diff options
Diffstat (limited to 'drivers/gpu/drm/amd/display')
328 files changed, 49249 insertions, 7015 deletions
diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig index f954bf61af28..87858bc57e64 100644 --- a/drivers/gpu/drm/amd/display/Kconfig +++ b/drivers/gpu/drm/amd/display/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: MIT menu "Display Engine Configuration" depends on DRM && DRM_AMDGPU @@ -6,35 +6,24 @@ config DRM_AMD_DC bool "AMD DC - Enable new display engine" default y select SND_HDA_COMPONENT if SND_HDA_CORE - select DRM_AMD_DC_DCN1_0 if X86 && !(KCOV_INSTRUMENT_ALL && KCOV_ENABLE_COMPARISONS) + select DRM_AMD_DC_DCN if (X86 || PPC64) && !(KCOV_INSTRUMENT_ALL && KCOV_ENABLE_COMPARISONS) help Choose this option if you want to use the new display engine support for AMDGPU. This adds required support for Vega and Raven ASICs. -config DRM_AMD_DC_DCN1_0 +config DRM_AMD_DC_DCN def_bool n help - RV family support for display engine + Raven, Navi and Renoir family support for display engine -config DRM_AMD_DC_DCN2_0 - bool "DCN 2.0 family" - default y - depends on DRM_AMD_DC && X86 - depends on DRM_AMD_DC_DCN1_0 - help - Choose this option if you want to have - Navi support for display engine - -config DRM_AMD_DC_DSC_SUPPORT - bool "DSC support" - default y - depends on DRM_AMD_DC && X86 - depends on DRM_AMD_DC_DCN1_0 - depends on DRM_AMD_DC_DCN2_0 +config DRM_AMD_DC_HDCP + bool "Enable HDCP support in DC" + depends on DRM_AMD_DC help - Choose this option if you want to have - Dynamic Stream Compression support + Choose this option + if you want to support + HDCP authentication config DEBUG_KERNEL_DC bool "Enable kgdb break in DC" diff --git a/drivers/gpu/drm/amd/display/Makefile b/drivers/gpu/drm/amd/display/Makefile index 496cee000f10..2633de77de5e 100644 --- a/drivers/gpu/drm/amd/display/Makefile +++ b/drivers/gpu/drm/amd/display/Makefile @@ -34,11 +34,20 @@ subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/freesync subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/color subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/info_packet subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/power +subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dmub/inc + +ifdef CONFIG_DRM_AMD_DC_HDCP +subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/hdcp +endif #TODO: remove when Timing Sync feature is complete subdir-ccflags-y += -DBUILD_FEATURE_TIMING_SYNC=0 -DAL_LIBS = amdgpu_dm dc modules/freesync modules/color modules/info_packet modules/power +DAL_LIBS = amdgpu_dm dc modules/freesync modules/color modules/info_packet modules/power dmub/src + +ifdef CONFIG_DRM_AMD_DC_HDCP +DAL_LIBS += modules/hdcp +endif AMD_DAL = $(addsuffix /Makefile, $(addprefix $(FULL_AMD_DISPLAY_PATH)/,$(DAL_LIBS))) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile b/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile index 94911871eb9b..9a3b7bf8ab0b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile @@ -31,6 +31,10 @@ ifneq ($(CONFIG_DRM_AMD_DC),) AMDGPUDM += amdgpu_dm_services.o amdgpu_dm_helpers.o amdgpu_dm_pp_smu.o endif +ifdef CONFIG_DRM_AMD_DC_HDCP +AMDGPUDM += amdgpu_dm_hdcp.o +endif + ifneq ($(CONFIG_DEBUG_FS),) AMDGPUDM += amdgpu_dm_crc.o amdgpu_dm_debugfs.o endif diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 4a29f72334d0..63e8a12a74bc 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -30,6 +30,10 @@ #include "dc.h" #include "dc/inc/core_types.h" #include "dal_asic_id.h" +#include "dmub/inc/dmub_srv.h" +#include "dc/inc/hw/dmcu.h" +#include "dc/inc/hw/abm.h" +#include "dc/dc_dmub_srv.h" #include "vid.h" #include "amdgpu.h" @@ -37,6 +41,10 @@ #include "amdgpu_ucode.h" #include "atom.h" #include "amdgpu_dm.h" +#ifdef CONFIG_DRM_AMD_DC_HDCP +#include "amdgpu_dm_hdcp.h" +#include <drm/drm_hdcp.h> +#endif #include "amdgpu_pm.h" #include "amd_shared.h" @@ -67,8 +75,9 @@ #include <drm/drm_edid.h> #include <drm/drm_vblank.h> #include <drm/drm_audio_component.h> +#include <drm/drm_hdcp.h> -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) #include "ivsrcid/dcn/irqsrcs_dcn_1_0.h" #include "dcn/dcn_1_0_offset.h" @@ -83,9 +92,18 @@ #include "modules/power/power_helpers.h" #include "modules/inc/mod_info_packet.h" +#define FIRMWARE_RENOIR_DMUB "amdgpu/renoir_dmcub.bin" +MODULE_FIRMWARE(FIRMWARE_RENOIR_DMUB); + #define FIRMWARE_RAVEN_DMCU "amdgpu/raven_dmcu.bin" MODULE_FIRMWARE(FIRMWARE_RAVEN_DMCU); +/* Number of bytes in PSP header for firmware. */ +#define PSP_HEADER_BYTES 0x100 + +/* Number of bytes in PSP footer for firmware. */ +#define PSP_FOOTER_BYTES 0x100 + /** * DOC: overview * @@ -143,6 +161,12 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, static void handle_cursor_update(struct drm_plane *plane, struct drm_plane_state *old_plane_state); +static void amdgpu_dm_set_psr_caps(struct dc_link *link); +static bool amdgpu_dm_psr_enable(struct dc_stream_state *stream); +static bool amdgpu_dm_link_setup_psr(struct dc_stream_state *stream); +static bool amdgpu_dm_psr_disable(struct dc_stream_state *stream); + + /* * dm_vblank_get_counter * @@ -263,6 +287,13 @@ static inline bool amdgpu_dm_vrr_active(struct dm_crtc_state *dm_state) dm_state->freesync_config.state == VRR_STATE_ACTIVE_FIXED; } +/** + * dm_pflip_high_irq() - Handle pageflip interrupt + * @interrupt_params: ignored + * + * Handles the pageflip interrupt by notifying all interested parties + * that the pageflip has been completed. + */ static void dm_pflip_high_irq(void *interrupt_params) { struct amdgpu_crtc *amdgpu_crtc; @@ -407,6 +438,13 @@ static void dm_vupdate_high_irq(void *interrupt_params) } } +/** + * dm_crtc_high_irq() - Handles CRTC interrupt + * @interrupt_params: ignored + * + * Handles the CRTC/VSYNC interrupt by notfying DRM's VBLANK + * event handler. + */ static void dm_crtc_high_irq(void *interrupt_params) { struct common_irq_params *irq_params = interrupt_params; @@ -454,6 +492,70 @@ static void dm_crtc_high_irq(void *interrupt_params) } } +#if defined(CONFIG_DRM_AMD_DC_DCN) +/** + * dm_dcn_crtc_high_irq() - Handles VStartup interrupt for DCN generation ASICs + * @interrupt params - interrupt parameters + * + * Notify DRM's vblank event handler at VSTARTUP + * + * Unlike DCE hardware, we trigger the handler at VSTARTUP. at which: + * * We are close enough to VUPDATE - the point of no return for hw + * * We are in the fixed portion of variable front porch when vrr is enabled + * * We are before VUPDATE, where double-buffered vrr registers are swapped + * + * It is therefore the correct place to signal vblank, send user flip events, + * and update VRR. + */ +static void dm_dcn_crtc_high_irq(void *interrupt_params) +{ + struct common_irq_params *irq_params = interrupt_params; + struct amdgpu_device *adev = irq_params->adev; + struct amdgpu_crtc *acrtc; + struct dm_crtc_state *acrtc_state; + unsigned long flags; + + acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VBLANK); + + if (!acrtc) + return; + + acrtc_state = to_dm_crtc_state(acrtc->base.state); + + DRM_DEBUG_DRIVER("crtc:%d, vupdate-vrr:%d\n", acrtc->crtc_id, + amdgpu_dm_vrr_active(acrtc_state)); + + amdgpu_dm_crtc_handle_crc_irq(&acrtc->base); + drm_crtc_handle_vblank(&acrtc->base); + + spin_lock_irqsave(&adev->ddev->event_lock, flags); + + if (acrtc_state->vrr_params.supported && + acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE) { + mod_freesync_handle_v_update( + adev->dm.freesync_module, + acrtc_state->stream, + &acrtc_state->vrr_params); + + dc_stream_adjust_vmin_vmax( + adev->dm.dc, + acrtc_state->stream, + &acrtc_state->vrr_params.adjust); + } + + if (acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED) { + if (acrtc->event) { + drm_crtc_send_vblank_event(&acrtc->base, acrtc->event); + acrtc->event = NULL; + drm_crtc_vblank_put(&acrtc->base); + } + acrtc->pflip_status = AMDGPU_FLIP_NONE; + } + + spin_unlock_irqrestore(&adev->ddev->event_lock, flags); +} +#endif + static int dm_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -643,14 +745,135 @@ void amdgpu_dm_audio_eld_notify(struct amdgpu_device *adev, int pin) } } +static int dm_dmub_hw_init(struct amdgpu_device *adev) +{ + const struct dmcub_firmware_header_v1_0 *hdr; + struct dmub_srv *dmub_srv = adev->dm.dmub_srv; + struct dmub_srv_fb_info *fb_info = adev->dm.dmub_fb_info; + const struct firmware *dmub_fw = adev->dm.dmub_fw; + struct dmcu *dmcu = adev->dm.dc->res_pool->dmcu; + struct abm *abm = adev->dm.dc->res_pool->abm; + struct dmub_srv_hw_params hw_params; + enum dmub_status status; + const unsigned char *fw_inst_const, *fw_bss_data; + uint32_t i, fw_inst_const_size, fw_bss_data_size; + bool has_hw_support; + + if (!dmub_srv) + /* DMUB isn't supported on the ASIC. */ + return 0; + + if (!fb_info) { + DRM_ERROR("No framebuffer info for DMUB service.\n"); + return -EINVAL; + } + + if (!dmub_fw) { + /* Firmware required for DMUB support. */ + DRM_ERROR("No firmware provided for DMUB.\n"); + return -EINVAL; + } + + status = dmub_srv_has_hw_support(dmub_srv, &has_hw_support); + if (status != DMUB_STATUS_OK) { + DRM_ERROR("Error checking HW support for DMUB: %d\n", status); + return -EINVAL; + } + + if (!has_hw_support) { + DRM_INFO("DMUB unsupported on ASIC\n"); + return 0; + } + + hdr = (const struct dmcub_firmware_header_v1_0 *)dmub_fw->data; + + fw_inst_const = dmub_fw->data + + le32_to_cpu(hdr->header.ucode_array_offset_bytes) + + PSP_HEADER_BYTES; + + fw_bss_data = dmub_fw->data + + le32_to_cpu(hdr->header.ucode_array_offset_bytes) + + le32_to_cpu(hdr->inst_const_bytes); + + /* Copy firmware and bios info into FB memory. */ + fw_inst_const_size = le32_to_cpu(hdr->inst_const_bytes) - + PSP_HEADER_BYTES - PSP_FOOTER_BYTES; + + fw_bss_data_size = le32_to_cpu(hdr->bss_data_bytes); + + memcpy(fb_info->fb[DMUB_WINDOW_0_INST_CONST].cpu_addr, fw_inst_const, + fw_inst_const_size); + memcpy(fb_info->fb[DMUB_WINDOW_2_BSS_DATA].cpu_addr, fw_bss_data, + fw_bss_data_size); + memcpy(fb_info->fb[DMUB_WINDOW_3_VBIOS].cpu_addr, adev->bios, + adev->bios_size); + + /* Reset regions that need to be reset. */ + memset(fb_info->fb[DMUB_WINDOW_4_MAILBOX].cpu_addr, 0, + fb_info->fb[DMUB_WINDOW_4_MAILBOX].size); + + memset(fb_info->fb[DMUB_WINDOW_5_TRACEBUFF].cpu_addr, 0, + fb_info->fb[DMUB_WINDOW_5_TRACEBUFF].size); + + memset(fb_info->fb[DMUB_WINDOW_6_FW_STATE].cpu_addr, 0, + fb_info->fb[DMUB_WINDOW_6_FW_STATE].size); + + /* Initialize hardware. */ + memset(&hw_params, 0, sizeof(hw_params)); + hw_params.fb_base = adev->gmc.fb_start; + hw_params.fb_offset = adev->gmc.aper_base; + + if (dmcu) + hw_params.psp_version = dmcu->psp_version; + + for (i = 0; i < fb_info->num_fb; ++i) + hw_params.fb[i] = &fb_info->fb[i]; + + status = dmub_srv_hw_init(dmub_srv, &hw_params); + if (status != DMUB_STATUS_OK) { + DRM_ERROR("Error initializing DMUB HW: %d\n", status); + return -EINVAL; + } + + /* Wait for firmware load to finish. */ + status = dmub_srv_wait_for_auto_load(dmub_srv, 100000); + if (status != DMUB_STATUS_OK) + DRM_WARN("Wait for DMUB auto-load failed: %d\n", status); + + /* Init DMCU and ABM if available. */ + if (dmcu && abm) { + dmcu->funcs->dmcu_init(dmcu); + abm->dmcu_is_running = dmcu->funcs->is_dmcu_initialized(dmcu); + } + + adev->dm.dc->ctx->dmub_srv = dc_dmub_srv_create(adev->dm.dc, dmub_srv); + if (!adev->dm.dc->ctx->dmub_srv) { + DRM_ERROR("Couldn't allocate DC DMUB server!\n"); + return -ENOMEM; + } + + DRM_INFO("DMUB hardware initialized: version=0x%08X\n", + adev->dm.dmcub_fw_version); + + return 0; +} + static int amdgpu_dm_init(struct amdgpu_device *adev) { struct dc_init_data init_data; +#ifdef CONFIG_DRM_AMD_DC_HDCP + struct dc_callback_init init_params; +#endif + int r; + adev->dm.ddev = adev->ddev; adev->dm.adev = adev; /* Zero all the fields */ memset(&init_data, 0, sizeof(init_data)); +#ifdef CONFIG_DRM_AMD_DC_HDCP + memset(&init_params, 0, sizeof(init_params)); +#endif mutex_init(&adev->dm.dc_lock); mutex_init(&adev->dm.audio_lock); @@ -683,22 +906,29 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) init_data.dce_environment = DCE_ENV_PRODUCTION_DRV; - /* - * TODO debug why this doesn't work on Raven - */ - if (adev->flags & AMD_IS_APU && - adev->asic_type >= CHIP_CARRIZO && - adev->asic_type < CHIP_RAVEN) + switch (adev->asic_type) { + case CHIP_CARRIZO: + case CHIP_STONEY: + case CHIP_RAVEN: + case CHIP_RENOIR: init_data.flags.gpu_vm_support = true; + break; + default: + break; + } if (amdgpu_dc_feature_mask & DC_FBC_MASK) init_data.flags.fbc_support = true; + if (amdgpu_dc_feature_mask & DC_MULTI_MON_PP_MCLK_SWITCH_MASK) + init_data.flags.multi_mon_pp_mclk_switch = true; + + if (amdgpu_dc_feature_mask & DC_DISABLE_FRACTIONAL_PWM_MASK) + init_data.flags.disable_fractional_pwm = true; + init_data.flags.power_down_display_on_boot = true; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 init_data.soc_bounding_box = adev->dm.soc_bounding_box; -#endif /* Display Core create. */ adev->dm.dc = dc_create(&init_data); @@ -710,6 +940,14 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) goto error; } + r = dm_dmub_hw_init(adev); + if (r) { + DRM_ERROR("DMUB interface failed to initialize: status=%d\n", r); + goto error; + } + + dc_hardware_init(adev->dm.dc); + adev->dm.freesync_module = mod_freesync_create(adev->dm.dc); if (!adev->dm.freesync_module) { DRM_ERROR( @@ -720,6 +958,18 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) amdgpu_dm_init_color_mod(); +#ifdef CONFIG_DRM_AMD_DC_HDCP + if (adev->asic_type >= CHIP_RAVEN) { + adev->dm.hdcp_workqueue = hdcp_create_workqueue(&adev->psp, &init_params.cp_psp, adev->dm.dc); + + if (!adev->dm.hdcp_workqueue) + DRM_ERROR("amdgpu: failed to initialize hdcp_workqueue.\n"); + else + DRM_DEBUG_DRIVER("amdgpu: hdcp_workqueue init done %p.\n", adev->dm.hdcp_workqueue); + + dc_init_callbacks(adev->dm.dc, &init_params); + } +#endif if (amdgpu_dm_initialize_drm_device(adev)) { DRM_ERROR( "amdgpu: failed to initialize sw for display support.\n"); @@ -761,6 +1011,25 @@ static void amdgpu_dm_fini(struct amdgpu_device *adev) amdgpu_dm_destroy_drm_device(&adev->dm); +#ifdef CONFIG_DRM_AMD_DC_HDCP + if (adev->dm.hdcp_workqueue) { + hdcp_destroy(adev->dm.hdcp_workqueue); + adev->dm.hdcp_workqueue = NULL; + } + + if (adev->dm.dc) + dc_deinit_callbacks(adev->dm.dc); +#endif + if (adev->dm.dc->ctx->dmub_srv) { + dc_dmub_srv_destroy(&adev->dm.dc->ctx->dmub_srv); + adev->dm.dc->ctx->dmub_srv = NULL; + } + + if (adev->dm.dmub_bo) + amdgpu_bo_free_kernel(&adev->dm.dmub_bo, + &adev->dm.dmub_bo_gpu_addr, + &adev->dm.dmub_bo_cpu_addr); + /* DC Destroy TODO: Replace destroy DAL */ if (adev->dm.dc) dc_destroy(&adev->dm.dc); @@ -809,6 +1078,9 @@ static int load_dmcu_fw(struct amdgpu_device *adev) case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_NAVI10: + case CHIP_NAVI14: + case CHIP_NAVI12: + case CHIP_RENOIR: return 0; case CHIP_RAVEN: if (ASICREV_IS_PICASSO(adev->external_rev_id)) @@ -868,9 +1140,160 @@ static int load_dmcu_fw(struct amdgpu_device *adev) return 0; } +static uint32_t amdgpu_dm_dmub_reg_read(void *ctx, uint32_t address) +{ + struct amdgpu_device *adev = ctx; + + return dm_read_reg(adev->dm.dc->ctx, address); +} + +static void amdgpu_dm_dmub_reg_write(void *ctx, uint32_t address, + uint32_t value) +{ + struct amdgpu_device *adev = ctx; + + return dm_write_reg(adev->dm.dc->ctx, address, value); +} + +static int dm_dmub_sw_init(struct amdgpu_device *adev) +{ + struct dmub_srv_create_params create_params; + struct dmub_srv_region_params region_params; + struct dmub_srv_region_info region_info; + struct dmub_srv_fb_params fb_params; + struct dmub_srv_fb_info *fb_info; + struct dmub_srv *dmub_srv; + const struct dmcub_firmware_header_v1_0 *hdr; + const char *fw_name_dmub; + enum dmub_asic dmub_asic; + enum dmub_status status; + int r; + + switch (adev->asic_type) { + case CHIP_RENOIR: + dmub_asic = DMUB_ASIC_DCN21; + fw_name_dmub = FIRMWARE_RENOIR_DMUB; + break; + + default: + /* ASIC doesn't support DMUB. */ + return 0; + } + + r = request_firmware_direct(&adev->dm.dmub_fw, fw_name_dmub, adev->dev); + if (r) { + DRM_ERROR("DMUB firmware loading failed: %d\n", r); + return 0; + } + + r = amdgpu_ucode_validate(adev->dm.dmub_fw); + if (r) { + DRM_ERROR("Couldn't validate DMUB firmware: %d\n", r); + return 0; + } + + if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { + DRM_WARN("Only PSP firmware loading is supported for DMUB\n"); + return 0; + } + + hdr = (const struct dmcub_firmware_header_v1_0 *)adev->dm.dmub_fw->data; + adev->firmware.ucode[AMDGPU_UCODE_ID_DMCUB].ucode_id = + AMDGPU_UCODE_ID_DMCUB; + adev->firmware.ucode[AMDGPU_UCODE_ID_DMCUB].fw = adev->dm.dmub_fw; + adev->firmware.fw_size += + ALIGN(le32_to_cpu(hdr->inst_const_bytes), PAGE_SIZE); + + adev->dm.dmcub_fw_version = le32_to_cpu(hdr->header.ucode_version); + + DRM_INFO("Loading DMUB firmware via PSP: version=0x%08X\n", + adev->dm.dmcub_fw_version); + + adev->dm.dmub_srv = kzalloc(sizeof(*adev->dm.dmub_srv), GFP_KERNEL); + dmub_srv = adev->dm.dmub_srv; + + if (!dmub_srv) { + DRM_ERROR("Failed to allocate DMUB service!\n"); + return -ENOMEM; + } + + memset(&create_params, 0, sizeof(create_params)); + create_params.user_ctx = adev; + create_params.funcs.reg_read = amdgpu_dm_dmub_reg_read; + create_params.funcs.reg_write = amdgpu_dm_dmub_reg_write; + create_params.asic = dmub_asic; + + /* Create the DMUB service. */ + status = dmub_srv_create(dmub_srv, &create_params); + if (status != DMUB_STATUS_OK) { + DRM_ERROR("Error creating DMUB service: %d\n", status); + return -EINVAL; + } + + /* Calculate the size of all the regions for the DMUB service. */ + memset(®ion_params, 0, sizeof(region_params)); + + region_params.inst_const_size = le32_to_cpu(hdr->inst_const_bytes) - + PSP_HEADER_BYTES - PSP_FOOTER_BYTES; + region_params.bss_data_size = le32_to_cpu(hdr->bss_data_bytes); + region_params.vbios_size = adev->bios_size; + region_params.fw_bss_data = + adev->dm.dmub_fw->data + + le32_to_cpu(hdr->header.ucode_array_offset_bytes) + + le32_to_cpu(hdr->inst_const_bytes); + + status = dmub_srv_calc_region_info(dmub_srv, ®ion_params, + ®ion_info); + + if (status != DMUB_STATUS_OK) { + DRM_ERROR("Error calculating DMUB region info: %d\n", status); + return -EINVAL; + } + + /* + * Allocate a framebuffer based on the total size of all the regions. + * TODO: Move this into GART. + */ + r = amdgpu_bo_create_kernel(adev, region_info.fb_size, PAGE_SIZE, + AMDGPU_GEM_DOMAIN_VRAM, &adev->dm.dmub_bo, + &adev->dm.dmub_bo_gpu_addr, + &adev->dm.dmub_bo_cpu_addr); + if (r) + return r; + + /* Rebase the regions on the framebuffer address. */ + memset(&fb_params, 0, sizeof(fb_params)); + fb_params.cpu_addr = adev->dm.dmub_bo_cpu_addr; + fb_params.gpu_addr = adev->dm.dmub_bo_gpu_addr; + fb_params.region_info = ®ion_info; + + adev->dm.dmub_fb_info = + kzalloc(sizeof(*adev->dm.dmub_fb_info), GFP_KERNEL); + fb_info = adev->dm.dmub_fb_info; + + if (!fb_info) { + DRM_ERROR( + "Failed to allocate framebuffer info for DMUB service!\n"); + return -ENOMEM; + } + + status = dmub_srv_calc_fb_info(dmub_srv, &fb_params, fb_info); + if (status != DMUB_STATUS_OK) { + DRM_ERROR("Error calculating DMUB FB info: %d\n", status); + return -EINVAL; + } + + return 0; +} + static int dm_sw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; + + r = dm_dmub_sw_init(adev); + if (r) + return r; return load_dmcu_fw(adev); } @@ -879,6 +1302,19 @@ static int dm_sw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + kfree(adev->dm.dmub_fb_info); + adev->dm.dmub_fb_info = NULL; + + if (adev->dm.dmub_srv) { + dmub_srv_destroy(adev->dm.dmub_srv); + adev->dm.dmub_srv = NULL; + } + + if (adev->dm.dmub_fw) { + release_firmware(adev->dm.dmub_fw); + adev->dm.dmub_fw = NULL; + } + if(adev->dm.fw_dmcu) { release_firmware(adev->dm.fw_dmcu); adev->dm.fw_dmcu = NULL; @@ -891,27 +1327,29 @@ static int detect_mst_link_for_all_connectors(struct drm_device *dev) { struct amdgpu_dm_connector *aconnector; struct drm_connector *connector; + struct drm_connector_list_iter iter; int ret = 0; - drm_modeset_lock(&dev->mode_config.connection_mutex, NULL); - - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + drm_connector_list_iter_begin(dev, &iter); + drm_for_each_connector_iter(connector, &iter) { aconnector = to_amdgpu_dm_connector(connector); if (aconnector->dc_link->type == dc_connection_mst_branch && aconnector->mst_mgr.aux) { DRM_DEBUG_DRIVER("DM_MST: starting TM on aconnector: %p [id: %d]\n", - aconnector, aconnector->base.base.id); + aconnector, + aconnector->base.base.id); ret = drm_dp_mst_topology_mgr_set_mst(&aconnector->mst_mgr, true); if (ret < 0) { DRM_ERROR("DM_MST: Failed to start MST\n"); - ((struct dc_link *)aconnector->dc_link)->type = dc_connection_single; - return ret; - } + aconnector->dc_link->type = + dc_connection_single; + break; } + } } + drm_connector_list_iter_end(&iter); - drm_modeset_unlock(&dev->mode_config.connection_mutex); return ret; } @@ -934,6 +1372,11 @@ static int dm_late_init(void *handle) params.backlight_lut_array_size = 16; params.backlight_lut_array = linear_lut; + /* Min backlight level after ABM reduction, Don't allow below 1% + * 0xFFFF x 0.01 = 0x28F + */ + params.min_abm_backlight = 0x28F; + /* todo will enable for navi10 */ if (adev->asic_type <= CHIP_RAVEN) { ret = dmcu_load_iram(dmcu, params); @@ -949,14 +1392,13 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend) { struct amdgpu_dm_connector *aconnector; struct drm_connector *connector; + struct drm_connector_list_iter iter; struct drm_dp_mst_topology_mgr *mgr; int ret; bool need_hotplug = false; - drm_modeset_lock(&dev->mode_config.connection_mutex, NULL); - - list_for_each_entry(connector, &dev->mode_config.connector_list, - head) { + drm_connector_list_iter_begin(dev, &iter); + drm_for_each_connector_iter(connector, &iter) { aconnector = to_amdgpu_dm_connector(connector); if (aconnector->dc_link->type != dc_connection_mst_branch || aconnector->mst_port) @@ -967,15 +1409,14 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend) if (suspend) { drm_dp_mst_topology_mgr_suspend(mgr); } else { - ret = drm_dp_mst_topology_mgr_resume(mgr); + ret = drm_dp_mst_topology_mgr_resume(mgr, true); if (ret < 0) { drm_dp_mst_topology_mgr_set_mst(mgr, false); need_hotplug = true; } } } - - drm_modeset_unlock(&dev->mode_config.connection_mutex); + drm_connector_list_iter_end(&iter); if (need_hotplug) drm_kms_helper_hotplug_event(dev); @@ -983,7 +1424,7 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend) /** * dm_hw_init() - Initialize DC device - * @handle: The base driver device containing the amdpgu_dm device. + * @handle: The base driver device containing the amdgpu_dm device. * * Initialize the &struct amdgpu_display_manager device. This involves calling * the initializers of each DM component, then populating the struct with them. @@ -1013,7 +1454,7 @@ static int dm_hw_init(void *handle) /** * dm_hw_fini() - Teardown DC device - * @handle: The base driver device containing the amdpgu_dm device. + * @handle: The base driver device containing the amdgpu_dm device. * * Teardown components within &struct amdgpu_display_manager that require * cleanup. This involves cleaning up the DRM device, DC, and any modules that @@ -1157,6 +1598,7 @@ static int dm_resume(void *handle) struct amdgpu_display_manager *dm = &adev->dm; struct amdgpu_dm_connector *aconnector; struct drm_connector *connector; + struct drm_connector_list_iter iter; struct drm_crtc *crtc; struct drm_crtc_state *new_crtc_state; struct dm_crtc_state *dm_new_crtc_state; @@ -1165,7 +1607,7 @@ static int dm_resume(void *handle) struct dm_plane_state *dm_new_plane_state; struct dm_atomic_state *dm_state = to_dm_atomic_state(dm->atomic_obj.state); enum dc_connection_type new_connection_type = dc_connection_none; - int i; + int i, r; /* Recreate dc_state - DC invalidates it when setting power state to S3. */ dc_release_state(dm_state->context); @@ -1173,23 +1615,29 @@ static int dm_resume(void *handle) /* TODO: Remove dc_state->dccg, use dc->dccg directly. */ dc_resource_state_construct(dm->dc, dm_state->context); + /* Before powering on DC we need to re-initialize DMUB. */ + r = dm_dmub_hw_init(adev); + if (r) + DRM_ERROR("DMUB interface failed to initialize: status=%d\n", r); + /* power on hardware */ dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D0); /* program HPD filter */ dc_resume(dm->dc); - /* On resume we need to rewrite the MSTM control bits to enamble MST*/ - s3_handle_mst(ddev, false); - /* * early enable HPD Rx IRQ, should be done before set mode as short * pulse interrupts are used for MST */ amdgpu_dm_irq_resume_early(adev); + /* On resume we need to rewrite the MSTM control bits to enable MST*/ + s3_handle_mst(ddev, false); + /* Do detection*/ - list_for_each_entry(connector, &ddev->mode_config.connector_list, head) { + drm_connector_list_iter_begin(ddev, &iter); + drm_for_each_connector_iter(connector, &iter) { aconnector = to_amdgpu_dm_connector(connector); /* @@ -1217,6 +1665,7 @@ static int dm_resume(void *handle) amdgpu_dm_update_connector_after_detect(aconnector); mutex_unlock(&aconnector->hpd_lock); } + drm_connector_list_iter_end(&iter); /* Force mode set in atomic commit */ for_each_new_crtc_in_state(dm->cached_state, crtc, new_crtc_state, i) @@ -1432,6 +1881,11 @@ amdgpu_dm_update_connector_after_detect(struct amdgpu_dm_connector *aconnector) dc_sink_release(aconnector->dc_sink); aconnector->dc_sink = NULL; aconnector->edid = NULL; +#ifdef CONFIG_DRM_AMD_DC_HDCP + /* Set CP to DESIRED if it was ENABLED, so we can re-enable it again on hotplug */ + if (connector->state->content_protection == DRM_MODE_CONTENT_PROTECTION_ENABLED) + connector->state->content_protection = DRM_MODE_CONTENT_PROTECTION_DESIRED; +#endif } mutex_unlock(&dev->mode_config.mutex); @@ -1446,6 +1900,9 @@ static void handle_hpd_irq(void *param) struct drm_connector *connector = &aconnector->base; struct drm_device *dev = connector->dev; enum dc_connection_type new_connection_type = dc_connection_none; +#ifdef CONFIG_DRM_AMD_DC_HDCP + struct amdgpu_device *adev = dev->dev_private; +#endif /* * In case of failure or MST no need to update connector status or notify the OS @@ -1453,6 +1910,10 @@ static void handle_hpd_irq(void *param) */ mutex_lock(&aconnector->hpd_lock); +#ifdef CONFIG_DRM_AMD_DC_HDCP + if (adev->asic_type >= CHIP_RAVEN) + hdcp_reset_display(adev->dm.hdcp_workqueue, aconnector->dc_link->link_index); +#endif if (aconnector->fake_enable) aconnector->fake_enable = false; @@ -1571,6 +2032,12 @@ static void handle_hpd_rx_irq(void *param) struct dc_link *dc_link = aconnector->dc_link; bool is_mst_root_connector = aconnector->mst_mgr.mst_state; enum dc_connection_type new_connection_type = dc_connection_none; +#ifdef CONFIG_DRM_AMD_DC_HDCP + union hpd_irq_data hpd_irq_data; + struct amdgpu_device *adev = dev->dev_private; + + memset(&hpd_irq_data, 0, sizeof(hpd_irq_data)); +#endif /* * TODO:Temporary add mutex to protect hpd interrupt not have a gpio @@ -1580,7 +2047,12 @@ static void handle_hpd_rx_irq(void *param) if (dc_link->type != dc_connection_mst_branch) mutex_lock(&aconnector->hpd_lock); + +#ifdef CONFIG_DRM_AMD_DC_HDCP + if (dc_link_handle_hpd_rx_irq(dc_link, &hpd_irq_data, NULL) && +#else if (dc_link_handle_hpd_rx_irq(dc_link, NULL, NULL) && +#endif !is_mst_root_connector) { /* Downstream Port status changed. */ if (!dc_link_detect_sink(dc_link, &new_connection_type)) @@ -1615,6 +2087,10 @@ static void handle_hpd_rx_irq(void *param) drm_kms_helper_hotplug_event(dev); } } +#ifdef CONFIG_DRM_AMD_DC_HDCP + if (hpd_irq_data.bytes.device_service_irq.bits.CP_IRQ) + hdcp_handle_cpirq(adev->dm.hdcp_workqueue, aconnector->base.index); +#endif if ((dc_link->cur_link_settings.lane_count != LANE_COUNT_UNKNOWN) || (dc_link->type == dc_connection_mst_branch)) dm_handle_hpd_rx_irq(aconnector); @@ -1769,7 +2245,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev) return 0; } -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) /* Register IRQ sources and initialize IRQ callbacks */ static int dcn10_register_irq_handlers(struct amdgpu_device *adev) { @@ -1815,35 +2291,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev) c_irq_params->irq_src = int_params.irq_source; amdgpu_dm_irq_register_interrupt(adev, &int_params, - dm_crtc_high_irq, c_irq_params); - } - - /* Use VUPDATE_NO_LOCK interrupt on DCN, which seems to correspond to - * the regular VUPDATE interrupt on DCE. We want DC_IRQ_SOURCE_VUPDATEx - * to trigger at end of each vblank, regardless of state of the lock, - * matching DCE behaviour. - */ - for (i = DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT; - i <= DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT + adev->mode_info.num_crtc - 1; - i++) { - r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DCE, i, &adev->vupdate_irq); - - if (r) { - DRM_ERROR("Failed to add vupdate irq id!\n"); - return r; - } - - int_params.int_context = INTERRUPT_HIGH_IRQ_CONTEXT; - int_params.irq_source = - dc_interrupt_to_irq_source(dc, i, 0); - - c_irq_params = &adev->dm.vupdate_params[int_params.irq_source - DC_IRQ_SOURCE_VUPDATE1]; - - c_irq_params->adev = adev; - c_irq_params->irq_src = int_params.irq_source; - - amdgpu_dm_irq_register_interrupt(adev, &int_params, - dm_vupdate_high_irq, c_irq_params); + dm_dcn_crtc_high_irq, c_irq_params); } /* Use GRPH_PFLIP interrupt */ @@ -2107,6 +2555,7 @@ static int amdgpu_dm_backlight_get_brightness(struct backlight_device *bd) } static const struct backlight_ops amdgpu_dm_backlight_ops = { + .options = BL_CORE_SUSPENDRESUME, .get_brightness = amdgpu_dm_backlight_get_brightness, .update_status = amdgpu_dm_backlight_update_status, }; @@ -2327,6 +2776,8 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) } else if (dc_link_detect(link, DETECT_REASON_BOOT)) { amdgpu_dm_update_connector_after_detect(aconnector); register_backlight_device(dm, link); + if (amdgpu_dc_feature_mask & DC_PSR_MASK) + amdgpu_dm_set_psr_caps(link); } @@ -2355,11 +2806,12 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) goto fail; } break; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) case CHIP_RAVEN: -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) + case CHIP_NAVI12: case CHIP_NAVI10: -#endif + case CHIP_NAVI14: + case CHIP_RENOIR: if (dcn10_register_irq_handlers(dm->adev)) { DRM_ERROR("DM: Failed to initialize IRQ\n"); goto fail; @@ -2428,8 +2880,7 @@ static ssize_t s3_debug_store(struct device *device, { int ret; int s3_state; - struct pci_dev *pdev = to_pci_dev(device); - struct drm_device *drm_dev = pci_get_drvdata(pdev); + struct drm_device *drm_dev = dev_get_drvdata(device); struct amdgpu_device *adev = drm_dev->dev_private; ret = kstrtoint(buf, 0, &s3_state); @@ -2506,20 +2957,29 @@ static int dm_early_init(void *handle) adev->mode_info.num_hpd = 6; adev->mode_info.num_dig = 6; break; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) case CHIP_RAVEN: adev->mode_info.num_crtc = 4; adev->mode_info.num_hpd = 4; adev->mode_info.num_dig = 4; break; #endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) case CHIP_NAVI10: + case CHIP_NAVI12: adev->mode_info.num_crtc = 6; adev->mode_info.num_hpd = 6; adev->mode_info.num_dig = 6; break; -#endif + case CHIP_NAVI14: + adev->mode_info.num_crtc = 5; + adev->mode_info.num_hpd = 5; + adev->mode_info.num_dig = 5; + break; + case CHIP_RENOIR: + adev->mode_info.num_crtc = 4; + adev->mode_info.num_hpd = 4; + adev->mode_info.num_dig = 4; + break; default: DRM_ERROR("Unsupported ASIC type: 0x%X\n", adev->asic_type); return -EINVAL; @@ -2665,7 +3125,7 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, const struct amdgpu_framebuffer *afb, const enum surface_pixel_format format, const enum dc_rotation_angle rotation, - const union plane_size *plane_size, + const struct plane_size *plane_size, const union dc_tiling_info *tiling_info, const uint64_t info, struct dc_plane_dcc_param *dcc, @@ -2691,8 +3151,8 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, return -EINVAL; input.format = format; - input.surface_size.width = plane_size->grph.surface_size.width; - input.surface_size.height = plane_size->grph.surface_size.height; + input.surface_size.width = plane_size->surface_size.width; + input.surface_size.height = plane_size->surface_size.height; input.swizzle_mode = tiling_info->gfx9.swizzle; if (rotation == ROTATION_ANGLE_0 || rotation == ROTATION_ANGLE_180) @@ -2710,9 +3170,9 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, return -EINVAL; dcc->enable = 1; - dcc->grph.meta_pitch = + dcc->meta_pitch = AMDGPU_TILING_GET(info, DCC_PITCH_MAX) + 1; - dcc->grph.independent_64b_blks = i64b; + dcc->independent_64b_blks = i64b; dcc_address = get_dcc_address(afb->address, info); address->grph.meta_addr.low_part = lower_32_bits(dcc_address); @@ -2728,7 +3188,7 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, const enum dc_rotation_angle rotation, const uint64_t tiling_flags, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, struct dc_plane_dcc_param *dcc, struct dc_plane_address *address) { @@ -2741,11 +3201,11 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, memset(address, 0, sizeof(*address)); if (format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) { - plane_size->grph.surface_size.x = 0; - plane_size->grph.surface_size.y = 0; - plane_size->grph.surface_size.width = fb->width; - plane_size->grph.surface_size.height = fb->height; - plane_size->grph.surface_pitch = + plane_size->surface_size.x = 0; + plane_size->surface_size.y = 0; + plane_size->surface_size.width = fb->width; + plane_size->surface_size.height = fb->height; + plane_size->surface_pitch = fb->pitches[0] / fb->format->cpp[0]; address->type = PLN_ADDR_TYPE_GRAPHICS; @@ -2754,20 +3214,20 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, } else if (format < SURFACE_PIXEL_FORMAT_INVALID) { uint64_t chroma_addr = afb->address + fb->offsets[1]; - plane_size->video.luma_size.x = 0; - plane_size->video.luma_size.y = 0; - plane_size->video.luma_size.width = fb->width; - plane_size->video.luma_size.height = fb->height; - plane_size->video.luma_pitch = + plane_size->surface_size.x = 0; + plane_size->surface_size.y = 0; + plane_size->surface_size.width = fb->width; + plane_size->surface_size.height = fb->height; + plane_size->surface_pitch = fb->pitches[0] / fb->format->cpp[0]; - plane_size->video.chroma_size.x = 0; - plane_size->video.chroma_size.y = 0; + plane_size->chroma_size.x = 0; + plane_size->chroma_size.y = 0; /* TODO: set these based on surface format */ - plane_size->video.chroma_size.width = fb->width / 2; - plane_size->video.chroma_size.height = fb->height / 2; + plane_size->chroma_size.width = fb->width / 2; + plane_size->chroma_size.height = fb->height / 2; - plane_size->video.chroma_pitch = + plane_size->chroma_pitch = fb->pitches[1] / fb->format->cpp[1]; address->type = PLN_ADDR_TYPE_VIDEO_PROGRESSIVE; @@ -2812,9 +3272,10 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, if (adev->asic_type == CHIP_VEGA10 || adev->asic_type == CHIP_VEGA12 || adev->asic_type == CHIP_VEGA20 || -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) adev->asic_type == CHIP_NAVI10 || -#endif + adev->asic_type == CHIP_NAVI14 || + adev->asic_type == CHIP_NAVI12 || + adev->asic_type == CHIP_RENOIR || adev->asic_type == CHIP_RAVEN) { /* Fill GFX9 params */ tiling_info->gfx9.num_pipes = @@ -2995,6 +3456,8 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev, plane_info->visible = true; plane_info->stereo_format = PLANE_STEREO_FORMAT_NONE; + plane_info->layer_index = 0; + ret = fill_plane_color_attributes(plane_state, plane_info->format, &plane_info->color_space); if (ret) @@ -3060,6 +3523,7 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev, dc_plane_state->global_alpha = plane_info.global_alpha; dc_plane_state->global_alpha_value = plane_info.global_alpha_value; dc_plane_state->dcc = plane_info.dcc; + dc_plane_state->layer_index = plane_info.layer_index; // Always returns 0 /* * Always set input transfer function, since plane state is refreshed @@ -3129,15 +3593,41 @@ static void update_stream_scaling_settings(const struct drm_display_mode *mode, static enum dc_color_depth convert_color_depth_from_display_info(const struct drm_connector *connector, - const struct drm_connector_state *state) + const struct drm_connector_state *state, + bool is_y420) { - uint32_t bpc = connector->display_info.bpc; + uint8_t bpc; + + if (is_y420) { + bpc = 8; + + /* Cap display bpc based on HDMI 2.0 HF-VSDB */ + if (connector->display_info.hdmi.y420_dc_modes & DRM_EDID_YCBCR420_DC_48) + bpc = 16; + else if (connector->display_info.hdmi.y420_dc_modes & DRM_EDID_YCBCR420_DC_36) + bpc = 12; + else if (connector->display_info.hdmi.y420_dc_modes & DRM_EDID_YCBCR420_DC_30) + bpc = 10; + } else { + bpc = (uint8_t)connector->display_info.bpc; + /* Assume 8 bpc by default if no bpc is specified. */ + bpc = bpc ? bpc : 8; + } if (!state) state = connector->state; if (state) { - bpc = state->max_bpc; + /* + * Cap display bpc based on the user requested value. + * + * The value for state->max_bpc may not correctly updated + * depending on when the connector gets added to the state + * or if this was called outside of atomic check, so it + * can't be used directly. + */ + bpc = min(bpc, state->max_requested_bpc); + /* Round down to the nearest even number. */ bpc = bpc - (bpc & 1); } @@ -3217,27 +3707,21 @@ get_output_color_space(const struct dc_crtc_timing *dc_crtc_timing) return color_space; } -static void reduce_mode_colour_depth(struct dc_crtc_timing *timing_out) -{ - if (timing_out->display_color_depth <= COLOR_DEPTH_888) - return; - - timing_out->display_color_depth--; -} - -static void adjust_colour_depth_from_display_info(struct dc_crtc_timing *timing_out, - const struct drm_display_info *info) +static bool adjust_colour_depth_from_display_info( + struct dc_crtc_timing *timing_out, + const struct drm_display_info *info) { + enum dc_color_depth depth = timing_out->display_color_depth; int normalized_clk; - if (timing_out->display_color_depth <= COLOR_DEPTH_888) - return; do { normalized_clk = timing_out->pix_clk_100hz / 10; /* YCbCr 4:2:0 requires additional adjustment of 1/2 */ if (timing_out->pixel_encoding == PIXEL_ENCODING_YCBCR420) normalized_clk /= 2; /* Adjusting pix clock following on HDMI spec based on colour depth */ - switch (timing_out->display_color_depth) { + switch (depth) { + case COLOR_DEPTH_888: + break; case COLOR_DEPTH_101010: normalized_clk = (normalized_clk * 30) / 24; break; @@ -3248,14 +3732,15 @@ static void adjust_colour_depth_from_display_info(struct dc_crtc_timing *timing_ normalized_clk = (normalized_clk * 48) / 24; break; default: - return; + /* The above depths are the only ones valid for HDMI. */ + return false; } - if (normalized_clk <= info->max_tmds_clock) - return; - reduce_mode_colour_depth(timing_out); - - } while (timing_out->display_color_depth > COLOR_DEPTH_888); - + if (normalized_clk <= info->max_tmds_clock) { + timing_out->display_color_depth = depth; + return true; + } + } while (--depth > COLOR_DEPTH_666); + return false; } static void fill_stream_properties_from_drm_display_mode( @@ -3267,8 +3752,12 @@ static void fill_stream_properties_from_drm_display_mode( { struct dc_crtc_timing *timing_out = &stream->timing; const struct drm_display_info *info = &connector->display_info; + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); + struct hdmi_vendor_infoframe hv_frame; + struct hdmi_avi_infoframe avi_frame; - memset(timing_out, 0, sizeof(struct dc_crtc_timing)); + memset(&hv_frame, 0, sizeof(hv_frame)); + memset(&avi_frame, 0, sizeof(avi_frame)); timing_out->h_border_left = 0; timing_out->h_border_right = 0; @@ -3278,6 +3767,9 @@ static void fill_stream_properties_from_drm_display_mode( if (drm_mode_is_420_only(info, mode_in) && stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR420; + else if (drm_mode_is_420_also(info, mode_in) + && aconnector->force_yuv420_output) + timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR420; else if ((connector->display_info.color_formats & DRM_COLOR_FORMAT_YCRCB444) && stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR444; @@ -3286,7 +3778,8 @@ static void fill_stream_properties_from_drm_display_mode( timing_out->timing_3d_format = TIMING_3D_FORMAT_NONE; timing_out->display_color_depth = convert_color_depth_from_display_info( - connector, connector_state); + connector, connector_state, + (timing_out->pixel_encoding == PIXEL_ENCODING_YCBCR420)); timing_out->scan_type = SCANNING_TYPE_NODATA; timing_out->hdmi_vic = 0; @@ -3302,6 +3795,13 @@ static void fill_stream_properties_from_drm_display_mode( timing_out->flags.VSYNC_POSITIVE_POLARITY = 1; } + if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) { + drm_hdmi_avi_infoframe_from_display_mode(&avi_frame, (struct drm_connector *)connector, mode_in); + timing_out->vic = avi_frame.video_code; + drm_hdmi_vendor_infoframe_from_display_mode(&hv_frame, (struct drm_connector *)connector, mode_in); + timing_out->hdmi_vic = hv_frame.vic; + } + timing_out->h_addressable = mode_in->crtc_hdisplay; timing_out->h_total = mode_in->crtc_htotal; timing_out->h_sync_width = @@ -3321,8 +3821,14 @@ static void fill_stream_properties_from_drm_display_mode( stream->out_transfer_func->type = TF_TYPE_PREDEFINED; stream->out_transfer_func->tf = TRANSFER_FUNCTION_SRGB; - if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) - adjust_colour_depth_from_display_info(timing_out, info); + if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) { + if (!adjust_colour_depth_from_display_info(timing_out, info) && + drm_mode_is_420_also(info, mode_in) && + timing_out->pixel_encoding != PIXEL_ENCODING_YCBCR420) { + timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR420; + adjust_colour_depth_from_display_info(timing_out, info); + } + } } static void fill_audio_info(struct audio_info *audio_info, @@ -3491,6 +3997,10 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, bool scale = dm_state ? (dm_state->scaling != RMX_OFF) : false; int mode_refresh; int preferred_refresh = 0; +#if defined(CONFIG_DRM_AMD_DC_DCN) + struct dsc_dec_dpcd_caps dsc_caps; +#endif + uint32_t link_bandwidth_kbps; struct dc_sink *sink = NULL; if (aconnector == NULL) { @@ -3518,6 +4028,9 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, stream->dm_stream_context = aconnector; + stream->timing.flags.LTE_340MCSC_SCRAMBLE = + drm_connector->display_info.hdmi.scdc.scrambling.low_rates; + list_for_each_entry(preferred_mode, &aconnector->base.modes, head) { /* Search for preferred mode */ if (preferred_mode->type & DRM_MODE_TYPE_PREFERRED) { @@ -3562,19 +4075,29 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, fill_stream_properties_from_drm_display_mode(stream, &mode, &aconnector->base, con_state, old_stream); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - /* stream->timing.flags.DSC = 0; */ - /* */ - /* if (aconnector->dc_link && */ - /* aconnector->dc_link->connector_signal == SIGNAL_TYPE_DISPLAY_PORT #<{(|&& */ - /* aconnector->dc_link->dpcd_caps.dsc_caps.dsc_basic_caps.is_dsc_supported|)}>#) */ - /* if (dc_dsc_compute_config(aconnector->dc_link->ctx->dc, */ - /* &aconnector->dc_link->dpcd_caps.dsc_caps, */ - /* dc_link_bandwidth_kbps(aconnector->dc_link, dc_link_get_link_cap(aconnector->dc_link)), */ - /* &stream->timing, */ - /* &stream->timing.dsc_cfg)) */ - /* stream->timing.flags.DSC = 1; */ + stream->timing.flags.DSC = 0; + + if (aconnector->dc_link && sink->sink_signal == SIGNAL_TYPE_DISPLAY_PORT) { +#if defined(CONFIG_DRM_AMD_DC_DCN) + dc_dsc_parse_dsc_dpcd(aconnector->dc_link->ctx->dc, + aconnector->dc_link->dpcd_caps.dsc_caps.dsc_basic_caps.raw, + aconnector->dc_link->dpcd_caps.dsc_caps.dsc_ext_caps.raw, + &dsc_caps); +#endif + link_bandwidth_kbps = dc_link_bandwidth_kbps(aconnector->dc_link, + dc_link_get_link_cap(aconnector->dc_link)); + +#if defined(CONFIG_DRM_AMD_DC_DCN) + if (dsc_caps.is_dsc_supported) + if (dc_dsc_compute_config(aconnector->dc_link->ctx->dc->res_pool->dscs[0], + &dsc_caps, + aconnector->dc_link->ctx->dc->debug.dsc_min_slice_height_override, + link_bandwidth_kbps, + &stream->timing, + &stream->timing.dsc_cfg)) + stream->timing.flags.DSC = 1; #endif + } update_stream_scaling_settings(&mode, dm_state, stream); @@ -3585,6 +4108,20 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, update_stream_signal(stream, sink); + if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) + mod_build_hf_vsif_infopacket(stream, &stream->vsp_infopacket, false, false); + if (stream->link->psr_feature_enabled) { + struct dc *core_dc = stream->link->ctx->dc; + + if (dc_is_dmcu_initialized(core_dc)) { + struct dmcu *dmcu = core_dc->res_pool->dmcu; + + stream->psr_version = dmcu->dmcu_version.psr_version; + mod_build_vsc_infopacket(stream, + &stream->vsc_infopacket, + &stream->use_vsc_sdp_for_colorimetry); + } + } finish: dc_sink_release(sink); @@ -3657,7 +4194,7 @@ dm_crtc_duplicate_state(struct drm_crtc *crtc) state->abm_level = cur->abm_level; state->vrr_supported = cur->vrr_supported; state->freesync_config = cur->freesync_config; - state->crc_enabled = cur->crc_enabled; + state->crc_src = cur->crc_src; state->cm_has_degamma = cur->cm_has_degamma; state->cm_is_degamma_srgb = cur->cm_is_degamma_srgb; @@ -3673,6 +4210,10 @@ static inline int dm_set_vupdate_irq(struct drm_crtc *crtc, bool enable) struct amdgpu_device *adev = crtc->dev->dev_private; int rc; + /* Do not set vupdate for DCN hardware */ + if (adev->family > AMDGPU_FAMILY_AI) + return 0; + irq_source = IRQ_TYPE_VUPDATE + acrtc->otg_inst; rc = dc_interrupt_set(adev->dm.dc, irq_source, enable) ? 0 : -EBUSY; @@ -3727,6 +4268,7 @@ static const struct drm_crtc_funcs amdgpu_dm_crtc_funcs = { .atomic_destroy_state = dm_crtc_destroy_state, .set_crc_source = amdgpu_dm_crtc_set_crc_source, .verify_crc_source = amdgpu_dm_crtc_verify_crc_source, + .get_crc_sources = amdgpu_dm_crtc_get_crc_sources, .enable_vblank = dm_enable_vblank, .disable_vblank = dm_disable_vblank, }; @@ -3915,7 +4457,8 @@ void amdgpu_dm_connector_funcs_reset(struct drm_connector *connector) state->underscan_hborder = 0; state->underscan_vborder = 0; state->base.max_requested_bpc = 8; - + state->vcpi_slots = 0; + state->pbn = 0; if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) state->abm_level = amdgpu_dm_abm_level; @@ -3943,7 +4486,8 @@ amdgpu_dm_connector_atomic_duplicate_state(struct drm_connector *connector) new_state->underscan_enable = state->underscan_enable; new_state->underscan_hborder = state->underscan_hborder; new_state->underscan_vborder = state->underscan_vborder; - + new_state->vcpi_slots = state->vcpi_slots; + new_state->pbn = state->pbn; return &new_state->base; } @@ -4059,8 +4603,8 @@ enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connec result = MODE_OK; else DRM_DEBUG_KMS("Mode %dx%d (clk %d) failed DC validation with error %d\n", - mode->vdisplay, mode->hdisplay, + mode->vdisplay, mode->clock, dc_result); @@ -4340,10 +4884,69 @@ static void dm_encoder_helper_disable(struct drm_encoder *encoder) } +static int convert_dc_color_depth_into_bpc (enum dc_color_depth display_color_depth) +{ + switch (display_color_depth) { + case COLOR_DEPTH_666: + return 6; + case COLOR_DEPTH_888: + return 8; + case COLOR_DEPTH_101010: + return 10; + case COLOR_DEPTH_121212: + return 12; + case COLOR_DEPTH_141414: + return 14; + case COLOR_DEPTH_161616: + return 16; + default: + break; + } + return 0; +} + static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder, struct drm_crtc_state *crtc_state, struct drm_connector_state *conn_state) { + struct drm_atomic_state *state = crtc_state->state; + struct drm_connector *connector = conn_state->connector; + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); + struct dm_connector_state *dm_new_connector_state = to_dm_connector_state(conn_state); + const struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode; + struct drm_dp_mst_topology_mgr *mst_mgr; + struct drm_dp_mst_port *mst_port; + enum dc_color_depth color_depth; + int clock, bpp = 0; + bool is_y420 = false; + + if (!aconnector->port || !aconnector->dc_sink) + return 0; + + mst_port = aconnector->port; + mst_mgr = &aconnector->mst_port->mst_mgr; + + if (!crtc_state->connectors_changed && !crtc_state->mode_changed) + return 0; + + if (!state->duplicated) { + is_y420 = drm_mode_is_420_also(&connector->display_info, adjusted_mode) && + aconnector->force_yuv420_output; + color_depth = convert_color_depth_from_display_info(connector, conn_state, + is_y420); + bpp = convert_dc_color_depth_into_bpc(color_depth) * 3; + clock = adjusted_mode->clock; + dm_new_connector_state->pbn = drm_dp_calc_pbn_mode(clock, bpp, false); + } + dm_new_connector_state->vcpi_slots = drm_dp_atomic_find_vcpi_slots(state, + mst_mgr, + mst_port, + dm_new_connector_state->pbn, + 0); + if (dm_new_connector_state->vcpi_slots < 0) { + DRM_DEBUG_ATOMIC("failed finding vcpi slots: %d\n", (int)dm_new_connector_state->vcpi_slots); + return dm_new_connector_state->vcpi_slots; + } return 0; } @@ -4352,6 +4955,71 @@ const struct drm_encoder_helper_funcs amdgpu_dm_encoder_helper_funcs = { .atomic_check = dm_encoder_helper_atomic_check }; +#if defined(CONFIG_DRM_AMD_DC_DCN) +static int dm_update_mst_vcpi_slots_for_dsc(struct drm_atomic_state *state, + struct dc_state *dc_state) +{ + struct dc_stream_state *stream = NULL; + struct drm_connector *connector; + struct drm_connector_state *new_con_state, *old_con_state; + struct amdgpu_dm_connector *aconnector; + struct dm_connector_state *dm_conn_state; + int i, j, clock, bpp; + int vcpi, pbn_div, pbn = 0; + + for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) { + + aconnector = to_amdgpu_dm_connector(connector); + + if (!aconnector->port) + continue; + + if (!new_con_state || !new_con_state->crtc) + continue; + + dm_conn_state = to_dm_connector_state(new_con_state); + + for (j = 0; j < dc_state->stream_count; j++) { + stream = dc_state->streams[j]; + if (!stream) + continue; + + if ((struct amdgpu_dm_connector*)stream->dm_stream_context == aconnector) + break; + + stream = NULL; + } + + if (!stream) + continue; + + if (stream->timing.flags.DSC != 1) { + drm_dp_mst_atomic_enable_dsc(state, + aconnector->port, + dm_conn_state->pbn, + 0, + false); + continue; + } + + pbn_div = dm_mst_get_pbn_divider(stream->link); + bpp = stream->timing.dsc_cfg.bits_per_pixel; + clock = stream->timing.pix_clk_100hz / 10; + pbn = drm_dp_calc_pbn_mode(clock, bpp, true); + vcpi = drm_dp_mst_atomic_enable_dsc(state, + aconnector->port, + pbn, pbn_div, + true); + if (vcpi < 0) + return vcpi; + + dm_conn_state->pbn = pbn; + dm_conn_state->vcpi_slots = vcpi; + } + return 0; +} +#endif + static void dm_drm_plane_reset(struct drm_plane *plane) { struct dm_plane_state *amdgpu_state = NULL; @@ -4439,14 +5107,14 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane, tv.num_shared = 1; list_add(&tv.head, &list); - r = ttm_eu_reserve_buffers(&ticket, &list, false, NULL, true); + r = ttm_eu_reserve_buffers(&ticket, &list, false, NULL); if (r) { dev_err(adev->dev, "fail to reserve bo (%d)\n", r); return r; } if (plane->type != DRM_PLANE_TYPE_CURSOR) - domain = amdgpu_display_supported_domains(adev); + domain = amdgpu_display_supported_domains(adev, rbo->flags); else domain = AMDGPU_GEM_DOMAIN_VRAM; @@ -4536,20 +5204,10 @@ static int dm_plane_atomic_check(struct drm_plane *plane, static int dm_plane_atomic_async_check(struct drm_plane *plane, struct drm_plane_state *new_plane_state) { - struct drm_plane_state *old_plane_state = - drm_atomic_get_old_plane_state(new_plane_state->state, plane); - /* Only support async updates on cursor planes. */ if (plane->type != DRM_PLANE_TYPE_CURSOR) return -EINVAL; - /* - * DRM calls prepare_fb and cleanup_fb on new_plane_state for - * async commits so don't allow fb changes. - */ - if (old_plane_state->fb != new_plane_state->fb) - return -EINVAL; - return 0; } @@ -4792,7 +5450,13 @@ static int to_drm_connector_type(enum signal_type st) static struct drm_encoder *amdgpu_dm_connector_to_encoder(struct drm_connector *connector) { - return drm_encoder_find(connector->dev, NULL, connector->encoder_ids[0]); + struct drm_encoder *encoder; + + /* There is only one encoder per connector */ + drm_connector_for_each_possible_encoder(connector, encoder) + return encoder; + + return NULL; } static void amdgpu_dm_get_native_mode(struct drm_connector *connector) @@ -5018,9 +5682,9 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, drm_connector_attach_max_bpc_property(&aconnector->base, 8, 16); - /* This defaults to the max in the range, but we want 8bpc. */ - aconnector->base.state->max_bpc = 8; - aconnector->base.state->max_requested_bpc = 8; + /* This defaults to the max in the range, but we want 8bpc for non-edp. */ + aconnector->base.state->max_bpc = (connector_type == DRM_MODE_CONNECTOR_eDP) ? 16 : 8; + aconnector->base.state->max_requested_bpc = aconnector->base.state->max_bpc; if (connector_type == DRM_MODE_CONNECTOR_eDP && dc_is_dmcu_initialized(adev->dm.dc)) { @@ -5037,6 +5701,10 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, drm_connector_attach_vrr_capable_property( &aconnector->base); +#ifdef CONFIG_DRM_AMD_DC_HDCP + if (adev->asic_type >= CHIP_RAVEN) + drm_connector_attach_content_protection_property(&aconnector->base, true); +#endif } } @@ -5144,11 +5812,12 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm, connector_type = to_drm_connector_type(link->connector_signal); - res = drm_connector_init( + res = drm_connector_init_with_ddc( dm->ddev, &aconnector->base, &amdgpu_dm_connector_funcs, - connector_type); + connector_type, + &i2c->base); if (res) { DRM_ERROR("connector_init failed\n"); @@ -5279,6 +5948,48 @@ is_scaling_state_different(const struct dm_connector_state *dm_state, return false; } +#ifdef CONFIG_DRM_AMD_DC_HDCP +static bool is_content_protection_different(struct drm_connector_state *state, + const struct drm_connector_state *old_state, + const struct drm_connector *connector, struct hdcp_workqueue *hdcp_w) +{ + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); + + if (old_state->hdcp_content_type != state->hdcp_content_type && + state->content_protection != DRM_MODE_CONTENT_PROTECTION_UNDESIRED) { + state->content_protection = DRM_MODE_CONTENT_PROTECTION_DESIRED; + return true; + } + + /* CP is being re enabled, ignore this */ + if (old_state->content_protection == DRM_MODE_CONTENT_PROTECTION_ENABLED && + state->content_protection == DRM_MODE_CONTENT_PROTECTION_DESIRED) { + state->content_protection = DRM_MODE_CONTENT_PROTECTION_ENABLED; + return false; + } + + /* S3 resume case, since old state will always be 0 (UNDESIRED) and the restored state will be ENABLED */ + if (old_state->content_protection == DRM_MODE_CONTENT_PROTECTION_UNDESIRED && + state->content_protection == DRM_MODE_CONTENT_PROTECTION_ENABLED) + state->content_protection = DRM_MODE_CONTENT_PROTECTION_DESIRED; + + /* Check if something is connected/enabled, otherwise we start hdcp but nothing is connected/enabled + * hot-plug, headless s3, dpms + */ + if (state->content_protection == DRM_MODE_CONTENT_PROTECTION_DESIRED && connector->dpms == DRM_MODE_DPMS_ON && + aconnector->dc_sink != NULL) + return true; + + if (old_state->content_protection == state->content_protection) + return false; + + if (state->content_protection == DRM_MODE_CONTENT_PROTECTION_UNDESIRED) + return true; + + return false; +} + +#endif static void remove_stream(struct amdgpu_device *adev, struct amdgpu_crtc *acrtc, struct dc_stream_state *stream) @@ -5620,6 +6331,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, uint32_t target_vblank, last_flip_vblank; bool vrr_active = amdgpu_dm_vrr_active(acrtc_state); bool pflip_present = false; + bool swizzle = true; struct { struct dc_surface_update surface_updates[MAX_SURFACES]; struct dc_plane_info plane_infos[MAX_SURFACES]; @@ -5665,6 +6377,9 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, dc_plane = dm_new_plane_state->dc_state; + if (dc_plane && !dc_plane->tiling_info.gfx9.swizzle) + swizzle = false; + bundle->surface_updates[planes_count].surface = dc_plane; if (new_pcrtc_state->color_mgmt_changed) { bundle->surface_updates[planes_count].gamma = dc_plane->gamma_correction; @@ -5693,11 +6408,11 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, * deadlock during GPU reset when this fence will not signal * but we hold reservation lock for the BO. */ - r = reservation_object_wait_timeout_rcu(abo->tbo.resv, true, + r = dma_resv_wait_timeout_rcu(abo->tbo.base.resv, true, false, msecs_to_jiffies(5000)); if (unlikely(r <= 0)) - DRM_ERROR("Waiting for fences timed out or interrupted!"); + DRM_ERROR("Waiting for fences timed out!"); /* * TODO This might fail and hence better not used, wait @@ -5721,8 +6436,13 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, bundle->surface_updates[planes_count].plane_info = &bundle->plane_infos[planes_count]; + /* + * Only allow immediate flips for fast updates that don't + * change FB pitch, DCC state, rotation or mirroing. + */ bundle->flip_addrs[planes_count].flip_immediate = - (crtc->state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) != 0; + crtc->state->async_flip && + acrtc_state->update_type == UPDATE_TYPE_FAST; timestamp_ns = ktime_get_ns(); bundle->flip_addrs[planes_count].flip_timestamp_in_us = div_u64(timestamp_ns, 1000); @@ -5814,6 +6534,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, /* Update the planes if changed or disable if we don't have any. */ if ((planes_count || acrtc_state->active_planes == 0) && acrtc_state->stream) { + bundle->stream_update.stream = acrtc_state->stream; if (new_pcrtc_state->mode_changed) { bundle->stream_update.src = acrtc_state->stream->src; bundle->stream_update.dst = acrtc_state->stream->dst; @@ -5849,14 +6570,29 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, &acrtc_state->vrr_params.adjust); spin_unlock_irqrestore(&pcrtc->dev->event_lock, flags); } - mutex_lock(&dm->dc_lock); + if ((acrtc_state->update_type > UPDATE_TYPE_FAST) && + acrtc_state->stream->link->psr_allow_active) + amdgpu_dm_psr_disable(acrtc_state->stream); + dc_commit_updates_for_stream(dm->dc, bundle->surface_updates, planes_count, acrtc_state->stream, &bundle->stream_update, dc_state); + + if ((acrtc_state->update_type > UPDATE_TYPE_FAST) && + acrtc_state->stream->psr_version && + !acrtc_state->stream->link->psr_feature_enabled) + amdgpu_dm_link_setup_psr(acrtc_state->stream); + else if ((acrtc_state->update_type == UPDATE_TYPE_FAST) && + acrtc_state->stream->link->psr_feature_enabled && + !acrtc_state->stream->link->psr_allow_active && + swizzle) { + amdgpu_dm_psr_enable(acrtc_state->stream); + } + mutex_unlock(&dm->dc_lock); } @@ -5967,6 +6703,9 @@ static void amdgpu_dm_enable_crtc_interrupts(struct drm_device *dev, struct drm_crtc *crtc; struct drm_crtc_state *old_crtc_state, *new_crtc_state; int i; +#ifdef CONFIG_DEBUG_FS + enum amdgpu_dm_pipe_crc_source source; +#endif for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { @@ -5992,9 +6731,11 @@ static void amdgpu_dm_enable_crtc_interrupts(struct drm_device *dev, #ifdef CONFIG_DEBUG_FS /* The stream has changed so CRC capture needs to re-enabled. */ - if (dm_new_crtc_state->crc_enabled) { - dm_new_crtc_state->crc_enabled = false; - amdgpu_dm_crtc_set_crc_source(crtc, "auto"); + source = dm_new_crtc_state->crc_src; + if (amdgpu_dm_is_valid_crc_source(source)) { + amdgpu_dm_crtc_configure_crc_source( + crtc, dm_new_crtc_state, + dm_new_crtc_state->crc_src); } #endif } @@ -6045,23 +6786,8 @@ static int amdgpu_dm_atomic_commit(struct drm_device *dev, if (dm_old_crtc_state->interrupts_enabled && (!dm_new_crtc_state->interrupts_enabled || - drm_atomic_crtc_needs_modeset(new_crtc_state))) { - /* - * Drop the extra vblank reference added by CRC - * capture if applicable. - */ - if (dm_new_crtc_state->crc_enabled) - drm_crtc_vblank_put(crtc); - - /* - * Only keep CRC capture enabled if there's - * still a stream for the CRTC. - */ - if (!dm_new_crtc_state->stream) - dm_new_crtc_state->crc_enabled = false; - + drm_atomic_crtc_needs_modeset(new_crtc_state))) manage_dm_interrupts(adev, acrtc, false); - } } /* * Add check here for SoC's that support hardware cursor plane, to @@ -6175,10 +6901,13 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) crtc->hwmode = new_crtc_state->mode; } else if (modereset_required(new_crtc_state)) { DRM_DEBUG_DRIVER("Atomic commit: RESET. crtc id %d:[%p]\n", acrtc->crtc_id, acrtc); - /* i.e. reset mode */ - if (dm_old_crtc_state->stream) + if (dm_old_crtc_state->stream) { + if (dm_old_crtc_state->stream->link->psr_allow_active) + amdgpu_dm_psr_disable(dm_old_crtc_state->stream); + remove_stream(adev, acrtc, dm_old_crtc_state->stream); + } } } /* for_each_crtc_in_state() */ @@ -6208,6 +6937,34 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) acrtc->otg_inst = status->primary_otg_inst; } } +#ifdef CONFIG_DRM_AMD_DC_HDCP + for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) { + struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state); + struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc); + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); + + new_crtc_state = NULL; + + if (acrtc) + new_crtc_state = drm_atomic_get_new_crtc_state(state, &acrtc->base); + + dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); + + if (dm_new_crtc_state && dm_new_crtc_state->stream == NULL && + connector->state->content_protection == DRM_MODE_CONTENT_PROTECTION_ENABLED) { + hdcp_reset_display(adev->dm.hdcp_workqueue, aconnector->dc_link->link_index); + new_con_state->content_protection = DRM_MODE_CONTENT_PROTECTION_DESIRED; + continue; + } + + if (is_content_protection_different(new_con_state, old_con_state, connector, adev->dm.hdcp_workqueue)) + hdcp_update_display( + adev->dm.hdcp_workqueue, aconnector->dc_link->link_index, aconnector, + new_con_state->hdcp_content_type, + new_con_state->content_protection == DRM_MODE_CONTENT_PROTECTION_DESIRED ? true + : false); + } +#endif /* Handle connector state changes */ for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) { @@ -6247,9 +7004,10 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) if (!scaling_changed && !abm_changed && !hdr_changed) continue; + stream_update.stream = dm_new_crtc_state->stream; if (scaling_changed) { update_stream_scaling_settings(&dm_new_con_state->base.crtc->mode, - dm_new_con_state, (struct dc_stream_state *)dm_new_crtc_state->stream); + dm_new_con_state, dm_new_crtc_state->stream); stream_update.src = dm_new_crtc_state->stream->src; stream_update.dst = dm_new_crtc_state->stream->dst; @@ -6310,7 +7068,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) amdgpu_dm_enable_crtc_interrupts(dev, state, true); for_each_new_crtc_in_state(state, crtc, new_crtc_state, j) - if (new_crtc_state->pageflip_flags & DRM_MODE_PAGE_FLIP_ASYNC) + if (new_crtc_state->async_flip) wait_for_vblank = false; /* update planes when needed per crtc*/ @@ -6994,31 +7752,34 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, int i, j, num_plane, ret = 0; struct drm_plane_state *old_plane_state, *new_plane_state; struct dm_plane_state *new_dm_plane_state, *old_dm_plane_state; - struct drm_crtc *new_plane_crtc, *old_plane_crtc; + struct drm_crtc *new_plane_crtc; struct drm_plane *plane; struct drm_crtc *crtc; struct drm_crtc_state *new_crtc_state, *old_crtc_state; struct dm_crtc_state *new_dm_crtc_state, *old_dm_crtc_state; struct dc_stream_status *status = NULL; - - struct dc_surface_update *updates; enum surface_update_type update_type = UPDATE_TYPE_FAST; + struct surface_info_bundle { + struct dc_surface_update surface_updates[MAX_SURFACES]; + struct dc_plane_info plane_infos[MAX_SURFACES]; + struct dc_scaling_info scaling_infos[MAX_SURFACES]; + struct dc_flip_addrs flip_addrs[MAX_SURFACES]; + struct dc_stream_update stream_update; + } *bundle; - updates = kcalloc(MAX_SURFACES, sizeof(*updates), GFP_KERNEL); + bundle = kzalloc(sizeof(*bundle), GFP_KERNEL); - if (!updates) { - DRM_ERROR("Failed to allocate plane updates\n"); + if (!bundle) { + DRM_ERROR("Failed to allocate update bundle\n"); /* Set type to FULL to avoid crashing in DC*/ update_type = UPDATE_TYPE_FULL; goto cleanup; } for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { - struct dc_scaling_info scaling_info; - struct dc_stream_update stream_update; - memset(&stream_update, 0, sizeof(stream_update)); + memset(bundle, 0, sizeof(struct surface_info_bundle)); new_dm_crtc_state = to_dm_crtc_state(new_crtc_state); old_dm_crtc_state = to_dm_crtc_state(old_crtc_state); @@ -7033,8 +7794,14 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, continue; for_each_oldnew_plane_in_state(state, plane, old_plane_state, new_plane_state, j) { + const struct amdgpu_framebuffer *amdgpu_fb = + to_amdgpu_framebuffer(new_plane_state->fb); + struct dc_plane_info *plane_info = &bundle->plane_infos[num_plane]; + struct dc_flip_addrs *flip_addr = &bundle->flip_addrs[num_plane]; + struct dc_scaling_info *scaling_info = &bundle->scaling_infos[num_plane]; + uint64_t tiling_flags; + new_plane_crtc = new_plane_state->crtc; - old_plane_crtc = old_plane_state->crtc; new_dm_plane_state = to_dm_plane_state(new_plane_state); old_dm_plane_state = to_dm_plane_state(old_plane_state); @@ -7049,32 +7816,49 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, if (crtc != new_plane_crtc) continue; - updates[num_plane].surface = new_dm_plane_state->dc_state; + bundle->surface_updates[num_plane].surface = + new_dm_plane_state->dc_state; if (new_crtc_state->mode_changed) { - stream_update.dst = new_dm_crtc_state->stream->dst; - stream_update.src = new_dm_crtc_state->stream->src; + bundle->stream_update.dst = new_dm_crtc_state->stream->dst; + bundle->stream_update.src = new_dm_crtc_state->stream->src; } if (new_crtc_state->color_mgmt_changed) { - updates[num_plane].gamma = + bundle->surface_updates[num_plane].gamma = new_dm_plane_state->dc_state->gamma_correction; - updates[num_plane].in_transfer_func = + bundle->surface_updates[num_plane].in_transfer_func = new_dm_plane_state->dc_state->in_transfer_func; - stream_update.gamut_remap = + bundle->stream_update.gamut_remap = &new_dm_crtc_state->stream->gamut_remap_matrix; - stream_update.output_csc_transform = + bundle->stream_update.output_csc_transform = &new_dm_crtc_state->stream->csc_color_matrix; - stream_update.out_transfer_func = + bundle->stream_update.out_transfer_func = new_dm_crtc_state->stream->out_transfer_func; } ret = fill_dc_scaling_info(new_plane_state, - &scaling_info); + scaling_info); if (ret) goto cleanup; - updates[num_plane].scaling_info = &scaling_info; + bundle->surface_updates[num_plane].scaling_info = scaling_info; + + if (amdgpu_fb) { + ret = get_fb_info(amdgpu_fb, &tiling_flags); + if (ret) + goto cleanup; + + ret = fill_dc_plane_info_and_addr( + dm->adev, new_plane_state, tiling_flags, + plane_info, + &flip_addr->address); + if (ret) + goto cleanup; + + bundle->surface_updates[num_plane].plane_info = plane_info; + bundle->surface_updates[num_plane].flip_addr = flip_addr; + } num_plane++; } @@ -7094,14 +7878,15 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, status = dc_stream_get_status_from_state(old_dm_state->context, new_dm_crtc_state->stream); - + bundle->stream_update.stream = new_dm_crtc_state->stream; /* * TODO: DC modifies the surface during this call so we need * to lock here - find a way to do this without locking. */ mutex_lock(&dm->dc_lock); - update_type = dc_check_update_surfaces_for_stream(dc, updates, num_plane, - &stream_update, status); + update_type = dc_check_update_surfaces_for_stream( + dc, bundle->surface_updates, num_plane, + &bundle->stream_update, status); mutex_unlock(&dm->dc_lock); if (update_type > UPDATE_TYPE_MED) { @@ -7111,12 +7896,35 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, } cleanup: - kfree(updates); + kfree(bundle); *out_type = update_type; return ret; } +static int add_affected_mst_dsc_crtcs(struct drm_atomic_state *state, struct drm_crtc *crtc) +{ + struct drm_connector *connector; + struct drm_connector_state *conn_state; + struct amdgpu_dm_connector *aconnector = NULL; + int i; + for_each_new_connector_in_state(state, connector, conn_state, i) { + if (conn_state->crtc != crtc) + continue; + + aconnector = to_amdgpu_dm_connector(connector); + if (!aconnector->port || !aconnector->mst_port) + aconnector = NULL; + else + break; + } + + if (!aconnector) + return 0; + + return drm_dp_mst_add_affected_dsc_crtcs(state, &aconnector->mst_port->mst_mgr); +} + /** * amdgpu_dm_atomic_check() - Atomic check implementation for AMDgpu DM. * @dev: The DRM device @@ -7169,6 +7977,16 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, if (ret) goto fail; + if (adev->asic_type >= CHIP_NAVI10) { + for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { + if (drm_atomic_crtc_needs_modeset(new_crtc_state)) { + ret = add_affected_mst_dsc_crtcs(state, crtc); + if (ret) + goto fail; + } + } + } + for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { if (!drm_atomic_crtc_needs_modeset(new_crtc_state) && !new_crtc_state->color_mgmt_changed && @@ -7272,6 +8090,26 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, if (ret) goto fail; + if (state->legacy_cursor_update) { + /* + * This is a fast cursor update coming from the plane update + * helper, check if it can be done asynchronously for better + * performance. + */ + state->async_update = + !drm_atomic_helper_async_check(dev, state); + + /* + * Skip the remaining global validation if this is an async + * update. Cursor updates can be done without affecting + * state or bandwidth calcs and this avoids the performance + * penalty of locking the private state object and + * allocating a new dc_state. + */ + if (state->async_update) + return 0; + } + /* Check scaling and underscan changes*/ /* TODO Removed scaling changes validation due to inability to commit * new stream into context w\o causing full reset. Need to @@ -7320,17 +8158,60 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, if (ret) goto fail; +#if defined(CONFIG_DRM_AMD_DC_DCN) + if (!compute_mst_dsc_configs_for_state(state, dm_state->context)) + goto fail; + + ret = dm_update_mst_vcpi_slots_for_dsc(state, dm_state->context); + if (ret) + goto fail; +#endif + + /* + * Perform validation of MST topology in the state: + * We need to perform MST atomic check before calling + * dc_validate_global_state(), or there is a chance + * to get stuck in an infinite loop and hang eventually. + */ + ret = drm_dp_mst_atomic_check(state); + if (ret) + goto fail; + if (dc_validate_global_state(dc, dm_state->context, false) != DC_OK) { ret = -EINVAL; goto fail; } - } else if (state->legacy_cursor_update) { + } else { /* - * This is a fast cursor update coming from the plane update - * helper, check if it can be done asynchronously for better - * performance. + * The commit is a fast update. Fast updates shouldn't change + * the DC context, affect global validation, and can have their + * commit work done in parallel with other commits not touching + * the same resource. If we have a new DC context as part of + * the DM atomic state from validation we need to free it and + * retain the existing one instead. */ - state->async_update = !drm_atomic_helper_async_check(dev, state); + struct dm_atomic_state *new_dm_state, *old_dm_state; + + new_dm_state = dm_atomic_get_new_state(state); + old_dm_state = dm_atomic_get_old_state(state); + + if (new_dm_state && old_dm_state) { + if (new_dm_state->context) + dc_release_state(new_dm_state->context); + + new_dm_state->context = old_dm_state->context; + + if (old_dm_state->context) + dc_retain_state(old_dm_state->context); + } + } + + /* Store the overall update type for use later in atomic check. */ + for_each_new_crtc_in_state (state, crtc, new_crtc_state, i) { + struct dm_crtc_state *dm_new_crtc_state = + to_dm_crtc_state(new_crtc_state); + + dm_new_crtc_state->update_type = (int)overall_update_type; } /* Must be success */ @@ -7461,3 +8342,113 @@ update: freesync_capable); } +static void amdgpu_dm_set_psr_caps(struct dc_link *link) +{ + uint8_t dpcd_data[EDP_PSR_RECEIVER_CAP_SIZE]; + + if (!(link->connector_signal & SIGNAL_TYPE_EDP)) + return; + if (link->type == dc_connection_none) + return; + if (dm_helpers_dp_read_dpcd(NULL, link, DP_PSR_SUPPORT, + dpcd_data, sizeof(dpcd_data))) { + link->psr_feature_enabled = dpcd_data[0] ? true:false; + DRM_INFO("PSR support:%d\n", link->psr_feature_enabled); + } +} + +/* + * amdgpu_dm_link_setup_psr() - configure psr link + * @stream: stream state + * + * Return: true if success + */ +static bool amdgpu_dm_link_setup_psr(struct dc_stream_state *stream) +{ + struct dc_link *link = NULL; + struct psr_config psr_config = {0}; + struct psr_context psr_context = {0}; + struct dc *dc = NULL; + bool ret = false; + + if (stream == NULL) + return false; + + link = stream->link; + dc = link->ctx->dc; + + psr_config.psr_version = dc->res_pool->dmcu->dmcu_version.psr_version; + + if (psr_config.psr_version > 0) { + psr_config.psr_exit_link_training_required = 0x1; + psr_config.psr_frame_capture_indication_req = 0; + psr_config.psr_rfb_setup_time = 0x37; + psr_config.psr_sdp_transmit_line_num_deadline = 0x20; + psr_config.allow_smu_optimizations = 0x0; + + ret = dc_link_setup_psr(link, stream, &psr_config, &psr_context); + + } + DRM_DEBUG_DRIVER("PSR link: %d\n", link->psr_feature_enabled); + + return ret; +} + +/* + * amdgpu_dm_psr_enable() - enable psr f/w + * @stream: stream state + * + * Return: true if success + */ +bool amdgpu_dm_psr_enable(struct dc_stream_state *stream) +{ + struct dc_link *link = stream->link; + unsigned int vsync_rate_hz = 0; + struct dc_static_screen_params params = {0}; + /* Calculate number of static frames before generating interrupt to + * enter PSR. + */ + // Init fail safe of 2 frames static + unsigned int num_frames_static = 2; + + DRM_DEBUG_DRIVER("Enabling psr...\n"); + + vsync_rate_hz = div64_u64(div64_u64(( + stream->timing.pix_clk_100hz * 100), + stream->timing.v_total), + stream->timing.h_total); + + /* Round up + * Calculate number of frames such that at least 30 ms of time has + * passed. + */ + if (vsync_rate_hz != 0) { + unsigned int frame_time_microsec = 1000000 / vsync_rate_hz; + num_frames_static = (30000 / frame_time_microsec) + 1; + } + + params.triggers.cursor_update = true; + params.triggers.overlay_update = true; + params.triggers.surface_update = true; + params.num_frames = num_frames_static; + + dc_stream_set_static_screen_params(link->ctx->dc, + &stream, 1, + ¶ms); + + return dc_link_set_psr_allow_active(link, true, false); +} + +/* + * amdgpu_dm_psr_disable() - disable psr f/w + * @stream: stream state + * + * Return: true if success + */ +static bool amdgpu_dm_psr_disable(struct dc_stream_state *stream) +{ + + DRM_DEBUG_DRIVER("Disabling psr...\n"); + + return dc_link_set_psr_allow_active(stream->link, false, true); +} diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index b89cbbfcc0e9..7ea9acb0358d 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -50,12 +50,15 @@ #include "irq_types.h" #include "signal_types.h" +#include "amdgpu_dm_crc.h" /* Forward declarations */ struct amdgpu_device; struct drm_device; struct amdgpu_dm_irq_handler_data; struct dc; +struct amdgpu_bo; +struct dmub_srv; struct common_irq_params { struct amdgpu_device *adev; @@ -107,6 +110,12 @@ struct amdgpu_dm_backlight_caps { * @display_indexes_num: Max number of display streams supported * @irq_handler_list_table_lock: Synchronizes access to IRQ tables * @backlight_dev: Backlight control device + * @backlight_link: Link on which to control backlight + * @backlight_caps: Capabilities of the backlight device + * @freesync_module: Module handling freesync calculations + * @fw_dmcu: Reference to DMCU firmware + * @dmcu_fw_version: Version of the DMCU firmware + * @soc_bounding_box: SOC bounding box values provided by gpu_info FW * @cached_state: Caches device atomic state for suspend/resume * @compressor: Frame buffer compression buffer. See &struct dm_comressor_info */ @@ -115,6 +124,57 @@ struct amdgpu_display_manager { struct dc *dc; /** + * @dmub_srv: + * + * DMUB service, used for controlling the DMUB on hardware + * that supports it. The pointer to the dmub_srv will be + * NULL on hardware that does not support it. + */ + struct dmub_srv *dmub_srv; + + /** + * @dmub_fb_info: + * + * Framebuffer regions for the DMUB. + */ + struct dmub_srv_fb_info *dmub_fb_info; + + /** + * @dmub_fw: + * + * DMUB firmware, required on hardware that has DMUB support. + */ + const struct firmware *dmub_fw; + + /** + * @dmub_bo: + * + * Buffer object for the DMUB. + */ + struct amdgpu_bo *dmub_bo; + + /** + * @dmub_bo_gpu_addr: + * + * GPU virtual address for the DMUB buffer object. + */ + u64 dmub_bo_gpu_addr; + + /** + * @dmub_bo_cpu_addr: + * + * CPU address for the DMUB buffer object. + */ + void *dmub_bo_cpu_addr; + + /** + * @dmcub_fw_version: + * + * DMCUB firmware version. + */ + uint32_t dmcub_fw_version; + + /** * @cgs_device: * * The Common Graphics Services device. It provides an interface for @@ -127,7 +187,7 @@ struct amdgpu_display_manager { u16 display_indexes_num; /** - * @atomic_obj + * @atomic_obj: * * In combination with &dm_atomic_state it helps manage * global atomic state that doesn't map cleanly into existing @@ -224,6 +284,9 @@ struct amdgpu_display_manager { struct amdgpu_dm_backlight_caps backlight_caps; struct mod_freesync *freesync_module; +#ifdef CONFIG_DRM_AMD_DC_HDCP + struct hdcp_workqueue *hdcp_workqueue; +#endif struct drm_atomic_state *cached_state; @@ -231,13 +294,13 @@ struct amdgpu_display_manager { const struct firmware *fw_dmcu; uint32_t dmcu_fw_version; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 /** + * @soc_bounding_box: + * * gpu_info FW provided soc bounding box struct or 0 if not * available in FW */ const struct gpu_info_soc_bounding_box_v1_0 *soc_bounding_box; -#endif }; struct amdgpu_dm_connector { @@ -267,6 +330,7 @@ struct amdgpu_dm_connector { struct drm_dp_mst_port *port; struct amdgpu_dm_connector *mst_port; struct amdgpu_encoder *mst_encoder; + struct drm_dp_aux *dsc_aux; /* TODO see if we can merge with ddc_bus or make a dm_connector */ struct amdgpu_i2c_adapter *i2c; @@ -286,6 +350,7 @@ struct amdgpu_dm_connector { uint32_t debugfs_dpcd_address; uint32_t debugfs_dpcd_size; #endif + bool force_yuv420_output; }; #define to_amdgpu_dm_connector(x) container_of(x, struct amdgpu_dm_connector, base) @@ -309,11 +374,12 @@ struct dm_crtc_state { bool cm_has_degamma; bool cm_is_degamma_srgb; + int update_type; int active_planes; bool interrupts_enabled; int crc_skip_count; - bool crc_enabled; + enum amdgpu_dm_pipe_crc_source crc_src; bool freesync_timing_changed; bool freesync_vrr_info_changed; @@ -345,6 +411,8 @@ struct dm_connector_state { bool underscan_enable; bool freesync_capable; uint8_t abm_level; + int vcpi_slots; + uint64_t pbn; }; #define to_dm_connector_state(x)\ @@ -380,19 +448,6 @@ void dm_restore_drm_connector_state(struct drm_device *dev, void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, struct edid *edid); -/* amdgpu_dm_crc.c */ -#ifdef CONFIG_DEBUG_FS -int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name); -int amdgpu_dm_crtc_verify_crc_source(struct drm_crtc *crtc, - const char *src_name, - size_t *values_cnt); -void amdgpu_dm_crtc_handle_crc_irq(struct drm_crtc *crtc); -#else -#define amdgpu_dm_crtc_set_crc_source NULL -#define amdgpu_dm_crtc_verify_crc_source NULL -#define amdgpu_dm_crtc_handle_crc_irq(x) -#endif - #define MAX_COLOR_LUT_ENTRIES 4096 /* Legacy gamm LUT users such as X doesn't like large LUT sizes */ #define MAX_COLOR_LEGACY_LUT_ENTRIES 256 diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c index b43bb7f90e4e..2233d293a707 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c @@ -210,6 +210,8 @@ static int __set_legacy_tf(struct dc_transfer_func *func, res = mod_color_calculate_regamma_params(func, gamma, true, has_rom, NULL); + dc_gamma_release(&gamma); + return res ? 0 : -ENOMEM; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c index bc67e6502733..eaad9099bc0b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c @@ -30,23 +30,57 @@ #include "amdgpu_dm.h" #include "dc.h" -enum amdgpu_dm_pipe_crc_source { - AMDGPU_DM_PIPE_CRC_SOURCE_NONE = 0, - AMDGPU_DM_PIPE_CRC_SOURCE_AUTO, - AMDGPU_DM_PIPE_CRC_SOURCE_MAX, - AMDGPU_DM_PIPE_CRC_SOURCE_INVALID = -1, +static const char *const pipe_crc_sources[] = { + "none", + "crtc", + "crtc dither", + "dprx", + "dprx dither", + "auto", }; static enum amdgpu_dm_pipe_crc_source dm_parse_crc_source(const char *source) { if (!source || !strcmp(source, "none")) return AMDGPU_DM_PIPE_CRC_SOURCE_NONE; - if (!strcmp(source, "auto")) - return AMDGPU_DM_PIPE_CRC_SOURCE_AUTO; + if (!strcmp(source, "auto") || !strcmp(source, "crtc")) + return AMDGPU_DM_PIPE_CRC_SOURCE_CRTC; + if (!strcmp(source, "dprx")) + return AMDGPU_DM_PIPE_CRC_SOURCE_DPRX; + if (!strcmp(source, "crtc dither")) + return AMDGPU_DM_PIPE_CRC_SOURCE_CRTC_DITHER; + if (!strcmp(source, "dprx dither")) + return AMDGPU_DM_PIPE_CRC_SOURCE_DPRX_DITHER; return AMDGPU_DM_PIPE_CRC_SOURCE_INVALID; } +static bool dm_is_crc_source_crtc(enum amdgpu_dm_pipe_crc_source src) +{ + return (src == AMDGPU_DM_PIPE_CRC_SOURCE_CRTC) || + (src == AMDGPU_DM_PIPE_CRC_SOURCE_CRTC_DITHER); +} + +static bool dm_is_crc_source_dprx(enum amdgpu_dm_pipe_crc_source src) +{ + return (src == AMDGPU_DM_PIPE_CRC_SOURCE_DPRX) || + (src == AMDGPU_DM_PIPE_CRC_SOURCE_DPRX_DITHER); +} + +static bool dm_need_crc_dither(enum amdgpu_dm_pipe_crc_source src) +{ + return (src == AMDGPU_DM_PIPE_CRC_SOURCE_CRTC_DITHER) || + (src == AMDGPU_DM_PIPE_CRC_SOURCE_DPRX_DITHER) || + (src == AMDGPU_DM_PIPE_CRC_SOURCE_NONE); +} + +const char *const *amdgpu_dm_crtc_get_crc_sources(struct drm_crtc *crtc, + size_t *count) +{ + *count = ARRAY_SIZE(pipe_crc_sources); + return pipe_crc_sources; +} + int amdgpu_dm_crtc_verify_crc_source(struct drm_crtc *crtc, const char *src_name, size_t *values_cnt) @@ -63,14 +97,57 @@ amdgpu_dm_crtc_verify_crc_source(struct drm_crtc *crtc, const char *src_name, return 0; } -int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) +int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc, + struct dm_crtc_state *dm_crtc_state, + enum amdgpu_dm_pipe_crc_source source) { struct amdgpu_device *adev = crtc->dev->dev_private; - struct dm_crtc_state *crtc_state = to_dm_crtc_state(crtc->state); - struct dc_stream_state *stream_state = crtc_state->stream; - bool enable; + struct dc_stream_state *stream_state = dm_crtc_state->stream; + bool enable = amdgpu_dm_is_valid_crc_source(source); + int ret = 0; + /* Configuration will be deferred to stream enable. */ + if (!stream_state) + return 0; + + mutex_lock(&adev->dm.dc_lock); + + /* Enable CRTC CRC generation if necessary. */ + if (dm_is_crc_source_crtc(source)) { + if (!dc_stream_configure_crc(stream_state->ctx->dc, + stream_state, enable, enable)) { + ret = -EINVAL; + goto unlock; + } + } + + /* Configure dithering */ + if (!dm_need_crc_dither(source)) { + dc_stream_set_dither_option(stream_state, DITHER_OPTION_TRUN8); + dc_stream_set_dyn_expansion(stream_state->ctx->dc, stream_state, + DYN_EXPANSION_DISABLE); + } else { + dc_stream_set_dither_option(stream_state, + DITHER_OPTION_DEFAULT); + dc_stream_set_dyn_expansion(stream_state->ctx->dc, stream_state, + DYN_EXPANSION_AUTO); + } + +unlock: + mutex_unlock(&adev->dm.dc_lock); + + return ret; +} + +int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) +{ enum amdgpu_dm_pipe_crc_source source = dm_parse_crc_source(src_name); + struct drm_crtc_commit *commit; + struct dm_crtc_state *crtc_state; + struct drm_dp_aux *aux = NULL; + bool enable = false; + bool enabled = false; + int ret = 0; if (source < 0) { DRM_DEBUG_DRIVER("Unknown CRC source %s for CRTC%d\n", @@ -78,41 +155,124 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) return -EINVAL; } - if (!stream_state) { - DRM_ERROR("No stream state for CRTC%d\n", crtc->index); - return -EINVAL; - } + ret = drm_modeset_lock(&crtc->mutex, NULL); + if (ret) + return ret; - enable = (source == AMDGPU_DM_PIPE_CRC_SOURCE_AUTO); + spin_lock(&crtc->commit_lock); + commit = list_first_entry_or_null(&crtc->commit_list, + struct drm_crtc_commit, commit_entry); + if (commit) + drm_crtc_commit_get(commit); + spin_unlock(&crtc->commit_lock); - mutex_lock(&adev->dm.dc_lock); - if (!dc_stream_configure_crc(stream_state->ctx->dc, stream_state, - enable, enable)) { - mutex_unlock(&adev->dm.dc_lock); - return -EINVAL; + if (commit) { + /* + * Need to wait for all outstanding programming to complete + * in commit tail since it can modify CRC related fields and + * hardware state. Since we're holding the CRTC lock we're + * guaranteed that no other commit work can be queued off + * before we modify the state below. + */ + ret = wait_for_completion_interruptible_timeout( + &commit->hw_done, 10 * HZ); + if (ret) + goto cleanup; } - /* When enabling CRC, we should also disable dithering. */ - dc_stream_set_dither_option(stream_state, - enable ? DITHER_OPTION_TRUN8 - : DITHER_OPTION_DEFAULT); + enable = amdgpu_dm_is_valid_crc_source(source); + crtc_state = to_dm_crtc_state(crtc->state); - mutex_unlock(&adev->dm.dc_lock); + /* + * USER REQ SRC | CURRENT SRC | BEHAVIOR + * ----------------------------- + * None | None | Do nothing + * None | CRTC | Disable CRTC CRC, set default to dither + * None | DPRX | Disable DPRX CRC, need 'aux', set default to dither + * None | CRTC DITHER | Disable CRTC CRC + * None | DPRX DITHER | Disable DPRX CRC, need 'aux' + * CRTC | XXXX | Enable CRTC CRC, no dither + * DPRX | XXXX | Enable DPRX CRC, need 'aux', no dither + * CRTC DITHER | XXXX | Enable CRTC CRC, set dither + * DPRX DITHER | XXXX | Enable DPRX CRC, need 'aux', set dither + */ + if (dm_is_crc_source_dprx(source) || + (source == AMDGPU_DM_PIPE_CRC_SOURCE_NONE && + dm_is_crc_source_dprx(crtc_state->crc_src))) { + struct amdgpu_dm_connector *aconn = NULL; + struct drm_connector *connector; + struct drm_connector_list_iter conn_iter; + + drm_connector_list_iter_begin(crtc->dev, &conn_iter); + drm_for_each_connector_iter(connector, &conn_iter) { + if (!connector->state || connector->state->crtc != crtc) + continue; + + aconn = to_amdgpu_dm_connector(connector); + break; + } + drm_connector_list_iter_end(&conn_iter); + + if (!aconn) { + DRM_DEBUG_DRIVER("No amd connector matching CRTC-%d\n", crtc->index); + ret = -EINVAL; + goto cleanup; + } + + aux = &aconn->dm_dp_aux.aux; + + if (!aux) { + DRM_DEBUG_DRIVER("No dp aux for amd connector\n"); + ret = -EINVAL; + goto cleanup; + } + } + + if (amdgpu_dm_crtc_configure_crc_source(crtc, crtc_state, source)) { + ret = -EINVAL; + goto cleanup; + } /* * Reading the CRC requires the vblank interrupt handler to be * enabled. Keep a reference until CRC capture stops. */ - if (!crtc_state->crc_enabled && enable) - drm_crtc_vblank_get(crtc); - else if (crtc_state->crc_enabled && !enable) + enabled = amdgpu_dm_is_valid_crc_source(crtc_state->crc_src); + if (!enabled && enable) { + ret = drm_crtc_vblank_get(crtc); + if (ret) + goto cleanup; + + if (dm_is_crc_source_dprx(source)) { + if (drm_dp_start_crc(aux, crtc)) { + DRM_DEBUG_DRIVER("dp start crc failed\n"); + ret = -EINVAL; + goto cleanup; + } + } + } else if (enabled && !enable) { drm_crtc_vblank_put(crtc); + if (dm_is_crc_source_dprx(source)) { + if (drm_dp_stop_crc(aux)) { + DRM_DEBUG_DRIVER("dp stop crc failed\n"); + ret = -EINVAL; + goto cleanup; + } + } + } - crtc_state->crc_enabled = enable; + crtc_state->crc_src = source; /* Reset crc_skipped on dm state */ crtc_state->crc_skip_count = 0; - return 0; + +cleanup: + if (commit) + drm_crtc_commit_put(commit); + + drm_modeset_unlock(&crtc->mutex); + + return ret; } /** @@ -135,7 +295,7 @@ void amdgpu_dm_crtc_handle_crc_irq(struct drm_crtc *crtc) stream_state = crtc_state->stream; /* Early return if CRC capture is not enabled. */ - if (!crtc_state->crc_enabled) + if (!amdgpu_dm_is_valid_crc_source(crtc_state->crc_src)) return; /* @@ -149,10 +309,12 @@ void amdgpu_dm_crtc_handle_crc_irq(struct drm_crtc *crtc) return; } - if (!dc_stream_get_crc(stream_state->ctx->dc, stream_state, - &crcs[0], &crcs[1], &crcs[2])) - return; + if (dm_is_crc_source_crtc(crtc_state->crc_src)) { + if (!dc_stream_get_crc(stream_state->ctx->dc, stream_state, + &crcs[0], &crcs[1], &crcs[2])) + return; - drm_crtc_add_crc_entry(crtc, true, - drm_crtc_accurate_vblank_count(crtc), crcs); + drm_crtc_add_crc_entry(crtc, true, + drm_crtc_accurate_vblank_count(crtc), crcs); + } } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h new file mode 100644 index 000000000000..f7d731797d3f --- /dev/null +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h @@ -0,0 +1,67 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef AMD_DAL_DEV_AMDGPU_DM_AMDGPU_DM_CRC_H_ +#define AMD_DAL_DEV_AMDGPU_DM_AMDGPU_DM_CRC_H_ + +struct drm_crtc; +struct dm_crtc_state; + +enum amdgpu_dm_pipe_crc_source { + AMDGPU_DM_PIPE_CRC_SOURCE_NONE = 0, + AMDGPU_DM_PIPE_CRC_SOURCE_CRTC, + AMDGPU_DM_PIPE_CRC_SOURCE_CRTC_DITHER, + AMDGPU_DM_PIPE_CRC_SOURCE_DPRX, + AMDGPU_DM_PIPE_CRC_SOURCE_DPRX_DITHER, + AMDGPU_DM_PIPE_CRC_SOURCE_MAX, + AMDGPU_DM_PIPE_CRC_SOURCE_INVALID = -1, +}; + +static inline bool amdgpu_dm_is_valid_crc_source(enum amdgpu_dm_pipe_crc_source source) +{ + return (source > AMDGPU_DM_PIPE_CRC_SOURCE_NONE) && + (source < AMDGPU_DM_PIPE_CRC_SOURCE_MAX); +} + +/* amdgpu_dm_crc.c */ +#ifdef CONFIG_DEBUG_FS +int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc, + struct dm_crtc_state *dm_crtc_state, + enum amdgpu_dm_pipe_crc_source source); +int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name); +int amdgpu_dm_crtc_verify_crc_source(struct drm_crtc *crtc, + const char *src_name, + size_t *values_cnt); +const char *const *amdgpu_dm_crtc_get_crc_sources(struct drm_crtc *crtc, + size_t *count); +void amdgpu_dm_crtc_handle_crc_irq(struct drm_crtc *crtc); +#else +#define amdgpu_dm_crtc_set_crc_source NULL +#define amdgpu_dm_crtc_verify_crc_source NULL +#define amdgpu_dm_crtc_get_crc_sources NULL +#define amdgpu_dm_crtc_handle_crc_irq(x) +#endif + +#endif /* AMD_DAL_DEV_AMDGPU_DM_AMDGPU_DM_CRC_H_ */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 36a1d794b4af..f81d3439ee8c 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -657,6 +657,7 @@ static ssize_t dp_phy_test_pattern_debugfs_write(struct file *f, const char __us dc_link_set_test_pattern( link, test_pattern, + DP_TEST_PATTERN_COLOR_SPACE_RGB, &link_training_settings, custom_pattern, 10); @@ -942,6 +943,52 @@ static const struct { {"aux_dpcd_data", &dp_dpcd_data_debugfs_fops} }; +/* + * Force YUV420 output if available from the given mode + */ +static int force_yuv420_output_set(void *data, u64 val) +{ + struct amdgpu_dm_connector *connector = data; + + connector->force_yuv420_output = (bool)val; + + return 0; +} + +/* + * Check if YUV420 is forced when available from the given mode + */ +static int force_yuv420_output_get(void *data, u64 *val) +{ + struct amdgpu_dm_connector *connector = data; + + *val = connector->force_yuv420_output; + + return 0; +} + +DEFINE_DEBUGFS_ATTRIBUTE(force_yuv420_output_fops, force_yuv420_output_get, + force_yuv420_output_set, "%llu\n"); + +/* + * Read PSR state + */ +static int psr_get(void *data, u64 *val) +{ + struct amdgpu_dm_connector *connector = data; + struct dc_link *link = connector->dc_link; + uint32_t psr_state = 0; + + dc_link_get_psr_state(link, &psr_state); + + *val = psr_state; + + return 0; +} + + +DEFINE_DEBUGFS_ATTRIBUTE(psr_fops, psr_get, NULL, "%llu\n"); + void connector_debugfs_init(struct amdgpu_dm_connector *connector) { int i; @@ -955,6 +1002,12 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector) dp_debugfs_entries[i].fops); } } + if (connector->base.connector_type == DRM_MODE_CONNECTOR_eDP) + debugfs_create_file_unsafe("psr_state", 0444, dir, connector, &psr_fops); + + debugfs_create_file_unsafe("force_yuv420_output", 0644, dir, connector, + &force_yuv420_output_fops); + } /* @@ -1053,9 +1106,33 @@ static int target_backlight_read(struct seq_file *m, void *data) return 0; } +static int mst_topo(struct seq_file *m, void *unused) +{ + struct drm_info_node *node = (struct drm_info_node *)m->private; + struct drm_device *dev = node->minor->dev; + struct drm_connector *connector; + struct drm_connector_list_iter conn_iter; + struct amdgpu_dm_connector *aconnector; + + drm_connector_list_iter_begin(dev, &conn_iter); + drm_for_each_connector_iter(connector, &conn_iter) { + if (connector->connector_type != DRM_MODE_CONNECTOR_DisplayPort) + continue; + + aconnector = to_amdgpu_dm_connector(connector); + + seq_printf(m, "\nMST topology for connector %d\n", aconnector->connector_id); + drm_dp_mst_dump_topology(m, &aconnector->mst_mgr); + } + drm_connector_list_iter_end(&conn_iter); + + return 0; +} + static const struct drm_info_list amdgpu_dm_debugfs_list[] = { {"amdgpu_current_backlight_pwm", ¤t_backlight_read}, {"amdgpu_target_backlight_pwm", &target_backlight_read}, + {"amdgpu_mst_topology", &mst_topo}, }; /* diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c new file mode 100644 index 000000000000..0acd3409dd6c --- /dev/null +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c @@ -0,0 +1,386 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "amdgpu_dm_hdcp.h" +#include "amdgpu.h" +#include "amdgpu_dm.h" +#include "dm_helpers.h" +#include <drm/drm_hdcp.h> + +static bool +lp_write_i2c(void *handle, uint32_t address, const uint8_t *data, uint32_t size) +{ + + struct dc_link *link = handle; + struct i2c_payload i2c_payloads[] = {{true, address, size, (void *)data} }; + struct i2c_command cmd = {i2c_payloads, 1, I2C_COMMAND_ENGINE_HW, link->dc->caps.i2c_speed_in_khz}; + + return dm_helpers_submit_i2c(link->ctx, link, &cmd); +} + +static bool +lp_read_i2c(void *handle, uint32_t address, uint8_t offset, uint8_t *data, uint32_t size) +{ + struct dc_link *link = handle; + + struct i2c_payload i2c_payloads[] = {{true, address, 1, &offset}, {false, address, size, data} }; + struct i2c_command cmd = {i2c_payloads, 2, I2C_COMMAND_ENGINE_HW, link->dc->caps.i2c_speed_in_khz}; + + return dm_helpers_submit_i2c(link->ctx, link, &cmd); +} + +static bool +lp_write_dpcd(void *handle, uint32_t address, const uint8_t *data, uint32_t size) +{ + struct dc_link *link = handle; + + return dm_helpers_dp_write_dpcd(link->ctx, link, address, data, size); +} + +static bool +lp_read_dpcd(void *handle, uint32_t address, uint8_t *data, uint32_t size) +{ + struct dc_link *link = handle; + + return dm_helpers_dp_read_dpcd(link->ctx, link, address, data, size); +} + +static void process_output(struct hdcp_workqueue *hdcp_work) +{ + struct mod_hdcp_output output = hdcp_work->output; + + if (output.callback_stop) + cancel_delayed_work(&hdcp_work->callback_dwork); + + if (output.callback_needed) + schedule_delayed_work(&hdcp_work->callback_dwork, + msecs_to_jiffies(output.callback_delay)); + + if (output.watchdog_timer_stop) + cancel_delayed_work(&hdcp_work->watchdog_timer_dwork); + + if (output.watchdog_timer_needed) + schedule_delayed_work(&hdcp_work->watchdog_timer_dwork, + msecs_to_jiffies(output.watchdog_timer_delay)); + + schedule_delayed_work(&hdcp_work->property_validate_dwork, msecs_to_jiffies(0)); +} + +void hdcp_update_display(struct hdcp_workqueue *hdcp_work, + unsigned int link_index, + struct amdgpu_dm_connector *aconnector, + uint8_t content_type, + bool enable_encryption) +{ + struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index]; + struct mod_hdcp_display *display = &hdcp_work[link_index].display; + struct mod_hdcp_link *link = &hdcp_work[link_index].link; + struct mod_hdcp_display_query query; + + mutex_lock(&hdcp_w->mutex); + hdcp_w->aconnector = aconnector; + + query.display = NULL; + mod_hdcp_query_display(&hdcp_w->hdcp, aconnector->base.index, &query); + + if (query.display != NULL) { + memcpy(display, query.display, sizeof(struct mod_hdcp_display)); + mod_hdcp_remove_display(&hdcp_w->hdcp, aconnector->base.index, &hdcp_w->output); + + hdcp_w->link.adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_0; + + if (enable_encryption) { + display->adjust.disable = 0; + if (content_type == DRM_MODE_HDCP_CONTENT_TYPE0) + hdcp_w->link.adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_0; + else if (content_type == DRM_MODE_HDCP_CONTENT_TYPE1) + hdcp_w->link.adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_1; + + schedule_delayed_work(&hdcp_w->property_validate_dwork, + msecs_to_jiffies(DRM_HDCP_CHECK_PERIOD_MS)); + } else { + display->adjust.disable = 1; + hdcp_w->encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + cancel_delayed_work(&hdcp_w->property_validate_dwork); + } + + display->state = MOD_HDCP_DISPLAY_ACTIVE; + } + + mod_hdcp_add_display(&hdcp_w->hdcp, link, display, &hdcp_w->output); + + process_output(hdcp_w); + mutex_unlock(&hdcp_w->mutex); +} + +static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work, + unsigned int link_index, + struct amdgpu_dm_connector *aconnector) +{ + struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index]; + + mutex_lock(&hdcp_w->mutex); + hdcp_w->aconnector = aconnector; + + mod_hdcp_remove_display(&hdcp_w->hdcp, aconnector->base.index, &hdcp_w->output); + + process_output(hdcp_w); + mutex_unlock(&hdcp_w->mutex); +} +void hdcp_reset_display(struct hdcp_workqueue *hdcp_work, unsigned int link_index) +{ + struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index]; + + mutex_lock(&hdcp_w->mutex); + + mod_hdcp_reset_connection(&hdcp_w->hdcp, &hdcp_w->output); + + cancel_delayed_work(&hdcp_w->property_validate_dwork); + hdcp_w->encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + + process_output(hdcp_w); + + mutex_unlock(&hdcp_w->mutex); +} + +void hdcp_handle_cpirq(struct hdcp_workqueue *hdcp_work, unsigned int link_index) +{ + struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index]; + + schedule_work(&hdcp_w->cpirq_work); +} + + + + +static void event_callback(struct work_struct *work) +{ + struct hdcp_workqueue *hdcp_work; + + hdcp_work = container_of(to_delayed_work(work), struct hdcp_workqueue, + callback_dwork); + + mutex_lock(&hdcp_work->mutex); + + cancel_delayed_work(&hdcp_work->watchdog_timer_dwork); + + mod_hdcp_process_event(&hdcp_work->hdcp, MOD_HDCP_EVENT_CALLBACK, + &hdcp_work->output); + + process_output(hdcp_work); + + mutex_unlock(&hdcp_work->mutex); + + +} +static void event_property_update(struct work_struct *work) +{ + + struct hdcp_workqueue *hdcp_work = container_of(work, struct hdcp_workqueue, property_update_work); + struct amdgpu_dm_connector *aconnector = hdcp_work->aconnector; + struct drm_device *dev = hdcp_work->aconnector->base.dev; + long ret; + + drm_modeset_lock(&dev->mode_config.connection_mutex, NULL); + mutex_lock(&hdcp_work->mutex); + + + if (aconnector->base.state->commit) { + ret = wait_for_completion_interruptible_timeout(&aconnector->base.state->commit->hw_done, 10 * HZ); + + if (ret == 0) { + DRM_ERROR("HDCP state unknown! Setting it to DESIRED"); + hdcp_work->encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + } + } + + if (hdcp_work->encryption_status != MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF) { + if (aconnector->base.state->hdcp_content_type == DRM_MODE_HDCP_CONTENT_TYPE0 && + hdcp_work->encryption_status <= MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE0_ON) + drm_hdcp_update_content_protection(&aconnector->base, DRM_MODE_CONTENT_PROTECTION_ENABLED); + else if (aconnector->base.state->hdcp_content_type == DRM_MODE_HDCP_CONTENT_TYPE1 && + hdcp_work->encryption_status == MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE1_ON) + drm_hdcp_update_content_protection(&aconnector->base, DRM_MODE_CONTENT_PROTECTION_ENABLED); + } else { + drm_hdcp_update_content_protection(&aconnector->base, DRM_MODE_CONTENT_PROTECTION_DESIRED); + } + + + mutex_unlock(&hdcp_work->mutex); + drm_modeset_unlock(&dev->mode_config.connection_mutex); +} + +static void event_property_validate(struct work_struct *work) +{ + struct hdcp_workqueue *hdcp_work = + container_of(to_delayed_work(work), struct hdcp_workqueue, property_validate_dwork); + struct mod_hdcp_display_query query; + struct amdgpu_dm_connector *aconnector = hdcp_work->aconnector; + + if (!aconnector) + return; + + mutex_lock(&hdcp_work->mutex); + + query.encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + mod_hdcp_query_display(&hdcp_work->hdcp, aconnector->base.index, &query); + + if (query.encryption_status != hdcp_work->encryption_status) { + hdcp_work->encryption_status = query.encryption_status; + schedule_work(&hdcp_work->property_update_work); + } + + mutex_unlock(&hdcp_work->mutex); +} + +static void event_watchdog_timer(struct work_struct *work) +{ + struct hdcp_workqueue *hdcp_work; + + hdcp_work = container_of(to_delayed_work(work), + struct hdcp_workqueue, + watchdog_timer_dwork); + + mutex_lock(&hdcp_work->mutex); + + mod_hdcp_process_event(&hdcp_work->hdcp, + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT, + &hdcp_work->output); + + process_output(hdcp_work); + + mutex_unlock(&hdcp_work->mutex); + +} + +static void event_cpirq(struct work_struct *work) +{ + struct hdcp_workqueue *hdcp_work; + + hdcp_work = container_of(work, struct hdcp_workqueue, cpirq_work); + + mutex_lock(&hdcp_work->mutex); + + mod_hdcp_process_event(&hdcp_work->hdcp, MOD_HDCP_EVENT_CPIRQ, &hdcp_work->output); + + process_output(hdcp_work); + + mutex_unlock(&hdcp_work->mutex); + +} + + +void hdcp_destroy(struct hdcp_workqueue *hdcp_work) +{ + int i = 0; + + for (i = 0; i < hdcp_work->max_link; i++) { + cancel_delayed_work_sync(&hdcp_work[i].callback_dwork); + cancel_delayed_work_sync(&hdcp_work[i].watchdog_timer_dwork); + } + + kfree(hdcp_work); + +} + +static void update_config(void *handle, struct cp_psp_stream_config *config) +{ + struct hdcp_workqueue *hdcp_work = handle; + struct amdgpu_dm_connector *aconnector = config->dm_stream_ctx; + int link_index = aconnector->dc_link->link_index; + struct mod_hdcp_display *display = &hdcp_work[link_index].display; + struct mod_hdcp_link *link = &hdcp_work[link_index].link; + + memset(display, 0, sizeof(*display)); + memset(link, 0, sizeof(*link)); + + display->index = aconnector->base.index; + + if (config->dpms_off) { + hdcp_remove_display(hdcp_work, link_index, aconnector); + return; + } + display->state = MOD_HDCP_DISPLAY_ACTIVE; + + if (aconnector->dc_sink != NULL) + link->mode = mod_hdcp_signal_type_to_operation_mode(aconnector->dc_sink->sink_signal); + + display->controller = CONTROLLER_ID_D0 + config->otg_inst; + display->dig_fe = config->stream_enc_inst; + link->dig_be = config->link_enc_inst; + link->ddc_line = aconnector->dc_link->ddc_hw_inst + 1; + link->dp.rev = aconnector->dc_link->dpcd_caps.dpcd_rev.raw; + display->adjust.disable = 1; + link->adjust.auth_delay = 2; + + hdcp_update_display(hdcp_work, link_index, aconnector, DRM_MODE_HDCP_CONTENT_TYPE0, false); +} + +struct hdcp_workqueue *hdcp_create_workqueue(void *psp_context, struct cp_psp *cp_psp, struct dc *dc) +{ + + int max_caps = dc->caps.max_links; + struct hdcp_workqueue *hdcp_work = kzalloc(max_caps*sizeof(*hdcp_work), GFP_KERNEL); + int i = 0; + + if (hdcp_work == NULL) + goto fail_alloc_context; + + hdcp_work->max_link = max_caps; + + for (i = 0; i < max_caps; i++) { + + mutex_init(&hdcp_work[i].mutex); + + INIT_WORK(&hdcp_work[i].cpirq_work, event_cpirq); + INIT_WORK(&hdcp_work[i].property_update_work, event_property_update); + INIT_DELAYED_WORK(&hdcp_work[i].callback_dwork, event_callback); + INIT_DELAYED_WORK(&hdcp_work[i].watchdog_timer_dwork, event_watchdog_timer); + INIT_DELAYED_WORK(&hdcp_work[i].property_validate_dwork, event_property_validate); + + hdcp_work[i].hdcp.config.psp.handle = psp_context; + hdcp_work[i].hdcp.config.ddc.handle = dc_get_link_at_index(dc, i); + hdcp_work[i].hdcp.config.ddc.funcs.write_i2c = lp_write_i2c; + hdcp_work[i].hdcp.config.ddc.funcs.read_i2c = lp_read_i2c; + hdcp_work[i].hdcp.config.ddc.funcs.write_dpcd = lp_write_dpcd; + hdcp_work[i].hdcp.config.ddc.funcs.read_dpcd = lp_read_dpcd; + } + + cp_psp->funcs.update_stream_config = update_config; + cp_psp->handle = hdcp_work; + + return hdcp_work; + +fail_alloc_context: + kfree(hdcp_work); + + return NULL; + + + +} + + + diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h new file mode 100644 index 000000000000..6abde86bce4a --- /dev/null +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h @@ -0,0 +1,69 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef AMDGPU_DM_AMDGPU_DM_HDCP_H_ +#define AMDGPU_DM_AMDGPU_DM_HDCP_H_ + +#include "mod_hdcp.h" +#include "hdcp.h" +#include "dc.h" +#include "dm_cp_psp.h" + +struct mod_hdcp; +struct mod_hdcp_link; +struct mod_hdcp_display; +struct cp_psp; + +struct hdcp_workqueue { + struct work_struct cpirq_work; + struct work_struct property_update_work; + struct delayed_work callback_dwork; + struct delayed_work watchdog_timer_dwork; + struct delayed_work property_validate_dwork; + struct amdgpu_dm_connector *aconnector; + struct mutex mutex; + + struct mod_hdcp hdcp; + struct mod_hdcp_output output; + struct mod_hdcp_display display; + struct mod_hdcp_link link; + + enum mod_hdcp_encryption_status encryption_status; + uint8_t max_link; +}; + +void hdcp_update_display(struct hdcp_workqueue *hdcp_work, + unsigned int link_index, + struct amdgpu_dm_connector *aconnector, + uint8_t content_type, + bool enable_encryption); + +void hdcp_reset_display(struct hdcp_workqueue *work, unsigned int link_index); +void hdcp_handle_cpirq(struct hdcp_workqueue *work, unsigned int link_index); +void hdcp_destroy(struct hdcp_workqueue *work); + +struct hdcp_workqueue *hdcp_create_workqueue(void *psp_context, struct cp_psp *cp_psp, struct dc *dc); + +#endif /* AMDGPU_DM_AMDGPU_DM_HDCP_H_ */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index a0ed0154a9f0..318b474ff20e 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -37,6 +37,7 @@ #include "dc.h" #include "amdgpu_dm.h" #include "amdgpu_dm_irq.h" +#include "amdgpu_dm_mst_types.h" #include "dm_helpers.h" @@ -97,11 +98,8 @@ enum dc_edid_status dm_helpers_parse_edid_caps( (struct edid *) edid->raw_edid); sad_count = drm_edid_to_sad((struct edid *) edid->raw_edid, &sads); - if (sad_count <= 0) { - DRM_INFO("SADs count is: %d, don't need to read it\n", - sad_count); + if (sad_count <= 0) return result; - } edid_caps->audio_mode_count = sad_count < DC_MAX_AUDIO_DESC_COUNT ? sad_count : DC_MAX_AUDIO_DESC_COUNT; for (i = 0; i < edid_caps->audio_mode_count; ++i) { @@ -183,19 +181,22 @@ bool dm_helpers_dp_mst_write_payload_allocation_table( bool enable) { struct amdgpu_dm_connector *aconnector; + struct dm_connector_state *dm_conn_state; struct drm_dp_mst_topology_mgr *mst_mgr; struct drm_dp_mst_port *mst_port; - int slots = 0; bool ret; - int clock; - int bpp = 0; - int pbn = 0; aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context; + /* Accessing the connector state is required for vcpi_slots allocation + * and directly relies on behaviour in commit check + * that blocks before commit guaranteeing that the state + * is not gonna be swapped while still in use in commit tail */ if (!aconnector || !aconnector->mst_port) return false; + dm_conn_state = to_dm_connector_state(aconnector->base.state); + mst_mgr = &aconnector->mst_port->mst_mgr; if (!mst_mgr->mst_state) @@ -204,42 +205,10 @@ bool dm_helpers_dp_mst_write_payload_allocation_table( mst_port = aconnector->port; if (enable) { - clock = stream->timing.pix_clk_100hz / 10; - - switch (stream->timing.display_color_depth) { - - case COLOR_DEPTH_666: - bpp = 6; - break; - case COLOR_DEPTH_888: - bpp = 8; - break; - case COLOR_DEPTH_101010: - bpp = 10; - break; - case COLOR_DEPTH_121212: - bpp = 12; - break; - case COLOR_DEPTH_141414: - bpp = 14; - break; - case COLOR_DEPTH_161616: - bpp = 16; - break; - default: - ASSERT(bpp != 0); - break; - } - - bpp = bpp * 3; - - /* TODO need to know link rate */ - - pbn = drm_dp_calc_pbn_mode(clock, bpp); - - slots = drm_dp_find_vcpi_slots(mst_mgr, pbn); - ret = drm_dp_mst_allocate_vcpi(mst_mgr, mst_port, pbn, slots); + ret = drm_dp_mst_allocate_vcpi(mst_mgr, mst_port, + dm_conn_state->pbn, + dm_conn_state->vcpi_slots); if (!ret) return false; @@ -247,7 +216,8 @@ bool dm_helpers_dp_mst_write_payload_allocation_table( drm_dp_mst_reset_vcpi_slots(mst_mgr, mst_port); } - ret = drm_dp_update_payload_part1(mst_mgr); + /* It's OK for this to fail */ + drm_dp_update_payload_part1(mst_mgr); /* mst_mgr->->payloads are VC payload notify MST branch using DPCD or * AUX message. The sequence is slot 1-63 allocated sequence for each @@ -256,9 +226,6 @@ bool dm_helpers_dp_mst_write_payload_allocation_table( get_payload_table(aconnector, proposed_table); - if (ret) - return false; - return true; } @@ -282,7 +249,7 @@ void dm_helpers_dp_mst_clear_payload_allocation_table( * Polls for ACT (allocation change trigger) handled and sends * ALLOCATE_PAYLOAD message. */ -bool dm_helpers_dp_mst_poll_for_allocation_change_trigger( +enum act_return_status dm_helpers_dp_mst_poll_for_allocation_change_trigger( struct dc_context *ctx, const struct dc_stream_state *stream) { @@ -293,19 +260,19 @@ bool dm_helpers_dp_mst_poll_for_allocation_change_trigger( aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context; if (!aconnector || !aconnector->mst_port) - return false; + return ACT_FAILED; mst_mgr = &aconnector->mst_port->mst_mgr; if (!mst_mgr->mst_state) - return false; + return ACT_FAILED; ret = drm_dp_check_act_status(mst_mgr); if (ret) - return false; + return ACT_FAILED; - return true; + return ACT_SUCCESS; } bool dm_helpers_dp_mst_send_payload_allocation( @@ -316,7 +283,6 @@ bool dm_helpers_dp_mst_send_payload_allocation( struct amdgpu_dm_connector *aconnector; struct drm_dp_mst_topology_mgr *mst_mgr; struct drm_dp_mst_port *mst_port; - int ret; aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context; @@ -330,10 +296,8 @@ bool dm_helpers_dp_mst_send_payload_allocation( if (!mst_mgr->mst_state) return false; - ret = drm_dp_update_payload_part2(mst_mgr); - - if (ret) - return false; + /* It's OK for this to fail */ + drm_dp_update_payload_part2(mst_mgr); if (!enable) drm_dp_mst_deallocate_vcpi(mst_mgr, mst_port); @@ -541,16 +505,32 @@ bool dm_helpers_submit_i2c( return result; } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT bool dm_helpers_dp_write_dsc_enable( struct dc_context *ctx, const struct dc_stream_state *stream, bool enable ) { + uint8_t enable_dsc = enable ? 1 : 0; + struct amdgpu_dm_connector *aconnector; + + if (!stream) + return false; + + if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context; + + if (!aconnector->dsc_aux) + return false; + + return (drm_dp_dpcd_write(aconnector->dsc_aux, DP_DSC_ENABLE, &enable_dsc, 1) >= 0); + } + + if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT) + return dm_helpers_dp_write_dpcd(ctx, stream->link, DP_DSC_ENABLE, &enable_dsc, 1); + return false; } -#endif bool dm_helpers_is_dp_sink_present(struct dc_link *link) { diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c index fa5d503d379c..cbcf504f73a5 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c @@ -111,17 +111,12 @@ static void init_handler_common_data(struct amdgpu_dm_irq_handler_data *hcd, */ static void dm_irq_work_func(struct work_struct *work) { - struct list_head *entry; struct irq_list_head *irq_list_head = container_of(work, struct irq_list_head, work); struct list_head *handler_list = &irq_list_head->head; struct amdgpu_dm_irq_handler_data *handler_data; - list_for_each(entry, handler_list) { - handler_data = list_entry(entry, - struct amdgpu_dm_irq_handler_data, - list); - + list_for_each_entry(handler_data, handler_list, list) { DRM_DEBUG_KMS("DM_IRQ: work_func: for dal_src=%d\n", handler_data->irq_source); @@ -528,19 +523,13 @@ static void amdgpu_dm_irq_immediate_work(struct amdgpu_device *adev, enum dc_irq_source irq_source) { struct amdgpu_dm_irq_handler_data *handler_data; - struct list_head *entry; unsigned long irq_table_flags; DM_IRQ_TABLE_LOCK(adev, irq_table_flags); - list_for_each( - entry, - &adev->dm.irq_handler_list_high_tab[irq_source]) { - - handler_data = list_entry(entry, - struct amdgpu_dm_irq_handler_data, - list); - + list_for_each_entry(handler_data, + &adev->dm.irq_handler_list_high_tab[irq_source], + list) { /* Call a subcomponent which registered for immediate * interrupt notification */ handler_data->handler(handler_data->handler_arg); @@ -732,8 +721,10 @@ void amdgpu_dm_hpd_init(struct amdgpu_device *adev) { struct drm_device *dev = adev->ddev; struct drm_connector *connector; + struct drm_connector_list_iter iter; - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + drm_connector_list_iter_begin(dev, &iter); + drm_for_each_connector_iter(connector, &iter) { struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); @@ -751,6 +742,7 @@ void amdgpu_dm_hpd_init(struct amdgpu_device *adev) true); } } + drm_connector_list_iter_end(&iter); } /** @@ -765,8 +757,10 @@ void amdgpu_dm_hpd_fini(struct amdgpu_device *adev) { struct drm_device *dev = adev->ddev; struct drm_connector *connector; + struct drm_connector_list_iter iter; - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + drm_connector_list_iter_begin(dev, &iter); + drm_for_each_connector_iter(connector, &iter) { struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); const struct dc_link *dc_link = amdgpu_dm_connector->dc_link; @@ -779,4 +773,5 @@ void amdgpu_dm_hpd_fini(struct amdgpu_device *adev) false); } } + drm_connector_list_iter_end(&iter); } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 6e205ee36ac3..5672f7765919 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -25,6 +25,7 @@ #include <linux/version.h> #include <drm/drm_atomic_helper.h> +#include <drm/drm_dp_mst_helper.h> #include "dm_services.h" #include "amdgpu.h" #include "amdgpu_dm.h" @@ -36,6 +37,14 @@ #include "dc_link_ddc.h" #include "i2caux_interface.h" +#if defined(CONFIG_DEBUG_FS) +#include "amdgpu_dm_debugfs.h" +#endif + + +#if defined(CONFIG_DRM_AMD_DC_DCN) +#include "dc/dcn20/dcn20_resource.h" +#endif /* #define TRACE_DPCD */ @@ -113,6 +122,7 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, result = -EIO; break; case AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY: + case AUX_CHANNEL_OPERATION_FAILED_ENGINE_ACQUIRE: result = -EBUSY; break; case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT: @@ -123,31 +133,14 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, return result; } -static enum drm_connector_status -dm_dp_mst_detect(struct drm_connector *connector, bool force) -{ - struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); - struct amdgpu_dm_connector *master = aconnector->mst_port; - - enum drm_connector_status status = - drm_dp_mst_detect_port( - connector, - &master->mst_mgr, - aconnector->port); - - return status; -} - static void dm_dp_mst_connector_destroy(struct drm_connector *connector) { struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); struct amdgpu_encoder *amdgpu_encoder = amdgpu_dm_connector->mst_encoder; - if (amdgpu_dm_connector->edid) { - kfree(amdgpu_dm_connector->edid); - amdgpu_dm_connector->edid = NULL; - } + kfree(amdgpu_dm_connector->edid); + amdgpu_dm_connector->edid = NULL; drm_encoder_cleanup(&amdgpu_encoder->base); kfree(amdgpu_encoder); @@ -156,17 +149,68 @@ dm_dp_mst_connector_destroy(struct drm_connector *connector) kfree(amdgpu_dm_connector); } +static int +amdgpu_dm_mst_connector_late_register(struct drm_connector *connector) +{ + struct amdgpu_dm_connector *amdgpu_dm_connector = + to_amdgpu_dm_connector(connector); + struct drm_dp_mst_port *port = amdgpu_dm_connector->port; + +#if defined(CONFIG_DEBUG_FS) + connector_debugfs_init(amdgpu_dm_connector); + amdgpu_dm_connector->debugfs_dpcd_address = 0; + amdgpu_dm_connector->debugfs_dpcd_size = 0; +#endif + + return drm_dp_mst_connector_late_register(connector, port); +} + +static void +amdgpu_dm_mst_connector_early_unregister(struct drm_connector *connector) +{ + struct amdgpu_dm_connector *amdgpu_dm_connector = + to_amdgpu_dm_connector(connector); + struct drm_dp_mst_port *port = amdgpu_dm_connector->port; + + drm_dp_mst_connector_early_unregister(connector, port); +} + static const struct drm_connector_funcs dm_dp_mst_connector_funcs = { - .detect = dm_dp_mst_detect, .fill_modes = drm_helper_probe_single_connector_modes, .destroy = dm_dp_mst_connector_destroy, .reset = amdgpu_dm_connector_funcs_reset, .atomic_duplicate_state = amdgpu_dm_connector_atomic_duplicate_state, .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, .atomic_set_property = amdgpu_dm_connector_atomic_set_property, - .atomic_get_property = amdgpu_dm_connector_atomic_get_property + .atomic_get_property = amdgpu_dm_connector_atomic_get_property, + .late_register = amdgpu_dm_mst_connector_late_register, + .early_unregister = amdgpu_dm_mst_connector_early_unregister, }; +#if defined(CONFIG_DRM_AMD_DC_DCN) +static bool validate_dsc_caps_on_connector(struct amdgpu_dm_connector *aconnector) +{ + struct dc_sink *dc_sink = aconnector->dc_sink; + struct drm_dp_mst_port *port = aconnector->port; + u8 dsc_caps[16] = { 0 }; + + aconnector->dsc_aux = drm_dp_mst_dsc_aux_for_port(port); + + if (!aconnector->dsc_aux) + return false; + + if (drm_dp_dpcd_read(aconnector->dsc_aux, DP_DSC_SUPPORT, dsc_caps, 16) < 0) + return false; + + if (!dc_dsc_parse_dsc_dpcd(aconnector->dc_link->ctx->dc, + dsc_caps, NULL, + &dc_sink->sink_dsc_caps.dsc_dec_caps)) + return false; + + return true; +} +#endif + static int dm_dp_mst_get_modes(struct drm_connector *connector) { struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); @@ -209,10 +253,16 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) /* dc_link_add_remote_sink returns a new reference */ aconnector->dc_sink = dc_sink; - if (aconnector->dc_sink) + if (aconnector->dc_sink) { amdgpu_dm_update_freesync_caps( connector, aconnector->edid); +#if defined(CONFIG_DRM_AMD_DC_DCN) + if (!validate_dsc_caps_on_connector(aconnector)) + memset(&aconnector->dc_sink->sink_dsc_caps, + 0, sizeof(aconnector->dc_sink->sink_dsc_caps)); +#endif + } } drm_connector_update_edid_property( @@ -223,17 +273,61 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) return ret; } -static struct drm_encoder *dm_mst_best_encoder(struct drm_connector *connector) +static struct drm_encoder * +dm_mst_atomic_best_encoder(struct drm_connector *connector, + struct drm_connector_state *connector_state) { - struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); + return &to_amdgpu_dm_connector(connector)->mst_encoder->base; +} - return &amdgpu_dm_connector->mst_encoder->base; +static int +dm_dp_mst_detect(struct drm_connector *connector, + struct drm_modeset_acquire_ctx *ctx, bool force) +{ + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); + struct amdgpu_dm_connector *master = aconnector->mst_port; + + return drm_dp_mst_detect_port(connector, ctx, &master->mst_mgr, + aconnector->port); +} + +static int dm_dp_mst_atomic_check(struct drm_connector *connector, + struct drm_atomic_state *state) +{ + struct drm_connector_state *new_conn_state = + drm_atomic_get_new_connector_state(state, connector); + struct drm_connector_state *old_conn_state = + drm_atomic_get_old_connector_state(state, connector); + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); + struct drm_crtc_state *new_crtc_state; + struct drm_dp_mst_topology_mgr *mst_mgr; + struct drm_dp_mst_port *mst_port; + + mst_port = aconnector->port; + mst_mgr = &aconnector->mst_port->mst_mgr; + + if (!old_conn_state->crtc) + return 0; + + if (new_conn_state->crtc) { + new_crtc_state = drm_atomic_get_new_crtc_state(state, new_conn_state->crtc); + if (!new_crtc_state || + !drm_atomic_crtc_needs_modeset(new_crtc_state) || + new_crtc_state->enable) + return 0; + } + + return drm_dp_atomic_release_vcpi_slots(state, + mst_mgr, + mst_port); } static const struct drm_connector_helper_funcs dm_dp_mst_connector_helper_funcs = { .get_modes = dm_dp_mst_get_modes, .mode_valid = amdgpu_dm_connector_mode_valid, - .best_encoder = dm_mst_best_encoder, + .atomic_best_encoder = dm_mst_atomic_best_encoder, + .detect_ctx = dm_dp_mst_detect, + .atomic_check = dm_dp_mst_atomic_check, }; static void amdgpu_dm_encoder_destroy(struct drm_encoder *encoder) @@ -388,13 +482,17 @@ void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, struct amdgpu_dm_connector *aconnector) { aconnector->dm_dp_aux.aux.name = "dmdc"; - aconnector->dm_dp_aux.aux.dev = dm->adev->dev; + aconnector->dm_dp_aux.aux.dev = aconnector->base.kdev; aconnector->dm_dp_aux.aux.transfer = dm_dp_aux_transfer; aconnector->dm_dp_aux.ddc_service = aconnector->dc_link->ddc; drm_dp_aux_register(&aconnector->dm_dp_aux.aux); drm_dp_cec_register_connector(&aconnector->dm_dp_aux.aux, - aconnector->base.name, dm->adev->dev); + &aconnector->base); + + if (aconnector->base.connector_type == DRM_MODE_CONNECTOR_eDP) + return; + aconnector->mst_mgr.cbs = &dm_mst_cbs; drm_dp_mst_topology_mgr_init( &aconnector->mst_mgr, @@ -405,3 +503,384 @@ void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, aconnector->connector_id); } +int dm_mst_get_pbn_divider(struct dc_link *link) +{ + if (!link) + return 0; + + return dc_link_bandwidth_kbps(link, + dc_link_get_link_cap(link)) / (8 * 1000 * 54); +} + +#if defined(CONFIG_DRM_AMD_DC_DCN) + +struct dsc_mst_fairness_params { + struct dc_crtc_timing *timing; + struct dc_sink *sink; + struct dc_dsc_bw_range bw_range; + bool compression_possible; + struct drm_dp_mst_port *port; +}; + +struct dsc_mst_fairness_vars { + int pbn; + bool dsc_enabled; + int bpp_x16; +}; + +static int kbps_to_peak_pbn(int kbps) +{ + u64 peak_kbps = kbps; + + peak_kbps *= 1006; + peak_kbps = div_u64(peak_kbps, 1000); + return (int) DIV64_U64_ROUND_UP(peak_kbps * 64, (54 * 8 * 1000)); +} + +static void set_dsc_configs_from_fairness_vars(struct dsc_mst_fairness_params *params, + struct dsc_mst_fairness_vars *vars, + int count) +{ + int i; + + for (i = 0; i < count; i++) { + memset(¶ms[i].timing->dsc_cfg, 0, sizeof(params[i].timing->dsc_cfg)); + if (vars[i].dsc_enabled && dc_dsc_compute_config( + params[i].sink->ctx->dc->res_pool->dscs[0], + ¶ms[i].sink->sink_dsc_caps.dsc_dec_caps, + params[i].sink->ctx->dc->debug.dsc_min_slice_height_override, + 0, + params[i].timing, + ¶ms[i].timing->dsc_cfg)) { + params[i].timing->flags.DSC = 1; + params[i].timing->dsc_cfg.bits_per_pixel = vars[i].bpp_x16; + } else { + params[i].timing->flags.DSC = 0; + } + } +} + +static int bpp_x16_from_pbn(struct dsc_mst_fairness_params param, int pbn) +{ + struct dc_dsc_config dsc_config; + u64 kbps; + + kbps = div_u64((u64)pbn * 994 * 8 * 54, 64); + dc_dsc_compute_config( + param.sink->ctx->dc->res_pool->dscs[0], + ¶m.sink->sink_dsc_caps.dsc_dec_caps, + param.sink->ctx->dc->debug.dsc_min_slice_height_override, + (int) kbps, param.timing, &dsc_config); + + return dsc_config.bits_per_pixel; +} + +static void increase_dsc_bpp(struct drm_atomic_state *state, + struct dc_link *dc_link, + struct dsc_mst_fairness_params *params, + struct dsc_mst_fairness_vars *vars, + int count) +{ + int i; + bool bpp_increased[MAX_PIPES]; + int initial_slack[MAX_PIPES]; + int min_initial_slack; + int next_index; + int remaining_to_increase = 0; + int pbn_per_timeslot; + int link_timeslots_used; + int fair_pbn_alloc; + + for (i = 0; i < count; i++) { + if (vars[i].dsc_enabled) { + initial_slack[i] = kbps_to_peak_pbn(params[i].bw_range.max_kbps) - vars[i].pbn; + bpp_increased[i] = false; + remaining_to_increase += 1; + } else { + initial_slack[i] = 0; + bpp_increased[i] = true; + } + } + + pbn_per_timeslot = dc_link_bandwidth_kbps(dc_link, + dc_link_get_link_cap(dc_link)) / (8 * 1000 * 54); + + while (remaining_to_increase) { + next_index = -1; + min_initial_slack = -1; + for (i = 0; i < count; i++) { + if (!bpp_increased[i]) { + if (min_initial_slack == -1 || min_initial_slack > initial_slack[i]) { + min_initial_slack = initial_slack[i]; + next_index = i; + } + } + } + + if (next_index == -1) + break; + + link_timeslots_used = 0; + + for (i = 0; i < count; i++) + link_timeslots_used += DIV_ROUND_UP(vars[i].pbn, pbn_per_timeslot); + + fair_pbn_alloc = (63 - link_timeslots_used) / remaining_to_increase * pbn_per_timeslot; + + if (initial_slack[next_index] > fair_pbn_alloc) { + vars[next_index].pbn += fair_pbn_alloc; + if (drm_dp_atomic_find_vcpi_slots(state, + params[next_index].port->mgr, + params[next_index].port, + vars[next_index].pbn, + dm_mst_get_pbn_divider(dc_link)) < 0) + return; + if (!drm_dp_mst_atomic_check(state)) { + vars[next_index].bpp_x16 = bpp_x16_from_pbn(params[next_index], vars[next_index].pbn); + } else { + vars[next_index].pbn -= fair_pbn_alloc; + if (drm_dp_atomic_find_vcpi_slots(state, + params[next_index].port->mgr, + params[next_index].port, + vars[next_index].pbn, + dm_mst_get_pbn_divider(dc_link)) < 0) + return; + } + } else { + vars[next_index].pbn += initial_slack[next_index]; + if (drm_dp_atomic_find_vcpi_slots(state, + params[next_index].port->mgr, + params[next_index].port, + vars[next_index].pbn, + dm_mst_get_pbn_divider(dc_link)) < 0) + return; + if (!drm_dp_mst_atomic_check(state)) { + vars[next_index].bpp_x16 = params[next_index].bw_range.max_target_bpp_x16; + } else { + vars[next_index].pbn -= initial_slack[next_index]; + if (drm_dp_atomic_find_vcpi_slots(state, + params[next_index].port->mgr, + params[next_index].port, + vars[next_index].pbn, + dm_mst_get_pbn_divider(dc_link)) < 0) + return; + } + } + + bpp_increased[next_index] = true; + remaining_to_increase--; + } +} + +static void try_disable_dsc(struct drm_atomic_state *state, + struct dc_link *dc_link, + struct dsc_mst_fairness_params *params, + struct dsc_mst_fairness_vars *vars, + int count) +{ + int i; + bool tried[MAX_PIPES]; + int kbps_increase[MAX_PIPES]; + int max_kbps_increase; + int next_index; + int remaining_to_try = 0; + + for (i = 0; i < count; i++) { + if (vars[i].dsc_enabled && vars[i].bpp_x16 == params[i].bw_range.max_target_bpp_x16) { + kbps_increase[i] = params[i].bw_range.stream_kbps - params[i].bw_range.max_kbps; + tried[i] = false; + remaining_to_try += 1; + } else { + kbps_increase[i] = 0; + tried[i] = true; + } + } + + while (remaining_to_try) { + next_index = -1; + max_kbps_increase = -1; + for (i = 0; i < count; i++) { + if (!tried[i]) { + if (max_kbps_increase == -1 || max_kbps_increase < kbps_increase[i]) { + max_kbps_increase = kbps_increase[i]; + next_index = i; + } + } + } + + if (next_index == -1) + break; + + vars[next_index].pbn = kbps_to_peak_pbn(params[next_index].bw_range.stream_kbps); + if (drm_dp_atomic_find_vcpi_slots(state, + params[next_index].port->mgr, + params[next_index].port, + vars[next_index].pbn, + 0) < 0) + return; + + if (!drm_dp_mst_atomic_check(state)) { + vars[next_index].dsc_enabled = false; + vars[next_index].bpp_x16 = 0; + } else { + vars[next_index].pbn = kbps_to_peak_pbn(params[next_index].bw_range.max_kbps); + if (drm_dp_atomic_find_vcpi_slots(state, + params[next_index].port->mgr, + params[next_index].port, + vars[next_index].pbn, + dm_mst_get_pbn_divider(dc_link)) < 0) + return; + } + + tried[next_index] = true; + remaining_to_try--; + } +} + +static bool compute_mst_dsc_configs_for_link(struct drm_atomic_state *state, + struct dc_state *dc_state, + struct dc_link *dc_link) +{ + int i; + struct dc_stream_state *stream; + struct dsc_mst_fairness_params params[MAX_PIPES]; + struct dsc_mst_fairness_vars vars[MAX_PIPES]; + struct amdgpu_dm_connector *aconnector; + int count = 0; + + memset(params, 0, sizeof(params)); + + /* Set up params */ + for (i = 0; i < dc_state->stream_count; i++) { + struct dc_dsc_policy dsc_policy = {0}; + + stream = dc_state->streams[i]; + + if (stream->link != dc_link) + continue; + + stream->timing.flags.DSC = 0; + + params[count].timing = &stream->timing; + params[count].sink = stream->sink; + aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context; + params[count].port = aconnector->port; + params[count].compression_possible = stream->sink->sink_dsc_caps.dsc_dec_caps.is_dsc_supported; + dc_dsc_get_policy_for_timing(params[count].timing, &dsc_policy); + if (!dc_dsc_compute_bandwidth_range( + stream->sink->ctx->dc->res_pool->dscs[0], + stream->sink->ctx->dc->debug.dsc_min_slice_height_override, + dsc_policy.min_target_bpp, + dsc_policy.max_target_bpp, + &stream->sink->sink_dsc_caps.dsc_dec_caps, + &stream->timing, ¶ms[count].bw_range)) + params[count].bw_range.stream_kbps = dc_bandwidth_in_kbps_from_timing(&stream->timing); + + count++; + } + /* Try no compression */ + for (i = 0; i < count; i++) { + vars[i].pbn = kbps_to_peak_pbn(params[i].bw_range.stream_kbps); + vars[i].dsc_enabled = false; + vars[i].bpp_x16 = 0; + if (drm_dp_atomic_find_vcpi_slots(state, + params[i].port->mgr, + params[i].port, + vars[i].pbn, + 0) < 0) + return false; + } + if (!drm_dp_mst_atomic_check(state)) { + set_dsc_configs_from_fairness_vars(params, vars, count); + return true; + } + + /* Try max compression */ + for (i = 0; i < count; i++) { + if (params[i].compression_possible) { + vars[i].pbn = kbps_to_peak_pbn(params[i].bw_range.min_kbps); + vars[i].dsc_enabled = true; + vars[i].bpp_x16 = params[i].bw_range.min_target_bpp_x16; + if (drm_dp_atomic_find_vcpi_slots(state, + params[i].port->mgr, + params[i].port, + vars[i].pbn, + dm_mst_get_pbn_divider(dc_link)) < 0) + return false; + } else { + vars[i].pbn = kbps_to_peak_pbn(params[i].bw_range.stream_kbps); + vars[i].dsc_enabled = false; + vars[i].bpp_x16 = 0; + if (drm_dp_atomic_find_vcpi_slots(state, + params[i].port->mgr, + params[i].port, + vars[i].pbn, + 0) < 0) + return false; + } + } + if (drm_dp_mst_atomic_check(state)) + return false; + + /* Optimize degree of compression */ + increase_dsc_bpp(state, dc_link, params, vars, count); + + try_disable_dsc(state, dc_link, params, vars, count); + + set_dsc_configs_from_fairness_vars(params, vars, count); + + return true; +} + +bool compute_mst_dsc_configs_for_state(struct drm_atomic_state *state, + struct dc_state *dc_state) +{ + int i, j; + struct dc_stream_state *stream; + bool computed_streams[MAX_PIPES]; + struct amdgpu_dm_connector *aconnector; + + for (i = 0; i < dc_state->stream_count; i++) + computed_streams[i] = false; + + for (i = 0; i < dc_state->stream_count; i++) { + stream = dc_state->streams[i]; + + if (stream->signal != SIGNAL_TYPE_DISPLAY_PORT_MST) + continue; + + aconnector = (struct amdgpu_dm_connector *)stream->dm_stream_context; + + if (!aconnector || !aconnector->dc_sink) + continue; + + if (!aconnector->dc_sink->sink_dsc_caps.dsc_dec_caps.is_dsc_supported) + continue; + + if (computed_streams[i]) + continue; + + mutex_lock(&aconnector->mst_mgr.lock); + if (!compute_mst_dsc_configs_for_link(state, dc_state, stream->link)) { + mutex_unlock(&aconnector->mst_mgr.lock); + return false; + } + mutex_unlock(&aconnector->mst_mgr.lock); + + for (j = 0; j < dc_state->stream_count; j++) { + if (dc_state->streams[j]->link == stream->link) + computed_streams[j] = true; + } + } + + for (i = 0; i < dc_state->stream_count; i++) { + stream = dc_state->streams[i]; + + if (stream->timing.flags.DSC == 1) + dcn20_add_dsc_to_stream_resource(stream->ctx->dc, dc_state, stream); + } + + return true; +} + +#endif diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h index 2da851b40042..d6813ce67bbd 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h @@ -29,7 +29,14 @@ struct amdgpu_display_manager; struct amdgpu_dm_connector; +int dm_mst_get_pbn_divider(struct dc_link *link); + void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, struct amdgpu_dm_connector *aconnector); +#if defined(CONFIG_DRM_AMD_DC_DCN) +bool compute_mst_dsc_configs_for_state(struct drm_atomic_state *state, + struct dc_state *dc_state); +#endif + #endif diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c index 592fa499c9f8..a2e1a73f66b8 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c @@ -151,18 +151,31 @@ static void get_default_clock_levels( static enum smu_clk_type dc_to_smu_clock_type( enum dm_pp_clock_type dm_pp_clk_type) { -#define DCCLK_MAP_SMUCLK(dcclk, smuclk) \ - [dcclk] = smuclk + enum smu_clk_type smu_clk_type = SMU_CLK_COUNT; - static int dc_clk_type_map[] = { - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_DISPLAY_CLK, SMU_DISPCLK), - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_ENGINE_CLK, SMU_GFXCLK), - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_MEMORY_CLK, SMU_MCLK), - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_DCEFCLK, SMU_DCEFCLK), - DCCLK_MAP_SMUCLK(DM_PP_CLOCK_TYPE_SOCCLK, SMU_SOCCLK), - }; + switch (dm_pp_clk_type) { + case DM_PP_CLOCK_TYPE_DISPLAY_CLK: + smu_clk_type = SMU_DISPCLK; + break; + case DM_PP_CLOCK_TYPE_ENGINE_CLK: + smu_clk_type = SMU_GFXCLK; + break; + case DM_PP_CLOCK_TYPE_MEMORY_CLK: + smu_clk_type = SMU_MCLK; + break; + case DM_PP_CLOCK_TYPE_DCEFCLK: + smu_clk_type = SMU_DCEFCLK; + break; + case DM_PP_CLOCK_TYPE_SOCCLK: + smu_clk_type = SMU_SOCCLK; + break; + default: + DRM_ERROR("DM_PPLIB: invalid clock type: %d!\n", + dm_pp_clk_type); + break; + } - return dc_clk_type_map[dm_pp_clk_type]; + return smu_clk_type; } static enum amd_pp_clock_type dc_to_pp_clock_type( @@ -329,12 +342,13 @@ bool dm_pp_get_clock_levels_by_type( if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->get_clock_by_type) { if (adev->powerplay.pp_funcs->get_clock_by_type(pp_handle, dc_to_pp_clock_type(clk_type), &pp_clks)) { - /* Error in pplib. Provide default values. */ + /* Error in pplib. Provide default values. */ + get_default_clock_levels(clk_type, dc_clks); return true; } - } else if (adev->smu.funcs && adev->smu.funcs->get_clock_by_type) { + } else if (adev->smu.ppt_funcs && adev->smu.ppt_funcs->get_clock_by_type) { if (smu_get_clock_by_type(&adev->smu, - dc_to_smu_clock_type(clk_type), + dc_to_pp_clock_type(clk_type), &pp_clks)) { get_default_clock_levels(clk_type, dc_clks); return true; @@ -352,7 +366,7 @@ bool dm_pp_get_clock_levels_by_type( validation_clks.memory_max_clock = 80000; validation_clks.level = 0; } - } else if (adev->smu.funcs && adev->smu.funcs->get_max_high_clocks) { + } else if (adev->smu.ppt_funcs && adev->smu.ppt_funcs->get_max_high_clocks) { if (smu_get_max_high_clocks(&adev->smu, &validation_clks)) { DRM_INFO("DM_PPLIB: Warning: using default validation clocks!\n"); validation_clks.engine_max_clock = 72000; @@ -419,7 +433,7 @@ bool dm_pp_get_clock_levels_by_type_with_latency( return false; } else if (adev->smu.ppt_funcs && adev->smu.ppt_funcs->get_clock_by_type_with_latency) { if (smu_get_clock_by_type_with_latency(&adev->smu, - dc_to_pp_clock_type(clk_type), + dc_to_smu_clock_type(clk_type), &pp_clks)) return false; } @@ -493,8 +507,8 @@ bool dm_pp_apply_clock_for_voltage_request( ret = adev->powerplay.pp_funcs->display_clock_voltage_request( adev->powerplay.pp_handle, &pp_clock_request); - else if (adev->smu.funcs && - adev->smu.funcs->display_clock_voltage_request) + else if (adev->smu.ppt_funcs && + adev->smu.ppt_funcs->display_clock_voltage_request) ret = smu_display_clock_voltage_request(&adev->smu, &pp_clock_request); if (ret) @@ -514,7 +528,7 @@ bool dm_pp_get_static_clocks( ret = adev->powerplay.pp_funcs->get_current_clocks( adev->powerplay.pp_handle, &pp_clk_info); - else if (adev->smu.funcs) + else if (adev->smu.ppt_funcs) ret = smu_get_current_clocks(&adev->smu, &pp_clk_info); if (ret) return false; @@ -576,10 +590,9 @@ void pp_rv_set_wm_ranges(struct pp_smu *pp, if (pp_funcs && pp_funcs->set_watermarks_for_clocks_ranges) pp_funcs->set_watermarks_for_clocks_ranges(pp_handle, &wm_with_clock_ranges); - else if (adev->smu.funcs && - adev->smu.funcs->set_watermarks_for_clock_ranges) + else smu_set_watermarks_for_clock_ranges(&adev->smu, - &wm_with_clock_ranges); + &wm_with_clock_ranges); } void pp_rv_set_pme_wa_enable(struct pp_smu *pp) @@ -591,7 +604,7 @@ void pp_rv_set_pme_wa_enable(struct pp_smu *pp) if (pp_funcs && pp_funcs->notify_smu_enable_pwe) pp_funcs->notify_smu_enable_pwe(pp_handle); - else if (adev->smu.funcs) + else if (adev->smu.ppt_funcs) smu_notify_smu_enable_pwe(&adev->smu); } @@ -652,7 +665,6 @@ enum pp_smu_status pp_nv_set_wm_ranges(struct pp_smu *pp, { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - struct smu_context *smu = &adev->smu; struct dm_pp_wm_sets_with_clock_ranges_soc15 wm_with_clock_ranges; struct dm_pp_clock_range_for_dmif_wm_set_soc15 *wm_dce_clocks = wm_with_clock_ranges.wm_dmif_clocks_ranges; @@ -695,15 +707,7 @@ enum pp_smu_status pp_nv_set_wm_ranges(struct pp_smu *pp, ranges->writer_wm_sets[i].min_drain_clk_mhz * 1000; } - if (!smu->funcs) - return PP_SMU_RESULT_UNSUPPORTED; - - /* 0: successful or smu.funcs->set_watermarks_for_clock_ranges = NULL; - * 1: fail - */ - if (smu_set_watermarks_for_clock_ranges(&adev->smu, - &wm_with_clock_ranges)) - return PP_SMU_RESULT_UNSUPPORTED; + smu_set_watermarks_for_clock_ranges(&adev->smu, &wm_with_clock_ranges); return PP_SMU_RESULT_OK; } @@ -714,10 +718,10 @@ enum pp_smu_status pp_nv_set_pme_wa_enable(struct pp_smu *pp) struct amdgpu_device *adev = ctx->driver_context; struct smu_context *smu = &adev->smu; - if (!smu->funcs) + if (!smu->ppt_funcs) return PP_SMU_RESULT_UNSUPPORTED; - /* 0: successful or smu.funcs->set_azalia_d3_pme = NULL; 1: fail */ + /* 0: successful or smu.ppt_funcs->set_azalia_d3_pme = NULL; 1: fail */ if (smu_set_azalia_d3_pme(smu)) return PP_SMU_RESULT_FAIL; @@ -730,10 +734,10 @@ enum pp_smu_status pp_nv_set_display_count(struct pp_smu *pp, int count) struct amdgpu_device *adev = ctx->driver_context; struct smu_context *smu = &adev->smu; - if (!smu->funcs) + if (!smu->ppt_funcs) return PP_SMU_RESULT_UNSUPPORTED; - /* 0: successful or smu.funcs->set_display_count = NULL; 1: fail */ + /* 0: successful or smu.ppt_funcs->set_display_count = NULL; 1: fail */ if (smu_set_display_count(smu, count)) return PP_SMU_RESULT_FAIL; @@ -746,10 +750,10 @@ enum pp_smu_status pp_nv_set_min_deep_sleep_dcfclk(struct pp_smu *pp, int mhz) struct amdgpu_device *adev = ctx->driver_context; struct smu_context *smu = &adev->smu; - if (!smu->funcs) + if (!smu->ppt_funcs) return PP_SMU_RESULT_UNSUPPORTED; - /* 0: successful or smu.funcs->set_deep_sleep_dcefclk = NULL;1: fail */ + /* 0: successful or smu.ppt_funcs->set_deep_sleep_dcefclk = NULL;1: fail */ if (smu_set_deep_sleep_dcefclk(smu, mhz)) return PP_SMU_RESULT_FAIL; @@ -764,13 +768,13 @@ enum pp_smu_status pp_nv_set_hard_min_dcefclk_by_freq( struct smu_context *smu = &adev->smu; struct pp_display_clock_request clock_req; - if (!smu->funcs) + if (!smu->ppt_funcs) return PP_SMU_RESULT_UNSUPPORTED; clock_req.clock_type = amd_pp_dcef_clock; clock_req.clock_freq_in_khz = mhz * 1000; - /* 0: successful or smu.funcs->display_clock_voltage_request = NULL + /* 0: successful or smu.ppt_funcs->display_clock_voltage_request = NULL * 1: fail */ if (smu_display_clock_voltage_request(smu, &clock_req)) @@ -786,13 +790,13 @@ enum pp_smu_status pp_nv_set_hard_min_uclk_by_freq(struct pp_smu *pp, int mhz) struct smu_context *smu = &adev->smu; struct pp_display_clock_request clock_req; - if (!smu->funcs) + if (!smu->ppt_funcs) return PP_SMU_RESULT_UNSUPPORTED; clock_req.clock_type = amd_pp_mem_clock; clock_req.clock_freq_in_khz = mhz * 1000; - /* 0: successful or smu.funcs->display_clock_voltage_request = NULL + /* 0: successful or smu.ppt_funcs->display_clock_voltage_request = NULL * 1: fail */ if (smu_display_clock_voltage_request(smu, &clock_req)) @@ -801,6 +805,19 @@ enum pp_smu_status pp_nv_set_hard_min_uclk_by_freq(struct pp_smu *pp, int mhz) return PP_SMU_RESULT_OK; } +enum pp_smu_status pp_nv_set_pstate_handshake_support( + struct pp_smu *pp, BOOLEAN pstate_handshake_supported) +{ + const struct dc_context *ctx = pp->dm; + struct amdgpu_device *adev = ctx->driver_context; + struct smu_context *smu = &adev->smu; + + if (smu_display_disable_memory_clock_switch(smu, !pstate_handshake_supported)) + return PP_SMU_RESULT_FAIL; + + return PP_SMU_RESULT_OK; +} + enum pp_smu_status pp_nv_set_voltage_by_freq(struct pp_smu *pp, enum pp_smu_nv_clock_id clock_id, int mhz) { @@ -809,7 +826,7 @@ enum pp_smu_status pp_nv_set_voltage_by_freq(struct pp_smu *pp, struct smu_context *smu = &adev->smu; struct pp_display_clock_request clock_req; - if (!smu->funcs) + if (!smu->ppt_funcs) return PP_SMU_RESULT_UNSUPPORTED; switch (clock_id) { @@ -827,7 +844,7 @@ enum pp_smu_status pp_nv_set_voltage_by_freq(struct pp_smu *pp, } clock_req.clock_freq_in_khz = mhz * 1000; - /* 0: successful or smu.funcs->display_clock_voltage_request = NULL + /* 0: successful or smu.ppt_funcs->display_clock_voltage_request = NULL * 1: fail */ if (smu_display_clock_voltage_request(smu, &clock_req)) @@ -843,13 +860,13 @@ enum pp_smu_status pp_nv_get_maximum_sustainable_clocks( struct amdgpu_device *adev = ctx->driver_context; struct smu_context *smu = &adev->smu; - if (!smu->funcs) + if (!smu->ppt_funcs) return PP_SMU_RESULT_UNSUPPORTED; - if (!smu->funcs->get_max_sustainable_clocks_by_dc) + if (!smu->ppt_funcs->get_max_sustainable_clocks_by_dc) return PP_SMU_RESULT_UNSUPPORTED; - if (!smu->funcs->get_max_sustainable_clocks_by_dc(smu, max_clocks)) + if (!smu_get_max_sustainable_clocks_by_dc(smu, max_clocks)) return PP_SMU_RESULT_OK; return PP_SMU_RESULT_FAIL; @@ -868,13 +885,95 @@ enum pp_smu_status pp_nv_get_uclk_dpm_states(struct pp_smu *pp, if (!smu->ppt_funcs->get_uclk_dpm_states) return PP_SMU_RESULT_UNSUPPORTED; - if (!smu->ppt_funcs->get_uclk_dpm_states(smu, + if (!smu_get_uclk_dpm_states(smu, clock_values_in_khz, num_states)) return PP_SMU_RESULT_OK; return PP_SMU_RESULT_FAIL; } +enum pp_smu_status pp_rn_get_dpm_clock_table( + struct pp_smu *pp, struct dpm_clocks *clock_table) +{ + const struct dc_context *ctx = pp->dm; + struct amdgpu_device *adev = ctx->driver_context; + struct smu_context *smu = &adev->smu; + + if (!smu->ppt_funcs) + return PP_SMU_RESULT_UNSUPPORTED; + + if (!smu->ppt_funcs->get_dpm_clock_table) + return PP_SMU_RESULT_UNSUPPORTED; + + if (!smu_get_dpm_clock_table(smu, clock_table)) + return PP_SMU_RESULT_OK; + + return PP_SMU_RESULT_FAIL; +} + +enum pp_smu_status pp_rn_set_wm_ranges(struct pp_smu *pp, + struct pp_smu_wm_range_sets *ranges) +{ + const struct dc_context *ctx = pp->dm; + struct amdgpu_device *adev = ctx->driver_context; + struct smu_context *smu = &adev->smu; + struct dm_pp_wm_sets_with_clock_ranges_soc15 wm_with_clock_ranges; + struct dm_pp_clock_range_for_dmif_wm_set_soc15 *wm_dce_clocks = + wm_with_clock_ranges.wm_dmif_clocks_ranges; + struct dm_pp_clock_range_for_mcif_wm_set_soc15 *wm_soc_clocks = + wm_with_clock_ranges.wm_mcif_clocks_ranges; + int32_t i; + + if (!smu->ppt_funcs) + return PP_SMU_RESULT_UNSUPPORTED; + + wm_with_clock_ranges.num_wm_dmif_sets = ranges->num_reader_wm_sets; + wm_with_clock_ranges.num_wm_mcif_sets = ranges->num_writer_wm_sets; + + for (i = 0; i < wm_with_clock_ranges.num_wm_dmif_sets; i++) { + if (ranges->reader_wm_sets[i].wm_inst > 3) + wm_dce_clocks[i].wm_set_id = WM_SET_A; + else + wm_dce_clocks[i].wm_set_id = + ranges->reader_wm_sets[i].wm_inst; + + wm_dce_clocks[i].wm_min_dcfclk_clk_in_khz = + ranges->reader_wm_sets[i].min_drain_clk_mhz; + + wm_dce_clocks[i].wm_max_dcfclk_clk_in_khz = + ranges->reader_wm_sets[i].max_drain_clk_mhz; + + wm_dce_clocks[i].wm_min_mem_clk_in_khz = + ranges->reader_wm_sets[i].min_fill_clk_mhz; + + wm_dce_clocks[i].wm_max_mem_clk_in_khz = + ranges->reader_wm_sets[i].max_fill_clk_mhz; + } + + for (i = 0; i < wm_with_clock_ranges.num_wm_mcif_sets; i++) { + if (ranges->writer_wm_sets[i].wm_inst > 3) + wm_soc_clocks[i].wm_set_id = WM_SET_A; + else + wm_soc_clocks[i].wm_set_id = + ranges->writer_wm_sets[i].wm_inst; + wm_soc_clocks[i].wm_min_socclk_clk_in_khz = + ranges->writer_wm_sets[i].min_fill_clk_mhz; + + wm_soc_clocks[i].wm_max_socclk_clk_in_khz = + ranges->writer_wm_sets[i].max_fill_clk_mhz; + + wm_soc_clocks[i].wm_min_mem_clk_in_khz = + ranges->writer_wm_sets[i].min_drain_clk_mhz; + + wm_soc_clocks[i].wm_max_mem_clk_in_khz = + ranges->writer_wm_sets[i].max_drain_clk_mhz; + } + + smu_set_watermarks_for_clock_ranges(&adev->smu, &wm_with_clock_ranges); + + return PP_SMU_RESULT_OK; +} + void dm_pp_get_funcs( struct dc_context *ctx, struct pp_smu_funcs *funcs) @@ -895,7 +994,6 @@ void dm_pp_get_funcs( funcs->rv_funcs.set_hard_min_fclk_by_freq = pp_rv_set_hard_min_fclk_by_freq; break; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 case DCN_VERSION_2_0: funcs->ctx.ver = PP_SMU_VER_NV; funcs->nv_funcs.pp_smu.dm = ctx; @@ -916,8 +1014,15 @@ void dm_pp_get_funcs( funcs->nv_funcs.get_maximum_sustainable_clocks = pp_nv_get_maximum_sustainable_clocks; /*todo compare data with window driver */ funcs->nv_funcs.get_uclk_dpm_states = pp_nv_get_uclk_dpm_states; + funcs->nv_funcs.set_pstate_handshake_support = pp_nv_set_pstate_handshake_support; + break; + + case DCN_VERSION_2_1: + funcs->ctx.ver = PP_SMU_VER_RN; + funcs->rn_funcs.pp_smu.dm = ctx; + funcs->rn_funcs.set_wm_ranges = pp_rn_set_wm_ranges; + funcs->rn_funcs.get_dpm_clock_table = pp_rn_get_dpm_clock_table; break; -#endif default: DRM_ERROR("smu version is not supported !\n"); break; diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile index 55ce5b657390..6e3dddc73246 100644 --- a/drivers/gpu/drm/amd/display/dc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/Makefile @@ -25,17 +25,11 @@ DC_LIBS = basics bios calcs clk_mgr dce gpio irq virtual -ifdef CONFIG_DRM_AMD_DC_DCN2_0 +ifdef CONFIG_DRM_AMD_DC_DCN DC_LIBS += dcn20 -endif - - -ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT DC_LIBS += dsc -endif - -ifdef CONFIG_DRM_AMD_DC_DCN1_0 DC_LIBS += dcn10 dml +DC_LIBS += dcn21 endif DC_LIBS += dce120 @@ -45,6 +39,10 @@ DC_LIBS += dce110 DC_LIBS += dce100 DC_LIBS += dce80 +ifdef CONFIG_DRM_AMD_DC_HDCP +DC_LIBS += hdcp +endif + AMD_DC = $(addsuffix /Makefile, $(addprefix $(FULL_AMD_DISPLAY_PATH)/dc/,$(DC_LIBS))) include $(AMD_DC) @@ -52,7 +50,7 @@ include $(AMD_DC) DISPLAY_CORE = dc.o dc_link.o dc_resource.o dc_hw_sequencer.o dc_sink.o \ dc_surface.o dc_link_hwss.o dc_link_dp.o dc_link_ddc.o dc_debug.o dc_stream.o -ifdef CONFIG_DRM_AMD_DC_DCN2_0 +ifdef CONFIG_DRM_AMD_DC_DCN DISPLAY_CORE += dc_vm_helper.o endif @@ -63,5 +61,6 @@ AMD_DM_REG_UPDATE = $(addprefix $(AMDDALPATH)/dc/,dc_helper.o) AMD_DISPLAY_FILES += $(AMD_DISPLAY_CORE) AMD_DISPLAY_FILES += $(AMD_DM_REG_UPDATE) - - +DC_DMUB += dc_dmub_srv.o +AMD_DISPLAY_DMUB = $(addprefix $(AMDDALPATH)/dc/,$(DC_DMUB)) +AMD_DISPLAY_FILES += $(AMD_DISPLAY_DMUB) diff --git a/drivers/gpu/drm/amd/display/dc/basics/Makefile b/drivers/gpu/drm/amd/display/dc/basics/Makefile index a50a76471107..7ad0cad0f4ef 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/Makefile +++ b/drivers/gpu/drm/amd/display/dc/basics/Makefile @@ -25,7 +25,7 @@ # subcomponents. BASICS = conversion.o fixpt31_32.o \ - log_helpers.o vector.o + log_helpers.o vector.o dc_common.o AMD_DAL_BASICS = $(addprefix $(AMDDALPATH)/dc/basics/,$(BASICS)) diff --git a/drivers/gpu/drm/amd/display/dc/basics/dc_common.c b/drivers/gpu/drm/amd/display/dc/basics/dc_common.c new file mode 100644 index 000000000000..b2fc4f8e6482 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/basics/dc_common.c @@ -0,0 +1,101 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "core_types.h" +#include "dc_common.h" +#include "basics/conversion.h" + +bool is_rgb_cspace(enum dc_color_space output_color_space) +{ + switch (output_color_space) { + case COLOR_SPACE_SRGB: + case COLOR_SPACE_SRGB_LIMITED: + case COLOR_SPACE_2020_RGB_FULLRANGE: + case COLOR_SPACE_2020_RGB_LIMITEDRANGE: + case COLOR_SPACE_ADOBERGB: + return true; + case COLOR_SPACE_YCBCR601: + case COLOR_SPACE_YCBCR709: + case COLOR_SPACE_YCBCR601_LIMITED: + case COLOR_SPACE_YCBCR709_LIMITED: + case COLOR_SPACE_2020_YCBCR: + return false; + default: + /* Add a case to switch */ + BREAK_TO_DEBUGGER(); + return false; + } +} + +bool is_lower_pipe_tree_visible(struct pipe_ctx *pipe_ctx) +{ + if (pipe_ctx->plane_state && pipe_ctx->plane_state->visible) + return true; + if (pipe_ctx->bottom_pipe && is_lower_pipe_tree_visible(pipe_ctx->bottom_pipe)) + return true; + return false; +} + +bool is_upper_pipe_tree_visible(struct pipe_ctx *pipe_ctx) +{ + if (pipe_ctx->plane_state && pipe_ctx->plane_state->visible) + return true; + if (pipe_ctx->top_pipe && is_upper_pipe_tree_visible(pipe_ctx->top_pipe)) + return true; + return false; +} + +bool is_pipe_tree_visible(struct pipe_ctx *pipe_ctx) +{ + if (pipe_ctx->plane_state && pipe_ctx->plane_state->visible) + return true; + if (pipe_ctx->top_pipe && is_upper_pipe_tree_visible(pipe_ctx->top_pipe)) + return true; + if (pipe_ctx->bottom_pipe && is_lower_pipe_tree_visible(pipe_ctx->bottom_pipe)) + return true; + return false; +} + +void build_prescale_params(struct dc_bias_and_scale *bias_and_scale, + const struct dc_plane_state *plane_state) +{ + if (plane_state->format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN + && plane_state->format != SURFACE_PIXEL_FORMAT_INVALID + && plane_state->input_csc_color_matrix.enable_adjustment + && plane_state->coeff_reduction_factor.value != 0) { + bias_and_scale->scale_blue = fixed_point_to_int_frac( + dc_fixpt_mul(plane_state->coeff_reduction_factor, + dc_fixpt_from_fraction(256, 255)), + 2, + 13); + bias_and_scale->scale_red = bias_and_scale->scale_blue; + bias_and_scale->scale_green = bias_and_scale->scale_blue; + } else { + bias_and_scale->scale_blue = 0x2000; + bias_and_scale->scale_red = 0x2000; + bias_and_scale->scale_green = 0x2000; + } +} + diff --git a/drivers/gpu/drm/amd/display/dc/basics/dc_common.h b/drivers/gpu/drm/amd/display/dc/basics/dc_common.h new file mode 100644 index 000000000000..7c0cbf47e8ce --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/basics/dc_common.h @@ -0,0 +1,42 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DAL_DC_COMMON_H__ +#define __DAL_DC_COMMON_H__ + +#include "core_types.h" + +bool is_rgb_cspace(enum dc_color_space output_color_space); + +bool is_lower_pipe_tree_visible(struct pipe_ctx *pipe_ctx); + +bool is_upper_pipe_tree_visible(struct pipe_ctx *pipe_ctx); + +bool is_pipe_tree_visible(struct pipe_ctx *pipe_ctx); + +void build_prescale_params(struct dc_bias_and_scale *bias_and_scale, + const struct dc_plane_state *plane_state); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c index 461eef1de124..008d4d11339d 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c @@ -111,7 +111,7 @@ struct dc_bios *bios_parser_create( return NULL; } -static void destruct(struct bios_parser *bp) +static void bios_parser_destruct(struct bios_parser *bp) { kfree(bp->base.bios_local_image); kfree(bp->base.integrated_info); @@ -126,7 +126,7 @@ static void bios_parser_destroy(struct dc_bios **dcb) return; } - destruct(bp); + bios_parser_destruct(bp); kfree(bp); *dcb = NULL; @@ -2189,7 +2189,7 @@ static uint32_t get_support_mask_for_device_id(struct device_id device_id) break; default: break; - }; + } /* Unidentified device ID, return empty support mask. */ return 0; @@ -2543,7 +2543,6 @@ static enum bp_result construct_integrated_info( /* Sort voltage table from low to high*/ if (result == BP_RESULT_OK) { - struct clock_voltage_caps temp = {0, 0}; uint32_t i; uint32_t j; @@ -2553,10 +2552,8 @@ static enum bp_result construct_integrated_info( info->disp_clk_voltage[j].max_supported_clk < info->disp_clk_voltage[j-1].max_supported_clk) { /* swap j and j - 1*/ - temp = info->disp_clk_voltage[j-1]; - info->disp_clk_voltage[j-1] = - info->disp_clk_voltage[j]; - info->disp_clk_voltage[j] = temp; + swap(info->disp_clk_voltage[j - 1], + info->disp_clk_voltage[j]); } } } @@ -2742,7 +2739,6 @@ static enum bp_result bios_get_board_layout_info( struct board_layout_info *board_layout_info) { unsigned int i; - struct bios_parser *bp; enum bp_result record_result; const unsigned int slot_index_to_vbios_id[MAX_BOARD_SLOTS] = { @@ -2751,7 +2747,6 @@ static enum bp_result bios_get_board_layout_info( 0, 0 }; - bp = BP_FROM_DCB(dcb); if (board_layout_info == NULL) { DC_LOG_DETECTION_EDID_PARSER("Invalid board_layout_info\n"); return BP_RESULT_BADINPUT; @@ -2796,8 +2791,6 @@ static const struct dc_vbios_funcs vbios_funcs = { .get_device_tag = bios_parser_get_device_tag, - .get_firmware_info = bios_parser_get_firmware_info, - .get_spread_spectrum_info = bios_parser_get_spread_spectrum_info, .get_ss_entry_number = bios_parser_get_ss_entry_number, @@ -2922,6 +2915,7 @@ static bool bios_parser_construct( dal_bios_parser_init_cmd_tbl_helper(&bp->cmd_helper, dce_version); bp->base.integrated_info = bios_parser_create_integrated_info(&bp->base); + bp->base.fw_info_valid = bios_parser_get_firmware_info(&bp->base, &bp->base.fw_info) == BP_RESULT_OK; return true; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index 6aa2e56dfb67..2f1c9584ac32 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -111,7 +111,7 @@ static struct atom_encoder_caps_record *get_encoder_cap_record( #define DATA_TABLES(table) (bp->master_data_tbl->listOfdatatables.table) -static void destruct(struct bios_parser *bp) +static void bios_parser2_destruct(struct bios_parser *bp) { kfree(bp->base.bios_local_image); kfree(bp->base.integrated_info); @@ -126,7 +126,7 @@ static void firmware_parser_destroy(struct dc_bios **dcb) return; } - destruct(bp); + bios_parser2_destruct(bp); kfree(bp); *dcb = NULL; @@ -294,11 +294,21 @@ static enum bp_result bios_parser_get_i2c_info(struct dc_bios *dcb, struct atom_display_object_path_v2 *object; struct atom_common_record_header *header; struct atom_i2c_record *record; + struct atom_i2c_record dummy_record = {0}; struct bios_parser *bp = BP_FROM_DCB(dcb); if (!info) return BP_RESULT_BADINPUT; + if (id.type == OBJECT_TYPE_GENERIC) { + dummy_record.i2c_id = id.id; + + if (get_gpio_i2c_info(bp, &dummy_record, info) == BP_RESULT_OK) + return BP_RESULT_OK; + else + return BP_RESULT_NORECORD; + } + object = get_bios_object(bp, id); if (!object) @@ -341,6 +351,7 @@ static enum bp_result get_gpio_i2c_info( struct atom_gpio_pin_lut_v2_1 *header; uint32_t count = 0; unsigned int table_index = 0; + bool find_valid = false; if (!info) return BP_RESULT_BADINPUT; @@ -368,33 +379,28 @@ static enum bp_result get_gpio_i2c_info( - sizeof(struct atom_common_table_header)) / sizeof(struct atom_gpio_pin_assignment); - table_index = record->i2c_id & I2C_HW_LANE_MUX; - - if (count < table_index) { - bool find_valid = false; - - for (table_index = 0; table_index < count; table_index++) { - if (((record->i2c_id & I2C_HW_CAP) == ( - header->gpio_pin[table_index].gpio_id & - I2C_HW_CAP)) && - ((record->i2c_id & I2C_HW_ENGINE_ID_MASK) == - (header->gpio_pin[table_index].gpio_id & - I2C_HW_ENGINE_ID_MASK)) && - ((record->i2c_id & I2C_HW_LANE_MUX) == - (header->gpio_pin[table_index].gpio_id & - I2C_HW_LANE_MUX))) { - /* still valid */ - find_valid = true; - break; - } + for (table_index = 0; table_index < count; table_index++) { + if (((record->i2c_id & I2C_HW_CAP) == ( + header->gpio_pin[table_index].gpio_id & + I2C_HW_CAP)) && + ((record->i2c_id & I2C_HW_ENGINE_ID_MASK) == + (header->gpio_pin[table_index].gpio_id & + I2C_HW_ENGINE_ID_MASK)) && + ((record->i2c_id & I2C_HW_LANE_MUX) == + (header->gpio_pin[table_index].gpio_id & + I2C_HW_LANE_MUX))) { + /* still valid */ + find_valid = true; + break; } - /* If we don't find the entry that we are looking for then - * we will return BP_Result_BadBiosTable. - */ - if (find_valid == false) - return BP_RESULT_BADBIOSTABLE; } + /* If we don't find the entry that we are looking for then + * we will return BP_Result_BadBiosTable. + */ + if (find_valid == false) + return BP_RESULT_BADBIOSTABLE; + /* get the GPIO_I2C_INFO */ info->i2c_hw_assist = (record->i2c_id & I2C_HW_CAP) ? true : false; info->i2c_line = record->i2c_id & I2C_HW_LANE_MUX; @@ -828,6 +834,7 @@ static enum bp_result bios_parser_get_spread_spectrum_info( case 1: return get_ss_info_v4_1(bp, signal, index, ss_info); case 2: + case 3: return get_ss_info_v4_2(bp, signal, index, ss_info); default: break; @@ -986,7 +993,7 @@ static uint32_t get_support_mask_for_device_id(struct device_id device_id) break; default: break; - }; + } /* Unidentified device ID, return empty support mask. */ return 0; @@ -1205,6 +1212,8 @@ static enum bp_result get_firmware_info_v3_1( bp->cmd_tbl.get_smu_clock_info(bp, SMU9_SYSPLL0_ID) * 10; } + info->oem_i2c_present = false; + return BP_RESULT_OK; } @@ -1283,6 +1292,13 @@ static enum bp_result get_firmware_info_v3_2( bp->cmd_tbl.get_smu_clock_info(bp, SMU11_SYSPLL3_0_ID) * 10; } + if (firmware_info->board_i2c_feature_id == 0x2) { + info->oem_i2c_present = true; + info->oem_i2c_obj_id = firmware_info->board_i2c_feature_gpio_id; + } else { + info->oem_i2c_present = false; + } + return BP_RESULT_OK; } @@ -1402,10 +1418,8 @@ static enum bp_result get_integrated_info_v11( info->ma_channel_number = info_v11->umachannelnumber; info->lvds_ss_percentage = le16_to_cpu(info_v11->lvds_ss_percentage); -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 info->dp_ss_control = le16_to_cpu(info_v11->reserved1); -#endif info->lvds_sspread_rate_in_10hz = le16_to_cpu(info_v11->lvds_ss_rate_10hz); info->hdmi_ss_percentage = @@ -1613,8 +1627,6 @@ static enum bp_result construct_integrated_info( struct atom_common_table_header *header; struct atom_data_revision revision; - - struct clock_voltage_caps temp = {0, 0}; uint32_t i; uint32_t j; @@ -1627,6 +1639,7 @@ static enum bp_result construct_integrated_info( /* Don't need to check major revision as they are all 1 */ switch (revision.minor) { case 11: + case 12: result = get_integrated_info_v11(bp, info); break; default: @@ -1644,10 +1657,8 @@ static enum bp_result construct_integrated_info( info->disp_clk_voltage[j-1].max_supported_clk ) { /* swap j and j - 1*/ - temp = info->disp_clk_voltage[j-1]; - info->disp_clk_voltage[j-1] = - info->disp_clk_voltage[j]; - info->disp_clk_voltage[j] = temp; + swap(info->disp_clk_voltage[j - 1], + info->disp_clk_voltage[j]); } } } @@ -1829,7 +1840,6 @@ static enum bp_result bios_get_board_layout_info( struct board_layout_info *board_layout_info) { unsigned int i; - struct bios_parser *bp; enum bp_result record_result; const unsigned int slot_index_to_vbios_id[MAX_BOARD_SLOTS] = { @@ -1838,7 +1848,6 @@ static enum bp_result bios_get_board_layout_info( 0, 0 }; - bp = BP_FROM_DCB(dcb); if (board_layout_info == NULL) { DC_LOG_DETECTION_EDID_PARSER("Invalid board_layout_info\n"); return BP_RESULT_BADINPUT; @@ -1881,8 +1890,6 @@ static const struct dc_vbios_funcs vbios_funcs = { .get_device_tag = bios_parser_get_device_tag, - .get_firmware_info = bios_parser_get_firmware_info, - .get_spread_spectrum_info = bios_parser_get_spread_spectrum_info, .get_ss_entry_number = bios_parser_get_ss_entry_number, @@ -1920,7 +1927,7 @@ static const struct dc_vbios_funcs vbios_funcs = { .get_board_layout_info = bios_get_board_layout_info, }; -static bool bios_parser_construct( +static bool bios_parser2_construct( struct bios_parser *bp, struct bp_init_data *init, enum dce_version dce_version) @@ -1998,6 +2005,7 @@ static bool bios_parser_construct( dal_bios_parser_init_cmd_tbl_helper2(&bp->cmd_helper, dce_version); bp->base.integrated_info = bios_parser_create_integrated_info(&bp->base); + bp->base.fw_info_valid = bios_parser_get_firmware_info(&bp->base, &bp->base.fw_info) == BP_RESULT_OK; return true; } @@ -2012,7 +2020,7 @@ struct dc_bios *firmware_parser_create( if (!bp) return NULL; - if (bios_parser_construct(bp, init, dce_version)) + if (bios_parser2_construct(bp, init, dce_version)) return &bp->base; kfree(bp); diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index bb2e8105e6ab..c4ba6e84db65 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -37,6 +37,8 @@ #include "bios_parser_types_internal2.h" #include "amdgpu.h" +#include "dc_dmub_srv.h" +#include "dc.h" #define DC_LOGGER \ bp->base.ctx->logger @@ -87,6 +89,10 @@ static enum bp_result encoder_control_digx_v1_5( struct bios_parser *bp, struct bp_encoder_control *cntl); +static enum bp_result encoder_control_fallback( + struct bios_parser *bp, + struct bp_encoder_control *cntl); + static void init_dig_encoder_control(struct bios_parser *bp) { uint32_t version = @@ -98,11 +104,26 @@ static void init_dig_encoder_control(struct bios_parser *bp) break; default: dm_output_to_console("Don't have dig_encoder_control for v%d\n", version); - bp->cmd_tbl.dig_encoder_control = NULL; + bp->cmd_tbl.dig_encoder_control = encoder_control_fallback; break; } } +static void encoder_control_dmcub( + struct dc_dmub_srv *dmcub, + struct dig_encoder_stream_setup_parameters_v1_5 *dig) +{ + struct dmub_rb_cmd_digx_encoder_control encoder_control = { 0 }; + + encoder_control.header.type = DMUB_CMD__VBIOS; + encoder_control.header.sub_type = DMUB_CMD__VBIOS_DIGX_ENCODER_CONTROL; + encoder_control.encoder_control.dig.stream_param = *dig; + + dc_dmub_srv_cmd_queue(dmcub, &encoder_control.header); + dc_dmub_srv_cmd_execute(dmcub); + dc_dmub_srv_wait_idle(dmcub); +} + static enum bp_result encoder_control_digx_v1_5( struct bios_parser *bp, struct bp_encoder_control *cntl) @@ -155,12 +176,30 @@ static enum bp_result encoder_control_digx_v1_5( break; } + if (bp->base.ctx->dc->ctx->dmub_srv && + bp->base.ctx->dc->debug.dmub_command_table) { + encoder_control_dmcub(bp->base.ctx->dmub_srv, ¶ms); + return BP_RESULT_OK; + } + if (EXEC_BIOS_CMD_TABLE(digxencodercontrol, params)) result = BP_RESULT_OK; return result; } +static enum bp_result encoder_control_fallback( + struct bios_parser *bp, + struct bp_encoder_control *cntl) +{ + if (bp->base.ctx->dc->ctx->dmub_srv && + bp->base.ctx->dc->debug.dmub_command_table) { + return encoder_control_digx_v1_5(bp, cntl); + } + + return BP_RESULT_FAILURE; +} + /***************************************************************************** ****************************************************************************** ** @@ -173,6 +212,10 @@ static enum bp_result transmitter_control_v1_6( struct bios_parser *bp, struct bp_transmitter_control *cntl); +static enum bp_result transmitter_control_fallback( + struct bios_parser *bp, + struct bp_transmitter_control *cntl); + static void init_transmitter_control(struct bios_parser *bp) { uint8_t frev; @@ -186,11 +229,27 @@ static void init_transmitter_control(struct bios_parser *bp) break; default: dm_output_to_console("Don't have transmitter_control for v%d\n", crev); - bp->cmd_tbl.transmitter_control = NULL; + bp->cmd_tbl.transmitter_control = transmitter_control_fallback; break; } } +static void transmitter_control_dmcub( + struct dc_dmub_srv *dmcub, + struct dig_transmitter_control_parameters_v1_6 *dig) +{ + struct dmub_rb_cmd_dig1_transmitter_control transmitter_control; + + transmitter_control.header.type = DMUB_CMD__VBIOS; + transmitter_control.header.sub_type = + DMUB_CMD__VBIOS_DIG1_TRANSMITTER_CONTROL; + transmitter_control.transmitter_control.dig = *dig; + + dc_dmub_srv_cmd_queue(dmcub, &transmitter_control.header); + dc_dmub_srv_cmd_execute(dmcub); + dc_dmub_srv_wait_idle(dmcub); +} + static enum bp_result transmitter_control_v1_6( struct bios_parser *bp, struct bp_transmitter_control *cntl) @@ -222,6 +281,11 @@ static enum bp_result transmitter_control_v1_6( __func__, ps.param.symclk_10khz); } + if (bp->base.ctx->dc->ctx->dmub_srv && + bp->base.ctx->dc->debug.dmub_command_table) { + transmitter_control_dmcub(bp->base.ctx->dmub_srv, &ps.param); + return BP_RESULT_OK; + } /*color_depth not used any more, driver has deep color factor in the Phyclk*/ if (EXEC_BIOS_CMD_TABLE(dig1transmittercontrol, ps)) @@ -229,6 +293,18 @@ static enum bp_result transmitter_control_v1_6( return result; } +static enum bp_result transmitter_control_fallback( + struct bios_parser *bp, + struct bp_transmitter_control *cntl) +{ + if (bp->base.ctx->dc->ctx->dmub_srv && + bp->base.ctx->dc->debug.dmub_command_table) { + return transmitter_control_v1_6(bp, cntl); + } + + return BP_RESULT_FAILURE; +} + /****************************************************************************** ****************************************************************************** ** @@ -241,6 +317,10 @@ static enum bp_result set_pixel_clock_v7( struct bios_parser *bp, struct bp_pixel_clock_parameters *bp_params); +static enum bp_result set_pixel_clock_fallback( + struct bios_parser *bp, + struct bp_pixel_clock_parameters *bp_params); + static void init_set_pixel_clock(struct bios_parser *bp) { switch (BIOS_CMD_TABLE_PARA_REVISION(setpixelclock)) { @@ -250,12 +330,25 @@ static void init_set_pixel_clock(struct bios_parser *bp) default: dm_output_to_console("Don't have set_pixel_clock for v%d\n", BIOS_CMD_TABLE_PARA_REVISION(setpixelclock)); - bp->cmd_tbl.set_pixel_clock = NULL; + bp->cmd_tbl.set_pixel_clock = set_pixel_clock_fallback; break; } } +static void set_pixel_clock_dmcub( + struct dc_dmub_srv *dmcub, + struct set_pixel_clock_parameter_v1_7 *clk) +{ + struct dmub_rb_cmd_set_pixel_clock pixel_clock = { 0 }; + pixel_clock.header.type = DMUB_CMD__VBIOS; + pixel_clock.header.sub_type = DMUB_CMD__VBIOS_SET_PIXEL_CLOCK; + pixel_clock.pixel_clock.clk = *clk; + + dc_dmub_srv_cmd_queue(dmcub, &pixel_clock.header); + dc_dmub_srv_cmd_execute(dmcub); + dc_dmub_srv_wait_idle(dmcub); +} static enum bp_result set_pixel_clock_v7( struct bios_parser *bp, @@ -331,12 +424,30 @@ static enum bp_result set_pixel_clock_v7( if (bp_params->signal_type == SIGNAL_TYPE_DVI_DUAL_LINK) clk.miscinfo |= PIXEL_CLOCK_V7_MISC_DVI_DUALLINK_EN; + if (bp->base.ctx->dc->ctx->dmub_srv && + bp->base.ctx->dc->debug.dmub_command_table) { + set_pixel_clock_dmcub(bp->base.ctx->dmub_srv, &clk); + return BP_RESULT_OK; + } + if (EXEC_BIOS_CMD_TABLE(setpixelclock, clk)) result = BP_RESULT_OK; } return result; } +static enum bp_result set_pixel_clock_fallback( + struct bios_parser *bp, + struct bp_pixel_clock_parameters *bp_params) +{ + if (bp->base.ctx->dc->ctx->dmub_srv && + bp->base.ctx->dc->debug.dmub_command_table) { + return set_pixel_clock_v7(bp, bp_params); + } + + return BP_RESULT_FAILURE; +} + /****************************************************************************** ****************************************************************************** ** @@ -569,6 +680,11 @@ static enum bp_result enable_disp_power_gating_v2_1( enum controller_id crtc_id, enum bp_pipe_control_action action); +static enum bp_result enable_disp_power_gating_fallback( + struct bios_parser *bp, + enum controller_id crtc_id, + enum bp_pipe_control_action action); + static void init_enable_disp_power_gating( struct bios_parser *bp) { @@ -580,11 +696,26 @@ static void init_enable_disp_power_gating( default: dm_output_to_console("Don't enable_disp_power_gating enable_crtc for v%d\n", BIOS_CMD_TABLE_PARA_REVISION(enabledisppowergating)); - bp->cmd_tbl.enable_disp_power_gating = NULL; + bp->cmd_tbl.enable_disp_power_gating = enable_disp_power_gating_fallback; break; } } +static void enable_disp_power_gating_dmcub( + struct dc_dmub_srv *dmcub, + struct enable_disp_power_gating_parameters_v2_1 *pwr) +{ + struct dmub_rb_cmd_enable_disp_power_gating power_gating; + + power_gating.header.type = DMUB_CMD__VBIOS; + power_gating.header.sub_type = DMUB_CMD__VBIOS_ENABLE_DISP_POWER_GATING; + power_gating.power_gating.pwr = *pwr; + + dc_dmub_srv_cmd_queue(dmcub, &power_gating.header); + dc_dmub_srv_cmd_execute(dmcub); + dc_dmub_srv_wait_idle(dmcub); +} + static enum bp_result enable_disp_power_gating_v2_1( struct bios_parser *bp, enum controller_id crtc_id, @@ -604,12 +735,32 @@ static enum bp_result enable_disp_power_gating_v2_1( ps.param.enable = bp->cmd_helper->disp_power_gating_action_to_atom(action); + if (bp->base.ctx->dc->ctx->dmub_srv && + bp->base.ctx->dc->debug.dmub_command_table) { + enable_disp_power_gating_dmcub(bp->base.ctx->dmub_srv, + &ps.param); + return BP_RESULT_OK; + } + if (EXEC_BIOS_CMD_TABLE(enabledisppowergating, ps.param)) result = BP_RESULT_OK; return result; } +static enum bp_result enable_disp_power_gating_fallback( + struct bios_parser *bp, + enum controller_id crtc_id, + enum bp_pipe_control_action action) +{ + if (bp->base.ctx->dc->ctx->dmub_srv && + bp->base.ctx->dc->debug.dmub_command_table) { + return enable_disp_power_gating_v2_1(bp, crtc_id, action); + } + + return BP_RESULT_FAILURE; +} + /****************************************************************************** ******************************************************************************* ** diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c index f9439dfc7b75..7388c987c595 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c @@ -55,18 +55,19 @@ bool dal_bios_parser_init_cmd_tbl_helper2( case DCE_VERSION_11_22: *h = dal_cmd_tbl_helper_dce112_get_table2(); return true; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) case DCN_VERSION_1_0: case DCN_VERSION_1_01: *h = dal_cmd_tbl_helper_dce112_get_table2(); return true; #endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) case DCN_VERSION_2_0: *h = dal_cmd_tbl_helper_dce112_get_table2(); return true; -#endif + case DCN_VERSION_2_1: + *h = dal_cmd_tbl_helper_dce112_get_table2(); + return true; case DCE_VERSION_12_0: case DCE_VERSION_12_1: *h = dal_cmd_tbl_helper_dce112_get_table2(); diff --git a/drivers/gpu/drm/amd/display/dc/bios/dce110/command_table_helper_dce110.c b/drivers/gpu/drm/amd/display/dc/bios/dce110/command_table_helper_dce110.c index ca24154468c7..11bf247bb180 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/dce110/command_table_helper_dce110.c +++ b/drivers/gpu/drm/amd/display/dc/bios/dce110/command_table_helper_dce110.c @@ -153,38 +153,10 @@ static uint8_t hpd_sel_to_atom(enum hpd_source_id id) static uint8_t dig_encoder_sel_to_atom(enum engine_id id) { - uint8_t atom_dig_encoder_sel = 0; - - switch (id) { - case ENGINE_ID_DIGA: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGA_SEL; - break; - case ENGINE_ID_DIGB: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGB_SEL; - break; - case ENGINE_ID_DIGC: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGC_SEL; - break; - case ENGINE_ID_DIGD: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGD_SEL; - break; - case ENGINE_ID_DIGE: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGE_SEL; - break; - case ENGINE_ID_DIGF: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGF_SEL; - break; - case ENGINE_ID_DIGG: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGG_SEL; - break; - case ENGINE_ID_UNKNOWN: - /* No DIG_FRONT is associated to DIG_BACKEND */ - atom_dig_encoder_sel = 0; - break; - default: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V5__DIGA_SEL; - break; - } + /* On any ASIC after DCE80, we manually program the DIG_FE + * selection (see connect_dig_be_to_fe function of the link + * encoder), so translation should always return 0 (no FE). + */ return 0; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper2_dce112.c b/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper2_dce112.c index 0237ae575068..755b6e33140a 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper2_dce112.c +++ b/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper2_dce112.c @@ -150,38 +150,10 @@ static uint8_t hpd_sel_to_atom(enum hpd_source_id id) static uint8_t dig_encoder_sel_to_atom(enum engine_id id) { - uint8_t atom_dig_encoder_sel = 0; - - switch (id) { - case ENGINE_ID_DIGA: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGA_SEL; - break; - case ENGINE_ID_DIGB: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGB_SEL; - break; - case ENGINE_ID_DIGC: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGC_SEL; - break; - case ENGINE_ID_DIGD: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGD_SEL; - break; - case ENGINE_ID_DIGE: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGE_SEL; - break; - case ENGINE_ID_DIGF: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGF_SEL; - break; - case ENGINE_ID_DIGG: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGG_SEL; - break; - case ENGINE_ID_UNKNOWN: - /* No DIG_FRONT is associated to DIG_BACKEND */ - atom_dig_encoder_sel = 0; - break; - default: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGA_SEL; - break; - } + /* On any ASIC after DCE80, we manually program the DIG_FE + * selection (see connect_dig_be_to_fe function of the link + * encoder), so translation should always return 0 (no FE). + */ return 0; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper_dce112.c b/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper_dce112.c index 452034f83e4c..06b4f7fa4a50 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper_dce112.c +++ b/drivers/gpu/drm/amd/display/dc/bios/dce112/command_table_helper_dce112.c @@ -150,38 +150,10 @@ static uint8_t hpd_sel_to_atom(enum hpd_source_id id) static uint8_t dig_encoder_sel_to_atom(enum engine_id id) { - uint8_t atom_dig_encoder_sel = 0; - - switch (id) { - case ENGINE_ID_DIGA: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGA_SEL; - break; - case ENGINE_ID_DIGB: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGB_SEL; - break; - case ENGINE_ID_DIGC: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGC_SEL; - break; - case ENGINE_ID_DIGD: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGD_SEL; - break; - case ENGINE_ID_DIGE: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGE_SEL; - break; - case ENGINE_ID_DIGF: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGF_SEL; - break; - case ENGINE_ID_DIGG: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGG_SEL; - break; - case ENGINE_ID_UNKNOWN: - /* No DIG_FRONT is associated to DIG_BACKEND */ - atom_dig_encoder_sel = 0; - break; - default: - atom_dig_encoder_sel = ATOM_TRANMSITTER_V6__DIGA_SEL; - break; - } + /* On any ASIC after DCE80, we manually program the DIG_FE + * selection (see connect_dig_be_to_fe function of the link + * encoder), so translation should always return 0 (no FE). + */ return 0; } diff --git a/drivers/gpu/drm/amd/display/dc/calcs/Makefile b/drivers/gpu/drm/amd/display/dc/calcs/Makefile index 95f332ee3e7e..4674aca8f206 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/Makefile +++ b/drivers/gpu/drm/amd/display/dc/calcs/Makefile @@ -1,5 +1,6 @@ # # Copyright 2017 Advanced Micro Devices, Inc. +# Copyright 2019 Raptor Engineering, LLC # # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the "Software"), @@ -24,21 +25,38 @@ # It calculates Bandwidth and Watermarks values for HW programming # -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),) - cc_stack_align := -mpreferred-stack-boundary=4 -else ifneq ($(call cc-option, -mstack-alignment=16),) - cc_stack_align := -mstack-alignment=16 +ifdef CONFIG_X86 +calcs_ccflags := -mhard-float -msse endif -calcs_ccflags := -mhard-float -msse $(cc_stack_align) +ifdef CONFIG_PPC64 +calcs_ccflags := -mhard-float -maltivec +endif + +ifdef CONFIG_CC_IS_GCC +ifeq ($(call cc-ifversion, -lt, 0701, y), y) +IS_OLD_GCC = 1 +endif +endif + +ifdef CONFIG_X86 +ifdef IS_OLD_GCC +# Stack alignment mismatch, proceed with caution. +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 +# (8B stack alignment). +calcs_ccflags += -mpreferred-stack-boundary=4 +else +calcs_ccflags += -msse2 +endif +endif -CFLAGS_dcn_calcs.o := $(calcs_ccflags) -CFLAGS_dcn_calc_auto.o := $(calcs_ccflags) -CFLAGS_dcn_calc_math.o := $(calcs_ccflags) -Wno-tautological-compare +CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calcs.o := $(calcs_ccflags) +CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_auto.o := $(calcs_ccflags) +CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_math.o := $(calcs_ccflags) -Wno-tautological-compare BW_CALCS = dce_calcs.o bw_fixed.o custom_float.o -ifdef CONFIG_DRM_AMD_DC_DCN1_0 +ifdef CONFIG_DRM_AMD_DC_DCN BW_CALCS += dcn_calcs.o dcn_calc_math.o dcn_calc_auto.o endif diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c index 7108d51a9c5b..5d081c42e81b 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c @@ -25,6 +25,7 @@ #include <linux/slab.h> +#include "resource.h" #include "dm_services.h" #include "dce_calcs.h" #include "dc.h" @@ -153,14 +154,14 @@ static void calculate_bandwidth( - if (data->d0_underlay_mode == bw_def_none) { d0_underlay_enable = 0; } - else { - d0_underlay_enable = 1; - } - if (data->d1_underlay_mode == bw_def_none) { d1_underlay_enable = 0; } - else { - d1_underlay_enable = 1; - } + if (data->d0_underlay_mode == bw_def_none) + d0_underlay_enable = false; + else + d0_underlay_enable = true; + if (data->d1_underlay_mode == bw_def_none) + d1_underlay_enable = false; + else + d1_underlay_enable = true; data->number_of_underlay_surfaces = d0_underlay_enable + d1_underlay_enable; switch (data->underlay_surface_type) { case bw_def_420: @@ -285,8 +286,8 @@ static void calculate_bandwidth( data->cursor_width_pixels[2] = bw_int_to_fixed(0); data->cursor_width_pixels[3] = bw_int_to_fixed(0); /* graphics surface parameters from spreadsheet*/ - fbc_enabled = 0; - lpt_enabled = 0; + fbc_enabled = false; + lpt_enabled = false; for (i = 4; i <= maximum_number_of_surfaces - 3; i++) { if (i < data->number_of_displays + 4) { if (i == 4 && data->d0_underlay_mode == bw_def_underlay_only) { @@ -337,9 +338,9 @@ static void calculate_bandwidth( data->access_one_channel_only[i] = 0; } if (data->fbc_en[i] == 1) { - fbc_enabled = 1; + fbc_enabled = true; if (data->lpt_en[i] == 1) { - lpt_enabled = 1; + lpt_enabled = true; } } data->cursor_width_pixels[i] = bw_int_to_fixed(vbios->cursor_width); @@ -2852,7 +2853,7 @@ static void populate_initial_data( data->src_height[num_displays * 2 + j] = bw_int_to_fixed(pipe[i].bottom_pipe->plane_res.scl_data.viewport.height); data->src_width[num_displays * 2 + j] = bw_int_to_fixed(pipe[i].bottom_pipe->plane_res.scl_data.viewport.width); data->pitch_in_pixels[num_displays * 2 + j] = bw_int_to_fixed( - pipe[i].bottom_pipe->plane_state->plane_size.grph.surface_pitch); + pipe[i].bottom_pipe->plane_state->plane_size.surface_pitch); data->h_taps[num_displays * 2 + j] = bw_int_to_fixed(pipe[i].bottom_pipe->plane_res.scl_data.taps.h_taps); data->v_taps[num_displays * 2 + j] = bw_int_to_fixed(pipe[i].bottom_pipe->plane_res.scl_data.taps.v_taps); data->h_scale_ratio[num_displays * 2 + j] = fixed31_32_to_bw_fixed( @@ -2977,6 +2978,32 @@ static void populate_initial_data( data->number_of_displays = num_displays; } +static bool all_displays_in_sync(const struct pipe_ctx pipe[], + int pipe_count) +{ + const struct pipe_ctx *active_pipes[MAX_PIPES]; + int i, num_active_pipes = 0; + + for (i = 0; i < pipe_count; i++) { + if (!pipe[i].stream || pipe[i].top_pipe) + continue; + + active_pipes[num_active_pipes++] = &pipe[i]; + } + + if (!num_active_pipes) + return false; + + for (i = 1; i < num_active_pipes; ++i) { + if (!resource_are_streams_timing_synchronizable( + active_pipes[0]->stream, active_pipes[i]->stream)) { + return false; + } + } + + return true; +} + /** * Return: * true - Display(s) configuration supported. @@ -2998,8 +3025,10 @@ bool bw_calcs(struct dc_context *ctx, populate_initial_data(pipe, pipe_count, data); - /*TODO: this should be taken out calcs output and assigned during timing sync for pplib use*/ - calcs_output->all_displays_in_sync = false; + if (ctx->dc->config.multi_mon_pp_mclk_switch) + calcs_output->all_displays_in_sync = all_displays_in_sync(pipe, pipe_count); + else + calcs_output->all_displays_in_sync = false; if (data->number_of_displays != 0) { uint8_t yclk_lvl, sclk_lvl; diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c index 38365dd911a3..1a37550731de 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c @@ -1,5 +1,6 @@ /* * Copyright 2017 Advanced Micro Devices, Inc. + * Copyright 2019 Raptor Engineering, LLC * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -53,13 +54,9 @@ * remain as-is as it provides us with a guarantee from HW that it is correct. */ -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 /* Defaults from spreadsheet rev#247. * RV2 delta: dram_clock_change_latency, max_num_dpp */ -#else -/* Defaults from spreadsheet rev#247 */ -#endif const struct dcn_soc_bounding_box dcn10_soc_defaults = { /* latencies */ .sr_exit_time = 17, /*us*/ @@ -329,7 +326,7 @@ static void pipe_ctx_to_e2e_pipe_params ( dcc_support_pixel_format(pipe->plane_state->format, &bpe) ? 1 : 0; } input->src.dcc_rate = 1; - input->src.meta_pitch = pipe->plane_state->dcc.grph.meta_pitch; + input->src.meta_pitch = pipe->plane_state->dcc.meta_pitch; input->src.source_scan = dm_horz; input->src.sw_mode = pipe->plane_state->tiling_info.gfx9.swizzle; @@ -626,7 +623,7 @@ static bool dcn_bw_apply_registry_override(struct dc *dc) { bool updated = false; - kernel_fpu_begin(); + DC_FP_START(); if ((int)(dc->dcn_soc->sr_exit_time * 1000) != dc->debug.sr_exit_time_ns && dc->debug.sr_exit_time_ns) { updated = true; @@ -662,7 +659,7 @@ static bool dcn_bw_apply_registry_override(struct dc *dc) dc->dcn_soc->dram_clock_change_latency = dc->debug.dram_clock_change_latency_ns / 1000.0; } - kernel_fpu_end(); + DC_FP_END(); return updated; } @@ -705,6 +702,17 @@ static void hack_bounding_box(struct dcn_bw_internal_vars *v, hack_force_pipe_split(v, context->streams[0]->timing.pix_clk_100hz); } + +unsigned int get_highest_allowed_voltage_level(uint32_t hw_internal_rev) +{ + /* for dali & pollock, the highest voltage level we want is 0 */ + if (ASICREV_IS_POLLOCK(hw_internal_rev) || ASICREV_IS_DALI(hw_internal_rev)) + return 0; + + /* we are ok with all levels */ + return 4; +} + bool dcn_validate_bandwidth( struct dc *dc, struct dc_state *context, @@ -731,7 +739,8 @@ bool dcn_validate_bandwidth( dcn_bw_sync_calcs_and_dml(dc); memset(v, 0, sizeof(*v)); - kernel_fpu_begin(); + DC_FP_START(); + v->sr_exit_time = dc->dcn_soc->sr_exit_time; v->sr_enter_plus_exit_time = dc->dcn_soc->sr_enter_plus_exit_time; v->urgent_latency = dc->dcn_soc->urgent_latency; @@ -1263,12 +1272,12 @@ bool dcn_validate_bandwidth( bw_limit = dc->dcn_soc->percent_disp_bw_limit * v->fabric_and_dram_bandwidth_vmax0p9; bw_limit_pass = (v->total_data_read_bandwidth / 1000.0) < bw_limit; - kernel_fpu_end(); + DC_FP_END(); PERFORMANCE_TRACE_END(); BW_VAL_TRACE_FINISH(); - if (bw_limit_pass && v->voltage_level != 5) + if (bw_limit_pass && v->voltage_level <= get_highest_allowed_voltage_level(dc->ctx->asic_id.hw_internal_rev)) return true; else return false; @@ -1426,37 +1435,49 @@ void dcn_bw_update_from_pplib(struct dc *dc) struct dc_context *ctx = dc->ctx; struct dm_pp_clock_levels_with_voltage fclks = {0}, dcfclks = {0}; bool res; + unsigned vmin0p65_idx, vmid0p72_idx, vnom0p8_idx, vmax0p9_idx; /* TODO: This is not the proper way to obtain fabric_and_dram_bandwidth, should be min(fclk, memclk) */ res = dm_pp_get_clock_levels_by_type_with_voltage( ctx, DM_PP_CLOCK_TYPE_FCLK, &fclks); - kernel_fpu_begin(); + DC_FP_START(); if (res) res = verify_clock_values(&fclks); if (res) { - ASSERT(fclks.num_levels >= 3); - dc->dcn_soc->fabric_and_dram_bandwidth_vmin0p65 = 32 * (fclks.data[0].clocks_in_khz / 1000.0) / 1000.0; - dc->dcn_soc->fabric_and_dram_bandwidth_vmid0p72 = dc->dcn_soc->number_of_channels * - (fclks.data[fclks.num_levels - (fclks.num_levels > 2 ? 3 : 2)].clocks_in_khz / 1000.0) - * ddr4_dram_factor_single_Channel / 1000.0; - dc->dcn_soc->fabric_and_dram_bandwidth_vnom0p8 = dc->dcn_soc->number_of_channels * - (fclks.data[fclks.num_levels - 2].clocks_in_khz / 1000.0) - * ddr4_dram_factor_single_Channel / 1000.0; - dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = dc->dcn_soc->number_of_channels * - (fclks.data[fclks.num_levels - 1].clocks_in_khz / 1000.0) - * ddr4_dram_factor_single_Channel / 1000.0; + ASSERT(fclks.num_levels); + + vmin0p65_idx = 0; + vmid0p72_idx = fclks.num_levels - + (fclks.num_levels > 2 ? 3 : (fclks.num_levels > 1 ? 2 : 1)); + vnom0p8_idx = fclks.num_levels - (fclks.num_levels > 1 ? 2 : 1); + vmax0p9_idx = fclks.num_levels - 1; + + dc->dcn_soc->fabric_and_dram_bandwidth_vmin0p65 = + 32 * (fclks.data[vmin0p65_idx].clocks_in_khz / 1000.0) / 1000.0; + dc->dcn_soc->fabric_and_dram_bandwidth_vmid0p72 = + dc->dcn_soc->number_of_channels * + (fclks.data[vmid0p72_idx].clocks_in_khz / 1000.0) + * ddr4_dram_factor_single_Channel / 1000.0; + dc->dcn_soc->fabric_and_dram_bandwidth_vnom0p8 = + dc->dcn_soc->number_of_channels * + (fclks.data[vnom0p8_idx].clocks_in_khz / 1000.0) + * ddr4_dram_factor_single_Channel / 1000.0; + dc->dcn_soc->fabric_and_dram_bandwidth_vmax0p9 = + dc->dcn_soc->number_of_channels * + (fclks.data[vmax0p9_idx].clocks_in_khz / 1000.0) + * ddr4_dram_factor_single_Channel / 1000.0; } else BREAK_TO_DEBUGGER(); - kernel_fpu_end(); + DC_FP_END(); res = dm_pp_get_clock_levels_by_type_with_voltage( ctx, DM_PP_CLOCK_TYPE_DCFCLK, &dcfclks); - kernel_fpu_begin(); + DC_FP_START(); if (res) res = verify_clock_values(&dcfclks); @@ -1469,7 +1490,7 @@ void dcn_bw_update_from_pplib(struct dc *dc) } else BREAK_TO_DEBUGGER(); - kernel_fpu_end(); + DC_FP_END(); } void dcn_bw_notify_pplib_of_wm_ranges(struct dc *dc) @@ -1484,11 +1505,11 @@ void dcn_bw_notify_pplib_of_wm_ranges(struct dc *dc) if (!pp || !pp->set_wm_ranges) return; - kernel_fpu_begin(); + DC_FP_START(); min_fclk_khz = dc->dcn_soc->fabric_and_dram_bandwidth_vmin0p65 * 1000000 / 32; min_dcfclk_khz = dc->dcn_soc->dcfclkv_min0p65 * 1000; socclk_khz = dc->dcn_soc->socclk * 1000; - kernel_fpu_end(); + DC_FP_END(); /* Now notify PPLib/SMU about which Watermarks sets they should select * depending on DPM state they are in. And update BW MGR GFX Engine and @@ -1539,7 +1560,7 @@ void dcn_bw_notify_pplib_of_wm_ranges(struct dc *dc) void dcn_bw_sync_calcs_and_dml(struct dc *dc) { - kernel_fpu_begin(); + DC_FP_START(); DC_LOG_BANDWIDTH_CALCS("sr_exit_time: %f ns\n" "sr_enter_plus_exit_time: %f ns\n" "urgent_latency: %f ns\n" @@ -1728,5 +1749,5 @@ void dcn_bw_sync_calcs_and_dml(struct dc *dc) dc->dml.ip.bug_forcing_LC_req_same_size_fixed = dc->dcn_ip->bug_forcing_luma_and_chroma_request_to_same_size_fixed == dcn_bw_yes; dc->dml.ip.dcfclk_cstate_latency = dc->dcn_ip->dcfclk_cstate_latency; - kernel_fpu_end(); + DC_FP_END(); } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile b/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile index 003c27767e9c..c0f6a8c7de7d 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/Makefile @@ -63,7 +63,7 @@ CLK_MGR_DCE120 = dce120_clk_mgr.o AMD_DAL_CLK_MGR_DCE120 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dce120/,$(CLK_MGR_DCE120)) AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCE120) -ifdef CONFIG_DRM_AMD_DC_DCN1_0 +ifdef CONFIG_DRM_AMD_DC_DCN ############################################################################### # DCN10 ############################################################################### @@ -72,9 +72,7 @@ CLK_MGR_DCN10 = rv1_clk_mgr.o rv1_clk_mgr_vbios_smu.o rv2_clk_mgr.o AMD_DAL_CLK_MGR_DCN10 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dcn10/,$(CLK_MGR_DCN10)) AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCN10) -endif -ifdef CONFIG_DRM_AMD_DC_DCN2_0 ############################################################################### # DCN20 ############################################################################### @@ -83,5 +81,19 @@ CLK_MGR_DCN20 = dcn20_clk_mgr.o AMD_DAL_CLK_MGR_DCN20 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dcn20/,$(CLK_MGR_DCN20)) AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCN20) + +############################################################################### +# DCN21 +############################################################################### +CLK_MGR_DCN21 = rn_clk_mgr.o rn_clk_mgr_vbios_smu.o + +# prevent build errors regarding soft-float vs hard-float FP ABI tags +# this code is currently unused on ppc64, as it applies to Renoir APUs only +ifdef CONFIG_PPC64 +CFLAGS_$(AMDDALPATH)/dc/clk_mgr/dcn21/rn_clk_mgr.o := $(call cc-option,-mno-gnu-attribute) endif +AMD_DAL_CLK_MGR_DCN21 = $(addprefix $(AMDDALPATH)/dc/clk_mgr/dcn21/,$(CLK_MGR_DCN21)) + +AMD_DISPLAY_FILES += $(AMD_DAL_CLK_MGR_DCN21) +endif diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c index 6b8fc5cbabb8..a78e5c74c79c 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c @@ -37,6 +37,7 @@ #include "dcn10/rv1_clk_mgr.h" #include "dcn10/rv2_clk_mgr.h" #include "dcn20/dcn20_clk_mgr.h" +#include "dcn21/rn_clk_mgr.h" int clk_mgr_helper_get_active_display_cnt( @@ -62,6 +63,31 @@ int clk_mgr_helper_get_active_display_cnt( return display_count; } +void clk_mgr_exit_optimized_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr) +{ + struct dc_link *edp_link = get_edp_link(dc); + + if (dc->hwss.exit_optimized_pwr_state) + dc->hwss.exit_optimized_pwr_state(dc, dc->current_state); + + if (edp_link) { + clk_mgr->psr_allow_active_cache = edp_link->psr_allow_active; + dc_link_set_psr_allow_active(edp_link, false, false); + } + +} + +void clk_mgr_optimize_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr) +{ + struct dc_link *edp_link = get_edp_link(dc); + + if (edp_link) + dc_link_set_psr_allow_active(edp_link, clk_mgr->psr_allow_active_cache, false); + + if (dc->hwss.optimize_pwr_state) + dc->hwss.optimize_pwr_state(dc, dc->current_state); + +} struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *pp_smu, struct dccg *dccg) { @@ -106,8 +132,19 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p dce120_clk_mgr_construct(ctx, clk_mgr); break; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) case FAMILY_RV: + if (ASICREV_IS_DALI(asic_id.hw_internal_rev) || + ASICREV_IS_POLLOCK(asic_id.hw_internal_rev)) { + /* TEMP: this check has to come before ASICREV_IS_RENOIR */ + /* which also incorrectly returns true for Dali/Pollock*/ + rv2_clk_mgr_construct(ctx, clk_mgr, pp_smu); + break; + } + if (ASICREV_IS_RENOIR(asic_id.hw_internal_rev)) { + rn_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg); + break; + } if (ASICREV_IS_RAVEN2(asic_id.hw_internal_rev)) { rv2_clk_mgr_construct(ctx, clk_mgr, pp_smu); break; @@ -118,13 +155,11 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p break; } break; -#endif /* Family RV */ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) case FAMILY_NV: dcn20_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg); break; -#endif /* Family NV */ +#endif /* Family RV and NV*/ default: ASSERT(0); /* Unknown Asic */ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c index 814450fefffa..26db1c5d4e4d 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c @@ -147,7 +147,7 @@ int dce_get_dp_ref_freq_khz(struct clk_mgr *clk_mgr_base) /* Calculate the current DFS clock, in kHz.*/ dp_ref_clk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR - * clk_mgr->dentist_vco_freq_khz) / target_div; + * clk_mgr->base.dentist_vco_freq_khz) / target_div; return dce_adjust_dp_ref_freq_for_ss(clk_mgr, dp_ref_clk_khz); } @@ -239,7 +239,7 @@ int dce_set_clock( /* Make sure requested clock isn't lower than minimum threshold*/ if (requested_clk_khz > 0) requested_clk_khz = max(requested_clk_khz, - clk_mgr_dce->dentist_vco_freq_khz / 64); + clk_mgr_dce->base.dentist_vco_freq_khz / 64); /* Prepare to program display clock*/ pxl_clk_params.target_pixel_clock_100hz = requested_clk_khz * 10; @@ -273,20 +273,14 @@ static void dce_clock_read_integrated_info(struct clk_mgr_internal *clk_mgr_dce) { struct dc_debug_options *debug = &clk_mgr_dce->base.ctx->dc->debug; struct dc_bios *bp = clk_mgr_dce->base.ctx->dc_bios; - struct integrated_info info = { { { 0 } } }; - struct dc_firmware_info fw_info = { { 0 } }; int i; if (bp->integrated_info) - info = *bp->integrated_info; - - clk_mgr_dce->dentist_vco_freq_khz = info.dentist_vco_freq; - if (clk_mgr_dce->dentist_vco_freq_khz == 0) { - bp->funcs->get_firmware_info(bp, &fw_info); - clk_mgr_dce->dentist_vco_freq_khz = - fw_info.smu_gpu_pll_output_freq; - if (clk_mgr_dce->dentist_vco_freq_khz == 0) - clk_mgr_dce->dentist_vco_freq_khz = 3600000; + clk_mgr_dce->base.dentist_vco_freq_khz = bp->integrated_info->dentist_vco_freq; + if (clk_mgr_dce->base.dentist_vco_freq_khz == 0) { + clk_mgr_dce->base.dentist_vco_freq_khz = bp->fw_info.smu_gpu_pll_output_freq; + if (clk_mgr_dce->base.dentist_vco_freq_khz == 0) + clk_mgr_dce->base.dentist_vco_freq_khz = 3600000; } /*update the maximum display clock for each power state*/ @@ -317,9 +311,10 @@ static void dce_clock_read_integrated_info(struct clk_mgr_internal *clk_mgr_dce) /*Do not allow bad VBIOS/SBIOS to override with invalid values, * check for > 100MHz*/ - if (info.disp_clk_voltage[i].max_supported_clk >= 100000) - clk_mgr_dce->max_clks_by_state[clk_state].display_clk_khz = - info.disp_clk_voltage[i].max_supported_clk; + if (bp->integrated_info) + if (bp->integrated_info->disp_clk_voltage[i].max_supported_clk >= 100000) + clk_mgr_dce->max_clks_by_state[clk_state].display_clk_khz = + bp->integrated_info->disp_clk_voltage[i].max_supported_clk; } if (!debug->disable_dfs_bypass && bp->integrated_info) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c index 5cc3acccda2a..b1e657e137a9 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c @@ -98,11 +98,14 @@ uint32_t dce110_get_min_vblank_time_us(const struct dc_state *context) struct dc_stream_state *stream = context->streams[j]; uint32_t vertical_blank_in_pixels = 0; uint32_t vertical_blank_time = 0; + uint32_t vertical_total_min = stream->timing.v_total; + struct dc_crtc_timing_adjust adjust = stream->adjust; + if (adjust.v_total_max != adjust.v_total_min) + vertical_total_min = adjust.v_total_min; vertical_blank_in_pixels = stream->timing.h_total * - (stream->timing.v_total + (vertical_total_min - stream->timing.v_addressable); - vertical_blank_time = vertical_blank_in_pixels * 10000 / stream->timing.pix_clk_100hz; @@ -171,6 +174,10 @@ void dce11_pplib_apply_display_requirements( struct dc_state *context) { struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg; + int memory_type_multiplier = MEMORY_TYPE_MULTIPLIER_CZ; + + if (dc->bw_vbios && dc->bw_vbios->memory_type == bw_def_hbm) + memory_type_multiplier = MEMORY_TYPE_HBM; pp_display_cfg->all_displays_in_sync = context->bw_ctx.bw.dce.all_displays_in_sync; @@ -183,8 +190,20 @@ void dce11_pplib_apply_display_requirements( pp_display_cfg->cpu_pstate_separation_time = context->bw_ctx.bw.dce.blackout_recovery_time_us; - pp_display_cfg->min_memory_clock_khz = context->bw_ctx.bw.dce.yclk_khz - / MEMORY_TYPE_MULTIPLIER_CZ; + /* + * TODO: determine whether the bandwidth has reached memory's limitation + * , then change minimum memory clock based on real-time bandwidth + * limitation. + */ + if (ASICREV_IS_VEGA20_P(dc->ctx->asic_id.hw_internal_rev) && (context->stream_count >= 2)) { + pp_display_cfg->min_memory_clock_khz = max(pp_display_cfg->min_memory_clock_khz, + (uint32_t) div64_s64( + div64_s64(dc->bw_vbios->high_yclk.value, + memory_type_multiplier), 10000)); + } else { + pp_display_cfg->min_memory_clock_khz = context->bw_ctx.bw.dce.yclk_khz + / memory_type_multiplier; + } pp_display_cfg->min_engine_clock_khz = determine_sclk_from_bounding_box( dc, diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c index 7c746ef1e32e..d031bd3d3072 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c @@ -72,8 +72,8 @@ int dce112_set_clock(struct clk_mgr *clk_mgr_base, int requested_clk_khz) struct clk_mgr_internal *clk_mgr_dce = TO_CLK_MGR_INTERNAL(clk_mgr_base); struct bp_set_dce_clock_parameters dce_clk_params; struct dc_bios *bp = clk_mgr_base->ctx->dc_bios; - struct dc *core_dc = clk_mgr_base->ctx->dc; - struct dmcu *dmcu = core_dc->res_pool->dmcu; + struct dc *dc = clk_mgr_base->ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; int actual_clock = requested_clk_khz; /* Prepare to program display clock*/ memset(&dce_clk_params, 0, sizeof(dce_clk_params)); @@ -81,7 +81,7 @@ int dce112_set_clock(struct clk_mgr *clk_mgr_base, int requested_clk_khz) /* Make sure requested clock isn't lower than minimum threshold*/ if (requested_clk_khz > 0) requested_clk_khz = max(requested_clk_khz, - clk_mgr_dce->dentist_vco_freq_khz / 62); + clk_mgr_dce->base.dentist_vco_freq_khz / 62); dce_clk_params.target_clock_frequency = requested_clk_khz; dce_clk_params.pll_id = CLOCK_SOURCE_ID_DFS; @@ -110,7 +110,7 @@ int dce112_set_clock(struct clk_mgr *clk_mgr_base, int requested_clk_khz) bp->funcs->set_dce_clock(bp, &dce_clk_params); - if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { if (clk_mgr_dce->dfs_bypass_disp_clk != actual_clock) dmcu->funcs->set_psr_wait_loop(dmcu, @@ -126,8 +126,8 @@ int dce112_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_clk_khz) { struct bp_set_dce_clock_parameters dce_clk_params; struct dc_bios *bp = clk_mgr->base.ctx->dc_bios; - struct dc *core_dc = clk_mgr->base.ctx->dc; - struct dmcu *dmcu = core_dc->res_pool->dmcu; + struct dc *dc = clk_mgr->base.ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; int actual_clock = requested_clk_khz; /* Prepare to program display clock*/ memset(&dce_clk_params, 0, sizeof(dce_clk_params)); @@ -135,7 +135,7 @@ int dce112_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_clk_khz) /* Make sure requested clock isn't lower than minimum threshold*/ if (requested_clk_khz > 0) requested_clk_khz = max(requested_clk_khz, - clk_mgr->dentist_vco_freq_khz / 62); + clk_mgr->base.dentist_vco_freq_khz / 62); dce_clk_params.target_clock_frequency = requested_clk_khz; dce_clk_params.pll_id = CLOCK_SOURCE_ID_DFS; @@ -152,7 +152,7 @@ int dce112_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_clk_khz) clk_mgr->cur_min_clks_state = DM_PP_CLOCKS_STATE_NOMINAL; - if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { if (clk_mgr->dfs_bypass_disp_clk != actual_clock) dmcu->funcs->set_psr_wait_loop(dmcu, diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c index caf8a4a4e442..3fab9296918a 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c @@ -34,6 +34,11 @@ #include "rv1_clk_mgr_vbios_smu.h" #include "rv1_clk_mgr_clk.h" +void rv1_init_clocks(struct clk_mgr *clk_mgr) +{ + memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); +} + static int rv1_determine_dppclk_threshold(struct clk_mgr_internal *clk_mgr, struct dc_clocks *new_clocks) { bool request_dpp_div = new_clocks->dispclk_khz > new_clocks->dppclk_khz; @@ -134,6 +139,9 @@ static void rv1_update_clocks(struct clk_mgr *clk_mgr_base, ASSERT(clk_mgr->pp_smu); + if (dc->work_arounds.skip_clock_update) + return; + pp_smu = &clk_mgr->pp_smu->rv_funcs; display_count = clk_mgr_helper_get_active_display_cnt(dc, context); @@ -232,6 +240,7 @@ static void rv1_enable_pme_wa(struct clk_mgr *clk_mgr_base) } static struct clk_mgr_funcs rv1_clk_funcs = { + .init_clocks = rv1_init_clocks, .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, .update_clocks = rv1_update_clocks, .enable_pme_wa = rv1_enable_pme_wa, @@ -246,7 +255,6 @@ void rv1_clk_mgr_construct(struct dc_context *ctx, struct clk_mgr_internal *clk_ { struct dc_debug_options *debug = &ctx->dc->debug; struct dc_bios *bp = ctx->dc_bios; - struct dc_firmware_info fw_info = { { 0 } }; clk_mgr->base.ctx = ctx; clk_mgr->pp_smu = pp_smu; @@ -261,12 +269,11 @@ void rv1_clk_mgr_construct(struct dc_context *ctx, struct clk_mgr_internal *clk_ clk_mgr->base.dprefclk_khz = 600000; if (bp->integrated_info) - clk_mgr->dentist_vco_freq_khz = bp->integrated_info->dentist_vco_freq; - if (clk_mgr->dentist_vco_freq_khz == 0) { - bp->funcs->get_firmware_info(bp, &fw_info); - clk_mgr->dentist_vco_freq_khz = fw_info.smu_gpu_pll_output_freq; - if (clk_mgr->dentist_vco_freq_khz == 0) - clk_mgr->dentist_vco_freq_khz = 3600000; + clk_mgr->base.dentist_vco_freq_khz = bp->integrated_info->dentist_vco_freq; + if (bp->fw_info_valid && clk_mgr->base.dentist_vco_freq_khz == 0) { + clk_mgr->base.dentist_vco_freq_khz = bp->fw_info.smu_gpu_pll_output_freq; + if (clk_mgr->base.dentist_vco_freq_khz == 0) + clk_mgr->base.dentist_vco_freq_khz = 3600000; } if (!debug->disable_dfs_bypass && bp->integrated_info) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c index 1897e91c8ccb..97b7f32294fd 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c @@ -88,8 +88,8 @@ int rv1_vbios_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, unsigned int rv1_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dispclk_khz) { int actual_dispclk_set_mhz = -1; - struct dc *core_dc = clk_mgr->base.ctx->dc; - struct dmcu *dmcu = core_dc->res_pool->dmcu; + struct dc *dc = clk_mgr->base.ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; /* Unit of SMU msg parameter is Mhz */ actual_dispclk_set_mhz = rv1_vbios_smu_send_msg_with_param( @@ -100,7 +100,7 @@ int rv1_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_di /* Actual dispclk set is returned in the parameter register */ actual_dispclk_set_mhz = REG_READ(MP1_SMN_C2PMSG_83) * 1000; - if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz) dmcu->funcs->set_psr_wait_loop(dmcu, diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c index 50bfb5921de0..49ce46b543ea 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c @@ -26,9 +26,8 @@ #include "dccg.h" #include "clk_mgr_internal.h" - -#include "dcn20/dcn20_clk_mgr.h" #include "dce100/dce_clk_mgr.h" +#include "dcn20_clk_mgr.h" #include "reg_helper.h" #include "core_types.h" #include "dm_helpers.h" @@ -102,30 +101,35 @@ uint32_t dentist_get_did_from_divider(int divider) } void dcn20_update_clocks_update_dpp_dto(struct clk_mgr_internal *clk_mgr, - struct dc_state *context) + struct dc_state *context, bool safe_to_lower) { int i; clk_mgr->dccg->ref_dppclk = clk_mgr->base.clks.dppclk_khz; for (i = 0; i < clk_mgr->base.ctx->dc->res_pool->pipe_count; i++) { - int dpp_inst, dppclk_khz; - - if (!context->res_ctx.pipe_ctx[i].plane_state) - continue; + int dpp_inst, dppclk_khz, prev_dppclk_khz; - dpp_inst = context->res_ctx.pipe_ctx[i].plane_res.dpp->inst; + /* Loop index will match dpp->inst if resource exists, + * and we want to avoid dependency on dpp object + */ + dpp_inst = i; dppclk_khz = context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; - clk_mgr->dccg->funcs->update_dpp_dto( - clk_mgr->dccg, dpp_inst, dppclk_khz); + + prev_dppclk_khz = clk_mgr->base.ctx->dc->current_state->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; + + if ((prev_dppclk_khz > dppclk_khz && safe_to_lower) || prev_dppclk_khz < dppclk_khz) { + clk_mgr->dccg->funcs->update_dpp_dto( + clk_mgr->dccg, dpp_inst, dppclk_khz); + } } } void dcn20_update_clocks_update_dentist(struct clk_mgr_internal *clk_mgr) { int dpp_divider = DENTIST_DIVIDER_RANGE_SCALE_FACTOR - * clk_mgr->dentist_vco_freq_khz / clk_mgr->base.clks.dppclk_khz; + * clk_mgr->base.dentist_vco_freq_khz / clk_mgr->base.clks.dppclk_khz; int disp_divider = DENTIST_DIVIDER_RANGE_SCALE_FACTOR - * clk_mgr->dentist_vco_freq_khz / clk_mgr->base.clks.dispclk_khz; + * clk_mgr->base.dentist_vco_freq_khz / clk_mgr->base.clks.dispclk_khz; uint32_t dppclk_wdivider = dentist_get_did_from_divider(dpp_divider); uint32_t dispclk_wdivider = dentist_get_did_from_divider(disp_divider); @@ -153,7 +157,20 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, bool enter_display_off = false; bool dpp_clock_lowered = false; struct dmcu *dmcu = clk_mgr_base->ctx->dc->res_pool->dmcu; + bool force_reset = false; + + if (dc->work_arounds.skip_clock_update) + return; + + if (clk_mgr_base->clks.dispclk_khz == 0 || + dc->debug.force_clock_mode & 0x1) { + //this is from resume or boot up, if forced_clock cfg option used, we bypass program dispclk and DPPCLK, but need set them for S3. + force_reset = true; + + dcn2_read_clocks_from_hw_dentist(clk_mgr_base); + //force_clock_mode 0x1: force reset the clock even it is the same clock as long as it is in Passive level. + } display_count = clk_mgr_helper_get_active_display_cnt(dc, context); if (dc->res_pool->pp_smu) pp_smu = &dc->res_pool->pp_smu->nv_funcs; @@ -172,6 +189,7 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, pp_smu->set_voltage_by_freq(&pp_smu->pp_smu, PP_SMU_NV_PHYCLK, clk_mgr_base->clks.phyclk_khz / 1000); } + if (dc->debug.force_min_dcfclk_mhz > 0) new_clocks->dcfclk_khz = (new_clocks->dcfclk_khz > (dc->debug.force_min_dcfclk_mhz * 1000)) ? new_clocks->dcfclk_khz : (dc->debug.force_min_dcfclk_mhz * 1000); @@ -196,6 +214,7 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, } if (should_update_pstate_support(safe_to_lower, new_clocks->p_state_change_support, clk_mgr_base->clks.p_state_change_support)) { + clk_mgr_base->clks.prev_p_state_change_support = clk_mgr_base->clks.p_state_change_support; clk_mgr_base->clks.p_state_change_support = new_clocks->p_state_change_support; if (pp_smu && pp_smu->set_pstate_handshake_support) pp_smu->set_pstate_handshake_support(&pp_smu->pp_smu, clk_mgr_base->clks.p_state_change_support); @@ -225,19 +244,22 @@ void dcn2_update_clocks(struct clk_mgr *clk_mgr_base, update_dispclk = true; } - if (dc->config.forced_clocks == false) { + + if (dc->config.forced_clocks == false || (force_reset && safe_to_lower)) { if (dpp_clock_lowered) { // if clock is being lowered, increase DTO before lowering refclk - dcn20_update_clocks_update_dpp_dto(clk_mgr, context); + dcn20_update_clocks_update_dpp_dto(clk_mgr, context, safe_to_lower); dcn20_update_clocks_update_dentist(clk_mgr); } else { // if clock is being raised, increase refclk before lowering DTO if (update_dppclk || update_dispclk) dcn20_update_clocks_update_dentist(clk_mgr); - if (update_dppclk) - dcn20_update_clocks_update_dpp_dto(clk_mgr, context); + // always update dtos unless clock is lowered and not safe to lower + if (new_clocks->dppclk_khz >= dc->current_state->bw_ctx.bw.dcn.clk.dppclk_khz) + dcn20_update_clocks_update_dpp_dto(clk_mgr, context, safe_to_lower); } } + if (update_dispclk && dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { /*update dmcu for wait_loop count*/ @@ -250,6 +272,8 @@ void dcn2_update_clocks_fpga(struct clk_mgr *clk_mgr, struct dc_state *context, bool safe_to_lower) { + struct clk_mgr_internal *clk_mgr_int = TO_CLK_MGR_INTERNAL(clk_mgr); + struct dc_clocks *new_clocks = &context->bw_ctx.bw.dcn.clk; /* Min fclk = 1.2GHz since all the extra scemi logic seems to run off of it */ int fclk_adj = new_clocks->fclk_khz > 1200000 ? new_clocks->fclk_khz : 1200000; @@ -287,14 +311,18 @@ void dcn2_update_clocks_fpga(struct clk_mgr *clk_mgr, clk_mgr->clks.dispclk_khz = new_clocks->dispclk_khz; } - /* Both fclk and dppclk ref are run on the same scemi clock so we - * need to keep the same value for both + /* Both fclk and ref_dppclk run on the same scemi clock. + * So take the higher value since the DPP DTO is typically programmed + * such that max dppclk is 1:1 with ref_dppclk. */ if (clk_mgr->clks.fclk_khz > clk_mgr->clks.dppclk_khz) clk_mgr->clks.dppclk_khz = clk_mgr->clks.fclk_khz; if (clk_mgr->clks.dppclk_khz > clk_mgr->clks.fclk_khz) clk_mgr->clks.fclk_khz = clk_mgr->clks.dppclk_khz; + // Both fclk and ref_dppclk run on the same scemi clock. + clk_mgr_int->dccg->ref_dppclk = clk_mgr->clks.fclk_khz; + dm_set_dcn_clocks(clk_mgr->ctx, &clk_mgr->clks); } @@ -303,6 +331,7 @@ void dcn2_init_clocks(struct clk_mgr *clk_mgr) memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); // Assumption is that boot state always supports pstate clk_mgr->clks.p_state_change_support = true; + clk_mgr->clks.prev_p_state_change_support = true; } void dcn2_enable_pme_wa(struct clk_mgr *clk_mgr_base) @@ -318,11 +347,82 @@ void dcn2_enable_pme_wa(struct clk_mgr *clk_mgr_base) } } + +void dcn2_read_clocks_from_hw_dentist(struct clk_mgr *clk_mgr_base) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + uint32_t dispclk_wdivider; + uint32_t dppclk_wdivider; + int disp_divider; + int dpp_divider; + + REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, &dispclk_wdivider); + REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DPPCLK_WDIVIDER, &dppclk_wdivider); + + disp_divider = dentist_get_divider_from_did(dispclk_wdivider); + dpp_divider = dentist_get_divider_from_did(dispclk_wdivider); + + if (disp_divider && dpp_divider) { + /* Calculate the current DFS clock, in kHz.*/ + clk_mgr_base->clks.dispclk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR + * clk_mgr->base.dentist_vco_freq_khz) / disp_divider; + + clk_mgr_base->clks.dppclk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR + * clk_mgr->base.dentist_vco_freq_khz) / dpp_divider; + } + +} + +void dcn2_get_clock(struct clk_mgr *clk_mgr, + struct dc_state *context, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg) +{ + + if (clock_type == DC_CLOCK_TYPE_DISPCLK) { + clock_cfg->max_clock_khz = context->bw_ctx.bw.dcn.clk.max_supported_dispclk_khz; + clock_cfg->min_clock_khz = DCN_MINIMUM_DISPCLK_Khz; + clock_cfg->current_clock_khz = clk_mgr->clks.dispclk_khz; + clock_cfg->bw_requirequired_clock_khz = context->bw_ctx.bw.dcn.clk.bw_dispclk_khz; + } + if (clock_type == DC_CLOCK_TYPE_DPPCLK) { + clock_cfg->max_clock_khz = context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz; + clock_cfg->min_clock_khz = DCN_MINIMUM_DPPCLK_Khz; + clock_cfg->current_clock_khz = clk_mgr->clks.dppclk_khz; + clock_cfg->bw_requirequired_clock_khz = context->bw_ctx.bw.dcn.clk.bw_dppclk_khz; + } +} + +static bool dcn2_are_clock_states_equal(struct dc_clocks *a, + struct dc_clocks *b) +{ + if (a->dispclk_khz != b->dispclk_khz) + return false; + else if (a->dppclk_khz != b->dppclk_khz) + return false; + else if (a->dcfclk_khz != b->dcfclk_khz) + return false; + else if (a->socclk_khz != b->socclk_khz) + return false; + else if (a->dcfclk_deep_sleep_khz != b->dcfclk_deep_sleep_khz) + return false; + else if (a->phyclk_khz != b->phyclk_khz) + return false; + else if (a->dramclk_khz != b->dramclk_khz) + return false; + else if (a->p_state_change_support != b->p_state_change_support) + return false; + + return true; +} + static struct clk_mgr_funcs dcn2_funcs = { .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, .update_clocks = dcn2_update_clocks, .init_clocks = dcn2_init_clocks, - .enable_pme_wa = dcn2_enable_pme_wa + .enable_pme_wa = dcn2_enable_pme_wa, + .get_clock = dcn2_get_clock, + .are_clock_states_equal = dcn2_are_clock_states_equal, }; @@ -350,7 +450,7 @@ void dcn20_clk_mgr_construct( if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { dcn2_funcs.update_clocks = dcn2_update_clocks_fpga; - clk_mgr->dentist_vco_freq_khz = 3850000; + clk_mgr->base.dentist_vco_freq_khz = 3850000; } else { /* DFS Slice 2 should be used for DPREFCLK */ @@ -374,15 +474,15 @@ void dcn20_clk_mgr_construct( pll_req = dc_fixpt_mul_int(pll_req, 100000); /* integer part is now VCO frequency in kHz */ - clk_mgr->dentist_vco_freq_khz = dc_fixpt_floor(pll_req); + clk_mgr->base.dentist_vco_freq_khz = dc_fixpt_floor(pll_req); /* in case we don't get a value from the register, use default */ - if (clk_mgr->dentist_vco_freq_khz == 0) - clk_mgr->dentist_vco_freq_khz = 3850000; + if (clk_mgr->base.dentist_vco_freq_khz == 0) + clk_mgr->base.dentist_vco_freq_khz = 3850000; /* Calculate the DPREFCLK in kHz.*/ clk_mgr->base.dprefclk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR - * clk_mgr->dentist_vco_freq_khz) / target_div; + * clk_mgr->base.dentist_vco_freq_khz) / target_div; } //Integrated_info table does not exist on dGPU projects so should not be referenced //anywhere in code for dGPUs. diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h index 5661a5a89847..0b9c045b0c8e 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.h @@ -34,7 +34,7 @@ void dcn2_update_clocks_fpga(struct clk_mgr *clk_mgr, struct dc_state *context, bool safe_to_lower); void dcn20_update_clocks_update_dpp_dto(struct clk_mgr_internal *clk_mgr, - struct dc_state *context); + struct dc_state *context, bool safe_to_lower); void dcn2_init_clocks(struct clk_mgr *clk_mgr); @@ -45,4 +45,14 @@ void dcn20_clk_mgr_construct(struct dc_context *ctx, uint32_t dentist_get_did_from_divider(int divider); +void dcn2_get_clock(struct clk_mgr *clk_mgr, + struct dc_state *context, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg); + +void dcn20_update_clocks_update_dentist(struct clk_mgr_internal *clk_mgr); + +void dcn2_read_clocks_from_hw_dentist(struct clk_mgr *clk_mgr_base); + + #endif //__DCN20_CLK_MGR_H__ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c new file mode 100644 index 000000000000..9ef3f7b91a1d --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -0,0 +1,758 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dccg.h" +#include "clk_mgr_internal.h" + + +#include "dcn20/dcn20_clk_mgr.h" +#include "rn_clk_mgr.h" + + +#include "dce100/dce_clk_mgr.h" +#include "rn_clk_mgr_vbios_smu.h" +#include "reg_helper.h" +#include "core_types.h" +#include "dm_helpers.h" + +#include "atomfirmware.h" +#include "clk/clk_10_0_2_offset.h" +#include "clk/clk_10_0_2_sh_mask.h" +#include "renoir_ip_offset.h" + + +/* Constants */ + +#define LPDDR_MEM_RETRAIN_LATENCY 4.977 /* Number obtained from LPDDR4 Training Counter Requirement doc */ + +/* Macros */ + +#define REG(reg_name) \ + (CLK_BASE.instance[0].segment[mm ## reg_name ## _BASE_IDX] + mm ## reg_name) + + +/* TODO: evaluate how to lower or disable all dcn clocks in screen off case */ +int rn_get_active_display_cnt_wa( + struct dc *dc, + struct dc_state *context) +{ + int i, display_count; + bool tmds_present = false; + + display_count = 0; + for (i = 0; i < context->stream_count; i++) { + const struct dc_stream_state *stream = context->streams[i]; + + if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A || + stream->signal == SIGNAL_TYPE_DVI_SINGLE_LINK || + stream->signal == SIGNAL_TYPE_DVI_DUAL_LINK) + tmds_present = true; + } + + for (i = 0; i < dc->link_count; i++) { + const struct dc_link *link = dc->links[i]; + + /* + * Only notify active stream or virtual stream. + * Need to notify virtual stream to work around + * headless case. HPD does not fire when system is in + * S0i2. + */ + /* abusing the fact that the dig and phy are coupled to see if the phy is enabled */ + if (link->connector_signal == SIGNAL_TYPE_VIRTUAL || + link->link_enc->funcs->is_dig_enabled(link->link_enc)) + display_count++; + } + + /* WA for hang on HDMI after display off back back on*/ + if (display_count == 0 && tmds_present) + display_count = 1; + + return display_count; +} + +void rn_update_clocks(struct clk_mgr *clk_mgr_base, + struct dc_state *context, + bool safe_to_lower) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + struct dc_clocks *new_clocks = &context->bw_ctx.bw.dcn.clk; + struct dc *dc = clk_mgr_base->ctx->dc; + int display_count; + bool update_dppclk = false; + bool update_dispclk = false; + bool dpp_clock_lowered = false; + + struct dmcu *dmcu = clk_mgr_base->ctx->dc->res_pool->dmcu; + + if (dc->work_arounds.skip_clock_update) + return; + + /* + * if it is safe to lower, but we are already in the lower state, we don't have to do anything + * also if safe to lower is false, we just go in the higher state + */ + if (safe_to_lower) { + /* check that we're not already in lower */ + if (clk_mgr_base->clks.pwr_state != DCN_PWR_STATE_LOW_POWER) { + + display_count = rn_get_active_display_cnt_wa(dc, context); + /* if we can go lower, go lower */ + if (display_count == 0) { + rn_vbios_smu_set_dcn_low_power_state(clk_mgr, DCN_PWR_STATE_LOW_POWER); + /* update power state */ + clk_mgr_base->clks.pwr_state = DCN_PWR_STATE_LOW_POWER; + } + } + } else { + /* check that we're not already in D0 */ + if (clk_mgr_base->clks.pwr_state != DCN_PWR_STATE_MISSION_MODE) { + rn_vbios_smu_set_dcn_low_power_state(clk_mgr, DCN_PWR_STATE_MISSION_MODE); + /* update power state */ + clk_mgr_base->clks.pwr_state = DCN_PWR_STATE_MISSION_MODE; + } + } + + if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, clk_mgr_base->clks.phyclk_khz)) { + clk_mgr_base->clks.phyclk_khz = new_clocks->phyclk_khz; + rn_vbios_smu_set_phyclk(clk_mgr, clk_mgr_base->clks.phyclk_khz); + } + + if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr_base->clks.dcfclk_khz)) { + clk_mgr_base->clks.dcfclk_khz = new_clocks->dcfclk_khz; + rn_vbios_smu_set_hard_min_dcfclk(clk_mgr, clk_mgr_base->clks.dcfclk_khz); + } + + if (should_set_clock(safe_to_lower, + new_clocks->dcfclk_deep_sleep_khz, clk_mgr_base->clks.dcfclk_deep_sleep_khz)) { + clk_mgr_base->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz; + rn_vbios_smu_set_min_deep_sleep_dcfclk(clk_mgr, clk_mgr_base->clks.dcfclk_deep_sleep_khz); + } + + // workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow. + if (!IS_DIAG_DC(dc->ctx->dce_environment)) { + if (new_clocks->dppclk_khz < 100000) + new_clocks->dppclk_khz = 100000; + } + + if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) { + if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz) + dpp_clock_lowered = true; + clk_mgr_base->clks.dppclk_khz = new_clocks->dppclk_khz; + update_dppclk = true; + } + + if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz)) { + clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz; + rn_vbios_smu_set_dispclk(clk_mgr, clk_mgr_base->clks.dispclk_khz); + + update_dispclk = true; + } + + if (dpp_clock_lowered) { + // increase per DPP DTO before lowering global dppclk + dcn20_update_clocks_update_dpp_dto(clk_mgr, context, safe_to_lower); + rn_vbios_smu_set_dppclk(clk_mgr, clk_mgr_base->clks.dppclk_khz); + } else { + // increase global DPPCLK before lowering per DPP DTO + if (update_dppclk || update_dispclk) + rn_vbios_smu_set_dppclk(clk_mgr, clk_mgr_base->clks.dppclk_khz); + // always update dtos unless clock is lowered and not safe to lower + if (new_clocks->dppclk_khz >= dc->current_state->bw_ctx.bw.dcn.clk.dppclk_khz) + dcn20_update_clocks_update_dpp_dto(clk_mgr, context, safe_to_lower); + } + + if (update_dispclk && + dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { + /*update dmcu for wait_loop count*/ + dmcu->funcs->set_psr_wait_loop(dmcu, + clk_mgr_base->clks.dispclk_khz / 1000 / 7); + } +} + + +static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr) +{ + /* get FbMult value */ + struct fixed31_32 pll_req; + unsigned int fbmult_frac_val = 0; + unsigned int fbmult_int_val = 0; + + + /* + * Register value of fbmult is in 8.16 format, we are converting to 31.32 + * to leverage the fix point operations available in driver + */ + + REG_GET(CLK1_CLK_PLL_REQ, FbMult_frac, &fbmult_frac_val); /* 16 bit fractional part*/ + REG_GET(CLK1_CLK_PLL_REQ, FbMult_int, &fbmult_int_val); /* 8 bit integer part */ + + pll_req = dc_fixpt_from_int(fbmult_int_val); + + /* + * since fractional part is only 16 bit in register definition but is 32 bit + * in our fix point definiton, need to shift left by 16 to obtain correct value + */ + pll_req.value |= fbmult_frac_val << 16; + + /* multiply by REFCLK period */ + pll_req = dc_fixpt_mul_int(pll_req, clk_mgr->dfs_ref_freq_khz); + + /* integer part is now VCO frequency in kHz */ + return dc_fixpt_floor(pll_req); +} + +static void rn_dump_clk_registers_internal(struct rn_clk_internal *internal, struct clk_mgr *clk_mgr_base) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + + internal->CLK1_CLK3_CURRENT_CNT = REG_READ(CLK1_CLK3_CURRENT_CNT); + internal->CLK1_CLK3_BYPASS_CNTL = REG_READ(CLK1_CLK3_BYPASS_CNTL); + + internal->CLK1_CLK3_DS_CNTL = REG_READ(CLK1_CLK3_DS_CNTL); //dcf deep sleep divider + internal->CLK1_CLK3_ALLOW_DS = REG_READ(CLK1_CLK3_ALLOW_DS); + + internal->CLK1_CLK1_CURRENT_CNT = REG_READ(CLK1_CLK1_CURRENT_CNT); + internal->CLK1_CLK1_BYPASS_CNTL = REG_READ(CLK1_CLK1_BYPASS_CNTL); + + internal->CLK1_CLK2_CURRENT_CNT = REG_READ(CLK1_CLK2_CURRENT_CNT); + internal->CLK1_CLK2_BYPASS_CNTL = REG_READ(CLK1_CLK2_BYPASS_CNTL); + + internal->CLK1_CLK0_CURRENT_CNT = REG_READ(CLK1_CLK0_CURRENT_CNT); + internal->CLK1_CLK0_BYPASS_CNTL = REG_READ(CLK1_CLK0_BYPASS_CNTL); +} + +/* This function collect raw clk register values */ +static void rn_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass, + struct clk_mgr *clk_mgr_base, struct clk_log_info *log_info) +{ + struct rn_clk_internal internal = {0}; + char *bypass_clks[5] = {"0x0 DFS", "0x1 REFCLK", "0x2 ERROR", "0x3 400 FCH", "0x4 600 FCH"}; + unsigned int chars_printed = 0; + unsigned int remaining_buffer = log_info->bufSize; + + rn_dump_clk_registers_internal(&internal, clk_mgr_base); + + regs_and_bypass->dcfclk = internal.CLK1_CLK3_CURRENT_CNT / 10; + regs_and_bypass->dcf_deep_sleep_divider = internal.CLK1_CLK3_DS_CNTL / 10; + regs_and_bypass->dcf_deep_sleep_allow = internal.CLK1_CLK3_ALLOW_DS; + regs_and_bypass->dprefclk = internal.CLK1_CLK2_CURRENT_CNT / 10; + regs_and_bypass->dispclk = internal.CLK1_CLK0_CURRENT_CNT / 10; + regs_and_bypass->dppclk = internal.CLK1_CLK1_CURRENT_CNT / 10; + + regs_and_bypass->dppclk_bypass = internal.CLK1_CLK1_BYPASS_CNTL & 0x0007; + if (regs_and_bypass->dppclk_bypass < 0 || regs_and_bypass->dppclk_bypass > 4) + regs_and_bypass->dppclk_bypass = 0; + regs_and_bypass->dcfclk_bypass = internal.CLK1_CLK3_BYPASS_CNTL & 0x0007; + if (regs_and_bypass->dcfclk_bypass < 0 || regs_and_bypass->dcfclk_bypass > 4) + regs_and_bypass->dcfclk_bypass = 0; + regs_and_bypass->dispclk_bypass = internal.CLK1_CLK0_BYPASS_CNTL & 0x0007; + if (regs_and_bypass->dispclk_bypass < 0 || regs_and_bypass->dispclk_bypass > 4) + regs_and_bypass->dispclk_bypass = 0; + regs_and_bypass->dprefclk_bypass = internal.CLK1_CLK2_BYPASS_CNTL & 0x0007; + if (regs_and_bypass->dprefclk_bypass < 0 || regs_and_bypass->dprefclk_bypass > 4) + regs_and_bypass->dprefclk_bypass = 0; + + if (log_info->enabled) { + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "clk_type,clk_value,deepsleep_cntl,deepsleep_allow,bypass\n"); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "dcfclk,%d,%d,%d,%s\n", + regs_and_bypass->dcfclk, + regs_and_bypass->dcf_deep_sleep_divider, + regs_and_bypass->dcf_deep_sleep_allow, + bypass_clks[(int) regs_and_bypass->dcfclk_bypass]); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "dprefclk,%d,N/A,N/A,%s\n", + regs_and_bypass->dprefclk, + bypass_clks[(int) regs_and_bypass->dprefclk_bypass]); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "dispclk,%d,N/A,N/A,%s\n", + regs_and_bypass->dispclk, + bypass_clks[(int) regs_and_bypass->dispclk_bypass]); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + //split + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "SPLIT\n"); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + // REGISTER VALUES + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "reg_name,value,clk_type\n"); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK3_CURRENT_CNT,%d,dcfclk\n", + internal.CLK1_CLK3_CURRENT_CNT); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK3_DS_CNTL,%d,dcf_deep_sleep_divider\n", + internal.CLK1_CLK3_DS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK3_ALLOW_DS,%d,dcf_deep_sleep_allow\n", + internal.CLK1_CLK3_ALLOW_DS); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK2_CURRENT_CNT,%d,dprefclk\n", + internal.CLK1_CLK2_CURRENT_CNT); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK0_CURRENT_CNT,%d,dispclk\n", + internal.CLK1_CLK0_CURRENT_CNT); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK1_CURRENT_CNT,%d,dppclk\n", + internal.CLK1_CLK1_CURRENT_CNT); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK3_BYPASS_CNTL,%d,dcfclk_bypass\n", + internal.CLK1_CLK3_BYPASS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK2_BYPASS_CNTL,%d,dprefclk_bypass\n", + internal.CLK1_CLK2_BYPASS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK0_BYPASS_CNTL,%d,dispclk_bypass\n", + internal.CLK1_CLK0_BYPASS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + + chars_printed = snprintf_count(log_info->pBuf, remaining_buffer, "CLK1_CLK1_BYPASS_CNTL,%d,dppclk_bypass\n", + internal.CLK1_CLK1_BYPASS_CNTL); + remaining_buffer -= chars_printed; + *log_info->sum_chars_printed += chars_printed; + log_info->pBuf += chars_printed; + } +} + +/* This function produce translated logical clk state values*/ +void rn_get_clk_states(struct clk_mgr *clk_mgr_base, struct clk_states *s) +{ + struct clk_state_registers_and_bypass sb = { 0 }; + struct clk_log_info log_info = { 0 }; + + rn_dump_clk_registers(&sb, clk_mgr_base, &log_info); + + s->dprefclk_khz = sb.dprefclk * 1000; +} + +void rn_enable_pme_wa(struct clk_mgr *clk_mgr_base) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + + rn_vbios_smu_enable_pme_wa(clk_mgr); +} + +void rn_init_clocks(struct clk_mgr *clk_mgr) +{ + memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); + // Assumption is that boot state always supports pstate + clk_mgr->clks.p_state_change_support = true; + clk_mgr->clks.prev_p_state_change_support = true; + clk_mgr->clks.pwr_state = DCN_PWR_STATE_UNKNOWN; +} + +void build_watermark_ranges(struct clk_bw_params *bw_params, struct pp_smu_wm_range_sets *ranges) +{ + int i, num_valid_sets; + + num_valid_sets = 0; + + for (i = 0; i < WM_SET_COUNT; i++) { + /* skip empty entries, the smu array has no holes*/ + if (!bw_params->wm_table.entries[i].valid) + continue; + + ranges->reader_wm_sets[num_valid_sets].wm_inst = bw_params->wm_table.entries[i].wm_inst; + ranges->reader_wm_sets[num_valid_sets].wm_type = bw_params->wm_table.entries[i].wm_type; + /* We will not select WM based on fclk, so leave it as unconstrained */ + ranges->reader_wm_sets[num_valid_sets].min_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->reader_wm_sets[num_valid_sets].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + /* dcfclk wil be used to select WM*/ + + if (ranges->reader_wm_sets[num_valid_sets].wm_type == WM_TYPE_PSTATE_CHG) { + if (i == 0) + ranges->reader_wm_sets[num_valid_sets].min_drain_clk_mhz = 0; + else { + /* add 1 to make it non-overlapping with next lvl */ + ranges->reader_wm_sets[num_valid_sets].min_drain_clk_mhz = bw_params->clk_table.entries[i - 1].dcfclk_mhz + 1; + } + ranges->reader_wm_sets[num_valid_sets].max_drain_clk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; + + } else { + /* unconstrained for memory retraining */ + ranges->reader_wm_sets[num_valid_sets].min_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->reader_wm_sets[num_valid_sets].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + + /* Modify previous watermark range to cover up to max */ + ranges->reader_wm_sets[num_valid_sets - 1].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + } + num_valid_sets++; + } + + ASSERT(num_valid_sets != 0); /* Must have at least one set of valid watermarks */ + ranges->num_reader_wm_sets = num_valid_sets; + + /* modify the min and max to make sure we cover the whole range*/ + ranges->reader_wm_sets[0].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->reader_wm_sets[0].min_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->reader_wm_sets[ranges->num_reader_wm_sets - 1].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + ranges->reader_wm_sets[ranges->num_reader_wm_sets - 1].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + + /* This is for writeback only, does not matter currently as no writeback support*/ + ranges->num_writer_wm_sets = 1; + ranges->writer_wm_sets[0].wm_inst = WM_A; + ranges->writer_wm_sets[0].min_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->writer_wm_sets[0].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + ranges->writer_wm_sets[0].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; + ranges->writer_wm_sets[0].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; + +} + +static void rn_notify_wm_ranges(struct clk_mgr *clk_mgr_base) +{ + struct dc_debug_options *debug = &clk_mgr_base->ctx->dc->debug; + struct pp_smu_wm_range_sets ranges = {0}; + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + struct pp_smu_funcs *pp_smu = clk_mgr->pp_smu; + + if (!debug->disable_pplib_wm_range) { + build_watermark_ranges(clk_mgr_base->bw_params, &ranges); + + /* Notify PP Lib/SMU which Watermarks to use for which clock ranges */ + if (pp_smu && pp_smu->rn_funcs.set_wm_ranges) + pp_smu->rn_funcs.set_wm_ranges(&pp_smu->rn_funcs.pp_smu, &ranges); + } + +} + +static bool rn_are_clock_states_equal(struct dc_clocks *a, + struct dc_clocks *b) +{ + if (a->dispclk_khz != b->dispclk_khz) + return false; + else if (a->dppclk_khz != b->dppclk_khz) + return false; + else if (a->dcfclk_khz != b->dcfclk_khz) + return false; + else if (a->dcfclk_deep_sleep_khz != b->dcfclk_deep_sleep_khz) + return false; + + return true; +} + + +static struct clk_mgr_funcs dcn21_funcs = { + .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, + .update_clocks = rn_update_clocks, + .init_clocks = rn_init_clocks, + .enable_pme_wa = rn_enable_pme_wa, + .are_clock_states_equal = rn_are_clock_states_equal, + .notify_wm_ranges = rn_notify_wm_ranges +}; + +struct clk_bw_params rn_bw_params = { + .vram_type = Ddr4MemType, + .num_channels = 1, + .clk_table = { + .entries = { + { + .voltage = 0, + .dcfclk_mhz = 400, + .fclk_mhz = 400, + .memclk_mhz = 800, + .socclk_mhz = 0, + }, + { + .voltage = 0, + .dcfclk_mhz = 483, + .fclk_mhz = 800, + .memclk_mhz = 1600, + .socclk_mhz = 0, + }, + { + .voltage = 0, + .dcfclk_mhz = 602, + .fclk_mhz = 1067, + .memclk_mhz = 1067, + .socclk_mhz = 0, + }, + { + .voltage = 0, + .dcfclk_mhz = 738, + .fclk_mhz = 1333, + .memclk_mhz = 1600, + .socclk_mhz = 0, + }, + }, + + .num_entries = 4, + }, + +}; + +struct wm_table ddr4_wm_table = { + .entries = { + { + .wm_inst = WM_A, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.72, + .sr_exit_time_us = 6.09, + .sr_enter_plus_exit_time_us = 7.14, + .valid = true, + }, + { + .wm_inst = WM_B, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.72, + .sr_exit_time_us = 10.12, + .sr_enter_plus_exit_time_us = 11.48, + .valid = true, + }, + { + .wm_inst = WM_C, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.72, + .sr_exit_time_us = 10.12, + .sr_enter_plus_exit_time_us = 11.48, + .valid = true, + }, + { + .wm_inst = WM_D, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.72, + .sr_exit_time_us = 10.12, + .sr_enter_plus_exit_time_us = 11.48, + .valid = true, + }, + } +}; + +struct wm_table lpddr4_wm_table = { + .entries = { + { + .wm_inst = WM_A, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.65333, + .sr_exit_time_us = 5.32, + .sr_enter_plus_exit_time_us = 6.38, + .valid = true, + }, + { + .wm_inst = WM_B, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.65333, + .sr_exit_time_us = 9.82, + .sr_enter_plus_exit_time_us = 11.196, + .valid = true, + }, + { + .wm_inst = WM_C, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.65333, + .sr_exit_time_us = 9.89, + .sr_enter_plus_exit_time_us = 11.24, + .valid = true, + }, + { + .wm_inst = WM_D, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.65333, + .sr_exit_time_us = 9.748, + .sr_enter_plus_exit_time_us = 11.102, + .valid = true, + }, + } +}; + + +static unsigned int find_dcfclk_for_voltage(struct dpm_clocks *clock_table, unsigned int voltage) +{ + int i; + + for (i = 0; i < PP_SMU_NUM_DCFCLK_DPM_LEVELS; i++) { + if (clock_table->DcfClocks[i].Vol == voltage) + return clock_table->DcfClocks[i].Freq; + } + + ASSERT(0); + return 0; +} + +static void rn_clk_mgr_helper_populate_bw_params(struct clk_bw_params *bw_params, struct dpm_clocks *clock_table, struct integrated_info *bios_info) +{ + int i, j = 0; + + j = -1; + + ASSERT(PP_SMU_NUM_FCLK_DPM_LEVELS <= MAX_NUM_DPM_LVL); + + /* Find lowest DPM, FCLK is filled in reverse order*/ + + for (i = PP_SMU_NUM_FCLK_DPM_LEVELS - 1; i >= 0; i--) { + if (clock_table->FClocks[i].Freq != 0) { + j = i; + break; + } + } + + if (j == -1) { + /* clock table is all 0s, just use our own hardcode */ + ASSERT(0); + return; + } + + bw_params->clk_table.num_entries = j + 1; + + for (i = 0; i < bw_params->clk_table.num_entries; i++, j--) { + bw_params->clk_table.entries[i].fclk_mhz = clock_table->FClocks[j].Freq; + bw_params->clk_table.entries[i].memclk_mhz = clock_table->MemClocks[j].Freq; + bw_params->clk_table.entries[i].voltage = clock_table->FClocks[j].Vol; + bw_params->clk_table.entries[i].dcfclk_mhz = find_dcfclk_for_voltage(clock_table, clock_table->FClocks[j].Vol); + } + + bw_params->vram_type = bios_info->memory_type; + bw_params->num_channels = bios_info->ma_channel_number; + + for (i = 0; i < WM_SET_COUNT; i++) { + bw_params->wm_table.entries[i].wm_inst = i; + + if (i >= bw_params->clk_table.num_entries) { + bw_params->wm_table.entries[i].valid = false; + continue; + } + + bw_params->wm_table.entries[i].wm_type = WM_TYPE_PSTATE_CHG; + bw_params->wm_table.entries[i].valid = true; + } + + if (bw_params->vram_type == LpDdr4MemType) { + /* + * WM set D will be re-purposed for memory retraining + */ + bw_params->wm_table.entries[WM_D].pstate_latency_us = LPDDR_MEM_RETRAIN_LATENCY; + bw_params->wm_table.entries[WM_D].wm_inst = WM_D; + bw_params->wm_table.entries[WM_D].wm_type = WM_TYPE_RETRAINING; + bw_params->wm_table.entries[WM_D].valid = true; + } + +} + +void rn_clk_mgr_construct( + struct dc_context *ctx, + struct clk_mgr_internal *clk_mgr, + struct pp_smu_funcs *pp_smu, + struct dccg *dccg) +{ + struct dc_debug_options *debug = &ctx->dc->debug; + struct dpm_clocks clock_table = { 0 }; + + clk_mgr->base.ctx = ctx; + clk_mgr->base.funcs = &dcn21_funcs; + + clk_mgr->pp_smu = pp_smu; + + clk_mgr->dccg = dccg; + clk_mgr->dfs_bypass_disp_clk = 0; + + clk_mgr->dprefclk_ss_percentage = 0; + clk_mgr->dprefclk_ss_divider = 1000; + clk_mgr->ss_on_dprefclk = false; + clk_mgr->dfs_ref_freq_khz = 48000; + + clk_mgr->smu_ver = rn_vbios_smu_get_smu_version(clk_mgr); + + if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { + dcn21_funcs.update_clocks = dcn2_update_clocks_fpga; + clk_mgr->base.dentist_vco_freq_khz = 3600000; + } else { + struct clk_log_info log_info = {0}; + + /* TODO: Check we get what we expect during bringup */ + clk_mgr->base.dentist_vco_freq_khz = get_vco_frequency_from_reg(clk_mgr); + + /* in case we don't get a value from the register, use default */ + if (clk_mgr->base.dentist_vco_freq_khz == 0) + clk_mgr->base.dentist_vco_freq_khz = 3600000; + + if (ctx->dc_bios->integrated_info->memory_type == LpDdr4MemType) { + rn_bw_params.wm_table = lpddr4_wm_table; + } else { + rn_bw_params.wm_table = ddr4_wm_table; + } + /* Saved clocks configured at boot for debug purposes */ + rn_dump_clk_registers(&clk_mgr->base.boot_snapshot, &clk_mgr->base, &log_info); + } + + clk_mgr->base.dprefclk_khz = 600000; + dce_clock_read_ss_info(clk_mgr); + + + clk_mgr->base.bw_params = &rn_bw_params; + + if (pp_smu && pp_smu->rn_funcs.get_dpm_clock_table) { + pp_smu->rn_funcs.get_dpm_clock_table(&pp_smu->rn_funcs.pp_smu, &clock_table); + if (ctx->dc_bios && ctx->dc_bios->integrated_info) { + rn_clk_mgr_helper_populate_bw_params (clk_mgr->base.bw_params, &clock_table, ctx->dc_bios->integrated_info); + } + } + + if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment) && clk_mgr->smu_ver >= 0x00371500) { + /* enable powerfeatures when displaycount goes to 0 */ + rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(clk_mgr, !debug->disable_48mhz_pwrdwn); + } +} + diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h new file mode 100644 index 000000000000..e4322fa5475b --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h @@ -0,0 +1,41 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __RN_CLK_MGR_H__ +#define __RN_CLK_MGR_H__ + +#include "clk_mgr.h" +#include "dm_pp_smu.h" + +struct rn_clk_registers { + uint32_t CLK1_CLK0_CURRENT_CNT; /* DPREFCLK */ +}; + +void rn_clk_mgr_construct(struct dc_context *ctx, + struct clk_mgr_internal *clk_mgr, + struct pp_smu_funcs *pp_smu, + struct dccg *dccg); + +#endif //__RN_CLK_MGR_H__ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c new file mode 100644 index 000000000000..6878aedf1d3e --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c @@ -0,0 +1,198 @@ +/* + * Copyright 2012-16 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "core_types.h" +#include "clk_mgr_internal.h" +#include "reg_helper.h" + +#include "renoir_ip_offset.h" + +#include "mp/mp_12_0_0_offset.h" +#include "mp/mp_12_0_0_sh_mask.h" + +#define REG(reg_name) \ + (MP0_BASE.instance[0].segment[mm ## reg_name ## _BASE_IDX] + mm ## reg_name) + +#define FN(reg_name, field) \ + FD(reg_name##__##field) + +#define VBIOSSMC_MSG_TestMessage 0x1 +#define VBIOSSMC_MSG_GetSmuVersion 0x2 +#define VBIOSSMC_MSG_PowerUpGfx 0x3 +#define VBIOSSMC_MSG_SetDispclkFreq 0x4 +#define VBIOSSMC_MSG_SetDprefclkFreq 0x5 +#define VBIOSSMC_MSG_PowerDownGfx 0x6 +#define VBIOSSMC_MSG_SetDppclkFreq 0x7 +#define VBIOSSMC_MSG_SetHardMinDcfclkByFreq 0x8 +#define VBIOSSMC_MSG_SetMinDeepSleepDcfclk 0x9 +#define VBIOSSMC_MSG_SetPhyclkVoltageByFreq 0xA +#define VBIOSSMC_MSG_GetFclkFrequency 0xB +#define VBIOSSMC_MSG_SetDisplayCount 0xC +#define VBIOSSMC_MSG_EnableTmdp48MHzRefclkPwrDown 0xD +#define VBIOSSMC_MSG_UpdatePmeRestore 0xE + +int rn_vbios_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, unsigned int msg_id, unsigned int param) +{ + /* First clear response register */ + REG_WRITE(MP1_SMN_C2PMSG_91, 0); + + /* Set the parameter register for the SMU message, unit is Mhz */ + REG_WRITE(MP1_SMN_C2PMSG_83, param); + + /* Trigger the message transaction by writing the message ID */ + REG_WRITE(MP1_SMN_C2PMSG_67, msg_id); + + REG_WAIT(MP1_SMN_C2PMSG_91, CONTENT, 1, 10, 200000); + + /* Actual dispclk set is returned in the parameter register */ + return REG_READ(MP1_SMN_C2PMSG_83); +} + +int rn_vbios_smu_get_smu_version(struct clk_mgr_internal *clk_mgr) +{ + return rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_GetSmuVersion, + 0); +} + + +int rn_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dispclk_khz) +{ + int actual_dispclk_set_mhz = -1; + struct dc *dc = clk_mgr->base.ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; + + /* Unit of SMU msg parameter is Mhz */ + actual_dispclk_set_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetDispclkFreq, + requested_dispclk_khz / 1000); + + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { + if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz) + dmcu->funcs->set_psr_wait_loop(dmcu, + actual_dispclk_set_mhz / 7); + } + } + + return actual_dispclk_set_mhz * 1000; +} + +int rn_vbios_smu_set_dprefclk(struct clk_mgr_internal *clk_mgr) +{ + int actual_dprefclk_set_mhz = -1; + + actual_dprefclk_set_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetDprefclkFreq, + clk_mgr->base.dprefclk_khz / 1000); + + /* TODO: add code for programing DP DTO, currently this is down by command table */ + + return actual_dprefclk_set_mhz * 1000; +} + +int rn_vbios_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_dcfclk_khz) +{ + int actual_dcfclk_set_mhz = -1; + + if (clk_mgr->smu_ver < 0x370c00) + return actual_dcfclk_set_mhz; + + actual_dcfclk_set_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetHardMinDcfclkByFreq, + requested_dcfclk_khz / 1000); + + return actual_dcfclk_set_mhz * 1000; +} + +int rn_vbios_smu_set_min_deep_sleep_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_min_ds_dcfclk_khz) +{ + int actual_min_ds_dcfclk_mhz = -1; + + if (clk_mgr->smu_ver < 0x370c00) + return actual_min_ds_dcfclk_mhz; + + actual_min_ds_dcfclk_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetMinDeepSleepDcfclk, + requested_min_ds_dcfclk_khz / 1000); + + return actual_min_ds_dcfclk_mhz * 1000; +} + +void rn_vbios_smu_set_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phyclk_khz) +{ + rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetPhyclkVoltageByFreq, + requested_phyclk_khz / 1000); +} + +int rn_vbios_smu_set_dppclk(struct clk_mgr_internal *clk_mgr, int requested_dpp_khz) +{ + int actual_dppclk_set_mhz = -1; + + actual_dppclk_set_mhz = rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetDppclkFreq, + requested_dpp_khz / 1000); + + return actual_dppclk_set_mhz * 1000; +} + +void rn_vbios_smu_set_dcn_low_power_state(struct clk_mgr_internal *clk_mgr, enum dcn_pwr_state state) +{ + int disp_count; + + if (state == DCN_PWR_STATE_LOW_POWER) + disp_count = 0; + else + disp_count = 1; + + rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_SetDisplayCount, + disp_count); +} + +void rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(struct clk_mgr_internal *clk_mgr, bool enable) +{ + rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_EnableTmdp48MHzRefclkPwrDown, + enable); +} + +void rn_vbios_smu_enable_pme_wa(struct clk_mgr_internal *clk_mgr) +{ + rn_vbios_smu_send_msg_with_param( + clk_mgr, + VBIOSSMC_MSG_UpdatePmeRestore, + 0); +} diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h new file mode 100644 index 000000000000..ccc01879c9d4 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h @@ -0,0 +1,40 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef DAL_DC_RN_CLK_MGR_VBIOS_SMU_H_ +#define DAL_DC_RN_CLK_MGR_VBIOS_SMU_H_ + +int rn_vbios_smu_get_smu_version(struct clk_mgr_internal *clk_mgr); +int rn_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dispclk_khz); +int rn_vbios_smu_set_dprefclk(struct clk_mgr_internal *clk_mgr); +int rn_vbios_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_dcfclk_khz); +int rn_vbios_smu_set_min_deep_sleep_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_min_ds_dcfclk_khz); +void rn_vbios_smu_set_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phyclk_khz); +int rn_vbios_smu_set_dppclk(struct clk_mgr_internal *clk_mgr, int requested_dpp_khz); +void rn_vbios_smu_set_dcn_low_power_state(struct clk_mgr_internal *clk_mgr, int display_count); +void rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(struct clk_mgr_internal *clk_mgr, bool enable); +void rn_vbios_smu_enable_pme_wa(struct clk_mgr_internal *clk_mgr); + +#endif /* DAL_DC_DCN10_RV1_CLK_MGR_VBIOS_SMU_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index cbc480a33376..04441dbcba76 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -58,21 +58,21 @@ #include "hubp.h" #include "dc_link_dp.h" +#include "dc_dmub_srv.h" -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #include "dsc.h" -#endif -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 #include "vm_helper.h" -#endif #include "dce/dce_i2c.h" +#define CTX \ + dc->ctx + #define DC_LOGGER \ dc->ctx->logger -const static char DC_BUILD_ID[] = "production-build"; +static const char DC_BUILD_ID[] = "production-build"; /** * DOC: Overview @@ -181,13 +181,25 @@ static bool create_links( link = link_create(&link_init_params); if (link) { - if (dc->config.edp_not_connected && - link->connector_signal == SIGNAL_TYPE_EDP) { - link_destroy(&link); - } else { + bool should_destory_link = false; + + if (link->connector_signal == SIGNAL_TYPE_EDP) { + if (dc->config.edp_not_connected) + should_destory_link = true; + else if (dc->debug.remove_disconnect_edp) { + enum dc_connection_type type; + dc_link_detect_sink(link, &type); + if (type == dc_connection_none) + should_destory_link = true; + } + } + + if (dc->config.force_enum_edp || !should_destory_link) { dc->links[dc->link_count] = link; link->dc = dc; ++dc->link_count; + } else { + link_destroy(&link); } } } @@ -275,11 +287,12 @@ bool dc_stream_adjust_vmin_vmax(struct dc *dc, struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe->stream == stream && pipe->stream_res.tg) { - pipe->stream->adjust = *adjust; dc->hwss.set_drr(&pipe, 1, adjust->v_total_min, - adjust->v_total_max); + adjust->v_total_max, + adjust->v_total_mid, + adjust->v_total_mid_frame_num); ret = true; } @@ -397,6 +410,27 @@ bool dc_stream_get_crc(struct dc *dc, struct dc_stream_state *stream, return false; } +void dc_stream_set_dyn_expansion(struct dc *dc, struct dc_stream_state *stream, + enum dc_dynamic_expansion option) +{ + /* OPP FMT dyn expansion updates*/ + int i = 0; + struct pipe_ctx *pipe_ctx; + + for (i = 0; i < MAX_PIPES; i++) { + if (dc->current_state->res_ctx.pipe_ctx[i].stream + == stream) { + pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; + pipe_ctx->stream_res.opp->dyn_expansion = option; + pipe_ctx->stream_res.opp->funcs->opp_set_dyn_expansion( + pipe_ctx->stream_res.opp, + COLOR_SPACE_YCBCR601, + stream->timing.display_color_depth, + stream->signal); + } + } +} + void dc_stream_set_dither_option(struct dc_stream_state *stream, enum dc_dither_option option) { @@ -476,10 +510,10 @@ bool dc_stream_program_csc_matrix(struct dc *dc, struct dc_stream_state *stream) return ret; } -void dc_stream_set_static_screen_events(struct dc *dc, +void dc_stream_set_static_screen_params(struct dc *dc, struct dc_stream_state **streams, int num_streams, - const struct dc_static_screen_events *events) + const struct dc_static_screen_params *params) { int i = 0; int j = 0; @@ -498,10 +532,10 @@ void dc_stream_set_static_screen_events(struct dc *dc, } } - dc->hwss.set_static_screen_control(pipes_affected, num_pipes_affected, events); + dc->hwss.set_static_screen_control(pipes_affected, num_pipes_affected, params); } -static void destruct(struct dc *dc) +static void dc_destruct(struct dc *dc) { if (dc->current_state) { dc_release_state(dc->current_state); @@ -534,7 +568,7 @@ static void destruct(struct dc *dc) kfree(dc->bw_dceip); dc->bw_dceip = NULL; -#ifdef CONFIG_DRM_AMD_DC_DCN1_0 +#ifdef CONFIG_DRM_AMD_DC_DCN kfree(dc->dcn_soc); dc->dcn_soc = NULL; @@ -542,32 +576,65 @@ static void destruct(struct dc *dc) dc->dcn_ip = NULL; #endif -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 kfree(dc->vm_helper); dc->vm_helper = NULL; -#endif } -static bool construct(struct dc *dc, +static bool dc_construct_ctx(struct dc *dc, + const struct dc_init_data *init_params) +{ + struct dc_context *dc_ctx; + enum dce_version dc_version = DCE_VERSION_UNKNOWN; + + dc_ctx = kzalloc(sizeof(*dc_ctx), GFP_KERNEL); + if (!dc_ctx) + return false; + + dc_ctx->cgs_device = init_params->cgs_device; + dc_ctx->driver_context = init_params->driver; + dc_ctx->dc = dc; + dc_ctx->asic_id = init_params->asic_id; + dc_ctx->dc_sink_id_count = 0; + dc_ctx->dc_stream_id_count = 0; + dc_ctx->dce_environment = init_params->dce_environment; + + /* Create logger */ + + dc_version = resource_parse_asic_id(init_params->asic_id); + dc_ctx->dce_version = dc_version; + + dc_ctx->perf_trace = dc_perf_trace_create(); + if (!dc_ctx->perf_trace) { + ASSERT_CRITICAL(false); + return false; + } + + dc->ctx = dc_ctx; + + return true; +} + +static bool dc_construct(struct dc *dc, const struct dc_init_data *init_params) { struct dc_context *dc_ctx; struct bw_calcs_dceip *dc_dceip; struct bw_calcs_vbios *dc_vbios; -#ifdef CONFIG_DRM_AMD_DC_DCN1_0 +#ifdef CONFIG_DRM_AMD_DC_DCN struct dcn_soc_bounding_box *dcn_soc; struct dcn_ip_params *dcn_ip; #endif - enum dce_version dc_version = DCE_VERSION_UNKNOWN; dc->config = init_params->flags; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 // Allocate memory for the vm_helper dc->vm_helper = kzalloc(sizeof(struct vm_helper), GFP_KERNEL); + if (!dc->vm_helper) { + dm_error("%s: failed to create dc->vm_helper\n", __func__); + goto fail; + } -#endif memcpy(&dc->bb_overrides, &init_params->bb_overrides, sizeof(dc->bb_overrides)); dc_dceip = kzalloc(sizeof(*dc_dceip), GFP_KERNEL); @@ -585,7 +652,7 @@ static bool construct(struct dc *dc, } dc->bw_vbios = dc_vbios; -#ifdef CONFIG_DRM_AMD_DC_DCN1_0 +#ifdef CONFIG_DRM_AMD_DC_DCN dcn_soc = kzalloc(sizeof(*dcn_soc), GFP_KERNEL); if (!dcn_soc) { dm_error("%s: failed to create dcn_soc\n", __func__); @@ -601,31 +668,15 @@ static bool construct(struct dc *dc, } dc->dcn_ip = dcn_ip; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 dc->soc_bounding_box = init_params->soc_bounding_box; #endif -#endif - dc_ctx = kzalloc(sizeof(*dc_ctx), GFP_KERNEL); - if (!dc_ctx) { + if (!dc_construct_ctx(dc, init_params)) { dm_error("%s: failed to create ctx\n", __func__); goto fail; } - dc_ctx->cgs_device = init_params->cgs_device; - dc_ctx->driver_context = init_params->driver; - dc_ctx->dc = dc; - dc_ctx->asic_id = init_params->asic_id; - dc_ctx->dc_sink_id_count = 0; - dc_ctx->dc_stream_id_count = 0; - dc->ctx = dc_ctx; - - /* Create logger */ - - dc_ctx->dce_environment = init_params->dce_environment; - - dc_version = resource_parse_asic_id(init_params->asic_id); - dc_ctx->dce_version = dc_version; + dc_ctx = dc->ctx; /* Resource should construct all asic specific resources. * This should be the only place where we need to parse the asic id @@ -640,7 +691,7 @@ static bool construct(struct dc *dc, bp_init_data.bios = init_params->asic_id.atombios_base_address; dc_ctx->dc_bios = dal_bios_parser_create( - &bp_init_data, dc_version); + &bp_init_data, dc_ctx->dce_version); if (!dc_ctx->dc_bios) { ASSERT_CRITICAL(false); @@ -648,17 +699,13 @@ static bool construct(struct dc *dc, } dc_ctx->created_bios = true; - } - - dc_ctx->perf_trace = dc_perf_trace_create(); - if (!dc_ctx->perf_trace) { - ASSERT_CRITICAL(false); - goto fail; } + + /* Create GPIO service */ dc_ctx->gpio_service = dal_gpio_service_create( - dc_version, + dc_ctx->dce_version, dc_ctx->dce_environment, dc_ctx); @@ -667,7 +714,7 @@ static bool construct(struct dc *dc, goto fail; } - dc->res_pool = dc_create_resource_pool(dc, init_params, dc_version); + dc->res_pool = dc_create_resource_pool(dc, init_params, dc_ctx->dce_version); if (!dc->res_pool) goto fail; @@ -675,6 +722,9 @@ static bool construct(struct dc *dc, if (!dc->clk_mgr) goto fail; + if (dc->res_pool->funcs->update_bw_bounding_box) + dc->res_pool->funcs->update_bw_bounding_box(dc, dc->clk_mgr->bw_params); + /* Creation of current_state must occur after dc->dml * is initialized in dc_create_resource_pool because * on creation it copies the contents of dc->dml @@ -695,12 +745,9 @@ static bool construct(struct dc *dc, return true; fail: - - destruct(dc); return false; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) static bool disable_all_writeback_pipes_for_stream( const struct dc *dc, struct dc_stream_state *stream, @@ -713,7 +760,6 @@ static bool disable_all_writeback_pipes_for_stream( return true; } -#endif static void disable_dangling_plane(struct dc *dc, struct dc_state *context) { @@ -739,11 +785,12 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) } if (should_disable && old_stream) { dc_rem_all_planes_for_stream(dc, old_stream, dangling_context); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) disable_all_writeback_pipes_for_stream(dc, old_stream, dangling_context); -#endif - dc->hwss.apply_ctx_for_surface(dc, old_stream, 0, dangling_context); + if (dc->hwss.apply_ctx_for_surface) + dc->hwss.apply_ctx_for_surface(dc, old_stream, 0, dangling_context); } + if (dc->hwss.program_front_end_for_ctx) + dc->hwss.program_front_end_for_ctx(dc, dangling_context); } current_ctx = dc->current_state; @@ -751,6 +798,33 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) dc_release_state(current_ctx); } +static void wait_for_no_pipes_pending(struct dc *dc, struct dc_state *context) +{ + int i; + int count = 0; + struct pipe_ctx *pipe; + PERF_TRACE(); + for (i = 0; i < MAX_PIPES; i++) { + pipe = &context->res_ctx.pipe_ctx[i]; + + if (!pipe->plane_state) + continue; + + /* Timeout 100 ms */ + while (count < 100000) { + /* Must set to false to start with, due to OR in update function */ + pipe->plane_state->status.is_flip_pending = false; + dc->hwss.update_pending_status(pipe); + if (!pipe->plane_state->status.is_flip_pending) + break; + udelay(1); + count++; + } + ASSERT(!pipe->plane_state->status.is_flip_pending); + } + PERF_TRACE(); +} + /******************************************************************************* * Public functions ******************************************************************************/ @@ -763,29 +837,38 @@ struct dc *dc_create(const struct dc_init_data *init_params) if (NULL == dc) goto alloc_fail; - if (false == construct(dc, init_params)) - goto construct_fail; + if (init_params->dce_environment == DCE_ENV_VIRTUAL_HW) { + if (false == dc_construct_ctx(dc, init_params)) { + dc_destruct(dc); + goto construct_fail; + } + } else { + if (false == dc_construct(dc, init_params)) { + dc_destruct(dc); + goto construct_fail; + } + + full_pipe_count = dc->res_pool->pipe_count; + if (dc->res_pool->underlay_pipe_index != NO_UNDERLAY_PIPE) + full_pipe_count--; + dc->caps.max_streams = min( + full_pipe_count, + dc->res_pool->stream_enc_count); - /*TODO: separate HW and SW initialization*/ - dc->hwss.init_hw(dc); + dc->optimize_seamless_boot_streams = 0; + dc->caps.max_links = dc->link_count; + dc->caps.max_audios = dc->res_pool->audio_count; + dc->caps.linear_pitch_alignment = 64; - full_pipe_count = dc->res_pool->pipe_count; - if (dc->res_pool->underlay_pipe_index != NO_UNDERLAY_PIPE) - full_pipe_count--; - dc->caps.max_streams = min( - full_pipe_count, - dc->res_pool->stream_enc_count); + dc->caps.max_dp_protocol_version = DP_VERSION_1_4; - dc->caps.max_links = dc->link_count; - dc->caps.max_audios = dc->res_pool->audio_count; - dc->caps.linear_pitch_alignment = 64; + if (dc->res_pool->dmcu != NULL) + dc->versions.dmcu_version = dc->res_pool->dmcu->dmcu_version; + } /* Populate versioning information */ dc->versions.dc_ver = DC_VER; - if (dc->res_pool->dmcu != NULL) - dc->versions.dmcu_version = dc->res_pool->dmcu->dmcu_version; - dc->build_id = DC_BUILD_ID; DC_LOG_DC("Display Core initialized\n"); @@ -801,14 +884,30 @@ alloc_fail: return NULL; } +void dc_hardware_init(struct dc *dc) +{ + if (dc->ctx->dce_environment != DCE_ENV_VIRTUAL_HW) + dc->hwss.init_hw(dc); +} + void dc_init_callbacks(struct dc *dc, const struct dc_callback_init *init_params) { +#ifdef CONFIG_DRM_AMD_DC_HDCP + dc->ctx->cp_psp = init_params->cp_psp; +#endif +} + +void dc_deinit_callbacks(struct dc *dc) +{ +#ifdef CONFIG_DRM_AMD_DC_HDCP + memset(&dc->ctx->cp_psp, 0, sizeof(dc->ctx->cp_psp)); +#endif } void dc_destroy(struct dc **dc) { - destruct(*dc); + dc_destruct(*dc); kfree(*dc); *dc = NULL; } @@ -882,15 +981,11 @@ static void program_timing_sync( /* set first pipe with plane as master */ for (j = 0; j < group_size; j++) { - struct pipe_ctx *temp; - if (pipe_set[j]->plane_state) { if (j == 0) break; - temp = pipe_set[0]; - pipe_set[0] = pipe_set[j]; - pipe_set[j] = temp; + swap(pipe_set[0], pipe_set[j]); break; } } @@ -947,31 +1042,87 @@ bool dc_validate_seamless_boot_timing(const struct dc *dc, struct dc_crtc_timing *crtc_timing) { struct timing_generator *tg; + struct stream_encoder *se = NULL; + + struct dc_crtc_timing hw_crtc_timing = {0}; + struct dc_link *link = sink->link; - unsigned int inst; + unsigned int i, enc_inst, tg_inst = 0; + + // Seamless port only support single DP and EDP so far + if (sink->sink_signal != SIGNAL_TYPE_DISPLAY_PORT && + sink->sink_signal != SIGNAL_TYPE_EDP) + return false; /* Check for enabled DIG to identify enabled display */ if (!link->link_enc->funcs->is_dig_enabled(link->link_enc)) return false; - /* Check for which front end is used by this encoder. - * Note the inst is 1 indexed, where 0 is undefined. - * Note that DIG_FE can source from different OTG but our - * current implementation always map 1-to-1, so this code makes - * the same assumption and doesn't check OTG source. - */ - inst = link->link_enc->funcs->get_dig_frontend(link->link_enc) - 1; + enc_inst = link->link_enc->funcs->get_dig_frontend(link->link_enc); + + if (enc_inst == ENGINE_ID_UNKNOWN) + return false; + + for (i = 0; i < dc->res_pool->stream_enc_count; i++) { + if (dc->res_pool->stream_enc[i]->id == enc_inst) { + + se = dc->res_pool->stream_enc[i]; + + tg_inst = dc->res_pool->stream_enc[i]->funcs->dig_source_otg( + dc->res_pool->stream_enc[i]); + break; + } + } + + // tg_inst not found + if (i == dc->res_pool->stream_enc_count) + return false; - /* Instance should be within the range of the pool */ - if (inst >= dc->res_pool->pipe_count) + if (tg_inst >= dc->res_pool->timing_generator_count) return false; - tg = dc->res_pool->timing_generators[inst]; + tg = dc->res_pool->timing_generators[tg_inst]; + + if (!tg->funcs->get_hw_timing) + return false; + + if (!tg->funcs->get_hw_timing(tg, &hw_crtc_timing)) + return false; + + if (crtc_timing->h_total != hw_crtc_timing.h_total) + return false; + + if (crtc_timing->h_border_left != hw_crtc_timing.h_border_left) + return false; + + if (crtc_timing->h_addressable != hw_crtc_timing.h_addressable) + return false; + + if (crtc_timing->h_border_right != hw_crtc_timing.h_border_right) + return false; + + if (crtc_timing->h_front_porch != hw_crtc_timing.h_front_porch) + return false; + + if (crtc_timing->h_sync_width != hw_crtc_timing.h_sync_width) + return false; + + if (crtc_timing->v_total != hw_crtc_timing.v_total) + return false; - if (!tg->funcs->is_matching_timing) + if (crtc_timing->v_border_top != hw_crtc_timing.v_border_top) return false; - if (!tg->funcs->is_matching_timing(tg, crtc_timing)) + if (crtc_timing->v_addressable != hw_crtc_timing.v_addressable) + return false; + + if (crtc_timing->v_border_bottom != hw_crtc_timing.v_border_bottom) + return false; + + if (crtc_timing->v_front_porch != hw_crtc_timing.v_front_porch) + return false; + + if (crtc_timing->v_sync_width != hw_crtc_timing.v_sync_width) return false; if (dc_is_dp_signal(link->connector_signal)) { @@ -979,10 +1130,25 @@ bool dc_validate_seamless_boot_timing(const struct dc *dc, dc->res_pool->dp_clock_source->funcs->get_pixel_clk_frequency_100hz( dc->res_pool->dp_clock_source, - inst, &pix_clk_100hz); + tg_inst, &pix_clk_100hz); if (crtc_timing->pix_clk_100hz != pix_clk_100hz) return false; + + if (!se->funcs->dp_get_pixel_format) + return false; + + if (!se->funcs->dp_get_pixel_format( + se, + &hw_crtc_timing.pixel_encoding, + &hw_crtc_timing.display_color_depth)) + return false; + + if (hw_crtc_timing.display_color_depth != crtc_timing->display_color_depth) + return false; + + if (hw_crtc_timing.pixel_encoding != crtc_timing->pixel_encoding) + return false; } return true; @@ -1035,24 +1201,25 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c for (i = 0; i < context->stream_count; i++) { if (context->streams[i]->apply_seamless_boot_optimization) - dc->optimize_seamless_boot = true; + dc->optimize_seamless_boot_streams++; } - if (!dc->optimize_seamless_boot) + if (dc->optimize_seamless_boot_streams == 0) dc->hwss.prepare_bandwidth(dc, context); /* re-program planes for existing stream, in case we need to * free up plane resource for later use */ - for (i = 0; i < context->stream_count; i++) { - if (context->streams[i]->mode_changed) - continue; + if (dc->hwss.apply_ctx_for_surface) + for (i = 0; i < context->stream_count; i++) { + if (context->streams[i]->mode_changed) + continue; - dc->hwss.apply_ctx_for_surface( - dc, context->streams[i], - context->stream_status[i].plane_count, - context); /* use new pipe config in new context */ - } + dc->hwss.apply_ctx_for_surface( + dc, context->streams[i], + context->stream_status[i].plane_count, + context); /* use new pipe config in new context */ + } /* Program hardware */ for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -1065,22 +1232,25 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c if (result != DC_OK) return result; - if (context->stream_count > 1) { + if (context->stream_count > 1 && !dc->debug.disable_timing_sync) { enable_timing_multisync(dc, context); program_timing_sync(dc, context); } /* Program all planes within new context*/ + if (dc->hwss.program_front_end_for_ctx) + dc->hwss.program_front_end_for_ctx(dc, context); for (i = 0; i < context->stream_count; i++) { const struct dc_link *link = context->streams[i]->link; if (!context->streams[i]->mode_changed) continue; - dc->hwss.apply_ctx_for_surface( - dc, context->streams[i], - context->stream_status[i].plane_count, - context); + if (dc->hwss.apply_ctx_for_surface) + dc->hwss.apply_ctx_for_surface( + dc, context->streams[i], + context->stream_status[i].plane_count, + context); /* * enable stereo @@ -1107,15 +1277,16 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c dc_enable_stereo(dc, context, dc_streams, context->stream_count); - if (!dc->optimize_seamless_boot) + if (dc->optimize_seamless_boot_streams == 0) { + /* Must wait for no flips to be pending before doing optimize bw */ + wait_for_no_pipes_pending(dc, context); /* pplib is notified if disp_num changed */ dc->hwss.optimize_bandwidth(dc, context); + } for (i = 0; i < context->stream_count; i++) context->streams[i]->mode_changed = false; - memset(&context->commit_hints, 0, sizeof(context->commit_hints)); - dc_release_state(dc->current_state); dc->current_state = context; @@ -1147,12 +1318,18 @@ bool dc_commit_state(struct dc *dc, struct dc_state *context) return (result == DC_OK); } +bool dc_is_hw_initialized(struct dc *dc) +{ + struct dc_bios *dcb = dc->ctx->dc_bios; + return dcb->funcs->is_accelerated_mode(dcb); +} + bool dc_post_update_surfaces_to_stream(struct dc *dc) { int i; struct dc_state *context = dc->current_state; - if (!dc->optimized_required || dc->optimize_seamless_boot) + if (!dc->optimized_required || dc->optimize_seamless_boot_streams > 0) return true; post_surface_trace(dc); @@ -1181,7 +1358,7 @@ struct dc_state *dc_create_state(struct dc *dc) * initialize and obtain IP and SOC the base DML instance from DC is * initially copied into every context */ -#ifdef CONFIG_DRM_AMD_DC_DCN1_0 +#ifdef CONFIG_DRM_AMD_DC_DCN memcpy(&context->bw_ctx.dml, &dc->dml, sizeof(struct display_mode_lib)); #endif @@ -1208,6 +1385,12 @@ struct dc_state *dc_copy_state(struct dc_state *src_ctx) if (cur_pipe->bottom_pipe) cur_pipe->bottom_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->bottom_pipe->pipe_idx]; + if (cur_pipe->prev_odm_pipe) + cur_pipe->prev_odm_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->prev_odm_pipe->pipe_idx]; + + if (cur_pipe->next_odm_pipe) + cur_pipe->next_odm_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->next_odm_pipe->pipe_idx]; + } for (i = 0; i < new_ctx->stream_count; i++) { @@ -1239,6 +1422,55 @@ void dc_release_state(struct dc_state *context) kref_put(&context->refcount, dc_state_free); } +bool dc_set_generic_gpio_for_stereo(bool enable, + struct gpio_service *gpio_service) +{ + enum gpio_result gpio_result = GPIO_RESULT_NON_SPECIFIC_ERROR; + struct gpio_pin_info pin_info; + struct gpio *generic; + struct gpio_generic_mux_config *config = kzalloc(sizeof(struct gpio_generic_mux_config), + GFP_KERNEL); + + if (!config) + return false; + pin_info = dal_gpio_get_generic_pin_info(gpio_service, GPIO_ID_GENERIC, 0); + + if (pin_info.mask == 0xFFFFFFFF || pin_info.offset == 0xFFFFFFFF) { + kfree(config); + return false; + } else { + generic = dal_gpio_service_create_generic_mux( + gpio_service, + pin_info.offset, + pin_info.mask); + } + + if (!generic) { + kfree(config); + return false; + } + + gpio_result = dal_gpio_open(generic, GPIO_MODE_OUTPUT); + + config->enable_output_from_mux = enable; + config->mux_select = GPIO_SIGNAL_SOURCE_PASS_THROUGH_STEREO_SYNC; + + if (gpio_result == GPIO_RESULT_OK) + gpio_result = dal_mux_setup_config(generic, config); + + if (gpio_result == GPIO_RESULT_OK) { + dal_gpio_close(generic); + dal_gpio_destroy_generic_mux(&generic); + kfree(config); + return true; + } else { + dal_gpio_close(generic); + dal_gpio_destroy_generic_mux(&generic); + kfree(config); + return false; + } +} + static bool is_surface_in_context( const struct dc_state *context, const struct dc_plane_state *plane_state) @@ -1299,14 +1531,9 @@ static enum surface_update_type get_plane_info_update_type(const struct dc_surfa elevate_update_type(&update_type, UPDATE_TYPE_MED); } - if (u->plane_info->sdr_white_level != u->surface->sdr_white_level) { - update_flags->bits.sdr_white_level = 1; - elevate_update_type(&update_type, UPDATE_TYPE_MED); - } - if (u->plane_info->dcc.enable != u->surface->dcc.enable - || u->plane_info->dcc.grph.independent_64b_blks != u->surface->dcc.grph.independent_64b_blks - || u->plane_info->dcc.grph.meta_pitch != u->surface->dcc.grph.meta_pitch) { + || u->plane_info->dcc.independent_64b_blks != u->surface->dcc.independent_64b_blks + || u->plane_info->dcc.meta_pitch != u->surface->dcc.meta_pitch) { update_flags->bits.dcc_change = 1; elevate_update_type(&update_type, UPDATE_TYPE_MED); } @@ -1320,9 +1547,8 @@ static enum surface_update_type get_plane_info_update_type(const struct dc_surfa elevate_update_type(&update_type, UPDATE_TYPE_FULL); } - if (u->plane_info->plane_size.grph.surface_pitch != u->surface->plane_size.grph.surface_pitch - || u->plane_info->plane_size.video.luma_pitch != u->surface->plane_size.video.luma_pitch - || u->plane_info->plane_size.video.chroma_pitch != u->surface->plane_size.video.chroma_pitch) { + if (u->plane_info->plane_size.surface_pitch != u->surface->plane_size.surface_pitch + || u->plane_info->plane_size.chroma_pitch != u->surface->plane_size.chroma_pitch) { update_flags->bits.plane_size_change = 1; elevate_update_type(&update_type, UPDATE_TYPE_MED); } @@ -1360,7 +1586,10 @@ static enum surface_update_type get_scaling_info_update_type( if (u->scaling_info->clip_rect.width != u->surface->clip_rect.width || u->scaling_info->clip_rect.height != u->surface->clip_rect.height || u->scaling_info->dst_rect.width != u->surface->dst_rect.width - || u->scaling_info->dst_rect.height != u->surface->dst_rect.height) { + || u->scaling_info->dst_rect.height != u->surface->dst_rect.height + || u->scaling_info->scaling_quality.integer_scaling != + u->surface->scaling_quality.integer_scaling + ) { update_flags->bits.scaling_change = 1; if ((u->scaling_info->dst_rect.width < u->surface->dst_rect.width @@ -1376,7 +1605,7 @@ static enum surface_update_type get_scaling_info_update_type( update_flags->bits.scaling_change = 1; if (u->scaling_info->src_rect.width > u->surface->src_rect.width - && u->scaling_info->src_rect.height > u->surface->src_rect.height) + || u->scaling_info->src_rect.height > u->surface->src_rect.height) /* Making src rect bigger requires a bandwidth change */ update_flags->bits.clock_change = 1; } @@ -1390,11 +1619,11 @@ static enum surface_update_type get_scaling_info_update_type( update_flags->bits.position_change = 1; if (update_flags->bits.clock_change - || update_flags->bits.bandwidth_change) + || update_flags->bits.bandwidth_change + || update_flags->bits.scaling_change) return UPDATE_TYPE_FULL; - if (update_flags->bits.scaling_change - || update_flags->bits.position_change) + if (update_flags->bits.position_change) return UPDATE_TYPE_MED; return UPDATE_TYPE_FAST; @@ -1408,20 +1637,15 @@ static enum surface_update_type det_surface_update(const struct dc *dc, enum surface_update_type overall_type = UPDATE_TYPE_FAST; union surface_update_flags *update_flags = &u->surface->update_flags; - update_flags->raw = 0; // Reset all flags - if (u->flip_addr) update_flags->bits.addr_update = 1; - if (!is_surface_in_context(context, u->surface)) { - update_flags->bits.new_plane = 1; + if (!is_surface_in_context(context, u->surface) || u->surface->force_full_update) { + update_flags->raw = 0xFFFFFFFF; return UPDATE_TYPE_FULL; } - if (u->surface->force_full_update) { - update_flags->bits.full_update = 1; - return UPDATE_TYPE_FULL; - } + update_flags->raw = 0; // Reset all flags type = get_plane_info_update_type(u); elevate_update_type(&overall_type, type); @@ -1453,6 +1677,12 @@ static enum surface_update_type det_surface_update(const struct dc *dc, update_flags->bits.gamma_change = 1; } + if (u->hdr_mult.value) + if (u->hdr_mult.value != u->surface->hdr_mult.value) { + update_flags->bits.hdr_mult = 1; + elevate_update_type(&overall_type, UPDATE_TYPE_MED); + } + if (update_flags->bits.in_transfer_func_change) { type = UPDATE_TYPE_MED; elevate_update_type(&overall_type, type); @@ -1479,40 +1709,45 @@ static enum surface_update_type check_update_surfaces_for_stream( enum surface_update_type overall_type = UPDATE_TYPE_FAST; if (stream_status == NULL || stream_status->plane_count != surface_count) - return UPDATE_TYPE_FULL; + overall_type = UPDATE_TYPE_FULL; /* some stream updates require passive update */ if (stream_update) { - if ((stream_update->src.height != 0) && - (stream_update->src.width != 0)) - return UPDATE_TYPE_FULL; + union stream_update_flags *su_flags = &stream_update->stream->update_flags; - if ((stream_update->dst.height != 0) && - (stream_update->dst.width != 0)) - return UPDATE_TYPE_FULL; + if ((stream_update->src.height != 0 && stream_update->src.width != 0) || + (stream_update->dst.height != 0 && stream_update->dst.width != 0) || + stream_update->integer_scaling_update) + su_flags->bits.scaling = 1; if (stream_update->out_transfer_func) - return UPDATE_TYPE_FULL; + su_flags->bits.out_tf = 1; if (stream_update->abm_level) - return UPDATE_TYPE_FULL; + su_flags->bits.abm_level = 1; if (stream_update->dpms_off) - return UPDATE_TYPE_FULL; + su_flags->bits.dpms_off = 1; + + if (stream_update->gamut_remap) + su_flags->bits.gamut_remap = 1; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) if (stream_update->wb_update) - return UPDATE_TYPE_FULL; -#endif + su_flags->bits.wb_update = 1; + if (su_flags->raw != 0) + overall_type = UPDATE_TYPE_FULL; + + if (stream_update->output_csc_transform || stream_update->output_color_space) + su_flags->bits.out_csc = 1; + + if (stream_update->dsc_config) + overall_type = UPDATE_TYPE_FULL; } for (i = 0 ; i < surface_count; i++) { enum surface_update_type type = det_surface_update(dc, &updates[i]); - if (type == UPDATE_TYPE_FULL) - return type; - elevate_update_type(&overall_type, type); } @@ -1534,13 +1769,29 @@ enum surface_update_type dc_check_update_surfaces_for_stream( int i; enum surface_update_type type; + if (stream_update) + stream_update->stream->update_flags.raw = 0; for (i = 0; i < surface_count; i++) updates[i].surface->update_flags.raw = 0; type = check_update_surfaces_for_stream(dc, updates, surface_count, stream_update, stream_status); - if (type == UPDATE_TYPE_FULL) + if (type == UPDATE_TYPE_FULL) { + if (stream_update) + stream_update->stream->update_flags.raw = 0xFFFFFFFF; for (i = 0; i < surface_count; i++) updates[i].surface->update_flags.raw = 0xFFFFFFFF; + } + + if (type == UPDATE_TYPE_FAST) { + // If there's an available clock comparator, we use that. + if (dc->clk_mgr->funcs->are_clock_states_equal) { + if (!dc->clk_mgr->funcs->are_clock_states_equal(&dc->clk_mgr->clks, &dc->current_state->bw_ctx.bw.dcn.clk)) + dc->optimized_required = true; + // Else we fallback to mem compare. + } else if (memcmp(&dc->current_state->bw_ctx.bw.dcn.clk, &dc->clk_mgr->clks, offsetof(struct dc_clocks, prev_p_state_change_support)) != 0) { + dc->optimized_required = true; + } + } return type; } @@ -1616,8 +1867,8 @@ static void copy_surface_update_to_plane( srf_update->plane_info->global_alpha_value; surface->dcc = srf_update->plane_info->dcc; - surface->sdr_white_level = - srf_update->plane_info->sdr_white_level; + surface->layer_index = + srf_update->plane_info->layer_index; } if (srf_update->gamma && @@ -1648,7 +1899,6 @@ static void copy_surface_update_to_plane( sizeof(struct dc_transfer_func_distributed_points)); } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) if (srf_update->func_shaper && (surface->in_shaper_func != srf_update->func_shaper)) @@ -1661,13 +1911,16 @@ static void copy_surface_update_to_plane( memcpy(surface->lut3d_func, srf_update->lut3d_func, sizeof(*surface->lut3d_func)); + if (srf_update->hdr_mult.value) + surface->hdr_mult = + srf_update->hdr_mult; + if (srf_update->blend_tf && (surface->blend_tf != srf_update->blend_tf)) memcpy(surface->blend_tf, srf_update->blend_tf, sizeof(*surface->blend_tf)); -#endif if (srf_update->input_csc_color_matrix) surface->input_csc_color_matrix = *srf_update->input_csc_color_matrix; @@ -1680,8 +1933,10 @@ static void copy_surface_update_to_plane( static void copy_stream_update_to_stream(struct dc *dc, struct dc_state *context, struct dc_stream_state *stream, - const struct dc_stream_update *update) + struct dc_stream_update *update) { + struct dc_context *dc_ctx = dc->ctx; + if (update == NULL || stream == NULL) return; @@ -1742,7 +1997,6 @@ static void copy_stream_update_to_stream(struct dc *dc, if (update->dither_option) stream->dither_option = *update->dither_option; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* update current stream with writeback info */ if (update->wb_update) { int i; @@ -1753,23 +2007,32 @@ static void copy_stream_update_to_stream(struct dc *dc, stream->writeback_info[i] = update->wb_update->writeback_info[i]; } -#endif -#if defined(CONFIG_DRM_AMD_DC_DSC_SUPPORT) if (update->dsc_config) { struct dc_dsc_config old_dsc_cfg = stream->timing.dsc_cfg; uint32_t old_dsc_enabled = stream->timing.flags.DSC; uint32_t enable_dsc = (update->dsc_config->num_slices_h != 0 && update->dsc_config->num_slices_v != 0); - stream->timing.dsc_cfg = *update->dsc_config; - stream->timing.flags.DSC = enable_dsc; - if (!dc->res_pool->funcs->validate_bandwidth(dc, context, - true)) { - stream->timing.dsc_cfg = old_dsc_cfg; - stream->timing.flags.DSC = old_dsc_enabled; + /* Use temporarry context for validating new DSC config */ + struct dc_state *dsc_validate_context = dc_create_state(dc); + + if (dsc_validate_context) { + dc_resource_state_copy_construct(dc->current_state, dsc_validate_context); + + stream->timing.dsc_cfg = *update->dsc_config; + stream->timing.flags.DSC = enable_dsc; + if (!dc->res_pool->funcs->validate_bandwidth(dc, dsc_validate_context, true)) { + stream->timing.dsc_cfg = old_dsc_cfg; + stream->timing.flags.DSC = old_dsc_enabled; + update->dsc_config = NULL; + } + + dc_release_state(dsc_validate_context); + } else { + DC_ERROR("Failed to allocate new validate context for DSC change\n"); + update->dsc_config = NULL; } } -#endif } static void commit_planes_do_stream_update(struct dc *dc, @@ -1779,22 +2042,21 @@ static void commit_planes_do_stream_update(struct dc *dc, struct dc_state *context) { int j; + bool should_program_abm; // Stream updates for (j = 0; j < dc->res_pool->pipe_count; j++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; - if (!pipe_ctx->top_pipe && - pipe_ctx->stream && - pipe_ctx->stream == stream) { + if (!pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe && pipe_ctx->stream == stream) { if (stream_update->periodic_interrupt0 && dc->hwss.setup_periodic_interrupt) - dc->hwss.setup_periodic_interrupt(pipe_ctx, VLINE0); + dc->hwss.setup_periodic_interrupt(dc, pipe_ctx, VLINE0); if (stream_update->periodic_interrupt1 && dc->hwss.setup_periodic_interrupt) - dc->hwss.setup_periodic_interrupt(pipe_ctx, VLINE1); + dc->hwss.setup_periodic_interrupt(dc, pipe_ctx, VLINE1); if ((stream_update->hdr_static_metadata && !stream->use_dynamic_meta) || stream_update->vrr_infopacket || @@ -1804,6 +2066,12 @@ static void commit_planes_do_stream_update(struct dc *dc, dc->hwss.update_info_frame(pipe_ctx); } + if (stream_update->hdr_static_metadata && + stream->use_dynamic_meta && + dc->hwss.set_dmdata_attributes && + pipe_ctx->stream->dmdata_address.quad_part != 0) + dc->hwss.set_dmdata_attributes(pipe_ctx); + if (stream_update->gamut_remap) dc_stream_set_gamut_remap(dc, stream); @@ -1811,54 +2079,65 @@ static void commit_planes_do_stream_update(struct dc *dc, dc_stream_program_csc_matrix(dc, stream); if (stream_update->dither_option) { -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); -#endif + struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe; resource_build_bit_depth_reduction_params(pipe_ctx->stream, &pipe_ctx->stream->bit_depth_params); pipe_ctx->stream_res.opp->funcs->opp_program_fmt(pipe_ctx->stream_res.opp, &stream->bit_depth_params, &stream->clamping); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - if (odm_pipe) + while (odm_pipe) { odm_pipe->stream_res.opp->funcs->opp_program_fmt(odm_pipe->stream_res.opp, &stream->bit_depth_params, &stream->clamping); -#endif + odm_pipe = odm_pipe->next_odm_pipe; + } } -#if defined(CONFIG_DRM_AMD_DC_DSC_SUPPORT) if (stream_update->dsc_config && dc->hwss.pipe_control_lock_global) { dc->hwss.pipe_control_lock_global(dc, pipe_ctx, true); dp_update_dsc_config(pipe_ctx); dc->hwss.pipe_control_lock_global(dc, pipe_ctx, false); } -#endif /* Full fe update*/ if (update_type == UPDATE_TYPE_FAST) continue; if (stream_update->dpms_off) { dc->hwss.pipe_control_lock(dc, pipe_ctx, true); + if (*stream_update->dpms_off) { - core_link_disable_stream(pipe_ctx, KEEP_ACQUIRED_RESOURCE); + core_link_disable_stream(pipe_ctx); + /* for dpms, keep acquired resources*/ + if (pipe_ctx->stream_res.audio && !dc->debug.az_endpoint_mute_only) + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + dc->hwss.optimize_bandwidth(dc, dc->current_state); } else { - dc->hwss.prepare_bandwidth(dc, dc->current_state); + if (dc->optimize_seamless_boot_streams == 0) + dc->hwss.prepare_bandwidth(dc, dc->current_state); + core_link_enable_stream(dc->current_state, pipe_ctx); } + dc->hwss.pipe_control_lock(dc, pipe_ctx, false); } if (stream_update->abm_level && pipe_ctx->stream_res.abm) { - if (pipe_ctx->stream_res.tg->funcs->is_blanked) { - // if otg funcs defined check if blanked before programming - if (!pipe_ctx->stream_res.tg->funcs->is_blanked(pipe_ctx->stream_res.tg)) + should_program_abm = true; + + // if otg funcs defined check if blanked before programming + if (pipe_ctx->stream_res.tg->funcs->is_blanked) + if (pipe_ctx->stream_res.tg->funcs->is_blanked(pipe_ctx->stream_res.tg)) + should_program_abm = false; + + if (should_program_abm) { + if (*stream_update->abm_level == ABM_LEVEL_IMMEDIATE_DISABLE) { + pipe_ctx->stream_res.abm->funcs->set_abm_immediate_disable(pipe_ctx->stream_res.abm); + } else { pipe_ctx->stream_res.abm->funcs->set_abm_level( pipe_ctx->stream_res.abm, stream->abm_level); - } else - pipe_ctx->stream_res.abm->funcs->set_abm_level( - pipe_ctx->stream_res.abm, stream->abm_level); + } + } } } } @@ -1875,7 +2154,7 @@ static void commit_planes_for_stream(struct dc *dc, int i, j; struct pipe_ctx *top_pipe_to_program = NULL; - if (dc->optimize_seamless_boot && surface_count > 0) { + if (dc->optimize_seamless_boot_streams > 0 && surface_count > 0) { /* Optimize seamless boot flag keeps clocks and watermarks high until * first flip. After first flip, optimization is required to lower * bandwidth. Important to note that it is expected UEFI will @@ -1884,12 +2163,14 @@ static void commit_planes_for_stream(struct dc *dc, */ if (stream->apply_seamless_boot_optimization) { stream->apply_seamless_boot_optimization = false; - dc->optimize_seamless_boot = false; - dc->optimized_required = true; + dc->optimize_seamless_boot_streams--; + + if (dc->optimize_seamless_boot_streams == 0) + dc->optimized_required = true; } } - if (update_type == UPDATE_TYPE_FULL && !dc->optimize_seamless_boot) { + if (update_type == UPDATE_TYPE_FULL && dc->optimize_seamless_boot_streams == 0) { dc->hwss.prepare_bandwidth(dc, context); context_clock_trace(dc, context); } @@ -1903,11 +2184,14 @@ static void commit_planes_for_stream(struct dc *dc, * In case of turning off screen, no need to program front end a second time. * just return after program blank. */ - dc->hwss.apply_ctx_for_surface(dc, stream, 0, context); + if (dc->hwss.apply_ctx_for_surface) + dc->hwss.apply_ctx_for_surface(dc, stream, 0, context); + if (dc->hwss.program_front_end_for_ctx) + dc->hwss.program_front_end_for_ctx(dc, context); + return; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) if (!IS_DIAG_DC(dc->ctx->dce_environment)) { for (i = 0; i < surface_count; i++) { struct dc_plane_state *plane_state = srf_updates[i].surface; @@ -1929,13 +2213,13 @@ static void commit_planes_for_stream(struct dc *dc, } } } -#endif // Update Type FULL, Surface updates for (j = 0; j < dc->res_pool->pipe_count; j++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; if (!pipe_ctx->top_pipe && + !pipe_ctx->prev_odm_pipe && pipe_ctx->stream && pipe_ctx->stream == stream) { struct dc_stream_status *stream_status = NULL; @@ -1949,7 +2233,6 @@ static void commit_planes_for_stream(struct dc *dc, if (update_type == UPDATE_TYPE_FAST) continue; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) ASSERT(!pipe_ctx->plane_state->triplebuffer_flips); if (dc->hwss.program_triplebuffer != NULL && @@ -1958,14 +2241,32 @@ static void commit_planes_for_stream(struct dc *dc, dc->hwss.program_triplebuffer( dc, pipe_ctx, pipe_ctx->plane_state->triplebuffer_flips); } -#endif stream_status = stream_get_status(context, pipe_ctx->stream); - dc->hwss.apply_ctx_for_surface( + if (dc->hwss.apply_ctx_for_surface) + dc->hwss.apply_ctx_for_surface( dc, pipe_ctx->stream, stream_status->plane_count, context); } } + if (dc->hwss.program_front_end_for_ctx && update_type != UPDATE_TYPE_FAST) { + dc->hwss.program_front_end_for_ctx(dc, context); +#ifdef CONFIG_DRM_AMD_DC_DCN + if (dc->debug.validate_dml_output) { + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx cur_pipe = context->res_ctx.pipe_ctx[i]; + if (cur_pipe.stream == NULL) + continue; + + cur_pipe.plane_res.hubp->funcs->validate_dml_output( + cur_pipe.plane_res.hubp, dc->ctx, + &context->res_ctx.pipe_ctx[i].rq_regs, + &context->res_ctx.pipe_ctx[i].dlg_regs, + &context->res_ctx.pipe_ctx[i].ttu_regs); + } + } +#endif + } // Update Type FAST, Surface updates if (update_type == UPDATE_TYPE_FAST) { @@ -1975,7 +2276,6 @@ static void commit_planes_for_stream(struct dc *dc, */ dc->hwss.pipe_control_lock(dc, top_pipe_to_program, true); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) if (dc->hwss.set_flip_control_gsl) for (i = 0; i < surface_count; i++) { struct dc_plane_state *plane_state = srf_updates[i].surface; @@ -1994,7 +2294,6 @@ static void commit_planes_for_stream(struct dc *dc, plane_state->flip_immediate); } } -#endif /* Perform requested Updates */ for (i = 0; i < surface_count; i++) { struct dc_plane_state *plane_state = srf_updates[i].surface; @@ -2007,7 +2306,6 @@ static void commit_planes_for_stream(struct dc *dc, if (pipe_ctx->plane_state != plane_state) continue; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /*program triple buffer after lock based on flip type*/ if (dc->hwss.program_triplebuffer != NULL && !dc->debug.disable_tri_buf) { @@ -2015,7 +2313,6 @@ static void commit_planes_for_stream(struct dc *dc, dc->hwss.program_triplebuffer( dc, pipe_ctx, plane_state->triplebuffer_flips); } -#endif if (srf_updates[i].flip_addr) dc->hwss.update_plane_addr(dc, pipe_ctx); } @@ -2050,7 +2347,7 @@ void dc_commit_updates_for_stream(struct dc *dc, enum surface_update_type update_type; struct dc_state *context; struct dc_context *dc_ctx = dc->ctx; - int i, j; + int i; stream_status = dc_stream_get_status(stream); context = dc->current_state; @@ -2088,16 +2385,6 @@ void dc_commit_updates_for_stream(struct dc *dc, copy_surface_update_to_plane(surface, &srf_updates[i]); - if (update_type >= UPDATE_TYPE_MED) { - for (j = 0; j < dc->res_pool->pipe_count; j++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; - - if (pipe_ctx->plane_state != surface) - continue; - - resource_build_scaling_params(pipe_ctx); - } - } } copy_stream_update_to_stream(dc, context, stream, stream_update); @@ -2175,18 +2462,22 @@ void dc_set_power_state( enum dc_acpi_cm_power_state power_state) { struct kref refcount; - struct display_mode_lib *dml = kzalloc(sizeof(struct display_mode_lib), - GFP_KERNEL); - - ASSERT(dml); - if (!dml) - return; + struct display_mode_lib *dml; switch (power_state) { case DC_ACPI_CM_POWER_STATE_D0: dc_resource_state_construct(dc, dc->current_state); + if (dc->ctx->dmub_srv) + dc_dmub_srv_wait_phy_init(dc->ctx->dmub_srv); + dc->hwss.init_hw(dc); + + if (dc->hwss.init_sys_ctx != NULL && + dc->vm_pa_config.valid) { + dc->hwss.init_sys_ctx(dc->hwseq, dc, &dc->vm_pa_config); + } + break; default: ASSERT(dc->current_state->stream_count == 0); @@ -2194,6 +2485,12 @@ void dc_set_power_state( * clean state, and dc hw programming optimizations will not * cause any trouble. */ + dml = kzalloc(sizeof(struct display_mode_lib), + GFP_KERNEL); + + ASSERT(dml); + if (!dml) + return; /* Preserve refcount */ refcount = dc->current_state->refcount; @@ -2207,10 +2504,10 @@ void dc_set_power_state( dc->current_state->refcount = refcount; dc->current_state->bw_ctx.dml = *dml; + kfree(dml); + break; } - - kfree(dml); } void dc_resume(struct dc *dc) @@ -2265,6 +2562,17 @@ bool dc_submit_i2c( cmd); } +bool dc_submit_i2c_oem( + struct dc *dc, + struct i2c_command *cmd) +{ + struct ddc_service *ddc = dc->res_pool->oem_device; + return dce_i2c_submit_command( + dc->res_pool, + ddc->ddc_pin, + cmd); +} + static bool link_add_remote_sink_helper(struct dc_link *dc_link, struct dc_sink *sink) { if (dc_link->sink_count >= MAX_SINKS_PER_LINK) { @@ -2387,3 +2695,14 @@ void get_clock_requirements_for_state(struct dc_state *state, struct AsicStateEx info->fClock = (unsigned int)state->bw_ctx.bw.dcn.clk.fclk_khz; info->phyClock = (unsigned int)state->bw_ctx.bw.dcn.clk.phyclk_khz; } +enum dc_status dc_set_clock(struct dc *dc, enum dc_clock_type clock_type, uint32_t clk_khz, uint32_t stepping) +{ + if (dc->hwss.set_clock) + return dc->hwss.set_clock(dc, clock_type, clk_khz, stepping); + return DC_ERROR_UNEXPECTED; +} +void dc_get_clock(struct dc *dc, enum dc_clock_type clock_type, struct dc_clock_config *clock_cfg) +{ + if (dc->hwss.get_clock) + dc->hwss.get_clock(dc, clock_type, clock_cfg); +} diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c index 5903e7822f98..502ed3c7959d 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c @@ -33,7 +33,6 @@ #include "core_status.h" #include "core_types.h" -#include "hw_sequencer.h" #include "resource.h" @@ -115,16 +114,16 @@ void pre_surface_trace( plane_state->clip_rect.height); SURFACE_TRACE( - "plane_state->plane_size.grph.surface_size.x = %d;\n" - "plane_state->plane_size.grph.surface_size.y = %d;\n" - "plane_state->plane_size.grph.surface_size.width = %d;\n" - "plane_state->plane_size.grph.surface_size.height = %d;\n" - "plane_state->plane_size.grph.surface_pitch = %d;\n", - plane_state->plane_size.grph.surface_size.x, - plane_state->plane_size.grph.surface_size.y, - plane_state->plane_size.grph.surface_size.width, - plane_state->plane_size.grph.surface_size.height, - plane_state->plane_size.grph.surface_pitch); + "plane_state->plane_size.surface_size.x = %d;\n" + "plane_state->plane_size.surface_size.y = %d;\n" + "plane_state->plane_size.surface_size.width = %d;\n" + "plane_state->plane_size.surface_size.height = %d;\n" + "plane_state->plane_size.surface_pitch = %d;\n", + plane_state->plane_size.surface_size.x, + plane_state->plane_size.surface_size.y, + plane_state->plane_size.surface_size.width, + plane_state->plane_size.surface_size.height, + plane_state->plane_size.surface_pitch); SURFACE_TRACE( @@ -202,20 +201,20 @@ void update_surface_trace( SURFACE_TRACE( "plane_info->color_space = %d;\n" "plane_info->format = %d;\n" - "plane_info->plane_size.grph.surface_pitch = %d;\n" - "plane_info->plane_size.grph.surface_size.height = %d;\n" - "plane_info->plane_size.grph.surface_size.width = %d;\n" - "plane_info->plane_size.grph.surface_size.x = %d;\n" - "plane_info->plane_size.grph.surface_size.y = %d;\n" + "plane_info->plane_size.surface_pitch = %d;\n" + "plane_info->plane_size.surface_size.height = %d;\n" + "plane_info->plane_size.surface_size.width = %d;\n" + "plane_info->plane_size.surface_size.x = %d;\n" + "plane_info->plane_size.surface_size.y = %d;\n" "plane_info->rotation = %d;\n" "plane_info->stereo_format = %d;\n", update->plane_info->color_space, update->plane_info->format, - update->plane_info->plane_size.grph.surface_pitch, - update->plane_info->plane_size.grph.surface_size.height, - update->plane_info->plane_size.grph.surface_size.width, - update->plane_info->plane_size.grph.surface_size.x, - update->plane_info->plane_size.grph.surface_size.y, + update->plane_info->plane_size.surface_pitch, + update->plane_info->plane_size.surface_size.height, + update->plane_info->plane_size.surface_size.width, + update->plane_info->plane_size.surface_size.x, + update->plane_info->plane_size.surface_size.y, update->plane_info->rotation, update->plane_info->stereo_format); @@ -310,14 +309,13 @@ void context_timing_trace( struct resource_context *res_ctx) { int i; - struct dc *core_dc = dc; int h_pos[MAX_PIPES] = {0}, v_pos[MAX_PIPES] = {0}; struct crtc_position position; - unsigned int underlay_idx = core_dc->res_pool->underlay_pipe_index; + unsigned int underlay_idx = dc->res_pool->underlay_pipe_index; DC_LOGGER_INIT(dc->ctx->logger); - for (i = 0; i < core_dc->res_pool->pipe_count; i++) { + for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[i]; /* get_position() returns CRTC vertical/horizontal counter * hence not applicable for underlay pipe @@ -329,7 +327,7 @@ void context_timing_trace( h_pos[i] = position.horizontal_count; v_pos[i] = position.vertical_count; } - for (i = 0; i < core_dc->res_pool->pipe_count; i++) { + for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[i]; if (pipe_ctx->stream == NULL || pipe_ctx->pipe_idx == underlay_idx) @@ -347,7 +345,7 @@ void context_clock_trace( struct dc *dc, struct dc_state *context) { -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) DC_LOGGER_INIT(dc->ctx->logger); CLOCK_TRACE("Current: dispclk_khz:%d max_dppclk_khz:%d dcfclk_khz:%d\n" "dcfclk_deep_sleep_khz:%d fclk_khz:%d socclk_khz:%d\n", diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index 355b4ba12796..a09119c10d7c 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -45,10 +45,7 @@ #include "dpcd_defs.h" #include "dmcu.h" #include "hw/clk_mgr.h" -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) -#include "resource.h" -#endif -#include "hw/clk_mgr.h" +#include "../dce/dmub_psr.h" #define DC_LOGGER_INIT(logger) @@ -78,12 +75,11 @@ enum { /******************************************************************************* * Private functions ******************************************************************************/ -static void destruct(struct dc_link *link) +static void dc_link_destruct(struct dc_link *link) { int i; if (link->hpd_gpio != NULL) { - dal_gpio_close(link->hpd_gpio); dal_gpio_destroy_irq(&link->hpd_gpio); link->hpd_gpio = NULL; } @@ -377,7 +373,7 @@ bool dc_link_is_dp_sink_present(struct dc_link *link) if (GPIO_RESULT_OK != dal_ddc_open( ddc, GPIO_MODE_INPUT, GPIO_DDC_CONFIG_TYPE_MODE_I2C)) { - dal_gpio_destroy_ddc(&ddc); + dal_ddc_close(ddc); return present; } @@ -524,7 +520,7 @@ static void link_disconnect_remap(struct dc_sink *prev_sink, struct dc_link *lin } -static void read_edp_current_link_settings_on_detect(struct dc_link *link) +static void read_current_link_settings_on_detect(struct dc_link *link) { union lane_count_set lane_count_set = { {0} }; uint8_t link_bw_set; @@ -559,17 +555,23 @@ static void read_edp_current_link_settings_on_detect(struct dc_link *link) &link_bw_set, sizeof(link_bw_set)); if (link_bw_set == 0) { - /* If standard link rates are not being used, - * Read DPCD 00115h to find the link rate set used - */ - core_link_read_dpcd(link, DP_LINK_RATE_SET, - &link_rate_set, sizeof(link_rate_set)); - - if (link_rate_set < link->dpcd_caps.edp_supported_link_rates_count) { - link->cur_link_settings.link_rate = - link->dpcd_caps.edp_supported_link_rates[link_rate_set]; - link->cur_link_settings.link_rate_set = link_rate_set; - link->cur_link_settings.use_link_rate_set = true; + if (link->connector_signal == SIGNAL_TYPE_EDP) { + /* If standard link rates are not being used, + * Read DPCD 00115h to find the edp link rate set used + */ + core_link_read_dpcd(link, DP_LINK_RATE_SET, + &link_rate_set, sizeof(link_rate_set)); + + // edp_supported_link_rates_count = 0 for DP + if (link_rate_set < link->dpcd_caps.edp_supported_link_rates_count) { + link->cur_link_settings.link_rate = + link->dpcd_caps.edp_supported_link_rates[link_rate_set]; + link->cur_link_settings.link_rate_set = link_rate_set; + link->cur_link_settings.use_link_rate_set = true; + } + } else { + // Link Rate not found. Seamless boot may not work. + ASSERT(false); } } else { link->cur_link_settings.link_rate = link_bw_set; @@ -684,6 +686,56 @@ static bool is_same_edid(struct dc_edid *old_edid, struct dc_edid *new_edid) return (memcmp(old_edid->raw_edid, new_edid->raw_edid, new_edid->length) == 0); } +static bool wait_for_alt_mode(struct dc_link *link) +{ + + /** + * something is terribly wrong if time out is > 200ms. (5Hz) + * 500 microseconds * 400 tries us 200 ms + **/ + unsigned int sleep_time_in_microseconds = 500; + unsigned int tries_allowed = 400; + bool is_in_alt_mode; + unsigned long long enter_timestamp; + unsigned long long finish_timestamp; + unsigned long long time_taken_in_ns; + int tries_taken; + + DC_LOGGER_INIT(link->ctx->logger); + + if (link->link_enc->funcs->is_in_alt_mode == NULL) + return true; + + is_in_alt_mode = link->link_enc->funcs->is_in_alt_mode(link->link_enc); + DC_LOG_WARNING("DP Alt mode state on HPD: %d\n", is_in_alt_mode); + + if (is_in_alt_mode) + return true; + + enter_timestamp = dm_get_timestamp(link->ctx); + + for (tries_taken = 0; tries_taken < tries_allowed; tries_taken++) { + udelay(sleep_time_in_microseconds); + /* ask the link if alt mode is enabled, if so return ok */ + if (link->link_enc->funcs->is_in_alt_mode(link->link_enc)) { + + finish_timestamp = dm_get_timestamp(link->ctx); + time_taken_in_ns = dm_get_elapse_time_in_ns( + link->ctx, finish_timestamp, enter_timestamp); + DC_LOG_WARNING("Alt mode entered finished after %llu ms\n", + div_u64(time_taken_in_ns, 1000000)); + return true; + } + + } + finish_timestamp = dm_get_timestamp(link->ctx); + time_taken_in_ns = dm_get_elapse_time_in_ns(link->ctx, finish_timestamp, + enter_timestamp); + DC_LOG_WARNING("Alt mode has timed out after %llu ms\n", + div_u64(time_taken_in_ns, 1000000)); + return false; +} + /** * dc_link_detect() - Detect if a sink is attached to a given link * @@ -692,7 +744,8 @@ static bool is_same_edid(struct dc_edid *old_edid, struct dc_edid *new_edid) * This does not create remote sinks but will trigger DM * to start MST detection if a branch is detected. */ -bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) +static bool dc_link_detect_helper(struct dc_link *link, + enum dc_detect_reason reason) { struct dc_sink_init_data sink_init_data = { 0 }; struct display_sink_capability sink_caps = { 0 }; @@ -707,6 +760,8 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) struct dpcd_caps prev_dpcd_caps; bool same_dpcd = true; enum dc_connection_type new_connection_type = dc_connection_none; + bool perform_dp_seamless_boot = false; + DC_LOGGER_INIT(link->ctx->logger); if (dc_is_virtual_signal(link->connector_signal)) @@ -763,15 +818,24 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) } case SIGNAL_TYPE_EDP: { - read_edp_current_link_settings_on_detect(link); detect_edp_sink_caps(link); - sink_caps.transaction_type = - DDC_TRANSACTION_TYPE_I2C_OVER_AUX; + read_current_link_settings_on_detect(link); + sink_caps.transaction_type = DDC_TRANSACTION_TYPE_I2C_OVER_AUX; sink_caps.signal = SIGNAL_TYPE_EDP; break; } case SIGNAL_TYPE_DISPLAY_PORT: { + + /* wa HPD high coming too early*/ + if (link->link_enc->features.flags.bits.DP_IS_USB_C == 1) { + + /* if alt mode times out, return false */ + if (wait_for_alt_mode(link) == false) { + return false; + } + } + if (!detect_dp( link, &sink_caps, @@ -787,25 +851,12 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) if (memcmp(&link->dpcd_caps, &prev_dpcd_caps, sizeof(struct dpcd_caps))) same_dpcd = false; } - /* Active dongle plug in without display or downstream unplug*/ + /* Active dongle downstream unplug*/ if (link->type == dc_connection_active_dongle && link->dpcd_caps.sink_count.bits.SINK_COUNT == 0) { - if (prev_sink != NULL) { + if (prev_sink != NULL) /* Downstream unplug */ dc_sink_release(prev_sink); - } else { - /* Empty dongle plug in */ - for (i = 0; i < LINK_TRAINING_MAX_VERIFY_RETRY; i++) { - int fail_count = 0; - - dp_verify_link_cap(link, - &link->reported_link_cap, - &fail_count); - - if (fail_count == 0) - break; - } - } return true; } @@ -817,12 +868,24 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) * empty which leads to allocate_mst_payload() has "0" * pbn_per_slot value leading to exception on dc_fixpt_div() */ - link->verified_link_cap = link->reported_link_cap; + dp_verify_mst_link_cap(link); + if (prev_sink != NULL) dc_sink_release(prev_sink); return false; } + // For seamless boot, to skip verify link cap, we read UEFI settings and set them as verified. + if (reason == DETECT_REASON_BOOT && + dc_ctx->dc->config.power_down_display_on_boot == false && + link->link_status.link_active == true) + perform_dp_seamless_boot = true; + + if (perform_dp_seamless_boot) { + read_current_link_settings_on_detect(link); + link->verified_link_cap = link->reported_link_cap; + } + break; } @@ -900,25 +963,17 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) same_edid = is_same_edid(&prev_sink->dc_edid, &sink->dc_edid); if (link->connector_signal == SIGNAL_TYPE_DISPLAY_PORT && - sink_caps.transaction_type == DDC_TRANSACTION_TYPE_I2C_OVER_AUX && - reason != DETECT_REASON_HPDRX) { + sink_caps.transaction_type == DDC_TRANSACTION_TYPE_I2C_OVER_AUX) { /* * TODO debug why Dell 2413 doesn't like * two link trainings */ - /* deal with non-mst cases */ - for (i = 0; i < LINK_TRAINING_MAX_VERIFY_RETRY; i++) { - int fail_count = 0; - - dp_verify_link_cap(link, - &link->reported_link_cap, - &fail_count); - - if (fail_count == 0) - break; - } - + // verify link cap for SST non-seamless boot + if (!perform_dp_seamless_boot) + dp_verify_link_cap_with_retries(link, + &link->reported_link_cap, + LINK_TRAINING_MAX_VERIFY_RETRY); } else { // If edid is the same, then discard new sink and revert back to original sink if (same_edid) { @@ -1007,6 +1062,23 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) dc_sink_release(prev_sink); return true; + +} + +bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) +{ + const struct dc *dc = link->dc; + bool ret; + + /* get out of low power state */ + clk_mgr_exit_optimized_pwr_state(dc, dc->clk_mgr); + + ret = dc_link_detect_helper(link, reason); + + /* Go back to power optimized state */ + clk_mgr_optimize_pwr_state(dc, dc->clk_mgr); + + return ret; } bool dc_link_get_hpd_state(struct dc_link *dc_link) @@ -1166,7 +1238,7 @@ static enum transmitter translate_encoder_to_transmitter( } } -static bool construct( +static bool dc_link_construct( struct dc_link *link, const struct link_init_data *init_params) { @@ -1188,6 +1260,9 @@ static bool construct( link->ctx = dc_ctx; link->link_index = init_params->link_index; + memset(&link->preferred_training_settings, 0, sizeof(struct dc_link_training_overrides)); + memset(&link->preferred_link_setting, 0, sizeof(struct dc_link_settings)); + link->link_id = bios->funcs->get_connector_id(bios, init_params->connector_index); if (link->link_id.type != OBJECT_TYPE_CONNECTOR) { @@ -1365,7 +1440,7 @@ struct dc_link *link_create(const struct link_init_data *init_params) if (NULL == link) goto alloc_fail; - if (false == construct(link, init_params)) + if (false == dc_link_construct(link, init_params)) goto construct_fail; return link; @@ -1379,62 +1454,11 @@ alloc_fail: void link_destroy(struct dc_link **link) { - destruct(*link); + dc_link_destruct(*link); kfree(*link); *link = NULL; } -static void dpcd_configure_panel_mode( - struct dc_link *link, - enum dp_panel_mode panel_mode) -{ - union dpcd_edp_config edp_config_set; - bool panel_mode_edp = false; - DC_LOGGER_INIT(link->ctx->logger); - - memset(&edp_config_set, '\0', sizeof(union dpcd_edp_config)); - - if (DP_PANEL_MODE_DEFAULT != panel_mode) { - - switch (panel_mode) { - case DP_PANEL_MODE_EDP: - case DP_PANEL_MODE_SPECIAL: - panel_mode_edp = true; - break; - - default: - break; - } - - /*set edp panel mode in receiver*/ - core_link_read_dpcd( - link, - DP_EDP_CONFIGURATION_SET, - &edp_config_set.raw, - sizeof(edp_config_set.raw)); - - if (edp_config_set.bits.PANEL_MODE_EDP - != panel_mode_edp) { - enum ddc_result result = DDC_RESULT_UNKNOWN; - - edp_config_set.bits.PANEL_MODE_EDP = - panel_mode_edp; - result = core_link_write_dpcd( - link, - DP_EDP_CONFIGURATION_SET, - &edp_config_set.raw, - sizeof(edp_config_set.raw)); - - ASSERT(result == DDC_RESULT_SUCESSFULL); - } - } - DC_LOG_DETECTION_DP_CAPS("Link: %d eDP panel mode supported: %d " - "eDP panel mode enabled: %d \n", - link->link_index, - link->dpcd_caps.panel_mode_edp, - panel_mode_edp); -} - static void enable_stream_features(struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; @@ -1465,21 +1489,22 @@ static enum dc_status enable_link_dp( bool skip_video_pattern; struct dc_link *link = stream->link; struct dc_link_settings link_settings = {0}; - enum dp_panel_mode panel_mode; + bool fec_enable; + int i; + bool apply_seamless_boot_optimization = false; + + // check for seamless boot + for (i = 0; i < state->stream_count; i++) { + if (state->streams[i]->apply_seamless_boot_optimization) { + apply_seamless_boot_optimization = true; + break; + } + } /* get link settings for video mode timing */ decide_link_settings(stream, &link_settings); if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP) { - /* If link settings are different than current and link already enabled - * then need to disable before programming to new rate. - */ - if (link->link_status.link_active && - (link->cur_link_settings.lane_count != link_settings.lane_count || - link->cur_link_settings.link_rate != link_settings.link_rate)) { - dp_disable_link_phy(link, pipe_ctx->stream->signal); - } - /*in case it is not on*/ link->dc->hwss.edp_power_control(link, true); link->dc->hwss.edp_wait_for_hpd_ready(link, true); @@ -1487,46 +1512,32 @@ static enum dc_status enable_link_dp( pipe_ctx->stream_res.pix_clk_params.requested_sym_clk = link_settings.link_rate * LINK_RATE_REF_FREQ_IN_KHZ; - state->clk_mgr->funcs->update_clocks(state->clk_mgr, state, false); - - dp_enable_link_phy( - link, - pipe_ctx->stream->signal, - pipe_ctx->clock_source->id, - &link_settings); - - if (stream->sink_patches.dppowerup_delay > 0) { - int delay_dp_power_up_in_ms = stream->sink_patches.dppowerup_delay; - - msleep(delay_dp_power_up_in_ms); - } - - panel_mode = dp_get_panel_mode(link); - dpcd_configure_panel_mode(link, panel_mode); + if (state->clk_mgr && !apply_seamless_boot_optimization) + state->clk_mgr->funcs->update_clocks(state->clk_mgr, state, false); skip_video_pattern = true; if (link_settings.link_rate == LINK_RATE_LOW) skip_video_pattern = false; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - dp_set_fec_ready(link, true); -#endif - if (perform_link_training_with_retries( - link, &link_settings, skip_video_pattern, - LINK_TRAINING_ATTEMPTS)) { + LINK_TRAINING_ATTEMPTS, + pipe_ctx, + pipe_ctx->stream->signal)) { link->cur_link_settings = link_settings; status = DC_OK; } else status = DC_FAIL_DP_LINK_TRAINING; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - dp_set_fec_enable(link, true); -#endif + if (link->preferred_training_settings.fec_enable != NULL) + fec_enable = *link->preferred_training_settings.fec_enable; + else + fec_enable = true; + + dp_set_fec_enable(link, fec_enable); return status; } @@ -2013,6 +2024,45 @@ static void write_i2c_redriver_setting( ASSERT(i2c_success); } +static void disable_link(struct dc_link *link, enum signal_type signal) +{ + /* + * TODO: implement call for dp_set_hw_test_pattern + * it is needed for compliance testing + */ + + /* Here we need to specify that encoder output settings + * need to be calculated as for the set mode, + * it will lead to querying dynamic link capabilities + * which should be done before enable output + */ + + if (dc_is_dp_signal(signal)) { + /* SST DP, eDP */ + if (dc_is_dp_sst_signal(signal)) + dp_disable_link_phy(link, signal); + else + dp_disable_link_phy_mst(link, signal); + + if (dc_is_dp_sst_signal(signal) || + link->mst_stream_alloc_table.stream_count == 0) { + dp_set_fec_enable(link, false); + dp_set_fec_ready(link, false); + } + } else { + if (signal != SIGNAL_TYPE_VIRTUAL) + link->link_enc->funcs->disable_output(link->link_enc, signal); + } + + if (signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + /* MST disable link only when no stream use the link */ + if (link->mst_stream_alloc_table.stream_count <= 0) + link->link_status.link_active = false; + } else { + link->link_status.link_active = false; + } +} + static void enable_link_hdmi(struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; @@ -2097,6 +2147,19 @@ static enum dc_status enable_link( struct pipe_ctx *pipe_ctx) { enum dc_status status = DC_ERROR_UNEXPECTED; + struct dc_stream_state *stream = pipe_ctx->stream; + struct dc_link *link = stream->link; + + /* There's some scenarios where driver is unloaded with display + * still enabled. When driver is reloaded, it may cause a display + * to not light up if there is a mismatch between old and new + * link settings. Need to call disable first before enabling at + * new link settings. + */ + if (link->link_status.link_active) { + disable_link(link, pipe_ctx->stream->signal); + } + switch (pipe_ctx->stream->signal) { case SIGNAL_TYPE_DISPLAY_PORT: status = enable_link_dp(state, pipe_ctx); @@ -2131,44 +2194,6 @@ static enum dc_status enable_link( return status; } -static void disable_link(struct dc_link *link, enum signal_type signal) -{ - /* - * TODO: implement call for dp_set_hw_test_pattern - * it is needed for compliance testing - */ - - /* here we need to specify that encoder output settings - * need to be calculated as for the set mode, - * it will lead to querying dynamic link capabilities - * which should be done before enable output */ - - if (dc_is_dp_signal(signal)) { - /* SST DP, eDP */ - if (dc_is_dp_sst_signal(signal)) - dp_disable_link_phy(link, signal); - else - dp_disable_link_phy_mst(link, signal); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - - if (dc_is_dp_sst_signal(signal) || - link->mst_stream_alloc_table.stream_count == 0) { - dp_set_fec_enable(link, false); - dp_set_fec_ready(link, false); - } -#endif - } else - link->link_enc->funcs->disable_output(link->link_enc, signal); - - if (signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { - /* MST disable link only when no stream use the link */ - if (link->mst_stream_alloc_table.stream_count <= 0) - link->link_status.link_active = false; - } else { - link->link_status.link_active = false; - } -} - static uint32_t get_timing_pixel_clock_100hz(const struct dc_crtc_timing *timing) { @@ -2205,7 +2230,7 @@ static bool dp_active_dongle_validate_timing( break; } - if (dongle_caps->dongle_type != DISPLAY_DONGLE_DP_HDMI_CONVERTER || + if (dpcd_caps->dongle_type != DISPLAY_DONGLE_DP_HDMI_CONVERTER || dongle_caps->extendedCapValid == false) return true; @@ -2305,9 +2330,9 @@ bool dc_link_set_backlight_level(const struct dc_link *link, uint32_t backlight_pwm_u16_16, uint32_t frame_ramp) { - struct dc *core_dc = link->ctx->dc; - struct abm *abm = core_dc->res_pool->abm; - struct dmcu *dmcu = core_dc->res_pool->dmcu; + struct dc *dc = link->ctx->dc; + struct abm *abm = dc->res_pool->abm; + struct dmcu *dmcu = dc->res_pool->dmcu; unsigned int controller_id = 0; bool use_smooth_brightness = true; int i; @@ -2325,22 +2350,22 @@ bool dc_link_set_backlight_level(const struct dc_link *link, if (dc_is_embedded_signal(link->connector_signal)) { for (i = 0; i < MAX_PIPES; i++) { - if (core_dc->current_state->res_ctx.pipe_ctx[i].stream) { - if (core_dc->current_state->res_ctx. + if (dc->current_state->res_ctx.pipe_ctx[i].stream) { + if (dc->current_state->res_ctx. pipe_ctx[i].stream->link == link) { /* DMCU -1 for all controller id values, * therefore +1 here */ controller_id = - core_dc->current_state-> + dc->current_state-> res_ctx.pipe_ctx[i].stream_res.tg->inst + 1; /* Disable brightness ramping when the display is blanked * as it can hang the DMCU */ - if (core_dc->current_state->res_ctx.pipe_ctx[i].plane_state == NULL) + if (dc->current_state->res_ctx.pipe_ctx[i].plane_state == NULL) frame_ramp = 0; } } @@ -2358,8 +2383,8 @@ bool dc_link_set_backlight_level(const struct dc_link *link, bool dc_link_set_abm_disable(const struct dc_link *link) { - struct dc *core_dc = link->ctx->dc; - struct abm *abm = core_dc->res_pool->abm; + struct dc *dc = link->ctx->dc; + struct abm *abm = dc->res_pool->abm; if ((abm == NULL) || (abm->funcs->set_backlight_level_pwm == NULL)) return false; @@ -2369,17 +2394,215 @@ bool dc_link_set_abm_disable(const struct dc_link *link) return true; } -bool dc_link_set_psr_enable(const struct dc_link *link, bool enable, bool wait) +bool dc_link_set_psr_allow_active(struct dc_link *link, bool allow_active, bool wait) { - struct dc *core_dc = link->ctx->dc; - struct dmcu *dmcu = core_dc->res_pool->dmcu; + struct dc *dc = link->ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; + struct dmub_psr *psr = dc->res_pool->psr; - if ((dmcu != NULL && dmcu->funcs->is_dmcu_initialized(dmcu)) && link->psr_enabled) - dmcu->funcs->set_psr_enable(dmcu, enable, wait); + if ((psr != NULL) && link->psr_feature_enabled) + psr->funcs->set_psr_enable(psr, allow_active); + else if ((dmcu != NULL && dmcu->funcs->is_dmcu_initialized(dmcu)) && link->psr_feature_enabled) + dmcu->funcs->set_psr_enable(dmcu, allow_active, wait); + + link->psr_allow_active = allow_active; return true; } +bool dc_link_get_psr_state(const struct dc_link *link, uint32_t *psr_state) +{ + struct dc *dc = link->ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; + struct dmub_psr *psr = dc->res_pool->psr; + + if (psr != NULL && link->psr_feature_enabled) + psr->funcs->get_psr_state(psr_state); + else if (dmcu != NULL && link->psr_feature_enabled) + dmcu->funcs->get_psr_state(dmcu, psr_state); + + return true; +} + +static inline enum physical_phy_id +transmitter_to_phy_id(enum transmitter transmitter_value) +{ + switch (transmitter_value) { + case TRANSMITTER_UNIPHY_A: + return PHYLD_0; + case TRANSMITTER_UNIPHY_B: + return PHYLD_1; + case TRANSMITTER_UNIPHY_C: + return PHYLD_2; + case TRANSMITTER_UNIPHY_D: + return PHYLD_3; + case TRANSMITTER_UNIPHY_E: + return PHYLD_4; + case TRANSMITTER_UNIPHY_F: + return PHYLD_5; + case TRANSMITTER_NUTMEG_CRT: + return PHYLD_6; + case TRANSMITTER_TRAVIS_CRT: + return PHYLD_7; + case TRANSMITTER_TRAVIS_LCD: + return PHYLD_8; + case TRANSMITTER_UNIPHY_G: + return PHYLD_9; + case TRANSMITTER_COUNT: + return PHYLD_COUNT; + case TRANSMITTER_UNKNOWN: + return PHYLD_UNKNOWN; + default: + WARN_ONCE(1, "Unknown transmitter value %d\n", + transmitter_value); + return PHYLD_UNKNOWN; + } +} + +bool dc_link_setup_psr(struct dc_link *link, + const struct dc_stream_state *stream, struct psr_config *psr_config, + struct psr_context *psr_context) +{ + struct dc *dc; + struct dmcu *dmcu; + struct dmub_psr *psr; + int i; + /* updateSinkPsrDpcdConfig*/ + union dpcd_psr_configuration psr_configuration; + + psr_context->controllerId = CONTROLLER_ID_UNDEFINED; + + if (!link) + return false; + + dc = link->ctx->dc; + dmcu = dc->res_pool->dmcu; + psr = dc->res_pool->psr; + + if (!dmcu && !psr) + return false; + + + memset(&psr_configuration, 0, sizeof(psr_configuration)); + + psr_configuration.bits.ENABLE = 1; + psr_configuration.bits.CRC_VERIFICATION = 1; + psr_configuration.bits.FRAME_CAPTURE_INDICATION = + psr_config->psr_frame_capture_indication_req; + + /* Check for PSR v2*/ + if (psr_config->psr_version == 0x2) { + /* For PSR v2 selective update. + * Indicates whether sink should start capturing + * immediately following active scan line, + * or starting with the 2nd active scan line. + */ + psr_configuration.bits.LINE_CAPTURE_INDICATION = 0; + /*For PSR v2, determines whether Sink should generate + * IRQ_HPD when CRC mismatch is detected. + */ + psr_configuration.bits.IRQ_HPD_WITH_CRC_ERROR = 1; + } + + dm_helpers_dp_write_dpcd( + link->ctx, + link, + 368, + &psr_configuration.raw, + sizeof(psr_configuration.raw)); + + psr_context->channel = link->ddc->ddc_pin->hw_info.ddc_channel; + psr_context->transmitterId = link->link_enc->transmitter; + psr_context->engineId = link->link_enc->preferred_engine; + + for (i = 0; i < MAX_PIPES; i++) { + if (dc->current_state->res_ctx.pipe_ctx[i].stream + == stream) { + /* dmcu -1 for all controller id values, + * therefore +1 here + */ + psr_context->controllerId = + dc->current_state->res_ctx. + pipe_ctx[i].stream_res.tg->inst + 1; + break; + } + } + + /* Hardcoded for now. Can be Pcie or Uniphy (or Unknown)*/ + psr_context->phyType = PHY_TYPE_UNIPHY; + /*PhyId is associated with the transmitter id*/ + psr_context->smuPhyId = + transmitter_to_phy_id(link->link_enc->transmitter); + + psr_context->crtcTimingVerticalTotal = stream->timing.v_total; + psr_context->vsync_rate_hz = div64_u64(div64_u64((stream-> + timing.pix_clk_100hz * 100), + stream->timing.v_total), + stream->timing.h_total); + + psr_context->psrSupportedDisplayConfig = true; + psr_context->psrExitLinkTrainingRequired = + psr_config->psr_exit_link_training_required; + psr_context->sdpTransmitLineNumDeadline = + psr_config->psr_sdp_transmit_line_num_deadline; + psr_context->psrFrameCaptureIndicationReq = + psr_config->psr_frame_capture_indication_req; + + psr_context->skipPsrWaitForPllLock = 0; /* only = 1 in KV */ + + psr_context->numberOfControllers = + link->dc->res_pool->timing_generator_count; + + psr_context->rfb_update_auto_en = true; + + /* 2 frames before enter PSR. */ + psr_context->timehyst_frames = 2; + /* half a frame + * (units in 100 lines, i.e. a value of 1 represents 100 lines) + */ + psr_context->hyst_lines = stream->timing.v_total / 2 / 100; + psr_context->aux_repeats = 10; + + psr_context->psr_level.u32all = 0; + +#if defined(CONFIG_DRM_AMD_DC_DCN) + /*skip power down the single pipe since it blocks the cstate*/ + if (ASICREV_IS_RAVEN(link->ctx->asic_id.hw_internal_rev)) + psr_context->psr_level.bits.SKIP_CRTC_DISABLE = true; +#endif + + /* SMU will perform additional powerdown sequence. + * For unsupported ASICs, set psr_level flag to skip PSR + * static screen notification to SMU. + * (Always set for DAL2, did not check ASIC) + */ + psr_context->allow_smu_optimizations = psr_config->allow_smu_optimizations; + + /* Complete PSR entry before aborting to prevent intermittent + * freezes on certain eDPs + */ + psr_context->psr_level.bits.DISABLE_PSR_ENTRY_ABORT = 1; + + /* Controls additional delay after remote frame capture before + * continuing power down, default = 0 + */ + psr_context->frame_delay = 0; + + if (psr) + link->psr_feature_enabled = psr->funcs->setup_psr(psr, link, psr_context); + else + link->psr_feature_enabled = dmcu->funcs->setup_psr(dmcu, link, psr_context); + + /* psr_enabled == 0 indicates setup_psr did not succeed, but this + * should not happen since firmware should be running at this point + */ + if (link->psr_feature_enabled == 0) + ASSERT(0); + + return true; + +} + const struct dc_link_status *dc_link_get_status(const struct dc_link *link) { return &link->link_status; @@ -2403,28 +2626,13 @@ static struct fixed31_32 get_pbn_per_slot(struct dc_stream_state *stream) return dc_fixpt_div_int(mbytes_per_sec, 54); } -static int get_color_depth(enum dc_color_depth color_depth) -{ - switch (color_depth) { - case COLOR_DEPTH_666: return 6; - case COLOR_DEPTH_888: return 8; - case COLOR_DEPTH_101010: return 10; - case COLOR_DEPTH_121212: return 12; - case COLOR_DEPTH_141414: return 14; - case COLOR_DEPTH_161616: return 16; - default: return 0; - } -} - static struct fixed31_32 get_pbn_from_timing(struct pipe_ctx *pipe_ctx) { - uint32_t bpc; uint64_t kbps; struct fixed31_32 peak_kbps; uint32_t numerator; uint32_t denominator; - bpc = get_color_depth(pipe_ctx->stream_res.pix_clk_params.color_depth); kbps = dc_bandwidth_in_kbps_from_timing(&pipe_ctx->stream->timing); /* @@ -2498,7 +2706,7 @@ static void update_mst_stream_alloc_table( /* convert link_mst_stream_alloc_table to dm dp_mst_stream_alloc_table * because stream_encoder is not exposed to dm */ -static enum dc_status allocate_mst_payload(struct pipe_ctx *pipe_ctx) +enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; @@ -2509,6 +2717,7 @@ static enum dc_status allocate_mst_payload(struct pipe_ctx *pipe_ctx) struct fixed31_32 pbn; struct fixed31_32 pbn_per_slot; uint8_t i; + enum act_return_status ret; DC_LOGGER_INIT(link->ctx->logger); /* enable_link_dp_mst already check link->enabled_stream_count @@ -2556,14 +2765,16 @@ static enum dc_status allocate_mst_payload(struct pipe_ctx *pipe_ctx) &link->mst_stream_alloc_table); /* send down message */ - dm_helpers_dp_mst_poll_for_allocation_change_trigger( + ret = dm_helpers_dp_mst_poll_for_allocation_change_trigger( stream->ctx, stream); - dm_helpers_dp_mst_send_payload_allocation( - stream->ctx, - stream, - true); + if (ret != ACT_LINK_LOST) { + dm_helpers_dp_mst_send_payload_allocation( + stream->ctx, + stream, + true); + } /* slot X.Y for only current stream */ pbn_per_slot = get_pbn_per_slot(stream); @@ -2656,15 +2867,83 @@ static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx) return DC_OK; } +enum dc_status dc_link_reallocate_mst_payload(struct dc_link *link) +{ + int i; + struct pipe_ctx *pipe_ctx; + + // Clear all of MST payload then reallocate + for (i = 0; i < MAX_PIPES; i++) { + pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i]; + + /* driver enable split pipe for external monitors + * we have to check pipe_ctx is split pipe or not + * If it's split pipe, driver using top pipe to + * reaallocate. + */ + if (!pipe_ctx || pipe_ctx->top_pipe) + continue; + + if (pipe_ctx->stream && pipe_ctx->stream->link == link && + pipe_ctx->stream->dpms_off == false && + pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + deallocate_mst_payload(pipe_ctx); + } + } + + for (i = 0; i < MAX_PIPES; i++) { + pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i]; + + if (!pipe_ctx || pipe_ctx->top_pipe) + continue; + + if (pipe_ctx->stream && pipe_ctx->stream->link == link && + pipe_ctx->stream->dpms_off == false && + pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + /* enable/disable PHY will clear connection between BE and FE + * need to restore it. + */ + link->link_enc->funcs->connect_dig_be_to_fe(link->link_enc, + pipe_ctx->stream_res.stream_enc->id, true); + dc_link_allocate_mst_payload(pipe_ctx); + } + } + + return DC_OK; +} + +#if defined(CONFIG_DRM_AMD_DC_HDCP) +static void update_psp_stream_config(struct pipe_ctx *pipe_ctx, bool dpms_off) +{ + struct cp_psp *cp_psp = &pipe_ctx->stream->ctx->cp_psp; + if (cp_psp && cp_psp->funcs.update_stream_config) { + struct cp_psp_stream_config config; + + memset(&config, 0, sizeof(config)); + + config.otg_inst = (uint8_t) pipe_ctx->stream_res.tg->inst; + config.stream_enc_inst = (uint8_t) pipe_ctx->stream_res.stream_enc->id; + config.link_enc_inst = pipe_ctx->stream->link->link_enc_hw_inst; + config.dpms_off = dpms_off; + config.dm_stream_ctx = pipe_ctx->stream->dm_stream_context; + cp_psp->funcs.update_stream_config(cp_psp->handle, &config); + } +} +#endif + void core_link_enable_stream( struct dc_state *state, struct pipe_ctx *pipe_ctx) { - struct dc *core_dc = pipe_ctx->stream->ctx->dc; + struct dc *dc = pipe_ctx->stream->ctx->dc; struct dc_stream_state *stream = pipe_ctx->stream; enum dc_status status; DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger); + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) && + dc_is_virtual_signal(pipe_ctx->stream->signal)) + return; + if (!dc_is_virtual_signal(pipe_ctx->stream->signal)) { stream->link->link_enc->funcs->setup( stream->link->link_enc, @@ -2680,6 +2959,7 @@ void core_link_enable_stream( pipe_ctx->stream_res.stream_enc, &stream->timing, stream->output_color_space, + stream->use_vsc_sdp_for_colorimetry, stream->link->dpcd_caps.dprx_feature.bits.SST_SPLIT_SDP_CAP); if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) @@ -2703,18 +2983,21 @@ void core_link_enable_stream( pipe_ctx->stream_res.stream_enc, &stream->timing); - if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { bool apply_edp_fast_boot_optimization = pipe_ctx->stream->apply_edp_fast_boot_optimization; pipe_ctx->stream->apply_edp_fast_boot_optimization = false; resource_build_info_frame(pipe_ctx); - core_dc->hwss.update_info_frame(pipe_ctx); + dc->hwss.update_info_frame(pipe_ctx); /* Do not touch link on seamless boot optimization. */ if (pipe_ctx->stream->apply_seamless_boot_optimization) { pipe_ctx->stream->dpms_off = false; +#if defined(CONFIG_DRM_AMD_DC_HDCP) + update_psp_stream_config(pipe_ctx, false); +#endif return; } @@ -2722,6 +3005,9 @@ void core_link_enable_stream( if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP && apply_edp_fast_boot_optimization) { pipe_ctx->stream->dpms_off = false; +#if defined(CONFIG_DRM_AMD_DC_HDCP) + update_psp_stream_config(pipe_ctx, false); +#endif return; } @@ -2747,7 +3033,7 @@ void core_link_enable_stream( } } - core_dc->hwss.enable_audio_stream(pipe_ctx); + dc->hwss.enable_audio_stream(pipe_ctx); /* turn off otg test pattern if enable */ if (pipe_ctx->stream_res.tg->funcs->set_test_pattern) @@ -2755,44 +3041,54 @@ void core_link_enable_stream( CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, COLOR_DEPTH_UNDEFINED); - core_dc->hwss.enable_stream(pipe_ctx); + if (pipe_ctx->stream->timing.flags.DSC) { + if (dc_is_dp_signal(pipe_ctx->stream->signal) || + dc_is_virtual_signal(pipe_ctx->stream->signal)) + dp_set_dsc_enable(pipe_ctx, true); + } + dc->hwss.enable_stream(pipe_ctx); + + /* Set DPS PPS SDP (AKA "info frames") */ + if (pipe_ctx->stream->timing.flags.DSC) { + if (dc_is_dp_signal(pipe_ctx->stream->signal) || + dc_is_virtual_signal(pipe_ctx->stream->signal)) + dp_set_dsc_pps_sdp(pipe_ctx, true); + } if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) - allocate_mst_payload(pipe_ctx); + dc_link_allocate_mst_payload(pipe_ctx); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - if (pipe_ctx->stream->timing.flags.DSC && - (dc_is_dp_signal(pipe_ctx->stream->signal) || - dc_is_virtual_signal(pipe_ctx->stream->signal))) { - dp_set_dsc_enable(pipe_ctx, true); - pipe_ctx->stream_res.tg->funcs->wait_for_state( - pipe_ctx->stream_res.tg, - CRTC_STATE_VBLANK); - } -#endif - core_dc->hwss.unblank_stream(pipe_ctx, + dc->hwss.unblank_stream(pipe_ctx, &pipe_ctx->stream->link->cur_link_settings); if (dc_is_dp_signal(pipe_ctx->stream->signal)) enable_stream_features(pipe_ctx); - } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - else { // if (IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) +#if defined(CONFIG_DRM_AMD_DC_HDCP) + update_psp_stream_config(pipe_ctx, false); +#endif + } else { // if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) if (dc_is_dp_signal(pipe_ctx->stream->signal) || dc_is_virtual_signal(pipe_ctx->stream->signal)) dp_set_dsc_enable(pipe_ctx, true); } -#endif } -void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option) +void core_link_disable_stream(struct pipe_ctx *pipe_ctx) { - struct dc *core_dc = pipe_ctx->stream->ctx->dc; + struct dc *dc = pipe_ctx->stream->ctx->dc; struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->sink->link; - core_dc->hwss.blank_stream(pipe_ctx); + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) && + dc_is_virtual_signal(pipe_ctx->stream->signal)) + return; + +#if defined(CONFIG_DRM_AMD_DC_HDCP) + update_psp_stream_config(pipe_ctx, true); +#endif + + dc->hwss.blank_stream(pipe_ctx); if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) deallocate_mst_payload(pipe_ctx); @@ -2821,25 +3117,23 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option) write_i2c_redriver_setting(pipe_ctx, false); } } - core_dc->hwss.disable_stream(pipe_ctx, option); + dc->hwss.disable_stream(pipe_ctx); disable_link(pipe_ctx->stream->link, pipe_ctx->stream->signal); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - if (pipe_ctx->stream->timing.flags.DSC && - dc_is_dp_signal(pipe_ctx->stream->signal)) { - dp_set_dsc_enable(pipe_ctx, false); + if (pipe_ctx->stream->timing.flags.DSC) { + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + dp_set_dsc_enable(pipe_ctx, false); } -#endif } void core_link_set_avmute(struct pipe_ctx *pipe_ctx, bool enable) { - struct dc *core_dc = pipe_ctx->stream->ctx->dc; + struct dc *dc = pipe_ctx->stream->ctx->dc; - if (pipe_ctx->stream->signal != SIGNAL_TYPE_HDMI_TYPE_A) + if (!dc_is_hdmi_signal(pipe_ctx->stream->signal)) return; - core_dc->hwss.set_avmute(pipe_ctx, enable); + dc->hwss.set_avmute(pipe_ctx, enable); } /** @@ -2897,13 +3191,11 @@ uint32_t dc_bandwidth_in_kbps_from_timing( uint32_t bits_per_channel = 0; uint32_t kbps; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT if (timing->flags.DSC) { kbps = (timing->pix_clk_100hz * timing->dsc_cfg.bits_per_pixel); kbps = kbps / 160 + ((kbps % 160) ? 1 : 0); return kbps; } -#endif switch (timing->display_color_depth) { case COLOR_DEPTH_666: @@ -2999,8 +3291,10 @@ void dc_link_set_preferred_link_settings(struct dc *dc, for (i = 0; i < MAX_PIPES; i++) { pipe = &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe->stream && pipe->stream->link) { - if (pipe->stream->link == link) + if (pipe->stream->link == link) { + link_stream = pipe->stream; break; + } } } @@ -3008,20 +3302,40 @@ void dc_link_set_preferred_link_settings(struct dc *dc, if (i == MAX_PIPES) return; - link_stream = link->dc->current_state->res_ctx.pipe_ctx[i].stream; - /* Cannot retrain link if backend is off */ if (link_stream->dpms_off) return; - if (link_stream) - decide_link_settings(link_stream, &store_settings); + decide_link_settings(link_stream, &store_settings); if ((store_settings.lane_count != LANE_COUNT_UNKNOWN) && (store_settings.link_rate != LINK_RATE_UNKNOWN)) dp_retrain_link_dp_test(link, &store_settings, false); } +void dc_link_set_preferred_training_settings(struct dc *dc, + struct dc_link_settings *link_setting, + struct dc_link_training_overrides *lt_overrides, + struct dc_link *link, + bool skip_immediate_retrain) +{ + if (lt_overrides != NULL) + link->preferred_training_settings = *lt_overrides; + else + memset(&link->preferred_training_settings, 0, sizeof(link->preferred_training_settings)); + + if (link_setting != NULL) { + link->preferred_link_setting = *link_setting; + } else { + link->preferred_link_setting.lane_count = LANE_COUNT_UNKNOWN; + link->preferred_link_setting.link_rate = LINK_RATE_UNKNOWN; + } + + /* Retrain now, or wait until next stream update to apply */ + if (skip_immediate_retrain == false) + dc_link_set_preferred_link_settings(dc, &link->preferred_link_setting, link); +} + void dc_link_enable_hpd(const struct dc_link *link) { dc_link_dp_enable_hpd(link); @@ -3032,9 +3346,9 @@ void dc_link_disable_hpd(const struct dc_link *link) dc_link_dp_disable_hpd(link); } - void dc_link_set_test_pattern(struct dc_link *link, enum dp_test_pattern test_pattern, + enum dp_test_pattern_color_space test_pattern_color_space, const struct link_training_settings *p_link_settings, const unsigned char *p_custom_pattern, unsigned int cust_pattern_size) @@ -3043,6 +3357,7 @@ void dc_link_set_test_pattern(struct dc_link *link, dc_link_dp_set_test_pattern( link, test_pattern, + test_pattern_color_space, p_link_settings, p_custom_pattern, cust_pattern_size); @@ -3058,7 +3373,6 @@ uint32_t dc_link_bandwidth_kbps( link_bw_kbps *= 8; /* 8 bits per byte*/ link_bw_kbps *= link_setting->lane_count; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT if (link->dpcd_caps.fec_cap.bits.FEC_CAPABLE) { /* Account for FEC overhead. * We have to do it based on caps, @@ -3083,7 +3397,6 @@ uint32_t dc_link_bandwidth_kbps( link_bw_kbps = mul_u64_u32_shr(BIT_ULL(32) * 970LL / 1000, link_bw_kbps, 32); } -#endif return link_bw_kbps; @@ -3097,3 +3410,10 @@ const struct dc_link_settings *dc_link_get_link_cap( return &link->preferred_link_setting; return &link->verified_link_cap; } + +void dc_link_overwrite_extended_receiver_cap( + struct dc_link *link) +{ + dp_overwrite_extended_receiver_cap(link); +} + diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c index e6da8506128b..a49c10d5df26 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c @@ -187,7 +187,7 @@ void dal_ddc_i2c_payloads_add( } -static void construct( +static void ddc_service_construct( struct ddc_service *ddc_service, struct ddc_service_init_data *init_data) { @@ -206,7 +206,10 @@ static void construct( ddc_service->ddc_pin = NULL; } else { hw_info.ddc_channel = i2c_info.i2c_line; - hw_info.hw_supported = i2c_info.i2c_hw_assist; + if (ddc_service->link != NULL) + hw_info.hw_supported = i2c_info.i2c_hw_assist; + else + hw_info.hw_supported = false; ddc_service->ddc_pin = dal_gpio_create_ddc( gpio_service, @@ -236,11 +239,11 @@ struct ddc_service *dal_ddc_service_create( if (!ddc_service) return NULL; - construct(ddc_service, init_data); + ddc_service_construct(ddc_service, init_data); return ddc_service; } -static void destruct(struct ddc_service *ddc) +static void ddc_service_destruct(struct ddc_service *ddc) { if (ddc->ddc_pin) dal_gpio_destroy_ddc(&ddc->ddc_pin); @@ -252,7 +255,7 @@ void dal_ddc_service_destroy(struct ddc_service **ddc) BREAK_TO_DEBUGGER(); return; } - destruct(*ddc); + ddc_service_destruct(*ddc); kfree(*ddc); *ddc = NULL; } @@ -294,7 +297,7 @@ static uint32_t defer_delay_converter_wa( { struct dc_link *link = ddc->link; - if (link->dpcd_caps.branch_dev_id == DP_BRANCH_DEVICE_ID_4 && + if (link->dpcd_caps.branch_dev_id == DP_BRANCH_DEVICE_ID_0080E1 && !memcmp(link->dpcd_caps.branch_dev_name, DP_DVI_CONVERTER_ID_4, sizeof(link->dpcd_caps.branch_dev_name))) @@ -374,6 +377,7 @@ void dal_ddc_service_i2c_query_dp_dual_mode_adaptor( enum display_dongle_type *dongle = &sink_cap->dongle_type; uint8_t type2_dongle_buf[DP_ADAPTOR_TYPE2_SIZE]; bool is_type2_dongle = false; + int retry_count = 2; struct dp_hdmi_dongle_signature_data *dongle_signature; /* Assume we have no valid DP passive dongle connected */ @@ -386,13 +390,24 @@ void dal_ddc_service_i2c_query_dp_dual_mode_adaptor( DP_HDMI_DONGLE_ADDRESS, type2_dongle_buf, sizeof(type2_dongle_buf))) { - *dongle = DISPLAY_DONGLE_DP_DVI_DONGLE; - sink_cap->max_hdmi_pixel_clock = DP_ADAPTOR_DVI_MAX_TMDS_CLK; + /* Passive HDMI dongles can sometimes fail here without retrying*/ + while (retry_count > 0) { + if (i2c_read(ddc, + DP_HDMI_DONGLE_ADDRESS, + type2_dongle_buf, + sizeof(type2_dongle_buf))) + break; + retry_count--; + } + if (retry_count == 0) { + *dongle = DISPLAY_DONGLE_DP_DVI_DONGLE; + sink_cap->max_hdmi_pixel_clock = DP_ADAPTOR_DVI_MAX_TMDS_CLK; - CONN_DATA_DETECT(ddc->link, type2_dongle_buf, sizeof(type2_dongle_buf), - "DP-DVI passive dongle %dMhz: ", - DP_ADAPTOR_DVI_MAX_TMDS_CLK / 1000); - return; + CONN_DATA_DETECT(ddc->link, type2_dongle_buf, sizeof(type2_dongle_buf), + "DP-DVI passive dongle %dMhz: ", + DP_ADAPTOR_DVI_MAX_TMDS_CLK / 1000); + return; + } } /* Check if Type 2 dongle.*/ @@ -496,7 +511,7 @@ bool dal_ddc_service_query_ddc_data( uint8_t *read_buf, uint32_t read_size) { - bool ret; + bool ret = false; uint32_t payload_size = dal_ddc_service_is_in_aux_transaction_mode(ddc) ? DEFAULT_AUX_MAX_DATA_SIZE : EDID_SEGMENT_SIZE; @@ -515,34 +530,32 @@ bool dal_ddc_service_query_ddc_data( /*TODO: len of payload data for i2c and aux is uint8!!!!, * but we want to read 256 over i2c!!!!*/ if (dal_ddc_service_is_in_aux_transaction_mode(ddc)) { - struct aux_payload write_payload = { - .i2c_over_aux = true, - .write = true, - .mot = true, - .address = address, - .length = write_size, - .data = write_buf, - .reply = NULL, - .defer_delay = get_defer_delay(ddc), - }; - - struct aux_payload read_payload = { - .i2c_over_aux = true, - .write = false, - .mot = false, - .address = address, - .length = read_size, - .data = read_buf, - .reply = NULL, - .defer_delay = get_defer_delay(ddc), - }; - - ret = dc_link_aux_transfer_with_retries(ddc, &write_payload); + struct aux_payload payload; + bool read_available = true; + + payload.i2c_over_aux = true; + payload.address = address; + payload.reply = NULL; + payload.defer_delay = get_defer_delay(ddc); + + if (write_size != 0) { + payload.write = true; + payload.mot = false; + payload.length = write_size; + payload.data = write_buf; + + ret = dal_ddc_submit_aux_command(ddc, &payload); + read_available = ret; + } - if (!ret) - return false; + if (read_size != 0 && read_available) { + payload.write = false; + payload.mot = false; + payload.length = read_size; + payload.data = read_buf; - ret = dc_link_aux_transfer_with_retries(ddc, &read_payload); + ret = dal_ddc_submit_aux_command(ddc, &payload); + } } else { struct i2c_payloads *payloads = dal_ddc_i2c_payloads_create(ddc->ctx, payloads_num); @@ -573,6 +586,41 @@ bool dal_ddc_service_query_ddc_data( return ret; } +bool dal_ddc_submit_aux_command(struct ddc_service *ddc, + struct aux_payload *payload) +{ + uint32_t retrieved = 0; + bool ret = false; + + if (!ddc) + return false; + + if (!payload) + return false; + + do { + struct aux_payload current_payload; + bool is_end_of_payload = (retrieved + DEFAULT_AUX_MAX_DATA_SIZE) > + payload->length ? true : false; + + current_payload.address = payload->address; + current_payload.data = &payload->data[retrieved]; + current_payload.defer_delay = payload->defer_delay; + current_payload.i2c_over_aux = payload->i2c_over_aux; + current_payload.length = is_end_of_payload ? + payload->length - retrieved : DEFAULT_AUX_MAX_DATA_SIZE; + current_payload.mot = !is_end_of_payload; + current_payload.reply = payload->reply; + current_payload.write = payload->write; + + ret = dc_link_aux_transfer_with_retries(ddc, ¤t_payload); + + retrieved += current_payload.length; + } while (retrieved < payload->length && ret == true); + + return ret; +} + /* dc_link_aux_transfer_raw() - Attempt to transfer * the given aux payload. This function does not perform * retries or handle error states. The reply is returned @@ -601,6 +649,19 @@ bool dc_link_aux_transfer_with_retries(struct ddc_service *ddc, return dce_aux_transfer_with_retries(ddc, payload); } + +uint32_t dc_link_aux_configure_timeout(struct ddc_service *ddc, + uint32_t timeout) +{ + uint32_t prev_timeout = 0; + struct ddc *ddc_pin = ddc->ddc_pin; + + if (ddc->ctx->dc->res_pool->engines[ddc_pin->pin_data->en]->funcs->configure_timeout) + prev_timeout = + ddc->ctx->dc->res_pool->engines[ddc_pin->pin_data->en]->funcs->configure_timeout(ddc, timeout); + return prev_timeout; +} + /*test only function*/ void dal_ddc_service_set_ddc_pin( struct ddc_service *ddc_service, diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index 2c7aaed907b9..cb731c1d30b1 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -4,12 +4,8 @@ #include "dc_link_dp.h" #include "dm_helpers.h" #include "opp.h" -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #include "dsc.h" -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "resource.h" -#endif #include "inc/core_types.h" #include "link_hwss.h" @@ -21,6 +17,9 @@ #define DC_LOGGER \ link->ctx->logger + +#define DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE 0x50 + /* maximum pre emphasis level allowed for each voltage swing level*/ static const enum dc_pre_emphasis voltage_swing_to_pre_emphasis[] = { PRE_EMPHASIS_LEVEL3, @@ -49,7 +48,7 @@ static struct dc_link_settings get_common_supported_link_settings( struct dc_link_settings link_setting_a, struct dc_link_settings link_setting_b); -static void wait_for_training_aux_rd_interval( +static uint32_t get_training_aux_rd_interval( struct dc_link *link, uint32_t default_wait_in_micro_secs) { @@ -68,15 +67,21 @@ static void wait_for_training_aux_rd_interval( sizeof(training_rd_interval)); if (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL) - default_wait_in_micro_secs = - training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000; + default_wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000; } - udelay(default_wait_in_micro_secs); + return default_wait_in_micro_secs; +} + +static void wait_for_training_aux_rd_interval( + struct dc_link *link, + uint32_t wait_in_micro_secs) +{ + udelay(wait_in_micro_secs); DC_LOG_HW_LINK_TRAINING("%s:\n wait = %d\n", __func__, - default_wait_in_micro_secs); + wait_in_micro_secs); } static void dpcd_set_training_pattern( @@ -95,27 +100,27 @@ static void dpcd_set_training_pattern( dpcd_pattern.v1_4.TRAINING_PATTERN_SET); } -static enum hw_dp_training_pattern get_supported_tp(struct dc_link *link) +static enum dc_dp_training_pattern get_supported_tp(struct dc_link *link) { - enum hw_dp_training_pattern highest_tp = HW_DP_TRAINING_PATTERN_2; + enum dc_dp_training_pattern highest_tp = DP_TRAINING_PATTERN_SEQUENCE_2; struct encoder_feature_support *features = &link->link_enc->features; struct dpcd_caps *dpcd_caps = &link->dpcd_caps; if (features->flags.bits.IS_TPS3_CAPABLE) - highest_tp = HW_DP_TRAINING_PATTERN_3; + highest_tp = DP_TRAINING_PATTERN_SEQUENCE_3; if (features->flags.bits.IS_TPS4_CAPABLE) - highest_tp = HW_DP_TRAINING_PATTERN_4; + highest_tp = DP_TRAINING_PATTERN_SEQUENCE_4; if (dpcd_caps->max_down_spread.bits.TPS4_SUPPORTED && - highest_tp >= HW_DP_TRAINING_PATTERN_4) - return HW_DP_TRAINING_PATTERN_4; + highest_tp >= DP_TRAINING_PATTERN_SEQUENCE_4) + return DP_TRAINING_PATTERN_SEQUENCE_4; if (dpcd_caps->max_ln_count.bits.TPS3_SUPPORTED && - highest_tp >= HW_DP_TRAINING_PATTERN_3) - return HW_DP_TRAINING_PATTERN_3; + highest_tp >= DP_TRAINING_PATTERN_SEQUENCE_3) + return DP_TRAINING_PATTERN_SEQUENCE_3; - return HW_DP_TRAINING_PATTERN_2; + return DP_TRAINING_PATTERN_SEQUENCE_2; } static void dpcd_set_link_settings( @@ -126,7 +131,7 @@ static void dpcd_set_link_settings( union down_spread_ctrl downspread = { {0} }; union lane_count_set lane_count_set = { {0} }; - enum hw_dp_training_pattern hw_tr_pattern; + enum dc_dp_training_pattern dp_tr_pattern; downspread.raw = (uint8_t) (lt_settings->link_settings.link_spread); @@ -134,21 +139,21 @@ static void dpcd_set_link_settings( lane_count_set.bits.LANE_COUNT_SET = lt_settings->link_settings.lane_count; - lane_count_set.bits.ENHANCED_FRAMING = 1; - + lane_count_set.bits.ENHANCED_FRAMING = lt_settings->enhanced_framing; lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = 0; - hw_tr_pattern = get_supported_tp(link); - if (hw_tr_pattern != HW_DP_TRAINING_PATTERN_4) { + dp_tr_pattern = get_supported_tp(link); + + if (dp_tr_pattern != DP_TRAINING_PATTERN_SEQUENCE_4) { lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED; } core_link_write_dpcd(link, DP_DOWNSPREAD_CTRL, - &downspread.raw, sizeof(downspread)); + &downspread.raw, sizeof(downspread)); core_link_write_dpcd(link, DP_LANE_COUNT_SET, - &lane_count_set.raw, 1); + &lane_count_set.raw, 1); if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_14 && lt_settings->link_settings.use_link_rate_set == true) { @@ -162,46 +167,47 @@ static void dpcd_set_link_settings( } if (rate) { - DC_LOG_HW_LINK_TRAINING("%s\n %x rate = %x\n %x lane = %x\n %x spread = %x\n", + DC_LOG_HW_LINK_TRAINING("%s\n %x rate = %x\n %x lane = %x framing = %x\n %x spread = %x\n", __func__, DP_LINK_BW_SET, lt_settings->link_settings.link_rate, DP_LANE_COUNT_SET, lt_settings->link_settings.lane_count, + lt_settings->enhanced_framing, DP_DOWNSPREAD_CTRL, lt_settings->link_settings.link_spread); } else { - DC_LOG_HW_LINK_TRAINING("%s\n %x rate set = %x\n %x lane = %x\n %x spread = %x\n", + DC_LOG_HW_LINK_TRAINING("%s\n %x rate set = %x\n %x lane = %x framing = %x\n %x spread = %x\n", __func__, DP_LINK_RATE_SET, lt_settings->link_settings.link_rate_set, DP_LANE_COUNT_SET, lt_settings->link_settings.lane_count, + lt_settings->enhanced_framing, DP_DOWNSPREAD_CTRL, lt_settings->link_settings.link_spread); } - } static enum dpcd_training_patterns - hw_training_pattern_to_dpcd_training_pattern( + dc_dp_training_pattern_to_dpcd_training_pattern( struct dc_link *link, - enum hw_dp_training_pattern pattern) + enum dc_dp_training_pattern pattern) { enum dpcd_training_patterns dpcd_tr_pattern = DPCD_TRAINING_PATTERN_VIDEOIDLE; switch (pattern) { - case HW_DP_TRAINING_PATTERN_1: + case DP_TRAINING_PATTERN_SEQUENCE_1: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_1; break; - case HW_DP_TRAINING_PATTERN_2: + case DP_TRAINING_PATTERN_SEQUENCE_2: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_2; break; - case HW_DP_TRAINING_PATTERN_3: + case DP_TRAINING_PATTERN_SEQUENCE_3: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_3; break; - case HW_DP_TRAINING_PATTERN_4: + case DP_TRAINING_PATTERN_SEQUENCE_4: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_4; break; default: @@ -212,37 +218,55 @@ static enum dpcd_training_patterns } return dpcd_tr_pattern; +} +static inline bool is_repeater(struct dc_link *link, uint32_t offset) +{ + return (!link->is_lttpr_mode_transparent && offset != 0); } static void dpcd_set_lt_pattern_and_lane_settings( struct dc_link *link, const struct link_training_settings *lt_settings, - enum hw_dp_training_pattern pattern) + enum dc_dp_training_pattern pattern, + uint32_t offset) { union dpcd_training_lane dpcd_lane[LANE_COUNT_DP_MAX] = { { {0} } }; - const uint32_t dpcd_base_lt_offset = - DP_TRAINING_PATTERN_SET; + + uint32_t dpcd_base_lt_offset; + uint8_t dpcd_lt_buffer[5] = {0}; union dpcd_training_pattern dpcd_pattern = { {0} }; uint32_t lane; uint32_t size_in_bytes; bool edp_workaround = false; /* TODO link_prop.INTERNAL */ + dpcd_base_lt_offset = DP_TRAINING_PATTERN_SET; + + if (is_repeater(link, offset)) + dpcd_base_lt_offset = DP_TRAINING_PATTERN_SET_PHY_REPEATER1 + + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); /***************************************************************** * DpcdAddress_TrainingPatternSet *****************************************************************/ dpcd_pattern.v1_4.TRAINING_PATTERN_SET = - hw_training_pattern_to_dpcd_training_pattern(link, pattern); + dc_dp_training_pattern_to_dpcd_training_pattern(link, pattern); - dpcd_lt_buffer[DP_TRAINING_PATTERN_SET - dpcd_base_lt_offset] + dpcd_lt_buffer[DP_TRAINING_PATTERN_SET - DP_TRAINING_PATTERN_SET] = dpcd_pattern.raw; - DC_LOG_HW_LINK_TRAINING("%s\n %x pattern = %x\n", - __func__, - DP_TRAINING_PATTERN_SET, - dpcd_pattern.v1_4.TRAINING_PATTERN_SET); - + if (is_repeater(link, offset)) { + DC_LOG_HW_LINK_TRAINING("%s\n LTTPR Repeater ID: %d\n 0x%X pattern = %x\n", + __func__, + offset, + dpcd_base_lt_offset, + dpcd_pattern.v1_4.TRAINING_PATTERN_SET); + } else { + DC_LOG_HW_LINK_TRAINING("%s\n 0x%X pattern = %x\n", + __func__, + dpcd_base_lt_offset, + dpcd_pattern.v1_4.TRAINING_PATTERN_SET); + } /***************************************************************** * DpcdAddress_Lane0Set -> DpcdAddress_Lane3Set *****************************************************************/ @@ -262,24 +286,35 @@ static void dpcd_set_lt_pattern_and_lane_settings( PRE_EMPHASIS_MAX_LEVEL ? 1 : 0); } - /* concatinate everything into one buffer*/ + /* concatenate everything into one buffer*/ size_in_bytes = lt_settings->link_settings.lane_count * sizeof(dpcd_lane[0]); // 0x00103 - 0x00102 memmove( - &dpcd_lt_buffer[DP_TRAINING_LANE0_SET - dpcd_base_lt_offset], + &dpcd_lt_buffer[DP_TRAINING_LANE0_SET - DP_TRAINING_PATTERN_SET], dpcd_lane, size_in_bytes); - DC_LOG_HW_LINK_TRAINING("%s:\n %x VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", - __func__, - DP_TRAINING_LANE0_SET, - dpcd_lane[0].bits.VOLTAGE_SWING_SET, - dpcd_lane[0].bits.PRE_EMPHASIS_SET, - dpcd_lane[0].bits.MAX_SWING_REACHED, - dpcd_lane[0].bits.MAX_PRE_EMPHASIS_REACHED); - + if (is_repeater(link, offset)) { + DC_LOG_HW_LINK_TRAINING("%s:\n LTTPR Repeater ID: %d\n" + " 0x%X VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", + __func__, + offset, + dpcd_base_lt_offset, + dpcd_lane[0].bits.VOLTAGE_SWING_SET, + dpcd_lane[0].bits.PRE_EMPHASIS_SET, + dpcd_lane[0].bits.MAX_SWING_REACHED, + dpcd_lane[0].bits.MAX_PRE_EMPHASIS_REACHED); + } else { + DC_LOG_HW_LINK_TRAINING("%s:\n 0x%X VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", + __func__, + dpcd_base_lt_offset, + dpcd_lane[0].bits.VOLTAGE_SWING_SET, + dpcd_lane[0].bits.PRE_EMPHASIS_SET, + dpcd_lane[0].bits.MAX_SWING_REACHED, + dpcd_lane[0].bits.MAX_PRE_EMPHASIS_REACHED); + } if (edp_workaround) { /* for eDP write in 2 parts because the 5-byte burst is * causing issues on some eDP panels (EPR#366724) @@ -346,12 +381,20 @@ static void update_drive_settings( { uint32_t lane; for (lane = 0; lane < src.link_settings.lane_count; lane++) { - dest->lane_settings[lane].VOLTAGE_SWING = - src.lane_settings[lane].VOLTAGE_SWING; - dest->lane_settings[lane].PRE_EMPHASIS = - src.lane_settings[lane].PRE_EMPHASIS; - dest->lane_settings[lane].POST_CURSOR2 = - src.lane_settings[lane].POST_CURSOR2; + if (dest->voltage_swing == NULL) + dest->lane_settings[lane].VOLTAGE_SWING = src.lane_settings[lane].VOLTAGE_SWING; + else + dest->lane_settings[lane].VOLTAGE_SWING = *dest->voltage_swing; + + if (dest->pre_emphasis == NULL) + dest->lane_settings[lane].PRE_EMPHASIS = src.lane_settings[lane].PRE_EMPHASIS; + else + dest->lane_settings[lane].PRE_EMPHASIS = *dest->pre_emphasis; + + if (dest->post_cursor2 == NULL) + dest->lane_settings[lane].POST_CURSOR2 = src.lane_settings[lane].POST_CURSOR2; + else + dest->lane_settings[lane].POST_CURSOR2 = *dest->post_cursor2; } } @@ -481,8 +524,12 @@ static void get_lane_status_and_drive_settings( const struct link_training_settings *link_training_setting, union lane_status *ln_status, union lane_align_status_updated *ln_status_updated, - struct link_training_settings *req_settings) + struct link_training_settings *req_settings, + uint32_t offset) { + unsigned int lane01_status_address = DP_LANE0_1_STATUS; + uint8_t lane_adjust_offset = 4; + unsigned int lane01_adjust_address; uint8_t dpcd_buf[6] = {0}; union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = { { {0} } }; struct link_training_settings request_settings = { {0} }; @@ -490,9 +537,16 @@ static void get_lane_status_and_drive_settings( memset(req_settings, '\0', sizeof(struct link_training_settings)); + if (is_repeater(link, offset)) { + lane01_status_address = + DP_LANE0_1_STATUS_PHY_REPEATER1 + + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + lane_adjust_offset = 3; + } + core_link_read_dpcd( link, - DP_LANE0_1_STATUS, + lane01_status_address, (uint8_t *)(dpcd_buf), sizeof(dpcd_buf)); @@ -503,22 +557,47 @@ static void get_lane_status_and_drive_settings( ln_status[lane].raw = get_nibble_at_index(&dpcd_buf[0], lane); dpcd_lane_adjust[lane].raw = - get_nibble_at_index(&dpcd_buf[4], lane); + get_nibble_at_index(&dpcd_buf[lane_adjust_offset], lane); } ln_status_updated->raw = dpcd_buf[2]; - DC_LOG_HW_LINK_TRAINING("%s:\n%x Lane01Status = %x\n %x Lane23Status = %x\n ", - __func__, - DP_LANE0_1_STATUS, dpcd_buf[0], - DP_LANE2_3_STATUS, dpcd_buf[1]); - - DC_LOG_HW_LINK_TRAINING("%s:\n %x Lane01AdjustRequest = %x\n %x Lane23AdjustRequest = %x\n", - __func__, - DP_ADJUST_REQUEST_LANE0_1, - dpcd_buf[4], - DP_ADJUST_REQUEST_LANE2_3, - dpcd_buf[5]); + if (is_repeater(link, offset)) { + DC_LOG_HW_LINK_TRAINING("%s:\n LTTPR Repeater ID: %d\n" + " 0x%X Lane01Status = %x\n 0x%X Lane23Status = %x\n ", + __func__, + offset, + lane01_status_address, dpcd_buf[0], + lane01_status_address + 1, dpcd_buf[1]); + } else { + DC_LOG_HW_LINK_TRAINING("%s:\n 0x%X Lane01Status = %x\n 0x%X Lane23Status = %x\n ", + __func__, + lane01_status_address, dpcd_buf[0], + lane01_status_address + 1, dpcd_buf[1]); + } + lane01_adjust_address = DP_ADJUST_REQUEST_LANE0_1; + + if (is_repeater(link, offset)) + lane01_adjust_address = DP_ADJUST_REQUEST_LANE0_1_PHY_REPEATER1 + + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + + if (is_repeater(link, offset)) { + DC_LOG_HW_LINK_TRAINING("%s:\n LTTPR Repeater ID: %d\n" + " 0x%X Lane01AdjustRequest = %x\n 0x%X Lane23AdjustRequest = %x\n", + __func__, + offset, + lane01_adjust_address, + dpcd_buf[lane_adjust_offset], + lane01_adjust_address + 1, + dpcd_buf[lane_adjust_offset + 1]); + } else { + DC_LOG_HW_LINK_TRAINING("%s:\n 0x%X Lane01AdjustRequest = %x\n 0x%X Lane23AdjustRequest = %x\n", + __func__, + lane01_adjust_address, + dpcd_buf[lane_adjust_offset], + lane01_adjust_address + 1, + dpcd_buf[lane_adjust_offset + 1]); + } /*copy to req_settings*/ request_settings.link_settings.lane_count = @@ -557,10 +636,18 @@ static void get_lane_status_and_drive_settings( static void dpcd_set_lane_settings( struct dc_link *link, - const struct link_training_settings *link_training_setting) + const struct link_training_settings *link_training_setting, + uint32_t offset) { union dpcd_training_lane dpcd_lane[LANE_COUNT_DP_MAX] = {{{0}}}; uint32_t lane; + unsigned int lane0_set_address; + + lane0_set_address = DP_TRAINING_LANE0_SET; + + if (is_repeater(link, offset)) + lane0_set_address = DP_TRAINING_LANE0_SET_PHY_REPEATER1 + + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); for (lane = 0; lane < (uint32_t)(link_training_setting-> @@ -583,7 +670,7 @@ static void dpcd_set_lane_settings( } core_link_write_dpcd(link, - DP_TRAINING_LANE0_SET, + lane0_set_address, (uint8_t *)(dpcd_lane), link_training_setting->link_settings.lane_count); @@ -606,14 +693,26 @@ static void dpcd_set_lane_settings( } */ - DC_LOG_HW_LINK_TRAINING("%s\n %x VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", - __func__, - DP_TRAINING_LANE0_SET, - dpcd_lane[0].bits.VOLTAGE_SWING_SET, - dpcd_lane[0].bits.PRE_EMPHASIS_SET, - dpcd_lane[0].bits.MAX_SWING_REACHED, - dpcd_lane[0].bits.MAX_PRE_EMPHASIS_REACHED); + if (is_repeater(link, offset)) { + DC_LOG_HW_LINK_TRAINING("%s\n LTTPR Repeater ID: %d\n" + " 0x%X VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", + __func__, + offset, + lane0_set_address, + dpcd_lane[0].bits.VOLTAGE_SWING_SET, + dpcd_lane[0].bits.PRE_EMPHASIS_SET, + dpcd_lane[0].bits.MAX_SWING_REACHED, + dpcd_lane[0].bits.MAX_PRE_EMPHASIS_REACHED); + } else { + DC_LOG_HW_LINK_TRAINING("%s\n 0x%X VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", + __func__, + lane0_set_address, + dpcd_lane[0].bits.VOLTAGE_SWING_SET, + dpcd_lane[0].bits.PRE_EMPHASIS_SET, + dpcd_lane[0].bits.MAX_SWING_REACHED, + dpcd_lane[0].bits.MAX_PRE_EMPHASIS_REACHED); + } link->cur_lane_setting = link_training_setting->lane_settings[0]; } @@ -633,17 +732,6 @@ static bool is_max_vs_reached( } -void dc_link_dp_set_drive_settings( - struct dc_link *link, - struct link_training_settings *lt_settings) -{ - /* program ASIC PHY settings*/ - dp_set_hw_lane_settings(link, lt_settings); - - /* Notify DP sink the PHY settings from source */ - dpcd_set_lane_settings(link, lt_settings); -} - static bool perform_post_lt_adj_req_sequence( struct dc_link *link, struct link_training_settings *lt_settings) @@ -676,7 +764,8 @@ static bool perform_post_lt_adj_req_sequence( lt_settings, dpcd_lane_status, &dpcd_lane_status_updated, - &req_settings); + &req_settings, + DPRX); if (dpcd_lane_status_updated.bits. POST_LT_ADJ_REQ_IN_PROGRESS == 0) @@ -733,6 +822,31 @@ static bool perform_post_lt_adj_req_sequence( } +/* Only used for channel equalization */ +static uint32_t translate_training_aux_read_interval(uint32_t dpcd_aux_read_interval) +{ + unsigned int aux_rd_interval_us = 400; + + switch (dpcd_aux_read_interval) { + case 0x01: + aux_rd_interval_us = 400; + break; + case 0x02: + aux_rd_interval_us = 4000; + break; + case 0x03: + aux_rd_interval_us = 8000; + break; + case 0x04: + aux_rd_interval_us = 16000; + break; + default: + break; + } + + return aux_rd_interval_us; +} + static enum link_training_result get_cr_failure(enum dc_lane_count ln_count, union lane_status *dpcd_lane_status) { @@ -751,37 +865,55 @@ static enum link_training_result get_cr_failure(enum dc_lane_count ln_count, static enum link_training_result perform_channel_equalization_sequence( struct dc_link *link, - struct link_training_settings *lt_settings) + struct link_training_settings *lt_settings, + uint32_t offset) { struct link_training_settings req_settings; - enum hw_dp_training_pattern hw_tr_pattern; + enum dc_dp_training_pattern tr_pattern; uint32_t retries_ch_eq; + uint32_t wait_time_microsec; enum dc_lane_count lane_count = lt_settings->link_settings.lane_count; union lane_align_status_updated dpcd_lane_status_updated = { {0} }; union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX] = { { {0} } }; - hw_tr_pattern = get_supported_tp(link); + /* Note: also check that TPS4 is a supported feature*/ - dp_set_hw_training_pattern(link, hw_tr_pattern); + tr_pattern = lt_settings->pattern_for_eq; + + if (is_repeater(link, offset)) + tr_pattern = DP_TRAINING_PATTERN_SEQUENCE_4; + + dp_set_hw_training_pattern(link, tr_pattern, offset); for (retries_ch_eq = 0; retries_ch_eq <= LINK_TRAINING_MAX_RETRY_COUNT; retries_ch_eq++) { - dp_set_hw_lane_settings(link, lt_settings); + dp_set_hw_lane_settings(link, lt_settings, offset); /* 2. update DPCD*/ if (!retries_ch_eq) /* EPR #361076 - write as a 5-byte burst, - * but only for the 1-st iteration*/ + * but only for the 1-st iteration + */ + dpcd_set_lt_pattern_and_lane_settings( link, lt_settings, - hw_tr_pattern); + tr_pattern, offset); else - dpcd_set_lane_settings(link, lt_settings); + dpcd_set_lane_settings(link, lt_settings, offset); /* 3. wait for receiver to lock-on*/ - wait_for_training_aux_rd_interval(link, 400); + wait_time_microsec = lt_settings->eq_pattern_time; + + if (is_repeater(link, offset)) + wait_time_microsec = + translate_training_aux_read_interval( + link->dpcd_caps.lttpr_caps.aux_rd_interval[offset - 1]); + + wait_for_training_aux_rd_interval( + link, + wait_time_microsec); /* 4. Read lane status and requested * drive settings as set by the sink*/ @@ -791,7 +923,8 @@ static enum link_training_result perform_channel_equalization_sequence( lt_settings, dpcd_lane_status, &dpcd_lane_status_updated, - &req_settings); + &req_settings, + offset); /* 5. check CR done*/ if (!is_cr_done(lane_count, dpcd_lane_status)) @@ -810,34 +943,26 @@ static enum link_training_result perform_channel_equalization_sequence( return LINK_TRAINING_EQ_FAIL_EQ; } +#define TRAINING_AUX_RD_INTERVAL 100 //us static enum link_training_result perform_clock_recovery_sequence( struct dc_link *link, - struct link_training_settings *lt_settings) + struct link_training_settings *lt_settings, + uint32_t offset) { uint32_t retries_cr; uint32_t retry_count; - uint32_t lane; + uint32_t wait_time_microsec; struct link_training_settings req_settings; - enum dc_lane_count lane_count = - lt_settings->link_settings.lane_count; - enum hw_dp_training_pattern hw_tr_pattern = HW_DP_TRAINING_PATTERN_1; + enum dc_lane_count lane_count = lt_settings->link_settings.lane_count; + enum dc_dp_training_pattern tr_pattern = DP_TRAINING_PATTERN_SEQUENCE_1; union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX]; union lane_align_status_updated dpcd_lane_status_updated; retries_cr = 0; retry_count = 0; - /* initial drive setting (VS/PE/PC2)*/ - for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { - lt_settings->lane_settings[lane].VOLTAGE_SWING = - VOLTAGE_SWING_LEVEL0; - lt_settings->lane_settings[lane].PRE_EMPHASIS = - PRE_EMPHASIS_DISABLED; - lt_settings->lane_settings[lane].POST_CURSOR2 = - POST_CURSOR2_DISABLED; - } - dp_set_hw_training_pattern(link, hw_tr_pattern); + dp_set_hw_training_pattern(link, tr_pattern, offset); /* najeeb - The synaptics MST hub can put the LT in * infinite loop by switching the VS @@ -845,7 +970,7 @@ static enum link_training_result perform_clock_recovery_sequence( /* between level 0 and level 1 continuously, here * we try for CR lock for LinkTrainingMaxCRRetry count*/ while ((retries_cr < LINK_TRAINING_MAX_RETRY_COUNT) && - (retry_count < LINK_TRAINING_MAX_CR_RETRY)) { + (retry_count < LINK_TRAINING_MAX_CR_RETRY)) { memset(&dpcd_lane_status, '\0', sizeof(dpcd_lane_status)); memset(&dpcd_lane_status_updated, '\0', @@ -854,25 +979,33 @@ static enum link_training_result perform_clock_recovery_sequence( /* 1. call HWSS to set lane settings*/ dp_set_hw_lane_settings( link, - lt_settings); + lt_settings, + offset); /* 2. update DPCD of the receiver*/ - if (!retries_cr) + if (!retry_count) /* EPR #361076 - write as a 5-byte burst, * but only for the 1-st iteration.*/ dpcd_set_lt_pattern_and_lane_settings( link, lt_settings, - hw_tr_pattern); + tr_pattern, + offset); else dpcd_set_lane_settings( link, - lt_settings); + lt_settings, + offset); /* 3. wait receiver to lock-on*/ + wait_time_microsec = lt_settings->cr_pattern_time; + + if (!link->is_lttpr_mode_transparent) + wait_time_microsec = TRAINING_AUX_RD_INTERVAL; + wait_for_training_aux_rd_interval( link, - 100); + wait_time_microsec); /* 4. Read lane status and requested drive * settings as set by the sink @@ -882,7 +1015,8 @@ static enum link_training_result perform_clock_recovery_sequence( lt_settings, dpcd_lane_status, &dpcd_lane_status_updated, - &req_settings); + &req_settings, + offset); /* 5. check CR done*/ if (is_cr_done(lane_count, dpcd_lane_status)) @@ -939,7 +1073,7 @@ static inline enum link_training_result perform_link_training_int( * TPS4 must be used instead of POST_LT_ADJ_REQ. */ if (link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED != 1 || - get_supported_tp(link) == HW_DP_TRAINING_PATTERN_4) + get_supported_tp(link) == DP_TRAINING_PATTERN_SEQUENCE_4) return status; if (status == LINK_TRAINING_SUCCESS && @@ -947,7 +1081,7 @@ static inline enum link_training_result perform_link_training_int( status = LINK_TRAINING_LQA_FAIL; lane_count_set.bits.LANE_COUNT_SET = lt_settings->link_settings.lane_count; - lane_count_set.bits.ENHANCED_FRAMING = 1; + lane_count_set.bits.ENHANCED_FRAMING = lt_settings->enhanced_framing; lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = 0; core_link_write_dpcd( @@ -959,24 +1093,29 @@ static inline enum link_training_result perform_link_training_int( return status; } -enum link_training_result dc_link_dp_perform_link_training( - struct dc_link *link, +static void initialize_training_settings( + struct dc_link *link, const struct dc_link_settings *link_setting, - bool skip_video_pattern) + const struct dc_link_training_overrides *overrides, + struct link_training_settings *lt_settings) { - enum link_training_result status = LINK_TRAINING_SUCCESS; + uint32_t lane; - char *link_rate = "Unknown"; - char *lt_result = "Unknown"; + memset(lt_settings, '\0', sizeof(struct link_training_settings)); - struct link_training_settings lt_settings; + /* Initialize link settings */ + lt_settings->link_settings.use_link_rate_set = link_setting->use_link_rate_set; + lt_settings->link_settings.link_rate_set = link_setting->link_rate_set; - memset(<_settings, '\0', sizeof(lt_settings)); + if (link->preferred_link_setting.link_rate != LINK_RATE_UNKNOWN) + lt_settings->link_settings.link_rate = link->preferred_link_setting.link_rate; + else + lt_settings->link_settings.link_rate = link_setting->link_rate; - lt_settings.link_settings.link_rate = link_setting->link_rate; - lt_settings.link_settings.lane_count = link_setting->lane_count; - lt_settings.link_settings.use_link_rate_set = link_setting->use_link_rate_set; - lt_settings.link_settings.link_rate_set = link_setting->link_rate_set; + if (link->preferred_link_setting.lane_count != LANE_COUNT_UNKNOWN) + lt_settings->link_settings.lane_count = link->preferred_link_setting.lane_count; + else + lt_settings->link_settings.lane_count = link_setting->lane_count; /*@todo[vdevulap] move SS to LS, should not be handled by displaypath*/ @@ -987,31 +1126,171 @@ enum link_training_result dc_link_dp_perform_link_training( * LINK_SPREAD_05_DOWNSPREAD_30KHZ : * LINK_SPREAD_DISABLED; */ + /* Initialize link spread */ if (link->dp_ss_off) - lt_settings.link_settings.link_spread = LINK_SPREAD_DISABLED; + lt_settings->link_settings.link_spread = LINK_SPREAD_DISABLED; + else if (overrides->downspread != NULL) + lt_settings->link_settings.link_spread + = *overrides->downspread + ? LINK_SPREAD_05_DOWNSPREAD_30KHZ + : LINK_SPREAD_DISABLED; else - lt_settings.link_settings.link_spread = LINK_SPREAD_05_DOWNSPREAD_30KHZ; + lt_settings->link_settings.link_spread = LINK_SPREAD_05_DOWNSPREAD_30KHZ; - /* 1. set link rate, lane count and spread*/ - dpcd_set_link_settings(link, <_settings); + /* Initialize lane settings overrides */ + if (overrides->voltage_swing != NULL) + lt_settings->voltage_swing = overrides->voltage_swing; - /* 2. perform link training (set link training done - * to false is done as well)*/ - status = perform_clock_recovery_sequence(link, <_settings); - if (status == LINK_TRAINING_SUCCESS) { - status = perform_channel_equalization_sequence(link, - <_settings); + if (overrides->pre_emphasis != NULL) + lt_settings->pre_emphasis = overrides->pre_emphasis; + + if (overrides->post_cursor2 != NULL) + lt_settings->post_cursor2 = overrides->post_cursor2; + + /* Initialize lane settings (VS/PE/PC2) */ + for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { + lt_settings->lane_settings[lane].VOLTAGE_SWING = + lt_settings->voltage_swing != NULL ? + *lt_settings->voltage_swing : + VOLTAGE_SWING_LEVEL0; + lt_settings->lane_settings[lane].PRE_EMPHASIS = + lt_settings->pre_emphasis != NULL ? + *lt_settings->pre_emphasis + : PRE_EMPHASIS_DISABLED; + lt_settings->lane_settings[lane].POST_CURSOR2 = + lt_settings->post_cursor2 != NULL ? + *lt_settings->post_cursor2 + : POST_CURSOR2_DISABLED; } - if ((status == LINK_TRAINING_SUCCESS) || !skip_video_pattern) { - status = perform_link_training_int(link, - <_settings, - status); + /* Initialize training timings */ + if (overrides->cr_pattern_time != NULL) + lt_settings->cr_pattern_time = *overrides->cr_pattern_time; + else + lt_settings->cr_pattern_time = get_training_aux_rd_interval(link, 100); + + if (overrides->eq_pattern_time != NULL) + lt_settings->eq_pattern_time = *overrides->eq_pattern_time; + else + lt_settings->eq_pattern_time = get_training_aux_rd_interval(link, 400); + + if (overrides->pattern_for_eq != NULL) + lt_settings->pattern_for_eq = *overrides->pattern_for_eq; + else + lt_settings->pattern_for_eq = get_supported_tp(link); + + if (overrides->enhanced_framing != NULL) + lt_settings->enhanced_framing = *overrides->enhanced_framing; + else + lt_settings->enhanced_framing = 1; +} + +static uint8_t convert_to_count(uint8_t lttpr_repeater_count) +{ + switch (lttpr_repeater_count) { + case 0x80: // 1 lttpr repeater + return 1; + case 0x40: // 2 lttpr repeaters + return 2; + case 0x20: // 3 lttpr repeaters + return 3; + case 0x10: // 4 lttpr repeaters + return 4; + case 0x08: // 5 lttpr repeaters + return 5; + case 0x04: // 6 lttpr repeaters + return 6; + case 0x02: // 7 lttpr repeaters + return 7; + case 0x01: // 8 lttpr repeaters + return 8; + default: + break; } + return 0; // invalid value +} - /* 6. print status message*/ - switch (lt_settings.link_settings.link_rate) { +static void configure_lttpr_mode(struct dc_link *link) +{ + /* aux timeout is already set to extended */ + /* RESET/SET lttpr mode to enable non transparent mode */ + uint8_t repeater_cnt; + uint32_t aux_interval_address; + uint8_t repeater_id; + enum dc_status result = DC_ERROR_UNEXPECTED; + uint8_t repeater_mode = DP_PHY_REPEATER_MODE_TRANSPARENT; + + DC_LOG_HW_LINK_TRAINING("%s\n Set LTTPR to Transparent Mode\n", __func__); + result = core_link_write_dpcd(link, + DP_PHY_REPEATER_MODE, + (uint8_t *)&repeater_mode, + sizeof(repeater_mode)); + + if (result == DC_OK) { + link->dpcd_caps.lttpr_caps.mode = repeater_mode; + } + + if (!link->is_lttpr_mode_transparent) { + + DC_LOG_HW_LINK_TRAINING("%s\n Set LTTPR to Non Transparent Mode\n", __func__); + + repeater_mode = DP_PHY_REPEATER_MODE_NON_TRANSPARENT; + result = core_link_write_dpcd(link, + DP_PHY_REPEATER_MODE, + (uint8_t *)&repeater_mode, + sizeof(repeater_mode)); + + if (result == DC_OK) { + link->dpcd_caps.lttpr_caps.mode = repeater_mode; + } + + repeater_cnt = convert_to_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + for (repeater_id = repeater_cnt; repeater_id > 0; repeater_id--) { + aux_interval_address = DP_TRAINING_AUX_RD_INTERVAL_PHY_REPEATER1 + + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (repeater_id - 1)); + core_link_read_dpcd( + link, + aux_interval_address, + (uint8_t *)&link->dpcd_caps.lttpr_caps.aux_rd_interval[repeater_id - 1], + sizeof(link->dpcd_caps.lttpr_caps.aux_rd_interval[repeater_id - 1])); + link->dpcd_caps.lttpr_caps.aux_rd_interval[repeater_id - 1] &= 0x7F; + } + } +} + +static void repeater_training_done(struct dc_link *link, uint32_t offset) +{ + union dpcd_training_pattern dpcd_pattern = { {0} }; + + const uint32_t dpcd_base_lt_offset = + DP_TRAINING_PATTERN_SET_PHY_REPEATER1 + + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + /* Set training not in progress*/ + dpcd_pattern.v1_4.TRAINING_PATTERN_SET = DPCD_TRAINING_PATTERN_VIDEOIDLE; + + core_link_write_dpcd( + link, + dpcd_base_lt_offset, + &dpcd_pattern.raw, + 1); + + DC_LOG_HW_LINK_TRAINING("%s\n LTTPR Id: %d 0x%X pattern = %x\n", + __func__, + offset, + dpcd_base_lt_offset, + dpcd_pattern.v1_4.TRAINING_PATTERN_SET); +} + +static void print_status_message( + struct dc_link *link, + const struct link_training_settings *lt_settings, + enum link_training_result status) +{ + char *link_rate = "Unknown"; + char *lt_result = "Unknown"; + char *lt_spread = "Disabled"; + switch (lt_settings->link_settings.link_rate) { case LINK_RATE_LOW: link_rate = "RBR"; break; @@ -1057,13 +1336,158 @@ enum link_training_result dc_link_dp_perform_link_training( break; } + switch (lt_settings->link_settings.link_spread) { + case LINK_SPREAD_DISABLED: + lt_spread = "Disabled"; + break; + case LINK_SPREAD_05_DOWNSPREAD_30KHZ: + lt_spread = "0.5% 30KHz"; + break; + case LINK_SPREAD_05_DOWNSPREAD_33KHZ: + lt_spread = "0.5% 33KHz"; + break; + default: + break; + } + /* Connectivity log: link training */ - CONN_MSG_LT(link, "%sx%d %s VS=%d, PE=%d", - link_rate, - lt_settings.link_settings.lane_count, - lt_result, - lt_settings.lane_settings[0].VOLTAGE_SWING, - lt_settings.lane_settings[0].PRE_EMPHASIS); + CONN_MSG_LT(link, "%sx%d %s VS=%d, PE=%d, DS=%s", + link_rate, + lt_settings->link_settings.lane_count, + lt_result, + lt_settings->lane_settings[0].VOLTAGE_SWING, + lt_settings->lane_settings[0].PRE_EMPHASIS, + lt_spread); +} + +void dc_link_dp_set_drive_settings( + struct dc_link *link, + struct link_training_settings *lt_settings) +{ + /* program ASIC PHY settings*/ + dp_set_hw_lane_settings(link, lt_settings, DPRX); + + /* Notify DP sink the PHY settings from source */ + dpcd_set_lane_settings(link, lt_settings, DPRX); +} + +bool dc_link_dp_perform_link_training_skip_aux( + struct dc_link *link, + const struct dc_link_settings *link_setting) +{ + struct link_training_settings lt_settings; + enum dc_dp_training_pattern pattern_for_cr = DP_TRAINING_PATTERN_SEQUENCE_1; + + initialize_training_settings( + link, + link_setting, + &link->preferred_training_settings, + <_settings); + + /* 1. Perform_clock_recovery_sequence. */ + + /* transmit training pattern for clock recovery */ + dp_set_hw_training_pattern(link, pattern_for_cr, DPRX); + + /* call HWSS to set lane settings*/ + dp_set_hw_lane_settings(link, <_settings, DPRX); + + /* wait receiver to lock-on*/ + wait_for_training_aux_rd_interval(link, lt_settings.cr_pattern_time); + + /* 2. Perform_channel_equalization_sequence. */ + + /* transmit training pattern for channel equalization. */ + dp_set_hw_training_pattern(link, lt_settings.pattern_for_eq, DPRX); + + /* call HWSS to set lane settings*/ + dp_set_hw_lane_settings(link, <_settings, DPRX); + + /* wait receiver to lock-on. */ + wait_for_training_aux_rd_interval(link, lt_settings.eq_pattern_time); + + /* 3. Perform_link_training_int. */ + + /* Mainlink output idle pattern. */ + dp_set_hw_test_pattern(link, DP_TEST_PATTERN_VIDEO_MODE, NULL, 0); + + print_status_message(link, <_settings, LINK_TRAINING_SUCCESS); + + return true; +} + +enum link_training_result dc_link_dp_perform_link_training( + struct dc_link *link, + const struct dc_link_settings *link_setting, + bool skip_video_pattern) +{ + enum link_training_result status = LINK_TRAINING_SUCCESS; + struct link_training_settings lt_settings; + + bool fec_enable; + uint8_t repeater_cnt; + uint8_t repeater_id; + + initialize_training_settings( + link, + link_setting, + &link->preferred_training_settings, + <_settings); + + /* 1. set link rate, lane count and spread. */ + dpcd_set_link_settings(link, <_settings); + + if (link->preferred_training_settings.fec_enable != NULL) + fec_enable = *link->preferred_training_settings.fec_enable; + else + fec_enable = true; + + dp_set_fec_ready(link, fec_enable); + + if (!link->is_lttpr_mode_transparent) { + /* Configure lttpr mode */ + configure_lttpr_mode(link); + + /* 2. perform link training (set link training done + * to false is done as well) + */ + repeater_cnt = convert_to_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + + for (repeater_id = repeater_cnt; (repeater_id > 0 && status == LINK_TRAINING_SUCCESS); + repeater_id--) { + status = perform_clock_recovery_sequence(link, <_settings, repeater_id); + + if (status != LINK_TRAINING_SUCCESS) + break; + + status = perform_channel_equalization_sequence(link, + <_settings, + repeater_id); + + if (status != LINK_TRAINING_SUCCESS) + break; + + repeater_training_done(link, repeater_id); + } + } + + if (status == LINK_TRAINING_SUCCESS) { + status = perform_clock_recovery_sequence(link, <_settings, DPRX); + if (status == LINK_TRAINING_SUCCESS) { + status = perform_channel_equalization_sequence(link, + <_settings, + DPRX); + } + } + + if ((status == LINK_TRAINING_SUCCESS) || !skip_video_pattern) { + status = perform_link_training_int(link, + <_settings, + status); + } + + /* 6. print status message*/ + print_status_message(link, <_settings, status); if (status != LINK_TRAINING_SUCCESS) link->ctx->dc->debug_data.ltFailCount++; @@ -1071,31 +1495,200 @@ enum link_training_result dc_link_dp_perform_link_training( return status; } - bool perform_link_training_with_retries( - struct dc_link *link, const struct dc_link_settings *link_setting, bool skip_video_pattern, - int attempts) + int attempts, + struct pipe_ctx *pipe_ctx, + enum signal_type signal) { uint8_t j; uint8_t delay_between_attempts = LINK_TRAINING_RETRY_DELAY; + struct dc_stream_state *stream = pipe_ctx->stream; + struct dc_link *link = stream->link; + enum dp_panel_mode panel_mode = dp_get_panel_mode(link); for (j = 0; j < attempts; ++j) { - if (dc_link_dp_perform_link_training( + dp_enable_link_phy( + link, + signal, + pipe_ctx->clock_source->id, + link_setting); + + if (stream->sink_patches.dppowerup_delay > 0) { + int delay_dp_power_up_in_ms = stream->sink_patches.dppowerup_delay; + + msleep(delay_dp_power_up_in_ms); + } + + dp_set_panel_mode(link, panel_mode); + + /* We need to do this before the link training to ensure the idle pattern in SST + * mode will be sent right after the link training + */ + link->link_enc->funcs->connect_dig_be_to_fe(link->link_enc, + pipe_ctx->stream_res.stream_enc->id, true); + + if (link->aux_access_disabled) { + dc_link_dp_perform_link_training_skip_aux(link, link_setting); + return true; + } else if (dc_link_dp_perform_link_training( link, link_setting, skip_video_pattern) == LINK_TRAINING_SUCCESS) return true; + /* latest link training still fail, skip delay and keep PHY on + */ + if (j == (attempts - 1)) + break; + + dp_disable_link_phy(link, signal); + msleep(delay_between_attempts); + delay_between_attempts += LINK_TRAINING_RETRY_DELAY; } return false; } +static enum clock_source_id get_clock_source_id(struct dc_link *link) +{ + enum clock_source_id dp_cs_id = CLOCK_SOURCE_ID_UNDEFINED; + struct clock_source *dp_cs = link->dc->res_pool->dp_clock_source; + + if (dp_cs != NULL) { + dp_cs_id = dp_cs->id; + } else { + /* + * dp clock source is not initialized for some reason. + * Should not happen, CLOCK_SOURCE_ID_EXTERNAL will be used + */ + ASSERT(dp_cs); + } + + return dp_cs_id; +} + +static void set_dp_mst_mode(struct dc_link *link, bool mst_enable) +{ + if (mst_enable == false && + link->type == dc_connection_mst_branch) { + /* Disable MST on link. Use only local sink. */ + dp_disable_link_phy_mst(link, link->connector_signal); + + link->type = dc_connection_single; + link->local_sink = link->remote_sinks[0]; + link->local_sink->sink_signal = SIGNAL_TYPE_DISPLAY_PORT; + } else if (mst_enable == true && + link->type == dc_connection_single && + link->remote_sinks[0] != NULL) { + /* Re-enable MST on link. */ + dp_disable_link_phy(link, link->connector_signal); + dp_enable_mst_on_sink(link, true); + + link->type = dc_connection_mst_branch; + link->local_sink->sink_signal = SIGNAL_TYPE_DISPLAY_PORT_MST; + } +} + +bool dc_link_dp_sync_lt_begin(struct dc_link *link) +{ + /* Begin Sync LT. During this time, + * DPCD:600h must not be powered down. + */ + link->sync_lt_in_progress = true; + + /*Clear any existing preferred settings.*/ + memset(&link->preferred_training_settings, 0, + sizeof(struct dc_link_training_overrides)); + memset(&link->preferred_link_setting, 0, + sizeof(struct dc_link_settings)); + + return true; +} + +enum link_training_result dc_link_dp_sync_lt_attempt( + struct dc_link *link, + struct dc_link_settings *link_settings, + struct dc_link_training_overrides *lt_overrides) +{ + struct link_training_settings lt_settings; + enum link_training_result lt_status = LINK_TRAINING_SUCCESS; + enum dp_panel_mode panel_mode = DP_PANEL_MODE_DEFAULT; + enum clock_source_id dp_cs_id = CLOCK_SOURCE_ID_EXTERNAL; + bool fec_enable = false; + + initialize_training_settings( + link, + link_settings, + lt_overrides, + <_settings); + + /* Setup MST Mode */ + if (lt_overrides->mst_enable) + set_dp_mst_mode(link, *lt_overrides->mst_enable); + + /* Disable link */ + dp_disable_link_phy(link, link->connector_signal); + + /* Enable link */ + dp_cs_id = get_clock_source_id(link); + dp_enable_link_phy(link, link->connector_signal, + dp_cs_id, link_settings); + + /* Set FEC enable */ + fec_enable = lt_overrides->fec_enable && *lt_overrides->fec_enable; + dp_set_fec_ready(link, fec_enable); + + if (lt_overrides->alternate_scrambler_reset) { + if (*lt_overrides->alternate_scrambler_reset) + panel_mode = DP_PANEL_MODE_EDP; + else + panel_mode = DP_PANEL_MODE_DEFAULT; + } else + panel_mode = dp_get_panel_mode(link); + + dp_set_panel_mode(link, panel_mode); + + /* Attempt to train with given link training settings */ + + /* Set link rate, lane count and spread. */ + dpcd_set_link_settings(link, <_settings); + + /* 2. perform link training (set link training done + * to false is done as well) + */ + lt_status = perform_clock_recovery_sequence(link, <_settings, DPRX); + if (lt_status == LINK_TRAINING_SUCCESS) { + lt_status = perform_channel_equalization_sequence(link, + <_settings, + DPRX); + } + + /* 3. Sync LT must skip TRAINING_PATTERN_SET:0 (video pattern)*/ + /* 4. print status message*/ + print_status_message(link, <_settings, lt_status); + + return lt_status; +} + +bool dc_link_dp_sync_lt_end(struct dc_link *link, bool link_down) +{ + /* If input parameter is set, shut down phy. + * Still shouldn't turn off dp_receiver (DPCD:600h) + */ + if (link_down == true) { + dp_disable_link_phy(link, link->connector_signal); + dp_set_fec_ready(link, false); + } + + link->sync_lt_in_progress = false; + return true; +} + static struct dc_link_settings get_max_link_cap(struct dc_link *link) { /* Set Default link settings */ @@ -1109,6 +1702,9 @@ static struct dc_link_settings get_max_link_cap(struct dc_link *link) if (link->link_enc->features.flags.bits.IS_HBR3_CAPABLE) max_link_cap.link_rate = LINK_RATE_HIGH3; + if (link->link_enc->funcs->get_max_link_cap) + link->link_enc->funcs->get_max_link_cap(link->link_enc, &max_link_cap); + /* Lower link settings based on sink's link cap */ if (link->reported_link_cap.lane_count < max_link_cap.lane_count) max_link_cap.lane_count = @@ -1120,6 +1716,22 @@ static struct dc_link_settings get_max_link_cap(struct dc_link *link) max_link_cap.link_spread) max_link_cap.link_spread = link->reported_link_cap.link_spread; + /* + * account for lttpr repeaters cap + * notes: repeaters do not snoop in the DPRX Capabilities addresses (3.6.3). + */ + if (!link->is_lttpr_mode_transparent) { + if (link->dpcd_caps.lttpr_caps.max_lane_count < max_link_cap.lane_count) + max_link_cap.lane_count = link->dpcd_caps.lttpr_caps.max_lane_count; + + if (link->dpcd_caps.lttpr_caps.max_link_rate < max_link_cap.link_rate) + max_link_cap.link_rate = link->dpcd_caps.lttpr_caps.max_link_rate; + + DC_LOG_HW_LINK_TRAINING("%s\n Training with LTTPR, max_lane count %d max_link rate %d \n", + __func__, + max_link_cap.lane_count, + max_link_cap.link_rate); + } return max_link_cap; } @@ -1250,7 +1862,6 @@ bool dp_verify_link_cap( bool success; bool skip_link_training; bool skip_video_pattern; - struct clock_source *dp_cs; enum clock_source_id dp_cs_id = CLOCK_SOURCE_ID_EXTERNAL; enum link_training_result status; union hpd_irq_data irq_data; @@ -1266,6 +1877,13 @@ bool dp_verify_link_cap( max_link_cap = get_max_link_cap(link); + /* Grant extended timeout request */ + if (!link->is_lttpr_mode_transparent && link->dpcd_caps.lttpr_caps.max_ext_timeout > 0) { + uint8_t grant = link->dpcd_caps.lttpr_caps.max_ext_timeout & 0x80; + + core_link_write_dpcd(link, DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT, &grant, sizeof(grant)); + } + /* TODO implement override and monitor patch later */ /* try to train the link from high to low to @@ -1274,18 +1892,18 @@ bool dp_verify_link_cap( /* disable PHY done possible by BIOS, will be done by driver itself */ dp_disable_link_phy(link, link->connector_signal); - dp_cs = link->dc->res_pool->dp_clock_source; - - if (dp_cs) - dp_cs_id = dp_cs->id; - else { - /* - * dp clock source is not initialized for some reason. - * Should not happen, CLOCK_SOURCE_ID_EXTERNAL will be used - */ - ASSERT(dp_cs); + /* Temporary Renoir-specific workaround for SWDEV-215184; + * PHY will sometimes be in bad state on hotplugging display from certain USB-C dongle, + * so add extra cycle of enabling and disabling the PHY before first link training. + */ + if (link->link_enc->features.flags.bits.DP_IS_USB_C && + link->dc->debug.usbc_combo_phy_reset_wa) { + dp_enable_link_phy(link, link->connector_signal, dp_cs_id, cur); + dp_disable_link_phy(link, link->connector_signal); } + dp_cs_id = get_clock_source_id(link); + /* link training starts with the maximum common settings * supported by both sink and ASIC. */ @@ -1354,6 +1972,49 @@ bool dp_verify_link_cap( return success; } +bool dp_verify_link_cap_with_retries( + struct dc_link *link, + struct dc_link_settings *known_limit_link_setting, + int attempts) +{ + uint8_t i = 0; + bool success = false; + + for (i = 0; i < attempts; i++) { + int fail_count = 0; + enum dc_connection_type type = dc_connection_none; + + memset(&link->verified_link_cap, 0, + sizeof(struct dc_link_settings)); + if (!dc_link_detect_sink(link, &type) || type == dc_connection_none) { + link->verified_link_cap.lane_count = LANE_COUNT_ONE; + link->verified_link_cap.link_rate = LINK_RATE_LOW; + link->verified_link_cap.link_spread = LINK_SPREAD_DISABLED; + break; + } else if (dp_verify_link_cap(link, + &link->reported_link_cap, + &fail_count) && fail_count == 0) { + success = true; + break; + } + msleep(10); + } + return success; +} + +bool dp_verify_mst_link_cap( + struct dc_link *link) +{ + struct dc_link_settings max_link_cap = {0}; + + max_link_cap = get_max_link_cap(link); + link->verified_link_cap = get_common_supported_link_settings( + link->reported_link_cap, + max_link_cap); + + return true; +} + static struct dc_link_settings get_common_supported_link_settings( struct dc_link_settings link_setting_a, struct dc_link_settings link_setting_b) @@ -1741,11 +2402,11 @@ static bool allow_hpd_rx_irq(const struct dc_link *link) return false; } -static bool handle_hpd_irq_psr_sink(const struct dc_link *link) +static bool handle_hpd_irq_psr_sink(struct dc_link *link) { union dpcd_psr_configuration psr_configuration; - if (!link->psr_enabled) + if (!link->psr_feature_enabled) return false; dm_helpers_dp_read_dpcd( @@ -1784,8 +2445,8 @@ static bool handle_hpd_irq_psr_sink(const struct dc_link *link) sizeof(psr_error_status.raw)); /* PSR error, disable and re-enable PSR */ - dc_link_set_psr_enable(link, false, true); - dc_link_set_psr_enable(link, true, true); + dc_link_set_psr_allow_active(link, false, true); + dc_link_set_psr_allow_active(link, true, true); return true; } else if (psr_sink_psr_status.bits.SINK_SELF_REFRESH_STATUS == @@ -1945,6 +2606,7 @@ static void dp_test_send_phy_test_pattern(struct dc_link *link) dc_link_dp_set_test_pattern( link, test_pattern, + DP_TEST_PATTERN_COLOR_SPACE_UNDEFINED, &link_training_settings, test_80_bit_pattern, (DP_TEST_80BIT_CUSTOM_PATTERN_79_72 - @@ -1956,6 +2618,8 @@ static void dp_test_send_link_test_pattern(struct dc_link *link) union link_test_pattern dpcd_test_pattern; union test_misc dpcd_test_params; enum dp_test_pattern test_pattern; + enum dp_test_pattern_color_space test_pattern_color_space = + DP_TEST_PATTERN_COLOR_SPACE_UNDEFINED; memset(&dpcd_test_pattern, 0, sizeof(dpcd_test_pattern)); memset(&dpcd_test_params, 0, sizeof(dpcd_test_params)); @@ -1990,14 +2654,105 @@ static void dp_test_send_link_test_pattern(struct dc_link *link) break; } + test_pattern_color_space = dpcd_test_params.bits.YCBCR_COEFS ? + DP_TEST_PATTERN_COLOR_SPACE_YCBCR709 : + DP_TEST_PATTERN_COLOR_SPACE_YCBCR601; + dc_link_dp_set_test_pattern( link, test_pattern, + test_pattern_color_space, NULL, NULL, 0); } +static void dp_test_get_audio_test_data(struct dc_link *link, bool disable_video) +{ + union audio_test_mode dpcd_test_mode = {0}; + struct audio_test_pattern_type dpcd_pattern_type = {0}; + union audio_test_pattern_period dpcd_pattern_period[AUDIO_CHANNELS_COUNT] = {0}; + enum dp_test_pattern test_pattern = DP_TEST_PATTERN_AUDIO_OPERATOR_DEFINED; + + struct pipe_ctx *pipes = link->dc->current_state->res_ctx.pipe_ctx; + struct pipe_ctx *pipe_ctx = &pipes[0]; + unsigned int channel_count; + unsigned int channel = 0; + unsigned int modes = 0; + unsigned int sampling_rate_in_hz = 0; + + // get audio test mode and test pattern parameters + core_link_read_dpcd( + link, + DP_TEST_AUDIO_MODE, + &dpcd_test_mode.raw, + sizeof(dpcd_test_mode)); + + core_link_read_dpcd( + link, + DP_TEST_AUDIO_PATTERN_TYPE, + &dpcd_pattern_type.value, + sizeof(dpcd_pattern_type)); + + channel_count = dpcd_test_mode.bits.channel_count + 1; + + // read pattern periods for requested channels when sawTooth pattern is requested + if (dpcd_pattern_type.value == AUDIO_TEST_PATTERN_SAWTOOTH || + dpcd_pattern_type.value == AUDIO_TEST_PATTERN_OPERATOR_DEFINED) { + + test_pattern = (dpcd_pattern_type.value == AUDIO_TEST_PATTERN_SAWTOOTH) ? + DP_TEST_PATTERN_AUDIO_SAWTOOTH : DP_TEST_PATTERN_AUDIO_OPERATOR_DEFINED; + // read period for each channel + for (channel = 0; channel < channel_count; channel++) { + core_link_read_dpcd( + link, + DP_TEST_AUDIO_PERIOD_CH1 + channel, + &dpcd_pattern_period[channel].raw, + sizeof(dpcd_pattern_period[channel])); + } + } + + // translate sampling rate + switch (dpcd_test_mode.bits.sampling_rate) { + case AUDIO_SAMPLING_RATE_32KHZ: + sampling_rate_in_hz = 32000; + break; + case AUDIO_SAMPLING_RATE_44_1KHZ: + sampling_rate_in_hz = 44100; + break; + case AUDIO_SAMPLING_RATE_48KHZ: + sampling_rate_in_hz = 48000; + break; + case AUDIO_SAMPLING_RATE_88_2KHZ: + sampling_rate_in_hz = 88200; + break; + case AUDIO_SAMPLING_RATE_96KHZ: + sampling_rate_in_hz = 96000; + break; + case AUDIO_SAMPLING_RATE_176_4KHZ: + sampling_rate_in_hz = 176400; + break; + case AUDIO_SAMPLING_RATE_192KHZ: + sampling_rate_in_hz = 192000; + break; + default: + sampling_rate_in_hz = 0; + break; + } + + link->audio_test_data.flags.test_requested = 1; + link->audio_test_data.flags.disable_video = disable_video; + link->audio_test_data.sampling_rate = sampling_rate_in_hz; + link->audio_test_data.channel_count = channel_count; + link->audio_test_data.pattern_type = test_pattern; + + if (test_pattern == DP_TEST_PATTERN_AUDIO_SAWTOOTH) { + for (modes = 0; modes < pipe_ctx->stream->audio_info.mode_count; modes++) { + link->audio_test_data.pattern_period[modes] = dpcd_pattern_period[modes].bits.pattern_period; + } + } +} + static void handle_automated_test(struct dc_link *link) { union test_request test_request; @@ -2027,6 +2782,12 @@ static void handle_automated_test(struct dc_link *link) dp_test_send_link_test_pattern(link); test_response.bits.ACK = 1; } + + if (test_request.bits.AUDIO_TEST_PATTERN) { + dp_test_get_audio_test_data(link, test_request.bits.TEST_AUDIO_DISABLED_VIDEO); + test_response.bits.ACK = 1; + } + if (test_request.bits.PHY_TEST_PATTERN) { dp_test_send_phy_test_pattern(link); test_response.bits.ACK = 1; @@ -2046,8 +2807,10 @@ bool dc_link_handle_hpd_rx_irq(struct dc_link *link, union hpd_irq_data *out_hpd union hpd_irq_data hpd_irq_dpcd_data = { { { {0} } } }; union device_service_irq device_service_clear = { { 0 } }; enum dc_status result; - bool status = false; + struct pipe_ctx *pipe_ctx; + struct dc_link_settings previous_link_settings; + int i; if (out_link_loss) *out_link_loss = false; @@ -2110,19 +2873,36 @@ bool dc_link_handle_hpd_rx_irq(struct dc_link *link, union hpd_irq_data *out_hpd /* For now we only handle 'Downstream port status' case. * If we got sink count changed it means * Downstream port status changed, - * then DM should call DC to do the detection. */ - if (hpd_rx_irq_check_link_loss_status( - link, - &hpd_irq_dpcd_data)) { + * then DM should call DC to do the detection. + * NOTE: Do not handle link loss on eDP since it is internal link*/ + if ((link->connector_signal != SIGNAL_TYPE_EDP) && + hpd_rx_irq_check_link_loss_status( + link, + &hpd_irq_dpcd_data)) { /* Connectivity log: link loss */ CONN_DATA_LINK_LOSS(link, hpd_irq_dpcd_data.raw, sizeof(hpd_irq_dpcd_data), "Status: "); - perform_link_training_with_retries(link, - &link->cur_link_settings, - true, LINK_TRAINING_ATTEMPTS); + for (i = 0; i < MAX_PIPES; i++) { + pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx && pipe_ctx->stream && pipe_ctx->stream->link == link) + break; + } + + if (pipe_ctx == NULL || pipe_ctx->stream == NULL) + return false; + + previous_link_settings = link->cur_link_settings; + + perform_link_training_with_retries(&previous_link_settings, + true, LINK_TRAINING_ATTEMPTS, + pipe_ctx, + pipe_ctx->stream->signal); + + if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) + dc_link_reallocate_mst_payload(link); status = false; if (out_link_loss) @@ -2156,6 +2936,11 @@ bool is_mst_supported(struct dc_link *link) union dpcd_rev rev; union mstm_cap cap; + if (link->preferred_training_settings.mst_enable && + *link->preferred_training_settings.mst_enable == false) { + return false; + } + rev.raw = 0; cap.raw = 0; @@ -2224,6 +3009,7 @@ static void get_active_converter_info( uint8_t data, struct dc_link *link) { union dp_downstream_port_present ds_port = { .byte = data }; + memset(&link->dpcd_caps.dongle_caps, 0, sizeof(link->dpcd_caps.dongle_caps)); /* decode converter info*/ if (!ds_port.fields.PORT_PRESENT) { @@ -2345,7 +3131,6 @@ static void dp_wa_power_up_0010FA(struct dc_link *link, uint8_t *dpcd_data, int length) { int retry = 0; - union dp_downstream_port_present ds_port = { 0 }; if (!link->dpcd_caps.dpcd_rev.raw) { do { @@ -2358,18 +3143,16 @@ static void dp_wa_power_up_0010FA(struct dc_link *link, uint8_t *dpcd_data, } while (retry++ < 4 && !link->dpcd_caps.dpcd_rev.raw); } - ds_port.byte = dpcd_data[DP_DOWNSTREAMPORT_PRESENT - - DP_DPCD_REV]; - if (link->dpcd_caps.dongle_type == DISPLAY_DONGLE_DP_VGA_CONVERTER) { switch (link->dpcd_caps.branch_dev_id) { - /* Some active dongles (DP-VGA, DP-DLDVI converters) power down + /* 0010FA active dongles (DP-VGA, DP-DLDVI converters) power down * all internal circuits including AUX communication preventing * reading DPCD table and EDID (spec violation). * Encoder will skip DP RX power down on disable_output to * keep receiver powered all the time.*/ - case DP_BRANCH_DEVICE_ID_1: - case DP_BRANCH_DEVICE_ID_4: + case DP_BRANCH_DEVICE_ID_0010FA: + case DP_BRANCH_DEVICE_ID_0080E1: + case DP_BRANCH_DEVICE_ID_00E04C: link->wa_flags.dp_keep_receiver_powered = true; break; @@ -2384,7 +3167,11 @@ static void dp_wa_power_up_0010FA(struct dc_link *link, uint8_t *dpcd_data, static bool retrieve_link_cap(struct dc_link *link) { - uint8_t dpcd_data[DP_ADAPTER_CAP - DP_DPCD_REV + 1]; + /* DP_ADAPTER_CAP - DP_DPCD_REV + 1 == 16 and also DP_DSC_BITS_PER_PIXEL_INC - DP_DSC_SUPPORT + 1 == 16, + * which means size 16 will be good for both of those DPCD register block reads + */ + uint8_t dpcd_data[16]; + uint8_t lttpr_dpcd_data[6]; /*Only need to read 1 byte starting from DP_DPRX_FEATURE_ENUMERATION_LIST. */ @@ -2400,7 +3187,19 @@ static bool retrieve_link_cap(struct dc_link *link) int i; struct dp_sink_hw_fw_revision dp_hw_fw_revision; + /* Set default timeout to 3.2ms and read LTTPR capabilities */ + bool ext_timeout_support = link->dc->caps.extended_aux_timeout_support && + !link->dc->config.disable_extended_timeout_support; + + link->is_lttpr_mode_transparent = true; + + if (ext_timeout_support) { + dc_link_aux_configure_timeout(link->ddc, + LINK_AUX_DEFAULT_EXTENDED_TIMEOUT_PERIOD); + } + memset(dpcd_data, '\0', sizeof(dpcd_data)); + memset(lttpr_dpcd_data, '\0', sizeof(lttpr_dpcd_data)); memset(&down_strm_port_count, '\0', sizeof(union down_stream_port_count)); memset(&edp_config_cap, '\0', @@ -2432,6 +3231,52 @@ static bool retrieve_link_cap(struct dc_link *link) return false; } + if (ext_timeout_support) { + + status = core_link_read_dpcd( + link, + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV, + lttpr_dpcd_data, + sizeof(lttpr_dpcd_data)); + + link->dpcd_caps.lttpr_caps.revision.raw = + lttpr_dpcd_data[DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.max_link_rate = + lttpr_dpcd_data[DP_MAX_LINK_RATE_PHY_REPEATER - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.phy_repeater_cnt = + lttpr_dpcd_data[DP_PHY_REPEATER_CNT - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.max_lane_count = + lttpr_dpcd_data[DP_MAX_LANE_COUNT_PHY_REPEATER - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.mode = + lttpr_dpcd_data[DP_PHY_REPEATER_MODE - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + link->dpcd_caps.lttpr_caps.max_ext_timeout = + lttpr_dpcd_data[DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT - + DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + + if (link->dpcd_caps.lttpr_caps.phy_repeater_cnt > 0 && + link->dpcd_caps.lttpr_caps.max_lane_count > 0 && + link->dpcd_caps.lttpr_caps.max_lane_count <= 4 && + link->dpcd_caps.lttpr_caps.revision.raw >= 0x14) { + link->is_lttpr_mode_transparent = false; + } else { + /*No lttpr reset timeout to its default value*/ + link->is_lttpr_mode_transparent = true; + dc_link_aux_configure_timeout(link->ddc, LINK_AUX_DEFAULT_TIMEOUT_PERIOD); + } + + CONN_DATA_DETECT(link, lttpr_dpcd_data, sizeof(lttpr_dpcd_data), "LTTPR Caps: "); + } + { union training_aux_rd_interval aux_rd_interval; @@ -2439,7 +3284,7 @@ static bool retrieve_link_cap(struct dc_link *link) dpcd_data[DP_TRAINING_AUX_RD_INTERVAL]; link->dpcd_caps.ext_receiver_cap_field_present = - aux_rd_interval.bits.EXT_RECEIVER_CAP_FIELD_PRESENT == 1 ? true:false; + aux_rd_interval.bits.EXT_RECEIVER_CAP_FIELD_PRESENT == 1; if (aux_rd_interval.bits.EXT_RECEIVER_CAP_FIELD_PRESENT == 1) { uint8_t ext_cap_data[16]; @@ -2570,7 +3415,6 @@ static bool retrieve_link_cap(struct dc_link *link) dp_hw_fw_revision.ieee_fw_rev, sizeof(dp_hw_fw_revision.ieee_fw_rev)); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT memset(&link->dpcd_caps.dsc_caps, '\0', sizeof(link->dpcd_caps.dsc_caps)); memset(&link->dpcd_caps.fec_cap, '\0', sizeof(link->dpcd_caps.fec_cap)); @@ -2592,7 +3436,6 @@ static bool retrieve_link_cap(struct dc_link *link) link->dpcd_caps.dsc_caps.dsc_ext_caps.raw, sizeof(link->dpcd_caps.dsc_caps.dsc_ext_caps.raw)); } -#endif /* Connectivity log: detection */ CONN_DATA_DETECT(link, dpcd_data, sizeof(dpcd_data), "Rx Caps: "); @@ -2600,6 +3443,68 @@ static bool retrieve_link_cap(struct dc_link *link) return true; } +bool dp_overwrite_extended_receiver_cap(struct dc_link *link) +{ + uint8_t dpcd_data[16]; + uint32_t read_dpcd_retry_cnt = 3; + enum dc_status status = DC_ERROR_UNEXPECTED; + union dp_downstream_port_present ds_port = { 0 }; + union down_stream_port_count down_strm_port_count; + union edp_configuration_cap edp_config_cap; + + int i; + + for (i = 0; i < read_dpcd_retry_cnt; i++) { + status = core_link_read_dpcd( + link, + DP_DPCD_REV, + dpcd_data, + sizeof(dpcd_data)); + if (status == DC_OK) + break; + } + + link->dpcd_caps.dpcd_rev.raw = + dpcd_data[DP_DPCD_REV - DP_DPCD_REV]; + + if (dpcd_data[DP_MAX_LANE_COUNT - DP_DPCD_REV] == 0) + return false; + + ds_port.byte = dpcd_data[DP_DOWNSTREAMPORT_PRESENT - + DP_DPCD_REV]; + + get_active_converter_info(ds_port.byte, link); + + down_strm_port_count.raw = dpcd_data[DP_DOWN_STREAM_PORT_COUNT - + DP_DPCD_REV]; + + link->dpcd_caps.allow_invalid_MSA_timing_param = + down_strm_port_count.bits.IGNORE_MSA_TIMING_PARAM; + + link->dpcd_caps.max_ln_count.raw = dpcd_data[ + DP_MAX_LANE_COUNT - DP_DPCD_REV]; + + link->dpcd_caps.max_down_spread.raw = dpcd_data[ + DP_MAX_DOWNSPREAD - DP_DPCD_REV]; + + link->reported_link_cap.lane_count = + link->dpcd_caps.max_ln_count.bits.MAX_LANE_COUNT; + link->reported_link_cap.link_rate = dpcd_data[ + DP_MAX_LINK_RATE - DP_DPCD_REV]; + link->reported_link_cap.link_spread = + link->dpcd_caps.max_down_spread.bits.MAX_DOWN_SPREAD ? + LINK_SPREAD_05_DOWNSPREAD_30KHZ : LINK_SPREAD_DISABLED; + + edp_config_cap.raw = dpcd_data[ + DP_EDP_CONFIGURATION_CAP - DP_DPCD_REV]; + link->dpcd_caps.panel_mode_edp = + edp_config_cap.bits.ALT_SCRAMBLER_RESET; + link->dpcd_caps.dpcd_display_control_capable = + edp_config_cap.bits.DPCD_DISPLAY_CONTROL_CAPABLE; + + return true; +} + bool detect_dp_sink_caps(struct dc_link *link) { return retrieve_link_cap(link); @@ -2714,21 +3619,20 @@ static bool is_dp_phy_pattern(enum dp_test_pattern test_pattern) static void set_crtc_test_pattern(struct dc_link *link, struct pipe_ctx *pipe_ctx, - enum dp_test_pattern test_pattern) + enum dp_test_pattern test_pattern, + enum dp_test_pattern_color_space test_pattern_color_space) { enum controller_dp_test_pattern controller_test_pattern; enum dc_color_depth color_depth = pipe_ctx-> stream->timing.display_color_depth; struct bit_depth_reduction_params params; struct output_pixel_processor *opp = pipe_ctx->stream_res.opp; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) int width = pipe_ctx->stream->timing.h_addressable + pipe_ctx->stream->timing.h_border_left + pipe_ctx->stream->timing.h_border_right; int height = pipe_ctx->stream->timing.v_addressable + pipe_ctx->stream->timing.v_border_bottom + pipe_ctx->stream->timing.v_border_top; -#endif memset(¶ms, 0, sizeof(params)); @@ -2772,17 +3676,42 @@ static void set_crtc_test_pattern(struct dc_link *link, if (pipe_ctx->stream_res.tg->funcs->set_test_pattern) pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg, controller_test_pattern, color_depth); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) else if (opp->funcs->opp_set_disp_pattern_generator) { - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + enum controller_dp_color_space controller_color_space; + int opp_cnt = 1; + int count; + + switch (test_pattern_color_space) { + case DP_TEST_PATTERN_COLOR_SPACE_RGB: + controller_color_space = CONTROLLER_DP_COLOR_SPACE_RGB; + break; + case DP_TEST_PATTERN_COLOR_SPACE_YCBCR601: + controller_color_space = CONTROLLER_DP_COLOR_SPACE_YCBCR601; + break; + case DP_TEST_PATTERN_COLOR_SPACE_YCBCR709: + controller_color_space = CONTROLLER_DP_COLOR_SPACE_YCBCR709; + break; + case DP_TEST_PATTERN_COLOR_SPACE_UNDEFINED: + default: + controller_color_space = CONTROLLER_DP_COLOR_SPACE_UDEFINED; + DC_LOG_ERROR("%s: Color space must be defined for test pattern", __func__); + ASSERT(0); + break; + } - if (bot_odm_pipe) { - struct output_pixel_processor *bot_opp = bot_odm_pipe->stream_res.opp; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; - bot_opp->funcs->opp_program_bit_depth_reduction(bot_opp, ¶ms); - width /= 2; - bot_opp->funcs->opp_set_disp_pattern_generator(bot_opp, + width /= opp_cnt; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; + + odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); + odm_opp->funcs->opp_set_disp_pattern_generator(odm_opp, controller_test_pattern, + controller_color_space, color_depth, NULL, width, @@ -2790,12 +3719,18 @@ static void set_crtc_test_pattern(struct dc_link *link, } opp->funcs->opp_set_disp_pattern_generator(opp, controller_test_pattern, + controller_color_space, color_depth, NULL, width, height); + /* wait for dpg to blank pixel data with test pattern */ + for (count = 0; count < 1000; count++) { + if (opp->funcs->dpg_is_blanked(opp)) + break; + udelay(100); + } } -#endif } break; case DP_TEST_PATTERN_VIDEO_MODE: @@ -2808,17 +3743,21 @@ static void set_crtc_test_pattern(struct dc_link *link, pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg, CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, color_depth); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) else if (opp->funcs->opp_set_disp_pattern_generator) { - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; - if (bot_odm_pipe) { - struct output_pixel_processor *bot_opp = bot_odm_pipe->stream_res.opp; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; - bot_opp->funcs->opp_program_bit_depth_reduction(bot_opp, ¶ms); - width /= 2; - bot_opp->funcs->opp_set_disp_pattern_generator(bot_opp, + width /= opp_cnt; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; + + odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); + odm_opp->funcs->opp_set_disp_pattern_generator(odm_opp, CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, + CONTROLLER_DP_COLOR_SPACE_UDEFINED, color_depth, NULL, width, @@ -2826,12 +3765,12 @@ static void set_crtc_test_pattern(struct dc_link *link, } opp->funcs->opp_set_disp_pattern_generator(opp, CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, + CONTROLLER_DP_COLOR_SPACE_UDEFINED, color_depth, NULL, width, height); } -#endif } break; @@ -2843,6 +3782,7 @@ static void set_crtc_test_pattern(struct dc_link *link, bool dc_link_dp_set_test_pattern( struct dc_link *link, enum dp_test_pattern test_pattern, + enum dp_test_pattern_color_space test_pattern_color_space, const struct link_training_settings *p_link_settings, const unsigned char *p_custom_pattern, unsigned int cust_pattern_size) @@ -2858,7 +3798,7 @@ bool dc_link_dp_set_test_pattern( memset(&training_pattern, 0, sizeof(training_pattern)); for (i = 0; i < MAX_PIPES; i++) { - if (pipes[i].stream->link == link) { + if (pipes[i].stream->link == link && !pipes[i].top_pipe && !pipes[i].prev_odm_pipe) { pipe_ctx = &pipes[i]; break; } @@ -2871,7 +3811,7 @@ bool dc_link_dp_set_test_pattern( if (link->test_pattern_enabled && test_pattern == DP_TEST_PATTERN_VIDEO_MODE) { /* Set CRTC Test Pattern */ - set_crtc_test_pattern(link, pipe_ctx, test_pattern); + set_crtc_test_pattern(link, pipe_ctx, test_pattern, test_pattern_color_space); dp_set_hw_test_pattern(link, test_pattern, (uint8_t *)p_custom_pattern, (uint32_t)cust_pattern_size); @@ -2894,8 +3834,8 @@ bool dc_link_dp_set_test_pattern( if (is_dp_phy_pattern(test_pattern)) { /* Set DPCD Lane Settings before running test pattern */ if (p_link_settings != NULL) { - dp_set_hw_lane_settings(link, p_link_settings); - dpcd_set_lane_settings(link, p_link_settings); + dp_set_hw_lane_settings(link, p_link_settings, DPRX); + dpcd_set_lane_settings(link, p_link_settings, DPRX); } /* Blank stream if running test pattern */ @@ -2985,8 +3925,38 @@ bool dc_link_dp_set_test_pattern( sizeof(training_pattern)); } } else { - /* CRTC Patterns */ - set_crtc_test_pattern(link, pipe_ctx, test_pattern); + enum dc_color_space color_space = COLOR_SPACE_UNKNOWN; + + switch (test_pattern_color_space) { + case DP_TEST_PATTERN_COLOR_SPACE_RGB: + color_space = COLOR_SPACE_SRGB; + if (test_pattern == DP_TEST_PATTERN_COLOR_SQUARES_CEA) + color_space = COLOR_SPACE_SRGB_LIMITED; + break; + + case DP_TEST_PATTERN_COLOR_SPACE_YCBCR601: + color_space = COLOR_SPACE_YCBCR601; + if (test_pattern == DP_TEST_PATTERN_COLOR_SQUARES_CEA) + color_space = COLOR_SPACE_YCBCR601_LIMITED; + break; + case DP_TEST_PATTERN_COLOR_SPACE_YCBCR709: + color_space = COLOR_SPACE_YCBCR709; + if (test_pattern == DP_TEST_PATTERN_COLOR_SQUARES_CEA) + color_space = COLOR_SPACE_YCBCR709_LIMITED; + break; + default: + break; + } + /* update MSA to requested color space */ + pipe_ctx->stream_res.stream_enc->funcs->dp_set_stream_attribute(pipe_ctx->stream_res.stream_enc, + &pipe_ctx->stream->timing, + color_space, + pipe_ctx->stream->use_vsc_sdp_for_colorimetry, + link->dpcd_caps.dprx_feature.bits.SST_SPLIT_SDP_CAP); + + /* CRTC Patterns */ + set_crtc_test_pattern(link, pipe_ctx, test_pattern, test_pattern_color_space); + /* Set Test Pattern state */ link->test_pattern_enabled = true; } @@ -3007,7 +3977,105 @@ void dp_enable_mst_on_sink(struct dc_link *link, bool enable) core_link_write_dpcd(link, DP_MSTM_CTRL, &mstmCntl, 1); } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT +void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode) +{ + union dpcd_edp_config edp_config_set; + bool panel_mode_edp = false; + + memset(&edp_config_set, '\0', sizeof(union dpcd_edp_config)); + + if (panel_mode != DP_PANEL_MODE_DEFAULT) { + + switch (panel_mode) { + case DP_PANEL_MODE_EDP: + case DP_PANEL_MODE_SPECIAL: + panel_mode_edp = true; + break; + + default: + break; + } + + /*set edp panel mode in receiver*/ + core_link_read_dpcd( + link, + DP_EDP_CONFIGURATION_SET, + &edp_config_set.raw, + sizeof(edp_config_set.raw)); + + if (edp_config_set.bits.PANEL_MODE_EDP + != panel_mode_edp) { + enum ddc_result result = DDC_RESULT_UNKNOWN; + + edp_config_set.bits.PANEL_MODE_EDP = + panel_mode_edp; + result = core_link_write_dpcd( + link, + DP_EDP_CONFIGURATION_SET, + &edp_config_set.raw, + sizeof(edp_config_set.raw)); + + ASSERT(result == DDC_RESULT_SUCESSFULL); + } + } + DC_LOG_DETECTION_DP_CAPS("Link: %d eDP panel mode supported: %d " + "eDP panel mode enabled: %d \n", + link->link_index, + link->dpcd_caps.panel_mode_edp, + panel_mode_edp); +} + +enum dp_panel_mode dp_get_panel_mode(struct dc_link *link) +{ + /* We need to explicitly check that connector + * is not DP. Some Travis_VGA get reported + * by video bios as DP. + */ + if (link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT) { + + switch (link->dpcd_caps.branch_dev_id) { + case DP_BRANCH_DEVICE_ID_0022B9: + /* alternate scrambler reset is required for Travis + * for the case when external chip does not + * provide sink device id, alternate scrambler + * scheme will be overriden later by querying + * Encoder features + */ + if (strncmp( + link->dpcd_caps.branch_dev_name, + DP_VGA_LVDS_CONVERTER_ID_2, + sizeof( + link->dpcd_caps. + branch_dev_name)) == 0) { + return DP_PANEL_MODE_SPECIAL; + } + break; + case DP_BRANCH_DEVICE_ID_00001A: + /* alternate scrambler reset is required for Travis + * for the case when external chip does not provide + * sink device id, alternate scrambler scheme will + * be overriden later by querying Encoder feature + */ + if (strncmp(link->dpcd_caps.branch_dev_name, + DP_VGA_LVDS_CONVERTER_ID_3, + sizeof( + link->dpcd_caps. + branch_dev_name)) == 0) { + return DP_PANEL_MODE_SPECIAL; + } + break; + default: + break; + } + } + + if (link->dpcd_caps.panel_mode_edp) { + return DP_PANEL_MODE_EDP; + } + + return DP_PANEL_MODE_DEFAULT; +} + void dp_set_fec_ready(struct dc_link *link, bool ready) { /* FEC has to be "set ready" before the link training. @@ -3024,7 +4092,7 @@ void dp_set_fec_ready(struct dc_link *link, bool ready) if (link_enc->funcs->fec_set_ready && link->dpcd_caps.fec_cap.bits.FEC_CAPABLE) { - if (link->fec_state == dc_link_fec_not_ready && ready) { + if (ready) { fec_config = 1; if (core_link_write_dpcd(link, DP_FEC_CONFIGURATION, @@ -3033,9 +4101,11 @@ void dp_set_fec_ready(struct dc_link *link, bool ready) link_enc->funcs->fec_set_ready(link_enc, true); link->fec_state = dc_link_fec_ready; } else { + link->link_enc->funcs->fec_set_ready(link->link_enc, false); + link->fec_state = dc_link_fec_not_ready; dm_error("dpcd write failed to set fec_ready"); } - } else if (link->fec_state == dc_link_fec_ready && !ready) { + } else if (link->fec_state == dc_link_fec_ready) { fec_config = 0; core_link_write_dpcd(link, DP_FEC_CONFIGURATION, @@ -3059,7 +4129,14 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) if (link_enc->funcs->fec_set_enable && link->dpcd_caps.fec_cap.bits.FEC_CAPABLE) { if (link->fec_state == dc_link_fec_ready && enable) { - msleep(1); + /* Accord to DP spec, FEC enable sequence can first + * be transmitted anytime after 1000 LL codes have + * been transmitted on the link after link training + * completion. Using 1 lane RBR should have the maximum + * time for transmitting 1000 LL codes which is 6.173 us. + * So use 7 microseconds delay instead. + */ + udelay(7); link_enc->funcs->fec_set_enable(link_enc, true); link->fec_state = dc_link_fec_enabled; } else if (link->fec_state == dc_link_fec_enabled && !enable) { @@ -3068,5 +4145,4 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) } } } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c index 2d019e1f6135..ddb855045767 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c @@ -12,12 +12,38 @@ #include "dc_link_ddc.h" #include "dm_helpers.h" #include "dpcd_defs.h" -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #include "dsc.h" -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "resource.h" -#endif + +static uint8_t convert_to_count(uint8_t lttpr_repeater_count) +{ + switch (lttpr_repeater_count) { + case 0x80: // 1 lttpr repeater + return 1; + case 0x40: // 2 lttpr repeaters + return 2; + case 0x20: // 3 lttpr repeaters + return 3; + case 0x10: // 4 lttpr repeaters + return 4; + case 0x08: // 5 lttpr repeaters + return 5; + case 0x04: // 6 lttpr repeaters + return 6; + case 0x02: // 7 lttpr repeaters + return 7; + case 0x01: // 8 lttpr repeaters + return 8; + default: + break; + } + return 0; // invalid value +} + +static inline bool is_immediate_downstream(struct dc_link *link, uint32_t offset) +{ + return (convert_to_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt) == offset); +} enum dc_status core_link_read_dpcd( struct dc_link *link, @@ -25,10 +51,11 @@ enum dc_status core_link_read_dpcd( uint8_t *data, uint32_t size) { - if (!dm_helpers_dp_read_dpcd(link->ctx, - link, - address, data, size)) - return DC_ERROR_UNEXPECTED; + if (!link->aux_access_disabled && + !dm_helpers_dp_read_dpcd(link->ctx, + link, address, data, size)) { + return DC_ERROR_UNEXPECTED; + } return DC_OK; } @@ -39,10 +66,11 @@ enum dc_status core_link_write_dpcd( const uint8_t *data, uint32_t size) { - if (!dm_helpers_dp_write_dpcd(link->ctx, - link, - address, data, size)) - return DC_ERROR_UNEXPECTED; + if (!link->aux_access_disabled && + !dm_helpers_dp_write_dpcd(link->ctx, + link, address, data, size)) { + return DC_ERROR_UNEXPECTED; + } return DC_OK; } @@ -53,6 +81,9 @@ void dp_receiver_power_ctrl(struct dc_link *link, bool on) state = on ? DP_POWER_STATE_D0 : DP_POWER_STATE_D3; + if (link->sync_lt_in_progress) + return; + core_link_write_dpcd(link, DP_SET_POWER, &state, sizeof(state)); } @@ -64,8 +95,8 @@ void dp_enable_link_phy( const struct dc_link_settings *link_settings) { struct link_encoder *link_enc = link->link_enc; - struct dc *core_dc = link->ctx->dc; - struct dmcu *dmcu = core_dc->res_pool->dmcu; + struct dc *dc = link->ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; struct pipe_ctx *pipes = link->dc->current_state->res_ctx.pipe_ctx; @@ -142,15 +173,20 @@ bool edp_receiver_ready_T9(struct dc_link *link) } bool edp_receiver_ready_T7(struct dc_link *link) { - unsigned int tries = 0; unsigned char sinkstatus = 0; unsigned char edpRev = 0; enum dc_status result = DC_OK; + /* use absolute time stamp to constrain max T7*/ + unsigned long long enter_timestamp = 0; + unsigned long long finish_timestamp = 0; + unsigned long long time_taken_in_ns = 0; + result = core_link_read_dpcd(link, DP_EDP_DPCD_REV, &edpRev, sizeof(edpRev)); if (result == DC_OK && edpRev < DP_EDP_12) return true; /* start from eDP version 1.2, SINK_STAUS indicate the sink is ready.*/ + enter_timestamp = dm_get_timestamp(link->ctx); do { sinkstatus = 0; result = core_link_read_dpcd(link, DP_SINK_STATUS, &sinkstatus, sizeof(sinkstatus)); @@ -158,15 +194,21 @@ bool edp_receiver_ready_T7(struct dc_link *link) break; if (result != DC_OK) break; - udelay(25); //MAx T7 is 50ms - } while (++tries < 300); + udelay(25); + finish_timestamp = dm_get_timestamp(link->ctx); + time_taken_in_ns = dm_get_elapse_time_in_ns(link->ctx, finish_timestamp, enter_timestamp); + } while (time_taken_in_ns < 50 * 1000000); //MAx T7 is 50ms + + if (link->local_sink->edid_caps.panel_patch.extra_t7_ms > 0) + udelay(link->local_sink->edid_caps.panel_patch.extra_t7_ms * 1000); + return result; } void dp_disable_link_phy(struct dc_link *link, enum signal_type signal) { - struct dc *core_dc = link->ctx->dc; - struct dmcu *dmcu = core_dc->res_pool->dmcu; + struct dc *dc = link->ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; if (!link->wa_flags.dp_keep_receiver_powered) dp_receiver_power_ctrl(link, false); @@ -203,21 +245,22 @@ void dp_disable_link_phy_mst(struct dc_link *link, enum signal_type signal) bool dp_set_hw_training_pattern( struct dc_link *link, - enum hw_dp_training_pattern pattern) + enum dc_dp_training_pattern pattern, + uint32_t offset) { enum dp_test_pattern test_pattern = DP_TEST_PATTERN_UNSUPPORTED; switch (pattern) { - case HW_DP_TRAINING_PATTERN_1: + case DP_TRAINING_PATTERN_SEQUENCE_1: test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN1; break; - case HW_DP_TRAINING_PATTERN_2: + case DP_TRAINING_PATTERN_SEQUENCE_2: test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN2; break; - case HW_DP_TRAINING_PATTERN_3: + case DP_TRAINING_PATTERN_SEQUENCE_3: test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN3; break; - case HW_DP_TRAINING_PATTERN_4: + case DP_TRAINING_PATTERN_SEQUENCE_4: test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN4; break; default: @@ -231,54 +274,18 @@ bool dp_set_hw_training_pattern( void dp_set_hw_lane_settings( struct dc_link *link, - const struct link_training_settings *link_settings) + const struct link_training_settings *link_settings, + uint32_t offset) { struct link_encoder *encoder = link->link_enc; + if (!link->is_lttpr_mode_transparent && !is_immediate_downstream(link, offset)) + return; + /* call Encoder to set lane settings */ encoder->funcs->dp_set_lane_settings(encoder, link_settings); } -enum dp_panel_mode dp_get_panel_mode(struct dc_link *link) -{ - /* We need to explicitly check that connector - * is not DP. Some Travis_VGA get reported - * by video bios as DP. - */ - if (link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT) { - - switch (link->dpcd_caps.branch_dev_id) { - case DP_BRANCH_DEVICE_ID_2: - if (strncmp( - link->dpcd_caps.branch_dev_name, - DP_VGA_LVDS_CONVERTER_ID_2, - sizeof( - link->dpcd_caps. - branch_dev_name)) == 0) { - return DP_PANEL_MODE_SPECIAL; - } - break; - case DP_BRANCH_DEVICE_ID_3: - if (strncmp(link->dpcd_caps.branch_dev_name, - DP_VGA_LVDS_CONVERTER_ID_3, - sizeof( - link->dpcd_caps. - branch_dev_name)) == 0) { - return DP_PANEL_MODE_SPECIAL; - } - break; - default: - break; - } - } - - if (link->dpcd_caps.panel_mode_edp) { - return DP_PANEL_MODE_EDP; - } - - return DP_PANEL_MODE_DEFAULT; -} - void dp_set_hw_test_pattern( struct dc_link *link, enum dp_test_pattern test_pattern, @@ -306,9 +313,10 @@ void dp_retrain_link_dp_test(struct dc_link *link, for (i = 0; i < MAX_PIPES; i++) { if (pipes[i].stream != NULL && - !pipes[i].top_pipe && + !pipes[i].top_pipe && !pipes[i].prev_odm_pipe && pipes[i].stream->link != NULL && - pipes[i].stream_res.stream_enc != NULL) { + pipes[i].stream_res.stream_enc != NULL && + pipes[i].stream->link == link) { udelay(100); pipes[i].stream_res.stream_enc->funcs->dp_blank( @@ -320,7 +328,9 @@ void dp_retrain_link_dp_test(struct dc_link *link, dp_receiver_power_ctrl(link, false); - link->dc->hwss.disable_stream(&pipes[i], KEEP_ACQUIRED_RESOURCE); + link->dc->hwss.disable_stream(&pipes[i]); + if ((&pipes[i])->stream_res.audio && !link->dc->debug.az_endpoint_mute_only) + (&pipes[i])->stream_res.audio->funcs->az_disable((&pipes[i])->stream_res.audio); link->link_enc->funcs->disable_output( link->link_enc, @@ -330,20 +340,12 @@ void dp_retrain_link_dp_test(struct dc_link *link, memset(&link->cur_link_settings, 0, sizeof(link->cur_link_settings)); - link->link_enc->funcs->enable_dp_output( - link->link_enc, - link_setting, - pipes[i].clock_source->id); - link->cur_link_settings = *link_setting; - - dp_receiver_power_ctrl(link, true); - perform_link_training_with_retries( - link, link_setting, skip_video_pattern, - LINK_TRAINING_ATTEMPTS); - + LINK_TRAINING_ATTEMPTS, + &pipes[i], + SIGNAL_TYPE_DISPLAY_PORT); link->dc->hwss.enable_stream(&pipes[i]); @@ -367,80 +369,98 @@ void dp_retrain_link_dp_test(struct dc_link *link, } } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #define DC_LOGGER \ dsc->ctx->logger static void dsc_optc_config_log(struct display_stream_compressor *dsc, struct dsc_optc_config *config) { - DC_LOG_DSC("Setting optc DSC config at DSC inst %d", dsc->inst); - DC_LOG_DSC("\n\tbytes_per_pixel %d\n\tis_pixel_format_444 %d\n\tslice_width %d", - config->bytes_per_pixel, - config->is_pixel_format_444, config->slice_width); + uint32_t precision = 1 << 28; + uint32_t bytes_per_pixel_int = config->bytes_per_pixel / precision; + uint32_t bytes_per_pixel_mod = config->bytes_per_pixel % precision; + uint64_t ll_bytes_per_pix_fraq = bytes_per_pixel_mod; + + /* 7 fractional digits decimal precision for bytes per pixel is enough because DSC + * bits per pixel precision is 1/16th of a pixel, which means bytes per pixel precision is + * 1/16/8 = 1/128 of a byte, or 0.0078125 decimal + */ + ll_bytes_per_pix_fraq *= 10000000; + ll_bytes_per_pix_fraq /= precision; + + DC_LOG_DSC("\tbytes_per_pixel 0x%08x (%d.%07d)", + config->bytes_per_pixel, bytes_per_pixel_int, (uint32_t)ll_bytes_per_pix_fraq); + DC_LOG_DSC("\tis_pixel_format_444 %d", config->is_pixel_format_444); + DC_LOG_DSC("\tslice_width %d", config->slice_width); } static bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable) { - struct dc *core_dc = pipe_ctx->stream->ctx->dc; + struct dc *dc = pipe_ctx->stream->ctx->dc; struct dc_stream_state *stream = pipe_ctx->stream; bool result = false; - if (IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) + if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) result = true; else - result = dm_helpers_dp_write_dsc_enable(core_dc->ctx, stream, enable); + result = dm_helpers_dp_write_dsc_enable(dc->ctx, stream, enable); return result; } -/* This has to be done after DSC was enabled on RX first, i.e. after dp_enable_dsc_on_rx() had been called +/* The stream with these settings can be sent (unblanked) only after DSC was enabled on RX first, + * i.e. after dp_enable_dsc_on_rx() had been called */ -static void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) +void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) { struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; - struct dc *core_dc = pipe_ctx->stream->ctx->dc; + struct dc *dc = pipe_ctx->stream->ctx->dc; struct dc_stream_state *stream = pipe_ctx->stream; - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; if (enable) { - /* TODO proper function */ struct dsc_config dsc_cfg; struct dsc_optc_config dsc_optc_cfg; enum optc_dsc_mode optc_dsc_mode; - uint8_t dsc_packed_pps[128]; /* Enable DSC hw block */ - dsc_cfg.pic_width = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; + dsc_cfg.pic_width = (stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right) / opp_cnt; dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; + ASSERT(dsc_cfg.dc_dsc_cfg.num_slices_h % opp_cnt == 0); + dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; - dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg, &dsc_packed_pps[0]); - if (odm_pipe) { - struct display_stream_compressor *bot_dsc = odm_pipe->stream_res.dsc; - uint8_t dsc_packed_pps_odm[128]; - - dsc_cfg.pic_width /= 2; - ASSERT(dsc_cfg.dc_dsc_cfg.num_slices_h % 2 == 0); - dsc_cfg.dc_dsc_cfg.num_slices_h /= 2; - dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg, &dsc_packed_pps_odm[0]); - bot_dsc->funcs->dsc_set_config(bot_dsc, &dsc_cfg, &dsc_optc_cfg, &dsc_packed_pps_odm[0]); - bot_dsc->funcs->dsc_enable(bot_dsc, odm_pipe->stream_res.opp->inst); - } + dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg); dsc->funcs->dsc_enable(dsc, pipe_ctx->stream_res.opp->inst); + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + struct display_stream_compressor *odm_dsc = odm_pipe->stream_res.dsc; + + odm_dsc->funcs->dsc_set_config(odm_dsc, &dsc_cfg, &dsc_optc_cfg); + odm_dsc->funcs->dsc_enable(odm_dsc, odm_pipe->stream_res.opp->inst); + } + dsc_cfg.dc_dsc_cfg.num_slices_h *= opp_cnt; + dsc_cfg.pic_width *= opp_cnt; optc_dsc_mode = dsc_optc_cfg.is_pixel_format_444 ? OPTC_DSC_ENABLED_444 : OPTC_DSC_ENABLED_NATIVE_SUBSAMPLED; - dsc_optc_config_log(dsc, &dsc_optc_cfg); /* Enable DSC in encoder */ - if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment) && pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config) + if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + DC_LOG_DSC("Setting stream encoder DSC config for engine %d:", (int)pipe_ctx->stream_res.stream_enc->id); + dsc_optc_config_log(dsc, &dsc_optc_cfg); pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(pipe_ctx->stream_res.stream_enc, optc_dsc_mode, dsc_optc_cfg.bytes_per_pixel, - dsc_optc_cfg.slice_width, - &dsc_packed_pps[0]); + dsc_optc_cfg.slice_width); + + /* PPS SDP is set elsewhere because it has to be done after DIG FE is connected to DIG BE */ + } /* Enable DSC in OPTC */ + DC_LOG_DSC("Setting optc DSC config for tg instance %d:", pipe_ctx->stream_res.tg->inst); + dsc_optc_config_log(dsc, &dsc_optc_cfg); pipe_ctx->stream_res.tg->funcs->set_dsc_config(pipe_ctx->stream_res.tg, optc_dsc_mode, dsc_optc_cfg.bytes_per_pixel, @@ -452,15 +472,18 @@ static void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) OPTC_DSC_DISABLED, 0, 0); /* disable DSC in stream encoder */ - if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { + if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config( pipe_ctx->stream_res.stream_enc, - OPTC_DSC_DISABLED, 0, 0, NULL); + OPTC_DSC_DISABLED, 0, 0); + + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.stream_enc, false, NULL); } /* disable DSC block */ pipe_ctx->stream_res.dsc->funcs->dsc_disable(pipe_ctx->stream_res.dsc); - if (odm_pipe) + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) odm_pipe->stream_res.dsc->funcs->dsc_disable(odm_pipe->stream_res.dsc); } } @@ -489,6 +512,50 @@ out: return result; } +bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable) +{ + struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; + struct dc *dc = pipe_ctx->stream->ctx->dc; + struct dc_stream_state *stream = pipe_ctx->stream; + + if (!pipe_ctx->stream->timing.flags.DSC || !dsc) + return false; + + if (enable) { + struct dsc_config dsc_cfg; + uint8_t dsc_packed_pps[128]; + + memset(&dsc_cfg, 0, sizeof(dsc_cfg)); + memset(dsc_packed_pps, 0, 128); + + /* Enable DSC hw block */ + dsc_cfg.pic_width = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; + dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; + dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; + dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; + + DC_LOG_DSC(" "); + dsc->funcs->dsc_get_packed_pps(dsc, &dsc_cfg, &dsc_packed_pps[0]); + if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + DC_LOG_DSC("Setting stream encoder DSC PPS SDP for engine %d\n", (int)pipe_ctx->stream_res.stream_enc->id); + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.stream_enc, + true, + &dsc_packed_pps[0]); + } + } else { + /* disable DSC PPS in stream encoder */ + if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.stream_enc, false, NULL); + } + } + + return true; +} + + bool dp_update_dsc_config(struct pipe_ctx *pipe_ctx) { struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; @@ -499,8 +566,7 @@ bool dp_update_dsc_config(struct pipe_ctx *pipe_ctx) return false; dp_set_dsc_on_stream(pipe_ctx, true); + dp_set_dsc_pps_sdp(pipe_ctx, true); return true; } -#endif - diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 2ceaab4fb5de..a0eb9e533a61 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -46,12 +46,11 @@ #include "dce100/dce100_resource.h" #include "dce110/dce110_resource.h" #include "dce112/dce112_resource.h" -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) #include "dcn10/dcn10_resource.h" #endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dcn20/dcn20_resource.h" -#endif +#include "dcn21/dcn21_resource.h" #include "dce120/dce120_resource.h" #define DC_LOGGER_INIT(logger) @@ -96,19 +95,19 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id) else dc_version = DCE_VERSION_12_0; break; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) case FAMILY_RV: dc_version = DCN_VERSION_1_0; if (ASICREV_IS_RAVEN2(asic_id.hw_internal_rev)) dc_version = DCN_VERSION_1_01; + if (ASICREV_IS_RENOIR(asic_id.hw_internal_rev)) + dc_version = DCN_VERSION_2_1; break; #endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) case FAMILY_NV: dc_version = DCN_VERSION_2_0; break; -#endif default: dc_version = DCE_VERSION_UNKNOWN; break; @@ -155,30 +154,29 @@ struct resource_pool *dc_create_resource_pool(struct dc *dc, init_data->num_virtual_links, dc); break; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) case DCN_VERSION_1_0: case DCN_VERSION_1_01: res_pool = dcn10_create_resource_pool(init_data, dc); break; -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) case DCN_VERSION_2_0: res_pool = dcn20_create_resource_pool(init_data, dc); break; + case DCN_VERSION_2_1: + res_pool = dcn21_create_resource_pool(init_data, dc); + break; #endif default: break; } - if (res_pool != NULL) { - struct dc_firmware_info fw_info = { { 0 } }; - if (dc->ctx->dc_bios->funcs->get_firmware_info(dc->ctx->dc_bios, - &fw_info) == BP_RESULT_OK) { + if (res_pool != NULL) { + if (dc->ctx->dc_bios->fw_info_valid) { res_pool->ref_clocks.xtalin_clock_inKhz = - fw_info.pll_info.crystal_frequency; + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; /* initialize with firmware data first, no all * ASIC have DCCG SW component. FPGA or * simulation need initialization of @@ -265,12 +263,10 @@ bool resource_construct( DC_ERR("DC: failed to create audio!\n"); return false; } - if (!aud->funcs->endpoint_valid(aud)) { aud->funcs->destroy(&aud); break; } - pool->audios[i] = aud; pool->audio_count++; } @@ -396,6 +392,9 @@ bool resource_are_streams_timing_synchronizable( if (stream1->view_format != stream2->view_format) return false; + if (stream1->ignore_msa_timing_param || stream2->ignore_msa_timing_param) + return false; + return true; } static bool is_dp_and_hdmi_sharable( @@ -941,11 +940,51 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx) } +/* + * When handling 270 rotation in mixed SLS mode, we have + * stream->timing.h_border_left that is non zero. If we are doing + * pipe-splitting, this h_border_left value gets added to recout.x and when it + * calls calculate_inits_and_adj_vp() and + * adjust_vp_and_init_for_seamless_clip(), it can cause viewport.height for a + * pipe to be incorrect. + * + * To fix this, instead of using stream->timing.h_border_left, we can use + * stream->dst.x to represent the border instead. So we will set h_border_left + * to 0 and shift the appropriate amount in stream->dst.x. We will then + * perform all calculations in resource_build_scaling_params() based on this + * and then restore the h_border_left and stream->dst.x to their original + * values. + * + * shift_border_left_to_dst() will shift the amount of h_border_left to + * stream->dst.x and set h_border_left to 0. restore_border_left_from_dst() + * will restore h_border_left and stream->dst.x back to their original values + * We also need to make sure pipe_ctx->plane_res.scl_data.h_active uses the + * original h_border_left value in its calculation. + */ +int shift_border_left_to_dst(struct pipe_ctx *pipe_ctx) +{ + int store_h_border_left = pipe_ctx->stream->timing.h_border_left; + + if (store_h_border_left) { + pipe_ctx->stream->timing.h_border_left = 0; + pipe_ctx->stream->dst.x += store_h_border_left; + } + return store_h_border_left; +} + +void restore_border_left_from_dst(struct pipe_ctx *pipe_ctx, + int store_h_border_left) +{ + pipe_ctx->stream->dst.x -= store_h_border_left; + pipe_ctx->stream->timing.h_border_left = store_h_border_left; +} + bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; struct dc_crtc_timing *timing = &pipe_ctx->stream->timing; bool res = false; + int store_h_border_left = shift_border_left_to_dst(pipe_ctx); DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger); /* Important: scaling ratio calculation requires pixel format, * lb depth calculation requires recout and taps require scaling ratios. @@ -958,8 +997,14 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) calculate_viewport(pipe_ctx); - if (pipe_ctx->plane_res.scl_data.viewport.height < 16 || pipe_ctx->plane_res.scl_data.viewport.width < 16) + if (pipe_ctx->plane_res.scl_data.viewport.height < 16 || + pipe_ctx->plane_res.scl_data.viewport.width < 16) { + if (store_h_border_left) { + restore_border_left_from_dst(pipe_ctx, + store_h_border_left); + } return false; + } calculate_recout(pipe_ctx); @@ -972,8 +1017,10 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.scl_data.recout.x += timing->h_border_left; pipe_ctx->plane_res.scl_data.recout.y += timing->v_border_top; - pipe_ctx->plane_res.scl_data.h_active = timing->h_addressable + timing->h_border_left + timing->h_border_right; - pipe_ctx->plane_res.scl_data.v_active = timing->v_addressable + timing->v_border_top + timing->v_border_bottom; + pipe_ctx->plane_res.scl_data.h_active = timing->h_addressable + + store_h_border_left + timing->h_border_right; + pipe_ctx->plane_res.scl_data.v_active = timing->v_addressable + + timing->v_border_top + timing->v_border_bottom; /* Taps calculations */ if (pipe_ctx->plane_res.xfm != NULL) @@ -983,6 +1030,8 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) if (pipe_ctx->plane_res.dpp != NULL) res = pipe_ctx->plane_res.dpp->funcs->dpp_get_optimal_number_of_taps( pipe_ctx->plane_res.dpp, &pipe_ctx->plane_res.scl_data, &plane_state->scaling_quality); + + if (!res) { /* Try 24 bpp linebuffer */ pipe_ctx->plane_res.scl_data.lb_params.depth = LB_PIXEL_DEPTH_24BPP; @@ -1018,6 +1067,9 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) plane_state->dst_rect.x, plane_state->dst_rect.y); + if (store_h_border_left) + restore_border_left_from_dst(pipe_ctx, store_h_border_left); + return res; } @@ -1103,25 +1155,21 @@ struct pipe_ctx *resource_get_head_pipe_for_stream( struct dc_stream_state *stream) { int i; + for (i = 0; i < MAX_PIPES; i++) { - if (res_ctx->pipe_ctx[i].stream == stream && - !res_ctx->pipe_ctx[i].top_pipe) { + if (res_ctx->pipe_ctx[i].stream == stream + && !res_ctx->pipe_ctx[i].top_pipe + && !res_ctx->pipe_ctx[i].prev_odm_pipe) return &res_ctx->pipe_ctx[i]; - break; - } } return NULL; } -static struct pipe_ctx *resource_get_tail_pipe_for_stream( +static struct pipe_ctx *resource_get_tail_pipe( struct resource_context *res_ctx, - struct dc_stream_state *stream) + struct pipe_ctx *head_pipe) { - struct pipe_ctx *head_pipe, *tail_pipe; - head_pipe = resource_get_head_pipe_for_stream(res_ctx, stream); - - if (!head_pipe) - return NULL; + struct pipe_ctx *tail_pipe; tail_pipe = head_pipe->bottom_pipe; @@ -1137,31 +1185,20 @@ static struct pipe_ctx *resource_get_tail_pipe_for_stream( * A free_pipe for a stream is defined here as a pipe * that has no surface attached yet */ -static struct pipe_ctx *acquire_free_pipe_for_stream( +static struct pipe_ctx *acquire_free_pipe_for_head( struct dc_state *context, const struct resource_pool *pool, - struct dc_stream_state *stream) + struct pipe_ctx *head_pipe) { int i; struct resource_context *res_ctx = &context->res_ctx; - struct pipe_ctx *head_pipe = NULL; - - /* Find head pipe, which has the back end set up*/ - - head_pipe = resource_get_head_pipe_for_stream(res_ctx, stream); - - if (!head_pipe) { - ASSERT(0); - return NULL; - } - if (!head_pipe->plane_state) return head_pipe; /* Re-use pipe already acquired for this stream if available*/ for (i = pool->pipe_count - 1; i >= 0; i--) { - if (res_ctx->pipe_ctx[i].stream == stream && + if (res_ctx->pipe_ctx[i].stream == head_pipe->stream && !res_ctx->pipe_ctx[i].plane_state) { return &res_ctx->pipe_ctx[i]; } @@ -1175,11 +1212,10 @@ static struct pipe_ctx *acquire_free_pipe_for_stream( if (!pool->funcs->acquire_idle_pipe_for_layer) return NULL; - return pool->funcs->acquire_idle_pipe_for_layer(context, pool, stream); - + return pool->funcs->acquire_idle_pipe_for_layer(context, pool, head_pipe->stream); } -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) static int acquire_first_split_pipe( struct resource_context *res_ctx, const struct resource_pool *pool, @@ -1190,7 +1226,7 @@ static int acquire_first_split_pipe( for (i = 0; i < pool->pipe_count; i++) { struct pipe_ctx *split_pipe = &res_ctx->pipe_ctx[i]; - if (split_pipe->top_pipe && !dc_res_is_odm_head_pipe(split_pipe) && + if (split_pipe->top_pipe && split_pipe->top_pipe->plane_state == split_pipe->plane_state) { split_pipe->top_pipe->bottom_pipe = split_pipe->bottom_pipe; if (split_pipe->bottom_pipe) @@ -1251,39 +1287,41 @@ bool dc_add_plane_to_context( return false; } - tail_pipe = resource_get_tail_pipe_for_stream(&context->res_ctx, stream); - ASSERT(tail_pipe); + /* retain new surface, but only once per stream */ + dc_plane_state_retain(plane_state); - free_pipe = acquire_free_pipe_for_stream(context, pool, stream); + while (head_pipe) { + tail_pipe = resource_get_tail_pipe(&context->res_ctx, head_pipe); + ASSERT(tail_pipe); -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) - if (!free_pipe) { - int pipe_idx = acquire_first_split_pipe(&context->res_ctx, pool, stream); - if (pipe_idx >= 0) - free_pipe = &context->res_ctx.pipe_ctx[pipe_idx]; - } -#endif - if (!free_pipe) - return false; + free_pipe = acquire_free_pipe_for_head(context, pool, head_pipe); - /* retain new surfaces */ - dc_plane_state_retain(plane_state); - free_pipe->plane_state = plane_state; - - if (head_pipe != free_pipe) { - free_pipe->stream_res.tg = tail_pipe->stream_res.tg; - free_pipe->stream_res.abm = tail_pipe->stream_res.abm; - free_pipe->stream_res.opp = tail_pipe->stream_res.opp; - free_pipe->stream_res.stream_enc = tail_pipe->stream_res.stream_enc; - free_pipe->stream_res.audio = tail_pipe->stream_res.audio; - free_pipe->clock_source = tail_pipe->clock_source; - free_pipe->top_pipe = tail_pipe; - tail_pipe->bottom_pipe = free_pipe; - } else if (free_pipe->bottom_pipe && free_pipe->bottom_pipe->plane_state == NULL) { - ASSERT(free_pipe->bottom_pipe->stream_res.opp != free_pipe->stream_res.opp); - free_pipe->bottom_pipe->plane_state = plane_state; - } + #if defined(CONFIG_DRM_AMD_DC_DCN) + if (!free_pipe) { + int pipe_idx = acquire_first_split_pipe(&context->res_ctx, pool, stream); + if (pipe_idx >= 0) + free_pipe = &context->res_ctx.pipe_ctx[pipe_idx]; + } + #endif + if (!free_pipe) { + dc_plane_state_release(plane_state); + return false; + } + free_pipe->plane_state = plane_state; + + if (head_pipe != free_pipe) { + free_pipe->stream_res.tg = tail_pipe->stream_res.tg; + free_pipe->stream_res.abm = tail_pipe->stream_res.abm; + free_pipe->stream_res.opp = tail_pipe->stream_res.opp; + free_pipe->stream_res.stream_enc = tail_pipe->stream_res.stream_enc; + free_pipe->stream_res.audio = tail_pipe->stream_res.audio; + free_pipe->clock_source = tail_pipe->clock_source; + free_pipe->top_pipe = tail_pipe; + tail_pipe->bottom_pipe = free_pipe; + } + head_pipe = head_pipe->next_odm_pipe; + } /* assign new surfaces*/ stream_status->plane_states[stream_status->plane_count] = plane_state; @@ -1292,35 +1330,6 @@ bool dc_add_plane_to_context( return true; } -struct pipe_ctx *dc_res_get_odm_bottom_pipe(struct pipe_ctx *pipe_ctx) -{ - struct pipe_ctx *bottom_pipe = pipe_ctx->bottom_pipe; - - /* ODM should only be updated once per otg */ - if (pipe_ctx->top_pipe) - return NULL; - - while (bottom_pipe) { - if (bottom_pipe->stream_res.opp != pipe_ctx->stream_res.opp) - break; - bottom_pipe = bottom_pipe->bottom_pipe; - } - - return bottom_pipe; -} - -bool dc_res_is_odm_head_pipe(struct pipe_ctx *pipe_ctx) -{ - struct pipe_ctx *top_pipe = pipe_ctx->top_pipe; - - if (!top_pipe) - return false; - if (top_pipe && top_pipe->stream_res.opp == pipe_ctx->stream_res.opp) - return false; - - return true; -} - bool dc_remove_plane_from_context( const struct dc *dc, struct dc_stream_state *stream, @@ -1347,12 +1356,6 @@ bool dc_remove_plane_from_context( struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; if (pipe_ctx->plane_state == plane_state) { - if (dc_res_is_odm_head_pipe(pipe_ctx)) { - pipe_ctx->plane_state = NULL; - pipe_ctx->bottom_pipe = NULL; - continue; - } - if (pipe_ctx->top_pipe) pipe_ctx->top_pipe->bottom_pipe = pipe_ctx->bottom_pipe; @@ -1367,13 +1370,10 @@ bool dc_remove_plane_from_context( * For head pipe detach surfaces from pipe for tail * pipe just zero it out */ - if (!pipe_ctx->top_pipe) { + if (!pipe_ctx->top_pipe) pipe_ctx->plane_state = NULL; - if (!dc_res_get_odm_bottom_pipe(pipe_ctx)) - pipe_ctx->bottom_pipe = NULL; - } else { + else memset(pipe_ctx, 0, sizeof(*pipe_ctx)); - } } } @@ -1568,6 +1568,9 @@ bool dc_is_stream_unchanged( if (!are_stream_backends_same(old_stream, stream)) return false; + if (old_stream->ignore_msa_timing_param != stream->ignore_msa_timing_param) + return false; + return true; } @@ -1657,26 +1660,28 @@ static int acquire_first_free_pipe( static struct audio *find_first_free_audio( struct resource_context *res_ctx, const struct resource_pool *pool, - enum engine_id id) + enum engine_id id, + enum dce_version dc_version) { - int i; - for (i = 0; i < pool->audio_count; i++) { + int i, available_audio_count; + + available_audio_count = pool->audio_count; + + for (i = 0; i < available_audio_count; i++) { if ((res_ctx->is_audio_acquired[i] == false) && (res_ctx->is_stream_enc_acquired[i] == true)) { /*we have enough audio endpoint, find the matching inst*/ if (id != i) continue; - return pool->audios[i]; } } - /* use engine id to find free audio */ - if ((id < pool->audio_count) && (res_ctx->is_audio_acquired[id] == false)) { + /* use engine id to find free audio */ + if ((id < available_audio_count) && (res_ctx->is_audio_acquired[id] == false)) { return pool->audios[id]; } - /*not found the matching one, first come first serve*/ - for (i = 0; i < pool->audio_count; i++) { + for (i = 0; i < available_audio_count; i++) { if (res_ctx->is_audio_acquired[i] == false) { return pool->audios[i]; } @@ -1736,50 +1741,45 @@ enum dc_status dc_remove_stream_from_ctx( { int i; struct dc_context *dc_ctx = dc->ctx; - struct pipe_ctx *del_pipe = NULL; - - /* Release primary pipe */ - for (i = 0; i < MAX_PIPES; i++) { - if (new_ctx->res_ctx.pipe_ctx[i].stream == stream && - !new_ctx->res_ctx.pipe_ctx[i].top_pipe) { - struct pipe_ctx *odm_pipe = - dc_res_get_odm_bottom_pipe(&new_ctx->res_ctx.pipe_ctx[i]); + struct pipe_ctx *del_pipe = resource_get_head_pipe_for_stream(&new_ctx->res_ctx, stream); + struct pipe_ctx *odm_pipe; - del_pipe = &new_ctx->res_ctx.pipe_ctx[i]; + if (!del_pipe) { + DC_ERROR("Pipe not found for stream %p !\n", stream); + return DC_ERROR_UNEXPECTED; + } - ASSERT(del_pipe->stream_res.stream_enc); - update_stream_engine_usage( - &new_ctx->res_ctx, - dc->res_pool, - del_pipe->stream_res.stream_enc, - false); + odm_pipe = del_pipe->next_odm_pipe; - if (del_pipe->stream_res.audio) - update_audio_usage( - &new_ctx->res_ctx, - dc->res_pool, - del_pipe->stream_res.audio, - false); + /* Release primary pipe */ + ASSERT(del_pipe->stream_res.stream_enc); + update_stream_engine_usage( + &new_ctx->res_ctx, + dc->res_pool, + del_pipe->stream_res.stream_enc, + false); - resource_unreference_clock_source(&new_ctx->res_ctx, - dc->res_pool, - del_pipe->clock_source); + if (del_pipe->stream_res.audio) + update_audio_usage( + &new_ctx->res_ctx, + dc->res_pool, + del_pipe->stream_res.audio, + false); - if (dc->res_pool->funcs->remove_stream_from_ctx) - dc->res_pool->funcs->remove_stream_from_ctx(dc, new_ctx, stream); + resource_unreference_clock_source(&new_ctx->res_ctx, + dc->res_pool, + del_pipe->clock_source); - memset(del_pipe, 0, sizeof(*del_pipe)); - if (odm_pipe) - memset(odm_pipe, 0, sizeof(*odm_pipe)); + if (dc->res_pool->funcs->remove_stream_from_ctx) + dc->res_pool->funcs->remove_stream_from_ctx(dc, new_ctx, stream); - break; - } - } + while (odm_pipe) { + struct pipe_ctx *next_odm_pipe = odm_pipe->next_odm_pipe; - if (!del_pipe) { - DC_ERROR("Pipe not found for stream %p !\n", stream); - return DC_ERROR_UNEXPECTED; + memset(odm_pipe, 0, sizeof(*odm_pipe)); + odm_pipe = next_odm_pipe; } + memset(del_pipe, 0, sizeof(*del_pipe)); for (i = 0; i < new_ctx->stream_count; i++) if (new_ctx->streams[i] == stream) @@ -1880,40 +1880,67 @@ static int acquire_resource_from_hw_enabled_state( struct dc_stream_state *stream) { struct dc_link *link = stream->link; - unsigned int inst; + unsigned int i, inst, tg_inst = 0; /* Check for enabled DIG to identify enabled display */ if (!link->link_enc->funcs->is_dig_enabled(link->link_enc)) return -1; - /* Check for which front end is used by this encoder. - * Note the inst is 1 indexed, where 0 is undefined. - * Note that DIG_FE can source from different OTG but our - * current implementation always map 1-to-1, so this code makes - * the same assumption and doesn't check OTG source. - */ - inst = link->link_enc->funcs->get_dig_frontend(link->link_enc) - 1; + inst = link->link_enc->funcs->get_dig_frontend(link->link_enc); + + if (inst == ENGINE_ID_UNKNOWN) + return -1; - /* Instance should be within the range of the pool */ - if (inst >= pool->pipe_count) + for (i = 0; i < pool->stream_enc_count; i++) { + if (pool->stream_enc[i]->id == inst) { + tg_inst = pool->stream_enc[i]->funcs->dig_source_otg( + pool->stream_enc[i]); + break; + } + } + + // tg_inst not found + if (i == pool->stream_enc_count) return -1; - if (!res_ctx->pipe_ctx[inst].stream) { - struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[inst]; + if (tg_inst >= pool->timing_generator_count) + return -1; + + if (!res_ctx->pipe_ctx[tg_inst].stream) { + struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[tg_inst]; + + pipe_ctx->stream_res.tg = pool->timing_generators[tg_inst]; + pipe_ctx->plane_res.mi = pool->mis[tg_inst]; + pipe_ctx->plane_res.hubp = pool->hubps[tg_inst]; + pipe_ctx->plane_res.ipp = pool->ipps[tg_inst]; + pipe_ctx->plane_res.xfm = pool->transforms[tg_inst]; + pipe_ctx->plane_res.dpp = pool->dpps[tg_inst]; + pipe_ctx->stream_res.opp = pool->opps[tg_inst]; + + if (pool->dpps[tg_inst]) { + pipe_ctx->plane_res.mpcc_inst = pool->dpps[tg_inst]->inst; - pipe_ctx->stream_res.tg = pool->timing_generators[inst]; - pipe_ctx->plane_res.mi = pool->mis[inst]; - pipe_ctx->plane_res.hubp = pool->hubps[inst]; - pipe_ctx->plane_res.ipp = pool->ipps[inst]; - pipe_ctx->plane_res.xfm = pool->transforms[inst]; - pipe_ctx->plane_res.dpp = pool->dpps[inst]; - pipe_ctx->stream_res.opp = pool->opps[inst]; - if (pool->dpps[inst]) - pipe_ctx->plane_res.mpcc_inst = pool->dpps[inst]->inst; - pipe_ctx->pipe_idx = inst; + // Read DPP->MPCC->OPP Pipe from HW State + if (pool->mpc->funcs->read_mpcc_state) { + struct mpcc_state s = {0}; + + pool->mpc->funcs->read_mpcc_state(pool->mpc, pipe_ctx->plane_res.mpcc_inst, &s); + + if (s.dpp_id < MAX_MPCC) + pool->mpc->mpcc_array[pipe_ctx->plane_res.mpcc_inst].dpp_id = s.dpp_id; + + if (s.bot_mpcc_id < MAX_MPCC) + pool->mpc->mpcc_array[pipe_ctx->plane_res.mpcc_inst].mpcc_bot = + &pool->mpc->mpcc_array[s.bot_mpcc_id]; + + if (s.opp_id < MAX_OPP) + pipe_ctx->stream_res.opp->mpc_tree_params.opp_id = s.opp_id; + } + } + pipe_ctx->pipe_idx = tg_inst; pipe_ctx->stream = stream; - return inst; + return tg_inst; } return -1; @@ -1961,7 +1988,7 @@ enum dc_status resource_map_pool_resources( /* acquire new resources */ pipe_idx = acquire_first_free_pipe(&context->res_ctx, pool, stream); -#ifdef CONFIG_DRM_AMD_DC_DCN1_0 +#ifdef CONFIG_DRM_AMD_DC_DCN if (pipe_idx < 0) pipe_idx = acquire_first_split_pipe(&context->res_ctx, pool, stream); #endif @@ -1988,7 +2015,7 @@ enum dc_status resource_map_pool_resources( dc_is_audio_capable_signal(pipe_ctx->stream->signal) && stream->audio_info.mode_count && stream->audio_info.flags.all) { pipe_ctx->stream_res.audio = find_first_free_audio( - &context->res_ctx, pool, pipe_ctx->stream_res.stream_enc->id); + &context->res_ctx, pool, pipe_ctx->stream_res.stream_enc->id, dc_ctx->dce_version); /* * Audio assigned in order first come first get. @@ -2039,6 +2066,13 @@ void dc_resource_state_construct( dst_ctx->clk_mgr = dc->clk_mgr; } + +bool dc_resource_is_dsc_encoding_supported(const struct dc *dc) +{ + return dc->res_pool->res_cap->num_dsc > 0; +} + + /** * dc_validate_global_state() - Determine if HW can support a given state * Checks HW resource availability and bandwidth requirement. @@ -2295,7 +2329,7 @@ static void set_avi_info_frame( if (color_space == COLOR_SPACE_SRGB || color_space == COLOR_SPACE_2020_RGB_FULLRANGE) { hdmi_info.bits.Q0_Q1 = RGB_QUANTIZATION_FULL_RANGE; - hdmi_info.bits.YQ0_YQ1 = YYC_QUANTIZATION_FULL_RANGE; + hdmi_info.bits.YQ0_YQ1 = YYC_QUANTIZATION_LIMITED_RANGE; } else if (color_space == COLOR_SPACE_SRGB_LIMITED || color_space == COLOR_SPACE_2020_RGB_LIMITEDRANGE) { hdmi_info.bits.Q0_Q1 = RGB_QUANTIZATION_LIMITED_RANGE; @@ -2475,6 +2509,12 @@ void dc_resource_state_copy_construct( if (cur_pipe->bottom_pipe) cur_pipe->bottom_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->bottom_pipe->pipe_idx]; + + if (cur_pipe->next_odm_pipe) + cur_pipe->next_odm_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->next_odm_pipe->pipe_idx]; + + if (cur_pipe->prev_odm_pipe) + cur_pipe->prev_odm_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->prev_odm_pipe->pipe_idx]; } for (i = 0; i < dst_ctx->stream_count; i++) { @@ -2755,9 +2795,8 @@ void resource_build_bit_depth_reduction_params(struct dc_stream_state *stream, enum dc_status dc_validate_stream(struct dc *dc, struct dc_stream_state *stream) { - struct dc *core_dc = dc; struct dc_link *link = stream->link; - struct timing_generator *tg = core_dc->res_pool->timing_generators[0]; + struct timing_generator *tg = dc->res_pool->timing_generators[0]; enum dc_status res = DC_OK; calculate_phy_pix_clks(stream); @@ -2820,3 +2859,48 @@ unsigned int resource_pixel_format_to_bpp(enum surface_pixel_format format) return -1; } } +static unsigned int get_max_audio_sample_rate(struct audio_mode *modes) +{ + if (modes) { + if (modes->sample_rates.rate.RATE_192) + return 192000; + if (modes->sample_rates.rate.RATE_176_4) + return 176400; + if (modes->sample_rates.rate.RATE_96) + return 96000; + if (modes->sample_rates.rate.RATE_88_2) + return 88200; + if (modes->sample_rates.rate.RATE_48) + return 48000; + if (modes->sample_rates.rate.RATE_44_1) + return 44100; + if (modes->sample_rates.rate.RATE_32) + return 32000; + } + /*original logic when no audio info*/ + return 441000; +} + +void get_audio_check(struct audio_info *aud_modes, + struct audio_check *audio_chk) +{ + unsigned int i; + unsigned int max_sample_rate = 0; + + if (aud_modes) { + audio_chk->audio_packet_type = 0x2;/*audio sample packet AP = .25 for layout0, 1 for layout1*/ + + audio_chk->max_audiosample_rate = 0; + for (i = 0; i < aud_modes->mode_count; i++) { + max_sample_rate = get_max_audio_sample_rate(&aud_modes->modes[i]); + if (audio_chk->max_audiosample_rate < max_sample_rate) + audio_chk->max_audiosample_rate = max_sample_rate; + /*dts takes the same as type 2: AP = 0.25*/ + } + /*check which one take more bandwidth*/ + if (audio_chk->max_audiosample_rate > 192000) + audio_chk->audio_packet_type = 0x9;/*AP =1*/ + audio_chk->acat = 0;/*not support*/ + } +} + diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_sink.c b/drivers/gpu/drm/amd/display/dc/core/dc_sink.c index 5cbfdf1c4b11..a249a0e5edd0 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_sink.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_sink.c @@ -33,7 +33,7 @@ * Private functions ******************************************************************************/ -static void destruct(struct dc_sink *sink) +static void dc_sink_destruct(struct dc_sink *sink) { if (sink->dc_container_id) { kfree(sink->dc_container_id); @@ -41,7 +41,7 @@ static void destruct(struct dc_sink *sink) } } -static bool construct(struct dc_sink *sink, const struct dc_sink_init_data *init_params) +static bool dc_sink_construct(struct dc_sink *sink, const struct dc_sink_init_data *init_params) { struct dc_link *link = init_params->link; @@ -75,7 +75,7 @@ void dc_sink_retain(struct dc_sink *sink) static void dc_sink_free(struct kref *kref) { struct dc_sink *sink = container_of(kref, struct dc_sink, refcount); - destruct(sink); + dc_sink_destruct(sink); kfree(sink); } @@ -91,7 +91,7 @@ struct dc_sink *dc_sink_create(const struct dc_sink_init_data *init_params) if (NULL == sink) goto alloc_fail; - if (false == construct(sink, init_params)) + if (false == dc_sink_construct(sink, init_params)) goto construct_fail; kref_init(&sink->refcount); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 352862370390..6ddbb00ed37a 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -27,14 +27,12 @@ #include <linux/slab.h> #include "dm_services.h" +#include "basics/dc_common.h" #include "dc.h" #include "core_types.h" #include "resource.h" #include "ipp.h" #include "timing_generator.h" -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) -#include "dcn10/dcn10_hw_sequencer.h" -#endif #define DC_LOGGER dc->ctx->logger @@ -58,7 +56,7 @@ void update_stream_signal(struct dc_stream_state *stream, struct dc_sink *sink) } } -static void construct(struct dc_stream_state *stream, +static void dc_stream_construct(struct dc_stream_state *stream, struct dc_sink *dc_sink_data) { uint32_t i = 0; @@ -108,7 +106,6 @@ static void construct(struct dc_stream_state *stream, /* EDID CAP translation for HDMI 2.0 */ stream->timing.flags.LTE_340MCSC_SCRAMBLE = dc_sink_data->edid_caps.lte_340mcsc_scramble; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT memset(&stream->timing.dsc_cfg, 0, sizeof(stream->timing.dsc_cfg)); stream->timing.dsc_cfg.num_slices_h = 0; stream->timing.dsc_cfg.num_slices_v = 0; @@ -117,7 +114,6 @@ static void construct(struct dc_stream_state *stream, stream->timing.dsc_cfg.linebuf_depth = 9; stream->timing.dsc_cfg.version_minor = 2; stream->timing.dsc_cfg.ycbcr422_simple = 0; -#endif update_stream_signal(stream, dc_sink_data); @@ -129,7 +125,7 @@ static void construct(struct dc_stream_state *stream, stream->ctx->dc_stream_id_count++; } -static void destruct(struct dc_stream_state *stream) +static void dc_stream_destruct(struct dc_stream_state *stream) { dc_sink_release(stream->sink); if (stream->out_transfer_func != NULL) { @@ -147,7 +143,7 @@ static void dc_stream_free(struct kref *kref) { struct dc_stream_state *stream = container_of(kref, struct dc_stream_state, refcount); - destruct(stream); + dc_stream_destruct(stream); kfree(stream); } @@ -170,7 +166,7 @@ struct dc_stream_state *dc_create_stream_for_sink( if (stream == NULL) return NULL; - construct(stream, sink); + dc_stream_construct(stream, sink); kref_init(&stream->refcount); @@ -237,7 +233,7 @@ struct dc_stream_status *dc_stream_get_status( static void delay_cursor_until_vupdate(struct pipe_ctx *pipe_ctx, struct dc *dc) { -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) unsigned int vupdate_line; unsigned int lines_to_vupdate, us_to_vupdate, vpos, nvpos; struct dc_stream_state *stream = pipe_ctx->stream; @@ -246,7 +242,7 @@ static void delay_cursor_until_vupdate(struct pipe_ctx *pipe_ctx, struct dc *dc) if (stream->ctx->asic_id.chip_family == FAMILY_RV && ASICREV_IS_RAVEN(stream->ctx->asic_id.hw_internal_rev)) { - vupdate_line = get_vupdate_offset_from_vsync(pipe_ctx); + vupdate_line = dc->hwss.get_vupdate_offset_from_vsync(pipe_ctx); if (!dc_stream_get_crtc_position(dc, &stream, 1, &vpos, &nvpos)) return; @@ -272,7 +268,7 @@ bool dc_stream_set_cursor_attributes( const struct dc_cursor_attributes *attributes) { int i; - struct dc *core_dc; + struct dc *dc; struct resource_context *res_ctx; struct pipe_ctx *pipe_to_program = NULL; @@ -290,8 +286,8 @@ bool dc_stream_set_cursor_attributes( return false; } - core_dc = stream->ctx->dc; - res_ctx = &core_dc->current_state->res_ctx; + dc = stream->ctx->dc; + res_ctx = &dc->current_state->res_ctx; stream->cursor_attributes = *attributes; for (i = 0; i < MAX_PIPES; i++) { @@ -303,17 +299,17 @@ bool dc_stream_set_cursor_attributes( if (!pipe_to_program) { pipe_to_program = pipe_ctx; - delay_cursor_until_vupdate(pipe_ctx, core_dc); - core_dc->hwss.pipe_control_lock(core_dc, pipe_to_program, true); + delay_cursor_until_vupdate(pipe_ctx, dc); + dc->hwss.pipe_control_lock(dc, pipe_to_program, true); } - core_dc->hwss.set_cursor_attribute(pipe_ctx); - if (core_dc->hwss.set_cursor_sdr_white_level) - core_dc->hwss.set_cursor_sdr_white_level(pipe_ctx); + dc->hwss.set_cursor_attribute(pipe_ctx); + if (dc->hwss.set_cursor_sdr_white_level) + dc->hwss.set_cursor_sdr_white_level(pipe_ctx); } if (pipe_to_program) - core_dc->hwss.pipe_control_lock(core_dc, pipe_to_program, false); + dc->hwss.pipe_control_lock(dc, pipe_to_program, false); return true; } @@ -323,7 +319,7 @@ bool dc_stream_set_cursor_position( const struct dc_cursor_position *position) { int i; - struct dc *core_dc; + struct dc *dc; struct resource_context *res_ctx; struct pipe_ctx *pipe_to_program = NULL; @@ -337,8 +333,8 @@ bool dc_stream_set_cursor_position( return false; } - core_dc = stream->ctx->dc; - res_ctx = &core_dc->current_state->res_ctx; + dc = stream->ctx->dc; + res_ctx = &dc->current_state->res_ctx; stream->cursor_position = *position; for (i = 0; i < MAX_PIPES; i++) { @@ -354,20 +350,19 @@ bool dc_stream_set_cursor_position( if (!pipe_to_program) { pipe_to_program = pipe_ctx; - delay_cursor_until_vupdate(pipe_ctx, core_dc); - core_dc->hwss.pipe_control_lock(core_dc, pipe_to_program, true); + delay_cursor_until_vupdate(pipe_ctx, dc); + dc->hwss.pipe_control_lock(dc, pipe_to_program, true); } - core_dc->hwss.set_cursor_position(pipe_ctx); + dc->hwss.set_cursor_position(pipe_ctx); } if (pipe_to_program) - core_dc->hwss.pipe_control_lock(core_dc, pipe_to_program, false); + dc->hwss.pipe_control_lock(dc, pipe_to_program, false); return true; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) bool dc_stream_add_writeback(struct dc *dc, struct dc_stream_state *stream, struct dc_writeback_info *wb_info) @@ -411,25 +406,30 @@ bool dc_stream_add_writeback(struct dc *dc, stream->writeback_info[stream->num_wb_info++] = *wb_info; } - if (!dc->hwss.update_bandwidth(dc, dc->current_state)) { - dm_error("DC: update_bandwidth failed!\n"); - return false; - } - - /* enable writeback */ if (dc->hwss.enable_writeback) { struct dc_stream_status *stream_status = dc_stream_get_status(stream); struct dwbc *dwb = dc->res_pool->dwbc[wb_info->dwb_pipe_inst]; + dwb->otg_inst = stream_status->primary_otg_inst; + } + if (IS_DIAG_DC(dc->ctx->dce_environment)) { + if (!dc->hwss.update_bandwidth(dc, dc->current_state)) { + dm_error("DC: update_bandwidth failed!\n"); + return false; + } - if (dwb->funcs->is_enabled(dwb)) { - /* writeback pipe already enabled, only need to update */ - dc->hwss.update_writeback(dc, stream_status, wb_info); - } else { - /* Enable writeback pipe from scratch*/ - dc->hwss.enable_writeback(dc, stream_status, wb_info); + /* enable writeback */ + if (dc->hwss.enable_writeback) { + struct dwbc *dwb = dc->res_pool->dwbc[wb_info->dwb_pipe_inst]; + + if (dwb->funcs->is_enabled(dwb)) { + /* writeback pipe already enabled, only need to update */ + dc->hwss.update_writeback(dc, wb_info, dc->current_state); + } else { + /* Enable writeback pipe from scratch*/ + dc->hwss.enable_writeback(dc, wb_info, dc->current_state); + } } } - return true; } @@ -468,26 +468,35 @@ bool dc_stream_remove_writeback(struct dc *dc, } stream->num_wb_info = j; - /* recalculate and apply DML parameters */ - if (!dc->hwss.update_bandwidth(dc, dc->current_state)) { - dm_error("DC: update_bandwidth failed!\n"); - return false; - } - - /* disable writeback */ - if (dc->hwss.disable_writeback) - dc->hwss.disable_writeback(dc, dwb_pipe_inst); + if (IS_DIAG_DC(dc->ctx->dce_environment)) { + /* recalculate and apply DML parameters */ + if (!dc->hwss.update_bandwidth(dc, dc->current_state)) { + dm_error("DC: update_bandwidth failed!\n"); + return false; + } + /* disable writeback */ + if (dc->hwss.disable_writeback) + dc->hwss.disable_writeback(dc, dwb_pipe_inst); + } return true; } -#endif +bool dc_stream_warmup_writeback(struct dc *dc, + int num_dwb, + struct dc_writeback_info *wb_info) +{ + if (dc->hwss.mmhubbub_warmup) + return dc->hwss.mmhubbub_warmup(dc, num_dwb, wb_info); + else + return false; +} uint32_t dc_stream_get_vblank_counter(const struct dc_stream_state *stream) { uint8_t i; - struct dc *core_dc = stream->ctx->dc; + struct dc *dc = stream->ctx->dc; struct resource_context *res_ctx = - &core_dc->current_state->res_ctx; + &dc->current_state->res_ctx; for (i = 0; i < MAX_PIPES; i++) { struct timing_generator *tg = res_ctx->pipe_ctx[i].stream_res.tg; @@ -544,9 +553,9 @@ bool dc_stream_get_scanoutpos(const struct dc_stream_state *stream, { uint8_t i; bool ret = false; - struct dc *core_dc = stream->ctx->dc; + struct dc *dc = stream->ctx->dc; struct resource_context *res_ctx = - &core_dc->current_state->res_ctx; + &dc->current_state->res_ctx; for (i = 0; i < MAX_PIPES; i++) { struct timing_generator *tg = res_ctx->pipe_ctx[i].stream_res.tg; @@ -566,10 +575,9 @@ bool dc_stream_get_scanoutpos(const struct dc_stream_state *stream, return ret; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) + bool dc_stream_dmdata_status_done(struct dc *dc, struct dc_stream_state *stream) { - bool status = true; struct pipe_ctx *pipe = NULL; int i; @@ -585,8 +593,7 @@ bool dc_stream_dmdata_status_done(struct dc *dc, struct dc_stream_state *stream) if (i == MAX_PIPES) return true; - status = dc->hwss.dmdata_status_done(pipe); - return status; + return dc->hwss.dmdata_status_done(pipe); } bool dc_stream_set_dynamic_metadata(struct dc *dc, @@ -597,6 +604,14 @@ bool dc_stream_set_dynamic_metadata(struct dc *dc, struct hubp *hubp; int i; + /* Dynamic metadata is only supported on HDMI or DP */ + if (!dc_is_hdmi_signal(stream->signal) && !dc_is_dp_signal(stream->signal)) + return false; + + /* Check hardware support */ + if (!dc->hwss.program_dmdata_engine) + return false; + for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream == stream) @@ -612,23 +627,7 @@ bool dc_stream_set_dynamic_metadata(struct dc *dc, pipe_ctx->stream->dmdata_address = attr->address; - if (pipe_ctx->stream_res.stream_enc && - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata != NULL) { - if (pipe_ctx->stream->dmdata_address.quad_part != 0) { - /* if using dynamic meta, don't set up generic infopackets */ - pipe_ctx->stream_res.encoder_info_frame.hdrsmd.valid = false; - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata( - pipe_ctx->stream_res.stream_enc, - true, pipe_ctx->plane_res.hubp->inst, - dc_is_dp_signal(pipe_ctx->stream->signal) ? - dmdata_dp : dmdata_hdmi); - } else - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata( - pipe_ctx->stream_res.stream_enc, - false, pipe_ctx->plane_res.hubp->inst, - dc_is_dp_signal(pipe_ctx->stream->signal) ? - dmdata_dp : dmdata_hdmi); - } + dc->hwss.program_dmdata_engine(pipe_ctx); if (hubp->funcs->dmdata_set_attributes != NULL && pipe_ctx->stream->dmdata_address.quad_part != 0) { @@ -637,7 +636,6 @@ bool dc_stream_set_dynamic_metadata(struct dc *dc, return true; } -#endif void dc_stream_log(const struct dc *dc, const struct dc_stream_state *stream) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c index f40e4fd52fa2..ea1229a3e2b2 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c @@ -37,7 +37,7 @@ /******************************************************************************* * Private functions ******************************************************************************/ -static void construct(struct dc_context *ctx, struct dc_plane_state *plane_state) +static void dc_plane_construct(struct dc_context *ctx, struct dc_plane_state *plane_state) { plane_state->ctx = ctx; @@ -50,7 +50,6 @@ static void construct(struct dc_context *ctx, struct dc_plane_state *plane_state plane_state->in_transfer_func->type = TF_TYPE_BYPASS; plane_state->in_transfer_func->ctx = ctx; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) plane_state->in_shaper_func = dc_create_transfer_func(); if (plane_state->in_shaper_func != NULL) { plane_state->in_shaper_func->type = TF_TYPE_BYPASS; @@ -60,7 +59,6 @@ static void construct(struct dc_context *ctx, struct dc_plane_state *plane_state plane_state->lut3d_func = dc_create_3dlut_func(); if (plane_state->lut3d_func != NULL) { plane_state->lut3d_func->ctx = ctx; - plane_state->lut3d_func->initialized = false; } plane_state->blend_tf = dc_create_transfer_func(); if (plane_state->blend_tf != NULL) { @@ -68,10 +66,9 @@ static void construct(struct dc_context *ctx, struct dc_plane_state *plane_state plane_state->blend_tf->ctx = ctx; } -#endif } -static void destruct(struct dc_plane_state *plane_state) +static void dc_plane_destruct(struct dc_plane_state *plane_state) { if (plane_state->gamma_correction != NULL) { dc_gamma_release(&plane_state->gamma_correction); @@ -81,7 +78,6 @@ static void destruct(struct dc_plane_state *plane_state) plane_state->in_transfer_func); plane_state->in_transfer_func = NULL; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) if (plane_state->in_shaper_func != NULL) { dc_transfer_func_release( plane_state->in_shaper_func); @@ -98,7 +94,6 @@ static void destruct(struct dc_plane_state *plane_state) plane_state->blend_tf = NULL; } -#endif } /******************************************************************************* @@ -113,16 +108,14 @@ void enable_surface_flip_reporting(struct dc_plane_state *plane_state, struct dc_plane_state *dc_create_plane_state(struct dc *dc) { - struct dc *core_dc = dc; - struct dc_plane_state *plane_state = kvzalloc(sizeof(*plane_state), - GFP_KERNEL); + GFP_KERNEL); if (NULL == plane_state) return NULL; kref_init(&plane_state->refcount); - construct(core_dc->ctx, plane_state); + dc_plane_construct(dc->ctx, plane_state); return plane_state; } @@ -142,7 +135,7 @@ const struct dc_plane_status *dc_plane_get_status( const struct dc_plane_state *plane_state) { const struct dc_plane_status *plane_status; - struct dc *core_dc; + struct dc *dc; int i; if (!plane_state || @@ -153,15 +146,15 @@ const struct dc_plane_status *dc_plane_get_status( } plane_status = &plane_state->status; - core_dc = plane_state->ctx->dc; + dc = plane_state->ctx->dc; - if (core_dc->current_state == NULL) + if (dc->current_state == NULL) return NULL; /* Find the current plane state and set its pending bit to false */ - for (i = 0; i < core_dc->res_pool->pipe_count; i++) { + for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = - &core_dc->current_state->res_ctx.pipe_ctx[i]; + &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->plane_state != plane_state) continue; @@ -171,14 +164,14 @@ const struct dc_plane_status *dc_plane_get_status( break; } - for (i = 0; i < core_dc->res_pool->pipe_count; i++) { + for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = - &core_dc->current_state->res_ctx.pipe_ctx[i]; + &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->plane_state != plane_state) continue; - core_dc->hwss.update_pending_status(pipe_ctx); + dc->hwss.update_pending_status(pipe_ctx); } return plane_status; @@ -192,7 +185,7 @@ void dc_plane_state_retain(struct dc_plane_state *plane_state) static void dc_plane_state_free(struct kref *kref) { struct dc_plane_state *plane_state = container_of(kref, struct dc_plane_state, refcount); - destruct(plane_state); + dc_plane_destruct(plane_state); kvfree(plane_state); } @@ -263,7 +256,6 @@ alloc_fail: return NULL; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) static void dc_3dlut_func_free(struct kref *kref) { struct dc_3dlut *lut = container_of(kref, struct dc_3dlut, refcount); @@ -279,7 +271,7 @@ struct dc_3dlut *dc_create_3dlut_func(void) goto alloc_fail; kref_init(&lut->refcount); - lut->initialized = false; + lut->state.raw = 0; return lut; @@ -297,6 +289,5 @@ void dc_3dlut_func_retain(struct dc_3dlut *lut) { kref_get(&lut->refcount); } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index e513028faefa..8ff25b5dd2f6 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -39,7 +39,7 @@ #include "inc/hw/dmcu.h" #include "dml/display_mode_lib.h" -#define DC_VER "3.2.35" +#define DC_VER "3.2.69" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -54,6 +54,10 @@ struct dc_versions { struct dmcu_version dmcu_version; }; +enum dp_protocol_version { + DP_VERSION_1_4, +}; + enum dc_plane_type { DC_PLANE_TYPE_INVALID, DC_PLANE_TYPE_DCE_RGB, @@ -111,18 +115,18 @@ struct dc_caps { bool force_dp_tps4_for_cp2520; bool disable_dp_clk_share; bool psp_setup_panel_mode; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 + bool extended_aux_timeout_support; + bool dmcub_support; bool hw_3d_lut; -#endif + enum dp_protocol_version max_dp_protocol_version; struct dc_plane_cap planes[MAX_PLANES]; }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dc_bug_wa { bool no_connect_phy_config; bool dedcn20_305_wa; + bool skip_clock_update; }; -#endif struct dc_dcc_surface_param { struct dc_size surface_size; @@ -153,11 +157,14 @@ struct dc_surface_dcc_cap { bool const_color_support; }; -struct dc_static_screen_events { - bool force_trigger; - bool cursor_update; - bool surface_update; - bool overlay_update; +struct dc_static_screen_params { + struct { + bool force_trigger; + bool cursor_update; + bool surface_update; + bool overlay_update; + } triggers; + unsigned int num_frames; }; @@ -218,14 +225,17 @@ struct dc_config { bool allow_seamless_boot_optimization; bool power_down_display_on_boot; bool edp_not_connected; + bool force_enum_edp; bool forced_clocks; - + bool disable_extended_timeout_support; // Used to disable extended timeout and lttpr feature as well + bool multi_mon_pp_mclk_switch; }; enum visual_confirm { VISUAL_CONFIRM_DISABLE = 0, VISUAL_CONFIRM_SURFACE = 1, VISUAL_CONFIRM_HDR = 2, + VISUAL_CONFIRM_MPCTREE = 4, }; enum dcc_option { @@ -244,6 +254,19 @@ enum wm_report_mode { WM_REPORT_DEFAULT = 0, WM_REPORT_OVERRIDE = 1, }; +enum dtm_pstate{ + dtm_level_p0 = 0,/*highest voltage*/ + dtm_level_p1, + dtm_level_p2, + dtm_level_p3, + dtm_level_p4,/*when active_display_count = 0*/ +}; + +enum dcn_pwr_state { + DCN_PWR_STATE_UNKNOWN = -1, + DCN_PWR_STATE_MISSION_MODE = 0, + DCN_PWR_STATE_LOW_POWER = 3, +}; /* * For any clocks that may differ per pipe @@ -251,7 +274,6 @@ enum wm_report_mode { */ struct dc_clocks { int dispclk_khz; - int max_supported_dppclk_khz; int dppclk_khz; int dcfclk_khz; int socclk_khz; @@ -260,6 +282,17 @@ struct dc_clocks { int phyclk_khz; int dramclk_khz; bool p_state_change_support; + enum dcn_pwr_state pwr_state; + /* + * Elements below are not compared for the purposes of + * optimization required + */ + bool prev_p_state_change_support; + enum dtm_pstate dtm_level; + int max_supported_dppclk_khz; + int max_supported_dispclk_khz; + int bw_dppclk_khz; /*a copy of dppclk_khz*/ + int bw_dispclk_khz; }; struct dc_bw_validation_profile { @@ -335,12 +368,14 @@ struct dc_debug_options { bool disable_dfs_bypass; bool disable_dpp_power_gate; bool disable_hubp_power_gate; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT bool disable_dsc_power_gate; -#endif + int dsc_min_slice_height_override; + int dsc_bpp_increment_div; + bool native422_support; bool disable_pplib_wm_range; enum wm_report_mode pplib_wm_report_mode; unsigned int min_disp_clk_khz; + unsigned int min_dpp_clk_khz; int sr_exit_time_dpm0_ns; int sr_enter_plus_exit_time_dpm0_ns; int sr_exit_time_ns; @@ -367,17 +402,30 @@ struct dc_debug_options { bool scl_reset_length10; bool hdmi20_disable; bool skip_detection_link_training; + bool remove_disconnect_edp; unsigned int force_odm_combine; //bit vector based on otg inst unsigned int force_fclk_khz; bool disable_tri_buf; + bool dmub_offload_enabled; + bool dmcub_emulation; + bool dmub_command_table; /* for testing only */ struct dc_bw_validation_profile bw_val_profile; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT bool disable_fec; -#endif + bool disable_48mhz_pwrdwn; /* This forces a hard min on the DCFCLK requested to SMU/PP * watermarks are not affected. */ unsigned int force_min_dcfclk_mhz; + bool disable_timing_sync; + bool cm_in_bypass; + int force_clock_mode;/*every mode change.*/ + + bool nv12_iflip_vm_wa; + bool disable_dram_clock_change_vactive_support; + bool validate_dml_output; + bool enable_dmcub_surface_flip; + bool usbc_combo_phy_reset_wa; + bool disable_dsc; }; struct dc_debug_data { @@ -386,7 +434,6 @@ struct dc_debug_data { uint32_t auxErrorCount; }; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 struct dc_phy_addr_space_config { struct { uint64_t start_addr; @@ -406,6 +453,7 @@ struct dc_phy_addr_space_config { } gart_config; bool valid; + uint64_t page_table_default_page_addr; }; struct dc_virtual_addr_space_config { @@ -415,7 +463,6 @@ struct dc_virtual_addr_space_config { uint32_t page_table_block_size_in_bytes; uint8_t page_table_depth; // 1 = 1 level, 2 = 2 level, etc. 0 = invalid }; -#endif struct dc_bounding_box_overrides { int sr_exit_time_ns; @@ -441,13 +488,9 @@ struct dc { struct dc_config config; struct dc_debug_options debug; struct dc_bounding_box_overrides bb_overrides; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dc_bug_wa work_arounds; -#endif struct dc_context *ctx; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 struct dc_phy_addr_space_config vm_pa_config; -#endif uint8_t link_count; struct dc_link *links[MAX_PIPES * 2]; @@ -463,7 +506,7 @@ struct dc { /* Inputs into BW and WM calculations. */ struct bw_calcs_dceip *bw_dceip; struct bw_calcs_vbios *bw_vbios; -#ifdef CONFIG_DRM_AMD_DC_DCN1_0 +#ifdef CONFIG_DRM_AMD_DC_DCN struct dcn_soc_bounding_box *dcn_soc; struct dcn_ip_params *dcn_ip; struct display_mode_lib dml; @@ -477,7 +520,7 @@ struct dc { bool optimized_required; /* Require to maintain clocks and bandwidth for UEFI enabled HW */ - bool optimize_seamless_boot; + int optimize_seamless_boot_streams; /* FBC compressor */ struct compressor *fbc_compressor; @@ -485,10 +528,8 @@ struct dc { struct dc_debug_data debug_data; const char *build_id; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 struct vm_helper *vm_helper; const struct gpu_info_soc_bounding_box_v1_0 *soc_bounding_box; -#endif }; enum frame_buffer_mode { @@ -520,30 +561,36 @@ struct dc_init_data { struct dc_bios *vbios_override; enum dce_environment dce_environment; + struct dmub_offload_funcs *dmub_if; + struct dc_reg_helper_state *dmub_offload; + struct dc_config flags; uint32_t log_mask; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 /** * gpu_info FW provided soc bounding box struct or 0 if not * available in FW */ const struct gpu_info_soc_bounding_box_v1_0 *soc_bounding_box; -#endif }; struct dc_callback_init { +#ifdef CONFIG_DRM_AMD_DC_HDCP + struct cp_psp cp_psp; +#else uint8_t reserved; +#endif }; struct dc *dc_create(const struct dc_init_data *init_params); +void dc_hardware_init(struct dc *dc); + int dc_get_vmid_use_vector(struct dc *dc); -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 void dc_setup_vm_context(struct dc *dc, struct dc_virtual_addr_space_config *va_config, int vmid); /* Returns the number of vmids supported */ int dc_setup_system_context(struct dc *dc, struct dc_phy_addr_space_config *pa_config); -#endif void dc_init_callbacks(struct dc *dc, const struct dc_callback_init *init_params); +void dc_deinit_callbacks(struct dc *dc); void dc_destroy(struct dc **dc); /******************************************************************************* @@ -597,9 +644,12 @@ enum dc_transfer_func_predefined { TRANSFER_FUNCTION_UNITY, TRANSFER_FUNCTION_HLG, TRANSFER_FUNCTION_HLG12, - TRANSFER_FUNCTION_GAMMA22 + TRANSFER_FUNCTION_GAMMA22, + TRANSFER_FUNCTION_GAMMA24, + TRANSFER_FUNCTION_GAMMA26 }; + struct dc_transfer_func { struct kref refcount; enum dc_transfer_func_type type; @@ -613,17 +663,29 @@ struct dc_transfer_func { }; }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) + +union dc_3dlut_state { + struct { + uint32_t initialized:1; /*if 3dlut is went through color module for initialization */ + uint32_t rmu_idx_valid:1; /*if mux settings are valid*/ + uint32_t rmu_mux_num:3; /*index of mux to use*/ + uint32_t mpc_rmu0_mux:4; /*select mpcc on mux, one of the following : mpcc0, mpcc1, mpcc2, mpcc3*/ + uint32_t mpc_rmu1_mux:4; + uint32_t mpc_rmu2_mux:4; + uint32_t reserved:15; + } bits; + uint32_t raw; +}; struct dc_3dlut { struct kref refcount; struct tetrahedral_params lut_3d; - uint32_t hdr_multiplier; - bool initialized; + struct fixed31_32 hdr_multiplier; + bool initialized; /*remove after diag fix*/ + union dc_3dlut_state state; struct dc_context *ctx; }; -#endif /* * This structure is filled in by dc_surface_get_status and contains * the last requested address and the currently active address so the called @@ -646,7 +708,7 @@ union surface_update_flags { uint32_t horizontal_mirror_change:1; uint32_t per_pixel_alpha_change:1; uint32_t global_alpha_change:1; - uint32_t sdr_white_level:1; + uint32_t hdr_mult:1; uint32_t rotation_change:1; uint32_t swizzle_change:1; uint32_t scaling_change:1; @@ -674,15 +736,13 @@ union surface_update_flags { struct dc_plane_state { struct dc_plane_address address; struct dc_plane_flip_time time; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) bool triplebuffer_flips; -#endif struct scaling_taps scaling_quality; struct rect src_rect; struct rect dst_rect; struct rect clip_rect; - union plane_size plane_size; + struct plane_size plane_size; union dc_tiling_info tiling_info; struct dc_plane_dcc_param dcc; @@ -692,18 +752,16 @@ struct dc_plane_state { struct dc_bias_and_scale *bias_and_scale; struct dc_csc_transform input_csc_color_matrix; struct fixed31_32 coeff_reduction_factor; - uint32_t sdr_white_level; + struct fixed31_32 hdr_mult; // TODO: No longer used, remove struct dc_hdr_static_metadata hdr_static_ctx; enum dc_color_space color_space; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dc_3dlut *lut3d_func; struct dc_transfer_func *in_shaper_func; struct dc_transfer_func *blend_tf; -#endif enum surface_pixel_format format; enum dc_rotation_angle rotation; @@ -716,6 +774,7 @@ struct dc_plane_state { bool visible; bool flip_immediate; bool horizontal_mirror; + int layer_index; union surface_update_flags update_flags; /* private to DC core */ @@ -731,20 +790,20 @@ struct dc_plane_state { }; struct dc_plane_info { - union plane_size plane_size; + struct plane_size plane_size; union dc_tiling_info tiling_info; struct dc_plane_dcc_param dcc; enum surface_pixel_format format; enum dc_rotation_angle rotation; enum plane_stereo_format stereo_format; enum dc_color_space color_space; - unsigned int sdr_white_level; bool horizontal_mirror; bool visible; bool per_pixel_alpha; bool global_alpha; int global_alpha_value; bool input_csc_enabled; + int layer_index; }; struct dc_scaling_info { @@ -761,7 +820,7 @@ struct dc_surface_update { const struct dc_flip_addrs *flip_addr; const struct dc_plane_info *plane_info; const struct dc_scaling_info *scaling_info; - + struct fixed31_32 hdr_mult; /* following updates require alloc/sleep/spin that is not isr safe, * null means no updates */ @@ -770,11 +829,9 @@ struct dc_surface_update { const struct dc_csc_transform *input_csc_color_matrix; const struct fixed31_32 *coeff_reduction_factor; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) const struct dc_transfer_func *func_shaper; const struct dc_3dlut *lut3d_func; const struct dc_transfer_func *blend_tf; -#endif }; /* @@ -795,11 +852,9 @@ void dc_transfer_func_retain(struct dc_transfer_func *dc_tf); void dc_transfer_func_release(struct dc_transfer_func *dc_tf); struct dc_transfer_func *dc_create_transfer_func(void); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dc_3dlut *dc_create_3dlut_func(void); void dc_3dlut_func_release(struct dc_3dlut *lut); void dc_3dlut_func_retain(struct dc_3dlut *lut); -#endif /* * This structure holds a surface address. There could be multiple addresses * in cases such as Stereo 3D, Planar YUV, etc. Other per-flip attributes such @@ -834,6 +889,9 @@ enum dc_status dc_validate_plane(struct dc *dc, const struct dc_plane_state *pla void get_clock_requirements_for_state(struct dc_state *state, struct AsicStateEx *info); +bool dc_set_generic_gpio_for_stereo(bool enable, + struct gpio_service *gpio_service); + /* * fast_validate: we return after determining if we can support the new state, * but before we populate the programming info @@ -858,6 +916,8 @@ void dc_resource_state_copy_construct_current( void dc_resource_state_destruct(struct dc_state *context); +bool dc_resource_is_dsc_encoding_supported(const struct dc *dc); + /* * TODO update to make it about validation sets * Set up streams and links associated to drive sinks @@ -913,10 +973,10 @@ struct dpcd_caps { bool panel_mode_edp; bool dpcd_display_control_capable; bool ext_receiver_cap_field_present; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT union dpcd_fec_capability fec_cap; struct dpcd_dsc_capabilities dsc_caps; -#endif + struct dc_lttpr_caps lttpr_caps; + }; #include "dc_link.h" @@ -937,14 +997,12 @@ struct dc_container_id { }; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct dc_sink_dsc_caps { // 'true' if these are virtual DPCD's DSC caps (immediately upstream of sink in MST topology), // 'false' if they are sink's DSC caps bool is_virtual_dpcd_dsc; struct dsc_dec_dpcd_caps dsc_dec_caps; }; -#endif /* * The sink structure contains EDID and other display device properties @@ -959,9 +1017,7 @@ struct dc_sink { struct stereo_3d_features features_3d[TIMING_3D_FORMAT_MAX]; bool converter_disable_audio; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct dc_sink_dsc_caps sink_dsc_caps; -#endif /* private to DC core */ struct dc_link *link; @@ -1019,11 +1075,12 @@ unsigned int dc_get_current_backlight_pwm(struct dc *dc); unsigned int dc_get_target_backlight_pwm(struct dc *dc); bool dc_is_dmcu_initialized(struct dc *dc); +bool dc_is_hw_initialized(struct dc *dc); -#if defined(CONFIG_DRM_AMD_DC_DSC_SUPPORT) +enum dc_status dc_set_clock(struct dc *dc, enum dc_clock_type clock_type, uint32_t clk_khz, uint32_t stepping); +void dc_get_clock(struct dc *dc, enum dc_clock_type clock_type, struct dc_clock_config *clock_cfg); /******************************************************************************* * DSC Interfaces ******************************************************************************/ #include "dc_dsc.h" -#endif #endif /* DC_INTERFACE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_bios_types.h b/drivers/gpu/drm/amd/display/dc/dc_bios_types.h index 78c3b300ec45..b1dd0d60d98e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_bios_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_bios_types.h @@ -61,9 +61,6 @@ struct dc_vbios_funcs { struct graphics_object_id connector_object_id, uint32_t device_tag_index, struct connector_device_tag_info *info); - enum bp_result (*get_firmware_info)( - struct dc_bios *bios, - struct dc_firmware_info *info); enum bp_result (*get_spread_spectrum_info)( struct dc_bios *bios, enum as_signal_type signal, @@ -152,6 +149,8 @@ struct dc_bios { struct dc_context *ctx; const struct bios_registers *regs; struct integrated_info *integrated_info; + struct dc_firmware_info fw_info; + bool fw_info_valid; }; #endif /* DC_BIOS_TYPES_H */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_ddc_types.h b/drivers/gpu/drm/amd/display/dc/dc_ddc_types.h index 4ef97f65e55d..4f8f576d5fcf 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_ddc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_ddc_types.h @@ -49,7 +49,8 @@ enum aux_channel_operation_result { AUX_CHANNEL_OPERATION_FAILED_REASON_UNKNOWN, AUX_CHANNEL_OPERATION_FAILED_INVALID_REPLY, AUX_CHANNEL_OPERATION_FAILED_TIMEOUT, - AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON + AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON, + AUX_CHANNEL_OPERATION_FAILED_ENGINE_ACQUIRE }; diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c new file mode 100644 index 000000000000..59c298a6484f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -0,0 +1,134 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dc.h" +#include "dc_dmub_srv.h" +#include "../dmub/inc/dmub_srv.h" + +static void dc_dmub_srv_construct(struct dc_dmub_srv *dc_srv, struct dc *dc, + struct dmub_srv *dmub) +{ + dc_srv->dmub = dmub; + dc_srv->ctx = dc->ctx; +} + +struct dc_dmub_srv *dc_dmub_srv_create(struct dc *dc, struct dmub_srv *dmub) +{ + struct dc_dmub_srv *dc_srv = + kzalloc(sizeof(struct dc_dmub_srv), GFP_KERNEL); + + if (dc_srv == NULL) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dc_dmub_srv_construct(dc_srv, dc, dmub); + + return dc_srv; +} + +void dc_dmub_srv_destroy(struct dc_dmub_srv **dmub_srv) +{ + if (*dmub_srv) { + kfree(*dmub_srv); + *dmub_srv = NULL; + } +} + +void dc_dmub_srv_cmd_queue(struct dc_dmub_srv *dc_dmub_srv, + struct dmub_cmd_header *cmd) +{ + struct dmub_srv *dmub = dc_dmub_srv->dmub; + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + enum dmub_status status; + + status = dmub_srv_cmd_queue(dmub, cmd); + if (status == DMUB_STATUS_OK) + return; + + if (status != DMUB_STATUS_QUEUE_FULL) + goto error; + + /* Execute and wait for queue to become empty again. */ + dc_dmub_srv_cmd_execute(dc_dmub_srv); + dc_dmub_srv_wait_idle(dc_dmub_srv); + + /* Requeue the command. */ + status = dmub_srv_cmd_queue(dmub, cmd); + if (status == DMUB_STATUS_OK) + return; + +error: + DC_ERROR("Error queuing DMUB command: status=%d\n", status); +} + +void dc_dmub_srv_cmd_execute(struct dc_dmub_srv *dc_dmub_srv) +{ + struct dmub_srv *dmub = dc_dmub_srv->dmub; + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + enum dmub_status status; + + status = dmub_srv_cmd_execute(dmub); + if (status != DMUB_STATUS_OK) + DC_ERROR("Error starting DMUB execution: status=%d\n", status); +} + +void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv) +{ + struct dmub_srv *dmub = dc_dmub_srv->dmub; + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + enum dmub_status status; + + status = dmub_srv_wait_for_idle(dmub, 100000); + if (status != DMUB_STATUS_OK) + DC_ERROR("Error waiting for DMUB idle: status=%d\n", status); +} + +void dc_dmub_srv_wait_phy_init(struct dc_dmub_srv *dc_dmub_srv) +{ + struct dmub_srv *dmub = dc_dmub_srv->dmub; + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + enum dmub_status status; + + for (;;) { + /* Wait up to a second for PHY init. */ + status = dmub_srv_wait_for_phy_init(dmub, 1000000); + if (status == DMUB_STATUS_OK) + /* Initialization OK */ + break; + + DC_ERROR("DMCUB PHY init failed: status=%d\n", status); + ASSERT(0); + + if (status != DMUB_STATUS_TIMEOUT) + /* + * Server likely initialized or we don't have + * DMCUB HW support - this won't end. + */ + break; + + /* Continue spinning so we don't hang the ASIC. */ + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h new file mode 100644 index 000000000000..754b6077539c --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h @@ -0,0 +1,60 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_DC_SRV_H_ +#define _DMUB_DC_SRV_H_ + +#include "os_types.h" +#include "../dmub/inc/dmub_cmd.h" + +struct dmub_srv; +struct dmub_cmd_header; + +struct dc_reg_helper_state { + bool gather_in_progress; + uint32_t same_addr_count; + bool should_burst_write; + union dmub_rb_cmd cmd_data; + unsigned int reg_seq_count; +}; + +struct dc_dmub_srv { + struct dmub_srv *dmub; + struct dc_reg_helper_state reg_helper_offload; + + struct dc_context *ctx; + void *dm; +}; + +void dc_dmub_srv_cmd_queue(struct dc_dmub_srv *dc_dmub_srv, + struct dmub_cmd_header *cmd); + +void dc_dmub_srv_cmd_execute(struct dc_dmub_srv *dc_dmub_srv); + +void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv); + +void dc_dmub_srv_wait_phy_init(struct dc_dmub_srv *dc_dmub_srv); + +#endif /* _DMUB_DC_SRV_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index dfcec4d3e9c0..dfe4472c9e40 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -90,6 +90,13 @@ enum dc_post_cursor2 { POST_CURSOR2_MAX_LEVEL = POST_CURSOR2_LEVEL3, }; +enum dc_dp_training_pattern { + DP_TRAINING_PATTERN_SEQUENCE_1 = 0, + DP_TRAINING_PATTERN_SEQUENCE_2, + DP_TRAINING_PATTERN_SEQUENCE_3, + DP_TRAINING_PATTERN_SEQUENCE_4, +}; + struct dc_link_settings { enum dc_lane_count lane_count; enum dc_link_rate link_rate; @@ -109,6 +116,21 @@ struct dc_link_training_settings { struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]; }; +struct dc_link_training_overrides { + enum dc_voltage_swing *voltage_swing; + enum dc_pre_emphasis *pre_emphasis; + enum dc_post_cursor2 *post_cursor2; + + uint16_t *cr_pattern_time; + uint16_t *eq_pattern_time; + enum dc_dp_training_pattern *pattern_for_eq; + + enum dc_link_spread *downspread; + bool *alternate_scrambler_reset; + bool *enhanced_framing; + bool *mst_enable; + bool *fec_enable; +}; union dpcd_rev { struct { @@ -447,13 +469,13 @@ union training_aux_rd_interval { /* Automated test structures */ union test_request { struct { - uint8_t LINK_TRAINING :1; - uint8_t LINK_TEST_PATTRN :1; - uint8_t EDID_READ :1; - uint8_t PHY_TEST_PATTERN :1; - uint8_t AUDIO_TEST_PATTERN :1; - uint8_t RESERVED :1; - uint8_t TEST_STEREO_3D :1; + uint8_t LINK_TRAINING :1; + uint8_t LINK_TEST_PATTRN :1; + uint8_t EDID_READ :1; + uint8_t PHY_TEST_PATTERN :1; + uint8_t RESERVED :1; + uint8_t AUDIO_TEST_PATTERN :1; + uint8_t TEST_AUDIO_DISABLED_VIDEO :1; } bits; uint8_t raw; }; @@ -500,19 +522,52 @@ union link_test_pattern { union test_misc { struct dpcd_test_misc_bits { - unsigned char SYNC_CLOCK :1; + unsigned char SYNC_CLOCK :1; /* dpcd_test_color_format */ - unsigned char CLR_FORMAT :2; + unsigned char CLR_FORMAT :2; /* dpcd_test_dyn_range */ - unsigned char DYN_RANGE :1; - unsigned char YCBCR :1; + unsigned char DYN_RANGE :1; + unsigned char YCBCR_COEFS :1; /* dpcd_test_bit_depth */ - unsigned char BPC :3; + unsigned char BPC :3; + } bits; + unsigned char raw; +}; + +union audio_test_mode { + struct { + unsigned char sampling_rate :4; + unsigned char channel_count :4; + } bits; + unsigned char raw; +}; + +union audio_test_pattern_period { + struct { + unsigned char pattern_period :4; + unsigned char reserved :4; } bits; unsigned char raw; }; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT +struct audio_test_pattern_type { + unsigned char value; +}; + +struct dp_audio_test_data_flags { + uint8_t test_requested :1; + uint8_t disable_video :1; +}; + +struct dp_audio_test_data { + + struct dp_audio_test_data_flags flags; + uint8_t sampling_rate; + uint8_t channel_count; + uint8_t pattern_type; + uint8_t pattern_period[8]; +}; + /* FEC capability DPCD register field bits-*/ union dpcd_fec_capability { struct { @@ -637,6 +692,5 @@ struct dpcd_dsc_capabilities { union dpcd_dsc_ext_capabilities dsc_ext_caps; }; -#endif /* CONFIG_DRM_AMD_DC_DSC_SUPPORT */ #endif /* DC_DP_TYPES_H */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dsc.h b/drivers/gpu/drm/amd/display/dc/dc_dsc.h index 6e42209f0e20..3800340a5b4f 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dsc.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dsc.h @@ -1,4 +1,3 @@ -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #ifndef DC_DSC_H_ #define DC_DSC_H_ /* @@ -30,6 +29,7 @@ #define DP_DSC_BRANCH_OVERALL_THROUGHPUT_0 0x0a0 /* DP 1.4a SCR */ #define DP_DSC_BRANCH_OVERALL_THROUGHPUT_1 0x0a1 #define DP_DSC_BRANCH_MAX_LINE_WIDTH 0x0a2 +#include "dc_types.h" struct dc_dsc_bw_range { uint32_t min_kbps; /* Bandwidth if min_target_bpp_x16 is used */ @@ -39,24 +39,45 @@ struct dc_dsc_bw_range { uint32_t stream_kbps; /* Uncompressed stream bandwidth */ }; +struct display_stream_compressor { + const struct dsc_funcs *funcs; + struct dc_context *ctx; + int inst; +}; + +struct dc_dsc_policy { + bool use_min_slices_h; + int max_slices_h; // Maximum available if 0 + int min_slice_height; // Must not be less than 8 + uint32_t max_target_bpp; + uint32_t min_target_bpp; +}; -bool dc_dsc_parse_dsc_dpcd(const uint8_t *dpcd_dsc_basic_data, +bool dc_dsc_parse_dsc_dpcd(const struct dc *dc, + const uint8_t *dpcd_dsc_basic_data, const uint8_t *dpcd_dsc_ext_data, struct dsc_dec_dpcd_caps *dsc_sink_caps); bool dc_dsc_compute_bandwidth_range( - const struct dc *dc, - const uint32_t min_kbps, - const uint32_t max_kbps, + const struct display_stream_compressor *dsc, + const uint32_t dsc_min_slice_height_override, + const uint32_t min_bpp, + const uint32_t max_bpp, const struct dsc_dec_dpcd_caps *dsc_sink_caps, const struct dc_crtc_timing *timing, struct dc_dsc_bw_range *range); bool dc_dsc_compute_config( - const struct dc *dc, + const struct display_stream_compressor *dsc, const struct dsc_dec_dpcd_caps *dsc_sink_caps, + const uint32_t dsc_min_slice_height_override, uint32_t target_bandwidth_kbps, const struct dc_crtc_timing *timing, struct dc_dsc_config *dsc_cfg); -#endif + +void dc_dsc_get_policy_for_timing(const struct dc_crtc_timing *timing, + struct dc_dsc_policy *policy); + +void dc_dsc_policy_set_max_target_bpp_limit(uint32_t limit); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c index 30b2f9edd42f..737048d8a96c 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c @@ -32,6 +32,74 @@ #include "dm_services.h" #include <stdarg.h> +#include "dc.h" +#include "dc_dmub_srv.h" + +static inline void submit_dmub_read_modify_write( + struct dc_reg_helper_state *offload, + const struct dc_context *ctx) +{ + struct dmub_rb_cmd_read_modify_write *cmd_buf = &offload->cmd_data.read_modify_write; + bool gather = false; + + offload->should_burst_write = + (offload->same_addr_count == (DMUB_READ_MODIFY_WRITE_SEQ__MAX - 1)); + cmd_buf->header.payload_bytes = + sizeof(struct dmub_cmd_read_modify_write_sequence) * offload->reg_seq_count; + + gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress; + ctx->dmub_srv->reg_helper_offload.gather_in_progress = false; + + dc_dmub_srv_cmd_queue(ctx->dmub_srv, &cmd_buf->header); + + ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather; + + memset(cmd_buf, 0, sizeof(*cmd_buf)); + + offload->reg_seq_count = 0; + offload->same_addr_count = 0; +} + +static inline void submit_dmub_burst_write( + struct dc_reg_helper_state *offload, + const struct dc_context *ctx) +{ + struct dmub_rb_cmd_burst_write *cmd_buf = &offload->cmd_data.burst_write; + bool gather = false; + + cmd_buf->header.payload_bytes = + sizeof(uint32_t) * offload->reg_seq_count; + + gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress; + ctx->dmub_srv->reg_helper_offload.gather_in_progress = false; + + dc_dmub_srv_cmd_queue(ctx->dmub_srv, &cmd_buf->header); + + ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather; + + memset(cmd_buf, 0, sizeof(*cmd_buf)); + + offload->reg_seq_count = 0; +} + +static inline void submit_dmub_reg_wait( + struct dc_reg_helper_state *offload, + const struct dc_context *ctx) +{ + struct dmub_rb_cmd_reg_wait *cmd_buf = &offload->cmd_data.reg_wait; + bool gather = false; + + gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress; + ctx->dmub_srv->reg_helper_offload.gather_in_progress = false; + + dc_dmub_srv_cmd_queue(ctx->dmub_srv, &cmd_buf->header); + + memset(cmd_buf, 0, sizeof(*cmd_buf)); + offload->reg_seq_count = 0; + + ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather; +} + struct dc_reg_value_masks { uint32_t value; uint32_t mask; @@ -77,6 +145,100 @@ static void set_reg_field_values(struct dc_reg_value_masks *field_value_mask, } } +static void dmub_flush_buffer_execute( + struct dc_reg_helper_state *offload, + const struct dc_context *ctx) +{ + submit_dmub_read_modify_write(offload, ctx); + dc_dmub_srv_cmd_execute(ctx->dmub_srv); +} + +static void dmub_flush_burst_write_buffer_execute( + struct dc_reg_helper_state *offload, + const struct dc_context *ctx) +{ + submit_dmub_burst_write(offload, ctx); + dc_dmub_srv_cmd_execute(ctx->dmub_srv); +} + +static bool dmub_reg_value_burst_set_pack(const struct dc_context *ctx, uint32_t addr, + uint32_t reg_val) +{ + struct dc_reg_helper_state *offload = &ctx->dmub_srv->reg_helper_offload; + struct dmub_rb_cmd_burst_write *cmd_buf = &offload->cmd_data.burst_write; + + /* flush command if buffer is full */ + if (offload->reg_seq_count == DMUB_BURST_WRITE_VALUES__MAX) + dmub_flush_burst_write_buffer_execute(offload, ctx); + + if (offload->cmd_data.cmd_common.header.type == DMUB_CMD__REG_SEQ_BURST_WRITE && + addr != cmd_buf->addr) { + dmub_flush_burst_write_buffer_execute(offload, ctx); + return false; + } + + cmd_buf->header.type = DMUB_CMD__REG_SEQ_BURST_WRITE; + cmd_buf->header.sub_type = 0; + cmd_buf->addr = addr; + cmd_buf->write_values[offload->reg_seq_count] = reg_val; + offload->reg_seq_count++; + + return true; +} + +static uint32_t dmub_reg_value_pack(const struct dc_context *ctx, uint32_t addr, + struct dc_reg_value_masks *field_value_mask) +{ + struct dc_reg_helper_state *offload = &ctx->dmub_srv->reg_helper_offload; + struct dmub_rb_cmd_read_modify_write *cmd_buf = &offload->cmd_data.read_modify_write; + struct dmub_cmd_read_modify_write_sequence *seq; + + /* flush command if buffer is full */ + if (offload->cmd_data.cmd_common.header.type != DMUB_CMD__REG_SEQ_BURST_WRITE && + offload->reg_seq_count == DMUB_READ_MODIFY_WRITE_SEQ__MAX) + dmub_flush_buffer_execute(offload, ctx); + + if (offload->should_burst_write) { + if (dmub_reg_value_burst_set_pack(ctx, addr, field_value_mask->value)) + return field_value_mask->value; + else + offload->should_burst_write = false; + } + + /* pack commands */ + cmd_buf->header.type = DMUB_CMD__REG_SEQ_READ_MODIFY_WRITE; + cmd_buf->header.sub_type = 0; + seq = &cmd_buf->seq[offload->reg_seq_count]; + + if (offload->reg_seq_count) { + if (cmd_buf->seq[offload->reg_seq_count - 1].addr == addr) + offload->same_addr_count++; + else + offload->same_addr_count = 0; + } + + seq->addr = addr; + seq->modify_mask = field_value_mask->mask; + seq->modify_value = field_value_mask->value; + offload->reg_seq_count++; + + return field_value_mask->value; +} + +static void dmub_reg_wait_done_pack(const struct dc_context *ctx, uint32_t addr, + uint32_t mask, uint32_t shift, uint32_t condition_value, uint32_t time_out_us) +{ + struct dc_reg_helper_state *offload = &ctx->dmub_srv->reg_helper_offload; + struct dmub_rb_cmd_reg_wait *cmd_buf = &offload->cmd_data.reg_wait; + + cmd_buf->header.type = DMUB_CMD__REG_REG_WAIT; + cmd_buf->header.sub_type = 0; + cmd_buf->reg_wait.addr = addr; + cmd_buf->reg_wait.condition_field_value = mask & (condition_value << shift); + cmd_buf->reg_wait.mask = mask; + cmd_buf->reg_wait.time_out_us = time_out_us; +} + uint32_t generic_reg_update_ex(const struct dc_context *ctx, uint32_t addr, int n, uint8_t shift1, uint32_t mask1, uint32_t field_value1, @@ -93,6 +255,11 @@ uint32_t generic_reg_update_ex(const struct dc_context *ctx, va_end(ap); + if (ctx->dmub_srv && + ctx->dmub_srv->reg_helper_offload.gather_in_progress) + return dmub_reg_value_pack(ctx, addr, &field_value_mask); + /* todo: return void so we can decouple code running in driver from register states */ + /* mmio write directly */ reg_val = dm_read_reg(ctx, addr); reg_val = (reg_val & ~field_value_mask.mask) | field_value_mask.value; @@ -118,6 +285,13 @@ uint32_t generic_reg_set_ex(const struct dc_context *ctx, /* mmio write directly */ reg_val = (reg_val & ~field_value_mask.mask) | field_value_mask.value; + + if (ctx->dmub_srv && + ctx->dmub_srv->reg_helper_offload.gather_in_progress) { + return dmub_reg_value_burst_set_pack(ctx, addr, reg_val); + /* todo: return void so we can decouple code running in driver from register states */ + } + dm_write_reg(ctx, addr, reg_val); return reg_val; } @@ -134,6 +308,14 @@ uint32_t dm_read_reg_func( return 0; } #endif + + if (ctx->dmub_srv && + ctx->dmub_srv->reg_helper_offload.gather_in_progress && + !ctx->dmub_srv->reg_helper_offload.should_burst_write) { + ASSERT(false); + return 0; + } + value = cgs_read_register(ctx->cgs_device, address); trace_amdgpu_dc_rreg(&ctx->perf_trace->read_count, address, value); @@ -299,7 +481,19 @@ void generic_reg_wait(const struct dc_context *ctx, uint32_t reg_val; int i; - /* something is terribly wrong if time out is > 200ms. (5Hz) */ + if (ctx->dmub_srv && + ctx->dmub_srv->reg_helper_offload.gather_in_progress) { + dmub_reg_wait_done_pack(ctx, addr, mask, shift, condition_value, + delay_between_poll_us * time_out_num_tries); + return; + } + + /* + * Something is terribly wrong if time out is > 3000ms. + * 3000ms is the maximum time needed for SMU to pass values back. + * This value comes from experiments. + * + */ ASSERT(delay_between_poll_us * time_out_num_tries <= 3000000); for (i = 0; i <= time_out_num_tries; i++) { @@ -346,12 +540,48 @@ uint32_t generic_read_indirect_reg(const struct dc_context *ctx, { uint32_t value = 0; + // when reg read, there should not be any offload. + if (ctx->dmub_srv && + ctx->dmub_srv->reg_helper_offload.gather_in_progress) { + ASSERT(false); + } + dm_write_reg(ctx, addr_index, index); value = dm_read_reg(ctx, addr_data); return value; } +uint32_t generic_indirect_reg_get(const struct dc_context *ctx, + uint32_t addr_index, uint32_t addr_data, + uint32_t index, int n, + uint8_t shift1, uint32_t mask1, uint32_t *field_value1, + ...) +{ + uint32_t shift, mask, *field_value; + uint32_t value = 0; + int i = 1; + + va_list ap; + + va_start(ap, field_value1); + + value = generic_read_indirect_reg(ctx, addr_index, addr_data, index); + *field_value1 = get_reg_field_value_ex(value, mask1, shift1); + + while (i < n) { + shift = va_arg(ap, uint32_t); + mask = va_arg(ap, uint32_t); + field_value = va_arg(ap, uint32_t *); + + *field_value = get_reg_field_value_ex(value, mask, shift); + i++; + } + + va_end(ap); + + return value; +} uint32_t generic_indirect_reg_update_ex(const struct dc_context *ctx, uint32_t addr_index, uint32_t addr_data, @@ -382,3 +612,68 @@ uint32_t generic_indirect_reg_update_ex(const struct dc_context *ctx, return reg_val; } + +void reg_sequence_start_gather(const struct dc_context *ctx) +{ + /* if reg sequence is supported and enabled, set flag to + * indicate we want to have REG_SET, REG_UPDATE macro build + * reg sequence command buffer rather than MMIO directly. + */ + + if (ctx->dmub_srv && ctx->dc->debug.dmub_offload_enabled) { + struct dc_reg_helper_state *offload = + &ctx->dmub_srv->reg_helper_offload; + + /* caller sequence mismatch. need to debug caller. offload will not work!!! */ + ASSERT(!offload->gather_in_progress); + + offload->gather_in_progress = true; + } +} + +void reg_sequence_start_execute(const struct dc_context *ctx) +{ + struct dc_reg_helper_state *offload; + + if (!ctx->dmub_srv) + return; + + offload = &ctx->dmub_srv->reg_helper_offload; + + if (offload && offload->gather_in_progress) { + offload->gather_in_progress = false; + offload->should_burst_write = false; + switch (offload->cmd_data.cmd_common.header.type) { + case DMUB_CMD__REG_SEQ_READ_MODIFY_WRITE: + submit_dmub_read_modify_write(offload, ctx); + break; + case DMUB_CMD__REG_REG_WAIT: + submit_dmub_reg_wait(offload, ctx); + break; + case DMUB_CMD__REG_SEQ_BURST_WRITE: + submit_dmub_burst_write(offload, ctx); + break; + default: + return; + } + + dc_dmub_srv_cmd_execute(ctx->dmub_srv); + } +} + +void reg_sequence_wait_done(const struct dc_context *ctx) +{ + /* callback to DM to poll for last submission done*/ + struct dc_reg_helper_state *offload; + + if (!ctx->dmub_srv) + return; + + offload = &ctx->dmub_srv->reg_helper_offload; + + if (offload && + ctx->dc->debug.dmub_offload_enabled && + !ctx->dc->debug.dmcub_emulation) { + dc_dmub_srv_wait_idle(ctx->dmub_srv); + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h index 22db5682aa6c..25c50bcab9e9 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -115,56 +115,25 @@ struct rect { int height; }; -union plane_size { - /* Grph or Video will be selected - * based on format above: - * Use Video structure if - * format >= DalPixelFormat_VideoBegin - * else use Grph structure +struct plane_size { + /* Graphic surface pitch in pixels. + * In LINEAR_GENERAL mode, pitch + * is 32 pixel aligned. */ - struct { - struct rect surface_size; - /* Graphic surface pitch in pixels. - * In LINEAR_GENERAL mode, pitch - * is 32 pixel aligned. - */ - int surface_pitch; - } grph; - - struct { - struct rect luma_size; - /* Graphic surface pitch in pixels. - * In LINEAR_GENERAL mode, pitch is - * 32 pixel aligned. - */ - int luma_pitch; - - struct rect chroma_size; - /* Graphic surface pitch in pixels. - * In LINEAR_GENERAL mode, pitch is - * 32 pixel aligned. - */ - int chroma_pitch; - } video; + int surface_pitch; + int chroma_pitch; + struct rect surface_size; + struct rect chroma_size; }; struct dc_plane_dcc_param { bool enable; - union { - struct { - int meta_pitch; - bool independent_64b_blks; - } grph; - - struct { - int meta_pitch_l; - bool independent_64b_blks_l; + int meta_pitch; + bool independent_64b_blks; - int meta_pitch_c; - bool independent_64b_blks_c; - } video; - }; + int meta_pitch_c; + bool independent_64b_blks_c; }; /*Displayable pixel format in fb*/ @@ -196,12 +165,10 @@ enum surface_pixel_format { /*swaped & float*/ SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F, /*grow graphics here if necessary */ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FIX, SURFACE_PIXEL_FORMAT_GRPH_BGR101111_FIX, SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FLOAT, SURFACE_PIXEL_FORMAT_GRPH_BGR101111_FLOAT, -#endif SURFACE_PIXEL_FORMAT_VIDEO_BEGIN, SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr = SURFACE_PIXEL_FORMAT_VIDEO_BEGIN, @@ -209,10 +176,8 @@ enum surface_pixel_format { SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCbCr, SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb, SURFACE_PIXEL_FORMAT_SUBSAMPLE_END, -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) SURFACE_PIXEL_FORMAT_VIDEO_ACrYCb2101010, SURFACE_PIXEL_FORMAT_VIDEO_CrYCbA1010102, -#endif SURFACE_PIXEL_FORMAT_VIDEO_AYCrCb8888, SURFACE_PIXEL_FORMAT_INVALID @@ -251,12 +216,10 @@ enum tile_split_values { DC_ROTATED_MICRO_TILING = 0x3, }; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 enum tripleBuffer_enable { DC_TRIPLEBUFFER_DISABLE = 0x0, DC_TRIPLEBUFFER_ENABLE = 0x1, }; -#endif /* TODO: These values come from hardware spec. We need to readdress this * if they ever change. @@ -456,13 +419,11 @@ struct dc_csc_transform { bool enable_adjustment; }; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 struct dc_rgb_fixed { struct fixed31_32 red; struct fixed31_32 green; struct fixed31_32 blue; }; -#endif struct dc_gamma { struct kref refcount; @@ -482,7 +443,6 @@ struct dc_gamma { * is_logical_identity indicates the given gamma ramp regardless of type is identity. */ bool is_identity; - bool is_logical_identity; }; /* Used by both ipp amd opp functions*/ @@ -498,10 +458,8 @@ enum dc_cursor_color_format { CURSOR_MODE_COLOR_1BIT_AND, CURSOR_MODE_COLOR_PRE_MULTIPLIED_ALPHA, CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA, -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) CURSOR_MODE_COLOR_64BIT_FP_PRE_MULTIPLIED, CURSOR_MODE_COLOR_64BIT_FP_UN_PRE_MULTIPLIED -#endif }; /* @@ -519,7 +477,8 @@ union dc_cursor_attribute_flags { uint32_t INVERT_PIXEL_DATA:1; uint32_t ZERO_EXPANSION:1; uint32_t MIN_MAX_INVERT:1; - uint32_t RESERVED:25; + uint32_t ENABLE_CURSOR_DEGAMMA:1; + uint32_t RESERVED:24; } bits; uint32_t value; }; @@ -607,6 +566,11 @@ enum dc_quantization_range { QUANTIZATION_RANGE_LIMITED }; +enum dc_dynamic_expansion { + DYN_EXPANSION_AUTO, + DYN_EXPANSION_DISABLE +}; + /* XFM */ /* used in struct dc_plane_state */ @@ -615,6 +579,7 @@ struct scaling_taps { uint32_t h_taps; uint32_t v_taps_c; uint32_t h_taps_c; + bool integer_scaling; }; enum dc_timing_standard { @@ -647,10 +612,8 @@ enum dc_color_depth { COLOR_DEPTH_121212, COLOR_DEPTH_141414, COLOR_DEPTH_161616, -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 COLOR_DEPTH_999, COLOR_DEPTH_111111, -#endif COLOR_DEPTH_COUNT }; @@ -711,9 +674,7 @@ struct dc_crtc_timing_flags { * rates less than or equal to 340Mcsc */ uint32_t LTE_340MCSC_SCRAMBLE:1; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT uint32_t DSC : 1; /* Use DSC with this timing */ -#endif }; enum dc_timing_3d_format { @@ -738,29 +699,6 @@ enum dc_timing_3d_format { TIMING_3D_FORMAT_MAX, }; -enum trigger_delay { - TRIGGER_DELAY_NEXT_PIXEL = 0, - TRIGGER_DELAY_NEXT_LINE, -}; - -enum crtc_event { - CRTC_EVENT_VSYNC_RISING = 0, - CRTC_EVENT_VSYNC_FALLING -}; - -struct crtc_trigger_info { - bool enabled; - struct dc_stream_state *event_source; - enum crtc_event event; - enum trigger_delay delay; -}; - -struct dc_crtc_timing_adjust { - uint32_t v_total_min; - uint32_t v_total_max; -}; - -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct dc_dsc_config { uint32_t num_slices_h; /* Number of DSC slices - horizontal */ uint32_t num_slices_v; /* Number of DSC slices - vertical */ @@ -771,7 +709,6 @@ struct dc_dsc_config { bool ycbcr422_simple; /* Tell DSC engine to convert YCbCr 4:2:2 to 'YCbCr 4:2:2 simple'. */ int32_t rc_buffer_size; /* DSC RC buffer block size in bytes */ }; -#endif struct dc_crtc_timing { uint32_t h_total; uint32_t h_border_left; @@ -798,11 +735,34 @@ struct dc_crtc_timing { enum scanning_type scan_type; struct dc_crtc_timing_flags flags; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct dc_dsc_config dsc_cfg; -#endif }; +enum trigger_delay { + TRIGGER_DELAY_NEXT_PIXEL = 0, + TRIGGER_DELAY_NEXT_LINE, +}; + +enum crtc_event { + CRTC_EVENT_VSYNC_RISING = 0, + CRTC_EVENT_VSYNC_FALLING +}; + +struct crtc_trigger_info { + bool enabled; + struct dc_stream_state *event_source; + enum crtc_event event; + enum trigger_delay delay; +}; + +struct dc_crtc_timing_adjust { + uint32_t v_total_min; + uint32_t v_total_max; + uint32_t v_total_mid; + uint32_t v_total_mid_frame_num; +}; + + /* Passed on init */ enum vram_type { VIDEO_MEMORY_TYPE_GDDR5 = 2, @@ -812,7 +772,6 @@ enum vram_type { VIDEO_MEMORY_TYPE_GDDR6 = 6, }; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 enum dwb_cnv_out_bpc { DWB_CNV_OUT_BPC_8BPC = 0, DWB_CNV_OUT_BPC_10BPC = 1, @@ -863,7 +822,6 @@ struct mcif_buf_params { unsigned int swlock; }; -#endif #define MAX_TG_COLOR_VALUE 0x3FF struct tg_color { diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index 6f0b80111e58..d25603128394 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -29,13 +29,11 @@ #include "dc_types.h" #include "grph_object_defs.h" -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT enum dc_link_fec_state { dc_link_fec_not_ready, dc_link_fec_ready, dc_link_fec_enabled }; -#endif struct dc_link_status { bool link_active; struct dpcd_caps *dpcd_caps; @@ -83,6 +81,9 @@ struct dc_link { bool is_hpd_filter_disabled; bool dp_ss_off; bool link_state_valid; + bool aux_access_disabled; + bool sync_lt_in_progress; + bool is_lttpr_mode_transparent; /* caps is the same as reported_link_cap. link_traing use * reported_link_cap. Will clean up. TODO @@ -92,6 +93,8 @@ struct dc_link { struct dc_link_settings cur_link_settings; struct dc_lane_settings cur_lane_setting; struct dc_link_settings preferred_link_setting; + struct dc_link_training_overrides preferred_training_settings; + struct dp_audio_test_data audio_test_data; uint8_t ddc_hw_inst; @@ -123,12 +126,14 @@ struct dc_link { unsigned short chip_caps; unsigned int dpcd_sink_count; enum edp_revision edp_revision; - bool psr_enabled; + bool psr_feature_enabled; + bool psr_allow_active; /* MST record stream using this link */ struct link_flags { bool dp_keep_receiver_powered; bool dp_skip_DID2; + bool dp_skip_reset_segment; } wa_flags; struct link_mst_stream_allocation_table mst_stream_alloc_table; @@ -136,9 +141,7 @@ struct dc_link { struct link_trace link_trace; struct gpio *hpd_gpio; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT enum dc_link_fec_state fec_state; -#endif }; const struct dc_link_status *dc_link_get_status(const struct dc_link *dc_link); @@ -155,6 +158,18 @@ static inline struct dc_link *dc_get_link_at_index(struct dc *dc, uint32_t link_ return dc->links[link_index]; } +static inline struct dc_link *get_edp_link(const struct dc *dc) +{ + int i; + + // report any eDP links, even unconnected DDI's + for (i = 0; i < dc->link_count; i++) { + if (dc->links[i]->connector_signal == SIGNAL_TYPE_EDP) + return dc->links[i]; + } + return NULL; +} + /* Set backlight level of an embedded panel (eDP, LVDS). * backlight_pwm_u16_16 is unsigned 32 bit with 16 bit integer * and 16 bit fractional, where 1.0 is max backlight value. @@ -167,7 +182,7 @@ int dc_link_get_backlight_level(const struct dc_link *dc_link); bool dc_link_set_abm_disable(const struct dc_link *dc_link); -bool dc_link_set_psr_enable(const struct dc_link *dc_link, bool enable, bool wait); +bool dc_link_set_psr_allow_active(struct dc_link *dc_link, bool enable, bool wait); bool dc_link_get_psr_state(const struct dc_link *dc_link, uint32_t *psr_state); @@ -189,6 +204,8 @@ enum dc_detect_reason { bool dc_link_detect(struct dc_link *dc_link, enum dc_detect_reason reason); bool dc_link_get_hpd_state(struct dc_link *dc_link); +enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx); +enum dc_status dc_link_reallocate_mst_payload(struct dc_link *link); /* Notify DC about DP RX Interrupt (aka Short Pulse Interrupt). * Return: @@ -217,11 +234,24 @@ void dc_link_dp_set_drive_settings( struct dc_link *link, struct link_training_settings *lt_settings); +bool dc_link_dp_perform_link_training_skip_aux( + struct dc_link *link, + const struct dc_link_settings *link_setting); + enum link_training_result dc_link_dp_perform_link_training( struct dc_link *link, const struct dc_link_settings *link_setting, bool skip_video_pattern); +bool dc_link_dp_sync_lt_begin(struct dc_link *link); + +enum link_training_result dc_link_dp_sync_lt_attempt( + struct dc_link *link, + struct dc_link_settings *link_setting, + struct dc_link_training_overrides *lt_settings); + +bool dc_link_dp_sync_lt_end(struct dc_link *link, bool link_down); + void dc_link_dp_enable_hpd(const struct dc_link *link); void dc_link_dp_disable_hpd(const struct dc_link *link); @@ -229,6 +259,7 @@ void dc_link_dp_disable_hpd(const struct dc_link *link); bool dc_link_dp_set_test_pattern( struct dc_link *link, enum dp_test_pattern test_pattern, + enum dp_test_pattern_color_space test_pattern_color_space, const struct link_training_settings *p_link_settings, const unsigned char *p_custom_pattern, unsigned int cust_pattern_size); @@ -251,10 +282,16 @@ void dc_link_perform_link_training(struct dc *dc, void dc_link_set_preferred_link_settings(struct dc *dc, struct dc_link_settings *link_setting, struct dc_link *link); +void dc_link_set_preferred_training_settings(struct dc *dc, + struct dc_link_settings *link_setting, + struct dc_link_training_overrides *lt_overrides, + struct dc_link *link, + bool skip_immediate_retrain); void dc_link_enable_hpd(const struct dc_link *link); void dc_link_disable_hpd(const struct dc_link *link); void dc_link_set_test_pattern(struct dc_link *link, enum dp_test_pattern test_pattern, + enum dp_test_pattern_color_space test_pattern_color_space, const struct link_training_settings *p_link_settings, const unsigned char *p_custom_pattern, unsigned int cust_pattern_size); @@ -265,11 +302,18 @@ uint32_t dc_link_bandwidth_kbps( const struct dc_link_settings *dc_link_get_link_cap( const struct dc_link *link); +void dc_link_overwrite_extended_receiver_cap( + struct dc_link *link); + bool dc_submit_i2c( struct dc *dc, uint32_t link_index, struct i2c_command *cmd); +bool dc_submit_i2c_oem( + struct dc *dc, + struct i2c_command *cmd); + uint32_t dc_bandwidth_in_kbps_from_timing( const struct dc_crtc_timing *timing); #endif /* DC_LINK_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index 0fa1c26bc20d..92096de79dec 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -52,7 +52,6 @@ struct freesync_context { bool dummy; }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) enum hubp_dmdata_mode { DMDATA_SW_MODE, DMDATA_HW_MODE @@ -82,9 +81,7 @@ struct dc_dmdata_attributes { /* An unbounded array of uint32s, represents software dmdata to be loaded */ uint32_t *dmdata_sw_data; }; -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dc_writeback_info { bool wb_enabled; int dwb_pipe_inst; @@ -96,7 +93,6 @@ struct dc_writeback_update { unsigned int num_wb_info; struct dc_writeback_info writeback_info[MAX_DWB_PIPES]; }; -#endif enum vertical_interrupt_ref_point { START_V_UPDATE = 0, @@ -113,6 +109,19 @@ struct periodic_interrupt_config { int lines_offset; }; +union stream_update_flags { + struct { + uint32_t scaling:1; + uint32_t out_tf:1; + uint32_t out_csc:1; + uint32_t abm_level:1; + uint32_t dpms_off:1; + uint32_t gamut_remap:1; + uint32_t wb_update:1; + } bits; + + uint32_t raw; +}; struct dc_stream_state { // sink is deprecated, new code should not reference @@ -149,6 +158,7 @@ struct dc_stream_state { enum view_3d_format view_format; + bool use_vsc_sdp_for_colorimetry; bool ignore_msa_timing_param; bool converter_disable_audio; uint8_t qs_bit; @@ -188,11 +198,9 @@ struct dc_stream_state { struct crtc_trigger_info triggered_crtc_reset; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* writeback */ unsigned int num_wb_info; struct dc_writeback_info writeback_info[MAX_DWB_PIPES]; -#endif /* Computed state bits */ bool mode_changed : 1; @@ -211,12 +219,15 @@ struct dc_stream_state { bool apply_seamless_boot_optimization; uint32_t stream_id; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT bool is_dsc_enabled; -#endif + union stream_update_flags update_flags; }; +#define ABM_LEVEL_IMMEDIATE_DISABLE 0xFFFFFFFF + struct dc_stream_update { + struct dc_stream_state *stream; + struct rect src; struct rect dst; struct dc_transfer_func *out_transfer_func; @@ -231,6 +242,7 @@ struct dc_stream_update { struct dc_info_packet *vsp_infopacket; bool *dpms_off; + bool integer_scaling_update; struct colorspace_transform *gamut_remap; enum dc_color_space *output_color_space; @@ -238,12 +250,8 @@ struct dc_stream_update { struct dc_csc_transform *output_csc_transform; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dc_writeback_update *wb_update; -#endif -#if defined(CONFIG_DRM_AMD_DC_DSC_SUPPORT) struct dc_dsc_config *dsc_config; -#endif }; bool dc_is_stream_unchanged( @@ -333,18 +341,23 @@ bool dc_add_all_planes_for_stream( int plane_count, struct dc_state *context); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) bool dc_stream_add_writeback(struct dc *dc, struct dc_stream_state *stream, struct dc_writeback_info *wb_info); + bool dc_stream_remove_writeback(struct dc *dc, struct dc_stream_state *stream, uint32_t dwb_pipe_inst); + +bool dc_stream_warmup_writeback(struct dc *dc, + int num_dwb, + struct dc_writeback_info *wb_info); + bool dc_stream_dmdata_status_done(struct dc *dc, struct dc_stream_state *stream); + bool dc_stream_set_dynamic_metadata(struct dc *dc, struct dc_stream_state *stream, struct dc_dmdata_attributes *dmdata_attr); -#endif enum dc_status dc_validate_stream(struct dc *dc, struct dc_stream_state *stream); @@ -426,10 +439,13 @@ bool dc_stream_get_crc(struct dc *dc, uint32_t *g_y, uint32_t *b_cb); -void dc_stream_set_static_screen_events(struct dc *dc, +void dc_stream_set_static_screen_params(struct dc *dc, struct dc_stream_state **stream, int num_streams, - const struct dc_static_screen_events *events); + const struct dc_static_screen_params *params); + +void dc_stream_set_dyn_expansion(struct dc *dc, struct dc_stream_state *stream, + enum dc_dynamic_expansion option); void dc_stream_set_dither_option(struct dc_stream_state *stream, enum dc_dither_option option); diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 6eabb6491a3d..e59532d98cb4 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -25,6 +25,10 @@ #ifndef DC_TYPES_H_ #define DC_TYPES_H_ +/* AND EdidUtility only needs a portion + * of this file, including the rest only + * causes additional issues. + */ #include "os_types.h" #include "fixed31_32.h" #include "irq_types.h" @@ -33,12 +37,17 @@ #include "dal_types.h" #include "grph_object_defs.h" +#ifdef CONFIG_DRM_AMD_DC_HDCP +#include "dm_cp_psp.h" +#endif + /* forward declarations */ struct dc_plane_state; struct dc_stream_state; struct dc_link; struct dc_sink; struct dal; +struct dc_dmub_srv; /******************************** * Environment definitions @@ -51,7 +60,12 @@ enum dce_environment { DCE_ENV_FPGA_MAXIMUS, /* Emulation on real HW or on FPGA. Used by Diagnostics, enforces * requirements of Diagnostics team. */ - DCE_ENV_DIAG + DCE_ENV_DIAG, + /* + * Guest VM system, DC HW may exist but is not virtualized and + * should not be used. SW support for VDI only. + */ + DCE_ENV_VIRTUAL_HW }; /* Note: use these macro definitions instead of direct comparison! */ @@ -100,6 +114,11 @@ struct dc_context { uint32_t dc_sink_id_count; uint32_t dc_stream_id_count; uint64_t fbc_gpu_addr; + struct dc_dmub_srv *dmub_srv; + +#ifdef CONFIG_DRM_AMD_DC_HDCP + struct cp_psp cp_psp; +#endif }; @@ -107,6 +126,7 @@ struct dc_context { #define DC_EDID_BLOCK_SIZE 128 #define MAX_SURFACE_NUM 4 #define NUM_PIXEL_FORMATS 10 +#define MAX_REPEATER_CNT 8 #include "dc_ddc_types.h" @@ -159,6 +179,12 @@ enum dc_edid_status { EDID_THE_SAME, }; +enum act_return_status { + ACT_SUCCESS, + ACT_LINK_LOST, + ACT_FAILED +}; + /* audio capability from EDID*/ struct dc_cea_audio_mode { uint8_t format_code; /* ucData[0] [6:3]*/ @@ -202,6 +228,8 @@ struct dc_panel_patch { unsigned int dppowerup_delay; unsigned int extra_t12_ms; unsigned int extra_delay_backlight_off; + unsigned int extra_t7_ms; + unsigned int manage_secondary_link; }; struct dc_edid_caps { @@ -383,6 +411,30 @@ enum dpcd_downstream_port_max_bpc { DOWN_STREAM_MAX_12BPC, DOWN_STREAM_MAX_16BPC }; + + +enum link_training_offset { + DPRX = 0, + LTTPR_PHY_REPEATER1 = 1, + LTTPR_PHY_REPEATER2 = 2, + LTTPR_PHY_REPEATER3 = 3, + LTTPR_PHY_REPEATER4 = 4, + LTTPR_PHY_REPEATER5 = 5, + LTTPR_PHY_REPEATER6 = 6, + LTTPR_PHY_REPEATER7 = 7, + LTTPR_PHY_REPEATER8 = 8 +}; + +struct dc_lttpr_caps { + union dpcd_rev revision; + uint8_t mode; + uint8_t max_lane_count; + uint8_t max_link_rate; + uint8_t phy_repeater_cnt; + uint8_t max_ext_timeout; + uint8_t aux_rd_interval[MAX_REPEATER_CNT - 1]; +}; + struct dc_dongle_caps { /* dongle type (DP converter, CV smart dongle) */ enum display_dongle_type dongle_type; @@ -421,7 +473,6 @@ enum display_content_type { DISPLAY_CONTENT_TYPE_GAME = 8 }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* writeback */ struct dwb_stereo_params { bool stereo_enabled; /* false: normal mode, true: 3D stereo */ @@ -452,7 +503,6 @@ struct dc_dwb_params { enum dwb_subsample_position subsample_position; struct dc_transfer_func *out_transfer_func; }; -#endif /* audio*/ @@ -554,15 +604,17 @@ struct audio_info { /* this field must be last in this struct */ struct audio_mode modes[DC_MAX_AUDIO_DESC_COUNT]; }; - +struct audio_check { + unsigned int audio_packet_type; + unsigned int max_audiosample_rate; + unsigned int acat; +}; enum dc_infoframe_type { DC_HDMI_INFOFRAME_TYPE_VENDOR = 0x81, DC_HDMI_INFOFRAME_TYPE_AVI = 0x82, DC_HDMI_INFOFRAME_TYPE_SPD = 0x83, DC_HDMI_INFOFRAME_TYPE_AUDIO = 0x84, -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT DC_DP_INFOFRAME_TYPE_PPS = 0x10, -#endif }; struct dc_info_packet { @@ -677,7 +729,7 @@ struct psr_context { /* The VSync rate in Hz used to calculate the * step size for smooth brightness feature */ - unsigned int vsyncRateHz; + unsigned int vsync_rate_hz; unsigned int skipPsrWaitForPllLock; unsigned int numberOfControllers; /* Unused, for future use. To indicate that first changed frame from @@ -725,7 +777,19 @@ struct AsicStateEx { unsigned int phyClock; }; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + +enum dc_clock_type { + DC_CLOCK_TYPE_DISPCLK = 0, + DC_CLOCK_TYPE_DPPCLK = 1, +}; + +struct dc_clock_config { + uint32_t max_clock_khz; + uint32_t min_clock_khz; + uint32_t bw_requirequired_clock_khz; + uint32_t current_clock_khz;/*current clock in use*/ +}; + /* DSC DPCD capabilities */ union dsc_slice_caps1 { struct { @@ -795,5 +859,5 @@ struct dsc_dec_dpcd_caps { uint32_t branch_overall_throughput_1_mps; /* In MPs */ uint32_t branch_max_line_width; }; -#endif + #endif /* DC_TYPES_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c index 58bd131d5b48..b8a3fc505c9b 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c @@ -77,6 +77,9 @@ static bool dce_abm_set_pipe(struct abm *abm, uint32_t controller_id) /* notifyDMCUMsg */ REG_UPDATE(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 1); + REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, + 1, 80000); + return true; } @@ -401,6 +404,10 @@ static bool dce_abm_init_backlight(struct abm *abm) /* Enable the backlight output */ REG_UPDATE(BL_PWM_CNTL, BL_PWM_EN, 1); + /* Disable fractional pwm if configured */ + REG_UPDATE(BL_PWM_CNTL, BL_PWM_FRACTIONAL_EN, + abm->ctx->dc->config.disable_fractional_pwm ? 0 : 1); + /* Unlock group 2 backlight registers */ REG_UPDATE(BL_PWM_GRP1_REG_LOCK, BL_PWM_GRP1_REG_LOCK, 0); @@ -489,9 +496,6 @@ void dce_abm_destroy(struct abm **abm) { struct dce_abm *abm_dce = TO_DCE_ABM(*abm); - if (abm_dce->base.dmcu_is_running == true) - abm_dce->base.funcs->set_abm_immediate_disable(*abm); - kfree(abm_dce); *abm = NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h index 7ba7e6f722f6..ba0caaffa24b 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h @@ -67,7 +67,6 @@ SRI(DC_ABM1_HGLS_REG_READ_PROGRESS, ABM, id), \ NBIO_SR(BIOS_SCRATCH_2) -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define ABM_DCN20_REG_LIST() \ ABM_COMMON_REG_LIST_DCE_BASE(), \ SR(DC_ABM1_HG_SAMPLE_RATE), \ @@ -81,7 +80,6 @@ SR(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES), \ SR(DC_ABM1_HGLS_REG_READ_PROGRESS), \ NBIO_SR(BIOS_SCRATCH_2) -#endif #define ABM_SF(reg_name, field_name, post_fix)\ .field_name = reg_name ## __ ## field_name ## post_fix @@ -163,9 +161,7 @@ ABM_SF(ABM0_DC_ABM1_HGLS_REG_READ_PROGRESS, \ ABM1_BL_REG_READ_MISSED_FRAME_CLEAR, mask_sh) -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define ABM_MASK_SH_LIST_DCN20(mask_sh) ABM_MASK_SH_LIST_DCE110(mask_sh) -#endif #define ABM_REG_FIELD_LIST(type) \ type ABM1_HG_NUM_OF_BINS_SEL; \ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c index 4a10a5d22c90..5a35495bc11d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c @@ -145,20 +145,20 @@ static void check_audio_bandwidth_hdmi( if (channel_count > 2) { /* Based on HDMI spec 1.3 Table 7.5 */ - if ((crtc_info->requested_pixel_clock <= 27000) && + if ((crtc_info->requested_pixel_clock_100Hz <= 270000) && (crtc_info->v_active <= 576) && !(crtc_info->interlaced) && !(crtc_info->pixel_repetition == 2 || crtc_info->pixel_repetition == 4)) { limit_freq_to_48_khz = true; - } else if ((crtc_info->requested_pixel_clock <= 27000) && + } else if ((crtc_info->requested_pixel_clock_100Hz <= 270000) && (crtc_info->v_active <= 576) && (crtc_info->interlaced) && (crtc_info->pixel_repetition == 2)) { limit_freq_to_88_2_khz = true; - } else if ((crtc_info->requested_pixel_clock <= 54000) && + } else if ((crtc_info->requested_pixel_clock_100Hz <= 540000) && (crtc_info->v_active <= 576) && !(crtc_info->interlaced)) { limit_freq_to_174_4_khz = true; @@ -613,6 +613,8 @@ void dce_aud_az_configure( AZ_REG_WRITE(AZALIA_F0_CODEC_PIN_CONTROL_SINK_INFO1, value); + DC_LOG_HW_AUDIO("\n\tAUDIO:az_configure: index: %u data, 0x%x, displayName %s: \n", + audio->inst, value, audio_info->display_name); /* *write the port ID: @@ -737,8 +739,8 @@ void dce_aud_az_configure( /* search pixel clock value for Azalia HDMI Audio */ static void get_azalia_clock_info_hdmi( - uint32_t crtc_pixel_clock_in_khz, - uint32_t actual_pixel_clock_in_khz, + uint32_t crtc_pixel_clock_100hz, + uint32_t actual_pixel_clock_100Hz, struct azalia_clock_info *azalia_clock_info) { /* audio_dto_phase= 24 * 10,000; @@ -749,11 +751,11 @@ static void get_azalia_clock_info_hdmi( /* audio_dto_module = PCLKFrequency * 10,000; * [khz] -> [100Hz] */ azalia_clock_info->audio_dto_module = - actual_pixel_clock_in_khz * 10; + actual_pixel_clock_100Hz; } static void get_azalia_clock_info_dp( - uint32_t requested_pixel_clock_in_khz, + uint32_t requested_pixel_clock_100Hz, const struct audio_pll_info *pll_info, struct azalia_clock_info *azalia_clock_info) { @@ -792,15 +794,15 @@ void dce_aud_wall_dto_setup( /* calculate DTO settings */ get_azalia_clock_info_hdmi( - crtc_info->requested_pixel_clock, - crtc_info->calculated_pixel_clock, + crtc_info->requested_pixel_clock_100Hz, + crtc_info->calculated_pixel_clock_100Hz, &clock_info); - DC_LOG_HW_AUDIO("\n%s:Input::requested_pixel_clock = %d"\ - "calculated_pixel_clock =%d\n"\ + DC_LOG_HW_AUDIO("\n%s:Input::requested_pixel_clock_100Hz = %d"\ + "calculated_pixel_clock_100Hz =%d\n"\ "audio_dto_module = %d audio_dto_phase =%d \n\n", __func__,\ - crtc_info->requested_pixel_clock,\ - crtc_info->calculated_pixel_clock,\ + crtc_info->requested_pixel_clock_100Hz,\ + crtc_info->calculated_pixel_clock_100Hz,\ clock_info.audio_dto_module,\ clock_info.audio_dto_phase); @@ -833,7 +835,7 @@ void dce_aud_wall_dto_setup( calculate DTO settings */ get_azalia_clock_info_dp( - crtc_info->requested_pixel_clock, + crtc_info->requested_pixel_clock_100Hz, pll_info, &clock_info); @@ -922,7 +924,6 @@ static const struct audio_funcs funcs = { .az_configure = dce_aud_az_configure, .destroy = dce_aud_destroy, }; - void dce_aud_destroy(struct audio **audio) { struct dce_audio *aud = DCE_AUD(*audio); @@ -936,7 +937,7 @@ struct audio *dce_audio_create( unsigned int inst, const struct dce_audio_registers *reg, const struct dce_audio_shift *shifts, - const struct dce_aduio_mask *masks + const struct dce_audio_mask *masks ) { struct dce_audio *audio = kzalloc(sizeof(*audio), GFP_KERNEL); @@ -953,7 +954,6 @@ struct audio *dce_audio_create( audio->regs = reg; audio->shifts = shifts; audio->masks = masks; - return &audio->base; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h index a0d5724aab31..1392fab0860b 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h @@ -101,7 +101,7 @@ struct dce_audio_shift { uint32_t DCCG_AUDIO_DTO1_USE_512FBR_DTO; }; -struct dce_aduio_mask { +struct dce_audio_mask { uint32_t AZALIA_ENDPOINT_REG_INDEX; uint32_t AZALIA_ENDPOINT_REG_DATA; @@ -125,7 +125,7 @@ struct dce_audio { struct audio base; const struct dce_audio_registers *regs; const struct dce_audio_shift *shifts; - const struct dce_aduio_mask *masks; + const struct dce_audio_mask *masks; }; struct audio *dce_audio_create( @@ -133,7 +133,7 @@ struct audio *dce_audio_create( unsigned int inst, const struct dce_audio_registers *reg, const struct dce_audio_shift *shifts, - const struct dce_aduio_mask *masks); + const struct dce_audio_mask *masks); void dce_aud_destroy(struct audio **audio); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c index f2295e780031..68c4049cbc2a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c @@ -30,6 +30,7 @@ #include "core_types.h" #include "dce_aux.h" #include "dce/dce_11_0_sh_mask.h" +#include "dm_event_log.h" #define CTX \ aux110->base.ctx @@ -41,6 +42,10 @@ #include "reg_helper.h" +#undef FN +#define FN(reg_name, field_name) \ + aux110->shift->field_name, aux110->mask->field_name + #define FROM_AUX_ENGINE(ptr) \ container_of((ptr), struct aux_engine_dce110, base) @@ -54,6 +59,16 @@ enum { AUX_TIMED_OUT_RETRY_COUNTER = 2, AUX_DEFER_RETRY_COUNTER = 6 }; + +#define TIME_OUT_INCREMENT 1016 +#define TIME_OUT_MULTIPLIER_8 8 +#define TIME_OUT_MULTIPLIER_16 16 +#define TIME_OUT_MULTIPLIER_32 32 +#define TIME_OUT_MULTIPLIER_64 64 +#define MAX_TIMEOUT_LENGTH 127 +#define DEFAULT_AUX_ENGINE_MULT 0 +#define DEFAULT_AUX_ENGINE_LENGTH 69 + static void release_engine( struct dce_aux *engine) { @@ -197,7 +212,7 @@ static void submit_channel_request( REG_UPDATE(AUX_INTERRUPT_CONTROL, AUX_SW_DONE_ACK, 1); REG_WAIT(AUX_SW_STATUS, AUX_SW_DONE, 0, - 10, aux110->timeout_period/10); + 10, aux110->polling_timeout_period/10); /* set the delay and the number of bytes to write */ @@ -252,6 +267,8 @@ static void submit_channel_request( } REG_UPDATE(AUX_SW_CONTROL, AUX_SW_GO, 1); + EVENT_LOG_AUX_REQ(engine->ddc->pin_data->en, EVENT_LOG_AUX_ORIGIN_NATIVE, + request->action, request->address, request->length, request->data); } static int read_channel_reply(struct dce_aux *engine, uint32_t size, @@ -324,7 +341,7 @@ static enum aux_channel_operation_result get_channel_status( /* poll to make sure that SW_DONE is asserted */ REG_WAIT(AUX_SW_STATUS, AUX_SW_DONE, 1, - 10, aux110->timeout_period/10); + 10, aux110->polling_timeout_period/10); value = REG_READ(AUX_SW_STATUS); /* in case HPD is LOW, exit AUX transaction */ @@ -383,7 +400,7 @@ static bool acquire( { enum gpio_result result; - if (!is_engine_available(engine)) + if ((engine == NULL) || !is_engine_available(engine)) return false; result = dal_ddc_open(ddc, GPIO_MODE_HARDWARE, @@ -411,20 +428,103 @@ void dce110_engine_destroy(struct dce_aux **engine) *engine = NULL; } + +static uint32_t dce_aux_configure_timeout(struct ddc_service *ddc, + uint32_t timeout_in_us) +{ + uint32_t multiplier = 0; + uint32_t length = 0; + uint32_t prev_length = 0; + uint32_t prev_mult = 0; + uint32_t prev_timeout_val = 0; + struct ddc *ddc_pin = ddc->ddc_pin; + struct dce_aux *aux_engine = ddc->ctx->dc->res_pool->engines[ddc_pin->pin_data->en]; + struct aux_engine_dce110 *aux110 = FROM_AUX_ENGINE(aux_engine); + + /* 1-Update polling timeout period */ + aux110->polling_timeout_period = timeout_in_us * SW_AUX_TIMEOUT_PERIOD_MULTIPLIER; + + /* 2-Update aux timeout period length and multiplier */ + if (timeout_in_us == 0) { + multiplier = DEFAULT_AUX_ENGINE_MULT; + length = DEFAULT_AUX_ENGINE_LENGTH; + } else if (timeout_in_us <= TIME_OUT_INCREMENT) { + multiplier = 0; + length = timeout_in_us/TIME_OUT_MULTIPLIER_8; + if (timeout_in_us % TIME_OUT_MULTIPLIER_8 != 0) + length++; + } else if (timeout_in_us <= 2 * TIME_OUT_INCREMENT) { + multiplier = 1; + length = timeout_in_us/TIME_OUT_MULTIPLIER_16; + if (timeout_in_us % TIME_OUT_MULTIPLIER_16 != 0) + length++; + } else if (timeout_in_us <= 4 * TIME_OUT_INCREMENT) { + multiplier = 2; + length = timeout_in_us/TIME_OUT_MULTIPLIER_32; + if (timeout_in_us % TIME_OUT_MULTIPLIER_32 != 0) + length++; + } else if (timeout_in_us > 4 * TIME_OUT_INCREMENT) { + multiplier = 3; + length = timeout_in_us/TIME_OUT_MULTIPLIER_64; + if (timeout_in_us % TIME_OUT_MULTIPLIER_64 != 0) + length++; + } + + length = (length < MAX_TIMEOUT_LENGTH) ? length : MAX_TIMEOUT_LENGTH; + + REG_GET_2(AUX_DPHY_RX_CONTROL1, AUX_RX_TIMEOUT_LEN, &prev_length, AUX_RX_TIMEOUT_LEN_MUL, &prev_mult); + + switch (prev_mult) { + case 0: + prev_timeout_val = prev_length * TIME_OUT_MULTIPLIER_8; + break; + case 1: + prev_timeout_val = prev_length * TIME_OUT_MULTIPLIER_16; + break; + case 2: + prev_timeout_val = prev_length * TIME_OUT_MULTIPLIER_32; + break; + case 3: + prev_timeout_val = prev_length * TIME_OUT_MULTIPLIER_64; + break; + default: + prev_timeout_val = DEFAULT_AUX_ENGINE_LENGTH * TIME_OUT_MULTIPLIER_8; + break; + } + + REG_UPDATE_SEQ_2(AUX_DPHY_RX_CONTROL1, AUX_RX_TIMEOUT_LEN, length, AUX_RX_TIMEOUT_LEN_MUL, multiplier); + + return prev_timeout_val; +} + +static struct dce_aux_funcs aux_functions = { + .configure_timeout = NULL, + .destroy = NULL, +}; + struct dce_aux *dce110_aux_engine_construct(struct aux_engine_dce110 *aux_engine110, struct dc_context *ctx, uint32_t inst, uint32_t timeout_period, - const struct dce110_aux_registers *regs) + const struct dce110_aux_registers *regs, + const struct dce110_aux_registers_mask *mask, + const struct dce110_aux_registers_shift *shift, + bool is_ext_aux_timeout_configurable) { aux_engine110->base.ddc = NULL; aux_engine110->base.ctx = ctx; aux_engine110->base.delay = 0; aux_engine110->base.max_defer_write_retry = 0; aux_engine110->base.inst = inst; - aux_engine110->timeout_period = timeout_period; + aux_engine110->polling_timeout_period = timeout_period; aux_engine110->regs = regs; + aux_engine110->mask = mask; + aux_engine110->shift = shift; + aux_engine110->base.funcs = &aux_functions; + if (is_ext_aux_timeout_configurable) + aux_engine110->base.funcs->configure_timeout = &dce_aux_configure_timeout; + return &aux_engine110->base; } @@ -461,8 +561,10 @@ int dce_aux_transfer_raw(struct ddc_service *ddc, memset(&aux_rep, 0, sizeof(aux_rep)); aux_engine = ddc->ctx->dc->res_pool->engines[ddc_pin->pin_data->en]; - if (!acquire(aux_engine, ddc_pin)) + if (!acquire(aux_engine, ddc_pin)) { + *operation_result = AUX_CHANNEL_OPERATION_FAILED_ENGINE_ACQUIRE; return -1; + } if (payload->i2c_over_aux) aux_req.type = AUX_TRANSACTION_TYPE_I2C; @@ -472,7 +574,7 @@ int dce_aux_transfer_raw(struct ddc_service *ddc, aux_req.action = i2caux_action_from_payload(payload); aux_req.address = payload->address; - aux_req.delay = payload->defer_delay * 10; + aux_req.delay = 0; aux_req.length = payload->length; aux_req.data = payload->data; @@ -480,9 +582,13 @@ int dce_aux_transfer_raw(struct ddc_service *ddc, *operation_result = get_channel_status(aux_engine, &returned_bytes); if (*operation_result == AUX_CHANNEL_OPERATION_SUCCEEDED) { - read_channel_reply(aux_engine, payload->length, + int bytes_replied = 0; + bytes_replied = read_channel_reply(aux_engine, payload->length, payload->data, payload->reply, &status); + EVENT_LOG_AUX_REP(aux_engine->ddc->pin_data->en, + EVENT_LOG_AUX_ORIGIN_NATIVE, *payload->reply, + bytes_replied, payload->data); res = returned_bytes; } else { res = -1; @@ -505,6 +611,8 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc, uint8_t reply; bool payload_reply = true; enum aux_channel_operation_result operation_result; + bool retry_on_defer = false; + int aux_ack_retries = 0, aux_defer_retries = 0, aux_i2c_defer_retries = 0, @@ -535,10 +643,19 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc, break; case AUX_TRANSACTION_REPLY_AUX_DEFER: - case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK: case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER: - if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES) + retry_on_defer = true; + /* fall through */ + case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK: + if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES) { goto fail; + } else { + if ((*payload->reply == AUX_TRANSACTION_REPLY_AUX_DEFER) || + (*payload->reply == AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER)) { + if (payload->defer_delay > 0) + msleep(payload->defer_delay); + } + } break; case AUX_TRANSACTION_REPLY_I2C_DEFER: @@ -562,19 +679,29 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc, break; case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT: - if (++aux_timeout_retries >= AUX_MAX_TIMEOUT_RETRIES) - goto fail; - else { - /* - * DP 1.4, 2.8.2: AUX Transaction Response/Reply Timeouts - * According to the DP spec there should be 3 retries total - * with a 400us wait inbetween each. Hardware already waits - * for 550us therefore no wait is required here. - */ + // Check whether a DEFER had occurred before the timeout. + // If so, treat timeout as a DEFER. + if (retry_on_defer) { + if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES) + goto fail; + else if (payload->defer_delay > 0) + msleep(payload->defer_delay); + } else { + if (++aux_timeout_retries >= AUX_MAX_TIMEOUT_RETRIES) + goto fail; + else { + /* + * DP 1.4, 2.8.2: AUX Transaction Response/Reply Timeouts + * According to the DP spec there should be 3 retries total + * with a 400us wait inbetween each. Hardware already waits + * for 550us therefore no wait is required here. + */ + } } break; case AUX_CHANNEL_OPERATION_FAILED_HPD_DISCON: + case AUX_CHANNEL_OPERATION_FAILED_ENGINE_ACQUIRE: case AUX_CHANNEL_OPERATION_FAILED_REASON_UNKNOWN: default: goto fail; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h index ed7fec8fe253..382465862f29 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.h @@ -29,15 +29,15 @@ #include "i2caux_interface.h" #include "inc/hw/aux_engine.h" -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 + #define AUX_COMMON_REG_LIST0(id)\ SRI(AUX_CONTROL, DP_AUX, id), \ SRI(AUX_ARB_CONTROL, DP_AUX, id), \ SRI(AUX_SW_DATA, DP_AUX, id), \ SRI(AUX_SW_CONTROL, DP_AUX, id), \ SRI(AUX_INTERRUPT_CONTROL, DP_AUX, id), \ + SRI(AUX_DPHY_RX_CONTROL1, DP_AUX, id), \ SRI(AUX_SW_STATUS, DP_AUX, id) -#endif #define AUX_COMMON_REG_LIST(id)\ SRI(AUX_CONTROL, DP_AUX, id), \ @@ -55,6 +55,7 @@ struct dce110_aux_registers { uint32_t AUX_SW_DATA; uint32_t AUX_SW_CONTROL; uint32_t AUX_INTERRUPT_CONTROL; + uint32_t AUX_DPHY_RX_CONTROL1; uint32_t AUX_SW_STATUS; uint32_t AUXN_IMPCAL; uint32_t AUXP_IMPCAL; @@ -62,6 +63,156 @@ struct dce110_aux_registers { uint32_t AUX_RESET_MASK; }; +#define DCE_AUX_REG_FIELD_LIST(type)\ + type AUX_EN;\ + type AUX_RESET;\ + type AUX_RESET_DONE;\ + type AUX_REG_RW_CNTL_STATUS;\ + type AUX_SW_USE_AUX_REG_REQ;\ + type AUX_SW_DONE_USING_AUX_REG;\ + type AUX_SW_AUTOINCREMENT_DISABLE;\ + type AUX_SW_DATA_RW;\ + type AUX_SW_INDEX;\ + type AUX_SW_GO;\ + type AUX_SW_DATA;\ + type AUX_SW_REPLY_BYTE_COUNT;\ + type AUX_SW_DONE;\ + type AUX_SW_DONE_ACK;\ + type AUXN_IMPCAL_ENABLE;\ + type AUXP_IMPCAL_ENABLE;\ + type AUXN_IMPCAL_OVERRIDE_ENABLE;\ + type AUXP_IMPCAL_OVERRIDE_ENABLE;\ + type AUX_RX_TIMEOUT_LEN;\ + type AUX_RX_TIMEOUT_LEN_MUL;\ + type AUXN_CALOUT_ERROR_AK;\ + type AUXP_CALOUT_ERROR_AK;\ + type AUX_SW_START_DELAY;\ + type AUX_SW_WR_BYTES + +#define DCE10_AUX_MASK_SH_LIST(mask_sh)\ + AUX_SF(AUX_CONTROL, AUX_EN, mask_sh),\ + AUX_SF(AUX_ARB_CONTROL, AUX_REG_RW_CNTL_STATUS, mask_sh),\ + AUX_SF(AUX_ARB_CONTROL, AUX_SW_USE_AUX_REG_REQ, mask_sh),\ + AUX_SF(AUX_ARB_CONTROL, AUX_SW_DONE_USING_AUX_REG, mask_sh),\ + AUX_SF(AUX_SW_CONTROL, AUX_SW_START_DELAY, mask_sh),\ + AUX_SF(AUX_SW_CONTROL, AUX_SW_WR_BYTES, mask_sh),\ + AUX_SF(AUX_SW_CONTROL, AUX_SW_GO, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_DATA_RW, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_INDEX, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_DATA, mask_sh),\ + AUX_SF(AUX_SW_STATUS, AUX_SW_REPLY_BYTE_COUNT, mask_sh),\ + AUX_SF(AUX_SW_STATUS, AUX_SW_DONE, mask_sh),\ + AUX_SF(AUX_INTERRUPT_CONTROL, AUX_SW_DONE_ACK, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_CALOUT_ERROR_AK, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_CALOUT_ERROR_AK, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_IMPCAL_ENABLE, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_IMPCAL_ENABLE, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_IMPCAL_OVERRIDE_ENABLE, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_IMPCAL_OVERRIDE_ENABLE, mask_sh) + +#define DCE_AUX_MASK_SH_LIST(mask_sh)\ + AUX_SF(AUX_CONTROL, AUX_EN, mask_sh),\ + AUX_SF(AUX_CONTROL, AUX_RESET, mask_sh),\ + AUX_SF(AUX_CONTROL, AUX_RESET_DONE, mask_sh),\ + AUX_SF(AUX_ARB_CONTROL, AUX_REG_RW_CNTL_STATUS, mask_sh),\ + AUX_SF(AUX_ARB_CONTROL, AUX_SW_USE_AUX_REG_REQ, mask_sh),\ + AUX_SF(AUX_ARB_CONTROL, AUX_SW_DONE_USING_AUX_REG, mask_sh),\ + AUX_SF(AUX_SW_CONTROL, AUX_SW_START_DELAY, mask_sh),\ + AUX_SF(AUX_SW_CONTROL, AUX_SW_WR_BYTES, mask_sh),\ + AUX_SF(AUX_SW_CONTROL, AUX_SW_GO, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_DATA_RW, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_INDEX, mask_sh),\ + AUX_SF(AUX_SW_DATA, AUX_SW_DATA, mask_sh),\ + AUX_SF(AUX_SW_STATUS, AUX_SW_REPLY_BYTE_COUNT, mask_sh),\ + AUX_SF(AUX_SW_STATUS, AUX_SW_DONE, mask_sh),\ + AUX_SF(AUX_INTERRUPT_CONTROL, AUX_SW_DONE_ACK, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_CALOUT_ERROR_AK, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_CALOUT_ERROR_AK, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_IMPCAL_ENABLE, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_IMPCAL_ENABLE, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_IMPCAL_OVERRIDE_ENABLE, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_IMPCAL_OVERRIDE_ENABLE, mask_sh) + +#define DCE12_AUX_MASK_SH_LIST(mask_sh)\ + AUX_SF(DP_AUX0_AUX_CONTROL, AUX_EN, mask_sh),\ + AUX_SF(DP_AUX0_AUX_CONTROL, AUX_RESET, mask_sh),\ + AUX_SF(DP_AUX0_AUX_CONTROL, AUX_RESET_DONE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_ARB_CONTROL, AUX_REG_RW_CNTL_STATUS, mask_sh),\ + AUX_SF(DP_AUX0_AUX_ARB_CONTROL, AUX_SW_USE_AUX_REG_REQ, mask_sh),\ + AUX_SF(DP_AUX0_AUX_ARB_CONTROL, AUX_SW_DONE_USING_AUX_REG, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_CONTROL, AUX_SW_START_DELAY, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_CONTROL, AUX_SW_WR_BYTES, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_CONTROL, AUX_SW_GO, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_DATA_RW, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_INDEX, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_DATA, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_STATUS, AUX_SW_REPLY_BYTE_COUNT, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_STATUS, AUX_SW_DONE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_INTERRUPT_CONTROL, AUX_SW_DONE_ACK, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_CALOUT_ERROR_AK, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_CALOUT_ERROR_AK, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_IMPCAL_ENABLE, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_IMPCAL_ENABLE, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_IMPCAL_OVERRIDE_ENABLE, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_IMPCAL_OVERRIDE_ENABLE, mask_sh) + +/* DCN10 MASK */ +#define DCN10_AUX_MASK_SH_LIST(mask_sh)\ + AUX_SF(DP_AUX0_AUX_CONTROL, AUX_EN, mask_sh),\ + AUX_SF(DP_AUX0_AUX_CONTROL, AUX_RESET, mask_sh),\ + AUX_SF(DP_AUX0_AUX_CONTROL, AUX_RESET_DONE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_ARB_CONTROL, AUX_REG_RW_CNTL_STATUS, mask_sh),\ + AUX_SF(DP_AUX0_AUX_ARB_CONTROL, AUX_SW_USE_AUX_REG_REQ, mask_sh),\ + AUX_SF(DP_AUX0_AUX_ARB_CONTROL, AUX_SW_DONE_USING_AUX_REG, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_CONTROL, AUX_SW_START_DELAY, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_CONTROL, AUX_SW_WR_BYTES, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_CONTROL, AUX_SW_GO, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_DATA_RW, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_INDEX, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_DATA, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_STATUS, AUX_SW_REPLY_BYTE_COUNT, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_STATUS, AUX_SW_DONE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_INTERRUPT_CONTROL, AUX_SW_DONE_ACK, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_CALOUT_ERROR_AK, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_CALOUT_ERROR_AK, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_IMPCAL_ENABLE, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_IMPCAL_ENABLE, mask_sh),\ + AUX_SF(AUXP_IMPCAL, AUXP_IMPCAL_OVERRIDE_ENABLE, mask_sh),\ + AUX_SF(AUXN_IMPCAL, AUXN_IMPCAL_OVERRIDE_ENABLE, mask_sh) + +/* for all other DCN */ +#define DCN_AUX_MASK_SH_LIST(mask_sh)\ + AUX_SF(DP_AUX0_AUX_CONTROL, AUX_EN, mask_sh),\ + AUX_SF(DP_AUX0_AUX_CONTROL, AUX_RESET, mask_sh),\ + AUX_SF(DP_AUX0_AUX_CONTROL, AUX_RESET_DONE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_ARB_CONTROL, AUX_REG_RW_CNTL_STATUS, mask_sh),\ + AUX_SF(DP_AUX0_AUX_ARB_CONTROL, AUX_SW_USE_AUX_REG_REQ, mask_sh),\ + AUX_SF(DP_AUX0_AUX_ARB_CONTROL, AUX_SW_DONE_USING_AUX_REG, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_CONTROL, AUX_SW_START_DELAY, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_CONTROL, AUX_SW_WR_BYTES, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_CONTROL, AUX_SW_GO, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_DATA_RW, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_AUTOINCREMENT_DISABLE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_INDEX, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_DATA, AUX_SW_DATA, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_STATUS, AUX_SW_REPLY_BYTE_COUNT, mask_sh),\ + AUX_SF(DP_AUX0_AUX_SW_STATUS, AUX_SW_DONE, mask_sh),\ + AUX_SF(DP_AUX0_AUX_INTERRUPT_CONTROL, AUX_SW_DONE_ACK, mask_sh),\ + AUX_SF(DP_AUX0_AUX_DPHY_RX_CONTROL1, AUX_RX_TIMEOUT_LEN, mask_sh),\ + AUX_SF(DP_AUX0_AUX_DPHY_RX_CONTROL1, AUX_RX_TIMEOUT_LEN_MUL, mask_sh) + +#define AUX_SF(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + enum { /* This is the timeout as defined in DP 1.2a, * 2.3.4 "Detailed uPacket TX AUX CH State Description". */ @@ -97,20 +248,34 @@ struct dce_aux { uint32_t max_defer_write_retry; bool acquire_reset; + struct dce_aux_funcs *funcs; }; +struct dce110_aux_registers_mask { + DCE_AUX_REG_FIELD_LIST(uint32_t); +}; + +struct dce110_aux_registers_shift { + DCE_AUX_REG_FIELD_LIST(uint8_t); +}; + + struct aux_engine_dce110 { struct dce_aux base; const struct dce110_aux_registers *regs; + const struct dce110_aux_registers_mask *mask; + const struct dce110_aux_registers_shift *shift; struct { uint32_t aux_control; uint32_t aux_arb_control; uint32_t aux_sw_data; uint32_t aux_sw_control; uint32_t aux_interrupt_control; + uint32_t aux_dphy_rx_control1; + uint32_t aux_dphy_rx_control0; uint32_t aux_sw_status; } addr; - uint32_t timeout_period; + uint32_t polling_timeout_period; }; struct aux_engine_dce110_init_data { @@ -120,12 +285,15 @@ struct aux_engine_dce110_init_data { const struct dce110_aux_registers *regs; }; -struct dce_aux *dce110_aux_engine_construct( - struct aux_engine_dce110 *aux_engine110, +struct dce_aux *dce110_aux_engine_construct(struct aux_engine_dce110 *aux_engine110, struct dc_context *ctx, uint32_t inst, uint32_t timeout_period, - const struct dce110_aux_registers *regs); + const struct dce110_aux_registers *regs, + + const struct dce110_aux_registers_mask *mask, + const struct dce110_aux_registers_shift *shift, + bool is_ext_aux_timeout_configurable); void dce110_engine_destroy(struct dce_aux **engine); @@ -139,4 +307,13 @@ int dce_aux_transfer_raw(struct ddc_service *ddc, bool dce_aux_transfer_with_retries(struct ddc_service *ddc, struct aux_payload *cmd); + +struct dce_aux_funcs { + uint32_t (*configure_timeout) + (struct ddc_service *ddc, + uint32_t timeout); + void (*destroy) + (struct aux_engine **ptr); +}; + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c index 5fae77e201d5..2e992fbc0d71 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c @@ -905,7 +905,7 @@ static bool dce112_program_pix_clk( struct dce110_clk_src *clk_src = TO_DCE110_CLK_SRC(clock_source); struct bp_pixel_clock_parameters bp_pc_params = {0}; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) if (IS_FPGA_MAXIMUS_DC(clock_source->ctx->dce_environment)) { unsigned int inst = pix_clk_params->controller_id - CONTROLLER_ID_D0; unsigned dp_dto_ref_100hz = 7000000; @@ -1004,7 +1004,6 @@ static bool get_pixel_clk_frequency_100hz( return false; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* this table is use to find *1.001 and /1.001 pixel rates from non-precise pixel rate */ struct pixel_rate_range_table_entry { @@ -1061,9 +1060,9 @@ static bool dcn20_program_pix_clk( static const struct clock_source_funcs dcn20_clk_src_funcs = { .cs_power_down = dce110_clock_source_power_down, .program_pix_clk = dcn20_program_pix_clk, - .get_pix_clk_dividers = dce112_get_pix_clk_dividers + .get_pix_clk_dividers = dce112_get_pix_clk_dividers, + .get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz }; -#endif /*****************************************/ /* Constructor */ @@ -1234,37 +1233,36 @@ static bool calc_pll_max_vco_construct( struct calc_pll_clock_source_init_data *init_data) { uint32_t i; - struct dc_firmware_info fw_info = { { 0 } }; + struct dc_firmware_info *fw_info; if (calc_pll_cs == NULL || init_data == NULL || init_data->bp == NULL) return false; - if (init_data->bp->funcs->get_firmware_info( - init_data->bp, - &fw_info) != BP_RESULT_OK) + if (!init_data->bp->fw_info_valid) return false; + fw_info = &init_data->bp->fw_info; calc_pll_cs->ctx = init_data->ctx; - calc_pll_cs->ref_freq_khz = fw_info.pll_info.crystal_frequency; + calc_pll_cs->ref_freq_khz = fw_info->pll_info.crystal_frequency; calc_pll_cs->min_vco_khz = - fw_info.pll_info.min_output_pxl_clk_pll_frequency; + fw_info->pll_info.min_output_pxl_clk_pll_frequency; calc_pll_cs->max_vco_khz = - fw_info.pll_info.max_output_pxl_clk_pll_frequency; + fw_info->pll_info.max_output_pxl_clk_pll_frequency; if (init_data->max_override_input_pxl_clk_pll_freq_khz != 0) calc_pll_cs->max_pll_input_freq_khz = init_data->max_override_input_pxl_clk_pll_freq_khz; else calc_pll_cs->max_pll_input_freq_khz = - fw_info.pll_info.max_input_pxl_clk_pll_frequency; + fw_info->pll_info.max_input_pxl_clk_pll_frequency; if (init_data->min_override_input_pxl_clk_pll_freq_khz != 0) calc_pll_cs->min_pll_input_freq_khz = init_data->min_override_input_pxl_clk_pll_freq_khz; else calc_pll_cs->min_pll_input_freq_khz = - fw_info.pll_info.min_input_pxl_clk_pll_frequency; + fw_info->pll_info.min_input_pxl_clk_pll_frequency; calc_pll_cs->min_pix_clock_pll_post_divider = init_data->min_pix_clk_pll_post_divider; @@ -1316,7 +1314,6 @@ bool dce110_clk_src_construct( const struct dce110_clk_src_shift *cs_shift, const struct dce110_clk_src_mask *cs_mask) { - struct dc_firmware_info fw_info = { { 0 } }; struct calc_pll_clock_source_init_data calc_pll_cs_init_data_hdmi; struct calc_pll_clock_source_init_data calc_pll_cs_init_data; @@ -1329,14 +1326,12 @@ bool dce110_clk_src_construct( clk_src->cs_shift = cs_shift; clk_src->cs_mask = cs_mask; - if (clk_src->bios->funcs->get_firmware_info( - clk_src->bios, &fw_info) != BP_RESULT_OK) { + if (!clk_src->bios->fw_info_valid) { ASSERT_CRITICAL(false); goto unexpected_failure; } - clk_src->ext_clk_khz = - fw_info.external_clock_source_frequency_for_dp; + clk_src->ext_clk_khz = clk_src->bios->fw_info.external_clock_source_frequency_for_dp; /* structure normally used with PLL ranges from ATOMBIOS; DS on by default */ calc_pll_cs_init_data.bp = bios; @@ -1376,7 +1371,7 @@ bool dce110_clk_src_construct( FRACT_FB_DIVIDER_DEC_POINTS_MAX_NUM; calc_pll_cs_init_data_hdmi.ctx = ctx; - clk_src->ref_freq_khz = fw_info.pll_info.crystal_frequency; + clk_src->ref_freq_khz = clk_src->bios->fw_info.pll_info.crystal_frequency; if (clk_src->base.id == CLOCK_SOURCE_ID_EXTERNAL) return true; @@ -1419,8 +1414,6 @@ bool dce112_clk_src_construct( const struct dce110_clk_src_shift *cs_shift, const struct dce110_clk_src_mask *cs_mask) { - struct dc_firmware_info fw_info = { { 0 } }; - clk_src->base.ctx = ctx; clk_src->bios = bios; clk_src->base.id = id; @@ -1430,18 +1423,16 @@ bool dce112_clk_src_construct( clk_src->cs_shift = cs_shift; clk_src->cs_mask = cs_mask; - if (clk_src->bios->funcs->get_firmware_info( - clk_src->bios, &fw_info) != BP_RESULT_OK) { + if (!clk_src->bios->fw_info_valid) { ASSERT_CRITICAL(false); return false; } - clk_src->ext_clk_khz = fw_info.external_clock_source_frequency_for_dp; + clk_src->ext_clk_khz = clk_src->bios->fw_info.external_clock_source_frequency_for_dp; return true; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) bool dcn20_clk_src_construct( struct dce110_clk_src *clk_src, struct dc_context *ctx, @@ -1457,4 +1448,3 @@ bool dcn20_clk_src_construct( return ret; } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h index adae03b1f3a7..51bd25079606 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h @@ -55,7 +55,6 @@ CS_SF(PHYPLLA_PIXCLK_RESYNC_CNTL, PHYPLLA_DCCG_DEEP_COLOR_CNTL, mask_sh),\ CS_SF(PHYPLLA_PIXCLK_RESYNC_CNTL, PHYPLLA_PIXCLK_DOUBLE_RATE_ENABLE, mask_sh) -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define CS_COMMON_REG_LIST_DCN2_0(index, pllid) \ SRI(PIXCLK_RESYNC_CNTL, PHYPLL, pllid),\ SRII(PHASE, DP_DTO, 0),\ @@ -76,17 +75,29 @@ SRII(PIXEL_RATE_CNTL, OTG, 3),\ SRII(PIXEL_RATE_CNTL, OTG, 4),\ SRII(PIXEL_RATE_CNTL, OTG, 5) -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) +#define CS_COMMON_REG_LIST_DCN2_1(index, pllid) \ + SRI(PIXCLK_RESYNC_CNTL, PHYPLL, pllid),\ + SRII(PHASE, DP_DTO, 0),\ + SRII(PHASE, DP_DTO, 1),\ + SRII(PHASE, DP_DTO, 2),\ + SRII(PHASE, DP_DTO, 3),\ + SRII(MODULO, DP_DTO, 0),\ + SRII(MODULO, DP_DTO, 1),\ + SRII(MODULO, DP_DTO, 2),\ + SRII(MODULO, DP_DTO, 3),\ + SRII(PIXEL_RATE_CNTL, OTG, 0),\ + SRII(PIXEL_RATE_CNTL, OTG, 1),\ + SRII(PIXEL_RATE_CNTL, OTG, 2),\ + SRII(PIXEL_RATE_CNTL, OTG, 3) + #define CS_COMMON_MASK_SH_LIST_DCN2_0(mask_sh)\ CS_SF(DP_DTO0_PHASE, DP_DTO0_PHASE, mask_sh),\ CS_SF(DP_DTO0_MODULO, DP_DTO0_MODULO, mask_sh),\ CS_SF(PHYPLLA_PIXCLK_RESYNC_CNTL, PHYPLLA_DCCG_DEEP_COLOR_CNTL, mask_sh),\ CS_SF(OTG0_PIXEL_RATE_CNTL, DP_DTO0_ENABLE, mask_sh) -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) #define CS_COMMON_REG_LIST_DCN1_0(index, pllid) \ SRI(PIXCLK_RESYNC_CNTL, PHYPLL, pllid),\ @@ -184,7 +195,6 @@ bool dce112_clk_src_construct( const struct dce110_clk_src_shift *cs_shift, const struct dce110_clk_src_mask *cs_mask); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) bool dcn20_clk_src_construct( struct dce110_clk_src *clk_src, struct dc_context *ctx, @@ -193,6 +203,5 @@ bool dcn20_clk_src_construct( const struct dce110_clk_src_regs *regs, const struct dce110_clk_src_shift *cs_shift, const struct dce110_clk_src_mask *cs_mask); -#endif #endif diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c index 0b86cee4876f..30d953acd016 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c @@ -59,6 +59,12 @@ #define MCP_BL_SET_PWM_FRAC 0x6A /* Enable or disable Fractional PWM */ #define MASTER_COMM_CNTL_REG__MASTER_COMM_INTERRUPT_MASK 0x00000001L +// PSP FW version +#define mmMP0_SMN_C2PMSG_58 0x1607A + +//Register access policy version +#define mmMP0_SMN_C2PMSG_91 0x1609B + static bool dce_dmcu_init(struct dmcu *dmcu) { // Do nothing @@ -318,7 +324,7 @@ static void dce_get_psr_wait_loop( return; } -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) static void dcn10_get_dmcu_version(struct dmcu *dmcu) { struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu); @@ -373,6 +379,7 @@ static bool dcn10_dmcu_init(struct dmcu *dmcu) const struct dc_config *config = &dmcu->ctx->dc->config; bool status = false; + PERF_TRACE(); /* Definition of DC_DMCU_SCRATCH * 0 : firmare not loaded * 1 : PSP load DMCU FW but not initialized @@ -429,9 +436,21 @@ static bool dcn10_dmcu_init(struct dmcu *dmcu) break; } + PERF_TRACE(); return status; } +static bool dcn21_dmcu_init(struct dmcu *dmcu) +{ + struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu); + uint32_t dmcub_psp_version = REG_READ(DMCUB_SCRATCH15); + + if (dmcu->auto_load_dmcu && dmcub_psp_version == 0) { + return false; + } + + return dcn10_dmcu_init(dmcu); +} static bool dcn10_dmcu_load_iram(struct dmcu *dmcu, unsigned int start_offset, @@ -518,9 +537,6 @@ static void dcn10_dmcu_set_psr_enable(struct dmcu *dmcu, bool enable, bool wait) if (dmcu->dmcu_state != DMCU_RUNNING) return; - dcn10_get_dmcu_psr_state(dmcu, &psr_state); - if (psr_state == 0 && !enable) - return; /* waitDMCUReadyForCmd */ REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, 0, dmcu_wait_reg_ready_interval, @@ -727,9 +743,7 @@ static bool dcn10_is_dmcu_initialized(struct dmcu *dmcu) return true; } -#endif //(CONFIG_DRM_AMD_DC_DCN1_0) -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) static bool dcn20_lock_phy(struct dmcu *dmcu) { @@ -777,7 +791,7 @@ static bool dcn20_unlock_phy(struct dmcu *dmcu) return true; } -#endif //(CONFIG_DRM_AMD_DC_DCN2_0) +#endif //(CONFIG_DRM_AMD_DC_DCN) static const struct dmcu_funcs dce_funcs = { .dmcu_init = dce_dmcu_init, @@ -790,7 +804,7 @@ static const struct dmcu_funcs dce_funcs = { .is_dmcu_initialized = dce_is_dmcu_initialized }; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) static const struct dmcu_funcs dcn10_funcs = { .dmcu_init = dcn10_dmcu_init, .load_iram = dcn10_dmcu_load_iram, @@ -801,9 +815,7 @@ static const struct dmcu_funcs dcn10_funcs = { .get_psr_wait_loop = dcn10_get_psr_wait_loop, .is_dmcu_initialized = dcn10_is_dmcu_initialized }; -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) static const struct dmcu_funcs dcn20_funcs = { .dmcu_init = dcn10_dmcu_init, .load_iram = dcn10_dmcu_load_iram, @@ -816,6 +828,19 @@ static const struct dmcu_funcs dcn20_funcs = { .lock_phy = dcn20_lock_phy, .unlock_phy = dcn20_unlock_phy }; + +static const struct dmcu_funcs dcn21_funcs = { + .dmcu_init = dcn21_dmcu_init, + .load_iram = dcn10_dmcu_load_iram, + .set_psr_enable = dcn10_dmcu_set_psr_enable, + .setup_psr = dcn10_dmcu_setup_psr, + .get_psr_state = dcn10_get_dmcu_psr_state, + .set_psr_wait_loop = dcn10_psr_wait_loop, + .get_psr_wait_loop = dcn10_get_psr_wait_loop, + .is_dmcu_initialized = dcn10_is_dmcu_initialized, + .lock_phy = dcn20_lock_phy, + .unlock_phy = dcn20_unlock_phy +}; #endif static void dce_dmcu_construct( @@ -836,6 +861,26 @@ static void dce_dmcu_construct( dmcu_dce->dmcu_mask = dmcu_mask; } +#if defined(CONFIG_DRM_AMD_DC_DCN) +static void dcn21_dmcu_construct( + struct dce_dmcu *dmcu_dce, + struct dc_context *ctx, + const struct dce_dmcu_registers *regs, + const struct dce_dmcu_shift *dmcu_shift, + const struct dce_dmcu_mask *dmcu_mask) +{ + uint32_t psp_version = 0; + + dce_dmcu_construct(dmcu_dce, ctx, regs, dmcu_shift, dmcu_mask); + + if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { + psp_version = dm_read_reg(ctx, mmMP0_SMN_C2PMSG_58); + dmcu_dce->base.auto_load_dmcu = ((psp_version & 0x00FF00FF) > 0x00110029); + dmcu_dce->base.psp_version = psp_version; + } +} +#endif + struct dmcu *dce_dmcu_create( struct dc_context *ctx, const struct dce_dmcu_registers *regs, @@ -857,7 +902,7 @@ struct dmcu *dce_dmcu_create( return &dmcu_dce->base; } -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) struct dmcu *dcn10_dmcu_create( struct dc_context *ctx, const struct dce_dmcu_registers *regs, @@ -878,9 +923,7 @@ struct dmcu *dcn10_dmcu_create( return &dmcu_dce->base; } -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dmcu *dcn20_dmcu_create( struct dc_context *ctx, const struct dce_dmcu_registers *regs, @@ -901,15 +944,33 @@ struct dmcu *dcn20_dmcu_create( return &dmcu_dce->base; } + +struct dmcu *dcn21_dmcu_create( + struct dc_context *ctx, + const struct dce_dmcu_registers *regs, + const struct dce_dmcu_shift *dmcu_shift, + const struct dce_dmcu_mask *dmcu_mask) +{ + struct dce_dmcu *dmcu_dce = kzalloc(sizeof(*dmcu_dce), GFP_KERNEL); + + if (dmcu_dce == NULL) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dcn21_dmcu_construct( + dmcu_dce, ctx, regs, dmcu_shift, dmcu_mask); + + dmcu_dce->base.funcs = &dcn21_funcs; + + return &dmcu_dce->base; +} #endif void dce_dmcu_destroy(struct dmcu **dmcu) { struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(*dmcu); - if (dmcu_dce->base.dmcu_state == DMCU_RUNNING) - dmcu_dce->base.funcs->set_psr_enable(*dmcu, false, true); - kfree(dmcu_dce); *dmcu = NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h index cc8587683b4b..5e044c2d3d6d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.h @@ -71,6 +71,10 @@ DMCU_COMMON_REG_LIST_DCE_BASE(), \ SR(DMU_MEM_PWR_CNTL) +#define DMCU_DCN20_REG_LIST()\ + DMCU_DCN10_REG_LIST(), \ + SR(DMCUB_SCRATCH15) + #define DMCU_SF(reg_name, field_name, post_fix)\ .field_name = reg_name ## __ ## field_name ## post_fix @@ -175,6 +179,7 @@ struct dce_dmcu_registers { uint32_t DMCU_INTERRUPT_TO_UC_EN_MASK; uint32_t SMU_INTERRUPT_CONTROL; uint32_t DC_DMCU_SCRATCH; + uint32_t DMCUB_SCRATCH15; }; struct dce_dmcu { @@ -261,13 +266,17 @@ struct dmcu *dcn10_dmcu_create( const struct dce_dmcu_shift *dmcu_shift, const struct dce_dmcu_mask *dmcu_mask); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dmcu *dcn20_dmcu_create( struct dc_context *ctx, const struct dce_dmcu_registers *regs, const struct dce_dmcu_shift *dmcu_shift, const struct dce_dmcu_mask *dmcu_mask); -#endif + +struct dmcu *dcn21_dmcu_create( + struct dc_context *ctx, + const struct dce_dmcu_registers *regs, + const struct dce_dmcu_shift *dmcu_shift, + const struct dce_dmcu_mask *dmcu_mask); void dce_dmcu_destroy(struct dmcu **dmcu); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.c b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.c index 0275d6d60da4..e1c5839a80dc 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.c @@ -25,7 +25,7 @@ #include "dce_hwseq.h" #include "reg_helper.h" -#include "hw_sequencer.h" +#include "hw_sequencer_private.h" #include "core_types.h" #define CTX \ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h index cb0a037b1c4a..c5aa1f48593a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h @@ -25,7 +25,7 @@ #ifndef __DCE_HWSEQ_H__ #define __DCE_HWSEQ_H__ -#include "hw_sequencer.h" +#include "dc_types.h" #define BL_REG_LIST()\ SR(LVTMA_PWRSEQ_CNTL), \ @@ -62,6 +62,10 @@ SRII(BLND_CONTROL, BLND, 4), \ SRII(BLND_CONTROL, BLND, 5) +#define HSWEQ_DCN_PIXEL_RATE_REG_LIST(blk, inst) \ + SRII(PIXEL_RATE_CNTL, blk, inst), \ + SRII(PHYPLL_PIXEL_RATE_CNTL, blk, inst) + #define HWSEQ_PIXEL_RATE_REG_LIST(blk) \ SRII(PIXEL_RATE_CNTL, blk, 0), \ SRII(PIXEL_RATE_CNTL, blk, 1), \ @@ -151,7 +155,10 @@ SR(DCCG_GATE_DISABLE_CNTL2), \ SR(DCFCLK_CNTL),\ SR(DCFCLK_CNTL), \ - SR(DC_MEM_GLOBAL_PWR_REQ_CNTL), \ + SR(DC_MEM_GLOBAL_PWR_REQ_CNTL) + + +#define MMHUB_DCN_REG_LIST()\ /* todo: get these from GVM instead of reading registers ourselves */\ MMHUB_SR(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32),\ MMHUB_SR(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32),\ @@ -166,10 +173,14 @@ MMHUB_SR(MC_VM_SYSTEM_APERTURE_LOW_ADDR),\ MMHUB_SR(MC_VM_SYSTEM_APERTURE_HIGH_ADDR) + #define HWSEQ_DCN1_REG_LIST()\ HWSEQ_DCN_REG_LIST(), \ - HWSEQ_PIXEL_RATE_REG_LIST(OTG), \ - HWSEQ_PHYPLL_REG_LIST(OTG), \ + MMHUB_DCN_REG_LIST(), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 0), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 1), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 2), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 3), \ SR(DCHUBBUB_SDPIF_FB_BASE),\ SR(DCHUBBUB_SDPIF_FB_OFFSET),\ SR(DCHUBBUB_SDPIF_AGP_BASE),\ @@ -199,11 +210,14 @@ SR(DC_IP_REQUEST_CNTL), \ BL_REG_LIST() -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define HWSEQ_DCN2_REG_LIST()\ HWSEQ_DCN_REG_LIST(), \ - HWSEQ_PIXEL_RATE_REG_LIST(OTG), \ - HWSEQ_PHYPLL_REG_LIST(OTG), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 0), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 1), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 2), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 3), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 4), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 5), \ SR(MICROSECOND_TIME_BASE_DIV), \ SR(MILLISECOND_TIME_BASE_DIV), \ SR(DISPCLK_FREQ_CHANGE_CNTL), \ @@ -227,8 +241,8 @@ SR(DOMAIN7_PG_CONFIG), \ SR(DOMAIN8_PG_CONFIG), \ SR(DOMAIN9_PG_CONFIG), \ - SR(DOMAIN10_PG_CONFIG), \ - SR(DOMAIN11_PG_CONFIG), \ +/* SR(DOMAIN10_PG_CONFIG), Navi1x HUBP5 not powergate-able*/\ +/* SR(DOMAIN11_PG_CONFIG), Navi1x DPP5 is not powergate-able */\ SR(DOMAIN16_PG_CONFIG), \ SR(DOMAIN17_PG_CONFIG), \ SR(DOMAIN18_PG_CONFIG), \ @@ -261,7 +275,57 @@ SR(D6VGA_CONTROL), \ SR(DC_IP_REQUEST_CNTL), \ BL_REG_LIST() -#endif + +#define HWSEQ_DCN21_REG_LIST()\ + HWSEQ_DCN_REG_LIST(), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 0), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 1), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 2), \ + HSWEQ_DCN_PIXEL_RATE_REG_LIST(OTG, 3), \ + MMHUB_DCN_REG_LIST(), \ + SR(MICROSECOND_TIME_BASE_DIV), \ + SR(MILLISECOND_TIME_BASE_DIV), \ + SR(DISPCLK_FREQ_CHANGE_CNTL), \ + SR(RBBMIF_TIMEOUT_DIS), \ + SR(RBBMIF_TIMEOUT_DIS_2), \ + SR(DCHUBBUB_CRC_CTRL), \ + SR(DPP_TOP0_DPP_CRC_CTRL), \ + SR(DPP_TOP0_DPP_CRC_VAL_B_A), \ + SR(DPP_TOP0_DPP_CRC_VAL_R_G), \ + SR(MPC_CRC_CTRL), \ + SR(MPC_CRC_RESULT_GB), \ + SR(MPC_CRC_RESULT_C), \ + SR(MPC_CRC_RESULT_AR), \ + SR(DOMAIN0_PG_CONFIG), \ + SR(DOMAIN1_PG_CONFIG), \ + SR(DOMAIN2_PG_CONFIG), \ + SR(DOMAIN3_PG_CONFIG), \ + SR(DOMAIN4_PG_CONFIG), \ + SR(DOMAIN5_PG_CONFIG), \ + SR(DOMAIN6_PG_CONFIG), \ + SR(DOMAIN7_PG_CONFIG), \ + SR(DOMAIN16_PG_CONFIG), \ + SR(DOMAIN17_PG_CONFIG), \ + SR(DOMAIN18_PG_CONFIG), \ + SR(DOMAIN0_PG_STATUS), \ + SR(DOMAIN1_PG_STATUS), \ + SR(DOMAIN2_PG_STATUS), \ + SR(DOMAIN3_PG_STATUS), \ + SR(DOMAIN4_PG_STATUS), \ + SR(DOMAIN5_PG_STATUS), \ + SR(DOMAIN6_PG_STATUS), \ + SR(DOMAIN7_PG_STATUS), \ + SR(DOMAIN16_PG_STATUS), \ + SR(DOMAIN17_PG_STATUS), \ + SR(DOMAIN18_PG_STATUS), \ + SR(D1VGA_CONTROL), \ + SR(D2VGA_CONTROL), \ + SR(D3VGA_CONTROL), \ + SR(D4VGA_CONTROL), \ + SR(D5VGA_CONTROL), \ + SR(D6VGA_CONTROL), \ + SR(DC_IP_REQUEST_CNTL), \ + BL_REG_LIST() struct dce_hwseq_registers { @@ -401,36 +465,34 @@ struct dce_hwseq_registers { HWS_SF1(blk, PHYPLL_PIXEL_RATE_CNTL, PHYPLL_PIXEL_RATE_SOURCE, mask_sh),\ HWS_SF1(blk, PHYPLL_PIXEL_RATE_CNTL, PIXEL_RATE_PLL_SOURCE, mask_sh) +#define HWSEQ_LVTMA_MASK_SH_LIST(mask_sh)\ + HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ + HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, mask_sh),\ + HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON_OVRD, mask_sh),\ + HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + #define HWSEQ_DCE8_MASK_SH_LIST(mask_sh)\ .DCFE_CLOCK_ENABLE = CRTC_DCFE_CLOCK_CONTROL__CRTC_DCFE_CLOCK_ENABLE ## mask_sh, \ HWS_SF(BLND_, V_UPDATE_LOCK, BLND_DCP_GRPH_V_UPDATE_LOCK, mask_sh),\ HWS_SF(BLND_, V_UPDATE_LOCK, BLND_SCL_V_UPDATE_LOCK, mask_sh),\ HWS_SF(BLND_, V_UPDATE_LOCK, BLND_DCP_GRPH_SURF_V_UPDATE_LOCK, mask_sh),\ HWS_SF(BLND_, CONTROL, BLND_MODE, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh),\ - HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_) + HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_),\ + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) #define HWSEQ_DCE10_MASK_SH_LIST(mask_sh)\ HWSEQ_DCEF_MASK_SH_LIST(mask_sh, DCFE_),\ HWSEQ_BLND_MASK_SH_LIST(mask_sh, BLND_),\ - HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_),\ + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) #define HWSEQ_DCE11_MASK_SH_LIST(mask_sh)\ HWSEQ_DCE10_MASK_SH_LIST(mask_sh),\ SF(DCFEV_CLOCK_CONTROL, DCFEV_CLOCK_ENABLE, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON_OVRD, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh),\ HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_) #define HWSEQ_DCE112_MASK_SH_LIST(mask_sh)\ HWSEQ_DCE10_MASK_SH_LIST(mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh),\ HWSEQ_PHYPLL_MASK_SH_LIST(mask_sh, CRTC0_) #define HWSEQ_GFX9_DCHUB_MASK_SH_LIST(mask_sh)\ @@ -438,18 +500,15 @@ struct dce_hwseq_registers { SF(DCHUB_FB_LOCATION, FB_BASE, mask_sh),\ SF(DCHUB_AGP_BASE, AGP_BASE, mask_sh),\ SF(DCHUB_AGP_BOT, AGP_BOT, mask_sh),\ - SF(DCHUB_AGP_TOP, AGP_TOP, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + SF(DCHUB_AGP_TOP, AGP_TOP, mask_sh) #define HWSEQ_DCE12_MASK_SH_LIST(mask_sh)\ HWSEQ_DCEF_MASK_SH_LIST(mask_sh, DCFE0_DCFE_),\ HWSEQ_BLND_MASK_SH_LIST(mask_sh, BLND0_BLND_),\ HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_),\ HWSEQ_PHYPLL_MASK_SH_LIST(mask_sh, CRTC0_),\ - HWSEQ_GFX9_DCHUB_MASK_SH_LIST(mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + HWSEQ_GFX9_DCHUB_MASK_SH_LIST(mask_sh),\ + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) #define HWSEQ_VG20_MASK_SH_LIST(mask_sh)\ HWSEQ_DCE12_MASK_SH_LIST(mask_sh),\ @@ -512,12 +571,8 @@ struct dce_hwseq_registers { HWS_SF(, D4VGA_CONTROL, D4VGA_MODE_ENABLE, mask_sh),\ HWS_SF(, VGA_TEST_CONTROL, VGA_TEST_ENABLE, mask_sh),\ HWS_SF(, VGA_TEST_CONTROL, VGA_TEST_RENDER_START, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON_OVRD, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define HWSEQ_DCN2_MASK_SH_LIST(mask_sh)\ HWSEQ_DCN_MASK_SH_LIST(mask_sh), \ HWS_SF(, DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, mask_sh), \ @@ -576,9 +631,50 @@ struct dce_hwseq_registers { HWS_SF(, DOMAIN20_PG_STATUS, DOMAIN20_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DOMAIN21_PG_STATUS, DOMAIN21_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DC_IP_REQUEST_CNTL, IP_REQUEST_EN, mask_sh), \ + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) + +#define HWSEQ_DCN21_MASK_SH_LIST(mask_sh)\ + HWSEQ_DCN_MASK_SH_LIST(mask_sh), \ + HWS_SF(, DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, mask_sh), \ + HWS_SF(, MMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32, PAGE_DIRECTORY_ENTRY_HI32, mask_sh),\ + HWS_SF(, MMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, PAGE_DIRECTORY_ENTRY_LO32, mask_sh),\ + HWS_SF(, DOMAIN0_PG_CONFIG, DOMAIN0_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN0_PG_CONFIG, DOMAIN0_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN1_PG_CONFIG, DOMAIN1_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN1_PG_CONFIG, DOMAIN1_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN2_PG_CONFIG, DOMAIN2_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN2_PG_CONFIG, DOMAIN2_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN3_PG_CONFIG, DOMAIN3_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN3_PG_CONFIG, DOMAIN3_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN4_PG_CONFIG, DOMAIN4_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN4_PG_CONFIG, DOMAIN4_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN5_PG_CONFIG, DOMAIN5_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN5_PG_CONFIG, DOMAIN5_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN6_PG_CONFIG, DOMAIN6_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN6_PG_CONFIG, DOMAIN6_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN7_PG_CONFIG, DOMAIN7_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN7_PG_CONFIG, DOMAIN7_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN16_PG_CONFIG, DOMAIN16_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN16_PG_CONFIG, DOMAIN16_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN17_PG_CONFIG, DOMAIN17_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN17_PG_CONFIG, DOMAIN17_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN18_PG_CONFIG, DOMAIN18_POWER_FORCEON, mask_sh), \ + HWS_SF(, DOMAIN18_PG_CONFIG, DOMAIN18_POWER_GATE, mask_sh), \ + HWS_SF(, DOMAIN0_PG_STATUS, DOMAIN0_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN1_PG_STATUS, DOMAIN1_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN2_PG_STATUS, DOMAIN2_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN3_PG_STATUS, DOMAIN3_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN4_PG_STATUS, DOMAIN4_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN5_PG_STATUS, DOMAIN5_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN6_PG_STATUS, DOMAIN6_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN7_PG_STATUS, DOMAIN7_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN16_PG_STATUS, DOMAIN16_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN17_PG_STATUS, DOMAIN17_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DOMAIN18_PG_STATUS, DOMAIN18_PGFSM_PWR_STATUS, mask_sh), \ + HWS_SF(, DC_IP_REQUEST_CNTL, IP_REQUEST_EN, mask_sh), \ + HWSEQ_LVTMA_MASK_SH_LIST(mask_sh), \ HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) -#endif #define HWSEQ_REG_FIELD_LIST(type) \ type DCFE_CLOCK_ENABLE; \ @@ -612,9 +708,9 @@ struct dce_hwseq_registers { type ENABLE_L1_TLB;\ type SYSTEM_ACCESS_MODE;\ type LVTMA_BLON;\ - type LVTMA_PWRSEQ_TARGET_STATE_R;\ type LVTMA_DIGON;\ - type LVTMA_DIGON_OVRD; + type LVTMA_DIGON_OVRD;\ + type LVTMA_PWRSEQ_TARGET_STATE_R; #define HWSEQ_DCN_REG_FIELD_LIST(type) \ type HUBP_VTG_SEL; \ @@ -715,6 +811,10 @@ enum blnd_mode { BLND_MODE_BLENDING,/* Alpha blending - blend 'current' and 'other' */ }; +struct dce_hwseq; +struct pipe_ctx; +struct clock_source; + void dce_enable_fe_clock(struct dce_hwseq *hwss, unsigned int inst, bool enable); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.c index 35a75398fcb4..dd41736bb5c4 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c.c @@ -31,7 +31,7 @@ bool dce_i2c_submit_command( struct i2c_command *cmd) { struct dce_i2c_hw *dce_i2c_hw; - struct dce_i2c_sw *dce_i2c_sw; + struct dce_i2c_sw dce_i2c_sw = {0}; if (!ddc) { BREAK_TO_DEBUGGER(); @@ -43,18 +43,15 @@ bool dce_i2c_submit_command( return false; } - /* The software engine is only available on dce8 */ - dce_i2c_sw = dce_i2c_acquire_i2c_sw_engine(pool, ddc); - - if (!dce_i2c_sw) { - dce_i2c_hw = acquire_i2c_hw_engine(pool, ddc); - - if (!dce_i2c_hw) - return false; + dce_i2c_hw = acquire_i2c_hw_engine(pool, ddc); + if (dce_i2c_hw) return dce_i2c_submit_command_hw(pool, ddc, cmd, dce_i2c_hw); - } - return dce_i2c_submit_command_sw(pool, ddc, cmd, dce_i2c_sw); + dce_i2c_sw.ctx = ddc->ctx; + if (dce_i2c_engine_acquire_sw(&dce_i2c_sw, ddc)) { + return dce_i2c_submit_command_sw(pool, ddc, cmd, &dce_i2c_sw); + } + return false; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c index a9061aaf1562..066188ba7949 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c @@ -25,6 +25,7 @@ #include <linux/delay.h> +#include "resource.h" #include "dce_i2c.h" #include "dce_i2c_hw.h" #include "reg_helper.h" @@ -99,31 +100,6 @@ static uint32_t get_hw_buffer_available_size( dce_i2c_hw->buffer_used_bytes; } -uint32_t get_reference_clock( - struct dc_bios *bios) -{ - struct dc_firmware_info info = { { 0 } }; - - if (bios->funcs->get_firmware_info(bios, &info) != BP_RESULT_OK) - return 0; - - return info.pll_info.crystal_frequency; -} - -static uint32_t get_speed( - const struct dce_i2c_hw *dce_i2c_hw) -{ - uint32_t pre_scale = 0; - - REG_GET(SPEED, DC_I2C_DDC1_PRESCALE, &pre_scale); - - /* [anaumov] it seems following is unnecessary */ - /*ASSERT(value.bits.DC_I2C_DDC1_PRESCALE);*/ - return pre_scale ? - dce_i2c_hw->reference_frequency / pre_scale : - dce_i2c_hw->default_speed; -} - static void process_channel_reply( struct dce_i2c_hw *dce_i2c_hw, struct i2c_payload *reply) @@ -288,16 +264,25 @@ static void set_speed( struct dce_i2c_hw *dce_i2c_hw, uint32_t speed) { + uint32_t xtal_ref_div = 0; + uint32_t prescale = 0; + + REG_GET(MICROSECOND_TIME_BASE_DIV, XTAL_REF_DIV, &xtal_ref_div); + + if (xtal_ref_div == 0) + xtal_ref_div = 2; + + prescale = ((dce_i2c_hw->reference_frequency * 2) / xtal_ref_div) / speed; if (speed) { if (dce_i2c_hw->masks->DC_I2C_DDC1_START_STOP_TIMING_CNTL) REG_UPDATE_N(SPEED, 3, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), dce_i2c_hw->reference_frequency / speed, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2, FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_START_STOP_TIMING_CNTL), speed > 50 ? 2:1); else REG_UPDATE_N(SPEED, 2, - FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), dce_i2c_hw->reference_frequency / speed, + FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_PRESCALE), prescale, FN(DC_I2C_DDC1_SPEED, DC_I2C_DDC1_THRESHOLD), 2); } } @@ -306,9 +291,7 @@ static bool setup_engine( struct dce_i2c_hw *dce_i2c_hw) { uint32_t i2c_setup_limit = I2C_SETUP_TIME_LIMIT_DCE; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) uint32_t reset_length = 0; -#endif /* we have checked I2c not used by DMCU, set SW use I2C REQ to 1 to indicate SW using it*/ REG_UPDATE(DC_I2C_ARBITRATION, DC_I2C_SW_USE_I2C_REG_REQ, 1); @@ -332,14 +315,12 @@ static bool setup_engine( REG_UPDATE_N(SETUP, 2, FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_TIME_LIMIT), i2c_setup_limit, FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_ENABLE), 1); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) } else { reset_length = dce_i2c_hw->send_reset_length; REG_UPDATE_N(SETUP, 3, FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_TIME_LIMIT), i2c_setup_limit, FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_SEND_RESET_LENGTH), reset_length, FN(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_ENABLE), 1); -#endif } /* Program HW priority * set to High - interrupt software I2C at any time @@ -358,9 +339,7 @@ static void release_engine( bool safe_to_reset; /* Restore original HW engine speed */ - - set_speed(dce_i2c_hw, dce_i2c_hw->original_speed); - + set_speed(dce_i2c_hw, dce_i2c_hw->default_speed); /* Reset HW engine */ { @@ -392,7 +371,6 @@ struct dce_i2c_hw *acquire_i2c_hw_engine( { uint32_t counter = 0; enum gpio_result result; - uint32_t current_speed; struct dce_i2c_hw *dce_i2c_hw = NULL; if (!ddc) @@ -401,7 +379,7 @@ struct dce_i2c_hw *acquire_i2c_hw_engine( if (ddc->hw_info.hw_supported) { enum gpio_ddc_line line = dal_ddc_get_line(ddc); - if (line < pool->pipe_count) + if (line < pool->res_cap->num_ddc) dce_i2c_hw = pool->hw_i2cs[line]; } @@ -430,11 +408,6 @@ struct dce_i2c_hw *acquire_i2c_hw_engine( dce_i2c_hw->ddc = ddc; - current_speed = get_speed(dce_i2c_hw); - - if (current_speed) - dce_i2c_hw->original_speed = current_speed; - if (!setup_engine(dce_i2c_hw)) { release_engine(dce_i2c_hw); return NULL; @@ -492,13 +465,9 @@ static void submit_channel_request_hw( static uint32_t get_transaction_timeout_hw( const struct dce_i2c_hw *dce_i2c_hw, - uint32_t length) + uint32_t length, + uint32_t speed) { - - uint32_t speed = get_speed(dce_i2c_hw); - - - uint32_t period_timeout; uint32_t num_of_clock_stretches; @@ -518,7 +487,8 @@ static uint32_t get_transaction_timeout_hw( bool dce_i2c_hw_engine_submit_payload( struct dce_i2c_hw *dce_i2c_hw, struct i2c_payload *payload, - bool middle_of_transaction) + bool middle_of_transaction, + uint32_t speed) { struct i2c_request_transaction_data request; @@ -556,7 +526,7 @@ bool dce_i2c_hw_engine_submit_payload( /* obtain timeout value before submitting request */ transaction_timeout = get_transaction_timeout_hw( - dce_i2c_hw, payload->length + 1); + dce_i2c_hw, payload->length + 1, speed); submit_channel_request_hw( dce_i2c_hw, &request); @@ -602,13 +572,11 @@ bool dce_i2c_submit_command_hw( struct i2c_payload *payload = cmd->payloads + index_of_payload; if (!dce_i2c_hw_engine_submit_payload( - dce_i2c_hw, payload, mot)) { + dce_i2c_hw, payload, mot, cmd->speed)) { result = false; break; } - - ++index_of_payload; } @@ -632,14 +600,13 @@ void dce_i2c_hw_construct( { dce_i2c_hw->ctx = ctx; dce_i2c_hw->engine_id = engine_id; - dce_i2c_hw->reference_frequency = get_reference_clock(ctx->dc_bios) >> 1; + dce_i2c_hw->reference_frequency = (ctx->dc_bios->fw_info.pll_info.crystal_frequency) >> 1; dce_i2c_hw->regs = regs; dce_i2c_hw->shifts = shifts; dce_i2c_hw->masks = masks; dce_i2c_hw->buffer_used_bytes = 0; dce_i2c_hw->transaction_count = 0; dce_i2c_hw->engine_keep_power_up_count = 1; - dce_i2c_hw->original_speed = DEFAULT_I2C_HW_SPEED; dce_i2c_hw->default_speed = DEFAULT_I2C_HW_SPEED; dce_i2c_hw->send_reset_length = 0; dce_i2c_hw->setup_limit = I2C_SETUP_TIME_LIMIT_DCE; @@ -654,9 +621,6 @@ void dce100_i2c_hw_construct( const struct dce_i2c_shift *shifts, const struct dce_i2c_mask *masks) { - - uint32_t xtal_ref_div = 0; - dce_i2c_hw_construct(dce_i2c_hw, ctx, engine_id, @@ -664,21 +628,6 @@ void dce100_i2c_hw_construct( shifts, masks); dce_i2c_hw->buffer_size = I2C_HW_BUFFER_SIZE_DCE100; - - REG_GET(MICROSECOND_TIME_BASE_DIV, XTAL_REF_DIV, &xtal_ref_div); - - if (xtal_ref_div == 0) - xtal_ref_div = 2; - - /*Calculating Reference Clock by divding original frequency by - * XTAL_REF_DIV. - * At upper level, uint32_t reference_frequency = - * dal_dce_i2c_get_reference_clock(as) >> 1 - * which already divided by 2. So we need x2 to get original - * reference clock from ppll_info - */ - dce_i2c_hw->reference_frequency = - (dce_i2c_hw->reference_frequency * 2) / xtal_ref_div; } void dce112_i2c_hw_construct( @@ -715,7 +664,6 @@ void dcn1_i2c_hw_construct( dce_i2c_hw->setup_limit = I2C_SETUP_TIME_LIMIT_DCN; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void dcn2_i2c_hw_construct( struct dce_i2c_hw *dce_i2c_hw, struct dc_context *ctx, @@ -734,4 +682,3 @@ void dcn2_i2c_hw_construct( if (ctx->dc->debug.scl_reset_length10) dce_i2c_hw->send_reset_length = I2C_SEND_RESET_LENGTH_10; } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.h b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.h index cb0234e5d597..fb055e6883c0 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.h @@ -177,9 +177,7 @@ struct dce_i2c_shift { uint8_t DC_I2C_INDEX; uint8_t DC_I2C_INDEX_WRITE; uint8_t XTAL_REF_DIV; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) uint8_t DC_I2C_DDC1_SEND_RESET_LENGTH; -#endif uint8_t DC_I2C_REG_RW_CNTL_STATUS; }; @@ -220,17 +218,13 @@ struct dce_i2c_mask { uint32_t DC_I2C_INDEX; uint32_t DC_I2C_INDEX_WRITE; uint32_t XTAL_REF_DIV; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) uint32_t DC_I2C_DDC1_SEND_RESET_LENGTH; -#endif uint32_t DC_I2C_REG_RW_CNTL_STATUS; }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define I2C_COMMON_MASK_SH_LIST_DCN2(mask_sh)\ I2C_COMMON_MASK_SH_LIST_DCE110(mask_sh),\ I2C_SF(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_SEND_RESET_LENGTH, mask_sh) -#endif struct dce_i2c_registers { uint32_t SETUP; @@ -262,7 +256,6 @@ struct i2c_request_transaction_data { struct dce_i2c_hw { struct ddc *ddc; - uint32_t original_speed; uint32_t engine_keep_power_up_count; uint32_t transaction_count; uint32_t buffer_used_bytes; @@ -312,7 +305,6 @@ void dcn1_i2c_hw_construct( const struct dce_i2c_shift *shifts, const struct dce_i2c_mask *masks); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void dcn2_i2c_hw_construct( struct dce_i2c_hw *dce_i2c_hw, struct dc_context *ctx, @@ -320,7 +312,6 @@ void dcn2_i2c_hw_construct( const struct dce_i2c_registers *regs, const struct dce_i2c_shift *shifts, const struct dce_i2c_mask *masks); -#endif bool dce_i2c_submit_command_hw( struct resource_pool *pool, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c index a5a11c251e25..87d8428df6c4 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c @@ -73,31 +73,6 @@ static void release_engine_dce_sw( dce_i2c_sw->ddc = NULL; } -static bool get_hw_supported_ddc_line( - struct ddc *ddc, - enum gpio_ddc_line *line) -{ - enum gpio_ddc_line line_found; - - *line = GPIO_DDC_LINE_UNKNOWN; - - if (!ddc) { - BREAK_TO_DEBUGGER(); - return false; - } - - if (!ddc->hw_info.hw_supported) - return false; - - line_found = dal_ddc_get_line(ddc); - - if (line_found >= GPIO_DDC_LINE_COUNT) - return false; - - *line = line_found; - - return true; -} static bool wait_for_scl_high_sw( struct dc_context *ctx, struct ddc *ddc, @@ -524,21 +499,3 @@ bool dce_i2c_submit_command_sw( return result; } -struct dce_i2c_sw *dce_i2c_acquire_i2c_sw_engine( - struct resource_pool *pool, - struct ddc *ddc) -{ - enum gpio_ddc_line line; - struct dce_i2c_sw *engine = NULL; - - if (get_hw_supported_ddc_line(ddc, &line)) - engine = pool->sw_i2cs[line]; - - if (!engine) - return NULL; - - if (!dce_i2c_engine_acquire_sw(engine, ddc)) - return NULL; - - return engine; -} diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.h b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.h index 5bbcdd455614..019fc47bb767 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.h @@ -49,9 +49,9 @@ bool dce_i2c_submit_command_sw( struct i2c_command *cmd, struct dce_i2c_sw *dce_i2c_sw); -struct dce_i2c_sw *dce_i2c_acquire_i2c_sw_engine( - struct resource_pool *pool, - struct ddc *ddc); +bool dce_i2c_engine_acquire_sw( + struct dce_i2c_sw *dce_i2c_sw, + struct ddc *ddc_handle); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c index a24a2bda8656..8aa937f496c4 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_mem_input.c @@ -148,7 +148,7 @@ static void dce_mi_program_pte_vm( pte->min_pte_before_flip_horiz_scan; REG_UPDATE(GRPH_PIPE_OUTSTANDING_REQUEST_LIMIT, - GRPH_PIPE_OUTSTANDING_REQUEST_LIMIT, 0xff); + GRPH_PIPE_OUTSTANDING_REQUEST_LIMIT, 0x7f); REG_UPDATE_3(DVMM_PTE_CONTROL, DVMM_PAGE_WIDTH, page_width, @@ -157,7 +157,7 @@ static void dce_mi_program_pte_vm( REG_UPDATE_2(DVMM_PTE_ARB_CONTROL, DVMM_PTE_REQ_PER_CHUNK, pte->pte_req_per_chunk, - DVMM_MAX_PTE_REQ_OUTSTANDING, 0xff); + DVMM_MAX_PTE_REQ_OUTSTANDING, 0x7f); } static void program_urgency_watermark( @@ -391,10 +391,10 @@ static void program_tiling( static void program_size_and_rotation( struct dce_mem_input *dce_mi, enum dc_rotation_angle rotation, - const union plane_size *plane_size) + const struct plane_size *plane_size) { - const struct rect *in_rect = &plane_size->grph.surface_size; - struct rect hw_rect = plane_size->grph.surface_size; + const struct rect *in_rect = &plane_size->surface_size; + struct rect hw_rect = plane_size->surface_size; const uint32_t rotation_angles[ROTATION_ANGLE_COUNT] = { [ROTATION_ANGLE_0] = 0, [ROTATION_ANGLE_90] = 1, @@ -423,7 +423,7 @@ static void program_size_and_rotation( GRPH_Y_END, hw_rect.height); REG_SET(GRPH_PITCH, 0, - GRPH_PITCH, plane_size->grph.surface_pitch); + GRPH_PITCH, plane_size->surface_pitch); REG_SET(HW_ROTATION, 0, GRPH_ROTATION_ANGLE, rotation_angles[rotation]); @@ -505,7 +505,7 @@ static void dce_mi_program_surface_config( struct mem_input *mi, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror) @@ -606,11 +606,11 @@ static void dce_mi_allocate_dmif( } if (dce_mi->wa.single_head_rdreq_dmif_limit) { - uint32_t eanble = (total_stream_num > 1) ? 0 : + uint32_t enable = (total_stream_num > 1) ? 0 : dce_mi->wa.single_head_rdreq_dmif_limit; REG_UPDATE(MC_HUB_RDREQ_DMIF_LIMIT, - ENABLE, eanble); + ENABLE, enable); } } @@ -636,11 +636,11 @@ static void dce_mi_free_dmif( 10, 3500); if (dce_mi->wa.single_head_rdreq_dmif_limit) { - uint32_t eanble = (total_stream_num > 1) ? 0 : + uint32_t enable = (total_stream_num > 1) ? 0 : dce_mi->wa.single_head_rdreq_dmif_limit; REG_UPDATE(MC_HUB_RDREQ_DMIF_LIMIT, - ENABLE, eanble); + ENABLE, enable); } } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c index 5e2b4d47c548..451574971b96 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c @@ -137,7 +137,7 @@ static void dce110_update_generic_info_packet( AFMT_GENERIC0_UPDATE, (packet_index == 0), AFMT_GENERIC2_UPDATE, (packet_index == 2)); } -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) if (REG(AFMT_VBI_PACKET_CONTROL1)) { switch (packet_index) { case 0: @@ -231,7 +231,7 @@ static void dce110_update_hdmi_info_packet( HDMI_GENERIC1_SEND, send, HDMI_GENERIC1_LINE, line); break; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) case 4: if (REG(HDMI_GENERIC_PACKET_CONTROL2)) REG_UPDATE_3(HDMI_GENERIC_PACKET_CONTROL2, @@ -275,9 +275,10 @@ static void dce110_stream_encoder_dp_set_stream_attribute( struct stream_encoder *enc, struct dc_crtc_timing *crtc_timing, enum dc_color_space output_color_space, + bool use_vsc_sdp_for_colorimetry, uint32_t enable_sdp_splitting) { -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) uint32_t h_active_start; uint32_t v_active_start; uint32_t misc0 = 0; @@ -329,7 +330,7 @@ static void dce110_stream_encoder_dp_set_stream_attribute( if (enc110->se_mask->DP_VID_M_DOUBLE_VALUE_EN) REG_UPDATE(DP_VID_TIMING, DP_VID_M_DOUBLE_VALUE_EN, 1); -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) if (enc110->se_mask->DP_VID_N_MUL) REG_UPDATE(DP_VID_TIMING, DP_VID_N_MUL, 1); #endif @@ -340,7 +341,7 @@ static void dce110_stream_encoder_dp_set_stream_attribute( break; } -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) if (REG(DP_MSA_MISC)) misc1 = REG_READ(DP_MSA_MISC); #endif @@ -374,7 +375,7 @@ static void dce110_stream_encoder_dp_set_stream_attribute( /* set dynamic range and YCbCr range */ -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) switch (hw_crtc_timing.display_color_depth) { case COLOR_DEPTH_666: colorimetry_bpc = 0; @@ -454,7 +455,7 @@ static void dce110_stream_encoder_dp_set_stream_attribute( DP_DYN_RANGE, dynamic_range_rgb, DP_YCBCR_RANGE, dynamic_range_ycbcr); -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) if (REG(DP_MSA_COLORIMETRY)) REG_SET(DP_MSA_COLORIMETRY, 0, DP_MSA_MISC0, misc0); @@ -489,7 +490,7 @@ static void dce110_stream_encoder_dp_set_stream_attribute( hw_crtc_timing.v_front_porch; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) /* start at begining of left border */ if (REG(DP_MSA_TIMING_PARAM2)) REG_SET_2(DP_MSA_TIMING_PARAM2, 0, @@ -786,7 +787,7 @@ static void dce110_stream_encoder_update_hdmi_info_packets( dce110_update_hdmi_info_packet(enc110, 3, &info_frame->hdrsmd); } -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) if (enc110->se_mask->HDMI_DB_DISABLE) { /* for bring up, disable dp double TODO */ if (REG(HDMI_DB_CONTROL)) @@ -824,7 +825,7 @@ static void dce110_stream_encoder_stop_hdmi_info_packets( HDMI_GENERIC1_LINE, 0, HDMI_GENERIC1_SEND, 0); -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) /* stop generic packets 2 & 3 on HDMI */ if (REG(HDMI_GENERIC_PACKET_CONTROL2)) REG_SET_6(HDMI_GENERIC_PACKET_CONTROL2, 0, @@ -1038,6 +1039,24 @@ static void dce110_stream_encoder_set_avmute( } +static void dce110_reset_hdmi_stream_attribute( + struct stream_encoder *enc) +{ + struct dce110_stream_encoder *enc110 = DCE110STRENC_FROM_STRENC(enc); + if (enc110->se_mask->HDMI_DATA_SCRAMBLE_EN) + REG_UPDATE_5(HDMI_CONTROL, + HDMI_PACKET_GEN_VERSION, 1, + HDMI_KEEPOUT_MODE, 1, + HDMI_DEEP_COLOR_ENABLE, 0, + HDMI_DATA_SCRAMBLE_EN, 0, + HDMI_CLOCK_CHANNEL_RATE, 0); + else + REG_UPDATE_3(HDMI_CONTROL, + HDMI_PACKET_GEN_VERSION, 1, + HDMI_KEEPOUT_MODE, 1, + HDMI_DEEP_COLOR_ENABLE, 0); +} + #define DP_SEC_AUD_N__DP_SEC_AUD_N__DEFAULT 0x8000 #define DP_SEC_TIMESTAMP__DP_SEC_TIMESTAMP_MODE__AUTO_CALC 1 @@ -1251,13 +1270,13 @@ static uint32_t calc_max_audio_packets_per_line( static void get_audio_clock_info( enum dc_color_depth color_depth, - uint32_t crtc_pixel_clock_in_khz, - uint32_t actual_pixel_clock_in_khz, + uint32_t crtc_pixel_clock_100Hz, + uint32_t actual_pixel_clock_100Hz, struct audio_clock_info *audio_clock_info) { const struct audio_clock_info *clock_info; uint32_t index; - uint32_t crtc_pixel_clock_in_10khz = crtc_pixel_clock_in_khz / 10; + uint32_t crtc_pixel_clock_in_10khz = crtc_pixel_clock_100Hz / 100; uint32_t audio_array_size; switch (color_depth) { @@ -1294,16 +1313,16 @@ static void get_audio_clock_info( } /* not found */ - if (actual_pixel_clock_in_khz == 0) - actual_pixel_clock_in_khz = crtc_pixel_clock_in_khz; + if (actual_pixel_clock_100Hz == 0) + actual_pixel_clock_100Hz = crtc_pixel_clock_100Hz; /* See HDMI spec the table entry under * pixel clock of "Other". */ audio_clock_info->pixel_clock_in_10khz = - actual_pixel_clock_in_khz / 10; - audio_clock_info->cts_32khz = actual_pixel_clock_in_khz; - audio_clock_info->cts_44khz = actual_pixel_clock_in_khz; - audio_clock_info->cts_48khz = actual_pixel_clock_in_khz; + actual_pixel_clock_100Hz / 100; + audio_clock_info->cts_32khz = actual_pixel_clock_100Hz / 10; + audio_clock_info->cts_44khz = actual_pixel_clock_100Hz / 10; + audio_clock_info->cts_48khz = actual_pixel_clock_100Hz / 10; audio_clock_info->n_32khz = 4096; audio_clock_info->n_44khz = 6272; @@ -1369,14 +1388,14 @@ static void dce110_se_setup_hdmi_audio( /* Program audio clock sample/regeneration parameters */ get_audio_clock_info(crtc_info->color_depth, - crtc_info->requested_pixel_clock, - crtc_info->calculated_pixel_clock, + crtc_info->requested_pixel_clock_100Hz, + crtc_info->calculated_pixel_clock_100Hz, &audio_clock_info); DC_LOG_HW_AUDIO( - "\n%s:Input::requested_pixel_clock = %d" \ - "calculated_pixel_clock = %d \n", __func__, \ - crtc_info->requested_pixel_clock, \ - crtc_info->calculated_pixel_clock); + "\n%s:Input::requested_pixel_clock_100Hz = %d" \ + "calculated_pixel_clock_100Hz = %d \n", __func__, \ + crtc_info->requested_pixel_clock_100Hz, \ + crtc_info->calculated_pixel_clock_100Hz); /* HDMI_ACR_32_0__HDMI_ACR_CTS_32_MASK */ REG_UPDATE(HDMI_ACR_32_0, HDMI_ACR_CTS_32, audio_clock_info.cts_32khz); @@ -1584,6 +1603,17 @@ static void dig_connect_to_otg( REG_UPDATE(DIG_FE_CNTL, DIG_SOURCE_SELECT, tg_inst); } +static unsigned int dig_source_otg( + struct stream_encoder *enc) +{ + uint32_t tg_inst = 0; + struct dce110_stream_encoder *enc110 = DCE110STRENC_FROM_STRENC(enc); + + REG_GET(DIG_FE_CNTL, DIG_SOURCE_SELECT, &tg_inst); + + return tg_inst; +} + static const struct stream_encoder_funcs dce110_str_enc_funcs = { .dp_set_stream_attribute = dce110_stream_encoder_dp_set_stream_attribute, @@ -1618,6 +1648,8 @@ static const struct stream_encoder_funcs dce110_str_enc_funcs = { .setup_stereo_sync = setup_stereo_sync, .set_avmute = dce110_stream_encoder_set_avmute, .dig_connect_to_otg = dig_connect_to_otg, + .hdmi_reset_stream_attribute = dce110_reset_hdmi_stream_attribute, + .dig_source_otg = dig_source_otg, }; void dce110_stream_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c new file mode 100644 index 000000000000..225955ec6d39 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c @@ -0,0 +1,220 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dmub_psr.h" +#include "dc.h" +#include "dc_dmub_srv.h" +#include "../../dmub/inc/dmub_srv.h" +#include "dmub_fw_state.h" +#include "core_types.h" +#include "ipp.h" + +#define MAX_PIPES 6 + +/** + * Get PSR state from firmware. + */ +static void dmub_get_psr_state(uint32_t *psr_state) +{ + // Not yet implemented + // Trigger GPINT interrupt from firmware +} + +/** + * Enable/Disable PSR. + */ +static void dmub_set_psr_enable(struct dmub_psr *dmub, bool enable) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = dmub->ctx; + + cmd.psr_enable.header.type = DMUB_CMD__PSR; + + if (enable) + cmd.psr_enable.header.sub_type = DMUB_CMD__PSR_ENABLE; + else + cmd.psr_enable.header.sub_type = DMUB_CMD__PSR_DISABLE; + + cmd.psr_enable.header.payload_bytes = 0; // Send header only + + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd.psr_enable.header); + dc_dmub_srv_cmd_execute(dc->dmub_srv); + dc_dmub_srv_wait_idle(dc->dmub_srv); +} + +/** + * Set PSR level. + */ +static void dmub_set_psr_level(struct dmub_psr *dmub, uint16_t psr_level) +{ + union dmub_rb_cmd cmd; + uint32_t psr_state = 0; + struct dc_context *dc = dmub->ctx; + + dmub_get_psr_state(&psr_state); + + if (psr_state == 0) + return; + + cmd.psr_set_level.header.type = DMUB_CMD__PSR; + cmd.psr_set_level.header.sub_type = DMUB_CMD__PSR_SET_LEVEL; + cmd.psr_set_level.header.payload_bytes = sizeof(struct dmub_cmd_psr_set_level_data); + cmd.psr_set_level.psr_set_level_data.psr_level = psr_level; + + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd.psr_set_level.header); + dc_dmub_srv_cmd_execute(dc->dmub_srv); + dc_dmub_srv_wait_idle(dc->dmub_srv); +} + +/** + * Setup PSR by programming phy registers and sending psr hw context values to firmware. + */ +static bool dmub_setup_psr(struct dmub_psr *dmub, + struct dc_link *link, + struct psr_context *psr_context) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = dmub->ctx; + struct dmub_cmd_psr_copy_settings_data *copy_settings_data + = &cmd.psr_copy_settings.psr_copy_settings_data; + struct pipe_ctx *pipe_ctx = NULL; + struct resource_context *res_ctx = &link->ctx->dc->current_state->res_ctx; + + for (int i = 0; i < MAX_PIPES; i++) { + if (res_ctx && + res_ctx->pipe_ctx[i].stream && + res_ctx->pipe_ctx[i].stream->link && + res_ctx->pipe_ctx[i].stream->link == link && + res_ctx->pipe_ctx[i].stream->link->connector_signal == SIGNAL_TYPE_EDP) { + pipe_ctx = &res_ctx->pipe_ctx[i]; + break; + } + } + + if (!pipe_ctx || + !&pipe_ctx->plane_res || + !&pipe_ctx->stream_res) + return false; + + // Program DP DPHY fast training registers + link->link_enc->funcs->psr_program_dp_dphy_fast_training(link->link_enc, + psr_context->psrExitLinkTrainingRequired); + + // Program DP_SEC_CNTL1 register to set transmission GPS0 line num and priority to high + link->link_enc->funcs->psr_program_secondary_packet(link->link_enc, + psr_context->sdpTransmitLineNumDeadline); + + cmd.psr_copy_settings.header.type = DMUB_CMD__PSR; + cmd.psr_copy_settings.header.sub_type = DMUB_CMD__PSR_COPY_SETTINGS; + cmd.psr_copy_settings.header.payload_bytes = sizeof(struct dmub_cmd_psr_copy_settings_data); + + // Hw insts + copy_settings_data->dpphy_inst = psr_context->phyType; + copy_settings_data->aux_inst = psr_context->channel; + copy_settings_data->digfe_inst = psr_context->engineId; + copy_settings_data->digbe_inst = psr_context->transmitterId; + + copy_settings_data->mpcc_inst = pipe_ctx->plane_res.mpcc_inst; + + if (pipe_ctx->plane_res.hubp) + copy_settings_data->hubp_inst = pipe_ctx->plane_res.hubp->inst; + else + copy_settings_data->hubp_inst = 0; + if (pipe_ctx->plane_res.dpp) + copy_settings_data->dpp_inst = pipe_ctx->plane_res.dpp->inst; + else + copy_settings_data->dpp_inst = 0; + if (pipe_ctx->stream_res.opp) + copy_settings_data->opp_inst = pipe_ctx->stream_res.opp->inst; + else + copy_settings_data->opp_inst = 0; + if (pipe_ctx->stream_res.tg) + copy_settings_data->otg_inst = pipe_ctx->stream_res.tg->inst; + else + copy_settings_data->otg_inst = 0; + + // Misc + copy_settings_data->psr_level = psr_context->psr_level.u32all; + copy_settings_data->hyst_frames = psr_context->timehyst_frames; + copy_settings_data->hyst_lines = psr_context->hyst_lines; + copy_settings_data->phy_type = psr_context->phyType; + copy_settings_data->aux_repeat = psr_context->aux_repeats; + copy_settings_data->smu_optimizations_en = psr_context->allow_smu_optimizations; + copy_settings_data->skip_wait_for_pll_lock = psr_context->skipPsrWaitForPllLock; + copy_settings_data->frame_delay = psr_context->frame_delay; + copy_settings_data->smu_phy_id = psr_context->smuPhyId; + copy_settings_data->num_of_controllers = psr_context->numberOfControllers; + copy_settings_data->frame_cap_ind = psr_context->psrFrameCaptureIndicationReq; + copy_settings_data->phy_num = psr_context->frame_delay & 0x7; + copy_settings_data->link_rate = psr_context->frame_delay & 0xF; + + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd.psr_copy_settings.header); + dc_dmub_srv_cmd_execute(dc->dmub_srv); + dc_dmub_srv_wait_idle(dc->dmub_srv); + + return true; +} + +static const struct dmub_psr_funcs psr_funcs = { + .set_psr_enable = dmub_set_psr_enable, + .setup_psr = dmub_setup_psr, + .get_psr_state = dmub_get_psr_state, + .set_psr_level = dmub_set_psr_level, +}; + +/** + * Construct PSR object. + */ +static void dmub_psr_construct(struct dmub_psr *psr, struct dc_context *ctx) +{ + psr->ctx = ctx; + psr->funcs = &psr_funcs; +} + +/** + * Allocate and initialize PSR object. + */ +struct dmub_psr *dmub_psr_create(struct dc_context *ctx) +{ + struct dmub_psr *psr = kzalloc(sizeof(struct dmub_psr), GFP_KERNEL); + + if (psr == NULL) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dmub_psr_construct(psr, ctx); + + return psr; +} + +/** + * Deallocate PSR object. + */ +void dmub_psr_destroy(struct dmub_psr **dmub) +{ + kfree(dmub); + *dmub = NULL; +} diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h new file mode 100644 index 000000000000..229958de3035 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.h @@ -0,0 +1,47 @@ +/* + * Copyright 2012-16 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_PSR_H_ +#define _DMUB_PSR_H_ + +#include "os_types.h" + +struct dmub_psr { + struct dc_context *ctx; + const struct dmub_psr_funcs *funcs; +}; + +struct dmub_psr_funcs { + void (*set_psr_enable)(struct dmub_psr *dmub, bool enable); + bool (*setup_psr)(struct dmub_psr *dmub, struct dc_link *link, struct psr_context *psr_context); + void (*get_psr_state)(uint32_t *psr_state); + void (*set_psr_level)(struct dmub_psr *dmub, uint16_t psr_level); +}; + +struct dmub_psr *dmub_psr_create(struct dc_context *ctx); +void dmub_psr_destroy(struct dmub_psr **dmub); + + +#endif /* _DCE_DMUB_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c index 799d36299c9b..753cb8edd996 100644 --- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c @@ -26,7 +26,6 @@ #include "dc.h" #include "core_types.h" #include "clk_mgr.h" -#include "hw_sequencer.h" #include "dce100_hw_sequencer.h" #include "resource.h" @@ -136,7 +135,7 @@ void dce100_hw_sequencer_construct(struct dc *dc) { dce110_hw_sequencer_construct(dc); - dc->hwss.enable_display_power_gating = dce100_enable_display_power_gating; + dc->hwseq->funcs.enable_display_power_gating = dce100_enable_display_power_gating; dc->hwss.prepare_bandwidth = dce100_prepare_bandwidth; dc->hwss.optimize_bandwidth = dce100_optimize_bandwidth; } diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.h index a6b80fdaa666..34518da20009 100644 --- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.h @@ -27,6 +27,7 @@ #define __DC_HWSS_DCE100_H__ #include "core_types.h" +#include "hw_sequencer_private.h" struct dc; struct dc_state; diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c index 6248c8455314..8f78bf9abbca 100644 --- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c @@ -304,7 +304,7 @@ static const struct dce_audio_shift audio_shift = { AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -399,6 +399,37 @@ static const struct dc_plane_cap plane_cap = { #define CC_DC_HDMI_STRAPS__AUDIO_STREAM_NUMBER__SHIFT 0x8 #endif +static int map_transmitter_id_to_phy_instance( + enum transmitter transmitter) +{ + switch (transmitter) { + case TRANSMITTER_UNIPHY_A: + return 0; + break; + case TRANSMITTER_UNIPHY_B: + return 1; + break; + case TRANSMITTER_UNIPHY_C: + return 2; + break; + case TRANSMITTER_UNIPHY_D: + return 3; + break; + case TRANSMITTER_UNIPHY_E: + return 4; + break; + case TRANSMITTER_UNIPHY_F: + return 5; + break; + case TRANSMITTER_UNIPHY_G: + return 6; + break; + default: + ASSERT(0); + return 0; + } +} + static void read_dce_straps( struct dc_context *ctx, struct resource_straps *straps) @@ -506,6 +537,14 @@ static const struct dce_mem_input_mask mi_masks = { .ENABLE = MC_HUB_RDREQ_DMIF_LIMIT__ENABLE_MASK }; +static const struct dce110_aux_registers_shift aux_shift = { + DCE10_AUX_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce110_aux_registers_mask aux_mask = { + DCE10_AUX_MASK_SH_LIST(_MASK) +}; + static struct mem_input *dce100_mem_input_create( struct dc_context *ctx, uint32_t inst) @@ -571,14 +610,18 @@ struct link_encoder *dce100_link_encoder_create( { struct dce110_link_encoder *enc110 = kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); + int link_regs_id; if (!enc110) return NULL; + link_regs_id = + map_transmitter_id_to_phy_instance(enc_init_data->transmitter); + dce110_link_encoder_construct(enc110, enc_init_data, &link_enc_feature, - &link_enc_regs[enc_init_data->transmitter], + &link_enc_regs[link_regs_id], &link_enc_aux_regs[enc_init_data->channel - 1], &link_enc_hpd_regs[enc_init_data->hpd_source]); return &enc110->base; @@ -611,7 +654,10 @@ struct dce_aux *dce100_aux_engine_create( dce110_aux_engine_construct(aux_engine, ctx, inst, SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD, - &aux_engine_regs[inst]); + &aux_engine_regs[inst], + &aux_mask, + &aux_shift, + ctx->dc->caps.extended_aux_timeout_support); return &aux_engine->base; } @@ -668,6 +714,7 @@ struct clock_source *dce100_clock_source_create( return &clk_src->base; } + kfree(clk_src); BREAK_TO_DEBUGGER(); return NULL; } @@ -678,7 +725,7 @@ void dce100_clock_source_destroy(struct clock_source **clk_src) *clk_src = NULL; } -static void destruct(struct dce110_resource_pool *pool) +static void dce100_resource_destruct(struct dce110_resource_pool *pool) { unsigned int i; @@ -838,7 +885,7 @@ static void dce100_destroy_resource_pool(struct resource_pool **pool) { struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool); - destruct(dce110_pool); + dce100_resource_destruct(dce110_pool); kfree(dce110_pool); *pool = NULL; } @@ -903,14 +950,13 @@ static const struct resource_funcs dce100_res_pool_funcs = { .find_first_free_match_stream_enc_for_link = dce100_find_first_free_match_stream_enc_for_link }; -static bool construct( +static bool dce100_resource_construct( uint8_t num_virtual_links, struct dc *dc, struct dce110_resource_pool *pool) { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -921,8 +967,7 @@ static bool construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce100_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); @@ -998,6 +1043,8 @@ static bool construct( dc->caps.max_cursor_size = 128; dc->caps.dual_link_dvi = true; dc->caps.disable_dp_clk_share = true; + dc->caps.extended_aux_timeout_support = false; + for (i = 0; i < pool->base.pipe_count; i++) { pool->base.timing_generators[i] = dce100_timing_generator_create( @@ -1075,7 +1122,7 @@ static bool construct( return true; res_create_fail: - destruct(pool); + dce100_resource_destruct(pool); return false; } @@ -1090,9 +1137,10 @@ struct resource_pool *dce100_create_resource_pool( if (!pool) return NULL; - if (construct(num_virtual_links, dc, pool)) + if (dce100_resource_construct(num_virtual_links, dc, pool)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index fafb4b470140..5b689273ff44 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -61,6 +61,8 @@ #include "atomfirmware.h" +#define GAMMA_HW_POINTS_NUM 256 + /* * All values are in milliseconds; * For eDP, after power-up/power/down, @@ -268,7 +270,7 @@ static void build_prescale_params(struct ipp_prescale_params *prescale_params, } static bool -dce110_set_input_transfer_func(struct pipe_ctx *pipe_ctx, +dce110_set_input_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state) { struct input_pixel_processor *ipp = pipe_ctx->plane_res.ipp; @@ -596,7 +598,7 @@ dce110_translate_regamma_to_hw_format(const struct dc_transfer_func *output_tf, } static bool -dce110_set_output_transfer_func(struct pipe_ctx *pipe_ctx, +dce110_set_output_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, const struct dc_stream_state *stream) { struct transform *xfm = pipe_ctx->plane_res.xfm; @@ -651,10 +653,9 @@ void dce110_enable_stream(struct pipe_ctx *pipe_ctx) { enum dc_lane_count lane_count = pipe_ctx->stream->link->cur_link_settings.lane_count; - struct dc_crtc_timing *timing = &pipe_ctx->stream->timing; struct dc_link *link = pipe_ctx->stream->link; - + const struct dc *dc = link->dc; uint32_t active_total_with_borders; uint32_t early_control = 0; @@ -667,29 +668,7 @@ void dce110_enable_stream(struct pipe_ctx *pipe_ctx) link->link_enc->funcs->connect_dig_be_to_fe(link->link_enc, pipe_ctx->stream_res.stream_enc->id, true); - /* update AVI info frame (HDMI, DP)*/ - /* TODO: FPGA may change to hwss.update_info_frame */ - -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - if (pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata != NULL && - pipe_ctx->plane_res.hubp != NULL) { - if (pipe_ctx->stream->dmdata_address.quad_part != 0) { - /* if using dynamic meta, don't set up generic infopackets */ - pipe_ctx->stream_res.encoder_info_frame.hdrsmd.valid = false; - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata( - pipe_ctx->stream_res.stream_enc, - true, pipe_ctx->plane_res.hubp->inst, - dc_is_dp_signal(pipe_ctx->stream->signal) ? - dmdata_dp : dmdata_hdmi); - } else - pipe_ctx->stream_res.stream_enc->funcs->set_dynamic_metadata( - pipe_ctx->stream_res.stream_enc, - false, pipe_ctx->plane_res.hubp->inst, - dc_is_dp_signal(pipe_ctx->stream->signal) ? - dmdata_dp : dmdata_hdmi); - } -#endif - dce110_update_info_frame(pipe_ctx); + dc->hwss.update_info_frame(pipe_ctx); /* enable early control to avoid corruption on DP monitor*/ active_total_with_borders = @@ -753,7 +732,7 @@ static enum bp_result link_transmitter_control( * @brief * eDP only. */ -void hwss_edp_wait_for_hpd_ready( +void dce110_edp_wait_for_hpd_ready( struct dc_link *link, bool power_up) { @@ -821,7 +800,7 @@ void hwss_edp_wait_for_hpd_ready( } } -void hwss_edp_power_control( +void dce110_edp_power_control( struct dc_link *link, bool power_up) { @@ -903,7 +882,7 @@ void hwss_edp_power_control( * @brief * eDP only. Control the backlight of the eDP panel */ -void hwss_edp_backlight_control( +void dce110_edp_backlight_control( struct dc_link *link, bool enable) { @@ -965,27 +944,23 @@ void hwss_edp_backlight_control( void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx) { /* notify audio driver for audio modes of monitor */ - struct dc *core_dc; - struct pp_smu_funcs *pp_smu = NULL; + struct dc *dc; struct clk_mgr *clk_mgr; unsigned int i, num_audio = 1; if (!pipe_ctx->stream) return; - core_dc = pipe_ctx->stream->ctx->dc; - clk_mgr = core_dc->clk_mgr; + dc = pipe_ctx->stream->ctx->dc; + clk_mgr = dc->clk_mgr; if (pipe_ctx->stream_res.audio && pipe_ctx->stream_res.audio->enabled == true) return; - if (core_dc->res_pool->pp_smu) - pp_smu = core_dc->res_pool->pp_smu; - if (pipe_ctx->stream_res.audio) { for (i = 0; i < MAX_PIPES; i++) { /*current_state not updated yet*/ - if (core_dc->current_state->res_ctx.pipe_ctx[i].stream_res.audio != NULL) + if (dc->current_state->res_ctx.pipe_ctx[i].stream_res.audio != NULL) num_audio++; } @@ -1003,10 +978,9 @@ void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx) } } -void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option) +void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx) { struct dc *dc; - struct pp_smu_funcs *pp_smu = NULL; struct clk_mgr *clk_mgr; if (!pipe_ctx || !pipe_ctx->stream) @@ -1023,27 +997,13 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option) if (pipe_ctx->stream_res.audio) { pipe_ctx->stream_res.audio->enabled = false; - if (dc->res_pool->pp_smu) - pp_smu = dc->res_pool->pp_smu; - - if (option != KEEP_ACQUIRED_RESOURCE || - !dc->debug.az_endpoint_mute_only) - /*only disalbe az_endpoint if power down or free*/ - pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); - if (dc_is_dp_signal(pipe_ctx->stream->signal)) pipe_ctx->stream_res.stream_enc->funcs->dp_audio_disable( pipe_ctx->stream_res.stream_enc); else pipe_ctx->stream_res.stream_enc->funcs->hdmi_audio_disable( pipe_ctx->stream_res.stream_enc); - /*don't free audio if it is from retrain or internal disable stream*/ - if (option == FREE_ACQUIRED_RESOURCE && dc->caps.dynamic_audio == true) { - /*we have to dynamic arbitrate the audio endpoints*/ - /*we free the resource, need reset is_audio_acquired*/ - update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, pipe_ctx->stream_res.audio, false); - pipe_ctx->stream_res.audio = NULL; - } + if (clk_mgr->funcs->enable_pme_wa) /*this is the first audio. apply the PME w/a in order to wake AZ from D3*/ clk_mgr->funcs->enable_pme_wa(clk_mgr); @@ -1056,21 +1016,24 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option) } } -void dce110_disable_stream(struct pipe_ctx *pipe_ctx, int option) +void dce110_disable_stream(struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; struct dc *dc = pipe_ctx->stream->ctx->dc; - if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) + if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) { pipe_ctx->stream_res.stream_enc->funcs->stop_hdmi_info_packets( pipe_ctx->stream_res.stream_enc); + pipe_ctx->stream_res.stream_enc->funcs->hdmi_reset_stream_attribute( + pipe_ctx->stream_res.stream_enc); + } if (dc_is_dp_signal(pipe_ctx->stream->signal)) pipe_ctx->stream_res.stream_enc->funcs->stop_dp_info_packets( pipe_ctx->stream_res.stream_enc); - dc->hwss.disable_audio_stream(pipe_ctx, option); + dc->hwss.disable_audio_stream(pipe_ctx); link->link_enc->funcs->connect_dig_be_to_fe( link->link_enc, @@ -1085,6 +1048,7 @@ void dce110_unblank_stream(struct pipe_ctx *pipe_ctx, struct encoder_unblank_param params = { { 0 } }; struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; + struct dce_hwseq *hws = link->dc->hwseq; /* only 3 items below are used by unblank */ params.timing = pipe_ctx->stream->timing; @@ -1094,7 +1058,7 @@ void dce110_unblank_stream(struct pipe_ctx *pipe_ctx, pipe_ctx->stream_res.stream_enc->funcs->dp_unblank(pipe_ctx->stream_res.stream_enc, ¶ms); if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) { - link->dc->hwss.edp_backlight_control(link, true); + hws->funcs.edp_backlight_control(link, true); } } @@ -1102,9 +1066,10 @@ void dce110_blank_stream(struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; + struct dce_hwseq *hws = link->dc->hwseq; if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) { - link->dc->hwss.edp_backlight_control(link, false); + hws->funcs.edp_backlight_control(link, false); dc_link_set_abm_disable(link); } @@ -1174,33 +1139,34 @@ static void build_audio_output( stream->timing.flags.INTERLACE; audio_output->crtc_info.refresh_rate = - (stream->timing.pix_clk_100hz*10000)/ + (stream->timing.pix_clk_100hz*100)/ (stream->timing.h_total*stream->timing.v_total); audio_output->crtc_info.color_depth = stream->timing.display_color_depth; - audio_output->crtc_info.requested_pixel_clock = - pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10; + audio_output->crtc_info.requested_pixel_clock_100Hz = + pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz; - audio_output->crtc_info.calculated_pixel_clock = - pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz / 10; + audio_output->crtc_info.calculated_pixel_clock_100Hz = + pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz; /*for HDMI, audio ACR is with deep color ratio factor*/ if (dc_is_hdmi_signal(pipe_ctx->stream->signal) && - audio_output->crtc_info.requested_pixel_clock == - (stream->timing.pix_clk_100hz / 10)) { + audio_output->crtc_info.requested_pixel_clock_100Hz == + (stream->timing.pix_clk_100hz)) { if (pipe_ctx->stream_res.pix_clk_params.pixel_encoding == PIXEL_ENCODING_YCBCR420) { - audio_output->crtc_info.requested_pixel_clock = - audio_output->crtc_info.requested_pixel_clock/2; - audio_output->crtc_info.calculated_pixel_clock = - pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz/20; + audio_output->crtc_info.requested_pixel_clock_100Hz = + audio_output->crtc_info.requested_pixel_clock_100Hz/2; + audio_output->crtc_info.calculated_pixel_clock_100Hz = + pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz/2; } } - if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT || - pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + if (state->clk_mgr && + (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT || + pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)) { audio_output->pll_info.dp_dto_source_clock_in_khz = state->clk_mgr->funcs->get_dp_ref_clk_frequency( state->clk_mgr); @@ -1260,7 +1226,7 @@ static void program_scaler(const struct dc *dc, { struct tg_color color = {0}; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) /* TOFPGA */ if (pipe_ctx->plane_res.xfm->funcs->transform_set_pixel_storage_depth == NULL) return; @@ -1359,12 +1325,11 @@ static enum dc_status apply_single_controller_ctx_to_hw( struct dc_stream_state *stream = pipe_ctx->stream; struct drr_params params = {0}; unsigned int event_triggers = 0; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); -#endif + struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe; + struct dce_hwseq *hws = dc->hwseq; - if (dc->hwss.disable_stream_gating) { - dc->hwss.disable_stream_gating(dc, pipe_ctx); + if (hws->funcs.disable_stream_gating) { + hws->funcs.disable_stream_gating(dc, pipe_ctx); } if (pipe_ctx->stream_res.audio != NULL) { @@ -1394,10 +1359,10 @@ static enum dc_status apply_single_controller_ctx_to_hw( /* */ /* Do not touch stream timing on seamless boot optimization. */ if (!pipe_ctx->stream->apply_seamless_boot_optimization) - dc->hwss.enable_stream_timing(pipe_ctx, context, dc); + hws->funcs.enable_stream_timing(pipe_ctx, context, dc); - if (dc->hwss.setup_vupdate_interrupt) - dc->hwss.setup_vupdate_interrupt(pipe_ctx); + if (hws->funcs.setup_vupdate_interrupt) + hws->funcs.setup_vupdate_interrupt(dc, pipe_ctx); params.vertical_total_min = stream->adjust.v_total_min; params.vertical_total_max = stream->adjust.v_total_max; @@ -1408,9 +1373,13 @@ static enum dc_status apply_single_controller_ctx_to_hw( // DRR should set trigger event to monitor surface update event if (stream->adjust.v_total_min != 0 && stream->adjust.v_total_max != 0) event_triggers = 0x80; + /* Event triggers and num frames initialized for DRR, but can be + * later updated for PSR use. Note DRR trigger events are generated + * regardless of whether num frames met. + */ if (pipe_ctx->stream_res.tg->funcs->set_static_screen_control) pipe_ctx->stream_res.tg->funcs->set_static_screen_control( - pipe_ctx->stream_res.tg, event_triggers); + pipe_ctx->stream_res.tg, event_triggers, 2); if (!dc_is_virtual_signal(pipe_ctx->stream->signal)) pipe_ctx->stream_res.stream_enc->funcs->dig_connect_to_otg( @@ -1427,8 +1396,7 @@ static enum dc_status apply_single_controller_ctx_to_hw( pipe_ctx->stream_res.opp, &stream->bit_depth_params, &stream->clamping); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - if (odm_pipe) { + while (odm_pipe) { odm_pipe->stream_res.opp->funcs->opp_set_dyn_expansion( odm_pipe->stream_res.opp, COLOR_SPACE_YCBCR601, @@ -1439,15 +1407,15 @@ static enum dc_status apply_single_controller_ctx_to_hw( odm_pipe->stream_res.opp, &stream->bit_depth_params, &stream->clamping); + odm_pipe = odm_pipe->next_odm_pipe; } -#endif if (!stream->dpms_off) core_link_enable_stream(context, pipe_ctx); pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->bottom_pipe != 0; - pipe_ctx->stream->link->psr_enabled = false; + pipe_ctx->stream->link->psr_feature_enabled = false; return DC_OK; } @@ -1457,8 +1425,6 @@ static enum dc_status apply_single_controller_ctx_to_hw( static void power_down_encoders(struct dc *dc) { int i; - enum connector_id connector_id; - enum signal_type signal = SIGNAL_TYPE_NONE; /* do not know BIOS back-front mapping, simply blank all. It will not * hurt for non-DP @@ -1469,15 +1435,15 @@ static void power_down_encoders(struct dc *dc) } for (i = 0; i < dc->link_count; i++) { - connector_id = dal_graphics_object_id_get_connector_id(dc->links[i]->link_id); - if ((connector_id == CONNECTOR_ID_DISPLAY_PORT) || - (connector_id == CONNECTOR_ID_EDP)) { + enum signal_type signal = dc->links[i]->connector_signal; + if ((signal == SIGNAL_TYPE_EDP) || + (signal == SIGNAL_TYPE_DISPLAY_PORT)) if (!dc->links[i]->wa_flags.dp_keep_receiver_powered) dp_receiver_power_ctrl(dc->links[i], false); - if (connector_id == CONNECTOR_ID_EDP) - signal = SIGNAL_TYPE_EDP; - } + + if (signal != SIGNAL_TYPE_EDP) + signal = SIGNAL_TYPE_NONE; dc->links[i]->link_enc->funcs->disable_output( dc->links[i]->link_enc, signal); @@ -1558,18 +1524,6 @@ static struct dc_stream_state *get_edp_stream(struct dc_state *context) return NULL; } -static struct dc_link *get_edp_link(struct dc *dc) -{ - int i; - - // report any eDP links, even unconnected DDI's - for (i = 0; i < dc->link_count; i++) { - if (dc->links[i]->connector_signal == SIGNAL_TYPE_EDP) - return dc->links[i]; - } - return NULL; -} - static struct dc_link *get_edp_link_with_sink( struct dc *dc, struct dc_state *context) @@ -1605,9 +1559,10 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context) bool can_apply_edp_fast_boot = false; bool can_apply_seamless_boot = false; bool keep_edp_vdd_on = false; + struct dce_hwseq *hws = dc->hwseq; - if (dc->hwss.init_pipes) - dc->hwss.init_pipes(dc, context); + if (hws->funcs.init_pipes) + hws->funcs.init_pipes(dc, context); edp_stream = get_edp_stream(context); @@ -1644,7 +1599,7 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context) if (!can_apply_edp_fast_boot && !can_apply_seamless_boot) { if (edp_link_with_sink && !keep_edp_vdd_on) { /*turn off backlight before DP_blank and encoder powered down*/ - dc->hwss.edp_backlight_control(edp_link_with_sink, false); + hws->funcs.edp_backlight_control(edp_link_with_sink, false); } /*resume from S3, no vbios posting, no need to power down again*/ power_down_all_hw_blocks(dc); @@ -1748,12 +1703,15 @@ void dce110_set_safe_displaymarks( ******************************************************************************/ static void set_drr(struct pipe_ctx **pipe_ctx, - int num_pipes, int vmin, int vmax) + int num_pipes, unsigned int vmin, unsigned int vmax, + unsigned int vmid, unsigned int vmid_frame_number) { int i = 0; struct drr_params params = {0}; // DRR should set trigger event to monitor surface update event unsigned int event_triggers = 0x80; + // Note DRR trigger events are generated regardless of whether num frames met. + unsigned int num_frames = 2; params.vertical_total_max = vmax; params.vertical_total_min = vmin; @@ -1769,7 +1727,7 @@ static void set_drr(struct pipe_ctx **pipe_ctx, if (vmax != 0 && vmin != 0) pipe_ctx[i]->stream_res.tg->funcs->set_static_screen_control( pipe_ctx[i]->stream_res.tg, - event_triggers); + event_triggers, num_frames); } } @@ -1786,30 +1744,31 @@ static void get_position(struct pipe_ctx **pipe_ctx, } static void set_static_screen_control(struct pipe_ctx **pipe_ctx, - int num_pipes, const struct dc_static_screen_events *events) + int num_pipes, const struct dc_static_screen_params *params) { unsigned int i; - unsigned int value = 0; + unsigned int triggers = 0; - if (events->overlay_update) - value |= 0x100; - if (events->surface_update) - value |= 0x80; - if (events->cursor_update) - value |= 0x2; - if (events->force_trigger) - value |= 0x1; + if (params->triggers.overlay_update) + triggers |= 0x100; + if (params->triggers.surface_update) + triggers |= 0x80; + if (params->triggers.cursor_update) + triggers |= 0x2; + if (params->triggers.force_trigger) + triggers |= 0x1; if (num_pipes) { struct dc *dc = pipe_ctx[0]->stream->ctx->dc; if (dc->fbc_compressor) - value |= 0x84; + triggers |= 0x84; } for (i = 0; i < num_pipes; i++) pipe_ctx[i]->stream_res.tg->funcs-> - set_static_screen_control(pipe_ctx[i]->stream_res.tg, value); + set_static_screen_control(pipe_ctx[i]->stream_res.tg, + triggers, params->num_frames); } /* @@ -1862,7 +1821,7 @@ static bool should_enable_fbc(struct dc *dc, return false; /* PSR should not be enabled */ - if (pipe_ctx->stream->link->psr_enabled) + if (pipe_ctx->stream->link->psr_feature_enabled) return false; /* Nothing to compress */ @@ -1932,8 +1891,25 @@ static void dce110_reset_hw_ctx_wrap( /* Disable if new stream is null. O/w, if stream is * disabled already, no need to disable again. */ - if (!pipe_ctx->stream || !pipe_ctx->stream->dpms_off) - core_link_disable_stream(pipe_ctx_old, FREE_ACQUIRED_RESOURCE); + if (!pipe_ctx->stream || !pipe_ctx->stream->dpms_off) { + core_link_disable_stream(pipe_ctx_old); + + /* free acquired resources*/ + if (pipe_ctx_old->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx_old->stream_res.audio->funcs-> + az_disable(pipe_ctx_old->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx_old->stream_res.audio, false); + pipe_ctx_old->stream_res.audio = NULL; + } + } + } pipe_ctx_old->stream_res.tg->funcs->set_blank(pipe_ctx_old->stream_res.tg, true); if (!hwss_wait_for_blank_complete(pipe_ctx_old->stream_res.tg)) { @@ -2041,13 +2017,14 @@ enum dc_status dce110_apply_ctx_to_hw( struct dc *dc, struct dc_state *context) { + struct dce_hwseq *hws = dc->hwseq; struct dc_bios *dcb = dc->ctx->dc_bios; enum dc_status status; int i; /* Reset old context */ /* look up the targets that have been removed since last commit */ - dc->hwss.reset_hw_ctx_wrap(dc, context); + hws->funcs.reset_hw_ctx_wrap(dc, context); /* Skip applying if no targets */ if (context->stream_count <= 0) @@ -2072,7 +2049,7 @@ enum dc_status dce110_apply_ctx_to_hw( continue; } - dc->hwss.enable_display_power_gating( + hws->funcs.enable_display_power_gating( dc, i, dc->ctx->dc_bios, PIPE_GATING_CONTROL_DISABLE); } @@ -2098,7 +2075,7 @@ enum dc_status dce110_apply_ctx_to_hw( if (pipe_ctx_old->stream && !pipe_need_reprogram(pipe_ctx_old, pipe_ctx)) continue; - if (pipe_ctx->top_pipe) + if (pipe_ctx->top_pipe || pipe_ctx->prev_odm_pipe) continue; status = apply_single_controller_ctx_to_hw( @@ -2381,19 +2358,20 @@ static void init_hw(struct dc *dc) struct transform *xfm; struct abm *abm; struct dmcu *dmcu; + struct dce_hwseq *hws = dc->hwseq; bp = dc->ctx->dc_bios; for (i = 0; i < dc->res_pool->pipe_count; i++) { xfm = dc->res_pool->transforms[i]; xfm->funcs->transform_reset(xfm); - dc->hwss.enable_display_power_gating( + hws->funcs.enable_display_power_gating( dc, i, bp, PIPE_GATING_CONTROL_INIT); - dc->hwss.enable_display_power_gating( + hws->funcs.enable_display_power_gating( dc, i, bp, PIPE_GATING_CONTROL_DISABLE); - dc->hwss.enable_display_pipe_clock_gating( + hws->funcs.enable_display_pipe_clock_gating( dc->ctx, true); } @@ -2475,17 +2453,15 @@ static void dce110_program_front_end_for_pipe( struct dc *dc, struct pipe_ctx *pipe_ctx) { struct mem_input *mi = pipe_ctx->plane_res.mi; - struct pipe_ctx *old_pipe = NULL; struct dc_plane_state *plane_state = pipe_ctx->plane_state; struct xfm_grph_csc_adjustment adjust; struct out_csc_color_matrix tbl_entry; unsigned int i; + struct dce_hwseq *hws = dc->hwseq; + DC_LOGGER_INIT(); memset(&tbl_entry, 0, sizeof(tbl_entry)); - if (dc->current_state) - old_pipe = &dc->current_state->res_ctx.pipe_ctx[pipe_ctx->pipe_idx]; - memset(&adjust, 0, sizeof(adjust)); adjust.gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS; @@ -2541,10 +2517,10 @@ static void dce110_program_front_end_for_pipe( if (pipe_ctx->plane_state->update_flags.bits.full_update || pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change || pipe_ctx->plane_state->update_flags.bits.gamma_change) - dc->hwss.set_input_transfer_func(pipe_ctx, pipe_ctx->plane_state); + hws->funcs.set_input_transfer_func(dc, pipe_ctx, pipe_ctx->plane_state); if (pipe_ctx->plane_state->update_flags.bits.full_update) - dc->hwss.set_output_transfer_func(pipe_ctx, pipe_ctx->stream); + hws->funcs.set_output_transfer_func(dc, pipe_ctx, pipe_ctx->stream); DC_LOG_SURFACE( "Pipe:%d %p: addr hi:0x%x, " @@ -2647,6 +2623,7 @@ static void dce110_apply_ctx_for_surface( static void dce110_power_down_fe(struct dc *dc, struct pipe_ctx *pipe_ctx) { + struct dce_hwseq *hws = dc->hwseq; int fe_idx = pipe_ctx->plane_res.mi ? pipe_ctx->plane_res.mi->inst : pipe_ctx->pipe_idx; @@ -2654,7 +2631,7 @@ static void dce110_power_down_fe(struct dc *dc, struct pipe_ctx *pipe_ctx) if (dc->current_state->res_ctx.pipe_ctx[fe_idx].stream) return; - dc->hwss.enable_display_power_gating( + hws->funcs.enable_display_power_gating( dc, fe_idx, dc->ctx->dc_bios, PIPE_GATING_CONTROL_ENABLE); dc->res_pool->transforms[fe_idx]->funcs->transform_reset( @@ -2743,14 +2720,10 @@ static const struct hw_sequencer_funcs dce110_funcs = { .program_gamut_remap = program_gamut_remap, .program_output_csc = program_output_csc, .init_hw = init_hw, - .init_pipes = init_pipes, .apply_ctx_to_hw = dce110_apply_ctx_to_hw, .apply_ctx_for_surface = dce110_apply_ctx_for_surface, .update_plane_addr = update_plane_addr, .update_pending_status = dce110_update_pending_status, - .set_input_transfer_func = dce110_set_input_transfer_func, - .set_output_transfer_func = dce110_set_output_transfer_func, - .power_down = dce110_power_down, .enable_accelerated_mode = dce110_enable_accelerated_mode, .enable_timing_synchronization = dce110_enable_timing_synchronization, .enable_per_frame_crtc_position_reset = dce110_enable_per_frame_crtc_position_reset, @@ -2761,8 +2734,6 @@ static const struct hw_sequencer_funcs dce110_funcs = { .blank_stream = dce110_blank_stream, .enable_audio_stream = dce110_enable_audio_stream, .disable_audio_stream = dce110_disable_audio_stream, - .enable_display_pipe_clock_gating = enable_display_pipe_clock_gating, - .enable_display_power_gating = dce110_enable_display_power_gating, .disable_plane = dce110_power_down_fe, .pipe_control_lock = dce_pipe_control_lock, .prepare_bandwidth = dce110_prepare_bandwidth, @@ -2770,22 +2741,33 @@ static const struct hw_sequencer_funcs dce110_funcs = { .set_drr = set_drr, .get_position = get_position, .set_static_screen_control = set_static_screen_control, - .reset_hw_ctx_wrap = dce110_reset_hw_ctx_wrap, - .enable_stream_timing = dce110_enable_stream_timing, - .disable_stream_gating = NULL, - .enable_stream_gating = NULL, .setup_stereo = NULL, .set_avmute = dce110_set_avmute, .wait_for_mpcc_disconnect = dce110_wait_for_mpcc_disconnect, - .edp_backlight_control = hwss_edp_backlight_control, - .edp_power_control = hwss_edp_power_control, - .edp_wait_for_hpd_ready = hwss_edp_wait_for_hpd_ready, + .edp_power_control = dce110_edp_power_control, + .edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready, .set_cursor_position = dce110_set_cursor_position, .set_cursor_attribute = dce110_set_cursor_attribute }; +static const struct hwseq_private_funcs dce110_private_funcs = { + .init_pipes = init_pipes, + .update_plane_addr = update_plane_addr, + .set_input_transfer_func = dce110_set_input_transfer_func, + .set_output_transfer_func = dce110_set_output_transfer_func, + .power_down = dce110_power_down, + .enable_display_pipe_clock_gating = enable_display_pipe_clock_gating, + .enable_display_power_gating = dce110_enable_display_power_gating, + .reset_hw_ctx_wrap = dce110_reset_hw_ctx_wrap, + .enable_stream_timing = dce110_enable_stream_timing, + .disable_stream_gating = NULL, + .enable_stream_gating = NULL, + .edp_backlight_control = dce110_edp_backlight_control, +}; + void dce110_hw_sequencer_construct(struct dc *dc) { dc->hwss = dce110_funcs; + dc->hwseq->funcs = dce110_private_funcs; } diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h index cd3e36d52a52..26a9c14a58b1 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h @@ -27,8 +27,8 @@ #define __DC_HWSS_DCE110_H__ #include "core_types.h" +#include "hw_sequencer_private.h" -#define GAMMA_HW_POINTS_NUM 256 struct dc; struct dc_state; struct dm_pp_display_configuration; @@ -42,7 +42,7 @@ enum dc_status dce110_apply_ctx_to_hw( void dce110_enable_stream(struct pipe_ctx *pipe_ctx); -void dce110_disable_stream(struct pipe_ctx *pipe_ctx, int option); +void dce110_disable_stream(struct pipe_ctx *pipe_ctx); void dce110_unblank_stream(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings); @@ -50,7 +50,7 @@ void dce110_unblank_stream(struct pipe_ctx *pipe_ctx, void dce110_blank_stream(struct pipe_ctx *pipe_ctx); void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx); -void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx, int option); +void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx); void dce110_update_info_frame(struct pipe_ctx *pipe_ctx); @@ -73,15 +73,15 @@ void dce110_optimize_bandwidth( void dp_receiver_power_ctrl(struct dc_link *link, bool on); -void hwss_edp_power_control( +void dce110_edp_power_control( struct dc_link *link, bool power_up); -void hwss_edp_backlight_control( +void dce110_edp_backlight_control( struct dc_link *link, bool enable); -void hwss_edp_wait_for_hpd_ready( +void dce110_edp_wait_for_hpd_ready( struct dc_link *link, bool power_up); diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_mem_input_v.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_mem_input_v.c index 9b9fc3d96c07..d54172d88f5f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_mem_input_v.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_mem_input_v.c @@ -229,26 +229,26 @@ static void program_tiling( static void program_size_and_rotation( struct dce_mem_input *mem_input110, enum dc_rotation_angle rotation, - const union plane_size *plane_size) + const struct plane_size *plane_size) { uint32_t value = 0; - union plane_size local_size = *plane_size; + struct plane_size local_size = *plane_size; if (rotation == ROTATION_ANGLE_90 || rotation == ROTATION_ANGLE_270) { - swap(local_size.video.luma_size.x, - local_size.video.luma_size.y); - swap(local_size.video.luma_size.width, - local_size.video.luma_size.height); - swap(local_size.video.chroma_size.x, - local_size.video.chroma_size.y); - swap(local_size.video.chroma_size.width, - local_size.video.chroma_size.height); + swap(local_size.surface_size.x, + local_size.surface_size.y); + swap(local_size.surface_size.width, + local_size.surface_size.height); + swap(local_size.chroma_size.x, + local_size.chroma_size.y); + swap(local_size.chroma_size.width, + local_size.chroma_size.height); } value = 0; - set_reg_field_value(value, local_size.video.luma_pitch, + set_reg_field_value(value, local_size.surface_pitch, UNP_GRPH_PITCH_L, GRPH_PITCH_L); dm_write_reg( @@ -257,7 +257,7 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.chroma_pitch, + set_reg_field_value(value, local_size.chroma_pitch, UNP_GRPH_PITCH_C, GRPH_PITCH_C); dm_write_reg( mem_input110->base.ctx, @@ -297,8 +297,8 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.luma_size.x + - local_size.video.luma_size.width, + set_reg_field_value(value, local_size.surface_size.x + + local_size.surface_size.width, UNP_GRPH_X_END_L, GRPH_X_END_L); dm_write_reg( mem_input110->base.ctx, @@ -306,8 +306,8 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.chroma_size.x + - local_size.video.chroma_size.width, + set_reg_field_value(value, local_size.chroma_size.x + + local_size.chroma_size.width, UNP_GRPH_X_END_C, GRPH_X_END_C); dm_write_reg( mem_input110->base.ctx, @@ -315,8 +315,8 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.luma_size.y + - local_size.video.luma_size.height, + set_reg_field_value(value, local_size.surface_size.y + + local_size.surface_size.height, UNP_GRPH_Y_END_L, GRPH_Y_END_L); dm_write_reg( mem_input110->base.ctx, @@ -324,8 +324,8 @@ static void program_size_and_rotation( value); value = 0; - set_reg_field_value(value, local_size.video.chroma_size.y + - local_size.video.chroma_size.height, + set_reg_field_value(value, local_size.chroma_size.y + + local_size.chroma_size.height, UNP_GRPH_Y_END_C, GRPH_Y_END_C); dm_write_reg( mem_input110->base.ctx, @@ -637,7 +637,7 @@ void dce_mem_input_v_program_surface_config( struct mem_input *mem_input, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizotal_mirror) diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c index 764329264c3b..bf14e9ab040c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c @@ -275,6 +275,14 @@ static const struct dce_stream_encoder_mask se_mask = { SE_COMMON_MASK_SH_LIST_DCE110(_MASK) }; +static const struct dce110_aux_registers_shift aux_shift = { + DCE_AUX_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce110_aux_registers_mask aux_mask = { + DCE_AUX_MASK_SH_LIST(_MASK) +}; + #define opp_regs(id)\ [id] = {\ OPP_DCE_110_REG_LIST(id),\ @@ -331,7 +339,7 @@ static const struct dce_audio_shift audio_shift = { AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -440,6 +448,37 @@ static const struct dc_plane_cap underlay_plane_cap = { #define CC_DC_HDMI_STRAPS__AUDIO_STREAM_NUMBER__SHIFT 0x8 #endif +static int map_transmitter_id_to_phy_instance( + enum transmitter transmitter) +{ + switch (transmitter) { + case TRANSMITTER_UNIPHY_A: + return 0; + break; + case TRANSMITTER_UNIPHY_B: + return 1; + break; + case TRANSMITTER_UNIPHY_C: + return 2; + break; + case TRANSMITTER_UNIPHY_D: + return 3; + break; + case TRANSMITTER_UNIPHY_E: + return 4; + break; + case TRANSMITTER_UNIPHY_F: + return 5; + break; + case TRANSMITTER_UNIPHY_G: + return 6; + break; + default: + ASSERT(0); + return 0; + } +} + static void read_dce_straps( struct dc_context *ctx, struct resource_straps *straps) @@ -617,14 +656,18 @@ static struct link_encoder *dce110_link_encoder_create( { struct dce110_link_encoder *enc110 = kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); + int link_regs_id; if (!enc110) return NULL; + link_regs_id = + map_transmitter_id_to_phy_instance(enc_init_data->transmitter); + dce110_link_encoder_construct(enc110, enc_init_data, &link_enc_feature, - &link_enc_regs[enc_init_data->transmitter], + &link_enc_regs[link_regs_id], &link_enc_aux_regs[enc_init_data->channel - 1], &link_enc_hpd_regs[enc_init_data->hpd_source]); return &enc110->base; @@ -657,7 +700,10 @@ struct dce_aux *dce110_aux_engine_create( dce110_aux_engine_construct(aux_engine, ctx, inst, SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD, - &aux_engine_regs[inst]); + &aux_engine_regs[inst], + &aux_mask, + &aux_shift, + ctx->dc->caps.extended_aux_timeout_support); return &aux_engine->base; } @@ -714,6 +760,7 @@ struct clock_source *dce110_clock_source_create( return &clk_src->base; } + kfree(clk_src); BREAK_TO_DEBUGGER(); return NULL; } @@ -735,7 +782,7 @@ void dce110_clock_source_destroy(struct clock_source **clk_src) *clk_src = NULL; } -static void destruct(struct dce110_resource_pool *pool) +static void dce110_resource_destruct(struct dce110_resource_pool *pool) { unsigned int i; @@ -1050,6 +1097,7 @@ static struct pipe_ctx *dce110_acquire_underlay( struct dc_stream_state *stream) { struct dc *dc = stream->ctx->dc; + struct dce_hwseq *hws = dc->hwseq; struct resource_context *res_ctx = &context->res_ctx; unsigned int underlay_idx = pool->underlay_pipe_index; struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[underlay_idx]; @@ -1070,7 +1118,7 @@ static struct pipe_ctx *dce110_acquire_underlay( struct tg_color black_color = {0}; struct dc_bios *dcb = dc->ctx->dc_bios; - dc->hwss.enable_display_power_gating( + hws->funcs.enable_display_power_gating( dc, pipe_ctx->stream_res.tg->inst, dcb, PIPE_GATING_CONTROL_DISABLE); @@ -1114,7 +1162,7 @@ static void dce110_destroy_resource_pool(struct resource_pool **pool) { struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool); - destruct(dce110_pool); + dce110_resource_destruct(dce110_pool); kfree(dce110_pool); *pool = NULL; } @@ -1266,7 +1314,7 @@ const struct resource_caps *dce110_resource_cap( return &carrizo_resource_cap; } -static bool construct( +static bool dce110_resource_construct( uint8_t num_virtual_links, struct dc *dc, struct dce110_resource_pool *pool, @@ -1274,7 +1322,6 @@ static bool construct( { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -1293,6 +1340,7 @@ static bool construct( dc->caps.i2c_speed_in_khz = 100; dc->caps.max_cursor_size = 128; dc->caps.is_apu = true; + dc->caps.extended_aux_timeout_support = false; /************************************************* * Create resources * @@ -1300,8 +1348,7 @@ static bool construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce110_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); @@ -1446,7 +1493,7 @@ static bool construct( return true; res_create_fail: - destruct(pool); + dce110_resource_destruct(pool); return false; } @@ -1461,9 +1508,10 @@ struct resource_pool *dce110_create_resource_pool( if (!pool) return NULL; - if (construct(num_virtual_links, dc, pool, asic_id)) + if (dce110_resource_construct(num_virtual_links, dc, pool, asic_id)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.c index 5f7c2c5641c4..1ea7db8eeb98 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.c @@ -469,22 +469,27 @@ void dce110_timing_generator_set_drr( void dce110_timing_generator_set_static_screen_control( struct timing_generator *tg, - uint32_t value) + uint32_t event_triggers, + uint32_t num_frames) { struct dce110_timing_generator *tg110 = DCE110TG_FROM_TG(tg); uint32_t static_screen_cntl = 0; uint32_t addr = 0; + // By register spec, it only takes 8 bit value + if (num_frames > 0xFF) + num_frames = 0xFF; + addr = CRTC_REG(mmCRTC_STATIC_SCREEN_CONTROL); static_screen_cntl = dm_read_reg(tg->ctx, addr); set_reg_field_value(static_screen_cntl, - value, + event_triggers, CRTC_STATIC_SCREEN_CONTROL, CRTC_STATIC_SCREEN_EVENT_MASK); set_reg_field_value(static_screen_cntl, - 2, + num_frames, CRTC_STATIC_SCREEN_CONTROL, CRTC_STATIC_SCREEN_FRAME_COUNT); diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.h b/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.h index 768ccf27ada9..d8a5ed7b485d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.h +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.h @@ -231,7 +231,8 @@ void dce110_timing_generator_set_drr( void dce110_timing_generator_set_static_screen_control( struct timing_generator *tg, - uint32_t value); + uint32_t event_triggers, + uint32_t num_frames); void dce110_timing_generator_get_crtc_scanoutpos( struct timing_generator *tg, diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c index 1e4a7c13f0ed..19873ee1f78d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c @@ -158,6 +158,6 @@ void dce112_hw_sequencer_construct(struct dc *dc) * structure */ dce110_hw_sequencer_construct(dc); - dc->hwss.enable_display_power_gating = dce112_enable_display_power_gating; + dc->hwseq->funcs.enable_display_power_gating = dce112_enable_display_power_gating; } diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.h index e646f4a37fa2..943f1b2c5b2f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.h @@ -27,6 +27,7 @@ #define __DC_HWSS_DCE112_H__ #include "core_types.h" +#include "hw_sequencer_private.h" struct dc; diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c index c6136e0ed1a4..700ad8b3e54b 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c @@ -172,6 +172,14 @@ static const struct dce_abm_mask abm_mask = { ABM_MASK_SH_LIST_DCE110(_MASK) }; +static const struct dce110_aux_registers_shift aux_shift = { + DCE_AUX_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce110_aux_registers_mask aux_mask = { + DCE_AUX_MASK_SH_LIST(_MASK) +}; + #define ipp_regs(id)\ [id] = {\ IPP_DCE110_REG_LIST_DCE_BASE(id)\ @@ -337,7 +345,7 @@ static const struct dce_audio_shift audio_shift = { AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -417,6 +425,37 @@ static const struct dc_plane_cap plane_cap = { #define CC_DC_HDMI_STRAPS__AUDIO_STREAM_NUMBER__SHIFT 0x8 #endif +static int map_transmitter_id_to_phy_instance( + enum transmitter transmitter) +{ + switch (transmitter) { + case TRANSMITTER_UNIPHY_A: + return 0; + break; + case TRANSMITTER_UNIPHY_B: + return 1; + break; + case TRANSMITTER_UNIPHY_C: + return 2; + break; + case TRANSMITTER_UNIPHY_D: + return 3; + break; + case TRANSMITTER_UNIPHY_E: + return 4; + break; + case TRANSMITTER_UNIPHY_F: + return 5; + break; + case TRANSMITTER_UNIPHY_G: + return 6; + break; + default: + ASSERT(0); + return 0; + } +} + static void read_dce_straps( struct dc_context *ctx, struct resource_straps *straps) @@ -575,14 +614,18 @@ struct link_encoder *dce112_link_encoder_create( { struct dce110_link_encoder *enc110 = kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); + int link_regs_id; if (!enc110) return NULL; + link_regs_id = + map_transmitter_id_to_phy_instance(enc_init_data->transmitter); + dce110_link_encoder_construct(enc110, enc_init_data, &link_enc_feature, - &link_enc_regs[enc_init_data->transmitter], + &link_enc_regs[link_regs_id], &link_enc_aux_regs[enc_init_data->channel - 1], &link_enc_hpd_regs[enc_init_data->hpd_source]); return &enc110->base; @@ -630,7 +673,10 @@ struct dce_aux *dce112_aux_engine_create( dce110_aux_engine_construct(aux_engine, ctx, inst, SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD, - &aux_engine_regs[inst]); + &aux_engine_regs[inst], + &aux_mask, + &aux_shift, + ctx->dc->caps.extended_aux_timeout_support); return &aux_engine->base; } @@ -687,6 +733,7 @@ struct clock_source *dce112_clock_source_create( return &clk_src->base; } + kfree(clk_src); BREAK_TO_DEBUGGER(); return NULL; } @@ -697,7 +744,7 @@ void dce112_clock_source_destroy(struct clock_source **clk_src) *clk_src = NULL; } -static void destruct(struct dce110_resource_pool *pool) +static void dce112_resource_destruct(struct dce110_resource_pool *pool) { unsigned int i; @@ -966,7 +1013,7 @@ static void dce112_destroy_resource_pool(struct resource_pool **pool) { struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool); - destruct(dce110_pool); + dce112_resource_destruct(dce110_pool); kfree(dce110_pool); *pool = NULL; } @@ -987,6 +1034,10 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) struct dm_pp_clock_levels_with_latency mem_clks = {0}; struct dm_pp_wm_sets_with_clock_ranges clk_ranges = {0}; struct dm_pp_clock_levels clks = {0}; + int memory_type_multiplier = MEMORY_TYPE_MULTIPLIER_CZ; + + if (dc->bw_vbios && dc->bw_vbios->memory_type == bw_def_hbm) + memory_type_multiplier = MEMORY_TYPE_HBM; /*do system clock TODO PPLIB: after PPLIB implement, * then remove old way @@ -1026,12 +1077,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) &clks); dc->bw_vbios->low_yclk = bw_frc_to_fixed( - clks.clocks_in_khz[0] * MEMORY_TYPE_MULTIPLIER_CZ, 1000); + clks.clocks_in_khz[0] * memory_type_multiplier, 1000); dc->bw_vbios->mid_yclk = bw_frc_to_fixed( - clks.clocks_in_khz[clks.num_levels>>1] * MEMORY_TYPE_MULTIPLIER_CZ, + clks.clocks_in_khz[clks.num_levels>>1] * memory_type_multiplier, 1000); dc->bw_vbios->high_yclk = bw_frc_to_fixed( - clks.clocks_in_khz[clks.num_levels-1] * MEMORY_TYPE_MULTIPLIER_CZ, + clks.clocks_in_khz[clks.num_levels-1] * memory_type_multiplier, 1000); return; @@ -1067,12 +1118,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) * YCLK = UMACLK*m_memoryTypeMultiplier */ dc->bw_vbios->low_yclk = bw_frc_to_fixed( - mem_clks.data[0].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, 1000); + mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->mid_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, + mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->high_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, + mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier, 1000); /* Now notify PPLib/SMU about which Watermarks sets they should select @@ -1135,7 +1186,7 @@ const struct resource_caps *dce112_resource_cap( return &polaris_10_resource_cap; } -static bool construct( +static bool dce112_resource_construct( uint8_t num_virtual_links, struct dc *dc, struct dce110_resource_pool *pool) @@ -1158,7 +1209,7 @@ static bool construct( dc->caps.i2c_speed_in_khz = 100; dc->caps.max_cursor_size = 128; dc->caps.dual_link_dvi = true; - + dc->caps.extended_aux_timeout_support = false; /************************************************* * Create resources * @@ -1321,7 +1372,7 @@ static bool construct( return true; res_create_fail: - destruct(pool); + dce112_resource_destruct(pool); return false; } @@ -1335,9 +1386,10 @@ struct resource_pool *dce112_create_resource_pool( if (!pool) return NULL; - if (construct(num_virtual_links, dc, pool)) + if (dce112_resource_construct(num_virtual_links, dc, pool)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c index 1ca30928025e..66a13aa39c95 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c @@ -265,7 +265,7 @@ void dce120_hw_sequencer_construct(struct dc *dc) * structure */ dce110_hw_sequencer_construct(dc); - dc->hwss.enable_display_power_gating = dce120_enable_display_power_gating; + dc->hwseq->funcs.enable_display_power_gating = dce120_enable_display_power_gating; dc->hwss.update_dchub = dce120_update_dchub; } diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.h index c51afbd0b012..bc024534732f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.h @@ -27,6 +27,7 @@ #define __DC_HWSS_DCE120_H__ #include "core_types.h" +#include "hw_sequencer_private.h" struct dc; diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c index 4a6ba3173a5a..53ab88ef71f5 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c @@ -63,8 +63,8 @@ #include "soc15_hw_ip.h" #include "vega10_ip_offset.h" #include "nbio/nbio_6_1_offset.h" -#include "mmhub/mmhub_9_4_0_offset.h" -#include "mmhub/mmhub_9_4_0_sh_mask.h" +#include "mmhub/mmhub_1_0_offset.h" +#include "mmhub/mmhub_1_0_sh_mask.h" #include "reg_helper.h" #include "dce100/dce100_resource.h" @@ -293,6 +293,14 @@ static const struct dce_stream_encoder_mask se_mask = { SE_COMMON_MASK_SH_LIST_DCE120(_MASK) }; +static const struct dce110_aux_registers_shift aux_shift = { + DCE12_AUX_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce110_aux_registers_mask aux_mask = { + DCE12_AUX_MASK_SH_LIST(_MASK) +}; + #define opp_regs(id)\ [id] = {\ OPP_DCE_120_REG_LIST(id),\ @@ -352,10 +360,41 @@ static const struct dce_audio_shift audio_shift = { DCE120_AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { DCE120_AUD_COMMON_MASK_SH_LIST(_MASK) }; +static int map_transmitter_id_to_phy_instance( + enum transmitter transmitter) +{ + switch (transmitter) { + case TRANSMITTER_UNIPHY_A: + return 0; + break; + case TRANSMITTER_UNIPHY_B: + return 1; + break; + case TRANSMITTER_UNIPHY_C: + return 2; + break; + case TRANSMITTER_UNIPHY_D: + return 3; + break; + case TRANSMITTER_UNIPHY_E: + return 4; + break; + case TRANSMITTER_UNIPHY_F: + return 5; + break; + case TRANSMITTER_UNIPHY_G: + return 6; + break; + default: + ASSERT(0); + return 0; + } +} + #define clk_src_regs(index, id)\ [index] = {\ CS_COMMON_REG_LIST_DCE_112(id),\ @@ -404,7 +443,10 @@ struct dce_aux *dce120_aux_engine_create( dce110_aux_engine_construct(aux_engine, ctx, inst, SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD, - &aux_engine_regs[inst]); + &aux_engine_regs[inst], + &aux_mask, + &aux_shift, + ctx->dc->caps.extended_aux_timeout_support); return &aux_engine->base; } @@ -500,6 +542,7 @@ static struct clock_source *dce120_clock_source_create( return &clk_src->base; } + kfree(clk_src); BREAK_TO_DEBUGGER(); return NULL; } @@ -544,7 +587,7 @@ static void dce120_transform_destroy(struct transform **xfm) *xfm = NULL; } -static void destruct(struct dce110_resource_pool *pool) +static void dce120_resource_destruct(struct dce110_resource_pool *pool) { unsigned int i; @@ -654,14 +697,18 @@ static struct link_encoder *dce120_link_encoder_create( { struct dce110_link_encoder *enc110 = kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); + int link_regs_id; if (!enc110) return NULL; + link_regs_id = + map_transmitter_id_to_phy_instance(enc_init_data->transmitter); + dce110_link_encoder_construct(enc110, enc_init_data, &link_enc_feature, - &link_enc_regs[enc_init_data->transmitter], + &link_enc_regs[link_regs_id], &link_enc_aux_regs[enc_init_data->channel - 1], &link_enc_hpd_regs[enc_init_data->hpd_source]); @@ -825,7 +872,7 @@ static void dce120_destroy_resource_pool(struct resource_pool **pool) { struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool); - destruct(dce110_pool); + dce120_resource_destruct(dce110_pool); kfree(dce110_pool); *pool = NULL; } @@ -847,6 +894,8 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) int i; unsigned int clk; unsigned int latency; + /*original logic in dal3*/ + int memory_type_multiplier = MEMORY_TYPE_MULTIPLIER_CZ; /*do system clock*/ if (!dm_pp_get_clock_levels_by_type_with_latency( @@ -905,13 +954,16 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc) * ALSO always convert UMA clock (from PPLIB) to YCLK (HW formula): * YCLK = UMACLK*m_memoryTypeMultiplier */ + if (dc->bw_vbios->memory_type == bw_def_hbm) + memory_type_multiplier = MEMORY_TYPE_HBM; + dc->bw_vbios->low_yclk = bw_frc_to_fixed( - mem_clks.data[0].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, 1000); + mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->mid_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, + mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier, 1000); dc->bw_vbios->high_yclk = bw_frc_to_fixed( - mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, + mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier, 1000); /* Now notify PPLib/SMU about which Watermarks sets they should select @@ -972,7 +1024,7 @@ static uint32_t read_pipe_fuses(struct dc_context *ctx) return value; } -static bool construct( +static bool dce120_resource_construct( uint8_t num_virtual_links, struct dc *dc, struct dce110_resource_pool *pool) @@ -1000,7 +1052,7 @@ static bool construct( dc->caps.max_cursor_size = 128; dc->caps.dual_link_dvi = true; dc->caps.psp_setup_panel_mode = true; - + dc->caps.extended_aux_timeout_support = false; dc->debug = debug_defaults; /************************************************* @@ -1185,7 +1237,7 @@ controller_create_fail: clk_src_create_fail: res_create_fail: - destruct(pool); + dce120_resource_destruct(pool); return false; } @@ -1200,9 +1252,10 @@ struct resource_pool *dce120_create_resource_pool( if (!pool) return NULL; - if (construct(num_virtual_links, dc, pool)) + if (dce120_resource_construct(num_virtual_links, dc, pool)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_timing_generator.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_timing_generator.c index 098e56962f2a..82bc4e192bbf 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_timing_generator.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_timing_generator.c @@ -819,13 +819,18 @@ void dce120_tg_set_colors(struct timing_generator *tg, static void dce120_timing_generator_set_static_screen_control( struct timing_generator *tg, - uint32_t value) + uint32_t event_triggers, + uint32_t num_frames) { struct dce110_timing_generator *tg110 = DCE110TG_FROM_TG(tg); + // By register spec, it only takes 8 bit value + if (num_frames > 0xFF) + num_frames = 0xFF; + CRTC_REG_UPDATE_2(CRTC0_CRTC_STATIC_SCREEN_CONTROL, - CRTC_STATIC_SCREEN_EVENT_MASK, value, - CRTC_STATIC_SCREEN_FRAME_COUNT, 2); + CRTC_STATIC_SCREEN_EVENT_MASK, event_triggers, + CRTC_STATIC_SCREEN_FRAME_COUNT, num_frames); } void dce120_timing_generator_set_test_pattern( diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c index c4543178ba20..893261c81854 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c @@ -74,7 +74,7 @@ void dce80_hw_sequencer_construct(struct dc *dc) { dce110_hw_sequencer_construct(dc); - dc->hwss.enable_display_power_gating = dce100_enable_display_power_gating; + dc->hwseq->funcs.enable_display_power_gating = dce100_enable_display_power_gating; dc->hwss.pipe_control_lock = dce_pipe_control_lock; dc->hwss.prepare_bandwidth = dce100_prepare_bandwidth; dc->hwss.optimize_bandwidth = dce100_optimize_bandwidth; diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.h index 7a1b31def66f..e43af832d00c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.h @@ -27,6 +27,7 @@ #define __DC_HWSS_DCE80_H__ #include "core_types.h" +#include "hw_sequencer_private.h" struct dc; diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c index 860a524ebcfa..2ad5c28c6e66 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c @@ -288,6 +288,14 @@ static const struct dce_opp_mask opp_mask = { OPP_COMMON_MASK_SH_LIST_DCE_80(_MASK) }; +static const struct dce110_aux_registers_shift aux_shift = { + DCE10_AUX_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce110_aux_registers_mask aux_mask = { + DCE10_AUX_MASK_SH_LIST(_MASK) +}; + #define aux_engine_regs(id)\ [id] = {\ AUX_COMMON_REG_LIST(id), \ @@ -322,7 +330,7 @@ static const struct dce_audio_shift audio_shift = { AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -431,6 +439,37 @@ static const struct dce_abm_mask abm_mask = { #define CC_DC_HDMI_STRAPS__AUDIO_STREAM_NUMBER__SHIFT 0x8 #endif +static int map_transmitter_id_to_phy_instance( + enum transmitter transmitter) +{ + switch (transmitter) { + case TRANSMITTER_UNIPHY_A: + return 0; + break; + case TRANSMITTER_UNIPHY_B: + return 1; + break; + case TRANSMITTER_UNIPHY_C: + return 2; + break; + case TRANSMITTER_UNIPHY_D: + return 3; + break; + case TRANSMITTER_UNIPHY_E: + return 4; + break; + case TRANSMITTER_UNIPHY_F: + return 5; + break; + case TRANSMITTER_UNIPHY_G: + return 6; + break; + default: + ASSERT(0); + return 0; + } +} + static void read_dce_straps( struct dc_context *ctx, struct resource_straps *straps) @@ -491,7 +530,10 @@ struct dce_aux *dce80_aux_engine_create( dce110_aux_engine_construct(aux_engine, ctx, inst, SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD, - &aux_engine_regs[inst]); + &aux_engine_regs[inst], + &aux_mask, + &aux_shift, + ctx->dc->caps.extended_aux_timeout_support); return &aux_engine->base; } @@ -669,14 +711,18 @@ struct link_encoder *dce80_link_encoder_create( { struct dce110_link_encoder *enc110 = kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); + int link_regs_id; if (!enc110) return NULL; + link_regs_id = + map_transmitter_id_to_phy_instance(enc_init_data->transmitter); + dce110_link_encoder_construct(enc110, enc_init_data, &link_enc_feature, - &link_enc_regs[enc_init_data->transmitter], + &link_enc_regs[link_regs_id], &link_enc_aux_regs[enc_init_data->channel - 1], &link_enc_hpd_regs[enc_init_data->hpd_source]); return &enc110->base; @@ -701,6 +747,7 @@ struct clock_source *dce80_clock_source_create( return &clk_src->base; } + kfree(clk_src); BREAK_TO_DEBUGGER(); return NULL; } @@ -726,7 +773,7 @@ static struct input_pixel_processor *dce80_ipp_create( return &ipp->base; } -static void destruct(struct dce110_resource_pool *pool) +static void dce80_resource_destruct(struct dce110_resource_pool *pool) { unsigned int i; @@ -854,7 +901,7 @@ static void dce80_destroy_resource_pool(struct resource_pool **pool) { struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool); - destruct(dce110_pool); + dce80_resource_destruct(dce110_pool); kfree(dce110_pool); *pool = NULL; } @@ -876,7 +923,6 @@ static bool dce80_construct( { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -895,6 +941,7 @@ static bool dce80_construct( dc->caps.i2c_speed_in_khz = 40; dc->caps.max_cursor_size = 128; dc->caps.dual_link_dvi = true; + dc->caps.extended_aux_timeout_support = false; /************************************************* * Create resources * @@ -902,8 +949,7 @@ static bool dce80_construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce80_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); @@ -1047,7 +1093,7 @@ static bool dce80_construct( return true; res_create_fail: - destruct(pool); + dce80_resource_destruct(pool); return false; } @@ -1075,7 +1121,6 @@ static bool dce81_construct( { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -1101,8 +1146,7 @@ static bool dce81_construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce80_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); @@ -1246,7 +1290,7 @@ static bool dce81_construct( return true; res_create_fail: - destruct(pool); + dce80_resource_destruct(pool); return false; } @@ -1274,7 +1318,6 @@ static bool dce83_construct( { unsigned int i; struct dc_context *ctx = dc->ctx; - struct dc_firmware_info info; struct dc_bios *bp; ctx->dc_bios->regs = &bios_regs; @@ -1300,8 +1343,7 @@ static bool dce83_construct( bp = ctx->dc_bios; - if ((bp->funcs->get_firmware_info(bp, &info) == BP_RESULT_OK) && - info.external_clock_source_frequency_for_dp != 0) { + if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) { pool->base.dp_clock_source = dce80_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true); @@ -1441,7 +1483,7 @@ static bool dce83_construct( return true; res_create_fail: - destruct(pool); + dce80_resource_destruct(pool); return false; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile index 032f872be89c..62ad1a11bff9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile @@ -22,7 +22,8 @@ # # Makefile for DCN. -DCN10 = dcn10_resource.o dcn10_ipp.o dcn10_hw_sequencer.o dcn10_hw_sequencer_debug.o \ +DCN10 = dcn10_init.o dcn10_resource.o dcn10_ipp.o dcn10_hw_sequencer.o \ + dcn10_hw_sequencer_debug.o \ dcn10_dpp.o dcn10_opp.o dcn10_optc.o \ dcn10_hubp.o dcn10_mpc.o \ dcn10_dpp_dscl.o dcn10_dpp_cm.o dcn10_cm_common.o \ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c index 7469333a2c8a..bbd6e01b3eca 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c @@ -343,8 +343,8 @@ bool cm_helper_translate_curve_to_hw_format( region_start = -MAX_LOW_POINT; region_end = NUMBER_REGIONS - MAX_LOW_POINT; } else { - /* 10 segments - * segment is from 2^-10 to 2^0 + /* 11 segments + * segment is from 2^-10 to 2^1 * There are less than 256 points, for optimization */ seg_distr[0] = 3; @@ -357,9 +357,10 @@ bool cm_helper_translate_curve_to_hw_format( seg_distr[7] = 4; seg_distr[8] = 4; seg_distr[9] = 4; + seg_distr[10] = 1; region_start = -10; - region_end = 0; + region_end = 1; } for (i = region_end - region_start; i < MAX_REGIONS_NUMBER ; i++) @@ -392,6 +393,10 @@ bool cm_helper_translate_curve_to_hw_format( rgb_resulted[hw_points - 1].green = output_tf->tf_pts.green[start_index]; rgb_resulted[hw_points - 1].blue = output_tf->tf_pts.blue[start_index]; + rgb_resulted[hw_points].red = rgb_resulted[hw_points - 1].red; + rgb_resulted[hw_points].green = rgb_resulted[hw_points - 1].green; + rgb_resulted[hw_points].blue = rgb_resulted[hw_points - 1].blue; + // All 3 color channels have same x corner_points[0].red.x = dc_fixpt_pow(dc_fixpt_from_int(2), dc_fixpt_from_int(region_start)); @@ -463,13 +468,6 @@ bool cm_helper_translate_curve_to_hw_format( i = 1; while (i != hw_points + 1) { - if (dc_fixpt_lt(rgb_plus_1->red, rgb->red)) - rgb_plus_1->red = rgb->red; - if (dc_fixpt_lt(rgb_plus_1->green, rgb->green)) - rgb_plus_1->green = rgb->green; - if (dc_fixpt_lt(rgb_plus_1->blue, rgb->blue)) - rgb_plus_1->blue = rgb->blue; - rgb->delta_red = dc_fixpt_sub(rgb_plus_1->red, rgb->red); rgb->delta_green = dc_fixpt_sub(rgb_plus_1->green, rgb->green); rgb->delta_blue = dc_fixpt_sub(rgb_plus_1->blue, rgb->blue); @@ -561,6 +559,10 @@ bool cm_helper_translate_curve_to_degamma_hw_format( rgb_resulted[hw_points - 1].green = output_tf->tf_pts.green[start_index]; rgb_resulted[hw_points - 1].blue = output_tf->tf_pts.blue[start_index]; + rgb_resulted[hw_points].red = rgb_resulted[hw_points - 1].red; + rgb_resulted[hw_points].green = rgb_resulted[hw_points - 1].green; + rgb_resulted[hw_points].blue = rgb_resulted[hw_points - 1].blue; + corner_points[0].red.x = dc_fixpt_pow(dc_fixpt_from_int(2), dc_fixpt_from_int(region_start)); corner_points[0].green.x = corner_points[0].red.x; @@ -623,13 +625,6 @@ bool cm_helper_translate_curve_to_degamma_hw_format( i = 1; while (i != hw_points + 1) { - if (dc_fixpt_lt(rgb_plus_1->red, rgb->red)) - rgb_plus_1->red = rgb->red; - if (dc_fixpt_lt(rgb_plus_1->green, rgb->green)) - rgb_plus_1->green = rgb->green; - if (dc_fixpt_lt(rgb_plus_1->blue, rgb->blue)) - rgb_plus_1->blue = rgb->blue; - rgb->delta_red = dc_fixpt_sub(rgb_plus_1->red, rgb->red); rgb->delta_green = dc_fixpt_sub(rgb_plus_1->green, rgb->green); rgb->delta_blue = dc_fixpt_sub(rgb_plus_1->blue, rgb->blue); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c index b95ec73fcae3..0e682b5aa3eb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c @@ -129,7 +129,7 @@ void dpp_set_gamut_remap_bypass(struct dcn10_dpp *dpp) #define IDENTITY_RATIO(ratio) (dc_fixpt_u2d19(ratio) == (1 << 19)) -static bool dpp_get_optimal_number_of_taps( +bool dpp1_get_optimal_number_of_taps( struct dpp *dpp, struct scaler_data *scl_data, const struct scaling_taps *in_taps) @@ -290,12 +290,8 @@ void dpp1_cnv_setup ( enum surface_pixel_format format, enum expansion_mode mode, struct dc_csc_transform input_csc_color_matrix, -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 enum dc_color_space input_color_space, struct cnv_alpha_2bit_lut *alpha_2bit_lut) -#else - enum dc_color_space input_color_space) -#endif { uint32_t pixel_format; uint32_t alpha_en; @@ -426,8 +422,9 @@ void dpp1_cnv_setup ( void dpp1_set_cursor_attributes( struct dpp *dpp_base, - enum dc_cursor_color_format color_format) + struct dc_cursor_attributes *cursor_attributes) { + enum dc_cursor_color_format color_format = cursor_attributes->color_format; struct dcn10_dpp *dpp = TO_DCN10_DPP(dpp_base); REG_UPDATE_2(CURSOR0_CONTROL, @@ -456,6 +453,19 @@ void dpp1_set_cursor_position( int src_y_offset = pos->y - pos->y_hotspot - param->viewport.y; uint32_t cur_en = pos->enable ? 1 : 0; + // Cursor width/height and hotspots need to be rotated for offset calculation + if (param->rotation == ROTATION_ANGLE_90 || param->rotation == ROTATION_ANGLE_270) { + swap(width, height); + if (param->rotation == ROTATION_ANGLE_90) { + src_x_offset = pos->x - pos->y_hotspot - param->viewport.x; + src_y_offset = pos->y - pos->x_hotspot - param->viewport.y; + } + } else if (param->rotation == ROTATION_ANGLE_180) { + src_x_offset = pos->x - param->viewport.x; + src_y_offset = pos->y - param->viewport.y; + } + + if (src_x_offset >= (int)param->viewport.width) cur_en = 0; /* not visible beyond right edge*/ @@ -507,7 +517,7 @@ static const struct dpp_funcs dcn10_dpp_funcs = { .dpp_read_state = dpp_read_state, .dpp_reset = dpp_reset, .dpp_set_scaler = dpp1_dscl_set_scaler_manual_scale, - .dpp_get_optimal_number_of_taps = dpp_get_optimal_number_of_taps, + .dpp_get_optimal_number_of_taps = dpp1_get_optimal_number_of_taps, .dpp_set_gamut_remap = dpp1_cm_set_gamut_remap, .dpp_set_csc_adjustment = dpp1_cm_set_output_csc_adjustment, .dpp_set_csc_default = dpp1_cm_set_output_csc_default, @@ -528,11 +538,9 @@ static const struct dpp_funcs dcn10_dpp_funcs = { .set_optional_cursor_attributes = dpp1_cnv_set_optional_cursor_attributes, .dpp_dppclk_control = dpp1_dppclk_control, .dpp_set_hdr_multiplier = dpp1_set_hdr_multiplier, -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) .dpp_program_blnd_lut = NULL, .dpp_program_shaper_lut = NULL, .dpp_program_3dlut = NULL -#endif }; static struct dpp_caps dcn10_dpp_cap = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h index 8a5517eebb7c..2edf566b3a72 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h @@ -1368,7 +1368,7 @@ enum dcn10_input_csc_select { void dpp1_set_cursor_attributes( struct dpp *dpp_base, - enum dc_cursor_color_format color_format); + struct dc_cursor_attributes *cursor_attributes); void dpp1_set_cursor_position( struct dpp *dpp_base, @@ -1486,12 +1486,8 @@ void dpp1_cnv_setup ( enum surface_pixel_format format, enum expansion_mode mode, struct dc_csc_transform input_csc_color_matrix, -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 enum dc_color_space input_color_space, struct cnv_alpha_2bit_lut *alpha_2bit_lut); -#else - enum dc_color_space input_color_space); -#endif void dpp1_full_bypass(struct dpp *dpp_base); @@ -1504,6 +1500,11 @@ void dpp1_set_hdr_multiplier( struct dpp *dpp_base, uint32_t multiplier); +bool dpp1_get_optimal_number_of_taps( + struct dpp *dpp, + struct scaler_data *scl_data, + const struct scaling_taps *in_taps); + void dpp1_construct(struct dcn10_dpp *dpp1, struct dc_context *ctx, uint32_t inst, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c index aa0c7a7d13a0..4d3f7d5e1473 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c @@ -88,26 +88,6 @@ enum dscl_mode_sel { DSCL_MODE_DSCL_BYPASS = 6 }; -static const struct dpp_input_csc_matrix dpp_input_csc_matrix[] = { - {COLOR_SPACE_SRGB, - {0x2000, 0, 0, 0, 0, 0x2000, 0, 0, 0, 0, 0x2000, 0} }, - {COLOR_SPACE_SRGB_LIMITED, - {0x2000, 0, 0, 0, 0, 0x2000, 0, 0, 0, 0, 0x2000, 0} }, - {COLOR_SPACE_YCBCR601, - {0x2cdd, 0x2000, 0, 0xe991, 0xe926, 0x2000, 0xf4fd, 0x10ef, - 0, 0x2000, 0x38b4, 0xe3a6} }, - {COLOR_SPACE_YCBCR601_LIMITED, - {0x3353, 0x2568, 0, 0xe400, 0xe5dc, 0x2568, 0xf367, 0x1108, - 0, 0x2568, 0x40de, 0xdd3a} }, - {COLOR_SPACE_YCBCR709, - {0x3265, 0x2000, 0, 0xe6ce, 0xf105, 0x2000, 0xfa01, 0xa7d, 0, - 0x2000, 0x3b61, 0xe24f} }, - - {COLOR_SPACE_YCBCR709_LIMITED, - {0x39a6, 0x2568, 0, 0xe0d6, 0xeedd, 0x2568, 0xf925, 0x9a8, 0, - 0x2568, 0x43ee, 0xdbb2} } -}; - static void program_gamut_remap( struct dcn10_dpp *dpp, const uint16_t *regval, @@ -352,6 +332,8 @@ void dpp1_cm_program_regamma_lut(struct dpp *dpp_base, uint32_t i; struct dcn10_dpp *dpp = TO_DCN10_DPP(dpp_base); + REG_SEQ_START(); + for (i = 0 ; i < num; i++) { REG_SET(CM_RGAM_LUT_DATA, 0, CM_RGAM_LUT_DATA, rgb[i].red_reg); REG_SET(CM_RGAM_LUT_DATA, 0, CM_RGAM_LUT_DATA, rgb[i].green_reg); @@ -626,10 +608,16 @@ void dpp1_set_degamma( case IPP_DEGAMMA_MODE_HW_xvYCC: REG_UPDATE(CM_DGAM_CONTROL, CM_DGAM_LUT_MODE, 2); break; + case IPP_DEGAMMA_MODE_USER_PWL: + REG_UPDATE(CM_DGAM_CONTROL, CM_DGAM_LUT_MODE, 3); + break; default: BREAK_TO_DEBUGGER(); break; } + + REG_SEQ_SUBMIT(); + REG_SEQ_WAIT_DONE(); } void dpp1_degamma_ram_select( @@ -731,10 +719,8 @@ void dpp1_full_bypass(struct dpp *dpp_base) /* COLOR_KEYER_CONTROL.COLOR_KEYER_EN = 0 this should be default */ if (dpp->tf_mask->CM_BYPASS_EN) REG_SET(CM_CONTROL, 0, CM_BYPASS_EN, 1); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) else REG_SET(CM_CONTROL, 0, CM_BYPASS, 1); -#endif /* Setting degamma bypass for now */ REG_SET(CM_DGAM_CONTROL, 0, CM_DGAM_LUT_MODE, 0); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c index d67e0abeee93..fce37c527a0b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c @@ -218,14 +218,12 @@ static void dpp1_dscl_set_lb( INTERLEAVE_EN, lb_params->interleave_en, /* Interleave source enable */ LB_DATA_FORMAT__ALPHA_EN, lb_params->alpha_en); /* Alpha enable */ } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) else { /* DSCL caps: pixel data processed in float format */ REG_SET_2(LB_DATA_FORMAT, 0, INTERLEAVE_EN, lb_params->interleave_en, /* Interleave source enable */ LB_DATA_FORMAT__ALPHA_EN, lb_params->alpha_en); /* Alpha enable */ } -#endif REG_SET_2(LB_MEMORY_CTRL, 0, MEMORY_CONFIG, mem_size_config, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dwb.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dwb.c index 374cc9acda3b..b6391a5ead78 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dwb.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dwb.c @@ -23,7 +23,7 @@ * */ -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) #include "reg_helper.h" #include "resource.h" @@ -109,9 +109,7 @@ const struct dwbc_funcs dcn10_dwbc_funcs = { .update = NULL, .set_stereo = NULL, .set_new_content = NULL, -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) .set_warmup = NULL, -#endif .dwb_set_scaler = NULL, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dwb.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dwb.h index c175edd0bae7..d56ea7c8171e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dwb.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dwb.h @@ -24,7 +24,7 @@ #ifndef __DC_DWBC_DCN10_H__ #define __DC_DWBC_DCN10_H__ -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) /* DCN */ #define BASE_INNER(seg) \ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c index a780057e2dbc..f36a0d8cedfe 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c @@ -104,7 +104,7 @@ void hubbub1_allow_self_refresh_control(struct hubbub *hubbub, bool allow) DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE, !allow); } -bool hububu1_is_allow_self_refresh_enabled(struct hubbub *hubbub) +bool hubbub1_is_allow_self_refresh_enabled(struct hubbub *hubbub) { struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub); uint32_t enable = 0; @@ -930,6 +930,9 @@ static bool hubbub1_get_dcc_compression_cap(struct hubbub *hubbub, output->grph.rgb.max_compressed_blk_size = 64; output->grph.rgb.independent_64b_blks = true; break; + default: + ASSERT(false); + break; } output->capable = true; @@ -945,6 +948,8 @@ static const struct hubbub_funcs hubbub1_funcs = { .get_dcc_compression_cap = hubbub1_get_dcc_compression_cap, .wm_read_state = hubbub1_wm_read_state, .program_watermarks = hubbub1_program_watermarks, + .is_allow_self_refresh_enabled = hubbub1_is_allow_self_refresh_enabled, + .allow_self_refresh_control = hubbub1_allow_self_refresh_control, }; void hubbub1_construct(struct hubbub *hubbub, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h index 7c2559c9ae23..af57751253de 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.h @@ -119,6 +119,26 @@ struct dcn_hubbub_registers { uint32_t DCN_VM_AGP_BOT; uint32_t DCN_VM_AGP_TOP; uint32_t DCN_VM_AGP_BASE; + uint32_t DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB; + uint32_t DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_NOM_A; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_NOM_B; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_NOM_C; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_NOM_D; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C; + uint32_t DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D; + uint32_t DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A; + uint32_t DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B; + uint32_t DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C; + uint32_t DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D; + uint32_t DCHUBBUB_ARB_HOSTVM_CNTL; + uint32_t DCHVM_CTRL0; + uint32_t DCHVM_MEM_CTRL; + uint32_t DCHVM_CLK_CTRL; + uint32_t DCHVM_RIOMMU_CTRL0; + uint32_t DCHVM_RIOMMU_STAT0; }; /* set field name */ @@ -196,7 +216,9 @@ struct dcn_hubbub_registers { type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A;\ type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B;\ type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C;\ - type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D + type DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D;\ + type DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB;\ + type DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB #define HUBBUB_STUTTER_REG_FIELD_LIST(type) \ type DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A;\ @@ -208,15 +230,62 @@ struct dcn_hubbub_registers { type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C;\ type DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D +#define HUBBUB_HVM_REG_FIELD_LIST(type) \ + type DCHUBBUB_ARB_MIN_REQ_OUTSTAND_COMMIT_THRESHOLD;\ + type DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_A;\ + type DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_B;\ + type DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_C;\ + type DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_D;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_A;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_B;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_C;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_D;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_B;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_C;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_D;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C;\ + type DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D;\ + type DCHUBBUB_ARB_FRAC_URG_BW_NOM_A;\ + type DCHUBBUB_ARB_FRAC_URG_BW_NOM_B;\ + type DCHUBBUB_ARB_FRAC_URG_BW_NOM_C;\ + type DCHUBBUB_ARB_FRAC_URG_BW_NOM_D;\ + type DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A;\ + type DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B;\ + type DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C;\ + type DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D;\ + type DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A;\ + type DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B;\ + type DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C;\ + type DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D;\ + type DCHUBBUB_ARB_MAX_QOS_COMMIT_THRESHOLD;\ + type HOSTVM_INIT_REQ; \ + type HVM_GPUVMRET_PWR_REQ_DIS; \ + type HVM_GPUVMRET_FORCE_REQ; \ + type HVM_GPUVMRET_POWER_STATUS; \ + type HVM_DISPCLK_R_GATE_DIS; \ + type HVM_DISPCLK_G_GATE_DIS; \ + type HVM_DCFCLK_R_GATE_DIS; \ + type HVM_DCFCLK_G_GATE_DIS; \ + type TR_REQ_REQCLKREQ_MODE; \ + type TW_RSP_COMPCLKREQ_MODE; \ + type HOSTVM_PREFETCH_REQ; \ + type HOSTVM_POWERSTATUS; \ + type RIOMMU_ACTIVE; \ + type HOSTVM_PREFETCH_DONE struct dcn_hubbub_shift { DCN_HUBBUB_REG_FIELD_LIST(uint8_t); HUBBUB_STUTTER_REG_FIELD_LIST(uint8_t); + HUBBUB_HVM_REG_FIELD_LIST(uint8_t); }; struct dcn_hubbub_mask { DCN_HUBBUB_REG_FIELD_LIST(uint32_t); HUBBUB_STUTTER_REG_FIELD_LIST(uint32_t); + HUBBUB_HVM_REG_FIELD_LIST(uint32_t); }; struct dc; @@ -247,7 +316,7 @@ void hubbub1_program_watermarks( void hubbub1_allow_self_refresh_control(struct hubbub *hubbub, bool allow); -bool hububu1_is_allow_self_refresh_enabled(struct hubbub *hubub); +bool hubbub1_is_allow_self_refresh_enabled(struct hubbub *hubub); void hubbub1_toggle_watermark_change_req( struct hubbub *hubbub); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c index 934bacc0c6ad..31b64733d693 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -163,7 +163,7 @@ void hubp1_program_tiling( void hubp1_program_size( struct hubp *hubp, enum surface_pixel_format format, - const union plane_size *plane_size, + const struct plane_size *plane_size, struct dc_plane_dcc_param *dcc) { struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); @@ -173,16 +173,16 @@ void hubp1_program_size( * 444 or 420 luma */ if (format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN && format < SURFACE_PIXEL_FORMAT_SUBSAMPLE_END) { - ASSERT(plane_size->video.chroma_pitch != 0); + ASSERT(plane_size->chroma_pitch != 0); /* Chroma pitch zero can cause system hang! */ - pitch = plane_size->video.luma_pitch - 1; - meta_pitch = dcc->video.meta_pitch_l - 1; - pitch_c = plane_size->video.chroma_pitch - 1; - meta_pitch_c = dcc->video.meta_pitch_c - 1; + pitch = plane_size->surface_pitch - 1; + meta_pitch = dcc->meta_pitch - 1; + pitch_c = plane_size->chroma_pitch - 1; + meta_pitch_c = dcc->meta_pitch_c - 1; } else { - pitch = plane_size->grph.surface_pitch - 1; - meta_pitch = dcc->grph.meta_pitch - 1; + pitch = plane_size->surface_pitch - 1; + meta_pitch = dcc->meta_pitch - 1; pitch_c = 0; meta_pitch_c = 0; } @@ -306,7 +306,6 @@ void hubp1_program_pixel_format( REG_UPDATE(DCSURF_SURFACE_CONFIG, SURFACE_PIXEL_FORMAT, 12); break; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) case SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FIX: REG_UPDATE(DCSURF_SURFACE_CONFIG, SURFACE_PIXEL_FORMAT, 112); @@ -327,7 +326,6 @@ void hubp1_program_pixel_format( REG_UPDATE(DCSURF_SURFACE_CONFIG, SURFACE_PIXEL_FORMAT, 119); break; -#endif default: BREAK_TO_DEBUGGER(); break; @@ -509,7 +507,7 @@ bool hubp1_program_surface_flip_and_addr( } void hubp1_dcc_control(struct hubp *hubp, bool enable, - bool independent_64b_blks) + enum hubp_ind_block_size independent_64b_blks) { uint32_t dcc_en = enable ? 1 : 0; uint32_t dcc_ind_64b_blk = independent_64b_blks ? 1 : 0; @@ -526,13 +524,13 @@ void hubp1_program_surface_config( struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, unsigned int compat_level) { - hubp1_dcc_control(hubp, dcc->enable, dcc->grph.independent_64b_blks); + hubp1_dcc_control(hubp, dcc->enable, dcc->independent_64b_blks); hubp1_program_tiling(hubp, tiling_info, format); hubp1_program_size(hubp, format, plane_size, dcc); hubp1_program_rotation(hubp, rotation, horizontal_mirror); @@ -841,9 +839,17 @@ void min_set_viewport( REG_SET_2(DCSURF_PRI_VIEWPORT_START_C, 0, PRI_VIEWPORT_X_START_C, viewport_c->x, PRI_VIEWPORT_Y_START_C, viewport_c->y); + + REG_SET_2(DCSURF_SEC_VIEWPORT_DIMENSION_C, 0, + SEC_VIEWPORT_WIDTH_C, viewport_c->width, + SEC_VIEWPORT_HEIGHT_C, viewport_c->height); + + REG_SET_2(DCSURF_SEC_VIEWPORT_START_C, 0, + SEC_VIEWPORT_X_START_C, viewport_c->x, + SEC_VIEWPORT_Y_START_C, viewport_c->y); } -void hubp1_read_state(struct hubp *hubp) +void hubp1_read_state_common(struct hubp *hubp) { struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); struct dcn_hubp_state *s = &hubp1->state; @@ -859,24 +865,6 @@ void hubp1_read_state(struct hubp *hubp) PRQ_EXPANSION_MODE, &rq_regs->prq_expansion_mode, MRQ_EXPANSION_MODE, &rq_regs->mrq_expansion_mode, CRQ_EXPANSION_MODE, &rq_regs->crq_expansion_mode); - REG_GET_8(DCHUBP_REQ_SIZE_CONFIG, - CHUNK_SIZE, &rq_regs->rq_regs_l.chunk_size, - MIN_CHUNK_SIZE, &rq_regs->rq_regs_l.min_chunk_size, - META_CHUNK_SIZE, &rq_regs->rq_regs_l.meta_chunk_size, - MIN_META_CHUNK_SIZE, &rq_regs->rq_regs_l.min_meta_chunk_size, - DPTE_GROUP_SIZE, &rq_regs->rq_regs_l.dpte_group_size, - MPTE_GROUP_SIZE, &rq_regs->rq_regs_l.mpte_group_size, - SWATH_HEIGHT, &rq_regs->rq_regs_l.swath_height, - PTE_ROW_HEIGHT_LINEAR, &rq_regs->rq_regs_l.pte_row_height_linear); - REG_GET_8(DCHUBP_REQ_SIZE_CONFIG_C, - CHUNK_SIZE_C, &rq_regs->rq_regs_c.chunk_size, - MIN_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_chunk_size, - META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.meta_chunk_size, - MIN_META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_meta_chunk_size, - DPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.dpte_group_size, - MPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.mpte_group_size, - SWATH_HEIGHT_C, &rq_regs->rq_regs_c.swath_height, - PTE_ROW_HEIGHT_LINEAR_C, &rq_regs->rq_regs_c.pte_row_height_linear); /* DLG - Per hubp */ REG_GET_2(BLANK_OFFSET_0, @@ -1024,14 +1012,47 @@ void hubp1_read_state(struct hubp *hubp) HUBP_TTU_DISABLE, &s->ttu_disable, HUBP_UNDERFLOW_STATUS, &s->underflow_status); + REG_GET(HUBP_CLK_CNTL, + HUBP_CLOCK_ENABLE, &s->clock_en); + REG_GET(DCN_GLOBAL_TTU_CNTL, MIN_TTU_VBLANK, &s->min_ttu_vblank); REG_GET_2(DCN_TTU_QOS_WM, QoS_LEVEL_LOW_WM, &s->qos_level_low_wm, QoS_LEVEL_HIGH_WM, &s->qos_level_high_wm); + } +void hubp1_read_state(struct hubp *hubp) +{ + struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); + struct dcn_hubp_state *s = &hubp1->state; + struct _vcs_dpi_display_rq_regs_st *rq_regs = &s->rq_regs; + + hubp1_read_state_common(hubp); + + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG, + CHUNK_SIZE, &rq_regs->rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, &rq_regs->rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, &rq_regs->rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, &rq_regs->rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, &rq_regs->rq_regs_l.dpte_group_size, + MPTE_GROUP_SIZE, &rq_regs->rq_regs_l.mpte_group_size, + SWATH_HEIGHT, &rq_regs->rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, &rq_regs->rq_regs_l.pte_row_height_linear); + + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG_C, + CHUNK_SIZE_C, &rq_regs->rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.dpte_group_size, + MPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.mpte_group_size, + SWATH_HEIGHT_C, &rq_regs->rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, &rq_regs->rq_regs_c.pte_row_height_linear); + +} enum cursor_pitch hubp1_get_cursor_pitch(unsigned int pitch) { enum cursor_pitch hw_pitch; @@ -1228,10 +1249,8 @@ static const struct hubp_funcs dcn10_hubp_funcs = { .hubp_get_underflow_status = hubp1_get_underflow_status, .hubp_init = hubp1_init, -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) .dmdata_set_attributes = NULL, .dmdata_load = NULL, -#endif }; /*****************************************/ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h index 31c8fdd3206c..780af5b3c16f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h @@ -47,6 +47,8 @@ SRI(DCSURF_SEC_VIEWPORT_START, HUBP, id), \ SRI(DCSURF_PRI_VIEWPORT_DIMENSION_C, HUBP, id), \ SRI(DCSURF_PRI_VIEWPORT_START_C, HUBP, id), \ + SRI(DCSURF_SEC_VIEWPORT_DIMENSION_C, HUBP, id), \ + SRI(DCSURF_SEC_VIEWPORT_START_C, HUBP, id), \ SRI(DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH, HUBPREQ, id),\ SRI(DCSURF_PRIMARY_SURFACE_ADDRESS, HUBPREQ, id),\ SRI(DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH, HUBPREQ, id),\ @@ -57,8 +59,12 @@ SRI(DCSURF_SECONDARY_META_SURFACE_ADDRESS, HUBPREQ, id),\ SRI(DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C, HUBPREQ, id),\ SRI(DCSURF_PRIMARY_SURFACE_ADDRESS_C, HUBPREQ, id),\ + SRI(DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH_C, HUBPREQ, id),\ + SRI(DCSURF_SECONDARY_SURFACE_ADDRESS_C, HUBPREQ, id),\ SRI(DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH_C, HUBPREQ, id),\ SRI(DCSURF_PRIMARY_META_SURFACE_ADDRESS_C, HUBPREQ, id),\ + SRI(DCSURF_SECONDARY_META_SURFACE_ADDRESS_HIGH_C, HUBPREQ, id),\ + SRI(DCSURF_SECONDARY_META_SURFACE_ADDRESS_C, HUBPREQ, id),\ SRI(DCSURF_SURFACE_INUSE, HUBPREQ, id),\ SRI(DCSURF_SURFACE_INUSE_HIGH, HUBPREQ, id),\ SRI(DCSURF_SURFACE_INUSE_C, HUBPREQ, id),\ @@ -125,8 +131,6 @@ SRI(DCN_VM_SYSTEM_APERTURE_LOW_ADDR_LSB, HUBPREQ, id),\ SRI(DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_MSB, HUBPREQ, id),\ SRI(DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB, HUBPREQ, id),\ - SR(DCHUBBUB_SDPIF_FB_BASE),\ - SR(DCHUBBUB_SDPIF_FB_OFFSET),\ SRI(CURSOR_SETTINS, HUBPREQ, id), \ SRI(CURSOR_SURFACE_ADDRESS_HIGH, CURSOR, id), \ SRI(CURSOR_SURFACE_ADDRESS, CURSOR, id), \ @@ -152,6 +156,8 @@ uint32_t DCSURF_SEC_VIEWPORT_START; \ uint32_t DCSURF_PRI_VIEWPORT_DIMENSION_C; \ uint32_t DCSURF_PRI_VIEWPORT_START_C; \ + uint32_t DCSURF_SEC_VIEWPORT_DIMENSION_C; \ + uint32_t DCSURF_SEC_VIEWPORT_START_C; \ uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH; \ uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS; \ uint32_t DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH; \ @@ -162,8 +168,12 @@ uint32_t DCSURF_SECONDARY_META_SURFACE_ADDRESS; \ uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C; \ uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_C; \ + uint32_t DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH_C; \ + uint32_t DCSURF_SECONDARY_SURFACE_ADDRESS_C; \ uint32_t DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH_C; \ uint32_t DCSURF_PRIMARY_META_SURFACE_ADDRESS_C; \ + uint32_t DCSURF_SECONDARY_META_SURFACE_ADDRESS_HIGH_C; \ + uint32_t DCSURF_SECONDARY_META_SURFACE_ADDRESS_C; \ uint32_t DCSURF_SURFACE_INUSE; \ uint32_t DCSURF_SURFACE_INUSE_HIGH; \ uint32_t DCSURF_SURFACE_INUSE_C; \ @@ -226,14 +236,6 @@ uint32_t DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB; \ uint32_t DCN_VM_SYSTEM_APERTURE_LOW_ADDR; \ uint32_t DCN_VM_SYSTEM_APERTURE_HIGH_ADDR; \ - uint32_t DCHUBBUB_SDPIF_FB_BASE; \ - uint32_t DCHUBBUB_SDPIF_FB_OFFSET; \ - uint32_t DCN_VM_FB_LOCATION_TOP; \ - uint32_t DCN_VM_FB_LOCATION_BASE; \ - uint32_t DCN_VM_FB_OFFSET; \ - uint32_t DCN_VM_AGP_BASE; \ - uint32_t DCN_VM_AGP_BOT; \ - uint32_t DCN_VM_AGP_TOP; \ uint32_t CURSOR_SETTINS; \ uint32_t CURSOR_SETTINGS; \ uint32_t CURSOR_SURFACE_ADDRESS_HIGH; \ @@ -249,7 +251,8 @@ .field_name = reg_name ## __ ## field_name ## post_fix /* Mask/shift struct generation macro for all ASICs (including those with reduced functionality) */ -#define HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh)\ +/*1.x, 2.x, and 3.x*/ +#define HUBP_MASK_SH_LIST_DCN_SHARE_COMMON(mask_sh)\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_BLANK_EN, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_TTU_DISABLE, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_UNDERFLOW_STATUS, mask_sh),\ @@ -265,7 +268,6 @@ HUBP_SF(HUBP0_DCSURF_ADDR_CONFIG, MAX_COMPRESSED_FRAGS, mask_sh),\ HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, SW_MODE, mask_sh),\ HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, META_LINEAR, mask_sh),\ - HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, RB_ALIGNED, mask_sh),\ HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, PIPE_ALIGNED, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_SURFACE_PITCH, PITCH, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_SURFACE_PITCH, META_PITCH, mask_sh),\ @@ -289,6 +291,10 @@ HUBP_SF(HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_C, PRI_VIEWPORT_HEIGHT_C, mask_sh),\ HUBP_SF(HUBP0_DCSURF_PRI_VIEWPORT_START_C, PRI_VIEWPORT_X_START_C, mask_sh),\ HUBP_SF(HUBP0_DCSURF_PRI_VIEWPORT_START_C, PRI_VIEWPORT_Y_START_C, mask_sh),\ + HUBP_SF(HUBP0_DCSURF_SEC_VIEWPORT_DIMENSION_C, SEC_VIEWPORT_WIDTH_C, mask_sh),\ + HUBP_SF(HUBP0_DCSURF_SEC_VIEWPORT_DIMENSION_C, SEC_VIEWPORT_HEIGHT_C, mask_sh),\ + HUBP_SF(HUBP0_DCSURF_SEC_VIEWPORT_START_C, SEC_VIEWPORT_X_START_C, mask_sh),\ + HUBP_SF(HUBP0_DCSURF_SEC_VIEWPORT_START_C, SEC_VIEWPORT_Y_START_C, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH, PRIMARY_SURFACE_ADDRESS_HIGH, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_PRIMARY_SURFACE_ADDRESS, PRIMARY_SURFACE_ADDRESS, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH, SECONDARY_SURFACE_ADDRESS_HIGH, mask_sh),\ @@ -299,8 +305,12 @@ HUBP_SF(HUBPREQ0_DCSURF_SECONDARY_META_SURFACE_ADDRESS, SECONDARY_META_SURFACE_ADDRESS, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C, PRIMARY_SURFACE_ADDRESS_HIGH_C, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_PRIMARY_SURFACE_ADDRESS_C, PRIMARY_SURFACE_ADDRESS_C, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH_C, SECONDARY_SURFACE_ADDRESS_HIGH_C, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_SECONDARY_SURFACE_ADDRESS_C, SECONDARY_SURFACE_ADDRESS_C, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH_C, PRIMARY_META_SURFACE_ADDRESS_HIGH_C, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_PRIMARY_META_SURFACE_ADDRESS_C, PRIMARY_META_SURFACE_ADDRESS_C, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_SECONDARY_META_SURFACE_ADDRESS_HIGH_C, SECONDARY_META_SURFACE_ADDRESS_HIGH_C, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_SECONDARY_META_SURFACE_ADDRESS_C, SECONDARY_META_SURFACE_ADDRESS_C, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_SURFACE_INUSE, SURFACE_INUSE_ADDRESS, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_SURFACE_INUSE_HIGH, SURFACE_INUSE_ADDRESS_HIGH, mask_sh),\ HUBP_SF(HUBPREQ0_DCSURF_SURFACE_INUSE_C, SURFACE_INUSE_ADDRESS_C, mask_sh),\ @@ -372,12 +382,17 @@ HUBP_SF(HUBPREQ0_DCN_SURF0_TTU_CNTL0, QoS_RAMP_DISABLE, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_SURF0_TTU_CNTL1, REFCYC_PER_REQ_DELIVERY_PRE, mask_sh),\ HUBP_SF(HUBP0_HUBP_CLK_CNTL, HUBP_CLOCK_ENABLE, mask_sh) - -#define HUBP_MASK_SH_LIST_DCN(mask_sh)\ - HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh),\ +/*2.x and 1.x only*/ +#define HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh)\ + HUBP_MASK_SH_LIST_DCN_SHARE_COMMON(mask_sh),\ + HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, RB_ALIGNED, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, MPTE_GROUP_SIZE, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, MPTE_GROUP_SIZE_C, mask_sh) +/*2.x and 1.x only*/ +#define HUBP_MASK_SH_LIST_DCN(mask_sh)\ + HUBP_MASK_SH_LIST_DCN_COMMON(mask_sh) + /* Mask/shift struct generation macro for ASICs with VM */ #define HUBP_MASK_SH_LIST_DCN_VM(mask_sh)\ HUBP_SF(HUBPREQ0_NOM_PARAMETERS_0, DST_Y_PER_PTE_ROW_NOM_L, mask_sh),\ @@ -412,8 +427,6 @@ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_LOW_ADDR_LSB, MC_VM_SYSTEM_APERTURE_LOW_ADDR_LSB, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_MSB, MC_VM_SYSTEM_APERTURE_HIGH_ADDR_MSB, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB, MC_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB, mask_sh),\ - HUBP_SF(DCHUBBUB_SDPIF_FB_BASE, SDPIF_FB_BASE, mask_sh),\ - HUBP_SF(DCHUBBUB_SDPIF_FB_OFFSET, SDPIF_FB_OFFSET, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, MC_VM_SYSTEM_APERTURE_DEFAULT_SYSTEM, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, mask_sh),\ @@ -434,7 +447,7 @@ HUBP_SF(CURSOR0_CURSOR_HOT_SPOT, CURSOR_HOT_SPOT_Y, mask_sh), \ HUBP_SF(CURSOR0_CURSOR_DST_OFFSET, CURSOR_DST_X_OFFSET, mask_sh) -#define DCN_HUBP_REG_FIELD_LIST(type) \ +#define DCN_HUBP_REG_FIELD_BASE_LIST(type) \ type HUBP_BLANK_EN;\ type HUBP_DISABLE;\ type HUBP_TTU_DISABLE;\ @@ -459,7 +472,6 @@ type ROTATION_ANGLE;\ type H_MIRROR_EN;\ type SURFACE_PIXEL_FORMAT;\ - type ALPHA_PLANE_EN;\ type SURFACE_FLIP_TYPE;\ type SURFACE_FLIP_MODE_FOR_STEREOSYNC;\ type SURFACE_FLIP_IN_STEREOSYNC;\ @@ -477,6 +489,10 @@ type PRI_VIEWPORT_HEIGHT_C; \ type PRI_VIEWPORT_X_START_C; \ type PRI_VIEWPORT_Y_START_C; \ + type SEC_VIEWPORT_WIDTH_C; \ + type SEC_VIEWPORT_HEIGHT_C; \ + type SEC_VIEWPORT_X_START_C; \ + type SEC_VIEWPORT_Y_START_C; \ type PRIMARY_SURFACE_ADDRESS_HIGH;\ type PRIMARY_SURFACE_ADDRESS;\ type SECONDARY_SURFACE_ADDRESS_HIGH;\ @@ -487,8 +503,12 @@ type SECONDARY_META_SURFACE_ADDRESS;\ type PRIMARY_SURFACE_ADDRESS_HIGH_C;\ type PRIMARY_SURFACE_ADDRESS_C;\ + type SECONDARY_SURFACE_ADDRESS_HIGH_C;\ + type SECONDARY_SURFACE_ADDRESS_C;\ type PRIMARY_META_SURFACE_ADDRESS_HIGH_C;\ type PRIMARY_META_SURFACE_ADDRESS_C;\ + type SECONDARY_META_SURFACE_ADDRESS_HIGH_C;\ + type SECONDARY_META_SURFACE_ADDRESS_C;\ type SURFACE_INUSE_ADDRESS;\ type SURFACE_INUSE_ADDRESS_HIGH;\ type SURFACE_INUSE_ADDRESS_C;\ @@ -589,18 +609,6 @@ type MC_VM_SYSTEM_APERTURE_HIGH_ADDR_LSB;\ type MC_VM_SYSTEM_APERTURE_LOW_ADDR;\ type MC_VM_SYSTEM_APERTURE_HIGH_ADDR;\ - type SDPIF_FB_TOP;\ - type SDPIF_FB_BASE;\ - type SDPIF_FB_OFFSET;\ - type SDPIF_AGP_BASE;\ - type SDPIF_AGP_BOT;\ - type SDPIF_AGP_TOP;\ - type FB_TOP;\ - type FB_BASE;\ - type FB_OFFSET;\ - type AGP_BASE;\ - type AGP_BOT;\ - type AGP_TOP;\ type DCN_VM_SYSTEM_APERTURE_DEFAULT_SYSTEM;\ type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB;\ type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB;\ @@ -632,6 +640,10 @@ type CURSOR_DST_X_OFFSET; \ type OUTPUT_FP +#define DCN_HUBP_REG_FIELD_LIST(type) \ + DCN_HUBP_REG_FIELD_BASE_LIST(type);\ + type ALPHA_PLANE_EN + struct dcn_mi_registers { HUBP_COMMON_REG_VARIABLE_LIST; }; @@ -658,6 +670,7 @@ struct dcn_hubp_state { uint32_t sw_mode; uint32_t dcc_en; uint32_t blank_en; + uint32_t clock_en; uint32_t underflow_status; uint32_t ttu_disable; uint32_t min_ttu_vblank; @@ -677,7 +690,7 @@ void hubp1_program_surface_config( struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, @@ -699,7 +712,7 @@ void hubp1_program_pixel_format( void hubp1_program_size( struct hubp *hubp, enum surface_pixel_format format, - const union plane_size *plane_size, + const struct plane_size *plane_size, struct dc_plane_dcc_param *dcc); void hubp1_program_rotation( @@ -714,15 +727,13 @@ void hubp1_program_tiling( void hubp1_dcc_control(struct hubp *hubp, bool enable, - bool independent_64b_blks); + enum hubp_ind_block_size independent_64b_blks); -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 bool hubp1_program_surface_flip_and_addr( struct hubp *hubp, const struct dc_plane_address *address, bool flip_immediate); -#endif bool hubp1_is_flip_pending(struct hubp *hubp); void hubp1_cursor_set_attributes( @@ -760,5 +771,6 @@ void hubp1_vready_workaround(struct hubp *hubp, struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest); void hubp1_init(struct hubp *hubp); +void hubp1_read_state_common(struct hubp *hubp); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index 2118ea21d7e9..1008ac8a0f2a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -25,17 +25,18 @@ #include <linux/delay.h> #include "dm_services.h" +#include "basics/dc_common.h" #include "core_types.h" #include "resource.h" #include "custom_float.h" #include "dcn10_hw_sequencer.h" -#include "dce110/dce110_hw_sequencer.h" +#include "dcn10_hw_sequencer_debug.h" #include "dce/dce_hwseq.h" #include "abm.h" #include "dmcu.h" #include "dcn10_optc.h" -#include "dcn10/dcn10_dpp.h" -#include "dcn10/dcn10_mpc.h" +#include "dcn10_dpp.h" +#include "dcn10_mpc.h" #include "timing_generator.h" #include "opp.h" #include "ipp.h" @@ -49,9 +50,7 @@ #include "clk_mgr.h" -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #include "dsc.h" -#endif #define DC_LOGGER_INIT(logger) @@ -68,6 +67,8 @@ #define DTN_INFO_MICRO_SEC(ref_cycle) \ print_microsec(dc_ctx, log_ctx, ref_cycle) +#define GAMMA_HW_POINTS_NUM 256 + void print_microsec(struct dc_context *dc_ctx, struct dc_log_buffer_ctx *log_ctx, uint32_t ref_cycle) @@ -81,6 +82,33 @@ void print_microsec(struct dc_context *dc_ctx, us_x10 % frac); } +static void dcn10_lock_all_pipes(struct dc *dc, + struct dc_state *context, + bool lock) +{ + struct pipe_ctx *pipe_ctx; + struct timing_generator *tg; + int i; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe_ctx = &context->res_ctx.pipe_ctx[i]; + tg = pipe_ctx->stream_res.tg; + /* + * Only lock the top pipe's tg to prevent redundant + * (un)locking. Also skip if pipe is disabled. + */ + if (pipe_ctx->top_pipe || + !pipe_ctx->stream || !pipe_ctx->plane_state || + !tg->funcs->is_tg_enabled(tg)) + continue; + + if (lock) + tg->funcs->lock(tg); + else + tg->funcs->unlock(tg); + } +} + static void log_mpc_crc(struct dc *dc, struct dc_log_buffer_ctx *log_ctx) { @@ -129,9 +157,8 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx) struct resource_pool *pool = dc->res_pool; int i; - DTN_INFO("HUBP: format addr_hi width height" - " rot mir sw_mode dcc_en blank_en ttu_dis underflow" - " min_ttu_vblank qos_low_wm qos_high_wm\n"); + DTN_INFO( + "HUBP: format addr_hi width height rot mir sw_mode dcc_en blank_en clock_en ttu_dis underflow min_ttu_vblank qos_low_wm qos_high_wm\n"); for (i = 0; i < pool->pipe_count; i++) { struct hubp *hubp = pool->hubps[i]; struct dcn_hubp_state *s = &(TO_DCN10_HUBP(hubp)->state); @@ -139,8 +166,7 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx) hubp->funcs->hubp_read_state(hubp); if (!s->blank_en) { - DTN_INFO("[%2d]: %5xh %6xh %5d %6d %2xh %2xh %6xh" - " %6d %8d %7d %8xh", + DTN_INFO("[%2d]: %5xh %6xh %5d %6d %2xh %2xh %6xh %6d %8d %8d %7d %8xh", hubp->inst, s->pixel_format, s->inuse_addr_hi, @@ -151,6 +177,7 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx) s->sw_mode, s->dcc_en, s->blank_en, + s->clock_en, s->ttu_disable, s->underflow_status); DTN_INFO_MICRO_SEC(s->min_ttu_vblank); @@ -308,21 +335,31 @@ void dcn10_log_hw_state(struct dc *dc, } DTN_INFO("\n"); - DTN_INFO("OTG: v_bs v_be v_ss v_se vpol vmax vmin vmax_sel vmin_sel" - " h_bs h_be h_ss h_se hpol htot vtot underflow\n"); + DTN_INFO("OTG: v_bs v_be v_ss v_se vpol vmax vmin vmax_sel vmin_sel h_bs h_be h_ss h_se hpol htot vtot underflow blank_en\n"); for (i = 0; i < pool->timing_generator_count; i++) { struct timing_generator *tg = pool->timing_generators[i]; struct dcn_otg_state s = {0}; - + /* Read shared OTG state registers for all DCNx */ optc1_read_otg_state(DCN10TG_FROM_TG(tg), &s); + /* + * For DCN2 and greater, a register on the OPP is used to + * determine if the CRTC is blanked instead of the OTG. So use + * dpg_is_blanked() if exists, otherwise fallback on otg. + * + * TODO: Implement DCN-specific read_otg_state hooks. + */ + if (pool->opps[i]->funcs->dpg_is_blanked) + s.blank_enabled = pool->opps[i]->funcs->dpg_is_blanked(pool->opps[i]); + else + s.blank_enabled = tg->funcs->is_blanked(tg); + //only print if OTG master is enabled if ((s.otg_enabled & 1) == 0) continue; - DTN_INFO("[%d]: %5d %5d %5d %5d %5d %5d %5d %9d %9d %5d %5d %5d" - " %5d %5d %5d %5d %9d\n", + DTN_INFO("[%d]: %5d %5d %5d %5d %5d %5d %5d %9d %9d %5d %5d %5d %5d %5d %5d %5d %9d %8d\n", tg->inst, s.v_blank_start, s.v_blank_end, @@ -340,7 +377,8 @@ void dcn10_log_hw_state(struct dc *dc, s.h_sync_a_pol, s.h_total, s.v_total, - s.underflow_occurred_status); + s.underflow_occurred_status, + s.blank_enabled); // Clear underflow for debug purposes // We want to keep underflow sticky bit on for the longevity tests outside of test environment. @@ -350,7 +388,6 @@ void dcn10_log_hw_state(struct dc *dc, } DTN_INFO("\n"); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT DTN_INFO("DSC: CLOCK_EN SLICE_WIDTH Bytes_pp\n"); for (i = 0; i < pool->res_cap->num_dsc; i++) { struct display_stream_compressor *dsc = pool->dscs[i]; @@ -387,7 +424,7 @@ void dcn10_log_hw_state(struct dc *dc, } DTN_INFO("\n"); - DTN_INFO("L_ENC: DPHY_FEC_EN DPHY_FEC_READY_SHADOW DPHY_FEC_ACTIVE_STATUS\n"); + DTN_INFO("L_ENC: DPHY_FEC_EN DPHY_FEC_READY_SHADOW DPHY_FEC_ACTIVE_STATUS DP_LINK_TRAINING_COMPLETE\n"); for (i = 0; i < dc->link_count; i++) { struct link_encoder *lenc = dc->links[i]->link_enc; @@ -395,16 +432,16 @@ void dcn10_log_hw_state(struct dc *dc, if (lenc->funcs->read_state) { lenc->funcs->read_state(lenc, &s); - DTN_INFO("[%-3d]: %-12d %-22d %-22d\n", + DTN_INFO("[%-3d]: %-12d %-22d %-22d %-25d\n", i, s.dphy_fec_en, s.dphy_fec_ready_shadow, - s.dphy_fec_active_status); + s.dphy_fec_active_status, + s.dp_link_training_complete); DTN_INFO("\n"); } } DTN_INFO("\n"); -#endif DTN_INFO("\nCALCULATED Clocks: dcfclk_khz:%d dcfclk_deep_sleep_khz:%d dispclk_khz:%d\n" "dppclk_khz:%d max_supported_dppclk_khz:%d fclk_khz:%d socclk_khz:%d\n\n", @@ -438,14 +475,14 @@ bool dcn10_did_underflow_occur(struct dc *dc, struct pipe_ctx *pipe_ctx) return false; } -static void enable_power_gating_plane( +void dcn10_enable_power_gating_plane( struct dce_hwseq *hws, bool enable) { - bool force_on = 1; /* disable power gating */ + bool force_on = true; /* disable power gating */ if (enable) - force_on = 0; + force_on = false; /* DCHUBP0/1/2/3 */ REG_UPDATE(DOMAIN0_PG_CONFIG, DOMAIN0_POWER_FORCEON, force_on); @@ -460,7 +497,7 @@ static void enable_power_gating_plane( REG_UPDATE(DOMAIN7_PG_CONFIG, DOMAIN7_POWER_FORCEON, force_on); } -static void disable_vga( +void dcn10_disable_vga( struct dce_hwseq *hws) { unsigned int in_vga1_mode = 0; @@ -493,7 +530,7 @@ static void disable_vga( REG_UPDATE(VGA_TEST_CONTROL, VGA_TEST_RENDER_START, 1); } -static void dpp_pg_control( +void dcn10_dpp_pg_control( struct dce_hwseq *hws, unsigned int dpp_inst, bool power_on) @@ -545,7 +582,7 @@ static void dpp_pg_control( } } -static void hubp_pg_control( +void dcn10_hubp_pg_control( struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on) @@ -605,8 +642,8 @@ static void power_on_plane( if (REG(DC_IP_REQUEST_CNTL)) { REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); - dpp_pg_control(hws, plane_id, true); - hubp_pg_control(hws, plane_id, true); + hws->funcs.dpp_pg_control(hws, plane_id, true); + hws->funcs.hubp_pg_control(hws, plane_id, true); REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); DC_LOG_DEBUG( @@ -627,7 +664,7 @@ static void undo_DEGVIDCN10_253_wa(struct dc *dc) REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); - hubp_pg_control(hws, 0, false); + hws->funcs.hubp_pg_control(hws, 0, false); REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); @@ -656,7 +693,7 @@ static void apply_DEGVIDCN10_253_wa(struct dc *dc) REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); - hubp_pg_control(hws, 0, true); + hws->funcs.hubp_pg_control(hws, 0, true); REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); @@ -664,10 +701,27 @@ static void apply_DEGVIDCN10_253_wa(struct dc *dc) hws->wa_state.DEGVIDCN10_253_applied = true; } -static void bios_golden_init(struct dc *dc) +void dcn10_bios_golden_init(struct dc *dc) { + struct dce_hwseq *hws = dc->hwseq; struct dc_bios *bp = dc->ctx->dc_bios; int i; + bool allow_self_fresh_force_enable = true; + + if (hws->funcs.s0i3_golden_init_wa && hws->funcs.s0i3_golden_init_wa(dc)) + return; + + if (dc->res_pool->hubbub->funcs->is_allow_self_refresh_enabled) + allow_self_fresh_force_enable = + dc->res_pool->hubbub->funcs->is_allow_self_refresh_enabled(dc->res_pool->hubbub); + + + /* WA for making DF sleep when idle after resume from S0i3. + * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE is set to 1 by + * command table, if DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 0 + * before calling command table and it changed to 1 after, + * it should be set back to 0. + */ /* initialize dcn global */ bp->funcs->enable_disp_power_gating(bp, @@ -678,6 +732,12 @@ static void bios_golden_init(struct dc *dc) bp->funcs->enable_disp_power_gating(bp, CONTROLLER_ID_D0 + i, ASIC_PIPE_DISABLE); } + + if (dc->res_pool->hubbub->funcs->allow_self_refresh_control) + if (allow_self_fresh_force_enable == false && + dc->res_pool->hubbub->funcs->is_allow_self_refresh_enabled(dc->res_pool->hubbub)) + dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, true); + } static void false_optc_underflow_wa( @@ -702,13 +762,14 @@ static void false_optc_underflow_wa( dc->hwss.wait_for_mpcc_disconnect(dc, dc->res_pool, old_pipe_ctx); } - tg->funcs->set_blank_data_double_buffer(tg, true); + if (tg->funcs->set_blank_data_double_buffer) + tg->funcs->set_blank_data_double_buffer(tg, true); if (tg->funcs->is_optc_underflow_occurred(tg) && !underflow) tg->funcs->clear_optc_underflow(tg); } -static enum dc_status dcn10_enable_stream_timing( +enum dc_status dcn10_enable_stream_timing( struct pipe_ctx *pipe_ctx, struct dc_state *context, struct dc *dc) @@ -799,6 +860,7 @@ static void dcn10_reset_back_end_for_pipe( struct dc_state *context) { int i; + struct dc_link *link; DC_LOGGER_INIT(dc->ctx->logger); if (pipe_ctx->stream_res.stream_enc == NULL) { pipe_ctx->stream = NULL; @@ -806,13 +868,31 @@ static void dcn10_reset_back_end_for_pipe( } if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - /* DPMS may already disable */ - if (!pipe_ctx->stream->dpms_off) - core_link_disable_stream(pipe_ctx, FREE_ACQUIRED_RESOURCE); - else if (pipe_ctx->stream_res.audio) { - dc->hwss.disable_audio_stream(pipe_ctx, FREE_ACQUIRED_RESOURCE); + link = pipe_ctx->stream->link; + /* DPMS may already disable or */ + /* dpms_off status is incorrect due to fastboot + * feature. When system resume from S4 with second + * screen only, the dpms_off would be true but + * VBIOS lit up eDP, so check link status too. + */ + if (!pipe_ctx->stream->dpms_off || link->link_status.link_active) + core_link_disable_stream(pipe_ctx); + else if (pipe_ctx->stream_res.audio) + dc->hwss.disable_audio_stream(pipe_ctx); + + if (pipe_ctx->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx->stream_res.audio, false); + pipe_ctx->stream_res.audio = NULL; + } } - } /* by upper caller loop, parent pipe: pipe0, will be reset last. @@ -823,6 +903,9 @@ static void dcn10_reset_back_end_for_pipe( pipe_ctx->stream_res.tg->funcs->disable_crtc(pipe_ctx->stream_res.tg); pipe_ctx->stream_res.tg->funcs->enable_optc_clock(pipe_ctx->stream_res.tg, false); + if (pipe_ctx->stream_res.tg->funcs->set_drr) + pipe_ctx->stream_res.tg->funcs->set_drr( + pipe_ctx->stream_res.tg, NULL); } for (i = 0; i < dc->res_pool->pipe_count; i++) @@ -939,8 +1022,9 @@ void dcn10_verify_allow_pstate_change_high(struct dc *dc) } /* trigger HW to start disconnect plane from stream on the next vsync */ -void hwss1_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn10_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) { + struct dce_hwseq *hws = dc->hwseq; struct hubp *hubp = pipe_ctx->plane_res.hubp; int dpp_id = pipe_ctx->plane_res.dpp->inst; struct mpc *mpc = dc->res_pool->mpc; @@ -965,10 +1049,10 @@ void hwss1_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) hubp->funcs->hubp_disconnect(hubp); if (dc->debug.sanity_checks) - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); } -static void plane_atomic_power_down(struct dc *dc, +void dcn10_plane_atomic_power_down(struct dc *dc, struct dpp *dpp, struct hubp *hubp) { @@ -978,8 +1062,8 @@ static void plane_atomic_power_down(struct dc *dc, if (REG(DC_IP_REQUEST_CNTL)) { REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); - dpp_pg_control(hws, dpp->inst, false); - hubp_pg_control(hws, hubp->inst, false); + hws->funcs.dpp_pg_control(hws, dpp->inst, false); + hws->funcs.hubp_pg_control(hws, hubp->inst, false); dpp->funcs->dpp_reset(dpp); REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); @@ -991,8 +1075,9 @@ static void plane_atomic_power_down(struct dc *dc, /* disable HW used by plane. * note: cannot disable until disconnect is complete */ -static void plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn10_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) { + struct dce_hwseq *hws = dc->hwseq; struct hubp *hubp = pipe_ctx->plane_res.hubp; struct dpp *dpp = pipe_ctx->plane_res.dpp; int opp_id = hubp->opp_id; @@ -1011,7 +1096,7 @@ static void plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) hubp->power_gated = true; dc->optimized_required = false; /* We're powering off, no need to optimize */ - plane_atomic_power_down(dc, + hws->funcs.plane_atomic_power_down(dc, pipe_ctx->plane_res.dpp, pipe_ctx->plane_res.hubp); @@ -1023,14 +1108,15 @@ static void plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) pipe_ctx->plane_state = NULL; } -static void dcn10_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn10_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) { + struct dce_hwseq *hws = dc->hwseq; DC_LOGGER_INIT(dc->ctx->logger); if (!pipe_ctx->plane_res.hubp || pipe_ctx->plane_res.hubp->power_gated) return; - plane_atomic_disable(dc, pipe_ctx); + hws->funcs.plane_atomic_disable(dc, pipe_ctx); apply_DEGVIDCN10_253_wa(dc); @@ -1038,9 +1124,10 @@ static void dcn10_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) pipe_ctx->pipe_idx); } -static void dcn10_init_pipes(struct dc *dc, struct dc_state *context) +void dcn10_init_pipes(struct dc *dc, struct dc_state *context) { int i; + struct dce_hwseq *hws = dc->hwseq; bool can_apply_seamless_boot = false; for (i = 0; i < context->stream_count; i++) { @@ -1065,15 +1152,28 @@ static void dcn10_init_pipes(struct dc *dc, struct dc_state *context) * command table. */ if (tg->funcs->is_tg_enabled(tg)) { - tg->funcs->lock(tg); - tg->funcs->set_blank(tg, true); - hwss_wait_for_blank_complete(tg); + if (hws->funcs.init_blank != NULL) { + hws->funcs.init_blank(dc, tg); + tg->funcs->lock(tg); + } else { + tg->funcs->lock(tg); + tg->funcs->set_blank(tg, true); + hwss_wait_for_blank_complete(tg); + } } } - /* Cannot reset the MPC mux if seamless boot */ - if (!can_apply_seamless_boot) - dc->res_pool->mpc->funcs->mpc_init(dc->res_pool->mpc); + /* num_opp will be equal to number of mpcc */ + for (i = 0; i < dc->res_pool->res_cap->num_opp; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + /* Cannot reset the MPC mux if seamless boot */ + if (pipe_ctx->stream != NULL && can_apply_seamless_boot) + continue; + + dc->res_pool->mpc->funcs->mpc_init_single_inst( + dc->res_pool->mpc, i); + } for (i = 0; i < dc->res_pool->pipe_count; i++) { struct timing_generator *tg = dc->res_pool->timing_generators[i]; @@ -1088,8 +1188,14 @@ static void dcn10_init_pipes(struct dc *dc, struct dc_state *context) if (can_apply_seamless_boot && pipe_ctx->stream != NULL && pipe_ctx->stream_res.tg->funcs->is_tg_enabled( - pipe_ctx->stream_res.tg)) + pipe_ctx->stream_res.tg)) { + // Enable double buffering for OTG_BLANK no matter if + // seamless boot is enabled or not to suppress global sync + // signals when OTG blanked. This is to prevent pipe from + // requesting data while in PSR. + tg->funcs->tg_init(tg); continue; + } /* Disable on the current state so the new one isn't cleared. */ pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -1111,12 +1217,12 @@ static void dcn10_init_pipes(struct dc *dc, struct dc_state *context) dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; - hwss1_plane_atomic_disconnect(dc, pipe_ctx); + hws->funcs.plane_atomic_disconnect(dc, pipe_ctx); if (tg->funcs->is_tg_enabled(tg)) tg->funcs->unlock(tg); - dcn10_disable_plane(dc, pipe_ctx); + dc->hwss.disable_plane(dc, pipe_ctx); pipe_ctx->stream_res.tg = NULL; pipe_ctx->plane_res.hubp = NULL; @@ -1125,15 +1231,24 @@ static void dcn10_init_pipes(struct dc *dc, struct dc_state *context) } } -static void dcn10_init_hw(struct dc *dc) +void dcn10_init_hw(struct dc *dc) { int i; struct abm *abm = dc->res_pool->abm; struct dmcu *dmcu = dc->res_pool->dmcu; struct dce_hwseq *hws = dc->hwseq; struct dc_bios *dcb = dc->ctx->dc_bios; + struct resource_pool *res_pool = dc->res_pool; + + if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) + dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); + + // Initialize the dccg + if (dc->res_pool->dccg && dc->res_pool->dccg->funcs->dccg_init) + dc->res_pool->dccg->funcs->dccg_init(res_pool->dccg); if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + REG_WRITE(REFCLK_CNTL, 0); REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); REG_WRITE(DIO_MEM_PWR_CTRL, 0); @@ -1147,31 +1262,40 @@ static void dcn10_init_hw(struct dc *dc) REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); } - enable_power_gating_plane(dc->hwseq, true); + //Enable ability to power gate / don't force power on permanently + hws->funcs.enable_power_gating_plane(hws, true); - /* end of FPGA. Below if real ASIC */ return; } - if (!dcb->funcs->is_accelerated_mode(dcb)) { - bool allow_self_fresh_force_enable = - hububu1_is_allow_self_refresh_enabled( - dc->res_pool->hubbub); + if (!dcb->funcs->is_accelerated_mode(dcb)) + hws->funcs.disable_vga(dc->hwseq); - bios_golden_init(dc); + hws->funcs.bios_golden_init(dc); + if (dc->ctx->dc_bios->fw_info_valid) { + res_pool->ref_clocks.xtalin_clock_inKhz = + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; - /* WA for making DF sleep when idle after resume from S0i3. - * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE is set to 1 by - * command table, if DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 0 - * before calling command table and it changed to 1 after, - * it should be set back to 0. - */ - if (allow_self_fresh_force_enable == false && - hububu1_is_allow_self_refresh_enabled(dc->res_pool->hubbub)) - hubbub1_allow_self_refresh_control(dc->res_pool->hubbub, true); + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + if (res_pool->dccg && res_pool->hubbub) { - disable_vga(dc->hwseq); - } + (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, + &res_pool->ref_clocks.dccg_ref_clock_inKhz); + + (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, + res_pool->ref_clocks.dccg_ref_clock_inKhz, + &res_pool->ref_clocks.dchub_ref_clock_inKhz); + } else { + // Not all ASICs have DCCG sw component + res_pool->ref_clocks.dccg_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; + res_pool->ref_clocks.dchub_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; + } + } + } else + ASSERT_CRITICAL(false); for (i = 0; i < dc->link_count; i++) { /* Power up AND update implementation according to the @@ -1188,6 +1312,11 @@ static void dcn10_init_hw(struct dc *dc) link->link_status.link_active = true; } + /* Power gate DSCs */ + for (i = 0; i < res_pool->res_cap->num_dsc; i++) + if (hws->funcs.dsc_pg_control != NULL) + hws->funcs.dsc_pg_control(hws, res_pool->dscs[i]->inst, false); + /* If taking control over from VBIOS, we may want to optimize our first * mode set, so we need to skip powering down pipes until we know which * pipes we want to use. @@ -1195,11 +1324,11 @@ static void dcn10_init_hw(struct dc *dc) * everything down. */ if (dcb->funcs->is_accelerated_mode(dcb) || dc->config.power_down_display_on_boot) { - dc->hwss.init_pipes(dc, dc->current_state); + hws->funcs.init_pipes(dc, dc->current_state); } - for (i = 0; i < dc->res_pool->audio_count; i++) { - struct audio *audio = dc->res_pool->audios[i]; + for (i = 0; i < res_pool->audio_count; i++) { + struct audio *audio = res_pool->audios[i]; audio->funcs->hw_init(audio); } @@ -1209,7 +1338,7 @@ static void dcn10_init_hw(struct dc *dc) abm->funcs->abm_init(abm); } - if (dmcu != NULL) + if (dmcu != NULL && !dmcu->auto_load_dmcu) dmcu->funcs->dmcu_init(dmcu); if (abm != NULL && dmcu != NULL) @@ -1227,16 +1356,19 @@ static void dcn10_init_hw(struct dc *dc) REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); } - enable_power_gating_plane(dc->hwseq, true); + hws->funcs.enable_power_gating_plane(dc->hwseq, true); + + if (dc->clk_mgr->funcs->notify_wm_ranges) + dc->clk_mgr->funcs->notify_wm_ranges(dc->clk_mgr); - memset(&dc->clk_mgr->clks, 0, sizeof(dc->clk_mgr->clks)); } -static void dcn10_reset_hw_ctx_wrap( +void dcn10_reset_hw_ctx_wrap( struct dc *dc, struct dc_state *context) { int i; + struct dce_hwseq *hws = dc->hwseq; /* Reset Back End*/ for (i = dc->res_pool->pipe_count - 1; i >= 0 ; i--) { @@ -1255,8 +1387,8 @@ static void dcn10_reset_hw_ctx_wrap( struct clock_source *old_clk = pipe_ctx_old->clock_source; dcn10_reset_back_end_for_pipe(dc, pipe_ctx_old, dc->current_state); - if (dc->hwss.enable_stream_gating) - dc->hwss.enable_stream_gating(dc, pipe_ctx); + if (hws->funcs.enable_stream_gating) + hws->funcs.enable_stream_gating(dc, pipe_ctx); if (old_clk) old_clk->funcs->cs_power_down(old_clk); } @@ -1289,9 +1421,7 @@ static bool patch_address_for_sbs_tb_stereo( return false; } - - -static void dcn10_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn10_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_ctx) { bool addr_patched = false; PHYSICAL_ADDRESS_LOC addr; @@ -1316,8 +1446,8 @@ static void dcn10_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_c pipe_ctx->plane_state->address.grph_stereo.left_addr = addr; } -static bool dcn10_set_input_transfer_func(struct pipe_ctx *pipe_ctx, - const struct dc_plane_state *plane_state) +bool dcn10_set_input_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, + const struct dc_plane_state *plane_state) { struct dpp *dpp_base = pipe_ctx->plane_res.dpp; const struct dc_transfer_func *tf = NULL; @@ -1349,6 +1479,11 @@ static bool dcn10_set_input_transfer_func(struct pipe_ctx *pipe_ctx, dpp_base->funcs->dpp_set_degamma(dpp_base, IPP_DEGAMMA_MODE_BYPASS); break; case TRANSFER_FUNCTION_PQ: + dpp_base->funcs->dpp_set_degamma(dpp_base, IPP_DEGAMMA_MODE_USER_PWL); + cm_helper_translate_curve_to_degamma_hw_format(tf, &dpp_base->degamma_params); + dpp_base->funcs->dpp_program_degamma_pwl(dpp_base, &dpp_base->degamma_params); + result = true; + break; default: result = false; break; @@ -1366,9 +1501,36 @@ static bool dcn10_set_input_transfer_func(struct pipe_ctx *pipe_ctx, return result; } -static bool -dcn10_set_output_transfer_func(struct pipe_ctx *pipe_ctx, - const struct dc_stream_state *stream) +#define MAX_NUM_HW_POINTS 0x200 + +static void log_tf(struct dc_context *ctx, + struct dc_transfer_func *tf, uint32_t hw_points_num) +{ + // DC_LOG_GAMMA is default logging of all hw points + // DC_LOG_ALL_GAMMA logs all points, not only hw points + // DC_LOG_ALL_TF_POINTS logs all channels of the tf + int i = 0; + + DC_LOGGER_INIT(ctx->logger); + DC_LOG_GAMMA("Gamma Correction TF"); + DC_LOG_ALL_GAMMA("Logging all tf points..."); + DC_LOG_ALL_TF_CHANNELS("Logging all channels..."); + + for (i = 0; i < hw_points_num; i++) { + DC_LOG_GAMMA("R\t%d\t%llu", i, tf->tf_pts.red[i].value); + DC_LOG_ALL_TF_CHANNELS("G\t%d\t%llu", i, tf->tf_pts.green[i].value); + DC_LOG_ALL_TF_CHANNELS("B\t%d\t%llu", i, tf->tf_pts.blue[i].value); + } + + for (i = hw_points_num; i < MAX_NUM_HW_POINTS; i++) { + DC_LOG_ALL_GAMMA("R\t%d\t%llu", i, tf->tf_pts.red[i].value); + DC_LOG_ALL_TF_CHANNELS("G\t%d\t%llu", i, tf->tf_pts.green[i].value); + DC_LOG_ALL_TF_CHANNELS("B\t%d\t%llu", i, tf->tf_pts.blue[i].value); + } +} + +bool dcn10_set_output_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, + const struct dc_stream_state *stream) { struct dpp *dpp = pipe_ctx->plane_res.dpp; @@ -1394,14 +1556,23 @@ dcn10_set_output_transfer_func(struct pipe_ctx *pipe_ctx, } else dpp->funcs->dpp_program_regamma_pwl(dpp, NULL, OPP_REGAMMA_BYPASS); + if (stream != NULL && stream->ctx != NULL && + stream->out_transfer_func != NULL) { + log_tf(stream->ctx, + stream->out_transfer_func, + dpp->regamma_params.hw_points_num); + } + return true; } -static void dcn10_pipe_control_lock( +void dcn10_pipe_control_lock( struct dc *dc, struct pipe_ctx *pipe, bool lock) { + struct dce_hwseq *hws = dc->hwseq; + /* use TG master update lock to lock everything on the TG * therefore only top pipe need to lock */ @@ -1409,7 +1580,7 @@ static void dcn10_pipe_control_lock( return; if (dc->debug.sanity_checks) - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); if (lock) pipe->stream_res.tg->funcs->lock(pipe->stream_res.tg); @@ -1417,7 +1588,7 @@ static void dcn10_pipe_control_lock( pipe->stream_res.tg->funcs->unlock(pipe->stream_res.tg); if (dc->debug.sanity_checks) - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); } static bool wait_for_reset_trigger_to_occur( @@ -1457,7 +1628,7 @@ static bool wait_for_reset_trigger_to_occur( return rc; } -static void dcn10_enable_timing_synchronization( +void dcn10_enable_timing_synchronization( struct dc *dc, int group_index, int group_size, @@ -1487,7 +1658,7 @@ static void dcn10_enable_timing_synchronization( DC_SYNC_INFO("Sync complete\n"); } -static void dcn10_enable_per_frame_crtc_position_reset( +void dcn10_enable_per_frame_crtc_position_reset( struct dc *dc, int group_size, struct pipe_ctx *grouped_pipes[]) @@ -1512,10 +1683,10 @@ static void dcn10_enable_per_frame_crtc_position_reset( } /*static void print_rq_dlg_ttu( - struct dc *core_dc, + struct dc *dc, struct pipe_ctx *pipe_ctx) { - DC_LOG_BANDWIDTH_CALCS(core_dc->ctx->logger, + DC_LOG_BANDWIDTH_CALCS(dc->ctx->logger, "\n============== DML TTU Output parameters [%d] ==============\n" "qos_level_low_wm: %d, \n" "qos_level_high_wm: %d, \n" @@ -1545,7 +1716,7 @@ static void dcn10_enable_per_frame_crtc_position_reset( pipe_ctx->ttu_regs.refcyc_per_req_delivery_pre_c ); - DC_LOG_BANDWIDTH_CALCS(core_dc->ctx->logger, + DC_LOG_BANDWIDTH_CALCS(dc->ctx->logger, "\n============== DML DLG Output parameters [%d] ==============\n" "refcyc_h_blank_end: %d, \n" "dlg_vblank_end: %d, \n" @@ -1580,7 +1751,7 @@ static void dcn10_enable_per_frame_crtc_position_reset( pipe_ctx->dlg_regs.refcyc_per_pte_group_nom_l ); - DC_LOG_BANDWIDTH_CALCS(core_dc->ctx->logger, + DC_LOG_BANDWIDTH_CALCS(dc->ctx->logger, "\ndst_y_per_meta_row_nom_l: %d, \n" "refcyc_per_meta_chunk_nom_l: %d, \n" "refcyc_per_line_delivery_pre_l: %d, \n" @@ -1610,7 +1781,7 @@ static void dcn10_enable_per_frame_crtc_position_reset( pipe_ctx->dlg_regs.refcyc_per_line_delivery_c ); - DC_LOG_BANDWIDTH_CALCS(core_dc->ctx->logger, + DC_LOG_BANDWIDTH_CALCS(dc->ctx->logger, "\n============== DML RQ Output parameters [%d] ==============\n" "chunk_size: %d \n" "min_chunk_size: %d \n" @@ -1725,7 +1896,7 @@ static void dcn10_enable_plane( struct dce_hwseq *hws = dc->hwseq; if (dc->debug.sanity_checks) { - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); } undo_DEGVIDCN10_253_wa(dc); @@ -1782,11 +1953,11 @@ static void dcn10_enable_plane( dcn10_program_pte_vm(hws, pipe_ctx->plane_res.hubp); if (dc->debug.sanity_checks) { - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); } } -static void program_gamut_remap(struct pipe_ctx *pipe_ctx) +void dcn10_program_gamut_remap(struct pipe_ctx *pipe_ctx) { int i = 0; struct dpp_grph_csc_adjustment adjust; @@ -1804,69 +1975,65 @@ static void program_gamut_remap(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.dpp->funcs->dpp_set_gamut_remap(pipe_ctx->plane_res.dpp, &adjust); } -static void dcn10_program_output_csc(struct dc *dc, - struct pipe_ctx *pipe_ctx, - enum dc_color_space colorspace, - uint16_t *matrix, - int opp_id) -{ - if (pipe_ctx->stream->csc_color_matrix.enable_adjustment == true) { - if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment != NULL) - pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment(pipe_ctx->plane_res.dpp, matrix); - } else { - if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default != NULL) - pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default(pipe_ctx->plane_res.dpp, colorspace); - } -} -bool is_lower_pipe_tree_visible(struct pipe_ctx *pipe_ctx) +static bool dcn10_is_rear_mpo_fix_required(struct pipe_ctx *pipe_ctx, enum dc_color_space colorspace) { - if (pipe_ctx->plane_state && pipe_ctx->plane_state->visible) - return true; - if (pipe_ctx->bottom_pipe && is_lower_pipe_tree_visible(pipe_ctx->bottom_pipe)) - return true; - return false; -} + if (pipe_ctx->plane_state && pipe_ctx->plane_state->layer_index > 0 && is_rgb_cspace(colorspace)) { + if (pipe_ctx->top_pipe) { + struct pipe_ctx *top = pipe_ctx->top_pipe; -bool is_upper_pipe_tree_visible(struct pipe_ctx *pipe_ctx) -{ - if (pipe_ctx->plane_state && pipe_ctx->plane_state->visible) - return true; - if (pipe_ctx->top_pipe && is_upper_pipe_tree_visible(pipe_ctx->top_pipe)) - return true; + while (top->top_pipe) + top = top->top_pipe; // Traverse to top pipe_ctx + if (top->plane_state && top->plane_state->layer_index == 0) + return true; // Front MPO plane not hidden + } + } return false; } -bool is_pipe_tree_visible(struct pipe_ctx *pipe_ctx) +static void dcn10_set_csc_adjustment_rgb_mpo_fix(struct pipe_ctx *pipe_ctx, uint16_t *matrix) { - if (pipe_ctx->plane_state && pipe_ctx->plane_state->visible) - return true; - if (pipe_ctx->top_pipe && is_upper_pipe_tree_visible(pipe_ctx->top_pipe)) - return true; - if (pipe_ctx->bottom_pipe && is_lower_pipe_tree_visible(pipe_ctx->bottom_pipe)) - return true; - return false; + // Override rear plane RGB bias to fix MPO brightness + uint16_t rgb_bias = matrix[3]; + + matrix[3] = 0; + matrix[7] = 0; + matrix[11] = 0; + pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment(pipe_ctx->plane_res.dpp, matrix); + matrix[3] = rgb_bias; + matrix[7] = rgb_bias; + matrix[11] = rgb_bias; } -bool is_rgb_cspace(enum dc_color_space output_color_space) +void dcn10_program_output_csc(struct dc *dc, + struct pipe_ctx *pipe_ctx, + enum dc_color_space colorspace, + uint16_t *matrix, + int opp_id) { - switch (output_color_space) { - case COLOR_SPACE_SRGB: - case COLOR_SPACE_SRGB_LIMITED: - case COLOR_SPACE_2020_RGB_FULLRANGE: - case COLOR_SPACE_2020_RGB_LIMITEDRANGE: - case COLOR_SPACE_ADOBERGB: - return true; - case COLOR_SPACE_YCBCR601: - case COLOR_SPACE_YCBCR709: - case COLOR_SPACE_YCBCR601_LIMITED: - case COLOR_SPACE_YCBCR709_LIMITED: - case COLOR_SPACE_2020_YCBCR: - return false; - default: - /* Add a case to switch */ - BREAK_TO_DEBUGGER(); - return false; + if (pipe_ctx->stream->csc_color_matrix.enable_adjustment == true) { + if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment != NULL) { + + /* MPO is broken with RGB colorspaces when OCSC matrix + * brightness offset >= 0 on DCN1 due to OCSC before MPC + * Blending adds offsets from front + rear to rear plane + * + * Fix is to set RGB bias to 0 on rear plane, top plane + * black value pixels add offset instead of rear + front + */ + + int16_t rgb_bias = matrix[3]; + // matrix[3/7/11] are all the same offset value + + if (rgb_bias > 0 && dcn10_is_rear_mpo_fix_required(pipe_ctx, colorspace)) { + dcn10_set_csc_adjustment_rgb_mpo_fix(pipe_ctx, matrix); + } else { + pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment(pipe_ctx->plane_res.dpp, matrix); + } + } + } else { + if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default != NULL) + pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default(pipe_ctx->plane_res.dpp, colorspace); } } @@ -1943,70 +2110,7 @@ void dcn10_get_hdr_visual_confirm_color( } } -static uint16_t fixed_point_to_int_frac( - struct fixed31_32 arg, - uint8_t integer_bits, - uint8_t fractional_bits) -{ - int32_t numerator; - int32_t divisor = 1 << fractional_bits; - - uint16_t result; - - uint16_t d = (uint16_t)dc_fixpt_floor( - dc_fixpt_abs( - arg)); - - if (d <= (uint16_t)(1 << integer_bits) - (1 / (uint16_t)divisor)) - numerator = (uint16_t)dc_fixpt_floor( - dc_fixpt_mul_int( - arg, - divisor)); - else { - numerator = dc_fixpt_floor( - dc_fixpt_sub( - dc_fixpt_from_int( - 1LL << integer_bits), - dc_fixpt_recip( - dc_fixpt_from_int( - divisor)))); - } - - if (numerator >= 0) - result = (uint16_t)numerator; - else - result = (uint16_t)( - (1 << (integer_bits + fractional_bits + 1)) + numerator); - - if ((result != 0) && dc_fixpt_lt( - arg, dc_fixpt_zero)) - result |= 1 << (integer_bits + fractional_bits); - - return result; -} - -void dcn10_build_prescale_params(struct dc_bias_and_scale *bias_and_scale, - const struct dc_plane_state *plane_state) -{ - if (plane_state->format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN - && plane_state->format != SURFACE_PIXEL_FORMAT_INVALID - && plane_state->input_csc_color_matrix.enable_adjustment - && plane_state->coeff_reduction_factor.value != 0) { - bias_and_scale->scale_blue = fixed_point_to_int_frac( - dc_fixpt_mul(plane_state->coeff_reduction_factor, - dc_fixpt_from_fraction(256, 255)), - 2, - 13); - bias_and_scale->scale_red = bias_and_scale->scale_blue; - bias_and_scale->scale_green = bias_and_scale->scale_blue; - } else { - bias_and_scale->scale_blue = 0x2000; - bias_and_scale->scale_red = 0x2000; - bias_and_scale->scale_green = 0x2000; - } -} - -static void update_dpp(struct dpp *dpp, struct dc_plane_state *plane_state) +static void dcn10_update_dpp(struct dpp *dpp, struct dc_plane_state *plane_state) { struct dc_bias_and_scale bns_params = {0}; @@ -2015,21 +2119,18 @@ static void update_dpp(struct dpp *dpp, struct dc_plane_state *plane_state) plane_state->format, EXPANSION_MODE_ZERO, plane_state->input_csc_color_matrix, -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 plane_state->color_space, NULL); -#else - plane_state->color_space); -#endif //set scale and bias registers - dcn10_build_prescale_params(&bns_params, plane_state); + build_prescale_params(&bns_params, plane_state); if (dpp->funcs->dpp_program_bias_and_scale) dpp->funcs->dpp_program_bias_and_scale(dpp, &bns_params); } -static void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) { + struct dce_hwseq *hws = dc->hwseq; struct hubp *hubp = pipe_ctx->plane_res.hubp; struct mpcc_blnd_cfg blnd_cfg = {{0}}; bool per_pixel_alpha = pipe_ctx->plane_state->per_pixel_alpha && pipe_ctx->bottom_pipe; @@ -2039,10 +2140,10 @@ static void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) struct mpc_tree *mpc_tree_params = &(pipe_ctx->stream_res.opp->mpc_tree_params); if (dc->debug.visual_confirm == VISUAL_CONFIRM_HDR) { - dcn10_get_hdr_visual_confirm_color( + hws->funcs.get_hdr_visual_confirm_color( pipe_ctx, &blnd_cfg.black_color); } else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SURFACE) { - dcn10_get_surface_visual_confirm_color( + hws->funcs.get_surface_visual_confirm_color( pipe_ctx, &blnd_cfg.black_color); } else { color_space_to_black_color( @@ -2124,15 +2225,16 @@ static void update_scaler(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.dpp, &pipe_ctx->plane_res.scl_data); } -void update_dchubp_dpp( +static void dcn10_update_dchubp_dpp( struct dc *dc, struct pipe_ctx *pipe_ctx, struct dc_state *context) { + struct dce_hwseq *hws = dc->hwseq; struct hubp *hubp = pipe_ctx->plane_res.hubp; struct dpp *dpp = pipe_ctx->plane_res.dpp; struct dc_plane_state *plane_state = pipe_ctx->plane_state; - union plane_size size = plane_state->plane_size; + struct plane_size size = plane_state->plane_size; unsigned int compat_level = 0; /* depends on DML calculation, DPP clock value may change dynamically */ @@ -2178,16 +2280,16 @@ void update_dchubp_dpp( &pipe_ctx->ttu_regs); } - size.grph.surface_size = pipe_ctx->plane_res.scl_data.viewport; + size.surface_size = pipe_ctx->plane_res.scl_data.viewport; if (plane_state->update_flags.bits.full_update || plane_state->update_flags.bits.bpp_change) - update_dpp(dpp, plane_state); + dcn10_update_dpp(dpp, plane_state); if (plane_state->update_flags.bits.full_update || plane_state->update_flags.bits.per_pixel_alpha_change || plane_state->update_flags.bits.global_alpha_change) - dc->hwss.update_mpcc(dc, pipe_ctx); + hws->funcs.update_mpcc(dc, pipe_ctx); if (plane_state->update_flags.bits.full_update || plane_state->update_flags.bits.per_pixel_alpha_change || @@ -2216,7 +2318,7 @@ void update_dchubp_dpp( if (plane_state->update_flags.bits.full_update) { /*gamut remap*/ - program_gamut_remap(pipe_ctx); + dc->hwss.program_gamut_remap(pipe_ctx); dc->hwss.program_output_csc(dc, pipe_ctx, @@ -2247,13 +2349,13 @@ void update_dchubp_dpp( hubp->power_gated = false; - dc->hwss.update_plane_addr(dc, pipe_ctx); + hws->funcs.update_plane_addr(dc, pipe_ctx); if (is_pipe_tree_visible(pipe_ctx)) hubp->funcs->set_blank(hubp, false); } -static void dcn10_blank_pixel_data( +void dcn10_blank_pixel_data( struct dc *dc, struct pipe_ctx *pipe_ctx, bool blank) @@ -2296,10 +2398,9 @@ static void dcn10_blank_pixel_data( } } -void set_hdr_multiplier(struct pipe_ctx *pipe_ctx) +void dcn10_set_hdr_multiplier(struct pipe_ctx *pipe_ctx) { - struct fixed31_32 multiplier = dc_fixpt_from_fraction( - pipe_ctx->plane_state->sdr_white_level, 80); + struct fixed31_32 multiplier = pipe_ctx->plane_state->hdr_mult; uint32_t hw_mult = 0x1f000; // 1.0 default multiplier struct custom_float_format fmt; @@ -2307,7 +2408,8 @@ void set_hdr_multiplier(struct pipe_ctx *pipe_ctx) fmt.mantissa_bits = 12; fmt.sign = true; - if (pipe_ctx->plane_state->sdr_white_level > 80) + + if (!dc_fixpt_eq(multiplier, dc_fixpt_from_int(0))) // check != 0 convert_to_custom_float_format(multiplier, &fmt, &hw_mult); pipe_ctx->plane_res.dpp->funcs->dpp_set_hdr_multiplier( @@ -2319,17 +2421,19 @@ void dcn10_program_pipe( struct pipe_ctx *pipe_ctx, struct dc_state *context) { + struct dce_hwseq *hws = dc->hwseq; + if (pipe_ctx->plane_state->update_flags.bits.full_update) dcn10_enable_plane(dc, pipe_ctx, context); - update_dchubp_dpp(dc, pipe_ctx, context); + dcn10_update_dchubp_dpp(dc, pipe_ctx, context); - set_hdr_multiplier(pipe_ctx); + hws->funcs.set_hdr_multiplier(pipe_ctx); if (pipe_ctx->plane_state->update_flags.bits.full_update || pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change || pipe_ctx->plane_state->update_flags.bits.gamma_change) - dc->hwss.set_input_transfer_func(pipe_ctx, pipe_ctx->plane_state); + hws->funcs.set_input_transfer_func(dc, pipe_ctx, pipe_ctx->plane_state); /* dcn10_translate_regamma_to_hw_format takes 750us to finish * only do gamma programming for full update. @@ -2338,14 +2442,16 @@ void dcn10_program_pipe( * doing heavy calculation and programming */ if (pipe_ctx->plane_state->update_flags.bits.full_update) - dc->hwss.set_output_transfer_func(pipe_ctx, pipe_ctx->stream); + hws->funcs.set_output_transfer_func(dc, pipe_ctx, pipe_ctx->stream); } -static void program_all_pipe_in_tree( +static void dcn10_program_all_pipe_in_tree( struct dc *dc, struct pipe_ctx *pipe_ctx, struct dc_state *context) { + struct dce_hwseq *hws = dc->hwseq; + if (pipe_ctx->top_pipe == NULL) { bool blank = !is_pipe_tree_visible(pipe_ctx); @@ -2359,18 +2465,20 @@ static void program_all_pipe_in_tree( pipe_ctx->stream_res.tg->funcs->set_vtg_params( pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); - dc->hwss.blank_pixel_data(dc, pipe_ctx, blank); + if (hws->funcs.setup_vupdate_interrupt) + hws->funcs.setup_vupdate_interrupt(dc, pipe_ctx); + hws->funcs.blank_pixel_data(dc, pipe_ctx, blank); } if (pipe_ctx->plane_state != NULL) - dcn10_program_pipe(dc, pipe_ctx, context); + hws->funcs.program_pipe(dc, pipe_ctx, context); if (pipe_ctx->bottom_pipe != NULL && pipe_ctx->bottom_pipe != pipe_ctx) - program_all_pipe_in_tree(dc, pipe_ctx->bottom_pipe, context); + dcn10_program_all_pipe_in_tree(dc, pipe_ctx->bottom_pipe, context); } -struct pipe_ctx *find_top_pipe_for_stream( +static struct pipe_ctx *dcn10_find_top_pipe_for_stream( struct dc *dc, struct dc_state *context, const struct dc_stream_state *stream) @@ -2388,25 +2496,26 @@ struct pipe_ctx *find_top_pipe_for_stream( if (pipe_ctx->stream != stream) continue; - if (!pipe_ctx->top_pipe) + if (!pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe) return pipe_ctx; } return NULL; } -static void dcn10_apply_ctx_for_surface( +void dcn10_apply_ctx_for_surface( struct dc *dc, const struct dc_stream_state *stream, int num_planes, struct dc_state *context) { + struct dce_hwseq *hws = dc->hwseq; int i; struct timing_generator *tg; uint32_t underflow_check_delay_us; bool removed_pipe[4] = { false }; bool interdependent_update = false; struct pipe_ctx *top_pipe_to_program = - find_top_pipe_for_stream(dc, context, stream); + dcn10_find_top_pipe_for_stream(dc, context, stream); DC_LOGGER_INIT(dc->ctx->logger); if (!top_pipe_to_program) @@ -2419,23 +2528,23 @@ static void dcn10_apply_ctx_for_surface( underflow_check_delay_us = dc->debug.underflow_assert_delay_us; - if (underflow_check_delay_us != 0xFFFFFFFF && dc->hwss.did_underflow_occur) - ASSERT(dc->hwss.did_underflow_occur(dc, top_pipe_to_program)); + if (underflow_check_delay_us != 0xFFFFFFFF && hws->funcs.did_underflow_occur) + ASSERT(hws->funcs.did_underflow_occur(dc, top_pipe_to_program)); if (interdependent_update) - lock_all_pipes(dc, context, true); + dcn10_lock_all_pipes(dc, context, true); else dcn10_pipe_control_lock(dc, top_pipe_to_program, true); if (underflow_check_delay_us != 0xFFFFFFFF) udelay(underflow_check_delay_us); - if (underflow_check_delay_us != 0xFFFFFFFF && dc->hwss.did_underflow_occur) - ASSERT(dc->hwss.did_underflow_occur(dc, top_pipe_to_program)); + if (underflow_check_delay_us != 0xFFFFFFFF && hws->funcs.did_underflow_occur) + ASSERT(hws->funcs.did_underflow_occur(dc, top_pipe_to_program)); if (num_planes == 0) { /* OTG blank before remove all front end */ - dc->hwss.blank_pixel_data(dc, top_pipe_to_program, true); + hws->funcs.blank_pixel_data(dc, top_pipe_to_program, true); } /* Disconnect unused mpcc */ @@ -2453,7 +2562,7 @@ static void dcn10_apply_ctx_for_surface( if (old_pipe_ctx->stream_res.tg == tg && old_pipe_ctx->plane_res.hubp && old_pipe_ctx->plane_res.hubp->opp_id != OPP_ID_INVALID) - dcn10_disable_plane(dc, old_pipe_ctx); + dc->hwss.disable_plane(dc, old_pipe_ctx); } if ((!pipe_ctx->plane_state || @@ -2461,7 +2570,7 @@ static void dcn10_apply_ctx_for_surface( old_pipe_ctx->plane_state && old_pipe_ctx->stream_res.tg == tg) { - dc->hwss.plane_atomic_disconnect(dc, old_pipe_ctx); + hws->funcs.plane_atomic_disconnect(dc, old_pipe_ctx); removed_pipe[i] = true; DC_LOG_DC("Reset mpcc for pipe %d\n", @@ -2470,13 +2579,11 @@ static void dcn10_apply_ctx_for_surface( } if (num_planes > 0) - program_all_pipe_in_tree(dc, top_pipe_to_program, context); + dcn10_program_all_pipe_in_tree(dc, top_pipe_to_program, context); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* Program secondary blending tree and writeback pipes */ - if ((stream->num_wb_info > 0) && (dc->hwss.program_all_writeback_pipes_in_tree)) - dc->hwss.program_all_writeback_pipes_in_tree(dc, stream, context); -#endif + if ((stream->num_wb_info > 0) && (hws->funcs.program_all_writeback_pipes_in_tree)) + hws->funcs.program_all_writeback_pipes_in_tree(dc, stream, context); if (interdependent_update) for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; @@ -2492,7 +2599,7 @@ static void dcn10_apply_ctx_for_surface( } if (interdependent_update) - lock_all_pipes(dc, context, false); + dcn10_lock_all_pipes(dc, context, false); else dcn10_pipe_control_lock(dc, top_pipe_to_program, false); @@ -2501,7 +2608,7 @@ static void dcn10_apply_ctx_for_surface( for (i = 0; i < dc->res_pool->pipe_count; i++) if (removed_pipe[i]) - dcn10_disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); + dc->hwss.disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); for (i = 0; i < dc->res_pool->pipe_count; i++) if (removed_pipe[i]) { @@ -2529,14 +2636,15 @@ static void dcn10_stereo_hw_frame_pack_wa(struct dc *dc, struct dc_state *contex } } -static void dcn10_prepare_bandwidth( +void dcn10_prepare_bandwidth( struct dc *dc, struct dc_state *context) { + struct dce_hwseq *hws = dc->hwseq; struct hubbub *hubbub = dc->res_pool->hubbub; if (dc->debug.sanity_checks) - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { if (context->stream_count == 0) @@ -2558,17 +2666,18 @@ static void dcn10_prepare_bandwidth( dcn_bw_notify_pplib_of_wm_ranges(dc); if (dc->debug.sanity_checks) - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); } -static void dcn10_optimize_bandwidth( +void dcn10_optimize_bandwidth( struct dc *dc, struct dc_state *context) { + struct dce_hwseq *hws = dc->hwseq; struct hubbub *hubbub = dc->res_pool->hubbub; if (dc->debug.sanity_checks) - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { if (context->stream_count == 0) @@ -2590,19 +2699,24 @@ static void dcn10_optimize_bandwidth( dcn_bw_notify_pplib_of_wm_ranges(dc); if (dc->debug.sanity_checks) - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); } -static void set_drr(struct pipe_ctx **pipe_ctx, - int num_pipes, int vmin, int vmax) +void dcn10_set_drr(struct pipe_ctx **pipe_ctx, + int num_pipes, unsigned int vmin, unsigned int vmax, + unsigned int vmid, unsigned int vmid_frame_number) { int i = 0; struct drr_params params = {0}; // DRR set trigger event mapped to OTG_TRIG_A (bit 11) for manual control flow unsigned int event_triggers = 0x800; + // Note DRR trigger events are generated regardless of whether num frames met. + unsigned int num_frames = 2; params.vertical_total_max = vmax; params.vertical_total_min = vmin; + params.vertical_total_mid = vmid; + params.vertical_total_mid_frame_num = vmid_frame_number; /* TODO: If multiple pipes are to be supported, you need * some GSL stuff. Static screen triggers may be programmed differently @@ -2614,11 +2728,11 @@ static void set_drr(struct pipe_ctx **pipe_ctx, if (vmax != 0 && vmin != 0) pipe_ctx[i]->stream_res.tg->funcs->set_static_screen_control( pipe_ctx[i]->stream_res.tg, - event_triggers); + event_triggers, num_frames); } } -static void get_position(struct pipe_ctx **pipe_ctx, +void dcn10_get_position(struct pipe_ctx **pipe_ctx, int num_pipes, struct crtc_position *position) { @@ -2630,22 +2744,23 @@ static void get_position(struct pipe_ctx **pipe_ctx, pipe_ctx[i]->stream_res.tg->funcs->get_position(pipe_ctx[i]->stream_res.tg, position); } -static void set_static_screen_control(struct pipe_ctx **pipe_ctx, - int num_pipes, const struct dc_static_screen_events *events) +void dcn10_set_static_screen_control(struct pipe_ctx **pipe_ctx, + int num_pipes, const struct dc_static_screen_params *params) { unsigned int i; - unsigned int value = 0; + unsigned int triggers = 0; - if (events->surface_update) - value |= 0x80; - if (events->cursor_update) - value |= 0x2; - if (events->force_trigger) - value |= 0x1; + if (params->triggers.surface_update) + triggers |= 0x80; + if (params->triggers.cursor_update) + triggers |= 0x2; + if (params->triggers.force_trigger) + triggers |= 0x1; for (i = 0; i < num_pipes; i++) pipe_ctx[i]->stream_res.tg->funcs-> - set_static_screen_control(pipe_ctx[i]->stream_res.tg, value); + set_static_screen_control(pipe_ctx[i]->stream_res.tg, + triggers, params->num_frames); } static void dcn10_config_stereo_parameters( @@ -2685,13 +2800,20 @@ static void dcn10_config_stereo_parameters( return; } -static void dcn10_setup_stereo(struct pipe_ctx *pipe_ctx, struct dc *dc) +void dcn10_setup_stereo(struct pipe_ctx *pipe_ctx, struct dc *dc) { struct crtc_stereo_flags flags = { 0 }; struct dc_stream_state *stream = pipe_ctx->stream; dcn10_config_stereo_parameters(stream, &flags); + if (stream->timing.timing_3d_format == TIMING_3D_FORMAT_SIDEBAND_FA) { + if (!dc_set_generic_gpio_for_stereo(true, dc->ctx->gpio_service)) + dc_set_generic_gpio_for_stereo(false, dc->ctx->gpio_service); + } else { + dc_set_generic_gpio_for_stereo(false, dc->ctx->gpio_service); + } + pipe_ctx->stream_res.opp->funcs->opp_program_stereo( pipe_ctx->stream_res.opp, flags.PROGRAM_STEREO == 1 ? true:false, @@ -2717,15 +2839,16 @@ static struct hubp *get_hubp_by_inst(struct resource_pool *res_pool, int mpcc_in return NULL; } -static void dcn10_wait_for_mpcc_disconnect( +void dcn10_wait_for_mpcc_disconnect( struct dc *dc, struct resource_pool *res_pool, struct pipe_ctx *pipe_ctx) { + struct dce_hwseq *hws = dc->hwseq; int mpcc_inst; if (dc->debug.sanity_checks) { - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); } if (!pipe_ctx->stream_res.opp) @@ -2742,12 +2865,12 @@ static void dcn10_wait_for_mpcc_disconnect( } if (dc->debug.sanity_checks) { - dcn10_verify_allow_pstate_change_high(dc); + hws->funcs.verify_allow_pstate_change_high(dc); } } -static bool dcn10_dummy_display_power_gating( +bool dcn10_dummy_display_power_gating( struct dc *dc, uint8_t controller_id, struct dc_bios *dcb, @@ -2756,7 +2879,7 @@ static bool dcn10_dummy_display_power_gating( return true; } -static void dcn10_update_pending_status(struct pipe_ctx *pipe_ctx) +void dcn10_update_pending_status(struct pipe_ctx *pipe_ctx) { struct dc_plane_state *plane_state = pipe_ctx->plane_state; struct timing_generator *tg = pipe_ctx->stream_res.tg; @@ -2780,19 +2903,42 @@ static void dcn10_update_pending_status(struct pipe_ctx *pipe_ctx) } } -static void dcn10_update_dchub(struct dce_hwseq *hws, struct dchub_init_data *dh_data) +void dcn10_update_dchub(struct dce_hwseq *hws, struct dchub_init_data *dh_data) { - if (hws->ctx->dc->res_pool->hubbub != NULL) { - struct hubp *hubp = hws->ctx->dc->res_pool->hubps[0]; + struct hubbub *hubbub = hws->ctx->dc->res_pool->hubbub; - if (hubp->funcs->hubp_update_dchub) - hubp->funcs->hubp_update_dchub(hubp, dh_data); - else - hubbub1_update_dchub(hws->ctx->dc->res_pool->hubbub, dh_data); + /* In DCN, this programming sequence is owned by the hubbub */ + hubbub->funcs->update_dchub(hubbub, dh_data); +} + +static bool dcn10_can_pipe_disable_cursor(struct pipe_ctx *pipe_ctx) +{ + struct pipe_ctx *test_pipe; + const struct rect *r1 = &pipe_ctx->plane_res.scl_data.recout, *r2; + int r1_r = r1->x + r1->width, r1_b = r1->y + r1->height, r2_r, r2_b; + + /** + * Disable the cursor if there's another pipe above this with a + * plane that contains this pipe's viewport to prevent double cursor + * and incorrect scaling artifacts. + */ + for (test_pipe = pipe_ctx->top_pipe; test_pipe; + test_pipe = test_pipe->top_pipe) { + if (!test_pipe->plane_state->visible) + continue; + + r2 = &test_pipe->plane_res.scl_data.recout; + r2_r = r2->x + r2->width; + r2_b = r2->y + r2->height; + + if (r1->x >= r2->x && r1->y >= r2->y && r1_r <= r2_r && r1_b <= r2_b) + return true; } + + return false; } -static void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) +void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) { struct dc_cursor_position pos_cpy = pipe_ctx->stream->cursor_position; struct hubp *hubp = pipe_ctx->plane_res.hubp; @@ -2806,35 +2952,108 @@ static void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx) .rotation = pipe_ctx->plane_state->rotation, .mirror = pipe_ctx->plane_state->horizontal_mirror }; - uint32_t x_plane = pipe_ctx->plane_state->dst_rect.x; - uint32_t y_plane = pipe_ctx->plane_state->dst_rect.y; - uint32_t x_offset = min(x_plane, pos_cpy.x); - uint32_t y_offset = min(y_plane, pos_cpy.y); + bool pipe_split_on = (pipe_ctx->top_pipe != NULL) || + (pipe_ctx->bottom_pipe != NULL); - pos_cpy.x -= x_offset; - pos_cpy.y -= y_offset; - pos_cpy.x_hotspot += (x_plane - x_offset); - pos_cpy.y_hotspot += (y_plane - y_offset); + int x_plane = pipe_ctx->plane_state->dst_rect.x; + int y_plane = pipe_ctx->plane_state->dst_rect.y; + int x_pos = pos_cpy.x; + int y_pos = pos_cpy.y; + + // translate cursor from stream space to plane space + x_pos = (x_pos - x_plane) * pipe_ctx->plane_state->src_rect.width / + pipe_ctx->plane_state->dst_rect.width; + y_pos = (y_pos - y_plane) * pipe_ctx->plane_state->src_rect.height / + pipe_ctx->plane_state->dst_rect.height; + + if (x_pos < 0) { + pos_cpy.x_hotspot -= x_pos; + x_pos = 0; + } + + if (y_pos < 0) { + pos_cpy.y_hotspot -= y_pos; + y_pos = 0; + } + + pos_cpy.x = (uint32_t)x_pos; + pos_cpy.y = (uint32_t)y_pos; if (pipe_ctx->plane_state->address.type == PLN_ADDR_TYPE_VIDEO_PROGRESSIVE) pos_cpy.enable = false; + if (pos_cpy.enable && dcn10_can_pipe_disable_cursor(pipe_ctx)) + pos_cpy.enable = false; + + // Swap axis and mirror horizontally + if (param.rotation == ROTATION_ANGLE_90) { + uint32_t temp_x = pos_cpy.x; + + pos_cpy.x = pipe_ctx->plane_res.scl_data.viewport.width - + (pos_cpy.y - pipe_ctx->plane_res.scl_data.viewport.x) + pipe_ctx->plane_res.scl_data.viewport.x; + pos_cpy.y = temp_x; + } + // Swap axis and mirror vertically + else if (param.rotation == ROTATION_ANGLE_270) { + uint32_t temp_y = pos_cpy.y; + int viewport_height = + pipe_ctx->plane_res.scl_data.viewport.height; + + if (pipe_split_on) { + if (pos_cpy.x > viewport_height) { + pos_cpy.x = pos_cpy.x - viewport_height; + pos_cpy.y = viewport_height - pos_cpy.x; + } else { + pos_cpy.y = 2 * viewport_height - pos_cpy.x; + } + } else + pos_cpy.y = viewport_height - pos_cpy.x; + pos_cpy.x = temp_y; + } + // Mirror horizontally and vertically + else if (param.rotation == ROTATION_ANGLE_180) { + int viewport_width = + pipe_ctx->plane_res.scl_data.viewport.width; + int viewport_x = + pipe_ctx->plane_res.scl_data.viewport.x; + + if (pipe_split_on) { + if (pos_cpy.x >= viewport_width + viewport_x) { + pos_cpy.x = 2 * viewport_width + - pos_cpy.x + 2 * viewport_x; + } else { + uint32_t temp_x = pos_cpy.x; + + pos_cpy.x = 2 * viewport_x - pos_cpy.x; + if (temp_x >= viewport_x + + (int)hubp->curs_attr.width || pos_cpy.x + <= (int)hubp->curs_attr.width + + pipe_ctx->plane_state->src_rect.x) { + pos_cpy.x = temp_x + viewport_width; + } + } + } else { + pos_cpy.x = viewport_width - pos_cpy.x + 2 * viewport_x; + } + pos_cpy.y = pipe_ctx->plane_res.scl_data.viewport.height - pos_cpy.y; + } + hubp->funcs->set_cursor_position(hubp, &pos_cpy, ¶m); dpp->funcs->set_cursor_position(dpp, &pos_cpy, ¶m, hubp->curs_attr.width, hubp->curs_attr.height); } -static void dcn10_set_cursor_attribute(struct pipe_ctx *pipe_ctx) +void dcn10_set_cursor_attribute(struct pipe_ctx *pipe_ctx) { struct dc_cursor_attributes *attributes = &pipe_ctx->stream->cursor_attributes; pipe_ctx->plane_res.hubp->funcs->set_cursor_attributes( pipe_ctx->plane_res.hubp, attributes); pipe_ctx->plane_res.dpp->funcs->set_cursor_attributes( - pipe_ctx->plane_res.dpp, attributes->color_format); + pipe_ctx->plane_res.dpp, attributes); } -static void dcn10_set_cursor_sdr_white_level(struct pipe_ctx *pipe_ctx) +void dcn10_set_cursor_sdr_white_level(struct pipe_ctx *pipe_ctx) { uint32_t sdr_white_level = pipe_ctx->stream->cursor_attributes.sdr_white_level; struct fixed31_32 multiplier; @@ -2861,12 +3080,12 @@ static void dcn10_set_cursor_sdr_white_level(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.dpp, &opt_attr); } -/** -* apply_front_porch_workaround TODO FPGA still need? -* -* This is a workaround for a bug that has existed since R5xx and has not been -* fixed keep Front porch at minimum 2 for Interlaced mode or 1 for progressive. -*/ +/* + * apply_front_porch_workaround TODO FPGA still need? + * + * This is a workaround for a bug that has existed since R5xx and has not been + * fixed keep Front porch at minimum 2 for Interlaced mode or 1 for progressive. + */ static void apply_front_porch_workaround( struct dc_crtc_timing *timing) { @@ -2879,7 +3098,7 @@ static void apply_front_porch_workaround( } } -int get_vupdate_offset_from_vsync(struct pipe_ctx *pipe_ctx) +int dcn10_get_vupdate_offset_from_vsync(struct pipe_ctx *pipe_ctx) { const struct dc_crtc_timing *dc_crtc_timing = &pipe_ctx->stream->timing; struct dc_crtc_timing patched_crtc_timing; @@ -2908,34 +3127,8 @@ int get_vupdate_offset_from_vsync(struct pipe_ctx *pipe_ctx) return vertical_line_start; } -void lock_all_pipes(struct dc *dc, - struct dc_state *context, - bool lock) -{ - struct pipe_ctx *pipe_ctx; - struct timing_generator *tg; - int i; - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - pipe_ctx = &context->res_ctx.pipe_ctx[i]; - tg = pipe_ctx->stream_res.tg; - /* - * Only lock the top pipe's tg to prevent redundant - * (un)locking. Also skip if pipe is disabled. - */ - if (pipe_ctx->top_pipe || - !pipe_ctx->stream || !pipe_ctx->plane_state || - !tg->funcs->is_tg_enabled(tg)) - continue; - - if (lock) - tg->funcs->lock(tg); - else - tg->funcs->unlock(tg); - } -} - -static void calc_vupdate_position( +static void dcn10_calc_vupdate_position( + struct dc *dc, struct pipe_ctx *pipe_ctx, uint32_t *start_line, uint32_t *end_line) @@ -2943,7 +3136,7 @@ static void calc_vupdate_position( const struct dc_crtc_timing *dc_crtc_timing = &pipe_ctx->stream->timing; int vline_int_offset_from_vupdate = pipe_ctx->stream->periodic_interrupt0.lines_offset; - int vupdate_offset_from_vsync = get_vupdate_offset_from_vsync(pipe_ctx); + int vupdate_offset_from_vsync = dc->hwss.get_vupdate_offset_from_vsync(pipe_ctx); int start_position; if (vline_int_offset_from_vupdate > 0) @@ -2964,7 +3157,8 @@ static void calc_vupdate_position( *end_line = 2; } -static void cal_vline_position( +static void dcn10_cal_vline_position( + struct dc *dc, struct pipe_ctx *pipe_ctx, enum vline_select vline, uint32_t *start_line, @@ -2979,7 +3173,8 @@ static void cal_vline_position( switch (ref_point) { case START_V_UPDATE: - calc_vupdate_position( + dcn10_calc_vupdate_position( + dc, pipe_ctx, start_line, end_line); @@ -2993,7 +3188,8 @@ static void cal_vline_position( } } -static void dcn10_setup_periodic_interrupt( +void dcn10_setup_periodic_interrupt( + struct dc *dc, struct pipe_ctx *pipe_ctx, enum vline_select vline) { @@ -3003,7 +3199,7 @@ static void dcn10_setup_periodic_interrupt( uint32_t start_line = 0; uint32_t end_line = 0; - cal_vline_position(pipe_ctx, vline, &start_line, &end_line); + dcn10_cal_vline_position(dc, pipe_ctx, vline, &start_line, &end_line); tg->funcs->setup_vertical_interrupt0(tg, start_line, end_line); @@ -3014,10 +3210,10 @@ static void dcn10_setup_periodic_interrupt( } } -static void dcn10_setup_vupdate_interrupt(struct pipe_ctx *pipe_ctx) +void dcn10_setup_vupdate_interrupt(struct dc *dc, struct pipe_ctx *pipe_ctx) { struct timing_generator *tg = pipe_ctx->stream_res.tg; - int start_line = get_vupdate_offset_from_vsync(pipe_ctx); + int start_line = dc->hwss.get_vupdate_offset_from_vsync(pipe_ctx); if (start_line < 0) { ASSERT(0); @@ -3028,12 +3224,13 @@ static void dcn10_setup_vupdate_interrupt(struct pipe_ctx *pipe_ctx) tg->funcs->setup_vertical_interrupt2(tg, start_line); } -static void dcn10_unblank_stream(struct pipe_ctx *pipe_ctx, +void dcn10_unblank_stream(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings) { struct encoder_unblank_param params = { { 0 } }; struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; + struct dce_hwseq *hws = link->dc->hwseq; /* only 3 items below are used by unblank */ params.timing = pipe_ctx->stream->timing; @@ -3047,11 +3244,11 @@ static void dcn10_unblank_stream(struct pipe_ctx *pipe_ctx, } if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) { - link->dc->hwss.edp_backlight_control(link, true); + hws->funcs.edp_backlight_control(link, true); } } -static void dcn10_send_immediate_sdp_message(struct pipe_ctx *pipe_ctx, +void dcn10_send_immediate_sdp_message(struct pipe_ctx *pipe_ctx, const uint8_t *custom_sdp_message, unsigned int sdp_message_size) { @@ -3062,66 +3259,53 @@ static void dcn10_send_immediate_sdp_message(struct pipe_ctx *pipe_ctx, sdp_message_size); } } +enum dc_status dcn10_set_clock(struct dc *dc, + enum dc_clock_type clock_type, + uint32_t clk_khz, + uint32_t stepping) +{ + struct dc_state *context = dc->current_state; + struct dc_clock_config clock_cfg = {0}; + struct dc_clocks *current_clocks = &context->bw_ctx.bw.dcn.clk; + + if (dc->clk_mgr && dc->clk_mgr->funcs->get_clock) + dc->clk_mgr->funcs->get_clock(dc->clk_mgr, + context, clock_type, &clock_cfg); + + if (!dc->clk_mgr->funcs->get_clock) + return DC_FAIL_UNSUPPORTED_1; + + if (clk_khz > clock_cfg.max_clock_khz) + return DC_FAIL_CLK_EXCEED_MAX; + + if (clk_khz < clock_cfg.min_clock_khz) + return DC_FAIL_CLK_BELOW_MIN; + + if (clk_khz < clock_cfg.bw_requirequired_clock_khz) + return DC_FAIL_CLK_BELOW_CFG_REQUIRED; + + /*update internal request clock for update clock use*/ + if (clock_type == DC_CLOCK_TYPE_DISPCLK) + current_clocks->dispclk_khz = clk_khz; + else if (clock_type == DC_CLOCK_TYPE_DPPCLK) + current_clocks->dppclk_khz = clk_khz; + else + return DC_ERROR_UNEXPECTED; + + if (dc->clk_mgr && dc->clk_mgr->funcs->update_clocks) + dc->clk_mgr->funcs->update_clocks(dc->clk_mgr, + context, true); + return DC_OK; -static const struct hw_sequencer_funcs dcn10_funcs = { - .program_gamut_remap = program_gamut_remap, - .init_hw = dcn10_init_hw, - .init_pipes = dcn10_init_pipes, - .apply_ctx_to_hw = dce110_apply_ctx_to_hw, - .apply_ctx_for_surface = dcn10_apply_ctx_for_surface, - .update_plane_addr = dcn10_update_plane_addr, - .plane_atomic_disconnect = hwss1_plane_atomic_disconnect, - .update_dchub = dcn10_update_dchub, - .update_mpcc = dcn10_update_mpcc, - .update_pending_status = dcn10_update_pending_status, - .set_input_transfer_func = dcn10_set_input_transfer_func, - .set_output_transfer_func = dcn10_set_output_transfer_func, - .program_output_csc = dcn10_program_output_csc, - .power_down = dce110_power_down, - .enable_accelerated_mode = dce110_enable_accelerated_mode, - .enable_timing_synchronization = dcn10_enable_timing_synchronization, - .enable_per_frame_crtc_position_reset = dcn10_enable_per_frame_crtc_position_reset, - .update_info_frame = dce110_update_info_frame, - .send_immediate_sdp_message = dcn10_send_immediate_sdp_message, - .enable_stream = dce110_enable_stream, - .disable_stream = dce110_disable_stream, - .unblank_stream = dcn10_unblank_stream, - .blank_stream = dce110_blank_stream, - .enable_audio_stream = dce110_enable_audio_stream, - .disable_audio_stream = dce110_disable_audio_stream, - .enable_display_power_gating = dcn10_dummy_display_power_gating, - .disable_plane = dcn10_disable_plane, - .blank_pixel_data = dcn10_blank_pixel_data, - .pipe_control_lock = dcn10_pipe_control_lock, - .prepare_bandwidth = dcn10_prepare_bandwidth, - .optimize_bandwidth = dcn10_optimize_bandwidth, - .reset_hw_ctx_wrap = dcn10_reset_hw_ctx_wrap, - .enable_stream_timing = dcn10_enable_stream_timing, - .set_drr = set_drr, - .get_position = get_position, - .set_static_screen_control = set_static_screen_control, - .setup_stereo = dcn10_setup_stereo, - .set_avmute = dce110_set_avmute, - .log_hw_state = dcn10_log_hw_state, - .get_hw_state = dcn10_get_hw_state, - .clear_status_bits = dcn10_clear_status_bits, - .wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect, - .edp_backlight_control = hwss_edp_backlight_control, - .edp_power_control = hwss_edp_power_control, - .edp_wait_for_hpd_ready = hwss_edp_wait_for_hpd_ready, - .set_cursor_position = dcn10_set_cursor_position, - .set_cursor_attribute = dcn10_set_cursor_attribute, - .set_cursor_sdr_white_level = dcn10_set_cursor_sdr_white_level, - .disable_stream_gating = NULL, - .enable_stream_gating = NULL, - .setup_periodic_interrupt = dcn10_setup_periodic_interrupt, - .setup_vupdate_interrupt = dcn10_setup_vupdate_interrupt, - .did_underflow_occur = dcn10_did_underflow_occur -}; - - -void dcn10_hw_sequencer_construct(struct dc *dc) -{ - dc->hwss = dcn10_funcs; } +void dcn10_get_clock(struct dc *dc, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg) +{ + struct dc_state *context = dc->current_state; + + if (dc->clk_mgr && dc->clk_mgr->funcs->get_clock) + dc->clk_mgr->funcs->get_clock(dc->clk_mgr, context, clock_type, clock_cfg); + +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h index d3616b1948cc..4d20f6586bb5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h @@ -27,68 +27,160 @@ #define __DC_HWSS_DCN10_H__ #include "core_types.h" +#include "hw_sequencer_private.h" struct dc; void dcn10_hw_sequencer_construct(struct dc *dc); -extern void fill_display_configs( - const struct dc_state *context, - struct dm_pp_display_configuration *pp_display_cfg); - -bool is_rgb_cspace(enum dc_color_space output_color_space); - -void hwss1_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx); - -void dcn10_verify_allow_pstate_change_high(struct dc *dc); +int dcn10_get_vupdate_offset_from_vsync(struct pipe_ctx *pipe_ctx); +void dcn10_setup_vupdate_interrupt(struct dc *dc, struct pipe_ctx *pipe_ctx); +enum dc_status dcn10_enable_stream_timing( + struct pipe_ctx *pipe_ctx, + struct dc_state *context, + struct dc *dc); +void dcn10_optimize_bandwidth( + struct dc *dc, + struct dc_state *context); +void dcn10_prepare_bandwidth( + struct dc *dc, + struct dc_state *context); +void dcn10_pipe_control_lock( + struct dc *dc, + struct pipe_ctx *pipe, + bool lock); +void dcn10_blank_pixel_data( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + bool blank); +void dcn10_unblank_stream(struct pipe_ctx *pipe_ctx, + struct dc_link_settings *link_settings); +void dcn10_program_output_csc(struct dc *dc, + struct pipe_ctx *pipe_ctx, + enum dc_color_space colorspace, + uint16_t *matrix, + int opp_id); +bool dcn10_set_output_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, + const struct dc_stream_state *stream); +bool dcn10_set_input_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, + const struct dc_plane_state *plane_state); +void dcn10_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn10_reset_hw_ctx_wrap( + struct dc *dc, + struct dc_state *context); +void dcn10_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn10_apply_ctx_for_surface( + struct dc *dc, + const struct dc_stream_state *stream, + int num_planes, + struct dc_state *context); +void dcn10_hubp_pg_control( + struct dce_hwseq *hws, + unsigned int hubp_inst, + bool power_on); +void dcn10_dpp_pg_control( + struct dce_hwseq *hws, + unsigned int dpp_inst, + bool power_on); +void dcn10_enable_power_gating_plane( + struct dce_hwseq *hws, + bool enable); +void dcn10_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn10_disable_vga( + struct dce_hwseq *hws); void dcn10_program_pipe( struct dc *dc, struct pipe_ctx *pipe_ctx, struct dc_state *context); - -void dcn10_get_hw_state( +void dcn10_program_gamut_remap(struct pipe_ctx *pipe_ctx); +void dcn10_init_hw(struct dc *dc); +void dcn10_init_pipes(struct dc *dc, struct dc_state *context); +enum dc_status dce110_apply_ctx_to_hw( + struct dc *dc, + struct dc_state *context); +void dcn10_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn10_update_dchub(struct dce_hwseq *hws, struct dchub_init_data *dh_data); +void dcn10_update_pending_status(struct pipe_ctx *pipe_ctx); +void dce110_power_down(struct dc *dc); +void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context); +void dcn10_enable_timing_synchronization( + struct dc *dc, + int group_index, + int group_size, + struct pipe_ctx *grouped_pipes[]); +void dcn10_enable_per_frame_crtc_position_reset( + struct dc *dc, + int group_size, + struct pipe_ctx *grouped_pipes[]); +void dce110_update_info_frame(struct pipe_ctx *pipe_ctx); +void dcn10_send_immediate_sdp_message(struct pipe_ctx *pipe_ctx, + const uint8_t *custom_sdp_message, + unsigned int sdp_message_size); +void dce110_blank_stream(struct pipe_ctx *pipe_ctx); +void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx); +void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx); +bool dcn10_dummy_display_power_gating( struct dc *dc, - char *pBuf, unsigned int bufSize, + uint8_t controller_id, + struct dc_bios *dcb, + enum pipe_gating_control power_gating); +void dcn10_set_drr(struct pipe_ctx **pipe_ctx, + int num_pipes, unsigned int vmin, unsigned int vmax, + unsigned int vmid, unsigned int vmid_frame_number); +void dcn10_get_position(struct pipe_ctx **pipe_ctx, + int num_pipes, + struct crtc_position *position); +void dcn10_set_static_screen_control(struct pipe_ctx **pipe_ctx, + int num_pipes, const struct dc_static_screen_params *params); +void dcn10_setup_stereo(struct pipe_ctx *pipe_ctx, struct dc *dc); +void dce110_set_avmute(struct pipe_ctx *pipe_ctx, bool enable); +void dcn10_log_hw_state(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx); +void dcn10_get_hw_state(struct dc *dc, + char *pBuf, + unsigned int bufSize, unsigned int mask); - void dcn10_clear_status_bits(struct dc *dc, unsigned int mask); - -bool is_lower_pipe_tree_visible(struct pipe_ctx *pipe_ctx); - -bool is_upper_pipe_tree_visible(struct pipe_ctx *pipe_ctx); - -bool is_pipe_tree_visible(struct pipe_ctx *pipe_ctx); - -void dcn10_program_pte_vm(struct dce_hwseq *hws, struct hubp *hubp); - -void set_hdr_multiplier(struct pipe_ctx *pipe_ctx); - +void dcn10_wait_for_mpcc_disconnect( + struct dc *dc, + struct resource_pool *res_pool, + struct pipe_ctx *pipe_ctx); +void dce110_edp_backlight_control( + struct dc_link *link, + bool enable); +void dce110_edp_power_control( + struct dc_link *link, + bool power_up); +void dce110_edp_wait_for_hpd_ready( + struct dc_link *link, + bool power_up); +void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx); +void dcn10_set_cursor_attribute(struct pipe_ctx *pipe_ctx); +void dcn10_set_cursor_sdr_white_level(struct pipe_ctx *pipe_ctx); +void dcn10_setup_periodic_interrupt( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + enum vline_select vline); +enum dc_status dcn10_set_clock(struct dc *dc, + enum dc_clock_type clock_type, + uint32_t clk_khz, + uint32_t stepping); +void dcn10_get_clock(struct dc *dc, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg); +bool dcn10_did_underflow_occur(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn10_bios_golden_init(struct dc *dc); +void dcn10_plane_atomic_power_down(struct dc *dc, + struct dpp *dpp, + struct hubp *hubp); void dcn10_get_surface_visual_confirm_color( const struct pipe_ctx *pipe_ctx, struct tg_color *color); - void dcn10_get_hdr_visual_confirm_color( struct pipe_ctx *pipe_ctx, struct tg_color *color); - -bool dcn10_did_underflow_occur(struct dc *dc, struct pipe_ctx *pipe_ctx); - -void update_dchubp_dpp( - struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct dc_state *context); - -struct pipe_ctx *find_top_pipe_for_stream( - struct dc *dc, - struct dc_state *context, - const struct dc_stream_state *stream); - -int get_vupdate_offset_from_vsync(struct pipe_ctx *pipe_ctx); - -void dcn10_build_prescale_params(struct dc_bias_and_scale *bias_and_scale, - const struct dc_plane_state *plane_state); -void lock_all_pipes(struct dc *dc, - struct dc_state *context, - bool lock); +void dcn10_set_hdr_multiplier(struct pipe_ctx *pipe_ctx); +void dcn10_verify_allow_pstate_change_high(struct dc *dc); #endif /* __DC_HWSS_DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c index 6e47444109d7..7f4766e45dff 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c @@ -45,7 +45,7 @@ #include "dcn10_cm_common.h" #include "clk_mgr.h" -static unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...) +unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...) { unsigned int ret_vsnprintf; unsigned int chars_printed; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.h new file mode 100644 index 000000000000..596f95c22e85 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.h @@ -0,0 +1,43 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_HWSS_DCN10_DEBUG_H__ +#define __DC_HWSS_DCN10_DEBUG_H__ + +#include "core_types.h" + +struct dc; + +void dcn10_clear_status_bits(struct dc *dc, unsigned int mask); + +void dcn10_log_hw_state(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx); + +void dcn10_get_hw_state(struct dc *dc, + char *pBuf, + unsigned int bufSize, + unsigned int mask); + +#endif /* __DC_HWSS_DCN10_DEBUG_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c new file mode 100644 index 000000000000..e7e5352ec424 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c @@ -0,0 +1,111 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "hw_sequencer_private.h" +#include "dce110/dce110_hw_sequencer.h" +#include "dcn10_hw_sequencer.h" + +static const struct hw_sequencer_funcs dcn10_funcs = { + .program_gamut_remap = dcn10_program_gamut_remap, + .init_hw = dcn10_init_hw, + .apply_ctx_to_hw = dce110_apply_ctx_to_hw, + .apply_ctx_for_surface = dcn10_apply_ctx_for_surface, + .update_plane_addr = dcn10_update_plane_addr, + .update_dchub = dcn10_update_dchub, + .update_pending_status = dcn10_update_pending_status, + .program_output_csc = dcn10_program_output_csc, + .enable_accelerated_mode = dce110_enable_accelerated_mode, + .enable_timing_synchronization = dcn10_enable_timing_synchronization, + .enable_per_frame_crtc_position_reset = dcn10_enable_per_frame_crtc_position_reset, + .update_info_frame = dce110_update_info_frame, + .send_immediate_sdp_message = dcn10_send_immediate_sdp_message, + .enable_stream = dce110_enable_stream, + .disable_stream = dce110_disable_stream, + .unblank_stream = dcn10_unblank_stream, + .blank_stream = dce110_blank_stream, + .enable_audio_stream = dce110_enable_audio_stream, + .disable_audio_stream = dce110_disable_audio_stream, + .disable_plane = dcn10_disable_plane, + .pipe_control_lock = dcn10_pipe_control_lock, + .prepare_bandwidth = dcn10_prepare_bandwidth, + .optimize_bandwidth = dcn10_optimize_bandwidth, + .set_drr = dcn10_set_drr, + .get_position = dcn10_get_position, + .set_static_screen_control = dcn10_set_static_screen_control, + .setup_stereo = dcn10_setup_stereo, + .set_avmute = dce110_set_avmute, + .log_hw_state = dcn10_log_hw_state, + .get_hw_state = dcn10_get_hw_state, + .clear_status_bits = dcn10_clear_status_bits, + .wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect, + .edp_power_control = dce110_edp_power_control, + .edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready, + .set_cursor_position = dcn10_set_cursor_position, + .set_cursor_attribute = dcn10_set_cursor_attribute, + .set_cursor_sdr_white_level = dcn10_set_cursor_sdr_white_level, + .setup_periodic_interrupt = dcn10_setup_periodic_interrupt, + .set_clock = dcn10_set_clock, + .get_clock = dcn10_get_clock, + .get_vupdate_offset_from_vsync = dcn10_get_vupdate_offset_from_vsync, +}; + +static const struct hwseq_private_funcs dcn10_private_funcs = { + .init_pipes = dcn10_init_pipes, + .update_plane_addr = dcn10_update_plane_addr, + .plane_atomic_disconnect = dcn10_plane_atomic_disconnect, + .program_pipe = dcn10_program_pipe, + .update_mpcc = dcn10_update_mpcc, + .set_input_transfer_func = dcn10_set_input_transfer_func, + .set_output_transfer_func = dcn10_set_output_transfer_func, + .power_down = dce110_power_down, + .enable_display_power_gating = dcn10_dummy_display_power_gating, + .blank_pixel_data = dcn10_blank_pixel_data, + .reset_hw_ctx_wrap = dcn10_reset_hw_ctx_wrap, + .enable_stream_timing = dcn10_enable_stream_timing, + .edp_backlight_control = dce110_edp_backlight_control, + .disable_stream_gating = NULL, + .enable_stream_gating = NULL, + .setup_vupdate_interrupt = dcn10_setup_vupdate_interrupt, + .did_underflow_occur = dcn10_did_underflow_occur, + .init_blank = NULL, + .disable_vga = dcn10_disable_vga, + .bios_golden_init = dcn10_bios_golden_init, + .plane_atomic_disable = dcn10_plane_atomic_disable, + .plane_atomic_power_down = dcn10_plane_atomic_power_down, + .enable_power_gating_plane = dcn10_enable_power_gating_plane, + .dpp_pg_control = dcn10_dpp_pg_control, + .hubp_pg_control = dcn10_hubp_pg_control, + .dsc_pg_control = NULL, + .get_surface_visual_confirm_color = dcn10_get_surface_visual_confirm_color, + .get_hdr_visual_confirm_color = dcn10_get_hdr_visual_confirm_color, + .set_hdr_multiplier = dcn10_set_hdr_multiplier, + .verify_allow_pstate_change_high = dcn10_verify_allow_pstate_change_high, +}; + +void dcn10_hw_sequencer_construct(struct dc *dc) +{ + dc->hwss = dcn10_funcs; + dc->hwseq->funcs = dcn10_private_funcs; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.h new file mode 100644 index 000000000000..8c6fd7b844a4 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.h @@ -0,0 +1,33 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_DCN10_INIT_H__ +#define __DC_DCN10_INIT_H__ + +struct dc; + +void dcn10_hw_sequencer_construct(struct dc *dc); + +#endif /* __DC_DCN10_INIT_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_ipp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_ipp.c index 0fb9e440cb9d..f05371c1fc36 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_ipp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_ipp.c @@ -53,11 +53,9 @@ static const struct ipp_funcs dcn10_ipp_funcs = { .ipp_destroy = dcn10_ipp_destroy }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) static const struct ipp_funcs dcn20_ipp_funcs = { .ipp_destroy = dcn10_ipp_destroy }; -#endif void dcn10_ipp_construct( struct dcn10_ipp *ippn10, @@ -76,7 +74,6 @@ void dcn10_ipp_construct( ippn10->ipp_mask = ipp_mask; } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void dcn20_ipp_construct( struct dcn10_ipp *ippn10, struct dc_context *ctx, @@ -93,4 +90,3 @@ void dcn20_ipp_construct( ippn10->ipp_shift = ipp_shift; ippn10->ipp_mask = ipp_mask; } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_ipp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_ipp.h index cfa24459242b..f0e0d07b0311 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_ipp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_ipp.h @@ -49,7 +49,6 @@ SRI(CURSOR_HOT_SPOT, CURSOR, id), \ SRI(CURSOR_DST_OFFSET, CURSOR, id) -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define IPP_REG_LIST_DCN20(id) \ IPP_REG_LIST_DCN(id), \ SRI(CURSOR_SETTINGS, HUBPREQ, id), \ @@ -60,7 +59,6 @@ SRI(CURSOR_POSITION, CURSOR0_, id), \ SRI(CURSOR_HOT_SPOT, CURSOR0_, id), \ SRI(CURSOR_DST_OFFSET, CURSOR0_, id) -#endif #define CURSOR0_CURSOR_CONTROL__CURSOR_2X_MAGNIFY__SHIFT 0x4 #define CURSOR0_CURSOR_CONTROL__CURSOR_2X_MAGNIFY_MASK 0x00000010L @@ -105,7 +103,6 @@ IPP_SF(CURSOR0_CURSOR_DST_OFFSET, CURSOR_DST_X_OFFSET, mask_sh), \ IPP_SF(CNVC_CFG0_FORMAT_CONTROL, OUTPUT_FP, mask_sh) -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define IPP_MASK_SH_LIST_DCN20(mask_sh) \ IPP_MASK_SH_LIST_DCN(mask_sh), \ IPP_SF(HUBPREQ0_CURSOR_SETTINGS, CURSOR0_DST_Y_OFFSET, mask_sh), \ @@ -124,7 +121,6 @@ IPP_SF(CURSOR0_0_CURSOR_HOT_SPOT, CURSOR_HOT_SPOT_X, mask_sh), \ IPP_SF(CURSOR0_0_CURSOR_HOT_SPOT, CURSOR_HOT_SPOT_Y, mask_sh), \ IPP_SF(CURSOR0_0_CURSOR_DST_OFFSET, CURSOR_DST_X_OFFSET, mask_sh) -#endif #define IPP_DCN10_REG_FIELD_LIST(type) \ type CNVC_SURFACE_PIXEL_FORMAT; \ @@ -196,13 +192,11 @@ void dcn10_ipp_construct(struct dcn10_ipp *ippn10, const struct dcn10_ipp_shift *ipp_shift, const struct dcn10_ipp_mask *ipp_mask); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void dcn20_ipp_construct(struct dcn10_ipp *ippn10, struct dc_context *ctx, int inst, const struct dcn10_ipp_registers *regs, const struct dcn10_ipp_shift *ipp_shift, const struct dcn10_ipp_mask *ipp_mask); -#endif #endif /* _DCN10_IPP_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c index 549d423a01f6..1a37c90e9d43 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c @@ -89,6 +89,7 @@ static const struct link_encoder_funcs dcn10_lnk_enc_funcs = { .disable_hpd = dcn10_link_encoder_disable_hpd, .is_dig_enabled = dcn10_is_dig_enabled, .get_dig_frontend = dcn10_get_dig_frontend, + .get_dig_mode = dcn10_get_dig_mode, .destroy = dcn10_link_encoder_destroy }; @@ -446,6 +447,46 @@ static uint8_t get_frontend_source( } } +unsigned int dcn10_get_dig_frontend(struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + int32_t value; + enum engine_id result; + + REG_GET(DIG_BE_CNTL, DIG_FE_SOURCE_SELECT, &value); + + switch (value) { + case DCN10_DIG_FE_SOURCE_SELECT_DIGA: + result = ENGINE_ID_DIGA; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGB: + result = ENGINE_ID_DIGB; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGC: + result = ENGINE_ID_DIGC; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGD: + result = ENGINE_ID_DIGD; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGE: + result = ENGINE_ID_DIGE; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGF: + result = ENGINE_ID_DIGF; + break; + case DCN10_DIG_FE_SOURCE_SELECT_DIGG: + result = ENGINE_ID_DIGG; + break; + default: + // invalid source select DIG + ASSERT(false); + result = ENGINE_ID_UNKNOWN; + } + + return result; + +} + void enc1_configure_encoder( struct dcn10_link_encoder *enc10, const struct dc_link_settings *link_settings) @@ -501,15 +542,6 @@ bool dcn10_is_dig_enabled(struct link_encoder *enc) return value; } -unsigned int dcn10_get_dig_frontend(struct link_encoder *enc) -{ - struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); - uint32_t value; - - REG_GET(DIG_BE_CNTL, DIG_FE_SOURCE_SELECT, &value); - return value; -} - static void link_encoder_disable(struct dcn10_link_encoder *enc10) { /* reset training pattern */ @@ -1366,3 +1398,25 @@ void dcn10_aux_initialize(struct dcn10_link_encoder *enc10) AUX_REG_UPDATE(AUX_DPHY_RX_CONTROL0, AUX_RX_RECEIVE_WINDOW, 0); } + +enum signal_type dcn10_get_dig_mode( + struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + uint32_t value; + REG_GET(DIG_BE_CNTL, DIG_MODE, &value); + switch (value) { + case 1: + return SIGNAL_TYPE_DISPLAY_PORT; + case 2: + return SIGNAL_TYPE_DVI_SINGLE_LINK; + case 3: + return SIGNAL_TYPE_HDMI_TYPE_A; + case 5: + return SIGNAL_TYPE_DISPLAY_PORT_MST; + default: + return SIGNAL_TYPE_NONE; + } + return SIGNAL_TYPE_NONE; +} + diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h index 33b2af1a181c..eb13589b9a81 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h @@ -72,9 +72,7 @@ struct dcn10_link_enc_aux_registers { uint32_t AUX_CONTROL; uint32_t AUX_DPHY_RX_CONTROL0; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 uint32_t AUX_DPHY_TX_CONTROL; -#endif }; struct dcn10_link_enc_hpd_registers { @@ -106,13 +104,46 @@ struct dcn10_link_enc_registers { uint32_t DP_DPHY_HBR2_PATTERN_CONTROL; uint32_t DP_SEC_CNTL1; uint32_t TMDS_CTL_BITS; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* DCCG */ uint32_t CLOCK_ENABLE; /* DIG */ uint32_t DIG_LANE_ENABLE; /* UNIPHY */ uint32_t CHANNEL_XBAR_CNTL; + /* DPCS */ + uint32_t RDPCSTX_PHY_CNTL3; + uint32_t RDPCSTX_PHY_CNTL4; + uint32_t RDPCSTX_PHY_CNTL5; + uint32_t RDPCSTX_PHY_CNTL6; + uint32_t RDPCSTX_PHY_CNTL7; + uint32_t RDPCSTX_PHY_CNTL8; + uint32_t RDPCSTX_PHY_CNTL9; + uint32_t RDPCSTX_PHY_CNTL10; + uint32_t RDPCSTX_PHY_CNTL11; + uint32_t RDPCSTX_PHY_CNTL12; + uint32_t RDPCSTX_PHY_CNTL13; + uint32_t RDPCSTX_PHY_CNTL14; + uint32_t RDPCSTX_PHY_CNTL15; + uint32_t RDPCSTX_CNTL; + uint32_t RDPCSTX_CLOCK_CNTL; + uint32_t RDPCSTX_PHY_CNTL0; + uint32_t RDPCSTX_PHY_CNTL2; + uint32_t RDPCSTX_PLL_UPDATE_DATA; + uint32_t RDPCS_TX_CR_ADDR; + uint32_t RDPCS_TX_CR_DATA; + uint32_t DPCSTX_TX_CLOCK_CNTL; + uint32_t DPCSTX_TX_CNTL; + uint32_t RDPCSTX_INTERRUPT_CONTROL; + uint32_t RDPCSTX_PHY_FUSE0; + uint32_t RDPCSTX_PHY_FUSE1; + uint32_t RDPCSTX_PHY_FUSE2; + uint32_t RDPCSTX_PHY_FUSE3; + uint32_t RDPCSTX_PHY_RX_LD_VAL; + uint32_t DPCSTX_DEBUG_CONFIG; + uint32_t RDPCSTX_DEBUG_CONFIG; + uint32_t RDPCSTX0_RDPCSTX_SCRATCH; + uint32_t RDPCSTX_DMCU_DPALT_DIS_BLOCK_REG; + uint32_t DCIO_SOFT_RESET; /* indirect registers */ uint32_t RAWLANE0_DIG_PCS_XF_RX_OVRD_IN_2; uint32_t RAWLANE0_DIG_PCS_XF_RX_OVRD_IN_3; @@ -122,7 +153,6 @@ struct dcn10_link_enc_registers { uint32_t RAWLANE2_DIG_PCS_XF_RX_OVRD_IN_3; uint32_t RAWLANE3_DIG_PCS_XF_RX_OVRD_IN_2; uint32_t RAWLANE3_DIG_PCS_XF_RX_OVRD_IN_3; -#endif }; #define LE_SF(reg_name, field_name, post_fix)\ @@ -228,7 +258,6 @@ struct dcn10_link_enc_registers { type AUX_LS_READ_EN;\ type AUX_RX_RECEIVE_WINDOW -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define DCN20_LINK_ENCODER_DPCS_REG_FIELD_LIST(type) \ type RDPCS_PHY_DP_TX0_DATA_EN;\ @@ -250,6 +279,10 @@ struct dcn10_link_enc_registers { type RDPCS_EXT_REFCLK_EN;\ type RDPCS_TX_FIFO_EN;\ type UNIPHY_LINK_ENABLE;\ + type UNIPHY_CHANNEL0_XBAR_SOURCE;\ + type UNIPHY_CHANNEL1_XBAR_SOURCE;\ + type UNIPHY_CHANNEL2_XBAR_SOURCE;\ + type UNIPHY_CHANNEL3_XBAR_SOURCE;\ type UNIPHY_CHANNEL0_INVERT;\ type UNIPHY_CHANNEL1_INVERT;\ type UNIPHY_CHANNEL2_INVERT;\ @@ -337,15 +370,46 @@ struct dcn10_link_enc_registers { type RDPCS_TX_FIFO_ERROR_MASK;\ type RDPCS_DPALT_DISABLE_TOGGLE_MASK;\ type RDPCS_DPALT_4LANE_TOGGLE_MASK;\ + type RDPCS_PHY_DPALT_DP4;\ + type RDPCS_PHY_DPALT_DISABLE;\ type RDPCS_PHY_DPALT_DISABLE_ACK;\ type RDPCS_PHY_DP_MPLLB_V2I;\ type RDPCS_PHY_DP_MPLLB_FREQ_VCO;\ + type RDPCS_PHY_DP_MPLLB_CP_INT_GS;\ + type RDPCS_PHY_RX_VREF_CTRL;\ type RDPCS_PHY_DP_MPLLB_CP_INT;\ type RDPCS_PHY_DP_MPLLB_CP_PROP;\ type RDPCS_PHY_RX_REF_LD_VAL;\ type RDPCS_PHY_RX_VCO_LD_VAL;\ type DPCSTX_DEBUG_CONFIG; \ - type RDPCSTX_DEBUG_CONFIG + type RDPCSTX_DEBUG_CONFIG; \ + type RDPCS_PHY_DP_TX0_EQ_MAIN;\ + type RDPCS_PHY_DP_TX0_EQ_PRE;\ + type RDPCS_PHY_DP_TX0_EQ_POST;\ + type RDPCS_PHY_DP_TX1_EQ_MAIN;\ + type RDPCS_PHY_DP_TX1_EQ_PRE;\ + type RDPCS_PHY_DP_TX1_EQ_POST;\ + type RDPCS_PHY_DP_TX2_EQ_MAIN;\ + type RDPCS_PHY_DP_MPLLB_CP_PROP_GS;\ + type RDPCS_PHY_DP_TX2_EQ_PRE;\ + type RDPCS_PHY_DP_TX2_EQ_POST;\ + type RDPCS_PHY_DP_TX3_EQ_MAIN;\ + type RDPCS_PHY_DCO_RANGE;\ + type RDPCS_PHY_DCO_FINETUNE;\ + type RDPCS_PHY_DP_TX3_EQ_PRE;\ + type RDPCS_PHY_DP_TX3_EQ_POST;\ + type RDPCS_PHY_SUP_PRE_HP;\ + type RDPCS_PHY_DP_TX0_VREGDRV_BYP;\ + type RDPCS_PHY_DP_TX1_VREGDRV_BYP;\ + type RDPCS_PHY_DP_TX2_VREGDRV_BYP;\ + type RDPCS_PHY_DP_TX3_VREGDRV_BYP;\ + type RDPCS_DMCU_DPALT_DIS_BLOCK_REG;\ + type UNIPHYA_SOFT_RESET;\ + type UNIPHYB_SOFT_RESET;\ + type UNIPHYC_SOFT_RESET;\ + type UNIPHYD_SOFT_RESET;\ + type UNIPHYE_SOFT_RESET;\ + type UNIPHYF_SOFT_RESET #define DCN20_LINK_ENCODER_REG_FIELD_LIST(type) \ type DIG_LANE0EN;\ @@ -374,20 +438,15 @@ struct dcn10_link_enc_registers { type AUX_TX_PRECHARGE_SYMBOLS; \ type AUX_MODE_DET_CHECK_DELAY;\ type DPCS_DBG_CBUS_DIS -#endif struct dcn10_link_enc_shift { DCN_LINK_ENCODER_REG_FIELD_LIST(uint8_t); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) DCN20_LINK_ENCODER_REG_FIELD_LIST(uint8_t); -#endif }; struct dcn10_link_enc_mask { DCN_LINK_ENCODER_REG_FIELD_LIST(uint32_t); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) DCN20_LINK_ENCODER_REG_FIELD_LIST(uint32_t); -#endif }; struct dcn10_link_encoder { @@ -514,4 +573,6 @@ unsigned int dcn10_get_dig_frontend(struct link_encoder *enc); void dcn10_aux_initialize(struct dcn10_link_encoder *enc10); +enum signal_type dcn10_get_dig_mode( + struct link_encoder *enc); #endif /* __DC_LINK_ENCODER__DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c index 0bca011ed7c9..04f863499cfb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c @@ -42,20 +42,27 @@ void mpc1_set_bg_color(struct mpc *mpc, int mpcc_id) { struct dcn10_mpc *mpc10 = TO_DCN10_MPC(mpc); + struct mpcc *bottommost_mpcc = mpc1_get_mpcc(mpc, mpcc_id); + uint32_t bg_r_cr, bg_g_y, bg_b_cb; + + /* find bottommost mpcc. */ + while (bottommost_mpcc->mpcc_bot) { + bottommost_mpcc = bottommost_mpcc->mpcc_bot; + } /* mpc color is 12 bit. tg_color is 10 bit */ /* todo: might want to use 16 bit to represent color and have each * hw block translate to correct color depth. */ - uint32_t bg_r_cr = bg_color->color_r_cr << 2; - uint32_t bg_g_y = bg_color->color_g_y << 2; - uint32_t bg_b_cb = bg_color->color_b_cb << 2; + bg_r_cr = bg_color->color_r_cr << 2; + bg_g_y = bg_color->color_g_y << 2; + bg_b_cb = bg_color->color_b_cb << 2; - REG_SET(MPCC_BG_R_CR[mpcc_id], 0, + REG_SET(MPCC_BG_R_CR[bottommost_mpcc->mpcc_id], 0, MPCC_BG_R_CR, bg_r_cr); - REG_SET(MPCC_BG_G_Y[mpcc_id], 0, + REG_SET(MPCC_BG_G_Y[bottommost_mpcc->mpcc_id], 0, MPCC_BG_G_Y, bg_g_y); - REG_SET(MPCC_BG_B_CB[mpcc_id], 0, + REG_SET(MPCC_BG_B_CB[bottommost_mpcc->mpcc_id], 0, MPCC_BG_B_CB, bg_b_cb); } @@ -211,7 +218,7 @@ struct mpcc *mpc1_insert_plane( } else { new_mpcc->mpcc_bot = NULL; REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf); - REG_UPDATE(MPCC_CONTROL[mpcc_id], MPCC_MODE, MPCC_BLEND_MODE_TOP_LAYER_PASSTHROUGH); + REG_UPDATE(MPCC_CONTROL[mpcc_id], MPCC_MODE, MPCC_BLEND_MODE_TOP_LAYER_ONLY); } REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, dpp_id); REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, tree->opp_id); @@ -364,6 +371,24 @@ void mpc1_mpc_init(struct mpc *mpc) } } +void mpc1_mpc_init_single_inst(struct mpc *mpc, unsigned int mpcc_id) +{ + struct dcn10_mpc *mpc10 = TO_DCN10_MPC(mpc); + int opp_id; + + REG_GET(MPCC_OPP_ID[mpcc_id], MPCC_OPP_ID, &opp_id); + + REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf); + REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf); + REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf); + + mpc1_init_mpcc(&(mpc->mpcc_array[mpcc_id]), mpcc_id); + + if (opp_id < MAX_OPP && REG(MUX[opp_id])) + REG_UPDATE(MUX[opp_id], MPC_OUT_MUX, 0xf); +} + + void mpc1_init_mpcc_list_from_hw( struct mpc *mpc, struct mpc_tree *tree) @@ -433,17 +458,16 @@ static const struct mpc_funcs dcn10_mpc_funcs = { .insert_plane = mpc1_insert_plane, .remove_mpcc = mpc1_remove_mpcc, .mpc_init = mpc1_mpc_init, + .mpc_init_single_inst = mpc1_mpc_init_single_inst, .get_mpcc_for_dpp = mpc1_get_mpcc_for_dpp, .wait_for_idle = mpc1_assert_idle_mpcc, .assert_mpcc_idle_before_connect = mpc1_assert_mpcc_idle_before_connect, .init_mpcc_list_from_hw = mpc1_init_mpcc_list_from_hw, .update_blending = mpc1_update_blending, -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) .set_denorm = NULL, .set_denorm_clamp = NULL, .set_output_csc = NULL, .set_output_gamma = NULL, -#endif }; void dcn10_mpc_construct(struct dcn10_mpc *mpc10, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h index d3d16c4cbea3..962a68e322ee 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h @@ -149,6 +149,10 @@ void mpc1_remove_mpcc( void mpc1_mpc_init( struct mpc *mpc); +void mpc1_mpc_init_single_inst( + struct mpc *mpc, + unsigned int mpcc_id); + void mpc1_assert_idle_mpcc( struct mpc *mpc, int id); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c index e9ebbbe256b4..d79718fde5a6 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c @@ -168,7 +168,10 @@ static void opp1_set_pixel_encoding( REG_UPDATE(FMT_CONTROL, FMT_PIXEL_ENCODING, 0); break; case PIXEL_ENCODING_YCBCR422: - REG_UPDATE(FMT_CONTROL, FMT_PIXEL_ENCODING, 1); + REG_UPDATE_3(FMT_CONTROL, + FMT_PIXEL_ENCODING, 1, + FMT_SUBSAMPLING_MODE, 2, + FMT_CBCR_BIT_REDUCTION_BYPASS, 0); break; case PIXEL_ENCODING_YCBCR420: REG_UPDATE(FMT_CONTROL, FMT_PIXEL_ENCODING, 2); @@ -237,6 +240,9 @@ void opp1_set_dyn_expansion( FMT_DYNAMIC_EXP_EN, 0, FMT_DYNAMIC_EXP_MODE, 0); + if (opp->dyn_expansion == DYN_EXPANSION_DISABLE) + return; + /*00 - 10-bit -> 12-bit dynamic expansion*/ /*01 - 8-bit -> 12-bit dynamic expansion*/ if (signal == SIGNAL_TYPE_HDMI_TYPE_A || @@ -367,11 +373,9 @@ void opp1_program_oppbuf( */ REG_UPDATE(OPPBUF_CONTROL, OPPBUF_PIXEL_REPETITION, oppbuf->pixel_repetition); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* Controls the number of padded pixels at the end of a segment */ if (REG(OPPBUF_CONTROL1)) REG_UPDATE(OPPBUF_CONTROL1, OPPBUF_NUM_SEGMENT_PADDED_PIXELS, oppbuf->num_segment_padded_pixels); -#endif } void opp1_pipe_clock_control(struct output_pixel_processor *opp, bool enable) @@ -398,9 +402,8 @@ static const struct opp_funcs dcn10_opp_funcs = { .opp_program_bit_depth_reduction = opp1_program_bit_depth_reduction, .opp_program_stereo = opp1_program_stereo, .opp_pipe_clock_control = opp1_pipe_clock_control, -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) .opp_set_disp_pattern_generator = NULL, -#endif + .dpg_is_blanked = NULL, .opp_destroy = opp1_destroy }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.h index 0f10adea000c..2c0ecfa5a643 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.h @@ -116,6 +116,8 @@ type FMT_RAND_G_SEED; \ type FMT_RAND_B_SEED; \ type FMT_PIXEL_ENCODING; \ + type FMT_SUBSAMPLING_MODE; \ + type FMT_CBCR_BIT_REDUCTION_BYPASS; \ type FMT_CLAMP_DATA_EN; \ type FMT_CLAMP_COLOR_FORMAT; \ type FMT_DYNAMIC_EXP_EN; \ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index a546c2bc9129..a9a43b397db9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -154,7 +154,7 @@ void optc1_program_timing( uint32_t h_sync_polarity, v_sync_polarity; uint32_t start_point = 0; uint32_t field_num = 0; - uint32_t h_div_2; + enum h_timing_div_mode h_div = H_TIMING_NO_DIV; struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -285,10 +285,11 @@ void optc1_program_timing( * of stereo handled in explicit call */ - h_div_2 = optc1_is_two_pixels_per_containter(&patched_crtc_timing); - REG_UPDATE(OTG_H_TIMING_CNTL, - OTG_H_TIMING_DIV_BY2, h_div_2 || optc1->comb_opp_id != 0xf); + if (optc1_is_two_pixels_per_containter(&patched_crtc_timing) || optc1->opp_count == 2) + h_div = H_TIMING_DIV_BY2; + REG_UPDATE(OTG_H_TIMING_CNTL, + OTG_H_TIMING_DIV_BY2, h_div); } void optc1_set_vtg_params(struct timing_generator *optc, @@ -456,11 +457,16 @@ static bool optc1_enable_crtc(struct timing_generator *optc) REG_UPDATE(CONTROL, VTG0_ENABLE, 1); + REG_SEQ_START(); + /* Enable CRTC */ REG_UPDATE_2(OTG_CONTROL, OTG_DISABLE_POINT_CNTL, 3, OTG_MASTER_EN, 1); + REG_SEQ_SUBMIT(); + REG_SEQ_WAIT_DONE(); + return true; } @@ -783,21 +789,26 @@ void optc1_set_early_control( void optc1_set_static_screen_control( struct timing_generator *optc, - uint32_t value) + uint32_t event_triggers, + uint32_t num_frames) { struct optc *optc1 = DCN10TG_FROM_TG(optc); + // By register spec, it only takes 8 bit value + if (num_frames > 0xFF) + num_frames = 0xFF; + /* Bit 8 is no longer applicable in RV for PSR case, * set bit 8 to 0 if given */ - if ((value & STATIC_SCREEN_EVENT_MASK_RANGETIMING_DOUBLE_BUFFER_UPDATE_EN) + if ((event_triggers & STATIC_SCREEN_EVENT_MASK_RANGETIMING_DOUBLE_BUFFER_UPDATE_EN) != 0) - value = value & + event_triggers = event_triggers & ~STATIC_SCREEN_EVENT_MASK_RANGETIMING_DOUBLE_BUFFER_UPDATE_EN; REG_SET_2(OTG_STATIC_SCREEN_CONTROL, 0, - OTG_STATIC_SCREEN_EVENT_MASK, value, - OTG_STATIC_SCREEN_FRAME_COUNT, 2); + OTG_STATIC_SCREEN_EVENT_MASK, event_triggers, + OTG_STATIC_SCREEN_FRAME_COUNT, num_frames); } void optc1_setup_manual_trigger(struct timing_generator *optc) @@ -824,6 +835,9 @@ void optc1_program_manual_trigger(struct timing_generator *optc) REG_SET(OTG_MANUAL_FLOW_CONTROL, 0, MANUAL_FLOW_CONTROL, 1); + + REG_SET(OTG_MANUAL_FLOW_CONTROL, 0, + MANUAL_FLOW_CONTROL, 0); } @@ -846,6 +860,18 @@ void optc1_set_drr( params->vertical_total_max > 0 && params->vertical_total_min > 0) { + if (params->vertical_total_mid != 0) { + + REG_SET(OTG_V_TOTAL_MID, 0, + OTG_V_TOTAL_MID, params->vertical_total_mid - 1); + + REG_UPDATE_2(OTG_V_TOTAL_CONTROL, + OTG_VTOTAL_MID_REPLACING_MAX_EN, 1, + OTG_VTOTAL_MID_FRAME_NUM, + (uint8_t)params->vertical_total_mid_frame_num); + + } + REG_SET(OTG_V_TOTAL_MAX, 0, OTG_V_TOTAL_MAX, params->vertical_total_max - 1); @@ -1214,59 +1240,25 @@ bool optc1_is_stereo_left_eye(struct timing_generator *optc) return ret; } -bool optc1_is_matching_timing(struct timing_generator *tg, - const struct dc_crtc_timing *otg_timing) +bool optc1_get_hw_timing(struct timing_generator *tg, + struct dc_crtc_timing *hw_crtc_timing) { - struct dc_crtc_timing hw_crtc_timing = {0}; struct dcn_otg_state s = {0}; - if (tg == NULL || otg_timing == NULL) + if (tg == NULL || hw_crtc_timing == NULL) return false; optc1_read_otg_state(DCN10TG_FROM_TG(tg), &s); - hw_crtc_timing.h_total = s.h_total + 1; - hw_crtc_timing.h_addressable = s.h_total - ((s.h_total - s.h_blank_start) + s.h_blank_end); - hw_crtc_timing.h_front_porch = s.h_total + 1 - s.h_blank_start; - hw_crtc_timing.h_sync_width = s.h_sync_a_end - s.h_sync_a_start; - - hw_crtc_timing.v_total = s.v_total + 1; - hw_crtc_timing.v_addressable = s.v_total - ((s.v_total - s.v_blank_start) + s.v_blank_end); - hw_crtc_timing.v_front_porch = s.v_total + 1 - s.v_blank_start; - hw_crtc_timing.v_sync_width = s.v_sync_a_end - s.v_sync_a_start; - - if (otg_timing->h_total != hw_crtc_timing.h_total) - return false; - - if (otg_timing->h_border_left != hw_crtc_timing.h_border_left) - return false; - - if (otg_timing->h_addressable != hw_crtc_timing.h_addressable) - return false; + hw_crtc_timing->h_total = s.h_total + 1; + hw_crtc_timing->h_addressable = s.h_total - ((s.h_total - s.h_blank_start) + s.h_blank_end); + hw_crtc_timing->h_front_porch = s.h_total + 1 - s.h_blank_start; + hw_crtc_timing->h_sync_width = s.h_sync_a_end - s.h_sync_a_start; - if (otg_timing->h_border_right != hw_crtc_timing.h_border_right) - return false; - - if (otg_timing->h_front_porch != hw_crtc_timing.h_front_porch) - return false; - - if (otg_timing->h_sync_width != hw_crtc_timing.h_sync_width) - return false; - - if (otg_timing->v_total != hw_crtc_timing.v_total) - return false; - - if (otg_timing->v_border_top != hw_crtc_timing.v_border_top) - return false; - - if (otg_timing->v_addressable != hw_crtc_timing.v_addressable) - return false; - - if (otg_timing->v_border_bottom != hw_crtc_timing.v_border_bottom) - return false; - - if (otg_timing->v_sync_width != hw_crtc_timing.v_sync_width) - return false; + hw_crtc_timing->v_total = s.v_total + 1; + hw_crtc_timing->v_addressable = s.v_total - ((s.v_total - s.v_blank_start) + s.v_blank_end); + hw_crtc_timing->v_front_porch = s.v_total + 1 - s.v_blank_start; + hw_crtc_timing->v_sync_width = s.v_sync_a_end - s.v_sync_a_start; return true; } @@ -1470,7 +1462,6 @@ static const struct timing_generator_funcs dcn10_tg_funcs = { .get_frame_count = optc1_get_vblank_counter, .get_scanoutpos = optc1_get_crtc_scanoutpos, .get_otg_active_size = optc1_get_otg_active_size, - .is_matching_timing = optc1_is_matching_timing, .set_early_control = optc1_set_early_control, /* used by enable_timing_synchronization. Not need for FPGA */ .wait_for_state = optc1_wait_for_state, @@ -1498,7 +1489,8 @@ static const struct timing_generator_funcs dcn10_tg_funcs = { .configure_crc = optc1_configure_crc, .set_vtg_params = optc1_set_vtg_params, .program_manual_trigger = optc1_program_manual_trigger, - .setup_manual_trigger = optc1_setup_manual_trigger + .setup_manual_trigger = optc1_setup_manual_trigger, + .get_hw_timing = optc1_get_hw_timing, }; void dcn10_timing_generator_init(struct optc *optc1) @@ -1513,10 +1505,8 @@ void dcn10_timing_generator_init(struct optc *optc1) optc1->min_v_blank_interlace = 5; optc1->min_h_sync_width = 8; optc1->min_v_sync_width = 1; - optc1->comb_opp_id = 0xf; } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT /* "Containter" vs. "pixel" is a concept within HW blocks, mostly those closer to the back-end. It works like this: * * - In most of the formats (RGB or YCbCr 4:4:4, 4:2:2 uncompressed and DSC 4:2:2 Simple) pixel rate is the same as @@ -1529,15 +1519,12 @@ void dcn10_timing_generator_init(struct optc *optc1) * to it) and has to be treated the same as 4:2:0, i.e. target containter rate has to be halved in this case as well. * */ -#endif bool optc1_is_two_pixels_per_containter(const struct dc_crtc_timing *timing) { bool two_pix = timing->pixel_encoding == PIXEL_ENCODING_YCBCR420; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT two_pix = two_pix || (timing->flags.DSC && timing->pixel_encoding == PIXEL_ENCODING_YCBCR422 && !timing->dsc_cfg.ycbcr422_simple); -#endif return two_pix; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h index 02599eb92ca6..f277656d5464 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h @@ -54,6 +54,7 @@ SRI(OTG_3D_STRUCTURE_CONTROL, OTG, inst),\ SRI(OTG_STEREO_STATUS, OTG, inst),\ SRI(OTG_V_TOTAL_MAX, OTG, inst),\ + SRI(OTG_V_TOTAL_MID, OTG, inst),\ SRI(OTG_V_TOTAL_MIN, OTG, inst),\ SRI(OTG_V_TOTAL_CONTROL, OTG, inst),\ SRI(OTG_TRIGA_CNTL, OTG, inst),\ @@ -125,6 +126,7 @@ struct dcn_optc_registers { uint32_t OTG_3D_STRUCTURE_CONTROL; uint32_t OTG_STEREO_STATUS; uint32_t OTG_V_TOTAL_MAX; + uint32_t OTG_V_TOTAL_MID; uint32_t OTG_V_TOTAL_MIN; uint32_t OTG_V_TOTAL_CONTROL; uint32_t OTG_TRIGA_CNTL; @@ -163,13 +165,11 @@ struct dcn_optc_registers { uint32_t OTG_CRC0_WINDOWB_X_CONTROL; uint32_t OTG_CRC0_WINDOWB_Y_CONTROL; uint32_t GSL_SOURCE_SELECT; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 uint32_t DWB_SOURCE_SELECT; uint32_t OTG_DSC_START_POSITION; uint32_t OPTC_DATA_FORMAT_CONTROL; uint32_t OPTC_BYTES_PER_PIXEL; uint32_t OPTC_WIDTH_CONTROL; -#endif }; #define TG_COMMON_MASK_SH_LIST_DCN(mask_sh)\ @@ -214,12 +214,15 @@ struct dcn_optc_registers { SF(OTG0_OTG_3D_STRUCTURE_CONTROL, OTG_3D_STRUCTURE_V_UPDATE_MODE, mask_sh),\ SF(OTG0_OTG_3D_STRUCTURE_CONTROL, OTG_3D_STRUCTURE_STEREO_SEL_OVR, mask_sh),\ SF(OTG0_OTG_V_TOTAL_MAX, OTG_V_TOTAL_MAX, mask_sh),\ + SF(OTG0_OTG_V_TOTAL_MID, OTG_V_TOTAL_MID, mask_sh),\ SF(OTG0_OTG_V_TOTAL_MIN, OTG_V_TOTAL_MIN, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_V_TOTAL_MIN_SEL, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_V_TOTAL_MAX_SEL, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_FORCE_LOCK_ON_EVENT, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_SET_V_TOTAL_MIN_MASK_EN, mask_sh),\ SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_SET_V_TOTAL_MIN_MASK, mask_sh),\ + SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_VTOTAL_MID_REPLACING_MAX_EN, mask_sh),\ + SF(OTG0_OTG_V_TOTAL_CONTROL, OTG_VTOTAL_MID_FRAME_NUM, mask_sh),\ SF(OTG0_OTG_FORCE_COUNT_NOW_CNTL, OTG_FORCE_COUNT_NOW_CLEAR, mask_sh),\ SF(OTG0_OTG_FORCE_COUNT_NOW_CNTL, OTG_FORCE_COUNT_NOW_MODE, mask_sh),\ SF(OTG0_OTG_FORCE_COUNT_NOW_CNTL, OTG_FORCE_COUNT_NOW_OCCURRED, mask_sh),\ @@ -348,9 +351,12 @@ struct dcn_optc_registers { type OTG_3D_STRUCTURE_V_UPDATE_MODE;\ type OTG_3D_STRUCTURE_STEREO_SEL_OVR;\ type OTG_V_TOTAL_MAX;\ + type OTG_V_TOTAL_MID;\ type OTG_V_TOTAL_MIN;\ type OTG_V_TOTAL_MIN_SEL;\ type OTG_V_TOTAL_MAX_SEL;\ + type OTG_VTOTAL_MID_REPLACING_MAX_EN;\ + type OTG_VTOTAL_MID_FRAME_NUM;\ type OTG_FORCE_LOCK_ON_EVENT;\ type OTG_SET_V_TOTAL_MIN_MASK_EN;\ type OTG_SET_V_TOTAL_MIN_MASK;\ @@ -448,7 +454,6 @@ struct dcn_optc_registers { type MANUAL_FLOW_CONTROL;\ type MANUAL_FLOW_CONTROL_SEL; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 #define TG_REG_FIELD_LIST(type) \ TG_REG_FIELD_LIST_DCN1_0(type)\ @@ -471,12 +476,6 @@ struct dcn_optc_registers { type OPTC_DWB0_SOURCE_SELECT;\ type OPTC_DWB1_SOURCE_SELECT; -#else - -#define TG_REG_FIELD_LIST(type) \ - TG_REG_FIELD_LIST_DCN1_0(type) - -#endif struct dcn_optc_shift { @@ -494,7 +493,7 @@ struct optc { const struct dcn_optc_shift *tg_shift; const struct dcn_optc_mask *tg_mask; - int comb_opp_id; + int opp_count; uint32_t max_h_total; uint32_t max_v_total; @@ -534,11 +533,15 @@ struct dcn_otg_state { uint32_t h_total; uint32_t underflow_occurred_status; uint32_t otg_enabled; + uint32_t blank_enabled; }; void optc1_read_otg_state(struct optc *optc1, struct dcn_otg_state *s); +bool optc1_get_hw_timing(struct timing_generator *tg, + struct dc_crtc_timing *hw_crtc_timing); + bool optc1_validate_timing( struct timing_generator *optc, const struct dc_crtc_timing *timing); @@ -622,7 +625,8 @@ void optc1_set_drr( void optc1_set_static_screen_control( struct timing_generator *optc, - uint32_t value); + uint32_t event_triggers, + uint32_t num_frames); void optc1_program_stereo(struct timing_generator *optc, const struct dc_crtc_timing *timing, struct crtc_stereo_flags *flags); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index a12530a3ab9c..3b71898e859e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -28,6 +28,8 @@ #include "dm_services.h" #include "dc.h" +#include "dcn10_init.h" + #include "resource.h" #include "include/irq_service_interface.h" #include "dcn10_resource.h" @@ -270,7 +272,7 @@ static const struct dce_audio_shift audio_shift = { DCE120_AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { DCE120_AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -319,6 +321,14 @@ static const struct dcn10_link_enc_mask le_mask = { LINK_ENCODER_MASK_SH_LIST_DCN10(_MASK) }; +static const struct dce110_aux_registers_shift aux_shift = { + DCN10_AUX_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce110_aux_registers_mask aux_mask = { + DCN10_AUX_MASK_SH_LIST(_MASK) +}; + #define ipp_regs(id)\ [id] = {\ IPP_REG_LIST_DCN10(id),\ @@ -471,6 +481,28 @@ static const struct dcn_hubbub_mask hubbub_mask = { HUBBUB_MASK_SH_LIST_DCN10(_MASK) }; +static int map_transmitter_id_to_phy_instance( + enum transmitter transmitter) +{ + switch (transmitter) { + case TRANSMITTER_UNIPHY_A: + return 0; + break; + case TRANSMITTER_UNIPHY_B: + return 1; + break; + case TRANSMITTER_UNIPHY_C: + return 2; + break; + case TRANSMITTER_UNIPHY_D: + return 3; + break; + default: + ASSERT(0); + return 0; + } +} + #define clk_src_regs(index, pllid)\ [index] = {\ CS_COMMON_REG_LIST_DCN1_0(index, pllid),\ @@ -642,7 +674,10 @@ struct dce_aux *dcn10_aux_engine_create( dce110_aux_engine_construct(aux_engine, ctx, inst, SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD, - &aux_engine_regs[inst]); + &aux_engine_regs[inst], + &aux_mask, + &aux_shift, + ctx->dc->caps.extended_aux_timeout_support); return &aux_engine->base; } @@ -751,14 +786,18 @@ struct link_encoder *dcn10_link_encoder_create( { struct dcn10_link_encoder *enc10 = kzalloc(sizeof(struct dcn10_link_encoder), GFP_KERNEL); + int link_regs_id; if (!enc10) return NULL; + link_regs_id = + map_transmitter_id_to_phy_instance(enc_init_data->transmitter); + dcn10_link_encoder_construct(enc10, enc_init_data, &link_enc_feature, - &link_enc_regs[enc_init_data->transmitter], + &link_enc_regs[link_regs_id], &link_enc_aux_regs[enc_init_data->channel - 1], &link_enc_hpd_regs[enc_init_data->hpd_source], &le_shift, @@ -786,6 +825,7 @@ struct clock_source *dcn10_clock_source_create( return &clk_src->base; } + kfree(clk_src); BREAK_TO_DEBUGGER(); return NULL; } @@ -881,7 +921,7 @@ static struct pp_smu_funcs *dcn10_pp_smu_create(struct dc_context *ctx) return pp_smu; } -static void destruct(struct dcn10_resource_pool *pool) +static void dcn10_resource_destruct(struct dcn10_resource_pool *pool) { unsigned int i; @@ -1128,7 +1168,7 @@ static void dcn10_destroy_resource_pool(struct resource_pool **pool) { struct dcn10_resource_pool *dcn10_pool = TO_DCN10_RES_POOL(*pool); - destruct(dcn10_pool); + dcn10_resource_destruct(dcn10_pool); kfree(dcn10_pool); *pool = NULL; } @@ -1267,7 +1307,7 @@ static uint32_t read_pipe_fuses(struct dc_context *ctx) return value; } -static bool construct( +static bool dcn10_resource_construct( uint8_t num_virtual_links, struct dc *dc, struct dcn10_resource_pool *pool) @@ -1307,6 +1347,8 @@ static bool construct( dc->caps.max_slave_planes = 1; dc->caps.is_apu = true; dc->caps.post_blend_color_processing = false; + dc->caps.extended_aux_timeout_support = false; + /* Raven DP PHY HBR2 eye diagram pattern is not stable. Use TP4 */ dc->caps.force_dp_tps4_for_cp2520 = true; @@ -1416,6 +1458,14 @@ static bool construct( pool->base.pp_smu = dcn10_pp_smu_create(ctx); + /* + * Right now SMU/PPLIB and DAL all have the AZ D3 force PME notification * + * implemented. So AZ D3 should work.For issue 197007. * + */ + if (pool->base.pp_smu != NULL + && pool->base.pp_smu->rv_funcs.set_pme_wa_enable != NULL) + dc->debug.az_endpoint_mute_only = false; + if (!dc->debug.disable_pplib_clock_request) dcn_bw_update_from_pplib(dc); dcn_bw_sync_calcs_and_dml(dc); @@ -1544,7 +1594,7 @@ static bool construct( fail: - destruct(pool); + dcn10_resource_destruct(pool); return false; } @@ -1559,9 +1609,10 @@ struct resource_pool *dcn10_create_resource_pool( if (!pool) return NULL; - if (construct(init_data->num_virtual_links, dc, pool)) + if (dcn10_resource_construct(init_data->num_virtual_links, dc, pool)) return &pool->base; + kfree(pool); BREAK_TO_DEBUGGER(); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c index b9ffbf6b58ff..376c4264d295 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c @@ -247,6 +247,7 @@ void enc1_stream_encoder_dp_set_stream_attribute( struct stream_encoder *enc, struct dc_crtc_timing *crtc_timing, enum dc_color_space output_color_space, + bool use_vsc_sdp_for_colorimetry, uint32_t enable_sdp_splitting) { uint32_t h_active_start; @@ -312,10 +313,7 @@ void enc1_stream_encoder_dp_set_stream_attribute( * Pixel Encoding/Colorimetry Format and that a Sink device shall ignore MISC1, bit 7, * and MISC0, bits 7:1 (MISC1, bit 7, and MISC0, bits 7:1, become "don't care"). */ - if ((hw_crtc_timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) || - (output_color_space == COLOR_SPACE_2020_YCBCR) || - (output_color_space == COLOR_SPACE_2020_RGB_FULLRANGE) || - (output_color_space == COLOR_SPACE_2020_RGB_LIMITEDRANGE)) + if (use_vsc_sdp_for_colorimetry) misc1 = misc1 | 0x40; else misc1 = misc1 & ~0x40; @@ -512,11 +510,12 @@ void enc1_stream_encoder_hdmi_set_stream_attribute( enc1_stream_encoder_set_stream_attribute_helper(enc1, crtc_timing); /* setup HDMI engine */ - REG_UPDATE_5(HDMI_CONTROL, + REG_UPDATE_6(HDMI_CONTROL, HDMI_PACKET_GEN_VERSION, 1, HDMI_KEEPOUT_MODE, 1, HDMI_DEEP_COLOR_ENABLE, 0, HDMI_DATA_SCRAMBLE_EN, 0, + HDMI_NO_EXTRA_NULL_PACKET_FILLED, 1, HDMI_CLOCK_CHANNEL_RATE, 0); @@ -1003,6 +1002,19 @@ void enc1_stream_encoder_set_avmute( REG_UPDATE(HDMI_GC, HDMI_GC_AVMUTE, value); } +void enc1_reset_hdmi_stream_attribute( + struct stream_encoder *enc) +{ + struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); + + REG_UPDATE_5(HDMI_CONTROL, + HDMI_PACKET_GEN_VERSION, 1, + HDMI_KEEPOUT_MODE, 1, + HDMI_DEEP_COLOR_ENABLE, 0, + HDMI_DATA_SCRAMBLE_EN, 0, + HDMI_CLOCK_CHANNEL_RATE, 0); +} + #define DP_SEC_AUD_N__DP_SEC_AUD_N__DEFAULT 0x8000 #define DP_SEC_TIMESTAMP__DP_SEC_TIMESTAMP_MODE__AUTO_CALC 1 @@ -1196,13 +1208,13 @@ static union audio_cea_channels speakers_to_channels( void get_audio_clock_info( enum dc_color_depth color_depth, - uint32_t crtc_pixel_clock_in_khz, - uint32_t actual_pixel_clock_in_khz, + uint32_t crtc_pixel_clock_100Hz, + uint32_t actual_pixel_clock_100Hz, struct audio_clock_info *audio_clock_info) { const struct audio_clock_info *clock_info; uint32_t index; - uint32_t crtc_pixel_clock_in_10khz = crtc_pixel_clock_in_khz / 10; + uint32_t crtc_pixel_clock_in_10khz = crtc_pixel_clock_100Hz / 100; uint32_t audio_array_size; switch (color_depth) { @@ -1239,16 +1251,16 @@ void get_audio_clock_info( } /* not found */ - if (actual_pixel_clock_in_khz == 0) - actual_pixel_clock_in_khz = crtc_pixel_clock_in_khz; + if (actual_pixel_clock_100Hz == 0) + actual_pixel_clock_100Hz = crtc_pixel_clock_100Hz; /* See HDMI spec the table entry under * pixel clock of "Other". */ audio_clock_info->pixel_clock_in_10khz = - actual_pixel_clock_in_khz / 10; - audio_clock_info->cts_32khz = actual_pixel_clock_in_khz; - audio_clock_info->cts_44khz = actual_pixel_clock_in_khz; - audio_clock_info->cts_48khz = actual_pixel_clock_in_khz; + actual_pixel_clock_100Hz / 100; + audio_clock_info->cts_32khz = actual_pixel_clock_100Hz / 10; + audio_clock_info->cts_44khz = actual_pixel_clock_100Hz / 10; + audio_clock_info->cts_48khz = actual_pixel_clock_100Hz / 10; audio_clock_info->n_32khz = 4096; audio_clock_info->n_44khz = 6272; @@ -1308,14 +1320,14 @@ static void enc1_se_setup_hdmi_audio( /* Program audio clock sample/regeneration parameters */ get_audio_clock_info(crtc_info->color_depth, - crtc_info->requested_pixel_clock, - crtc_info->calculated_pixel_clock, + crtc_info->requested_pixel_clock_100Hz, + crtc_info->calculated_pixel_clock_100Hz, &audio_clock_info); DC_LOG_HW_AUDIO( - "\n%s:Input::requested_pixel_clock = %d" \ - "calculated_pixel_clock = %d \n", __func__, \ - crtc_info->requested_pixel_clock, \ - crtc_info->calculated_pixel_clock); + "\n%s:Input::requested_pixel_clock_100Hz = %d" \ + "calculated_pixel_clock_100Hz = %d \n", __func__, \ + crtc_info->requested_pixel_clock_100Hz, \ + crtc_info->calculated_pixel_clock_100Hz); /* HDMI_ACR_32_0__HDMI_ACR_CTS_32_MASK */ REG_UPDATE(HDMI_ACR_32_0, HDMI_ACR_CTS_32, audio_clock_info.cts_32khz); @@ -1528,6 +1540,77 @@ void enc1_dig_connect_to_otg( REG_UPDATE(DIG_FE_CNTL, DIG_SOURCE_SELECT, tg_inst); } +unsigned int enc1_dig_source_otg( + struct stream_encoder *enc) +{ + uint32_t tg_inst = 0; + struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); + + REG_GET(DIG_FE_CNTL, DIG_SOURCE_SELECT, &tg_inst); + + return tg_inst; +} + +bool enc1_stream_encoder_dp_get_pixel_format( + struct stream_encoder *enc, + enum dc_pixel_encoding *encoding, + enum dc_color_depth *depth) +{ + uint32_t hw_encoding = 0; + uint32_t hw_depth = 0; + struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); + + if (enc == NULL || + encoding == NULL || + depth == NULL) + return false; + + REG_GET_2(DP_PIXEL_FORMAT, + DP_PIXEL_ENCODING, &hw_encoding, + DP_COMPONENT_DEPTH, &hw_depth); + + switch (hw_depth) { + case DP_COMPONENT_PIXEL_DEPTH_6BPC: + *depth = COLOR_DEPTH_666; + break; + case DP_COMPONENT_PIXEL_DEPTH_8BPC: + *depth = COLOR_DEPTH_888; + break; + case DP_COMPONENT_PIXEL_DEPTH_10BPC: + *depth = COLOR_DEPTH_101010; + break; + case DP_COMPONENT_PIXEL_DEPTH_12BPC: + *depth = COLOR_DEPTH_121212; + break; + case DP_COMPONENT_PIXEL_DEPTH_16BPC: + *depth = COLOR_DEPTH_161616; + break; + default: + *depth = COLOR_DEPTH_UNDEFINED; + break; + } + + switch (hw_encoding) { + case DP_PIXEL_ENCODING_TYPE_RGB444: + *encoding = PIXEL_ENCODING_RGB; + break; + case DP_PIXEL_ENCODING_TYPE_YCBCR422: + *encoding = PIXEL_ENCODING_YCBCR422; + break; + case DP_PIXEL_ENCODING_TYPE_YCBCR444: + case DP_PIXEL_ENCODING_TYPE_Y_ONLY: + *encoding = PIXEL_ENCODING_YCBCR444; + break; + case DP_PIXEL_ENCODING_TYPE_YCBCR420: + *encoding = PIXEL_ENCODING_YCBCR420; + break; + default: + *encoding = PIXEL_ENCODING_UNDEFINED; + break; + } + return true; +} + static const struct stream_encoder_funcs dcn10_str_enc_funcs = { .dp_set_stream_attribute = enc1_stream_encoder_dp_set_stream_attribute, @@ -1562,6 +1645,10 @@ static const struct stream_encoder_funcs dcn10_str_enc_funcs = { .setup_stereo_sync = enc1_setup_stereo_sync, .set_avmute = enc1_stream_encoder_set_avmute, .dig_connect_to_otg = enc1_dig_connect_to_otg, + .hdmi_reset_stream_attribute = enc1_reset_hdmi_stream_attribute, + .dig_source_otg = enc1_dig_source_otg, + + .dp_get_pixel_format = enc1_stream_encoder_dp_get_pixel_format, }; void dcn10_stream_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h index bc2b4af9543b..f9b9e221c698 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.h @@ -89,7 +89,8 @@ SRI(DP_VID_STREAM_CNTL, DP, id), \ SRI(DP_VID_TIMING, DP, id), \ SRI(DP_SEC_AUD_N, DP, id), \ - SRI(DP_SEC_TIMESTAMP, DP, id) + SRI(DP_SEC_TIMESTAMP, DP, id), \ + SRI(DIG_CLOCK_PATTERN, DIG, id) #define SE_DCN_REG_LIST(id)\ SE_COMMON_DCN_REG_LIST(id) @@ -162,14 +163,13 @@ struct dcn10_stream_enc_registers { uint32_t DP_MSA_TIMING_PARAM3; uint32_t DP_MSA_TIMING_PARAM4; uint32_t HDMI_DB_CONTROL; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) uint32_t DP_DSC_CNTL; uint32_t DP_DSC_BYTES_PER_PIXEL; uint32_t DME_CONTROL; uint32_t DP_SEC_METADATA_TRANSMISSION; uint32_t HDMI_METADATA_PACKET_CONTROL; uint32_t DP_SEC_FRAMING4; -#endif + uint32_t DIG_CLOCK_PATTERN; }; @@ -189,6 +189,7 @@ struct dcn10_stream_enc_registers { SE_SF(DIG0_HDMI_CONTROL, HDMI_DEEP_COLOR_ENABLE, mask_sh),\ SE_SF(DIG0_HDMI_CONTROL, HDMI_DEEP_COLOR_DEPTH, mask_sh),\ SE_SF(DIG0_HDMI_CONTROL, HDMI_DATA_SCRAMBLE_EN, mask_sh),\ + SE_SF(DIG0_HDMI_CONTROL, HDMI_NO_EXTRA_NULL_PACKET_FILLED, mask_sh),\ SE_SF(DIG0_HDMI_VBI_PACKET_CONTROL, HDMI_GC_CONT, mask_sh),\ SE_SF(DIG0_HDMI_VBI_PACKET_CONTROL, HDMI_GC_SEND, mask_sh),\ SE_SF(DIG0_HDMI_VBI_PACKET_CONTROL, HDMI_NULL_SEND, mask_sh),\ @@ -297,7 +298,8 @@ struct dcn10_stream_enc_registers { SE_SF(DP0_DP_MSA_TIMING_PARAM4, DP_MSA_VHEIGHT, mask_sh),\ SE_SF(DIG0_HDMI_DB_CONTROL, HDMI_DB_DISABLE, mask_sh),\ SE_SF(DP0_DP_VID_TIMING, DP_VID_N_MUL, mask_sh),\ - SE_SF(DIG0_DIG_FE_CNTL, DIG_SOURCE_SELECT, mask_sh) + SE_SF(DIG0_DIG_FE_CNTL, DIG_SOURCE_SELECT, mask_sh),\ + SE_SF(DIG0_DIG_CLOCK_PATTERN, DIG_CLOCK_PATTERN, mask_sh) #define SE_COMMON_MASK_SH_LIST_SOC(mask_sh)\ SE_COMMON_MASK_SH_LIST_SOC_BASE(mask_sh) @@ -374,6 +376,7 @@ struct dcn10_stream_enc_registers { type HDMI_GC_SEND;\ type HDMI_NULL_SEND;\ type HDMI_DATA_SCRAMBLE_EN;\ + type HDMI_NO_EXTRA_NULL_PACKET_FILLED;\ type HDMI_AUDIO_INFO_SEND;\ type AFMT_AUDIO_INFO_UPDATE;\ type HDMI_AUDIO_INFO_LINE;\ @@ -458,9 +461,9 @@ struct dcn10_stream_enc_registers { type HDMI_DB_DISABLE;\ type DP_VID_N_MUL;\ type DP_VID_M_DOUBLE_VALUE_EN;\ - type DIG_SOURCE_SELECT + type DIG_SOURCE_SELECT;\ + type DIG_CLOCK_PATTERN -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define SE_REG_FIELD_LIST_DCN2_0(type) \ type DP_DSC_MODE;\ type DP_DSC_SLICE_WIDTH;\ @@ -479,20 +482,15 @@ struct dcn10_stream_enc_registers { type DOLBY_VISION_EN;\ type DP_PIXEL_COMBINE;\ type DP_SST_SDP_SPLITTING -#endif struct dcn10_stream_encoder_shift { SE_REG_FIELD_LIST_DCN1_0(uint8_t); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) SE_REG_FIELD_LIST_DCN2_0(uint8_t); -#endif }; struct dcn10_stream_encoder_mask { SE_REG_FIELD_LIST_DCN1_0(uint32_t); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) SE_REG_FIELD_LIST_DCN2_0(uint32_t); -#endif }; struct dcn10_stream_encoder { @@ -520,6 +518,7 @@ void enc1_stream_encoder_dp_set_stream_attribute( struct stream_encoder *enc, struct dc_crtc_timing *crtc_timing, enum dc_color_space output_color_space, + bool use_vsc_sdp_for_colorimetry, uint32_t enable_sdp_splitting); void enc1_stream_encoder_hdmi_set_stream_attribute( @@ -592,6 +591,9 @@ void enc1_dig_connect_to_otg( struct stream_encoder *enc, int tg_inst); +unsigned int enc1_dig_source_otg( + struct stream_encoder *enc); + void enc1_stream_encoder_set_stream_attribute_helper( struct dcn10_stream_encoder *enc1, struct dc_crtc_timing *crtc_timing); @@ -605,8 +607,16 @@ void enc1_se_enable_dp_audio( void get_audio_clock_info( enum dc_color_depth color_depth, - uint32_t crtc_pixel_clock_in_khz, - uint32_t actual_pixel_clock_in_khz, + uint32_t crtc_pixel_clock_100Hz, + uint32_t actual_pixel_clock_100Hz, struct audio_clock_info *audio_clock_info); +void enc1_reset_hdmi_stream_attribute( + struct stream_encoder *enc); + +bool enc1_stream_encoder_dp_get_pixel_format( + struct stream_encoder *enc, + enum dc_pixel_encoding *encoding, + enum dc_color_depth *depth); + #endif /* __DC_STREAM_ENCODER_DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile index e9721a906592..5fcaf78334ff 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile @@ -1,22 +1,38 @@ +# SPDX-License-Identifier: MIT # # Makefile for DCN. -DCN20 = dcn20_resource.o dcn20_hwseq.o dcn20_dpp.o dcn20_dpp_cm.o dcn20_hubp.o \ +DCN20 = dcn20_resource.o dcn20_init.o dcn20_hwseq.o dcn20_dpp.o dcn20_dpp_cm.o dcn20_hubp.o \ dcn20_mpc.o dcn20_opp.o dcn20_hubbub.o dcn20_optc.o dcn20_mmhubbub.o \ dcn20_stream_encoder.o dcn20_link_encoder.o dcn20_dccg.o \ dcn20_vmid.o dcn20_dwb.o dcn20_dwb_scl.o -ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT DCN20 += dcn20_dsc.o + +ifdef CONFIG_X86 +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -msse +endif + +ifdef CONFIG_PPC64 +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o := -mhard-float -maltivec endif -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),) - cc_stack_align := -mpreferred-stack-boundary=4 -else ifneq ($(call cc-option, -mstack-alignment=16),) - cc_stack_align := -mstack-alignment=16 +ifdef CONFIG_CC_IS_GCC +ifeq ($(call cc-ifversion, -lt, 0701, y), y) +IS_OLD_GCC = 1 +endif endif -CFLAGS_dcn20_resource.o := -mhard-float -msse $(cc_stack_align) +ifdef CONFIG_X86 +ifdef IS_OLD_GCC +# Stack alignment mismatch, proceed with caution. +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 +# (8B stack alignment). +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -mpreferred-stack-boundary=4 +else +CFLAGS_$(AMDDALPATH)/dc/dcn20/dcn20_resource.o += -msse2 +endif +endif AMD_DAL_DCN20 = $(addprefix $(AMDDALPATH)/dc/dcn20/,$(DCN20)) diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c index 31aa6ee5cd5b..50bffbfdd394 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.c @@ -50,20 +50,20 @@ void dccg2_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk) if (dccg->ref_dppclk && req_dppclk) { int ref_dppclk = dccg->ref_dppclk; + int modulo, phase; - ASSERT(req_dppclk <= ref_dppclk); - /* need to clamp to 8 bits */ - if (ref_dppclk > 0xff) { - int divider = (ref_dppclk + 0xfe) / 0xff; + // phase / modulo = dpp pipe clk / dpp global clk + modulo = 0xff; // use FF at the end + phase = ((modulo * req_dppclk) + ref_dppclk - 1) / ref_dppclk; - ref_dppclk /= divider; - req_dppclk = (req_dppclk + divider - 1) / divider; - if (req_dppclk > ref_dppclk) - req_dppclk = ref_dppclk; + if (phase > 0xff) { + ASSERT(false); + phase = 0xff; } + REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0, - DPPCLK0_DTO_PHASE, req_dppclk, - DPPCLK0_DTO_MODULO, ref_dppclk); + DPPCLK0_DTO_PHASE, phase, + DPPCLK0_DTO_MODULO, modulo); REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_ENABLE[dpp_inst], 1); } else { @@ -96,32 +96,6 @@ void dccg2_get_dccg_ref_freq(struct dccg *dccg, void dccg2_init(struct dccg *dccg) { - struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); - - // Fallthrough intentional to program all available dpp_dto's - switch (dccg_dcn->base.ctx->dc->res_pool->pipe_count) { - case 6: - REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_DB_EN[5], 1); - /* Fall through */ - case 5: - REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_DB_EN[4], 1); - /* Fall through */ - case 4: - REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_DB_EN[3], 1); - /* Fall through */ - case 3: - REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_DB_EN[2], 1); - /* Fall through */ - case 2: - REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_DB_EN[1], 1); - /* Fall through */ - case 1: - REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_DB_EN[0], 1); - break; - default: - ASSERT(false); - break; - } } static const struct dccg_funcs dccg2_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c index 9bc5dd23d297..13e057d7ee93 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c @@ -72,6 +72,21 @@ void dpp20_read_state(struct dpp *dpp_base, } } +void dpp2_power_on_obuf( + struct dpp *dpp_base, + bool power_on) +{ + struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); + + REG_UPDATE(CM_MEM_PWR_CTRL, SHARED_MEM_PWR_DIS, power_on == true ? 1:0); + + REG_UPDATE(OBUF_MEM_PWR_CTRL, + OBUF_MEM_PWR_FORCE, power_on == true ? 0:1); + + REG_UPDATE(DSCL_MEM_PWR_CTRL, + LUT_MEM_PWR_FORCE, power_on == true ? 0:1); +} + void dpp2_dummy_program_input_lut( struct dpp *dpp_base, const struct dc_gamma *gamma) @@ -89,7 +104,7 @@ static void dpp2_cnv_setup ( uint32_t pixel_format = 0; uint32_t alpha_en = 1; enum dc_color_space color_space = COLOR_SPACE_SRGB; - enum dcn10_input_csc_select select = INPUT_CSC_SELECT_BYPASS; + enum dcn20_input_csc_select select = DCN2_ICSC_SELECT_BYPASS; bool force_disable_cursor = false; struct out_csc_color_matrix tbl_entry; uint32_t is_2bit = 0; @@ -130,25 +145,25 @@ static void dpp2_cnv_setup ( force_disable_cursor = false; pixel_format = 65; color_space = COLOR_SPACE_YCBCR709; - select = INPUT_CSC_SELECT_ICSC; + select = DCN2_ICSC_SELECT_ICSC_A; break; case SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb: force_disable_cursor = true; pixel_format = 64; color_space = COLOR_SPACE_YCBCR709; - select = INPUT_CSC_SELECT_ICSC; + select = DCN2_ICSC_SELECT_ICSC_A; break; case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCbCr: force_disable_cursor = true; pixel_format = 67; color_space = COLOR_SPACE_YCBCR709; - select = INPUT_CSC_SELECT_ICSC; + select = DCN2_ICSC_SELECT_ICSC_A; break; case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb: force_disable_cursor = true; pixel_format = 66; color_space = COLOR_SPACE_YCBCR709; - select = INPUT_CSC_SELECT_ICSC; + select = DCN2_ICSC_SELECT_ICSC_A; break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: pixel_format = 22; @@ -162,7 +177,7 @@ static void dpp2_cnv_setup ( case SURFACE_PIXEL_FORMAT_VIDEO_AYCrCb8888: pixel_format = 12; color_space = COLOR_SPACE_YCBCR709; - select = INPUT_CSC_SELECT_ICSC; + select = DCN2_ICSC_SELECT_ICSC_A; break; case SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FIX: pixel_format = 112; @@ -173,13 +188,13 @@ static void dpp2_cnv_setup ( case SURFACE_PIXEL_FORMAT_VIDEO_ACrYCb2101010: pixel_format = 114; color_space = COLOR_SPACE_YCBCR709; - select = INPUT_CSC_SELECT_ICSC; + select = DCN2_ICSC_SELECT_ICSC_A; is_2bit = 1; break; case SURFACE_PIXEL_FORMAT_VIDEO_CrYCbA1010102: pixel_format = 115; color_space = COLOR_SPACE_YCBCR709; - select = INPUT_CSC_SELECT_ICSC; + select = DCN2_ICSC_SELECT_ICSC_A; is_2bit = 1; break; case SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FLOAT: @@ -212,13 +227,13 @@ static void dpp2_cnv_setup ( tbl_entry.color_space = input_color_space; if (color_space >= COLOR_SPACE_YCBCR601) - select = INPUT_CSC_SELECT_ICSC; + select = DCN2_ICSC_SELECT_ICSC_A; else - select = INPUT_CSC_SELECT_BYPASS; + select = DCN2_ICSC_SELECT_BYPASS; - dpp1_program_input_csc(dpp_base, color_space, select, &tbl_entry); + dpp2_program_input_csc(dpp_base, color_space, select, &tbl_entry); } else - dpp1_program_input_csc(dpp_base, color_space, select, NULL); + dpp2_program_input_csc(dpp_base, color_space, select, NULL); if (force_disable_cursor) { REG_UPDATE(CURSOR_CONTROL, @@ -227,6 +242,7 @@ static void dpp2_cnv_setup ( CUR0_ENABLE, 0); } + dpp2_power_on_obuf(dpp_base, true); } @@ -326,14 +342,18 @@ void dpp2_cnv_set_alpha_keyer( void dpp2_set_cursor_attributes( struct dpp *dpp_base, - enum dc_cursor_color_format color_format) + struct dc_cursor_attributes *cursor_attributes) { + enum dc_cursor_color_format color_format = cursor_attributes->color_format; struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); int cur_rom_en = 0; if (color_format == CURSOR_MODE_COLOR_PRE_MULTIPLIED_ALPHA || - color_format == CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA) - cur_rom_en = 1; + color_format == CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA) { + if (cursor_attributes->attribute_flags.bits.ENABLE_CURSOR_DEGAMMA) { + cur_rom_en = 1; + } + } REG_UPDATE_3(CURSOR0_CONTROL, CUR0_MODE, color_format, @@ -356,13 +376,6 @@ bool dpp2_get_optimal_number_of_taps( struct scaler_data *scl_data, const struct scaling_taps *in_taps) { - uint32_t pixel_width; - - if (scl_data->viewport.width > scl_data->recout.width) - pixel_width = scl_data->recout.width; - else - pixel_width = scl_data->viewport.width; - /* Some ASICs does not support FP16 scaling, so we reject modes require this*/ if (scl_data->viewport.width != scl_data->h_active && scl_data->viewport.height != scl_data->v_active && @@ -444,8 +457,8 @@ static struct dpp_funcs dcn20_dpp_funcs = { .dpp_read_state = dpp20_read_state, .dpp_reset = dpp_reset, .dpp_set_scaler = dpp1_dscl_set_scaler_manual_scale, - .dpp_get_optimal_number_of_taps = dpp2_get_optimal_number_of_taps, - .dpp_set_gamut_remap = dpp1_cm_set_gamut_remap, + .dpp_get_optimal_number_of_taps = dpp1_get_optimal_number_of_taps, + .dpp_set_gamut_remap = dpp2_cm_set_gamut_remap, .dpp_set_csc_adjustment = NULL, .dpp_set_csc_default = NULL, .dpp_program_regamma_pwl = oppn20_dummy_program_regamma_pwl, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h index 59b67ed57c19..27610251c57f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h @@ -30,16 +30,20 @@ #define TO_DCN20_DPP(dpp)\ container_of(dpp, struct dcn20_dpp, base) -#define TF_REG_LIST_DCN20(id) \ - TF_REG_LIST_DCN(id), \ +#define TF_REG_LIST_DCN20_COMMON_UPDATED(id) \ SRI(CM_BLNDGAM_LUT_WRITE_EN_MASK, CM, id), \ + SRI(CM_BLNDGAM_RAMB_SLOPE_CNTL_B, CM, id), \ + SRI(CM_BLNDGAM_RAMB_SLOPE_CNTL_G, CM, id), \ + SRI(CM_BLNDGAM_RAMB_SLOPE_CNTL_R, CM, id), \ + SRI(CM_BLNDGAM_RAMA_SLOPE_CNTL_B, CM, id), \ + SRI(CM_BLNDGAM_RAMA_SLOPE_CNTL_G, CM, id), \ + SRI(CM_BLNDGAM_RAMA_SLOPE_CNTL_R, CM, id) + +#define TF_REG_LIST_DCN20_COMMON(id) \ SRI(CM_BLNDGAM_CONTROL, CM, id), \ SRI(CM_BLNDGAM_RAMB_START_CNTL_B, CM, id), \ SRI(CM_BLNDGAM_RAMB_START_CNTL_G, CM, id), \ SRI(CM_BLNDGAM_RAMB_START_CNTL_R, CM, id), \ - SRI(CM_BLNDGAM_RAMB_SLOPE_CNTL_B, CM, id), \ - SRI(CM_BLNDGAM_RAMB_SLOPE_CNTL_G, CM, id), \ - SRI(CM_BLNDGAM_RAMB_SLOPE_CNTL_R, CM, id), \ SRI(CM_BLNDGAM_RAMB_END_CNTL1_B, CM, id), \ SRI(CM_BLNDGAM_RAMB_END_CNTL2_B, CM, id), \ SRI(CM_BLNDGAM_RAMB_END_CNTL1_G, CM, id), \ @@ -66,9 +70,6 @@ SRI(CM_BLNDGAM_RAMA_START_CNTL_B, CM, id), \ SRI(CM_BLNDGAM_RAMA_START_CNTL_G, CM, id), \ SRI(CM_BLNDGAM_RAMA_START_CNTL_R, CM, id), \ - SRI(CM_BLNDGAM_RAMA_SLOPE_CNTL_B, CM, id), \ - SRI(CM_BLNDGAM_RAMA_SLOPE_CNTL_G, CM, id), \ - SRI(CM_BLNDGAM_RAMA_SLOPE_CNTL_R, CM, id), \ SRI(CM_BLNDGAM_RAMA_END_CNTL1_B, CM, id), \ SRI(CM_BLNDGAM_RAMA_END_CNTL2_B, CM, id), \ SRI(CM_BLNDGAM_RAMA_END_CNTL1_G, CM, id), \ @@ -147,7 +148,22 @@ SRI(CM_SHAPER_RAMA_REGION_28_29, CM, id), \ SRI(CM_SHAPER_RAMA_REGION_30_31, CM, id), \ SRI(CM_SHAPER_RAMA_REGION_32_33, CM, id), \ - SRI(CM_SHAPER_LUT_INDEX, CM, id), \ + SRI(CM_SHAPER_LUT_INDEX, CM, id) + +#define TF_REG_LIST_DCN20_COMMON_APPEND(id) \ + SRI(CM_GAMUT_REMAP_B_C11_C12, CM, id),\ + SRI(CM_GAMUT_REMAP_B_C13_C14, CM, id),\ + SRI(CM_GAMUT_REMAP_B_C21_C22, CM, id),\ + SRI(CM_GAMUT_REMAP_B_C23_C24, CM, id),\ + SRI(CM_GAMUT_REMAP_B_C31_C32, CM, id),\ + SRI(CM_GAMUT_REMAP_B_C33_C34, CM, id),\ + SRI(CM_ICSC_B_C11_C12, CM, id), \ + SRI(CM_ICSC_B_C33_C34, CM, id) + +#define TF_REG_LIST_DCN20(id) \ + TF_REG_LIST_DCN(id), \ + TF_REG_LIST_DCN20_COMMON(id), \ + TF_REG_LIST_DCN20_COMMON_UPDATED(id), \ SRI(CURSOR_CONTROL, CURSOR0_, id), \ SRI(ALPHA_2BIT_LUT, CNVC_CFG, id), \ SRI(FCNV_FP_BIAS_R, CNVC_CFG, id), \ @@ -162,29 +178,45 @@ SRI(COLOR_KEYER_GREEN, CNVC_CFG, id), \ SRI(COLOR_KEYER_BLUE, CNVC_CFG, id), \ SRI(CM_SHAPER_LUT_DATA, CM, id), \ - SRI(CURSOR_CONTROL, CURSOR0_, id) + SRI(CURSOR_CONTROL, CURSOR0_, id),\ + SRI(OBUF_MEM_PWR_CTRL, DSCL, id),\ + SRI(DSCL_MEM_PWR_CTRL, DSCL, id) -#define TF_REG_LIST_SH_MASK_DCN20(mask_sh)\ - TF_REG_LIST_SH_MASK_DCN(mask_sh), \ + +#define TF_REG_LIST_SH_MASK_DCN20_UPDATED(mask_sh)\ + TF_SF(CM0_CM_BLNDGAM_RAMB_SLOPE_CNTL_B, CM_BLNDGAM_RAMB_EXP_REGION_LINEAR_SLOPE_B, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMB_SLOPE_CNTL_G, CM_BLNDGAM_RAMB_EXP_REGION_LINEAR_SLOPE_G, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMB_SLOPE_CNTL_R, CM_BLNDGAM_RAMB_EXP_REGION_LINEAR_SLOPE_R, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL1_B, CM_BLNDGAM_RAMB_EXP_REGION_END_B, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL1_G, CM_BLNDGAM_RAMB_EXP_REGION_END_G, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL1_R, CM_BLNDGAM_RAMB_EXP_REGION_END_R, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMA_SLOPE_CNTL_B, CM_BLNDGAM_RAMA_EXP_REGION_LINEAR_SLOPE_B, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMA_SLOPE_CNTL_G, CM_BLNDGAM_RAMA_EXP_REGION_LINEAR_SLOPE_G, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMA_SLOPE_CNTL_R, CM_BLNDGAM_RAMA_EXP_REGION_LINEAR_SLOPE_R, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL1_B, CM_BLNDGAM_RAMA_EXP_REGION_END_B, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL1_G, CM_BLNDGAM_RAMA_EXP_REGION_END_G, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL1_R, CM_BLNDGAM_RAMA_EXP_REGION_END_R, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL2_B, CM_BLNDGAM_RAMA_EXP_REGION_END_BASE_B, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL2_G, CM_BLNDGAM_RAMA_EXP_REGION_END_BASE_G, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL2_R, CM_BLNDGAM_RAMA_EXP_REGION_END_BASE_R, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_CONTROL, CM_BLNDGAM_LUT_MODE, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_LUT_WRITE_EN_MASK, CM_BLNDGAM_LUT_WRITE_EN_MASK, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_LUT_WRITE_EN_MASK, CM_BLNDGAM_LUT_WRITE_SEL, mask_sh), \ + TF_SF(CM0_CM_BLNDGAM_LUT_WRITE_EN_MASK, CM_BLNDGAM_CONFIG_STATUS, mask_sh), \ + TF_SF(CM0_CM_SHAPER_CONTROL, CM_SHAPER_LUT_MODE, mask_sh) + + +#define TF_REG_LIST_SH_MASK_DCN20_COMMON(mask_sh)\ + TF_SF(CM0_CM_3DLUT_MODE, CM_3DLUT_MODE, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_START_CNTL_B, CM_BLNDGAM_RAMB_EXP_REGION_START_B, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_START_CNTL_B, CM_BLNDGAM_RAMB_EXP_REGION_START_SEGMENT_B, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_START_CNTL_G, CM_BLNDGAM_RAMB_EXP_REGION_START_G, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_START_CNTL_G, CM_BLNDGAM_RAMB_EXP_REGION_START_SEGMENT_G, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_START_CNTL_R, CM_BLNDGAM_RAMB_EXP_REGION_START_R, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_START_CNTL_R, CM_BLNDGAM_RAMB_EXP_REGION_START_SEGMENT_R, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMB_SLOPE_CNTL_B, CM_BLNDGAM_RAMB_EXP_REGION_LINEAR_SLOPE_B, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMB_SLOPE_CNTL_G, CM_BLNDGAM_RAMB_EXP_REGION_LINEAR_SLOPE_G, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMB_SLOPE_CNTL_R, CM_BLNDGAM_RAMB_EXP_REGION_LINEAR_SLOPE_R, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL1_B, CM_BLNDGAM_RAMB_EXP_REGION_END_B, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL2_B, CM_BLNDGAM_RAMB_EXP_REGION_END_SLOPE_B, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL2_B, CM_BLNDGAM_RAMB_EXP_REGION_END_BASE_B, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL1_G, CM_BLNDGAM_RAMB_EXP_REGION_END_G, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL2_G, CM_BLNDGAM_RAMB_EXP_REGION_END_SLOPE_G, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL2_G, CM_BLNDGAM_RAMB_EXP_REGION_END_BASE_G, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL1_R, CM_BLNDGAM_RAMB_EXP_REGION_END_R, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL2_R, CM_BLNDGAM_RAMB_EXP_REGION_END_SLOPE_R, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMB_END_CNTL2_R, CM_BLNDGAM_RAMB_EXP_REGION_END_BASE_R, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_REGION_0_1, CM_BLNDGAM_RAMB_EXP_REGION0_LUT_OFFSET, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_REGION_0_1, CM_BLNDGAM_RAMB_EXP_REGION0_NUM_SEGMENTS, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMB_REGION_0_1, CM_BLNDGAM_RAMB_EXP_REGION1_LUT_OFFSET, mask_sh), \ @@ -259,18 +291,9 @@ TF_SF(CM0_CM_BLNDGAM_RAMA_START_CNTL_G, CM_BLNDGAM_RAMA_EXP_REGION_START_SEGMENT_G, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_START_CNTL_R, CM_BLNDGAM_RAMA_EXP_REGION_START_R, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_START_CNTL_R, CM_BLNDGAM_RAMA_EXP_REGION_START_SEGMENT_R, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMA_SLOPE_CNTL_B, CM_BLNDGAM_RAMA_EXP_REGION_LINEAR_SLOPE_B, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMA_SLOPE_CNTL_G, CM_BLNDGAM_RAMA_EXP_REGION_LINEAR_SLOPE_G, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMA_SLOPE_CNTL_R, CM_BLNDGAM_RAMA_EXP_REGION_LINEAR_SLOPE_R, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL1_B, CM_BLNDGAM_RAMA_EXP_REGION_END_B, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL2_B, CM_BLNDGAM_RAMA_EXP_REGION_END_SLOPE_B, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL2_B, CM_BLNDGAM_RAMA_EXP_REGION_END_BASE_B, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL1_G, CM_BLNDGAM_RAMA_EXP_REGION_END_G, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL2_G, CM_BLNDGAM_RAMA_EXP_REGION_END_SLOPE_G, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL2_G, CM_BLNDGAM_RAMA_EXP_REGION_END_BASE_G, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL1_R, CM_BLNDGAM_RAMA_EXP_REGION_END_R, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL2_R, CM_BLNDGAM_RAMA_EXP_REGION_END_SLOPE_R, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_RAMA_END_CNTL2_R, CM_BLNDGAM_RAMA_EXP_REGION_END_BASE_R, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_REGION_0_1, CM_BLNDGAM_RAMA_EXP_REGION0_LUT_OFFSET, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_REGION_0_1, CM_BLNDGAM_RAMA_EXP_REGION0_NUM_SEGMENTS, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_REGION_0_1, CM_BLNDGAM_RAMA_EXP_REGION1_LUT_OFFSET, mask_sh), \ @@ -339,9 +362,6 @@ TF_SF(CM0_CM_BLNDGAM_RAMA_REGION_32_33, CM_BLNDGAM_RAMA_EXP_REGION32_NUM_SEGMENTS, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_REGION_32_33, CM_BLNDGAM_RAMA_EXP_REGION33_LUT_OFFSET, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_RAMA_REGION_32_33, CM_BLNDGAM_RAMA_EXP_REGION33_NUM_SEGMENTS, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_LUT_WRITE_EN_MASK, CM_BLNDGAM_LUT_WRITE_EN_MASK, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_LUT_WRITE_EN_MASK, CM_BLNDGAM_LUT_WRITE_SEL, mask_sh), \ - TF_SF(CM0_CM_BLNDGAM_LUT_WRITE_EN_MASK, CM_BLNDGAM_CONFIG_STATUS, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_LUT_INDEX, CM_BLNDGAM_LUT_INDEX, mask_sh), \ TF_SF(CM0_CM_BLNDGAM_LUT_DATA, CM_BLNDGAM_LUT_DATA, mask_sh), \ TF_SF(CM0_CM_MEM_PWR_CTRL, BLNDGAM_MEM_PWR_FORCE, mask_sh), \ @@ -354,7 +374,6 @@ TF_SF(CM0_CM_3DLUT_READ_WRITE_CONTROL, CM_3DLUT_WRITE_EN_MASK, mask_sh), \ TF_SF(CM0_CM_3DLUT_READ_WRITE_CONTROL, CM_3DLUT_RAM_SEL, mask_sh), \ TF_SF(CM0_CM_3DLUT_READ_WRITE_CONTROL, CM_3DLUT_30BIT_EN, mask_sh), \ - TF_SF(CM0_CM_3DLUT_READ_WRITE_CONTROL, CM_3DLUT_CONFIG_STATUS, mask_sh), \ TF_SF(CM0_CM_3DLUT_READ_WRITE_CONTROL, CM_3DLUT_READ_SEL, mask_sh), \ TF_SF(CM0_CM_SHAPER_CONTROL, CM_SHAPER_LUT_MODE, mask_sh), \ TF_SF(CM0_CM_SHAPER_RAMB_START_CNTL_B, CM_SHAPER_RAMB_EXP_REGION_START_B, mask_sh), \ @@ -519,9 +538,14 @@ TF_SF(CM0_CM_SHAPER_RAMA_REGION_32_33, CM_SHAPER_RAMA_EXP_REGION33_NUM_SEGMENTS, mask_sh), \ TF_SF(CM0_CM_SHAPER_LUT_WRITE_EN_MASK, CM_SHAPER_LUT_WRITE_EN_MASK, mask_sh), \ TF_SF(CM0_CM_SHAPER_LUT_WRITE_EN_MASK, CM_SHAPER_LUT_WRITE_SEL, mask_sh), \ - TF_SF(CM0_CM_SHAPER_LUT_WRITE_EN_MASK, CM_SHAPER_CONFIG_STATUS, mask_sh), \ TF_SF(CM0_CM_SHAPER_LUT_INDEX, CM_SHAPER_LUT_INDEX, mask_sh), \ - TF_SF(CM0_CM_SHAPER_LUT_DATA, CM_SHAPER_LUT_DATA, mask_sh), \ + TF_SF(CM0_CM_SHAPER_LUT_DATA, CM_SHAPER_LUT_DATA, mask_sh) + + +#define TF_REG_LIST_SH_MASK_DCN20(mask_sh)\ + TF_REG_LIST_SH_MASK_DCN(mask_sh), \ + TF_REG_LIST_SH_MASK_DCN20_COMMON(mask_sh), \ + TF_REG_LIST_SH_MASK_DCN20_UPDATED(mask_sh), \ TF_SF(CM0_CM_DGAM_LUT_WRITE_EN_MASK, CM_DGAM_CONFIG_STATUS, mask_sh), \ TF_SF(CM0_CM_CONTROL, CM_BYPASS, mask_sh), \ TF_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_MODE, mask_sh), \ @@ -554,11 +578,33 @@ TF_SF(CNVC_CFG0_COLOR_KEYER_BLUE, COLOR_KEYER_BLUE_HIGH, mask_sh), \ TF_SF(CNVC_CUR0_CURSOR0_CONTROL, CUR0_PIX_INV_MODE, mask_sh), \ TF_SF(CNVC_CUR0_CURSOR0_CONTROL, CUR0_PIXEL_ALPHA_MOD_EN, mask_sh), \ - TF_SF(CNVC_CUR0_CURSOR0_CONTROL, CUR0_ROM_EN, mask_sh) + TF_SF(CNVC_CUR0_CURSOR0_CONTROL, CUR0_ROM_EN, mask_sh),\ + TF_SF(DSCL0_OBUF_MEM_PWR_CTRL, OBUF_MEM_PWR_FORCE, mask_sh),\ + TF_SF(DSCL0_DSCL_MEM_PWR_CTRL, LUT_MEM_PWR_FORCE, mask_sh) + +/* DPP CM debug status register: + * + * Status index including current ICSC, Gamut Remap Mode is 9 + * ICSC Mode: [4..3] + * Gamut Remap Mode: [10..9] + */ +#define CM_TEST_DEBUG_DATA_STATUS_IDX 9 + +#define TF_DEBUG_REG_LIST_SH_DCN20 \ + TF_DEBUG_REG_LIST_SH_DCN10, \ + .CM_TEST_DEBUG_DATA_ICSC_MODE = 3, \ + .CM_TEST_DEBUG_DATA_GAMUT_REMAP_MODE = 9 + +#define TF_DEBUG_REG_LIST_MASK_DCN20 \ + TF_DEBUG_REG_LIST_MASK_DCN10, \ + .CM_TEST_DEBUG_DATA_ICSC_MODE = 0x18, \ + .CM_TEST_DEBUG_DATA_GAMUT_REMAP_MODE = 0x600 #define TF_REG_FIELD_LIST_DCN2_0(type) \ TF_REG_FIELD_LIST(type) \ type CM_BLNDGAM_LUT_DATA; \ + type CM_TEST_DEBUG_DATA_ICSC_MODE; \ + type CM_TEST_DEBUG_DATA_GAMUT_REMAP_MODE; \ type FORMAT_CNV16; \ type CNVC_BYPASS_MSB_ALIGN; \ type CLAMP_POSITIVE; \ @@ -585,7 +631,10 @@ type COLOR_KEYER_BLUE_HIGH; \ type CUR0_PIX_INV_MODE; \ type CUR0_PIXEL_ALPHA_MOD_EN; \ - type CUR0_ROM_EN + type CUR0_ROM_EN;\ + type OBUF_MEM_PWR_FORCE;\ + type LUT_MEM_PWR_FORCE + struct dcn2_dpp_shift { TF_REG_FIELD_LIST_DCN2_0(uint8_t); @@ -609,10 +658,23 @@ struct dcn2_dpp_mask { uint32_t COLOR_KEYER_ALPHA; \ uint32_t COLOR_KEYER_RED; \ uint32_t COLOR_KEYER_GREEN; \ - uint32_t COLOR_KEYER_BLUE + uint32_t COLOR_KEYER_BLUE; \ + uint32_t OBUF_MEM_PWR_CTRL; \ + uint32_t DSCL_MEM_PWR_CTRL + +#define DPP_DCN2_REG_VARIABLE_LIST_CM_APPEND \ + uint32_t CM_GAMUT_REMAP_B_C11_C12; \ + uint32_t CM_GAMUT_REMAP_B_C13_C14; \ + uint32_t CM_GAMUT_REMAP_B_C21_C22; \ + uint32_t CM_GAMUT_REMAP_B_C23_C24; \ + uint32_t CM_GAMUT_REMAP_B_C31_C32; \ + uint32_t CM_GAMUT_REMAP_B_C33_C34; \ + uint32_t CM_ICSC_B_C11_C12; \ + uint32_t CM_ICSC_B_C33_C34 struct dcn2_dpp_registers { DPP_DCN2_REG_VARIABLE_LIST; + DPP_DCN2_REG_VARIABLE_LIST_CM_APPEND; }; struct dcn20_dpp { @@ -634,6 +696,18 @@ struct dcn20_dpp { struct pwl_params pwl_data; }; +enum dcn20_input_csc_select { + DCN2_ICSC_SELECT_BYPASS = 0, + DCN2_ICSC_SELECT_ICSC_A = 1, + DCN2_ICSC_SELECT_ICSC_B = 2 +}; + +enum dcn20_gamut_remap_select { + DCN2_GAMUT_REMAP_BYPASS = 0, + DCN2_GAMUT_REMAP_COEF_A = 1, + DCN2_GAMUT_REMAP_COEF_B = 2 +}; + void dpp20_read_state(struct dpp *dpp_base, struct dcn_dpp_state *s); @@ -645,6 +719,16 @@ void dpp2_set_degamma( struct dpp *dpp_base, enum ipp_degamma_mode mode); +void dpp2_cm_set_gamut_remap( + struct dpp *dpp_base, + const struct dpp_grph_csc_adjustment *adjust); + +void dpp2_program_input_csc( + struct dpp *dpp_base, + enum dc_color_space color_space, + enum dcn20_input_csc_select input_select, + const struct out_csc_color_matrix *tbl_entry); + bool dpp20_program_blnd_lut( struct dpp *dpp_base, const struct pwl_params *params); @@ -668,7 +752,7 @@ void dscl2_calc_lb_num_partitions( void dpp2_set_cursor_attributes( struct dpp *dpp_base, - enum dc_cursor_color_format color_format); + struct dc_cursor_attributes *cursor_attributes); void dpp2_dummy_program_input_lut( struct dpp *dpp_base, @@ -683,11 +767,6 @@ void dpp2_set_hdr_multiplier( struct dpp *dpp_base, uint32_t multiplier); -bool dpp2_get_optimal_number_of_taps( - struct dpp *dpp, - struct scaler_data *scl_data, - const struct scaling_taps *in_taps); - bool dpp2_construct(struct dcn20_dpp *dpp2, struct dc_context *ctx, uint32_t inst, @@ -695,4 +774,7 @@ bool dpp2_construct(struct dcn20_dpp *dpp2, const struct dcn2_dpp_shift *tf_shift, const struct dcn2_dpp_mask *tf_mask); +void dpp2_power_on_obuf( + struct dpp *dpp_base, + bool power_on); #endif /* __DC_HWSS_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c index e28b8e7bedf5..8dc3d1f73984 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c @@ -36,6 +36,9 @@ #define REG(reg)\ dpp->tf_regs->reg +#define IND_REG(index) \ + (index) + #define CTX \ dpp->base.ctx @@ -44,15 +47,17 @@ dpp->tf_shift->field_name, dpp->tf_mask->field_name - - - static void dpp2_enable_cm_block( struct dpp *dpp_base) { struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); - REG_UPDATE(CM_CONTROL, CM_BYPASS, 0); + unsigned int cm_bypass_mode = 0; + //Temp, put CM in bypass mode + if (dpp_base->ctx->dc->debug.cm_in_bypass) + cm_bypass_mode = 1; + + REG_UPDATE(CM_CONTROL, CM_BYPASS, cm_bypass_mode); } @@ -144,12 +149,164 @@ void dpp2_set_degamma( case IPP_DEGAMMA_MODE_HW_xvYCC: REG_UPDATE(CM_DGAM_CONTROL, CM_DGAM_LUT_MODE, 2); break; + case IPP_DEGAMMA_MODE_USER_PWL: + REG_UPDATE(CM_DGAM_CONTROL, CM_DGAM_LUT_MODE, 3); + break; default: BREAK_TO_DEBUGGER(); break; } } +static void program_gamut_remap( + struct dcn20_dpp *dpp, + const uint16_t *regval, + enum dcn20_gamut_remap_select select) +{ + uint32_t cur_select = 0; + struct color_matrices_reg gam_regs; + + if (regval == NULL || select == DCN2_GAMUT_REMAP_BYPASS) { + REG_SET(CM_GAMUT_REMAP_CONTROL, 0, + CM_GAMUT_REMAP_MODE, 0); + return; + } + + /* determine which gamut_remap coefficients (A or B) we are using + * currently. select the alternate set to double buffer + * the update so gamut_remap is updated on frame boundary + */ + IX_REG_GET(CM_TEST_DEBUG_INDEX, CM_TEST_DEBUG_DATA, + CM_TEST_DEBUG_DATA_STATUS_IDX, + CM_TEST_DEBUG_DATA_GAMUT_REMAP_MODE, &cur_select); + + /* value stored in dbg reg will be 1 greater than mode we want */ + if (cur_select != DCN2_GAMUT_REMAP_COEF_A) + select = DCN2_GAMUT_REMAP_COEF_A; + else + select = DCN2_GAMUT_REMAP_COEF_B; + + gam_regs.shifts.csc_c11 = dpp->tf_shift->CM_GAMUT_REMAP_C11; + gam_regs.masks.csc_c11 = dpp->tf_mask->CM_GAMUT_REMAP_C11; + gam_regs.shifts.csc_c12 = dpp->tf_shift->CM_GAMUT_REMAP_C12; + gam_regs.masks.csc_c12 = dpp->tf_mask->CM_GAMUT_REMAP_C12; + + if (select == DCN2_GAMUT_REMAP_COEF_A) { + gam_regs.csc_c11_c12 = REG(CM_GAMUT_REMAP_C11_C12); + gam_regs.csc_c33_c34 = REG(CM_GAMUT_REMAP_C33_C34); + } else { + gam_regs.csc_c11_c12 = REG(CM_GAMUT_REMAP_B_C11_C12); + gam_regs.csc_c33_c34 = REG(CM_GAMUT_REMAP_B_C33_C34); + } + + cm_helper_program_color_matrices( + dpp->base.ctx, + regval, + &gam_regs); + + REG_SET( + CM_GAMUT_REMAP_CONTROL, 0, + CM_GAMUT_REMAP_MODE, select); + +} + +void dpp2_cm_set_gamut_remap( + struct dpp *dpp_base, + const struct dpp_grph_csc_adjustment *adjust) +{ + struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); + int i = 0; + + if (adjust->gamut_adjust_type != GRAPHICS_GAMUT_ADJUST_TYPE_SW) + /* Bypass if type is bypass or hw */ + program_gamut_remap(dpp, NULL, DCN2_GAMUT_REMAP_BYPASS); + else { + struct fixed31_32 arr_matrix[12]; + uint16_t arr_reg_val[12]; + + for (i = 0; i < 12; i++) + arr_matrix[i] = adjust->temperature_matrix[i]; + + convert_float_matrix( + arr_reg_val, arr_matrix, 12); + + program_gamut_remap(dpp, arr_reg_val, DCN2_GAMUT_REMAP_COEF_A); + } +} + +void dpp2_program_input_csc( + struct dpp *dpp_base, + enum dc_color_space color_space, + enum dcn20_input_csc_select input_select, + const struct out_csc_color_matrix *tbl_entry) +{ + struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); + int i; + int arr_size = sizeof(dpp_input_csc_matrix)/sizeof(struct dpp_input_csc_matrix); + const uint16_t *regval = NULL; + uint32_t cur_select = 0; + enum dcn20_input_csc_select select; + struct color_matrices_reg icsc_regs; + + if (input_select == DCN2_ICSC_SELECT_BYPASS) { + REG_SET(CM_ICSC_CONTROL, 0, CM_ICSC_MODE, 0); + return; + } + + if (tbl_entry == NULL) { + for (i = 0; i < arr_size; i++) + if (dpp_input_csc_matrix[i].color_space == color_space) { + regval = dpp_input_csc_matrix[i].regval; + break; + } + + if (regval == NULL) { + BREAK_TO_DEBUGGER(); + return; + } + } else { + regval = tbl_entry->regval; + } + + /* determine which CSC coefficients (A or B) we are using + * currently. select the alternate set to double buffer + * the CSC update so CSC is updated on frame boundary + */ + IX_REG_GET(CM_TEST_DEBUG_INDEX, CM_TEST_DEBUG_DATA, + CM_TEST_DEBUG_DATA_STATUS_IDX, + CM_TEST_DEBUG_DATA_ICSC_MODE, &cur_select); + + if (cur_select != DCN2_ICSC_SELECT_ICSC_A) + select = DCN2_ICSC_SELECT_ICSC_A; + else + select = DCN2_ICSC_SELECT_ICSC_B; + + icsc_regs.shifts.csc_c11 = dpp->tf_shift->CM_ICSC_C11; + icsc_regs.masks.csc_c11 = dpp->tf_mask->CM_ICSC_C11; + icsc_regs.shifts.csc_c12 = dpp->tf_shift->CM_ICSC_C12; + icsc_regs.masks.csc_c12 = dpp->tf_mask->CM_ICSC_C12; + + if (select == DCN2_ICSC_SELECT_ICSC_A) { + + icsc_regs.csc_c11_c12 = REG(CM_ICSC_C11_C12); + icsc_regs.csc_c33_c34 = REG(CM_ICSC_C33_C34); + + } else { + + icsc_regs.csc_c11_c12 = REG(CM_ICSC_B_C11_C12); + icsc_regs.csc_c33_c34 = REG(CM_ICSC_B_C33_C34); + + } + + cm_helper_program_color_matrices( + dpp->base.ctx, + regval, + &icsc_regs); + + REG_SET(CM_ICSC_CONTROL, 0, + CM_ICSC_MODE, select); +} + static void dpp20_power_on_blnd_lut( struct dpp *dpp_base, bool power_on) diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c index ffd0014ec3b5..6bdfee20b6a7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c @@ -23,13 +23,12 @@ * */ -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #include "reg_helper.h" #include "dcn20_dsc.h" #include "dsc/dscc_types.h" static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_config *pps); -static bool dsc_prepare_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, +static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals, struct dsc_optc_config *dsc_optc_cfg); static void dsc_init_reg_values(struct dsc_reg_values *reg_vals); static void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params); @@ -42,7 +41,8 @@ static void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock static void dsc2_read_state(struct display_stream_compressor *dsc, struct dcn_dsc_state *s); static bool dsc2_validate_stream(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg); static void dsc2_set_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, - struct dsc_optc_config *dsc_optc_cfg, uint8_t *dsc_packed_pps); + struct dsc_optc_config *dsc_optc_cfg); +static bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps); static void dsc2_enable(struct display_stream_compressor *dsc, int opp_pipe); static void dsc2_disable(struct display_stream_compressor *dsc); @@ -51,6 +51,7 @@ const struct dsc_funcs dcn20_dsc_funcs = { .dsc_read_state = dsc2_read_state, .dsc_validate_stream = dsc2_validate_stream, .dsc_set_config = dsc2_set_config, + .dsc_get_packed_pps = dsc2_get_packed_pps, .dsc_enable = dsc2_enable, .dsc_disable = dsc2_disable, }; @@ -117,7 +118,7 @@ static void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock dsc_enc_caps->color_formats.bits.RGB = 1; dsc_enc_caps->color_formats.bits.YCBCR_444 = 1; dsc_enc_caps->color_formats.bits.YCBCR_SIMPLE_422 = 1; - dsc_enc_caps->color_formats.bits.YCBCR_NATIVE_422 = 1; + dsc_enc_caps->color_formats.bits.YCBCR_NATIVE_422 = 0; dsc_enc_caps->color_formats.bits.YCBCR_NATIVE_420 = 1; dsc_enc_caps->color_depth.bits.COLOR_DEPTH_8_BPC = 1; @@ -162,46 +163,79 @@ static void dsc2_read_state(struct display_stream_compressor *dsc, struct dcn_ds static bool dsc2_validate_stream(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg) { struct dsc_optc_config dsc_optc_cfg; + struct dcn20_dsc *dsc20 = TO_DCN20_DSC(dsc); - if (dsc_cfg->pic_width > TO_DCN20_DSC(dsc)->max_image_width) + if (dsc_cfg->pic_width > dsc20->max_image_width) return false; - return dsc_prepare_config(dsc, dsc_cfg, &dsc_optc_cfg); + return dsc_prepare_config(dsc_cfg, &dsc20->reg_vals, &dsc_optc_cfg); } -static void dsc_config_log(struct display_stream_compressor *dsc, - const struct dsc_config *config) +static void dsc_config_log(struct display_stream_compressor *dsc, const struct dsc_config *config) { - DC_LOG_DSC("Setting DSC Config at DSC inst %d", dsc->inst); - DC_LOG_DSC("\n\tnum_slices_h %d\n\tnum_slices_v %d\n\tbits_per_pixel %d\n\tcolor_depth %d", - config->dc_dsc_cfg.num_slices_h, - config->dc_dsc_cfg.num_slices_v, + DC_LOG_DSC("\tnum_slices_h %d", config->dc_dsc_cfg.num_slices_h); + DC_LOG_DSC("\tnum_slices_v %d", config->dc_dsc_cfg.num_slices_v); + DC_LOG_DSC("\tbits_per_pixel %d (%d.%04d)", config->dc_dsc_cfg.bits_per_pixel, - config->color_depth); + config->dc_dsc_cfg.bits_per_pixel / 16, + ((config->dc_dsc_cfg.bits_per_pixel % 16) * 10000) / 16); + DC_LOG_DSC("\tcolor_depth %d", config->color_depth); } static void dsc2_set_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, - struct dsc_optc_config *dsc_optc_cfg, uint8_t *dsc_packed_pps) + struct dsc_optc_config *dsc_optc_cfg) { bool is_config_ok; struct dcn20_dsc *dsc20 = TO_DCN20_DSC(dsc); + DC_LOG_DSC(" "); + DC_LOG_DSC("Setting DSC Config at DSC inst %d", dsc->inst); dsc_config_log(dsc, dsc_cfg); - is_config_ok = dsc_prepare_config(dsc, dsc_cfg, dsc_optc_cfg); + is_config_ok = dsc_prepare_config(dsc_cfg, &dsc20->reg_vals, dsc_optc_cfg); ASSERT(is_config_ok); - drm_dsc_pps_payload_pack((struct drm_dsc_picture_parameter_set *)dsc_packed_pps, &dsc20->reg_vals.pps); + DC_LOG_DSC("programming DSC Picture Parameter Set (PPS):"); dsc_log_pps(dsc, &dsc20->reg_vals.pps); dsc_write_to_registers(dsc, &dsc20->reg_vals); } +static bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps) +{ + bool is_config_ok; + struct dsc_reg_values dsc_reg_vals; + struct dsc_optc_config dsc_optc_cfg; + + memset(&dsc_reg_vals, 0, sizeof(dsc_reg_vals)); + memset(&dsc_optc_cfg, 0, sizeof(dsc_optc_cfg)); + + DC_LOG_DSC("Getting packed DSC PPS for DSC Config:"); + dsc_config_log(dsc, dsc_cfg); + DC_LOG_DSC("DSC Picture Parameter Set (PPS):"); + is_config_ok = dsc_prepare_config(dsc_cfg, &dsc_reg_vals, &dsc_optc_cfg); + ASSERT(is_config_ok); + drm_dsc_pps_payload_pack((struct drm_dsc_picture_parameter_set *)dsc_packed_pps, &dsc_reg_vals.pps); + dsc_log_pps(dsc, &dsc_reg_vals.pps); + + return is_config_ok; +} + + static void dsc2_enable(struct display_stream_compressor *dsc, int opp_pipe) { struct dcn20_dsc *dsc20 = TO_DCN20_DSC(dsc); + int dsc_clock_en; + int dsc_fw_config; + int enabled_opp_pipe; - /* TODO Check if DSC alreay in use? */ - DC_LOG_DSC("enable DSC at opp pipe %d", opp_pipe); + DC_LOG_DSC("enable DSC %d at opp pipe %d", dsc->inst, opp_pipe); + + REG_GET(DSC_TOP_CONTROL, DSC_CLOCK_EN, &dsc_clock_en); + REG_GET_2(DSCRM_DSC_FORWARD_CONFIG, DSCRM_DSC_FORWARD_EN, &dsc_fw_config, DSCRM_DSC_OPP_PIPE_SOURCE, &enabled_opp_pipe); + if ((dsc_clock_en || dsc_fw_config) && enabled_opp_pipe != opp_pipe) { + DC_LOG_DSC("ERROR: DSC %d at opp pipe %d already enabled!", dsc->inst, enabled_opp_pipe); + ASSERT(0); + } REG_UPDATE(DSC_TOP_CONTROL, DSC_CLOCK_EN, 1); @@ -215,8 +249,18 @@ static void dsc2_enable(struct display_stream_compressor *dsc, int opp_pipe) static void dsc2_disable(struct display_stream_compressor *dsc) { struct dcn20_dsc *dsc20 = TO_DCN20_DSC(dsc); + int dsc_clock_en; + int dsc_fw_config; + int enabled_opp_pipe; - DC_LOG_DSC("disable DSC"); + DC_LOG_DSC("disable DSC %d", dsc->inst); + + REG_GET(DSC_TOP_CONTROL, DSC_CLOCK_EN, &dsc_clock_en); + REG_GET_2(DSCRM_DSC_FORWARD_CONFIG, DSCRM_DSC_FORWARD_EN, &dsc_fw_config, DSCRM_DSC_OPP_PIPE_SOURCE, &enabled_opp_pipe); + if (!dsc_clock_en || !dsc_fw_config) { + DC_LOG_DSC("ERROR: DSC %d at opp pipe %d already disabled!", dsc->inst, enabled_opp_pipe); + ASSERT(0); + } REG_UPDATE(DSCRM_DSC_FORWARD_CONFIG, DSCRM_DSC_FORWARD_EN, 0); @@ -232,7 +276,6 @@ static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_co int i; int bits_per_pixel = pps->bits_per_pixel; - DC_LOG_DSC("programming DSC Picture Parameter Set (PPS):"); DC_LOG_DSC("\tdsc_version_major %d", pps->dsc_version_major); DC_LOG_DSC("\tdsc_version_minor %d", pps->dsc_version_minor); DC_LOG_DSC("\tbits_per_component %d", pps->bits_per_component); @@ -282,13 +325,11 @@ static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_co } } -static bool dsc_prepare_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, +static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals, struct dsc_optc_config *dsc_optc_cfg) { struct dsc_parameters dsc_params; - struct dcn20_dsc *dsc20 = TO_DCN20_DSC(dsc); - /* Validate input parameters */ ASSERT(dsc_cfg->dc_dsc_cfg.num_slices_h); ASSERT(dsc_cfg->dc_dsc_cfg.num_slices_v); @@ -302,7 +343,7 @@ static bool dsc_prepare_config(struct display_stream_compressor *dsc, const stru dsc_cfg->dc_dsc_cfg.linebuf_depth == 0))); ASSERT(96 <= dsc_cfg->dc_dsc_cfg.bits_per_pixel && dsc_cfg->dc_dsc_cfg.bits_per_pixel <= 0x3ff); // 6.0 <= bits_per_pixel <= 63.9375 - if (!dsc_cfg->dc_dsc_cfg.num_slices_v || !dsc_cfg->dc_dsc_cfg.num_slices_v || + if (!dsc_cfg->dc_dsc_cfg.num_slices_v || !dsc_cfg->dc_dsc_cfg.num_slices_h || !(dsc_cfg->dc_dsc_cfg.version_minor == 1 || dsc_cfg->dc_dsc_cfg.version_minor == 2) || !dsc_cfg->pic_width || !dsc_cfg->pic_height || !((dsc_cfg->dc_dsc_cfg.version_minor == 1 && // v1.1 line buffer depth range: @@ -315,54 +356,54 @@ static bool dsc_prepare_config(struct display_stream_compressor *dsc, const stru return false; } - dsc_init_reg_values(&dsc20->reg_vals); + dsc_init_reg_values(dsc_reg_vals); /* Copy input config */ - dsc20->reg_vals.pixel_format = dsc_dc_pixel_encoding_to_dsc_pixel_format(dsc_cfg->pixel_encoding, dsc_cfg->dc_dsc_cfg.ycbcr422_simple); - dsc20->reg_vals.num_slices_h = dsc_cfg->dc_dsc_cfg.num_slices_h; - dsc20->reg_vals.num_slices_v = dsc_cfg->dc_dsc_cfg.num_slices_v; - dsc20->reg_vals.pps.dsc_version_minor = dsc_cfg->dc_dsc_cfg.version_minor; - dsc20->reg_vals.pps.pic_width = dsc_cfg->pic_width; - dsc20->reg_vals.pps.pic_height = dsc_cfg->pic_height; - dsc20->reg_vals.pps.bits_per_component = dsc_dc_color_depth_to_dsc_bits_per_comp(dsc_cfg->color_depth); - dsc20->reg_vals.pps.block_pred_enable = dsc_cfg->dc_dsc_cfg.block_pred_enable; - dsc20->reg_vals.pps.line_buf_depth = dsc_cfg->dc_dsc_cfg.linebuf_depth; - dsc20->reg_vals.alternate_ich_encoding_en = dsc20->reg_vals.pps.dsc_version_minor == 1 ? 0 : 1; + dsc_reg_vals->pixel_format = dsc_dc_pixel_encoding_to_dsc_pixel_format(dsc_cfg->pixel_encoding, dsc_cfg->dc_dsc_cfg.ycbcr422_simple); + dsc_reg_vals->num_slices_h = dsc_cfg->dc_dsc_cfg.num_slices_h; + dsc_reg_vals->num_slices_v = dsc_cfg->dc_dsc_cfg.num_slices_v; + dsc_reg_vals->pps.dsc_version_minor = dsc_cfg->dc_dsc_cfg.version_minor; + dsc_reg_vals->pps.pic_width = dsc_cfg->pic_width; + dsc_reg_vals->pps.pic_height = dsc_cfg->pic_height; + dsc_reg_vals->pps.bits_per_component = dsc_dc_color_depth_to_dsc_bits_per_comp(dsc_cfg->color_depth); + dsc_reg_vals->pps.block_pred_enable = dsc_cfg->dc_dsc_cfg.block_pred_enable; + dsc_reg_vals->pps.line_buf_depth = dsc_cfg->dc_dsc_cfg.linebuf_depth; + dsc_reg_vals->alternate_ich_encoding_en = dsc_reg_vals->pps.dsc_version_minor == 1 ? 0 : 1; // TODO: in addition to validating slice height (pic height must be divisible by slice height), // see what happens when the same condition doesn't apply for slice_width/pic_width. - dsc20->reg_vals.pps.slice_width = dsc_cfg->pic_width / dsc_cfg->dc_dsc_cfg.num_slices_h; - dsc20->reg_vals.pps.slice_height = dsc_cfg->pic_height / dsc_cfg->dc_dsc_cfg.num_slices_v; + dsc_reg_vals->pps.slice_width = dsc_cfg->pic_width / dsc_cfg->dc_dsc_cfg.num_slices_h; + dsc_reg_vals->pps.slice_height = dsc_cfg->pic_height / dsc_cfg->dc_dsc_cfg.num_slices_v; - ASSERT(dsc20->reg_vals.pps.slice_height * dsc_cfg->dc_dsc_cfg.num_slices_v == dsc_cfg->pic_height); - if (!(dsc20->reg_vals.pps.slice_height * dsc_cfg->dc_dsc_cfg.num_slices_v == dsc_cfg->pic_height)) { + ASSERT(dsc_reg_vals->pps.slice_height * dsc_cfg->dc_dsc_cfg.num_slices_v == dsc_cfg->pic_height); + if (!(dsc_reg_vals->pps.slice_height * dsc_cfg->dc_dsc_cfg.num_slices_v == dsc_cfg->pic_height)) { dm_output_to_console("%s: pix height %d not divisible by num_slices_v %d\n\n", __func__, dsc_cfg->pic_height, dsc_cfg->dc_dsc_cfg.num_slices_v); return false; } - dsc20->reg_vals.bpp_x32 = dsc_cfg->dc_dsc_cfg.bits_per_pixel << 1; - if (dsc20->reg_vals.pixel_format == DSC_PIXFMT_NATIVE_YCBCR420 || dsc20->reg_vals.pixel_format == DSC_PIXFMT_NATIVE_YCBCR422) - dsc20->reg_vals.pps.bits_per_pixel = dsc20->reg_vals.bpp_x32; + dsc_reg_vals->bpp_x32 = dsc_cfg->dc_dsc_cfg.bits_per_pixel << 1; + if (dsc_reg_vals->pixel_format == DSC_PIXFMT_NATIVE_YCBCR420 || dsc_reg_vals->pixel_format == DSC_PIXFMT_NATIVE_YCBCR422) + dsc_reg_vals->pps.bits_per_pixel = dsc_reg_vals->bpp_x32; else - dsc20->reg_vals.pps.bits_per_pixel = dsc20->reg_vals.bpp_x32 >> 1; + dsc_reg_vals->pps.bits_per_pixel = dsc_reg_vals->bpp_x32 >> 1; - dsc20->reg_vals.pps.convert_rgb = dsc20->reg_vals.pixel_format == DSC_PIXFMT_RGB ? 1 : 0; - dsc20->reg_vals.pps.native_422 = (dsc20->reg_vals.pixel_format == DSC_PIXFMT_NATIVE_YCBCR422); - dsc20->reg_vals.pps.native_420 = (dsc20->reg_vals.pixel_format == DSC_PIXFMT_NATIVE_YCBCR420); - dsc20->reg_vals.pps.simple_422 = (dsc20->reg_vals.pixel_format == DSC_PIXFMT_SIMPLE_YCBCR422); + dsc_reg_vals->pps.convert_rgb = dsc_reg_vals->pixel_format == DSC_PIXFMT_RGB ? 1 : 0; + dsc_reg_vals->pps.native_422 = (dsc_reg_vals->pixel_format == DSC_PIXFMT_NATIVE_YCBCR422); + dsc_reg_vals->pps.native_420 = (dsc_reg_vals->pixel_format == DSC_PIXFMT_NATIVE_YCBCR420); + dsc_reg_vals->pps.simple_422 = (dsc_reg_vals->pixel_format == DSC_PIXFMT_SIMPLE_YCBCR422); - if (dscc_compute_dsc_parameters(&dsc20->reg_vals.pps, &dsc_params)) { + if (dscc_compute_dsc_parameters(&dsc_reg_vals->pps, &dsc_params)) { dm_output_to_console("%s: DSC config failed\n", __func__); return false; } - dsc_update_from_dsc_parameters(&dsc20->reg_vals, &dsc_params); + dsc_update_from_dsc_parameters(dsc_reg_vals, &dsc_params); dsc_optc_cfg->bytes_per_pixel = dsc_params.bytes_per_pixel; - dsc_optc_cfg->slice_width = dsc20->reg_vals.pps.slice_width; - dsc_optc_cfg->is_pixel_format_444 = dsc20->reg_vals.pixel_format == DSC_PIXFMT_RGB || - dsc20->reg_vals.pixel_format == DSC_PIXFMT_YCBCR444 || - dsc20->reg_vals.pixel_format == DSC_PIXFMT_SIMPLE_YCBCR422; + dsc_optc_cfg->slice_width = dsc_reg_vals->pps.slice_width; + dsc_optc_cfg->is_pixel_format_444 = dsc_reg_vals->pixel_format == DSC_PIXFMT_RGB || + dsc_reg_vals->pixel_format == DSC_PIXFMT_YCBCR444 || + dsc_reg_vals->pixel_format == DSC_PIXFMT_SIMPLE_YCBCR422; return true; } @@ -427,6 +468,8 @@ static void dsc_init_reg_values(struct dsc_reg_values *reg_vals) { int i; + memset(reg_vals, 0, sizeof(struct dsc_reg_values)); + /* Non-PPS values */ reg_vals->dsc_clock_enable = 1; reg_vals->dsc_clock_gating_disable = 0; @@ -436,7 +479,7 @@ static void dsc_init_reg_values(struct dsc_reg_values *reg_vals) reg_vals->ich_reset_at_eol = 0; reg_vals->alternate_ich_encoding_en = 0; reg_vals->rc_buffer_model_size = 0; - reg_vals->disable_ich = 0; + /*reg_vals->disable_ich = 0;*/ reg_vals->dsc_dbg_en = 0; for (i = 0; i < 4; i++) @@ -518,9 +561,11 @@ static void dsc_write_to_registers(struct display_stream_compressor *dsc, const ALTERNATE_ICH_ENCODING_EN, reg_vals->alternate_ich_encoding_en, NUMBER_OF_SLICES_IN_VERTICAL_DIRECTION, reg_vals->num_slices_v - 1); - REG_SET_2(DSCC_CONFIG1, 0, + REG_SET(DSCC_CONFIG1, 0, + DSCC_RATE_CONTROL_BUFFER_MODEL_SIZE, reg_vals->rc_buffer_model_size); + /*REG_SET_2(DSCC_CONFIG1, 0, DSCC_RATE_CONTROL_BUFFER_MODEL_SIZE, reg_vals->rc_buffer_model_size, - DSCC_DISABLE_ICH, reg_vals->disable_ich); + DSCC_DISABLE_ICH, reg_vals->disable_ich);*/ REG_SET_4(DSCC_INTERRUPT_CONTROL_STATUS, 0, DSCC_RATE_CONTROL_BUFFER_MODEL0_OVERFLOW_OCCURRED_INT_EN, reg_vals->rc_buffer_model_overflow_int_en[0], @@ -691,4 +736,3 @@ static void dsc_write_to_registers(struct display_stream_compressor *dsc, const } } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h index 168865a16288..9855a7ed0387 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h @@ -21,7 +21,6 @@ * Authors: AMD * */ -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #ifndef __DCN20_DSC_H__ #define __DCN20_DSC_H__ @@ -103,7 +102,7 @@ DSC_SF(DSCC0_DSCC_CONFIG0, ALTERNATE_ICH_ENCODING_EN, mask_sh), \ DSC_SF(DSCC0_DSCC_CONFIG0, NUMBER_OF_SLICES_IN_VERTICAL_DIRECTION, mask_sh), \ DSC_SF(DSCC0_DSCC_CONFIG1, DSCC_RATE_CONTROL_BUFFER_MODEL_SIZE, mask_sh), \ - DSC_SF(DSCC0_DSCC_CONFIG1, DSCC_DISABLE_ICH, mask_sh), \ + /*DSC_SF(DSCC0_DSCC_CONFIG1, DSCC_DISABLE_ICH, mask_sh),*/ \ DSC_SF(DSCC0_DSCC_STATUS, DSCC_DOUBLE_BUFFER_REG_UPDATE_PENDING, mask_sh), \ DSC_SF(DSCC0_DSCC_INTERRUPT_CONTROL_STATUS, DSCC_RATE_BUFFER0_OVERFLOW_OCCURRED, mask_sh), \ DSC_SF(DSCC0_DSCC_INTERRUPT_CONTROL_STATUS, DSCC_RATE_BUFFER1_OVERFLOW_OCCURRED, mask_sh), \ @@ -278,7 +277,7 @@ type ALTERNATE_ICH_ENCODING_EN; \ type NUMBER_OF_SLICES_IN_VERTICAL_DIRECTION; \ type DSCC_RATE_CONTROL_BUFFER_MODEL_SIZE; \ - type DSCC_DISABLE_ICH; \ + /*type DSCC_DISABLE_ICH;*/ \ type DSCC_DOUBLE_BUFFER_REG_UPDATE_PENDING; \ type DSCC_RATE_BUFFER0_OVERFLOW_OCCURRED; \ type DSCC_RATE_BUFFER1_OVERFLOW_OCCURRED; \ @@ -572,4 +571,3 @@ void dsc2_construct(struct dcn20_dsc *dsc, #endif -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c index cd8bc92ce3ba..880954ac0b02 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c @@ -722,7 +722,6 @@ bool dwb_program_horz_scalar(struct dcn20_dwbc *dwbc20, struct scaling_taps num_taps) { uint32_t h_ratio_luma = 1; - uint32_t h_ratio_chroma = 1; uint32_t h_taps_luma = num_taps.h_taps; uint32_t h_taps_chroma = num_taps.h_taps_c; int32_t h_init_phase_luma = 0; @@ -747,7 +746,6 @@ bool dwb_program_horz_scalar(struct dcn20_dwbc *dwbc20, h_ratio_luma = -1; else h_ratio_luma = dc_fixpt_u3d19(tmp_h_ratio_luma) << 5; - h_ratio_chroma = h_ratio_luma * 2; /*Program ratio*/ REG_UPDATE(WBSCL_HORZ_FILTER_SCALE_RATIO, WBSCL_H_SCALE_RATIO, h_ratio_luma); @@ -803,7 +801,6 @@ bool dwb_program_vert_scalar(struct dcn20_dwbc *dwbc20, enum dwb_subsample_position subsample_position) { uint32_t v_ratio_luma = 1; - uint32_t v_ratio_chroma = 1; uint32_t v_taps_luma = num_taps.v_taps; uint32_t v_taps_chroma = num_taps.v_taps_c; int32_t v_init_phase_luma = 0; @@ -827,7 +824,6 @@ bool dwb_program_vert_scalar(struct dcn20_dwbc *dwbc20, v_ratio_luma = -1; else v_ratio_luma = dc_fixpt_u3d19(tmp_v_ratio_luma) << 5; - v_ratio_chroma = v_ratio_luma * 2; /*Program ratio*/ REG_UPDATE(WBSCL_VERT_FILTER_SCALE_RATIO, WBSCL_V_SCALE_RATIO, v_ratio_luma); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c index 6e2dbd03f9bf..9235f7d29454 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c @@ -26,6 +26,7 @@ #include "dcn20_hubbub.h" #include "reg_helper.h" +#include "clk_mgr.h" #define REG(reg)\ hubbub1->regs->reg @@ -185,14 +186,13 @@ static void hubbub2_get_blk256_size(unsigned int *blk256_width, unsigned int *bl } static void hubbub2_det_request_size( + unsigned int detile_buf_size, unsigned int height, unsigned int width, unsigned int bpe, bool *req128_horz_wc, bool *req128_vert_wc) { - unsigned int detile_buf_size = 164 * 1024; /* 164KB for DCN1.0 */ - unsigned int blk256_height = 0; unsigned int blk256_width = 0; unsigned int swath_bytes_horz_wc, swath_bytes_vert_wc; @@ -235,7 +235,8 @@ bool hubbub2_get_dcc_compression_cap(struct hubbub *hubbub, &segment_order_horz, &segment_order_vert)) return false; - hubbub2_det_request_size(input->surface_size.height, input->surface_size.width, + hubbub2_det_request_size(TO_DCN20_HUBBUB(hubbub)->detile_buf_size, + input->surface_size.height, input->surface_size.width, bpe, &req128_horz_wc, &req128_vert_wc); if (!req128_horz_wc && !req128_vert_wc) { @@ -292,6 +293,9 @@ bool hubbub2_get_dcc_compression_cap(struct hubbub *hubbub, output->grph.rgb.max_compressed_blk_size = 64; output->grph.rgb.independent_64b_blks = true; break; + default: + ASSERT(false); + break; } output->capable = true; output->const_color_support = true; @@ -379,6 +383,11 @@ int hubbub2_init_dchub_sys_ctx(struct hubbub *hubbub, REG_SET(DCN_VM_AGP_BASE, 0, AGP_BASE, pa_config->system_aperture.agp_base >> 24); + REG_SET(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB, 0, + DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB, (pa_config->page_table_default_page_addr >> 44) & 0xF); + REG_SET(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB, 0, + DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB, (pa_config->page_table_default_page_addr >> 12) & 0xFFFFFFFF); + if (pa_config->gart_config.page_table_start_addr != pa_config->gart_config.page_table_end_addr) { phys_config.page_table_start_addr = pa_config->gart_config.page_table_start_addr >> 12; phys_config.page_table_end_addr = pa_config->gart_config.page_table_end_addr >> 12; @@ -397,54 +406,67 @@ void hubbub2_update_dchub(struct hubbub *hubbub, { struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); - if (REG(DCHUBBUB_SDPIF_FB_TOP) == 0) { - ASSERT(false); - /*should not come here*/ + if (REG(DCN_VM_FB_LOCATION_TOP) == 0) return; - } - /* TODO: port code from dal2 */ + switch (dh_data->fb_mode) { case FRAME_BUFFER_MODE_ZFB_ONLY: /*For ZFB case need to put DCHUB FB BASE and TOP upside down to indicate ZFB mode*/ - REG_UPDATE(DCHUBBUB_SDPIF_FB_TOP, - SDPIF_FB_TOP, 0); - - REG_UPDATE(DCHUBBUB_SDPIF_FB_BASE, - SDPIF_FB_BASE, 0x0FFFF); - - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BASE, - SDPIF_AGP_BASE, dh_data->zfb_phys_addr_base >> 22); - - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BOT, - SDPIF_AGP_BOT, dh_data->zfb_mc_base_addr >> 22); - - REG_UPDATE(DCHUBBUB_SDPIF_AGP_TOP, - SDPIF_AGP_TOP, (dh_data->zfb_mc_base_addr + - dh_data->zfb_size_in_byte - 1) >> 22); + REG_UPDATE(DCN_VM_FB_LOCATION_TOP, + FB_TOP, 0); + + REG_UPDATE(DCN_VM_FB_LOCATION_BASE, + FB_BASE, 0xFFFFFF); + + /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ + REG_UPDATE(DCN_VM_AGP_BASE, + AGP_BASE, dh_data->zfb_phys_addr_base >> 24); + + /*This field defines the bottom range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_BOT, + AGP_BOT, dh_data->zfb_mc_base_addr >> 24); + + /*This field defines the top range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_TOP, + AGP_TOP, (dh_data->zfb_mc_base_addr + + dh_data->zfb_size_in_byte - 1) >> 24); break; case FRAME_BUFFER_MODE_MIXED_ZFB_AND_LOCAL: /*Should not touch FB LOCATION (done by VBIOS on AsicInit table)*/ - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BASE, - SDPIF_AGP_BASE, dh_data->zfb_phys_addr_base >> 22); + /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ + REG_UPDATE(DCN_VM_AGP_BASE, + AGP_BASE, dh_data->zfb_phys_addr_base >> 24); - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BOT, - SDPIF_AGP_BOT, dh_data->zfb_mc_base_addr >> 22); + /*This field defines the bottom range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_BOT, + AGP_BOT, dh_data->zfb_mc_base_addr >> 24); - REG_UPDATE(DCHUBBUB_SDPIF_AGP_TOP, - SDPIF_AGP_TOP, (dh_data->zfb_mc_base_addr + - dh_data->zfb_size_in_byte - 1) >> 22); + /*This field defines the top range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_TOP, + AGP_TOP, (dh_data->zfb_mc_base_addr + + dh_data->zfb_size_in_byte - 1) >> 24); break; case FRAME_BUFFER_MODE_LOCAL_ONLY: - /*Should not touch FB LOCATION (done by VBIOS on AsicInit table)*/ - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BASE, - SDPIF_AGP_BASE, 0); + /*Should not touch FB LOCATION (should be done by VBIOS)*/ + + /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ + REG_UPDATE(DCN_VM_AGP_BASE, + AGP_BASE, 0); - REG_UPDATE(DCHUBBUB_SDPIF_AGP_BOT, - SDPIF_AGP_BOT, 0X03FFFF); + /*This field defines the bottom range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_BOT, + AGP_BOT, 0xFFFFFF); - REG_UPDATE(DCHUBBUB_SDPIF_AGP_TOP, - SDPIF_AGP_TOP, 0); + /*This field defines the top range of the AGP aperture and represents the 24*/ + /*MSBs, bits [47:24] of the 48 address bits*/ + REG_UPDATE(DCN_VM_AGP_TOP, + AGP_TOP, 0); break; default: break; @@ -553,13 +575,23 @@ static void hubbub2_program_watermarks( */ hubbub1_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); hubbub1_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + + /* + * There's a special case when going from p-state support to p-state unsupported + * here we are going to LOWER watermarks to go to dummy p-state only, but this has + * to be done prepare_bandwidth, not optimize + */ + if (hubbub1->base.ctx->dc->clk_mgr->clks.prev_p_state_change_support == true && + hubbub1->base.ctx->dc->clk_mgr->clks.p_state_change_support == false) + safe_to_lower = true; + hubbub1_program_pstate_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); REG_SET(DCHUBBUB_ARB_SAT_LEVEL, 0, DCHUBBUB_ARB_SAT_LEVEL, 60 * refclk_mhz); REG_UPDATE(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 180); - hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); + hubbub->funcs->allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); } static const struct hubbub_funcs hubbub2_funcs = { @@ -572,6 +604,8 @@ static const struct hubbub_funcs hubbub2_funcs = { .wm_read_state = hubbub2_wm_read_state, .get_dchub_ref_freq = hubbub2_get_dchub_ref_freq, .program_watermarks = hubbub2_program_watermarks, + .is_allow_self_refresh_enabled = hubbub1_is_allow_self_refresh_enabled, + .allow_self_refresh_control = hubbub1_allow_self_refresh_control, }; void hubbub2_construct(struct dcn20_hubbub *hubbub, @@ -589,4 +623,5 @@ void hubbub2_construct(struct dcn20_hubbub *hubbub, hubbub->masks = hubbub_mask; hubbub->debug_test_index_pstate = 0xB; + hubbub->detile_buf_size = 164 * 1024; /* 164KB for DCN2.0 */ } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h index a7b6ca26a9ad..501532dd523a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h @@ -29,13 +29,21 @@ #include "dcn10/dcn10_hubbub.h" #include "dcn20_vmid.h" +#define HUBBUB_REG_LIST_DCN20_COMMON()\ + HUBBUB_REG_LIST_DCN_COMMON(), \ + SR(DCHUBBUB_CRC_CTRL), \ + SR(DCN_VM_FB_LOCATION_BASE),\ + SR(DCN_VM_FB_LOCATION_TOP),\ + SR(DCN_VM_FB_OFFSET),\ + SR(DCN_VM_AGP_BOT),\ + SR(DCN_VM_AGP_TOP),\ + SR(DCN_VM_AGP_BASE) + #define TO_DCN20_HUBBUB(hubbub)\ container_of(hubbub, struct dcn20_hubbub, base) -#define HUBBUB_REG_LIST_DCN20(id)\ +#define HUBBUB_REG_LIST_DCN20_COMMON()\ HUBBUB_REG_LIST_DCN_COMMON(), \ - HUBBUB_VM_REG_LIST(), \ - HUBBUB_SR_WATERMARK_REG_LIST(), \ SR(DCHUBBUB_CRC_CTRL), \ SR(DCN_VM_FB_LOCATION_BASE),\ SR(DCN_VM_FB_LOCATION_TOP),\ @@ -44,6 +52,14 @@ SR(DCN_VM_AGP_TOP),\ SR(DCN_VM_AGP_BASE) +#define HUBBUB_REG_LIST_DCN20(id)\ + HUBBUB_REG_LIST_DCN20_COMMON(), \ + HUBBUB_SR_WATERMARK_REG_LIST(), \ + HUBBUB_VM_REG_LIST(),\ + SR(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB),\ + SR(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB) + + #define HUBBUB_MASK_SH_LIST_DCN20(mask_sh)\ HUBBUB_MASK_SH_LIST_DCN_COMMON(mask_sh), \ HUBBUB_MASK_SH_LIST_STUTTER(mask_sh), \ @@ -53,7 +69,9 @@ HUBBUB_SF(DCN_VM_FB_OFFSET, FB_OFFSET, mask_sh), \ HUBBUB_SF(DCN_VM_AGP_BOT, AGP_BOT, mask_sh), \ HUBBUB_SF(DCN_VM_AGP_TOP, AGP_TOP, mask_sh), \ - HUBBUB_SF(DCN_VM_AGP_BASE, AGP_BASE, mask_sh) + HUBBUB_SF(DCN_VM_AGP_BASE, AGP_BASE, mask_sh), \ + HUBBUB_SF(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB, DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_MSB, mask_sh), \ + HUBBUB_SF(DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB, DCN_VM_PROTECTION_FAULT_DEFAULT_ADDR_LSB, mask_sh) struct dcn20_hubbub { struct hubbub base; @@ -63,6 +81,7 @@ struct dcn20_hubbub { unsigned int debug_test_index_pstate; struct dcn_watermark_set watermarks; struct dcn20_vmid vmid[16]; + unsigned int detile_buf_size; }; void hubbub2_construct(struct dcn20_hubbub *hubbub, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c index d3f7dd374d50..84d7ac5dd206 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c @@ -30,6 +30,8 @@ #include "reg_helper.h" #include "basics/conversion.h" +#define DC_LOGGER_INIT(logger) + #define REG(reg)\ hubp2->hubp_regs->reg @@ -40,81 +42,6 @@ #define FN(reg_name, field_name) \ hubp2->hubp_shift->field_name, hubp2->hubp_mask->field_name -void hubp2_update_dchub( - struct hubp *hubp, - struct dchub_init_data *dh_data) -{ - struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); - if (REG(DCN_VM_FB_LOCATION_TOP) == 0) - return; - - switch (dh_data->fb_mode) { - case FRAME_BUFFER_MODE_ZFB_ONLY: - /*For ZFB case need to put DCHUB FB BASE and TOP upside down to indicate ZFB mode*/ - REG_UPDATE(DCN_VM_FB_LOCATION_TOP, - FB_TOP, 0); - - REG_UPDATE(DCN_VM_FB_LOCATION_BASE, - FB_BASE, 0xFFFFFF); - - /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ - REG_UPDATE(DCN_VM_AGP_BASE, - AGP_BASE, dh_data->zfb_phys_addr_base >> 24); - - /*This field defines the bottom range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_BOT, - AGP_BOT, dh_data->zfb_mc_base_addr >> 24); - - /*This field defines the top range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_TOP, - AGP_TOP, (dh_data->zfb_mc_base_addr + - dh_data->zfb_size_in_byte - 1) >> 24); - break; - case FRAME_BUFFER_MODE_MIXED_ZFB_AND_LOCAL: - /*Should not touch FB LOCATION (done by VBIOS on AsicInit table)*/ - - /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ - REG_UPDATE(DCN_VM_AGP_BASE, - AGP_BASE, dh_data->zfb_phys_addr_base >> 24); - - /*This field defines the bottom range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_BOT, - AGP_BOT, dh_data->zfb_mc_base_addr >> 24); - - /*This field defines the top range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_TOP, - AGP_TOP, (dh_data->zfb_mc_base_addr + - dh_data->zfb_size_in_byte - 1) >> 24); - break; - case FRAME_BUFFER_MODE_LOCAL_ONLY: - /*Should not touch FB LOCATION (should be done by VBIOS)*/ - - /*This field defines the 24 MSBs, bits [47:24] of the 48 bit AGP Base*/ - REG_UPDATE(DCN_VM_AGP_BASE, - AGP_BASE, 0); - - /*This field defines the bottom range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_BOT, - AGP_BOT, 0xFFFFFF); - - /*This field defines the top range of the AGP aperture and represents the 24*/ - /*MSBs, bits [47:24] of the 48 address bits*/ - REG_UPDATE(DCN_VM_AGP_TOP, - AGP_TOP, 0); - break; - default: - break; - } - - dh_data->dchub_initialzied = true; - dh_data->dchub_info_valid = false; -} - void hubp2_set_vm_system_aperture_settings(struct hubp *hubp, struct vm_system_aperture_param *apt) { @@ -156,7 +83,85 @@ void hubp2_program_deadline( { struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); - hubp1_program_deadline(hubp, dlg_attr, ttu_attr); + /* DLG - Per hubp */ + REG_SET_2(BLANK_OFFSET_0, 0, + REFCYC_H_BLANK_END, dlg_attr->refcyc_h_blank_end, + DLG_V_BLANK_END, dlg_attr->dlg_vblank_end); + + REG_SET(BLANK_OFFSET_1, 0, + MIN_DST_Y_NEXT_START, dlg_attr->min_dst_y_next_start); + + REG_SET(DST_DIMENSIONS, 0, + REFCYC_PER_HTOTAL, dlg_attr->refcyc_per_htotal); + + REG_SET_2(DST_AFTER_SCALER, 0, + REFCYC_X_AFTER_SCALER, dlg_attr->refcyc_x_after_scaler, + DST_Y_AFTER_SCALER, dlg_attr->dst_y_after_scaler); + + REG_SET(REF_FREQ_TO_PIX_FREQ, 0, + REF_FREQ_TO_PIX_FREQ, dlg_attr->ref_freq_to_pix_freq); + + /* DLG - Per luma/chroma */ + REG_SET(VBLANK_PARAMETERS_1, 0, + REFCYC_PER_PTE_GROUP_VBLANK_L, dlg_attr->refcyc_per_pte_group_vblank_l); + + if (REG(NOM_PARAMETERS_0)) + REG_SET(NOM_PARAMETERS_0, 0, + DST_Y_PER_PTE_ROW_NOM_L, dlg_attr->dst_y_per_pte_row_nom_l); + + if (REG(NOM_PARAMETERS_1)) + REG_SET(NOM_PARAMETERS_1, 0, + REFCYC_PER_PTE_GROUP_NOM_L, dlg_attr->refcyc_per_pte_group_nom_l); + + REG_SET(NOM_PARAMETERS_4, 0, + DST_Y_PER_META_ROW_NOM_L, dlg_attr->dst_y_per_meta_row_nom_l); + + REG_SET(NOM_PARAMETERS_5, 0, + REFCYC_PER_META_CHUNK_NOM_L, dlg_attr->refcyc_per_meta_chunk_nom_l); + + REG_SET_2(PER_LINE_DELIVERY, 0, + REFCYC_PER_LINE_DELIVERY_L, dlg_attr->refcyc_per_line_delivery_l, + REFCYC_PER_LINE_DELIVERY_C, dlg_attr->refcyc_per_line_delivery_c); + + REG_SET(VBLANK_PARAMETERS_2, 0, + REFCYC_PER_PTE_GROUP_VBLANK_C, dlg_attr->refcyc_per_pte_group_vblank_c); + + if (REG(NOM_PARAMETERS_2)) + REG_SET(NOM_PARAMETERS_2, 0, + DST_Y_PER_PTE_ROW_NOM_C, dlg_attr->dst_y_per_pte_row_nom_c); + + if (REG(NOM_PARAMETERS_3)) + REG_SET(NOM_PARAMETERS_3, 0, + REFCYC_PER_PTE_GROUP_NOM_C, dlg_attr->refcyc_per_pte_group_nom_c); + + REG_SET(NOM_PARAMETERS_6, 0, + DST_Y_PER_META_ROW_NOM_C, dlg_attr->dst_y_per_meta_row_nom_c); + + REG_SET(NOM_PARAMETERS_7, 0, + REFCYC_PER_META_CHUNK_NOM_C, dlg_attr->refcyc_per_meta_chunk_nom_c); + + /* TTU - per hubp */ + REG_SET_2(DCN_TTU_QOS_WM, 0, + QoS_LEVEL_LOW_WM, ttu_attr->qos_level_low_wm, + QoS_LEVEL_HIGH_WM, ttu_attr->qos_level_high_wm); + + /* TTU - per luma/chroma */ + /* Assumed surf0 is luma and 1 is chroma */ + + REG_SET_3(DCN_SURF0_TTU_CNTL0, 0, + REFCYC_PER_REQ_DELIVERY, ttu_attr->refcyc_per_req_delivery_l, + QoS_LEVEL_FIXED, ttu_attr->qos_level_fixed_l, + QoS_RAMP_DISABLE, ttu_attr->qos_ramp_disable_l); + + REG_SET_3(DCN_SURF1_TTU_CNTL0, 0, + REFCYC_PER_REQ_DELIVERY, ttu_attr->refcyc_per_req_delivery_c, + QoS_LEVEL_FIXED, ttu_attr->qos_level_fixed_c, + QoS_RAMP_DISABLE, ttu_attr->qos_ramp_disable_c); + + REG_SET_3(DCN_CUR0_TTU_CNTL0, 0, + REFCYC_PER_REQ_DELIVERY, ttu_attr->refcyc_per_req_delivery_cur0, + QoS_LEVEL_FIXED, ttu_attr->qos_level_fixed_cur0, + QoS_RAMP_DISABLE, ttu_attr->qos_ramp_disable_cur0); REG_SET(FLIP_PARAMETERS_1, 0, REFCYC_PER_PTE_GROUP_FLIP_L, dlg_attr->refcyc_per_pte_group_flip_l); @@ -184,6 +189,39 @@ void hubp2_vready_at_or_After_vsync(struct hubp *hubp, REG_UPDATE(DCHUBP_CNTL, HUBP_VREADY_AT_OR_AFTER_VSYNC, value); } +void hubp2_program_requestor( + struct hubp *hubp, + struct _vcs_dpi_display_rq_regs_st *rq_regs) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_UPDATE(HUBPRET_CONTROL, + DET_BUF_PLANE1_BASE_ADDRESS, rq_regs->plane1_base_address); + REG_SET_4(DCN_EXPANSION_MODE, 0, + DRQ_EXPANSION_MODE, rq_regs->drq_expansion_mode, + PRQ_EXPANSION_MODE, rq_regs->prq_expansion_mode, + MRQ_EXPANSION_MODE, rq_regs->mrq_expansion_mode, + CRQ_EXPANSION_MODE, rq_regs->crq_expansion_mode); + REG_SET_8(DCHUBP_REQ_SIZE_CONFIG, 0, + CHUNK_SIZE, rq_regs->rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, rq_regs->rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, rq_regs->rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, rq_regs->rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, rq_regs->rq_regs_l.dpte_group_size, + MPTE_GROUP_SIZE, rq_regs->rq_regs_l.mpte_group_size, + SWATH_HEIGHT, rq_regs->rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, rq_regs->rq_regs_l.pte_row_height_linear); + REG_SET_8(DCHUBP_REQ_SIZE_CONFIG_C, 0, + CHUNK_SIZE_C, rq_regs->rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, rq_regs->rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, rq_regs->rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, rq_regs->rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, rq_regs->rq_regs_c.dpte_group_size, + MPTE_GROUP_SIZE_C, rq_regs->rq_regs_c.mpte_group_size, + SWATH_HEIGHT_C, rq_regs->rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, rq_regs->rq_regs_c.pte_row_height_linear); +} + static void hubp2_setup( struct hubp *hubp, struct _vcs_dpi_display_dlg_regs_st *dlg_attr, @@ -196,7 +234,7 @@ static void hubp2_setup( */ hubp2_vready_at_or_After_vsync(hubp, pipe_dest); - hubp1_program_requestor(hubp, rq_regs); + hubp2_program_requestor(hubp, rq_regs); hubp2_program_deadline(hubp, dlg_attr, ttu_attr); } @@ -283,11 +321,203 @@ static void hubp2_program_tiling( PIPE_ALIGNED, 0); } +void hubp2_program_size( + struct hubp *hubp, + enum surface_pixel_format format, + const struct plane_size *plane_size, + struct dc_plane_dcc_param *dcc) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t pitch, meta_pitch, pitch_c, meta_pitch_c; + bool use_pitch_c = false; + + /* Program data and meta surface pitch (calculation from addrlib) + * 444 or 420 luma + */ + use_pitch_c = format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN + && format < SURFACE_PIXEL_FORMAT_SUBSAMPLE_END; + if (use_pitch_c) { + ASSERT(plane_size->chroma_pitch != 0); + /* Chroma pitch zero can cause system hang! */ + + pitch = plane_size->surface_pitch - 1; + meta_pitch = dcc->meta_pitch - 1; + pitch_c = plane_size->chroma_pitch - 1; + meta_pitch_c = dcc->meta_pitch_c - 1; + } else { + pitch = plane_size->surface_pitch - 1; + meta_pitch = dcc->meta_pitch - 1; + pitch_c = 0; + meta_pitch_c = 0; + } + + if (!dcc->enable) { + meta_pitch = 0; + meta_pitch_c = 0; + } + + REG_UPDATE_2(DCSURF_SURFACE_PITCH, + PITCH, pitch, META_PITCH, meta_pitch); + + use_pitch_c = format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN; + if (use_pitch_c) + REG_UPDATE_2(DCSURF_SURFACE_PITCH_C, + PITCH_C, pitch_c, META_PITCH_C, meta_pitch_c); +} + +void hubp2_program_rotation( + struct hubp *hubp, + enum dc_rotation_angle rotation, + bool horizontal_mirror) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t mirror; + + + if (horizontal_mirror) + mirror = 1; + else + mirror = 0; + + /* Program rotation angle and horz mirror - no mirror */ + if (rotation == ROTATION_ANGLE_0) + REG_UPDATE_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, 0, + H_MIRROR_EN, mirror); + else if (rotation == ROTATION_ANGLE_90) + REG_UPDATE_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, 1, + H_MIRROR_EN, mirror); + else if (rotation == ROTATION_ANGLE_180) + REG_UPDATE_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, 2, + H_MIRROR_EN, mirror); + else if (rotation == ROTATION_ANGLE_270) + REG_UPDATE_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, 3, + H_MIRROR_EN, mirror); +} + +void hubp2_dcc_control(struct hubp *hubp, bool enable, + enum hubp_ind_block_size independent_64b_blks) +{ + uint32_t dcc_en = enable ? 1 : 0; + uint32_t dcc_ind_64b_blk = independent_64b_blks ? 1 : 0; + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_UPDATE_4(DCSURF_SURFACE_CONTROL, + PRIMARY_SURFACE_DCC_EN, dcc_en, + PRIMARY_SURFACE_DCC_IND_64B_BLK, dcc_ind_64b_blk, + SECONDARY_SURFACE_DCC_EN, dcc_en, + SECONDARY_SURFACE_DCC_IND_64B_BLK, dcc_ind_64b_blk); +} + +void hubp2_program_pixel_format( + struct hubp *hubp, + enum surface_pixel_format format) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t red_bar = 3; + uint32_t blue_bar = 2; + + /* swap for ABGR format */ + if (format == SURFACE_PIXEL_FORMAT_GRPH_ABGR8888 + || format == SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010 + || format == SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010_XR_BIAS + || format == SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F) { + red_bar = 2; + blue_bar = 3; + } + + REG_UPDATE_2(HUBPRET_CONTROL, + CROSSBAR_SRC_CB_B, blue_bar, + CROSSBAR_SRC_CR_R, red_bar); + + /* Mapping is same as ipp programming (cnvc) */ + + switch (format) { + case SURFACE_PIXEL_FORMAT_GRPH_ARGB1555: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 1); + break; + case SURFACE_PIXEL_FORMAT_GRPH_RGB565: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 3); + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB8888: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR8888: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 8); + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB2101010: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR2101010_XR_BIAS: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 10); + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 22); + break; + case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F:/*we use crossbar already*/ + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 24); + break; + + case SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 65); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 64); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCbCr: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 67); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 66); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_AYCrCb8888: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 12); + break; + case SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FIX: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 112); + break; + case SURFACE_PIXEL_FORMAT_GRPH_BGR101111_FIX: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 113); + break; + case SURFACE_PIXEL_FORMAT_VIDEO_ACrYCb2101010: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 114); + break; + case SURFACE_PIXEL_FORMAT_GRPH_RGB111110_FLOAT: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 118); + break; + case SURFACE_PIXEL_FORMAT_GRPH_BGR101111_FLOAT: + REG_UPDATE(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, 119); + break; + default: + BREAK_TO_DEBUGGER(); + break; + } + + /* don't see the need of program the xbar in DCN 1.0 */ +} + void hubp2_program_surface_config( struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, @@ -295,11 +525,11 @@ void hubp2_program_surface_config( { struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); - hubp1_dcc_control(hubp, dcc->enable, dcc->grph.independent_64b_blks); + hubp2_dcc_control(hubp, dcc->enable, dcc->independent_64b_blks); hubp2_program_tiling(hubp2, tiling_info, format); - hubp1_program_size(hubp, format, plane_size, dcc); - hubp1_program_rotation(hubp, rotation, horizontal_mirror); - hubp1_program_pixel_format(hubp, format); + hubp2_program_size(hubp, format, plane_size, dcc); + hubp2_program_rotation(hubp, rotation, horizontal_mirror); + hubp2_program_pixel_format(hubp, format); } enum cursor_lines_per_chunk hubp2_get_lines_per_chunk( @@ -652,30 +882,702 @@ void hubp2_set_flip_control_surface_gsl(struct hubp *hubp, bool enable) REG_UPDATE(DCSURF_FLIP_CONTROL2, SURFACE_GSL_ENABLE, enable ? 1 : 0); } +bool hubp2_is_flip_pending(struct hubp *hubp) +{ + uint32_t flip_pending = 0; + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + struct dc_plane_address earliest_inuse_address; + + REG_GET(DCSURF_FLIP_CONTROL, + SURFACE_FLIP_PENDING, &flip_pending); + + REG_GET(DCSURF_SURFACE_EARLIEST_INUSE, + SURFACE_EARLIEST_INUSE_ADDRESS, &earliest_inuse_address.grph.addr.low_part); + + REG_GET(DCSURF_SURFACE_EARLIEST_INUSE_HIGH, + SURFACE_EARLIEST_INUSE_ADDRESS_HIGH, &earliest_inuse_address.grph.addr.high_part); + + if (flip_pending) + return true; + + if (earliest_inuse_address.grph.addr.quad_part != hubp->request_address.grph.addr.quad_part) + return true; + + return false; +} + +void hubp2_set_blank(struct hubp *hubp, bool blank) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t blank_en = blank ? 1 : 0; + + REG_UPDATE_2(DCHUBP_CNTL, + HUBP_BLANK_EN, blank_en, + HUBP_TTU_DISABLE, blank_en); + + if (blank) { + uint32_t reg_val = REG_READ(DCHUBP_CNTL); + + if (reg_val) { + /* init sequence workaround: in case HUBP is + * power gated, this wait would timeout. + * + * we just wrote reg_val to non-0, if it stay 0 + * it means HUBP is gated + */ + REG_WAIT(DCHUBP_CNTL, + HUBP_NO_OUTSTANDING_REQ, 1, + 1, 200); + } + + hubp->mpcc_id = 0xf; + hubp->opp_id = OPP_ID_INVALID; + } +} + +void hubp2_cursor_set_position( + struct hubp *hubp, + const struct dc_cursor_position *pos, + const struct dc_cursor_mi_param *param) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + int src_x_offset = pos->x - pos->x_hotspot - param->viewport.x; + int src_y_offset = pos->y - pos->y_hotspot - param->viewport.y; + int x_hotspot = pos->x_hotspot; + int y_hotspot = pos->y_hotspot; + int cursor_height = (int)hubp->curs_attr.height; + int cursor_width = (int)hubp->curs_attr.width; + uint32_t dst_x_offset; + uint32_t cur_en = pos->enable ? 1 : 0; + + /* + * Guard aganst cursor_set_position() from being called with invalid + * attributes + * + * TODO: Look at combining cursor_set_position() and + * cursor_set_attributes() into cursor_update() + */ + if (hubp->curs_attr.address.quad_part == 0) + return; + + // Rotated cursor width/height and hotspots tweaks for offset calculation + if (param->rotation == ROTATION_ANGLE_90 || param->rotation == ROTATION_ANGLE_270) { + swap(cursor_height, cursor_width); + if (param->rotation == ROTATION_ANGLE_90) { + src_x_offset = pos->x - pos->y_hotspot - param->viewport.x; + src_y_offset = pos->y - pos->x_hotspot - param->viewport.y; + } + } else if (param->rotation == ROTATION_ANGLE_180) { + src_x_offset = pos->x - param->viewport.x; + src_y_offset = pos->y - param->viewport.y; + } + + if (param->mirror) { + x_hotspot = param->viewport.width - x_hotspot; + src_x_offset = param->viewport.x + param->viewport.width - src_x_offset; + } + + dst_x_offset = (src_x_offset >= 0) ? src_x_offset : 0; + dst_x_offset *= param->ref_clk_khz; + dst_x_offset /= param->pixel_clk_khz; + + ASSERT(param->h_scale_ratio.value); + + if (param->h_scale_ratio.value) + dst_x_offset = dc_fixpt_floor(dc_fixpt_div( + dc_fixpt_from_int(dst_x_offset), + param->h_scale_ratio)); + + if (src_x_offset >= (int)param->viewport.width) + cur_en = 0; /* not visible beyond right edge*/ + + if (src_x_offset + cursor_width <= 0) + cur_en = 0; /* not visible beyond left edge*/ + + if (src_y_offset >= (int)param->viewport.height) + cur_en = 0; /* not visible beyond bottom edge*/ + + if (src_y_offset + cursor_height <= 0) + cur_en = 0; /* not visible beyond top edge*/ + + if (cur_en && REG_READ(CURSOR_SURFACE_ADDRESS) == 0) + hubp->funcs->set_cursor_attributes(hubp, &hubp->curs_attr); + + REG_UPDATE(CURSOR_CONTROL, + CURSOR_ENABLE, cur_en); + + REG_SET_2(CURSOR_POSITION, 0, + CURSOR_X_POSITION, pos->x, + CURSOR_Y_POSITION, pos->y); + + REG_SET_2(CURSOR_HOT_SPOT, 0, + CURSOR_HOT_SPOT_X, x_hotspot, + CURSOR_HOT_SPOT_Y, y_hotspot); + + REG_SET(CURSOR_DST_OFFSET, 0, + CURSOR_DST_X_OFFSET, dst_x_offset); + /* TODO Handle surface pixel formats other than 4:4:4 */ +} + +void hubp2_clk_cntl(struct hubp *hubp, bool enable) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + uint32_t clk_enable = enable ? 1 : 0; + + REG_UPDATE(HUBP_CLK_CNTL, HUBP_CLOCK_ENABLE, clk_enable); +} + +void hubp2_vtg_sel(struct hubp *hubp, uint32_t otg_inst) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_UPDATE(DCHUBP_CNTL, HUBP_VTG_SEL, otg_inst); +} + +void hubp2_clear_underflow(struct hubp *hubp) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + + REG_UPDATE(DCHUBP_CNTL, HUBP_UNDERFLOW_CLEAR, 1); +} + +void hubp2_read_state_common(struct hubp *hubp) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + struct dcn_hubp_state *s = &hubp2->state; + struct _vcs_dpi_display_dlg_regs_st *dlg_attr = &s->dlg_attr; + struct _vcs_dpi_display_ttu_regs_st *ttu_attr = &s->ttu_attr; + struct _vcs_dpi_display_rq_regs_st *rq_regs = &s->rq_regs; + + /* Requester */ + REG_GET(HUBPRET_CONTROL, + DET_BUF_PLANE1_BASE_ADDRESS, &rq_regs->plane1_base_address); + REG_GET_4(DCN_EXPANSION_MODE, + DRQ_EXPANSION_MODE, &rq_regs->drq_expansion_mode, + PRQ_EXPANSION_MODE, &rq_regs->prq_expansion_mode, + MRQ_EXPANSION_MODE, &rq_regs->mrq_expansion_mode, + CRQ_EXPANSION_MODE, &rq_regs->crq_expansion_mode); + + /* DLG - Per hubp */ + REG_GET_2(BLANK_OFFSET_0, + REFCYC_H_BLANK_END, &dlg_attr->refcyc_h_blank_end, + DLG_V_BLANK_END, &dlg_attr->dlg_vblank_end); + + REG_GET(BLANK_OFFSET_1, + MIN_DST_Y_NEXT_START, &dlg_attr->min_dst_y_next_start); + + REG_GET(DST_DIMENSIONS, + REFCYC_PER_HTOTAL, &dlg_attr->refcyc_per_htotal); + + REG_GET_2(DST_AFTER_SCALER, + REFCYC_X_AFTER_SCALER, &dlg_attr->refcyc_x_after_scaler, + DST_Y_AFTER_SCALER, &dlg_attr->dst_y_after_scaler); + + if (REG(PREFETCH_SETTINS)) + REG_GET_2(PREFETCH_SETTINS, + DST_Y_PREFETCH, &dlg_attr->dst_y_prefetch, + VRATIO_PREFETCH, &dlg_attr->vratio_prefetch); + else + REG_GET_2(PREFETCH_SETTINGS, + DST_Y_PREFETCH, &dlg_attr->dst_y_prefetch, + VRATIO_PREFETCH, &dlg_attr->vratio_prefetch); + + REG_GET_2(VBLANK_PARAMETERS_0, + DST_Y_PER_VM_VBLANK, &dlg_attr->dst_y_per_vm_vblank, + DST_Y_PER_ROW_VBLANK, &dlg_attr->dst_y_per_row_vblank); + + REG_GET(REF_FREQ_TO_PIX_FREQ, + REF_FREQ_TO_PIX_FREQ, &dlg_attr->ref_freq_to_pix_freq); + + /* DLG - Per luma/chroma */ + REG_GET(VBLANK_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_VBLANK_L, &dlg_attr->refcyc_per_pte_group_vblank_l); + + REG_GET(VBLANK_PARAMETERS_3, + REFCYC_PER_META_CHUNK_VBLANK_L, &dlg_attr->refcyc_per_meta_chunk_vblank_l); + + if (REG(NOM_PARAMETERS_0)) + REG_GET(NOM_PARAMETERS_0, + DST_Y_PER_PTE_ROW_NOM_L, &dlg_attr->dst_y_per_pte_row_nom_l); + + if (REG(NOM_PARAMETERS_1)) + REG_GET(NOM_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_NOM_L, &dlg_attr->refcyc_per_pte_group_nom_l); + + REG_GET(NOM_PARAMETERS_4, + DST_Y_PER_META_ROW_NOM_L, &dlg_attr->dst_y_per_meta_row_nom_l); + + REG_GET(NOM_PARAMETERS_5, + REFCYC_PER_META_CHUNK_NOM_L, &dlg_attr->refcyc_per_meta_chunk_nom_l); + + REG_GET_2(PER_LINE_DELIVERY_PRE, + REFCYC_PER_LINE_DELIVERY_PRE_L, &dlg_attr->refcyc_per_line_delivery_pre_l, + REFCYC_PER_LINE_DELIVERY_PRE_C, &dlg_attr->refcyc_per_line_delivery_pre_c); + + REG_GET_2(PER_LINE_DELIVERY, + REFCYC_PER_LINE_DELIVERY_L, &dlg_attr->refcyc_per_line_delivery_l, + REFCYC_PER_LINE_DELIVERY_C, &dlg_attr->refcyc_per_line_delivery_c); + + if (REG(PREFETCH_SETTINS_C)) + REG_GET(PREFETCH_SETTINS_C, + VRATIO_PREFETCH_C, &dlg_attr->vratio_prefetch_c); + else + REG_GET(PREFETCH_SETTINGS_C, + VRATIO_PREFETCH_C, &dlg_attr->vratio_prefetch_c); + + REG_GET(VBLANK_PARAMETERS_2, + REFCYC_PER_PTE_GROUP_VBLANK_C, &dlg_attr->refcyc_per_pte_group_vblank_c); + + REG_GET(VBLANK_PARAMETERS_4, + REFCYC_PER_META_CHUNK_VBLANK_C, &dlg_attr->refcyc_per_meta_chunk_vblank_c); + + if (REG(NOM_PARAMETERS_2)) + REG_GET(NOM_PARAMETERS_2, + DST_Y_PER_PTE_ROW_NOM_C, &dlg_attr->dst_y_per_pte_row_nom_c); + + if (REG(NOM_PARAMETERS_3)) + REG_GET(NOM_PARAMETERS_3, + REFCYC_PER_PTE_GROUP_NOM_C, &dlg_attr->refcyc_per_pte_group_nom_c); + + REG_GET(NOM_PARAMETERS_6, + DST_Y_PER_META_ROW_NOM_C, &dlg_attr->dst_y_per_meta_row_nom_c); + + REG_GET(NOM_PARAMETERS_7, + REFCYC_PER_META_CHUNK_NOM_C, &dlg_attr->refcyc_per_meta_chunk_nom_c); + + /* TTU - per hubp */ + REG_GET_2(DCN_TTU_QOS_WM, + QoS_LEVEL_LOW_WM, &ttu_attr->qos_level_low_wm, + QoS_LEVEL_HIGH_WM, &ttu_attr->qos_level_high_wm); + + REG_GET_2(DCN_GLOBAL_TTU_CNTL, + MIN_TTU_VBLANK, &ttu_attr->min_ttu_vblank, + QoS_LEVEL_FLIP, &ttu_attr->qos_level_flip); + + /* TTU - per luma/chroma */ + /* Assumed surf0 is luma and 1 is chroma */ + + REG_GET_3(DCN_SURF0_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr->refcyc_per_req_delivery_l, + QoS_LEVEL_FIXED, &ttu_attr->qos_level_fixed_l, + QoS_RAMP_DISABLE, &ttu_attr->qos_ramp_disable_l); + + REG_GET(DCN_SURF0_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, + &ttu_attr->refcyc_per_req_delivery_pre_l); + + REG_GET_3(DCN_SURF1_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr->refcyc_per_req_delivery_c, + QoS_LEVEL_FIXED, &ttu_attr->qos_level_fixed_c, + QoS_RAMP_DISABLE, &ttu_attr->qos_ramp_disable_c); + + REG_GET(DCN_SURF1_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, + &ttu_attr->refcyc_per_req_delivery_pre_c); + + /* Rest of hubp */ + REG_GET(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, &s->pixel_format); + + REG_GET(DCSURF_SURFACE_EARLIEST_INUSE_HIGH, + SURFACE_EARLIEST_INUSE_ADDRESS_HIGH, &s->inuse_addr_hi); + + REG_GET(DCSURF_SURFACE_EARLIEST_INUSE, + SURFACE_EARLIEST_INUSE_ADDRESS, &s->inuse_addr_lo); + + REG_GET_2(DCSURF_PRI_VIEWPORT_DIMENSION, + PRI_VIEWPORT_WIDTH, &s->viewport_width, + PRI_VIEWPORT_HEIGHT, &s->viewport_height); + + REG_GET_2(DCSURF_SURFACE_CONFIG, + ROTATION_ANGLE, &s->rotation_angle, + H_MIRROR_EN, &s->h_mirror_en); + + REG_GET(DCSURF_TILING_CONFIG, + SW_MODE, &s->sw_mode); + + REG_GET(DCSURF_SURFACE_CONTROL, + PRIMARY_SURFACE_DCC_EN, &s->dcc_en); + + REG_GET_3(DCHUBP_CNTL, + HUBP_BLANK_EN, &s->blank_en, + HUBP_TTU_DISABLE, &s->ttu_disable, + HUBP_UNDERFLOW_STATUS, &s->underflow_status); + + REG_GET(HUBP_CLK_CNTL, + HUBP_CLOCK_ENABLE, &s->clock_en); + + REG_GET(DCN_GLOBAL_TTU_CNTL, + MIN_TTU_VBLANK, &s->min_ttu_vblank); + + REG_GET_2(DCN_TTU_QOS_WM, + QoS_LEVEL_LOW_WM, &s->qos_level_low_wm, + QoS_LEVEL_HIGH_WM, &s->qos_level_high_wm); + +} + +void hubp2_read_state(struct hubp *hubp) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + struct dcn_hubp_state *s = &hubp2->state; + struct _vcs_dpi_display_rq_regs_st *rq_regs = &s->rq_regs; + + hubp2_read_state_common(hubp); + + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG, + CHUNK_SIZE, &rq_regs->rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, &rq_regs->rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, &rq_regs->rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, &rq_regs->rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, &rq_regs->rq_regs_l.dpte_group_size, + MPTE_GROUP_SIZE, &rq_regs->rq_regs_l.mpte_group_size, + SWATH_HEIGHT, &rq_regs->rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, &rq_regs->rq_regs_l.pte_row_height_linear); + + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG_C, + CHUNK_SIZE_C, &rq_regs->rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, &rq_regs->rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.dpte_group_size, + MPTE_GROUP_SIZE_C, &rq_regs->rq_regs_c.mpte_group_size, + SWATH_HEIGHT_C, &rq_regs->rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, &rq_regs->rq_regs_c.pte_row_height_linear); + +} + +void hubp2_validate_dml_output(struct hubp *hubp, + struct dc_context *ctx, + struct _vcs_dpi_display_rq_regs_st *dml_rq_regs, + struct _vcs_dpi_display_dlg_regs_st *dml_dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *dml_ttu_attr) +{ + struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); + struct _vcs_dpi_display_rq_regs_st rq_regs = {0}; + struct _vcs_dpi_display_dlg_regs_st dlg_attr = {0}; + struct _vcs_dpi_display_ttu_regs_st ttu_attr = {0}; + DC_LOGGER_INIT(ctx->logger); + DC_LOG_DEBUG("DML Validation | Running Validation"); + + /* Requestor Regs */ + REG_GET(HUBPRET_CONTROL, + DET_BUF_PLANE1_BASE_ADDRESS, &rq_regs.plane1_base_address); + REG_GET_4(DCN_EXPANSION_MODE, + DRQ_EXPANSION_MODE, &rq_regs.drq_expansion_mode, + PRQ_EXPANSION_MODE, &rq_regs.prq_expansion_mode, + MRQ_EXPANSION_MODE, &rq_regs.mrq_expansion_mode, + CRQ_EXPANSION_MODE, &rq_regs.crq_expansion_mode); + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG, + CHUNK_SIZE, &rq_regs.rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, &rq_regs.rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, &rq_regs.rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, &rq_regs.rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, &rq_regs.rq_regs_l.dpte_group_size, + MPTE_GROUP_SIZE, &rq_regs.rq_regs_l.mpte_group_size, + SWATH_HEIGHT, &rq_regs.rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, &rq_regs.rq_regs_l.pte_row_height_linear); + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG_C, + CHUNK_SIZE_C, &rq_regs.rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, &rq_regs.rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, &rq_regs.rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, &rq_regs.rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, &rq_regs.rq_regs_c.dpte_group_size, + MPTE_GROUP_SIZE_C, &rq_regs.rq_regs_c.mpte_group_size, + SWATH_HEIGHT_C, &rq_regs.rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, &rq_regs.rq_regs_c.pte_row_height_linear); + + if (rq_regs.plane1_base_address != dml_rq_regs->plane1_base_address) + DC_LOG_DEBUG("DML Validation | HUBPRET_CONTROL:DET_BUF_PLANE1_BASE_ADDRESS - Expected: %u Actual: %u\n", + dml_rq_regs->plane1_base_address, rq_regs.plane1_base_address); + if (rq_regs.drq_expansion_mode != dml_rq_regs->drq_expansion_mode) + DC_LOG_DEBUG("DML Validation | DCN_EXPANSION_MODE:DRQ_EXPANSION_MODE - Expected: %u Actual: %u\n", + dml_rq_regs->drq_expansion_mode, rq_regs.drq_expansion_mode); + if (rq_regs.prq_expansion_mode != dml_rq_regs->prq_expansion_mode) + DC_LOG_DEBUG("DML Validation | DCN_EXPANSION_MODE:MRQ_EXPANSION_MODE - Expected: %u Actual: %u\n", + dml_rq_regs->prq_expansion_mode, rq_regs.prq_expansion_mode); + if (rq_regs.mrq_expansion_mode != dml_rq_regs->mrq_expansion_mode) + DC_LOG_DEBUG("DML Validation | DCN_EXPANSION_MODE:DET_BUF_PLANE1_BASE_ADDRESS - Expected: %u Actual: %u\n", + dml_rq_regs->mrq_expansion_mode, rq_regs.mrq_expansion_mode); + if (rq_regs.crq_expansion_mode != dml_rq_regs->crq_expansion_mode) + DC_LOG_DEBUG("DML Validation | DCN_EXPANSION_MODE:CRQ_EXPANSION_MODE - Expected: %u Actual: %u\n", + dml_rq_regs->crq_expansion_mode, rq_regs.crq_expansion_mode); + + if (rq_regs.rq_regs_l.chunk_size != dml_rq_regs->rq_regs_l.chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:CHUNK_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.chunk_size, rq_regs.rq_regs_l.chunk_size); + if (rq_regs.rq_regs_l.min_chunk_size != dml_rq_regs->rq_regs_l.min_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:MIN_CHUNK_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.min_chunk_size, rq_regs.rq_regs_l.min_chunk_size); + if (rq_regs.rq_regs_l.meta_chunk_size != dml_rq_regs->rq_regs_l.meta_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:META_CHUNK_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.meta_chunk_size, rq_regs.rq_regs_l.meta_chunk_size); + if (rq_regs.rq_regs_l.min_meta_chunk_size != dml_rq_regs->rq_regs_l.min_meta_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:MIN_META_CHUNK_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.min_meta_chunk_size, rq_regs.rq_regs_l.min_meta_chunk_size); + if (rq_regs.rq_regs_l.dpte_group_size != dml_rq_regs->rq_regs_l.dpte_group_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:DPTE_GROUP_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.dpte_group_size, rq_regs.rq_regs_l.dpte_group_size); + if (rq_regs.rq_regs_l.mpte_group_size != dml_rq_regs->rq_regs_l.mpte_group_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:MPTE_GROUP_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.mpte_group_size, rq_regs.rq_regs_l.mpte_group_size); + if (rq_regs.rq_regs_l.swath_height != dml_rq_regs->rq_regs_l.swath_height) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:SWATH_HEIGHT - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.swath_height, rq_regs.rq_regs_l.swath_height); + if (rq_regs.rq_regs_l.pte_row_height_linear != dml_rq_regs->rq_regs_l.pte_row_height_linear) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:PTE_ROW_HEIGHT_LINEAR - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.pte_row_height_linear, rq_regs.rq_regs_l.pte_row_height_linear); + + if (rq_regs.rq_regs_c.chunk_size != dml_rq_regs->rq_regs_c.chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:CHUNK_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.chunk_size, rq_regs.rq_regs_c.chunk_size); + if (rq_regs.rq_regs_c.min_chunk_size != dml_rq_regs->rq_regs_c.min_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:MIN_CHUNK_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.min_chunk_size, rq_regs.rq_regs_c.min_chunk_size); + if (rq_regs.rq_regs_c.meta_chunk_size != dml_rq_regs->rq_regs_c.meta_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:META_CHUNK_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.meta_chunk_size, rq_regs.rq_regs_c.meta_chunk_size); + if (rq_regs.rq_regs_c.min_meta_chunk_size != dml_rq_regs->rq_regs_c.min_meta_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:MIN_META_CHUNK_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.min_meta_chunk_size, rq_regs.rq_regs_c.min_meta_chunk_size); + if (rq_regs.rq_regs_c.dpte_group_size != dml_rq_regs->rq_regs_c.dpte_group_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:DPTE_GROUP_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.dpte_group_size, rq_regs.rq_regs_c.dpte_group_size); + if (rq_regs.rq_regs_c.mpte_group_size != dml_rq_regs->rq_regs_c.mpte_group_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:MPTE_GROUP_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.mpte_group_size, rq_regs.rq_regs_c.mpte_group_size); + if (rq_regs.rq_regs_c.swath_height != dml_rq_regs->rq_regs_c.swath_height) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:SWATH_HEIGHT_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.swath_height, rq_regs.rq_regs_c.swath_height); + if (rq_regs.rq_regs_c.pte_row_height_linear != dml_rq_regs->rq_regs_c.pte_row_height_linear) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:PTE_ROW_HEIGHT_LINEAR_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.pte_row_height_linear, rq_regs.rq_regs_c.pte_row_height_linear); + + /* DLG - Per hubp */ + REG_GET_2(BLANK_OFFSET_0, + REFCYC_H_BLANK_END, &dlg_attr.refcyc_h_blank_end, + DLG_V_BLANK_END, &dlg_attr.dlg_vblank_end); + REG_GET(BLANK_OFFSET_1, + MIN_DST_Y_NEXT_START, &dlg_attr.min_dst_y_next_start); + REG_GET(DST_DIMENSIONS, + REFCYC_PER_HTOTAL, &dlg_attr.refcyc_per_htotal); + REG_GET_2(DST_AFTER_SCALER, + REFCYC_X_AFTER_SCALER, &dlg_attr.refcyc_x_after_scaler, + DST_Y_AFTER_SCALER, &dlg_attr.dst_y_after_scaler); + REG_GET(REF_FREQ_TO_PIX_FREQ, + REF_FREQ_TO_PIX_FREQ, &dlg_attr.ref_freq_to_pix_freq); + + if (dlg_attr.refcyc_h_blank_end != dml_dlg_attr->refcyc_h_blank_end) + DC_LOG_DEBUG("DML Validation | BLANK_OFFSET_0:REFCYC_H_BLANK_END - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_h_blank_end, dlg_attr.refcyc_h_blank_end); + if (dlg_attr.dlg_vblank_end != dml_dlg_attr->dlg_vblank_end) + DC_LOG_DEBUG("DML Validation | BLANK_OFFSET_0:DLG_V_BLANK_END - Expected: %u Actual: %u\n", + dml_dlg_attr->dlg_vblank_end, dlg_attr.dlg_vblank_end); + if (dlg_attr.min_dst_y_next_start != dml_dlg_attr->min_dst_y_next_start) + DC_LOG_DEBUG("DML Validation | BLANK_OFFSET_1:MIN_DST_Y_NEXT_START - Expected: %u Actual: %u\n", + dml_dlg_attr->min_dst_y_next_start, dlg_attr.min_dst_y_next_start); + if (dlg_attr.refcyc_per_htotal != dml_dlg_attr->refcyc_per_htotal) + DC_LOG_DEBUG("DML Validation | DST_DIMENSIONS:REFCYC_PER_HTOTAL - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_htotal, dlg_attr.refcyc_per_htotal); + if (dlg_attr.refcyc_x_after_scaler != dml_dlg_attr->refcyc_x_after_scaler) + DC_LOG_DEBUG("DML Validation | DST_AFTER_SCALER:REFCYC_X_AFTER_SCALER - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_x_after_scaler, dlg_attr.refcyc_x_after_scaler); + if (dlg_attr.dst_y_after_scaler != dml_dlg_attr->dst_y_after_scaler) + DC_LOG_DEBUG("DML Validation | DST_AFTER_SCALER:DST_Y_AFTER_SCALER - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_after_scaler, dlg_attr.dst_y_after_scaler); + if (dlg_attr.ref_freq_to_pix_freq != dml_dlg_attr->ref_freq_to_pix_freq) + DC_LOG_DEBUG("DML Validation | REF_FREQ_TO_PIX_FREQ:REF_FREQ_TO_PIX_FREQ - Expected: %u Actual: %u\n", + dml_dlg_attr->ref_freq_to_pix_freq, dlg_attr.ref_freq_to_pix_freq); + + /* DLG - Per luma/chroma */ + REG_GET(VBLANK_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_VBLANK_L, &dlg_attr.refcyc_per_pte_group_vblank_l); + if (REG(NOM_PARAMETERS_0)) + REG_GET(NOM_PARAMETERS_0, + DST_Y_PER_PTE_ROW_NOM_L, &dlg_attr.dst_y_per_pte_row_nom_l); + if (REG(NOM_PARAMETERS_1)) + REG_GET(NOM_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_NOM_L, &dlg_attr.refcyc_per_pte_group_nom_l); + REG_GET(NOM_PARAMETERS_4, + DST_Y_PER_META_ROW_NOM_L, &dlg_attr.dst_y_per_meta_row_nom_l); + REG_GET(NOM_PARAMETERS_5, + REFCYC_PER_META_CHUNK_NOM_L, &dlg_attr.refcyc_per_meta_chunk_nom_l); + REG_GET_2(PER_LINE_DELIVERY, + REFCYC_PER_LINE_DELIVERY_L, &dlg_attr.refcyc_per_line_delivery_l, + REFCYC_PER_LINE_DELIVERY_C, &dlg_attr.refcyc_per_line_delivery_c); + REG_GET_2(PER_LINE_DELIVERY_PRE, + REFCYC_PER_LINE_DELIVERY_PRE_L, &dlg_attr.refcyc_per_line_delivery_pre_l, + REFCYC_PER_LINE_DELIVERY_PRE_C, &dlg_attr.refcyc_per_line_delivery_pre_c); + REG_GET(VBLANK_PARAMETERS_2, + REFCYC_PER_PTE_GROUP_VBLANK_C, &dlg_attr.refcyc_per_pte_group_vblank_c); + if (REG(NOM_PARAMETERS_2)) + REG_GET(NOM_PARAMETERS_2, + DST_Y_PER_PTE_ROW_NOM_C, &dlg_attr.dst_y_per_pte_row_nom_c); + if (REG(NOM_PARAMETERS_3)) + REG_GET(NOM_PARAMETERS_3, + REFCYC_PER_PTE_GROUP_NOM_C, &dlg_attr.refcyc_per_pte_group_nom_c); + REG_GET(NOM_PARAMETERS_6, + DST_Y_PER_META_ROW_NOM_C, &dlg_attr.dst_y_per_meta_row_nom_c); + REG_GET(NOM_PARAMETERS_7, + REFCYC_PER_META_CHUNK_NOM_C, &dlg_attr.refcyc_per_meta_chunk_nom_c); + REG_GET(VBLANK_PARAMETERS_3, + REFCYC_PER_META_CHUNK_VBLANK_L, &dlg_attr.refcyc_per_meta_chunk_vblank_l); + REG_GET(VBLANK_PARAMETERS_4, + REFCYC_PER_META_CHUNK_VBLANK_C, &dlg_attr.refcyc_per_meta_chunk_vblank_c); + + if (dlg_attr.refcyc_per_pte_group_vblank_l != dml_dlg_attr->refcyc_per_pte_group_vblank_l) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_1:REFCYC_PER_PTE_GROUP_VBLANK_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_vblank_l, dlg_attr.refcyc_per_pte_group_vblank_l); + if (dlg_attr.dst_y_per_pte_row_nom_l != dml_dlg_attr->dst_y_per_pte_row_nom_l) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_0:DST_Y_PER_PTE_ROW_NOM_L - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_per_pte_row_nom_l, dlg_attr.dst_y_per_pte_row_nom_l); + if (dlg_attr.refcyc_per_pte_group_nom_l != dml_dlg_attr->refcyc_per_pte_group_nom_l) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_1:REFCYC_PER_PTE_GROUP_NOM_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_nom_l, dlg_attr.refcyc_per_pte_group_nom_l); + if (dlg_attr.dst_y_per_meta_row_nom_l != dml_dlg_attr->dst_y_per_meta_row_nom_l) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_4:DST_Y_PER_META_ROW_NOM_L - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_per_meta_row_nom_l, dlg_attr.dst_y_per_meta_row_nom_l); + if (dlg_attr.refcyc_per_meta_chunk_nom_l != dml_dlg_attr->refcyc_per_meta_chunk_nom_l) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_5:REFCYC_PER_META_CHUNK_NOM_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_nom_l, dlg_attr.refcyc_per_meta_chunk_nom_l); + if (dlg_attr.refcyc_per_line_delivery_l != dml_dlg_attr->refcyc_per_line_delivery_l) + DC_LOG_DEBUG("DML Validation | PER_LINE_DELIVERY:REFCYC_PER_LINE_DELIVERY_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_line_delivery_l, dlg_attr.refcyc_per_line_delivery_l); + if (dlg_attr.refcyc_per_line_delivery_c != dml_dlg_attr->refcyc_per_line_delivery_c) + DC_LOG_DEBUG("DML Validation | PER_LINE_DELIVERY:REFCYC_PER_LINE_DELIVERY_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_line_delivery_c, dlg_attr.refcyc_per_line_delivery_c); + if (dlg_attr.refcyc_per_pte_group_vblank_c != dml_dlg_attr->refcyc_per_pte_group_vblank_c) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_2:REFCYC_PER_PTE_GROUP_VBLANK_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_vblank_c, dlg_attr.refcyc_per_pte_group_vblank_c); + if (dlg_attr.dst_y_per_pte_row_nom_c != dml_dlg_attr->dst_y_per_pte_row_nom_c) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_2:DST_Y_PER_PTE_ROW_NOM_C - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_per_pte_row_nom_c, dlg_attr.dst_y_per_pte_row_nom_c); + if (dlg_attr.refcyc_per_pte_group_nom_c != dml_dlg_attr->refcyc_per_pte_group_nom_c) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_3:REFCYC_PER_PTE_GROUP_NOM_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_nom_c, dlg_attr.refcyc_per_pte_group_nom_c); + if (dlg_attr.dst_y_per_meta_row_nom_c != dml_dlg_attr->dst_y_per_meta_row_nom_c) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_6:DST_Y_PER_META_ROW_NOM_C - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_per_meta_row_nom_c, dlg_attr.dst_y_per_meta_row_nom_c); + if (dlg_attr.refcyc_per_meta_chunk_nom_c != dml_dlg_attr->refcyc_per_meta_chunk_nom_c) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_7:REFCYC_PER_META_CHUNK_NOM_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_nom_c, dlg_attr.refcyc_per_meta_chunk_nom_c); + if (dlg_attr.refcyc_per_line_delivery_pre_l != dml_dlg_attr->refcyc_per_line_delivery_pre_l) + DC_LOG_DEBUG("DML Validation | PER_LINE_DELIVERY_PRE:REFCYC_PER_LINE_DELIVERY_PRE_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_line_delivery_pre_l, dlg_attr.refcyc_per_line_delivery_pre_l); + if (dlg_attr.refcyc_per_line_delivery_pre_c != dml_dlg_attr->refcyc_per_line_delivery_pre_c) + DC_LOG_DEBUG("DML Validation | PER_LINE_DELIVERY_PRE:REFCYC_PER_LINE_DELIVERY_PRE_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_line_delivery_pre_c, dlg_attr.refcyc_per_line_delivery_pre_c); + if (dlg_attr.refcyc_per_meta_chunk_vblank_l != dml_dlg_attr->refcyc_per_meta_chunk_vblank_l) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_3:REFCYC_PER_META_CHUNK_VBLANK_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_vblank_l, dlg_attr.refcyc_per_meta_chunk_vblank_l); + if (dlg_attr.refcyc_per_meta_chunk_vblank_c != dml_dlg_attr->refcyc_per_meta_chunk_vblank_c) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_4:REFCYC_PER_META_CHUNK_VBLANK_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_vblank_c, dlg_attr.refcyc_per_meta_chunk_vblank_c); + + /* TTU - per hubp */ + REG_GET_2(DCN_TTU_QOS_WM, + QoS_LEVEL_LOW_WM, &ttu_attr.qos_level_low_wm, + QoS_LEVEL_HIGH_WM, &ttu_attr.qos_level_high_wm); + + if (ttu_attr.qos_level_low_wm != dml_ttu_attr->qos_level_low_wm) + DC_LOG_DEBUG("DML Validation | DCN_TTU_QOS_WM:QoS_LEVEL_LOW_WM - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_low_wm, ttu_attr.qos_level_low_wm); + if (ttu_attr.qos_level_high_wm != dml_ttu_attr->qos_level_high_wm) + DC_LOG_DEBUG("DML Validation | DCN_TTU_QOS_WM:QoS_LEVEL_HIGH_WM - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_high_wm, ttu_attr.qos_level_high_wm); + + /* TTU - per luma/chroma */ + /* Assumed surf0 is luma and 1 is chroma */ + REG_GET_3(DCN_SURF0_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr.refcyc_per_req_delivery_l, + QoS_LEVEL_FIXED, &ttu_attr.qos_level_fixed_l, + QoS_RAMP_DISABLE, &ttu_attr.qos_ramp_disable_l); + REG_GET_3(DCN_SURF1_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr.refcyc_per_req_delivery_c, + QoS_LEVEL_FIXED, &ttu_attr.qos_level_fixed_c, + QoS_RAMP_DISABLE, &ttu_attr.qos_ramp_disable_c); + REG_GET_3(DCN_CUR0_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr.refcyc_per_req_delivery_cur0, + QoS_LEVEL_FIXED, &ttu_attr.qos_level_fixed_cur0, + QoS_RAMP_DISABLE, &ttu_attr.qos_ramp_disable_cur0); + REG_GET(FLIP_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_FLIP_L, &dlg_attr.refcyc_per_pte_group_flip_l); + REG_GET(DCN_CUR0_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, &ttu_attr.refcyc_per_req_delivery_pre_cur0); + REG_GET(DCN_CUR1_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, &ttu_attr.refcyc_per_req_delivery_pre_cur1); + REG_GET(DCN_SURF0_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, &ttu_attr.refcyc_per_req_delivery_pre_l); + REG_GET(DCN_SURF1_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, &ttu_attr.refcyc_per_req_delivery_pre_c); + + if (ttu_attr.refcyc_per_req_delivery_l != dml_ttu_attr->refcyc_per_req_delivery_l) + DC_LOG_DEBUG("DML Validation | DCN_SURF0_TTU_CNTL0:REFCYC_PER_REQ_DELIVERY - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_l, ttu_attr.refcyc_per_req_delivery_l); + if (ttu_attr.qos_level_fixed_l != dml_ttu_attr->qos_level_fixed_l) + DC_LOG_DEBUG("DML Validation | DCN_SURF0_TTU_CNTL0:QoS_LEVEL_FIXED - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_fixed_l, ttu_attr.qos_level_fixed_l); + if (ttu_attr.qos_ramp_disable_l != dml_ttu_attr->qos_ramp_disable_l) + DC_LOG_DEBUG("DML Validation | DCN_SURF0_TTU_CNTL0:QoS_RAMP_DISABLE - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_ramp_disable_l, ttu_attr.qos_ramp_disable_l); + if (ttu_attr.refcyc_per_req_delivery_c != dml_ttu_attr->refcyc_per_req_delivery_c) + DC_LOG_DEBUG("DML Validation | DCN_SURF1_TTU_CNTL0:REFCYC_PER_REQ_DELIVERY - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_c, ttu_attr.refcyc_per_req_delivery_c); + if (ttu_attr.qos_level_fixed_c != dml_ttu_attr->qos_level_fixed_c) + DC_LOG_DEBUG("DML Validation | DCN_SURF1_TTU_CNTL0:QoS_LEVEL_FIXED - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_fixed_c, ttu_attr.qos_level_fixed_c); + if (ttu_attr.qos_ramp_disable_c != dml_ttu_attr->qos_ramp_disable_c) + DC_LOG_DEBUG("DML Validation | DCN_SURF1_TTU_CNTL0:QoS_RAMP_DISABLE - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_ramp_disable_c, ttu_attr.qos_ramp_disable_c); + if (ttu_attr.refcyc_per_req_delivery_cur0 != dml_ttu_attr->refcyc_per_req_delivery_cur0) + DC_LOG_DEBUG("DML Validation | DCN_CUR0_TTU_CNTL0:REFCYC_PER_REQ_DELIVERY - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_cur0, ttu_attr.refcyc_per_req_delivery_cur0); + if (ttu_attr.qos_level_fixed_cur0 != dml_ttu_attr->qos_level_fixed_cur0) + DC_LOG_DEBUG("DML Validation | DCN_CUR0_TTU_CNTL0:QoS_LEVEL_FIXED - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_fixed_cur0, ttu_attr.qos_level_fixed_cur0); + if (ttu_attr.qos_ramp_disable_cur0 != dml_ttu_attr->qos_ramp_disable_cur0) + DC_LOG_DEBUG("DML Validation | DCN_CUR0_TTU_CNTL0:QoS_RAMP_DISABLE - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_ramp_disable_cur0, ttu_attr.qos_ramp_disable_cur0); + if (dlg_attr.refcyc_per_pte_group_flip_l != dml_dlg_attr->refcyc_per_pte_group_flip_l) + DC_LOG_DEBUG("DML Validation | FLIP_PARAMETERS_1:REFCYC_PER_PTE_GROUP_FLIP_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_flip_l, dlg_attr.refcyc_per_pte_group_flip_l); + if (ttu_attr.refcyc_per_req_delivery_pre_cur0 != dml_ttu_attr->refcyc_per_req_delivery_pre_cur0) + DC_LOG_DEBUG("DML Validation | DCN_CUR0_TTU_CNTL1:REFCYC_PER_REQ_DELIVERY_PRE - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_pre_cur0, ttu_attr.refcyc_per_req_delivery_pre_cur0); + if (ttu_attr.refcyc_per_req_delivery_pre_cur1 != dml_ttu_attr->refcyc_per_req_delivery_pre_cur1) + DC_LOG_DEBUG("DML Validation | DCN_CUR1_TTU_CNTL1:REFCYC_PER_REQ_DELIVERY_PRE - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_pre_cur1, ttu_attr.refcyc_per_req_delivery_pre_cur1); + if (ttu_attr.refcyc_per_req_delivery_pre_l != dml_ttu_attr->refcyc_per_req_delivery_pre_l) + DC_LOG_DEBUG("DML Validation | DCN_SURF0_TTU_CNTL1:REFCYC_PER_REQ_DELIVERY_PRE - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_pre_l, ttu_attr.refcyc_per_req_delivery_pre_l); + if (ttu_attr.refcyc_per_req_delivery_pre_c != dml_ttu_attr->refcyc_per_req_delivery_pre_c) + DC_LOG_DEBUG("DML Validation | DCN_SURF1_TTU_CNTL1:REFCYC_PER_REQ_DELIVERY_PRE - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_pre_c, ttu_attr.refcyc_per_req_delivery_pre_c); +} + static struct hubp_funcs dcn20_hubp_funcs = { .hubp_enable_tripleBuffer = hubp2_enable_triplebuffer, .hubp_is_triplebuffer_enabled = hubp2_is_triplebuffer_enabled, .hubp_program_surface_flip_and_addr = hubp2_program_surface_flip_and_addr, .hubp_program_surface_config = hubp2_program_surface_config, - .hubp_is_flip_pending = hubp1_is_flip_pending, + .hubp_is_flip_pending = hubp2_is_flip_pending, .hubp_setup = hubp2_setup, .hubp_setup_interdependent = hubp2_setup_interdependent, .hubp_set_vm_system_aperture_settings = hubp2_set_vm_system_aperture_settings, - .set_blank = hubp1_set_blank, - .dcc_control = hubp1_dcc_control, - .hubp_update_dchub = hubp2_update_dchub, + .set_blank = hubp2_set_blank, + .dcc_control = hubp2_dcc_control, .mem_program_viewport = min_set_viewport, .set_cursor_attributes = hubp2_cursor_set_attributes, - .set_cursor_position = hubp1_cursor_set_position, - .hubp_clk_cntl = hubp1_clk_cntl, - .hubp_vtg_sel = hubp1_vtg_sel, + .set_cursor_position = hubp2_cursor_set_position, + .hubp_clk_cntl = hubp2_clk_cntl, + .hubp_vtg_sel = hubp2_vtg_sel, .dmdata_set_attributes = hubp2_dmdata_set_attributes, .dmdata_load = hubp2_dmdata_load, .dmdata_status_done = hubp2_dmdata_status_done, - .hubp_read_state = hubp1_read_state, - .hubp_clear_underflow = hubp1_clear_underflow, + .hubp_read_state = hubp2_read_state, + .hubp_clear_underflow = hubp2_clear_underflow, .hubp_set_flip_control_surface_gsl = hubp2_set_flip_control_surface_gsl, .hubp_init = hubp1_init, + .validate_dml_output = hubp2_validate_dml_output, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h index d5acc348be22..8c04a3606a54 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h @@ -38,12 +38,6 @@ SRI(PREFETCH_SETTINGS_C, HUBPREQ, id),\ SRI(DCN_VM_SYSTEM_APERTURE_LOW_ADDR, HUBPREQ, id),\ SRI(DCN_VM_SYSTEM_APERTURE_HIGH_ADDR, HUBPREQ, id),\ - SR(DCN_VM_FB_LOCATION_TOP),\ - SR(DCN_VM_FB_LOCATION_BASE),\ - SR(DCN_VM_FB_OFFSET),\ - SR(DCN_VM_AGP_BASE),\ - SR(DCN_VM_AGP_BOT),\ - SR(DCN_VM_AGP_TOP),\ SRI(CURSOR_SETTINGS, HUBPREQ, id), \ SRI(CURSOR_SURFACE_ADDRESS_HIGH, CURSOR0_, id), \ SRI(CURSOR_SURFACE_ADDRESS, CURSOR0_, id), \ @@ -72,8 +66,8 @@ SR(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB),\ SR(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB) -#define HUBP_MASK_SH_LIST_DCN2_COMMON(mask_sh)\ - HUBP_MASK_SH_LIST_DCN(mask_sh),\ +#define HUBP_MASK_SH_LIST_DCN2_SHARE_COMMON(mask_sh)\ + HUBP_MASK_SH_LIST_DCN_SHARE_COMMON(mask_sh),\ HUBP_MASK_SH_LIST_DCN_VM(mask_sh),\ HUBP_SF(HUBP0_DCSURF_SURFACE_CONFIG, ROTATION_ANGLE, mask_sh),\ HUBP_SF(HUBP0_DCSURF_SURFACE_CONFIG, H_MIRROR_EN, mask_sh),\ @@ -82,12 +76,6 @@ HUBP_SF(HUBPREQ0_PREFETCH_SETTINGS_C, VRATIO_PREFETCH_C, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_LOW_ADDR, MC_VM_SYSTEM_APERTURE_LOW_ADDR, mask_sh),\ HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_HIGH_ADDR, MC_VM_SYSTEM_APERTURE_HIGH_ADDR, mask_sh),\ - HUBP_SF(DCN_VM_FB_LOCATION_TOP, FB_TOP, mask_sh),\ - HUBP_SF(DCN_VM_FB_LOCATION_BASE, FB_BASE, mask_sh),\ - HUBP_SF(DCN_VM_FB_OFFSET, FB_OFFSET, mask_sh),\ - HUBP_SF(DCN_VM_AGP_BASE, AGP_BASE, mask_sh),\ - HUBP_SF(DCN_VM_AGP_BOT, AGP_BOT, mask_sh),\ - HUBP_SF(DCN_VM_AGP_TOP, AGP_TOP, mask_sh),\ HUBP_SF(HUBPREQ0_CURSOR_SETTINGS, CURSOR0_DST_Y_OFFSET, mask_sh), \ HUBP_SF(HUBPREQ0_CURSOR_SETTINGS, CURSOR0_CHUNK_HDL_ADJUST, mask_sh), \ HUBP_SF(CURSOR0_0_CURSOR_SURFACE_ADDRESS_HIGH, CURSOR_SURFACE_ADDRESS_HIGH, mask_sh), \ @@ -127,13 +115,21 @@ HUBP_SF(HUBPREQ0_DCSURF_FLIP_CONTROL2, SURFACE_TRIPLE_BUFFER_ENABLE, mask_sh),\ HUBP_SF(HUBPREQ0_VMID_SETTINGS_0, VMID, mask_sh) +/*DCN2.x and DCN1.x*/ +#define HUBP_MASK_SH_LIST_DCN2_COMMON(mask_sh)\ + HUBP_MASK_SH_LIST_DCN2_SHARE_COMMON(mask_sh),\ + HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, RB_ALIGNED, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, MPTE_GROUP_SIZE, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG_C, MPTE_GROUP_SIZE_C, mask_sh) + +/*DCN2.0 specific*/ #define HUBP_MASK_SH_LIST_DCN20(mask_sh)\ HUBP_MASK_SH_LIST_DCN2_COMMON(mask_sh),\ HUBP_SF(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, DCN_VM_SYSTEM_APERTURE_DEFAULT_SYSTEM, mask_sh),\ HUBP_SF(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, mask_sh),\ HUBP_SF(DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, mask_sh) - +/*DCN2.x */ #define DCN2_HUBP_REG_COMMON_VARIABLE_LIST \ HUBP_COMMON_REG_VARIABLE_LIST; \ uint32_t DMDATA_ADDRESS_HIGH; \ @@ -149,14 +145,20 @@ uint32_t FLIP_PARAMETERS_2;\ uint32_t DCN_CUR1_TTU_CNTL0;\ uint32_t DCN_CUR1_TTU_CNTL1;\ - uint32_t VMID_SETTINGS_0;\ + uint32_t VMID_SETTINGS_0 + + +#define DCN21_HUBP_REG_COMMON_VARIABLE_LIST \ + DCN2_HUBP_REG_COMMON_VARIABLE_LIST; \ uint32_t FLIP_PARAMETERS_3;\ uint32_t FLIP_PARAMETERS_4;\ + uint32_t FLIP_PARAMETERS_5;\ + uint32_t FLIP_PARAMETERS_6;\ uint32_t VBLANK_PARAMETERS_5;\ uint32_t VBLANK_PARAMETERS_6 #define DCN2_HUBP_REG_FIELD_VARIABLE_LIST(type) \ - DCN_HUBP_REG_FIELD_LIST(type); \ + DCN_HUBP_REG_FIELD_BASE_LIST(type); \ type DMDATA_ADDRESS_HIGH;\ type DMDATA_MODE;\ type DMDATA_UPDATED;\ @@ -180,17 +182,27 @@ type SURFACE_TRIPLE_BUFFER_ENABLE;\ type VMID +#define DCN21_HUBP_REG_FIELD_VARIABLE_LIST(type) \ + DCN2_HUBP_REG_FIELD_VARIABLE_LIST(type);\ + type REFCYC_PER_VM_GROUP_FLIP;\ + type REFCYC_PER_VM_REQ_FLIP;\ + type REFCYC_PER_VM_GROUP_VBLANK;\ + type REFCYC_PER_VM_REQ_VBLANK;\ + type REFCYC_PER_PTE_GROUP_FLIP_C; \ + type REFCYC_PER_META_CHUNK_FLIP_C; \ + type VM_GROUP_SIZE + struct dcn_hubp2_registers { - DCN2_HUBP_REG_COMMON_VARIABLE_LIST; + DCN21_HUBP_REG_COMMON_VARIABLE_LIST; }; struct dcn_hubp2_shift { - DCN2_HUBP_REG_FIELD_VARIABLE_LIST(uint8_t); + DCN21_HUBP_REG_FIELD_VARIABLE_LIST(uint8_t); }; struct dcn_hubp2_mask { - DCN2_HUBP_REG_FIELD_VARIABLE_LIST(uint32_t); + DCN21_HUBP_REG_FIELD_VARIABLE_LIST(uint32_t); }; struct dcn20_hubp { @@ -217,10 +229,6 @@ void hubp2_setup_interdependent( void hubp2_vready_at_or_After_vsync(struct hubp *hubp, struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest); -void hubp2_update_dchub( - struct hubp *hubp, - struct dchub_init_data *dh_data); - void hubp2_cursor_set_attributes( struct hubp *hubp, const struct dc_cursor_attributes *attr); @@ -262,16 +270,53 @@ bool hubp2_program_surface_flip_and_addr( const struct dc_plane_address *address, bool flip_immediate); +void hubp2_dcc_control(struct hubp *hubp, bool enable, + enum hubp_ind_block_size independent_64b_blks); + +void hubp2_program_size( + struct hubp *hubp, + enum surface_pixel_format format, + const struct plane_size *plane_size, + struct dc_plane_dcc_param *dcc); + +void hubp2_program_rotation( + struct hubp *hubp, + enum dc_rotation_angle rotation, + bool horizontal_mirror); + +void hubp2_program_pixel_format( + struct hubp *hubp, + enum surface_pixel_format format); + void hubp2_program_surface_config( struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, unsigned int compat_level); +bool hubp2_is_flip_pending(struct hubp *hubp); + +void hubp2_set_blank(struct hubp *hubp, bool blank); + +void hubp2_cursor_set_position( + struct hubp *hubp, + const struct dc_cursor_position *pos, + const struct dc_cursor_mi_param *param); + +void hubp2_clk_cntl(struct hubp *hubp, bool enable); + +void hubp2_vtg_sel(struct hubp *hubp, uint32_t otg_inst); + +void hubp2_clear_underflow(struct hubp *hubp); + +void hubp2_read_state_common(struct hubp *hubp); + +void hubp2_read_state(struct hubp *hubp); + #endif /* __DC_MEM_INPUT_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index d810c8940129..a444fed94184 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -25,17 +25,15 @@ #include <linux/delay.h> #include "dm_services.h" +#include "basics/dc_common.h" #include "dm_helpers.h" #include "core_types.h" #include "resource.h" -#include "dcn20/dcn20_resource.h" -#include "dce110/dce110_hw_sequencer.h" -#include "dcn10/dcn10_hw_sequencer.h" +#include "dcn20_resource.h" #include "dcn20_hwseq.h" #include "dce/dce_hwseq.h" -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT -#include "dcn20/dcn20_dsc.h" -#endif +#include "dcn20_dsc.h" +#include "dcn20_optc.h" #include "abm.h" #include "clk_mgr.h" #include "dmcu.h" @@ -45,10 +43,9 @@ #include "ipp.h" #include "mpc.h" #include "mcif_wb.h" +#include "dchubbub.h" #include "reg_helper.h" #include "dcn10/dcn10_cm_common.h" -#include "dcn10/dcn10_hubbub.h" -#include "dcn10/dcn10_optc.h" #include "dc_link_dp.h" #include "vm_helper.h" #include "dccg.h" @@ -64,58 +61,166 @@ #define FN(reg_name, field_name) \ hws->shifts->field_name, hws->masks->field_name -static void bios_golden_init(struct dc *dc) +static int find_free_gsl_group(const struct dc *dc) { - struct dc_bios *bp = dc->ctx->dc_bios; - int i; + if (dc->res_pool->gsl_groups.gsl_0 == 0) + return 1; + if (dc->res_pool->gsl_groups.gsl_1 == 0) + return 2; + if (dc->res_pool->gsl_groups.gsl_2 == 0) + return 3; - /* initialize dcn global */ - bp->funcs->enable_disp_power_gating(bp, - CONTROLLER_ID_D0, ASIC_PIPE_INIT); + return 0; +} - for (i = 0; i < dc->res_pool->pipe_count; i++) { - /* initialize dcn per pipe */ - bp->funcs->enable_disp_power_gating(bp, - CONTROLLER_ID_D0 + i, ASIC_PIPE_DISABLE); +/* NOTE: This is not a generic setup_gsl function (hence the suffix as_lock) + * This is only used to lock pipes in pipe splitting case with immediate flip + * Ordinary MPC/OTG locks suppress VUPDATE which doesn't help with immediate, + * so we get tearing with freesync since we cannot flip multiple pipes + * atomically. + * We use GSL for this: + * - immediate flip: find first available GSL group if not already assigned + * program gsl with that group, set current OTG as master + * and always us 0x4 = AND of flip_ready from all pipes + * - vsync flip: disable GSL if used + * + * Groups in stream_res are stored as +1 from HW registers, i.e. + * gsl_0 <=> pipe_ctx->stream_res.gsl_group == 1 + * Using a magic value like -1 would require tracking all inits/resets + */ +static void dcn20_setup_gsl_group_as_lock( + const struct dc *dc, + struct pipe_ctx *pipe_ctx, + bool enable) +{ + struct gsl_params gsl; + int group_idx; + + memset(&gsl, 0, sizeof(struct gsl_params)); + + if (enable) { + /* return if group already assigned since GSL was set up + * for vsync flip, we would unassign so it can't be "left over" + */ + if (pipe_ctx->stream_res.gsl_group > 0) + return; + + group_idx = find_free_gsl_group(dc); + ASSERT(group_idx != 0); + pipe_ctx->stream_res.gsl_group = group_idx; + + /* set gsl group reg field and mark resource used */ + switch (group_idx) { + case 1: + gsl.gsl0_en = 1; + dc->res_pool->gsl_groups.gsl_0 = 1; + break; + case 2: + gsl.gsl1_en = 1; + dc->res_pool->gsl_groups.gsl_1 = 1; + break; + case 3: + gsl.gsl2_en = 1; + dc->res_pool->gsl_groups.gsl_2 = 1; + break; + default: + BREAK_TO_DEBUGGER(); + return; // invalid case + } + gsl.gsl_master_en = 1; + } else { + group_idx = pipe_ctx->stream_res.gsl_group; + if (group_idx == 0) + return; // if not in use, just return + + pipe_ctx->stream_res.gsl_group = 0; + + /* unset gsl group reg field and mark resource free */ + switch (group_idx) { + case 1: + gsl.gsl0_en = 0; + dc->res_pool->gsl_groups.gsl_0 = 0; + break; + case 2: + gsl.gsl1_en = 0; + dc->res_pool->gsl_groups.gsl_1 = 0; + break; + case 3: + gsl.gsl2_en = 0; + dc->res_pool->gsl_groups.gsl_2 = 0; + break; + default: + BREAK_TO_DEBUGGER(); + return; + } + gsl.gsl_master_en = 0; } + + /* at this point we want to program whether it's to enable or disable */ + if (pipe_ctx->stream_res.tg->funcs->set_gsl != NULL && + pipe_ctx->stream_res.tg->funcs->set_gsl_source_select != NULL) { + pipe_ctx->stream_res.tg->funcs->set_gsl( + pipe_ctx->stream_res.tg, + &gsl); + + pipe_ctx->stream_res.tg->funcs->set_gsl_source_select( + pipe_ctx->stream_res.tg, group_idx, enable ? 4 : 0); + } else + BREAK_TO_DEBUGGER(); +} + +void dcn20_set_flip_control_gsl( + struct pipe_ctx *pipe_ctx, + bool flip_immediate) +{ + if (pipe_ctx && pipe_ctx->plane_res.hubp->funcs->hubp_set_flip_control_surface_gsl) + pipe_ctx->plane_res.hubp->funcs->hubp_set_flip_control_surface_gsl( + pipe_ctx->plane_res.hubp, flip_immediate); + } -static void enable_power_gating_plane( +void dcn20_enable_power_gating_plane( struct dce_hwseq *hws, bool enable) { - bool force_on = 1; /* disable power gating */ + bool force_on = true; /* disable power gating */ if (enable) - force_on = 0; + force_on = false; /* DCHUBP0/1/2/3/4/5 */ REG_UPDATE(DOMAIN0_PG_CONFIG, DOMAIN0_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN2_PG_CONFIG, DOMAIN2_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN4_PG_CONFIG, DOMAIN4_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN6_PG_CONFIG, DOMAIN6_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN8_PG_CONFIG, DOMAIN8_POWER_FORCEON, force_on); - /*Do not power gate DCHUB5, should be left at HW default, power on permanently*/ - /*REG_UPDATE(DOMAIN10_PG_CONFIG, DOMAIN10_POWER_FORCEON, force_on);*/ + if (REG(DOMAIN8_PG_CONFIG)) + REG_UPDATE(DOMAIN8_PG_CONFIG, DOMAIN8_POWER_FORCEON, force_on); + if (REG(DOMAIN10_PG_CONFIG)) + REG_UPDATE(DOMAIN10_PG_CONFIG, DOMAIN8_POWER_FORCEON, force_on); /* DPP0/1/2/3/4/5 */ REG_UPDATE(DOMAIN1_PG_CONFIG, DOMAIN1_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN3_PG_CONFIG, DOMAIN3_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN5_PG_CONFIG, DOMAIN5_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN7_PG_CONFIG, DOMAIN7_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN9_PG_CONFIG, DOMAIN9_POWER_FORCEON, force_on); - /*Do not power gate DPP5, should be left at HW default, power on permanently*/ - /*REG_UPDATE(DOMAIN11_PG_CONFIG, DOMAIN11_POWER_FORCEON, force_on);*/ + if (REG(DOMAIN9_PG_CONFIG)) + REG_UPDATE(DOMAIN9_PG_CONFIG, DOMAIN9_POWER_FORCEON, force_on); + if (REG(DOMAIN11_PG_CONFIG)) + REG_UPDATE(DOMAIN11_PG_CONFIG, DOMAIN9_POWER_FORCEON, force_on); + /* DCS0/1/2/3/4/5 */ REG_UPDATE(DOMAIN16_PG_CONFIG, DOMAIN16_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN17_PG_CONFIG, DOMAIN17_POWER_FORCEON, force_on); REG_UPDATE(DOMAIN18_PG_CONFIG, DOMAIN18_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN19_PG_CONFIG, DOMAIN19_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN20_PG_CONFIG, DOMAIN20_POWER_FORCEON, force_on); - REG_UPDATE(DOMAIN21_PG_CONFIG, DOMAIN21_POWER_FORCEON, force_on); + if (REG(DOMAIN19_PG_CONFIG)) + REG_UPDATE(DOMAIN19_PG_CONFIG, DOMAIN19_POWER_FORCEON, force_on); + if (REG(DOMAIN20_PG_CONFIG)) + REG_UPDATE(DOMAIN20_PG_CONFIG, DOMAIN20_POWER_FORCEON, force_on); + if (REG(DOMAIN21_PG_CONFIG)) + REG_UPDATE(DOMAIN21_PG_CONFIG, DOMAIN21_POWER_FORCEON, force_on); } -static void dcn20_dccg_init(struct dce_hwseq *hws) +void dcn20_dccg_init(struct dce_hwseq *hws) { /* * set MICROSECOND_TIME_BASE_DIV @@ -139,7 +244,7 @@ static void dcn20_dccg_init(struct dce_hwseq *hws) REG_WRITE(DISPCLK_FREQ_CHANGE_CNTL, 0x801003c); } -static void disable_vga( +void dcn20_disable_vga( struct dce_hwseq *hws) { REG_WRITE(D1VGA_CONTROL, 0); @@ -150,23 +255,24 @@ static void disable_vga( REG_WRITE(D6VGA_CONTROL, 0); } -void dcn20_program_tripleBuffer( +void dcn20_program_triple_buffer( const struct dc *dc, struct pipe_ctx *pipe_ctx, - bool enableTripleBuffer) + bool enable_triple_buffer) { if (pipe_ctx->plane_res.hubp && pipe_ctx->plane_res.hubp->funcs) { pipe_ctx->plane_res.hubp->funcs->hubp_enable_tripleBuffer( pipe_ctx->plane_res.hubp, - enableTripleBuffer); + enable_triple_buffer); } } /* Blank pixel data during initialization */ -static void dcn20_init_blank( +void dcn20_init_blank( struct dc *dc, struct timing_generator *tg) { + struct dce_hwseq *hws = dc->hwseq; enum dc_color_space color_space; struct tg_color black_color = {0}; struct output_pixel_processor *opp = NULL; @@ -197,6 +303,7 @@ static void dcn20_init_blank( opp->funcs->opp_set_disp_pattern_generator( opp, CONTROLLER_DP_TEST_PATTERN_SOLID_COLOR, + CONTROLLER_DP_COLOR_SPACE_UDEFINED, COLOR_DEPTH_UNDEFINED, &black_color, otg_active_width, @@ -206,17 +313,17 @@ static void dcn20_init_blank( bottom_opp->funcs->opp_set_disp_pattern_generator( bottom_opp, CONTROLLER_DP_TEST_PATTERN_SOLID_COLOR, + CONTROLLER_DP_COLOR_SPACE_UDEFINED, COLOR_DEPTH_UNDEFINED, &black_color, otg_active_width, otg_active_height); } - dcn20_hwss_wait_for_blank_complete(opp); + hws->funcs.wait_for_blank_complete(opp); } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT -static void dcn20_dsc_pg_control( +void dcn20_dsc_pg_control( struct dce_hwseq *hws, unsigned int dsc_inst, bool power_on) @@ -292,9 +399,8 @@ static void dcn20_dsc_pg_control( if (org_ip_request_cntl == 0) REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); } -#endif -static void dcn20_dpp_pg_control( +void dcn20_dpp_pg_control( struct dce_hwseq *hws, unsigned int dpp_inst, bool power_on) @@ -368,7 +474,7 @@ static void dcn20_dpp_pg_control( } -static void dcn20_hubp_pg_control( +void dcn20_hubp_pg_control( struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on) @@ -442,34 +548,12 @@ static void dcn20_hubp_pg_control( } - -static void dcn20_plane_atomic_power_down(struct dc *dc, struct pipe_ctx *pipe_ctx) -{ - struct dce_hwseq *hws = dc->hwseq; - struct dpp *dpp = pipe_ctx->plane_res.dpp; - - DC_LOGGER_INIT(dc->ctx->logger); - - if (REG(DC_IP_REQUEST_CNTL)) { - REG_SET(DC_IP_REQUEST_CNTL, 0, - IP_REQUEST_EN, 1); - dcn20_dpp_pg_control(hws, dpp->inst, false); - dcn20_hubp_pg_control(hws, pipe_ctx->plane_res.hubp->inst, false); - dpp->funcs->dpp_reset(dpp); - REG_SET(DC_IP_REQUEST_CNTL, 0, - IP_REQUEST_EN, 0); - DC_LOG_DEBUG( - "Power gated front end %d\n", pipe_ctx->pipe_idx); - } -} - - - /* disable HW used by plane. * note: cannot disable until disconnect is complete */ -static void dcn20_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn20_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) { + struct dce_hwseq *hws = dc->hwseq; struct hubp *hubp = pipe_ctx->plane_res.hubp; struct dpp *dpp = pipe_ctx->plane_res.dpp; @@ -488,9 +572,10 @@ static void dcn20_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) dpp->funcs->dpp_dppclk_control(dpp, false, false); hubp->power_gated = true; - dc->optimized_required = false; /* We're powering off, no need to optimize */ - dcn20_plane_atomic_power_down(dc, pipe_ctx); + hws->funcs.plane_atomic_power_down(dc, + pipe_ctx->plane_res.dpp, + pipe_ctx->plane_res.hubp); pipe_ctx->stream = NULL; memset(&pipe_ctx->stream_res, 0, sizeof(pipe_ctx->stream_res)); @@ -514,212 +599,18 @@ void dcn20_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) pipe_ctx->pipe_idx); } -static void dcn20_init_hw(struct dc *dc) -{ - int i, j; - struct abm *abm = dc->res_pool->abm; - struct dmcu *dmcu = dc->res_pool->dmcu; - struct dce_hwseq *hws = dc->hwseq; - struct dc_bios *dcb = dc->ctx->dc_bios; - struct resource_pool *res_pool = dc->res_pool; - struct dc_state *context = dc->current_state; - struct dc_firmware_info fw_info = { { 0 } }; - - if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) - dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); - - // Initialize the dccg - if (res_pool->dccg->funcs->dccg_init) - res_pool->dccg->funcs->dccg_init(res_pool->dccg); - - //Enable ability to power gate / don't force power on permanently - enable_power_gating_plane(dc->hwseq, true); - - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF); - REG_WRITE(RBBMIF_TIMEOUT_DIS_2, 0xFFFFFFFF); - - dcn20_dccg_init(hws); - - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2); - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); - REG_WRITE(REFCLK_CNTL, 0); - } else { - if (!dcb->funcs->is_accelerated_mode(dcb)) { - bios_golden_init(dc); - if (dc->ctx->dc_bios->funcs->get_firmware_info( - dc->ctx->dc_bios, &fw_info) == BP_RESULT_OK) { - res_pool->ref_clocks.xtalin_clock_inKhz = fw_info.pll_info.crystal_frequency; - - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (res_pool->dccg && res_pool->hubbub) { - - (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, - fw_info.pll_info.crystal_frequency, - &res_pool->ref_clocks.dccg_ref_clock_inKhz); - - (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, - res_pool->ref_clocks.dccg_ref_clock_inKhz, - &res_pool->ref_clocks.dchub_ref_clock_inKhz); - } else { - // Not all ASICs have DCCG sw component - res_pool->ref_clocks.dccg_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - res_pool->ref_clocks.dchub_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - } - } - } else - ASSERT_CRITICAL(false); - disable_vga(dc->hwseq); - } - - for (i = 0; i < dc->link_count; i++) { - /* Power up AND update implementation according to the - * required signal (which may be different from the - * default signal on connector). - */ - struct dc_link *link = dc->links[i]; - - link->link_enc->funcs->hw_init(link->link_enc); - } - } - - /* Blank pixel data with OPP DPG */ - for (i = 0; i < dc->res_pool->timing_generator_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - - if (tg->funcs->is_tg_enabled(tg)) { - dcn20_init_blank(dc, tg); - } - } - - for (i = 0; i < res_pool->timing_generator_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - - if (tg->funcs->is_tg_enabled(tg)) - tg->funcs->lock(tg); - } - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct dpp *dpp = res_pool->dpps[i]; - - dpp->funcs->dpp_reset(dpp); - } - - /* Reset all MPCC muxes */ - res_pool->mpc->funcs->mpc_init(res_pool->mpc); - - /* initialize OPP mpc_tree parameter */ - for (i = 0; i < dc->res_pool->res_cap->num_opp; i++) { - res_pool->opps[i]->mpc_tree_params.opp_id = res_pool->opps[i]->inst; - res_pool->opps[i]->mpc_tree_params.opp_list = NULL; - for (j = 0; j < MAX_PIPES; j++) - res_pool->opps[i]->mpcc_disconnect_pending[j] = false; - } - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - struct hubp *hubp = dc->res_pool->hubps[i]; - struct dpp *dpp = dc->res_pool->dpps[i]; - - pipe_ctx->stream_res.tg = tg; - pipe_ctx->pipe_idx = i; - - pipe_ctx->plane_res.hubp = hubp; - pipe_ctx->plane_res.dpp = dpp; - pipe_ctx->plane_res.mpcc_inst = dpp->inst; - hubp->mpcc_id = dpp->inst; - hubp->opp_id = OPP_ID_INVALID; - hubp->power_gated = false; - pipe_ctx->stream_res.opp = NULL; - - hubp->funcs->hubp_init(hubp); - - //dc->res_pool->opps[i]->mpc_tree_params.opp_id = dc->res_pool->opps[i]->inst; - //dc->res_pool->opps[i]->mpc_tree_params.opp_list = NULL; - dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; - pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; - /*to do*/ - hwss1_plane_atomic_disconnect(dc, pipe_ctx); - } - - /* initialize DWB pointer to MCIF_WB */ - for (i = 0; i < res_pool->res_cap->num_dwb; i++) - res_pool->dwbc[i]->mcif = res_pool->mcif_wb[i]; - - for (i = 0; i < dc->res_pool->timing_generator_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - - if (tg->funcs->is_tg_enabled(tg)) - tg->funcs->unlock(tg); - } - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - - dc->hwss.disable_plane(dc, pipe_ctx); - - pipe_ctx->stream_res.tg = NULL; - pipe_ctx->plane_res.hubp = NULL; - } - - for (i = 0; i < dc->res_pool->timing_generator_count; i++) { - struct timing_generator *tg = dc->res_pool->timing_generators[i]; - - tg->funcs->tg_init(tg); - } - - /* end of FPGA. Below if real ASIC */ - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - return; - - - for (i = 0; i < res_pool->audio_count; i++) { - struct audio *audio = res_pool->audios[i]; - - audio->funcs->hw_init(audio); - } - - if (abm != NULL) { - abm->funcs->init_backlight(abm); - abm->funcs->abm_init(abm); - } - - if (dmcu != NULL) - dmcu->funcs->dmcu_init(dmcu); - - if (abm != NULL && dmcu != NULL) - abm->dmcu_is_running = dmcu->funcs->is_dmcu_initialized(dmcu); - - /* power AFMT HDMI memory TODO: may move to dis/en output save power*/ - REG_WRITE(DIO_MEM_PWR_CTRL, 0); - - if (!dc->debug.disable_clock_gate) { - /* enable all DCN clock gating */ - REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0); - - REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0); - - REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); - } - -} - enum dc_status dcn20_enable_stream_timing( struct pipe_ctx *pipe_ctx, struct dc_state *context, struct dc *dc) { + struct dce_hwseq *hws = dc->hwseq; struct dc_stream_state *stream = pipe_ctx->stream; struct drr_params params = {0}; unsigned int event_triggers = 0; - - -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); -#endif + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + int opp_inst[MAX_PIPES] = { pipe_ctx->stream_res.opp->inst }; /* by upper caller loop, pipe0 is parent pipe and be called first. * back end is set up by for pipe0. Other children pipe share back end @@ -730,12 +621,17 @@ enum dc_status dcn20_enable_stream_timing( /* TODO check if timing_changed, disable stream if timing changed */ - if (odm_pipe) + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + opp_inst[opp_cnt] = odm_pipe->stream_res.opp->inst; + opp_cnt++; + } + + if (opp_cnt > 1) pipe_ctx->stream_res.tg->funcs->set_odm_combine( pipe_ctx->stream_res.tg, - odm_pipe->stream_res.opp->inst, - pipe_ctx->stream->timing.h_addressable/2, - pipe_ctx->stream->timing.pixel_encoding); + opp_inst, opp_cnt, + &pipe_ctx->stream->timing); + /* HW program guide assume display already disable * by unplug sequence. OTG assume stop. */ @@ -759,11 +655,7 @@ enum dc_status dcn20_enable_stream_timing( pipe_ctx->stream->signal, true); - if (pipe_ctx->stream_res.tg->funcs->setup_global_lock) - pipe_ctx->stream_res.tg->funcs->setup_global_lock( - pipe_ctx->stream_res.tg); - - if (odm_pipe) + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) odm_pipe->stream_res.opp->funcs->opp_pipe_clock_control( odm_pipe->stream_res.opp, true); @@ -772,7 +664,7 @@ enum dc_status dcn20_enable_stream_timing( pipe_ctx->stream_res.opp, true); - dc->hwss.blank_pixel_data(dc, pipe_ctx, true); + hws->funcs.blank_pixel_data(dc, pipe_ctx, true); /* VTG is within DCHUB command block. DCFCLK is always on */ if (false == pipe_ctx->stream_res.tg->funcs->enable_crtc(pipe_ctx->stream_res.tg)) { @@ -780,10 +672,12 @@ enum dc_status dcn20_enable_stream_timing( return DC_ERROR_UNEXPECTED; } - dcn20_hwss_wait_for_blank_complete(pipe_ctx->stream_res.opp); + hws->funcs.wait_for_blank_complete(pipe_ctx->stream_res.opp); params.vertical_total_min = stream->adjust.v_total_min; params.vertical_total_max = stream->adjust.v_total_max; + params.vertical_total_mid = stream->adjust.v_total_mid; + params.vertical_total_mid_frame_num = stream->adjust.v_total_mid_frame_num; if (pipe_ctx->stream_res.tg->funcs->set_drr) pipe_ctx->stream_res.tg->funcs->set_drr( pipe_ctx->stream_res.tg, ¶ms); @@ -791,9 +685,13 @@ enum dc_status dcn20_enable_stream_timing( // DRR should set trigger event to monitor surface update event if (stream->adjust.v_total_min != 0 && stream->adjust.v_total_max != 0) event_triggers = 0x80; + /* Event triggers and num frames initialized for DRR, but can be + * later updated for PSR use. Note DRR trigger events are generated + * regardless of whether num frames met. + */ if (pipe_ctx->stream_res.tg->funcs->set_static_screen_control) pipe_ctx->stream_res.tg->funcs->set_static_screen_control( - pipe_ctx->stream_res.tg, event_triggers); + pipe_ctx->stream_res.tg, event_triggers, 2); /* TODO program crtc source select for non-virtual signal*/ /* TODO program FMT */ @@ -814,6 +712,10 @@ void dcn20_program_output_csc(struct dc *dc, { struct mpc *mpc = dc->res_pool->mpc; enum mpc_output_csc_mode ocsc_mode = MPC_OUTPUT_CSC_COEF_A; + int mpcc_id = pipe_ctx->plane_res.hubp->inst; + + if (mpc->funcs->power_on_mpc_mem_pwr) + mpc->funcs->power_on_mpc_mem_pwr(mpc, mpcc_id, true); if (pipe_ctx->stream->csc_color_matrix.enable_adjustment == true) { if (mpc->funcs->set_output_csc != NULL) @@ -830,7 +732,7 @@ void dcn20_program_output_csc(struct dc *dc, } } -bool dcn20_set_output_transfer_func(struct pipe_ctx *pipe_ctx, +bool dcn20_set_output_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, const struct dc_stream_state *stream) { int mpcc_id = pipe_ctx->plane_res.hubp->inst; @@ -843,7 +745,9 @@ bool dcn20_set_output_transfer_func(struct pipe_ctx *pipe_ctx, * if programming for all pipes is required then remove condition * pipe_ctx->top_pipe == NULL ,but then fix the diagnostic. */ - if ((pipe_ctx->top_pipe == NULL || dc_res_is_odm_head_pipe(pipe_ctx)) + if (mpc->funcs->power_on_mpc_mem_pwr) + mpc->funcs->power_on_mpc_mem_pwr(mpc, mpcc_id, true); + if (pipe_ctx->top_pipe == NULL && mpc->funcs->set_output_gamma && stream->out_transfer_func) { if (stream->out_transfer_func->type == TF_TYPE_HWPWL) params = &stream->out_transfer_func->pwl; @@ -867,7 +771,7 @@ bool dcn20_set_output_transfer_func(struct pipe_ctx *pipe_ctx, return true; } -static bool dcn20_set_blend_lut( +bool dcn20_set_blend_lut( struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state) { struct dpp *dpp_base = pipe_ctx->plane_res.dpp; @@ -889,7 +793,7 @@ static bool dcn20_set_blend_lut( return result; } -static bool dcn20_set_shaper_3dlut( +bool dcn20_set_shaper_3dlut( struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state) { struct dpp *dpp_base = pipe_ctx->plane_res.dpp; @@ -909,26 +813,20 @@ static bool dcn20_set_shaper_3dlut( result = dpp_base->funcs->dpp_program_shaper_lut(dpp_base, shaper_lut); if (plane_state->lut3d_func && - plane_state->lut3d_func->initialized == true) + plane_state->lut3d_func->state.bits.initialized == 1) result = dpp_base->funcs->dpp_program_3dlut(dpp_base, &plane_state->lut3d_func->lut_3d); else result = dpp_base->funcs->dpp_program_3dlut(dpp_base, NULL); - if (plane_state->lut3d_func && - plane_state->lut3d_func->initialized == true && - plane_state->lut3d_func->hdr_multiplier != 0) - dpp_base->funcs->dpp_set_hdr_multiplier(dpp_base, - plane_state->lut3d_func->hdr_multiplier); - else - dpp_base->funcs->dpp_set_hdr_multiplier(dpp_base, 0x1f000); - return result; } -bool dcn20_set_input_transfer_func(struct pipe_ctx *pipe_ctx, - const struct dc_plane_state *plane_state) +bool dcn20_set_input_transfer_func(struct dc *dc, + struct pipe_ctx *pipe_ctx, + const struct dc_plane_state *plane_state) { + struct dce_hwseq *hws = dc->hwseq; struct dpp *dpp_base = pipe_ctx->plane_res.dpp; const struct dc_transfer_func *tf = NULL; bool result = true; @@ -937,8 +835,8 @@ bool dcn20_set_input_transfer_func(struct pipe_ctx *pipe_ctx, if (dpp_base == NULL || plane_state == NULL) return false; - dcn20_set_shaper_3dlut(pipe_ctx, plane_state); - dcn20_set_blend_lut(pipe_ctx, plane_state); + hws->funcs.set_shaper_3dlut(pipe_ctx, plane_state); + hws->funcs.set_blend_lut(pipe_ctx, plane_state); if (plane_state->in_transfer_func) tf = plane_state->in_transfer_func; @@ -983,6 +881,11 @@ bool dcn20_set_input_transfer_func(struct pipe_ctx *pipe_ctx, IPP_DEGAMMA_MODE_BYPASS); break; case TRANSFER_FUNCTION_PQ: + dpp_base->funcs->dpp_set_degamma(dpp_base, IPP_DEGAMMA_MODE_USER_PWL); + cm_helper_translate_curve_to_degamma_hw_format(tf, &dpp_base->degamma_params); + dpp_base->funcs->dpp_program_degamma_pwl(dpp_base, &dpp_base->degamma_params); + result = true; + break; default: result = false; break; @@ -1003,16 +906,22 @@ bool dcn20_set_input_transfer_func(struct pipe_ctx *pipe_ctx, return result; } -static void dcn20_update_odm(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx) +void dcn20_update_odm(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx) { - struct pipe_ctx *combine_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + int opp_inst[MAX_PIPES] = { pipe_ctx->stream_res.opp->inst }; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + opp_inst[opp_cnt] = odm_pipe->stream_res.opp->inst; + opp_cnt++; + } - if (combine_pipe) + if (opp_cnt > 1) pipe_ctx->stream_res.tg->funcs->set_odm_combine( pipe_ctx->stream_res.tg, - combine_pipe->stream_res.opp->inst, - pipe_ctx->plane_res.scl_data.h_active, - pipe_ctx->stream->timing.pixel_encoding); + opp_inst, opp_cnt, + &pipe_ctx->stream->timing); else pipe_ctx->stream_res.tg->funcs->set_odm_bypass( pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); @@ -1028,23 +937,32 @@ void dcn20_blank_pixel_data( struct dc_stream_state *stream = pipe_ctx->stream; enum dc_color_space color_space = stream->output_color_space; enum controller_dp_test_pattern test_pattern = CONTROLLER_DP_TEST_PATTERN_SOLID_COLOR; - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); + enum controller_dp_color_space test_pattern_color_space = CONTROLLER_DP_COLOR_SPACE_UDEFINED; + struct pipe_ctx *odm_pipe; + int odm_cnt = 1; int width = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; int height = stream->timing.v_addressable + stream->timing.v_border_bottom + stream->timing.v_border_top; + if (stream->link->test_pattern_enabled) + return; + /* get opp dpg blank color */ color_space_to_black_color(dc, color_space, &black_color); - if (bot_odm_pipe) - width = width / 2; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + odm_cnt++; + + width = width / odm_cnt; if (blank) { if (stream_res->abm) stream_res->abm->funcs->set_abm_immediate_disable(stream_res->abm); - if (dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE) + if (dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE) { test_pattern = CONTROLLER_DP_TEST_PATTERN_COLORSQUARES; + test_pattern_color_space = CONTROLLER_DP_COLOR_SPACE_RGB; + } } else { test_pattern = CONTROLLER_DP_TEST_PATTERN_VIDEOMODE; } @@ -1052,16 +970,18 @@ void dcn20_blank_pixel_data( stream_res->opp->funcs->opp_set_disp_pattern_generator( stream_res->opp, test_pattern, + test_pattern_color_space, stream->timing.display_color_depth, &black_color, width, height); - if (bot_odm_pipe) { - bot_odm_pipe->stream_res.opp->funcs->opp_set_disp_pattern_generator( - bot_odm_pipe->stream_res.opp, - dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE ? + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + odm_pipe->stream_res.opp->funcs->opp_set_disp_pattern_generator( + odm_pipe->stream_res.opp, + dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE && blank ? CONTROLLER_DP_TEST_PATTERN_COLORRAMP : test_pattern, + test_pattern_color_space, stream->timing.display_color_depth, &black_color, width, @@ -1106,6 +1026,9 @@ void dcn20_enable_plane( /* enable DCFCLK current DCHUB */ pipe_ctx->plane_res.hubp->funcs->hubp_clk_cntl(pipe_ctx->plane_res.hubp, true); + /* initialize HUBP on power up */ + pipe_ctx->plane_res.hubp->funcs->hubp_init(pipe_ctx->plane_res.hubp); + /* make sure OPP_PIPE_CLOCK_EN = 1 */ pipe_ctx->stream_res.opp->funcs->opp_pipe_clock_control( pipe_ctx->stream_res.opp, @@ -1166,67 +1089,6 @@ void dcn20_enable_plane( } -static void dcn20_program_pipe( - struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct dc_state *context) -{ - pipe_ctx->plane_state->update_flags.bits.full_update = - context->commit_hints.full_update_needed ? 1 : pipe_ctx->plane_state->update_flags.bits.full_update; - - if (pipe_ctx->plane_state->update_flags.bits.full_update) - dcn20_enable_plane(dc, pipe_ctx, context); - - update_dchubp_dpp(dc, pipe_ctx, context); - - set_hdr_multiplier(pipe_ctx); - - if (pipe_ctx->plane_state->update_flags.bits.full_update || - pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change || - pipe_ctx->plane_state->update_flags.bits.gamma_change) - dc->hwss.set_input_transfer_func(pipe_ctx, pipe_ctx->plane_state); - - /* dcn10_translate_regamma_to_hw_format takes 750us to finish - * only do gamma programming for full update. - * TODO: This can be further optimized/cleaned up - * Always call this for now since it does memcmp inside before - * doing heavy calculation and programming - */ - if (pipe_ctx->plane_state->update_flags.bits.full_update) - dc->hwss.set_output_transfer_func(pipe_ctx, pipe_ctx->stream); -} - -static void dcn20_program_all_pipe_in_tree( - struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct dc_state *context) -{ - if (pipe_ctx->top_pipe == NULL) { - bool blank = !is_pipe_tree_visible(pipe_ctx); - - pipe_ctx->stream_res.tg->funcs->program_global_sync( - pipe_ctx->stream_res.tg, - pipe_ctx->pipe_dlg_param.vready_offset, - pipe_ctx->pipe_dlg_param.vstartup_start, - pipe_ctx->pipe_dlg_param.vupdate_offset, - pipe_ctx->pipe_dlg_param.vupdate_width); - - pipe_ctx->stream_res.tg->funcs->set_vtg_params( - pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); - - dc->hwss.blank_pixel_data(dc, pipe_ctx, blank); - - if (dc->hwss.update_odm) - dc->hwss.update_odm(dc, context, pipe_ctx); - } - - if (pipe_ctx->plane_state != NULL) - dcn20_program_pipe(dc, pipe_ctx, context); - - if (pipe_ctx->bottom_pipe != NULL && pipe_ctx->bottom_pipe != pipe_ctx) - dcn20_program_all_pipe_in_tree(dc, pipe_ctx->bottom_pipe, context); -} - void dcn20_pipe_control_lock_global( struct dc *dc, struct pipe_ctx *pipe, @@ -1266,14 +1128,22 @@ void dcn20_pipe_control_lock( flip_immediate = pipe->plane_state->flip_immediate; if (flip_immediate && lock) { - while (pipe->plane_res.hubp->funcs->hubp_is_flip_pending(pipe->plane_res.hubp)) { + const int TIMEOUT_FOR_FLIP_PENDING = 100000; + int i; + + for (i = 0; i < TIMEOUT_FOR_FLIP_PENDING; ++i) { + if (!pipe->plane_res.hubp->funcs->hubp_is_flip_pending(pipe->plane_res.hubp)) + break; udelay(1); } - if (pipe->bottom_pipe != NULL) - while (pipe->bottom_pipe->plane_res.hubp->funcs->hubp_is_flip_pending(pipe->bottom_pipe->plane_res.hubp)) { + if (pipe->bottom_pipe != NULL) { + for (i = 0; i < TIMEOUT_FOR_FLIP_PENDING; ++i) { + if (!pipe->bottom_pipe->plane_res.hubp->funcs->hubp_is_flip_pending(pipe->bottom_pipe->plane_res.hubp)) + break; udelay(1); } + } } /* In flip immediate and pipe splitting case, we need to use GSL @@ -1297,100 +1167,492 @@ void dcn20_pipe_control_lock( } } -static void dcn20_apply_ctx_for_surface( - struct dc *dc, - const struct dc_stream_state *stream, - int num_planes, - struct dc_state *context) +static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx *new_pipe) { + new_pipe->update_flags.raw = 0; - int i; - struct timing_generator *tg; - bool removed_pipe[6] = { false }; - bool interdependent_update = false; - struct pipe_ctx *top_pipe_to_program = - find_top_pipe_for_stream(dc, context, stream); - DC_LOGGER_INIT(dc->ctx->logger); - - if (!top_pipe_to_program) + /* Exit on unchanged, unused pipe */ + if (!old_pipe->plane_state && !new_pipe->plane_state) + return; + /* Detect pipe enable/disable */ + if (!old_pipe->plane_state && new_pipe->plane_state) { + new_pipe->update_flags.bits.enable = 1; + new_pipe->update_flags.bits.mpcc = 1; + new_pipe->update_flags.bits.dppclk = 1; + new_pipe->update_flags.bits.hubp_interdependent = 1; + new_pipe->update_flags.bits.hubp_rq_dlg_ttu = 1; + new_pipe->update_flags.bits.gamut_remap = 1; + new_pipe->update_flags.bits.scaler = 1; + new_pipe->update_flags.bits.viewport = 1; + if (!new_pipe->top_pipe && !new_pipe->prev_odm_pipe) { + new_pipe->update_flags.bits.odm = 1; + new_pipe->update_flags.bits.global_sync = 1; + } + return; + } + if (old_pipe->plane_state && !new_pipe->plane_state) { + new_pipe->update_flags.bits.disable = 1; return; + } - tg = top_pipe_to_program->stream_res.tg; + /* Detect top pipe only changes */ + if (!new_pipe->top_pipe && !new_pipe->prev_odm_pipe) { + /* Detect odm changes */ + if ((old_pipe->next_odm_pipe && new_pipe->next_odm_pipe + && old_pipe->next_odm_pipe->pipe_idx != new_pipe->next_odm_pipe->pipe_idx) + || (!old_pipe->next_odm_pipe && new_pipe->next_odm_pipe) + || (old_pipe->next_odm_pipe && !new_pipe->next_odm_pipe) + || old_pipe->stream_res.opp != new_pipe->stream_res.opp) + new_pipe->update_flags.bits.odm = 1; + + /* Detect global sync changes */ + if (old_pipe->pipe_dlg_param.vready_offset != new_pipe->pipe_dlg_param.vready_offset + || old_pipe->pipe_dlg_param.vstartup_start != new_pipe->pipe_dlg_param.vstartup_start + || old_pipe->pipe_dlg_param.vupdate_offset != new_pipe->pipe_dlg_param.vupdate_offset + || old_pipe->pipe_dlg_param.vupdate_width != new_pipe->pipe_dlg_param.vupdate_width) + new_pipe->update_flags.bits.global_sync = 1; + } - interdependent_update = top_pipe_to_program->plane_state && - top_pipe_to_program->plane_state->update_flags.bits.full_update; + /* + * Detect opp / tg change, only set on change, not on enable + * Assume mpcc inst = pipe index, if not this code needs to be updated + * since mpcc is what is affected by these. In fact all of our sequence + * makes this assumption at the moment with how hubp reset is matched to + * same index mpcc reset. + */ + if (old_pipe->stream_res.opp != new_pipe->stream_res.opp) + new_pipe->update_flags.bits.opp_changed = 1; + if (old_pipe->stream_res.tg != new_pipe->stream_res.tg) + new_pipe->update_flags.bits.tg_changed = 1; + + /* Detect mpcc blending changes, only dpp inst and bot matter here */ + if (old_pipe->plane_res.dpp != new_pipe->plane_res.dpp + || old_pipe->stream_res.opp != new_pipe->stream_res.opp + || (!old_pipe->bottom_pipe && new_pipe->bottom_pipe) + || (old_pipe->bottom_pipe && !new_pipe->bottom_pipe) + || (old_pipe->bottom_pipe && new_pipe->bottom_pipe + && old_pipe->bottom_pipe->plane_res.mpcc_inst + != new_pipe->bottom_pipe->plane_res.mpcc_inst)) + new_pipe->update_flags.bits.mpcc = 1; + + /* Detect dppclk change */ + if (old_pipe->plane_res.bw.dppclk_khz != new_pipe->plane_res.bw.dppclk_khz) + new_pipe->update_flags.bits.dppclk = 1; + + /* Check for scl update */ + if (memcmp(&old_pipe->plane_res.scl_data, &new_pipe->plane_res.scl_data, sizeof(struct scaler_data))) + new_pipe->update_flags.bits.scaler = 1; + /* Check for vp update */ + if (memcmp(&old_pipe->plane_res.scl_data.viewport, &new_pipe->plane_res.scl_data.viewport, sizeof(struct rect)) + || memcmp(&old_pipe->plane_res.scl_data.viewport_c, + &new_pipe->plane_res.scl_data.viewport_c, sizeof(struct rect))) + new_pipe->update_flags.bits.viewport = 1; + + /* Detect dlg/ttu/rq updates */ + { + struct _vcs_dpi_display_dlg_regs_st old_dlg_attr = old_pipe->dlg_regs; + struct _vcs_dpi_display_ttu_regs_st old_ttu_attr = old_pipe->ttu_regs; + struct _vcs_dpi_display_dlg_regs_st *new_dlg_attr = &new_pipe->dlg_regs; + struct _vcs_dpi_display_ttu_regs_st *new_ttu_attr = &new_pipe->ttu_regs; + + /* Detect pipe interdependent updates */ + if (old_dlg_attr.dst_y_prefetch != new_dlg_attr->dst_y_prefetch || + old_dlg_attr.vratio_prefetch != new_dlg_attr->vratio_prefetch || + old_dlg_attr.vratio_prefetch_c != new_dlg_attr->vratio_prefetch_c || + old_dlg_attr.dst_y_per_vm_vblank != new_dlg_attr->dst_y_per_vm_vblank || + old_dlg_attr.dst_y_per_row_vblank != new_dlg_attr->dst_y_per_row_vblank || + old_dlg_attr.dst_y_per_vm_flip != new_dlg_attr->dst_y_per_vm_flip || + old_dlg_attr.dst_y_per_row_flip != new_dlg_attr->dst_y_per_row_flip || + old_dlg_attr.refcyc_per_meta_chunk_vblank_l != new_dlg_attr->refcyc_per_meta_chunk_vblank_l || + old_dlg_attr.refcyc_per_meta_chunk_vblank_c != new_dlg_attr->refcyc_per_meta_chunk_vblank_c || + old_dlg_attr.refcyc_per_meta_chunk_flip_l != new_dlg_attr->refcyc_per_meta_chunk_flip_l || + old_dlg_attr.refcyc_per_line_delivery_pre_l != new_dlg_attr->refcyc_per_line_delivery_pre_l || + old_dlg_attr.refcyc_per_line_delivery_pre_c != new_dlg_attr->refcyc_per_line_delivery_pre_c || + old_ttu_attr.refcyc_per_req_delivery_pre_l != new_ttu_attr->refcyc_per_req_delivery_pre_l || + old_ttu_attr.refcyc_per_req_delivery_pre_c != new_ttu_attr->refcyc_per_req_delivery_pre_c || + old_ttu_attr.refcyc_per_req_delivery_pre_cur0 != new_ttu_attr->refcyc_per_req_delivery_pre_cur0 || + old_ttu_attr.refcyc_per_req_delivery_pre_cur1 != new_ttu_attr->refcyc_per_req_delivery_pre_cur1 || + old_ttu_attr.min_ttu_vblank != new_ttu_attr->min_ttu_vblank || + old_ttu_attr.qos_level_flip != new_ttu_attr->qos_level_flip) { + old_dlg_attr.dst_y_prefetch = new_dlg_attr->dst_y_prefetch; + old_dlg_attr.vratio_prefetch = new_dlg_attr->vratio_prefetch; + old_dlg_attr.vratio_prefetch_c = new_dlg_attr->vratio_prefetch_c; + old_dlg_attr.dst_y_per_vm_vblank = new_dlg_attr->dst_y_per_vm_vblank; + old_dlg_attr.dst_y_per_row_vblank = new_dlg_attr->dst_y_per_row_vblank; + old_dlg_attr.dst_y_per_vm_flip = new_dlg_attr->dst_y_per_vm_flip; + old_dlg_attr.dst_y_per_row_flip = new_dlg_attr->dst_y_per_row_flip; + old_dlg_attr.refcyc_per_meta_chunk_vblank_l = new_dlg_attr->refcyc_per_meta_chunk_vblank_l; + old_dlg_attr.refcyc_per_meta_chunk_vblank_c = new_dlg_attr->refcyc_per_meta_chunk_vblank_c; + old_dlg_attr.refcyc_per_meta_chunk_flip_l = new_dlg_attr->refcyc_per_meta_chunk_flip_l; + old_dlg_attr.refcyc_per_line_delivery_pre_l = new_dlg_attr->refcyc_per_line_delivery_pre_l; + old_dlg_attr.refcyc_per_line_delivery_pre_c = new_dlg_attr->refcyc_per_line_delivery_pre_c; + old_ttu_attr.refcyc_per_req_delivery_pre_l = new_ttu_attr->refcyc_per_req_delivery_pre_l; + old_ttu_attr.refcyc_per_req_delivery_pre_c = new_ttu_attr->refcyc_per_req_delivery_pre_c; + old_ttu_attr.refcyc_per_req_delivery_pre_cur0 = new_ttu_attr->refcyc_per_req_delivery_pre_cur0; + old_ttu_attr.refcyc_per_req_delivery_pre_cur1 = new_ttu_attr->refcyc_per_req_delivery_pre_cur1; + old_ttu_attr.min_ttu_vblank = new_ttu_attr->min_ttu_vblank; + old_ttu_attr.qos_level_flip = new_ttu_attr->qos_level_flip; + new_pipe->update_flags.bits.hubp_interdependent = 1; + } + /* Detect any other updates to ttu/rq/dlg */ + if (memcmp(&old_dlg_attr, &new_pipe->dlg_regs, sizeof(old_dlg_attr)) || + memcmp(&old_ttu_attr, &new_pipe->ttu_regs, sizeof(old_ttu_attr)) || + memcmp(&old_pipe->rq_regs, &new_pipe->rq_regs, sizeof(old_pipe->rq_regs))) + new_pipe->update_flags.bits.hubp_rq_dlg_ttu = 1; + } +} - if (interdependent_update) - lock_all_pipes(dc, context, true); - else - dcn20_pipe_control_lock(dc, top_pipe_to_program, true); +static void dcn20_update_dchubp_dpp( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + struct dc_state *context) +{ + struct dce_hwseq *hws = dc->hwseq; + struct hubp *hubp = pipe_ctx->plane_res.hubp; + struct dpp *dpp = pipe_ctx->plane_res.dpp; + struct dc_plane_state *plane_state = pipe_ctx->plane_state; + bool viewport_changed = false; - if (num_planes == 0) { - /* OTG blank before remove all front end */ - dc->hwss.blank_pixel_data(dc, top_pipe_to_program, true); + if (pipe_ctx->update_flags.bits.dppclk) + dpp->funcs->dpp_dppclk_control(dpp, false, true); + + /* TODO: Need input parameter to tell current DCHUB pipe tie to which OTG + * VTG is within DCHUBBUB which is commond block share by each pipe HUBP. + * VTG is 1:1 mapping with OTG. Each pipe HUBP will select which VTG + */ + if (pipe_ctx->update_flags.bits.hubp_rq_dlg_ttu) { + hubp->funcs->hubp_vtg_sel(hubp, pipe_ctx->stream_res.tg->inst); + + hubp->funcs->hubp_setup( + hubp, + &pipe_ctx->dlg_regs, + &pipe_ctx->ttu_regs, + &pipe_ctx->rq_regs, + &pipe_ctx->pipe_dlg_param); + } + if (pipe_ctx->update_flags.bits.hubp_interdependent) + hubp->funcs->hubp_setup_interdependent( + hubp, + &pipe_ctx->dlg_regs, + &pipe_ctx->ttu_regs); + + if (pipe_ctx->update_flags.bits.enable || + plane_state->update_flags.bits.bpp_change || + plane_state->update_flags.bits.input_csc_change || + plane_state->update_flags.bits.color_space_change || + plane_state->update_flags.bits.coeff_reduction_change) { + struct dc_bias_and_scale bns_params = {0}; + + // program the input csc + dpp->funcs->dpp_setup(dpp, + plane_state->format, + EXPANSION_MODE_ZERO, + plane_state->input_csc_color_matrix, + plane_state->color_space, + NULL); + + if (dpp->funcs->dpp_program_bias_and_scale) { + //TODO :for CNVC set scale and bias registers if necessary + build_prescale_params(&bns_params, plane_state); + dpp->funcs->dpp_program_bias_and_scale(dpp, &bns_params); + } } - /* Disconnect unused mpcc */ - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - struct pipe_ctx *old_pipe_ctx = - &dc->current_state->res_ctx.pipe_ctx[i]; - /* - * Powergate reused pipes that are not powergated - * fairly hacky right now, using opp_id as indicator - * TODO: After move dc_post to dc_update, this will - * be removed. - */ - if (pipe_ctx->plane_state && !old_pipe_ctx->plane_state) { - if (old_pipe_ctx->stream_res.tg == tg && - old_pipe_ctx->plane_res.hubp && - old_pipe_ctx->plane_res.hubp->opp_id != OPP_ID_INVALID) - dcn20_disable_plane(dc, old_pipe_ctx); + if (pipe_ctx->update_flags.bits.mpcc + || plane_state->update_flags.bits.global_alpha_change + || plane_state->update_flags.bits.per_pixel_alpha_change) { + // MPCC inst is equal to pipe index in practice + int mpcc_inst = hubp->inst; + int opp_inst; + int opp_count = dc->res_pool->pipe_count; + + for (opp_inst = 0; opp_inst < opp_count; opp_inst++) { + if (dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst]) { + dc->res_pool->mpc->funcs->wait_for_idle(dc->res_pool->mpc, mpcc_inst); + dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst] = false; + break; + } } + hws->funcs.update_mpcc(dc, pipe_ctx); + } - if ((!pipe_ctx->plane_state || - pipe_ctx->stream_res.tg != old_pipe_ctx->stream_res.tg) && - old_pipe_ctx->plane_state && - old_pipe_ctx->stream_res.tg == tg) { + if (pipe_ctx->update_flags.bits.scaler || + plane_state->update_flags.bits.scaling_change || + plane_state->update_flags.bits.position_change || + plane_state->update_flags.bits.per_pixel_alpha_change || + pipe_ctx->stream->update_flags.bits.scaling) { + pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->plane_state->per_pixel_alpha; + ASSERT(pipe_ctx->plane_res.scl_data.lb_params.depth == LB_PIXEL_DEPTH_30BPP); + /* scaler configuration */ + pipe_ctx->plane_res.dpp->funcs->dpp_set_scaler( + pipe_ctx->plane_res.dpp, &pipe_ctx->plane_res.scl_data); + } - dc->hwss.plane_atomic_disconnect(dc, old_pipe_ctx); - removed_pipe[i] = true; + if (pipe_ctx->update_flags.bits.viewport || + (context == dc->current_state && plane_state->update_flags.bits.scaling_change) || + (context == dc->current_state && pipe_ctx->stream->update_flags.bits.scaling)) { - DC_LOG_DC("Reset mpcc for pipe %d\n", - old_pipe_ctx->pipe_idx); - } + hubp->funcs->mem_program_viewport( + hubp, + &pipe_ctx->plane_res.scl_data.viewport, + &pipe_ctx->plane_res.scl_data.viewport_c); + viewport_changed = true; + } + + /* Any updates are handled in dc interface, just need to apply existing for plane enable */ + if ((pipe_ctx->update_flags.bits.enable || pipe_ctx->update_flags.bits.opp_changed || + pipe_ctx->update_flags.bits.scaler || pipe_ctx->update_flags.bits.viewport) + && pipe_ctx->stream->cursor_attributes.address.quad_part != 0) { + dc->hwss.set_cursor_position(pipe_ctx); + dc->hwss.set_cursor_attribute(pipe_ctx); + + if (dc->hwss.set_cursor_sdr_white_level) + dc->hwss.set_cursor_sdr_white_level(pipe_ctx); + } + + /* Any updates are handled in dc interface, just need + * to apply existing for plane enable / opp change */ + if (pipe_ctx->update_flags.bits.enable || pipe_ctx->update_flags.bits.opp_changed + || pipe_ctx->stream->update_flags.bits.gamut_remap + || pipe_ctx->stream->update_flags.bits.out_csc) { + /* dpp/cm gamut remap*/ + dc->hwss.program_gamut_remap(pipe_ctx); + + /*call the dcn2 method which uses mpc csc*/ + dc->hwss.program_output_csc(dc, + pipe_ctx, + pipe_ctx->stream->output_color_space, + pipe_ctx->stream->csc_color_matrix.matrix, + hubp->opp_id); + } + + if (pipe_ctx->update_flags.bits.enable || + pipe_ctx->update_flags.bits.opp_changed || + plane_state->update_flags.bits.pixel_format_change || + plane_state->update_flags.bits.horizontal_mirror_change || + plane_state->update_flags.bits.rotation_change || + plane_state->update_flags.bits.swizzle_change || + plane_state->update_flags.bits.dcc_change || + plane_state->update_flags.bits.bpp_change || + plane_state->update_flags.bits.scaling_change || + plane_state->update_flags.bits.plane_size_change) { + struct plane_size size = plane_state->plane_size; + + size.surface_size = pipe_ctx->plane_res.scl_data.viewport; + hubp->funcs->hubp_program_surface_config( + hubp, + plane_state->format, + &plane_state->tiling_info, + &size, + plane_state->rotation, + &plane_state->dcc, + plane_state->horizontal_mirror, + 0); + hubp->power_gated = false; + } + + if (hubp->funcs->apply_PLAT_54186_wa && viewport_changed) + hubp->funcs->apply_PLAT_54186_wa(hubp, &plane_state->address); + + if (pipe_ctx->update_flags.bits.enable || plane_state->update_flags.bits.addr_update) + hws->funcs.update_plane_addr(dc, pipe_ctx); + + + + if (pipe_ctx->update_flags.bits.enable) + hubp->funcs->set_blank(hubp, false); +} + + +static void dcn20_program_pipe( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + struct dc_state *context) +{ + struct dce_hwseq *hws = dc->hwseq; + /* Only need to unblank on top pipe */ + if ((pipe_ctx->update_flags.bits.enable || pipe_ctx->stream->update_flags.bits.abm_level) + && !pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe) + hws->funcs.blank_pixel_data(dc, pipe_ctx, !pipe_ctx->plane_state->visible); + + if (pipe_ctx->update_flags.bits.global_sync) { + pipe_ctx->stream_res.tg->funcs->program_global_sync( + pipe_ctx->stream_res.tg, + pipe_ctx->pipe_dlg_param.vready_offset, + pipe_ctx->pipe_dlg_param.vstartup_start, + pipe_ctx->pipe_dlg_param.vupdate_offset, + pipe_ctx->pipe_dlg_param.vupdate_width); + + pipe_ctx->stream_res.tg->funcs->set_vtg_params( + pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); + + if (hws->funcs.setup_vupdate_interrupt) + hws->funcs.setup_vupdate_interrupt(dc, pipe_ctx); + } + + if (pipe_ctx->update_flags.bits.odm) + hws->funcs.update_odm(dc, context, pipe_ctx); + + if (pipe_ctx->update_flags.bits.enable) + dcn20_enable_plane(dc, pipe_ctx, context); + + if (pipe_ctx->update_flags.raw || pipe_ctx->plane_state->update_flags.raw || pipe_ctx->stream->update_flags.raw) + dcn20_update_dchubp_dpp(dc, pipe_ctx, context); + + if (pipe_ctx->update_flags.bits.enable + || pipe_ctx->plane_state->update_flags.bits.hdr_mult) + hws->funcs.set_hdr_multiplier(pipe_ctx); + + if (pipe_ctx->update_flags.bits.enable || + pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change || + pipe_ctx->plane_state->update_flags.bits.gamma_change) + hws->funcs.set_input_transfer_func(dc, pipe_ctx, pipe_ctx->plane_state); + + /* dcn10_translate_regamma_to_hw_format takes 750us to finish + * only do gamma programming for powering on, internal memcmp to avoid + * updating on slave planes + */ + if (pipe_ctx->update_flags.bits.enable || pipe_ctx->stream->update_flags.bits.out_tf) + hws->funcs.set_output_transfer_func(dc, pipe_ctx, pipe_ctx->stream); + + /* If the pipe has been enabled or has a different opp, we + * should reprogram the fmt. This deals with cases where + * interation between mpc and odm combine on different streams + * causes a different pipe to be chosen to odm combine with. + */ + if (pipe_ctx->update_flags.bits.enable + || pipe_ctx->update_flags.bits.opp_changed) { + + pipe_ctx->stream_res.opp->funcs->opp_set_dyn_expansion( + pipe_ctx->stream_res.opp, + COLOR_SPACE_YCBCR601, + pipe_ctx->stream->timing.display_color_depth, + pipe_ctx->stream->signal); + + pipe_ctx->stream_res.opp->funcs->opp_program_fmt( + pipe_ctx->stream_res.opp, + &pipe_ctx->stream->bit_depth_params, + &pipe_ctx->stream->clamping); } +} + +static bool does_pipe_need_lock(struct pipe_ctx *pipe) +{ + if ((pipe->plane_state && pipe->plane_state->update_flags.raw) + || pipe->update_flags.raw) + return true; + if (pipe->bottom_pipe) + return does_pipe_need_lock(pipe->bottom_pipe); + + return false; +} - if (num_planes > 0) - dcn20_program_all_pipe_in_tree(dc, top_pipe_to_program, context); +void dcn20_program_front_end_for_ctx( + struct dc *dc, + struct dc_state *context) +{ + const unsigned int TIMEOUT_FOR_PIPE_ENABLE_MS = 100; + int i; + struct dce_hwseq *hws = dc->hwseq; + bool pipe_locked[MAX_PIPES] = {false}; + DC_LOGGER_INIT(dc->ctx->logger); - /* Program secondary blending tree and writeback pipes */ - if ((stream->num_wb_info > 0) && (dc->hwss.program_all_writeback_pipes_in_tree)) - dc->hwss.program_all_writeback_pipes_in_tree(dc, stream, context); + /* Carry over GSL groups in case the context is changing. */ + for (i = 0; i < dc->res_pool->pipe_count; i++) + if (context->res_ctx.pipe_ctx[i].stream == dc->current_state->res_ctx.pipe_ctx[i].stream) + context->res_ctx.pipe_ctx[i].stream_res.gsl_group = + dc->current_state->res_ctx.pipe_ctx[i].stream_res.gsl_group; - if (interdependent_update) - for (i = 0; i < dc->res_pool->pipe_count; i++) { + /* Set pipe update flags and lock pipes */ + for (i = 0; i < dc->res_pool->pipe_count; i++) + dcn20_detect_pipe_changes(&dc->current_state->res_ctx.pipe_ctx[i], + &context->res_ctx.pipe_ctx[i]); + for (i = 0; i < dc->res_pool->pipe_count; i++) + if (!context->res_ctx.pipe_ctx[i].top_pipe && + does_pipe_need_lock(&context->res_ctx.pipe_ctx[i])) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - /* Skip inactive pipes and ones already updated */ - if (!pipe_ctx->stream || pipe_ctx->stream == stream || - !pipe_ctx->plane_state || !tg->funcs->is_tg_enabled(tg)) - continue; + if (pipe_ctx->update_flags.bits.tg_changed || pipe_ctx->update_flags.bits.enable) + dc->hwss.pipe_control_lock(dc, pipe_ctx, true); + if (!pipe_ctx->update_flags.bits.enable) + dc->hwss.pipe_control_lock(dc, &dc->current_state->res_ctx.pipe_ctx[i], true); + pipe_locked[i] = true; + } + + /* OTG blank before disabling all front ends */ + for (i = 0; i < dc->res_pool->pipe_count; i++) + if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable + && !context->res_ctx.pipe_ctx[i].top_pipe + && !context->res_ctx.pipe_ctx[i].prev_odm_pipe + && context->res_ctx.pipe_ctx[i].stream) + hws->funcs.blank_pixel_data(dc, &context->res_ctx.pipe_ctx[i], true); - pipe_ctx->plane_res.hubp->funcs->hubp_setup_interdependent( - pipe_ctx->plane_res.hubp, - &pipe_ctx->dlg_regs, - &pipe_ctx->ttu_regs); + /* Disconnect mpcc */ + for (i = 0; i < dc->res_pool->pipe_count; i++) + if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable + || context->res_ctx.pipe_ctx[i].update_flags.bits.opp_changed) { + hws->funcs.plane_atomic_disconnect(dc, &dc->current_state->res_ctx.pipe_ctx[i]); + DC_LOG_DC("Reset mpcc for pipe %d\n", dc->current_state->res_ctx.pipe_ctx[i].pipe_idx); } - if (interdependent_update) - lock_all_pipes(dc, context, false); - else - dcn20_pipe_control_lock(dc, top_pipe_to_program, false); + /* + * Program all updated pipes, order matters for mpcc setup. Start with + * top pipe and program all pipes that follow in order + */ + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->plane_state && !pipe->top_pipe) { + while (pipe) { + dcn20_program_pipe(dc, pipe, context); + pipe = pipe->bottom_pipe; + } + /* Program secondary blending tree and writeback pipes */ + pipe = &context->res_ctx.pipe_ctx[i]; + if (!pipe->prev_odm_pipe && pipe->stream->num_wb_info > 0 + && (pipe->update_flags.raw || pipe->plane_state->update_flags.raw || pipe->stream->update_flags.raw) + && hws->funcs.program_all_writeback_pipes_in_tree) + hws->funcs.program_all_writeback_pipes_in_tree(dc, pipe->stream, context); + } + } + + /* Unlock all locked pipes */ + for (i = 0; i < dc->res_pool->pipe_count; i++) + if (pipe_locked[i]) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (pipe_ctx->update_flags.bits.tg_changed || pipe_ctx->update_flags.bits.enable) + dc->hwss.pipe_control_lock(dc, pipe_ctx, false); + if (!pipe_ctx->update_flags.bits.enable) + dc->hwss.pipe_control_lock(dc, &dc->current_state->res_ctx.pipe_ctx[i], false); + } for (i = 0; i < dc->res_pool->pipe_count; i++) - if (removed_pipe[i]) - dcn20_disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); + if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable) + dc->hwss.disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); + + /* + * If we are enabling a pipe, we need to wait for pending clear as this is a critical + * part of the enable operation otherwise, DM may request an immediate flip which + * will cause HW to perform an "immediate enable" (as opposed to "vsync enable") which + * is unsupported on DCN. + */ + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->plane_state && !pipe->top_pipe && pipe->update_flags.bits.enable) { + struct hubp *hubp = pipe->plane_res.hubp; + int j = 0; + + for (j = 0; j < TIMEOUT_FOR_PIPE_ENABLE_MS*1000 + && hubp->funcs->hubp_is_flip_pending(hubp); j++) + mdelay(1); + } + } + + /* WA to apply WM setting*/ + if (dc->hwseq->wa.DEGVIDCN21) + dc->res_pool->hubbub->funcs->apply_DEDCN21_147_wa(dc->res_pool->hubbub); } @@ -1400,16 +1662,16 @@ void dcn20_prepare_bandwidth( { struct hubbub *hubbub = dc->res_pool->hubbub; + dc->clk_mgr->funcs->update_clocks( + dc->clk_mgr, + context, + false); + /* program dchubbub watermarks */ hubbub->funcs->program_watermarks(hubbub, &context->bw_ctx.bw.dcn.watermarks, dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, false); - - dc->clk_mgr->funcs->update_clocks( - dc->clk_mgr, - context, - false); } void dcn20_optimize_bandwidth( @@ -1435,6 +1697,7 @@ bool dcn20_update_bandwidth( struct dc_state *context) { int i; + struct dce_hwseq *hws = dc->hwseq; /* recalculate DML parameters */ if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false)) @@ -1463,7 +1726,11 @@ bool dcn20_update_bandwidth( pipe_ctx->stream_res.tg->funcs->set_vtg_params( pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); - dc->hwss.blank_pixel_data(dc, pipe_ctx, blank); + if (pipe_ctx->prev_odm_pipe == NULL) + hws->funcs.blank_pixel_data(dc, pipe_ctx, blank); + + if (hws->funcs.setup_vupdate_interrupt) + hws->funcs.setup_vupdate_interrupt(dc, pipe_ctx); } pipe_ctx->plane_res.hubp->funcs->hubp_setup( @@ -1477,10 +1744,10 @@ bool dcn20_update_bandwidth( return true; } -static void dcn20_enable_writeback( +void dcn20_enable_writeback( struct dc *dc, - const struct dc_stream_status *stream_status, - struct dc_writeback_info *wb_info) + struct dc_writeback_info *wb_info, + struct dc_state *context) { struct dwbc *dwb; struct mcif_wb *mcif_wb; @@ -1492,12 +1759,11 @@ static void dcn20_enable_writeback( mcif_wb = dc->res_pool->mcif_wb[wb_info->dwb_pipe_inst]; /* set the OPTC source mux */ - ASSERT(stream_status->primary_otg_inst < MAX_PIPES); - optc = dc->res_pool->timing_generators[stream_status->primary_otg_inst]; + optc = dc->res_pool->timing_generators[dwb->otg_inst]; optc->funcs->set_dwb_source(optc, wb_info->dwb_pipe_inst); /* set MCIF_WB buffer and arbitration configuration */ mcif_wb->funcs->config_mcif_buf(mcif_wb, &wb_info->mcif_buf_params, wb_info->dwb_params.dest_height); - mcif_wb->funcs->config_mcif_arb(mcif_wb, &dc->current_state->bw_ctx.bw.dcn.bw_writeback.mcif_wb_arb[wb_info->dwb_pipe_inst]); + mcif_wb->funcs->config_mcif_arb(mcif_wb, &context->bw_ctx.bw.dcn.bw_writeback.mcif_wb_arb[wb_info->dwb_pipe_inst]); /* Enable MCIF_WB */ mcif_wb->funcs->enable_mcif(mcif_wb); /* Enable DWB */ @@ -1520,7 +1786,7 @@ void dcn20_disable_writeback( mcif_wb->funcs->disable_mcif(mcif_wb); } -bool dcn20_hwss_wait_for_blank_complete( +bool dcn20_wait_for_blank_complete( struct output_pixel_processor *opp) { int counter; @@ -1549,32 +1815,34 @@ bool dcn20_dmdata_status_done(struct pipe_ctx *pipe_ctx) return hubp->funcs->dmdata_status_done(hubp); } -static void dcn20_disable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn20_disable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx) { -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct dce_hwseq *hws = dc->hwseq; - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); if (pipe_ctx->stream_res.dsc) { + struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe; + dcn20_dsc_pg_control(hws, pipe_ctx->stream_res.dsc->inst, true); - if (bot_odm_pipe) - dcn20_dsc_pg_control(hws, bot_odm_pipe->stream_res.dsc->inst, true); + while (odm_pipe) { + dcn20_dsc_pg_control(hws, odm_pipe->stream_res.dsc->inst, true); + odm_pipe = odm_pipe->next_odm_pipe; + } } -#endif } -static void dcn20_enable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn20_enable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx) { -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct dce_hwseq *hws = dc->hwseq; - struct pipe_ctx *bot_odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); if (pipe_ctx->stream_res.dsc) { + struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe; + dcn20_dsc_pg_control(hws, pipe_ctx->stream_res.dsc->inst, false); - if (bot_odm_pipe) - dcn20_dsc_pg_control(hws, bot_odm_pipe->stream_res.dsc->inst, false); + while (odm_pipe) { + dcn20_dsc_pg_control(hws, odm_pipe->stream_res.dsc->inst, false); + odm_pipe = odm_pipe->next_odm_pipe; + } } -#endif } void dcn20_set_dmdata_attributes(struct pipe_ctx *pipe_ctx) @@ -1597,12 +1865,7 @@ void dcn20_set_dmdata_attributes(struct pipe_ctx *pipe_ctx) hubp->funcs->dmdata_set_attributes(hubp, &attr); } -void dcn20_disable_stream(struct pipe_ctx *pipe_ctx, int option) -{ - dce110_disable_stream(pipe_ctx, option); -} - -static void dcn20_init_vm_ctx( +void dcn20_init_vm_ctx( struct dce_hwseq *hws, struct dc *dc, struct dc_virtual_addr_space_config *va_config, @@ -1624,7 +1887,7 @@ static void dcn20_init_vm_ctx( dc->res_pool->hubbub->funcs->init_vm_ctx(dc->res_pool->hubbub, &config, vmid); } -static int dcn20_init_sys_ctx(struct dce_hwseq *hws, struct dc *dc, struct dc_phy_addr_space_config *pa_config) +int dcn20_init_sys_ctx(struct dce_hwseq *hws, struct dc *dc, struct dc_phy_addr_space_config *pa_config) { struct dcn_hubbub_phys_addr_config config; @@ -1637,6 +1900,7 @@ static int dcn20_init_sys_ctx(struct dce_hwseq *hws, struct dc *dc, struct dc_ph config.gart_config.page_table_start_addr = pa_config->gart_config.page_table_start_addr; config.gart_config.page_table_end_addr = pa_config->gart_config.page_table_end_addr; config.gart_config.page_table_base_addr = pa_config->gart_config.page_table_base_addr; + config.page_table_default_page_addr = pa_config->page_table_default_page_addr; return dc->res_pool->hubbub->funcs->init_dchub_sys_ctx(dc->res_pool->hubbub, &config); } @@ -1667,8 +1931,7 @@ static bool patch_address_for_sbs_tb_stereo( return false; } - -static void dcn20_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn20_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_ctx) { bool addr_patched = false; PHYSICAL_ADDRESS_LOC addr; @@ -1702,30 +1965,35 @@ void dcn20_unblank_stream(struct pipe_ctx *pipe_ctx, struct encoder_unblank_param params = { { 0 } }; struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; - params.odm = dc_res_get_odm_bottom_pipe(pipe_ctx); + struct dce_hwseq *hws = link->dc->hwseq; + struct pipe_ctx *odm_pipe; + params.opp_cnt = 1; + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + params.opp_cnt++; + } /* only 3 items below are used by unblank */ params.timing = pipe_ctx->stream->timing; params.link_settings.link_rate = link_settings->link_rate; if (dc_is_dp_signal(pipe_ctx->stream->signal)) { - if (optc1_is_two_pixels_per_containter(&stream->timing) || params.odm) + if (optc2_is_two_pixels_per_containter(&stream->timing) || params.opp_cnt > 1) params.timing.pix_clk_100hz /= 2; pipe_ctx->stream_res.stream_enc->funcs->dp_set_odm_combine( - pipe_ctx->stream_res.stream_enc, params.odm); + pipe_ctx->stream_res.stream_enc, params.opp_cnt > 1); pipe_ctx->stream_res.stream_enc->funcs->dp_unblank(pipe_ctx->stream_res.stream_enc, ¶ms); } if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) { - link->dc->hwss.edp_backlight_control(link, true); + hws->funcs.edp_backlight_control(link, true); } } -void dcn20_setup_vupdate_interrupt(struct pipe_ctx *pipe_ctx) +void dcn20_setup_vupdate_interrupt(struct dc *dc, struct pipe_ctx *pipe_ctx) { struct timing_generator *tg = pipe_ctx->stream_res.tg; - int start_line = get_vupdate_offset_from_vsync(pipe_ctx); + int start_line = dc->hwss.get_vupdate_offset_from_vsync(pipe_ctx); if (start_line < 0) start_line = 0; @@ -1740,6 +2008,7 @@ static void dcn20_reset_back_end_for_pipe( struct dc_state *context) { int i; + struct dc_link *link; DC_LOGGER_INIT(dc->ctx->logger); if (pipe_ctx->stream_res.stream_enc == NULL) { pipe_ctx->stream = NULL; @@ -1747,17 +2016,36 @@ static void dcn20_reset_back_end_for_pipe( } if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - /* DPMS may already disable */ - if (!pipe_ctx->stream->dpms_off) - core_link_disable_stream(pipe_ctx, FREE_ACQUIRED_RESOURCE); - else if (pipe_ctx->stream_res.audio) { - dc->hwss.disable_audio_stream(pipe_ctx, FREE_ACQUIRED_RESOURCE); + link = pipe_ctx->stream->link; + /* DPMS may already disable or */ + /* dpms_off status is incorrect due to fastboot + * feature. When system resume from S4 with second + * screen only, the dpms_off would be true but + * VBIOS lit up eDP, so check link status too. + */ + if (!pipe_ctx->stream->dpms_off || link->link_status.link_active) + core_link_disable_stream(pipe_ctx); + else if (pipe_ctx->stream_res.audio) + dc->hwss.disable_audio_stream(pipe_ctx); + + /* free acquired resources */ + if (pipe_ctx->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx->stream_res.audio, false); + pipe_ctx->stream_res.audio = NULL; + } } } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - else if (pipe_ctx->stream_res.dsc) + else if (pipe_ctx->stream_res.dsc) { dp_set_dsc_enable(pipe_ctx, false); -#endif + } /* by upper caller loop, parent pipe: pipe0, will be reset last. * back end share by all pipes and will be disable only when disable @@ -1770,6 +2058,10 @@ static void dcn20_reset_back_end_for_pipe( if (pipe_ctx->stream_res.tg->funcs->set_odm_bypass) pipe_ctx->stream_res.tg->funcs->set_odm_bypass( pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); + + if (pipe_ctx->stream_res.tg->funcs->set_drr) + pipe_ctx->stream_res.tg->funcs->set_drr( + pipe_ctx->stream_res.tg, NULL); } for (i = 0; i < dc->res_pool->pipe_count; i++) @@ -1784,11 +2076,12 @@ static void dcn20_reset_back_end_for_pipe( pipe_ctx->pipe_idx, pipe_ctx->stream_res.tg->inst); } -static void dcn20_reset_hw_ctx_wrap( +void dcn20_reset_hw_ctx_wrap( struct dc *dc, struct dc_state *context) { int i; + struct dce_hwseq *hws = dc->hwseq; /* Reset Back End*/ for (i = dc->res_pool->pipe_count - 1; i >= 0 ; i--) { @@ -1799,7 +2092,7 @@ static void dcn20_reset_hw_ctx_wrap( if (!pipe_ctx_old->stream) continue; - if (pipe_ctx_old->top_pipe) + if (pipe_ctx_old->top_pipe || pipe_ctx_old->prev_odm_pipe) continue; if (!pipe_ctx->stream || @@ -1807,19 +2100,42 @@ static void dcn20_reset_hw_ctx_wrap( struct clock_source *old_clk = pipe_ctx_old->clock_source; dcn20_reset_back_end_for_pipe(dc, pipe_ctx_old, dc->current_state); - if (dc->hwss.enable_stream_gating) - dc->hwss.enable_stream_gating(dc, pipe_ctx); + if (hws->funcs.enable_stream_gating) + hws->funcs.enable_stream_gating(dc, pipe_ctx); if (old_clk) old_clk->funcs->cs_power_down(old_clk); } } } -static void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn20_get_mpctree_visual_confirm_color( + struct pipe_ctx *pipe_ctx, + struct tg_color *color) +{ + const struct tg_color pipe_colors[6] = { + {MAX_TG_COLOR_VALUE, 0, 0}, // red + {MAX_TG_COLOR_VALUE, 0, MAX_TG_COLOR_VALUE}, // yellow + {0, MAX_TG_COLOR_VALUE, 0}, // blue + {MAX_TG_COLOR_VALUE / 2, 0, MAX_TG_COLOR_VALUE / 2}, // purple + {0, 0, MAX_TG_COLOR_VALUE}, // green + {MAX_TG_COLOR_VALUE, MAX_TG_COLOR_VALUE * 2 / 3, 0}, // orange + }; + + struct pipe_ctx *top_pipe = pipe_ctx; + + while (top_pipe->top_pipe) { + top_pipe = top_pipe->top_pipe; + } + + *color = pipe_colors[top_pipe->pipe_idx]; +} + +void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) { + struct dce_hwseq *hws = dc->hwseq; struct hubp *hubp = pipe_ctx->plane_res.hubp; struct mpcc_blnd_cfg blnd_cfg = { {0} }; - bool per_pixel_alpha = pipe_ctx->plane_state->per_pixel_alpha && pipe_ctx->bottom_pipe; + bool per_pixel_alpha = pipe_ctx->plane_state->per_pixel_alpha; int mpcc_id; struct mpcc *new_mpcc; struct mpc *mpc = dc->res_pool->mpc; @@ -1827,10 +2143,13 @@ static void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) // input to MPCC is always RGB, by default leave black_color at 0 if (dc->debug.visual_confirm == VISUAL_CONFIRM_HDR) { - dcn10_get_hdr_visual_confirm_color( + hws->funcs.get_hdr_visual_confirm_color( pipe_ctx, &blnd_cfg.black_color); } else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SURFACE) { - dcn10_get_surface_visual_confirm_color( + hws->funcs.get_surface_visual_confirm_color( + pipe_ctx, &blnd_cfg.black_color); + } else if (dc->debug.visual_confirm == VISUAL_CONFIRM_MPCTREE) { + dcn20_get_mpctree_visual_confirm_color( pipe_ctx, &blnd_cfg.black_color); } @@ -1864,12 +2183,6 @@ static void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) */ mpcc_id = hubp->inst; - /* If there is no full update, don't need to touch MPC tree*/ - if (!pipe_ctx->plane_state->update_flags.bits.full_update) { - mpc->funcs->update_blending(mpc, &blnd_cfg, mpcc_id); - return; - } - /* check if this MPCC is already being used */ new_mpcc = mpc->funcs->get_mpcc_for_dpp(mpc_tree_params, mpcc_id); /* remove MPCC if being used */ @@ -1894,156 +2207,191 @@ static void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) hubp->mpcc_id = mpcc_id; } -static int find_free_gsl_group(const struct dc *dc) +void dcn20_enable_stream(struct pipe_ctx *pipe_ctx) { - if (dc->res_pool->gsl_groups.gsl_0 == 0) - return 1; - if (dc->res_pool->gsl_groups.gsl_1 == 0) - return 2; - if (dc->res_pool->gsl_groups.gsl_2 == 0) - return 3; + enum dc_lane_count lane_count = + pipe_ctx->stream->link->cur_link_settings.lane_count; - return 0; -} + struct dc_crtc_timing *timing = &pipe_ctx->stream->timing; + struct dc_link *link = pipe_ctx->stream->link; -/* NOTE: This is not a generic setup_gsl function (hence the suffix as_lock) - * This is only used to lock pipes in pipe splitting case with immediate flip - * Ordinary MPC/OTG locks suppress VUPDATE which doesn't help with immediate, - * so we get tearing with freesync since we cannot flip multiple pipes - * atomically. - * We use GSL for this: - * - immediate flip: find first available GSL group if not already assigned - * program gsl with that group, set current OTG as master - * and always us 0x4 = AND of flip_ready from all pipes - * - vsync flip: disable GSL if used - * - * Groups in stream_res are stored as +1 from HW registers, i.e. - * gsl_0 <=> pipe_ctx->stream_res.gsl_group == 1 - * Using a magic value like -1 would require tracking all inits/resets - */ -void dcn20_setup_gsl_group_as_lock( - const struct dc *dc, - struct pipe_ctx *pipe_ctx, - bool enable) -{ - struct gsl_params gsl; - int group_idx; + uint32_t active_total_with_borders; + uint32_t early_control = 0; + struct timing_generator *tg = pipe_ctx->stream_res.tg; - memset(&gsl, 0, sizeof(struct gsl_params)); + /* For MST, there are multiply stream go to only one link. + * connect DIG back_end to front_end while enable_stream and + * disconnect them during disable_stream + * BY this, it is logic clean to separate stream and link + */ + link->link_enc->funcs->connect_dig_be_to_fe(link->link_enc, + pipe_ctx->stream_res.stream_enc->id, true); - if (enable) { - /* return if group already assigned since GSL was set up - * for vsync flip, we would unassign so it can't be "left over" - */ - if (pipe_ctx->stream_res.gsl_group > 0) - return; + if (pipe_ctx->plane_state && pipe_ctx->plane_state->flip_immediate != 1) { + if (link->dc->hwss.program_dmdata_engine) + link->dc->hwss.program_dmdata_engine(pipe_ctx); + } - group_idx = find_free_gsl_group(dc); - ASSERT(group_idx != 0); - pipe_ctx->stream_res.gsl_group = group_idx; + link->dc->hwss.update_info_frame(pipe_ctx); - /* set gsl group reg field and mark resource used */ - switch (group_idx) { - case 1: - gsl.gsl0_en = 1; - dc->res_pool->gsl_groups.gsl_0 = 1; - break; - case 2: - gsl.gsl1_en = 1; - dc->res_pool->gsl_groups.gsl_1 = 1; - break; - case 3: - gsl.gsl2_en = 1; - dc->res_pool->gsl_groups.gsl_2 = 1; - break; - default: - BREAK_TO_DEBUGGER(); - return; // invalid case - } - gsl.gsl_master_en = 1; - } else { - group_idx = pipe_ctx->stream_res.gsl_group; - if (group_idx == 0) - return; // if not in use, just return + /* enable early control to avoid corruption on DP monitor*/ + active_total_with_borders = + timing->h_addressable + + timing->h_border_left + + timing->h_border_right; - pipe_ctx->stream_res.gsl_group = 0; + if (lane_count != 0) + early_control = active_total_with_borders % lane_count; - /* unset gsl group reg field and mark resource free */ - switch (group_idx) { - case 1: - gsl.gsl0_en = 0; - dc->res_pool->gsl_groups.gsl_0 = 0; - break; - case 2: - gsl.gsl1_en = 0; - dc->res_pool->gsl_groups.gsl_1 = 0; - break; - case 3: - gsl.gsl2_en = 0; - dc->res_pool->gsl_groups.gsl_2 = 0; - break; - default: - BREAK_TO_DEBUGGER(); - return; - } - gsl.gsl_master_en = 0; - } + if (early_control == 0) + early_control = lane_count; - /* at this point we want to program whether it's to enable or disable */ - if (pipe_ctx->stream_res.tg->funcs->set_gsl != NULL && - pipe_ctx->stream_res.tg->funcs->set_gsl_source_select != NULL) { - pipe_ctx->stream_res.tg->funcs->set_gsl( - pipe_ctx->stream_res.tg, - &gsl); + tg->funcs->set_early_control(tg, early_control); - pipe_ctx->stream_res.tg->funcs->set_gsl_source_select( - pipe_ctx->stream_res.tg, group_idx, enable ? 4 : 0); - } else - BREAK_TO_DEBUGGER(); + /* enable audio only within mode set */ + if (pipe_ctx->stream_res.audio != NULL) { + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + pipe_ctx->stream_res.stream_enc->funcs->dp_audio_enable(pipe_ctx->stream_res.stream_enc); + } } -static void dcn20_set_flip_control_gsl( - struct pipe_ctx *pipe_ctx, - bool flip_immediate) +void dcn20_program_dmdata_engine(struct pipe_ctx *pipe_ctx) { - if (pipe_ctx && pipe_ctx->plane_res.hubp->funcs->hubp_set_flip_control_surface_gsl) - pipe_ctx->plane_res.hubp->funcs->hubp_set_flip_control_surface_gsl( - pipe_ctx->plane_res.hubp, flip_immediate); + struct dc_stream_state *stream = pipe_ctx->stream; + struct hubp *hubp = pipe_ctx->plane_res.hubp; + bool enable = false; + struct stream_encoder *stream_enc = pipe_ctx->stream_res.stream_enc; + enum dynamic_metadata_mode mode = dc_is_dp_signal(stream->signal) + ? dmdata_dp + : dmdata_hdmi; + + /* if using dynamic meta, don't set up generic infopackets */ + if (pipe_ctx->stream->dmdata_address.quad_part != 0) { + pipe_ctx->stream_res.encoder_info_frame.hdrsmd.valid = false; + enable = true; + } + + if (!hubp) + return; + if (!stream_enc || !stream_enc->funcs->set_dynamic_metadata) + return; + + stream_enc->funcs->set_dynamic_metadata(stream_enc, enable, + hubp->inst, mode); } -void dcn20_hw_sequencer_construct(struct dc *dc) +void dcn20_fpga_init_hw(struct dc *dc) { - dcn10_hw_sequencer_construct(dc); - dc->hwss.init_hw = dcn20_init_hw; - dc->hwss.init_pipes = NULL; - dc->hwss.unblank_stream = dcn20_unblank_stream; - dc->hwss.update_plane_addr = dcn20_update_plane_addr; - dc->hwss.disable_plane = dcn20_disable_plane, - dc->hwss.enable_stream_timing = dcn20_enable_stream_timing; - dc->hwss.program_triplebuffer = dcn20_program_tripleBuffer; - dc->hwss.set_input_transfer_func = dcn20_set_input_transfer_func; - dc->hwss.set_output_transfer_func = dcn20_set_output_transfer_func; - dc->hwss.apply_ctx_for_surface = dcn20_apply_ctx_for_surface; - dc->hwss.pipe_control_lock = dcn20_pipe_control_lock; - dc->hwss.pipe_control_lock_global = dcn20_pipe_control_lock_global; - dc->hwss.optimize_bandwidth = dcn20_optimize_bandwidth; - dc->hwss.prepare_bandwidth = dcn20_prepare_bandwidth; - dc->hwss.update_bandwidth = dcn20_update_bandwidth; - dc->hwss.enable_writeback = dcn20_enable_writeback; - dc->hwss.disable_writeback = dcn20_disable_writeback; - dc->hwss.program_output_csc = dcn20_program_output_csc; - dc->hwss.update_odm = dcn20_update_odm; - dc->hwss.blank_pixel_data = dcn20_blank_pixel_data; - dc->hwss.dmdata_status_done = dcn20_dmdata_status_done; - dc->hwss.disable_stream = dcn20_disable_stream; - dc->hwss.init_sys_ctx = dcn20_init_sys_ctx; - dc->hwss.init_vm_ctx = dcn20_init_vm_ctx; - dc->hwss.disable_stream_gating = dcn20_disable_stream_gating; - dc->hwss.enable_stream_gating = dcn20_enable_stream_gating; - dc->hwss.setup_vupdate_interrupt = dcn20_setup_vupdate_interrupt; - dc->hwss.reset_hw_ctx_wrap = dcn20_reset_hw_ctx_wrap; - dc->hwss.update_mpcc = dcn20_update_mpcc; - dc->hwss.set_flip_control_gsl = dcn20_set_flip_control_gsl; - dc->hwss.did_underflow_occur = dcn10_did_underflow_occur; + int i, j; + struct dce_hwseq *hws = dc->hwseq; + struct resource_pool *res_pool = dc->res_pool; + struct dc_state *context = dc->current_state; + + if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) + dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); + + // Initialize the dccg + if (res_pool->dccg->funcs->dccg_init) + res_pool->dccg->funcs->dccg_init(res_pool->dccg); + + //Enable ability to power gate / don't force power on permanently + hws->funcs.enable_power_gating_plane(hws, true); + + // Specific to FPGA dccg and registers + REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF); + REG_WRITE(RBBMIF_TIMEOUT_DIS_2, 0xFFFFFFFF); + + hws->funcs.dccg_init(hws); + + REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2); + REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); + REG_WRITE(REFCLK_CNTL, 0); + // + + + /* Blank pixel data with OPP DPG */ + for (i = 0; i < dc->res_pool->timing_generator_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + + if (tg->funcs->is_tg_enabled(tg)) + dcn20_init_blank(dc, tg); + } + + for (i = 0; i < res_pool->timing_generator_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + + if (tg->funcs->is_tg_enabled(tg)) + tg->funcs->lock(tg); + } + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct dpp *dpp = res_pool->dpps[i]; + + dpp->funcs->dpp_reset(dpp); + } + + /* Reset all MPCC muxes */ + res_pool->mpc->funcs->mpc_init(res_pool->mpc); + + /* initialize OPP mpc_tree parameter */ + for (i = 0; i < dc->res_pool->res_cap->num_opp; i++) { + res_pool->opps[i]->mpc_tree_params.opp_id = res_pool->opps[i]->inst; + res_pool->opps[i]->mpc_tree_params.opp_list = NULL; + for (j = 0; j < MAX_PIPES; j++) + res_pool->opps[i]->mpcc_disconnect_pending[j] = false; + } + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + struct hubp *hubp = dc->res_pool->hubps[i]; + struct dpp *dpp = dc->res_pool->dpps[i]; + + pipe_ctx->stream_res.tg = tg; + pipe_ctx->pipe_idx = i; + + pipe_ctx->plane_res.hubp = hubp; + pipe_ctx->plane_res.dpp = dpp; + pipe_ctx->plane_res.mpcc_inst = dpp->inst; + hubp->mpcc_id = dpp->inst; + hubp->opp_id = OPP_ID_INVALID; + hubp->power_gated = false; + pipe_ctx->stream_res.opp = NULL; + + hubp->funcs->hubp_init(hubp); + + //dc->res_pool->opps[i]->mpc_tree_params.opp_id = dc->res_pool->opps[i]->inst; + //dc->res_pool->opps[i]->mpc_tree_params.opp_list = NULL; + dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; + pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; + /*to do*/ + hws->funcs.plane_atomic_disconnect(dc, pipe_ctx); + } + + /* initialize DWB pointer to MCIF_WB */ + for (i = 0; i < res_pool->res_cap->num_dwb; i++) + res_pool->dwbc[i]->mcif = res_pool->mcif_wb[i]; + + for (i = 0; i < dc->res_pool->timing_generator_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + + if (tg->funcs->is_tg_enabled(tg)) + tg->funcs->unlock(tg); + } + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + dc->hwss.disable_plane(dc, pipe_ctx); + + pipe_ctx->stream_res.tg = NULL; + pipe_ctx->plane_res.hubp = NULL; + } + + for (i = 0; i < dc->res_pool->timing_generator_count; i++) { + struct timing_generator *tg = dc->res_pool->timing_generators[i]; + + tg->funcs->tg_init(tg); + } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h index 2b0409454073..02c9be5ebd47 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h @@ -26,78 +26,112 @@ #ifndef __DC_HWSS_DCN20_H__ #define __DC_HWSS_DCN20_H__ -struct dc; +#include "hw_sequencer_private.h" -void dcn20_hw_sequencer_construct(struct dc *dc); - -enum dc_status dcn20_enable_stream_timing( - struct pipe_ctx *pipe_ctx, - struct dc_state *context, - struct dc *dc); - -void dcn20_blank_pixel_data( +bool dcn20_set_blend_lut( + struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state); +bool dcn20_set_shaper_3dlut( + struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state); +void dcn20_program_front_end_for_ctx( struct dc *dc, - struct pipe_ctx *pipe_ctx, - bool blank); - + struct dc_state *context); +void dcn20_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx); +bool dcn20_set_input_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, + const struct dc_plane_state *plane_state); +bool dcn20_set_output_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, + const struct dc_stream_state *stream); void dcn20_program_output_csc(struct dc *dc, struct pipe_ctx *pipe_ctx, enum dc_color_space colorspace, uint16_t *matrix, int opp_id); - +void dcn20_enable_stream(struct pipe_ctx *pipe_ctx); +void dcn20_unblank_stream(struct pipe_ctx *pipe_ctx, + struct dc_link_settings *link_settings); +void dcn20_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_blank_pixel_data( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + bool blank); +void dcn20_pipe_control_lock( + struct dc *dc, + struct pipe_ctx *pipe, + bool lock); +void dcn20_pipe_control_lock_global( + struct dc *dc, + struct pipe_ctx *pipe, + bool lock); void dcn20_prepare_bandwidth( struct dc *dc, struct dc_state *context); - void dcn20_optimize_bandwidth( struct dc *dc, struct dc_state *context); - bool dcn20_update_bandwidth( struct dc *dc, struct dc_state *context); - +void dcn20_reset_hw_ctx_wrap( + struct dc *dc, + struct dc_state *context); +enum dc_status dcn20_enable_stream_timing( + struct pipe_ctx *pipe_ctx, + struct dc_state *context, + struct dc *dc); +void dcn20_disable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_enable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_setup_vupdate_interrupt(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_init_blank( + struct dc *dc, + struct timing_generator *tg); +void dcn20_disable_vga( + struct dce_hwseq *hws); +void dcn20_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_enable_power_gating_plane( + struct dce_hwseq *hws, + bool enable); +void dcn20_dpp_pg_control( + struct dce_hwseq *hws, + unsigned int dpp_inst, + bool power_on); +void dcn20_hubp_pg_control( + struct dce_hwseq *hws, + unsigned int hubp_inst, + bool power_on); +void dcn20_program_triple_buffer( + const struct dc *dc, + struct pipe_ctx *pipe_ctx, + bool enable_triple_buffer); +void dcn20_enable_writeback( + struct dc *dc, + struct dc_writeback_info *wb_info, + struct dc_state *context); void dcn20_disable_writeback( struct dc *dc, unsigned int dwb_pipe_inst); - -bool dcn20_hwss_wait_for_blank_complete( - struct output_pixel_processor *opp); - -bool dcn20_set_output_transfer_func(struct pipe_ctx *pipe_ctx, - const struct dc_stream_state *stream); - -bool dcn20_set_input_transfer_func(struct pipe_ctx *pipe_ctx, - const struct dc_plane_state *plane_state); - +void dcn20_update_odm(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx); bool dcn20_dmdata_status_done(struct pipe_ctx *pipe_ctx); - +void dcn20_program_dmdata_engine(struct pipe_ctx *pipe_ctx); void dcn20_set_dmdata_attributes(struct pipe_ctx *pipe_ctx); - -void dcn20_disable_stream(struct pipe_ctx *pipe_ctx, int option); - -void dcn20_program_tripleBuffer( - const struct dc *dc, +void dcn20_init_vm_ctx( + struct dce_hwseq *hws, + struct dc *dc, + struct dc_virtual_addr_space_config *va_config, + int vmid); +void dcn20_set_flip_control_gsl( struct pipe_ctx *pipe_ctx, - bool enableTripleBuffer); - -void dcn20_setup_vupdate_interrupt(struct pipe_ctx *pipe_ctx); - -void dcn20_pipe_control_lock_global( + bool flip_immediate); +void dcn20_dsc_pg_control( + struct dce_hwseq *hws, + unsigned int dsc_inst, + bool power_on); +void dcn20_fpga_init_hw(struct dc *dc); +bool dcn20_wait_for_blank_complete( + struct output_pixel_processor *opp); +void dcn20_dccg_init(struct dce_hwseq *hws); +int dcn20_init_sys_ctx(struct dce_hwseq *hws, struct dc *dc, - struct pipe_ctx *pipe, - bool lock); -void dcn20_setup_gsl_group_as_lock(const struct dc *dc, - struct pipe_ctx *pipe_ctx, - bool enable); -void dcn20_pipe_control_lock( - struct dc *dc, - struct pipe_ctx *pipe, - bool lock); -void dcn20_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx); -void dcn20_enable_plane( - struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct dc_state *context); + struct dc_phy_addr_space_config *pa_config); + #endif /* __DC_HWSS_DCN20_H__ */ + diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c new file mode 100644 index 000000000000..d51e02fdab4d --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c @@ -0,0 +1,133 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dce110/dce110_hw_sequencer.h" +#include "dcn10/dcn10_hw_sequencer.h" +#include "dcn20_hwseq.h" + +static const struct hw_sequencer_funcs dcn20_funcs = { + .program_gamut_remap = dcn10_program_gamut_remap, + .init_hw = dcn10_init_hw, + .apply_ctx_to_hw = dce110_apply_ctx_to_hw, + .apply_ctx_for_surface = NULL, + .program_front_end_for_ctx = dcn20_program_front_end_for_ctx, + .update_plane_addr = dcn20_update_plane_addr, + .update_dchub = dcn10_update_dchub, + .update_pending_status = dcn10_update_pending_status, + .program_output_csc = dcn20_program_output_csc, + .enable_accelerated_mode = dce110_enable_accelerated_mode, + .enable_timing_synchronization = dcn10_enable_timing_synchronization, + .enable_per_frame_crtc_position_reset = dcn10_enable_per_frame_crtc_position_reset, + .update_info_frame = dce110_update_info_frame, + .send_immediate_sdp_message = dcn10_send_immediate_sdp_message, + .enable_stream = dcn20_enable_stream, + .disable_stream = dce110_disable_stream, + .unblank_stream = dcn20_unblank_stream, + .blank_stream = dce110_blank_stream, + .enable_audio_stream = dce110_enable_audio_stream, + .disable_audio_stream = dce110_disable_audio_stream, + .disable_plane = dcn20_disable_plane, + .pipe_control_lock = dcn20_pipe_control_lock, + .pipe_control_lock_global = dcn20_pipe_control_lock_global, + .prepare_bandwidth = dcn20_prepare_bandwidth, + .optimize_bandwidth = dcn20_optimize_bandwidth, + .update_bandwidth = dcn20_update_bandwidth, + .set_drr = dcn10_set_drr, + .get_position = dcn10_get_position, + .set_static_screen_control = dcn10_set_static_screen_control, + .setup_stereo = dcn10_setup_stereo, + .set_avmute = dce110_set_avmute, + .log_hw_state = dcn10_log_hw_state, + .get_hw_state = dcn10_get_hw_state, + .clear_status_bits = dcn10_clear_status_bits, + .wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect, + .edp_power_control = dce110_edp_power_control, + .edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready, + .set_cursor_position = dcn10_set_cursor_position, + .set_cursor_attribute = dcn10_set_cursor_attribute, + .set_cursor_sdr_white_level = dcn10_set_cursor_sdr_white_level, + .setup_periodic_interrupt = dcn10_setup_periodic_interrupt, + .set_clock = dcn10_set_clock, + .get_clock = dcn10_get_clock, + .program_triplebuffer = dcn20_program_triple_buffer, + .enable_writeback = dcn20_enable_writeback, + .disable_writeback = dcn20_disable_writeback, + .dmdata_status_done = dcn20_dmdata_status_done, + .program_dmdata_engine = dcn20_program_dmdata_engine, + .set_dmdata_attributes = dcn20_set_dmdata_attributes, + .init_sys_ctx = dcn20_init_sys_ctx, + .init_vm_ctx = dcn20_init_vm_ctx, + .set_flip_control_gsl = dcn20_set_flip_control_gsl, + .get_vupdate_offset_from_vsync = dcn10_get_vupdate_offset_from_vsync, +}; + +static const struct hwseq_private_funcs dcn20_private_funcs = { + .init_pipes = dcn10_init_pipes, + .update_plane_addr = dcn20_update_plane_addr, + .plane_atomic_disconnect = dcn10_plane_atomic_disconnect, + .update_mpcc = dcn20_update_mpcc, + .set_input_transfer_func = dcn20_set_input_transfer_func, + .set_output_transfer_func = dcn20_set_output_transfer_func, + .power_down = dce110_power_down, + .enable_display_power_gating = dcn10_dummy_display_power_gating, + .blank_pixel_data = dcn20_blank_pixel_data, + .reset_hw_ctx_wrap = dcn20_reset_hw_ctx_wrap, + .enable_stream_timing = dcn20_enable_stream_timing, + .edp_backlight_control = dce110_edp_backlight_control, + .disable_stream_gating = dcn20_disable_stream_gating, + .enable_stream_gating = dcn20_enable_stream_gating, + .setup_vupdate_interrupt = dcn20_setup_vupdate_interrupt, + .did_underflow_occur = dcn10_did_underflow_occur, + .init_blank = dcn20_init_blank, + .disable_vga = dcn20_disable_vga, + .bios_golden_init = dcn10_bios_golden_init, + .plane_atomic_disable = dcn20_plane_atomic_disable, + .plane_atomic_power_down = dcn10_plane_atomic_power_down, + .enable_power_gating_plane = dcn20_enable_power_gating_plane, + .dpp_pg_control = dcn20_dpp_pg_control, + .hubp_pg_control = dcn20_hubp_pg_control, + .dsc_pg_control = NULL, + .update_odm = dcn20_update_odm, + .dsc_pg_control = dcn20_dsc_pg_control, + .get_surface_visual_confirm_color = dcn10_get_surface_visual_confirm_color, + .get_hdr_visual_confirm_color = dcn10_get_hdr_visual_confirm_color, + .set_hdr_multiplier = dcn10_set_hdr_multiplier, + .verify_allow_pstate_change_high = dcn10_verify_allow_pstate_change_high, + .wait_for_blank_complete = dcn20_wait_for_blank_complete, + .dccg_init = dcn20_dccg_init, + .set_blend_lut = dcn20_set_blend_lut, + .set_shaper_3dlut = dcn20_set_shaper_3dlut, +}; + +void dcn20_hw_sequencer_construct(struct dc *dc) +{ + dc->hwss = dcn20_funcs; + dc->hwseq->funcs = dcn20_private_funcs; + + if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + dc->hwss.init_hw = dcn20_fpga_init_hw; + dc->hwseq->funcs.init_pipes = NULL; + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.h new file mode 100644 index 000000000000..12277797cd71 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.h @@ -0,0 +1,33 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_DCN20_INIT_H__ +#define __DC_DCN20_INIT_H__ + +struct dc; + +void dcn20_hw_sequencer_construct(struct dc *dc); + +#endif /* __DC_DCN20_INIT_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c index f495582e9e87..e4ac73035c84 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c @@ -168,10 +168,8 @@ static struct mpll_cfg dcn2_mpll_cfg[] = { void enc2_fec_set_enable(struct link_encoder *enc, bool enable) { struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT DC_LOG_DSC("%s FEC at link encoder inst %d", enable ? "Enabling" : "Disabling", enc->id.enum_id); -#endif REG_UPDATE(DP_DPHY_CNTL, DPHY_FEC_EN, enable); } @@ -192,7 +190,6 @@ bool enc2_fec_is_active(struct link_encoder *enc) return (active != 0); } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT /* this function reads dsc related register fields to be logged later in dcn10_log_hw_state * into a dcn_dsc_state struct. */ @@ -203,8 +200,8 @@ void link_enc2_read_state(struct link_encoder *enc, struct link_enc_state *s) REG_GET(DP_DPHY_CNTL, DPHY_FEC_EN, &s->dphy_fec_en); REG_GET(DP_DPHY_CNTL, DPHY_FEC_READY_SHADOW, &s->dphy_fec_ready_shadow); REG_GET(DP_DPHY_CNTL, DPHY_FEC_ACTIVE_STATUS, &s->dphy_fec_active_status); + REG_GET(DP_LINK_CNTL, DP_LINK_TRAINING_COMPLETE, &s->dp_link_training_complete); } -#endif static bool update_cfg_data( struct dcn10_link_encoder *enc10, @@ -315,9 +312,7 @@ void enc2_hw_init(struct link_encoder *enc) } static const struct link_encoder_funcs dcn20_link_enc_funcs = { -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT .read_state = link_enc2_read_state, -#endif .validate_output_with_stream = dcn10_link_encoder_validate_output_with_stream, .hw_init = enc2_hw_init, @@ -341,6 +336,7 @@ static const struct link_encoder_funcs dcn20_link_enc_funcs = { .fec_set_enable = enc2_fec_set_enable, .fec_set_ready = enc2_fec_set_ready, .fec_is_active = enc2_fec_is_active, + .get_dig_mode = dcn10_get_dig_mode, .get_dig_frontend = dcn10_get_dig_frontend, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h index 3736b5548a25..8cab8107fd94 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h @@ -33,7 +33,142 @@ SRI(AUX_DPHY_TX_CONTROL, DP_AUX, id) #define UNIPHY_MASK_SH_LIST(mask_sh)\ - LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_LINK_ENABLE, mask_sh) + LE_SF(SYMCLKA_CLOCK_ENABLE, SYMCLKA_CLOCK_ENABLE, mask_sh),\ + LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_LINK_ENABLE, mask_sh),\ + LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_CHANNEL0_XBAR_SOURCE, mask_sh),\ + LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_CHANNEL1_XBAR_SOURCE, mask_sh),\ + LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_CHANNEL2_XBAR_SOURCE, mask_sh),\ + LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_CHANNEL3_XBAR_SOURCE, mask_sh) + +#define DPCS_MASK_SH_LIST(mask_sh)\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX0_CLK_RDY, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX0_DATA_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX1_CLK_RDY, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX1_DATA_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX2_CLK_RDY, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX2_DATA_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX3_CLK_RDY, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX3_DATA_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL4, RDPCS_PHY_DP_TX0_TERM_CTRL, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL4, RDPCS_PHY_DP_TX1_TERM_CTRL, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL4, RDPCS_PHY_DP_TX2_TERM_CTRL, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL4, RDPCS_PHY_DP_TX3_TERM_CTRL, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL11, RDPCS_PHY_DP_MPLLB_MULTIPLIER, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL5, RDPCS_PHY_DP_TX0_WIDTH, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL5, RDPCS_PHY_DP_TX0_RATE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL5, RDPCS_PHY_DP_TX1_WIDTH, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL5, RDPCS_PHY_DP_TX1_RATE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_TX2_PSTATE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_TX3_PSTATE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_TX2_MPLL_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_TX3_MPLL_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL7, RDPCS_PHY_DP_MPLLB_FRACN_QUOT, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL7, RDPCS_PHY_DP_MPLLB_FRACN_DEN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL8, RDPCS_PHY_DP_MPLLB_SSC_PEAK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL9, RDPCS_PHY_DP_MPLLB_SSC_UP_SPREAD, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL9, RDPCS_PHY_DP_MPLLB_SSC_STEPSIZE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL10, RDPCS_PHY_DP_MPLLB_FRACN_REM, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL11, RDPCS_PHY_DP_REF_CLK_MPLLB_DIV, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL11, RDPCS_PHY_HDMI_MPLLB_HDMI_DIV, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL12, RDPCS_PHY_DP_MPLLB_SSC_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL12, RDPCS_PHY_DP_MPLLB_DIV5_CLK_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL12, RDPCS_PHY_DP_MPLLB_TX_CLK_DIV, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL12, RDPCS_PHY_DP_MPLLB_WORD_DIV2_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL12, RDPCS_PHY_DP_MPLLB_STATE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL13, RDPCS_PHY_DP_MPLLB_DIV_CLK_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL13, RDPCS_PHY_DP_MPLLB_DIV_MULTIPLIER, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL14, RDPCS_PHY_DP_MPLLB_FRACN_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL14, RDPCS_PHY_DP_MPLLB_PMIX_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CNTL, RDPCS_TX_FIFO_LANE0_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CNTL, RDPCS_TX_FIFO_LANE1_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CNTL, RDPCS_TX_FIFO_LANE2_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CNTL, RDPCS_TX_FIFO_LANE3_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CNTL, RDPCS_TX_FIFO_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CNTL, RDPCS_TX_FIFO_RD_START_DELAY, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CLOCK_CNTL, RDPCS_EXT_REFCLK_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CLOCK_CNTL, RDPCS_SRAMCLK_BYPASS, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CLOCK_CNTL, RDPCS_SRAMCLK_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CLOCK_CNTL, RDPCS_SRAMCLK_CLOCK_ON, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CLOCK_CNTL, RDPCS_SYMCLK_DIV2_CLOCK_ON, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CLOCK_CNTL, RDPCS_SYMCLK_DIV2_GATE_DIS, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_CLOCK_CNTL, RDPCS_SYMCLK_DIV2_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX0_DISABLE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX1_DISABLE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX2_DISABLE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX3_DISABLE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX0_REQ, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX1_REQ, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX2_REQ, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX3_REQ, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX0_ACK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX1_ACK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX2_ACK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX3_ACK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX0_RESET, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX1_RESET, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX2_RESET, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL3, RDPCS_PHY_DP_TX3_RESET, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL0, RDPCS_PHY_RESET, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL0, RDPCS_PHY_CR_MUX_SEL, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL0, RDPCS_PHY_REF_RANGE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL0, RDPCS_SRAM_BYPASS, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL0, RDPCS_SRAM_EXT_LD_DONE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL0, RDPCS_PHY_HDMIMODE_ENABLE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL0, RDPCS_SRAM_INIT_DONE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL2, RDPCS_PHY_DP4_POR, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PLL_UPDATE_DATA, RDPCS_PLL_UPDATE_DATA, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_INTERRUPT_CONTROL, RDPCS_REG_FIFO_ERROR_MASK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_INTERRUPT_CONTROL, RDPCS_TX_FIFO_ERROR_MASK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_INTERRUPT_CONTROL, RDPCS_DPALT_DISABLE_TOGGLE_MASK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_INTERRUPT_CONTROL, RDPCS_DPALT_4LANE_TOGGLE_MASK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCS_TX_CR_ADDR, RDPCS_TX_CR_ADDR, mask_sh),\ + LE_SF(RDPCSTX0_RDPCS_TX_CR_DATA, RDPCS_TX_CR_DATA, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_DP_MPLLB_V2I, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_DP_TX0_EQ_MAIN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_DP_TX0_EQ_PRE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_DP_TX0_EQ_POST, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_DP_MPLLB_FREQ_VCO, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE1, RDPCS_PHY_DP_MPLLB_CP_INT, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE1, RDPCS_PHY_DP_MPLLB_CP_PROP, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE1, RDPCS_PHY_DP_TX1_EQ_MAIN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE1, RDPCS_PHY_DP_TX1_EQ_PRE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE1, RDPCS_PHY_DP_TX1_EQ_POST, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE2, RDPCS_PHY_DP_TX2_EQ_MAIN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE2, RDPCS_PHY_DP_TX2_EQ_PRE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE2, RDPCS_PHY_DP_TX2_EQ_POST, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DP_TX3_EQ_MAIN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DCO_FINETUNE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DCO_RANGE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DP_TX3_EQ_PRE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DP_TX3_EQ_POST, mask_sh),\ + LE_SF(DPCSTX0_DPCSTX_TX_CLOCK_CNTL, DPCS_SYMCLK_CLOCK_ON, mask_sh),\ + LE_SF(DPCSTX0_DPCSTX_TX_CLOCK_CNTL, DPCS_SYMCLK_GATE_DIS, mask_sh),\ + LE_SF(DPCSTX0_DPCSTX_TX_CLOCK_CNTL, DPCS_SYMCLK_EN, mask_sh),\ + LE_SF(DPCSTX0_DPCSTX_TX_CNTL, DPCS_TX_DATA_SWAP, mask_sh),\ + LE_SF(DPCSTX0_DPCSTX_TX_CNTL, DPCS_TX_DATA_ORDER_INVERT, mask_sh),\ + LE_SF(DPCSTX0_DPCSTX_TX_CNTL, DPCS_TX_FIFO_EN, mask_sh),\ + LE_SF(DPCSTX0_DPCSTX_TX_CNTL, DPCS_TX_FIFO_RD_START_DELAY, mask_sh),\ + LE_SF(DPCSTX0_DPCSTX_DEBUG_CONFIG, DPCS_DBG_CBUS_DIS, mask_sh) + +#define DPCS_DCN2_MASK_SH_LIST(mask_sh)\ + DPCS_MASK_SH_LIST(mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_RX_LD_VAL, RDPCS_PHY_RX_REF_LD_VAL, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_RX_LD_VAL, RDPCS_PHY_RX_VCO_LD_VAL, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DISABLE_ACK, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_TX0_PSTATE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_TX1_PSTATE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_TX0_MPLL_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_TX1_MPLL_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_REF_CLK_EN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL5, RDPCS_PHY_DP_TX2_WIDTH, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL5, RDPCS_PHY_DP_TX2_RATE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL5, RDPCS_PHY_DP_TX3_WIDTH, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL5, RDPCS_PHY_DP_TX3_RATE, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYA_SOFT_RESET, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYB_SOFT_RESET, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYC_SOFT_RESET, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYD_SOFT_RESET, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYE_SOFT_RESET, mask_sh) #define LINK_ENCODER_MASK_SH_LIST_DCN20(mask_sh)\ LINK_ENCODER_MASK_SH_LIST_DCN10(mask_sh),\ @@ -63,6 +198,49 @@ SRI(CLOCK_ENABLE, SYMCLK, id), \ SRI(CHANNEL_XBAR_CNTL, UNIPHY, id) +#define DPCS_DCN2_CMN_REG_LIST(id) \ + SRI(DIG_LANE_ENABLE, DIG, id), \ + SRI(TMDS_CTL_BITS, DIG, id), \ + SRI(RDPCSTX_PHY_CNTL3, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL4, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL5, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL6, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL7, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL8, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL9, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL10, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL11, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL12, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL13, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL14, RDPCSTX, id), \ + SRI(RDPCSTX_CNTL, RDPCSTX, id), \ + SRI(RDPCSTX_CLOCK_CNTL, RDPCSTX, id), \ + SRI(RDPCSTX_INTERRUPT_CONTROL, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL0, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_CNTL2, RDPCSTX, id), \ + SRI(RDPCSTX_PLL_UPDATE_DATA, RDPCSTX, id), \ + SRI(RDPCS_TX_CR_ADDR, RDPCSTX, id), \ + SRI(RDPCS_TX_CR_DATA, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_FUSE0, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_FUSE1, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_FUSE2, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_FUSE3, RDPCSTX, id), \ + SRI(DPCSTX_TX_CLOCK_CNTL, DPCSTX, id), \ + SRI(DPCSTX_TX_CNTL, DPCSTX, id), \ + SRI(DPCSTX_DEBUG_CONFIG, DPCSTX, id), \ + SRI(RDPCSTX_DEBUG_CONFIG, RDPCSTX, id), \ + SR(RDPCSTX0_RDPCSTX_SCRATCH) + + +#define DPCS_DCN2_REG_LIST(id) \ + DPCS_DCN2_CMN_REG_LIST(id), \ + SRI(RDPCSTX_PHY_RX_LD_VAL, RDPCSTX, id),\ + SRI(RDPCSTX_DMCU_DPALT_DIS_BLOCK_REG, RDPCSTX, id) + +#define LE_DCN2_REG_LIST(id) \ + LE_DCN10_REG_LIST(id), \ + SR(DCIO_SOFT_RESET) + struct mpll_cfg { uint32_t mpllb_ana_v2i; uint32_t mpllb_ana_freq_vco; @@ -91,6 +269,13 @@ struct mpll_cfg { uint32_t ref_range; uint32_t ref_clk; bool hdmimode_enable; + bool sup_pre_hp; + bool dp_tx0_vergdrv_byp; + bool dp_tx1_vergdrv_byp; + bool dp_tx2_vergdrv_byp; + bool dp_tx3_vergdrv_byp; + + }; struct dpcssys_phy_seq_cfg { @@ -151,9 +336,7 @@ void enc2_fec_set_ready(struct link_encoder *enc, bool ready); bool enc2_fec_is_active(struct link_encoder *enc); void enc2_hw_init(struct link_encoder *enc); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT void link_enc2_read_state(struct link_encoder *enc, struct link_enc_state *s); -#endif void dcn20_link_encoder_enable_dp_output( struct link_encoder *enc, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c index 240749e4cf83..de9c857ab3e9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c @@ -33,6 +33,9 @@ #define REG(reg)\ mpc20->mpc_regs->reg +#define IND_REG(index) \ + (index) + #define CTX \ mpc20->base.ctx @@ -132,19 +135,33 @@ void mpc2_set_output_csc( const uint16_t *regval, enum mpc_output_csc_mode ocsc_mode) { + uint32_t cur_mode; struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); struct color_matrices_reg ocsc_regs; - REG_SET(CSC_MODE[opp_id], 0, MPC_OCSC_MODE, ocsc_mode); - - if (ocsc_mode == MPC_OUTPUT_CSC_DISABLE) + if (ocsc_mode == MPC_OUTPUT_CSC_DISABLE) { + REG_SET(CSC_MODE[opp_id], 0, MPC_OCSC_MODE, ocsc_mode); return; + } if (regval == NULL) { BREAK_TO_DEBUGGER(); return; } + /* determine which CSC coefficients (A or B) we are using + * currently. select the alternate set to double buffer + * the CSC update so CSC is updated on frame boundary + */ + IX_REG_GET(MPC_OCSC_TEST_DEBUG_INDEX, MPC_OCSC_TEST_DEBUG_DATA, + MPC_OCSC_TEST_DEBUG_DATA_STATUS_IDX, + MPC_OCSC_TEST_DEBUG_DATA_OCSC_MODE, &cur_mode); + + if (cur_mode != MPC_OUTPUT_CSC_COEF_A) + ocsc_mode = MPC_OUTPUT_CSC_COEF_A; + else + ocsc_mode = MPC_OUTPUT_CSC_COEF_B; + ocsc_regs.shifts.csc_c11 = mpc20->mpc_shift->MPC_OCSC_C11_A; ocsc_regs.masks.csc_c11 = mpc20->mpc_mask->MPC_OCSC_C11_A; ocsc_regs.shifts.csc_c12 = mpc20->mpc_shift->MPC_OCSC_C12_A; @@ -157,10 +174,13 @@ void mpc2_set_output_csc( ocsc_regs.csc_c11_c12 = REG(CSC_C11_C12_B[opp_id]); ocsc_regs.csc_c33_c34 = REG(CSC_C33_C34_B[opp_id]); } + cm_helper_program_color_matrices( mpc20->base.ctx, regval, &ocsc_regs); + + REG_SET(CSC_MODE[opp_id], 0, MPC_OCSC_MODE, ocsc_mode); } void mpc2_set_ocsc_default( @@ -169,14 +189,16 @@ void mpc2_set_ocsc_default( enum dc_color_space color_space, enum mpc_output_csc_mode ocsc_mode) { + uint32_t cur_mode; struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); uint32_t arr_size; struct color_matrices_reg ocsc_regs; const uint16_t *regval = NULL; - REG_SET(CSC_MODE[opp_id], 0, MPC_OCSC_MODE, ocsc_mode); - if (ocsc_mode == MPC_OUTPUT_CSC_DISABLE) + if (ocsc_mode == MPC_OUTPUT_CSC_DISABLE) { + REG_SET(CSC_MODE[opp_id], 0, MPC_OCSC_MODE, ocsc_mode); return; + } regval = find_color_matrix(color_space, &arr_size); @@ -185,6 +207,19 @@ void mpc2_set_ocsc_default( return; } + /* determine which CSC coefficients (A or B) we are using + * currently. select the alternate set to double buffer + * the CSC update so CSC is updated on frame boundary + */ + IX_REG_GET(MPC_OCSC_TEST_DEBUG_INDEX, MPC_OCSC_TEST_DEBUG_DATA, + MPC_OCSC_TEST_DEBUG_DATA_STATUS_IDX, + MPC_OCSC_TEST_DEBUG_DATA_OCSC_MODE, &cur_mode); + + if (cur_mode != MPC_OUTPUT_CSC_COEF_A) + ocsc_mode = MPC_OUTPUT_CSC_COEF_A; + else + ocsc_mode = MPC_OUTPUT_CSC_COEF_B; + ocsc_regs.shifts.csc_c11 = mpc20->mpc_shift->MPC_OCSC_C11_A; ocsc_regs.masks.csc_c11 = mpc20->mpc_mask->MPC_OCSC_C11_A; ocsc_regs.shifts.csc_c12 = mpc20->mpc_shift->MPC_OCSC_C12_A; @@ -203,6 +238,8 @@ void mpc2_set_ocsc_default( mpc20->base.ctx, regval, &ocsc_regs); + + REG_SET(CSC_MODE[opp_id], 0, MPC_OCSC_MODE, ocsc_mode); } static void mpc2_ogam_get_reg_field( @@ -233,14 +270,14 @@ static void mpc2_ogam_get_reg_field( reg->masks.exp_resion_start_segment = mpc20->mpc_mask->MPCC_OGAM_RAMA_EXP_REGION_START_SEGMENT_B; } -static void mpc20_power_on_ogam_lut( +void mpc20_power_on_ogam_lut( struct mpc *mpc, int mpcc_id, bool power_on) { struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); REG_SET(MPCC_MEM_PWR_CTRL[mpcc_id], 0, - MPCC_OGAM_MEM_PWR_FORCE, power_on == true ? 0:1); + MPCC_OGAM_MEM_PWR_DIS, power_on == true ? 1:0); } @@ -345,6 +382,9 @@ static void mpc20_program_ogam_pwl( uint32_t i; struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); + PERF_TRACE(); + REG_SEQ_START(); + for (i = 0 ; i < num; i++) { REG_SET(MPCC_OGAM_LUT_DATA[mpcc_id], 0, MPCC_OGAM_LUT_DATA, rgb[i].red_reg); REG_SET(MPCC_OGAM_LUT_DATA[mpcc_id], 0, MPCC_OGAM_LUT_DATA, rgb[i].green_reg); @@ -368,6 +408,11 @@ void apply_DEDCN20_305_wa( { struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); + if (mpc->ctx->dc->debug.cm_in_bypass) { + REG_SET(MPCC_OGAM_MODE[mpcc_id], 0, MPCC_OGAM_MODE, 0); + return; + } + if (mpc->ctx->dc->work_arounds.dedcn20_305_wa == false) { /*hw fixed in new review*/ return; @@ -390,10 +435,16 @@ void mpc2_set_output_gamma( enum dc_lut_mode next_mode; struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); + if (mpc->ctx->dc->debug.cm_in_bypass) { + REG_SET(MPCC_OGAM_MODE[mpcc_id], 0, MPCC_OGAM_MODE, 0); + return; + } + if (params == NULL) { REG_SET(MPCC_OGAM_MODE[mpcc_id], 0, MPCC_OGAM_MODE, 0); return; } + current_mode = mpc20_get_ogam_current(mpc, mpcc_id); if (current_mode == LUT_BYPASS || current_mode == LUT_RAM_A) next_mode = LUT_RAM_B; @@ -435,24 +486,28 @@ void mpc2_assert_mpcc_idle_before_connect(struct mpc *mpc, int mpcc_id) { struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); unsigned int top_sel, mpc_busy, mpc_idle, mpc_disabled; - REG_GET(MPCC_STATUS[mpcc_id], MPCC_DISABLED, &mpc_disabled); - - if (mpc_disabled) { - ASSERT(0); - return; - } REG_GET(MPCC_TOP_SEL[mpcc_id], MPCC_TOP_SEL, &top_sel); - if (top_sel == 0xf) { - REG_GET_2(MPCC_STATUS[mpcc_id], - MPCC_BUSY, &mpc_busy, - MPCC_IDLE, &mpc_idle); + REG_GET_3(MPCC_STATUS[mpcc_id], + MPCC_BUSY, &mpc_busy, + MPCC_IDLE, &mpc_idle, + MPCC_DISABLED, &mpc_disabled); - ASSERT(mpc_busy == 0); - ASSERT(mpc_idle == 1); + if (top_sel == 0xf) { + ASSERT(!mpc_busy); + ASSERT(mpc_idle); + ASSERT(mpc_disabled); + } else { + ASSERT(!mpc_disabled); + ASSERT(!mpc_idle); } + + REG_SEQ_SUBMIT(); + PERF_TRACE(); + REG_SEQ_WAIT_DONE(); + PERF_TRACE(); } static void mpc2_init_mpcc(struct mpcc *mpcc, int mpcc_inst) @@ -488,6 +543,7 @@ const struct mpc_funcs dcn20_mpc_funcs = { .insert_plane = mpc1_insert_plane, .remove_mpcc = mpc1_remove_mpcc, .mpc_init = mpc1_mpc_init, + .mpc_init_single_inst = mpc1_mpc_init_single_inst, .update_blending = mpc2_update_blending, .get_mpcc_for_dpp = mpc2_get_mpcc_for_dpp, .wait_for_idle = mpc2_assert_idle_mpcc, @@ -498,6 +554,7 @@ const struct mpc_funcs dcn20_mpc_funcs = { .set_output_csc = mpc2_set_output_csc, .set_ocsc_default = mpc2_set_ocsc_default, .set_output_gamma = mpc2_set_output_gamma, + .power_on_mpc_mem_pwr = mpc20_power_on_ogam_lut, }; void dcn20_mpc_construct(struct dcn20_mpc *mpc20, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h index 9750095d2d73..c78fd5123497 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h @@ -80,6 +80,10 @@ SRII(DENORM_CLAMP_G_Y, MPC_OUT, inst),\ SRII(DENORM_CLAMP_B_CB, MPC_OUT, inst) +#define MPC_DBG_REG_LIST_DCN2_0() \ + SR(MPC_OCSC_TEST_DEBUG_DATA),\ + SR(MPC_OCSC_TEST_DEBUG_INDEX) + #define MPC_REG_VARIABLE_LIST_DCN2_0 \ MPC_COMMON_REG_VARIABLE_LIST \ uint32_t MPCC_TOP_GAIN[MAX_MPCC]; \ @@ -118,6 +122,8 @@ uint32_t MPCC_OGAM_LUT_RAM_CONTROL[MAX_MPCC];\ uint32_t MPCC_OGAM_LUT_DATA[MAX_MPCC];\ uint32_t MPCC_OGAM_MODE[MAX_MPCC];\ + uint32_t MPC_OCSC_TEST_DEBUG_DATA;\ + uint32_t MPC_OCSC_TEST_DEBUG_INDEX;\ uint32_t CSC_MODE[MAX_OPP]; \ uint32_t CSC_C11_C12_A[MAX_OPP]; \ uint32_t CSC_C33_C34_A[MAX_OPP]; \ @@ -134,6 +140,7 @@ SF(MPCC0_MPCC_TOP_GAIN, MPCC_TOP_GAIN, mask_sh),\ SF(MPCC0_MPCC_BOT_GAIN_INSIDE, MPCC_BOT_GAIN_INSIDE, mask_sh),\ SF(MPCC0_MPCC_BOT_GAIN_OUTSIDE, MPCC_BOT_GAIN_OUTSIDE, mask_sh),\ + SF(MPC_OCSC_TEST_DEBUG_INDEX, MPC_OCSC_TEST_DEBUG_INDEX, mask_sh),\ SF(MPC_OUT0_CSC_MODE, MPC_OCSC_MODE, mask_sh),\ SF(MPC_OUT0_CSC_C11_C12_A, MPC_OCSC_C11_A, mask_sh),\ SF(MPC_OUT0_CSC_C11_C12_A, MPC_OCSC_C12_A, mask_sh),\ @@ -159,6 +166,7 @@ SF(MPCC_OGAM0_MPCC_OGAM_RAMB_START_CNTL_B, MPCC_OGAM_RAMB_EXP_REGION_START_B, mask_sh),\ SF(MPCC_OGAM0_MPCC_OGAM_RAMB_START_CNTL_B, MPCC_OGAM_RAMB_EXP_REGION_START_SEGMENT_B, mask_sh),\ SF(MPCC0_MPCC_MEM_PWR_CTRL, MPCC_OGAM_MEM_PWR_FORCE, mask_sh),\ + SF(MPCC0_MPCC_MEM_PWR_CTRL, MPCC_OGAM_MEM_PWR_DIS, mask_sh),\ SF(MPCC_OGAM0_MPCC_OGAM_LUT_INDEX, MPCC_OGAM_LUT_INDEX, mask_sh),\ SF(MPCC_OGAM0_MPCC_OGAM_LUT_RAM_CONTROL, MPCC_OGAM_LUT_WRITE_EN_MASK, mask_sh),\ SF(MPCC_OGAM0_MPCC_OGAM_LUT_RAM_CONTROL, MPCC_OGAM_LUT_RAM_SEL, mask_sh),\ @@ -173,6 +181,20 @@ SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MAX_B_CB, mask_sh),\ SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MIN_B_CB, mask_sh) +/* + * DCN2 MPC_OCSC debug status register: + * + * Status index including current OCSC Mode is 1 + * OCSC Mode: [1..0] + */ +#define MPC_OCSC_TEST_DEBUG_DATA_STATUS_IDX 1 + +#define MPC_DEBUG_REG_LIST_SH_DCN20 \ + .MPC_OCSC_TEST_DEBUG_DATA_OCSC_MODE = 0 + +#define MPC_DEBUG_REG_LIST_MASK_DCN20 \ + .MPC_OCSC_TEST_DEBUG_DATA_OCSC_MODE = 0x3 + #define MPC_REG_FIELD_LIST_DCN2_0(type) \ MPC_REG_FIELD_LIST(type)\ type MPCC_BG_BPC;\ @@ -180,6 +202,8 @@ type MPCC_TOP_GAIN;\ type MPCC_BOT_GAIN_INSIDE;\ type MPCC_BOT_GAIN_OUTSIDE;\ + type MPC_OCSC_TEST_DEBUG_DATA_OCSC_MODE;\ + type MPC_OCSC_TEST_DEBUG_INDEX;\ type MPC_OCSC_MODE;\ type MPC_OCSC_C11_A;\ type MPC_OCSC_C12_A;\ @@ -217,7 +241,8 @@ type MPC_OUT_DENORM_CLAMP_MIN_G_Y;\ type MPC_OUT_DENORM_CLAMP_MAX_B_CB;\ type MPC_OUT_DENORM_CLAMP_MIN_B_CB;\ - type MPCC_DISABLED; + type MPCC_DISABLED;\ + type MPCC_OGAM_MEM_PWR_DIS; struct dcn20_mpc_registers { MPC_REG_VARIABLE_LIST_DCN2_0 @@ -282,4 +307,5 @@ void mpc2_set_output_gamma( void mpc2_assert_idle_mpcc(struct mpc *mpc, int id); void mpc2_assert_mpcc_idle_before_connect(struct mpc *mpc, int mpcc_id); +void mpc20_power_on_ogam_lut(struct mpc *mpc, int mpcc_id, bool power_on); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c index d9e7c711a71c..023cc71fad0f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.c @@ -41,6 +41,7 @@ void opp2_set_disp_pattern_generator( struct output_pixel_processor *opp, enum controller_dp_test_pattern test_pattern, + enum controller_dp_color_space color_space, enum dc_color_depth color_depth, const struct tg_color *solid_color, int width, @@ -100,9 +101,22 @@ void opp2_set_disp_pattern_generator( TEST_PATTERN_DYN_RANGE_CEA : TEST_PATTERN_DYN_RANGE_VESA); + switch (color_space) { + case CONTROLLER_DP_COLOR_SPACE_YCBCR601: + mode = TEST_PATTERN_MODE_COLORSQUARES_YCBCR601; + break; + case CONTROLLER_DP_COLOR_SPACE_YCBCR709: + mode = TEST_PATTERN_MODE_COLORSQUARES_YCBCR709; + break; + case CONTROLLER_DP_COLOR_SPACE_RGB: + default: + mode = TEST_PATTERN_MODE_COLORSQUARES_RGB; + break; + } + REG_UPDATE_6(DPG_CONTROL, DPG_EN, 1, - DPG_MODE, TEST_PATTERN_MODE_COLORSQUARES_RGB, + DPG_MODE, mode, DPG_DYNAMIC_RANGE, dyn_range, DPG_BIT_DEPTH, bit_depth, DPG_VRES, 6, @@ -332,7 +346,6 @@ static struct opp_funcs dcn20_opp_funcs = { .opp_set_disp_pattern_generator = opp2_set_disp_pattern_generator, .dpg_is_blanked = opp2_dpg_is_blanked, .opp_dpg_set_blank_color = opp2_dpg_set_blank_color, - .opp_convert_pti = NULL, .opp_destroy = opp1_destroy, .opp_program_left_edge_extra_pixel = opp2_program_left_edge_extra_pixel, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h index abd8de9a78f8..4093bec172c1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_opp.h @@ -140,6 +140,7 @@ void dcn20_opp_construct(struct dcn20_opp *oppn20, void opp2_set_disp_pattern_generator( struct output_pixel_processor *opp, enum controller_dp_test_pattern test_pattern, + enum controller_dp_color_space color_space, enum dc_color_depth color_depth, const struct tg_color *solid_color, int width, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c index 1ae973962d53..d875b0c38fde 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c @@ -59,11 +59,16 @@ bool optc2_enable_crtc(struct timing_generator *optc) REG_UPDATE(CONTROL, VTG0_ENABLE, 1); + REG_SEQ_START(); + /* Enable CRTC */ REG_UPDATE_2(OTG_CONTROL, OTG_DISABLE_POINT_CNTL, 3, OTG_MASTER_EN, 1); + REG_SEQ_SUBMIT(); + REG_SEQ_WAIT_DONE(); + return true; } @@ -167,7 +172,6 @@ void optc2_set_gsl_source_select( } } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT /* DSC encoder frame start controls: x = h position, line_num = # of lines from vstartup */ void optc2_set_dsc_encoder_frame_start(struct timing_generator *optc, int x_position, @@ -191,15 +195,6 @@ void optc2_set_dsc_config(struct timing_generator *optc, uint32_t dsc_slice_width) { struct optc *optc1 = DCN10TG_FROM_TG(optc); - uint32_t data_format = 0; - /* skip if dsc mode is not changed */ - data_format = dm_read_reg(CTX, REG(OPTC_DATA_FORMAT_CONTROL)); - - data_format = data_format & 0x30; /* bit5:4 */ - data_format = data_format >> 4; - - if (data_format == dsc_mode) - return; REG_UPDATE(OPTC_DATA_FORMAT_CONTROL, OPTC_DSC_MODE, dsc_mode); @@ -210,13 +205,12 @@ void optc2_set_dsc_config(struct timing_generator *optc, REG_UPDATE(OPTC_WIDTH_CONTROL, OPTC_DSC_SLICE_WIDTH, dsc_slice_width); } -#endif -/** - * PTI i think is already done somewhere else for 2ka - * (opp?, please double check. - * OPTC side only has 1 register to set for PTI_ENABLE) - */ +/*TEMP: Need to figure out inheritance model here.*/ +bool optc2_is_two_pixels_per_containter(const struct dc_crtc_timing *timing) +{ + return optc1_is_two_pixels_per_containter(timing); +} void optc2_set_odm_bypass(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing) @@ -224,28 +218,31 @@ void optc2_set_odm_bypass(struct timing_generator *optc, struct optc *optc1 = DCN10TG_FROM_TG(optc); uint32_t h_div_2 = 0; - optc1->comb_opp_id = 0xf; REG_SET_3(OPTC_DATA_SOURCE_SELECT, 0, OPTC_NUM_OF_INPUT_SEGMENT, 0, OPTC_SEG0_SRC_SEL, optc->inst, OPTC_SEG1_SRC_SEL, 0xf); REG_WRITE(OTG_H_TIMING_CNTL, 0); - h_div_2 = optc1_is_two_pixels_per_containter(dc_crtc_timing); + h_div_2 = optc2_is_two_pixels_per_containter(dc_crtc_timing); REG_UPDATE(OTG_H_TIMING_CNTL, OTG_H_TIMING_DIV_BY2, h_div_2); REG_SET(OPTC_MEMORY_CONFIG, 0, OPTC_MEM_SEL, 0); + optc1->opp_count = 1; } -void optc2_set_odm_combine(struct timing_generator *optc, int combine_opp_id, - int mpcc_hactive, enum dc_pixel_encoding pixel_encoding) +void optc2_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_cnt, + struct dc_crtc_timing *timing) { struct optc *optc1 = DCN10TG_FROM_TG(optc); - /* 2 pieces of memory required for up to 5120 displays, 4 for up to 8192 */ - int memory_mask = mpcc_hactive <= 2560 ? 0x3 : 0xf; + int mpcc_hactive = (timing->h_addressable + timing->h_border_left + timing->h_border_right) + / opp_cnt; + uint32_t memory_mask; uint32_t data_fmt = 0; + ASSERT(opp_cnt == 2); + /* TODO: In pseudocode but does not affect maximus, delete comment if we dont need on asic * REG_SET(OTG_GLOBAL_CONTROL2, 0, GLOBAL_UPDATE_LOCK_EN, 1); * Program OTG register MASTER_UPDATE_LOCK_DB_X/Y to the position before DP frame start @@ -253,27 +250,35 @@ void optc2_set_odm_combine(struct timing_generator *optc, int combine_opp_id, * MASTER_UPDATE_LOCK_DB_X, 160, * MASTER_UPDATE_LOCK_DB_Y, 240); */ + + /* 2 pieces of memory required for up to 5120 displays, 4 for up to 8192, + * however, for ODM combine we can simplify by always using 4. + * To make sure there's no overlap, each instance "reserves" 2 memories and + * they are uniquely combined here. + */ + memory_mask = 0x3 << (opp_id[0] * 2) | 0x3 << (opp_id[1] * 2); + if (REG(OPTC_MEMORY_CONFIG)) REG_SET(OPTC_MEMORY_CONFIG, 0, - OPTC_MEM_SEL, memory_mask << (optc->inst * 4)); + OPTC_MEM_SEL, memory_mask); - if (pixel_encoding == PIXEL_ENCODING_YCBCR422) + if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422) data_fmt = 1; - else if (pixel_encoding == PIXEL_ENCODING_YCBCR420) + else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) data_fmt = 2; REG_UPDATE(OPTC_DATA_FORMAT_CONTROL, OPTC_DATA_FORMAT, data_fmt); REG_SET_3(OPTC_DATA_SOURCE_SELECT, 0, OPTC_NUM_OF_INPUT_SEGMENT, 1, - OPTC_SEG0_SRC_SEL, optc->inst, - OPTC_SEG1_SRC_SEL, combine_opp_id); + OPTC_SEG0_SRC_SEL, opp_id[0], + OPTC_SEG1_SRC_SEL, opp_id[1]); REG_UPDATE(OPTC_WIDTH_CONTROL, OPTC_SEGMENT_WIDTH, mpcc_hactive); REG_SET(OTG_H_TIMING_CNTL, 0, OTG_H_TIMING_DIV_BY2, 1); - optc1->comb_opp_id = combine_opp_id; + optc1->opp_count = opp_cnt; } void optc2_get_optc_source(struct timing_generator *optc, @@ -293,6 +298,10 @@ void optc2_get_optc_source(struct timing_generator *optc, *num_of_src_opp = 2; else *num_of_src_opp = 1; + + /* Work around VBIOS not updating OPTC_NUM_OF_INPUT_SEGMENT */ + if (*src_opp_id_1 == 0xf) + *num_of_src_opp = 1; } void optc2_set_dwb_source(struct timing_generator *optc, @@ -339,65 +348,6 @@ void optc2_triplebuffer_unlock(struct timing_generator *optc) } - -void optc2_setup_global_lock(struct timing_generator *optc) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - uint32_t v_blank_start = 0; - uint32_t h_blank_start = 0, h_total = 0; - - REG_SET(OTG_GLOBAL_CONTROL1, 0, MASTER_UPDATE_LOCK_DB_EN, 1); - - REG_SET(OTG_GLOBAL_CONTROL2, 0, DIG_UPDATE_LOCATION, 20); - - REG_GET(OTG_V_BLANK_START_END, OTG_V_BLANK_START, &v_blank_start); - - REG_GET(OTG_H_BLANK_START_END, OTG_H_BLANK_START, &h_blank_start); - - REG_GET(OTG_H_TOTAL, OTG_H_TOTAL, &h_total); - REG_UPDATE_2(OTG_GLOBAL_CONTROL1, - MASTER_UPDATE_LOCK_DB_X, - h_blank_start - 200 - 1, - MASTER_UPDATE_LOCK_DB_Y, - v_blank_start - 1); -} - -void optc2_lock_global(struct timing_generator *optc) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - - REG_UPDATE(OTG_GLOBAL_CONTROL2, GLOBAL_UPDATE_LOCK_EN, 1); - - REG_SET(OTG_GLOBAL_CONTROL0, 0, - OTG_MASTER_UPDATE_LOCK_SEL, optc->inst); - REG_SET(OTG_MASTER_UPDATE_LOCK, 0, - OTG_MASTER_UPDATE_LOCK, 1); - - /* Should be fast, status does not update on maximus */ - if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS) - REG_WAIT(OTG_MASTER_UPDATE_LOCK, - UPDATE_LOCK_STATUS, 1, - 1, 10); -} - -void optc2_lock(struct timing_generator *optc) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - - REG_UPDATE(OTG_GLOBAL_CONTROL2, GLOBAL_UPDATE_LOCK_EN, 0); - - REG_SET(OTG_GLOBAL_CONTROL0, 0, - OTG_MASTER_UPDATE_LOCK_SEL, optc->inst); - REG_SET(OTG_MASTER_UPDATE_LOCK, 0, - OTG_MASTER_UPDATE_LOCK, 1); - - /* Should be fast, status does not update on maximus */ - if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS) - REG_WAIT(OTG_MASTER_UPDATE_LOCK, - UPDATE_LOCK_STATUS, 1, - 1, 10); -} - void optc2_lock_doublebuffer_enable(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -440,14 +390,8 @@ void optc2_setup_manual_trigger(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); - REG_SET(OTG_MANUAL_FLOW_CONTROL, 0, - MANUAL_FLOW_CONTROL, 1); - - REG_SET(OTG_GLOBAL_CONTROL2, 0, - MANUAL_FLOW_CONTROL_SEL, optc->inst); - REG_SET_8(OTG_TRIGA_CNTL, 0, - OTG_TRIGA_SOURCE_SELECT, 22, + OTG_TRIGA_SOURCE_SELECT, 21, OTG_TRIGA_SOURCE_PIPE_SELECT, optc->inst, OTG_TRIGA_RISING_EDGE_DETECT_CNTL, 1, OTG_TRIGA_FALLING_EDGE_DETECT_CNTL, 0, @@ -492,10 +436,8 @@ static struct timing_generator_funcs dcn20_tg_funcs = { .triplebuffer_lock = optc2_triplebuffer_lock, .triplebuffer_unlock = optc2_triplebuffer_unlock, .disable_reset_trigger = optc1_disable_reset_trigger, - .lock = optc2_lock, + .lock = optc1_lock, .unlock = optc1_unlock, - .lock_global = optc2_lock_global, - .setup_global_lock = optc2_setup_global_lock, .lock_doublebuffer_enable = optc2_lock_doublebuffer_enable, .lock_doublebuffer_disable = optc2_lock_doublebuffer_disable, .enable_optc_clock = optc1_enable_optc_clock, @@ -511,9 +453,7 @@ static struct timing_generator_funcs dcn20_tg_funcs = { .setup_global_swap_lock = NULL, .get_crc = optc1_get_crc, .configure_crc = optc1_configure_crc, -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT .set_dsc_config = optc2_set_dsc_config, -#endif .set_dwb_source = optc2_set_dwb_source, .set_odm_bypass = optc2_set_odm_bypass, .set_odm_combine = optc2_set_odm_combine, @@ -522,7 +462,8 @@ static struct timing_generator_funcs dcn20_tg_funcs = { .set_gsl_source_select = optc2_set_gsl_source_select, .set_vtg_params = optc1_set_vtg_params, .program_manual_trigger = optc2_program_manual_trigger, - .setup_manual_trigger = optc2_setup_manual_trigger + .setup_manual_trigger = optc2_setup_manual_trigger, + .get_hw_timing = optc1_get_hw_timing, }; void dcn20_timing_generator_init(struct optc *optc1) @@ -537,6 +478,5 @@ void dcn20_timing_generator_init(struct optc *optc1) optc1->min_v_blank_interlace = 5; optc1->min_h_sync_width = 4;// Minimum HSYNC = 8 pixels asked By HW in the first place for no actual reason. Oculus Rift S will not light up with 8 as it's hsyncWidth is 6. Changing it to 4 to fix that issue. optc1->min_v_sync_width = 1; - optc1->comb_opp_id = 0xf; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h index ebf07c582da2..239cc40ae474 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h @@ -86,18 +86,16 @@ void optc2_set_gsl_source_select(struct timing_generator *optc, int group_idx, uint32_t gsl_ready_signal); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT void optc2_set_dsc_config(struct timing_generator *optc, enum optc_dsc_mode dsc_mode, uint32_t dsc_bytes_per_pixel, uint32_t dsc_slice_width); -#endif void optc2_set_odm_bypass(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing); -void optc2_set_odm_combine(struct timing_generator *optc, int combine_opp_id, - int mpcc_hactive, enum dc_pixel_encoding pixel_encoding); +void optc2_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_cnt, + struct dc_crtc_timing *timing); void optc2_get_optc_source(struct timing_generator *optc, uint32_t *num_of_src_opp, @@ -106,11 +104,9 @@ void optc2_get_optc_source(struct timing_generator *optc, void optc2_triplebuffer_lock(struct timing_generator *optc); void optc2_triplebuffer_unlock(struct timing_generator *optc); -void optc2_lock(struct timing_generator *optc); -void optc2_lock_global(struct timing_generator *optc); -void optc2_setup_global_lock(struct timing_generator *optc); void optc2_lock_doublebuffer_disable(struct timing_generator *optc); void optc2_lock_doublebuffer_enable(struct timing_generator *optc); +void optc2_setup_manual_trigger(struct timing_generator *optc); void optc2_program_manual_trigger(struct timing_generator *optc); - +bool optc2_is_two_pixels_per_containter(const struct dc_crtc_timing *timing); #endif /* __DC_OPTC_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index b949e202d6cb..85f90f3e24cb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -1,5 +1,6 @@ /* * Copyright 2016 Advanced Micro Devices, Inc. + * Copyright 2019 Raptor Engineering, LLC * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -28,6 +29,8 @@ #include "dm_services.h" #include "dc.h" +#include "dcn20_init.h" + #include "resource.h" #include "include/irq_service_interface.h" #include "dcn20/dcn20_resource.h" @@ -45,9 +48,7 @@ #include "dcn10/dcn10_resource.h" #include "dcn20_opp.h" -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #include "dcn20_dsc.h" -#endif #include "dcn20_link_encoder.h" #include "dcn20_stream_encoder.h" @@ -59,11 +60,14 @@ #include "dml/display_mode_vba.h" #include "dcn20_dccg.h" #include "dcn20_vmid.h" +#include "dc_link_ddc.h" #include "navi10_ip_offset.h" #include "dcn/dcn_2_0_0_offset.h" #include "dcn/dcn_2_0_0_sh_mask.h" +#include "dpcs/dpcs_2_0_0_offset.h" +#include "dpcs/dpcs_2_0_0_sh_mask.h" #include "nbio/nbio_2_3_offset.h" @@ -82,7 +86,6 @@ #include "amdgpu_socbb.h" -#define SOC_BOUNDING_BOX_VALID false #define DC_LOGGER_INIT(logger) struct _vcs_dpi_ip_params_st dcn2_0_ip = { @@ -93,11 +96,7 @@ struct _vcs_dpi_ip_params_st dcn2_0_ip = { .hostvm_max_page_table_levels = 4, .hostvm_cached_page_table_levels = 0, .pte_group_size_bytes = 2048, -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT .num_dsc = 6, -#else - .num_dsc = 0, -#endif .rob_buffer_size_kbytes = 168, .det_buffer_size_kbytes = 164, .dpte_buffer_size_in_pte_reqs_luma = 84, @@ -156,8 +155,187 @@ struct _vcs_dpi_ip_params_st dcn2_0_ip = { .xfc_fill_constant_bytes = 0, }; -struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { 0 }; - +struct _vcs_dpi_ip_params_st dcn2_0_nv14_ip = { + .odm_capable = 1, + .gpuvm_enable = 0, + .hostvm_enable = 0, + .gpuvm_max_page_table_levels = 4, + .hostvm_max_page_table_levels = 4, + .hostvm_cached_page_table_levels = 0, + .num_dsc = 5, + .rob_buffer_size_kbytes = 168, + .det_buffer_size_kbytes = 164, + .dpte_buffer_size_in_pte_reqs_luma = 84, + .dpte_buffer_size_in_pte_reqs_chroma = 42,//todo + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_enable = 1, + .max_page_table_levels = 4, + .pte_chunk_size_kbytes = 2, + .meta_chunk_size_kbytes = 2, + .writeback_chunk_size_kbytes = 2, + .line_buffer_size_bits = 789504, + .is_line_buffer_bpp_fixed = 0, + .line_buffer_fixed_bpp = 0, + .dcc_supported = true, + .max_line_buffer_lines = 12, + .writeback_luma_buffer_size_kbytes = 12, + .writeback_chroma_buffer_size_kbytes = 8, + .writeback_chroma_line_buffer_width_pixels = 4, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_taps = 12, + .writeback_max_vscl_taps = 12, + .writeback_line_buffer_luma_buffer_size = 0, + .writeback_line_buffer_chroma_buffer_size = 14643, + .cursor_buffer_size = 8, + .cursor_chunk_size = 2, + .max_num_otg = 5, + .max_num_dpp = 5, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 8, + .max_vscl_ratio = 8, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.10, + .min_vblank_lines = 32, // + .dppclk_delay_subtotal = 77, // + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_scl = 50, + .dppclk_delay_cnvc_formatter = 8, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 87, // + .dcfclk_cstate_latency = 10, // SRExitTime + .max_inter_dcn_tile_repeaters = 8, + .xfc_supported = true, + .xfc_fill_bw_overhead_percent = 10.0, + .xfc_fill_constant_bytes = 0, + .ptoi_supported = 0 +}; + +struct _vcs_dpi_soc_bounding_box_st dcn2_0_soc = { + /* Defaults that get patched on driver load from firmware. */ + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 560.0, + .fabricclk_mhz = 560.0, + .dispclk_mhz = 513.0, + .dppclk_mhz = 513.0, + .phyclk_mhz = 540.0, + .socclk_mhz = 560.0, + .dscclk_mhz = 171.0, + .dram_speed_mts = 8960.0, + }, + { + .state = 1, + .dcfclk_mhz = 694.0, + .fabricclk_mhz = 694.0, + .dispclk_mhz = 642.0, + .dppclk_mhz = 642.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 694.0, + .dscclk_mhz = 214.0, + .dram_speed_mts = 11104.0, + }, + { + .state = 2, + .dcfclk_mhz = 875.0, + .fabricclk_mhz = 875.0, + .dispclk_mhz = 734.0, + .dppclk_mhz = 734.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 875.0, + .dscclk_mhz = 245.0, + .dram_speed_mts = 14000.0, + }, + { + .state = 3, + .dcfclk_mhz = 1000.0, + .fabricclk_mhz = 1000.0, + .dispclk_mhz = 1100.0, + .dppclk_mhz = 1100.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1000.0, + .dscclk_mhz = 367.0, + .dram_speed_mts = 16000.0, + }, + { + .state = 4, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + /*Extra state, no dispclk ramping*/ + { + .state = 5, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 16000.0, + }, + }, + .num_states = 5, + .sr_exit_time_us = 8.6, + .sr_enter_plus_exit_time_us = 10.9, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 40.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .ideal_dram_bw_after_urgent_percent = 40.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 2, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 131, + .urgent_out_of_order_return_per_channel_bytes = 256, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 16, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 404.0, + .dummy_pstate_latency_us = 5.0, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3850, + .xfc_bus_transport_time_us = 20, + .xfc_xbuf_latency_tolerance_us = 4, + .use_urgent_burst_bw = 0 +}; + +struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv12_soc = { 0 }; #ifndef mmDP0_DP_DPHY_INTERNAL_CTRL #define mmDP0_DP_DPHY_INTERNAL_CTRL 0x210f @@ -314,7 +492,7 @@ static const struct dce_audio_shift audio_shift = { DCE120_AUD_COMMON_MASK_SH_LIST(__SHIFT) }; -static const struct dce_aduio_mask audio_mask = { +static const struct dce_audio_mask audio_mask = { DCE120_AUD_COMMON_MASK_SH_LIST(_MASK) }; @@ -373,6 +551,7 @@ static const struct dcn10_link_enc_hpd_registers link_enc_hpd_regs[] = { [id] = {\ LE_DCN10_REG_LIST(id), \ UNIPHY_DCN2_REG_LIST(phyid), \ + DPCS_DCN2_REG_LIST(id), \ SRI(DP_DPHY_INTERNAL_CTRL, DP, id) \ } @@ -386,11 +565,13 @@ static const struct dcn10_link_enc_registers link_enc_regs[] = { }; static const struct dcn10_link_enc_shift le_shift = { - LINK_ENCODER_MASK_SH_LIST_DCN20(__SHIFT) + LINK_ENCODER_MASK_SH_LIST_DCN20(__SHIFT),\ + DPCS_DCN2_MASK_SH_LIST(__SHIFT) }; static const struct dcn10_link_enc_mask le_mask = { - LINK_ENCODER_MASK_SH_LIST_DCN20(_MASK) + LINK_ENCODER_MASK_SH_LIST_DCN20(_MASK),\ + DPCS_DCN2_MASK_SH_LIST(_MASK) }; #define ipp_regs(id)\ @@ -457,6 +638,7 @@ static const struct dce110_aux_registers aux_engine_regs[] = { #define tf_regs(id)\ [id] = {\ TF_REG_LIST_DCN20(id),\ + TF_REG_LIST_DCN20_COMMON_APPEND(id),\ } static const struct dcn2_dpp_registers tf_regs[] = { @@ -469,11 +651,13 @@ static const struct dcn2_dpp_registers tf_regs[] = { }; static const struct dcn2_dpp_shift tf_shift = { - TF_REG_LIST_SH_MASK_DCN20(__SHIFT) + TF_REG_LIST_SH_MASK_DCN20(__SHIFT), + TF_DEBUG_REG_LIST_SH_DCN20 }; static const struct dcn2_dpp_mask tf_mask = { - TF_REG_LIST_SH_MASK_DCN20(_MASK) + TF_REG_LIST_SH_MASK_DCN20(_MASK), + TF_DEBUG_REG_LIST_MASK_DCN20 }; #define dwbc_regs_dcn2(id)\ @@ -523,14 +707,17 @@ static const struct dcn20_mpc_registers mpc_regs = { MPC_OUT_MUX_REG_LIST_DCN2_0(3), MPC_OUT_MUX_REG_LIST_DCN2_0(4), MPC_OUT_MUX_REG_LIST_DCN2_0(5), + MPC_DBG_REG_LIST_DCN2_0() }; static const struct dcn20_mpc_shift mpc_shift = { - MPC_COMMON_MASK_SH_LIST_DCN2_0(__SHIFT) + MPC_COMMON_MASK_SH_LIST_DCN2_0(__SHIFT), + MPC_DEBUG_REG_LIST_SH_DCN20 }; static const struct dcn20_mpc_mask mpc_mask = { - MPC_COMMON_MASK_SH_LIST_DCN2_0(_MASK) + MPC_COMMON_MASK_SH_LIST_DCN2_0(_MASK), + MPC_DEBUG_REG_LIST_MASK_DCN20 }; #define tg_regs(id)\ @@ -620,7 +807,42 @@ static const struct dcn20_vmid_mask vmid_masks = { DCN20_VMID_MASK_SH_LIST(_MASK) }; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT +static const struct dce110_aux_registers_shift aux_shift = { + DCN_AUX_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce110_aux_registers_mask aux_mask = { + DCN_AUX_MASK_SH_LIST(_MASK) +}; + +static int map_transmitter_id_to_phy_instance( + enum transmitter transmitter) +{ + switch (transmitter) { + case TRANSMITTER_UNIPHY_A: + return 0; + break; + case TRANSMITTER_UNIPHY_B: + return 1; + break; + case TRANSMITTER_UNIPHY_C: + return 2; + break; + case TRANSMITTER_UNIPHY_D: + return 3; + break; + case TRANSMITTER_UNIPHY_E: + return 4; + break; + case TRANSMITTER_UNIPHY_F: + return 5; + break; + default: + ASSERT(0); + return 0; + } +} + #define dsc_regsDCN20(id)\ [id] = {\ DSC_REG_LIST_DCN20(id)\ @@ -642,7 +864,6 @@ static const struct dcn20_dsc_shift dsc_shift = { static const struct dcn20_dsc_mask dsc_mask = { DSC_REG_LIST_SH_MASK_DCN20(_MASK) }; -#endif static const struct dccg_registers dccg_regs = { DCCG_REG_LIST_DCN2() @@ -666,9 +887,7 @@ static const struct resource_caps res_cap_nv10 = { .num_dwb = 1, .num_ddc = 6, .num_vmid = 16, -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT .num_dsc = 6, -#endif }; static const struct dc_plane_cap plane_cap = { @@ -695,6 +914,18 @@ static const struct dc_plane_cap plane_cap = { .fp16 = 1 } }; +static const struct resource_caps res_cap_nv14 = { + .num_timing_generator = 5, + .num_opp = 5, + .num_video_plane = 5, + .num_audio = 6, + .num_stream_encoder = 5, + .num_pll = 5, + .num_dwb = 1, + .num_ddc = 5, + .num_vmid = 16, + .num_dsc = 5, +}; static const struct dc_debug_options debug_defaults_drv = { .disable_dmcu = true, @@ -703,7 +934,7 @@ static const struct dc_debug_options debug_defaults_drv = { .clock_trace = true, .disable_pplib_clock_request = true, .pipe_split_policy = MPC_SPLIT_DYNAMIC, - .force_single_disp_pipe_split = true, + .force_single_disp_pipe_split = false, .disable_dcc = DCC_ENABLE, .vsr_support = true, .performance_trace = false, @@ -800,7 +1031,10 @@ struct dce_aux *dcn20_aux_engine_create( dce110_aux_engine_construct(aux_engine, ctx, inst, SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD, - &aux_engine_regs[inst]); + &aux_engine_regs[inst], + &aux_mask, + &aux_shift, + ctx->dc->caps.extended_aux_timeout_support); return &aux_engine->base; } @@ -920,14 +1154,18 @@ struct link_encoder *dcn20_link_encoder_create( { struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); + int link_regs_id; if (!enc20) return NULL; + link_regs_id = + map_transmitter_id_to_phy_instance(enc_init_data->transmitter); + dcn20_link_encoder_construct(enc20, enc_init_data, &link_enc_feature, - &link_enc_regs[enc_init_data->transmitter], + &link_enc_regs[link_regs_id], &link_enc_aux_regs[enc_init_data->channel - 1], &link_enc_hpd_regs[enc_init_data->hpd_source], &le_shift, @@ -955,6 +1193,7 @@ struct clock_source *dcn20_clock_source_create( return &clk_src->base; } + kfree(clk_src); BREAK_TO_DEBUGGER(); return NULL; } @@ -984,6 +1223,11 @@ struct stream_encoder *dcn20_stream_encoder_create( if (!enc1) return NULL; + if (ASICREV_IS_NAVI14_M(ctx->asic_id.hw_internal_rev)) { + if (eng_id >= ENGINE_ID_DIGD) + eng_id++; + } + dcn20_stream_encoder_construct(enc1, ctx, ctx->dc_bios, eng_id, &stream_enc_regs[eng_id], &se_shift, &se_mask); @@ -1031,13 +1275,14 @@ static const struct resource_create_funcs res_create_maximus_funcs = { .create_hwseq = dcn20_hwseq_create, }; +static void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu); + void dcn20_clock_source_destroy(struct clock_source **clk_src) { kfree(TO_DCE110_CLK_SRC(*clk_src)); *clk_src = NULL; } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct display_stream_compressor *dcn20_dsc_create( struct dc_context *ctx, uint32_t inst) @@ -1060,9 +1305,8 @@ void dcn20_dsc_destroy(struct display_stream_compressor **dsc) *dsc = NULL; } -#endif -static void destruct(struct dcn20_resource_pool *pool) +static void dcn20_resource_destruct(struct dcn20_resource_pool *pool) { unsigned int i; @@ -1073,12 +1317,10 @@ static void destruct(struct dcn20_resource_pool *pool) } } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT for (i = 0; i < pool->base.res_cap->num_dsc; i++) { if (pool->base.dscs[i] != NULL) dcn20_dsc_destroy(&pool->base.dscs[i]); } -#endif if (pool->base.mpc != NULL) { kfree(TO_DCN20_MPC(pool->base.mpc)); @@ -1171,6 +1413,8 @@ static void destruct(struct dcn20_resource_pool *pool) if (pool->base.pp_smu != NULL) dcn20_pp_smu_destroy(&pool->base.pp_smu); + if (pool->base.oem_device != NULL) + dal_ddc_service_destroy(&pool->base.oem_device); } struct hubp *dcn20_hubp_create( @@ -1197,7 +1441,11 @@ static void get_pixel_clock_parameters( struct pixel_clk_params *pixel_clk_params) { const struct dc_stream_state *stream = pipe_ctx->stream; - bool odm_combine = dc_res_get_odm_bottom_pipe(pipe_ctx) != NULL; + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; pixel_clk_params->requested_pix_clk_100hz = stream->timing.pix_clk_100hz; pixel_clk_params->encoder_object_id = stream->link->link_enc->id; @@ -1215,7 +1463,9 @@ static void get_pixel_clock_parameters( if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR422) pixel_clk_params->color_depth = COLOR_DEPTH_888; - if (optc1_is_two_pixels_per_containter(&stream->timing) || odm_combine) + if (opp_cnt == 4) + pixel_clk_params->requested_pix_clk_100hz /= 4; + else if (optc2_is_two_pixels_per_containter(&stream->timing) || opp_cnt == 2) pixel_clk_params->requested_pix_clk_100hz /= 2; if (stream->timing.timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING) @@ -1281,17 +1531,23 @@ enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state return status; } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT static void acquire_dsc(struct resource_context *res_ctx, const struct resource_pool *pool, - struct display_stream_compressor **dsc) + struct display_stream_compressor **dsc, + int pipe_idx) { int i; ASSERT(*dsc == NULL); *dsc = NULL; + if (pool->res_cap->num_dsc == pool->res_cap->num_opp) { + *dsc = pool->dscs[pipe_idx]; + res_ctx->is_dsc_acquired[pipe_idx] = true; + return; + } + /* Find first free DSC */ for (i = 0; i < pool->res_cap->num_dsc; i++) if (!res_ctx->is_dsc_acquired[i]) { @@ -1315,11 +1571,9 @@ static void release_dsc(struct resource_context *res_ctx, } } -#endif -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT -static enum dc_status add_dsc_to_stream_resource(struct dc *dc, +enum dc_status dcn20_add_dsc_to_stream_resource(struct dc *dc, struct dc_state *dc_ctx, struct dc_stream_state *dc_stream) { @@ -1334,11 +1588,13 @@ static enum dc_status add_dsc_to_stream_resource(struct dc *dc, if (pipe_ctx->stream != dc_stream) continue; - acquire_dsc(&dc_ctx->res_ctx, pool, &pipe_ctx->stream_res.dsc); + if (pipe_ctx->stream_res.dsc) + continue; + + acquire_dsc(&dc_ctx->res_ctx, pool, &pipe_ctx->stream_res.dsc, i); /* The number of DSCs can be less than the number of pipes */ if (!pipe_ctx->stream_res.dsc) { - dm_output_to_console("No DSCs available\n"); result = DC_NO_DSC_RESOURCE; } @@ -1359,24 +1615,17 @@ static enum dc_status remove_dsc_from_stream_resource(struct dc *dc, for (i = 0; i < MAX_PIPES; i++) { if (new_ctx->res_ctx.pipe_ctx[i].stream == dc_stream && !new_ctx->res_ctx.pipe_ctx[i].top_pipe) { pipe_ctx = &new_ctx->res_ctx.pipe_ctx[i]; - break; + + if (pipe_ctx->stream_res.dsc) + release_dsc(&new_ctx->res_ctx, dc->res_pool, &pipe_ctx->stream_res.dsc); } } if (!pipe_ctx) return DC_ERROR_UNEXPECTED; - - if (pipe_ctx->stream_res.dsc) { - struct pipe_ctx *odm_pipe = dc_res_get_odm_bottom_pipe(pipe_ctx); - - release_dsc(&new_ctx->res_ctx, dc->res_pool, &pipe_ctx->stream_res.dsc); - if (odm_pipe) - release_dsc(&new_ctx->res_ctx, dc->res_pool, &odm_pipe->stream_res.dsc); - } - - return DC_OK; + else + return DC_OK; } -#endif enum dc_status dcn20_add_stream_to_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream) @@ -1388,11 +1637,9 @@ enum dc_status dcn20_add_stream_to_ctx(struct dc *dc, struct dc_state *new_ctx, if (result == DC_OK) result = resource_map_phy_clock_resources(dc, new_ctx, dc_stream); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT /* Get a DSC if required and available */ if (result == DC_OK && dc_stream->timing.flags.DSC) - result = add_dsc_to_stream_resource(dc, new_ctx, dc_stream); -#endif + result = dcn20_add_dsc_to_stream_resource(dc, new_ctx, dc_stream); if (result == DC_OK) result = dcn20_build_mapped_resource(dc, new_ctx, dc_stream); @@ -1405,9 +1652,7 @@ enum dc_status dcn20_remove_stream_from_ctx(struct dc *dc, struct dc_state *new_ { enum dc_status result = DC_OK; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT result = remove_dsc_from_stream_resource(dc, new_ctx, dc_stream); -#endif return result; } @@ -1473,17 +1718,87 @@ static void swizzle_to_dml_params( } } -static bool dcn20_split_stream_for_combine( +bool dcn20_split_stream_for_odm( + struct resource_context *res_ctx, + const struct resource_pool *pool, + struct pipe_ctx *prev_odm_pipe, + struct pipe_ctx *next_odm_pipe) +{ + int pipe_idx = next_odm_pipe->pipe_idx; + + *next_odm_pipe = *prev_odm_pipe; + + next_odm_pipe->pipe_idx = pipe_idx; + next_odm_pipe->plane_res.mi = pool->mis[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.hubp = pool->hubps[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.ipp = pool->ipps[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.xfm = pool->transforms[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.dpp = pool->dpps[next_odm_pipe->pipe_idx]; + next_odm_pipe->plane_res.mpcc_inst = pool->dpps[next_odm_pipe->pipe_idx]->inst; + next_odm_pipe->stream_res.dsc = NULL; + if (prev_odm_pipe->next_odm_pipe && prev_odm_pipe->next_odm_pipe != next_odm_pipe) { + next_odm_pipe->next_odm_pipe = prev_odm_pipe->next_odm_pipe; + next_odm_pipe->next_odm_pipe->prev_odm_pipe = next_odm_pipe; + } + prev_odm_pipe->next_odm_pipe = next_odm_pipe; + next_odm_pipe->prev_odm_pipe = prev_odm_pipe; + ASSERT(next_odm_pipe->top_pipe == NULL); + + if (prev_odm_pipe->plane_state) { + struct scaler_data *sd = &prev_odm_pipe->plane_res.scl_data; + int new_width; + + /* HACTIVE halved for odm combine */ + sd->h_active /= 2; + /* Calculate new vp and recout for left pipe */ + /* Need at least 16 pixels width per side */ + if (sd->recout.x + 16 >= sd->h_active) + return false; + new_width = sd->h_active - sd->recout.x; + sd->viewport.width -= dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz, sd->recout.width - new_width)); + sd->viewport_c.width -= dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz_c, sd->recout.width - new_width)); + sd->recout.width = new_width; + + /* Calculate new vp and recout for right pipe */ + sd = &next_odm_pipe->plane_res.scl_data; + /* HACTIVE halved for odm combine */ + sd->h_active /= 2; + /* Need at least 16 pixels width per side */ + if (new_width <= 16) + return false; + new_width = sd->recout.width + sd->recout.x - sd->h_active; + sd->viewport.width -= dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz, sd->recout.width - new_width)); + sd->viewport_c.width -= dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz_c, sd->recout.width - new_width)); + sd->recout.width = new_width; + sd->viewport.x += dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz, sd->h_active - sd->recout.x)); + sd->viewport_c.x += dc_fixpt_floor(dc_fixpt_mul_int( + sd->ratios.horz_c, sd->h_active - sd->recout.x)); + sd->recout.x = 0; + } + next_odm_pipe->stream_res.opp = pool->opps[next_odm_pipe->pipe_idx]; + if (next_odm_pipe->stream->timing.flags.DSC == 1) { + acquire_dsc(res_ctx, pool, &next_odm_pipe->stream_res.dsc, next_odm_pipe->pipe_idx); + ASSERT(next_odm_pipe->stream_res.dsc); + if (next_odm_pipe->stream_res.dsc == NULL) + return false; + } + + return true; +} + +void dcn20_split_stream_for_mpc( struct resource_context *res_ctx, const struct resource_pool *pool, struct pipe_ctx *primary_pipe, - struct pipe_ctx *secondary_pipe, - bool is_odm_combine) + struct pipe_ctx *secondary_pipe) { int pipe_idx = secondary_pipe->pipe_idx; - struct scaler_data *sd = &primary_pipe->plane_res.scl_data; struct pipe_ctx *sec_bot_pipe = secondary_pipe->bottom_pipe; - int new_width; *secondary_pipe = *primary_pipe; secondary_pipe->bottom_pipe = sec_bot_pipe; @@ -1495,9 +1810,7 @@ static bool dcn20_split_stream_for_combine( secondary_pipe->plane_res.xfm = pool->transforms[secondary_pipe->pipe_idx]; secondary_pipe->plane_res.dpp = pool->dpps[secondary_pipe->pipe_idx]; secondary_pipe->plane_res.mpcc_inst = pool->dpps[secondary_pipe->pipe_idx]->inst; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT secondary_pipe->stream_res.dsc = NULL; -#endif if (primary_pipe->bottom_pipe && primary_pipe->bottom_pipe != secondary_pipe) { ASSERT(!secondary_pipe->bottom_pipe); secondary_pipe->bottom_pipe = primary_pipe->bottom_pipe; @@ -1506,57 +1819,9 @@ static bool dcn20_split_stream_for_combine( primary_pipe->bottom_pipe = secondary_pipe; secondary_pipe->top_pipe = primary_pipe; - if (is_odm_combine) { - if (primary_pipe->plane_state) { - /* HACTIVE halved for odm combine */ - sd->h_active /= 2; - /* Copy scl_data to secondary pipe */ - secondary_pipe->plane_res.scl_data = *sd; - - /* Calculate new vp and recout for left pipe */ - /* Need at least 16 pixels width per side */ - if (sd->recout.x + 16 >= sd->h_active) - return false; - new_width = sd->h_active - sd->recout.x; - sd->viewport.width -= dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz, sd->recout.width - new_width)); - sd->viewport_c.width -= dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz_c, sd->recout.width - new_width)); - sd->recout.width = new_width; - - /* Calculate new vp and recout for right pipe */ - sd = &secondary_pipe->plane_res.scl_data; - new_width = sd->recout.width + sd->recout.x - sd->h_active; - /* Need at least 16 pixels width per side */ - if (new_width <= 16) - return false; - sd->viewport.width -= dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz, sd->recout.width - new_width)); - sd->viewport_c.width -= dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz_c, sd->recout.width - new_width)); - sd->recout.width = new_width; - sd->viewport.x += dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz, sd->h_active - sd->recout.x)); - sd->viewport_c.x += dc_fixpt_floor(dc_fixpt_mul_int( - sd->ratios.horz_c, sd->h_active - sd->recout.x)); - sd->recout.x = 0; - } - secondary_pipe->stream_res.opp = pool->opps[secondary_pipe->pipe_idx]; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - if (secondary_pipe->stream->timing.flags.DSC == 1) { - acquire_dsc(res_ctx, pool, &secondary_pipe->stream_res.dsc); - ASSERT(secondary_pipe->stream_res.dsc); - if (secondary_pipe->stream_res.dsc == NULL) - return false; - } -#endif - } else { - ASSERT(primary_pipe->plane_state); - resource_build_scaling_params(primary_pipe); - resource_build_scaling_params(secondary_pipe); - } - - return true; + ASSERT(primary_pipe->plane_state); + resource_build_scaling_params(primary_pipe); + resource_build_scaling_params(secondary_pipe); } void dcn20_populate_dml_writeback_from_context( @@ -1596,11 +1861,28 @@ void dcn20_populate_dml_writeback_from_context( } +static int get_num_odm_heads(struct pipe_ctx *pipe) +{ + int odm_head_count = 0; + struct pipe_ctx *next_pipe = pipe->next_odm_pipe; + while (next_pipe) { + odm_head_count++; + next_pipe = next_pipe->next_odm_pipe; + } + pipe = pipe->prev_odm_pipe; + while (pipe) { + odm_head_count++; + pipe = pipe->prev_odm_pipe; + } + return odm_head_count ? odm_head_count + 1 : 0; +} + int dcn20_populate_dml_pipes_from_context( - struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes) + struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes) { int pipe_cnt, i; bool synchronized_vblank = true; + struct resource_context *res_ctx = &context->res_ctx; for (i = 0, pipe_cnt = -1; i < dc->res_pool->pipe_count; i++) { if (!res_ctx->pipe_ctx[i].stream) @@ -1610,7 +1892,7 @@ int dcn20_populate_dml_pipes_from_context( pipe_cnt = i; continue; } - if (!resource_are_streams_timing_synchronizable( + if (dc->debug.disable_timing_sync || !resource_are_streams_timing_synchronizable( res_ctx->pipe_ctx[pipe_cnt].stream, res_ctx->pipe_ctx[i].stream)) { synchronized_vblank = false; @@ -1620,25 +1902,30 @@ int dcn20_populate_dml_pipes_from_context( for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing = &res_ctx->pipe_ctx[i].stream->timing; + unsigned int v_total; + unsigned int front_porch; int output_bpc; if (!res_ctx->pipe_ctx[i].stream) continue; + + v_total = timing->v_total; + front_porch = timing->v_front_porch; /* todo: pipes[pipe_cnt].pipe.src.dynamic_metadata_enable = 0; pipes[pipe_cnt].pipe.src.dcc = 0; pipes[pipe_cnt].pipe.src.vm = 0;*/ -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + pipes[pipe_cnt].clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; + pipes[pipe_cnt].dout.dsc_enable = res_ctx->pipe_ctx[i].stream->timing.flags.DSC; /* todo: rotation?*/ pipes[pipe_cnt].dout.dsc_slices = res_ctx->pipe_ctx[i].stream->timing.dsc_cfg.num_slices_h; -#endif if (res_ctx->pipe_ctx[i].stream->use_dynamic_meta) { pipes[pipe_cnt].pipe.src.dynamic_metadata_enable = true; /* 1/2 vblank */ pipes[pipe_cnt].pipe.src.dynamic_metadata_lines_before_active = - (timing->v_total - timing->v_addressable + (v_total - timing->v_addressable - timing->v_border_top - timing->v_border_bottom) / 2; /* 36 bytes dp, 32 hdmi */ pipes[pipe_cnt].pipe.src.dynamic_metadata_xmit_bytes = @@ -1652,13 +1939,13 @@ int dcn20_populate_dml_pipes_from_context( - timing->h_addressable - timing->h_border_left - timing->h_border_right; - pipes[pipe_cnt].pipe.dest.vblank_start = timing->v_total - timing->v_front_porch; + pipes[pipe_cnt].pipe.dest.vblank_start = v_total - front_porch; pipes[pipe_cnt].pipe.dest.vblank_end = pipes[pipe_cnt].pipe.dest.vblank_start - timing->v_addressable - timing->v_border_top - timing->v_border_bottom; pipes[pipe_cnt].pipe.dest.htotal = timing->h_total; - pipes[pipe_cnt].pipe.dest.vtotal = timing->v_total; + pipes[pipe_cnt].pipe.dest.vtotal = v_total; pipes[pipe_cnt].pipe.dest.hactive = timing->h_addressable; pipes[pipe_cnt].pipe.dest.vactive = timing->v_addressable; pipes[pipe_cnt].pipe.dest.interlaced = timing->flags.INTERLACE; @@ -1669,6 +1956,24 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].dout.dp_lanes = 4; pipes[pipe_cnt].pipe.dest.vtotal_min = res_ctx->pipe_ctx[i].stream->adjust.v_total_min; pipes[pipe_cnt].pipe.dest.vtotal_max = res_ctx->pipe_ctx[i].stream->adjust.v_total_max; + switch (get_num_odm_heads(&res_ctx->pipe_ctx[i])) { + case 2: + pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_2to1; + break; + default: + pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_disabled; + } + pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; + if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state + == res_ctx->pipe_ctx[i].plane_state) + pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].top_pipe->pipe_idx; + else if (res_ctx->pipe_ctx[i].prev_odm_pipe) { + struct pipe_ctx *first_pipe = res_ctx->pipe_ctx[i].prev_odm_pipe; + + while (first_pipe->prev_odm_pipe) + first_pipe = first_pipe->prev_odm_pipe; + pipes[pipe_cnt].pipe.src.hsplit_grp = first_pipe->pipe_idx; + } switch (res_ctx->pipe_ctx[i].stream->signal) { case SIGNAL_TYPE_DISPLAY_PORT_MST: @@ -1708,20 +2013,17 @@ int dcn20_populate_dml_pipes_from_context( case COLOR_DEPTH_161616: output_bpc = 16; break; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 case COLOR_DEPTH_999: output_bpc = 9; break; case COLOR_DEPTH_111111: output_bpc = 11; break; -#endif default: output_bpc = 8; break; } - switch (res_ctx->pipe_ctx[i].stream->timing.pixel_encoding) { case PIXEL_ENCODING_RGB: case PIXEL_ENCODING_YCBCR444: @@ -1730,7 +2032,7 @@ int dcn20_populate_dml_pipes_from_context( break; case PIXEL_ENCODING_YCBCR420: pipes[pipe_cnt].dout.output_format = dm_420; - pipes[pipe_cnt].dout.output_bpp = (output_bpc * 3) / 2; + pipes[pipe_cnt].dout.output_bpp = (output_bpc * 3.0) / 2; break; case PIXEL_ENCODING_YCBCR422: if (true) /* todo */ @@ -1743,10 +2045,9 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].dout.output_format = dm_444; pipes[pipe_cnt].dout.output_bpp = output_bpc * 3; } - pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].pipe_idx; - if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state - == res_ctx->pipe_ctx[i].plane_state) - pipes[pipe_cnt].pipe.src.hsplit_grp = res_ctx->pipe_ctx[i].top_pipe->pipe_idx; + + if (res_ctx->pipe_ctx[i].stream->timing.flags.DSC) + pipes[pipe_cnt].dout.output_bpp = res_ctx->pipe_ctx[i].stream->timing.dsc_cfg.bits_per_pixel / 16.0; /* todo: default max for now, until there is logic reflecting this in dc*/ pipes[pipe_cnt].dout.output_bpc = 12; @@ -1770,6 +2071,10 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].pipe.src.viewport_height = timing->v_addressable; if (pipes[pipe_cnt].pipe.src.viewport_height > 1080) pipes[pipe_cnt].pipe.src.viewport_height = 1080; + pipes[pipe_cnt].pipe.src.surface_height_y = pipes[pipe_cnt].pipe.src.viewport_height; + pipes[pipe_cnt].pipe.src.surface_width_y = pipes[pipe_cnt].pipe.src.viewport_width; + pipes[pipe_cnt].pipe.src.surface_height_c = pipes[pipe_cnt].pipe.src.viewport_height; + pipes[pipe_cnt].pipe.src.surface_width_c = pipes[pipe_cnt].pipe.src.viewport_width; pipes[pipe_cnt].pipe.src.data_pitch = ((pipes[pipe_cnt].pipe.src.viewport_width + 63) / 64) * 64; /* linear sw only */ pipes[pipe_cnt].pipe.src.source_format = dm_444_32; pipes[pipe_cnt].pipe.dest.recout_width = pipes[pipe_cnt].pipe.src.viewport_width; /*vp_width/hratio*/ @@ -1784,8 +2089,8 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].pipe.scale_taps.vtaps = 1; pipes[pipe_cnt].pipe.src.is_hsplit = 0; pipes[pipe_cnt].pipe.dest.odm_combine = 0; - pipes[pipe_cnt].pipe.dest.vtotal_min = timing->v_total; - pipes[pipe_cnt].pipe.dest.vtotal_max = timing->v_total; + pipes[pipe_cnt].pipe.dest.vtotal_min = v_total; + pipes[pipe_cnt].pipe.dest.vtotal_max = v_total; } else { struct dc_plane_state *pln = res_ctx->pipe_ctx[i].plane_state; struct scaler_data *scl = &res_ctx->pipe_ctx[i].plane_res.scl_data; @@ -1795,14 +2100,6 @@ int dcn20_populate_dml_pipes_from_context( && res_ctx->pipe_ctx[i].bottom_pipe->plane_state == pln) || (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state == pln); - pipes[pipe_cnt].pipe.dest.odm_combine = (res_ctx->pipe_ctx[i].bottom_pipe - && res_ctx->pipe_ctx[i].bottom_pipe->plane_state == pln - && res_ctx->pipe_ctx[i].bottom_pipe->stream_res.opp - != res_ctx->pipe_ctx[i].stream_res.opp) - || (res_ctx->pipe_ctx[i].top_pipe - && res_ctx->pipe_ctx[i].top_pipe->plane_state == pln - && res_ctx->pipe_ctx[i].top_pipe->stream_res.opp - != res_ctx->pipe_ctx[i].stream_res.opp); pipes[pipe_cnt].pipe.src.source_scan = pln->rotation == ROTATION_ANGLE_90 || pln->rotation == ROTATION_ANGLE_270 ? dm_vert : dm_horz; pipes[pipe_cnt].pipe.src.viewport_y_y = scl->viewport.y; @@ -1811,14 +2108,18 @@ int dcn20_populate_dml_pipes_from_context( pipes[pipe_cnt].pipe.src.viewport_width_c = scl->viewport_c.width; pipes[pipe_cnt].pipe.src.viewport_height = scl->viewport.height; pipes[pipe_cnt].pipe.src.viewport_height_c = scl->viewport_c.height; + pipes[pipe_cnt].pipe.src.surface_width_y = pln->plane_size.surface_size.width; + pipes[pipe_cnt].pipe.src.surface_height_y = pln->plane_size.surface_size.height; + pipes[pipe_cnt].pipe.src.surface_width_c = pln->plane_size.chroma_size.width; + pipes[pipe_cnt].pipe.src.surface_height_c = pln->plane_size.chroma_size.height; if (pln->format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) { - pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.video.luma_pitch; - pipes[pipe_cnt].pipe.src.data_pitch_c = pln->plane_size.video.chroma_pitch; - pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.video.meta_pitch_l; - pipes[pipe_cnt].pipe.src.meta_pitch_c = pln->dcc.video.meta_pitch_c; + pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.surface_pitch; + pipes[pipe_cnt].pipe.src.data_pitch_c = pln->plane_size.chroma_pitch; + pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.meta_pitch; + pipes[pipe_cnt].pipe.src.meta_pitch_c = pln->dcc.meta_pitch_c; } else { - pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.grph.surface_pitch; - pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.grph.meta_pitch; + pipes[pipe_cnt].pipe.src.data_pitch = pln->plane_size.surface_pitch; + pipes[pipe_cnt].pipe.src.meta_pitch = pln->dcc.meta_pitch; } pipes[pipe_cnt].pipe.src.dcc = pln->dcc.enable; pipes[pipe_cnt].pipe.dest.recout_width = scl->recout.width; @@ -1976,8 +2277,7 @@ void dcn20_set_mcif_arb_params( } } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT -static bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx) +bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx) { int i; @@ -1986,53 +2286,144 @@ static bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx) struct pipe_ctx *pipe_ctx = &new_ctx->res_ctx.pipe_ctx[i]; struct dc_stream_state *stream = pipe_ctx->stream; struct dsc_config dsc_cfg; + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; /* Only need to validate top pipe */ - if (pipe_ctx->top_pipe || !stream || !stream->timing.flags.DSC) + if (pipe_ctx->top_pipe || pipe_ctx->prev_odm_pipe || !stream || !stream->timing.flags.DSC) continue; - dsc_cfg.pic_width = stream->timing.h_addressable + stream->timing.h_border_left - + stream->timing.h_border_right; + dsc_cfg.pic_width = (stream->timing.h_addressable + stream->timing.h_border_left + + stream->timing.h_border_right) / opp_cnt; dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; - if (dc_res_get_odm_bottom_pipe(pipe_ctx)) - dsc_cfg.pic_width /= 2; dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; dsc_cfg.color_depth = stream->timing.display_color_depth; dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; + dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; if (!pipe_ctx->stream_res.dsc->funcs->dsc_validate_stream(pipe_ctx->stream_res.dsc, &dsc_cfg)) return false; } return true; } -#endif -bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) +struct pipe_ctx *dcn20_find_secondary_pipe(struct dc *dc, + struct resource_context *res_ctx, + const struct resource_pool *pool, + const struct pipe_ctx *primary_pipe) { - bool out = false; + struct pipe_ctx *secondary_pipe = NULL; + + if (dc && primary_pipe) { + int j; + int preferred_pipe_idx = 0; + + /* first check the prev dc state: + * if this primary pipe has a bottom pipe in prev. state + * and if the bottom pipe is still available (which it should be), + * pick that pipe as secondary + * Same logic applies for ODM pipes. Since mpo is not allowed with odm + * check in else case. + */ + if (dc->current_state->res_ctx.pipe_ctx[primary_pipe->pipe_idx].bottom_pipe) { + preferred_pipe_idx = dc->current_state->res_ctx.pipe_ctx[primary_pipe->pipe_idx].bottom_pipe->pipe_idx; + if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) { + secondary_pipe = &res_ctx->pipe_ctx[preferred_pipe_idx]; + secondary_pipe->pipe_idx = preferred_pipe_idx; + } + } else if (dc->current_state->res_ctx.pipe_ctx[primary_pipe->pipe_idx].next_odm_pipe) { + preferred_pipe_idx = dc->current_state->res_ctx.pipe_ctx[primary_pipe->pipe_idx].next_odm_pipe->pipe_idx; + if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) { + secondary_pipe = &res_ctx->pipe_ctx[preferred_pipe_idx]; + secondary_pipe->pipe_idx = preferred_pipe_idx; + } + } - BW_VAL_TRACE_SETUP(); + /* + * if this primary pipe does not have a bottom pipe in prev. state + * start backward and find a pipe that did not used to be a bottom pipe in + * prev. dc state. This way we make sure we keep the same assignment as + * last state and will not have to reprogram every pipe + */ + if (secondary_pipe == NULL) { + for (j = dc->res_pool->pipe_count - 1; j >= 0; j--) { + if (dc->current_state->res_ctx.pipe_ctx[j].top_pipe == NULL + && dc->current_state->res_ctx.pipe_ctx[j].prev_odm_pipe == NULL) { + preferred_pipe_idx = j; + + if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) { + secondary_pipe = &res_ctx->pipe_ctx[preferred_pipe_idx]; + secondary_pipe->pipe_idx = preferred_pipe_idx; + break; + } + } + } + } + /* + * We should never hit this assert unless assignments are shuffled around + * if this happens we will prob. hit a vsync tdr + */ + ASSERT(secondary_pipe); + /* + * search backwards for the second pipe to keep pipe + * assignment more consistent + */ + if (secondary_pipe == NULL) { + for (j = dc->res_pool->pipe_count - 1; j >= 0; j--) { + preferred_pipe_idx = j; + + if (res_ctx->pipe_ctx[preferred_pipe_idx].stream == NULL) { + secondary_pipe = &res_ctx->pipe_ctx[preferred_pipe_idx]; + secondary_pipe->pipe_idx = preferred_pipe_idx; + break; + } + } + } + } - int pipe_cnt, i, pipe_idx, vlevel, vlevel_unsplit; - int pipe_split_from[MAX_PIPES]; - bool odm_capable = context->bw_ctx.dml.ip.odm_capable; - bool force_split = false; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - bool failed_non_odm_dsc = false; -#endif - int split_threshold = dc->res_pool->pipe_count / 2; - bool avoid_split = dc->debug.pipe_split_policy != MPC_SPLIT_DYNAMIC; - display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_KERNEL); - DC_LOGGER_INIT(dc->ctx->logger); + return secondary_pipe; +} - BW_VAL_TRACE_COUNT(); +void dcn20_merge_pipes_for_validate( + struct dc *dc, + struct dc_state *context) +{ + int i; - ASSERT(pipes); - if (!pipes) - return false; + /* merge previously split odm pipes since mode support needs to make the decision */ + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + struct pipe_ctx *odm_pipe = pipe->next_odm_pipe; + + if (pipe->prev_odm_pipe) + continue; + pipe->next_odm_pipe = NULL; + while (odm_pipe) { + struct pipe_ctx *next_odm_pipe = odm_pipe->next_odm_pipe; + + odm_pipe->plane_state = NULL; + odm_pipe->stream = NULL; + odm_pipe->top_pipe = NULL; + odm_pipe->bottom_pipe = NULL; + odm_pipe->prev_odm_pipe = NULL; + odm_pipe->next_odm_pipe = NULL; + if (odm_pipe->stream_res.dsc) + release_dsc(&context->res_ctx, dc->res_pool, &odm_pipe->stream_res.dsc); + /* Clear plane_res and stream_res */ + memset(&odm_pipe->plane_res, 0, sizeof(odm_pipe->plane_res)); + memset(&odm_pipe->stream_res, 0, sizeof(odm_pipe->stream_res)); + odm_pipe = next_odm_pipe; + } + if (pipe->plane_state) + resource_build_scaling_params(pipe); + } + + /* merge previously mpc split pipes since mode support needs to make the decision */ for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; struct pipe_ctx *hsplit_pipe = pipe->bottom_pipe; @@ -2040,7 +2431,6 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, if (!hsplit_pipe || hsplit_pipe->plane_state != pipe->plane_state) continue; - /* merge previously split pipe since mode support needs to make the decision */ pipe->bottom_pipe = hsplit_pipe->bottom_pipe; if (hsplit_pipe->bottom_pipe) hsplit_pipe->bottom_pipe->top_pipe = pipe; @@ -2048,60 +2438,26 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, hsplit_pipe->stream = NULL; hsplit_pipe->top_pipe = NULL; hsplit_pipe->bottom_pipe = NULL; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - if (hsplit_pipe->stream_res.dsc && hsplit_pipe->stream_res.dsc != pipe->stream_res.dsc) - release_dsc(&context->res_ctx, dc->res_pool, &hsplit_pipe->stream_res.dsc); -#endif + /* Clear plane_res and stream_res */ memset(&hsplit_pipe->plane_res, 0, sizeof(hsplit_pipe->plane_res)); memset(&hsplit_pipe->stream_res, 0, sizeof(hsplit_pipe->stream_res)); if (pipe->plane_state) resource_build_scaling_params(pipe); } +} - if (dc->res_pool->funcs->populate_dml_pipes) - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, - &context->res_ctx, pipes); - else - pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, - &context->res_ctx, pipes); - - if (!pipe_cnt) { - BW_VAL_TRACE_SKIP(pass); - out = true; - goto validate_out; - } - - context->bw_ctx.dml.ip.odm_capable = 0; - - vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); - - context->bw_ctx.dml.ip.odm_capable = odm_capable; - -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - /* 1 dsc per stream dsc validation */ - if (vlevel <= context->bw_ctx.dml.soc.num_states) - if (!dcn20_validate_dsc(dc, context)) { - failed_non_odm_dsc = true; - vlevel = context->bw_ctx.dml.soc.num_states + 1; - } -#endif - - if (vlevel > context->bw_ctx.dml.soc.num_states && odm_capable) - vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); - - if (vlevel > context->bw_ctx.dml.soc.num_states) - goto validate_fail; - - if ((context->stream_count > split_threshold && dc->current_state->stream_count <= split_threshold) - || (context->stream_count <= split_threshold && dc->current_state->stream_count > split_threshold)) - context->commit_hints.full_update_needed = true; - - /*initialize pipe_just_split_from to invalid idx*/ - for (i = 0; i < MAX_PIPES; i++) - pipe_split_from[i] = -1; +int dcn20_validate_apply_pipe_split_flags( + struct dc *dc, + struct dc_state *context, + int vlevel, + bool *split) +{ + int i, pipe_idx, vlevel_split; + bool force_split = false; + bool avoid_split = dc->debug.pipe_split_policy != MPC_SPLIT_DYNAMIC; - /* Single display only conditionals get set here */ + /* Single display loop, exits if there is more than one display */ for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; bool exit_loop = false; @@ -2128,49 +2484,113 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, if (exit_loop) break; } - - if (context->stream_count > split_threshold) + /* TODO: fix dc bugs and remove this split threshold thing */ + if (context->stream_count > dc->res_pool->pipe_count / 2) avoid_split = true; - vlevel_unsplit = vlevel; + /* Avoid split loop looks for lowest voltage level that allows most unsplit pipes possible */ + if (avoid_split) { + for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + for (vlevel_split = vlevel; vlevel <= context->bw_ctx.dml.soc.num_states; vlevel++) + if (context->bw_ctx.dml.vba.NoOfDPP[vlevel][0][pipe_idx] == 1) + break; + /* Impossible to not split this pipe */ + if (vlevel > context->bw_ctx.dml.soc.num_states) + vlevel = vlevel_split; + pipe_idx++; + } + context->bw_ctx.dml.vba.maxMpcComb = 0; + } + + /* Split loop sets which pipe should be split based on dml outputs and dc flags */ for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + if (!context->res_ctx.pipe_ctx[i].stream) continue; - for (; vlevel_unsplit <= context->bw_ctx.dml.soc.num_states; vlevel_unsplit++) - if (context->bw_ctx.dml.vba.NoOfDPP[vlevel_unsplit][0][pipe_idx] == 1) - break; + + if (force_split || context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] > 1) + split[i] = true; + if ((pipe->stream->view_format == + VIEW_3D_FORMAT_SIDE_BY_SIDE || + pipe->stream->view_format == + VIEW_3D_FORMAT_TOP_AND_BOTTOM) && + (pipe->stream->timing.timing_3d_format == + TIMING_3D_FORMAT_TOP_AND_BOTTOM || + pipe->stream->timing.timing_3d_format == + TIMING_3D_FORMAT_SIDE_BY_SIDE)) + split[i] = true; + if (dc->debug.force_odm_combine & (1 << pipe->stream_res.tg->inst)) { + split[i] = true; + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_idx] = dm_odm_combine_mode_2to1; + } + context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx] = + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_idx]; + /* Adjust dppclk when split is forced, do not bother with dispclk */ + if (split[i] && context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] == 1) + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] /= 2; pipe_idx++; } + return vlevel; +} + +bool dcn20_fast_validate_bw( + struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *pipe_cnt_out, + int *pipe_split_from, + int *vlevel_out) +{ + bool out = false; + bool split[MAX_PIPES] = { false }; + int pipe_cnt, i, pipe_idx, vlevel; + + ASSERT(pipes); + if (!pipes) + return false; + + dcn20_merge_pipes_for_validate(dc, context); + + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes); + + *pipe_cnt_out = pipe_cnt; + + if (!pipe_cnt) { + out = true; + goto validate_out; + } + + vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); + + if (vlevel > context->bw_ctx.dml.soc.num_states) + goto validate_fail; + + vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, vlevel, split); + + /*initialize pipe_just_split_from to invalid idx*/ + for (i = 0; i < MAX_PIPES; i++) + pipe_split_from[i] = -1; + for (i = 0, pipe_idx = -1; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; struct pipe_ctx *hsplit_pipe = pipe->bottom_pipe; - bool need_split = true; - bool need_split3d; if (!pipe->stream || pipe_split_from[i] >= 0) continue; pipe_idx++; - if (dc->debug.force_odm_combine & (1 << pipe->stream_res.tg->inst)) { - force_split = true; - context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx] = true; - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_idx] = true; - } - if (force_split && context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] == 1) - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] /= 2; - if (dc->config.forced_clocks == true) { - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] = - context->bw_ctx.dml.soc.clock_limits[0].dppclk_mhz; - } if (!pipe->top_pipe && !pipe->plane_state && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]) { - hsplit_pipe = find_idle_secondary_pipe(&context->res_ctx, dc->res_pool, pipe); + hsplit_pipe = dcn20_find_secondary_pipe(dc, &context->res_ctx, dc->res_pool, pipe); ASSERT(hsplit_pipe); - if (!dcn20_split_stream_for_combine( + if (!dcn20_split_stream_for_odm( &context->res_ctx, dc->res_pool, - pipe, hsplit_pipe, - true)) + pipe, hsplit_pipe)) goto validate_fail; pipe_split_from[hsplit_pipe->pipe_idx] = pipe_idx; dcn20_build_mapped_resource(dc, context, pipe->stream); @@ -2182,40 +2602,30 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, if (pipe->top_pipe && pipe->plane_state == pipe->top_pipe->plane_state) continue; - need_split3d = ((pipe->stream->view_format == - VIEW_3D_FORMAT_SIDE_BY_SIDE || - pipe->stream->view_format == - VIEW_3D_FORMAT_TOP_AND_BOTTOM) && - (pipe->stream->timing.timing_3d_format == - TIMING_3D_FORMAT_TOP_AND_BOTTOM || - pipe->stream->timing.timing_3d_format == - TIMING_3D_FORMAT_SIDE_BY_SIDE)); - - if (avoid_split && vlevel_unsplit <= context->bw_ctx.dml.soc.num_states && !force_split && !need_split3d) { - need_split = false; - vlevel = vlevel_unsplit; - context->bw_ctx.dml.vba.maxMpcComb = 0; - } else - need_split = context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] == 2; - /* We do not support mpo + odm at the moment */ if (hsplit_pipe && hsplit_pipe->plane_state != pipe->plane_state && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]) goto validate_fail; - if (need_split3d || need_split || force_split) { + if (split[i]) { if (!hsplit_pipe || hsplit_pipe->plane_state != pipe->plane_state) { /* pipe not split previously needs split */ - hsplit_pipe = find_idle_secondary_pipe(&context->res_ctx, dc->res_pool, pipe); - ASSERT(hsplit_pipe || force_split); - if (!hsplit_pipe) + hsplit_pipe = dcn20_find_secondary_pipe(dc, &context->res_ctx, dc->res_pool, pipe); + ASSERT(hsplit_pipe); + if (!hsplit_pipe) { + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] *= 2; continue; - - if (!dcn20_split_stream_for_combine( + } + if (context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]) { + if (!dcn20_split_stream_for_odm( + &context->res_ctx, dc->res_pool, + pipe, hsplit_pipe)) + goto validate_fail; + dcn20_build_mapped_resource(dc, context, pipe->stream); + } else + dcn20_split_stream_for_mpc( &context->res_ctx, dc->res_pool, - pipe, hsplit_pipe, - context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx])) - goto validate_fail; + pipe, hsplit_pipe); pipe_split_from[hsplit_pipe->pipe_idx] = pipe_idx; } } else if (hsplit_pipe && hsplit_pipe->plane_state == pipe->plane_state) { @@ -2223,22 +2633,33 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, ASSERT(0); } } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT /* Actual dsc count per stream dsc validation*/ - if (failed_non_odm_dsc && !dcn20_validate_dsc(dc, context)) { + if (!dcn20_validate_dsc(dc, context)) { context->bw_ctx.dml.vba.ValidationStatus[context->bw_ctx.dml.vba.soc.num_states] = DML_FAIL_DSC_VALIDATION_FAILURE; goto validate_fail; } -#endif - BW_VAL_TRACE_END_VOLTAGE_LEVEL(); + *vlevel_out = vlevel; - if (fast_validate) { - BW_VAL_TRACE_SKIP(fast); - out = true; - goto validate_out; - } + out = true; + goto validate_out; + +validate_fail: + out = false; + +validate_out: + return out; +} + +static void dcn20_calculate_wm( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *out_pipe_cnt, + int *pipe_split_from, + int vlevel) +{ + int pipe_cnt, i, pipe_idx; for (i = 0, pipe_idx = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { if (!context->res_ctx.pipe_ctx[i].stream) @@ -2252,7 +2673,7 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx]; if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_idx] == pipe_idx) pipes[pipe_cnt].pipe.dest.odm_combine = - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_idx]; + context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]; else pipes[pipe_cnt].pipe.dest.odm_combine = 0; pipe_idx++; @@ -2261,26 +2682,34 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_split_from[i]]; if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_split_from[i]] == pipe_split_from[i]) pipes[pipe_cnt].pipe.dest.odm_combine = - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_split_from[i]]; + context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_split_from[i]]; else pipes[pipe_cnt].pipe.dest.odm_combine = 0; } + if (dc->config.forced_clocks) { pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dispclk_mhz; pipes[pipe_cnt].clks_cfg.dppclk_mhz = context->bw_ctx.dml.soc.clock_limits[0].dppclk_mhz; } + if (dc->debug.min_disp_clk_khz > pipes[pipe_cnt].clks_cfg.dispclk_mhz * 1000) + pipes[pipe_cnt].clks_cfg.dispclk_mhz = dc->debug.min_disp_clk_khz / 1000.0; + if (dc->debug.min_dpp_clk_khz > pipes[pipe_cnt].clks_cfg.dppclk_mhz * 1000) + pipes[pipe_cnt].clks_cfg.dppclk_mhz = dc->debug.min_dpp_clk_khz / 1000.0; + pipe_cnt++; } if (pipe_cnt != pipe_idx) { if (dc->res_pool->funcs->populate_dml_pipes) pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, - &context->res_ctx, pipes); + context, pipes); else pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, - &context->res_ctx, pipes); + context, pipes); } + *out_pipe_cnt = pipe_cnt; + pipes[0].clks_cfg.voltage = vlevel; pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].dcfclk_mhz; pipes[0].clks_cfg.socclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].socclk_mhz; @@ -2296,6 +2725,9 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.b.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.b.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.b.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; if (vlevel < 2) { pipes[0].clks_cfg.voltage = 2; @@ -2307,6 +2739,8 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.c.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.c.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.c.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; if (vlevel < 3) { pipes[0].clks_cfg.voltage = 3; @@ -2318,6 +2752,8 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.d.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.d.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; pipes[0].clks_cfg.voltage = vlevel; pipes[0].clks_cfg.dcfclk_mhz = context->bw_ctx.dml.soc.clock_limits[vlevel].dcfclk_mhz; @@ -2327,6 +2763,19 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.a.pte_meta_urgent_ns = get_wm_memory_trip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + context->bw_ctx.bw.dcn.watermarks.a.frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; +} + +void dcn20_calculate_dlg_params( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int vlevel) +{ + int i, j, pipe_idx, pipe_idx_unsplit; + bool visited[MAX_PIPES] = { 0 }; + /* Writeback MCIF_WB arbitration parameters */ dc->res_pool->funcs->set_mcif_arb_params(dc, context, pipes, pipe_cnt); @@ -2341,26 +2790,63 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, != dm_dram_clock_change_unsupported; context->bw_ctx.bw.dcn.clk.dppclk_khz = 0; - BW_VAL_TRACE_END_WATERMARKS(); + /* + * An artifact of dml pipe split/odm is that pipes get merged back together for + * calculation. Therefore we need to only extract for first pipe in ascending index order + * and copy into the other split half. + */ + for (i = 0, pipe_idx = 0, pipe_idx_unsplit = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + if (!visited[pipe_idx]) { + display_pipe_source_params_st *src = &pipes[pipe_idx].pipe.src; + display_pipe_dest_params_st *dst = &pipes[pipe_idx].pipe.dest; + + dst->vstartup_start = context->bw_ctx.dml.vba.VStartup[pipe_idx_unsplit]; + dst->vupdate_offset = context->bw_ctx.dml.vba.VUpdateOffsetPix[pipe_idx_unsplit]; + dst->vupdate_width = context->bw_ctx.dml.vba.VUpdateWidthPix[pipe_idx_unsplit]; + dst->vready_offset = context->bw_ctx.dml.vba.VReadyOffsetPix[pipe_idx_unsplit]; + /* + * j iterates inside pipes array, unlike i which iterates inside + * pipe_ctx array + */ + if (src->is_hsplit) + for (j = pipe_idx + 1; j < pipe_cnt; j++) { + display_pipe_source_params_st *src_j = &pipes[j].pipe.src; + display_pipe_dest_params_st *dst_j = &pipes[j].pipe.dest; + + if (src_j->is_hsplit && !visited[j] + && src->hsplit_grp == src_j->hsplit_grp) { + dst_j->vstartup_start = context->bw_ctx.dml.vba.VStartup[pipe_idx_unsplit]; + dst_j->vupdate_offset = context->bw_ctx.dml.vba.VUpdateOffsetPix[pipe_idx_unsplit]; + dst_j->vupdate_width = context->bw_ctx.dml.vba.VUpdateWidthPix[pipe_idx_unsplit]; + dst_j->vready_offset = context->bw_ctx.dml.vba.VReadyOffsetPix[pipe_idx_unsplit]; + visited[j] = true; + } + } + visited[pipe_idx] = true; + pipe_idx_unsplit++; + } + pipe_idx++; + } for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { if (!context->res_ctx.pipe_ctx[i].stream) continue; - pipes[pipe_idx].pipe.dest.vstartup_start = context->bw_ctx.dml.vba.VStartup[pipe_idx]; - pipes[pipe_idx].pipe.dest.vupdate_offset = context->bw_ctx.dml.vba.VUpdateOffsetPix[pipe_idx]; - pipes[pipe_idx].pipe.dest.vupdate_width = context->bw_ctx.dml.vba.VUpdateWidthPix[pipe_idx]; - pipes[pipe_idx].pipe.dest.vready_offset = context->bw_ctx.dml.vba.VReadyOffsetPix[pipe_idx]; if (context->bw_ctx.bw.dcn.clk.dppclk_khz < pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000) context->bw_ctx.bw.dcn.clk.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - context->res_ctx.pipe_ctx[i].stream_res.dscclk_khz = - context->bw_ctx.dml.vba.DSCCLK_calculated[pipe_idx] * 1000; -#endif + ASSERT(visited[pipe_idx]); context->res_ctx.pipe_ctx[i].pipe_dlg_param = pipes[pipe_idx].pipe.dest; pipe_idx++; } + /*save a original dppclock copy*/ + context->bw_ctx.bw.dcn.clk.bw_dppclk_khz = context->bw_ctx.bw.dcn.clk.dppclk_khz; + context->bw_ctx.bw.dcn.clk.bw_dispclk_khz = context->bw_ctx.bw.dcn.clk.dispclk_khz; + context->bw_ctx.bw.dcn.clk.max_supported_dppclk_khz = context->bw_ctx.dml.soc.clock_limits[vlevel].dppclk_mhz * 1000; + context->bw_ctx.bw.dcn.clk.max_supported_dispclk_khz = context->bw_ctx.dml.soc.clock_limits[vlevel].dispclk_mhz * 1000; for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { bool cstate_en = context->bw_ctx.dml.vba.PrefetchMode[vlevel][context->bw_ctx.dml.vba.maxMpcComb] != 2; @@ -2383,8 +2869,43 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, pipes[pipe_idx].pipe); pipe_idx++; } +} + +static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *context, + bool fast_validate) +{ + bool out = false; + + BW_VAL_TRACE_SETUP(); + + int vlevel = 0; + int pipe_split_from[MAX_PIPES]; + int pipe_cnt = 0; + display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_KERNEL); + DC_LOGGER_INIT(dc->ctx->logger); + + BW_VAL_TRACE_COUNT(); + + out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel); + + if (pipe_cnt == 0) + goto validate_out; + + if (!out) + goto validate_fail; + + BW_VAL_TRACE_END_VOLTAGE_LEVEL(); + + if (fast_validate) { + BW_VAL_TRACE_SKIP(fast); + goto validate_out; + } + + dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel); + dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + + BW_VAL_TRACE_END_WATERMARKS(); - out = true; goto validate_out; validate_fail: @@ -2402,6 +2923,58 @@ validate_out: return out; } + +bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, + bool fast_validate) +{ + bool voltage_supported = false; + bool full_pstate_supported = false; + bool dummy_pstate_supported = false; + double p_state_latency_us; + + DC_FP_START(); + p_state_latency_us = context->bw_ctx.dml.soc.dram_clock_change_latency_us; + context->bw_ctx.dml.soc.disable_dram_clock_change_vactive_support = + dc->debug.disable_dram_clock_change_vactive_support; + + if (fast_validate) { + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, true); + + DC_FP_END(); + return voltage_supported; + } + + // Best case, we support full UCLK switch latency + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false); + full_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; + + if (context->bw_ctx.dml.soc.dummy_pstate_latency_us == 0 || + (voltage_supported && full_pstate_supported)) { + context->bw_ctx.bw.dcn.clk.p_state_change_support = full_pstate_supported; + goto restore_dml_state; + } + + // Fallback: Try to only support G6 temperature read latency + context->bw_ctx.dml.soc.dram_clock_change_latency_us = context->bw_ctx.dml.soc.dummy_pstate_latency_us; + + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false); + dummy_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; + + if (voltage_supported && dummy_pstate_supported) { + context->bw_ctx.bw.dcn.clk.p_state_change_support = false; + goto restore_dml_state; + } + + // ERROR: fallback is supposed to always work. + ASSERT(false); + +restore_dml_state: + context->bw_ctx.dml.soc.dram_clock_change_latency_us = p_state_latency_us; + + DC_FP_END(); + return voltage_supported; +} + struct pipe_ctx *dcn20_acquire_idle_pipe_for_layer( struct dc_state *state, const struct resource_pool *pool, @@ -2443,7 +3016,7 @@ static void dcn20_destroy_resource_pool(struct resource_pool **pool) { struct dcn20_resource_pool *dcn20_pool = TO_DCN20_RES_POOL(*pool); - destruct(dcn20_pool); + dcn20_resource_destruct(dcn20_pool); kfree(dcn20_pool); *pool = NULL; } @@ -2482,6 +3055,7 @@ static struct resource_funcs dcn20_res_pool_funcs = { .populate_dml_writeback_from_context = dcn20_populate_dml_writeback_from_context, .get_default_swizzle_mode = dcn20_get_default_swizzle_mode, .set_mcif_arb_params = dcn20_set_mcif_arb_params, + .populate_dml_pipes = dcn20_populate_dml_pipes_from_context, .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link }; @@ -2490,8 +3064,6 @@ bool dcn20_dwbc_create(struct dc_context *ctx, struct resource_pool *pool) int i; uint32_t pipe_count = pool->res_cap->num_dwb; - ASSERT(pipe_count > 0); - for (i = 0; i < pipe_count; i++) { struct dcn20_dwbc *dwbc20 = kzalloc(sizeof(struct dcn20_dwbc), GFP_KERNEL); @@ -2537,7 +3109,7 @@ bool dcn20_mmhubbub_create(struct dc_context *ctx, struct resource_pool *pool) return true; } -struct pp_smu_funcs *dcn20_pp_smu_create(struct dc_context *ctx) +static struct pp_smu_funcs *dcn20_pp_smu_create(struct dc_context *ctx) { struct pp_smu_funcs *pp_smu = kzalloc(sizeof(*pp_smu), GFP_KERNEL); @@ -2552,7 +3124,7 @@ struct pp_smu_funcs *dcn20_pp_smu_create(struct dc_context *ctx) return pp_smu; } -void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu) +static void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu) { if (pp_smu && *pp_smu) { kfree(*pp_smu); @@ -2560,7 +3132,7 @@ void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu) } } -static void cap_soc_clocks( +void dcn20_cap_soc_clocks( struct _vcs_dpi_soc_bounding_box_st *bb, struct pp_smu_nv_clock_table max_clocks) { @@ -2576,9 +3148,6 @@ static void cap_soc_clocks( && max_clocks.uClockInKhz != 0) bb->clock_limits[i].dram_speed_mts = (max_clocks.uClockInKhz / 1000) * 16; - // HACK: Force every uclk to max for now to "disable" uclk switching. - bb->clock_limits[i].dram_speed_mts = (max_clocks.uClockInKhz / 1000) * 16; - if ((bb->clock_limits[i].fabricclk_mhz > (max_clocks.fabricClockInKhz / 1000)) && max_clocks.fabricClockInKhz != 0) bb->clock_limits[i].fabricclk_mhz = (max_clocks.fabricClockInKhz / 1000); @@ -2630,10 +3199,10 @@ static void cap_soc_clocks( } } -static void update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb, +void dcn20_update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb, struct pp_smu_nv_clock_table *max_clocks, unsigned int *uclk_states, unsigned int num_states) { - struct _vcs_dpi_voltage_scaling_st calculated_states[MAX_CLOCK_LIMIT_STATES] = {0}; + struct _vcs_dpi_voltage_scaling_st calculated_states[MAX_CLOCK_LIMIT_STATES]; int i; int num_calculated_states = 0; int min_dcfclk = 0; @@ -2641,12 +3210,18 @@ static void update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_ if (num_states == 0) return; + memset(calculated_states, 0, sizeof(calculated_states)); + if (dc->bb_overrides.min_dcfclk_mhz > 0) min_dcfclk = dc->bb_overrides.min_dcfclk_mhz; - else - // Accounting for SOC/DCF relationship, we can go as high as - // 506Mhz in Vmin. We need to code 507 since SMU will round down to 506. - min_dcfclk = 507; + else { + if (ASICREV_IS_NAVI12_P(dc->ctx->asic_id.hw_internal_rev)) + min_dcfclk = 310; + else + // Accounting for SOC/DCF relationship, we can go as high as + // 506Mhz in Vmin. + min_dcfclk = 506; + } for (i = 0; i < num_states; i++) { int min_fclk_required_by_uclk; @@ -2674,6 +3249,10 @@ static void update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_ num_calculated_states++; } + calculated_states[num_calculated_states - 1].socclk_mhz = max_clocks->socClockInKhz / 1000; + calculated_states[num_calculated_states - 1].fabricclk_mhz = max_clocks->socClockInKhz / 1000; + calculated_states[num_calculated_states - 1].dcfclk_mhz = max_clocks->dcfClockInKhz / 1000; + memcpy(bb->clock_limits, calculated_states, sizeof(bb->clock_limits)); bb->num_states = num_calculated_states; @@ -2682,9 +3261,8 @@ static void update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_ bb->clock_limits[num_calculated_states].state = bb->num_states; } -static void patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb) +void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb) { - kernel_fpu_begin(); if ((int)(bb->sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns && dc->bb_overrides.sr_exit_time_ns) { bb->sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0; @@ -2708,7 +3286,31 @@ static void patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_s bb->dram_clock_change_latency_us = dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; } - kernel_fpu_end(); +} + +static struct _vcs_dpi_soc_bounding_box_st *get_asic_rev_soc_bb( + uint32_t hw_internal_rev) +{ + if (ASICREV_IS_NAVI12_P(hw_internal_rev)) + return &dcn2_0_nv12_soc; + + return &dcn2_0_soc; +} + +static struct _vcs_dpi_ip_params_st *get_asic_rev_ip_params( + uint32_t hw_internal_rev) +{ + /* NV14 */ + if (ASICREV_IS_NAVI14_M(hw_internal_rev)) + return &dcn2_0_nv14_ip; + + /* NV12 and NV10 */ + return &dcn2_0_ip; +} + +static enum dml_project get_dml_project_version(uint32_t hw_internal_rev) +{ + return DML_PROJECT_NAVI10v2; } #define fixed16_to_double(x) (((double) x) / ((double) (1 << 16))) @@ -2718,113 +3320,119 @@ static bool init_soc_bounding_box(struct dc *dc, struct dcn20_resource_pool *pool) { const struct gpu_info_soc_bounding_box_v1_0 *bb = dc->soc_bounding_box; + struct _vcs_dpi_soc_bounding_box_st *loaded_bb = + get_asic_rev_soc_bb(dc->ctx->asic_id.hw_internal_rev); + struct _vcs_dpi_ip_params_st *loaded_ip = + get_asic_rev_ip_params(dc->ctx->asic_id.hw_internal_rev); + DC_LOGGER_INIT(dc->ctx->logger); - if (!bb && !SOC_BOUNDING_BOX_VALID) { + /* TODO: upstream NV12 bounding box when its launched */ + if (!bb && ASICREV_IS_NAVI12_P(dc->ctx->asic_id.hw_internal_rev)) { DC_LOG_ERROR("%s: not valid soc bounding box/n", __func__); return false; } - if (bb && !SOC_BOUNDING_BOX_VALID) { + if (bb && ASICREV_IS_NAVI12_P(dc->ctx->asic_id.hw_internal_rev)) { int i; - dcn2_0_soc.sr_exit_time_us = + dcn2_0_nv12_soc.sr_exit_time_us = fixed16_to_double_to_cpu(bb->sr_exit_time_us); - dcn2_0_soc.sr_enter_plus_exit_time_us = + dcn2_0_nv12_soc.sr_enter_plus_exit_time_us = fixed16_to_double_to_cpu(bb->sr_enter_plus_exit_time_us); - dcn2_0_soc.urgent_latency_us = + dcn2_0_nv12_soc.urgent_latency_us = fixed16_to_double_to_cpu(bb->urgent_latency_us); - dcn2_0_soc.urgent_latency_pixel_data_only_us = + dcn2_0_nv12_soc.urgent_latency_pixel_data_only_us = fixed16_to_double_to_cpu(bb->urgent_latency_pixel_data_only_us); - dcn2_0_soc.urgent_latency_pixel_mixed_with_vm_data_us = + dcn2_0_nv12_soc.urgent_latency_pixel_mixed_with_vm_data_us = fixed16_to_double_to_cpu(bb->urgent_latency_pixel_mixed_with_vm_data_us); - dcn2_0_soc.urgent_latency_vm_data_only_us = + dcn2_0_nv12_soc.urgent_latency_vm_data_only_us = fixed16_to_double_to_cpu(bb->urgent_latency_vm_data_only_us); - dcn2_0_soc.urgent_out_of_order_return_per_channel_pixel_only_bytes = + dcn2_0_nv12_soc.urgent_out_of_order_return_per_channel_pixel_only_bytes = le32_to_cpu(bb->urgent_out_of_order_return_per_channel_pixel_only_bytes); - dcn2_0_soc.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = + dcn2_0_nv12_soc.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = le32_to_cpu(bb->urgent_out_of_order_return_per_channel_pixel_and_vm_bytes); - dcn2_0_soc.urgent_out_of_order_return_per_channel_vm_only_bytes = + dcn2_0_nv12_soc.urgent_out_of_order_return_per_channel_vm_only_bytes = le32_to_cpu(bb->urgent_out_of_order_return_per_channel_vm_only_bytes); - dcn2_0_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = + dcn2_0_nv12_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = fixed16_to_double_to_cpu(bb->pct_ideal_dram_sdp_bw_after_urgent_pixel_only); - dcn2_0_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = + dcn2_0_nv12_soc.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = fixed16_to_double_to_cpu(bb->pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm); - dcn2_0_soc.pct_ideal_dram_sdp_bw_after_urgent_vm_only = + dcn2_0_nv12_soc.pct_ideal_dram_sdp_bw_after_urgent_vm_only = fixed16_to_double_to_cpu(bb->pct_ideal_dram_sdp_bw_after_urgent_vm_only); - dcn2_0_soc.max_avg_sdp_bw_use_normal_percent = + dcn2_0_nv12_soc.max_avg_sdp_bw_use_normal_percent = fixed16_to_double_to_cpu(bb->max_avg_sdp_bw_use_normal_percent); - dcn2_0_soc.max_avg_dram_bw_use_normal_percent = + dcn2_0_nv12_soc.max_avg_dram_bw_use_normal_percent = fixed16_to_double_to_cpu(bb->max_avg_dram_bw_use_normal_percent); - dcn2_0_soc.writeback_latency_us = + dcn2_0_nv12_soc.writeback_latency_us = fixed16_to_double_to_cpu(bb->writeback_latency_us); - dcn2_0_soc.ideal_dram_bw_after_urgent_percent = + dcn2_0_nv12_soc.ideal_dram_bw_after_urgent_percent = fixed16_to_double_to_cpu(bb->ideal_dram_bw_after_urgent_percent); - dcn2_0_soc.max_request_size_bytes = + dcn2_0_nv12_soc.max_request_size_bytes = le32_to_cpu(bb->max_request_size_bytes); - dcn2_0_soc.dram_channel_width_bytes = + dcn2_0_nv12_soc.dram_channel_width_bytes = le32_to_cpu(bb->dram_channel_width_bytes); - dcn2_0_soc.fabric_datapath_to_dcn_data_return_bytes = + dcn2_0_nv12_soc.fabric_datapath_to_dcn_data_return_bytes = le32_to_cpu(bb->fabric_datapath_to_dcn_data_return_bytes); - dcn2_0_soc.dcn_downspread_percent = + dcn2_0_nv12_soc.dcn_downspread_percent = fixed16_to_double_to_cpu(bb->dcn_downspread_percent); - dcn2_0_soc.downspread_percent = + dcn2_0_nv12_soc.downspread_percent = fixed16_to_double_to_cpu(bb->downspread_percent); - dcn2_0_soc.dram_page_open_time_ns = + dcn2_0_nv12_soc.dram_page_open_time_ns = fixed16_to_double_to_cpu(bb->dram_page_open_time_ns); - dcn2_0_soc.dram_rw_turnaround_time_ns = + dcn2_0_nv12_soc.dram_rw_turnaround_time_ns = fixed16_to_double_to_cpu(bb->dram_rw_turnaround_time_ns); - dcn2_0_soc.dram_return_buffer_per_channel_bytes = + dcn2_0_nv12_soc.dram_return_buffer_per_channel_bytes = le32_to_cpu(bb->dram_return_buffer_per_channel_bytes); - dcn2_0_soc.round_trip_ping_latency_dcfclk_cycles = + dcn2_0_nv12_soc.round_trip_ping_latency_dcfclk_cycles = le32_to_cpu(bb->round_trip_ping_latency_dcfclk_cycles); - dcn2_0_soc.urgent_out_of_order_return_per_channel_bytes = + dcn2_0_nv12_soc.urgent_out_of_order_return_per_channel_bytes = le32_to_cpu(bb->urgent_out_of_order_return_per_channel_bytes); - dcn2_0_soc.channel_interleave_bytes = + dcn2_0_nv12_soc.channel_interleave_bytes = le32_to_cpu(bb->channel_interleave_bytes); - dcn2_0_soc.num_banks = + dcn2_0_nv12_soc.num_banks = le32_to_cpu(bb->num_banks); - dcn2_0_soc.num_chans = + dcn2_0_nv12_soc.num_chans = le32_to_cpu(bb->num_chans); - dcn2_0_soc.vmm_page_size_bytes = + dcn2_0_nv12_soc.vmm_page_size_bytes = le32_to_cpu(bb->vmm_page_size_bytes); - dcn2_0_soc.dram_clock_change_latency_us = + dcn2_0_nv12_soc.dram_clock_change_latency_us = fixed16_to_double_to_cpu(bb->dram_clock_change_latency_us); // HACK!! Lower uclock latency switch time so we don't switch - dcn2_0_soc.dram_clock_change_latency_us = 10; - dcn2_0_soc.writeback_dram_clock_change_latency_us = + dcn2_0_nv12_soc.dram_clock_change_latency_us = 10; + dcn2_0_nv12_soc.writeback_dram_clock_change_latency_us = fixed16_to_double_to_cpu(bb->writeback_dram_clock_change_latency_us); - dcn2_0_soc.return_bus_width_bytes = + dcn2_0_nv12_soc.return_bus_width_bytes = le32_to_cpu(bb->return_bus_width_bytes); - dcn2_0_soc.dispclk_dppclk_vco_speed_mhz = + dcn2_0_nv12_soc.dispclk_dppclk_vco_speed_mhz = le32_to_cpu(bb->dispclk_dppclk_vco_speed_mhz); - dcn2_0_soc.xfc_bus_transport_time_us = + dcn2_0_nv12_soc.xfc_bus_transport_time_us = le32_to_cpu(bb->xfc_bus_transport_time_us); - dcn2_0_soc.xfc_xbuf_latency_tolerance_us = + dcn2_0_nv12_soc.xfc_xbuf_latency_tolerance_us = le32_to_cpu(bb->xfc_xbuf_latency_tolerance_us); - dcn2_0_soc.use_urgent_burst_bw = + dcn2_0_nv12_soc.use_urgent_burst_bw = le32_to_cpu(bb->use_urgent_burst_bw); - dcn2_0_soc.num_states = + dcn2_0_nv12_soc.num_states = le32_to_cpu(bb->num_states); - for (i = 0; i < dcn2_0_soc.num_states; i++) { - dcn2_0_soc.clock_limits[i].state = + for (i = 0; i < dcn2_0_nv12_soc.num_states; i++) { + dcn2_0_nv12_soc.clock_limits[i].state = le32_to_cpu(bb->clock_limits[i].state); - dcn2_0_soc.clock_limits[i].dcfclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].dcfclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].dcfclk_mhz); - dcn2_0_soc.clock_limits[i].fabricclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].fabricclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].fabricclk_mhz); - dcn2_0_soc.clock_limits[i].dispclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].dispclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].dispclk_mhz); - dcn2_0_soc.clock_limits[i].dppclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].dppclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].dppclk_mhz); - dcn2_0_soc.clock_limits[i].phyclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].phyclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].phyclk_mhz); - dcn2_0_soc.clock_limits[i].socclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].socclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].socclk_mhz); - dcn2_0_soc.clock_limits[i].dscclk_mhz = + dcn2_0_nv12_soc.clock_limits[i].dscclk_mhz = fixed16_to_double_to_cpu(bb->clock_limits[i].dscclk_mhz); - dcn2_0_soc.clock_limits[i].dram_speed_mts = + dcn2_0_nv12_soc.clock_limits[i].dram_speed_mts = fixed16_to_double_to_cpu(bb->clock_limits[i].dram_speed_mts); } } @@ -2833,7 +3441,6 @@ static bool init_soc_bounding_box(struct dc *dc, struct pp_smu_nv_clock_table max_clocks = {0}; unsigned int uclk_states[8] = {0}; unsigned int num_states = 0; - int i; enum pp_smu_status status; bool clock_limits_available = false; bool uclk_states_available = false; @@ -2855,24 +3462,20 @@ static bool init_soc_bounding_box(struct dc *dc, clock_limits_available = (status == PP_SMU_RESULT_OK); } - // HACK: Use the max uclk_states value for all elements. - for (i = 0; i < num_states; i++) - uclk_states[i] = uclk_states[num_states - 1]; - if (clock_limits_available && uclk_states_available && num_states) - update_bounding_box(dc, &dcn2_0_soc, &max_clocks, uclk_states, num_states); + dcn20_update_bounding_box(dc, loaded_bb, &max_clocks, uclk_states, num_states); else if (clock_limits_available) - cap_soc_clocks(&dcn2_0_soc, max_clocks); + dcn20_cap_soc_clocks(loaded_bb, max_clocks); } - dcn2_0_ip.max_num_otg = pool->base.res_cap->num_timing_generator; - dcn2_0_ip.max_num_dpp = pool->base.pipe_count; - patch_bounding_box(dc, &dcn2_0_soc); + loaded_ip->max_num_otg = pool->base.res_cap->num_timing_generator; + loaded_ip->max_num_dpp = pool->base.pipe_count; + dcn20_patch_bounding_box(dc, loaded_bb); return true; } -static bool construct( +static bool dcn20_resource_construct( uint8_t num_virtual_links, struct dc *dc, struct dcn20_resource_pool *pool) @@ -2880,19 +3483,33 @@ static bool construct( int i; struct dc_context *ctx = dc->ctx; struct irq_service_init_data init_data; + struct ddc_service_init_data ddc_init_data; + struct _vcs_dpi_soc_bounding_box_st *loaded_bb = + get_asic_rev_soc_bb(ctx->asic_id.hw_internal_rev); + struct _vcs_dpi_ip_params_st *loaded_ip = + get_asic_rev_ip_params(ctx->asic_id.hw_internal_rev); + enum dml_project dml_project_version = + get_dml_project_version(ctx->asic_id.hw_internal_rev); - ctx->dc_bios->regs = &bios_regs; + DC_FP_START(); - pool->base.res_cap = &res_cap_nv10; + ctx->dc_bios->regs = &bios_regs; pool->base.funcs = &dcn20_res_pool_funcs; + if (ASICREV_IS_NAVI14_M(ctx->asic_id.hw_internal_rev)) { + pool->base.res_cap = &res_cap_nv14; + pool->base.pipe_count = 5; + pool->base.mpcc_count = 5; + } else { + pool->base.res_cap = &res_cap_nv10; + pool->base.pipe_count = 6; + pool->base.mpcc_count = 6; + } /************************************************* * Resource + asic cap harcoding * *************************************************/ pool->base.underlay_pipe_index = NO_UNDERLAY_PIPE; - pool->base.pipe_count = 6; - pool->base.mpcc_count = 6; dc->caps.max_downscale_ratio = 200; dc->caps.i2c_speed_in_khz = 100; dc->caps.max_cursor_size = 256; @@ -2902,6 +3519,7 @@ static bool construct( dc->caps.post_blend_color_processing = true; dc->caps.force_dp_tps4_for_cp2520 = true; dc->caps.hw_3d_lut = true; + dc->caps.extended_aux_timeout_support = true; if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) { dc->debug = debug_defaults_drv; @@ -2998,7 +3616,7 @@ static bool construct( goto create_fail; } - dml_init_instance(&dc->dml, &dcn2_0_soc, &dcn2_0_ip, DML_PROJECT_NAVI10); + dml_init_instance(&dc->dml, loaded_bb, loaded_ip, dml_project_version); if (!dc->debug.disable_pplib_wm_range) { struct pp_smu_wm_range_sets ranges = {0}; @@ -3006,7 +3624,7 @@ static bool construct( ranges.num_reader_wm_sets = 0; - if (dcn2_0_soc.num_states == 1) { + if (loaded_bb->num_states == 1) { ranges.reader_wm_sets[0].wm_inst = i; ranges.reader_wm_sets[0].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; ranges.reader_wm_sets[0].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; @@ -3014,13 +3632,13 @@ static bool construct( ranges.reader_wm_sets[0].max_fill_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; ranges.num_reader_wm_sets = 1; - } else if (dcn2_0_soc.num_states > 1) { - for (i = 0; i < 4 && i < dcn2_0_soc.num_states; i++) { + } else if (loaded_bb->num_states > 1) { + for (i = 0; i < 4 && i < loaded_bb->num_states; i++) { ranges.reader_wm_sets[i].wm_inst = i; ranges.reader_wm_sets[i].min_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MIN; ranges.reader_wm_sets[i].max_drain_clk_mhz = PP_SMU_WM_SET_RANGE_CLK_UNCONSTRAINED_MAX; - ranges.reader_wm_sets[i].min_fill_clk_mhz = (i > 0) ? (dcn2_0_soc.clock_limits[i - 1].dram_speed_mts / 16) + 1 : 0; - ranges.reader_wm_sets[i].max_fill_clk_mhz = dcn2_0_soc.clock_limits[i].dram_speed_mts / 16; + ranges.reader_wm_sets[i].min_fill_clk_mhz = (i > 0) ? (loaded_bb->clock_limits[i - 1].dram_speed_mts / 16) + 1 : 0; + ranges.reader_wm_sets[i].max_fill_clk_mhz = loaded_bb->clock_limits[i].dram_speed_mts / 16; ranges.num_reader_wm_sets = i + 1; } @@ -3127,7 +3745,6 @@ static bool construct( goto create_fail; } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT for (i = 0; i < pool->base.res_cap->num_dsc; i++) { pool->base.dscs[i] = dcn20_dsc_create(ctx, i); if (pool->base.dscs[i] == NULL) { @@ -3136,7 +3753,6 @@ static bool construct( goto create_fail; } } -#endif if (!dcn20_dwbc_create(ctx, &pool->base)) { BREAK_TO_DEBUGGER(); @@ -3163,11 +3779,24 @@ static bool construct( dc->cap_funcs = cap_funcs; + if (dc->ctx->dc_bios->fw_info.oem_i2c_present) { + ddc_init_data.ctx = dc->ctx; + ddc_init_data.link = NULL; + ddc_init_data.id.id = dc->ctx->dc_bios->fw_info.oem_i2c_obj_id; + ddc_init_data.id.enum_id = 0; + ddc_init_data.id.type = OBJECT_TYPE_GENERIC; + pool->base.oem_device = dal_ddc_service_create(&ddc_init_data); + } else { + pool->base.oem_device = NULL; + } + + DC_FP_END(); return true; create_fail: - destruct(pool); + DC_FP_END(); + dcn20_resource_destruct(pool); return false; } @@ -3182,7 +3811,7 @@ struct resource_pool *dcn20_create_resource_pool( if (!pool) return NULL; - if (construct(init_data->num_virtual_links, dc, pool)) + if (dcn20_resource_construct(init_data->num_virtual_links, dc, pool)) return &pool->base; BREAK_TO_DEBUGGER(); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h index b5a75289f444..f5893840b79b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h @@ -50,7 +50,7 @@ unsigned int dcn20_calc_max_scaled_time( enum mmhubbub_wbif_mode mode, unsigned int urgent_watermark); int dcn20_populate_dml_pipes_from_context( - struct dc *dc, struct resource_context *res_ctx, display_e2e_pipe_params_st *pipes); + struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes); struct pipe_ctx *dcn20_acquire_idle_pipe_for_layer( struct dc_state *state, const struct resource_pool *pool, @@ -95,9 +95,12 @@ struct display_stream_compressor *dcn20_dsc_create( struct dc_context *ctx, uint32_t inst); void dcn20_dsc_destroy(struct display_stream_compressor **dsc); -struct pp_smu_funcs *dcn20_pp_smu_create(struct dc_context *ctx); -void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu); - +void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb); +void dcn20_cap_soc_clocks( + struct _vcs_dpi_soc_bounding_box_st *bb, + struct pp_smu_nv_clock_table max_clocks); +void dcn20_update_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb, + struct pp_smu_nv_clock_table *max_clocks, unsigned int *uclk_states, unsigned int num_states); struct hubp *dcn20_hubp_create( struct dc_context *ctx, uint32_t inst); @@ -116,9 +119,45 @@ void dcn20_set_mcif_arb_params( display_e2e_pipe_params_st *pipes, int pipe_cnt); bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate); +void dcn20_merge_pipes_for_validate( + struct dc *dc, + struct dc_state *context); +int dcn20_validate_apply_pipe_split_flags( + struct dc *dc, + struct dc_state *context, + int vlevel, + bool *split); +bool dcn20_validate_dsc(struct dc *dc, struct dc_state *new_ctx); +void dcn20_split_stream_for_mpc( + struct resource_context *res_ctx, + const struct resource_pool *pool, + struct pipe_ctx *primary_pipe, + struct pipe_ctx *secondary_pipe); +bool dcn20_split_stream_for_odm( + struct resource_context *res_ctx, + const struct resource_pool *pool, + struct pipe_ctx *prev_odm_pipe, + struct pipe_ctx *next_odm_pipe); +struct pipe_ctx *dcn20_find_secondary_pipe(struct dc *dc, + struct resource_context *res_ctx, + const struct resource_pool *pool, + const struct pipe_ctx *primary_pipe); +bool dcn20_fast_validate_bw( + struct dc *dc, + struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *pipe_cnt_out, + int *pipe_split_from, + int *vlevel_out); +void dcn20_calculate_dlg_params( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int pipe_cnt, + int vlevel); enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state *context, struct dc_stream_state *stream); enum dc_status dcn20_add_stream_to_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); +enum dc_status dcn20_add_dsc_to_stream_resource(struct dc *dc, struct dc_state *dc_ctx, struct dc_stream_state *dc_stream); enum dc_status dcn20_remove_stream_from_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); enum dc_status dcn20_get_default_swizzle_mode(struct dc_plane_state *plane_state); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c index f5bcffc426b8..9b70a1e7b962 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c @@ -205,11 +205,9 @@ static void enc2_stream_encoder_stop_hdmi_info_packets( HDMI_GENERIC7_LINE, 0); } -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - /* Update GSP7 SDP 128 byte long */ -static void enc2_send_gsp7_128_info_packet( +static void enc2_update_gsp7_128_info_packet( struct dcn10_stream_encoder *enc1, const struct dc_info_packet_128 *info_packet) { @@ -277,18 +275,9 @@ static void enc2_send_gsp7_128_info_packet( static void enc2_dp_set_dsc_config(struct stream_encoder *enc, enum optc_dsc_mode dsc_mode, uint32_t dsc_bytes_per_pixel, - uint32_t dsc_slice_width, - uint8_t *dsc_packed_pps) + uint32_t dsc_slice_width) { struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); - uint32_t dsc_value = 0; - - dsc_value = REG_READ(DP_DSC_CNTL); - - /* dsc disable skip */ - if ((dsc_value & 0x3) == 0x0) - return; - REG_UPDATE_2(DP_DSC_CNTL, DP_DSC_MODE, dsc_mode, @@ -296,8 +285,16 @@ static void enc2_dp_set_dsc_config(struct stream_encoder *enc, REG_SET(DP_DSC_BYTES_PER_PIXEL, 0, DP_DSC_BYTES_PER_PIXEL, dsc_bytes_per_pixel); +} + - if (dsc_mode != OPTC_DSC_DISABLED) { +static void enc2_dp_set_dsc_pps_info_packet(struct stream_encoder *enc, + bool enable, + uint8_t *dsc_packed_pps) +{ + struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); + + if (enable) { struct dc_info_packet_128 pps_sdp; ASSERT(dsc_packed_pps); @@ -309,7 +306,7 @@ static void enc2_dp_set_dsc_config(struct stream_encoder *enc, pps_sdp.hb2 = 127; pps_sdp.hb3 = 0; memcpy(&pps_sdp.sb[0], dsc_packed_pps, sizeof(pps_sdp.sb)); - enc2_send_gsp7_128_info_packet(enc1, &pps_sdp); + enc2_update_gsp7_128_info_packet(enc1, &pps_sdp); /* Enable Generic Stream Packet 7 (GSP) transmission */ //REG_UPDATE(DP_SEC_CNTL, @@ -340,9 +337,8 @@ static void enc2_dp_set_dsc_config(struct stream_encoder *enc, REG_UPDATE(DP_SEC_CNTL2, DP_SEC_GSP7_PPS, 0); } } -#endif -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + /* this function read dsc related register fields to be logged later in dcn10_log_hw_state * into a dcn_dsc_state struct. */ @@ -363,7 +359,6 @@ static void enc2_read_state(struct stream_encoder *enc, struct enc_state *s) REG_GET(DP_SEC_CNTL, DP_SEC_STREAM_ENABLE, &s->sec_stream_enable); } } -#endif /* Set Dynamic Metadata-configuration. * enable_dme: TRUE: enables Dynamic Metadata Enfine, FALSE: disables DME @@ -373,7 +368,7 @@ static void enc2_read_state(struct stream_encoder *enc, struct enc_state *s) * * Ensure the OTG master update lock is set when changing DME configuration. */ -static void enc2_set_dynamic_metadata(struct stream_encoder *enc, +void enc2_set_dynamic_metadata(struct stream_encoder *enc, bool enable_dme, uint32_t hubp_requestor_id, enum dynamic_metadata_mode dmdata_mode) @@ -443,10 +438,8 @@ static bool is_two_pixels_per_containter(const struct dc_crtc_timing *timing) { bool two_pix = timing->pixel_encoding == PIXEL_ENCODING_YCBCR420; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT two_pix = two_pix || (timing->flags.DSC && timing->pixel_encoding == PIXEL_ENCODING_YCBCR422 && !timing->dsc_cfg.ycbcr422_simple); -#endif return two_pix; } @@ -463,7 +456,7 @@ void enc2_stream_encoder_dp_unblank( uint64_t m_vid_l = n_vid; /* YCbCr 4:2:0 : Computed VID_M will be 2X the input rate */ - if (is_two_pixels_per_containter(¶m->timing) || param->odm) { + if (is_two_pixels_per_containter(¶m->timing) || param->opp_cnt > 1) { /*this logic should be the same in get_pixel_clock_parameters() */ n_multiply = 1; } @@ -495,15 +488,23 @@ void enc2_stream_encoder_dp_unblank( DP_VID_N_MUL, n_multiply); } - /* set DIG_START to 0x1 to reset FIFO */ + /* make sure stream is disabled before resetting steer fifo */ + REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_ENABLE, false); + REG_WAIT(DP_VID_STREAM_CNTL, DP_VID_STREAM_STATUS, 0, 10, 5000); + /* set DIG_START to 0x1 to reset FIFO */ REG_UPDATE(DIG_FE_CNTL, DIG_START, 1); + udelay(1); /* write 0 to take the FIFO out of reset */ REG_UPDATE(DIG_FE_CNTL, DIG_START, 0); - /* switch DP encoder to CRTC data */ + /* switch DP encoder to CRTC data, but reset it the fifo first. It may happen + * that it overflows during mode transition, and sometimes doesn't recover. + */ + REG_UPDATE(DP_STEER_FIFO, DP_STEER_FIFO_RESET, 1); + udelay(10); REG_UPDATE(DP_STEER_FIFO, DP_STEER_FIFO_RESET, 0); @@ -536,11 +537,16 @@ void enc2_stream_encoder_dp_set_stream_attribute( struct stream_encoder *enc, struct dc_crtc_timing *crtc_timing, enum dc_color_space output_color_space, + bool use_vsc_sdp_for_colorimetry, uint32_t enable_sdp_splitting) { struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); - enc1_stream_encoder_dp_set_stream_attribute(enc, crtc_timing, output_color_space, enable_sdp_splitting); + enc1_stream_encoder_dp_set_stream_attribute(enc, + crtc_timing, + output_color_space, + use_vsc_sdp_for_colorimetry, + enable_sdp_splitting); REG_UPDATE(DP_SEC_FRAMING4, DP_SST_SDP_SPLITTING, enable_sdp_splitting); @@ -563,6 +569,8 @@ static const struct stream_encoder_funcs dcn20_str_enc_funcs = { enc2_stream_encoder_stop_hdmi_info_packets, .update_dp_info_packets = enc2_stream_encoder_update_dp_info_packets, + .send_immediate_sdp_message = + enc1_stream_encoder_send_immediate_sdp_message, .stop_dp_info_packets = enc1_stream_encoder_stop_dp_info_packets, .dp_blank = @@ -580,14 +588,16 @@ static const struct stream_encoder_funcs dcn20_str_enc_funcs = { .setup_stereo_sync = enc1_setup_stereo_sync, .set_avmute = enc1_stream_encoder_set_avmute, .dig_connect_to_otg = enc1_dig_connect_to_otg, -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - .enc_read_state = enc2_read_state, -#endif + .dig_source_otg = enc1_dig_source_otg, -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + .dp_get_pixel_format = + enc1_stream_encoder_dp_get_pixel_format, + + .enc_read_state = enc2_read_state, .dp_set_dsc_config = enc2_dp_set_dsc_config, -#endif + .dp_set_dsc_pps_info_packet = enc2_dp_set_dsc_pps_info_packet, .set_dynamic_metadata = enc2_set_dynamic_metadata, + .hdmi_reset_stream_attribute = enc1_reset_hdmi_stream_attribute, }; void dcn20_stream_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.h index 6d40e8c9b78f..d2a805bd4573 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.h @@ -98,10 +98,16 @@ void enc2_stream_encoder_dp_set_stream_attribute( struct stream_encoder *enc, struct dc_crtc_timing *crtc_timing, enum dc_color_space output_color_space, + bool use_vsc_sdp_for_colorimetry, uint32_t enable_sdp_splitting); void enc2_stream_encoder_dp_unblank( struct stream_encoder *enc, const struct encoder_unblank_param *param); +void enc2_set_dynamic_metadata(struct stream_encoder *enc, + bool enable_dme, + uint32_t hubp_requestor_id, + enum dynamic_metadata_mode dmdata_mode); + #endif /* __DC_STREAM_ENCODER_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile new file mode 100644 index 000000000000..07684d3e375a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile @@ -0,0 +1,35 @@ +# SPDX-License-Identifier: MIT +# +# Makefile for DCN21. + +DCN21 = dcn21_init.o dcn21_hubp.o dcn21_hubbub.o dcn21_resource.o \ + dcn21_hwseq.o dcn21_link_encoder.o + +ifdef CONFIG_X86 +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -msse +endif + +ifdef CONFIG_PPC64 +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o := -mhard-float -maltivec +endif + +ifdef CONFIG_CC_IS_GCC +ifeq ($(call cc-ifversion, -lt, 0701, y), y) +IS_OLD_GCC = 1 +endif +endif + +ifdef CONFIG_X86 +ifdef IS_OLD_GCC +# Stack alignment mismatch, proceed with caution. +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 +# (8B stack alignment). +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -mpreferred-stack-boundary=4 +else +CFLAGS_$(AMDDALPATH)/dc/dcn21/dcn21_resource.o += -msse2 +endif +endif + +AMD_DAL_DCN21 = $(addprefix $(AMDDALPATH)/dc/dcn21/,$(DCN21)) + +AMD_DISPLAY_FILES += $(AMD_DAL_DCN21) diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c new file mode 100644 index 000000000000..f546260c15b7 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c @@ -0,0 +1,657 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include <linux/delay.h> +#include "dm_services.h" +#include "dcn20/dcn20_hubbub.h" +#include "dcn21_hubbub.h" +#include "reg_helper.h" + +#define REG(reg)\ + hubbub1->regs->reg +#define DC_LOGGER \ + hubbub1->base.ctx->logger +#define CTX \ + hubbub1->base.ctx + +#undef FN +#define FN(reg_name, field_name) \ + hubbub1->shifts->field_name, hubbub1->masks->field_name + +#define REG(reg)\ + hubbub1->regs->reg + +#define CTX \ + hubbub1->base.ctx + +#undef FN +#define FN(reg_name, field_name) \ + hubbub1->shifts->field_name, hubbub1->masks->field_name + +#ifdef NUM_VMID +#undef NUM_VMID +#endif +#define NUM_VMID 16 + +static uint32_t convert_and_clamp( + uint32_t wm_ns, + uint32_t refclk_mhz, + uint32_t clamp_value) +{ + uint32_t ret_val = 0; + ret_val = wm_ns * refclk_mhz; + ret_val /= 1000; + + if (ret_val > clamp_value) + ret_val = clamp_value; + + return ret_val; +} + +void dcn21_dchvm_init(struct hubbub *hubbub) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + uint32_t riommu_active; + int i; + + //Init DCHVM block + REG_UPDATE(DCHVM_CTRL0, HOSTVM_INIT_REQ, 1); + + //Poll until RIOMMU_ACTIVE = 1 + for (i = 0; i < 100; i++) { + REG_GET(DCHVM_RIOMMU_STAT0, RIOMMU_ACTIVE, &riommu_active); + + if (riommu_active) + break; + else + udelay(5); + } + + if (riommu_active) { + //Reflect the power status of DCHUBBUB + REG_UPDATE(DCHVM_RIOMMU_CTRL0, HOSTVM_POWERSTATUS, 1); + + //Start rIOMMU prefetching + REG_UPDATE(DCHVM_RIOMMU_CTRL0, HOSTVM_PREFETCH_REQ, 1); + + // Enable dynamic clock gating + REG_UPDATE_4(DCHVM_CLK_CTRL, + HVM_DISPCLK_R_GATE_DIS, 0, + HVM_DISPCLK_G_GATE_DIS, 0, + HVM_DCFCLK_R_GATE_DIS, 0, + HVM_DCFCLK_G_GATE_DIS, 0); + + //Poll until HOSTVM_PREFETCH_DONE = 1 + REG_WAIT(DCHVM_RIOMMU_STAT0, HOSTVM_PREFETCH_DONE, 1, 5, 100); + } +} + +int hubbub21_init_dchub(struct hubbub *hubbub, + struct dcn_hubbub_phys_addr_config *pa_config) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + struct dcn_vmid_page_table_config phys_config; + + REG_SET(DCN_VM_FB_LOCATION_BASE, 0, + FB_BASE, pa_config->system_aperture.fb_base >> 24); + REG_SET(DCN_VM_FB_LOCATION_TOP, 0, + FB_TOP, pa_config->system_aperture.fb_top >> 24); + REG_SET(DCN_VM_FB_OFFSET, 0, + FB_OFFSET, pa_config->system_aperture.fb_offset >> 24); + REG_SET(DCN_VM_AGP_BOT, 0, + AGP_BOT, pa_config->system_aperture.agp_bot >> 24); + REG_SET(DCN_VM_AGP_TOP, 0, + AGP_TOP, pa_config->system_aperture.agp_top >> 24); + REG_SET(DCN_VM_AGP_BASE, 0, + AGP_BASE, pa_config->system_aperture.agp_base >> 24); + + if (pa_config->gart_config.page_table_start_addr != pa_config->gart_config.page_table_end_addr) { + phys_config.page_table_start_addr = pa_config->gart_config.page_table_start_addr >> 12; + phys_config.page_table_end_addr = pa_config->gart_config.page_table_end_addr >> 12; + phys_config.page_table_base_addr = pa_config->gart_config.page_table_base_addr | 1; //Note: hack + phys_config.depth = 0; + phys_config.block_size = 0; + // Init VMID 0 based on PA config + dcn20_vmid_setup(&hubbub1->vmid[0], &phys_config); + } + + dcn21_dchvm_init(hubbub); + + return NUM_VMID; +} + +void hubbub21_program_urgent_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + uint32_t prog_wm_value; + + /* Repeat for water mark set A, B, C and D. */ + /* clock state A */ + if (safe_to_lower || watermarks->a.urgent_ns > hubbub1->watermarks.a.urgent_ns) { + hubbub1->watermarks.a.urgent_ns = watermarks->a.urgent_ns; + prog_wm_value = convert_and_clamp(watermarks->a.urgent_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_A, prog_wm_value); + + DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_A calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->a.urgent_ns, prog_wm_value); + } + + /* determine the transfer time for a quantity of data for a particular requestor.*/ + if (safe_to_lower || watermarks->a.frac_urg_bw_flip + > hubbub1->watermarks.a.frac_urg_bw_flip) { + hubbub1->watermarks.a.frac_urg_bw_flip = watermarks->a.frac_urg_bw_flip; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, 0, + DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, watermarks->a.frac_urg_bw_flip); + } + + if (safe_to_lower || watermarks->a.frac_urg_bw_nom + > hubbub1->watermarks.a.frac_urg_bw_nom) { + hubbub1->watermarks.a.frac_urg_bw_nom = watermarks->a.frac_urg_bw_nom; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, 0, + DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, watermarks->a.frac_urg_bw_nom); + } + if (safe_to_lower || watermarks->a.urgent_latency_ns > hubbub1->watermarks.a.urgent_latency_ns) { + hubbub1->watermarks.a.urgent_latency_ns = watermarks->a.urgent_latency_ns; + prog_wm_value = convert_and_clamp(watermarks->a.urgent_latency_ns, + refclk_mhz, 0x1fffff); + REG_SET(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A, 0, + DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A, prog_wm_value); + } + + /* clock state B */ + if (safe_to_lower || watermarks->b.urgent_ns > hubbub1->watermarks.b.urgent_ns) { + hubbub1->watermarks.b.urgent_ns = watermarks->b.urgent_ns; + prog_wm_value = convert_and_clamp(watermarks->b.urgent_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_B, prog_wm_value); + + DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_B calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->b.urgent_ns, prog_wm_value); + } + + /* determine the transfer time for a quantity of data for a particular requestor.*/ + if (safe_to_lower || watermarks->a.frac_urg_bw_flip + > hubbub1->watermarks.a.frac_urg_bw_flip) { + hubbub1->watermarks.a.frac_urg_bw_flip = watermarks->a.frac_urg_bw_flip; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, 0, + DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, watermarks->a.frac_urg_bw_flip); + } + + if (safe_to_lower || watermarks->a.frac_urg_bw_nom + > hubbub1->watermarks.a.frac_urg_bw_nom) { + hubbub1->watermarks.a.frac_urg_bw_nom = watermarks->a.frac_urg_bw_nom; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, 0, + DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, watermarks->a.frac_urg_bw_nom); + } + + if (safe_to_lower || watermarks->b.urgent_latency_ns > hubbub1->watermarks.b.urgent_latency_ns) { + hubbub1->watermarks.b.urgent_latency_ns = watermarks->b.urgent_latency_ns; + prog_wm_value = convert_and_clamp(watermarks->b.urgent_latency_ns, + refclk_mhz, 0x1fffff); + REG_SET(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B, 0, + DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B, prog_wm_value); + } + + /* clock state C */ + if (safe_to_lower || watermarks->c.urgent_ns > hubbub1->watermarks.c.urgent_ns) { + hubbub1->watermarks.c.urgent_ns = watermarks->c.urgent_ns; + prog_wm_value = convert_and_clamp(watermarks->c.urgent_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_C, prog_wm_value); + + DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_C calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->c.urgent_ns, prog_wm_value); + } + + /* determine the transfer time for a quantity of data for a particular requestor.*/ + if (safe_to_lower || watermarks->a.frac_urg_bw_flip + > hubbub1->watermarks.a.frac_urg_bw_flip) { + hubbub1->watermarks.a.frac_urg_bw_flip = watermarks->a.frac_urg_bw_flip; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, 0, + DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, watermarks->a.frac_urg_bw_flip); + } + + if (safe_to_lower || watermarks->a.frac_urg_bw_nom + > hubbub1->watermarks.a.frac_urg_bw_nom) { + hubbub1->watermarks.a.frac_urg_bw_nom = watermarks->a.frac_urg_bw_nom; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, 0, + DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, watermarks->a.frac_urg_bw_nom); + } + + if (safe_to_lower || watermarks->c.urgent_latency_ns > hubbub1->watermarks.c.urgent_latency_ns) { + hubbub1->watermarks.c.urgent_latency_ns = watermarks->c.urgent_latency_ns; + prog_wm_value = convert_and_clamp(watermarks->c.urgent_latency_ns, + refclk_mhz, 0x1fffff); + REG_SET(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C, 0, + DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C, prog_wm_value); + } + + /* clock state D */ + if (safe_to_lower || watermarks->d.urgent_ns > hubbub1->watermarks.d.urgent_ns) { + hubbub1->watermarks.d.urgent_ns = watermarks->d.urgent_ns; + prog_wm_value = convert_and_clamp(watermarks->d.urgent_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, 0, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_D, prog_wm_value); + + DC_LOG_BANDWIDTH_CALCS("URGENCY_WATERMARK_D calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->d.urgent_ns, prog_wm_value); + } + + /* determine the transfer time for a quantity of data for a particular requestor.*/ + if (safe_to_lower || watermarks->a.frac_urg_bw_flip + > hubbub1->watermarks.a.frac_urg_bw_flip) { + hubbub1->watermarks.a.frac_urg_bw_flip = watermarks->a.frac_urg_bw_flip; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, 0, + DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, watermarks->a.frac_urg_bw_flip); + } + + if (safe_to_lower || watermarks->a.frac_urg_bw_nom + > hubbub1->watermarks.a.frac_urg_bw_nom) { + hubbub1->watermarks.a.frac_urg_bw_nom = watermarks->a.frac_urg_bw_nom; + + REG_SET(DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, 0, + DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, watermarks->a.frac_urg_bw_nom); + } + + if (safe_to_lower || watermarks->d.urgent_latency_ns > hubbub1->watermarks.d.urgent_latency_ns) { + hubbub1->watermarks.d.urgent_latency_ns = watermarks->d.urgent_latency_ns; + prog_wm_value = convert_and_clamp(watermarks->d.urgent_latency_ns, + refclk_mhz, 0x1fffff); + REG_SET(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D, 0, + DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D, prog_wm_value); + } +} + +void hubbub21_program_stutter_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + uint32_t prog_wm_value; + + /* clock state A */ + if (safe_to_lower || watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns + > hubbub1->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns) { + hubbub1->watermarks.a.cstate_pstate.cstate_enter_plus_exit_ns = + watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_A calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->a.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); + } + + if (safe_to_lower || watermarks->a.cstate_pstate.cstate_exit_ns + > hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns) { + hubbub1->watermarks.a.cstate_pstate.cstate_exit_ns = + watermarks->a.cstate_pstate.cstate_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->a.cstate_pstate.cstate_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_A calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->a.cstate_pstate.cstate_exit_ns, prog_wm_value); + } + + /* clock state B */ + if (safe_to_lower || watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns + > hubbub1->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns) { + hubbub1->watermarks.b.cstate_pstate.cstate_enter_plus_exit_ns = + watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_B, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_B calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->b.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); + } + + if (safe_to_lower || watermarks->b.cstate_pstate.cstate_exit_ns + > hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns) { + hubbub1->watermarks.b.cstate_pstate.cstate_exit_ns = + watermarks->b.cstate_pstate.cstate_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->b.cstate_pstate.cstate_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_B calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->b.cstate_pstate.cstate_exit_ns, prog_wm_value); + } + + /* clock state C */ + if (safe_to_lower || watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns + > hubbub1->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns) { + hubbub1->watermarks.c.cstate_pstate.cstate_enter_plus_exit_ns = + watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_C, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_C calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->c.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); + } + + if (safe_to_lower || watermarks->c.cstate_pstate.cstate_exit_ns + > hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns) { + hubbub1->watermarks.c.cstate_pstate.cstate_exit_ns = + watermarks->c.cstate_pstate.cstate_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->c.cstate_pstate.cstate_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_C calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->c.cstate_pstate.cstate_exit_ns, prog_wm_value); + } + + /* clock state D */ + if (safe_to_lower || watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns + > hubbub1->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns) { + hubbub1->watermarks.d.cstate_pstate.cstate_enter_plus_exit_ns = + watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, 0, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_D, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_ENTER_EXIT_WATERMARK_D calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->d.cstate_pstate.cstate_enter_plus_exit_ns, prog_wm_value); + } + + if (safe_to_lower || watermarks->d.cstate_pstate.cstate_exit_ns + > hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns) { + hubbub1->watermarks.d.cstate_pstate.cstate_exit_ns = + watermarks->d.cstate_pstate.cstate_exit_ns; + prog_wm_value = convert_and_clamp( + watermarks->d.cstate_pstate.cstate_exit_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, 0, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("SR_EXIT_WATERMARK_D calculated =%d\n" + "HW register value = 0x%x\n", + watermarks->d.cstate_pstate.cstate_exit_ns, prog_wm_value); + } +} + +void hubbub21_program_pstate_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + uint32_t prog_wm_value; + + /* clock state A */ + if (safe_to_lower || watermarks->a.cstate_pstate.pstate_change_ns + > hubbub1->watermarks.a.cstate_pstate.pstate_change_ns) { + hubbub1->watermarks.a.cstate_pstate.pstate_change_ns = + watermarks->a.cstate_pstate.pstate_change_ns; + prog_wm_value = convert_and_clamp( + watermarks->a.cstate_pstate.pstate_change_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_A calculated =%d\n" + "HW register value = 0x%x\n\n", + watermarks->a.cstate_pstate.pstate_change_ns, prog_wm_value); + } + + /* clock state B */ + if (safe_to_lower || watermarks->b.cstate_pstate.pstate_change_ns + > hubbub1->watermarks.b.cstate_pstate.pstate_change_ns) { + hubbub1->watermarks.b.cstate_pstate.pstate_change_ns = + watermarks->b.cstate_pstate.pstate_change_ns; + prog_wm_value = convert_and_clamp( + watermarks->b.cstate_pstate.pstate_change_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_B calculated =%d\n" + "HW register value = 0x%x\n\n", + watermarks->b.cstate_pstate.pstate_change_ns, prog_wm_value); + } + + /* clock state C */ + if (safe_to_lower || watermarks->c.cstate_pstate.pstate_change_ns + > hubbub1->watermarks.c.cstate_pstate.pstate_change_ns) { + hubbub1->watermarks.c.cstate_pstate.pstate_change_ns = + watermarks->c.cstate_pstate.pstate_change_ns; + prog_wm_value = convert_and_clamp( + watermarks->c.cstate_pstate.pstate_change_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_C calculated =%d\n" + "HW register value = 0x%x\n\n", + watermarks->c.cstate_pstate.pstate_change_ns, prog_wm_value); + } + + /* clock state D */ + if (safe_to_lower || watermarks->d.cstate_pstate.pstate_change_ns + > hubbub1->watermarks.d.cstate_pstate.pstate_change_ns) { + hubbub1->watermarks.d.cstate_pstate.pstate_change_ns = + watermarks->d.cstate_pstate.pstate_change_ns; + prog_wm_value = convert_and_clamp( + watermarks->d.cstate_pstate.pstate_change_ns, + refclk_mhz, 0x1fffff); + REG_SET_2(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, 0, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, prog_wm_value, + DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, prog_wm_value); + DC_LOG_BANDWIDTH_CALCS("DRAM_CLK_CHANGE_WATERMARK_D calculated =%d\n" + "HW register value = 0x%x\n\n", + watermarks->d.cstate_pstate.pstate_change_ns, prog_wm_value); + } +} + +void hubbub21_program_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + + hubbub21_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + hubbub21_program_stutter_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + hubbub21_program_pstate_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower); + + /* + * The DCHub arbiter has a mechanism to dynamically rate limit the DCHub request stream to the fabric. + * If the memory controller is fully utilized and the DCHub requestors are + * well ahead of their amortized schedule, then it is safe to prevent the next winner + * from being committed and sent to the fabric. + * The utilization of the memory controller is approximated by ensuring that + * the number of outstanding requests is greater than a threshold specified + * by the ARB_MIN_REQ_OUTSTANDING. To determine that the DCHub requestors are well ahead of the amortized schedule, + * the slack of the next winner is compared with the ARB_SAT_LEVEL in DLG RefClk cycles. + * + * TODO: Revisit request limit after figure out right number. request limit for Renoir isn't decided yet, set maximum value (0x1FF) + * to turn off it for now. + */ + REG_SET(DCHUBBUB_ARB_SAT_LEVEL, 0, + DCHUBBUB_ARB_SAT_LEVEL, 60 * refclk_mhz); + REG_UPDATE_2(DCHUBBUB_ARB_DF_REQ_OUTSTAND, + DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 0x1FF, + DCHUBBUB_ARB_MIN_REQ_OUTSTAND_COMMIT_THRESHOLD, 0xA); + REG_UPDATE(DCHUBBUB_ARB_HOSTVM_CNTL, + DCHUBBUB_ARB_MAX_QOS_COMMIT_THRESHOLD, 0xF); + + hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); +} + +void hubbub21_wm_read_state(struct hubbub *hubbub, + struct dcn_hubbub_wm *wm) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + struct dcn_hubbub_wm_set *s; + + memset(wm, 0, sizeof(struct dcn_hubbub_wm)); + + s = &wm->sets[0]; + s->wm_set = 0; + REG_GET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, &s->data_urgent); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, &s->sr_enter); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, &s->sr_exit); + + REG_GET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, &s->dram_clk_chanage); + + s = &wm->sets[1]; + s->wm_set = 1; + REG_GET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, &s->data_urgent); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, &s->sr_enter); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, &s->sr_exit); + + REG_GET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, &s->dram_clk_chanage); + + s = &wm->sets[2]; + s->wm_set = 2; + REG_GET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, &s->data_urgent); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, &s->sr_enter); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, &s->sr_exit); + + REG_GET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, &s->dram_clk_chanage); + + s = &wm->sets[3]; + s->wm_set = 3; + REG_GET(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, + DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, &s->data_urgent); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, + DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, &s->sr_enter); + + REG_GET(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, + DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, &s->sr_exit); + + REG_GET(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, + DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, &s->dram_clk_chanage); +} + +void hubbub21_apply_DEDCN21_147_wa(struct hubbub *hubbub) +{ + struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); + uint32_t prog_wm_value; + + prog_wm_value = REG_READ(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A); + REG_WRITE(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, prog_wm_value); +} + +static const struct hubbub_funcs hubbub21_funcs = { + .update_dchub = hubbub2_update_dchub, + .init_dchub_sys_ctx = hubbub21_init_dchub, + .init_vm_ctx = hubbub2_init_vm_ctx, + .dcc_support_swizzle = hubbub2_dcc_support_swizzle, + .dcc_support_pixel_format = hubbub2_dcc_support_pixel_format, + .get_dcc_compression_cap = hubbub2_get_dcc_compression_cap, + .wm_read_state = hubbub21_wm_read_state, + .get_dchub_ref_freq = hubbub2_get_dchub_ref_freq, + .program_watermarks = hubbub21_program_watermarks, + .apply_DEDCN21_147_wa = hubbub21_apply_DEDCN21_147_wa, +}; + +void hubbub21_construct(struct dcn20_hubbub *hubbub, + struct dc_context *ctx, + const struct dcn_hubbub_registers *hubbub_regs, + const struct dcn_hubbub_shift *hubbub_shift, + const struct dcn_hubbub_mask *hubbub_mask) +{ + hubbub->base.ctx = ctx; + + hubbub->base.funcs = &hubbub21_funcs; + + hubbub->regs = hubbub_regs; + hubbub->shifts = hubbub_shift; + hubbub->masks = hubbub_mask; + + hubbub->debug_test_index_pstate = 0xB; + hubbub->detile_buf_size = 164 * 1024; /* 164KB for DCN2.0 */ +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h new file mode 100644 index 000000000000..c4840dfb1fa5 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.h @@ -0,0 +1,146 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef DAL_DC_DCN21_DCN21_HUBBUB_H_ +#define DAL_DC_DCN21_DCN21_HUBBUB_H_ + +#include "dcn20/dcn20_hubbub.h" + +#define HUBBUB_HVM_REG_LIST() \ + SR(DCHUBBUB_ARB_FRAC_URG_BW_NOM_A),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_NOM_B),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_NOM_C),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_NOM_D),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C),\ + SR(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D),\ + SR(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A),\ + SR(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B),\ + SR(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C),\ + SR(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D),\ + SR(DCHUBBUB_ARB_HOSTVM_CNTL), \ + SR(DCHVM_CTRL0), \ + SR(DCHVM_MEM_CTRL), \ + SR(DCHVM_CLK_CTRL), \ + SR(DCHVM_RIOMMU_CTRL0), \ + SR(DCHVM_RIOMMU_STAT0) + +#define HUBBUB_REG_LIST_DCN21()\ + HUBBUB_REG_LIST_DCN20_COMMON(), \ + HUBBUB_SR_WATERMARK_REG_LIST(), \ + HUBBUB_HVM_REG_LIST() + +#define HUBBUB_MASK_SH_LIST_HVM(mask_sh) \ + HUBBUB_SF(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND_COMMIT_THRESHOLD, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, DCHUBBUB_ARB_FRAC_URG_BW_FLIP_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, DCHUBBUB_ARB_FRAC_URG_BW_NOM_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, DCHUBBUB_ARB_FRAC_URG_BW_NOM_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, DCHUBBUB_ARB_FRAC_URG_BW_NOM_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, DCHUBBUB_ARB_FRAC_URG_BW_NOM_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_A, DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_B, DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_C, DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_DATA_URGENCY_WATERMARK_D, DCHUBBUB_ARB_VM_ROW_URGENCY_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_A, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_B, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_C, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_ENTER_WATERMARK_D, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_ENTER_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_A, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_B, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_C, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_SR_EXIT_WATERMARK_D, DCHUBBUB_ARB_VM_ROW_ALLOW_SR_EXIT_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, DCHUBBUB_ARB_VM_ROW_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_HOSTVM_CNTL, DCHUBBUB_ARB_MAX_QOS_COMMIT_THRESHOLD, mask_sh), \ + HUBBUB_SF(DCHVM_CTRL0, HOSTVM_INIT_REQ, mask_sh), \ + HUBBUB_SF(DCHVM_MEM_CTRL, HVM_GPUVMRET_PWR_REQ_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_MEM_CTRL, HVM_GPUVMRET_FORCE_REQ, mask_sh), \ + HUBBUB_SF(DCHVM_MEM_CTRL, HVM_GPUVMRET_POWER_STATUS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, HVM_DISPCLK_R_GATE_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, HVM_DISPCLK_G_GATE_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, HVM_DCFCLK_R_GATE_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, HVM_DCFCLK_G_GATE_DIS, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, TR_REQ_REQCLKREQ_MODE, mask_sh), \ + HUBBUB_SF(DCHVM_CLK_CTRL, TW_RSP_COMPCLKREQ_MODE, mask_sh), \ + HUBBUB_SF(DCHVM_RIOMMU_CTRL0, HOSTVM_PREFETCH_REQ, mask_sh), \ + HUBBUB_SF(DCHVM_RIOMMU_CTRL0, HOSTVM_POWERSTATUS, mask_sh), \ + HUBBUB_SF(DCHVM_RIOMMU_STAT0, RIOMMU_ACTIVE, mask_sh), \ + HUBBUB_SF(DCHVM_RIOMMU_STAT0, HOSTVM_PREFETCH_DONE, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_A, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_B, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_C, mask_sh), \ + HUBBUB_SF(DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D, DCHUBBUB_ARB_REFCYC_PER_TRIP_TO_MEMORY_D, mask_sh) + +#define HUBBUB_MASK_SH_LIST_DCN21(mask_sh)\ + HUBBUB_MASK_SH_LIST_HVM(mask_sh), \ + HUBBUB_MASK_SH_LIST_DCN_COMMON(mask_sh), \ + HUBBUB_MASK_SH_LIST_STUTTER(mask_sh), \ + HUBBUB_SF(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, mask_sh), \ + HUBBUB_SF(DCN_VM_FB_LOCATION_BASE, FB_BASE, mask_sh), \ + HUBBUB_SF(DCN_VM_FB_LOCATION_TOP, FB_TOP, mask_sh), \ + HUBBUB_SF(DCN_VM_FB_OFFSET, FB_OFFSET, mask_sh), \ + HUBBUB_SF(DCN_VM_AGP_BOT, AGP_BOT, mask_sh), \ + HUBBUB_SF(DCN_VM_AGP_TOP, AGP_TOP, mask_sh), \ + HUBBUB_SF(DCN_VM_AGP_BASE, AGP_BASE, mask_sh) + +void dcn21_dchvm_init(struct hubbub *hubbub); +int hubbub21_init_dchub(struct hubbub *hubbub, + struct dcn_hubbub_phys_addr_config *pa_config); +void hubbub21_program_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower); +void hubbub21_program_urgent_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower); +void hubbub21_program_stutter_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower); +void hubbub21_program_pstate_watermarks( + struct hubbub *hubbub, + struct dcn_watermark_set *watermarks, + unsigned int refclk_mhz, + bool safe_to_lower); + +void hubbub21_wm_read_state(struct hubbub *hubbub, + struct dcn_hubbub_wm *wm); + +void hubbub21_construct(struct dcn20_hubbub *hubbub, + struct dc_context *ctx, + const struct dcn_hubbub_registers *hubbub_regs, + const struct dcn_hubbub_shift *hubbub_shift, + const struct dcn_hubbub_mask *hubbub_mask); + +#endif /* DAL_DC_DCN21_DCN21_HUBBUB_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c new file mode 100644 index 000000000000..cf09b9335728 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c @@ -0,0 +1,960 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dcn10/dcn10_hubp.h" +#include "dcn21_hubp.h" + +#include "dm_services.h" +#include "reg_helper.h" + +#include "dc_dmub_srv.h" + +#define DC_LOGGER_INIT(logger) + +#define REG(reg)\ + hubp21->hubp_regs->reg + +#define CTX \ + hubp21->base.ctx + +#undef FN +#define FN(reg_name, field_name) \ + hubp21->hubp_shift->field_name, hubp21->hubp_mask->field_name + +/* + * In DCN2.1, the non-double buffered version of the following 4 DLG registers are used in RTL. + * As a result, if S/W updates any of these registers during a mode change, + * the current frame before the mode change will use the new value right away + * and can lead to generating incorrect request deadlines and incorrect TTU/QoS behavior. + * + * REFCYC_PER_VM_GROUP_FLIP[22:0] + * REFCYC_PER_VM_GROUP_VBLANK[22:0] + * REFCYC_PER_VM_REQ_FLIP[22:0] + * REFCYC_PER_VM_REQ_VBLANK[22:0] + * + * REFCYC_PER_VM_*_FLIP affects the deadline of the VM requests generated + * when flipping to a new surface + * + * REFCYC_PER_VM_*_VBLANK affects the deadline of the VM requests generated + * during prefetch period of a frame. The prefetch starts at a pre-determined + * number of lines before the display active per frame + * + * DCN may underflow due to incorrectly programming these registers + * during VM stage of prefetch/iflip. First lines of display active + * or a sub-region of active using a new surface will be corrupted + * until the VM data returns at flip/mode change transitions + * + * Work around: + * workaround is always opt to use the more aggressive settings. + * On any mode switch, if the new reg values are smaller than the current values, + * then update the regs with the new values. + * + * Link to the ticket: http://ontrack-internal.amd.com/browse/DEDCN21-142 + * + */ +void apply_DEDCN21_142_wa_for_hostvm_deadline( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + uint32_t cur_value; + + REG_GET(VBLANK_PARAMETERS_5, REFCYC_PER_VM_GROUP_VBLANK, &cur_value); + if (cur_value > dlg_attr->refcyc_per_vm_group_vblank) + REG_SET(VBLANK_PARAMETERS_5, 0, + REFCYC_PER_VM_GROUP_VBLANK, dlg_attr->refcyc_per_vm_group_vblank); + + REG_GET(VBLANK_PARAMETERS_6, + REFCYC_PER_VM_REQ_VBLANK, + &cur_value); + if (cur_value > dlg_attr->refcyc_per_vm_req_vblank) + REG_SET(VBLANK_PARAMETERS_6, 0, + REFCYC_PER_VM_REQ_VBLANK, dlg_attr->refcyc_per_vm_req_vblank); + + REG_GET(FLIP_PARAMETERS_3, REFCYC_PER_VM_GROUP_FLIP, &cur_value); + if (cur_value > dlg_attr->refcyc_per_vm_group_flip) + REG_SET(FLIP_PARAMETERS_3, 0, + REFCYC_PER_VM_GROUP_FLIP, dlg_attr->refcyc_per_vm_group_flip); + + REG_GET(FLIP_PARAMETERS_4, REFCYC_PER_VM_REQ_FLIP, &cur_value); + if (cur_value > dlg_attr->refcyc_per_vm_req_flip) + REG_SET(FLIP_PARAMETERS_4, 0, + REFCYC_PER_VM_REQ_FLIP, dlg_attr->refcyc_per_vm_req_flip); + + REG_SET(FLIP_PARAMETERS_5, 0, + REFCYC_PER_PTE_GROUP_FLIP_C, dlg_attr->refcyc_per_pte_group_flip_c); + REG_SET(FLIP_PARAMETERS_6, 0, + REFCYC_PER_META_CHUNK_FLIP_C, dlg_attr->refcyc_per_meta_chunk_flip_c); +} + +void hubp21_program_deadline( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *ttu_attr) +{ + hubp2_program_deadline(hubp, dlg_attr, ttu_attr); + + apply_DEDCN21_142_wa_for_hostvm_deadline(hubp, dlg_attr); +} + +void hubp21_program_requestor( + struct hubp *hubp, + struct _vcs_dpi_display_rq_regs_st *rq_regs) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + + REG_UPDATE(HUBPRET_CONTROL, + DET_BUF_PLANE1_BASE_ADDRESS, rq_regs->plane1_base_address); + REG_SET_4(DCN_EXPANSION_MODE, 0, + DRQ_EXPANSION_MODE, rq_regs->drq_expansion_mode, + PRQ_EXPANSION_MODE, rq_regs->prq_expansion_mode, + MRQ_EXPANSION_MODE, rq_regs->mrq_expansion_mode, + CRQ_EXPANSION_MODE, rq_regs->crq_expansion_mode); + REG_SET_8(DCHUBP_REQ_SIZE_CONFIG, 0, + CHUNK_SIZE, rq_regs->rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, rq_regs->rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, rq_regs->rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, rq_regs->rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, rq_regs->rq_regs_l.dpte_group_size, + VM_GROUP_SIZE, rq_regs->rq_regs_l.mpte_group_size, + SWATH_HEIGHT, rq_regs->rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, rq_regs->rq_regs_l.pte_row_height_linear); + REG_SET_7(DCHUBP_REQ_SIZE_CONFIG_C, 0, + CHUNK_SIZE_C, rq_regs->rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, rq_regs->rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, rq_regs->rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, rq_regs->rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, rq_regs->rq_regs_c.dpte_group_size, + SWATH_HEIGHT_C, rq_regs->rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, rq_regs->rq_regs_c.pte_row_height_linear); +} + +static void hubp21_setup( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *ttu_attr, + struct _vcs_dpi_display_rq_regs_st *rq_regs, + struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest) +{ + /* otg is locked when this func is called. Register are double buffered. + * disable the requestors is not needed + */ + + hubp2_vready_at_or_After_vsync(hubp, pipe_dest); + hubp21_program_requestor(hubp, rq_regs); + hubp21_program_deadline(hubp, dlg_attr, ttu_attr); + +} + +void hubp21_set_viewport( + struct hubp *hubp, + const struct rect *viewport, + const struct rect *viewport_c) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + + REG_SET_2(DCSURF_PRI_VIEWPORT_DIMENSION, 0, + PRI_VIEWPORT_WIDTH, viewport->width, + PRI_VIEWPORT_HEIGHT, viewport->height); + + REG_SET_2(DCSURF_PRI_VIEWPORT_START, 0, + PRI_VIEWPORT_X_START, viewport->x, + PRI_VIEWPORT_Y_START, viewport->y); + + /*for stereo*/ + REG_SET_2(DCSURF_SEC_VIEWPORT_DIMENSION, 0, + SEC_VIEWPORT_WIDTH, viewport->width, + SEC_VIEWPORT_HEIGHT, viewport->height); + + REG_SET_2(DCSURF_SEC_VIEWPORT_START, 0, + SEC_VIEWPORT_X_START, viewport->x, + SEC_VIEWPORT_Y_START, viewport->y); + + /* DC supports NV12 only at the moment */ + REG_SET_2(DCSURF_PRI_VIEWPORT_DIMENSION_C, 0, + PRI_VIEWPORT_WIDTH_C, viewport_c->width, + PRI_VIEWPORT_HEIGHT_C, viewport_c->height); + + REG_SET_2(DCSURF_PRI_VIEWPORT_START_C, 0, + PRI_VIEWPORT_X_START_C, viewport_c->x, + PRI_VIEWPORT_Y_START_C, viewport_c->y); + + REG_SET_2(DCSURF_SEC_VIEWPORT_DIMENSION_C, 0, + SEC_VIEWPORT_WIDTH_C, viewport_c->width, + SEC_VIEWPORT_HEIGHT_C, viewport_c->height); + + REG_SET_2(DCSURF_SEC_VIEWPORT_START_C, 0, + SEC_VIEWPORT_X_START_C, viewport_c->x, + SEC_VIEWPORT_Y_START_C, viewport_c->y); +} + +static void hubp21_apply_PLAT_54186_wa( + struct hubp *hubp, + const struct dc_plane_address *address) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + struct dc_debug_options *debug = &hubp->ctx->dc->debug; + unsigned int chroma_bpe = 2; + unsigned int luma_addr_high_part = 0; + unsigned int row_height = 0; + unsigned int chroma_pitch = 0; + unsigned int viewport_c_height = 0; + unsigned int viewport_c_width = 0; + unsigned int patched_viewport_height = 0; + unsigned int patched_viewport_width = 0; + unsigned int rotation_angle = 0; + unsigned int pix_format = 0; + unsigned int h_mirror_en = 0; + unsigned int tile_blk_size = 64 * 1024; /* 64KB for 64KB SW, 4KB for 4KB SW */ + + + if (!debug->nv12_iflip_vm_wa) + return; + + REG_GET(DCHUBP_REQ_SIZE_CONFIG_C, + PTE_ROW_HEIGHT_LINEAR_C, &row_height); + + REG_GET_2(DCSURF_PRI_VIEWPORT_DIMENSION_C, + PRI_VIEWPORT_WIDTH_C, &viewport_c_width, + PRI_VIEWPORT_HEIGHT_C, &viewport_c_height); + + REG_GET(DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C, + PRIMARY_SURFACE_ADDRESS_HIGH_C, &luma_addr_high_part); + + REG_GET(DCSURF_SURFACE_PITCH_C, + PITCH_C, &chroma_pitch); + + chroma_pitch += 1; + + REG_GET_3(DCSURF_SURFACE_CONFIG, + SURFACE_PIXEL_FORMAT, &pix_format, + ROTATION_ANGLE, &rotation_angle, + H_MIRROR_EN, &h_mirror_en); + + /* reset persistent cached data */ + hubp21->PLAT_54186_wa_chroma_addr_offset = 0; + /* apply wa only for NV12 surface with scatter gather enabled with viewport > 512 along + * the vertical direction*/ + if (address->type != PLN_ADDR_TYPE_VIDEO_PROGRESSIVE || + address->video_progressive.luma_addr.high_part == 0xf4) + return; + + if ((rotation_angle == ROTATION_ANGLE_0 || rotation_angle == ROTATION_ANGLE_180) + && viewport_c_height <= 512) + return; + + if ((rotation_angle == ROTATION_ANGLE_90 || rotation_angle == ROTATION_ANGLE_270) + && viewport_c_width <= 512) + return; + + switch (rotation_angle) { + case ROTATION_ANGLE_0: /* 0 degree rotation */ + row_height = 128; + patched_viewport_height = (viewport_c_height / row_height + 1) * row_height + 1; + patched_viewport_width = viewport_c_width; + hubp21->PLAT_54186_wa_chroma_addr_offset = 0; + break; + case ROTATION_ANGLE_180: /* 180 degree rotation */ + row_height = 128; + patched_viewport_height = viewport_c_height + row_height; + patched_viewport_width = viewport_c_width; + hubp21->PLAT_54186_wa_chroma_addr_offset = 0 - chroma_pitch * row_height * chroma_bpe; + break; + case ROTATION_ANGLE_90: /* 90 degree rotation */ + row_height = 256; + if (h_mirror_en) { + patched_viewport_height = viewport_c_height; + patched_viewport_width = viewport_c_width + row_height; + hubp21->PLAT_54186_wa_chroma_addr_offset = 0; + } else { + patched_viewport_height = viewport_c_height; + patched_viewport_width = viewport_c_width + row_height; + hubp21->PLAT_54186_wa_chroma_addr_offset = 0 - tile_blk_size; + } + break; + case ROTATION_ANGLE_270: /* 270 degree rotation */ + row_height = 256; + if (h_mirror_en) { + patched_viewport_height = viewport_c_height; + patched_viewport_width = viewport_c_width + row_height; + hubp21->PLAT_54186_wa_chroma_addr_offset = 0 - tile_blk_size; + } else { + patched_viewport_height = viewport_c_height; + patched_viewport_width = viewport_c_width + row_height; + hubp21->PLAT_54186_wa_chroma_addr_offset = 0; + } + break; + default: + ASSERT(0); + break; + } + + /* catch cases where viewport keep growing */ + ASSERT(patched_viewport_height && patched_viewport_height < 5000); + ASSERT(patched_viewport_width && patched_viewport_width < 5000); + + REG_UPDATE_2(DCSURF_PRI_VIEWPORT_DIMENSION_C, + PRI_VIEWPORT_WIDTH_C, patched_viewport_width, + PRI_VIEWPORT_HEIGHT_C, patched_viewport_height); +} + +void hubp21_set_vm_system_aperture_settings(struct hubp *hubp, + struct vm_system_aperture_param *apt) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + + PHYSICAL_ADDRESS_LOC mc_vm_apt_default; + PHYSICAL_ADDRESS_LOC mc_vm_apt_low; + PHYSICAL_ADDRESS_LOC mc_vm_apt_high; + + // The format of default addr is 48:12 of the 48 bit addr + mc_vm_apt_default.quad_part = apt->sys_default.quad_part >> 12; + + // The format of high/low are 48:18 of the 48 bit addr + mc_vm_apt_low.quad_part = apt->sys_low.quad_part >> 18; + mc_vm_apt_high.quad_part = apt->sys_high.quad_part >> 18; + + REG_SET(DCN_VM_SYSTEM_APERTURE_LOW_ADDR, 0, + MC_VM_SYSTEM_APERTURE_LOW_ADDR, mc_vm_apt_low.quad_part); + + REG_SET(DCN_VM_SYSTEM_APERTURE_HIGH_ADDR, 0, + MC_VM_SYSTEM_APERTURE_HIGH_ADDR, mc_vm_apt_high.quad_part); + + REG_SET_2(DCN_VM_MX_L1_TLB_CNTL, 0, + ENABLE_L1_TLB, 1, + SYSTEM_ACCESS_MODE, 0x3); +} + +void hubp21_validate_dml_output(struct hubp *hubp, + struct dc_context *ctx, + struct _vcs_dpi_display_rq_regs_st *dml_rq_regs, + struct _vcs_dpi_display_dlg_regs_st *dml_dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *dml_ttu_attr) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + struct _vcs_dpi_display_rq_regs_st rq_regs = {0}; + struct _vcs_dpi_display_dlg_regs_st dlg_attr = {0}; + struct _vcs_dpi_display_ttu_regs_st ttu_attr = {0}; + DC_LOGGER_INIT(ctx->logger); + DC_LOG_DEBUG("DML Validation | Running Validation"); + + /* Requester - Per hubp */ + REG_GET(HUBPRET_CONTROL, + DET_BUF_PLANE1_BASE_ADDRESS, &rq_regs.plane1_base_address); + REG_GET_4(DCN_EXPANSION_MODE, + DRQ_EXPANSION_MODE, &rq_regs.drq_expansion_mode, + PRQ_EXPANSION_MODE, &rq_regs.prq_expansion_mode, + MRQ_EXPANSION_MODE, &rq_regs.mrq_expansion_mode, + CRQ_EXPANSION_MODE, &rq_regs.crq_expansion_mode); + REG_GET_8(DCHUBP_REQ_SIZE_CONFIG, + CHUNK_SIZE, &rq_regs.rq_regs_l.chunk_size, + MIN_CHUNK_SIZE, &rq_regs.rq_regs_l.min_chunk_size, + META_CHUNK_SIZE, &rq_regs.rq_regs_l.meta_chunk_size, + MIN_META_CHUNK_SIZE, &rq_regs.rq_regs_l.min_meta_chunk_size, + DPTE_GROUP_SIZE, &rq_regs.rq_regs_l.dpte_group_size, + VM_GROUP_SIZE, &rq_regs.rq_regs_l.mpte_group_size, + SWATH_HEIGHT, &rq_regs.rq_regs_l.swath_height, + PTE_ROW_HEIGHT_LINEAR, &rq_regs.rq_regs_l.pte_row_height_linear); + REG_GET_7(DCHUBP_REQ_SIZE_CONFIG_C, + CHUNK_SIZE_C, &rq_regs.rq_regs_c.chunk_size, + MIN_CHUNK_SIZE_C, &rq_regs.rq_regs_c.min_chunk_size, + META_CHUNK_SIZE_C, &rq_regs.rq_regs_c.meta_chunk_size, + MIN_META_CHUNK_SIZE_C, &rq_regs.rq_regs_c.min_meta_chunk_size, + DPTE_GROUP_SIZE_C, &rq_regs.rq_regs_c.dpte_group_size, + SWATH_HEIGHT_C, &rq_regs.rq_regs_c.swath_height, + PTE_ROW_HEIGHT_LINEAR_C, &rq_regs.rq_regs_c.pte_row_height_linear); + + if (rq_regs.plane1_base_address != dml_rq_regs->plane1_base_address) + DC_LOG_DEBUG("DML Validation | HUBPRET_CONTROL:DET_BUF_PLANE1_BASE_ADDRESS - Expected: %u Actual: %u\n", + dml_rq_regs->plane1_base_address, rq_regs.plane1_base_address); + if (rq_regs.drq_expansion_mode != dml_rq_regs->drq_expansion_mode) + DC_LOG_DEBUG("DML Validation | DCN_EXPANSION_MODE:DRQ_EXPANSION_MODE - Expected: %u Actual: %u\n", + dml_rq_regs->drq_expansion_mode, rq_regs.drq_expansion_mode); + if (rq_regs.prq_expansion_mode != dml_rq_regs->prq_expansion_mode) + DC_LOG_DEBUG("DML Validation | DCN_EXPANSION_MODE:MRQ_EXPANSION_MODE - Expected: %u Actual: %u\n", + dml_rq_regs->prq_expansion_mode, rq_regs.prq_expansion_mode); + if (rq_regs.mrq_expansion_mode != dml_rq_regs->mrq_expansion_mode) + DC_LOG_DEBUG("DML Validation | DCN_EXPANSION_MODE:DET_BUF_PLANE1_BASE_ADDRESS - Expected: %u Actual: %u\n", + dml_rq_regs->mrq_expansion_mode, rq_regs.mrq_expansion_mode); + if (rq_regs.crq_expansion_mode != dml_rq_regs->crq_expansion_mode) + DC_LOG_DEBUG("DML Validation | DCN_EXPANSION_MODE:CRQ_EXPANSION_MODE - Expected: %u Actual: %u\n", + dml_rq_regs->crq_expansion_mode, rq_regs.crq_expansion_mode); + + if (rq_regs.rq_regs_l.chunk_size != dml_rq_regs->rq_regs_l.chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:CHUNK_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.chunk_size, rq_regs.rq_regs_l.chunk_size); + if (rq_regs.rq_regs_l.min_chunk_size != dml_rq_regs->rq_regs_l.min_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:MIN_CHUNK_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.min_chunk_size, rq_regs.rq_regs_l.min_chunk_size); + if (rq_regs.rq_regs_l.meta_chunk_size != dml_rq_regs->rq_regs_l.meta_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:META_CHUNK_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.meta_chunk_size, rq_regs.rq_regs_l.meta_chunk_size); + if (rq_regs.rq_regs_l.min_meta_chunk_size != dml_rq_regs->rq_regs_l.min_meta_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:MIN_META_CHUNK_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.min_meta_chunk_size, rq_regs.rq_regs_l.min_meta_chunk_size); + if (rq_regs.rq_regs_l.dpte_group_size != dml_rq_regs->rq_regs_l.dpte_group_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:DPTE_GROUP_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.dpte_group_size, rq_regs.rq_regs_l.dpte_group_size); + if (rq_regs.rq_regs_l.mpte_group_size != dml_rq_regs->rq_regs_l.mpte_group_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:VM_GROUP_SIZE - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.mpte_group_size, rq_regs.rq_regs_l.mpte_group_size); + if (rq_regs.rq_regs_l.swath_height != dml_rq_regs->rq_regs_l.swath_height) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:SWATH_HEIGHT - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.swath_height, rq_regs.rq_regs_l.swath_height); + if (rq_regs.rq_regs_l.pte_row_height_linear != dml_rq_regs->rq_regs_l.pte_row_height_linear) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG_C:PTE_ROW_HEIGHT_LINEAR - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_l.pte_row_height_linear, rq_regs.rq_regs_l.pte_row_height_linear); + + if (rq_regs.rq_regs_c.chunk_size != dml_rq_regs->rq_regs_c.chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:CHUNK_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.chunk_size, rq_regs.rq_regs_c.chunk_size); + if (rq_regs.rq_regs_c.min_chunk_size != dml_rq_regs->rq_regs_c.min_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:MIN_CHUNK_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.min_chunk_size, rq_regs.rq_regs_c.min_chunk_size); + if (rq_regs.rq_regs_c.meta_chunk_size != dml_rq_regs->rq_regs_c.meta_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:META_CHUNK_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.meta_chunk_size, rq_regs.rq_regs_c.meta_chunk_size); + if (rq_regs.rq_regs_c.min_meta_chunk_size != dml_rq_regs->rq_regs_c.min_meta_chunk_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:MIN_META_CHUNK_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.min_meta_chunk_size, rq_regs.rq_regs_c.min_meta_chunk_size); + if (rq_regs.rq_regs_c.dpte_group_size != dml_rq_regs->rq_regs_c.dpte_group_size) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:DPTE_GROUP_SIZE_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.dpte_group_size, rq_regs.rq_regs_c.dpte_group_size); + if (rq_regs.rq_regs_c.swath_height != dml_rq_regs->rq_regs_c.swath_height) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:SWATH_HEIGHT_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.swath_height, rq_regs.rq_regs_c.swath_height); + if (rq_regs.rq_regs_c.pte_row_height_linear != dml_rq_regs->rq_regs_c.pte_row_height_linear) + DC_LOG_DEBUG("DML Validation | DCHUBP_REQ_SIZE_CONFIG:PTE_ROW_HEIGHT_LINEAR_C - Expected: %u Actual: %u\n", + dml_rq_regs->rq_regs_c.pte_row_height_linear, rq_regs.rq_regs_c.pte_row_height_linear); + + + /* DLG - Per hubp */ + REG_GET_2(BLANK_OFFSET_0, + REFCYC_H_BLANK_END, &dlg_attr.refcyc_h_blank_end, + DLG_V_BLANK_END, &dlg_attr.dlg_vblank_end); + REG_GET(BLANK_OFFSET_1, + MIN_DST_Y_NEXT_START, &dlg_attr.min_dst_y_next_start); + REG_GET(DST_DIMENSIONS, + REFCYC_PER_HTOTAL, &dlg_attr.refcyc_per_htotal); + REG_GET_2(DST_AFTER_SCALER, + REFCYC_X_AFTER_SCALER, &dlg_attr.refcyc_x_after_scaler, + DST_Y_AFTER_SCALER, &dlg_attr.dst_y_after_scaler); + REG_GET(REF_FREQ_TO_PIX_FREQ, + REF_FREQ_TO_PIX_FREQ, &dlg_attr.ref_freq_to_pix_freq); + + if (dlg_attr.refcyc_h_blank_end != dml_dlg_attr->refcyc_h_blank_end) + DC_LOG_DEBUG("DML Validation | BLANK_OFFSET_0:REFCYC_H_BLANK_END - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_h_blank_end, dlg_attr.refcyc_h_blank_end); + if (dlg_attr.dlg_vblank_end != dml_dlg_attr->dlg_vblank_end) + DC_LOG_DEBUG("DML Validation | BLANK_OFFSET_0:DLG_V_BLANK_END - Expected: %u Actual: %u\n", + dml_dlg_attr->dlg_vblank_end, dlg_attr.dlg_vblank_end); + if (dlg_attr.min_dst_y_next_start != dml_dlg_attr->min_dst_y_next_start) + DC_LOG_DEBUG("DML Validation | BLANK_OFFSET_1:MIN_DST_Y_NEXT_START - Expected: %u Actual: %u\n", + dml_dlg_attr->min_dst_y_next_start, dlg_attr.min_dst_y_next_start); + if (dlg_attr.refcyc_per_htotal != dml_dlg_attr->refcyc_per_htotal) + DC_LOG_DEBUG("DML Validation | DST_DIMENSIONS:REFCYC_PER_HTOTAL - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_htotal, dlg_attr.refcyc_per_htotal); + if (dlg_attr.refcyc_x_after_scaler != dml_dlg_attr->refcyc_x_after_scaler) + DC_LOG_DEBUG("DML Validation | DST_AFTER_SCALER:REFCYC_X_AFTER_SCALER - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_x_after_scaler, dlg_attr.refcyc_x_after_scaler); + if (dlg_attr.dst_y_after_scaler != dml_dlg_attr->dst_y_after_scaler) + DC_LOG_DEBUG("DML Validation | DST_AFTER_SCALER:DST_Y_AFTER_SCALER - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_after_scaler, dlg_attr.dst_y_after_scaler); + if (dlg_attr.ref_freq_to_pix_freq != dml_dlg_attr->ref_freq_to_pix_freq) + DC_LOG_DEBUG("DML Validation | REF_FREQ_TO_PIX_FREQ:REF_FREQ_TO_PIX_FREQ - Expected: %u Actual: %u\n", + dml_dlg_attr->ref_freq_to_pix_freq, dlg_attr.ref_freq_to_pix_freq); + + /* DLG - Per luma/chroma */ + REG_GET(VBLANK_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_VBLANK_L, &dlg_attr.refcyc_per_pte_group_vblank_l); + if (REG(NOM_PARAMETERS_0)) + REG_GET(NOM_PARAMETERS_0, + DST_Y_PER_PTE_ROW_NOM_L, &dlg_attr.dst_y_per_pte_row_nom_l); + if (REG(NOM_PARAMETERS_1)) + REG_GET(NOM_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_NOM_L, &dlg_attr.refcyc_per_pte_group_nom_l); + REG_GET(NOM_PARAMETERS_4, + DST_Y_PER_META_ROW_NOM_L, &dlg_attr.dst_y_per_meta_row_nom_l); + REG_GET(NOM_PARAMETERS_5, + REFCYC_PER_META_CHUNK_NOM_L, &dlg_attr.refcyc_per_meta_chunk_nom_l); + REG_GET_2(PER_LINE_DELIVERY, + REFCYC_PER_LINE_DELIVERY_L, &dlg_attr.refcyc_per_line_delivery_l, + REFCYC_PER_LINE_DELIVERY_C, &dlg_attr.refcyc_per_line_delivery_c); + REG_GET_2(PER_LINE_DELIVERY_PRE, + REFCYC_PER_LINE_DELIVERY_PRE_L, &dlg_attr.refcyc_per_line_delivery_pre_l, + REFCYC_PER_LINE_DELIVERY_PRE_C, &dlg_attr.refcyc_per_line_delivery_pre_c); + REG_GET(VBLANK_PARAMETERS_2, + REFCYC_PER_PTE_GROUP_VBLANK_C, &dlg_attr.refcyc_per_pte_group_vblank_c); + if (REG(NOM_PARAMETERS_2)) + REG_GET(NOM_PARAMETERS_2, + DST_Y_PER_PTE_ROW_NOM_C, &dlg_attr.dst_y_per_pte_row_nom_c); + if (REG(NOM_PARAMETERS_3)) + REG_GET(NOM_PARAMETERS_3, + REFCYC_PER_PTE_GROUP_NOM_C, &dlg_attr.refcyc_per_pte_group_nom_c); + REG_GET(NOM_PARAMETERS_6, + DST_Y_PER_META_ROW_NOM_C, &dlg_attr.dst_y_per_meta_row_nom_c); + REG_GET(NOM_PARAMETERS_7, + REFCYC_PER_META_CHUNK_NOM_C, &dlg_attr.refcyc_per_meta_chunk_nom_c); + REG_GET(VBLANK_PARAMETERS_3, + REFCYC_PER_META_CHUNK_VBLANK_L, &dlg_attr.refcyc_per_meta_chunk_vblank_l); + REG_GET(VBLANK_PARAMETERS_4, + REFCYC_PER_META_CHUNK_VBLANK_C, &dlg_attr.refcyc_per_meta_chunk_vblank_c); + + if (dlg_attr.refcyc_per_pte_group_vblank_l != dml_dlg_attr->refcyc_per_pte_group_vblank_l) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_1:REFCYC_PER_PTE_GROUP_VBLANK_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_vblank_l, dlg_attr.refcyc_per_pte_group_vblank_l); + if (dlg_attr.dst_y_per_pte_row_nom_l != dml_dlg_attr->dst_y_per_pte_row_nom_l) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_0:DST_Y_PER_PTE_ROW_NOM_L - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_per_pte_row_nom_l, dlg_attr.dst_y_per_pte_row_nom_l); + if (dlg_attr.refcyc_per_pte_group_nom_l != dml_dlg_attr->refcyc_per_pte_group_nom_l) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_1:REFCYC_PER_PTE_GROUP_NOM_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_nom_l, dlg_attr.refcyc_per_pte_group_nom_l); + if (dlg_attr.dst_y_per_meta_row_nom_l != dml_dlg_attr->dst_y_per_meta_row_nom_l) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_4:DST_Y_PER_META_ROW_NOM_L - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_per_meta_row_nom_l, dlg_attr.dst_y_per_meta_row_nom_l); + if (dlg_attr.refcyc_per_meta_chunk_nom_l != dml_dlg_attr->refcyc_per_meta_chunk_nom_l) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_5:REFCYC_PER_META_CHUNK_NOM_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_nom_l, dlg_attr.refcyc_per_meta_chunk_nom_l); + if (dlg_attr.refcyc_per_line_delivery_l != dml_dlg_attr->refcyc_per_line_delivery_l) + DC_LOG_DEBUG("DML Validation | PER_LINE_DELIVERY:REFCYC_PER_LINE_DELIVERY_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_line_delivery_l, dlg_attr.refcyc_per_line_delivery_l); + if (dlg_attr.refcyc_per_line_delivery_c != dml_dlg_attr->refcyc_per_line_delivery_c) + DC_LOG_DEBUG("DML Validation | PER_LINE_DELIVERY:REFCYC_PER_LINE_DELIVERY_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_line_delivery_c, dlg_attr.refcyc_per_line_delivery_c); + if (dlg_attr.refcyc_per_pte_group_vblank_c != dml_dlg_attr->refcyc_per_pte_group_vblank_c) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_2:REFCYC_PER_PTE_GROUP_VBLANK_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_vblank_c, dlg_attr.refcyc_per_pte_group_vblank_c); + if (dlg_attr.dst_y_per_pte_row_nom_c != dml_dlg_attr->dst_y_per_pte_row_nom_c) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_2:DST_Y_PER_PTE_ROW_NOM_C - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_per_pte_row_nom_c, dlg_attr.dst_y_per_pte_row_nom_c); + if (dlg_attr.refcyc_per_pte_group_nom_c != dml_dlg_attr->refcyc_per_pte_group_nom_c) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_3:REFCYC_PER_PTE_GROUP_NOM_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_nom_c, dlg_attr.refcyc_per_pte_group_nom_c); + if (dlg_attr.dst_y_per_meta_row_nom_c != dml_dlg_attr->dst_y_per_meta_row_nom_c) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_6:DST_Y_PER_META_ROW_NOM_C - Expected: %u Actual: %u\n", + dml_dlg_attr->dst_y_per_meta_row_nom_c, dlg_attr.dst_y_per_meta_row_nom_c); + if (dlg_attr.refcyc_per_meta_chunk_nom_c != dml_dlg_attr->refcyc_per_meta_chunk_nom_c) + DC_LOG_DEBUG("DML Validation | NOM_PARAMETERS_7:REFCYC_PER_META_CHUNK_NOM_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_nom_c, dlg_attr.refcyc_per_meta_chunk_nom_c); + if (dlg_attr.refcyc_per_line_delivery_pre_l != dml_dlg_attr->refcyc_per_line_delivery_pre_l) + DC_LOG_DEBUG("DML Validation | PER_LINE_DELIVERY_PRE:REFCYC_PER_LINE_DELIVERY_PRE_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_line_delivery_pre_l, dlg_attr.refcyc_per_line_delivery_pre_l); + if (dlg_attr.refcyc_per_line_delivery_pre_c != dml_dlg_attr->refcyc_per_line_delivery_pre_c) + DC_LOG_DEBUG("DML Validation | PER_LINE_DELIVERY_PRE:REFCYC_PER_LINE_DELIVERY_PRE_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_line_delivery_pre_c, dlg_attr.refcyc_per_line_delivery_pre_c); + if (dlg_attr.refcyc_per_meta_chunk_vblank_l != dml_dlg_attr->refcyc_per_meta_chunk_vblank_l) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_3:REFCYC_PER_META_CHUNK_VBLANK_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_vblank_l, dlg_attr.refcyc_per_meta_chunk_vblank_l); + if (dlg_attr.refcyc_per_meta_chunk_vblank_c != dml_dlg_attr->refcyc_per_meta_chunk_vblank_c) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_4:REFCYC_PER_META_CHUNK_VBLANK_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_vblank_c, dlg_attr.refcyc_per_meta_chunk_vblank_c); + + /* TTU - per hubp */ + REG_GET_2(DCN_TTU_QOS_WM, + QoS_LEVEL_LOW_WM, &ttu_attr.qos_level_low_wm, + QoS_LEVEL_HIGH_WM, &ttu_attr.qos_level_high_wm); + + if (ttu_attr.qos_level_low_wm != dml_ttu_attr->qos_level_low_wm) + DC_LOG_DEBUG("DML Validation | DCN_TTU_QOS_WM:QoS_LEVEL_LOW_WM - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_low_wm, ttu_attr.qos_level_low_wm); + if (ttu_attr.qos_level_high_wm != dml_ttu_attr->qos_level_high_wm) + DC_LOG_DEBUG("DML Validation | DCN_TTU_QOS_WM:QoS_LEVEL_HIGH_WM - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_high_wm, ttu_attr.qos_level_high_wm); + + /* TTU - per luma/chroma */ + /* Assumed surf0 is luma and 1 is chroma */ + REG_GET_3(DCN_SURF0_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr.refcyc_per_req_delivery_l, + QoS_LEVEL_FIXED, &ttu_attr.qos_level_fixed_l, + QoS_RAMP_DISABLE, &ttu_attr.qos_ramp_disable_l); + REG_GET_3(DCN_SURF1_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr.refcyc_per_req_delivery_c, + QoS_LEVEL_FIXED, &ttu_attr.qos_level_fixed_c, + QoS_RAMP_DISABLE, &ttu_attr.qos_ramp_disable_c); + REG_GET_3(DCN_CUR0_TTU_CNTL0, + REFCYC_PER_REQ_DELIVERY, &ttu_attr.refcyc_per_req_delivery_cur0, + QoS_LEVEL_FIXED, &ttu_attr.qos_level_fixed_cur0, + QoS_RAMP_DISABLE, &ttu_attr.qos_ramp_disable_cur0); + REG_GET(FLIP_PARAMETERS_1, + REFCYC_PER_PTE_GROUP_FLIP_L, &dlg_attr.refcyc_per_pte_group_flip_l); + REG_GET(DCN_CUR0_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, &ttu_attr.refcyc_per_req_delivery_pre_cur0); + REG_GET(DCN_CUR1_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, &ttu_attr.refcyc_per_req_delivery_pre_cur1); + REG_GET(DCN_SURF0_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, &ttu_attr.refcyc_per_req_delivery_pre_l); + REG_GET(DCN_SURF1_TTU_CNTL1, + REFCYC_PER_REQ_DELIVERY_PRE, &ttu_attr.refcyc_per_req_delivery_pre_c); + + if (ttu_attr.refcyc_per_req_delivery_l != dml_ttu_attr->refcyc_per_req_delivery_l) + DC_LOG_DEBUG("DML Validation | DCN_SURF0_TTU_CNTL0:REFCYC_PER_REQ_DELIVERY - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_l, ttu_attr.refcyc_per_req_delivery_l); + if (ttu_attr.qos_level_fixed_l != dml_ttu_attr->qos_level_fixed_l) + DC_LOG_DEBUG("DML Validation | DCN_SURF0_TTU_CNTL0:QoS_LEVEL_FIXED - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_fixed_l, ttu_attr.qos_level_fixed_l); + if (ttu_attr.qos_ramp_disable_l != dml_ttu_attr->qos_ramp_disable_l) + DC_LOG_DEBUG("DML Validation | DCN_SURF0_TTU_CNTL0:QoS_RAMP_DISABLE - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_ramp_disable_l, ttu_attr.qos_ramp_disable_l); + if (ttu_attr.refcyc_per_req_delivery_c != dml_ttu_attr->refcyc_per_req_delivery_c) + DC_LOG_DEBUG("DML Validation | DCN_SURF1_TTU_CNTL0:REFCYC_PER_REQ_DELIVERY - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_c, ttu_attr.refcyc_per_req_delivery_c); + if (ttu_attr.qos_level_fixed_c != dml_ttu_attr->qos_level_fixed_c) + DC_LOG_DEBUG("DML Validation | DCN_SURF1_TTU_CNTL0:QoS_LEVEL_FIXED - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_fixed_c, ttu_attr.qos_level_fixed_c); + if (ttu_attr.qos_ramp_disable_c != dml_ttu_attr->qos_ramp_disable_c) + DC_LOG_DEBUG("DML Validation | DCN_SURF1_TTU_CNTL0:QoS_RAMP_DISABLE - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_ramp_disable_c, ttu_attr.qos_ramp_disable_c); + if (ttu_attr.refcyc_per_req_delivery_cur0 != dml_ttu_attr->refcyc_per_req_delivery_cur0) + DC_LOG_DEBUG("DML Validation | DCN_CUR0_TTU_CNTL0:REFCYC_PER_REQ_DELIVERY - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_cur0, ttu_attr.refcyc_per_req_delivery_cur0); + if (ttu_attr.qos_level_fixed_cur0 != dml_ttu_attr->qos_level_fixed_cur0) + DC_LOG_DEBUG("DML Validation | DCN_CUR0_TTU_CNTL0:QoS_LEVEL_FIXED - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_level_fixed_cur0, ttu_attr.qos_level_fixed_cur0); + if (ttu_attr.qos_ramp_disable_cur0 != dml_ttu_attr->qos_ramp_disable_cur0) + DC_LOG_DEBUG("DML Validation | DCN_CUR0_TTU_CNTL0:QoS_RAMP_DISABLE - Expected: %u Actual: %u\n", + dml_ttu_attr->qos_ramp_disable_cur0, ttu_attr.qos_ramp_disable_cur0); + if (dlg_attr.refcyc_per_pte_group_flip_l != dml_dlg_attr->refcyc_per_pte_group_flip_l) + DC_LOG_DEBUG("DML Validation | FLIP_PARAMETERS_1:REFCYC_PER_PTE_GROUP_FLIP_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_flip_l, dlg_attr.refcyc_per_pte_group_flip_l); + if (ttu_attr.refcyc_per_req_delivery_pre_cur0 != dml_ttu_attr->refcyc_per_req_delivery_pre_cur0) + DC_LOG_DEBUG("DML Validation | DCN_CUR0_TTU_CNTL1:REFCYC_PER_REQ_DELIVERY_PRE - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_pre_cur0, ttu_attr.refcyc_per_req_delivery_pre_cur0); + if (ttu_attr.refcyc_per_req_delivery_pre_cur1 != dml_ttu_attr->refcyc_per_req_delivery_pre_cur1) + DC_LOG_DEBUG("DML Validation | DCN_CUR1_TTU_CNTL1:REFCYC_PER_REQ_DELIVERY_PRE - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_pre_cur1, ttu_attr.refcyc_per_req_delivery_pre_cur1); + if (ttu_attr.refcyc_per_req_delivery_pre_l != dml_ttu_attr->refcyc_per_req_delivery_pre_l) + DC_LOG_DEBUG("DML Validation | DCN_SURF0_TTU_CNTL1:REFCYC_PER_REQ_DELIVERY_PRE - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_pre_l, ttu_attr.refcyc_per_req_delivery_pre_l); + if (ttu_attr.refcyc_per_req_delivery_pre_c != dml_ttu_attr->refcyc_per_req_delivery_pre_c) + DC_LOG_DEBUG("DML Validation | DCN_SURF1_TTU_CNTL1:REFCYC_PER_REQ_DELIVERY_PRE - Expected: %u Actual: %u\n", + dml_ttu_attr->refcyc_per_req_delivery_pre_c, ttu_attr.refcyc_per_req_delivery_pre_c); + + /* Host VM deadline regs */ + REG_GET(VBLANK_PARAMETERS_5, + REFCYC_PER_VM_GROUP_VBLANK, &dlg_attr.refcyc_per_vm_group_vblank); + REG_GET(VBLANK_PARAMETERS_6, + REFCYC_PER_VM_REQ_VBLANK, &dlg_attr.refcyc_per_vm_req_vblank); + REG_GET(FLIP_PARAMETERS_3, + REFCYC_PER_VM_GROUP_FLIP, &dlg_attr.refcyc_per_vm_group_flip); + REG_GET(FLIP_PARAMETERS_4, + REFCYC_PER_VM_REQ_FLIP, &dlg_attr.refcyc_per_vm_req_flip); + REG_GET(FLIP_PARAMETERS_5, + REFCYC_PER_PTE_GROUP_FLIP_C, &dlg_attr.refcyc_per_pte_group_flip_c); + REG_GET(FLIP_PARAMETERS_6, + REFCYC_PER_META_CHUNK_FLIP_C, &dlg_attr.refcyc_per_meta_chunk_flip_c); + REG_GET(FLIP_PARAMETERS_2, + REFCYC_PER_META_CHUNK_FLIP_L, &dlg_attr.refcyc_per_meta_chunk_flip_l); + + if (dlg_attr.refcyc_per_vm_group_vblank != dml_dlg_attr->refcyc_per_vm_group_vblank) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_5:REFCYC_PER_VM_GROUP_VBLANK - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_vm_group_vblank, dlg_attr.refcyc_per_vm_group_vblank); + if (dlg_attr.refcyc_per_vm_req_vblank != dml_dlg_attr->refcyc_per_vm_req_vblank) + DC_LOG_DEBUG("DML Validation | VBLANK_PARAMETERS_6:REFCYC_PER_VM_REQ_VBLANK - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_vm_req_vblank, dlg_attr.refcyc_per_vm_req_vblank); + if (dlg_attr.refcyc_per_vm_group_flip != dml_dlg_attr->refcyc_per_vm_group_flip) + DC_LOG_DEBUG("DML Validation | FLIP_PARAMETERS_3:REFCYC_PER_VM_GROUP_FLIP - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_vm_group_flip, dlg_attr.refcyc_per_vm_group_flip); + if (dlg_attr.refcyc_per_vm_req_flip != dml_dlg_attr->refcyc_per_vm_req_flip) + DC_LOG_DEBUG("DML Validation | FLIP_PARAMETERS_4:REFCYC_PER_VM_REQ_FLIP - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_vm_req_flip, dlg_attr.refcyc_per_vm_req_flip); + if (dlg_attr.refcyc_per_pte_group_flip_c != dml_dlg_attr->refcyc_per_pte_group_flip_c) + DC_LOG_DEBUG("DML Validation | FLIP_PARAMETERS_5:REFCYC_PER_PTE_GROUP_FLIP_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_pte_group_flip_c, dlg_attr.refcyc_per_pte_group_flip_c); + if (dlg_attr.refcyc_per_meta_chunk_flip_c != dml_dlg_attr->refcyc_per_meta_chunk_flip_c) + DC_LOG_DEBUG("DML Validation | FLIP_PARAMETERS_6:REFCYC_PER_META_CHUNK_FLIP_C - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_flip_c, dlg_attr.refcyc_per_meta_chunk_flip_c); + if (dlg_attr.refcyc_per_meta_chunk_flip_l != dml_dlg_attr->refcyc_per_meta_chunk_flip_l) + DC_LOG_DEBUG("DML Validation | FLIP_PARAMETERS_2:REFCYC_PER_META_CHUNK_FLIP_L - Expected: %u Actual: %u\n", + dml_dlg_attr->refcyc_per_meta_chunk_flip_l, dlg_attr.refcyc_per_meta_chunk_flip_l); +} + +static void program_surface_flip_and_addr(struct hubp *hubp, struct surface_flip_registers *flip_regs) +{ + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + + REG_UPDATE_3(DCSURF_FLIP_CONTROL, + SURFACE_FLIP_TYPE, flip_regs->immediate, + SURFACE_FLIP_MODE_FOR_STEREOSYNC, flip_regs->grph_stereo, + SURFACE_FLIP_IN_STEREOSYNC, flip_regs->grph_stereo); + + REG_UPDATE(VMID_SETTINGS_0, + VMID, flip_regs->vmid); + + REG_UPDATE_8(DCSURF_SURFACE_CONTROL, + PRIMARY_SURFACE_TMZ, flip_regs->tmz_surface, + PRIMARY_SURFACE_TMZ_C, flip_regs->tmz_surface, + PRIMARY_META_SURFACE_TMZ, flip_regs->tmz_surface, + PRIMARY_META_SURFACE_TMZ_C, flip_regs->tmz_surface, + SECONDARY_SURFACE_TMZ, flip_regs->tmz_surface, + SECONDARY_SURFACE_TMZ_C, flip_regs->tmz_surface, + SECONDARY_META_SURFACE_TMZ, flip_regs->tmz_surface, + SECONDARY_META_SURFACE_TMZ_C, flip_regs->tmz_surface); + + REG_SET(DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH_C, 0, + PRIMARY_META_SURFACE_ADDRESS_HIGH_C, + flip_regs->DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH_C); + + REG_SET(DCSURF_PRIMARY_META_SURFACE_ADDRESS_C, 0, + PRIMARY_META_SURFACE_ADDRESS_C, + flip_regs->DCSURF_PRIMARY_META_SURFACE_ADDRESS_C); + + REG_SET(DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH, 0, + PRIMARY_META_SURFACE_ADDRESS_HIGH, + flip_regs->DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH); + + REG_SET(DCSURF_PRIMARY_META_SURFACE_ADDRESS, 0, + PRIMARY_META_SURFACE_ADDRESS, + flip_regs->DCSURF_PRIMARY_META_SURFACE_ADDRESS); + + REG_SET(DCSURF_SECONDARY_META_SURFACE_ADDRESS_HIGH, 0, + SECONDARY_META_SURFACE_ADDRESS_HIGH, + flip_regs->DCSURF_SECONDARY_META_SURFACE_ADDRESS_HIGH); + + REG_SET(DCSURF_SECONDARY_META_SURFACE_ADDRESS, 0, + SECONDARY_META_SURFACE_ADDRESS, + flip_regs->DCSURF_SECONDARY_META_SURFACE_ADDRESS); + + + REG_SET(DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH, 0, + SECONDARY_SURFACE_ADDRESS_HIGH, + flip_regs->DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH); + + REG_SET(DCSURF_SECONDARY_SURFACE_ADDRESS, 0, + SECONDARY_SURFACE_ADDRESS, + flip_regs->DCSURF_SECONDARY_SURFACE_ADDRESS); + + + REG_SET(DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C, 0, + PRIMARY_SURFACE_ADDRESS_HIGH_C, + flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C); + + REG_SET(DCSURF_PRIMARY_SURFACE_ADDRESS_C, 0, + PRIMARY_SURFACE_ADDRESS_C, + flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_C); + + REG_SET(DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH, 0, + PRIMARY_SURFACE_ADDRESS_HIGH, + flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH); + + REG_SET(DCSURF_PRIMARY_SURFACE_ADDRESS, 0, + PRIMARY_SURFACE_ADDRESS, + flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS); +} + +void dmcub_PLAT_54186_wa(struct hubp *hubp, struct surface_flip_registers *flip_regs) +{ + struct dc_dmub_srv *dmcub = hubp->ctx->dmub_srv; + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + struct dmub_rb_cmd_PLAT_54186_wa PLAT_54186_wa = { 0 }; + + PLAT_54186_wa.header.type = DMUB_CMD__PLAT_54186_WA; + PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS = flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS; + PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS_C = flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_C; + PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH = flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH; + PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C = flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C; + PLAT_54186_wa.flip.flip_params.grph_stereo = flip_regs->grph_stereo; + PLAT_54186_wa.flip.flip_params.hubp_inst = hubp->inst; + PLAT_54186_wa.flip.flip_params.immediate = flip_regs->immediate; + PLAT_54186_wa.flip.flip_params.tmz_surface = flip_regs->tmz_surface; + PLAT_54186_wa.flip.flip_params.vmid = flip_regs->vmid; + + PERF_TRACE(); // TODO: remove after performance is stable. + dc_dmub_srv_cmd_queue(dmcub, &PLAT_54186_wa.header); + PERF_TRACE(); // TODO: remove after performance is stable. + dc_dmub_srv_cmd_execute(dmcub); + PERF_TRACE(); // TODO: remove after performance is stable. + dc_dmub_srv_wait_idle(dmcub); + PERF_TRACE(); // TODO: remove after performance is stable. +} + +bool hubp21_program_surface_flip_and_addr( + struct hubp *hubp, + const struct dc_plane_address *address, + bool flip_immediate) +{ + struct dc_debug_options *debug = &hubp->ctx->dc->debug; + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + struct surface_flip_registers flip_regs = { 0 }; + + flip_regs.vmid = address->vmid; + + switch (address->type) { + case PLN_ADDR_TYPE_GRAPHICS: + if (address->grph.addr.quad_part == 0) { + BREAK_TO_DEBUGGER(); + break; + } + + if (address->grph.meta_addr.quad_part != 0) { + flip_regs.DCSURF_PRIMARY_META_SURFACE_ADDRESS = + address->grph.meta_addr.low_part; + flip_regs.DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH = + address->grph.meta_addr.high_part; + } + + flip_regs.DCSURF_PRIMARY_SURFACE_ADDRESS = + address->grph.addr.low_part; + flip_regs.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH = + address->grph.addr.high_part; + break; + case PLN_ADDR_TYPE_VIDEO_PROGRESSIVE: + if (address->video_progressive.luma_addr.quad_part == 0 + || address->video_progressive.chroma_addr.quad_part == 0) + break; + + if (address->video_progressive.luma_meta_addr.quad_part != 0) { + flip_regs.DCSURF_PRIMARY_META_SURFACE_ADDRESS = + address->video_progressive.luma_meta_addr.low_part; + flip_regs.DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH = + address->video_progressive.luma_meta_addr.high_part; + + flip_regs.DCSURF_PRIMARY_META_SURFACE_ADDRESS_C = + address->video_progressive.chroma_meta_addr.low_part; + flip_regs.DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH_C = + address->video_progressive.chroma_meta_addr.high_part; + } + + flip_regs.DCSURF_PRIMARY_SURFACE_ADDRESS = + address->video_progressive.luma_addr.low_part; + flip_regs.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH = + address->video_progressive.luma_addr.high_part; + + if (debug->nv12_iflip_vm_wa) { + flip_regs.DCSURF_PRIMARY_SURFACE_ADDRESS_C = + address->video_progressive.chroma_addr.low_part + hubp21->PLAT_54186_wa_chroma_addr_offset; + } else + flip_regs.DCSURF_PRIMARY_SURFACE_ADDRESS_C = + address->video_progressive.chroma_addr.low_part; + + flip_regs.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C = + address->video_progressive.chroma_addr.high_part; + + break; + case PLN_ADDR_TYPE_GRPH_STEREO: + if (address->grph_stereo.left_addr.quad_part == 0) + break; + if (address->grph_stereo.right_addr.quad_part == 0) + break; + + flip_regs.grph_stereo = true; + + if (address->grph_stereo.right_meta_addr.quad_part != 0) { + flip_regs.DCSURF_SECONDARY_META_SURFACE_ADDRESS = + address->grph_stereo.right_meta_addr.low_part; + flip_regs.DCSURF_SECONDARY_META_SURFACE_ADDRESS_HIGH = + address->grph_stereo.right_meta_addr.high_part; + } + + if (address->grph_stereo.left_meta_addr.quad_part != 0) { + flip_regs.DCSURF_PRIMARY_META_SURFACE_ADDRESS = + address->grph_stereo.left_meta_addr.low_part; + flip_regs.DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH = + address->grph_stereo.left_meta_addr.high_part; + } + + flip_regs.DCSURF_PRIMARY_SURFACE_ADDRESS = + address->grph_stereo.left_addr.low_part; + flip_regs.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH = + address->grph_stereo.left_addr.high_part; + + flip_regs.DCSURF_SECONDARY_SURFACE_ADDRESS = + address->grph_stereo.right_addr.low_part; + flip_regs.DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH = + address->grph_stereo.right_addr.high_part; + + break; + default: + BREAK_TO_DEBUGGER(); + break; + } + + flip_regs.tmz_surface = address->tmz_surface; + flip_regs.immediate = flip_immediate; + + if (hubp->ctx->dc->debug.enable_dmcub_surface_flip && address->type == PLN_ADDR_TYPE_VIDEO_PROGRESSIVE) + dmcub_PLAT_54186_wa(hubp, &flip_regs); + else + program_surface_flip_and_addr(hubp, &flip_regs); + + hubp->request_address = *address; + + return true; +} + +void hubp21_init(struct hubp *hubp) +{ + // DEDCN21-133: Inconsistent row starting line for flip between DPTE and Meta + // This is a chicken bit to enable the ECO fix. + + struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); + //hubp[i].HUBPREQ_DEBUG.HUBPREQ_DEBUG[26] = 1; + REG_WRITE(HUBPREQ_DEBUG, 1 << 26); +} +static struct hubp_funcs dcn21_hubp_funcs = { + .hubp_enable_tripleBuffer = hubp2_enable_triplebuffer, + .hubp_is_triplebuffer_enabled = hubp2_is_triplebuffer_enabled, + .hubp_program_surface_flip_and_addr = hubp21_program_surface_flip_and_addr, + .hubp_program_surface_config = hubp1_program_surface_config, + .hubp_is_flip_pending = hubp1_is_flip_pending, + .hubp_setup = hubp21_setup, + .hubp_setup_interdependent = hubp2_setup_interdependent, + .hubp_set_vm_system_aperture_settings = hubp21_set_vm_system_aperture_settings, + .set_blank = hubp1_set_blank, + .dcc_control = hubp1_dcc_control, + .mem_program_viewport = hubp21_set_viewport, + .apply_PLAT_54186_wa = hubp21_apply_PLAT_54186_wa, + .set_cursor_attributes = hubp2_cursor_set_attributes, + .set_cursor_position = hubp1_cursor_set_position, + .hubp_clk_cntl = hubp1_clk_cntl, + .hubp_vtg_sel = hubp1_vtg_sel, + .dmdata_set_attributes = hubp2_dmdata_set_attributes, + .dmdata_load = hubp2_dmdata_load, + .dmdata_status_done = hubp2_dmdata_status_done, + .hubp_read_state = hubp1_read_state, + .hubp_clear_underflow = hubp1_clear_underflow, + .hubp_set_flip_control_surface_gsl = hubp2_set_flip_control_surface_gsl, + .hubp_init = hubp21_init, + .validate_dml_output = hubp21_validate_dml_output, +}; + +bool hubp21_construct( + struct dcn21_hubp *hubp21, + struct dc_context *ctx, + uint32_t inst, + const struct dcn_hubp2_registers *hubp_regs, + const struct dcn_hubp2_shift *hubp_shift, + const struct dcn_hubp2_mask *hubp_mask) +{ + hubp21->base.funcs = &dcn21_hubp_funcs; + hubp21->base.ctx = ctx; + hubp21->hubp_regs = hubp_regs; + hubp21->hubp_shift = hubp_shift; + hubp21->hubp_mask = hubp_mask; + hubp21->base.inst = inst; + hubp21->base.opp_id = OPP_ID_INVALID; + hubp21->base.mpcc_id = 0xf; + + return true; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.h new file mode 100644 index 000000000000..9873b6cbc5ba --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.h @@ -0,0 +1,134 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef DAL_DC_DCN21_DCN21_HUBP_H_ +#define DAL_DC_DCN21_DCN21_HUBP_H_ + +#include "../dcn20/dcn20_hubp.h" +#include "../dcn10/dcn10_hubp.h" + +#define TO_DCN21_HUBP(hubp)\ + container_of(hubp, struct dcn21_hubp, base) + +#define HUBP_REG_LIST_DCN21(id)\ + HUBP_REG_LIST_DCN2_COMMON(id),\ + SRI(FLIP_PARAMETERS_3, HUBPREQ, id),\ + SRI(FLIP_PARAMETERS_4, HUBPREQ, id),\ + SRI(FLIP_PARAMETERS_5, HUBPREQ, id),\ + SRI(FLIP_PARAMETERS_6, HUBPREQ, id),\ + SRI(VBLANK_PARAMETERS_5, HUBPREQ, id),\ + SRI(VBLANK_PARAMETERS_6, HUBPREQ, id) + +#define HUBP_MASK_SH_LIST_DCN21_COMMON(mask_sh)\ + HUBP_MASK_SH_LIST_DCN_SHARE_COMMON(mask_sh),\ + HUBP_MASK_SH_LIST_DCN_VM(mask_sh),\ + HUBP_SF(HUBP0_DCSURF_SURFACE_CONFIG, ROTATION_ANGLE, mask_sh),\ + HUBP_SF(HUBP0_DCSURF_SURFACE_CONFIG, H_MIRROR_EN, mask_sh),\ + HUBP_SF(HUBPREQ0_PREFETCH_SETTINGS, DST_Y_PREFETCH, mask_sh),\ + HUBP_SF(HUBPREQ0_PREFETCH_SETTINGS, VRATIO_PREFETCH, mask_sh),\ + HUBP_SF(HUBPREQ0_PREFETCH_SETTINGS_C, VRATIO_PREFETCH_C, mask_sh),\ + HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_LOW_ADDR, MC_VM_SYSTEM_APERTURE_LOW_ADDR, mask_sh),\ + HUBP_SF(HUBPREQ0_DCN_VM_SYSTEM_APERTURE_HIGH_ADDR, MC_VM_SYSTEM_APERTURE_HIGH_ADDR, mask_sh),\ + HUBP_SF(HUBPREQ0_CURSOR_SETTINGS, CURSOR0_DST_Y_OFFSET, mask_sh), \ + HUBP_SF(HUBPREQ0_CURSOR_SETTINGS, CURSOR0_CHUNK_HDL_ADJUST, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_SURFACE_ADDRESS_HIGH, CURSOR_SURFACE_ADDRESS_HIGH, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_SURFACE_ADDRESS, CURSOR_SURFACE_ADDRESS, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_SIZE, CURSOR_WIDTH, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_SIZE, CURSOR_HEIGHT, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_MODE, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_2X_MAGNIFY, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_PITCH, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_LINES_PER_CHUNK, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_CONTROL, CURSOR_ENABLE, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_POSITION, CURSOR_X_POSITION, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_POSITION, CURSOR_Y_POSITION, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_HOT_SPOT, CURSOR_HOT_SPOT_X, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_HOT_SPOT, CURSOR_HOT_SPOT_Y, mask_sh), \ + HUBP_SF(CURSOR0_0_CURSOR_DST_OFFSET, CURSOR_DST_X_OFFSET, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_ADDRESS_HIGH, DMDATA_ADDRESS_HIGH, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_CNTL, DMDATA_MODE, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_CNTL, DMDATA_UPDATED, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_CNTL, DMDATA_REPEAT, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_CNTL, DMDATA_SIZE, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_SW_CNTL, DMDATA_SW_UPDATED, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_SW_CNTL, DMDATA_SW_REPEAT, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_SW_CNTL, DMDATA_SW_SIZE, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_QOS_CNTL, DMDATA_QOS_MODE, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_QOS_CNTL, DMDATA_QOS_LEVEL, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_QOS_CNTL, DMDATA_DL_DELTA, mask_sh), \ + HUBP_SF(CURSOR0_0_DMDATA_STATUS, DMDATA_DONE, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_0, DST_Y_PER_VM_FLIP, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_0, DST_Y_PER_ROW_FLIP, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_1, REFCYC_PER_PTE_GROUP_FLIP_L, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_2, REFCYC_PER_META_CHUNK_FLIP_L, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_VREADY_AT_OR_AFTER_VSYNC, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_DISABLE_STOP_DATA_DURING_VM, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_FLIP_CONTROL, HUBPREQ_MASTER_UPDATE_LOCK_STATUS, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_FLIP_CONTROL2, SURFACE_GSL_ENABLE, mask_sh),\ + HUBP_SF(HUBPREQ0_DCSURF_FLIP_CONTROL2, SURFACE_TRIPLE_BUFFER_ENABLE, mask_sh),\ + HUBP_SF(HUBPREQ0_VMID_SETTINGS_0, VMID, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_3, REFCYC_PER_VM_GROUP_FLIP, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_4, REFCYC_PER_VM_REQ_FLIP, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_5, REFCYC_PER_PTE_GROUP_FLIP_C, mask_sh),\ + HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_6, REFCYC_PER_META_CHUNK_FLIP_C, mask_sh),\ + HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_5, REFCYC_PER_VM_GROUP_VBLANK, mask_sh),\ + HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_6, REFCYC_PER_VM_REQ_VBLANK, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh) + +#define HUBP_MASK_SH_LIST_DCN21(mask_sh)\ + HUBP_MASK_SH_LIST_DCN21_COMMON(mask_sh),\ + HUBP_SF(HUBP0_DCSURF_TILING_CONFIG, RB_ALIGNED, mask_sh) + + +struct dcn21_hubp { + struct hubp base; + struct dcn_hubp_state state; + const struct dcn_hubp2_registers *hubp_regs; + const struct dcn_hubp2_shift *hubp_shift; + const struct dcn_hubp2_mask *hubp_mask; + int PLAT_54186_wa_chroma_addr_offset; +}; + +bool hubp21_construct( + struct dcn21_hubp *hubp21, + struct dc_context *ctx, + uint32_t inst, + const struct dcn_hubp2_registers *hubp_regs, + const struct dcn_hubp2_shift *hubp_shift, + const struct dcn_hubp2_mask *hubp_mask); + +void apply_DEDCN21_142_wa_for_hostvm_deadline( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr); + +void hubp21_program_deadline( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *ttu_attr); + +void hubp21_program_requestor( + struct hubp *hubp, + struct _vcs_dpi_display_rq_regs_st *rq_regs); +#endif /* DAL_DC_DCN21_DCN21_HUBP_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c new file mode 100644 index 000000000000..081ad8e43d58 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c @@ -0,0 +1,114 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dm_services.h" +#include "dm_helpers.h" +#include "core_types.h" +#include "resource.h" +#include "dce/dce_hwseq.h" +#include "dcn21_hwseq.h" +#include "vmid.h" +#include "reg_helper.h" +#include "hw/clk_mgr.h" + + +#define DC_LOGGER_INIT(logger) + +#define CTX \ + hws->ctx +#define REG(reg)\ + hws->regs->reg + +#undef FN +#define FN(reg_name, field_name) \ + hws->shifts->field_name, hws->masks->field_name + +/* Temporary read settings, future will get values from kmd directly */ +static void mmhub_update_page_table_config(struct dcn_hubbub_phys_addr_config *config, + struct dce_hwseq *hws) +{ + uint32_t page_table_base_hi; + uint32_t page_table_base_lo; + + REG_GET(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32, + PAGE_DIRECTORY_ENTRY_HI32, &page_table_base_hi); + REG_GET(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, + PAGE_DIRECTORY_ENTRY_LO32, &page_table_base_lo); + + config->gart_config.page_table_base_addr = ((uint64_t)page_table_base_hi << 32) | page_table_base_lo; + +} + +int dcn21_init_sys_ctx(struct dce_hwseq *hws, struct dc *dc, struct dc_phy_addr_space_config *pa_config) +{ + struct dcn_hubbub_phys_addr_config config; + + config.system_aperture.fb_top = pa_config->system_aperture.fb_top; + config.system_aperture.fb_offset = pa_config->system_aperture.fb_offset; + config.system_aperture.fb_base = pa_config->system_aperture.fb_base; + config.system_aperture.agp_top = pa_config->system_aperture.agp_top; + config.system_aperture.agp_bot = pa_config->system_aperture.agp_bot; + config.system_aperture.agp_base = pa_config->system_aperture.agp_base; + config.gart_config.page_table_start_addr = pa_config->gart_config.page_table_start_addr; + config.gart_config.page_table_end_addr = pa_config->gart_config.page_table_end_addr; + config.gart_config.page_table_base_addr = pa_config->gart_config.page_table_base_addr; + + mmhub_update_page_table_config(&config, hws); + + return dc->res_pool->hubbub->funcs->init_dchub_sys_ctx(dc->res_pool->hubbub, &config); +} + +// work around for Renoir s0i3, if register is programmed, bypass golden init. + +bool dcn21_s0i3_golden_init_wa(struct dc *dc) +{ + struct dce_hwseq *hws = dc->hwseq; + uint32_t value = 0; + + value = REG_READ(MICROSECOND_TIME_BASE_DIV); + + return value != 0x00120464; +} + +void dcn21_exit_optimized_pwr_state( + const struct dc *dc, + struct dc_state *context) +{ + dc->clk_mgr->funcs->update_clocks( + dc->clk_mgr, + context, + false); +} + +void dcn21_optimize_pwr_state( + const struct dc *dc, + struct dc_state *context) +{ + dc->clk_mgr->funcs->update_clocks( + dc->clk_mgr, + context, + true); +} + diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h new file mode 100644 index 000000000000..182736096123 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.h @@ -0,0 +1,47 @@ +/* +* Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_HWSS_DCN21_H__ +#define __DC_HWSS_DCN21_H__ + +#include "hw_sequencer_private.h" + +struct dc; + +int dcn21_init_sys_ctx(struct dce_hwseq *hws, + struct dc *dc, + struct dc_phy_addr_space_config *pa_config); + +bool dcn21_s0i3_golden_init_wa(struct dc *dc); + +void dcn21_exit_optimized_pwr_state( + const struct dc *dc, + struct dc_state *context); + +void dcn21_optimize_pwr_state( + const struct dc *dc, + struct dc_state *context); + +#endif /* __DC_HWSS_DCN21_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c new file mode 100644 index 000000000000..4861aa5c59ae --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c @@ -0,0 +1,142 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dce110/dce110_hw_sequencer.h" +#include "dcn10/dcn10_hw_sequencer.h" +#include "dcn20/dcn20_hwseq.h" +#include "dcn21_hwseq.h" + +static const struct hw_sequencer_funcs dcn21_funcs = { + .program_gamut_remap = dcn10_program_gamut_remap, + .init_hw = dcn10_init_hw, + .apply_ctx_to_hw = dce110_apply_ctx_to_hw, + .apply_ctx_for_surface = NULL, + .program_front_end_for_ctx = dcn20_program_front_end_for_ctx, + .update_plane_addr = dcn20_update_plane_addr, + .update_dchub = dcn10_update_dchub, + .update_pending_status = dcn10_update_pending_status, + .program_output_csc = dcn20_program_output_csc, + .enable_accelerated_mode = dce110_enable_accelerated_mode, + .enable_timing_synchronization = dcn10_enable_timing_synchronization, + .enable_per_frame_crtc_position_reset = dcn10_enable_per_frame_crtc_position_reset, + .update_info_frame = dce110_update_info_frame, + .send_immediate_sdp_message = dcn10_send_immediate_sdp_message, + .enable_stream = dcn20_enable_stream, + .disable_stream = dce110_disable_stream, + .unblank_stream = dcn20_unblank_stream, + .blank_stream = dce110_blank_stream, + .enable_audio_stream = dce110_enable_audio_stream, + .disable_audio_stream = dce110_disable_audio_stream, + .disable_plane = dcn20_disable_plane, + .pipe_control_lock = dcn20_pipe_control_lock, + .pipe_control_lock_global = dcn20_pipe_control_lock_global, + .prepare_bandwidth = dcn20_prepare_bandwidth, + .optimize_bandwidth = dcn20_optimize_bandwidth, + .update_bandwidth = dcn20_update_bandwidth, + .set_drr = dcn10_set_drr, + .get_position = dcn10_get_position, + .set_static_screen_control = dcn10_set_static_screen_control, + .setup_stereo = dcn10_setup_stereo, + .set_avmute = dce110_set_avmute, + .log_hw_state = dcn10_log_hw_state, + .get_hw_state = dcn10_get_hw_state, + .clear_status_bits = dcn10_clear_status_bits, + .wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect, + .edp_power_control = dce110_edp_power_control, + .edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready, + .set_cursor_position = dcn10_set_cursor_position, + .set_cursor_attribute = dcn10_set_cursor_attribute, + .set_cursor_sdr_white_level = dcn10_set_cursor_sdr_white_level, + .setup_periodic_interrupt = dcn10_setup_periodic_interrupt, + .set_clock = dcn10_set_clock, + .get_clock = dcn10_get_clock, + .program_triplebuffer = dcn20_program_triple_buffer, + .enable_writeback = dcn20_enable_writeback, + .disable_writeback = dcn20_disable_writeback, + .dmdata_status_done = dcn20_dmdata_status_done, + .program_dmdata_engine = dcn20_program_dmdata_engine, + .set_dmdata_attributes = dcn20_set_dmdata_attributes, + .init_sys_ctx = dcn21_init_sys_ctx, + .init_vm_ctx = dcn20_init_vm_ctx, + .set_flip_control_gsl = dcn20_set_flip_control_gsl, + .optimize_pwr_state = dcn21_optimize_pwr_state, + .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, + .get_vupdate_offset_from_vsync = dcn10_get_vupdate_offset_from_vsync, + .set_cursor_position = dcn10_set_cursor_position, + .set_cursor_attribute = dcn10_set_cursor_attribute, + .set_cursor_sdr_white_level = dcn10_set_cursor_sdr_white_level, + .optimize_pwr_state = dcn21_optimize_pwr_state, + .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, +}; + +static const struct hwseq_private_funcs dcn21_private_funcs = { + .init_pipes = dcn10_init_pipes, + .update_plane_addr = dcn20_update_plane_addr, + .plane_atomic_disconnect = dcn10_plane_atomic_disconnect, + .update_mpcc = dcn20_update_mpcc, + .set_input_transfer_func = dcn20_set_input_transfer_func, + .set_output_transfer_func = dcn20_set_output_transfer_func, + .power_down = dce110_power_down, + .enable_display_power_gating = dcn10_dummy_display_power_gating, + .blank_pixel_data = dcn20_blank_pixel_data, + .reset_hw_ctx_wrap = dcn20_reset_hw_ctx_wrap, + .enable_stream_timing = dcn20_enable_stream_timing, + .edp_backlight_control = dce110_edp_backlight_control, + .disable_stream_gating = dcn20_disable_stream_gating, + .enable_stream_gating = dcn20_enable_stream_gating, + .setup_vupdate_interrupt = dcn20_setup_vupdate_interrupt, + .did_underflow_occur = dcn10_did_underflow_occur, + .init_blank = dcn20_init_blank, + .disable_vga = dcn20_disable_vga, + .bios_golden_init = dcn10_bios_golden_init, + .plane_atomic_disable = dcn20_plane_atomic_disable, + .plane_atomic_power_down = dcn10_plane_atomic_power_down, + .enable_power_gating_plane = dcn20_enable_power_gating_plane, + .dpp_pg_control = dcn20_dpp_pg_control, + .hubp_pg_control = dcn20_hubp_pg_control, + .dsc_pg_control = NULL, + .update_odm = dcn20_update_odm, + .dsc_pg_control = dcn20_dsc_pg_control, + .get_surface_visual_confirm_color = dcn10_get_surface_visual_confirm_color, + .get_hdr_visual_confirm_color = dcn10_get_hdr_visual_confirm_color, + .set_hdr_multiplier = dcn10_set_hdr_multiplier, + .verify_allow_pstate_change_high = dcn10_verify_allow_pstate_change_high, + .s0i3_golden_init_wa = dcn21_s0i3_golden_init_wa, + .wait_for_blank_complete = dcn20_wait_for_blank_complete, + .dccg_init = dcn20_dccg_init, + .set_blend_lut = dcn20_set_blend_lut, + .set_shaper_3dlut = dcn20_set_shaper_3dlut, +}; + +void dcn21_hw_sequencer_construct(struct dc *dc) +{ + dc->hwss = dcn21_funcs; + dc->hwseq->funcs = dcn21_private_funcs; + + if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + dc->hwss.init_hw = dcn20_fpga_init_hw; + dc->hwseq->funcs.init_pipes = NULL; + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.h new file mode 100644 index 000000000000..3ed24292648a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.h @@ -0,0 +1,33 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_DCN21_INIT_H__ +#define __DC_DCN21_INIT_H__ + +struct dc; + +void dcn21_hw_sequencer_construct(struct dc *dc); + +#endif /* __DC_DCN20_INIT_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c new file mode 100644 index 000000000000..e45683ac871a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c @@ -0,0 +1,468 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "reg_helper.h" + +#include <linux/delay.h> +#include "core_types.h" +#include "link_encoder.h" +#include "dcn21_link_encoder.h" +#include "stream_encoder.h" + +#include "i2caux_interface.h" +#include "dc_bios_types.h" + +#include "gpio_service_interface.h" + +#define CTX \ + enc10->base.ctx +#define DC_LOGGER \ + enc10->base.ctx->logger + +#define REG(reg)\ + (enc10->link_regs->reg) + +#undef FN +#define FN(reg_name, field_name) \ + enc10->link_shift->field_name, enc10->link_mask->field_name + +#define IND_REG(index) \ + (enc10->link_regs->index) + +static struct mpll_cfg dcn21_mpll_cfg_ref[] = { + // RBR + { + .hdmimode_enable = 0, + .ref_range = 1, + .ref_clk_mpllb_div = 1, + .mpllb_ssc_en = 1, + .mpllb_div5_clk_en = 1, + .mpllb_multiplier = 238, + .mpllb_fracn_en = 0, + .mpllb_fracn_quot = 0, + .mpllb_fracn_rem = 0, + .mpllb_fracn_den = 1, + .mpllb_ssc_up_spread = 0, + .mpllb_ssc_peak = 44237, + .mpllb_ssc_stepsize = 59454, + .mpllb_div_clk_en = 0, + .mpllb_div_multiplier = 0, + .mpllb_hdmi_div = 0, + .mpllb_tx_clk_div = 2, + .tx_vboost_lvl = 5, + .mpllb_pmix_en = 1, + .mpllb_word_div2_en = 0, + .mpllb_ana_v2i = 2, + .mpllb_ana_freq_vco = 2, + .mpllb_ana_cp_int = 9, + .mpllb_ana_cp_prop = 15, + .hdmi_pixel_clk_div = 0, + }, + // HBR + { + .hdmimode_enable = 0, + .ref_range = 1, + .ref_clk_mpllb_div = 1, + .mpllb_ssc_en = 1, + .mpllb_div5_clk_en = 1, + .mpllb_multiplier = 192, + .mpllb_fracn_en = 1, + .mpllb_fracn_quot = 32768, + .mpllb_fracn_rem = 0, + .mpllb_fracn_den = 1, + .mpllb_ssc_up_spread = 0, + .mpllb_ssc_peak = 36864, + .mpllb_ssc_stepsize = 49545, + .mpllb_div_clk_en = 0, + .mpllb_div_multiplier = 0, + .mpllb_hdmi_div = 0, + .mpllb_tx_clk_div = 1, + .tx_vboost_lvl = 5, + .mpllb_pmix_en = 1, + .mpllb_word_div2_en = 0, + .mpllb_ana_v2i = 2, + .mpllb_ana_freq_vco = 3, + .mpllb_ana_cp_int = 9, + .mpllb_ana_cp_prop = 15, + .hdmi_pixel_clk_div = 0, + }, + //HBR2 + { + .hdmimode_enable = 0, + .ref_range = 1, + .ref_clk_mpllb_div = 1, + .mpllb_ssc_en = 1, + .mpllb_div5_clk_en = 1, + .mpllb_multiplier = 192, + .mpllb_fracn_en = 1, + .mpllb_fracn_quot = 32768, + .mpllb_fracn_rem = 0, + .mpllb_fracn_den = 1, + .mpllb_ssc_up_spread = 0, + .mpllb_ssc_peak = 36864, + .mpllb_ssc_stepsize = 49545, + .mpllb_div_clk_en = 0, + .mpllb_div_multiplier = 0, + .mpllb_hdmi_div = 0, + .mpllb_tx_clk_div = 0, + .tx_vboost_lvl = 5, + .mpllb_pmix_en = 1, + .mpllb_word_div2_en = 0, + .mpllb_ana_v2i = 2, + .mpllb_ana_freq_vco = 3, + .mpllb_ana_cp_int = 9, + .mpllb_ana_cp_prop = 15, + .hdmi_pixel_clk_div = 0, + }, + //HBR3 + { + .hdmimode_enable = 0, + .ref_range = 1, + .ref_clk_mpllb_div = 1, + .mpllb_ssc_en = 1, + .mpllb_div5_clk_en = 1, + .mpllb_multiplier = 304, + .mpllb_fracn_en = 1, + .mpllb_fracn_quot = 49152, + .mpllb_fracn_rem = 0, + .mpllb_fracn_den = 1, + .mpllb_ssc_up_spread = 0, + .mpllb_ssc_peak = 55296, + .mpllb_ssc_stepsize = 74318, + .mpllb_div_clk_en = 0, + .mpllb_div_multiplier = 0, + .mpllb_hdmi_div = 0, + .mpllb_tx_clk_div = 0, + .tx_vboost_lvl = 5, + .mpllb_pmix_en = 1, + .mpllb_word_div2_en = 0, + .mpllb_ana_v2i = 2, + .mpllb_ana_freq_vco = 1, + .mpllb_ana_cp_int = 7, + .mpllb_ana_cp_prop = 16, + .hdmi_pixel_clk_div = 0, + }, +}; + + +static bool update_cfg_data( + struct dcn10_link_encoder *enc10, + const struct dc_link_settings *link_settings, + struct dpcssys_phy_seq_cfg *cfg) +{ + int i; + + cfg->load_sram_fw = false; + cfg->use_calibration_setting = true; + + //TODO: need to implement a proper lane mapping for Renoir. + for (i = 0; i < 4; i++) + cfg->lane_en[i] = true; + + switch (link_settings->link_rate) { + case LINK_RATE_LOW: + cfg->mpll_cfg = dcn21_mpll_cfg_ref[0]; + break; + case LINK_RATE_HIGH: + cfg->mpll_cfg = dcn21_mpll_cfg_ref[1]; + break; + case LINK_RATE_HIGH2: + cfg->mpll_cfg = dcn21_mpll_cfg_ref[2]; + break; + case LINK_RATE_HIGH3: + cfg->mpll_cfg = dcn21_mpll_cfg_ref[3]; + break; + default: + DC_LOG_ERROR("%s: No supported link rate found %X!\n", + __func__, link_settings->link_rate); + return false; + } + + return true; +} + +void dcn21_link_encoder_get_max_link_cap(struct link_encoder *enc, + struct dc_link_settings *link_settings) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + uint32_t value; + + REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DP4, &value); + + if (!value && link_settings->lane_count > LANE_COUNT_TWO) + link_settings->lane_count = LANE_COUNT_TWO; +} + +bool dcn21_link_encoder_is_in_alt_mode(struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + uint32_t value; + + REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DISABLE, &value); + + // if value == 1 alt mode is disabled, otherwise it is enabled + return !value; +} + +bool dcn21_link_encoder_acquire_phy(struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + int value; + + if (enc->features.flags.bits.DP_IS_USB_C) { + REG_GET(RDPCSTX_PHY_CNTL6, + RDPCS_PHY_DPALT_DISABLE, &value); + + if (value == 1) { + ASSERT(0); + return false; + } + REG_UPDATE(RDPCSTX_PHY_CNTL6, + RDPCS_PHY_DPALT_DISABLE_ACK, 0); + + udelay(40); + + REG_GET(RDPCSTX_PHY_CNTL6, + RDPCS_PHY_DPALT_DISABLE, &value); + if (value == 1) { + ASSERT(0); + REG_UPDATE(RDPCSTX_PHY_CNTL6, + RDPCS_PHY_DPALT_DISABLE_ACK, 1); + return false; + } + } + + REG_UPDATE(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_REF_CLK_EN, 1); + + return true; +} + + + +static void dcn21_link_encoder_release_phy(struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + + if (enc->features.flags.bits.DP_IS_USB_C) { + REG_UPDATE(RDPCSTX_PHY_CNTL6, + RDPCS_PHY_DPALT_DISABLE_ACK, 1); + } + + REG_UPDATE(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DP_REF_CLK_EN, 0); + +} + +void dcn21_link_encoder_enable_dp_output( + struct link_encoder *enc, + const struct dc_link_settings *link_settings, + enum clock_source_id clock_source) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + struct dcn21_link_encoder *enc21 = (struct dcn21_link_encoder *) enc10; + struct dpcssys_phy_seq_cfg *cfg = &enc21->phy_seq_cfg; + + if (!dcn21_link_encoder_acquire_phy(enc)) + return; + + if (!enc->ctx->dc->debug.avoid_vbios_exec_table) { + dcn10_link_encoder_enable_dp_output(enc, link_settings, clock_source); + return; + } + + if (!update_cfg_data(enc10, link_settings, cfg)) + return; + + enc1_configure_encoder(enc10, link_settings); + + dcn10_link_encoder_setup(enc, SIGNAL_TYPE_DISPLAY_PORT); + +} + +void dcn21_link_encoder_enable_dp_mst_output( + struct link_encoder *enc, + const struct dc_link_settings *link_settings, + enum clock_source_id clock_source) +{ + if (!dcn21_link_encoder_acquire_phy(enc)) + return; + + dcn10_link_encoder_enable_dp_mst_output(enc, link_settings, clock_source); +} + +void dcn21_link_encoder_disable_output( + struct link_encoder *enc, + enum signal_type signal) +{ + dcn10_link_encoder_disable_output(enc, signal); + + if (dc_is_dp_signal(signal)) + dcn21_link_encoder_release_phy(enc); +} + + +static const struct link_encoder_funcs dcn21_link_enc_funcs = { + .read_state = link_enc2_read_state, + .validate_output_with_stream = + dcn10_link_encoder_validate_output_with_stream, + .hw_init = enc2_hw_init, + .setup = dcn10_link_encoder_setup, + .enable_tmds_output = dcn10_link_encoder_enable_tmds_output, + .enable_dp_output = dcn21_link_encoder_enable_dp_output, + .enable_dp_mst_output = dcn21_link_encoder_enable_dp_mst_output, + .disable_output = dcn21_link_encoder_disable_output, + .dp_set_lane_settings = dcn10_link_encoder_dp_set_lane_settings, + .dp_set_phy_pattern = dcn10_link_encoder_dp_set_phy_pattern, + .update_mst_stream_allocation_table = + dcn10_link_encoder_update_mst_stream_allocation_table, + .psr_program_dp_dphy_fast_training = + dcn10_psr_program_dp_dphy_fast_training, + .psr_program_secondary_packet = dcn10_psr_program_secondary_packet, + .connect_dig_be_to_fe = dcn10_link_encoder_connect_dig_be_to_fe, + .enable_hpd = dcn10_link_encoder_enable_hpd, + .disable_hpd = dcn10_link_encoder_disable_hpd, + .is_dig_enabled = dcn10_is_dig_enabled, + .destroy = dcn10_link_encoder_destroy, + .fec_set_enable = enc2_fec_set_enable, + .fec_set_ready = enc2_fec_set_ready, + .fec_is_active = enc2_fec_is_active, + .get_dig_frontend = dcn10_get_dig_frontend, + .is_in_alt_mode = dcn21_link_encoder_is_in_alt_mode, + .get_max_link_cap = dcn21_link_encoder_get_max_link_cap, +}; + +void dcn21_link_encoder_construct( + struct dcn21_link_encoder *enc21, + const struct encoder_init_data *init_data, + const struct encoder_feature_support *enc_features, + const struct dcn10_link_enc_registers *link_regs, + const struct dcn10_link_enc_aux_registers *aux_regs, + const struct dcn10_link_enc_hpd_registers *hpd_regs, + const struct dcn10_link_enc_shift *link_shift, + const struct dcn10_link_enc_mask *link_mask) +{ + struct bp_encoder_cap_info bp_cap_info = {0}; + const struct dc_vbios_funcs *bp_funcs = init_data->ctx->dc_bios->funcs; + enum bp_result result = BP_RESULT_OK; + struct dcn10_link_encoder *enc10 = &enc21->enc10; + + enc10->base.funcs = &dcn21_link_enc_funcs; + enc10->base.ctx = init_data->ctx; + enc10->base.id = init_data->encoder; + + enc10->base.hpd_source = init_data->hpd_source; + enc10->base.connector = init_data->connector; + + enc10->base.preferred_engine = ENGINE_ID_UNKNOWN; + + enc10->base.features = *enc_features; + + enc10->base.transmitter = init_data->transmitter; + + /* set the flag to indicate whether driver poll the I2C data pin + * while doing the DP sink detect + */ + +/* if (dal_adapter_service_is_feature_supported(as, + FEATURE_DP_SINK_DETECT_POLL_DATA_PIN)) + enc10->base.features.flags.bits. + DP_SINK_DETECT_POLL_DATA_PIN = true;*/ + + enc10->base.output_signals = + SIGNAL_TYPE_DVI_SINGLE_LINK | + SIGNAL_TYPE_DVI_DUAL_LINK | + SIGNAL_TYPE_LVDS | + SIGNAL_TYPE_DISPLAY_PORT | + SIGNAL_TYPE_DISPLAY_PORT_MST | + SIGNAL_TYPE_EDP | + SIGNAL_TYPE_HDMI_TYPE_A; + + /* For DCE 8.0 and 8.1, by design, UNIPHY is hardwired to DIG_BE. + * SW always assign DIG_FE 1:1 mapped to DIG_FE for non-MST UNIPHY. + * SW assign DIG_FE to non-MST UNIPHY first and MST last. So prefer + * DIG is per UNIPHY and used by SST DP, eDP, HDMI, DVI and LVDS. + * Prefer DIG assignment is decided by board design. + * For DCE 8.0, there are only max 6 UNIPHYs, we assume board design + * and VBIOS will filter out 7 UNIPHY for DCE 8.0. + * By this, adding DIGG should not hurt DCE 8.0. + * This will let DCE 8.1 share DCE 8.0 as much as possible + */ + + enc10->link_regs = link_regs; + enc10->aux_regs = aux_regs; + enc10->hpd_regs = hpd_regs; + enc10->link_shift = link_shift; + enc10->link_mask = link_mask; + + switch (enc10->base.transmitter) { + case TRANSMITTER_UNIPHY_A: + enc10->base.preferred_engine = ENGINE_ID_DIGA; + break; + case TRANSMITTER_UNIPHY_B: + enc10->base.preferred_engine = ENGINE_ID_DIGB; + break; + case TRANSMITTER_UNIPHY_C: + enc10->base.preferred_engine = ENGINE_ID_DIGC; + break; + case TRANSMITTER_UNIPHY_D: + enc10->base.preferred_engine = ENGINE_ID_DIGD; + break; + case TRANSMITTER_UNIPHY_E: + enc10->base.preferred_engine = ENGINE_ID_DIGE; + break; + case TRANSMITTER_UNIPHY_F: + enc10->base.preferred_engine = ENGINE_ID_DIGF; + break; + case TRANSMITTER_UNIPHY_G: + enc10->base.preferred_engine = ENGINE_ID_DIGG; + break; + default: + ASSERT_CRITICAL(false); + enc10->base.preferred_engine = ENGINE_ID_UNKNOWN; + } + + /* default to one to mirror Windows behavior */ + enc10->base.features.flags.bits.HDMI_6GB_EN = 1; + + result = bp_funcs->get_encoder_cap_info(enc10->base.ctx->dc_bios, + enc10->base.id, &bp_cap_info); + + /* Override features with DCE-specific values */ + if (result == BP_RESULT_OK) { + enc10->base.features.flags.bits.IS_HBR2_CAPABLE = + bp_cap_info.DP_HBR2_EN; + enc10->base.features.flags.bits.IS_HBR3_CAPABLE = + bp_cap_info.DP_HBR3_EN; + enc10->base.features.flags.bits.HDMI_6GB_EN = bp_cap_info.HDMI_6GB_EN; + enc10->base.features.flags.bits.DP_IS_USB_C = + bp_cap_info.DP_IS_USB_C; + } else { + DC_LOG_WARNING("%s: Failed to get encoder_cap_info from VBIOS with error code %d!\n", + __func__, + result); + } + if (enc10->base.ctx->dc->debug.hdmi20_disable) { + enc10->base.features.flags.bits.HDMI_6GB_EN = 0; + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.h new file mode 100644 index 000000000000..033d5d76f195 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.h @@ -0,0 +1,100 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_LINK_ENCODER__DCN21_H__ +#define __DC_LINK_ENCODER__DCN21_H__ + +#include "dcn20/dcn20_link_encoder.h" + +struct dcn21_link_encoder { + struct dcn10_link_encoder enc10; + struct dpcssys_phy_seq_cfg phy_seq_cfg; +}; + +#define DPCS_DCN21_MASK_SH_LIST(mask_sh)\ + DPCS_DCN2_MASK_SH_LIST(mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_TX_VBOOST_LVL, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE2, RDPCS_PHY_DP_MPLLB_CP_PROP_GS, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_RX_VREF_CTRL, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_DP_MPLLB_CP_INT_GS, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_DMCU_DPALT_DIS_BLOCK_REG, RDPCS_DMCU_DPALT_DIS_BLOCK_REG, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL15, RDPCS_PHY_SUP_PRE_HP, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL15, RDPCS_PHY_DP_TX0_VREGDRV_BYP, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL15, RDPCS_PHY_DP_TX1_VREGDRV_BYP, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL15, RDPCS_PHY_DP_TX2_VREGDRV_BYP, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL15, RDPCS_PHY_DP_TX3_VREGDRV_BYP, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DP4, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DISABLE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_DP_TX0_EQ_MAIN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_DP_TX0_EQ_PRE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE0, RDPCS_PHY_DP_TX0_EQ_POST, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE1, RDPCS_PHY_DP_TX1_EQ_MAIN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE1, RDPCS_PHY_DP_TX1_EQ_PRE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE1, RDPCS_PHY_DP_TX1_EQ_POST, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE2, RDPCS_PHY_DP_TX2_EQ_MAIN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE2, RDPCS_PHY_DP_TX2_EQ_PRE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE2, RDPCS_PHY_DP_TX2_EQ_POST, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DP_TX3_EQ_MAIN, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DCO_FINETUNE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DCO_RANGE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DP_TX3_EQ_PRE, mask_sh),\ + LE_SF(RDPCSTX0_RDPCSTX_PHY_FUSE3, RDPCS_PHY_DP_TX3_EQ_POST, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYA_SOFT_RESET, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYB_SOFT_RESET, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYC_SOFT_RESET, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYD_SOFT_RESET, mask_sh),\ + LE_SF(DCIO_SOFT_RESET, UNIPHYE_SOFT_RESET, mask_sh) + +#define DPCS_DCN21_REG_LIST(id) \ + DPCS_DCN2_REG_LIST(id),\ + SRI(RDPCSTX_PHY_CNTL15, RDPCSTX, id),\ + SRI(RDPCSTX_DMCU_DPALT_DIS_BLOCK_REG, RDPCSTX, id) + +#define LINK_ENCODER_MASK_SH_LIST_DCN21(mask_sh)\ + LINK_ENCODER_MASK_SH_LIST_DCN20(mask_sh),\ + LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_CHANNEL0_XBAR_SOURCE, mask_sh),\ + LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_CHANNEL1_XBAR_SOURCE, mask_sh),\ + LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_CHANNEL2_XBAR_SOURCE, mask_sh),\ + LE_SF(UNIPHYA_CHANNEL_XBAR_CNTL, UNIPHY_CHANNEL3_XBAR_SOURCE, mask_sh), \ + SRI(RDPCSTX_PHY_FUSE2, RDPCSTX, id), \ + SRI(RDPCSTX_PHY_FUSE3, RDPCSTX, id), \ + SR(RDPCSTX0_RDPCSTX_SCRATCH) + +void dcn21_link_encoder_enable_dp_output( + struct link_encoder *enc, + const struct dc_link_settings *link_settings, + enum clock_source_id clock_source); + +void dcn21_link_encoder_construct( + struct dcn21_link_encoder *enc21, + const struct encoder_init_data *init_data, + const struct encoder_feature_support *enc_features, + const struct dcn10_link_enc_registers *link_regs, + const struct dcn10_link_enc_aux_registers *aux_regs, + const struct dcn10_link_enc_hpd_registers *hpd_regs, + const struct dcn10_link_enc_shift *link_shift, + const struct dcn10_link_enc_mask *link_mask); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c new file mode 100644 index 000000000000..33d0a176841a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -0,0 +1,1938 @@ +/* +* Copyright 2018 Advanced Micro Devices, Inc. + * Copyright 2019 Raptor Engineering, LLC + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include <linux/slab.h> + +#include "dm_services.h" +#include "dc.h" + +#include "dcn21_init.h" + +#include "resource.h" +#include "include/irq_service_interface.h" +#include "dcn20/dcn20_resource.h" + +#include "clk_mgr.h" +#include "dcn10/dcn10_hubp.h" +#include "dcn10/dcn10_ipp.h" +#include "dcn20/dcn20_hubbub.h" +#include "dcn20/dcn20_mpc.h" +#include "dcn20/dcn20_hubp.h" +#include "dcn21_hubp.h" +#include "irq/dcn21/irq_service_dcn21.h" +#include "dcn20/dcn20_dpp.h" +#include "dcn20/dcn20_optc.h" +#include "dcn21/dcn21_hwseq.h" +#include "dce110/dce110_hw_sequencer.h" +#include "dcn20/dcn20_opp.h" +#include "dcn20/dcn20_dsc.h" +#include "dcn21/dcn21_link_encoder.h" +#include "dcn20/dcn20_stream_encoder.h" +#include "dce/dce_clock_source.h" +#include "dce/dce_audio.h" +#include "dce/dce_hwseq.h" +#include "virtual/virtual_stream_encoder.h" +#include "dce110/dce110_resource.h" +#include "dml/display_mode_vba.h" +#include "dcn20/dcn20_dccg.h" +#include "dcn21_hubbub.h" +#include "dcn10/dcn10_resource.h" +#include "dce110/dce110_resource.h" + +#include "dcn20/dcn20_dwb.h" +#include "dcn20/dcn20_mmhubbub.h" +#include "dpcs/dpcs_2_1_0_offset.h" +#include "dpcs/dpcs_2_1_0_sh_mask.h" + +#include "renoir_ip_offset.h" +#include "dcn/dcn_2_1_0_offset.h" +#include "dcn/dcn_2_1_0_sh_mask.h" + +#include "nbio/nbio_7_0_offset.h" + +#include "mmhub/mmhub_2_0_0_offset.h" +#include "mmhub/mmhub_2_0_0_sh_mask.h" + +#include "reg_helper.h" +#include "dce/dce_abm.h" +#include "dce/dce_dmcu.h" +#include "dce/dce_aux.h" +#include "dce/dce_i2c.h" +#include "dcn21_resource.h" +#include "vm_helper.h" +#include "dcn20/dcn20_vmid.h" +#include "../dce/dmub_psr.h" + +#define SOC_BOUNDING_BOX_VALID false +#define DC_LOGGER_INIT(logger) + + +struct _vcs_dpi_ip_params_st dcn2_1_ip = { + .odm_capable = 1, + .gpuvm_enable = 1, + .hostvm_enable = 1, + .gpuvm_max_page_table_levels = 1, + .hostvm_max_page_table_levels = 4, + .hostvm_cached_page_table_levels = 2, + .num_dsc = 3, + .rob_buffer_size_kbytes = 168, + .det_buffer_size_kbytes = 164, + .dpte_buffer_size_in_pte_reqs_luma = 44, + .dpte_buffer_size_in_pte_reqs_chroma = 42,//todo + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_enable = 1, + .max_page_table_levels = 4, + .pte_chunk_size_kbytes = 2, + .meta_chunk_size_kbytes = 2, + .writeback_chunk_size_kbytes = 2, + .line_buffer_size_bits = 789504, + .is_line_buffer_bpp_fixed = 0, + .line_buffer_fixed_bpp = 0, + .dcc_supported = true, + .max_line_buffer_lines = 12, + .writeback_luma_buffer_size_kbytes = 12, + .writeback_chroma_buffer_size_kbytes = 8, + .writeback_chroma_line_buffer_width_pixels = 4, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_taps = 12, + .writeback_max_vscl_taps = 12, + .writeback_line_buffer_luma_buffer_size = 0, + .writeback_line_buffer_chroma_buffer_size = 14643, + .cursor_buffer_size = 8, + .cursor_chunk_size = 2, + .max_num_otg = 4, + .max_num_dpp = 4, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 4, + .max_vscl_ratio = 4, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.10, + .min_vblank_lines = 32, // + .dppclk_delay_subtotal = 77, // + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_scl = 50, + .dppclk_delay_cnvc_formatter = 8, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 87, // + .dcfclk_cstate_latency = 10, // SRExitTime + .max_inter_dcn_tile_repeaters = 8, + + .xfc_supported = false, + .xfc_fill_bw_overhead_percent = 10.0, + .xfc_fill_constant_bytes = 0, + .ptoi_supported = 0 +}; + +struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 304.0, + .fabricclk_mhz = 600.0, + .dispclk_mhz = 618.0, + .dppclk_mhz = 440.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 205.67, + .dram_speed_mts = 1600.0, + }, + { + .state = 1, + .dcfclk_mhz = 304.0, + .fabricclk_mhz = 600.0, + .dispclk_mhz = 618.0, + .dppclk_mhz = 618.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 205.67, + .dram_speed_mts = 1600.0, + }, + { + .state = 2, + .dcfclk_mhz = 608.0, + .fabricclk_mhz = 1066.0, + .dispclk_mhz = 888.0, + .dppclk_mhz = 888.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 278.0, + .dscclk_mhz = 287.67, + .dram_speed_mts = 2133.0, + }, + { + .state = 3, + .dcfclk_mhz = 676.0, + .fabricclk_mhz = 1600.0, + .dispclk_mhz = 1015.0, + .dppclk_mhz = 1015.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 715.0, + .dscclk_mhz = 318.334, + .dram_speed_mts = 4266.0, + }, + { + .state = 4, + .dcfclk_mhz = 810.0, + .fabricclk_mhz = 1600.0, + .dispclk_mhz = 1395.0, + .dppclk_mhz = 1285.0, + .phyclk_mhz = 1325.0, + .socclk_mhz = 953.0, + .dscclk_mhz = 489.0, + .dram_speed_mts = 4266.0, + }, + /*Extra state, no dispclk ramping*/ + { + .state = 5, + .dcfclk_mhz = 810.0, + .fabricclk_mhz = 1600.0, + .dispclk_mhz = 1395.0, + .dppclk_mhz = 1285.0, + .phyclk_mhz = 1325.0, + .socclk_mhz = 953.0, + .dscclk_mhz = 489.0, + .dram_speed_mts = 4266.0, + }, + + }, + + .sr_exit_time_us = 12.5, + .sr_enter_plus_exit_time_us = 17.0, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 75.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 60.0, + .max_avg_dram_bw_use_normal_percent = 100.0, + .writeback_latency_us = 12.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 4, + .fabric_datapath_to_dcn_data_return_bytes = 32, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.5, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 128, + .urgent_out_of_order_return_per_channel_bytes = 4096, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 4, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 23.84, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3600, + .xfc_bus_transport_time_us = 4, + .xfc_xbuf_latency_tolerance_us = 4, + .use_urgent_burst_bw = 1, + .num_states = 5 +}; + +#ifndef MAX +#define MAX(X, Y) ((X) > (Y) ? (X) : (Y)) +#endif +#ifndef MIN +#define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) +#endif + +/* begin ********************* + * macros to expend register list macro defined in HW object header file */ + +/* DCN */ +/* TODO awful hack. fixup dcn20_dwb.h */ +#undef BASE_INNER +#define BASE_INNER(seg) DMU_BASE__INST0_SEG ## seg + +#define BASE(seg) BASE_INNER(seg) + +#define SR(reg_name)\ + .reg_name = BASE(mm ## reg_name ## _BASE_IDX) + \ + mm ## reg_name + +#define SRI(reg_name, block, id)\ + .reg_name = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +#define SRIR(var_name, reg_name, block, id)\ + .var_name = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +#define SRII(reg_name, block, id)\ + .reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +#define DCCG_SRII(reg_name, block, id)\ + .block ## _ ## reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +/* NBIO */ +#define NBIO_BASE_INNER(seg) \ + NBIF0_BASE__INST0_SEG ## seg + +#define NBIO_BASE(seg) \ + NBIO_BASE_INNER(seg) + +#define NBIO_SR(reg_name)\ + .reg_name = NBIO_BASE(mm ## reg_name ## _BASE_IDX) + \ + mm ## reg_name + +/* MMHUB */ +#define MMHUB_BASE_INNER(seg) \ + MMHUB_BASE__INST0_SEG ## seg + +#define MMHUB_BASE(seg) \ + MMHUB_BASE_INNER(seg) + +#define MMHUB_SR(reg_name)\ + .reg_name = MMHUB_BASE(mmMM ## reg_name ## _BASE_IDX) + \ + mmMM ## reg_name + +#define clk_src_regs(index, pllid)\ +[index] = {\ + CS_COMMON_REG_LIST_DCN2_1(index, pllid),\ +} + +static const struct dce110_clk_src_regs clk_src_regs[] = { + clk_src_regs(0, A), + clk_src_regs(1, B), + clk_src_regs(2, C), + clk_src_regs(3, D), + clk_src_regs(4, E), +}; + +static const struct dce110_clk_src_shift cs_shift = { + CS_COMMON_MASK_SH_LIST_DCN2_0(__SHIFT) +}; + +static const struct dce110_clk_src_mask cs_mask = { + CS_COMMON_MASK_SH_LIST_DCN2_0(_MASK) +}; + +static const struct bios_registers bios_regs = { + NBIO_SR(BIOS_SCRATCH_3), + NBIO_SR(BIOS_SCRATCH_6) +}; + +static const struct dce_dmcu_registers dmcu_regs = { + DMCU_DCN20_REG_LIST() +}; + +static const struct dce_dmcu_shift dmcu_shift = { + DMCU_MASK_SH_LIST_DCN10(__SHIFT) +}; + +static const struct dce_dmcu_mask dmcu_mask = { + DMCU_MASK_SH_LIST_DCN10(_MASK) +}; + +static const struct dce_abm_registers abm_regs = { + ABM_DCN20_REG_LIST() +}; + +static const struct dce_abm_shift abm_shift = { + ABM_MASK_SH_LIST_DCN20(__SHIFT) +}; + +static const struct dce_abm_mask abm_mask = { + ABM_MASK_SH_LIST_DCN20(_MASK) +}; + +#define audio_regs(id)\ +[id] = {\ + AUD_COMMON_REG_LIST(id)\ +} + +static const struct dce_audio_registers audio_regs[] = { + audio_regs(0), + audio_regs(1), + audio_regs(2), + audio_regs(3), + audio_regs(4), + audio_regs(5), +}; + +#define DCE120_AUD_COMMON_MASK_SH_LIST(mask_sh)\ + SF(AZF0ENDPOINT0_AZALIA_F0_CODEC_ENDPOINT_INDEX, AZALIA_ENDPOINT_REG_INDEX, mask_sh),\ + SF(AZF0ENDPOINT0_AZALIA_F0_CODEC_ENDPOINT_DATA, AZALIA_ENDPOINT_REG_DATA, mask_sh),\ + AUD_COMMON_MASK_SH_LIST_BASE(mask_sh) + +static const struct dce_audio_shift audio_shift = { + DCE120_AUD_COMMON_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_audio_mask audio_mask = { + DCE120_AUD_COMMON_MASK_SH_LIST(_MASK) +}; + +static const struct dccg_registers dccg_regs = { + DCCG_COMMON_REG_LIST_DCN_BASE() +}; + +static const struct dccg_shift dccg_shift = { + DCCG_MASK_SH_LIST_DCN2(__SHIFT) +}; + +static const struct dccg_mask dccg_mask = { + DCCG_MASK_SH_LIST_DCN2(_MASK) +}; + +#define opp_regs(id)\ +[id] = {\ + OPP_REG_LIST_DCN20(id),\ +} + +static const struct dcn20_opp_registers opp_regs[] = { + opp_regs(0), + opp_regs(1), + opp_regs(2), + opp_regs(3), + opp_regs(4), + opp_regs(5), +}; + +static const struct dcn20_opp_shift opp_shift = { + OPP_MASK_SH_LIST_DCN20(__SHIFT) +}; + +static const struct dcn20_opp_mask opp_mask = { + OPP_MASK_SH_LIST_DCN20(_MASK) +}; + +#define tg_regs(id)\ +[id] = {TG_COMMON_REG_LIST_DCN2_0(id)} + +static const struct dcn_optc_registers tg_regs[] = { + tg_regs(0), + tg_regs(1), + tg_regs(2), + tg_regs(3) +}; + +static const struct dcn_optc_shift tg_shift = { + TG_COMMON_MASK_SH_LIST_DCN2_0(__SHIFT) +}; + +static const struct dcn_optc_mask tg_mask = { + TG_COMMON_MASK_SH_LIST_DCN2_0(_MASK) +}; + +static const struct dcn20_mpc_registers mpc_regs = { + MPC_REG_LIST_DCN2_0(0), + MPC_REG_LIST_DCN2_0(1), + MPC_REG_LIST_DCN2_0(2), + MPC_REG_LIST_DCN2_0(3), + MPC_REG_LIST_DCN2_0(4), + MPC_REG_LIST_DCN2_0(5), + MPC_OUT_MUX_REG_LIST_DCN2_0(0), + MPC_OUT_MUX_REG_LIST_DCN2_0(1), + MPC_OUT_MUX_REG_LIST_DCN2_0(2), + MPC_OUT_MUX_REG_LIST_DCN2_0(3), + MPC_DBG_REG_LIST_DCN2_0() +}; + +static const struct dcn20_mpc_shift mpc_shift = { + MPC_COMMON_MASK_SH_LIST_DCN2_0(__SHIFT), + MPC_DEBUG_REG_LIST_SH_DCN20 +}; + +static const struct dcn20_mpc_mask mpc_mask = { + MPC_COMMON_MASK_SH_LIST_DCN2_0(_MASK), + MPC_DEBUG_REG_LIST_MASK_DCN20 +}; + +#define hubp_regs(id)\ +[id] = {\ + HUBP_REG_LIST_DCN21(id)\ +} + +static const struct dcn_hubp2_registers hubp_regs[] = { + hubp_regs(0), + hubp_regs(1), + hubp_regs(2), + hubp_regs(3) +}; + +static const struct dcn_hubp2_shift hubp_shift = { + HUBP_MASK_SH_LIST_DCN21(__SHIFT) +}; + +static const struct dcn_hubp2_mask hubp_mask = { + HUBP_MASK_SH_LIST_DCN21(_MASK) +}; + +static const struct dcn_hubbub_registers hubbub_reg = { + HUBBUB_REG_LIST_DCN21() +}; + +static const struct dcn_hubbub_shift hubbub_shift = { + HUBBUB_MASK_SH_LIST_DCN21(__SHIFT) +}; + +static const struct dcn_hubbub_mask hubbub_mask = { + HUBBUB_MASK_SH_LIST_DCN21(_MASK) +}; + + +#define vmid_regs(id)\ +[id] = {\ + DCN20_VMID_REG_LIST(id)\ +} + +static const struct dcn_vmid_registers vmid_regs[] = { + vmid_regs(0), + vmid_regs(1), + vmid_regs(2), + vmid_regs(3), + vmid_regs(4), + vmid_regs(5), + vmid_regs(6), + vmid_regs(7), + vmid_regs(8), + vmid_regs(9), + vmid_regs(10), + vmid_regs(11), + vmid_regs(12), + vmid_regs(13), + vmid_regs(14), + vmid_regs(15) +}; + +static const struct dcn20_vmid_shift vmid_shifts = { + DCN20_VMID_MASK_SH_LIST(__SHIFT) +}; + +static const struct dcn20_vmid_mask vmid_masks = { + DCN20_VMID_MASK_SH_LIST(_MASK) +}; + +#define dsc_regsDCN20(id)\ +[id] = {\ + DSC_REG_LIST_DCN20(id)\ +} + +static const struct dcn20_dsc_registers dsc_regs[] = { + dsc_regsDCN20(0), + dsc_regsDCN20(1), + dsc_regsDCN20(2), + dsc_regsDCN20(3), + dsc_regsDCN20(4), + dsc_regsDCN20(5) +}; + +static const struct dcn20_dsc_shift dsc_shift = { + DSC_REG_LIST_SH_MASK_DCN20(__SHIFT) +}; + +static const struct dcn20_dsc_mask dsc_mask = { + DSC_REG_LIST_SH_MASK_DCN20(_MASK) +}; + +#define ipp_regs(id)\ +[id] = {\ + IPP_REG_LIST_DCN20(id),\ +} + +static const struct dcn10_ipp_registers ipp_regs[] = { + ipp_regs(0), + ipp_regs(1), + ipp_regs(2), + ipp_regs(3), +}; + +static const struct dcn10_ipp_shift ipp_shift = { + IPP_MASK_SH_LIST_DCN20(__SHIFT) +}; + +static const struct dcn10_ipp_mask ipp_mask = { + IPP_MASK_SH_LIST_DCN20(_MASK), +}; + +#define opp_regs(id)\ +[id] = {\ + OPP_REG_LIST_DCN20(id),\ +} + + +#define aux_engine_regs(id)\ +[id] = {\ + AUX_COMMON_REG_LIST0(id), \ + .AUXN_IMPCAL = 0, \ + .AUXP_IMPCAL = 0, \ + .AUX_RESET_MASK = DP_AUX0_AUX_CONTROL__AUX_RESET_MASK, \ +} + +static const struct dce110_aux_registers aux_engine_regs[] = { + aux_engine_regs(0), + aux_engine_regs(1), + aux_engine_regs(2), + aux_engine_regs(3), + aux_engine_regs(4), +}; + +#define tf_regs(id)\ +[id] = {\ + TF_REG_LIST_DCN20(id),\ + TF_REG_LIST_DCN20_COMMON_APPEND(id),\ +} + +static const struct dcn2_dpp_registers tf_regs[] = { + tf_regs(0), + tf_regs(1), + tf_regs(2), + tf_regs(3), +}; + +static const struct dcn2_dpp_shift tf_shift = { + TF_REG_LIST_SH_MASK_DCN20(__SHIFT), + TF_DEBUG_REG_LIST_SH_DCN20 +}; + +static const struct dcn2_dpp_mask tf_mask = { + TF_REG_LIST_SH_MASK_DCN20(_MASK), + TF_DEBUG_REG_LIST_MASK_DCN20 +}; + +#define stream_enc_regs(id)\ +[id] = {\ + SE_DCN2_REG_LIST(id)\ +} + +static const struct dcn10_stream_enc_registers stream_enc_regs[] = { + stream_enc_regs(0), + stream_enc_regs(1), + stream_enc_regs(2), + stream_enc_regs(3), + stream_enc_regs(4), +}; + +static const struct dce110_aux_registers_shift aux_shift = { + DCN_AUX_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce110_aux_registers_mask aux_mask = { + DCN_AUX_MASK_SH_LIST(_MASK) +}; + +static const struct dcn10_stream_encoder_shift se_shift = { + SE_COMMON_MASK_SH_LIST_DCN20(__SHIFT) +}; + +static const struct dcn10_stream_encoder_mask se_mask = { + SE_COMMON_MASK_SH_LIST_DCN20(_MASK) +}; + +static void dcn21_pp_smu_destroy(struct pp_smu_funcs **pp_smu); + +static int dcn21_populate_dml_pipes_from_context( + struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes); + +static struct input_pixel_processor *dcn21_ipp_create( + struct dc_context *ctx, uint32_t inst) +{ + struct dcn10_ipp *ipp = + kzalloc(sizeof(struct dcn10_ipp), GFP_KERNEL); + + if (!ipp) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dcn20_ipp_construct(ipp, ctx, inst, + &ipp_regs[inst], &ipp_shift, &ipp_mask); + return &ipp->base; +} + +static struct dpp *dcn21_dpp_create( + struct dc_context *ctx, + uint32_t inst) +{ + struct dcn20_dpp *dpp = + kzalloc(sizeof(struct dcn20_dpp), GFP_KERNEL); + + if (!dpp) + return NULL; + + if (dpp2_construct(dpp, ctx, inst, + &tf_regs[inst], &tf_shift, &tf_mask)) + return &dpp->base; + + BREAK_TO_DEBUGGER(); + kfree(dpp); + return NULL; +} + +static struct dce_aux *dcn21_aux_engine_create( + struct dc_context *ctx, + uint32_t inst) +{ + struct aux_engine_dce110 *aux_engine = + kzalloc(sizeof(struct aux_engine_dce110), GFP_KERNEL); + + if (!aux_engine) + return NULL; + + dce110_aux_engine_construct(aux_engine, ctx, inst, + SW_AUX_TIMEOUT_PERIOD_MULTIPLIER * AUX_TIMEOUT_PERIOD, + &aux_engine_regs[inst], + &aux_mask, + &aux_shift, + ctx->dc->caps.extended_aux_timeout_support); + + return &aux_engine->base; +} + +#define i2c_inst_regs(id) { I2C_HW_ENGINE_COMMON_REG_LIST(id) } + +static const struct dce_i2c_registers i2c_hw_regs[] = { + i2c_inst_regs(1), + i2c_inst_regs(2), + i2c_inst_regs(3), + i2c_inst_regs(4), + i2c_inst_regs(5), +}; + +static const struct dce_i2c_shift i2c_shifts = { + I2C_COMMON_MASK_SH_LIST_DCN2(__SHIFT) +}; + +static const struct dce_i2c_mask i2c_masks = { + I2C_COMMON_MASK_SH_LIST_DCN2(_MASK) +}; + +struct dce_i2c_hw *dcn21_i2c_hw_create( + struct dc_context *ctx, + uint32_t inst) +{ + struct dce_i2c_hw *dce_i2c_hw = + kzalloc(sizeof(struct dce_i2c_hw), GFP_KERNEL); + + if (!dce_i2c_hw) + return NULL; + + dcn2_i2c_hw_construct(dce_i2c_hw, ctx, inst, + &i2c_hw_regs[inst], &i2c_shifts, &i2c_masks); + + return dce_i2c_hw; +} + +static const struct resource_caps res_cap_rn = { + .num_timing_generator = 4, + .num_opp = 4, + .num_video_plane = 4, + .num_audio = 4, // 4 audio endpoints. 4 audio streams + .num_stream_encoder = 5, + .num_pll = 5, // maybe 3 because the last two used for USB-c + .num_dwb = 1, + .num_ddc = 5, + .num_vmid = 1, + .num_dsc = 3, +}; + +#ifdef DIAGS_BUILD +static const struct resource_caps res_cap_rn_FPGA_4pipe = { + .num_timing_generator = 4, + .num_opp = 4, + .num_video_plane = 4, + .num_audio = 7, + .num_stream_encoder = 4, + .num_pll = 4, + .num_dwb = 1, + .num_ddc = 4, + .num_dsc = 0, +}; + +static const struct resource_caps res_cap_rn_FPGA_2pipe_dsc = { + .num_timing_generator = 2, + .num_opp = 2, + .num_video_plane = 2, + .num_audio = 7, + .num_stream_encoder = 2, + .num_pll = 4, + .num_dwb = 1, + .num_ddc = 4, + .num_dsc = 2, +}; +#endif + +static const struct dc_plane_cap plane_cap = { + .type = DC_PLANE_TYPE_DCN_UNIVERSAL, + .blends_with_above = true, + .blends_with_below = true, + .per_pixel_alpha = true, + + .pixel_format_support = { + .argb8888 = true, + .nv12 = true, + .fp16 = true + }, + + .max_upscale_factor = { + .argb8888 = 16000, + .nv12 = 16000, + .fp16 = 16000 + }, + + .max_downscale_factor = { + .argb8888 = 250, + .nv12 = 250, + .fp16 = 250 + } +}; + +static const struct dc_debug_options debug_defaults_drv = { + .disable_dmcu = true, + .force_abm_enable = false, + .timing_trace = false, + .clock_trace = true, + .disable_pplib_clock_request = true, + .pipe_split_policy = MPC_SPLIT_AVOID_MULT_DISP, + .force_single_disp_pipe_split = false, + .disable_dcc = DCC_ENABLE, + .vsr_support = true, + .performance_trace = false, + .max_downscale_src_width = 4096, + .disable_pplib_wm_range = false, + .scl_reset_length10 = true, + .sanity_checks = true, + .disable_48mhz_pwrdwn = false, + .nv12_iflip_vm_wa = true, + .usbc_combo_phy_reset_wa = true +}; + +static const struct dc_debug_options debug_defaults_diags = { + .disable_dmcu = true, + .force_abm_enable = false, + .timing_trace = true, + .clock_trace = true, + .disable_dpp_power_gate = true, + .disable_hubp_power_gate = true, + .disable_clock_gate = true, + .disable_pplib_clock_request = true, + .disable_pplib_wm_range = true, + .disable_stutter = true, + .disable_48mhz_pwrdwn = true, +}; + +enum dcn20_clk_src_array_id { + DCN20_CLK_SRC_PLL0, + DCN20_CLK_SRC_PLL1, + DCN20_CLK_SRC_PLL2, + DCN20_CLK_SRC_TOTAL_DCN21 +}; + +static void dcn21_resource_destruct(struct dcn21_resource_pool *pool) +{ + unsigned int i; + + for (i = 0; i < pool->base.stream_enc_count; i++) { + if (pool->base.stream_enc[i] != NULL) { + kfree(DCN10STRENC_FROM_STRENC(pool->base.stream_enc[i])); + pool->base.stream_enc[i] = NULL; + } + } + + for (i = 0; i < pool->base.res_cap->num_dsc; i++) { + if (pool->base.dscs[i] != NULL) + dcn20_dsc_destroy(&pool->base.dscs[i]); + } + + if (pool->base.mpc != NULL) { + kfree(TO_DCN20_MPC(pool->base.mpc)); + pool->base.mpc = NULL; + } + if (pool->base.hubbub != NULL) { + kfree(pool->base.hubbub); + pool->base.hubbub = NULL; + } + for (i = 0; i < pool->base.pipe_count; i++) { + if (pool->base.dpps[i] != NULL) + dcn20_dpp_destroy(&pool->base.dpps[i]); + + if (pool->base.ipps[i] != NULL) + pool->base.ipps[i]->funcs->ipp_destroy(&pool->base.ipps[i]); + + if (pool->base.hubps[i] != NULL) { + kfree(TO_DCN20_HUBP(pool->base.hubps[i])); + pool->base.hubps[i] = NULL; + } + + if (pool->base.irqs != NULL) { + dal_irq_service_destroy(&pool->base.irqs); + } + } + + for (i = 0; i < pool->base.res_cap->num_ddc; i++) { + if (pool->base.engines[i] != NULL) + dce110_engine_destroy(&pool->base.engines[i]); + if (pool->base.hw_i2cs[i] != NULL) { + kfree(pool->base.hw_i2cs[i]); + pool->base.hw_i2cs[i] = NULL; + } + if (pool->base.sw_i2cs[i] != NULL) { + kfree(pool->base.sw_i2cs[i]); + pool->base.sw_i2cs[i] = NULL; + } + } + + for (i = 0; i < pool->base.res_cap->num_opp; i++) { + if (pool->base.opps[i] != NULL) + pool->base.opps[i]->funcs->opp_destroy(&pool->base.opps[i]); + } + + for (i = 0; i < pool->base.res_cap->num_timing_generator; i++) { + if (pool->base.timing_generators[i] != NULL) { + kfree(DCN10TG_FROM_TG(pool->base.timing_generators[i])); + pool->base.timing_generators[i] = NULL; + } + } + + for (i = 0; i < pool->base.res_cap->num_dwb; i++) { + if (pool->base.dwbc[i] != NULL) { + kfree(TO_DCN20_DWBC(pool->base.dwbc[i])); + pool->base.dwbc[i] = NULL; + } + if (pool->base.mcif_wb[i] != NULL) { + kfree(TO_DCN20_MMHUBBUB(pool->base.mcif_wb[i])); + pool->base.mcif_wb[i] = NULL; + } + } + + for (i = 0; i < pool->base.audio_count; i++) { + if (pool->base.audios[i]) + dce_aud_destroy(&pool->base.audios[i]); + } + + for (i = 0; i < pool->base.clk_src_count; i++) { + if (pool->base.clock_sources[i] != NULL) { + dcn20_clock_source_destroy(&pool->base.clock_sources[i]); + pool->base.clock_sources[i] = NULL; + } + } + + if (pool->base.dp_clock_source != NULL) { + dcn20_clock_source_destroy(&pool->base.dp_clock_source); + pool->base.dp_clock_source = NULL; + } + + + if (pool->base.abm != NULL) + dce_abm_destroy(&pool->base.abm); + + if (pool->base.dmcu != NULL) + dce_dmcu_destroy(&pool->base.dmcu); + + if (pool->base.dccg != NULL) + dcn_dccg_destroy(&pool->base.dccg); + + if (pool->base.pp_smu != NULL) + dcn21_pp_smu_destroy(&pool->base.pp_smu); +} + + +static void calculate_wm_set_for_vlevel( + int vlevel, + struct wm_range_table_entry *table_entry, + struct dcn_watermarks *wm_set, + struct display_mode_lib *dml, + display_e2e_pipe_params_st *pipes, + int pipe_cnt) +{ + double dram_clock_change_latency_cached = dml->soc.dram_clock_change_latency_us; + + ASSERT(vlevel < dml->soc.num_states); + /* only pipe 0 is read for voltage and dcf/soc clocks */ + pipes[0].clks_cfg.voltage = vlevel; + pipes[0].clks_cfg.dcfclk_mhz = dml->soc.clock_limits[vlevel].dcfclk_mhz; + pipes[0].clks_cfg.socclk_mhz = dml->soc.clock_limits[vlevel].socclk_mhz; + + dml->soc.dram_clock_change_latency_us = table_entry->pstate_latency_us; + dml->soc.sr_exit_time_us = table_entry->sr_exit_time_us; + dml->soc.sr_enter_plus_exit_time_us = table_entry->sr_enter_plus_exit_time_us; + + wm_set->urgent_ns = get_wm_urgent(dml, pipes, pipe_cnt) * 1000; + wm_set->cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(dml, pipes, pipe_cnt) * 1000; + wm_set->cstate_pstate.cstate_exit_ns = get_wm_stutter_exit(dml, pipes, pipe_cnt) * 1000; + wm_set->cstate_pstate.pstate_change_ns = get_wm_dram_clock_change(dml, pipes, pipe_cnt) * 1000; + wm_set->pte_meta_urgent_ns = get_wm_memory_trip(dml, pipes, pipe_cnt) * 1000; + wm_set->frac_urg_bw_nom = get_fraction_of_urgent_bandwidth(dml, pipes, pipe_cnt) * 1000; + wm_set->frac_urg_bw_flip = get_fraction_of_urgent_bandwidth_imm_flip(dml, pipes, pipe_cnt) * 1000; + wm_set->urgent_latency_ns = get_urgent_latency(dml, pipes, pipe_cnt) * 1000; + dml->soc.dram_clock_change_latency_us = dram_clock_change_latency_cached; + +} + +static void patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb) +{ + int i; + + DC_FP_START(); + + if (dc->bb_overrides.sr_exit_time_ns) { + for (i = 0; i < WM_SET_COUNT; i++) { + dc->clk_mgr->bw_params->wm_table.entries[i].sr_exit_time_us = + dc->bb_overrides.sr_exit_time_ns / 1000.0; + } + } + + if (dc->bb_overrides.sr_enter_plus_exit_time_ns) { + for (i = 0; i < WM_SET_COUNT; i++) { + dc->clk_mgr->bw_params->wm_table.entries[i].sr_enter_plus_exit_time_us = + dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; + } + } + + if (dc->bb_overrides.urgent_latency_ns) { + bb->urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; + } + + if (dc->bb_overrides.dram_clock_change_latency_ns) { + for (i = 0; i < WM_SET_COUNT; i++) { + dc->clk_mgr->bw_params->wm_table.entries[i].pstate_latency_us = + dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; + } + } + + DC_FP_END(); +} + +void dcn21_calculate_wm( + struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + int *out_pipe_cnt, + int *pipe_split_from, + int vlevel_req) +{ + int pipe_cnt, i, pipe_idx; + int vlevel, vlevel_max; + struct wm_range_table_entry *table_entry; + struct clk_bw_params *bw_params = dc->clk_mgr->bw_params; + + ASSERT(bw_params); + + patch_bounding_box(dc, &context->bw_ctx.dml.soc); + + for (i = 0, pipe_idx = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + pipes[pipe_cnt].clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; + pipes[pipe_cnt].clks_cfg.dispclk_mhz = context->bw_ctx.dml.vba.RequiredDISPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb]; + + if (pipe_split_from[i] < 0) { + pipes[pipe_cnt].clks_cfg.dppclk_mhz = + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx]; + if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_idx] == pipe_idx) + pipes[pipe_cnt].pipe.dest.odm_combine = + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel_req][pipe_idx]; + else + pipes[pipe_cnt].pipe.dest.odm_combine = 0; + pipe_idx++; + } else { + pipes[pipe_cnt].clks_cfg.dppclk_mhz = + context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel_req][context->bw_ctx.dml.vba.maxMpcComb][pipe_split_from[i]]; + if (context->bw_ctx.dml.vba.BlendingAndTiming[pipe_split_from[i]] == pipe_split_from[i]) + pipes[pipe_cnt].pipe.dest.odm_combine = + context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel_req][pipe_split_from[i]]; + else + pipes[pipe_cnt].pipe.dest.odm_combine = 0; + } + pipe_cnt++; + } + + if (pipe_cnt != pipe_idx) { + if (dc->res_pool->funcs->populate_dml_pipes) + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, + context, pipes); + else + pipe_cnt = dcn21_populate_dml_pipes_from_context(dc, + context, pipes); + } + + *out_pipe_cnt = pipe_cnt; + + vlevel_max = bw_params->clk_table.num_entries - 1; + + + /* WM Set D */ + table_entry = &bw_params->wm_table.entries[WM_D]; + if (table_entry->wm_type == WM_TYPE_RETRAINING) + vlevel = 0; + else + vlevel = vlevel_max; + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.d, + &context->bw_ctx.dml, pipes, pipe_cnt); + /* WM Set C */ + table_entry = &bw_params->wm_table.entries[WM_C]; + vlevel = MIN(MAX(vlevel_req, 2), vlevel_max); + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.c, + &context->bw_ctx.dml, pipes, pipe_cnt); + /* WM Set B */ + table_entry = &bw_params->wm_table.entries[WM_B]; + vlevel = MIN(MAX(vlevel_req, 1), vlevel_max); + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.b, + &context->bw_ctx.dml, pipes, pipe_cnt); + + /* WM Set A */ + table_entry = &bw_params->wm_table.entries[WM_A]; + vlevel = MIN(vlevel_req, vlevel_max); + calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.a, + &context->bw_ctx.dml, pipes, pipe_cnt); +} + + +bool dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context, + bool fast_validate) +{ + bool out = false; + + BW_VAL_TRACE_SETUP(); + + int vlevel = 0; + int pipe_split_from[MAX_PIPES]; + int pipe_cnt = 0; + display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_KERNEL); + DC_LOGGER_INIT(dc->ctx->logger); + + BW_VAL_TRACE_COUNT(); + + out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel); + + if (pipe_cnt == 0) + goto validate_out; + + if (!out) + goto validate_fail; + + BW_VAL_TRACE_END_VOLTAGE_LEVEL(); + + if (fast_validate) { + BW_VAL_TRACE_SKIP(fast); + goto validate_out; + } + + dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel); + dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + + BW_VAL_TRACE_END_WATERMARKS(); + + goto validate_out; + +validate_fail: + DC_LOG_WARNING("Mode Validation Warning: %s failed validation.\n", + dml_get_status_message(context->bw_ctx.dml.vba.ValidationStatus[context->bw_ctx.dml.vba.soc.num_states])); + + BW_VAL_TRACE_SKIP(fail); + out = false; + +validate_out: + kfree(pipes); + + BW_VAL_TRACE_FINISH(); + + return out; +} +static void dcn21_destroy_resource_pool(struct resource_pool **pool) +{ + struct dcn21_resource_pool *dcn21_pool = TO_DCN21_RES_POOL(*pool); + + dcn21_resource_destruct(dcn21_pool); + kfree(dcn21_pool); + *pool = NULL; +} + +static struct clock_source *dcn21_clock_source_create( + struct dc_context *ctx, + struct dc_bios *bios, + enum clock_source_id id, + const struct dce110_clk_src_regs *regs, + bool dp_clk_src) +{ + struct dce110_clk_src *clk_src = + kzalloc(sizeof(struct dce110_clk_src), GFP_KERNEL); + + if (!clk_src) + return NULL; + + if (dcn20_clk_src_construct(clk_src, ctx, bios, id, + regs, &cs_shift, &cs_mask)) { + clk_src->base.dp_clk_src = dp_clk_src; + return &clk_src->base; + } + + BREAK_TO_DEBUGGER(); + return NULL; +} + +static struct hubp *dcn21_hubp_create( + struct dc_context *ctx, + uint32_t inst) +{ + struct dcn21_hubp *hubp21 = + kzalloc(sizeof(struct dcn21_hubp), GFP_KERNEL); + + if (!hubp21) + return NULL; + + if (hubp21_construct(hubp21, ctx, inst, + &hubp_regs[inst], &hubp_shift, &hubp_mask)) + return &hubp21->base; + + BREAK_TO_DEBUGGER(); + kfree(hubp21); + return NULL; +} + +static struct hubbub *dcn21_hubbub_create(struct dc_context *ctx) +{ + int i; + + struct dcn20_hubbub *hubbub = kzalloc(sizeof(struct dcn20_hubbub), + GFP_KERNEL); + + if (!hubbub) + return NULL; + + hubbub21_construct(hubbub, ctx, + &hubbub_reg, + &hubbub_shift, + &hubbub_mask); + + for (i = 0; i < res_cap_rn.num_vmid; i++) { + struct dcn20_vmid *vmid = &hubbub->vmid[i]; + + vmid->ctx = ctx; + + vmid->regs = &vmid_regs[i]; + vmid->shifts = &vmid_shifts; + vmid->masks = &vmid_masks; + } + + return &hubbub->base; +} + +struct output_pixel_processor *dcn21_opp_create( + struct dc_context *ctx, uint32_t inst) +{ + struct dcn20_opp *opp = + kzalloc(sizeof(struct dcn20_opp), GFP_KERNEL); + + if (!opp) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dcn20_opp_construct(opp, ctx, inst, + &opp_regs[inst], &opp_shift, &opp_mask); + return &opp->base; +} + +struct timing_generator *dcn21_timing_generator_create( + struct dc_context *ctx, + uint32_t instance) +{ + struct optc *tgn10 = + kzalloc(sizeof(struct optc), GFP_KERNEL); + + if (!tgn10) + return NULL; + + tgn10->base.inst = instance; + tgn10->base.ctx = ctx; + + tgn10->tg_regs = &tg_regs[instance]; + tgn10->tg_shift = &tg_shift; + tgn10->tg_mask = &tg_mask; + + dcn20_timing_generator_init(tgn10); + + return &tgn10->base; +} + +struct mpc *dcn21_mpc_create(struct dc_context *ctx) +{ + struct dcn20_mpc *mpc20 = kzalloc(sizeof(struct dcn20_mpc), + GFP_KERNEL); + + if (!mpc20) + return NULL; + + dcn20_mpc_construct(mpc20, ctx, + &mpc_regs, + &mpc_shift, + &mpc_mask, + 6); + + return &mpc20->base; +} + +static void read_dce_straps( + struct dc_context *ctx, + struct resource_straps *straps) +{ + generic_reg_get(ctx, mmDC_PINSTRAPS + BASE(mmDC_PINSTRAPS_BASE_IDX), + FN(DC_PINSTRAPS, DC_PINSTRAPS_AUDIO), &straps->dc_pinstraps_audio); + +} + + +struct display_stream_compressor *dcn21_dsc_create( + struct dc_context *ctx, uint32_t inst) +{ + struct dcn20_dsc *dsc = + kzalloc(sizeof(struct dcn20_dsc), GFP_KERNEL); + + if (!dsc) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dsc2_construct(dsc, ctx, inst, &dsc_regs[inst], &dsc_shift, &dsc_mask); + return &dsc->base; +} + +static void update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) +{ + struct dcn21_resource_pool *pool = TO_DCN21_RES_POOL(dc->res_pool); + struct clk_limit_table *clk_table = &bw_params->clk_table; + int i; + + dcn2_1_ip.max_num_otg = pool->base.res_cap->num_timing_generator; + dcn2_1_ip.max_num_dpp = pool->base.pipe_count; + dcn2_1_soc.num_chans = bw_params->num_channels; + + for (i = 0; i < clk_table->num_entries; i++) { + + dcn2_1_soc.clock_limits[i].state = i; + dcn2_1_soc.clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; + dcn2_1_soc.clock_limits[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; + dcn2_1_soc.clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz; + dcn2_1_soc.clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; + } + dcn2_1_soc.clock_limits[i] = dcn2_1_soc.clock_limits[i - 1]; + dcn2_1_soc.num_states = i; + + // diags does not retrieve proper values from SMU, do not update DML instance for diags + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) && !IS_DIAG_DC(dc->ctx->dce_environment)) + dml_init_instance(&dc->dml, &dcn2_1_soc, &dcn2_1_ip, DML_PROJECT_DCN21); +} + +/* Temporary Place holder until we can get them from fuse */ +static struct dpm_clocks dummy_clocks = { + .DcfClocks = { + {.Freq = 400, .Vol = 1}, + {.Freq = 483, .Vol = 1}, + {.Freq = 602, .Vol = 1}, + {.Freq = 738, .Vol = 1} }, + .SocClocks = { + {.Freq = 300, .Vol = 1}, + {.Freq = 400, .Vol = 1}, + {.Freq = 400, .Vol = 1}, + {.Freq = 400, .Vol = 1} }, + .FClocks = { + {.Freq = 400, .Vol = 1}, + {.Freq = 800, .Vol = 1}, + {.Freq = 1067, .Vol = 1}, + {.Freq = 1600, .Vol = 1} }, + .MemClocks = { + {.Freq = 800, .Vol = 1}, + {.Freq = 1600, .Vol = 1}, + {.Freq = 1067, .Vol = 1}, + {.Freq = 1600, .Vol = 1} }, + +}; + +static enum pp_smu_status dummy_set_wm_ranges(struct pp_smu *pp, + struct pp_smu_wm_range_sets *ranges) +{ + return PP_SMU_RESULT_OK; +} + +static enum pp_smu_status dummy_get_dpm_clock_table(struct pp_smu *pp, + struct dpm_clocks *clock_table) +{ + *clock_table = dummy_clocks; + return PP_SMU_RESULT_OK; +} + +static struct pp_smu_funcs *dcn21_pp_smu_create(struct dc_context *ctx) +{ + struct pp_smu_funcs *pp_smu = kzalloc(sizeof(*pp_smu), GFP_KERNEL); + + if (!pp_smu) + return pp_smu; + + if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment) || IS_DIAG_DC(ctx->dce_environment)) { + pp_smu->ctx.ver = PP_SMU_VER_RN; + pp_smu->rn_funcs.get_dpm_clock_table = dummy_get_dpm_clock_table; + pp_smu->rn_funcs.set_wm_ranges = dummy_set_wm_ranges; + } else { + + dm_pp_get_funcs(ctx, pp_smu); + + if (pp_smu->ctx.ver != PP_SMU_VER_RN) + pp_smu = memset(pp_smu, 0, sizeof(struct pp_smu_funcs)); + } + + return pp_smu; +} + +static void dcn21_pp_smu_destroy(struct pp_smu_funcs **pp_smu) +{ + if (pp_smu && *pp_smu) { + kfree(*pp_smu); + *pp_smu = NULL; + } +} + +static struct audio *dcn21_create_audio( + struct dc_context *ctx, unsigned int inst) +{ + return dce_audio_create(ctx, inst, + &audio_regs[inst], &audio_shift, &audio_mask); +} + +static struct dc_cap_funcs cap_funcs = { + .get_dcc_compression_cap = dcn20_get_dcc_compression_cap +}; + +struct stream_encoder *dcn21_stream_encoder_create( + enum engine_id eng_id, + struct dc_context *ctx) +{ + struct dcn10_stream_encoder *enc1 = + kzalloc(sizeof(struct dcn10_stream_encoder), GFP_KERNEL); + + if (!enc1) + return NULL; + + dcn20_stream_encoder_construct(enc1, ctx, ctx->dc_bios, eng_id, + &stream_enc_regs[eng_id], + &se_shift, &se_mask); + + return &enc1->base; +} + +static const struct dce_hwseq_registers hwseq_reg = { + HWSEQ_DCN21_REG_LIST() +}; + +static const struct dce_hwseq_shift hwseq_shift = { + HWSEQ_DCN21_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_hwseq_mask hwseq_mask = { + HWSEQ_DCN21_MASK_SH_LIST(_MASK) +}; + +static struct dce_hwseq *dcn21_hwseq_create( + struct dc_context *ctx) +{ + struct dce_hwseq *hws = kzalloc(sizeof(struct dce_hwseq), GFP_KERNEL); + + if (hws) { + hws->ctx = ctx; + hws->regs = &hwseq_reg; + hws->shifts = &hwseq_shift; + hws->masks = &hwseq_mask; + hws->wa.DEGVIDCN21 = true; + } + return hws; +} + +static const struct resource_create_funcs res_create_funcs = { + .read_dce_straps = read_dce_straps, + .create_audio = dcn21_create_audio, + .create_stream_encoder = dcn21_stream_encoder_create, + .create_hwseq = dcn21_hwseq_create, +}; + +static const struct resource_create_funcs res_create_maximus_funcs = { + .read_dce_straps = NULL, + .create_audio = NULL, + .create_stream_encoder = NULL, + .create_hwseq = dcn21_hwseq_create, +}; + +static const struct encoder_feature_support link_enc_feature = { + .max_hdmi_deep_color = COLOR_DEPTH_121212, + .max_hdmi_pixel_clock = 600000, + .hdmi_ycbcr420_supported = true, + .dp_ycbcr420_supported = true, + .flags.bits.IS_HBR2_CAPABLE = true, + .flags.bits.IS_HBR3_CAPABLE = true, + .flags.bits.IS_TPS3_CAPABLE = true, + .flags.bits.IS_TPS4_CAPABLE = true +}; + + +#define link_regs(id, phyid)\ +[id] = {\ + LE_DCN2_REG_LIST(id), \ + UNIPHY_DCN2_REG_LIST(phyid), \ + DPCS_DCN21_REG_LIST(id), \ + SRI(DP_DPHY_INTERNAL_CTRL, DP, id) \ +} + +static const struct dcn10_link_enc_registers link_enc_regs[] = { + link_regs(0, A), + link_regs(1, B), + link_regs(2, C), + link_regs(3, D), + link_regs(4, E), +}; + +#define aux_regs(id)\ +[id] = {\ + DCN2_AUX_REG_LIST(id)\ +} + +static const struct dcn10_link_enc_aux_registers link_enc_aux_regs[] = { + aux_regs(0), + aux_regs(1), + aux_regs(2), + aux_regs(3), + aux_regs(4) +}; + +#define hpd_regs(id)\ +[id] = {\ + HPD_REG_LIST(id)\ +} + +static const struct dcn10_link_enc_hpd_registers link_enc_hpd_regs[] = { + hpd_regs(0), + hpd_regs(1), + hpd_regs(2), + hpd_regs(3), + hpd_regs(4) +}; + +static const struct dcn10_link_enc_shift le_shift = { + LINK_ENCODER_MASK_SH_LIST_DCN20(__SHIFT),\ + DPCS_DCN21_MASK_SH_LIST(__SHIFT) +}; + +static const struct dcn10_link_enc_mask le_mask = { + LINK_ENCODER_MASK_SH_LIST_DCN20(_MASK),\ + DPCS_DCN21_MASK_SH_LIST(_MASK) +}; + +static int map_transmitter_id_to_phy_instance( + enum transmitter transmitter) +{ + switch (transmitter) { + case TRANSMITTER_UNIPHY_A: + return 0; + break; + case TRANSMITTER_UNIPHY_B: + return 1; + break; + case TRANSMITTER_UNIPHY_C: + return 2; + break; + case TRANSMITTER_UNIPHY_D: + return 3; + break; + case TRANSMITTER_UNIPHY_E: + return 4; + break; + default: + ASSERT(0); + return 0; + } +} + +static struct link_encoder *dcn21_link_encoder_create( + const struct encoder_init_data *enc_init_data) +{ + struct dcn21_link_encoder *enc21 = + kzalloc(sizeof(struct dcn21_link_encoder), GFP_KERNEL); + int link_regs_id; + + if (!enc21) + return NULL; + + link_regs_id = + map_transmitter_id_to_phy_instance(enc_init_data->transmitter); + + dcn21_link_encoder_construct(enc21, + enc_init_data, + &link_enc_feature, + &link_enc_regs[link_regs_id], + &link_enc_aux_regs[enc_init_data->channel - 1], + &link_enc_hpd_regs[enc_init_data->hpd_source], + &le_shift, + &le_mask); + + return &enc21->enc10.base; +} +#define CTX ctx + +#define REG(reg_name) \ + (DCN_BASE.instance[0].segment[mm ## reg_name ## _BASE_IDX] + mm ## reg_name) + +static uint32_t read_pipe_fuses(struct dc_context *ctx) +{ + uint32_t value = REG_READ(CC_DC_PIPE_DIS); + /* RV1 support max 4 pipes */ + value = value & 0xf; + return value; +} + +static int dcn21_populate_dml_pipes_from_context( + struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes) +{ + uint32_t pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes); + int i; + struct resource_context *res_ctx = &context->res_ctx; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + + if (!res_ctx->pipe_ctx[i].stream) + continue; + + pipes[i].pipe.src.hostvm = 1; + pipes[i].pipe.src.gpuvm = 1; + } + + return pipe_cnt; +} + +static struct resource_funcs dcn21_res_pool_funcs = { + .destroy = dcn21_destroy_resource_pool, + .link_enc_create = dcn21_link_encoder_create, + .validate_bandwidth = dcn21_validate_bandwidth, + .populate_dml_pipes = dcn21_populate_dml_pipes_from_context, + .add_stream_to_ctx = dcn20_add_stream_to_ctx, + .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, + .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .populate_dml_writeback_from_context = dcn20_populate_dml_writeback_from_context, + .get_default_swizzle_mode = dcn20_get_default_swizzle_mode, + .set_mcif_arb_params = dcn20_set_mcif_arb_params, + .find_first_free_match_stream_enc_for_link = dcn10_find_first_free_match_stream_enc_for_link, + .update_bw_bounding_box = update_bw_bounding_box +}; + +static bool dcn21_resource_construct( + uint8_t num_virtual_links, + struct dc *dc, + struct dcn21_resource_pool *pool) +{ + int i, j; + struct dc_context *ctx = dc->ctx; + struct irq_service_init_data init_data; + uint32_t pipe_fuses = read_pipe_fuses(ctx); + uint32_t num_pipes; + + ctx->dc_bios->regs = &bios_regs; + + pool->base.res_cap = &res_cap_rn; +#ifdef DIAGS_BUILD + if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) + //pool->base.res_cap = &res_cap_nv10_FPGA_2pipe_dsc; + pool->base.res_cap = &res_cap_rn_FPGA_4pipe; +#endif + + pool->base.funcs = &dcn21_res_pool_funcs; + + /************************************************* + * Resource + asic cap harcoding * + *************************************************/ + pool->base.underlay_pipe_index = NO_UNDERLAY_PIPE; + + /* max pipe num for ASIC before check pipe fuses */ + pool->base.pipe_count = pool->base.res_cap->num_timing_generator; + + dc->caps.max_downscale_ratio = 200; + dc->caps.i2c_speed_in_khz = 100; + dc->caps.max_cursor_size = 256; + dc->caps.dmdata_alloc_size = 2048; + dc->caps.hw_3d_lut = true; + + dc->caps.max_slave_planes = 1; + dc->caps.post_blend_color_processing = true; + dc->caps.force_dp_tps4_for_cp2520 = true; + dc->caps.extended_aux_timeout_support = true; + dc->caps.dmcub_support = true; + + if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) + dc->debug = debug_defaults_drv; + else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { + pool->base.pipe_count = 4; + dc->debug = debug_defaults_diags; + } else + dc->debug = debug_defaults_diags; + + // Init the vm_helper + if (dc->vm_helper) + vm_helper_init(dc->vm_helper, 16); + + /************************************************* + * Create resources * + *************************************************/ + + pool->base.clock_sources[DCN20_CLK_SRC_PLL0] = + dcn21_clock_source_create(ctx, ctx->dc_bios, + CLOCK_SOURCE_COMBO_PHY_PLL0, + &clk_src_regs[0], false); + pool->base.clock_sources[DCN20_CLK_SRC_PLL1] = + dcn21_clock_source_create(ctx, ctx->dc_bios, + CLOCK_SOURCE_COMBO_PHY_PLL1, + &clk_src_regs[1], false); + pool->base.clock_sources[DCN20_CLK_SRC_PLL2] = + dcn21_clock_source_create(ctx, ctx->dc_bios, + CLOCK_SOURCE_COMBO_PHY_PLL2, + &clk_src_regs[2], false); + + pool->base.clk_src_count = DCN20_CLK_SRC_TOTAL_DCN21; + + /* todo: not reuse phy_pll registers */ + pool->base.dp_clock_source = + dcn21_clock_source_create(ctx, ctx->dc_bios, + CLOCK_SOURCE_ID_DP_DTO, + &clk_src_regs[0], true); + + for (i = 0; i < pool->base.clk_src_count; i++) { + if (pool->base.clock_sources[i] == NULL) { + dm_error("DC: failed to create clock sources!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + } + + pool->base.dccg = dccg2_create(ctx, &dccg_regs, &dccg_shift, &dccg_mask); + if (pool->base.dccg == NULL) { + dm_error("DC: failed to create dccg!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + + pool->base.dmcu = dcn21_dmcu_create(ctx, + &dmcu_regs, + &dmcu_shift, + &dmcu_mask); + if (pool->base.dmcu == NULL) { + dm_error("DC: failed to create dmcu!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + + // Leave as NULL to not affect current dmcu psr programming sequence + // Will be uncommented when functionality is confirmed to be working + pool->base.psr = NULL; + + pool->base.abm = dce_abm_create(ctx, + &abm_regs, + &abm_shift, + &abm_mask); + if (pool->base.abm == NULL) { + dm_error("DC: failed to create abm!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + + pool->base.pp_smu = dcn21_pp_smu_create(ctx); + + num_pipes = dcn2_1_ip.max_num_dpp; + + for (i = 0; i < dcn2_1_ip.max_num_dpp; i++) + if (pipe_fuses & 1 << i) + num_pipes--; + dcn2_1_ip.max_num_dpp = num_pipes; + dcn2_1_ip.max_num_otg = num_pipes; + + dml_init_instance(&dc->dml, &dcn2_1_soc, &dcn2_1_ip, DML_PROJECT_DCN21); + + init_data.ctx = dc->ctx; + pool->base.irqs = dal_irq_service_dcn21_create(&init_data); + if (!pool->base.irqs) + goto create_fail; + + j = 0; + /* mem input -> ipp -> dpp -> opp -> TG */ + for (i = 0; i < pool->base.pipe_count; i++) { + /* if pipe is disabled, skip instance of HW pipe, + * i.e, skip ASIC register instance + */ + if ((pipe_fuses & (1 << i)) != 0) + continue; + + pool->base.hubps[j] = dcn21_hubp_create(ctx, i); + if (pool->base.hubps[j] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC: failed to create memory input!\n"); + goto create_fail; + } + + pool->base.ipps[j] = dcn21_ipp_create(ctx, i); + if (pool->base.ipps[j] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC: failed to create input pixel processor!\n"); + goto create_fail; + } + + pool->base.dpps[j] = dcn21_dpp_create(ctx, i); + if (pool->base.dpps[j] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC: failed to create dpps!\n"); + goto create_fail; + } + + pool->base.opps[j] = dcn21_opp_create(ctx, i); + if (pool->base.opps[j] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC: failed to create output pixel processor!\n"); + goto create_fail; + } + + pool->base.timing_generators[j] = dcn21_timing_generator_create( + ctx, i); + if (pool->base.timing_generators[j] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create tg!\n"); + goto create_fail; + } + j++; + } + + for (i = 0; i < pool->base.res_cap->num_ddc; i++) { + pool->base.engines[i] = dcn21_aux_engine_create(ctx, i); + if (pool->base.engines[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC:failed to create aux engine!!\n"); + goto create_fail; + } + pool->base.hw_i2cs[i] = dcn21_i2c_hw_create(ctx, i); + if (pool->base.hw_i2cs[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error( + "DC:failed to create hw i2c!!\n"); + goto create_fail; + } + pool->base.sw_i2cs[i] = NULL; + } + + pool->base.timing_generator_count = j; + pool->base.pipe_count = j; + pool->base.mpcc_count = j; + + pool->base.mpc = dcn21_mpc_create(ctx); + if (pool->base.mpc == NULL) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create mpc!\n"); + goto create_fail; + } + + pool->base.hubbub = dcn21_hubbub_create(ctx); + if (pool->base.hubbub == NULL) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create hubbub!\n"); + goto create_fail; + } + + for (i = 0; i < pool->base.res_cap->num_dsc; i++) { + pool->base.dscs[i] = dcn21_dsc_create(ctx, i); + if (pool->base.dscs[i] == NULL) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create display stream compressor %d!\n", i); + goto create_fail; + } + } + + if (!dcn20_dwbc_create(ctx, &pool->base)) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create dwbc!\n"); + goto create_fail; + } + if (!dcn20_mmhubbub_create(ctx, &pool->base)) { + BREAK_TO_DEBUGGER(); + dm_error("DC: failed to create mcif_wb!\n"); + goto create_fail; + } + + if (!resource_construct(num_virtual_links, dc, &pool->base, + (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? + &res_create_funcs : &res_create_maximus_funcs))) + goto create_fail; + + dcn21_hw_sequencer_construct(dc); + + dc->caps.max_planes = pool->base.pipe_count; + + for (i = 0; i < dc->caps.max_planes; ++i) + dc->caps.planes[i] = plane_cap; + + dc->cap_funcs = cap_funcs; + + return true; + +create_fail: + + dcn21_resource_destruct(pool); + + return false; +} + +struct resource_pool *dcn21_create_resource_pool( + const struct dc_init_data *init_data, + struct dc *dc) +{ + struct dcn21_resource_pool *pool = + kzalloc(sizeof(struct dcn21_resource_pool), GFP_KERNEL); + + if (!pool) + return NULL; + + if (dcn21_resource_construct(init_data->num_virtual_links, dc, pool)) + return &pool->base; + + BREAK_TO_DEBUGGER(); + kfree(pool); + return NULL; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h new file mode 100644 index 000000000000..a27355171bca --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.h @@ -0,0 +1,45 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DCN21_RESOURCE_H_ +#define _DCN21_RESOURCE_H_ + +#include "core_types.h" + +#define TO_DCN21_RES_POOL(pool)\ + container_of(pool, struct dcn21_resource_pool, base) + +struct dc; +struct resource_pool; +struct _vcs_dpi_display_pipe_params_st; + +struct dcn21_resource_pool { + struct resource_pool base; +}; +struct resource_pool *dcn21_create_resource_pool( + const struct dc_init_data *init_data, + struct dc *dc); + +#endif /* _DCN21_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h b/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h new file mode 100644 index 000000000000..626d22d437f4 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h @@ -0,0 +1,49 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef DM_CP_PSP_IF__H +#define DM_CP_PSP_IF__H + +struct dc_link; + +struct cp_psp_stream_config { + uint8_t otg_inst; + uint8_t link_enc_inst; + uint8_t stream_enc_inst; + void *dm_stream_ctx; + bool dpms_off; +}; + +struct cp_psp_funcs { + void (*update_stream_config)(void *handle, struct cp_psp_stream_config *config); +}; + +struct cp_psp { + void *handle; + struct cp_psp_funcs funcs; +}; + + +#endif /* DM_CP_PSP_IF__H */ diff --git a/drivers/gpu/drm/amd/display/dc/dm_helpers.h b/drivers/gpu/drm/amd/display/dc/dm_helpers.h index b6b4333737f2..8bde1d688f2e 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_helpers.h +++ b/drivers/gpu/drm/amd/display/dc/dm_helpers.h @@ -74,7 +74,7 @@ void dm_helpers_dp_mst_clear_payload_allocation_table( /* * Polls for ACT (allocation change trigger) handled and */ -bool dm_helpers_dp_mst_poll_for_allocation_change_trigger( +enum act_return_status dm_helpers_dp_mst_poll_for_allocation_change_trigger( struct dc_context *ctx, const struct dc_stream_state *stream); /* @@ -118,13 +118,11 @@ bool dm_helpers_submit_i2c( const struct dc_link *link, struct i2c_command *cmd); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT bool dm_helpers_dp_write_dsc_enable( struct dc_context *ctx, const struct dc_stream_state *stream, bool enable ); -#endif bool dm_helpers_is_dp_sink_present( struct dc_link *link); diff --git a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h index 680689cab5dd..ae608c329366 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h +++ b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h @@ -41,9 +41,8 @@ enum pp_smu_ver { */ PP_SMU_UNSUPPORTED, PP_SMU_VER_RV, -#ifndef CONFIG_TRIM_DRM_AMD_DC_DCN2_0 PP_SMU_VER_NV, -#endif + PP_SMU_VER_RN, PP_SMU_VER_MAX }; @@ -140,7 +139,6 @@ struct pp_smu_funcs_rv { void (*set_pme_wa_enable)(struct pp_smu *pp); }; -#ifndef CONFIG_TRIM_DRM_AMD_DC_DCN2_0 /* Used by pp_smu_funcs_nv.set_voltage_by_freq * */ @@ -244,15 +242,51 @@ struct pp_smu_funcs_nv { enum pp_smu_status (*set_pstate_handshake_support)(struct pp_smu *pp, BOOLEAN pstate_handshake_supported); }; -#endif + +#define PP_SMU_NUM_SOCCLK_DPM_LEVELS 8 +#define PP_SMU_NUM_DCFCLK_DPM_LEVELS 8 +#define PP_SMU_NUM_FCLK_DPM_LEVELS 4 +#define PP_SMU_NUM_MEMCLK_DPM_LEVELS 4 + +struct dpm_clock { + uint32_t Freq; // In MHz + uint32_t Vol; // Millivolts with 2 fractional bits +}; + + +/* this is a copy of the structure defined in smuxx_driver_if.h*/ +struct dpm_clocks { + struct dpm_clock DcfClocks[PP_SMU_NUM_DCFCLK_DPM_LEVELS]; + struct dpm_clock SocClocks[PP_SMU_NUM_SOCCLK_DPM_LEVELS]; + struct dpm_clock FClocks[PP_SMU_NUM_FCLK_DPM_LEVELS]; + struct dpm_clock MemClocks[PP_SMU_NUM_MEMCLK_DPM_LEVELS]; +}; + + +struct pp_smu_funcs_rn { + struct pp_smu pp_smu; + + /* + * reader and writer WM's are sent together as part of one table + * + * PPSMC_MSG_SetDriverDramAddrHigh + * PPSMC_MSG_SetDriverDramAddrLow + * PPSMC_MSG_TransferTableDram2Smu + * + */ + enum pp_smu_status (*set_wm_ranges)(struct pp_smu *pp, + struct pp_smu_wm_range_sets *ranges); + + enum pp_smu_status (*get_dpm_clock_table) (struct pp_smu *pp, + struct dpm_clocks *clock_table); +}; struct pp_smu_funcs { struct pp_smu ctx; union { struct pp_smu_funcs_rv rv_funcs; -#ifndef CONFIG_TRIM_DRM_AMD_DC_DCN2_0 struct pp_smu_funcs_nv nv_funcs; -#endif + struct pp_smu_funcs_rn rn_funcs; }; }; diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h index b426ba02b793..968ff1fef486 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_services.h +++ b/drivers/gpu/drm/amd/display/dc/dm_services.h @@ -40,6 +40,9 @@ #undef DEPRECATED +struct dmub_srv; +struct dc_dmub_srv; + irq_handler_idx dm_register_interrupt( struct dc_context *ctx, struct dc_interrupt_params *int_params, @@ -139,6 +142,13 @@ uint32_t generic_reg_update_ex(const struct dc_context *ctx, uint32_t addr, int n, uint8_t shift1, uint32_t mask1, uint32_t field_value1, ...); +struct dc_dmub_srv *dc_dmub_srv_create(struct dc *dc, struct dmub_srv *dmub); +void dc_dmub_srv_destroy(struct dc_dmub_srv **dmub_srv); + +void reg_sequence_start_gather(const struct dc_context *ctx); +void reg_sequence_start_execute(const struct dc_context *ctx); +void reg_sequence_wait_done(const struct dc_context *ctx); + #define FD(reg_field) reg_field ## __SHIFT, \ reg_field ## _MASK @@ -151,6 +161,7 @@ void generic_reg_wait(const struct dc_context *ctx, unsigned int delay_between_poll_us, unsigned int time_out_num_tries, const char *func_name, int line); +unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...); /* These macros need to be used with soc15 registers in order to retrieve * the actual offset. diff --git a/drivers/gpu/drm/amd/display/dc/dm_services_types.h b/drivers/gpu/drm/amd/display/dc/dm_services_types.h index a3d1be20dd9d..b52ba6ffabe1 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_services_types.h +++ b/drivers/gpu/drm/amd/display/dc/dm_services_types.h @@ -220,6 +220,7 @@ struct dm_bl_data_point { }; /* Total size of the structure should not exceed 256 bytes */ +#define BL_DATA_POINTS 99 struct dm_acpi_atif_backlight_caps { uint16_t size; /* Bytes 0-1 (2 bytes) */ uint16_t flags; /* Byted 2-3 (2 bytes) */ @@ -229,7 +230,7 @@ struct dm_acpi_atif_backlight_caps { uint8_t min_input_signal; /* Byte 7 */ uint8_t max_input_signal; /* Byte 8 */ uint8_t num_data_points; /* Byte 9 */ - struct dm_bl_data_point data_points[99]; /* Bytes 10-207 (198 bytes)*/ + struct dm_bl_data_point data_points[BL_DATA_POINTS]; /* Bytes 10-207 (198 bytes)*/ }; enum dm_acpi_display_type { diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index 0bb7a20675c4..7ee8b8460a9b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -1,5 +1,6 @@ # # Copyright 2017 Advanced Micro Devices, Inc. +# Copyright 2019 Raptor Engineering, LLC # # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the "Software"), @@ -24,35 +25,56 @@ # It provides the general basic services required by other DAL # subcomponents. -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),) - cc_stack_align := -mpreferred-stack-boundary=4 -else ifneq ($(call cc-option, -mstack-alignment=16),) - cc_stack_align := -mstack-alignment=16 +ifdef CONFIG_X86 +dml_ccflags := -mhard-float -msse endif -dml_ccflags := -mhard-float -msse $(cc_stack_align) +ifdef CONFIG_PPC64 +dml_ccflags := -mhard-float -maltivec +endif -CFLAGS_display_mode_lib.o := $(dml_ccflags) +ifdef CONFIG_CC_IS_GCC +ifeq ($(call cc-ifversion, -lt, 0701, y), y) +IS_OLD_GCC = 1 +endif +endif -ifdef CONFIG_DRM_AMD_DC_DCN2_0 -CFLAGS_display_mode_vba.o := $(dml_ccflags) -CFLAGS_display_mode_vba_20.o := $(dml_ccflags) -CFLAGS_display_rq_dlg_calc_20.o := $(dml_ccflags) +ifdef CONFIG_X86 +ifdef IS_OLD_GCC +# Stack alignment mismatch, proceed with caution. +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 +# (8B stack alignment). +dml_ccflags += -mpreferred-stack-boundary=4 +else +dml_ccflags += -msse2 endif -ifdef CONFIG_DRM_AMD_DCN3AG -CFLAGS_display_mode_vba_3ag.o := $(dml_ccflags) endif -CFLAGS_dml1_display_rq_dlg_calc.o := $(dml_ccflags) -CFLAGS_display_rq_dlg_helpers.o := $(dml_ccflags) -CFLAGS_dml_common_defs.o := $(dml_ccflags) + +CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags) + +ifdef CONFIG_DRM_AMD_DC_DCN +CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20v2.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_mode_vba_21.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_rq_dlg_calc_21.o := $(dml_ccflags) +endif +CFLAGS_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dml_common_defs.o := $(dml_ccflags) DML = display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o \ dml_common_defs.o -ifdef CONFIG_DRM_AMD_DC_DCN2_0 +ifdef CONFIG_DRM_AMD_DC_DCN DML += display_mode_vba.o dcn20/display_rq_dlg_calc_20.o dcn20/display_mode_vba_20.o +DML += dcn20/display_rq_dlg_calc_20v2.o dcn20/display_mode_vba_20v2.o +DML += dcn21/display_rq_dlg_calc_21.o dcn21/display_mode_vba_21.o endif + AMD_DAL_DML = $(addprefix $(AMDDALPATH)/dc/dml/,$(DML)) AMD_DISPLAY_FILES += $(AMD_DAL_DML) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20.c index 649883777f62..45f028986a8d 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20.c @@ -38,6 +38,7 @@ #define BPP_INVALID 0 #define BPP_BLENDED_PIPE 0xffffffff +#define DCN20_MAX_420_IMAGE_WIDTH 4096 static double adjust_ReturnBW( struct display_mode_lib *mode_lib, @@ -937,7 +938,7 @@ static unsigned int CalculateVMAndRowBytes( *MetaRowByte = 0; } - if (SurfaceTiling == dm_sw_linear || SurfaceTiling == dm_sw_gfx7_2d_thin_gl || SurfaceTiling == dm_sw_gfx7_2d_thin_lvp) { + if (SurfaceTiling == dm_sw_linear || SurfaceTiling == dm_sw_gfx7_2d_thin_gl || SurfaceTiling == dm_sw_gfx7_2d_thin_l_vp) { MacroTileSizeBytes = 256; MacroTileHeight = BlockHeight256Bytes; } else if (SurfaceTiling == dm_sw_4kb_s || SurfaceTiling == dm_sw_4kb_s_x @@ -1335,11 +1336,11 @@ static void dml20_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPer else mode_lib->vba.SwathWidthSingleDPPY[k] = mode_lib->vba.ViewportHeight[k]; - if (mode_lib->vba.ODMCombineEnabled[k] == true) + if (mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) MainPlaneDoesODMCombine = true; for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) if (mode_lib->vba.BlendingAndTiming[k] == j - && mode_lib->vba.ODMCombineEnabled[j] == true) + && mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) MainPlaneDoesODMCombine = true; if (MainPlaneDoesODMCombine == true) @@ -2577,7 +2578,9 @@ static void dml20_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPer mode_lib->vba.MinActiveDRAMClockChangeMargin + mode_lib->vba.DRAMClockChangeLatency; - if (mode_lib->vba.MinActiveDRAMClockChangeMargin > 0) { + if (mode_lib->vba.DRAMClockChangeSupportsVActive && + mode_lib->vba.MinActiveDRAMClockChangeMargin > 60) { + mode_lib->vba.DRAMClockChangeWatermark += 25; mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive; } else { if (mode_lib->vba.SynchronizedVBlank || mode_lib->vba.NumberOfActivePlanes == 1) { @@ -2846,12 +2849,12 @@ static void dml20_DisplayPipeConfiguration(struct display_mode_lib *mode_lib) SwathWidth = mode_lib->vba.ViewportHeight[k]; } - if (mode_lib->vba.ODMCombineEnabled[k] == true) { + if (mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) { MainPlaneDoesODMCombine = true; } for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { if (mode_lib->vba.BlendingAndTiming[k] == j - && mode_lib->vba.ODMCombineEnabled[j] == true) { + && mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) { MainPlaneDoesODMCombine = true; } } @@ -3346,7 +3349,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l == dm_420_10)) || (((mode_lib->vba.SurfaceTiling[k] == dm_sw_gfx7_2d_thin_gl || mode_lib->vba.SurfaceTiling[k] - == dm_sw_gfx7_2d_thin_lvp) + == dm_sw_gfx7_2d_thin_l_vp) && !((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 || mode_lib->vba.SourcePixelFormat[k] @@ -3444,10 +3447,10 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l locals->FabricAndDRAMBandwidthPerState[i] * 1000) * locals->PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelDataOnly / 100; - locals->ReturnBWPerState[i] = locals->ReturnBWToDCNPerState; + locals->ReturnBWPerState[i][0] = locals->ReturnBWToDCNPerState; if (locals->DCCEnabledInAnyPlane == true && locals->ReturnBWToDCNPerState > locals->DCFCLKPerState[i] * locals->ReturnBusWidth / 4) { - locals->ReturnBWPerState[i] = dml_min(locals->ReturnBWPerState[i], + locals->ReturnBWPerState[i][0] = dml_min(locals->ReturnBWPerState[i][0], locals->ReturnBWToDCNPerState * 4 * (1 - locals->UrgentLatency / ((locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 / (locals->ReturnBWToDCNPerState - locals->DCFCLKPerState[i] @@ -3458,7 +3461,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024); if (locals->DCCEnabledInAnyPlane && locals->CriticalPoint > 1 && locals->CriticalPoint < 4) { - locals->ReturnBWPerState[i] = dml_min(locals->ReturnBWPerState[i], + locals->ReturnBWPerState[i][0] = dml_min(locals->ReturnBWPerState[i][0], 4 * locals->ReturnBWToDCNPerState * (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * locals->UrgentLatency / @@ -3470,7 +3473,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l locals->DCFCLKPerState[i], locals->FabricAndDRAMBandwidthPerState[i] * 1000); if (locals->DCCEnabledInAnyPlane == true && locals->ReturnBWToDCNPerState > locals->DCFCLKPerState[i] * locals->ReturnBusWidth / 4) { - locals->ReturnBWPerState[i] = dml_min(locals->ReturnBWPerState[i], + locals->ReturnBWPerState[i][0] = dml_min(locals->ReturnBWPerState[i][0], locals->ReturnBWToDCNPerState * 4 * (1 - locals->UrgentLatency / ((locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 / (locals->ReturnBWToDCNPerState - locals->DCFCLKPerState[i] @@ -3481,7 +3484,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024); if (locals->DCCEnabledInAnyPlane && locals->CriticalPoint > 1 && locals->CriticalPoint < 4) { - locals->ReturnBWPerState[i] = dml_min(locals->ReturnBWPerState[i], + locals->ReturnBWPerState[i][0] = dml_min(locals->ReturnBWPerState[i][0], 4 * locals->ReturnBWToDCNPerState * (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * locals->UrgentLatency / @@ -3519,12 +3522,12 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i] = (mode_lib->vba.RoundTripPingLatencyCycles + 32.0) / mode_lib->vba.DCFCLKPerState[i] - + locals->UrgentOutOfOrderReturnPerChannel * mode_lib->vba.NumberOfChannels / locals->ReturnBWPerState[i]; - if ((mode_lib->vba.ROBBufferSizeInKByte - mode_lib->vba.PixelChunkSizeInKByte) * 1024.0 / locals->ReturnBWPerState[i] + + locals->UrgentOutOfOrderReturnPerChannel * mode_lib->vba.NumberOfChannels / locals->ReturnBWPerState[i][0]; + if ((mode_lib->vba.ROBBufferSizeInKByte - mode_lib->vba.PixelChunkSizeInKByte) * 1024.0 / locals->ReturnBWPerState[i][0] > locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i]) { - locals->ROBSupport[i] = true; + locals->ROBSupport[i][0] = true; } else { - locals->ROBSupport[i] = false; + locals->ROBSupport[i][0] = false; } } /*Writeback Mode Support Check*/ @@ -3892,16 +3895,22 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l && i == mode_lib->vba.soc.num_states) mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine = mode_lib->vba.PixelClock[k] / 2 * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); - if (mode_lib->vba.ODMCapability == false || mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine <= mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity) { - locals->ODMCombineEnablePerState[i][k] = false; - mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine; - } else { - locals->ODMCombineEnablePerState[i][k] = true; - mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + + locals->ODMCombineEnablePerState[i][k] = false; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine; + if (mode_lib->vba.ODMCapability) { + if (locals->PlaneRequiredDISPCLKWithoutODMCombine > mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity) { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } else if (locals->HActive[k] > DCN20_MAX_420_IMAGE_WIDTH && locals->OutputFormat[k] == dm_420) { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } } + if (locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) <= mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity && locals->SwathWidthYSingleDPP[k] <= locals->MaximumSwathWidth[k] - && locals->ODMCombineEnablePerState[i][k] == false) { + && locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_disabled) { locals->NoOfDPP[i][j][k] = 1; locals->RequiredDPPCLK[i][j][k] = locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); @@ -3990,16 +3999,16 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l /*Viewport Size Check*/ for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { - locals->ViewportSizeSupport[i] = true; + locals->ViewportSizeSupport[i][0] = true; for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { - if (locals->ODMCombineEnablePerState[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { if (dml_min(locals->SwathWidthYSingleDPP[k], dml_round(mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k])) > locals->MaximumSwathWidth[k]) { - locals->ViewportSizeSupport[i] = false; + locals->ViewportSizeSupport[i][0] = false; } } else { if (locals->SwathWidthYSingleDPP[k] / 2.0 > locals->MaximumSwathWidth[k]) { - locals->ViewportSizeSupport[i] = false; + locals->ViewportSizeSupport[i][0] = false; } } } @@ -4181,8 +4190,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.DSCFormatFactor = 1; } if (locals->RequiresDSC[i][k] == true) { - if (locals->ODMCombineEnablePerState[i][k] - == true) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { if (mode_lib->vba.PixelClockBackEnd[k] / 6.0 / mode_lib->vba.DSCFormatFactor > (1.0 - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * mode_lib->vba.MaxDSCCLK[i]) { locals->DSCCLKRequiredMoreThanSupported[i] = @@ -4205,7 +4213,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.TotalDSCUnitsRequired = 0.0; for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { if (locals->RequiresDSC[i][k] == true) { - if (locals->ODMCombineEnablePerState[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { mode_lib->vba.TotalDSCUnitsRequired = mode_lib->vba.TotalDSCUnitsRequired + 2.0; } else { @@ -4247,7 +4255,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.bpp = locals->OutputBppPerState[i][k]; } if (locals->RequiresDSC[i][k] == true && mode_lib->vba.bpp != 0.0) { - if (locals->ODMCombineEnablePerState[i][k] == false) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_disabled) { locals->DSCDelayPerState[i][k] = dscceComputeDelay( mode_lib->vba.DSCInputBitPerComponent[k], @@ -4290,7 +4298,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { for (j = 0; j < 2; j++) { for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { - if (locals->ODMCombineEnablePerState[i][k] == true) + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) locals->SwathWidthYPerState[i][j][k] = dml_min(locals->SwathWidthYSingleDPP[k], dml_round(locals->HActive[k] / 2 * locals->HRatio[k])); else locals->SwathWidthYPerState[i][j][k] = locals->SwathWidthYSingleDPP[k] / locals->NoOfDPP[i][j][k]; @@ -4343,28 +4351,28 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l locals->EffectiveDETLBLinesLuma = dml_floor(locals->LinesInDETLuma + dml_min( locals->LinesInDETLuma * locals->RequiredDISPCLK[i][j] * locals->BytePerPixelInDETY[k] * - locals->PSCL_FACTOR[k] / locals->ReturnBWPerState[i], + locals->PSCL_FACTOR[k] / locals->ReturnBWPerState[i][0], locals->EffectiveLBLatencyHidingSourceLinesLuma), locals->SwathHeightYPerState[i][j][k]); locals->EffectiveDETLBLinesChroma = dml_floor(locals->LinesInDETChroma + dml_min( locals->LinesInDETChroma * locals->RequiredDISPCLK[i][j] * locals->BytePerPixelInDETC[k] * - locals->PSCL_FACTOR_CHROMA[k] / locals->ReturnBWPerState[i], + locals->PSCL_FACTOR_CHROMA[k] / locals->ReturnBWPerState[i][0], locals->EffectiveLBLatencyHidingSourceLinesChroma), locals->SwathHeightCPerState[i][j][k]); if (locals->BytePerPixelInDETC[k] == 0) { locals->UrgentLatencySupportUsPerState[i][j][k] = locals->EffectiveDETLBLinesLuma * (locals->HTotal[k] / locals->PixelClock[k]) / locals->VRatio[k] - locals->EffectiveDETLBLinesLuma * locals->SwathWidthYPerState[i][j][k] * - dml_ceil(locals->BytePerPixelInDETY[k], 1) / (locals->ReturnBWPerState[i] / locals->NoOfDPP[i][j][k]); + dml_ceil(locals->BytePerPixelInDETY[k], 1) / (locals->ReturnBWPerState[i][0] / locals->NoOfDPP[i][j][k]); } else { locals->UrgentLatencySupportUsPerState[i][j][k] = dml_min( locals->EffectiveDETLBLinesLuma * (locals->HTotal[k] / locals->PixelClock[k]) / locals->VRatio[k] - locals->EffectiveDETLBLinesLuma * locals->SwathWidthYPerState[i][j][k] * - dml_ceil(locals->BytePerPixelInDETY[k], 1) / (locals->ReturnBWPerState[i] / locals->NoOfDPP[i][j][k]), + dml_ceil(locals->BytePerPixelInDETY[k], 1) / (locals->ReturnBWPerState[i][0] / locals->NoOfDPP[i][j][k]), locals->EffectiveDETLBLinesChroma * (locals->HTotal[k] / locals->PixelClock[k]) / (locals->VRatio[k] / 2) - locals->EffectiveDETLBLinesChroma * locals->SwathWidthYPerState[i][j][k] / 2 * - dml_ceil(locals->BytePerPixelInDETC[k], 2) / (locals->ReturnBWPerState[i] / locals->NoOfDPP[i][j][k])); + dml_ceil(locals->BytePerPixelInDETC[k], 2) / (locals->ReturnBWPerState[i][0] / locals->NoOfDPP[i][j][k])); } } } @@ -4404,14 +4412,14 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l locals->SwathHeightYThisState[k] = locals->SwathHeightYPerState[i][j][k]; locals->SwathHeightCThisState[k] = locals->SwathHeightCPerState[i][j][k]; locals->SwathWidthYThisState[k] = locals->SwathWidthYPerState[i][j][k]; - mode_lib->vba.ProjectedDCFCLKDeepSleep = dml_max( - mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], mode_lib->vba.PixelClock[k] / 16.0); if (mode_lib->vba.BytePerPixelInDETC[k] == 0.0) { if (mode_lib->vba.VRatio[k] <= 1.0) { - mode_lib->vba.ProjectedDCFCLKDeepSleep = + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = dml_max( - mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], 1.1 * dml_ceil( mode_lib->vba.BytePerPixelInDETY[k], @@ -4421,9 +4429,9 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l * mode_lib->vba.PixelClock[k] / mode_lib->vba.NoOfDPP[i][j][k]); } else { - mode_lib->vba.ProjectedDCFCLKDeepSleep = + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = dml_max( - mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], 1.1 * dml_ceil( mode_lib->vba.BytePerPixelInDETY[k], @@ -4434,9 +4442,9 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l } } else { if (mode_lib->vba.VRatio[k] <= 1.0) { - mode_lib->vba.ProjectedDCFCLKDeepSleep = + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = dml_max( - mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], 1.1 * dml_ceil( mode_lib->vba.BytePerPixelInDETY[k], @@ -4446,9 +4454,9 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l * mode_lib->vba.PixelClock[k] / mode_lib->vba.NoOfDPP[i][j][k]); } else { - mode_lib->vba.ProjectedDCFCLKDeepSleep = + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = dml_max( - mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], 1.1 * dml_ceil( mode_lib->vba.BytePerPixelInDETY[k], @@ -4458,9 +4466,9 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l * mode_lib->vba.RequiredDPPCLK[i][j][k]); } if (mode_lib->vba.VRatio[k] / 2.0 <= 1.0) { - mode_lib->vba.ProjectedDCFCLKDeepSleep = + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = dml_max( - mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], 1.1 * dml_ceil( mode_lib->vba.BytePerPixelInDETC[k], @@ -4471,9 +4479,9 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l * mode_lib->vba.PixelClock[k] / mode_lib->vba.NoOfDPP[i][j][k]); } else { - mode_lib->vba.ProjectedDCFCLKDeepSleep = + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = dml_max( - mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], 1.1 * dml_ceil( mode_lib->vba.BytePerPixelInDETC[k], @@ -4509,7 +4517,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l &mode_lib->vba.PTEBufferSizeNotExceededY[i][j][k], &mode_lib->vba.dpte_row_height[k], &mode_lib->vba.meta_row_height[k]); - mode_lib->vba.PrefetchLinesY[k] = CalculatePrefetchSourceLines( + mode_lib->vba.PrefetchLinesY[0][0][k] = CalculatePrefetchSourceLines( mode_lib, mode_lib->vba.VRatio[k], mode_lib->vba.vtaps[k], @@ -4548,7 +4556,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l &mode_lib->vba.PTEBufferSizeNotExceededC[i][j][k], &mode_lib->vba.dpte_row_height_chroma[k], &mode_lib->vba.meta_row_height_chroma[k]); - mode_lib->vba.PrefetchLinesC[k] = CalculatePrefetchSourceLines( + mode_lib->vba.PrefetchLinesC[0][0][k] = CalculatePrefetchSourceLines( mode_lib, mode_lib->vba.VRatio[k] / 2.0, mode_lib->vba.VTAPsChroma[k], @@ -4562,14 +4570,14 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.PDEAndMetaPTEBytesPerFrameC = 0.0; mode_lib->vba.MetaRowBytesC = 0.0; mode_lib->vba.DPTEBytesPerRowC = 0.0; - locals->PrefetchLinesC[k] = 0.0; + locals->PrefetchLinesC[0][0][k] = 0.0; locals->PTEBufferSizeNotExceededC[i][j][k] = true; locals->PTEBufferSizeInRequestsForLuma = mode_lib->vba.PTEBufferSizeInRequestsLuma + mode_lib->vba.PTEBufferSizeInRequestsChroma; } - locals->PDEAndMetaPTEBytesPerFrame[k] = + locals->PDEAndMetaPTEBytesPerFrame[0][0][k] = mode_lib->vba.PDEAndMetaPTEBytesPerFrameY + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC; - locals->MetaRowBytes[k] = mode_lib->vba.MetaRowBytesY + mode_lib->vba.MetaRowBytesC; - locals->DPTEBytesPerRow[k] = mode_lib->vba.DPTEBytesPerRowY + mode_lib->vba.DPTEBytesPerRowC; + locals->MetaRowBytes[0][0][k] = mode_lib->vba.MetaRowBytesY + mode_lib->vba.MetaRowBytesC; + locals->DPTEBytesPerRow[0][0][k] = mode_lib->vba.DPTEBytesPerRowY + mode_lib->vba.DPTEBytesPerRowC; CalculateActiveRowBandwidth( mode_lib->vba.GPUVMEnable, @@ -4596,14 +4604,14 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l + mode_lib->vba.TotalNumberOfDCCActiveDPP[i][j] * mode_lib->vba.MetaChunkSize) * 1024.0 - / mode_lib->vba.ReturnBWPerState[i]; + / mode_lib->vba.ReturnBWPerState[i][0]; if (mode_lib->vba.GPUVMEnable == true) { mode_lib->vba.ExtraLatency = mode_lib->vba.ExtraLatency + mode_lib->vba.TotalNumberOfActiveDPP[i][j] * mode_lib->vba.PTEGroupSize - / mode_lib->vba.ReturnBWPerState[i]; + / mode_lib->vba.ReturnBWPerState[i][0]; } - mode_lib->vba.TimeCalc = 24.0 / mode_lib->vba.ProjectedDCFCLKDeepSleep; + mode_lib->vba.TimeCalc = 24.0 / mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0]; for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { if (mode_lib->vba.BlendingAndTiming[k] == k) { @@ -4653,7 +4661,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l } for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { - locals->MaximumVStartup[k] = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] + locals->MaximumVStartup[0][0][k] = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] - dml_max(1.0, dml_ceil(locals->WritebackDelay[i][k] / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]), 1.0)); } @@ -4698,7 +4706,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.RequiredDPPCLK[i][j][k], mode_lib->vba.RequiredDISPCLK[i][j], mode_lib->vba.PixelClock[k], - mode_lib->vba.ProjectedDCFCLKDeepSleep, + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], mode_lib->vba.DSCDelayPerState[i][k], mode_lib->vba.NoOfDPP[i][j][k], mode_lib->vba.ScalerEnabled[k], @@ -4716,7 +4724,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l - mode_lib->vba.VActive[k], mode_lib->vba.HTotal[k], mode_lib->vba.MaxInterDCNTileRepeaters, - mode_lib->vba.MaximumVStartup[k], + mode_lib->vba.MaximumVStartup[0][0][k], mode_lib->vba.GPUVMMaxPageTableLevels, mode_lib->vba.GPUVMEnable, mode_lib->vba.DynamicMetadataEnable[k], @@ -4726,15 +4734,15 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.UrgentLatencyPixelDataOnly, mode_lib->vba.ExtraLatency, mode_lib->vba.TimeCalc, - mode_lib->vba.PDEAndMetaPTEBytesPerFrame[k], - mode_lib->vba.MetaRowBytes[k], - mode_lib->vba.DPTEBytesPerRow[k], - mode_lib->vba.PrefetchLinesY[k], + mode_lib->vba.PDEAndMetaPTEBytesPerFrame[0][0][k], + mode_lib->vba.MetaRowBytes[0][0][k], + mode_lib->vba.DPTEBytesPerRow[0][0][k], + mode_lib->vba.PrefetchLinesY[0][0][k], mode_lib->vba.SwathWidthYPerState[i][j][k], mode_lib->vba.BytePerPixelInDETY[k], mode_lib->vba.PrefillY[k], mode_lib->vba.MaxNumSwY[k], - mode_lib->vba.PrefetchLinesC[k], + mode_lib->vba.PrefetchLinesC[0][0][k], mode_lib->vba.BytePerPixelInDETC[k], mode_lib->vba.PrefillC[k], mode_lib->vba.MaxNumSwC[k], @@ -4765,19 +4773,19 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l locals->prefetch_vm_bw_valid = true; locals->prefetch_row_bw_valid = true; for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { - if (locals->PDEAndMetaPTEBytesPerFrame[k] == 0) + if (locals->PDEAndMetaPTEBytesPerFrame[0][0][k] == 0) locals->prefetch_vm_bw[k] = 0; else if (locals->LinesForMetaPTE[k] > 0) - locals->prefetch_vm_bw[k] = locals->PDEAndMetaPTEBytesPerFrame[k] + locals->prefetch_vm_bw[k] = locals->PDEAndMetaPTEBytesPerFrame[0][0][k] / (locals->LinesForMetaPTE[k] * locals->HTotal[k] / locals->PixelClock[k]); else { locals->prefetch_vm_bw[k] = 0; locals->prefetch_vm_bw_valid = false; } - if (locals->MetaRowBytes[k] + locals->DPTEBytesPerRow[k] == 0) + if (locals->MetaRowBytes[0][0][k] + locals->DPTEBytesPerRow[0][0][k] == 0) locals->prefetch_row_bw[k] = 0; else if (locals->LinesForMetaAndDPTERow[k] > 0) - locals->prefetch_row_bw[k] = (locals->MetaRowBytes[k] + locals->DPTEBytesPerRow[k]) + locals->prefetch_row_bw[k] = (locals->MetaRowBytes[0][0][k] + locals->DPTEBytesPerRow[0][0][k]) / (locals->LinesForMetaAndDPTERow[k] * locals->HTotal[k] / locals->PixelClock[k]); else { locals->prefetch_row_bw[k] = 0; @@ -4796,13 +4804,13 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.RequiredPrefetchPixelDataBWLuma[i][j][k]) + mode_lib->vba.meta_row_bw[k] + mode_lib->vba.dpte_row_bw[k]); } - locals->BandwidthWithoutPrefetchSupported[i] = true; - if (mode_lib->vba.MaximumReadBandwidthWithoutPrefetch > locals->ReturnBWPerState[i]) { - locals->BandwidthWithoutPrefetchSupported[i] = false; + locals->BandwidthWithoutPrefetchSupported[i][0] = true; + if (mode_lib->vba.MaximumReadBandwidthWithoutPrefetch > locals->ReturnBWPerState[i][0]) { + locals->BandwidthWithoutPrefetchSupported[i][0] = false; } locals->PrefetchSupported[i][j] = true; - if (mode_lib->vba.MaximumReadBandwidthWithPrefetch > locals->ReturnBWPerState[i]) { + if (mode_lib->vba.MaximumReadBandwidthWithPrefetch > locals->ReturnBWPerState[i][0]) { locals->PrefetchSupported[i][j] = false; } for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { @@ -4827,7 +4835,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l if (mode_lib->vba.PrefetchSupported[i][j] == true && mode_lib->vba.VRatioInPrefetchSupported[i][j] == true) { mode_lib->vba.BandwidthAvailableForImmediateFlip = - mode_lib->vba.ReturnBWPerState[i]; + mode_lib->vba.ReturnBWPerState[i][0]; for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { mode_lib->vba.BandwidthAvailableForImmediateFlip = mode_lib->vba.BandwidthAvailableForImmediateFlip @@ -4841,9 +4849,9 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { mode_lib->vba.ImmediateFlipBytes[k] = - mode_lib->vba.PDEAndMetaPTEBytesPerFrame[k] - + mode_lib->vba.MetaRowBytes[k] - + mode_lib->vba.DPTEBytesPerRow[k]; + mode_lib->vba.PDEAndMetaPTEBytesPerFrame[0][0][k] + + mode_lib->vba.MetaRowBytes[0][0][k] + + mode_lib->vba.DPTEBytesPerRow[0][0][k]; } } mode_lib->vba.TotImmediateFlipBytes = 0.0; @@ -4871,9 +4879,9 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l / mode_lib->vba.PixelClock[k], mode_lib->vba.VRatio[k], mode_lib->vba.Tno_bw[k], - mode_lib->vba.PDEAndMetaPTEBytesPerFrame[k], - mode_lib->vba.MetaRowBytes[k], - mode_lib->vba.DPTEBytesPerRow[k], + mode_lib->vba.PDEAndMetaPTEBytesPerFrame[0][0][k], + mode_lib->vba.MetaRowBytes[0][0][k], + mode_lib->vba.DPTEBytesPerRow[0][0][k], mode_lib->vba.DCCEnable[k], mode_lib->vba.dpte_row_height[k], mode_lib->vba.meta_row_height[k], @@ -4898,7 +4906,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l } mode_lib->vba.ImmediateFlipSupportedForState[i][j] = true; if (mode_lib->vba.total_dcn_read_bw_with_flip - > mode_lib->vba.ReturnBWPerState[i]) { + > mode_lib->vba.ReturnBWPerState[i][0]) { mode_lib->vba.ImmediateFlipSupportedForState[i][j] = false; } for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { @@ -4917,13 +4925,13 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; k++) mode_lib->vba.MaxTotalVActiveRDBandwidth = mode_lib->vba.MaxTotalVActiveRDBandwidth + mode_lib->vba.ReadBandwidth[k]; for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { - mode_lib->vba.MaxTotalVerticalActiveAvailableBandwidth[i] = dml_min(mode_lib->vba.ReturnBusWidth * + mode_lib->vba.MaxTotalVerticalActiveAvailableBandwidth[i][0] = dml_min(mode_lib->vba.ReturnBusWidth * mode_lib->vba.DCFCLKPerState[i], mode_lib->vba.FabricAndDRAMBandwidthPerState[i] * 1000) * mode_lib->vba.MaxAveragePercentOfIdealDRAMBWDisplayCanUseInNormalSystemOperation / 100; - if (mode_lib->vba.MaxTotalVActiveRDBandwidth <= mode_lib->vba.MaxTotalVerticalActiveAvailableBandwidth[i]) - mode_lib->vba.TotalVerticalActiveBandwidthSupport[i] = true; + if (mode_lib->vba.MaxTotalVActiveRDBandwidth <= mode_lib->vba.MaxTotalVerticalActiveAvailableBandwidth[i][0]) + mode_lib->vba.TotalVerticalActiveBandwidthSupport[i][0] = true; else - mode_lib->vba.TotalVerticalActiveBandwidthSupport[i] = false; + mode_lib->vba.TotalVerticalActiveBandwidthSupport[i][0] = false; } /*PTE Buffer Size Check*/ @@ -5011,7 +5019,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l status = DML_FAIL_SCALE_RATIO_TAP; } else if (mode_lib->vba.SourceFormatPixelAndScanSupport != true) { status = DML_FAIL_SOURCE_PIXEL_FORMAT; - } else if (locals->ViewportSizeSupport[i] != true) { + } else if (locals->ViewportSizeSupport[i][0] != true) { status = DML_FAIL_VIEWPORT_SIZE; } else if (locals->DIOSupport[i] != true) { status = DML_FAIL_DIO_SUPPORT; @@ -5021,7 +5029,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l status = DML_FAIL_DSC_CLK_REQUIRED; } else if (locals->UrgentLatencySupport[i][j] != true) { status = DML_FAIL_URGENT_LATENCY; - } else if (locals->ROBSupport[i] != true) { + } else if (locals->ROBSupport[i][0] != true) { status = DML_FAIL_REORDERING_BUFFER; } else if (locals->DISPCLK_DPPCLK_Support[i][j] != true) { status = DML_FAIL_DISPCLK_DPPCLK; @@ -5041,7 +5049,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l status = DML_FAIL_PITCH_SUPPORT; } else if (locals->PrefetchSupported[i][j] != true) { status = DML_FAIL_PREFETCH_SUPPORT; - } else if (locals->TotalVerticalActiveBandwidthSupport[i] != true) { + } else if (locals->TotalVerticalActiveBandwidthSupport[i][0] != true) { status = DML_FAIL_TOTAL_V_ACTIVE_BW; } else if (locals->VRatioInPrefetchSupported[i][j] != true) { status = DML_FAIL_V_RATIO_PREFETCH; @@ -5087,7 +5095,7 @@ void dml20_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.DRAMSpeed = mode_lib->vba.DRAMSpeedPerState[mode_lib->vba.VoltageLevel]; mode_lib->vba.FabricClock = mode_lib->vba.FabricClockPerState[mode_lib->vba.VoltageLevel]; mode_lib->vba.SOCCLK = mode_lib->vba.SOCCLKPerState[mode_lib->vba.VoltageLevel]; - mode_lib->vba.ReturnBW = locals->ReturnBWPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.ReturnBW = locals->ReturnBWPerState[mode_lib->vba.VoltageLevel][0]; mode_lib->vba.FabricAndDRAMBandwidth = locals->FabricAndDRAMBandwidthPerState[mode_lib->vba.VoltageLevel]; for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { if (mode_lib->vba.BlendingAndTiming[k] == k) { diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c new file mode 100644 index 000000000000..485a9c62ec58 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c @@ -0,0 +1,5155 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "../display_mode_lib.h" +#include "display_mode_vba_20v2.h" +#include "../dml_inline_defs.h" + +/* + * NOTE: + * This file is gcc-parseable HW gospel, coming straight from HW engineers. + * + * It doesn't adhere to Linux kernel style and sometimes will do things in odd + * ways. Unless there is something clearly wrong with it the code should + * remain as-is as it provides us with a guarantee from HW that it is correct. + */ + +#define BPP_INVALID 0 +#define BPP_BLENDED_PIPE 0xffffffff +#define DCN20_MAX_DSC_IMAGE_WIDTH 5184 +#define DCN20_MAX_420_IMAGE_WIDTH 4096 + +static double adjust_ReturnBW( + struct display_mode_lib *mode_lib, + double ReturnBW, + bool DCCEnabledAnyPlane, + double ReturnBandwidthToDCN); +static unsigned int dscceComputeDelay( + unsigned int bpc, + double bpp, + unsigned int sliceWidth, + unsigned int numSlices, + enum output_format_class pixelFormat); +static unsigned int dscComputeDelay(enum output_format_class pixelFormat); +static bool CalculateDelayAfterScaler( + struct display_mode_lib *mode_lib, + double ReturnBW, + double ReadBandwidthPlaneLuma, + double ReadBandwidthPlaneChroma, + double TotalDataReadBandwidth, + double DisplayPipeLineDeliveryTimeLuma, + double DisplayPipeLineDeliveryTimeChroma, + double DPPCLK, + double DISPCLK, + double PixelClock, + unsigned int DSCDelay, + unsigned int DPPPerPlane, + bool ScalerEnabled, + unsigned int NumberOfCursors, + double DPPCLKDelaySubtotal, + double DPPCLKDelaySCL, + double DPPCLKDelaySCLLBOnly, + double DPPCLKDelayCNVCFormater, + double DPPCLKDelayCNVCCursor, + double DISPCLKDelaySubtotal, + unsigned int ScalerRecoutWidth, + enum output_format_class OutputFormat, + unsigned int HTotal, + unsigned int SwathWidthSingleDPPY, + double BytePerPixelDETY, + double BytePerPixelDETC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + double *DSTXAfterScaler, + double *DSTYAfterScaler + ); +// Super monster function with some 45 argument +static bool CalculatePrefetchSchedule( + struct display_mode_lib *mode_lib, + double DPPCLK, + double DISPCLK, + double PixelClock, + double DCFCLKDeepSleep, + unsigned int DPPPerPlane, + unsigned int NumberOfCursors, + unsigned int VBlank, + unsigned int HTotal, + unsigned int MaxInterDCNTileRepeaters, + unsigned int VStartup, + unsigned int PageTableLevels, + bool GPUVMEnable, + bool DynamicMetadataEnable, + unsigned int DynamicMetadataLinesBeforeActiveRequired, + unsigned int DynamicMetadataTransmittedBytes, + bool DCCEnable, + double UrgentLatencyPixelDataOnly, + double UrgentExtraLatency, + double TCalc, + unsigned int PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + double PrefetchSourceLinesY, + unsigned int SwathWidthY, + double BytePerPixelDETY, + double VInitPreFillY, + unsigned int MaxNumSwathY, + double PrefetchSourceLinesC, + double BytePerPixelDETC, + double VInitPreFillC, + unsigned int MaxNumSwathC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double TWait, + bool XFCEnabled, + double XFCRemoteSurfaceFlipDelay, + bool InterlaceEnable, + bool ProgressiveToInterlaceUnitInOPP, + double DSTXAfterScaler, + double DSTYAfterScaler, + double *DestinationLinesForPrefetch, + double *PrefetchBandwidth, + double *DestinationLinesToRequestVMInVBlank, + double *DestinationLinesToRequestRowInVBlank, + double *VRatioPrefetchY, + double *VRatioPrefetchC, + double *RequiredPrefetchPixDataBW, + double *Tno_bw, + unsigned int *VUpdateOffsetPix, + double *VUpdateWidthPix, + double *VReadyOffsetPix); +static double RoundToDFSGranularityUp(double Clock, double VCOSpeed); +static double RoundToDFSGranularityDown(double Clock, double VCOSpeed); +static double CalculatePrefetchSourceLines( + struct display_mode_lib *mode_lib, + double VRatio, + double vtaps, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + unsigned int SwathHeight, + unsigned int ViewportYStart, + double *VInitPreFill, + unsigned int *MaxNumSwath); +static unsigned int CalculateVMAndRowBytes( + struct display_mode_lib *mode_lib, + bool DCCEnable, + unsigned int BlockHeight256Bytes, + unsigned int BlockWidth256Bytes, + enum source_format_class SourcePixelFormat, + unsigned int SurfaceTiling, + unsigned int BytePerPixel, + enum scan_direction_class ScanDirection, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + unsigned int SwathWidthY, + bool GPUVMEnable, + unsigned int VMMPageSize, + unsigned int PTEBufferSizeInRequestsLuma, + unsigned int PDEProcessingBufIn64KBReqs, + unsigned int Pitch, + unsigned int DCCMetaPitch, + unsigned int *MacroTileWidth, + unsigned int *MetaRowByte, + unsigned int *PixelPTEBytesPerRow, + bool *PTEBufferSizeNotExceeded, + unsigned int *dpte_row_height, + unsigned int *meta_row_height); +static double CalculateTWait( + unsigned int PrefetchMode, + double DRAMClockChangeLatency, + double UrgentLatencyPixelDataOnly, + double SREnterPlusExitTime); +static double CalculateRemoteSurfaceFlipDelay( + struct display_mode_lib *mode_lib, + double VRatio, + double SwathWidth, + double Bpp, + double LineTime, + double XFCTSlvVupdateOffset, + double XFCTSlvVupdateWidth, + double XFCTSlvVreadyOffset, + double XFCXBUFLatencyTolerance, + double XFCFillBWOverhead, + double XFCSlvChunkSize, + double XFCBusTransportTime, + double TCalc, + double TWait, + double *SrcActiveDrainRate, + double *TInitXFill, + double *TslvChk); +static void CalculateActiveRowBandwidth( + bool GPUVMEnable, + enum source_format_class SourcePixelFormat, + double VRatio, + bool DCCEnable, + double LineTime, + unsigned int MetaRowByteLuma, + unsigned int MetaRowByteChroma, + unsigned int meta_row_height_luma, + unsigned int meta_row_height_chroma, + unsigned int PixelPTEBytesPerRowLuma, + unsigned int PixelPTEBytesPerRowChroma, + unsigned int dpte_row_height_luma, + unsigned int dpte_row_height_chroma, + double *meta_row_bw, + double *dpte_row_bw, + double *qual_row_bw); +static void CalculateFlipSchedule( + struct display_mode_lib *mode_lib, + double UrgentExtraLatency, + double UrgentLatencyPixelDataOnly, + unsigned int GPUVMMaxPageTableLevels, + bool GPUVMEnable, + double BandwidthAvailableForImmediateFlip, + unsigned int TotImmediateFlipBytes, + enum source_format_class SourcePixelFormat, + unsigned int ImmediateFlipBytes, + double LineTime, + double VRatio, + double Tno_bw, + double PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + bool DCCEnable, + unsigned int dpte_row_height, + unsigned int meta_row_height, + double qual_row_bw, + double *DestinationLinesToRequestVMInImmediateFlip, + double *DestinationLinesToRequestRowInImmediateFlip, + double *final_flip_bw, + bool *ImmediateFlipSupportedForPipe); +static double CalculateWriteBackDelay( + enum source_format_class WritebackPixelFormat, + double WritebackHRatio, + double WritebackVRatio, + unsigned int WritebackLumaHTaps, + unsigned int WritebackLumaVTaps, + unsigned int WritebackChromaHTaps, + unsigned int WritebackChromaVTaps, + unsigned int WritebackDestinationWidth); + +static void dml20v2_DisplayPipeConfiguration(struct display_mode_lib *mode_lib); +static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation( + struct display_mode_lib *mode_lib); + +void dml20v2_recalculate(struct display_mode_lib *mode_lib) +{ + ModeSupportAndSystemConfiguration(mode_lib); + mode_lib->vba.FabricAndDRAMBandwidth = dml_min( + mode_lib->vba.DRAMSpeed * mode_lib->vba.NumberOfChannels * mode_lib->vba.DRAMChannelWidth, + mode_lib->vba.FabricClock * mode_lib->vba.FabricDatapathToDCNDataReturn) / 1000.0; + PixelClockAdjustmentForProgressiveToInterlaceUnit(mode_lib); + dml20v2_DisplayPipeConfiguration(mode_lib); + dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation(mode_lib); +} + +static double adjust_ReturnBW( + struct display_mode_lib *mode_lib, + double ReturnBW, + bool DCCEnabledAnyPlane, + double ReturnBandwidthToDCN) +{ + double CriticalCompression; + + if (DCCEnabledAnyPlane + && ReturnBandwidthToDCN + > mode_lib->vba.DCFCLK * mode_lib->vba.ReturnBusWidth / 4.0) + ReturnBW = + dml_min( + ReturnBW, + ReturnBandwidthToDCN * 4 + * (1.0 + - mode_lib->vba.UrgentLatencyPixelDataOnly + / ((mode_lib->vba.ROBBufferSizeInKByte + - mode_lib->vba.PixelChunkSizeInKByte) + * 1024 + / ReturnBandwidthToDCN + - mode_lib->vba.DCFCLK + * mode_lib->vba.ReturnBusWidth + / 4) + + mode_lib->vba.UrgentLatencyPixelDataOnly)); + + CriticalCompression = 2.0 * mode_lib->vba.ReturnBusWidth * mode_lib->vba.DCFCLK + * mode_lib->vba.UrgentLatencyPixelDataOnly + / (ReturnBandwidthToDCN * mode_lib->vba.UrgentLatencyPixelDataOnly + + (mode_lib->vba.ROBBufferSizeInKByte + - mode_lib->vba.PixelChunkSizeInKByte) + * 1024); + + if (DCCEnabledAnyPlane && CriticalCompression > 1.0 && CriticalCompression < 4.0) + ReturnBW = + dml_min( + ReturnBW, + 4.0 * ReturnBandwidthToDCN + * (mode_lib->vba.ROBBufferSizeInKByte + - mode_lib->vba.PixelChunkSizeInKByte) + * 1024 + * mode_lib->vba.ReturnBusWidth + * mode_lib->vba.DCFCLK + * mode_lib->vba.UrgentLatencyPixelDataOnly + / dml_pow( + (ReturnBandwidthToDCN + * mode_lib->vba.UrgentLatencyPixelDataOnly + + (mode_lib->vba.ROBBufferSizeInKByte + - mode_lib->vba.PixelChunkSizeInKByte) + * 1024), + 2)); + + return ReturnBW; +} + +static unsigned int dscceComputeDelay( + unsigned int bpc, + double bpp, + unsigned int sliceWidth, + unsigned int numSlices, + enum output_format_class pixelFormat) +{ + // valid bpc = source bits per component in the set of {8, 10, 12} + // valid bpp = increments of 1/16 of a bit + // min = 6/7/8 in N420/N422/444, respectively + // max = such that compression is 1:1 + //valid sliceWidth = number of pixels per slice line, must be less than or equal to 5184/numSlices (or 4096/numSlices in 420 mode) + //valid numSlices = number of slices in the horiziontal direction per DSC engine in the set of {1, 2, 3, 4} + //valid pixelFormat = pixel/color format in the set of {:N444_RGB, :S422, :N422, :N420} + + // fixed value + unsigned int rcModelSize = 8192; + + // N422/N420 operate at 2 pixels per clock + unsigned int pixelsPerClock, lstall, D, initalXmitDelay, w, s, ix, wx, p, l0, a, ax, l, + Delay, pixels; + + if (pixelFormat == dm_n422 || pixelFormat == dm_420) + pixelsPerClock = 2; + // #all other modes operate at 1 pixel per clock + else + pixelsPerClock = 1; + + //initial transmit delay as per PPS + initalXmitDelay = dml_round(rcModelSize / 2.0 / bpp / pixelsPerClock); + + //compute ssm delay + if (bpc == 8) + D = 81; + else if (bpc == 10) + D = 89; + else + D = 113; + + //divide by pixel per cycle to compute slice width as seen by DSC + w = sliceWidth / pixelsPerClock; + + //422 mode has an additional cycle of delay + if (pixelFormat == dm_s422) + s = 1; + else + s = 0; + + //main calculation for the dscce + ix = initalXmitDelay + 45; + wx = (w + 2) / 3; + p = 3 * wx - w; + l0 = ix / w; + a = ix + p * l0; + ax = (a + 2) / 3 + D + 6 + 1; + l = (ax + wx - 1) / wx; + if ((ix % w) == 0 && p != 0) + lstall = 1; + else + lstall = 0; + Delay = l * wx * (numSlices - 1) + ax + s + lstall + 22; + + //dsc processes 3 pixel containers per cycle and a container can contain 1 or 2 pixels + pixels = Delay * 3 * pixelsPerClock; + return pixels; +} + +static unsigned int dscComputeDelay(enum output_format_class pixelFormat) +{ + unsigned int Delay = 0; + + if (pixelFormat == dm_420) { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 0; + // dscc - input deserializer + Delay = Delay + 3; + // dscc gets pixels every other cycle + Delay = Delay + 2; + // dscc - input cdc fifo + Delay = Delay + 12; + // dscc gets pixels every other cycle + Delay = Delay + 13; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 7; + // dscc gets pixels every other cycle + Delay = Delay + 3; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output serializer + Delay = Delay + 1; + // sft + Delay = Delay + 1; + } else if (pixelFormat == dm_n422) { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 1; + // dscc - input deserializer + Delay = Delay + 5; + // dscc - input cdc fifo + Delay = Delay + 25; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 10; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output serializer + Delay = Delay + 1; + // sft + Delay = Delay + 1; + } else { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 0; + // dscc - input deserializer + Delay = Delay + 3; + // dscc - input cdc fifo + Delay = Delay + 12; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 7; + // dscc - output serializer + Delay = Delay + 1; + // dscc - cdc uncertainty + Delay = Delay + 2; + // sft + Delay = Delay + 1; + } + + return Delay; +} + +static bool CalculateDelayAfterScaler( + struct display_mode_lib *mode_lib, + double ReturnBW, + double ReadBandwidthPlaneLuma, + double ReadBandwidthPlaneChroma, + double TotalDataReadBandwidth, + double DisplayPipeLineDeliveryTimeLuma, + double DisplayPipeLineDeliveryTimeChroma, + double DPPCLK, + double DISPCLK, + double PixelClock, + unsigned int DSCDelay, + unsigned int DPPPerPlane, + bool ScalerEnabled, + unsigned int NumberOfCursors, + double DPPCLKDelaySubtotal, + double DPPCLKDelaySCL, + double DPPCLKDelaySCLLBOnly, + double DPPCLKDelayCNVCFormater, + double DPPCLKDelayCNVCCursor, + double DISPCLKDelaySubtotal, + unsigned int ScalerRecoutWidth, + enum output_format_class OutputFormat, + unsigned int HTotal, + unsigned int SwathWidthSingleDPPY, + double BytePerPixelDETY, + double BytePerPixelDETC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + double *DSTXAfterScaler, + double *DSTYAfterScaler + ) +{ + unsigned int DPPCycles, DISPCLKCycles; + double DataFabricLineDeliveryTimeLuma; + double DataFabricLineDeliveryTimeChroma; + double DSTTotalPixelsAfterScaler; + + DataFabricLineDeliveryTimeLuma = SwathWidthSingleDPPY * SwathHeightY * dml_ceil(BytePerPixelDETY, 1) / (mode_lib->vba.ReturnBW * ReadBandwidthPlaneLuma / TotalDataReadBandwidth); + mode_lib->vba.LastPixelOfLineExtraWatermark = dml_max(mode_lib->vba.LastPixelOfLineExtraWatermark, DataFabricLineDeliveryTimeLuma - DisplayPipeLineDeliveryTimeLuma); + + if (BytePerPixelDETC != 0) { + DataFabricLineDeliveryTimeChroma = SwathWidthSingleDPPY / 2 * SwathHeightC * dml_ceil(BytePerPixelDETC, 2) / (mode_lib->vba.ReturnBW * ReadBandwidthPlaneChroma / TotalDataReadBandwidth); + mode_lib->vba.LastPixelOfLineExtraWatermark = dml_max(mode_lib->vba.LastPixelOfLineExtraWatermark, DataFabricLineDeliveryTimeChroma - DisplayPipeLineDeliveryTimeChroma); + } + + if (ScalerEnabled) + DPPCycles = DPPCLKDelaySubtotal + DPPCLKDelaySCL; + else + DPPCycles = DPPCLKDelaySubtotal + DPPCLKDelaySCLLBOnly; + + DPPCycles = DPPCycles + DPPCLKDelayCNVCFormater + NumberOfCursors * DPPCLKDelayCNVCCursor; + + DISPCLKCycles = DISPCLKDelaySubtotal; + + if (DPPCLK == 0.0 || DISPCLK == 0.0) + return true; + + *DSTXAfterScaler = DPPCycles * PixelClock / DPPCLK + DISPCLKCycles * PixelClock / DISPCLK + + DSCDelay; + + if (DPPPerPlane > 1) + *DSTXAfterScaler = *DSTXAfterScaler + ScalerRecoutWidth; + + if (OutputFormat == dm_420 || (Interlace && ProgressiveToInterlaceUnitInOPP)) + *DSTYAfterScaler = 1; + else + *DSTYAfterScaler = 0; + + DSTTotalPixelsAfterScaler = ((double) (*DSTYAfterScaler * HTotal)) + *DSTXAfterScaler; + *DSTYAfterScaler = dml_floor(DSTTotalPixelsAfterScaler / HTotal, 1); + *DSTXAfterScaler = DSTTotalPixelsAfterScaler - ((double) (*DSTYAfterScaler * HTotal)); + + return true; +} + +static bool CalculatePrefetchSchedule( + struct display_mode_lib *mode_lib, + double DPPCLK, + double DISPCLK, + double PixelClock, + double DCFCLKDeepSleep, + unsigned int DPPPerPlane, + unsigned int NumberOfCursors, + unsigned int VBlank, + unsigned int HTotal, + unsigned int MaxInterDCNTileRepeaters, + unsigned int VStartup, + unsigned int PageTableLevels, + bool GPUVMEnable, + bool DynamicMetadataEnable, + unsigned int DynamicMetadataLinesBeforeActiveRequired, + unsigned int DynamicMetadataTransmittedBytes, + bool DCCEnable, + double UrgentLatencyPixelDataOnly, + double UrgentExtraLatency, + double TCalc, + unsigned int PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + double PrefetchSourceLinesY, + unsigned int SwathWidthY, + double BytePerPixelDETY, + double VInitPreFillY, + unsigned int MaxNumSwathY, + double PrefetchSourceLinesC, + double BytePerPixelDETC, + double VInitPreFillC, + unsigned int MaxNumSwathC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double TWait, + bool XFCEnabled, + double XFCRemoteSurfaceFlipDelay, + bool InterlaceEnable, + bool ProgressiveToInterlaceUnitInOPP, + double DSTXAfterScaler, + double DSTYAfterScaler, + double *DestinationLinesForPrefetch, + double *PrefetchBandwidth, + double *DestinationLinesToRequestVMInVBlank, + double *DestinationLinesToRequestRowInVBlank, + double *VRatioPrefetchY, + double *VRatioPrefetchC, + double *RequiredPrefetchPixDataBW, + double *Tno_bw, + unsigned int *VUpdateOffsetPix, + double *VUpdateWidthPix, + double *VReadyOffsetPix) +{ + bool MyError = false; + double TotalRepeaterDelayTime; + double Tdm, LineTime, Tsetup; + double dst_y_prefetch_equ; + double Tsw_oto; + double prefetch_bw_oto; + double Tvm_oto; + double Tr0_oto; + double Tpre_oto; + double dst_y_prefetch_oto; + double TimeForFetchingMetaPTE = 0; + double TimeForFetchingRowInVBlank = 0; + double LinesToRequestPrefetchPixelData = 0; + + *VUpdateOffsetPix = dml_ceil(HTotal / 4.0, 1); + TotalRepeaterDelayTime = MaxInterDCNTileRepeaters * (2.0 / DPPCLK + 3.0 / DISPCLK); + *VUpdateWidthPix = (14.0 / DCFCLKDeepSleep + 12.0 / DPPCLK + TotalRepeaterDelayTime) + * PixelClock; + + *VReadyOffsetPix = dml_max( + 150.0 / DPPCLK, + TotalRepeaterDelayTime + 20.0 / DCFCLKDeepSleep + 10.0 / DPPCLK) + * PixelClock; + + Tsetup = (double) (*VUpdateOffsetPix + *VUpdateWidthPix + *VReadyOffsetPix) / PixelClock; + + LineTime = (double) HTotal / PixelClock; + + if (DynamicMetadataEnable) { + double Tdmbf, Tdmec, Tdmsks; + + Tdm = dml_max(0.0, UrgentExtraLatency - TCalc); + Tdmbf = DynamicMetadataTransmittedBytes / 4.0 / DISPCLK; + Tdmec = LineTime; + if (DynamicMetadataLinesBeforeActiveRequired == 0) + Tdmsks = VBlank * LineTime / 2.0; + else + Tdmsks = DynamicMetadataLinesBeforeActiveRequired * LineTime; + if (InterlaceEnable && !ProgressiveToInterlaceUnitInOPP) + Tdmsks = Tdmsks / 2; + if (VStartup * LineTime + < Tsetup + TWait + UrgentExtraLatency + Tdmbf + Tdmec + Tdmsks) { + MyError = true; + } + } else + Tdm = 0; + + if (GPUVMEnable) { + if (PageTableLevels == 4) + *Tno_bw = UrgentExtraLatency + UrgentLatencyPixelDataOnly; + else if (PageTableLevels == 3) + *Tno_bw = UrgentExtraLatency; + else + *Tno_bw = 0; + } else if (DCCEnable) + *Tno_bw = LineTime; + else + *Tno_bw = LineTime / 4; + + dst_y_prefetch_equ = VStartup - dml_max(TCalc + TWait, XFCRemoteSurfaceFlipDelay) / LineTime + - (Tsetup + Tdm) / LineTime + - (DSTYAfterScaler + DSTXAfterScaler / HTotal); + + Tsw_oto = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime; + + prefetch_bw_oto = (MetaRowByte + PixelPTEBytesPerRow + + PrefetchSourceLinesY * SwathWidthY * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * SwathWidthY / 2 * dml_ceil(BytePerPixelDETC, 2)) + / Tsw_oto; + + if (GPUVMEnable == true) { + Tvm_oto = + dml_max( + *Tno_bw + PDEAndMetaPTEBytesFrame / prefetch_bw_oto, + dml_max( + UrgentExtraLatency + + UrgentLatencyPixelDataOnly + * (PageTableLevels + - 1), + LineTime / 4.0)); + } else + Tvm_oto = LineTime / 4.0; + + if ((GPUVMEnable == true || DCCEnable == true)) { + Tr0_oto = dml_max( + (MetaRowByte + PixelPTEBytesPerRow) / prefetch_bw_oto, + dml_max(UrgentLatencyPixelDataOnly, dml_max(LineTime - Tvm_oto, LineTime / 4))); + } else + Tr0_oto = LineTime - Tvm_oto; + + Tpre_oto = Tvm_oto + Tr0_oto + Tsw_oto; + + dst_y_prefetch_oto = Tpre_oto / LineTime; + + if (dst_y_prefetch_oto < dst_y_prefetch_equ) + *DestinationLinesForPrefetch = dst_y_prefetch_oto; + else + *DestinationLinesForPrefetch = dst_y_prefetch_equ; + + *DestinationLinesForPrefetch = dml_floor(4.0 * (*DestinationLinesForPrefetch + 0.125), 1) + / 4; + + dml_print("DML: VStartup: %d\n", VStartup); + dml_print("DML: TCalc: %f\n", TCalc); + dml_print("DML: TWait: %f\n", TWait); + dml_print("DML: XFCRemoteSurfaceFlipDelay: %f\n", XFCRemoteSurfaceFlipDelay); + dml_print("DML: LineTime: %f\n", LineTime); + dml_print("DML: Tsetup: %f\n", Tsetup); + dml_print("DML: Tdm: %f\n", Tdm); + dml_print("DML: DSTYAfterScaler: %f\n", DSTYAfterScaler); + dml_print("DML: DSTXAfterScaler: %f\n", DSTXAfterScaler); + dml_print("DML: HTotal: %d\n", HTotal); + + *PrefetchBandwidth = 0; + *DestinationLinesToRequestVMInVBlank = 0; + *DestinationLinesToRequestRowInVBlank = 0; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBW = 0; + if (*DestinationLinesForPrefetch > 1) { + *PrefetchBandwidth = (PDEAndMetaPTEBytesFrame + 2 * MetaRowByte + + 2 * PixelPTEBytesPerRow + + PrefetchSourceLinesY * SwathWidthY * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * SwathWidthY / 2 + * dml_ceil(BytePerPixelDETC, 2)) + / (*DestinationLinesForPrefetch * LineTime - *Tno_bw); + if (GPUVMEnable) { + TimeForFetchingMetaPTE = + dml_max( + *Tno_bw + + (double) PDEAndMetaPTEBytesFrame + / *PrefetchBandwidth, + dml_max( + UrgentExtraLatency + + UrgentLatencyPixelDataOnly + * (PageTableLevels + - 1), + LineTime / 4)); + } else { + if (NumberOfCursors > 0 || XFCEnabled) + TimeForFetchingMetaPTE = LineTime / 4; + else + TimeForFetchingMetaPTE = 0.0; + } + + if ((GPUVMEnable == true || DCCEnable == true)) { + TimeForFetchingRowInVBlank = + dml_max( + (MetaRowByte + PixelPTEBytesPerRow) + / *PrefetchBandwidth, + dml_max( + UrgentLatencyPixelDataOnly, + dml_max( + LineTime + - TimeForFetchingMetaPTE, + LineTime + / 4.0))); + } else { + if (NumberOfCursors > 0 || XFCEnabled) + TimeForFetchingRowInVBlank = LineTime - TimeForFetchingMetaPTE; + else + TimeForFetchingRowInVBlank = 0.0; + } + + *DestinationLinesToRequestVMInVBlank = dml_floor( + 4.0 * (TimeForFetchingMetaPTE / LineTime + 0.125), + 1) / 4.0; + + *DestinationLinesToRequestRowInVBlank = dml_floor( + 4.0 * (TimeForFetchingRowInVBlank / LineTime + 0.125), + 1) / 4.0; + + LinesToRequestPrefetchPixelData = + *DestinationLinesForPrefetch + - ((NumberOfCursors > 0 || GPUVMEnable + || DCCEnable) ? + (*DestinationLinesToRequestVMInVBlank + + *DestinationLinesToRequestRowInVBlank) : + 0.0); + + if (LinesToRequestPrefetchPixelData > 0) { + + *VRatioPrefetchY = (double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData; + *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); + if ((SwathHeightY > 4) && (VInitPreFillY > 3)) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillY - 3.0) / 2.0) { + *VRatioPrefetchY = + dml_max( + (double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData, + (double) MaxNumSwathY + * SwathHeightY + / (LinesToRequestPrefetchPixelData + - (VInitPreFillY + - 3.0) + / 2.0)); + *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); + } else { + MyError = true; + *VRatioPrefetchY = 0; + } + } + + *VRatioPrefetchC = (double) PrefetchSourceLinesC + / LinesToRequestPrefetchPixelData; + *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); + + if ((SwathHeightC > 4)) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillC - 3.0) / 2.0) { + *VRatioPrefetchC = + dml_max( + *VRatioPrefetchC, + (double) MaxNumSwathC + * SwathHeightC + / (LinesToRequestPrefetchPixelData + - (VInitPreFillC + - 3.0) + / 2.0)); + *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); + } else { + MyError = true; + *VRatioPrefetchC = 0; + } + } + + *RequiredPrefetchPixDataBW = + DPPPerPlane + * ((double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData + * dml_ceil( + BytePerPixelDETY, + 1) + + (double) PrefetchSourceLinesC + / LinesToRequestPrefetchPixelData + * dml_ceil( + BytePerPixelDETC, + 2) + / 2) + * SwathWidthY / LineTime; + } else { + MyError = true; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBW = 0; + } + + } else { + MyError = true; + } + + if (MyError) { + *PrefetchBandwidth = 0; + TimeForFetchingMetaPTE = 0; + TimeForFetchingRowInVBlank = 0; + *DestinationLinesToRequestVMInVBlank = 0; + *DestinationLinesToRequestRowInVBlank = 0; + *DestinationLinesForPrefetch = 0; + LinesToRequestPrefetchPixelData = 0; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBW = 0; + } + + return MyError; +} + +static double RoundToDFSGranularityUp(double Clock, double VCOSpeed) +{ + return VCOSpeed * 4 / dml_floor(VCOSpeed * 4 / Clock, 1); +} + +static double RoundToDFSGranularityDown(double Clock, double VCOSpeed) +{ + return VCOSpeed * 4 / dml_ceil(VCOSpeed * 4 / Clock, 1); +} + +static double CalculatePrefetchSourceLines( + struct display_mode_lib *mode_lib, + double VRatio, + double vtaps, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + unsigned int SwathHeight, + unsigned int ViewportYStart, + double *VInitPreFill, + unsigned int *MaxNumSwath) +{ + unsigned int MaxPartialSwath; + + if (ProgressiveToInterlaceUnitInOPP) + *VInitPreFill = dml_floor((VRatio + vtaps + 1) / 2.0, 1); + else + *VInitPreFill = dml_floor((VRatio + vtaps + 1 + Interlace * 0.5 * VRatio) / 2.0, 1); + + if (!mode_lib->vba.IgnoreViewportPositioning) { + + *MaxNumSwath = dml_ceil((*VInitPreFill - 1.0) / SwathHeight, 1) + 1.0; + + if (*VInitPreFill > 1.0) + MaxPartialSwath = (unsigned int) (*VInitPreFill - 2) % SwathHeight; + else + MaxPartialSwath = (unsigned int) (*VInitPreFill + SwathHeight - 2) + % SwathHeight; + MaxPartialSwath = dml_max(1U, MaxPartialSwath); + + } else { + + if (ViewportYStart != 0) + dml_print( + "WARNING DML: using viewport y position of 0 even though actual viewport y position is non-zero in prefetch source lines calculation\n"); + + *MaxNumSwath = dml_ceil(*VInitPreFill / SwathHeight, 1); + + if (*VInitPreFill > 1.0) + MaxPartialSwath = (unsigned int) (*VInitPreFill - 1) % SwathHeight; + else + MaxPartialSwath = (unsigned int) (*VInitPreFill + SwathHeight - 1) + % SwathHeight; + } + + return *MaxNumSwath * SwathHeight + MaxPartialSwath; +} + +static unsigned int CalculateVMAndRowBytes( + struct display_mode_lib *mode_lib, + bool DCCEnable, + unsigned int BlockHeight256Bytes, + unsigned int BlockWidth256Bytes, + enum source_format_class SourcePixelFormat, + unsigned int SurfaceTiling, + unsigned int BytePerPixel, + enum scan_direction_class ScanDirection, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + unsigned int SwathWidth, + bool GPUVMEnable, + unsigned int VMMPageSize, + unsigned int PTEBufferSizeInRequestsLuma, + unsigned int PDEProcessingBufIn64KBReqs, + unsigned int Pitch, + unsigned int DCCMetaPitch, + unsigned int *MacroTileWidth, + unsigned int *MetaRowByte, + unsigned int *PixelPTEBytesPerRow, + bool *PTEBufferSizeNotExceeded, + unsigned int *dpte_row_height, + unsigned int *meta_row_height) +{ + unsigned int MetaRequestHeight; + unsigned int MetaRequestWidth; + unsigned int MetaSurfWidth; + unsigned int MetaSurfHeight; + unsigned int MPDEBytesFrame; + unsigned int MetaPTEBytesFrame; + unsigned int DCCMetaSurfaceBytes; + + unsigned int MacroTileSizeBytes; + unsigned int MacroTileHeight; + unsigned int DPDE0BytesFrame; + unsigned int ExtraDPDEBytesFrame; + unsigned int PDEAndMetaPTEBytesFrame; + + if (DCCEnable == true) { + MetaRequestHeight = 8 * BlockHeight256Bytes; + MetaRequestWidth = 8 * BlockWidth256Bytes; + if (ScanDirection == dm_horz) { + *meta_row_height = MetaRequestHeight; + MetaSurfWidth = dml_ceil((double) SwathWidth - 1, MetaRequestWidth) + + MetaRequestWidth; + *MetaRowByte = MetaSurfWidth * MetaRequestHeight * BytePerPixel / 256.0; + } else { + *meta_row_height = MetaRequestWidth; + MetaSurfHeight = dml_ceil((double) SwathWidth - 1, MetaRequestHeight) + + MetaRequestHeight; + *MetaRowByte = MetaSurfHeight * MetaRequestWidth * BytePerPixel / 256.0; + } + if (ScanDirection == dm_horz) { + DCCMetaSurfaceBytes = DCCMetaPitch + * (dml_ceil(ViewportHeight - 1, 64 * BlockHeight256Bytes) + + 64 * BlockHeight256Bytes) * BytePerPixel + / 256; + } else { + DCCMetaSurfaceBytes = DCCMetaPitch + * (dml_ceil( + (double) ViewportHeight - 1, + 64 * BlockHeight256Bytes) + + 64 * BlockHeight256Bytes) * BytePerPixel + / 256; + } + if (GPUVMEnable == true) { + MetaPTEBytesFrame = (dml_ceil( + (double) (DCCMetaSurfaceBytes - VMMPageSize) + / (8 * VMMPageSize), + 1) + 1) * 64; + MPDEBytesFrame = 128 * (mode_lib->vba.GPUVMMaxPageTableLevels - 1); + } else { + MetaPTEBytesFrame = 0; + MPDEBytesFrame = 0; + } + } else { + MetaPTEBytesFrame = 0; + MPDEBytesFrame = 0; + *MetaRowByte = 0; + } + + if (SurfaceTiling == dm_sw_linear || SurfaceTiling == dm_sw_gfx7_2d_thin_gl || SurfaceTiling == dm_sw_gfx7_2d_thin_l_vp) { + MacroTileSizeBytes = 256; + MacroTileHeight = BlockHeight256Bytes; + } else if (SurfaceTiling == dm_sw_4kb_s || SurfaceTiling == dm_sw_4kb_s_x + || SurfaceTiling == dm_sw_4kb_d || SurfaceTiling == dm_sw_4kb_d_x) { + MacroTileSizeBytes = 4096; + MacroTileHeight = 4 * BlockHeight256Bytes; + } else if (SurfaceTiling == dm_sw_64kb_s || SurfaceTiling == dm_sw_64kb_s_t + || SurfaceTiling == dm_sw_64kb_s_x || SurfaceTiling == dm_sw_64kb_d + || SurfaceTiling == dm_sw_64kb_d_t || SurfaceTiling == dm_sw_64kb_d_x + || SurfaceTiling == dm_sw_64kb_r_x) { + MacroTileSizeBytes = 65536; + MacroTileHeight = 16 * BlockHeight256Bytes; + } else { + MacroTileSizeBytes = 262144; + MacroTileHeight = 32 * BlockHeight256Bytes; + } + *MacroTileWidth = MacroTileSizeBytes / BytePerPixel / MacroTileHeight; + + if (GPUVMEnable == true && mode_lib->vba.GPUVMMaxPageTableLevels > 1) { + if (ScanDirection == dm_horz) { + DPDE0BytesFrame = + 64 + * (dml_ceil( + ((Pitch + * (dml_ceil( + ViewportHeight + - 1, + MacroTileHeight) + + MacroTileHeight) + * BytePerPixel) + - MacroTileSizeBytes) + / (8 + * 2097152), + 1) + 1); + } else { + DPDE0BytesFrame = + 64 + * (dml_ceil( + ((Pitch + * (dml_ceil( + (double) SwathWidth + - 1, + MacroTileHeight) + + MacroTileHeight) + * BytePerPixel) + - MacroTileSizeBytes) + / (8 + * 2097152), + 1) + 1); + } + ExtraDPDEBytesFrame = 128 * (mode_lib->vba.GPUVMMaxPageTableLevels - 2); + } else { + DPDE0BytesFrame = 0; + ExtraDPDEBytesFrame = 0; + } + + PDEAndMetaPTEBytesFrame = MetaPTEBytesFrame + MPDEBytesFrame + DPDE0BytesFrame + + ExtraDPDEBytesFrame; + + if (GPUVMEnable == true) { + unsigned int PTERequestSize; + unsigned int PixelPTEReqHeight; + unsigned int PixelPTEReqWidth; + double FractionOfPTEReturnDrop; + unsigned int EffectivePDEProcessingBufIn64KBReqs; + + if (SurfaceTiling == dm_sw_linear) { + PixelPTEReqHeight = 1; + PixelPTEReqWidth = 8.0 * VMMPageSize / BytePerPixel; + PTERequestSize = 64; + FractionOfPTEReturnDrop = 0; + } else if (MacroTileSizeBytes == 4096) { + PixelPTEReqHeight = MacroTileHeight; + PixelPTEReqWidth = 8 * *MacroTileWidth; + PTERequestSize = 64; + if (ScanDirection == dm_horz) + FractionOfPTEReturnDrop = 0; + else + FractionOfPTEReturnDrop = 7 / 8; + } else if (VMMPageSize == 4096 && MacroTileSizeBytes > 4096) { + PixelPTEReqHeight = 16 * BlockHeight256Bytes; + PixelPTEReqWidth = 16 * BlockWidth256Bytes; + PTERequestSize = 128; + FractionOfPTEReturnDrop = 0; + } else { + PixelPTEReqHeight = MacroTileHeight; + PixelPTEReqWidth = 8 * *MacroTileWidth; + PTERequestSize = 64; + FractionOfPTEReturnDrop = 0; + } + + if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) + EffectivePDEProcessingBufIn64KBReqs = PDEProcessingBufIn64KBReqs / 2; + else + EffectivePDEProcessingBufIn64KBReqs = PDEProcessingBufIn64KBReqs; + + if (SurfaceTiling == dm_sw_linear) { + *dpte_row_height = + dml_min( + 128, + 1 + << (unsigned int) dml_floor( + dml_log2( + dml_min( + (double) PTEBufferSizeInRequestsLuma + * PixelPTEReqWidth, + EffectivePDEProcessingBufIn64KBReqs + * 65536.0 + / BytePerPixel) + / Pitch), + 1)); + *PixelPTEBytesPerRow = PTERequestSize + * (dml_ceil( + (double) (Pitch * *dpte_row_height - 1) + / PixelPTEReqWidth, + 1) + 1); + } else if (ScanDirection == dm_horz) { + *dpte_row_height = PixelPTEReqHeight; + *PixelPTEBytesPerRow = PTERequestSize + * (dml_ceil(((double) SwathWidth - 1) / PixelPTEReqWidth, 1) + + 1); + } else { + *dpte_row_height = dml_min(PixelPTEReqWidth, *MacroTileWidth); + *PixelPTEBytesPerRow = PTERequestSize + * (dml_ceil( + ((double) SwathWidth - 1) + / PixelPTEReqHeight, + 1) + 1); + } + if (*PixelPTEBytesPerRow * (1 - FractionOfPTEReturnDrop) + <= 64 * PTEBufferSizeInRequestsLuma) { + *PTEBufferSizeNotExceeded = true; + } else { + *PTEBufferSizeNotExceeded = false; + } + } else { + *PixelPTEBytesPerRow = 0; + *PTEBufferSizeNotExceeded = true; + } + + return PDEAndMetaPTEBytesFrame; +} + +static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation( + struct display_mode_lib *mode_lib) +{ + unsigned int j, k; + + mode_lib->vba.WritebackDISPCLK = 0.0; + mode_lib->vba.DISPCLKWithRamping = 0; + mode_lib->vba.DISPCLKWithoutRamping = 0; + mode_lib->vba.GlobalDPPCLK = 0.0; + + // dml_ml->vba.DISPCLK and dml_ml->vba.DPPCLK Calculation + // + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.WritebackEnable[k]) { + mode_lib->vba.WritebackDISPCLK = + dml_max( + mode_lib->vba.WritebackDISPCLK, + CalculateWriteBackDISPCLK( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.PixelClock[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.WritebackChromaLineBufferWidth)); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.HRatio[k] > 1) { + mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / dml_ceil( + mode_lib->vba.htaps[k] + / 6.0, + 1)); + } else { + mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + + mode_lib->vba.DPPCLKUsingSingleDPPLuma = + mode_lib->vba.PixelClock[k] + * dml_max( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + dml_max( + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / mode_lib->vba.PSCL_THROUGHPUT_LUMA[k], + 1.0)); + + if ((mode_lib->vba.htaps[k] > 6 || mode_lib->vba.vtaps[k] > 6) + && mode_lib->vba.DPPCLKUsingSingleDPPLuma + < 2 * mode_lib->vba.PixelClock[k]) { + mode_lib->vba.DPPCLKUsingSingleDPPLuma = 2 * mode_lib->vba.PixelClock[k]; + } + + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] = 0.0; + mode_lib->vba.DPPCLKUsingSingleDPP[k] = + mode_lib->vba.DPPCLKUsingSingleDPPLuma; + } else { + if (mode_lib->vba.HRatio[k] > 1) { + mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] = + dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / 2 + / dml_ceil( + mode_lib->vba.HTAPsChroma[k] + / 6.0, + 1.0)); + } else { + mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + mode_lib->vba.DPPCLKUsingSingleDPPChroma = + mode_lib->vba.PixelClock[k] + * dml_max( + mode_lib->vba.VTAPsChroma[k] + / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k] + / 2), + dml_max( + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / 4 + / mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k], + 1.0)); + + if ((mode_lib->vba.HTAPsChroma[k] > 6 || mode_lib->vba.VTAPsChroma[k] > 6) + && mode_lib->vba.DPPCLKUsingSingleDPPChroma + < 2 * mode_lib->vba.PixelClock[k]) { + mode_lib->vba.DPPCLKUsingSingleDPPChroma = 2 + * mode_lib->vba.PixelClock[k]; + } + + mode_lib->vba.DPPCLKUsingSingleDPP[k] = dml_max( + mode_lib->vba.DPPCLKUsingSingleDPPLuma, + mode_lib->vba.DPPCLKUsingSingleDPPChroma); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] != k) + continue; + if (mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DISPCLKWithRamping = + dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.PixelClock[k] / 2 + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100) + * (1 + + mode_lib->vba.DISPCLKRampingMargin + / 100)); + mode_lib->vba.DISPCLKWithoutRamping = + dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.PixelClock[k] / 2 + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100)); + } else if (!mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DISPCLKWithRamping = + dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.PixelClock[k] + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100) + * (1 + + mode_lib->vba.DISPCLKRampingMargin + / 100)); + mode_lib->vba.DISPCLKWithoutRamping = + dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.PixelClock[k] + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100)); + } + } + + mode_lib->vba.DISPCLKWithRamping = dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.WritebackDISPCLK); + mode_lib->vba.DISPCLKWithoutRamping = dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.WritebackDISPCLK); + + ASSERT(mode_lib->vba.DISPCLKDPPCLKVCOSpeed != 0); + mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity = RoundToDFSGranularityUp( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity = RoundToDFSGranularityUp( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.MaxDispclkRoundedToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.soc.clock_limits[mode_lib->vba.soc.num_states].dispclk_mhz, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + if (mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity + > mode_lib->vba.MaxDispclkRoundedToDFSGranularity) { + mode_lib->vba.DISPCLK_calculated = + mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity; + } else if (mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity + > mode_lib->vba.MaxDispclkRoundedToDFSGranularity) { + mode_lib->vba.DISPCLK_calculated = mode_lib->vba.MaxDispclkRoundedToDFSGranularity; + } else { + mode_lib->vba.DISPCLK_calculated = + mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity; + } + DTRACE(" dispclk_mhz (calculated) = %f", mode_lib->vba.DISPCLK_calculated); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.DPPPerPlane[k] == 0) { + mode_lib->vba.DPPCLK_calculated[k] = 0; + } else { + mode_lib->vba.DPPCLK_calculated[k] = mode_lib->vba.DPPCLKUsingSingleDPP[k] + / mode_lib->vba.DPPPerPlane[k] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100); + } + mode_lib->vba.GlobalDPPCLK = dml_max( + mode_lib->vba.GlobalDPPCLK, + mode_lib->vba.DPPCLK_calculated[k]); + } + mode_lib->vba.GlobalDPPCLK = RoundToDFSGranularityUp( + mode_lib->vba.GlobalDPPCLK, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.DPPCLK_calculated[k] = mode_lib->vba.GlobalDPPCLK / 255 + * dml_ceil( + mode_lib->vba.DPPCLK_calculated[k] * 255 + / mode_lib->vba.GlobalDPPCLK, + 1); + DTRACE(" dppclk_mhz[%i] (calculated) = %f", k, mode_lib->vba.DPPCLK_calculated[k]); + } + + // Urgent Watermark + mode_lib->vba.DCCEnabledAnyPlane = false; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + if (mode_lib->vba.DCCEnable[k]) + mode_lib->vba.DCCEnabledAnyPlane = true; + + mode_lib->vba.ReturnBandwidthToDCN = dml_min( + mode_lib->vba.ReturnBusWidth * mode_lib->vba.DCFCLK, + mode_lib->vba.FabricAndDRAMBandwidth * 1000) + * mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelDataOnly / 100; + + mode_lib->vba.ReturnBW = mode_lib->vba.ReturnBandwidthToDCN; + mode_lib->vba.ReturnBW = adjust_ReturnBW( + mode_lib, + mode_lib->vba.ReturnBW, + mode_lib->vba.DCCEnabledAnyPlane, + mode_lib->vba.ReturnBandwidthToDCN); + + // Let's do this calculation again?? + mode_lib->vba.ReturnBandwidthToDCN = dml_min( + mode_lib->vba.ReturnBusWidth * mode_lib->vba.DCFCLK, + mode_lib->vba.FabricAndDRAMBandwidth * 1000); + mode_lib->vba.ReturnBW = adjust_ReturnBW( + mode_lib, + mode_lib->vba.ReturnBW, + mode_lib->vba.DCCEnabledAnyPlane, + mode_lib->vba.ReturnBandwidthToDCN); + + DTRACE(" dcfclk_mhz = %f", mode_lib->vba.DCFCLK); + DTRACE(" return_bw_to_dcn = %f", mode_lib->vba.ReturnBandwidthToDCN); + DTRACE(" return_bus_bw = %f", mode_lib->vba.ReturnBW); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + bool MainPlaneDoesODMCombine = false; + + if (mode_lib->vba.SourceScan[k] == dm_horz) + mode_lib->vba.SwathWidthSingleDPPY[k] = mode_lib->vba.ViewportWidth[k]; + else + mode_lib->vba.SwathWidthSingleDPPY[k] = mode_lib->vba.ViewportHeight[k]; + + if (mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) + MainPlaneDoesODMCombine = true; + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) + if (mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) + MainPlaneDoesODMCombine = true; + + if (MainPlaneDoesODMCombine == true) + mode_lib->vba.SwathWidthY[k] = dml_min( + (double) mode_lib->vba.SwathWidthSingleDPPY[k], + dml_round( + mode_lib->vba.HActive[k] / 2.0 + * mode_lib->vba.HRatio[k])); + else { + if (mode_lib->vba.DPPPerPlane[k] == 0) { + mode_lib->vba.SwathWidthY[k] = 0; + } else { + mode_lib->vba.SwathWidthY[k] = mode_lib->vba.SwathWidthSingleDPPY[k] + / mode_lib->vba.DPPPerPlane[k]; + } + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + mode_lib->vba.BytePerPixelDETY[k] = 8; + mode_lib->vba.BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + mode_lib->vba.BytePerPixelDETY[k] = 4; + mode_lib->vba.BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + mode_lib->vba.BytePerPixelDETY[k] = 2; + mode_lib->vba.BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8) { + mode_lib->vba.BytePerPixelDETY[k] = 1; + mode_lib->vba.BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + mode_lib->vba.BytePerPixelDETY[k] = 1; + mode_lib->vba.BytePerPixelDETC[k] = 2; + } else { // dm_420_10 + mode_lib->vba.BytePerPixelDETY[k] = 4.0 / 3.0; + mode_lib->vba.BytePerPixelDETC[k] = 8.0 / 3.0; + } + } + + mode_lib->vba.TotalDataReadBandwidth = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.ReadBandwidthPlaneLuma[k] = mode_lib->vba.SwathWidthSingleDPPY[k] + * dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k]; + mode_lib->vba.ReadBandwidthPlaneChroma[k] = mode_lib->vba.SwathWidthSingleDPPY[k] + / 2 * dml_ceil(mode_lib->vba.BytePerPixelDETC[k], 2) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k] / 2; + DTRACE( + " read_bw[%i] = %fBps", + k, + mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k]); + mode_lib->vba.TotalDataReadBandwidth += mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k]; + } + + mode_lib->vba.TotalDCCActiveDPP = 0; + mode_lib->vba.TotalActiveDPP = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.TotalActiveDPP = mode_lib->vba.TotalActiveDPP + + mode_lib->vba.DPPPerPlane[k]; + if (mode_lib->vba.DCCEnable[k]) + mode_lib->vba.TotalDCCActiveDPP = mode_lib->vba.TotalDCCActiveDPP + + mode_lib->vba.DPPPerPlane[k]; + } + + mode_lib->vba.UrgentRoundTripAndOutOfOrderLatency = + (mode_lib->vba.RoundTripPingLatencyCycles + 32) / mode_lib->vba.DCFCLK + + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelDataOnly + * mode_lib->vba.NumberOfChannels + / mode_lib->vba.ReturnBW; + + mode_lib->vba.LastPixelOfLineExtraWatermark = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.VRatio[k] <= 1.0) + mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k] = + (double) mode_lib->vba.SwathWidthY[k] + * mode_lib->vba.DPPPerPlane[k] + / mode_lib->vba.HRatio[k] + / mode_lib->vba.PixelClock[k]; + else + mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k] = + (double) mode_lib->vba.SwathWidthY[k] + / mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] + / mode_lib->vba.DPPCLK[k]; + + if (mode_lib->vba.BytePerPixelDETC[k] == 0) + mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k] = 0.0; + else if (mode_lib->vba.VRatio[k] / 2.0 <= 1.0) + mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k] = + mode_lib->vba.SwathWidthY[k] / 2.0 + * mode_lib->vba.DPPPerPlane[k] + / (mode_lib->vba.HRatio[k] / 2.0) + / mode_lib->vba.PixelClock[k]; + else + mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k] = + mode_lib->vba.SwathWidthY[k] / 2.0 + / mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] + / mode_lib->vba.DPPCLK[k]; + } + + mode_lib->vba.UrgentExtraLatency = mode_lib->vba.UrgentRoundTripAndOutOfOrderLatency + + (mode_lib->vba.TotalActiveDPP * mode_lib->vba.PixelChunkSizeInKByte + + mode_lib->vba.TotalDCCActiveDPP + * mode_lib->vba.MetaChunkSize) * 1024.0 + / mode_lib->vba.ReturnBW; + + if (mode_lib->vba.GPUVMEnable) + mode_lib->vba.UrgentExtraLatency += mode_lib->vba.TotalActiveDPP + * mode_lib->vba.PTEGroupSize / mode_lib->vba.ReturnBW; + + mode_lib->vba.UrgentWatermark = mode_lib->vba.UrgentLatencyPixelDataOnly + + mode_lib->vba.LastPixelOfLineExtraWatermark + + mode_lib->vba.UrgentExtraLatency; + + DTRACE(" urgent_extra_latency = %fus", mode_lib->vba.UrgentExtraLatency); + DTRACE(" wm_urgent = %fus", mode_lib->vba.UrgentWatermark); + + mode_lib->vba.UrgentLatency = mode_lib->vba.UrgentLatencyPixelDataOnly; + + mode_lib->vba.TotalActiveWriteback = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.WritebackEnable[k]) + mode_lib->vba.TotalActiveWriteback = mode_lib->vba.TotalActiveWriteback + mode_lib->vba.ActiveWritebacksPerPlane[k]; + } + + if (mode_lib->vba.TotalActiveWriteback <= 1) + mode_lib->vba.WritebackUrgentWatermark = mode_lib->vba.WritebackLatency; + else + mode_lib->vba.WritebackUrgentWatermark = mode_lib->vba.WritebackLatency + + mode_lib->vba.WritebackChunkSize * 1024.0 / 32 + / mode_lib->vba.SOCCLK; + + DTRACE(" wm_wb_urgent = %fus", mode_lib->vba.WritebackUrgentWatermark); + + // NB P-State/DRAM Clock Change Watermark + mode_lib->vba.DRAMClockChangeWatermark = mode_lib->vba.DRAMClockChangeLatency + + mode_lib->vba.UrgentWatermark; + + DTRACE(" wm_pstate_change = %fus", mode_lib->vba.DRAMClockChangeWatermark); + + DTRACE(" calculating wb pstate watermark"); + DTRACE(" total wb outputs %d", mode_lib->vba.TotalActiveWriteback); + DTRACE(" socclk frequency %f Mhz", mode_lib->vba.SOCCLK); + + if (mode_lib->vba.TotalActiveWriteback <= 1) + mode_lib->vba.WritebackDRAMClockChangeWatermark = + mode_lib->vba.DRAMClockChangeLatency + + mode_lib->vba.WritebackLatency; + else + mode_lib->vba.WritebackDRAMClockChangeWatermark = + mode_lib->vba.DRAMClockChangeLatency + + mode_lib->vba.WritebackLatency + + mode_lib->vba.WritebackChunkSize * 1024.0 / 32 + / mode_lib->vba.SOCCLK; + + DTRACE(" wm_wb_pstate %fus", mode_lib->vba.WritebackDRAMClockChangeWatermark); + + // Stutter Efficiency + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.LinesInDETY[k] = mode_lib->vba.DETBufferSizeY[k] + / mode_lib->vba.BytePerPixelDETY[k] / mode_lib->vba.SwathWidthY[k]; + mode_lib->vba.LinesInDETYRoundedDownToSwath[k] = dml_floor( + mode_lib->vba.LinesInDETY[k], + mode_lib->vba.SwathHeightY[k]); + mode_lib->vba.FullDETBufferingTimeY[k] = + mode_lib->vba.LinesInDETYRoundedDownToSwath[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + / mode_lib->vba.VRatio[k]; + if (mode_lib->vba.BytePerPixelDETC[k] > 0) { + mode_lib->vba.LinesInDETC[k] = mode_lib->vba.DETBufferSizeC[k] + / mode_lib->vba.BytePerPixelDETC[k] + / (mode_lib->vba.SwathWidthY[k] / 2); + mode_lib->vba.LinesInDETCRoundedDownToSwath[k] = dml_floor( + mode_lib->vba.LinesInDETC[k], + mode_lib->vba.SwathHeightC[k]); + mode_lib->vba.FullDETBufferingTimeC[k] = + mode_lib->vba.LinesInDETCRoundedDownToSwath[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + / (mode_lib->vba.VRatio[k] / 2); + } else { + mode_lib->vba.LinesInDETC[k] = 0; + mode_lib->vba.LinesInDETCRoundedDownToSwath[k] = 0; + mode_lib->vba.FullDETBufferingTimeC[k] = 999999; + } + } + + mode_lib->vba.MinFullDETBufferingTime = 999999.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.FullDETBufferingTimeY[k] + < mode_lib->vba.MinFullDETBufferingTime) { + mode_lib->vba.MinFullDETBufferingTime = + mode_lib->vba.FullDETBufferingTimeY[k]; + mode_lib->vba.FrameTimeForMinFullDETBufferingTime = + (double) mode_lib->vba.VTotal[k] * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } + if (mode_lib->vba.FullDETBufferingTimeC[k] + < mode_lib->vba.MinFullDETBufferingTime) { + mode_lib->vba.MinFullDETBufferingTime = + mode_lib->vba.FullDETBufferingTimeC[k]; + mode_lib->vba.FrameTimeForMinFullDETBufferingTime = + (double) mode_lib->vba.VTotal[k] * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } + } + + mode_lib->vba.AverageReadBandwidthGBytePerSecond = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.DCCEnable[k]) { + mode_lib->vba.AverageReadBandwidthGBytePerSecond = + mode_lib->vba.AverageReadBandwidthGBytePerSecond + + mode_lib->vba.ReadBandwidthPlaneLuma[k] + / mode_lib->vba.DCCRate[k] + / 1000 + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + / mode_lib->vba.DCCRate[k] + / 1000; + } else { + mode_lib->vba.AverageReadBandwidthGBytePerSecond = + mode_lib->vba.AverageReadBandwidthGBytePerSecond + + mode_lib->vba.ReadBandwidthPlaneLuma[k] + / 1000 + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + / 1000; + } + if (mode_lib->vba.DCCEnable[k]) { + mode_lib->vba.AverageReadBandwidthGBytePerSecond = + mode_lib->vba.AverageReadBandwidthGBytePerSecond + + mode_lib->vba.ReadBandwidthPlaneLuma[k] + / 1000 / 256 + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + / 1000 / 256; + } + if (mode_lib->vba.GPUVMEnable) { + mode_lib->vba.AverageReadBandwidthGBytePerSecond = + mode_lib->vba.AverageReadBandwidthGBytePerSecond + + mode_lib->vba.ReadBandwidthPlaneLuma[k] + / 1000 / 512 + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + / 1000 / 512; + } + } + + mode_lib->vba.PartOfBurstThatFitsInROB = + dml_min( + mode_lib->vba.MinFullDETBufferingTime + * mode_lib->vba.TotalDataReadBandwidth, + mode_lib->vba.ROBBufferSizeInKByte * 1024 + * mode_lib->vba.TotalDataReadBandwidth + / (mode_lib->vba.AverageReadBandwidthGBytePerSecond + * 1000)); + mode_lib->vba.StutterBurstTime = mode_lib->vba.PartOfBurstThatFitsInROB + * (mode_lib->vba.AverageReadBandwidthGBytePerSecond * 1000) + / mode_lib->vba.TotalDataReadBandwidth / mode_lib->vba.ReturnBW + + (mode_lib->vba.MinFullDETBufferingTime + * mode_lib->vba.TotalDataReadBandwidth + - mode_lib->vba.PartOfBurstThatFitsInROB) + / (mode_lib->vba.DCFCLK * 64); + if (mode_lib->vba.TotalActiveWriteback == 0) { + mode_lib->vba.StutterEfficiencyNotIncludingVBlank = (1 + - (mode_lib->vba.SRExitTime + mode_lib->vba.StutterBurstTime) + / mode_lib->vba.MinFullDETBufferingTime) * 100; + } else { + mode_lib->vba.StutterEfficiencyNotIncludingVBlank = 0; + } + + mode_lib->vba.SmallestVBlank = 999999; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.SynchronizedVBlank || mode_lib->vba.NumberOfActivePlanes == 1) { + mode_lib->vba.VBlankTime = (double) (mode_lib->vba.VTotal[k] + - mode_lib->vba.VActive[k]) * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.VBlankTime = 0; + } + mode_lib->vba.SmallestVBlank = dml_min( + mode_lib->vba.SmallestVBlank, + mode_lib->vba.VBlankTime); + } + + mode_lib->vba.StutterEfficiency = (mode_lib->vba.StutterEfficiencyNotIncludingVBlank / 100 + * (mode_lib->vba.FrameTimeForMinFullDETBufferingTime + - mode_lib->vba.SmallestVBlank) + + mode_lib->vba.SmallestVBlank) + / mode_lib->vba.FrameTimeForMinFullDETBufferingTime * 100; + + // dml_ml->vba.DCFCLK Deep Sleep + mode_lib->vba.DCFCLKDeepSleep = 8.0; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; k++) { + if (mode_lib->vba.BytePerPixelDETC[k] > 0) { + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = + dml_max( + 1.1 * mode_lib->vba.SwathWidthY[k] + * dml_ceil( + mode_lib->vba.BytePerPixelDETY[k], + 1) / 32 + / mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k], + 1.1 * mode_lib->vba.SwathWidthY[k] / 2.0 + * dml_ceil( + mode_lib->vba.BytePerPixelDETC[k], + 2) / 32 + / mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k]); + } else + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = 1.1 * mode_lib->vba.SwathWidthY[k] + * dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1) / 64.0 + / mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k]; + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = dml_max( + mode_lib->vba.DCFCLKDeepSleepPerPlane[k], + mode_lib->vba.PixelClock[k] / 16.0); + mode_lib->vba.DCFCLKDeepSleep = dml_max( + mode_lib->vba.DCFCLKDeepSleep, + mode_lib->vba.DCFCLKDeepSleepPerPlane[k]); + + DTRACE( + " dcfclk_deepsleep_per_plane[%i] = %fMHz", + k, + mode_lib->vba.DCFCLKDeepSleepPerPlane[k]); + } + + DTRACE(" dcfclk_deepsleep_mhz = %fMHz", mode_lib->vba.DCFCLKDeepSleep); + + // Stutter Watermark + mode_lib->vba.StutterExitWatermark = mode_lib->vba.SRExitTime + + mode_lib->vba.LastPixelOfLineExtraWatermark + + mode_lib->vba.UrgentExtraLatency + 10 / mode_lib->vba.DCFCLKDeepSleep; + mode_lib->vba.StutterEnterPlusExitWatermark = mode_lib->vba.SREnterPlusExitTime + + mode_lib->vba.LastPixelOfLineExtraWatermark + + mode_lib->vba.UrgentExtraLatency; + + DTRACE(" wm_cstate_exit = %fus", mode_lib->vba.StutterExitWatermark); + DTRACE(" wm_cstate_enter_exit = %fus", mode_lib->vba.StutterEnterPlusExitWatermark); + + // Urgent Latency Supported + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.EffectiveDETPlusLBLinesLuma = + dml_floor( + mode_lib->vba.LinesInDETY[k] + + dml_min( + mode_lib->vba.LinesInDETY[k] + * mode_lib->vba.DPPCLK[k] + * mode_lib->vba.BytePerPixelDETY[k] + * mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] + / (mode_lib->vba.ReturnBW + / mode_lib->vba.DPPPerPlane[k]), + (double) mode_lib->vba.EffectiveLBLatencyHidingSourceLinesLuma), + mode_lib->vba.SwathHeightY[k]); + + mode_lib->vba.UrgentLatencySupportUsLuma = mode_lib->vba.EffectiveDETPlusLBLinesLuma + * (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + / mode_lib->vba.VRatio[k] + - mode_lib->vba.EffectiveDETPlusLBLinesLuma + * mode_lib->vba.SwathWidthY[k] + * mode_lib->vba.BytePerPixelDETY[k] + / (mode_lib->vba.ReturnBW + / mode_lib->vba.DPPPerPlane[k]); + + if (mode_lib->vba.BytePerPixelDETC[k] > 0) { + mode_lib->vba.EffectiveDETPlusLBLinesChroma = + dml_floor( + mode_lib->vba.LinesInDETC[k] + + dml_min( + mode_lib->vba.LinesInDETC[k] + * mode_lib->vba.DPPCLK[k] + * mode_lib->vba.BytePerPixelDETC[k] + * mode_lib->vba.PSCL_THROUGHPUT_CHROMA[k] + / (mode_lib->vba.ReturnBW + / mode_lib->vba.DPPPerPlane[k]), + (double) mode_lib->vba.EffectiveLBLatencyHidingSourceLinesChroma), + mode_lib->vba.SwathHeightC[k]); + mode_lib->vba.UrgentLatencySupportUsChroma = + mode_lib->vba.EffectiveDETPlusLBLinesChroma + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + / (mode_lib->vba.VRatio[k] / 2) + - mode_lib->vba.EffectiveDETPlusLBLinesChroma + * (mode_lib->vba.SwathWidthY[k] + / 2) + * mode_lib->vba.BytePerPixelDETC[k] + / (mode_lib->vba.ReturnBW + / mode_lib->vba.DPPPerPlane[k]); + mode_lib->vba.UrgentLatencySupportUs[k] = dml_min( + mode_lib->vba.UrgentLatencySupportUsLuma, + mode_lib->vba.UrgentLatencySupportUsChroma); + } else { + mode_lib->vba.UrgentLatencySupportUs[k] = + mode_lib->vba.UrgentLatencySupportUsLuma; + } + } + + mode_lib->vba.MinUrgentLatencySupportUs = 999999; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.MinUrgentLatencySupportUs = dml_min( + mode_lib->vba.MinUrgentLatencySupportUs, + mode_lib->vba.UrgentLatencySupportUs[k]); + } + + // Non-Urgent Latency Tolerance + mode_lib->vba.NonUrgentLatencyTolerance = mode_lib->vba.MinUrgentLatencySupportUs + - mode_lib->vba.UrgentWatermark; + + // DSCCLK + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if ((mode_lib->vba.BlendingAndTiming[k] != k) || !mode_lib->vba.DSCEnabled[k]) { + mode_lib->vba.DSCCLK_calculated[k] = 0.0; + } else { + if (mode_lib->vba.OutputFormat[k] == dm_420 + || mode_lib->vba.OutputFormat[k] == dm_n422) + mode_lib->vba.DSCFormatFactor = 2; + else + mode_lib->vba.DSCFormatFactor = 1; + if (mode_lib->vba.ODMCombineEnabled[k]) + mode_lib->vba.DSCCLK_calculated[k] = + mode_lib->vba.PixelClockBackEnd[k] / 6 + / mode_lib->vba.DSCFormatFactor + / (1 + - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100); + else + mode_lib->vba.DSCCLK_calculated[k] = + mode_lib->vba.PixelClockBackEnd[k] / 3 + / mode_lib->vba.DSCFormatFactor + / (1 + - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100); + } + } + + // DSC Delay + // TODO + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + double bpp = mode_lib->vba.OutputBpp[k]; + unsigned int slices = mode_lib->vba.NumberOfDSCSlices[k]; + + if (mode_lib->vba.DSCEnabled[k] && bpp != 0) { + if (!mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DSCDelay[k] = + dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + bpp, + dml_ceil( + (double) mode_lib->vba.HActive[k] + / mode_lib->vba.NumberOfDSCSlices[k], + 1), + slices, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k]); + } else { + mode_lib->vba.DSCDelay[k] = + 2 + * (dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + bpp, + dml_ceil( + (double) mode_lib->vba.HActive[k] + / mode_lib->vba.NumberOfDSCSlices[k], + 1), + slices / 2.0, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k])); + } + mode_lib->vba.DSCDelay[k] = mode_lib->vba.DSCDelay[k] + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.PixelClockBackEnd[k]; + } else { + mode_lib->vba.DSCDelay[k] = 0; + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) // NumberOfPlanes + if (j != k && mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.DSCEnabled[j]) + mode_lib->vba.DSCDelay[k] = mode_lib->vba.DSCDelay[j]; + + // Prefetch + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + unsigned int PDEAndMetaPTEBytesFrameY; + unsigned int PixelPTEBytesPerRowY; + unsigned int MetaRowByteY; + unsigned int MetaRowByteC; + unsigned int PDEAndMetaPTEBytesFrameC; + unsigned int PixelPTEBytesPerRowC; + + Calculate256BBlockSizes( + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1), + dml_ceil(mode_lib->vba.BytePerPixelDETC[k], 2), + &mode_lib->vba.BlockHeight256BytesY[k], + &mode_lib->vba.BlockHeight256BytesC[k], + &mode_lib->vba.BlockWidth256BytesY[k], + &mode_lib->vba.BlockWidth256BytesC[k]); + PDEAndMetaPTEBytesFrameY = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + mode_lib->vba.BlockHeight256BytesY[k], + mode_lib->vba.BlockWidth256BytesY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + mode_lib->vba.SwathWidthY[k], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsLuma, + mode_lib->vba.PDEProcessingBufIn64KBReqs, + mode_lib->vba.PitchY[k], + mode_lib->vba.DCCMetaPitchY[k], + &mode_lib->vba.MacroTileWidthY[k], + &MetaRowByteY, + &PixelPTEBytesPerRowY, + &mode_lib->vba.PTEBufferSizeNotExceeded[mode_lib->vba.VoltageLevel][0], + &mode_lib->vba.dpte_row_height[k], + &mode_lib->vba.meta_row_height[k]); + mode_lib->vba.PrefetchSourceLinesY[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.vtaps[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.ViewportYStartY[k], + &mode_lib->vba.VInitPreFillY[k], + &mode_lib->vba.MaxNumSwathY[k]); + + if ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_8)) { + PDEAndMetaPTEBytesFrameC = + CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + mode_lib->vba.BlockHeight256BytesC[k], + mode_lib->vba.BlockWidth256BytesC[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil( + mode_lib->vba.BytePerPixelDETC[k], + 2), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k] / 2, + mode_lib->vba.ViewportHeight[k] / 2, + mode_lib->vba.SwathWidthY[k] / 2, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsLuma, + mode_lib->vba.PDEProcessingBufIn64KBReqs, + mode_lib->vba.PitchC[k], + 0, + &mode_lib->vba.MacroTileWidthC[k], + &MetaRowByteC, + &PixelPTEBytesPerRowC, + &mode_lib->vba.PTEBufferSizeNotExceeded[mode_lib->vba.VoltageLevel][0], + &mode_lib->vba.dpte_row_height_chroma[k], + &mode_lib->vba.meta_row_height_chroma[k]); + mode_lib->vba.PrefetchSourceLinesC[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k] / 2, + mode_lib->vba.VTAPsChroma[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightC[k], + mode_lib->vba.ViewportYStartC[k], + &mode_lib->vba.VInitPreFillC[k], + &mode_lib->vba.MaxNumSwathC[k]); + } else { + PixelPTEBytesPerRowC = 0; + PDEAndMetaPTEBytesFrameC = 0; + MetaRowByteC = 0; + mode_lib->vba.MaxNumSwathC[k] = 0; + mode_lib->vba.PrefetchSourceLinesC[k] = 0; + } + + mode_lib->vba.PixelPTEBytesPerRow[k] = PixelPTEBytesPerRowY + PixelPTEBytesPerRowC; + mode_lib->vba.PDEAndMetaPTEBytesFrame[k] = PDEAndMetaPTEBytesFrameY + + PDEAndMetaPTEBytesFrameC; + mode_lib->vba.MetaRowByte[k] = MetaRowByteY + MetaRowByteC; + + CalculateActiveRowBandwidth( + mode_lib->vba.GPUVMEnable, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + MetaRowByteY, + MetaRowByteC, + mode_lib->vba.meta_row_height[k], + mode_lib->vba.meta_row_height_chroma[k], + PixelPTEBytesPerRowY, + PixelPTEBytesPerRowC, + mode_lib->vba.dpte_row_height[k], + mode_lib->vba.dpte_row_height_chroma[k], + &mode_lib->vba.meta_row_bw[k], + &mode_lib->vba.dpte_row_bw[k], + &mode_lib->vba.qual_row_bw[k]); + } + + mode_lib->vba.TCalc = 24.0 / mode_lib->vba.DCFCLKDeepSleep; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] = + mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k]) + / mode_lib->vba.DISPCLK; + } else + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] = 0; + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { + if (mode_lib->vba.BlendingAndTiming[j] == k + && mode_lib->vba.WritebackEnable[j] == true) { + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] = + dml_max( + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k], + mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[j], + mode_lib->vba.WritebackHRatio[j], + mode_lib->vba.WritebackVRatio[j], + mode_lib->vba.WritebackLumaHTaps[j], + mode_lib->vba.WritebackLumaVTaps[j], + mode_lib->vba.WritebackChromaHTaps[j], + mode_lib->vba.WritebackChromaVTaps[j], + mode_lib->vba.WritebackDestinationWidth[j]) + / mode_lib->vba.DISPCLK); + } + } + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) + if (mode_lib->vba.BlendingAndTiming[k] == j) + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] = + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][j]; + + mode_lib->vba.VStartupLines = 13; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.MaxVStartupLines[k] = + mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] + - dml_max( + 1.0, + dml_ceil( + mode_lib->vba.WritebackDelay[mode_lib->vba.VoltageLevel][k] + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1)); + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + mode_lib->vba.MaximumMaxVStartupLines = dml_max( + mode_lib->vba.MaximumMaxVStartupLines, + mode_lib->vba.MaxVStartupLines[k]); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.cursor_bw[k] = 0.0; + for (j = 0; j < mode_lib->vba.NumberOfCursors[k]; ++j) + mode_lib->vba.cursor_bw[k] += mode_lib->vba.CursorWidth[k][j] + * mode_lib->vba.CursorBPP[k][j] / 8.0 + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k]; + } + + do { + double MaxTotalRDBandwidth = 0; + bool DestinationLineTimesForPrefetchLessThan2 = false; + bool VRatioPrefetchMoreThan4 = false; + bool prefetch_vm_bw_valid = true; + bool prefetch_row_bw_valid = true; + double TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.SREnterPlusExitTime); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.XFCEnabled[k] == true) { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = + CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.SwathWidthY[k], + dml_ceil( + mode_lib->vba.BytePerPixelDETY[k], + 1), + mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TCalc, + TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + } else { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = 0; + } + + CalculateDelayAfterScaler(mode_lib, mode_lib->vba.ReturnBW, mode_lib->vba.ReadBandwidthPlaneLuma[k], mode_lib->vba.ReadBandwidthPlaneChroma[k], mode_lib->vba.TotalDataReadBandwidth, + mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k], mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k], + mode_lib->vba.DPPCLK[k], mode_lib->vba.DISPCLK, mode_lib->vba.PixelClock[k], mode_lib->vba.DSCDelay[k], mode_lib->vba.DPPPerPlane[k], mode_lib->vba.ScalerEnabled[k], mode_lib->vba.NumberOfCursors[k], + mode_lib->vba.DPPCLKDelaySubtotal, mode_lib->vba.DPPCLKDelaySCL, mode_lib->vba.DPPCLKDelaySCLLBOnly, mode_lib->vba.DPPCLKDelayCNVCFormater, mode_lib->vba.DPPCLKDelayCNVCCursor, mode_lib->vba.DISPCLKDelaySubtotal, + mode_lib->vba.SwathWidthY[k] / mode_lib->vba.HRatio[k], mode_lib->vba.OutputFormat[k], mode_lib->vba.HTotal[k], + mode_lib->vba.SwathWidthSingleDPPY[k], mode_lib->vba.BytePerPixelDETY[k], mode_lib->vba.BytePerPixelDETC[k], mode_lib->vba.SwathHeightY[k], mode_lib->vba.SwathHeightC[k], mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, &mode_lib->vba.DSTXAfterScaler[k], &mode_lib->vba.DSTYAfterScaler[k]); + + mode_lib->vba.ErrorResult[k] = + CalculatePrefetchSchedule( + mode_lib, + mode_lib->vba.DPPCLK[k], + mode_lib->vba.DISPCLK, + mode_lib->vba.PixelClock[k], + mode_lib->vba.DCFCLKDeepSleep, + mode_lib->vba.DPPPerPlane[k], + mode_lib->vba.NumberOfCursors[k], + mode_lib->vba.VTotal[k] + - mode_lib->vba.VActive[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.MaxInterDCNTileRepeaters, + dml_min( + mode_lib->vba.VStartupLines, + mode_lib->vba.MaxVStartupLines[k]), + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.DynamicMetadataEnable[k], + mode_lib->vba.DynamicMetadataLinesBeforeActiveRequired[k], + mode_lib->vba.DynamicMetadataTransmittedBytes[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.TCalc, + mode_lib->vba.PDEAndMetaPTEBytesFrame[k], + mode_lib->vba.MetaRowByte[k], + mode_lib->vba.PixelPTEBytesPerRow[k], + mode_lib->vba.PrefetchSourceLinesY[k], + mode_lib->vba.SwathWidthY[k], + mode_lib->vba.BytePerPixelDETY[k], + mode_lib->vba.VInitPreFillY[k], + mode_lib->vba.MaxNumSwathY[k], + mode_lib->vba.PrefetchSourceLinesC[k], + mode_lib->vba.BytePerPixelDETC[k], + mode_lib->vba.VInitPreFillC[k], + mode_lib->vba.MaxNumSwathC[k], + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + TWait, + mode_lib->vba.XFCEnabled[k], + mode_lib->vba.XFCRemoteSurfaceFlipDelay, + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.DSTXAfterScaler[k], + mode_lib->vba.DSTYAfterScaler[k], + &mode_lib->vba.DestinationLinesForPrefetch[k], + &mode_lib->vba.PrefetchBandwidth[k], + &mode_lib->vba.DestinationLinesToRequestVMInVBlank[k], + &mode_lib->vba.DestinationLinesToRequestRowInVBlank[k], + &mode_lib->vba.VRatioPrefetchY[k], + &mode_lib->vba.VRatioPrefetchC[k], + &mode_lib->vba.RequiredPrefetchPixDataBWLuma[k], + &mode_lib->vba.Tno_bw[k], + &mode_lib->vba.VUpdateOffsetPix[k], + &mode_lib->vba.VUpdateWidthPix[k], + &mode_lib->vba.VReadyOffsetPix[k]); + + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.VStartup[k] = dml_min( + mode_lib->vba.VStartupLines, + mode_lib->vba.MaxVStartupLines[k]); + if (mode_lib->vba.VStartupRequiredWhenNotEnoughTimeForDynamicMetadata + != 0) { + mode_lib->vba.VStartup[k] = + mode_lib->vba.VStartupRequiredWhenNotEnoughTimeForDynamicMetadata; + } + } else { + mode_lib->vba.VStartup[k] = + dml_min( + mode_lib->vba.VStartupLines, + mode_lib->vba.MaxVStartupLines[mode_lib->vba.BlendingAndTiming[k]]); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + + if (mode_lib->vba.PDEAndMetaPTEBytesFrame[k] == 0) + mode_lib->vba.prefetch_vm_bw[k] = 0; + else if (mode_lib->vba.DestinationLinesToRequestVMInVBlank[k] > 0) { + mode_lib->vba.prefetch_vm_bw[k] = + (double) mode_lib->vba.PDEAndMetaPTEBytesFrame[k] + / (mode_lib->vba.DestinationLinesToRequestVMInVBlank[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + } else { + mode_lib->vba.prefetch_vm_bw[k] = 0; + prefetch_vm_bw_valid = false; + } + if (mode_lib->vba.MetaRowByte[k] + mode_lib->vba.PixelPTEBytesPerRow[k] + == 0) + mode_lib->vba.prefetch_row_bw[k] = 0; + else if (mode_lib->vba.DestinationLinesToRequestRowInVBlank[k] > 0) { + mode_lib->vba.prefetch_row_bw[k] = + (double) (mode_lib->vba.MetaRowByte[k] + + mode_lib->vba.PixelPTEBytesPerRow[k]) + / (mode_lib->vba.DestinationLinesToRequestRowInVBlank[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + } else { + mode_lib->vba.prefetch_row_bw[k] = 0; + prefetch_row_bw_valid = false; + } + + MaxTotalRDBandwidth = + MaxTotalRDBandwidth + mode_lib->vba.cursor_bw[k] + + dml_max( + mode_lib->vba.prefetch_vm_bw[k], + dml_max( + mode_lib->vba.prefetch_row_bw[k], + dml_max( + mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k], + mode_lib->vba.RequiredPrefetchPixDataBWLuma[k]) + + mode_lib->vba.meta_row_bw[k] + + mode_lib->vba.dpte_row_bw[k])); + + if (mode_lib->vba.DestinationLinesForPrefetch[k] < 2) + DestinationLineTimesForPrefetchLessThan2 = true; + if (mode_lib->vba.VRatioPrefetchY[k] > 4 + || mode_lib->vba.VRatioPrefetchC[k] > 4) + VRatioPrefetchMoreThan4 = true; + } + + if (MaxTotalRDBandwidth <= mode_lib->vba.ReturnBW && prefetch_vm_bw_valid + && prefetch_row_bw_valid && !VRatioPrefetchMoreThan4 + && !DestinationLineTimesForPrefetchLessThan2) + mode_lib->vba.PrefetchModeSupported = true; + else { + mode_lib->vba.PrefetchModeSupported = false; + dml_print( + "DML: CalculatePrefetchSchedule ***failed***. Bandwidth violation. Results are NOT valid\n"); + } + + if (mode_lib->vba.PrefetchModeSupported == true) { + double final_flip_bw[DC__NUM_DPP__MAX]; + unsigned int ImmediateFlipBytes[DC__NUM_DPP__MAX]; + double total_dcn_read_bw_with_flip = 0; + + mode_lib->vba.BandwidthAvailableForImmediateFlip = mode_lib->vba.ReturnBW; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = + mode_lib->vba.BandwidthAvailableForImmediateFlip + - mode_lib->vba.cursor_bw[k] + - dml_max( + mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k] + + mode_lib->vba.qual_row_bw[k], + mode_lib->vba.PrefetchBandwidth[k]); + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + ImmediateFlipBytes[k] = 0; + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + ImmediateFlipBytes[k] = + mode_lib->vba.PDEAndMetaPTEBytesFrame[k] + + mode_lib->vba.MetaRowByte[k] + + mode_lib->vba.PixelPTEBytesPerRow[k]; + } + } + mode_lib->vba.TotImmediateFlipBytes = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + mode_lib->vba.TotImmediateFlipBytes = + mode_lib->vba.TotImmediateFlipBytes + + ImmediateFlipBytes[k]; + } + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + CalculateFlipSchedule( + mode_lib, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.BandwidthAvailableForImmediateFlip, + mode_lib->vba.TotImmediateFlipBytes, + mode_lib->vba.SourcePixelFormat[k], + ImmediateFlipBytes[k], + mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.Tno_bw[k], + mode_lib->vba.PDEAndMetaPTEBytesFrame[k], + mode_lib->vba.MetaRowByte[k], + mode_lib->vba.PixelPTEBytesPerRow[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.dpte_row_height[k], + mode_lib->vba.meta_row_height[k], + mode_lib->vba.qual_row_bw[k], + &mode_lib->vba.DestinationLinesToRequestVMInImmediateFlip[k], + &mode_lib->vba.DestinationLinesToRequestRowInImmediateFlip[k], + &final_flip_bw[k], + &mode_lib->vba.ImmediateFlipSupportedForPipe[k]); + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + total_dcn_read_bw_with_flip = + total_dcn_read_bw_with_flip + + mode_lib->vba.cursor_bw[k] + + dml_max( + mode_lib->vba.prefetch_vm_bw[k], + dml_max( + mode_lib->vba.prefetch_row_bw[k], + final_flip_bw[k] + + dml_max( + mode_lib->vba.ReadBandwidthPlaneLuma[k] + + mode_lib->vba.ReadBandwidthPlaneChroma[k], + mode_lib->vba.RequiredPrefetchPixDataBWLuma[k]))); + } + mode_lib->vba.ImmediateFlipSupported = true; + if (total_dcn_read_bw_with_flip > mode_lib->vba.ReturnBW) { + mode_lib->vba.ImmediateFlipSupported = false; + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ImmediateFlipSupportedForPipe[k] == false) { + mode_lib->vba.ImmediateFlipSupported = false; + } + } + } else { + mode_lib->vba.ImmediateFlipSupported = false; + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ErrorResult[k]) { + mode_lib->vba.PrefetchModeSupported = false; + dml_print( + "DML: CalculatePrefetchSchedule ***failed***. Prefetch schedule violation. Results are NOT valid\n"); + } + } + + mode_lib->vba.VStartupLines = mode_lib->vba.VStartupLines + 1; + } while (!((mode_lib->vba.PrefetchModeSupported + && (!mode_lib->vba.ImmediateFlipSupport + || mode_lib->vba.ImmediateFlipSupported)) + || mode_lib->vba.MaximumMaxVStartupLines < mode_lib->vba.VStartupLines)); + + //Display Pipeline Delivery Time in Prefetch + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.VRatioPrefetchY[k] <= 1) { + mode_lib->vba.DisplayPipeLineDeliveryTimeLumaPrefetch[k] = + mode_lib->vba.SwathWidthY[k] * mode_lib->vba.DPPPerPlane[k] + / mode_lib->vba.HRatio[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.DisplayPipeLineDeliveryTimeLumaPrefetch[k] = + mode_lib->vba.SwathWidthY[k] + / mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] + / mode_lib->vba.DPPCLK[k]; + } + if (mode_lib->vba.BytePerPixelDETC[k] == 0) { + mode_lib->vba.DisplayPipeLineDeliveryTimeChromaPrefetch[k] = 0; + } else { + if (mode_lib->vba.VRatioPrefetchC[k] <= 1) { + mode_lib->vba.DisplayPipeLineDeliveryTimeChromaPrefetch[k] = + mode_lib->vba.SwathWidthY[k] + * mode_lib->vba.DPPPerPlane[k] + / mode_lib->vba.HRatio[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.DisplayPipeLineDeliveryTimeChromaPrefetch[k] = + mode_lib->vba.SwathWidthY[k] + / mode_lib->vba.PSCL_THROUGHPUT_LUMA[k] + / mode_lib->vba.DPPCLK[k]; + } + } + } + + // Min TTUVBlank + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 0) { + mode_lib->vba.AllowDRAMClockChangeDuringVBlank[k] = true; + mode_lib->vba.AllowDRAMSelfRefreshDuringVBlank[k] = true; + mode_lib->vba.MinTTUVBlank[k] = dml_max( + mode_lib->vba.DRAMClockChangeWatermark, + dml_max( + mode_lib->vba.StutterEnterPlusExitWatermark, + mode_lib->vba.UrgentWatermark)); + } else if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 1) { + mode_lib->vba.AllowDRAMClockChangeDuringVBlank[k] = false; + mode_lib->vba.AllowDRAMSelfRefreshDuringVBlank[k] = true; + mode_lib->vba.MinTTUVBlank[k] = dml_max( + mode_lib->vba.StutterEnterPlusExitWatermark, + mode_lib->vba.UrgentWatermark); + } else { + mode_lib->vba.AllowDRAMClockChangeDuringVBlank[k] = false; + mode_lib->vba.AllowDRAMSelfRefreshDuringVBlank[k] = false; + mode_lib->vba.MinTTUVBlank[k] = mode_lib->vba.UrgentWatermark; + } + if (!mode_lib->vba.DynamicMetadataEnable[k]) + mode_lib->vba.MinTTUVBlank[k] = mode_lib->vba.TCalc + + mode_lib->vba.MinTTUVBlank[k]; + } + + // DCC Configuration + mode_lib->vba.ActiveDPPs = 0; + // NB P-State/DRAM Clock Change Support + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.ActiveDPPs = mode_lib->vba.ActiveDPPs + mode_lib->vba.DPPPerPlane[k]; + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + double EffectiveLBLatencyHidingY; + double EffectiveLBLatencyHidingC; + double DPPOutputBufferLinesY; + double DPPOutputBufferLinesC; + double DPPOPPBufferingY; + double MaxDETBufferingTimeY; + double ActiveDRAMClockChangeLatencyMarginY; + + mode_lib->vba.LBLatencyHidingSourceLinesY = + dml_min( + mode_lib->vba.MaxLineBufferLines, + (unsigned int) dml_floor( + (double) mode_lib->vba.LineBufferSize + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.SwathWidthY[k] + / dml_max( + mode_lib->vba.HRatio[k], + 1.0)), + 1)) - (mode_lib->vba.vtaps[k] - 1); + + mode_lib->vba.LBLatencyHidingSourceLinesC = + dml_min( + mode_lib->vba.MaxLineBufferLines, + (unsigned int) dml_floor( + (double) mode_lib->vba.LineBufferSize + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.SwathWidthY[k] + / 2.0 + / dml_max( + mode_lib->vba.HRatio[k] + / 2, + 1.0)), + 1)) + - (mode_lib->vba.VTAPsChroma[k] - 1); + + EffectiveLBLatencyHidingY = mode_lib->vba.LBLatencyHidingSourceLinesY + / mode_lib->vba.VRatio[k] + * (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]); + + EffectiveLBLatencyHidingC = mode_lib->vba.LBLatencyHidingSourceLinesC + / (mode_lib->vba.VRatio[k] / 2) + * (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]); + + if (mode_lib->vba.SwathWidthY[k] > 2 * mode_lib->vba.DPPOutputBufferPixels) { + DPPOutputBufferLinesY = mode_lib->vba.DPPOutputBufferPixels + / mode_lib->vba.SwathWidthY[k]; + } else if (mode_lib->vba.SwathWidthY[k] > mode_lib->vba.DPPOutputBufferPixels) { + DPPOutputBufferLinesY = 0.5; + } else { + DPPOutputBufferLinesY = 1; + } + + if (mode_lib->vba.SwathWidthY[k] / 2 > 2 * mode_lib->vba.DPPOutputBufferPixels) { + DPPOutputBufferLinesC = mode_lib->vba.DPPOutputBufferPixels + / (mode_lib->vba.SwathWidthY[k] / 2); + } else if (mode_lib->vba.SwathWidthY[k] / 2 > mode_lib->vba.DPPOutputBufferPixels) { + DPPOutputBufferLinesC = 0.5; + } else { + DPPOutputBufferLinesC = 1; + } + + DPPOPPBufferingY = (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * (DPPOutputBufferLinesY + mode_lib->vba.OPPOutputBufferLines); + MaxDETBufferingTimeY = mode_lib->vba.FullDETBufferingTimeY[k] + + (mode_lib->vba.LinesInDETY[k] + - mode_lib->vba.LinesInDETYRoundedDownToSwath[k]) + / mode_lib->vba.SwathHeightY[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + + ActiveDRAMClockChangeLatencyMarginY = DPPOPPBufferingY + EffectiveLBLatencyHidingY + + MaxDETBufferingTimeY - mode_lib->vba.DRAMClockChangeWatermark; + + if (mode_lib->vba.ActiveDPPs > 1) { + ActiveDRAMClockChangeLatencyMarginY = + ActiveDRAMClockChangeLatencyMarginY + - (1 - 1 / (mode_lib->vba.ActiveDPPs - 1)) + * mode_lib->vba.SwathHeightY[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + } + + if (mode_lib->vba.BytePerPixelDETC[k] > 0) { + double DPPOPPBufferingC = (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + * (DPPOutputBufferLinesC + + mode_lib->vba.OPPOutputBufferLines); + double MaxDETBufferingTimeC = + mode_lib->vba.FullDETBufferingTimeC[k] + + (mode_lib->vba.LinesInDETC[k] + - mode_lib->vba.LinesInDETCRoundedDownToSwath[k]) + / mode_lib->vba.SwathHeightC[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + double ActiveDRAMClockChangeLatencyMarginC = DPPOPPBufferingC + + EffectiveLBLatencyHidingC + MaxDETBufferingTimeC + - mode_lib->vba.DRAMClockChangeWatermark; + + if (mode_lib->vba.ActiveDPPs > 1) { + ActiveDRAMClockChangeLatencyMarginC = + ActiveDRAMClockChangeLatencyMarginC + - (1 + - 1 + / (mode_lib->vba.ActiveDPPs + - 1)) + * mode_lib->vba.SwathHeightC[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]); + } + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = dml_min( + ActiveDRAMClockChangeLatencyMarginY, + ActiveDRAMClockChangeLatencyMarginC); + } else { + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = + ActiveDRAMClockChangeLatencyMarginY; + } + + if (mode_lib->vba.WritebackEnable[k]) { + double WritebackDRAMClockChangeLatencyMargin; + + if (mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + WritebackDRAMClockChangeLatencyMargin = + (double) (mode_lib->vba.WritebackInterfaceLumaBufferSize + + mode_lib->vba.WritebackInterfaceChromaBufferSize) + / (mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + * 4) + - mode_lib->vba.WritebackDRAMClockChangeWatermark; + } else if (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + WritebackDRAMClockChangeLatencyMargin = + dml_min( + (double) mode_lib->vba.WritebackInterfaceLumaBufferSize + * 8.0 / 10, + 2.0 + * mode_lib->vba.WritebackInterfaceChromaBufferSize + * 8 / 10) + / (mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k])) + - mode_lib->vba.WritebackDRAMClockChangeWatermark; + } else { + WritebackDRAMClockChangeLatencyMargin = + dml_min( + (double) mode_lib->vba.WritebackInterfaceLumaBufferSize, + 2.0 + * mode_lib->vba.WritebackInterfaceChromaBufferSize) + / (mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k])) + - mode_lib->vba.WritebackDRAMClockChangeWatermark; + } + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = dml_min( + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k], + WritebackDRAMClockChangeLatencyMargin); + } + } + + mode_lib->vba.MinActiveDRAMClockChangeMargin = 999999; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] + < mode_lib->vba.MinActiveDRAMClockChangeMargin) { + mode_lib->vba.MinActiveDRAMClockChangeMargin = + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k]; + } + } + + mode_lib->vba.MinActiveDRAMClockChangeLatencySupported = + mode_lib->vba.MinActiveDRAMClockChangeMargin + + mode_lib->vba.DRAMClockChangeLatency; + + if (mode_lib->vba.DRAMClockChangeSupportsVActive && + mode_lib->vba.MinActiveDRAMClockChangeMargin > 60) { + mode_lib->vba.DRAMClockChangeWatermark += 25; + mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive; + } else if (mode_lib->vba.DummyPStateCheck && + mode_lib->vba.MinActiveDRAMClockChangeMargin > 0) { + mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive; + } else { + if (mode_lib->vba.SynchronizedVBlank || mode_lib->vba.NumberOfActivePlanes == 1) { + mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vblank; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (!mode_lib->vba.AllowDRAMClockChangeDuringVBlank[k]) { + mode_lib->vba.DRAMClockChangeSupport[0][0] = + dm_dram_clock_change_unsupported; + } + } + } else { + mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_unsupported; + } + } + for (k = 0; k <= mode_lib->vba.soc.num_states; k++) + for (j = 0; j < 2; j++) + mode_lib->vba.DRAMClockChangeSupport[k][j] = mode_lib->vba.DRAMClockChangeSupport[0][0]; + + //XFC Parameters: + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.XFCEnabled[k] == true) { + double TWait; + + mode_lib->vba.XFCSlaveVUpdateOffset[k] = mode_lib->vba.XFCTSlvVupdateOffset; + mode_lib->vba.XFCSlaveVupdateWidth[k] = mode_lib->vba.XFCTSlvVupdateWidth; + mode_lib->vba.XFCSlaveVReadyOffset[k] = mode_lib->vba.XFCTSlvVreadyOffset; + TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.SREnterPlusExitTime); + mode_lib->vba.XFCRemoteSurfaceFlipDelay = CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.SwathWidthY[k], + dml_ceil(mode_lib->vba.BytePerPixelDETY[k], 1), + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TCalc, + TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + mode_lib->vba.XFCRemoteSurfaceFlipLatency[k] = + dml_floor( + mode_lib->vba.XFCRemoteSurfaceFlipDelay + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + mode_lib->vba.XFCTransferDelay[k] = + dml_ceil( + mode_lib->vba.XFCBusTransportTime + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + mode_lib->vba.XFCPrechargeDelay[k] = + dml_ceil( + (mode_lib->vba.XFCBusTransportTime + + mode_lib->vba.TInitXFill + + mode_lib->vba.TslvChk) + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + mode_lib->vba.InitFillLevel = mode_lib->vba.XFCXBUFLatencyTolerance + * mode_lib->vba.SrcActiveDrainRate; + mode_lib->vba.FinalFillMargin = + (mode_lib->vba.DestinationLinesToRequestVMInVBlank[k] + + mode_lib->vba.DestinationLinesToRequestRowInVBlank[k]) + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k] + * mode_lib->vba.SrcActiveDrainRate + + mode_lib->vba.XFCFillConstant; + mode_lib->vba.FinalFillLevel = mode_lib->vba.XFCRemoteSurfaceFlipDelay + * mode_lib->vba.SrcActiveDrainRate + + mode_lib->vba.FinalFillMargin; + mode_lib->vba.RemainingFillLevel = dml_max( + 0.0, + mode_lib->vba.FinalFillLevel - mode_lib->vba.InitFillLevel); + mode_lib->vba.TFinalxFill = mode_lib->vba.RemainingFillLevel + / (mode_lib->vba.SrcActiveDrainRate + * mode_lib->vba.XFCFillBWOverhead / 100); + mode_lib->vba.XFCPrefetchMargin[k] = + mode_lib->vba.XFCRemoteSurfaceFlipDelay + + mode_lib->vba.TFinalxFill + + (mode_lib->vba.DestinationLinesToRequestVMInVBlank[k] + + mode_lib->vba.DestinationLinesToRequestRowInVBlank[k]) + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.XFCSlaveVUpdateOffset[k] = 0; + mode_lib->vba.XFCSlaveVupdateWidth[k] = 0; + mode_lib->vba.XFCSlaveVReadyOffset[k] = 0; + mode_lib->vba.XFCRemoteSurfaceFlipLatency[k] = 0; + mode_lib->vba.XFCPrechargeDelay[k] = 0; + mode_lib->vba.XFCTransferDelay[k] = 0; + mode_lib->vba.XFCPrefetchMargin[k] = 0; + } + } + { + unsigned int VStartupMargin = 0; + bool FirstMainPlane = true; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + unsigned int Margin = (mode_lib->vba.MaxVStartupLines[k] - mode_lib->vba.VStartup[k]) + * mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]; + + if (FirstMainPlane) { + VStartupMargin = Margin; + FirstMainPlane = false; + } else + VStartupMargin = dml_min(VStartupMargin, Margin); + } + + if (mode_lib->vba.UseMaximumVStartup) { + if (mode_lib->vba.VTotal_Max[k] == mode_lib->vba.VTotal[k]) { + //only use max vstart if it is not drr or lateflip. + mode_lib->vba.VStartup[k] = mode_lib->vba.MaxVStartupLines[mode_lib->vba.BlendingAndTiming[k]]; + } + } + } +} +} + +static void dml20v2_DisplayPipeConfiguration(struct display_mode_lib *mode_lib) +{ + double BytePerPixDETY; + double BytePerPixDETC; + double Read256BytesBlockHeightY; + double Read256BytesBlockHeightC; + double Read256BytesBlockWidthY; + double Read256BytesBlockWidthC; + double MaximumSwathHeightY; + double MaximumSwathHeightC; + double MinimumSwathHeightY; + double MinimumSwathHeightC; + double SwathWidth; + double SwathWidthGranularityY; + double SwathWidthGranularityC; + double RoundedUpMaxSwathSizeBytesY; + double RoundedUpMaxSwathSizeBytesC; + unsigned int j, k; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + bool MainPlaneDoesODMCombine = false; + + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + BytePerPixDETY = 8; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + BytePerPixDETY = 4; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + BytePerPixDETY = 2; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8) { + BytePerPixDETY = 1; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + BytePerPixDETY = 1; + BytePerPixDETC = 2; + } else { + BytePerPixDETY = 4.0 / 3.0; + BytePerPixDETC = 8.0 / 3.0; + } + + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + Read256BytesBlockHeightY = 1; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + Read256BytesBlockHeightY = 4; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + Read256BytesBlockHeightY = 8; + } else { + Read256BytesBlockHeightY = 16; + } + Read256BytesBlockWidthY = 256 / dml_ceil(BytePerPixDETY, 1) + / Read256BytesBlockHeightY; + Read256BytesBlockHeightC = 0; + Read256BytesBlockWidthC = 0; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + Read256BytesBlockHeightY = 1; + Read256BytesBlockHeightC = 1; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + Read256BytesBlockHeightY = 16; + Read256BytesBlockHeightC = 8; + } else { + Read256BytesBlockHeightY = 8; + Read256BytesBlockHeightC = 8; + } + Read256BytesBlockWidthY = 256 / dml_ceil(BytePerPixDETY, 1) + / Read256BytesBlockHeightY; + Read256BytesBlockWidthC = 256 / dml_ceil(BytePerPixDETC, 2) + / Read256BytesBlockHeightC; + } + + if (mode_lib->vba.SourceScan[k] == dm_horz) { + MaximumSwathHeightY = Read256BytesBlockHeightY; + MaximumSwathHeightC = Read256BytesBlockHeightC; + } else { + MaximumSwathHeightY = Read256BytesBlockWidthY; + MaximumSwathHeightC = Read256BytesBlockWidthC; + } + + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + || (mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_t + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s_x) + && mode_lib->vba.SourceScan[k] == dm_horz)) { + MinimumSwathHeightY = MaximumSwathHeightY; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8 + && mode_lib->vba.SourceScan[k] != dm_horz) { + MinimumSwathHeightY = MaximumSwathHeightY; + } else { + MinimumSwathHeightY = MaximumSwathHeightY / 2.0; + } + MinimumSwathHeightC = MaximumSwathHeightC; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + MinimumSwathHeightY = MaximumSwathHeightY; + MinimumSwathHeightC = MaximumSwathHeightC; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8 + && mode_lib->vba.SourceScan[k] == dm_horz) { + MinimumSwathHeightY = MaximumSwathHeightY / 2.0; + MinimumSwathHeightC = MaximumSwathHeightC; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10 + && mode_lib->vba.SourceScan[k] == dm_horz) { + MinimumSwathHeightC = MaximumSwathHeightC / 2.0; + MinimumSwathHeightY = MaximumSwathHeightY; + } else { + MinimumSwathHeightY = MaximumSwathHeightY; + MinimumSwathHeightC = MaximumSwathHeightC; + } + } + + if (mode_lib->vba.SourceScan[k] == dm_horz) { + SwathWidth = mode_lib->vba.ViewportWidth[k]; + } else { + SwathWidth = mode_lib->vba.ViewportHeight[k]; + } + + if (mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) { + MainPlaneDoesODMCombine = true; + } + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { + if (mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) { + MainPlaneDoesODMCombine = true; + } + } + + if (MainPlaneDoesODMCombine == true) { + SwathWidth = dml_min( + SwathWidth, + mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k]); + } else { + if (mode_lib->vba.DPPPerPlane[k] == 0) + SwathWidth = 0; + else + SwathWidth = SwathWidth / mode_lib->vba.DPPPerPlane[k]; + } + + SwathWidthGranularityY = 256 / dml_ceil(BytePerPixDETY, 1) / MaximumSwathHeightY; + RoundedUpMaxSwathSizeBytesY = (dml_ceil( + (double) (SwathWidth - 1), + SwathWidthGranularityY) + SwathWidthGranularityY) * BytePerPixDETY + * MaximumSwathHeightY; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + RoundedUpMaxSwathSizeBytesY = dml_ceil(RoundedUpMaxSwathSizeBytesY, 256) + + 256; + } + if (MaximumSwathHeightC > 0) { + SwathWidthGranularityC = 256.0 / dml_ceil(BytePerPixDETC, 2) + / MaximumSwathHeightC; + RoundedUpMaxSwathSizeBytesC = (dml_ceil( + (double) (SwathWidth / 2.0 - 1), + SwathWidthGranularityC) + SwathWidthGranularityC) + * BytePerPixDETC * MaximumSwathHeightC; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + RoundedUpMaxSwathSizeBytesC = dml_ceil( + RoundedUpMaxSwathSizeBytesC, + 256) + 256; + } + } else + RoundedUpMaxSwathSizeBytesC = 0.0; + + if (RoundedUpMaxSwathSizeBytesY + RoundedUpMaxSwathSizeBytesC + <= mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0) { + mode_lib->vba.SwathHeightY[k] = MaximumSwathHeightY; + mode_lib->vba.SwathHeightC[k] = MaximumSwathHeightC; + } else { + mode_lib->vba.SwathHeightY[k] = MinimumSwathHeightY; + mode_lib->vba.SwathHeightC[k] = MinimumSwathHeightC; + } + + if (mode_lib->vba.SwathHeightC[k] == 0) { + mode_lib->vba.DETBufferSizeY[k] = mode_lib->vba.DETBufferSizeInKByte * 1024; + mode_lib->vba.DETBufferSizeC[k] = 0; + } else if (mode_lib->vba.SwathHeightY[k] <= mode_lib->vba.SwathHeightC[k]) { + mode_lib->vba.DETBufferSizeY[k] = mode_lib->vba.DETBufferSizeInKByte + * 1024.0 / 2; + mode_lib->vba.DETBufferSizeC[k] = mode_lib->vba.DETBufferSizeInKByte + * 1024.0 / 2; + } else { + mode_lib->vba.DETBufferSizeY[k] = mode_lib->vba.DETBufferSizeInKByte + * 1024.0 * 2 / 3; + mode_lib->vba.DETBufferSizeC[k] = mode_lib->vba.DETBufferSizeInKByte + * 1024.0 / 3; + } + } +} + +static double CalculateTWait( + unsigned int PrefetchMode, + double DRAMClockChangeLatency, + double UrgentLatencyPixelDataOnly, + double SREnterPlusExitTime) +{ + if (PrefetchMode == 0) { + return dml_max( + DRAMClockChangeLatency + UrgentLatencyPixelDataOnly, + dml_max(SREnterPlusExitTime, UrgentLatencyPixelDataOnly)); + } else if (PrefetchMode == 1) { + return dml_max(SREnterPlusExitTime, UrgentLatencyPixelDataOnly); + } else { + return UrgentLatencyPixelDataOnly; + } +} + +static double CalculateRemoteSurfaceFlipDelay( + struct display_mode_lib *mode_lib, + double VRatio, + double SwathWidth, + double Bpp, + double LineTime, + double XFCTSlvVupdateOffset, + double XFCTSlvVupdateWidth, + double XFCTSlvVreadyOffset, + double XFCXBUFLatencyTolerance, + double XFCFillBWOverhead, + double XFCSlvChunkSize, + double XFCBusTransportTime, + double TCalc, + double TWait, + double *SrcActiveDrainRate, + double *TInitXFill, + double *TslvChk) +{ + double TSlvSetup, AvgfillRate, result; + + *SrcActiveDrainRate = VRatio * SwathWidth * Bpp / LineTime; + TSlvSetup = XFCTSlvVupdateOffset + XFCTSlvVupdateWidth + XFCTSlvVreadyOffset; + *TInitXFill = XFCXBUFLatencyTolerance / (1 + XFCFillBWOverhead / 100); + AvgfillRate = *SrcActiveDrainRate * (1 + XFCFillBWOverhead / 100); + *TslvChk = XFCSlvChunkSize / AvgfillRate; + dml_print( + "DML::CalculateRemoteSurfaceFlipDelay: SrcActiveDrainRate: %f\n", + *SrcActiveDrainRate); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TSlvSetup: %f\n", TSlvSetup); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TInitXFill: %f\n", *TInitXFill); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: AvgfillRate: %f\n", AvgfillRate); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TslvChk: %f\n", *TslvChk); + result = 2 * XFCBusTransportTime + TSlvSetup + TCalc + TWait + *TslvChk + *TInitXFill; // TODO: This doesn't seem to match programming guide + dml_print("DML::CalculateRemoteSurfaceFlipDelay: RemoteSurfaceFlipDelay: %f\n", result); + return result; +} + +static double CalculateWriteBackDelay( + enum source_format_class WritebackPixelFormat, + double WritebackHRatio, + double WritebackVRatio, + unsigned int WritebackLumaHTaps, + unsigned int WritebackLumaVTaps, + unsigned int WritebackChromaHTaps, + unsigned int WritebackChromaVTaps, + unsigned int WritebackDestinationWidth) +{ + double CalculateWriteBackDelay = + dml_max( + dml_ceil(WritebackLumaHTaps / 4.0, 1) / WritebackHRatio, + WritebackLumaVTaps * dml_ceil(1.0 / WritebackVRatio, 1) + * dml_ceil( + WritebackDestinationWidth + / 4.0, + 1) + + dml_ceil(1.0 / WritebackVRatio, 1) + * (dml_ceil( + WritebackLumaVTaps + / 4.0, + 1) + 4)); + + if (WritebackPixelFormat != dm_444_32) { + CalculateWriteBackDelay = + dml_max( + CalculateWriteBackDelay, + dml_max( + dml_ceil( + WritebackChromaHTaps + / 2.0, + 1) + / (2 + * WritebackHRatio), + WritebackChromaVTaps + * dml_ceil( + 1 + / (2 + * WritebackVRatio), + 1) + * dml_ceil( + WritebackDestinationWidth + / 2.0 + / 2.0, + 1) + + dml_ceil( + 1 + / (2 + * WritebackVRatio), + 1) + * (dml_ceil( + WritebackChromaVTaps + / 4.0, + 1) + + 4))); + } + return CalculateWriteBackDelay; +} + +static void CalculateActiveRowBandwidth( + bool GPUVMEnable, + enum source_format_class SourcePixelFormat, + double VRatio, + bool DCCEnable, + double LineTime, + unsigned int MetaRowByteLuma, + unsigned int MetaRowByteChroma, + unsigned int meta_row_height_luma, + unsigned int meta_row_height_chroma, + unsigned int PixelPTEBytesPerRowLuma, + unsigned int PixelPTEBytesPerRowChroma, + unsigned int dpte_row_height_luma, + unsigned int dpte_row_height_chroma, + double *meta_row_bw, + double *dpte_row_bw, + double *qual_row_bw) +{ + if (DCCEnable != true) { + *meta_row_bw = 0; + } else if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *meta_row_bw = VRatio * MetaRowByteLuma / (meta_row_height_luma * LineTime) + + VRatio / 2 * MetaRowByteChroma + / (meta_row_height_chroma * LineTime); + } else { + *meta_row_bw = VRatio * MetaRowByteLuma / (meta_row_height_luma * LineTime); + } + + if (GPUVMEnable != true) { + *dpte_row_bw = 0; + } else if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *dpte_row_bw = VRatio * PixelPTEBytesPerRowLuma / (dpte_row_height_luma * LineTime) + + VRatio / 2 * PixelPTEBytesPerRowChroma + / (dpte_row_height_chroma * LineTime); + } else { + *dpte_row_bw = VRatio * PixelPTEBytesPerRowLuma / (dpte_row_height_luma * LineTime); + } + + if ((SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10)) { + *qual_row_bw = *meta_row_bw + *dpte_row_bw; + } else { + *qual_row_bw = 0; + } +} + +static void CalculateFlipSchedule( + struct display_mode_lib *mode_lib, + double UrgentExtraLatency, + double UrgentLatencyPixelDataOnly, + unsigned int GPUVMMaxPageTableLevels, + bool GPUVMEnable, + double BandwidthAvailableForImmediateFlip, + unsigned int TotImmediateFlipBytes, + enum source_format_class SourcePixelFormat, + unsigned int ImmediateFlipBytes, + double LineTime, + double VRatio, + double Tno_bw, + double PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + bool DCCEnable, + unsigned int dpte_row_height, + unsigned int meta_row_height, + double qual_row_bw, + double *DestinationLinesToRequestVMInImmediateFlip, + double *DestinationLinesToRequestRowInImmediateFlip, + double *final_flip_bw, + bool *ImmediateFlipSupportedForPipe) +{ + double min_row_time = 0.0; + + if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *DestinationLinesToRequestVMInImmediateFlip = 0.0; + *DestinationLinesToRequestRowInImmediateFlip = 0.0; + *final_flip_bw = qual_row_bw; + *ImmediateFlipSupportedForPipe = true; + } else { + double TimeForFetchingMetaPTEImmediateFlip; + double TimeForFetchingRowInVBlankImmediateFlip; + + if (GPUVMEnable == true) { + mode_lib->vba.ImmediateFlipBW[0] = BandwidthAvailableForImmediateFlip + * ImmediateFlipBytes / TotImmediateFlipBytes; + TimeForFetchingMetaPTEImmediateFlip = + dml_max( + Tno_bw + + PDEAndMetaPTEBytesFrame + / mode_lib->vba.ImmediateFlipBW[0], + dml_max( + UrgentExtraLatency + + UrgentLatencyPixelDataOnly + * (GPUVMMaxPageTableLevels + - 1), + LineTime / 4.0)); + } else { + TimeForFetchingMetaPTEImmediateFlip = 0; + } + + *DestinationLinesToRequestVMInImmediateFlip = dml_floor( + 4.0 * (TimeForFetchingMetaPTEImmediateFlip / LineTime + 0.125), + 1) / 4.0; + + if ((GPUVMEnable == true || DCCEnable == true)) { + mode_lib->vba.ImmediateFlipBW[0] = BandwidthAvailableForImmediateFlip + * ImmediateFlipBytes / TotImmediateFlipBytes; + TimeForFetchingRowInVBlankImmediateFlip = dml_max( + (MetaRowByte + PixelPTEBytesPerRow) + / mode_lib->vba.ImmediateFlipBW[0], + dml_max(UrgentLatencyPixelDataOnly, LineTime / 4.0)); + } else { + TimeForFetchingRowInVBlankImmediateFlip = 0; + } + + *DestinationLinesToRequestRowInImmediateFlip = dml_floor( + 4.0 * (TimeForFetchingRowInVBlankImmediateFlip / LineTime + 0.125), + 1) / 4.0; + + if (GPUVMEnable == true) { + *final_flip_bw = + dml_max( + PDEAndMetaPTEBytesFrame + / (*DestinationLinesToRequestVMInImmediateFlip + * LineTime), + (MetaRowByte + PixelPTEBytesPerRow) + / (TimeForFetchingRowInVBlankImmediateFlip + * LineTime)); + } else if (MetaRowByte + PixelPTEBytesPerRow > 0) { + *final_flip_bw = (MetaRowByte + PixelPTEBytesPerRow) + / (TimeForFetchingRowInVBlankImmediateFlip * LineTime); + } else { + *final_flip_bw = 0; + } + + if (GPUVMEnable && !DCCEnable) + min_row_time = dpte_row_height * LineTime / VRatio; + else if (!GPUVMEnable && DCCEnable) + min_row_time = meta_row_height * LineTime / VRatio; + else + min_row_time = dml_min(dpte_row_height, meta_row_height) * LineTime + / VRatio; + + if (*DestinationLinesToRequestVMInImmediateFlip >= 8 + || *DestinationLinesToRequestRowInImmediateFlip >= 16 + || TimeForFetchingMetaPTEImmediateFlip + + 2 * TimeForFetchingRowInVBlankImmediateFlip + > min_row_time) + *ImmediateFlipSupportedForPipe = false; + else + *ImmediateFlipSupportedForPipe = true; + } +} + +static unsigned int TruncToValidBPP( + double DecimalBPP, + bool DSCEnabled, + enum output_encoder_class Output, + enum output_format_class Format, + unsigned int DSCInputBitPerComponent) +{ + if (Output == dm_hdmi) { + if (Format == dm_420) { + if (DecimalBPP >= 18) + return 18; + else if (DecimalBPP >= 15) + return 15; + else if (DecimalBPP >= 12) + return 12; + else + return BPP_INVALID; + } else if (Format == dm_444) { + if (DecimalBPP >= 36) + return 36; + else if (DecimalBPP >= 30) + return 30; + else if (DecimalBPP >= 24) + return 24; + else if (DecimalBPP >= 18) + return 18; + else + return BPP_INVALID; + } else { + if (DecimalBPP / 1.5 >= 24) + return 24; + else if (DecimalBPP / 1.5 >= 20) + return 20; + else if (DecimalBPP / 1.5 >= 16) + return 16; + else + return BPP_INVALID; + } + } else { + if (DSCEnabled) { + if (Format == dm_420) { + if (DecimalBPP < 6) + return BPP_INVALID; + else if (DecimalBPP >= 1.5 * DSCInputBitPerComponent - 1 / 16) + return 1.5 * DSCInputBitPerComponent - 1 / 16; + else + return dml_floor(16 * DecimalBPP, 1) / 16; + } else if (Format == dm_n422) { + if (DecimalBPP < 7) + return BPP_INVALID; + else if (DecimalBPP >= 2 * DSCInputBitPerComponent - 1 / 16) + return 2 * DSCInputBitPerComponent - 1 / 16; + else + return dml_floor(16 * DecimalBPP, 1) / 16; + } else { + if (DecimalBPP < 8) + return BPP_INVALID; + else if (DecimalBPP >= 3 * DSCInputBitPerComponent - 1 / 16) + return 3 * DSCInputBitPerComponent - 1 / 16; + else + return dml_floor(16 * DecimalBPP, 1) / 16; + } + } else if (Format == dm_420) { + if (DecimalBPP >= 18) + return 18; + else if (DecimalBPP >= 15) + return 15; + else if (DecimalBPP >= 12) + return 12; + else + return BPP_INVALID; + } else if (Format == dm_s422 || Format == dm_n422) { + if (DecimalBPP >= 24) + return 24; + else if (DecimalBPP >= 20) + return 20; + else if (DecimalBPP >= 16) + return 16; + else + return BPP_INVALID; + } else { + if (DecimalBPP >= 36) + return 36; + else if (DecimalBPP >= 30) + return 30; + else if (DecimalBPP >= 24) + return 24; + else if (DecimalBPP >= 18) + return 18; + else + return BPP_INVALID; + } + } +} + +void dml20v2_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) +{ + struct vba_vars_st *locals = &mode_lib->vba; + + int i; + unsigned int j, k, m; + + /*MODE SUPPORT, VOLTAGE STATE AND SOC CONFIGURATION*/ + + /*Scale Ratio, taps Support Check*/ + + mode_lib->vba.ScaleRatioAndTapsSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.ScalerEnabled[k] == false + && ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8) + || mode_lib->vba.HRatio[k] != 1.0 + || mode_lib->vba.htaps[k] != 1.0 + || mode_lib->vba.VRatio[k] != 1.0 + || mode_lib->vba.vtaps[k] != 1.0)) { + mode_lib->vba.ScaleRatioAndTapsSupport = false; + } else if (mode_lib->vba.vtaps[k] < 1.0 || mode_lib->vba.vtaps[k] > 8.0 + || mode_lib->vba.htaps[k] < 1.0 || mode_lib->vba.htaps[k] > 8.0 + || (mode_lib->vba.htaps[k] > 1.0 + && (mode_lib->vba.htaps[k] % 2) == 1) + || mode_lib->vba.HRatio[k] > mode_lib->vba.MaxHSCLRatio + || mode_lib->vba.VRatio[k] > mode_lib->vba.MaxVSCLRatio + || mode_lib->vba.HRatio[k] > mode_lib->vba.htaps[k] + || mode_lib->vba.VRatio[k] > mode_lib->vba.vtaps[k] + || (mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8 + && (mode_lib->vba.HRatio[k] / 2.0 + > mode_lib->vba.HTAPsChroma[k] + || mode_lib->vba.VRatio[k] / 2.0 + > mode_lib->vba.VTAPsChroma[k]))) { + mode_lib->vba.ScaleRatioAndTapsSupport = false; + } + } + /*Source Format, Pixel Format and Scan Support Check*/ + + mode_lib->vba.SourceFormatPixelAndScanSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if ((mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + && mode_lib->vba.SourceScan[k] != dm_horz) + || ((mode_lib->vba.SurfaceTiling[k] == dm_sw_4kb_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_4kb_d_x + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d_t + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d_x + || mode_lib->vba.SurfaceTiling[k] == dm_sw_var_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_var_d_x) + && mode_lib->vba.SourcePixelFormat[k] != dm_444_64) + || (mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_r_x + && (mode_lib->vba.SourcePixelFormat[k] == dm_mono_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_10)) + || (((mode_lib->vba.SurfaceTiling[k] == dm_sw_gfx7_2d_thin_gl + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_gfx7_2d_thin_l_vp) + && !((mode_lib->vba.SourcePixelFormat[k] + == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] + == dm_444_32) + && mode_lib->vba.SourceScan[k] + == dm_horz + && mode_lib->vba.SupportGFX7CompatibleTilingIn32bppAnd64bpp + == true + && mode_lib->vba.DCCEnable[k] + == false)) + || (mode_lib->vba.DCCEnable[k] == true + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_linear + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_10)))) { + mode_lib->vba.SourceFormatPixelAndScanSupport = false; + } + } + /*Bandwidth Support Check*/ + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + locals->BytePerPixelInDETY[k] = 8.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + locals->BytePerPixelInDETY[k] = 4.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16) { + locals->BytePerPixelInDETY[k] = 2.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_mono_8) { + locals->BytePerPixelInDETY[k] = 1.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + locals->BytePerPixelInDETY[k] = 1.0; + locals->BytePerPixelInDETC[k] = 2.0; + } else { + locals->BytePerPixelInDETY[k] = 4.0 / 3; + locals->BytePerPixelInDETC[k] = 8.0 / 3; + } + if (mode_lib->vba.SourceScan[k] == dm_horz) { + locals->SwathWidthYSingleDPP[k] = mode_lib->vba.ViewportWidth[k]; + } else { + locals->SwathWidthYSingleDPP[k] = mode_lib->vba.ViewportHeight[k]; + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->ReadBandwidthLuma[k] = locals->SwathWidthYSingleDPP[k] * dml_ceil(locals->BytePerPixelInDETY[k], 1.0) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k]; + locals->ReadBandwidthChroma[k] = locals->SwathWidthYSingleDPP[k] / 2 * dml_ceil(locals->BytePerPixelInDETC[k], 2.0) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k] / 2.0; + locals->ReadBandwidth[k] = locals->ReadBandwidthLuma[k] + locals->ReadBandwidthChroma[k]; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 4.0; + } else if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 3.0; + } else if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 1.5; + } else { + locals->WriteBandwidth[k] = 0.0; + } + } + mode_lib->vba.DCCEnabledInAnyPlane = false; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.DCCEnable[k] == true) { + mode_lib->vba.DCCEnabledInAnyPlane = true; + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->FabricAndDRAMBandwidthPerState[i] = dml_min( + mode_lib->vba.DRAMSpeedPerState[i] * mode_lib->vba.NumberOfChannels + * mode_lib->vba.DRAMChannelWidth, + mode_lib->vba.FabricClockPerState[i] + * mode_lib->vba.FabricDatapathToDCNDataReturn) / 1000; + locals->ReturnBWToDCNPerState = dml_min(locals->ReturnBusWidth * locals->DCFCLKPerState[i], + locals->FabricAndDRAMBandwidthPerState[i] * 1000) + * locals->PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelDataOnly / 100; + + locals->ReturnBWPerState[i][0] = locals->ReturnBWToDCNPerState; + + if (locals->DCCEnabledInAnyPlane == true && locals->ReturnBWToDCNPerState > locals->DCFCLKPerState[i] * locals->ReturnBusWidth / 4) { + locals->ReturnBWPerState[i][0] = dml_min(locals->ReturnBWPerState[i][0], + locals->ReturnBWToDCNPerState * 4 * (1 - locals->UrgentLatency / + ((locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 + / (locals->ReturnBWToDCNPerState - locals->DCFCLKPerState[i] + * locals->ReturnBusWidth / 4) + locals->UrgentLatency))); + } + locals->CriticalPoint = 2 * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * + locals->UrgentLatency / (locals->ReturnBWToDCNPerState * locals->UrgentLatency + + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024); + + if (locals->DCCEnabledInAnyPlane && locals->CriticalPoint > 1 && locals->CriticalPoint < 4) { + locals->ReturnBWPerState[i][0] = dml_min(locals->ReturnBWPerState[i][0], + 4 * locals->ReturnBWToDCNPerState * + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 + * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * locals->UrgentLatency / + dml_pow((locals->ReturnBWToDCNPerState * locals->UrgentLatency + + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024), 2)); + } + + locals->ReturnBWToDCNPerState = dml_min(locals->ReturnBusWidth * + locals->DCFCLKPerState[i], locals->FabricAndDRAMBandwidthPerState[i] * 1000); + + if (locals->DCCEnabledInAnyPlane == true && locals->ReturnBWToDCNPerState > locals->DCFCLKPerState[i] * locals->ReturnBusWidth / 4) { + locals->ReturnBWPerState[i][0] = dml_min(locals->ReturnBWPerState[i][0], + locals->ReturnBWToDCNPerState * 4 * (1 - locals->UrgentLatency / + ((locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 + / (locals->ReturnBWToDCNPerState - locals->DCFCLKPerState[i] + * locals->ReturnBusWidth / 4) + locals->UrgentLatency))); + } + locals->CriticalPoint = 2 * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * + locals->UrgentLatency / (locals->ReturnBWToDCNPerState * locals->UrgentLatency + + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024); + + if (locals->DCCEnabledInAnyPlane && locals->CriticalPoint > 1 && locals->CriticalPoint < 4) { + locals->ReturnBWPerState[i][0] = dml_min(locals->ReturnBWPerState[i][0], + 4 * locals->ReturnBWToDCNPerState * + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024 + * locals->ReturnBusWidth * locals->DCFCLKPerState[i] * locals->UrgentLatency / + dml_pow((locals->ReturnBWToDCNPerState * locals->UrgentLatency + + (locals->ROBBufferSizeInKByte - locals->PixelChunkSizeInKByte) * 1024), 2)); + } + } + /*Writeback Latency support check*/ + + mode_lib->vba.WritebackLatencySupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + if (locals->WriteBandwidth[k] + > (mode_lib->vba.WritebackInterfaceLumaBufferSize + + mode_lib->vba.WritebackInterfaceChromaBufferSize) + / mode_lib->vba.WritebackLatency) { + mode_lib->vba.WritebackLatencySupport = false; + } + } else { + if (locals->WriteBandwidth[k] + > 1.5 + * dml_min( + mode_lib->vba.WritebackInterfaceLumaBufferSize, + 2.0 + * mode_lib->vba.WritebackInterfaceChromaBufferSize) + / mode_lib->vba.WritebackLatency) { + mode_lib->vba.WritebackLatencySupport = false; + } + } + } + } + /*Re-ordering Buffer Support Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i] = + (mode_lib->vba.RoundTripPingLatencyCycles + 32.0) / mode_lib->vba.DCFCLKPerState[i] + + locals->UrgentOutOfOrderReturnPerChannel * mode_lib->vba.NumberOfChannels / locals->ReturnBWPerState[i][0]; + if ((mode_lib->vba.ROBBufferSizeInKByte - mode_lib->vba.PixelChunkSizeInKByte) * 1024.0 / locals->ReturnBWPerState[i][0] + > locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i]) { + locals->ROBSupport[i][0] = true; + } else { + locals->ROBSupport[i][0] = false; + } + } + /*Writeback Mode Support Check*/ + + mode_lib->vba.TotalNumberOfActiveWriteback = 0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.ActiveWritebacksPerPlane[k] == 0) + mode_lib->vba.ActiveWritebacksPerPlane[k] = 1; + mode_lib->vba.TotalNumberOfActiveWriteback = + mode_lib->vba.TotalNumberOfActiveWriteback + + mode_lib->vba.ActiveWritebacksPerPlane[k]; + } + } + mode_lib->vba.WritebackModeSupport = true; + if (mode_lib->vba.TotalNumberOfActiveWriteback > mode_lib->vba.MaxNumWriteback) { + mode_lib->vba.WritebackModeSupport = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.Writeback10bpc420Supported != true + && mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + mode_lib->vba.WritebackModeSupport = false; + } + } + /*Writeback Scale Ratio and Taps Support Check*/ + + mode_lib->vba.WritebackScaleRatioAndTapsSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.WritebackLumaAndChromaScalingSupported == false + && (mode_lib->vba.WritebackHRatio[k] != 1.0 + || mode_lib->vba.WritebackVRatio[k] != 1.0)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (mode_lib->vba.WritebackHRatio[k] > mode_lib->vba.WritebackMaxHSCLRatio + || mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackMaxVSCLRatio + || mode_lib->vba.WritebackHRatio[k] + < mode_lib->vba.WritebackMinHSCLRatio + || mode_lib->vba.WritebackVRatio[k] + < mode_lib->vba.WritebackMinVSCLRatio + || mode_lib->vba.WritebackLumaHTaps[k] + > mode_lib->vba.WritebackMaxHSCLTaps + || mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackMaxVSCLTaps + || mode_lib->vba.WritebackHRatio[k] + > mode_lib->vba.WritebackLumaHTaps[k] + || mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackLumaVTaps[k] + || (mode_lib->vba.WritebackLumaHTaps[k] > 2.0 + && ((mode_lib->vba.WritebackLumaHTaps[k] % 2) + == 1)) + || (mode_lib->vba.WritebackPixelFormat[k] != dm_444_32 + && (mode_lib->vba.WritebackChromaHTaps[k] + > mode_lib->vba.WritebackMaxHSCLTaps + || mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackMaxVSCLTaps + || 2.0 + * mode_lib->vba.WritebackHRatio[k] + > mode_lib->vba.WritebackChromaHTaps[k] + || 2.0 + * mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackChromaVTaps[k] + || (mode_lib->vba.WritebackChromaHTaps[k] > 2.0 + && ((mode_lib->vba.WritebackChromaHTaps[k] % 2) == 1))))) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (mode_lib->vba.WritebackVRatio[k] < 1.0) { + mode_lib->vba.WritebackLumaVExtra = + dml_max(1.0 - 2.0 / dml_ceil(1.0 / mode_lib->vba.WritebackVRatio[k], 1.0), 0.0); + } else { + mode_lib->vba.WritebackLumaVExtra = -1; + } + if ((mode_lib->vba.WritebackPixelFormat[k] == dm_444_32 + && mode_lib->vba.WritebackLumaVTaps[k] + > (mode_lib->vba.WritebackLineBufferLumaBufferSize + + mode_lib->vba.WritebackLineBufferChromaBufferSize) + / 3.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_8 + && mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackLineBufferLumaBufferSize + * 8.0 / 10.0 / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10 + && mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackLineBufferLumaBufferSize + * 8.0 / 10.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (2.0 * mode_lib->vba.WritebackVRatio[k] < 1) { + mode_lib->vba.WritebackChromaVExtra = 0.0; + } else { + mode_lib->vba.WritebackChromaVExtra = -1; + } + if ((mode_lib->vba.WritebackPixelFormat[k] == dm_420_8 + && mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackLineBufferChromaBufferSize + * 8.0 / 10.0 / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackChromaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10 + && mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackLineBufferChromaBufferSize + * 8.0 / 10.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackChromaVExtra)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + } + } + /*Maximum DISPCLK/DPPCLK Support check*/ + + mode_lib->vba.WritebackRequiredDISPCLK = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + mode_lib->vba.WritebackRequiredDISPCLK = + dml_max( + mode_lib->vba.WritebackRequiredDISPCLK, + CalculateWriteBackDISPCLK( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.PixelClock[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.WritebackChromaLineBufferWidth)); + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.HRatio[k] > 1.0) { + locals->PSCL_FACTOR[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / dml_ceil( + mode_lib->vba.htaps[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_FACTOR[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + if (locals->BytePerPixelInDETC[k] == 0.0) { + locals->PSCL_FACTOR_CHROMA[k] = 0.0; + locals->MinDPPCLKUsingSingleDPP[k] = + mode_lib->vba.PixelClock[k] + * dml_max3( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_FACTOR[k], + 1.0); + if ((mode_lib->vba.htaps[k] > 6.0 || mode_lib->vba.vtaps[k] > 6.0) + && locals->MinDPPCLKUsingSingleDPP[k] + < 2.0 * mode_lib->vba.PixelClock[k]) { + locals->MinDPPCLKUsingSingleDPP[k] = 2.0 + * mode_lib->vba.PixelClock[k]; + } + } else { + if (mode_lib->vba.HRatio[k] / 2.0 > 1.0) { + locals->PSCL_FACTOR_CHROMA[k] = + dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / 2.0 + / dml_ceil( + mode_lib->vba.HTAPsChroma[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_FACTOR_CHROMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + locals->MinDPPCLKUsingSingleDPP[k] = + mode_lib->vba.PixelClock[k] + * dml_max5( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_FACTOR[k], + mode_lib->vba.VTAPsChroma[k] + / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k] + / 2.0), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / 4.0 + / locals->PSCL_FACTOR_CHROMA[k], + 1.0); + if ((mode_lib->vba.htaps[k] > 6.0 || mode_lib->vba.vtaps[k] > 6.0 + || mode_lib->vba.HTAPsChroma[k] > 6.0 + || mode_lib->vba.VTAPsChroma[k] > 6.0) + && locals->MinDPPCLKUsingSingleDPP[k] + < 2.0 * mode_lib->vba.PixelClock[k]) { + locals->MinDPPCLKUsingSingleDPP[k] = 2.0 + * mode_lib->vba.PixelClock[k]; + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + Calculate256BBlockSizes( + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + dml_ceil(locals->BytePerPixelInDETC[k], 2.0), + &locals->Read256BlockHeightY[k], + &locals->Read256BlockHeightC[k], + &locals->Read256BlockWidthY[k], + &locals->Read256BlockWidthC[k]); + if (mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MaxSwathHeightY[k] = locals->Read256BlockHeightY[k]; + locals->MaxSwathHeightC[k] = locals->Read256BlockHeightC[k]; + } else { + locals->MaxSwathHeightY[k] = locals->Read256BlockWidthY[k]; + locals->MaxSwathHeightC[k] = locals->Read256BlockWidthC[k]; + } + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + || (mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_t + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s_x) + && mode_lib->vba.SourceScan[k] == dm_horz)) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + } else { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k] + / 2.0; + } + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8 + && mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k] + / 2.0; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10 + && mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k] + / 2.0; + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + } else { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } + } + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + mode_lib->vba.MaximumSwathWidthSupport = 8192.0; + } else { + mode_lib->vba.MaximumSwathWidthSupport = 5120.0; + } + mode_lib->vba.MaximumSwathWidthInDETBuffer = + dml_min( + mode_lib->vba.MaximumSwathWidthSupport, + mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0 + / (locals->BytePerPixelInDETY[k] + * locals->MinSwathHeightY[k] + + locals->BytePerPixelInDETC[k] + / 2.0 + * locals->MinSwathHeightC[k])); + if (locals->BytePerPixelInDETC[k] == 0.0) { + mode_lib->vba.MaximumSwathWidthInLineBuffer = + mode_lib->vba.LineBufferSize + * dml_max(mode_lib->vba.HRatio[k], 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.vtaps[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k], + 1.0) + - 2, + 0.0)); + } else { + mode_lib->vba.MaximumSwathWidthInLineBuffer = + dml_min( + mode_lib->vba.LineBufferSize + * dml_max( + mode_lib->vba.HRatio[k], + 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.vtaps[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k], + 1.0) + - 2, + 0.0)), + 2.0 * mode_lib->vba.LineBufferSize + * dml_max( + mode_lib->vba.HRatio[k] + / 2.0, + 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.VTAPsChroma[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k] + / 2.0, + 1.0) + - 2, + 0.0))); + } + locals->MaximumSwathWidth[k] = dml_min( + mode_lib->vba.MaximumSwathWidthInDETBuffer, + mode_lib->vba.MaximumSwathWidthInLineBuffer); + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + double MaxMaxDispclkRoundedDown = RoundToDFSGranularityDown( + mode_lib->vba.MaxDispclk[mode_lib->vba.soc.num_states], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + + for (j = 0; j < 2; j++) { + mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.MaxDispclk[i], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.MaxDppclk[i], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + locals->RequiredDISPCLK[i][j] = 0.0; + locals->DISPCLK_DPPCLK_Support[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine = + mode_lib->vba.PixelClock[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + * (1.0 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + if (mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine >= mode_lib->vba.MaxDispclk[i] + && i == mode_lib->vba.soc.num_states) + mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine = mode_lib->vba.PixelClock[k] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + + mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine = mode_lib->vba.PixelClock[k] / 2 + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * (1 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + if (mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine >= mode_lib->vba.MaxDispclk[i] + && i == mode_lib->vba.soc.num_states) + mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine = mode_lib->vba.PixelClock[k] / 2 + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + + locals->ODMCombineEnablePerState[i][k] = false; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine; + if (mode_lib->vba.ODMCapability) { + if (locals->PlaneRequiredDISPCLKWithoutODMCombine > MaxMaxDispclkRoundedDown) { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } else if (locals->DSCEnabled[k] && (locals->HActive[k] > DCN20_MAX_DSC_IMAGE_WIDTH)) { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } else if (locals->HActive[k] > DCN20_MAX_420_IMAGE_WIDTH && locals->OutputFormat[k] == dm_420) { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } + } + + if (locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) <= mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity + && locals->SwathWidthYSingleDPP[k] <= locals->MaximumSwathWidth[k] + && locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_disabled) { + locals->NoOfDPP[i][j][k] = 1; + locals->RequiredDPPCLK[i][j][k] = + locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } else { + locals->NoOfDPP[i][j][k] = 2; + locals->RequiredDPPCLK[i][j][k] = + locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) / 2.0; + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.PlaneRequiredDISPCLK); + if ((locals->MinDPPCLKUsingSingleDPP[k] / locals->NoOfDPP[i][j][k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + > mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity) + || (mode_lib->vba.PlaneRequiredDISPCLK > mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity)) { + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + } + locals->TotalNumberOfActiveDPP[i][j] = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + if (j == 1) { + while (locals->TotalNumberOfActiveDPP[i][j] < mode_lib->vba.MaxNumDPP + && locals->TotalNumberOfActiveDPP[i][j] < 2 * mode_lib->vba.NumberOfActivePlanes) { + double BWOfNonSplitPlaneOfMaximumBandwidth; + unsigned int NumberOfNonSplitPlaneOfMaximumBandwidth; + + BWOfNonSplitPlaneOfMaximumBandwidth = 0; + NumberOfNonSplitPlaneOfMaximumBandwidth = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->ReadBandwidth[k] > BWOfNonSplitPlaneOfMaximumBandwidth && locals->NoOfDPP[i][j][k] == 1) { + BWOfNonSplitPlaneOfMaximumBandwidth = locals->ReadBandwidth[k]; + NumberOfNonSplitPlaneOfMaximumBandwidth = k; + } + } + locals->NoOfDPP[i][j][NumberOfNonSplitPlaneOfMaximumBandwidth] = 2; + locals->RequiredDPPCLK[i][j][NumberOfNonSplitPlaneOfMaximumBandwidth] = + locals->MinDPPCLKUsingSingleDPP[NumberOfNonSplitPlaneOfMaximumBandwidth] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100) / 2; + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + 1; + } + } + if (locals->TotalNumberOfActiveDPP[i][j] > mode_lib->vba.MaxNumDPP) { + locals->RequiredDISPCLK[i][j] = 0.0; + locals->DISPCLK_DPPCLK_Support[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->ODMCombineEnablePerState[i][k] = false; + if (locals->SwathWidthYSingleDPP[k] <= locals->MaximumSwathWidth[k]) { + locals->NoOfDPP[i][j][k] = 1; + locals->RequiredDPPCLK[i][j][k] = locals->MinDPPCLKUsingSingleDPP[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } else { + locals->NoOfDPP[i][j][k] = 2; + locals->RequiredDPPCLK[i][j][k] = locals->MinDPPCLKUsingSingleDPP[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) / 2.0; + } + if (i != mode_lib->vba.soc.num_states) { + mode_lib->vba.PlaneRequiredDISPCLK = + mode_lib->vba.PixelClock[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + * (1.0 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + } else { + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PixelClock[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.PlaneRequiredDISPCLK); + if (locals->MinDPPCLKUsingSingleDPP[k] / locals->NoOfDPP[i][j][k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + > mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity + || mode_lib->vba.PlaneRequiredDISPCLK > mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity) + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + locals->TotalNumberOfActiveDPP[i][j] = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.WritebackRequiredDISPCLK); + if (mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity + < mode_lib->vba.WritebackRequiredDISPCLK) { + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + } + } + /*Viewport Size Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->ViewportSizeSupport[i][0] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { + if (dml_min(locals->SwathWidthYSingleDPP[k], dml_round(mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k])) + > locals->MaximumSwathWidth[k]) { + locals->ViewportSizeSupport[i][0] = false; + } + } else { + if (locals->SwathWidthYSingleDPP[k] / 2.0 > locals->MaximumSwathWidth[k]) { + locals->ViewportSizeSupport[i][0] = false; + } + } + } + } + /*Total Available Pipes Support Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + if (locals->TotalNumberOfActiveDPP[i][j] <= mode_lib->vba.MaxNumDPP) + locals->TotalAvailablePipesSupport[i][j] = true; + else + locals->TotalAvailablePipesSupport[i][j] = false; + } + } + /*Total Available OTG Support Check*/ + + mode_lib->vba.TotalNumberOfActiveOTG = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.TotalNumberOfActiveOTG = mode_lib->vba.TotalNumberOfActiveOTG + + 1.0; + } + } + if (mode_lib->vba.TotalNumberOfActiveOTG <= mode_lib->vba.MaxNumOTG) { + mode_lib->vba.NumberOfOTGSupport = true; + } else { + mode_lib->vba.NumberOfOTGSupport = false; + } + /*Display IO and DSC Support Check*/ + + mode_lib->vba.NonsupportedDSCInputBPC = false; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (!(mode_lib->vba.DSCInputBitPerComponent[k] == 12.0 + || mode_lib->vba.DSCInputBitPerComponent[k] == 10.0 + || mode_lib->vba.DSCInputBitPerComponent[k] == 8.0)) { + mode_lib->vba.NonsupportedDSCInputBPC = true; + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->RequiresDSC[i][k] = 0; + locals->RequiresFEC[i][k] = 0; + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.Output[k] == dm_hdmi) { + locals->RequiresDSC[i][k] = 0; + locals->RequiresFEC[i][k] = 0; + locals->OutputBppPerState[i][k] = TruncToValidBPP( + dml_min(600.0, mode_lib->vba.PHYCLKPerState[i]) / mode_lib->vba.PixelClockBackEnd[k] * 24, + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + } else if (mode_lib->vba.Output[k] == dm_dp + || mode_lib->vba.Output[k] == dm_edp) { + if (mode_lib->vba.Output[k] == dm_edp) { + mode_lib->vba.EffectiveFECOverhead = 0.0; + } else { + mode_lib->vba.EffectiveFECOverhead = + mode_lib->vba.FECOverhead; + } + if (mode_lib->vba.PHYCLKPerState[i] >= 270.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 270.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 270.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = mode_lib->vba.Outbpp; + } + if (mode_lib->vba.Outbpp == BPP_INVALID && mode_lib->vba.PHYCLKPerState[i] >= 540.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 540.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 540.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = mode_lib->vba.Outbpp; + } + if (mode_lib->vba.Outbpp == BPP_INVALID + && mode_lib->vba.PHYCLKPerState[i] + >= 810.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 810.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 810.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true || mode_lib->vba.Outbpp == BPP_INVALID) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = + mode_lib->vba.Outbpp; + } + } + } else { + locals->OutputBppPerState[i][k] = BPP_BLENDED_PIPE; + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->DIOSupport[i] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->OutputBppPerState[i][k] == BPP_INVALID + || (mode_lib->vba.OutputFormat[k] == dm_420 + && mode_lib->vba.Interlace[k] == true + && mode_lib->vba.ProgressiveToInterlaceUnitInOPP == true)) { + locals->DIOSupport[i] = false; + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->DSCCLKRequiredMoreThanSupported[i] = false; + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if ((mode_lib->vba.Output[k] == dm_dp + || mode_lib->vba.Output[k] == dm_edp)) { + if (mode_lib->vba.OutputFormat[k] == dm_420 + || mode_lib->vba.OutputFormat[k] + == dm_n422) { + mode_lib->vba.DSCFormatFactor = 2; + } else { + mode_lib->vba.DSCFormatFactor = 1; + } + if (locals->RequiresDSC[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { + if (mode_lib->vba.PixelClockBackEnd[k] / 6.0 / mode_lib->vba.DSCFormatFactor + > (1.0 - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * mode_lib->vba.MaxDSCCLK[i]) { + locals->DSCCLKRequiredMoreThanSupported[i] = + true; + } + } else { + if (mode_lib->vba.PixelClockBackEnd[k] / 3.0 / mode_lib->vba.DSCFormatFactor + > (1.0 - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * mode_lib->vba.MaxDSCCLK[i]) { + locals->DSCCLKRequiredMoreThanSupported[i] = + true; + } + } + } + } + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->NotEnoughDSCUnits[i] = false; + mode_lib->vba.TotalDSCUnitsRequired = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->RequiresDSC[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { + mode_lib->vba.TotalDSCUnitsRequired = + mode_lib->vba.TotalDSCUnitsRequired + 2.0; + } else { + mode_lib->vba.TotalDSCUnitsRequired = + mode_lib->vba.TotalDSCUnitsRequired + 1.0; + } + } + } + if (mode_lib->vba.TotalDSCUnitsRequired > mode_lib->vba.NumberOfDSC) { + locals->NotEnoughDSCUnits[i] = true; + } + } + /*DSC Delay per state*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] != k) { + mode_lib->vba.slices = 0; + } else if (locals->RequiresDSC[i][k] == 0 + || locals->RequiresDSC[i][k] == false) { + mode_lib->vba.slices = 0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 3200.0) { + mode_lib->vba.slices = dml_ceil( + mode_lib->vba.PixelClockBackEnd[k] / 400.0, + 4.0); + } else if (mode_lib->vba.PixelClockBackEnd[k] > 1360.0) { + mode_lib->vba.slices = 8.0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 680.0) { + mode_lib->vba.slices = 4.0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 340.0) { + mode_lib->vba.slices = 2.0; + } else { + mode_lib->vba.slices = 1.0; + } + if (locals->OutputBppPerState[i][k] == BPP_BLENDED_PIPE + || locals->OutputBppPerState[i][k] == BPP_INVALID) { + mode_lib->vba.bpp = 0.0; + } else { + mode_lib->vba.bpp = locals->OutputBppPerState[i][k]; + } + if (locals->RequiresDSC[i][k] == true && mode_lib->vba.bpp != 0.0) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_disabled) { + locals->DSCDelayPerState[i][k] = + dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + mode_lib->vba.bpp, + dml_ceil( + mode_lib->vba.HActive[k] + / mode_lib->vba.slices, + 1.0), + mode_lib->vba.slices, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k]); + } else { + locals->DSCDelayPerState[i][k] = + 2.0 * (dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + mode_lib->vba.bpp, + dml_ceil(mode_lib->vba.HActive[k] / mode_lib->vba.slices, 1.0), + mode_lib->vba.slices / 2, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay(mode_lib->vba.OutputFormat[k])); + } + locals->DSCDelayPerState[i][k] = + locals->DSCDelayPerState[i][k] * mode_lib->vba.PixelClock[k] / mode_lib->vba.PixelClockBackEnd[k]; + } else { + locals->DSCDelayPerState[i][k] = 0.0; + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + for (j = 0; j <= mode_lib->vba.NumberOfActivePlanes - 1; j++) { + if (mode_lib->vba.BlendingAndTiming[k] == m && locals->RequiresDSC[i][m] == true) + locals->DSCDelayPerState[i][k] = locals->DSCDelayPerState[i][m]; + } + } + } + } + + //Prefetch Check + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) + locals->SwathWidthYPerState[i][j][k] = dml_min(locals->SwathWidthYSingleDPP[k], dml_round(locals->HActive[k] / 2 * locals->HRatio[k])); + else + locals->SwathWidthYPerState[i][j][k] = locals->SwathWidthYSingleDPP[k] / locals->NoOfDPP[i][j][k]; + locals->SwathWidthGranularityY = 256 / dml_ceil(locals->BytePerPixelInDETY[k], 1) / locals->MaxSwathHeightY[k]; + locals->RoundedUpMaxSwathSizeBytesY = (dml_ceil(locals->SwathWidthYPerState[i][j][k] - 1, locals->SwathWidthGranularityY) + + locals->SwathWidthGranularityY) * locals->BytePerPixelInDETY[k] * locals->MaxSwathHeightY[k]; + if (locals->SourcePixelFormat[k] == dm_420_10) { + locals->RoundedUpMaxSwathSizeBytesY = dml_ceil(locals->RoundedUpMaxSwathSizeBytesY, 256) + 256; + } + if (locals->MaxSwathHeightC[k] > 0) { + locals->SwathWidthGranularityC = 256 / dml_ceil(locals->BytePerPixelInDETC[k], 2) / locals->MaxSwathHeightC[k]; + + locals->RoundedUpMaxSwathSizeBytesC = (dml_ceil(locals->SwathWidthYPerState[i][j][k] / 2 - 1, locals->SwathWidthGranularityC) + + locals->SwathWidthGranularityC) * locals->BytePerPixelInDETC[k] * locals->MaxSwathHeightC[k]; + } + if (locals->SourcePixelFormat[k] == dm_420_10) { + locals->RoundedUpMaxSwathSizeBytesC = dml_ceil(locals->RoundedUpMaxSwathSizeBytesC, 256) + 256; + } else { + locals->RoundedUpMaxSwathSizeBytesC = 0; + } + + if (locals->RoundedUpMaxSwathSizeBytesY + locals->RoundedUpMaxSwathSizeBytesC <= locals->DETBufferSizeInKByte * 1024 / 2) { + locals->SwathHeightYPerState[i][j][k] = locals->MaxSwathHeightY[k]; + locals->SwathHeightCPerState[i][j][k] = locals->MaxSwathHeightC[k]; + } else { + locals->SwathHeightYPerState[i][j][k] = locals->MinSwathHeightY[k]; + locals->SwathHeightCPerState[i][j][k] = locals->MinSwathHeightC[k]; + } + + if (locals->BytePerPixelInDETC[k] == 0) { + locals->LinesInDETLuma = locals->DETBufferSizeInKByte * 1024 / locals->BytePerPixelInDETY[k] / locals->SwathWidthYPerState[i][j][k]; + locals->LinesInDETChroma = 0; + } else if (locals->SwathHeightYPerState[i][j][k] <= locals->SwathHeightCPerState[i][j][k]) { + locals->LinesInDETLuma = locals->DETBufferSizeInKByte * 1024 / 2 / locals->BytePerPixelInDETY[k] / + locals->SwathWidthYPerState[i][j][k]; + locals->LinesInDETChroma = locals->DETBufferSizeInKByte * 1024 / 2 / locals->BytePerPixelInDETC[k] / (locals->SwathWidthYPerState[i][j][k] / 2); + } else { + locals->LinesInDETLuma = locals->DETBufferSizeInKByte * 1024 * 2 / 3 / locals->BytePerPixelInDETY[k] / locals->SwathWidthYPerState[i][j][k]; + locals->LinesInDETChroma = locals->DETBufferSizeInKByte * 1024 / 3 / locals->BytePerPixelInDETY[k] / (locals->SwathWidthYPerState[i][j][k] / 2); + } + + locals->EffectiveLBLatencyHidingSourceLinesLuma = dml_min(locals->MaxLineBufferLines, + dml_floor(locals->LineBufferSize / locals->LBBitPerPixel[k] / (locals->SwathWidthYPerState[i][j][k] + / dml_max(locals->HRatio[k], 1)), 1)) - (locals->vtaps[k] - 1); + + locals->EffectiveLBLatencyHidingSourceLinesChroma = dml_min(locals->MaxLineBufferLines, + dml_floor(locals->LineBufferSize / locals->LBBitPerPixel[k] + / (locals->SwathWidthYPerState[i][j][k] / 2 + / dml_max(locals->HRatio[k] / 2, 1)), 1)) - (locals->VTAPsChroma[k] - 1); + + locals->EffectiveDETLBLinesLuma = dml_floor(locals->LinesInDETLuma + dml_min( + locals->LinesInDETLuma * locals->RequiredDISPCLK[i][j] * locals->BytePerPixelInDETY[k] * + locals->PSCL_FACTOR[k] / locals->ReturnBWPerState[i][0], + locals->EffectiveLBLatencyHidingSourceLinesLuma), + locals->SwathHeightYPerState[i][j][k]); + + locals->EffectiveDETLBLinesChroma = dml_floor(locals->LinesInDETChroma + dml_min( + locals->LinesInDETChroma * locals->RequiredDISPCLK[i][j] * locals->BytePerPixelInDETC[k] * + locals->PSCL_FACTOR_CHROMA[k] / locals->ReturnBWPerState[i][0], + locals->EffectiveLBLatencyHidingSourceLinesChroma), + locals->SwathHeightCPerState[i][j][k]); + + if (locals->BytePerPixelInDETC[k] == 0) { + locals->UrgentLatencySupportUsPerState[i][j][k] = locals->EffectiveDETLBLinesLuma * (locals->HTotal[k] / locals->PixelClock[k]) + / locals->VRatio[k] - locals->EffectiveDETLBLinesLuma * locals->SwathWidthYPerState[i][j][k] * + dml_ceil(locals->BytePerPixelInDETY[k], 1) / (locals->ReturnBWPerState[i][0] / locals->NoOfDPP[i][j][k]); + } else { + locals->UrgentLatencySupportUsPerState[i][j][k] = dml_min( + locals->EffectiveDETLBLinesLuma * (locals->HTotal[k] / locals->PixelClock[k]) + / locals->VRatio[k] - locals->EffectiveDETLBLinesLuma * locals->SwathWidthYPerState[i][j][k] * + dml_ceil(locals->BytePerPixelInDETY[k], 1) / (locals->ReturnBWPerState[i][0] / locals->NoOfDPP[i][j][k]), + locals->EffectiveDETLBLinesChroma * (locals->HTotal[k] / locals->PixelClock[k]) / (locals->VRatio[k] / 2) - + locals->EffectiveDETLBLinesChroma * locals->SwathWidthYPerState[i][j][k] / 2 * + dml_ceil(locals->BytePerPixelInDETC[k], 2) / (locals->ReturnBWPerState[i][0] / locals->NoOfDPP[i][j][k])); + } + } + } + } + + for (i = 0; i <= locals->soc.num_states; i++) { + for (j = 0; j < 2; j++) { + locals->UrgentLatencySupport[i][j] = true; + for (k = 0; k < locals->NumberOfActivePlanes; k++) { + if (locals->UrgentLatencySupportUsPerState[i][j][k] < locals->UrgentLatency) + locals->UrgentLatencySupport[i][j] = false; + } + } + } + + + /*Prefetch Check*/ + for (i = 0; i <= locals->soc.num_states; i++) { + for (j = 0; j < 2; j++) { + locals->TotalNumberOfDCCActiveDPP[i][j] = 0; + for (k = 0; k < locals->NumberOfActivePlanes; k++) { + if (locals->DCCEnable[k] == true) { + locals->TotalNumberOfDCCActiveDPP[i][j] = + locals->TotalNumberOfDCCActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + } + } + } + } + + CalculateMinAndMaxPrefetchMode(locals->AllowDRAMSelfRefreshOrDRAMClockChangeInVblank, &locals->MinPrefetchMode, &locals->MaxPrefetchMode); + + locals->MaxTotalVActiveRDBandwidth = 0; + for (k = 0; k < locals->NumberOfActivePlanes; k++) { + locals->MaxTotalVActiveRDBandwidth = locals->MaxTotalVActiveRDBandwidth + locals->ReadBandwidth[k]; + } + + for (i = 0; i <= locals->soc.num_states; i++) { + for (j = 0; j < 2; j++) { + for (k = 0; k < locals->NumberOfActivePlanes; k++) { + locals->NoOfDPPThisState[k] = locals->NoOfDPP[i][j][k]; + locals->RequiredDPPCLKThisState[k] = locals->RequiredDPPCLK[i][j][k]; + locals->SwathHeightYThisState[k] = locals->SwathHeightYPerState[i][j][k]; + locals->SwathHeightCThisState[k] = locals->SwathHeightCPerState[i][j][k]; + locals->SwathWidthYThisState[k] = locals->SwathWidthYPerState[i][j][k]; + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], + mode_lib->vba.PixelClock[k] / 16.0); + if (mode_lib->vba.BytePerPixelInDETC[k] == 0.0) { + if (mode_lib->vba.VRatio[k] <= 1.0) { + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETY[k], + 1.0) + / 64.0 + * mode_lib->vba.HRatio[k] + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.NoOfDPP[i][j][k]); + } else { + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETY[k], + 1.0) + / 64.0 + * mode_lib->vba.PSCL_FACTOR[k] + * mode_lib->vba.RequiredDPPCLK[i][j][k]); + } + } else { + if (mode_lib->vba.VRatio[k] <= 1.0) { + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETY[k], + 1.0) + / 32.0 + * mode_lib->vba.HRatio[k] + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.NoOfDPP[i][j][k]); + } else { + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETY[k], + 1.0) + / 32.0 + * mode_lib->vba.PSCL_FACTOR[k] + * mode_lib->vba.RequiredDPPCLK[i][j][k]); + } + if (mode_lib->vba.VRatio[k] / 2.0 <= 1.0) { + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETC[k], + 2.0) + / 32.0 + * mode_lib->vba.HRatio[k] + / 2.0 + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.NoOfDPP[i][j][k]); + } else { + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0] = + dml_max( + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], + 1.1 + * dml_ceil( + mode_lib->vba.BytePerPixelInDETC[k], + 2.0) + / 32.0 + * mode_lib->vba.PSCL_FACTOR_CHROMA[k] + * mode_lib->vba.RequiredDPPCLK[i][j][k]); + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameY = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + mode_lib->vba.Read256BlockHeightY[k], + mode_lib->vba.Read256BlockWidthY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(mode_lib->vba.BytePerPixelInDETY[k], 1.0), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + mode_lib->vba.SwathWidthYPerState[i][j][k], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsLuma, + mode_lib->vba.PDEProcessingBufIn64KBReqs, + mode_lib->vba.PitchY[k], + mode_lib->vba.DCCMetaPitchY[k], + &mode_lib->vba.MacroTileWidthY[k], + &mode_lib->vba.MetaRowBytesY, + &mode_lib->vba.DPTEBytesPerRowY, + &mode_lib->vba.PTEBufferSizeNotExceededY[i][j][k], + &mode_lib->vba.dpte_row_height[k], + &mode_lib->vba.meta_row_height[k]); + mode_lib->vba.PrefetchLinesY[0][0][k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.vtaps[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightYPerState[i][j][k], + mode_lib->vba.ViewportYStartY[k], + &mode_lib->vba.PrefillY[k], + &mode_lib->vba.MaxNumSwY[k]); + if ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8)) { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + mode_lib->vba.Read256BlockHeightY[k], + mode_lib->vba.Read256BlockWidthY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(mode_lib->vba.BytePerPixelInDETC[k], 2.0), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k] / 2.0, + mode_lib->vba.ViewportHeight[k] / 2.0, + mode_lib->vba.SwathWidthYPerState[i][j][k] / 2.0, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsLuma, + mode_lib->vba.PDEProcessingBufIn64KBReqs, + mode_lib->vba.PitchC[k], + 0.0, + &mode_lib->vba.MacroTileWidthC[k], + &mode_lib->vba.MetaRowBytesC, + &mode_lib->vba.DPTEBytesPerRowC, + &mode_lib->vba.PTEBufferSizeNotExceededC[i][j][k], + &mode_lib->vba.dpte_row_height_chroma[k], + &mode_lib->vba.meta_row_height_chroma[k]); + mode_lib->vba.PrefetchLinesC[0][0][k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k] / 2.0, + mode_lib->vba.VTAPsChroma[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightCPerState[i][j][k], + mode_lib->vba.ViewportYStartC[k], + &mode_lib->vba.PrefillC[k], + &mode_lib->vba.MaxNumSwC[k]); + } else { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC = 0.0; + mode_lib->vba.MetaRowBytesC = 0.0; + mode_lib->vba.DPTEBytesPerRowC = 0.0; + locals->PrefetchLinesC[0][0][k] = 0.0; + locals->PTEBufferSizeNotExceededC[i][j][k] = true; + locals->PTEBufferSizeInRequestsForLuma = mode_lib->vba.PTEBufferSizeInRequestsLuma + mode_lib->vba.PTEBufferSizeInRequestsChroma; + } + locals->PDEAndMetaPTEBytesPerFrame[0][0][k] = + mode_lib->vba.PDEAndMetaPTEBytesPerFrameY + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC; + locals->MetaRowBytes[0][0][k] = mode_lib->vba.MetaRowBytesY + mode_lib->vba.MetaRowBytesC; + locals->DPTEBytesPerRow[0][0][k] = mode_lib->vba.DPTEBytesPerRowY + mode_lib->vba.DPTEBytesPerRowC; + + CalculateActiveRowBandwidth( + mode_lib->vba.GPUVMEnable, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.MetaRowBytesY, + mode_lib->vba.MetaRowBytesC, + mode_lib->vba.meta_row_height[k], + mode_lib->vba.meta_row_height_chroma[k], + mode_lib->vba.DPTEBytesPerRowY, + mode_lib->vba.DPTEBytesPerRowC, + mode_lib->vba.dpte_row_height[k], + mode_lib->vba.dpte_row_height_chroma[k], + &mode_lib->vba.meta_row_bw[k], + &mode_lib->vba.dpte_row_bw[k], + &mode_lib->vba.qual_row_bw[k]); + } + mode_lib->vba.ExtraLatency = + mode_lib->vba.UrgentRoundTripAndOutOfOrderLatencyPerState[i] + + (mode_lib->vba.TotalNumberOfActiveDPP[i][j] + * mode_lib->vba.PixelChunkSizeInKByte + + mode_lib->vba.TotalNumberOfDCCActiveDPP[i][j] + * mode_lib->vba.MetaChunkSize) + * 1024.0 + / mode_lib->vba.ReturnBWPerState[i][0]; + if (mode_lib->vba.GPUVMEnable == true) { + mode_lib->vba.ExtraLatency = mode_lib->vba.ExtraLatency + + mode_lib->vba.TotalNumberOfActiveDPP[i][j] + * mode_lib->vba.PTEGroupSize + / mode_lib->vba.ReturnBWPerState[i][0]; + } + mode_lib->vba.TimeCalc = 24.0 / mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0]; + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WritebackDelay[i][k] = mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k]) / locals->RequiredDISPCLK[i][j]; + } else { + locals->WritebackDelay[i][k] = 0.0; + } + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + if (mode_lib->vba.BlendingAndTiming[m] == k + && mode_lib->vba.WritebackEnable[m] + == true) { + locals->WritebackDelay[i][k] = dml_max(locals->WritebackDelay[i][k], + mode_lib->vba.WritebackLatency + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[m], + mode_lib->vba.WritebackHRatio[m], + mode_lib->vba.WritebackVRatio[m], + mode_lib->vba.WritebackLumaHTaps[m], + mode_lib->vba.WritebackLumaVTaps[m], + mode_lib->vba.WritebackChromaHTaps[m], + mode_lib->vba.WritebackChromaVTaps[m], + mode_lib->vba.WritebackDestinationWidth[m]) / locals->RequiredDISPCLK[i][j]); + } + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + if (mode_lib->vba.BlendingAndTiming[k] == m) { + locals->WritebackDelay[i][k] = locals->WritebackDelay[i][m]; + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m < locals->NumberOfCursors[k]; m++) + locals->cursor_bw[k] = locals->NumberOfCursors[k] * locals->CursorWidth[k][m] * locals->CursorBPP[k][m] + / 8 / (locals->HTotal[k] / locals->PixelClock[k]) * locals->VRatio[k]; + } + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->MaximumVStartup[0][0][k] = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] + - dml_max(1.0, dml_ceil(locals->WritebackDelay[i][k] / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]), 1.0)); + } + + mode_lib->vba.NextPrefetchMode = mode_lib->vba.MinPrefetchMode; + do { + mode_lib->vba.PrefetchMode[i][j] = mode_lib->vba.NextPrefetchMode; + mode_lib->vba.NextPrefetchMode = mode_lib->vba.NextPrefetchMode + 1; + + mode_lib->vba.TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[i][j], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.SREnterPlusExitTime); + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + + if (mode_lib->vba.XFCEnabled[k] == true) { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = + CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + locals->SwathWidthYPerState[i][j][k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TimeCalc, + mode_lib->vba.TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + } else { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = 0.0; + } + + CalculateDelayAfterScaler(mode_lib, mode_lib->vba.ReturnBWPerState[i][0], mode_lib->vba.ReadBandwidthLuma[k], mode_lib->vba.ReadBandwidthChroma[k], mode_lib->vba.MaxTotalVActiveRDBandwidth, + mode_lib->vba.DisplayPipeLineDeliveryTimeLuma[k], mode_lib->vba.DisplayPipeLineDeliveryTimeChroma[k], + mode_lib->vba.RequiredDPPCLK[i][j][k], mode_lib->vba.RequiredDISPCLK[i][j], mode_lib->vba.PixelClock[k], mode_lib->vba.DSCDelayPerState[i][k], mode_lib->vba.NoOfDPP[i][j][k], mode_lib->vba.ScalerEnabled[k], mode_lib->vba.NumberOfCursors[k], + mode_lib->vba.DPPCLKDelaySubtotal, mode_lib->vba.DPPCLKDelaySCL, mode_lib->vba.DPPCLKDelaySCLLBOnly, mode_lib->vba.DPPCLKDelayCNVCFormater, mode_lib->vba.DPPCLKDelayCNVCCursor, mode_lib->vba.DISPCLKDelaySubtotal, + mode_lib->vba.SwathWidthYPerState[i][j][k] / mode_lib->vba.HRatio[k], mode_lib->vba.OutputFormat[k], mode_lib->vba.HTotal[k], + mode_lib->vba.SwathWidthYSingleDPP[k], mode_lib->vba.BytePerPixelInDETY[k], mode_lib->vba.BytePerPixelInDETC[k], mode_lib->vba.SwathHeightYThisState[k], mode_lib->vba.SwathHeightCThisState[k], mode_lib->vba.Interlace[k], mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + &mode_lib->vba.DSTXAfterScaler[k], &mode_lib->vba.DSTYAfterScaler[k]); + + mode_lib->vba.IsErrorResult[i][j][k] = + CalculatePrefetchSchedule( + mode_lib, + mode_lib->vba.RequiredDPPCLK[i][j][k], + mode_lib->vba.RequiredDISPCLK[i][j], + mode_lib->vba.PixelClock[k], + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], + mode_lib->vba.NoOfDPP[i][j][k], + mode_lib->vba.NumberOfCursors[k], + mode_lib->vba.VTotal[k] + - mode_lib->vba.VActive[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.MaxInterDCNTileRepeaters, + mode_lib->vba.MaximumVStartup[0][0][k], + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.DynamicMetadataEnable[k], + mode_lib->vba.DynamicMetadataLinesBeforeActiveRequired[k], + mode_lib->vba.DynamicMetadataTransmittedBytes[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.ExtraLatency, + mode_lib->vba.TimeCalc, + mode_lib->vba.PDEAndMetaPTEBytesPerFrame[0][0][k], + mode_lib->vba.MetaRowBytes[0][0][k], + mode_lib->vba.DPTEBytesPerRow[0][0][k], + mode_lib->vba.PrefetchLinesY[0][0][k], + mode_lib->vba.SwathWidthYPerState[i][j][k], + mode_lib->vba.BytePerPixelInDETY[k], + mode_lib->vba.PrefillY[k], + mode_lib->vba.MaxNumSwY[k], + mode_lib->vba.PrefetchLinesC[0][0][k], + mode_lib->vba.BytePerPixelInDETC[k], + mode_lib->vba.PrefillC[k], + mode_lib->vba.MaxNumSwC[k], + mode_lib->vba.SwathHeightYPerState[i][j][k], + mode_lib->vba.SwathHeightCPerState[i][j][k], + mode_lib->vba.TWait, + mode_lib->vba.XFCEnabled[k], + mode_lib->vba.XFCRemoteSurfaceFlipDelay, + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.DSTXAfterScaler[k], + mode_lib->vba.DSTYAfterScaler[k], + &mode_lib->vba.LineTimesForPrefetch[k], + &mode_lib->vba.PrefetchBW[k], + &mode_lib->vba.LinesForMetaPTE[k], + &mode_lib->vba.LinesForMetaAndDPTERow[k], + &mode_lib->vba.VRatioPreY[i][j][k], + &mode_lib->vba.VRatioPreC[i][j][k], + &mode_lib->vba.RequiredPrefetchPixelDataBWLuma[i][j][k], + &mode_lib->vba.Tno_bw[k], + &mode_lib->vba.VUpdateOffsetPix[k], + &mode_lib->vba.VUpdateWidthPix[k], + &mode_lib->vba.VReadyOffsetPix[k]); + } + mode_lib->vba.MaximumReadBandwidthWithoutPrefetch = 0.0; + mode_lib->vba.MaximumReadBandwidthWithPrefetch = 0.0; + locals->prefetch_vm_bw_valid = true; + locals->prefetch_row_bw_valid = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->PDEAndMetaPTEBytesPerFrame[0][0][k] == 0) + locals->prefetch_vm_bw[k] = 0; + else if (locals->LinesForMetaPTE[k] > 0) + locals->prefetch_vm_bw[k] = locals->PDEAndMetaPTEBytesPerFrame[0][0][k] + / (locals->LinesForMetaPTE[k] * locals->HTotal[k] / locals->PixelClock[k]); + else { + locals->prefetch_vm_bw[k] = 0; + locals->prefetch_vm_bw_valid = false; + } + if (locals->MetaRowBytes[0][0][k] + locals->DPTEBytesPerRow[0][0][k] == 0) + locals->prefetch_row_bw[k] = 0; + else if (locals->LinesForMetaAndDPTERow[k] > 0) + locals->prefetch_row_bw[k] = (locals->MetaRowBytes[0][0][k] + locals->DPTEBytesPerRow[0][0][k]) + / (locals->LinesForMetaAndDPTERow[k] * locals->HTotal[k] / locals->PixelClock[k]); + else { + locals->prefetch_row_bw[k] = 0; + locals->prefetch_row_bw_valid = false; + } + + mode_lib->vba.MaximumReadBandwidthWithoutPrefetch = mode_lib->vba.MaximumReadBandwidthWithPrefetch + + mode_lib->vba.cursor_bw[k] + mode_lib->vba.ReadBandwidth[k] + mode_lib->vba.meta_row_bw[k] + mode_lib->vba.dpte_row_bw[k]; + mode_lib->vba.MaximumReadBandwidthWithPrefetch = + mode_lib->vba.MaximumReadBandwidthWithPrefetch + + mode_lib->vba.cursor_bw[k] + + dml_max3( + mode_lib->vba.prefetch_vm_bw[k], + mode_lib->vba.prefetch_row_bw[k], + dml_max(mode_lib->vba.ReadBandwidth[k], + mode_lib->vba.RequiredPrefetchPixelDataBWLuma[i][j][k]) + + mode_lib->vba.meta_row_bw[k] + mode_lib->vba.dpte_row_bw[k]); + } + locals->BandwidthWithoutPrefetchSupported[i][0] = true; + if (mode_lib->vba.MaximumReadBandwidthWithoutPrefetch > locals->ReturnBWPerState[i][0]) { + locals->BandwidthWithoutPrefetchSupported[i][0] = false; + } + + locals->PrefetchSupported[i][j] = true; + if (mode_lib->vba.MaximumReadBandwidthWithPrefetch > locals->ReturnBWPerState[i][0]) { + locals->PrefetchSupported[i][j] = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->LineTimesForPrefetch[k] < 2.0 + || locals->LinesForMetaPTE[k] >= 8.0 + || locals->LinesForMetaAndDPTERow[k] >= 16.0 + || mode_lib->vba.IsErrorResult[i][j][k] == true) { + locals->PrefetchSupported[i][j] = false; + } + } + locals->VRatioInPrefetchSupported[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->VRatioPreY[i][j][k] > 4.0 + || locals->VRatioPreC[i][j][k] > 4.0 + || mode_lib->vba.IsErrorResult[i][j][k] == true) { + locals->VRatioInPrefetchSupported[i][j] = false; + } + } + } while ((locals->PrefetchSupported[i][j] != true || locals->VRatioInPrefetchSupported[i][j] != true) + && mode_lib->vba.NextPrefetchMode < mode_lib->vba.MaxPrefetchMode); + + if (mode_lib->vba.PrefetchSupported[i][j] == true + && mode_lib->vba.VRatioInPrefetchSupported[i][j] == true) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = + mode_lib->vba.ReturnBWPerState[i][0]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = + mode_lib->vba.BandwidthAvailableForImmediateFlip + - mode_lib->vba.cursor_bw[k] + - dml_max( + mode_lib->vba.ReadBandwidth[k] + mode_lib->vba.qual_row_bw[k], + mode_lib->vba.PrefetchBW[k]); + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.ImmediateFlipBytes[k] = 0.0; + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + mode_lib->vba.ImmediateFlipBytes[k] = + mode_lib->vba.PDEAndMetaPTEBytesPerFrame[0][0][k] + + mode_lib->vba.MetaRowBytes[0][0][k] + + mode_lib->vba.DPTEBytesPerRow[0][0][k]; + } + } + mode_lib->vba.TotImmediateFlipBytes = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + mode_lib->vba.TotImmediateFlipBytes = + mode_lib->vba.TotImmediateFlipBytes + + mode_lib->vba.ImmediateFlipBytes[k]; + } + } + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + CalculateFlipSchedule( + mode_lib, + mode_lib->vba.ExtraLatency, + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.BandwidthAvailableForImmediateFlip, + mode_lib->vba.TotImmediateFlipBytes, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.ImmediateFlipBytes[k], + mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.Tno_bw[k], + mode_lib->vba.PDEAndMetaPTEBytesPerFrame[0][0][k], + mode_lib->vba.MetaRowBytes[0][0][k], + mode_lib->vba.DPTEBytesPerRow[0][0][k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.dpte_row_height[k], + mode_lib->vba.meta_row_height[k], + mode_lib->vba.qual_row_bw[k], + &mode_lib->vba.DestinationLinesToRequestVMInImmediateFlip[k], + &mode_lib->vba.DestinationLinesToRequestRowInImmediateFlip[k], + &mode_lib->vba.final_flip_bw[k], + &mode_lib->vba.ImmediateFlipSupportedForPipe[k]); + } + mode_lib->vba.total_dcn_read_bw_with_flip = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.total_dcn_read_bw_with_flip = + mode_lib->vba.total_dcn_read_bw_with_flip + + mode_lib->vba.cursor_bw[k] + + dml_max3( + mode_lib->vba.prefetch_vm_bw[k], + mode_lib->vba.prefetch_row_bw[k], + mode_lib->vba.final_flip_bw[k] + + dml_max( + mode_lib->vba.ReadBandwidth[k], + mode_lib->vba.RequiredPrefetchPixelDataBWLuma[i][j][k])); + } + mode_lib->vba.ImmediateFlipSupportedForState[i][j] = true; + if (mode_lib->vba.total_dcn_read_bw_with_flip + > mode_lib->vba.ReturnBWPerState[i][0]) { + mode_lib->vba.ImmediateFlipSupportedForState[i][j] = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.ImmediateFlipSupportedForPipe[k] == false) { + mode_lib->vba.ImmediateFlipSupportedForState[i][j] = false; + } + } + } else { + mode_lib->vba.ImmediateFlipSupportedForState[i][j] = false; + } + } + } + + /*Vertical Active BW support*/ + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + mode_lib->vba.MaxTotalVerticalActiveAvailableBandwidth[i][0] = dml_min(mode_lib->vba.ReturnBusWidth * + mode_lib->vba.DCFCLKPerState[i], mode_lib->vba.FabricAndDRAMBandwidthPerState[i] * 1000) * + mode_lib->vba.MaxAveragePercentOfIdealDRAMBWDisplayCanUseInNormalSystemOperation / 100; + if (mode_lib->vba.MaxTotalVActiveRDBandwidth <= mode_lib->vba.MaxTotalVerticalActiveAvailableBandwidth[i][0]) + mode_lib->vba.TotalVerticalActiveBandwidthSupport[i][0] = true; + else + mode_lib->vba.TotalVerticalActiveBandwidthSupport[i][0] = false; + } + + /*PTE Buffer Size Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + locals->PTEBufferSizeNotExceeded[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->PTEBufferSizeNotExceededY[i][j][k] == false + || locals->PTEBufferSizeNotExceededC[i][j][k] == false) { + locals->PTEBufferSizeNotExceeded[i][j] = false; + } + } + } + } + /*Cursor Support Check*/ + mode_lib->vba.CursorSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (j = 0; j < 2; j++) { + if (mode_lib->vba.CursorWidth[k][j] > 0.0) { + if (dml_floor( + dml_floor( + mode_lib->vba.CursorBufferSize + - mode_lib->vba.CursorChunkSize, + mode_lib->vba.CursorChunkSize) * 1024.0 + / (mode_lib->vba.CursorWidth[k][j] + * mode_lib->vba.CursorBPP[k][j] + / 8.0), + 1.0) + * (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + / mode_lib->vba.VRatio[k] < mode_lib->vba.UrgentLatencyPixelDataOnly + || (mode_lib->vba.CursorBPP[k][j] == 64.0 + && mode_lib->vba.Cursor64BppSupport == false)) { + mode_lib->vba.CursorSupport = false; + } + } + } + } + /*Valid Pitch Check*/ + + mode_lib->vba.PitchSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->AlignedYPitch[k] = dml_ceil( + dml_max(mode_lib->vba.PitchY[k], mode_lib->vba.ViewportWidth[k]), + locals->MacroTileWidthY[k]); + if (locals->AlignedYPitch[k] > mode_lib->vba.PitchY[k]) { + mode_lib->vba.PitchSupport = false; + } + if (mode_lib->vba.DCCEnable[k] == true) { + locals->AlignedDCCMetaPitch[k] = dml_ceil( + dml_max( + mode_lib->vba.DCCMetaPitchY[k], + mode_lib->vba.ViewportWidth[k]), + 64.0 * locals->Read256BlockWidthY[k]); + } else { + locals->AlignedDCCMetaPitch[k] = mode_lib->vba.DCCMetaPitchY[k]; + } + if (locals->AlignedDCCMetaPitch[k] > mode_lib->vba.DCCMetaPitchY[k]) { + mode_lib->vba.PitchSupport = false; + } + if (mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8) { + locals->AlignedCPitch[k] = dml_ceil( + dml_max( + mode_lib->vba.PitchC[k], + mode_lib->vba.ViewportWidth[k] / 2.0), + locals->MacroTileWidthC[k]); + } else { + locals->AlignedCPitch[k] = mode_lib->vba.PitchC[k]; + } + if (locals->AlignedCPitch[k] > mode_lib->vba.PitchC[k]) { + mode_lib->vba.PitchSupport = false; + } + } + /*Mode Support, Voltage State and SOC Configuration*/ + + for (i = mode_lib->vba.soc.num_states; i >= 0; i--) { + for (j = 0; j < 2; j++) { + enum dm_validation_status status = DML_VALIDATION_OK; + + if (mode_lib->vba.ScaleRatioAndTapsSupport != true) { + status = DML_FAIL_SCALE_RATIO_TAP; + } else if (mode_lib->vba.SourceFormatPixelAndScanSupport != true) { + status = DML_FAIL_SOURCE_PIXEL_FORMAT; + } else if (locals->ViewportSizeSupport[i][0] != true) { + status = DML_FAIL_VIEWPORT_SIZE; + } else if (locals->DIOSupport[i] != true) { + status = DML_FAIL_DIO_SUPPORT; + } else if (locals->NotEnoughDSCUnits[i] != false) { + status = DML_FAIL_NOT_ENOUGH_DSC; + } else if (locals->DSCCLKRequiredMoreThanSupported[i] != false) { + status = DML_FAIL_DSC_CLK_REQUIRED; + } else if (locals->UrgentLatencySupport[i][j] != true) { + status = DML_FAIL_URGENT_LATENCY; + } else if (locals->ROBSupport[i][0] != true) { + status = DML_FAIL_REORDERING_BUFFER; + } else if (locals->DISPCLK_DPPCLK_Support[i][j] != true) { + status = DML_FAIL_DISPCLK_DPPCLK; + } else if (locals->TotalAvailablePipesSupport[i][j] != true) { + status = DML_FAIL_TOTAL_AVAILABLE_PIPES; + } else if (mode_lib->vba.NumberOfOTGSupport != true) { + status = DML_FAIL_NUM_OTG; + } else if (mode_lib->vba.WritebackModeSupport != true) { + status = DML_FAIL_WRITEBACK_MODE; + } else if (mode_lib->vba.WritebackLatencySupport != true) { + status = DML_FAIL_WRITEBACK_LATENCY; + } else if (mode_lib->vba.WritebackScaleRatioAndTapsSupport != true) { + status = DML_FAIL_WRITEBACK_SCALE_RATIO_TAP; + } else if (mode_lib->vba.CursorSupport != true) { + status = DML_FAIL_CURSOR_SUPPORT; + } else if (mode_lib->vba.PitchSupport != true) { + status = DML_FAIL_PITCH_SUPPORT; + } else if (locals->PrefetchSupported[i][j] != true) { + status = DML_FAIL_PREFETCH_SUPPORT; + } else if (locals->TotalVerticalActiveBandwidthSupport[i][0] != true) { + status = DML_FAIL_TOTAL_V_ACTIVE_BW; + } else if (locals->VRatioInPrefetchSupported[i][j] != true) { + status = DML_FAIL_V_RATIO_PREFETCH; + } else if (locals->PTEBufferSizeNotExceeded[i][j] != true) { + status = DML_FAIL_PTE_BUFFER_SIZE; + } else if (mode_lib->vba.NonsupportedDSCInputBPC != false) { + status = DML_FAIL_DSC_INPUT_BPC; + } + + if (status == DML_VALIDATION_OK) { + locals->ModeSupport[i][j] = true; + } else { + locals->ModeSupport[i][j] = false; + } + locals->ValidationStatus[i] = status; + } + } + { + unsigned int MaximumMPCCombine = 0; + mode_lib->vba.VoltageLevel = mode_lib->vba.soc.num_states + 1; + for (i = mode_lib->vba.VoltageOverrideLevel; i <= mode_lib->vba.soc.num_states; i++) { + if (locals->ModeSupport[i][0] == true || locals->ModeSupport[i][1] == true) { + mode_lib->vba.VoltageLevel = i; + if (locals->ModeSupport[i][1] == true && (locals->ModeSupport[i][0] == false + || mode_lib->vba.WhenToDoMPCCombine == dm_mpc_always_when_possible)) { + MaximumMPCCombine = 1; + } else { + MaximumMPCCombine = 0; + } + break; + } + } + mode_lib->vba.ImmediateFlipSupport = + locals->ImmediateFlipSupportedForState[mode_lib->vba.VoltageLevel][MaximumMPCCombine]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.DPPPerPlane[k] = locals->NoOfDPP[mode_lib->vba.VoltageLevel][MaximumMPCCombine][k]; + locals->DPPCLK[k] = locals->RequiredDPPCLK[mode_lib->vba.VoltageLevel][MaximumMPCCombine][k]; + } + mode_lib->vba.DISPCLK = locals->RequiredDISPCLK[mode_lib->vba.VoltageLevel][MaximumMPCCombine]; + mode_lib->vba.maxMpcComb = MaximumMPCCombine; + } + mode_lib->vba.DCFCLK = mode_lib->vba.DCFCLKPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.DRAMSpeed = mode_lib->vba.DRAMSpeedPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.FabricClock = mode_lib->vba.FabricClockPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.SOCCLK = mode_lib->vba.SOCCLKPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.ReturnBW = locals->ReturnBWPerState[mode_lib->vba.VoltageLevel][0]; + mode_lib->vba.FabricAndDRAMBandwidth = locals->FabricAndDRAMBandwidthPerState[mode_lib->vba.VoltageLevel]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.ODMCombineEnabled[k] = + locals->ODMCombineEnablePerState[mode_lib->vba.VoltageLevel][k]; + } else { + mode_lib->vba.ODMCombineEnabled[k] = 0; + } + mode_lib->vba.DSCEnabled[k] = + locals->RequiresDSC[mode_lib->vba.VoltageLevel][k]; + mode_lib->vba.OutputBpp[k] = + locals->OutputBppPerState[mode_lib->vba.VoltageLevel][k]; + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.h new file mode 100644 index 000000000000..a989d3ca1e99 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.h @@ -0,0 +1,32 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DCN20V2_DISPLAY_MODE_VBA_H_ +#define _DCN20V2_DISPLAY_MODE_VBA_H_ + +void dml20v2_recalculate(struct display_mode_lib *mode_lib); +void dml20v2_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c index 878bf4782ce6..ca807846032f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c @@ -107,10 +107,10 @@ static unsigned int get_bytes_per_element(enum source_format_class source_format static bool is_dual_plane(enum source_format_class source_format) { - bool ret_val = 0; + bool ret_val = false; if ((source_format == dm_420_8) || (source_format == dm_420_10)) - ret_val = 1; + ret_val = true; return ret_val; } @@ -207,7 +207,7 @@ static void extract_rq_regs(struct display_mode_lib *mode_lib, rq_regs->rq_regs_l.swath_height = dml_log2(rq_param.dlg.rq_l.swath_height); rq_regs->rq_regs_c.swath_height = dml_log2(rq_param.dlg.rq_c.swath_height); - // FIXME: take the max between luma, chroma chunk size? + // TODO: take the max between luma, chroma chunk size? // okay for now, as we are setting chunk_bytes to 8kb anyways if (rq_param.sizing.rq_l.chunk_bytes >= 32 * 1024) { //32kb rq_regs->drq_expansion_mode = 0; @@ -240,8 +240,8 @@ static void handle_det_buf_split(struct display_mode_lib *mode_lib, unsigned int swath_bytes_c = 0; unsigned int full_swath_bytes_packed_l = 0; unsigned int full_swath_bytes_packed_c = 0; - bool req128_l = 0; - bool req128_c = 0; + bool req128_l = false; + bool req128_c = false; bool surf_linear = (pipe_src_param.sw_mode == dm_sw_linear); bool surf_vert = (pipe_src_param.source_scan == dm_vert); unsigned int log2_swath_height_l = 0; @@ -264,13 +264,13 @@ static void handle_det_buf_split(struct display_mode_lib *mode_lib, total_swath_bytes = 2 * full_swath_bytes_packed_l + 2 * full_swath_bytes_packed_c; if (total_swath_bytes <= detile_buf_size_in_bytes) { //full 256b request - req128_l = 0; - req128_c = 0; + req128_l = false; + req128_c = false; swath_bytes_l = full_swath_bytes_packed_l; swath_bytes_c = full_swath_bytes_packed_c; } else { //128b request (for luma only for yuv420 8bpc) - req128_l = 1; - req128_c = 0; + req128_l = true; + req128_c = false; swath_bytes_l = full_swath_bytes_packed_l / 2; swath_bytes_c = full_swath_bytes_packed_c; } @@ -280,9 +280,9 @@ static void handle_det_buf_split(struct display_mode_lib *mode_lib, total_swath_bytes = 2 * full_swath_bytes_packed_l; if (total_swath_bytes <= detile_buf_size_in_bytes) - req128_l = 0; + req128_l = false; else - req128_l = 1; + req128_l = true; swath_bytes_l = total_swath_bytes; swath_bytes_c = 0; @@ -670,14 +670,14 @@ static void get_surf_rq_param(struct display_mode_lib *mode_lib, const display_pipe_source_params_st pipe_src_param, bool is_chroma) { - bool mode_422 = 0; + bool mode_422 = false; unsigned int vp_width = 0; unsigned int vp_height = 0; unsigned int data_pitch = 0; unsigned int meta_pitch = 0; unsigned int ppe = mode_422 ? 2 : 1; - // FIXME check if ppe apply for both luma and chroma in 422 case + // TODO check if ppe apply for both luma and chroma in 422 case if (is_chroma) { vp_width = pipe_src_param.viewport_width_c / ppe; vp_height = pipe_src_param.viewport_height_c; @@ -929,8 +929,7 @@ static void dml20_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib, min_dst_y_ttu_vblank = min_ttu_vblank * pclk_freq_in_mhz / (double) htotal; dlg_vblank_start = interlaced ? (vblank_start / 2) : vblank_start; - disp_dlg_regs->min_dst_y_next_start = (unsigned int) (((double) dlg_vblank_start - + min_dst_y_ttu_vblank) * dml_pow(2, 2)); + disp_dlg_regs->min_dst_y_next_start = (unsigned int) ((double) dlg_vblank_start * dml_pow(2, 2)); ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int) dml_pow(2, 18)); dml_print("DML_DLG: %s: min_dcfclk_mhz = %3.2f\n", @@ -959,7 +958,7 @@ static void dml20_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib, // Source // dcc_en = src.dcc; dual_plane = is_dual_plane((enum source_format_class)(src->source_format)); - mode_422 = 0; // FIXME + mode_422 = false; // TODO access_dir = (src->source_scan == dm_vert); // vp access direction: horizontal or vertical accessed // bytes_per_element_l = get_bytes_per_element(source_format_class(src.source_format), 0); // bytes_per_element_c = get_bytes_per_element(source_format_class(src.source_format), 1); @@ -1655,7 +1654,7 @@ static void calculate_ttu_cursor(struct display_mode_lib *mode_lib, cur_width_ub = dml_ceil((double) cur_src_width / (double) cur_req_width, 1) * (double) cur_req_width; cur_req_per_width = cur_width_ub / (double) cur_req_width; - hactive_cur = (double) cur_src_width / hscl_ratio; // FIXME: oswin to think about what to do for cursor + hactive_cur = (double) cur_src_width / hscl_ratio; // TODO: oswin to think about what to do for cursor if (vratio_pre_l <= 1.0) { *refcyc_per_req_delivery_pre_cur = hactive_cur * ref_freq_to_pix_freq diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c new file mode 100644 index 000000000000..287b7a0ad108 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c @@ -0,0 +1,1701 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "../display_mode_lib.h" +#include "../display_mode_vba.h" +#include "display_rq_dlg_calc_20v2.h" + +// Function: dml20v2_rq_dlg_get_rq_params +// Calculate requestor related parameters that register definition agnostic +// (i.e. this layer does try to separate real values from register definition) +// Input: +// pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.) +// Output: +// rq_param - values that can be used to setup RQ (e.g. swath_height, plane1_addr, etc.) +// +static void dml20v2_rq_dlg_get_rq_params( + struct display_mode_lib *mode_lib, + display_rq_params_st * rq_param, + const display_pipe_source_params_st pipe_src_param); + +// Function: dml20v2_rq_dlg_get_dlg_params +// Calculate deadline related parameters +// +static void dml20v2_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib, + const display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + display_dlg_regs_st *disp_dlg_regs, + display_ttu_regs_st *disp_ttu_regs, + const display_rq_dlg_params_st rq_dlg_param, + const display_dlg_sys_params_st dlg_sys_param, + const bool cstate_en, + const bool pstate_en); +/* + * NOTE: + * This file is gcc-parseable HW gospel, coming straight from HW engineers. + * + * It doesn't adhere to Linux kernel style and sometimes will do things in odd + * ways. Unless there is something clearly wrong with it the code should + * remain as-is as it provides us with a guarantee from HW that it is correct. + */ + +static void calculate_ttu_cursor(struct display_mode_lib *mode_lib, + double *refcyc_per_req_delivery_pre_cur, + double *refcyc_per_req_delivery_cur, + double refclk_freq_in_mhz, + double ref_freq_to_pix_freq, + double hscale_pixel_rate_l, + double hscl_ratio, + double vratio_pre_l, + double vratio_l, + unsigned int cur_width, + enum cursor_bpp cur_bpp); + +#include "../dml_inline_defs.h" + +static unsigned int get_bytes_per_element(enum source_format_class source_format, bool is_chroma) +{ + unsigned int ret_val = 0; + + if (source_format == dm_444_16) { + if (!is_chroma) + ret_val = 2; + } else if (source_format == dm_444_32) { + if (!is_chroma) + ret_val = 4; + } else if (source_format == dm_444_64) { + if (!is_chroma) + ret_val = 8; + } else if (source_format == dm_420_8) { + if (is_chroma) + ret_val = 2; + else + ret_val = 1; + } else if (source_format == dm_420_10) { + if (is_chroma) + ret_val = 4; + else + ret_val = 2; + } else if (source_format == dm_444_8) { + ret_val = 1; + } + return ret_val; +} + +static bool is_dual_plane(enum source_format_class source_format) +{ + bool ret_val = false; + + if ((source_format == dm_420_8) || (source_format == dm_420_10)) + ret_val = true; + + return ret_val; +} + +static double get_refcyc_per_delivery(struct display_mode_lib *mode_lib, + double refclk_freq_in_mhz, + double pclk_freq_in_mhz, + bool odm_combine, + unsigned int recout_width, + unsigned int hactive, + double vratio, + double hscale_pixel_rate, + unsigned int delivery_width, + unsigned int req_per_swath_ub) +{ + double refcyc_per_delivery = 0.0; + + if (vratio <= 1.0) { + if (odm_combine) + refcyc_per_delivery = (double) refclk_freq_in_mhz + * dml_min((double) recout_width, (double) hactive / 2.0) + / pclk_freq_in_mhz / (double) req_per_swath_ub; + else + refcyc_per_delivery = (double) refclk_freq_in_mhz * (double) recout_width + / pclk_freq_in_mhz / (double) req_per_swath_ub; + } else { + refcyc_per_delivery = (double) refclk_freq_in_mhz * (double) delivery_width + / (double) hscale_pixel_rate / (double) req_per_swath_ub; + } + + dml_print("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, refclk_freq_in_mhz); + dml_print("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, pclk_freq_in_mhz); + dml_print("DML_DLG: %s: recout_width = %d\n", __func__, recout_width); + dml_print("DML_DLG: %s: vratio = %3.2f\n", __func__, vratio); + dml_print("DML_DLG: %s: req_per_swath_ub = %d\n", __func__, req_per_swath_ub); + dml_print("DML_DLG: %s: refcyc_per_delivery= %3.2f\n", __func__, refcyc_per_delivery); + + return refcyc_per_delivery; + +} + +static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_size) +{ + if (tile_size == dm_256k_tile) + return (256 * 1024); + else if (tile_size == dm_64k_tile) + return (64 * 1024); + else + return (4 * 1024); +} + +static void extract_rq_sizing_regs(struct display_mode_lib *mode_lib, + display_data_rq_regs_st *rq_regs, + const display_data_rq_sizing_params_st rq_sizing) +{ + dml_print("DML_DLG: %s: rq_sizing param\n", __func__); + print__data_rq_sizing_params_st(mode_lib, rq_sizing); + + rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10; + + if (rq_sizing.min_chunk_bytes == 0) + rq_regs->min_chunk_size = 0; + else + rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1; + + rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10; + if (rq_sizing.min_meta_chunk_bytes == 0) + rq_regs->min_meta_chunk_size = 0; + else + rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1; + + rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6; + rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6; +} + +static void extract_rq_regs(struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_rq_params_st rq_param) +{ + unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024; + unsigned int detile_buf_plane1_addr = 0; + + extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l); + + rq_regs->rq_regs_l.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_l.dpte_row_height), + 1) - 3; + + if (rq_param.yuv420) { + extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c); + rq_regs->rq_regs_c.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_c.dpte_row_height), + 1) - 3; + } + + rq_regs->rq_regs_l.swath_height = dml_log2(rq_param.dlg.rq_l.swath_height); + rq_regs->rq_regs_c.swath_height = dml_log2(rq_param.dlg.rq_c.swath_height); + + // TODO: take the max between luma, chroma chunk size? + // okay for now, as we are setting chunk_bytes to 8kb anyways + if (rq_param.sizing.rq_l.chunk_bytes >= 32 * 1024) { //32kb + rq_regs->drq_expansion_mode = 0; + } else { + rq_regs->drq_expansion_mode = 2; + } + rq_regs->prq_expansion_mode = 1; + rq_regs->mrq_expansion_mode = 1; + rq_regs->crq_expansion_mode = 1; + + if (rq_param.yuv420) { + if ((double) rq_param.misc.rq_l.stored_swath_bytes + / (double) rq_param.misc.rq_c.stored_swath_bytes <= 1.5) { + detile_buf_plane1_addr = (detile_buf_size_in_bytes / 2.0 / 64.0); // half to chroma + } else { + detile_buf_plane1_addr = dml_round_to_multiple((unsigned int) ((2.0 * detile_buf_size_in_bytes) / 3.0), + 256, + 0) / 64.0; // 2/3 to chroma + } + } + rq_regs->plane1_base_address = detile_buf_plane1_addr; +} + +static void handle_det_buf_split(struct display_mode_lib *mode_lib, + display_rq_params_st *rq_param, + const display_pipe_source_params_st pipe_src_param) +{ + unsigned int total_swath_bytes = 0; + unsigned int swath_bytes_l = 0; + unsigned int swath_bytes_c = 0; + unsigned int full_swath_bytes_packed_l = 0; + unsigned int full_swath_bytes_packed_c = 0; + bool req128_l = false; + bool req128_c = false; + bool surf_linear = (pipe_src_param.sw_mode == dm_sw_linear); + bool surf_vert = (pipe_src_param.source_scan == dm_vert); + unsigned int log2_swath_height_l = 0; + unsigned int log2_swath_height_c = 0; + unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024; + + full_swath_bytes_packed_l = rq_param->misc.rq_l.full_swath_bytes; + full_swath_bytes_packed_c = rq_param->misc.rq_c.full_swath_bytes; + + if (rq_param->yuv420_10bpc) { + full_swath_bytes_packed_l = dml_round_to_multiple(rq_param->misc.rq_l.full_swath_bytes * 2 / 3, + 256, + 1) + 256; + full_swath_bytes_packed_c = dml_round_to_multiple(rq_param->misc.rq_c.full_swath_bytes * 2 / 3, + 256, + 1) + 256; + } + + if (rq_param->yuv420) { + total_swath_bytes = 2 * full_swath_bytes_packed_l + 2 * full_swath_bytes_packed_c; + + if (total_swath_bytes <= detile_buf_size_in_bytes) { //full 256b request + req128_l = false; + req128_c = false; + swath_bytes_l = full_swath_bytes_packed_l; + swath_bytes_c = full_swath_bytes_packed_c; + } else { //128b request (for luma only for yuv420 8bpc) + req128_l = true; + req128_c = false; + swath_bytes_l = full_swath_bytes_packed_l / 2; + swath_bytes_c = full_swath_bytes_packed_c; + } + // Note: assumption, the config that pass in will fit into + // the detiled buffer. + } else { + total_swath_bytes = 2 * full_swath_bytes_packed_l; + + if (total_swath_bytes <= detile_buf_size_in_bytes) + req128_l = false; + else + req128_l = true; + + swath_bytes_l = total_swath_bytes; + swath_bytes_c = 0; + } + rq_param->misc.rq_l.stored_swath_bytes = swath_bytes_l; + rq_param->misc.rq_c.stored_swath_bytes = swath_bytes_c; + + if (surf_linear) { + log2_swath_height_l = 0; + log2_swath_height_c = 0; + } else if (!surf_vert) { + log2_swath_height_l = dml_log2(rq_param->misc.rq_l.blk256_height) - req128_l; + log2_swath_height_c = dml_log2(rq_param->misc.rq_c.blk256_height) - req128_c; + } else { + log2_swath_height_l = dml_log2(rq_param->misc.rq_l.blk256_width) - req128_l; + log2_swath_height_c = dml_log2(rq_param->misc.rq_c.blk256_width) - req128_c; + } + rq_param->dlg.rq_l.swath_height = 1 << log2_swath_height_l; + rq_param->dlg.rq_c.swath_height = 1 << log2_swath_height_c; + + dml_print("DML_DLG: %s: req128_l = %0d\n", __func__, req128_l); + dml_print("DML_DLG: %s: req128_c = %0d\n", __func__, req128_c); + dml_print("DML_DLG: %s: full_swath_bytes_packed_l = %0d\n", + __func__, + full_swath_bytes_packed_l); + dml_print("DML_DLG: %s: full_swath_bytes_packed_c = %0d\n", + __func__, + full_swath_bytes_packed_c); +} + +static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, + display_data_rq_dlg_params_st *rq_dlg_param, + display_data_rq_misc_params_st *rq_misc_param, + display_data_rq_sizing_params_st *rq_sizing_param, + unsigned int vp_width, + unsigned int vp_height, + unsigned int data_pitch, + unsigned int meta_pitch, + unsigned int source_format, + unsigned int tiling, + unsigned int macro_tile_size, + unsigned int source_scan, + unsigned int is_chroma) +{ + bool surf_linear = (tiling == dm_sw_linear); + bool surf_vert = (source_scan == dm_vert); + + unsigned int bytes_per_element; + unsigned int bytes_per_element_y = get_bytes_per_element((enum source_format_class)(source_format), + false); + unsigned int bytes_per_element_c = get_bytes_per_element((enum source_format_class)(source_format), + true); + + unsigned int blk256_width = 0; + unsigned int blk256_height = 0; + + unsigned int blk256_width_y = 0; + unsigned int blk256_height_y = 0; + unsigned int blk256_width_c = 0; + unsigned int blk256_height_c = 0; + unsigned int log2_bytes_per_element; + unsigned int log2_blk256_width; + unsigned int log2_blk256_height; + unsigned int blk_bytes; + unsigned int log2_blk_bytes; + unsigned int log2_blk_height; + unsigned int log2_blk_width; + unsigned int log2_meta_req_bytes; + unsigned int log2_meta_req_height; + unsigned int log2_meta_req_width; + unsigned int meta_req_width; + unsigned int meta_req_height; + unsigned int log2_meta_row_height; + unsigned int meta_row_width_ub; + unsigned int log2_meta_chunk_bytes; + unsigned int log2_meta_chunk_height; + + //full sized meta chunk width in unit of data elements + unsigned int log2_meta_chunk_width; + unsigned int log2_min_meta_chunk_bytes; + unsigned int min_meta_chunk_width; + unsigned int meta_chunk_width; + unsigned int meta_chunk_per_row_int; + unsigned int meta_row_remainder; + unsigned int meta_chunk_threshold; + unsigned int meta_blk_bytes; + unsigned int meta_blk_height; + unsigned int meta_blk_width; + unsigned int meta_surface_bytes; + unsigned int vmpg_bytes; + unsigned int meta_pte_req_per_frame_ub; + unsigned int meta_pte_bytes_per_frame_ub; + const unsigned int log2_vmpg_bytes = dml_log2(mode_lib->soc.vmm_page_size_bytes); + const unsigned int dpte_buf_in_pte_reqs = mode_lib->ip.dpte_buffer_size_in_pte_reqs_luma; + const unsigned int pde_proc_buffer_size_64k_reqs = + mode_lib->ip.pde_proc_buffer_size_64k_reqs; + + unsigned int log2_vmpg_height = 0; + unsigned int log2_vmpg_width = 0; + unsigned int log2_dpte_req_height_ptes = 0; + unsigned int log2_dpte_req_height = 0; + unsigned int log2_dpte_req_width = 0; + unsigned int log2_dpte_row_height_linear = 0; + unsigned int log2_dpte_row_height = 0; + unsigned int log2_dpte_group_width = 0; + unsigned int dpte_row_width_ub = 0; + unsigned int dpte_req_height = 0; + unsigned int dpte_req_width = 0; + unsigned int dpte_group_width = 0; + unsigned int log2_dpte_group_bytes = 0; + unsigned int log2_dpte_group_length = 0; + unsigned int pde_buf_entries; + bool yuv420 = (source_format == dm_420_8 || source_format == dm_420_10); + + Calculate256BBlockSizes((enum source_format_class)(source_format), + (enum dm_swizzle_mode)(tiling), + bytes_per_element_y, + bytes_per_element_c, + &blk256_height_y, + &blk256_height_c, + &blk256_width_y, + &blk256_width_c); + + if (!is_chroma) { + blk256_width = blk256_width_y; + blk256_height = blk256_height_y; + bytes_per_element = bytes_per_element_y; + } else { + blk256_width = blk256_width_c; + blk256_height = blk256_height_c; + bytes_per_element = bytes_per_element_c; + } + + log2_bytes_per_element = dml_log2(bytes_per_element); + + dml_print("DML_DLG: %s: surf_linear = %d\n", __func__, surf_linear); + dml_print("DML_DLG: %s: surf_vert = %d\n", __func__, surf_vert); + dml_print("DML_DLG: %s: blk256_width = %d\n", __func__, blk256_width); + dml_print("DML_DLG: %s: blk256_height = %d\n", __func__, blk256_height); + + log2_blk256_width = dml_log2((double) blk256_width); + log2_blk256_height = dml_log2((double) blk256_height); + blk_bytes = surf_linear ? + 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); + log2_blk_bytes = dml_log2((double) blk_bytes); + log2_blk_height = 0; + log2_blk_width = 0; + + // remember log rule + // "+" in log is multiply + // "-" in log is divide + // "/2" is like square root + // blk is vertical biased + if (tiling != dm_sw_linear) + log2_blk_height = log2_blk256_height + + dml_ceil((double) (log2_blk_bytes - 8) / 2.0, 1); + else + log2_blk_height = 0; // blk height of 1 + + log2_blk_width = log2_blk_bytes - log2_bytes_per_element - log2_blk_height; + + if (!surf_vert) { + rq_dlg_param->swath_width_ub = dml_round_to_multiple(vp_width - 1, blk256_width, 1) + + blk256_width; + rq_dlg_param->req_per_swath_ub = rq_dlg_param->swath_width_ub >> log2_blk256_width; + } else { + rq_dlg_param->swath_width_ub = dml_round_to_multiple(vp_height - 1, blk256_height, 1) + + blk256_height; + rq_dlg_param->req_per_swath_ub = rq_dlg_param->swath_width_ub >> log2_blk256_height; + } + + if (!surf_vert) + rq_misc_param->full_swath_bytes = rq_dlg_param->swath_width_ub * blk256_height + * bytes_per_element; + else + rq_misc_param->full_swath_bytes = rq_dlg_param->swath_width_ub * blk256_width + * bytes_per_element; + + rq_misc_param->blk256_height = blk256_height; + rq_misc_param->blk256_width = blk256_width; + + // ------- + // meta + // ------- + log2_meta_req_bytes = 6; // meta request is 64b and is 8x8byte meta element + + // each 64b meta request for dcn is 8x8 meta elements and + // a meta element covers one 256b block of the the data surface. + log2_meta_req_height = log2_blk256_height + 3; // meta req is 8x8 byte, each byte represent 1 blk256 + log2_meta_req_width = log2_meta_req_bytes + 8 - log2_bytes_per_element + - log2_meta_req_height; + meta_req_width = 1 << log2_meta_req_width; + meta_req_height = 1 << log2_meta_req_height; + log2_meta_row_height = 0; + meta_row_width_ub = 0; + + // the dimensions of a meta row are meta_row_width x meta_row_height in elements. + // calculate upper bound of the meta_row_width + if (!surf_vert) { + log2_meta_row_height = log2_meta_req_height; + meta_row_width_ub = dml_round_to_multiple(vp_width - 1, meta_req_width, 1) + + meta_req_width; + rq_dlg_param->meta_req_per_row_ub = meta_row_width_ub / meta_req_width; + } else { + log2_meta_row_height = log2_meta_req_width; + meta_row_width_ub = dml_round_to_multiple(vp_height - 1, meta_req_height, 1) + + meta_req_height; + rq_dlg_param->meta_req_per_row_ub = meta_row_width_ub / meta_req_height; + } + rq_dlg_param->meta_bytes_per_row_ub = rq_dlg_param->meta_req_per_row_ub * 64; + + rq_dlg_param->meta_row_height = 1 << log2_meta_row_height; + + log2_meta_chunk_bytes = dml_log2(rq_sizing_param->meta_chunk_bytes); + log2_meta_chunk_height = log2_meta_row_height; + + //full sized meta chunk width in unit of data elements + log2_meta_chunk_width = log2_meta_chunk_bytes + 8 - log2_bytes_per_element + - log2_meta_chunk_height; + log2_min_meta_chunk_bytes = dml_log2(rq_sizing_param->min_meta_chunk_bytes); + min_meta_chunk_width = 1 + << (log2_min_meta_chunk_bytes + 8 - log2_bytes_per_element + - log2_meta_chunk_height); + meta_chunk_width = 1 << log2_meta_chunk_width; + meta_chunk_per_row_int = (unsigned int) (meta_row_width_ub / meta_chunk_width); + meta_row_remainder = meta_row_width_ub % meta_chunk_width; + meta_chunk_threshold = 0; + meta_blk_bytes = 4096; + meta_blk_height = blk256_height * 64; + meta_blk_width = meta_blk_bytes * 256 / bytes_per_element / meta_blk_height; + meta_surface_bytes = meta_pitch + * (dml_round_to_multiple(vp_height - 1, meta_blk_height, 1) + meta_blk_height) + * bytes_per_element / 256; + vmpg_bytes = mode_lib->soc.vmm_page_size_bytes; + meta_pte_req_per_frame_ub = (dml_round_to_multiple(meta_surface_bytes - vmpg_bytes, + 8 * vmpg_bytes, + 1) + 8 * vmpg_bytes) / (8 * vmpg_bytes); + meta_pte_bytes_per_frame_ub = meta_pte_req_per_frame_ub * 64; //64B mpte request + rq_dlg_param->meta_pte_bytes_per_frame_ub = meta_pte_bytes_per_frame_ub; + + dml_print("DML_DLG: %s: meta_blk_height = %d\n", __func__, meta_blk_height); + dml_print("DML_DLG: %s: meta_blk_width = %d\n", __func__, meta_blk_width); + dml_print("DML_DLG: %s: meta_surface_bytes = %d\n", __func__, meta_surface_bytes); + dml_print("DML_DLG: %s: meta_pte_req_per_frame_ub = %d\n", + __func__, + meta_pte_req_per_frame_ub); + dml_print("DML_DLG: %s: meta_pte_bytes_per_frame_ub = %d\n", + __func__, + meta_pte_bytes_per_frame_ub); + + if (!surf_vert) + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_width; + else + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_height; + + if (meta_row_remainder <= meta_chunk_threshold) + rq_dlg_param->meta_chunks_per_row_ub = meta_chunk_per_row_int + 1; + else + rq_dlg_param->meta_chunks_per_row_ub = meta_chunk_per_row_int + 2; + + // ------ + // dpte + // ------ + if (surf_linear) { + log2_vmpg_height = 0; // one line high + } else { + log2_vmpg_height = (log2_vmpg_bytes - 8) / 2 + log2_blk256_height; + } + log2_vmpg_width = log2_vmpg_bytes - log2_bytes_per_element - log2_vmpg_height; + + // only 3 possible shapes for dpte request in dimensions of ptes: 8x1, 4x2, 2x4. + if (surf_linear) { //one 64B PTE request returns 8 PTEs + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_vmpg_width + 3; + log2_dpte_req_height = 0; + } else if (log2_blk_bytes == 12) { //4KB tile means 4kB page size + //one 64B req gives 8x1 PTEs for 4KB tile + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_blk_width + 3; + log2_dpte_req_height = log2_blk_height + 0; + } else if ((log2_blk_bytes >= 16) && (log2_vmpg_bytes == 12)) { // tile block >= 64KB + //two 64B reqs of 2x4 PTEs give 16 PTEs to cover 64KB + log2_dpte_req_height_ptes = 4; + log2_dpte_req_width = log2_blk256_width + 4; // log2_64KB_width + log2_dpte_req_height = log2_blk256_height + 4; // log2_64KB_height + } else { //64KB page size and must 64KB tile block + //one 64B req gives 8x1 PTEs for 64KB tile + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_blk_width + 3; + log2_dpte_req_height = log2_blk_height + 0; + } + + // The dpte request dimensions in data elements is dpte_req_width x dpte_req_height + // log2_vmpg_width is how much 1 pte represent, now calculating how much a 64b pte req represent + // That depends on the pte shape (i.e. 8x1, 4x2, 2x4) + //log2_dpte_req_height = log2_vmpg_height + log2_dpte_req_height_ptes; + //log2_dpte_req_width = log2_vmpg_width + log2_dpte_req_width_ptes; + dpte_req_height = 1 << log2_dpte_req_height; + dpte_req_width = 1 << log2_dpte_req_width; + + // calculate pitch dpte row buffer can hold + // round the result down to a power of two. + pde_buf_entries = yuv420 ? (pde_proc_buffer_size_64k_reqs >> 1) : pde_proc_buffer_size_64k_reqs; + if (surf_linear) { + unsigned int dpte_row_height; + + log2_dpte_row_height_linear = dml_floor(dml_log2(dml_min(64 * 1024 * pde_buf_entries + / bytes_per_element, + dpte_buf_in_pte_reqs + * dpte_req_width) + / data_pitch), + 1); + + ASSERT(log2_dpte_row_height_linear >= 3); + + if (log2_dpte_row_height_linear > 7) + log2_dpte_row_height_linear = 7; + + log2_dpte_row_height = log2_dpte_row_height_linear; + // For linear, the dpte row is pitch dependent and the pte requests wrap at the pitch boundary. + // the dpte_row_width_ub is the upper bound of data_pitch*dpte_row_height in elements with this unique buffering. + dpte_row_height = 1 << log2_dpte_row_height; + dpte_row_width_ub = dml_round_to_multiple(data_pitch * dpte_row_height - 1, + dpte_req_width, + 1) + dpte_req_width; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_width; + } else { + // the upper bound of the dpte_row_width without dependency on viewport position follows. + // for tiled mode, row height is the same as req height and row store up to vp size upper bound + if (!surf_vert) { + log2_dpte_row_height = log2_dpte_req_height; + dpte_row_width_ub = dml_round_to_multiple(vp_width - 1, dpte_req_width, 1) + + dpte_req_width; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_width; + } else { + log2_dpte_row_height = + (log2_blk_width < log2_dpte_req_width) ? + log2_blk_width : log2_dpte_req_width; + dpte_row_width_ub = dml_round_to_multiple(vp_height - 1, dpte_req_height, 1) + + dpte_req_height; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_height; + } + } + if (log2_blk_bytes >= 16 && log2_vmpg_bytes == 12) // tile block >= 64KB + rq_dlg_param->dpte_bytes_per_row_ub = rq_dlg_param->dpte_req_per_row_ub * 128; //2*64B dpte request + else + rq_dlg_param->dpte_bytes_per_row_ub = rq_dlg_param->dpte_req_per_row_ub * 64; //64B dpte request + + rq_dlg_param->dpte_row_height = 1 << log2_dpte_row_height; + + // the dpte_group_bytes is reduced for the specific case of vertical + // access of a tile surface that has dpte request of 8x1 ptes. + if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + rq_sizing_param->dpte_group_bytes = 512; + else + //full size + rq_sizing_param->dpte_group_bytes = 2048; + + //since pte request size is 64byte, the number of data pte requests per full sized group is as follows. + log2_dpte_group_bytes = dml_log2(rq_sizing_param->dpte_group_bytes); + log2_dpte_group_length = log2_dpte_group_bytes - 6; //length in 64b requests + + // full sized data pte group width in elements + if (!surf_vert) + log2_dpte_group_width = log2_dpte_group_length + log2_dpte_req_width; + else + log2_dpte_group_width = log2_dpte_group_length + log2_dpte_req_height; + + //But if the tile block >=64KB and the page size is 4KB, then each dPTE request is 2*64B + if ((log2_blk_bytes >= 16) && (log2_vmpg_bytes == 12)) // tile block >= 64KB + log2_dpte_group_width = log2_dpte_group_width - 1; + + dpte_group_width = 1 << log2_dpte_group_width; + + // since dpte groups are only aligned to dpte_req_width and not dpte_group_width, + // the upper bound for the dpte groups per row is as follows. + rq_dlg_param->dpte_groups_per_row_ub = dml_ceil((double) dpte_row_width_ub / dpte_group_width, + 1); +} + +static void get_surf_rq_param(struct display_mode_lib *mode_lib, + display_data_rq_sizing_params_st *rq_sizing_param, + display_data_rq_dlg_params_st *rq_dlg_param, + display_data_rq_misc_params_st *rq_misc_param, + const display_pipe_source_params_st pipe_src_param, + bool is_chroma) +{ + bool mode_422 = false; + unsigned int vp_width = 0; + unsigned int vp_height = 0; + unsigned int data_pitch = 0; + unsigned int meta_pitch = 0; + unsigned int ppe = mode_422 ? 2 : 1; + + // TODO check if ppe apply for both luma and chroma in 422 case + if (is_chroma) { + vp_width = pipe_src_param.viewport_width_c / ppe; + vp_height = pipe_src_param.viewport_height_c; + data_pitch = pipe_src_param.data_pitch_c; + meta_pitch = pipe_src_param.meta_pitch_c; + } else { + vp_width = pipe_src_param.viewport_width / ppe; + vp_height = pipe_src_param.viewport_height; + data_pitch = pipe_src_param.data_pitch; + meta_pitch = pipe_src_param.meta_pitch; + } + + rq_sizing_param->chunk_bytes = 8192; + + if (rq_sizing_param->chunk_bytes == 64 * 1024) + rq_sizing_param->min_chunk_bytes = 0; + else + rq_sizing_param->min_chunk_bytes = 1024; + + rq_sizing_param->meta_chunk_bytes = 2048; + rq_sizing_param->min_meta_chunk_bytes = 256; + + rq_sizing_param->mpte_group_bytes = 2048; + + get_meta_and_pte_attr(mode_lib, + rq_dlg_param, + rq_misc_param, + rq_sizing_param, + vp_width, + vp_height, + data_pitch, + meta_pitch, + pipe_src_param.source_format, + pipe_src_param.sw_mode, + pipe_src_param.macro_tile_size, + pipe_src_param.source_scan, + is_chroma); +} + +static void dml20v2_rq_dlg_get_rq_params(struct display_mode_lib *mode_lib, + display_rq_params_st *rq_param, + const display_pipe_source_params_st pipe_src_param) +{ + // get param for luma surface + rq_param->yuv420 = pipe_src_param.source_format == dm_420_8 + || pipe_src_param.source_format == dm_420_10; + rq_param->yuv420_10bpc = pipe_src_param.source_format == dm_420_10; + + get_surf_rq_param(mode_lib, + &(rq_param->sizing.rq_l), + &(rq_param->dlg.rq_l), + &(rq_param->misc.rq_l), + pipe_src_param, + 0); + + if (is_dual_plane((enum source_format_class)(pipe_src_param.source_format))) { + // get param for chroma surface + get_surf_rq_param(mode_lib, + &(rq_param->sizing.rq_c), + &(rq_param->dlg.rq_c), + &(rq_param->misc.rq_c), + pipe_src_param, + 1); + } + + // calculate how to split the det buffer space between luma and chroma + handle_det_buf_split(mode_lib, rq_param, pipe_src_param); + print__rq_params_st(mode_lib, *rq_param); +} + +void dml20v2_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_pipe_params_st pipe_param) +{ + display_rq_params_st rq_param = {0}; + + memset(rq_regs, 0, sizeof(*rq_regs)); + dml20v2_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param.src); + extract_rq_regs(mode_lib, rq_regs, rq_param); + + print__rq_regs_st(mode_lib, *rq_regs); +} + +// Note: currently taken in as is. +// Nice to decouple code from hw register implement and extract code that are repeated for luma and chroma. +static void dml20v2_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib, + const display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + display_dlg_regs_st *disp_dlg_regs, + display_ttu_regs_st *disp_ttu_regs, + const display_rq_dlg_params_st rq_dlg_param, + const display_dlg_sys_params_st dlg_sys_param, + const bool cstate_en, + const bool pstate_en) +{ + const display_pipe_source_params_st *src = &e2e_pipe_param[pipe_idx].pipe.src; + const display_pipe_dest_params_st *dst = &e2e_pipe_param[pipe_idx].pipe.dest; + const display_output_params_st *dout = &e2e_pipe_param[pipe_idx].dout; + const display_clocks_and_cfg_st *clks = &e2e_pipe_param[pipe_idx].clks_cfg; + const scaler_ratio_depth_st *scl = &e2e_pipe_param[pipe_idx].pipe.scale_ratio_depth; + const scaler_taps_st *taps = &e2e_pipe_param[pipe_idx].pipe.scale_taps; + + // ------------------------- + // Section 1.15.2.1: OTG dependent Params + // ------------------------- + // Timing + unsigned int htotal = dst->htotal; +// unsigned int hblank_start = dst.hblank_start; // TODO: Remove + unsigned int hblank_end = dst->hblank_end; + unsigned int vblank_start = dst->vblank_start; + unsigned int vblank_end = dst->vblank_end; + unsigned int min_vblank = mode_lib->ip.min_vblank_lines; + + double dppclk_freq_in_mhz = clks->dppclk_mhz; + double dispclk_freq_in_mhz = clks->dispclk_mhz; + double refclk_freq_in_mhz = clks->refclk_mhz; + double pclk_freq_in_mhz = dst->pixel_rate_mhz; + bool interlaced = dst->interlaced; + + double ref_freq_to_pix_freq = refclk_freq_in_mhz / pclk_freq_in_mhz; + + double min_dcfclk_mhz; + double t_calc_us; + double min_ttu_vblank; + + double min_dst_y_ttu_vblank; + unsigned int dlg_vblank_start; + bool dual_plane; + bool mode_422; + unsigned int access_dir; + unsigned int vp_height_l; + unsigned int vp_width_l; + unsigned int vp_height_c; + unsigned int vp_width_c; + + // Scaling + unsigned int htaps_l; + unsigned int htaps_c; + double hratio_l; + double hratio_c; + double vratio_l; + double vratio_c; + bool scl_enable; + + double line_time_in_us; + // double vinit_l; + // double vinit_c; + // double vinit_bot_l; + // double vinit_bot_c; + + // unsigned int swath_height_l; + unsigned int swath_width_ub_l; + // unsigned int dpte_bytes_per_row_ub_l; + unsigned int dpte_groups_per_row_ub_l; + // unsigned int meta_pte_bytes_per_frame_ub_l; + // unsigned int meta_bytes_per_row_ub_l; + + // unsigned int swath_height_c; + unsigned int swath_width_ub_c; + // unsigned int dpte_bytes_per_row_ub_c; + unsigned int dpte_groups_per_row_ub_c; + + unsigned int meta_chunks_per_row_ub_l; + unsigned int meta_chunks_per_row_ub_c; + unsigned int vupdate_offset; + unsigned int vupdate_width; + unsigned int vready_offset; + + unsigned int dppclk_delay_subtotal; + unsigned int dispclk_delay_subtotal; + unsigned int pixel_rate_delay_subtotal; + + unsigned int vstartup_start; + unsigned int dst_x_after_scaler; + unsigned int dst_y_after_scaler; + double line_wait; + double dst_y_prefetch; + double dst_y_per_vm_vblank; + double dst_y_per_row_vblank; + double dst_y_per_vm_flip; + double dst_y_per_row_flip; + double min_dst_y_per_vm_vblank; + double min_dst_y_per_row_vblank; + double lsw; + double vratio_pre_l; + double vratio_pre_c; + unsigned int req_per_swath_ub_l; + unsigned int req_per_swath_ub_c; + unsigned int meta_row_height_l; + unsigned int meta_row_height_c; + unsigned int swath_width_pixels_ub_l; + unsigned int swath_width_pixels_ub_c; + unsigned int scaler_rec_in_width_l; + unsigned int scaler_rec_in_width_c; + unsigned int dpte_row_height_l; + unsigned int dpte_row_height_c; + double hscale_pixel_rate_l; + double hscale_pixel_rate_c; + double min_hratio_fact_l; + double min_hratio_fact_c; + double refcyc_per_line_delivery_pre_l; + double refcyc_per_line_delivery_pre_c; + double refcyc_per_line_delivery_l; + double refcyc_per_line_delivery_c; + + double refcyc_per_req_delivery_pre_l; + double refcyc_per_req_delivery_pre_c; + double refcyc_per_req_delivery_l; + double refcyc_per_req_delivery_c; + + unsigned int full_recout_width; + double xfc_transfer_delay; + double xfc_precharge_delay; + double xfc_remote_surface_flip_latency; + double xfc_dst_y_delta_drq_limit; + double xfc_prefetch_margin; + double refcyc_per_req_delivery_pre_cur0; + double refcyc_per_req_delivery_cur0; + double refcyc_per_req_delivery_pre_cur1; + double refcyc_per_req_delivery_cur1; + + memset(disp_dlg_regs, 0, sizeof(*disp_dlg_regs)); + memset(disp_ttu_regs, 0, sizeof(*disp_ttu_regs)); + + dml_print("DML_DLG: %s: cstate_en = %d\n", __func__, cstate_en); + dml_print("DML_DLG: %s: pstate_en = %d\n", __func__, pstate_en); + + dml_print("DML_DLG: %s: dppclk_freq_in_mhz = %3.2f\n", __func__, dppclk_freq_in_mhz); + dml_print("DML_DLG: %s: dispclk_freq_in_mhz = %3.2f\n", __func__, dispclk_freq_in_mhz); + dml_print("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, refclk_freq_in_mhz); + dml_print("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, pclk_freq_in_mhz); + dml_print("DML_DLG: %s: interlaced = %d\n", __func__, interlaced); + ASSERT(ref_freq_to_pix_freq < 4.0); + + disp_dlg_regs->ref_freq_to_pix_freq = + (unsigned int) (ref_freq_to_pix_freq * dml_pow(2, 19)); + disp_dlg_regs->refcyc_per_htotal = (unsigned int) (ref_freq_to_pix_freq * (double) htotal + * dml_pow(2, 8)); + disp_dlg_regs->dlg_vblank_end = interlaced ? (vblank_end / 2) : vblank_end; // 15 bits + disp_dlg_regs->refcyc_h_blank_end = (unsigned int) ((double) hblank_end + * (double) ref_freq_to_pix_freq); + ASSERT(disp_dlg_regs->refcyc_h_blank_end < (unsigned int) dml_pow(2, 13)); + + min_dcfclk_mhz = dlg_sys_param.deepsleep_dcfclk_mhz; + t_calc_us = get_tcalc(mode_lib, e2e_pipe_param, num_pipes); + min_ttu_vblank = get_min_ttu_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + min_dst_y_ttu_vblank = min_ttu_vblank * pclk_freq_in_mhz / (double) htotal; + dlg_vblank_start = interlaced ? (vblank_start / 2) : vblank_start; + + disp_dlg_regs->min_dst_y_next_start = (unsigned int) (((double) dlg_vblank_start + + min_dst_y_ttu_vblank) * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int) dml_pow(2, 18)); + + dml_print("DML_DLG: %s: min_dcfclk_mhz = %3.2f\n", + __func__, + min_dcfclk_mhz); + dml_print("DML_DLG: %s: min_ttu_vblank = %3.2f\n", + __func__, + min_ttu_vblank); + dml_print("DML_DLG: %s: min_dst_y_ttu_vblank = %3.2f\n", + __func__, + min_dst_y_ttu_vblank); + dml_print("DML_DLG: %s: t_calc_us = %3.2f\n", + __func__, + t_calc_us); + dml_print("DML_DLG: %s: disp_dlg_regs->min_dst_y_next_start = 0x%0x\n", + __func__, + disp_dlg_regs->min_dst_y_next_start); + dml_print("DML_DLG: %s: ref_freq_to_pix_freq = %3.2f\n", + __func__, + ref_freq_to_pix_freq); + + // ------------------------- + // Section 1.15.2.2: Prefetch, Active and TTU + // ------------------------- + // Prefetch Calc + // Source +// dcc_en = src.dcc; + dual_plane = is_dual_plane((enum source_format_class)(src->source_format)); + mode_422 = false; // TODO + access_dir = (src->source_scan == dm_vert); // vp access direction: horizontal or vertical accessed +// bytes_per_element_l = get_bytes_per_element(source_format_class(src.source_format), 0); +// bytes_per_element_c = get_bytes_per_element(source_format_class(src.source_format), 1); + vp_height_l = src->viewport_height; + vp_width_l = src->viewport_width; + vp_height_c = src->viewport_height_c; + vp_width_c = src->viewport_width_c; + + // Scaling + htaps_l = taps->htaps; + htaps_c = taps->htaps_c; + hratio_l = scl->hscl_ratio; + hratio_c = scl->hscl_ratio_c; + vratio_l = scl->vscl_ratio; + vratio_c = scl->vscl_ratio_c; + scl_enable = scl->scl_enable; + + line_time_in_us = (htotal / pclk_freq_in_mhz); +// vinit_l = scl.vinit; +// vinit_c = scl.vinit_c; +// vinit_bot_l = scl.vinit_bot; +// vinit_bot_c = scl.vinit_bot_c; + +// unsigned int swath_height_l = rq_dlg_param.rq_l.swath_height; + swath_width_ub_l = rq_dlg_param.rq_l.swath_width_ub; +// unsigned int dpte_bytes_per_row_ub_l = rq_dlg_param.rq_l.dpte_bytes_per_row_ub; + dpte_groups_per_row_ub_l = rq_dlg_param.rq_l.dpte_groups_per_row_ub; +// unsigned int meta_pte_bytes_per_frame_ub_l = rq_dlg_param.rq_l.meta_pte_bytes_per_frame_ub; +// unsigned int meta_bytes_per_row_ub_l = rq_dlg_param.rq_l.meta_bytes_per_row_ub; + +// unsigned int swath_height_c = rq_dlg_param.rq_c.swath_height; + swath_width_ub_c = rq_dlg_param.rq_c.swath_width_ub; + // dpte_bytes_per_row_ub_c = rq_dlg_param.rq_c.dpte_bytes_per_row_ub; + dpte_groups_per_row_ub_c = rq_dlg_param.rq_c.dpte_groups_per_row_ub; + + meta_chunks_per_row_ub_l = rq_dlg_param.rq_l.meta_chunks_per_row_ub; + meta_chunks_per_row_ub_c = rq_dlg_param.rq_c.meta_chunks_per_row_ub; + vupdate_offset = dst->vupdate_offset; + vupdate_width = dst->vupdate_width; + vready_offset = dst->vready_offset; + + dppclk_delay_subtotal = mode_lib->ip.dppclk_delay_subtotal; + dispclk_delay_subtotal = mode_lib->ip.dispclk_delay_subtotal; + + if (scl_enable) + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_scl; + else + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_scl_lb_only; + + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_cnvc_formatter + + src->num_cursors * mode_lib->ip.dppclk_delay_cnvc_cursor; + + if (dout->dsc_enable) { + double dsc_delay = get_dsc_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dispclk_delay_subtotal += dsc_delay; + } + + pixel_rate_delay_subtotal = dppclk_delay_subtotal * pclk_freq_in_mhz / dppclk_freq_in_mhz + + dispclk_delay_subtotal * pclk_freq_in_mhz / dispclk_freq_in_mhz; + + vstartup_start = dst->vstartup_start; + if (interlaced) { + if (vstartup_start / 2.0 + - (double) (vready_offset + vupdate_width + vupdate_offset) / htotal + <= vblank_end / 2.0) + disp_dlg_regs->vready_after_vcount0 = 1; + else + disp_dlg_regs->vready_after_vcount0 = 0; + } else { + if (vstartup_start + - (double) (vready_offset + vupdate_width + vupdate_offset) / htotal + <= vblank_end) + disp_dlg_regs->vready_after_vcount0 = 1; + else + disp_dlg_regs->vready_after_vcount0 = 0; + } + + // TODO: Where is this coming from? + if (interlaced) + vstartup_start = vstartup_start / 2; + + // TODO: What if this min_vblank doesn't match the value in the dml_config_settings.cpp? + if (vstartup_start >= min_vblank) { + dml_print("WARNING: DML_DLG: %s: vblank_start=%d vblank_end=%d\n", + __func__, + vblank_start, + vblank_end); + dml_print("WARNING: DML_DLG: %s: vstartup_start=%d should be less than min_vblank=%d\n", + __func__, + vstartup_start, + min_vblank); + min_vblank = vstartup_start + 1; + dml_print("WARNING: DML_DLG: %s: vstartup_start=%d should be less than min_vblank=%d\n", + __func__, + vstartup_start, + min_vblank); + } + + dst_x_after_scaler = get_dst_x_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dst_y_after_scaler = get_dst_y_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dml_print("DML_DLG: %s: htotal = %d\n", __func__, htotal); + dml_print("DML_DLG: %s: pixel_rate_delay_subtotal = %d\n", + __func__, + pixel_rate_delay_subtotal); + dml_print("DML_DLG: %s: dst_x_after_scaler = %d\n", + __func__, + dst_x_after_scaler); + dml_print("DML_DLG: %s: dst_y_after_scaler = %d\n", + __func__, + dst_y_after_scaler); + + // Lwait + line_wait = mode_lib->soc.urgent_latency_us; + if (cstate_en) + line_wait = dml_max(mode_lib->soc.sr_enter_plus_exit_time_us, line_wait); + if (pstate_en) + line_wait = dml_max(mode_lib->soc.dram_clock_change_latency_us + + mode_lib->soc.urgent_latency_us, + line_wait); + line_wait = line_wait / line_time_in_us; + + dst_y_prefetch = get_dst_y_prefetch(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dml_print("DML_DLG: %s: dst_y_prefetch (after rnd) = %3.2f\n", __func__, dst_y_prefetch); + + dst_y_per_vm_vblank = get_dst_y_per_vm_vblank(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + dst_y_per_row_vblank = get_dst_y_per_row_vblank(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + dst_y_per_vm_flip = get_dst_y_per_vm_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dst_y_per_row_flip = get_dst_y_per_row_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + min_dst_y_per_vm_vblank = 8.0; + min_dst_y_per_row_vblank = 16.0; + + // magic! + if (htotal <= 75) { + min_vblank = 300; + min_dst_y_per_vm_vblank = 100.0; + min_dst_y_per_row_vblank = 100.0; + } + + dml_print("DML_DLG: %s: dst_y_per_vm_vblank = %3.2f\n", __func__, dst_y_per_vm_vblank); + dml_print("DML_DLG: %s: dst_y_per_row_vblank = %3.2f\n", __func__, dst_y_per_row_vblank); + + ASSERT(dst_y_per_vm_vblank < min_dst_y_per_vm_vblank); + ASSERT(dst_y_per_row_vblank < min_dst_y_per_row_vblank); + + ASSERT(dst_y_prefetch > (dst_y_per_vm_vblank + dst_y_per_row_vblank)); + lsw = dst_y_prefetch - (dst_y_per_vm_vblank + dst_y_per_row_vblank); + + dml_print("DML_DLG: %s: lsw = %3.2f\n", __func__, lsw); + + vratio_pre_l = get_vratio_prefetch_l(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + vratio_pre_c = get_vratio_prefetch_c(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dml_print("DML_DLG: %s: vratio_pre_l=%3.2f\n", __func__, vratio_pre_l); + dml_print("DML_DLG: %s: vratio_pre_c=%3.2f\n", __func__, vratio_pre_c); + + // Active + req_per_swath_ub_l = rq_dlg_param.rq_l.req_per_swath_ub; + req_per_swath_ub_c = rq_dlg_param.rq_c.req_per_swath_ub; + meta_row_height_l = rq_dlg_param.rq_l.meta_row_height; + meta_row_height_c = rq_dlg_param.rq_c.meta_row_height; + swath_width_pixels_ub_l = 0; + swath_width_pixels_ub_c = 0; + scaler_rec_in_width_l = 0; + scaler_rec_in_width_c = 0; + dpte_row_height_l = rq_dlg_param.rq_l.dpte_row_height; + dpte_row_height_c = rq_dlg_param.rq_c.dpte_row_height; + + if (mode_422) { + swath_width_pixels_ub_l = swath_width_ub_l * 2; // *2 for 2 pixel per element + swath_width_pixels_ub_c = swath_width_ub_c * 2; + } else { + swath_width_pixels_ub_l = swath_width_ub_l * 1; + swath_width_pixels_ub_c = swath_width_ub_c * 1; + } + + hscale_pixel_rate_l = 0.; + hscale_pixel_rate_c = 0.; + min_hratio_fact_l = 1.0; + min_hratio_fact_c = 1.0; + + if (htaps_l <= 1) + min_hratio_fact_l = 2.0; + else if (htaps_l <= 6) { + if ((hratio_l * 2.0) > 4.0) + min_hratio_fact_l = 4.0; + else + min_hratio_fact_l = hratio_l * 2.0; + } else { + if (hratio_l > 4.0) + min_hratio_fact_l = 4.0; + else + min_hratio_fact_l = hratio_l; + } + + hscale_pixel_rate_l = min_hratio_fact_l * dppclk_freq_in_mhz; + + if (htaps_c <= 1) + min_hratio_fact_c = 2.0; + else if (htaps_c <= 6) { + if ((hratio_c * 2.0) > 4.0) + min_hratio_fact_c = 4.0; + else + min_hratio_fact_c = hratio_c * 2.0; + } else { + if (hratio_c > 4.0) + min_hratio_fact_c = 4.0; + else + min_hratio_fact_c = hratio_c; + } + + hscale_pixel_rate_c = min_hratio_fact_c * dppclk_freq_in_mhz; + + refcyc_per_line_delivery_pre_l = 0.; + refcyc_per_line_delivery_pre_c = 0.; + refcyc_per_line_delivery_l = 0.; + refcyc_per_line_delivery_c = 0.; + + refcyc_per_req_delivery_pre_l = 0.; + refcyc_per_req_delivery_pre_c = 0.; + refcyc_per_req_delivery_l = 0.; + refcyc_per_req_delivery_c = 0.; + + full_recout_width = 0; + // In ODM + if (src->is_hsplit) { + // This "hack" is only allowed (and valid) for MPC combine. In ODM + // combine, you MUST specify the full_recout_width...according to Oswin + if (dst->full_recout_width == 0 && !dst->odm_combine) { + dml_print("DML_DLG: %s: Warning: full_recout_width not set in hsplit mode\n", + __func__); + full_recout_width = dst->recout_width * 2; // assume half split for dcn1 + } else + full_recout_width = dst->full_recout_width; + } else + full_recout_width = dst->recout_width; + + // As of DCN2, mpc_combine and odm_combine are mutually exclusive + refcyc_per_line_delivery_pre_l = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_l, + hscale_pixel_rate_l, + swath_width_pixels_ub_l, + 1); // per line + + refcyc_per_line_delivery_l = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_l, + hscale_pixel_rate_l, + swath_width_pixels_ub_l, + 1); // per line + + dml_print("DML_DLG: %s: full_recout_width = %d\n", + __func__, + full_recout_width); + dml_print("DML_DLG: %s: hscale_pixel_rate_l = %3.2f\n", + __func__, + hscale_pixel_rate_l); + dml_print("DML_DLG: %s: refcyc_per_line_delivery_pre_l = %3.2f\n", + __func__, + refcyc_per_line_delivery_pre_l); + dml_print("DML_DLG: %s: refcyc_per_line_delivery_l = %3.2f\n", + __func__, + refcyc_per_line_delivery_l); + + if (dual_plane) { + refcyc_per_line_delivery_pre_c = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_c, + hscale_pixel_rate_c, + swath_width_pixels_ub_c, + 1); // per line + + refcyc_per_line_delivery_c = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_c, + hscale_pixel_rate_c, + swath_width_pixels_ub_c, + 1); // per line + + dml_print("DML_DLG: %s: refcyc_per_line_delivery_pre_c = %3.2f\n", + __func__, + refcyc_per_line_delivery_pre_c); + dml_print("DML_DLG: %s: refcyc_per_line_delivery_c = %3.2f\n", + __func__, + refcyc_per_line_delivery_c); + } + + // TTU - Luma / Chroma + if (access_dir) { // vertical access + scaler_rec_in_width_l = vp_height_l; + scaler_rec_in_width_c = vp_height_c; + } else { + scaler_rec_in_width_l = vp_width_l; + scaler_rec_in_width_c = vp_width_c; + } + + refcyc_per_req_delivery_pre_l = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_l, + hscale_pixel_rate_l, + scaler_rec_in_width_l, + req_per_swath_ub_l); // per req + refcyc_per_req_delivery_l = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_l, + hscale_pixel_rate_l, + scaler_rec_in_width_l, + req_per_swath_ub_l); // per req + + dml_print("DML_DLG: %s: refcyc_per_req_delivery_pre_l = %3.2f\n", + __func__, + refcyc_per_req_delivery_pre_l); + dml_print("DML_DLG: %s: refcyc_per_req_delivery_l = %3.2f\n", + __func__, + refcyc_per_req_delivery_l); + + ASSERT(refcyc_per_req_delivery_pre_l < dml_pow(2, 13)); + ASSERT(refcyc_per_req_delivery_l < dml_pow(2, 13)); + + if (dual_plane) { + refcyc_per_req_delivery_pre_c = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_c, + hscale_pixel_rate_c, + scaler_rec_in_width_c, + req_per_swath_ub_c); // per req + refcyc_per_req_delivery_c = get_refcyc_per_delivery(mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_c, + hscale_pixel_rate_c, + scaler_rec_in_width_c, + req_per_swath_ub_c); // per req + + dml_print("DML_DLG: %s: refcyc_per_req_delivery_pre_c = %3.2f\n", + __func__, + refcyc_per_req_delivery_pre_c); + dml_print("DML_DLG: %s: refcyc_per_req_delivery_c = %3.2f\n", + __func__, + refcyc_per_req_delivery_c); + + ASSERT(refcyc_per_req_delivery_pre_c < dml_pow(2, 13)); + ASSERT(refcyc_per_req_delivery_c < dml_pow(2, 13)); + } + + // XFC + xfc_transfer_delay = get_xfc_transfer_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + xfc_precharge_delay = get_xfc_precharge_delay(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + xfc_remote_surface_flip_latency = get_xfc_remote_surface_flip_latency(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + xfc_dst_y_delta_drq_limit = xfc_remote_surface_flip_latency; + xfc_prefetch_margin = get_xfc_prefetch_margin(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + + // TTU - Cursor + refcyc_per_req_delivery_pre_cur0 = 0.0; + refcyc_per_req_delivery_cur0 = 0.0; + if (src->num_cursors > 0) { + calculate_ttu_cursor(mode_lib, + &refcyc_per_req_delivery_pre_cur0, + &refcyc_per_req_delivery_cur0, + refclk_freq_in_mhz, + ref_freq_to_pix_freq, + hscale_pixel_rate_l, + scl->hscl_ratio, + vratio_pre_l, + vratio_l, + src->cur0_src_width, + (enum cursor_bpp)(src->cur0_bpp)); + } + + refcyc_per_req_delivery_pre_cur1 = 0.0; + refcyc_per_req_delivery_cur1 = 0.0; + if (src->num_cursors > 1) { + calculate_ttu_cursor(mode_lib, + &refcyc_per_req_delivery_pre_cur1, + &refcyc_per_req_delivery_cur1, + refclk_freq_in_mhz, + ref_freq_to_pix_freq, + hscale_pixel_rate_l, + scl->hscl_ratio, + vratio_pre_l, + vratio_l, + src->cur1_src_width, + (enum cursor_bpp)(src->cur1_bpp)); + } + + // TTU - Misc + // all hard-coded + + // Assignment to register structures + disp_dlg_regs->dst_y_after_scaler = dst_y_after_scaler; // in terms of line + disp_dlg_regs->refcyc_x_after_scaler = dst_x_after_scaler * ref_freq_to_pix_freq; // in terms of refclk + ASSERT(disp_dlg_regs->refcyc_x_after_scaler < (unsigned int) dml_pow(2, 13)); + disp_dlg_regs->dst_y_prefetch = (unsigned int) (dst_y_prefetch * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_vm_vblank = (unsigned int) (dst_y_per_vm_vblank * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_row_vblank = (unsigned int) (dst_y_per_row_vblank * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_vm_flip = (unsigned int) (dst_y_per_vm_flip * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_row_flip = (unsigned int) (dst_y_per_row_flip * dml_pow(2, 2)); + + disp_dlg_regs->vratio_prefetch = (unsigned int) (vratio_pre_l * dml_pow(2, 19)); + disp_dlg_regs->vratio_prefetch_c = (unsigned int) (vratio_pre_c * dml_pow(2, 19)); + + disp_dlg_regs->refcyc_per_pte_group_vblank_l = + (unsigned int) (dst_y_per_row_vblank * (double) htotal + * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_l); + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int) dml_pow(2, 13)); + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_vblank_c = (unsigned int) (dst_y_per_row_vblank + * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_c); + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_c + < (unsigned int) dml_pow(2, 13)); + } + + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l = + (unsigned int) (dst_y_per_row_vblank * (double) htotal + * ref_freq_to_pix_freq / (double) meta_chunks_per_row_ub_l); + ASSERT(disp_dlg_regs->refcyc_per_meta_chunk_vblank_l < (unsigned int) dml_pow(2, 13)); + + disp_dlg_regs->refcyc_per_meta_chunk_vblank_c = + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l; // dcc for 4:2:0 is not supported in dcn1.0. assigned to be the same as _l for now + + disp_dlg_regs->refcyc_per_pte_group_flip_l = (unsigned int) (dst_y_per_row_flip * htotal + * ref_freq_to_pix_freq) / dpte_groups_per_row_ub_l; + disp_dlg_regs->refcyc_per_meta_chunk_flip_l = (unsigned int) (dst_y_per_row_flip * htotal + * ref_freq_to_pix_freq) / meta_chunks_per_row_ub_l; + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_flip_c = (unsigned int) (dst_y_per_row_flip + * htotal * ref_freq_to_pix_freq) / dpte_groups_per_row_ub_c; + disp_dlg_regs->refcyc_per_meta_chunk_flip_c = (unsigned int) (dst_y_per_row_flip + * htotal * ref_freq_to_pix_freq) / meta_chunks_per_row_ub_c; + } + + disp_dlg_regs->dst_y_per_pte_row_nom_l = (unsigned int) ((double) dpte_row_height_l + / (double) vratio_l * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->dst_y_per_pte_row_nom_l < (unsigned int) dml_pow(2, 17)); + + if (dual_plane) { + disp_dlg_regs->dst_y_per_pte_row_nom_c = (unsigned int) ((double) dpte_row_height_c + / (double) vratio_c * dml_pow(2, 2)); + if (disp_dlg_regs->dst_y_per_pte_row_nom_c >= (unsigned int) dml_pow(2, 17)) { + dml_print("DML_DLG: %s: Warning dst_y_per_pte_row_nom_c %u larger than supported by register format U15.2 %u\n", + __func__, + disp_dlg_regs->dst_y_per_pte_row_nom_c, + (unsigned int) dml_pow(2, 17) - 1); + } + } + + disp_dlg_regs->dst_y_per_meta_row_nom_l = (unsigned int) ((double) meta_row_height_l + / (double) vratio_l * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->dst_y_per_meta_row_nom_l < (unsigned int) dml_pow(2, 17)); + + disp_dlg_regs->dst_y_per_meta_row_nom_c = disp_dlg_regs->dst_y_per_meta_row_nom_l; // TODO: dcc for 4:2:0 is not supported in dcn1.0. assigned to be the same as _l for now + + disp_dlg_regs->refcyc_per_pte_group_nom_l = (unsigned int) ((double) dpte_row_height_l + / (double) vratio_l * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_l); + if (disp_dlg_regs->refcyc_per_pte_group_nom_l >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_pte_group_nom_l = dml_pow(2, 23) - 1; + disp_dlg_regs->refcyc_per_meta_chunk_nom_l = (unsigned int) ((double) meta_row_height_l + / (double) vratio_l * (double) htotal * ref_freq_to_pix_freq + / (double) meta_chunks_per_row_ub_l); + if (disp_dlg_regs->refcyc_per_meta_chunk_nom_l >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_meta_chunk_nom_l = dml_pow(2, 23) - 1; + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_nom_c = + (unsigned int) ((double) dpte_row_height_c / (double) vratio_c + * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_c); + if (disp_dlg_regs->refcyc_per_pte_group_nom_c >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_pte_group_nom_c = dml_pow(2, 23) - 1; + + // TODO: Is this the right calculation? Does htotal need to be halved? + disp_dlg_regs->refcyc_per_meta_chunk_nom_c = + (unsigned int) ((double) meta_row_height_c / (double) vratio_c + * (double) htotal * ref_freq_to_pix_freq + / (double) meta_chunks_per_row_ub_c); + if (disp_dlg_regs->refcyc_per_meta_chunk_nom_c >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_meta_chunk_nom_c = dml_pow(2, 23) - 1; + } + + disp_dlg_regs->refcyc_per_line_delivery_pre_l = (unsigned int) dml_floor(refcyc_per_line_delivery_pre_l, + 1); + disp_dlg_regs->refcyc_per_line_delivery_l = (unsigned int) dml_floor(refcyc_per_line_delivery_l, + 1); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_l < (unsigned int) dml_pow(2, 13)); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_l < (unsigned int) dml_pow(2, 13)); + + disp_dlg_regs->refcyc_per_line_delivery_pre_c = (unsigned int) dml_floor(refcyc_per_line_delivery_pre_c, + 1); + disp_dlg_regs->refcyc_per_line_delivery_c = (unsigned int) dml_floor(refcyc_per_line_delivery_c, + 1); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_c < (unsigned int) dml_pow(2, 13)); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_c < (unsigned int) dml_pow(2, 13)); + + disp_dlg_regs->chunk_hdl_adjust_cur0 = 3; + disp_dlg_regs->dst_y_offset_cur0 = 0; + disp_dlg_regs->chunk_hdl_adjust_cur1 = 3; + disp_dlg_regs->dst_y_offset_cur1 = 0; + + disp_dlg_regs->xfc_reg_transfer_delay = xfc_transfer_delay; + disp_dlg_regs->xfc_reg_precharge_delay = xfc_precharge_delay; + disp_dlg_regs->xfc_reg_remote_surface_flip_latency = xfc_remote_surface_flip_latency; + disp_dlg_regs->xfc_reg_prefetch_margin = dml_ceil(xfc_prefetch_margin * refclk_freq_in_mhz, + 1); + + // slave has to have this value also set to off + if (src->xfc_enable && !src->xfc_slave) + disp_dlg_regs->dst_y_delta_drq_limit = dml_ceil(xfc_dst_y_delta_drq_limit, 1); + else + disp_dlg_regs->dst_y_delta_drq_limit = 0x7fff; // off + + disp_ttu_regs->refcyc_per_req_delivery_pre_l = (unsigned int) (refcyc_per_req_delivery_pre_l + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_l = (unsigned int) (refcyc_per_req_delivery_l + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_c = (unsigned int) (refcyc_per_req_delivery_pre_c + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_c = (unsigned int) (refcyc_per_req_delivery_c + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_cur0 = + (unsigned int) (refcyc_per_req_delivery_pre_cur0 * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_cur0 = (unsigned int) (refcyc_per_req_delivery_cur0 + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_cur1 = + (unsigned int) (refcyc_per_req_delivery_pre_cur1 * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_cur1 = (unsigned int) (refcyc_per_req_delivery_cur1 + * dml_pow(2, 10)); + disp_ttu_regs->qos_level_low_wm = 0; + ASSERT(disp_ttu_regs->qos_level_low_wm < dml_pow(2, 14)); + disp_ttu_regs->qos_level_high_wm = (unsigned int) (4.0 * (double) htotal + * ref_freq_to_pix_freq); + /*ASSERT(disp_ttu_regs->qos_level_high_wm < dml_pow(2, 14));*/ + + disp_ttu_regs->qos_level_flip = 14; + disp_ttu_regs->qos_level_fixed_l = 8; + disp_ttu_regs->qos_level_fixed_c = 8; + disp_ttu_regs->qos_level_fixed_cur0 = 8; + disp_ttu_regs->qos_ramp_disable_l = 0; + disp_ttu_regs->qos_ramp_disable_c = 0; + disp_ttu_regs->qos_ramp_disable_cur0 = 0; + + disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz; + ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24)); + + print__ttu_regs_st(mode_lib, *disp_ttu_regs); + print__dlg_regs_st(mode_lib, *disp_dlg_regs); +} + +void dml20v2_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib, + display_dlg_regs_st *dlg_regs, + display_ttu_regs_st *ttu_regs, + display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + const bool cstate_en, + const bool pstate_en, + const bool vm_en, + const bool ignore_viewport_pos, + const bool immediate_flip_support) +{ + display_rq_params_st rq_param = {0}; + display_dlg_sys_params_st dlg_sys_param = {0}; + + // Get watermark and Tex. + dlg_sys_param.t_urg_wm_us = get_wm_urgent(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.deepsleep_dcfclk_mhz = get_clk_dcf_deepsleep(mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.t_extra_us = get_urgent_extra_latency(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.mem_trip_us = get_wm_memory_trip(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.t_mclk_wm_us = get_wm_dram_clock_change(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.t_sr_wm_us = get_wm_stutter_enter_exit(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.total_flip_bw = get_total_immediate_flip_bw(mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.total_flip_bytes = get_total_immediate_flip_bytes(mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.t_srx_delay_us = mode_lib->ip.dcfclk_cstate_latency + / dlg_sys_param.deepsleep_dcfclk_mhz; // TODO: Deprecated + + print__dlg_sys_params_st(mode_lib, dlg_sys_param); + + // system parameter calculation done + + dml_print("DML_DLG: Calculation for pipe[%d] start\n\n", pipe_idx); + dml20v2_rq_dlg_get_rq_params(mode_lib, &rq_param, e2e_pipe_param[pipe_idx].pipe.src); + dml20v2_rq_dlg_get_dlg_params(mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx, + dlg_regs, + ttu_regs, + rq_param.dlg, + dlg_sys_param, + cstate_en, + pstate_en); + dml_print("DML_DLG: Calculation for pipe[%d] end\n", pipe_idx); +} + +static void calculate_ttu_cursor(struct display_mode_lib *mode_lib, + double *refcyc_per_req_delivery_pre_cur, + double *refcyc_per_req_delivery_cur, + double refclk_freq_in_mhz, + double ref_freq_to_pix_freq, + double hscale_pixel_rate_l, + double hscl_ratio, + double vratio_pre_l, + double vratio_l, + unsigned int cur_width, + enum cursor_bpp cur_bpp) +{ + unsigned int cur_src_width = cur_width; + unsigned int cur_req_size = 0; + unsigned int cur_req_width = 0; + double cur_width_ub = 0.0; + double cur_req_per_width = 0.0; + double hactive_cur = 0.0; + + ASSERT(cur_src_width <= 256); + + *refcyc_per_req_delivery_pre_cur = 0.0; + *refcyc_per_req_delivery_cur = 0.0; + if (cur_src_width > 0) { + unsigned int cur_bit_per_pixel = 0; + + if (cur_bpp == dm_cur_2bit) { + cur_req_size = 64; // byte + cur_bit_per_pixel = 2; + } else { // 32bit + cur_bit_per_pixel = 32; + if (cur_src_width >= 1 && cur_src_width <= 16) + cur_req_size = 64; + else if (cur_src_width >= 17 && cur_src_width <= 31) + cur_req_size = 128; + else + cur_req_size = 256; + } + + cur_req_width = (double) cur_req_size / ((double) cur_bit_per_pixel / 8.0); + cur_width_ub = dml_ceil((double) cur_src_width / (double) cur_req_width, 1) + * (double) cur_req_width; + cur_req_per_width = cur_width_ub / (double) cur_req_width; + hactive_cur = (double) cur_src_width / hscl_ratio; // TODO: oswin to think about what to do for cursor + + if (vratio_pre_l <= 1.0) { + *refcyc_per_req_delivery_pre_cur = hactive_cur * ref_freq_to_pix_freq + / (double) cur_req_per_width; + } else { + *refcyc_per_req_delivery_pre_cur = (double) refclk_freq_in_mhz + * (double) cur_src_width / hscale_pixel_rate_l + / (double) cur_req_per_width; + } + + ASSERT(*refcyc_per_req_delivery_pre_cur < dml_pow(2, 13)); + + if (vratio_l <= 1.0) { + *refcyc_per_req_delivery_cur = hactive_cur * ref_freq_to_pix_freq + / (double) cur_req_per_width; + } else { + *refcyc_per_req_delivery_cur = (double) refclk_freq_in_mhz + * (double) cur_src_width / hscale_pixel_rate_l + / (double) cur_req_per_width; + } + + dml_print("DML_DLG: %s: cur_req_width = %d\n", + __func__, + cur_req_width); + dml_print("DML_DLG: %s: cur_width_ub = %3.2f\n", + __func__, + cur_width_ub); + dml_print("DML_DLG: %s: cur_req_per_width = %3.2f\n", + __func__, + cur_req_per_width); + dml_print("DML_DLG: %s: hactive_cur = %3.2f\n", + __func__, + hactive_cur); + dml_print("DML_DLG: %s: refcyc_per_req_delivery_pre_cur = %3.2f\n", + __func__, + *refcyc_per_req_delivery_pre_cur); + dml_print("DML_DLG: %s: refcyc_per_req_delivery_cur = %3.2f\n", + __func__, + *refcyc_per_req_delivery_cur); + + ASSERT(*refcyc_per_req_delivery_cur < dml_pow(2, 13)); + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h new file mode 100644 index 000000000000..0378406bf7e7 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h @@ -0,0 +1,74 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DML20V2_DISPLAY_RQ_DLG_CALC_H__ +#define __DML20V2_DISPLAY_RQ_DLG_CALC_H__ + +#include "../dml_common_defs.h" +#include "../display_rq_dlg_helpers.h" + +struct display_mode_lib; + + +// Function: dml_rq_dlg_get_rq_reg +// Main entry point for test to get the register values out of this DML class. +// This function calls <get_rq_param> and <extract_rq_regs> fucntions to calculate +// and then populate the rq_regs struct +// Input: +// pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.) +// Output: +// rq_regs - struct that holds all the RQ registers field value. +// See also: <display_rq_regs_st> +void dml20v2_rq_dlg_get_rq_reg( + struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_pipe_params_st pipe_param); + + +// Function: dml_rq_dlg_get_dlg_reg +// Calculate and return DLG and TTU register struct given the system setting +// Output: +// dlg_regs - output DLG register struct +// ttu_regs - output DLG TTU register struct +// Input: +// e2e_pipe_param - "compacted" array of e2e pipe param struct +// num_pipes - num of active "pipe" or "route" +// pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg +// cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered. +// Added for legacy or unrealistic timing tests. +void dml20v2_rq_dlg_get_dlg_reg( + struct display_mode_lib *mode_lib, + display_dlg_regs_st *dlg_regs, + display_ttu_regs_st *ttu_regs, + display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + const bool cstate_en, + const bool pstate_en, + const bool vm_en, + const bool ignore_viewport_pos, + const bool immediate_flip_support); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c new file mode 100644 index 000000000000..e6617c958bb8 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c @@ -0,0 +1,6134 @@ +/* + * Copyright 2017 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + + +#include "../display_mode_lib.h" +#include "../dml_inline_defs.h" +#include "../display_mode_vba.h" +#include "display_mode_vba_21.h" + + +/* + * NOTE: + * This file is gcc-parsable HW gospel, coming straight from HW engineers. + * + * It doesn't adhere to Linux kernel style and sometimes will do things in odd + * ways. Unless there is something clearly wrong with it the code should + * remain as-is as it provides us with a guarantee from HW that it is correct. + */ +typedef struct { + double DPPCLK; + double DISPCLK; + double PixelClock; + double DCFCLKDeepSleep; + unsigned int DPPPerPlane; + bool ScalerEnabled; + enum scan_direction_class SourceScan; + unsigned int BlockWidth256BytesY; + unsigned int BlockHeight256BytesY; + unsigned int BlockWidth256BytesC; + unsigned int BlockHeight256BytesC; + unsigned int InterlaceEnable; + unsigned int NumberOfCursors; + unsigned int VBlank; + unsigned int HTotal; +} Pipe; + +typedef struct { + bool Enable; + unsigned int MaxPageTableLevels; + unsigned int CachedPageTableLevels; +} HostVM; + +#define BPP_INVALID 0 +#define BPP_BLENDED_PIPE 0xffffffff +#define DCN21_MAX_DSC_IMAGE_WIDTH 5184 +#define DCN21_MAX_420_IMAGE_WIDTH 4096 + +static void DisplayPipeConfiguration(struct display_mode_lib *mode_lib); +static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation( + struct display_mode_lib *mode_lib); +static unsigned int dscceComputeDelay( + unsigned int bpc, + double bpp, + unsigned int sliceWidth, + unsigned int numSlices, + enum output_format_class pixelFormat); +static unsigned int dscComputeDelay(enum output_format_class pixelFormat); +// Super monster function with some 45 argument +static bool CalculatePrefetchSchedule( + struct display_mode_lib *mode_lib, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + Pipe *myPipe, + unsigned int DSCDelay, + double DPPCLKDelaySubtotal, + double DPPCLKDelaySCL, + double DPPCLKDelaySCLLBOnly, + double DPPCLKDelayCNVCFormater, + double DPPCLKDelayCNVCCursor, + double DISPCLKDelaySubtotal, + unsigned int ScalerRecoutWidth, + enum output_format_class OutputFormat, + unsigned int MaxInterDCNTileRepeaters, + unsigned int VStartup, + unsigned int MaxVStartup, + unsigned int GPUVMPageTableLevels, + bool GPUVMEnable, + HostVM *myHostVM, + bool DynamicMetadataEnable, + int DynamicMetadataLinesBeforeActiveRequired, + unsigned int DynamicMetadataTransmittedBytes, + bool DCCEnable, + double UrgentLatency, + double UrgentExtraLatency, + double TCalc, + unsigned int PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + double PrefetchSourceLinesY, + unsigned int SwathWidthY, + double BytePerPixelDETY, + double VInitPreFillY, + unsigned int MaxNumSwathY, + double PrefetchSourceLinesC, + double BytePerPixelDETC, + double VInitPreFillC, + unsigned int MaxNumSwathC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double TWait, + bool XFCEnabled, + double XFCRemoteSurfaceFlipDelay, + bool ProgressiveToInterlaceUnitInOPP, + double *DSTXAfterScaler, + double *DSTYAfterScaler, + double *DestinationLinesForPrefetch, + double *PrefetchBandwidth, + double *DestinationLinesToRequestVMInVBlank, + double *DestinationLinesToRequestRowInVBlank, + double *VRatioPrefetchY, + double *VRatioPrefetchC, + double *RequiredPrefetchPixDataBWLuma, + double *RequiredPrefetchPixDataBWChroma, + unsigned int *VStartupRequiredWhenNotEnoughTimeForDynamicMetadata, + double *Tno_bw, + double *prefetch_vmrow_bw, + unsigned int *swath_width_luma_ub, + unsigned int *swath_width_chroma_ub, + unsigned int *VUpdateOffsetPix, + double *VUpdateWidthPix, + double *VReadyOffsetPix); +static double RoundToDFSGranularityUp(double Clock, double VCOSpeed); +static double RoundToDFSGranularityDown(double Clock, double VCOSpeed); +static double CalculateDCCConfiguration( + bool DCCEnabled, + bool DCCProgrammingAssumesScanDirectionUnknown, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + double DETBufferSize, + unsigned int RequestHeight256Byte, + unsigned int SwathHeight, + enum dm_swizzle_mode TilingFormat, + unsigned int BytePerPixel, + enum scan_direction_class ScanOrientation, + unsigned int *MaxUncompressedBlock, + unsigned int *MaxCompressedBlock, + unsigned int *Independent64ByteBlock); +static double CalculatePrefetchSourceLines( + struct display_mode_lib *mode_lib, + double VRatio, + double vtaps, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + unsigned int SwathHeight, + unsigned int ViewportYStart, + double *VInitPreFill, + unsigned int *MaxNumSwath); +static unsigned int CalculateVMAndRowBytes( + struct display_mode_lib *mode_lib, + bool DCCEnable, + unsigned int BlockHeight256Bytes, + unsigned int BlockWidth256Bytes, + enum source_format_class SourcePixelFormat, + unsigned int SurfaceTiling, + unsigned int BytePerPixel, + enum scan_direction_class ScanDirection, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + unsigned int SwathWidthY, + bool GPUVMEnable, + bool HostVMEnable, + unsigned int HostVMMaxPageTableLevels, + unsigned int HostVMCachedPageTableLevels, + unsigned int VMMPageSize, + unsigned int PTEBufferSizeInRequests, + unsigned int Pitch, + unsigned int DCCMetaPitch, + unsigned int *MacroTileWidth, + unsigned int *MetaRowByte, + unsigned int *PixelPTEBytesPerRow, + bool *PTEBufferSizeNotExceeded, + unsigned int *dpte_row_width_ub, + unsigned int *dpte_row_height, + unsigned int *MetaRequestWidth, + unsigned int *MetaRequestHeight, + unsigned int *meta_row_width, + unsigned int *meta_row_height, + unsigned int *vm_group_bytes, + unsigned int *dpte_group_bytes, + unsigned int *PixelPTEReqWidth, + unsigned int *PixelPTEReqHeight, + unsigned int *PTERequestSize, + unsigned int *DPDE0BytesFrame, + unsigned int *MetaPTEBytesFrame); + +static double CalculateTWait( + unsigned int PrefetchMode, + double DRAMClockChangeLatency, + double UrgentLatency, + double SREnterPlusExitTime); +static double CalculateRemoteSurfaceFlipDelay( + struct display_mode_lib *mode_lib, + double VRatio, + double SwathWidth, + double Bpp, + double LineTime, + double XFCTSlvVupdateOffset, + double XFCTSlvVupdateWidth, + double XFCTSlvVreadyOffset, + double XFCXBUFLatencyTolerance, + double XFCFillBWOverhead, + double XFCSlvChunkSize, + double XFCBusTransportTime, + double TCalc, + double TWait, + double *SrcActiveDrainRate, + double *TInitXFill, + double *TslvChk); +static void CalculateActiveRowBandwidth( + bool GPUVMEnable, + enum source_format_class SourcePixelFormat, + double VRatio, + bool DCCEnable, + double LineTime, + unsigned int MetaRowByteLuma, + unsigned int MetaRowByteChroma, + unsigned int meta_row_height_luma, + unsigned int meta_row_height_chroma, + unsigned int PixelPTEBytesPerRowLuma, + unsigned int PixelPTEBytesPerRowChroma, + unsigned int dpte_row_height_luma, + unsigned int dpte_row_height_chroma, + double *meta_row_bw, + double *dpte_row_bw); +static void CalculateFlipSchedule( + struct display_mode_lib *mode_lib, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + double UrgentExtraLatency, + double UrgentLatency, + unsigned int GPUVMMaxPageTableLevels, + bool HostVMEnable, + unsigned int HostVMMaxPageTableLevels, + unsigned int HostVMCachedPageTableLevels, + bool GPUVMEnable, + double PDEAndMetaPTEBytesPerFrame, + double MetaRowBytes, + double DPTEBytesPerRow, + double BandwidthAvailableForImmediateFlip, + unsigned int TotImmediateFlipBytes, + enum source_format_class SourcePixelFormat, + double LineTime, + double VRatio, + double Tno_bw, + bool DCCEnable, + unsigned int dpte_row_height, + unsigned int meta_row_height, + unsigned int dpte_row_height_chroma, + unsigned int meta_row_height_chroma, + double *DestinationLinesToRequestVMInImmediateFlip, + double *DestinationLinesToRequestRowInImmediateFlip, + double *final_flip_bw, + bool *ImmediateFlipSupportedForPipe); +static double CalculateWriteBackDelay( + enum source_format_class WritebackPixelFormat, + double WritebackHRatio, + double WritebackVRatio, + unsigned int WritebackLumaHTaps, + unsigned int WritebackLumaVTaps, + unsigned int WritebackChromaHTaps, + unsigned int WritebackChromaVTaps, + unsigned int WritebackDestinationWidth); +static void CalculateWatermarksAndDRAMSpeedChangeSupport( + struct display_mode_lib *mode_lib, + unsigned int PrefetchMode, + unsigned int NumberOfActivePlanes, + unsigned int MaxLineBufferLines, + unsigned int LineBufferSize, + unsigned int DPPOutputBufferPixels, + double DETBufferSizeInKByte, + unsigned int WritebackInterfaceLumaBufferSize, + unsigned int WritebackInterfaceChromaBufferSize, + double DCFCLK, + double UrgentOutOfOrderReturn, + double ReturnBW, + bool GPUVMEnable, + int dpte_group_bytes[], + unsigned int MetaChunkSize, + double UrgentLatency, + double ExtraLatency, + double WritebackLatency, + double WritebackChunkSize, + double SOCCLK, + double DRAMClockChangeLatency, + double SRExitTime, + double SREnterPlusExitTime, + double DCFCLKDeepSleep, + int DPPPerPlane[], + bool DCCEnable[], + double DPPCLK[], + double SwathWidthSingleDPPY[], + unsigned int SwathHeightY[], + double ReadBandwidthPlaneLuma[], + unsigned int SwathHeightC[], + double ReadBandwidthPlaneChroma[], + unsigned int LBBitPerPixel[], + double SwathWidthY[], + double HRatio[], + unsigned int vtaps[], + unsigned int VTAPsChroma[], + double VRatio[], + unsigned int HTotal[], + double PixelClock[], + unsigned int BlendingAndTiming[], + double BytePerPixelDETY[], + double BytePerPixelDETC[], + bool WritebackEnable[], + enum source_format_class WritebackPixelFormat[], + double WritebackDestinationWidth[], + double WritebackDestinationHeight[], + double WritebackSourceHeight[], + enum clock_change_support *DRAMClockChangeSupport, + double *UrgentWatermark, + double *WritebackUrgentWatermark, + double *DRAMClockChangeWatermark, + double *WritebackDRAMClockChangeWatermark, + double *StutterExitWatermark, + double *StutterEnterPlusExitWatermark, + double *MinActiveDRAMClockChangeLatencySupported); +static void CalculateDCFCLKDeepSleep( + struct display_mode_lib *mode_lib, + unsigned int NumberOfActivePlanes, + double BytePerPixelDETY[], + double BytePerPixelDETC[], + double VRatio[], + double SwathWidthY[], + int DPPPerPlane[], + double HRatio[], + double PixelClock[], + double PSCL_THROUGHPUT[], + double PSCL_THROUGHPUT_CHROMA[], + double DPPCLK[], + double *DCFCLKDeepSleep); +static void CalculateDETBufferSize( + double DETBufferSizeInKByte, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double *DETBufferSizeY, + double *DETBufferSizeC); +static void CalculateUrgentBurstFactor( + unsigned int DETBufferSizeInKByte, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + unsigned int SwathWidthY, + double LineTime, + double UrgentLatency, + double CursorBufferSize, + unsigned int CursorWidth, + unsigned int CursorBPP, + double VRatio, + double VRatioPreY, + double VRatioPreC, + double BytePerPixelInDETY, + double BytePerPixelInDETC, + double *UrgentBurstFactorCursor, + double *UrgentBurstFactorCursorPre, + double *UrgentBurstFactorLuma, + double *UrgentBurstFactorLumaPre, + double *UrgentBurstFactorChroma, + double *UrgentBurstFactorChromaPre, + unsigned int *NotEnoughUrgentLatencyHiding, + unsigned int *NotEnoughUrgentLatencyHidingPre); + +static void CalculatePixelDeliveryTimes( + unsigned int NumberOfActivePlanes, + double VRatio[], + double VRatioPrefetchY[], + double VRatioPrefetchC[], + unsigned int swath_width_luma_ub[], + unsigned int swath_width_chroma_ub[], + int DPPPerPlane[], + double HRatio[], + double PixelClock[], + double PSCL_THROUGHPUT[], + double PSCL_THROUGHPUT_CHROMA[], + double DPPCLK[], + double BytePerPixelDETC[], + enum scan_direction_class SourceScan[], + unsigned int BlockWidth256BytesY[], + unsigned int BlockHeight256BytesY[], + unsigned int BlockWidth256BytesC[], + unsigned int BlockHeight256BytesC[], + double DisplayPipeLineDeliveryTimeLuma[], + double DisplayPipeLineDeliveryTimeChroma[], + double DisplayPipeLineDeliveryTimeLumaPrefetch[], + double DisplayPipeLineDeliveryTimeChromaPrefetch[], + double DisplayPipeRequestDeliveryTimeLuma[], + double DisplayPipeRequestDeliveryTimeChroma[], + double DisplayPipeRequestDeliveryTimeLumaPrefetch[], + double DisplayPipeRequestDeliveryTimeChromaPrefetch[]); + +static void CalculateMetaAndPTETimes( + unsigned int NumberOfActivePlanes, + bool GPUVMEnable, + unsigned int MetaChunkSize, + unsigned int MinMetaChunkSizeBytes, + unsigned int GPUVMMaxPageTableLevels, + unsigned int HTotal[], + double VRatio[], + double VRatioPrefetchY[], + double VRatioPrefetchC[], + double DestinationLinesToRequestRowInVBlank[], + double DestinationLinesToRequestRowInImmediateFlip[], + double DestinationLinesToRequestVMInVBlank[], + double DestinationLinesToRequestVMInImmediateFlip[], + bool DCCEnable[], + double PixelClock[], + double BytePerPixelDETY[], + double BytePerPixelDETC[], + enum scan_direction_class SourceScan[], + unsigned int dpte_row_height[], + unsigned int dpte_row_height_chroma[], + unsigned int meta_row_width[], + unsigned int meta_row_height[], + unsigned int meta_req_width[], + unsigned int meta_req_height[], + int dpte_group_bytes[], + unsigned int PTERequestSizeY[], + unsigned int PTERequestSizeC[], + unsigned int PixelPTEReqWidthY[], + unsigned int PixelPTEReqHeightY[], + unsigned int PixelPTEReqWidthC[], + unsigned int PixelPTEReqHeightC[], + unsigned int dpte_row_width_luma_ub[], + unsigned int dpte_row_width_chroma_ub[], + unsigned int vm_group_bytes[], + unsigned int dpde0_bytes_per_frame_ub_l[], + unsigned int dpde0_bytes_per_frame_ub_c[], + unsigned int meta_pte_bytes_per_frame_ub_l[], + unsigned int meta_pte_bytes_per_frame_ub_c[], + double DST_Y_PER_PTE_ROW_NOM_L[], + double DST_Y_PER_PTE_ROW_NOM_C[], + double DST_Y_PER_META_ROW_NOM_L[], + double TimePerMetaChunkNominal[], + double TimePerMetaChunkVBlank[], + double TimePerMetaChunkFlip[], + double time_per_pte_group_nom_luma[], + double time_per_pte_group_vblank_luma[], + double time_per_pte_group_flip_luma[], + double time_per_pte_group_nom_chroma[], + double time_per_pte_group_vblank_chroma[], + double time_per_pte_group_flip_chroma[], + double TimePerVMGroupVBlank[], + double TimePerVMGroupFlip[], + double TimePerVMRequestVBlank[], + double TimePerVMRequestFlip[]); + +static double CalculateExtraLatency( + double UrgentRoundTripAndOutOfOrderLatency, + int TotalNumberOfActiveDPP, + int PixelChunkSizeInKByte, + int TotalNumberOfDCCActiveDPP, + int MetaChunkSize, + double ReturnBW, + bool GPUVMEnable, + bool HostVMEnable, + int NumberOfActivePlanes, + int NumberOfDPP[], + int dpte_group_bytes[], + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + int HostVMMaxPageTableLevels, + int HostVMCachedPageTableLevels); + +void dml21_recalculate(struct display_mode_lib *mode_lib) +{ + ModeSupportAndSystemConfiguration(mode_lib); + PixelClockAdjustmentForProgressiveToInterlaceUnit(mode_lib); + DisplayPipeConfiguration(mode_lib); + DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation(mode_lib); +} + +static unsigned int dscceComputeDelay( + unsigned int bpc, + double bpp, + unsigned int sliceWidth, + unsigned int numSlices, + enum output_format_class pixelFormat) +{ + // valid bpc = source bits per component in the set of {8, 10, 12} + // valid bpp = increments of 1/16 of a bit + // min = 6/7/8 in N420/N422/444, respectively + // max = such that compression is 1:1 + //valid sliceWidth = number of pixels per slice line, must be less than or equal to 5184/numSlices (or 4096/numSlices in 420 mode) + //valid numSlices = number of slices in the horiziontal direction per DSC engine in the set of {1, 2, 3, 4} + //valid pixelFormat = pixel/color format in the set of {:N444_RGB, :S422, :N422, :N420} + + // fixed value + unsigned int rcModelSize = 8192; + + // N422/N420 operate at 2 pixels per clock + unsigned int pixelsPerClock, lstall, D, initalXmitDelay, w, S, ix, wx, p, l0, a, ax, l, + Delay, pixels; + + if (pixelFormat == dm_n422 || pixelFormat == dm_420) + pixelsPerClock = 2; + // #all other modes operate at 1 pixel per clock + else + pixelsPerClock = 1; + + //initial transmit delay as per PPS + initalXmitDelay = dml_round(rcModelSize / 2.0 / bpp / pixelsPerClock); + + //compute ssm delay + if (bpc == 8) + D = 81; + else if (bpc == 10) + D = 89; + else + D = 113; + + //divide by pixel per cycle to compute slice width as seen by DSC + w = sliceWidth / pixelsPerClock; + + //422 mode has an additional cycle of delay + if (pixelFormat == dm_s422) + S = 1; + else + S = 0; + + //main calculation for the dscce + ix = initalXmitDelay + 45; + wx = (w + 2) / 3; + p = 3 * wx - w; + l0 = ix / w; + a = ix + p * l0; + ax = (a + 2) / 3 + D + 6 + 1; + l = (ax + wx - 1) / wx; + if ((ix % w) == 0 && p != 0) + lstall = 1; + else + lstall = 0; + Delay = l * wx * (numSlices - 1) + ax + S + lstall + 22; + + //dsc processes 3 pixel containers per cycle and a container can contain 1 or 2 pixels + pixels = Delay * 3 * pixelsPerClock; + return pixels; +} + +static unsigned int dscComputeDelay(enum output_format_class pixelFormat) +{ + unsigned int Delay = 0; + + if (pixelFormat == dm_420) { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 0; + // dscc - input deserializer + Delay = Delay + 3; + // dscc gets pixels every other cycle + Delay = Delay + 2; + // dscc - input cdc fifo + Delay = Delay + 12; + // dscc gets pixels every other cycle + Delay = Delay + 13; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 7; + // dscc gets pixels every other cycle + Delay = Delay + 3; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output serializer + Delay = Delay + 1; + // sft + Delay = Delay + 1; + } else if (pixelFormat == dm_n422) { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 1; + // dscc - input deserializer + Delay = Delay + 5; + // dscc - input cdc fifo + Delay = Delay + 25; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 10; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output serializer + Delay = Delay + 1; + // sft + Delay = Delay + 1; + } else { + // sfr + Delay = Delay + 2; + // dsccif + Delay = Delay + 0; + // dscc - input deserializer + Delay = Delay + 3; + // dscc - input cdc fifo + Delay = Delay + 12; + // dscc - cdc uncertainty + Delay = Delay + 2; + // dscc - output cdc fifo + Delay = Delay + 7; + // dscc - output serializer + Delay = Delay + 1; + // dscc - cdc uncertainty + Delay = Delay + 2; + // sft + Delay = Delay + 1; + } + + return Delay; +} + +static bool CalculatePrefetchSchedule( + struct display_mode_lib *mode_lib, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + Pipe *myPipe, + unsigned int DSCDelay, + double DPPCLKDelaySubtotal, + double DPPCLKDelaySCL, + double DPPCLKDelaySCLLBOnly, + double DPPCLKDelayCNVCFormater, + double DPPCLKDelayCNVCCursor, + double DISPCLKDelaySubtotal, + unsigned int ScalerRecoutWidth, + enum output_format_class OutputFormat, + unsigned int MaxInterDCNTileRepeaters, + unsigned int VStartup, + unsigned int MaxVStartup, + unsigned int GPUVMPageTableLevels, + bool GPUVMEnable, + HostVM *myHostVM, + bool DynamicMetadataEnable, + int DynamicMetadataLinesBeforeActiveRequired, + unsigned int DynamicMetadataTransmittedBytes, + bool DCCEnable, + double UrgentLatency, + double UrgentExtraLatency, + double TCalc, + unsigned int PDEAndMetaPTEBytesFrame, + unsigned int MetaRowByte, + unsigned int PixelPTEBytesPerRow, + double PrefetchSourceLinesY, + unsigned int SwathWidthY, + double BytePerPixelDETY, + double VInitPreFillY, + unsigned int MaxNumSwathY, + double PrefetchSourceLinesC, + double BytePerPixelDETC, + double VInitPreFillC, + unsigned int MaxNumSwathC, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double TWait, + bool XFCEnabled, + double XFCRemoteSurfaceFlipDelay, + bool ProgressiveToInterlaceUnitInOPP, + double *DSTXAfterScaler, + double *DSTYAfterScaler, + double *DestinationLinesForPrefetch, + double *PrefetchBandwidth, + double *DestinationLinesToRequestVMInVBlank, + double *DestinationLinesToRequestRowInVBlank, + double *VRatioPrefetchY, + double *VRatioPrefetchC, + double *RequiredPrefetchPixDataBWLuma, + double *RequiredPrefetchPixDataBWChroma, + unsigned int *VStartupRequiredWhenNotEnoughTimeForDynamicMetadata, + double *Tno_bw, + double *prefetch_vmrow_bw, + unsigned int *swath_width_luma_ub, + unsigned int *swath_width_chroma_ub, + unsigned int *VUpdateOffsetPix, + double *VUpdateWidthPix, + double *VReadyOffsetPix) +{ + bool MyError = false; + unsigned int DPPCycles, DISPCLKCycles; + double DSTTotalPixelsAfterScaler, TotalRepeaterDelayTime; + double Tdm, LineTime, Tsetup; + double dst_y_prefetch_equ; + double Tsw_oto; + double prefetch_bw_oto; + double Tvm_oto; + double Tr0_oto; + double Tvm_oto_lines; + double Tr0_oto_lines; + double Tsw_oto_lines; + double dst_y_prefetch_oto; + double TimeForFetchingMetaPTE = 0; + double TimeForFetchingRowInVBlank = 0; + double LinesToRequestPrefetchPixelData = 0; + double HostVMInefficiencyFactor; + unsigned int HostVMDynamicLevels; + + if (GPUVMEnable == true && myHostVM->Enable == true) { + HostVMInefficiencyFactor = + PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData + / PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly; + HostVMDynamicLevels = myHostVM->MaxPageTableLevels + - myHostVM->CachedPageTableLevels; + } else { + HostVMInefficiencyFactor = 1; + HostVMDynamicLevels = 0; + } + + if (myPipe->ScalerEnabled) + DPPCycles = DPPCLKDelaySubtotal + DPPCLKDelaySCL; + else + DPPCycles = DPPCLKDelaySubtotal + DPPCLKDelaySCLLBOnly; + + DPPCycles = DPPCycles + DPPCLKDelayCNVCFormater + myPipe->NumberOfCursors * DPPCLKDelayCNVCCursor; + + DISPCLKCycles = DISPCLKDelaySubtotal; + + if (myPipe->DPPCLK == 0.0 || myPipe->DISPCLK == 0.0) + return true; + + *DSTXAfterScaler = DPPCycles * myPipe->PixelClock / myPipe->DPPCLK + + DISPCLKCycles * myPipe->PixelClock / myPipe->DISPCLK + DSCDelay; + + if (myPipe->DPPPerPlane > 1) + *DSTXAfterScaler = *DSTXAfterScaler + ScalerRecoutWidth; + + if (OutputFormat == dm_420 || (myPipe->InterlaceEnable && ProgressiveToInterlaceUnitInOPP)) + *DSTYAfterScaler = 1; + else + *DSTYAfterScaler = 0; + + DSTTotalPixelsAfterScaler = ((double) (*DSTYAfterScaler * myPipe->HTotal)) + *DSTXAfterScaler; + *DSTYAfterScaler = dml_floor(DSTTotalPixelsAfterScaler / myPipe->HTotal, 1); + *DSTXAfterScaler = DSTTotalPixelsAfterScaler - ((double) (*DSTYAfterScaler * myPipe->HTotal)); + + *VUpdateOffsetPix = dml_ceil(myPipe->HTotal / 4.0, 1); + TotalRepeaterDelayTime = MaxInterDCNTileRepeaters * (2.0 / myPipe->DPPCLK + 3.0 / myPipe->DISPCLK); + *VUpdateWidthPix = (14.0 / myPipe->DCFCLKDeepSleep + 12.0 / myPipe->DPPCLK + TotalRepeaterDelayTime) + * myPipe->PixelClock; + + *VReadyOffsetPix = dml_max( + 150.0 / myPipe->DPPCLK, + TotalRepeaterDelayTime + 20.0 / myPipe->DCFCLKDeepSleep + 10.0 / myPipe->DPPCLK) + * myPipe->PixelClock; + + Tsetup = (double) (*VUpdateOffsetPix + *VUpdateWidthPix + *VReadyOffsetPix) / myPipe->PixelClock; + + LineTime = (double) myPipe->HTotal / myPipe->PixelClock; + + if (DynamicMetadataEnable) { + double Tdmbf, Tdmec, Tdmsks; + + Tdm = dml_max(0.0, UrgentExtraLatency - TCalc); + Tdmbf = DynamicMetadataTransmittedBytes / 4.0 / myPipe->DISPCLK; + Tdmec = LineTime; + if (DynamicMetadataLinesBeforeActiveRequired == -1) + Tdmsks = myPipe->VBlank * LineTime / 2.0; + else + Tdmsks = DynamicMetadataLinesBeforeActiveRequired * LineTime; + if (myPipe->InterlaceEnable && !ProgressiveToInterlaceUnitInOPP) + Tdmsks = Tdmsks / 2; + if (VStartup * LineTime + < Tsetup + TWait + UrgentExtraLatency + Tdmbf + Tdmec + Tdmsks) { + MyError = true; + *VStartupRequiredWhenNotEnoughTimeForDynamicMetadata = (Tsetup + TWait + + UrgentExtraLatency + Tdmbf + Tdmec + Tdmsks) / LineTime; + } else + *VStartupRequiredWhenNotEnoughTimeForDynamicMetadata = 0.0; + } else + Tdm = 0; + + if (GPUVMEnable) { + if (GPUVMPageTableLevels >= 3) + *Tno_bw = UrgentExtraLatency + UrgentLatency * ((GPUVMPageTableLevels - 2) * (myHostVM->MaxPageTableLevels + 1) - 1); + else + *Tno_bw = 0; + } else if (!DCCEnable) + *Tno_bw = LineTime; + else + *Tno_bw = LineTime / 4; + + dst_y_prefetch_equ = VStartup - dml_max(TCalc + TWait, XFCRemoteSurfaceFlipDelay) / LineTime + - (Tsetup + Tdm) / LineTime + - (*DSTYAfterScaler + *DSTXAfterScaler / myPipe->HTotal); + + Tsw_oto = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime; + + if (myPipe->SourceScan == dm_horz) { + *swath_width_luma_ub = dml_ceil(SwathWidthY - 1, myPipe->BlockWidth256BytesY) + myPipe->BlockWidth256BytesY; + *swath_width_chroma_ub = dml_ceil(SwathWidthY / 2 - 1, myPipe->BlockWidth256BytesC) + myPipe->BlockWidth256BytesC; + } else { + *swath_width_luma_ub = dml_ceil(SwathWidthY - 1, myPipe->BlockHeight256BytesY) + myPipe->BlockHeight256BytesY; + *swath_width_chroma_ub = dml_ceil(SwathWidthY / 2 - 1, myPipe->BlockHeight256BytesC) + myPipe->BlockHeight256BytesC; + } + + prefetch_bw_oto = (PrefetchSourceLinesY * *swath_width_luma_ub * dml_ceil(BytePerPixelDETY, 1) + PrefetchSourceLinesC * *swath_width_chroma_ub * dml_ceil(BytePerPixelDETC, 2)) / Tsw_oto; + + + if (GPUVMEnable == true) { + Tvm_oto = dml_max(*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / prefetch_bw_oto, + dml_max(UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1), + LineTime / 4.0)); + } else + Tvm_oto = LineTime / 4.0; + + if ((GPUVMEnable == true || DCCEnable == true)) { + Tr0_oto = dml_max( + (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / prefetch_bw_oto, + dml_max(UrgentLatency * (HostVMDynamicLevels + 1), dml_max(LineTime - Tvm_oto, LineTime / 4))); + } else + Tr0_oto = (LineTime - Tvm_oto) / 2.0; + + Tvm_oto_lines = dml_ceil(4 * Tvm_oto / LineTime, 1) / 4.0; + Tr0_oto_lines = dml_ceil(4 * Tr0_oto / LineTime, 1) / 4.0; + Tsw_oto_lines = dml_ceil(4 * Tsw_oto / LineTime, 1) / 4.0; + dst_y_prefetch_oto = Tvm_oto_lines + 2 * Tr0_oto_lines + Tsw_oto_lines + 0.75; + + dst_y_prefetch_equ = dml_floor(4.0 * (dst_y_prefetch_equ + 0.125), 1) / 4.0; + + if (dst_y_prefetch_oto < dst_y_prefetch_equ) + *DestinationLinesForPrefetch = dst_y_prefetch_oto; + else + *DestinationLinesForPrefetch = dst_y_prefetch_equ; + + dml_print("DML: VStartup: %d\n", VStartup); + dml_print("DML: TCalc: %f\n", TCalc); + dml_print("DML: TWait: %f\n", TWait); + dml_print("DML: XFCRemoteSurfaceFlipDelay: %f\n", XFCRemoteSurfaceFlipDelay); + dml_print("DML: LineTime: %f\n", LineTime); + dml_print("DML: Tsetup: %f\n", Tsetup); + dml_print("DML: Tdm: %f\n", Tdm); + dml_print("DML: DSTYAfterScaler: %f\n", *DSTYAfterScaler); + dml_print("DML: DSTXAfterScaler: %f\n", *DSTXAfterScaler); + dml_print("DML: HTotal: %d\n", myPipe->HTotal); + + *PrefetchBandwidth = 0; + *DestinationLinesToRequestVMInVBlank = 0; + *DestinationLinesToRequestRowInVBlank = 0; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBWLuma = 0; + if (*DestinationLinesForPrefetch > 1) { + double PrefetchBandwidth1 = (PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor + 2 * MetaRowByte + + 2 * PixelPTEBytesPerRow * HostVMInefficiencyFactor + + PrefetchSourceLinesY * *swath_width_luma_ub * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * *swath_width_chroma_ub * dml_ceil(BytePerPixelDETC, 2)) + / (*DestinationLinesForPrefetch * LineTime - *Tno_bw); + + double PrefetchBandwidth2 = (PDEAndMetaPTEBytesFrame * + HostVMInefficiencyFactor + PrefetchSourceLinesY * + *swath_width_luma_ub * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * *swath_width_chroma_ub * + dml_ceil(BytePerPixelDETC, 2)) / + (*DestinationLinesForPrefetch * LineTime - *Tno_bw - 2 * + UrgentLatency * (1 + HostVMDynamicLevels)); + + double PrefetchBandwidth3 = (2 * MetaRowByte + 2 * PixelPTEBytesPerRow + * HostVMInefficiencyFactor + PrefetchSourceLinesY * + *swath_width_luma_ub * dml_ceil(BytePerPixelDETY, 1) + + PrefetchSourceLinesC * *swath_width_chroma_ub * + dml_ceil(BytePerPixelDETC, 2)) / + (*DestinationLinesForPrefetch * LineTime - + UrgentExtraLatency - UrgentLatency * (GPUVMPageTableLevels + * (HostVMDynamicLevels + 1) - 1)); + + double PrefetchBandwidth4 = (PrefetchSourceLinesY * *swath_width_luma_ub * + dml_ceil(BytePerPixelDETY, 1) + PrefetchSourceLinesC * + *swath_width_chroma_ub * dml_ceil(BytePerPixelDETC, 2)) / + (*DestinationLinesForPrefetch * LineTime - + UrgentExtraLatency - UrgentLatency * (GPUVMPageTableLevels + * (HostVMDynamicLevels + 1) - 1) - 2 * UrgentLatency * + (1 + HostVMDynamicLevels)); + + if (VStartup == MaxVStartup && (PrefetchBandwidth1 > 4 * prefetch_bw_oto) && (*DestinationLinesForPrefetch - dml_ceil(Tsw_oto_lines, 1) / 4.0 - 0.75) * LineTime - *Tno_bw > 0) { + PrefetchBandwidth1 = (PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor + 2 * MetaRowByte + 2 * PixelPTEBytesPerRow * HostVMInefficiencyFactor) / ((*DestinationLinesForPrefetch - dml_ceil(Tsw_oto_lines, 1) / 4.0 - 0.75) * LineTime - *Tno_bw); + } + if (*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / PrefetchBandwidth1 >= UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1) && (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / PrefetchBandwidth1 >= UrgentLatency * (1 + HostVMDynamicLevels)) { + *PrefetchBandwidth = PrefetchBandwidth1; + } else if (*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / PrefetchBandwidth2 >= UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1) && (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / PrefetchBandwidth2 < UrgentLatency * (1 + HostVMDynamicLevels)) { + *PrefetchBandwidth = PrefetchBandwidth2; + } else if (*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / PrefetchBandwidth3 < UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1) && (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / PrefetchBandwidth3 >= UrgentLatency * (1 + HostVMDynamicLevels)) { + *PrefetchBandwidth = PrefetchBandwidth3; + } else { + *PrefetchBandwidth = PrefetchBandwidth4; + } + + if (GPUVMEnable) { + TimeForFetchingMetaPTE = dml_max(*Tno_bw + (double) PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / *PrefetchBandwidth, + dml_max(UrgentExtraLatency + UrgentLatency * (GPUVMPageTableLevels * (HostVMDynamicLevels + 1) - 1), LineTime / 4)); + } else { +// 5/30/2018 - This was an optimization requested from Sy but now NumberOfCursors is no longer a factor +// so if this needs to be reinstated, then it should be officially done in the VBA code as well. +// if (mode_lib->NumberOfCursors > 0 || XFCEnabled) + TimeForFetchingMetaPTE = LineTime / 4; +// else +// TimeForFetchingMetaPTE = 0.0; + } + + if ((GPUVMEnable == true || DCCEnable == true)) { + TimeForFetchingRowInVBlank = + dml_max( + (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) + / *PrefetchBandwidth, + dml_max( + UrgentLatency * (1 + HostVMDynamicLevels), + dml_max( + (LineTime + - TimeForFetchingMetaPTE) / 2.0, + LineTime + / 4.0))); + } else { +// See note above dated 5/30/2018 +// if (NumberOfCursors > 0 || XFCEnabled) + TimeForFetchingRowInVBlank = (LineTime - TimeForFetchingMetaPTE) / 2.0; +// else // TODO: Did someone else add this?? +// TimeForFetchingRowInVBlank = 0.0; + } + + *DestinationLinesToRequestVMInVBlank = dml_ceil(4.0 * TimeForFetchingMetaPTE / LineTime, 1.0) / 4.0; + + *DestinationLinesToRequestRowInVBlank = dml_ceil(4.0 * TimeForFetchingRowInVBlank / LineTime, 1.0) / 4.0; + + LinesToRequestPrefetchPixelData = *DestinationLinesForPrefetch +// See note above dated 5/30/2018 +// - ((NumberOfCursors > 0 || GPUVMEnable || DCCEnable) ? + - ((GPUVMEnable || DCCEnable) ? + (*DestinationLinesToRequestVMInVBlank + 2 * *DestinationLinesToRequestRowInVBlank) : + 0.0); // TODO: Did someone else add this?? + + if (LinesToRequestPrefetchPixelData > 0) { + + *VRatioPrefetchY = (double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData; + *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); + if ((SwathHeightY > 4) && (VInitPreFillY > 3)) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillY - 3.0) / 2.0) { + *VRatioPrefetchY = + dml_max( + (double) PrefetchSourceLinesY + / LinesToRequestPrefetchPixelData, + (double) MaxNumSwathY + * SwathHeightY + / (LinesToRequestPrefetchPixelData + - (VInitPreFillY + - 3.0) + / 2.0)); + *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); + } else { + MyError = true; + *VRatioPrefetchY = 0; + } + } + + *VRatioPrefetchC = (double) PrefetchSourceLinesC + / LinesToRequestPrefetchPixelData; + *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); + + if ((SwathHeightC > 4)) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillC - 3.0) / 2.0) { + *VRatioPrefetchC = + dml_max( + *VRatioPrefetchC, + (double) MaxNumSwathC + * SwathHeightC + / (LinesToRequestPrefetchPixelData + - (VInitPreFillC + - 3.0) + / 2.0)); + *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); + } else { + MyError = true; + *VRatioPrefetchC = 0; + } + } + + *RequiredPrefetchPixDataBWLuma = myPipe->DPPPerPlane + * (double) PrefetchSourceLinesY / LinesToRequestPrefetchPixelData + * dml_ceil(BytePerPixelDETY, 1) + * *swath_width_luma_ub / LineTime; + *RequiredPrefetchPixDataBWChroma = myPipe->DPPPerPlane + * (double) PrefetchSourceLinesC / LinesToRequestPrefetchPixelData + * dml_ceil(BytePerPixelDETC, 2) + * *swath_width_chroma_ub / LineTime; + } else { + MyError = true; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBWLuma = 0; + *RequiredPrefetchPixDataBWChroma = 0; + } + + dml_print("DML: Tvm: %fus\n", TimeForFetchingMetaPTE); + dml_print("DML: Tr0: %fus\n", TimeForFetchingRowInVBlank); + dml_print("DML: Tsw: %fus\n", (double)(*DestinationLinesForPrefetch) * LineTime - TimeForFetchingMetaPTE - TimeForFetchingRowInVBlank); + dml_print("DML: Tpre: %fus\n", (double)(*DestinationLinesForPrefetch) * LineTime); + dml_print("DML: row_bytes = dpte_row_bytes (per_pipe) = PixelPTEBytesPerRow = : %d\n", PixelPTEBytesPerRow); + + } else { + MyError = true; + } + + { + double prefetch_vm_bw; + double prefetch_row_bw; + + if (PDEAndMetaPTEBytesFrame == 0) { + prefetch_vm_bw = 0; + } else if (*DestinationLinesToRequestVMInVBlank > 0) { + prefetch_vm_bw = PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / (*DestinationLinesToRequestVMInVBlank * LineTime); + } else { + prefetch_vm_bw = 0; + MyError = true; + } + if (MetaRowByte + PixelPTEBytesPerRow == 0) { + prefetch_row_bw = 0; + } else if (*DestinationLinesToRequestRowInVBlank > 0) { + prefetch_row_bw = (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / (*DestinationLinesToRequestRowInVBlank * LineTime); + } else { + prefetch_row_bw = 0; + MyError = true; + } + + *prefetch_vmrow_bw = dml_max(prefetch_vm_bw, prefetch_row_bw); + } + + if (MyError) { + *PrefetchBandwidth = 0; + TimeForFetchingMetaPTE = 0; + TimeForFetchingRowInVBlank = 0; + *DestinationLinesToRequestVMInVBlank = 0; + *DestinationLinesToRequestRowInVBlank = 0; + *DestinationLinesForPrefetch = 0; + LinesToRequestPrefetchPixelData = 0; + *VRatioPrefetchY = 0; + *VRatioPrefetchC = 0; + *RequiredPrefetchPixDataBWLuma = 0; + *RequiredPrefetchPixDataBWChroma = 0; + } + + return MyError; +} + +static double RoundToDFSGranularityUp(double Clock, double VCOSpeed) +{ + return VCOSpeed * 4 / dml_floor(VCOSpeed * 4 / Clock, 1); +} + +static double RoundToDFSGranularityDown(double Clock, double VCOSpeed) +{ + return VCOSpeed * 4 / dml_ceil(VCOSpeed * 4 / Clock, 1); +} + +static double CalculateDCCConfiguration( + bool DCCEnabled, + bool DCCProgrammingAssumesScanDirectionUnknown, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + double DETBufferSize, + unsigned int RequestHeight256Byte, + unsigned int SwathHeight, + enum dm_swizzle_mode TilingFormat, + unsigned int BytePerPixel, + enum scan_direction_class ScanOrientation, + unsigned int *MaxUncompressedBlock, + unsigned int *MaxCompressedBlock, + unsigned int *Independent64ByteBlock) +{ + double MaximumDCCCompressionSurface = 0.0; + enum { + REQ_256Bytes, + REQ_128BytesNonContiguous, + REQ_128BytesContiguous, + REQ_NA + } Request = REQ_NA; + + if (DCCEnabled == true) { + if (DCCProgrammingAssumesScanDirectionUnknown == true) { + if (DETBufferSize >= RequestHeight256Byte * ViewportWidth * BytePerPixel + && DETBufferSize + >= 256 / RequestHeight256Byte + * ViewportHeight) { + Request = REQ_256Bytes; + } else if ((DETBufferSize + < RequestHeight256Byte * ViewportWidth * BytePerPixel + && (BytePerPixel == 2 || BytePerPixel == 4)) + || (DETBufferSize + < 256 / RequestHeight256Byte + * ViewportHeight + && BytePerPixel == 8 + && (TilingFormat == dm_sw_4kb_d + || TilingFormat + == dm_sw_4kb_d_x + || TilingFormat + == dm_sw_var_d + || TilingFormat + == dm_sw_var_d_x + || TilingFormat + == dm_sw_64kb_d + || TilingFormat + == dm_sw_64kb_d_x + || TilingFormat + == dm_sw_64kb_d_t + || TilingFormat + == dm_sw_64kb_r_x))) { + Request = REQ_128BytesNonContiguous; + } else { + Request = REQ_128BytesContiguous; + } + } else { + if (BytePerPixel == 1) { + if (ScanOrientation == dm_vert || SwathHeight == 16) { + Request = REQ_256Bytes; + } else { + Request = REQ_128BytesContiguous; + } + } else if (BytePerPixel == 2) { + if ((ScanOrientation == dm_vert && SwathHeight == 16) || (ScanOrientation != dm_vert && SwathHeight == 8)) { + Request = REQ_256Bytes; + } else if (ScanOrientation == dm_vert) { + Request = REQ_128BytesContiguous; + } else { + Request = REQ_128BytesNonContiguous; + } + } else if (BytePerPixel == 4) { + if (SwathHeight == 8) { + Request = REQ_256Bytes; + } else if (ScanOrientation == dm_vert) { + Request = REQ_128BytesContiguous; + } else { + Request = REQ_128BytesNonContiguous; + } + } else if (BytePerPixel == 8) { + if (TilingFormat == dm_sw_4kb_d || TilingFormat == dm_sw_4kb_d_x + || TilingFormat == dm_sw_var_d + || TilingFormat == dm_sw_var_d_x + || TilingFormat == dm_sw_64kb_d + || TilingFormat == dm_sw_64kb_d_x + || TilingFormat == dm_sw_64kb_d_t + || TilingFormat == dm_sw_64kb_r_x) { + if ((ScanOrientation == dm_vert && SwathHeight == 8) + || (ScanOrientation != dm_vert + && SwathHeight == 4)) { + Request = REQ_256Bytes; + } else if (ScanOrientation != dm_vert) { + Request = REQ_128BytesContiguous; + } else { + Request = REQ_128BytesNonContiguous; + } + } else { + if (ScanOrientation != dm_vert || SwathHeight == 8) { + Request = REQ_256Bytes; + } else { + Request = REQ_128BytesContiguous; + } + } + } + } + } else { + Request = REQ_NA; + } + + if (Request == REQ_256Bytes) { + *MaxUncompressedBlock = 256; + *MaxCompressedBlock = 256; + *Independent64ByteBlock = false; + MaximumDCCCompressionSurface = 4.0; + } else if (Request == REQ_128BytesContiguous) { + *MaxUncompressedBlock = 128; + *MaxCompressedBlock = 128; + *Independent64ByteBlock = false; + MaximumDCCCompressionSurface = 2.0; + } else if (Request == REQ_128BytesNonContiguous) { + *MaxUncompressedBlock = 256; + *MaxCompressedBlock = 64; + *Independent64ByteBlock = true; + MaximumDCCCompressionSurface = 4.0; + } else { + *MaxUncompressedBlock = 0; + *MaxCompressedBlock = 0; + *Independent64ByteBlock = 0; + MaximumDCCCompressionSurface = 0.0; + } + + return MaximumDCCCompressionSurface; +} + +static double CalculatePrefetchSourceLines( + struct display_mode_lib *mode_lib, + double VRatio, + double vtaps, + bool Interlace, + bool ProgressiveToInterlaceUnitInOPP, + unsigned int SwathHeight, + unsigned int ViewportYStart, + double *VInitPreFill, + unsigned int *MaxNumSwath) +{ + unsigned int MaxPartialSwath; + + if (ProgressiveToInterlaceUnitInOPP) + *VInitPreFill = dml_floor((VRatio + vtaps + 1) / 2.0, 1); + else + *VInitPreFill = dml_floor((VRatio + vtaps + 1 + Interlace * 0.5 * VRatio) / 2.0, 1); + + if (!mode_lib->vba.IgnoreViewportPositioning) { + + *MaxNumSwath = dml_ceil((*VInitPreFill - 1.0) / SwathHeight, 1) + 1.0; + + if (*VInitPreFill > 1.0) + MaxPartialSwath = (unsigned int) (*VInitPreFill - 2) % SwathHeight; + else + MaxPartialSwath = (unsigned int) (*VInitPreFill + SwathHeight - 2) + % SwathHeight; + MaxPartialSwath = dml_max(1U, MaxPartialSwath); + + } else { + + if (ViewportYStart != 0) + dml_print( + "WARNING DML: using viewport y position of 0 even though actual viewport y position is non-zero in prefetch source lines calculation\n"); + + *MaxNumSwath = dml_ceil(*VInitPreFill / SwathHeight, 1); + + if (*VInitPreFill > 1.0) + MaxPartialSwath = (unsigned int) (*VInitPreFill - 1) % SwathHeight; + else + MaxPartialSwath = (unsigned int) (*VInitPreFill + SwathHeight - 1) + % SwathHeight; + } + + return *MaxNumSwath * SwathHeight + MaxPartialSwath; +} + +static unsigned int CalculateVMAndRowBytes( + struct display_mode_lib *mode_lib, + bool DCCEnable, + unsigned int BlockHeight256Bytes, + unsigned int BlockWidth256Bytes, + enum source_format_class SourcePixelFormat, + unsigned int SurfaceTiling, + unsigned int BytePerPixel, + enum scan_direction_class ScanDirection, + unsigned int ViewportWidth, + unsigned int ViewportHeight, + unsigned int SwathWidth, + bool GPUVMEnable, + bool HostVMEnable, + unsigned int HostVMMaxPageTableLevels, + unsigned int HostVMCachedPageTableLevels, + unsigned int VMMPageSize, + unsigned int PTEBufferSizeInRequests, + unsigned int Pitch, + unsigned int DCCMetaPitch, + unsigned int *MacroTileWidth, + unsigned int *MetaRowByte, + unsigned int *PixelPTEBytesPerRow, + bool *PTEBufferSizeNotExceeded, + unsigned int *dpte_row_width_ub, + unsigned int *dpte_row_height, + unsigned int *MetaRequestWidth, + unsigned int *MetaRequestHeight, + unsigned int *meta_row_width, + unsigned int *meta_row_height, + unsigned int *vm_group_bytes, + unsigned int *dpte_group_bytes, + unsigned int *PixelPTEReqWidth, + unsigned int *PixelPTEReqHeight, + unsigned int *PTERequestSize, + unsigned int *DPDE0BytesFrame, + unsigned int *MetaPTEBytesFrame) +{ + unsigned int MPDEBytesFrame; + unsigned int DCCMetaSurfaceBytes; + unsigned int MacroTileSizeBytes; + unsigned int MacroTileHeight; + unsigned int ExtraDPDEBytesFrame; + unsigned int PDEAndMetaPTEBytesFrame; + unsigned int PixelPTEReqHeightPTEs; + + if (DCCEnable == true) { + *MetaRequestHeight = 8 * BlockHeight256Bytes; + *MetaRequestWidth = 8 * BlockWidth256Bytes; + if (ScanDirection == dm_horz) { + *meta_row_height = *MetaRequestHeight; + *meta_row_width = dml_ceil((double) SwathWidth - 1, *MetaRequestWidth) + + *MetaRequestWidth; + *MetaRowByte = *meta_row_width * *MetaRequestHeight * BytePerPixel / 256.0; + } else { + *meta_row_height = *MetaRequestWidth; + *meta_row_width = dml_ceil((double) SwathWidth - 1, *MetaRequestHeight) + + *MetaRequestHeight; + *MetaRowByte = *meta_row_width * *MetaRequestWidth * BytePerPixel / 256.0; + } + if (ScanDirection == dm_horz) { + DCCMetaSurfaceBytes = DCCMetaPitch + * (dml_ceil(ViewportHeight - 1, 64 * BlockHeight256Bytes) + + 64 * BlockHeight256Bytes) * BytePerPixel + / 256; + } else { + DCCMetaSurfaceBytes = DCCMetaPitch + * (dml_ceil( + (double) ViewportHeight - 1, + 64 * BlockHeight256Bytes) + + 64 * BlockHeight256Bytes) * BytePerPixel + / 256; + } + if (GPUVMEnable == true) { + *MetaPTEBytesFrame = (dml_ceil( + (double) (DCCMetaSurfaceBytes - VMMPageSize) + / (8 * VMMPageSize), + 1) + 1) * 64; + MPDEBytesFrame = 128 * ((mode_lib->vba.GPUVMMaxPageTableLevels + 1) * (mode_lib->vba.HostVMMaxPageTableLevels + 1) - 2); + } else { + *MetaPTEBytesFrame = 0; + MPDEBytesFrame = 0; + } + } else { + *MetaPTEBytesFrame = 0; + MPDEBytesFrame = 0; + *MetaRowByte = 0; + } + + if (SurfaceTiling == dm_sw_linear || SurfaceTiling == dm_sw_gfx7_2d_thin_gl || SurfaceTiling == dm_sw_gfx7_2d_thin_l_vp) { + MacroTileSizeBytes = 256; + MacroTileHeight = BlockHeight256Bytes; + } else if (SurfaceTiling == dm_sw_4kb_s || SurfaceTiling == dm_sw_4kb_s_x + || SurfaceTiling == dm_sw_4kb_d || SurfaceTiling == dm_sw_4kb_d_x) { + MacroTileSizeBytes = 4096; + MacroTileHeight = 4 * BlockHeight256Bytes; + } else if (SurfaceTiling == dm_sw_64kb_s || SurfaceTiling == dm_sw_64kb_s_t + || SurfaceTiling == dm_sw_64kb_s_x || SurfaceTiling == dm_sw_64kb_d + || SurfaceTiling == dm_sw_64kb_d_t || SurfaceTiling == dm_sw_64kb_d_x + || SurfaceTiling == dm_sw_64kb_r_x) { + MacroTileSizeBytes = 65536; + MacroTileHeight = 16 * BlockHeight256Bytes; + } else { + MacroTileSizeBytes = 262144; + MacroTileHeight = 32 * BlockHeight256Bytes; + } + *MacroTileWidth = MacroTileSizeBytes / BytePerPixel / MacroTileHeight; + + if (GPUVMEnable == true && (mode_lib->vba.GPUVMMaxPageTableLevels + 1) * (mode_lib->vba.HostVMMaxPageTableLevels + 1) > 2) { + if (ScanDirection == dm_horz) { + *DPDE0BytesFrame = 64 * (dml_ceil(((Pitch * (dml_ceil(ViewportHeight - 1, MacroTileHeight) + MacroTileHeight) * BytePerPixel) - MacroTileSizeBytes) / (8 * 2097152), 1) + 1); + } else { + *DPDE0BytesFrame = 64 * (dml_ceil(((Pitch * (dml_ceil((double) SwathWidth - 1, MacroTileHeight) + MacroTileHeight) * BytePerPixel) - MacroTileSizeBytes) / (8 * 2097152), 1) + 1); + } + ExtraDPDEBytesFrame = 128 * ((mode_lib->vba.GPUVMMaxPageTableLevels + 1) * (mode_lib->vba.HostVMMaxPageTableLevels + 1) - 3); + } else { + *DPDE0BytesFrame = 0; + ExtraDPDEBytesFrame = 0; + } + + PDEAndMetaPTEBytesFrame = *MetaPTEBytesFrame + MPDEBytesFrame + *DPDE0BytesFrame + + ExtraDPDEBytesFrame; + + if (HostVMEnable == true) { + PDEAndMetaPTEBytesFrame = PDEAndMetaPTEBytesFrame * (1 + 8 * (HostVMMaxPageTableLevels - HostVMCachedPageTableLevels)); + } + + if (GPUVMEnable == true) { + double FractionOfPTEReturnDrop; + + if (SurfaceTiling == dm_sw_linear) { + PixelPTEReqHeightPTEs = 1; + *PixelPTEReqHeight = 1; + *PixelPTEReqWidth = 8.0 * VMMPageSize / BytePerPixel; + *PTERequestSize = 64; + FractionOfPTEReturnDrop = 0; + } else if (MacroTileSizeBytes == 4096) { + PixelPTEReqHeightPTEs = 1; + *PixelPTEReqHeight = MacroTileHeight; + *PixelPTEReqWidth = 8 * *MacroTileWidth; + *PTERequestSize = 64; + if (ScanDirection == dm_horz) + FractionOfPTEReturnDrop = 0; + else + FractionOfPTEReturnDrop = 7 / 8; + } else if (VMMPageSize == 4096 && MacroTileSizeBytes > 4096) { + PixelPTEReqHeightPTEs = 16; + *PixelPTEReqHeight = 16 * BlockHeight256Bytes; + *PixelPTEReqWidth = 16 * BlockWidth256Bytes; + *PTERequestSize = 128; + FractionOfPTEReturnDrop = 0; + } else { + PixelPTEReqHeightPTEs = 1; + *PixelPTEReqHeight = MacroTileHeight; + *PixelPTEReqWidth = 8 * *MacroTileWidth; + *PTERequestSize = 64; + FractionOfPTEReturnDrop = 0; + } + + if (SurfaceTiling == dm_sw_linear) { + *dpte_row_height = dml_min(128, + 1 << (unsigned int) dml_floor( + dml_log2( + (double) PTEBufferSizeInRequests * *PixelPTEReqWidth / Pitch), + 1)); + *dpte_row_width_ub = (dml_ceil((double) (Pitch * *dpte_row_height - 1) / *PixelPTEReqWidth, 1) + 1) * *PixelPTEReqWidth; + *PixelPTEBytesPerRow = *dpte_row_width_ub / *PixelPTEReqWidth * *PTERequestSize; + } else if (ScanDirection == dm_horz) { + *dpte_row_height = *PixelPTEReqHeight; + *dpte_row_width_ub = (dml_ceil((double) (SwathWidth - 1) / *PixelPTEReqWidth, 1) + 1) * *PixelPTEReqWidth; + *PixelPTEBytesPerRow = *dpte_row_width_ub / *PixelPTEReqWidth * *PTERequestSize; + } else { + *dpte_row_height = dml_min(*PixelPTEReqWidth, *MacroTileWidth); + *dpte_row_width_ub = (dml_ceil((double) (SwathWidth - 1) / *PixelPTEReqHeight, 1) + 1) * *PixelPTEReqHeight; + *PixelPTEBytesPerRow = *dpte_row_width_ub / *PixelPTEReqHeight * *PTERequestSize; + } + if (*PixelPTEBytesPerRow * (1 - FractionOfPTEReturnDrop) + <= 64 * PTEBufferSizeInRequests) { + *PTEBufferSizeNotExceeded = true; + } else { + *PTEBufferSizeNotExceeded = false; + } + } else { + *PixelPTEBytesPerRow = 0; + *PTEBufferSizeNotExceeded = true; + } + dml_print("DML: vm_bytes = meta_pte_bytes_per_frame (per_pipe) = MetaPTEBytesFrame = : %d\n", *MetaPTEBytesFrame); + + if (HostVMEnable == true) { + *PixelPTEBytesPerRow = *PixelPTEBytesPerRow * (1 + 8 * (HostVMMaxPageTableLevels - HostVMCachedPageTableLevels)); + } + + if (HostVMEnable == true) { + *vm_group_bytes = 512; + *dpte_group_bytes = 512; + } else if (GPUVMEnable == true) { + *vm_group_bytes = 2048; + if (SurfaceTiling != dm_sw_linear && PixelPTEReqHeightPTEs == 1 && ScanDirection != dm_horz) { + *dpte_group_bytes = 512; + } else { + *dpte_group_bytes = 2048; + } + } else { + *vm_group_bytes = 0; + *dpte_group_bytes = 0; + } + + return PDEAndMetaPTEBytesFrame; +} + +static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation( + struct display_mode_lib *mode_lib) +{ + struct vba_vars_st *locals = &mode_lib->vba; + unsigned int j, k; + + mode_lib->vba.WritebackDISPCLK = 0.0; + mode_lib->vba.DISPCLKWithRamping = 0; + mode_lib->vba.DISPCLKWithoutRamping = 0; + mode_lib->vba.GlobalDPPCLK = 0.0; + + // DISPCLK and DPPCLK Calculation + // + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.WritebackEnable[k]) { + mode_lib->vba.WritebackDISPCLK = + dml_max( + mode_lib->vba.WritebackDISPCLK, + CalculateWriteBackDISPCLK( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.PixelClock[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.WritebackChromaLineBufferWidth)); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.HRatio[k] > 1) { + locals->PSCL_THROUGHPUT_LUMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / dml_ceil( + mode_lib->vba.htaps[k] + / 6.0, + 1)); + } else { + locals->PSCL_THROUGHPUT_LUMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + + mode_lib->vba.DPPCLKUsingSingleDPPLuma = + mode_lib->vba.PixelClock[k] + * dml_max( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + dml_max( + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_THROUGHPUT_LUMA[k], + 1.0)); + + if ((mode_lib->vba.htaps[k] > 6 || mode_lib->vba.vtaps[k] > 6) + && mode_lib->vba.DPPCLKUsingSingleDPPLuma + < 2 * mode_lib->vba.PixelClock[k]) { + mode_lib->vba.DPPCLKUsingSingleDPPLuma = 2 * mode_lib->vba.PixelClock[k]; + } + + if ((mode_lib->vba.SourcePixelFormat[k] != dm_420_8 + && mode_lib->vba.SourcePixelFormat[k] != dm_420_10)) { + locals->PSCL_THROUGHPUT_CHROMA[k] = 0.0; + locals->DPPCLKUsingSingleDPP[k] = + mode_lib->vba.DPPCLKUsingSingleDPPLuma; + } else { + if (mode_lib->vba.HRatio[k] > 1) { + locals->PSCL_THROUGHPUT_CHROMA[k] = + dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / 2 + / dml_ceil( + mode_lib->vba.HTAPsChroma[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_THROUGHPUT_CHROMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + mode_lib->vba.DPPCLKUsingSingleDPPChroma = + mode_lib->vba.PixelClock[k] + * dml_max( + mode_lib->vba.VTAPsChroma[k] + / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k] + / 2), + dml_max( + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / 4 + / locals->PSCL_THROUGHPUT_CHROMA[k], + 1.0)); + + if ((mode_lib->vba.HTAPsChroma[k] > 6 || mode_lib->vba.VTAPsChroma[k] > 6) + && mode_lib->vba.DPPCLKUsingSingleDPPChroma + < 2 * mode_lib->vba.PixelClock[k]) { + mode_lib->vba.DPPCLKUsingSingleDPPChroma = 2 + * mode_lib->vba.PixelClock[k]; + } + + locals->DPPCLKUsingSingleDPP[k] = dml_max( + mode_lib->vba.DPPCLKUsingSingleDPPLuma, + mode_lib->vba.DPPCLKUsingSingleDPPChroma); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] != k) + continue; + if (mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DISPCLKWithRamping = + dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.PixelClock[k] / 2 + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100) + * (1 + + mode_lib->vba.DISPCLKRampingMargin + / 100)); + mode_lib->vba.DISPCLKWithoutRamping = + dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.PixelClock[k] / 2 + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100)); + } else if (!mode_lib->vba.ODMCombineEnabled[k]) { + mode_lib->vba.DISPCLKWithRamping = + dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.PixelClock[k] + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100) + * (1 + + mode_lib->vba.DISPCLKRampingMargin + / 100)); + mode_lib->vba.DISPCLKWithoutRamping = + dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.PixelClock[k] + * (1 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100)); + } + } + + mode_lib->vba.DISPCLKWithRamping = dml_max( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.WritebackDISPCLK); + mode_lib->vba.DISPCLKWithoutRamping = dml_max( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.WritebackDISPCLK); + + ASSERT(mode_lib->vba.DISPCLKDPPCLKVCOSpeed != 0); + mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity = RoundToDFSGranularityUp( + mode_lib->vba.DISPCLKWithRamping, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity = RoundToDFSGranularityUp( + mode_lib->vba.DISPCLKWithoutRamping, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.MaxDispclkRoundedToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.soc.clock_limits[mode_lib->vba.soc.num_states - 1].dispclk_mhz, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + if (mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity + > mode_lib->vba.MaxDispclkRoundedToDFSGranularity) { + mode_lib->vba.DISPCLK_calculated = + mode_lib->vba.DISPCLKWithoutRampingRoundedToDFSGranularity; + } else if (mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity + > mode_lib->vba.MaxDispclkRoundedToDFSGranularity) { + mode_lib->vba.DISPCLK_calculated = mode_lib->vba.MaxDispclkRoundedToDFSGranularity; + } else { + mode_lib->vba.DISPCLK_calculated = + mode_lib->vba.DISPCLKWithRampingRoundedToDFSGranularity; + } + DTRACE(" dispclk_mhz (calculated) = %f", mode_lib->vba.DISPCLK_calculated); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.DPPCLK_calculated[k] = locals->DPPCLKUsingSingleDPP[k] + / mode_lib->vba.DPPPerPlane[k] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100); + mode_lib->vba.GlobalDPPCLK = dml_max( + mode_lib->vba.GlobalDPPCLK, + mode_lib->vba.DPPCLK_calculated[k]); + } + mode_lib->vba.GlobalDPPCLK = RoundToDFSGranularityUp( + mode_lib->vba.GlobalDPPCLK, + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.DPPCLK_calculated[k] = mode_lib->vba.GlobalDPPCLK / 255 + * dml_ceil( + mode_lib->vba.DPPCLK_calculated[k] * 255 + / mode_lib->vba.GlobalDPPCLK, + 1); + DTRACE(" dppclk_mhz[%i] (calculated) = %f", k, mode_lib->vba.DPPCLK_calculated[k]); + } + + // Urgent and B P-State/DRAM Clock Change Watermark + DTRACE(" dcfclk_mhz = %f", mode_lib->vba.DCFCLK); + DTRACE(" return_bw_to_dcn = %f", mode_lib->vba.ReturnBandwidthToDCN); + DTRACE(" return_bus_bw = %f", mode_lib->vba.ReturnBW); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + bool MainPlaneDoesODMCombine = false; + + if (mode_lib->vba.SourceScan[k] == dm_horz) + locals->SwathWidthSingleDPPY[k] = mode_lib->vba.ViewportWidth[k]; + else + locals->SwathWidthSingleDPPY[k] = mode_lib->vba.ViewportHeight[k]; + + if (mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) + MainPlaneDoesODMCombine = true; + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) + if (mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) + MainPlaneDoesODMCombine = true; + + if (MainPlaneDoesODMCombine == true) + locals->SwathWidthY[k] = dml_min( + (double) locals->SwathWidthSingleDPPY[k], + dml_round( + mode_lib->vba.HActive[k] / 2.0 + * mode_lib->vba.HRatio[k])); + else + locals->SwathWidthY[k] = locals->SwathWidthSingleDPPY[k] + / mode_lib->vba.DPPPerPlane[k]; + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + locals->BytePerPixelDETY[k] = 8; + locals->BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + locals->BytePerPixelDETY[k] = 4; + locals->BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16 || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16) { + locals->BytePerPixelDETY[k] = 2; + locals->BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8 || mode_lib->vba.SourcePixelFormat[k] == dm_mono_8) { + locals->BytePerPixelDETY[k] = 1; + locals->BytePerPixelDETC[k] = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + locals->BytePerPixelDETY[k] = 1; + locals->BytePerPixelDETC[k] = 2; + } else { // dm_420_10 + locals->BytePerPixelDETY[k] = 4.0 / 3.0; + locals->BytePerPixelDETC[k] = 8.0 / 3.0; + } + } + + mode_lib->vba.TotalDataReadBandwidth = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + locals->ReadBandwidthPlaneLuma[k] = locals->SwathWidthSingleDPPY[k] + * dml_ceil(locals->BytePerPixelDETY[k], 1) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k]; + locals->ReadBandwidthPlaneChroma[k] = locals->SwathWidthSingleDPPY[k] + / 2 * dml_ceil(locals->BytePerPixelDETC[k], 2) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) + * mode_lib->vba.VRatio[k] / 2; + DTRACE( + " read_bw[%i] = %fBps", + k, + locals->ReadBandwidthPlaneLuma[k] + + locals->ReadBandwidthPlaneChroma[k]); + mode_lib->vba.TotalDataReadBandwidth += locals->ReadBandwidthPlaneLuma[k] + + locals->ReadBandwidthPlaneChroma[k]; + } + + // DCFCLK Deep Sleep + CalculateDCFCLKDeepSleep( + mode_lib, + mode_lib->vba.NumberOfActivePlanes, + locals->BytePerPixelDETY, + locals->BytePerPixelDETC, + mode_lib->vba.VRatio, + locals->SwathWidthY, + mode_lib->vba.DPPPerPlane, + mode_lib->vba.HRatio, + mode_lib->vba.PixelClock, + locals->PSCL_THROUGHPUT_LUMA, + locals->PSCL_THROUGHPUT_CHROMA, + locals->DPPCLK, + &mode_lib->vba.DCFCLKDeepSleep); + + // DSCCLK + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if ((mode_lib->vba.BlendingAndTiming[k] != k) || !mode_lib->vba.DSCEnabled[k]) { + locals->DSCCLK_calculated[k] = 0.0; + } else { + if (mode_lib->vba.OutputFormat[k] == dm_420 + || mode_lib->vba.OutputFormat[k] == dm_n422) + mode_lib->vba.DSCFormatFactor = 2; + else + mode_lib->vba.DSCFormatFactor = 1; + if (mode_lib->vba.ODMCombineEnabled[k]) + locals->DSCCLK_calculated[k] = + mode_lib->vba.PixelClockBackEnd[k] / 6 + / mode_lib->vba.DSCFormatFactor + / (1 + - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100); + else + locals->DSCCLK_calculated[k] = + mode_lib->vba.PixelClockBackEnd[k] / 3 + / mode_lib->vba.DSCFormatFactor + / (1 + - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100); + } + } + + // DSC Delay + // TODO + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + double bpp = mode_lib->vba.OutputBpp[k]; + unsigned int slices = mode_lib->vba.NumberOfDSCSlices[k]; + + if (mode_lib->vba.DSCEnabled[k] && bpp != 0) { + if (!mode_lib->vba.ODMCombineEnabled[k]) { + locals->DSCDelay[k] = + dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + bpp, + dml_ceil( + (double) mode_lib->vba.HActive[k] + / mode_lib->vba.NumberOfDSCSlices[k], + 1), + slices, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k]); + } else { + locals->DSCDelay[k] = + 2 + * (dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + bpp, + dml_ceil( + (double) mode_lib->vba.HActive[k] + / mode_lib->vba.NumberOfDSCSlices[k], + 1), + slices / 2.0, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k])); + } + locals->DSCDelay[k] = locals->DSCDelay[k] + * mode_lib->vba.PixelClock[k] + / mode_lib->vba.PixelClockBackEnd[k]; + } else { + locals->DSCDelay[k] = 0; + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) // NumberOfPlanes + if (j != k && mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.DSCEnabled[j]) + locals->DSCDelay[k] = locals->DSCDelay[j]; + + // Prefetch + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + unsigned int PDEAndMetaPTEBytesFrameY; + unsigned int PixelPTEBytesPerRowY; + unsigned int MetaRowByteY; + unsigned int MetaRowByteC; + unsigned int PDEAndMetaPTEBytesFrameC; + unsigned int PixelPTEBytesPerRowC; + bool PTEBufferSizeNotExceededY; + bool PTEBufferSizeNotExceededC; + + Calculate256BBlockSizes( + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelDETY[k], 1), + dml_ceil(locals->BytePerPixelDETC[k], 2), + &locals->BlockHeight256BytesY[k], + &locals->BlockHeight256BytesC[k], + &locals->BlockWidth256BytesY[k], + &locals->BlockWidth256BytesC[k]); + + locals->PrefetchSourceLinesY[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.vtaps[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.ViewportYStartY[k], + &locals->VInitPreFillY[k], + &locals->MaxNumSwathY[k]); + + if ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_8)) { + PDEAndMetaPTEBytesFrameC = + CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + locals->BlockHeight256BytesC[k], + locals->BlockWidth256BytesC[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil( + locals->BytePerPixelDETC[k], + 2), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k] / 2, + mode_lib->vba.ViewportHeight[k] / 2, + locals->SwathWidthY[k] / 2, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsChroma, + mode_lib->vba.PitchC[k], + mode_lib->vba.DCCMetaPitchC[k], + &locals->MacroTileWidthC[k], + &MetaRowByteC, + &PixelPTEBytesPerRowC, + &PTEBufferSizeNotExceededC, + &locals->dpte_row_width_chroma_ub[k], + &locals->dpte_row_height_chroma[k], + &locals->meta_req_width_chroma[k], + &locals->meta_req_height_chroma[k], + &locals->meta_row_width_chroma[k], + &locals->meta_row_height_chroma[k], + &locals->vm_group_bytes_chroma, + &locals->dpte_group_bytes_chroma, + &locals->PixelPTEReqWidthC[k], + &locals->PixelPTEReqHeightC[k], + &locals->PTERequestSizeC[k], + &locals->dpde0_bytes_per_frame_ub_c[k], + &locals->meta_pte_bytes_per_frame_ub_c[k]); + + locals->PrefetchSourceLinesC[k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k] / 2, + mode_lib->vba.VTAPsChroma[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + mode_lib->vba.SwathHeightC[k], + mode_lib->vba.ViewportYStartC[k], + &locals->VInitPreFillC[k], + &locals->MaxNumSwathC[k]); + } else { + PixelPTEBytesPerRowC = 0; + PDEAndMetaPTEBytesFrameC = 0; + MetaRowByteC = 0; + locals->MaxNumSwathC[k] = 0; + locals->PrefetchSourceLinesC[k] = 0; + locals->PTEBufferSizeInRequestsForLuma = mode_lib->vba.PTEBufferSizeInRequestsLuma + mode_lib->vba.PTEBufferSizeInRequestsChroma; + } + + PDEAndMetaPTEBytesFrameY = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + locals->BlockHeight256BytesY[k], + locals->BlockWidth256BytesY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelDETY[k], 1), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + locals->SwathWidthY[k], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.VMMPageSize, + locals->PTEBufferSizeInRequestsForLuma, + mode_lib->vba.PitchY[k], + mode_lib->vba.DCCMetaPitchY[k], + &locals->MacroTileWidthY[k], + &MetaRowByteY, + &PixelPTEBytesPerRowY, + &PTEBufferSizeNotExceededY, + &locals->dpte_row_width_luma_ub[k], + &locals->dpte_row_height[k], + &locals->meta_req_width[k], + &locals->meta_req_height[k], + &locals->meta_row_width[k], + &locals->meta_row_height[k], + &locals->vm_group_bytes[k], + &locals->dpte_group_bytes[k], + &locals->PixelPTEReqWidthY[k], + &locals->PixelPTEReqHeightY[k], + &locals->PTERequestSizeY[k], + &locals->dpde0_bytes_per_frame_ub_l[k], + &locals->meta_pte_bytes_per_frame_ub_l[k]); + + locals->PixelPTEBytesPerRow[k] = PixelPTEBytesPerRowY + PixelPTEBytesPerRowC; + locals->PDEAndMetaPTEBytesFrame[k] = PDEAndMetaPTEBytesFrameY + + PDEAndMetaPTEBytesFrameC; + locals->MetaRowByte[k] = MetaRowByteY + MetaRowByteC; + + CalculateActiveRowBandwidth( + mode_lib->vba.GPUVMEnable, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + MetaRowByteY, + MetaRowByteC, + locals->meta_row_height[k], + locals->meta_row_height_chroma[k], + PixelPTEBytesPerRowY, + PixelPTEBytesPerRowC, + locals->dpte_row_height[k], + locals->dpte_row_height_chroma[k], + &locals->meta_row_bw[k], + &locals->dpte_row_bw[k]); + } + + mode_lib->vba.TotalDCCActiveDPP = 0; + mode_lib->vba.TotalActiveDPP = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.TotalActiveDPP = mode_lib->vba.TotalActiveDPP + + mode_lib->vba.DPPPerPlane[k]; + if (mode_lib->vba.DCCEnable[k]) + mode_lib->vba.TotalDCCActiveDPP = mode_lib->vba.TotalDCCActiveDPP + + mode_lib->vba.DPPPerPlane[k]; + } + + mode_lib->vba.UrgentOutOfOrderReturnPerChannel = dml_max3( + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelDataOnly, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelMixedWithVMData, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelVMDataOnly); + + mode_lib->vba.UrgentRoundTripAndOutOfOrderLatency = + (mode_lib->vba.RoundTripPingLatencyCycles + 32) / mode_lib->vba.DCFCLK + + mode_lib->vba.UrgentOutOfOrderReturnPerChannel + * mode_lib->vba.NumberOfChannels + / mode_lib->vba.ReturnBW; + + mode_lib->vba.UrgentExtraLatency = CalculateExtraLatency( + mode_lib->vba.UrgentRoundTripAndOutOfOrderLatency, + mode_lib->vba.TotalActiveDPP, + mode_lib->vba.PixelChunkSizeInKByte, + mode_lib->vba.TotalDCCActiveDPP, + mode_lib->vba.MetaChunkSize, + mode_lib->vba.ReturnBW, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.DPPPerPlane, + locals->dpte_group_bytes, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels); + + + mode_lib->vba.TCalc = 24.0 / mode_lib->vba.DCFCLKDeepSleep; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] = + mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k]) + / mode_lib->vba.DISPCLK; + } else + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] = 0; + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { + if (mode_lib->vba.BlendingAndTiming[j] == k + && mode_lib->vba.WritebackEnable[j] == true) { + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] = + dml_max( + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k], + mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[j], + mode_lib->vba.WritebackHRatio[j], + mode_lib->vba.WritebackVRatio[j], + mode_lib->vba.WritebackLumaHTaps[j], + mode_lib->vba.WritebackLumaVTaps[j], + mode_lib->vba.WritebackChromaHTaps[j], + mode_lib->vba.WritebackChromaVTaps[j], + mode_lib->vba.WritebackDestinationWidth[j]) + / mode_lib->vba.DISPCLK); + } + } + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) + if (mode_lib->vba.BlendingAndTiming[k] == j) + locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] = + locals->WritebackDelay[mode_lib->vba.VoltageLevel][j]; + + mode_lib->vba.VStartupLines = 13; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + locals->MaxVStartupLines[k] = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] - dml_max(1.0, dml_ceil(locals->WritebackDelay[mode_lib->vba.VoltageLevel][k] / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]), 1)); + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) + locals->MaximumMaxVStartupLines = dml_max(locals->MaximumMaxVStartupLines, locals->MaxVStartupLines[k]); + + // We don't really care to iterate between the various prefetch modes + //mode_lib->vba.PrefetchERROR = CalculateMinAndMaxPrefetchMode(mode_lib->vba.AllowDRAMSelfRefreshOrDRAMClockChangeInVblank, &mode_lib->vba.MinPrefetchMode, &mode_lib->vba.MaxPrefetchMode); + mode_lib->vba.UrgentLatency = dml_max3(mode_lib->vba.UrgentLatencyPixelDataOnly, mode_lib->vba.UrgentLatencyPixelMixedWithVMData, mode_lib->vba.UrgentLatencyVMDataOnly); + + do { + double MaxTotalRDBandwidth = 0; + double MaxTotalRDBandwidthNoUrgentBurst = 0; + bool DestinationLineTimesForPrefetchLessThan2 = false; + bool VRatioPrefetchMoreThan4 = false; + double TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.SREnterPlusExitTime); + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + Pipe myPipe; + HostVM myHostVM; + + if (mode_lib->vba.XFCEnabled[k] == true) { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = + CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + locals->SwathWidthY[k], + dml_ceil( + locals->BytePerPixelDETY[k], + 1), + mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TCalc, + TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + } else { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = 0; + } + + myPipe.DPPCLK = locals->DPPCLK[k]; + myPipe.DISPCLK = mode_lib->vba.DISPCLK; + myPipe.PixelClock = mode_lib->vba.PixelClock[k]; + myPipe.DCFCLKDeepSleep = mode_lib->vba.DCFCLKDeepSleep; + myPipe.DPPPerPlane = mode_lib->vba.DPPPerPlane[k]; + myPipe.ScalerEnabled = mode_lib->vba.ScalerEnabled[k]; + myPipe.SourceScan = mode_lib->vba.SourceScan[k]; + myPipe.BlockWidth256BytesY = locals->BlockWidth256BytesY[k]; + myPipe.BlockHeight256BytesY = locals->BlockHeight256BytesY[k]; + myPipe.BlockWidth256BytesC = locals->BlockWidth256BytesC[k]; + myPipe.BlockHeight256BytesC = locals->BlockHeight256BytesC[k]; + myPipe.InterlaceEnable = mode_lib->vba.Interlace[k]; + myPipe.NumberOfCursors = mode_lib->vba.NumberOfCursors[k]; + myPipe.VBlank = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k]; + myPipe.HTotal = mode_lib->vba.HTotal[k]; + + + myHostVM.Enable = mode_lib->vba.HostVMEnable; + myHostVM.MaxPageTableLevels = mode_lib->vba.HostVMMaxPageTableLevels; + myHostVM.CachedPageTableLevels = mode_lib->vba.HostVMCachedPageTableLevels; + + mode_lib->vba.ErrorResult[k] = + CalculatePrefetchSchedule( + mode_lib, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + &myPipe, + locals->DSCDelay[k], + mode_lib->vba.DPPCLKDelaySubtotal, + mode_lib->vba.DPPCLKDelaySCL, + mode_lib->vba.DPPCLKDelaySCLLBOnly, + mode_lib->vba.DPPCLKDelayCNVCFormater, + mode_lib->vba.DPPCLKDelayCNVCCursor, + mode_lib->vba.DISPCLKDelaySubtotal, + (unsigned int) (locals->SwathWidthY[k] + / mode_lib->vba.HRatio[k]), + mode_lib->vba.OutputFormat[k], + mode_lib->vba.MaxInterDCNTileRepeaters, + dml_min(mode_lib->vba.VStartupLines, locals->MaxVStartupLines[k]), + locals->MaxVStartupLines[k], + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + &myHostVM, + mode_lib->vba.DynamicMetadataEnable[k], + mode_lib->vba.DynamicMetadataLinesBeforeActiveRequired[k], + mode_lib->vba.DynamicMetadataTransmittedBytes[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.UrgentLatency, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.TCalc, + locals->PDEAndMetaPTEBytesFrame[k], + locals->MetaRowByte[k], + locals->PixelPTEBytesPerRow[k], + locals->PrefetchSourceLinesY[k], + locals->SwathWidthY[k], + locals->BytePerPixelDETY[k], + locals->VInitPreFillY[k], + locals->MaxNumSwathY[k], + locals->PrefetchSourceLinesC[k], + locals->BytePerPixelDETC[k], + locals->VInitPreFillC[k], + locals->MaxNumSwathC[k], + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + TWait, + mode_lib->vba.XFCEnabled[k], + mode_lib->vba.XFCRemoteSurfaceFlipDelay, + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + &locals->DSTXAfterScaler[k], + &locals->DSTYAfterScaler[k], + &locals->DestinationLinesForPrefetch[k], + &locals->PrefetchBandwidth[k], + &locals->DestinationLinesToRequestVMInVBlank[k], + &locals->DestinationLinesToRequestRowInVBlank[k], + &locals->VRatioPrefetchY[k], + &locals->VRatioPrefetchC[k], + &locals->RequiredPrefetchPixDataBWLuma[k], + &locals->RequiredPrefetchPixDataBWChroma[k], + &locals->VStartupRequiredWhenNotEnoughTimeForDynamicMetadata, + &locals->Tno_bw[k], + &locals->prefetch_vmrow_bw[k], + &locals->swath_width_luma_ub[k], + &locals->swath_width_chroma_ub[k], + &mode_lib->vba.VUpdateOffsetPix[k], + &mode_lib->vba.VUpdateWidthPix[k], + &mode_lib->vba.VReadyOffsetPix[k]); + if (mode_lib->vba.BlendingAndTiming[k] == k) { + locals->VStartup[k] = dml_min( + mode_lib->vba.VStartupLines, + locals->MaxVStartupLines[k]); + if (locals->VStartupRequiredWhenNotEnoughTimeForDynamicMetadata + != 0) { + locals->VStartup[k] = + locals->VStartupRequiredWhenNotEnoughTimeForDynamicMetadata; + } + } else { + locals->VStartup[k] = + dml_min( + mode_lib->vba.VStartupLines, + locals->MaxVStartupLines[mode_lib->vba.BlendingAndTiming[k]]); + } + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + unsigned int m; + + locals->cursor_bw[k] = 0; + locals->cursor_bw_pre[k] = 0; + for (m = 0; m < mode_lib->vba.NumberOfCursors[k]; m++) { + locals->cursor_bw[k] += mode_lib->vba.CursorWidth[k][m] * mode_lib->vba.CursorBPP[k][m] / 8.0 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k]; + locals->cursor_bw_pre[k] += mode_lib->vba.CursorWidth[k][m] * mode_lib->vba.CursorBPP[k][m] / 8.0 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * locals->VRatioPrefetchY[k]; + } + + CalculateUrgentBurstFactor( + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + locals->SwathWidthY[k], + mode_lib->vba.HTotal[k] / + mode_lib->vba.PixelClock[k], + mode_lib->vba.UrgentLatency, + mode_lib->vba.CursorBufferSize, + mode_lib->vba.CursorWidth[k][0] + mode_lib->vba.CursorWidth[k][1], + dml_max(mode_lib->vba.CursorBPP[k][0], mode_lib->vba.CursorBPP[k][1]), + mode_lib->vba.VRatio[k], + locals->VRatioPrefetchY[k], + locals->VRatioPrefetchC[k], + locals->BytePerPixelDETY[k], + locals->BytePerPixelDETC[k], + &locals->UrgentBurstFactorCursor[k], + &locals->UrgentBurstFactorCursorPre[k], + &locals->UrgentBurstFactorLuma[k], + &locals->UrgentBurstFactorLumaPre[k], + &locals->UrgentBurstFactorChroma[k], + &locals->UrgentBurstFactorChromaPre[k], + &locals->NotEnoughUrgentLatencyHiding, + &locals->NotEnoughUrgentLatencyHidingPre); + + if (mode_lib->vba.UseUrgentBurstBandwidth == false) { + locals->UrgentBurstFactorLuma[k] = 1; + locals->UrgentBurstFactorChroma[k] = 1; + locals->UrgentBurstFactorCursor[k] = 1; + locals->UrgentBurstFactorLumaPre[k] = 1; + locals->UrgentBurstFactorChromaPre[k] = 1; + locals->UrgentBurstFactorCursorPre[k] = 1; + } + + MaxTotalRDBandwidth = MaxTotalRDBandwidth + + dml_max3(locals->prefetch_vmrow_bw[k], + locals->ReadBandwidthPlaneLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthPlaneChroma[k] * locals->UrgentBurstFactorChroma[k] + locals->cursor_bw[k] + * locals->UrgentBurstFactorCursor[k] + locals->meta_row_bw[k] + locals->dpte_row_bw[k], + locals->RequiredPrefetchPixDataBWLuma[k] * locals->UrgentBurstFactorLumaPre[k] + locals->RequiredPrefetchPixDataBWChroma[k] + * locals->UrgentBurstFactorChromaPre[k] + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + + MaxTotalRDBandwidthNoUrgentBurst = MaxTotalRDBandwidthNoUrgentBurst + + dml_max3(locals->prefetch_vmrow_bw[k], + locals->ReadBandwidthPlaneLuma[k] + locals->ReadBandwidthPlaneChroma[k] + locals->cursor_bw[k] + + locals->meta_row_bw[k] + locals->dpte_row_bw[k], + locals->RequiredPrefetchPixDataBWLuma[k] + locals->RequiredPrefetchPixDataBWChroma[k] + locals->cursor_bw_pre[k]); + + if (locals->DestinationLinesForPrefetch[k] < 2) + DestinationLineTimesForPrefetchLessThan2 = true; + if (locals->VRatioPrefetchY[k] > 4 || locals->VRatioPrefetchC[k] > 4) + VRatioPrefetchMoreThan4 = true; + } + mode_lib->vba.FractionOfUrgentBandwidth = MaxTotalRDBandwidthNoUrgentBurst / mode_lib->vba.ReturnBW; + + if (MaxTotalRDBandwidth <= mode_lib->vba.ReturnBW && locals->NotEnoughUrgentLatencyHiding == 0 && locals->NotEnoughUrgentLatencyHidingPre == 0 && !VRatioPrefetchMoreThan4 + && !DestinationLineTimesForPrefetchLessThan2) + mode_lib->vba.PrefetchModeSupported = true; + else { + mode_lib->vba.PrefetchModeSupported = false; + dml_print( + "DML: CalculatePrefetchSchedule ***failed***. Bandwidth violation. Results are NOT valid\n"); + } + + if (mode_lib->vba.PrefetchModeSupported == true) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = mode_lib->vba.ReturnBW; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = + mode_lib->vba.BandwidthAvailableForImmediateFlip + - dml_max( + locals->ReadBandwidthPlaneLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthPlaneChroma[k] * locals->UrgentBurstFactorChroma[k] + + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k], + locals->RequiredPrefetchPixDataBWLuma[k] * locals->UrgentBurstFactorLumaPre[k] + + locals->RequiredPrefetchPixDataBWChroma[k] * locals->UrgentBurstFactorChromaPre[k] + + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + } + + mode_lib->vba.TotImmediateFlipBytes = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.TotImmediateFlipBytes = mode_lib->vba.TotImmediateFlipBytes + locals->PDEAndMetaPTEBytesFrame[k] + locals->MetaRowByte[k] + locals->PixelPTEBytesPerRow[k]; + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + CalculateFlipSchedule( + mode_lib, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.GPUVMEnable, + locals->PDEAndMetaPTEBytesFrame[k], + locals->MetaRowByte[k], + locals->PixelPTEBytesPerRow[k], + mode_lib->vba.BandwidthAvailableForImmediateFlip, + mode_lib->vba.TotImmediateFlipBytes, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.VRatio[k], + locals->Tno_bw[k], + mode_lib->vba.DCCEnable[k], + locals->dpte_row_height[k], + locals->meta_row_height[k], + locals->dpte_row_height_chroma[k], + locals->meta_row_height_chroma[k], + &locals->DestinationLinesToRequestVMInImmediateFlip[k], + &locals->DestinationLinesToRequestRowInImmediateFlip[k], + &locals->final_flip_bw[k], + &locals->ImmediateFlipSupportedForPipe[k]); + } + mode_lib->vba.total_dcn_read_bw_with_flip = 0.0; + mode_lib->vba.total_dcn_read_bw_with_flip_no_urgent_burst = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + mode_lib->vba.total_dcn_read_bw_with_flip = + mode_lib->vba.total_dcn_read_bw_with_flip + dml_max3( + locals->prefetch_vmrow_bw[k], + locals->final_flip_bw[k] + locals->ReadBandwidthLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthChroma[k] * locals->UrgentBurstFactorChroma[k] + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k], + locals->final_flip_bw[k] + locals->RequiredPrefetchPixDataBWLuma[k] * locals->UrgentBurstFactorLumaPre[k] + + locals->RequiredPrefetchPixDataBWChroma[k] * locals->UrgentBurstFactorChromaPre[k] + + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + mode_lib->vba.total_dcn_read_bw_with_flip_no_urgent_burst = + mode_lib->vba.total_dcn_read_bw_with_flip_no_urgent_burst + + dml_max3(locals->prefetch_vmrow_bw[k], + locals->final_flip_bw[k] + locals->ReadBandwidthPlaneLuma[k] + locals->ReadBandwidthPlaneChroma[k] + locals->cursor_bw[k], + locals->final_flip_bw[k] + locals->RequiredPrefetchPixDataBWLuma[k] + locals->RequiredPrefetchPixDataBWChroma[k] + locals->cursor_bw_pre[k]); + + } + mode_lib->vba.FractionOfUrgentBandwidthImmediateFlip = mode_lib->vba.total_dcn_read_bw_with_flip_no_urgent_burst / mode_lib->vba.ReturnBW; + + mode_lib->vba.ImmediateFlipSupported = true; + if (mode_lib->vba.total_dcn_read_bw_with_flip > mode_lib->vba.ReturnBW) { + mode_lib->vba.ImmediateFlipSupported = false; + } + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->ImmediateFlipSupportedForPipe[k] == false) { + mode_lib->vba.ImmediateFlipSupported = false; + } + } + } else { + mode_lib->vba.ImmediateFlipSupported = false; + } + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ErrorResult[k]) { + mode_lib->vba.PrefetchModeSupported = false; + dml_print( + "DML: CalculatePrefetchSchedule ***failed***. Prefetch schedule violation. Results are NOT valid\n"); + } + } + + mode_lib->vba.VStartupLines = mode_lib->vba.VStartupLines + 1; + } while (!((mode_lib->vba.PrefetchModeSupported + && ((!mode_lib->vba.ImmediateFlipSupport && !mode_lib->vba.HostVMEnable) + || mode_lib->vba.ImmediateFlipSupported)) + || locals->MaximumMaxVStartupLines < mode_lib->vba.VStartupLines)); + + //Watermarks and NB P-State/DRAM Clock Change Support + { + enum clock_change_support DRAMClockChangeSupport; // dummy + CalculateWatermarksAndDRAMSpeedChangeSupport( + mode_lib, + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.MaxLineBufferLines, + mode_lib->vba.LineBufferSize, + mode_lib->vba.DPPOutputBufferPixels, + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.WritebackInterfaceLumaBufferSize, + mode_lib->vba.WritebackInterfaceChromaBufferSize, + mode_lib->vba.DCFCLK, + mode_lib->vba.UrgentOutOfOrderReturnPerChannel * mode_lib->vba.NumberOfChannels, + mode_lib->vba.ReturnBW, + mode_lib->vba.GPUVMEnable, + locals->dpte_group_bytes, + mode_lib->vba.MetaChunkSize, + mode_lib->vba.UrgentLatency, + mode_lib->vba.UrgentExtraLatency, + mode_lib->vba.WritebackLatency, + mode_lib->vba.WritebackChunkSize, + mode_lib->vba.SOCCLK, + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.SRExitTime, + mode_lib->vba.SREnterPlusExitTime, + mode_lib->vba.DCFCLKDeepSleep, + mode_lib->vba.DPPPerPlane, + mode_lib->vba.DCCEnable, + locals->DPPCLK, + locals->SwathWidthSingleDPPY, + mode_lib->vba.SwathHeightY, + locals->ReadBandwidthPlaneLuma, + mode_lib->vba.SwathHeightC, + locals->ReadBandwidthPlaneChroma, + mode_lib->vba.LBBitPerPixel, + locals->SwathWidthY, + mode_lib->vba.HRatio, + mode_lib->vba.vtaps, + mode_lib->vba.VTAPsChroma, + mode_lib->vba.VRatio, + mode_lib->vba.HTotal, + mode_lib->vba.PixelClock, + mode_lib->vba.BlendingAndTiming, + locals->BytePerPixelDETY, + locals->BytePerPixelDETC, + mode_lib->vba.WritebackEnable, + mode_lib->vba.WritebackPixelFormat, + mode_lib->vba.WritebackDestinationWidth, + mode_lib->vba.WritebackDestinationHeight, + mode_lib->vba.WritebackSourceHeight, + &DRAMClockChangeSupport, + &mode_lib->vba.UrgentWatermark, + &mode_lib->vba.WritebackUrgentWatermark, + &mode_lib->vba.DRAMClockChangeWatermark, + &mode_lib->vba.WritebackDRAMClockChangeWatermark, + &mode_lib->vba.StutterExitWatermark, + &mode_lib->vba.StutterEnterPlusExitWatermark, + &mode_lib->vba.MinActiveDRAMClockChangeLatencySupported); + } + + + //Display Pipeline Delivery Time in Prefetch, Groups + CalculatePixelDeliveryTimes( + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.VRatio, + locals->VRatioPrefetchY, + locals->VRatioPrefetchC, + locals->swath_width_luma_ub, + locals->swath_width_chroma_ub, + mode_lib->vba.DPPPerPlane, + mode_lib->vba.HRatio, + mode_lib->vba.PixelClock, + locals->PSCL_THROUGHPUT_LUMA, + locals->PSCL_THROUGHPUT_CHROMA, + locals->DPPCLK, + locals->BytePerPixelDETC, + mode_lib->vba.SourceScan, + locals->BlockWidth256BytesY, + locals->BlockHeight256BytesY, + locals->BlockWidth256BytesC, + locals->BlockHeight256BytesC, + locals->DisplayPipeLineDeliveryTimeLuma, + locals->DisplayPipeLineDeliveryTimeChroma, + locals->DisplayPipeLineDeliveryTimeLumaPrefetch, + locals->DisplayPipeLineDeliveryTimeChromaPrefetch, + locals->DisplayPipeRequestDeliveryTimeLuma, + locals->DisplayPipeRequestDeliveryTimeChroma, + locals->DisplayPipeRequestDeliveryTimeLumaPrefetch, + locals->DisplayPipeRequestDeliveryTimeChromaPrefetch); + + CalculateMetaAndPTETimes( + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.MetaChunkSize, + mode_lib->vba.MinMetaChunkSizeBytes, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.HTotal, + mode_lib->vba.VRatio, + locals->VRatioPrefetchY, + locals->VRatioPrefetchC, + locals->DestinationLinesToRequestRowInVBlank, + locals->DestinationLinesToRequestRowInImmediateFlip, + locals->DestinationLinesToRequestVMInVBlank, + locals->DestinationLinesToRequestVMInImmediateFlip, + mode_lib->vba.DCCEnable, + mode_lib->vba.PixelClock, + locals->BytePerPixelDETY, + locals->BytePerPixelDETC, + mode_lib->vba.SourceScan, + locals->dpte_row_height, + locals->dpte_row_height_chroma, + locals->meta_row_width, + locals->meta_row_height, + locals->meta_req_width, + locals->meta_req_height, + locals->dpte_group_bytes, + locals->PTERequestSizeY, + locals->PTERequestSizeC, + locals->PixelPTEReqWidthY, + locals->PixelPTEReqHeightY, + locals->PixelPTEReqWidthC, + locals->PixelPTEReqHeightC, + locals->dpte_row_width_luma_ub, + locals->dpte_row_width_chroma_ub, + locals->vm_group_bytes, + locals->dpde0_bytes_per_frame_ub_l, + locals->dpde0_bytes_per_frame_ub_c, + locals->meta_pte_bytes_per_frame_ub_l, + locals->meta_pte_bytes_per_frame_ub_c, + locals->DST_Y_PER_PTE_ROW_NOM_L, + locals->DST_Y_PER_PTE_ROW_NOM_C, + locals->DST_Y_PER_META_ROW_NOM_L, + locals->TimePerMetaChunkNominal, + locals->TimePerMetaChunkVBlank, + locals->TimePerMetaChunkFlip, + locals->time_per_pte_group_nom_luma, + locals->time_per_pte_group_vblank_luma, + locals->time_per_pte_group_flip_luma, + locals->time_per_pte_group_nom_chroma, + locals->time_per_pte_group_vblank_chroma, + locals->time_per_pte_group_flip_chroma, + locals->TimePerVMGroupVBlank, + locals->TimePerVMGroupFlip, + locals->TimePerVMRequestVBlank, + locals->TimePerVMRequestFlip); + + + // Min TTUVBlank + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 0) { + locals->AllowDRAMClockChangeDuringVBlank[k] = true; + locals->AllowDRAMSelfRefreshDuringVBlank[k] = true; + locals->MinTTUVBlank[k] = dml_max( + mode_lib->vba.DRAMClockChangeWatermark, + dml_max( + mode_lib->vba.StutterEnterPlusExitWatermark, + mode_lib->vba.UrgentWatermark)); + } else if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 1) { + locals->AllowDRAMClockChangeDuringVBlank[k] = false; + locals->AllowDRAMSelfRefreshDuringVBlank[k] = true; + locals->MinTTUVBlank[k] = dml_max( + mode_lib->vba.StutterEnterPlusExitWatermark, + mode_lib->vba.UrgentWatermark); + } else { + locals->AllowDRAMClockChangeDuringVBlank[k] = false; + locals->AllowDRAMSelfRefreshDuringVBlank[k] = false; + locals->MinTTUVBlank[k] = mode_lib->vba.UrgentWatermark; + } + if (!mode_lib->vba.DynamicMetadataEnable[k]) + locals->MinTTUVBlank[k] = mode_lib->vba.TCalc + + locals->MinTTUVBlank[k]; + } + + // DCC Configuration + mode_lib->vba.ActiveDPPs = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + locals->MaximumDCCCompressionYSurface[k] = CalculateDCCConfiguration( + mode_lib->vba.DCCEnable[k], + false, // We should always know the direction DCCProgrammingAssumesScanDirectionUnknown, + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + mode_lib->vba.DETBufferSizeInKByte * 1024, + locals->BlockHeight256BytesY[k], + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SurfaceTiling[k], + locals->BytePerPixelDETY[k], + mode_lib->vba.SourceScan[k], + &locals->DCCYMaxUncompressedBlock[k], + &locals->DCCYMaxCompressedBlock[k], + &locals->DCCYIndependent64ByteBlock[k]); + } + + //XFC Parameters: + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.XFCEnabled[k] == true) { + double TWait; + + locals->XFCSlaveVUpdateOffset[k] = mode_lib->vba.XFCTSlvVupdateOffset; + locals->XFCSlaveVupdateWidth[k] = mode_lib->vba.XFCTSlvVupdateWidth; + locals->XFCSlaveVReadyOffset[k] = mode_lib->vba.XFCTSlvVreadyOffset; + TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.SREnterPlusExitTime); + mode_lib->vba.XFCRemoteSurfaceFlipDelay = CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + locals->SwathWidthY[k], + dml_ceil(locals->BytePerPixelDETY[k], 1), + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TCalc, + TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + locals->XFCRemoteSurfaceFlipLatency[k] = + dml_floor( + mode_lib->vba.XFCRemoteSurfaceFlipDelay + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + locals->XFCTransferDelay[k] = + dml_ceil( + mode_lib->vba.XFCBusTransportTime + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + locals->XFCPrechargeDelay[k] = + dml_ceil( + (mode_lib->vba.XFCBusTransportTime + + mode_lib->vba.TInitXFill + + mode_lib->vba.TslvChk) + / (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]), + 1); + mode_lib->vba.InitFillLevel = mode_lib->vba.XFCXBUFLatencyTolerance + * mode_lib->vba.SrcActiveDrainRate; + mode_lib->vba.FinalFillMargin = + (locals->DestinationLinesToRequestVMInVBlank[k] + + locals->DestinationLinesToRequestRowInVBlank[k]) + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k] + * mode_lib->vba.SrcActiveDrainRate + + mode_lib->vba.XFCFillConstant; + mode_lib->vba.FinalFillLevel = mode_lib->vba.XFCRemoteSurfaceFlipDelay + * mode_lib->vba.SrcActiveDrainRate + + mode_lib->vba.FinalFillMargin; + mode_lib->vba.RemainingFillLevel = dml_max( + 0.0, + mode_lib->vba.FinalFillLevel - mode_lib->vba.InitFillLevel); + mode_lib->vba.TFinalxFill = mode_lib->vba.RemainingFillLevel + / (mode_lib->vba.SrcActiveDrainRate + * mode_lib->vba.XFCFillBWOverhead / 100); + locals->XFCPrefetchMargin[k] = + mode_lib->vba.XFCRemoteSurfaceFlipDelay + + mode_lib->vba.TFinalxFill + + (locals->DestinationLinesToRequestVMInVBlank[k] + + locals->DestinationLinesToRequestRowInVBlank[k]) + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } else { + locals->XFCSlaveVUpdateOffset[k] = 0; + locals->XFCSlaveVupdateWidth[k] = 0; + locals->XFCSlaveVReadyOffset[k] = 0; + locals->XFCRemoteSurfaceFlipLatency[k] = 0; + locals->XFCPrechargeDelay[k] = 0; + locals->XFCTransferDelay[k] = 0; + locals->XFCPrefetchMargin[k] = 0; + } + } + + // Stutter Efficiency + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + CalculateDETBufferSize( + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + &locals->DETBufferSizeY[k], + &locals->DETBufferSizeC[k]); + + locals->LinesInDETY[k] = locals->DETBufferSizeY[k] + / locals->BytePerPixelDETY[k] / locals->SwathWidthY[k]; + locals->LinesInDETYRoundedDownToSwath[k] = dml_floor( + locals->LinesInDETY[k], + mode_lib->vba.SwathHeightY[k]); + locals->FullDETBufferingTimeY[k] = + locals->LinesInDETYRoundedDownToSwath[k] + * (mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) + / mode_lib->vba.VRatio[k]; + } + + mode_lib->vba.StutterPeriod = 999999.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->FullDETBufferingTimeY[k] < mode_lib->vba.StutterPeriod) { + mode_lib->vba.StutterPeriod = locals->FullDETBufferingTimeY[k]; + mode_lib->vba.FrameTimeForMinFullDETBufferingTime = + (double) mode_lib->vba.VTotal[k] * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + locals->BytePerPixelYCriticalPlane = dml_ceil(locals->BytePerPixelDETY[k], 1); + locals->SwathWidthYCriticalPlane = locals->SwathWidthY[k]; + locals->LinesToFinishSwathTransferStutterCriticalPlane = + mode_lib->vba.SwathHeightY[k] - (locals->LinesInDETY[k] - locals->LinesInDETYRoundedDownToSwath[k]); + } + } + + mode_lib->vba.AverageReadBandwidth = 0.0; + mode_lib->vba.TotalRowReadBandwidth = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + unsigned int DCCRateLimit; + + if (mode_lib->vba.DCCEnable[k]) { + if (locals->DCCYMaxCompressedBlock[k] == 256) + DCCRateLimit = 4; + else + DCCRateLimit = 2; + + mode_lib->vba.AverageReadBandwidth = + mode_lib->vba.AverageReadBandwidth + + (locals->ReadBandwidthPlaneLuma[k] + locals->ReadBandwidthPlaneChroma[k]) / + dml_min(mode_lib->vba.DCCRate[k], DCCRateLimit); + } else { + mode_lib->vba.AverageReadBandwidth = + mode_lib->vba.AverageReadBandwidth + + locals->ReadBandwidthPlaneLuma[k] + + locals->ReadBandwidthPlaneChroma[k]; + } + mode_lib->vba.TotalRowReadBandwidth = mode_lib->vba.TotalRowReadBandwidth + + locals->meta_row_bw[k] + locals->dpte_row_bw[k]; + } + + mode_lib->vba.AverageDCCCompressionRate = mode_lib->vba.TotalDataReadBandwidth / mode_lib->vba.AverageReadBandwidth; + + mode_lib->vba.PartOfBurstThatFitsInROB = + dml_min( + mode_lib->vba.StutterPeriod + * mode_lib->vba.TotalDataReadBandwidth, + mode_lib->vba.ROBBufferSizeInKByte * 1024 + * mode_lib->vba.AverageDCCCompressionRate); + mode_lib->vba.StutterBurstTime = mode_lib->vba.PartOfBurstThatFitsInROB + / mode_lib->vba.AverageDCCCompressionRate / mode_lib->vba.ReturnBW + + (mode_lib->vba.StutterPeriod * mode_lib->vba.TotalDataReadBandwidth + - mode_lib->vba.PartOfBurstThatFitsInROB) + / (mode_lib->vba.DCFCLK * 64) + + mode_lib->vba.StutterPeriod * mode_lib->vba.TotalRowReadBandwidth / mode_lib->vba.ReturnBW; + mode_lib->vba.StutterBurstTime = dml_max( + mode_lib->vba.StutterBurstTime, + (locals->LinesToFinishSwathTransferStutterCriticalPlane * locals->BytePerPixelYCriticalPlane * + locals->SwathWidthYCriticalPlane / mode_lib->vba.ReturnBW) + ); + + mode_lib->vba.TotalActiveWriteback = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + mode_lib->vba.TotalActiveWriteback = mode_lib->vba.TotalActiveWriteback + 1; + } + } + + if (mode_lib->vba.TotalActiveWriteback == 0) { + mode_lib->vba.StutterEfficiencyNotIncludingVBlank = (1 + - (mode_lib->vba.SRExitTime + mode_lib->vba.StutterBurstTime) + / mode_lib->vba.StutterPeriod) * 100; + } else { + mode_lib->vba.StutterEfficiencyNotIncludingVBlank = 0; + } + + mode_lib->vba.SmallestVBlank = 999999; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.SynchronizedVBlank || mode_lib->vba.NumberOfActivePlanes == 1) { + mode_lib->vba.VBlankTime = (double) (mode_lib->vba.VTotal[k] + - mode_lib->vba.VActive[k]) * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]; + } else { + mode_lib->vba.VBlankTime = 0; + } + mode_lib->vba.SmallestVBlank = dml_min( + mode_lib->vba.SmallestVBlank, + mode_lib->vba.VBlankTime); + } + + mode_lib->vba.StutterEfficiency = (mode_lib->vba.StutterEfficiencyNotIncludingVBlank / 100 + * (mode_lib->vba.FrameTimeForMinFullDETBufferingTime + - mode_lib->vba.SmallestVBlank) + + mode_lib->vba.SmallestVBlank) + / mode_lib->vba.FrameTimeForMinFullDETBufferingTime * 100; +} + +static void DisplayPipeConfiguration(struct display_mode_lib *mode_lib) +{ + // Display Pipe Configuration + double BytePerPixDETY; + double BytePerPixDETC; + double Read256BytesBlockHeightY; + double Read256BytesBlockHeightC; + double Read256BytesBlockWidthY; + double Read256BytesBlockWidthC; + double MaximumSwathHeightY; + double MaximumSwathHeightC; + double MinimumSwathHeightY; + double MinimumSwathHeightC; + double SwathWidth; + double SwathWidthGranularityY; + double SwathWidthGranularityC; + double RoundedUpMaxSwathSizeBytesY; + double RoundedUpMaxSwathSizeBytesC; + unsigned int j, k; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + bool MainPlaneDoesODMCombine = false; + + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + BytePerPixDETY = 8; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + BytePerPixDETY = 4; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + BytePerPixDETY = 2; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8) { + BytePerPixDETY = 1; + BytePerPixDETC = 0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + BytePerPixDETY = 1; + BytePerPixDETC = 2; + } else { + BytePerPixDETY = 4.0 / 3.0; + BytePerPixDETC = 8.0 / 3.0; + } + + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + Read256BytesBlockHeightY = 1; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + Read256BytesBlockHeightY = 4; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16) { + Read256BytesBlockHeightY = 8; + } else { + Read256BytesBlockHeightY = 16; + } + Read256BytesBlockWidthY = 256 / dml_ceil(BytePerPixDETY, 1) + / Read256BytesBlockHeightY; + Read256BytesBlockHeightC = 0; + Read256BytesBlockWidthC = 0; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + Read256BytesBlockHeightY = 1; + Read256BytesBlockHeightC = 1; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + Read256BytesBlockHeightY = 16; + Read256BytesBlockHeightC = 8; + } else { + Read256BytesBlockHeightY = 8; + Read256BytesBlockHeightC = 8; + } + Read256BytesBlockWidthY = 256 / dml_ceil(BytePerPixDETY, 1) + / Read256BytesBlockHeightY; + Read256BytesBlockWidthC = 256 / dml_ceil(BytePerPixDETC, 2) + / Read256BytesBlockHeightC; + } + + if (mode_lib->vba.SourceScan[k] == dm_horz) { + MaximumSwathHeightY = Read256BytesBlockHeightY; + MaximumSwathHeightC = Read256BytesBlockHeightC; + } else { + MaximumSwathHeightY = Read256BytesBlockWidthY; + MaximumSwathHeightC = Read256BytesBlockWidthC; + } + + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + || (mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_t + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s_x) + && mode_lib->vba.SourceScan[k] == dm_horz)) { + MinimumSwathHeightY = MaximumSwathHeightY; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_8 + && mode_lib->vba.SourceScan[k] != dm_horz) { + MinimumSwathHeightY = MaximumSwathHeightY; + } else { + MinimumSwathHeightY = MaximumSwathHeightY / 2.0; + } + MinimumSwathHeightC = MaximumSwathHeightC; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + MinimumSwathHeightY = MaximumSwathHeightY; + MinimumSwathHeightC = MaximumSwathHeightC; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8 + && mode_lib->vba.SourceScan[k] == dm_horz) { + MinimumSwathHeightY = MaximumSwathHeightY / 2.0; + MinimumSwathHeightC = MaximumSwathHeightC; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10 + && mode_lib->vba.SourceScan[k] == dm_horz) { + MinimumSwathHeightC = MaximumSwathHeightC / 2.0; + MinimumSwathHeightY = MaximumSwathHeightY; + } else { + MinimumSwathHeightY = MaximumSwathHeightY; + MinimumSwathHeightC = MaximumSwathHeightC; + } + } + + if (mode_lib->vba.SourceScan[k] == dm_horz) { + SwathWidth = mode_lib->vba.ViewportWidth[k]; + } else { + SwathWidth = mode_lib->vba.ViewportHeight[k]; + } + + if (mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) { + MainPlaneDoesODMCombine = true; + } + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { + if (mode_lib->vba.BlendingAndTiming[k] == j + && mode_lib->vba.ODMCombineEnabled[k] == dm_odm_combine_mode_2to1) { + MainPlaneDoesODMCombine = true; + } + } + + if (MainPlaneDoesODMCombine == true) { + SwathWidth = dml_min( + SwathWidth, + mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k]); + } else { + SwathWidth = SwathWidth / mode_lib->vba.DPPPerPlane[k]; + } + + SwathWidthGranularityY = 256 / dml_ceil(BytePerPixDETY, 1) / MaximumSwathHeightY; + RoundedUpMaxSwathSizeBytesY = (dml_ceil( + (double) (SwathWidth - 1), + SwathWidthGranularityY) + SwathWidthGranularityY) * BytePerPixDETY + * MaximumSwathHeightY; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + RoundedUpMaxSwathSizeBytesY = dml_ceil(RoundedUpMaxSwathSizeBytesY, 256) + + 256; + } + if (MaximumSwathHeightC > 0) { + SwathWidthGranularityC = 256.0 / dml_ceil(BytePerPixDETC, 2) + / MaximumSwathHeightC; + RoundedUpMaxSwathSizeBytesC = (dml_ceil( + (double) (SwathWidth / 2.0 - 1), + SwathWidthGranularityC) + SwathWidthGranularityC) + * BytePerPixDETC * MaximumSwathHeightC; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + RoundedUpMaxSwathSizeBytesC = dml_ceil( + RoundedUpMaxSwathSizeBytesC, + 256) + 256; + } + } else + RoundedUpMaxSwathSizeBytesC = 0.0; + + if (RoundedUpMaxSwathSizeBytesY + RoundedUpMaxSwathSizeBytesC + <= mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0) { + mode_lib->vba.SwathHeightY[k] = MaximumSwathHeightY; + mode_lib->vba.SwathHeightC[k] = MaximumSwathHeightC; + } else { + mode_lib->vba.SwathHeightY[k] = MinimumSwathHeightY; + mode_lib->vba.SwathHeightC[k] = MinimumSwathHeightC; + } + + CalculateDETBufferSize( + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.SwathHeightY[k], + mode_lib->vba.SwathHeightC[k], + &mode_lib->vba.DETBufferSizeY[k], + &mode_lib->vba.DETBufferSizeC[k]); + } +} + +static double CalculateTWait( + unsigned int PrefetchMode, + double DRAMClockChangeLatency, + double UrgentLatency, + double SREnterPlusExitTime) +{ + if (PrefetchMode == 0) { + return dml_max( + DRAMClockChangeLatency + UrgentLatency, + dml_max(SREnterPlusExitTime, UrgentLatency)); + } else if (PrefetchMode == 1) { + return dml_max(SREnterPlusExitTime, UrgentLatency); + } else { + return UrgentLatency; + } +} + +static double CalculateRemoteSurfaceFlipDelay( + struct display_mode_lib *mode_lib, + double VRatio, + double SwathWidth, + double Bpp, + double LineTime, + double XFCTSlvVupdateOffset, + double XFCTSlvVupdateWidth, + double XFCTSlvVreadyOffset, + double XFCXBUFLatencyTolerance, + double XFCFillBWOverhead, + double XFCSlvChunkSize, + double XFCBusTransportTime, + double TCalc, + double TWait, + double *SrcActiveDrainRate, + double *TInitXFill, + double *TslvChk) +{ + double TSlvSetup, AvgfillRate, result; + + *SrcActiveDrainRate = VRatio * SwathWidth * Bpp / LineTime; + TSlvSetup = XFCTSlvVupdateOffset + XFCTSlvVupdateWidth + XFCTSlvVreadyOffset; + *TInitXFill = XFCXBUFLatencyTolerance / (1 + XFCFillBWOverhead / 100); + AvgfillRate = *SrcActiveDrainRate * (1 + XFCFillBWOverhead / 100); + *TslvChk = XFCSlvChunkSize / AvgfillRate; + dml_print( + "DML::CalculateRemoteSurfaceFlipDelay: SrcActiveDrainRate: %f\n", + *SrcActiveDrainRate); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TSlvSetup: %f\n", TSlvSetup); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TInitXFill: %f\n", *TInitXFill); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: AvgfillRate: %f\n", AvgfillRate); + dml_print("DML::CalculateRemoteSurfaceFlipDelay: TslvChk: %f\n", *TslvChk); + result = 2 * XFCBusTransportTime + TSlvSetup + TCalc + TWait + *TslvChk + *TInitXFill; // TODO: This doesn't seem to match programming guide + dml_print("DML::CalculateRemoteSurfaceFlipDelay: RemoteSurfaceFlipDelay: %f\n", result); + return result; +} + +static double CalculateWriteBackDelay( + enum source_format_class WritebackPixelFormat, + double WritebackHRatio, + double WritebackVRatio, + unsigned int WritebackLumaHTaps, + unsigned int WritebackLumaVTaps, + unsigned int WritebackChromaHTaps, + unsigned int WritebackChromaVTaps, + unsigned int WritebackDestinationWidth) +{ + double CalculateWriteBackDelay = + dml_max( + dml_ceil(WritebackLumaHTaps / 4.0, 1) / WritebackHRatio, + WritebackLumaVTaps * dml_ceil(1.0 / WritebackVRatio, 1) + * dml_ceil( + WritebackDestinationWidth + / 4.0, + 1) + + dml_ceil(1.0 / WritebackVRatio, 1) + * (dml_ceil( + WritebackLumaVTaps + / 4.0, + 1) + 4)); + + if (WritebackPixelFormat != dm_444_32) { + CalculateWriteBackDelay = + dml_max( + CalculateWriteBackDelay, + dml_max( + dml_ceil( + WritebackChromaHTaps + / 2.0, + 1) + / (2 + * WritebackHRatio), + WritebackChromaVTaps + * dml_ceil( + 1 + / (2 + * WritebackVRatio), + 1) + * dml_ceil( + WritebackDestinationWidth + / 2.0 + / 2.0, + 1) + + dml_ceil( + 1 + / (2 + * WritebackVRatio), + 1) + * (dml_ceil( + WritebackChromaVTaps + / 4.0, + 1) + + 4))); + } + return CalculateWriteBackDelay; +} + +static void CalculateActiveRowBandwidth( + bool GPUVMEnable, + enum source_format_class SourcePixelFormat, + double VRatio, + bool DCCEnable, + double LineTime, + unsigned int MetaRowByteLuma, + unsigned int MetaRowByteChroma, + unsigned int meta_row_height_luma, + unsigned int meta_row_height_chroma, + unsigned int PixelPTEBytesPerRowLuma, + unsigned int PixelPTEBytesPerRowChroma, + unsigned int dpte_row_height_luma, + unsigned int dpte_row_height_chroma, + double *meta_row_bw, + double *dpte_row_bw) +{ + if (DCCEnable != true) { + *meta_row_bw = 0; + } else if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *meta_row_bw = VRatio * MetaRowByteLuma / (meta_row_height_luma * LineTime) + + VRatio / 2 * MetaRowByteChroma + / (meta_row_height_chroma * LineTime); + } else { + *meta_row_bw = VRatio * MetaRowByteLuma / (meta_row_height_luma * LineTime); + } + + if (GPUVMEnable != true) { + *dpte_row_bw = 0; + } else if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + *dpte_row_bw = VRatio * PixelPTEBytesPerRowLuma / (dpte_row_height_luma * LineTime) + + VRatio / 2 * PixelPTEBytesPerRowChroma + / (dpte_row_height_chroma * LineTime); + } else { + *dpte_row_bw = VRatio * PixelPTEBytesPerRowLuma / (dpte_row_height_luma * LineTime); + } +} + +static void CalculateFlipSchedule( + struct display_mode_lib *mode_lib, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + double UrgentExtraLatency, + double UrgentLatency, + unsigned int GPUVMMaxPageTableLevels, + bool HostVMEnable, + unsigned int HostVMMaxPageTableLevels, + unsigned int HostVMCachedPageTableLevels, + bool GPUVMEnable, + double PDEAndMetaPTEBytesPerFrame, + double MetaRowBytes, + double DPTEBytesPerRow, + double BandwidthAvailableForImmediateFlip, + unsigned int TotImmediateFlipBytes, + enum source_format_class SourcePixelFormat, + double LineTime, + double VRatio, + double Tno_bw, + bool DCCEnable, + unsigned int dpte_row_height, + unsigned int meta_row_height, + unsigned int dpte_row_height_chroma, + unsigned int meta_row_height_chroma, + double *DestinationLinesToRequestVMInImmediateFlip, + double *DestinationLinesToRequestRowInImmediateFlip, + double *final_flip_bw, + bool *ImmediateFlipSupportedForPipe) +{ + double min_row_time = 0.0; + unsigned int HostVMDynamicLevels; + double TimeForFetchingMetaPTEImmediateFlip; + double TimeForFetchingRowInVBlankImmediateFlip; + double ImmediateFlipBW; + double HostVMInefficiencyFactor; + + if (GPUVMEnable == true && HostVMEnable == true) { + HostVMInefficiencyFactor = + PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData + / PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly; + HostVMDynamicLevels = HostVMMaxPageTableLevels - HostVMCachedPageTableLevels; + } else { + HostVMInefficiencyFactor = 1; + HostVMDynamicLevels = 0; + } + + ImmediateFlipBW = (PDEAndMetaPTEBytesPerFrame + MetaRowBytes + DPTEBytesPerRow) + * BandwidthAvailableForImmediateFlip / TotImmediateFlipBytes; + + if (GPUVMEnable == true) { + TimeForFetchingMetaPTEImmediateFlip = dml_max3( + Tno_bw + PDEAndMetaPTEBytesPerFrame * HostVMInefficiencyFactor / ImmediateFlipBW, + UrgentExtraLatency + UrgentLatency * (GPUVMMaxPageTableLevels * (HostVMDynamicLevels + 1) - 1), + LineTime / 4.0); + } else { + TimeForFetchingMetaPTEImmediateFlip = 0; + } + + *DestinationLinesToRequestVMInImmediateFlip = dml_ceil(4.0 * (TimeForFetchingMetaPTEImmediateFlip / LineTime), 1) / 4.0; + if ((GPUVMEnable == true || DCCEnable == true)) { + TimeForFetchingRowInVBlankImmediateFlip = dml_max3((MetaRowBytes + DPTEBytesPerRow) * HostVMInefficiencyFactor / ImmediateFlipBW, UrgentLatency * (HostVMDynamicLevels + 1), LineTime / 4); + } else { + TimeForFetchingRowInVBlankImmediateFlip = 0; + } + + *DestinationLinesToRequestRowInImmediateFlip = dml_ceil(4.0 * (TimeForFetchingRowInVBlankImmediateFlip / LineTime), 1) / 4.0; + *final_flip_bw = dml_max(PDEAndMetaPTEBytesPerFrame * HostVMInefficiencyFactor / (*DestinationLinesToRequestVMInImmediateFlip * LineTime), (MetaRowBytes + DPTEBytesPerRow) * HostVMInefficiencyFactor / (*DestinationLinesToRequestRowInImmediateFlip * LineTime)); + if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { + if (GPUVMEnable == true && DCCEnable != true) { + min_row_time = dml_min( + dpte_row_height * LineTime / VRatio, + dpte_row_height_chroma * LineTime / (VRatio / 2)); + } else if (GPUVMEnable != true && DCCEnable == true) { + min_row_time = dml_min( + meta_row_height * LineTime / VRatio, + meta_row_height_chroma * LineTime / (VRatio / 2)); + } else { + min_row_time = dml_min4( + dpte_row_height * LineTime / VRatio, + meta_row_height * LineTime / VRatio, + dpte_row_height_chroma * LineTime / (VRatio / 2), + meta_row_height_chroma * LineTime / (VRatio / 2)); + } + } else { + if (GPUVMEnable == true && DCCEnable != true) { + min_row_time = dpte_row_height * LineTime / VRatio; + } else if (GPUVMEnable != true && DCCEnable == true) { + min_row_time = meta_row_height * LineTime / VRatio; + } else { + min_row_time = dml_min( + dpte_row_height * LineTime / VRatio, + meta_row_height * LineTime / VRatio); + } + } + + if (*DestinationLinesToRequestVMInImmediateFlip >= 32 + || *DestinationLinesToRequestRowInImmediateFlip >= 16 + || TimeForFetchingMetaPTEImmediateFlip + 2 * TimeForFetchingRowInVBlankImmediateFlip > min_row_time) { + *ImmediateFlipSupportedForPipe = false; + } else { + *ImmediateFlipSupportedForPipe = true; + } +} + +static unsigned int TruncToValidBPP( + double DecimalBPP, + double DesiredBPP, + bool DSCEnabled, + enum output_encoder_class Output, + enum output_format_class Format, + unsigned int DSCInputBitPerComponent) +{ + if (Output == dm_hdmi) { + if (Format == dm_420) { + if (DecimalBPP >= 18 && (DesiredBPP == 0 || DesiredBPP == 18)) + return 18; + else if (DecimalBPP >= 15 && (DesiredBPP == 0 || DesiredBPP == 15)) + return 15; + else if (DecimalBPP >= 12 && (DesiredBPP == 0 || DesiredBPP == 12)) + return 12; + else + return BPP_INVALID; + } else if (Format == dm_444) { + if (DecimalBPP >= 36 && (DesiredBPP == 0 || DesiredBPP == 36)) + return 36; + else if (DecimalBPP >= 30 && (DesiredBPP == 0 || DesiredBPP == 30)) + return 30; + else if (DecimalBPP >= 24 && (DesiredBPP == 0 || DesiredBPP == 24)) + return 24; + else if (DecimalBPP >= 18 && (DesiredBPP == 0 || DesiredBPP == 18)) + return 18; + else + return BPP_INVALID; + } else { + if (DecimalBPP / 1.5 >= 24 && (DesiredBPP == 0 || DesiredBPP == 24)) + return 24; + else if (DecimalBPP / 1.5 >= 20 && (DesiredBPP == 0 || DesiredBPP == 20)) + return 20; + else if (DecimalBPP / 1.5 >= 16 && (DesiredBPP == 0 || DesiredBPP == 16)) + return 16; + else + return BPP_INVALID; + } + } else { + if (DSCEnabled) { + if (Format == dm_420) { + if (DesiredBPP == 0) { + if (DecimalBPP < 6) + return BPP_INVALID; + else if (DecimalBPP >= 1.5 * DSCInputBitPerComponent - 1.0 / 16.0) + return 1.5 * DSCInputBitPerComponent - 1.0 / 16.0; + else + return dml_floor(16 * DecimalBPP, 1) / 16.0; + } else { + if (DecimalBPP < 6 + || DesiredBPP < 6 + || DesiredBPP > 1.5 * DSCInputBitPerComponent - 1.0 / 16.0 + || DecimalBPP < DesiredBPP) { + return BPP_INVALID; + } else { + return DesiredBPP; + } + } + } else if (Format == dm_n422) { + if (DesiredBPP == 0) { + if (DecimalBPP < 7) + return BPP_INVALID; + else if (DecimalBPP >= 2 * DSCInputBitPerComponent - 1.0 / 16.0) + return 2 * DSCInputBitPerComponent - 1.0 / 16.0; + else + return dml_floor(16 * DecimalBPP, 1) / 16.0; + } else { + if (DecimalBPP < 7 + || DesiredBPP < 7 + || DesiredBPP > 2 * DSCInputBitPerComponent - 1.0 / 16.0 + || DecimalBPP < DesiredBPP) { + return BPP_INVALID; + } else { + return DesiredBPP; + } + } + } else { + if (DesiredBPP == 0) { + if (DecimalBPP < 8) + return BPP_INVALID; + else if (DecimalBPP >= 3 * DSCInputBitPerComponent - 1.0 / 16.0) + return 3 * DSCInputBitPerComponent - 1.0 / 16.0; + else + return dml_floor(16 * DecimalBPP, 1) / 16.0; + } else { + if (DecimalBPP < 8 + || DesiredBPP < 8 + || DesiredBPP > 3 * DSCInputBitPerComponent - 1.0 / 16.0 + || DecimalBPP < DesiredBPP) { + return BPP_INVALID; + } else { + return DesiredBPP; + } + } + } + } else if (Format == dm_420) { + if (DecimalBPP >= 18 && (DesiredBPP == 0 || DesiredBPP == 18)) + return 18; + else if (DecimalBPP >= 15 && (DesiredBPP == 0 || DesiredBPP == 15)) + return 15; + else if (DecimalBPP >= 12 && (DesiredBPP == 0 || DesiredBPP == 12)) + return 12; + else + return BPP_INVALID; + } else if (Format == dm_s422 || Format == dm_n422) { + if (DecimalBPP >= 24 && (DesiredBPP == 0 || DesiredBPP == 24)) + return 24; + else if (DecimalBPP >= 20 && (DesiredBPP == 0 || DesiredBPP == 20)) + return 20; + else if (DecimalBPP >= 16 && (DesiredBPP == 0 || DesiredBPP == 16)) + return 16; + else + return BPP_INVALID; + } else { + if (DecimalBPP >= 36 && (DesiredBPP == 0 || DesiredBPP == 36)) + return 36; + else if (DecimalBPP >= 30 && (DesiredBPP == 0 || DesiredBPP == 30)) + return 30; + else if (DecimalBPP >= 24 && (DesiredBPP == 0 || DesiredBPP == 24)) + return 24; + else if (DecimalBPP >= 18 && (DesiredBPP == 0 || DesiredBPP == 18)) + return 18; + else + return BPP_INVALID; + } + } +} + +void dml21_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) +{ + struct vba_vars_st *locals = &mode_lib->vba; + + int i; + unsigned int j, k, m; + + /*MODE SUPPORT, VOLTAGE STATE AND SOC CONFIGURATION*/ + + /*Scale Ratio, taps Support Check*/ + + mode_lib->vba.ScaleRatioAndTapsSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.ScalerEnabled[k] == false + && ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8) + || mode_lib->vba.HRatio[k] != 1.0 + || mode_lib->vba.htaps[k] != 1.0 + || mode_lib->vba.VRatio[k] != 1.0 + || mode_lib->vba.vtaps[k] != 1.0)) { + mode_lib->vba.ScaleRatioAndTapsSupport = false; + } else if (mode_lib->vba.vtaps[k] < 1.0 || mode_lib->vba.vtaps[k] > 8.0 + || mode_lib->vba.htaps[k] < 1.0 || mode_lib->vba.htaps[k] > 8.0 + || (mode_lib->vba.htaps[k] > 1.0 + && (mode_lib->vba.htaps[k] % 2) == 1) + || mode_lib->vba.HRatio[k] > mode_lib->vba.MaxHSCLRatio + || mode_lib->vba.VRatio[k] > mode_lib->vba.MaxVSCLRatio + || mode_lib->vba.HRatio[k] > mode_lib->vba.htaps[k] + || mode_lib->vba.VRatio[k] > mode_lib->vba.vtaps[k] + || (mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8 + && (mode_lib->vba.HRatio[k] / 2.0 + > mode_lib->vba.HTAPsChroma[k] + || mode_lib->vba.VRatio[k] / 2.0 + > mode_lib->vba.VTAPsChroma[k]))) { + mode_lib->vba.ScaleRatioAndTapsSupport = false; + } + } + /*Source Format, Pixel Format and Scan Support Check*/ + + mode_lib->vba.SourceFormatPixelAndScanSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if ((mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + && mode_lib->vba.SourceScan[k] != dm_horz) + || ((mode_lib->vba.SurfaceTiling[k] == dm_sw_4kb_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_4kb_d_x + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d_t + || mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_d_x + || mode_lib->vba.SurfaceTiling[k] == dm_sw_var_d + || mode_lib->vba.SurfaceTiling[k] == dm_sw_var_d_x) + && mode_lib->vba.SourcePixelFormat[k] != dm_444_64) + || (mode_lib->vba.SurfaceTiling[k] == dm_sw_64kb_r_x + && (mode_lib->vba.SourcePixelFormat[k] == dm_mono_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_10)) + || (((mode_lib->vba.SurfaceTiling[k] == dm_sw_gfx7_2d_thin_gl + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_gfx7_2d_thin_l_vp) + && !((mode_lib->vba.SourcePixelFormat[k] + == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] + == dm_444_32) + && mode_lib->vba.SourceScan[k] + == dm_horz + && mode_lib->vba.SupportGFX7CompatibleTilingIn32bppAnd64bpp + == true + && mode_lib->vba.DCCEnable[k] + == false)) + || (mode_lib->vba.DCCEnable[k] == true + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_linear + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_8 + || mode_lib->vba.SourcePixelFormat[k] + == dm_420_10)))) { + mode_lib->vba.SourceFormatPixelAndScanSupport = false; + } + } + /*Bandwidth Support Check*/ + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.SourcePixelFormat[k] == dm_444_64) { + locals->BytePerPixelInDETY[k] = 8.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_32) { + locals->BytePerPixelInDETY[k] = 4.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16) { + locals->BytePerPixelInDETY[k] = 2.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_mono_8) { + locals->BytePerPixelInDETY[k] = 1.0; + locals->BytePerPixelInDETC[k] = 0.0; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8) { + locals->BytePerPixelInDETY[k] = 1.0; + locals->BytePerPixelInDETC[k] = 2.0; + } else { + locals->BytePerPixelInDETY[k] = 4.0 / 3; + locals->BytePerPixelInDETC[k] = 8.0 / 3; + } + if (mode_lib->vba.SourceScan[k] == dm_horz) { + locals->SwathWidthYSingleDPP[k] = mode_lib->vba.ViewportWidth[k]; + } else { + locals->SwathWidthYSingleDPP[k] = mode_lib->vba.ViewportHeight[k]; + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->ReadBandwidthLuma[k] = locals->SwathWidthYSingleDPP[k] * dml_ceil(locals->BytePerPixelInDETY[k], 1.0) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k]; + locals->ReadBandwidthChroma[k] = locals->SwathWidthYSingleDPP[k] / 2 * dml_ceil(locals->BytePerPixelInDETC[k], 2.0) + / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k] / 2.0; + locals->ReadBandwidth[k] = locals->ReadBandwidthLuma[k] + locals->ReadBandwidthChroma[k]; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 4.0; + } else if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 3.0; + } else if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WriteBandwidth[k] = mode_lib->vba.WritebackDestinationWidth[k] + * mode_lib->vba.WritebackDestinationHeight[k] + / (mode_lib->vba.WritebackSourceHeight[k] + * mode_lib->vba.HTotal[k] + / mode_lib->vba.PixelClock[k]) * 1.5; + } else { + locals->WriteBandwidth[k] = 0.0; + } + } + mode_lib->vba.DCCEnabledInAnyPlane = false; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.DCCEnable[k] == true) { + mode_lib->vba.DCCEnabledInAnyPlane = true; + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->IdealSDPPortBandwidthPerState[i][0] = dml_min3( + mode_lib->vba.ReturnBusWidth * mode_lib->vba.DCFCLKPerState[i], + mode_lib->vba.DRAMSpeedPerState[i] * mode_lib->vba.NumberOfChannels + * mode_lib->vba.DRAMChannelWidth, + mode_lib->vba.FabricClockPerState[i] + * mode_lib->vba.FabricDatapathToDCNDataReturn); + if (mode_lib->vba.HostVMEnable == false) { + locals->ReturnBWPerState[i][0] = locals->IdealSDPPortBandwidthPerState[i][0] + * mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelDataOnly / 100.0; + } else { + locals->ReturnBWPerState[i][0] = locals->IdealSDPPortBandwidthPerState[i][0] + * mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData / 100.0; + } + } + /*Writeback Latency support check*/ + + mode_lib->vba.WritebackLatencySupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.WritebackPixelFormat[k] == dm_444_32) { + if (locals->WriteBandwidth[k] + > (mode_lib->vba.WritebackInterfaceLumaBufferSize + + mode_lib->vba.WritebackInterfaceChromaBufferSize) + / mode_lib->vba.WritebackLatency) { + mode_lib->vba.WritebackLatencySupport = false; + } + } else { + if (locals->WriteBandwidth[k] + > 1.5 + * dml_min( + mode_lib->vba.WritebackInterfaceLumaBufferSize, + 2.0 + * mode_lib->vba.WritebackInterfaceChromaBufferSize) + / mode_lib->vba.WritebackLatency) { + mode_lib->vba.WritebackLatencySupport = false; + } + } + } + } + /*Re-ordering Buffer Support Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i] = + (mode_lib->vba.RoundTripPingLatencyCycles + 32.0) / mode_lib->vba.DCFCLKPerState[i] + + dml_max3(mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelDataOnly, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelMixedWithVMData, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelVMDataOnly) + * mode_lib->vba.NumberOfChannels / locals->ReturnBWPerState[i][0]; + if ((mode_lib->vba.ROBBufferSizeInKByte - mode_lib->vba.PixelChunkSizeInKByte) * 1024.0 / locals->ReturnBWPerState[i][0] + > locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i]) { + locals->ROBSupport[i][0] = true; + } else { + locals->ROBSupport[i][0] = false; + } + } + /*Writeback Mode Support Check*/ + + mode_lib->vba.TotalNumberOfActiveWriteback = 0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.ActiveWritebacksPerPlane[k] == 0) + mode_lib->vba.ActiveWritebacksPerPlane[k] = 1; + mode_lib->vba.TotalNumberOfActiveWriteback = + mode_lib->vba.TotalNumberOfActiveWriteback + + mode_lib->vba.ActiveWritebacksPerPlane[k]; + } + } + mode_lib->vba.WritebackModeSupport = true; + if (mode_lib->vba.TotalNumberOfActiveWriteback > mode_lib->vba.MaxNumWriteback) { + mode_lib->vba.WritebackModeSupport = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true + && mode_lib->vba.Writeback10bpc420Supported != true + && mode_lib->vba.WritebackPixelFormat[k] == dm_420_10) { + mode_lib->vba.WritebackModeSupport = false; + } + } + /*Writeback Scale Ratio and Taps Support Check*/ + + mode_lib->vba.WritebackScaleRatioAndTapsSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + if (mode_lib->vba.WritebackLumaAndChromaScalingSupported == false + && (mode_lib->vba.WritebackHRatio[k] != 1.0 + || mode_lib->vba.WritebackVRatio[k] != 1.0)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (mode_lib->vba.WritebackHRatio[k] > mode_lib->vba.WritebackMaxHSCLRatio + || mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackMaxVSCLRatio + || mode_lib->vba.WritebackHRatio[k] + < mode_lib->vba.WritebackMinHSCLRatio + || mode_lib->vba.WritebackVRatio[k] + < mode_lib->vba.WritebackMinVSCLRatio + || mode_lib->vba.WritebackLumaHTaps[k] + > mode_lib->vba.WritebackMaxHSCLTaps + || mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackMaxVSCLTaps + || mode_lib->vba.WritebackHRatio[k] + > mode_lib->vba.WritebackLumaHTaps[k] + || mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackLumaVTaps[k] + || (mode_lib->vba.WritebackLumaHTaps[k] > 2.0 + && ((mode_lib->vba.WritebackLumaHTaps[k] % 2) + == 1)) + || (mode_lib->vba.WritebackPixelFormat[k] != dm_444_32 + && (mode_lib->vba.WritebackChromaHTaps[k] + > mode_lib->vba.WritebackMaxHSCLTaps + || mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackMaxVSCLTaps + || 2.0 + * mode_lib->vba.WritebackHRatio[k] + > mode_lib->vba.WritebackChromaHTaps[k] + || 2.0 + * mode_lib->vba.WritebackVRatio[k] + > mode_lib->vba.WritebackChromaVTaps[k] + || (mode_lib->vba.WritebackChromaHTaps[k] > 2.0 + && ((mode_lib->vba.WritebackChromaHTaps[k] % 2) == 1))))) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (mode_lib->vba.WritebackVRatio[k] < 1.0) { + mode_lib->vba.WritebackLumaVExtra = + dml_max(1.0 - 2.0 / dml_ceil(1.0 / mode_lib->vba.WritebackVRatio[k], 1.0), 0.0); + } else { + mode_lib->vba.WritebackLumaVExtra = -1; + } + if ((mode_lib->vba.WritebackPixelFormat[k] == dm_444_32 + && mode_lib->vba.WritebackLumaVTaps[k] + > (mode_lib->vba.WritebackLineBufferLumaBufferSize + + mode_lib->vba.WritebackLineBufferChromaBufferSize) + / 3.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_8 + && mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackLineBufferLumaBufferSize + * 8.0 / 10.0 / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10 + && mode_lib->vba.WritebackLumaVTaps[k] + > mode_lib->vba.WritebackLineBufferLumaBufferSize + * 8.0 / 10.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackLumaVExtra)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + if (2.0 * mode_lib->vba.WritebackVRatio[k] < 1) { + mode_lib->vba.WritebackChromaVExtra = 0.0; + } else { + mode_lib->vba.WritebackChromaVExtra = -1; + } + if ((mode_lib->vba.WritebackPixelFormat[k] == dm_420_8 + && mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackLineBufferChromaBufferSize + * 8.0 / 10.0 / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackChromaVExtra) + || (mode_lib->vba.WritebackPixelFormat[k] == dm_420_10 + && mode_lib->vba.WritebackChromaVTaps[k] + > mode_lib->vba.WritebackLineBufferChromaBufferSize + * 8.0 / 10.0 + / mode_lib->vba.WritebackDestinationWidth[k] + - mode_lib->vba.WritebackChromaVExtra)) { + mode_lib->vba.WritebackScaleRatioAndTapsSupport = false; + } + } + } + /*Maximum DISPCLK/DPPCLK Support check*/ + + mode_lib->vba.WritebackRequiredDISPCLK = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.WritebackEnable[k] == true) { + mode_lib->vba.WritebackRequiredDISPCLK = + dml_max( + mode_lib->vba.WritebackRequiredDISPCLK, + CalculateWriteBackDISPCLK( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.PixelClock[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k], + mode_lib->vba.HTotal[k], + mode_lib->vba.WritebackChromaLineBufferWidth)); + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.HRatio[k] > 1.0) { + locals->PSCL_FACTOR[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / dml_ceil( + mode_lib->vba.htaps[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_FACTOR[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + if (locals->BytePerPixelInDETC[k] == 0.0) { + locals->PSCL_FACTOR_CHROMA[k] = 0.0; + locals->MinDPPCLKUsingSingleDPP[k] = + mode_lib->vba.PixelClock[k] + * dml_max3( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_FACTOR[k], + 1.0); + if ((mode_lib->vba.htaps[k] > 6.0 || mode_lib->vba.vtaps[k] > 6.0) + && locals->MinDPPCLKUsingSingleDPP[k] + < 2.0 * mode_lib->vba.PixelClock[k]) { + locals->MinDPPCLKUsingSingleDPP[k] = 2.0 + * mode_lib->vba.PixelClock[k]; + } + } else { + if (mode_lib->vba.HRatio[k] / 2.0 > 1.0) { + locals->PSCL_FACTOR_CHROMA[k] = + dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput + * mode_lib->vba.HRatio[k] + / 2.0 + / dml_ceil( + mode_lib->vba.HTAPsChroma[k] + / 6.0, + 1.0)); + } else { + locals->PSCL_FACTOR_CHROMA[k] = dml_min( + mode_lib->vba.MaxDCHUBToPSCLThroughput, + mode_lib->vba.MaxPSCLToLBThroughput); + } + locals->MinDPPCLKUsingSingleDPP[k] = + mode_lib->vba.PixelClock[k] + * dml_max5( + mode_lib->vba.vtaps[k] / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k]), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / locals->PSCL_FACTOR[k], + mode_lib->vba.VTAPsChroma[k] + / 6.0 + * dml_min( + 1.0, + mode_lib->vba.HRatio[k] + / 2.0), + mode_lib->vba.HRatio[k] + * mode_lib->vba.VRatio[k] + / 4.0 + / locals->PSCL_FACTOR_CHROMA[k], + 1.0); + if ((mode_lib->vba.htaps[k] > 6.0 || mode_lib->vba.vtaps[k] > 6.0 + || mode_lib->vba.HTAPsChroma[k] > 6.0 + || mode_lib->vba.VTAPsChroma[k] > 6.0) + && locals->MinDPPCLKUsingSingleDPP[k] + < 2.0 * mode_lib->vba.PixelClock[k]) { + locals->MinDPPCLKUsingSingleDPP[k] = 2.0 + * mode_lib->vba.PixelClock[k]; + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + Calculate256BBlockSizes( + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + dml_ceil(locals->BytePerPixelInDETC[k], 2.0), + &locals->Read256BlockHeightY[k], + &locals->Read256BlockHeightC[k], + &locals->Read256BlockWidthY[k], + &locals->Read256BlockWidthC[k]); + if (mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MaxSwathHeightY[k] = locals->Read256BlockHeightY[k]; + locals->MaxSwathHeightC[k] = locals->Read256BlockHeightC[k]; + } else { + locals->MaxSwathHeightY[k] = locals->Read256BlockWidthY[k]; + locals->MaxSwathHeightC[k] = locals->Read256BlockWidthC[k]; + } + if ((mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_32 + || mode_lib->vba.SourcePixelFormat[k] == dm_444_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_16 + || mode_lib->vba.SourcePixelFormat[k] == dm_mono_8)) { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear + || (mode_lib->vba.SourcePixelFormat[k] == dm_444_64 + && (mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_4kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_t + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_64kb_s_x + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s + || mode_lib->vba.SurfaceTiling[k] + == dm_sw_var_s_x) + && mode_lib->vba.SourceScan[k] == dm_horz)) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + } else { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k] + / 2.0; + } + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else { + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_8 + && mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k] + / 2.0; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } else if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10 + && mode_lib->vba.SourceScan[k] == dm_horz) { + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k] + / 2.0; + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + } else { + locals->MinSwathHeightY[k] = locals->MaxSwathHeightY[k]; + locals->MinSwathHeightC[k] = locals->MaxSwathHeightC[k]; + } + } + if (mode_lib->vba.SurfaceTiling[k] == dm_sw_linear) { + mode_lib->vba.MaximumSwathWidthSupport = 8192.0; + } else { + mode_lib->vba.MaximumSwathWidthSupport = 5120.0; + } + mode_lib->vba.MaximumSwathWidthInDETBuffer = + dml_min( + mode_lib->vba.MaximumSwathWidthSupport, + mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0 + / (locals->BytePerPixelInDETY[k] + * locals->MinSwathHeightY[k] + + locals->BytePerPixelInDETC[k] + / 2.0 + * locals->MinSwathHeightC[k])); + if (locals->BytePerPixelInDETC[k] == 0.0) { + mode_lib->vba.MaximumSwathWidthInLineBuffer = + mode_lib->vba.LineBufferSize + * dml_max(mode_lib->vba.HRatio[k], 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.vtaps[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k], + 1.0) + - 2, + 0.0)); + } else { + mode_lib->vba.MaximumSwathWidthInLineBuffer = + dml_min( + mode_lib->vba.LineBufferSize + * dml_max( + mode_lib->vba.HRatio[k], + 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.vtaps[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k], + 1.0) + - 2, + 0.0)), + 2.0 * mode_lib->vba.LineBufferSize + * dml_max( + mode_lib->vba.HRatio[k] + / 2.0, + 1.0) + / mode_lib->vba.LBBitPerPixel[k] + / (mode_lib->vba.VTAPsChroma[k] + + dml_max( + dml_ceil( + mode_lib->vba.VRatio[k] + / 2.0, + 1.0) + - 2, + 0.0))); + } + locals->MaximumSwathWidth[k] = dml_min( + mode_lib->vba.MaximumSwathWidthInDETBuffer, + mode_lib->vba.MaximumSwathWidthInLineBuffer); + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + double MaxMaxDispclkRoundedDown = RoundToDFSGranularityDown( + mode_lib->vba.MaxDispclk[mode_lib->vba.soc.num_states], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + + for (j = 0; j < 2; j++) { + mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.MaxDispclk[i], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity = RoundToDFSGranularityDown( + mode_lib->vba.MaxDppclk[i], + mode_lib->vba.DISPCLKDPPCLKVCOSpeed); + locals->RequiredDISPCLK[i][j] = 0.0; + locals->DISPCLK_DPPCLK_Support[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine = + mode_lib->vba.PixelClock[k] + * (1.0 + + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading + / 100.0) + * (1.0 + + mode_lib->vba.DISPCLKRampingMargin + / 100.0); + if (mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine >= mode_lib->vba.MaxDispclk[i] + && i == mode_lib->vba.soc.num_states) + mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine = mode_lib->vba.PixelClock[k] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + + mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine = mode_lib->vba.PixelClock[k] / 2 + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * (1 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + if (mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine >= mode_lib->vba.MaxDispclk[i] + && i == mode_lib->vba.soc.num_states) + mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine = mode_lib->vba.PixelClock[k] / 2 + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + + locals->ODMCombineEnablePerState[i][k] = false; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithoutODMCombine; + if (mode_lib->vba.ODMCapability) { + if (locals->PlaneRequiredDISPCLKWithoutODMCombine > MaxMaxDispclkRoundedDown) { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } else if (locals->DSCEnabled[k] && (locals->HActive[k] > DCN21_MAX_DSC_IMAGE_WIDTH)) { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } else if (locals->HActive[k] > DCN21_MAX_420_IMAGE_WIDTH && locals->OutputFormat[k] == dm_420) { + locals->ODMCombineEnablePerState[i][k] = true; + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PlaneRequiredDISPCLKWithODMCombine; + } + } + + if (locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) <= mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity + && locals->SwathWidthYSingleDPP[k] <= locals->MaximumSwathWidth[k] + && locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_disabled) { + locals->NoOfDPP[i][j][k] = 1; + locals->RequiredDPPCLK[i][j][k] = + locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } else { + locals->NoOfDPP[i][j][k] = 2; + locals->RequiredDPPCLK[i][j][k] = + locals->MinDPPCLKUsingSingleDPP[k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) / 2.0; + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.PlaneRequiredDISPCLK); + if ((locals->MinDPPCLKUsingSingleDPP[k] / locals->NoOfDPP[i][j][k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + > mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity) + || (mode_lib->vba.PlaneRequiredDISPCLK > mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity)) { + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + } + locals->TotalNumberOfActiveDPP[i][j] = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + if (j == 1) { + while (locals->TotalNumberOfActiveDPP[i][j] < mode_lib->vba.MaxNumDPP + && locals->TotalNumberOfActiveDPP[i][j] < 2 * mode_lib->vba.NumberOfActivePlanes) { + double BWOfNonSplitPlaneOfMaximumBandwidth; + unsigned int NumberOfNonSplitPlaneOfMaximumBandwidth; + + BWOfNonSplitPlaneOfMaximumBandwidth = 0; + NumberOfNonSplitPlaneOfMaximumBandwidth = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->ReadBandwidth[k] > BWOfNonSplitPlaneOfMaximumBandwidth && locals->NoOfDPP[i][j][k] == 1) { + BWOfNonSplitPlaneOfMaximumBandwidth = locals->ReadBandwidth[k]; + NumberOfNonSplitPlaneOfMaximumBandwidth = k; + } + } + locals->NoOfDPP[i][j][NumberOfNonSplitPlaneOfMaximumBandwidth] = 2; + locals->RequiredDPPCLK[i][j][NumberOfNonSplitPlaneOfMaximumBandwidth] = + locals->MinDPPCLKUsingSingleDPP[NumberOfNonSplitPlaneOfMaximumBandwidth] + * (1 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100) / 2; + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + 1; + } + } + if (locals->TotalNumberOfActiveDPP[i][j] > mode_lib->vba.MaxNumDPP) { + locals->RequiredDISPCLK[i][j] = 0.0; + locals->DISPCLK_DPPCLK_Support[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->ODMCombineEnablePerState[i][k] = false; + if (locals->SwathWidthYSingleDPP[k] <= locals->MaximumSwathWidth[k]) { + locals->NoOfDPP[i][j][k] = 1; + locals->RequiredDPPCLK[i][j][k] = locals->MinDPPCLKUsingSingleDPP[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } else { + locals->NoOfDPP[i][j][k] = 2; + locals->RequiredDPPCLK[i][j][k] = locals->MinDPPCLKUsingSingleDPP[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) / 2.0; + } + if (i != mode_lib->vba.soc.num_states) { + mode_lib->vba.PlaneRequiredDISPCLK = + mode_lib->vba.PixelClock[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + * (1.0 + mode_lib->vba.DISPCLKRampingMargin / 100.0); + } else { + mode_lib->vba.PlaneRequiredDISPCLK = mode_lib->vba.PixelClock[k] + * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0); + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.PlaneRequiredDISPCLK); + if (locals->MinDPPCLKUsingSingleDPP[k] / locals->NoOfDPP[i][j][k] * (1.0 + mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) + > mode_lib->vba.MaxDppclkRoundedDownToDFSGranularity + || mode_lib->vba.PlaneRequiredDISPCLK > mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity) + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + locals->TotalNumberOfActiveDPP[i][j] = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) + locals->TotalNumberOfActiveDPP[i][j] = locals->TotalNumberOfActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + } + locals->RequiredDISPCLK[i][j] = dml_max( + locals->RequiredDISPCLK[i][j], + mode_lib->vba.WritebackRequiredDISPCLK); + if (mode_lib->vba.MaxDispclkRoundedDownToDFSGranularity + < mode_lib->vba.WritebackRequiredDISPCLK) { + locals->DISPCLK_DPPCLK_Support[i][j] = false; + } + } + } + /*Viewport Size Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->ViewportSizeSupport[i][0] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { + if (dml_min(locals->SwathWidthYSingleDPP[k], dml_round(mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k])) + > locals->MaximumSwathWidth[k]) { + locals->ViewportSizeSupport[i][0] = false; + } + } else { + if (locals->SwathWidthYSingleDPP[k] / 2.0 > locals->MaximumSwathWidth[k]) { + locals->ViewportSizeSupport[i][0] = false; + } + } + } + } + /*Total Available Pipes Support Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + if (locals->TotalNumberOfActiveDPP[i][j] <= mode_lib->vba.MaxNumDPP) + locals->TotalAvailablePipesSupport[i][j] = true; + else + locals->TotalAvailablePipesSupport[i][j] = false; + } + } + /*Total Available OTG Support Check*/ + + mode_lib->vba.TotalNumberOfActiveOTG = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.TotalNumberOfActiveOTG = mode_lib->vba.TotalNumberOfActiveOTG + + 1.0; + } + } + if (mode_lib->vba.TotalNumberOfActiveOTG <= mode_lib->vba.MaxNumOTG) { + mode_lib->vba.NumberOfOTGSupport = true; + } else { + mode_lib->vba.NumberOfOTGSupport = false; + } + /*Display IO and DSC Support Check*/ + + mode_lib->vba.NonsupportedDSCInputBPC = false; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (!(mode_lib->vba.DSCInputBitPerComponent[k] == 12.0 + || mode_lib->vba.DSCInputBitPerComponent[k] == 10.0 + || mode_lib->vba.DSCInputBitPerComponent[k] == 8.0)) { + mode_lib->vba.NonsupportedDSCInputBPC = true; + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = 0; + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.Output[k] == dm_hdmi) { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = 0; + locals->OutputBppPerState[i][k] = TruncToValidBPP( + dml_min(600.0, mode_lib->vba.PHYCLKPerState[i]) / mode_lib->vba.PixelClockBackEnd[k] * 24, + mode_lib->vba.ForcedOutputLinkBPP[k], + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + } else if (mode_lib->vba.Output[k] == dm_dp + || mode_lib->vba.Output[k] == dm_edp) { + if (mode_lib->vba.Output[k] == dm_edp) { + mode_lib->vba.EffectiveFECOverhead = 0.0; + } else { + mode_lib->vba.EffectiveFECOverhead = + mode_lib->vba.FECOverhead; + } + if (mode_lib->vba.PHYCLKPerState[i] >= 270.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 270.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 270.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = mode_lib->vba.Outbpp; + } + if (mode_lib->vba.Outbpp == BPP_INVALID && mode_lib->vba.PHYCLKPerState[i] >= 540.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 540.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 540.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = mode_lib->vba.Outbpp; + } + if (mode_lib->vba.Outbpp == BPP_INVALID + && mode_lib->vba.PHYCLKPerState[i] + >= 810.0) { + mode_lib->vba.Outbpp = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * 810.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + false, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + mode_lib->vba.OutbppDSC = TruncToValidBPP( + (1.0 - mode_lib->vba.Downspreading / 100.0) * (1.0 - mode_lib->vba.EffectiveFECOverhead / 100.0) * 810.0 + * mode_lib->vba.OutputLinkDPLanes[k] / mode_lib->vba.PixelClockBackEnd[k] * 8.0, + mode_lib->vba.ForcedOutputLinkBPP[k], + true, + mode_lib->vba.Output[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.DSCInputBitPerComponent[k]); + if (mode_lib->vba.DSCEnabled[k] == true || mode_lib->vba.Outbpp == BPP_INVALID) { + locals->RequiresDSC[i][k] = true; + if (mode_lib->vba.Output[k] == dm_dp) { + locals->RequiresFEC[i][k] = true; + } else { + locals->RequiresFEC[i][k] = false; + } + mode_lib->vba.Outbpp = mode_lib->vba.OutbppDSC; + } else { + locals->RequiresDSC[i][k] = false; + locals->RequiresFEC[i][k] = false; + } + locals->OutputBppPerState[i][k] = + mode_lib->vba.Outbpp; + } + } + } else { + locals->OutputBppPerState[i][k] = BPP_BLENDED_PIPE; + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->DIOSupport[i] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->OutputBppPerState[i][k] == BPP_INVALID + || (mode_lib->vba.OutputFormat[k] == dm_420 + && mode_lib->vba.Interlace[k] == true + && mode_lib->vba.ProgressiveToInterlaceUnitInOPP == true)) { + locals->DIOSupport[i] = false; + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->DSCCLKRequiredMoreThanSupported[i] = false; + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if ((mode_lib->vba.Output[k] == dm_dp + || mode_lib->vba.Output[k] == dm_edp)) { + if (mode_lib->vba.OutputFormat[k] == dm_420 + || mode_lib->vba.OutputFormat[k] + == dm_n422) { + mode_lib->vba.DSCFormatFactor = 2; + } else { + mode_lib->vba.DSCFormatFactor = 1; + } + if (locals->RequiresDSC[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { + if (mode_lib->vba.PixelClockBackEnd[k] / 6.0 / mode_lib->vba.DSCFormatFactor + > (1.0 - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * mode_lib->vba.MaxDSCCLK[i]) { + locals->DSCCLKRequiredMoreThanSupported[i] = + true; + } + } else { + if (mode_lib->vba.PixelClockBackEnd[k] / 3.0 / mode_lib->vba.DSCFormatFactor + > (1.0 - mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading / 100.0) * mode_lib->vba.MaxDSCCLK[i]) { + locals->DSCCLKRequiredMoreThanSupported[i] = + true; + } + } + } + } + } + } + } + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + locals->NotEnoughDSCUnits[i] = false; + mode_lib->vba.TotalDSCUnitsRequired = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->RequiresDSC[i][k] == true) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { + mode_lib->vba.TotalDSCUnitsRequired = + mode_lib->vba.TotalDSCUnitsRequired + 2.0; + } else { + mode_lib->vba.TotalDSCUnitsRequired = + mode_lib->vba.TotalDSCUnitsRequired + 1.0; + } + } + } + if (mode_lib->vba.TotalDSCUnitsRequired > mode_lib->vba.NumberOfDSC) { + locals->NotEnoughDSCUnits[i] = true; + } + } + /*DSC Delay per state*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] != k) { + mode_lib->vba.slices = 0; + } else if (locals->RequiresDSC[i][k] == 0 + || locals->RequiresDSC[i][k] == false) { + mode_lib->vba.slices = 0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 3200.0) { + mode_lib->vba.slices = dml_ceil( + mode_lib->vba.PixelClockBackEnd[k] / 400.0, + 4.0); + } else if (mode_lib->vba.PixelClockBackEnd[k] > 1360.0) { + mode_lib->vba.slices = 8.0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 680.0) { + mode_lib->vba.slices = 4.0; + } else if (mode_lib->vba.PixelClockBackEnd[k] > 340.0) { + mode_lib->vba.slices = 2.0; + } else { + mode_lib->vba.slices = 1.0; + } + if (locals->OutputBppPerState[i][k] == BPP_BLENDED_PIPE + || locals->OutputBppPerState[i][k] == BPP_INVALID) { + mode_lib->vba.bpp = 0.0; + } else { + mode_lib->vba.bpp = locals->OutputBppPerState[i][k]; + } + if (locals->RequiresDSC[i][k] == true && mode_lib->vba.bpp != 0.0) { + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_disabled) { + locals->DSCDelayPerState[i][k] = + dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + mode_lib->vba.bpp, + dml_ceil( + mode_lib->vba.HActive[k] + / mode_lib->vba.slices, + 1.0), + mode_lib->vba.slices, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay( + mode_lib->vba.OutputFormat[k]); + } else { + locals->DSCDelayPerState[i][k] = + 2.0 * (dscceComputeDelay( + mode_lib->vba.DSCInputBitPerComponent[k], + mode_lib->vba.bpp, + dml_ceil(mode_lib->vba.HActive[k] / mode_lib->vba.slices, 1.0), + mode_lib->vba.slices / 2, + mode_lib->vba.OutputFormat[k]) + + dscComputeDelay(mode_lib->vba.OutputFormat[k])); + } + locals->DSCDelayPerState[i][k] = + locals->DSCDelayPerState[i][k] * mode_lib->vba.PixelClock[k] / mode_lib->vba.PixelClockBackEnd[k]; + } else { + locals->DSCDelayPerState[i][k] = 0.0; + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + for (j = 0; j <= mode_lib->vba.NumberOfActivePlanes - 1; j++) { + if (mode_lib->vba.BlendingAndTiming[k] == m && locals->RequiresDSC[i][m] == true) + locals->DSCDelayPerState[i][k] = locals->DSCDelayPerState[i][m]; + } + } + } + } + + //Prefetch Check + for (i = 0; i <= mode_lib->vba.soc.num_states; ++i) { + for (j = 0; j <= 1; ++j) { + locals->TotalNumberOfDCCActiveDPP[i][j] = 0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (mode_lib->vba.DCCEnable[k] == true) + locals->TotalNumberOfDCCActiveDPP[i][j] = locals->TotalNumberOfDCCActiveDPP[i][j] + locals->NoOfDPP[i][j][k]; + } + } + } + + mode_lib->vba.UrgentLatency = dml_max3( + mode_lib->vba.UrgentLatencyPixelDataOnly, + mode_lib->vba.UrgentLatencyPixelMixedWithVMData, + mode_lib->vba.UrgentLatencyVMDataOnly); + mode_lib->vba.PrefetchERROR = CalculateMinAndMaxPrefetchMode( + mode_lib->vba.AllowDRAMSelfRefreshOrDRAMClockChangeInVblank, + &mode_lib->vba.MinPrefetchMode, + &mode_lib->vba.MaxPrefetchMode); + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->RequiredDPPCLKThisState[k] = locals->RequiredDPPCLK[i][j][k]; + locals->NoOfDPPThisState[k] = locals->NoOfDPP[i][j][k]; + if (locals->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1) { + locals->SwathWidthYThisState[k] = + dml_min(locals->SwathWidthYSingleDPP[k], dml_round(mode_lib->vba.HActive[k] / 2.0 * mode_lib->vba.HRatio[k])); + } else { + locals->SwathWidthYThisState[k] = locals->SwathWidthYSingleDPP[k] / locals->NoOfDPP[i][j][k]; + } + mode_lib->vba.SwathWidthGranularityY = 256.0 + / dml_ceil(locals->BytePerPixelInDETY[k], 1.0) + / locals->MaxSwathHeightY[k]; + mode_lib->vba.RoundedUpMaxSwathSizeBytesY = + (dml_ceil(locals->SwathWidthYThisState[k] - 1.0, mode_lib->vba.SwathWidthGranularityY) + + mode_lib->vba.SwathWidthGranularityY) * locals->BytePerPixelInDETY[k] * locals->MaxSwathHeightY[k]; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + mode_lib->vba.RoundedUpMaxSwathSizeBytesY = dml_ceil( + mode_lib->vba.RoundedUpMaxSwathSizeBytesY, + 256.0) + 256; + } + if (locals->MaxSwathHeightC[k] > 0.0) { + mode_lib->vba.SwathWidthGranularityC = 256.0 / dml_ceil(locals->BytePerPixelInDETC[k], 2.0) / locals->MaxSwathHeightC[k]; + mode_lib->vba.RoundedUpMaxSwathSizeBytesC = (dml_ceil(locals->SwathWidthYThisState[k] / 2.0 - 1.0, mode_lib->vba.SwathWidthGranularityC) + + mode_lib->vba.SwathWidthGranularityC) * locals->BytePerPixelInDETC[k] * locals->MaxSwathHeightC[k]; + if (mode_lib->vba.SourcePixelFormat[k] == dm_420_10) { + mode_lib->vba.RoundedUpMaxSwathSizeBytesC = dml_ceil(mode_lib->vba.RoundedUpMaxSwathSizeBytesC, 256.0) + 256; + } + } else { + mode_lib->vba.RoundedUpMaxSwathSizeBytesC = 0.0; + } + if (mode_lib->vba.RoundedUpMaxSwathSizeBytesY + mode_lib->vba.RoundedUpMaxSwathSizeBytesC + <= mode_lib->vba.DETBufferSizeInKByte * 1024.0 / 2.0) { + locals->SwathHeightYThisState[k] = locals->MaxSwathHeightY[k]; + locals->SwathHeightCThisState[k] = locals->MaxSwathHeightC[k]; + } else { + locals->SwathHeightYThisState[k] = + locals->MinSwathHeightY[k]; + locals->SwathHeightCThisState[k] = + locals->MinSwathHeightC[k]; + } + } + + CalculateDCFCLKDeepSleep( + mode_lib, + mode_lib->vba.NumberOfActivePlanes, + locals->BytePerPixelInDETY, + locals->BytePerPixelInDETC, + mode_lib->vba.VRatio, + locals->SwathWidthYThisState, + locals->NoOfDPPThisState, + mode_lib->vba.HRatio, + mode_lib->vba.PixelClock, + locals->PSCL_FACTOR, + locals->PSCL_FACTOR_CHROMA, + locals->RequiredDPPCLKThisState, + &mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0]); + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if ((mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8)) { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + locals->Read256BlockHeightC[k], + locals->Read256BlockWidthC[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelInDETC[k], 2.0), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k] / 2.0, + mode_lib->vba.ViewportHeight[k] / 2.0, + locals->SwathWidthYThisState[k] / 2.0, + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.VMMPageSize, + mode_lib->vba.PTEBufferSizeInRequestsChroma, + mode_lib->vba.PitchC[k], + 0.0, + &locals->MacroTileWidthC[k], + &mode_lib->vba.MetaRowBytesC, + &mode_lib->vba.DPTEBytesPerRowC, + &locals->PTEBufferSizeNotExceededC[i][j][k], + locals->dpte_row_width_chroma_ub, + &locals->dpte_row_height_chroma[k], + &locals->meta_req_width_chroma[k], + &locals->meta_req_height_chroma[k], + &locals->meta_row_width_chroma[k], + &locals->meta_row_height_chroma[k], + &locals->vm_group_bytes_chroma, + &locals->dpte_group_bytes_chroma, + locals->PixelPTEReqWidthC, + locals->PixelPTEReqHeightC, + locals->PTERequestSizeC, + locals->dpde0_bytes_per_frame_ub_c, + locals->meta_pte_bytes_per_frame_ub_c); + locals->PrefetchLinesC[0][0][k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k]/2, + mode_lib->vba.VTAPsChroma[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + locals->SwathHeightCThisState[k], + mode_lib->vba.ViewportYStartC[k], + &locals->PrefillC[k], + &locals->MaxNumSwC[k]); + locals->PTEBufferSizeInRequestsForLuma = mode_lib->vba.PTEBufferSizeInRequestsLuma; + } else { + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC = 0.0; + mode_lib->vba.MetaRowBytesC = 0.0; + mode_lib->vba.DPTEBytesPerRowC = 0.0; + locals->PrefetchLinesC[0][0][k] = 0.0; + locals->PTEBufferSizeNotExceededC[i][j][k] = true; + locals->PTEBufferSizeInRequestsForLuma = mode_lib->vba.PTEBufferSizeInRequestsLuma + mode_lib->vba.PTEBufferSizeInRequestsChroma; + } + mode_lib->vba.PDEAndMetaPTEBytesPerFrameY = CalculateVMAndRowBytes( + mode_lib, + mode_lib->vba.DCCEnable[k], + locals->Read256BlockHeightY[k], + locals->Read256BlockWidthY[k], + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.SurfaceTiling[k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + mode_lib->vba.SourceScan[k], + mode_lib->vba.ViewportWidth[k], + mode_lib->vba.ViewportHeight[k], + locals->SwathWidthYThisState[k], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.VMMPageSize, + locals->PTEBufferSizeInRequestsForLuma, + mode_lib->vba.PitchY[k], + mode_lib->vba.DCCMetaPitchY[k], + &locals->MacroTileWidthY[k], + &mode_lib->vba.MetaRowBytesY, + &mode_lib->vba.DPTEBytesPerRowY, + &locals->PTEBufferSizeNotExceededY[i][j][k], + locals->dpte_row_width_luma_ub, + &locals->dpte_row_height[k], + &locals->meta_req_width[k], + &locals->meta_req_height[k], + &locals->meta_row_width[k], + &locals->meta_row_height[k], + &locals->vm_group_bytes[k], + &locals->dpte_group_bytes[k], + locals->PixelPTEReqWidthY, + locals->PixelPTEReqHeightY, + locals->PTERequestSizeY, + locals->dpde0_bytes_per_frame_ub_l, + locals->meta_pte_bytes_per_frame_ub_l); + locals->PrefetchLinesY[0][0][k] = CalculatePrefetchSourceLines( + mode_lib, + mode_lib->vba.VRatio[k], + mode_lib->vba.vtaps[k], + mode_lib->vba.Interlace[k], + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + locals->SwathHeightYThisState[k], + mode_lib->vba.ViewportYStartY[k], + &locals->PrefillY[k], + &locals->MaxNumSwY[k]); + locals->PDEAndMetaPTEBytesPerFrame[0][0][k] = + mode_lib->vba.PDEAndMetaPTEBytesPerFrameY + mode_lib->vba.PDEAndMetaPTEBytesPerFrameC; + locals->MetaRowBytes[0][0][k] = mode_lib->vba.MetaRowBytesY + mode_lib->vba.MetaRowBytesC; + locals->DPTEBytesPerRow[0][0][k] = mode_lib->vba.DPTEBytesPerRowY + mode_lib->vba.DPTEBytesPerRowC; + + CalculateActiveRowBandwidth( + mode_lib->vba.GPUVMEnable, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.VRatio[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.HTotal[k] / + mode_lib->vba.PixelClock[k], + mode_lib->vba.MetaRowBytesY, + mode_lib->vba.MetaRowBytesC, + locals->meta_row_height[k], + locals->meta_row_height_chroma[k], + mode_lib->vba.DPTEBytesPerRowY, + mode_lib->vba.DPTEBytesPerRowC, + locals->dpte_row_height[k], + locals->dpte_row_height_chroma[k], + &locals->meta_row_bw[k], + &locals->dpte_row_bw[k]); + } + mode_lib->vba.ExtraLatency = CalculateExtraLatency( + locals->UrgentRoundTripAndOutOfOrderLatencyPerState[i], + locals->TotalNumberOfActiveDPP[i][j], + mode_lib->vba.PixelChunkSizeInKByte, + locals->TotalNumberOfDCCActiveDPP[i][j], + mode_lib->vba.MetaChunkSize, + locals->ReturnBWPerState[i][0], + mode_lib->vba.GPUVMEnable, + mode_lib->vba.HostVMEnable, + mode_lib->vba.NumberOfActivePlanes, + locals->NoOfDPPThisState, + locals->dpte_group_bytes, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels); + + mode_lib->vba.TimeCalc = 24.0 / mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + if (mode_lib->vba.WritebackEnable[k] == true) { + locals->WritebackDelay[i][k] = mode_lib->vba.WritebackLatency + + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[k], + mode_lib->vba.WritebackHRatio[k], + mode_lib->vba.WritebackVRatio[k], + mode_lib->vba.WritebackLumaHTaps[k], + mode_lib->vba.WritebackLumaVTaps[k], + mode_lib->vba.WritebackChromaHTaps[k], + mode_lib->vba.WritebackChromaVTaps[k], + mode_lib->vba.WritebackDestinationWidth[k]) / locals->RequiredDISPCLK[i][j]; + } else { + locals->WritebackDelay[i][k] = 0.0; + } + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + if (mode_lib->vba.BlendingAndTiming[m] == k + && mode_lib->vba.WritebackEnable[m] + == true) { + locals->WritebackDelay[i][k] = dml_max(locals->WritebackDelay[i][k], + mode_lib->vba.WritebackLatency + CalculateWriteBackDelay( + mode_lib->vba.WritebackPixelFormat[m], + mode_lib->vba.WritebackHRatio[m], + mode_lib->vba.WritebackVRatio[m], + mode_lib->vba.WritebackLumaHTaps[m], + mode_lib->vba.WritebackLumaVTaps[m], + mode_lib->vba.WritebackChromaHTaps[m], + mode_lib->vba.WritebackChromaVTaps[m], + mode_lib->vba.WritebackDestinationWidth[m]) / locals->RequiredDISPCLK[i][j]); + } + } + } + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + for (m = 0; m <= mode_lib->vba.NumberOfActivePlanes - 1; m++) { + if (mode_lib->vba.BlendingAndTiming[k] == m) { + locals->WritebackDelay[i][k] = locals->WritebackDelay[i][m]; + } + } + } + mode_lib->vba.MaxMaxVStartup[0][0] = 0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->MaximumVStartup[0][0][k] = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k] + - dml_max(1.0, dml_ceil(locals->WritebackDelay[i][k] / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]), 1.0)); + mode_lib->vba.MaxMaxVStartup[0][0] = dml_max(mode_lib->vba.MaxMaxVStartup[0][0], locals->MaximumVStartup[0][0][k]); + } + + mode_lib->vba.NextPrefetchMode = mode_lib->vba.MinPrefetchMode; + mode_lib->vba.NextMaxVStartup = mode_lib->vba.MaxMaxVStartup[0][0]; + do { + mode_lib->vba.PrefetchMode[i][j] = mode_lib->vba.NextPrefetchMode; + mode_lib->vba.MaxVStartup = mode_lib->vba.NextMaxVStartup; + + mode_lib->vba.TWait = CalculateTWait( + mode_lib->vba.PrefetchMode[i][j], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.SREnterPlusExitTime); + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + Pipe myPipe; + HostVM myHostVM; + + if (mode_lib->vba.XFCEnabled[k] == true) { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = + CalculateRemoteSurfaceFlipDelay( + mode_lib, + mode_lib->vba.VRatio[k], + locals->SwathWidthYThisState[k], + dml_ceil(locals->BytePerPixelInDETY[k], 1.0), + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.XFCTSlvVupdateOffset, + mode_lib->vba.XFCTSlvVupdateWidth, + mode_lib->vba.XFCTSlvVreadyOffset, + mode_lib->vba.XFCXBUFLatencyTolerance, + mode_lib->vba.XFCFillBWOverhead, + mode_lib->vba.XFCSlvChunkSize, + mode_lib->vba.XFCBusTransportTime, + mode_lib->vba.TimeCalc, + mode_lib->vba.TWait, + &mode_lib->vba.SrcActiveDrainRate, + &mode_lib->vba.TInitXFill, + &mode_lib->vba.TslvChk); + } else { + mode_lib->vba.XFCRemoteSurfaceFlipDelay = 0.0; + } + + myPipe.DPPCLK = locals->RequiredDPPCLK[i][j][k]; + myPipe.DISPCLK = locals->RequiredDISPCLK[i][j]; + myPipe.PixelClock = mode_lib->vba.PixelClock[k]; + myPipe.DCFCLKDeepSleep = mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0]; + myPipe.DPPPerPlane = locals->NoOfDPP[i][j][k]; + myPipe.ScalerEnabled = mode_lib->vba.ScalerEnabled[k]; + myPipe.SourceScan = mode_lib->vba.SourceScan[k]; + myPipe.BlockWidth256BytesY = locals->Read256BlockWidthY[k]; + myPipe.BlockHeight256BytesY = locals->Read256BlockHeightY[k]; + myPipe.BlockWidth256BytesC = locals->Read256BlockWidthC[k]; + myPipe.BlockHeight256BytesC = locals->Read256BlockHeightC[k]; + myPipe.InterlaceEnable = mode_lib->vba.Interlace[k]; + myPipe.NumberOfCursors = mode_lib->vba.NumberOfCursors[k]; + myPipe.VBlank = mode_lib->vba.VTotal[k] - mode_lib->vba.VActive[k]; + myPipe.HTotal = mode_lib->vba.HTotal[k]; + + + myHostVM.Enable = mode_lib->vba.HostVMEnable; + myHostVM.MaxPageTableLevels = mode_lib->vba.HostVMMaxPageTableLevels; + myHostVM.CachedPageTableLevels = mode_lib->vba.HostVMCachedPageTableLevels; + + + mode_lib->vba.IsErrorResult[i][j][k] = CalculatePrefetchSchedule( + mode_lib, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + &myPipe, + locals->DSCDelayPerState[i][k], + mode_lib->vba.DPPCLKDelaySubtotal, + mode_lib->vba.DPPCLKDelaySCL, + mode_lib->vba.DPPCLKDelaySCLLBOnly, + mode_lib->vba.DPPCLKDelayCNVCFormater, + mode_lib->vba.DPPCLKDelayCNVCCursor, + mode_lib->vba.DISPCLKDelaySubtotal, + locals->SwathWidthYThisState[k] / mode_lib->vba.HRatio[k], + mode_lib->vba.OutputFormat[k], + mode_lib->vba.MaxInterDCNTileRepeaters, + dml_min(mode_lib->vba.MaxVStartup, locals->MaximumVStartup[0][0][k]), + locals->MaximumVStartup[0][0][k], + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.GPUVMEnable, + &myHostVM, + mode_lib->vba.DynamicMetadataEnable[k], + mode_lib->vba.DynamicMetadataLinesBeforeActiveRequired[k], + mode_lib->vba.DynamicMetadataTransmittedBytes[k], + mode_lib->vba.DCCEnable[k], + mode_lib->vba.UrgentLatency, + mode_lib->vba.ExtraLatency, + mode_lib->vba.TimeCalc, + locals->PDEAndMetaPTEBytesPerFrame[0][0][k], + locals->MetaRowBytes[0][0][k], + locals->DPTEBytesPerRow[0][0][k], + locals->PrefetchLinesY[0][0][k], + locals->SwathWidthYThisState[k], + locals->BytePerPixelInDETY[k], + locals->PrefillY[k], + locals->MaxNumSwY[k], + locals->PrefetchLinesC[0][0][k], + locals->BytePerPixelInDETC[k], + locals->PrefillC[k], + locals->MaxNumSwC[k], + locals->SwathHeightYThisState[k], + locals->SwathHeightCThisState[k], + mode_lib->vba.TWait, + mode_lib->vba.XFCEnabled[k], + mode_lib->vba.XFCRemoteSurfaceFlipDelay, + mode_lib->vba.ProgressiveToInterlaceUnitInOPP, + &locals->dst_x_after_scaler, + &locals->dst_y_after_scaler, + &locals->LineTimesForPrefetch[k], + &locals->PrefetchBW[k], + &locals->LinesForMetaPTE[k], + &locals->LinesForMetaAndDPTERow[k], + &locals->VRatioPreY[i][j][k], + &locals->VRatioPreC[i][j][k], + &locals->RequiredPrefetchPixelDataBWLuma[i][j][k], + &locals->RequiredPrefetchPixelDataBWChroma[i][j][k], + &locals->VStartupRequiredWhenNotEnoughTimeForDynamicMetadata, + &locals->Tno_bw[k], + &locals->prefetch_vmrow_bw[k], + locals->swath_width_luma_ub, + locals->swath_width_chroma_ub, + &mode_lib->vba.VUpdateOffsetPix[k], + &mode_lib->vba.VUpdateWidthPix[k], + &mode_lib->vba.VReadyOffsetPix[k]); + } + mode_lib->vba.MaximumReadBandwidthWithoutPrefetch = 0.0; + mode_lib->vba.MaximumReadBandwidthWithPrefetch = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + unsigned int m; + + locals->cursor_bw[k] = 0; + locals->cursor_bw_pre[k] = 0; + for (m = 0; m < mode_lib->vba.NumberOfCursors[k]; m++) { + locals->cursor_bw[k] = mode_lib->vba.CursorWidth[k][m] * mode_lib->vba.CursorBPP[k][m] + / 8.0 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * mode_lib->vba.VRatio[k]; + locals->cursor_bw_pre[k] = mode_lib->vba.CursorWidth[k][m] * mode_lib->vba.CursorBPP[k][m] + / 8.0 / (mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k]) * locals->VRatioPreY[i][j][k]; + } + + CalculateUrgentBurstFactor( + mode_lib->vba.DETBufferSizeInKByte, + locals->SwathHeightYThisState[k], + locals->SwathHeightCThisState[k], + locals->SwathWidthYThisState[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.UrgentLatency, + mode_lib->vba.CursorBufferSize, + mode_lib->vba.CursorWidth[k][0] + mode_lib->vba.CursorWidth[k][1], + dml_max(mode_lib->vba.CursorBPP[k][0], mode_lib->vba.CursorBPP[k][1]), + mode_lib->vba.VRatio[k], + locals->VRatioPreY[i][j][k], + locals->VRatioPreC[i][j][k], + locals->BytePerPixelInDETY[k], + locals->BytePerPixelInDETC[k], + &locals->UrgentBurstFactorCursor[k], + &locals->UrgentBurstFactorCursorPre[k], + &locals->UrgentBurstFactorLuma[k], + &locals->UrgentBurstFactorLumaPre[k], + &locals->UrgentBurstFactorChroma[k], + &locals->UrgentBurstFactorChromaPre[k], + &locals->NotEnoughUrgentLatencyHiding, + &locals->NotEnoughUrgentLatencyHidingPre); + + if (mode_lib->vba.UseUrgentBurstBandwidth == false) { + locals->UrgentBurstFactorCursor[k] = 1; + locals->UrgentBurstFactorCursorPre[k] = 1; + locals->UrgentBurstFactorLuma[k] = 1; + locals->UrgentBurstFactorLumaPre[k] = 1; + locals->UrgentBurstFactorChroma[k] = 1; + locals->UrgentBurstFactorChromaPre[k] = 1; + } + + mode_lib->vba.MaximumReadBandwidthWithoutPrefetch = mode_lib->vba.MaximumReadBandwidthWithoutPrefetch + + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k] + locals->ReadBandwidthLuma[k] + * locals->UrgentBurstFactorLuma[k] + locals->ReadBandwidthChroma[k] + * locals->UrgentBurstFactorChroma[k] + locals->meta_row_bw[k] + locals->dpte_row_bw[k]; + mode_lib->vba.MaximumReadBandwidthWithPrefetch = mode_lib->vba.MaximumReadBandwidthWithPrefetch + + dml_max3(locals->prefetch_vmrow_bw[k], + locals->ReadBandwidthLuma[k] * locals->UrgentBurstFactorLuma[k] + locals->ReadBandwidthChroma[k] + * locals->UrgentBurstFactorChroma[k] + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k] + + locals->meta_row_bw[k] + locals->dpte_row_bw[k], + locals->RequiredPrefetchPixelDataBWLuma[i][j][k] * locals->UrgentBurstFactorLumaPre[k] + + locals->RequiredPrefetchPixelDataBWChroma[i][j][k] * locals->UrgentBurstFactorChromaPre[k] + + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + } + locals->BandwidthWithoutPrefetchSupported[i][0] = true; + if (mode_lib->vba.MaximumReadBandwidthWithoutPrefetch > locals->ReturnBWPerState[i][0] + || locals->NotEnoughUrgentLatencyHiding == 1) { + locals->BandwidthWithoutPrefetchSupported[i][0] = false; + } + + locals->PrefetchSupported[i][j] = true; + if (mode_lib->vba.MaximumReadBandwidthWithPrefetch > locals->ReturnBWPerState[i][0] + || locals->NotEnoughUrgentLatencyHiding == 1 + || locals->NotEnoughUrgentLatencyHidingPre == 1) { + locals->PrefetchSupported[i][j] = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->LineTimesForPrefetch[k] < 2.0 + || locals->LinesForMetaPTE[k] >= 32.0 + || locals->LinesForMetaAndDPTERow[k] >= 16.0 + || mode_lib->vba.IsErrorResult[i][j][k] == true) { + locals->PrefetchSupported[i][j] = false; + } + } + locals->VRatioInPrefetchSupported[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->VRatioPreY[i][j][k] > 4.0 + || locals->VRatioPreC[i][j][k] > 4.0 + || mode_lib->vba.IsErrorResult[i][j][k] == true) { + locals->VRatioInPrefetchSupported[i][j] = false; + } + } + mode_lib->vba.AnyLinesForVMOrRowTooLarge = false; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (locals->LinesForMetaAndDPTERow[k] >= 16 || locals->LinesForMetaPTE[k] >= 32) { + mode_lib->vba.AnyLinesForVMOrRowTooLarge = true; + } + } + + if (mode_lib->vba.MaxVStartup <= 13 || mode_lib->vba.AnyLinesForVMOrRowTooLarge == false) { + mode_lib->vba.NextMaxVStartup = mode_lib->vba.MaxMaxVStartup[0][0]; + mode_lib->vba.NextPrefetchMode = mode_lib->vba.NextPrefetchMode + 1; + } else { + mode_lib->vba.NextMaxVStartup = mode_lib->vba.NextMaxVStartup - 1; + } + } while ((locals->PrefetchSupported[i][j] != true || locals->VRatioInPrefetchSupported[i][j] != true) + && (mode_lib->vba.NextMaxVStartup != mode_lib->vba.MaxMaxVStartup[0][0] + || mode_lib->vba.NextPrefetchMode < mode_lib->vba.MaxPrefetchMode)); + + if (locals->PrefetchSupported[i][j] == true && locals->VRatioInPrefetchSupported[i][j] == true) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = locals->ReturnBWPerState[i][0]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.BandwidthAvailableForImmediateFlip = mode_lib->vba.BandwidthAvailableForImmediateFlip + - dml_max(locals->ReadBandwidthLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthChroma[k] * locals->UrgentBurstFactorChroma[k] + + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k], + locals->RequiredPrefetchPixelDataBWLuma[i][j][k] * locals->UrgentBurstFactorLumaPre[k] + + locals->RequiredPrefetchPixelDataBWChroma[i][j][k] * locals->UrgentBurstFactorChromaPre[k] + + locals->cursor_bw_pre[k] * locals->UrgentBurstFactorCursorPre[k]); + } + mode_lib->vba.TotImmediateFlipBytes = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.TotImmediateFlipBytes = mode_lib->vba.TotImmediateFlipBytes + + locals->PDEAndMetaPTEBytesPerFrame[0][0][k] + locals->MetaRowBytes[0][0][k] + locals->DPTEBytesPerRow[0][0][k]; + } + + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + CalculateFlipSchedule( + mode_lib, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + mode_lib->vba.PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + mode_lib->vba.ExtraLatency, + mode_lib->vba.UrgentLatency, + mode_lib->vba.GPUVMMaxPageTableLevels, + mode_lib->vba.HostVMEnable, + mode_lib->vba.HostVMMaxPageTableLevels, + mode_lib->vba.HostVMCachedPageTableLevels, + mode_lib->vba.GPUVMEnable, + locals->PDEAndMetaPTEBytesPerFrame[0][0][k], + locals->MetaRowBytes[0][0][k], + locals->DPTEBytesPerRow[0][0][k], + mode_lib->vba.BandwidthAvailableForImmediateFlip, + mode_lib->vba.TotImmediateFlipBytes, + mode_lib->vba.SourcePixelFormat[k], + mode_lib->vba.HTotal[k] / mode_lib->vba.PixelClock[k], + mode_lib->vba.VRatio[k], + locals->Tno_bw[k], + mode_lib->vba.DCCEnable[k], + locals->dpte_row_height[k], + locals->meta_row_height[k], + locals->dpte_row_height_chroma[k], + locals->meta_row_height_chroma[k], + &locals->DestinationLinesToRequestVMInImmediateFlip[k], + &locals->DestinationLinesToRequestRowInImmediateFlip[k], + &locals->final_flip_bw[k], + &locals->ImmediateFlipSupportedForPipe[k]); + } + mode_lib->vba.total_dcn_read_bw_with_flip = 0.0; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.total_dcn_read_bw_with_flip = mode_lib->vba.total_dcn_read_bw_with_flip + dml_max3( + locals->prefetch_vmrow_bw[k], + locals->final_flip_bw[k] + locals->ReadBandwidthLuma[k] * locals->UrgentBurstFactorLuma[k] + + locals->ReadBandwidthChroma[k] * locals->UrgentBurstFactorChroma[k] + + locals->cursor_bw[k] * locals->UrgentBurstFactorCursor[k], + locals->final_flip_bw[k] + locals->RequiredPrefetchPixelDataBWLuma[i][j][k] + * locals->UrgentBurstFactorLumaPre[k] + locals->RequiredPrefetchPixelDataBWChroma[i][j][k] + * locals->UrgentBurstFactorChromaPre[k] + locals->cursor_bw_pre[k] + * locals->UrgentBurstFactorCursorPre[k]); + } + locals->ImmediateFlipSupportedForState[i][j] = true; + if (mode_lib->vba.total_dcn_read_bw_with_flip + > locals->ReturnBWPerState[i][0]) { + locals->ImmediateFlipSupportedForState[i][j] = false; + } + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->ImmediateFlipSupportedForPipe[k] == false) { + locals->ImmediateFlipSupportedForState[i][j] = false; + } + } + } else { + locals->ImmediateFlipSupportedForState[i][j] = false; + } + mode_lib->vba.UrgentOutOfOrderReturnPerChannel = dml_max3( + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelDataOnly, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelPixelMixedWithVMData, + mode_lib->vba.UrgentOutOfOrderReturnPerChannelVMDataOnly); + CalculateWatermarksAndDRAMSpeedChangeSupport( + mode_lib, + mode_lib->vba.PrefetchMode[i][j], + mode_lib->vba.NumberOfActivePlanes, + mode_lib->vba.MaxLineBufferLines, + mode_lib->vba.LineBufferSize, + mode_lib->vba.DPPOutputBufferPixels, + mode_lib->vba.DETBufferSizeInKByte, + mode_lib->vba.WritebackInterfaceLumaBufferSize, + mode_lib->vba.WritebackInterfaceChromaBufferSize, + mode_lib->vba.DCFCLKPerState[i], + mode_lib->vba.UrgentOutOfOrderReturnPerChannel * mode_lib->vba.NumberOfChannels, + locals->ReturnBWPerState[i][0], + mode_lib->vba.GPUVMEnable, + locals->dpte_group_bytes, + mode_lib->vba.MetaChunkSize, + mode_lib->vba.UrgentLatency, + mode_lib->vba.ExtraLatency, + mode_lib->vba.WritebackLatency, + mode_lib->vba.WritebackChunkSize, + mode_lib->vba.SOCCLKPerState[i], + mode_lib->vba.DRAMClockChangeLatency, + mode_lib->vba.SRExitTime, + mode_lib->vba.SREnterPlusExitTime, + mode_lib->vba.ProjectedDCFCLKDeepSleep[0][0], + locals->NoOfDPPThisState, + mode_lib->vba.DCCEnable, + locals->RequiredDPPCLKThisState, + locals->SwathWidthYSingleDPP, + locals->SwathHeightYThisState, + locals->ReadBandwidthLuma, + locals->SwathHeightCThisState, + locals->ReadBandwidthChroma, + mode_lib->vba.LBBitPerPixel, + locals->SwathWidthYThisState, + mode_lib->vba.HRatio, + mode_lib->vba.vtaps, + mode_lib->vba.VTAPsChroma, + mode_lib->vba.VRatio, + mode_lib->vba.HTotal, + mode_lib->vba.PixelClock, + mode_lib->vba.BlendingAndTiming, + locals->BytePerPixelInDETY, + locals->BytePerPixelInDETC, + mode_lib->vba.WritebackEnable, + mode_lib->vba.WritebackPixelFormat, + mode_lib->vba.WritebackDestinationWidth, + mode_lib->vba.WritebackDestinationHeight, + mode_lib->vba.WritebackSourceHeight, + &locals->DRAMClockChangeSupport[i][j], + &mode_lib->vba.UrgentWatermark, + &mode_lib->vba.WritebackUrgentWatermark, + &mode_lib->vba.DRAMClockChangeWatermark, + &mode_lib->vba.WritebackDRAMClockChangeWatermark, + &mode_lib->vba.StutterExitWatermark, + &mode_lib->vba.StutterEnterPlusExitWatermark, + &mode_lib->vba.MinActiveDRAMClockChangeLatencySupported); + } + } + + /*Vertical Active BW support*/ + { + double MaxTotalVActiveRDBandwidth = 0.0; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + MaxTotalVActiveRDBandwidth = MaxTotalVActiveRDBandwidth + locals->ReadBandwidth[k]; + } + for (i = 0; i <= mode_lib->vba.soc.num_states; ++i) { + locals->MaxTotalVerticalActiveAvailableBandwidth[i][0] = dml_min( + locals->IdealSDPPortBandwidthPerState[i][0] * + mode_lib->vba.MaxAveragePercentOfIdealSDPPortBWDisplayCanUseInNormalSystemOperation + / 100.0, mode_lib->vba.DRAMSpeedPerState[i] * + mode_lib->vba.NumberOfChannels * + mode_lib->vba.DRAMChannelWidth * + mode_lib->vba.MaxAveragePercentOfIdealDRAMBWDisplayCanUseInNormalSystemOperation + / 100.0); + + if (MaxTotalVActiveRDBandwidth <= locals->MaxTotalVerticalActiveAvailableBandwidth[i][0]) { + locals->TotalVerticalActiveBandwidthSupport[i][0] = true; + } else { + locals->TotalVerticalActiveBandwidthSupport[i][0] = false; + } + } + } + + /*PTE Buffer Size Check*/ + + for (i = 0; i <= mode_lib->vba.soc.num_states; i++) { + for (j = 0; j < 2; j++) { + locals->PTEBufferSizeNotExceeded[i][j] = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (locals->PTEBufferSizeNotExceededY[i][j][k] == false + || locals->PTEBufferSizeNotExceededC[i][j][k] == false) { + locals->PTEBufferSizeNotExceeded[i][j] = false; + } + } + } + } + /*Cursor Support Check*/ + + mode_lib->vba.CursorSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.CursorWidth[k][0] > 0.0) { + for (m = 0; m < mode_lib->vba.NumberOfCursors[k]; m++) { + if (mode_lib->vba.CursorBPP[k][m] == 64 && mode_lib->vba.Cursor64BppSupport == false) { + mode_lib->vba.CursorSupport = false; + } + } + } + } + /*Valid Pitch Check*/ + + mode_lib->vba.PitchSupport = true; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + locals->AlignedYPitch[k] = dml_ceil( + dml_max(mode_lib->vba.PitchY[k], mode_lib->vba.ViewportWidth[k]), + locals->MacroTileWidthY[k]); + if (locals->AlignedYPitch[k] > mode_lib->vba.PitchY[k]) { + mode_lib->vba.PitchSupport = false; + } + if (mode_lib->vba.DCCEnable[k] == true) { + locals->AlignedDCCMetaPitch[k] = dml_ceil( + dml_max( + mode_lib->vba.DCCMetaPitchY[k], + mode_lib->vba.ViewportWidth[k]), + 64.0 * locals->Read256BlockWidthY[k]); + } else { + locals->AlignedDCCMetaPitch[k] = mode_lib->vba.DCCMetaPitchY[k]; + } + if (locals->AlignedDCCMetaPitch[k] > mode_lib->vba.DCCMetaPitchY[k]) { + mode_lib->vba.PitchSupport = false; + } + if (mode_lib->vba.SourcePixelFormat[k] != dm_444_64 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_32 + && mode_lib->vba.SourcePixelFormat[k] != dm_444_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_16 + && mode_lib->vba.SourcePixelFormat[k] != dm_mono_8) { + locals->AlignedCPitch[k] = dml_ceil( + dml_max( + mode_lib->vba.PitchC[k], + mode_lib->vba.ViewportWidth[k] / 2.0), + locals->MacroTileWidthC[k]); + } else { + locals->AlignedCPitch[k] = mode_lib->vba.PitchC[k]; + } + if (locals->AlignedCPitch[k] > mode_lib->vba.PitchC[k]) { + mode_lib->vba.PitchSupport = false; + } + } + /*Mode Support, Voltage State and SOC Configuration*/ + + for (i = mode_lib->vba.soc.num_states; i >= 0; i--) { + for (j = 0; j < 2; j++) { + enum dm_validation_status status = DML_VALIDATION_OK; + + if (mode_lib->vba.ScaleRatioAndTapsSupport != true) { + status = DML_FAIL_SCALE_RATIO_TAP; + } else if (mode_lib->vba.SourceFormatPixelAndScanSupport != true) { + status = DML_FAIL_SOURCE_PIXEL_FORMAT; + } else if (locals->ViewportSizeSupport[i][0] != true) { + status = DML_FAIL_VIEWPORT_SIZE; + } else if (locals->DIOSupport[i] != true) { + status = DML_FAIL_DIO_SUPPORT; + } else if (locals->NotEnoughDSCUnits[i] != false) { + status = DML_FAIL_NOT_ENOUGH_DSC; + } else if (locals->DSCCLKRequiredMoreThanSupported[i] != false) { + status = DML_FAIL_DSC_CLK_REQUIRED; + } else if (locals->ROBSupport[i][0] != true) { + status = DML_FAIL_REORDERING_BUFFER; + } else if (locals->DISPCLK_DPPCLK_Support[i][j] != true) { + status = DML_FAIL_DISPCLK_DPPCLK; + } else if (locals->TotalAvailablePipesSupport[i][j] != true) { + status = DML_FAIL_TOTAL_AVAILABLE_PIPES; + } else if (mode_lib->vba.NumberOfOTGSupport != true) { + status = DML_FAIL_NUM_OTG; + } else if (mode_lib->vba.WritebackModeSupport != true) { + status = DML_FAIL_WRITEBACK_MODE; + } else if (mode_lib->vba.WritebackLatencySupport != true) { + status = DML_FAIL_WRITEBACK_LATENCY; + } else if (mode_lib->vba.WritebackScaleRatioAndTapsSupport != true) { + status = DML_FAIL_WRITEBACK_SCALE_RATIO_TAP; + } else if (mode_lib->vba.CursorSupport != true) { + status = DML_FAIL_CURSOR_SUPPORT; + } else if (mode_lib->vba.PitchSupport != true) { + status = DML_FAIL_PITCH_SUPPORT; + } else if (locals->TotalVerticalActiveBandwidthSupport[i][0] != true) { + status = DML_FAIL_TOTAL_V_ACTIVE_BW; + } else if (locals->PTEBufferSizeNotExceeded[i][j] != true) { + status = DML_FAIL_PTE_BUFFER_SIZE; + } else if (mode_lib->vba.NonsupportedDSCInputBPC != false) { + status = DML_FAIL_DSC_INPUT_BPC; + } else if ((mode_lib->vba.HostVMEnable != false + && locals->ImmediateFlipSupportedForState[i][j] != true)) { + status = DML_FAIL_HOST_VM_IMMEDIATE_FLIP; + } else if (locals->PrefetchSupported[i][j] != true) { + status = DML_FAIL_PREFETCH_SUPPORT; + } else if (locals->VRatioInPrefetchSupported[i][j] != true) { + status = DML_FAIL_V_RATIO_PREFETCH; + } + + if (status == DML_VALIDATION_OK) { + locals->ModeSupport[i][j] = true; + } else { + locals->ModeSupport[i][j] = false; + } + locals->ValidationStatus[i] = status; + } + } + { + unsigned int MaximumMPCCombine = 0; + mode_lib->vba.VoltageLevel = mode_lib->vba.soc.num_states + 1; + for (i = mode_lib->vba.VoltageOverrideLevel; i <= mode_lib->vba.soc.num_states; i++) { + if (locals->ModeSupport[i][0] == true || locals->ModeSupport[i][1] == true) { + mode_lib->vba.VoltageLevel = i; + if (locals->ModeSupport[i][1] == true && (locals->ModeSupport[i][0] == false + || mode_lib->vba.WhenToDoMPCCombine == dm_mpc_always_when_possible + || (mode_lib->vba.WhenToDoMPCCombine == dm_mpc_reduce_voltage_and_clocks + && ((locals->DRAMClockChangeSupport[i][1] == dm_dram_clock_change_vactive + && locals->DRAMClockChangeSupport[i][0] != dm_dram_clock_change_vactive) + || (locals->DRAMClockChangeSupport[i][1] == dm_dram_clock_change_vblank + && locals->DRAMClockChangeSupport[i][0] == dm_dram_clock_change_unsupported))))) { + MaximumMPCCombine = 1; + } else { + MaximumMPCCombine = 0; + } + break; + } + } + mode_lib->vba.ImmediateFlipSupport = + locals->ImmediateFlipSupportedForState[mode_lib->vba.VoltageLevel][MaximumMPCCombine]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + mode_lib->vba.DPPPerPlane[k] = locals->NoOfDPP[mode_lib->vba.VoltageLevel][MaximumMPCCombine][k]; + locals->DPPCLK[k] = locals->RequiredDPPCLK[mode_lib->vba.VoltageLevel][MaximumMPCCombine][k]; + } + mode_lib->vba.DISPCLK = locals->RequiredDISPCLK[mode_lib->vba.VoltageLevel][MaximumMPCCombine]; + mode_lib->vba.maxMpcComb = MaximumMPCCombine; + } + mode_lib->vba.DCFCLK = mode_lib->vba.DCFCLKPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.DRAMSpeed = mode_lib->vba.DRAMSpeedPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.FabricClock = mode_lib->vba.FabricClockPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.SOCCLK = mode_lib->vba.SOCCLKPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.ReturnBW = locals->ReturnBWPerState[mode_lib->vba.VoltageLevel][0]; + for (k = 0; k <= mode_lib->vba.NumberOfActivePlanes - 1; k++) { + if (mode_lib->vba.BlendingAndTiming[k] == k) { + mode_lib->vba.ODMCombineEnabled[k] = + locals->ODMCombineEnablePerState[mode_lib->vba.VoltageLevel][k]; + } else { + mode_lib->vba.ODMCombineEnabled[k] = false; + } + mode_lib->vba.DSCEnabled[k] = + locals->RequiresDSC[mode_lib->vba.VoltageLevel][k]; + mode_lib->vba.OutputBpp[k] = + locals->OutputBppPerState[mode_lib->vba.VoltageLevel][k]; + } +} + +static void CalculateWatermarksAndDRAMSpeedChangeSupport( + struct display_mode_lib *mode_lib, + unsigned int PrefetchMode, + unsigned int NumberOfActivePlanes, + unsigned int MaxLineBufferLines, + unsigned int LineBufferSize, + unsigned int DPPOutputBufferPixels, + double DETBufferSizeInKByte, + unsigned int WritebackInterfaceLumaBufferSize, + unsigned int WritebackInterfaceChromaBufferSize, + double DCFCLK, + double UrgentOutOfOrderReturn, + double ReturnBW, + bool GPUVMEnable, + int dpte_group_bytes[], + unsigned int MetaChunkSize, + double UrgentLatency, + double ExtraLatency, + double WritebackLatency, + double WritebackChunkSize, + double SOCCLK, + double DRAMClockChangeLatency, + double SRExitTime, + double SREnterPlusExitTime, + double DCFCLKDeepSleep, + int DPPPerPlane[], + bool DCCEnable[], + double DPPCLK[], + double SwathWidthSingleDPPY[], + unsigned int SwathHeightY[], + double ReadBandwidthPlaneLuma[], + unsigned int SwathHeightC[], + double ReadBandwidthPlaneChroma[], + unsigned int LBBitPerPixel[], + double SwathWidthY[], + double HRatio[], + unsigned int vtaps[], + unsigned int VTAPsChroma[], + double VRatio[], + unsigned int HTotal[], + double PixelClock[], + unsigned int BlendingAndTiming[], + double BytePerPixelDETY[], + double BytePerPixelDETC[], + bool WritebackEnable[], + enum source_format_class WritebackPixelFormat[], + double WritebackDestinationWidth[], + double WritebackDestinationHeight[], + double WritebackSourceHeight[], + enum clock_change_support *DRAMClockChangeSupport, + double *UrgentWatermark, + double *WritebackUrgentWatermark, + double *DRAMClockChangeWatermark, + double *WritebackDRAMClockChangeWatermark, + double *StutterExitWatermark, + double *StutterEnterPlusExitWatermark, + double *MinActiveDRAMClockChangeLatencySupported) +{ + double EffectiveLBLatencyHidingY; + double EffectiveLBLatencyHidingC; + double DPPOutputBufferLinesY; + double DPPOutputBufferLinesC; + double DETBufferSizeY; + double DETBufferSizeC; + double LinesInDETY[DC__NUM_DPP__MAX]; + double LinesInDETC; + unsigned int LinesInDETYRoundedDownToSwath[DC__NUM_DPP__MAX]; + unsigned int LinesInDETCRoundedDownToSwath; + double FullDETBufferingTimeY[DC__NUM_DPP__MAX]; + double FullDETBufferingTimeC; + double ActiveDRAMClockChangeLatencyMarginY; + double ActiveDRAMClockChangeLatencyMarginC; + double WritebackDRAMClockChangeLatencyMargin; + double PlaneWithMinActiveDRAMClockChangeMargin; + double SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank; + double FullDETBufferingTimeYStutterCriticalPlane = 0; + double TimeToFinishSwathTransferStutterCriticalPlane = 0; + unsigned int k, j; + + mode_lib->vba.TotalActiveDPP = 0; + mode_lib->vba.TotalDCCActiveDPP = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + mode_lib->vba.TotalActiveDPP = mode_lib->vba.TotalActiveDPP + DPPPerPlane[k]; + if (DCCEnable[k] == true) { + mode_lib->vba.TotalDCCActiveDPP = mode_lib->vba.TotalDCCActiveDPP + DPPPerPlane[k]; + } + } + + mode_lib->vba.TotalDataReadBandwidth = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + mode_lib->vba.TotalDataReadBandwidth = mode_lib->vba.TotalDataReadBandwidth + + ReadBandwidthPlaneLuma[k] + ReadBandwidthPlaneChroma[k]; + } + + *UrgentWatermark = UrgentLatency + ExtraLatency; + + *DRAMClockChangeWatermark = DRAMClockChangeLatency + *UrgentWatermark; + + mode_lib->vba.TotalActiveWriteback = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (WritebackEnable[k] == true) { + mode_lib->vba.TotalActiveWriteback = mode_lib->vba.TotalActiveWriteback + 1; + } + } + + if (mode_lib->vba.TotalActiveWriteback <= 1) { + *WritebackUrgentWatermark = WritebackLatency; + } else { + *WritebackUrgentWatermark = WritebackLatency + + WritebackChunkSize * 1024.0 / 32.0 / SOCCLK; + } + + if (mode_lib->vba.TotalActiveWriteback <= 1) { + *WritebackDRAMClockChangeWatermark = DRAMClockChangeLatency + WritebackLatency; + } else { + *WritebackDRAMClockChangeWatermark = DRAMClockChangeLatency + WritebackLatency + + WritebackChunkSize * 1024.0 / 32.0 / SOCCLK; + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + + mode_lib->vba.LBLatencyHidingSourceLinesY = dml_min((double) MaxLineBufferLines, + dml_floor(LineBufferSize / LBBitPerPixel[k] / (SwathWidthY[k] / dml_max(HRatio[k], 1.0)), 1)) + - (vtaps[k] - 1); + + mode_lib->vba.LBLatencyHidingSourceLinesC = dml_min((double) MaxLineBufferLines, + dml_floor(LineBufferSize / LBBitPerPixel[k] / (SwathWidthY[k] / 2 / dml_max(HRatio[k] / 2, 1.0)), 1)) + - (VTAPsChroma[k] - 1); + + EffectiveLBLatencyHidingY = mode_lib->vba.LBLatencyHidingSourceLinesY / VRatio[k] + * (HTotal[k] / PixelClock[k]); + + EffectiveLBLatencyHidingC = mode_lib->vba.LBLatencyHidingSourceLinesC + / (VRatio[k] / 2) * (HTotal[k] / PixelClock[k]); + + if (SwathWidthY[k] > 2 * DPPOutputBufferPixels) { + DPPOutputBufferLinesY = (double) DPPOutputBufferPixels / SwathWidthY[k]; + } else if (SwathWidthY[k] > DPPOutputBufferPixels) { + DPPOutputBufferLinesY = 0.5; + } else { + DPPOutputBufferLinesY = 1; + } + + if (SwathWidthY[k] / 2.0 > 2 * DPPOutputBufferPixels) { + DPPOutputBufferLinesC = (double) DPPOutputBufferPixels + / (SwathWidthY[k] / 2.0); + } else if (SwathWidthY[k] / 2.0 > DPPOutputBufferPixels) { + DPPOutputBufferLinesC = 0.5; + } else { + DPPOutputBufferLinesC = 1; + } + + CalculateDETBufferSize( + DETBufferSizeInKByte, + SwathHeightY[k], + SwathHeightC[k], + &DETBufferSizeY, + &DETBufferSizeC); + + LinesInDETY[k] = DETBufferSizeY / BytePerPixelDETY[k] / SwathWidthY[k]; + LinesInDETYRoundedDownToSwath[k] = dml_floor(LinesInDETY[k], SwathHeightY[k]); + FullDETBufferingTimeY[k] = LinesInDETYRoundedDownToSwath[k] + * (HTotal[k] / PixelClock[k]) / VRatio[k]; + if (BytePerPixelDETC[k] > 0) { + LinesInDETC = DETBufferSizeC / BytePerPixelDETC[k] / (SwathWidthY[k] / 2.0); + LinesInDETCRoundedDownToSwath = dml_floor(LinesInDETC, SwathHeightC[k]); + FullDETBufferingTimeC = LinesInDETCRoundedDownToSwath + * (HTotal[k] / PixelClock[k]) / (VRatio[k] / 2); + } else { + LinesInDETC = 0; + FullDETBufferingTimeC = 999999; + } + + ActiveDRAMClockChangeLatencyMarginY = HTotal[k] / PixelClock[k] + * DPPOutputBufferLinesY + EffectiveLBLatencyHidingY + + FullDETBufferingTimeY[k] - *DRAMClockChangeWatermark; + + if (NumberOfActivePlanes > 1) { + ActiveDRAMClockChangeLatencyMarginY = ActiveDRAMClockChangeLatencyMarginY + - (1 - 1.0 / NumberOfActivePlanes) * SwathHeightY[k] * HTotal[k] / PixelClock[k] / VRatio[k]; + } + + if (BytePerPixelDETC[k] > 0) { + ActiveDRAMClockChangeLatencyMarginC = HTotal[k] / PixelClock[k] + * DPPOutputBufferLinesC + EffectiveLBLatencyHidingC + + FullDETBufferingTimeC - *DRAMClockChangeWatermark; + if (NumberOfActivePlanes > 1) { + ActiveDRAMClockChangeLatencyMarginC = ActiveDRAMClockChangeLatencyMarginC + - (1 - 1.0 / NumberOfActivePlanes) * SwathHeightC[k] * HTotal[k] / PixelClock[k] / (VRatio[k] / 2); + } + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = dml_min( + ActiveDRAMClockChangeLatencyMarginY, + ActiveDRAMClockChangeLatencyMarginC); + } else { + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = ActiveDRAMClockChangeLatencyMarginY; + } + + if (WritebackEnable[k] == true) { + if (WritebackPixelFormat[k] == dm_444_32) { + WritebackDRAMClockChangeLatencyMargin = (WritebackInterfaceLumaBufferSize + + WritebackInterfaceChromaBufferSize) / (WritebackDestinationWidth[k] + * WritebackDestinationHeight[k] / (WritebackSourceHeight[k] * HTotal[k] + / PixelClock[k]) * 4) - *WritebackDRAMClockChangeWatermark; + } else { + WritebackDRAMClockChangeLatencyMargin = dml_min( + WritebackInterfaceLumaBufferSize * 8.0 / 10, + 2 * WritebackInterfaceChromaBufferSize * 8.0 / 10) / (WritebackDestinationWidth[k] + * WritebackDestinationHeight[k] / (WritebackSourceHeight[k] * HTotal[k] / PixelClock[k])) + - *WritebackDRAMClockChangeWatermark; + } + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] = dml_min( + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k], + WritebackDRAMClockChangeLatencyMargin); + } + } + + mode_lib->vba.MinActiveDRAMClockChangeMargin = 999999; + PlaneWithMinActiveDRAMClockChangeMargin = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] + < mode_lib->vba.MinActiveDRAMClockChangeMargin) { + mode_lib->vba.MinActiveDRAMClockChangeMargin = + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k]; + if (BlendingAndTiming[k] == k) { + PlaneWithMinActiveDRAMClockChangeMargin = k; + } else { + for (j = 0; j < NumberOfActivePlanes; ++j) { + if (BlendingAndTiming[k] == j) { + PlaneWithMinActiveDRAMClockChangeMargin = j; + } + } + } + } + } + + *MinActiveDRAMClockChangeLatencySupported = mode_lib->vba.MinActiveDRAMClockChangeMargin + DRAMClockChangeLatency; + + SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank = 999999; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (!((k == PlaneWithMinActiveDRAMClockChangeMargin) && (BlendingAndTiming[k] == k)) + && !(BlendingAndTiming[k] == PlaneWithMinActiveDRAMClockChangeMargin) + && mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] + < SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank) { + SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank = + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k]; + } + } + + mode_lib->vba.TotalNumberOfActiveOTG = 0; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (BlendingAndTiming[k] == k) { + mode_lib->vba.TotalNumberOfActiveOTG = mode_lib->vba.TotalNumberOfActiveOTG + 1; + } + } + + if (mode_lib->vba.MinActiveDRAMClockChangeMargin > 0) { + *DRAMClockChangeSupport = dm_dram_clock_change_vactive; + } else if (((mode_lib->vba.SynchronizedVBlank == true + || mode_lib->vba.TotalNumberOfActiveOTG == 1 + || SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank > 0) + && PrefetchMode == 0)) { + *DRAMClockChangeSupport = dm_dram_clock_change_vblank; + } else { + *DRAMClockChangeSupport = dm_dram_clock_change_unsupported; + } + + FullDETBufferingTimeYStutterCriticalPlane = FullDETBufferingTimeY[0]; + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (FullDETBufferingTimeY[k] <= FullDETBufferingTimeYStutterCriticalPlane) { + TimeToFinishSwathTransferStutterCriticalPlane = (SwathHeightY[k] + - (LinesInDETY[k] - LinesInDETYRoundedDownToSwath[k])) + * (HTotal[k] / PixelClock[k]) / VRatio[k]; + } + } + + *StutterExitWatermark = SRExitTime + mode_lib->vba.LastPixelOfLineExtraWatermark + + ExtraLatency + 10 / DCFCLKDeepSleep; + *StutterEnterPlusExitWatermark = dml_max( + SREnterPlusExitTime + mode_lib->vba.LastPixelOfLineExtraWatermark + + ExtraLatency + 10 / DCFCLKDeepSleep, + TimeToFinishSwathTransferStutterCriticalPlane); + +} + +static void CalculateDCFCLKDeepSleep( + struct display_mode_lib *mode_lib, + unsigned int NumberOfActivePlanes, + double BytePerPixelDETY[], + double BytePerPixelDETC[], + double VRatio[], + double SwathWidthY[], + int DPPPerPlane[], + double HRatio[], + double PixelClock[], + double PSCL_THROUGHPUT[], + double PSCL_THROUGHPUT_CHROMA[], + double DPPCLK[], + double *DCFCLKDeepSleep) +{ + unsigned int k; + double DisplayPipeLineDeliveryTimeLuma; + double DisplayPipeLineDeliveryTimeChroma; + //double DCFCLKDeepSleepPerPlane[DC__NUM_DPP__MAX]; + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (VRatio[k] <= 1) { + DisplayPipeLineDeliveryTimeLuma = SwathWidthY[k] * DPPPerPlane[k] + / HRatio[k] / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeLuma = SwathWidthY[k] / PSCL_THROUGHPUT[k] + / DPPCLK[k]; + } + if (BytePerPixelDETC[k] == 0) { + DisplayPipeLineDeliveryTimeChroma = 0; + } else { + if (VRatio[k] / 2 <= 1) { + DisplayPipeLineDeliveryTimeChroma = SwathWidthY[k] / 2.0 + * DPPPerPlane[k] / (HRatio[k] / 2) / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeChroma = SwathWidthY[k] / 2.0 + / PSCL_THROUGHPUT_CHROMA[k] / DPPCLK[k]; + } + } + + if (BytePerPixelDETC[k] > 0) { + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = dml_max( + 1.1 * SwathWidthY[k] * dml_ceil(BytePerPixelDETY[k], 1) + / 32.0 / DisplayPipeLineDeliveryTimeLuma, + 1.1 * SwathWidthY[k] / 2.0 + * dml_ceil(BytePerPixelDETC[k], 2) / 32.0 + / DisplayPipeLineDeliveryTimeChroma); + } else { + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = 1.1 * SwathWidthY[k] + * dml_ceil(BytePerPixelDETY[k], 1) / 64.0 + / DisplayPipeLineDeliveryTimeLuma; + } + mode_lib->vba.DCFCLKDeepSleepPerPlane[k] = dml_max( + mode_lib->vba.DCFCLKDeepSleepPerPlane[k], + PixelClock[k] / 16); + + } + + *DCFCLKDeepSleep = 8; + for (k = 0; k < NumberOfActivePlanes; ++k) { + *DCFCLKDeepSleep = dml_max( + *DCFCLKDeepSleep, + mode_lib->vba.DCFCLKDeepSleepPerPlane[k]); + } +} + +static void CalculateDETBufferSize( + double DETBufferSizeInKByte, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + double *DETBufferSizeY, + double *DETBufferSizeC) +{ + if (SwathHeightC == 0) { + *DETBufferSizeY = DETBufferSizeInKByte * 1024; + *DETBufferSizeC = 0; + } else if (SwathHeightY <= SwathHeightC) { + *DETBufferSizeY = DETBufferSizeInKByte * 1024 / 2; + *DETBufferSizeC = DETBufferSizeInKByte * 1024 / 2; + } else { + *DETBufferSizeY = DETBufferSizeInKByte * 1024 * 2 / 3; + *DETBufferSizeC = DETBufferSizeInKByte * 1024 / 3; + } +} + +static void CalculateUrgentBurstFactor( + unsigned int DETBufferSizeInKByte, + unsigned int SwathHeightY, + unsigned int SwathHeightC, + unsigned int SwathWidthY, + double LineTime, + double UrgentLatency, + double CursorBufferSize, + unsigned int CursorWidth, + unsigned int CursorBPP, + double VRatio, + double VRatioPreY, + double VRatioPreC, + double BytePerPixelInDETY, + double BytePerPixelInDETC, + double *UrgentBurstFactorCursor, + double *UrgentBurstFactorCursorPre, + double *UrgentBurstFactorLuma, + double *UrgentBurstFactorLumaPre, + double *UrgentBurstFactorChroma, + double *UrgentBurstFactorChromaPre, + unsigned int *NotEnoughUrgentLatencyHiding, + unsigned int *NotEnoughUrgentLatencyHidingPre) +{ + double LinesInDETLuma; + double LinesInDETChroma; + unsigned int LinesInCursorBuffer; + double CursorBufferSizeInTime; + double CursorBufferSizeInTimePre; + double DETBufferSizeInTimeLuma; + double DETBufferSizeInTimeLumaPre; + double DETBufferSizeInTimeChroma; + double DETBufferSizeInTimeChromaPre; + double DETBufferSizeY; + double DETBufferSizeC; + + *NotEnoughUrgentLatencyHiding = 0; + *NotEnoughUrgentLatencyHidingPre = 0; + + if (CursorWidth > 0) { + LinesInCursorBuffer = 1 << (unsigned int) dml_floor( + dml_log2(CursorBufferSize * 1024.0 / (CursorWidth * CursorBPP / 8.0)), 1.0); + CursorBufferSizeInTime = LinesInCursorBuffer * LineTime / VRatio; + if (CursorBufferSizeInTime - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHiding = 1; + *UrgentBurstFactorCursor = 0; + } else { + *UrgentBurstFactorCursor = CursorBufferSizeInTime + / (CursorBufferSizeInTime - UrgentLatency); + } + if (VRatioPreY > 0) { + CursorBufferSizeInTimePre = LinesInCursorBuffer * LineTime / VRatioPreY; + if (CursorBufferSizeInTimePre - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHidingPre = 1; + *UrgentBurstFactorCursorPre = 0; + } else { + *UrgentBurstFactorCursorPre = CursorBufferSizeInTimePre + / (CursorBufferSizeInTimePre - UrgentLatency); + } + } else { + *UrgentBurstFactorCursorPre = 1; + } + } + + CalculateDETBufferSize( + DETBufferSizeInKByte, + SwathHeightY, + SwathHeightC, + &DETBufferSizeY, + &DETBufferSizeC); + + LinesInDETLuma = DETBufferSizeY / BytePerPixelInDETY / SwathWidthY; + DETBufferSizeInTimeLuma = dml_floor(LinesInDETLuma, SwathHeightY) * LineTime / VRatio; + if (DETBufferSizeInTimeLuma - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHiding = 1; + *UrgentBurstFactorLuma = 0; + } else { + *UrgentBurstFactorLuma = DETBufferSizeInTimeLuma + / (DETBufferSizeInTimeLuma - UrgentLatency); + } + if (VRatioPreY > 0) { + DETBufferSizeInTimeLumaPre = dml_floor(LinesInDETLuma, SwathHeightY) * LineTime + / VRatioPreY; + if (DETBufferSizeInTimeLumaPre - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHidingPre = 1; + *UrgentBurstFactorLumaPre = 0; + } else { + *UrgentBurstFactorLumaPre = DETBufferSizeInTimeLumaPre + / (DETBufferSizeInTimeLumaPre - UrgentLatency); + } + } else { + *UrgentBurstFactorLumaPre = 1; + } + + if (BytePerPixelInDETC > 0) { + LinesInDETChroma = DETBufferSizeC / BytePerPixelInDETC / (SwathWidthY / 2); + DETBufferSizeInTimeChroma = dml_floor(LinesInDETChroma, SwathHeightC) * LineTime + / (VRatio / 2); + if (DETBufferSizeInTimeChroma - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHiding = 1; + *UrgentBurstFactorChroma = 0; + } else { + *UrgentBurstFactorChroma = DETBufferSizeInTimeChroma + / (DETBufferSizeInTimeChroma - UrgentLatency); + } + if (VRatioPreC > 0) { + DETBufferSizeInTimeChromaPre = dml_floor(LinesInDETChroma, SwathHeightC) + * LineTime / VRatioPreC; + if (DETBufferSizeInTimeChromaPre - UrgentLatency <= 0) { + *NotEnoughUrgentLatencyHidingPre = 1; + *UrgentBurstFactorChromaPre = 0; + } else { + *UrgentBurstFactorChromaPre = DETBufferSizeInTimeChromaPre + / (DETBufferSizeInTimeChromaPre - UrgentLatency); + } + } else { + *UrgentBurstFactorChromaPre = 1; + } + } +} + +static void CalculatePixelDeliveryTimes( + unsigned int NumberOfActivePlanes, + double VRatio[], + double VRatioPrefetchY[], + double VRatioPrefetchC[], + unsigned int swath_width_luma_ub[], + unsigned int swath_width_chroma_ub[], + int DPPPerPlane[], + double HRatio[], + double PixelClock[], + double PSCL_THROUGHPUT[], + double PSCL_THROUGHPUT_CHROMA[], + double DPPCLK[], + double BytePerPixelDETC[], + enum scan_direction_class SourceScan[], + unsigned int BlockWidth256BytesY[], + unsigned int BlockHeight256BytesY[], + unsigned int BlockWidth256BytesC[], + unsigned int BlockHeight256BytesC[], + double DisplayPipeLineDeliveryTimeLuma[], + double DisplayPipeLineDeliveryTimeChroma[], + double DisplayPipeLineDeliveryTimeLumaPrefetch[], + double DisplayPipeLineDeliveryTimeChromaPrefetch[], + double DisplayPipeRequestDeliveryTimeLuma[], + double DisplayPipeRequestDeliveryTimeChroma[], + double DisplayPipeRequestDeliveryTimeLumaPrefetch[], + double DisplayPipeRequestDeliveryTimeChromaPrefetch[]) +{ + double req_per_swath_ub; + unsigned int k; + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (VRatio[k] <= 1) { + DisplayPipeLineDeliveryTimeLuma[k] = swath_width_luma_ub[k] * DPPPerPlane[k] + / HRatio[k] / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeLuma[k] = swath_width_luma_ub[k] + / PSCL_THROUGHPUT[k] / DPPCLK[k]; + } + + if (BytePerPixelDETC[k] == 0) { + DisplayPipeLineDeliveryTimeChroma[k] = 0; + } else { + if (VRatio[k] / 2 <= 1) { + DisplayPipeLineDeliveryTimeChroma[k] = swath_width_chroma_ub[k] + * DPPPerPlane[k] / (HRatio[k] / 2) / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeChroma[k] = swath_width_chroma_ub[k] + / PSCL_THROUGHPUT_CHROMA[k] / DPPCLK[k]; + } + } + + if (VRatioPrefetchY[k] <= 1) { + DisplayPipeLineDeliveryTimeLumaPrefetch[k] = swath_width_luma_ub[k] + * DPPPerPlane[k] / HRatio[k] / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeLumaPrefetch[k] = swath_width_luma_ub[k] + / PSCL_THROUGHPUT[k] / DPPCLK[k]; + } + + if (BytePerPixelDETC[k] == 0) { + DisplayPipeLineDeliveryTimeChromaPrefetch[k] = 0; + } else { + if (VRatioPrefetchC[k] <= 1) { + DisplayPipeLineDeliveryTimeChromaPrefetch[k] = + swath_width_chroma_ub[k] * DPPPerPlane[k] + / (HRatio[k] / 2) / PixelClock[k]; + } else { + DisplayPipeLineDeliveryTimeChromaPrefetch[k] = + swath_width_chroma_ub[k] / PSCL_THROUGHPUT_CHROMA[k] / DPPCLK[k]; + } + } + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (SourceScan[k] == dm_horz) { + req_per_swath_ub = swath_width_luma_ub[k] / BlockWidth256BytesY[k]; + } else { + req_per_swath_ub = swath_width_luma_ub[k] / BlockHeight256BytesY[k]; + } + DisplayPipeRequestDeliveryTimeLuma[k] = DisplayPipeLineDeliveryTimeLuma[k] + / req_per_swath_ub; + DisplayPipeRequestDeliveryTimeLumaPrefetch[k] = + DisplayPipeLineDeliveryTimeLumaPrefetch[k] / req_per_swath_ub; + if (BytePerPixelDETC[k] == 0) { + DisplayPipeRequestDeliveryTimeChroma[k] = 0; + DisplayPipeRequestDeliveryTimeChromaPrefetch[k] = 0; + } else { + if (SourceScan[k] == dm_horz) { + req_per_swath_ub = swath_width_chroma_ub[k] + / BlockWidth256BytesC[k]; + } else { + req_per_swath_ub = swath_width_chroma_ub[k] + / BlockHeight256BytesC[k]; + } + DisplayPipeRequestDeliveryTimeChroma[k] = + DisplayPipeLineDeliveryTimeChroma[k] / req_per_swath_ub; + DisplayPipeRequestDeliveryTimeChromaPrefetch[k] = + DisplayPipeLineDeliveryTimeChromaPrefetch[k] / req_per_swath_ub; + } + } +} + +static void CalculateMetaAndPTETimes( + unsigned int NumberOfActivePlanes, + bool GPUVMEnable, + unsigned int MetaChunkSize, + unsigned int MinMetaChunkSizeBytes, + unsigned int GPUVMMaxPageTableLevels, + unsigned int HTotal[], + double VRatio[], + double VRatioPrefetchY[], + double VRatioPrefetchC[], + double DestinationLinesToRequestRowInVBlank[], + double DestinationLinesToRequestRowInImmediateFlip[], + double DestinationLinesToRequestVMInVBlank[], + double DestinationLinesToRequestVMInImmediateFlip[], + bool DCCEnable[], + double PixelClock[], + double BytePerPixelDETY[], + double BytePerPixelDETC[], + enum scan_direction_class SourceScan[], + unsigned int dpte_row_height[], + unsigned int dpte_row_height_chroma[], + unsigned int meta_row_width[], + unsigned int meta_row_height[], + unsigned int meta_req_width[], + unsigned int meta_req_height[], + int dpte_group_bytes[], + unsigned int PTERequestSizeY[], + unsigned int PTERequestSizeC[], + unsigned int PixelPTEReqWidthY[], + unsigned int PixelPTEReqHeightY[], + unsigned int PixelPTEReqWidthC[], + unsigned int PixelPTEReqHeightC[], + unsigned int dpte_row_width_luma_ub[], + unsigned int dpte_row_width_chroma_ub[], + unsigned int vm_group_bytes[], + unsigned int dpde0_bytes_per_frame_ub_l[], + unsigned int dpde0_bytes_per_frame_ub_c[], + unsigned int meta_pte_bytes_per_frame_ub_l[], + unsigned int meta_pte_bytes_per_frame_ub_c[], + double DST_Y_PER_PTE_ROW_NOM_L[], + double DST_Y_PER_PTE_ROW_NOM_C[], + double DST_Y_PER_META_ROW_NOM_L[], + double TimePerMetaChunkNominal[], + double TimePerMetaChunkVBlank[], + double TimePerMetaChunkFlip[], + double time_per_pte_group_nom_luma[], + double time_per_pte_group_vblank_luma[], + double time_per_pte_group_flip_luma[], + double time_per_pte_group_nom_chroma[], + double time_per_pte_group_vblank_chroma[], + double time_per_pte_group_flip_chroma[], + double TimePerVMGroupVBlank[], + double TimePerVMGroupFlip[], + double TimePerVMRequestVBlank[], + double TimePerVMRequestFlip[]) +{ + unsigned int meta_chunk_width; + unsigned int min_meta_chunk_width; + unsigned int meta_chunk_per_row_int; + unsigned int meta_row_remainder; + unsigned int meta_chunk_threshold; + unsigned int meta_chunks_per_row_ub; + unsigned int dpte_group_width_luma; + unsigned int dpte_group_width_chroma; + unsigned int dpte_groups_per_row_luma_ub; + unsigned int dpte_groups_per_row_chroma_ub; + unsigned int num_group_per_lower_vm_stage; + unsigned int num_req_per_lower_vm_stage; + unsigned int k; + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (GPUVMEnable == true) { + DST_Y_PER_PTE_ROW_NOM_L[k] = dpte_row_height[k] / VRatio[k]; + if (BytePerPixelDETC[k] == 0) { + DST_Y_PER_PTE_ROW_NOM_C[k] = 0; + } else { + DST_Y_PER_PTE_ROW_NOM_C[k] = dpte_row_height_chroma[k] / (VRatio[k] / 2); + } + } else { + DST_Y_PER_PTE_ROW_NOM_L[k] = 0; + DST_Y_PER_PTE_ROW_NOM_C[k] = 0; + } + if (DCCEnable[k] == true) { + DST_Y_PER_META_ROW_NOM_L[k] = meta_row_height[k] / VRatio[k]; + } else { + DST_Y_PER_META_ROW_NOM_L[k] = 0; + } + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (DCCEnable[k] == true) { + meta_chunk_width = MetaChunkSize * 1024 * 256 + / dml_ceil(BytePerPixelDETY[k], 1) / meta_row_height[k]; + min_meta_chunk_width = MinMetaChunkSizeBytes * 256 + / dml_ceil(BytePerPixelDETY[k], 1) / meta_row_height[k]; + meta_chunk_per_row_int = meta_row_width[k] / meta_chunk_width; + meta_row_remainder = meta_row_width[k] % meta_chunk_width; + if (SourceScan[k] == dm_horz) { + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_width[k]; + } else { + meta_chunk_threshold = 2 * min_meta_chunk_width + - meta_req_height[k]; + } + if (meta_row_remainder <= meta_chunk_threshold) { + meta_chunks_per_row_ub = meta_chunk_per_row_int + 1; + } else { + meta_chunks_per_row_ub = meta_chunk_per_row_int + 2; + } + TimePerMetaChunkNominal[k] = meta_row_height[k] / VRatio[k] * HTotal[k] + / PixelClock[k] / meta_chunks_per_row_ub; + TimePerMetaChunkVBlank[k] = DestinationLinesToRequestRowInVBlank[k] + * HTotal[k] / PixelClock[k] / meta_chunks_per_row_ub; + TimePerMetaChunkFlip[k] = DestinationLinesToRequestRowInImmediateFlip[k] + * HTotal[k] / PixelClock[k] / meta_chunks_per_row_ub; + } else { + TimePerMetaChunkNominal[k] = 0; + TimePerMetaChunkVBlank[k] = 0; + TimePerMetaChunkFlip[k] = 0; + } + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (GPUVMEnable == true) { + if (SourceScan[k] == dm_horz) { + dpte_group_width_luma = dpte_group_bytes[k] / PTERequestSizeY[k] + * PixelPTEReqWidthY[k]; + } else { + dpte_group_width_luma = dpte_group_bytes[k] / PTERequestSizeY[k] + * PixelPTEReqHeightY[k]; + } + dpte_groups_per_row_luma_ub = dml_ceil( + dpte_row_width_luma_ub[k] / dpte_group_width_luma, + 1); + time_per_pte_group_nom_luma[k] = DST_Y_PER_PTE_ROW_NOM_L[k] * HTotal[k] + / PixelClock[k] / dpte_groups_per_row_luma_ub; + time_per_pte_group_vblank_luma[k] = DestinationLinesToRequestRowInVBlank[k] + * HTotal[k] / PixelClock[k] / dpte_groups_per_row_luma_ub; + time_per_pte_group_flip_luma[k] = + DestinationLinesToRequestRowInImmediateFlip[k] * HTotal[k] + / PixelClock[k] + / dpte_groups_per_row_luma_ub; + if (BytePerPixelDETC[k] == 0) { + time_per_pte_group_nom_chroma[k] = 0; + time_per_pte_group_vblank_chroma[k] = 0; + time_per_pte_group_flip_chroma[k] = 0; + } else { + if (SourceScan[k] == dm_horz) { + dpte_group_width_chroma = dpte_group_bytes[k] + / PTERequestSizeC[k] * PixelPTEReqWidthC[k]; + } else { + dpte_group_width_chroma = dpte_group_bytes[k] + / PTERequestSizeC[k] + * PixelPTEReqHeightC[k]; + } + dpte_groups_per_row_chroma_ub = dml_ceil( + dpte_row_width_chroma_ub[k] + / dpte_group_width_chroma, + 1); + time_per_pte_group_nom_chroma[k] = DST_Y_PER_PTE_ROW_NOM_C[k] + * HTotal[k] / PixelClock[k] + / dpte_groups_per_row_chroma_ub; + time_per_pte_group_vblank_chroma[k] = + DestinationLinesToRequestRowInVBlank[k] * HTotal[k] + / PixelClock[k] + / dpte_groups_per_row_chroma_ub; + time_per_pte_group_flip_chroma[k] = + DestinationLinesToRequestRowInImmediateFlip[k] + * HTotal[k] / PixelClock[k] + / dpte_groups_per_row_chroma_ub; + } + } else { + time_per_pte_group_nom_luma[k] = 0; + time_per_pte_group_vblank_luma[k] = 0; + time_per_pte_group_flip_luma[k] = 0; + time_per_pte_group_nom_chroma[k] = 0; + time_per_pte_group_vblank_chroma[k] = 0; + time_per_pte_group_flip_chroma[k] = 0; + } + } + + for (k = 0; k < NumberOfActivePlanes; ++k) { + if (GPUVMEnable == true && (DCCEnable[k] == true || GPUVMMaxPageTableLevels > 1)) { + if (DCCEnable[k] == false) { + if (BytePerPixelDETC[k] > 0) { + num_group_per_lower_vm_stage = + dml_ceil((double) (dpde0_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (dpde0_bytes_per_frame_ub_c[k]) / (double) (vm_group_bytes[k]), 1); + } else { + num_group_per_lower_vm_stage = + dml_ceil((double) (dpde0_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1); + } + } else { + if (GPUVMMaxPageTableLevels == 1) { + if (BytePerPixelDETC[k] > 0) { + num_group_per_lower_vm_stage = + dml_ceil((double) (meta_pte_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (meta_pte_bytes_per_frame_ub_c[k]) / (double) (vm_group_bytes[k]), 1); + } else { + num_group_per_lower_vm_stage = + dml_ceil((double) (meta_pte_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1); + } + } else { + if (BytePerPixelDETC[k] > 0) { + num_group_per_lower_vm_stage = + dml_ceil((double) (dpde0_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (dpde0_bytes_per_frame_ub_c[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (meta_pte_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (meta_pte_bytes_per_frame_ub_c[k]) / (double) (vm_group_bytes[k]), 1); + } else { + num_group_per_lower_vm_stage = + dml_ceil((double) (dpde0_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1) + + dml_ceil((double) (meta_pte_bytes_per_frame_ub_l[k]) / (double) (vm_group_bytes[k]), 1); + } + } + } + + if (DCCEnable[k] == false) { + if (BytePerPixelDETC[k] > 0) { + num_req_per_lower_vm_stage = dpde0_bytes_per_frame_ub_l[k] + / 64 + dpde0_bytes_per_frame_ub_c[k] / 64; + } else { + num_req_per_lower_vm_stage = dpde0_bytes_per_frame_ub_l[k] + / 64; + } + } else { + if (GPUVMMaxPageTableLevels == 1) { + if (BytePerPixelDETC[k] > 0) { + num_req_per_lower_vm_stage = meta_pte_bytes_per_frame_ub_l[k] / 64 + + meta_pte_bytes_per_frame_ub_c[k] / 64; + } else { + num_req_per_lower_vm_stage = meta_pte_bytes_per_frame_ub_l[k] / 64; + } + } else { + if (BytePerPixelDETC[k] > 0) { + num_req_per_lower_vm_stage = dpde0_bytes_per_frame_ub_l[k] / 64 + + dpde0_bytes_per_frame_ub_c[k] / 64 + + meta_pte_bytes_per_frame_ub_l[k] / 64 + + meta_pte_bytes_per_frame_ub_c[k] / 64; + } else { + num_req_per_lower_vm_stage = dpde0_bytes_per_frame_ub_l[k] / 64 + + meta_pte_bytes_per_frame_ub_l[k] / 64; + } + } + } + + TimePerVMGroupVBlank[k] = DestinationLinesToRequestVMInVBlank[k] * HTotal[k] + / PixelClock[k] / num_group_per_lower_vm_stage; + TimePerVMGroupFlip[k] = DestinationLinesToRequestVMInImmediateFlip[k] + * HTotal[k] / PixelClock[k] / num_group_per_lower_vm_stage; + TimePerVMRequestVBlank[k] = DestinationLinesToRequestVMInVBlank[k] + * HTotal[k] / PixelClock[k] / num_req_per_lower_vm_stage; + TimePerVMRequestFlip[k] = DestinationLinesToRequestVMInImmediateFlip[k] + * HTotal[k] / PixelClock[k] / num_req_per_lower_vm_stage; + + if (GPUVMMaxPageTableLevels > 2) { + TimePerVMGroupVBlank[k] = TimePerVMGroupVBlank[k] / 2; + TimePerVMGroupFlip[k] = TimePerVMGroupFlip[k] / 2; + TimePerVMRequestVBlank[k] = TimePerVMRequestVBlank[k] / 2; + TimePerVMRequestFlip[k] = TimePerVMRequestFlip[k] / 2; + } + + } else { + TimePerVMGroupVBlank[k] = 0; + TimePerVMGroupFlip[k] = 0; + TimePerVMRequestVBlank[k] = 0; + TimePerVMRequestFlip[k] = 0; + } + } +} + +static double CalculateExtraLatency( + double UrgentRoundTripAndOutOfOrderLatency, + int TotalNumberOfActiveDPP, + int PixelChunkSizeInKByte, + int TotalNumberOfDCCActiveDPP, + int MetaChunkSize, + double ReturnBW, + bool GPUVMEnable, + bool HostVMEnable, + int NumberOfActivePlanes, + int NumberOfDPP[], + int dpte_group_bytes[], + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData, + double PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly, + int HostVMMaxPageTableLevels, + int HostVMCachedPageTableLevels) +{ + double CalculateExtraLatency; + double HostVMInefficiencyFactor; + int HostVMDynamicLevels; + + if (GPUVMEnable && HostVMEnable) { + HostVMInefficiencyFactor = + PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyPixelMixedWithVMData + / PercentOfIdealDRAMFabricAndSDPPortBWReceivedAfterUrgLatencyVMDataOnly; + HostVMDynamicLevels = HostVMMaxPageTableLevels - HostVMCachedPageTableLevels; + } else { + HostVMInefficiencyFactor = 1; + HostVMDynamicLevels = 0; + } + + CalculateExtraLatency = UrgentRoundTripAndOutOfOrderLatency + + (TotalNumberOfActiveDPP * PixelChunkSizeInKByte + + TotalNumberOfDCCActiveDPP * MetaChunkSize) * 1024.0 + / ReturnBW; + + if (GPUVMEnable) { + int k; + + for (k = 0; k < NumberOfActivePlanes; k++) { + CalculateExtraLatency = CalculateExtraLatency + + NumberOfDPP[k] * dpte_group_bytes[k] + * (1 + 8 * HostVMDynamicLevels) + * HostVMInefficiencyFactor / ReturnBW; + } + } + return CalculateExtraLatency; +} + diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.h b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.h new file mode 100644 index 000000000000..fb9548a2f894 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.h @@ -0,0 +1,32 @@ +/* + * Copyright 2017 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DML21_DISPLAY_MODE_VBA_H__ +#define __DML21_DISPLAY_MODE_VBA_H__ + +void dml21_recalculate(struct display_mode_lib *mode_lib); +void dml21_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib); + +#endif /* _DML21_DISPLAY_MODE_VBA_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c new file mode 100644 index 000000000000..a38baa73d484 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c @@ -0,0 +1,1821 @@ +/* + * Copyright 2017 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + + +#include "../display_mode_lib.h" +#include "../display_mode_vba.h" +#include "../dml_inline_defs.h" +#include "display_rq_dlg_calc_21.h" + +/* + * NOTE: + * This file is gcc-parseable HW gospel, coming straight from HW engineers. + * + * It doesn't adhere to Linux kernel style and sometimes will do things in odd + * ways. Unless there is something clearly wrong with it the code should + * remain as-is as it provides us with a guarantee from HW that it is correct. + */ + +static void calculate_ttu_cursor( + struct display_mode_lib *mode_lib, + double *refcyc_per_req_delivery_pre_cur, + double *refcyc_per_req_delivery_cur, + double refclk_freq_in_mhz, + double ref_freq_to_pix_freq, + double hscale_pixel_rate_l, + double hscl_ratio, + double vratio_pre_l, + double vratio_l, + unsigned int cur_width, + enum cursor_bpp cur_bpp); + +static unsigned int get_bytes_per_element(enum source_format_class source_format, bool is_chroma) +{ + unsigned int ret_val = 0; + + if (source_format == dm_444_16) { + if (!is_chroma) + ret_val = 2; + } else if (source_format == dm_444_32) { + if (!is_chroma) + ret_val = 4; + } else if (source_format == dm_444_64) { + if (!is_chroma) + ret_val = 8; + } else if (source_format == dm_420_8) { + if (is_chroma) + ret_val = 2; + else + ret_val = 1; + } else if (source_format == dm_420_10) { + if (is_chroma) + ret_val = 4; + else + ret_val = 2; + } else if (source_format == dm_444_8) { + ret_val = 1; + } + return ret_val; +} + +static bool is_dual_plane(enum source_format_class source_format) +{ + bool ret_val = false; + + if ((source_format == dm_420_8) || (source_format == dm_420_10)) + ret_val = true; + + return ret_val; +} + +static double get_refcyc_per_delivery( + struct display_mode_lib *mode_lib, + double refclk_freq_in_mhz, + double pclk_freq_in_mhz, + bool odm_combine, + unsigned int recout_width, + unsigned int hactive, + double vratio, + double hscale_pixel_rate, + unsigned int delivery_width, + unsigned int req_per_swath_ub) +{ + double refcyc_per_delivery = 0.0; + + if (vratio <= 1.0) { + if (odm_combine) + refcyc_per_delivery = (double) refclk_freq_in_mhz + * dml_min((double) recout_width, (double) hactive / 2.0) + / pclk_freq_in_mhz / (double) req_per_swath_ub; + else + refcyc_per_delivery = (double) refclk_freq_in_mhz * (double) recout_width + / pclk_freq_in_mhz / (double) req_per_swath_ub; + } else { + refcyc_per_delivery = (double) refclk_freq_in_mhz * (double) delivery_width + / (double) hscale_pixel_rate / (double) req_per_swath_ub; + } + + dml_print("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, refclk_freq_in_mhz); + dml_print("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, pclk_freq_in_mhz); + dml_print("DML_DLG: %s: recout_width = %d\n", __func__, recout_width); + dml_print("DML_DLG: %s: vratio = %3.2f\n", __func__, vratio); + dml_print("DML_DLG: %s: req_per_swath_ub = %d\n", __func__, req_per_swath_ub); + dml_print("DML_DLG: %s: refcyc_per_delivery= %3.2f\n", __func__, refcyc_per_delivery); + + return refcyc_per_delivery; + +} + +static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_size) +{ + if (tile_size == dm_256k_tile) + return (256 * 1024); + else if (tile_size == dm_64k_tile) + return (64 * 1024); + else + return (4 * 1024); +} + +static void extract_rq_sizing_regs( + struct display_mode_lib *mode_lib, + display_data_rq_regs_st *rq_regs, + const display_data_rq_sizing_params_st rq_sizing) +{ + dml_print("DML_DLG: %s: rq_sizing param\n", __func__); + print__data_rq_sizing_params_st(mode_lib, rq_sizing); + + rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10; + + if (rq_sizing.min_chunk_bytes == 0) + rq_regs->min_chunk_size = 0; + else + rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1; + + rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10; + if (rq_sizing.min_meta_chunk_bytes == 0) + rq_regs->min_meta_chunk_size = 0; + else + rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1; + + rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6; + rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6; +} + +static void extract_rq_regs( + struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_rq_params_st rq_param) +{ + unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024; + unsigned int detile_buf_plane1_addr = 0; + + extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l); + + rq_regs->rq_regs_l.pte_row_height_linear = dml_floor( + dml_log2(rq_param.dlg.rq_l.dpte_row_height), + 1) - 3; + + if (rq_param.yuv420) { + extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c); + rq_regs->rq_regs_c.pte_row_height_linear = dml_floor( + dml_log2(rq_param.dlg.rq_c.dpte_row_height), + 1) - 3; + } + + rq_regs->rq_regs_l.swath_height = dml_log2(rq_param.dlg.rq_l.swath_height); + rq_regs->rq_regs_c.swath_height = dml_log2(rq_param.dlg.rq_c.swath_height); + + // FIXME: take the max between luma, chroma chunk size? + // okay for now, as we are setting chunk_bytes to 8kb anyways + if (rq_param.sizing.rq_l.chunk_bytes >= 32 * 1024) { //32kb + rq_regs->drq_expansion_mode = 0; + } else { + rq_regs->drq_expansion_mode = 2; + } + rq_regs->prq_expansion_mode = 1; + rq_regs->mrq_expansion_mode = 1; + rq_regs->crq_expansion_mode = 1; + + if (rq_param.yuv420) { + if ((double) rq_param.misc.rq_l.stored_swath_bytes + / (double) rq_param.misc.rq_c.stored_swath_bytes <= 1.5) { + detile_buf_plane1_addr = (detile_buf_size_in_bytes / 2.0 / 64.0); // half to chroma + } else { + detile_buf_plane1_addr = dml_round_to_multiple( + (unsigned int) ((2.0 * detile_buf_size_in_bytes) / 3.0), + 256, + 0) / 64.0; // 2/3 to chroma + } + } + rq_regs->plane1_base_address = detile_buf_plane1_addr; +} + +static void handle_det_buf_split( + struct display_mode_lib *mode_lib, + display_rq_params_st *rq_param, + const display_pipe_source_params_st pipe_src_param) +{ + unsigned int total_swath_bytes = 0; + unsigned int swath_bytes_l = 0; + unsigned int swath_bytes_c = 0; + unsigned int full_swath_bytes_packed_l = 0; + unsigned int full_swath_bytes_packed_c = 0; + bool req128_l = false; + bool req128_c = false; + bool surf_linear = (pipe_src_param.sw_mode == dm_sw_linear); + bool surf_vert = (pipe_src_param.source_scan == dm_vert); + unsigned int log2_swath_height_l = 0; + unsigned int log2_swath_height_c = 0; + unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024; + + full_swath_bytes_packed_l = rq_param->misc.rq_l.full_swath_bytes; + full_swath_bytes_packed_c = rq_param->misc.rq_c.full_swath_bytes; + + if (rq_param->yuv420_10bpc) { + full_swath_bytes_packed_l = dml_round_to_multiple( + rq_param->misc.rq_l.full_swath_bytes * 2 / 3, + 256, + 1) + 256; + full_swath_bytes_packed_c = dml_round_to_multiple( + rq_param->misc.rq_c.full_swath_bytes * 2 / 3, + 256, + 1) + 256; + } + + if (rq_param->yuv420) { + total_swath_bytes = 2 * full_swath_bytes_packed_l + 2 * full_swath_bytes_packed_c; + + if (total_swath_bytes <= detile_buf_size_in_bytes) { //full 256b request + req128_l = false; + req128_c = false; + swath_bytes_l = full_swath_bytes_packed_l; + swath_bytes_c = full_swath_bytes_packed_c; + } else { //128b request (for luma only for yuv420 8bpc) + req128_l = true; + req128_c = false; + swath_bytes_l = full_swath_bytes_packed_l / 2; + swath_bytes_c = full_swath_bytes_packed_c; + } + // Note: assumption, the config that pass in will fit into + // the detiled buffer. + } else { + total_swath_bytes = 2 * full_swath_bytes_packed_l; + + if (total_swath_bytes <= detile_buf_size_in_bytes) + req128_l = false; + else + req128_l = true; + + swath_bytes_l = total_swath_bytes; + swath_bytes_c = 0; + } + rq_param->misc.rq_l.stored_swath_bytes = swath_bytes_l; + rq_param->misc.rq_c.stored_swath_bytes = swath_bytes_c; + + if (surf_linear) { + log2_swath_height_l = 0; + log2_swath_height_c = 0; + } else if (!surf_vert) { + log2_swath_height_l = dml_log2(rq_param->misc.rq_l.blk256_height) - req128_l; + log2_swath_height_c = dml_log2(rq_param->misc.rq_c.blk256_height) - req128_c; + } else { + log2_swath_height_l = dml_log2(rq_param->misc.rq_l.blk256_width) - req128_l; + log2_swath_height_c = dml_log2(rq_param->misc.rq_c.blk256_width) - req128_c; + } + rq_param->dlg.rq_l.swath_height = 1 << log2_swath_height_l; + rq_param->dlg.rq_c.swath_height = 1 << log2_swath_height_c; + + dml_print("DML_DLG: %s: req128_l = %0d\n", __func__, req128_l); + dml_print("DML_DLG: %s: req128_c = %0d\n", __func__, req128_c); + dml_print( + "DML_DLG: %s: full_swath_bytes_packed_l = %0d\n", + __func__, + full_swath_bytes_packed_l); + dml_print( + "DML_DLG: %s: full_swath_bytes_packed_c = %0d\n", + __func__, + full_swath_bytes_packed_c); +} + +static void get_meta_and_pte_attr( + struct display_mode_lib *mode_lib, + display_data_rq_dlg_params_st *rq_dlg_param, + display_data_rq_misc_params_st *rq_misc_param, + display_data_rq_sizing_params_st *rq_sizing_param, + unsigned int vp_width, + unsigned int vp_height, + unsigned int data_pitch, + unsigned int meta_pitch, + unsigned int source_format, + unsigned int tiling, + unsigned int macro_tile_size, + unsigned int source_scan, + unsigned int hostvm_enable, + unsigned int is_chroma) +{ + bool surf_linear = (tiling == dm_sw_linear); + bool surf_vert = (source_scan == dm_vert); + + unsigned int bytes_per_element; + unsigned int bytes_per_element_y = get_bytes_per_element( + (enum source_format_class) (source_format), + false); + unsigned int bytes_per_element_c = get_bytes_per_element( + (enum source_format_class) (source_format), + true); + + unsigned int blk256_width = 0; + unsigned int blk256_height = 0; + + unsigned int blk256_width_y = 0; + unsigned int blk256_height_y = 0; + unsigned int blk256_width_c = 0; + unsigned int blk256_height_c = 0; + unsigned int log2_bytes_per_element; + unsigned int log2_blk256_width; + unsigned int log2_blk256_height; + unsigned int blk_bytes; + unsigned int log2_blk_bytes; + unsigned int log2_blk_height; + unsigned int log2_blk_width; + unsigned int log2_meta_req_bytes; + unsigned int log2_meta_req_height; + unsigned int log2_meta_req_width; + unsigned int meta_req_width; + unsigned int meta_req_height; + unsigned int log2_meta_row_height; + unsigned int meta_row_width_ub; + unsigned int log2_meta_chunk_bytes; + unsigned int log2_meta_chunk_height; + + //full sized meta chunk width in unit of data elements + unsigned int log2_meta_chunk_width; + unsigned int log2_min_meta_chunk_bytes; + unsigned int min_meta_chunk_width; + unsigned int meta_chunk_width; + unsigned int meta_chunk_per_row_int; + unsigned int meta_row_remainder; + unsigned int meta_chunk_threshold; + unsigned int meta_blk_bytes; + unsigned int meta_blk_height; + unsigned int meta_blk_width; + unsigned int meta_surface_bytes; + unsigned int vmpg_bytes; + unsigned int meta_pte_req_per_frame_ub; + unsigned int meta_pte_bytes_per_frame_ub; + const unsigned int log2_vmpg_bytes = dml_log2(mode_lib->soc.vmm_page_size_bytes); + const unsigned int dpte_buf_in_pte_reqs = + mode_lib->ip.dpte_buffer_size_in_pte_reqs_luma + mode_lib->ip.dpte_buffer_size_in_pte_reqs_chroma; + const unsigned int pde_proc_buffer_size_64k_reqs = + mode_lib->ip.pde_proc_buffer_size_64k_reqs; + + unsigned int log2_vmpg_height = 0; + unsigned int log2_vmpg_width = 0; + unsigned int log2_dpte_req_height_ptes = 0; + unsigned int log2_dpte_req_height = 0; + unsigned int log2_dpte_req_width = 0; + unsigned int log2_dpte_row_height_linear = 0; + unsigned int log2_dpte_row_height = 0; + unsigned int log2_dpte_group_width = 0; + unsigned int dpte_row_width_ub = 0; + unsigned int dpte_req_height = 0; + unsigned int dpte_req_width = 0; + unsigned int dpte_group_width = 0; + unsigned int log2_dpte_group_bytes = 0; + unsigned int log2_dpte_group_length = 0; + unsigned int pde_buf_entries; + bool yuv420 = (source_format == dm_420_8 || source_format == dm_420_10); + + Calculate256BBlockSizes( + (enum source_format_class) (source_format), + (enum dm_swizzle_mode) (tiling), + bytes_per_element_y, + bytes_per_element_c, + &blk256_height_y, + &blk256_height_c, + &blk256_width_y, + &blk256_width_c); + + if (!is_chroma) { + blk256_width = blk256_width_y; + blk256_height = blk256_height_y; + bytes_per_element = bytes_per_element_y; + } else { + blk256_width = blk256_width_c; + blk256_height = blk256_height_c; + bytes_per_element = bytes_per_element_c; + } + + log2_bytes_per_element = dml_log2(bytes_per_element); + + dml_print("DML_DLG: %s: surf_linear = %d\n", __func__, surf_linear); + dml_print("DML_DLG: %s: surf_vert = %d\n", __func__, surf_vert); + dml_print("DML_DLG: %s: blk256_width = %d\n", __func__, blk256_width); + dml_print("DML_DLG: %s: blk256_height = %d\n", __func__, blk256_height); + + log2_blk256_width = dml_log2((double) blk256_width); + log2_blk256_height = dml_log2((double) blk256_height); + blk_bytes = surf_linear ? + 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); + log2_blk_bytes = dml_log2((double) blk_bytes); + log2_blk_height = 0; + log2_blk_width = 0; + + // remember log rule + // "+" in log is multiply + // "-" in log is divide + // "/2" is like square root + // blk is vertical biased + if (tiling != dm_sw_linear) + log2_blk_height = log2_blk256_height + + dml_ceil((double) (log2_blk_bytes - 8) / 2.0, 1); + else + log2_blk_height = 0; // blk height of 1 + + log2_blk_width = log2_blk_bytes - log2_bytes_per_element - log2_blk_height; + + if (!surf_vert) { + rq_dlg_param->swath_width_ub = dml_round_to_multiple(vp_width - 1, blk256_width, 1) + + blk256_width; + rq_dlg_param->req_per_swath_ub = rq_dlg_param->swath_width_ub >> log2_blk256_width; + } else { + rq_dlg_param->swath_width_ub = dml_round_to_multiple( + vp_height - 1, + blk256_height, + 1) + blk256_height; + rq_dlg_param->req_per_swath_ub = rq_dlg_param->swath_width_ub >> log2_blk256_height; + } + + if (!surf_vert) + rq_misc_param->full_swath_bytes = rq_dlg_param->swath_width_ub * blk256_height + * bytes_per_element; + else + rq_misc_param->full_swath_bytes = rq_dlg_param->swath_width_ub * blk256_width + * bytes_per_element; + + rq_misc_param->blk256_height = blk256_height; + rq_misc_param->blk256_width = blk256_width; + + // ------- + // meta + // ------- + log2_meta_req_bytes = 6; // meta request is 64b and is 8x8byte meta element + + // each 64b meta request for dcn is 8x8 meta elements and + // a meta element covers one 256b block of the the data surface. + log2_meta_req_height = log2_blk256_height + 3; // meta req is 8x8 byte, each byte represent 1 blk256 + log2_meta_req_width = log2_meta_req_bytes + 8 - log2_bytes_per_element + - log2_meta_req_height; + meta_req_width = 1 << log2_meta_req_width; + meta_req_height = 1 << log2_meta_req_height; + log2_meta_row_height = 0; + meta_row_width_ub = 0; + + // the dimensions of a meta row are meta_row_width x meta_row_height in elements. + // calculate upper bound of the meta_row_width + if (!surf_vert) { + log2_meta_row_height = log2_meta_req_height; + meta_row_width_ub = dml_round_to_multiple(vp_width - 1, meta_req_width, 1) + + meta_req_width; + rq_dlg_param->meta_req_per_row_ub = meta_row_width_ub / meta_req_width; + } else { + log2_meta_row_height = log2_meta_req_width; + meta_row_width_ub = dml_round_to_multiple(vp_height - 1, meta_req_height, 1) + + meta_req_height; + rq_dlg_param->meta_req_per_row_ub = meta_row_width_ub / meta_req_height; + } + rq_dlg_param->meta_bytes_per_row_ub = rq_dlg_param->meta_req_per_row_ub * 64; + + rq_dlg_param->meta_row_height = 1 << log2_meta_row_height; + + log2_meta_chunk_bytes = dml_log2(rq_sizing_param->meta_chunk_bytes); + log2_meta_chunk_height = log2_meta_row_height; + + //full sized meta chunk width in unit of data elements + log2_meta_chunk_width = log2_meta_chunk_bytes + 8 - log2_bytes_per_element + - log2_meta_chunk_height; + log2_min_meta_chunk_bytes = dml_log2(rq_sizing_param->min_meta_chunk_bytes); + min_meta_chunk_width = 1 + << (log2_min_meta_chunk_bytes + 8 - log2_bytes_per_element + - log2_meta_chunk_height); + meta_chunk_width = 1 << log2_meta_chunk_width; + meta_chunk_per_row_int = (unsigned int) (meta_row_width_ub / meta_chunk_width); + meta_row_remainder = meta_row_width_ub % meta_chunk_width; + meta_chunk_threshold = 0; + meta_blk_bytes = 4096; + meta_blk_height = blk256_height * 64; + meta_blk_width = meta_blk_bytes * 256 / bytes_per_element / meta_blk_height; + meta_surface_bytes = meta_pitch + * (dml_round_to_multiple(vp_height - 1, meta_blk_height, 1) + + meta_blk_height) * bytes_per_element / 256; + vmpg_bytes = mode_lib->soc.vmm_page_size_bytes; + meta_pte_req_per_frame_ub = (dml_round_to_multiple( + meta_surface_bytes - vmpg_bytes, + 8 * vmpg_bytes, + 1) + 8 * vmpg_bytes) / (8 * vmpg_bytes); + meta_pte_bytes_per_frame_ub = meta_pte_req_per_frame_ub * 64; //64B mpte request + rq_dlg_param->meta_pte_bytes_per_frame_ub = meta_pte_bytes_per_frame_ub; + + dml_print("DML_DLG: %s: meta_blk_height = %d\n", __func__, meta_blk_height); + dml_print("DML_DLG: %s: meta_blk_width = %d\n", __func__, meta_blk_width); + dml_print("DML_DLG: %s: meta_surface_bytes = %d\n", __func__, meta_surface_bytes); + dml_print( + "DML_DLG: %s: meta_pte_req_per_frame_ub = %d\n", + __func__, + meta_pte_req_per_frame_ub); + dml_print( + "DML_DLG: %s: meta_pte_bytes_per_frame_ub = %d\n", + __func__, + meta_pte_bytes_per_frame_ub); + + if (!surf_vert) + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_width; + else + meta_chunk_threshold = 2 * min_meta_chunk_width - meta_req_height; + + if (meta_row_remainder <= meta_chunk_threshold) + rq_dlg_param->meta_chunks_per_row_ub = meta_chunk_per_row_int + 1; + else + rq_dlg_param->meta_chunks_per_row_ub = meta_chunk_per_row_int + 2; + + // ------ + // dpte + // ------ + if (surf_linear) { + log2_vmpg_height = 0; // one line high + } else { + log2_vmpg_height = (log2_vmpg_bytes - 8) / 2 + log2_blk256_height; + } + log2_vmpg_width = log2_vmpg_bytes - log2_bytes_per_element - log2_vmpg_height; + + // only 3 possible shapes for dpte request in dimensions of ptes: 8x1, 4x2, 2x4. + if (surf_linear) { //one 64B PTE request returns 8 PTEs + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_vmpg_width + 3; + log2_dpte_req_height = 0; + } else if (log2_blk_bytes == 12) { //4KB tile means 4kB page size + //one 64B req gives 8x1 PTEs for 4KB tile + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_blk_width + 3; + log2_dpte_req_height = log2_blk_height + 0; + } else if ((log2_blk_bytes >= 16) && (log2_vmpg_bytes == 12)) { // tile block >= 64KB + //two 64B reqs of 2x4 PTEs give 16 PTEs to cover 64KB + log2_dpte_req_height_ptes = 4; + log2_dpte_req_width = log2_blk256_width + 4; // log2_64KB_width + log2_dpte_req_height = log2_blk256_height + 4; // log2_64KB_height + } else { //64KB page size and must 64KB tile block + //one 64B req gives 8x1 PTEs for 64KB tile + log2_dpte_req_height_ptes = 0; + log2_dpte_req_width = log2_blk_width + 3; + log2_dpte_req_height = log2_blk_height + 0; + } + + // The dpte request dimensions in data elements is dpte_req_width x dpte_req_height + // log2_vmpg_width is how much 1 pte represent, now calculating how much a 64b pte req represent + // That depends on the pte shape (i.e. 8x1, 4x2, 2x4) + //log2_dpte_req_height = log2_vmpg_height + log2_dpte_req_height_ptes; + //log2_dpte_req_width = log2_vmpg_width + log2_dpte_req_width_ptes; + dpte_req_height = 1 << log2_dpte_req_height; + dpte_req_width = 1 << log2_dpte_req_width; + + // calculate pitch dpte row buffer can hold + // round the result down to a power of two. + pde_buf_entries = + yuv420 ? (pde_proc_buffer_size_64k_reqs >> 1) : pde_proc_buffer_size_64k_reqs; + if (surf_linear) { + unsigned int dpte_row_height; + + log2_dpte_row_height_linear = dml_floor( + dml_log2( + dml_min( + 64 * 1024 * pde_buf_entries + / bytes_per_element, + dpte_buf_in_pte_reqs + * dpte_req_width) + / data_pitch), + 1); + + ASSERT(log2_dpte_row_height_linear >= 3); + + if (log2_dpte_row_height_linear > 7) + log2_dpte_row_height_linear = 7; + + log2_dpte_row_height = log2_dpte_row_height_linear; + // For linear, the dpte row is pitch dependent and the pte requests wrap at the pitch boundary. + // the dpte_row_width_ub is the upper bound of data_pitch*dpte_row_height in elements with this unique buffering. + dpte_row_height = 1 << log2_dpte_row_height; + dpte_row_width_ub = dml_round_to_multiple( + data_pitch * dpte_row_height - 1, + dpte_req_width, + 1) + dpte_req_width; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_width; + } else { + // the upper bound of the dpte_row_width without dependency on viewport position follows. + // for tiled mode, row height is the same as req height and row store up to vp size upper bound + if (!surf_vert) { + log2_dpte_row_height = log2_dpte_req_height; + dpte_row_width_ub = dml_round_to_multiple(vp_width - 1, dpte_req_width, 1) + + dpte_req_width; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_width; + } else { + log2_dpte_row_height = + (log2_blk_width < log2_dpte_req_width) ? + log2_blk_width : log2_dpte_req_width; + dpte_row_width_ub = dml_round_to_multiple(vp_height - 1, dpte_req_height, 1) + + dpte_req_height; + rq_dlg_param->dpte_req_per_row_ub = dpte_row_width_ub / dpte_req_height; + } + } + if (log2_blk_bytes >= 16 && log2_vmpg_bytes == 12) // tile block >= 64KB + rq_dlg_param->dpte_bytes_per_row_ub = rq_dlg_param->dpte_req_per_row_ub * 128; //2*64B dpte request + else + rq_dlg_param->dpte_bytes_per_row_ub = rq_dlg_param->dpte_req_per_row_ub * 64; //64B dpte request + + rq_dlg_param->dpte_row_height = 1 << log2_dpte_row_height; + + // the dpte_group_bytes is reduced for the specific case of vertical + // access of a tile surface that has dpte request of 8x1 ptes. + + if (hostvm_enable) + rq_sizing_param->dpte_group_bytes = 512; + else { + if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group + rq_sizing_param->dpte_group_bytes = 512; + else + //full size + rq_sizing_param->dpte_group_bytes = 2048; + } + + //since pte request size is 64byte, the number of data pte requests per full sized group is as follows. + log2_dpte_group_bytes = dml_log2(rq_sizing_param->dpte_group_bytes); + log2_dpte_group_length = log2_dpte_group_bytes - 6; //length in 64b requests + + // full sized data pte group width in elements + if (!surf_vert) + log2_dpte_group_width = log2_dpte_group_length + log2_dpte_req_width; + else + log2_dpte_group_width = log2_dpte_group_length + log2_dpte_req_height; + + //But if the tile block >=64KB and the page size is 4KB, then each dPTE request is 2*64B + if ((log2_blk_bytes >= 16) && (log2_vmpg_bytes == 12)) // tile block >= 64KB + log2_dpte_group_width = log2_dpte_group_width - 1; + + dpte_group_width = 1 << log2_dpte_group_width; + + // since dpte groups are only aligned to dpte_req_width and not dpte_group_width, + // the upper bound for the dpte groups per row is as follows. + rq_dlg_param->dpte_groups_per_row_ub = dml_ceil( + (double) dpte_row_width_ub / dpte_group_width, + 1); +} + +static void get_surf_rq_param( + struct display_mode_lib *mode_lib, + display_data_rq_sizing_params_st *rq_sizing_param, + display_data_rq_dlg_params_st *rq_dlg_param, + display_data_rq_misc_params_st *rq_misc_param, + const display_pipe_params_st pipe_param, + bool is_chroma) +{ + bool mode_422 = false; + unsigned int vp_width = 0; + unsigned int vp_height = 0; + unsigned int data_pitch = 0; + unsigned int meta_pitch = 0; + unsigned int ppe = mode_422 ? 2 : 1; + + // FIXME check if ppe apply for both luma and chroma in 422 case + if (is_chroma) { + vp_width = pipe_param.src.viewport_width_c / ppe; + vp_height = pipe_param.src.viewport_height_c; + data_pitch = pipe_param.src.data_pitch_c; + meta_pitch = pipe_param.src.meta_pitch_c; + } else { + vp_width = pipe_param.src.viewport_width / ppe; + vp_height = pipe_param.src.viewport_height; + data_pitch = pipe_param.src.data_pitch; + meta_pitch = pipe_param.src.meta_pitch; + } + + if (pipe_param.dest.odm_combine) { + unsigned int access_dir; + unsigned int full_src_vp_width; + unsigned int hactive_half; + unsigned int src_hactive_half; + access_dir = (pipe_param.src.source_scan == dm_vert); // vp access direction: horizontal or vertical accessed + hactive_half = pipe_param.dest.hactive / 2; + if (is_chroma) { + full_src_vp_width = pipe_param.scale_ratio_depth.hscl_ratio_c * pipe_param.dest.full_recout_width; + src_hactive_half = pipe_param.scale_ratio_depth.hscl_ratio_c * hactive_half; + } else { + full_src_vp_width = pipe_param.scale_ratio_depth.hscl_ratio * pipe_param.dest.full_recout_width; + src_hactive_half = pipe_param.scale_ratio_depth.hscl_ratio * hactive_half; + } + + if (access_dir == 0) { + vp_width = dml_min(full_src_vp_width, src_hactive_half); + dml_print("DML_DLG: %s: vp_width = %d\n", __func__, vp_width); + } else { + vp_height = dml_min(full_src_vp_width, src_hactive_half); + dml_print("DML_DLG: %s: vp_height = %d\n", __func__, vp_height); + + } + dml_print("DML_DLG: %s: full_src_vp_width = %d\n", __func__, full_src_vp_width); + dml_print("DML_DLG: %s: hactive_half = %d\n", __func__, hactive_half); + dml_print("DML_DLG: %s: src_hactive_half = %d\n", __func__, src_hactive_half); + } + rq_sizing_param->chunk_bytes = 8192; + + if (rq_sizing_param->chunk_bytes == 64 * 1024) + rq_sizing_param->min_chunk_bytes = 0; + else + rq_sizing_param->min_chunk_bytes = 1024; + + rq_sizing_param->meta_chunk_bytes = 2048; + rq_sizing_param->min_meta_chunk_bytes = 256; + + if (pipe_param.src.hostvm) + rq_sizing_param->mpte_group_bytes = 512; + else + rq_sizing_param->mpte_group_bytes = 2048; + + get_meta_and_pte_attr( + mode_lib, + rq_dlg_param, + rq_misc_param, + rq_sizing_param, + vp_width, + vp_height, + data_pitch, + meta_pitch, + pipe_param.src.source_format, + pipe_param.src.sw_mode, + pipe_param.src.macro_tile_size, + pipe_param.src.source_scan, + pipe_param.src.hostvm, + is_chroma); +} + +static void dml_rq_dlg_get_rq_params( + struct display_mode_lib *mode_lib, + display_rq_params_st *rq_param, + const display_pipe_params_st pipe_param) +{ + // get param for luma surface + rq_param->yuv420 = pipe_param.src.source_format == dm_420_8 + || pipe_param.src.source_format == dm_420_10; + rq_param->yuv420_10bpc = pipe_param.src.source_format == dm_420_10; + + get_surf_rq_param( + mode_lib, + &(rq_param->sizing.rq_l), + &(rq_param->dlg.rq_l), + &(rq_param->misc.rq_l), + pipe_param, + 0); + + if (is_dual_plane((enum source_format_class) (pipe_param.src.source_format))) { + // get param for chroma surface + get_surf_rq_param( + mode_lib, + &(rq_param->sizing.rq_c), + &(rq_param->dlg.rq_c), + &(rq_param->misc.rq_c), + pipe_param, + 1); + } + + // calculate how to split the det buffer space between luma and chroma + handle_det_buf_split(mode_lib, rq_param, pipe_param.src); + print__rq_params_st(mode_lib, *rq_param); +} + +void dml21_rq_dlg_get_rq_reg( + struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_pipe_params_st pipe_param) +{ + display_rq_params_st rq_param = {0}; + + memset(rq_regs, 0, sizeof(*rq_regs)); + dml_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param); + extract_rq_regs(mode_lib, rq_regs, rq_param); + + print__rq_regs_st(mode_lib, *rq_regs); +} + +// Note: currently taken in as is. +// Nice to decouple code from hw register implement and extract code that are repeated for luma and chroma. +static void dml_rq_dlg_get_dlg_params( + struct display_mode_lib *mode_lib, + const display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + display_dlg_regs_st *disp_dlg_regs, + display_ttu_regs_st *disp_ttu_regs, + const display_rq_dlg_params_st rq_dlg_param, + const display_dlg_sys_params_st dlg_sys_param, + const bool cstate_en, + const bool pstate_en) +{ + const display_pipe_source_params_st *src = &e2e_pipe_param[pipe_idx].pipe.src; + const display_pipe_dest_params_st *dst = &e2e_pipe_param[pipe_idx].pipe.dest; + const display_output_params_st *dout = &e2e_pipe_param[pipe_idx].dout; + const display_clocks_and_cfg_st *clks = &e2e_pipe_param[pipe_idx].clks_cfg; + const scaler_ratio_depth_st *scl = &e2e_pipe_param[pipe_idx].pipe.scale_ratio_depth; + const scaler_taps_st *taps = &e2e_pipe_param[pipe_idx].pipe.scale_taps; + + // ------------------------- + // Section 1.15.2.1: OTG dependent Params + // ------------------------- + // Timing + unsigned int htotal = dst->htotal; + // unsigned int hblank_start = dst.hblank_start; // TODO: Remove + unsigned int hblank_end = dst->hblank_end; + unsigned int vblank_start = dst->vblank_start; + unsigned int vblank_end = dst->vblank_end; + unsigned int min_vblank = mode_lib->ip.min_vblank_lines; + + double dppclk_freq_in_mhz = clks->dppclk_mhz; + double dispclk_freq_in_mhz = clks->dispclk_mhz; + double refclk_freq_in_mhz = clks->refclk_mhz; + double pclk_freq_in_mhz = dst->pixel_rate_mhz; + bool interlaced = dst->interlaced; + + double ref_freq_to_pix_freq = refclk_freq_in_mhz / pclk_freq_in_mhz; + + double min_dcfclk_mhz; + double t_calc_us; + double min_ttu_vblank; + + double min_dst_y_ttu_vblank; + unsigned int dlg_vblank_start; + bool dual_plane; + bool mode_422; + unsigned int access_dir; + unsigned int vp_height_l; + unsigned int vp_width_l; + unsigned int vp_height_c; + unsigned int vp_width_c; + + // Scaling + unsigned int htaps_l; + unsigned int htaps_c; + double hratio_l; + double hratio_c; + double vratio_l; + double vratio_c; + bool scl_enable; + + double line_time_in_us; + // double vinit_l; + // double vinit_c; + // double vinit_bot_l; + // double vinit_bot_c; + + // unsigned int swath_height_l; + unsigned int swath_width_ub_l; + // unsigned int dpte_bytes_per_row_ub_l; + unsigned int dpte_groups_per_row_ub_l; + // unsigned int meta_pte_bytes_per_frame_ub_l; + // unsigned int meta_bytes_per_row_ub_l; + + // unsigned int swath_height_c; + unsigned int swath_width_ub_c; + // unsigned int dpte_bytes_per_row_ub_c; + unsigned int dpte_groups_per_row_ub_c; + + unsigned int meta_chunks_per_row_ub_l; + unsigned int meta_chunks_per_row_ub_c; + unsigned int vupdate_offset; + unsigned int vupdate_width; + unsigned int vready_offset; + + unsigned int dppclk_delay_subtotal; + unsigned int dispclk_delay_subtotal; + unsigned int pixel_rate_delay_subtotal; + + unsigned int vstartup_start; + unsigned int dst_x_after_scaler; + unsigned int dst_y_after_scaler; + double line_wait; + double dst_y_prefetch; + double dst_y_per_vm_vblank; + double dst_y_per_row_vblank; + double dst_y_per_vm_flip; + double dst_y_per_row_flip; + double max_dst_y_per_vm_vblank; + double max_dst_y_per_row_vblank; + double lsw; + double vratio_pre_l; + double vratio_pre_c; + unsigned int req_per_swath_ub_l; + unsigned int req_per_swath_ub_c; + unsigned int meta_row_height_l; + unsigned int meta_row_height_c; + unsigned int swath_width_pixels_ub_l; + unsigned int swath_width_pixels_ub_c; + unsigned int scaler_rec_in_width_l; + unsigned int scaler_rec_in_width_c; + unsigned int dpte_row_height_l; + unsigned int dpte_row_height_c; + double hscale_pixel_rate_l; + double hscale_pixel_rate_c; + double min_hratio_fact_l; + double min_hratio_fact_c; + double refcyc_per_line_delivery_pre_l; + double refcyc_per_line_delivery_pre_c; + double refcyc_per_line_delivery_l; + double refcyc_per_line_delivery_c; + + double refcyc_per_req_delivery_pre_l; + double refcyc_per_req_delivery_pre_c; + double refcyc_per_req_delivery_l; + double refcyc_per_req_delivery_c; + + unsigned int full_recout_width; + double xfc_transfer_delay; + double xfc_precharge_delay; + double xfc_remote_surface_flip_latency; + double xfc_dst_y_delta_drq_limit; + double xfc_prefetch_margin; + double refcyc_per_req_delivery_pre_cur0; + double refcyc_per_req_delivery_cur0; + double refcyc_per_req_delivery_pre_cur1; + double refcyc_per_req_delivery_cur1; + + memset(disp_dlg_regs, 0, sizeof(*disp_dlg_regs)); + memset(disp_ttu_regs, 0, sizeof(*disp_ttu_regs)); + + dml_print("DML_DLG: %s: cstate_en = %d\n", __func__, cstate_en); + dml_print("DML_DLG: %s: pstate_en = %d\n", __func__, pstate_en); + + dml_print("DML_DLG: %s: dppclk_freq_in_mhz = %3.2f\n", __func__, dppclk_freq_in_mhz); + dml_print("DML_DLG: %s: dispclk_freq_in_mhz = %3.2f\n", __func__, dispclk_freq_in_mhz); + dml_print("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, refclk_freq_in_mhz); + dml_print("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, pclk_freq_in_mhz); + dml_print("DML_DLG: %s: interlaced = %d\n", __func__, interlaced); + ASSERT(ref_freq_to_pix_freq < 4.0); + + disp_dlg_regs->ref_freq_to_pix_freq = + (unsigned int) (ref_freq_to_pix_freq * dml_pow(2, 19)); + disp_dlg_regs->refcyc_per_htotal = (unsigned int) (ref_freq_to_pix_freq * (double) htotal + * dml_pow(2, 8)); + disp_dlg_regs->dlg_vblank_end = interlaced ? (vblank_end / 2) : vblank_end; // 15 bits + disp_dlg_regs->refcyc_h_blank_end = (unsigned int) ((double) hblank_end + * (double) ref_freq_to_pix_freq); + ASSERT(disp_dlg_regs->refcyc_h_blank_end < (unsigned int)dml_pow(2, 13)); + + min_dcfclk_mhz = dlg_sys_param.deepsleep_dcfclk_mhz; + t_calc_us = get_tcalc(mode_lib, e2e_pipe_param, num_pipes); + min_ttu_vblank = get_min_ttu_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + min_dst_y_ttu_vblank = min_ttu_vblank * pclk_freq_in_mhz / (double) htotal; + dlg_vblank_start = interlaced ? (vblank_start / 2) : vblank_start; + + disp_dlg_regs->min_dst_y_next_start = (unsigned int) (((double) dlg_vblank_start) * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int)dml_pow(2, 18)); + + dml_print( + "DML_DLG: %s: min_dcfclk_mhz = %3.2f\n", + __func__, + min_dcfclk_mhz); + dml_print( + "DML_DLG: %s: min_ttu_vblank = %3.2f\n", + __func__, + min_ttu_vblank); + dml_print( + "DML_DLG: %s: min_dst_y_ttu_vblank = %3.2f\n", + __func__, + min_dst_y_ttu_vblank); + dml_print( + "DML_DLG: %s: t_calc_us = %3.2f\n", + __func__, + t_calc_us); + dml_print( + "DML_DLG: %s: disp_dlg_regs->min_dst_y_next_start = 0x%0x\n", + __func__, + disp_dlg_regs->min_dst_y_next_start); + dml_print( + "DML_DLG: %s: ref_freq_to_pix_freq = %3.2f\n", + __func__, + ref_freq_to_pix_freq); + + // ------------------------- + // Section 1.15.2.2: Prefetch, Active and TTU + // ------------------------- + // Prefetch Calc + // Source + // dcc_en = src.dcc; + dual_plane = is_dual_plane((enum source_format_class) (src->source_format)); + mode_422 = false; // FIXME + access_dir = (src->source_scan == dm_vert); // vp access direction: horizontal or vertical accessed + // bytes_per_element_l = get_bytes_per_element(source_format_class(src.source_format), 0); + // bytes_per_element_c = get_bytes_per_element(source_format_class(src.source_format), 1); + vp_height_l = src->viewport_height; + vp_width_l = src->viewport_width; + vp_height_c = src->viewport_height_c; + vp_width_c = src->viewport_width_c; + + // Scaling + htaps_l = taps->htaps; + htaps_c = taps->htaps_c; + hratio_l = scl->hscl_ratio; + hratio_c = scl->hscl_ratio_c; + vratio_l = scl->vscl_ratio; + vratio_c = scl->vscl_ratio_c; + scl_enable = scl->scl_enable; + + line_time_in_us = (htotal / pclk_freq_in_mhz); + swath_width_ub_l = rq_dlg_param.rq_l.swath_width_ub; + dpte_groups_per_row_ub_l = rq_dlg_param.rq_l.dpte_groups_per_row_ub; + swath_width_ub_c = rq_dlg_param.rq_c.swath_width_ub; + dpte_groups_per_row_ub_c = rq_dlg_param.rq_c.dpte_groups_per_row_ub; + + meta_chunks_per_row_ub_l = rq_dlg_param.rq_l.meta_chunks_per_row_ub; + meta_chunks_per_row_ub_c = rq_dlg_param.rq_c.meta_chunks_per_row_ub; + vupdate_offset = dst->vupdate_offset; + vupdate_width = dst->vupdate_width; + vready_offset = dst->vready_offset; + + dppclk_delay_subtotal = mode_lib->ip.dppclk_delay_subtotal; + dispclk_delay_subtotal = mode_lib->ip.dispclk_delay_subtotal; + + if (scl_enable) + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_scl; + else + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_scl_lb_only; + + dppclk_delay_subtotal += mode_lib->ip.dppclk_delay_cnvc_formatter + + src->num_cursors * mode_lib->ip.dppclk_delay_cnvc_cursor; + + if (dout->dsc_enable) { + double dsc_delay = get_dsc_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dispclk_delay_subtotal += dsc_delay; + } + + pixel_rate_delay_subtotal = dppclk_delay_subtotal * pclk_freq_in_mhz / dppclk_freq_in_mhz + + dispclk_delay_subtotal * pclk_freq_in_mhz / dispclk_freq_in_mhz; + + vstartup_start = dst->vstartup_start; + if (interlaced) { + if (vstartup_start / 2.0 + - (double) (vready_offset + vupdate_width + vupdate_offset) / htotal + <= vblank_end / 2.0) + disp_dlg_regs->vready_after_vcount0 = 1; + else + disp_dlg_regs->vready_after_vcount0 = 0; + } else { + if (vstartup_start + - (double) (vready_offset + vupdate_width + vupdate_offset) / htotal + <= vblank_end) + disp_dlg_regs->vready_after_vcount0 = 1; + else + disp_dlg_regs->vready_after_vcount0 = 0; + } + + // TODO: Where is this coming from? + if (interlaced) + vstartup_start = vstartup_start / 2; + + // TODO: What if this min_vblank doesn't match the value in the dml_config_settings.cpp? + if (vstartup_start >= min_vblank) { + dml_print( + "WARNING: DML_DLG: %s: vblank_start=%d vblank_end=%d\n", + __func__, + vblank_start, + vblank_end); + dml_print( + "WARNING: DML_DLG: %s: vstartup_start=%d should be less than min_vblank=%d\n", + __func__, + vstartup_start, + min_vblank); + min_vblank = vstartup_start + 1; + dml_print( + "WARNING: DML_DLG: %s: vstartup_start=%d should be less than min_vblank=%d\n", + __func__, + vstartup_start, + min_vblank); + } + + dst_x_after_scaler = get_dst_x_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dst_y_after_scaler = get_dst_y_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dml_print("DML_DLG: %s: htotal = %d\n", __func__, htotal); + dml_print( + "DML_DLG: %s: pixel_rate_delay_subtotal = %d\n", + __func__, + pixel_rate_delay_subtotal); + dml_print( + "DML_DLG: %s: dst_x_after_scaler = %d\n", + __func__, + dst_x_after_scaler); + dml_print( + "DML_DLG: %s: dst_y_after_scaler = %d\n", + __func__, + dst_y_after_scaler); + + // Lwait + // TODO: Should this be urgent_latency_pixel_mixed_with_vm_data_us? + line_wait = mode_lib->soc.urgent_latency_pixel_data_only_us; + if (cstate_en) + line_wait = dml_max(mode_lib->soc.sr_enter_plus_exit_time_us, line_wait); + if (pstate_en) + line_wait = dml_max( + mode_lib->soc.dram_clock_change_latency_us + + mode_lib->soc.urgent_latency_pixel_data_only_us, // TODO: Should this be urgent_latency_pixel_mixed_with_vm_data_us? + line_wait); + line_wait = line_wait / line_time_in_us; + + dst_y_prefetch = get_dst_y_prefetch(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dml_print("DML_DLG: %s: dst_y_prefetch (after rnd) = %3.2f\n", __func__, dst_y_prefetch); + + dst_y_per_vm_vblank = get_dst_y_per_vm_vblank( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + dst_y_per_row_vblank = get_dst_y_per_row_vblank( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + dst_y_per_vm_flip = get_dst_y_per_vm_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + dst_y_per_row_flip = get_dst_y_per_row_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + max_dst_y_per_vm_vblank = 32.0; + max_dst_y_per_row_vblank = 16.0; + + // magic! + if (htotal <= 75) { + min_vblank = 300; + max_dst_y_per_vm_vblank = 100.0; + max_dst_y_per_row_vblank = 100.0; + } + + dml_print("DML_DLG: %s: dst_y_per_vm_flip = %3.2f\n", __func__, dst_y_per_vm_flip); + dml_print("DML_DLG: %s: dst_y_per_row_flip = %3.2f\n", __func__, dst_y_per_row_flip); + dml_print("DML_DLG: %s: dst_y_per_vm_vblank = %3.2f\n", __func__, dst_y_per_vm_vblank); + dml_print("DML_DLG: %s: dst_y_per_row_vblank = %3.2f\n", __func__, dst_y_per_row_vblank); + + ASSERT(dst_y_per_vm_vblank < max_dst_y_per_vm_vblank); + ASSERT(dst_y_per_row_vblank < max_dst_y_per_row_vblank); + + ASSERT(dst_y_prefetch > (dst_y_per_vm_vblank + dst_y_per_row_vblank)); + lsw = dst_y_prefetch - (dst_y_per_vm_vblank + dst_y_per_row_vblank); + + dml_print("DML_DLG: %s: lsw = %3.2f\n", __func__, lsw); + + vratio_pre_l = get_vratio_prefetch_l(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + vratio_pre_c = get_vratio_prefetch_c(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + + dml_print("DML_DLG: %s: vratio_pre_l=%3.2f\n", __func__, vratio_pre_l); + dml_print("DML_DLG: %s: vratio_pre_c=%3.2f\n", __func__, vratio_pre_c); + + // Active + req_per_swath_ub_l = rq_dlg_param.rq_l.req_per_swath_ub; + req_per_swath_ub_c = rq_dlg_param.rq_c.req_per_swath_ub; + meta_row_height_l = rq_dlg_param.rq_l.meta_row_height; + meta_row_height_c = rq_dlg_param.rq_c.meta_row_height; + swath_width_pixels_ub_l = 0; + swath_width_pixels_ub_c = 0; + scaler_rec_in_width_l = 0; + scaler_rec_in_width_c = 0; + dpte_row_height_l = rq_dlg_param.rq_l.dpte_row_height; + dpte_row_height_c = rq_dlg_param.rq_c.dpte_row_height; + + if (mode_422) { + swath_width_pixels_ub_l = swath_width_ub_l * 2; // *2 for 2 pixel per element + swath_width_pixels_ub_c = swath_width_ub_c * 2; + } else { + swath_width_pixels_ub_l = swath_width_ub_l * 1; + swath_width_pixels_ub_c = swath_width_ub_c * 1; + } + + hscale_pixel_rate_l = 0.; + hscale_pixel_rate_c = 0.; + min_hratio_fact_l = 1.0; + min_hratio_fact_c = 1.0; + + if (htaps_l <= 1) + min_hratio_fact_l = 2.0; + else if (htaps_l <= 6) { + if ((hratio_l * 2.0) > 4.0) + min_hratio_fact_l = 4.0; + else + min_hratio_fact_l = hratio_l * 2.0; + } else { + if (hratio_l > 4.0) + min_hratio_fact_l = 4.0; + else + min_hratio_fact_l = hratio_l; + } + + hscale_pixel_rate_l = min_hratio_fact_l * dppclk_freq_in_mhz; + + if (htaps_c <= 1) + min_hratio_fact_c = 2.0; + else if (htaps_c <= 6) { + if ((hratio_c * 2.0) > 4.0) + min_hratio_fact_c = 4.0; + else + min_hratio_fact_c = hratio_c * 2.0; + } else { + if (hratio_c > 4.0) + min_hratio_fact_c = 4.0; + else + min_hratio_fact_c = hratio_c; + } + + hscale_pixel_rate_c = min_hratio_fact_c * dppclk_freq_in_mhz; + + refcyc_per_line_delivery_pre_l = 0.; + refcyc_per_line_delivery_pre_c = 0.; + refcyc_per_line_delivery_l = 0.; + refcyc_per_line_delivery_c = 0.; + + refcyc_per_req_delivery_pre_l = 0.; + refcyc_per_req_delivery_pre_c = 0.; + refcyc_per_req_delivery_l = 0.; + refcyc_per_req_delivery_c = 0.; + + full_recout_width = 0; + // In ODM + if (src->is_hsplit) { + // This "hack" is only allowed (and valid) for MPC combine. In ODM + // combine, you MUST specify the full_recout_width...according to Oswin + if (dst->full_recout_width == 0 && !dst->odm_combine) { + dml_print( + "DML_DLG: %s: Warning: full_recout_width not set in hsplit mode\n", + __func__); + full_recout_width = dst->recout_width * 2; // assume half split for dcn1 + } else + full_recout_width = dst->full_recout_width; + } else + full_recout_width = dst->recout_width; + + // As of DCN2, mpc_combine and odm_combine are mutually exclusive + refcyc_per_line_delivery_pre_l = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_l, + hscale_pixel_rate_l, + swath_width_pixels_ub_l, + 1); // per line + + refcyc_per_line_delivery_l = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_l, + hscale_pixel_rate_l, + swath_width_pixels_ub_l, + 1); // per line + + dml_print("DML_DLG: %s: full_recout_width = %d\n", __func__, full_recout_width); + dml_print( + "DML_DLG: %s: hscale_pixel_rate_l = %3.2f\n", + __func__, + hscale_pixel_rate_l); + dml_print( + "DML_DLG: %s: refcyc_per_line_delivery_pre_l = %3.2f\n", + __func__, + refcyc_per_line_delivery_pre_l); + dml_print( + "DML_DLG: %s: refcyc_per_line_delivery_l = %3.2f\n", + __func__, + refcyc_per_line_delivery_l); + + if (dual_plane) { + refcyc_per_line_delivery_pre_c = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_c, + hscale_pixel_rate_c, + swath_width_pixels_ub_c, + 1); // per line + + refcyc_per_line_delivery_c = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_c, + hscale_pixel_rate_c, + swath_width_pixels_ub_c, + 1); // per line + + dml_print( + "DML_DLG: %s: refcyc_per_line_delivery_pre_c = %3.2f\n", + __func__, + refcyc_per_line_delivery_pre_c); + dml_print( + "DML_DLG: %s: refcyc_per_line_delivery_c = %3.2f\n", + __func__, + refcyc_per_line_delivery_c); + } + + // TTU - Luma / Chroma + if (access_dir) { // vertical access + scaler_rec_in_width_l = vp_height_l; + scaler_rec_in_width_c = vp_height_c; + } else { + scaler_rec_in_width_l = vp_width_l; + scaler_rec_in_width_c = vp_width_c; + } + + refcyc_per_req_delivery_pre_l = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_l, + hscale_pixel_rate_l, + scaler_rec_in_width_l, + req_per_swath_ub_l); // per req + refcyc_per_req_delivery_l = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_l, + hscale_pixel_rate_l, + scaler_rec_in_width_l, + req_per_swath_ub_l); // per req + + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_pre_l = %3.2f\n", + __func__, + refcyc_per_req_delivery_pre_l); + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_l = %3.2f\n", + __func__, + refcyc_per_req_delivery_l); + + ASSERT(refcyc_per_req_delivery_pre_l < dml_pow(2, 13)); + ASSERT(refcyc_per_req_delivery_l < dml_pow(2, 13)); + + if (dual_plane) { + refcyc_per_req_delivery_pre_c = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_pre_c, + hscale_pixel_rate_c, + scaler_rec_in_width_c, + req_per_swath_ub_c); // per req + refcyc_per_req_delivery_c = get_refcyc_per_delivery( + mode_lib, + refclk_freq_in_mhz, + pclk_freq_in_mhz, + dst->odm_combine, + full_recout_width, + dst->hactive, + vratio_c, + hscale_pixel_rate_c, + scaler_rec_in_width_c, + req_per_swath_ub_c); // per req + + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_pre_c = %3.2f\n", + __func__, + refcyc_per_req_delivery_pre_c); + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_c = %3.2f\n", + __func__, + refcyc_per_req_delivery_c); + + ASSERT(refcyc_per_req_delivery_pre_c < dml_pow(2, 13)); + ASSERT(refcyc_per_req_delivery_c < dml_pow(2, 13)); + } + + // XFC + xfc_transfer_delay = get_xfc_transfer_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + xfc_precharge_delay = get_xfc_precharge_delay( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + xfc_remote_surface_flip_latency = get_xfc_remote_surface_flip_latency( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + xfc_dst_y_delta_drq_limit = xfc_remote_surface_flip_latency; + xfc_prefetch_margin = get_xfc_prefetch_margin( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx); + + // TTU - Cursor + refcyc_per_req_delivery_pre_cur0 = 0.0; + refcyc_per_req_delivery_cur0 = 0.0; + if (src->num_cursors > 0) { + calculate_ttu_cursor( + mode_lib, + &refcyc_per_req_delivery_pre_cur0, + &refcyc_per_req_delivery_cur0, + refclk_freq_in_mhz, + ref_freq_to_pix_freq, + hscale_pixel_rate_l, + scl->hscl_ratio, + vratio_pre_l, + vratio_l, + src->cur0_src_width, + (enum cursor_bpp) (src->cur0_bpp)); + } + + refcyc_per_req_delivery_pre_cur1 = 0.0; + refcyc_per_req_delivery_cur1 = 0.0; + if (src->num_cursors > 1) { + calculate_ttu_cursor( + mode_lib, + &refcyc_per_req_delivery_pre_cur1, + &refcyc_per_req_delivery_cur1, + refclk_freq_in_mhz, + ref_freq_to_pix_freq, + hscale_pixel_rate_l, + scl->hscl_ratio, + vratio_pre_l, + vratio_l, + src->cur1_src_width, + (enum cursor_bpp) (src->cur1_bpp)); + } + + // TTU - Misc + // all hard-coded + + // Assignment to register structures + disp_dlg_regs->dst_y_after_scaler = dst_y_after_scaler; // in terms of line + disp_dlg_regs->refcyc_x_after_scaler = dst_x_after_scaler * ref_freq_to_pix_freq; // in terms of refclk + ASSERT(disp_dlg_regs->refcyc_x_after_scaler < (unsigned int)dml_pow(2, 13)); + disp_dlg_regs->dst_y_prefetch = (unsigned int) (dst_y_prefetch * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_vm_vblank = (unsigned int) (dst_y_per_vm_vblank * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_row_vblank = (unsigned int) (dst_y_per_row_vblank * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_vm_flip = (unsigned int) (dst_y_per_vm_flip * dml_pow(2, 2)); + disp_dlg_regs->dst_y_per_row_flip = (unsigned int) (dst_y_per_row_flip * dml_pow(2, 2)); + + disp_dlg_regs->vratio_prefetch = (unsigned int) (vratio_pre_l * dml_pow(2, 19)); + disp_dlg_regs->vratio_prefetch_c = (unsigned int) (vratio_pre_c * dml_pow(2, 19)); + + dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_vblank = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_vblank); + dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_vblank = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_vblank); + dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_flip); + dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_flip); + + disp_dlg_regs->refcyc_per_pte_group_vblank_l = + (unsigned int) (dst_y_per_row_vblank * (double) htotal + * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_l); + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)dml_pow(2, 13)); + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_vblank_c = (unsigned int) (dst_y_per_row_vblank + * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_c); + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_c + < (unsigned int)dml_pow(2, 13)); + } + + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l = + (unsigned int) (dst_y_per_row_vblank * (double) htotal + * ref_freq_to_pix_freq / (double) meta_chunks_per_row_ub_l); + ASSERT(disp_dlg_regs->refcyc_per_meta_chunk_vblank_l < (unsigned int)dml_pow(2, 13)); + + disp_dlg_regs->refcyc_per_meta_chunk_vblank_c = + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l; // dcc for 4:2:0 is not supported in dcn1.0. assigned to be the same as _l for now + + disp_dlg_regs->refcyc_per_pte_group_flip_l = (unsigned int) (dst_y_per_row_flip * htotal + * ref_freq_to_pix_freq) / dpte_groups_per_row_ub_l; + disp_dlg_regs->refcyc_per_meta_chunk_flip_l = (unsigned int) (dst_y_per_row_flip * htotal + * ref_freq_to_pix_freq) / meta_chunks_per_row_ub_l; + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_flip_c = (unsigned int) (dst_y_per_row_flip + * htotal * ref_freq_to_pix_freq) / dpte_groups_per_row_ub_c; + disp_dlg_regs->refcyc_per_meta_chunk_flip_c = (unsigned int) (dst_y_per_row_flip + * htotal * ref_freq_to_pix_freq) / meta_chunks_per_row_ub_c; + } + + disp_dlg_regs->refcyc_per_vm_group_vblank = get_refcyc_per_vm_group_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; + disp_dlg_regs->refcyc_per_vm_group_flip = get_refcyc_per_vm_group_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; + disp_dlg_regs->refcyc_per_vm_req_vblank = get_refcyc_per_vm_req_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; + disp_dlg_regs->refcyc_per_vm_req_flip = get_refcyc_per_vm_req_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; + + // Clamp to max for now + if (disp_dlg_regs->refcyc_per_vm_group_vblank >= (unsigned int)dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_vm_group_vblank = dml_pow(2, 23) - 1; + + if (disp_dlg_regs->refcyc_per_vm_group_flip >= (unsigned int)dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_vm_group_flip = dml_pow(2, 23) - 1; + + if (disp_dlg_regs->refcyc_per_vm_req_vblank >= (unsigned int)dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_vm_req_vblank = dml_pow(2, 23) - 1; + + if (disp_dlg_regs->refcyc_per_vm_req_flip >= (unsigned int)dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_vm_req_flip = dml_pow(2, 23) - 1; + disp_dlg_regs->dst_y_per_pte_row_nom_l = (unsigned int) ((double) dpte_row_height_l + / (double) vratio_l * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->dst_y_per_pte_row_nom_l < (unsigned int)dml_pow(2, 17)); + + if (dual_plane) { + disp_dlg_regs->dst_y_per_pte_row_nom_c = (unsigned int) ((double) dpte_row_height_c + / (double) vratio_c * dml_pow(2, 2)); + if (disp_dlg_regs->dst_y_per_pte_row_nom_c >= (unsigned int) dml_pow(2, 17)) { + dml_print( + "DML_DLG: %s: Warning dst_y_per_pte_row_nom_c %u larger than supported by register format U15.2 %u\n", + __func__, + disp_dlg_regs->dst_y_per_pte_row_nom_c, + (unsigned int)dml_pow(2, 17) - 1); + } + } + + disp_dlg_regs->dst_y_per_meta_row_nom_l = (unsigned int) ((double) meta_row_height_l + / (double) vratio_l * dml_pow(2, 2)); + ASSERT(disp_dlg_regs->dst_y_per_meta_row_nom_l < (unsigned int)dml_pow(2, 17)); + + disp_dlg_regs->dst_y_per_meta_row_nom_c = disp_dlg_regs->dst_y_per_meta_row_nom_l; // TODO: dcc for 4:2:0 is not supported in dcn1.0. assigned to be the same as _l for now + + dml_print( + "DML: Trow: %fus\n", + line_time_in_us * (double)dpte_row_height_l / (double)vratio_l); + + disp_dlg_regs->refcyc_per_pte_group_nom_l = (unsigned int) ((double) dpte_row_height_l + / (double) vratio_l * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_l); + if (disp_dlg_regs->refcyc_per_pte_group_nom_l >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_pte_group_nom_l = dml_pow(2, 23) - 1; + disp_dlg_regs->refcyc_per_meta_chunk_nom_l = (unsigned int) ((double) meta_row_height_l + / (double) vratio_l * (double) htotal * ref_freq_to_pix_freq + / (double) meta_chunks_per_row_ub_l); + if (disp_dlg_regs->refcyc_per_meta_chunk_nom_l >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_meta_chunk_nom_l = dml_pow(2, 23) - 1; + + if (dual_plane) { + disp_dlg_regs->refcyc_per_pte_group_nom_c = + (unsigned int) ((double) dpte_row_height_c / (double) vratio_c + * (double) htotal * ref_freq_to_pix_freq + / (double) dpte_groups_per_row_ub_c); + if (disp_dlg_regs->refcyc_per_pte_group_nom_c >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_pte_group_nom_c = dml_pow(2, 23) - 1; + + // TODO: Is this the right calculation? Does htotal need to be halved? + disp_dlg_regs->refcyc_per_meta_chunk_nom_c = + (unsigned int) ((double) meta_row_height_c / (double) vratio_c + * (double) htotal * ref_freq_to_pix_freq + / (double) meta_chunks_per_row_ub_c); + if (disp_dlg_regs->refcyc_per_meta_chunk_nom_c >= (unsigned int) dml_pow(2, 23)) + disp_dlg_regs->refcyc_per_meta_chunk_nom_c = dml_pow(2, 23) - 1; + } + + disp_dlg_regs->refcyc_per_line_delivery_pre_l = (unsigned int) dml_floor( + refcyc_per_line_delivery_pre_l, 1); + disp_dlg_regs->refcyc_per_line_delivery_l = (unsigned int) dml_floor( + refcyc_per_line_delivery_l, 1); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_l < (unsigned int)dml_pow(2, 13)); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_l < (unsigned int)dml_pow(2, 13)); + + disp_dlg_regs->refcyc_per_line_delivery_pre_c = (unsigned int) dml_floor( + refcyc_per_line_delivery_pre_c, 1); + disp_dlg_regs->refcyc_per_line_delivery_c = (unsigned int) dml_floor( + refcyc_per_line_delivery_c, 1); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_c < (unsigned int)dml_pow(2, 13)); + ASSERT(disp_dlg_regs->refcyc_per_line_delivery_c < (unsigned int)dml_pow(2, 13)); + + disp_dlg_regs->chunk_hdl_adjust_cur0 = 3; + disp_dlg_regs->dst_y_offset_cur0 = 0; + disp_dlg_regs->chunk_hdl_adjust_cur1 = 3; + disp_dlg_regs->dst_y_offset_cur1 = 0; + + disp_dlg_regs->xfc_reg_transfer_delay = xfc_transfer_delay; + disp_dlg_regs->xfc_reg_precharge_delay = xfc_precharge_delay; + disp_dlg_regs->xfc_reg_remote_surface_flip_latency = xfc_remote_surface_flip_latency; + disp_dlg_regs->xfc_reg_prefetch_margin = dml_ceil( + xfc_prefetch_margin * refclk_freq_in_mhz, 1); + + // slave has to have this value also set to off + if (src->xfc_enable && !src->xfc_slave) + disp_dlg_regs->dst_y_delta_drq_limit = dml_ceil(xfc_dst_y_delta_drq_limit, 1); + else + disp_dlg_regs->dst_y_delta_drq_limit = 0x7fff; // off + + disp_ttu_regs->refcyc_per_req_delivery_pre_l = (unsigned int) (refcyc_per_req_delivery_pre_l + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_l = (unsigned int) (refcyc_per_req_delivery_l + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_c = (unsigned int) (refcyc_per_req_delivery_pre_c + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_c = (unsigned int) (refcyc_per_req_delivery_c + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_cur0 = + (unsigned int) (refcyc_per_req_delivery_pre_cur0 * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_cur0 = (unsigned int) (refcyc_per_req_delivery_cur0 + * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_pre_cur1 = + (unsigned int) (refcyc_per_req_delivery_pre_cur1 * dml_pow(2, 10)); + disp_ttu_regs->refcyc_per_req_delivery_cur1 = (unsigned int) (refcyc_per_req_delivery_cur1 + * dml_pow(2, 10)); + disp_ttu_regs->qos_level_low_wm = 0; + ASSERT(disp_ttu_regs->qos_level_low_wm < dml_pow(2, 14)); + disp_ttu_regs->qos_level_high_wm = (unsigned int) (4.0 * (double) htotal + * ref_freq_to_pix_freq); + ASSERT(disp_ttu_regs->qos_level_high_wm < dml_pow(2, 14)); + + disp_ttu_regs->qos_level_flip = 14; + disp_ttu_regs->qos_level_fixed_l = 8; + disp_ttu_regs->qos_level_fixed_c = 8; + disp_ttu_regs->qos_level_fixed_cur0 = 8; + disp_ttu_regs->qos_ramp_disable_l = 0; + disp_ttu_regs->qos_ramp_disable_c = 0; + disp_ttu_regs->qos_ramp_disable_cur0 = 0; + + disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz; + ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24)); + + print__ttu_regs_st(mode_lib, *disp_ttu_regs); + print__dlg_regs_st(mode_lib, *disp_dlg_regs); +} + +void dml21_rq_dlg_get_dlg_reg( + struct display_mode_lib *mode_lib, + display_dlg_regs_st *dlg_regs, + display_ttu_regs_st *ttu_regs, + display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + const bool cstate_en, + const bool pstate_en, + const bool vm_en, + const bool ignore_viewport_pos, + const bool immediate_flip_support) +{ + display_rq_params_st rq_param = {0}; + display_dlg_sys_params_st dlg_sys_param = {0}; + + // Get watermark and Tex. + dlg_sys_param.t_urg_wm_us = get_wm_urgent(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.deepsleep_dcfclk_mhz = get_clk_dcf_deepsleep( + mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.t_extra_us = get_urgent_extra_latency(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.mem_trip_us = get_wm_memory_trip(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.t_mclk_wm_us = get_wm_dram_clock_change(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.t_sr_wm_us = get_wm_stutter_enter_exit(mode_lib, e2e_pipe_param, num_pipes); + dlg_sys_param.total_flip_bw = get_total_immediate_flip_bw( + mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.total_flip_bytes = get_total_immediate_flip_bytes( + mode_lib, + e2e_pipe_param, + num_pipes); + dlg_sys_param.t_srx_delay_us = mode_lib->ip.dcfclk_cstate_latency + / dlg_sys_param.deepsleep_dcfclk_mhz; // TODO: Deprecated + + print__dlg_sys_params_st(mode_lib, dlg_sys_param); + + // system parameter calculation done + + dml_print("DML_DLG: Calculation for pipe[%d] start\n\n", pipe_idx); + dml_rq_dlg_get_rq_params(mode_lib, &rq_param, e2e_pipe_param[pipe_idx].pipe); + dml_rq_dlg_get_dlg_params( + mode_lib, + e2e_pipe_param, + num_pipes, + pipe_idx, + dlg_regs, + ttu_regs, + rq_param.dlg, + dlg_sys_param, + cstate_en, + pstate_en); + dml_print("DML_DLG: Calculation for pipe[%d] end\n", pipe_idx); +} + +void dml_rq_dlg_get_arb_params(struct display_mode_lib *mode_lib, display_arb_params_st *arb_param) +{ + memset(arb_param, 0, sizeof(*arb_param)); + arb_param->max_req_outstanding = 256; + arb_param->min_req_outstanding = 68; + arb_param->sat_level_us = 60; +} + +static void calculate_ttu_cursor( + struct display_mode_lib *mode_lib, + double *refcyc_per_req_delivery_pre_cur, + double *refcyc_per_req_delivery_cur, + double refclk_freq_in_mhz, + double ref_freq_to_pix_freq, + double hscale_pixel_rate_l, + double hscl_ratio, + double vratio_pre_l, + double vratio_l, + unsigned int cur_width, + enum cursor_bpp cur_bpp) +{ + unsigned int cur_src_width = cur_width; + unsigned int cur_req_size = 0; + unsigned int cur_req_width = 0; + double cur_width_ub = 0.0; + double cur_req_per_width = 0.0; + double hactive_cur = 0.0; + + ASSERT(cur_src_width <= 256); + + *refcyc_per_req_delivery_pre_cur = 0.0; + *refcyc_per_req_delivery_cur = 0.0; + if (cur_src_width > 0) { + unsigned int cur_bit_per_pixel = 0; + + if (cur_bpp == dm_cur_2bit) { + cur_req_size = 64; // byte + cur_bit_per_pixel = 2; + } else { // 32bit + cur_bit_per_pixel = 32; + if (cur_src_width >= 1 && cur_src_width <= 16) + cur_req_size = 64; + else if (cur_src_width >= 17 && cur_src_width <= 31) + cur_req_size = 128; + else + cur_req_size = 256; + } + + cur_req_width = (double) cur_req_size / ((double) cur_bit_per_pixel / 8.0); + cur_width_ub = dml_ceil((double) cur_src_width / (double) cur_req_width, 1) + * (double) cur_req_width; + cur_req_per_width = cur_width_ub / (double) cur_req_width; + hactive_cur = (double) cur_src_width / hscl_ratio; // FIXME: oswin to think about what to do for cursor + + if (vratio_pre_l <= 1.0) { + *refcyc_per_req_delivery_pre_cur = hactive_cur * ref_freq_to_pix_freq + / (double) cur_req_per_width; + } else { + *refcyc_per_req_delivery_pre_cur = (double) refclk_freq_in_mhz + * (double) cur_src_width / hscale_pixel_rate_l + / (double) cur_req_per_width; + } + + ASSERT(*refcyc_per_req_delivery_pre_cur < dml_pow(2, 13)); + + if (vratio_l <= 1.0) { + *refcyc_per_req_delivery_cur = hactive_cur * ref_freq_to_pix_freq + / (double) cur_req_per_width; + } else { + *refcyc_per_req_delivery_cur = (double) refclk_freq_in_mhz + * (double) cur_src_width / hscale_pixel_rate_l + / (double) cur_req_per_width; + } + + dml_print( + "DML_DLG: %s: cur_req_width = %d\n", + __func__, + cur_req_width); + dml_print( + "DML_DLG: %s: cur_width_ub = %3.2f\n", + __func__, + cur_width_ub); + dml_print( + "DML_DLG: %s: cur_req_per_width = %3.2f\n", + __func__, + cur_req_per_width); + dml_print( + "DML_DLG: %s: hactive_cur = %3.2f\n", + __func__, + hactive_cur); + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_pre_cur = %3.2f\n", + __func__, + *refcyc_per_req_delivery_pre_cur); + dml_print( + "DML_DLG: %s: refcyc_per_req_delivery_cur = %3.2f\n", + __func__, + *refcyc_per_req_delivery_cur); + + ASSERT(*refcyc_per_req_delivery_cur < dml_pow(2, 13)); + } +} + diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h new file mode 100644 index 000000000000..83e95f8cbff2 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h @@ -0,0 +1,73 @@ +/* + * Copyright 2017 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DML21_DISPLAY_RQ_DLG_CALC_H__ +#define __DML21_DISPLAY_RQ_DLG_CALC_H__ + +#include "../dml_common_defs.h" +#include "../display_rq_dlg_helpers.h" + +struct display_mode_lib; + + +// Function: dml_rq_dlg_get_rq_reg +// Main entry point for test to get the register values out of this DML class. +// This function calls <get_rq_param> and <extract_rq_regs> functions to calculate +// and then populate the rq_regs struct +// Input: +// pipe_src_param - pipe source configuration (e.g. vp, pitch, etc.) +// Output: +// rq_regs - struct that holds all the RQ registers field value. +// See also: <display_rq_regs_st> +void dml21_rq_dlg_get_rq_reg( + struct display_mode_lib *mode_lib, + display_rq_regs_st *rq_regs, + const display_pipe_params_st pipe_param); + +// Function: dml_rq_dlg_get_dlg_reg +// Calculate and return DLG and TTU register struct given the system setting +// Output: +// dlg_regs - output DLG register struct +// ttu_regs - output DLG TTU register struct +// Input: +// e2e_pipe_param - "compacted" array of e2e pipe param struct +// num_pipes - num of active "pipe" or "route" +// pipe_idx - index that identifies the e2e_pipe_param that corresponding to this dlg +// cstate - 0: when calculate min_ttu_vblank it is assumed cstate is not required. 1: Normal mode, cstate is considered. +// Added for legacy or unrealistic timing tests. +void dml21_rq_dlg_get_dlg_reg( + struct display_mode_lib *mode_lib, + display_dlg_regs_st *dlg_regs, + display_ttu_regs_st *ttu_regs, + display_e2e_pipe_params_st *e2e_pipe_param, + const unsigned int num_pipes, + const unsigned int pipe_idx, + const bool cstate_en, + const bool pstate_en, + const bool vm_en, + const bool ignore_viewport_pos, + const bool immediate_flip_support); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h index 0c2fab1e93b6..bfc2f39bd1ef 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h @@ -37,11 +37,14 @@ enum source_format_class { dm_444_64 = 2, dm_420_8 = 3, dm_420_10 = 4, - dm_422_8 = 5, - dm_422_10 = 6, - dm_444_8 = 7, + dm_420_12 = 5, + dm_422_8 = 6, + dm_422_10 = 7, + dm_444_8 = 8, dm_mono_8 = dm_444_8, - dm_mono_16 = dm_444_16 + dm_mono_16 = dm_444_16, + dm_rgbe = 9, + dm_rgbe_alpha = 10, }; enum output_bpc_class { dm_out_6 = 0, dm_out_8 = 1, dm_out_10 = 2, dm_out_12 = 3, dm_out_16 = 4 @@ -82,8 +85,8 @@ enum dm_swizzle_mode { dm_sw_var_s_x = 29, dm_sw_var_d_x = 30, dm_sw_64kb_r_x, - dm_sw_gfx7_2d_thin_lvp, - dm_sw_gfx7_2d_thin_gl + dm_sw_gfx7_2d_thin_l_vp, + dm_sw_gfx7_2d_thin_gl, }; enum lb_depth { dm_lb_10 = 0, dm_lb_8 = 1, dm_lb_6 = 2, dm_lb_12 = 3, dm_lb_16 = 4, @@ -112,7 +115,12 @@ enum output_standard { enum mpc_combine_affinity { dm_mpc_always_when_possible, dm_mpc_reduce_voltage, - dm_mpc_reduce_voltage_and_clocks + dm_mpc_reduce_voltage_and_clocks, + dm_mpc_never +}; + +enum RequestType { + REQ_256Bytes, REQ_128BytesNonContiguous, REQ_128BytesContiguous, REQ_NA }; enum self_refresh_affinity { @@ -131,9 +139,7 @@ enum dm_validation_status { DML_FAIL_DIO_SUPPORT, DML_FAIL_NOT_ENOUGH_DSC, DML_FAIL_DSC_CLK_REQUIRED, -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT DML_FAIL_DSC_VALIDATION_FAILURE, -#endif DML_FAIL_URGENT_LATENCY, DML_FAIL_REORDERING_BUFFER, DML_FAIL_DISPCLK_DPPCLK, @@ -157,4 +163,22 @@ enum writeback_config { dm_whole_buffer_for_single_stream_interleave, }; +enum odm_combine_mode { + dm_odm_combine_mode_disabled, + dm_odm_combine_mode_2to1, + dm_odm_combine_mode_4to1, +}; + +enum odm_combine_policy { + dm_odm_combine_policy_dal, + dm_odm_combine_policy_none, + dm_odm_combine_policy_2to1, + dm_odm_combine_policy_4to1, +}; + +enum immediate_flip_requirement { + dm_immediate_flip_not_required, + dm_immediate_flip_required, +}; + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c index 91810c7d5cf5..2689401a03a3 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c @@ -25,19 +25,33 @@ #include "display_mode_lib.h" #include "dc_features.h" -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dcn20/display_mode_vba_20.h" #include "dcn20/display_rq_dlg_calc_20.h" -#endif +#include "dcn20/display_mode_vba_20v2.h" +#include "dcn20/display_rq_dlg_calc_20v2.h" +#include "dcn21/display_mode_vba_21.h" +#include "dcn21/display_rq_dlg_calc_21.h" -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) const struct dml_funcs dml20_funcs = { .validate = dml20_ModeSupportAndSystemConfigurationFull, .recalculate = dml20_recalculate, .rq_dlg_get_dlg_reg = dml20_rq_dlg_get_dlg_reg, .rq_dlg_get_rq_reg = dml20_rq_dlg_get_rq_reg }; -#endif + +const struct dml_funcs dml20v2_funcs = { + .validate = dml20v2_ModeSupportAndSystemConfigurationFull, + .recalculate = dml20v2_recalculate, + .rq_dlg_get_dlg_reg = dml20v2_rq_dlg_get_dlg_reg, + .rq_dlg_get_rq_reg = dml20v2_rq_dlg_get_rq_reg +}; + +const struct dml_funcs dml21_funcs = { + .validate = dml21_ModeSupportAndSystemConfigurationFull, + .recalculate = dml21_recalculate, + .rq_dlg_get_dlg_reg = dml21_rq_dlg_get_dlg_reg, + .rq_dlg_get_rq_reg = dml21_rq_dlg_get_rq_reg +}; void dml_init_instance(struct display_mode_lib *lib, const struct _vcs_dpi_soc_bounding_box_st *soc_bb, @@ -48,11 +62,16 @@ void dml_init_instance(struct display_mode_lib *lib, lib->ip = *ip_params; lib->project = project; switch (project) { -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 case DML_PROJECT_NAVI10: lib->funcs = dml20_funcs; break; -#endif + case DML_PROJECT_NAVI10v2: + lib->funcs = dml20v2_funcs; + break; + case DML_PROJECT_DCN21: + lib->funcs = dml21_funcs; + break; + default: break; } diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h index 5bf13d67f289..cf2758ca5b02 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h @@ -27,16 +27,14 @@ #include "dml_common_defs.h" -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 #include "display_mode_vba.h" -#endif enum dml_project { DML_PROJECT_UNDEFINED, DML_PROJECT_RAVEN1, -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 DML_PROJECT_NAVI10, -#endif + DML_PROJECT_NAVI10v2, + DML_PROJECT_DCN21, }; struct display_mode_lib; @@ -66,9 +64,7 @@ struct display_mode_lib { struct _vcs_dpi_ip_params_st ip; struct _vcs_dpi_soc_bounding_box_st soc; enum dml_project project; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 struct vba_vars_st vba; -#endif struct dal_logger *logger; struct dml_funcs funcs; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h index 5678472546ab..658f81e757e9 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h @@ -63,6 +63,7 @@ struct _vcs_dpi_voltage_scaling_st { double dispclk_mhz; double phyclk_mhz; double dppclk_mhz; + double dtbclk_mhz; }; struct _vcs_dpi_soc_bounding_box_st { @@ -99,7 +100,9 @@ struct _vcs_dpi_soc_bounding_box_st { unsigned int num_chans; unsigned int vmm_page_size_bytes; unsigned int hostvm_min_page_size_bytes; + unsigned int gpuvm_min_page_size_bytes; double dram_clock_change_latency_us; + double dummy_pstate_latency_us; double writeback_dram_clock_change_latency_us; unsigned int return_bus_width_bytes; unsigned int voltage_override; @@ -108,6 +111,10 @@ struct _vcs_dpi_soc_bounding_box_st { int use_urgent_burst_bw; unsigned int num_states; struct _vcs_dpi_voltage_scaling_st clock_limits[MAX_CLOCK_LIMIT_STATES]; + bool do_urgent_latency_adjustment; + double urgent_latency_adjustment_fabric_clock_component_us; + double urgent_latency_adjustment_fabric_clock_reference_mhz; + bool disable_dram_clock_change_vactive_support; }; struct _vcs_dpi_ip_params_st { @@ -141,7 +148,6 @@ struct _vcs_dpi_ip_params_st { unsigned int writeback_interface_buffer_size_kbytes; unsigned int writeback_line_buffer_buffer_size; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 unsigned int writeback_10bpc420_supported; double writeback_max_hscl_ratio; double writeback_max_vscl_ratio; @@ -151,7 +157,6 @@ struct _vcs_dpi_ip_params_st { unsigned int writeback_max_vscl_taps; unsigned int writeback_line_buffer_luma_buffer_size; unsigned int writeback_line_buffer_chroma_buffer_size; -#endif unsigned int max_page_table_levels; unsigned int max_num_dpp; @@ -210,6 +215,7 @@ struct _vcs_dpi_display_pipe_source_params_st { int source_format; unsigned char dcc; unsigned int dcc_rate; + unsigned int dcc_rate_chroma; unsigned char dcc_use_global; unsigned char vm; bool gpuvm; // gpuvm enabled @@ -221,6 +227,10 @@ struct _vcs_dpi_display_pipe_source_params_st { int source_scan; int sw_mode; int macro_tile_size; + unsigned int surface_width_y; + unsigned int surface_height_y; + unsigned int surface_width_c; + unsigned int surface_height_c; unsigned int viewport_width; unsigned int viewport_height; unsigned int viewport_y_y; @@ -265,7 +275,7 @@ struct writeback_st { struct _vcs_dpi_display_output_params_st { int dp_lanes; - int output_bpp; + double output_bpp; int dsc_enable; int wb_enable; int num_active_wb; @@ -273,6 +283,7 @@ struct _vcs_dpi_display_output_params_st { int output_type; int output_format; int dsc_slices; + int max_audio_sample_rate; struct writeback_st wb; }; @@ -314,10 +325,11 @@ struct _vcs_dpi_display_pipe_dest_params_st { unsigned int vupdate_width; unsigned int vready_offset; unsigned char interlaced; + unsigned char embedded; double pixel_rate_mhz; unsigned char synchronized_vblank_all_planes; unsigned char otg_inst; - unsigned char odm_combine; + unsigned int odm_combine; unsigned char use_maximum_vstartup; unsigned int vtotal_max; unsigned int vtotal_min; @@ -396,6 +408,7 @@ struct _vcs_dpi_display_rq_misc_params_st { struct _vcs_dpi_display_rq_params_st { unsigned char yuv420; unsigned char yuv420_10bpc; + unsigned char rgbe_alpha; display_rq_misc_params_st misc; display_rq_sizing_params_st sizing; display_rq_dlg_params_st dlg; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c index 4d2a1262d9db..b3c96d9b472f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c @@ -23,7 +23,6 @@ * */ -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 #include "display_mode_lib.h" #include "display_mode_vba.h" @@ -222,13 +221,17 @@ static void fetch_socbb_params(struct display_mode_lib *mode_lib) mode_lib->vba.SRExitTime = soc->sr_exit_time_us; mode_lib->vba.SREnterPlusExitTime = soc->sr_enter_plus_exit_time_us; mode_lib->vba.DRAMClockChangeLatency = soc->dram_clock_change_latency_us; + mode_lib->vba.DummyPStateCheck = soc->dram_clock_change_latency_us == soc->dummy_pstate_latency_us; + mode_lib->vba.DRAMClockChangeSupportsVActive = !soc->disable_dram_clock_change_vactive_support || + mode_lib->vba.DummyPStateCheck; + mode_lib->vba.Downspreading = soc->downspread_percent; mode_lib->vba.DRAMChannelWidth = soc->dram_channel_width_bytes; // new! mode_lib->vba.FabricDatapathToDCNDataReturn = soc->fabric_datapath_to_dcn_data_return_bytes; // new! mode_lib->vba.DISPCLKDPPCLKDSCCLKDownSpreading = soc->dcn_downspread_percent; // new mode_lib->vba.DISPCLKDPPCLKVCOSpeed = soc->dispclk_dppclk_vco_speed_mhz; // new mode_lib->vba.VMMPageSize = soc->vmm_page_size_bytes; - mode_lib->vba.GPUVMMinPageSize = soc->vmm_page_size_bytes / 1024; + mode_lib->vba.GPUVMMinPageSize = soc->gpuvm_min_page_size_bytes / 1024; mode_lib->vba.HostVMMinPageSize = soc->hostvm_min_page_size_bytes / 1024; // Set the voltage scaling clocks as the defaults. Most of these will // be set to different values by the test @@ -261,7 +264,17 @@ static void fetch_socbb_params(struct display_mode_lib *mode_lib) mode_lib->vba.DRAMSpeedPerState[i] = soc->clock_limits[i].dram_speed_mts; //mode_lib->vba.DRAMSpeedPerState[i] = soc->clock_limits[i].dram_speed_mhz; mode_lib->vba.MaxDispclk[i] = soc->clock_limits[i].dispclk_mhz; + mode_lib->vba.DTBCLKPerState[i] = soc->clock_limits[i].dtbclk_mhz; } + mode_lib->vba.MinVoltageLevel = 0; + mode_lib->vba.MaxVoltageLevel = mode_lib->vba.soc.num_states; + + mode_lib->vba.DoUrgentLatencyAdjustment = + soc->do_urgent_latency_adjustment; + mode_lib->vba.UrgentLatencyAdjustmentFabricClockComponent = + soc->urgent_latency_adjustment_fabric_clock_component_us; + mode_lib->vba.UrgentLatencyAdjustmentFabricClockReference = + soc->urgent_latency_adjustment_fabric_clock_reference_mhz; } static void fetch_ip_params(struct display_mode_lib *mode_lib) @@ -296,8 +309,6 @@ static void fetch_ip_params(struct display_mode_lib *mode_lib) mode_lib->vba.WritebackInterfaceBufferSize = ip->writeback_interface_buffer_size_kbytes; mode_lib->vba.WritebackLineBufferSize = ip->writeback_line_buffer_buffer_size; - mode_lib->vba.MinVoltageLevel = 0; - mode_lib->vba.MaxVoltageLevel = 5; mode_lib->vba.WritebackChromaLineBufferWidth = ip->writeback_chroma_line_buffer_width_pixels; @@ -368,6 +379,7 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) mode_lib->vba.pipe_plane[j] = mode_lib->vba.NumberOfActivePlanes; + mode_lib->vba.EmbeddedPanel[mode_lib->vba.NumberOfActivePlanes] = dst->embedded; mode_lib->vba.DPPPerPlane[mode_lib->vba.NumberOfActivePlanes] = 1; mode_lib->vba.SourceScan[mode_lib->vba.NumberOfActivePlanes] = (enum scan_direction_class) (src->source_scan); @@ -385,8 +397,10 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) src->viewport_y_c; mode_lib->vba.PitchY[mode_lib->vba.NumberOfActivePlanes] = src->data_pitch; mode_lib->vba.SurfaceHeightY[mode_lib->vba.NumberOfActivePlanes] = src->viewport_height; + mode_lib->vba.SurfaceWidthY[mode_lib->vba.NumberOfActivePlanes] = src->viewport_width; mode_lib->vba.PitchC[mode_lib->vba.NumberOfActivePlanes] = src->data_pitch_c; mode_lib->vba.SurfaceHeightC[mode_lib->vba.NumberOfActivePlanes] = src->viewport_height_c; + mode_lib->vba.SurfaceWidthC[mode_lib->vba.NumberOfActivePlanes] = src->viewport_width_c; mode_lib->vba.DCCMetaPitchY[mode_lib->vba.NumberOfActivePlanes] = src->meta_pitch; mode_lib->vba.DCCMetaPitchC[mode_lib->vba.NumberOfActivePlanes] = src->meta_pitch_c; mode_lib->vba.HRatio[mode_lib->vba.NumberOfActivePlanes] = scl->hscl_ratio; @@ -410,8 +424,8 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) ip->dcc_supported : src->dcc && ip->dcc_supported; mode_lib->vba.DCCRate[mode_lib->vba.NumberOfActivePlanes] = src->dcc_rate; /* TODO: Needs to be set based on src->dcc_rate_luma/chroma */ - mode_lib->vba.DCCRateLuma[mode_lib->vba.NumberOfActivePlanes] = 0; - mode_lib->vba.DCCRateChroma[mode_lib->vba.NumberOfActivePlanes] = 0; + mode_lib->vba.DCCRateLuma[mode_lib->vba.NumberOfActivePlanes] = src->dcc_rate; + mode_lib->vba.DCCRateChroma[mode_lib->vba.NumberOfActivePlanes] = src->dcc_rate_chroma; mode_lib->vba.SourcePixelFormat[mode_lib->vba.NumberOfActivePlanes] = (enum source_format_class) (src->source_format); @@ -425,6 +439,8 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) dst->odm_combine; mode_lib->vba.OutputFormat[mode_lib->vba.NumberOfActivePlanes] = (enum output_format_class) (dout->output_format); + mode_lib->vba.OutputBpp[mode_lib->vba.NumberOfActivePlanes] = + dout->output_bpp; mode_lib->vba.Output[mode_lib->vba.NumberOfActivePlanes] = (enum output_encoder_class) (dout->output_type); @@ -437,7 +453,7 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) dout->dp_lanes; /* TODO: Needs to be set based on dout->audio.audio_sample_rate_khz/sample_layout */ mode_lib->vba.AudioSampleRate[mode_lib->vba.NumberOfActivePlanes] = - 44.1 * 1000; + dout->max_audio_sample_rate; mode_lib->vba.AudioSampleLayout[mode_lib->vba.NumberOfActivePlanes] = 1; mode_lib->vba.DRAMClockChangeLatencyOverride = 0.0; @@ -457,6 +473,10 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) dout->wb.wb_dst_width; mode_lib->vba.WritebackDestinationHeight[mode_lib->vba.NumberOfActivePlanes] = dout->wb.wb_dst_height; + mode_lib->vba.WritebackHRatio[mode_lib->vba.NumberOfActivePlanes] = + dout->wb.wb_hratio; + mode_lib->vba.WritebackVRatio[mode_lib->vba.NumberOfActivePlanes] = + dout->wb.wb_vratio; mode_lib->vba.WritebackPixelFormat[mode_lib->vba.NumberOfActivePlanes] = (enum source_format_class) (dout->wb.wb_pixel_format); mode_lib->vba.WritebackHTaps[mode_lib->vba.NumberOfActivePlanes] = @@ -568,6 +588,8 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) if (src->is_hsplit) { for (k = j + 1; k < mode_lib->vba.cache_num_pipes; ++k) { display_pipe_source_params_st *src_k = &pipes[k].pipe.src; + display_pipe_dest_params_st *dst_k = &pipes[k].pipe.dest; + display_output_params_st *dout_k = &pipes[j].dout; if (src_k->is_hsplit && !visited[k] && src->hsplit_grp == src_k->hsplit_grp) { @@ -575,12 +597,21 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) mode_lib->vba.NumberOfActivePlanes; mode_lib->vba.DPPPerPlane[mode_lib->vba.NumberOfActivePlanes]++; if (mode_lib->vba.SourceScan[mode_lib->vba.NumberOfActivePlanes] - == dm_horz) + == dm_horz) { mode_lib->vba.ViewportWidth[mode_lib->vba.NumberOfActivePlanes] += src_k->viewport_width; - else + mode_lib->vba.ViewportWidthChroma[mode_lib->vba.NumberOfActivePlanes] += + src_k->viewport_width; + mode_lib->vba.ScalerRecoutWidth[mode_lib->vba.NumberOfActivePlanes] += + dst_k->recout_width; + } else { mode_lib->vba.ViewportHeight[mode_lib->vba.NumberOfActivePlanes] += src_k->viewport_height; + mode_lib->vba.ViewportHeightChroma[mode_lib->vba.NumberOfActivePlanes] += + src_k->viewport_height; + } + mode_lib->vba.NumberOfDSCSlices[mode_lib->vba.NumberOfActivePlanes] += + dout_k->dsc_slices; visited[k] = true; } @@ -786,7 +817,9 @@ void ModeSupportAndSystemConfiguration(struct display_mode_lib *mode_lib) unsigned int total_pipes = 0; mode_lib->vba.VoltageLevel = mode_lib->vba.cache_pipes[0].clks_cfg.voltage; - mode_lib->vba.ReturnBW = mode_lib->vba.ReturnBWPerState[mode_lib->vba.VoltageLevel]; + mode_lib->vba.ReturnBW = mode_lib->vba.ReturnBWPerState[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb]; + if (mode_lib->vba.ReturnBW == 0) + mode_lib->vba.ReturnBW = mode_lib->vba.ReturnBWPerState[mode_lib->vba.VoltageLevel][0]; mode_lib->vba.FabricAndDRAMBandwidth = mode_lib->vba.FabricAndDRAMBandwidthPerState[mode_lib->vba.VoltageLevel]; fetch_socbb_params(mode_lib); @@ -836,4 +869,3 @@ double CalculateWriteBackDISPCLK( return CalculateWriteBackDISPCLK; } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h index 0347f74cda3a..2875efd85467 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h @@ -23,7 +23,6 @@ * */ -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 #ifndef __DML2_DISPLAY_MODE_VBA_H__ #define __DML2_DISPLAY_MODE_VBA_H__ @@ -155,7 +154,10 @@ struct vba_vars_st { double UrgentLatencySupportUsChroma; unsigned int DSCFormatFactor; + bool DummyPStateCheck; + bool DRAMClockChangeSupportsVActive; bool PrefetchModeSupported; + bool PrefetchAndImmediateFlipSupported; enum self_refresh_affinity AllowDRAMSelfRefreshOrDRAMClockChangeInVblank; // Mode Support only double XFCRemoteSurfaceFlipDelay; double TInitXFill; @@ -290,6 +292,7 @@ struct vba_vars_st { double PixelClock[DC__NUM_DPP__MAX]; double PixelClockBackEnd[DC__NUM_DPP__MAX]; bool DCCEnable[DC__NUM_DPP__MAX]; + bool FECEnable[DC__NUM_DPP__MAX]; unsigned int DCCMetaPitchY[DC__NUM_DPP__MAX]; unsigned int DCCMetaPitchC[DC__NUM_DPP__MAX]; enum scan_direction_class SourceScan[DC__NUM_DPP__MAX]; @@ -316,7 +319,7 @@ struct vba_vars_st { unsigned int DynamicMetadataTransmittedBytes[DC__NUM_DPP__MAX]; double DCCRate[DC__NUM_DPP__MAX]; double AverageDCCCompressionRate; - bool ODMCombineEnabled[DC__NUM_DPP__MAX]; + enum odm_combine_mode ODMCombineEnabled[DC__NUM_DPP__MAX]; double OutputBpp[DC__NUM_DPP__MAX]; bool DSCEnabled[DC__NUM_DPP__MAX]; unsigned int DSCInputBitPerComponent[DC__NUM_DPP__MAX]; @@ -344,6 +347,7 @@ struct vba_vars_st { unsigned int EffectiveLBLatencyHidingSourceLinesChroma; double BandwidthAvailableForImmediateFlip; unsigned int PrefetchMode[DC__VOLTAGE_STATES + 1][2]; + unsigned int PrefetchModePerState[DC__VOLTAGE_STATES + 1][2]; unsigned int MinPrefetchMode; unsigned int MaxPrefetchMode; bool AnyLinesForVMOrRowTooLarge; @@ -385,6 +389,7 @@ struct vba_vars_st { /* vba mode support */ /*inputs*/ + bool EmbeddedPanel[DC__NUM_DPP__MAX]; bool SupportGFX7CompatibleTilingIn32bppAnd64bpp; double MaxHSCLRatio; double MaxVSCLRatio; @@ -392,9 +397,11 @@ struct vba_vars_st { bool WritebackLumaAndChromaScalingSupported; bool Cursor64BppSupport; double DCFCLKPerState[DC__VOLTAGE_STATES + 1]; + double DCFCLKState[DC__VOLTAGE_STATES + 1][2]; double FabricClockPerState[DC__VOLTAGE_STATES + 1]; double SOCCLKPerState[DC__VOLTAGE_STATES + 1]; double PHYCLKPerState[DC__VOLTAGE_STATES + 1]; + double DTBCLKPerState[DC__VOLTAGE_STATES + 1]; double MaxDppclk[DC__VOLTAGE_STATES + 1]; double MaxDSCCLK[DC__VOLTAGE_STATES + 1]; double DRAMSpeedPerState[DC__VOLTAGE_STATES + 1]; @@ -439,7 +446,7 @@ struct vba_vars_st { double OutputLinkDPLanes[DC__NUM_DPP__MAX]; double ForcedOutputLinkBPP[DC__NUM_DPP__MAX]; // Mode Support only double ImmediateFlipBW[DC__NUM_DPP__MAX]; - double MaxMaxVStartup; + double MaxMaxVStartup[DC__VOLTAGE_STATES + 1][2]; double WritebackLumaVExtra; double WritebackChromaVExtra; @@ -466,7 +473,7 @@ struct vba_vars_st { double RoundedUpMaxSwathSizeBytesC; double EffectiveDETLBLinesLuma; double EffectiveDETLBLinesChroma; - double ProjectedDCFCLKDeepSleep; + double ProjectedDCFCLKDeepSleep[DC__VOLTAGE_STATES + 1][2]; double PDEAndMetaPTEBytesPerFrameY; double PDEAndMetaPTEBytesPerFrameC; unsigned int MetaRowBytesY; @@ -484,11 +491,11 @@ struct vba_vars_st { double FractionOfUrgentBandwidthImmediateFlip; // Mode Support debugging output /* ms locals */ - double IdealSDPPortBandwidthPerState[DC__VOLTAGE_STATES + 1]; + double IdealSDPPortBandwidthPerState[DC__VOLTAGE_STATES + 1][2]; unsigned int NoOfDPP[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; int NoOfDPPThisState[DC__NUM_DPP__MAX]; - bool ODMCombineEnablePerState[DC__VOLTAGE_STATES + 1][DC__NUM_DPP__MAX]; - unsigned int SwathWidthYThisState[DC__NUM_DPP__MAX]; + enum odm_combine_mode ODMCombineEnablePerState[DC__VOLTAGE_STATES + 1][DC__NUM_DPP__MAX]; + double SwathWidthYThisState[DC__NUM_DPP__MAX]; unsigned int SwathHeightCPerState[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; unsigned int SwathHeightYThisState[DC__NUM_DPP__MAX]; unsigned int SwathHeightCThisState[DC__NUM_DPP__MAX]; @@ -500,7 +507,7 @@ struct vba_vars_st { double RequiredDPPCLKThisState[DC__NUM_DPP__MAX]; bool PTEBufferSizeNotExceededY[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; bool PTEBufferSizeNotExceededC[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; - bool BandwidthWithoutPrefetchSupported[DC__VOLTAGE_STATES + 1]; + bool BandwidthWithoutPrefetchSupported[DC__VOLTAGE_STATES + 1][2]; bool PrefetchSupported[DC__VOLTAGE_STATES + 1][2]; bool VRatioInPrefetchSupported[DC__VOLTAGE_STATES + 1][2]; double RequiredDISPCLK[DC__VOLTAGE_STATES + 1][2]; @@ -509,21 +516,22 @@ struct vba_vars_st { unsigned int TotalNumberOfActiveDPP[DC__VOLTAGE_STATES + 1][2]; unsigned int TotalNumberOfDCCActiveDPP[DC__VOLTAGE_STATES + 1][2]; bool ModeSupport[DC__VOLTAGE_STATES + 1][2]; - double ReturnBWPerState[DC__VOLTAGE_STATES + 1]; + double ReturnBWPerState[DC__VOLTAGE_STATES + 1][2]; bool DIOSupport[DC__VOLTAGE_STATES + 1]; bool NotEnoughDSCUnits[DC__VOLTAGE_STATES + 1]; bool DSCCLKRequiredMoreThanSupported[DC__VOLTAGE_STATES + 1]; + bool DTBCLKRequiredMoreThanSupported[DC__VOLTAGE_STATES + 1]; double UrgentRoundTripAndOutOfOrderLatencyPerState[DC__VOLTAGE_STATES + 1]; - bool ROBSupport[DC__VOLTAGE_STATES + 1]; + bool ROBSupport[DC__VOLTAGE_STATES + 1][2]; bool PTEBufferSizeNotExceeded[DC__VOLTAGE_STATES + 1][2]; - bool TotalVerticalActiveBandwidthSupport[DC__VOLTAGE_STATES + 1]; - double MaxTotalVerticalActiveAvailableBandwidth[DC__VOLTAGE_STATES + 1]; + bool TotalVerticalActiveBandwidthSupport[DC__VOLTAGE_STATES + 1][2]; + double MaxTotalVerticalActiveAvailableBandwidth[DC__VOLTAGE_STATES + 1][2]; double PrefetchBW[DC__NUM_DPP__MAX]; - double PDEAndMetaPTEBytesPerFrame[DC__NUM_DPP__MAX]; - double MetaRowBytes[DC__NUM_DPP__MAX]; - double DPTEBytesPerRow[DC__NUM_DPP__MAX]; - double PrefetchLinesY[DC__NUM_DPP__MAX]; - double PrefetchLinesC[DC__NUM_DPP__MAX]; + double PDEAndMetaPTEBytesPerFrame[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double MetaRowBytes[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double DPTEBytesPerRow[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double PrefetchLinesY[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double PrefetchLinesC[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; unsigned int MaxNumSwY[DC__NUM_DPP__MAX]; unsigned int MaxNumSwC[DC__NUM_DPP__MAX]; double PrefillY[DC__NUM_DPP__MAX]; @@ -532,7 +540,7 @@ struct vba_vars_st { double LinesForMetaPTE[DC__NUM_DPP__MAX]; double LinesForMetaAndDPTERow[DC__NUM_DPP__MAX]; double MinDPPCLKUsingSingleDPP[DC__NUM_DPP__MAX]; - unsigned int SwathWidthYSingleDPP[DC__NUM_DPP__MAX]; + double SwathWidthYSingleDPP[DC__NUM_DPP__MAX]; double BytePerPixelInDETY[DC__NUM_DPP__MAX]; double BytePerPixelInDETC[DC__NUM_DPP__MAX]; bool RequiresDSC[DC__VOLTAGE_STATES + 1][DC__NUM_DPP__MAX]; @@ -540,7 +548,7 @@ struct vba_vars_st { double RequiresFEC[DC__VOLTAGE_STATES + 1][DC__NUM_DPP__MAX]; double OutputBppPerState[DC__VOLTAGE_STATES + 1][DC__NUM_DPP__MAX]; double DSCDelayPerState[DC__VOLTAGE_STATES + 1][DC__NUM_DPP__MAX]; - bool ViewportSizeSupport[DC__VOLTAGE_STATES + 1]; + bool ViewportSizeSupport[DC__VOLTAGE_STATES + 1][2]; unsigned int Read256BlockHeightY[DC__NUM_DPP__MAX]; unsigned int Read256BlockWidthY[DC__NUM_DPP__MAX]; unsigned int Read256BlockHeightC[DC__NUM_DPP__MAX]; @@ -555,7 +563,7 @@ struct vba_vars_st { double WriteBandwidth[DC__NUM_DPP__MAX]; double PSCL_FACTOR[DC__NUM_DPP__MAX]; double PSCL_FACTOR_CHROMA[DC__NUM_DPP__MAX]; - double MaximumVStartup[DC__NUM_DPP__MAX]; + double MaximumVStartup[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; unsigned int MacroTileWidthY[DC__NUM_DPP__MAX]; unsigned int MacroTileWidthC[DC__NUM_DPP__MAX]; double AlignedDCCMetaPitch[DC__NUM_DPP__MAX]; @@ -572,7 +580,7 @@ struct vba_vars_st { bool ImmediateFlipSupportedForState[DC__VOLTAGE_STATES + 1][2]; double WritebackDelay[DC__VOLTAGE_STATES + 1][DC__NUM_DPP__MAX]; unsigned int vm_group_bytes[DC__NUM_DPP__MAX]; - long dpte_group_bytes[DC__NUM_DPP__MAX]; + unsigned int dpte_group_bytes[DC__NUM_DPP__MAX]; unsigned int dpte_row_height[DC__NUM_DPP__MAX]; unsigned int meta_req_height[DC__NUM_DPP__MAX]; unsigned int meta_req_width[DC__NUM_DPP__MAX]; @@ -598,13 +606,14 @@ struct vba_vars_st { double UrgentBurstFactorChroma[DC__NUM_DPP__MAX]; double UrgentBurstFactorChromaPre[DC__NUM_DPP__MAX]; + bool MPCCombine[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; double SwathWidthCSingleDPP[DC__NUM_DPP__MAX]; double MaximumSwathWidthInLineBufferLuma; double MaximumSwathWidthInLineBufferChroma; double MaximumSwathWidthLuma[DC__NUM_DPP__MAX]; double MaximumSwathWidthChroma[DC__NUM_DPP__MAX]; - bool odm_combine_dummy[DC__NUM_DPP__MAX]; + enum odm_combine_mode odm_combine_dummy[DC__NUM_DPP__MAX]; double dummy1[DC__NUM_DPP__MAX]; double dummy2[DC__NUM_DPP__MAX]; double dummy3[DC__NUM_DPP__MAX]; @@ -614,9 +623,9 @@ struct vba_vars_st { double dummy7[DC__NUM_DPP__MAX]; double dummy8[DC__NUM_DPP__MAX]; unsigned int dummyinteger1ms[DC__NUM_DPP__MAX]; - unsigned int dummyinteger2ms[DC__NUM_DPP__MAX]; + double dummyinteger2ms[DC__NUM_DPP__MAX]; unsigned int dummyinteger3[DC__NUM_DPP__MAX]; - unsigned int dummyinteger4; + unsigned int dummyinteger4[DC__NUM_DPP__MAX]; unsigned int dummyinteger5; unsigned int dummyinteger6; unsigned int dummyinteger7; @@ -625,6 +634,10 @@ struct vba_vars_st { unsigned int dummyinteger10; unsigned int dummyinteger11; unsigned int dummyinteger12; + unsigned int dummyintegerarr1[DC__NUM_DPP__MAX]; + unsigned int dummyintegerarr2[DC__NUM_DPP__MAX]; + unsigned int dummyintegerarr3[DC__NUM_DPP__MAX]; + unsigned int dummyintegerarr4[DC__NUM_DPP__MAX]; bool dummysinglestring; bool SingleDPPViewportSizeSupportPerPlane[DC__NUM_DPP__MAX]; double PlaneRequiredDISPCLKWithODMCombine2To1; @@ -632,18 +645,19 @@ struct vba_vars_st { unsigned int TotalNumberOfSingleDPPPlanes[DC__VOLTAGE_STATES + 1][2]; bool LinkDSCEnable; bool ODMCombine4To1SupportCheckOK[DC__VOLTAGE_STATES + 1]; - bool ODMCombineEnableThisState[DC__NUM_DPP__MAX]; - unsigned int SwathWidthCThisState[DC__NUM_DPP__MAX]; + enum odm_combine_mode ODMCombineEnableThisState[DC__NUM_DPP__MAX]; + double SwathWidthCThisState[DC__NUM_DPP__MAX]; bool ViewportSizeSupportPerPlane[DC__NUM_DPP__MAX]; double AlignedDCCMetaPitchY[DC__NUM_DPP__MAX]; double AlignedDCCMetaPitchC[DC__NUM_DPP__MAX]; unsigned int NotEnoughUrgentLatencyHiding; unsigned int NotEnoughUrgentLatencyHidingPre; - long PTEBufferSizeInRequestsForLuma; + int PTEBufferSizeInRequestsForLuma; + int PTEBufferSizeInRequestsForChroma; // Missing from VBA - long dpte_group_bytes_chroma; + int dpte_group_bytes_chroma; unsigned int vm_group_bytes_chroma; double dst_x_after_scaler; double dst_y_after_scaler; @@ -668,8 +682,8 @@ struct vba_vars_st { double MinTTUVBlank[DC__NUM_DPP__MAX]; double BytePerPixelDETY[DC__NUM_DPP__MAX]; double BytePerPixelDETC[DC__NUM_DPP__MAX]; - unsigned int SwathWidthY[DC__NUM_DPP__MAX]; - unsigned int SwathWidthSingleDPPY[DC__NUM_DPP__MAX]; + double SwathWidthY[DC__NUM_DPP__MAX]; + double SwathWidthSingleDPPY[DC__NUM_DPP__MAX]; double CursorRequestDeliveryTime[DC__NUM_DPP__MAX]; double CursorRequestDeliveryTimePrefetch[DC__NUM_DPP__MAX]; double ReadBandwidthPlaneLuma[DC__NUM_DPP__MAX]; @@ -745,12 +759,12 @@ struct vba_vars_st { double LinesInDETY[DC__NUM_DPP__MAX]; double LinesInDETYRoundedDownToSwath[DC__NUM_DPP__MAX]; - unsigned int SwathWidthSingleDPPC[DC__NUM_DPP__MAX]; - unsigned int SwathWidthC[DC__NUM_DPP__MAX]; + double SwathWidthSingleDPPC[DC__NUM_DPP__MAX]; + double SwathWidthC[DC__NUM_DPP__MAX]; unsigned int BytePerPixelY[DC__NUM_DPP__MAX]; unsigned int BytePerPixelC[DC__NUM_DPP__MAX]; - long dummyinteger1; - long dummyinteger2; + unsigned int dummyinteger1; + unsigned int dummyinteger2; double FinalDRAMClockChangeLatency; double Tdmdl_vm[DC__NUM_DPP__MAX]; double Tdmdl[DC__NUM_DPP__MAX]; @@ -764,6 +778,7 @@ struct vba_vars_st { unsigned int DCCCMaxCompressedBlock[DC__NUM_DPP__MAX]; unsigned int DCCCIndependent64ByteBlock[DC__NUM_DPP__MAX]; double VStartupMargin; + bool NotEnoughTimeForDynamicMetadata; /* Missing from VBA */ unsigned int MaximumMaxVStartupLines; @@ -787,6 +802,9 @@ struct vba_vars_st { unsigned int PDEProcessingBufIn64KBReqs; double MaxTotalVActiveRDBandwidth; + bool DoUrgentLatencyAdjustment; + double UrgentLatencyAdjustmentFabricClockComponent; + double UrgentLatencyAdjustmentFabricClockReference; double MinUrgentLatencySupportUs; double MinFullDETBufferingTime; double AverageReadBandwidthGBytePerSecond; @@ -796,11 +814,13 @@ struct vba_vars_st { unsigned int ViewportHeightChroma[DC__NUM_DPP__MAX]; double HRatioChroma[DC__NUM_DPP__MAX]; double VRatioChroma[DC__NUM_DPP__MAX]; - long WritebackSourceWidth[DC__NUM_DPP__MAX]; + int WritebackSourceWidth[DC__NUM_DPP__MAX]; bool ModeIsSupported; bool ODMCombine4To1Supported; + unsigned int SurfaceWidthY[DC__NUM_DPP__MAX]; + unsigned int SurfaceWidthC[DC__NUM_DPP__MAX]; unsigned int SurfaceHeightY[DC__NUM_DPP__MAX]; unsigned int SurfaceHeightC[DC__NUM_DPP__MAX]; unsigned int WritebackHTaps[DC__NUM_DPP__MAX]; @@ -830,6 +850,58 @@ struct vba_vars_st { unsigned int MaxNumHDMIFRLOutputs; int AudioSampleRate[DC__NUM_DPP__MAX]; int AudioSampleLayout[DC__NUM_DPP__MAX]; + + int PercentMarginOverMinimumRequiredDCFCLK; + bool DynamicMetadataSupported[DC__VOLTAGE_STATES + 1][2]; + enum immediate_flip_requirement ImmediateFlipRequirement; + double DETBufferSizeYThisState[DC__NUM_DPP__MAX]; + double DETBufferSizeCThisState[DC__NUM_DPP__MAX]; + bool NoUrgentLatencyHiding[DC__NUM_DPP__MAX]; + bool NoUrgentLatencyHidingPre[DC__NUM_DPP__MAX]; + int swath_width_luma_ub_this_state[DC__NUM_DPP__MAX]; + int swath_width_chroma_ub_this_state[DC__NUM_DPP__MAX]; + double UrgLatency[DC__VOLTAGE_STATES + 1]; + double VActiveCursorBandwidth[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double VActivePixelBandwidth[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + bool NoTimeForPrefetch[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + bool NoTimeForDynamicMetadata[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double dpte_row_bandwidth[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double meta_row_bandwidth[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double DETBufferSizeYAllStates[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double DETBufferSizeCAllStates[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + int swath_width_luma_ub_all_states[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + int swath_width_chroma_ub_all_states[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + bool NotUrgentLatencyHiding[DC__VOLTAGE_STATES + 1][2]; + unsigned int SwathHeightYAllStates[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + unsigned int SwathHeightCAllStates[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + unsigned int SwathWidthYAllStates[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + unsigned int SwathWidthCAllStates[DC__VOLTAGE_STATES + 1][2][DC__NUM_DPP__MAX]; + double TotalDPTERowBandwidth[DC__VOLTAGE_STATES + 1][2]; + double TotalMetaRowBandwidth[DC__VOLTAGE_STATES + 1][2]; + double TotalVActiveCursorBandwidth[DC__VOLTAGE_STATES + 1][2]; + double TotalVActivePixelBandwidth[DC__VOLTAGE_STATES + 1][2]; + bool UseMinimumRequiredDCFCLK; + double WritebackDelayTime[DC__NUM_DPP__MAX]; + unsigned int DCCYIndependentBlock[DC__NUM_DPP__MAX]; + unsigned int DCCCIndependentBlock[DC__NUM_DPP__MAX]; + unsigned int dummyinteger15; + unsigned int dummyinteger16; + unsigned int dummyinteger17; + unsigned int dummyinteger18; + unsigned int dummyinteger19; + unsigned int dummyinteger20; + unsigned int dummyinteger21; + unsigned int dummyinteger22; + unsigned int dummyinteger23; + unsigned int dummyinteger24; + unsigned int dummyinteger25; + unsigned int dummyinteger26; + unsigned int dummyinteger27; + unsigned int dummyinteger28; + unsigned int dummyinteger29; + bool dummystring[DC__NUM_DPP__MAX]; + double BPP; + enum odm_combine_policy ODMCombinePolicy; }; bool CalculateMinAndMaxPrefetchMode( @@ -851,4 +923,3 @@ double CalculateWriteBackDISPCLK( unsigned int WritebackChromaLineBufferWidth); #endif /* _DML2_DISPLAY_MODE_VBA_H_ */ -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c b/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c index ad8571f5a142..4c3e9cc30167 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c @@ -243,7 +243,7 @@ void dml1_extract_rq_regs( rq_regs->rq_regs_l.swath_height = dml_log2(rq_param.dlg.rq_l.swath_height); rq_regs->rq_regs_c.swath_height = dml_log2(rq_param.dlg.rq_c.swath_height); - /* FIXME: take the max between luma, chroma chunk size? + /* TODO: take the max between luma, chroma chunk size? * okay for now, as we are setting chunk_bytes to 8kb anyways */ if (rq_param.sizing.rq_l.chunk_bytes >= 32 * 1024) { /*32kb */ @@ -602,7 +602,7 @@ static void get_surf_rq_param( unsigned int log2_dpte_group_length; unsigned int func_meta_row_height, func_dpte_row_height; - /* FIXME check if ppe apply for both luma and chroma in 422 case */ + /* TODO check if ppe apply for both luma and chroma in 422 case */ if (is_chroma) { vp_width = pipe_src_param.viewport_width_c / ppe; vp_height = pipe_src_param.viewport_height_c; @@ -1141,7 +1141,7 @@ void dml1_rq_dlg_get_dlg_params( ASSERT(disp_dlg_regs->refcyc_h_blank_end < (unsigned int) dml_pow(2, 13)); disp_dlg_regs->dlg_vblank_end = interlaced ? (vblank_end / 2) : vblank_end; /* 15 bits */ - prefetch_xy_calc_in_dcfclk = 24.0; /* FIXME: ip_param */ + prefetch_xy_calc_in_dcfclk = 24.0; /* TODO: ip_param */ min_dcfclk_mhz = dlg_sys_param.deepsleep_dcfclk_mhz; t_calc_us = prefetch_xy_calc_in_dcfclk / min_dcfclk_mhz; min_ttu_vblank = dlg_sys_param.t_urg_wm_us; @@ -1182,7 +1182,7 @@ void dml1_rq_dlg_get_dlg_params( dcc_en = e2e_pipe_param.pipe.src.dcc; dual_plane = is_dual_plane( (enum source_format_class) e2e_pipe_param.pipe.src.source_format); - mode_422 = 0; /* FIXME */ + mode_422 = 0; /* TODO */ access_dir = (e2e_pipe_param.pipe.src.source_scan == dm_vert); /* vp access direction: horizontal or vertical accessed */ bytes_per_element_l = get_bytes_per_element( (enum source_format_class) e2e_pipe_param.pipe.src.source_format, @@ -1837,7 +1837,7 @@ void dml1_rq_dlg_get_dlg_params( cur0_width_ub = dml_ceil((double) cur0_src_width / (double) cur0_req_width, 1) * (double) cur0_req_width; cur0_req_per_width = cur0_width_ub / (double) cur0_req_width; - hactive_cur0 = (double) cur0_src_width / hratios_cur0; /* FIXME: oswin to think about what to do for cursor */ + hactive_cur0 = (double) cur0_src_width / hratios_cur0; /* TODO: oswin to think about what to do for cursor */ if (vratio_pre_l <= 1.0) { refcyc_per_req_delivery_pre_cur0 = hactive_cur0 * ref_freq_to_pix_freq diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_common_defs.c b/drivers/gpu/drm/amd/display/dc/dml/dml_common_defs.c index b953b02a1512..723af0b2dda0 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_common_defs.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dml_common_defs.c @@ -24,7 +24,7 @@ */ #include "dml_common_defs.h" -#include "../calcs/dcn_calc_math.h" +#include "dcn_calc_math.h" #include "dml_inline_defs.h" diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h index eca140da13d8..ded71ea82413 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h @@ -27,7 +27,7 @@ #define __DML_INLINE_DEFS_H__ #include "dml_common_defs.h" -#include "../calcs/dcn_calc_math.h" +#include "dcn_calc_math.h" #include "dml_logger.h" static inline double dml_min(double a, double b) diff --git a/drivers/gpu/drm/amd/display/dc/dsc/Makefile b/drivers/gpu/drm/amd/display/dc/dsc/Makefile index e019cd9447e8..3f66868df171 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dsc/Makefile @@ -1,18 +1,35 @@ +# SPDX-License-Identifier: MIT # # Makefile for the 'dsc' sub-component of DAL. -ifneq ($(call cc-option, -mpreferred-stack-boundary=4),) - cc_stack_align := -mpreferred-stack-boundary=4 -else ifneq ($(call cc-option, -mstack-alignment=16),) - cc_stack_align := -mstack-alignment=16 +ifdef CONFIG_X86 +dsc_ccflags := -mhard-float -msse endif -dsc_ccflags := -mhard-float -msse $(cc_stack_align) +ifdef CONFIG_PPC64 +dsc_ccflags := -mhard-float -maltivec +endif + +ifdef CONFIG_CC_IS_GCC +ifeq ($(call cc-ifversion, -lt, 0701, y), y) +IS_OLD_GCC = 1 +endif +endif + +ifdef CONFIG_X86 +ifdef IS_OLD_GCC +# Stack alignment mismatch, proceed with caution. +# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 +# (8B stack alignment). +dsc_ccflags += -mpreferred-stack-boundary=4 +else +dsc_ccflags += -msse2 +endif +endif -CFLAGS_rc_calc.o := $(dsc_ccflags) -CFLAGS_rc_calc_dpi.o := $(dsc_ccflags) -CFLAGS_codec_main_amd.o := $(dsc_ccflags) -CFLAGS_dc_dsc.o := $(dsc_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc.o := $(dsc_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc_dpi.o := $(dsc_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dsc/dc_dsc.o := $(dsc_ccflags) DSC = dc_dsc.o rc_calc.o rc_calc_dpi.o diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c index ef5f84a144c3..87d682d25278 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c @@ -22,14 +22,69 @@ * Author: AMD */ -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT -#include "dc.h" -#include "core_types.h" +#include "dc_hw_types.h" #include "dsc.h" #include <drm/drm_dp_helper.h> +#include "dc.h" /* This module's internal functions */ +/* default DSC policy target bitrate limit is 16bpp */ +static uint32_t dsc_policy_max_target_bpp_limit = 16; + +static uint32_t dc_dsc_bandwidth_in_kbps_from_timing( + const struct dc_crtc_timing *timing) +{ + uint32_t bits_per_channel = 0; + uint32_t kbps; + + if (timing->flags.DSC) { + kbps = (timing->pix_clk_100hz * timing->dsc_cfg.bits_per_pixel); + kbps = kbps / 160 + ((kbps % 160) ? 1 : 0); + return kbps; + } + + switch (timing->display_color_depth) { + case COLOR_DEPTH_666: + bits_per_channel = 6; + break; + case COLOR_DEPTH_888: + bits_per_channel = 8; + break; + case COLOR_DEPTH_101010: + bits_per_channel = 10; + break; + case COLOR_DEPTH_121212: + bits_per_channel = 12; + break; + case COLOR_DEPTH_141414: + bits_per_channel = 14; + break; + case COLOR_DEPTH_161616: + bits_per_channel = 16; + break; + default: + break; + } + + ASSERT(bits_per_channel != 0); + + kbps = timing->pix_clk_100hz / 10; + kbps *= bits_per_channel; + + if (timing->flags.Y_ONLY != 1) { + /*Only YOnly make reduce bandwidth by 1/3 compares to RGB*/ + kbps *= 3; + if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) + kbps /= 2; + else if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR422) + kbps = kbps * 2 / 3; + } + + return kbps; + +} + static bool dsc_buff_block_size_from_dpcd(int dpcd_buff_block_size, int *buff_block_size) { @@ -161,16 +216,19 @@ static bool dsc_bpp_increment_div_from_dpcd(int bpp_increment_dpcd, uint32_t *bp } static void get_dsc_enc_caps( - const struct dc *dc, + const struct display_stream_compressor *dsc, struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz) { // This is a static HW query, so we can use any DSC - struct display_stream_compressor *dsc = dc->res_pool->dscs[0]; memset(dsc_enc_caps, 0, sizeof(struct dsc_enc_caps)); - if (dsc) - dsc->funcs->dsc_get_enc_caps(dsc_enc_caps, pixel_clock_100Hz); + if (dsc) { + if (!dsc->ctx->dc->debug.disable_dsc) + dsc->funcs->dsc_get_enc_caps(dsc_enc_caps, pixel_clock_100Hz); + if (dsc->ctx->dc->debug.native422_support) + dsc_enc_caps->color_formats.bits.YCBCR_NATIVE_422 = 1; + } } /* Returns 'false' if no intersection was found for at least one capablity. @@ -241,14 +299,6 @@ static bool intersect_dsc_caps( return true; } -struct dc_dsc_policy { - bool use_min_slices_h; - int max_slices_h; // Maximum available if 0 - int num_slices_v; - int max_target_bpp; - int min_target_bpp; // Minimum target bits per pixel -}; - static inline uint32_t dsc_div_by_10_round_up(uint32_t value) { return (value + 9) / 10; @@ -270,19 +320,6 @@ static inline uint32_t calc_dsc_bpp_x16(uint32_t stream_bandwidth_kbps, uint32_t return dsc_target_bpp_x16; } -const struct dc_dsc_policy dsc_policy = { - .use_min_slices_h = true, // DSC Policy: Use minimum number of slices that fits the pixel clock - .max_slices_h = 0, // DSC Policy: Use max available slices (in our case 4 for or 8, depending on the mode) - /* DSC Policy: Number of vertical slices set to 2 for no particular reason. - * Seems small enough to not affect the quality too much, while still providing some error - * propagation control (which may also help debugging). - */ - .num_slices_v = 16, - .max_target_bpp = 16, - .min_target_bpp = 8, -}; - - /* Get DSC bandwidth range based on [min_bpp, max_bpp] target bitrate range, and timing's pixel clock * and uncompressed bandwidth. */ @@ -294,7 +331,7 @@ static void get_dsc_bandwidth_range( struct dc_dsc_bw_range *range) { /* native stream bandwidth */ - range->stream_kbps = dc_bandwidth_in_kbps_from_timing(timing); + range->stream_kbps = dc_dsc_bandwidth_in_kbps_from_timing(timing); /* max dsc target bpp */ range->max_kbps = dsc_div_by_10_round_up(max_bpp * timing->pix_clk_100hz); @@ -516,6 +553,7 @@ static bool setup_dsc_config( const struct dsc_enc_caps *dsc_enc_caps, int target_bandwidth_kbps, const struct dc_crtc_timing *timing, + int min_slice_height_override, struct dc_dsc_config *dsc_cfg) { struct dsc_enc_caps dsc_common_caps; @@ -528,11 +566,13 @@ static bool setup_dsc_config( int sink_per_slice_throughput_mps; int branch_max_throughput_mps = 0; bool is_dsc_possible = false; - int num_slices_v; int pic_height; + int slice_height; + struct dc_dsc_policy policy; memset(dsc_cfg, 0, sizeof(struct dc_dsc_config)); + dc_dsc_get_policy_for_timing(timing, &policy); pic_width = timing->h_addressable + timing->h_border_left + timing->h_border_right; pic_height = timing->v_addressable + timing->v_border_top + timing->v_border_bottom; @@ -548,7 +588,12 @@ static bool setup_dsc_config( goto done; if (target_bandwidth_kbps > 0) { - is_dsc_possible = decide_dsc_target_bpp_x16(&dsc_policy, &dsc_common_caps, target_bandwidth_kbps, timing, &target_bpp); + is_dsc_possible = decide_dsc_target_bpp_x16( + &policy, + &dsc_common_caps, + target_bandwidth_kbps, + timing, + &target_bpp); dsc_cfg->bits_per_pixel = target_bpp; } if (!is_dsc_possible) @@ -615,7 +660,7 @@ static bool setup_dsc_config( if (!is_dsc_possible) goto done; - // DSC slicing + // Slice width (i.e. number of slices per line) max_slices_h = get_max_dsc_slices(dsc_common_caps.slice_caps); while (max_slices_h > 0) { @@ -650,20 +695,20 @@ static bool setup_dsc_config( if (!is_dsc_possible) goto done; - if (dsc_policy.use_min_slices_h) { + if (policy.use_min_slices_h) { if (min_slices_h > 0) num_slices_h = min_slices_h; else if (max_slices_h > 0) { // Fall back to max slices if min slices is not working out - if (dsc_policy.max_slices_h) - num_slices_h = min(dsc_policy.max_slices_h, max_slices_h); + if (policy.max_slices_h) + num_slices_h = min(policy.max_slices_h, max_slices_h); else num_slices_h = max_slices_h; } else is_dsc_possible = false; } else { if (max_slices_h > 0) { - if (dsc_policy.max_slices_h) - num_slices_h = min(dsc_policy.max_slices_h, max_slices_h); + if (policy.max_slices_h) + num_slices_h = min(policy.max_slices_h, max_slices_h); else num_slices_h = max_slices_h; } else if (min_slices_h > 0) // Fall back to min slices if max slices is not possible @@ -678,29 +723,29 @@ static bool setup_dsc_config( dsc_cfg->num_slices_h = num_slices_h; slice_width = pic_width / num_slices_h; - // Vertical number of slices: start from policy and pick the first one that height is divisible by. + is_dsc_possible = slice_width <= dsc_common_caps.max_slice_width; + if (!is_dsc_possible) + goto done; + + // Slice height (i.e. number of slices per column): start with policy and pick the first one that height is divisible by. // For 4:2:0 make sure the slice height is divisible by 2 as well. - num_slices_v = dsc_policy.num_slices_v; - if (num_slices_v < 1) - num_slices_v = 1; - - while (num_slices_v >= 1) { - if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) { - int slice_height = pic_height / num_slices_v; - if (pic_height % num_slices_v == 0 && slice_height % 2 == 0) - break; - } else if (pic_height % num_slices_v == 0) - break; + if (min_slice_height_override == 0) + slice_height = min(policy.min_slice_height, pic_height); + else + slice_height = min(min_slice_height_override, pic_height); - num_slices_v--; - } + while (slice_height < pic_height && (pic_height % slice_height != 0 || + (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420 && slice_height % 2 != 0))) + slice_height++; - dsc_cfg->num_slices_v = num_slices_v; + if (timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) // For the case when pic_height < dsc_policy.min_sice_height + is_dsc_possible = (slice_height % 2 == 0); - is_dsc_possible = slice_width <= dsc_common_caps.max_slice_width; if (!is_dsc_possible) goto done; + dsc_cfg->num_slices_v = pic_height/slice_height; + // Final decission: can we do DSC or not? if (is_dsc_possible) { // Fill out the rest of DSC settings @@ -716,7 +761,7 @@ done: return is_dsc_possible; } -bool dc_dsc_parse_dsc_dpcd(const uint8_t *dpcd_dsc_basic_data, const uint8_t *dpcd_dsc_ext_data, struct dsc_dec_dpcd_caps *dsc_sink_caps) +bool dc_dsc_parse_dsc_dpcd(const struct dc *dc, const uint8_t *dpcd_dsc_basic_data, const uint8_t *dpcd_dsc_ext_data, struct dsc_dec_dpcd_caps *dsc_sink_caps) { if (!dpcd_dsc_basic_data) return false; @@ -769,6 +814,23 @@ bool dc_dsc_parse_dsc_dpcd(const uint8_t *dpcd_dsc_basic_data, const uint8_t *dp if (!dsc_bpp_increment_div_from_dpcd(dpcd_dsc_basic_data[DP_DSC_BITS_PER_PIXEL_INC - DP_DSC_SUPPORT], &dsc_sink_caps->bpp_increment_div)) return false; + if (dc->debug.dsc_bpp_increment_div) { + /* dsc_bpp_increment_div should onl be 1, 2, 4, 8 or 16, but rather than rejecting invalid values, + * we'll accept all and get it into range. This also makes the above check against 0 redundant, + * but that one stresses out the override will be only used if it's not 0. + */ + if (dc->debug.dsc_bpp_increment_div >= 1) + dsc_sink_caps->bpp_increment_div = 1; + if (dc->debug.dsc_bpp_increment_div >= 2) + dsc_sink_caps->bpp_increment_div = 2; + if (dc->debug.dsc_bpp_increment_div >= 4) + dsc_sink_caps->bpp_increment_div = 4; + if (dc->debug.dsc_bpp_increment_div >= 8) + dsc_sink_caps->bpp_increment_div = 8; + if (dc->debug.dsc_bpp_increment_div >= 16) + dsc_sink_caps->bpp_increment_div = 16; + } + /* Extended caps */ if (dpcd_dsc_ext_data == NULL) { // Extended DPCD DSC data can be null, e.g. because it doesn't apply to SST dsc_sink_caps->branch_overall_throughput_0_mps = 0; @@ -809,7 +871,8 @@ bool dc_dsc_parse_dsc_dpcd(const uint8_t *dpcd_dsc_basic_data, const uint8_t *dp * If DSC is not possible, leave '*range' untouched. */ bool dc_dsc_compute_bandwidth_range( - const struct dc *dc, + const struct display_stream_compressor *dsc, + const uint32_t dsc_min_slice_height_override, const uint32_t min_bpp, const uint32_t max_bpp, const struct dsc_dec_dpcd_caps *dsc_sink_caps, @@ -821,16 +884,14 @@ bool dc_dsc_compute_bandwidth_range( struct dsc_enc_caps dsc_common_caps; struct dc_dsc_config config; - get_dsc_enc_caps(dc, &dsc_enc_caps, timing->pix_clk_100hz); + get_dsc_enc_caps(dsc, &dsc_enc_caps, timing->pix_clk_100hz); is_dsc_possible = intersect_dsc_caps(dsc_sink_caps, &dsc_enc_caps, timing->pixel_encoding, &dsc_common_caps); if (is_dsc_possible) - is_dsc_possible = setup_dsc_config(dsc_sink_caps, - &dsc_enc_caps, - 0, - timing, &config); + is_dsc_possible = setup_dsc_config(dsc_sink_caps, &dsc_enc_caps, 0, timing, + dsc_min_slice_height_override, &config); if (is_dsc_possible) get_dsc_bandwidth_range(min_bpp, max_bpp, &dsc_common_caps, timing, range); @@ -839,8 +900,9 @@ bool dc_dsc_compute_bandwidth_range( } bool dc_dsc_compute_config( - const struct dc *dc, + const struct display_stream_compressor *dsc, const struct dsc_dec_dpcd_caps *dsc_sink_caps, + const uint32_t dsc_min_slice_height_override, uint32_t target_bandwidth_kbps, const struct dc_crtc_timing *timing, struct dc_dsc_config *dsc_cfg) @@ -848,11 +910,74 @@ bool dc_dsc_compute_config( bool is_dsc_possible = false; struct dsc_enc_caps dsc_enc_caps; - get_dsc_enc_caps(dc, &dsc_enc_caps, timing->pix_clk_100hz); + get_dsc_enc_caps(dsc, &dsc_enc_caps, timing->pix_clk_100hz); is_dsc_possible = setup_dsc_config(dsc_sink_caps, &dsc_enc_caps, target_bandwidth_kbps, - timing, dsc_cfg); + timing, dsc_min_slice_height_override, dsc_cfg); return is_dsc_possible; } -#endif /* CONFIG_DRM_AMD_DC_DSC_SUPPORT */ + +void dc_dsc_get_policy_for_timing(const struct dc_crtc_timing *timing, struct dc_dsc_policy *policy) +{ + uint32_t bpc = 0; + + policy->min_target_bpp = 0; + policy->max_target_bpp = 0; + + /* DSC Policy: Use minimum number of slices that fits the pixel clock */ + policy->use_min_slices_h = true; + + /* DSC Policy: Use max available slices + * (in our case 4 for or 8, depending on the mode) + */ + policy->max_slices_h = 0; + + /* DSC Policy: Use slice height recommended + * by VESA DSC Spreadsheet user guide + */ + policy->min_slice_height = 108; + + /* DSC Policy: follow DP specs with an internal upper limit to 16 bpp + * for better interoperability + */ + switch (timing->display_color_depth) { + case COLOR_DEPTH_888: + bpc = 8; + break; + case COLOR_DEPTH_101010: + bpc = 10; + break; + case COLOR_DEPTH_121212: + bpc = 12; + break; + default: + return; + } + switch (timing->pixel_encoding) { + case PIXEL_ENCODING_RGB: + case PIXEL_ENCODING_YCBCR444: + case PIXEL_ENCODING_YCBCR422: /* assume no YCbCr422 native support */ + /* DP specs limits to 8 */ + policy->min_target_bpp = 8; + /* DP specs limits to 3 x bpc */ + policy->max_target_bpp = 3 * bpc; + break; + case PIXEL_ENCODING_YCBCR420: + /* DP specs limits to 6 */ + policy->min_target_bpp = 6; + /* DP specs limits to 1.5 x bpc assume bpc is an even number */ + policy->max_target_bpp = bpc * 3 / 2; + break; + default: + return; + } + /* internal upper limit, default 16 bpp */ + if (policy->max_target_bpp > dsc_policy_max_target_bpp_limit) + policy->max_target_bpp = dsc_policy_max_target_bpp_limit; +} + +void dc_dsc_policy_set_max_target_bpp_limit(uint32_t limit) +{ + dsc_policy_max_target_bpp_limit = limit; +} diff --git a/drivers/gpu/drm/amd/display/dc/dsc/drm_dsc_dc.c b/drivers/gpu/drm/amd/display/dc/dsc/drm_dsc_dc.c deleted file mode 100644 index 340ef4d41ebd..000000000000 --- a/drivers/gpu/drm/amd/display/dc/dsc/drm_dsc_dc.c +++ /dev/null @@ -1,388 +0,0 @@ -// SPDX-License-Identifier: MIT -/* - * Copyright © 2018 Intel Corp - * - * Author: - * Manasi Navare <manasi.d.navare@intel.com> - */ - -/* DC versions of linux includes */ -#include <include/drm_dsc_dc.h> - -#define EXPORT_SYMBOL(symbol) /* nothing */ -#define BUILD_BUG_ON(cond) /* nothing */ -#define DIV_ROUND_UP(a, b) (((b) + (a) - 1) / (b)) -#define ERANGE -1 -#define DRM_DEBUG_KMS(msg) /* nothing */ -#define cpu_to_be16(__x) little_to_big(__x) - -static unsigned short little_to_big(int data) -{ - /* Swap lower and upper byte. DMCU uses big endian format. */ - return (0xff & (data >> 8)) + ((data & 0xff) << 8); -} - -/* - * Everything below this comment was copied directly from drm_dsc.c. - * Only the functions needed in DC are included. - * Please keep this file synced with upstream. - */ - -/** - * DOC: dsc helpers - * - * These functions contain some common logic and helpers to deal with VESA - * Display Stream Compression standard required for DSC on Display Port/eDP or - * MIPI display interfaces. - */ - -/** - * drm_dsc_pps_payload_pack() - Populates the DSC PPS - * - * @pps_payload: - * Bitwise struct for DSC Picture Parameter Set. This is defined - * by &struct drm_dsc_picture_parameter_set - * @dsc_cfg: - * DSC Configuration data filled by driver as defined by - * &struct drm_dsc_config - * - * DSC source device sends a picture parameter set (PPS) containing the - * information required by the sink to decode the compressed frame. Driver - * populates the DSC PPS struct using the DSC configuration parameters in - * the order expected by the DSC Display Sink device. For the DSC, the sink - * device expects the PPS payload in big endian format for fields - * that span more than 1 byte. - */ -void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_payload, - const struct drm_dsc_config *dsc_cfg) -{ - int i; - - /* Protect against someone accidently changing struct size */ - BUILD_BUG_ON(sizeof(*pps_payload) != - DP_SDP_PPS_HEADER_PAYLOAD_BYTES_MINUS_1 + 1); - - memset(pps_payload, 0, sizeof(*pps_payload)); - - /* PPS 0 */ - pps_payload->dsc_version = - dsc_cfg->dsc_version_minor | - dsc_cfg->dsc_version_major << DSC_PPS_VERSION_MAJOR_SHIFT; - - /* PPS 1, 2 is 0 */ - - /* PPS 3 */ - pps_payload->pps_3 = - dsc_cfg->line_buf_depth | - dsc_cfg->bits_per_component << DSC_PPS_BPC_SHIFT; - - /* PPS 4 */ - pps_payload->pps_4 = - ((dsc_cfg->bits_per_pixel & DSC_PPS_BPP_HIGH_MASK) >> - DSC_PPS_MSB_SHIFT) | - dsc_cfg->vbr_enable << DSC_PPS_VBR_EN_SHIFT | - dsc_cfg->simple_422 << DSC_PPS_SIMPLE422_SHIFT | - dsc_cfg->convert_rgb << DSC_PPS_CONVERT_RGB_SHIFT | - dsc_cfg->block_pred_enable << DSC_PPS_BLOCK_PRED_EN_SHIFT; - - /* PPS 5 */ - pps_payload->bits_per_pixel_low = - (dsc_cfg->bits_per_pixel & DSC_PPS_LSB_MASK); - - /* - * The DSC panel expects the PPS packet to have big endian format - * for data spanning 2 bytes. Use a macro cpu_to_be16() to convert - * to big endian format. If format is little endian, it will swap - * bytes to convert to Big endian else keep it unchanged. - */ - - /* PPS 6, 7 */ - pps_payload->pic_height = cpu_to_be16(dsc_cfg->pic_height); - - /* PPS 8, 9 */ - pps_payload->pic_width = cpu_to_be16(dsc_cfg->pic_width); - - /* PPS 10, 11 */ - pps_payload->slice_height = cpu_to_be16(dsc_cfg->slice_height); - - /* PPS 12, 13 */ - pps_payload->slice_width = cpu_to_be16(dsc_cfg->slice_width); - - /* PPS 14, 15 */ - pps_payload->chunk_size = cpu_to_be16(dsc_cfg->slice_chunk_size); - - /* PPS 16 */ - pps_payload->initial_xmit_delay_high = - ((dsc_cfg->initial_xmit_delay & - DSC_PPS_INIT_XMIT_DELAY_HIGH_MASK) >> - DSC_PPS_MSB_SHIFT); - - /* PPS 17 */ - pps_payload->initial_xmit_delay_low = - (dsc_cfg->initial_xmit_delay & DSC_PPS_LSB_MASK); - - /* PPS 18, 19 */ - pps_payload->initial_dec_delay = - cpu_to_be16(dsc_cfg->initial_dec_delay); - - /* PPS 20 is 0 */ - - /* PPS 21 */ - pps_payload->initial_scale_value = - dsc_cfg->initial_scale_value; - - /* PPS 22, 23 */ - pps_payload->scale_increment_interval = - cpu_to_be16(dsc_cfg->scale_increment_interval); - - /* PPS 24 */ - pps_payload->scale_decrement_interval_high = - ((dsc_cfg->scale_decrement_interval & - DSC_PPS_SCALE_DEC_INT_HIGH_MASK) >> - DSC_PPS_MSB_SHIFT); - - /* PPS 25 */ - pps_payload->scale_decrement_interval_low = - (dsc_cfg->scale_decrement_interval & DSC_PPS_LSB_MASK); - - /* PPS 26[7:0], PPS 27[7:5] RESERVED */ - - /* PPS 27 */ - pps_payload->first_line_bpg_offset = - dsc_cfg->first_line_bpg_offset; - - /* PPS 28, 29 */ - pps_payload->nfl_bpg_offset = - cpu_to_be16(dsc_cfg->nfl_bpg_offset); - - /* PPS 30, 31 */ - pps_payload->slice_bpg_offset = - cpu_to_be16(dsc_cfg->slice_bpg_offset); - - /* PPS 32, 33 */ - pps_payload->initial_offset = - cpu_to_be16(dsc_cfg->initial_offset); - - /* PPS 34, 35 */ - pps_payload->final_offset = cpu_to_be16(dsc_cfg->final_offset); - - /* PPS 36 */ - pps_payload->flatness_min_qp = dsc_cfg->flatness_min_qp; - - /* PPS 37 */ - pps_payload->flatness_max_qp = dsc_cfg->flatness_max_qp; - - /* PPS 38, 39 */ - pps_payload->rc_model_size = - cpu_to_be16(DSC_RC_MODEL_SIZE_CONST); - - /* PPS 40 */ - pps_payload->rc_edge_factor = DSC_RC_EDGE_FACTOR_CONST; - - /* PPS 41 */ - pps_payload->rc_quant_incr_limit0 = - dsc_cfg->rc_quant_incr_limit0; - - /* PPS 42 */ - pps_payload->rc_quant_incr_limit1 = - dsc_cfg->rc_quant_incr_limit1; - - /* PPS 43 */ - pps_payload->rc_tgt_offset = DSC_RC_TGT_OFFSET_LO_CONST | - DSC_RC_TGT_OFFSET_HI_CONST << DSC_PPS_RC_TGT_OFFSET_HI_SHIFT; - - /* PPS 44 - 57 */ - for (i = 0; i < DSC_NUM_BUF_RANGES - 1; i++) - pps_payload->rc_buf_thresh[i] = - dsc_cfg->rc_buf_thresh[i]; - - /* PPS 58 - 87 */ - /* - * For DSC sink programming the RC Range parameter fields - * are as follows: Min_qp[15:11], max_qp[10:6], offset[5:0] - */ - for (i = 0; i < DSC_NUM_BUF_RANGES; i++) { - pps_payload->rc_range_parameters[i] = - ((dsc_cfg->rc_range_params[i].range_min_qp << - DSC_PPS_RC_RANGE_MINQP_SHIFT) | - (dsc_cfg->rc_range_params[i].range_max_qp << - DSC_PPS_RC_RANGE_MAXQP_SHIFT) | - (dsc_cfg->rc_range_params[i].range_bpg_offset)); - pps_payload->rc_range_parameters[i] = - cpu_to_be16(pps_payload->rc_range_parameters[i]); - } - - /* PPS 88 */ - pps_payload->native_422_420 = dsc_cfg->native_422 | - dsc_cfg->native_420 << DSC_PPS_NATIVE_420_SHIFT; - - /* PPS 89 */ - pps_payload->second_line_bpg_offset = - dsc_cfg->second_line_bpg_offset; - - /* PPS 90, 91 */ - pps_payload->nsl_bpg_offset = - cpu_to_be16(dsc_cfg->nsl_bpg_offset); - - /* PPS 92, 93 */ - pps_payload->second_line_offset_adj = - cpu_to_be16(dsc_cfg->second_line_offset_adj); - - /* PPS 94 - 127 are O */ -} -EXPORT_SYMBOL(drm_dsc_pps_payload_pack); - -/** - * drm_dsc_compute_rc_parameters() - Write rate control - * parameters to the dsc configuration defined in - * &struct drm_dsc_config in accordance with the DSC 1.2 - * specification. Some configuration fields must be present - * beforehand. - * - * @vdsc_cfg: - * DSC Configuration data partially filled by driver - */ -int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg) -{ - unsigned long groups_per_line = 0; - unsigned long groups_total = 0; - unsigned long num_extra_mux_bits = 0; - unsigned long slice_bits = 0; - unsigned long hrd_delay = 0; - unsigned long final_scale = 0; - unsigned long rbs_min = 0; - - if (vdsc_cfg->native_420 || vdsc_cfg->native_422) { - /* Number of groups used to code each line of a slice */ - groups_per_line = DIV_ROUND_UP(vdsc_cfg->slice_width / 2, - DSC_RC_PIXELS_PER_GROUP); - - /* chunksize in Bytes */ - vdsc_cfg->slice_chunk_size = DIV_ROUND_UP(vdsc_cfg->slice_width / 2 * - vdsc_cfg->bits_per_pixel, - (8 * 16)); - } else { - /* Number of groups used to code each line of a slice */ - groups_per_line = DIV_ROUND_UP(vdsc_cfg->slice_width, - DSC_RC_PIXELS_PER_GROUP); - - /* chunksize in Bytes */ - vdsc_cfg->slice_chunk_size = DIV_ROUND_UP(vdsc_cfg->slice_width * - vdsc_cfg->bits_per_pixel, - (8 * 16)); - } - - if (vdsc_cfg->convert_rgb) - num_extra_mux_bits = 3 * (vdsc_cfg->mux_word_size + - (4 * vdsc_cfg->bits_per_component + 4) - - 2); - else if (vdsc_cfg->native_422) - num_extra_mux_bits = 4 * vdsc_cfg->mux_word_size + - (4 * vdsc_cfg->bits_per_component + 4) + - 3 * (4 * vdsc_cfg->bits_per_component) - 2; - else - num_extra_mux_bits = 3 * vdsc_cfg->mux_word_size + - (4 * vdsc_cfg->bits_per_component + 4) + - 2 * (4 * vdsc_cfg->bits_per_component) - 2; - /* Number of bits in one Slice */ - slice_bits = 8 * vdsc_cfg->slice_chunk_size * vdsc_cfg->slice_height; - - while ((num_extra_mux_bits > 0) && - ((slice_bits - num_extra_mux_bits) % vdsc_cfg->mux_word_size)) - num_extra_mux_bits--; - - if (groups_per_line < vdsc_cfg->initial_scale_value - 8) - vdsc_cfg->initial_scale_value = groups_per_line + 8; - - /* scale_decrement_interval calculation according to DSC spec 1.11 */ - if (vdsc_cfg->initial_scale_value > 8) - vdsc_cfg->scale_decrement_interval = groups_per_line / - (vdsc_cfg->initial_scale_value - 8); - else - vdsc_cfg->scale_decrement_interval = DSC_SCALE_DECREMENT_INTERVAL_MAX; - - vdsc_cfg->final_offset = vdsc_cfg->rc_model_size - - (vdsc_cfg->initial_xmit_delay * - vdsc_cfg->bits_per_pixel + 8) / 16 + num_extra_mux_bits; - - if (vdsc_cfg->final_offset >= vdsc_cfg->rc_model_size) { - DRM_DEBUG_KMS("FinalOfs < RcModelSze for this InitialXmitDelay\n"); - return -ERANGE; - } - - final_scale = (vdsc_cfg->rc_model_size * 8) / - (vdsc_cfg->rc_model_size - vdsc_cfg->final_offset); - if (vdsc_cfg->slice_height > 1) - /* - * NflBpgOffset is 16 bit value with 11 fractional bits - * hence we multiply by 2^11 for preserving the - * fractional part - */ - vdsc_cfg->nfl_bpg_offset = DIV_ROUND_UP((vdsc_cfg->first_line_bpg_offset << 11), - (vdsc_cfg->slice_height - 1)); - else - vdsc_cfg->nfl_bpg_offset = 0; - - /* 2^16 - 1 */ - if (vdsc_cfg->nfl_bpg_offset > 65535) { - DRM_DEBUG_KMS("NflBpgOffset is too large for this slice height\n"); - return -ERANGE; - } - - /* Number of groups used to code the entire slice */ - groups_total = groups_per_line * vdsc_cfg->slice_height; - - /* slice_bpg_offset is 16 bit value with 11 fractional bits */ - vdsc_cfg->slice_bpg_offset = DIV_ROUND_UP(((vdsc_cfg->rc_model_size - - vdsc_cfg->initial_offset + - num_extra_mux_bits) << 11), - groups_total); - - if (final_scale > 9) { - /* - * ScaleIncrementInterval = - * finaloffset/((NflBpgOffset + SliceBpgOffset)*8(finalscale - 1.125)) - * as (NflBpgOffset + SliceBpgOffset) has 11 bit fractional value, - * we need divide by 2^11 from pstDscCfg values - */ - vdsc_cfg->scale_increment_interval = - (vdsc_cfg->final_offset * (1 << 11)) / - ((vdsc_cfg->nfl_bpg_offset + - vdsc_cfg->slice_bpg_offset) * - (final_scale - 9)); - } else { - /* - * If finalScaleValue is less than or equal to 9, a value of 0 should - * be used to disable the scale increment at the end of the slice - */ - vdsc_cfg->scale_increment_interval = 0; - } - - if (vdsc_cfg->scale_increment_interval > 65535) { - DRM_DEBUG_KMS("ScaleIncrementInterval is large for slice height\n"); - return -ERANGE; - } - - /* - * DSC spec mentions that bits_per_pixel specifies the target - * bits/pixel (bpp) rate that is used by the encoder, - * in steps of 1/16 of a bit per pixel - */ - rbs_min = vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset + - DIV_ROUND_UP(vdsc_cfg->initial_xmit_delay * - vdsc_cfg->bits_per_pixel, 16) + - groups_per_line * vdsc_cfg->first_line_bpg_offset; - - hrd_delay = DIV_ROUND_UP((rbs_min * 16), vdsc_cfg->bits_per_pixel); - vdsc_cfg->rc_bits = (hrd_delay * vdsc_cfg->bits_per_pixel) / 16; - vdsc_cfg->initial_dec_delay = hrd_delay - vdsc_cfg->initial_xmit_delay; - - /* As per DSC spec v1.2a recommendation: */ - if (vdsc_cfg->native_420) - vdsc_cfg->second_line_offset_adj = 512; - else - vdsc_cfg->second_line_offset_adj = 0; - - return 0; -} -EXPORT_SYMBOL(drm_dsc_compute_rc_parameters); diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dscc_types.h b/drivers/gpu/drm/amd/display/dc/dsc/dscc_types.h index 020ad8f685ea..9f70e87b3ecb 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dscc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dsc/dscc_types.h @@ -1,4 +1,3 @@ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* * Copyright 2017 Advanced Micro Devices, Inc. @@ -51,4 +50,3 @@ int dscc_compute_dsc_parameters(const struct drm_dsc_config *pps, struct dsc_par #endif -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dsc/qp_tables.h b/drivers/gpu/drm/amd/display/dc/dsc/qp_tables.h index f66d006eac5d..e5fac9f4181d 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/qp_tables.h +++ b/drivers/gpu/drm/amd/display/dc/dsc/qp_tables.h @@ -1,4 +1,3 @@ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* * Copyright 2017 Advanced Micro Devices, Inc. @@ -703,4 +702,3 @@ const qp_table qp_table_422_8bpc_max = { { 16, { 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3, 4} } }; -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c index ca51e83f8764..03ae15946c6d 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c @@ -1,4 +1,3 @@ -#if defined(CONFIG_DRM_AMD_DC_DSC_SUPPORT) /* * Copyright 2017 Advanced Micro Devices, Inc. @@ -177,7 +176,6 @@ void calc_rc_params(struct rc_params *rc, enum colour_mode cm, enum bits_per_com { float bpp_group; float initial_xmit_delay_factor; - int source_bpp; int padding_pixels; int i; @@ -217,8 +215,6 @@ void calc_rc_params(struct rc_params *rc, enum colour_mode cm, enum bits_per_com rc->initial_xmit_delay++; } - source_bpp = MODE_SELECT(bpc * 3, bpc * 2, bpc * 1.5); - rc->flatness_min_qp = ((bpc == BPC_8) ? (3) : ((bpc == BPC_10) ? (7) : (11))) - ((minor_version == 1 && cm == CM_444) ? 1 : 0); rc->flatness_max_qp = ((bpc == BPC_8) ? (12) : ((bpc == BPC_10) ? (16) : (20))) - ((minor_version == 1 && cm == CM_444) ? 1 : 0); rc->flatness_det_thresh = 2 << (bpc - 8); @@ -255,4 +251,3 @@ void calc_rc_params(struct rc_params *rc, enum colour_mode cm, enum bits_per_com rc->rc_buf_thresh[13] = 8064; } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.h b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.h index f1d6e793bc61..b6b1f09c2009 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.h +++ b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.h @@ -1,4 +1,3 @@ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* * Copyright 2017 Advanced Micro Devices, Inc. @@ -82,4 +81,3 @@ void calc_rc_params(struct rc_params *rc, enum colour_mode cm, enum bits_per_com #endif -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc_dpi.c b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc_dpi.c index 73172fd0b529..1f6e63b71456 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc_dpi.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc_dpi.c @@ -1,4 +1,3 @@ -#if defined(CONFIG_DRM_AMD_DC_DSC_SUPPORT) /* * Copyright 2012-17 Advanced Micro Devices, Inc. * @@ -144,4 +143,3 @@ int dscc_compute_dsc_parameters(const struct drm_dsc_config *pps, struct dsc_par return ret; } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/Makefile b/drivers/gpu/drm/amd/display/dc/gpio/Makefile index c3d92878875d..202baa210cc8 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/Makefile +++ b/drivers/gpu/drm/amd/display/dc/gpio/Makefile @@ -24,7 +24,7 @@ # It provides the control and status of HW GPIO pins. GPIO = gpio_base.o gpio_service.o hw_factory.o \ - hw_gpio.o hw_hpd.o hw_ddc.o hw_translate.o + hw_gpio.o hw_hpd.o hw_ddc.o hw_generic.o hw_translate.o AMD_DAL_GPIO = $(addprefix $(AMDDALPATH)/dc/gpio/,$(GPIO)) @@ -61,26 +61,32 @@ AMD_DISPLAY_FILES += $(AMD_DAL_GPIO_DCE120) ############################################################################### # DCN 1x ############################################################################### -ifdef CONFIG_DRM_AMD_DC_DCN1_0 +ifdef CONFIG_DRM_AMD_DC_DCN GPIO_DCN10 = hw_translate_dcn10.o hw_factory_dcn10.o AMD_DAL_GPIO_DCN10 = $(addprefix $(AMDDALPATH)/dc/gpio/dcn10/,$(GPIO_DCN10)) AMD_DISPLAY_FILES += $(AMD_DAL_GPIO_DCN10) -endif ############################################################################### # DCN 2 ############################################################################### -ifdef CONFIG_DRM_AMD_DC_DCN2_0 GPIO_DCN20 = hw_translate_dcn20.o hw_factory_dcn20.o AMD_DAL_GPIO_DCN20 = $(addprefix $(AMDDALPATH)/dc/gpio/dcn20/,$(GPIO_DCN20)) AMD_DISPLAY_FILES += $(AMD_DAL_GPIO_DCN20) -endif ############################################################################### +# DCN 21 +############################################################################### +GPIO_DCN21 = hw_translate_dcn21.o hw_factory_dcn21.o + +AMD_DAL_GPIO_DCN21 = $(addprefix $(AMDDALPATH)/dc/gpio/dcn21/,$(GPIO_DCN21)) + +AMD_DISPLAY_FILES += $(AMD_DAL_GPIO_DCN21) +endif +############################################################################### # Diagnostics on FPGA ############################################################################### GPIO_DIAG_FPGA = hw_translate_diag.o hw_factory_diag.o diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce110/hw_factory_dce110.c b/drivers/gpu/drm/amd/display/dc/gpio/dce110/hw_factory_dce110.c index 20d81bca119c..66e4841f41e4 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dce110/hw_factory_dce110.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dce110/hw_factory_dce110.c @@ -24,9 +24,15 @@ */ #include "dm_services.h" + #include "include/gpio_types.h" #include "../hw_factory.h" +#include "../hw_gpio.h" +#include "../hw_ddc.h" +#include "../hw_hpd.h" +#include "../hw_generic.h" + #include "hw_factory_dce110.h" #include "dce/dce_11_0_d.h" @@ -143,12 +149,12 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) } static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = NULL, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = NULL, .define_hpd_registers = define_hpd_registers, .define_ddc_registers = define_ddc_registers }; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c b/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c index ea3f888e5c65..cf98aa827a9a 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dce120/hw_factory_dce120.c @@ -27,10 +27,10 @@ #include "include/gpio_types.h" #include "../hw_factory.h" - #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" #include "hw_factory_dce120.h" @@ -164,12 +164,12 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) /* fucntion table */ static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = NULL, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = NULL, .define_hpd_registers = define_hpd_registers, .define_ddc_registers = define_ddc_registers }; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dce80/hw_factory_dce80.c b/drivers/gpu/drm/amd/display/dc/gpio/dce80/hw_factory_dce80.c index 48b67866377e..496d3ffb74bb 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dce80/hw_factory_dce80.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dce80/hw_factory_dce80.c @@ -32,10 +32,12 @@ #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" #include "dce/dce_8_0_d.h" #include "dce/dce_8_0_sh_mask.h" + #define REG(reg_name)\ mm ## reg_name @@ -147,12 +149,12 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) } static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = NULL, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = NULL, .define_hpd_registers = define_hpd_registers, .define_ddc_registers = define_ddc_registers }; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c index 32aa47a04a0d..b38c96c9fed3 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn10/hw_factory_dcn10.c @@ -31,6 +31,7 @@ #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" #include "hw_factory_dcn10.h" @@ -121,6 +122,42 @@ static const struct ddc_sh_mask ddc_mask = { DDC_MASK_SH_LIST(_MASK) }; +#include "../generic_regs.h" + +/* set field name */ +#define SF_GENERIC(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +#define generic_regs(id) \ +{\ + GENERIC_REG_LIST(id)\ +} + +static const struct generic_registers generic_regs[] = { + generic_regs(A), + generic_regs(B), +}; + +static const struct generic_sh_mask generic_shift[] = { + GENERIC_MASK_SH_LIST(__SHIFT, A), + GENERIC_MASK_SH_LIST(__SHIFT, B), +}; + +static const struct generic_sh_mask generic_mask[] = { + GENERIC_MASK_SH_LIST(_MASK, A), + GENERIC_MASK_SH_LIST(_MASK, B), +}; + +static void define_generic_registers(struct hw_gpio_pin *pin, uint32_t en) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(pin); + + generic->regs = &generic_regs[en]; + generic->shifts = &generic_shift[en]; + generic->masks = &generic_mask[en]; + generic->base.regs = &generic_regs[en].gpio; +} + static void define_ddc_registers( struct hw_gpio_pin *pin, uint32_t en) @@ -159,14 +196,15 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) /* fucntion table */ static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = dal_hw_generic_init, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = dal_hw_generic_get_pin, .define_hpd_registers = define_hpd_registers, - .define_ddc_registers = define_ddc_registers + .define_ddc_registers = define_ddc_registers, + .define_generic_registers = define_generic_registers }; /* * dal_hw_factory_dcn10_init diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c index abd76d855375..83f798cb8b21 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.c @@ -22,7 +22,6 @@ * Authors: AMD * */ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dm_services.h" #include "include/gpio_types.h" #include "../hw_factory.h" @@ -31,6 +30,7 @@ #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" #include "hw_factory_dcn20.h" @@ -109,6 +109,12 @@ static const struct ddc_registers ddc_data_regs_dcn[] = { ddc_data_regs_dcn2(4), ddc_data_regs_dcn2(5), ddc_data_regs_dcn2(6), + { + DDC_GPIO_VGA_REG_LIST(DATA), + .ddc_setup = 0, + .phy_aux_cntl = 0, + .dc_gpio_aux_ctrl_5 = 0 + } }; static const struct ddc_registers ddc_clk_regs_dcn[] = { @@ -118,6 +124,12 @@ static const struct ddc_registers ddc_clk_regs_dcn[] = { ddc_clk_regs_dcn2(4), ddc_clk_regs_dcn2(5), ddc_clk_regs_dcn2(6), + { + DDC_GPIO_VGA_REG_LIST(CLK), + .ddc_setup = 0, + .phy_aux_cntl = 0, + .dc_gpio_aux_ctrl_5 = 0 + } }; static const struct ddc_sh_mask ddc_shift[] = { @@ -138,6 +150,32 @@ static const struct ddc_sh_mask ddc_mask[] = { DDC_MASK_SH_LIST_DCN2(_MASK, 6) }; +#include "../generic_regs.h" + +/* set field name */ +#define SF_GENERIC(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +#define generic_regs(id) \ +{\ + GENERIC_REG_LIST(id)\ +} + +static const struct generic_registers generic_regs[] = { + generic_regs(A), + generic_regs(B), +}; + +static const struct generic_sh_mask generic_shift[] = { + GENERIC_MASK_SH_LIST(__SHIFT, A), + GENERIC_MASK_SH_LIST(__SHIFT, B), +}; + +static const struct generic_sh_mask generic_mask[] = { + GENERIC_MASK_SH_LIST(_MASK, A), + GENERIC_MASK_SH_LIST(_MASK, B), +}; + static void define_ddc_registers( struct hw_gpio_pin *pin, uint32_t en) @@ -173,17 +211,27 @@ static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) hpd->base.regs = &hpd_regs[en].gpio; } +static void define_generic_registers(struct hw_gpio_pin *pin, uint32_t en) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(pin); + + generic->regs = &generic_regs[en]; + generic->shifts = &generic_shift[en]; + generic->masks = &generic_mask[en]; + generic->base.regs = &generic_regs[en].gpio; +} /* fucntion table */ static const struct hw_factory_funcs funcs = { - .create_ddc_data = dal_hw_ddc_create, - .create_ddc_clock = dal_hw_ddc_create, - .create_generic = NULL, - .create_hpd = dal_hw_hpd_create, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = dal_hw_ddc_init, + .init_generic = dal_hw_generic_init, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = dal_hw_generic_get_pin, .define_hpd_registers = define_hpd_registers, - .define_ddc_registers = define_ddc_registers + .define_ddc_registers = define_ddc_registers, + .define_generic_registers = define_generic_registers, }; /* * dal_hw_factory_dcn10_init @@ -209,4 +257,3 @@ void dal_hw_factory_dcn20_init(struct hw_factory *factory) factory->funcs = &funcs; } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.h b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.h index 43a4ce7aa3bf..0fd9b315bd7a 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_factory_dcn20.h @@ -22,7 +22,6 @@ * Authors: AMD * */ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #ifndef __DAL_HW_FACTORY_DCN20_H__ #define __DAL_HW_FACTORY_DCN20_H__ @@ -30,4 +29,3 @@ void dal_hw_factory_dcn20_init(struct hw_factory *factory); #endif /* __DAL_HW_FACTORY_DCN20_H__ */ -#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.c index b393cc13298a..52ba62b3b5e4 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.c @@ -26,7 +26,6 @@ /* * Pre-requisites: headers required by header of this unit */ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "hw_translate_dcn20.h" #include "dm_services.h" @@ -71,7 +70,7 @@ static bool offset_to_id( { switch (offset) { /* GENERIC */ - case REG(DC_GENERICA): + case REG(DC_GPIO_GENERIC_A): *id = GPIO_ID_GENERIC; switch (mask) { case DC_GPIO_GENERIC_A__DC_GPIO_GENERICA_A_MASK: @@ -379,4 +378,3 @@ void dal_hw_translate_dcn20_init(struct hw_translate *tr) tr->funcs = &funcs; } -#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.h b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.h index 01f52c7bed86..5f7a35530e26 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn20/hw_translate_dcn20.h @@ -22,7 +22,6 @@ * Authors: AMD * */ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #ifndef __DAL_HW_TRANSLATE_DCN20_H__ #define __DAL_HW_TRANSLATE_DCN20_H__ @@ -32,4 +31,3 @@ struct hw_translate; void dal_hw_translate_dcn20_init(struct hw_translate *tr); #endif /* __DAL_HW_TRANSLATE_DCN20_H__ */ -#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c new file mode 100644 index 000000000000..907c5911eb9e --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.c @@ -0,0 +1,240 @@ +/* + * Copyright 2013-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "dm_services.h" +#include "include/gpio_types.h" +#include "../hw_factory.h" + + +#include "../hw_gpio.h" +#include "../hw_ddc.h" +#include "../hw_hpd.h" +#include "../hw_generic.h" + +#include "hw_factory_dcn21.h" + +#include "dcn/dcn_2_1_0_offset.h" +#include "dcn/dcn_2_1_0_sh_mask.h" +#include "renoir_ip_offset.h" + +#include "reg_helper.h" +#include "../hpd_regs.h" +/* begin ********************* + * macros to expend register list macro defined in HW object header file */ + +/* DCN */ +#define block HPD +#define reg_num 0 + +#undef BASE_INNER +#define BASE_INNER(seg) DMU_BASE__INST0_SEG ## seg + +#define BASE(seg) BASE_INNER(seg) + + + +#define REG(reg_name)\ + BASE(mm ## reg_name ## _BASE_IDX) + mm ## reg_name + +#define SF_HPD(reg_name, field_name, post_fix)\ + .field_name = HPD0_ ## reg_name ## __ ## field_name ## post_fix + +#define REGI(reg_name, block, id)\ + BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + +#define SF(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +/* macros to expend register list macro defined in HW object header file + * end *********************/ + + + +#define hpd_regs(id) \ +{\ + HPD_REG_LIST(id)\ +} + +static const struct hpd_registers hpd_regs[] = { + hpd_regs(0), + hpd_regs(1), + hpd_regs(2), + hpd_regs(3), + hpd_regs(4), +}; + +static const struct hpd_sh_mask hpd_shift = { + HPD_MASK_SH_LIST(__SHIFT) +}; + +static const struct hpd_sh_mask hpd_mask = { + HPD_MASK_SH_LIST(_MASK) +}; + +#include "../ddc_regs.h" + + /* set field name */ +#define SF_DDC(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +static const struct ddc_registers ddc_data_regs_dcn[] = { + ddc_data_regs_dcn2(1), + ddc_data_regs_dcn2(2), + ddc_data_regs_dcn2(3), + ddc_data_regs_dcn2(4), + ddc_data_regs_dcn2(5), +}; + +static const struct ddc_registers ddc_clk_regs_dcn[] = { + ddc_clk_regs_dcn2(1), + ddc_clk_regs_dcn2(2), + ddc_clk_regs_dcn2(3), + ddc_clk_regs_dcn2(4), + ddc_clk_regs_dcn2(5), +}; + +static const struct ddc_sh_mask ddc_shift[] = { + DDC_MASK_SH_LIST_DCN2(__SHIFT, 1), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 2), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 3), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 4), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 5), + DDC_MASK_SH_LIST_DCN2(__SHIFT, 6) +}; + +static const struct ddc_sh_mask ddc_mask[] = { + DDC_MASK_SH_LIST_DCN2(_MASK, 1), + DDC_MASK_SH_LIST_DCN2(_MASK, 2), + DDC_MASK_SH_LIST_DCN2(_MASK, 3), + DDC_MASK_SH_LIST_DCN2(_MASK, 4), + DDC_MASK_SH_LIST_DCN2(_MASK, 5), + DDC_MASK_SH_LIST_DCN2(_MASK, 6) +}; + +#include "../generic_regs.h" + +/* set field name */ +#define SF_GENERIC(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +#define generic_regs(id) \ +{\ + GENERIC_REG_LIST(id)\ +} + +static const struct generic_registers generic_regs[] = { + generic_regs(A), +}; + +static const struct generic_sh_mask generic_shift[] = { + GENERIC_MASK_SH_LIST(__SHIFT, A), +}; + +static const struct generic_sh_mask generic_mask[] = { + GENERIC_MASK_SH_LIST(_MASK, A), +}; + +static void define_generic_registers(struct hw_gpio_pin *pin, uint32_t en) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(pin); + + generic->regs = &generic_regs[en]; + generic->shifts = &generic_shift[en]; + generic->masks = &generic_mask[en]; + generic->base.regs = &generic_regs[en].gpio; +} + +static void define_ddc_registers( + struct hw_gpio_pin *pin, + uint32_t en) +{ + struct hw_ddc *ddc = HW_DDC_FROM_BASE(pin); + + switch (pin->id) { + case GPIO_ID_DDC_DATA: + ddc->regs = &ddc_data_regs_dcn[en]; + ddc->base.regs = &ddc_data_regs_dcn[en].gpio; + break; + case GPIO_ID_DDC_CLOCK: + ddc->regs = &ddc_clk_regs_dcn[en]; + ddc->base.regs = &ddc_clk_regs_dcn[en].gpio; + break; + default: + ASSERT_CRITICAL(false); + return; + } + + ddc->shifts = &ddc_shift[en]; + ddc->masks = &ddc_mask[en]; + +} + +static void define_hpd_registers(struct hw_gpio_pin *pin, uint32_t en) +{ + struct hw_hpd *hpd = HW_HPD_FROM_BASE(pin); + + hpd->regs = &hpd_regs[en]; + hpd->shifts = &hpd_shift; + hpd->masks = &hpd_mask; + hpd->base.regs = &hpd_regs[en].gpio; +} + + +/* fucntion table */ +static const struct hw_factory_funcs funcs = { + .init_ddc_data = dal_hw_ddc_init, + .init_generic = dal_hw_generic_init, + .init_hpd = dal_hw_hpd_init, + .get_ddc_pin = dal_hw_ddc_get_pin, + .get_hpd_pin = dal_hw_hpd_get_pin, + .get_generic_pin = dal_hw_generic_get_pin, + .define_hpd_registers = define_hpd_registers, + .define_ddc_registers = define_ddc_registers, + .define_generic_registers = define_generic_registers +}; +/* + * dal_hw_factory_dcn10_init + * + * @brief + * Initialize HW factory function pointers and pin info + * + * @param + * struct hw_factory *factory - [out] struct of function pointers + */ +void dal_hw_factory_dcn21_init(struct hw_factory *factory) +{ + /*TODO check ASIC CAPs*/ + factory->number_of_pins[GPIO_ID_DDC_DATA] = 8; + factory->number_of_pins[GPIO_ID_DDC_CLOCK] = 8; + factory->number_of_pins[GPIO_ID_GENERIC] = 4; + factory->number_of_pins[GPIO_ID_HPD] = 6; + factory->number_of_pins[GPIO_ID_GPIO_PAD] = 28; + factory->number_of_pins[GPIO_ID_VIP_PAD] = 0; + factory->number_of_pins[GPIO_ID_SYNC] = 0; + factory->number_of_pins[GPIO_ID_GSL] = 0;/*add this*/ + + factory->funcs = &funcs; +} + diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.h b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.h new file mode 100644 index 000000000000..4949e0c7fa06 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_factory_dcn21.h @@ -0,0 +1,31 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __DAL_HW_FACTORY_DCN21_H__ +#define __DAL_HW_FACTORY_DCN21_H__ + +/* Initialize HW factory function pointers and pin info */ +void dal_hw_factory_dcn21_init(struct hw_factory *factory); + +#endif /* __DAL_HW_FACTORY_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.c new file mode 100644 index 000000000000..291966efe63d --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.c @@ -0,0 +1,383 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +/* + * Pre-requisites: headers required by header of this unit + */ +#include "hw_translate_dcn21.h" + +#include "dm_services.h" +#include "include/gpio_types.h" +#include "../hw_translate.h" + +#include "dcn/dcn_2_1_0_offset.h" +#include "dcn/dcn_2_1_0_sh_mask.h" +#include "renoir_ip_offset.h" + + + + +/* begin ********************* + * macros to expend register list macro defined in HW object header file */ + +/* DCN */ +#define block HPD +#define reg_num 0 + +#undef BASE_INNER +#define BASE_INNER(seg) DMU_BASE__INST0_SEG ## seg + +#define BASE(seg) BASE_INNER(seg) + +#undef REG +#define REG(reg_name)\ + BASE(mm ## reg_name ## _BASE_IDX) + mm ## reg_name +#define SF_HPD(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +/* macros to expend register list macro defined in HW object header file + * end *********************/ + + +static bool offset_to_id( + uint32_t offset, + uint32_t mask, + enum gpio_id *id, + uint32_t *en) +{ + switch (offset) { + /* GENERIC */ + case REG(DC_GPIO_GENERIC_A): + *id = GPIO_ID_GENERIC; + switch (mask) { + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICA_A_MASK: + *en = GPIO_GENERIC_A; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICB_A_MASK: + *en = GPIO_GENERIC_B; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICC_A_MASK: + *en = GPIO_GENERIC_C; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICD_A_MASK: + *en = GPIO_GENERIC_D; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICE_A_MASK: + *en = GPIO_GENERIC_E; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICF_A_MASK: + *en = GPIO_GENERIC_F; + return true; + case DC_GPIO_GENERIC_A__DC_GPIO_GENERICG_A_MASK: + *en = GPIO_GENERIC_G; + return true; + default: + ASSERT_CRITICAL(false); +#ifdef PALLADIUM_SUPPORTED + *en = GPIO_DDC_LINE_DDC1; + return true; +#endif + return false; + } + break; + /* HPD */ + case REG(DC_GPIO_HPD_A): + *id = GPIO_ID_HPD; + switch (mask) { + case DC_GPIO_HPD_A__DC_GPIO_HPD1_A_MASK: + *en = GPIO_HPD_1; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD2_A_MASK: + *en = GPIO_HPD_2; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD3_A_MASK: + *en = GPIO_HPD_3; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD4_A_MASK: + *en = GPIO_HPD_4; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD5_A_MASK: + *en = GPIO_HPD_5; + return true; + case DC_GPIO_HPD_A__DC_GPIO_HPD6_A_MASK: + *en = GPIO_HPD_6; + return true; + default: + ASSERT_CRITICAL(false); + return false; + } + break; + /* REG(DC_GPIO_GENLK_MASK */ + case REG(DC_GPIO_GENLK_A): + *id = GPIO_ID_GSL; + switch (mask) { + case DC_GPIO_GENLK_A__DC_GPIO_GENLK_CLK_A_MASK: + *en = GPIO_GSL_GENLOCK_CLOCK; + return true; + case DC_GPIO_GENLK_A__DC_GPIO_GENLK_VSYNC_A_MASK: + *en = GPIO_GSL_GENLOCK_VSYNC; + return true; + case DC_GPIO_GENLK_A__DC_GPIO_SWAPLOCK_A_A_MASK: + *en = GPIO_GSL_SWAPLOCK_A; + return true; + case DC_GPIO_GENLK_A__DC_GPIO_SWAPLOCK_B_A_MASK: + *en = GPIO_GSL_SWAPLOCK_B; + return true; + default: + ASSERT_CRITICAL(false); + return false; + } + break; + /* DDC */ + /* we don't care about the GPIO_ID for DDC + * in DdcHandle it will use GPIO_ID_DDC_DATA/GPIO_ID_DDC_CLOCK + * directly in the create method */ + case REG(DC_GPIO_DDC1_A): + *en = GPIO_DDC_LINE_DDC1; + return true; + case REG(DC_GPIO_DDC2_A): + *en = GPIO_DDC_LINE_DDC2; + return true; + case REG(DC_GPIO_DDC3_A): + *en = GPIO_DDC_LINE_DDC3; + return true; + case REG(DC_GPIO_DDC4_A): + *en = GPIO_DDC_LINE_DDC4; + return true; + case REG(DC_GPIO_DDC5_A): + *en = GPIO_DDC_LINE_DDC5; + return true; + case REG(DC_GPIO_DDCVGA_A): + *en = GPIO_DDC_LINE_DDC_VGA; + return true; + +// case REG(DC_GPIO_I2CPAD_A): not exit +// case REG(DC_GPIO_PWRSEQ_A): +// case REG(DC_GPIO_PAD_STRENGTH_1): +// case REG(DC_GPIO_PAD_STRENGTH_2): +// case REG(DC_GPIO_DEBUG): + /* UNEXPECTED */ + default: +// case REG(DC_GPIO_SYNCA_A): not exist +#ifdef PALLADIUM_SUPPORTED + *id = GPIO_ID_HPD; + *en = GPIO_DDC_LINE_DDC1; + return true; +#endif + ASSERT_CRITICAL(false); + return false; + } +} + +static bool id_to_offset( + enum gpio_id id, + uint32_t en, + struct gpio_pin_info *info) +{ + bool result = true; + + switch (id) { + case GPIO_ID_DDC_DATA: + info->mask = DC_GPIO_DDC5_A__DC_GPIO_DDC5DATA_A_MASK; + switch (en) { + case GPIO_DDC_LINE_DDC1: + info->offset = REG(DC_GPIO_DDC1_A); + break; + case GPIO_DDC_LINE_DDC2: + info->offset = REG(DC_GPIO_DDC2_A); + break; + case GPIO_DDC_LINE_DDC3: + info->offset = REG(DC_GPIO_DDC3_A); + break; + case GPIO_DDC_LINE_DDC4: + info->offset = REG(DC_GPIO_DDC4_A); + break; + case GPIO_DDC_LINE_DDC5: + info->offset = REG(DC_GPIO_DDC5_A); + break; + case GPIO_DDC_LINE_DDC_VGA: + info->offset = REG(DC_GPIO_DDCVGA_A); + break; + case GPIO_DDC_LINE_I2C_PAD: + default: + ASSERT_CRITICAL(false); + result = false; + } + break; + case GPIO_ID_DDC_CLOCK: + info->mask = DC_GPIO_DDC5_A__DC_GPIO_DDC5CLK_A_MASK; + switch (en) { + case GPIO_DDC_LINE_DDC1: + info->offset = REG(DC_GPIO_DDC1_A); + break; + case GPIO_DDC_LINE_DDC2: + info->offset = REG(DC_GPIO_DDC2_A); + break; + case GPIO_DDC_LINE_DDC3: + info->offset = REG(DC_GPIO_DDC3_A); + break; + case GPIO_DDC_LINE_DDC4: + info->offset = REG(DC_GPIO_DDC4_A); + break; + case GPIO_DDC_LINE_DDC5: + info->offset = REG(DC_GPIO_DDC5_A); + break; + case GPIO_DDC_LINE_DDC_VGA: + info->offset = REG(DC_GPIO_DDCVGA_A); + break; + case GPIO_DDC_LINE_I2C_PAD: + default: + ASSERT_CRITICAL(false); + result = false; + } + break; + case GPIO_ID_GENERIC: + info->offset = REG(DC_GPIO_GENERIC_A); + switch (en) { + case GPIO_GENERIC_A: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICA_A_MASK; + break; + case GPIO_GENERIC_B: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICB_A_MASK; + break; + case GPIO_GENERIC_C: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICC_A_MASK; + break; + case GPIO_GENERIC_D: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICD_A_MASK; + break; + case GPIO_GENERIC_E: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICE_A_MASK; + break; + case GPIO_GENERIC_F: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICF_A_MASK; + break; + case GPIO_GENERIC_G: + info->mask = DC_GPIO_GENERIC_A__DC_GPIO_GENERICG_A_MASK; + break; + default: + ASSERT_CRITICAL(false); + result = false; + } + break; + case GPIO_ID_HPD: + info->offset = REG(DC_GPIO_HPD_A); + switch (en) { + case GPIO_HPD_1: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD1_A_MASK; + break; + case GPIO_HPD_2: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD2_A_MASK; + break; + case GPIO_HPD_3: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD3_A_MASK; + break; + case GPIO_HPD_4: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD4_A_MASK; + break; + case GPIO_HPD_5: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD5_A_MASK; + break; + case GPIO_HPD_6: + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD6_A_MASK; + break; + default: + ASSERT_CRITICAL(false); +#ifdef PALLADIUM_SUPPORTED + info->mask = DC_GPIO_HPD_A__DC_GPIO_HPD1_A_MASK; + result = true; +#endif + result = false; + } + break; + case GPIO_ID_GSL: + switch (en) { + case GPIO_GSL_GENLOCK_CLOCK: + /*not implmented*/ + ASSERT_CRITICAL(false); + result = false; + break; + case GPIO_GSL_GENLOCK_VSYNC: + /*not implmented*/ + ASSERT_CRITICAL(false); + result = false; + break; + case GPIO_GSL_SWAPLOCK_A: + /*not implmented*/ + ASSERT_CRITICAL(false); + result = false; + break; + case GPIO_GSL_SWAPLOCK_B: + /*not implmented*/ + ASSERT_CRITICAL(false); + result = false; + + break; + default: + ASSERT_CRITICAL(false); + result = false; + } + break; + case GPIO_ID_SYNC: + case GPIO_ID_VIP_PAD: + default: + ASSERT_CRITICAL(false); + result = false; + } + + if (result) { + info->offset_y = info->offset + 2; + info->offset_en = info->offset + 1; + info->offset_mask = info->offset - 1; + + info->mask_y = info->mask; + info->mask_en = info->mask; + info->mask_mask = info->mask; + } + + return result; +} + +/* function table */ +static const struct hw_translate_funcs funcs = { + .offset_to_id = offset_to_id, + .id_to_offset = id_to_offset, +}; + +/* + * dal_hw_translate_dcn10_init + * + * @brief + * Initialize Hw translate function pointers. + * + * @param + * struct hw_translate *tr - [out] struct of function pointers + * + */ +void dal_hw_translate_dcn21_init(struct hw_translate *tr) +{ + tr->funcs = &funcs; +} + diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.h b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.h new file mode 100644 index 000000000000..9462b0a65200 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn21/hw_translate_dcn21.h @@ -0,0 +1,33 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __DAL_HW_TRANSLATE_DCN21_H__ +#define __DAL_HW_TRANSLATE_DCN21_H__ + +struct hw_translate; + +/* Initialize Hw translate function pointers */ +void dal_hw_translate_dcn21_init(struct hw_translate *tr); + +#endif /* __DAL_HW_TRANSLATE_DCN21_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/gpio/ddc_regs.h b/drivers/gpu/drm/amd/display/dc/gpio/ddc_regs.h index f91e85b04956..308a543178a5 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/ddc_regs.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/ddc_regs.h @@ -48,13 +48,11 @@ DDC_GPIO_REG_LIST(cd,id),\ .ddc_setup = REG(DC_I2C_DDC ## id ## _SETUP) -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define DDC_REG_LIST_DCN2(cd, id) \ DDC_GPIO_REG_LIST(cd, id),\ .ddc_setup = REG(DC_I2C_DDC ## id ## _SETUP),\ .phy_aux_cntl = REG(PHY_AUX_CNTL), \ .dc_gpio_aux_ctrl_5 = REG(DC_GPIO_AUX_CTRL_5) -#endif #define DDC_GPIO_VGA_REG_LIST_ENTRY(type,cd)\ .type ## _reg = REG(DC_GPIO_DDCVGA_ ## type),\ @@ -90,13 +88,11 @@ DDC_GPIO_I2C_REG_LIST(cd),\ .ddc_setup = 0 -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define DDC_I2C_REG_LIST_DCN2(cd) \ DDC_GPIO_I2C_REG_LIST(cd),\ .ddc_setup = 0,\ .phy_aux_cntl = REG(PHY_AUX_CNTL), \ .dc_gpio_aux_ctrl_5 = REG(DC_GPIO_AUX_CTRL_5) -#endif #define DDC_MASK_SH_LIST_COMMON(mask_sh) \ SF_DDC(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_ENABLE, mask_sh),\ SF_DDC(DC_I2C_DDC1_SETUP, DC_I2C_DDC1_EDID_DETECT_ENABLE, mask_sh),\ @@ -110,22 +106,18 @@ SF_DDC(DC_GPIO_I2CPAD_MASK, DC_GPIO_SDA_PD_DIS, mask_sh),\ SF_DDC(DC_GPIO_I2CPAD_MASK, DC_GPIO_SCL_PD_DIS, mask_sh) -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define DDC_MASK_SH_LIST_DCN2(mask_sh, cd) \ {DDC_MASK_SH_LIST_COMMON(mask_sh),\ 0,\ 0,\ (PHY_AUX_CNTL__AUX## cd ##_PAD_RXSEL## mask_sh),\ (DC_GPIO_AUX_CTRL_5__DDC_PAD## cd ##_I2CMODE## mask_sh)} -#endif struct ddc_registers { struct gpio_registers gpio; uint32_t ddc_setup; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) uint32_t phy_aux_cntl; uint32_t dc_gpio_aux_ctrl_5; -#endif }; struct ddc_sh_mask { @@ -140,11 +132,9 @@ struct ddc_sh_mask { /* i2cpad_mask */ uint32_t DC_GPIO_SDA_PD_DIS; uint32_t DC_GPIO_SCL_PD_DIS; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) //phy_aux_cntl uint32_t AUX_PAD_RXSEL; uint32_t DDC_PAD_I2CMODE; -#endif }; @@ -180,7 +170,6 @@ struct ddc_sh_mask { {\ DDC_I2C_REG_LIST(SCL)\ } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define ddc_data_regs_dcn2(id) \ {\ DDC_REG_LIST_DCN2(DATA, id)\ @@ -200,7 +189,6 @@ struct ddc_sh_mask { {\ DDC_REG_LIST_DCN2(SCL)\ } -#endif #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_GPIO_DDC_REGS_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/gpio/diagnostics/hw_factory_diag.c b/drivers/gpu/drm/amd/display/dc/gpio/diagnostics/hw_factory_diag.c index 26695b963c58..df68430aeb0c 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/diagnostics/hw_factory_diag.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/diagnostics/hw_factory_diag.c @@ -38,15 +38,13 @@ #include "../hw_gpio.h" #include "../hw_ddc.h" #include "../hw_hpd.h" +#include "../hw_generic.h" /* function table */ static const struct hw_factory_funcs funcs = { - .create_ddc_data = NULL, - .create_ddc_clock = NULL, - .create_generic = NULL, - .create_hpd = NULL, - .create_sync = NULL, - .create_gsl = NULL, + .init_ddc_data = NULL, + .init_generic = NULL, + .init_hpd = NULL, }; void dal_hw_factory_diag_fpga_init(struct hw_factory *factory) diff --git a/drivers/gpu/drm/amd/display/dc/gpio/generic_regs.h b/drivers/gpu/drm/amd/display/dc/gpio/generic_regs.h new file mode 100644 index 000000000000..8c05295c05c2 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/generic_regs.h @@ -0,0 +1,66 @@ +/* + * Copyright 2012-16 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef DRIVERS_GPU_DRM_AMD_DC_DEV_DC_GPIO_GENERIC_REGS_H_ +#define DRIVERS_GPU_DRM_AMD_DC_DEV_DC_GPIO_GENERIC_REGS_H_ + +#include "gpio_regs.h" + +#define GENERIC_GPIO_REG_LIST_ENTRY(type, cd, id) \ + .type ## _reg = REG(DC_GPIO_GENERIC_## type),\ + .type ## _mask = DC_GPIO_GENERIC_ ## type ## __DC_GPIO_GENERIC ## id ## _ ## type ## _MASK,\ + .type ## _shift = DC_GPIO_GENERIC_ ## type ## __DC_GPIO_GENERIC ## id ## _ ## type ## __SHIFT + +#define GENERIC_GPIO_REG_LIST(id) \ + {\ + GENERIC_GPIO_REG_LIST_ENTRY(MASK, cd, id),\ + GENERIC_GPIO_REG_LIST_ENTRY(A, cd, id),\ + GENERIC_GPIO_REG_LIST_ENTRY(EN, cd, id),\ + GENERIC_GPIO_REG_LIST_ENTRY(Y, cd, id)\ + } + +#define GENERIC_REG_LIST(id) \ + GENERIC_GPIO_REG_LIST(id), \ + .mux = REG(DC_GENERIC ## id),\ + +#define GENERIC_MASK_SH_LIST(mask_sh, cd) \ + {(DC_GENERIC ## cd ##__GENERIC ## cd ##_EN## mask_sh),\ + (DC_GENERIC ## cd ##__GENERIC ## cd ##_SEL## mask_sh)} + +struct generic_registers { + struct gpio_registers gpio; + uint32_t mux; +}; + +struct generic_sh_mask { + /* enable */ + uint32_t GENERIC_EN; + /* select */ + uint32_t GENERIC_SEL; + +}; + + +#endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_GPIO_GENERIC_REGS_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c index d03b38e80d9b..f67c18375bfd 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c @@ -67,10 +67,14 @@ enum gpio_result dal_gpio_open_ex( return GPIO_RESULT_ALREADY_OPENED; } + // No action if allocation failed during gpio construct + if (!gpio->hw_container.ddc) { + ASSERT_CRITICAL(false); + return GPIO_RESULT_NON_SPECIFIC_ERROR; + } gpio->mode = mode; - return dal_gpio_service_open( - gpio->service, gpio->id, gpio->en, mode, &gpio->pin); + return dal_gpio_service_open(gpio); } enum gpio_result dal_gpio_get_value( @@ -231,6 +235,21 @@ enum gpio_pin_output_state dal_gpio_get_output_state( return gpio->output_state; } +struct hw_ddc *dal_gpio_get_ddc(struct gpio *gpio) +{ + return gpio->hw_container.ddc; +} + +struct hw_hpd *dal_gpio_get_hpd(struct gpio *gpio) +{ + return gpio->hw_container.hpd; +} + +struct hw_generic *dal_gpio_get_generic(struct gpio *gpio) +{ + return gpio->hw_container.generic; +} + void dal_gpio_close( struct gpio *gpio) { @@ -267,6 +286,30 @@ struct gpio *dal_gpio_create( gpio->mode = GPIO_MODE_UNKNOWN; gpio->output_state = output_state; + //initialize hw_container union based on id + switch (gpio->id) { + case GPIO_ID_DDC_DATA: + gpio->service->factory.funcs->init_ddc_data(&gpio->hw_container.ddc, service->ctx, id, en); + break; + case GPIO_ID_DDC_CLOCK: + gpio->service->factory.funcs->init_ddc_data(&gpio->hw_container.ddc, service->ctx, id, en); + break; + case GPIO_ID_GENERIC: + gpio->service->factory.funcs->init_generic(&gpio->hw_container.generic, service->ctx, id, en); + break; + case GPIO_ID_HPD: + gpio->service->factory.funcs->init_hpd(&gpio->hw_container.hpd, service->ctx, id, en); + break; + // TODO: currently gpio for sync and gsl does not get created, might need it later + case GPIO_ID_SYNC: + break; + case GPIO_ID_GSL: + break; + default: + ASSERT_CRITICAL(false); + gpio->pin = NULL; + } + return gpio; } @@ -278,7 +321,32 @@ void dal_gpio_destroy( return; } - dal_gpio_close(*gpio); + switch ((*gpio)->id) { + case GPIO_ID_DDC_DATA: + kfree((*gpio)->hw_container.ddc); + (*gpio)->hw_container.ddc = NULL; + break; + case GPIO_ID_DDC_CLOCK: + //TODO: might want to change it to init_ddc_clock + kfree((*gpio)->hw_container.ddc); + (*gpio)->hw_container.ddc = NULL; + break; + case GPIO_ID_GENERIC: + kfree((*gpio)->hw_container.generic); + (*gpio)->hw_container.generic = NULL; + break; + case GPIO_ID_HPD: + kfree((*gpio)->hw_container.hpd); + (*gpio)->hw_container.hpd = NULL; + break; + // TODO: currently gpio for sync and gsl does not get created, might need it later + case GPIO_ID_SYNC: + break; + case GPIO_ID_GSL: + break; + default: + break; + } kfree(*gpio); diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c index a7fab44f66b6..92280cc05e2d 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c @@ -141,6 +141,57 @@ struct gpio *dal_gpio_service_create_irq( return dal_gpio_create_irq(service, id, en); } +struct gpio *dal_gpio_service_create_generic_mux( + struct gpio_service *service, + uint32_t offset, + uint32_t mask) +{ + enum gpio_id id; + uint32_t en; + struct gpio *generic; + + if (!service->translate.funcs->offset_to_id(offset, mask, &id, &en)) { + ASSERT_CRITICAL(false); + return NULL; + } + + generic = dal_gpio_create( + service, id, en, GPIO_PIN_OUTPUT_STATE_DEFAULT); + + return generic; +} + +void dal_gpio_destroy_generic_mux( + struct gpio **mux) +{ + if (!mux || !*mux) { + ASSERT_CRITICAL(false); + return; + } + + dal_gpio_destroy(mux); + kfree(*mux); + + *mux = NULL; +} + +struct gpio_pin_info dal_gpio_get_generic_pin_info( + struct gpio_service *service, + enum gpio_id id, + uint32_t en) +{ + struct gpio_pin_info pin; + + if (service->translate.funcs->id_to_offset) { + service->translate.funcs->id_to_offset(id, en, &pin); + } else { + pin.mask = 0xFFFFFFFF; + pin.offset = 0xFFFFFFFF; + } + + return pin; +} + void dal_gpio_service_destroy( struct gpio_service **ptr) { @@ -165,6 +216,21 @@ void dal_gpio_service_destroy( *ptr = NULL; } +enum gpio_result dal_mux_setup_config( + struct gpio *mux, + struct gpio_generic_mux_config *config) +{ + struct gpio_config_data config_data; + + if (!config) + return GPIO_RESULT_INVALID_DATA; + + config_data.config.generic_mux = *config; + config_data.type = GPIO_CONFIG_TYPE_GENERIC_MUX; + + return dal_gpio_set_config(mux, &config_data); +} + /* * @brief * Private API. @@ -223,13 +289,15 @@ enum gpio_result dal_gpio_service_unlock( } enum gpio_result dal_gpio_service_open( - struct gpio_service *service, - enum gpio_id id, - uint32_t en, - enum gpio_mode mode, - struct hw_gpio_pin **ptr) + struct gpio *gpio) { - struct hw_gpio_pin *pin; + struct gpio_service *service = gpio->service; + enum gpio_id id = gpio->id; + uint32_t en = gpio->en; + enum gpio_mode mode = gpio->mode; + + struct hw_gpio_pin **pin = &gpio->pin; + if (!service->busyness[id]) { ASSERT_CRITICAL(false); @@ -243,50 +311,43 @@ enum gpio_result dal_gpio_service_open( switch (id) { case GPIO_ID_DDC_DATA: - pin = service->factory.funcs->create_ddc_data( - service->ctx, id, en); - service->factory.funcs->define_ddc_registers(pin, en); + *pin = service->factory.funcs->get_ddc_pin(gpio); + service->factory.funcs->define_ddc_registers(*pin, en); break; case GPIO_ID_DDC_CLOCK: - pin = service->factory.funcs->create_ddc_clock( - service->ctx, id, en); - service->factory.funcs->define_ddc_registers(pin, en); + *pin = service->factory.funcs->get_ddc_pin(gpio); + service->factory.funcs->define_ddc_registers(*pin, en); break; case GPIO_ID_GENERIC: - pin = service->factory.funcs->create_generic( - service->ctx, id, en); + *pin = service->factory.funcs->get_generic_pin(gpio); + service->factory.funcs->define_generic_registers(*pin, en); break; case GPIO_ID_HPD: - pin = service->factory.funcs->create_hpd( - service->ctx, id, en); - service->factory.funcs->define_hpd_registers(pin, en); + *pin = service->factory.funcs->get_hpd_pin(gpio); + service->factory.funcs->define_hpd_registers(*pin, en); break; + + //TODO: gsl and sync support? create_sync and create_gsl are NULL case GPIO_ID_SYNC: - pin = service->factory.funcs->create_sync( - service->ctx, id, en); - break; case GPIO_ID_GSL: - pin = service->factory.funcs->create_gsl( - service->ctx, id, en); break; default: ASSERT_CRITICAL(false); return GPIO_RESULT_NON_SPECIFIC_ERROR; } - if (!pin) { + if (!*pin) { ASSERT_CRITICAL(false); return GPIO_RESULT_NON_SPECIFIC_ERROR; } - if (!pin->funcs->open(pin, mode)) { + if (!(*pin)->funcs->open(*pin, mode)) { ASSERT_CRITICAL(false); - dal_gpio_service_close(service, &pin); + dal_gpio_service_close(service, pin); return GPIO_RESULT_OPEN_FAILED; } set_pin_busy(service, id, en); - *ptr = pin; return GPIO_RESULT_OK; } @@ -308,11 +369,10 @@ void dal_gpio_service_close( pin->funcs->close(pin); - pin->funcs->destroy(ptr); + *ptr = NULL; } } - enum dc_irq_source dal_irq_get_source( const struct gpio *irq) { @@ -399,7 +459,6 @@ void dal_gpio_destroy_irq( return; } - dal_gpio_close(*irq); dal_gpio_destroy(irq); kfree(*irq); diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h index 0c678af75331..b9775a131ecd 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h @@ -42,11 +42,7 @@ struct gpio_service { }; enum gpio_result dal_gpio_service_open( - struct gpio_service *service, - enum gpio_id id, - uint32_t en, - enum gpio_mode mode, - struct hw_gpio_pin **ptr); + struct gpio *gpio); void dal_gpio_service_close( struct gpio_service *service, diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.c index 408857d19c84..1ae153eab31d 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.c @@ -28,6 +28,7 @@ #include "dm_services.h" +#include "include/gpio_interface.h" #include "include/gpio_types.h" #include "hw_gpio.h" #include "hw_ddc.h" @@ -45,18 +46,20 @@ #define REG(reg)\ (ddc->regs->reg) -static void destruct( +struct gpio; + +static void dal_hw_ddc_destruct( struct hw_ddc *pin) { dal_hw_gpio_destruct(&pin->base); } -static void destroy( +static void dal_hw_ddc_destroy( struct hw_gpio_pin **ptr) { struct hw_ddc *pin = HW_DDC_FROM_BASE(*ptr); - destruct(pin); + dal_hw_ddc_destruct(pin); kfree(pin); @@ -147,7 +150,6 @@ static enum gpio_result set_config( AUX_PAD1_MODE, 0); } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) if (ddc->regs->dc_gpio_aux_ctrl_5 != 0) { REG_UPDATE(dc_gpio_aux_ctrl_5, DDC_PAD_I2CMODE, 1); } @@ -155,7 +157,6 @@ static enum gpio_result set_config( if (ddc->regs->phy_aux_cntl != 0) { REG_UPDATE(phy_aux_cntl, AUX_PAD_RXSEL, 1); } -#endif return GPIO_RESULT_OK; case GPIO_DDC_CONFIG_TYPE_MODE_AUX: /* set the AUX pad mode */ @@ -163,12 +164,10 @@ static enum gpio_result set_config( REG_SET(gpio.MASK_reg, regval, AUX_PAD1_MODE, 1); } -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) if (ddc->regs->dc_gpio_aux_ctrl_5 != 0) { REG_UPDATE(dc_gpio_aux_ctrl_5, DDC_PAD_I2CMODE, 0); } -#endif return GPIO_RESULT_OK; case GPIO_DDC_CONFIG_TYPE_POLL_FOR_CONNECT: @@ -208,7 +207,7 @@ static enum gpio_result set_config( } static const struct hw_gpio_pin_funcs funcs = { - .destroy = destroy, + .destroy = dal_hw_ddc_destroy, .open = dal_hw_gpio_open, .get_value = dal_hw_gpio_get_value, .set_value = dal_hw_gpio_set_value, @@ -217,7 +216,7 @@ static const struct hw_gpio_pin_funcs funcs = { .close = dal_hw_gpio_close, }; -static void construct( +static void dal_hw_ddc_construct( struct hw_ddc *ddc, enum gpio_id id, uint32_t en, @@ -227,24 +226,29 @@ static void construct( ddc->base.base.funcs = &funcs; } -struct hw_gpio_pin *dal_hw_ddc_create( +void dal_hw_ddc_init( + struct hw_ddc **hw_ddc, struct dc_context *ctx, enum gpio_id id, uint32_t en) { - struct hw_ddc *pin; - if ((en < GPIO_DDC_LINE_MIN) || (en > GPIO_DDC_LINE_MAX)) { ASSERT_CRITICAL(false); - return NULL; + *hw_ddc = NULL; } - pin = kzalloc(sizeof(struct hw_ddc), GFP_KERNEL); - if (!pin) { + *hw_ddc = kzalloc(sizeof(struct hw_ddc), GFP_KERNEL); + if (!*hw_ddc) { ASSERT_CRITICAL(false); - return NULL; + return; } - construct(pin, id, en, ctx); - return &pin->base.base; + dal_hw_ddc_construct(*hw_ddc, id, en, ctx); +} + +struct hw_gpio_pin *dal_hw_ddc_get_pin(struct gpio *gpio) +{ + struct hw_ddc *hw_ddc = dal_gpio_get_ddc(gpio); + + return &hw_ddc->base.base; } diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.h b/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.h index 9690e2a885d7..cc30e65df431 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_ddc.h @@ -38,9 +38,12 @@ struct hw_ddc { #define HW_DDC_FROM_BASE(hw_gpio) \ container_of((HW_GPIO_FROM_BASE(hw_gpio)), struct hw_ddc, base) -struct hw_gpio_pin *dal_hw_ddc_create( +void dal_hw_ddc_init( + struct hw_ddc **hw_ddc, struct dc_context *ctx, enum gpio_id id, uint32_t en); +struct hw_gpio_pin *dal_hw_ddc_get_pin(struct gpio *gpio); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c index 78f528f92907..d2d36d48caaa 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c @@ -45,12 +45,11 @@ #include "dce80/hw_factory_dce80.h" #include "dce110/hw_factory_dce110.h" #include "dce120/hw_factory_dce120.h" -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) #include "dcn10/hw_factory_dcn10.h" #endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dcn20/hw_factory_dcn20.h" -#endif +#include "dcn21/hw_factory_dcn21.h" #include "diagnostics/hw_factory_diag.h" @@ -87,17 +86,18 @@ bool dal_hw_factory_init( case DCE_VERSION_12_1: dal_hw_factory_dce120_init(factory); return true; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) case DCN_VERSION_1_0: case DCN_VERSION_1_01: dal_hw_factory_dcn10_init(factory); return true; -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) case DCN_VERSION_2_0: dal_hw_factory_dcn20_init(factory); return true; + case DCN_VERSION_2_1: + dal_hw_factory_dcn21_init(factory); + return true; #endif default: diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h index 6e4dd3521935..e15b037f3bcd 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.h @@ -28,41 +28,44 @@ struct hw_gpio_pin; struct hw_hpd; +struct hw_ddc; +struct hw_generic; +struct gpio; struct hw_factory { uint32_t number_of_pins[GPIO_ID_COUNT]; const struct hw_factory_funcs { - struct hw_gpio_pin *(*create_ddc_data)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_ddc_clock)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_generic)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_hpd)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_sync)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); - struct hw_gpio_pin *(*create_gsl)( - struct dc_context *ctx, - enum gpio_id id, - uint32_t en); + void (*init_ddc_data)( + struct hw_ddc **hw_ddc, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en); + void (*init_generic)( + struct hw_generic **hw_generic, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en); + void (*init_hpd)( + struct hw_hpd **hw_hpd, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en); + struct hw_gpio_pin *(*get_hpd_pin)( + struct gpio *gpio); + struct hw_gpio_pin *(*get_ddc_pin)( + struct gpio *gpio); + struct hw_gpio_pin *(*get_generic_pin)( + struct gpio *gpio); void (*define_hpd_registers)( struct hw_gpio_pin *pin, uint32_t en); void (*define_ddc_registers)( struct hw_gpio_pin *pin, uint32_t en); + void (*define_generic_registers)( + struct hw_gpio_pin *pin, + uint32_t en); } *funcs; }; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.c new file mode 100644 index 000000000000..f9e847e6555d --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.c @@ -0,0 +1,129 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include <linux/slab.h> + +#include "dm_services.h" + +#include "include/gpio_interface.h" +#include "include/gpio_types.h" +#include "hw_gpio.h" +#include "hw_generic.h" + +#include "reg_helper.h" +#include "generic_regs.h" + +#undef FN +#define FN(reg_name, field_name) \ + generic->shifts->field_name, generic->masks->field_name + +#define CTX \ + generic->base.base.ctx +#define REG(reg)\ + (generic->regs->reg) + +struct gpio; + +static void dal_hw_generic_destruct( + struct hw_generic *pin) +{ + dal_hw_gpio_destruct(&pin->base); +} + +static void dal_hw_generic_destroy( + struct hw_gpio_pin **ptr) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(*ptr); + + dal_hw_generic_destruct(generic); + + kfree(generic); + + *ptr = NULL; +} + +static enum gpio_result set_config( + struct hw_gpio_pin *ptr, + const struct gpio_config_data *config_data) +{ + struct hw_generic *generic = HW_GENERIC_FROM_BASE(ptr); + + if (!config_data) + return GPIO_RESULT_INVALID_DATA; + + REG_UPDATE_2(mux, + GENERIC_EN, config_data->config.generic_mux.enable_output_from_mux, + GENERIC_SEL, config_data->config.generic_mux.mux_select); + + return GPIO_RESULT_OK; +} + +static const struct hw_gpio_pin_funcs funcs = { + .destroy = dal_hw_generic_destroy, + .open = dal_hw_gpio_open, + .get_value = dal_hw_gpio_get_value, + .set_value = dal_hw_gpio_set_value, + .set_config = set_config, + .change_mode = dal_hw_gpio_change_mode, + .close = dal_hw_gpio_close, +}; + +static void dal_hw_generic_construct( + struct hw_generic *pin, + enum gpio_id id, + uint32_t en, + struct dc_context *ctx) +{ + dal_hw_gpio_construct(&pin->base, id, en, ctx); + pin->base.base.funcs = &funcs; +} + +void dal_hw_generic_init( + struct hw_generic **hw_generic, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en) +{ + if ((en < GPIO_DDC_LINE_MIN) || (en > GPIO_DDC_LINE_MAX)) { + ASSERT_CRITICAL(false); + *hw_generic = NULL; + } + + *hw_generic = kzalloc(sizeof(struct hw_generic), GFP_KERNEL); + if (!*hw_generic) { + ASSERT_CRITICAL(false); + return; + } + + dal_hw_generic_construct(*hw_generic, id, en, ctx); +} + + +struct hw_gpio_pin *dal_hw_generic_get_pin(struct gpio *gpio) +{ + struct hw_generic *hw_generic = dal_gpio_get_generic(gpio); + + return &hw_generic->base.base; +} diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.h b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.h new file mode 100644 index 000000000000..bd6ffeb5e9df --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_generic.h @@ -0,0 +1,50 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DAL_HW_generic_H__ +#define __DAL_HW_generic_H__ + +#include "generic_regs.h" +#include "hw_gpio.h" + +struct hw_generic { + struct hw_gpio base; + const struct generic_registers *regs; + const struct generic_sh_mask *shifts; + const struct generic_sh_mask *masks; +}; + +#define HW_GENERIC_FROM_BASE(hw_gpio) \ + container_of((HW_GPIO_FROM_BASE(hw_gpio)), struct hw_generic, base) + +void dal_hw_generic_init( + struct hw_generic **hw_generic, + struct dc_context *ctx, + enum gpio_id id, + uint32_t en); + +struct hw_gpio_pin *dal_hw_generic_get_pin(struct gpio *gpio); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.c index 5e11d748e6f3..692f29de7797 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.c @@ -27,6 +27,7 @@ #include "dm_services.h" +#include "include/gpio_interface.h" #include "include/gpio_types.h" #include "hw_gpio.h" #include "hw_hpd.h" @@ -43,14 +44,7 @@ #define REG(reg)\ (hpd->regs->reg) -static void dal_hw_hpd_construct( - struct hw_hpd *pin, - enum gpio_id id, - uint32_t en, - struct dc_context *ctx) -{ - dal_hw_gpio_construct(&pin->base, id, en, ctx); -} +struct gpio; static void dal_hw_hpd_destruct( struct hw_hpd *pin) @@ -58,19 +52,12 @@ static void dal_hw_hpd_destruct( dal_hw_gpio_destruct(&pin->base); } - -static void destruct( - struct hw_hpd *hpd) -{ - dal_hw_hpd_destruct(hpd); -} - -static void destroy( +static void dal_hw_hpd_destroy( struct hw_gpio_pin **ptr) { struct hw_hpd *hpd = HW_HPD_FROM_BASE(*ptr); - destruct(hpd); + dal_hw_hpd_destruct(hpd); kfree(hpd); @@ -117,7 +104,7 @@ static enum gpio_result set_config( } static const struct hw_gpio_pin_funcs funcs = { - .destroy = destroy, + .destroy = dal_hw_hpd_destroy, .open = dal_hw_gpio_open, .get_value = get_value, .set_value = dal_hw_gpio_set_value, @@ -126,39 +113,39 @@ static const struct hw_gpio_pin_funcs funcs = { .close = dal_hw_gpio_close, }; -static void construct( - struct hw_hpd *hpd, +static void dal_hw_hpd_construct( + struct hw_hpd *pin, enum gpio_id id, uint32_t en, struct dc_context *ctx) { - dal_hw_hpd_construct(hpd, id, en, ctx); - hpd->base.base.funcs = &funcs; + dal_hw_gpio_construct(&pin->base, id, en, ctx); + pin->base.base.funcs = &funcs; } -struct hw_gpio_pin *dal_hw_hpd_create( +void dal_hw_hpd_init( + struct hw_hpd **hw_hpd, struct dc_context *ctx, enum gpio_id id, uint32_t en) { - struct hw_hpd *hpd; - - if (id != GPIO_ID_HPD) { + if ((en < GPIO_DDC_LINE_MIN) || (en > GPIO_DDC_LINE_MAX)) { ASSERT_CRITICAL(false); - return NULL; + *hw_hpd = NULL; } - if ((en < GPIO_HPD_MIN) || (en > GPIO_HPD_MAX)) { + *hw_hpd = kzalloc(sizeof(struct hw_hpd), GFP_KERNEL); + if (!*hw_hpd) { ASSERT_CRITICAL(false); - return NULL; + return; } - hpd = kzalloc(sizeof(struct hw_hpd), GFP_KERNEL); - if (!hpd) { - ASSERT_CRITICAL(false); - return NULL; - } + dal_hw_hpd_construct(*hw_hpd, id, en, ctx); +} + +struct hw_gpio_pin *dal_hw_hpd_get_pin(struct gpio *gpio) +{ + struct hw_hpd *hw_hpd = dal_gpio_get_hpd(gpio); - construct(hpd, id, en, ctx); - return &hpd->base.base; + return &hw_hpd->base.base; } diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.h b/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.h index 4ab7a208f781..e7d8b3bb016c 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_hpd.h @@ -38,9 +38,12 @@ struct hw_hpd { #define HW_HPD_FROM_BASE(hw_gpio) \ container_of((HW_GPIO_FROM_BASE(hw_gpio)), struct hw_hpd, base) -struct hw_gpio_pin *dal_hw_hpd_create( +void dal_hw_hpd_init( + struct hw_hpd **hw_hpd, struct dc_context *ctx, enum gpio_id id, uint32_t en); +struct hw_gpio_pin *dal_hw_hpd_get_pin(struct gpio *gpio); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c index c35fe201d335..5d396657a1ee 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c @@ -43,12 +43,11 @@ #include "dce80/hw_translate_dce80.h" #include "dce110/hw_translate_dce110.h" #include "dce120/hw_translate_dce120.h" -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) #include "dcn10/hw_translate_dcn10.h" #endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dcn20/hw_translate_dcn20.h" -#endif +#include "dcn21/hw_translate_dcn21.h" #include "diagnostics/hw_translate_diag.h" @@ -82,17 +81,18 @@ bool dal_hw_translate_init( case DCE_VERSION_12_1: dal_hw_translate_dce120_init(translate); return true; -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) case DCN_VERSION_1_0: case DCN_VERSION_1_01: dal_hw_translate_dcn10_init(translate); return true; -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) case DCN_VERSION_2_0: dal_hw_translate_dcn20_init(translate); return true; + case DCN_VERSION_2_1: + dal_hw_translate_dcn21_init(translate); + return true; #endif default: diff --git a/drivers/gpu/drm/amd/display/dc/hdcp/Makefile b/drivers/gpu/drm/amd/display/dc/hdcp/Makefile new file mode 100644 index 000000000000..4170b6eb9ec0 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/hdcp/Makefile @@ -0,0 +1,28 @@ +# Copyright 2019 Advanced Micro Devices, Inc. +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the "Software"), +# to deal in the Software without restriction, including without limitation +# the rights to use, copy, modify, merge, publish, distribute, sublicense, +# and/or sell copies of the Software, and to permit persons to whom the +# Software is furnished to do so, subject to the following conditions: +# +# The above copyright notice and this permission notice shall be included in +# all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR +# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +# OTHER DEALINGS IN THE SOFTWARE. +# +# Makefile for the 'hdcp' sub-component of DAL. +# + +HDCP_MSG = hdcp_msg.o + +AMD_DAL_HDCP_MSG = $(addprefix $(AMDDALPATH)/dc/hdcp/,$(HDCP_MSG)) + +AMD_DISPLAY_FILES += $(AMD_DAL_HDCP_MSG) diff --git a/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c b/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c new file mode 100644 index 000000000000..6f730b5bfe42 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c @@ -0,0 +1,324 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include <linux/slab.h> + +#include "dm_services.h" +#include "dm_helpers.h" +#include "include/hdcp_types.h" +#include "include/i2caux_interface.h" +#include "include/signal_types.h" +#include "core_types.h" +#include "dc_link_ddc.h" +#include "link_hwss.h" + +#define DC_LOGGER \ + link->ctx->logger +#define HDCP14_KSV_SIZE 5 +#define HDCP14_MAX_KSV_FIFO_SIZE 127*HDCP14_KSV_SIZE + +static const bool hdcp_cmd_is_read[] = { + [HDCP_MESSAGE_ID_READ_BKSV] = true, + [HDCP_MESSAGE_ID_READ_RI_R0] = true, + [HDCP_MESSAGE_ID_READ_PJ] = true, + [HDCP_MESSAGE_ID_WRITE_AKSV] = false, + [HDCP_MESSAGE_ID_WRITE_AINFO] = false, + [HDCP_MESSAGE_ID_WRITE_AN] = false, + [HDCP_MESSAGE_ID_READ_VH_X] = true, + [HDCP_MESSAGE_ID_READ_VH_0] = true, + [HDCP_MESSAGE_ID_READ_VH_1] = true, + [HDCP_MESSAGE_ID_READ_VH_2] = true, + [HDCP_MESSAGE_ID_READ_VH_3] = true, + [HDCP_MESSAGE_ID_READ_VH_4] = true, + [HDCP_MESSAGE_ID_READ_BCAPS] = true, + [HDCP_MESSAGE_ID_READ_BSTATUS] = true, + [HDCP_MESSAGE_ID_READ_KSV_FIFO] = true, + [HDCP_MESSAGE_ID_READ_BINFO] = true, + [HDCP_MESSAGE_ID_HDCP2VERSION] = true, + [HDCP_MESSAGE_ID_RX_CAPS] = true, + [HDCP_MESSAGE_ID_WRITE_AKE_INIT] = false, + [HDCP_MESSAGE_ID_READ_AKE_SEND_CERT] = true, + [HDCP_MESSAGE_ID_WRITE_AKE_NO_STORED_KM] = false, + [HDCP_MESSAGE_ID_WRITE_AKE_STORED_KM] = false, + [HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME] = true, + [HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO] = true, + [HDCP_MESSAGE_ID_WRITE_LC_INIT] = false, + [HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME] = true, + [HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS] = false, + [HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST] = true, + [HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK] = false, + [HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE] = false, + [HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY] = true, + [HDCP_MESSAGE_ID_READ_RXSTATUS] = true, + [HDCP_MESSAGE_ID_WRITE_CONTENT_STREAM_TYPE] = false +}; + +static const uint8_t hdcp_i2c_offsets[] = { + [HDCP_MESSAGE_ID_READ_BKSV] = 0x0, + [HDCP_MESSAGE_ID_READ_RI_R0] = 0x8, + [HDCP_MESSAGE_ID_READ_PJ] = 0xA, + [HDCP_MESSAGE_ID_WRITE_AKSV] = 0x10, + [HDCP_MESSAGE_ID_WRITE_AINFO] = 0x15, + [HDCP_MESSAGE_ID_WRITE_AN] = 0x18, + [HDCP_MESSAGE_ID_READ_VH_X] = 0x20, + [HDCP_MESSAGE_ID_READ_VH_0] = 0x20, + [HDCP_MESSAGE_ID_READ_VH_1] = 0x24, + [HDCP_MESSAGE_ID_READ_VH_2] = 0x28, + [HDCP_MESSAGE_ID_READ_VH_3] = 0x2C, + [HDCP_MESSAGE_ID_READ_VH_4] = 0x30, + [HDCP_MESSAGE_ID_READ_BCAPS] = 0x40, + [HDCP_MESSAGE_ID_READ_BSTATUS] = 0x41, + [HDCP_MESSAGE_ID_READ_KSV_FIFO] = 0x43, + [HDCP_MESSAGE_ID_READ_BINFO] = 0xFF, + [HDCP_MESSAGE_ID_HDCP2VERSION] = 0x50, + [HDCP_MESSAGE_ID_WRITE_AKE_INIT] = 0x60, + [HDCP_MESSAGE_ID_READ_AKE_SEND_CERT] = 0x80, + [HDCP_MESSAGE_ID_WRITE_AKE_NO_STORED_KM] = 0x60, + [HDCP_MESSAGE_ID_WRITE_AKE_STORED_KM] = 0x60, + [HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME] = 0x80, + [HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO] = 0x80, + [HDCP_MESSAGE_ID_WRITE_LC_INIT] = 0x60, + [HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME] = 0x80, + [HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS] = 0x60, + [HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST] = 0x80, + [HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK] = 0x60, + [HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE] = 0x60, + [HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY] = 0x80, + [HDCP_MESSAGE_ID_READ_RXSTATUS] = 0x70 +}; + +struct protection_properties { + bool supported; + bool (*process_transaction)( + struct dc_link *link, + struct hdcp_protection_message *message_info); +}; + +static const struct protection_properties non_supported_protection = { + .supported = false +}; + +static bool hdmi_14_process_transaction( + struct dc_link *link, + struct hdcp_protection_message *message_info) +{ + uint8_t *buff = NULL; + bool result; + const uint8_t hdcp_i2c_addr_link_primary = 0x3a; /* 0x74 >> 1*/ + const uint8_t hdcp_i2c_addr_link_secondary = 0x3b; /* 0x76 >> 1*/ + struct i2c_command i2c_command; + uint8_t offset = hdcp_i2c_offsets[message_info->msg_id]; + struct i2c_payload i2c_payloads[] = { + { true, 0, 1, &offset }, + /* actual hdcp payload, will be filled later, zeroed for now*/ + { 0 } + }; + + switch (message_info->link) { + case HDCP_LINK_SECONDARY: + i2c_payloads[0].address = hdcp_i2c_addr_link_secondary; + i2c_payloads[1].address = hdcp_i2c_addr_link_secondary; + break; + case HDCP_LINK_PRIMARY: + default: + i2c_payloads[0].address = hdcp_i2c_addr_link_primary; + i2c_payloads[1].address = hdcp_i2c_addr_link_primary; + break; + } + + if (hdcp_cmd_is_read[message_info->msg_id]) { + i2c_payloads[1].write = false; + i2c_command.number_of_payloads = ARRAY_SIZE(i2c_payloads); + i2c_payloads[1].length = message_info->length; + i2c_payloads[1].data = message_info->data; + } else { + i2c_command.number_of_payloads = 1; + buff = kzalloc(message_info->length + 1, GFP_KERNEL); + + if (!buff) + return false; + + buff[0] = offset; + memmove(&buff[1], message_info->data, message_info->length); + i2c_payloads[0].length = message_info->length + 1; + i2c_payloads[0].data = buff; + } + + i2c_command.payloads = i2c_payloads; + i2c_command.engine = I2C_COMMAND_ENGINE_HW;//only HW + i2c_command.speed = link->ddc->ctx->dc->caps.i2c_speed_in_khz; + + result = dm_helpers_submit_i2c( + link->ctx, + link, + &i2c_command); + kfree(buff); + + return result; +} + +static const struct protection_properties hdmi_14_protection = { + .supported = true, + .process_transaction = hdmi_14_process_transaction +}; + +static const uint32_t hdcp_dpcd_addrs[] = { + [HDCP_MESSAGE_ID_READ_BKSV] = 0x68000, + [HDCP_MESSAGE_ID_READ_RI_R0] = 0x68005, + [HDCP_MESSAGE_ID_READ_PJ] = 0xFFFFFFFF, + [HDCP_MESSAGE_ID_WRITE_AKSV] = 0x68007, + [HDCP_MESSAGE_ID_WRITE_AINFO] = 0x6803B, + [HDCP_MESSAGE_ID_WRITE_AN] = 0x6800c, + [HDCP_MESSAGE_ID_READ_VH_X] = 0x68014, + [HDCP_MESSAGE_ID_READ_VH_0] = 0x68014, + [HDCP_MESSAGE_ID_READ_VH_1] = 0x68018, + [HDCP_MESSAGE_ID_READ_VH_2] = 0x6801c, + [HDCP_MESSAGE_ID_READ_VH_3] = 0x68020, + [HDCP_MESSAGE_ID_READ_VH_4] = 0x68024, + [HDCP_MESSAGE_ID_READ_BCAPS] = 0x68028, + [HDCP_MESSAGE_ID_READ_BSTATUS] = 0x68029, + [HDCP_MESSAGE_ID_READ_KSV_FIFO] = 0x6802c, + [HDCP_MESSAGE_ID_READ_BINFO] = 0x6802a, + [HDCP_MESSAGE_ID_RX_CAPS] = 0x6921d, + [HDCP_MESSAGE_ID_WRITE_AKE_INIT] = 0x69000, + [HDCP_MESSAGE_ID_READ_AKE_SEND_CERT] = 0x6900b, + [HDCP_MESSAGE_ID_WRITE_AKE_NO_STORED_KM] = 0x69220, + [HDCP_MESSAGE_ID_WRITE_AKE_STORED_KM] = 0x692a0, + [HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME] = 0x692c0, + [HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO] = 0x692e0, + [HDCP_MESSAGE_ID_WRITE_LC_INIT] = 0x692f0, + [HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME] = 0x692f8, + [HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS] = 0x69318, + [HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST] = 0x69330, + [HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK] = 0x693e0, + [HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE] = 0x693f0, + [HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY] = 0x69473, + [HDCP_MESSAGE_ID_READ_RXSTATUS] = 0x69493, + [HDCP_MESSAGE_ID_WRITE_CONTENT_STREAM_TYPE] = 0x69494 +}; + +static bool dpcd_access_helper( + struct dc_link *link, + uint32_t length, + uint8_t *data, + uint32_t dpcd_addr, + bool is_read) +{ + enum dc_status status; + uint32_t cur_length = 0; + uint32_t offset = 0; + uint32_t ksv_read_size = 0x6803b - 0x6802c; + + /* Read KSV, need repeatedly handle */ + if (dpcd_addr == 0x6802c) { + if (length % HDCP14_KSV_SIZE) { + DC_LOG_ERROR("%s: KsvFifo Size(%d) is not a multiple of HDCP14_KSV_SIZE(%d)\n", + __func__, + length, + HDCP14_KSV_SIZE); + } + if (length > HDCP14_MAX_KSV_FIFO_SIZE) { + DC_LOG_ERROR("%s: KsvFifo Size(%d) is greater than HDCP14_MAX_KSV_FIFO_SIZE(%d)\n", + __func__, + length, + HDCP14_MAX_KSV_FIFO_SIZE); + } + + DC_LOG_ERROR("%s: Reading %d Ksv(s) from KsvFifo\n", + __func__, + length / HDCP14_KSV_SIZE); + + while (length > 0) { + if (length > ksv_read_size) { + status = core_link_read_dpcd( + link, + dpcd_addr + offset, + data + offset, + ksv_read_size); + + data += ksv_read_size; + length -= ksv_read_size; + } else { + status = core_link_read_dpcd( + link, + dpcd_addr + offset, + data + offset, + length); + + data += length; + length = 0; + } + + if (status != DC_OK) + return false; + } + } else { + while (length > 0) { + if (length > DEFAULT_AUX_MAX_DATA_SIZE) + cur_length = DEFAULT_AUX_MAX_DATA_SIZE; + else + cur_length = length; + + if (is_read) { + status = core_link_read_dpcd( + link, + dpcd_addr + offset, + data + offset, + cur_length); + } else { + status = core_link_write_dpcd( + link, + dpcd_addr + offset, + data + offset, + cur_length); + } + + if (status != DC_OK) + return false; + + length -= cur_length; + offset += cur_length; + } + } + return true; +} + +static bool dp_11_process_transaction( + struct dc_link *link, + struct hdcp_protection_message *message_info) +{ + return dpcd_access_helper( + link, + message_info->length, + message_info->data, + hdcp_dpcd_addrs[message_info->msg_id], + hdcp_cmd_is_read[message_info->msg_id]); +} + +static const struct protection_properties dp_11_protection = { + .supported = true, + .process_transaction = dp_11_process_transaction +}; + diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_status.h b/drivers/gpu/drm/amd/display/dc/inc/core_status.h index 0a094d7c9380..4ead89dd7c41 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_status.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_status.h @@ -43,11 +43,12 @@ enum dc_status { DC_FAIL_BANDWIDTH_VALIDATE = 13, /* BW and Watermark validation */ DC_FAIL_SCALING = 14, DC_FAIL_DP_LINK_TRAINING = 15, -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 DC_FAIL_DSC_VALIDATE = 16, DC_NO_DSC_RESOURCE = 17, -#endif DC_FAIL_UNSUPPORTED_1 = 18, + DC_FAIL_CLK_EXCEED_MAX = 21, + DC_FAIL_CLK_BELOW_MIN = 22, /*THIS IS MIN PER IP*/ + DC_FAIL_CLK_BELOW_CFG_REQUIRED = 23, /*THIS IS hard_min in PPLIB*/ DC_ERROR_UNEXPECTED = -1 }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index a148ffde8b12..f285b76888fb 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -33,13 +33,11 @@ #include "dc_bios_types.h" #include "mem_input.h" #include "hubp.h" -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) #include "mpc.h" #endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #include "dwb.h" #include "mcif_wb.h" -#endif #define MAX_CLOCK_SOURCES 7 @@ -52,7 +50,9 @@ void enable_surface_flip_reporting(struct dc_plane_state *plane_state, #include "clock_source.h" #include "audio.h" #include "dm_pp_smu.h" - +#ifdef CONFIG_DRM_AMD_DC_HDCP +#include "dm_cp_psp.h" +#endif /************ link *****************/ struct link_init_data { @@ -63,11 +63,6 @@ struct link_init_data { TODO: remove it when DC is complete. */ }; -enum { - FREE_ACQUIRED_RESOURCE = 0, - KEEP_ACQUIRED_RESOURCE = 1, -}; - struct dc_link *link_create(const struct link_init_data *init_params); void link_destroy(struct dc_link **link); @@ -82,7 +77,7 @@ void core_link_enable_stream( struct dc_state *state, struct pipe_ctx *pipe_ctx); -void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option); +void core_link_disable_stream(struct pipe_ctx *pipe_ctx); void core_link_set_avmute(struct pipe_ctx *pipe_ctx, bool enable); /********** DAL Core*********************/ @@ -92,6 +87,7 @@ void core_link_set_avmute(struct pipe_ctx *pipe_ctx, bool enable); struct resource_pool; struct dc_state; struct resource_context; +struct clk_bw_params; struct resource_funcs { void (*destroy)(struct resource_pool **pool); @@ -105,7 +101,7 @@ struct resource_funcs { int (*populate_dml_pipes)( struct dc *dc, - struct resource_context *res_ctx, + struct dc_state *context, display_e2e_pipe_params_st *pipes); enum dc_status (*validate_global)( @@ -135,7 +131,6 @@ struct resource_funcs { struct resource_context *res_ctx, const struct resource_pool *pool, struct dc_stream_state *stream); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void (*populate_dml_writeback_from_context)( struct dc *dc, struct resource_context *res_ctx, @@ -146,7 +141,9 @@ struct resource_funcs { struct dc_state *context, display_e2e_pipe_params_st *pipes, int pipe_cnt); -#endif + void (*update_bw_bounding_box)( + struct dc *dc, + struct clk_bw_params *bw_params); }; @@ -175,7 +172,6 @@ struct resource_pool { struct dce_i2c_sw *sw_i2cs[MAX_PIPES]; bool i2c_hw_buffer_in_use; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dwbc *dwbc[MAX_DWB_PIPES]; struct mcif_wb *mcif_wb[MAX_DWB_PIPES]; struct { @@ -183,11 +179,8 @@ struct resource_pool { unsigned int gsl_1:1; unsigned int gsl_2:1; } gsl_groups; -#endif -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct display_stream_compressor *dscs[MAX_PIPES]; -#endif unsigned int pipe_count; unsigned int underlay_pipe_index; @@ -201,9 +194,7 @@ struct resource_pool { unsigned int timing_generator_count; unsigned int mpcc_count; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) unsigned int writeback_pipe_count; -#endif /* * reserved clock source for DP */ @@ -221,9 +212,12 @@ struct resource_pool { struct abm *abm; struct dmcu *dmcu; + struct dmub_psr *psr; const struct resource_funcs *funcs; const struct resource_caps *res_cap; + + struct ddc_service *oem_device; }; struct dcn_fe_bandwidth { @@ -233,10 +227,7 @@ struct dcn_fe_bandwidth { struct stream_resource { struct output_pixel_processor *opp; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct display_stream_compressor *dsc; - int dscclk_khz; -#endif struct timing_generator *tg; struct stream_encoder *stream_enc; struct audio *audio; @@ -245,12 +236,10 @@ struct stream_resource { struct encoder_info_frame encoder_info_frame; struct abm *abm; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* There are only (num_pipes+1)/2 groups. 0 means unassigned, * otherwise it's using group number 'gsl_group-1' */ uint8_t gsl_group; -#endif }; struct plane_resource { @@ -299,18 +288,18 @@ struct pipe_ctx { struct pipe_ctx *top_pipe; struct pipe_ctx *bottom_pipe; + struct pipe_ctx *next_odm_pipe; + struct pipe_ctx *prev_odm_pipe; -#ifdef CONFIG_DRM_AMD_DC_DCN1_0 +#ifdef CONFIG_DRM_AMD_DC_DCN struct _vcs_dpi_display_dlg_regs_st dlg_regs; struct _vcs_dpi_display_ttu_regs_st ttu_regs; struct _vcs_dpi_display_rq_regs_st rq_regs; struct _vcs_dpi_display_pipe_dest_params_st pipe_dlg_param; #endif union pipe_update_flags update_flags; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 struct dwbc *dwbc; struct mcif_wb *mcif_wb; -#endif }; struct resource_context { @@ -319,9 +308,7 @@ struct resource_context { bool is_audio_acquired[MAX_PIPES]; uint8_t clock_source_ref_count[MAX_CLOCK_SOURCES]; uint8_t dp_clock_source_ref_count; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 bool is_dsc_acquired[MAX_PIPES]; -#endif }; struct dce_bw_output { @@ -341,18 +328,14 @@ struct dce_bw_output { int blackout_recovery_time_us; }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dcn_bw_writeback { struct mcif_arb_params mcif_wb_arb[MAX_DWB_PIPES]; }; -#endif struct dcn_bw_output { struct dc_clocks clk; struct dcn_watermark_set watermarks; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dcn_bw_writeback bw_writeback; -#endif }; union bw_output { @@ -386,16 +369,12 @@ struct dc_state { /* Note: these are big structures, do *not* put on stack! */ struct dm_pp_display_configuration pp_display_cfg; -#ifdef CONFIG_DRM_AMD_DC_DCN1_0 +#ifdef CONFIG_DRM_AMD_DC_DCN struct dcn_bw_internal_vars dcn_bw_vars; #endif struct clk_mgr *clk_mgr; - struct { - bool full_update_needed : 1; - } commit_hints; - struct kref refcount; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h index b1fab251c09b..de2d160114db 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_ddc.h @@ -95,6 +95,9 @@ bool dal_ddc_service_query_ddc_data( uint8_t *read_buf, uint32_t read_size); +bool dal_ddc_submit_aux_command(struct ddc_service *ddc, + struct aux_payload *payload); + int dc_link_aux_transfer_raw(struct ddc_service *ddc, struct aux_payload *payload, enum aux_channel_operation_result *operation_result); @@ -102,6 +105,9 @@ int dc_link_aux_transfer_raw(struct ddc_service *ddc, bool dc_link_aux_transfer_with_retries(struct ddc_service *ddc, struct aux_payload *payload); +uint32_t dc_link_aux_configure_timeout(struct ddc_service *ddc, + uint32_t timeout); + void dal_ddc_service_write_scdc_data( struct ddc_service *ddc_service, uint32_t pix_clk, diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h index 2d95eff94239..8b1f0ce6c2a7 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h @@ -28,6 +28,8 @@ #define LINK_TRAINING_ATTEMPTS 4 #define LINK_TRAINING_RETRY_DELAY 50 /* ms */ +#define LINK_AUX_DEFAULT_EXTENDED_TIMEOUT_PERIOD 3200 /*us*/ +#define LINK_AUX_DEFAULT_TIMEOUT_PERIOD 552 /*us*/ struct dc_link; struct dc_stream_state; @@ -38,6 +40,14 @@ bool dp_verify_link_cap( struct dc_link_settings *known_limit_link_setting, int *fail_count); +bool dp_verify_link_cap_with_retries( + struct dc_link *link, + struct dc_link_settings *known_limit_link_setting, + int attempts); + +bool dp_verify_mst_link_cap( + struct dc_link *link); + bool dp_validate_mode_timing( struct dc_link *link, const struct dc_crtc_timing *timing); @@ -47,10 +57,11 @@ void decide_link_settings( struct dc_link_settings *link_setting); bool perform_link_training_with_retries( - struct dc_link *link, const struct dc_link_settings *link_setting, bool skip_video_pattern, - int attempts); + int attempts, + struct pipe_ctx *pipe_ctx, + enum signal_type signal); bool is_mst_supported(struct dc_link *link); @@ -62,11 +73,16 @@ bool is_dp_active_dongle(const struct dc_link *link); void dp_enable_mst_on_sink(struct dc_link *link, bool enable); -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT +enum dp_panel_mode dp_get_panel_mode(struct dc_link *link); +void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode); + +bool dp_overwrite_extended_receiver_cap(struct dc_link *link); + void dp_set_fec_ready(struct dc_link *link, bool ready); void dp_set_fec_enable(struct dc_link *link, bool enable); bool dp_set_dsc_enable(struct pipe_ctx *pipe_ctx, bool enable); +bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable); +void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable); bool dp_update_dsc_config(struct pipe_ctx *pipe_ctx); -#endif #endif /* __DC_LINK_DP_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_math.h b/drivers/gpu/drm/amd/display/dc/inc/dcn_calc_math.h index 45a07eeffbb6..45a07eeffbb6 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_math.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dcn_calc_math.h diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/aux_engine.h b/drivers/gpu/drm/amd/display/dc/inc/hw/aux_engine.h index e79cd4e92919..e77b3a76766d 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/aux_engine.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/aux_engine.h @@ -140,6 +140,9 @@ struct write_command_context { struct aux_engine_funcs { + bool (*configure_timeout)( + struct ddc_service *ddc, + uint32_t timeout); void (*destroy)( struct aux_engine **ptr); bool (*acquire_engine)( diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h index 36ebd5bc7863..ac530c057ddd 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h @@ -28,6 +28,129 @@ #include "dc.h" +#define DCN_MINIMUM_DISPCLK_Khz 100000 +#define DCN_MINIMUM_DPPCLK_Khz 100000 + +/* Constants */ +#define DDR4_DRAM_WIDTH 64 +#define WM_A 0 +#define WM_B 1 +#define WM_C 2 +#define WM_D 3 +#define WM_SET_COUNT 4 + +#define DCN_MINIMUM_DISPCLK_Khz 100000 +#define DCN_MINIMUM_DPPCLK_Khz 100000 + +/* Will these bw structures be ASIC specific? */ + +#define MAX_NUM_DPM_LVL 8 +#define WM_SET_COUNT 4 + + +struct clk_limit_table_entry { + unsigned int voltage; /* milivolts withh 2 fractional bits */ + unsigned int dcfclk_mhz; + unsigned int fclk_mhz; + unsigned int memclk_mhz; + unsigned int socclk_mhz; +}; + +/* This table is contiguous */ +struct clk_limit_table { + struct clk_limit_table_entry entries[MAX_NUM_DPM_LVL]; + unsigned int num_entries; +}; + +struct wm_range_table_entry { + unsigned int wm_inst; + unsigned int wm_type; + double pstate_latency_us; + double sr_exit_time_us; + double sr_enter_plus_exit_time_us; + bool valid; +}; + + +struct clk_log_info { + bool enabled; + char *pBuf; + unsigned int bufSize; + unsigned int *sum_chars_printed; +}; + +struct clk_state_registers_and_bypass { + uint32_t dcfclk; + uint32_t dcf_deep_sleep_divider; + uint32_t dcf_deep_sleep_allow; + uint32_t dprefclk; + uint32_t dispclk; + uint32_t dppclk; + + uint32_t dppclk_bypass; + uint32_t dcfclk_bypass; + uint32_t dprefclk_bypass; + uint32_t dispclk_bypass; +}; + +struct rv1_clk_internal { + uint32_t CLK0_CLK8_CURRENT_CNT; //dcfclk + uint32_t CLK0_CLK8_DS_CNTL; //dcf_deep_sleep_divider + uint32_t CLK0_CLK8_ALLOW_DS; //dcf_deep_sleep_allow + uint32_t CLK0_CLK10_CURRENT_CNT; //dprefclk + uint32_t CLK0_CLK11_CURRENT_CNT; //dispclk + + uint32_t CLK0_CLK8_BYPASS_CNTL; //dcfclk bypass + uint32_t CLK0_CLK10_BYPASS_CNTL; //dprefclk bypass + uint32_t CLK0_CLK11_BYPASS_CNTL; //dispclk bypass +}; + +struct rn_clk_internal { + uint32_t CLK1_CLK0_CURRENT_CNT; //dispclk + uint32_t CLK1_CLK1_CURRENT_CNT; //dppclk + uint32_t CLK1_CLK2_CURRENT_CNT; //dprefclk + uint32_t CLK1_CLK3_CURRENT_CNT; //dcfclk + uint32_t CLK1_CLK3_DS_CNTL; //dcf_deep_sleep_divider + uint32_t CLK1_CLK3_ALLOW_DS; //dcf_deep_sleep_allow + + uint32_t CLK1_CLK0_BYPASS_CNTL; //dispclk bypass + uint32_t CLK1_CLK1_BYPASS_CNTL; //dppclk bypass + uint32_t CLK1_CLK2_BYPASS_CNTL; //dprefclk bypass + uint32_t CLK1_CLK3_BYPASS_CNTL; //dcfclk bypass + +}; + +/* For dtn logging and debugging */ +struct clk_state_registers { + uint32_t CLK0_CLK8_CURRENT_CNT; //dcfclk + uint32_t CLK0_CLK8_DS_CNTL; //dcf_deep_sleep_divider + uint32_t CLK0_CLK8_ALLOW_DS; //dcf_deep_sleep_allow + uint32_t CLK0_CLK10_CURRENT_CNT; //dprefclk + uint32_t CLK0_CLK11_CURRENT_CNT; //dispclk +}; + +/* TODO: combine this with the above */ +struct clk_bypass { + uint32_t dcfclk_bypass; + uint32_t dispclk_pypass; + uint32_t dprefclk_bypass; +}; +/* + * This table is not contiguous, can have holes, each + * entry correspond to one set of WM. For example if + * we have 2 DPM and LPDDR, we will WM set A, B and + * D occupied, C will be emptry. + */ +struct wm_table { + struct wm_range_table_entry entries[WM_SET_COUNT]; +}; + +struct clk_bw_params { + unsigned int vram_type; + unsigned int num_channels; + struct clk_limit_table clk_table; + struct wm_table wm_table; +}; /* Public interfaces */ struct clk_states { @@ -51,13 +174,25 @@ struct clk_mgr_funcs { void (*init_clocks)(struct clk_mgr *clk_mgr); void (*enable_pme_wa) (struct clk_mgr *clk_mgr); + void (*get_clock)(struct clk_mgr *clk_mgr, + struct dc_state *context, + enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg); + + bool (*are_clock_states_equal) (struct dc_clocks *a, + struct dc_clocks *b); + void (*notify_wm_ranges)(struct clk_mgr *clk_mgr); }; struct clk_mgr { struct dc_context *ctx; struct clk_mgr_funcs *funcs; struct dc_clocks clks; + bool psr_allow_active_cache; int dprefclk_khz; // Used by program pixel clock in clock source funcs, need to figureout where this goes + int dentist_vco_freq_khz; + struct clk_state_registers_and_bypass boot_snapshot; + struct clk_bw_params *bw_params; }; /* forward declarations */ @@ -67,4 +202,8 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p void dc_destroy_clk_mgr(struct clk_mgr *clk_mgr); +void clk_mgr_exit_optimized_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr); + +void clk_mgr_optimize_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr); + #endif /* __DAL_CLK_MGR_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h index 0835ac041acf..862952c0286a 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h @@ -64,6 +64,8 @@ enum dentist_divider_range { *************************************************************************************** */ +/* Macros */ + #define TO_CLK_MGR_INTERNAL(clk_mgr)\ container_of(clk_mgr, struct clk_mgr_internal, base) @@ -94,12 +96,10 @@ enum dentist_divider_range { .MP1_SMN_C2PMSG_83 = mmMP1_SMN_C2PMSG_83, \ .MP1_SMN_C2PMSG_67 = mmMP1_SMN_C2PMSG_67 -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 #define CLK_REG_LIST_NV10() \ SR(DENTIST_DISPCLK_CNTL), \ CLK_SRI(CLK3_CLK_PLL_REQ, CLK3, 0), \ CLK_SRI(CLK3_CLK2_DFS_CNTL, CLK3, 0) -#endif #define CLK_SF(reg_name, field_name, post_fix)\ .field_name = reg_name ## __ ## field_name ## post_fix @@ -118,7 +118,6 @@ enum dentist_divider_range { CLK_SF(MP1_SMN_C2PMSG_83, CONTENT, mask_sh),\ CLK_SF(MP1_SMN_C2PMSG_91, CONTENT, mask_sh), -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 #define CLK_COMMON_MASK_SH_LIST_DCN20_BASE(mask_sh) \ CLK_COMMON_MASK_SH_LIST_DCN_COMMON_BASE(mask_sh),\ CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DPPCLK_WDIVIDER, mask_sh),\ @@ -128,7 +127,6 @@ enum dentist_divider_range { CLK_COMMON_MASK_SH_LIST_DCN20_BASE(mask_sh),\ CLK_SF(CLK3_0_CLK3_CLK_PLL_REQ, FbMult_int, mask_sh),\ CLK_SF(CLK3_0_CLK3_CLK_PLL_REQ, FbMult_frac, mask_sh) -#endif #define CLK_REG_FIELD_LIST(type) \ type DPREFCLK_SRC_SEL; \ @@ -141,30 +139,24 @@ enum dentist_divider_range { ****************** Clock Manager Private Structures *********************************** *************************************************************************************** */ -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 #define CLK20_REG_FIELD_LIST(type) \ type DENTIST_DPPCLK_WDIVIDER; \ type DENTIST_DPPCLK_CHG_DONE; \ type FbMult_int; \ type FbMult_frac; -#endif #define VBIOS_SMU_REG_FIELD_LIST(type) \ type CONTENT; struct clk_mgr_shift { CLK_REG_FIELD_LIST(uint8_t) -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 CLK20_REG_FIELD_LIST(uint8_t) -#endif VBIOS_SMU_REG_FIELD_LIST(uint32_t) }; struct clk_mgr_mask { CLK_REG_FIELD_LIST(uint32_t) -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 CLK20_REG_FIELD_LIST(uint32_t) -#endif VBIOS_SMU_REG_FIELD_LIST(uint32_t) }; @@ -172,16 +164,29 @@ struct clk_mgr_registers { uint32_t DPREFCLK_CNTL; uint32_t DENTIST_DISPCLK_CNTL; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 uint32_t CLK3_CLK2_DFS_CNTL; uint32_t CLK3_CLK_PLL_REQ; -#endif uint32_t MP1_SMN_C2PMSG_67; uint32_t MP1_SMN_C2PMSG_83; uint32_t MP1_SMN_C2PMSG_91; }; +enum clock_type { + clock_type_dispclk = 1, + clock_type_dcfclk, + clock_type_socclk, + clock_type_pixelclk, + clock_type_phyclk, + clock_type_dppclk, + clock_type_fclk, + clock_type_dcfdsclk, + clock_type_dscclk, + clock_type_uclk, + clock_type_dramclk, +}; + + struct state_dependent_clocks { int display_clk_khz; int pixel_clk_khz; @@ -189,6 +194,7 @@ struct state_dependent_clocks { struct clk_mgr_internal { struct clk_mgr base; + int smu_ver; struct pp_smu_funcs *pp_smu; struct clk_mgr_internal_funcs *funcs; @@ -207,12 +213,12 @@ struct clk_mgr_internal { struct state_dependent_clocks max_clks_by_state[DM_PP_CLOCKS_MAX_STATES]; /*TODO: figure out which of the below fields should be here vs in asic specific portion */ - int dentist_vco_freq_khz; - /* Cache the status of DFS-bypass feature*/ bool dfs_bypass_enabled; /* True if the DFS-bypass feature is enabled and active. */ bool dfs_bypass_active; + + uint32_t dfs_ref_freq_khz; /* * Cache the display clock returned by VBIOS if DFS-bypass is enabled. * This is basically "Crystal Frequency In KHz" (XTALIN) frequency @@ -276,8 +282,14 @@ static inline bool should_set_clock(bool safe_to_lower, int calc_clk, int cur_cl static inline bool should_update_pstate_support(bool safe_to_lower, bool calc_support, bool cur_support) { - // Whenever we are transitioning pstate support, we always want to notify prior to committing state - return (calc_support != cur_support) ? !safe_to_lower : false; + if (cur_support != calc_support) { + if (calc_support == true && safe_to_lower) + return true; + else if (calc_support == false && !safe_to_lower) + return true; + } + + return false; } int clk_mgr_helper_get_active_display_cnt( diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h index 9502478c4a1b..c0dc1d0f5cae 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -52,7 +52,6 @@ struct dcn_hubbub_wm { struct dcn_hubbub_wm_set sets[4]; }; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 enum dcn_hubbub_page_table_depth { DCN_PAGE_TABLE_DEPTH_1_LEVEL, DCN_PAGE_TABLE_DEPTH_2_LEVEL, @@ -80,6 +79,8 @@ struct dcn_hubbub_phys_addr_config { uint64_t page_table_end_addr; uint64_t page_table_base_addr; } gart_config; + + uint64_t page_table_default_page_addr; }; struct dcn_hubbub_virt_addr_config { @@ -99,13 +100,11 @@ struct hubbub_addr_config { } default_addrs; }; -#endif struct hubbub_funcs { void (*update_dchub)( struct hubbub *hubbub, struct dchub_init_data *dh_data); -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 int (*init_dchub_sys_ctx)( struct hubbub *hubbub, struct dcn_hubbub_phys_addr_config *pa_config); @@ -114,7 +113,6 @@ struct hubbub_funcs { struct dcn_hubbub_virt_addr_config *va_config, int vmid); -#endif bool (*get_dcc_compression_cap)(struct hubbub *hubbub, const struct dc_dcc_surface_param *input, struct dc_surface_dcc_cap *output); @@ -141,6 +139,11 @@ struct hubbub_funcs { struct dcn_watermark_set *watermarks, unsigned int refclk_mhz, bool safe_to_lower); + + bool (*is_allow_self_refresh_enabled)(struct hubbub *hubbub); + void (*allow_self_refresh_control)(struct hubbub *hubbub, bool allow); + + void (*apply_DEDCN21_147_wa)(struct hubbub *hubbub); }; struct hubbub { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h index c68f0ce346c7..5315f1f86b21 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h @@ -52,6 +52,8 @@ struct dmcu { enum dmcu_state dmcu_state; struct dmcu_version dmcu_version; unsigned int cached_wait_loop_number; + uint32_t psp_version; + bool auto_load_dmcu; }; struct dmcu_funcs { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h index 60c671fcf186..45ef390ae052 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h @@ -36,13 +36,10 @@ struct dpp { struct dpp_caps *caps; struct pwl_params regamma_params; struct pwl_params degamma_params; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dpp_cursor_attributes cur_attr; -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct pwl_params shaper_params; -#endif + bool cm_bypass_mode; }; struct dpp_input_csc_matrix { @@ -50,12 +47,31 @@ struct dpp_input_csc_matrix { uint16_t regval[12]; }; +static const struct dpp_input_csc_matrix dpp_input_csc_matrix[] = { + {COLOR_SPACE_SRGB, + {0x2000, 0, 0, 0, 0, 0x2000, 0, 0, 0, 0, 0x2000, 0} }, + {COLOR_SPACE_SRGB_LIMITED, + {0x2000, 0, 0, 0, 0, 0x2000, 0, 0, 0, 0, 0x2000, 0} }, + {COLOR_SPACE_YCBCR601, + {0x2cdd, 0x2000, 0, 0xe991, 0xe926, 0x2000, 0xf4fd, 0x10ef, + 0, 0x2000, 0x38b4, 0xe3a6} }, + {COLOR_SPACE_YCBCR601_LIMITED, + {0x3353, 0x2568, 0, 0xe400, 0xe5dc, 0x2568, 0xf367, 0x1108, + 0, 0x2568, 0x40de, 0xdd3a} }, + {COLOR_SPACE_YCBCR709, + {0x3265, 0x2000, 0, 0xe6ce, 0xf105, 0x2000, 0xfa01, 0xa7d, 0, + 0x2000, 0x3b61, 0xe24f} }, + + {COLOR_SPACE_YCBCR709_LIMITED, + {0x39a6, 0x2568, 0, 0xe0d6, 0xeedd, 0x2568, 0xf925, 0x9a8, 0, + 0x2568, 0x43ee, 0xdbb2} } +}; + struct dpp_grph_csc_adjustment { struct fixed31_32 temperature_matrix[CSC_TEMPERATURE_MATRIX_SIZE]; enum graphics_gamut_adjust_type gamut_adjust_type; }; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 struct cnv_color_keyer_params { int color_keyer_en; int color_keyer_mode; @@ -81,7 +97,6 @@ struct cnv_alpha_2bit_lut { int lut2; int lut3; }; -#endif struct dcn_dpp_state { uint32_t is_enabled; @@ -189,18 +204,14 @@ struct dpp_funcs { enum surface_pixel_format format, enum expansion_mode mode, struct dc_csc_transform input_csc_color_matrix, -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 enum dc_color_space input_color_space, struct cnv_alpha_2bit_lut *alpha_2bit_lut); -#else - enum dc_color_space input_color_space); -#endif void (*dpp_full_bypass)(struct dpp *dpp_base); void (*set_cursor_attributes)( struct dpp *dpp_base, - enum dc_cursor_color_format color_format); + struct dc_cursor_attributes *cursor_attributes); void (*set_cursor_position)( struct dpp *dpp_base, @@ -223,7 +234,6 @@ struct dpp_funcs { bool dppclk_div, bool enable); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) bool (*dpp_program_blnd_lut)( struct dpp *dpp, const struct pwl_params *params); @@ -236,7 +246,6 @@ struct dpp_funcs { void (*dpp_cnv_set_alpha_keyer)( struct dpp *dpp_base, struct cnv_color_keyer_params *color_keyer); -#endif }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h index c905d020b59e..c59740084ebc 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h @@ -22,13 +22,16 @@ * Authors: AMD * */ -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT #ifndef __DAL_DSC_H__ #define __DAL_DSC_H__ #include "dc_dsc.h" #include "dc_hw_types.h" -#include "dc_dp_types.h" +#include "dc_types.h" +/* do not include any other headers + * or else it might break Edid Utility functionality. + */ + /* Input parameters for configuring DSC from the outside of DSC */ struct dsc_config { @@ -81,21 +84,16 @@ struct dsc_enc_caps { uint32_t bpp_increment_div; /* bpp increment divisor, e.g. if 16, it's 1/16th of a bit */ }; -struct display_stream_compressor { - const struct dsc_funcs *funcs; - struct dc_context *ctx; - int inst; -}; - struct dsc_funcs { void (*dsc_get_enc_caps)(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz); void (*dsc_read_state)(struct display_stream_compressor *dsc, struct dcn_dsc_state *s); bool (*dsc_validate_stream)(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg); void (*dsc_set_config)(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, - struct dsc_optc_config *dsc_optc_cfg, uint8_t *dsc_packed_pps); + struct dsc_optc_config *dsc_optc_cfg); + bool (*dsc_get_packed_pps)(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, + uint8_t *dsc_packed_pps); void (*dsc_enable)(struct display_stream_compressor *dsc, int opp_pipe); void (*dsc_disable)(struct display_stream_compressor *dsc); }; #endif -#endif diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h index a3409294ae0c..459f95f52486 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dwb.h @@ -45,38 +45,21 @@ enum dwb_source { dwb_src_scl = 0, /* for DCE7x/9x, DCN won't support. */ dwb_src_blnd, /* for DCE7x/9x */ dwb_src_fmt, /* for DCE7x/9x */ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) dwb_src_otg0 = 0x100, /* for DCN1.x/DCN2.x, register: mmDWB_SOURCE_SELECT */ dwb_src_otg1, /* for DCN1.x/DCN2.x */ dwb_src_otg2, /* for DCN1.x/DCN2.x */ dwb_src_otg3, /* for DCN1.x/DCN2.x */ -#else - dwb_src_otg0 = 0x100, /* for DCN1.x, register: mmDWB_SOURCE_SELECT */ - dwb_src_otg1, /* for DCN1.x */ - dwb_src_otg2, /* for DCN1.x */ - dwb_src_otg3, /* for DCN1.x */ -#endif - dwb_src_mpc0 = 0x200, /* for DCN2, register: mmMPC_DWB0_MUX, mmMPC_DWB1_MUX, mmMPC_DWB2_MUX */ - dwb_src_mpc1, /* for DCN2 */ - dwb_src_mpc2, /* for DCN2 */ - dwb_src_mpc3, /* for DCN2 */ - dwb_src_mpc4, /* for DCN2 */ }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* DCN1.x, DCN2.x support 2 pipes */ -#else -/* DCN1.x supports 2 pipes */ -#endif enum dwb_pipe { dwb_pipe0 = 0, -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) dwb_pipe1, #endif dwb_pipe_max_num, }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) enum dwb_frame_capture_enable { DWB_FRAME_CAPTURE_DISABLE = 0, DWB_FRAME_CAPTURE_ENABLE = 1, @@ -89,9 +72,7 @@ enum wbscl_coef_filter_type_sel { WBSCL_COEF_CHROMA_HORZ_FILTER = 3 }; -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dwb_warmup_params { bool warmup_en; /* false: normal mode, true: enable pattern generator */ bool warmup_mode; /* false: 420, true: 444 */ @@ -100,7 +81,6 @@ struct dwb_warmup_params { int warmup_width; /* Pattern width (pixels) */ int warmup_height; /* Pattern height (lines) */ }; -#endif struct dwb_caps { enum dce_version hw_version; /* DCN engine version. */ @@ -133,7 +113,8 @@ struct dwbc { int wb_src_plane_inst;/*hubp, mpcc, inst*/ bool update_privacymask; uint32_t mask_id; - + int otg_inst; + bool mvc_cfg; }; struct dwbc_funcs { @@ -162,13 +143,11 @@ struct dwbc_funcs { struct dwbc *dwbc, bool is_new_content); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void (*set_warmup)( struct dwbc *dwbc, struct dwb_warmup_params *warmup_params); -#endif bool (*get_dwb_status)( struct dwbc *dwbc); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/gpio.h b/drivers/gpu/drm/amd/display/dc/inc/hw/gpio.h index 90d0148430fb..5253dc8b15f8 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/gpio.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/gpio.h @@ -28,12 +28,22 @@ #include "gpio_types.h" + +union gpio_hw_container { + struct hw_ddc *ddc; + struct hw_generic *generic; + struct hw_hpd *hpd; +}; + struct gpio { struct gpio_service *service; struct hw_gpio_pin *pin; enum gpio_id id; uint32_t en; + + union gpio_hw_container hw_container; enum gpio_mode mode; + /* when GPIO comes from VBIOS, it has defined output state */ enum gpio_pin_output_state output_state; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h index 51bff8717cc9..2cb8466e657b 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h @@ -38,15 +38,18 @@ enum cursor_pitch { }; enum cursor_lines_per_chunk { -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) CURSOR_LINE_PER_CHUNK_1 = 0, /* new for DCN2 */ -#endif CURSOR_LINE_PER_CHUNK_2 = 1, CURSOR_LINE_PER_CHUNK_4, CURSOR_LINE_PER_CHUNK_8, CURSOR_LINE_PER_CHUNK_16 }; +enum hubp_ind_block_size { + hubp_ind_block_unconstrained = 0, + hubp_ind_block_64b, +}; + struct hubp { const struct hubp_funcs *funcs; struct dc_context *ctx; @@ -60,6 +63,26 @@ struct hubp { bool power_gated; }; +struct surface_flip_registers { + uint32_t DCSURF_SURFACE_CONTROL; + uint32_t DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH; + uint32_t DCSURF_PRIMARY_META_SURFACE_ADDRESS; + uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH; + uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS; + uint32_t DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH_C; + uint32_t DCSURF_PRIMARY_META_SURFACE_ADDRESS_C; + uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C; + uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_C; + uint32_t DCSURF_SECONDARY_META_SURFACE_ADDRESS_HIGH; + uint32_t DCSURF_SECONDARY_META_SURFACE_ADDRESS; + uint32_t DCSURF_SECONDARY_SURFACE_ADDRESS_HIGH; + uint32_t DCSURF_SECONDARY_SURFACE_ADDRESS; + bool tmz_surface; + bool immediate; + uint8_t vmid; + bool grph_stereo; +}; + struct hubp_funcs { void (*hubp_setup)( struct hubp *hubp, @@ -74,12 +97,16 @@ struct hubp_funcs { struct _vcs_dpi_display_ttu_regs_st *ttu_regs); void (*dcc_control)(struct hubp *hubp, bool enable, - bool independent_64b_blks); + enum hubp_ind_block_size blk_size); + void (*mem_program_viewport)( struct hubp *hubp, const struct rect *viewport, const struct rect *viewport_c); + void (*apply_PLAT_54186_wa)(struct hubp *hubp, + const struct dc_plane_address *address); + bool (*hubp_program_surface_flip_and_addr)( struct hubp *hubp, const struct dc_plane_address *address, @@ -103,7 +130,7 @@ struct hubp_funcs { struct hubp *hubp, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror, @@ -111,9 +138,6 @@ struct hubp_funcs { bool (*hubp_is_flip_pending)(struct hubp *hubp); - void (*hubp_update_dchub)(struct hubp *hubp, - struct dchub_init_data *dh_data); - void (*set_blank)(struct hubp *hubp, bool blank); void (*set_hubp_blank_en)(struct hubp *hubp, bool blank); @@ -136,7 +160,6 @@ struct hubp_funcs { unsigned int (*hubp_get_underflow_status)(struct hubp *hubp); void (*hubp_init)(struct hubp *hubp); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void (*dmdata_set_attributes)( struct hubp *hubp, const struct dc_dmdata_attributes *attr); @@ -156,7 +179,13 @@ struct hubp_funcs { void (*hubp_set_flip_control_surface_gsl)( struct hubp *hubp, bool enable); -#endif + + void (*validate_dml_output)( + struct hubp *hubp, + struct dc_context *ctx, + struct _vcs_dpi_display_rq_regs_st *dml_rq_regs, + struct _vcs_dpi_display_dlg_regs_st *dml_dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *dml_ttu_attr); }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h index f82365e2d03c..75d419081e76 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h @@ -36,9 +36,7 @@ #define MAX_AUDIOS 7 #define MAX_PIPES 6 -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define MAX_DWB_PIPES 1 -#endif struct gamma_curve { uint32_t offset; @@ -81,7 +79,6 @@ struct pwl_result_data { uint32_t delta_blue_reg; }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dc_rgb { uint32_t red; uint32_t green; @@ -110,7 +107,6 @@ struct tetrahedral_params { bool use_12bits; }; -#endif /* arr_curve_points - regamma regions/segments specification * arr_points - beginning and end point specified separately (only one on DCE) @@ -195,13 +191,11 @@ enum opp_regamma { OPP_REGAMMA_USER }; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT enum optc_dsc_mode { OPTC_DSC_DISABLED = 0, OPTC_DSC_ENABLED_444 = 1, /* 'RGB 444' or 'Simple YCbCr 4:2:2' (4:2:2 upsampled to 4:4:4) */ OPTC_DSC_ENABLED_NATIVE_SUBSAMPLED = 2 /* Native 4:2:2 or 4:2:0 */ }; -#endif struct dc_bias_and_scale { uint16_t scale_red; @@ -224,12 +218,8 @@ enum test_pattern_mode { TEST_PATTERN_MODE_VERTICALBARS, TEST_PATTERN_MODE_HORIZONTALBARS, TEST_PATTERN_MODE_SINGLERAMP_RGB, -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) TEST_PATTERN_MODE_DUALRAMP_RGB, TEST_PATTERN_MODE_XR_BIAS_RGB -#else - TEST_PATTERN_MODE_DUALRAMP_RGB -#endif }; enum test_pattern_color_format { @@ -255,6 +245,13 @@ enum controller_dp_test_pattern { CONTROLLER_DP_TEST_PATTERN_SOLID_COLOR }; +enum controller_dp_color_space { + CONTROLLER_DP_COLOR_SPACE_RGB, + CONTROLLER_DP_COLOR_SPACE_YCBCR601, + CONTROLLER_DP_COLOR_SPACE_YCBCR709, + CONTROLLER_DP_COLOR_SPACE_UDEFINED +}; + enum dc_lut_mode { LUT_BYPASS, LUT_RAM_A, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h index e5e8640a9ef3..fb748f082c56 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h @@ -113,26 +113,21 @@ struct link_encoder { struct encoder_feature_support features; enum transmitter transmitter; enum hpd_source_id hpd_source; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 bool usbc_combo_phy; -#endif }; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct link_enc_state { uint32_t dphy_fec_en; uint32_t dphy_fec_ready_shadow; uint32_t dphy_fec_active_status; + uint32_t dp_link_training_complete; }; -#endif struct link_encoder_funcs { -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT void (*read_state)( struct link_encoder *enc, struct link_enc_state *s); -#endif bool (*validate_output_with_stream)( struct link_encoder *enc, const struct dc_stream_state *stream); void (*hw_init)(struct link_encoder *enc); @@ -174,7 +169,6 @@ struct link_encoder_funcs { unsigned int (*get_dig_frontend)(struct link_encoder *enc); void (*destroy)(struct link_encoder **enc); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void (*fec_set_enable)(struct link_encoder *enc, bool enable); @@ -182,7 +176,13 @@ struct link_encoder_funcs { bool ready); bool (*fec_is_active)(struct link_encoder *enc); -#endif + bool (*is_in_alt_mode) (struct link_encoder *enc); + + void (*get_max_link_cap)(struct link_encoder *enc, + struct dc_link_settings *link_settings); + + enum signal_type (*get_dig_mode)( + struct link_encoder *enc); }; #endif /* LINK_ENCODER_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h index da89c2edb07c..2e2310f1901a 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h @@ -40,6 +40,9 @@ struct cstate_pstate_watermarks_st { struct dcn_watermarks { uint32_t pte_meta_urgent_ns; uint32_t urgent_ns; + uint32_t frac_urg_bw_nom; + uint32_t frac_urg_bw_flip; + int32_t urgent_latency_ns; struct cstate_pstate_watermarks_st cstate_pstate; }; @@ -149,7 +152,7 @@ struct mem_input_funcs { struct mem_input *mem_input, enum surface_pixel_format format, union dc_tiling_info *tiling_info, - union plane_size *plane_size, + struct plane_size *plane_size, enum dc_rotation_angle rotation, struct dc_plane_dcc_param *dcc, bool horizontal_mirror); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h index 45b94e319cd4..094afc4c8173 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h @@ -31,9 +31,7 @@ #define MAX_MPCC 6 #define MAX_OPP 6 -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define MAX_DWB 1 -#endif enum mpc_output_csc_mode { MPC_OUTPUT_CSC_DISABLE = 0, @@ -66,14 +64,12 @@ struct mpcc_blnd_cfg { int global_alpha; bool overlap_only; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) /* MPCC top/bottom gain settings */ int bottom_gain_mode; int background_color_bpc; int top_gain; int bottom_inside_gain; int bottom_outside_gain; -#endif }; struct mpcc_sm_cfg { @@ -90,7 +86,6 @@ struct mpcc_sm_cfg { int force_next_field_polarity; }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct mpc_denorm_clamp { int clamp_max_r_cr; int clamp_min_r_cr; @@ -99,7 +94,6 @@ struct mpc_denorm_clamp { int clamp_max_b_cb; int clamp_min_b_cb; }; -#endif /* * MPCC connection and blending configuration for a single MPCC instance. @@ -126,9 +120,8 @@ struct mpc { struct dc_context *ctx; struct mpcc mpcc_array[MAX_MPCC]; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct pwl_params blender_params; -#endif + bool cm_bypass_mode; }; struct mpcc_state { @@ -198,6 +191,9 @@ struct mpc_funcs { * Return: void */ void (*mpc_init)(struct mpc *mpc); + void (*mpc_init_single_inst)( + struct mpc *mpc, + unsigned int mpcc_id); /* * Update the blending configuration for a specified MPCC. @@ -226,7 +222,6 @@ struct mpc_funcs { struct mpc *mpc, struct mpc_tree *tree); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void (*set_denorm)(struct mpc *mpc, int opp_id, enum dc_color_depth output_depth); @@ -250,7 +245,10 @@ struct mpc_funcs { struct mpc *mpc, int mpcc_id, const struct pwl_params *params); -#endif + void (*power_on_mpc_mem_pwr)( + struct mpc *mpc, + int mpcc_id, + bool power_on); }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h index 5d8a7bcccc6f..7575564b2265 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h @@ -208,6 +208,7 @@ struct output_pixel_processor { struct mpc_tree mpc_tree_params; bool mpcc_disconnect_pending[MAX_PIPES]; const struct opp_funcs *funcs; + uint32_t dyn_expansion; }; enum fmt_stereo_action { @@ -262,9 +263,7 @@ struct oppbuf_params { enum oppbuf_display_segmentation mso_segmentation; uint32_t mso_overlap_pixel_num; uint32_t pixel_repetition; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) uint32_t num_segment_padded_pixels; -#endif }; struct opp_funcs { @@ -304,10 +303,10 @@ struct opp_funcs { struct output_pixel_processor *opp, bool enable); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void (*opp_set_disp_pattern_generator)( struct output_pixel_processor *opp, enum controller_dp_test_pattern test_pattern, + enum controller_dp_color_space color_space, enum dc_color_depth color_depth, const struct tg_color *solid_color, int width, @@ -316,11 +315,6 @@ struct opp_funcs { bool (*dpg_is_blanked)( struct output_pixel_processor *opp); - void (*opp_convert_pti)( - struct output_pixel_processor *opp, - bool enable, - bool polarity); - void (*opp_dpg_set_blank_color)( struct output_pixel_processor *opp, const struct tg_color *color); @@ -328,7 +322,6 @@ struct opp_funcs { void (*opp_program_left_edge_extra_pixel)( struct output_pixel_processor *opp, bool count); -#endif }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h index ed7d9588b309..351b387ad606 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h @@ -65,13 +65,11 @@ struct audio_clock_info { uint32_t cts_48khz; }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) enum dynamic_metadata_mode { dmdata_dp, dmdata_hdmi, dmdata_dolby_vision }; -#endif struct encoder_info_frame { /* auxiliary video information */ @@ -90,9 +88,7 @@ struct encoder_info_frame { struct encoder_unblank_param { struct dc_link_settings link_settings; struct dc_crtc_timing timing; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 - bool odm; -#endif + int opp_cnt; }; struct encoder_set_dp_phy_pattern_param { @@ -109,7 +105,6 @@ struct stream_encoder { enum engine_id id; }; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT struct enc_state { uint32_t dsc_mode; // DISABLED 0; 1 or 2 indicate enabled state. uint32_t dsc_slice_width; @@ -119,16 +114,13 @@ struct enc_state { uint32_t sec_gsp_pps_enable; uint32_t sec_stream_enable; }; -#endif struct stream_encoder_funcs { - #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT - void (*enc_read_state)(struct stream_encoder *enc, struct enc_state *s); - #endif void (*dp_set_stream_attribute)( struct stream_encoder *enc, struct dc_crtc_timing *crtc_timing, enum dc_color_space output_color_space, + bool use_vsc_sdp_for_colorimetry, uint32_t enable_sdp_splitting); void (*hdmi_set_stream_attribute)( @@ -211,15 +203,28 @@ struct stream_encoder_funcs { struct stream_encoder *enc, int tg_inst); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT + void (*hdmi_reset_stream_attribute)( + struct stream_encoder *enc); + + unsigned int (*dig_source_otg)( + struct stream_encoder *enc); + + bool (*dp_get_pixel_format)( + struct stream_encoder *enc, + enum dc_pixel_encoding *encoding, + enum dc_color_depth *depth); + + void (*enc_read_state)(struct stream_encoder *enc, struct enc_state *s); + void (*dp_set_dsc_config)( struct stream_encoder *enc, enum optc_dsc_mode dsc_mode, uint32_t dsc_bytes_per_pixel, - uint32_t dsc_slice_width, - uint8_t *dsc_packed_pps); -#endif + uint32_t dsc_slice_width); + + void (*dp_set_dsc_pps_info_packet)(struct stream_encoder *enc, + bool enable, + uint8_t *dsc_packed_pps); void (*set_dynamic_metadata)(struct stream_encoder *enc, bool enable, @@ -229,7 +234,6 @@ struct stream_encoder_funcs { void (*dp_set_odm_combine)( struct stream_encoder *enc, bool odm_combine); -#endif }; #endif /* STREAM_ENCODER_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h index 5e93bc0e8ff9..e5e7d94026fc 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h @@ -59,6 +59,8 @@ struct gsl_params { struct drr_params { uint32_t vertical_total_min; uint32_t vertical_total_max; + uint32_t vertical_total_mid; + uint32_t vertical_total_mid_frame_num; bool immediate_flip; }; @@ -96,6 +98,11 @@ enum crc_selection { INTERSECT_WINDOW_NOT_A_NOT_B, }; +enum h_timing_div_mode { + H_TIMING_NO_DIV, + H_TIMING_DIV_BY2, +}; + struct crc_params { /* Regions used to calculate CRC*/ uint16_t windowa_x_start; @@ -184,16 +191,12 @@ struct timing_generator_funcs { bool (*did_triggered_reset_occur)(struct timing_generator *tg); void (*setup_global_swap_lock)(struct timing_generator *tg, const struct dcp_gsl_params *gsl_params); - void (*setup_global_lock)(struct timing_generator *tg); void (*unlock)(struct timing_generator *tg); void (*lock)(struct timing_generator *tg); - void (*lock_global)(struct timing_generator *tg); void (*lock_doublebuffer_disable)(struct timing_generator *tg); void (*lock_doublebuffer_enable)(struct timing_generator *tg); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void(*triplebuffer_unlock)(struct timing_generator *tg); void(*triplebuffer_lock)(struct timing_generator *tg); -#endif void (*enable_reset_trigger)(struct timing_generator *tg, int source_tg_inst); void (*enable_crtc_reset)(struct timing_generator *tg, @@ -205,7 +208,8 @@ struct timing_generator_funcs { bool enable, const struct dc_crtc_timing *timing); void (*set_drr)(struct timing_generator *tg, const struct drr_params *params); void (*set_static_screen_control)(struct timing_generator *tg, - uint32_t value); + uint32_t event_triggers, + uint32_t num_frames); void (*set_test_pattern)( struct timing_generator *tg, enum controller_dp_test_pattern test_pattern, @@ -230,7 +234,6 @@ struct timing_generator_funcs { bool (*is_optc_underflow_occurred)(struct timing_generator *tg); void (*clear_optc_underflow)(struct timing_generator *tg); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) void (*set_dwb_source)(struct timing_generator *optc, uint32_t dwb_pipe_inst); @@ -238,7 +241,6 @@ struct timing_generator_funcs { uint32_t *num_of_input_segments, uint32_t *seg0_src_sel, uint32_t *seg1_src_sel); -#endif /** * Configure CRCs for the given timing generator. Return false if TG is @@ -256,25 +258,23 @@ struct timing_generator_funcs { void (*program_manual_trigger)(struct timing_generator *optc); void (*setup_manual_trigger)(struct timing_generator *optc); + bool (*get_hw_timing)(struct timing_generator *optc, + struct dc_crtc_timing *hw_crtc_timing); void (*set_vtg_params)(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing); -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT void (*set_dsc_config)(struct timing_generator *optc, enum optc_dsc_mode dsc_mode, uint32_t dsc_bytes_per_pixel, uint32_t dsc_slice_width); -#endif - void (*set_odm_bypass)(struct timing_generator *tg, const struct dc_crtc_timing *dc_crtc_timing); - void (*set_odm_combine)(struct timing_generator *tg, int combine_opp_id, - int mpcc_hactive, enum dc_pixel_encoding pixel_encoding); + void (*set_odm_bypass)(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing); + void (*set_odm_combine)(struct timing_generator *optc, int *opp_id, int opp_cnt, + struct dc_crtc_timing *timing); void (*set_gsl)(struct timing_generator *optc, const struct gsl_params *params); void (*set_gsl_source_select)(struct timing_generator *optc, int group_idx, uint32_t gsl_ready_signal); -#endif }; #endif diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h index 4d56d48a3179..209118f9f193 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h @@ -32,268 +32,160 @@ #include "inc/hw/link_encoder.h" #include "core_status.h" -enum pipe_gating_control { - PIPE_GATING_CONTROL_DISABLE = 0, - PIPE_GATING_CONTROL_ENABLE, - PIPE_GATING_CONTROL_INIT -}; - enum vline_select { VLINE0, VLINE1 }; -struct dce_hwseq_wa { - bool blnd_crtc_trigger; - bool DEGVIDCN10_253; - bool false_optc_underflow; - bool DEGVIDCN10_254; -}; - -struct hwseq_wa_state { - bool DEGVIDCN10_253_applied; -}; - -struct dce_hwseq { - struct dc_context *ctx; - const struct dce_hwseq_registers *regs; - const struct dce_hwseq_shift *shifts; - const struct dce_hwseq_mask *masks; - struct dce_hwseq_wa wa; - struct hwseq_wa_state wa_state; -}; - struct pipe_ctx; struct dc_state; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) struct dc_stream_status; struct dc_writeback_info; -#endif struct dchub_init_data; -struct dc_static_screen_events; +struct dc_static_screen_params; struct resource_pool; -struct resource_context; -struct stream_resource; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 struct dc_phy_addr_space_config; struct dc_virtual_addr_space_config; -#endif +struct dpp; +struct dce_hwseq; struct hw_sequencer_funcs { + /* Embedded Display Related */ + void (*edp_power_control)(struct dc_link *link, bool enable); + void (*edp_wait_for_hpd_ready)(struct dc_link *link, bool power_up); - void (*disable_stream_gating)(struct dc *dc, struct pipe_ctx *pipe_ctx); - - void (*enable_stream_gating)(struct dc *dc, struct pipe_ctx *pipe_ctx); - + /* Pipe Programming Related */ void (*init_hw)(struct dc *dc); - - void (*init_pipes)(struct dc *dc, struct dc_state *context); - - enum dc_status (*apply_ctx_to_hw)( - struct dc *dc, struct dc_state *context); - - void (*reset_hw_ctx_wrap)( - struct dc *dc, struct dc_state *context); - - void (*apply_ctx_for_surface)( - struct dc *dc, + void (*enable_accelerated_mode)(struct dc *dc, + struct dc_state *context); + enum dc_status (*apply_ctx_to_hw)(struct dc *dc, + struct dc_state *context); + void (*disable_plane)(struct dc *dc, struct pipe_ctx *pipe_ctx); + void (*apply_ctx_for_surface)(struct dc *dc, const struct dc_stream_state *stream, - int num_planes, + int num_planes, struct dc_state *context); + void (*program_front_end_for_ctx)(struct dc *dc, struct dc_state *context); - - void (*program_gamut_remap)( + void (*update_plane_addr)(const struct dc *dc, struct pipe_ctx *pipe_ctx); - - void (*program_output_csc)(struct dc *dc, - struct pipe_ctx *pipe_ctx, - enum dc_color_space colorspace, - uint16_t *matrix, - int opp_id); - -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - void (*program_triplebuffer)( - const struct dc *dc, - struct pipe_ctx *pipe_ctx, - bool enableTripleBuffer); - void (*set_flip_control_gsl)( - struct pipe_ctx *pipe_ctx, - bool flip_immediate); -#endif - - void (*update_plane_addr)( - const struct dc *dc, - struct pipe_ctx *pipe_ctx); - - void (*plane_atomic_disconnect)( - struct dc *dc, - struct pipe_ctx *pipe_ctx); - - void (*update_dchub)( - struct dce_hwseq *hws, - struct dchub_init_data *dh_data); - -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 - int (*init_sys_ctx)( - struct dce_hwseq *hws, - struct dc *dc, - struct dc_phy_addr_space_config *pa_config); - void (*init_vm_ctx)( - struct dce_hwseq *hws, - struct dc *dc, - struct dc_virtual_addr_space_config *va_config, - int vmid); -#endif - void (*update_mpcc)( - struct dc *dc, - struct pipe_ctx *pipe_ctx); - - void (*update_pending_status)( + void (*update_dchub)(struct dce_hwseq *hws, + struct dchub_init_data *dh_data); + void (*wait_for_mpcc_disconnect)(struct dc *dc, + struct resource_pool *res_pool, struct pipe_ctx *pipe_ctx); - - bool (*set_input_transfer_func)( - struct pipe_ctx *pipe_ctx, - const struct dc_plane_state *plane_state); - - bool (*set_output_transfer_func)( - struct pipe_ctx *pipe_ctx, - const struct dc_stream_state *stream); - - void (*power_down)(struct dc *dc); - - void (*enable_accelerated_mode)(struct dc *dc, struct dc_state *context); - - void (*enable_timing_synchronization)( - struct dc *dc, - int group_index, - int group_size, - struct pipe_ctx *grouped_pipes[]); - - void (*enable_per_frame_crtc_position_reset)( - struct dc *dc, - int group_size, + void (*program_triplebuffer)(const struct dc *dc, + struct pipe_ctx *pipe_ctx, bool enableTripleBuffer); + void (*update_pending_status)(struct pipe_ctx *pipe_ctx); + + /* Pipe Lock Related */ + void (*pipe_control_lock_global)(struct dc *dc, + struct pipe_ctx *pipe, bool lock); + void (*pipe_control_lock)(struct dc *dc, + struct pipe_ctx *pipe, bool lock); + void (*set_flip_control_gsl)(struct pipe_ctx *pipe_ctx, + bool flip_immediate); + + /* Timing Related */ + void (*get_position)(struct pipe_ctx **pipe_ctx, int num_pipes, + struct crtc_position *position); + int (*get_vupdate_offset_from_vsync)(struct pipe_ctx *pipe_ctx); + void (*enable_per_frame_crtc_position_reset)(struct dc *dc, + int group_size, struct pipe_ctx *grouped_pipes[]); + void (*enable_timing_synchronization)(struct dc *dc, + int group_index, int group_size, struct pipe_ctx *grouped_pipes[]); + void (*setup_periodic_interrupt)(struct dc *dc, + struct pipe_ctx *pipe_ctx, + enum vline_select vline); + void (*set_drr)(struct pipe_ctx **pipe_ctx, int num_pipes, + unsigned int vmin, unsigned int vmax, + unsigned int vmid, unsigned int vmid_frame_number); + void (*set_static_screen_control)(struct pipe_ctx **pipe_ctx, + int num_pipes, + const struct dc_static_screen_params *events); - void (*enable_display_pipe_clock_gating)( - struct dc_context *ctx, - bool clock_gating); - - bool (*enable_display_power_gating)( - struct dc *dc, - uint8_t controller_id, - struct dc_bios *dcb, - enum pipe_gating_control power_gating); - - void (*disable_plane)(struct dc *dc, struct pipe_ctx *pipe_ctx); - - void (*update_info_frame)(struct pipe_ctx *pipe_ctx); - - void (*send_immediate_sdp_message)( - struct pipe_ctx *pipe_ctx, - const uint8_t *custom_sdp_message, - unsigned int sdp_message_size); - + /* Stream Related */ void (*enable_stream)(struct pipe_ctx *pipe_ctx); - - void (*disable_stream)(struct pipe_ctx *pipe_ctx, - int option); - + void (*disable_stream)(struct pipe_ctx *pipe_ctx); + void (*blank_stream)(struct pipe_ctx *pipe_ctx); void (*unblank_stream)(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings); - void (*blank_stream)(struct pipe_ctx *pipe_ctx); - - void (*enable_audio_stream)(struct pipe_ctx *pipe_ctx); + /* Bandwidth Related */ + void (*prepare_bandwidth)(struct dc *dc, struct dc_state *context); + bool (*update_bandwidth)(struct dc *dc, struct dc_state *context); + void (*optimize_bandwidth)(struct dc *dc, struct dc_state *context); - void (*disable_audio_stream)(struct pipe_ctx *pipe_ctx, int option); + /* Infopacket Related */ + void (*set_avmute)(struct pipe_ctx *pipe_ctx, bool enable); + void (*send_immediate_sdp_message)( + struct pipe_ctx *pipe_ctx, + const uint8_t *custom_sdp_message, + unsigned int sdp_message_size); + void (*update_info_frame)(struct pipe_ctx *pipe_ctx); + void (*set_dmdata_attributes)(struct pipe_ctx *pipe); + void (*program_dmdata_engine)(struct pipe_ctx *pipe_ctx); + bool (*dmdata_status_done)(struct pipe_ctx *pipe_ctx); - void (*pipe_control_lock)( - struct dc *dc, - struct pipe_ctx *pipe, - bool lock); + /* Cursor Related */ + void (*set_cursor_position)(struct pipe_ctx *pipe); + void (*set_cursor_attribute)(struct pipe_ctx *pipe); + void (*set_cursor_sdr_white_level)(struct pipe_ctx *pipe); - void (*pipe_control_lock_global)( - struct dc *dc, - struct pipe_ctx *pipe, - bool lock); - void (*blank_pixel_data)( - struct dc *dc, - struct pipe_ctx *pipe_ctx, - bool blank); + /* Colour Related */ + void (*program_gamut_remap)(struct pipe_ctx *pipe_ctx); + void (*program_output_csc)(struct dc *dc, struct pipe_ctx *pipe_ctx, + enum dc_color_space colorspace, + uint16_t *matrix, int opp_id); - void (*prepare_bandwidth)( + /* VM Related */ + int (*init_sys_ctx)(struct dce_hwseq *hws, struct dc *dc, - struct dc_state *context); - void (*optimize_bandwidth)( + struct dc_phy_addr_space_config *pa_config); + void (*init_vm_ctx)(struct dce_hwseq *hws, struct dc *dc, - struct dc_state *context); + struct dc_virtual_addr_space_config *va_config, + int vmid); -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - bool (*update_bandwidth)( - struct dc *dc, + /* Writeback Related */ + void (*update_writeback)(struct dc *dc, + struct dc_writeback_info *wb_info, struct dc_state *context); - bool (*dmdata_status_done)(struct pipe_ctx *pipe_ctx); -#endif - - void (*set_drr)(struct pipe_ctx **pipe_ctx, int num_pipes, - int vmin, int vmax); - - void (*get_position)(struct pipe_ctx **pipe_ctx, int num_pipes, - struct crtc_position *position); + void (*enable_writeback)(struct dc *dc, + struct dc_writeback_info *wb_info, + struct dc_state *context); + void (*disable_writeback)(struct dc *dc, + unsigned int dwb_pipe_inst); - void (*set_static_screen_control)(struct pipe_ctx **pipe_ctx, - int num_pipes, const struct dc_static_screen_events *events); + bool (*mmhubbub_warmup)(struct dc *dc, + unsigned int num_dwb, + struct dc_writeback_info *wb_info); - enum dc_status (*enable_stream_timing)( - struct pipe_ctx *pipe_ctx, - struct dc_state *context, - struct dc *dc); + /* Clock Related */ + enum dc_status (*set_clock)(struct dc *dc, + enum dc_clock_type clock_type, + uint32_t clk_khz, uint32_t stepping); + void (*get_clock)(struct dc *dc, enum dc_clock_type clock_type, + struct dc_clock_config *clock_cfg); + void (*optimize_pwr_state)(const struct dc *dc, + struct dc_state *context); + void (*exit_optimized_pwr_state)(const struct dc *dc, + struct dc_state *context); - void (*setup_stereo)( - struct pipe_ctx *pipe_ctx, - struct dc *dc); + /* Audio Related */ + void (*enable_audio_stream)(struct pipe_ctx *pipe_ctx); + void (*disable_audio_stream)(struct pipe_ctx *pipe_ctx); - void (*set_avmute)(struct pipe_ctx *pipe_ctx, bool enable); + /* Stereo 3D Related */ + void (*setup_stereo)(struct pipe_ctx *pipe_ctx, struct dc *dc); - void (*log_hw_state)(struct dc *dc, - struct dc_log_buffer_ctx *log_ctx); - void (*get_hw_state)(struct dc *dc, char *pBuf, unsigned int bufSize, unsigned int mask); + /* HW State Logging Related */ + void (*log_hw_state)(struct dc *dc, struct dc_log_buffer_ctx *log_ctx); + void (*get_hw_state)(struct dc *dc, char *pBuf, + unsigned int bufSize, unsigned int mask); void (*clear_status_bits)(struct dc *dc, unsigned int mask); - void (*wait_for_mpcc_disconnect)(struct dc *dc, - struct resource_pool *res_pool, - struct pipe_ctx *pipe_ctx); - - void (*edp_power_control)( - struct dc_link *link, - bool enable); - void (*edp_backlight_control)( - struct dc_link *link, - bool enable); - void (*edp_wait_for_hpd_ready)(struct dc_link *link, bool power_up); - void (*set_cursor_position)(struct pipe_ctx *pipe); - void (*set_cursor_attribute)(struct pipe_ctx *pipe); - void (*set_cursor_sdr_white_level)(struct pipe_ctx *pipe); - - void (*setup_periodic_interrupt)(struct pipe_ctx *pipe_ctx, enum vline_select vline); - void (*setup_vupdate_interrupt)(struct pipe_ctx *pipe_ctx); - bool (*did_underflow_occur)(struct dc *dc, struct pipe_ctx *pipe_ctx); - -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) - void (*update_odm)(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx); - void (*program_all_writeback_pipes_in_tree)( - struct dc *dc, - const struct dc_stream_state *stream, - struct dc_state *context); - void (*update_writeback)(struct dc *dc, - const struct dc_stream_status *stream_status, - struct dc_writeback_info *wb_info); - void (*enable_writeback)(struct dc *dc, - const struct dc_stream_status *stream_status, - struct dc_writeback_info *wb_info); - void (*disable_writeback)(struct dc *dc, - unsigned int dwb_pipe_inst); -#endif }; void color_space_to_black_color( diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h new file mode 100644 index 000000000000..ecf566378ccd --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h @@ -0,0 +1,156 @@ +/* + * Copyright 2015 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_HW_SEQUENCER_PRIVATE_H__ +#define __DC_HW_SEQUENCER_PRIVATE_H__ + +#include "dc_types.h" + +enum pipe_gating_control { + PIPE_GATING_CONTROL_DISABLE = 0, + PIPE_GATING_CONTROL_ENABLE, + PIPE_GATING_CONTROL_INIT +}; + +struct dce_hwseq_wa { + bool blnd_crtc_trigger; + bool DEGVIDCN10_253; + bool false_optc_underflow; + bool DEGVIDCN10_254; + bool DEGVIDCN21; +}; + +struct hwseq_wa_state { + bool DEGVIDCN10_253_applied; +}; + +struct pipe_ctx; +struct dc_state; +struct dc_stream_status; +struct dc_writeback_info; +struct dchub_init_data; +struct dc_static_screen_params; +struct resource_pool; +struct resource_context; +struct stream_resource; +struct dc_phy_addr_space_config; +struct dc_virtual_addr_space_config; +struct hubp; +struct dpp; +struct dce_hwseq; +struct timing_generator; +struct tg_color; +struct output_pixel_processor; + +struct hwseq_private_funcs { + + void (*disable_stream_gating)(struct dc *dc, struct pipe_ctx *pipe_ctx); + void (*enable_stream_gating)(struct dc *dc, struct pipe_ctx *pipe_ctx); + void (*init_pipes)(struct dc *dc, struct dc_state *context); + void (*reset_hw_ctx_wrap)(struct dc *dc, struct dc_state *context); + void (*update_plane_addr)(const struct dc *dc, + struct pipe_ctx *pipe_ctx); + void (*plane_atomic_disconnect)(struct dc *dc, + struct pipe_ctx *pipe_ctx); + void (*update_mpcc)(struct dc *dc, struct pipe_ctx *pipe_ctx); + bool (*set_input_transfer_func)(struct dc *dc, + struct pipe_ctx *pipe_ctx, + const struct dc_plane_state *plane_state); + bool (*set_output_transfer_func)(struct dc *dc, + struct pipe_ctx *pipe_ctx, + const struct dc_stream_state *stream); + void (*power_down)(struct dc *dc); + void (*enable_display_pipe_clock_gating)(struct dc_context *ctx, + bool clock_gating); + bool (*enable_display_power_gating)(struct dc *dc, + uint8_t controller_id, + struct dc_bios *dcb, + enum pipe_gating_control power_gating); + void (*blank_pixel_data)(struct dc *dc, + struct pipe_ctx *pipe_ctx, + bool blank); + enum dc_status (*enable_stream_timing)( + struct pipe_ctx *pipe_ctx, + struct dc_state *context, + struct dc *dc); + void (*edp_backlight_control)(struct dc_link *link, + bool enable); + void (*setup_vupdate_interrupt)(struct dc *dc, + struct pipe_ctx *pipe_ctx); + bool (*did_underflow_occur)(struct dc *dc, struct pipe_ctx *pipe_ctx); + void (*init_blank)(struct dc *dc, struct timing_generator *tg); + void (*disable_vga)(struct dce_hwseq *hws); + void (*bios_golden_init)(struct dc *dc); + void (*plane_atomic_power_down)(struct dc *dc, + struct dpp *dpp, + struct hubp *hubp); + void (*plane_atomic_disable)(struct dc *dc, struct pipe_ctx *pipe_ctx); + void (*enable_power_gating_plane)(struct dce_hwseq *hws, + bool enable); + void (*dpp_pg_control)(struct dce_hwseq *hws, + unsigned int dpp_inst, + bool power_on); + void (*hubp_pg_control)(struct dce_hwseq *hws, + unsigned int hubp_inst, + bool power_on); + void (*dsc_pg_control)(struct dce_hwseq *hws, + unsigned int dsc_inst, + bool power_on); + void (*update_odm)(struct dc *dc, struct dc_state *context, + struct pipe_ctx *pipe_ctx); + void (*program_all_writeback_pipes_in_tree)(struct dc *dc, + const struct dc_stream_state *stream, + struct dc_state *context); + bool (*s0i3_golden_init_wa)(struct dc *dc); + void (*get_surface_visual_confirm_color)( + const struct pipe_ctx *pipe_ctx, + struct tg_color *color); + void (*get_hdr_visual_confirm_color)(struct pipe_ctx *pipe_ctx, + struct tg_color *color); + void (*set_hdr_multiplier)(struct pipe_ctx *pipe_ctx); + void (*verify_allow_pstate_change_high)(struct dc *dc); + void (*program_pipe)(struct dc *dc, + struct pipe_ctx *pipe_ctx, + struct dc_state *context); + bool (*wait_for_blank_complete)(struct output_pixel_processor *opp); + void (*dccg_init)(struct dce_hwseq *hws); + bool (*set_blend_lut)(struct pipe_ctx *pipe_ctx, + const struct dc_plane_state *plane_state); + bool (*set_shaper_3dlut)(struct pipe_ctx *pipe_ctx, + const struct dc_plane_state *plane_state); +}; + +struct dce_hwseq { + struct dc_context *ctx; + const struct dce_hwseq_registers *regs; + const struct dce_hwseq_shift *shifts; + const struct dce_hwseq_mask *masks; + struct dce_hwseq_wa wa; + struct hwseq_wa_state wa_state; + struct hwseq_private_funcs funcs; + +}; + +#endif /* __DC_HW_SEQUENCER_PRIVATE_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h b/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h index 30be7bb4a01a..9af7ee5bc8ee 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h @@ -60,11 +60,13 @@ void dp_disable_link_phy_mst(struct dc_link *link, enum signal_type signal); bool dp_set_hw_training_pattern( struct dc_link *link, - enum hw_dp_training_pattern pattern); + enum dc_dp_training_pattern pattern, + uint32_t offset); void dp_set_hw_lane_settings( struct dc_link *link, - const struct link_training_settings *link_settings); + const struct link_training_settings *link_settings, + uint32_t offset); void dp_set_hw_test_pattern( struct dc_link *link, @@ -72,8 +74,6 @@ void dp_set_hw_test_pattern( uint8_t *custom_pattern, uint32_t custom_pattern_size); -enum dp_panel_mode dp_get_panel_mode(struct dc_link *link); - void dp_retrain_link_dp_test(struct dc_link *link, struct dc_link_settings *link_setting, bool skip_video_pattern); diff --git a/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h b/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h index 8503d9cc4763..2470405e996b 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h +++ b/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h @@ -458,7 +458,14 @@ uint32_t generic_reg_get8(const struct dc_context *ctx, uint32_t addr, #define IX_REG_READ(index_reg_name, data_reg_name, index) \ generic_read_indirect_reg(CTX, REG(index_reg_name), REG(data_reg_name), IND_REG(index)) +#define IX_REG_GET_N(index_reg_name, data_reg_name, index, n, ...) \ + generic_indirect_reg_get(CTX, REG(index_reg_name), REG(data_reg_name), \ + IND_REG(index), \ + n, __VA_ARGS__) +#define IX_REG_GET(index_reg_name, data_reg_name, index, field, val) \ + IX_REG_GET_N(index_reg_name, data_reg_name, index, 1, \ + FN(data_reg_name, field), val) #define IX_REG_UPDATE_N(index_reg_name, data_reg_name, index, n, ...) \ generic_indirect_reg_update_ex(CTX, \ @@ -479,10 +486,35 @@ uint32_t generic_read_indirect_reg(const struct dc_context *ctx, uint32_t addr_index, uint32_t addr_data, uint32_t index); +uint32_t generic_indirect_reg_get(const struct dc_context *ctx, + uint32_t addr_index, uint32_t addr_data, + uint32_t index, int n, + uint8_t shift1, uint32_t mask1, uint32_t *field_value1, + ...); + uint32_t generic_indirect_reg_update_ex(const struct dc_context *ctx, uint32_t addr_index, uint32_t addr_data, uint32_t index, uint32_t reg_val, int n, uint8_t shift1, uint32_t mask1, uint32_t field_value1, ...); +/* register offload macros + * + * instead of MMIO to register directly, in some cases we want + * to gather register sequence and execute the register sequence + * from another thread so we optimize time required for lengthy ops + */ + +/* start gathering register sequence */ +#define REG_SEQ_START() \ + reg_sequence_start_gather(CTX) + +/* start execution of register sequence gathered since REG_SEQ_START */ +#define REG_SEQ_SUBMIT() \ + reg_sequence_start_execute(CTX) + +/* wait for the last REG_SEQ_SUBMIT to finish */ +#define REG_SEQ_WAIT_DONE() \ + reg_sequence_wait_done(CTX) + #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_REG_HELPER_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index 47f81072d7e9..5ae8ada154ef 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -31,6 +31,8 @@ #include "dm_pp_smu.h" #define MEMORY_TYPE_MULTIPLIER_CZ 4 +#define MEMORY_TYPE_HBM 2 + enum dce_version resource_parse_asic_id( struct hw_asic_id asic_id); @@ -44,12 +46,8 @@ struct resource_caps { int num_pll; int num_dwb; int num_ddc; -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 int num_vmid; -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT int num_dsc; -#endif -#endif }; struct resource_straps { @@ -179,7 +177,6 @@ void update_audio_usage( unsigned int resource_pixel_format_to_bpp(enum surface_pixel_format format); -struct pipe_ctx *dc_res_get_odm_bottom_pipe(struct pipe_ctx *pipe_ctx); -bool dc_res_is_odm_head_pipe(struct pipe_ctx *pipe_ctx); - +void get_audio_check(struct audio_info *aud_modes, + struct audio_check *aud_chk); #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/irq/Makefile b/drivers/gpu/drm/amd/display/dc/irq/Makefile index ad87c2f093e2..0f682ac53bb2 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/Makefile +++ b/drivers/gpu/drm/amd/display/dc/irq/Makefile @@ -60,20 +60,26 @@ AMD_DISPLAY_FILES += $(AMD_DAL_IRQ_DCE12) ############################################################################### # DCN 1x ############################################################################### -ifdef CONFIG_DRM_AMD_DC_DCN1_0 +ifdef CONFIG_DRM_AMD_DC_DCN IRQ_DCN1 = irq_service_dcn10.o AMD_DAL_IRQ_DCN1 = $(addprefix $(AMDDALPATH)/dc/irq/dcn10/,$(IRQ_DCN1)) AMD_DISPLAY_FILES += $(AMD_DAL_IRQ_DCN1) -endif ############################################################################### # DCN 20 ############################################################################### -ifdef CONFIG_DRM_AMD_DC_DCN2_0 IRQ_DCN2 = irq_service_dcn20.o AMD_DAL_IRQ_DCN2 = $(addprefix $(AMDDALPATH)/dc/irq/dcn20/,$(IRQ_DCN2)) AMD_DISPLAY_FILES += $(AMD_DAL_IRQ_DCN2) +############################################################################### +# DCN 21 +############################################################################### +IRQ_DCN21 = irq_service_dcn21.o + +AMD_DAL_IRQ_DCN21= $(addprefix $(AMDDALPATH)/dc/irq/dcn21/,$(IRQ_DCN21)) + +AMD_DISPLAY_FILES += $(AMD_DAL_IRQ_DCN21) endif diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c b/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c index 1a581c464345..378cc11aa047 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c @@ -204,7 +204,7 @@ bool dce110_vblank_set(struct irq_service *irq_service, bool enable) { struct dc_context *dc_ctx = irq_service->ctx; - struct dc *core_dc = irq_service->ctx->dc; + struct dc *dc = irq_service->ctx->dc; enum dc_irq_source dal_irq_src = dc_interrupt_to_irq_source(irq_service->ctx->dc, info->src_id, @@ -212,7 +212,7 @@ bool dce110_vblank_set(struct irq_service *irq_service, uint8_t pipe_offset = dal_irq_src - IRQ_TYPE_VBLANK; struct timing_generator *tg = - core_dc->current_state->res_ctx.pipe_ctx[pipe_offset].stream_res.tg; + dc->current_state->res_ctx.pipe_ctx[pipe_offset].stream_res.tg; if (enable) { if (!tg || !tg->funcs->arm_vert_intr(tg, 2)) { @@ -403,7 +403,7 @@ static const struct irq_service_funcs irq_service_funcs_dce110 = { .to_dal_irq_source = to_dal_irq_source_dce110 }; -static void construct(struct irq_service *irq_service, +static void dce110_irq_construct(struct irq_service *irq_service, struct irq_service_init_data *init_data) { dal_irq_service_construct(irq_service, init_data); @@ -421,6 +421,6 @@ dal_irq_service_dce110_create(struct irq_service_init_data *init_data) if (!irq_service) return NULL; - construct(irq_service, init_data); + dce110_irq_construct(irq_service, init_data); return irq_service; } diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c b/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c index 15380336cb51..2fe4703395f3 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c @@ -273,7 +273,7 @@ static const struct irq_service_funcs irq_service_funcs_dce120 = { .to_dal_irq_source = to_dal_irq_source_dce110 }; -static void construct( +static void dce120_irq_construct( struct irq_service *irq_service, struct irq_service_init_data *init_data) { @@ -292,6 +292,6 @@ struct irq_service *dal_irq_service_dce120_create( if (!irq_service) return NULL; - construct(irq_service, init_data); + dce120_irq_construct(irq_service, init_data); return irq_service; } diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c b/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c index 281fee8ad1e5..17e426b80a00 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c @@ -283,7 +283,7 @@ static const struct irq_service_funcs irq_service_funcs_dce80 = { .to_dal_irq_source = to_dal_irq_source_dce110 }; -static void construct( +static void dce80_irq_construct( struct irq_service *irq_service, struct irq_service_init_data *init_data) { @@ -302,7 +302,7 @@ struct irq_service *dal_irq_service_dce80_create( if (!irq_service) return NULL; - construct(irq_service, init_data); + dce80_irq_construct(irq_service, init_data); return irq_service; } diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c b/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c index cc8e7dedccce..f956b3bde680 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c @@ -355,7 +355,7 @@ static const struct irq_service_funcs irq_service_funcs_dcn10 = { .to_dal_irq_source = to_dal_irq_source_dcn10 }; -static void construct( +static void dcn10_irq_construct( struct irq_service *irq_service, struct irq_service_init_data *init_data) { @@ -374,6 +374,6 @@ struct irq_service *dal_irq_service_dcn10_create( if (!irq_service) return NULL; - construct(irq_service, init_data); + dcn10_irq_construct(irq_service, init_data); return irq_service; } diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c b/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c index 3cc0f2a1f77c..2a1fea501f8c 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c @@ -167,6 +167,11 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = { .ack = NULL }; +static const struct irq_source_info_funcs vupdate_no_lock_irq_info_funcs = { + .set = NULL, + .ack = NULL +}; + #undef BASE_INNER #define BASE_INNER(seg) DCN_BASE__INST0_SEG ## seg @@ -221,12 +226,15 @@ static const struct irq_source_info_funcs vblank_irq_info_funcs = { .funcs = &pflip_irq_info_funcs\ } -#define vupdate_int_entry(reg_num)\ +/* vupdate_no_lock_int_entry maps to DC_IRQ_SOURCE_VUPDATEx, to match semantic + * of DCE's DC_IRQ_SOURCE_VUPDATEx. + */ +#define vupdate_no_lock_int_entry(reg_num)\ [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\ IRQ_REG_ENTRY(OTG, reg_num,\ - OTG_GLOBAL_SYNC_STATUS, VUPDATE_INT_EN,\ - OTG_GLOBAL_SYNC_STATUS, VUPDATE_EVENT_CLEAR),\ - .funcs = &vblank_irq_info_funcs\ + OTG_GLOBAL_SYNC_STATUS, VUPDATE_NO_LOCK_INT_EN,\ + OTG_GLOBAL_SYNC_STATUS, VUPDATE_NO_LOCK_EVENT_CLEAR),\ + .funcs = &vupdate_no_lock_irq_info_funcs\ } #define vblank_int_entry(reg_num)\ @@ -333,12 +341,12 @@ irq_source_info_dcn20[DAL_IRQ_SOURCES_NUMBER] = { dc_underflow_int_entry(6), [DC_IRQ_SOURCE_DMCU_SCP] = dummy_irq_entry(), [DC_IRQ_SOURCE_VBIOS_SW] = dummy_irq_entry(), - vupdate_int_entry(0), - vupdate_int_entry(1), - vupdate_int_entry(2), - vupdate_int_entry(3), - vupdate_int_entry(4), - vupdate_int_entry(5), + vupdate_no_lock_int_entry(0), + vupdate_no_lock_int_entry(1), + vupdate_no_lock_int_entry(2), + vupdate_no_lock_int_entry(3), + vupdate_no_lock_int_entry(4), + vupdate_no_lock_int_entry(5), vblank_int_entry(0), vblank_int_entry(1), vblank_int_entry(2), @@ -351,7 +359,7 @@ static const struct irq_service_funcs irq_service_funcs_dcn20 = { .to_dal_irq_source = to_dal_irq_source_dcn20 }; -static void construct( +static void dcn20_irq_construct( struct irq_service *irq_service, struct irq_service_init_data *init_data) { @@ -370,6 +378,6 @@ struct irq_service *dal_irq_service_dcn20_create( if (!irq_service) return NULL; - construct(irq_service, init_data); + dcn20_irq_construct(irq_service, init_data); return irq_service; } diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c new file mode 100644 index 000000000000..1b971265418b --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c @@ -0,0 +1,374 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include <linux/slab.h> + +#include "dm_services.h" + +#include "include/logger_interface.h" + +#include "../dce110/irq_service_dce110.h" + +#include "dcn/dcn_2_1_0_offset.h" +#include "dcn/dcn_2_1_0_sh_mask.h" +#include "renoir_ip_offset.h" + + +#include "irq_service_dcn21.h" + +#include "ivsrcid/dcn/irqsrcs_dcn_1_0.h" + +enum dc_irq_source to_dal_irq_source_dcn21( + struct irq_service *irq_service, + uint32_t src_id, + uint32_t ext_id) +{ + switch (src_id) { + case DCN_1_0__SRCID__DC_D1_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK1; + case DCN_1_0__SRCID__DC_D2_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK2; + case DCN_1_0__SRCID__DC_D3_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK3; + case DCN_1_0__SRCID__DC_D4_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK4; + case DCN_1_0__SRCID__DC_D5_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK5; + case DCN_1_0__SRCID__DC_D6_OTG_VSTARTUP: + return DC_IRQ_SOURCE_VBLANK6; + case DCN_1_0__SRCID__HUBP0_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP1; + case DCN_1_0__SRCID__HUBP1_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP2; + case DCN_1_0__SRCID__HUBP2_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP3; + case DCN_1_0__SRCID__HUBP3_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP4; + case DCN_1_0__SRCID__HUBP4_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP5; + case DCN_1_0__SRCID__HUBP5_FLIP_INTERRUPT: + return DC_IRQ_SOURCE_PFLIP6; + case DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE1; + case DCN_1_0__SRCID__OTG1_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE2; + case DCN_1_0__SRCID__OTG2_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE3; + case DCN_1_0__SRCID__OTG3_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE4; + case DCN_1_0__SRCID__OTG4_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE5; + case DCN_1_0__SRCID__OTG5_IHC_V_UPDATE_NO_LOCK_INTERRUPT: + return DC_IRQ_SOURCE_VUPDATE6; + + case DCN_1_0__SRCID__DC_HPD1_INT: + /* generic src_id for all HPD and HPDRX interrupts */ + switch (ext_id) { + case DCN_1_0__CTXID__DC_HPD1_INT: + return DC_IRQ_SOURCE_HPD1; + case DCN_1_0__CTXID__DC_HPD2_INT: + return DC_IRQ_SOURCE_HPD2; + case DCN_1_0__CTXID__DC_HPD3_INT: + return DC_IRQ_SOURCE_HPD3; + case DCN_1_0__CTXID__DC_HPD4_INT: + return DC_IRQ_SOURCE_HPD4; + case DCN_1_0__CTXID__DC_HPD5_INT: + return DC_IRQ_SOURCE_HPD5; + case DCN_1_0__CTXID__DC_HPD6_INT: + return DC_IRQ_SOURCE_HPD6; + case DCN_1_0__CTXID__DC_HPD1_RX_INT: + return DC_IRQ_SOURCE_HPD1RX; + case DCN_1_0__CTXID__DC_HPD2_RX_INT: + return DC_IRQ_SOURCE_HPD2RX; + case DCN_1_0__CTXID__DC_HPD3_RX_INT: + return DC_IRQ_SOURCE_HPD3RX; + case DCN_1_0__CTXID__DC_HPD4_RX_INT: + return DC_IRQ_SOURCE_HPD4RX; + case DCN_1_0__CTXID__DC_HPD5_RX_INT: + return DC_IRQ_SOURCE_HPD5RX; + case DCN_1_0__CTXID__DC_HPD6_RX_INT: + return DC_IRQ_SOURCE_HPD6RX; + default: + return DC_IRQ_SOURCE_INVALID; + } + break; + + default: + break; + } + return DC_IRQ_SOURCE_INVALID; +} + +static bool hpd_ack( + struct irq_service *irq_service, + const struct irq_source_info *info) +{ + uint32_t addr = info->status_reg; + uint32_t value = dm_read_reg(irq_service->ctx, addr); + uint32_t current_status = + get_reg_field_value( + value, + HPD0_DC_HPD_INT_STATUS, + DC_HPD_SENSE_DELAYED); + + dal_irq_service_ack_generic(irq_service, info); + + value = dm_read_reg(irq_service->ctx, info->enable_reg); + + set_reg_field_value( + value, + current_status ? 0 : 1, + HPD0_DC_HPD_INT_CONTROL, + DC_HPD_INT_POLARITY); + + dm_write_reg(irq_service->ctx, info->enable_reg, value); + + return true; +} + +static const struct irq_source_info_funcs hpd_irq_info_funcs = { + .set = NULL, + .ack = hpd_ack +}; + +static const struct irq_source_info_funcs hpd_rx_irq_info_funcs = { + .set = NULL, + .ack = NULL +}; + +static const struct irq_source_info_funcs pflip_irq_info_funcs = { + .set = NULL, + .ack = NULL +}; + +static const struct irq_source_info_funcs vblank_irq_info_funcs = { + .set = NULL, + .ack = NULL +}; + +#undef BASE_INNER +#define BASE_INNER(seg) DMU_BASE__INST0_SEG ## seg + +/* compile time expand base address. */ +#define BASE(seg) \ + BASE_INNER(seg) + + +#define SRI(reg_name, block, id)\ + BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## id ## _ ## reg_name + + +#define IRQ_REG_ENTRY(block, reg_num, reg1, mask1, reg2, mask2)\ + .enable_reg = SRI(reg1, block, reg_num),\ + .enable_mask = \ + block ## reg_num ## _ ## reg1 ## __ ## mask1 ## _MASK,\ + .enable_value = {\ + block ## reg_num ## _ ## reg1 ## __ ## mask1 ## _MASK,\ + ~block ## reg_num ## _ ## reg1 ## __ ## mask1 ## _MASK \ + },\ + .ack_reg = SRI(reg2, block, reg_num),\ + .ack_mask = \ + block ## reg_num ## _ ## reg2 ## __ ## mask2 ## _MASK,\ + .ack_value = \ + block ## reg_num ## _ ## reg2 ## __ ## mask2 ## _MASK \ + + + +#define hpd_int_entry(reg_num)\ + [DC_IRQ_SOURCE_HPD1 + reg_num] = {\ + IRQ_REG_ENTRY(HPD, reg_num,\ + DC_HPD_INT_CONTROL, DC_HPD_INT_EN,\ + DC_HPD_INT_CONTROL, DC_HPD_INT_ACK),\ + .status_reg = SRI(DC_HPD_INT_STATUS, HPD, reg_num),\ + .funcs = &hpd_irq_info_funcs\ + } + +#define hpd_rx_int_entry(reg_num)\ + [DC_IRQ_SOURCE_HPD1RX + reg_num] = {\ + IRQ_REG_ENTRY(HPD, reg_num,\ + DC_HPD_INT_CONTROL, DC_HPD_RX_INT_EN,\ + DC_HPD_INT_CONTROL, DC_HPD_RX_INT_ACK),\ + .status_reg = SRI(DC_HPD_INT_STATUS, HPD, reg_num),\ + .funcs = &hpd_rx_irq_info_funcs\ + } +#define pflip_int_entry(reg_num)\ + [DC_IRQ_SOURCE_PFLIP1 + reg_num] = {\ + IRQ_REG_ENTRY(HUBPREQ, reg_num,\ + DCSURF_SURFACE_FLIP_INTERRUPT, SURFACE_FLIP_INT_MASK,\ + DCSURF_SURFACE_FLIP_INTERRUPT, SURFACE_FLIP_CLEAR),\ + .funcs = &pflip_irq_info_funcs\ + } + +#define vupdate_int_entry(reg_num)\ + [DC_IRQ_SOURCE_VUPDATE1 + reg_num] = {\ + IRQ_REG_ENTRY(OTG, reg_num,\ + OTG_GLOBAL_SYNC_STATUS, VUPDATE_INT_EN,\ + OTG_GLOBAL_SYNC_STATUS, VUPDATE_EVENT_CLEAR),\ + .funcs = &vblank_irq_info_funcs\ + } + +#define vblank_int_entry(reg_num)\ + [DC_IRQ_SOURCE_VBLANK1 + reg_num] = {\ + IRQ_REG_ENTRY(OTG, reg_num,\ + OTG_GLOBAL_SYNC_STATUS, VSTARTUP_INT_EN,\ + OTG_GLOBAL_SYNC_STATUS, VSTARTUP_EVENT_CLEAR),\ + .funcs = &vblank_irq_info_funcs\ + } + +#define dummy_irq_entry() \ + {\ + .funcs = &dummy_irq_info_funcs\ + } + +#define i2c_int_entry(reg_num) \ + [DC_IRQ_SOURCE_I2C_DDC ## reg_num] = dummy_irq_entry() + +#define dp_sink_int_entry(reg_num) \ + [DC_IRQ_SOURCE_DPSINK ## reg_num] = dummy_irq_entry() + +#define gpio_pad_int_entry(reg_num) \ + [DC_IRQ_SOURCE_GPIOPAD ## reg_num] = dummy_irq_entry() + +#define dc_underflow_int_entry(reg_num) \ + [DC_IRQ_SOURCE_DC ## reg_num ## UNDERFLOW] = dummy_irq_entry() + +static const struct irq_source_info_funcs dummy_irq_info_funcs = { + .set = dal_irq_service_dummy_set, + .ack = dal_irq_service_dummy_ack +}; + +static const struct irq_source_info +irq_source_info_dcn21[DAL_IRQ_SOURCES_NUMBER] = { + [DC_IRQ_SOURCE_INVALID] = dummy_irq_entry(), + hpd_int_entry(0), + hpd_int_entry(1), + hpd_int_entry(2), + hpd_int_entry(3), + hpd_int_entry(4), + hpd_rx_int_entry(0), + hpd_rx_int_entry(1), + hpd_rx_int_entry(2), + hpd_rx_int_entry(3), + hpd_rx_int_entry(4), + i2c_int_entry(1), + i2c_int_entry(2), + i2c_int_entry(3), + i2c_int_entry(4), + i2c_int_entry(5), + i2c_int_entry(6), + dp_sink_int_entry(1), + dp_sink_int_entry(2), + dp_sink_int_entry(3), + dp_sink_int_entry(4), + dp_sink_int_entry(5), + dp_sink_int_entry(6), + [DC_IRQ_SOURCE_TIMER] = dummy_irq_entry(), + pflip_int_entry(0), + pflip_int_entry(1), + pflip_int_entry(2), + pflip_int_entry(3), + [DC_IRQ_SOURCE_PFLIP5] = dummy_irq_entry(), + [DC_IRQ_SOURCE_PFLIP6] = dummy_irq_entry(), + [DC_IRQ_SOURCE_PFLIP_UNDERLAY0] = dummy_irq_entry(), + gpio_pad_int_entry(0), + gpio_pad_int_entry(1), + gpio_pad_int_entry(2), + gpio_pad_int_entry(3), + gpio_pad_int_entry(4), + gpio_pad_int_entry(5), + gpio_pad_int_entry(6), + gpio_pad_int_entry(7), + gpio_pad_int_entry(8), + gpio_pad_int_entry(9), + gpio_pad_int_entry(10), + gpio_pad_int_entry(11), + gpio_pad_int_entry(12), + gpio_pad_int_entry(13), + gpio_pad_int_entry(14), + gpio_pad_int_entry(15), + gpio_pad_int_entry(16), + gpio_pad_int_entry(17), + gpio_pad_int_entry(18), + gpio_pad_int_entry(19), + gpio_pad_int_entry(20), + gpio_pad_int_entry(21), + gpio_pad_int_entry(22), + gpio_pad_int_entry(23), + gpio_pad_int_entry(24), + gpio_pad_int_entry(25), + gpio_pad_int_entry(26), + gpio_pad_int_entry(27), + gpio_pad_int_entry(28), + gpio_pad_int_entry(29), + gpio_pad_int_entry(30), + dc_underflow_int_entry(1), + dc_underflow_int_entry(2), + dc_underflow_int_entry(3), + dc_underflow_int_entry(4), + dc_underflow_int_entry(5), + dc_underflow_int_entry(6), + [DC_IRQ_SOURCE_DMCU_SCP] = dummy_irq_entry(), + [DC_IRQ_SOURCE_VBIOS_SW] = dummy_irq_entry(), + vupdate_int_entry(0), + vupdate_int_entry(1), + vupdate_int_entry(2), + vupdate_int_entry(3), + vupdate_int_entry(4), + vupdate_int_entry(5), + vblank_int_entry(0), + vblank_int_entry(1), + vblank_int_entry(2), + vblank_int_entry(3), + vblank_int_entry(4), + vblank_int_entry(5), +}; + +static const struct irq_service_funcs irq_service_funcs_dcn21 = { + .to_dal_irq_source = to_dal_irq_source_dcn21 +}; + +static void dcn21_irq_construct( + struct irq_service *irq_service, + struct irq_service_init_data *init_data) +{ + dal_irq_service_construct(irq_service, init_data); + + irq_service->info = irq_source_info_dcn21; + irq_service->funcs = &irq_service_funcs_dcn21; +} + +struct irq_service *dal_irq_service_dcn21_create( + struct irq_service_init_data *init_data) +{ + struct irq_service *irq_service = kzalloc(sizeof(*irq_service), + GFP_KERNEL); + + if (!irq_service) + return NULL; + + dcn21_irq_construct(irq_service, init_data); + return irq_service; +} diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.h b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.h new file mode 100644 index 000000000000..da2bd0e93d7a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.h @@ -0,0 +1,34 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DAL_IRQ_SERVICE_DCN21_H__ +#define __DAL_IRQ_SERVICE_DCN21_H__ + +#include "../irq_service.h" + +struct irq_service *dal_irq_service_dcn21_create( + struct irq_service_init_data *init_data); + +#endif diff --git a/drivers/gpu/drm/amd/display/dc/irq/irq_service.c b/drivers/gpu/drm/amd/display/dc/irq/irq_service.c index 0878550a8178..33053b9fe6bd 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/irq_service.c +++ b/drivers/gpu/drm/amd/display/dc/irq/irq_service.c @@ -38,7 +38,7 @@ #include "dce120/irq_service_dce120.h" -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) #include "dcn10/irq_service_dcn10.h" #endif diff --git a/drivers/gpu/drm/amd/display/dc/os_types.h b/drivers/gpu/drm/amd/display/dc/os_types.h index 30ec80ac6fc8..c34eba19860a 100644 --- a/drivers/gpu/drm/amd/display/dc/os_types.h +++ b/drivers/gpu/drm/amd/display/dc/os_types.h @@ -1,5 +1,6 @@ /* * Copyright 2012-16 Advanced Micro Devices, Inc. + * Copyright 2019 Raptor Engineering, LLC * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -29,6 +30,7 @@ #include <linux/kgdb.h> #include <linux/kref.h> #include <linux/types.h> +#include <linux/slab.h> #include <asm/byteorder.h> @@ -48,8 +50,39 @@ #define dm_error(fmt, ...) DRM_ERROR(fmt, ##__VA_ARGS__) -#if defined(CONFIG_DRM_AMD_DC_DCN1_0) +#if defined(CONFIG_DRM_AMD_DC_DCN) +#if defined(CONFIG_X86) #include <asm/fpu/api.h> +#define DC_FP_START() kernel_fpu_begin() +#define DC_FP_END() kernel_fpu_end() +#elif defined(CONFIG_PPC64) +#include <asm/switch_to.h> +#include <asm/cputable.h> +#define DC_FP_START() { \ + if (cpu_has_feature(CPU_FTR_VSX_COMP)) { \ + preempt_disable(); \ + enable_kernel_vsx(); \ + } else if (cpu_has_feature(CPU_FTR_ALTIVEC_COMP)) { \ + preempt_disable(); \ + enable_kernel_altivec(); \ + } else if (!cpu_has_feature(CPU_FTR_FPU_UNAVAILABLE)) { \ + preempt_disable(); \ + enable_kernel_fp(); \ + } \ +} +#define DC_FP_END() { \ + if (cpu_has_feature(CPU_FTR_VSX_COMP)) { \ + disable_kernel_vsx(); \ + preempt_enable(); \ + } else if (cpu_has_feature(CPU_FTR_ALTIVEC_COMP)) { \ + disable_kernel_altivec(); \ + preempt_enable(); \ + } else if (!cpu_has_feature(CPU_FTR_FPU_UNAVAILABLE)) { \ + disable_kernel_fp(); \ + preempt_enable(); \ + } \ +} +#endif #endif /* diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c index c9a6dd878d9b..b8040da94b9d 100644 --- a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c @@ -32,6 +32,7 @@ static void virtual_stream_encoder_dp_set_stream_attribute( struct stream_encoder *enc, struct dc_crtc_timing *crtc_timing, enum dc_color_space output_color_space, + bool use_vsc_sdp_for_colorimetry, uint32_t enable_sdp_splitting) {} static void virtual_stream_encoder_hdmi_set_stream_attribute( @@ -77,22 +78,18 @@ static void virtual_audio_mute_control( struct stream_encoder *enc, bool mute) {} -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT +static void virtual_stream_encoder_reset_hdmi_stream_attribute( + struct stream_encoder *enc) +{} + static void virtual_enc_dp_set_odm_combine( struct stream_encoder *enc, bool odm_combine) {} -#endif -#endif static const struct stream_encoder_funcs virtual_str_enc_funcs = { -#ifdef CONFIG_DRM_AMD_DC_DCN2_0 -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT .dp_set_odm_combine = virtual_enc_dp_set_odm_combine, -#endif -#endif .dp_set_stream_attribute = virtual_stream_encoder_dp_set_stream_attribute, .hdmi_set_stream_attribute = @@ -116,6 +113,7 @@ static const struct stream_encoder_funcs virtual_str_enc_funcs = { .audio_mute_control = virtual_audio_mute_control, .set_avmute = virtual_stream_encoder_set_avmute, + .hdmi_reset_stream_attribute = virtual_stream_encoder_reset_hdmi_stream_attribute, }; bool virtual_stream_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h new file mode 100644 index 000000000000..cd9532b4f14d --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -0,0 +1,289 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_CMD_H_ +#define _DMUB_CMD_H_ + +#include "dmub_types.h" +#include "dmub_cmd_dal.h" +#include "dmub_cmd_vbios.h" +#include "atomfirmware.h" + +#define DMUB_RB_CMD_SIZE 64 +#define DMUB_RB_MAX_ENTRY 128 +#define DMUB_RB_SIZE (DMUB_RB_CMD_SIZE * DMUB_RB_MAX_ENTRY) +#define REG_SET_MASK 0xFFFF + + +/* + * Command IDs should be treated as stable ABI. + * Do not reuse or modify IDs. + */ + +enum dmub_cmd_type { + DMUB_CMD__NULL = 0, + DMUB_CMD__REG_SEQ_READ_MODIFY_WRITE = 1, + DMUB_CMD__REG_SEQ_FIELD_UPDATE_SEQ = 2, + DMUB_CMD__REG_SEQ_BURST_WRITE = 3, + DMUB_CMD__REG_REG_WAIT = 4, + DMUB_CMD__PLAT_54186_WA = 5, + DMUB_CMD__PSR = 64, + DMUB_CMD__VBIOS = 128, +}; + +#pragma pack(push, 1) + +struct dmub_cmd_header { + unsigned int type : 8; + unsigned int sub_type : 8; + unsigned int reserved0 : 8; + unsigned int payload_bytes : 6; /* up to 60 bytes */ + unsigned int reserved1 : 2; +}; + +/* + * Read modify write + * + * 60 payload bytes can hold up to 5 sets of read modify writes, + * each take 3 dwords. + * + * number of sequences = header.payload_bytes / sizeof(struct dmub_cmd_read_modify_write_sequence) + * + * modify_mask = 0xffff'ffff means all fields are going to be updated. in this case + * command parser will skip the read and we can use modify_mask = 0xffff'ffff as reg write + */ +struct dmub_cmd_read_modify_write_sequence { + uint32_t addr; + uint32_t modify_mask; + uint32_t modify_value; +}; + +#define DMUB_READ_MODIFY_WRITE_SEQ__MAX 5 +struct dmub_rb_cmd_read_modify_write { + struct dmub_cmd_header header; // type = DMUB_CMD__REG_SEQ_READ_MODIFY_WRITE + struct dmub_cmd_read_modify_write_sequence seq[DMUB_READ_MODIFY_WRITE_SEQ__MAX]; +}; + +/* + * Update a register with specified masks and values sequeunce + * + * 60 payload bytes can hold address + up to 7 sets of mask/value combo, each take 2 dword + * + * number of field update sequence = (header.payload_bytes - sizeof(addr)) / sizeof(struct read_modify_write_sequence) + * + * + * USE CASE: + * 1. auto-increment register where additional read would update pointer and produce wrong result + * 2. toggle a bit without read in the middle + */ + +struct dmub_cmd_reg_field_update_sequence { + uint32_t modify_mask; // 0xffff'ffff to skip initial read + uint32_t modify_value; +}; + +#define DMUB_REG_FIELD_UPDATE_SEQ__MAX 7 + +struct dmub_rb_cmd_reg_field_update_sequence { + struct dmub_cmd_header header; + uint32_t addr; + struct dmub_cmd_reg_field_update_sequence seq[DMUB_REG_FIELD_UPDATE_SEQ__MAX]; +}; + + +/* + * Burst write + * + * support use case such as writing out LUTs. + * + * 60 payload bytes can hold up to 14 values to write to given address + * + * number of payload = header.payload_bytes / sizeof(struct read_modify_write_sequence) + */ +#define DMUB_BURST_WRITE_VALUES__MAX 14 +struct dmub_rb_cmd_burst_write { + struct dmub_cmd_header header; // type = DMUB_CMD__REG_SEQ_BURST_WRITE + uint32_t addr; + uint32_t write_values[DMUB_BURST_WRITE_VALUES__MAX]; +}; + + +struct dmub_rb_cmd_common { + struct dmub_cmd_header header; + uint8_t cmd_buffer[DMUB_RB_CMD_SIZE - sizeof(struct dmub_cmd_header)]; +}; + +struct dmub_cmd_reg_wait_data { + uint32_t addr; + uint32_t mask; + uint32_t condition_field_value; + uint32_t time_out_us; +}; + +struct dmub_rb_cmd_reg_wait { + struct dmub_cmd_header header; + struct dmub_cmd_reg_wait_data reg_wait; +}; + +#ifndef PHYSICAL_ADDRESS_LOC +#define PHYSICAL_ADDRESS_LOC union large_integer +#endif + +struct dmub_cmd_PLAT_54186_wa { + uint32_t DCSURF_SURFACE_CONTROL; + uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH; + uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS; + uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C; + uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_C; + struct { + uint8_t hubp_inst : 4; + uint8_t tmz_surface : 1; + uint8_t immediate :1; + uint8_t vmid : 4; + uint8_t grph_stereo : 1; + uint32_t reserved : 21; + } flip_params; + uint32_t reserved[9]; +}; + +struct dmub_rb_cmd_PLAT_54186_wa { + struct dmub_cmd_header header; + struct dmub_cmd_PLAT_54186_wa flip; +}; + +struct dmub_cmd_digx_encoder_control_data { + union dig_encoder_control_parameters_v1_5 dig; +}; + +struct dmub_rb_cmd_digx_encoder_control { + struct dmub_cmd_header header; + struct dmub_cmd_digx_encoder_control_data encoder_control; +}; + +struct dmub_cmd_set_pixel_clock_data { + struct set_pixel_clock_parameter_v1_7 clk; +}; + +struct dmub_rb_cmd_set_pixel_clock { + struct dmub_cmd_header header; + struct dmub_cmd_set_pixel_clock_data pixel_clock; +}; + +struct dmub_cmd_enable_disp_power_gating_data { + struct enable_disp_power_gating_parameters_v2_1 pwr; +}; + +struct dmub_rb_cmd_enable_disp_power_gating { + struct dmub_cmd_header header; + struct dmub_cmd_enable_disp_power_gating_data power_gating; +}; + +struct dmub_cmd_dig1_transmitter_control_data { + struct dig_transmitter_control_parameters_v1_6 dig; +}; + +struct dmub_rb_cmd_dig1_transmitter_control { + struct dmub_cmd_header header; + struct dmub_cmd_dig1_transmitter_control_data transmitter_control; +}; + +struct dmub_rb_cmd_dpphy_init { + struct dmub_cmd_header header; + uint8_t reserved[60]; +}; + +struct dmub_cmd_psr_copy_settings_data { + uint16_t psr_level; + uint8_t hubp_inst; + uint8_t dpp_inst; + uint8_t mpcc_inst; + uint8_t opp_inst; + uint8_t otg_inst; + uint8_t digfe_inst; + uint8_t digbe_inst; + uint8_t dpphy_inst; + uint8_t aux_inst; + uint8_t hyst_frames; + uint8_t hyst_lines; + uint8_t phy_num; + uint8_t phy_type; + uint8_t aux_repeat; + uint8_t smu_optimizations_en; + uint8_t skip_wait_for_pll_lock; + uint8_t frame_delay; + uint8_t smu_phy_id; + uint8_t num_of_controllers; + uint8_t link_rate; + uint8_t frame_cap_ind; +}; + +struct dmub_rb_cmd_psr_copy_settings { + struct dmub_cmd_header header; + struct dmub_cmd_psr_copy_settings_data psr_copy_settings_data; +}; + +struct dmub_cmd_psr_set_level_data { + uint16_t psr_level; +}; + +struct dmub_rb_cmd_psr_set_level { + struct dmub_cmd_header header; + struct dmub_cmd_psr_set_level_data psr_set_level_data; +}; + +struct dmub_rb_cmd_psr_enable { + struct dmub_cmd_header header; +}; + +struct dmub_cmd_psr_setup_data { + enum psr_version version; // PSR version 1 or 2 +}; + +struct dmub_rb_cmd_psr_setup { + struct dmub_cmd_header header; + struct dmub_cmd_psr_setup_data psr_setup_data; +}; + +union dmub_rb_cmd { + struct dmub_rb_cmd_read_modify_write read_modify_write; + struct dmub_rb_cmd_reg_field_update_sequence reg_field_update_seq; + struct dmub_rb_cmd_burst_write burst_write; + struct dmub_rb_cmd_reg_wait reg_wait; + struct dmub_rb_cmd_common cmd_common; + struct dmub_rb_cmd_digx_encoder_control digx_encoder_control; + struct dmub_rb_cmd_set_pixel_clock set_pixel_clock; + struct dmub_rb_cmd_enable_disp_power_gating enable_disp_power_gating; + struct dmub_rb_cmd_dpphy_init dpphy_init; + struct dmub_rb_cmd_dig1_transmitter_control dig1_transmitter_control; + struct dmub_rb_cmd_psr_enable psr_enable; + struct dmub_rb_cmd_psr_copy_settings psr_copy_settings; + struct dmub_rb_cmd_psr_set_level psr_set_level; + struct dmub_rb_cmd_PLAT_54186_wa PLAT_54186_wa; + struct dmub_rb_cmd_psr_setup psr_setup; +}; + +#pragma pack(pop) + +#endif /* _DMUB_CMD_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h new file mode 100644 index 000000000000..7b69eb37f762 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h @@ -0,0 +1,48 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_CMD_DAL_H_ +#define _DMUB_CMD_DAL_H_ + +/* + * Command IDs should be treated as stable ABI. + * Do not reuse or modify IDs. + */ + +enum dmub_cmd_psr_type { + DMUB_CMD__PSR_SETUP = 0, + DMUB_CMD__PSR_COPY_SETTINGS = 1, + DMUB_CMD__PSR_ENABLE = 2, + DMUB_CMD__PSR_DISABLE = 3, + DMUB_CMD__PSR_SET_LEVEL = 4, +}; + +enum psr_version { + PSR_VERSION_1 = 0x10, // PSR Version 1 + PSR_VERSION_2 = 0x20, // PSR Version 2, includes selective update + PSR_VERSION_2_Y_COORD = 0x21, // PSR Version 2, includes Y-coordinate support for SU +}; + +#endif /* _DMUB_CMD_DAL_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_vbios.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_vbios.h new file mode 100644 index 000000000000..b6deb8e2590f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_vbios.h @@ -0,0 +1,41 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_CMD_VBIOS_H_ +#define _DMUB_CMD_VBIOS_H_ + +/* + * Command IDs should be treated as stable ABI. + * Do not reuse or modify IDs. + */ + +enum dmub_cmd_vbios_type { + DMUB_CMD__VBIOS_DIGX_ENCODER_CONTROL = 0, + DMUB_CMD__VBIOS_DIG1_TRANSMITTER_CONTROL = 1, + DMUB_CMD__VBIOS_SET_PIXEL_CLOCK = 2, + DMUB_CMD__VBIOS_ENABLE_DISP_POWER_GATING = 3, +}; + +#endif /* _DMUB_CMD_VBIOS_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_fw_meta.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_fw_meta.h new file mode 100644 index 000000000000..242ec257998c --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_fw_meta.h @@ -0,0 +1,63 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef _DMUB_META_H_ +#define _DMUB_META_H_ + +#include "dmub_types.h" + +#pragma pack(push, 1) + +/* Magic value for identifying dmub_fw_meta_info */ +#define DMUB_FW_META_MAGIC 0x444D5542 + +/* Offset from the end of the file to the dmub_fw_meta_info */ +#define DMUB_FW_META_OFFSET 0x24 + +/** + * struct dmub_fw_meta_info - metadata associated with fw binary + * + * NOTE: This should be considered a stable API. Fields should + * not be repurposed or reordered. New fields should be + * added instead to extend the structure. + * + * @magic_value: magic value identifying DMUB firmware meta info + * @fw_region_size: size of the firmware state region + * @trace_buffer_size: size of the tracebuffer region + */ +struct dmub_fw_meta_info { + uint32_t magic_value; + uint32_t fw_region_size; + uint32_t trace_buffer_size; +}; + +/* Ensure that the structure remains 64 bytes. */ +union dmub_fw_meta { + struct dmub_fw_meta_info info; + uint8_t reserved[64]; +}; + +#pragma pack(pop) + +#endif /* _DMUB_META_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_rb.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_rb.h new file mode 100644 index 000000000000..df875fdd2ab0 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_rb.h @@ -0,0 +1,154 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_RB_H_ +#define _DMUB_RB_H_ + +#include "dmub_types.h" +#include "dmub_cmd.h" + +#if defined(__cplusplus) +extern "C" { +#endif + +struct dmub_cmd_header; + +struct dmub_rb_init_params { + void *ctx; + void *base_address; + uint32_t capacity; +}; + +struct dmub_rb { + void *base_address; + uint32_t data_count; + uint32_t rptr; + uint32_t wrpt; + uint32_t capacity; + + void *ctx; + void *dmub; +}; + + +static inline bool dmub_rb_empty(struct dmub_rb *rb) +{ + return (rb->wrpt == rb->rptr); +} + +static inline bool dmub_rb_full(struct dmub_rb *rb) +{ + uint32_t data_count; + + if (rb->wrpt >= rb->rptr) + data_count = rb->wrpt - rb->rptr; + else + data_count = rb->capacity - (rb->rptr - rb->wrpt); + + return (data_count == (rb->capacity - DMUB_RB_CMD_SIZE)); +} + +static inline bool dmub_rb_push_front(struct dmub_rb *rb, + const struct dmub_cmd_header *cmd) +{ + uint64_t volatile *dst = (uint64_t volatile *)(rb->base_address) + rb->wrpt / sizeof(uint64_t); + const uint64_t *src = (const uint64_t *)cmd; + int i; + + if (dmub_rb_full(rb)) + return false; + + // copying data + for (i = 0; i < DMUB_RB_CMD_SIZE / sizeof(uint64_t); i++) + *dst++ = *src++; + + rb->wrpt += DMUB_RB_CMD_SIZE; + + if (rb->wrpt >= rb->capacity) + rb->wrpt %= rb->capacity; + + return true; +} + +static inline bool dmub_rb_front(struct dmub_rb *rb, + struct dmub_cmd_header *cmd) +{ + uint8_t *rd_ptr = (uint8_t *)rb->base_address + rb->rptr; + + if (dmub_rb_empty(rb)) + return false; + + dmub_memcpy(cmd, rd_ptr, DMUB_RB_CMD_SIZE); + + return true; +} + +static inline bool dmub_rb_pop_front(struct dmub_rb *rb) +{ + if (dmub_rb_empty(rb)) + return false; + + rb->rptr += DMUB_RB_CMD_SIZE; + + if (rb->rptr >= rb->capacity) + rb->rptr %= rb->capacity; + + return true; +} + +static inline void dmub_rb_flush_pending(const struct dmub_rb *rb) +{ + uint32_t rptr = rb->rptr; + uint32_t wptr = rb->wrpt; + + while (rptr != wptr) { + uint64_t volatile *data = (uint64_t volatile *)rb->base_address + rptr / sizeof(uint64_t); + //uint64_t volatile *p = (uint64_t volatile *)data; + uint64_t temp; + int i; + + for (i = 0; i < DMUB_RB_CMD_SIZE / sizeof(uint64_t); i++) + temp = *data++; + + rptr += DMUB_RB_CMD_SIZE; + if (rptr >= rb->capacity) + rptr %= rb->capacity; + } +} + +static inline void dmub_rb_init(struct dmub_rb *rb, + struct dmub_rb_init_params *init_params) +{ + rb->base_address = init_params->base_address; + rb->capacity = init_params->capacity; + rb->rptr = 0; + rb->wrpt = 0; +} + +#if defined(__cplusplus) +} +#endif + +#endif /* _DMUB_RB_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h new file mode 100644 index 000000000000..f8917594036a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h @@ -0,0 +1,523 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_SRV_H_ +#define _DMUB_SRV_H_ + +/** + * DOC: DMUB interface and operation + * + * DMUB is the interface to the display DMCUB microcontroller on DCN hardware. + * It delegates hardware initialization and command submission to the + * microcontroller. DMUB is the shortname for DMCUB. + * + * This interface is not thread-safe. Ensure that all access to the interface + * is properly synchronized by the caller. + * + * Initialization and usage of the DMUB service should be done in the + * steps given below: + * + * 1. dmub_srv_create() + * 2. dmub_srv_has_hw_support() + * 3. dmub_srv_calc_region_info() + * 4. dmub_srv_hw_init() + * + * The call to dmub_srv_create() is required to use the server. + * + * The calls to dmub_srv_has_hw_support() and dmub_srv_calc_region_info() + * are helpers to query cache window size and allocate framebuffer(s) + * for the cache windows. + * + * The call to dmub_srv_hw_init() programs the DMCUB registers to prepare + * for command submission. Commands can be queued via dmub_srv_cmd_queue() + * and executed via dmub_srv_cmd_execute(). + * + * If the queue is full the dmub_srv_wait_for_idle() call can be used to + * wait until the queue has been cleared. + * + * Destroying the DMUB service can be done by calling dmub_srv_destroy(). + * This does not clear DMUB hardware state, only software state. + * + * The interface is intended to be standalone and should not depend on any + * other component within DAL. + */ + +#include "dmub_types.h" +#include "dmub_cmd.h" +#include "dmub_rb.h" + +#if defined(__cplusplus) +extern "C" { +#endif + +/* Forward declarations */ +struct dmub_srv; +struct dmub_cmd_header; +struct dmub_srv_common_regs; + +/* enum dmub_status - return code for dmcub functions */ +enum dmub_status { + DMUB_STATUS_OK = 0, + DMUB_STATUS_NO_CTX, + DMUB_STATUS_QUEUE_FULL, + DMUB_STATUS_TIMEOUT, + DMUB_STATUS_INVALID, +}; + +/* enum dmub_asic - dmub asic identifier */ +enum dmub_asic { + DMUB_ASIC_NONE = 0, + DMUB_ASIC_DCN20, + DMUB_ASIC_DCN21, + DMUB_ASIC_MAX, +}; + +/* enum dmub_window_id - dmub window identifier */ +enum dmub_window_id { + DMUB_WINDOW_0_INST_CONST = 0, + DMUB_WINDOW_1_STACK, + DMUB_WINDOW_2_BSS_DATA, + DMUB_WINDOW_3_VBIOS, + DMUB_WINDOW_4_MAILBOX, + DMUB_WINDOW_5_TRACEBUFF, + DMUB_WINDOW_6_FW_STATE, + DMUB_WINDOW_7_RESERVED, + DMUB_WINDOW_TOTAL, +}; + +/** + * struct dmub_region - dmub hw memory region + * @base: base address for region, must be 256 byte aligned + * @top: top address for region + */ +struct dmub_region { + uint32_t base; + uint32_t top; +}; + +/** + * struct dmub_window - dmub hw cache window + * @off: offset to the fb memory in gpu address space + * @r: region in uc address space for cache window + */ +struct dmub_window { + union dmub_addr offset; + struct dmub_region region; +}; + +/** + * struct dmub_fb - defines a dmub framebuffer memory region + * @cpu_addr: cpu virtual address for the region, NULL if invalid + * @gpu_addr: gpu virtual address for the region, NULL if invalid + * @size: size of the region in bytes, zero if invalid + */ +struct dmub_fb { + void *cpu_addr; + uint64_t gpu_addr; + uint32_t size; +}; + +/** + * struct dmub_srv_region_params - params used for calculating dmub regions + * @inst_const_size: size of the fw inst const section + * @bss_data_size: size of the fw bss data section + * @vbios_size: size of the vbios data + * @fw_bss_data: raw firmware bss data section + */ +struct dmub_srv_region_params { + uint32_t inst_const_size; + uint32_t bss_data_size; + uint32_t vbios_size; + const uint8_t *fw_bss_data; +}; + +/** + * struct dmub_srv_region_info - output region info from the dmub service + * @fb_size: required minimum fb size for all regions, aligned to 4096 bytes + * @num_regions: number of regions used by the dmub service + * @regions: region info + * + * The regions are aligned such that they can be all placed within the + * same framebuffer but they can also be placed into different framebuffers. + * + * The size of each region can be calculated by the caller: + * size = reg.top - reg.base + * + * Care must be taken when performing custom allocations to ensure that each + * region base address is 256 byte aligned. + */ +struct dmub_srv_region_info { + uint32_t fb_size; + uint8_t num_regions; + struct dmub_region regions[DMUB_WINDOW_TOTAL]; +}; + +/** + * struct dmub_srv_fb_params - parameters used for driver fb setup + * @region_info: region info calculated by dmub service + * @cpu_addr: base cpu address for the framebuffer + * @gpu_addr: base gpu virtual address for the framebuffer + */ +struct dmub_srv_fb_params { + const struct dmub_srv_region_info *region_info; + void *cpu_addr; + uint64_t gpu_addr; +}; + +/** + * struct dmub_srv_fb_info - output fb info from the dmub service + * @num_fbs: number of required dmub framebuffers + * @fbs: fb data for each region + * + * Output from the dmub service helper that can be used by the + * driver to prepare dmub_fb that can be passed into the dmub + * hw init service. + * + * Assumes that all regions are within the same framebuffer + * and have been setup according to the region_info generated + * by the dmub service. + */ +struct dmub_srv_fb_info { + uint8_t num_fb; + struct dmub_fb fb[DMUB_WINDOW_TOTAL]; +}; + +/** + * struct dmub_srv_base_funcs - Driver specific base callbacks + */ +struct dmub_srv_base_funcs { + /** + * @reg_read: + * + * Hook for reading a register. + * + * Return: The 32-bit register value from the given address. + */ + uint32_t (*reg_read)(void *ctx, uint32_t address); + + /** + * @reg_write: + * + * Hook for writing a value to the register specified by address. + */ + void (*reg_write)(void *ctx, uint32_t address, uint32_t value); +}; + +/** + * struct dmub_srv_hw_funcs - hardware sequencer funcs for dmub + */ +struct dmub_srv_hw_funcs { + /* private: internal use only */ + + void (*init)(struct dmub_srv *dmub); + + void (*reset)(struct dmub_srv *dmub); + + void (*reset_release)(struct dmub_srv *dmub); + + void (*backdoor_load)(struct dmub_srv *dmub, + const struct dmub_window *cw0, + const struct dmub_window *cw1); + + void (*setup_windows)(struct dmub_srv *dmub, + const struct dmub_window *cw2, + const struct dmub_window *cw3, + const struct dmub_window *cw4, + const struct dmub_window *cw5, + const struct dmub_window *cw6); + + void (*setup_mailbox)(struct dmub_srv *dmub, + const struct dmub_region *inbox1); + + uint32_t (*get_inbox1_rptr)(struct dmub_srv *dmub); + + void (*set_inbox1_wptr)(struct dmub_srv *dmub, uint32_t wptr_offset); + + bool (*is_supported)(struct dmub_srv *dmub); + + bool (*is_hw_init)(struct dmub_srv *dmub); + + bool (*is_phy_init)(struct dmub_srv *dmub); + + bool (*is_auto_load_done)(struct dmub_srv *dmub); +}; + +/** + * struct dmub_srv_create_params - params for dmub service creation + * @base_funcs: driver supplied base routines + * @hw_funcs: optional overrides for hw funcs + * @user_ctx: context data for callback funcs + * @asic: driver supplied asic + * @is_virtual: false for hw support only + */ +struct dmub_srv_create_params { + struct dmub_srv_base_funcs funcs; + struct dmub_srv_hw_funcs *hw_funcs; + void *user_ctx; + enum dmub_asic asic; + bool is_virtual; +}; + +/* + * struct dmub_srv_hw_params - params for dmub hardware initialization + * @fb: framebuffer info for each region + * @fb_base: base of the framebuffer aperture + * @fb_offset: offset of the framebuffer aperture + * @psp_version: psp version to pass for DMCU init + * @load_inst_const: true if DMUB should load inst const fw + */ +struct dmub_srv_hw_params { + struct dmub_fb *fb[DMUB_WINDOW_TOTAL]; + uint64_t fb_base; + uint64_t fb_offset; + uint32_t psp_version; + bool load_inst_const; +}; + +/** + * struct dmub_srv - software state for dmcub + * @asic: dmub asic identifier + * @user_ctx: user provided context for the dmub_srv + * @is_virtual: false if hardware support only + * @fw_state: dmub firmware state pointer + */ +struct dmub_srv { + enum dmub_asic asic; + void *user_ctx; + bool is_virtual; + volatile const struct dmub_fw_state *fw_state; + + /* private: internal use only */ + const struct dmub_srv_common_regs *regs; + + struct dmub_srv_base_funcs funcs; + struct dmub_srv_hw_funcs hw_funcs; + struct dmub_rb inbox1_rb; + + bool sw_init; + bool hw_init; + + uint64_t fb_base; + uint64_t fb_offset; + uint32_t psp_version; +}; + +/** + * dmub_srv_create() - creates the DMUB service. + * @dmub: the dmub service + * @params: creation parameters for the service + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_create(struct dmub_srv *dmub, + const struct dmub_srv_create_params *params); + +/** + * dmub_srv_destroy() - destroys the DMUB service. + * @dmub: the dmub service + */ +void dmub_srv_destroy(struct dmub_srv *dmub); + +/** + * dmub_srv_calc_region_info() - retreives region info from the dmub service + * @dmub: the dmub service + * @params: parameters used to calculate region locations + * @info_out: the output region info from dmub + * + * Calculates the base and top address for all relevant dmub regions + * using the parameters given (if any). + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status +dmub_srv_calc_region_info(struct dmub_srv *dmub, + const struct dmub_srv_region_params *params, + struct dmub_srv_region_info *out); + +/** + * dmub_srv_calc_region_info() - retreives fb info from the dmub service + * @dmub: the dmub service + * @params: parameters used to calculate fb locations + * @info_out: the output fb info from dmub + * + * Calculates the base and top address for all relevant dmub regions + * using the parameters given (if any). + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_calc_fb_info(struct dmub_srv *dmub, + const struct dmub_srv_fb_params *params, + struct dmub_srv_fb_info *out); + +/** + * dmub_srv_has_hw_support() - returns hw support state for dmcub + * @dmub: the dmub service + * @is_supported: hw support state + * + * Queries the hardware for DMCUB support and returns the result. + * + * Can be called before dmub_srv_hw_init(). + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_has_hw_support(struct dmub_srv *dmub, + bool *is_supported); + +/** + * dmub_srv_is_hw_init() - returns hardware init state + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_is_hw_init(struct dmub_srv *dmub, bool *is_hw_init); + +/** + * dmub_srv_hw_init() - initializes the underlying DMUB hardware + * @dmub: the dmub service + * @params: params for hardware initialization + * + * Resets the DMUB hardware and performs backdoor loading of the + * required cache regions based on the input framebuffer regions. + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_NO_CTX - dmcub context not initialized + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, + const struct dmub_srv_hw_params *params); + +/** + * dmub_srv_hw_reset() - puts the DMUB hardware in reset state if initialized + * @dmub: the dmub service + * + * Before destroying the DMUB service or releasing the backing framebuffer + * memory we'll need to put the DMCUB into reset first. + * + * A subsequent call to dmub_srv_hw_init() will re-enable the DMCUB. + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub); + +/** + * dmub_srv_cmd_queue() - queues a command to the DMUB + * @dmub: the dmub service + * @cmd: the command to queue + * + * Queues a command to the DMUB service but does not begin execution + * immediately. + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_QUEUE_FULL - no remaining room in queue + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub, + const struct dmub_cmd_header *cmd); + +/** + * dmub_srv_cmd_execute() - Executes a queued sequence to the dmub + * @dmub: the dmub service + * + * Begins execution of queued commands on the dmub. + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_cmd_execute(struct dmub_srv *dmub); + +/** + * dmub_srv_wait_for_auto_load() - Waits for firmware auto load to complete + * @dmub: the dmub service + * @timeout_us: the maximum number of microseconds to wait + * + * Waits until firmware has been autoloaded by the DMCUB. The maximum + * wait time is given in microseconds to prevent spinning forever. + * + * On ASICs without firmware autoload support this function will return + * immediately. + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_TIMEOUT - wait for phy init timed out + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_wait_for_auto_load(struct dmub_srv *dmub, + uint32_t timeout_us); + +/** + * dmub_srv_wait_for_phy_init() - Waits for DMUB PHY init to complete + * @dmub: the dmub service + * @timeout_us: the maximum number of microseconds to wait + * + * Waits until the PHY has been initialized by the DMUB. The maximum + * wait time is given in microseconds to prevent spinning forever. + * + * On ASICs without PHY init support this function will return + * immediately. + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_TIMEOUT - wait for phy init timed out + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_wait_for_phy_init(struct dmub_srv *dmub, + uint32_t timeout_us); + +/** + * dmub_srv_wait_for_idle() - Waits for the DMUB to be idle + * @dmub: the dmub service + * @timeout_us: the maximum number of microseconds to wait + * + * Waits until the DMUB buffer is empty and all commands have + * finished processing. The maximum wait time is given in + * microseconds to prevent spinning forever. + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_TIMEOUT - wait for buffer to flush timed out + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_wait_for_idle(struct dmub_srv *dmub, + uint32_t timeout_us); + +#if defined(__cplusplus) +} +#endif + +#endif /* _DMUB_SRV_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_trace_buffer.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_trace_buffer.h new file mode 100644 index 000000000000..6b3ee42db350 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_trace_buffer.h @@ -0,0 +1,69 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef _DMUB_TRACE_BUFFER_H_ +#define _DMUB_TRACE_BUFFER_H_ + +#include "dmub_types.h" + +#define LOAD_DMCU_FW 1 +#define LOAD_PHY_FW 2 + + +enum dmucb_trace_code { + DMCUB__UNKNOWN, + DMCUB__MAIN_BEGIN, + DMCUB__PHY_INIT_BEGIN, + DMCUB__PHY_FW_SRAM_LOAD_BEGIN, + DMCUB__PHY_FW_SRAM_LOAD_END, + DMCUB__PHY_INIT_POLL_DONE, + DMCUB__PHY_INIT_END, + DMCUB__DMCU_ERAM_LOAD_BEGIN, + DMCUB__DMCU_ERAM_LOAD_END, + DMCUB__DMCU_ISR_LOAD_BEGIN, + DMCUB__DMCU_ISR_LOAD_END, + DMCUB__MAIN_IDLE, + DMCUB__PERF_TRACE, + DMCUB__PG_DONE, +}; + +struct dmcub_trace_buf_entry { + enum dmucb_trace_code trace_code; + uint32_t tick_count; + uint32_t param0; + uint32_t param1; +}; + +#define TRACE_BUF_SIZE (1024) //1 kB +#define PERF_TRACE_MAX_ENTRY ((TRACE_BUF_SIZE - 8)/sizeof(struct dmcub_trace_buf_entry)) + + +struct dmcub_trace_buf { + uint32_t entry_count; + uint32_t clk_freq; + struct dmcub_trace_buf_entry entries[PERF_TRACE_MAX_ENTRY]; +}; + + +#endif /* _DMUB_TRACE_BUFFER_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_types.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_types.h new file mode 100644 index 000000000000..41d524b0db2f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_types.h @@ -0,0 +1,64 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_TYPES_H_ +#define _DMUB_TYPES_H_ + +/* Basic type definitions. */ +#include <asm/byteorder.h> +#include <linux/types.h> +#include <linux/string.h> +#include <linux/delay.h> +#include <stdarg.h> + +#if defined(__cplusplus) +extern "C" { +#endif + +#ifndef dmub_memcpy +#define dmub_memcpy(dest, source, bytes) memcpy((dest), (source), (bytes)) +#endif + +#ifndef dmub_memset +#define dmub_memset(dest, val, bytes) memset((dest), (val), (bytes)) +#endif + +#ifndef dmub_udelay +#define dmub_udelay(microseconds) udelay(microseconds) +#endif + +union dmub_addr { + struct { + uint32_t low_part; + uint32_t high_part; + } u; + uint64_t quad_part; +}; + +#if defined(__cplusplus) +} +#endif + +#endif /* _DMUB_TYPES_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/Makefile b/drivers/gpu/drm/amd/display/dmub/src/Makefile new file mode 100644 index 000000000000..e08dfeea24b0 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/Makefile @@ -0,0 +1,27 @@ +# +# Copyright 2019 Advanced Micro Devices, Inc. +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the "Software"), +# to deal in the Software without restriction, including without limitation +# the rights to use, copy, modify, merge, publish, distribute, sublicense, +# and/or sell copies of the Software, and to permit persons to whom the +# Software is furnished to do so, subject to the following conditions: +# +# The above copyright notice and this permission notice shall be included in +# all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR +# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +# OTHER DEALINGS IN THE SOFTWARE. +# + +DMUB = dmub_srv.o dmub_reg.o dmub_dcn20.o dmub_dcn21.o + +AMD_DAL_DMUB = $(addprefix $(AMDDALPATH)/dmub/src/,$(DMUB)) + +AMD_DISPLAY_FILES += $(AMD_DAL_DMUB) diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c new file mode 100644 index 000000000000..b2ca8e0dbac9 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c @@ -0,0 +1,219 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "../inc/dmub_srv.h" +#include "dmub_reg.h" +#include "dmub_dcn20.h" + +#include "dcn/dcn_2_0_0_offset.h" +#include "dcn/dcn_2_0_0_sh_mask.h" +#include "soc15_hw_ip.h" +#include "vega10_ip_offset.h" + +#define BASE_INNER(seg) DCN_BASE__INST0_SEG##seg +#define CTX dmub +#define REGS dmub->regs + +/* Registers. */ + +const struct dmub_srv_common_regs dmub_srv_dcn20_regs = { +#define DMUB_SR(reg) REG_OFFSET(reg), + { DMUB_COMMON_REGS() }, +#undef DMUB_SR + +#define DMUB_SF(reg, field) FD_MASK(reg, field), + { DMUB_COMMON_FIELDS() }, +#undef DMUB_SF + +#define DMUB_SF(reg, field) FD_SHIFT(reg, field), + { DMUB_COMMON_FIELDS() }, +#undef DMUB_SF +}; + +/* Shared functions. */ + +static void dmub_dcn20_get_fb_base_offset(struct dmub_srv *dmub, + uint64_t *fb_base, + uint64_t *fb_offset) +{ + uint32_t tmp; + + REG_GET(DCN_VM_FB_LOCATION_BASE, FB_BASE, &tmp); + *fb_base = (uint64_t)tmp << 24; + + REG_GET(DCN_VM_FB_OFFSET, FB_OFFSET, &tmp); + *fb_offset = (uint64_t)tmp << 24; +} + +static inline void dmub_dcn20_translate_addr(const union dmub_addr *addr_in, + uint64_t fb_base, + uint64_t fb_offset, + union dmub_addr *addr_out) +{ + addr_out->quad_part = addr_in->quad_part - fb_base + fb_offset; +} + +void dmub_dcn20_reset(struct dmub_srv *dmub) +{ + REG_UPDATE(DMCUB_CNTL, DMCUB_SOFT_RESET, 1); + REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0); + REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); + REG_WRITE(DMCUB_INBOX1_RPTR, 0); + REG_WRITE(DMCUB_INBOX1_WPTR, 0); +} + +void dmub_dcn20_reset_release(struct dmub_srv *dmub) +{ + REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 0); + REG_WRITE(DMCUB_SCRATCH15, dmub->psp_version & 0x001100FF); + REG_UPDATE_2(DMCUB_CNTL, DMCUB_ENABLE, 1, DMCUB_TRACEPORT_EN, 1); + REG_UPDATE(DMCUB_CNTL, DMCUB_SOFT_RESET, 0); +} + +void dmub_dcn20_backdoor_load(struct dmub_srv *dmub, + const struct dmub_window *cw0, + const struct dmub_window *cw1) +{ + union dmub_addr offset; + uint64_t fb_base, fb_offset; + + dmub_dcn20_get_fb_base_offset(dmub, &fb_base, &fb_offset); + + REG_UPDATE(DMCUB_SEC_CNTL, DMCUB_SEC_RESET, 1); + REG_UPDATE_2(DMCUB_MEM_CNTL, DMCUB_MEM_READ_SPACE, 0x3, + DMCUB_MEM_WRITE_SPACE, 0x3); + + dmub_dcn20_translate_addr(&cw0->offset, fb_base, fb_offset, &offset); + + REG_WRITE(DMCUB_REGION3_CW0_OFFSET, offset.u.low_part); + REG_WRITE(DMCUB_REGION3_CW0_OFFSET_HIGH, offset.u.high_part); + REG_WRITE(DMCUB_REGION3_CW0_BASE_ADDRESS, cw0->region.base); + REG_SET_2(DMCUB_REGION3_CW0_TOP_ADDRESS, 0, + DMCUB_REGION3_CW0_TOP_ADDRESS, cw0->region.top, + DMCUB_REGION3_CW0_ENABLE, 1); + + dmub_dcn20_translate_addr(&cw1->offset, fb_base, fb_offset, &offset); + + REG_WRITE(DMCUB_REGION3_CW1_OFFSET, offset.u.low_part); + REG_WRITE(DMCUB_REGION3_CW1_OFFSET_HIGH, offset.u.high_part); + REG_WRITE(DMCUB_REGION3_CW1_BASE_ADDRESS, cw1->region.base); + REG_SET_2(DMCUB_REGION3_CW1_TOP_ADDRESS, 0, + DMCUB_REGION3_CW1_TOP_ADDRESS, cw1->region.top, + DMCUB_REGION3_CW1_ENABLE, 1); + + REG_UPDATE_2(DMCUB_SEC_CNTL, DMCUB_SEC_RESET, 0, DMCUB_MEM_UNIT_ID, + 0x20); +} + +void dmub_dcn20_setup_windows(struct dmub_srv *dmub, + const struct dmub_window *cw2, + const struct dmub_window *cw3, + const struct dmub_window *cw4, + const struct dmub_window *cw5, + const struct dmub_window *cw6) +{ + union dmub_addr offset; + uint64_t fb_base, fb_offset; + + dmub_dcn20_get_fb_base_offset(dmub, &fb_base, &fb_offset); + + dmub_dcn20_translate_addr(&cw2->offset, fb_base, fb_offset, &offset); + + REG_WRITE(DMCUB_REGION3_CW2_OFFSET, offset.u.low_part); + REG_WRITE(DMCUB_REGION3_CW2_OFFSET_HIGH, offset.u.high_part); + REG_WRITE(DMCUB_REGION3_CW2_BASE_ADDRESS, cw2->region.base); + REG_SET_2(DMCUB_REGION3_CW2_TOP_ADDRESS, 0, + DMCUB_REGION3_CW2_TOP_ADDRESS, cw2->region.top, + DMCUB_REGION3_CW2_ENABLE, 1); + + dmub_dcn20_translate_addr(&cw3->offset, fb_base, fb_offset, &offset); + + REG_WRITE(DMCUB_REGION3_CW3_OFFSET, offset.u.low_part); + REG_WRITE(DMCUB_REGION3_CW3_OFFSET_HIGH, offset.u.high_part); + REG_WRITE(DMCUB_REGION3_CW3_BASE_ADDRESS, cw3->region.base); + REG_SET_2(DMCUB_REGION3_CW3_TOP_ADDRESS, 0, + DMCUB_REGION3_CW3_TOP_ADDRESS, cw3->region.top, + DMCUB_REGION3_CW3_ENABLE, 1); + + /* TODO: Move this to CW4. */ + dmub_dcn20_translate_addr(&cw4->offset, fb_base, fb_offset, &offset); + + REG_WRITE(DMCUB_REGION4_OFFSET, offset.u.low_part); + REG_WRITE(DMCUB_REGION4_OFFSET_HIGH, offset.u.high_part); + REG_SET_2(DMCUB_REGION4_TOP_ADDRESS, 0, DMCUB_REGION4_TOP_ADDRESS, + cw4->region.top - cw4->region.base - 1, DMCUB_REGION4_ENABLE, + 1); + + dmub_dcn20_translate_addr(&cw5->offset, fb_base, fb_offset, &offset); + + REG_WRITE(DMCUB_REGION3_CW5_OFFSET, offset.u.low_part); + REG_WRITE(DMCUB_REGION3_CW5_OFFSET_HIGH, offset.u.high_part); + REG_WRITE(DMCUB_REGION3_CW5_BASE_ADDRESS, cw5->region.base); + REG_SET_2(DMCUB_REGION3_CW5_TOP_ADDRESS, 0, + DMCUB_REGION3_CW5_TOP_ADDRESS, cw5->region.top, + DMCUB_REGION3_CW5_ENABLE, 1); + + dmub_dcn20_translate_addr(&cw6->offset, fb_base, fb_offset, &offset); + + REG_WRITE(DMCUB_REGION3_CW6_OFFSET, offset.u.low_part); + REG_WRITE(DMCUB_REGION3_CW6_OFFSET_HIGH, offset.u.high_part); + REG_WRITE(DMCUB_REGION3_CW6_BASE_ADDRESS, cw6->region.base); + REG_SET_2(DMCUB_REGION3_CW6_TOP_ADDRESS, 0, + DMCUB_REGION3_CW6_TOP_ADDRESS, cw6->region.top, + DMCUB_REGION3_CW6_ENABLE, 1); +} + +void dmub_dcn20_setup_mailbox(struct dmub_srv *dmub, + const struct dmub_region *inbox1) +{ + /* TODO: Use CW4 instead of region 4. */ + + REG_WRITE(DMCUB_INBOX1_BASE_ADDRESS, 0x80000000); + REG_WRITE(DMCUB_INBOX1_SIZE, inbox1->top - inbox1->base); +} + +uint32_t dmub_dcn20_get_inbox1_rptr(struct dmub_srv *dmub) +{ + return REG_READ(DMCUB_INBOX1_RPTR); +} + +void dmub_dcn20_set_inbox1_wptr(struct dmub_srv *dmub, uint32_t wptr_offset) +{ + REG_WRITE(DMCUB_INBOX1_WPTR, wptr_offset); +} + +bool dmub_dcn20_is_hw_init(struct dmub_srv *dmub) +{ + return REG_READ(DMCUB_REGION3_CW2_BASE_ADDRESS) != 0; +} + +bool dmub_dcn20_is_supported(struct dmub_srv *dmub) +{ + uint32_t supported = 0; + + REG_GET(CC_DC_PIPE_DIS, DC_DMCUB_ENABLE, &supported); + + return supported; +} diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h new file mode 100644 index 000000000000..04b0fa13153d --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h @@ -0,0 +1,186 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_DCN20_H_ +#define _DMUB_DCN20_H_ + +#include "../inc/dmub_types.h" + +struct dmub_srv; + +/* DCN20 register definitions. */ + +#define DMUB_COMMON_REGS() \ + DMUB_SR(DMCUB_CNTL) \ + DMUB_SR(DMCUB_MEM_CNTL) \ + DMUB_SR(DMCUB_SEC_CNTL) \ + DMUB_SR(DMCUB_INBOX1_BASE_ADDRESS) \ + DMUB_SR(DMCUB_INBOX1_SIZE) \ + DMUB_SR(DMCUB_INBOX1_RPTR) \ + DMUB_SR(DMCUB_INBOX1_WPTR) \ + DMUB_SR(DMCUB_REGION3_CW0_OFFSET) \ + DMUB_SR(DMCUB_REGION3_CW1_OFFSET) \ + DMUB_SR(DMCUB_REGION3_CW2_OFFSET) \ + DMUB_SR(DMCUB_REGION3_CW3_OFFSET) \ + DMUB_SR(DMCUB_REGION3_CW4_OFFSET) \ + DMUB_SR(DMCUB_REGION3_CW5_OFFSET) \ + DMUB_SR(DMCUB_REGION3_CW6_OFFSET) \ + DMUB_SR(DMCUB_REGION3_CW7_OFFSET) \ + DMUB_SR(DMCUB_REGION3_CW0_OFFSET_HIGH) \ + DMUB_SR(DMCUB_REGION3_CW1_OFFSET_HIGH) \ + DMUB_SR(DMCUB_REGION3_CW2_OFFSET_HIGH) \ + DMUB_SR(DMCUB_REGION3_CW3_OFFSET_HIGH) \ + DMUB_SR(DMCUB_REGION3_CW4_OFFSET_HIGH) \ + DMUB_SR(DMCUB_REGION3_CW5_OFFSET_HIGH) \ + DMUB_SR(DMCUB_REGION3_CW6_OFFSET_HIGH) \ + DMUB_SR(DMCUB_REGION3_CW7_OFFSET_HIGH) \ + DMUB_SR(DMCUB_REGION3_CW0_BASE_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW1_BASE_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW2_BASE_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW3_BASE_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW4_BASE_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW5_BASE_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW6_BASE_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW7_BASE_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW0_TOP_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW1_TOP_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW2_TOP_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW3_TOP_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW4_TOP_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW5_TOP_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW6_TOP_ADDRESS) \ + DMUB_SR(DMCUB_REGION3_CW7_TOP_ADDRESS) \ + DMUB_SR(DMCUB_REGION4_OFFSET) \ + DMUB_SR(DMCUB_REGION4_OFFSET_HIGH) \ + DMUB_SR(DMCUB_REGION4_TOP_ADDRESS) \ + DMUB_SR(DMCUB_SCRATCH0) \ + DMUB_SR(DMCUB_SCRATCH1) \ + DMUB_SR(DMCUB_SCRATCH2) \ + DMUB_SR(DMCUB_SCRATCH3) \ + DMUB_SR(DMCUB_SCRATCH4) \ + DMUB_SR(DMCUB_SCRATCH5) \ + DMUB_SR(DMCUB_SCRATCH6) \ + DMUB_SR(DMCUB_SCRATCH7) \ + DMUB_SR(DMCUB_SCRATCH8) \ + DMUB_SR(DMCUB_SCRATCH9) \ + DMUB_SR(DMCUB_SCRATCH10) \ + DMUB_SR(DMCUB_SCRATCH11) \ + DMUB_SR(DMCUB_SCRATCH12) \ + DMUB_SR(DMCUB_SCRATCH13) \ + DMUB_SR(DMCUB_SCRATCH14) \ + DMUB_SR(DMCUB_SCRATCH15) \ + DMUB_SR(CC_DC_PIPE_DIS) \ + DMUB_SR(MMHUBBUB_SOFT_RESET) \ + DMUB_SR(DCN_VM_FB_LOCATION_BASE) \ + DMUB_SR(DCN_VM_FB_OFFSET) + +#define DMUB_COMMON_FIELDS() \ + DMUB_SF(DMCUB_CNTL, DMCUB_ENABLE) \ + DMUB_SF(DMCUB_CNTL, DMCUB_SOFT_RESET) \ + DMUB_SF(DMCUB_CNTL, DMCUB_TRACEPORT_EN) \ + DMUB_SF(DMCUB_MEM_CNTL, DMCUB_MEM_READ_SPACE) \ + DMUB_SF(DMCUB_MEM_CNTL, DMCUB_MEM_WRITE_SPACE) \ + DMUB_SF(DMCUB_SEC_CNTL, DMCUB_SEC_RESET) \ + DMUB_SF(DMCUB_SEC_CNTL, DMCUB_MEM_UNIT_ID) \ + DMUB_SF(DMCUB_REGION3_CW0_TOP_ADDRESS, DMCUB_REGION3_CW0_TOP_ADDRESS) \ + DMUB_SF(DMCUB_REGION3_CW0_TOP_ADDRESS, DMCUB_REGION3_CW0_ENABLE) \ + DMUB_SF(DMCUB_REGION3_CW1_TOP_ADDRESS, DMCUB_REGION3_CW1_TOP_ADDRESS) \ + DMUB_SF(DMCUB_REGION3_CW1_TOP_ADDRESS, DMCUB_REGION3_CW1_ENABLE) \ + DMUB_SF(DMCUB_REGION3_CW2_TOP_ADDRESS, DMCUB_REGION3_CW2_TOP_ADDRESS) \ + DMUB_SF(DMCUB_REGION3_CW2_TOP_ADDRESS, DMCUB_REGION3_CW2_ENABLE) \ + DMUB_SF(DMCUB_REGION3_CW3_TOP_ADDRESS, DMCUB_REGION3_CW3_TOP_ADDRESS) \ + DMUB_SF(DMCUB_REGION3_CW3_TOP_ADDRESS, DMCUB_REGION3_CW3_ENABLE) \ + DMUB_SF(DMCUB_REGION3_CW4_TOP_ADDRESS, DMCUB_REGION3_CW4_TOP_ADDRESS) \ + DMUB_SF(DMCUB_REGION3_CW4_TOP_ADDRESS, DMCUB_REGION3_CW4_ENABLE) \ + DMUB_SF(DMCUB_REGION3_CW5_TOP_ADDRESS, DMCUB_REGION3_CW5_TOP_ADDRESS) \ + DMUB_SF(DMCUB_REGION3_CW5_TOP_ADDRESS, DMCUB_REGION3_CW5_ENABLE) \ + DMUB_SF(DMCUB_REGION3_CW6_TOP_ADDRESS, DMCUB_REGION3_CW6_TOP_ADDRESS) \ + DMUB_SF(DMCUB_REGION3_CW6_TOP_ADDRESS, DMCUB_REGION3_CW6_ENABLE) \ + DMUB_SF(DMCUB_REGION3_CW7_TOP_ADDRESS, DMCUB_REGION3_CW7_TOP_ADDRESS) \ + DMUB_SF(DMCUB_REGION3_CW7_TOP_ADDRESS, DMCUB_REGION3_CW7_ENABLE) \ + DMUB_SF(DMCUB_REGION4_TOP_ADDRESS, DMCUB_REGION4_TOP_ADDRESS) \ + DMUB_SF(DMCUB_REGION4_TOP_ADDRESS, DMCUB_REGION4_ENABLE) \ + DMUB_SF(CC_DC_PIPE_DIS, DC_DMCUB_ENABLE) \ + DMUB_SF(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET) \ + DMUB_SF(DCN_VM_FB_LOCATION_BASE, FB_BASE) \ + DMUB_SF(DCN_VM_FB_OFFSET, FB_OFFSET) + +struct dmub_srv_common_reg_offset { +#define DMUB_SR(reg) uint32_t reg; + DMUB_COMMON_REGS() +#undef DMUB_SR +}; + +struct dmub_srv_common_reg_shift { +#define DMUB_SF(reg, field) uint8_t reg##__##field; + DMUB_COMMON_FIELDS() +#undef DMUB_SF +}; + +struct dmub_srv_common_reg_mask { +#define DMUB_SF(reg, field) uint32_t reg##__##field; + DMUB_COMMON_FIELDS() +#undef DMUB_SF +}; + +struct dmub_srv_common_regs { + const struct dmub_srv_common_reg_offset offset; + const struct dmub_srv_common_reg_mask mask; + const struct dmub_srv_common_reg_shift shift; +}; + +extern const struct dmub_srv_common_regs dmub_srv_dcn20_regs; + +/* Hardware functions. */ + +void dmub_dcn20_init(struct dmub_srv *dmub); + +void dmub_dcn20_reset(struct dmub_srv *dmub); + +void dmub_dcn20_reset_release(struct dmub_srv *dmub); + +void dmub_dcn20_backdoor_load(struct dmub_srv *dmub, + const struct dmub_window *cw0, + const struct dmub_window *cw1); + +void dmub_dcn20_setup_windows(struct dmub_srv *dmub, + const struct dmub_window *cw2, + const struct dmub_window *cw3, + const struct dmub_window *cw4, + const struct dmub_window *cw5, + const struct dmub_window *cw6); + +void dmub_dcn20_setup_mailbox(struct dmub_srv *dmub, + const struct dmub_region *inbox1); + +uint32_t dmub_dcn20_get_inbox1_rptr(struct dmub_srv *dmub); + +void dmub_dcn20_set_inbox1_wptr(struct dmub_srv *dmub, uint32_t wptr_offset); + +bool dmub_dcn20_is_hw_init(struct dmub_srv *dmub); + +bool dmub_dcn20_is_supported(struct dmub_srv *dmub); + +#endif /* _DMUB_DCN20_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c new file mode 100644 index 000000000000..5bed9fcd6b5c --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c @@ -0,0 +1,64 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "../inc/dmub_srv.h" +#include "dmub_reg.h" +#include "dmub_dcn21.h" + +#include "dcn/dcn_2_1_0_offset.h" +#include "dcn/dcn_2_1_0_sh_mask.h" +#include "renoir_ip_offset.h" + +#define BASE_INNER(seg) DMU_BASE__INST0_SEG##seg +#define CTX dmub +#define REGS dmub->regs + +/* Registers. */ + +const struct dmub_srv_common_regs dmub_srv_dcn21_regs = { +#define DMUB_SR(reg) REG_OFFSET(reg), + { DMUB_COMMON_REGS() }, +#undef DMUB_SR + +#define DMUB_SF(reg, field) FD_MASK(reg, field), + { DMUB_COMMON_FIELDS() }, +#undef DMUB_SF + +#define DMUB_SF(reg, field) FD_SHIFT(reg, field), + { DMUB_COMMON_FIELDS() }, +#undef DMUB_SF +}; + +/* Shared functions. */ + +bool dmub_dcn21_is_auto_load_done(struct dmub_srv *dmub) +{ + return (REG_READ(DMCUB_SCRATCH0) == 3); +} + +bool dmub_dcn21_is_phy_init(struct dmub_srv *dmub) +{ + return REG_READ(DMCUB_SCRATCH10) == 0; +} diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h new file mode 100644 index 000000000000..2bbea237137b --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h @@ -0,0 +1,41 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_DCN21_H_ +#define _DMUB_DCN21_H_ + +#include "dmub_dcn20.h" + +/* Registers. */ + +extern const struct dmub_srv_common_regs dmub_srv_dcn21_regs; + +/* Hardware functions. */ + +bool dmub_dcn21_is_auto_load_done(struct dmub_srv *dmub); + +bool dmub_dcn21_is_phy_init(struct dmub_srv *dmub); + +#endif /* _DMUB_DCN21_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.c new file mode 100644 index 000000000000..4094eca212f0 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.c @@ -0,0 +1,109 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dmub_reg.h" +#include "../inc/dmub_srv.h" + +struct dmub_reg_value_masks { + uint32_t value; + uint32_t mask; +}; + +static inline void +set_reg_field_value_masks(struct dmub_reg_value_masks *field_value_mask, + uint32_t value, uint32_t mask, uint8_t shift) +{ + field_value_mask->value = + (field_value_mask->value & ~mask) | (mask & (value << shift)); + field_value_mask->mask = field_value_mask->mask | mask; +} + +static void set_reg_field_values(struct dmub_reg_value_masks *field_value_mask, + uint32_t addr, int n, uint8_t shift1, + uint32_t mask1, uint32_t field_value1, + va_list ap) +{ + uint32_t shift, mask, field_value; + int i = 1; + + /* gather all bits value/mask getting updated in this register */ + set_reg_field_value_masks(field_value_mask, field_value1, mask1, + shift1); + + while (i < n) { + shift = va_arg(ap, uint32_t); + mask = va_arg(ap, uint32_t); + field_value = va_arg(ap, uint32_t); + + set_reg_field_value_masks(field_value_mask, field_value, mask, + shift); + i++; + } +} + +static inline uint32_t get_reg_field_value_ex(uint32_t reg_value, uint32_t mask, + uint8_t shift) +{ + return (mask & reg_value) >> shift; +} + +void dmub_reg_update(struct dmub_srv *srv, uint32_t addr, int n, uint8_t shift1, + uint32_t mask1, uint32_t field_value1, ...) +{ + struct dmub_reg_value_masks field_value_mask = { 0 }; + uint32_t reg_val; + va_list ap; + + va_start(ap, field_value1); + set_reg_field_values(&field_value_mask, addr, n, shift1, mask1, + field_value1, ap); + va_end(ap); + + reg_val = srv->funcs.reg_read(srv->user_ctx, addr); + reg_val = (reg_val & ~field_value_mask.mask) | field_value_mask.value; + srv->funcs.reg_write(srv->user_ctx, addr, reg_val); +} + +void dmub_reg_set(struct dmub_srv *srv, uint32_t addr, uint32_t reg_val, int n, + uint8_t shift1, uint32_t mask1, uint32_t field_value1, ...) +{ + struct dmub_reg_value_masks field_value_mask = { 0 }; + va_list ap; + + va_start(ap, field_value1); + set_reg_field_values(&field_value_mask, addr, n, shift1, mask1, + field_value1, ap); + va_end(ap); + + reg_val = (reg_val & ~field_value_mask.mask) | field_value_mask.value; + srv->funcs.reg_write(srv->user_ctx, addr, reg_val); +} + +void dmub_reg_get(struct dmub_srv *srv, uint32_t addr, uint8_t shift, + uint32_t mask, uint32_t *field_value) +{ + uint32_t reg_val = srv->funcs.reg_read(srv->user_ctx, addr); + *field_value = get_reg_field_value_ex(reg_val, mask, shift); +} diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.h new file mode 100644 index 000000000000..c1f4030929a4 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.h @@ -0,0 +1,124 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_REG_H_ +#define _DMUB_REG_H_ + +#include "../inc/dmub_types.h" + +struct dmub_srv; + +/* Register offset and field lookup. */ + +#define BASE(seg) BASE_INNER(seg) + +#define REG_OFFSET(reg_name) (BASE(mm##reg_name##_BASE_IDX) + mm##reg_name) + +#define FD_SHIFT(reg_name, field) reg_name##__##field##__SHIFT + +#define FD_MASK(reg_name, field) reg_name##__##field##_MASK + +#define REG(reg) (REGS)->offset.reg + +#define FD(reg_field) (REGS)->shift.reg_field, (REGS)->mask.reg_field + +#define FN(reg_name, field) FD(reg_name##__##field) + +/* Register reads and writes. */ + +#define REG_READ(reg) ((CTX)->funcs.reg_read((CTX)->user_ctx, REG(reg))) + +#define REG_WRITE(reg, val) \ + ((CTX)->funcs.reg_write((CTX)->user_ctx, REG(reg), (val))) + +/* Register field setting. */ + +#define REG_SET_N(reg_name, n, initial_val, ...) \ + dmub_reg_set(CTX, REG(reg_name), initial_val, n, __VA_ARGS__) + +#define REG_SET(reg_name, initial_val, field, val) \ + REG_SET_N(reg_name, 1, initial_val, \ + FN(reg_name, field), val) + +#define REG_SET_2(reg, init_value, f1, v1, f2, v2) \ + REG_SET_N(reg, 2, init_value, \ + FN(reg, f1), v1, \ + FN(reg, f2), v2) + +#define REG_SET_3(reg, init_value, f1, v1, f2, v2, f3, v3) \ + REG_SET_N(reg, 3, init_value, \ + FN(reg, f1), v1, \ + FN(reg, f2), v2, \ + FN(reg, f3), v3) + +#define REG_SET_4(reg, init_value, f1, v1, f2, v2, f3, v3, f4, v4) \ + REG_SET_N(reg, 4, init_value, \ + FN(reg, f1), v1, \ + FN(reg, f2), v2, \ + FN(reg, f3), v3, \ + FN(reg, f4), v4) + +/* Register field updating. */ + +#define REG_UPDATE_N(reg_name, n, ...)\ + dmub_reg_update(CTX, REG(reg_name), n, __VA_ARGS__) + +#define REG_UPDATE(reg_name, field, val) \ + REG_UPDATE_N(reg_name, 1, \ + FN(reg_name, field), val) + +#define REG_UPDATE_2(reg, f1, v1, f2, v2) \ + REG_UPDATE_N(reg, 2,\ + FN(reg, f1), v1,\ + FN(reg, f2), v2) + +#define REG_UPDATE_3(reg, f1, v1, f2, v2, f3, v3) \ + REG_UPDATE_N(reg, 3, \ + FN(reg, f1), v1, \ + FN(reg, f2), v2, \ + FN(reg, f3), v3) + +#define REG_UPDATE_4(reg, f1, v1, f2, v2, f3, v3, f4, v4) \ + REG_UPDATE_N(reg, 4, \ + FN(reg, f1), v1, \ + FN(reg, f2), v2, \ + FN(reg, f3), v3, \ + FN(reg, f4), v4) + +/* Register field getting. */ + +#define REG_GET(reg_name, field, val) \ + dmub_reg_get(CTX, REG(reg_name), FN(reg_name, field), val) + +void dmub_reg_set(struct dmub_srv *srv, uint32_t addr, uint32_t reg_val, int n, + uint8_t shift1, uint32_t mask1, uint32_t field_value1, ...); + +void dmub_reg_update(struct dmub_srv *srv, uint32_t addr, int n, uint8_t shift1, + uint32_t mask1, uint32_t field_value1, ...); + +void dmub_reg_get(struct dmub_srv *srv, uint32_t addr, uint8_t shift, + uint32_t mask, uint32_t *field_value); + +#endif /* _DMUB_REG_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c new file mode 100644 index 000000000000..85a518bf8a76 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -0,0 +1,524 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "../inc/dmub_srv.h" +#include "dmub_dcn20.h" +#include "dmub_dcn21.h" +#include "dmub_fw_meta.h" +#include "os_types.h" +/* + * Note: the DMUB service is standalone. No additional headers should be + * added below or above this line unless they reside within the DMUB + * folder. + */ + +/* Alignment for framebuffer memory. */ +#define DMUB_FB_ALIGNMENT (1024 * 1024) + +/* Stack size. */ +#define DMUB_STACK_SIZE (128 * 1024) + +/* Context size. */ +#define DMUB_CONTEXT_SIZE (512 * 1024) + +/* Mailbox size */ +#define DMUB_MAILBOX_SIZE (DMUB_RB_SIZE) + +/* Default state size if meta is absent. */ +#define DMUB_FW_STATE_SIZE (1024) + +/* Default tracebuffer size if meta is absent. */ +#define DMUB_TRACE_BUFFER_SIZE (1024) + +/* Number of windows in use. */ +#define DMUB_NUM_WINDOWS (DMUB_WINDOW_6_FW_STATE + 1) +/* Base addresses. */ + +#define DMUB_CW0_BASE (0x60000000) +#define DMUB_CW1_BASE (0x61000000) +#define DMUB_CW3_BASE (0x63000000) +#define DMUB_CW5_BASE (0x65000000) +#define DMUB_CW6_BASE (0x66000000) + +static inline uint32_t dmub_align(uint32_t val, uint32_t factor) +{ + return (val + factor - 1) / factor * factor; +} + +static void dmub_flush_buffer_mem(const struct dmub_fb *fb) +{ + const uint8_t *base = (const uint8_t *)fb->cpu_addr; + uint8_t buf[64]; + uint32_t pos, end; + + /** + * Read 64-byte chunks since we don't want to store a + * large temporary buffer for this purpose. + */ + end = fb->size / sizeof(buf) * sizeof(buf); + + for (pos = 0; pos < end; pos += sizeof(buf)) + dmub_memcpy(buf, base + pos, sizeof(buf)); + + /* Read anything leftover into the buffer. */ + if (end < fb->size) + dmub_memcpy(buf, base + pos, fb->size - end); +} + +static const struct dmub_fw_meta_info * +dmub_get_fw_meta_info(const uint8_t *fw_bss_data, uint32_t fw_bss_data_size) +{ + const union dmub_fw_meta *meta; + + if (fw_bss_data == NULL) + return NULL; + + if (fw_bss_data_size < sizeof(union dmub_fw_meta) + DMUB_FW_META_OFFSET) + return NULL; + + meta = (const union dmub_fw_meta *)(fw_bss_data + fw_bss_data_size - + DMUB_FW_META_OFFSET - + sizeof(union dmub_fw_meta)); + + if (meta->info.magic_value != DMUB_FW_META_MAGIC) + return NULL; + + return &meta->info; +} + +static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic) +{ + struct dmub_srv_hw_funcs *funcs = &dmub->hw_funcs; + + switch (asic) { + case DMUB_ASIC_DCN20: + case DMUB_ASIC_DCN21: + dmub->regs = &dmub_srv_dcn20_regs; + + funcs->reset = dmub_dcn20_reset; + funcs->reset_release = dmub_dcn20_reset_release; + funcs->backdoor_load = dmub_dcn20_backdoor_load; + funcs->setup_windows = dmub_dcn20_setup_windows; + funcs->setup_mailbox = dmub_dcn20_setup_mailbox; + funcs->get_inbox1_rptr = dmub_dcn20_get_inbox1_rptr; + funcs->set_inbox1_wptr = dmub_dcn20_set_inbox1_wptr; + funcs->is_supported = dmub_dcn20_is_supported; + funcs->is_hw_init = dmub_dcn20_is_hw_init; + + if (asic == DMUB_ASIC_DCN21) { + dmub->regs = &dmub_srv_dcn21_regs; + + funcs->is_auto_load_done = dmub_dcn21_is_auto_load_done; + funcs->is_phy_init = dmub_dcn21_is_phy_init; + } + break; + + default: + return false; + } + + return true; +} + +enum dmub_status dmub_srv_create(struct dmub_srv *dmub, + const struct dmub_srv_create_params *params) +{ + enum dmub_status status = DMUB_STATUS_OK; + + dmub_memset(dmub, 0, sizeof(*dmub)); + + dmub->funcs = params->funcs; + dmub->user_ctx = params->user_ctx; + dmub->asic = params->asic; + dmub->is_virtual = params->is_virtual; + + /* Setup asic dependent hardware funcs. */ + if (!dmub_srv_hw_setup(dmub, params->asic)) { + status = DMUB_STATUS_INVALID; + goto cleanup; + } + + /* Override (some) hardware funcs based on user params. */ + if (params->hw_funcs) { + if (params->hw_funcs->get_inbox1_rptr) + dmub->hw_funcs.get_inbox1_rptr = + params->hw_funcs->get_inbox1_rptr; + + if (params->hw_funcs->set_inbox1_wptr) + dmub->hw_funcs.set_inbox1_wptr = + params->hw_funcs->set_inbox1_wptr; + + if (params->hw_funcs->is_supported) + dmub->hw_funcs.is_supported = + params->hw_funcs->is_supported; + } + + /* Sanity checks for required hw func pointers. */ + if (!dmub->hw_funcs.get_inbox1_rptr || + !dmub->hw_funcs.set_inbox1_wptr) { + status = DMUB_STATUS_INVALID; + goto cleanup; + } + +cleanup: + if (status == DMUB_STATUS_OK) + dmub->sw_init = true; + else + dmub_srv_destroy(dmub); + + return status; +} + +void dmub_srv_destroy(struct dmub_srv *dmub) +{ + dmub_memset(dmub, 0, sizeof(*dmub)); +} + +enum dmub_status +dmub_srv_calc_region_info(struct dmub_srv *dmub, + const struct dmub_srv_region_params *params, + struct dmub_srv_region_info *out) +{ + struct dmub_region *inst = &out->regions[DMUB_WINDOW_0_INST_CONST]; + struct dmub_region *stack = &out->regions[DMUB_WINDOW_1_STACK]; + struct dmub_region *data = &out->regions[DMUB_WINDOW_2_BSS_DATA]; + struct dmub_region *bios = &out->regions[DMUB_WINDOW_3_VBIOS]; + struct dmub_region *mail = &out->regions[DMUB_WINDOW_4_MAILBOX]; + struct dmub_region *trace_buff = &out->regions[DMUB_WINDOW_5_TRACEBUFF]; + struct dmub_region *fw_state = &out->regions[DMUB_WINDOW_6_FW_STATE]; + const struct dmub_fw_meta_info *fw_info; + uint32_t fw_state_size = DMUB_FW_STATE_SIZE; + uint32_t trace_buffer_size = DMUB_TRACE_BUFFER_SIZE; + + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + memset(out, 0, sizeof(*out)); + + out->num_regions = DMUB_NUM_WINDOWS; + + inst->base = 0x0; + inst->top = inst->base + params->inst_const_size; + + data->base = dmub_align(inst->top, 256); + data->top = data->base + params->bss_data_size; + + /* + * All cache windows below should be aligned to the size + * of the DMCUB cache line, 64 bytes. + */ + + stack->base = dmub_align(data->top, 256); + stack->top = stack->base + DMUB_STACK_SIZE + DMUB_CONTEXT_SIZE; + + bios->base = dmub_align(stack->top, 256); + bios->top = bios->base + params->vbios_size; + + mail->base = dmub_align(bios->top, 256); + mail->top = mail->base + DMUB_MAILBOX_SIZE; + + fw_info = dmub_get_fw_meta_info(params->fw_bss_data, + params->bss_data_size); + + if (fw_info) { + fw_state_size = fw_info->fw_region_size; + trace_buffer_size = fw_info->trace_buffer_size; + } + + trace_buff->base = dmub_align(mail->top, 256); + trace_buff->top = trace_buff->base + dmub_align(trace_buffer_size, 64); + + fw_state->base = dmub_align(trace_buff->top, 256); + fw_state->top = fw_state->base + dmub_align(fw_state_size, 64); + + out->fb_size = dmub_align(fw_state->top, 4096); + + return DMUB_STATUS_OK; +} + +enum dmub_status dmub_srv_calc_fb_info(struct dmub_srv *dmub, + const struct dmub_srv_fb_params *params, + struct dmub_srv_fb_info *out) +{ + uint8_t *cpu_base; + uint64_t gpu_base; + uint32_t i; + + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + memset(out, 0, sizeof(*out)); + + if (params->region_info->num_regions != DMUB_NUM_WINDOWS) + return DMUB_STATUS_INVALID; + + cpu_base = (uint8_t *)params->cpu_addr; + gpu_base = params->gpu_addr; + + for (i = 0; i < DMUB_NUM_WINDOWS; ++i) { + const struct dmub_region *reg = + ¶ms->region_info->regions[i]; + + out->fb[i].cpu_addr = cpu_base + reg->base; + out->fb[i].gpu_addr = gpu_base + reg->base; + out->fb[i].size = reg->top - reg->base; + } + + out->num_fb = DMUB_NUM_WINDOWS; + + return DMUB_STATUS_OK; +} + +enum dmub_status dmub_srv_has_hw_support(struct dmub_srv *dmub, + bool *is_supported) +{ + *is_supported = false; + + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + if (dmub->hw_funcs.is_supported) + *is_supported = dmub->hw_funcs.is_supported(dmub); + + return DMUB_STATUS_OK; +} + +enum dmub_status dmub_srv_is_hw_init(struct dmub_srv *dmub, bool *is_hw_init) +{ + *is_hw_init = false; + + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + if (!dmub->hw_init) + return DMUB_STATUS_OK; + + if (dmub->hw_funcs.is_hw_init) + *is_hw_init = dmub->hw_funcs.is_hw_init(dmub); + + return DMUB_STATUS_OK; +} + +enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, + const struct dmub_srv_hw_params *params) +{ + struct dmub_fb *inst_fb = params->fb[DMUB_WINDOW_0_INST_CONST]; + struct dmub_fb *stack_fb = params->fb[DMUB_WINDOW_1_STACK]; + struct dmub_fb *data_fb = params->fb[DMUB_WINDOW_2_BSS_DATA]; + struct dmub_fb *bios_fb = params->fb[DMUB_WINDOW_3_VBIOS]; + struct dmub_fb *mail_fb = params->fb[DMUB_WINDOW_4_MAILBOX]; + struct dmub_fb *tracebuff_fb = params->fb[DMUB_WINDOW_5_TRACEBUFF]; + struct dmub_fb *fw_state_fb = params->fb[DMUB_WINDOW_6_FW_STATE]; + + struct dmub_rb_init_params rb_params; + struct dmub_window cw0, cw1, cw2, cw3, cw4, cw5, cw6; + struct dmub_region inbox1; + + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + dmub->fb_base = params->fb_base; + dmub->fb_offset = params->fb_offset; + dmub->psp_version = params->psp_version; + + if (inst_fb && data_fb) { + cw0.offset.quad_part = inst_fb->gpu_addr; + cw0.region.base = DMUB_CW0_BASE; + cw0.region.top = cw0.region.base + inst_fb->size - 1; + + cw1.offset.quad_part = stack_fb->gpu_addr; + cw1.region.base = DMUB_CW1_BASE; + cw1.region.top = cw1.region.base + stack_fb->size - 1; + + /** + * Read back all the instruction memory so we don't hang the + * DMCUB when backdoor loading if the write from x86 hasn't been + * flushed yet. This only occurs in backdoor loading. + */ + dmub_flush_buffer_mem(inst_fb); + + if (params->load_inst_const && dmub->hw_funcs.backdoor_load) + dmub->hw_funcs.backdoor_load(dmub, &cw0, &cw1); + } + + if (dmub->hw_funcs.reset) + dmub->hw_funcs.reset(dmub); + + if (inst_fb && data_fb && bios_fb && mail_fb && tracebuff_fb && + fw_state_fb) { + cw2.offset.quad_part = data_fb->gpu_addr; + cw2.region.base = DMUB_CW0_BASE + inst_fb->size; + cw2.region.top = cw2.region.base + data_fb->size; + + cw3.offset.quad_part = bios_fb->gpu_addr; + cw3.region.base = DMUB_CW3_BASE; + cw3.region.top = cw3.region.base + bios_fb->size; + + cw4.offset.quad_part = mail_fb->gpu_addr; + cw4.region.base = cw3.region.top + 1; + cw4.region.top = cw4.region.base + mail_fb->size; + + inbox1.base = cw4.region.base; + inbox1.top = cw4.region.top; + + cw5.offset.quad_part = tracebuff_fb->gpu_addr; + cw5.region.base = DMUB_CW5_BASE; + cw5.region.top = cw5.region.base + tracebuff_fb->size; + + cw6.offset.quad_part = fw_state_fb->gpu_addr; + cw6.region.base = DMUB_CW6_BASE; + cw6.region.top = cw6.region.base + fw_state_fb->size; + + dmub->fw_state = fw_state_fb->cpu_addr; + + if (dmub->hw_funcs.setup_windows) + dmub->hw_funcs.setup_windows(dmub, &cw2, &cw3, &cw4, + &cw5, &cw6); + + if (dmub->hw_funcs.setup_mailbox) + dmub->hw_funcs.setup_mailbox(dmub, &inbox1); + } + + if (mail_fb) { + dmub_memset(&rb_params, 0, sizeof(rb_params)); + rb_params.ctx = dmub; + rb_params.base_address = mail_fb->cpu_addr; + rb_params.capacity = DMUB_RB_SIZE; + + dmub_rb_init(&dmub->inbox1_rb, &rb_params); + } + + if (dmub->hw_funcs.reset_release) + dmub->hw_funcs.reset_release(dmub); + + dmub->hw_init = true; + + return DMUB_STATUS_OK; +} + +enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub) +{ + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + if (dmub->hw_init == false) + return DMUB_STATUS_OK; + + if (dmub->hw_funcs.reset) + dmub->hw_funcs.reset(dmub); + + dmub->hw_init = false; + + return DMUB_STATUS_OK; +} + +enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub, + const struct dmub_cmd_header *cmd) +{ + if (!dmub->hw_init) + return DMUB_STATUS_INVALID; + + if (dmub_rb_push_front(&dmub->inbox1_rb, cmd)) + return DMUB_STATUS_OK; + + return DMUB_STATUS_QUEUE_FULL; +} + +enum dmub_status dmub_srv_cmd_execute(struct dmub_srv *dmub) +{ + if (!dmub->hw_init) + return DMUB_STATUS_INVALID; + + /** + * Read back all the queued commands to ensure that they've + * been flushed to framebuffer memory. Otherwise DMCUB might + * read back stale, fully invalid or partially invalid data. + */ + dmub_rb_flush_pending(&dmub->inbox1_rb); + + dmub->hw_funcs.set_inbox1_wptr(dmub, dmub->inbox1_rb.wrpt); + return DMUB_STATUS_OK; +} + +enum dmub_status dmub_srv_wait_for_auto_load(struct dmub_srv *dmub, + uint32_t timeout_us) +{ + uint32_t i; + + if (!dmub->hw_init) + return DMUB_STATUS_INVALID; + + if (!dmub->hw_funcs.is_auto_load_done) + return DMUB_STATUS_OK; + + for (i = 0; i <= timeout_us; i += 100) { + if (dmub->hw_funcs.is_auto_load_done(dmub)) + return DMUB_STATUS_OK; + + udelay(100); + } + + return DMUB_STATUS_TIMEOUT; +} + +enum dmub_status dmub_srv_wait_for_phy_init(struct dmub_srv *dmub, + uint32_t timeout_us) +{ + uint32_t i = 0; + + if (!dmub->hw_init) + return DMUB_STATUS_INVALID; + + if (!dmub->hw_funcs.is_phy_init) + return DMUB_STATUS_OK; + + for (i = 0; i <= timeout_us; i += 10) { + if (dmub->hw_funcs.is_phy_init(dmub)) + return DMUB_STATUS_OK; + + udelay(10); + } + + return DMUB_STATUS_TIMEOUT; +} + +enum dmub_status dmub_srv_wait_for_idle(struct dmub_srv *dmub, + uint32_t timeout_us) +{ + uint32_t i; + + if (!dmub->hw_init) + return DMUB_STATUS_INVALID; + + for (i = 0; i <= timeout_us; ++i) { + dmub->inbox1_rb.rptr = dmub->hw_funcs.get_inbox1_rptr(dmub); + if (dmub_rb_empty(&dmub->inbox1_rb)) + return DMUB_STATUS_OK; + + udelay(1); + } + + return DMUB_STATUS_TIMEOUT; +} diff --git a/drivers/gpu/drm/amd/display/include/audio_types.h b/drivers/gpu/drm/amd/display/include/audio_types.h index 6364fbc24cfe..66a54da0641c 100644 --- a/drivers/gpu/drm/amd/display/include/audio_types.h +++ b/drivers/gpu/drm/amd/display/include/audio_types.h @@ -38,8 +38,8 @@ struct audio_crtc_info { uint32_t h_active; uint32_t v_active; uint32_t pixel_repetition; - uint32_t requested_pixel_clock; /* in KHz */ - uint32_t calculated_pixel_clock; /* in KHz */ + uint32_t requested_pixel_clock_100Hz; /* in 100Hz */ + uint32_t calculated_pixel_clock_100Hz; /* in 100Hz */ uint32_t refresh_rate; enum dc_color_depth color_depth; bool interlaced; diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h index 887e6a8597c4..a2903985b9e8 100644 --- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h +++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h @@ -134,24 +134,50 @@ #define PICASSO_A0 0x41 /* DCN1_01 */ #define RAVEN2_A0 0x81 +#define RAVEN2_15D8_REV_94 0x94 +#define RAVEN2_15D8_REV_95 0x95 +#define RAVEN2_15D8_REV_E3 0xE3 +#define RAVEN2_15D8_REV_E4 0xE4 +#define RAVEN2_15D8_REV_E9 0xE9 +#define RAVEN2_15D8_REV_EA 0xEA +#define RAVEN2_15D8_REV_EB 0xEB #define RAVEN1_F0 0xF0 #define RAVEN_UNKNOWN 0xFF - +#ifndef ASICREV_IS_RAVEN #define ASICREV_IS_RAVEN(eChipRev) ((eChipRev >= RAVEN_A0) && eChipRev < RAVEN_UNKNOWN) -#define ASICREV_IS_PICASSO(eChipRev) ((eChipRev >= PICASSO_A0) && (eChipRev < RAVEN2_A0)) -#define ASICREV_IS_RAVEN2(eChipRev) ((eChipRev >= RAVEN2_A0) && (eChipRev < 0xF0)) - +#endif +#define ASICREV_IS_PICASSO(eChipRev) ((eChipRev >= PICASSO_A0) && (eChipRev < RAVEN2_A0)) +#ifndef ASICREV_IS_RAVEN2 +#define ASICREV_IS_RAVEN2(eChipRev) ((eChipRev >= RAVEN2_A0) && (eChipRev < RAVEN1_F0)) +#endif #define ASICREV_IS_RV1_F0(eChipRev) ((eChipRev >= RAVEN1_F0) && (eChipRev < RAVEN_UNKNOWN)) - +#define ASICREV_IS_DALI(eChipRev) ((eChipRev == RAVEN2_15D8_REV_E3) \ + || (eChipRev == RAVEN2_15D8_REV_E4)) +#define ASICREV_IS_POLLOCK(eChipRev) (eChipRev == RAVEN2_15D8_REV_94 \ + || eChipRev == RAVEN2_15D8_REV_95 \ + || eChipRev == RAVEN2_15D8_REV_E9 \ + || eChipRev == RAVEN2_15D8_REV_EA \ + || eChipRev == RAVEN2_15D8_REV_EB) #define FAMILY_RV 142 /* DCN 1*/ -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) #define FAMILY_NV 143 /* DCN 2*/ -#endif +enum { + NV_NAVI10_P_A0 = 1, + NV_NAVI12_P_A0 = 10, + NV_NAVI14_M_A0 = 20, + NV_UNKNOWN = 0xFF +}; + +#define ASICREV_IS_NAVI10_P(eChipRev) (eChipRev < NV_NAVI12_P_A0) +#define ASICREV_IS_NAVI12_P(eChipRev) ((eChipRev >= NV_NAVI12_P_A0) && (eChipRev < NV_NAVI14_M_A0)) +#define ASICREV_IS_NAVI14_M(eChipRev) ((eChipRev >= NV_NAVI14_M_A0) && (eChipRev < NV_UNKNOWN)) +#define RENOIR_A0 0x91 +#define DEVICE_ID_RENOIR_1636 0x1636 // Renoir +#define ASICREV_IS_RENOIR(eChipRev) ((eChipRev >= RENOIR_A0) && (eChipRev < 0xFF)) /* * ASIC chip ID diff --git a/drivers/gpu/drm/amd/display/include/dal_types.h b/drivers/gpu/drm/amd/display/include/dal_types.h index 1e3ce4d847ae..0b6859189ca7 100644 --- a/drivers/gpu/drm/amd/display/include/dal_types.h +++ b/drivers/gpu/drm/amd/display/include/dal_types.h @@ -46,9 +46,8 @@ enum dce_version { DCE_VERSION_MAX, DCN_VERSION_1_0, DCN_VERSION_1_01, -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) DCN_VERSION_2_0, -#endif + DCN_VERSION_2_1, DCN_VERSION_MAX }; diff --git a/drivers/gpu/drm/amd/display/include/ddc_service_types.h b/drivers/gpu/drm/amd/display/include/ddc_service_types.h index d968956a10cd..9ad49da50a17 100644 --- a/drivers/gpu/drm/amd/display/include/ddc_service_types.h +++ b/drivers/gpu/drm/amd/display/include/ddc_service_types.h @@ -25,10 +25,14 @@ #ifndef __DAL_DDC_SERVICE_TYPES_H__ #define __DAL_DDC_SERVICE_TYPES_H__ -#define DP_BRANCH_DEVICE_ID_1 0x0010FA -#define DP_BRANCH_DEVICE_ID_2 0x0022B9 -#define DP_BRANCH_DEVICE_ID_3 0x00001A -#define DP_BRANCH_DEVICE_ID_4 0x0080e1 +/* 0010FA dongles (ST Micro) external converter chip id */ +#define DP_BRANCH_DEVICE_ID_0010FA 0x0010FA +/* 0022B9 external converter chip id */ +#define DP_BRANCH_DEVICE_ID_0022B9 0x0022B9 +#define DP_BRANCH_DEVICE_ID_00001A 0x00001A +#define DP_BRANCH_DEVICE_ID_0080E1 0x0080e1 +#define DP_BRANCH_DEVICE_ID_90CC24 0x90CC24 +#define DP_BRANCH_DEVICE_ID_00E04C 0x00E04C enum ddc_result { DDC_RESULT_UNKNOWN = 0, diff --git a/drivers/gpu/drm/amd/display/include/gpio_interface.h b/drivers/gpu/drm/amd/display/include/gpio_interface.h index 7de64195dc33..5e888a093c16 100644 --- a/drivers/gpu/drm/amd/display/include/gpio_interface.h +++ b/drivers/gpu/drm/amd/display/include/gpio_interface.h @@ -93,8 +93,17 @@ enum sync_source dal_gpio_get_sync_source( enum gpio_pin_output_state dal_gpio_get_output_state( const struct gpio *gpio); +struct hw_ddc *dal_gpio_get_ddc(struct gpio *gpio); + +struct hw_hpd *dal_gpio_get_hpd(struct gpio *gpio); + +struct hw_generic *dal_gpio_get_generic(struct gpio *gpio); + /* Close the handle */ void dal_gpio_close( struct gpio *gpio); + + + #endif diff --git a/drivers/gpu/drm/amd/display/include/gpio_service_interface.h b/drivers/gpu/drm/amd/display/include/gpio_service_interface.h index f40259bade40..9c55d247227e 100644 --- a/drivers/gpu/drm/amd/display/include/gpio_service_interface.h +++ b/drivers/gpu/drm/amd/display/include/gpio_service_interface.h @@ -51,13 +51,29 @@ struct gpio *dal_gpio_service_create_irq( uint32_t offset, uint32_t mask); +struct gpio *dal_gpio_service_create_generic_mux( + struct gpio_service *service, + uint32_t offset, + uint32_t mask); + +void dal_gpio_destroy_generic_mux( + struct gpio **mux); + +enum gpio_result dal_mux_setup_config( + struct gpio *mux, + struct gpio_generic_mux_config *config); + +struct gpio_pin_info dal_gpio_get_generic_pin_info( + struct gpio_service *service, + enum gpio_id id, + uint32_t en); + struct ddc *dal_gpio_create_ddc( struct gpio_service *service, uint32_t offset, uint32_t mask, struct gpio_ddc_hw_info *info); - void dal_gpio_destroy_ddc( struct ddc **ddc); diff --git a/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h b/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h index f312834fef50..d51de94e4bc3 100644 --- a/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h +++ b/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h @@ -178,7 +178,8 @@ struct dc_firmware_info { uint32_t default_engine_clk; /* in KHz */ uint32_t dp_phy_ref_clk; /* in KHz - DCE12 only */ uint32_t i2c_engine_ref_clk; /* in KHz - DCE12 only */ - + bool oem_i2c_present; + uint8_t oem_i2c_obj_id; }; diff --git a/drivers/gpu/drm/amd/display/include/hdcp_types.h b/drivers/gpu/drm/amd/display/include/hdcp_types.h new file mode 100644 index 000000000000..f31e6befc8d6 --- /dev/null +++ b/drivers/gpu/drm/amd/display/include/hdcp_types.h @@ -0,0 +1,96 @@ +/* +* Copyright 2019 Advanced Micro Devices, Inc. +* +* Permission is hereby granted, free of charge, to any person obtaining a +* copy of this software and associated documentation files (the "Software"), +* to deal in the Software without restriction, including without limitation +* the rights to use, copy, modify, merge, publish, distribute, sublicense, +* and/or sell copies of the Software, and to permit persons to whom the +* Software is furnished to do so, subject to the following conditions: +* +* The above copyright notice and this permission notice shall be included in +* all copies or substantial portions of the Software. +* +* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +* THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR +* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +* OTHER DEALINGS IN THE SOFTWARE. +* +* Authors: AMD +* +*/ + +#ifndef __DC_HDCP_TYPES_H__ +#define __DC_HDCP_TYPES_H__ + +enum hdcp_message_id { + HDCP_MESSAGE_ID_INVALID = -1, + + /* HDCP 1.4 */ + + HDCP_MESSAGE_ID_READ_BKSV = 0, + /* HDMI is called Ri', DP is called R0' */ + HDCP_MESSAGE_ID_READ_RI_R0, + HDCP_MESSAGE_ID_READ_PJ, + HDCP_MESSAGE_ID_WRITE_AKSV, + HDCP_MESSAGE_ID_WRITE_AINFO, + HDCP_MESSAGE_ID_WRITE_AN, + HDCP_MESSAGE_ID_READ_VH_X, + HDCP_MESSAGE_ID_READ_VH_0, + HDCP_MESSAGE_ID_READ_VH_1, + HDCP_MESSAGE_ID_READ_VH_2, + HDCP_MESSAGE_ID_READ_VH_3, + HDCP_MESSAGE_ID_READ_VH_4, + HDCP_MESSAGE_ID_READ_BCAPS, + HDCP_MESSAGE_ID_READ_BSTATUS, + HDCP_MESSAGE_ID_READ_KSV_FIFO, + HDCP_MESSAGE_ID_READ_BINFO, + + /* HDCP 2.2 */ + + HDCP_MESSAGE_ID_HDCP2VERSION, + HDCP_MESSAGE_ID_RX_CAPS, + HDCP_MESSAGE_ID_WRITE_AKE_INIT, + HDCP_MESSAGE_ID_READ_AKE_SEND_CERT, + HDCP_MESSAGE_ID_WRITE_AKE_NO_STORED_KM, + HDCP_MESSAGE_ID_WRITE_AKE_STORED_KM, + HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME, + HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO, + HDCP_MESSAGE_ID_WRITE_LC_INIT, + HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME, + HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS, + HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, + HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK, + HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE, + HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY, + HDCP_MESSAGE_ID_READ_RXSTATUS, + HDCP_MESSAGE_ID_WRITE_CONTENT_STREAM_TYPE, + + HDCP_MESSAGE_ID_MAX +}; + +enum hdcp_version { + HDCP_Unknown = 0, + HDCP_VERSION_14, + HDCP_VERSION_22, +}; + +enum hdcp_link { + HDCP_LINK_PRIMARY, + HDCP_LINK_SECONDARY +}; + +struct hdcp_protection_message { + enum hdcp_version version; + /* relevant only for DVI */ + enum hdcp_link link; + enum hdcp_message_id msg_id; + uint32_t length; + uint8_t max_retries; + uint8_t *data; +}; + +#endif diff --git a/drivers/gpu/drm/amd/display/include/i2caux_interface.h b/drivers/gpu/drm/amd/display/include/i2caux_interface.h index bb012cb1a9f5..c7fbb9c3ad6b 100644 --- a/drivers/gpu/drm/amd/display/include/i2caux_interface.h +++ b/drivers/gpu/drm/amd/display/include/i2caux_interface.h @@ -42,7 +42,7 @@ struct aux_payload { bool write; bool mot; uint32_t address; - uint8_t length; + uint32_t length; uint8_t *data; /* * used to return the reply type of the transaction diff --git a/drivers/gpu/drm/amd/display/include/link_service_types.h b/drivers/gpu/drm/amd/display/include/link_service_types.h index 80f0d93cfd94..4869d4562e4d 100644 --- a/drivers/gpu/drm/amd/display/include/link_service_types.h +++ b/drivers/gpu/drm/amd/display/include/link_service_types.h @@ -71,14 +71,17 @@ enum link_training_result { struct link_training_settings { struct dc_link_settings link_settings; struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]; - bool allow_invalid_msa_timing_param; -}; -enum hw_dp_training_pattern { - HW_DP_TRAINING_PATTERN_1 = 0, - HW_DP_TRAINING_PATTERN_2, - HW_DP_TRAINING_PATTERN_3, - HW_DP_TRAINING_PATTERN_4 + enum dc_voltage_swing *voltage_swing; + enum dc_pre_emphasis *pre_emphasis; + enum dc_post_cursor2 *post_cursor2; + + uint16_t cr_pattern_time; + uint16_t eq_pattern_time; + enum dc_dp_training_pattern pattern_for_eq; + + bool enhanced_framing; + bool allow_invalid_msa_timing_param; }; /*TODO: Move this enum test harness*/ @@ -120,6 +123,13 @@ enum dp_test_pattern { DP_TEST_PATTERN_UNSUPPORTED }; +enum dp_test_pattern_color_space { + DP_TEST_PATTERN_COLOR_SPACE_RGB, + DP_TEST_PATTERN_COLOR_SPACE_YCBCR601, + DP_TEST_PATTERN_COLOR_SPACE_YCBCR709, + DP_TEST_PATTERN_COLOR_SPACE_UNDEFINED +}; + enum dp_panel_mode { /* not required */ DP_PANEL_MODE_DEFAULT, diff --git a/drivers/gpu/drm/amd/display/include/logger_interface.h b/drivers/gpu/drm/amd/display/include/logger_interface.h index a0b68c266dab..6e008de25629 100644 --- a/drivers/gpu/drm/amd/display/include/logger_interface.h +++ b/drivers/gpu/drm/amd/display/include/logger_interface.h @@ -155,4 +155,6 @@ void context_clock_trace( #define DISPLAY_STATS_END(entry) (void)(entry) +#define LOG_GAMMA_WRITE(msg, ...) + #endif /* __DAL_LOGGER_INTERFACE_H__ */ diff --git a/drivers/gpu/drm/amd/display/include/logger_types.h b/drivers/gpu/drm/amd/display/include/logger_types.h index ea8d445816b8..89a709267019 100644 --- a/drivers/gpu/drm/amd/display/include/logger_types.h +++ b/drivers/gpu/drm/amd/display/include/logger_types.h @@ -63,12 +63,11 @@ #define DC_LOG_IF_TRACE(...) pr_debug("[IF_TRACE]:"__VA_ARGS__) #define DC_LOG_PERF_TRACE(...) DRM_DEBUG_KMS(__VA_ARGS__) #define DC_LOG_RETIMER_REDRIVER(...) DRM_DEBUG_KMS(__VA_ARGS__) -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT +#define DC_LOG_GAMMA(...) pr_debug("[GAMMA]:"__VA_ARGS__) +#define DC_LOG_ALL_GAMMA(...) pr_debug("[GAMMA]:"__VA_ARGS__) +#define DC_LOG_ALL_TF_CHANNELS(...) pr_debug("[GAMMA]:"__VA_ARGS__) #define DC_LOG_DSC(...) DRM_DEBUG_KMS(__VA_ARGS__) -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN3_0) || defined(CONFIG_DRM_AMD_DC_DCN2_0) #define DC_LOG_DWB(...) DRM_DEBUG_KMS(__VA_ARGS__) -#endif struct dal_logger; @@ -113,10 +112,12 @@ enum dc_log_type { LOG_PERF_TRACE, LOG_DISPLAYSTATS, LOG_HDMI_RETIMER_REDRIVER, -#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT LOG_DSC, -#endif LOG_DWB, + LOG_GAMMA_DEBUG, + LOG_MAX_HW_POINTS, + LOG_ALL_TF_CHANNELS, + LOG_SAMPLE_1DLUT, LOG_SECTION_TOTAL_COUNT }; diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c index 88898935a5e6..cac09d500fda 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c @@ -30,7 +30,6 @@ #include "opp.h" #include "color_gamma.h" - #define NUM_PTS_IN_REGION 16 #define NUM_REGIONS 32 #define MAX_HW_POINTS (NUM_PTS_IN_REGION*NUM_REGIONS) @@ -40,6 +39,33 @@ static struct hw_x_point coordinates_x[MAX_HW_POINTS + 2]; static struct fixed31_32 pq_table[MAX_HW_POINTS + 2]; static struct fixed31_32 de_pq_table[MAX_HW_POINTS + 2]; +// these are helpers for calculations to reduce stack usage +// do not depend on these being preserved across calls +static struct fixed31_32 scratch_1; +static struct fixed31_32 scratch_2; +static struct translate_from_linear_space_args scratch_gamma_args; + +/* Helper to optimize gamma calculation, only use in translate_from_linear, in + * particular the dc_fixpt_pow function which is very expensive + * The idea is that our regions for X points are exponential and currently they all use + * the same number of points (NUM_PTS_IN_REGION) and in each region every point + * is exactly 2x the one at the same index in the previous region. In other words + * X[i] = 2 * X[i-NUM_PTS_IN_REGION] for i>=16 + * The other fact is that (2x)^gamma = 2^gamma * x^gamma + * So we compute and save x^gamma for the first 16 regions, and for every next region + * just multiply with 2^gamma which can be computed once, and save the result so we + * recursively compute all the values. + */ +static struct fixed31_32 pow_buffer[NUM_PTS_IN_REGION]; +static struct fixed31_32 gamma_of_2; // 2^gamma +int pow_buffer_ptr = -1; + /*sRGB 709 2.2 2.4 P3*/ +static const int32_t gamma_numerator01[] = { 31308, 180000, 0, 0, 0}; +static const int32_t gamma_numerator02[] = { 12920, 4500, 0, 0, 0}; +static const int32_t gamma_numerator03[] = { 55, 99, 0, 0, 0}; +static const int32_t gamma_numerator04[] = { 55, 99, 0, 0, 0}; +static const int32_t gamma_numerator05[] = { 2400, 2200, 2200, 2400, 2600}; + static bool pq_initialized; /* = false; */ static bool de_pq_initialized; /* = false; */ @@ -71,6 +97,18 @@ void setup_x_points_distribution(void) } } +void log_x_points_distribution(struct dal_logger *logger) +{ + int i = 0; + + if (logger != NULL) { + LOG_GAMMA_WRITE("Log X Distribution\n"); + + for (i = 0; i < MAX_HW_POINTS; i++) + LOG_GAMMA_WRITE("%llu\n", coordinates_x[i].x.value); + } +} + static void compute_pq(struct fixed31_32 in_x, struct fixed31_32 *out_y) { /* consts for PQ gamma formula. */ @@ -116,6 +154,7 @@ static void compute_de_pq(struct fixed31_32 in_x, struct fixed31_32 *out_y) struct fixed31_32 l_pow_m1; struct fixed31_32 base, div; + struct fixed31_32 base2; if (dc_fixpt_lt(in_x, dc_fixpt_zero)) @@ -125,69 +164,80 @@ static void compute_de_pq(struct fixed31_32 in_x, struct fixed31_32 *out_y) dc_fixpt_div(dc_fixpt_one, m2)); base = dc_fixpt_sub(l_pow_m1, c1); - if (dc_fixpt_lt(base, dc_fixpt_zero)) - base = dc_fixpt_zero; - div = dc_fixpt_sub(c2, dc_fixpt_mul(c3, l_pow_m1)); - *out_y = dc_fixpt_pow(dc_fixpt_div(base, div), - dc_fixpt_div(dc_fixpt_one, m1)); + base2 = dc_fixpt_div(base, div); + //avoid complex numbers + if (dc_fixpt_lt(base2, dc_fixpt_zero)) + base2 = dc_fixpt_sub(dc_fixpt_zero, base2); + + + *out_y = dc_fixpt_pow(base2, dc_fixpt_div(dc_fixpt_one, m1)); } + /*de gamma, none linear to linear*/ -static void compute_hlg_oetf(struct fixed31_32 in_x, bool is_light0_12, struct fixed31_32 *out_y) +static void compute_hlg_eotf(struct fixed31_32 in_x, + struct fixed31_32 *out_y, + uint32_t sdr_white_level, uint32_t max_luminance_nits) { struct fixed31_32 a; struct fixed31_32 b; struct fixed31_32 c; struct fixed31_32 threshold; - struct fixed31_32 reference_white_level; + struct fixed31_32 x; + struct fixed31_32 scaling_factor = + dc_fixpt_from_fraction(max_luminance_nits, sdr_white_level); a = dc_fixpt_from_fraction(17883277, 100000000); - if (is_light0_12) { - /*light 0-12*/ - b = dc_fixpt_from_fraction(28466892, 100000000); - c = dc_fixpt_from_fraction(55991073, 100000000); - threshold = dc_fixpt_one; - reference_white_level = dc_fixpt_half; + b = dc_fixpt_from_fraction(28466892, 100000000); + c = dc_fixpt_from_fraction(55991073, 100000000); + threshold = dc_fixpt_from_fraction(1, 2); + + if (dc_fixpt_lt(in_x, threshold)) { + x = dc_fixpt_mul(in_x, in_x); + x = dc_fixpt_div_int(x, 3); } else { - /*light 0-1*/ - b = dc_fixpt_from_fraction(2372241, 100000000); - c = dc_fixpt_add(dc_fixpt_one, dc_fixpt_from_fraction(429347, 100000000)); - threshold = dc_fixpt_from_fraction(1, 12); - reference_white_level = dc_fixpt_pow(dc_fixpt_from_fraction(3, 1), dc_fixpt_half); + x = dc_fixpt_sub(in_x, c); + x = dc_fixpt_div(x, a); + x = dc_fixpt_exp(x); + x = dc_fixpt_add(x, b); + x = dc_fixpt_div_int(x, 12); } - if (dc_fixpt_lt(threshold, in_x)) - *out_y = dc_fixpt_add(c, dc_fixpt_mul(a, dc_fixpt_log(dc_fixpt_sub(in_x, b)))); - else - *out_y = dc_fixpt_mul(dc_fixpt_pow(in_x, dc_fixpt_half), reference_white_level); + *out_y = dc_fixpt_mul(x, scaling_factor); + } /*re gamma, linear to none linear*/ -static void compute_hlg_eotf(struct fixed31_32 in_x, bool is_light0_12, struct fixed31_32 *out_y) +static void compute_hlg_oetf(struct fixed31_32 in_x, struct fixed31_32 *out_y, + uint32_t sdr_white_level, uint32_t max_luminance_nits) { struct fixed31_32 a; struct fixed31_32 b; struct fixed31_32 c; - struct fixed31_32 reference_white_level; + struct fixed31_32 threshold; + struct fixed31_32 x; + struct fixed31_32 scaling_factor = + dc_fixpt_from_fraction(sdr_white_level, max_luminance_nits); a = dc_fixpt_from_fraction(17883277, 100000000); - if (is_light0_12) { - /*light 0-12*/ - b = dc_fixpt_from_fraction(28466892, 100000000); - c = dc_fixpt_from_fraction(55991073, 100000000); - reference_white_level = dc_fixpt_from_fraction(4, 1); + b = dc_fixpt_from_fraction(28466892, 100000000); + c = dc_fixpt_from_fraction(55991073, 100000000); + threshold = dc_fixpt_from_fraction(1, 12); + x = dc_fixpt_mul(in_x, scaling_factor); + + + if (dc_fixpt_lt(x, threshold)) { + x = dc_fixpt_mul(x, dc_fixpt_from_fraction(3, 1)); + *out_y = dc_fixpt_pow(x, dc_fixpt_half); } else { - /*light 0-1*/ - b = dc_fixpt_from_fraction(2372241, 100000000); - c = dc_fixpt_add(dc_fixpt_one, dc_fixpt_from_fraction(429347, 100000000)); - reference_white_level = dc_fixpt_from_fraction(1, 3); + x = dc_fixpt_mul(x, dc_fixpt_from_fraction(12, 1)); + x = dc_fixpt_sub(x, b); + x = dc_fixpt_log(x); + x = dc_fixpt_mul(a, x); + *out_y = dc_fixpt_add(x, c); } - if (dc_fixpt_lt(dc_fixpt_half, in_x)) - *out_y = dc_fixpt_add(dc_fixpt_exp(dc_fixpt_div(dc_fixpt_sub(in_x, c), a)), b); - else - *out_y = dc_fixpt_mul(dc_fixpt_pow(in_x, dc_fixpt_from_fraction(2, 1)), reference_white_level); } @@ -243,95 +293,144 @@ struct dividers { struct fixed31_32 divider3; }; -enum gamma_type_index { - gamma_type_index_2_4, - gamma_type_index_2_2, - gamma_type_index_2_2_flat -}; -static void build_coefficients(struct gamma_coefficients *coefficients, enum gamma_type_index type) +static bool build_coefficients(struct gamma_coefficients *coefficients, enum dc_transfer_func_predefined type) { - static const int32_t numerator01[] = { 31308, 180000, 0}; - static const int32_t numerator02[] = { 12920, 4500, 0}; - static const int32_t numerator03[] = { 55, 99, 0}; - static const int32_t numerator04[] = { 55, 99, 0}; - static const int32_t numerator05[] = { 2400, 2200, 2200}; uint32_t i = 0; uint32_t index = 0; + bool ret = true; - if (type == gamma_type_index_2_2) + if (type == TRANSFER_FUNCTION_SRGB) + index = 0; + else if (type == TRANSFER_FUNCTION_BT709) index = 1; - else if (type == gamma_type_index_2_2_flat) + else if (type == TRANSFER_FUNCTION_GAMMA22) index = 2; + else if (type == TRANSFER_FUNCTION_GAMMA24) + index = 3; + else if (type == TRANSFER_FUNCTION_GAMMA26) + index = 4; + else { + ret = false; + goto release; + } do { coefficients->a0[i] = dc_fixpt_from_fraction( - numerator01[index], 10000000); + gamma_numerator01[index], 10000000); coefficients->a1[i] = dc_fixpt_from_fraction( - numerator02[index], 1000); + gamma_numerator02[index], 1000); coefficients->a2[i] = dc_fixpt_from_fraction( - numerator03[index], 1000); + gamma_numerator03[index], 1000); coefficients->a3[i] = dc_fixpt_from_fraction( - numerator04[index], 1000); + gamma_numerator04[index], 1000); coefficients->user_gamma[i] = dc_fixpt_from_fraction( - numerator05[index], 1000); + gamma_numerator05[index], 1000); ++i; } while (i != ARRAY_SIZE(coefficients->a0)); +release: + return ret; } static struct fixed31_32 translate_from_linear_space( - struct fixed31_32 arg, - struct fixed31_32 a0, - struct fixed31_32 a1, - struct fixed31_32 a2, - struct fixed31_32 a3, - struct fixed31_32 gamma) + struct translate_from_linear_space_args *args) { const struct fixed31_32 one = dc_fixpt_from_int(1); - if (dc_fixpt_lt(one, arg)) + if (dc_fixpt_le(one, args->arg)) return one; - if (dc_fixpt_le(arg, dc_fixpt_neg(a0))) + if (dc_fixpt_le(args->arg, dc_fixpt_neg(args->a0))) { + scratch_1 = dc_fixpt_add(one, args->a3); + scratch_2 = dc_fixpt_pow( + dc_fixpt_neg(args->arg), + dc_fixpt_recip(args->gamma)); + scratch_1 = dc_fixpt_mul(scratch_1, scratch_2); + scratch_1 = dc_fixpt_sub(args->a2, scratch_1); + + return scratch_1; + } else if (dc_fixpt_le(args->a0, args->arg)) { + if (pow_buffer_ptr == 0) { + gamma_of_2 = dc_fixpt_pow(dc_fixpt_from_int(2), + dc_fixpt_recip(args->gamma)); + } + scratch_1 = dc_fixpt_add(one, args->a3); + if (pow_buffer_ptr < 16) + scratch_2 = dc_fixpt_pow(args->arg, + dc_fixpt_recip(args->gamma)); + else + scratch_2 = dc_fixpt_mul(gamma_of_2, + pow_buffer[pow_buffer_ptr%16]); + + if (pow_buffer_ptr != -1) { + pow_buffer[pow_buffer_ptr%16] = scratch_2; + pow_buffer_ptr++; + } + + scratch_1 = dc_fixpt_mul(scratch_1, scratch_2); + scratch_1 = dc_fixpt_sub(scratch_1, args->a2); + + return scratch_1; + } + else + return dc_fixpt_mul(args->arg, args->a1); +} + + +static struct fixed31_32 translate_from_linear_space_long( + struct translate_from_linear_space_args *args) +{ + const struct fixed31_32 one = dc_fixpt_from_int(1); + + if (dc_fixpt_lt(one, args->arg)) + return one; + + if (dc_fixpt_le(args->arg, dc_fixpt_neg(args->a0))) return dc_fixpt_sub( - a2, + args->a2, dc_fixpt_mul( dc_fixpt_add( one, - a3), + args->a3), dc_fixpt_pow( - dc_fixpt_neg(arg), - dc_fixpt_recip(gamma)))); - else if (dc_fixpt_le(a0, arg)) + dc_fixpt_neg(args->arg), + dc_fixpt_recip(args->gamma)))); + else if (dc_fixpt_le(args->a0, args->arg)) return dc_fixpt_sub( dc_fixpt_mul( dc_fixpt_add( one, - a3), + args->a3), dc_fixpt_pow( - arg, - dc_fixpt_recip(gamma))), - a2); + args->arg, + dc_fixpt_recip(args->gamma))), + args->a2); else return dc_fixpt_mul( - arg, - a1); + args->arg, + args->a1); } -static struct fixed31_32 calculate_gamma22(struct fixed31_32 arg) +static struct fixed31_32 calculate_gamma22(struct fixed31_32 arg, bool use_eetf) { struct fixed31_32 gamma = dc_fixpt_from_fraction(22, 10); - return translate_from_linear_space(arg, - dc_fixpt_zero, - dc_fixpt_zero, - dc_fixpt_zero, - dc_fixpt_zero, - gamma); + scratch_gamma_args.arg = arg; + scratch_gamma_args.a0 = dc_fixpt_zero; + scratch_gamma_args.a1 = dc_fixpt_zero; + scratch_gamma_args.a2 = dc_fixpt_zero; + scratch_gamma_args.a3 = dc_fixpt_zero; + scratch_gamma_args.gamma = gamma; + + if (use_eetf) + return translate_from_linear_space_long(&scratch_gamma_args); + + return translate_from_linear_space(&scratch_gamma_args); } + static struct fixed31_32 translate_to_linear_space( struct fixed31_32 arg, struct fixed31_32 a0, @@ -365,18 +464,19 @@ static struct fixed31_32 translate_to_linear_space( return linear; } -static inline struct fixed31_32 translate_from_linear_space_ex( +static struct fixed31_32 translate_from_linear_space_ex( struct fixed31_32 arg, struct gamma_coefficients *coeff, uint32_t color_index) { - return translate_from_linear_space( - arg, - coeff->a0[color_index], - coeff->a1[color_index], - coeff->a2[color_index], - coeff->a3[color_index], - coeff->user_gamma[color_index]); + scratch_gamma_args.arg = arg; + scratch_gamma_args.a0 = coeff->a0[color_index]; + scratch_gamma_args.a1 = coeff->a1[color_index]; + scratch_gamma_args.a2 = coeff->a2[color_index]; + scratch_gamma_args.a3 = coeff->a3[color_index]; + scratch_gamma_args.gamma = coeff->user_gamma[color_index]; + + return translate_from_linear_space(&scratch_gamma_args); } @@ -709,30 +809,42 @@ static void build_de_pq(struct pwl_float_data_ex *de_pq, } } -static void build_regamma(struct pwl_float_data_ex *rgb_regamma, +static bool build_regamma(struct pwl_float_data_ex *rgb_regamma, uint32_t hw_points_num, - const struct hw_x_point *coordinate_x, enum gamma_type_index type) + const struct hw_x_point *coordinate_x, enum dc_transfer_func_predefined type) { uint32_t i; + bool ret = false; - struct gamma_coefficients coeff; + struct gamma_coefficients *coeff; struct pwl_float_data_ex *rgb = rgb_regamma; const struct hw_x_point *coord_x = coordinate_x; - build_coefficients(&coeff, type); + coeff = kvzalloc(sizeof(*coeff), GFP_KERNEL); + if (!coeff) + goto release; + + if (!build_coefficients(coeff, type)) + goto release; + memset(pow_buffer, 0, NUM_PTS_IN_REGION * sizeof(struct fixed31_32)); + pow_buffer_ptr = 0; // see variable definition for more info i = 0; - - while (i != hw_points_num + 1) { + while (i <= hw_points_num) { /*TODO use y vs r,g,b*/ rgb->r = translate_from_linear_space_ex( - coord_x->x, &coeff, 0); + coord_x->x, coeff, 0); rgb->g = rgb->r; rgb->b = rgb->r; ++coord_x; ++rgb; ++i; } + pow_buffer_ptr = -1; // reset back to no optimize + ret = true; +release: + kfree(coeff); + return ret; } static void hermite_spline_eetf(struct fixed31_32 input_x, @@ -830,7 +942,6 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma, struct fixed31_32 max_display; struct fixed31_32 min_display; struct fixed31_32 max_content; - struct fixed31_32 min_content; struct fixed31_32 clip = dc_fixpt_one; struct fixed31_32 output; bool use_eetf = false; @@ -844,7 +955,6 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma, max_display = dc_fixpt_from_int(fs_params->max_display); min_display = dc_fixpt_from_fraction(fs_params->min_display, 10000); max_content = dc_fixpt_from_int(fs_params->max_content); - min_content = dc_fixpt_from_fraction(fs_params->min_content, 10000); sdr_white_level = dc_fixpt_from_int(fs_params->sdr_white_level); if (fs_params->min_display > 1000) // cap at 0.1 at the bottom @@ -852,16 +962,14 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma, if (fs_params->max_display < 100) // cap at 100 at the top max_display = dc_fixpt_from_int(100); - if (fs_params->min_content < fs_params->min_display) - use_eetf = true; - else - min_content = min_display; - + // only max used, we don't adjust min luminance if (fs_params->max_content > fs_params->max_display) use_eetf = true; else max_content = max_display; + if (!use_eetf) + pow_buffer_ptr = 0; // see var definition for more info rgb += 32; // first 32 points have problems with fixed point, too small coord_x += 32; for (i = 32; i <= hw_points_num; i++) { @@ -880,7 +988,7 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma, if (dc_fixpt_lt(scaledX, dc_fixpt_zero)) output = dc_fixpt_zero; else - output = calculate_gamma22(scaledX); + output = calculate_gamma22(scaledX, use_eetf); rgb->r = output; rgb->g = output; @@ -900,19 +1008,23 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma, ++coord_x; ++rgb; } + pow_buffer_ptr = -1; return true; } -static void build_degamma(struct pwl_float_data_ex *curve, +static bool build_degamma(struct pwl_float_data_ex *curve, uint32_t hw_points_num, - const struct hw_x_point *coordinate_x, enum gamma_type_index type) + const struct hw_x_point *coordinate_x, enum dc_transfer_func_predefined type) { uint32_t i; struct gamma_coefficients coeff; uint32_t begin_index, end_index; + bool ret = false; + + if (!build_coefficients(&coeff, type)) + goto release; - build_coefficients(&coeff, type); i = 0; /* X points is 2^-25 to 2^7 @@ -941,11 +1053,19 @@ static void build_degamma(struct pwl_float_data_ex *curve, curve[i].b = dc_fixpt_one; i++; } + ret = true; +release: + return ret; } + + + + static void build_hlg_degamma(struct pwl_float_data_ex *degamma, uint32_t hw_points_num, - const struct hw_x_point *coordinate_x, bool is_light0_12) + const struct hw_x_point *coordinate_x, + uint32_t sdr_white_level, uint32_t max_luminance_nits) { uint32_t i; @@ -953,9 +1073,9 @@ static void build_hlg_degamma(struct pwl_float_data_ex *degamma, const struct hw_x_point *coord_x = coordinate_x; i = 0; - + //check when i == 434 while (i != hw_points_num + 1) { - compute_hlg_oetf(coord_x->x, is_light0_12, &rgb->r); + compute_hlg_eotf(coord_x->x, &rgb->r, sdr_white_level, max_luminance_nits); rgb->g = rgb->r; rgb->b = rgb->r; ++coord_x; @@ -964,9 +1084,11 @@ static void build_hlg_degamma(struct pwl_float_data_ex *degamma, } } + static void build_hlg_regamma(struct pwl_float_data_ex *regamma, uint32_t hw_points_num, - const struct hw_x_point *coordinate_x, bool is_light0_12) + const struct hw_x_point *coordinate_x, + uint32_t sdr_white_level, uint32_t max_luminance_nits) { uint32_t i; @@ -975,8 +1097,9 @@ static void build_hlg_regamma(struct pwl_float_data_ex *regamma, i = 0; + //when i == 471 while (i != hw_points_num + 1) { - compute_hlg_eotf(coord_x->x, is_light0_12, &rgb->r); + compute_hlg_oetf(coord_x->x, &rgb->r, sdr_white_level, max_luminance_nits); rgb->g = rgb->r; rgb->b = rgb->r; ++coord_x; @@ -1550,124 +1673,6 @@ static bool map_regamma_hw_to_x_user( #define _EXTRA_POINTS 3 -bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf, - const struct dc_gamma *ramp, bool mapUserRamp, bool canRomBeUsed, - const struct freesync_hdr_tf_params *fs_params) -{ - struct dc_transfer_func_distributed_points *tf_pts = &output_tf->tf_pts; - struct dividers dividers; - - struct pwl_float_data *rgb_user = NULL; - struct pwl_float_data_ex *rgb_regamma = NULL; - struct gamma_pixel *axis_x = NULL; - struct pixel_gamma_point *coeff = NULL; - enum dc_transfer_func_predefined tf = TRANSFER_FUNCTION_SRGB; - bool ret = false; - - if (output_tf->type == TF_TYPE_BYPASS) - return false; - - /* we can use hardcoded curve for plain SRGB TF */ - if (output_tf->type == TF_TYPE_PREDEFINED && canRomBeUsed == true && - output_tf->tf == TRANSFER_FUNCTION_SRGB) { - if (ramp == NULL) - return true; - if ((ramp->is_logical_identity) || - (!mapUserRamp && ramp->type == GAMMA_RGB_256)) - return true; - } - - output_tf->type = TF_TYPE_DISTRIBUTED_POINTS; - - if (ramp && (mapUserRamp || ramp->type != GAMMA_RGB_256)) { - rgb_user = kvcalloc(ramp->num_entries + _EXTRA_POINTS, - sizeof(*rgb_user), - GFP_KERNEL); - if (!rgb_user) - goto rgb_user_alloc_fail; - - axis_x = kvcalloc(ramp->num_entries + 3, sizeof(*axis_x), - GFP_KERNEL); - if (!axis_x) - goto axis_x_alloc_fail; - - dividers.divider1 = dc_fixpt_from_fraction(3, 2); - dividers.divider2 = dc_fixpt_from_int(2); - dividers.divider3 = dc_fixpt_from_fraction(5, 2); - - build_evenly_distributed_points( - axis_x, - ramp->num_entries, - dividers); - - if (ramp->type == GAMMA_RGB_256 && mapUserRamp) - scale_gamma(rgb_user, ramp, dividers); - else if (ramp->type == GAMMA_RGB_FLOAT_1024) - scale_gamma_dx(rgb_user, ramp, dividers); - } - - rgb_regamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, - sizeof(*rgb_regamma), - GFP_KERNEL); - if (!rgb_regamma) - goto rgb_regamma_alloc_fail; - - coeff = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*coeff), - GFP_KERNEL); - if (!coeff) - goto coeff_alloc_fail; - - tf = output_tf->tf; - if (tf == TRANSFER_FUNCTION_PQ) { - tf_pts->end_exponent = 7; - tf_pts->x_point_at_y1_red = 125; - tf_pts->x_point_at_y1_green = 125; - tf_pts->x_point_at_y1_blue = 125; - - build_pq(rgb_regamma, - MAX_HW_POINTS, - coordinates_x, - output_tf->sdr_ref_white_level); - } else if (tf == TRANSFER_FUNCTION_GAMMA22 && - fs_params != NULL && fs_params->skip_tm == 0) { - build_freesync_hdr(rgb_regamma, - MAX_HW_POINTS, - coordinates_x, - fs_params); - } else { - tf_pts->end_exponent = 0; - tf_pts->x_point_at_y1_red = 1; - tf_pts->x_point_at_y1_green = 1; - tf_pts->x_point_at_y1_blue = 1; - - build_regamma(rgb_regamma, - MAX_HW_POINTS, - coordinates_x, tf == TRANSFER_FUNCTION_SRGB ? gamma_type_index_2_4 : - tf == TRANSFER_FUNCTION_GAMMA22 ? - gamma_type_index_2_2_flat : gamma_type_index_2_2); - } - map_regamma_hw_to_x_user(ramp, coeff, rgb_user, - coordinates_x, axis_x, rgb_regamma, - MAX_HW_POINTS, tf_pts, - (mapUserRamp || (ramp && ramp->type != GAMMA_RGB_256)) && - (ramp && ramp->type != GAMMA_CS_TFM_1D)); - - if (ramp && ramp->type == GAMMA_CS_TFM_1D) - apply_lut_1d(ramp, MAX_HW_POINTS, tf_pts); - - ret = true; - - kvfree(coeff); -coeff_alloc_fail: - kvfree(rgb_regamma); -rgb_regamma_alloc_fail: - kvfree(axis_x); -axis_x_alloc_fail: - kvfree(rgb_user); -rgb_user_alloc_fail: - return ret; -} - bool calculate_user_regamma_coeff(struct dc_transfer_func *output_tf, const struct regamma_lut *regamma) { @@ -1845,13 +1850,19 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, MAX_HW_POINTS, coordinates_x); else if (tf == TRANSFER_FUNCTION_SRGB || - tf == TRANSFER_FUNCTION_BT709) + tf == TRANSFER_FUNCTION_BT709 || + tf == TRANSFER_FUNCTION_GAMMA22 || + tf == TRANSFER_FUNCTION_GAMMA24 || + tf == TRANSFER_FUNCTION_GAMMA26) build_degamma(curve, MAX_HW_POINTS, coordinates_x, - tf == TRANSFER_FUNCTION_SRGB ? - gamma_type_index_2_4 : tf == TRANSFER_FUNCTION_GAMMA22 ? - gamma_type_index_2_2_flat : gamma_type_index_2_2); + tf); + else if (tf == TRANSFER_FUNCTION_HLG) + build_hlg_degamma(curve, + MAX_HW_POINTS, + coordinates_x, + 80, 1000); else if (tf == TRANSFER_FUNCTION_LINEAR) { // just copy coordinates_x into curve i = 0; @@ -1869,10 +1880,28 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, tf_pts->x_point_at_y1_green = 1; tf_pts->x_point_at_y1_blue = 1; - map_regamma_hw_to_x_user(ramp, coeff, rgb_user, - coordinates_x, axis_x, curve, - MAX_HW_POINTS, tf_pts, - mapUserRamp && ramp && ramp->type == GAMMA_RGB_256); + if (input_tf->tf == TRANSFER_FUNCTION_PQ) { + /* just copy current rgb_regamma into tf_pts */ + struct pwl_float_data_ex *curvePt = curve; + int i = 0; + + while (i <= MAX_HW_POINTS) { + tf_pts->red[i] = curvePt->r; + tf_pts->green[i] = curvePt->g; + tf_pts->blue[i] = curvePt->b; + ++curvePt; + ++i; + } + } else { + //clamps to 0-1 + map_regamma_hw_to_x_user(ramp, coeff, rgb_user, + coordinates_x, axis_x, curve, + MAX_HW_POINTS, tf_pts, + mapUserRamp && ramp && ramp->type == GAMMA_RGB_256); + } + + + if (ramp->type == GAMMA_CUSTOM) apply_lut_1d(ramp, MAX_HW_POINTS, tf_pts); @@ -1891,14 +1920,14 @@ rgb_user_alloc_fail: return ret; } - -bool mod_color_calculate_curve(enum dc_transfer_func_predefined trans, +static bool calculate_curve(enum dc_transfer_func_predefined trans, struct dc_transfer_func_distributed_points *points, + struct pwl_float_data_ex *rgb_regamma, + const struct freesync_hdr_tf_params *fs_params, uint32_t sdr_ref_white_level) { uint32_t i; bool ret = false; - struct pwl_float_data_ex *rgb_regamma = NULL; if (trans == TRANSFER_FUNCTION_UNITY || trans == TRANSFER_FUNCTION_LINEAR) { @@ -1908,42 +1937,50 @@ bool mod_color_calculate_curve(enum dc_transfer_func_predefined trans, points->x_point_at_y1_blue = 1; for (i = 0; i <= MAX_HW_POINTS ; i++) { - points->red[i] = coordinates_x[i].x; - points->green[i] = coordinates_x[i].x; - points->blue[i] = coordinates_x[i].x; + rgb_regamma[i].r = coordinates_x[i].x; + rgb_regamma[i].g = coordinates_x[i].x; + rgb_regamma[i].b = coordinates_x[i].x; } + ret = true; } else if (trans == TRANSFER_FUNCTION_PQ) { - rgb_regamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, - sizeof(*rgb_regamma), - GFP_KERNEL); - if (!rgb_regamma) - goto rgb_regamma_alloc_fail; points->end_exponent = 7; points->x_point_at_y1_red = 125; points->x_point_at_y1_green = 125; points->x_point_at_y1_blue = 125; - build_pq(rgb_regamma, MAX_HW_POINTS, coordinates_x, sdr_ref_white_level); - for (i = 0; i <= MAX_HW_POINTS ; i++) { - points->red[i] = rgb_regamma[i].r; - points->green[i] = rgb_regamma[i].g; - points->blue[i] = rgb_regamma[i].b; - } + ret = true; + } else if (trans == TRANSFER_FUNCTION_GAMMA22 && + fs_params != NULL && fs_params->skip_tm == 0) { + build_freesync_hdr(rgb_regamma, + MAX_HW_POINTS, + coordinates_x, + fs_params); - kvfree(rgb_regamma); - } else if (trans == TRANSFER_FUNCTION_SRGB || - trans == TRANSFER_FUNCTION_BT709) { - rgb_regamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, - sizeof(*rgb_regamma), - GFP_KERNEL); - if (!rgb_regamma) - goto rgb_regamma_alloc_fail; + ret = true; + } else if (trans == TRANSFER_FUNCTION_HLG) { + points->end_exponent = 4; + points->x_point_at_y1_red = 12; + points->x_point_at_y1_green = 12; + points->x_point_at_y1_blue = 12; + + build_hlg_regamma(rgb_regamma, + MAX_HW_POINTS, + coordinates_x, + 80, 1000); + + ret = true; + } else { + // trans == TRANSFER_FUNCTION_SRGB + // trans == TRANSFER_FUNCTION_BT709 + // trans == TRANSFER_FUNCTION_GAMMA22 + // trans == TRANSFER_FUNCTION_GAMMA24 + // trans == TRANSFER_FUNCTION_GAMMA26 points->end_exponent = 0; points->x_point_at_y1_red = 1; points->x_point_at_y1_green = 1; @@ -1952,42 +1989,112 @@ bool mod_color_calculate_curve(enum dc_transfer_func_predefined trans, build_regamma(rgb_regamma, MAX_HW_POINTS, coordinates_x, - trans == TRANSFER_FUNCTION_SRGB ? - gamma_type_index_2_4 : trans == TRANSFER_FUNCTION_GAMMA22 ? - gamma_type_index_2_2_flat : gamma_type_index_2_2); - for (i = 0; i <= MAX_HW_POINTS ; i++) { - points->red[i] = rgb_regamma[i].r; - points->green[i] = rgb_regamma[i].g; - points->blue[i] = rgb_regamma[i].b; - } + trans); + ret = true; + } - kvfree(rgb_regamma); - } else if (trans == TRANSFER_FUNCTION_HLG || - trans == TRANSFER_FUNCTION_HLG12) { - rgb_regamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, - sizeof(*rgb_regamma), - GFP_KERNEL); - if (!rgb_regamma) - goto rgb_regamma_alloc_fail; + return ret; +} - build_hlg_regamma(rgb_regamma, - MAX_HW_POINTS, - coordinates_x, - trans == TRANSFER_FUNCTION_HLG12 ? true:false); - for (i = 0; i <= MAX_HW_POINTS ; i++) { - points->red[i] = rgb_regamma[i].r; - points->green[i] = rgb_regamma[i].g; - points->blue[i] = rgb_regamma[i].b; - } - ret = true; - kvfree(rgb_regamma); +bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf, + const struct dc_gamma *ramp, bool mapUserRamp, bool canRomBeUsed, + const struct freesync_hdr_tf_params *fs_params) +{ + struct dc_transfer_func_distributed_points *tf_pts = &output_tf->tf_pts; + struct dividers dividers; + + struct pwl_float_data *rgb_user = NULL; + struct pwl_float_data_ex *rgb_regamma = NULL; + struct gamma_pixel *axis_x = NULL; + struct pixel_gamma_point *coeff = NULL; + enum dc_transfer_func_predefined tf = TRANSFER_FUNCTION_SRGB; + bool ret = false; + + if (output_tf->type == TF_TYPE_BYPASS) + return false; + + /* we can use hardcoded curve for plain SRGB TF */ + if (output_tf->type == TF_TYPE_PREDEFINED && canRomBeUsed == true && + output_tf->tf == TRANSFER_FUNCTION_SRGB) { + if (ramp == NULL) + return true; + if ((ramp->is_identity && ramp->type != GAMMA_CS_TFM_1D) || + (!mapUserRamp && ramp->type == GAMMA_RGB_256)) + return true; + } + + output_tf->type = TF_TYPE_DISTRIBUTED_POINTS; + + if (ramp && ramp->type != GAMMA_CS_TFM_1D && + (mapUserRamp || ramp->type != GAMMA_RGB_256)) { + rgb_user = kvcalloc(ramp->num_entries + _EXTRA_POINTS, + sizeof(*rgb_user), + GFP_KERNEL); + if (!rgb_user) + goto rgb_user_alloc_fail; + + axis_x = kvcalloc(ramp->num_entries + 3, sizeof(*axis_x), + GFP_KERNEL); + if (!axis_x) + goto axis_x_alloc_fail; + + dividers.divider1 = dc_fixpt_from_fraction(3, 2); + dividers.divider2 = dc_fixpt_from_int(2); + dividers.divider3 = dc_fixpt_from_fraction(5, 2); + + build_evenly_distributed_points( + axis_x, + ramp->num_entries, + dividers); + + if (ramp->type == GAMMA_RGB_256 && mapUserRamp) + scale_gamma(rgb_user, ramp, dividers); + else if (ramp->type == GAMMA_RGB_FLOAT_1024) + scale_gamma_dx(rgb_user, ramp, dividers); + } + + rgb_regamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, + sizeof(*rgb_regamma), + GFP_KERNEL); + if (!rgb_regamma) + goto rgb_regamma_alloc_fail; + + coeff = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*coeff), + GFP_KERNEL); + if (!coeff) + goto coeff_alloc_fail; + + tf = output_tf->tf; + + ret = calculate_curve(tf, + tf_pts, + rgb_regamma, + fs_params, + output_tf->sdr_ref_white_level); + + if (ret) { + map_regamma_hw_to_x_user(ramp, coeff, rgb_user, + coordinates_x, axis_x, rgb_regamma, + MAX_HW_POINTS, tf_pts, + (mapUserRamp || (ramp && ramp->type != GAMMA_RGB_256)) && + (ramp && ramp->type != GAMMA_CS_TFM_1D)); + + if (ramp && ramp->type == GAMMA_CS_TFM_1D) + apply_lut_1d(ramp, MAX_HW_POINTS, tf_pts); } + + kvfree(coeff); +coeff_alloc_fail: + kvfree(rgb_regamma); rgb_regamma_alloc_fail: + kvfree(axis_x); +axis_x_alloc_fail: + kvfree(rgb_user); +rgb_user_alloc_fail: return ret; } - bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, struct dc_transfer_func_distributed_points *points) { @@ -2024,8 +2131,10 @@ bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, kvfree(rgb_degamma); } else if (trans == TRANSFER_FUNCTION_SRGB || - trans == TRANSFER_FUNCTION_BT709 || - trans == TRANSFER_FUNCTION_GAMMA22) { + trans == TRANSFER_FUNCTION_BT709 || + trans == TRANSFER_FUNCTION_GAMMA22 || + trans == TRANSFER_FUNCTION_GAMMA24 || + trans == TRANSFER_FUNCTION_GAMMA26) { rgb_degamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*rgb_degamma), GFP_KERNEL); @@ -2035,9 +2144,7 @@ bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, build_degamma(rgb_degamma, MAX_HW_POINTS, coordinates_x, - trans == TRANSFER_FUNCTION_SRGB ? - gamma_type_index_2_4 : trans == TRANSFER_FUNCTION_GAMMA22 ? - gamma_type_index_2_2_flat : gamma_type_index_2_2); + trans); for (i = 0; i <= MAX_HW_POINTS ; i++) { points->red[i] = rgb_degamma[i].r; points->green[i] = rgb_degamma[i].g; @@ -2046,8 +2153,7 @@ bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, ret = true; kvfree(rgb_degamma); - } else if (trans == TRANSFER_FUNCTION_HLG || - trans == TRANSFER_FUNCTION_HLG12) { + } else if (trans == TRANSFER_FUNCTION_HLG) { rgb_degamma = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*rgb_degamma), GFP_KERNEL); @@ -2057,7 +2163,7 @@ bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, build_hlg_degamma(rgb_degamma, MAX_HW_POINTS, coordinates_x, - trans == TRANSFER_FUNCTION_HLG12 ? true:false); + 80, 1000); for (i = 0; i <= MAX_HW_POINTS ; i++) { points->red[i] = rgb_degamma[i].r; points->green[i] = rgb_degamma[i].g; @@ -2074,5 +2180,3 @@ bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, rgb_degamma_alloc_fail: return ret; } - - diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.h b/drivers/gpu/drm/amd/display/modules/color/color_gamma.h index 369953fafadf..9994817a9a03 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.h +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.h @@ -82,7 +82,17 @@ struct freesync_hdr_tf_params { unsigned int skip_tm; // skip tm }; +struct translate_from_linear_space_args { + struct fixed31_32 arg; + struct fixed31_32 a0; + struct fixed31_32 a1; + struct fixed31_32 a2; + struct fixed31_32 a3; + struct fixed31_32 gamma; +}; + void setup_x_points_distribution(void); +void log_x_points_distribution(struct dal_logger *logger); void precompute_pq(void); void precompute_de_pq(void); @@ -93,10 +103,6 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf, bool mod_color_calculate_degamma_params(struct dc_transfer_func *output_tf, const struct dc_gamma *ramp, bool mapUserRamp); -bool mod_color_calculate_curve(enum dc_transfer_func_predefined trans, - struct dc_transfer_func_distributed_points *points, - uint32_t sdr_ref_white_level); - bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, struct dc_transfer_func_distributed_points *points); diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index 7c20171a3b6d..b9992ebf77a6 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -37,8 +37,8 @@ #define STATIC_SCREEN_RAMP_DELTA_REFRESH_RATE_PER_FRAME ((1000 / 60) * 65) /* Number of elements in the render times cache array */ #define RENDER_TIMES_MAX_COUNT 10 -/* Threshold to exit BTR (to avoid frequent enter-exits at the lower limit) */ -#define BTR_EXIT_MARGIN 2000 +/* Threshold to exit/exit BTR (to avoid frequent enter-exits at the lower limit) */ +#define BTR_MAX_MARGIN 2500 /* Threshold to change BTR multiplier (to avoid frequent changes) */ #define BTR_DRIFT_MARGIN 2000 /*Threshold to exit fixed refresh rate*/ @@ -52,93 +52,6 @@ struct core_freesync { struct dc *dc; }; -void setFieldWithMask(unsigned char *dest, unsigned int mask, unsigned int value) -{ - unsigned int shift = 0; - - if (!mask || !dest) - return; - - while (!((mask >> shift) & 1)) - shift++; - - //reset - *dest = *dest & ~mask; - //set - //dont let value span past mask - value = value & (mask >> shift); - //insert value - *dest = *dest | (value << shift); -} - -// VTEM Byte Offset -#define VRR_VTEM_PB0 0 -#define VRR_VTEM_PB1 1 -#define VRR_VTEM_PB2 2 -#define VRR_VTEM_PB3 3 -#define VRR_VTEM_PB4 4 -#define VRR_VTEM_PB5 5 -#define VRR_VTEM_PB6 6 - -#define VRR_VTEM_MD0 7 -#define VRR_VTEM_MD1 8 -#define VRR_VTEM_MD2 9 -#define VRR_VTEM_MD3 10 - - -// VTEM Byte Masks -//PB0 -#define MASK__VRR_VTEM_PB0__RESERVED0 0x01 -#define MASK__VRR_VTEM_PB0__SYNC 0x02 -#define MASK__VRR_VTEM_PB0__VFR 0x04 -#define MASK__VRR_VTEM_PB0__AFR 0x08 -#define MASK__VRR_VTEM_PB0__DS_TYPE 0x30 - //0: Periodic pseudo-static EM Data Set - //1: Periodic dynamic EM Data Set - //2: Unique EM Data Set - //3: Reserved -#define MASK__VRR_VTEM_PB0__END 0x40 -#define MASK__VRR_VTEM_PB0__NEW 0x80 - -//PB1 -#define MASK__VRR_VTEM_PB1__RESERVED1 0xFF - -//PB2 -#define MASK__VRR_VTEM_PB2__ORGANIZATION_ID 0xFF - //0: This is a Vendor Specific EM Data Set - //1: This EM Data Set is defined by This Specification (HDMI 2.1 r102.clean) - //2: This EM Data Set is defined by CTA-861-G - //3: This EM Data Set is defined by VESA -//PB3 -#define MASK__VRR_VTEM_PB3__DATA_SET_TAG_MSB 0xFF -//PB4 -#define MASK__VRR_VTEM_PB4__DATA_SET_TAG_LSB 0xFF -//PB5 -#define MASK__VRR_VTEM_PB5__DATA_SET_LENGTH_MSB 0xFF -//PB6 -#define MASK__VRR_VTEM_PB6__DATA_SET_LENGTH_LSB 0xFF - - - -//PB7-27 (20 bytes): -//PB7 = MD0 -#define MASK__VRR_VTEM_MD0__VRR_EN 0x01 -#define MASK__VRR_VTEM_MD0__M_CONST 0x02 -#define MASK__VRR_VTEM_MD0__RESERVED2 0x0C -#define MASK__VRR_VTEM_MD0__FVA_FACTOR_M1 0xF0 - -//MD1 -#define MASK__VRR_VTEM_MD1__BASE_VFRONT 0xFF - -//MD2 -#define MASK__VRR_VTEM_MD2__BASE_REFRESH_RATE_98 0x03 -#define MASK__VRR_VTEM_MD2__RB 0x04 -#define MASK__VRR_VTEM_MD2__RESERVED3 0xF8 - -//MD3 -#define MASK__VRR_VTEM_MD3__BASE_REFRESH_RATE_07 0xFF - - #define MOD_FREESYNC_TO_CORE(mod_freesync)\ container_of(mod_freesync, struct core_freesync, public) @@ -209,7 +122,7 @@ static unsigned int calc_v_total_from_refresh( const struct dc_stream_state *stream, unsigned int refresh_in_uhz) { - unsigned int v_total = stream->timing.v_total; + unsigned int v_total; unsigned int frame_duration_in_ns; frame_duration_in_ns = @@ -321,6 +234,10 @@ static void update_v_total_for_static_ramp( current_duration_in_us) * (stream->timing.pix_clk_100hz / 10)), stream->timing.h_total), 1000); + /* v_total cannot be less than nominal */ + if (v_total < stream->timing.v_total) + v_total = stream->timing.v_total; + in_out_vrr->adjust.v_total_min = v_total; in_out_vrr->adjust.v_total_max = v_total; } @@ -337,24 +254,22 @@ static void apply_below_the_range(struct core_freesync *core_freesync, unsigned int delta_from_mid_point_in_us_1 = 0xFFFFFFFF; unsigned int delta_from_mid_point_in_us_2 = 0xFFFFFFFF; unsigned int frames_to_insert = 0; - unsigned int min_frame_duration_in_ns = 0; - unsigned int max_render_time_in_us = in_out_vrr->max_duration_in_us; unsigned int delta_from_mid_point_delta_in_us; - - min_frame_duration_in_ns = ((unsigned int) (div64_u64( - (1000000000ULL * 1000000), - in_out_vrr->max_refresh_in_uhz))); + unsigned int max_render_time_in_us = + in_out_vrr->max_duration_in_us - in_out_vrr->btr.margin_in_us; /* Program BTR */ - if (last_render_time_in_us + BTR_EXIT_MARGIN < max_render_time_in_us) { + if ((last_render_time_in_us + in_out_vrr->btr.margin_in_us / 2) < max_render_time_in_us) { /* Exit Below the Range */ if (in_out_vrr->btr.btr_active) { in_out_vrr->btr.frame_counter = 0; in_out_vrr->btr.btr_active = false; } - } else if (last_render_time_in_us > max_render_time_in_us) { + } else if (last_render_time_in_us > (max_render_time_in_us + in_out_vrr->btr.margin_in_us / 2)) { /* Enter Below the Range */ - in_out_vrr->btr.btr_active = true; + if (!in_out_vrr->btr.btr_active) { + in_out_vrr->btr.btr_active = true; + } } /* BTR set to "not active" so disengage */ @@ -410,7 +325,9 @@ static void apply_below_the_range(struct core_freesync *core_freesync, /* Choose number of frames to insert based on how close it * can get to the mid point of the variable range. */ - if (delta_from_mid_point_in_us_1 < delta_from_mid_point_in_us_2) { + if ((frame_time_in_us / mid_point_frames_ceil) > in_out_vrr->min_duration_in_us && + (delta_from_mid_point_in_us_1 < delta_from_mid_point_in_us_2 || + mid_point_frames_floor < 2)) { frames_to_insert = mid_point_frames_ceil; delta_from_mid_point_delta_in_us = delta_from_mid_point_in_us_2 - delta_from_mid_point_in_us_1; @@ -426,7 +343,7 @@ static void apply_below_the_range(struct core_freesync *core_freesync, if (in_out_vrr->btr.frames_to_insert != 0 && delta_from_mid_point_delta_in_us < BTR_DRIFT_MARGIN) { if (((last_render_time_in_us / in_out_vrr->btr.frames_to_insert) < - in_out_vrr->max_duration_in_us) && + max_render_time_in_us) && ((last_render_time_in_us / in_out_vrr->btr.frames_to_insert) > in_out_vrr->min_duration_in_us)) frames_to_insert = in_out_vrr->btr.frames_to_insert; @@ -435,6 +352,12 @@ static void apply_below_the_range(struct core_freesync *core_freesync, /* Either we've calculated the number of frames to insert, * or we need to insert min duration frames */ + if (last_render_time_in_us / frames_to_insert < + in_out_vrr->min_duration_in_us){ + frames_to_insert -= (frames_to_insert > 1) ? + 1 : 0; + } + if (frames_to_insert > 0) inserted_frame_duration_in_us = last_render_time_in_us / frames_to_insert; @@ -458,7 +381,7 @@ static void apply_fixed_refresh(struct core_freesync *core_freesync, bool update = false; unsigned int max_render_time_in_us = in_out_vrr->max_duration_in_us; - //Compute the exit refresh rate and exit frame duration + /* Compute the exit refresh rate and exit frame duration */ unsigned int exit_refresh_rate_in_milli_hz = ((1000000000/max_render_time_in_us) + (1000*FIXED_REFRESH_EXIT_MARGIN_IN_HZ)); unsigned int exit_frame_duration_in_us = 1000000000/exit_refresh_rate_in_milli_hz; @@ -568,22 +491,64 @@ bool mod_freesync_get_v_position(struct mod_freesync *mod_freesync, return false; } -static void build_vrr_infopacket_header_vtem(enum signal_type signal, +static void build_vrr_infopacket_data(const struct mod_vrr_params *vrr, + struct dc_info_packet *infopacket) +{ + /* PB1 = 0x1A (24bit AMD IEEE OUI (0x00001A) - Byte 0) */ + infopacket->sb[1] = 0x1A; + + /* PB2 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 1) */ + infopacket->sb[2] = 0x00; + + /* PB3 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 2) */ + infopacket->sb[3] = 0x00; + + /* PB4 = Reserved */ + + /* PB5 = Reserved */ + + /* PB6 = [Bits 7:3 = Reserved] */ + + /* PB6 = [Bit 0 = FreeSync Supported] */ + if (vrr->state != VRR_STATE_UNSUPPORTED) + infopacket->sb[6] |= 0x01; + + /* PB6 = [Bit 1 = FreeSync Enabled] */ + if (vrr->state != VRR_STATE_DISABLED && + vrr->state != VRR_STATE_UNSUPPORTED) + infopacket->sb[6] |= 0x02; + + /* PB6 = [Bit 2 = FreeSync Active] */ + if (vrr->state == VRR_STATE_ACTIVE_VARIABLE || + vrr->state == VRR_STATE_ACTIVE_FIXED) + infopacket->sb[6] |= 0x04; + + /* PB7 = FreeSync Minimum refresh rate (Hz) */ + infopacket->sb[7] = (unsigned char)(vrr->min_refresh_in_uhz / 1000000); + + /* PB8 = FreeSync Maximum refresh rate (Hz) + * Note: We should never go above the field rate of the mode timing set. + */ + infopacket->sb[8] = (unsigned char)(vrr->max_refresh_in_uhz / 1000000); + + + //FreeSync HDR + infopacket->sb[9] = 0; + infopacket->sb[10] = 0; +} + +static void build_vrr_infopacket_fs2_data(enum color_transfer_func app_tf, struct dc_info_packet *infopacket) { - // HEADER - - // HB0, HB1, HB2 indicates PacketType VTEMPacket - infopacket->hb0 = 0x7F; - infopacket->hb1 = 0xC0; - infopacket->hb2 = 0x00; //sequence_index - - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB0], MASK__VRR_VTEM_PB0__VFR, 1); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB2], MASK__VRR_VTEM_PB2__ORGANIZATION_ID, 1); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB3], MASK__VRR_VTEM_PB3__DATA_SET_TAG_MSB, 0); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB4], MASK__VRR_VTEM_PB4__DATA_SET_TAG_LSB, 1); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB5], MASK__VRR_VTEM_PB5__DATA_SET_LENGTH_MSB, 0); - setFieldWithMask(&infopacket->sb[VRR_VTEM_PB6], MASK__VRR_VTEM_PB6__DATA_SET_LENGTH_LSB, 4); + if (app_tf != TRANSFER_FUNC_UNKNOWN) { + infopacket->valid = true; + + infopacket->sb[6] |= 0x08; // PB6 = [Bit 3 = Native Color Active] + + if (app_tf == TRANSFER_FUNC_GAMMA_22) { + infopacket->sb[9] |= 0x04; // PB6 = [Bit 2 = Gamma 2.2 EOTF Active] + } + } } static void build_vrr_infopacket_header_v1(enum signal_type signal, @@ -684,105 +649,6 @@ static void build_vrr_infopacket_header_v2(enum signal_type signal, } } -static void build_vrr_vtem_infopacket_data(const struct dc_stream_state *stream, - const struct mod_vrr_params *vrr, - struct dc_info_packet *infopacket) -{ - unsigned int fieldRateInHz; - - if (vrr->state == VRR_STATE_ACTIVE_VARIABLE || - vrr->state == VRR_STATE_ACTIVE_FIXED) { - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD0], MASK__VRR_VTEM_MD0__VRR_EN, 1); - } else { - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD0], MASK__VRR_VTEM_MD0__VRR_EN, 0); - } - - if (!stream->timing.vic) { - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD1], MASK__VRR_VTEM_MD1__BASE_VFRONT, - stream->timing.v_front_porch); - - - /* TODO: In dal2, we check mode flags for a reduced blanking timing. - * Need a way to relay that information to this function. - * if("ReducedBlanking") - * { - * setFieldWithMask(&infopacket->sb[VRR_VTEM_MD2], MASK__VRR_VTEM_MD2__RB, 1; - * } - */ - - //TODO: DAL2 does FixPoint and rounding. Here we might need to account for that - fieldRateInHz = (stream->timing.pix_clk_100hz * 100)/ - (stream->timing.h_total * stream->timing.v_total); - - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD2], MASK__VRR_VTEM_MD2__BASE_REFRESH_RATE_98, - fieldRateInHz >> 8); - setFieldWithMask(&infopacket->sb[VRR_VTEM_MD3], MASK__VRR_VTEM_MD3__BASE_REFRESH_RATE_07, - fieldRateInHz); - - } - infopacket->valid = true; -} - -static void build_vrr_infopacket_data(const struct mod_vrr_params *vrr, - struct dc_info_packet *infopacket) -{ - /* PB1 = 0x1A (24bit AMD IEEE OUI (0x00001A) - Byte 0) */ - infopacket->sb[1] = 0x1A; - - /* PB2 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 1) */ - infopacket->sb[2] = 0x00; - - /* PB3 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 2) */ - infopacket->sb[3] = 0x00; - - /* PB4 = Reserved */ - - /* PB5 = Reserved */ - - /* PB6 = [Bits 7:3 = Reserved] */ - - /* PB6 = [Bit 0 = FreeSync Supported] */ - if (vrr->state != VRR_STATE_UNSUPPORTED) - infopacket->sb[6] |= 0x01; - - /* PB6 = [Bit 1 = FreeSync Enabled] */ - if (vrr->state != VRR_STATE_DISABLED && - vrr->state != VRR_STATE_UNSUPPORTED) - infopacket->sb[6] |= 0x02; - - /* PB6 = [Bit 2 = FreeSync Active] */ - if (vrr->state == VRR_STATE_ACTIVE_VARIABLE || - vrr->state == VRR_STATE_ACTIVE_FIXED) - infopacket->sb[6] |= 0x04; - - /* PB7 = FreeSync Minimum refresh rate (Hz) */ - infopacket->sb[7] = (unsigned char)(vrr->min_refresh_in_uhz / 1000000); - - /* PB8 = FreeSync Maximum refresh rate (Hz) - * Note: We should never go above the field rate of the mode timing set. - */ - infopacket->sb[8] = (unsigned char)(vrr->max_refresh_in_uhz / 1000000); - - - //FreeSync HDR - infopacket->sb[9] = 0; - infopacket->sb[10] = 0; -} - -static void build_vrr_infopacket_fs2_data(enum color_transfer_func app_tf, - struct dc_info_packet *infopacket) -{ - if (app_tf != TRANSFER_FUNC_UNKNOWN) { - infopacket->valid = true; - - infopacket->sb[6] |= 0x08; // PB6 = [Bit 3 = Native Color Active] - - if (app_tf == TRANSFER_FUNC_GAMMA_22) { - infopacket->sb[9] |= 0x04; // PB6 = [Bit 2 = Gamma 2.2 EOTF Active] - } - } -} - static void build_vrr_infopacket_checksum(unsigned int *payload_size, struct dc_info_packet *infopacket) { @@ -835,21 +701,6 @@ static void build_vrr_infopacket_v2(enum signal_type signal, infopacket->valid = true; } -static void build_vrr_infopacket_vtem(const struct dc_stream_state *stream, - const struct mod_vrr_params *vrr, - struct dc_info_packet *infopacket) -{ - //VTEM info packet for HdmiVrr - - memset(infopacket, 0, sizeof(struct dc_info_packet)); - - //VTEM Packet is structured differently - build_vrr_infopacket_header_vtem(stream->signal, infopacket); - build_vrr_vtem_infopacket_data(stream, vrr, infopacket); - - infopacket->valid = true; -} - void mod_freesync_build_vrr_infopacket(struct mod_freesync *mod_freesync, const struct dc_stream_state *stream, const struct mod_vrr_params *vrr, @@ -862,16 +713,13 @@ void mod_freesync_build_vrr_infopacket(struct mod_freesync *mod_freesync, * Check if Freesync is supported. Return if false. If true, * set the corresponding bit in the info packet */ - if (!vrr->supported || (!vrr->send_info_frame && packet_type != PACKET_TYPE_VTEM)) + if (!vrr->supported || (!vrr->send_info_frame)) return; switch (packet_type) { case PACKET_TYPE_FS2: build_vrr_infopacket_v2(stream->signal, vrr, app_tf, infopacket); break; - case PACKET_TYPE_VTEM: - build_vrr_infopacket_vtem(stream, vrr, infopacket); - break; case PACKET_TYPE_VRR: case PACKET_TYPE_FS1: default: @@ -887,8 +735,8 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, struct core_freesync *core_freesync = NULL; unsigned long long nominal_field_rate_in_uhz = 0; unsigned int refresh_range = 0; - unsigned int min_refresh_in_uhz = 0; - unsigned int max_refresh_in_uhz = 0; + unsigned long long min_refresh_in_uhz = 0; + unsigned long long max_refresh_in_uhz = 0; if (mod_freesync == NULL) return; @@ -899,6 +747,10 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, nominal_field_rate_in_uhz = mod_freesync_calc_nominal_field_rate(stream); + /* Rounded to the nearest Hz */ + nominal_field_rate_in_uhz = 1000000ULL * + div_u64(nominal_field_rate_in_uhz + 500000, 1000000); + min_refresh_in_uhz = in_config->min_refresh_in_uhz; max_refresh_in_uhz = in_config->max_refresh_in_uhz; @@ -915,7 +767,7 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, min_refresh_in_uhz = nominal_field_rate_in_uhz; if (!vrr_settings_require_update(core_freesync, - in_config, min_refresh_in_uhz, max_refresh_in_uhz, + in_config, (unsigned int)min_refresh_in_uhz, (unsigned int)max_refresh_in_uhz, in_out_vrr)) return; @@ -931,36 +783,45 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, return; } else { - in_out_vrr->min_refresh_in_uhz = min_refresh_in_uhz; + in_out_vrr->min_refresh_in_uhz = (unsigned int)min_refresh_in_uhz; in_out_vrr->max_duration_in_us = calc_duration_in_us_from_refresh_in_uhz( - min_refresh_in_uhz); + (unsigned int)min_refresh_in_uhz); - in_out_vrr->max_refresh_in_uhz = max_refresh_in_uhz; + in_out_vrr->max_refresh_in_uhz = (unsigned int)max_refresh_in_uhz; in_out_vrr->min_duration_in_us = calc_duration_in_us_from_refresh_in_uhz( - max_refresh_in_uhz); + (unsigned int)max_refresh_in_uhz); refresh_range = in_out_vrr->max_refresh_in_uhz - in_out_vrr->min_refresh_in_uhz; + in_out_vrr->btr.margin_in_us = in_out_vrr->max_duration_in_us - + 2 * in_out_vrr->min_duration_in_us; + if (in_out_vrr->btr.margin_in_us > BTR_MAX_MARGIN) + in_out_vrr->btr.margin_in_us = BTR_MAX_MARGIN; + in_out_vrr->supported = true; } in_out_vrr->fixed.ramping_active = in_config->ramping; in_out_vrr->btr.btr_enabled = in_config->btr; + if (in_out_vrr->max_refresh_in_uhz < 2 * in_out_vrr->min_refresh_in_uhz) in_out_vrr->btr.btr_enabled = false; + in_out_vrr->btr.btr_active = false; in_out_vrr->btr.inserted_duration_in_us = 0; in_out_vrr->btr.frames_to_insert = 0; in_out_vrr->btr.frame_counter = 0; + in_out_vrr->fixed.fixed_active = false; + in_out_vrr->fixed.target_refresh_in_uhz = 0; + in_out_vrr->btr.mid_point_in_us = - in_out_vrr->min_duration_in_us + - (in_out_vrr->max_duration_in_us - - in_out_vrr->min_duration_in_us) / 2; + (in_out_vrr->min_duration_in_us + + in_out_vrr->max_duration_in_us) / 2; if (in_out_vrr->state == VRR_STATE_UNSUPPORTED) { in_out_vrr->adjust.v_total_min = stream->timing.v_total; @@ -973,6 +834,7 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, in_out_vrr->adjust.v_total_max = stream->timing.v_total; } else if (in_out_vrr->state == VRR_STATE_ACTIVE_VARIABLE && refresh_range >= MIN_REFRESH_RANGE_IN_US) { + in_out_vrr->adjust.v_total_min = calc_v_total_from_refresh(stream, in_out_vrr->max_refresh_in_uhz); @@ -1130,13 +992,9 @@ void mod_freesync_get_settings(struct mod_freesync *mod_freesync, unsigned int *inserted_frames, unsigned int *inserted_duration_in_us) { - struct core_freesync *core_freesync = NULL; - if (mod_freesync == NULL) return; - core_freesync = MOD_FREESYNC_TO_CORE(mod_freesync); - if (vrr->supported) { *v_total_min = vrr->adjust.v_total_min; *v_total_max = vrr->adjust.v_total_max; @@ -1151,14 +1009,13 @@ unsigned long long mod_freesync_calc_nominal_field_rate( const struct dc_stream_state *stream) { unsigned long long nominal_field_rate_in_uhz = 0; + unsigned int total = stream->timing.h_total * stream->timing.v_total; - /* Calculate nominal field rate for stream */ + /* Calculate nominal field rate for stream, rounded up to nearest integer */ nominal_field_rate_in_uhz = stream->timing.pix_clk_100hz / 10; nominal_field_rate_in_uhz *= 1000ULL * 1000ULL * 1000ULL; - nominal_field_rate_in_uhz = div_u64(nominal_field_rate_in_uhz, - stream->timing.h_total); - nominal_field_rate_in_uhz = div_u64(nominal_field_rate_in_uhz, - stream->timing.v_total); + + nominal_field_rate_in_uhz = div_u64(nominal_field_rate_in_uhz, total); return nominal_field_rate_in_uhz; } diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/Makefile b/drivers/gpu/drm/amd/display/modules/hdcp/Makefile new file mode 100644 index 000000000000..904424da01b5 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/Makefile @@ -0,0 +1,33 @@ +# +# Copyright 2019 Advanced Micro Devices, Inc. +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the "Software"), +# to deal in the Software without restriction, including without limitation +# the rights to use, copy, modify, merge, publish, distribute, sublicense, +# and/or sell copies of the Software, and to permit persons to whom the +# Software is furnished to do so, subject to the following conditions: +# +# The above copyright notice and this permission notice shall be included in +# all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR +# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +# OTHER DEALINGS IN THE SOFTWARE. +# +# +# Makefile for the 'hdcp' sub-module of DAL. +# + +HDCP = hdcp_ddc.o hdcp_log.o hdcp_psp.o hdcp.o \ + hdcp1_execution.o hdcp1_transition.o \ + hdcp2_execution.o hdcp2_transition.o + +AMD_DAL_HDCP = $(addprefix $(AMDDALPATH)/modules/hdcp/,$(HDCP)) +#$(info ************ DAL-HDCP_MAKEFILE ************) + +AMD_DISPLAY_FILES += $(AMD_DAL_HDCP) diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c new file mode 100644 index 000000000000..8aa528e874c4 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c @@ -0,0 +1,509 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "hdcp.h" + +static void push_error_status(struct mod_hdcp *hdcp, + enum mod_hdcp_status status) +{ + struct mod_hdcp_trace *trace = &hdcp->connection.trace; + + if (trace->error_count < MAX_NUM_OF_ERROR_TRACE) { + trace->errors[trace->error_count].status = status; + trace->errors[trace->error_count].state_id = hdcp->state.id; + trace->error_count++; + HDCP_ERROR_TRACE(hdcp, status); + } + + if (is_hdcp1(hdcp)) { + hdcp->connection.hdcp1_retry_count++; + } else if (is_hdcp2(hdcp)) { + hdcp->connection.hdcp2_retry_count++; + } +} + +static uint8_t is_cp_desired_hdcp1(struct mod_hdcp *hdcp) +{ + int i, is_auth_needed = 0; + + /* if all displays on the link don't need authentication, + * hdcp is not desired + */ + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { + if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && + !hdcp->connection.displays[i].adjust.disable) { + is_auth_needed = 1; + break; + } + } + + return (hdcp->connection.hdcp1_retry_count < MAX_NUM_OF_ATTEMPTS) && + is_auth_needed && + !hdcp->connection.link.adjust.hdcp1.disable; +} + +static uint8_t is_cp_desired_hdcp2(struct mod_hdcp *hdcp) +{ + int i, is_auth_needed = 0; + + /* if all displays on the link don't need authentication, + * hdcp is not desired + */ + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { + if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_INACTIVE && + !hdcp->connection.displays[i].adjust.disable) { + is_auth_needed = 1; + break; + } + } + + return (hdcp->connection.hdcp2_retry_count < MAX_NUM_OF_ATTEMPTS) && + is_auth_needed && + !hdcp->connection.link.adjust.hdcp2.disable && + !hdcp->connection.is_hdcp2_revoked; +} + +static enum mod_hdcp_status execution(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + union mod_hdcp_transition_input *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (is_in_initialized_state(hdcp)) { + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + /* initialize transition input */ + memset(input, 0, sizeof(union mod_hdcp_transition_input)); + } else if (is_in_cp_not_desired_state(hdcp)) { + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + /* update topology event if hdcp is not desired */ + status = mod_hdcp_add_display_topology(hdcp); + } else if (is_in_hdcp1_states(hdcp)) { + status = mod_hdcp_hdcp1_execution(hdcp, event_ctx, &input->hdcp1); + } else if (is_in_hdcp1_dp_states(hdcp)) { + status = mod_hdcp_hdcp1_dp_execution(hdcp, + event_ctx, &input->hdcp1); + } else if (is_in_hdcp2_states(hdcp)) { + status = mod_hdcp_hdcp2_execution(hdcp, event_ctx, &input->hdcp2); + } else if (is_in_hdcp2_dp_states(hdcp)) { + status = mod_hdcp_hdcp2_dp_execution(hdcp, + event_ctx, &input->hdcp2); + } +out: + return status; +} + +static enum mod_hdcp_status transition(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + union mod_hdcp_transition_input *input, + struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->unexpected_event) + goto out; + + if (is_in_initialized_state(hdcp)) { + if (is_dp_hdcp(hdcp)) + if (is_cp_desired_hdcp2(hdcp)) { + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A0_DETERMINE_RX_HDCP_CAPABLE); + } else if (is_cp_desired_hdcp1(hdcp)) { + callback_in_ms(0, output); + set_state_id(hdcp, output, D1_A0_DETERMINE_RX_HDCP_CAPABLE); + } else { + callback_in_ms(0, output); + set_state_id(hdcp, output, HDCP_CP_NOT_DESIRED); + } + else if (is_hdmi_dvi_sl_hdcp(hdcp)) + if (is_cp_desired_hdcp2(hdcp)) { + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A0_KNOWN_HDCP2_CAPABLE_RX); + } else if (is_cp_desired_hdcp1(hdcp)) { + callback_in_ms(0, output); + set_state_id(hdcp, output, H1_A0_WAIT_FOR_ACTIVE_RX); + } else { + callback_in_ms(0, output); + set_state_id(hdcp, output, HDCP_CP_NOT_DESIRED); + } + else { + callback_in_ms(0, output); + set_state_id(hdcp, output, HDCP_CP_NOT_DESIRED); + } + } else if (is_in_cp_not_desired_state(hdcp)) { + increment_stay_counter(hdcp); + } else if (is_in_hdcp1_states(hdcp)) { + status = mod_hdcp_hdcp1_transition(hdcp, + event_ctx, &input->hdcp1, output); + } else if (is_in_hdcp1_dp_states(hdcp)) { + status = mod_hdcp_hdcp1_dp_transition(hdcp, + event_ctx, &input->hdcp1, output); + } else if (is_in_hdcp2_states(hdcp)) { + status = mod_hdcp_hdcp2_transition(hdcp, + event_ctx, &input->hdcp2, output); + } else if (is_in_hdcp2_dp_states(hdcp)) { + status = mod_hdcp_hdcp2_dp_transition(hdcp, + event_ctx, &input->hdcp2, output); + } else { + status = MOD_HDCP_STATUS_INVALID_STATE; + } +out: + return status; +} + +static enum mod_hdcp_status reset_authentication(struct mod_hdcp *hdcp, + struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (is_hdcp1(hdcp)) { + if (hdcp->auth.trans_input.hdcp1.create_session != UNKNOWN) { + /* TODO - update psp to unify create session failure + * recovery between hdcp1 and 2. + */ + mod_hdcp_hdcp1_destroy_session(hdcp); + + } + if (hdcp->auth.trans_input.hdcp1.add_topology == PASS) { + status = mod_hdcp_remove_display_topology(hdcp); + if (status != MOD_HDCP_STATUS_SUCCESS) { + output->callback_needed = 0; + output->watchdog_timer_needed = 0; + goto out; + } + } + HDCP_TOP_RESET_AUTH_TRACE(hdcp); + memset(&hdcp->auth, 0, sizeof(struct mod_hdcp_authentication)); + memset(&hdcp->state, 0, sizeof(struct mod_hdcp_state)); + set_state_id(hdcp, output, HDCP_INITIALIZED); + } else if (is_hdcp2(hdcp)) { + if (hdcp->auth.trans_input.hdcp2.create_session == PASS) { + status = mod_hdcp_hdcp2_destroy_session(hdcp); + if (status != MOD_HDCP_STATUS_SUCCESS) { + output->callback_needed = 0; + output->watchdog_timer_needed = 0; + goto out; + } + } + if (hdcp->auth.trans_input.hdcp2.add_topology == PASS) { + status = mod_hdcp_remove_display_topology(hdcp); + if (status != MOD_HDCP_STATUS_SUCCESS) { + output->callback_needed = 0; + output->watchdog_timer_needed = 0; + goto out; + } + } + HDCP_TOP_RESET_AUTH_TRACE(hdcp); + memset(&hdcp->auth, 0, sizeof(struct mod_hdcp_authentication)); + memset(&hdcp->state, 0, sizeof(struct mod_hdcp_state)); + set_state_id(hdcp, output, HDCP_INITIALIZED); + } else if (is_in_cp_not_desired_state(hdcp)) { + status = mod_hdcp_remove_display_topology(hdcp); + if (status != MOD_HDCP_STATUS_SUCCESS) { + output->callback_needed = 0; + output->watchdog_timer_needed = 0; + goto out; + } + HDCP_TOP_RESET_AUTH_TRACE(hdcp); + memset(&hdcp->auth, 0, sizeof(struct mod_hdcp_authentication)); + memset(&hdcp->state, 0, sizeof(struct mod_hdcp_state)); + set_state_id(hdcp, output, HDCP_INITIALIZED); + } + +out: + /* stop callback and watchdog requests from previous authentication*/ + output->watchdog_timer_stop = 1; + output->callback_stop = 1; + return status; +} + +static enum mod_hdcp_status reset_connection(struct mod_hdcp *hdcp, + struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + memset(output, 0, sizeof(struct mod_hdcp_output)); + + status = reset_authentication(hdcp, output); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + + if (current_state(hdcp) != HDCP_UNINITIALIZED) { + HDCP_TOP_RESET_CONN_TRACE(hdcp); + set_state_id(hdcp, output, HDCP_UNINITIALIZED); + } + memset(&hdcp->connection, 0, sizeof(hdcp->connection)); +out: + return status; +} + +/* + * Implementation of functions in mod_hdcp.h + */ +size_t mod_hdcp_get_memory_size(void) +{ + return sizeof(struct mod_hdcp); +} + +enum mod_hdcp_status mod_hdcp_setup(struct mod_hdcp *hdcp, + struct mod_hdcp_config *config) +{ + struct mod_hdcp_output output; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + memset(hdcp, 0, sizeof(struct mod_hdcp)); + memset(&output, 0, sizeof(output)); + hdcp->config = *config; + HDCP_TOP_INTERFACE_TRACE(hdcp); + status = reset_connection(hdcp, &output); + if (status != MOD_HDCP_STATUS_SUCCESS) + push_error_status(hdcp, status); + return status; +} + +enum mod_hdcp_status mod_hdcp_teardown(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + struct mod_hdcp_output output; + + HDCP_TOP_INTERFACE_TRACE(hdcp); + memset(&output, 0, sizeof(output)); + status = reset_connection(hdcp, &output); + if (status == MOD_HDCP_STATUS_SUCCESS) + memset(hdcp, 0, sizeof(struct mod_hdcp)); + else + push_error_status(hdcp, status); + return status; +} + +enum mod_hdcp_status mod_hdcp_add_display(struct mod_hdcp *hdcp, + struct mod_hdcp_link *link, struct mod_hdcp_display *display, + struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + struct mod_hdcp_display *display_container = NULL; + + HDCP_TOP_INTERFACE_TRACE_WITH_INDEX(hdcp, display->index); + memset(output, 0, sizeof(struct mod_hdcp_output)); + + /* skip inactive display */ + if (display->state != MOD_HDCP_DISPLAY_ACTIVE) { + status = MOD_HDCP_STATUS_SUCCESS; + goto out; + } + + /* check existing display container */ + if (get_active_display_at_index(hdcp, display->index)) { + status = MOD_HDCP_STATUS_SUCCESS; + goto out; + } + + /* find an empty display container */ + display_container = get_empty_display_container(hdcp); + if (!display_container) { + status = MOD_HDCP_STATUS_DISPLAY_OUT_OF_BOUND; + goto out; + } + + /* reset existing authentication status */ + status = reset_authentication(hdcp, output); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + + /* add display to connection */ + hdcp->connection.link = *link; + *display_container = *display; + + /* reset retry counters */ + reset_retry_counts(hdcp); + + /* reset error trace */ + memset(&hdcp->connection.trace, 0, sizeof(hdcp->connection.trace)); + + /* request authentication */ + if (current_state(hdcp) != HDCP_INITIALIZED) + set_state_id(hdcp, output, HDCP_INITIALIZED); + callback_in_ms(hdcp->connection.link.adjust.auth_delay * 1000, output); +out: + if (status != MOD_HDCP_STATUS_SUCCESS) + push_error_status(hdcp, status); + + return status; +} + +enum mod_hdcp_status mod_hdcp_remove_display(struct mod_hdcp *hdcp, + uint8_t index, struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + struct mod_hdcp_display *display = NULL; + + HDCP_TOP_INTERFACE_TRACE_WITH_INDEX(hdcp, index); + memset(output, 0, sizeof(struct mod_hdcp_output)); + + /* find display in connection */ + display = get_active_display_at_index(hdcp, index); + if (!display) { + status = MOD_HDCP_STATUS_SUCCESS; + goto out; + } + + /* stop current authentication */ + status = reset_authentication(hdcp, output); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + + /* remove display */ + display->state = MOD_HDCP_DISPLAY_INACTIVE; + + /* clear retry counters */ + reset_retry_counts(hdcp); + + /* reset error trace */ + memset(&hdcp->connection.trace, 0, sizeof(hdcp->connection.trace)); + + /* request authentication for remaining displays*/ + if (get_active_display_count(hdcp) > 0) + callback_in_ms(hdcp->connection.link.adjust.auth_delay * 1000, + output); +out: + if (status != MOD_HDCP_STATUS_SUCCESS) + push_error_status(hdcp, status); + return status; +} + +enum mod_hdcp_status mod_hdcp_query_display(struct mod_hdcp *hdcp, + uint8_t index, struct mod_hdcp_display_query *query) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + struct mod_hdcp_display *display = NULL; + + /* find display in connection */ + display = get_active_display_at_index(hdcp, index); + if (!display) { + status = MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; + goto out; + } + + /* populate query */ + query->link = &hdcp->connection.link; + query->display = display; + query->trace = &hdcp->connection.trace; + query->encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + + if (is_display_encryption_enabled(display)) { + if (is_hdcp1(hdcp)) { + query->encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP1_ON; + } else if (is_hdcp2(hdcp)) { + if (query->link->adjust.hdcp2.force_type == MOD_HDCP_FORCE_TYPE_0) + query->encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE0_ON; + else if (query->link->adjust.hdcp2.force_type == MOD_HDCP_FORCE_TYPE_1) + query->encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE1_ON; + else + query->encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP2_ON; + } + } else { + query->encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + } + +out: + return status; +} + +enum mod_hdcp_status mod_hdcp_reset_connection(struct mod_hdcp *hdcp, + struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + HDCP_TOP_INTERFACE_TRACE(hdcp); + status = reset_connection(hdcp, output); + if (status != MOD_HDCP_STATUS_SUCCESS) + push_error_status(hdcp, status); + + return status; +} + +enum mod_hdcp_status mod_hdcp_process_event(struct mod_hdcp *hdcp, + enum mod_hdcp_event event, struct mod_hdcp_output *output) +{ + enum mod_hdcp_status exec_status, trans_status, reset_status, status; + struct mod_hdcp_event_context event_ctx; + + HDCP_EVENT_TRACE(hdcp, event); + memset(output, 0, sizeof(struct mod_hdcp_output)); + memset(&event_ctx, 0, sizeof(struct mod_hdcp_event_context)); + event_ctx.event = event; + + /* execute and transition */ + exec_status = execution(hdcp, &event_ctx, &hdcp->auth.trans_input); + trans_status = transition( + hdcp, &event_ctx, &hdcp->auth.trans_input, output); + if (trans_status == MOD_HDCP_STATUS_SUCCESS) { + status = MOD_HDCP_STATUS_SUCCESS; + } else if (exec_status == MOD_HDCP_STATUS_SUCCESS) { + status = MOD_HDCP_STATUS_INTERNAL_POLICY_FAILURE; + push_error_status(hdcp, status); + } else { + status = exec_status; + push_error_status(hdcp, status); + } + + /* reset authentication if needed */ + if (trans_status == MOD_HDCP_STATUS_RESET_NEEDED) { + HDCP_FULL_DDC_TRACE(hdcp); + reset_status = reset_authentication(hdcp, output); + if (reset_status != MOD_HDCP_STATUS_SUCCESS) + push_error_status(hdcp, reset_status); + } + return status; +} + +enum mod_hdcp_operation_mode mod_hdcp_signal_type_to_operation_mode( + enum signal_type signal) +{ + enum mod_hdcp_operation_mode mode = MOD_HDCP_MODE_OFF; + + switch (signal) { + case SIGNAL_TYPE_DVI_SINGLE_LINK: + case SIGNAL_TYPE_HDMI_TYPE_A: + mode = MOD_HDCP_MODE_DEFAULT; + break; + case SIGNAL_TYPE_EDP: + case SIGNAL_TYPE_DISPLAY_PORT: + mode = MOD_HDCP_MODE_DP; + break; + case SIGNAL_TYPE_DISPLAY_PORT_MST: + mode = MOD_HDCP_MODE_DP_MST; + break; + default: + break; + } + + return mode; +} diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h new file mode 100644 index 000000000000..af78e4f1be68 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h @@ -0,0 +1,587 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef HDCP_H_ +#define HDCP_H_ + +#include "mod_hdcp.h" +#include "hdcp_log.h" + +#include <drm/drm_hdcp.h> +#include <drm/drm_dp_helper.h> + +enum mod_hdcp_trans_input_result { + UNKNOWN = 0, + PASS, + FAIL +}; + +struct mod_hdcp_transition_input_hdcp1 { + uint8_t bksv_read; + uint8_t bksv_validation; + uint8_t add_topology; + uint8_t create_session; + uint8_t an_write; + uint8_t aksv_write; + uint8_t ainfo_write; + uint8_t bcaps_read; + uint8_t r0p_read; + uint8_t rx_validation; + uint8_t encryption; + uint8_t link_maintenance; + uint8_t ready_check; + uint8_t bstatus_read; + uint8_t max_cascade_check; + uint8_t max_devs_check; + uint8_t device_count_check; + uint8_t ksvlist_read; + uint8_t vp_read; + uint8_t ksvlist_vp_validation; + + uint8_t hdcp_capable_dp; + uint8_t binfo_read_dp; + uint8_t r0p_available_dp; + uint8_t link_integrity_check; + uint8_t reauth_request_check; + uint8_t stream_encryption_dp; +}; + +struct mod_hdcp_transition_input_hdcp2 { + uint8_t hdcp2version_read; + uint8_t hdcp2_capable_check; + uint8_t add_topology; + uint8_t create_session; + uint8_t ake_init_prepare; + uint8_t ake_init_write; + uint8_t rxstatus_read; + uint8_t ake_cert_available; + uint8_t ake_cert_read; + uint8_t ake_cert_validation; + uint8_t stored_km_write; + uint8_t no_stored_km_write; + uint8_t h_prime_available; + uint8_t h_prime_read; + uint8_t pairing_available; + uint8_t pairing_info_read; + uint8_t h_prime_validation; + uint8_t lc_init_prepare; + uint8_t lc_init_write; + uint8_t l_prime_available_poll; + uint8_t l_prime_read; + uint8_t l_prime_validation; + uint8_t eks_prepare; + uint8_t eks_write; + uint8_t enable_encryption; + uint8_t reauth_request_check; + uint8_t rx_id_list_read; + uint8_t device_count_check; + uint8_t rx_id_list_validation; + uint8_t repeater_auth_ack_write; + uint8_t prepare_stream_manage; + uint8_t stream_manage_write; + uint8_t stream_ready_available; + uint8_t stream_ready_read; + uint8_t stream_ready_validation; + + uint8_t rx_caps_read_dp; + uint8_t content_stream_type_write; + uint8_t link_integrity_check_dp; + uint8_t stream_encryption_dp; +}; + +union mod_hdcp_transition_input { + struct mod_hdcp_transition_input_hdcp1 hdcp1; + struct mod_hdcp_transition_input_hdcp2 hdcp2; +}; + +struct mod_hdcp_message_hdcp1 { + uint8_t an[8]; + uint8_t aksv[5]; + uint8_t ainfo; + uint8_t bksv[5]; + uint16_t r0p; + uint8_t bcaps; + uint16_t bstatus; + uint8_t ksvlist[635]; + uint16_t ksvlist_size; + uint8_t vp[20]; + + uint16_t binfo_dp; +}; + +struct mod_hdcp_message_hdcp2 { + uint8_t hdcp2version_hdmi; + uint8_t rxcaps_dp[3]; + uint8_t rxstatus[2]; + + uint8_t ake_init[12]; + uint8_t ake_cert[534]; + uint8_t ake_no_stored_km[129]; + uint8_t ake_stored_km[33]; + uint8_t ake_h_prime[33]; + uint8_t ake_pairing_info[17]; + uint8_t lc_init[9]; + uint8_t lc_l_prime[33]; + uint8_t ske_eks[25]; + uint8_t rx_id_list[177]; // 22 + 5 * 31 + uint16_t rx_id_list_size; + uint8_t repeater_auth_ack[17]; + uint8_t repeater_auth_stream_manage[68]; // 6 + 2 * 31 + uint16_t stream_manage_size; + uint8_t repeater_auth_stream_ready[33]; + uint8_t rxstatus_dp; + uint8_t content_stream_type_dp[2]; +}; + +union mod_hdcp_message { + struct mod_hdcp_message_hdcp1 hdcp1; + struct mod_hdcp_message_hdcp2 hdcp2; +}; + +struct mod_hdcp_auth_counters { + uint8_t stream_management_retry_count; +}; + +/* contains values per connection */ +struct mod_hdcp_connection { + struct mod_hdcp_link link; + struct mod_hdcp_display displays[MAX_NUM_OF_DISPLAYS]; + uint8_t is_repeater; + uint8_t is_km_stored; + uint8_t is_hdcp2_revoked; + struct mod_hdcp_trace trace; + uint8_t hdcp1_retry_count; + uint8_t hdcp2_retry_count; +}; + +/* contains values per authentication cycle */ +struct mod_hdcp_authentication { + uint32_t id; + union mod_hdcp_message msg; + union mod_hdcp_transition_input trans_input; + struct mod_hdcp_auth_counters count; +}; + +/* contains values per state change */ +struct mod_hdcp_state { + uint8_t id; + uint32_t stay_count; +}; + +/* per event in a state */ +struct mod_hdcp_event_context { + enum mod_hdcp_event event; + uint8_t rx_id_list_ready; + uint8_t unexpected_event; +}; + +struct mod_hdcp { + /* per link */ + struct mod_hdcp_config config; + /* per connection */ + struct mod_hdcp_connection connection; + /* per authentication attempt */ + struct mod_hdcp_authentication auth; + /* per state in an authentication */ + struct mod_hdcp_state state; + /* reserved memory buffer */ + uint8_t buf[2025]; +}; + +enum mod_hdcp_initial_state_id { + HDCP_UNINITIALIZED = 0x0, + HDCP_INITIAL_STATE_START = HDCP_UNINITIALIZED, + HDCP_INITIALIZED, + HDCP_CP_NOT_DESIRED, + HDCP_INITIAL_STATE_END = HDCP_CP_NOT_DESIRED +}; + +enum mod_hdcp_hdcp1_state_id { + HDCP1_STATE_START = HDCP_INITIAL_STATE_END, + H1_A0_WAIT_FOR_ACTIVE_RX, + H1_A1_EXCHANGE_KSVS, + H1_A2_COMPUTATIONS_A3_VALIDATE_RX_A6_TEST_FOR_REPEATER, + H1_A45_AUTHENTICATED, + H1_A8_WAIT_FOR_READY, + H1_A9_READ_KSV_LIST, + HDCP1_STATE_END = H1_A9_READ_KSV_LIST +}; + +enum mod_hdcp_hdcp1_dp_state_id { + HDCP1_DP_STATE_START = HDCP1_STATE_END, + D1_A0_DETERMINE_RX_HDCP_CAPABLE, + D1_A1_EXCHANGE_KSVS, + D1_A23_WAIT_FOR_R0_PRIME, + D1_A2_COMPUTATIONS_A3_VALIDATE_RX_A5_TEST_FOR_REPEATER, + D1_A4_AUTHENTICATED, + D1_A6_WAIT_FOR_READY, + D1_A7_READ_KSV_LIST, + HDCP1_DP_STATE_END = D1_A7_READ_KSV_LIST, +}; + +enum mod_hdcp_hdcp2_state_id { + HDCP2_STATE_START = HDCP1_DP_STATE_END, + H2_A0_KNOWN_HDCP2_CAPABLE_RX, + H2_A1_SEND_AKE_INIT, + H2_A1_VALIDATE_AKE_CERT, + H2_A1_SEND_NO_STORED_KM, + H2_A1_READ_H_PRIME, + H2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME, + H2_A1_SEND_STORED_KM, + H2_A1_VALIDATE_H_PRIME, + H2_A2_LOCALITY_CHECK, + H2_A3_EXCHANGE_KS_AND_TEST_FOR_REPEATER, + H2_ENABLE_ENCRYPTION, + H2_A5_AUTHENTICATED, + H2_A6_WAIT_FOR_RX_ID_LIST, + H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK, + H2_A9_SEND_STREAM_MANAGEMENT, + H2_A9_VALIDATE_STREAM_READY, + HDCP2_STATE_END = H2_A9_VALIDATE_STREAM_READY, +}; + +enum mod_hdcp_hdcp2_dp_state_id { + HDCP2_DP_STATE_START = HDCP2_STATE_END, + D2_A0_DETERMINE_RX_HDCP_CAPABLE, + D2_A1_SEND_AKE_INIT, + D2_A1_VALIDATE_AKE_CERT, + D2_A1_SEND_NO_STORED_KM, + D2_A1_READ_H_PRIME, + D2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME, + D2_A1_SEND_STORED_KM, + D2_A1_VALIDATE_H_PRIME, + D2_A2_LOCALITY_CHECK, + D2_A34_EXCHANGE_KS_AND_TEST_FOR_REPEATER, + D2_SEND_CONTENT_STREAM_TYPE, + D2_ENABLE_ENCRYPTION, + D2_A5_AUTHENTICATED, + D2_A6_WAIT_FOR_RX_ID_LIST, + D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK, + D2_A9_SEND_STREAM_MANAGEMENT, + D2_A9_VALIDATE_STREAM_READY, + HDCP2_DP_STATE_END = D2_A9_VALIDATE_STREAM_READY, + HDCP_STATE_END = HDCP2_DP_STATE_END, +}; + +/* hdcp1 executions and transitions */ +typedef enum mod_hdcp_status (*mod_hdcp_action)(struct mod_hdcp *hdcp); +uint8_t mod_hdcp_execute_and_set( + mod_hdcp_action func, uint8_t *flag, + enum mod_hdcp_status *status, struct mod_hdcp *hdcp, char *str); +enum mod_hdcp_status mod_hdcp_hdcp1_execution(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input); +enum mod_hdcp_status mod_hdcp_hdcp1_dp_execution(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input); +enum mod_hdcp_status mod_hdcp_hdcp1_transition(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input, + struct mod_hdcp_output *output); +enum mod_hdcp_status mod_hdcp_hdcp1_dp_transition(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input, + struct mod_hdcp_output *output); + +/* hdcp2 executions and transitions */ +enum mod_hdcp_status mod_hdcp_hdcp2_execution(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input); +enum mod_hdcp_status mod_hdcp_hdcp2_dp_execution(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input); +enum mod_hdcp_status mod_hdcp_hdcp2_transition(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input, + struct mod_hdcp_output *output); +enum mod_hdcp_status mod_hdcp_hdcp2_dp_transition(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input, + struct mod_hdcp_output *output); + +/* log functions */ +void mod_hdcp_dump_binary_message(uint8_t *msg, uint32_t msg_size, + uint8_t *buf, uint32_t buf_size); +/* TODO: add adjustment log */ + +/* psp functions */ +enum mod_hdcp_status mod_hdcp_add_display_topology( + struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_remove_display_topology( + struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp1_destroy_session(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp1_validate_rx(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp1_enable_encryption(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp1_validate_ksvlist_vp(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption( + struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp1_link_maintenance(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp1_get_link_encryption_status(struct mod_hdcp *hdcp, + enum mod_hdcp_encryption_status *encryption_status); +enum mod_hdcp_status mod_hdcp_hdcp2_create_session(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_destroy_session(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_prepare_ake_init(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_validate_ake_cert(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_validate_h_prime(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_prepare_lc_init(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_validate_l_prime(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_prepare_eks(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_enable_encryption(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_validate_rx_id_list(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption( + struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_prepare_stream_management( + struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_validate_stream_ready( + struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_hdcp2_get_link_encryption_status(struct mod_hdcp *hdcp, + enum mod_hdcp_encryption_status *encryption_status); + +/* ddc functions */ +enum mod_hdcp_status mod_hdcp_read_bksv(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_bcaps(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_bstatus(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_r0p(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_ksvlist(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_vp(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_binfo(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_aksv(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_ainfo(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_an(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_hdcp2version(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_rxcaps(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_rxstatus(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_ake_cert(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_h_prime(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_pairing_info(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_l_prime(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_rx_id_list(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_read_stream_ready(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_ake_init(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_no_stored_km(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_stored_km(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_lc_init(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_eks(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_repeater_auth_ack(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_stream_manage(struct mod_hdcp *hdcp); +enum mod_hdcp_status mod_hdcp_write_content_type(struct mod_hdcp *hdcp); + +/* hdcp version helpers */ +static inline uint8_t is_dp_hdcp(struct mod_hdcp *hdcp) +{ + return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP || + hdcp->connection.link.mode == MOD_HDCP_MODE_DP_MST); +} + +static inline uint8_t is_dp_mst_hdcp(struct mod_hdcp *hdcp) +{ + return (hdcp->connection.link.mode == MOD_HDCP_MODE_DP_MST); +} + +static inline uint8_t is_hdmi_dvi_sl_hdcp(struct mod_hdcp *hdcp) +{ + return (hdcp->connection.link.mode == MOD_HDCP_MODE_DEFAULT); +} + +/* hdcp state helpers */ +static inline uint8_t current_state(struct mod_hdcp *hdcp) +{ + return hdcp->state.id; +} + +static inline void set_state_id(struct mod_hdcp *hdcp, + struct mod_hdcp_output *output, uint8_t id) +{ + memset(&hdcp->state, 0, sizeof(hdcp->state)); + hdcp->state.id = id; + /* callback timer should be reset per state */ + output->callback_stop = 1; + output->watchdog_timer_stop = 1; + HDCP_NEXT_STATE_TRACE(hdcp, id, output); +} + +static inline uint8_t is_in_hdcp1_states(struct mod_hdcp *hdcp) +{ + return (current_state(hdcp) > HDCP1_STATE_START && + current_state(hdcp) <= HDCP1_STATE_END); +} + +static inline uint8_t is_in_hdcp1_dp_states(struct mod_hdcp *hdcp) +{ + return (current_state(hdcp) > HDCP1_DP_STATE_START && + current_state(hdcp) <= HDCP1_DP_STATE_END); +} + +static inline uint8_t is_in_hdcp2_states(struct mod_hdcp *hdcp) +{ + return (current_state(hdcp) > HDCP2_STATE_START && + current_state(hdcp) <= HDCP2_STATE_END); +} + +static inline uint8_t is_in_hdcp2_dp_states(struct mod_hdcp *hdcp) +{ + return (current_state(hdcp) > HDCP2_DP_STATE_START && + current_state(hdcp) <= HDCP2_DP_STATE_END); +} + +static inline uint8_t is_hdcp1(struct mod_hdcp *hdcp) +{ + return (is_in_hdcp1_states(hdcp) || is_in_hdcp1_dp_states(hdcp)); +} + +static inline uint8_t is_hdcp2(struct mod_hdcp *hdcp) +{ + return (is_in_hdcp2_states(hdcp) || is_in_hdcp2_dp_states(hdcp)); +} + +static inline uint8_t is_in_cp_not_desired_state(struct mod_hdcp *hdcp) +{ + return current_state(hdcp) == HDCP_CP_NOT_DESIRED; +} + +static inline uint8_t is_in_initialized_state(struct mod_hdcp *hdcp) +{ + return current_state(hdcp) == HDCP_INITIALIZED; +} + +/* transition operation helpers */ +static inline void increment_stay_counter(struct mod_hdcp *hdcp) +{ + hdcp->state.stay_count++; +} + +static inline void fail_and_restart_in_ms(uint16_t time, + enum mod_hdcp_status *status, + struct mod_hdcp_output *output) +{ + output->callback_needed = 1; + output->callback_delay = time; + output->watchdog_timer_needed = 0; + output->watchdog_timer_delay = 0; + *status = MOD_HDCP_STATUS_RESET_NEEDED; +} + +static inline void callback_in_ms(uint16_t time, struct mod_hdcp_output *output) +{ + output->callback_needed = 1; + output->callback_delay = time; +} + +static inline void set_watchdog_in_ms(struct mod_hdcp *hdcp, uint16_t time, + struct mod_hdcp_output *output) +{ + output->watchdog_timer_needed = 1; + output->watchdog_timer_delay = time; +} + +/* connection topology helpers */ +static inline uint8_t is_display_active(struct mod_hdcp_display *display) +{ + return display->state >= MOD_HDCP_DISPLAY_ACTIVE; +} + +static inline uint8_t is_display_added(struct mod_hdcp_display *display) +{ + return display->state >= MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; +} + +static inline uint8_t is_display_encryption_enabled(struct mod_hdcp_display *display) +{ + return display->state >= MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; +} + +static inline uint8_t get_active_display_count(struct mod_hdcp *hdcp) +{ + uint8_t added_count = 0; + uint8_t i; + + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) + if (is_display_active(&hdcp->connection.displays[i])) + added_count++; + return added_count; +} + +static inline uint8_t get_added_display_count(struct mod_hdcp *hdcp) +{ + uint8_t added_count = 0; + uint8_t i; + + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) + if (is_display_added(&hdcp->connection.displays[i])) + added_count++; + return added_count; +} + +static inline struct mod_hdcp_display *get_first_added_display( + struct mod_hdcp *hdcp) +{ + uint8_t i; + struct mod_hdcp_display *display = NULL; + + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) + if (is_display_added(&hdcp->connection.displays[i])) { + display = &hdcp->connection.displays[i]; + break; + } + return display; +} + +static inline struct mod_hdcp_display *get_active_display_at_index( + struct mod_hdcp *hdcp, uint8_t index) +{ + uint8_t i; + struct mod_hdcp_display *display = NULL; + + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) + if (hdcp->connection.displays[i].index == index && + is_display_active(&hdcp->connection.displays[i])) { + display = &hdcp->connection.displays[i]; + break; + } + return display; +} + +static inline struct mod_hdcp_display *get_empty_display_container( + struct mod_hdcp *hdcp) +{ + uint8_t i; + struct mod_hdcp_display *display = NULL; + + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) + if (!is_display_active(&hdcp->connection.displays[i])) { + display = &hdcp->connection.displays[i]; + break; + } + return display; +} + +static inline void reset_retry_counts(struct mod_hdcp *hdcp) +{ + hdcp->connection.hdcp1_retry_count = 0; + hdcp->connection.hdcp2_retry_count = 0; +} + +#endif /* HDCP_H_ */ diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c new file mode 100644 index 000000000000..37670db64855 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c @@ -0,0 +1,529 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "hdcp.h" + +static inline enum mod_hdcp_status validate_bksv(struct mod_hdcp *hdcp) +{ + uint64_t n = 0; + uint8_t count = 0; + + memcpy(&n, hdcp->auth.msg.hdcp1.bksv, sizeof(uint64_t)); + + while (n) { + count++; + n &= (n - 1); + } + return (count == 20) ? MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP1_INVALID_BKSV; +} + +static inline enum mod_hdcp_status check_ksv_ready(struct mod_hdcp *hdcp) +{ + if (is_dp_hdcp(hdcp)) + return (hdcp->auth.msg.hdcp1.bstatus & DP_BSTATUS_READY) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP1_KSV_LIST_NOT_READY; + return (hdcp->auth.msg.hdcp1.bcaps & DRM_HDCP_DDC_BCAPS_KSV_FIFO_READY) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP1_KSV_LIST_NOT_READY; +} + +static inline enum mod_hdcp_status check_hdcp_capable_dp(struct mod_hdcp *hdcp) +{ + return (hdcp->auth.msg.hdcp1.bcaps & DP_BCAPS_HDCP_CAPABLE) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP1_NOT_CAPABLE; +} + +static inline enum mod_hdcp_status check_r0p_available_dp(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + if (is_dp_hdcp(hdcp)) { + status = (hdcp->auth.msg.hdcp1.bstatus & + DP_BSTATUS_R0_PRIME_READY) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP1_R0_PRIME_PENDING; + } else { + status = MOD_HDCP_STATUS_INVALID_OPERATION; + } + return status; +} + +static inline enum mod_hdcp_status check_link_integrity_dp( + struct mod_hdcp *hdcp) +{ + return (hdcp->auth.msg.hdcp1.bstatus & + DP_BSTATUS_LINK_FAILURE) ? + MOD_HDCP_STATUS_HDCP1_LINK_INTEGRITY_FAILURE : + MOD_HDCP_STATUS_SUCCESS; +} + +static inline enum mod_hdcp_status check_no_reauthentication_request_dp( + struct mod_hdcp *hdcp) +{ + return (hdcp->auth.msg.hdcp1.bstatus & DP_BSTATUS_REAUTH_REQ) ? + MOD_HDCP_STATUS_HDCP1_REAUTH_REQUEST_ISSUED : + MOD_HDCP_STATUS_SUCCESS; +} + +static inline enum mod_hdcp_status check_no_max_cascade(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = DRM_HDCP_MAX_CASCADE_EXCEEDED(hdcp->auth.msg.hdcp1.binfo_dp >> 8) + ? MOD_HDCP_STATUS_HDCP1_MAX_CASCADE_EXCEEDED_FAILURE + : MOD_HDCP_STATUS_SUCCESS; + else + status = DRM_HDCP_MAX_CASCADE_EXCEEDED(hdcp->auth.msg.hdcp1.bstatus >> 8) + ? MOD_HDCP_STATUS_HDCP1_MAX_CASCADE_EXCEEDED_FAILURE + : MOD_HDCP_STATUS_SUCCESS; + return status; +} + +static inline enum mod_hdcp_status check_no_max_devs(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = DRM_HDCP_MAX_DEVICE_EXCEEDED(hdcp->auth.msg.hdcp1.binfo_dp) ? + MOD_HDCP_STATUS_HDCP1_MAX_DEVS_EXCEEDED_FAILURE : + MOD_HDCP_STATUS_SUCCESS; + else + status = DRM_HDCP_MAX_DEVICE_EXCEEDED(hdcp->auth.msg.hdcp1.bstatus) ? + MOD_HDCP_STATUS_HDCP1_MAX_DEVS_EXCEEDED_FAILURE : + MOD_HDCP_STATUS_SUCCESS; + return status; +} + +static inline uint8_t get_device_count(struct mod_hdcp *hdcp) +{ + return is_dp_hdcp(hdcp) ? + DRM_HDCP_NUM_DOWNSTREAM(hdcp->auth.msg.hdcp1.binfo_dp) : + DRM_HDCP_NUM_DOWNSTREAM(hdcp->auth.msg.hdcp1.bstatus); +} + +static inline enum mod_hdcp_status check_device_count(struct mod_hdcp *hdcp) +{ + /* device count must be greater than or equal to tracked hdcp displays */ + return (get_device_count(hdcp) < get_added_display_count(hdcp)) ? + MOD_HDCP_STATUS_HDCP1_DEVICE_COUNT_MISMATCH_FAILURE : + MOD_HDCP_STATUS_SUCCESS; +} + +static enum mod_hdcp_status wait_for_active_rx(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_read_bksv, + &input->bksv_read, &status, + hdcp, "bksv_read")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_read_bcaps, + &input->bcaps_read, &status, + hdcp, "bcaps_read")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status exchange_ksvs(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_add_display_topology, + &input->add_topology, &status, + hdcp, "add_topology")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp1_create_session, + &input->create_session, &status, + hdcp, "create_session")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_write_an, + &input->an_write, &status, + hdcp, "an_write")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_write_aksv, + &input->aksv_write, &status, + hdcp, "aksv_write")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_read_bksv, + &input->bksv_read, &status, + hdcp, "bksv_read")) + goto out; + if (!mod_hdcp_execute_and_set(validate_bksv, + &input->bksv_validation, &status, + hdcp, "bksv_validation")) + goto out; + if (hdcp->auth.msg.hdcp1.ainfo) { + if (!mod_hdcp_execute_and_set(mod_hdcp_write_ainfo, + &input->ainfo_write, &status, + hdcp, "ainfo_write")) + goto out; + } +out: + return status; +} + +static enum mod_hdcp_status computations_validate_rx_test_for_repeater( + struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_read_r0p, + &input->r0p_read, &status, + hdcp, "r0p_read")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp1_validate_rx, + &input->rx_validation, &status, + hdcp, "rx_validation")) + goto out; + if (hdcp->connection.is_repeater) { + if (!hdcp->connection.link.adjust.hdcp1.postpone_encryption) + if (!mod_hdcp_execute_and_set( + mod_hdcp_hdcp1_enable_encryption, + &input->encryption, &status, + hdcp, "encryption")) + goto out; + } else { + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp1_enable_encryption, + &input->encryption, &status, + hdcp, "encryption")) + goto out; + if (is_dp_mst_hdcp(hdcp)) + if (!mod_hdcp_execute_and_set( + mod_hdcp_hdcp1_enable_dp_stream_encryption, + &input->stream_encryption_dp, &status, + hdcp, "stream_encryption_dp")) + goto out; + } +out: + return status; +} + +static enum mod_hdcp_status authenticated(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp1_link_maintenance, + &input->link_maintenance, &status, + hdcp, "link_maintenance")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status wait_for_ready(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ && + event_ctx->event != MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (is_dp_hdcp(hdcp)) { + if (!mod_hdcp_execute_and_set(mod_hdcp_read_bstatus, + &input->bstatus_read, &status, + hdcp, "bstatus_read")) + goto out; + if (!mod_hdcp_execute_and_set(check_link_integrity_dp, + &input->link_integrity_check, &status, + hdcp, "link_integrity_check")) + goto out; + if (!mod_hdcp_execute_and_set(check_no_reauthentication_request_dp, + &input->reauth_request_check, &status, + hdcp, "reauth_request_check")) + goto out; + } else { + if (!mod_hdcp_execute_and_set(mod_hdcp_read_bcaps, + &input->bcaps_read, &status, + hdcp, "bcaps_read")) + goto out; + } + if (!mod_hdcp_execute_and_set(check_ksv_ready, + &input->ready_check, &status, + hdcp, "ready_check")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status read_ksv_list(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + uint8_t device_count; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (is_dp_hdcp(hdcp)) { + if (!mod_hdcp_execute_and_set(mod_hdcp_read_binfo, + &input->binfo_read_dp, &status, + hdcp, "binfo_read_dp")) + goto out; + } else { + if (!mod_hdcp_execute_and_set(mod_hdcp_read_bstatus, + &input->bstatus_read, &status, + hdcp, "bstatus_read")) + goto out; + } + if (!mod_hdcp_execute_and_set(check_no_max_cascade, + &input->max_cascade_check, &status, + hdcp, "max_cascade_check")) + goto out; + if (!mod_hdcp_execute_and_set(check_no_max_devs, + &input->max_devs_check, &status, + hdcp, "max_devs_check")) + goto out; + if (!mod_hdcp_execute_and_set(check_device_count, + &input->device_count_check, &status, + hdcp, "device_count_check")) + goto out; + device_count = get_device_count(hdcp); + hdcp->auth.msg.hdcp1.ksvlist_size = device_count*5; + if (!mod_hdcp_execute_and_set(mod_hdcp_read_ksvlist, + &input->ksvlist_read, &status, + hdcp, "ksvlist_read")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_read_vp, + &input->vp_read, &status, + hdcp, "vp_read")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp1_validate_ksvlist_vp, + &input->ksvlist_vp_validation, &status, + hdcp, "ksvlist_vp_validation")) + goto out; + if (input->encryption != PASS) + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp1_enable_encryption, + &input->encryption, &status, + hdcp, "encryption")) + goto out; + if (is_dp_mst_hdcp(hdcp)) + if (!mod_hdcp_execute_and_set( + mod_hdcp_hdcp1_enable_dp_stream_encryption, + &input->stream_encryption_dp, &status, + hdcp, "stream_encryption_dp")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status determine_rx_hdcp_capable_dp(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_read_bcaps, + &input->bcaps_read, &status, + hdcp, "bcaps_read")) + goto out; + if (!mod_hdcp_execute_and_set(check_hdcp_capable_dp, + &input->hdcp_capable_dp, &status, + hdcp, "hdcp_capable_dp")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status wait_for_r0_prime_dp(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CPIRQ && + event_ctx->event != MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_read_bstatus, + &input->bstatus_read, &status, + hdcp, "bstatus_read")) + goto out; + if (!mod_hdcp_execute_and_set(check_r0p_available_dp, + &input->r0p_available_dp, &status, + hdcp, "r0p_available_dp")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status authenticated_dp(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CPIRQ) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_read_bstatus, + &input->bstatus_read, &status, + hdcp, "bstatus_read")) + goto out; + if (!mod_hdcp_execute_and_set(check_link_integrity_dp, + &input->link_integrity_check, &status, + hdcp, "link_integrity_check")) + goto out; + if (!mod_hdcp_execute_and_set(check_no_reauthentication_request_dp, + &input->reauth_request_check, &status, + hdcp, "reauth_request_check")) + goto out; +out: + return status; +} + +uint8_t mod_hdcp_execute_and_set( + mod_hdcp_action func, uint8_t *flag, + enum mod_hdcp_status *status, struct mod_hdcp *hdcp, char *str) +{ + *status = func(hdcp); + if (*status == MOD_HDCP_STATUS_SUCCESS && *flag != PASS) { + HDCP_INPUT_PASS_TRACE(hdcp, str); + *flag = PASS; + } else if (*status != MOD_HDCP_STATUS_SUCCESS && *flag != FAIL) { + HDCP_INPUT_FAIL_TRACE(hdcp, str); + *flag = FAIL; + } + return (*status == MOD_HDCP_STATUS_SUCCESS); +} + +enum mod_hdcp_status mod_hdcp_hdcp1_execution(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + switch (current_state(hdcp)) { + case H1_A0_WAIT_FOR_ACTIVE_RX: + status = wait_for_active_rx(hdcp, event_ctx, input); + break; + case H1_A1_EXCHANGE_KSVS: + status = exchange_ksvs(hdcp, event_ctx, input); + break; + case H1_A2_COMPUTATIONS_A3_VALIDATE_RX_A6_TEST_FOR_REPEATER: + status = computations_validate_rx_test_for_repeater(hdcp, + event_ctx, input); + break; + case H1_A45_AUTHENTICATED: + status = authenticated(hdcp, event_ctx, input); + break; + case H1_A8_WAIT_FOR_READY: + status = wait_for_ready(hdcp, event_ctx, input); + break; + case H1_A9_READ_KSV_LIST: + status = read_ksv_list(hdcp, event_ctx, input); + break; + default: + status = MOD_HDCP_STATUS_INVALID_STATE; + break; + } + + return status; +} + +extern enum mod_hdcp_status mod_hdcp_hdcp1_dp_execution(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + switch (current_state(hdcp)) { + case D1_A0_DETERMINE_RX_HDCP_CAPABLE: + status = determine_rx_hdcp_capable_dp(hdcp, event_ctx, input); + break; + case D1_A1_EXCHANGE_KSVS: + status = exchange_ksvs(hdcp, event_ctx, input); + break; + case D1_A23_WAIT_FOR_R0_PRIME: + status = wait_for_r0_prime_dp(hdcp, event_ctx, input); + break; + case D1_A2_COMPUTATIONS_A3_VALIDATE_RX_A5_TEST_FOR_REPEATER: + status = computations_validate_rx_test_for_repeater( + hdcp, event_ctx, input); + break; + case D1_A4_AUTHENTICATED: + status = authenticated_dp(hdcp, event_ctx, input); + break; + case D1_A6_WAIT_FOR_READY: + status = wait_for_ready(hdcp, event_ctx, input); + break; + case D1_A7_READ_KSV_LIST: + status = read_ksv_list(hdcp, event_ctx, input); + break; + default: + status = MOD_HDCP_STATUS_INVALID_STATE; + break; + } + + return status; +} diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c new file mode 100644 index 000000000000..76edcbe51f71 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_transition.c @@ -0,0 +1,319 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "hdcp.h" + +enum mod_hdcp_status mod_hdcp_hdcp1_transition(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input, + struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + struct mod_hdcp_connection *conn = &hdcp->connection; + struct mod_hdcp_link_adjustment *adjust = &hdcp->connection.link.adjust; + + switch (current_state(hdcp)) { + case H1_A0_WAIT_FOR_ACTIVE_RX: + if (input->bksv_read != PASS || input->bcaps_read != PASS) { + /* 1A-04: repeatedly attempts on port access failure */ + callback_in_ms(500, output); + increment_stay_counter(hdcp); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, H1_A1_EXCHANGE_KSVS); + break; + case H1_A1_EXCHANGE_KSVS: + if (input->add_topology != PASS || + input->create_session != PASS) { + /* out of sync with psp state */ + adjust->hdcp1.disable = 1; + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->an_write != PASS || + input->aksv_write != PASS || + input->bksv_read != PASS || + input->bksv_validation != PASS || + input->ainfo_write == FAIL) { + /* 1A-05: consider invalid bksv a failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(300, output); + set_state_id(hdcp, output, + H1_A2_COMPUTATIONS_A3_VALIDATE_RX_A6_TEST_FOR_REPEATER); + break; + case H1_A2_COMPUTATIONS_A3_VALIDATE_RX_A6_TEST_FOR_REPEATER: + if (input->bcaps_read != PASS || + input->r0p_read != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->rx_validation != PASS) { + /* 1A-06: consider invalid r0' a failure */ + /* 1A-08: consider bksv listed in SRM a failure */ + /* + * some slow RX will fail rx validation when it is + * not ready. give it more time to react before retry. + */ + fail_and_restart_in_ms(1000, &status, output); + break; + } else if (!conn->is_repeater && input->encryption != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + if (conn->is_repeater) { + callback_in_ms(0, output); + set_watchdog_in_ms(hdcp, 5000, output); + set_state_id(hdcp, output, H1_A8_WAIT_FOR_READY); + } else { + callback_in_ms(0, output); + set_state_id(hdcp, output, H1_A45_AUTHENTICATED); + HDCP_FULL_DDC_TRACE(hdcp); + } + break; + case H1_A45_AUTHENTICATED: + if (input->link_maintenance != PASS) { + /* 1A-07: consider invalid ri' a failure */ + /* 1A-07a: consider read ri' not returned a failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(500, output); + increment_stay_counter(hdcp); + break; + case H1_A8_WAIT_FOR_READY: + if (input->ready_check != PASS) { + if (event_ctx->event == + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + /* 1B-03: fail hdcp on ksv list READY timeout */ + /* prevent black screen in next attempt */ + adjust->hdcp1.postpone_encryption = 1; + fail_and_restart_in_ms(0, &status, output); + } else { + /* continue ksv list READY polling*/ + callback_in_ms(500, output); + increment_stay_counter(hdcp); + } + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, H1_A9_READ_KSV_LIST); + break; + case H1_A9_READ_KSV_LIST: + if (input->bstatus_read != PASS || + input->max_cascade_check != PASS || + input->max_devs_check != PASS || + input->device_count_check != PASS || + input->ksvlist_read != PASS || + input->vp_read != PASS || + input->ksvlist_vp_validation != PASS || + input->encryption != PASS) { + /* 1B-06: consider MAX_CASCADE_EXCEEDED a failure */ + /* 1B-05: consider MAX_DEVS_EXCEEDED a failure */ + /* 1B-04: consider invalid v' a failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, H1_A45_AUTHENTICATED); + HDCP_FULL_DDC_TRACE(hdcp); + break; + default: + status = MOD_HDCP_STATUS_INVALID_STATE; + fail_and_restart_in_ms(0, &status, output); + break; + } + + return status; +} + +enum mod_hdcp_status mod_hdcp_hdcp1_dp_transition(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp1 *input, + struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + struct mod_hdcp_connection *conn = &hdcp->connection; + struct mod_hdcp_link_adjustment *adjust = &hdcp->connection.link.adjust; + + switch (current_state(hdcp)) { + case D1_A0_DETERMINE_RX_HDCP_CAPABLE: + if (input->bcaps_read != PASS) { + /* 1A-04: no authentication on bcaps read failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->hdcp_capable_dp != PASS) { + adjust->hdcp1.disable = 1; + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, D1_A1_EXCHANGE_KSVS); + break; + case D1_A1_EXCHANGE_KSVS: + if (input->add_topology != PASS || + input->create_session != PASS) { + /* out of sync with psp state */ + adjust->hdcp1.disable = 1; + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->an_write != PASS || + input->aksv_write != PASS || + input->bksv_read != PASS || + input->bksv_validation != PASS || + input->ainfo_write == FAIL) { + /* 1A-05: consider invalid bksv a failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } + set_watchdog_in_ms(hdcp, 100, output); + set_state_id(hdcp, output, D1_A23_WAIT_FOR_R0_PRIME); + break; + case D1_A23_WAIT_FOR_R0_PRIME: + if (input->bstatus_read != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->r0p_available_dp != PASS) { + if (event_ctx->event == MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) + fail_and_restart_in_ms(0, &status, output); + else + increment_stay_counter(hdcp); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, D1_A2_COMPUTATIONS_A3_VALIDATE_RX_A5_TEST_FOR_REPEATER); + break; + case D1_A2_COMPUTATIONS_A3_VALIDATE_RX_A5_TEST_FOR_REPEATER: + if (input->r0p_read != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->rx_validation != PASS) { + if (hdcp->state.stay_count < 2) { + /* allow 2 additional retries */ + callback_in_ms(0, output); + increment_stay_counter(hdcp); + } else { + /* + * 1A-06: consider invalid r0' a failure + * after 3 attempts. + * 1A-08: consider bksv listed in SRM a failure + */ + /* + * some slow RX will fail rx validation when it is + * not ready. give it more time to react before retry. + */ + fail_and_restart_in_ms(1000, &status, output); + } + break; + } else if ((!conn->is_repeater && input->encryption != PASS) || + (!conn->is_repeater && is_dp_mst_hdcp(hdcp) && input->stream_encryption_dp != PASS)) { + fail_and_restart_in_ms(0, &status, output); + break; + } + if (conn->is_repeater) { + set_watchdog_in_ms(hdcp, 5000, output); + set_state_id(hdcp, output, D1_A6_WAIT_FOR_READY); + } else { + set_state_id(hdcp, output, D1_A4_AUTHENTICATED); + HDCP_FULL_DDC_TRACE(hdcp); + } + break; + case D1_A4_AUTHENTICATED: + if (input->link_integrity_check != PASS || + input->reauth_request_check != PASS) { + /* 1A-07: restart hdcp on a link integrity failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } + break; + case D1_A6_WAIT_FOR_READY: + if (input->link_integrity_check == FAIL || + input->reauth_request_check == FAIL) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->ready_check != PASS) { + if (event_ctx->event == + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + /* 1B-04: fail hdcp on ksv list READY timeout */ + /* prevent black screen in next attempt */ + adjust->hdcp1.postpone_encryption = 1; + fail_and_restart_in_ms(0, &status, output); + } else { + increment_stay_counter(hdcp); + } + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, D1_A7_READ_KSV_LIST); + break; + case D1_A7_READ_KSV_LIST: + if (input->binfo_read_dp != PASS || + input->max_cascade_check != PASS || + input->max_devs_check != PASS) { + /* 1B-06: consider MAX_DEVS_EXCEEDED a failure */ + /* 1B-07: consider MAX_CASCADE_EXCEEDED a failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->device_count_check != PASS) { + /* + * some slow dongle doesn't update + * device count as soon as downstream is connected. + * give it more time to react. + */ + adjust->hdcp1.postpone_encryption = 1; + fail_and_restart_in_ms(1000, &status, output); + break; + } else if (input->ksvlist_read != PASS || + input->vp_read != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->ksvlist_vp_validation != PASS) { + if (hdcp->state.stay_count < 2) { + /* allow 2 additional retries */ + callback_in_ms(0, output); + increment_stay_counter(hdcp); + } else { + /* + * 1B-05: consider invalid v' a failure + * after 3 attempts. + */ + fail_and_restart_in_ms(0, &status, output); + } + break; + } else if (input->encryption != PASS || + (is_dp_mst_hdcp(hdcp) && input->stream_encryption_dp != PASS)) { + fail_and_restart_in_ms(0, &status, output); + break; + } + set_state_id(hdcp, output, D1_A4_AUTHENTICATED); + HDCP_FULL_DDC_TRACE(hdcp); + break; + default: + fail_and_restart_in_ms(0, &status, output); + break; + } + + return status; +} diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c new file mode 100644 index 000000000000..f730b94ac3c0 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c @@ -0,0 +1,886 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include <linux/delay.h> + +#include "hdcp.h" + +static inline enum mod_hdcp_status check_receiver_id_list_ready(struct mod_hdcp *hdcp) +{ + uint8_t is_ready = 0; + + if (is_dp_hdcp(hdcp)) + is_ready = HDCP_2_2_DP_RXSTATUS_READY(hdcp->auth.msg.hdcp2.rxstatus_dp) ? 1 : 0; + else + is_ready = (HDCP_2_2_HDMI_RXSTATUS_READY(hdcp->auth.msg.hdcp2.rxstatus[0]) && + (HDCP_2_2_HDMI_RXSTATUS_MSG_SZ_HI(hdcp->auth.msg.hdcp2.rxstatus[1]) << 8 | + hdcp->auth.msg.hdcp2.rxstatus[0])) ? 1 : 0; + return is_ready ? MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_NOT_READY; +} + +static inline enum mod_hdcp_status check_hdcp2_capable(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = (hdcp->auth.msg.hdcp2.rxcaps_dp[2] & HDCP_2_2_RX_CAPS_VERSION_VAL) && + HDCP_2_2_DP_HDCP_CAPABLE(hdcp->auth.msg.hdcp2.rxcaps_dp[0]) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_NOT_CAPABLE; + else + status = (hdcp->auth.msg.hdcp2.hdcp2version_hdmi & HDCP_2_2_HDMI_SUPPORT_MASK) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_NOT_CAPABLE; + return status; +} + +static inline enum mod_hdcp_status check_reauthentication_request( + struct mod_hdcp *hdcp) +{ + uint8_t ret = 0; + + if (is_dp_hdcp(hdcp)) + ret = HDCP_2_2_DP_RXSTATUS_REAUTH_REQ(hdcp->auth.msg.hdcp2.rxstatus_dp) ? + MOD_HDCP_STATUS_HDCP2_REAUTH_REQUEST : + MOD_HDCP_STATUS_SUCCESS; + else + ret = HDCP_2_2_HDMI_RXSTATUS_REAUTH_REQ(hdcp->auth.msg.hdcp2.rxstatus[0]) ? + MOD_HDCP_STATUS_HDCP2_REAUTH_REQUEST : + MOD_HDCP_STATUS_SUCCESS; + return ret; +} + +static inline enum mod_hdcp_status check_link_integrity_failure_dp( + struct mod_hdcp *hdcp) +{ + return HDCP_2_2_DP_RXSTATUS_LINK_FAILED(hdcp->auth.msg.hdcp2.rxstatus_dp) ? + MOD_HDCP_STATUS_HDCP2_REAUTH_LINK_INTEGRITY_FAILURE : + MOD_HDCP_STATUS_SUCCESS; +} + +static enum mod_hdcp_status check_ake_cert_available(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + uint16_t size; + + if (is_dp_hdcp(hdcp)) { + status = MOD_HDCP_STATUS_SUCCESS; + } else { + status = mod_hdcp_read_rxstatus(hdcp); + if (status == MOD_HDCP_STATUS_SUCCESS) { + size = HDCP_2_2_HDMI_RXSTATUS_MSG_SZ_HI(hdcp->auth.msg.hdcp2.rxstatus[1]) << 8 | + hdcp->auth.msg.hdcp2.rxstatus[0]; + status = (size == sizeof(hdcp->auth.msg.hdcp2.ake_cert)) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_AKE_CERT_PENDING; + } + } + return status; +} + +static enum mod_hdcp_status check_h_prime_available(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + uint8_t size; + + status = mod_hdcp_read_rxstatus(hdcp); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + + if (is_dp_hdcp(hdcp)) { + status = HDCP_2_2_DP_RXSTATUS_H_PRIME(hdcp->auth.msg.hdcp2.rxstatus_dp) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_H_PRIME_PENDING; + } else { + size = HDCP_2_2_HDMI_RXSTATUS_MSG_SZ_HI(hdcp->auth.msg.hdcp2.rxstatus[1]) << 8 | + hdcp->auth.msg.hdcp2.rxstatus[0]; + status = (size == sizeof(hdcp->auth.msg.hdcp2.ake_h_prime)) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_H_PRIME_PENDING; + } +out: + return status; +} + +static enum mod_hdcp_status check_pairing_info_available(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + uint8_t size; + + status = mod_hdcp_read_rxstatus(hdcp); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + + if (is_dp_hdcp(hdcp)) { + status = HDCP_2_2_DP_RXSTATUS_PAIRING(hdcp->auth.msg.hdcp2.rxstatus_dp) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_PAIRING_INFO_PENDING; + } else { + size = HDCP_2_2_HDMI_RXSTATUS_MSG_SZ_HI(hdcp->auth.msg.hdcp2.rxstatus[1]) << 8 | + hdcp->auth.msg.hdcp2.rxstatus[0]; + status = (size == sizeof(hdcp->auth.msg.hdcp2.ake_pairing_info)) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_PAIRING_INFO_PENDING; + } +out: + return status; +} + +static enum mod_hdcp_status poll_l_prime_available(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + uint8_t size; + uint16_t max_wait = 20; // units of ms + uint16_t num_polls = 5; + uint16_t wait_time = max_wait / num_polls; + + if (is_dp_hdcp(hdcp)) + status = MOD_HDCP_STATUS_INVALID_OPERATION; + else + for (; num_polls; num_polls--) { + msleep(wait_time); + + status = mod_hdcp_read_rxstatus(hdcp); + if (status != MOD_HDCP_STATUS_SUCCESS) + break; + + size = HDCP_2_2_HDMI_RXSTATUS_MSG_SZ_HI(hdcp->auth.msg.hdcp2.rxstatus[1]) << 8 | + hdcp->auth.msg.hdcp2.rxstatus[0]; + status = (size == sizeof(hdcp->auth.msg.hdcp2.lc_l_prime)) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_L_PRIME_PENDING; + if (status == MOD_HDCP_STATUS_SUCCESS) + break; + } + return status; +} + +static enum mod_hdcp_status check_stream_ready_available(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + uint8_t size; + + if (is_dp_hdcp(hdcp)) { + status = MOD_HDCP_STATUS_INVALID_OPERATION; + } else { + status = mod_hdcp_read_rxstatus(hdcp); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + size = HDCP_2_2_HDMI_RXSTATUS_MSG_SZ_HI(hdcp->auth.msg.hdcp2.rxstatus[1]) << 8 | + hdcp->auth.msg.hdcp2.rxstatus[0]; + status = (size == sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_ready)) ? + MOD_HDCP_STATUS_SUCCESS : + MOD_HDCP_STATUS_HDCP2_STREAM_READY_PENDING; + } +out: + return status; +} + +static inline uint8_t get_device_count(struct mod_hdcp *hdcp) +{ + return HDCP_2_2_DEV_COUNT_LO(hdcp->auth.msg.hdcp2.rx_id_list[2]) + + (HDCP_2_2_DEV_COUNT_HI(hdcp->auth.msg.hdcp2.rx_id_list[1]) << 4); +} + +static enum mod_hdcp_status check_device_count(struct mod_hdcp *hdcp) +{ + /* device count must be greater than or equal to tracked hdcp displays */ + return (get_device_count(hdcp) < get_added_display_count(hdcp)) ? + MOD_HDCP_STATUS_HDCP2_DEVICE_COUNT_MISMATCH_FAILURE : + MOD_HDCP_STATUS_SUCCESS; +} + +static uint8_t process_rxstatus(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input, + enum mod_hdcp_status *status) +{ + if (!mod_hdcp_execute_and_set(mod_hdcp_read_rxstatus, + &input->rxstatus_read, status, + hdcp, "rxstatus_read")) + goto out; + if (!mod_hdcp_execute_and_set(check_reauthentication_request, + &input->reauth_request_check, status, + hdcp, "reauth_request_check")) + goto out; + if (is_dp_hdcp(hdcp)) { + if (!mod_hdcp_execute_and_set(check_link_integrity_failure_dp, + &input->link_integrity_check_dp, status, + hdcp, "link_integrity_check_dp")) + goto out; + } + if (hdcp->connection.is_repeater) + if (check_receiver_id_list_ready(hdcp) == + MOD_HDCP_STATUS_SUCCESS) { + HDCP_INPUT_PASS_TRACE(hdcp, "rx_id_list_ready"); + event_ctx->rx_id_list_ready = 1; + if (is_dp_hdcp(hdcp)) + hdcp->auth.msg.hdcp2.rx_id_list_size = + sizeof(hdcp->auth.msg.hdcp2.rx_id_list); + else + hdcp->auth.msg.hdcp2.rx_id_list_size = + HDCP_2_2_HDMI_RXSTATUS_MSG_SZ_HI(hdcp->auth.msg.hdcp2.rxstatus[1]) << 8 | + hdcp->auth.msg.hdcp2.rxstatus[0]; + } +out: + return (*status == MOD_HDCP_STATUS_SUCCESS); +} + +static enum mod_hdcp_status known_hdcp2_capable_rx(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + if (!mod_hdcp_execute_and_set(mod_hdcp_read_hdcp2version, + &input->hdcp2version_read, &status, + hdcp, "hdcp2version_read")) + goto out; + if (!mod_hdcp_execute_and_set(check_hdcp2_capable, + &input->hdcp2_capable_check, &status, + hdcp, "hdcp2_capable")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status send_ake_init(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + if (!mod_hdcp_execute_and_set(mod_hdcp_add_display_topology, + &input->add_topology, &status, + hdcp, "add_topology")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_create_session, + &input->create_session, &status, + hdcp, "create_session")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_prepare_ake_init, + &input->ake_init_prepare, &status, + hdcp, "ake_init_prepare")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_write_ake_init, + &input->ake_init_write, &status, + hdcp, "ake_init_write")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status validate_ake_cert(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (is_hdmi_dvi_sl_hdcp(hdcp)) + if (!mod_hdcp_execute_and_set(check_ake_cert_available, + &input->ake_cert_available, &status, + hdcp, "ake_cert_available")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_read_ake_cert, + &input->ake_cert_read, &status, + hdcp, "ake_cert_read")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_validate_ake_cert, + &input->ake_cert_validation, &status, + hdcp, "ake_cert_validation")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status send_no_stored_km(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_write_no_stored_km, + &input->no_stored_km_write, &status, + hdcp, "no_stored_km_write")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status read_h_prime(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ && + event_ctx->event != MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(check_h_prime_available, + &input->h_prime_available, &status, + hdcp, "h_prime_available")) + goto out; + + if (!mod_hdcp_execute_and_set(mod_hdcp_read_h_prime, + &input->h_prime_read, &status, + hdcp, "h_prime_read")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status read_pairing_info_and_validate_h_prime( + struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ && + event_ctx->event != MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(check_pairing_info_available, + &input->pairing_available, &status, + hdcp, "pairing_available")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_read_pairing_info, + &input->pairing_info_read, &status, + hdcp, "pairing_info_read")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_validate_h_prime, + &input->h_prime_validation, &status, + hdcp, "h_prime_validation")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status send_stored_km(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_write_stored_km, + &input->stored_km_write, &status, + hdcp, "stored_km_write")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status validate_h_prime(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ && + event_ctx->event != MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(check_h_prime_available, + &input->h_prime_available, &status, + hdcp, "h_prime_available")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_read_h_prime, + &input->h_prime_read, &status, + hdcp, "h_prime_read")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_validate_h_prime, + &input->h_prime_validation, &status, + hdcp, "h_prime_validation")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status locality_check(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_prepare_lc_init, + &input->lc_init_prepare, &status, + hdcp, "lc_init_prepare")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_write_lc_init, + &input->lc_init_write, &status, + hdcp, "lc_init_write")) + goto out; + if (is_dp_hdcp(hdcp)) + msleep(16); + else + if (!mod_hdcp_execute_and_set(poll_l_prime_available, + &input->l_prime_available_poll, &status, + hdcp, "l_prime_available_poll")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_read_l_prime, + &input->l_prime_read, &status, + hdcp, "l_prime_read")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_validate_l_prime, + &input->l_prime_validation, &status, + hdcp, "l_prime_validation")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status exchange_ks_and_test_for_repeater(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_prepare_eks, + &input->eks_prepare, &status, + hdcp, "eks_prepare")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_write_eks, + &input->eks_write, &status, + hdcp, "eks_write")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status enable_encryption(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ) { + event_ctx->unexpected_event = 1; + goto out; + } + if (event_ctx->event == MOD_HDCP_EVENT_CPIRQ) { + process_rxstatus(hdcp, event_ctx, input, &status); + goto out; + } + + if (is_hdmi_dvi_sl_hdcp(hdcp)) { + if (!process_rxstatus(hdcp, event_ctx, input, &status)) + goto out; + if (event_ctx->rx_id_list_ready) + goto out; + } + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_enable_encryption, + &input->enable_encryption, &status, + hdcp, "enable_encryption")) + goto out; + if (is_dp_mst_hdcp(hdcp)) { + if (!mod_hdcp_execute_and_set( + mod_hdcp_hdcp2_enable_dp_stream_encryption, + &input->stream_encryption_dp, &status, + hdcp, "stream_encryption_dp")) + goto out; + } +out: + return status; +} + +static enum mod_hdcp_status authenticated(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!process_rxstatus(hdcp, event_ctx, input, &status)) + goto out; + if (event_ctx->rx_id_list_ready) + goto out; +out: + return status; +} + +static enum mod_hdcp_status wait_for_rx_id_list(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ && + event_ctx->event != MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!process_rxstatus(hdcp, event_ctx, input, &status)) + goto out; + if (!event_ctx->rx_id_list_ready) { + status = MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_NOT_READY; + goto out; + } +out: + return status; +} + +static enum mod_hdcp_status verify_rx_id_list_and_send_ack(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ) { + event_ctx->unexpected_event = 1; + goto out; + } + if (event_ctx->event == MOD_HDCP_EVENT_CPIRQ) { + process_rxstatus(hdcp, event_ctx, input, &status); + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_read_rx_id_list, + &input->rx_id_list_read, + &status, hdcp, "receiver_id_list_read")) + goto out; + if (!mod_hdcp_execute_and_set(check_device_count, + &input->device_count_check, + &status, hdcp, "device_count_check")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_validate_rx_id_list, + &input->rx_id_list_validation, + &status, hdcp, "rx_id_list_validation")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_write_repeater_auth_ack, + &input->repeater_auth_ack_write, + &status, hdcp, "repeater_auth_ack_write")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status send_stream_management(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ) { + event_ctx->unexpected_event = 1; + goto out; + } + if (event_ctx->event == MOD_HDCP_EVENT_CPIRQ) { + process_rxstatus(hdcp, event_ctx, input, &status); + goto out; + } + + if (is_hdmi_dvi_sl_hdcp(hdcp)) { + if (!process_rxstatus(hdcp, event_ctx, input, &status)) + goto out; + if (event_ctx->rx_id_list_ready) + goto out; + } + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_prepare_stream_management, + &input->prepare_stream_manage, + &status, hdcp, "prepare_stream_manage")) + goto out; + + if (!mod_hdcp_execute_and_set(mod_hdcp_write_stream_manage, + &input->stream_manage_write, + &status, hdcp, "stream_manage_write")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status validate_stream_ready(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ && + event_ctx->event != MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + event_ctx->unexpected_event = 1; + goto out; + } + if (event_ctx->event == MOD_HDCP_EVENT_CPIRQ) { + process_rxstatus(hdcp, event_ctx, input, &status); + goto out; + } + + if (is_hdmi_dvi_sl_hdcp(hdcp)) { + if (!process_rxstatus(hdcp, event_ctx, input, &status)) + goto out; + if (event_ctx->rx_id_list_ready) { + goto out; + } + } + if (is_hdmi_dvi_sl_hdcp(hdcp)) + if (!mod_hdcp_execute_and_set(check_stream_ready_available, + &input->stream_ready_available, + &status, hdcp, "stream_ready_available")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_read_stream_ready, + &input->stream_ready_read, + &status, hdcp, "stream_ready_read")) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_validate_stream_ready, + &input->stream_ready_validation, + &status, hdcp, "stream_ready_validation")) + goto out; + +out: + return status; +} + +static enum mod_hdcp_status determine_rx_hdcp_capable_dp(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!mod_hdcp_execute_and_set(mod_hdcp_read_rxcaps, + &input->rx_caps_read_dp, + &status, hdcp, "rx_caps_read_dp")) + goto out; + if (!mod_hdcp_execute_and_set(check_hdcp2_capable, + &input->hdcp2_capable_check, &status, + hdcp, "hdcp2_capable_check")) + goto out; +out: + return status; +} + +static enum mod_hdcp_status send_content_stream_type_dp(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK && + event_ctx->event != MOD_HDCP_EVENT_CPIRQ) { + event_ctx->unexpected_event = 1; + goto out; + } + + if (!process_rxstatus(hdcp, event_ctx, input, &status)) + goto out; + if (!mod_hdcp_execute_and_set(mod_hdcp_write_content_type, + &input->content_stream_type_write, &status, + hdcp, "content_stream_type_write")) + goto out; +out: + return status; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_execution(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + switch (current_state(hdcp)) { + case H2_A0_KNOWN_HDCP2_CAPABLE_RX: + status = known_hdcp2_capable_rx(hdcp, event_ctx, input); + break; + case H2_A1_SEND_AKE_INIT: + status = send_ake_init(hdcp, event_ctx, input); + break; + case H2_A1_VALIDATE_AKE_CERT: + status = validate_ake_cert(hdcp, event_ctx, input); + break; + case H2_A1_SEND_NO_STORED_KM: + status = send_no_stored_km(hdcp, event_ctx, input); + break; + case H2_A1_READ_H_PRIME: + status = read_h_prime(hdcp, event_ctx, input); + break; + case H2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME: + status = read_pairing_info_and_validate_h_prime(hdcp, + event_ctx, input); + break; + case H2_A1_SEND_STORED_KM: + status = send_stored_km(hdcp, event_ctx, input); + break; + case H2_A1_VALIDATE_H_PRIME: + status = validate_h_prime(hdcp, event_ctx, input); + break; + case H2_A2_LOCALITY_CHECK: + status = locality_check(hdcp, event_ctx, input); + break; + case H2_A3_EXCHANGE_KS_AND_TEST_FOR_REPEATER: + status = exchange_ks_and_test_for_repeater(hdcp, event_ctx, input); + break; + case H2_ENABLE_ENCRYPTION: + status = enable_encryption(hdcp, event_ctx, input); + break; + case H2_A5_AUTHENTICATED: + status = authenticated(hdcp, event_ctx, input); + break; + case H2_A6_WAIT_FOR_RX_ID_LIST: + status = wait_for_rx_id_list(hdcp, event_ctx, input); + break; + case H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK: + status = verify_rx_id_list_and_send_ack(hdcp, event_ctx, input); + break; + case H2_A9_SEND_STREAM_MANAGEMENT: + status = send_stream_management(hdcp, event_ctx, input); + break; + case H2_A9_VALIDATE_STREAM_READY: + status = validate_stream_ready(hdcp, event_ctx, input); + break; + default: + status = MOD_HDCP_STATUS_INVALID_STATE; + break; + } + + return status; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_dp_execution(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + switch (current_state(hdcp)) { + case D2_A0_DETERMINE_RX_HDCP_CAPABLE: + status = determine_rx_hdcp_capable_dp(hdcp, event_ctx, input); + break; + case D2_A1_SEND_AKE_INIT: + status = send_ake_init(hdcp, event_ctx, input); + break; + case D2_A1_VALIDATE_AKE_CERT: + status = validate_ake_cert(hdcp, event_ctx, input); + break; + case D2_A1_SEND_NO_STORED_KM: + status = send_no_stored_km(hdcp, event_ctx, input); + break; + case D2_A1_READ_H_PRIME: + status = read_h_prime(hdcp, event_ctx, input); + break; + case D2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME: + status = read_pairing_info_and_validate_h_prime(hdcp, + event_ctx, input); + break; + case D2_A1_SEND_STORED_KM: + status = send_stored_km(hdcp, event_ctx, input); + break; + case D2_A1_VALIDATE_H_PRIME: + status = validate_h_prime(hdcp, event_ctx, input); + break; + case D2_A2_LOCALITY_CHECK: + status = locality_check(hdcp, event_ctx, input); + break; + case D2_A34_EXCHANGE_KS_AND_TEST_FOR_REPEATER: + status = exchange_ks_and_test_for_repeater(hdcp, + event_ctx, input); + break; + case D2_SEND_CONTENT_STREAM_TYPE: + status = send_content_stream_type_dp(hdcp, event_ctx, input); + break; + case D2_ENABLE_ENCRYPTION: + status = enable_encryption(hdcp, event_ctx, input); + break; + case D2_A5_AUTHENTICATED: + status = authenticated(hdcp, event_ctx, input); + break; + case D2_A6_WAIT_FOR_RX_ID_LIST: + status = wait_for_rx_id_list(hdcp, event_ctx, input); + break; + case D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK: + status = verify_rx_id_list_and_send_ack(hdcp, event_ctx, input); + break; + case D2_A9_SEND_STREAM_MANAGEMENT: + status = send_stream_management(hdcp, event_ctx, input); + break; + case D2_A9_VALIDATE_STREAM_READY: + status = validate_stream_ready(hdcp, event_ctx, input); + break; + default: + status = MOD_HDCP_STATUS_INVALID_STATE; + break; + } + + return status; +} diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c new file mode 100644 index 000000000000..8cae3e3aacd5 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c @@ -0,0 +1,679 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "hdcp.h" + +enum mod_hdcp_status mod_hdcp_hdcp2_transition(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input, + struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + struct mod_hdcp_connection *conn = &hdcp->connection; + struct mod_hdcp_link_adjustment *adjust = &hdcp->connection.link.adjust; + + switch (current_state(hdcp)) { + case H2_A0_KNOWN_HDCP2_CAPABLE_RX: + if (input->hdcp2version_read != PASS || + input->hdcp2_capable_check != PASS) { + adjust->hdcp2.disable = 1; + callback_in_ms(0, output); + set_state_id(hdcp, output, HDCP_INITIALIZED); + } else { + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A1_SEND_AKE_INIT); + } + break; + case H2_A1_SEND_AKE_INIT: + if (input->add_topology != PASS || + input->create_session != PASS || + input->ake_init_prepare != PASS) { + /* out of sync with psp state */ + adjust->hdcp2.disable = 1; + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->ake_init_write != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + set_watchdog_in_ms(hdcp, 100, output); + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A1_VALIDATE_AKE_CERT); + break; + case H2_A1_VALIDATE_AKE_CERT: + if (input->ake_cert_available != PASS) { + if (event_ctx->event == + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + /* 1A-08: consider ake timeout a failure */ + /* some hdmi receivers are not ready for HDCP + * immediately after video becomes active, + * delay 1s before retry on first HDCP message + * timeout. + */ + fail_and_restart_in_ms(1000, &status, output); + } else { + /* continue ake cert polling*/ + callback_in_ms(10, output); + increment_stay_counter(hdcp); + } + break; + } else if (input->ake_cert_read != PASS || + input->ake_cert_validation != PASS) { + /* + * 1A-09: consider invalid ake cert a failure + * 1A-10: consider receiver id listed in SRM a failure + */ + fail_and_restart_in_ms(0, &status, output); + break; + } + if (conn->is_km_stored && + !adjust->hdcp2.force_no_stored_km) { + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A1_SEND_STORED_KM); + } else { + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A1_SEND_NO_STORED_KM); + } + break; + case H2_A1_SEND_NO_STORED_KM: + if (input->no_stored_km_write != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + if (adjust->hdcp2.increase_h_prime_timeout) + set_watchdog_in_ms(hdcp, 2000, output); + else + set_watchdog_in_ms(hdcp, 1000, output); + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A1_READ_H_PRIME); + break; + case H2_A1_READ_H_PRIME: + if (input->h_prime_available != PASS) { + if (event_ctx->event == + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + /* 1A-11-3: consider h' timeout a failure */ + fail_and_restart_in_ms(1000, &status, output); + } else { + /* continue h' polling */ + callback_in_ms(100, output); + increment_stay_counter(hdcp); + } + break; + } else if (input->h_prime_read != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + set_watchdog_in_ms(hdcp, 200, output); + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME); + break; + case H2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME: + if (input->pairing_available != PASS) { + if (event_ctx->event == + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + /* 1A-12: consider pairing info timeout + * a failure + */ + fail_and_restart_in_ms(0, &status, output); + } else { + /* continue pairing info polling */ + callback_in_ms(20, output); + increment_stay_counter(hdcp); + } + break; + } else if (input->pairing_info_read != PASS || + input->h_prime_validation != PASS) { + /* 1A-11-1: consider invalid h' a failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A2_LOCALITY_CHECK); + break; + case H2_A1_SEND_STORED_KM: + if (input->stored_km_write != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + set_watchdog_in_ms(hdcp, 200, output); + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A1_VALIDATE_H_PRIME); + break; + case H2_A1_VALIDATE_H_PRIME: + if (input->h_prime_available != PASS) { + if (event_ctx->event == + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + /* 1A-11-2: consider h' timeout a failure */ + fail_and_restart_in_ms(1000, &status, output); + } else { + /* continue h' polling */ + callback_in_ms(20, output); + increment_stay_counter(hdcp); + } + break; + } else if (input->h_prime_read != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->h_prime_validation != PASS) { + /* 1A-11-1: consider invalid h' a failure */ + adjust->hdcp2.force_no_stored_km = 1; + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A2_LOCALITY_CHECK); + break; + case H2_A2_LOCALITY_CHECK: + if (hdcp->state.stay_count > 10 || + input->lc_init_prepare != PASS || + input->lc_init_write != PASS || + input->l_prime_available_poll != PASS || + input->l_prime_read != PASS) { + /* + * 1A-05: consider disconnection after LC init a failure + * 1A-13-1: consider invalid l' a failure + * 1A-13-2: consider l' timeout a failure + */ + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->l_prime_validation != PASS) { + callback_in_ms(0, output); + increment_stay_counter(hdcp); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A3_EXCHANGE_KS_AND_TEST_FOR_REPEATER); + break; + case H2_A3_EXCHANGE_KS_AND_TEST_FOR_REPEATER: + if (input->eks_prepare != PASS || + input->eks_write != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + if (conn->is_repeater) { + set_watchdog_in_ms(hdcp, 3000, output); + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A6_WAIT_FOR_RX_ID_LIST); + } else { + /* some CTS equipment requires a delay GREATER than + * 200 ms, so delay 210 ms instead of 200 ms + */ + callback_in_ms(210, output); + set_state_id(hdcp, output, H2_ENABLE_ENCRYPTION); + } + break; + case H2_ENABLE_ENCRYPTION: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS) { + /* + * 1A-07: restart hdcp on REAUTH_REQ + * 1B-08: restart hdcp on REAUTH_REQ + */ + fail_and_restart_in_ms(0, &status, output); + break; + } else if (event_ctx->rx_id_list_ready && conn->is_repeater) { + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + } else if (input->enable_encryption != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A5_AUTHENTICATED); + HDCP_FULL_DDC_TRACE(hdcp); + break; + case H2_A5_AUTHENTICATED: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (event_ctx->rx_id_list_ready && conn->is_repeater) { + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + } + callback_in_ms(500, output); + increment_stay_counter(hdcp); + break; + case H2_A6_WAIT_FOR_RX_ID_LIST: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (!event_ctx->rx_id_list_ready) { + if (event_ctx->event == MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + /* 1B-02: consider rx id list timeout a failure */ + /* some CTS equipment's actual timeout + * measurement is slightly greater than 3000 ms. + * Delay 100 ms to ensure it is fully timeout + * before re-authentication. + */ + fail_and_restart_in_ms(100, &status, output); + } else { + callback_in_ms(300, output); + increment_stay_counter(hdcp); + } + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + case H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS || + input->rx_id_list_read != PASS || + input->device_count_check != PASS || + input->rx_id_list_validation != PASS || + input->repeater_auth_ack_write != PASS) { + /* 1B-03: consider invalid v' a failure + * 1B-04: consider MAX_DEVS_EXCEEDED a failure + * 1B-05: consider MAX_CASCADE_EXCEEDED a failure + * 1B-06: consider invalid seq_num_V a failure + * 1B-09: consider seq_num_V rollover a failure + */ + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A9_SEND_STREAM_MANAGEMENT); + break; + case H2_A9_SEND_STREAM_MANAGEMENT: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (event_ctx->rx_id_list_ready && conn->is_repeater) { + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + } else if (input->prepare_stream_manage != PASS || + input->stream_manage_write != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + set_watchdog_in_ms(hdcp, 100, output); + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A9_VALIDATE_STREAM_READY); + break; + case H2_A9_VALIDATE_STREAM_READY: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (event_ctx->rx_id_list_ready && conn->is_repeater) { + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + } else if (input->stream_ready_available != PASS) { + if (event_ctx->event == MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) { + /* 1B-10-2: restart content stream management on + * stream ready timeout + */ + hdcp->auth.count.stream_management_retry_count++; + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A9_SEND_STREAM_MANAGEMENT); + } else { + callback_in_ms(10, output); + increment_stay_counter(hdcp); + } + break; + } else if (input->stream_ready_read != PASS || + input->stream_ready_validation != PASS) { + /* + * 1B-10-1: restart content stream management + * on invalid M' + */ + if (hdcp->auth.count.stream_management_retry_count > 10) { + fail_and_restart_in_ms(0, &status, output); + } else { + hdcp->auth.count.stream_management_retry_count++; + callback_in_ms(0, output); + set_state_id(hdcp, output, H2_A9_SEND_STREAM_MANAGEMENT); + } + break; + } + callback_in_ms(200, output); + set_state_id(hdcp, output, H2_ENABLE_ENCRYPTION); + break; + default: + status = MOD_HDCP_STATUS_INVALID_STATE; + fail_and_restart_in_ms(0, &status, output); + break; + } + + return status; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_dp_transition(struct mod_hdcp *hdcp, + struct mod_hdcp_event_context *event_ctx, + struct mod_hdcp_transition_input_hdcp2 *input, + struct mod_hdcp_output *output) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + struct mod_hdcp_connection *conn = &hdcp->connection; + struct mod_hdcp_link_adjustment *adjust = &hdcp->connection.link.adjust; + + switch (current_state(hdcp)) { + case D2_A0_DETERMINE_RX_HDCP_CAPABLE: + if (input->rx_caps_read_dp != PASS || + input->hdcp2_capable_check != PASS) { + adjust->hdcp2.disable = 1; + callback_in_ms(0, output); + set_state_id(hdcp, output, HDCP_INITIALIZED); + } else { + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A1_SEND_AKE_INIT); + } + break; + case D2_A1_SEND_AKE_INIT: + if (input->add_topology != PASS || + input->create_session != PASS || + input->ake_init_prepare != PASS) { + /* out of sync with psp state */ + adjust->hdcp2.disable = 1; + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->ake_init_write != PASS) { + /* possibly display not ready */ + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(100, output); + set_state_id(hdcp, output, D2_A1_VALIDATE_AKE_CERT); + break; + case D2_A1_VALIDATE_AKE_CERT: + if (input->ake_cert_read != PASS || + input->ake_cert_validation != PASS) { + /* + * 1A-08: consider invalid ake cert a failure + * 1A-09: consider receiver id listed in SRM a failure + */ + fail_and_restart_in_ms(0, &status, output); + break; + } + if (conn->is_km_stored && + !adjust->hdcp2.force_no_stored_km) { + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A1_SEND_STORED_KM); + } else { + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A1_SEND_NO_STORED_KM); + } + break; + case D2_A1_SEND_NO_STORED_KM: + if (input->no_stored_km_write != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + if (adjust->hdcp2.increase_h_prime_timeout) + set_watchdog_in_ms(hdcp, 2000, output); + else + set_watchdog_in_ms(hdcp, 1000, output); + set_state_id(hdcp, output, D2_A1_READ_H_PRIME); + break; + case D2_A1_READ_H_PRIME: + if (input->h_prime_available != PASS) { + if (event_ctx->event == + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) + /* 1A-10-3: consider h' timeout a failure */ + fail_and_restart_in_ms(1000, &status, output); + else + increment_stay_counter(hdcp); + break; + } else if (input->h_prime_read != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + set_watchdog_in_ms(hdcp, 200, output); + set_state_id(hdcp, output, D2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME); + break; + case D2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME: + if (input->pairing_available != PASS) { + if (event_ctx->event == + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) + /* + * 1A-11: consider pairing info timeout + * a failure + */ + fail_and_restart_in_ms(0, &status, output); + else + increment_stay_counter(hdcp); + break; + } else if (input->pairing_info_read != PASS || + input->h_prime_validation != PASS) { + /* 1A-10-1: consider invalid h' a failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A2_LOCALITY_CHECK); + break; + case D2_A1_SEND_STORED_KM: + if (input->stored_km_write != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + set_watchdog_in_ms(hdcp, 200, output); + set_state_id(hdcp, output, D2_A1_VALIDATE_H_PRIME); + break; + case D2_A1_VALIDATE_H_PRIME: + if (input->h_prime_available != PASS) { + if (event_ctx->event == + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) + /* 1A-10-2: consider h' timeout a failure */ + fail_and_restart_in_ms(1000, &status, output); + else + increment_stay_counter(hdcp); + break; + } else if (input->h_prime_read != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->h_prime_validation != PASS) { + /* 1A-10-1: consider invalid h' a failure */ + adjust->hdcp2.force_no_stored_km = 1; + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A2_LOCALITY_CHECK); + break; + case D2_A2_LOCALITY_CHECK: + if (hdcp->state.stay_count > 10 || + input->lc_init_prepare != PASS || + input->lc_init_write != PASS || + input->l_prime_read != PASS) { + /* 1A-12: consider invalid l' a failure */ + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->l_prime_validation != PASS) { + callback_in_ms(0, output); + increment_stay_counter(hdcp); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A34_EXCHANGE_KS_AND_TEST_FOR_REPEATER); + break; + case D2_A34_EXCHANGE_KS_AND_TEST_FOR_REPEATER: + if (input->eks_prepare != PASS || + input->eks_write != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + if (conn->is_repeater) { + set_watchdog_in_ms(hdcp, 3000, output); + set_state_id(hdcp, output, D2_A6_WAIT_FOR_RX_ID_LIST); + } else { + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_SEND_CONTENT_STREAM_TYPE); + } + break; + case D2_SEND_CONTENT_STREAM_TYPE: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS || + input->link_integrity_check_dp != PASS || + input->content_stream_type_write != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(210, output); + set_state_id(hdcp, output, D2_ENABLE_ENCRYPTION); + break; + case D2_ENABLE_ENCRYPTION: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS || + input->link_integrity_check_dp != PASS) { + /* + * 1A-07: restart hdcp on REAUTH_REQ + * 1B-08: restart hdcp on REAUTH_REQ + */ + fail_and_restart_in_ms(0, &status, output); + break; + } else if (event_ctx->rx_id_list_ready && conn->is_repeater) { + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + } else if (input->enable_encryption != PASS || + (is_dp_mst_hdcp(hdcp) && input->stream_encryption_dp != PASS)) { + fail_and_restart_in_ms(0, &status, output); + break; + } + set_state_id(hdcp, output, D2_A5_AUTHENTICATED); + HDCP_FULL_DDC_TRACE(hdcp); + break; + case D2_A5_AUTHENTICATED: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (input->link_integrity_check_dp != PASS) { + if (hdcp->connection.hdcp2_retry_count >= 1) + adjust->hdcp2.force_type = MOD_HDCP_FORCE_TYPE_0; + fail_and_restart_in_ms(0, &status, output); + break; + } else if (event_ctx->rx_id_list_ready && conn->is_repeater) { + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + } + increment_stay_counter(hdcp); + break; + case D2_A6_WAIT_FOR_RX_ID_LIST: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS || + input->link_integrity_check_dp != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (!event_ctx->rx_id_list_ready) { + if (event_ctx->event == MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) + /* 1B-02: consider rx id list timeout a failure */ + fail_and_restart_in_ms(0, &status, output); + else + increment_stay_counter(hdcp); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + case D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS || + input->link_integrity_check_dp != PASS || + input->rx_id_list_read != PASS || + input->device_count_check != PASS || + input->rx_id_list_validation != PASS || + input->repeater_auth_ack_write != PASS) { + /* + * 1B-03: consider invalid v' a failure + * 1B-04: consider MAX_DEVS_EXCEEDED a failure + * 1B-05: consider MAX_CASCADE_EXCEEDED a failure + * 1B-06: consider invalid seq_num_V a failure + * 1B-09: consider seq_num_V rollover a failure + */ + fail_and_restart_in_ms(0, &status, output); + break; + } + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A9_SEND_STREAM_MANAGEMENT); + break; + case D2_A9_SEND_STREAM_MANAGEMENT: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS || + input->link_integrity_check_dp != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (event_ctx->rx_id_list_ready) { + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + } else if (input->prepare_stream_manage != PASS || + input->stream_manage_write != PASS) { + if (event_ctx->event == MOD_HDCP_EVENT_CALLBACK) + fail_and_restart_in_ms(0, &status, output); + else + increment_stay_counter(hdcp); + break; + } + callback_in_ms(100, output); + set_state_id(hdcp, output, D2_A9_VALIDATE_STREAM_READY); + break; + case D2_A9_VALIDATE_STREAM_READY: + if (input->rxstatus_read != PASS || + input->reauth_request_check != PASS || + input->link_integrity_check_dp != PASS) { + fail_and_restart_in_ms(0, &status, output); + break; + } else if (event_ctx->rx_id_list_ready) { + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK); + break; + } else if (input->stream_ready_read != PASS || + input->stream_ready_validation != PASS) { + /* + * 1B-10-1: restart content stream management + * on invalid M' + * 1B-10-2: consider stream ready timeout a failure + */ + if (hdcp->auth.count.stream_management_retry_count > 10) { + fail_and_restart_in_ms(0, &status, output); + } else if (event_ctx->event == MOD_HDCP_EVENT_CALLBACK) { + hdcp->auth.count.stream_management_retry_count++; + callback_in_ms(0, output); + set_state_id(hdcp, output, D2_A9_SEND_STREAM_MANAGEMENT); + } else { + increment_stay_counter(hdcp); + } + break; + } + callback_in_ms(200, output); + set_state_id(hdcp, output, D2_ENABLE_ENCRYPTION); + break; + default: + status = MOD_HDCP_STATUS_INVALID_STATE; + fail_and_restart_in_ms(0, &status, output); + break; + } + return status; +} diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c new file mode 100644 index 000000000000..ff9d54812e62 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c @@ -0,0 +1,631 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "hdcp.h" + +#define MIN(a, b) ((a) < (b) ? (a) : (b)) +#define HDCP_I2C_ADDR 0x3a /* 0x74 >> 1*/ +#define KSV_READ_SIZE 0xf /* 0x6803b - 0x6802c */ +#define HDCP_MAX_AUX_TRANSACTION_SIZE 16 + +enum mod_hdcp_ddc_message_id { + MOD_HDCP_MESSAGE_ID_INVALID = -1, + + /* HDCP 1.4 */ + + MOD_HDCP_MESSAGE_ID_READ_BKSV = 0, + MOD_HDCP_MESSAGE_ID_READ_RI_R0, + MOD_HDCP_MESSAGE_ID_WRITE_AKSV, + MOD_HDCP_MESSAGE_ID_WRITE_AINFO, + MOD_HDCP_MESSAGE_ID_WRITE_AN, + MOD_HDCP_MESSAGE_ID_READ_VH_X, + MOD_HDCP_MESSAGE_ID_READ_VH_0, + MOD_HDCP_MESSAGE_ID_READ_VH_1, + MOD_HDCP_MESSAGE_ID_READ_VH_2, + MOD_HDCP_MESSAGE_ID_READ_VH_3, + MOD_HDCP_MESSAGE_ID_READ_VH_4, + MOD_HDCP_MESSAGE_ID_READ_BCAPS, + MOD_HDCP_MESSAGE_ID_READ_BSTATUS, + MOD_HDCP_MESSAGE_ID_READ_KSV_FIFO, + MOD_HDCP_MESSAGE_ID_READ_BINFO, + + /* HDCP 2.2 */ + + MOD_HDCP_MESSAGE_ID_HDCP2VERSION, + MOD_HDCP_MESSAGE_ID_RX_CAPS, + MOD_HDCP_MESSAGE_ID_WRITE_AKE_INIT, + MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_CERT, + MOD_HDCP_MESSAGE_ID_WRITE_AKE_NO_STORED_KM, + MOD_HDCP_MESSAGE_ID_WRITE_AKE_STORED_KM, + MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME, + MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO, + MOD_HDCP_MESSAGE_ID_WRITE_LC_INIT, + MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME, + MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS, + MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, + MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK, + MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE, + MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY, + MOD_HDCP_MESSAGE_ID_READ_RXSTATUS, + MOD_HDCP_MESSAGE_ID_WRITE_CONTENT_STREAM_TYPE, + + MOD_HDCP_MESSAGE_ID_MAX +}; + +static const uint8_t hdcp_i2c_offsets[] = { + [MOD_HDCP_MESSAGE_ID_READ_BKSV] = 0x0, + [MOD_HDCP_MESSAGE_ID_READ_RI_R0] = 0x8, + [MOD_HDCP_MESSAGE_ID_WRITE_AKSV] = 0x10, + [MOD_HDCP_MESSAGE_ID_WRITE_AINFO] = 0x15, + [MOD_HDCP_MESSAGE_ID_WRITE_AN] = 0x18, + [MOD_HDCP_MESSAGE_ID_READ_VH_X] = 0x20, + [MOD_HDCP_MESSAGE_ID_READ_VH_0] = 0x20, + [MOD_HDCP_MESSAGE_ID_READ_VH_1] = 0x24, + [MOD_HDCP_MESSAGE_ID_READ_VH_2] = 0x28, + [MOD_HDCP_MESSAGE_ID_READ_VH_3] = 0x2C, + [MOD_HDCP_MESSAGE_ID_READ_VH_4] = 0x30, + [MOD_HDCP_MESSAGE_ID_READ_BCAPS] = 0x40, + [MOD_HDCP_MESSAGE_ID_READ_BSTATUS] = 0x41, + [MOD_HDCP_MESSAGE_ID_READ_KSV_FIFO] = 0x43, + [MOD_HDCP_MESSAGE_ID_READ_BINFO] = 0xFF, + [MOD_HDCP_MESSAGE_ID_HDCP2VERSION] = 0x50, + [MOD_HDCP_MESSAGE_ID_WRITE_AKE_INIT] = 0x60, + [MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_CERT] = 0x80, + [MOD_HDCP_MESSAGE_ID_WRITE_AKE_NO_STORED_KM] = 0x60, + [MOD_HDCP_MESSAGE_ID_WRITE_AKE_STORED_KM] = 0x60, + [MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME] = 0x80, + [MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO] = 0x80, + [MOD_HDCP_MESSAGE_ID_WRITE_LC_INIT] = 0x60, + [MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME] = 0x80, + [MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS] = 0x60, + [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST] = 0x80, + [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK] = 0x60, + [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE] = 0x60, + [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY] = 0x80, + [MOD_HDCP_MESSAGE_ID_READ_RXSTATUS] = 0x70, + [MOD_HDCP_MESSAGE_ID_WRITE_CONTENT_STREAM_TYPE] = 0x0 +}; + +static const uint32_t hdcp_dpcd_addrs[] = { + [MOD_HDCP_MESSAGE_ID_READ_BKSV] = 0x68000, + [MOD_HDCP_MESSAGE_ID_READ_RI_R0] = 0x68005, + [MOD_HDCP_MESSAGE_ID_WRITE_AKSV] = 0x68007, + [MOD_HDCP_MESSAGE_ID_WRITE_AINFO] = 0x6803B, + [MOD_HDCP_MESSAGE_ID_WRITE_AN] = 0x6800c, + [MOD_HDCP_MESSAGE_ID_READ_VH_X] = 0x68014, + [MOD_HDCP_MESSAGE_ID_READ_VH_0] = 0x68014, + [MOD_HDCP_MESSAGE_ID_READ_VH_1] = 0x68018, + [MOD_HDCP_MESSAGE_ID_READ_VH_2] = 0x6801c, + [MOD_HDCP_MESSAGE_ID_READ_VH_3] = 0x68020, + [MOD_HDCP_MESSAGE_ID_READ_VH_4] = 0x68024, + [MOD_HDCP_MESSAGE_ID_READ_BCAPS] = 0x68028, + [MOD_HDCP_MESSAGE_ID_READ_BSTATUS] = 0x68029, + [MOD_HDCP_MESSAGE_ID_READ_KSV_FIFO] = 0x6802c, + [MOD_HDCP_MESSAGE_ID_READ_BINFO] = 0x6802a, + [MOD_HDCP_MESSAGE_ID_RX_CAPS] = 0x6921d, + [MOD_HDCP_MESSAGE_ID_WRITE_AKE_INIT] = 0x69000, + [MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_CERT] = 0x6900b, + [MOD_HDCP_MESSAGE_ID_WRITE_AKE_NO_STORED_KM] = 0x69220, + [MOD_HDCP_MESSAGE_ID_WRITE_AKE_STORED_KM] = 0x692a0, + [MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME] = 0x692c0, + [MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO] = 0x692e0, + [MOD_HDCP_MESSAGE_ID_WRITE_LC_INIT] = 0x692f0, + [MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME] = 0x692f8, + [MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS] = 0x69318, + [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST] = 0x69330, + [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK] = 0x693e0, + [MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE] = 0x693f0, + [MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY] = 0x69473, + [MOD_HDCP_MESSAGE_ID_READ_RXSTATUS] = 0x69493, + [MOD_HDCP_MESSAGE_ID_WRITE_CONTENT_STREAM_TYPE] = 0x69494 +}; + +static enum mod_hdcp_status read(struct mod_hdcp *hdcp, + enum mod_hdcp_ddc_message_id msg_id, + uint8_t *buf, + uint32_t buf_len) +{ + bool success = true; + uint32_t cur_size = 0; + uint32_t data_offset = 0; + + if (is_dp_hdcp(hdcp)) { + while (buf_len > 0) { + cur_size = MIN(buf_len, HDCP_MAX_AUX_TRANSACTION_SIZE); + success = hdcp->config.ddc.funcs.read_dpcd(hdcp->config.ddc.handle, + hdcp_dpcd_addrs[msg_id] + data_offset, + buf + data_offset, + cur_size); + + if (!success) + break; + + buf_len -= cur_size; + data_offset += cur_size; + } + } else { + success = hdcp->config.ddc.funcs.read_i2c( + hdcp->config.ddc.handle, + HDCP_I2C_ADDR, + hdcp_i2c_offsets[msg_id], + buf, + (uint32_t)buf_len); + } + + return success ? MOD_HDCP_STATUS_SUCCESS : MOD_HDCP_STATUS_DDC_FAILURE; +} + +static enum mod_hdcp_status read_repeatedly(struct mod_hdcp *hdcp, + enum mod_hdcp_ddc_message_id msg_id, + uint8_t *buf, + uint32_t buf_len, + uint8_t read_size) +{ + enum mod_hdcp_status status = MOD_HDCP_STATUS_DDC_FAILURE; + uint32_t cur_size = 0; + uint32_t data_offset = 0; + + while (buf_len > 0) { + cur_size = MIN(buf_len, read_size); + status = read(hdcp, msg_id, buf + data_offset, cur_size); + + if (status != MOD_HDCP_STATUS_SUCCESS) + break; + + buf_len -= cur_size; + data_offset += cur_size; + } + + return status; +} + +static enum mod_hdcp_status write(struct mod_hdcp *hdcp, + enum mod_hdcp_ddc_message_id msg_id, + uint8_t *buf, + uint32_t buf_len) +{ + bool success = true; + uint32_t cur_size = 0; + uint32_t data_offset = 0; + + if (is_dp_hdcp(hdcp)) { + while (buf_len > 0) { + cur_size = MIN(buf_len, HDCP_MAX_AUX_TRANSACTION_SIZE); + success = hdcp->config.ddc.funcs.write_dpcd( + hdcp->config.ddc.handle, + hdcp_dpcd_addrs[msg_id] + data_offset, + buf + data_offset, + cur_size); + + if (!success) + break; + + buf_len -= cur_size; + data_offset += cur_size; + } + } else { + hdcp->buf[0] = hdcp_i2c_offsets[msg_id]; + memmove(&hdcp->buf[1], buf, buf_len); + success = hdcp->config.ddc.funcs.write_i2c( + hdcp->config.ddc.handle, + HDCP_I2C_ADDR, + hdcp->buf, + (uint32_t)(buf_len+1)); + } + + return success ? MOD_HDCP_STATUS_SUCCESS : MOD_HDCP_STATUS_DDC_FAILURE; +} + +enum mod_hdcp_status mod_hdcp_read_bksv(struct mod_hdcp *hdcp) +{ + return read(hdcp, MOD_HDCP_MESSAGE_ID_READ_BKSV, + hdcp->auth.msg.hdcp1.bksv, + sizeof(hdcp->auth.msg.hdcp1.bksv)); +} + +enum mod_hdcp_status mod_hdcp_read_bcaps(struct mod_hdcp *hdcp) +{ + return read(hdcp, MOD_HDCP_MESSAGE_ID_READ_BCAPS, + &hdcp->auth.msg.hdcp1.bcaps, + sizeof(hdcp->auth.msg.hdcp1.bcaps)); +} + +enum mod_hdcp_status mod_hdcp_read_bstatus(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_BSTATUS, + (uint8_t *)&hdcp->auth.msg.hdcp1.bstatus, + 1); + else + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_BSTATUS, + (uint8_t *)&hdcp->auth.msg.hdcp1.bstatus, + sizeof(hdcp->auth.msg.hdcp1.bstatus)); + return status; +} + +enum mod_hdcp_status mod_hdcp_read_r0p(struct mod_hdcp *hdcp) +{ + return read(hdcp, MOD_HDCP_MESSAGE_ID_READ_RI_R0, + (uint8_t *)&hdcp->auth.msg.hdcp1.r0p, + sizeof(hdcp->auth.msg.hdcp1.r0p)); +} + +/* special case, reading repeatedly at the same address, don't use read() */ +enum mod_hdcp_status mod_hdcp_read_ksvlist(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = read_repeatedly(hdcp, MOD_HDCP_MESSAGE_ID_READ_KSV_FIFO, + hdcp->auth.msg.hdcp1.ksvlist, + hdcp->auth.msg.hdcp1.ksvlist_size, + KSV_READ_SIZE); + else + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_KSV_FIFO, + (uint8_t *)&hdcp->auth.msg.hdcp1.ksvlist, + hdcp->auth.msg.hdcp1.ksvlist_size); + return status; +} + +enum mod_hdcp_status mod_hdcp_read_vp(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_VH_0, + &hdcp->auth.msg.hdcp1.vp[0], 4); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_VH_1, + &hdcp->auth.msg.hdcp1.vp[4], 4); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_VH_2, + &hdcp->auth.msg.hdcp1.vp[8], 4); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_VH_3, + &hdcp->auth.msg.hdcp1.vp[12], 4); + if (status != MOD_HDCP_STATUS_SUCCESS) + goto out; + + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_VH_4, + &hdcp->auth.msg.hdcp1.vp[16], 4); +out: + return status; +} + +enum mod_hdcp_status mod_hdcp_read_binfo(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_BINFO, + (uint8_t *)&hdcp->auth.msg.hdcp1.binfo_dp, + sizeof(hdcp->auth.msg.hdcp1.binfo_dp)); + else + status = MOD_HDCP_STATUS_INVALID_OPERATION; + + return status; +} + +enum mod_hdcp_status mod_hdcp_write_aksv(struct mod_hdcp *hdcp) +{ + return write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_AKSV, + hdcp->auth.msg.hdcp1.aksv, + sizeof(hdcp->auth.msg.hdcp1.aksv)); +} + +enum mod_hdcp_status mod_hdcp_write_ainfo(struct mod_hdcp *hdcp) +{ + return write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_AINFO, + &hdcp->auth.msg.hdcp1.ainfo, + sizeof(hdcp->auth.msg.hdcp1.ainfo)); +} + +enum mod_hdcp_status mod_hdcp_write_an(struct mod_hdcp *hdcp) +{ + return write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_AN, + hdcp->auth.msg.hdcp1.an, + sizeof(hdcp->auth.msg.hdcp1.an)); +} + +enum mod_hdcp_status mod_hdcp_read_hdcp2version(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = MOD_HDCP_STATUS_INVALID_OPERATION; + else + status = read(hdcp, MOD_HDCP_MESSAGE_ID_HDCP2VERSION, + &hdcp->auth.msg.hdcp2.hdcp2version_hdmi, + sizeof(hdcp->auth.msg.hdcp2.hdcp2version_hdmi)); + + return status; +} + +enum mod_hdcp_status mod_hdcp_read_rxcaps(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (!is_dp_hdcp(hdcp)) + status = MOD_HDCP_STATUS_INVALID_OPERATION; + else + status = read(hdcp, MOD_HDCP_MESSAGE_ID_RX_CAPS, + hdcp->auth.msg.hdcp2.rxcaps_dp, + sizeof(hdcp->auth.msg.hdcp2.rxcaps_dp)); + + return status; +} + +enum mod_hdcp_status mod_hdcp_read_rxstatus(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) { + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_RXSTATUS, + &hdcp->auth.msg.hdcp2.rxstatus_dp, + 1); + } else { + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_RXSTATUS, + (uint8_t *)&hdcp->auth.msg.hdcp2.rxstatus, + sizeof(hdcp->auth.msg.hdcp2.rxstatus)); + } + return status; +} + +enum mod_hdcp_status mod_hdcp_read_ake_cert(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) { + hdcp->auth.msg.hdcp2.ake_cert[0] = 3; + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_CERT, + hdcp->auth.msg.hdcp2.ake_cert+1, + sizeof(hdcp->auth.msg.hdcp2.ake_cert)-1); + + } else { + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_CERT, + hdcp->auth.msg.hdcp2.ake_cert, + sizeof(hdcp->auth.msg.hdcp2.ake_cert)); + } + return status; +} + +enum mod_hdcp_status mod_hdcp_read_h_prime(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) { + hdcp->auth.msg.hdcp2.ake_h_prime[0] = 7; + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME, + hdcp->auth.msg.hdcp2.ake_h_prime+1, + sizeof(hdcp->auth.msg.hdcp2.ake_h_prime)-1); + + } else { + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_H_PRIME, + hdcp->auth.msg.hdcp2.ake_h_prime, + sizeof(hdcp->auth.msg.hdcp2.ake_h_prime)); + } + return status; +} + +enum mod_hdcp_status mod_hdcp_read_pairing_info(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) { + hdcp->auth.msg.hdcp2.ake_pairing_info[0] = 8; + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO, + hdcp->auth.msg.hdcp2.ake_pairing_info+1, + sizeof(hdcp->auth.msg.hdcp2.ake_pairing_info)-1); + + } else { + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_AKE_SEND_PAIRING_INFO, + hdcp->auth.msg.hdcp2.ake_pairing_info, + sizeof(hdcp->auth.msg.hdcp2.ake_pairing_info)); + } + return status; +} + +enum mod_hdcp_status mod_hdcp_read_l_prime(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) { + hdcp->auth.msg.hdcp2.lc_l_prime[0] = 10; + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME, + hdcp->auth.msg.hdcp2.lc_l_prime+1, + sizeof(hdcp->auth.msg.hdcp2.lc_l_prime)-1); + + } else { + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME, + hdcp->auth.msg.hdcp2.lc_l_prime, + sizeof(hdcp->auth.msg.hdcp2.lc_l_prime)); + } + return status; +} + +enum mod_hdcp_status mod_hdcp_read_rx_id_list(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) { + hdcp->auth.msg.hdcp2.rx_id_list[0] = 12; + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, + hdcp->auth.msg.hdcp2.rx_id_list+1, + sizeof(hdcp->auth.msg.hdcp2.rx_id_list)-1); + + } else { + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, + hdcp->auth.msg.hdcp2.rx_id_list, + hdcp->auth.msg.hdcp2.rx_id_list_size); + } + return status; +} + +enum mod_hdcp_status mod_hdcp_read_stream_ready(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) { + hdcp->auth.msg.hdcp2.repeater_auth_stream_ready[0] = 17; + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY, + hdcp->auth.msg.hdcp2.repeater_auth_stream_ready+1, + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_ready)-1); + + } else { + status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_STREAM_READY, + hdcp->auth.msg.hdcp2.repeater_auth_stream_ready, + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_ready)); + } + return status; +} + +enum mod_hdcp_status mod_hdcp_write_ake_init(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_AKE_INIT, + hdcp->auth.msg.hdcp2.ake_init+1, + sizeof(hdcp->auth.msg.hdcp2.ake_init)-1); + else + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_AKE_INIT, + hdcp->auth.msg.hdcp2.ake_init, + sizeof(hdcp->auth.msg.hdcp2.ake_init)); + return status; +} + +enum mod_hdcp_status mod_hdcp_write_no_stored_km(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_AKE_NO_STORED_KM, + hdcp->auth.msg.hdcp2.ake_no_stored_km+1, + sizeof(hdcp->auth.msg.hdcp2.ake_no_stored_km)-1); + else + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_AKE_NO_STORED_KM, + hdcp->auth.msg.hdcp2.ake_no_stored_km, + sizeof(hdcp->auth.msg.hdcp2.ake_no_stored_km)); + return status; +} + +enum mod_hdcp_status mod_hdcp_write_stored_km(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_AKE_STORED_KM, + hdcp->auth.msg.hdcp2.ake_stored_km+1, + sizeof(hdcp->auth.msg.hdcp2.ake_stored_km)-1); + else + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_AKE_STORED_KM, + hdcp->auth.msg.hdcp2.ake_stored_km, + sizeof(hdcp->auth.msg.hdcp2.ake_stored_km)); + return status; +} + +enum mod_hdcp_status mod_hdcp_write_lc_init(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_LC_INIT, + hdcp->auth.msg.hdcp2.lc_init+1, + sizeof(hdcp->auth.msg.hdcp2.lc_init)-1); + else + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_LC_INIT, + hdcp->auth.msg.hdcp2.lc_init, + sizeof(hdcp->auth.msg.hdcp2.lc_init)); + return status; +} + +enum mod_hdcp_status mod_hdcp_write_eks(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = write(hdcp, + MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS, + hdcp->auth.msg.hdcp2.ske_eks+1, + sizeof(hdcp->auth.msg.hdcp2.ske_eks)-1); + else + status = write(hdcp, + MOD_HDCP_MESSAGE_ID_WRITE_SKE_SEND_EKS, + hdcp->auth.msg.hdcp2.ske_eks, + sizeof(hdcp->auth.msg.hdcp2.ske_eks)); + return status; +} + +enum mod_hdcp_status mod_hdcp_write_repeater_auth_ack(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK, + hdcp->auth.msg.hdcp2.repeater_auth_ack+1, + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_ack)-1); + else + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_SEND_ACK, + hdcp->auth.msg.hdcp2.repeater_auth_ack, + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_ack)); + return status; +} + +enum mod_hdcp_status mod_hdcp_write_stream_manage(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = write(hdcp, + MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE, + hdcp->auth.msg.hdcp2.repeater_auth_stream_manage+1, + hdcp->auth.msg.hdcp2.stream_manage_size-1); + else + status = write(hdcp, + MOD_HDCP_MESSAGE_ID_WRITE_REPEATER_AUTH_STREAM_MANAGE, + hdcp->auth.msg.hdcp2.repeater_auth_stream_manage, + hdcp->auth.msg.hdcp2.stream_manage_size); + return status; +} + +enum mod_hdcp_status mod_hdcp_write_content_type(struct mod_hdcp *hdcp) +{ + enum mod_hdcp_status status; + + if (is_dp_hdcp(hdcp)) + status = write(hdcp, MOD_HDCP_MESSAGE_ID_WRITE_CONTENT_STREAM_TYPE, + hdcp->auth.msg.hdcp2.content_stream_type_dp+1, + sizeof(hdcp->auth.msg.hdcp2.content_stream_type_dp)-1); + else + status = MOD_HDCP_STATUS_INVALID_OPERATION; + return status; +} diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c new file mode 100644 index 000000000000..724ebcee9a19 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c @@ -0,0 +1,281 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + + +#include "hdcp.h" + +void mod_hdcp_dump_binary_message(uint8_t *msg, uint32_t msg_size, + uint8_t *buf, uint32_t buf_size) +{ + const uint8_t bytes_per_line = 16, + byte_size = 3, + newline_size = 1, + terminator_size = 1; + uint32_t line_count = msg_size / bytes_per_line, + trailing_bytes = msg_size % bytes_per_line; + uint32_t target_size = (byte_size * bytes_per_line + newline_size) * line_count + + byte_size * trailing_bytes + newline_size + terminator_size; + uint32_t buf_pos = 0; + uint32_t i = 0; + + if (buf_size >= target_size) { + for (i = 0; i < msg_size; i++) { + if (i % bytes_per_line == 0) + buf[buf_pos++] = '\n'; + sprintf(&buf[buf_pos], "%02X ", msg[i]); + buf_pos += byte_size; + } + buf[buf_pos++] = '\0'; + } +} + +char *mod_hdcp_status_to_str(int32_t status) +{ + switch (status) { + case MOD_HDCP_STATUS_SUCCESS: + return "MOD_HDCP_STATUS_SUCCESS"; + case MOD_HDCP_STATUS_FAILURE: + return "MOD_HDCP_STATUS_FAILURE"; + case MOD_HDCP_STATUS_RESET_NEEDED: + return "MOD_HDCP_STATUS_RESET_NEEDED"; + case MOD_HDCP_STATUS_DISPLAY_OUT_OF_BOUND: + return "MOD_HDCP_STATUS_DISPLAY_OUT_OF_BOUND"; + case MOD_HDCP_STATUS_DISPLAY_NOT_FOUND: + return "MOD_HDCP_STATUS_DISPLAY_NOT_FOUND"; + case MOD_HDCP_STATUS_INVALID_STATE: + return "MOD_HDCP_STATUS_INVALID_STATE"; + case MOD_HDCP_STATUS_NOT_IMPLEMENTED: + return "MOD_HDCP_STATUS_NOT_IMPLEMENTED"; + case MOD_HDCP_STATUS_INTERNAL_POLICY_FAILURE: + return "MOD_HDCP_STATUS_INTERNAL_POLICY_FAILURE"; + case MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE: + return "MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE"; + case MOD_HDCP_STATUS_CREATE_PSP_SERVICE_FAILURE: + return "MOD_HDCP_STATUS_CREATE_PSP_SERVICE_FAILURE"; + case MOD_HDCP_STATUS_DESTROY_PSP_SERVICE_FAILURE: + return "MOD_HDCP_STATUS_DESTROY_PSP_SERVICE_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_CREATE_SESSION_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_CREATE_SESSION_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_DESTROY_SESSION_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_DESTROY_SESSION_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_VALIDATE_ENCRYPTION_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_VALIDATE_ENCRYPTION_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_NOT_HDCP_REPEATER: + return "MOD_HDCP_STATUS_HDCP1_NOT_HDCP_REPEATER"; + case MOD_HDCP_STATUS_HDCP1_NOT_CAPABLE: + return "MOD_HDCP_STATUS_HDCP1_NOT_CAPABLE"; + case MOD_HDCP_STATUS_HDCP1_R0_PRIME_PENDING: + return "MOD_HDCP_STATUS_HDCP1_R0_PRIME_PENDING"; + case MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_KSV_LIST_NOT_READY: + return "MOD_HDCP_STATUS_HDCP1_KSV_LIST_NOT_READY"; + case MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION: + return "MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION"; + case MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_MAX_CASCADE_EXCEEDED_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_MAX_CASCADE_EXCEEDED_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_MAX_DEVS_EXCEEDED_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_MAX_DEVS_EXCEEDED_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_DEVICE_COUNT_MISMATCH_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_DEVICE_COUNT_MISMATCH_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_LINK_INTEGRITY_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_LINK_INTEGRITY_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_REAUTH_REQUEST_ISSUED: + return "MOD_HDCP_STATUS_HDCP1_REAUTH_REQUEST_ISSUED"; + case MOD_HDCP_STATUS_HDCP1_LINK_MAINTENANCE_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_LINK_MAINTENANCE_FAILURE"; + case MOD_HDCP_STATUS_HDCP1_INVALID_BKSV: + return "MOD_HDCP_STATUS_HDCP1_INVALID_BKSV"; + case MOD_HDCP_STATUS_DDC_FAILURE: + return "MOD_HDCP_STATUS_DDC_FAILURE"; + case MOD_HDCP_STATUS_INVALID_OPERATION: + return "MOD_HDCP_STATUS_INVALID_OPERATION"; + case MOD_HDCP_STATUS_HDCP2_NOT_CAPABLE: + return "MOD_HDCP_STATUS_HDCP2_NOT_CAPABLE"; + case MOD_HDCP_STATUS_HDCP2_CREATE_SESSION_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_CREATE_SESSION_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_DESTROY_SESSION_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_DESTROY_SESSION_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_PREP_AKE_INIT_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_PREP_AKE_INIT_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_AKE_CERT_PENDING: + return "MOD_HDCP_STATUS_HDCP2_AKE_CERT_PENDING"; + case MOD_HDCP_STATUS_HDCP2_H_PRIME_PENDING: + return "MOD_HDCP_STATUS_HDCP2_H_PRIME_PENDING"; + case MOD_HDCP_STATUS_HDCP2_PAIRING_INFO_PENDING: + return "MOD_HDCP_STATUS_HDCP2_PAIRING_INFO_PENDING"; + case MOD_HDCP_STATUS_HDCP2_VALIDATE_AKE_CERT_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_VALIDATE_AKE_CERT_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_AKE_CERT_REVOKED: + return "MOD_HDCP_STATUS_HDCP2_AKE_CERT_REVOKED"; + case MOD_HDCP_STATUS_HDCP2_VALIDATE_H_PRIME_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_VALIDATE_H_PRIME_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_VALIDATE_PAIRING_INFO_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_VALIDATE_PAIRING_INFO_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_PREP_LC_INIT_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_PREP_LC_INIT_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_L_PRIME_PENDING: + return "MOD_HDCP_STATUS_HDCP2_L_PRIME_PENDING"; + case MOD_HDCP_STATUS_HDCP2_VALIDATE_L_PRIME_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_VALIDATE_L_PRIME_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_PREP_EKS_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_PREP_EKS_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_ENABLE_ENCRYPTION_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_ENABLE_ENCRYPTION_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_VALIDATE_RX_ID_LIST_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_VALIDATE_RX_ID_LIST_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_REVOKED: + return "MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_REVOKED"; + case MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_NOT_READY: + return "MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_NOT_READY"; + case MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION: + return "MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION"; + case MOD_HDCP_STATUS_HDCP2_STREAM_READY_PENDING: + return "MOD_HDCP_STATUS_HDCP2_STREAM_READY_PENDING"; + case MOD_HDCP_STATUS_HDCP2_VALIDATE_STREAM_READY_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_VALIDATE_STREAM_READY_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_PREPARE_STREAM_MANAGEMENT_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_PREPARE_STREAM_MANAGEMENT_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_REAUTH_REQUEST: + return "MOD_HDCP_STATUS_HDCP2_REAUTH_REQUEST"; + case MOD_HDCP_STATUS_HDCP2_REAUTH_LINK_INTEGRITY_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_REAUTH_LINK_INTEGRITY_FAILURE"; + case MOD_HDCP_STATUS_HDCP2_DEVICE_COUNT_MISMATCH_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_DEVICE_COUNT_MISMATCH_FAILURE"; + default: + return "MOD_HDCP_STATUS_UNKNOWN"; + } +} + +char *mod_hdcp_state_id_to_str(int32_t id) +{ + switch (id) { + case HDCP_UNINITIALIZED: + return "HDCP_UNINITIALIZED"; + case HDCP_INITIALIZED: + return "HDCP_INITIALIZED"; + case HDCP_CP_NOT_DESIRED: + return "HDCP_CP_NOT_DESIRED"; + case H1_A0_WAIT_FOR_ACTIVE_RX: + return "H1_A0_WAIT_FOR_ACTIVE_RX"; + case H1_A1_EXCHANGE_KSVS: + return "H1_A1_EXCHANGE_KSVS"; + case H1_A2_COMPUTATIONS_A3_VALIDATE_RX_A6_TEST_FOR_REPEATER: + return "H1_A2_COMPUTATIONS_A3_VALIDATE_RX_A6_TEST_FOR_REPEATER"; + case H1_A45_AUTHENTICATED: + return "H1_A45_AUTHENTICATED"; + case H1_A8_WAIT_FOR_READY: + return "H1_A8_WAIT_FOR_READY"; + case H1_A9_READ_KSV_LIST: + return "H1_A9_READ_KSV_LIST"; + case D1_A0_DETERMINE_RX_HDCP_CAPABLE: + return "D1_A0_DETERMINE_RX_HDCP_CAPABLE"; + case D1_A1_EXCHANGE_KSVS: + return "D1_A1_EXCHANGE_KSVS"; + case D1_A23_WAIT_FOR_R0_PRIME: + return "D1_A23_WAIT_FOR_R0_PRIME"; + case D1_A2_COMPUTATIONS_A3_VALIDATE_RX_A5_TEST_FOR_REPEATER: + return "D1_A2_COMPUTATIONS_A3_VALIDATE_RX_A5_TEST_FOR_REPEATER"; + case D1_A4_AUTHENTICATED: + return "D1_A4_AUTHENTICATED"; + case D1_A6_WAIT_FOR_READY: + return "D1_A6_WAIT_FOR_READY"; + case D1_A7_READ_KSV_LIST: + return "D1_A7_READ_KSV_LIST"; + case H2_A0_KNOWN_HDCP2_CAPABLE_RX: + return "H2_A0_KNOWN_HDCP2_CAPABLE_RX"; + case H2_A1_SEND_AKE_INIT: + return "H2_A1_SEND_AKE_INIT"; + case H2_A1_VALIDATE_AKE_CERT: + return "H2_A1_VALIDATE_AKE_CERT"; + case H2_A1_SEND_NO_STORED_KM: + return "H2_A1_SEND_NO_STORED_KM"; + case H2_A1_READ_H_PRIME: + return "H2_A1_READ_H_PRIME"; + case H2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME: + return "H2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME"; + case H2_A1_SEND_STORED_KM: + return "H2_A1_SEND_STORED_KM"; + case H2_A1_VALIDATE_H_PRIME: + return "H2_A1_VALIDATE_H_PRIME"; + case H2_A2_LOCALITY_CHECK: + return "H2_A2_LOCALITY_CHECK"; + case H2_A3_EXCHANGE_KS_AND_TEST_FOR_REPEATER: + return "H2_A3_EXCHANGE_KS_AND_TEST_FOR_REPEATER"; + case H2_ENABLE_ENCRYPTION: + return "H2_ENABLE_ENCRYPTION"; + case H2_A5_AUTHENTICATED: + return "H2_A5_AUTHENTICATED"; + case H2_A6_WAIT_FOR_RX_ID_LIST: + return "H2_A6_WAIT_FOR_RX_ID_LIST"; + case H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK: + return "H2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK"; + case H2_A9_SEND_STREAM_MANAGEMENT: + return "H2_A9_SEND_STREAM_MANAGEMENT"; + case H2_A9_VALIDATE_STREAM_READY: + return "H2_A9_VALIDATE_STREAM_READY"; + case D2_A0_DETERMINE_RX_HDCP_CAPABLE: + return "D2_A0_DETERMINE_RX_HDCP_CAPABLE"; + case D2_A1_SEND_AKE_INIT: + return "D2_A1_SEND_AKE_INIT"; + case D2_A1_VALIDATE_AKE_CERT: + return "D2_A1_VALIDATE_AKE_CERT"; + case D2_A1_SEND_NO_STORED_KM: + return "D2_A1_SEND_NO_STORED_KM"; + case D2_A1_READ_H_PRIME: + return "D2_A1_READ_H_PRIME"; + case D2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME: + return "D2_A1_READ_PAIRING_INFO_AND_VALIDATE_H_PRIME"; + case D2_A1_SEND_STORED_KM: + return "D2_A1_SEND_STORED_KM"; + case D2_A1_VALIDATE_H_PRIME: + return "D2_A1_VALIDATE_H_PRIME"; + case D2_A2_LOCALITY_CHECK: + return "D2_A2_LOCALITY_CHECK"; + case D2_A34_EXCHANGE_KS_AND_TEST_FOR_REPEATER: + return "D2_A34_EXCHANGE_KS_AND_TEST_FOR_REPEATER"; + case D2_SEND_CONTENT_STREAM_TYPE: + return "D2_SEND_CONTENT_STREAM_TYPE"; + case D2_ENABLE_ENCRYPTION: + return "D2_ENABLE_ENCRYPTION"; + case D2_A5_AUTHENTICATED: + return "D2_A5_AUTHENTICATED"; + case D2_A6_WAIT_FOR_RX_ID_LIST: + return "D2_A6_WAIT_FOR_RX_ID_LIST"; + case D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK: + return "D2_A78_VERIFY_RX_ID_LIST_AND_SEND_ACK"; + case D2_A9_SEND_STREAM_MANAGEMENT: + return "D2_A9_SEND_STREAM_MANAGEMENT"; + case D2_A9_VALIDATE_STREAM_READY: + return "D2_A9_VALIDATE_STREAM_READY"; + default: + return "UNKNOWN_STATE_ID"; + }; +} + diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h new file mode 100644 index 000000000000..ff91373ebada --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h @@ -0,0 +1,193 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef MOD_HDCP_LOG_H_ +#define MOD_HDCP_LOG_H_ + +#ifdef CONFIG_DRM_AMD_DC_HDCP +#define HDCP_LOG_ERR(hdcp, ...) DRM_WARN(__VA_ARGS__) +#define HDCP_LOG_VER(hdcp, ...) DRM_DEBUG_KMS(__VA_ARGS__) +#define HDCP_LOG_FSM(hdcp, ...) DRM_DEBUG_KMS(__VA_ARGS__) +#define HDCP_LOG_TOP(hdcp, ...) pr_debug("[HDCP_TOP]:"__VA_ARGS__) +#define HDCP_LOG_DDC(hdcp, ...) pr_debug("[HDCP_DDC]:"__VA_ARGS__) +#endif + +/* default logs */ +#define HDCP_ERROR_TRACE(hdcp, status) \ + HDCP_LOG_ERR(hdcp, \ + "[Link %d] WARNING %s IN STATE %s", \ + hdcp->config.index, \ + mod_hdcp_status_to_str(status), \ + mod_hdcp_state_id_to_str(hdcp->state.id)) +#define HDCP_HDCP1_ENABLED_TRACE(hdcp, displayIndex) \ + HDCP_LOG_VER(hdcp, \ + "[Link %d] HDCP 1.4 enabled on display %d", \ + hdcp->config.index, displayIndex) +#define HDCP_HDCP2_ENABLED_TRACE(hdcp, displayIndex) \ + HDCP_LOG_VER(hdcp, \ + "[Link %d] HDCP 2.2 enabled on display %d", \ + hdcp->config.index, displayIndex) +/* state machine logs */ +#define HDCP_REMOVE_DISPLAY_TRACE(hdcp, displayIndex) \ + HDCP_LOG_FSM(hdcp, \ + "[Link %d] HDCP_REMOVE_DISPLAY index %d", \ + hdcp->config.index, displayIndex) +#define HDCP_INPUT_PASS_TRACE(hdcp, str) \ + HDCP_LOG_FSM(hdcp, \ + "[Link %d]\tPASS %s", \ + hdcp->config.index, str) +#define HDCP_INPUT_FAIL_TRACE(hdcp, str) \ + HDCP_LOG_FSM(hdcp, \ + "[Link %d]\tFAIL %s", \ + hdcp->config.index, str) +#define HDCP_NEXT_STATE_TRACE(hdcp, id, output) do { \ + if (output->watchdog_timer_needed) \ + HDCP_LOG_FSM(hdcp, \ + "[Link %d] > %s with %d ms watchdog", \ + hdcp->config.index, \ + mod_hdcp_state_id_to_str(id), output->watchdog_timer_delay); \ + else \ + HDCP_LOG_FSM(hdcp, \ + "[Link %d] > %s", hdcp->config.index, \ + mod_hdcp_state_id_to_str(id)); \ +} while (0) +#define HDCP_TIMEOUT_TRACE(hdcp) \ + HDCP_LOG_FSM(hdcp, "[Link %d] --> TIMEOUT", hdcp->config.index) +#define HDCP_CPIRQ_TRACE(hdcp) \ + HDCP_LOG_FSM(hdcp, "[Link %d] --> CPIRQ", hdcp->config.index) +#define HDCP_EVENT_TRACE(hdcp, event) \ + if (event == MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) \ + HDCP_TIMEOUT_TRACE(hdcp); \ + else if (event == MOD_HDCP_EVENT_CPIRQ) \ + HDCP_CPIRQ_TRACE(hdcp) +/* TODO: find some way to tell if logging is off to save time */ +#define HDCP_DDC_READ_TRACE(hdcp, msg_name, msg, msg_size) do { \ + mod_hdcp_dump_binary_message(msg, msg_size, hdcp->buf, \ + sizeof(hdcp->buf)); \ + HDCP_LOG_DDC(hdcp, "[Link %d] Read %s%s", hdcp->config.index, \ + msg_name, hdcp->buf); \ +} while (0) +#define HDCP_DDC_WRITE_TRACE(hdcp, msg_name, msg, msg_size) do { \ + mod_hdcp_dump_binary_message(msg, msg_size, hdcp->buf, \ + sizeof(hdcp->buf)); \ + HDCP_LOG_DDC(hdcp, "[Link %d] Write %s%s", \ + hdcp->config.index, msg_name,\ + hdcp->buf); \ +} while (0) +#define HDCP_FULL_DDC_TRACE(hdcp) do { \ + if (is_hdcp1(hdcp)) { \ + HDCP_DDC_READ_TRACE(hdcp, "BKSV", hdcp->auth.msg.hdcp1.bksv, \ + sizeof(hdcp->auth.msg.hdcp1.bksv)); \ + HDCP_DDC_READ_TRACE(hdcp, "BCAPS", &hdcp->auth.msg.hdcp1.bcaps, \ + sizeof(hdcp->auth.msg.hdcp1.bcaps)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "AN", hdcp->auth.msg.hdcp1.an, \ + sizeof(hdcp->auth.msg.hdcp1.an)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "AKSV", hdcp->auth.msg.hdcp1.aksv, \ + sizeof(hdcp->auth.msg.hdcp1.aksv)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "AINFO", &hdcp->auth.msg.hdcp1.ainfo, \ + sizeof(hdcp->auth.msg.hdcp1.ainfo)); \ + HDCP_DDC_READ_TRACE(hdcp, "RI' / R0'", \ + (uint8_t *)&hdcp->auth.msg.hdcp1.r0p, \ + sizeof(hdcp->auth.msg.hdcp1.r0p)); \ + HDCP_DDC_READ_TRACE(hdcp, "BINFO", \ + (uint8_t *)&hdcp->auth.msg.hdcp1.binfo_dp, \ + sizeof(hdcp->auth.msg.hdcp1.binfo_dp)); \ + HDCP_DDC_READ_TRACE(hdcp, "KSVLIST", hdcp->auth.msg.hdcp1.ksvlist, \ + hdcp->auth.msg.hdcp1.ksvlist_size); \ + HDCP_DDC_READ_TRACE(hdcp, "V'", hdcp->auth.msg.hdcp1.vp, \ + sizeof(hdcp->auth.msg.hdcp1.vp)); \ + } else { \ + HDCP_DDC_READ_TRACE(hdcp, "HDCP2Version", \ + &hdcp->auth.msg.hdcp2.hdcp2version_hdmi, \ + sizeof(hdcp->auth.msg.hdcp2.hdcp2version_hdmi)); \ + HDCP_DDC_READ_TRACE(hdcp, "Rx Caps", hdcp->auth.msg.hdcp2.rxcaps_dp, \ + sizeof(hdcp->auth.msg.hdcp2.rxcaps_dp)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "AKE Init", hdcp->auth.msg.hdcp2.ake_init, \ + sizeof(hdcp->auth.msg.hdcp2.ake_init)); \ + HDCP_DDC_READ_TRACE(hdcp, "AKE Cert", hdcp->auth.msg.hdcp2.ake_cert, \ + sizeof(hdcp->auth.msg.hdcp2.ake_cert)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "Stored KM", \ + hdcp->auth.msg.hdcp2.ake_stored_km, \ + sizeof(hdcp->auth.msg.hdcp2.ake_stored_km)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "No Stored KM", \ + hdcp->auth.msg.hdcp2.ake_no_stored_km, \ + sizeof(hdcp->auth.msg.hdcp2.ake_no_stored_km)); \ + HDCP_DDC_READ_TRACE(hdcp, "H'", hdcp->auth.msg.hdcp2.ake_h_prime, \ + sizeof(hdcp->auth.msg.hdcp2.ake_h_prime)); \ + HDCP_DDC_READ_TRACE(hdcp, "Pairing Info", \ + hdcp->auth.msg.hdcp2.ake_pairing_info, \ + sizeof(hdcp->auth.msg.hdcp2.ake_pairing_info)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "LC Init", hdcp->auth.msg.hdcp2.lc_init, \ + sizeof(hdcp->auth.msg.hdcp2.lc_init)); \ + HDCP_DDC_READ_TRACE(hdcp, "L'", hdcp->auth.msg.hdcp2.lc_l_prime, \ + sizeof(hdcp->auth.msg.hdcp2.lc_l_prime)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "Exchange KS", hdcp->auth.msg.hdcp2.ske_eks, \ + sizeof(hdcp->auth.msg.hdcp2.ske_eks)); \ + HDCP_DDC_READ_TRACE(hdcp, "Rx Status", \ + (uint8_t *)&hdcp->auth.msg.hdcp2.rxstatus, \ + sizeof(hdcp->auth.msg.hdcp2.rxstatus)); \ + HDCP_DDC_READ_TRACE(hdcp, "Rx Id List", \ + hdcp->auth.msg.hdcp2.rx_id_list, \ + hdcp->auth.msg.hdcp2.rx_id_list_size); \ + HDCP_DDC_WRITE_TRACE(hdcp, "Rx Id List Ack", \ + hdcp->auth.msg.hdcp2.repeater_auth_ack, \ + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_ack)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "Content Stream Management", \ + hdcp->auth.msg.hdcp2.repeater_auth_stream_manage, \ + hdcp->auth.msg.hdcp2.stream_manage_size); \ + HDCP_DDC_READ_TRACE(hdcp, "Stream Ready", \ + hdcp->auth.msg.hdcp2.repeater_auth_stream_ready, \ + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_ready)); \ + HDCP_DDC_WRITE_TRACE(hdcp, "Content Stream Type", \ + hdcp->auth.msg.hdcp2.content_stream_type_dp, \ + sizeof(hdcp->auth.msg.hdcp2.content_stream_type_dp)); \ + } \ +} while (0) +#define HDCP_TOP_ADD_DISPLAY_TRACE(hdcp, i) \ + HDCP_LOG_TOP(hdcp, "[Link %d]\tadd display %d", \ + hdcp->config.index, i) +#define HDCP_TOP_REMOVE_DISPLAY_TRACE(hdcp, i) \ + HDCP_LOG_TOP(hdcp, "[Link %d]\tremove display %d", \ + hdcp->config.index, i) +#define HDCP_TOP_HDCP1_DESTROY_SESSION_TRACE(hdcp) \ + HDCP_LOG_TOP(hdcp, "[Link %d]\tdestroy hdcp1 session", \ + hdcp->config.index) +#define HDCP_TOP_HDCP2_DESTROY_SESSION_TRACE(hdcp) \ + HDCP_LOG_TOP(hdcp, "[Link %d]\tdestroy hdcp2 session", \ + hdcp->config.index) +#define HDCP_TOP_RESET_AUTH_TRACE(hdcp) \ + HDCP_LOG_TOP(hdcp, "[Link %d]\treset authentication", hdcp->config.index) +#define HDCP_TOP_RESET_CONN_TRACE(hdcp) \ + HDCP_LOG_TOP(hdcp, "[Link %d]\treset connection", hdcp->config.index) +#define HDCP_TOP_INTERFACE_TRACE(hdcp) do { \ + HDCP_LOG_TOP(hdcp, "\n"); \ + HDCP_LOG_TOP(hdcp, "[Link %d] %s", hdcp->config.index, __func__); \ +} while (0) +#define HDCP_TOP_INTERFACE_TRACE_WITH_INDEX(hdcp, i) do { \ + HDCP_LOG_TOP(hdcp, "\n"); \ + HDCP_LOG_TOP(hdcp, "[Link %d] %s display %d", hdcp->config.index, __func__, i); \ +} while (0) + +#endif // MOD_HDCP_LOG_H_ diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c new file mode 100644 index 000000000000..7911dc157d5a --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c @@ -0,0 +1,832 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#define MAX_NUM_DISPLAYS 24 + + +#include "hdcp.h" + +#include "amdgpu.h" +#include "hdcp_psp.h" + +static void hdcp2_message_init(struct mod_hdcp *hdcp, + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *in) +{ + in->session_handle = hdcp->auth.id; + in->prepare.msg1_id = TA_HDCP_HDCP2_MSG_ID__NULL_MESSAGE; + in->prepare.msg2_id = TA_HDCP_HDCP2_MSG_ID__NULL_MESSAGE; + in->process.msg1_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__NULL_MESSAGE; + in->process.msg1_desc.msg_size = 0; + in->process.msg2_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__NULL_MESSAGE; + in->process.msg2_desc.msg_size = 0; + in->process.msg3_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__NULL_MESSAGE; + in->process.msg3_desc.msg_size = 0; +} +enum mod_hdcp_status mod_hdcp_remove_display_topology(struct mod_hdcp *hdcp) +{ + + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_dtm_shared_memory *dtm_cmd; + struct mod_hdcp_display *display = NULL; + uint8_t i; + + dtm_cmd = (struct ta_dtm_shared_memory *)psp->dtm_context.dtm_shared_buf; + + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { + if (is_display_added(&(hdcp->connection.displays[i]))) { + + memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); + + display = &hdcp->connection.displays[i]; + + dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_UPDATE_V2; + dtm_cmd->dtm_in_message.topology_update_v2.display_handle = display->index; + dtm_cmd->dtm_in_message.topology_update_v2.is_active = 0; + dtm_cmd->dtm_status = TA_DTM_STATUS__GENERIC_FAILURE; + + psp_dtm_invoke(psp, dtm_cmd->cmd_id); + + if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) + return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; + + display->state = MOD_HDCP_DISPLAY_ACTIVE; + HDCP_TOP_REMOVE_DISPLAY_TRACE(hdcp, display->index); + } + } + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_add_display_topology(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_dtm_shared_memory *dtm_cmd; + struct mod_hdcp_display *display = NULL; + struct mod_hdcp_link *link = &hdcp->connection.link; + uint8_t i; + + if (!psp->dtm_context.dtm_initialized) { + DRM_ERROR("Failed to add display topology, DTM TA is not initialized."); + return MOD_HDCP_STATUS_FAILURE; + } + + dtm_cmd = (struct ta_dtm_shared_memory *)psp->dtm_context.dtm_shared_buf; + + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { + if (hdcp->connection.displays[i].state == MOD_HDCP_DISPLAY_ACTIVE) { + display = &hdcp->connection.displays[i]; + + memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); + + dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_UPDATE_V2; + dtm_cmd->dtm_in_message.topology_update_v2.display_handle = display->index; + dtm_cmd->dtm_in_message.topology_update_v2.is_active = 1; + dtm_cmd->dtm_in_message.topology_update_v2.controller = display->controller; + dtm_cmd->dtm_in_message.topology_update_v2.ddc_line = link->ddc_line; + dtm_cmd->dtm_in_message.topology_update_v2.dig_be = link->dig_be; + dtm_cmd->dtm_in_message.topology_update_v2.dig_fe = display->dig_fe; + dtm_cmd->dtm_in_message.topology_update_v2.dp_mst_vcid = display->vc_id; + dtm_cmd->dtm_in_message.topology_update_v2.max_hdcp_supported_version = + TA_DTM_HDCP_VERSION_MAX_SUPPORTED__2_2; + dtm_cmd->dtm_status = TA_DTM_STATUS__GENERIC_FAILURE; + + psp_dtm_invoke(psp, dtm_cmd->cmd_id); + + if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) + return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; + + display->state = MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; + HDCP_TOP_ADD_DISPLAY_TRACE(hdcp, display->index); + } + } + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp) +{ + + struct psp_context *psp = hdcp->config.psp.handle; + struct mod_hdcp_display *display = get_first_added_display(hdcp); + struct ta_hdcp_shared_memory *hdcp_cmd; + + if (!psp->hdcp_context.hdcp_initialized) { + DRM_ERROR("Failed to create hdcp session. HDCP TA is not initialized."); + return MOD_HDCP_STATUS_FAILURE; + } + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->in_msg.hdcp1_create_session.display_handle = display->index; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP1_CREATE_SESSION; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + hdcp->auth.id = hdcp_cmd->out_msg.hdcp1_create_session.session_handle; + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP1_CREATE_SESSION_FAILURE; + + hdcp->auth.msg.hdcp1.ainfo = hdcp_cmd->out_msg.hdcp1_create_session.ainfo_primary; + memcpy(hdcp->auth.msg.hdcp1.aksv, hdcp_cmd->out_msg.hdcp1_create_session.aksv_primary, + sizeof(hdcp->auth.msg.hdcp1.aksv)); + memcpy(hdcp->auth.msg.hdcp1.an, hdcp_cmd->out_msg.hdcp1_create_session.an_primary, + sizeof(hdcp->auth.msg.hdcp1.an)); + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp1_destroy_session(struct mod_hdcp *hdcp) +{ + + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->in_msg.hdcp1_destroy_session.session_handle = hdcp->auth.id; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP1_DESTROY_SESSION; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP1_DESTROY_SESSION_FAILURE; + + HDCP_TOP_HDCP1_DESTROY_SESSION_TRACE(hdcp); + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp1_validate_rx(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->in_msg.hdcp1_first_part_authentication.session_handle = hdcp->auth.id; + + memcpy(hdcp_cmd->in_msg.hdcp1_first_part_authentication.bksv_primary, hdcp->auth.msg.hdcp1.bksv, + TA_HDCP__HDCP1_KSV_SIZE); + + hdcp_cmd->in_msg.hdcp1_first_part_authentication.r0_prime_primary = hdcp->auth.msg.hdcp1.r0p; + hdcp_cmd->in_msg.hdcp1_first_part_authentication.bcaps = hdcp->auth.msg.hdcp1.bcaps; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP1_FIRST_PART_AUTHENTICATION; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE; + + if (hdcp_cmd->out_msg.hdcp1_first_part_authentication.authentication_status == + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_FIRST_PART_COMPLETE) { + /* needs second part of authentication */ + hdcp->connection.is_repeater = 1; + } else if (hdcp_cmd->out_msg.hdcp1_first_part_authentication.authentication_status == + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_AUTHENTICATED) { + hdcp->connection.is_repeater = 0; + } else + return MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE; + + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp1_enable_encryption(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct mod_hdcp_display *display = get_first_added_display(hdcp); + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->in_msg.hdcp1_enable_encryption.session_handle = hdcp->auth.id; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP1_ENABLE_ENCRYPTION; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION; + + if (!is_dp_mst_hdcp(hdcp)) { + display->state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; + HDCP_HDCP1_ENABLED_TRACE(hdcp, display->index); + } + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp1_validate_ksvlist_vp(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->in_msg.hdcp1_second_part_authentication.session_handle = hdcp->auth.id; + + hdcp_cmd->in_msg.hdcp1_second_part_authentication.ksv_list_size = hdcp->auth.msg.hdcp1.ksvlist_size; + memcpy(hdcp_cmd->in_msg.hdcp1_second_part_authentication.ksv_list, hdcp->auth.msg.hdcp1.ksvlist, + hdcp->auth.msg.hdcp1.ksvlist_size); + + memcpy(hdcp_cmd->in_msg.hdcp1_second_part_authentication.v_prime, hdcp->auth.msg.hdcp1.vp, + sizeof(hdcp->auth.msg.hdcp1.vp)); + + hdcp_cmd->in_msg.hdcp1_second_part_authentication.bstatus_binfo = + is_dp_hdcp(hdcp) ? hdcp->auth.msg.hdcp1.binfo_dp : hdcp->auth.msg.hdcp1.bstatus; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP1_SECOND_PART_AUTHENTICATION; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE; + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption(struct mod_hdcp *hdcp) +{ + + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + int i = 0; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { + + if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || + hdcp->connection.displays[i].adjust.disable) + continue; + + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->in_msg.hdcp1_enable_dp_stream_encryption.session_handle = hdcp->auth.id; + hdcp_cmd->in_msg.hdcp1_enable_dp_stream_encryption.display_handle = hdcp->connection.displays[i].index; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP1_ENABLE_DP_STREAM_ENCRYPTION; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE; + + hdcp->connection.displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; + HDCP_HDCP1_ENABLED_TRACE(hdcp, hdcp->connection.displays[i].index); + } + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp1_link_maintenance(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->in_msg.hdcp1_get_encryption_status.session_handle = hdcp->auth.id; + + hdcp_cmd->out_msg.hdcp1_get_encryption_status.protection_level = 0; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP1_GET_ENCRYPTION_STATUS; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP1_LINK_MAINTENANCE_FAILURE; + + return (hdcp_cmd->out_msg.hdcp1_get_encryption_status.protection_level == 1) + ? MOD_HDCP_STATUS_SUCCESS + : MOD_HDCP_STATUS_HDCP1_LINK_MAINTENANCE_FAILURE; +} + +enum mod_hdcp_status mod_hdcp_hdcp1_get_link_encryption_status(struct mod_hdcp *hdcp, + enum mod_hdcp_encryption_status *encryption_status) +{ + *encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + + if (mod_hdcp_hdcp1_link_maintenance(hdcp) != MOD_HDCP_STATUS_SUCCESS) + return MOD_HDCP_STATUS_FAILURE; + + *encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP1_ON; + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_create_session(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct mod_hdcp_display *display = get_first_added_display(hdcp); + + if (!psp->hdcp_context.hdcp_initialized) { + DRM_ERROR("Failed to create hdcp session, HDCP TA is not initialized"); + return MOD_HDCP_STATUS_FAILURE; + } + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + if (!display) + return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; + + hdcp_cmd->in_msg.hdcp2_create_session_v2.display_handle = display->index; + + if (hdcp->connection.link.adjust.hdcp2.force_type == MOD_HDCP_FORCE_TYPE_0) + hdcp_cmd->in_msg.hdcp2_create_session_v2.negotiate_content_type = + TA_HDCP2_CONTENT_TYPE_NEGOTIATION_TYPE__FORCE_TYPE0; + else if (hdcp->connection.link.adjust.hdcp2.force_type == MOD_HDCP_FORCE_TYPE_1) + hdcp_cmd->in_msg.hdcp2_create_session_v2.negotiate_content_type = + TA_HDCP2_CONTENT_TYPE_NEGOTIATION_TYPE__FORCE_TYPE1; + else if (hdcp->connection.link.adjust.hdcp2.force_type == MOD_HDCP_FORCE_TYPE_MAX) + hdcp_cmd->in_msg.hdcp2_create_session_v2.negotiate_content_type = + TA_HDCP2_CONTENT_TYPE_NEGOTIATION_TYPE__MAX_SUPPORTED; + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_CREATE_SESSION_V2; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_CREATE_SESSION_FAILURE; + + hdcp->auth.id = hdcp_cmd->out_msg.hdcp2_create_session_v2.session_handle; + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_destroy_session(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->in_msg.hdcp2_destroy_session.session_handle = hdcp->auth.id; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_DESTROY_SESSION; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_DESTROY_SESSION_FAILURE; + + HDCP_TOP_HDCP2_DESTROY_SESSION_TRACE(hdcp); + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_prepare_ake_init(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + msg_out = &hdcp_cmd->out_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; + msg_in->prepare.msg1_id = TA_HDCP_HDCP2_MSG_ID__AKE_INIT; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_PREP_AKE_INIT_FAILURE; + + memcpy(&hdcp->auth.msg.hdcp2.ake_init[0], &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.ake_init)); + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_validate_ake_cert(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + msg_out = &hdcp_cmd->out_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + msg_in->process.msg1_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__AKE_SEND_CERT; + msg_in->process.msg1_desc.msg_size = TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_CERT; + + memcpy(&msg_in->process.receiver_message[0], hdcp->auth.msg.hdcp2.ake_cert, + sizeof(hdcp->auth.msg.hdcp2.ake_cert)); + + msg_in->prepare.msg1_id = TA_HDCP_HDCP2_MSG_ID__AKE_NO_STORED_KM; + msg_in->prepare.msg2_id = TA_HDCP_HDCP2_MSG_ID__AKE_STORED_KM; + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_VALIDATE_AKE_CERT_FAILURE; + + memcpy(hdcp->auth.msg.hdcp2.ake_no_stored_km, &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.ake_no_stored_km)); + + memcpy(hdcp->auth.msg.hdcp2.ake_stored_km, + &msg_out->prepare.transmitter_message[sizeof(hdcp->auth.msg.hdcp2.ake_no_stored_km)], + sizeof(hdcp->auth.msg.hdcp2.ake_stored_km)); + + if (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) { + hdcp->connection.is_km_stored = msg_out->process.is_km_stored ? 1 : 0; + hdcp->connection.is_repeater = msg_out->process.is_repeater ? 1 : 0; + return MOD_HDCP_STATUS_SUCCESS; + } + + return MOD_HDCP_STATUS_FAILURE; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_validate_h_prime(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + msg_out = &hdcp_cmd->out_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + msg_in->process.msg1_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__AKE_SEND_H_PRIME; + msg_in->process.msg1_desc.msg_size = TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_H_PRIME; + + memcpy(&msg_in->process.receiver_message[0], hdcp->auth.msg.hdcp2.ake_h_prime, + sizeof(hdcp->auth.msg.hdcp2.ake_h_prime)); + + if (!hdcp->connection.is_km_stored) { + msg_in->process.msg2_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__AKE_SEND_PAIRING_INFO; + msg_in->process.msg2_desc.msg_size = TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_PAIRING_INFO; + memcpy(&msg_in->process.receiver_message[sizeof(hdcp->auth.msg.hdcp2.ake_h_prime)], + hdcp->auth.msg.hdcp2.ake_pairing_info, sizeof(hdcp->auth.msg.hdcp2.ake_pairing_info)); + } + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_VALIDATE_H_PRIME_FAILURE; + + if (msg_out->process.msg1_status != TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_VALIDATE_H_PRIME_FAILURE; + else if (!hdcp->connection.is_km_stored && + msg_out->process.msg2_status != TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_VALIDATE_PAIRING_INFO_FAILURE; + + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_prepare_lc_init(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + msg_out = &hdcp_cmd->out_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + msg_in->prepare.msg1_id = TA_HDCP_HDCP2_MSG_ID__LC_INIT; + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_PREP_LC_INIT_FAILURE; + + memcpy(hdcp->auth.msg.hdcp2.lc_init, &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.lc_init)); + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_validate_l_prime(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + msg_out = &hdcp_cmd->out_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + msg_in->process.msg1_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__LC_SEND_L_PRIME; + msg_in->process.msg1_desc.msg_size = TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__LC_SEND_L_PRIME; + + memcpy(&msg_in->process.receiver_message[0], hdcp->auth.msg.hdcp2.lc_l_prime, + sizeof(hdcp->auth.msg.hdcp2.lc_l_prime)); + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_VALIDATE_L_PRIME_FAILURE; + + if (msg_out->process.msg1_status != TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_VALIDATE_L_PRIME_FAILURE; + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_prepare_eks(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + msg_out = &hdcp_cmd->out_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + msg_in->prepare.msg1_id = TA_HDCP_HDCP2_MSG_ID__SKE_SEND_EKS; + + if (is_dp_hdcp(hdcp)) + msg_in->prepare.msg2_id = TA_HDCP_HDCP2_MSG_ID__SIGNAL_CONTENT_STREAM_TYPE_DP; + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_PREP_EKS_FAILURE; + + memcpy(hdcp->auth.msg.hdcp2.ske_eks, &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.ske_eks)); + msg_out->prepare.msg1_desc.msg_size = sizeof(hdcp->auth.msg.hdcp2.ske_eks); + + if (is_dp_hdcp(hdcp)) { + memcpy(hdcp->auth.msg.hdcp2.content_stream_type_dp, + &msg_out->prepare.transmitter_message[sizeof(hdcp->auth.msg.hdcp2.ske_eks)], + sizeof(hdcp->auth.msg.hdcp2.content_stream_type_dp)); + } + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_enable_encryption(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct mod_hdcp_display *display = get_first_added_display(hdcp); + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + if (!display) + return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; + + hdcp_cmd->in_msg.hdcp1_enable_encryption.session_handle = hdcp->auth.id; + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_SET_ENCRYPTION; + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_ENABLE_ENCRYPTION_FAILURE; + + if (!is_dp_mst_hdcp(hdcp)) { + display->state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; + HDCP_HDCP2_ENABLED_TRACE(hdcp, display->index); + } + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_validate_rx_id_list(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + msg_out = &hdcp_cmd->out_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + msg_in->process.msg1_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__REPEATERAUTH_SEND_RECEIVERID_LIST; + msg_in->process.msg1_desc.msg_size = sizeof(hdcp->auth.msg.hdcp2.rx_id_list); + memcpy(&msg_in->process.receiver_message[0], hdcp->auth.msg.hdcp2.rx_id_list, + sizeof(hdcp->auth.msg.hdcp2.rx_id_list)); + + msg_in->prepare.msg1_id = TA_HDCP_HDCP2_MSG_ID__REPEATERAUTH_SEND_ACK; + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_VALIDATE_RX_ID_LIST_FAILURE; + + memcpy(hdcp->auth.msg.hdcp2.repeater_auth_ack, &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_ack)); + + if (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) { + hdcp->connection.is_km_stored = msg_out->process.is_km_stored ? 1 : 0; + hdcp->connection.is_repeater = msg_out->process.is_repeater ? 1 : 0; + return MOD_HDCP_STATUS_SUCCESS; + } + + + return MOD_HDCP_STATUS_HDCP2_VALIDATE_RX_ID_LIST_FAILURE; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + uint8_t i; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { + if (hdcp->connection.displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || + hdcp->connection.displays[i].adjust.disable) + continue; + hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.display_handle = hdcp->connection.displays[i].index; + hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.session_handle = hdcp->auth.id; + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_ENABLE_DP_STREAM_ENCRYPTION; + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + break; + + hdcp->connection.displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; + HDCP_HDCP2_ENABLED_TRACE(hdcp, hdcp->connection.displays[i].index); + } + + return (hdcp_cmd->hdcp_status == TA_HDCP_STATUS__SUCCESS) ? MOD_HDCP_STATUS_SUCCESS + : MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_prepare_stream_management(struct mod_hdcp *hdcp) +{ + + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + msg_out = &hdcp_cmd->out_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + msg_in->prepare.msg1_id = TA_HDCP_HDCP2_MSG_ID__REPEATERAUTH_STREAM_MANAGE; + + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_HDCP2_PREPARE_STREAM_MANAGEMENT_FAILURE; + + hdcp->auth.msg.hdcp2.stream_manage_size = msg_out->prepare.msg1_desc.msg_size; + + memcpy(hdcp->auth.msg.hdcp2.repeater_auth_stream_manage, &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_manage)); + + return MOD_HDCP_STATUS_SUCCESS; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_validate_stream_ready(struct mod_hdcp *hdcp) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + msg_in = &hdcp_cmd->in_msg.hdcp2_prepare_process_authentication_message_v2; + msg_out = &hdcp_cmd->out_msg.hdcp2_prepare_process_authentication_message_v2; + + hdcp2_message_init(hdcp, msg_in); + + msg_in->process.msg1_desc.msg_id = TA_HDCP_HDCP2_MSG_ID__REPEATERAUTH_STREAM_READY; + + msg_in->process.msg1_desc.msg_size = sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_ready); + + memcpy(&msg_in->process.receiver_message[0], hdcp->auth.msg.hdcp2.repeater_auth_stream_ready, + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_ready)); + + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + return (hdcp_cmd->hdcp_status == TA_HDCP_STATUS__SUCCESS) && + (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) + ? MOD_HDCP_STATUS_SUCCESS + : MOD_HDCP_STATUS_HDCP2_VALIDATE_STREAM_READY_FAILURE; +} + +enum mod_hdcp_status mod_hdcp_hdcp2_get_link_encryption_status(struct mod_hdcp *hdcp, + enum mod_hdcp_encryption_status *encryption_status) +{ + struct psp_context *psp = hdcp->config.psp.handle; + struct ta_hdcp_shared_memory *hdcp_cmd; + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + + hdcp_cmd->in_msg.hdcp2_get_encryption_status.session_handle = hdcp->auth.id; + hdcp_cmd->out_msg.hdcp2_get_encryption_status.protection_level = 0; + hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_GET_ENCRYPTION_STATUS; + *encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + + psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); + + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + return MOD_HDCP_STATUS_FAILURE; + + if (hdcp_cmd->out_msg.hdcp2_get_encryption_status.protection_level == 1) { + if (hdcp_cmd->out_msg.hdcp2_get_encryption_status.hdcp2_type == TA_HDCP2_CONTENT_TYPE__TYPE1) + *encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE1_ON; + else + *encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE0_ON; + } + + return MOD_HDCP_STATUS_SUCCESS; +} diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h new file mode 100644 index 000000000000..82a5e997d573 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h @@ -0,0 +1,466 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef MODULES_HDCP_HDCP_PSP_H_ +#define MODULES_HDCP_HDCP_PSP_H_ + +/* + * NOTE: These parameters are a one-to-one copy of the + * parameters required by PSP + */ +enum bgd_security_hdcp_encryption_level { + HDCP_ENCRYPTION_LEVEL__INVALID = 0, + HDCP_ENCRYPTION_LEVEL__OFF, + HDCP_ENCRYPTION_LEVEL__ON +}; + +enum bgd_security_hdcp2_content_type { + HDCP2_CONTENT_TYPE__INVALID = 0, + HDCP2_CONTENT_TYPE__TYPE0, + HDCP2_CONTENT_TYPE__TYPE1 +}; +enum ta_dtm_command { + TA_DTM_COMMAND__UNUSED_1 = 1, + TA_DTM_COMMAND__TOPOLOGY_UPDATE_V2, + TA_DTM_COMMAND__TOPOLOGY_ASSR_ENABLE +}; + +/* DTM related enumerations */ +/**********************************************************/ + +enum ta_dtm_status { + TA_DTM_STATUS__SUCCESS = 0x00, + TA_DTM_STATUS__GENERIC_FAILURE = 0x01, + TA_DTM_STATUS__INVALID_PARAMETER = 0x02, + TA_DTM_STATUS__NULL_POINTER = 0x3 +}; + +/* input/output structures for DTM commands */ +/**********************************************************/ +/** + * Input structures + */ +enum ta_dtm_hdcp_version_max_supported { + TA_DTM_HDCP_VERSION_MAX_SUPPORTED__NONE = 0, + TA_DTM_HDCP_VERSION_MAX_SUPPORTED__1_x = 10, + TA_DTM_HDCP_VERSION_MAX_SUPPORTED__2_0 = 20, + TA_DTM_HDCP_VERSION_MAX_SUPPORTED__2_1 = 21, + TA_DTM_HDCP_VERSION_MAX_SUPPORTED__2_2 = 22, + TA_DTM_HDCP_VERSION_MAX_SUPPORTED__2_3 = 23 +}; + +struct ta_dtm_topology_update_input_v2 { + /* display handle is unique across the driver and is used to identify a display */ + /* for all security interfaces which reference displays such as HDCP */ + uint32_t display_handle; + uint32_t is_active; + uint32_t is_miracast; + uint32_t controller; + uint32_t ddc_line; + uint32_t dig_be; + uint32_t dig_fe; + uint32_t dp_mst_vcid; + uint32_t is_assr; + uint32_t max_hdcp_supported_version; +}; + +struct ta_dtm_topology_assr_enable { + uint32_t display_topology_dig_be_index; +}; + +/** + * Output structures + */ + +/* No output structures yet */ + +union ta_dtm_cmd_input { + struct ta_dtm_topology_update_input_v2 topology_update_v2; + struct ta_dtm_topology_assr_enable topology_assr_enable; +}; + +union ta_dtm_cmd_output { + uint32_t reserved; +}; + +struct ta_dtm_shared_memory { + uint32_t cmd_id; + uint32_t resp_id; + enum ta_dtm_status dtm_status; + uint32_t reserved; + union ta_dtm_cmd_input dtm_in_message; + union ta_dtm_cmd_output dtm_out_message; +}; + +int psp_cmd_submit_buf(struct psp_context *psp, struct amdgpu_firmware_info *ucode, struct psp_gfx_cmd_resp *cmd, + uint64_t fence_mc_addr); + +enum ta_hdcp_command { + TA_HDCP_COMMAND__INITIALIZE, + TA_HDCP_COMMAND__HDCP1_CREATE_SESSION, + TA_HDCP_COMMAND__HDCP1_DESTROY_SESSION, + TA_HDCP_COMMAND__HDCP1_FIRST_PART_AUTHENTICATION, + TA_HDCP_COMMAND__HDCP1_SECOND_PART_AUTHENTICATION, + TA_HDCP_COMMAND__HDCP1_ENABLE_ENCRYPTION, + TA_HDCP_COMMAND__HDCP1_ENABLE_DP_STREAM_ENCRYPTION, + TA_HDCP_COMMAND__HDCP1_GET_ENCRYPTION_STATUS, + TA_HDCP_COMMAND__UNUSED_1, + TA_HDCP_COMMAND__HDCP2_DESTROY_SESSION, + TA_HDCP_COMMAND__UNUSED_2, + TA_HDCP_COMMAND__HDCP2_SET_ENCRYPTION, + TA_HDCP_COMMAND__HDCP2_GET_ENCRYPTION_STATUS, + TA_HDCP_COMMAND__UNUSED_3, + TA_HDCP_COMMAND__HDCP2_CREATE_SESSION_V2, + TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2, + TA_HDCP_COMMAND__HDCP2_ENABLE_DP_STREAM_ENCRYPTION +}; + +enum ta_hdcp2_msg_id { + TA_HDCP_HDCP2_MSG_ID__NULL_MESSAGE = 1, + TA_HDCP_HDCP2_MSG_ID__AKE_INIT = 2, + TA_HDCP_HDCP2_MSG_ID__AKE_SEND_CERT = 3, + TA_HDCP_HDCP2_MSG_ID__AKE_NO_STORED_KM = 4, + TA_HDCP_HDCP2_MSG_ID__AKE_STORED_KM = 5, + TA_HDCP_HDCP2_MSG_ID__AKE_SEND_RRX = 6, + TA_HDCP_HDCP2_MSG_ID__AKE_SEND_H_PRIME = 7, + TA_HDCP_HDCP2_MSG_ID__AKE_SEND_PAIRING_INFO = 8, + TA_HDCP_HDCP2_MSG_ID__LC_INIT = 9, + TA_HDCP_HDCP2_MSG_ID__LC_SEND_L_PRIME = 10, + TA_HDCP_HDCP2_MSG_ID__SKE_SEND_EKS = 11, + TA_HDCP_HDCP2_MSG_ID__REPEATERAUTH_SEND_RECEIVERID_LIST = 12, + TA_HDCP_HDCP2_MSG_ID__RTT_READY = 13, + TA_HDCP_HDCP2_MSG_ID__RTT_CHALLENGE = 14, + TA_HDCP_HDCP2_MSG_ID__REPEATERAUTH_SEND_ACK = 15, + TA_HDCP_HDCP2_MSG_ID__REPEATERAUTH_STREAM_MANAGE = 16, + TA_HDCP_HDCP2_MSG_ID__REPEATERAUTH_STREAM_READY = 17, + TA_HDCP_HDCP2_MSG_ID__RECEIVER_AUTH_STATUS = 18, + TA_HDCP_HDCP2_MSG_ID__AKE_TRANSMITTER_INFO = 19, + TA_HDCP_HDCP2_MSG_ID__AKE_RECEIVER_INFO = 20, + TA_HDCP_HDCP2_MSG_ID__SIGNAL_CONTENT_STREAM_TYPE_DP = 129 +}; + +enum ta_hdcp2_hdcp2_msg_id_max_size { + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__NULL_MESSAGE = 0, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_INIT = 12, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_CERT = 534, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_NO_STORED_KM = 129, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_STORED_KM = 33, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_RRX = 9, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_H_PRIME = 33, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_PAIRING_INFO = 17, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__LC_INIT = 9, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__LC_SEND_L_PRIME = 33, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__SKE_SEND_EKS = 25, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__REPEATERAUTH_SEND_RECEIVERID_LIST = 181, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__RTT_READY = 1, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__RTT_CHALLENGE = 17, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__REPEATERAUTH_SEND_RACK = 17, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__REPEATERAUTH_STREAM_MANAGE = 13, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__REPEATERAUTH_STREAM_READY = 33, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__RECEIVER_AUTH_STATUS = 4, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_TRANSMITTER_INFO = 6, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_RECEIVER_INFO = 6, + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__SIGNAL_CONTENT_STREAM_TYPE_DP = 1 +}; + +/* HDCP related enumerations */ +/**********************************************************/ +#define TA_HDCP__INVALID_SESSION 0xFFFF +#define TA_HDCP__HDCP1_AN_SIZE 8 +#define TA_HDCP__HDCP1_KSV_SIZE 5 +#define TA_HDCP__HDCP1_KSV_LIST_MAX_ENTRIES 127 +#define TA_HDCP__HDCP1_V_PRIME_SIZE 20 +#define TA_HDCP__HDCP2_TX_BUF_MAX_SIZE \ + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_NO_STORED_KM + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_STORED_KM + 6 + +// 64 bits boundaries +#define TA_HDCP__HDCP2_RX_BUF_MAX_SIZE \ + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_CERT + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_RECEIVER_INFO + 4 + +enum ta_hdcp_status { + TA_HDCP_STATUS__SUCCESS = 0x00, + TA_HDCP_STATUS__GENERIC_FAILURE = 0x01, + TA_HDCP_STATUS__NULL_POINTER = 0x02, + TA_HDCP_STATUS__FAILED_ALLOCATING_SESSION = 0x03, + TA_HDCP_STATUS__FAILED_SETUP_TX = 0x04, + TA_HDCP_STATUS__INVALID_PARAMETER = 0x05, + TA_HDCP_STATUS__VHX_ERROR = 0x06, + TA_HDCP_STATUS__SESSION_NOT_CLOSED_PROPERLY = 0x07, + TA_HDCP_STATUS__SRM_FAILURE = 0x08, + TA_HDCP_STATUS__MST_AUTHENTICATED_ALREADY_STARTED = 0x09, + TA_HDCP_STATUS__AKE_SEND_CERT_FAILURE = 0x0A, + TA_HDCP_STATUS__AKE_NO_STORED_KM_FAILURE = 0x0B, + TA_HDCP_STATUS__AKE_SEND_HPRIME_FAILURE = 0x0C, + TA_HDCP_STATUS__LC_SEND_LPRIME_FAILURE = 0x0D, + TA_HDCP_STATUS__SKE_SEND_EKS_FAILURE = 0x0E, + TA_HDCP_STATUS__REPAUTH_SEND_RXIDLIST_FAILURE = 0x0F, + TA_HDCP_STATUS__REPAUTH_STREAM_READY_FAILURE = 0x10, + TA_HDCP_STATUS__ASD_GENERIC_FAILURE = 0x11, + TA_HDCP_STATUS__UNWRAP_SECRET_FAILURE = 0x12, + TA_HDCP_STATUS__ENABLE_ENCR_FAILURE = 0x13, + TA_HDCP_STATUS__DISABLE_ENCR_FAILURE = 0x14, + TA_HDCP_STATUS__NOT_ENOUGH_MEMORY_FAILURE = 0x15, + TA_HDCP_STATUS__UNKNOWN_MESSAGE = 0x16, + TA_HDCP_STATUS__TOO_MANY_STREAM = 0x17 +}; + +enum ta_hdcp_authentication_status { + TA_HDCP_AUTHENTICATION_STATUS__NOT_STARTED = 0x00, + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_FIRST_PART_FAILED = 0x01, + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_FIRST_PART_COMPLETE = 0x02, + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_SECOND_PART_FAILED = 0x03, + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_AUTHENTICATED = 0x04, + TA_HDCP_AUTHENTICATION_STATUS__HDCP22_AUTHENTICATION_PENDING = 0x06, + TA_HDCP_AUTHENTICATION_STATUS__HDCP22_AUTHENTICATION_FAILED = 0x07, + TA_HDCP_AUTHENTICATION_STATUS__HDCP22_AUTHENTICATED = 0x08, + TA_HDCP_AUTHENTICATION_STATUS__HDCP1_KSV_VALIDATION_FAILED = 0x09 +}; + +enum ta_hdcp2_msg_authentication_status { + TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS = 0, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__KM_NOT_AVAILABLE, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__UNUSED, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__INVALID = 100, // everything above does not fail the request + TA_HDCP2_MSG_AUTHENTICATION_STATUS__NOT_ENOUGH_MEMORY, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__NOT_EXPECTED_MSG, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__SIGNATURE_CERTIFICAT_ERROR, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__INCORRECT_HDCP_VERSION, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__UNKNOWN_MESSAGE, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__INVALID_HMAC, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__INVALID_TOPOLOGY, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__INVALID_SEQ_NUM, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__INVALID_SIZE, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__INVALID_LENGTH, + TA_HDCP2_MSG_AUTHENTICATION_STATUS__REAUTH_REQUEST +}; + +enum ta_hdcp_content_type { + TA_HDCP2_CONTENT_TYPE__TYPE0 = 1, + TA_HDCP2_CONTENT_TYPE__TYPE1, +}; + +enum ta_hdcp_content_type_negotiation_type { + TA_HDCP2_CONTENT_TYPE_NEGOTIATION_TYPE__FORCE_TYPE0 = 1, + TA_HDCP2_CONTENT_TYPE_NEGOTIATION_TYPE__FORCE_TYPE1, + TA_HDCP2_CONTENT_TYPE_NEGOTIATION_TYPE__MAX_SUPPORTED +}; + +enum ta_hdcp2_version { + TA_HDCP2_VERSION_UNKNOWN = 0, + TA_HDCP2_VERSION_2_0 = 20, + TA_HDCP2_VERSION_2_1 = 21, + TA_HDCP2_VERSION_2_2 = 22 +}; + +/* input/output structures for HDCP commands */ +/**********************************************************/ +struct ta_hdcp_cmd_hdcp1_create_session_input { + uint8_t display_handle; +}; + +struct ta_hdcp_cmd_hdcp1_create_session_output { + uint32_t session_handle; + uint8_t an_primary[TA_HDCP__HDCP1_AN_SIZE]; + uint8_t aksv_primary[TA_HDCP__HDCP1_KSV_SIZE]; + uint8_t ainfo_primary; + uint8_t an_secondary[TA_HDCP__HDCP1_AN_SIZE]; + uint8_t aksv_secondary[TA_HDCP__HDCP1_KSV_SIZE]; + uint8_t ainfo_secondary; +}; + +struct ta_hdcp_cmd_hdcp1_destroy_session_input { + uint32_t session_handle; +}; + +struct ta_hdcp_cmd_hdcp1_first_part_authentication_input { + uint32_t session_handle; + uint8_t bksv_primary[TA_HDCP__HDCP1_KSV_SIZE]; + uint8_t bksv_secondary[TA_HDCP__HDCP1_KSV_SIZE]; + uint8_t bcaps; + uint16_t r0_prime_primary; + uint16_t r0_prime_secondary; +}; + +struct ta_hdcp_cmd_hdcp1_first_part_authentication_output { + enum ta_hdcp_authentication_status authentication_status; +}; + +struct ta_hdcp_cmd_hdcp1_second_part_authentication_input { + uint32_t session_handle; + uint16_t bstatus_binfo; + uint8_t ksv_list[TA_HDCP__HDCP1_KSV_LIST_MAX_ENTRIES][TA_HDCP__HDCP1_KSV_SIZE]; + uint32_t ksv_list_size; + uint8_t pj_prime; + uint8_t v_prime[TA_HDCP__HDCP1_V_PRIME_SIZE]; +}; + +struct ta_hdcp_cmd_hdcp1_second_part_authentication_output { + enum ta_hdcp_authentication_status authentication_status; +}; + +struct ta_hdcp_cmd_hdcp1_enable_encryption_input { + uint32_t session_handle; +}; + +struct ta_hdcp_cmd_hdcp1_enable_dp_stream_encryption_input { + uint32_t session_handle; + uint32_t display_handle; +}; + +struct ta_hdcp_cmd_hdcp1_get_encryption_status_input { + uint32_t session_handle; +}; + +struct ta_hdcp_cmd_hdcp1_get_encryption_status_output { + uint32_t protection_level; +}; + +struct ta_hdcp_cmd_hdcp2_create_session_input_v2 { + uint32_t display_handle; + enum ta_hdcp_content_type_negotiation_type negotiate_content_type; +}; + +struct ta_hdcp_cmd_hdcp2_create_session_output_v2 { + uint32_t session_handle; +}; + +struct ta_hdcp_cmd_hdcp2_destroy_session_input { + uint32_t session_handle; +}; + +struct ta_hdcp_cmd_hdcp2_authentication_message_v2 { + enum ta_hdcp2_msg_id msg_id; + uint32_t msg_size; +}; + +struct ta_hdcp_cmd_hdcp2_process_authentication_message_input_v2 { + struct ta_hdcp_cmd_hdcp2_authentication_message_v2 msg1_desc; + struct ta_hdcp_cmd_hdcp2_authentication_message_v2 msg2_desc; + struct ta_hdcp_cmd_hdcp2_authentication_message_v2 msg3_desc; + uint8_t receiver_message[TA_HDCP__HDCP2_RX_BUF_MAX_SIZE]; +}; + +struct ta_hdcp_cmd_hdcp2_process_authentication_message_output_v2 { + uint32_t hdcp_version; + uint32_t is_km_stored; + uint32_t is_locality_precompute_support; + uint32_t is_repeater; + enum ta_hdcp2_msg_authentication_status msg1_status; + enum ta_hdcp2_msg_authentication_status msg2_status; + enum ta_hdcp2_msg_authentication_status msg3_status; +}; + +struct ta_hdcp_cmd_hdcp2_prepare_authentication_message_input_v2 { + enum ta_hdcp2_msg_id msg1_id; + enum ta_hdcp2_msg_id msg2_id; +}; + +struct ta_hdcp_cmd_hdcp2_prepare_authentication_message_output_v2 { + enum ta_hdcp2_msg_authentication_status msg1_status; + enum ta_hdcp2_msg_authentication_status msg2_status; + struct ta_hdcp_cmd_hdcp2_authentication_message_v2 msg1_desc; + struct ta_hdcp_cmd_hdcp2_authentication_message_v2 msg2_desc; + uint8_t transmitter_message[TA_HDCP__HDCP2_TX_BUF_MAX_SIZE]; +}; + +struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 { + uint32_t session_handle; + struct ta_hdcp_cmd_hdcp2_process_authentication_message_input_v2 process; + struct ta_hdcp_cmd_hdcp2_prepare_authentication_message_input_v2 prepare; +}; + +struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 { + uint32_t authentication_status; + struct ta_hdcp_cmd_hdcp2_process_authentication_message_output_v2 process; + struct ta_hdcp_cmd_hdcp2_prepare_authentication_message_output_v2 prepare; +}; + +struct ta_hdcp_cmd_hdcp2_set_encryption_input { + uint32_t session_handle; +}; + +struct ta_hdcp_cmd_hdcp2_get_encryption_status_input { + uint32_t session_handle; +}; + +struct ta_hdcp_cmd_hdcp2_get_encryption_status_output { + enum ta_hdcp_content_type hdcp2_type; + uint32_t protection_level; +}; + +struct ta_hdcp_cmd_hdcp2_enable_dp_stream_encryption_input { + uint32_t session_handle; + uint32_t display_handle; +}; + +/**********************************************************/ +/* Common input structure for HDCP callbacks */ +union ta_hdcp_cmd_input { + struct ta_hdcp_cmd_hdcp1_create_session_input hdcp1_create_session; + struct ta_hdcp_cmd_hdcp1_destroy_session_input hdcp1_destroy_session; + struct ta_hdcp_cmd_hdcp1_first_part_authentication_input hdcp1_first_part_authentication; + struct ta_hdcp_cmd_hdcp1_second_part_authentication_input hdcp1_second_part_authentication; + struct ta_hdcp_cmd_hdcp1_enable_encryption_input hdcp1_enable_encryption; + struct ta_hdcp_cmd_hdcp1_enable_dp_stream_encryption_input hdcp1_enable_dp_stream_encryption; + struct ta_hdcp_cmd_hdcp1_get_encryption_status_input hdcp1_get_encryption_status; + struct ta_hdcp_cmd_hdcp2_destroy_session_input hdcp2_destroy_session; + struct ta_hdcp_cmd_hdcp2_set_encryption_input hdcp2_set_encryption; + struct ta_hdcp_cmd_hdcp2_get_encryption_status_input hdcp2_get_encryption_status; + struct ta_hdcp_cmd_hdcp2_create_session_input_v2 hdcp2_create_session_v2; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 + hdcp2_prepare_process_authentication_message_v2; + struct ta_hdcp_cmd_hdcp2_enable_dp_stream_encryption_input hdcp2_enable_dp_stream_encryption; +}; + +/* Common output structure for HDCP callbacks */ +union ta_hdcp_cmd_output { + struct ta_hdcp_cmd_hdcp1_create_session_output hdcp1_create_session; + struct ta_hdcp_cmd_hdcp1_first_part_authentication_output hdcp1_first_part_authentication; + struct ta_hdcp_cmd_hdcp1_second_part_authentication_output hdcp1_second_part_authentication; + struct ta_hdcp_cmd_hdcp1_get_encryption_status_output hdcp1_get_encryption_status; + struct ta_hdcp_cmd_hdcp2_get_encryption_status_output hdcp2_get_encryption_status; + struct ta_hdcp_cmd_hdcp2_create_session_output_v2 hdcp2_create_session_v2; + struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 + hdcp2_prepare_process_authentication_message_v2; +}; +/**********************************************************/ + +struct ta_hdcp_shared_memory { + uint32_t cmd_id; + enum ta_hdcp_status hdcp_status; + uint32_t reserved; + union ta_hdcp_cmd_input in_msg; + union ta_hdcp_cmd_output out_msg; +}; + +enum psp_status { + PSP_STATUS__SUCCESS = 0, + PSP_STATUS__ERROR_INVALID_PARAMS, + PSP_STATUS__ERROR_GENERIC, + PSP_STATUS__ERROR_OUT_OF_MEMORY, + PSP_STATUS__ERROR_UNSUPPORTED_FEATURE +}; + +#endif /* MODULES_HDCP_HDCP_PSP_H_ */ diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h index dcef85994c45..dbe7835aabcf 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h @@ -92,6 +92,7 @@ struct mod_vrr_params_btr { uint32_t inserted_duration_in_us; uint32_t frames_to_insert; uint32_t frame_counter; + uint32_t margin_in_us; }; struct mod_vrr_params_fixed_refresh { @@ -173,4 +174,6 @@ bool mod_freesync_is_valid_range(struct mod_freesync *mod_freesync, uint32_t min_refresh_request_in_uhz, uint32_t max_refresh_request_in_uhz); + + #endif diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h new file mode 100644 index 000000000000..f2a0e1a064da --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h @@ -0,0 +1,298 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef MOD_HDCP_H_ +#define MOD_HDCP_H_ + +#include "os_types.h" +#include "signal_types.h" + +/* Forward Declarations */ +struct mod_hdcp; + +#define MAX_NUM_OF_DISPLAYS 6 +#define MAX_NUM_OF_ATTEMPTS 4 +#define MAX_NUM_OF_ERROR_TRACE 10 + +/* detailed return status */ +enum mod_hdcp_status { + MOD_HDCP_STATUS_SUCCESS = 0, + MOD_HDCP_STATUS_FAILURE, + MOD_HDCP_STATUS_RESET_NEEDED, + MOD_HDCP_STATUS_DISPLAY_OUT_OF_BOUND, + MOD_HDCP_STATUS_DISPLAY_NOT_FOUND, + MOD_HDCP_STATUS_INVALID_STATE, + MOD_HDCP_STATUS_NOT_IMPLEMENTED, + MOD_HDCP_STATUS_INTERNAL_POLICY_FAILURE, + MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE, + MOD_HDCP_STATUS_CREATE_PSP_SERVICE_FAILURE, + MOD_HDCP_STATUS_DESTROY_PSP_SERVICE_FAILURE, + MOD_HDCP_STATUS_HDCP1_CREATE_SESSION_FAILURE, + MOD_HDCP_STATUS_HDCP1_DESTROY_SESSION_FAILURE, + MOD_HDCP_STATUS_HDCP1_VALIDATE_ENCRYPTION_FAILURE, + MOD_HDCP_STATUS_HDCP1_NOT_HDCP_REPEATER, + MOD_HDCP_STATUS_HDCP1_NOT_CAPABLE, + MOD_HDCP_STATUS_HDCP1_R0_PRIME_PENDING, + MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE, + MOD_HDCP_STATUS_HDCP1_KSV_LIST_NOT_READY, + MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE, + MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION, + MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE, + MOD_HDCP_STATUS_HDCP1_MAX_CASCADE_EXCEEDED_FAILURE, + MOD_HDCP_STATUS_HDCP1_MAX_DEVS_EXCEEDED_FAILURE, + MOD_HDCP_STATUS_HDCP1_DEVICE_COUNT_MISMATCH_FAILURE, + MOD_HDCP_STATUS_HDCP1_LINK_INTEGRITY_FAILURE, + MOD_HDCP_STATUS_HDCP1_REAUTH_REQUEST_ISSUED, + MOD_HDCP_STATUS_HDCP1_LINK_MAINTENANCE_FAILURE, + MOD_HDCP_STATUS_HDCP1_INVALID_BKSV, + MOD_HDCP_STATUS_DDC_FAILURE, /* TODO: specific errors */ + MOD_HDCP_STATUS_INVALID_OPERATION, + MOD_HDCP_STATUS_HDCP2_NOT_CAPABLE, + MOD_HDCP_STATUS_HDCP2_CREATE_SESSION_FAILURE, + MOD_HDCP_STATUS_HDCP2_DESTROY_SESSION_FAILURE, + MOD_HDCP_STATUS_HDCP2_PREP_AKE_INIT_FAILURE, + MOD_HDCP_STATUS_HDCP2_AKE_CERT_PENDING, + MOD_HDCP_STATUS_HDCP2_H_PRIME_PENDING, + MOD_HDCP_STATUS_HDCP2_PAIRING_INFO_PENDING, + MOD_HDCP_STATUS_HDCP2_VALIDATE_AKE_CERT_FAILURE, + MOD_HDCP_STATUS_HDCP2_AKE_CERT_REVOKED, + MOD_HDCP_STATUS_HDCP2_VALIDATE_H_PRIME_FAILURE, + MOD_HDCP_STATUS_HDCP2_VALIDATE_PAIRING_INFO_FAILURE, + MOD_HDCP_STATUS_HDCP2_PREP_LC_INIT_FAILURE, + MOD_HDCP_STATUS_HDCP2_L_PRIME_PENDING, + MOD_HDCP_STATUS_HDCP2_VALIDATE_L_PRIME_FAILURE, + MOD_HDCP_STATUS_HDCP2_PREP_EKS_FAILURE, + MOD_HDCP_STATUS_HDCP2_ENABLE_ENCRYPTION_FAILURE, + MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_NOT_READY, + MOD_HDCP_STATUS_HDCP2_VALIDATE_RX_ID_LIST_FAILURE, + MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_REVOKED, + MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION, + MOD_HDCP_STATUS_HDCP2_STREAM_READY_PENDING, + MOD_HDCP_STATUS_HDCP2_VALIDATE_STREAM_READY_FAILURE, + MOD_HDCP_STATUS_HDCP2_PREPARE_STREAM_MANAGEMENT_FAILURE, + MOD_HDCP_STATUS_HDCP2_REAUTH_REQUEST, + MOD_HDCP_STATUS_HDCP2_REAUTH_LINK_INTEGRITY_FAILURE, + MOD_HDCP_STATUS_HDCP2_DEVICE_COUNT_MISMATCH_FAILURE, +}; + +struct mod_hdcp_displayport { + uint8_t rev; + uint8_t assr_supported; +}; + +struct mod_hdcp_hdmi { + uint8_t reserved; +}; +enum mod_hdcp_operation_mode { + MOD_HDCP_MODE_OFF, + MOD_HDCP_MODE_DEFAULT, + MOD_HDCP_MODE_DP, + MOD_HDCP_MODE_DP_MST +}; + +enum mod_hdcp_display_state { + MOD_HDCP_DISPLAY_INACTIVE = 0, + MOD_HDCP_DISPLAY_ACTIVE, + MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED, + MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED +}; + +struct mod_hdcp_ddc { + void *handle; + struct { + bool (*read_i2c)(void *handle, + uint32_t address, + uint8_t offset, + uint8_t *data, + uint32_t size); + bool (*write_i2c)(void *handle, + uint32_t address, + const uint8_t *data, + uint32_t size); + bool (*read_dpcd)(void *handle, + uint32_t address, + uint8_t *data, + uint32_t size); + bool (*write_dpcd)(void *handle, + uint32_t address, + const uint8_t *data, + uint32_t size); + } funcs; +}; + +struct mod_hdcp_psp { + void *handle; + void *funcs; +}; + +struct mod_hdcp_display_adjustment { + uint8_t disable : 1; + uint8_t reserved : 7; +}; + +struct mod_hdcp_link_adjustment_hdcp1 { + uint8_t disable : 1; + uint8_t postpone_encryption : 1; + uint8_t reserved : 6; +}; + +enum mod_hdcp_force_hdcp_type { + MOD_HDCP_FORCE_TYPE_MAX = 0, + MOD_HDCP_FORCE_TYPE_0, + MOD_HDCP_FORCE_TYPE_1 +}; + +struct mod_hdcp_link_adjustment_hdcp2 { + uint8_t disable : 1; + uint8_t force_type : 2; + uint8_t force_no_stored_km : 1; + uint8_t increase_h_prime_timeout: 1; + uint8_t reserved : 3; +}; + +struct mod_hdcp_link_adjustment { + uint8_t auth_delay; + struct mod_hdcp_link_adjustment_hdcp1 hdcp1; + struct mod_hdcp_link_adjustment_hdcp2 hdcp2; +}; + +struct mod_hdcp_error { + enum mod_hdcp_status status; + uint8_t state_id; +}; + +struct mod_hdcp_trace { + struct mod_hdcp_error errors[MAX_NUM_OF_ERROR_TRACE]; + uint8_t error_count; +}; + +enum mod_hdcp_encryption_status { + MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF = 0, + MOD_HDCP_ENCRYPTION_STATUS_HDCP1_ON, + MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE0_ON, + MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE1_ON, + MOD_HDCP_ENCRYPTION_STATUS_HDCP2_ON +}; + +/* per link events dm has to notify to hdcp module */ +enum mod_hdcp_event { + MOD_HDCP_EVENT_CALLBACK = 0, + MOD_HDCP_EVENT_WATCHDOG_TIMEOUT, + MOD_HDCP_EVENT_CPIRQ +}; + +/* output flags from module requesting timer operations */ +struct mod_hdcp_output { + uint8_t callback_needed; + uint8_t callback_stop; + uint8_t watchdog_timer_needed; + uint8_t watchdog_timer_stop; + uint16_t callback_delay; + uint16_t watchdog_timer_delay; +}; + +/* used to represent per display info */ +struct mod_hdcp_display { + enum mod_hdcp_display_state state; + uint8_t index; + uint8_t controller; + uint8_t dig_fe; + union { + uint8_t vc_id; + }; + struct mod_hdcp_display_adjustment adjust; +}; + +/* used to represent per link info */ +/* in case a link has multiple displays, they share the same link info */ +struct mod_hdcp_link { + enum mod_hdcp_operation_mode mode; + uint8_t dig_be; + uint8_t ddc_line; + union { + struct mod_hdcp_displayport dp; + struct mod_hdcp_hdmi hdmi; + }; + struct mod_hdcp_link_adjustment adjust; +}; + +/* a query structure for a display's hdcp information */ +struct mod_hdcp_display_query { + const struct mod_hdcp_display *display; + const struct mod_hdcp_link *link; + const struct mod_hdcp_trace *trace; + enum mod_hdcp_encryption_status encryption_status; +}; + +/* contains values per on external display configuration change */ +struct mod_hdcp_config { + struct mod_hdcp_psp psp; + struct mod_hdcp_ddc ddc; + uint8_t index; +}; + +struct mod_hdcp; + +/* dm allocates memory of mod_hdcp per dc_link on dm init based on memory size*/ +size_t mod_hdcp_get_memory_size(void); + +/* called per link on link creation */ +enum mod_hdcp_status mod_hdcp_setup(struct mod_hdcp *hdcp, + struct mod_hdcp_config *config); + +/* called per link on link destroy */ +enum mod_hdcp_status mod_hdcp_teardown(struct mod_hdcp *hdcp); + +/* called per display on cp_desired set to true */ +enum mod_hdcp_status mod_hdcp_add_display(struct mod_hdcp *hdcp, + struct mod_hdcp_link *link, struct mod_hdcp_display *display, + struct mod_hdcp_output *output); + +/* called per display on cp_desired set to false */ +enum mod_hdcp_status mod_hdcp_remove_display(struct mod_hdcp *hdcp, + uint8_t index, struct mod_hdcp_output *output); + +/* called to query hdcp information on a specific index */ +enum mod_hdcp_status mod_hdcp_query_display(struct mod_hdcp *hdcp, + uint8_t index, struct mod_hdcp_display_query *query); + +/* called per link on connectivity change */ +enum mod_hdcp_status mod_hdcp_reset_connection(struct mod_hdcp *hdcp, + struct mod_hdcp_output *output); + +/* called per link on events (i.e. callback, watchdog, CP_IRQ) */ +enum mod_hdcp_status mod_hdcp_process_event(struct mod_hdcp *hdcp, + enum mod_hdcp_event event, struct mod_hdcp_output *output); + +/* called to convert enum mod_hdcp_status to c string */ +char *mod_hdcp_status_to_str(int32_t status); + +/* called to convert state id to c string */ +char *mod_hdcp_state_id_to_str(int32_t id); + +/* called to convert signal type to operation mode */ +enum mod_hdcp_operation_mode mod_hdcp_signal_type_to_operation_mode( + enum signal_type signal); +#endif /* MOD_HDCP_H_ */ diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h index 5b1c9a4c7643..42cbeffac640 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h @@ -26,13 +26,18 @@ #ifndef MOD_INFO_PACKET_H_ #define MOD_INFO_PACKET_H_ +#include "dm_services.h" #include "mod_shared.h" - //Forward Declarations struct dc_stream_state; struct dc_info_packet; +struct mod_vrr_params; void mod_build_vsc_infopacket(const struct dc_stream_state *stream, - struct dc_info_packet *info_packet); + struct dc_info_packet *info_packet, + bool *use_vsc_sdp_for_colorimetry); + +void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream, + struct dc_info_packet *info_packet, int ALLMEnabled, int ALLMValue); #endif diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h b/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h index b45f7d65e76a..fe2117904329 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h @@ -45,7 +45,6 @@ enum vrr_packet_type { PACKET_TYPE_VTEM }; -#if defined(CONFIG_DRM_AMD_DC_DCN2_0) union lut3d_control_flags { unsigned int raw; struct { @@ -104,6 +103,5 @@ struct lut3d_settings { enum lut3d_control_gamut_map map2; enum lut3d_control_rotation_mode rotation2; }; -#endif #endif /* MOD_SHARED_H_ */ diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c index bc13c552797f..6a8a056424b8 100644 --- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c +++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c @@ -27,8 +27,92 @@ #include "core_types.h" #include "dc_types.h" #include "mod_shared.h" +#include "mod_freesync.h" +#include "dc.h" + +enum vsc_packet_revision { + vsc_packet_undefined = 0, + //01h = VSC SDP supports only 3D stereo. + vsc_packet_rev1 = 1, + //02h = 3D stereo + PSR. + vsc_packet_rev2 = 2, + //03h = 3D stereo + PSR2. + vsc_packet_rev3 = 3, + //04h = 3D stereo + PSR/PSR2 + Y-coordinate. + vsc_packet_rev4 = 4, + //05h = 3D stereo + PSR/PSR2 + Y-coordinate + Pixel Encoding/Colorimetry Format + vsc_packet_rev5 = 5, +}; #define HDMI_INFOFRAME_TYPE_VENDOR 0x81 +#define HF_VSIF_VERSION 1 + +// VTEM Byte Offset +#define VTEM_PB0 0 +#define VTEM_PB1 1 +#define VTEM_PB2 2 +#define VTEM_PB3 3 +#define VTEM_PB4 4 +#define VTEM_PB5 5 +#define VTEM_PB6 6 + +#define VTEM_MD0 7 +#define VTEM_MD1 8 +#define VTEM_MD2 9 +#define VTEM_MD3 10 + + +// VTEM Byte Masks +//PB0 +#define MASK_VTEM_PB0__RESERVED0 0x01 +#define MASK_VTEM_PB0__SYNC 0x02 +#define MASK_VTEM_PB0__VFR 0x04 +#define MASK_VTEM_PB0__AFR 0x08 +#define MASK_VTEM_PB0__DS_TYPE 0x30 + //0: Periodic pseudo-static EM Data Set + //1: Periodic dynamic EM Data Set + //2: Unique EM Data Set + //3: Reserved +#define MASK_VTEM_PB0__END 0x40 +#define MASK_VTEM_PB0__NEW 0x80 + +//PB1 +#define MASK_VTEM_PB1__RESERVED1 0xFF + +//PB2 +#define MASK_VTEM_PB2__ORGANIZATION_ID 0xFF + //0: This is a Vendor Specific EM Data Set + //1: This EM Data Set is defined by This Specification (HDMI 2.1 r102.clean) + //2: This EM Data Set is defined by CTA-861-G + //3: This EM Data Set is defined by VESA +//PB3 +#define MASK_VTEM_PB3__DATA_SET_TAG_MSB 0xFF +//PB4 +#define MASK_VTEM_PB4__DATA_SET_TAG_LSB 0xFF +//PB5 +#define MASK_VTEM_PB5__DATA_SET_LENGTH_MSB 0xFF +//PB6 +#define MASK_VTEM_PB6__DATA_SET_LENGTH_LSB 0xFF + + + +//PB7-27 (20 bytes): +//PB7 = MD0 +#define MASK_VTEM_MD0__VRR_EN 0x01 +#define MASK_VTEM_MD0__M_CONST 0x02 +#define MASK_VTEM_MD0__RESERVED2 0x0C +#define MASK_VTEM_MD0__FVA_FACTOR_M1 0xF0 + +//MD1 +#define MASK_VTEM_MD1__BASE_VFRONT 0xFF + +//MD2 +#define MASK_VTEM_MD2__BASE_REFRESH_RATE_98 0x03 +#define MASK_VTEM_MD2__RB 0x04 +#define MASK_VTEM_MD2__RESERVED3 0xF8 + +//MD3 +#define MASK_VTEM_MD3__BASE_REFRESH_RATE_07 0xFF enum ColorimetryRGBDP { ColorimetryRGB_DP_sRGB = 0, @@ -46,35 +130,41 @@ enum ColorimetryYCCDP { }; void mod_build_vsc_infopacket(const struct dc_stream_state *stream, - struct dc_info_packet *info_packet) + struct dc_info_packet *info_packet, + bool *use_vsc_sdp_for_colorimetry) { - unsigned int vscPacketRevision = 0; + unsigned int vsc_packet_revision = vsc_packet_undefined; unsigned int i; unsigned int pixelEncoding = 0; unsigned int colorimetryFormat = 0; bool stereo3dSupport = false; + /* Initialize first, later if infopacket is valid determine if VSC SDP + * should be used to signal colorimetry format and pixel encoding. + */ + *use_vsc_sdp_for_colorimetry = false; + if (stream->timing.timing_3d_format != TIMING_3D_FORMAT_NONE && stream->view_format != VIEW_3D_FORMAT_NONE) { - vscPacketRevision = 1; + vsc_packet_revision = vsc_packet_rev1; stereo3dSupport = true; } /*VSC packet set to 2 when DP revision >= 1.2*/ if (stream->psr_version != 0) - vscPacketRevision = 2; + vsc_packet_revision = vsc_packet_rev2; /* Update to revision 5 for extended colorimetry support for DPCD 1.4+ */ if (stream->link->dpcd_caps.dpcd_rev.raw >= 0x14 && stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED) - vscPacketRevision = 5; + vsc_packet_revision = vsc_packet_rev5; /* VSC packet not needed based on the features * supported by this DP display */ - if (vscPacketRevision == 0) + if (vsc_packet_revision == vsc_packet_undefined) return; - if (vscPacketRevision == 0x2) { + if (vsc_packet_revision == vsc_packet_rev2) { /* Secondary-data Packet ID = 0*/ info_packet->hb0 = 0x00; /* 07h - Packet Type Value indicating Video @@ -96,7 +186,7 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, info_packet->valid = true; } - if (vscPacketRevision == 0x1) { + if (vsc_packet_revision == vsc_packet_rev1) { info_packet->hb0 = 0x00; // Secondary-data Packet ID = 0 info_packet->hb1 = 0x07; // 07h = Packet Type Value indicating Video Stream Configuration packet @@ -167,7 +257,7 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, * the Pixel Encoding/Colorimetry Format and that a Sink device must ignore MISC1, bit 7, and * MISC0, bits 7:1 (MISC1, bit 7. and MISC0, bits 7:1 become "don't care").) */ - if (vscPacketRevision == 0x5) { + if (vsc_packet_revision == vsc_packet_rev5) { /* Secondary-data Packet ID = 0 */ info_packet->hb0 = 0x00; /* 07h - Packet Type Value indicating Video Stream Configuration packet */ @@ -179,6 +269,13 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, info_packet->valid = true; + /* If we are using VSC SDP revision 05h, use this to signal for + * colorimetry format and pixel encoding. HW should later be + * programmed to set MSA MISC1 bit 6 to indicate ignore + * colorimetry format and pixel encoding in the MSA. + */ + *use_vsc_sdp_for_colorimetry = true; + /* Set VSC SDP fields for pixel encoding and colorimetry format from DP 1.3 specs * Data Bytes DB 18~16 * Bits 3:0 (Colorimetry Format) | Bits 7:4 (Pixel Encoding) @@ -323,6 +420,102 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, */ info_packet->sb[18] = 0; } +} + +/** + ***************************************************************************** + * Function: mod_build_hf_vsif_infopacket + * + * @brief + * Prepare HDMI Vendor Specific info frame. + * Follows HDMI Spec to build up Vendor Specific info frame + * + * @param [in] stream: contains data we may need to construct VSIF (i.e. timing_3d_format, etc.) + * @param [out] info_packet: output structure where to store VSIF + ***************************************************************************** + */ +void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream, + struct dc_info_packet *info_packet, int ALLMEnabled, int ALLMValue) +{ + unsigned int length = 5; + bool hdmi_vic_mode = false; + uint8_t checksum = 0; + uint32_t i = 0; + enum dc_timing_3d_format format; + bool bALLM = (bool)ALLMEnabled; + bool bALLMVal = (bool)ALLMValue; + + info_packet->valid = false; + format = stream->timing.timing_3d_format; + if (stream->view_format == VIEW_3D_FORMAT_NONE) + format = TIMING_3D_FORMAT_NONE; + + if (stream->timing.hdmi_vic != 0 + && stream->timing.h_total >= 3840 + && stream->timing.v_total >= 2160 + && format == TIMING_3D_FORMAT_NONE) + hdmi_vic_mode = true; + + if ((format == TIMING_3D_FORMAT_NONE) && !hdmi_vic_mode && !bALLM) + return; + + info_packet->sb[1] = 0x03; + info_packet->sb[2] = 0x0C; + info_packet->sb[3] = 0x00; + if (bALLM) { + info_packet->sb[1] = 0xD8; + info_packet->sb[2] = 0x5D; + info_packet->sb[3] = 0xC4; + info_packet->sb[4] = HF_VSIF_VERSION; + } + + if (format != TIMING_3D_FORMAT_NONE) + info_packet->sb[4] = (2 << 5); + + else if (hdmi_vic_mode) + info_packet->sb[4] = (1 << 5); + + switch (format) { + case TIMING_3D_FORMAT_HW_FRAME_PACKING: + case TIMING_3D_FORMAT_SW_FRAME_PACKING: + info_packet->sb[5] = (0x0 << 4); + break; + + case TIMING_3D_FORMAT_SIDE_BY_SIDE: + case TIMING_3D_FORMAT_SBS_SW_PACKED: + info_packet->sb[5] = (0x8 << 4); + length = 6; + break; + + case TIMING_3D_FORMAT_TOP_AND_BOTTOM: + case TIMING_3D_FORMAT_TB_SW_PACKED: + info_packet->sb[5] = (0x6 << 4); + break; + + default: + break; + } + + if (hdmi_vic_mode) + info_packet->sb[5] = stream->timing.hdmi_vic; + + info_packet->hb0 = HDMI_INFOFRAME_TYPE_VENDOR; + info_packet->hb1 = 0x01; + info_packet->hb2 = (uint8_t) (length); + + if (bALLM) + info_packet->sb[5] = (info_packet->sb[5] & ~0x02) | (bALLMVal << 1); + + checksum += info_packet->hb0; + checksum += info_packet->hb1; + checksum += info_packet->hb2; + + for (i = 1; i <= length; i++) + checksum += info_packet->sb[i]; + + info_packet->sb[0] = (uint8_t) (0x100 - checksum); + + info_packet->valid = true; } diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c index b3810b864676..e75a4bb94488 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c @@ -66,6 +66,39 @@ static const unsigned char abm_config[abm_defines_max_config][abm_defines_max_le { 3, 6, 10, 12 }, /* Alt #3 - Super aggressiveness */ }; +struct abm_parameters { + unsigned char min_reduction; + unsigned char max_reduction; + unsigned char bright_pos_gain; + unsigned char dark_pos_gain; + unsigned char brightness_gain; + unsigned char contrast_factor; + unsigned char deviation_gain; + unsigned char min_knee; + unsigned char max_knee; +}; + +static const struct abm_parameters abm_settings_config0[abm_defines_max_level] = { +// min_red max_red bright_pos dark_pos brightness_gain contrast deviation min_knee max_knee + {0xff, 0xbf, 0x20, 0x00, 0xff, 0x99, 0xb3, 0x40, 0xe0}, + {0xde, 0x85, 0x20, 0x00, 0xff, 0x90, 0xa8, 0x40, 0xdf}, + {0xb0, 0x50, 0x20, 0x00, 0xc0, 0x88, 0x78, 0x70, 0xa0}, + {0x82, 0x40, 0x20, 0x00, 0x00, 0xff, 0xb3, 0x70, 0x70}, +}; + +static const struct abm_parameters abm_settings_config1[abm_defines_max_level] = { +// min_red max_red bright_pos dark_pos brightness_gain contrast deviation min_knee max_knee + {0xf0, 0xd9, 0x20, 0x00, 0x00, 0xff, 0xb3, 0x70, 0x70}, + {0xcd, 0xa5, 0x20, 0x00, 0x00, 0xff, 0xb3, 0x70, 0x70}, + {0x99, 0x65, 0x20, 0x00, 0x00, 0xff, 0xb3, 0x70, 0x70}, + {0x82, 0x4d, 0x20, 0x00, 0x00, 0xff, 0xb3, 0x70, 0x70}, +}; + +static const struct abm_parameters * const abm_settings[] = { + abm_settings_config0, + abm_settings_config1, +}; + #define NUM_AMBI_LEVEL 5 #define NUM_AGGR_LEVEL 4 #define NUM_POWER_FN_SEGS 8 @@ -82,7 +115,7 @@ static const unsigned char abm_config[abm_defines_max_config][abm_defines_max_le /* NOTE: iRAM is 256B in size */ struct iram_table_v_2 { /* flags */ - uint16_t flags; /* 0x00 U16 */ + uint16_t min_abm_backlight; /* 0x00 U16 */ /* parameters for ABM2.0 algorithm */ uint8_t min_reduction[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x02 U0.8 */ @@ -107,10 +140,10 @@ struct iram_table_v_2 { /* For reading PSR State directly from IRAM */ uint8_t psr_state; /* 0xf0 */ - uint8_t dmcu_mcp_interface_version; /* 0xf1 */ - uint8_t dmcu_abm_feature_version; /* 0xf2 */ - uint8_t dmcu_psr_feature_version; /* 0xf3 */ - uint16_t dmcu_version; /* 0xf4 */ + uint8_t dmcu_mcp_interface_version; /* 0xf1 */ + uint8_t dmcu_abm_feature_version; /* 0xf2 */ + uint8_t dmcu_psr_feature_version; /* 0xf3 */ + uint16_t dmcu_version; /* 0xf4 */ uint8_t dmcu_state; /* 0xf6 */ uint16_t blRampReduction; /* 0xf7 */ @@ -131,40 +164,43 @@ struct iram_table_v_2_2 { uint8_t max_reduction[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x16 U0.8 */ uint8_t bright_pos_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x2a U2.6 */ uint8_t dark_pos_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x3e U2.6 */ - uint8_t hybridFactor[NUM_AGGR_LEVEL]; /* 0x52 U0.8 */ - uint8_t contrastFactor[NUM_AGGR_LEVEL]; /* 0x56 U0.8 */ - uint8_t deviation_gain[NUM_AGGR_LEVEL]; /* 0x5a U0.8 */ - uint8_t iir_curve[NUM_AMBI_LEVEL]; /* 0x5e U0.8 */ - uint8_t pad[29]; /* 0x63 U0.8 */ + uint8_t hybrid_factor[NUM_AGGR_LEVEL]; /* 0x52 U0.8 */ + uint8_t contrast_factor[NUM_AGGR_LEVEL]; /* 0x56 U0.8 */ + uint8_t deviation_gain[NUM_AGGR_LEVEL]; /* 0x5a U0.8 */ + uint8_t iir_curve[NUM_AMBI_LEVEL]; /* 0x5e U0.8 */ + uint8_t min_knee[NUM_AGGR_LEVEL]; /* 0x63 U0.8 */ + uint8_t max_knee[NUM_AGGR_LEVEL]; /* 0x67 U0.8 */ + uint16_t min_abm_backlight; /* 0x6b U16 */ + uint8_t pad[19]; /* 0x6d U0.8 */ /* parameters for crgb conversion */ - uint16_t crgb_thresh[NUM_POWER_FN_SEGS]; /* 0x80 U3.13 */ - uint16_t crgb_offset[NUM_POWER_FN_SEGS]; /* 0x90 U1.15 */ - uint16_t crgb_slope[NUM_POWER_FN_SEGS]; /* 0xa0 U4.12 */ + uint16_t crgb_thresh[NUM_POWER_FN_SEGS]; /* 0x80 U3.13 */ + uint16_t crgb_offset[NUM_POWER_FN_SEGS]; /* 0x90 U1.15 */ + uint16_t crgb_slope[NUM_POWER_FN_SEGS]; /* 0xa0 U4.12 */ /* parameters for custom curve */ /* thresholds for brightness --> backlight */ - uint16_t backlight_thresholds[NUM_BL_CURVE_SEGS]; /* 0xb0 U16.0 */ + uint16_t backlight_thresholds[NUM_BL_CURVE_SEGS]; /* 0xb0 U16.0 */ /* offsets for brightness --> backlight */ - uint16_t backlight_offsets[NUM_BL_CURVE_SEGS]; /* 0xd0 U16.0 */ + uint16_t backlight_offsets[NUM_BL_CURVE_SEGS]; /* 0xd0 U16.0 */ /* For reading PSR State directly from IRAM */ - uint8_t psr_state; /* 0xf0 */ - uint8_t dmcu_mcp_interface_version; /* 0xf1 */ - uint8_t dmcu_abm_feature_version; /* 0xf2 */ - uint8_t dmcu_psr_feature_version; /* 0xf3 */ - uint16_t dmcu_version; /* 0xf4 */ - uint8_t dmcu_state; /* 0xf6 */ - - uint8_t dummy1; /* 0xf7 */ - uint8_t dummy2; /* 0xf8 */ - uint8_t dummy3; /* 0xf9 */ - uint8_t dummy4; /* 0xfa */ - uint8_t dummy5; /* 0xfb */ - uint8_t dummy6; /* 0xfc */ - uint8_t dummy7; /* 0xfd */ - uint8_t dummy8; /* 0xfe */ - uint8_t dummy9; /* 0xff */ + uint8_t psr_state; /* 0xf0 */ + uint8_t dmcu_mcp_interface_version; /* 0xf1 */ + uint8_t dmcu_abm_feature_version; /* 0xf2 */ + uint8_t dmcu_psr_feature_version; /* 0xf3 */ + uint16_t dmcu_version; /* 0xf4 */ + uint8_t dmcu_state; /* 0xf6 */ + + uint8_t dummy1; /* 0xf7 */ + uint8_t dummy2; /* 0xf8 */ + uint8_t dummy3; /* 0xf9 */ + uint8_t dummy4; /* 0xfa */ + uint8_t dummy5; /* 0xfb */ + uint8_t dummy6; /* 0xfc */ + uint8_t dummy7; /* 0xfd */ + uint8_t dummy8; /* 0xfe */ + uint8_t dummy9; /* 0xff */ }; #pragma pack(pop) @@ -236,7 +272,8 @@ void fill_iram_v_2(struct iram_table_v_2 *ram_table, struct dmcu_iram_parameters { unsigned int set = params.set; - ram_table->flags = 0x0; + ram_table->min_abm_backlight = + cpu_to_be16(params.min_abm_backlight); ram_table->deviation_gain = 0xb3; ram_table->blRampReduction = @@ -410,6 +447,9 @@ void fill_iram_v_2_2(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parame ram_table->flags = 0x0; + ram_table->min_abm_backlight = + cpu_to_be16(params.min_abm_backlight); + ram_table->deviation_gain[0] = 0xb3; ram_table->deviation_gain[1] = 0xa8; ram_table->deviation_gain[2] = 0x98; @@ -501,15 +541,76 @@ void fill_iram_v_2_2(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parame ram_table->dark_pos_gain[4][2] = 0x00; ram_table->dark_pos_gain[4][3] = 0x00; - ram_table->hybridFactor[0] = 0xff; - ram_table->hybridFactor[1] = 0xff; - ram_table->hybridFactor[2] = 0xff; - ram_table->hybridFactor[3] = 0xc0; + ram_table->hybrid_factor[0] = 0xff; + ram_table->hybrid_factor[1] = 0xff; + ram_table->hybrid_factor[2] = 0xff; + ram_table->hybrid_factor[3] = 0xc0; - ram_table->contrastFactor[0] = 0x99; - ram_table->contrastFactor[1] = 0x99; - ram_table->contrastFactor[2] = 0x90; - ram_table->contrastFactor[3] = 0x80; + ram_table->contrast_factor[0] = 0x99; + ram_table->contrast_factor[1] = 0x99; + ram_table->contrast_factor[2] = 0x90; + ram_table->contrast_factor[3] = 0x80; + + ram_table->iir_curve[0] = 0x65; + ram_table->iir_curve[1] = 0x65; + ram_table->iir_curve[2] = 0x65; + ram_table->iir_curve[3] = 0x65; + ram_table->iir_curve[4] = 0x65; + + //Gamma 2.2 + ram_table->crgb_thresh[0] = cpu_to_be16(0x127c); + ram_table->crgb_thresh[1] = cpu_to_be16(0x151b); + ram_table->crgb_thresh[2] = cpu_to_be16(0x17d5); + ram_table->crgb_thresh[3] = cpu_to_be16(0x1a56); + ram_table->crgb_thresh[4] = cpu_to_be16(0x1c83); + ram_table->crgb_thresh[5] = cpu_to_be16(0x1e72); + ram_table->crgb_thresh[6] = cpu_to_be16(0x20f0); + ram_table->crgb_thresh[7] = cpu_to_be16(0x232b); + ram_table->crgb_offset[0] = cpu_to_be16(0x2999); + ram_table->crgb_offset[1] = cpu_to_be16(0x3999); + ram_table->crgb_offset[2] = cpu_to_be16(0x4666); + ram_table->crgb_offset[3] = cpu_to_be16(0x5999); + ram_table->crgb_offset[4] = cpu_to_be16(0x6333); + ram_table->crgb_offset[5] = cpu_to_be16(0x7800); + ram_table->crgb_offset[6] = cpu_to_be16(0x8c00); + ram_table->crgb_offset[7] = cpu_to_be16(0xa000); + ram_table->crgb_slope[0] = cpu_to_be16(0x3609); + ram_table->crgb_slope[1] = cpu_to_be16(0x2dfa); + ram_table->crgb_slope[2] = cpu_to_be16(0x27ea); + ram_table->crgb_slope[3] = cpu_to_be16(0x235d); + ram_table->crgb_slope[4] = cpu_to_be16(0x2042); + ram_table->crgb_slope[5] = cpu_to_be16(0x1dc3); + ram_table->crgb_slope[6] = cpu_to_be16(0x1b1a); + ram_table->crgb_slope[7] = cpu_to_be16(0x1910); + + fill_backlight_transform_table_v_2_2( + params, ram_table); +} + +void fill_iram_v_2_3(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parameters params) +{ + unsigned int i, j; + unsigned int set = params.set; + + ram_table->flags = 0x0; + + ram_table->min_abm_backlight = + cpu_to_be16(params.min_abm_backlight); + + for (i = 0; i < NUM_AGGR_LEVEL; i++) { + ram_table->hybrid_factor[i] = abm_settings[set][i].brightness_gain; + ram_table->contrast_factor[i] = abm_settings[set][i].contrast_factor; + ram_table->deviation_gain[i] = abm_settings[set][i].deviation_gain; + ram_table->min_knee[i] = abm_settings[set][i].min_knee; + ram_table->max_knee[i] = abm_settings[set][i].max_knee; + + for (j = 0; j < NUM_AMBI_LEVEL; j++) { + ram_table->min_reduction[j][i] = abm_settings[set][i].min_reduction; + ram_table->max_reduction[j][i] = abm_settings[set][i].max_reduction; + ram_table->bright_pos_gain[j][i] = abm_settings[set][i].bright_pos_gain; + ram_table->dark_pos_gain[j][i] = abm_settings[set][i].dark_pos_gain; + } + } ram_table->iir_curve[0] = 0x65; ram_table->iir_curve[1] = 0x65; @@ -561,7 +662,16 @@ bool dmcu_load_iram(struct dmcu *dmcu, memset(&ram_table, 0, sizeof(ram_table)); - if (dmcu->dmcu_version.abm_version == 0x22) { + if (dmcu->dmcu_version.abm_version == 0x24) { + fill_iram_v_2_3((struct iram_table_v_2_2 *)ram_table, params); + result = dmcu->funcs->load_iram( + dmcu, 0, (char *)(&ram_table), IRAM_RESERVE_AREA_START_V2_2); + } else if (dmcu->dmcu_version.abm_version == 0x23) { + fill_iram_v_2_3((struct iram_table_v_2_2 *)ram_table, params); + + result = dmcu->funcs->load_iram( + dmcu, 0, (char *)(&ram_table), IRAM_RESERVE_AREA_START_V2_2); + } else if (dmcu->dmcu_version.abm_version == 0x22) { fill_iram_v_2_2((struct iram_table_v_2_2 *)ram_table, params); result = dmcu->funcs->load_iram( @@ -581,3 +691,4 @@ bool dmcu_load_iram(struct dmcu *dmcu, return result; } + diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h index da5df00fedce..e54157026330 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h @@ -38,6 +38,7 @@ struct dmcu_iram_parameters { unsigned int backlight_lut_array_size; unsigned int backlight_ramping_reduction; unsigned int backlight_ramping_start; + unsigned int min_abm_backlight; unsigned int set; }; |
