diff options
Diffstat (limited to 'drivers/gpu')
251 files changed, 9830 insertions, 4358 deletions
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index bcbc4234893a..39d5f7562f1c 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -337,6 +337,8 @@ source "drivers/gpu/drm/vboxvideo/Kconfig" source "drivers/gpu/drm/lima/Kconfig" +source "drivers/gpu/drm/panfrost/Kconfig" + source "drivers/gpu/drm/aspeed/Kconfig" # Keep legacy drivers last diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index 7ebae3d45505..3d0c75cd687c 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -112,4 +112,5 @@ obj-$(CONFIG_DRM_TVE200) += tve200/ obj-$(CONFIG_DRM_XEN) += xen/ obj-$(CONFIG_DRM_VBOXVIDEO) += vboxvideo/ obj-$(CONFIG_DRM_LIMA) += lima/ +obj-$(CONFIG_DRM_PANFROST) += panfrost/ obj-$(CONFIG_DRM_ASPEED_GFX) += aspeed/ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 3f08be7a913e..d47354997e3c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3192,11 +3192,16 @@ static int amdgpu_device_recover_vram(struct amdgpu_device *adev) break; if (fence) { - r = dma_fence_wait_timeout(fence, false, tmo); + tmo = dma_fence_wait_timeout(fence, false, tmo); dma_fence_put(fence); fence = next; - if (r <= 0) + if (tmo == 0) { + r = -ETIMEDOUT; break; + } else if (tmo < 0) { + r = tmo; + break; + } } else { fence = next; } @@ -3207,8 +3212,8 @@ static int amdgpu_device_recover_vram(struct amdgpu_device *adev) tmo = dma_fence_wait_timeout(fence, false, tmo); dma_fence_put(fence); - if (r <= 0 || tmo <= 0) { - DRM_ERROR("recover vram bo from shadow failed\n"); + if (r < 0 || tmo <= 0) { + DRM_ERROR("recover vram bo from shadow failed, r is %ld, tmo is %ld\n", r, tmo); return -EIO; } @@ -3652,6 +3657,7 @@ static void amdgpu_device_get_min_pci_speed_width(struct amdgpu_device *adev, struct pci_dev *pdev = adev->pdev; enum pci_bus_speed cur_speed; enum pcie_link_width cur_width; + u32 ret = 1; *speed = PCI_SPEED_UNKNOWN; *width = PCIE_LNK_WIDTH_UNKNOWN; @@ -3659,6 +3665,10 @@ static void amdgpu_device_get_min_pci_speed_width(struct amdgpu_device *adev, while (pdev) { cur_speed = pcie_get_speed_cap(pdev); cur_width = pcie_get_width_cap(pdev); + ret = pcie_bandwidth_available(adev->pdev, NULL, + NULL, &cur_width); + if (!ret) + cur_width = PCIE_LNK_WIDTH_RESRV; if (cur_speed != PCI_SPEED_UNKNOWN) { if (*speed == PCI_SPEED_UNKNOWN) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 0b8ef2d27d6b..fe393a46f881 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -35,6 +35,7 @@ #include "amdgpu_trace.h" #define AMDGPU_IB_TEST_TIMEOUT msecs_to_jiffies(1000) +#define AMDGPU_IB_TEST_GFX_XGMI_TIMEOUT msecs_to_jiffies(2000) /* * IB @@ -344,6 +345,8 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev) * cost waiting for it coming back under RUNTIME only */ tmo_gfx = 8 * AMDGPU_IB_TEST_TIMEOUT; + } else if (adev->gmc.xgmi.hive_id) { + tmo_gfx = AMDGPU_IB_TEST_GFX_XGMI_TIMEOUT; } for (i = 0; i < adev->num_rings; ++i) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 107756dd1b25..ba67d1023264 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -2455,8 +2455,6 @@ static int gfx_v9_0_rlc_resume(struct amdgpu_device *adev) /* disable CG */ WREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL, 0); - adev->gfx.rlc.funcs->reset(adev); - gfx_v9_0_init_pg(adev); if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index b3cdbf79f47b..2fee3063a0d6 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -320,6 +320,7 @@ static const struct kfd_deviceid supported_devices[] = { { 0x9876, &carrizo_device_info }, /* Carrizo */ { 0x9877, &carrizo_device_info }, /* Carrizo */ { 0x15DD, &raven_device_info }, /* Raven */ + { 0x15D8, &raven_device_info }, /* Raven */ #endif { 0x67A0, &hawaii_device_info }, /* Hawaii */ { 0x67A1, &hawaii_device_info }, /* Hawaii */ 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 054681113d71..923cb5ca5a57 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -4950,6 +4950,7 @@ static void handle_cursor_update(struct drm_plane *plane, amdgpu_crtc->cursor_width = plane->state->crtc_w; amdgpu_crtc->cursor_height = plane->state->crtc_h; + memset(&attributes, 0, sizeof(attributes)); attributes.address.high_part = upper_32_bits(address); attributes.address.low_part = lower_32_bits(address); attributes.width = plane->state->crtc_w; @@ -5861,9 +5862,11 @@ static void get_freesync_config_for_crtc( struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(new_con_state->base.connector); struct drm_display_mode *mode = &new_crtc_state->base.mode; + int vrefresh = drm_mode_vrefresh(mode); new_crtc_state->vrr_supported = new_con_state->freesync_capable && - aconnector->min_vfreq <= drm_mode_vrefresh(mode); + vrefresh >= aconnector->min_vfreq && + vrefresh <= aconnector->max_vfreq; if (new_crtc_state->vrr_supported) { new_crtc_state->stream->ignore_msa_timing_param = true; 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 3b1e38e110ee..3ef68a249f4d 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -2673,12 +2673,18 @@ void core_link_enable_stream( void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option) { struct dc *core_dc = pipe_ctx->stream->ctx->dc; + struct dc_stream_state *stream = pipe_ctx->stream; core_dc->hwss.blank_stream(pipe_ctx); if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) deallocate_mst_payload(pipe_ctx); + if (dc_is_hdmi_signal(pipe_ctx->stream->signal)) + dal_ddc_service_write_scdc_data( + stream->link->ddc, 0, + stream->timing.flags.LTE_340MCSC_SCRAMBLE); + core_dc->hwss.disable_stream(pipe_ctx, option); disable_link(pipe_ctx->stream->link, pipe_ctx->stream->signal); 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 683829466a44..0ba68d41b9c3 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -1150,28 +1150,9 @@ void hubp1_cursor_set_position( REG_UPDATE(CURSOR_CONTROL, CURSOR_ENABLE, cur_en); - //account for cases where we see negative offset relative to overlay plane - if (src_x_offset < 0 && src_y_offset < 0) { - REG_SET_2(CURSOR_POSITION, 0, - CURSOR_X_POSITION, 0, - CURSOR_Y_POSITION, 0); - x_hotspot -= src_x_offset; - y_hotspot -= src_y_offset; - } else if (src_x_offset < 0) { - REG_SET_2(CURSOR_POSITION, 0, - CURSOR_X_POSITION, 0, - CURSOR_Y_POSITION, pos->y); - x_hotspot -= src_x_offset; - } else if (src_y_offset < 0) { - REG_SET_2(CURSOR_POSITION, 0, + REG_SET_2(CURSOR_POSITION, 0, CURSOR_X_POSITION, pos->x, - CURSOR_Y_POSITION, 0); - y_hotspot -= src_y_offset; - } else { - REG_SET_2(CURSOR_POSITION, 0, - CURSOR_X_POSITION, pos->x, - CURSOR_Y_POSITION, pos->y); - } + CURSOR_Y_POSITION, pos->y); REG_SET_2(CURSOR_HOT_SPOT, 0, CURSOR_HOT_SPOT_X, x_hotspot, diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c index 34f3782ff018..39a547084e90 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c @@ -91,6 +91,12 @@ static void vega20_set_default_registry_data(struct pp_hwmgr *hwmgr) * MP0CLK DS */ data->registry_data.disallowed_features = 0xE0041C00; + /* ECC feature should be disabled on old SMUs */ + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetSmuVersion); + hwmgr->smu_version = smum_get_argument(hwmgr); + if (hwmgr->smu_version < 0x282100) + data->registry_data.disallowed_features |= FEATURE_ECC_MASK; + data->registry_data.od_state_in_dc_support = 0; data->registry_data.thermal_support = 1; data->registry_data.skip_baco_hardware = 0; @@ -357,6 +363,7 @@ static void vega20_init_dpm_defaults(struct pp_hwmgr *hwmgr) data->smu_features[GNLD_DS_MP1CLK].smu_feature_id = FEATURE_DS_MP1CLK_BIT; data->smu_features[GNLD_DS_MP0CLK].smu_feature_id = FEATURE_DS_MP0CLK_BIT; data->smu_features[GNLD_XGMI].smu_feature_id = FEATURE_XGMI_BIT; + data->smu_features[GNLD_ECC].smu_feature_id = FEATURE_ECC_BIT; for (i = 0; i < GNLD_FEATURES_MAX; i++) { data->smu_features[i].smu_feature_bitmap = @@ -3030,7 +3037,8 @@ static int vega20_get_ppfeature_status(struct pp_hwmgr *hwmgr, char *buf) "FCLK_DS", "MP1CLK_DS", "MP0CLK_DS", - "XGMI"}; + "XGMI", + "ECC"}; static const char *output_title[] = { "FEATURES", "BITMASK", @@ -3472,6 +3480,7 @@ static int vega20_apply_clocks_adjust_rules(struct pp_hwmgr *hwmgr) struct vega20_single_dpm_table *dpm_table; bool vblank_too_short = false; bool disable_mclk_switching; + bool disable_fclk_switching; uint32_t i, latency; disable_mclk_switching = ((1 < hwmgr->display_config->num_display) && @@ -3547,13 +3556,20 @@ static int vega20_apply_clocks_adjust_rules(struct pp_hwmgr *hwmgr) if (hwmgr->display_config->nb_pstate_switch_disable) dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; + if ((disable_mclk_switching && + (dpm_table->dpm_state.hard_min_level == dpm_table->dpm_levels[dpm_table->count - 1].value)) || + hwmgr->display_config->min_mem_set_clock / 100 >= dpm_table->dpm_levels[dpm_table->count - 1].value) + disable_fclk_switching = true; + else + disable_fclk_switching = false; + /* fclk */ dpm_table = &(data->dpm_table.fclk_table); dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[0].value; dpm_table->dpm_state.soft_max_level = VG20_CLOCK_MAX_DEFAULT; dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[0].value; dpm_table->dpm_state.hard_max_level = VG20_CLOCK_MAX_DEFAULT; - if (hwmgr->display_config->nb_pstate_switch_disable) + if (hwmgr->display_config->nb_pstate_switch_disable || disable_fclk_switching) dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; /* vclk */ diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h index f37a28afd574..2c3125f82b24 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h @@ -80,6 +80,7 @@ enum { GNLD_DS_MP1CLK, GNLD_DS_MP0CLK, GNLD_XGMI, + GNLD_ECC, GNLD_FEATURES_MAX }; diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h index 63d5cf691549..195c4ae67058 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if.h @@ -99,7 +99,7 @@ #define FEATURE_DS_MP1CLK_BIT 30 #define FEATURE_DS_MP0CLK_BIT 31 #define FEATURE_XGMI_BIT 32 -#define FEATURE_SPARE_33_BIT 33 +#define FEATURE_ECC_BIT 33 #define FEATURE_SPARE_34_BIT 34 #define FEATURE_SPARE_35_BIT 35 #define FEATURE_SPARE_36_BIT 36 @@ -165,7 +165,8 @@ #define FEATURE_DS_FCLK_MASK (1 << FEATURE_DS_FCLK_BIT ) #define FEATURE_DS_MP1CLK_MASK (1 << FEATURE_DS_MP1CLK_BIT ) #define FEATURE_DS_MP0CLK_MASK (1 << FEATURE_DS_MP0CLK_BIT ) -#define FEATURE_XGMI_MASK (1 << FEATURE_XGMI_BIT ) +#define FEATURE_XGMI_MASK (1ULL << FEATURE_XGMI_BIT ) +#define FEATURE_ECC_MASK (1ULL << FEATURE_ECC_BIT ) #define DPM_OVERRIDE_DISABLE_SOCCLK_PID 0x00000001 #define DPM_OVERRIDE_DISABLE_UCLK_PID 0x00000002 diff --git a/drivers/gpu/drm/aspeed/Kconfig b/drivers/gpu/drm/aspeed/Kconfig index 42b74d18a41b..cccab520e02f 100644 --- a/drivers/gpu/drm/aspeed/Kconfig +++ b/drivers/gpu/drm/aspeed/Kconfig @@ -1,11 +1,11 @@ config DRM_ASPEED_GFX tristate "ASPEED BMC Display Controller" depends on DRM && OF + depends on (COMPILE_TEST || ARCH_ASPEED) select DRM_KMS_HELPER select DRM_KMS_CMA_HELPER - select DRM_PANEL - select DMA_CMA - select CMA + select DMA_CMA if HAVE_DMA_CONTIGUOUS + select CMA if HAVE_DMA_CONTIGUOUS select MFD_SYSCON help Chose this option if you have an ASPEED AST2500 SOC Display diff --git a/drivers/gpu/drm/bochs/bochs.h b/drivers/gpu/drm/bochs/bochs.h index 049d058571d4..341cc9d1bab4 100644 --- a/drivers/gpu/drm/bochs/bochs.h +++ b/drivers/gpu/drm/bochs/bochs.h @@ -7,6 +7,7 @@ #include <drm/drm_crtc_helper.h> #include <drm/drm_encoder.h> #include <drm/drm_fb_helper.h> +#include <drm/drm_simple_kms_helper.h> #include <drm/drm_gem.h> @@ -69,9 +70,8 @@ struct bochs_device { struct edid *edid; /* drm */ - struct drm_device *dev; - struct drm_crtc crtc; - struct drm_encoder encoder; + struct drm_device *dev; + struct drm_simple_display_pipe pipe; struct drm_connector connector; /* ttm */ diff --git a/drivers/gpu/drm/bochs/bochs_kms.c b/drivers/gpu/drm/bochs/bochs_kms.c index 485f9cf05e8b..5e905f50449d 100644 --- a/drivers/gpu/drm/bochs/bochs_kms.c +++ b/drivers/gpu/drm/bochs/bochs_kms.c @@ -22,76 +22,55 @@ MODULE_PARM_DESC(defy, "default y resolution"); /* ---------------------------------------------------------------------- */ -static void bochs_crtc_mode_set_nofb(struct drm_crtc *crtc) +static const uint32_t bochs_formats[] = { + DRM_FORMAT_XRGB8888, + DRM_FORMAT_BGRX8888, +}; + +static void bochs_plane_update(struct bochs_device *bochs, + struct drm_plane_state *state) { - struct bochs_device *bochs = - container_of(crtc, struct bochs_device, crtc); + struct bochs_bo *bo; - bochs_hw_setmode(bochs, &crtc->mode); + if (!state->fb || !bochs->stride) + return; + + bo = gem_to_bochs_bo(state->fb->obj[0]); + bochs_hw_setbase(bochs, + state->crtc_x, + state->crtc_y, + bo->bo.offset); + bochs_hw_setformat(bochs, state->fb->format); } -static void bochs_crtc_atomic_enable(struct drm_crtc *crtc, - struct drm_crtc_state *old_crtc_state) +static void bochs_pipe_enable(struct drm_simple_display_pipe *pipe, + struct drm_crtc_state *crtc_state, + struct drm_plane_state *plane_state) { + struct bochs_device *bochs = pipe->crtc.dev->dev_private; + + bochs_hw_setmode(bochs, &crtc_state->mode); + bochs_plane_update(bochs, plane_state); } -static void bochs_crtc_atomic_flush(struct drm_crtc *crtc, - struct drm_crtc_state *old_crtc_state) +static void bochs_pipe_update(struct drm_simple_display_pipe *pipe, + struct drm_plane_state *old_state) { - struct drm_device *dev = crtc->dev; - struct drm_pending_vblank_event *event; + struct bochs_device *bochs = pipe->crtc.dev->dev_private; + struct drm_crtc *crtc = &pipe->crtc; - if (crtc->state && crtc->state->event) { - unsigned long irqflags; + bochs_plane_update(bochs, pipe->plane.state); - spin_lock_irqsave(&dev->event_lock, irqflags); - event = crtc->state->event; + if (crtc->state->event) { + spin_lock_irq(&crtc->dev->event_lock); + drm_crtc_send_vblank_event(crtc, crtc->state->event); crtc->state->event = NULL; - drm_crtc_send_vblank_event(crtc, event); - spin_unlock_irqrestore(&dev->event_lock, irqflags); + spin_unlock_irq(&crtc->dev->event_lock); } } - -/* These provide the minimum set of functions required to handle a CRTC */ -static const struct drm_crtc_funcs bochs_crtc_funcs = { - .set_config = drm_atomic_helper_set_config, - .destroy = drm_crtc_cleanup, - .page_flip = drm_atomic_helper_page_flip, - .reset = drm_atomic_helper_crtc_reset, - .atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state, - .atomic_destroy_state = drm_atomic_helper_crtc_destroy_state, -}; - -static const struct drm_crtc_helper_funcs bochs_helper_funcs = { - .mode_set_nofb = bochs_crtc_mode_set_nofb, - .atomic_enable = bochs_crtc_atomic_enable, - .atomic_flush = bochs_crtc_atomic_flush, -}; - -static const uint32_t bochs_formats[] = { - DRM_FORMAT_XRGB8888, - DRM_FORMAT_BGRX8888, -}; - -static void bochs_plane_atomic_update(struct drm_plane *plane, - struct drm_plane_state *old_state) -{ - struct bochs_device *bochs = plane->dev->dev_private; - struct bochs_bo *bo; - - if (!plane->state->fb) - return; - bo = gem_to_bochs_bo(plane->state->fb->obj[0]); - bochs_hw_setbase(bochs, - plane->state->crtc_x, - plane->state->crtc_y, - bo->bo.offset); - bochs_hw_setformat(bochs, plane->state->fb->format); -} - -static int bochs_plane_prepare_fb(struct drm_plane *plane, - struct drm_plane_state *new_state) +static int bochs_pipe_prepare_fb(struct drm_simple_display_pipe *pipe, + struct drm_plane_state *new_state) { struct bochs_bo *bo; @@ -101,8 +80,8 @@ static int bochs_plane_prepare_fb(struct drm_plane *plane, return bochs_bo_pin(bo, TTM_PL_FLAG_VRAM); } -static void bochs_plane_cleanup_fb(struct drm_plane *plane, - struct drm_plane_state *old_state) +static void bochs_pipe_cleanup_fb(struct drm_simple_display_pipe *pipe, + struct drm_plane_state *old_state) { struct bochs_bo *bo; @@ -112,73 +91,13 @@ static void bochs_plane_cleanup_fb(struct drm_plane *plane, bochs_bo_unpin(bo); } -static const struct drm_plane_helper_funcs bochs_plane_helper_funcs = { - .atomic_update = bochs_plane_atomic_update, - .prepare_fb = bochs_plane_prepare_fb, - .cleanup_fb = bochs_plane_cleanup_fb, -}; - -static const struct drm_plane_funcs bochs_plane_funcs = { - .update_plane = drm_atomic_helper_update_plane, - .disable_plane = drm_atomic_helper_disable_plane, - .destroy = drm_primary_helper_destroy, - .reset = drm_atomic_helper_plane_reset, - .atomic_duplicate_state = drm_atomic_helper_plane_duplicate_state, - .atomic_destroy_state = drm_atomic_helper_plane_destroy_state, -}; - -static struct drm_plane *bochs_primary_plane(struct drm_device *dev) -{ - struct drm_plane *primary; - int ret; - - primary = kzalloc(sizeof(*primary), GFP_KERNEL); - if (primary == NULL) { - DRM_DEBUG_KMS("Failed to allocate primary plane\n"); - return NULL; - } - - ret = drm_universal_plane_init(dev, primary, 0, - &bochs_plane_funcs, - bochs_formats, - ARRAY_SIZE(bochs_formats), - NULL, - DRM_PLANE_TYPE_PRIMARY, NULL); - if (ret) { - kfree(primary); - return NULL; - } - - drm_plane_helper_add(primary, &bochs_plane_helper_funcs); - return primary; -} - -static void bochs_crtc_init(struct drm_device *dev) -{ - struct bochs_device *bochs = dev->dev_private; - struct drm_crtc *crtc = &bochs->crtc; - struct drm_plane *primary = bochs_primary_plane(dev); - - drm_crtc_init_with_planes(dev, crtc, primary, NULL, - &bochs_crtc_funcs, NULL); - drm_crtc_helper_add(crtc, &bochs_helper_funcs); -} - -static const struct drm_encoder_funcs bochs_encoder_encoder_funcs = { - .destroy = drm_encoder_cleanup, +static const struct drm_simple_display_pipe_funcs bochs_pipe_funcs = { + .enable = bochs_pipe_enable, + .update = bochs_pipe_update, + .prepare_fb = bochs_pipe_prepare_fb, + .cleanup_fb = bochs_pipe_cleanup_fb, }; -static void bochs_encoder_init(struct drm_device *dev) -{ - struct bochs_device *bochs = dev->dev_private; - struct drm_encoder *encoder = &bochs->encoder; - - encoder->possible_crtcs = 0x1; - drm_encoder_init(dev, encoder, &bochs_encoder_encoder_funcs, - DRM_MODE_ENCODER_DAC, NULL); -} - - static int bochs_connector_get_modes(struct drm_connector *connector) { struct bochs_device *bochs = @@ -278,11 +197,14 @@ int bochs_kms_init(struct bochs_device *bochs) bochs->dev->mode_config.funcs = &bochs_mode_funcs; - bochs_crtc_init(bochs->dev); - bochs_encoder_init(bochs->dev); bochs_connector_init(bochs->dev); - drm_connector_attach_encoder(&bochs->connector, - &bochs->encoder); + drm_simple_display_pipe_init(bochs->dev, + &bochs->pipe, + &bochs_pipe_funcs, + bochs_formats, + ARRAY_SIZE(bochs_formats), + NULL, + &bochs->connector); drm_mode_config_reset(bochs->dev); diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c index a63e5f0dae56..db761329a1e3 100644 --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c @@ -1037,6 +1037,31 @@ void dw_hdmi_phy_i2c_write(struct dw_hdmi *hdmi, unsigned short data, } EXPORT_SYMBOL_GPL(dw_hdmi_phy_i2c_write); +/* Filter out invalid setups to avoid configuring SCDC and scrambling */ +static bool dw_hdmi_support_scdc(struct dw_hdmi *hdmi) +{ + struct drm_display_info *display = &hdmi->connector.display_info; + + /* Completely disable SCDC support for older controllers */ + if (hdmi->version < 0x200a) + return false; + + /* Disable if SCDC is not supported, or if an HF-VSDB block is absent */ + if (!display->hdmi.scdc.supported || + !display->hdmi.scdc.scrambling.supported) + return false; + + /* + * Disable if display only support low TMDS rates and scrambling + * for low rates is not supported either + */ + if (!display->hdmi.scdc.scrambling.low_rates && + display->max_tmds_clock <= 340000) + return false; + + return true; +} + /* * HDMI2.0 Specifies the following procedure for High TMDS Bit Rates: * - The Source shall suspend transmission of the TMDS clock and data @@ -1055,7 +1080,7 @@ void dw_hdmi_set_high_tmds_clock_ratio(struct dw_hdmi *hdmi) unsigned long mtmdsclock = hdmi->hdmi_data.video_mode.mtmdsclock; /* Control for TMDS Bit Period/TMDS Clock-Period Ratio */ - if (hdmi->connector.display_info.hdmi.scdc.supported) { + if (dw_hdmi_support_scdc(hdmi)) { if (mtmdsclock > HDMI14_MAX_TMDSCLK) drm_scdc_set_high_tmds_clock_ratio(hdmi->ddc, 1); else @@ -1579,8 +1604,9 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi, /* Set up HDMI_FC_INVIDCONF */ inv_val = (hdmi->hdmi_data.hdcp_enable || - vmode->mtmdsclock > HDMI14_MAX_TMDSCLK || - hdmi_info->scdc.scrambling.low_rates ? + (dw_hdmi_support_scdc(hdmi) && + (vmode->mtmdsclock > HDMI14_MAX_TMDSCLK || + hdmi_info->scdc.scrambling.low_rates)) ? HDMI_FC_INVIDCONF_HDCP_KEEPOUT_ACTIVE : HDMI_FC_INVIDCONF_HDCP_KEEPOUT_INACTIVE); @@ -1646,7 +1672,7 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi, } /* Scrambling Control */ - if (hdmi_info->scdc.supported) { + if (dw_hdmi_support_scdc(hdmi)) { if (vmode->mtmdsclock > HDMI14_MAX_TMDSCLK || hdmi_info->scdc.scrambling.low_rates) { /* diff --git a/drivers/gpu/drm/bridge/ti-tfp410.c b/drivers/gpu/drm/bridge/ti-tfp410.c index 285be4a0f4bd..8b0e71bd3ca7 100644 --- a/drivers/gpu/drm/bridge/ti-tfp410.c +++ b/drivers/gpu/drm/bridge/ti-tfp410.c @@ -29,8 +29,10 @@ struct tfp410 { struct drm_connector connector; unsigned int connector_type; + u32 bus_format; struct i2c_adapter *ddc; struct gpio_desc *hpd; + int hpd_irq; struct delayed_work hpd_work; struct gpio_desc *powerdown; @@ -124,8 +126,10 @@ static int tfp410_attach(struct drm_bridge *bridge) return -ENODEV; } - if (dvi->hpd) + if (dvi->hpd_irq >= 0) dvi->connector.polled = DRM_CONNECTOR_POLL_HPD; + else + dvi->connector.polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; drm_connector_helper_add(&dvi->connector, &tfp410_con_helper_funcs); @@ -136,6 +140,9 @@ static int tfp410_attach(struct drm_bridge *bridge) return ret; } + drm_display_info_set_bus_formats(&dvi->connector.display_info, + &dvi->bus_format, 1); + drm_connector_attach_encoder(&dvi->connector, bridge->encoder); @@ -194,6 +201,7 @@ static int tfp410_parse_timings(struct tfp410 *dvi, bool i2c) struct drm_bridge_timings *timings = &dvi->timings; struct device_node *ep; u32 pclk_sample = 0; + u32 bus_width = 24; s32 deskew = 0; /* Start with defaults. */ @@ -218,6 +226,7 @@ static int tfp410_parse_timings(struct tfp410 *dvi, bool i2c) /* Get the sampling edge from the endpoint. */ of_property_read_u32(ep, "pclk-sample", &pclk_sample); + of_property_read_u32(ep, "bus-width", &bus_width); of_node_put(ep); timings->input_bus_flags = DRM_BUS_FLAG_DE_HIGH; @@ -235,6 +244,17 @@ static int tfp410_parse_timings(struct tfp410 *dvi, bool i2c) return -EINVAL; } + switch (bus_width) { + case 12: + dvi->bus_format = MEDIA_BUS_FMT_RGB888_2X12_LE; + break; + case 24: + dvi->bus_format = MEDIA_BUS_FMT_RGB888_1X24; + break; + default: + return -EINVAL; + } + /* Get the setup and hold time from vendor-specific properties. */ of_property_read_u32(dvi->dev->of_node, "ti,deskew", (u32 *)&deskew); if (deskew < -4 || deskew > 3) @@ -324,10 +344,15 @@ static int tfp410_init(struct device *dev, bool i2c) return PTR_ERR(dvi->powerdown); } - if (dvi->hpd) { + if (dvi->hpd) + dvi->hpd_irq = gpiod_to_irq(dvi->hpd); + else + dvi->hpd_irq = -ENXIO; + + if (dvi->hpd_irq >= 0) { INIT_DELAYED_WORK(&dvi->hpd_work, tfp410_hpd_work_func); - ret = devm_request_threaded_irq(dev, gpiod_to_irq(dvi->hpd), + ret = devm_request_threaded_irq(dev, dvi->hpd_irq, NULL, tfp410_hpd_irq_thread, IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | IRQF_ONESHOT, "hdmi-hpd", dvi); diff --git a/drivers/gpu/drm/cirrus/cirrus.c b/drivers/gpu/drm/cirrus/cirrus.c index 5095b8ce52c2..be4ea370ba31 100644 --- a/drivers/gpu/drm/cirrus/cirrus.c +++ b/drivers/gpu/drm/cirrus/cirrus.c @@ -307,16 +307,16 @@ static int cirrus_fb_blit_rect(struct drm_framebuffer *fb, return -ENOMEM; if (cirrus->cpp == fb->format->cpp[0]) - drm_fb_memcpy_dstclip(__io_virt(cirrus->vram), + drm_fb_memcpy_dstclip(cirrus->vram, vmap, fb, rect); else if (fb->format->cpp[0] == 4 && cirrus->cpp == 2) - drm_fb_xrgb8888_to_rgb565_dstclip(__io_virt(cirrus->vram), + drm_fb_xrgb8888_to_rgb565_dstclip(cirrus->vram, cirrus->pitch, vmap, fb, rect, false); else if (fb->format->cpp[0] == 4 && cirrus->cpp == 3) - drm_fb_xrgb8888_to_rgb888_dstclip(__io_virt(cirrus->vram), + drm_fb_xrgb8888_to_rgb888_dstclip(cirrus->vram, cirrus->pitch, vmap, fb, rect); diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 86efd2da37f9..2e0cb4246cbd 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -1034,7 +1034,7 @@ disable_outputs(struct drm_device *dev, struct drm_atomic_state *old_state) funcs->atomic_disable(crtc, old_crtc_state); else if (funcs->disable) funcs->disable(crtc); - else + else if (funcs->dpms) funcs->dpms(crtc, DRM_MODE_DPMS_OFF); if (!(dev->irq_enabled && dev->num_crtcs)) @@ -1277,10 +1277,9 @@ void drm_atomic_helper_commit_modeset_enables(struct drm_device *dev, if (new_crtc_state->enable) { DRM_DEBUG_ATOMIC("enabling [CRTC:%d:%s]\n", crtc->base.id, crtc->name); - if (funcs->atomic_enable) funcs->atomic_enable(crtc, old_crtc_state); - else + else if (funcs->commit) funcs->commit(crtc); } } diff --git a/drivers/gpu/drm/drm_client.c b/drivers/gpu/drm/drm_client.c index 9b2bd28dde0a..f20d1dda3961 100644 --- a/drivers/gpu/drm/drm_client.c +++ b/drivers/gpu/drm/drm_client.c @@ -69,7 +69,8 @@ EXPORT_SYMBOL(drm_client_close); * @name: Client name * @funcs: DRM client functions (optional) * - * This initialises the client and opens a &drm_file. Use drm_client_add() to complete the process. + * This initialises the client and opens a &drm_file. + * Use drm_client_register() to complete the process. * The caller needs to hold a reference on @dev before calling this function. * The client is freed when the &drm_device is unregistered. See drm_client_release(). * @@ -108,16 +109,16 @@ err_put_module: EXPORT_SYMBOL(drm_client_init); /** - * drm_client_add - Add client to the device list + * drm_client_register - Register client * @client: DRM client * * Add the client to the &drm_device client list to activate its callbacks. * @client must be initialized by a call to drm_client_init(). After - * drm_client_add() it is no longer permissible to call drm_client_release() + * drm_client_register() it is no longer permissible to call drm_client_release() * directly (outside the unregister callback), instead cleanup will happen * automatically on driver unload. */ -void drm_client_add(struct drm_client_dev *client) +void drm_client_register(struct drm_client_dev *client) { struct drm_device *dev = client->dev; @@ -125,7 +126,7 @@ void drm_client_add(struct drm_client_dev *client) list_add(&client->list, &dev->clientlist); mutex_unlock(&dev->clientlist_mutex); } -EXPORT_SYMBOL(drm_client_add); +EXPORT_SYMBOL(drm_client_register); /** * drm_client_release - Release DRM client resources diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 84791dd4a90d..4de4b9d59d49 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -1961,7 +1961,7 @@ static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper, best_depth = fmt->depth; } } - if (sizes.surface_depth != best_depth) { + if (sizes.surface_depth != best_depth && best_depth) { DRM_INFO("requested bpp %d, scaled depth down to %d", sizes.surface_bpp, best_depth); sizes.surface_depth = best_depth; @@ -2559,6 +2559,194 @@ static void drm_setup_crtc_rotation(struct drm_fb_helper *fb_helper, fb_helper->sw_rotations |= DRM_MODE_ROTATE_0; } +static struct drm_fb_helper_crtc * +drm_fb_helper_crtc(struct drm_fb_helper *fb_helper, struct drm_crtc *crtc) +{ + int i; + + for (i = 0; i < fb_helper->crtc_count; i++) + if (fb_helper->crtc_info[i].mode_set.crtc == crtc) + return &fb_helper->crtc_info[i]; + + return NULL; +} + +/* Try to read the BIOS display configuration and use it for the initial config */ +static bool drm_fb_helper_firmware_config(struct drm_fb_helper *fb_helper, + struct drm_fb_helper_crtc **crtcs, + struct drm_display_mode **modes, + struct drm_fb_offset *offsets, + bool *enabled, int width, int height) +{ + struct drm_device *dev = fb_helper->dev; + unsigned int count = min(fb_helper->connector_count, BITS_PER_LONG); + unsigned long conn_configured, conn_seq; + int i, j; + bool *save_enabled; + bool fallback = true, ret = true; + int num_connectors_enabled = 0; + int num_connectors_detected = 0; + struct drm_modeset_acquire_ctx ctx; + + save_enabled = kcalloc(count, sizeof(bool), GFP_KERNEL); + if (!save_enabled) + return false; + + drm_modeset_acquire_init(&ctx, 0); + + while (drm_modeset_lock_all_ctx(dev, &ctx) != 0) + drm_modeset_backoff(&ctx); + + memcpy(save_enabled, enabled, count); + conn_seq = GENMASK(count - 1, 0); + conn_configured = 0; +retry: + for (i = 0; i < count; i++) { + struct drm_fb_helper_connector *fb_conn; + struct drm_connector *connector; + struct drm_encoder *encoder; + struct drm_fb_helper_crtc *new_crtc; + + fb_conn = fb_helper->connector_info[i]; + connector = fb_conn->connector; + + if (conn_configured & BIT(i)) + continue; + + /* First pass, only consider tiled connectors */ + if (conn_seq == GENMASK(count - 1, 0) && !connector->has_tile) + continue; + + if (connector->status == connector_status_connected) + num_connectors_detected++; + + if (!enabled[i]) { + DRM_DEBUG_KMS("connector %s not enabled, skipping\n", + connector->name); + conn_configured |= BIT(i); + continue; + } + + if (connector->force == DRM_FORCE_OFF) { + DRM_DEBUG_KMS("connector %s is disabled by user, skipping\n", + connector->name); + enabled[i] = false; + continue; + } + + encoder = connector->state->best_encoder; + if (!encoder || WARN_ON(!connector->state->crtc)) { + if (connector->force > DRM_FORCE_OFF) + goto bail; + + DRM_DEBUG_KMS("connector %s has no encoder or crtc, skipping\n", + connector->name); + enabled[i] = false; + conn_configured |= BIT(i); + continue; + } + + num_connectors_enabled++; + + new_crtc = drm_fb_helper_crtc(fb_helper, connector->state->crtc); + + /* + * Make sure we're not trying to drive multiple connectors + * with a single CRTC, since our cloning support may not + * match the BIOS. + */ + for (j = 0; j < count; j++) { + if (crtcs[j] == new_crtc) { + DRM_DEBUG_KMS("fallback: cloned configuration\n"); + goto bail; + } + } + + DRM_DEBUG_KMS("looking for cmdline mode on connector %s\n", + connector->name); + + /* go for command line mode first */ + modes[i] = drm_pick_cmdline_mode(fb_conn); + + /* try for preferred next */ + if (!modes[i]) { + DRM_DEBUG_KMS("looking for preferred mode on connector %s %d\n", + connector->name, connector->has_tile); + modes[i] = drm_has_preferred_mode(fb_conn, width, + height); + } + + /* No preferred mode marked by the EDID? Are there any modes? */ + if (!modes[i] && !list_empty(&connector->modes)) { + DRM_DEBUG_KMS("using first mode listed on connector %s\n", + connector->name); + modes[i] = list_first_entry(&connector->modes, + struct drm_display_mode, + head); + } + + /* last resort: use current mode */ + if (!modes[i]) { + /* + * IMPORTANT: We want to use the adjusted mode (i.e. + * after the panel fitter upscaling) as the initial + * config, not the input mode, which is what crtc->mode + * usually contains. But since our current + * code puts a mode derived from the post-pfit timings + * into crtc->mode this works out correctly. + * + * This is crtc->mode and not crtc->state->mode for the + * fastboot check to work correctly. + */ + DRM_DEBUG_KMS("looking for current mode on connector %s\n", + connector->name); + modes[i] = &connector->state->crtc->mode; + } + crtcs[i] = new_crtc; + + DRM_DEBUG_KMS("connector %s on [CRTC:%d:%s]: %dx%d%s\n", + connector->name, + connector->state->crtc->base.id, + connector->state->crtc->name, + modes[i]->hdisplay, modes[i]->vdisplay, + modes[i]->flags & DRM_MODE_FLAG_INTERLACE ? "i" : ""); + + fallback = false; + conn_configured |= BIT(i); + } + + if (conn_configured != conn_seq) { /* repeat until no more are found */ + conn_seq = conn_configured; + goto retry; + } + + /* + * If the BIOS didn't enable everything it could, fall back to have the + * same user experiencing of lighting up as much as possible like the + * fbdev helper library. + */ + if (num_connectors_enabled != num_connectors_detected && + num_connectors_enabled < dev->mode_config.num_crtc) { + DRM_DEBUG_KMS("fallback: Not all outputs enabled\n"); + DRM_DEBUG_KMS("Enabled: %i, detected: %i\n", num_connectors_enabled, + num_connectors_detected); + fallback = true; + } + + if (fallback) { +bail: + DRM_DEBUG_KMS("Not using firmware configuration\n"); + memcpy(enabled, save_enabled, count); + ret = false; + } + + drm_modeset_drop_locks(&ctx); + drm_modeset_acquire_fini(&ctx); + + kfree(save_enabled); + return ret; +} + static void drm_setup_crtcs(struct drm_fb_helper *fb_helper, u32 width, u32 height) { @@ -2591,10 +2779,8 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper, DRM_DEBUG_KMS("No connectors reported connected with modes\n"); drm_enable_connectors(fb_helper, enabled); - if (!(fb_helper->funcs->initial_config && - fb_helper->funcs->initial_config(fb_helper, crtcs, modes, - offsets, - enabled, width, height))) { + if (!drm_fb_helper_firmware_config(fb_helper, crtcs, modes, offsets, + enabled, width, height)) { memset(modes, 0, fb_helper->connector_count*sizeof(modes[0])); memset(crtcs, 0, fb_helper->connector_count*sizeof(crtcs[0])); memset(offsets, 0, fb_helper->connector_count*sizeof(offsets[0])); @@ -3322,7 +3508,7 @@ int drm_fbdev_generic_setup(struct drm_device *dev, unsigned int preferred_bpp) if (ret) DRM_DEV_DEBUG(dev->dev, "client hotplug ret=%d\n", ret); - drm_client_add(&fb_helper->client); + drm_client_register(&fb_helper->client); return 0; } diff --git a/drivers/gpu/drm/drm_format_helper.c b/drivers/gpu/drm/drm_format_helper.c index 00d716f14173..a18da35145b7 100644 --- a/drivers/gpu/drm/drm_format_helper.c +++ b/drivers/gpu/drm/drm_format_helper.c @@ -10,23 +10,17 @@ #include <linux/module.h> #include <linux/slab.h> +#include <linux/io.h> #include <drm/drm_format_helper.h> #include <drm/drm_framebuffer.h> #include <drm/drm_fourcc.h> #include <drm/drm_rect.h> -static void drm_fb_memcpy_lines(void *dst, unsigned int dst_pitch, - void *src, unsigned int src_pitch, - unsigned int linelength, unsigned int lines) +static unsigned int clip_offset(struct drm_rect *clip, + unsigned int pitch, unsigned int cpp) { - int line; - - for (line = 0; line < lines; line++) { - memcpy(dst, src, linelength); - src += src_pitch; - dst += dst_pitch; - } + return clip->y1 * pitch + clip->x1 * cpp; } /** @@ -43,35 +37,44 @@ void drm_fb_memcpy(void *dst, void *vaddr, struct drm_framebuffer *fb, struct drm_rect *clip) { unsigned int cpp = drm_format_plane_cpp(fb->format->format, 0); - unsigned int offset = (clip->y1 * fb->pitches[0]) + (clip->x1 * cpp); size_t len = (clip->x2 - clip->x1) * cpp; + unsigned int y, lines = clip->y2 - clip->y1; - drm_fb_memcpy_lines(dst, len, - vaddr + offset, fb->pitches[0], - len, clip->y2 - clip->y1); + vaddr += clip_offset(clip, fb->pitches[0], cpp); + for (y = 0; y < lines; y++) { + memcpy(dst, vaddr, len); + vaddr += fb->pitches[0]; + dst += len; + } } EXPORT_SYMBOL(drm_fb_memcpy); /** * drm_fb_memcpy_dstclip - Copy clip buffer - * @dst: Destination buffer + * @dst: Destination buffer (iomem) * @vaddr: Source buffer * @fb: DRM framebuffer * @clip: Clip rectangle area to copy * * This function applies clipping on dst, i.e. the destination is a - * full framebuffer but only the clip rect content is copied over. + * full (iomem) framebuffer but only the clip rect content is copied over. */ -void drm_fb_memcpy_dstclip(void *dst, void *vaddr, struct drm_framebuffer *fb, +void drm_fb_memcpy_dstclip(void __iomem *dst, void *vaddr, + struct drm_framebuffer *fb, struct drm_rect *clip) { unsigned int cpp = drm_format_plane_cpp(fb->format->format, 0); - unsigned int offset = (clip->y1 * fb->pitches[0]) + (clip->x1 * cpp); + unsigned int offset = clip_offset(clip, fb->pitches[0], cpp); size_t len = (clip->x2 - clip->x1) * cpp; + unsigned int y, lines = clip->y2 - clip->y1; - drm_fb_memcpy_lines(dst + offset, fb->pitches[0], - vaddr + offset, fb->pitches[0], - len, clip->y2 - clip->y1); + vaddr += offset; + dst += offset; + for (y = 0; y < lines; y++) { + memcpy_toio(dst, vaddr, len); + vaddr += fb->pitches[0]; + dst += fb->pitches[0]; + } } EXPORT_SYMBOL(drm_fb_memcpy_dstclip); @@ -110,42 +113,22 @@ void drm_fb_swab16(u16 *dst, void *vaddr, struct drm_framebuffer *fb, } EXPORT_SYMBOL(drm_fb_swab16); -static void drm_fb_xrgb8888_to_rgb565_lines(void *dst, unsigned int dst_pitch, - void *src, unsigned int src_pitch, - unsigned int src_linelength, - unsigned int lines, - bool swap) +static void drm_fb_xrgb8888_to_rgb565_line(u16 *dbuf, u32 *sbuf, + unsigned int pixels, + bool swab) { - unsigned int linepixels = src_linelength / sizeof(u32); - unsigned int x, y; - u32 *sbuf; - u16 *dbuf, val16; - - /* - * The cma memory is write-combined so reads are uncached. - * Speed up by fetching one line at a time. - */ - sbuf = kmalloc(src_linelength, GFP_KERNEL); - if (!sbuf) - return; - - for (y = 0; y < lines; y++) { - memcpy(sbuf, src, src_linelength); - dbuf = dst; - for (x = 0; x < linepixels; x++) { - val16 = ((sbuf[x] & 0x00F80000) >> 8) | - ((sbuf[x] & 0x0000FC00) >> 5) | - ((sbuf[x] & 0x000000F8) >> 3); - if (swap) - *dbuf++ = swab16(val16); - else - *dbuf++ = val16; - } - src += src_pitch; - dst += dst_pitch; + unsigned int x; + u16 val16; + + for (x = 0; x < pixels; x++) { + val16 = ((sbuf[x] & 0x00F80000) >> 8) | + ((sbuf[x] & 0x0000FC00) >> 5) | + ((sbuf[x] & 0x000000F8) >> 3); + if (swab) + dbuf[x] = swab16(val16); + else + dbuf[x] = val16; } - - kfree(sbuf); } /** @@ -154,7 +137,7 @@ static void drm_fb_xrgb8888_to_rgb565_lines(void *dst, unsigned int dst_pitch, * @vaddr: XRGB8888 source buffer * @fb: DRM framebuffer * @clip: Clip rectangle area to copy - * @swap: Swap bytes + * @swab: Swap bytes * * Drivers can use this function for RGB565 devices that don't natively * support XRGB8888. @@ -164,109 +147,124 @@ static void drm_fb_xrgb8888_to_rgb565_lines(void *dst, unsigned int dst_pitch, */ void drm_fb_xrgb8888_to_rgb565(void *dst, void *vaddr, struct drm_framebuffer *fb, - struct drm_rect *clip, bool swap) + struct drm_rect *clip, bool swab) { - unsigned int src_offset = (clip->y1 * fb->pitches[0]) - + (clip->x1 * sizeof(u32)); - size_t src_len = (clip->x2 - clip->x1) * sizeof(u32); - size_t dst_len = (clip->x2 - clip->x1) * sizeof(u16); - - drm_fb_xrgb8888_to_rgb565_lines(dst, dst_len, - vaddr + src_offset, fb->pitches[0], - src_len, clip->y2 - clip->y1, - swap); + size_t linepixels = clip->x2 - clip->x1; + size_t src_len = linepixels * sizeof(u32); + size_t dst_len = linepixels * sizeof(u16); + unsigned y, lines = clip->y2 - clip->y1; + void *sbuf; + + /* + * The cma memory is write-combined so reads are uncached. + * Speed up by fetching one line at a time. + */ + sbuf = kmalloc(src_len, GFP_KERNEL); + if (!sbuf) + return; + + vaddr += clip_offset(clip, fb->pitches[0], sizeof(u32)); + for (y = 0; y < lines; y++) { + memcpy(sbuf, vaddr, src_len); + drm_fb_xrgb8888_to_rgb565_line(dst, sbuf, linepixels, swab); + vaddr += fb->pitches[0]; + dst += dst_len; + } + + kfree(sbuf); } EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb565); /** * drm_fb_xrgb8888_to_rgb565_dstclip - Convert XRGB8888 to RGB565 clip buffer - * @dst: RGB565 destination buffer + * @dst: RGB565 destination buffer (iomem) * @dst_pitch: destination buffer pitch * @vaddr: XRGB8888 source buffer * @fb: DRM framebuffer * @clip: Clip rectangle area to copy - * @swap: Swap bytes + * @swab: Swap bytes * * Drivers can use this function for RGB565 devices that don't natively * support XRGB8888. * * This function applies clipping on dst, i.e. the destination is a - * full framebuffer but only the clip rect content is copied over. + * full (iomem) framebuffer but only the clip rect content is copied over. */ -void drm_fb_xrgb8888_to_rgb565_dstclip(void *dst, unsigned int dst_pitch, +void drm_fb_xrgb8888_to_rgb565_dstclip(void __iomem *dst, unsigned int dst_pitch, void *vaddr, struct drm_framebuffer *fb, - struct drm_rect *clip, bool swap) -{ - unsigned int src_offset = (clip->y1 * fb->pitches[0]) - + (clip->x1 * sizeof(u32)); - unsigned int dst_offset = (clip->y1 * dst_pitch) - + (clip->x1 * sizeof(u16)); - size_t src_len = (clip->x2 - clip->x1) * sizeof(u32); - - drm_fb_xrgb8888_to_rgb565_lines(dst + dst_offset, dst_pitch, - vaddr + src_offset, fb->pitches[0], - src_len, clip->y2 - clip->y1, - swap); -} -EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb565_dstclip); - -static void drm_fb_xrgb8888_to_rgb888_lines(void *dst, unsigned int dst_pitch, - void *src, unsigned int src_pitch, - unsigned int src_linelength, - unsigned int lines) + struct drm_rect *clip, bool swab) { - unsigned int linepixels = src_linelength / 3; - unsigned int x, y; - u32 *sbuf; - u8 *dbuf; + size_t linepixels = clip->x2 - clip->x1; + size_t dst_len = linepixels * sizeof(u16); + unsigned y, lines = clip->y2 - clip->y1; + void *dbuf; - sbuf = kmalloc(src_linelength, GFP_KERNEL); - if (!sbuf) + dbuf = kmalloc(dst_len, GFP_KERNEL); + if (!dbuf) return; + vaddr += clip_offset(clip, fb->pitches[0], sizeof(u32)); + dst += clip_offset(clip, dst_pitch, sizeof(u16)); for (y = 0; y < lines; y++) { - memcpy(sbuf, src, src_linelength); - dbuf = dst; - for (x = 0; x < linepixels; x++) { - *dbuf++ = (sbuf[x] & 0x000000FF) >> 0; - *dbuf++ = (sbuf[x] & 0x0000FF00) >> 8; - *dbuf++ = (sbuf[x] & 0x00FF0000) >> 16; - } - src += src_pitch; - dst += dst_pitch; + drm_fb_xrgb8888_to_rgb565_line(dbuf, vaddr, linepixels, swab); + memcpy_toio(dst, dbuf, dst_len); + vaddr += fb->pitches[0]; + dst += dst_len; } - kfree(sbuf); + kfree(dbuf); +} +EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb565_dstclip); + +static void drm_fb_xrgb8888_to_rgb888_line(u8 *dbuf, u32 *sbuf, + unsigned int pixels) +{ + unsigned int x; + + for (x = 0; x < pixels; x++) { + *dbuf++ = (sbuf[x] & 0x000000FF) >> 0; + *dbuf++ = (sbuf[x] & 0x0000FF00) >> 8; + *dbuf++ = (sbuf[x] & 0x00FF0000) >> 16; + } } /** * drm_fb_xrgb8888_to_rgb888_dstclip - Convert XRGB8888 to RGB888 clip buffer - * @dst: RGB565 destination buffer + * @dst: RGB565 destination buffer (iomem) * @dst_pitch: destination buffer pitch * @vaddr: XRGB8888 source buffer * @fb: DRM framebuffer * @clip: Clip rectangle area to copy - * @dstclip: Clip destination too. * * Drivers can use this function for RGB888 devices that don't natively * support XRGB8888. * * This function applies clipping on dst, i.e. the destination is a - * full framebuffer but only the clip rect content is copied over. + * full (iomem) framebuffer but only the clip rect content is copied over. */ -void drm_fb_xrgb8888_to_rgb888_dstclip(void *dst, unsigned int dst_pitch, +void drm_fb_xrgb8888_to_rgb888_dstclip(void __iomem *dst, unsigned int dst_pitch, void *vaddr, struct drm_framebuffer *fb, struct drm_rect *clip) { - unsigned int src_offset = (clip->y1 * fb->pitches[0]) - + (clip->x1 * sizeof(u32)); - unsigned int dst_offset = (clip->y1 * dst_pitch) - + (clip->x1 * 3); - size_t src_len = (clip->x2 - clip->x1) * sizeof(u32); - - drm_fb_xrgb8888_to_rgb888_lines(dst + dst_offset, dst_pitch, - vaddr + src_offset, fb->pitches[0], - src_len, clip->y2 - clip->y1); + size_t linepixels = clip->x2 - clip->x1; + size_t dst_len = linepixels * 3; + unsigned y, lines = clip->y2 - clip->y1; + void *dbuf; + + dbuf = kmalloc(dst_len, GFP_KERNEL); + if (!dbuf) + return; + + vaddr += clip_offset(clip, fb->pitches[0], sizeof(u32)); + dst += clip_offset(clip, dst_pitch, sizeof(u16)); + for (y = 0; y < lines; y++) { + drm_fb_xrgb8888_to_rgb888_line(dbuf, vaddr, linepixels); + memcpy_toio(dst, dbuf, dst_len); + vaddr += fb->pitches[0]; + dst += dst_len; + } + + kfree(dbuf); } EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb888_dstclip); diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index 52c0a837a3b2..fae4676707b6 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -646,6 +646,85 @@ void drm_gem_put_pages(struct drm_gem_object *obj, struct page **pages, } EXPORT_SYMBOL(drm_gem_put_pages); +static int objects_lookup(struct drm_file *filp, u32 *handle, int count, + struct drm_gem_object **objs) +{ + int i, ret = 0; + struct drm_gem_object *obj; + + spin_lock(&filp->table_lock); + + for (i = 0; i < count; i++) { + /* Check if we currently have a reference on the object */ + obj = idr_find(&filp->object_idr, handle[i]); + if (!obj) { + ret = -ENOENT; + break; + } + drm_gem_object_get(obj); + objs[i] = obj; + } + spin_unlock(&filp->table_lock); + + return ret; +} + +/** + * drm_gem_objects_lookup - look up GEM objects from an array of handles + * @filp: DRM file private date + * @bo_handles: user pointer to array of userspace handle + * @count: size of handle array + * @objs_out: returned pointer to array of drm_gem_object pointers + * + * Takes an array of userspace handles and returns a newly allocated array of + * GEM objects. + * + * For a single handle lookup, use drm_gem_object_lookup(). + * + * Returns: + * + * @objs filled in with GEM object pointers. Returned GEM objects need to be + * released with drm_gem_object_put(). -ENOENT is returned on a lookup + * failure. 0 is returned on success. + * + */ +int drm_gem_objects_lookup(struct drm_file *filp, void __user *bo_handles, + int count, struct drm_gem_object ***objs_out) +{ + int ret; + u32 *handles; + struct drm_gem_object **objs; + + if (!count) + return 0; + + objs = kvmalloc_array(count, sizeof(struct drm_gem_object *), + GFP_KERNEL | __GFP_ZERO); + if (!objs) + return -ENOMEM; + + handles = kvmalloc_array(count, sizeof(u32), GFP_KERNEL); + if (!handles) { + ret = -ENOMEM; + goto out; + } + + if (copy_from_user(handles, bo_handles, count * sizeof(u32))) { + ret = -EFAULT; + DRM_DEBUG("Failed to copy in GEM handles\n"); + goto out; + } + + ret = objects_lookup(filp, handles, count, objs); + *objs_out = objs; + +out: + kvfree(handles); + return ret; + +} +EXPORT_SYMBOL(drm_gem_objects_lookup); + /** * drm_gem_object_lookup - look up a GEM object from its handle * @filp: DRM file private date @@ -655,21 +734,15 @@ EXPORT_SYMBOL(drm_gem_put_pages); * * A reference to the object named by the handle if such exists on @filp, NULL * otherwise. + * + * If looking up an array of handles, use drm_gem_objects_lookup(). */ struct drm_gem_object * drm_gem_object_lookup(struct drm_file *filp, u32 handle) { - struct drm_gem_object *obj; - - spin_lock(&filp->table_lock); - - /* Check if we currently have a reference on the object */ - obj = idr_find(&filp->object_idr, handle); - if (obj) - drm_gem_object_get(obj); - - spin_unlock(&filp->table_lock); + struct drm_gem_object *obj = NULL; + objects_lookup(filp, &handle, 1, &obj); return obj; } EXPORT_SYMBOL(drm_gem_object_lookup); @@ -1294,3 +1367,96 @@ drm_gem_unlock_reservations(struct drm_gem_object **objs, int count, ww_acquire_fini(acquire_ctx); } EXPORT_SYMBOL(drm_gem_unlock_reservations); + +/** + * drm_gem_fence_array_add - Adds the fence to an array of fences to be + * waited on, deduplicating fences from the same context. + * + * @fence_array array of dma_fence * for the job to block on. + * @fence the dma_fence to add to the list of dependencies. + * + * Returns: + * 0 on success, or an error on failing to expand the array. + */ +int drm_gem_fence_array_add(struct xarray *fence_array, + struct dma_fence *fence) +{ + struct dma_fence *entry; + unsigned long index; + u32 id = 0; + int ret; + + if (!fence) + return 0; + + /* Deduplicate if we already depend on a fence from the same context. + * This lets the size of the array of deps scale with the number of + * engines involved, rather than the number of BOs. + */ + xa_for_each(fence_array, index, entry) { + if (entry->context != fence->context) + continue; + + if (dma_fence_is_later(fence, entry)) { + dma_fence_put(entry); + xa_store(fence_array, index, fence, GFP_KERNEL); + } else { + dma_fence_put(fence); + } + return 0; + } + + ret = xa_alloc(fence_array, &id, fence, xa_limit_32b, GFP_KERNEL); + if (ret != 0) + dma_fence_put(fence); + + return ret; +} +EXPORT_SYMBOL(drm_gem_fence_array_add); + +/** + * drm_gem_fence_array_add_implicit - Adds the implicit dependencies tracked + * in the GEM object's reservation object to an array of dma_fences for use in + * scheduling a rendering job. + * + * This should be called after drm_gem_lock_reservations() on your array of + * GEM objects used in the job but before updating the reservations with your + * own fences. + * + * @fence_array array of dma_fence * for the job to block on. + * @obj the gem object to add new dependencies from. + * @write whether the job might write the object (so we need to depend on + * shared fences in the reservation object). + */ +int drm_gem_fence_array_add_implicit(struct xarray *fence_array, + struct drm_gem_object *obj, + bool write) +{ + int ret; + struct dma_fence **fences; + unsigned int i, fence_count; + + if (!write) { + struct dma_fence *fence = + reservation_object_get_excl_rcu(obj->resv); + + return drm_gem_fence_array_add(fence_array, fence); + } + + ret = reservation_object_get_fences_rcu(obj->resv, NULL, + &fence_count, &fences); + if (ret || !fence_count) + return ret; + + for (i = 0; i < fence_count; i++) { + ret = drm_gem_fence_array_add(fence_array, fences[i]); + if (ret) + break; + } + + for (; i < fence_count; i++) + dma_fence_put(fences[i]); + kfree(fences); + return ret; +} +EXPORT_SYMBOL(drm_gem_fence_array_add_implicit); diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index d337f161909c..ce8a70875bd5 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -508,13 +508,6 @@ int drm_version(struct drm_device *dev, void *data, return err; } -static inline bool -drm_render_driver_and_ioctl(const struct drm_device *dev, u32 flags) -{ - return drm_core_check_feature(dev, DRIVER_RENDER) && - (flags & DRM_RENDER_ALLOW); -} - /** * drm_ioctl_permit - Check ioctl permissions against caller * @@ -529,19 +522,14 @@ drm_render_driver_and_ioctl(const struct drm_device *dev, u32 flags) */ int drm_ioctl_permit(u32 flags, struct drm_file *file_priv) { - const struct drm_device *dev = file_priv->minor->dev; - /* ROOT_ONLY is only for CAP_SYS_ADMIN */ if (unlikely((flags & DRM_ROOT_ONLY) && !capable(CAP_SYS_ADMIN))) return -EACCES; - /* AUTH is only for master ... */ - if (unlikely((flags & DRM_AUTH) && drm_is_primary_client(file_priv))) { - /* authenticated ones, or render capable on DRM_RENDER_ALLOW. */ - if (!file_priv->authenticated && - !drm_render_driver_and_ioctl(dev, flags)) - return -EACCES; - } + /* AUTH is only for authenticated or render client */ + if (unlikely((flags & DRM_AUTH) && !drm_is_render_client(file_priv) && + !file_priv->authenticated)) + return -EACCES; /* MASTER is only for master or control clients */ if (unlikely((flags & DRM_MASTER) && diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c index 4a1c2023ccf0..1a346ae1599d 100644 --- a/drivers/gpu/drm/drm_mode_config.c +++ b/drivers/gpu/drm/drm_mode_config.c @@ -297,8 +297,9 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) return -ENOMEM; dev->mode_config.prop_crtc_id = prop; - prop = drm_property_create(dev, DRM_MODE_PROP_BLOB, "FB_DAMAGE_CLIPS", - 0); + prop = drm_property_create(dev, + DRM_MODE_PROP_ATOMIC | DRM_MODE_PROP_BLOB, + "FB_DAMAGE_CLIPS", 0); if (!prop) return -ENOMEM; dev->mode_config.prop_fb_damage_clips = prop; diff --git a/drivers/gpu/drm/i915/.gitignore b/drivers/gpu/drm/i915/.gitignore new file mode 100644 index 000000000000..cff45d81f42f --- /dev/null +++ b/drivers/gpu/drm/i915/.gitignore @@ -0,0 +1 @@ +header_test_*.c diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 60de05f3fa60..fbcb0904f4a8 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -32,10 +32,13 @@ CFLAGS_intel_fbdev.o = $(call cc-disable-warning, override-init) subdir-ccflags-y += \ $(call as-instr,movntdqa (%eax)$(comma)%xmm0,-DCONFIG_AS_MOVNTDQA) +# Extra header tests +include $(src)/Makefile.header-test + # Please keep these build lists sorted! # core driver code -i915-y := i915_drv.o \ +i915-y += i915_drv.o \ i915_irq.o \ i915_memcpy.o \ i915_mm.o \ @@ -57,15 +60,6 @@ i915-$(CONFIG_COMPAT) += i915_ioc32.o i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o intel_pipe_crc.o i915-$(CONFIG_PERF_EVENTS) += i915_pmu.o -# Test the headers are compilable as standalone units -i915-$(CONFIG_DRM_I915_WERROR) += \ - test_i915_active_types_standalone.o \ - test_i915_gem_context_types_standalone.o \ - test_i915_timeline_types_standalone.o \ - test_intel_context_types_standalone.o \ - test_intel_engine_types_standalone.o \ - test_intel_workarounds_types_standalone.o - # GEM code i915-y += \ i915_active.o \ diff --git a/drivers/gpu/drm/i915/Makefile.header-test b/drivers/gpu/drm/i915/Makefile.header-test new file mode 100644 index 000000000000..c1c391816fa7 --- /dev/null +++ b/drivers/gpu/drm/i915/Makefile.header-test @@ -0,0 +1,47 @@ +# SPDX-License-Identifier: MIT +# Copyright © 2019 Intel Corporation + +# Test the headers are compilable as standalone units +header_test := \ + i915_active_types.h \ + i915_gem_context_types.h \ + i915_priolist_types.h \ + i915_scheduler_types.h \ + i915_timeline_types.h \ + intel_atomic_plane.h \ + intel_audio.h \ + intel_cdclk.h \ + intel_color.h \ + intel_connector.h \ + intel_context_types.h \ + intel_crt.h \ + intel_csr.h \ + intel_ddi.h \ + intel_dp.h \ + intel_dvo.h \ + intel_engine_types.h \ + intel_fbc.h \ + intel_fbdev.h \ + intel_frontbuffer.h \ + intel_hdcp.h \ + intel_hdmi.h \ + intel_lspcon.h \ + intel_lvds.h \ + intel_panel.h \ + intel_pipe_crc.h \ + intel_pm.h \ + intel_psr.h \ + intel_sdvo.h \ + intel_sprite.h \ + intel_tv.h \ + intel_workarounds_types.h + +quiet_cmd_header_test = HDRTEST $@ + cmd_header_test = echo "\#include \"$(<F)\"" > $@ + +header_test_%.c: %.h + $(call cmd,header_test) + +i915-$(CONFIG_DRM_I915_WERROR) += $(foreach h,$(header_test),$(patsubst %.h,header_test_%.o,$(h))) + +clean-files += $(foreach h,$(header_test),$(patsubst %.h,header_test_%.c,$(h))) diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index cf4a1ecf6853..ab002cfd3cab 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -1077,6 +1077,7 @@ static int cmd_handler_pipe_control(struct parser_exec_state *s) bool index_mode = false; unsigned int post_sync; int ret = 0; + u32 hws_pga, val; post_sync = (cmd_val(s, 1) & PIPE_CONTROL_POST_SYNC_OP_MASK) >> 14; @@ -1100,6 +1101,15 @@ static int cmd_handler_pipe_control(struct parser_exec_state *s) index_mode = true; ret |= cmd_address_audit(s, gma, sizeof(u64), index_mode); + if (ret) + return ret; + if (index_mode) { + hws_pga = s->vgpu->hws_pga[s->ring_id]; + gma = hws_pga + gma; + patch_value(s, cmd_ptr(s, 2), gma); + val = cmd_val(s, 1) & (~(1 << 21)); + patch_value(s, cmd_ptr(s, 1), val); + } } } } @@ -1317,8 +1327,14 @@ static int gen8_update_plane_mmio_from_mi_display_flip( info->tile_val << 10); } - vgpu_vreg_t(vgpu, PIPE_FRMCOUNT_G4X(info->pipe))++; - intel_vgpu_trigger_virtual_event(vgpu, info->event); + if (info->plane == PLANE_PRIMARY) + vgpu_vreg_t(vgpu, PIPE_FLIPCOUNT_G4X(info->pipe))++; + + if (info->async_flip) + intel_vgpu_trigger_virtual_event(vgpu, info->event); + else + set_bit(info->event, vgpu->irq.flip_done_event[info->pipe]); + return 0; } @@ -1437,7 +1453,7 @@ static inline int cmd_address_audit(struct parser_exec_state *s, } if (index_mode) { - if (guest_gma >= I915_GTT_PAGE_SIZE / sizeof(u64)) { + if (guest_gma >= I915_GTT_PAGE_SIZE) { ret = -EFAULT; goto err; } @@ -1563,6 +1579,7 @@ static int cmd_handler_mi_flush_dw(struct parser_exec_state *s) unsigned long gma; bool index_mode = false; int ret = 0; + u32 hws_pga, val; /* Check post-sync and ppgtt bit */ if (((cmd_val(s, 0) >> 14) & 0x3) && (cmd_val(s, 1) & (1 << 2))) { @@ -1573,6 +1590,15 @@ static int cmd_handler_mi_flush_dw(struct parser_exec_state *s) if (cmd_val(s, 0) & (1 << 21)) index_mode = true; ret = cmd_address_audit(s, gma, sizeof(u64), index_mode); + if (ret) + return ret; + if (index_mode) { + hws_pga = s->vgpu->hws_pga[s->ring_id]; + gma = hws_pga + gma; + patch_value(s, cmd_ptr(s, 1), gma); + val = cmd_val(s, 0) & (~(1 << 21)); + patch_value(s, cmd_ptr(s, 0), val); + } } /* Check notify bit */ if ((cmd_val(s, 0) & (1 << 8))) diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c index 035479e273be..e1c313da6c00 100644 --- a/drivers/gpu/drm/i915/gvt/display.c +++ b/drivers/gpu/drm/i915/gvt/display.c @@ -407,7 +407,6 @@ static void emulate_vblank_on_pipe(struct intel_vgpu *vgpu, int pipe) if (!pipe_is_enabled(vgpu, pipe)) continue; - vgpu_vreg_t(vgpu, PIPE_FLIPCOUNT_G4X(pipe))++; intel_vgpu_trigger_virtual_event(vgpu, event); } @@ -448,7 +447,7 @@ void intel_gvt_emulate_vblank(struct intel_gvt *gvt) /** * intel_vgpu_emulate_hotplug - trigger hotplug event for vGPU * @vgpu: a vGPU - * @conncted: link state + * @connected: link state * * This function is used to trigger hotplug interrupt for vGPU * diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c index f27edf17b4ab..4e1e425189ba 100644 --- a/drivers/gpu/drm/i915/gvt/dmabuf.c +++ b/drivers/gpu/drm/i915/gvt/dmabuf.c @@ -209,7 +209,7 @@ static int vgpu_get_plane_info(struct drm_device *dev, struct drm_i915_private *dev_priv = to_i915(dev); struct intel_vgpu_primary_plane_format p; struct intel_vgpu_cursor_plane_format c; - int ret; + int ret, tile_height = 1; if (plane_id == DRM_PLANE_TYPE_PRIMARY) { ret = intel_vgpu_decode_primary_plane(vgpu, &p); @@ -228,19 +228,19 @@ static int vgpu_get_plane_info(struct drm_device *dev, break; case PLANE_CTL_TILED_X: info->drm_format_mod = I915_FORMAT_MOD_X_TILED; + tile_height = 8; break; case PLANE_CTL_TILED_Y: info->drm_format_mod = I915_FORMAT_MOD_Y_TILED; + tile_height = 32; break; case PLANE_CTL_TILED_YF: info->drm_format_mod = I915_FORMAT_MOD_Yf_TILED; + tile_height = 32; break; default: gvt_vgpu_err("invalid tiling mode: %x\n", p.tiled); } - - info->size = (((p.stride * p.height * p.bpp) / 8) + - (PAGE_SIZE - 1)) >> PAGE_SHIFT; } else if (plane_id == DRM_PLANE_TYPE_CURSOR) { ret = intel_vgpu_decode_cursor_plane(vgpu, &c); if (ret) @@ -262,14 +262,13 @@ static int vgpu_get_plane_info(struct drm_device *dev, info->x_hot = UINT_MAX; info->y_hot = UINT_MAX; } - - info->size = (((info->stride * c.height * c.bpp) / 8) - + (PAGE_SIZE - 1)) >> PAGE_SHIFT; } else { gvt_vgpu_err("invalid plane id:%d\n", plane_id); return -EINVAL; } + info->size = (info->stride * roundup(info->height, tile_height) + + PAGE_SIZE - 1) >> PAGE_SHIFT; if (info->size == 0) { gvt_vgpu_err("fb size is zero\n"); return -EINVAL; diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c index 1a93472cb34e..f21b8fb5b37e 100644 --- a/drivers/gpu/drm/i915/gvt/execlist.c +++ b/drivers/gpu/drm/i915/gvt/execlist.c @@ -526,12 +526,13 @@ static void init_vgpu_execlist(struct intel_vgpu *vgpu, int ring_id) vgpu_vreg(vgpu, ctx_status_ptr_reg) = ctx_status_ptr.dw; } -static void clean_execlist(struct intel_vgpu *vgpu, unsigned long engine_mask) +static void clean_execlist(struct intel_vgpu *vgpu, + intel_engine_mask_t engine_mask) { - unsigned int tmp; struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; struct intel_engine_cs *engine; struct intel_vgpu_submission *s = &vgpu->submission; + intel_engine_mask_t tmp; for_each_engine_masked(engine, dev_priv, engine_mask, tmp) { kfree(s->ring_scan_buffer[engine->id]); @@ -541,18 +542,18 @@ static void clean_execlist(struct intel_vgpu *vgpu, unsigned long engine_mask) } static void reset_execlist(struct intel_vgpu *vgpu, - unsigned long engine_mask) + intel_engine_mask_t engine_mask) { struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; struct intel_engine_cs *engine; - unsigned int tmp; + intel_engine_mask_t tmp; for_each_engine_masked(engine, dev_priv, engine_mask, tmp) init_vgpu_execlist(vgpu, engine->id); } static int init_execlist(struct intel_vgpu *vgpu, - unsigned long engine_mask) + intel_engine_mask_t engine_mask) { reset_execlist(vgpu, engine_mask); return 0; diff --git a/drivers/gpu/drm/i915/gvt/execlist.h b/drivers/gpu/drm/i915/gvt/execlist.h index 714d709829a2..5ccc2c695848 100644 --- a/drivers/gpu/drm/i915/gvt/execlist.h +++ b/drivers/gpu/drm/i915/gvt/execlist.h @@ -180,6 +180,6 @@ int intel_vgpu_init_execlist(struct intel_vgpu *vgpu); int intel_vgpu_submit_execlist(struct intel_vgpu *vgpu, int ring_id); void intel_vgpu_reset_execlist(struct intel_vgpu *vgpu, - unsigned long engine_mask); + intel_engine_mask_t engine_mask); #endif /*_GVT_EXECLIST_H_*/ diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index c7103dd2d8d5..c2f7d20f6346 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -750,14 +750,20 @@ static void ppgtt_free_spt(struct intel_vgpu_ppgtt_spt *spt) static void ppgtt_free_all_spt(struct intel_vgpu *vgpu) { - struct intel_vgpu_ppgtt_spt *spt; + struct intel_vgpu_ppgtt_spt *spt, *spn; struct radix_tree_iter iter; - void **slot; + LIST_HEAD(all_spt); + void __rcu **slot; + rcu_read_lock(); radix_tree_for_each_slot(slot, &vgpu->gtt.spt_tree, &iter, 0) { spt = radix_tree_deref_slot(slot); - ppgtt_free_spt(spt); + list_move(&spt->post_shadow_list, &all_spt); } + rcu_read_unlock(); + + list_for_each_entry_safe(spt, spn, &all_spt, post_shadow_list) + ppgtt_free_spt(spt); } static int ppgtt_handle_guest_write_page_table_bytes( @@ -1882,7 +1888,11 @@ struct intel_vgpu_mm *intel_vgpu_create_ppgtt_mm(struct intel_vgpu *vgpu, } list_add_tail(&mm->ppgtt_mm.list, &vgpu->gtt.ppgtt_mm_list_head); + + mutex_lock(&gvt->gtt.ppgtt_mm_lock); list_add_tail(&mm->ppgtt_mm.lru_list, &gvt->gtt.ppgtt_mm_lru_list_head); + mutex_unlock(&gvt->gtt.ppgtt_mm_lock); + return mm; } @@ -1942,7 +1952,7 @@ void _intel_vgpu_mm_release(struct kref *mm_ref) */ void intel_vgpu_unpin_mm(struct intel_vgpu_mm *mm) { - atomic_dec(&mm->pincount); + atomic_dec_if_positive(&mm->pincount); } /** @@ -1967,9 +1977,10 @@ int intel_vgpu_pin_mm(struct intel_vgpu_mm *mm) if (ret) return ret; + mutex_lock(&mm->vgpu->gvt->gtt.ppgtt_mm_lock); list_move_tail(&mm->ppgtt_mm.lru_list, &mm->vgpu->gvt->gtt.ppgtt_mm_lru_list_head); - + mutex_unlock(&mm->vgpu->gvt->gtt.ppgtt_mm_lock); } return 0; @@ -1980,6 +1991,8 @@ static int reclaim_one_ppgtt_mm(struct intel_gvt *gvt) struct intel_vgpu_mm *mm; struct list_head *pos, *n; + mutex_lock(&gvt->gtt.ppgtt_mm_lock); + list_for_each_safe(pos, n, &gvt->gtt.ppgtt_mm_lru_list_head) { mm = container_of(pos, struct intel_vgpu_mm, ppgtt_mm.lru_list); @@ -1987,9 +2000,11 @@ static int reclaim_one_ppgtt_mm(struct intel_gvt *gvt) continue; list_del_init(&mm->ppgtt_mm.lru_list); + mutex_unlock(&gvt->gtt.ppgtt_mm_lock); invalidate_ppgtt_mm(mm); return 1; } + mutex_unlock(&gvt->gtt.ppgtt_mm_lock); return 0; } @@ -2489,6 +2504,7 @@ static void clean_spt_oos(struct intel_gvt *gvt) list_for_each_safe(pos, n, >t->oos_page_free_list_head) { oos_page = container_of(pos, struct intel_vgpu_oos_page, list); list_del(&oos_page->list); + free_page((unsigned long)oos_page->mem); kfree(oos_page); } } @@ -2509,6 +2525,12 @@ static int setup_spt_oos(struct intel_gvt *gvt) ret = -ENOMEM; goto fail; } + oos_page->mem = (void *)__get_free_pages(GFP_KERNEL, 0); + if (!oos_page->mem) { + ret = -ENOMEM; + kfree(oos_page); + goto fail; + } INIT_LIST_HEAD(&oos_page->list); INIT_LIST_HEAD(&oos_page->vm_list); @@ -2659,6 +2681,7 @@ int intel_gvt_init_gtt(struct intel_gvt *gvt) } } INIT_LIST_HEAD(&gvt->gtt.ppgtt_mm_lru_list_head); + mutex_init(&gvt->gtt.ppgtt_mm_lock); return 0; } @@ -2699,7 +2722,9 @@ void intel_vgpu_invalidate_ppgtt(struct intel_vgpu *vgpu) list_for_each_safe(pos, n, &vgpu->gtt.ppgtt_mm_list_head) { mm = container_of(pos, struct intel_vgpu_mm, ppgtt_mm.list); if (mm->type == INTEL_GVT_MM_PPGTT) { + mutex_lock(&vgpu->gvt->gtt.ppgtt_mm_lock); list_del_init(&mm->ppgtt_mm.lru_list); + mutex_unlock(&vgpu->gvt->gtt.ppgtt_mm_lock); if (mm->ppgtt_mm.shadowed) invalidate_ppgtt_mm(mm); } diff --git a/drivers/gpu/drm/i915/gvt/gtt.h b/drivers/gpu/drm/i915/gvt/gtt.h index d8cb04cc946d..32c573aea494 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.h +++ b/drivers/gpu/drm/i915/gvt/gtt.h @@ -88,6 +88,7 @@ struct intel_gvt_gtt { void (*mm_free_page_table)(struct intel_vgpu_mm *mm); struct list_head oos_page_use_list_head; struct list_head oos_page_free_list_head; + struct mutex ppgtt_mm_lock; struct list_head ppgtt_mm_lru_list_head; struct page *scratch_page; @@ -221,7 +222,7 @@ struct intel_vgpu_oos_page { struct list_head list; struct list_head vm_list; int id; - unsigned char mem[I915_GTT_PAGE_SIZE]; + void *mem; }; #define GTT_ENTRY_NUM_IN_ONE_PAGE 512 diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h index 8bce09de4b82..f5a328b5290a 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.h +++ b/drivers/gpu/drm/i915/gvt/gvt.h @@ -94,7 +94,6 @@ struct intel_vgpu_fence { struct intel_vgpu_mmio { void *vreg; - void *sreg; }; #define INTEL_GVT_MAX_BAR_NUM 4 @@ -111,11 +110,9 @@ struct intel_vgpu_cfg_space { #define vgpu_cfg_space(vgpu) ((vgpu)->cfg_space.virtual_cfg_space) -#define INTEL_GVT_MAX_PIPE 4 - struct intel_vgpu_irq { bool irq_warn_once[INTEL_GVT_EVENT_MAX]; - DECLARE_BITMAP(flip_done_event[INTEL_GVT_MAX_PIPE], + DECLARE_BITMAP(flip_done_event[I915_MAX_PIPES], INTEL_GVT_EVENT_MAX); }; @@ -144,9 +141,9 @@ enum { struct intel_vgpu_submission_ops { const char *name; - int (*init)(struct intel_vgpu *vgpu, unsigned long engine_mask); - void (*clean)(struct intel_vgpu *vgpu, unsigned long engine_mask); - void (*reset)(struct intel_vgpu *vgpu, unsigned long engine_mask); + int (*init)(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask); + void (*clean)(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask); + void (*reset)(struct intel_vgpu *vgpu, intel_engine_mask_t engine_mask); }; struct intel_vgpu_submission { @@ -449,10 +446,6 @@ void intel_vgpu_write_fence(struct intel_vgpu *vgpu, (*(u64 *)(vgpu->mmio.vreg + i915_mmio_reg_offset(reg))) #define vgpu_vreg64(vgpu, offset) \ (*(u64 *)(vgpu->mmio.vreg + (offset))) -#define vgpu_sreg_t(vgpu, reg) \ - (*(u32 *)(vgpu->mmio.sreg + i915_mmio_reg_offset(reg))) -#define vgpu_sreg(vgpu, offset) \ - (*(u32 *)(vgpu->mmio.sreg + (offset))) #define for_each_active_vgpu(gvt, vgpu, id) \ idr_for_each_entry((&(gvt)->vgpu_idr), (vgpu), (id)) \ @@ -488,7 +481,7 @@ struct intel_vgpu *intel_gvt_create_vgpu(struct intel_gvt *gvt, void intel_gvt_destroy_vgpu(struct intel_vgpu *vgpu); void intel_gvt_release_vgpu(struct intel_vgpu *vgpu); void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr, - unsigned int engine_mask); + intel_engine_mask_t engine_mask); void intel_gvt_reset_vgpu(struct intel_vgpu *vgpu); void intel_gvt_activate_vgpu(struct intel_vgpu *vgpu); void intel_gvt_deactivate_vgpu(struct intel_vgpu *vgpu); diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index dbc749617922..18f01eeb2510 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -311,7 +311,7 @@ static int mul_force_wake_write(struct intel_vgpu *vgpu, static int gdrst_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - unsigned int engine_mask = 0; + intel_engine_mask_t engine_mask = 0; u32 data; write_vreg(vgpu, offset, p_data, bytes); @@ -750,18 +750,19 @@ static int pri_surf_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; - unsigned int index = DSPSURF_TO_PIPE(offset); - i915_reg_t surflive_reg = DSPSURFLIVE(index); - int flip_event[] = { - [PIPE_A] = PRIMARY_A_FLIP_DONE, - [PIPE_B] = PRIMARY_B_FLIP_DONE, - [PIPE_C] = PRIMARY_C_FLIP_DONE, - }; + u32 pipe = DSPSURF_TO_PIPE(offset); + int event = SKL_FLIP_EVENT(pipe, PLANE_PRIMARY); write_vreg(vgpu, offset, p_data, bytes); - vgpu_vreg_t(vgpu, surflive_reg) = vgpu_vreg(vgpu, offset); + vgpu_vreg_t(vgpu, DSPSURFLIVE(pipe)) = vgpu_vreg(vgpu, offset); + + vgpu_vreg_t(vgpu, PIPE_FLIPCOUNT_G4X(pipe))++; + + if (vgpu_vreg_t(vgpu, DSPCNTR(pipe)) & PLANE_CTL_ASYNC_FLIP) + intel_vgpu_trigger_virtual_event(vgpu, event); + else + set_bit(event, vgpu->irq.flip_done_event[pipe]); - set_bit(flip_event[index], vgpu->irq.flip_done_event[index]); return 0; } @@ -771,18 +772,42 @@ static int pri_surf_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, static int spr_surf_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - unsigned int index = SPRSURF_TO_PIPE(offset); - i915_reg_t surflive_reg = SPRSURFLIVE(index); - int flip_event[] = { - [PIPE_A] = SPRITE_A_FLIP_DONE, - [PIPE_B] = SPRITE_B_FLIP_DONE, - [PIPE_C] = SPRITE_C_FLIP_DONE, - }; + u32 pipe = SPRSURF_TO_PIPE(offset); + int event = SKL_FLIP_EVENT(pipe, PLANE_SPRITE0); + + write_vreg(vgpu, offset, p_data, bytes); + vgpu_vreg_t(vgpu, SPRSURFLIVE(pipe)) = vgpu_vreg(vgpu, offset); + + if (vgpu_vreg_t(vgpu, SPRCTL(pipe)) & PLANE_CTL_ASYNC_FLIP) + intel_vgpu_trigger_virtual_event(vgpu, event); + else + set_bit(event, vgpu->irq.flip_done_event[pipe]); + + return 0; +} + +static int reg50080_mmio_write(struct intel_vgpu *vgpu, + unsigned int offset, void *p_data, + unsigned int bytes) +{ + struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + enum pipe pipe = REG_50080_TO_PIPE(offset); + enum plane_id plane = REG_50080_TO_PLANE(offset); + int event = SKL_FLIP_EVENT(pipe, plane); write_vreg(vgpu, offset, p_data, bytes); - vgpu_vreg_t(vgpu, surflive_reg) = vgpu_vreg(vgpu, offset); + if (plane == PLANE_PRIMARY) { + vgpu_vreg_t(vgpu, DSPSURFLIVE(pipe)) = vgpu_vreg(vgpu, offset); + vgpu_vreg_t(vgpu, PIPE_FLIPCOUNT_G4X(pipe))++; + } else { + vgpu_vreg_t(vgpu, SPRSURFLIVE(pipe)) = vgpu_vreg(vgpu, offset); + } + + if ((vgpu_vreg(vgpu, offset) & REG50080_FLIP_TYPE_MASK) == REG50080_FLIP_TYPE_ASYNC) + intel_vgpu_trigger_virtual_event(vgpu, event); + else + set_bit(event, vgpu->irq.flip_done_event[pipe]); - set_bit(flip_event[index], vgpu->irq.flip_done_event[index]); return 0; } @@ -1969,6 +1994,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_DH(DSPSURF(PIPE_A), D_ALL, NULL, pri_surf_mmio_write); MMIO_D(DSPOFFSET(PIPE_A), D_ALL); MMIO_D(DSPSURFLIVE(PIPE_A), D_ALL); + MMIO_DH(REG_50080(PIPE_A, PLANE_PRIMARY), D_ALL, NULL, + reg50080_mmio_write); MMIO_D(DSPCNTR(PIPE_B), D_ALL); MMIO_D(DSPADDR(PIPE_B), D_ALL); @@ -1978,6 +2005,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_DH(DSPSURF(PIPE_B), D_ALL, NULL, pri_surf_mmio_write); MMIO_D(DSPOFFSET(PIPE_B), D_ALL); MMIO_D(DSPSURFLIVE(PIPE_B), D_ALL); + MMIO_DH(REG_50080(PIPE_B, PLANE_PRIMARY), D_ALL, NULL, + reg50080_mmio_write); MMIO_D(DSPCNTR(PIPE_C), D_ALL); MMIO_D(DSPADDR(PIPE_C), D_ALL); @@ -1987,6 +2016,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_DH(DSPSURF(PIPE_C), D_ALL, NULL, pri_surf_mmio_write); MMIO_D(DSPOFFSET(PIPE_C), D_ALL); MMIO_D(DSPSURFLIVE(PIPE_C), D_ALL); + MMIO_DH(REG_50080(PIPE_C, PLANE_PRIMARY), D_ALL, NULL, + reg50080_mmio_write); MMIO_D(SPRCTL(PIPE_A), D_ALL); MMIO_D(SPRLINOFF(PIPE_A), D_ALL); @@ -2000,6 +2031,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_D(SPROFFSET(PIPE_A), D_ALL); MMIO_D(SPRSCALE(PIPE_A), D_ALL); MMIO_D(SPRSURFLIVE(PIPE_A), D_ALL); + MMIO_DH(REG_50080(PIPE_A, PLANE_SPRITE0), D_ALL, NULL, + reg50080_mmio_write); MMIO_D(SPRCTL(PIPE_B), D_ALL); MMIO_D(SPRLINOFF(PIPE_B), D_ALL); @@ -2013,6 +2046,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_D(SPROFFSET(PIPE_B), D_ALL); MMIO_D(SPRSCALE(PIPE_B), D_ALL); MMIO_D(SPRSURFLIVE(PIPE_B), D_ALL); + MMIO_DH(REG_50080(PIPE_B, PLANE_SPRITE0), D_ALL, NULL, + reg50080_mmio_write); MMIO_D(SPRCTL(PIPE_C), D_ALL); MMIO_D(SPRLINOFF(PIPE_C), D_ALL); @@ -2026,6 +2061,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_D(SPROFFSET(PIPE_C), D_ALL); MMIO_D(SPRSCALE(PIPE_C), D_ALL); MMIO_D(SPRSURFLIVE(PIPE_C), D_ALL); + MMIO_DH(REG_50080(PIPE_C, PLANE_SPRITE0), D_ALL, NULL, + reg50080_mmio_write); MMIO_D(HTOTAL(TRANSCODER_A), D_ALL); MMIO_D(HBLANK(TRANSCODER_A), D_ALL); @@ -2827,26 +2864,26 @@ static int init_skl_mmio_info(struct intel_gvt *gvt) MMIO_DH(DBUF_CTL, D_SKL_PLUS, NULL, gen9_dbuf_ctl_mmio_write); - MMIO_D(_MMIO(0xa210), D_SKL_PLUS); + MMIO_D(GEN9_PG_ENABLE, D_SKL_PLUS); MMIO_D(GEN9_MEDIA_PG_IDLE_HYSTERESIS, D_SKL_PLUS); MMIO_D(GEN9_RENDER_PG_IDLE_HYSTERESIS, D_SKL_PLUS); MMIO_DFH(GEN9_GAMT_ECO_REG_RW_IA, D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); - MMIO_DH(_MMIO(0x4ddc), D_SKL_PLUS, NULL, NULL); - MMIO_DH(_MMIO(0x42080), D_SKL_PLUS, NULL, NULL); - MMIO_D(_MMIO(0x45504), D_SKL_PLUS); - MMIO_D(_MMIO(0x45520), D_SKL_PLUS); - MMIO_D(_MMIO(0x46000), D_SKL_PLUS); - MMIO_DH(_MMIO(0x46010), D_SKL_PLUS, NULL, skl_lcpll_write); - MMIO_DH(_MMIO(0x46014), D_SKL_PLUS, NULL, skl_lcpll_write); - MMIO_D(_MMIO(0x6C040), D_SKL_PLUS); - MMIO_D(_MMIO(0x6C048), D_SKL_PLUS); - MMIO_D(_MMIO(0x6C050), D_SKL_PLUS); - MMIO_D(_MMIO(0x6C044), D_SKL_PLUS); - MMIO_D(_MMIO(0x6C04C), D_SKL_PLUS); - MMIO_D(_MMIO(0x6C054), D_SKL_PLUS); - MMIO_D(_MMIO(0x6c058), D_SKL_PLUS); - MMIO_D(_MMIO(0x6c05c), D_SKL_PLUS); - MMIO_DH(_MMIO(0x6c060), D_SKL_PLUS, dpll_status_read, NULL); + MMIO_DH(MMCD_MISC_CTRL, D_SKL_PLUS, NULL, NULL); + MMIO_DH(CHICKEN_PAR1_1, D_SKL_PLUS, NULL, NULL); + MMIO_D(DC_STATE_EN, D_SKL_PLUS); + MMIO_D(DC_STATE_DEBUG, D_SKL_PLUS); + MMIO_D(CDCLK_CTL, D_SKL_PLUS); + MMIO_DH(LCPLL1_CTL, D_SKL_PLUS, NULL, skl_lcpll_write); + MMIO_DH(LCPLL2_CTL, D_SKL_PLUS, NULL, skl_lcpll_write); + MMIO_D(_MMIO(_DPLL1_CFGCR1), D_SKL_PLUS); + MMIO_D(_MMIO(_DPLL2_CFGCR1), D_SKL_PLUS); + MMIO_D(_MMIO(_DPLL3_CFGCR1), D_SKL_PLUS); + MMIO_D(_MMIO(_DPLL1_CFGCR2), D_SKL_PLUS); + MMIO_D(_MMIO(_DPLL2_CFGCR2), D_SKL_PLUS); + MMIO_D(_MMIO(_DPLL3_CFGCR2), D_SKL_PLUS); + MMIO_D(DPLL_CTRL1, D_SKL_PLUS); + MMIO_D(DPLL_CTRL2, D_SKL_PLUS); + MMIO_DH(DPLL_STATUS, D_SKL_PLUS, dpll_status_read, NULL); MMIO_DH(SKL_PS_WIN_POS(PIPE_A, 0), D_SKL_PLUS, NULL, pf_write); MMIO_DH(SKL_PS_WIN_POS(PIPE_A, 1), D_SKL_PLUS, NULL, pf_write); @@ -2965,40 +3002,41 @@ static int init_skl_mmio_info(struct intel_gvt *gvt) MMIO_DH(_MMIO(_REG_701C4(PIPE_C, 3)), D_SKL_PLUS, NULL, NULL); MMIO_DH(_MMIO(_REG_701C4(PIPE_C, 4)), D_SKL_PLUS, NULL, NULL); - MMIO_D(_MMIO(0x70380), D_SKL_PLUS); - MMIO_D(_MMIO(0x71380), D_SKL_PLUS); + MMIO_D(_MMIO(_PLANE_CTL_3_A), D_SKL_PLUS); + MMIO_D(_MMIO(_PLANE_CTL_3_B), D_SKL_PLUS); MMIO_D(_MMIO(0x72380), D_SKL_PLUS); MMIO_D(_MMIO(0x7239c), D_SKL_PLUS); - MMIO_D(_MMIO(0x7039c), D_SKL_PLUS); + MMIO_D(_MMIO(_PLANE_SURF_3_A), D_SKL_PLUS); - MMIO_D(_MMIO(0x8f074), D_SKL_PLUS); - MMIO_D(_MMIO(0x8f004), D_SKL_PLUS); - MMIO_D(_MMIO(0x8f034), D_SKL_PLUS); + MMIO_D(CSR_SSP_BASE, D_SKL_PLUS); + MMIO_D(CSR_HTP_SKL, D_SKL_PLUS); + MMIO_D(CSR_LAST_WRITE, D_SKL_PLUS); - MMIO_D(_MMIO(0xb11c), D_SKL_PLUS); + MMIO_D(BDW_SCRATCH1, D_SKL_PLUS); - MMIO_D(_MMIO(0x51000), D_SKL_PLUS); - MMIO_D(_MMIO(0x6c00c), D_SKL_PLUS); + MMIO_D(SKL_DFSM, D_SKL_PLUS); + MMIO_D(DISPIO_CR_TX_BMU_CR0, D_SKL_PLUS); - MMIO_F(_MMIO(0xc800), 0x7f8, F_CMD_ACCESS, 0, 0, D_SKL_PLUS, + MMIO_F(GEN9_GFX_MOCS(0), 0x7f8, F_CMD_ACCESS, 0, 0, D_SKL_PLUS, NULL, NULL); - MMIO_F(_MMIO(0xb020), 0x80, F_CMD_ACCESS, 0, 0, D_SKL_PLUS, + MMIO_F(GEN7_L3CNTLREG2, 0x80, F_CMD_ACCESS, 0, 0, D_SKL_PLUS, NULL, NULL); MMIO_D(RPM_CONFIG0, D_SKL_PLUS); MMIO_D(_MMIO(0xd08), D_SKL_PLUS); MMIO_D(RC6_LOCATION, D_SKL_PLUS); - MMIO_DFH(_MMIO(0x20e0), D_SKL_PLUS, F_MODE_MASK, NULL, NULL); - MMIO_DFH(_MMIO(0x20ec), D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS, + MMIO_DFH(GEN7_FF_SLICE_CS_CHICKEN1, D_SKL_PLUS, F_MODE_MASK, + NULL, NULL); + MMIO_DFH(GEN9_CS_DEBUG_MODE1, D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL); /* TRTT */ - MMIO_DFH(_MMIO(0x4de0), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); - MMIO_DFH(_MMIO(0x4de4), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); - MMIO_DFH(_MMIO(0x4de8), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); - MMIO_DFH(_MMIO(0x4dec), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); - MMIO_DFH(_MMIO(0x4df0), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); - MMIO_DFH(_MMIO(0x4df4), D_SKL_PLUS, F_CMD_ACCESS, + MMIO_DFH(TRVATTL3PTRDW(0), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); + MMIO_DFH(TRVATTL3PTRDW(1), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); + MMIO_DFH(TRVATTL3PTRDW(2), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); + MMIO_DFH(TRVATTL3PTRDW(3), D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); + MMIO_DFH(TRVADR, D_SKL_PLUS, F_CMD_ACCESS, NULL, NULL); + MMIO_DFH(TRTTE, D_SKL_PLUS, F_CMD_ACCESS, NULL, gen9_trtte_write); MMIO_DH(_MMIO(0x4dfc), D_SKL_PLUS, NULL, gen9_trtt_chicken_write); @@ -3011,7 +3049,7 @@ static int init_skl_mmio_info(struct intel_gvt *gvt) MMIO_DH(DMA_CTRL, D_SKL_PLUS, NULL, dma_ctrl_write); MMIO_D(_MMIO(0x65900), D_SKL_PLUS); - MMIO_D(_MMIO(0x1082c0), D_SKL_PLUS); + MMIO_D(GEN6_STOLEN_RESERVED, D_SKL_PLUS); MMIO_D(_MMIO(0x4068), D_SKL_PLUS); MMIO_D(_MMIO(0x67054), D_SKL_PLUS); MMIO_D(_MMIO(0x6e560), D_SKL_PLUS); @@ -3042,8 +3080,8 @@ static int init_skl_mmio_info(struct intel_gvt *gvt) MMIO_DFH(GEN9_WM_CHICKEN3, D_SKL_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL); - MMIO_D(_MMIO(0x4ab8), D_KBL | D_CFL); - MMIO_D(_MMIO(0x2248), D_SKL_PLUS); + MMIO_D(GAMT_CHKN_BIT_REG, D_KBL); + MMIO_D(GEN9_CTX_PREEMPT_REG, D_KBL | D_SKL); return 0; } @@ -3265,7 +3303,7 @@ void intel_gvt_clean_mmio_info(struct intel_gvt *gvt) /* Special MMIO blocks. */ static struct gvt_mmio_block mmio_blocks[] = { {D_SKL_PLUS, _MMIO(CSR_MMIO_START_RANGE), 0x3000, NULL, NULL}, - {D_ALL, _MMIO(MCHBAR_MIRROR_BASE_SNB), 0x40000, NULL, NULL}, + {D_ALL, MCHBAR_MIRROR_REG_BASE, 0x4000, NULL, NULL}, {D_ALL, _MMIO(VGT_PVINFO_PAGE), VGT_PVINFO_SIZE, pvinfo_mmio_read, pvinfo_mmio_write}, {D_ALL, LGC_PALETTE(PIPE_A, 0), 1024, NULL, NULL}, @@ -3489,12 +3527,11 @@ int intel_vgpu_mmio_reg_rw(struct intel_vgpu *vgpu, unsigned int offset, return mmio_info->read(vgpu, offset, pdata, bytes); else { u64 ro_mask = mmio_info->ro_mask; - u32 old_vreg = 0, old_sreg = 0; + u32 old_vreg = 0; u64 data = 0; if (intel_gvt_mmio_has_mode_mask(gvt, mmio_info->offset)) { old_vreg = vgpu_vreg(vgpu, offset); - old_sreg = vgpu_sreg(vgpu, offset); } if (likely(!ro_mask)) @@ -3516,8 +3553,6 @@ int intel_vgpu_mmio_reg_rw(struct intel_vgpu *vgpu, unsigned int offset, vgpu_vreg(vgpu, offset) = (old_vreg & ~mask) | (vgpu_vreg(vgpu, offset) & mask); - vgpu_sreg(vgpu, offset) = (old_sreg & ~mask) - | (vgpu_sreg(vgpu, offset) & mask); } } diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index d5fcc447d22f..a68addf95c23 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -905,7 +905,7 @@ static inline bool intel_vgpu_in_aperture(struct intel_vgpu *vgpu, u64 off) static int intel_vgpu_aperture_rw(struct intel_vgpu *vgpu, u64 off, void *buf, unsigned long count, bool is_write) { - void *aperture_va; + void __iomem *aperture_va; if (!intel_vgpu_in_aperture(vgpu, off) || !intel_vgpu_in_aperture(vgpu, off + count)) { @@ -920,9 +920,9 @@ static int intel_vgpu_aperture_rw(struct intel_vgpu *vgpu, u64 off, return -EIO; if (is_write) - memcpy(aperture_va + offset_in_page(off), buf, count); + memcpy_toio(aperture_va + offset_in_page(off), buf, count); else - memcpy(buf, aperture_va + offset_in_page(off), count); + memcpy_fromio(buf, aperture_va + offset_in_page(off), count); io_mapping_unmap(aperture_va); diff --git a/drivers/gpu/drm/i915/gvt/mmio.c b/drivers/gpu/drm/i915/gvt/mmio.c index ed4df2f6d60b..a55178884d67 100644 --- a/drivers/gpu/drm/i915/gvt/mmio.c +++ b/drivers/gpu/drm/i915/gvt/mmio.c @@ -239,7 +239,6 @@ void intel_vgpu_reset_mmio(struct intel_vgpu *vgpu, bool dmlr) if (dmlr) { memcpy(vgpu->mmio.vreg, mmio, info->mmio_size); - memcpy(vgpu->mmio.sreg, mmio, info->mmio_size); vgpu_vreg_t(vgpu, GEN6_GT_THREAD_STATUS_REG) = 0; @@ -280,7 +279,6 @@ void intel_vgpu_reset_mmio(struct intel_vgpu *vgpu, bool dmlr) * touched */ memcpy(vgpu->mmio.vreg, mmio, GVT_GEN8_MMIO_RESET_OFFSET); - memcpy(vgpu->mmio.sreg, mmio, GVT_GEN8_MMIO_RESET_OFFSET); } } @@ -296,12 +294,10 @@ int intel_vgpu_init_mmio(struct intel_vgpu *vgpu) { const struct intel_gvt_device_info *info = &vgpu->gvt->device_info; - vgpu->mmio.vreg = vzalloc(array_size(info->mmio_size, 2)); + vgpu->mmio.vreg = vzalloc(info->mmio_size); if (!vgpu->mmio.vreg) return -ENOMEM; - vgpu->mmio.sreg = vgpu->mmio.vreg + info->mmio_size; - intel_vgpu_reset_mmio(vgpu, true); return 0; @@ -315,5 +311,5 @@ int intel_vgpu_init_mmio(struct intel_vgpu *vgpu) void intel_vgpu_clean_mmio(struct intel_vgpu *vgpu) { vfree(vgpu->mmio.vreg); - vgpu->mmio.vreg = vgpu->mmio.sreg = NULL; + vgpu->mmio.vreg = NULL; } diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c index 76630fbe51b6..e7e14c842be4 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.c +++ b/drivers/gpu/drm/i915/gvt/mmio_context.c @@ -68,7 +68,7 @@ static struct engine_mmio gen8_engine_mmio_list[] __cacheline_aligned = { {BCS0, RING_MI_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2209c */ {BCS0, RING_INSTPM(BLT_RING_BASE), 0xffff, false}, /* 0x220c0 */ {BCS0, RING_HWSTAM(BLT_RING_BASE), 0x0, false}, /* 0x22098 */ - {BCS0, RING_EXCC(BLT_RING_BASE), 0x0, false}, /* 0x22028 */ + {BCS0, RING_EXCC(BLT_RING_BASE), 0xffff, false}, /* 0x22028 */ {RCS0, INVALID_MMIO_REG, 0, false } /* Terminated */ }; @@ -119,7 +119,7 @@ static struct engine_mmio gen9_engine_mmio_list[] __cacheline_aligned = { {BCS0, RING_MI_MODE(BLT_RING_BASE), 0xffff, false}, /* 0x2209c */ {BCS0, RING_INSTPM(BLT_RING_BASE), 0xffff, false}, /* 0x220c0 */ {BCS0, RING_HWSTAM(BLT_RING_BASE), 0x0, false}, /* 0x22098 */ - {BCS0, RING_EXCC(BLT_RING_BASE), 0x0, false}, /* 0x22028 */ + {BCS0, RING_EXCC(BLT_RING_BASE), 0xffff, false}, /* 0x22028 */ {VCS1, RING_EXCC(GEN8_BSD2_RING_BASE), 0xffff, false}, /* 0x1c028 */ diff --git a/drivers/gpu/drm/i915/gvt/reg.h b/drivers/gpu/drm/i915/gvt/reg.h index 428d252344f1..3de5b643b266 100644 --- a/drivers/gpu/drm/i915/gvt/reg.h +++ b/drivers/gpu/drm/i915/gvt/reg.h @@ -60,6 +60,37 @@ #define _REG_701C0(pipe, plane) (0x701c0 + pipe * 0x1000 + (plane - 1) * 0x100) #define _REG_701C4(pipe, plane) (0x701c4 + pipe * 0x1000 + (plane - 1) * 0x100) +#define SKL_FLIP_EVENT(pipe, plane) (PRIMARY_A_FLIP_DONE + (plane) * 3 + (pipe)) + +#define PLANE_CTL_ASYNC_FLIP (1 << 9) +#define REG50080_FLIP_TYPE_MASK 0x3 +#define REG50080_FLIP_TYPE_ASYNC 0x1 + +#define REG_50080(_pipe, _plane) ({ \ + typeof(_pipe) (p) = (_pipe); \ + typeof(_plane) (q) = (_plane); \ + (((p) == PIPE_A) ? (((q) == PLANE_PRIMARY) ? (_MMIO(0x50080)) : \ + (_MMIO(0x50090))) : \ + (((p) == PIPE_B) ? (((q) == PLANE_PRIMARY) ? (_MMIO(0x50088)) : \ + (_MMIO(0x50098))) : \ + (((p) == PIPE_C) ? (((q) == PLANE_PRIMARY) ? (_MMIO(0x5008C)) : \ + (_MMIO(0x5009C))) : \ + (_MMIO(0x50080))))); }) + +#define REG_50080_TO_PIPE(_reg) ({ \ + typeof(_reg) (reg) = (_reg); \ + (((reg) == 0x50080 || (reg) == 0x50090) ? (PIPE_A) : \ + (((reg) == 0x50088 || (reg) == 0x50098) ? (PIPE_B) : \ + (((reg) == 0x5008C || (reg) == 0x5009C) ? (PIPE_C) : \ + (INVALID_PIPE)))); }) + +#define REG_50080_TO_PLANE(_reg) ({ \ + typeof(_reg) (reg) = (_reg); \ + (((reg) == 0x50080 || (reg) == 0x50088 || (reg) == 0x5008C) ? \ + (PLANE_PRIMARY) : \ + (((reg) == 0x50090 || (reg) == 0x50098 || (reg) == 0x5009C) ? \ + (PLANE_SPRITE0) : (I915_MAX_PLANES))); }) + #define GFX_MODE_BIT_SET_IN_MASK(val, bit) \ ((((bit) & 0xffff0000) == 0) && !!((val) & (((bit) << 16)))) @@ -95,4 +126,7 @@ #define RING_GFX_MODE(base) _MMIO((base) + 0x29c) #define VF_GUARDBAND _MMIO(0x83a4) +/* define the effective range of MCHBAR register on Sandybridge+ */ +#define MCHBAR_MIRROR_REG_BASE _MMIO(MCHBAR_MIRROR_BASE_SNB + 0x4000) + #endif diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index 3faf2438b9bc..8998fa5ab198 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -346,7 +346,7 @@ static int set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload, int i = 0; if (mm->type != INTEL_GVT_MM_PPGTT || !mm->ppgtt_mm.shadowed) - return -1; + return -EINVAL; if (mm->ppgtt_mm.root_entry_type == GTT_TYPE_PPGTT_ROOT_L4_ENTRY) { px_dma(&ppgtt->pml4) = mm->ppgtt_mm.shadow_pdps[0]; @@ -410,12 +410,6 @@ int intel_gvt_scan_and_shadow_workload(struct intel_vgpu_workload *workload) if (workload->shadow) return 0; - ret = set_context_ppgtt_from_shadow(workload, shadow_ctx); - if (ret < 0) { - gvt_vgpu_err("workload shadow ppgtt isn't ready\n"); - return ret; - } - /* pin shadow context by gvt even the shadow context will be pinned * when i915 alloc request. That is because gvt will update the guest * context from shadow context when workload is completed, and at that @@ -677,6 +671,9 @@ static int dispatch_workload(struct intel_vgpu_workload *workload) { struct intel_vgpu *vgpu = workload->vgpu; struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; + struct intel_vgpu_submission *s = &vgpu->submission; + struct i915_gem_context *shadow_ctx = s->shadow_ctx; + struct i915_request *rq; int ring_id = workload->ring_id; int ret; @@ -686,6 +683,12 @@ static int dispatch_workload(struct intel_vgpu_workload *workload) mutex_lock(&vgpu->vgpu_lock); mutex_lock(&dev_priv->drm.struct_mutex); + ret = set_context_ppgtt_from_shadow(workload, shadow_ctx); + if (ret < 0) { + gvt_vgpu_err("workload shadow ppgtt isn't ready\n"); + goto err_req; + } + ret = intel_gvt_workload_req_alloc(workload); if (ret) goto err_req; @@ -702,6 +705,14 @@ static int dispatch_workload(struct intel_vgpu_workload *workload) ret = prepare_workload(workload); out: + if (ret) { + /* We might still need to add request with + * clean ctx to retire it properly.. + */ + rq = fetch_and_zero(&workload->req); + i915_request_put(rq); + } + if (!IS_ERR_OR_NULL(workload->req)) { gvt_dbg_sched("ring id %d submit workload to i915 %p\n", ring_id, workload->req); @@ -738,7 +749,8 @@ static struct intel_vgpu_workload *pick_next_workload( goto out; } - if (list_empty(workload_q_head(scheduler->current_vgpu, ring_id))) + if (!scheduler->current_vgpu->active || + list_empty(workload_q_head(scheduler->current_vgpu, ring_id))) goto out; /* @@ -838,13 +850,13 @@ static void update_guest_context(struct intel_vgpu_workload *workload) } void intel_vgpu_clean_workloads(struct intel_vgpu *vgpu, - unsigned long engine_mask) + intel_engine_mask_t engine_mask) { struct intel_vgpu_submission *s = &vgpu->submission; struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; struct intel_engine_cs *engine; struct intel_vgpu_workload *pos, *n; - unsigned int tmp; + intel_engine_mask_t tmp; /* free the unsubmited workloads in the queues. */ for_each_engine_masked(engine, dev_priv, engine_mask, tmp) { @@ -1137,7 +1149,7 @@ void intel_vgpu_clean_submission(struct intel_vgpu *vgpu) * */ void intel_vgpu_reset_submission(struct intel_vgpu *vgpu, - unsigned long engine_mask) + intel_engine_mask_t engine_mask) { struct intel_vgpu_submission *s = &vgpu->submission; @@ -1227,7 +1239,7 @@ out_shadow_ctx: * */ int intel_vgpu_select_submission_ops(struct intel_vgpu *vgpu, - unsigned long engine_mask, + intel_engine_mask_t engine_mask, unsigned int interface) { struct intel_vgpu_submission *s = &vgpu->submission; @@ -1473,8 +1485,9 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, intel_runtime_pm_put_unchecked(dev_priv); } - if (ret && (vgpu_is_vm_unhealthy(ret))) { - enter_failsafe_mode(vgpu, GVT_FAILSAFE_GUEST_ERR); + if (ret) { + if (vgpu_is_vm_unhealthy(ret)) + enter_failsafe_mode(vgpu, GVT_FAILSAFE_GUEST_ERR); intel_vgpu_destroy_workload(workload); return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h index 0635b2c4bed7..90c6756f5453 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.h +++ b/drivers/gpu/drm/i915/gvt/scheduler.h @@ -142,12 +142,12 @@ void intel_gvt_wait_vgpu_idle(struct intel_vgpu *vgpu); int intel_vgpu_setup_submission(struct intel_vgpu *vgpu); void intel_vgpu_reset_submission(struct intel_vgpu *vgpu, - unsigned long engine_mask); + intel_engine_mask_t engine_mask); void intel_vgpu_clean_submission(struct intel_vgpu *vgpu); int intel_vgpu_select_submission_ops(struct intel_vgpu *vgpu, - unsigned long engine_mask, + intel_engine_mask_t engine_mask, unsigned int interface); extern const struct intel_vgpu_submission_ops @@ -160,6 +160,6 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id, void intel_vgpu_destroy_workload(struct intel_vgpu_workload *workload); void intel_vgpu_clean_workloads(struct intel_vgpu *vgpu, - unsigned long engine_mask); + intel_engine_mask_t engine_mask); #endif diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c index 314e40121e47..44ce3c2b9ac1 100644 --- a/drivers/gpu/drm/i915/gvt/vgpu.c +++ b/drivers/gpu/drm/i915/gvt/vgpu.c @@ -526,11 +526,11 @@ struct intel_vgpu *intel_gvt_create_vgpu(struct intel_gvt *gvt, * GPU engines. For FLR, engine_mask is ignored. */ void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr, - unsigned int engine_mask) + intel_engine_mask_t engine_mask) { struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_workload_scheduler *scheduler = &gvt->scheduler; - unsigned int resetting_eng = dmlr ? ALL_ENGINES : engine_mask; + intel_engine_mask_t resetting_eng = dmlr ? ALL_ENGINES : engine_mask; gvt_dbg_core("------------------------------------------\n"); gvt_dbg_core("resseting vgpu%d, dmlr %d, engine_mask %08x\n", diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 652f65d2e131..5823ffb17821 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -26,14 +26,21 @@ * */ -#include <linux/sort.h> #include <linux/sched/mm.h> +#include <linux/sort.h> + #include <drm/drm_debugfs.h> #include <drm/drm_fourcc.h> -#include "intel_drv.h" -#include "intel_guc_submission.h" #include "i915_reset.h" +#include "intel_dp.h" +#include "intel_drv.h" +#include "intel_fbc.h" +#include "intel_guc_submission.h" +#include "intel_hdcp.h" +#include "intel_hdmi.h" +#include "intel_pm.h" +#include "intel_psr.h" static inline struct drm_i915_private *node_to_i915(struct drm_info_node *node) { @@ -826,11 +833,11 @@ static int i915_interrupt_info(struct seq_file *m, void *data) } else if (!HAS_PCH_SPLIT(dev_priv)) { seq_printf(m, "Interrupt enable: %08x\n", - I915_READ(IER)); + I915_READ(GEN2_IER)); seq_printf(m, "Interrupt identity: %08x\n", - I915_READ(IIR)); + I915_READ(GEN2_IIR)); seq_printf(m, "Interrupt mask: %08x\n", - I915_READ(IMR)); + I915_READ(GEN2_IMR)); for_each_pipe(dev_priv, pipe) seq_printf(m, "Pipe %c stat: %08x\n", pipe_name(pipe), @@ -2087,8 +2094,8 @@ static int i915_llc(struct seq_file *m, void *data) const bool edram = INTEL_GEN(dev_priv) > 8; seq_printf(m, "LLC: %s\n", yesno(HAS_LLC(dev_priv))); - seq_printf(m, "%s: %lluMB\n", edram ? "eDRAM" : "eLLC", - intel_uncore_edram_size(dev_priv)/1024/1024); + seq_printf(m, "%s: %uMB\n", edram ? "eDRAM" : "eLLC", + dev_priv->edram_size_mb); return 0; } @@ -2245,7 +2252,7 @@ static int i915_guc_stage_pool(struct seq_file *m, void *data) const struct intel_guc *guc = &dev_priv->guc; struct guc_stage_desc *desc = guc->stage_desc_pool_vaddr; struct intel_guc_client *client = guc->execbuf_client; - unsigned int tmp; + intel_engine_mask_t tmp; int index; if (!USES_GUC_SUBMISSION(dev_priv)) @@ -4786,7 +4793,10 @@ static int i915_dsc_fec_support_show(struct seq_file *m, void *data) ret = drm_modeset_lock(&dev->mode_config.connection_mutex, &ctx); if (ret) { - ret = -EINTR; + if (ret == -EDEADLK && !drm_modeset_backoff(&ctx)) { + try_again = true; + continue; + } break; } crtc = connector->state->crtc; @@ -4811,6 +4821,8 @@ static int i915_dsc_fec_support_show(struct seq_file *m, void *data) yesno(crtc_state->dsc_params.compression_enable)); seq_printf(m, "DSC_Sink_Support: %s\n", yesno(drm_dp_sink_supports_dsc(intel_dp->dsc_dpcd))); + seq_printf(m, "Force_DSC_Enable: %s\n", + yesno(intel_dp->force_dsc_en)); if (!intel_dp_is_edp(intel_dp)) seq_printf(m, "FEC_Sink_Support: %s\n", yesno(drm_dp_sink_supports_fec(intel_dp->fec_capable))); diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index bbe1a5d56480..1ad88e6d7c04 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -48,12 +48,19 @@ #include <drm/i915_drm.h> #include "i915_drv.h" -#include "i915_trace.h" #include "i915_pmu.h" -#include "i915_reset.h" #include "i915_query.h" +#include "i915_reset.h" +#include "i915_trace.h" #include "i915_vgpu.h" +#include "intel_audio.h" +#include "intel_cdclk.h" +#include "intel_csr.h" +#include "intel_dp.h" #include "intel_drv.h" +#include "intel_fbdev.h" +#include "intel_pm.h" +#include "intel_sprite.h" #include "intel_uc.h" #include "intel_workarounds.h" @@ -868,10 +875,13 @@ static int i915_driver_init_early(struct drm_i915_private *dev_priv) if (i915_inject_load_failure()) return -ENODEV; + intel_device_info_subplatform_init(dev_priv); + + intel_uncore_init_early(&dev_priv->uncore); + spin_lock_init(&dev_priv->irq_lock); spin_lock_init(&dev_priv->gpu_error.lock); mutex_init(&dev_priv->backlight_lock); - spin_lock_init(&dev_priv->uncore.lock); mutex_init(&dev_priv->sb_lock); mutex_init(&dev_priv->av_mutex); @@ -954,7 +964,7 @@ static int i915_driver_init_mmio(struct drm_i915_private *dev_priv) if (i915_get_bridge_dev(dev_priv)) return -EIO; - ret = intel_uncore_init(&dev_priv->uncore); + ret = intel_uncore_init_mmio(&dev_priv->uncore); if (ret < 0) goto err_bridge; @@ -963,7 +973,7 @@ static int i915_driver_init_mmio(struct drm_i915_private *dev_priv) intel_device_info_init_mmio(dev_priv); - intel_uncore_prune(&dev_priv->uncore); + intel_uncore_prune_mmio_domains(&dev_priv->uncore); intel_uc_init_mmio(dev_priv); @@ -977,7 +987,7 @@ static int i915_driver_init_mmio(struct drm_i915_private *dev_priv) err_uncore: intel_teardown_mchbar(dev_priv); - intel_uncore_fini(&dev_priv->uncore); + intel_uncore_fini_mmio(&dev_priv->uncore); err_bridge: pci_dev_put(dev_priv->bridge_dev); @@ -991,7 +1001,7 @@ err_bridge: static void i915_driver_cleanup_mmio(struct drm_i915_private *dev_priv) { intel_teardown_mchbar(dev_priv); - intel_uncore_fini(&dev_priv->uncore); + intel_uncore_fini_mmio(&dev_priv->uncore); pci_dev_put(dev_priv->bridge_dev); } @@ -1441,6 +1451,45 @@ intel_get_dram_info(struct drm_i915_private *dev_priv) dram_info->ranks, yesno(dram_info->is_16gb_dimm)); } +static u32 gen9_edram_size_mb(struct drm_i915_private *dev_priv, u32 cap) +{ + const unsigned int ways[8] = { 4, 8, 12, 16, 16, 16, 16, 16 }; + const unsigned int sets[4] = { 1, 1, 2, 2 }; + + return EDRAM_NUM_BANKS(cap) * + ways[EDRAM_WAYS_IDX(cap)] * + sets[EDRAM_SETS_IDX(cap)]; +} + +static void edram_detect(struct drm_i915_private *dev_priv) +{ + u32 edram_cap = 0; + + if (!(IS_HASWELL(dev_priv) || + IS_BROADWELL(dev_priv) || + INTEL_GEN(dev_priv) >= 9)) + return; + + edram_cap = __raw_uncore_read32(&dev_priv->uncore, HSW_EDRAM_CAP); + + /* NB: We can't write IDICR yet because we don't have gt funcs set up */ + + if (!(edram_cap & EDRAM_ENABLED)) + return; + + /* + * The needed capability bits for size calculation are not there with + * pre gen9 so return 128MB always. + */ + if (INTEL_GEN(dev_priv) < 9) + dev_priv->edram_size_mb = 128; + else + dev_priv->edram_size_mb = + gen9_edram_size_mb(dev_priv, edram_cap); + + DRM_INFO("Found %uMB of eDRAM\n", dev_priv->edram_size_mb); +} + /** * i915_driver_init_hw - setup state requiring device access * @dev_priv: device private @@ -1483,6 +1532,9 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv) intel_sanitize_options(dev_priv); + /* needs to be done before ggtt probe */ + edram_detect(dev_priv); + i915_perf_init(dev_priv); ret = i915_ggtt_probe_hw(dev_priv); @@ -1718,10 +1770,12 @@ static void i915_welcome_messages(struct drm_i915_private *dev_priv) if (drm_debug & DRM_UT_DRIVER) { struct drm_printer p = drm_debug_printer("i915 device info:"); - drm_printf(&p, "pciid=0x%04x rev=0x%02x platform=%s gen=%i\n", + drm_printf(&p, "pciid=0x%04x rev=0x%02x platform=%s (subplatform=0x%x) gen=%i\n", INTEL_DEVID(dev_priv), INTEL_REVID(dev_priv), intel_platform_name(INTEL_INFO(dev_priv)->platform), + intel_subplatform(RUNTIME_INFO(dev_priv), + INTEL_INFO(dev_priv)->platform), INTEL_GEN(dev_priv)); intel_device_info_dump_flags(INTEL_INFO(dev_priv), &p); @@ -1764,8 +1818,6 @@ i915_driver_create(struct pci_dev *pdev, const struct pci_device_id *ent) memcpy(device_info, match_info, sizeof(*device_info)); RUNTIME_INFO(i915)->device_id = pdev->device; - BUILD_BUG_ON(INTEL_MAX_PLATFORMS > - BITS_PER_TYPE(device_info->platform_mask)); BUG_ON(device_info->gen > BITS_PER_TYPE(device_info->gen_mask)); return i915; @@ -1862,6 +1914,13 @@ void i915_driver_unload(struct drm_device *dev) i915_driver_unregister(dev_priv); + /* + * After unregistering the device to prevent any new users, cancel + * all in-flight requests so that we can quickly unbind the active + * resources. + */ + i915_gem_set_wedged(dev_priv); + /* Flush any external code that still may be under the RCU lock */ synchronize_rcu(); diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 25c264e55d3c..066fd2a12851 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -66,13 +66,14 @@ #include "intel_device_info.h" #include "intel_display.h" #include "intel_dpll_mgr.h" +#include "intel_frontbuffer.h" #include "intel_lrc.h" #include "intel_opregion.h" #include "intel_ringbuffer.h" +#include "intel_uc.h" #include "intel_uncore.h" #include "intel_wopcm.h" #include "intel_workarounds.h" -#include "intel_uc.h" #include "i915_gem.h" #include "i915_gem_context.h" @@ -92,8 +93,8 @@ #define DRIVER_NAME "i915" #define DRIVER_DESC "Intel Graphics" -#define DRIVER_DATE "20190328" -#define DRIVER_TIMESTAMP 1553776914 +#define DRIVER_DATE "20190417" +#define DRIVER_TIMESTAMP 1555492067 /* Use I915_STATE_WARN(x) and I915_STATE_WARN_ON() (rather than WARN() and * WARN_ON()) for hw state sanity checks to check for unexpected conditions @@ -282,7 +283,8 @@ struct drm_i915_display_funcs { void (*get_cdclk)(struct drm_i915_private *dev_priv, struct intel_cdclk_state *cdclk_state); void (*set_cdclk)(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state); + const struct intel_cdclk_state *cdclk_state, + enum pipe pipe); int (*get_fifo_size)(struct drm_i915_private *dev_priv, enum i9xx_plane_id i9xx_plane); int (*compute_pipe_wm)(struct intel_crtc_state *cstate); @@ -325,6 +327,7 @@ struct drm_i915_display_funcs { /* display clock increase/decrease */ /* pll clock increase/decrease */ + int (*color_check)(struct intel_crtc_state *crtc_state); /* * Program double buffered color management registers during * vblank evasion. The registers should then latch during the @@ -373,14 +376,6 @@ enum i915_cache_level { #define I915_COLOR_UNEVICTABLE (-1) /* a non-vma sharing the address space */ -enum fb_op_origin { - ORIGIN_GTT, - ORIGIN_CPU, - ORIGIN_CS, - ORIGIN_FLIP, - ORIGIN_DIRTYFB, -}; - struct intel_fbc { /* This is always the inner lock when overlapping with struct_mutex and * it's the outer lock when overlapping with stolen_lock. */ @@ -1628,6 +1623,8 @@ struct drm_i915_private { struct intel_cdclk_state actual; /* The current hardware cdclk state */ struct intel_cdclk_state hw; + + int force_min_cdclk; } cdclk; /** @@ -1706,8 +1703,11 @@ struct drm_i915_private { struct intel_l3_parity l3_parity; - /* Cannot be determined by PCIID. You must always read a register. */ - u32 edram_cap; + /* + * edram size in MB. + * Cannot be determined by PCIID. You must always read a register. + */ + u32 edram_size_mb; /* * Protects RPS/RC6 register access and PCU communication. @@ -1747,6 +1747,7 @@ struct drm_i915_private { * */ struct mutex av_mutex; + int audio_power_refcount; struct { struct mutex mutex; @@ -1994,7 +1995,6 @@ struct drm_i915_private { /* Abstract the submission mechanism (legacy ringbuffer or execlists) away */ struct { - void (*resume)(struct drm_i915_private *); void (*cleanup_engine)(struct intel_engine_cs *engine); struct i915_gt_timelines { @@ -2298,7 +2298,69 @@ static inline unsigned int i915_sg_segment_size(void) #define IS_REVID(p, since, until) \ (INTEL_REVID(p) >= (since) && INTEL_REVID(p) <= (until)) -#define IS_PLATFORM(dev_priv, p) (INTEL_INFO(dev_priv)->platform_mask & BIT(p)) +static __always_inline unsigned int +__platform_mask_index(const struct intel_runtime_info *info, + enum intel_platform p) +{ + const unsigned int pbits = + BITS_PER_TYPE(info->platform_mask[0]) - INTEL_SUBPLATFORM_BITS; + + /* Expand the platform_mask array if this fails. */ + BUILD_BUG_ON(INTEL_MAX_PLATFORMS > + pbits * ARRAY_SIZE(info->platform_mask)); + + return p / pbits; +} + +static __always_inline unsigned int +__platform_mask_bit(const struct intel_runtime_info *info, + enum intel_platform p) +{ + const unsigned int pbits = + BITS_PER_TYPE(info->platform_mask[0]) - INTEL_SUBPLATFORM_BITS; + + return p % pbits + INTEL_SUBPLATFORM_BITS; +} + +static inline u32 +intel_subplatform(const struct intel_runtime_info *info, enum intel_platform p) +{ + const unsigned int pi = __platform_mask_index(info, p); + + return info->platform_mask[pi] & INTEL_SUBPLATFORM_BITS; +} + +static __always_inline bool +IS_PLATFORM(const struct drm_i915_private *i915, enum intel_platform p) +{ + const struct intel_runtime_info *info = RUNTIME_INFO(i915); + const unsigned int pi = __platform_mask_index(info, p); + const unsigned int pb = __platform_mask_bit(info, p); + + BUILD_BUG_ON(!__builtin_constant_p(p)); + + return info->platform_mask[pi] & BIT(pb); +} + +static __always_inline bool +IS_SUBPLATFORM(const struct drm_i915_private *i915, + enum intel_platform p, unsigned int s) +{ + const struct intel_runtime_info *info = RUNTIME_INFO(i915); + const unsigned int pi = __platform_mask_index(info, p); + const unsigned int pb = __platform_mask_bit(info, p); + const unsigned int msb = BITS_PER_TYPE(info->platform_mask[0]) - 1; + const u32 mask = info->platform_mask[pi]; + + BUILD_BUG_ON(!__builtin_constant_p(p)); + BUILD_BUG_ON(!__builtin_constant_p(s)); + BUILD_BUG_ON((s) >= INTEL_SUBPLATFORM_BITS); + + /* Shift and test on the MSB position so sign flag can be used. */ + return ((mask << (msb - pb)) & (mask << (msb - s))) & BIT(msb); +} + +#define IS_MOBILE(dev_priv) (INTEL_INFO(dev_priv)->is_mobile) #define IS_I830(dev_priv) IS_PLATFORM(dev_priv, INTEL_I830) #define IS_I845G(dev_priv) IS_PLATFORM(dev_priv, INTEL_I845G) @@ -2313,11 +2375,11 @@ static inline unsigned int i915_sg_segment_size(void) #define IS_G45(dev_priv) IS_PLATFORM(dev_priv, INTEL_G45) #define IS_GM45(dev_priv) IS_PLATFORM(dev_priv, INTEL_GM45) #define IS_G4X(dev_priv) (IS_G45(dev_priv) || IS_GM45(dev_priv)) -#define IS_PINEVIEW_G(dev_priv) (INTEL_DEVID(dev_priv) == 0xa001) -#define IS_PINEVIEW_M(dev_priv) (INTEL_DEVID(dev_priv) == 0xa011) #define IS_PINEVIEW(dev_priv) IS_PLATFORM(dev_priv, INTEL_PINEVIEW) #define IS_G33(dev_priv) IS_PLATFORM(dev_priv, INTEL_G33) -#define IS_IRONLAKE_M(dev_priv) (INTEL_DEVID(dev_priv) == 0x0046) +#define IS_IRONLAKE(dev_priv) IS_PLATFORM(dev_priv, INTEL_IRONLAKE) +#define IS_IRONLAKE_M(dev_priv) \ + (IS_PLATFORM(dev_priv, INTEL_IRONLAKE) && IS_MOBILE(dev_priv)) #define IS_IVYBRIDGE(dev_priv) IS_PLATFORM(dev_priv, INTEL_IVYBRIDGE) #define IS_IVB_GT1(dev_priv) (IS_IVYBRIDGE(dev_priv) && \ INTEL_INFO(dev_priv)->gt == 1) @@ -2333,46 +2395,34 @@ static inline unsigned int i915_sg_segment_size(void) #define IS_CANNONLAKE(dev_priv) IS_PLATFORM(dev_priv, INTEL_CANNONLAKE) #define IS_ICELAKE(dev_priv) IS_PLATFORM(dev_priv, INTEL_ICELAKE) #define IS_ELKHARTLAKE(dev_priv) IS_PLATFORM(dev_priv, INTEL_ELKHARTLAKE) -#define IS_MOBILE(dev_priv) (INTEL_INFO(dev_priv)->is_mobile) #define IS_HSW_EARLY_SDV(dev_priv) (IS_HASWELL(dev_priv) && \ (INTEL_DEVID(dev_priv) & 0xFF00) == 0x0C00) -#define IS_BDW_ULT(dev_priv) (IS_BROADWELL(dev_priv) && \ - ((INTEL_DEVID(dev_priv) & 0xf) == 0x6 || \ - (INTEL_DEVID(dev_priv) & 0xf) == 0xb || \ - (INTEL_DEVID(dev_priv) & 0xf) == 0xe)) -/* ULX machines are also considered ULT. */ -#define IS_BDW_ULX(dev_priv) (IS_BROADWELL(dev_priv) && \ - (INTEL_DEVID(dev_priv) & 0xf) == 0xe) +#define IS_BDW_ULT(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULT) +#define IS_BDW_ULX(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULX) #define IS_BDW_GT3(dev_priv) (IS_BROADWELL(dev_priv) && \ INTEL_INFO(dev_priv)->gt == 3) -#define IS_HSW_ULT(dev_priv) (IS_HASWELL(dev_priv) && \ - (INTEL_DEVID(dev_priv) & 0xFF00) == 0x0A00) +#define IS_HSW_ULT(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_HASWELL, INTEL_SUBPLATFORM_ULT) #define IS_HSW_GT3(dev_priv) (IS_HASWELL(dev_priv) && \ INTEL_INFO(dev_priv)->gt == 3) #define IS_HSW_GT1(dev_priv) (IS_HASWELL(dev_priv) && \ INTEL_INFO(dev_priv)->gt == 1) /* ULX machines are also considered ULT. */ -#define IS_HSW_ULX(dev_priv) (INTEL_DEVID(dev_priv) == 0x0A0E || \ - INTEL_DEVID(dev_priv) == 0x0A1E) -#define IS_SKL_ULT(dev_priv) (INTEL_DEVID(dev_priv) == 0x1906 || \ - INTEL_DEVID(dev_priv) == 0x1913 || \ - INTEL_DEVID(dev_priv) == 0x1916 || \ - INTEL_DEVID(dev_priv) == 0x1921 || \ - INTEL_DEVID(dev_priv) == 0x1926) -#define IS_SKL_ULX(dev_priv) (INTEL_DEVID(dev_priv) == 0x190E || \ - INTEL_DEVID(dev_priv) == 0x1915 || \ - INTEL_DEVID(dev_priv) == 0x191E) -#define IS_KBL_ULT(dev_priv) (INTEL_DEVID(dev_priv) == 0x5906 || \ - INTEL_DEVID(dev_priv) == 0x5913 || \ - INTEL_DEVID(dev_priv) == 0x5916 || \ - INTEL_DEVID(dev_priv) == 0x5921 || \ - INTEL_DEVID(dev_priv) == 0x5926) -#define IS_KBL_ULX(dev_priv) (INTEL_DEVID(dev_priv) == 0x590E || \ - INTEL_DEVID(dev_priv) == 0x5915 || \ - INTEL_DEVID(dev_priv) == 0x591E) -#define IS_AML_ULX(dev_priv) (INTEL_DEVID(dev_priv) == 0x591C || \ - INTEL_DEVID(dev_priv) == 0x87C0 || \ - INTEL_DEVID(dev_priv) == 0x87CA) +#define IS_HSW_ULX(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_HASWELL, INTEL_SUBPLATFORM_ULX) +#define IS_SKL_ULT(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULT) +#define IS_SKL_ULX(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULX) +#define IS_KBL_ULT(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULT) +#define IS_KBL_ULX(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULX) +#define IS_AML_ULX(dev_priv) \ + (IS_SUBPLATFORM(dev_priv, INTEL_KABYLAKE, INTEL_SUBPLATFORM_AML) || \ + IS_SUBPLATFORM(dev_priv, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_AML)) #define IS_SKL_GT2(dev_priv) (IS_SKYLAKE(dev_priv) && \ INTEL_INFO(dev_priv)->gt == 2) #define IS_SKL_GT3(dev_priv) (IS_SKYLAKE(dev_priv) && \ @@ -2383,16 +2433,16 @@ static inline unsigned int i915_sg_segment_size(void) INTEL_INFO(dev_priv)->gt == 2) #define IS_KBL_GT3(dev_priv) (IS_KABYLAKE(dev_priv) && \ INTEL_INFO(dev_priv)->gt == 3) -#define IS_CFL_ULT(dev_priv) (IS_COFFEELAKE(dev_priv) && \ - (INTEL_DEVID(dev_priv) & 0x00F0) == 0x00A0) +#define IS_CFL_ULT(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_ULT) #define IS_CFL_GT2(dev_priv) (IS_COFFEELAKE(dev_priv) && \ INTEL_INFO(dev_priv)->gt == 2) #define IS_CFL_GT3(dev_priv) (IS_COFFEELAKE(dev_priv) && \ INTEL_INFO(dev_priv)->gt == 3) -#define IS_CNL_WITH_PORT_F(dev_priv) (IS_CANNONLAKE(dev_priv) && \ - (INTEL_DEVID(dev_priv) & 0x0004) == 0x0004) -#define IS_ICL_WITH_PORT_F(dev_priv) (IS_ICELAKE(dev_priv) && \ - INTEL_DEVID(dev_priv) != 0x8A51) +#define IS_CNL_WITH_PORT_F(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_CANNONLAKE, INTEL_SUBPLATFORM_PORTF) +#define IS_ICL_WITH_PORT_F(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_ICELAKE, INTEL_SUBPLATFORM_PORTF) #define IS_ALPHA_SUPPORT(intel_info) ((intel_info)->is_alpha_support) @@ -2451,7 +2501,6 @@ static inline unsigned int i915_sg_segment_size(void) #define IS_GEN9_LP(dev_priv) (IS_GEN(dev_priv, 9) && IS_LP(dev_priv)) #define IS_GEN9_BC(dev_priv) (IS_GEN(dev_priv, 9) && !IS_LP(dev_priv)) -#define ALL_ENGINES (~0u) #define HAS_ENGINE(dev_priv, id) (INTEL_INFO(dev_priv)->engine_mask & BIT(id)) #define ENGINE_INSTANCES_MASK(dev_priv, first, count) ({ \ @@ -2467,7 +2516,7 @@ static inline unsigned int i915_sg_segment_size(void) #define HAS_LLC(dev_priv) (INTEL_INFO(dev_priv)->has_llc) #define HAS_SNOOP(dev_priv) (INTEL_INFO(dev_priv)->has_snoop) -#define HAS_EDRAM(dev_priv) (!!((dev_priv)->edram_cap & EDRAM_ENABLED)) +#define HAS_EDRAM(dev_priv) ((dev_priv)->edram_size_mb) #define HAS_WT(dev_priv) ((IS_HASWELL(dev_priv) || \ IS_BROADWELL(dev_priv)) && HAS_EDRAM(dev_priv)) @@ -2860,6 +2909,7 @@ static inline void i915_gem_drain_workqueue(struct drm_i915_private *i915) int pass = 2; do { rcu_barrier(); + i915_gem_drain_freed_objects(i915); drain_workqueue(i915->wq); } while (--pass); } diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index e506e43cfade..0a818a60ad31 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -50,6 +50,7 @@ #include "intel_drv.h" #include "intel_frontbuffer.h" #include "intel_mocs.h" +#include "intel_pm.h" #include "intel_workarounds.h" static void i915_gem_flush_free_objects(struct drm_i915_private *i915); @@ -308,7 +309,7 @@ static void __start_cpu_write(struct drm_i915_gem_object *obj) obj->cache_dirty = true; } -static void +void __i915_gem_object_release_shmem(struct drm_i915_gem_object *obj, struct sg_table *pages, bool needs_clflush) @@ -2202,7 +2203,6 @@ i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj, struct page *page; __i915_gem_object_release_shmem(obj, pages, true); - i915_gem_gtt_finish_pages(obj, pages); if (i915_gem_object_needs_bit17_swizzle(obj)) @@ -2789,7 +2789,11 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj, u64 remain, offset; unsigned int pg; - /* Before we instantiate/pin the backing store for our use, we + /* Caller already validated user args */ + GEM_BUG_ON(!access_ok(user_data, arg->size)); + + /* + * Before we instantiate/pin the backing store for our use, we * can prepopulate the shmemfs filp efficiently using a write into * the pagecache. We avoid the penalty of instantiating all the * pages, important if the user is just writing to a few and never @@ -2803,7 +2807,8 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj, if (obj->mm.madv != I915_MADV_WILLNEED) return -EFAULT; - /* Before the pages are instantiated the object is treated as being + /* + * Before the pages are instantiated the object is treated as being * in the CPU domain. The pages will be clflushed as required before * use, and we can freely write into the pages directly. If userspace * races pwrite with any other operation; corruption will ensue - @@ -2819,20 +2824,32 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj, struct page *page; void *data, *vaddr; int err; + char c; len = PAGE_SIZE - pg; if (len > remain) len = remain; + /* Prefault the user page to reduce potential recursion */ + err = __get_user(c, user_data); + if (err) + return err; + + err = __get_user(c, user_data + len - 1); + if (err) + return err; + err = pagecache_write_begin(obj->base.filp, mapping, offset, len, 0, &page, &data); if (err < 0) return err; - vaddr = kmap(page); - unwritten = copy_from_user(vaddr + pg, user_data, len); - kunmap(page); + vaddr = kmap_atomic(page); + unwritten = __copy_from_user_inatomic(vaddr + pg, + user_data, + len); + kunmap_atomic(vaddr); err = pagecache_write_end(obj->base.filp, mapping, offset, len, len - unwritten, @@ -2840,8 +2857,9 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj, if (err < 0) return err; + /* We don't handle -EFAULT, leave it to the caller to check */ if (unwritten) - return -EFAULT; + return -ENODEV; remain -= len; user_data += len; @@ -3824,16 +3842,16 @@ i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj, return vma; } -static __always_inline unsigned int __busy_read_flag(unsigned int id) +static __always_inline u32 __busy_read_flag(u8 id) { - if (id == I915_ENGINE_CLASS_INVALID) - return 0xffff0000; + if (id == (u8)I915_ENGINE_CLASS_INVALID) + return 0xffff0000u; GEM_BUG_ON(id >= 16); - return 0x10000 << id; + return 0x10000u << id; } -static __always_inline unsigned int __busy_write_id(unsigned int id) +static __always_inline u32 __busy_write_id(u8 id) { /* * The uABI guarantees an active writer is also amongst the read @@ -3844,15 +3862,14 @@ static __always_inline unsigned int __busy_write_id(unsigned int id) * last_read - hence we always set both read and write busy for * last_write. */ - if (id == I915_ENGINE_CLASS_INVALID) - return 0xffffffff; + if (id == (u8)I915_ENGINE_CLASS_INVALID) + return 0xffffffffu; return (id + 1) | __busy_read_flag(id); } static __always_inline unsigned int -__busy_set_if_active(const struct dma_fence *fence, - unsigned int (*flag)(unsigned int id)) +__busy_set_if_active(const struct dma_fence *fence, u32 (*flag)(u8 id)) { const struct i915_request *rq; @@ -3872,6 +3889,8 @@ __busy_set_if_active(const struct dma_fence *fence, if (i915_request_completed(rq)) return 0; + /* Beware type-expansion follies! */ + BUILD_BUG_ON(!typecheck(u8, rq->engine->uabi_class)); return flag(rq->engine->uabi_class); } @@ -4494,7 +4513,7 @@ void i915_gem_resume(struct drm_i915_private *i915) * guarantee that the context image is complete. So let's just reset * it and start again. */ - i915->gt.resume(i915); + intel_gt_resume(i915); if (i915_gem_init_hw(i915)) goto err_wedged; @@ -4834,13 +4853,10 @@ int i915_gem_init(struct drm_i915_private *dev_priv) dev_priv->mm.unordered_timeline = dma_fence_context_alloc(1); - if (HAS_LOGICAL_RING_CONTEXTS(dev_priv)) { - dev_priv->gt.resume = intel_lr_context_resume; + if (HAS_LOGICAL_RING_CONTEXTS(dev_priv)) dev_priv->gt.cleanup_engine = intel_logical_ring_cleanup; - } else { - dev_priv->gt.resume = intel_legacy_submission_resume; + else dev_priv->gt.cleanup_engine = intel_engine_cleanup; - } i915_timelines_init(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_gem.h b/drivers/gpu/drm/i915/i915_gem.h index 5c073fe73664..9074eb1e843f 100644 --- a/drivers/gpu/drm/i915/i915_gem.h +++ b/drivers/gpu/drm/i915/i915_gem.h @@ -73,8 +73,6 @@ struct drm_i915_private; #define GEM_TRACE_DUMP_ON(expr) BUILD_BUG_ON_INVALID(expr) #endif -#define I915_NUM_ENGINES 8 - #define I915_GEM_IDLE_TIMEOUT (HZ / 5) void i915_gem_park(struct drm_i915_private *i915); diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index 662da485e15f..dd728b26b5aa 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -562,7 +562,7 @@ static void init_contexts(struct drm_i915_private *i915) static bool needs_preempt_context(struct drm_i915_private *i915) { - return HAS_LOGICAL_RING_PREEMPTION(i915); + return HAS_EXECLISTS(i915); } int i915_gem_contexts_init(struct drm_i915_private *dev_priv) @@ -858,9 +858,9 @@ static void cb_retire(struct i915_active *base) kfree(cb); } -I915_SELFTEST_DECLARE(static unsigned long context_barrier_inject_fault); +I915_SELFTEST_DECLARE(static intel_engine_mask_t context_barrier_inject_fault); static int context_barrier_task(struct i915_gem_context *ctx, - unsigned long engines, + intel_engine_mask_t engines, int (*emit)(struct i915_request *rq, void *data), void (*task)(void *data), void *data) @@ -922,7 +922,7 @@ static int context_barrier_task(struct i915_gem_context *ctx, } int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915, - unsigned long mask) + intel_engine_mask_t mask) { struct intel_engine_cs *engine; @@ -969,10 +969,10 @@ int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915, return 0; } -static int get_ppgtt(struct i915_gem_context *ctx, +static int get_ppgtt(struct drm_i915_file_private *file_priv, + struct i915_gem_context *ctx, struct drm_i915_gem_context_param *args) { - struct drm_i915_file_private *file_priv = ctx->file_priv; struct i915_hw_ppgtt *ppgtt; int ret; @@ -1028,6 +1028,7 @@ static int emit_ppgtt_update(struct i915_request *rq, void *data) { struct i915_hw_ppgtt *ppgtt = rq->gem_context->ppgtt; struct intel_engine_cs *engine = rq->engine; + u32 base = engine->mmio_base; u32 *cs; int i; @@ -1040,9 +1041,9 @@ static int emit_ppgtt_update(struct i915_request *rq, void *data) *cs++ = MI_LOAD_REGISTER_IMM(2); - *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(engine, 0)); + *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(base, 0)); *cs++ = upper_32_bits(pd_daddr); - *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(engine, 0)); + *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(base, 0)); *cs++ = lower_32_bits(pd_daddr); *cs++ = MI_NOOP; @@ -1056,9 +1057,9 @@ static int emit_ppgtt_update(struct i915_request *rq, void *data) for (i = GEN8_3LVL_PDPES; i--; ) { const dma_addr_t pd_daddr = i915_page_dir_dma_addr(ppgtt, i); - *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(engine, i)); + *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(base, i)); *cs++ = upper_32_bits(pd_daddr); - *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(engine, i)); + *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(base, i)); *cs++ = lower_32_bits(pd_daddr); } *cs++ = MI_NOOP; @@ -1071,10 +1072,10 @@ static int emit_ppgtt_update(struct i915_request *rq, void *data) return 0; } -static int set_ppgtt(struct i915_gem_context *ctx, +static int set_ppgtt(struct drm_i915_file_private *file_priv, + struct i915_gem_context *ctx, struct drm_i915_gem_context_param *args) { - struct drm_i915_file_private *file_priv = ctx->file_priv; struct i915_hw_ppgtt *ppgtt, *old; int err; @@ -1166,7 +1167,7 @@ static int gen8_modify_rpcs(struct intel_context *ce, struct intel_sseu sseu) { struct drm_i915_private *i915 = ce->engine->i915; - struct i915_request *rq, *prev; + struct i915_request *rq; intel_wakeref_t wakeref; int ret; @@ -1191,16 +1192,7 @@ gen8_modify_rpcs(struct intel_context *ce, struct intel_sseu sseu) } /* Queue this switch after all other activity by this context. */ - prev = i915_active_request_raw(&ce->ring->timeline->last_request, - &i915->drm.struct_mutex); - if (prev && !i915_request_completed(prev)) { - ret = i915_request_await_dma_fence(rq, &prev->fence); - if (ret < 0) - goto out_add; - } - - /* Order all following requests to be after. */ - ret = i915_timeline_set_barrier(ce->ring->timeline, rq); + ret = i915_active_request_set(&ce->ring->timeline->last_request, rq); if (ret) goto out_add; @@ -1394,8 +1386,8 @@ static int set_sseu(struct i915_gem_context *ctx, return -EINVAL; engine = intel_engine_lookup_user(i915, - user_sseu.engine_class, - user_sseu.engine_instance); + user_sseu.engine.engine_class, + user_sseu.engine.engine_instance); if (!engine) return -EINVAL; @@ -1416,7 +1408,8 @@ static int set_sseu(struct i915_gem_context *ctx, return 0; } -static int ctx_setparam(struct i915_gem_context *ctx, +static int ctx_setparam(struct drm_i915_file_private *fpriv, + struct i915_gem_context *ctx, struct drm_i915_gem_context_param *args) { int ret = 0; @@ -1485,7 +1478,7 @@ static int ctx_setparam(struct i915_gem_context *ctx, break; case I915_CONTEXT_PARAM_VM: - ret = set_ppgtt(ctx, args); + ret = set_ppgtt(fpriv, ctx, args); break; case I915_CONTEXT_PARAM_BAN_PERIOD: @@ -1513,7 +1506,7 @@ static int create_setparam(struct i915_user_extension __user *ext, void *data) if (local.param.ctx_id) return -EINVAL; - return ctx_setparam(arg->ctx, &local.param); + return ctx_setparam(arg->fpriv, arg->ctx, &local.param); } static const i915_user_extension_fn create_extensions[] = { @@ -1633,8 +1626,8 @@ static int get_sseu(struct i915_gem_context *ctx, return -EINVAL; engine = intel_engine_lookup_user(ctx->i915, - user_sseu.engine_class, - user_sseu.engine_instance); + user_sseu.engine.engine_class, + user_sseu.engine.engine_instance); if (!engine) return -EINVAL; @@ -1712,7 +1705,7 @@ int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data, break; case I915_CONTEXT_PARAM_VM: - ret = get_ppgtt(ctx, args); + ret = get_ppgtt(file_priv, ctx, args); break; case I915_CONTEXT_PARAM_BAN_PERIOD: @@ -1737,7 +1730,7 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data, if (!ctx) return -ENOENT; - ret = ctx_setparam(ctx, args); + ret = ctx_setparam(file_priv, ctx, args); i915_gem_context_put(ctx); return ret; diff --git a/drivers/gpu/drm/i915/i915_gem_context.h b/drivers/gpu/drm/i915/i915_gem_context.h index edc6ba3f0288..23dcb01bfd82 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.h +++ b/drivers/gpu/drm/i915/i915_gem_context.h @@ -142,7 +142,7 @@ void i915_gem_context_close(struct drm_file *file); int i915_switch_context(struct i915_request *rq); int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915, - unsigned long engine_mask); + intel_engine_mask_t engine_mask); void i915_gem_context_release(struct kref *ctx_ref); struct i915_gem_context * diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 736c845eb77f..8f460cc4cc1f 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -1228,7 +1228,7 @@ static int gen8_init_scratch(struct i915_address_space *vm) vm->scratch_pte = gen8_pte_encode(vm->scratch_page.daddr, I915_CACHE_LLC, - PTE_READ_ONLY); + vm->has_read_only); vm->scratch_pt = alloc_pt(vm); if (IS_ERR(vm->scratch_pt)) { @@ -1548,8 +1548,13 @@ static struct i915_hw_ppgtt *gen8_ppgtt_create(struct drm_i915_private *i915) ppgtt_init(i915, ppgtt); - /* From bdw, there is support for read-only pages in the PPGTT. */ - ppgtt->vm.has_read_only = true; + /* + * From bdw, there is hw support for read-only pages in the PPGTT. + * + * Gen11 has HSDES#:1807136187 unresolved. Disable ro support + * for now. + */ + ppgtt->vm.has_read_only = INTEL_GEN(i915) != 11; /* There are only few exceptions for gen >=6. chv and bxt. * And we are not sure about the latter so play safe for now. diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h index 83ded9fc761a..f597f35b109b 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.h +++ b/drivers/gpu/drm/i915/i915_gem_gtt.h @@ -390,7 +390,7 @@ struct i915_hw_ppgtt { struct i915_address_space vm; struct kref ref; - unsigned long pd_dirty_engines; + intel_engine_mask_t pd_dirty_engines; union { struct i915_pml4 pml4; /* GEN8+ & 48b PPGTT */ struct i915_page_directory_pointer pdp; /* GEN8+ */ diff --git a/drivers/gpu/drm/i915/i915_gem_object.h b/drivers/gpu/drm/i915/i915_gem_object.h index 1a24dc97e4fd..ca93a40c0c87 100644 --- a/drivers/gpu/drm/i915/i915_gem_object.h +++ b/drivers/gpu/drm/i915/i915_gem_object.h @@ -502,4 +502,8 @@ void i915_gem_object_set_cache_coherency(struct drm_i915_gem_object *obj, unsigned int cache_level); void i915_gem_object_flush_if_display(struct drm_i915_gem_object *obj); +void __i915_gem_object_release_shmem(struct drm_i915_gem_object *obj, + struct sg_table *pages, + bool needs_clflush); + #endif diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c index ad0087127144..215bf3fef10c 100644 --- a/drivers/gpu/drm/i915/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/i915_gem_userptr.c @@ -673,9 +673,7 @@ i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj, if (!pages) return; - if (obj->mm.madv != I915_MADV_WILLNEED) - obj->mm.dirty = false; - + __i915_gem_object_release_shmem(obj, pages, true); i915_gem_gtt_finish_pages(obj, pages); for_each_sgt_page(page, sgt_iter, pages) { diff --git a/drivers/gpu/drm/i915/i915_globals.c b/drivers/gpu/drm/i915/i915_globals.c index 2f5c72e2a9d1..81e5c2ce336b 100644 --- a/drivers/gpu/drm/i915/i915_globals.c +++ b/drivers/gpu/drm/i915/i915_globals.c @@ -17,6 +17,33 @@ static LIST_HEAD(globals); +static atomic_t active; +static atomic_t epoch; +static struct park_work { + struct rcu_work work; + int epoch; +} park; + +static void i915_globals_shrink(void) +{ + struct i915_global *global; + + /* + * kmem_cache_shrink() discards empty slabs and reorders partially + * filled slabs to prioritise allocating from the mostly full slabs, + * with the aim of reducing fragmentation. + */ + list_for_each_entry(global, &globals, link) + global->shrink(); +} + +static void __i915_globals_park(struct work_struct *work) +{ + /* Confirm nothing woke up in the last grace period */ + if (park.epoch == atomic_read(&epoch)) + i915_globals_shrink(); +} + void __init i915_global_register(struct i915_global *global) { GEM_BUG_ON(!global->shrink); @@ -57,44 +84,12 @@ int __init i915_globals_init(void) } } + INIT_RCU_WORK(&park.work, __i915_globals_park); return 0; } -static void i915_globals_shrink(void) -{ - struct i915_global *global; - - /* - * kmem_cache_shrink() discards empty slabs and reorders partially - * filled slabs to prioritise allocating from the mostly full slabs, - * with the aim of reducing fragmentation. - */ - list_for_each_entry(global, &globals, link) - global->shrink(); -} - -static atomic_t active; -static atomic_t epoch; -struct park_work { - struct rcu_work work; - int epoch; -}; - -static void __i915_globals_park(struct work_struct *work) -{ - struct park_work *wrk = container_of(work, typeof(*wrk), work.work); - - /* Confirm nothing woke up in the last grace period */ - if (wrk->epoch == atomic_read(&epoch)) - i915_globals_shrink(); - - kfree(wrk); -} - void i915_globals_park(void) { - struct park_work *wrk; - /* * Defer shrinking the global slab caches (and other work) until * after a RCU grace period has completed with no activity. This @@ -107,13 +102,8 @@ void i915_globals_park(void) if (!atomic_dec_and_test(&active)) return; - wrk = kmalloc(sizeof(*wrk), GFP_KERNEL); - if (!wrk) - return; - - wrk->epoch = atomic_inc_return(&epoch); - INIT_RCU_WORK(&wrk->work, __i915_globals_park); - queue_rcu_work(system_wq, &wrk->work); + park.epoch = atomic_inc_return(&epoch); + queue_rcu_work(system_wq, &park.work); } void i915_globals_unpark(void) @@ -125,8 +115,8 @@ void i915_globals_unpark(void) void __exit i915_globals_exit(void) { /* Flush any residual park_work */ - rcu_barrier(); - flush_scheduled_work(); + atomic_inc(&epoch); + flush_rcu_work(&park.work); __i915_globals_cleanup(); diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index a2a98ccda421..f51ff683dd2e 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -677,6 +677,9 @@ static void __err_print_to_sgl(struct drm_i915_error_state_buf *m, err_printf(m, "Reset count: %u\n", error->reset_count); err_printf(m, "Suspend count: %u\n", error->suspend_count); err_printf(m, "Platform: %s\n", intel_platform_name(error->device_info.platform)); + err_printf(m, "Subplatform: 0x%x\n", + intel_subplatform(&error->runtime_info, + error->device_info.platform)); err_print_pciid(m, m->i915); err_printf(m, "IOMMU enabled?: %d\n", error->iommu); @@ -1093,7 +1096,7 @@ static u32 capture_error_bo(struct drm_i915_error_buffer *err, * It's only a small step better than a random number in its current form. */ static u32 i915_error_generate_code(struct i915_gpu_state *error, - unsigned long engine_mask) + intel_engine_mask_t engine_mask) { /* * IPEHR would be an ideal way to detect errors, as it's the gross @@ -1212,20 +1215,23 @@ static void error_record_engine_registers(struct i915_gpu_state *error, ee->vm_info.gfx_mode = I915_READ(RING_MODE_GEN7(engine)); - if (IS_GEN(dev_priv, 6)) + if (IS_GEN(dev_priv, 6)) { ee->vm_info.pp_dir_base = ENGINE_READ(engine, RING_PP_DIR_BASE_READ); - else if (IS_GEN(dev_priv, 7)) + } else if (IS_GEN(dev_priv, 7)) { ee->vm_info.pp_dir_base = - ENGINE_READ(engine, RING_PP_DIR_BASE); - else if (INTEL_GEN(dev_priv) >= 8) + ENGINE_READ(engine, RING_PP_DIR_BASE); + } else if (INTEL_GEN(dev_priv) >= 8) { + u32 base = engine->mmio_base; + for (i = 0; i < 4; i++) { ee->vm_info.pdp[i] = - I915_READ(GEN8_RING_PDP_UDW(engine, i)); + I915_READ(GEN8_RING_PDP_UDW(base, i)); ee->vm_info.pdp[i] <<= 32; ee->vm_info.pdp[i] |= - I915_READ(GEN8_RING_PDP_LDW(engine, i)); + I915_READ(GEN8_RING_PDP_LDW(base, i)); } + } } } @@ -1629,16 +1635,17 @@ static void capture_reg_state(struct i915_gpu_state *error) error->gtier[0] = I915_READ(GTIER); error->ngtier = 1; } else if (IS_GEN(dev_priv, 2)) { - error->ier = I915_READ16(IER); + error->ier = I915_READ16(GEN2_IER); } else if (!IS_VALLEYVIEW(dev_priv)) { - error->ier = I915_READ(IER); + error->ier = I915_READ(GEN2_IER); } error->eir = I915_READ(EIR); error->pgtbl_er = I915_READ(PGTBL_ER); } static const char * -error_msg(struct i915_gpu_state *error, unsigned long engines, const char *msg) +error_msg(struct i915_gpu_state *error, + intel_engine_mask_t engines, const char *msg) { int len; int i; @@ -1648,7 +1655,7 @@ error_msg(struct i915_gpu_state *error, unsigned long engines, const char *msg) engines &= ~BIT(i); len = scnprintf(error->error_msg, sizeof(error->error_msg), - "GPU HANG: ecode %d:%lx:0x%08x", + "GPU HANG: ecode %d:%x:0x%08x", INTEL_GEN(error->i915), engines, i915_error_generate_code(error, engines)); if (engines) { @@ -1787,7 +1794,7 @@ i915_capture_gpu_state(struct drm_i915_private *i915) * to pick up. */ void i915_capture_error_state(struct drm_i915_private *i915, - unsigned long engine_mask, + intel_engine_mask_t engine_mask, const char *msg) { static bool warned; diff --git a/drivers/gpu/drm/i915/i915_gpu_error.h b/drivers/gpu/drm/i915/i915_gpu_error.h index 302a14240b45..5dc761e85d9d 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.h +++ b/drivers/gpu/drm/i915/i915_gpu_error.h @@ -263,7 +263,7 @@ void i915_error_printf(struct drm_i915_error_state_buf *e, const char *f, ...); struct i915_gpu_state *i915_capture_gpu_state(struct drm_i915_private *i915); void i915_capture_error_state(struct drm_i915_private *dev_priv, - unsigned long engine_mask, + intel_engine_mask_t engine_mask, const char *error_msg); static inline struct i915_gpu_state * diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 455b2bf691b5..b92cfd69134b 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -28,16 +28,19 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt -#include <linux/sysrq.h> -#include <linux/slab.h> -#include <linux/cpuidle.h> #include <linux/circ_buf.h> -#include <drm/drm_irq.h> +#include <linux/cpuidle.h> +#include <linux/slab.h> +#include <linux/sysrq.h> + #include <drm/drm_drv.h> +#include <drm/drm_irq.h> #include <drm/i915_drm.h> + #include "i915_drv.h" #include "i915_trace.h" #include "intel_drv.h" +#include "intel_psr.h" /** * DOC: interrupt handling @@ -133,92 +136,120 @@ static const u32 hpd_icp[HPD_NUM_PINS] = { [HPD_PORT_F] = SDE_TC4_HOTPLUG_ICP }; -/* IIR can theoretically queue up two events. Be paranoid. */ -#define GEN8_IRQ_RESET_NDX(type, which) do { \ - I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \ - POSTING_READ(GEN8_##type##_IMR(which)); \ - I915_WRITE(GEN8_##type##_IER(which), 0); \ - I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \ - POSTING_READ(GEN8_##type##_IIR(which)); \ - I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \ - POSTING_READ(GEN8_##type##_IIR(which)); \ -} while (0) - -#define GEN3_IRQ_RESET(type) do { \ - I915_WRITE(type##IMR, 0xffffffff); \ - POSTING_READ(type##IMR); \ - I915_WRITE(type##IER, 0); \ - I915_WRITE(type##IIR, 0xffffffff); \ - POSTING_READ(type##IIR); \ - I915_WRITE(type##IIR, 0xffffffff); \ - POSTING_READ(type##IIR); \ -} while (0) - -#define GEN2_IRQ_RESET(type) do { \ - I915_WRITE16(type##IMR, 0xffff); \ - POSTING_READ16(type##IMR); \ - I915_WRITE16(type##IER, 0); \ - I915_WRITE16(type##IIR, 0xffff); \ - POSTING_READ16(type##IIR); \ - I915_WRITE16(type##IIR, 0xffff); \ - POSTING_READ16(type##IIR); \ -} while (0) +static void gen3_irq_reset(struct intel_uncore *uncore, i915_reg_t imr, + i915_reg_t iir, i915_reg_t ier) +{ + intel_uncore_write(uncore, imr, 0xffffffff); + intel_uncore_posting_read(uncore, imr); + + intel_uncore_write(uncore, ier, 0); + + /* IIR can theoretically queue up two events. Be paranoid. */ + intel_uncore_write(uncore, iir, 0xffffffff); + intel_uncore_posting_read(uncore, iir); + intel_uncore_write(uncore, iir, 0xffffffff); + intel_uncore_posting_read(uncore, iir); +} + +static void gen2_irq_reset(struct intel_uncore *uncore) +{ + intel_uncore_write16(uncore, GEN2_IMR, 0xffff); + intel_uncore_posting_read16(uncore, GEN2_IMR); + + intel_uncore_write16(uncore, GEN2_IER, 0); + + /* IIR can theoretically queue up two events. Be paranoid. */ + intel_uncore_write16(uncore, GEN2_IIR, 0xffff); + intel_uncore_posting_read16(uncore, GEN2_IIR); + intel_uncore_write16(uncore, GEN2_IIR, 0xffff); + intel_uncore_posting_read16(uncore, GEN2_IIR); +} + +#define GEN8_IRQ_RESET_NDX(uncore, type, which) \ +({ \ + unsigned int which_ = which; \ + gen3_irq_reset((uncore), GEN8_##type##_IMR(which_), \ + GEN8_##type##_IIR(which_), GEN8_##type##_IER(which_)); \ +}) + +#define GEN3_IRQ_RESET(uncore, type) \ + gen3_irq_reset((uncore), type##IMR, type##IIR, type##IER) + +#define GEN2_IRQ_RESET(uncore) \ + gen2_irq_reset(uncore) /* * We should clear IMR at preinstall/uninstall, and just check at postinstall. */ -static void gen3_assert_iir_is_zero(struct drm_i915_private *dev_priv, - i915_reg_t reg) +static void gen3_assert_iir_is_zero(struct intel_uncore *uncore, i915_reg_t reg) { - u32 val = I915_READ(reg); + u32 val = intel_uncore_read(uncore, reg); if (val == 0) return; WARN(1, "Interrupt register 0x%x is not zero: 0x%08x\n", i915_mmio_reg_offset(reg), val); - I915_WRITE(reg, 0xffffffff); - POSTING_READ(reg); - I915_WRITE(reg, 0xffffffff); - POSTING_READ(reg); + intel_uncore_write(uncore, reg, 0xffffffff); + intel_uncore_posting_read(uncore, reg); + intel_uncore_write(uncore, reg, 0xffffffff); + intel_uncore_posting_read(uncore, reg); } -static void gen2_assert_iir_is_zero(struct drm_i915_private *dev_priv, - i915_reg_t reg) +static void gen2_assert_iir_is_zero(struct intel_uncore *uncore) { - u16 val = I915_READ16(reg); + u16 val = intel_uncore_read16(uncore, GEN2_IIR); if (val == 0) return; WARN(1, "Interrupt register 0x%x is not zero: 0x%08x\n", - i915_mmio_reg_offset(reg), val); - I915_WRITE16(reg, 0xffff); - POSTING_READ16(reg); - I915_WRITE16(reg, 0xffff); - POSTING_READ16(reg); -} - -#define GEN8_IRQ_INIT_NDX(type, which, imr_val, ier_val) do { \ - gen3_assert_iir_is_zero(dev_priv, GEN8_##type##_IIR(which)); \ - I915_WRITE(GEN8_##type##_IER(which), (ier_val)); \ - I915_WRITE(GEN8_##type##_IMR(which), (imr_val)); \ - POSTING_READ(GEN8_##type##_IMR(which)); \ -} while (0) - -#define GEN3_IRQ_INIT(type, imr_val, ier_val) do { \ - gen3_assert_iir_is_zero(dev_priv, type##IIR); \ - I915_WRITE(type##IER, (ier_val)); \ - I915_WRITE(type##IMR, (imr_val)); \ - POSTING_READ(type##IMR); \ -} while (0) - -#define GEN2_IRQ_INIT(type, imr_val, ier_val) do { \ - gen2_assert_iir_is_zero(dev_priv, type##IIR); \ - I915_WRITE16(type##IER, (ier_val)); \ - I915_WRITE16(type##IMR, (imr_val)); \ - POSTING_READ16(type##IMR); \ -} while (0) + i915_mmio_reg_offset(GEN2_IIR), val); + intel_uncore_write16(uncore, GEN2_IIR, 0xffff); + intel_uncore_posting_read16(uncore, GEN2_IIR); + intel_uncore_write16(uncore, GEN2_IIR, 0xffff); + intel_uncore_posting_read16(uncore, GEN2_IIR); +} + +static void gen3_irq_init(struct intel_uncore *uncore, + i915_reg_t imr, u32 imr_val, + i915_reg_t ier, u32 ier_val, + i915_reg_t iir) +{ + gen3_assert_iir_is_zero(uncore, iir); + + intel_uncore_write(uncore, ier, ier_val); + intel_uncore_write(uncore, imr, imr_val); + intel_uncore_posting_read(uncore, imr); +} + +static void gen2_irq_init(struct intel_uncore *uncore, + u32 imr_val, u32 ier_val) +{ + gen2_assert_iir_is_zero(uncore); + + intel_uncore_write16(uncore, GEN2_IER, ier_val); + intel_uncore_write16(uncore, GEN2_IMR, imr_val); + intel_uncore_posting_read16(uncore, GEN2_IMR); +} + +#define GEN8_IRQ_INIT_NDX(uncore, type, which, imr_val, ier_val) \ +({ \ + unsigned int which_ = which; \ + gen3_irq_init((uncore), \ + GEN8_##type##_IMR(which_), imr_val, \ + GEN8_##type##_IER(which_), ier_val, \ + GEN8_##type##_IIR(which_)); \ +}) + +#define GEN3_IRQ_INIT(uncore, type, imr_val, ier_val) \ + gen3_irq_init((uncore), \ + type##IMR, imr_val, \ + type##IER, ier_val, \ + type##IIR) + +#define GEN2_IRQ_INIT(uncore, imr_val, ier_val) \ + gen2_irq_init((uncore), imr_val, ier_val) static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir); static void gen9_guc_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir); @@ -366,24 +397,41 @@ static i915_reg_t gen6_pm_iir(struct drm_i915_private *dev_priv) return INTEL_GEN(dev_priv) >= 8 ? GEN8_GT_IIR(2) : GEN6_PMIIR; } -static i915_reg_t gen6_pm_imr(struct drm_i915_private *dev_priv) +static void write_pm_imr(struct drm_i915_private *dev_priv) { - if (INTEL_GEN(dev_priv) >= 11) - return GEN11_GPM_WGBOXPERF_INTR_MASK; - else if (INTEL_GEN(dev_priv) >= 8) - return GEN8_GT_IMR(2); - else - return GEN6_PMIMR; + i915_reg_t reg; + u32 mask = dev_priv->pm_imr; + + if (INTEL_GEN(dev_priv) >= 11) { + reg = GEN11_GPM_WGBOXPERF_INTR_MASK; + /* pm is in upper half */ + mask = mask << 16; + } else if (INTEL_GEN(dev_priv) >= 8) { + reg = GEN8_GT_IMR(2); + } else { + reg = GEN6_PMIMR; + } + + I915_WRITE(reg, mask); + POSTING_READ(reg); } -static i915_reg_t gen6_pm_ier(struct drm_i915_private *dev_priv) +static void write_pm_ier(struct drm_i915_private *dev_priv) { - if (INTEL_GEN(dev_priv) >= 11) - return GEN11_GPM_WGBOXPERF_INTR_ENABLE; - else if (INTEL_GEN(dev_priv) >= 8) - return GEN8_GT_IER(2); - else - return GEN6_PMIER; + i915_reg_t reg; + u32 mask = dev_priv->pm_ier; + + if (INTEL_GEN(dev_priv) >= 11) { + reg = GEN11_GPM_WGBOXPERF_INTR_ENABLE; + /* pm is in upper half */ + mask = mask << 16; + } else if (INTEL_GEN(dev_priv) >= 8) { + reg = GEN8_GT_IER(2); + } else { + reg = GEN6_PMIER; + } + + I915_WRITE(reg, mask); } /** @@ -408,8 +456,7 @@ static void snb_update_pm_irq(struct drm_i915_private *dev_priv, if (new_val != dev_priv->pm_imr) { dev_priv->pm_imr = new_val; - I915_WRITE(gen6_pm_imr(dev_priv), dev_priv->pm_imr); - POSTING_READ(gen6_pm_imr(dev_priv)); + write_pm_imr(dev_priv); } } @@ -450,7 +497,7 @@ static void gen6_enable_pm_irq(struct drm_i915_private *dev_priv, u32 enable_mas lockdep_assert_held(&dev_priv->irq_lock); dev_priv->pm_ier |= enable_mask; - I915_WRITE(gen6_pm_ier(dev_priv), dev_priv->pm_ier); + write_pm_ier(dev_priv); gen6_unmask_pm_irq(dev_priv, enable_mask); /* unmask_pm_irq provides an implicit barrier (POSTING_READ) */ } @@ -461,7 +508,7 @@ static void gen6_disable_pm_irq(struct drm_i915_private *dev_priv, u32 disable_m dev_priv->pm_ier &= ~disable_mask; __gen6_mask_pm_irq(dev_priv, disable_mask); - I915_WRITE(gen6_pm_ier(dev_priv), dev_priv->pm_ier); + write_pm_ier(dev_priv); /* though a barrier is missing here, but don't really need a one */ } @@ -1470,7 +1517,7 @@ gen8_cs_irq_handler(struct intel_engine_cs *engine, u32 iir) if (iir & GT_RENDER_USER_INTERRUPT) { intel_engine_breadcrumbs_irq(engine); - tasklet |= USES_GUC_SUBMISSION(engine->i915); + tasklet |= intel_engine_needs_breadcrumb_tasklet(engine); } if (tasklet) @@ -1793,6 +1840,25 @@ static void i9xx_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, /* The RPS events need forcewake, so we add them to a work queue and mask their * IMR bits until the work is done. Other interrupts can be processed without * the work queue. */ +static void gen11_rps_irq_handler(struct drm_i915_private *i915, u32 pm_iir) +{ + struct intel_rps *rps = &i915->gt_pm.rps; + const u32 events = i915->pm_rps_events & pm_iir; + + lockdep_assert_held(&i915->irq_lock); + + if (unlikely(!events)) + return; + + gen6_mask_pm_irq(i915, events); + + if (!rps->interrupts_enabled) + return; + + rps->pm_iir |= events; + schedule_work(&rps->work); +} + static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir) { struct intel_rps *rps = &dev_priv->gt_pm.rps; @@ -2946,7 +3012,7 @@ gen11_other_irq_handler(struct drm_i915_private * const i915, const u8 instance, const u16 iir) { if (instance == OTHER_GTPM_INSTANCE) - return gen6_rps_irq_handler(i915, iir); + return gen11_rps_irq_handler(i915, iir); WARN_ONCE(1, "unhandled other interrupt instance=0x%x, iir=0x%x\n", instance, iir); @@ -3003,14 +3069,8 @@ gen11_gt_bank_handler(struct drm_i915_private * const i915, intr_dw = raw_reg_read(regs, GEN11_GT_INTR_DW(bank)); - if (unlikely(!intr_dw)) { - DRM_ERROR("GT_INTR_DW%u blank!\n", bank); - return; - } - for_each_set_bit(bit, &intr_dw, 32) { - const u32 ident = gen11_gt_engine_identity(i915, - bank, bit); + const u32 ident = gen11_gt_engine_identity(i915, bank, bit); gen11_gt_identity_handler(i915, ident); } @@ -3305,10 +3365,12 @@ static void i945gm_vblank_work_fini(struct drm_i915_private *dev_priv) static void ibx_irq_reset(struct drm_i915_private *dev_priv) { + struct intel_uncore *uncore = &dev_priv->uncore; + if (HAS_PCH_NOP(dev_priv)) return; - GEN3_IRQ_RESET(SDE); + GEN3_IRQ_RESET(uncore, SDE); if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv)) I915_WRITE(SERR_INT, 0xffffffff); @@ -3336,13 +3398,17 @@ static void ibx_irq_pre_postinstall(struct drm_device *dev) static void gen5_gt_irq_reset(struct drm_i915_private *dev_priv) { - GEN3_IRQ_RESET(GT); + struct intel_uncore *uncore = &dev_priv->uncore; + + GEN3_IRQ_RESET(uncore, GT); if (INTEL_GEN(dev_priv) >= 6) - GEN3_IRQ_RESET(GEN6_PM); + GEN3_IRQ_RESET(uncore, GEN6_PM); } static void vlv_display_irq_reset(struct drm_i915_private *dev_priv) { + struct intel_uncore *uncore = &dev_priv->uncore; + if (IS_CHERRYVIEW(dev_priv)) I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK_CHV); else @@ -3353,12 +3419,14 @@ static void vlv_display_irq_reset(struct drm_i915_private *dev_priv) i9xx_pipestat_irq_reset(dev_priv); - GEN3_IRQ_RESET(VLV_); + GEN3_IRQ_RESET(uncore, VLV_); dev_priv->irq_mask = ~0u; } static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv) { + struct intel_uncore *uncore = &dev_priv->uncore; + u32 pipestat_mask; u32 enable_mask; enum pipe pipe; @@ -3383,7 +3451,7 @@ static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv) dev_priv->irq_mask = ~enable_mask; - GEN3_IRQ_INIT(VLV_, dev_priv->irq_mask, enable_mask); + GEN3_IRQ_INIT(uncore, VLV_, dev_priv->irq_mask, enable_mask); } /* drm_dma.h hooks @@ -3391,8 +3459,9 @@ static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv) static void ironlake_irq_reset(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; - GEN3_IRQ_RESET(DE); + GEN3_IRQ_RESET(uncore, DE); if (IS_GEN(dev_priv, 7)) I915_WRITE(GEN7_ERR_INT, 0xffffffff); @@ -3423,15 +3492,18 @@ static void valleyview_irq_reset(struct drm_device *dev) static void gen8_gt_irq_reset(struct drm_i915_private *dev_priv) { - GEN8_IRQ_RESET_NDX(GT, 0); - GEN8_IRQ_RESET_NDX(GT, 1); - GEN8_IRQ_RESET_NDX(GT, 2); - GEN8_IRQ_RESET_NDX(GT, 3); + struct intel_uncore *uncore = &dev_priv->uncore; + + GEN8_IRQ_RESET_NDX(uncore, GT, 0); + GEN8_IRQ_RESET_NDX(uncore, GT, 1); + GEN8_IRQ_RESET_NDX(uncore, GT, 2); + GEN8_IRQ_RESET_NDX(uncore, GT, 3); } static void gen8_irq_reset(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; int pipe; gen8_master_intr_disable(dev_priv->uncore.regs); @@ -3444,11 +3516,11 @@ static void gen8_irq_reset(struct drm_device *dev) for_each_pipe(dev_priv, pipe) if (intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PIPE(pipe))) - GEN8_IRQ_RESET_NDX(DE_PIPE, pipe); + GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe); - GEN3_IRQ_RESET(GEN8_DE_PORT_); - GEN3_IRQ_RESET(GEN8_DE_MISC_); - GEN3_IRQ_RESET(GEN8_PCU_); + GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_); + GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_); + GEN3_IRQ_RESET(uncore, GEN8_PCU_); if (HAS_PCH_SPLIT(dev_priv)) ibx_irq_reset(dev_priv); @@ -3474,6 +3546,7 @@ static void gen11_gt_irq_reset(struct drm_i915_private *dev_priv) static void gen11_irq_reset(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_uncore *uncore = &dev_priv->uncore; int pipe; gen11_master_intr_disable(dev_priv->uncore.regs); @@ -3488,21 +3561,23 @@ static void gen11_irq_reset(struct drm_device *dev) for_each_pipe(dev_priv, pipe) if (intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PIPE(pipe))) - GEN8_IRQ_RESET_NDX(DE_PIPE, pipe); + GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe); - GEN3_IRQ_RESET(GEN8_DE_PORT_); - GEN3_IRQ_RESET(GEN8_DE_MISC_); - GEN3_IRQ_RESET(GEN11_DE_HPD_); - GEN3_IRQ_RESET(GEN11_GU_MISC_); - GEN3_IRQ_RESET(GEN8_PCU_); + GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_); + GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_); + GEN3_IRQ_RESET(uncore, GEN11_DE_HPD_); + GEN3_IRQ_RESET(uncore, GEN11_GU_MISC_); + GEN3_IRQ_RESET(uncore, GEN8_PCU_); if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) - GEN3_IRQ_RESET(SDE); + GEN3_IRQ_RESET(uncore, SDE); } void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv, u8 pipe_mask) { + struct intel_uncore *uncore = &dev_priv->uncore; + u32 extra_ier = GEN8_PIPE_VBLANK | GEN8_PIPE_FIFO_UNDERRUN; enum pipe pipe; @@ -3514,7 +3589,7 @@ void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv, } for_each_pipe_masked(dev_priv, pipe, pipe_mask) - GEN8_IRQ_INIT_NDX(DE_PIPE, pipe, + GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe, dev_priv->de_irq_mask[pipe], ~dev_priv->de_irq_mask[pipe] | extra_ier); @@ -3524,6 +3599,7 @@ void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv, void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv, u8 pipe_mask) { + struct intel_uncore *uncore = &dev_priv->uncore; enum pipe pipe; spin_lock_irq(&dev_priv->irq_lock); @@ -3534,7 +3610,7 @@ void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv, } for_each_pipe_masked(dev_priv, pipe, pipe_mask) - GEN8_IRQ_RESET_NDX(DE_PIPE, pipe); + GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe); spin_unlock_irq(&dev_priv->irq_lock); @@ -3545,13 +3621,14 @@ void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv, static void cherryview_irq_reset(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; I915_WRITE(GEN8_MASTER_IRQ, 0); POSTING_READ(GEN8_MASTER_IRQ); gen8_gt_irq_reset(dev_priv); - GEN3_IRQ_RESET(GEN8_PCU_); + GEN3_IRQ_RESET(uncore, GEN8_PCU_); spin_lock_irq(&dev_priv->irq_lock); if (dev_priv->display_irqs_enabled) @@ -3823,7 +3900,7 @@ static void ibx_irq_postinstall(struct drm_device *dev) else mask = SDE_GMBUS_CPT; - gen3_assert_iir_is_zero(dev_priv, SDEIIR); + gen3_assert_iir_is_zero(&dev_priv->uncore, SDEIIR); I915_WRITE(SDEIMR, ~mask); if (HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv) || @@ -3836,6 +3913,7 @@ static void ibx_irq_postinstall(struct drm_device *dev) static void gen5_gt_irq_postinstall(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; u32 pm_irqs, gt_irqs; pm_irqs = gt_irqs = 0; @@ -3854,7 +3932,7 @@ static void gen5_gt_irq_postinstall(struct drm_device *dev) gt_irqs |= GT_BLT_USER_INTERRUPT | GT_BSD_USER_INTERRUPT; } - GEN3_IRQ_INIT(GT, dev_priv->gt_irq_mask, gt_irqs); + GEN3_IRQ_INIT(uncore, GT, dev_priv->gt_irq_mask, gt_irqs); if (INTEL_GEN(dev_priv) >= 6) { /* @@ -3867,13 +3945,14 @@ static void gen5_gt_irq_postinstall(struct drm_device *dev) } dev_priv->pm_imr = 0xffffffff; - GEN3_IRQ_INIT(GEN6_PM, dev_priv->pm_imr, pm_irqs); + GEN3_IRQ_INIT(uncore, GEN6_PM, dev_priv->pm_imr, pm_irqs); } } static int ironlake_irq_postinstall(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; u32 display_mask, extra_mask; if (INTEL_GEN(dev_priv) >= 7) { @@ -3892,7 +3971,7 @@ static int ironlake_irq_postinstall(struct drm_device *dev) } if (IS_HASWELL(dev_priv)) { - gen3_assert_iir_is_zero(dev_priv, EDP_PSR_IIR); + gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR); intel_psr_irq_control(dev_priv, dev_priv->psr.debug); display_mask |= DE_EDP_PSR_INT_HSW; } @@ -3901,7 +3980,8 @@ static int ironlake_irq_postinstall(struct drm_device *dev) ibx_irq_pre_postinstall(dev); - GEN3_IRQ_INIT(DE, dev_priv->irq_mask, display_mask | extra_mask); + GEN3_IRQ_INIT(uncore, DE, dev_priv->irq_mask, + display_mask | extra_mask); gen5_gt_irq_postinstall(dev); @@ -3971,6 +4051,8 @@ static int valleyview_irq_postinstall(struct drm_device *dev) static void gen8_gt_irq_postinstall(struct drm_i915_private *dev_priv) { + struct intel_uncore *uncore = &dev_priv->uncore; + /* These are interrupts we'll toggle with the ring mask register */ u32 gt_interrupts[] = { (GT_RENDER_USER_INTERRUPT << GEN8_RCS_IRQ_SHIFT | @@ -3991,18 +4073,20 @@ static void gen8_gt_irq_postinstall(struct drm_i915_private *dev_priv) dev_priv->pm_ier = 0x0; dev_priv->pm_imr = ~dev_priv->pm_ier; - GEN8_IRQ_INIT_NDX(GT, 0, ~gt_interrupts[0], gt_interrupts[0]); - GEN8_IRQ_INIT_NDX(GT, 1, ~gt_interrupts[1], gt_interrupts[1]); + GEN8_IRQ_INIT_NDX(uncore, GT, 0, ~gt_interrupts[0], gt_interrupts[0]); + GEN8_IRQ_INIT_NDX(uncore, GT, 1, ~gt_interrupts[1], gt_interrupts[1]); /* * RPS interrupts will get enabled/disabled on demand when RPS itself * is enabled/disabled. Same wil be the case for GuC interrupts. */ - GEN8_IRQ_INIT_NDX(GT, 2, dev_priv->pm_imr, dev_priv->pm_ier); - GEN8_IRQ_INIT_NDX(GT, 3, ~gt_interrupts[3], gt_interrupts[3]); + GEN8_IRQ_INIT_NDX(uncore, GT, 2, dev_priv->pm_imr, dev_priv->pm_ier); + GEN8_IRQ_INIT_NDX(uncore, GT, 3, ~gt_interrupts[3], gt_interrupts[3]); } static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) { + struct intel_uncore *uncore = &dev_priv->uncore; + u32 de_pipe_masked = GEN8_PIPE_CDCLK_CRC_DONE; u32 de_pipe_enables; u32 de_port_masked = GEN8_AUX_CHANNEL_A; @@ -4038,7 +4122,7 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) else if (IS_BROADWELL(dev_priv)) de_port_enables |= GEN8_PORT_DP_A_HOTPLUG; - gen3_assert_iir_is_zero(dev_priv, EDP_PSR_IIR); + gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR); intel_psr_irq_control(dev_priv, dev_priv->psr.debug); for_each_pipe(dev_priv, pipe) { @@ -4046,20 +4130,21 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) if (intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PIPE(pipe))) - GEN8_IRQ_INIT_NDX(DE_PIPE, pipe, + GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe, dev_priv->de_irq_mask[pipe], de_pipe_enables); } - GEN3_IRQ_INIT(GEN8_DE_PORT_, ~de_port_masked, de_port_enables); - GEN3_IRQ_INIT(GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked); + GEN3_IRQ_INIT(uncore, GEN8_DE_PORT_, ~de_port_masked, de_port_enables); + GEN3_IRQ_INIT(uncore, GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked); if (INTEL_GEN(dev_priv) >= 11) { u32 de_hpd_masked = 0; u32 de_hpd_enables = GEN11_DE_TC_HOTPLUG_MASK | GEN11_DE_TBT_HOTPLUG_MASK; - GEN3_IRQ_INIT(GEN11_DE_HPD_, ~de_hpd_masked, de_hpd_enables); + GEN3_IRQ_INIT(uncore, GEN11_DE_HPD_, ~de_hpd_masked, + de_hpd_enables); gen11_hpd_detection_setup(dev_priv); } else if (IS_GEN9_LP(dev_priv)) { bxt_hpd_detection_setup(dev_priv); @@ -4122,7 +4207,7 @@ static void icp_irq_postinstall(struct drm_device *dev) I915_WRITE(SDEIER, 0xffffffff); POSTING_READ(SDEIER); - gen3_assert_iir_is_zero(dev_priv, SDEIIR); + gen3_assert_iir_is_zero(&dev_priv->uncore, SDEIIR); I915_WRITE(SDEIMR, ~mask); icp_hpd_detection_setup(dev_priv); @@ -4131,6 +4216,7 @@ static void icp_irq_postinstall(struct drm_device *dev) static int gen11_irq_postinstall(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_uncore *uncore = &dev_priv->uncore; u32 gu_misc_masked = GEN11_GU_MISC_GSE; if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) @@ -4139,7 +4225,7 @@ static int gen11_irq_postinstall(struct drm_device *dev) gen11_gt_irq_postinstall(dev_priv); gen8_de_irq_postinstall(dev_priv); - GEN3_IRQ_INIT(GEN11_GU_MISC_, ~gu_misc_masked, gu_misc_masked); + GEN3_IRQ_INIT(uncore, GEN11_GU_MISC_, ~gu_misc_masked, gu_misc_masked); I915_WRITE(GEN11_DISPLAY_INT_CTL, GEN11_DISPLAY_IRQ_ENABLE); @@ -4169,15 +4255,17 @@ static int cherryview_irq_postinstall(struct drm_device *dev) static void i8xx_irq_reset(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; i9xx_pipestat_irq_reset(dev_priv); - GEN2_IRQ_RESET(); + GEN2_IRQ_RESET(uncore); } static int i8xx_irq_postinstall(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; u16 enable_mask; I915_WRITE16(EMR, ~(I915_ERROR_PAGE_TABLE | @@ -4195,7 +4283,7 @@ static int i8xx_irq_postinstall(struct drm_device *dev) I915_MASTER_ERROR_INTERRUPT | I915_USER_INTERRUPT; - GEN2_IRQ_INIT(, dev_priv->irq_mask, enable_mask); + GEN2_IRQ_INIT(uncore, dev_priv->irq_mask, enable_mask); /* Interrupt setup is already guaranteed to be single-threaded, this is * just to make the assert_spin_locked check happy. */ @@ -4299,7 +4387,7 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg) u16 eir = 0, eir_stuck = 0; u16 iir; - iir = I915_READ16(IIR); + iir = I915_READ16(GEN2_IIR); if (iir == 0) break; @@ -4312,7 +4400,7 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg) if (iir & I915_MASTER_ERROR_INTERRUPT) i8xx_error_irq_ack(dev_priv, &eir, &eir_stuck); - I915_WRITE16(IIR, iir); + I915_WRITE16(GEN2_IIR, iir); if (iir & I915_USER_INTERRUPT) intel_engine_breadcrumbs_irq(dev_priv->engine[RCS0]); @@ -4331,6 +4419,7 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg) static void i915_irq_reset(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; if (I915_HAS_HOTPLUG(dev_priv)) { i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0); @@ -4339,12 +4428,13 @@ static void i915_irq_reset(struct drm_device *dev) i9xx_pipestat_irq_reset(dev_priv); - GEN3_IRQ_RESET(); + GEN3_IRQ_RESET(uncore, GEN2_); } static int i915_irq_postinstall(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; u32 enable_mask; I915_WRITE(EMR, ~(I915_ERROR_PAGE_TABLE | @@ -4371,7 +4461,7 @@ static int i915_irq_postinstall(struct drm_device *dev) dev_priv->irq_mask &= ~I915_DISPLAY_PORT_INTERRUPT; } - GEN3_IRQ_INIT(, dev_priv->irq_mask, enable_mask); + GEN3_IRQ_INIT(uncore, GEN2_, dev_priv->irq_mask, enable_mask); /* Interrupt setup is already guaranteed to be single-threaded, this is * just to make the assert_spin_locked check happy. */ @@ -4403,7 +4493,7 @@ static irqreturn_t i915_irq_handler(int irq, void *arg) u32 hotplug_status = 0; u32 iir; - iir = I915_READ(IIR); + iir = I915_READ(GEN2_IIR); if (iir == 0) break; @@ -4420,7 +4510,7 @@ static irqreturn_t i915_irq_handler(int irq, void *arg) if (iir & I915_MASTER_ERROR_INTERRUPT) i9xx_error_irq_ack(dev_priv, &eir, &eir_stuck); - I915_WRITE(IIR, iir); + I915_WRITE(GEN2_IIR, iir); if (iir & I915_USER_INTERRUPT) intel_engine_breadcrumbs_irq(dev_priv->engine[RCS0]); @@ -4442,18 +4532,20 @@ static irqreturn_t i915_irq_handler(int irq, void *arg) static void i965_irq_reset(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0); I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT)); i9xx_pipestat_irq_reset(dev_priv); - GEN3_IRQ_RESET(); + GEN3_IRQ_RESET(uncore, GEN2_); } static int i965_irq_postinstall(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_uncore *uncore = &dev_priv->uncore; u32 enable_mask; u32 error_mask; @@ -4491,7 +4583,7 @@ static int i965_irq_postinstall(struct drm_device *dev) if (IS_G4X(dev_priv)) enable_mask |= I915_BSD_USER_INTERRUPT; - GEN3_IRQ_INIT(, dev_priv->irq_mask, enable_mask); + GEN3_IRQ_INIT(uncore, GEN2_, dev_priv->irq_mask, enable_mask); /* Interrupt setup is already guaranteed to be single-threaded, this is * just to make the assert_spin_locked check happy. */ @@ -4549,7 +4641,7 @@ static irqreturn_t i965_irq_handler(int irq, void *arg) u32 hotplug_status = 0; u32 iir; - iir = I915_READ(IIR); + iir = I915_READ(GEN2_IIR); if (iir == 0) break; @@ -4565,7 +4657,7 @@ static irqreturn_t i965_irq_handler(int irq, void *arg) if (iir & I915_MASTER_ERROR_INTERRUPT) i9xx_error_irq_ack(dev_priv, &eir, &eir_stuck); - I915_WRITE(IIR, iir); + I915_WRITE(GEN2_IIR, iir); if (iir & I915_USER_INTERRUPT) intel_engine_breadcrumbs_irq(dev_priv->engine[RCS0]); @@ -4623,6 +4715,10 @@ void intel_irq_init(struct drm_i915_private *dev_priv) GEN6_PM_RP_DOWN_THRESHOLD | GEN6_PM_RP_DOWN_TIMEOUT); + /* We share the register with other engine */ + if (INTEL_GEN(dev_priv) > 9) + GEM_WARN_ON(dev_priv->pm_rps_events & 0xffff0000); + rps->pm_intrmsk_mbz = 0; /* diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index a7e1611af26d..f893c2cbce15 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -31,8 +31,9 @@ #include "i915_drv.h" #include "i915_globals.h" #include "i915_selftest.h" +#include "intel_fbdev.h" -#define PLATFORM(x) .platform = (x), .platform_mask = BIT(x) +#define PLATFORM(x) .platform = (x) #define GEN(x) .gen = (x), .gen_mask = BIT((x) - 1) #define I845_PIPE_OFFSETS \ @@ -116,8 +117,16 @@ [PIPE_C] = IVB_CURSOR_C_OFFSET, \ } -#define BDW_COLORS \ - .color = { .degamma_lut_size = 512, .gamma_lut_size = 512 } +#define I9XX_COLORS \ + .color = { .gamma_lut_size = 256 } +#define I965_COLORS \ + .color = { .gamma_lut_size = 129, \ + .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ + } +#define ILK_COLORS \ + .color = { .gamma_lut_size = 1024 } +#define IVB_COLORS \ + .color = { .degamma_lut_size = 1024, .gamma_lut_size = 1024 } #define CHV_COLORS \ .color = { .degamma_lut_size = 65, .gamma_lut_size = 257, \ .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ @@ -150,6 +159,7 @@ .has_coherent_ggtt = false, \ I9XX_PIPE_OFFSETS, \ I9XX_CURSOR_OFFSETS, \ + I9XX_COLORS, \ GEN_DEFAULT_PAGE_SIZES #define I845_FEATURES \ @@ -166,6 +176,7 @@ .has_coherent_ggtt = false, \ I845_PIPE_OFFSETS, \ I845_CURSOR_OFFSETS, \ + I9XX_COLORS, \ GEN_DEFAULT_PAGE_SIZES static const struct intel_device_info intel_i830_info = { @@ -199,6 +210,7 @@ static const struct intel_device_info intel_i865g_info = { .has_coherent_ggtt = true, \ I9XX_PIPE_OFFSETS, \ I9XX_CURSOR_OFFSETS, \ + I9XX_COLORS, \ GEN_DEFAULT_PAGE_SIZES static const struct intel_device_info intel_i915g_info = { @@ -257,7 +269,14 @@ static const struct intel_device_info intel_g33_info = { .display.has_overlay = 1, }; -static const struct intel_device_info intel_pineview_info = { +static const struct intel_device_info intel_pineview_g_info = { + GEN3_FEATURES, + PLATFORM(INTEL_PINEVIEW), + .display.has_hotplug = 1, + .display.has_overlay = 1, +}; + +static const struct intel_device_info intel_pineview_m_info = { GEN3_FEATURES, PLATFORM(INTEL_PINEVIEW), .is_mobile = 1, @@ -276,6 +295,7 @@ static const struct intel_device_info intel_pineview_info = { .has_coherent_ggtt = true, \ I9XX_PIPE_OFFSETS, \ I9XX_CURSOR_OFFSETS, \ + I965_COLORS, \ GEN_DEFAULT_PAGE_SIZES static const struct intel_device_info intel_i965g_info = { @@ -325,6 +345,7 @@ static const struct intel_device_info intel_gm45_info = { .has_rc6 = 0, \ I9XX_PIPE_OFFSETS, \ I9XX_CURSOR_OFFSETS, \ + ILK_COLORS, \ GEN_DEFAULT_PAGE_SIZES static const struct intel_device_info intel_ironlake_d_info = { @@ -353,6 +374,7 @@ static const struct intel_device_info intel_ironlake_m_info = { .ppgtt_size = 31, \ I9XX_PIPE_OFFSETS, \ I9XX_CURSOR_OFFSETS, \ + ILK_COLORS, \ GEN_DEFAULT_PAGE_SIZES #define SNB_D_PLATFORM \ @@ -399,6 +421,7 @@ static const struct intel_device_info intel_sandybridge_m_gt2_info = { .ppgtt_size = 31, \ IVB_PIPE_OFFSETS, \ IVB_CURSOR_OFFSETS, \ + IVB_COLORS, \ GEN_DEFAULT_PAGE_SIZES #define IVB_D_PLATFORM \ @@ -457,6 +480,7 @@ static const struct intel_device_info intel_valleyview_info = { .display_mmio_offset = VLV_DISPLAY_BASE, I9XX_PIPE_OFFSETS, I9XX_CURSOR_OFFSETS, + I965_COLORS, GEN_DEFAULT_PAGE_SIZES, }; @@ -494,7 +518,6 @@ static const struct intel_device_info intel_haswell_gt3_info = { #define GEN8_FEATURES \ G75_FEATURES, \ GEN(8), \ - BDW_COLORS, \ .page_sizes = I915_GTT_PAGE_SIZE_4K | \ I915_GTT_PAGE_SIZE_2M, \ .has_logical_ring_contexts = 1, \ @@ -629,7 +652,7 @@ static const struct intel_device_info intel_skylake_gt4_info = { .display.has_ipc = 1, \ HSW_PIPE_OFFSETS, \ IVB_CURSOR_OFFSETS, \ - BDW_COLORS, \ + IVB_COLORS, \ GEN9_DEFAULT_PAGE_SIZES static const struct intel_device_info intel_broxton_info = { @@ -761,7 +784,8 @@ static const struct pci_device_id pciidlist[] = { INTEL_I965GM_IDS(&intel_i965gm_info), INTEL_GM45_IDS(&intel_gm45_info), INTEL_G45_IDS(&intel_g45_info), - INTEL_PINEVIEW_IDS(&intel_pineview_info), + INTEL_PINEVIEW_G_IDS(&intel_pineview_g_info), + INTEL_PINEVIEW_M_IDS(&intel_pineview_m_info), INTEL_IRONLAKE_D_IDS(&intel_ironlake_d_info), INTEL_IRONLAKE_M_IDS(&intel_ironlake_m_info), INTEL_SNB_D_GT1_IDS(&intel_sandybridge_d_gt1_info), diff --git a/drivers/gpu/drm/i915/i915_priolist_types.h b/drivers/gpu/drm/i915/i915_priolist_types.h new file mode 100644 index 000000000000..cc44ebd3b553 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_priolist_types.h @@ -0,0 +1,42 @@ +/* + * SPDX-License-Identifier: MIT + * + * Copyright © 2018 Intel Corporation + */ + +#ifndef _I915_PRIOLIST_TYPES_H_ +#define _I915_PRIOLIST_TYPES_H_ + +#include <linux/list.h> +#include <linux/rbtree.h> + +#include <uapi/drm/i915_drm.h> + +enum { + I915_PRIORITY_MIN = I915_CONTEXT_MIN_USER_PRIORITY - 1, + I915_PRIORITY_NORMAL = I915_CONTEXT_DEFAULT_PRIORITY, + I915_PRIORITY_MAX = I915_CONTEXT_MAX_USER_PRIORITY + 1, + + I915_PRIORITY_INVALID = INT_MIN +}; + +#define I915_USER_PRIORITY_SHIFT 3 +#define I915_USER_PRIORITY(x) ((x) << I915_USER_PRIORITY_SHIFT) + +#define I915_PRIORITY_COUNT BIT(I915_USER_PRIORITY_SHIFT) +#define I915_PRIORITY_MASK (I915_PRIORITY_COUNT - 1) + +#define I915_PRIORITY_WAIT ((u8)BIT(0)) +#define I915_PRIORITY_NEWCLIENT ((u8)BIT(1)) +#define I915_PRIORITY_NOSEMAPHORE ((u8)BIT(2)) + +#define __NO_PREEMPTION (I915_PRIORITY_WAIT) + +struct i915_priolist { + struct list_head requests[I915_PRIORITY_COUNT]; + struct rb_node node; + unsigned long used; + int priority; +}; + +#endif /* _I915_PRIOLIST_TYPES_H_ */ diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index c866379a521b..b74824f0b5b1 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -439,8 +439,8 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define RING_PP_DIR_DCLV(base) _MMIO((base) + 0x220) #define PP_DIR_DCLV_2G 0xffffffff -#define GEN8_RING_PDP_UDW(engine, n) _MMIO((engine)->mmio_base + 0x270 + (n) * 8 + 4) -#define GEN8_RING_PDP_LDW(engine, n) _MMIO((engine)->mmio_base + 0x270 + (n) * 8) +#define GEN8_RING_PDP_UDW(base, n) _MMIO((base) + 0x270 + (n) * 8 + 4) +#define GEN8_RING_PDP_LDW(base, n) _MMIO((base) + 0x270 + (n) * 8) #define GEN8_R_PWR_CLK_STATE _MMIO(0x20C8) #define GEN8_RPCS_ENABLE (1 << 31) @@ -2446,8 +2446,10 @@ enum i915_power_well_id { #define RING_HWS_PGA(base) _MMIO((base) + 0x80) #define RING_HWS_PGA_GEN6(base) _MMIO((base) + 0x2080) #define RING_RESET_CTL(base) _MMIO((base) + 0xd0) -#define RESET_CTL_REQUEST_RESET (1 << 0) -#define RESET_CTL_READY_TO_RESET (1 << 1) +#define RESET_CTL_CAT_ERROR REG_BIT(2) +#define RESET_CTL_READY_TO_RESET REG_BIT(1) +#define RESET_CTL_REQUEST_RESET REG_BIT(0) + #define RING_SEMA_WAIT_POLL(base) _MMIO((base) + 0x24c) #define HSW_GTT_CACHE_EN _MMIO(0x4024) @@ -2713,10 +2715,10 @@ enum i915_power_well_id { #define VLV_GU_CTL0 _MMIO(VLV_DISPLAY_BASE + 0x2030) #define VLV_GU_CTL1 _MMIO(VLV_DISPLAY_BASE + 0x2034) #define SCPD0 _MMIO(0x209c) /* 915+ only */ -#define IER _MMIO(0x20a0) -#define IIR _MMIO(0x20a4) -#define IMR _MMIO(0x20a8) -#define ISR _MMIO(0x20ac) +#define GEN2_IER _MMIO(0x20a0) +#define GEN2_IIR _MMIO(0x20a4) +#define GEN2_IMR _MMIO(0x20a8) +#define GEN2_ISR _MMIO(0x20ac) #define VLV_GUNIT_CLOCK_GATE _MMIO(VLV_DISPLAY_BASE + 0x2060) #define GINT_DIS (1 << 22) #define GCFG_DIS (1 << 8) @@ -4209,42 +4211,6 @@ enum { #define PIPESRC(trans) _MMIO_TRANS2(trans, _PIPEASRC) #define PIPE_MULT(trans) _MMIO_TRANS2(trans, _PIPE_MULT_A) -/* VLV eDP PSR registers */ -#define _PSRCTLA (VLV_DISPLAY_BASE + 0x60090) -#define _PSRCTLB (VLV_DISPLAY_BASE + 0x61090) -#define VLV_EDP_PSR_ENABLE (1 << 0) -#define VLV_EDP_PSR_RESET (1 << 1) -#define VLV_EDP_PSR_MODE_MASK (7 << 2) -#define VLV_EDP_PSR_MODE_HW_TIMER (1 << 3) -#define VLV_EDP_PSR_MODE_SW_TIMER (1 << 2) -#define VLV_EDP_PSR_SINGLE_FRAME_UPDATE (1 << 7) -#define VLV_EDP_PSR_ACTIVE_ENTRY (1 << 8) -#define VLV_EDP_PSR_SRC_TRANSMITTER_STATE (1 << 9) -#define VLV_EDP_PSR_DBL_FRAME (1 << 10) -#define VLV_EDP_PSR_FRAME_COUNT_MASK (0xff << 16) -#define VLV_EDP_PSR_IDLE_FRAME_SHIFT 16 -#define VLV_PSRCTL(pipe) _MMIO_PIPE(pipe, _PSRCTLA, _PSRCTLB) - -#define _VSCSDPA (VLV_DISPLAY_BASE + 0x600a0) -#define _VSCSDPB (VLV_DISPLAY_BASE + 0x610a0) -#define VLV_EDP_PSR_SDP_FREQ_MASK (3 << 30) -#define VLV_EDP_PSR_SDP_FREQ_ONCE (1 << 31) -#define VLV_EDP_PSR_SDP_FREQ_EVFRAME (1 << 30) -#define VLV_VSCSDP(pipe) _MMIO_PIPE(pipe, _VSCSDPA, _VSCSDPB) - -#define _PSRSTATA (VLV_DISPLAY_BASE + 0x60094) -#define _PSRSTATB (VLV_DISPLAY_BASE + 0x61094) -#define VLV_EDP_PSR_LAST_STATE_MASK (7 << 3) -#define VLV_EDP_PSR_CURR_STATE_MASK 7 -#define VLV_EDP_PSR_DISABLED (0 << 0) -#define VLV_EDP_PSR_INACTIVE (1 << 0) -#define VLV_EDP_PSR_IN_TRANS_TO_ACTIVE (2 << 0) -#define VLV_EDP_PSR_ACTIVE_NORFB_UP (3 << 0) -#define VLV_EDP_PSR_ACTIVE_SF_UPDATE (4 << 0) -#define VLV_EDP_PSR_EXIT (5 << 0) -#define VLV_EDP_PSR_IN_TRANS (1 << 7) -#define VLV_PSRSTAT(pipe) _MMIO_PIPE(pipe, _PSRSTATA, _PSRSTATB) - /* HSW+ eDP PSR registers */ #define HSW_EDP_PSR_BASE 0x64800 #define BDW_EDP_PSR_BASE 0x6f800 @@ -5795,6 +5761,10 @@ enum { #define PIPEFRAMEPIXEL(pipe) _MMIO_PIPE2(pipe, _PIPEAFRAMEPIXEL) #define PIPESTAT(pipe) _MMIO_PIPE2(pipe, _PIPEASTAT) +#define _PIPEAGCMAX 0x70010 +#define _PIPEBGCMAX 0x71010 +#define PIPEGCMAX(pipe, i) _MMIO_PIPE2(pipe, _PIPEAGCMAX + (i) * 4) + #define _PIPE_MISC_A 0x70030 #define _PIPE_MISC_B 0x71030 #define PIPEMISC_YUV420_ENABLE (1 << 27) @@ -7209,11 +7179,21 @@ enum { #define _LGC_PALETTE_B 0x4a800 #define LGC_PALETTE(pipe, i) _MMIO(_PIPE(pipe, _LGC_PALETTE_A, _LGC_PALETTE_B) + (i) * 4) +/* ilk/snb precision palette */ +#define _PREC_PALETTE_A 0x4b000 +#define _PREC_PALETTE_B 0x4c000 +#define PREC_PALETTE(pipe, i) _MMIO(_PIPE(pipe, _PREC_PALETTE_A, _PREC_PALETTE_B) + (i) * 4) + +#define _PREC_PIPEAGCMAX 0x4d000 +#define _PREC_PIPEBGCMAX 0x4d010 +#define PREC_PIPEGCMAX(pipe, i) _MMIO(_PIPE(pipe, _PIPEAGCMAX, _PIPEBGCMAX) + (i) * 4) + #define _GAMMA_MODE_A 0x4a480 #define _GAMMA_MODE_B 0x4ac80 #define GAMMA_MODE(pipe) _MMIO_PIPE(pipe, _GAMMA_MODE_A, _GAMMA_MODE_B) #define PRE_CSC_GAMMA_ENABLE (1 << 31) #define POST_CSC_GAMMA_ENABLE (1 << 30) +#define GAMMA_MODE_MODE_MASK (3 << 0) #define GAMMA_MODE_MODE_8BIT (0 << 0) #define GAMMA_MODE_MODE_10BIT (1 << 0) #define GAMMA_MODE_MODE_12BIT (2 << 0) @@ -8709,8 +8689,9 @@ enum { #define GEN9_MEDIA_PG_IDLE_HYSTERESIS _MMIO(0xA0C4) #define GEN9_RENDER_PG_IDLE_HYSTERESIS _MMIO(0xA0C8) #define GEN9_PG_ENABLE _MMIO(0xA210) -#define GEN9_RENDER_PG_ENABLE (1 << 0) -#define GEN9_MEDIA_PG_ENABLE (1 << 1) +#define GEN9_RENDER_PG_ENABLE REG_BIT(0) +#define GEN9_MEDIA_PG_ENABLE REG_BIT(1) +#define GEN11_MEDIA_SAMPLER_PG_ENABLE REG_BIT(2) #define GEN8_PUSHBUS_CONTROL _MMIO(0xA248) #define GEN8_PUSHBUS_ENABLE _MMIO(0xA250) #define GEN8_PUSHBUS_SHIFT _MMIO(0xA25C) @@ -8725,6 +8706,11 @@ enum { #define GEN6_PMIER _MMIO(0x4402C) #define GEN6_PM_MBOX_EVENT (1 << 25) #define GEN6_PM_THERMAL_EVENT (1 << 24) + +/* + * For Gen11 these are in the upper word of the GPM_WGBOXPERF + * registers. Shifting is handled on accessing the imr and ier. + */ #define GEN6_PM_RP_DOWN_TIMEOUT (1 << 6) #define GEN6_PM_RP_UP_THRESHOLD (1 << 5) #define GEN6_PM_RP_DOWN_THRESHOLD (1 << 4) @@ -10127,6 +10113,7 @@ enum skl_power_gate { #define PAL_PREC_SPLIT_MODE (1 << 31) #define PAL_PREC_AUTO_INCREMENT (1 << 15) #define PAL_PREC_INDEX_VALUE_MASK (0x3ff << 0) +#define PAL_PREC_INDEX_VALUE(x) ((x) << 0) #define _PAL_PREC_DATA_A 0x4A404 #define _PAL_PREC_DATA_B 0x4AC04 #define _PAL_PREC_DATA_C 0x4B404 @@ -10144,6 +10131,7 @@ enum skl_power_gate { #define PREC_PAL_DATA(pipe) _MMIO_PIPE(pipe, _PAL_PREC_DATA_A, _PAL_PREC_DATA_B) #define PREC_PAL_GC_MAX(pipe, i) _MMIO(_PIPE(pipe, _PAL_PREC_GC_MAX_A, _PAL_PREC_GC_MAX_B) + (i) * 4) #define PREC_PAL_EXT_GC_MAX(pipe, i) _MMIO(_PIPE(pipe, _PAL_PREC_EXT_GC_MAX_A, _PAL_PREC_EXT_GC_MAX_B) + (i) * 4) +#define PREC_PAL_EXT2_GC_MAX(pipe, i) _MMIO(_PIPE(pipe, _PAL_PREC_EXT2_GC_MAX_A, _PAL_PREC_EXT2_GC_MAX_B) + (i) * 4) #define _PRE_CSC_GAMC_INDEX_A 0x4A484 #define _PRE_CSC_GAMC_INDEX_B 0x4AC84 diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index e9c2094ab8ea..b836721d3b13 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -29,10 +29,11 @@ #include <linux/sched/clock.h> #include <linux/sched/signal.h> -#include "i915_drv.h" #include "i915_active.h" +#include "i915_drv.h" #include "i915_globals.h" #include "i915_reset.h" +#include "intel_pm.h" struct execute_cb { struct list_head link; @@ -100,6 +101,7 @@ static void i915_fence_release(struct dma_fence *fence) * caught trying to reuse dead objects. */ i915_sw_fence_fini(&rq->submit); + i915_sw_fence_fini(&rq->semaphore); kmem_cache_free(global.slab_requests, rq); } @@ -551,6 +553,36 @@ submit_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state) return NOTIFY_DONE; } +static int __i915_sw_fence_call +semaphore_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state) +{ + struct i915_request *request = + container_of(fence, typeof(*request), semaphore); + + switch (state) { + case FENCE_COMPLETE: + /* + * We only check a small portion of our dependencies + * and so cannot guarantee that there remains no + * semaphore chain across all. Instead of opting + * for the full NOSEMAPHORE boost, we go for the + * smaller (but still preempting) boost of + * NEWCLIENT. This will be enough to boost over + * a busywaiting request (as that cannot be + * NEWCLIENT) without accidentally boosting + * a busywait over real work elsewhere. + */ + i915_schedule_bump_priority(request, I915_PRIORITY_NEWCLIENT); + break; + + case FENCE_FREE: + i915_request_put(request); + break; + } + + return NOTIFY_DONE; +} + static void ring_retire_requests(struct intel_ring *ring) { struct i915_request *rq, *rn; @@ -583,11 +615,6 @@ out: return kmem_cache_alloc(global.slab_requests, GFP_KERNEL); } -static int add_timeline_barrier(struct i915_request *rq) -{ - return i915_request_await_active_request(rq, &rq->timeline->barrier); -} - /** * i915_request_alloc - allocate a request structure * @@ -706,6 +733,7 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx) /* We bump the ref for the fence chain */ i915_sw_fence_init(&i915_request_get(rq)->submit, submit_notify); + i915_sw_fence_init(&i915_request_get(rq)->semaphore, semaphore_notify); i915_sched_node_init(&rq->sched); @@ -737,10 +765,6 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx) */ rq->head = rq->ring->emit; - ret = add_timeline_barrier(rq); - if (ret) - goto err_unwind; - ret = engine->request_alloc(rq); if (ret) goto err_unwind; @@ -751,7 +775,10 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx) rq->infix = rq->ring->emit; /* end of header; start of user payload */ /* Check that we didn't interrupt ourselves with a new request */ + lockdep_assert_held(&rq->timeline->mutex); GEM_BUG_ON(rq->timeline->seqno != rq->fence.seqno); + rq->cookie = lockdep_pin_lock(&rq->timeline->mutex); + return rq; err_unwind: @@ -783,6 +810,18 @@ emit_semaphore_wait(struct i915_request *to, GEM_BUG_ON(!from->timeline->has_initial_breadcrumb); GEM_BUG_ON(INTEL_GEN(to->i915) < 8); + /* Just emit the first semaphore we see as request space is limited. */ + if (to->sched.semaphores & from->engine->mask) + return i915_sw_fence_await_dma_fence(&to->submit, + &from->fence, 0, + I915_FENCE_GFP); + + err = i915_sw_fence_await_dma_fence(&to->semaphore, + &from->fence, 0, + I915_FENCE_GFP); + if (err < 0) + return err; + /* We need to pin the signaler's HWSP until we are finished reading. */ err = i915_timeline_read_hwsp(from, to, &hwsp_offset); if (err) @@ -814,7 +853,8 @@ emit_semaphore_wait(struct i915_request *to, *cs++ = 0; intel_ring_advance(to, cs); - to->sched.flags |= I915_SCHED_HAS_SEMAPHORE; + to->sched.semaphores |= from->engine->mask; + to->sched.flags |= I915_SCHED_HAS_SEMAPHORE_CHAIN; return 0; } @@ -1063,6 +1103,8 @@ void i915_request_add(struct i915_request *request) engine->name, request->fence.context, request->fence.seqno); lockdep_assert_held(&request->timeline->mutex); + lockdep_unpin_lock(&request->timeline->mutex, request->cookie); + trace_i915_request_add(request); /* @@ -1110,6 +1152,7 @@ void i915_request_add(struct i915_request *request) * run at the earliest possible convenience. */ local_bh_disable(); + i915_sw_fence_commit(&request->semaphore); rcu_read_lock(); /* RCU serialisation for set-wedged protection */ if (engine->schedule) { struct i915_sched_attr attr = request->gem_context->sched; @@ -1126,7 +1169,7 @@ void i915_request_add(struct i915_request *request) * far in the distance past over useful work, we keep a history * of any semaphore use along our dependency chain. */ - if (!(request->sched.flags & I915_SCHED_HAS_SEMAPHORE)) + if (!(request->sched.flags & I915_SCHED_HAS_SEMAPHORE_CHAIN)) attr.priority |= I915_PRIORITY_NOSEMAPHORE; /* @@ -1316,7 +1359,9 @@ long i915_request_wait(struct i915_request *rq, if (flags & I915_WAIT_PRIORITY) { if (!i915_request_started(rq) && INTEL_GEN(rq->i915) >= 6) gen6_rps_boost(rq); + local_bh_disable(); /* suspend tasklets for reprioritisation */ i915_schedule_bump_priority(rq, I915_PRIORITY_WAIT); + local_bh_enable(); /* kick tasklets en masse */ } wait.tsk = current; diff --git a/drivers/gpu/drm/i915/i915_request.h b/drivers/gpu/drm/i915/i915_request.h index cd6c130964cd..a982664618c2 100644 --- a/drivers/gpu/drm/i915/i915_request.h +++ b/drivers/gpu/drm/i915/i915_request.h @@ -26,6 +26,7 @@ #define I915_REQUEST_H #include <linux/dma-fence.h> +#include <linux/lockdep.h> #include "i915_gem.h" #include "i915_scheduler.h" @@ -121,6 +122,15 @@ struct i915_request { unsigned long rcustate; /* + * We pin the timeline->mutex while constructing the request to + * ensure that no caller accidentally drops it during construction. + * The timeline->mutex must be held to ensure that only this caller + * can use the ring and manipulate the associated timeline during + * construction. + */ + struct pin_cookie cookie; + + /* * Fences for the various phases in the request's lifetime. * * The submit fence is used to await upon all of the request's @@ -133,6 +143,7 @@ struct i915_request { struct i915_sw_dma_fence_cb dmaq; }; struct list_head execute_cb; + struct i915_sw_fence semaphore; /* * A list of everyone we wait upon, and everyone who waits upon us. diff --git a/drivers/gpu/drm/i915/i915_reset.c b/drivers/gpu/drm/i915/i915_reset.c index 2f25ed702ba0..677d59304e78 100644 --- a/drivers/gpu/drm/i915/i915_reset.c +++ b/drivers/gpu/drm/i915/i915_reset.c @@ -18,6 +18,26 @@ /* XXX How to handle concurrent GGTT updates using tiling registers? */ #define RESET_UNDER_STOP_MACHINE 0 +static void rmw_set(struct intel_uncore *uncore, i915_reg_t reg, u32 set) +{ + intel_uncore_rmw(uncore, reg, 0, set); +} + +static void rmw_clear(struct intel_uncore *uncore, i915_reg_t reg, u32 clr) +{ + intel_uncore_rmw(uncore, reg, clr, 0); +} + +static void rmw_set_fw(struct intel_uncore *uncore, i915_reg_t reg, u32 set) +{ + intel_uncore_rmw_fw(uncore, reg, 0, set); +} + +static void rmw_clear_fw(struct intel_uncore *uncore, i915_reg_t reg, u32 clr) +{ + intel_uncore_rmw_fw(uncore, reg, clr, 0); +} + static void engine_skip_context(struct i915_request *rq) { struct intel_engine_cs *engine = rq->engine; @@ -119,7 +139,7 @@ void i915_reset_request(struct i915_request *rq, bool guilty) static void gen3_stop_engine(struct intel_engine_cs *engine) { - struct drm_i915_private *dev_priv = engine->i915; + struct intel_uncore *uncore = engine->uncore; const u32 base = engine->mmio_base; GEM_TRACE("%s\n", engine->name); @@ -127,32 +147,35 @@ static void gen3_stop_engine(struct intel_engine_cs *engine) if (intel_engine_stop_cs(engine)) GEM_TRACE("%s: timed out on STOP_RING\n", engine->name); - I915_WRITE_FW(RING_HEAD(base), I915_READ_FW(RING_TAIL(base))); - POSTING_READ_FW(RING_HEAD(base)); /* paranoia */ + intel_uncore_write_fw(uncore, + RING_HEAD(base), + intel_uncore_read_fw(uncore, RING_TAIL(base))); + intel_uncore_posting_read_fw(uncore, RING_HEAD(base)); /* paranoia */ - I915_WRITE_FW(RING_HEAD(base), 0); - I915_WRITE_FW(RING_TAIL(base), 0); - POSTING_READ_FW(RING_TAIL(base)); + intel_uncore_write_fw(uncore, RING_HEAD(base), 0); + intel_uncore_write_fw(uncore, RING_TAIL(base), 0); + intel_uncore_posting_read_fw(uncore, RING_TAIL(base)); /* The ring must be empty before it is disabled */ - I915_WRITE_FW(RING_CTL(base), 0); + intel_uncore_write_fw(uncore, RING_CTL(base), 0); /* Check acts as a post */ - if (I915_READ_FW(RING_HEAD(base))) + if (intel_uncore_read_fw(uncore, RING_HEAD(base))) GEM_TRACE("%s: ring head [%x] not parked\n", - engine->name, I915_READ_FW(RING_HEAD(base))); + engine->name, + intel_uncore_read_fw(uncore, RING_HEAD(base))); } static void i915_stop_engines(struct drm_i915_private *i915, - unsigned int engine_mask) + intel_engine_mask_t engine_mask) { struct intel_engine_cs *engine; - enum intel_engine_id id; + intel_engine_mask_t tmp; if (INTEL_GEN(i915) < 3) return; - for_each_engine_masked(engine, i915, engine_mask, id) + for_each_engine_masked(engine, i915, engine_mask, tmp) gen3_stop_engine(engine); } @@ -165,7 +188,7 @@ static bool i915_in_reset(struct pci_dev *pdev) } static int i915_do_reset(struct drm_i915_private *i915, - unsigned int engine_mask, + intel_engine_mask_t engine_mask, unsigned int retry) { struct pci_dev *pdev = i915->drm.pdev; @@ -194,7 +217,7 @@ static bool g4x_reset_complete(struct pci_dev *pdev) } static int g33_do_reset(struct drm_i915_private *i915, - unsigned int engine_mask, + intel_engine_mask_t engine_mask, unsigned int retry) { struct pci_dev *pdev = i915->drm.pdev; @@ -203,17 +226,17 @@ static int g33_do_reset(struct drm_i915_private *i915, return wait_for_atomic(g4x_reset_complete(pdev), 50); } -static int g4x_do_reset(struct drm_i915_private *dev_priv, - unsigned int engine_mask, +static int g4x_do_reset(struct drm_i915_private *i915, + intel_engine_mask_t engine_mask, unsigned int retry) { - struct pci_dev *pdev = dev_priv->drm.pdev; + struct pci_dev *pdev = i915->drm.pdev; + struct intel_uncore *uncore = &i915->uncore; int ret; /* WaVcpClkGateDisableForMediaReset:ctg,elk */ - I915_WRITE_FW(VDECCLK_GATE_D, - I915_READ(VDECCLK_GATE_D) | VCP_UNIT_CLOCK_GATE_DISABLE); - POSTING_READ_FW(VDECCLK_GATE_D); + rmw_set_fw(uncore, VDECCLK_GATE_D, VCP_UNIT_CLOCK_GATE_DISABLE); + intel_uncore_posting_read_fw(uncore, VDECCLK_GATE_D); pci_write_config_byte(pdev, I915_GDRST, GRDOM_MEDIA | GRDOM_RESET_ENABLE); @@ -234,18 +257,17 @@ static int g4x_do_reset(struct drm_i915_private *dev_priv, out: pci_write_config_byte(pdev, I915_GDRST, 0); - I915_WRITE_FW(VDECCLK_GATE_D, - I915_READ(VDECCLK_GATE_D) & ~VCP_UNIT_CLOCK_GATE_DISABLE); - POSTING_READ_FW(VDECCLK_GATE_D); + rmw_clear_fw(uncore, VDECCLK_GATE_D, VCP_UNIT_CLOCK_GATE_DISABLE); + intel_uncore_posting_read_fw(uncore, VDECCLK_GATE_D); return ret; } -static int ironlake_do_reset(struct drm_i915_private *dev_priv, - unsigned int engine_mask, +static int ironlake_do_reset(struct drm_i915_private *i915, + intel_engine_mask_t engine_mask, unsigned int retry) { - struct intel_uncore *uncore = &dev_priv->uncore; + struct intel_uncore *uncore = &i915->uncore; int ret; intel_uncore_write_fw(uncore, ILK_GDSR, @@ -277,10 +299,10 @@ out: } /* Reset the hardware domains (GENX_GRDOM_*) specified by mask */ -static int gen6_hw_domain_reset(struct drm_i915_private *dev_priv, +static int gen6_hw_domain_reset(struct drm_i915_private *i915, u32 hw_domain_mask) { - struct intel_uncore *uncore = &dev_priv->uncore; + struct intel_uncore *uncore = &i915->uncore; int err; /* @@ -303,7 +325,7 @@ static int gen6_hw_domain_reset(struct drm_i915_private *dev_priv, } static int gen6_reset_engines(struct drm_i915_private *i915, - unsigned int engine_mask, + intel_engine_mask_t engine_mask, unsigned int retry) { struct intel_engine_cs *engine; @@ -319,7 +341,7 @@ static int gen6_reset_engines(struct drm_i915_private *i915, if (engine_mask == ALL_ENGINES) { hw_mask = GEN6_GRDOM_FULL; } else { - unsigned int tmp; + intel_engine_mask_t tmp; hw_mask = 0; for_each_engine_masked(engine, i915, engine_mask, tmp) { @@ -331,11 +353,10 @@ static int gen6_reset_engines(struct drm_i915_private *i915, return gen6_hw_domain_reset(i915, hw_mask); } -static u32 gen11_lock_sfc(struct drm_i915_private *dev_priv, - struct intel_engine_cs *engine) +static u32 gen11_lock_sfc(struct intel_engine_cs *engine) { - struct intel_uncore *uncore = &dev_priv->uncore; - u8 vdbox_sfc_access = RUNTIME_INFO(dev_priv)->vdbox_sfc_access; + struct intel_uncore *uncore = engine->uncore; + u8 vdbox_sfc_access = RUNTIME_INFO(engine->i915)->vdbox_sfc_access; i915_reg_t sfc_forced_lock, sfc_forced_lock_ack; u32 sfc_forced_lock_bit, sfc_forced_lock_ack_bit; i915_reg_t sfc_usage; @@ -382,7 +403,7 @@ static u32 gen11_lock_sfc(struct drm_i915_private *dev_priv, * ends up being locked to the engine we want to reset, we have to reset * it as well (we will unlock it once the reset sequence is completed). */ - intel_uncore_rmw_or_fw(uncore, sfc_forced_lock, sfc_forced_lock_bit); + rmw_set_fw(uncore, sfc_forced_lock, sfc_forced_lock_bit); if (__intel_wait_for_register_fw(uncore, sfc_forced_lock_ack, @@ -399,10 +420,10 @@ static u32 gen11_lock_sfc(struct drm_i915_private *dev_priv, return 0; } -static void gen11_unlock_sfc(struct drm_i915_private *dev_priv, - struct intel_engine_cs *engine) +static void gen11_unlock_sfc(struct intel_engine_cs *engine) { - u8 vdbox_sfc_access = RUNTIME_INFO(dev_priv)->vdbox_sfc_access; + struct intel_uncore *uncore = engine->uncore; + u8 vdbox_sfc_access = RUNTIME_INFO(engine->i915)->vdbox_sfc_access; i915_reg_t sfc_forced_lock; u32 sfc_forced_lock_bit; @@ -424,12 +445,11 @@ static void gen11_unlock_sfc(struct drm_i915_private *dev_priv, return; } - I915_WRITE_FW(sfc_forced_lock, - I915_READ_FW(sfc_forced_lock) & ~sfc_forced_lock_bit); + rmw_clear_fw(uncore, sfc_forced_lock, sfc_forced_lock_bit); } static int gen11_reset_engines(struct drm_i915_private *i915, - unsigned int engine_mask, + intel_engine_mask_t engine_mask, unsigned int retry) { const u32 hw_engine_mask[] = { @@ -443,7 +463,7 @@ static int gen11_reset_engines(struct drm_i915_private *i915, [VECS1] = GEN11_GRDOM_VECS2, }; struct intel_engine_cs *engine; - unsigned int tmp; + intel_engine_mask_t tmp; u32 hw_mask; int ret; @@ -454,7 +474,7 @@ static int gen11_reset_engines(struct drm_i915_private *i915, for_each_engine_masked(engine, i915, engine_mask, tmp) { GEM_BUG_ON(engine->id >= ARRAY_SIZE(hw_engine_mask)); hw_mask |= hw_engine_mask[engine->id]; - hw_mask |= gen11_lock_sfc(i915, engine); + hw_mask |= gen11_lock_sfc(engine); } } @@ -462,46 +482,62 @@ static int gen11_reset_engines(struct drm_i915_private *i915, if (engine_mask != ALL_ENGINES) for_each_engine_masked(engine, i915, engine_mask, tmp) - gen11_unlock_sfc(i915, engine); + gen11_unlock_sfc(engine); return ret; } static int gen8_engine_reset_prepare(struct intel_engine_cs *engine) { - struct intel_uncore *uncore = &engine->i915->uncore; + struct intel_uncore *uncore = engine->uncore; + const i915_reg_t reg = RING_RESET_CTL(engine->mmio_base); + u32 request, mask, ack; int ret; - intel_uncore_write_fw(uncore, RING_RESET_CTL(engine->mmio_base), - _MASKED_BIT_ENABLE(RESET_CTL_REQUEST_RESET)); + ack = intel_uncore_read_fw(uncore, reg); + if (ack & RESET_CTL_CAT_ERROR) { + /* + * For catastrophic errors, ready-for-reset sequence + * needs to be bypassed: HAS#396813 + */ + request = RESET_CTL_CAT_ERROR; + mask = RESET_CTL_CAT_ERROR; + + /* Catastrophic errors need to be cleared by HW */ + ack = 0; + } else if (!(ack & RESET_CTL_READY_TO_RESET)) { + request = RESET_CTL_REQUEST_RESET; + mask = RESET_CTL_READY_TO_RESET; + ack = RESET_CTL_READY_TO_RESET; + } else { + return 0; + } - ret = __intel_wait_for_register_fw(uncore, - RING_RESET_CTL(engine->mmio_base), - RESET_CTL_READY_TO_RESET, - RESET_CTL_READY_TO_RESET, - 700, 0, - NULL); + intel_uncore_write_fw(uncore, reg, _MASKED_BIT_ENABLE(request)); + ret = __intel_wait_for_register_fw(uncore, reg, mask, ack, + 700, 0, NULL); if (ret) - DRM_ERROR("%s: reset request timeout\n", engine->name); + DRM_ERROR("%s reset request timed out: {request: %08x, RESET_CTL: %08x}\n", + engine->name, request, + intel_uncore_read_fw(uncore, reg)); return ret; } static void gen8_engine_reset_cancel(struct intel_engine_cs *engine) { - struct drm_i915_private *dev_priv = engine->i915; - - I915_WRITE_FW(RING_RESET_CTL(engine->mmio_base), - _MASKED_BIT_DISABLE(RESET_CTL_REQUEST_RESET)); + intel_uncore_write_fw(engine->uncore, + RING_RESET_CTL(engine->mmio_base), + _MASKED_BIT_DISABLE(RESET_CTL_REQUEST_RESET)); } static int gen8_reset_engines(struct drm_i915_private *i915, - unsigned int engine_mask, + intel_engine_mask_t engine_mask, unsigned int retry) { struct intel_engine_cs *engine; const bool reset_non_ready = retry >= 1; - unsigned int tmp; + intel_engine_mask_t tmp; int ret; for_each_engine_masked(engine, i915, engine_mask, tmp) { @@ -537,7 +573,7 @@ skip_reset: } typedef int (*reset_func)(struct drm_i915_private *, - unsigned int engine_mask, + intel_engine_mask_t engine_mask, unsigned int retry); static reset_func intel_get_gpu_reset(struct drm_i915_private *i915) @@ -558,7 +594,8 @@ static reset_func intel_get_gpu_reset(struct drm_i915_private *i915) return NULL; } -int intel_gpu_reset(struct drm_i915_private *i915, unsigned int engine_mask) +int intel_gpu_reset(struct drm_i915_private *i915, + intel_engine_mask_t engine_mask) { const int retries = engine_mask == ALL_ENGINES ? RESET_MAX_RETRIES : 1; reset_func reset; @@ -646,7 +683,7 @@ static void reset_prepare_engine(struct intel_engine_cs *engine) * written to the powercontext is undefined and so we may lose * GPU state upon resume, i.e. fail to restart after a reset. */ - intel_uncore_forcewake_get(&engine->i915->uncore, FORCEWAKE_ALL); + intel_uncore_forcewake_get(engine->uncore, FORCEWAKE_ALL); engine->reset.prepare(engine); } @@ -692,7 +729,8 @@ static void gt_revoke(struct drm_i915_private *i915) revoke_mmaps(i915); } -static int gt_reset(struct drm_i915_private *i915, unsigned int stalled_mask) +static int gt_reset(struct drm_i915_private *i915, + intel_engine_mask_t stalled_mask) { struct intel_engine_cs *engine; enum intel_engine_id id; @@ -717,7 +755,7 @@ static int gt_reset(struct drm_i915_private *i915, unsigned int stalled_mask) static void reset_finish_engine(struct intel_engine_cs *engine) { engine->reset.finish(engine); - intel_uncore_forcewake_put(&engine->i915->uncore, FORCEWAKE_ALL); + intel_uncore_forcewake_put(engine->uncore, FORCEWAKE_ALL); } struct i915_gpu_restart { @@ -951,7 +989,8 @@ bool i915_gem_unset_wedged(struct drm_i915_private *i915) return result; } -static int do_reset(struct drm_i915_private *i915, unsigned int stalled_mask) +static int do_reset(struct drm_i915_private *i915, + intel_engine_mask_t stalled_mask) { int err, i; @@ -986,7 +1025,7 @@ static int do_reset(struct drm_i915_private *i915, unsigned int stalled_mask) * - re-init display */ void i915_reset(struct drm_i915_private *i915, - unsigned int stalled_mask, + intel_engine_mask_t stalled_mask, const char *reason) { struct i915_gpu_error *error = &i915->gpu_error; @@ -1173,49 +1212,50 @@ static void i915_reset_device(struct drm_i915_private *i915, kobject_uevent_env(kobj, KOBJ_CHANGE, reset_done_event); } -static void clear_register(struct drm_i915_private *dev_priv, i915_reg_t reg) +static void clear_register(struct intel_uncore *uncore, i915_reg_t reg) { - I915_WRITE(reg, I915_READ(reg)); + intel_uncore_rmw(uncore, reg, 0, 0); } -void i915_clear_error_registers(struct drm_i915_private *dev_priv) +void i915_clear_error_registers(struct drm_i915_private *i915) { + struct intel_uncore *uncore = &i915->uncore; u32 eir; - if (!IS_GEN(dev_priv, 2)) - clear_register(dev_priv, PGTBL_ER); + if (!IS_GEN(i915, 2)) + clear_register(uncore, PGTBL_ER); - if (INTEL_GEN(dev_priv) < 4) - clear_register(dev_priv, IPEIR(RENDER_RING_BASE)); + if (INTEL_GEN(i915) < 4) + clear_register(uncore, IPEIR(RENDER_RING_BASE)); else - clear_register(dev_priv, IPEIR_I965); + clear_register(uncore, IPEIR_I965); - clear_register(dev_priv, EIR); - eir = I915_READ(EIR); + clear_register(uncore, EIR); + eir = intel_uncore_read(uncore, EIR); if (eir) { /* * some errors might have become stuck, * mask them. */ DRM_DEBUG_DRIVER("EIR stuck: 0x%08x, masking\n", eir); - I915_WRITE(EMR, I915_READ(EMR) | eir); - I915_WRITE(IIR, I915_MASTER_ERROR_INTERRUPT); + rmw_set(uncore, EMR, eir); + intel_uncore_write(uncore, GEN2_IIR, + I915_MASTER_ERROR_INTERRUPT); } - if (INTEL_GEN(dev_priv) >= 8) { - I915_WRITE(GEN8_RING_FAULT_REG, - I915_READ(GEN8_RING_FAULT_REG) & ~RING_FAULT_VALID); - POSTING_READ(GEN8_RING_FAULT_REG); - } else if (INTEL_GEN(dev_priv) >= 6) { + if (INTEL_GEN(i915) >= 8) { + rmw_clear(uncore, GEN8_RING_FAULT_REG, RING_FAULT_VALID); + intel_uncore_posting_read(uncore, GEN8_RING_FAULT_REG); + } else if (INTEL_GEN(i915) >= 6) { struct intel_engine_cs *engine; enum intel_engine_id id; - for_each_engine(engine, dev_priv, id) { - I915_WRITE(RING_FAULT_REG(engine), - I915_READ(RING_FAULT_REG(engine)) & - ~RING_FAULT_VALID); + for_each_engine(engine, i915, id) { + rmw_clear(uncore, + RING_FAULT_REG(engine), RING_FAULT_VALID); + intel_uncore_posting_read(uncore, + RING_FAULT_REG(engine)); } - POSTING_READ(RING_FAULT_REG(dev_priv->engine[RCS0])); } } @@ -1233,14 +1273,14 @@ void i915_clear_error_registers(struct drm_i915_private *dev_priv) * of a ring dump etc.). */ void i915_handle_error(struct drm_i915_private *i915, - u32 engine_mask, + intel_engine_mask_t engine_mask, unsigned long flags, const char *fmt, ...) { struct i915_gpu_error *error = &i915->gpu_error; struct intel_engine_cs *engine; intel_wakeref_t wakeref; - unsigned int tmp; + intel_engine_mask_t tmp; char error_msg[80]; char *msg = NULL; diff --git a/drivers/gpu/drm/i915/i915_reset.h b/drivers/gpu/drm/i915/i915_reset.h index 16f2389f656f..3c0450289b8f 100644 --- a/drivers/gpu/drm/i915/i915_reset.h +++ b/drivers/gpu/drm/i915/i915_reset.h @@ -11,13 +11,16 @@ #include <linux/types.h> #include <linux/srcu.h> +#include "intel_engine_types.h" + struct drm_i915_private; +struct i915_request; struct intel_engine_cs; struct intel_guc; __printf(4, 5) void i915_handle_error(struct drm_i915_private *i915, - u32 engine_mask, + intel_engine_mask_t engine_mask, unsigned long flags, const char *fmt, ...); #define I915_ERROR_CAPTURE BIT(0) @@ -25,7 +28,7 @@ void i915_handle_error(struct drm_i915_private *i915, void i915_clear_error_registers(struct drm_i915_private *i915); void i915_reset(struct drm_i915_private *i915, - unsigned int stalled_mask, + intel_engine_mask_t stalled_mask, const char *reason); int i915_reset_engine(struct intel_engine_cs *engine, const char *reason); @@ -41,7 +44,8 @@ int i915_terminally_wedged(struct drm_i915_private *i915); bool intel_has_gpu_reset(struct drm_i915_private *i915); bool intel_has_reset_engine(struct drm_i915_private *i915); -int intel_gpu_reset(struct drm_i915_private *i915, u32 engine_mask); +int intel_gpu_reset(struct drm_i915_private *i915, + intel_engine_mask_t engine_mask); int intel_reset_guc(struct drm_i915_private *i915); diff --git a/drivers/gpu/drm/i915/i915_scheduler.c b/drivers/gpu/drm/i915/i915_scheduler.c index e0f609d01564..39bc4f54e272 100644 --- a/drivers/gpu/drm/i915/i915_scheduler.c +++ b/drivers/gpu/drm/i915/i915_scheduler.c @@ -41,6 +41,7 @@ void i915_sched_node_init(struct i915_sched_node *node) INIT_LIST_HEAD(&node->waiters_list); INIT_LIST_HEAD(&node->link); node->attr.priority = I915_PRIORITY_INVALID; + node->semaphores = 0; node->flags = 0; } @@ -63,7 +64,7 @@ bool __i915_sched_node_add_dependency(struct i915_sched_node *node, { bool ret = false; - spin_lock(&schedule_lock); + spin_lock_irq(&schedule_lock); if (!node_signaled(signal)) { INIT_LIST_HEAD(&dep->dfs_link); @@ -73,14 +74,14 @@ bool __i915_sched_node_add_dependency(struct i915_sched_node *node, dep->flags = flags; /* Keep track of whether anyone on this chain has a semaphore */ - if (signal->flags & I915_SCHED_HAS_SEMAPHORE && + if (signal->flags & I915_SCHED_HAS_SEMAPHORE_CHAIN && !node_started(signal)) - node->flags |= I915_SCHED_HAS_SEMAPHORE; + node->flags |= I915_SCHED_HAS_SEMAPHORE_CHAIN; ret = true; } - spin_unlock(&schedule_lock); + spin_unlock_irq(&schedule_lock); return ret; } @@ -107,7 +108,7 @@ void i915_sched_node_fini(struct i915_sched_node *node) GEM_BUG_ON(!list_empty(&node->link)); - spin_lock(&schedule_lock); + spin_lock_irq(&schedule_lock); /* * Everyone we depended upon (the fences we wait to be signaled) @@ -134,7 +135,7 @@ void i915_sched_node_fini(struct i915_sched_node *node) i915_dependency_free(dep); } - spin_unlock(&schedule_lock); + spin_unlock_irq(&schedule_lock); } static inline struct i915_priolist *to_priolist(struct rb_node *rb) @@ -355,7 +356,7 @@ static void __i915_schedule(struct i915_request *rq, memset(&cache, 0, sizeof(cache)); engine = rq->engine; - spin_lock_irq(&engine->timeline.lock); + spin_lock(&engine->timeline.lock); /* Fifo and depth-first replacement ensure our deps execute before us */ list_for_each_entry_safe_reverse(dep, p, &dfs, dfs_link) { @@ -406,32 +407,33 @@ static void __i915_schedule(struct i915_request *rq, tasklet_hi_schedule(&engine->execlists.tasklet); } - spin_unlock_irq(&engine->timeline.lock); + spin_unlock(&engine->timeline.lock); } void i915_schedule(struct i915_request *rq, const struct i915_sched_attr *attr) { - spin_lock(&schedule_lock); + spin_lock_irq(&schedule_lock); __i915_schedule(rq, attr); - spin_unlock(&schedule_lock); + spin_unlock_irq(&schedule_lock); } void i915_schedule_bump_priority(struct i915_request *rq, unsigned int bump) { struct i915_sched_attr attr; + unsigned long flags; GEM_BUG_ON(bump & ~I915_PRIORITY_MASK); if (READ_ONCE(rq->sched.attr.priority) == I915_PRIORITY_INVALID) return; - spin_lock_bh(&schedule_lock); + spin_lock_irqsave(&schedule_lock, flags); attr = rq->sched.attr; attr.priority |= bump; __i915_schedule(rq, &attr); - spin_unlock_bh(&schedule_lock); + spin_unlock_irqrestore(&schedule_lock, flags); } void __i915_priolist_free(struct i915_priolist *p) diff --git a/drivers/gpu/drm/i915/i915_scheduler.h b/drivers/gpu/drm/i915/i915_scheduler.h index 9a1d257f3d6e..07d243acf553 100644 --- a/drivers/gpu/drm/i915/i915_scheduler.h +++ b/drivers/gpu/drm/i915/i915_scheduler.h @@ -8,92 +8,10 @@ #define _I915_SCHEDULER_H_ #include <linux/bitops.h> +#include <linux/list.h> #include <linux/kernel.h> -#include <uapi/drm/i915_drm.h> - -struct drm_i915_private; -struct i915_request; -struct intel_engine_cs; - -enum { - I915_PRIORITY_MIN = I915_CONTEXT_MIN_USER_PRIORITY - 1, - I915_PRIORITY_NORMAL = I915_CONTEXT_DEFAULT_PRIORITY, - I915_PRIORITY_MAX = I915_CONTEXT_MAX_USER_PRIORITY + 1, - - I915_PRIORITY_INVALID = INT_MIN -}; - -#define I915_USER_PRIORITY_SHIFT 3 -#define I915_USER_PRIORITY(x) ((x) << I915_USER_PRIORITY_SHIFT) - -#define I915_PRIORITY_COUNT BIT(I915_USER_PRIORITY_SHIFT) -#define I915_PRIORITY_MASK (I915_PRIORITY_COUNT - 1) - -#define I915_PRIORITY_WAIT ((u8)BIT(0)) -#define I915_PRIORITY_NEWCLIENT ((u8)BIT(1)) -#define I915_PRIORITY_NOSEMAPHORE ((u8)BIT(2)) - -#define __NO_PREEMPTION (I915_PRIORITY_WAIT) - -struct i915_sched_attr { - /** - * @priority: execution and service priority - * - * All clients are equal, but some are more equal than others! - * - * Requests from a context with a greater (more positive) value of - * @priority will be executed before those with a lower @priority - * value, forming a simple QoS. - * - * The &drm_i915_private.kernel_context is assigned the lowest priority. - */ - int priority; -}; - -/* - * "People assume that time is a strict progression of cause to effect, but - * actually, from a nonlinear, non-subjective viewpoint, it's more like a big - * ball of wibbly-wobbly, timey-wimey ... stuff." -The Doctor, 2015 - * - * Requests exist in a complex web of interdependencies. Each request - * has to wait for some other request to complete before it is ready to be run - * (e.g. we have to wait until the pixels have been rendering into a texture - * before we can copy from it). We track the readiness of a request in terms - * of fences, but we also need to keep the dependency tree for the lifetime - * of the request (beyond the life of an individual fence). We use the tree - * at various points to reorder the requests whilst keeping the requests - * in order with respect to their various dependencies. - * - * There is no active component to the "scheduler". As we know the dependency - * DAG of each request, we are able to insert it into a sorted queue when it - * is ready, and are able to reorder its portion of the graph to accommodate - * dynamic priority changes. - */ -struct i915_sched_node { - struct list_head signalers_list; /* those before us, we depend upon */ - struct list_head waiters_list; /* those after us, they depend upon us */ - struct list_head link; - struct i915_sched_attr attr; - unsigned int flags; -#define I915_SCHED_HAS_SEMAPHORE BIT(0) -}; - -struct i915_dependency { - struct i915_sched_node *signaler; - struct list_head signal_link; - struct list_head wait_link; - struct list_head dfs_link; - unsigned long flags; -#define I915_DEPENDENCY_ALLOC BIT(0) -}; - -struct i915_priolist { - struct list_head requests[I915_PRIORITY_COUNT]; - struct rb_node node; - unsigned long used; - int priority; -}; +#include "i915_scheduler_types.h" #define priolist_for_each_request(it, plist, idx) \ for (idx = 0; idx < ARRAY_SIZE((plist)->requests); idx++) \ diff --git a/drivers/gpu/drm/i915/i915_scheduler_types.h b/drivers/gpu/drm/i915/i915_scheduler_types.h new file mode 100644 index 000000000000..f1af3916a808 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_scheduler_types.h @@ -0,0 +1,72 @@ +/* + * SPDX-License-Identifier: MIT + * + * Copyright © 2018 Intel Corporation + */ + +#ifndef _I915_SCHEDULER_TYPES_H_ +#define _I915_SCHEDULER_TYPES_H_ + +#include <linux/list.h> + +#include "i915_priolist_types.h" +#include "intel_engine_types.h" + +struct drm_i915_private; +struct i915_request; +struct intel_engine_cs; + +struct i915_sched_attr { + /** + * @priority: execution and service priority + * + * All clients are equal, but some are more equal than others! + * + * Requests from a context with a greater (more positive) value of + * @priority will be executed before those with a lower @priority + * value, forming a simple QoS. + * + * The &drm_i915_private.kernel_context is assigned the lowest priority. + */ + int priority; +}; + +/* + * "People assume that time is a strict progression of cause to effect, but + * actually, from a nonlinear, non-subjective viewpoint, it's more like a big + * ball of wibbly-wobbly, timey-wimey ... stuff." -The Doctor, 2015 + * + * Requests exist in a complex web of interdependencies. Each request + * has to wait for some other request to complete before it is ready to be run + * (e.g. we have to wait until the pixels have been rendering into a texture + * before we can copy from it). We track the readiness of a request in terms + * of fences, but we also need to keep the dependency tree for the lifetime + * of the request (beyond the life of an individual fence). We use the tree + * at various points to reorder the requests whilst keeping the requests + * in order with respect to their various dependencies. + * + * There is no active component to the "scheduler". As we know the dependency + * DAG of each request, we are able to insert it into a sorted queue when it + * is ready, and are able to reorder its portion of the graph to accommodate + * dynamic priority changes. + */ +struct i915_sched_node { + struct list_head signalers_list; /* those before us, we depend upon */ + struct list_head waiters_list; /* those after us, they depend upon us */ + struct list_head link; + struct i915_sched_attr attr; + unsigned int flags; +#define I915_SCHED_HAS_SEMAPHORE_CHAIN BIT(0) + intel_engine_mask_t semaphores; +}; + +struct i915_dependency { + struct i915_sched_node *signaler; + struct list_head signal_link; + struct list_head wait_link; + struct list_head dfs_link; + unsigned long flags; +#define I915_DEPENDENCY_ALLOC BIT(0) +}; + +#endif /* _I915_SCHEDULER_TYPES_H_ */ diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c index d2f2a9c2fabd..95f3dab1b229 100644 --- a/drivers/gpu/drm/i915/i915_suspend.c +++ b/drivers/gpu/drm/i915/i915_suspend.c @@ -25,8 +25,10 @@ */ #include <drm/i915_drm.h> -#include "intel_drv.h" + #include "i915_reg.h" +#include "intel_drv.h" +#include "intel_fbc.h" static void i915_save_display(struct drm_i915_private *dev_priv) { diff --git a/drivers/gpu/drm/i915/i915_timeline.c b/drivers/gpu/drm/i915/i915_timeline.c index 2f4907364920..5fbea0892f33 100644 --- a/drivers/gpu/drm/i915/i915_timeline.c +++ b/drivers/gpu/drm/i915/i915_timeline.c @@ -253,7 +253,6 @@ int i915_timeline_init(struct drm_i915_private *i915, spin_lock_init(&timeline->lock); mutex_init(&timeline->mutex); - INIT_ACTIVE_REQUEST(&timeline->barrier); INIT_ACTIVE_REQUEST(&timeline->last_request); INIT_LIST_HEAD(&timeline->requests); @@ -326,7 +325,6 @@ void i915_timeline_fini(struct i915_timeline *timeline) { GEM_BUG_ON(timeline->pin_count); GEM_BUG_ON(!list_empty(&timeline->requests)); - GEM_BUG_ON(i915_active_request_isset(&timeline->barrier)); i915_syncmap_free(&timeline->sync); diff --git a/drivers/gpu/drm/i915/i915_timeline.h b/drivers/gpu/drm/i915/i915_timeline.h index c1e47a423d85..27668a1a69a3 100644 --- a/drivers/gpu/drm/i915/i915_timeline.h +++ b/drivers/gpu/drm/i915/i915_timeline.h @@ -27,6 +27,7 @@ #include <linux/lockdep.h> +#include "i915_active.h" #include "i915_syncmap.h" #include "i915_timeline_types.h" @@ -109,19 +110,4 @@ void i915_timelines_init(struct drm_i915_private *i915); void i915_timelines_park(struct drm_i915_private *i915); void i915_timelines_fini(struct drm_i915_private *i915); -/** - * i915_timeline_set_barrier - orders submission between different timelines - * @timeline: timeline to set the barrier on - * @rq: request after which new submissions can proceed - * - * Sets the passed in request as the serialization point for all subsequent - * submissions on @timeline. Subsequent requests will not be submitted to GPU - * until the barrier has been completed. - */ -static inline int -i915_timeline_set_barrier(struct i915_timeline *tl, struct i915_request *rq) -{ - return i915_active_request_set(&tl->barrier, rq); -} - #endif diff --git a/drivers/gpu/drm/i915/i915_timeline_types.h b/drivers/gpu/drm/i915/i915_timeline_types.h index 12ba3c573aa0..5256a0b5c5f7 100644 --- a/drivers/gpu/drm/i915/i915_timeline_types.h +++ b/drivers/gpu/drm/i915/i915_timeline_types.h @@ -9,9 +9,10 @@ #include <linux/list.h> #include <linux/kref.h> +#include <linux/mutex.h> #include <linux/types.h> -#include "i915_active.h" +#include "i915_active_types.h" struct drm_i915_private; struct i915_vma; @@ -60,16 +61,6 @@ struct i915_timeline { */ struct i915_syncmap *sync; - /** - * Barrier provides the ability to serialize ordering between different - * timelines. - * - * Users can call i915_timeline_set_barrier which will make all - * subsequent submissions to this timeline be executed only after the - * barrier has been completed. - */ - struct i915_active_request barrier; - struct list_head link; struct drm_i915_private *i915; diff --git a/drivers/gpu/drm/i915/icl_dsi.c b/drivers/gpu/drm/i915/icl_dsi.c index b67ffaa283dc..9d962ea1e635 100644 --- a/drivers/gpu/drm/i915/icl_dsi.c +++ b/drivers/gpu/drm/i915/icl_dsi.c @@ -25,9 +25,13 @@ * Jani Nikula <jani.nikula@intel.com> */ -#include <drm/drm_mipi_dsi.h> #include <drm/drm_atomic_helper.h> +#include <drm/drm_mipi_dsi.h> + +#include "intel_connector.h" +#include "intel_ddi.h" #include "intel_dsi.h" +#include "intel_panel.h" static inline int header_credits_available(struct drm_i915_private *dev_priv, enum transcoder dsi_trans) @@ -323,6 +327,21 @@ static void gen11_dsi_program_esc_clk_div(struct intel_encoder *encoder) } } +static void get_dsi_io_power_domains(struct drm_i915_private *dev_priv, + struct intel_dsi *intel_dsi) +{ + enum port port; + + for_each_dsi_port(port, intel_dsi->ports) { + WARN_ON(intel_dsi->io_wakeref[port]); + intel_dsi->io_wakeref[port] = + intel_display_power_get(dev_priv, + port == PORT_A ? + POWER_DOMAIN_PORT_DDI_A_IO : + POWER_DOMAIN_PORT_DDI_B_IO); + } +} + static void gen11_dsi_enable_io_power(struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); @@ -336,13 +355,7 @@ static void gen11_dsi_enable_io_power(struct intel_encoder *encoder) I915_WRITE(ICL_DSI_IO_MODECTL(port), tmp); } - for_each_dsi_port(port, intel_dsi->ports) { - intel_dsi->io_wakeref[port] = - intel_display_power_get(dev_priv, - port == PORT_A ? - POWER_DOMAIN_PORT_DDI_A_IO : - POWER_DOMAIN_PORT_DDI_B_IO); - } + get_dsi_io_power_domains(dev_priv, intel_dsi); } static void gen11_dsi_power_up_lanes(struct intel_encoder *encoder) @@ -589,6 +602,12 @@ static void gen11_dsi_map_pll(struct intel_encoder *encoder, val |= DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, port); } I915_WRITE(DPCLKA_CFGCR0_ICL, val); + + for_each_dsi_port(port, intel_dsi->ports) { + val &= ~DPCLKA_CFGCR0_DDI_CLK_OFF(port); + } + I915_WRITE(DPCLKA_CFGCR0_ICL, val); + POSTING_READ(DPCLKA_CFGCR0_ICL); mutex_unlock(&dev_priv->dpll_lock); @@ -1119,7 +1138,7 @@ static void gen11_dsi_disable_port(struct intel_encoder *encoder) DRM_ERROR("DDI port:%c buffer not idle\n", port_name(port)); } - gen11_dsi_ungate_clocks(encoder); + gen11_dsi_gate_clocks(encoder); } static void gen11_dsi_disable_io_power(struct intel_encoder *encoder) @@ -1133,13 +1152,11 @@ static void gen11_dsi_disable_io_power(struct intel_encoder *encoder) intel_wakeref_t wakeref; wakeref = fetch_and_zero(&intel_dsi->io_wakeref[port]); - if (wakeref) { - intel_display_power_put(dev_priv, - port == PORT_A ? - POWER_DOMAIN_PORT_DDI_A_IO : - POWER_DOMAIN_PORT_DDI_B_IO, - wakeref); - } + intel_display_power_put(dev_priv, + port == PORT_A ? + POWER_DOMAIN_PORT_DDI_A_IO : + POWER_DOMAIN_PORT_DDI_B_IO, + wakeref); } /* set mode to DDI */ @@ -1219,20 +1236,11 @@ static int gen11_dsi_compute_config(struct intel_encoder *encoder, return 0; } -static u64 gen11_dsi_get_power_domains(struct intel_encoder *encoder, - struct intel_crtc_state *crtc_state) +static void gen11_dsi_get_power_domains(struct intel_encoder *encoder, + struct intel_crtc_state *crtc_state) { - struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); - u64 domains = 0; - enum port port; - - for_each_dsi_port(port, intel_dsi->ports) - if (port == PORT_A) - domains |= BIT_ULL(POWER_DOMAIN_PORT_DDI_A_IO); - else - domains |= BIT_ULL(POWER_DOMAIN_PORT_DDI_B_IO); - - return domains; + get_dsi_io_power_domains(to_i915(encoder->base.dev), + enc_to_intel_dsi(&encoder->base)); } static bool gen11_dsi_get_hw_state(struct intel_encoder *encoder, diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c index b844e8840c6f..8c8fae32ec50 100644 --- a/drivers/gpu/drm/i915/intel_atomic.c +++ b/drivers/gpu/drm/i915/intel_atomic.c @@ -35,6 +35,8 @@ #include <drm/drm_plane_helper.h> #include "intel_drv.h" +#include "intel_hdcp.h" +#include "intel_sprite.h" /** * intel_digital_connector_atomic_get_property - hook for connector->atomic_get_property. diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c index 9d32a6fcf840..d11681d71add 100644 --- a/drivers/gpu/drm/i915/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/intel_atomic_plane.c @@ -35,7 +35,10 @@ #include <drm/drm_fourcc.h> #include <drm/drm_plane_helper.h> +#include "intel_atomic_plane.h" #include "intel_drv.h" +#include "intel_pm.h" +#include "intel_sprite.h" struct intel_plane *intel_plane_alloc(void) { diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.h b/drivers/gpu/drm/i915/intel_atomic_plane.h new file mode 100644 index 000000000000..14678620440f --- /dev/null +++ b/drivers/gpu/drm/i915/intel_atomic_plane.h @@ -0,0 +1,40 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_ATOMIC_PLANE_H__ +#define __INTEL_ATOMIC_PLANE_H__ + +struct drm_plane; +struct intel_atomic_state; +struct intel_crtc; +struct intel_crtc_state; +struct intel_plane; +struct intel_plane_state; + +extern const struct drm_plane_helper_funcs intel_plane_helper_funcs; + +void intel_update_plane(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state); +void intel_update_slave(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state); +void intel_disable_plane(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state); +struct intel_plane *intel_plane_alloc(void); +void intel_plane_free(struct intel_plane *plane); +struct drm_plane_state *intel_plane_duplicate_state(struct drm_plane *plane); +void intel_plane_destroy_state(struct drm_plane *plane, + struct drm_plane_state *state); +void skl_update_planes_on_crtc(struct intel_atomic_state *state, + struct intel_crtc *crtc); +void i9xx_update_planes_on_crtc(struct intel_atomic_state *state, + struct intel_crtc *crtc); +int intel_plane_atomic_check_with_state(const struct intel_crtc_state *old_crtc_state, + struct intel_crtc_state *crtc_state, + const struct intel_plane_state *old_plane_state, + struct intel_plane_state *intel_state); + +#endif /* __INTEL_ATOMIC_PLANE_H__ */ diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c index 502b57ce72ab..bca4cc025d3d 100644 --- a/drivers/gpu/drm/i915/intel_audio.c +++ b/drivers/gpu/drm/i915/intel_audio.c @@ -21,14 +21,16 @@ * DEALINGS IN THE SOFTWARE. */ -#include <linux/kernel.h> #include <linux/component.h> +#include <linux/kernel.h> + +#include <drm/drm_edid.h> #include <drm/i915_component.h> #include <drm/intel_lpe_audio.h> -#include "intel_drv.h" -#include <drm/drm_edid.h> #include "i915_drv.h" +#include "intel_audio.h" +#include "intel_drv.h" /** * DOC: High Definition Audio over HDMI and Display Port @@ -741,18 +743,78 @@ void intel_init_audio_hooks(struct drm_i915_private *dev_priv) } } +static void glk_force_audio_cdclk(struct drm_i915_private *dev_priv, + bool enable) +{ + struct drm_modeset_acquire_ctx ctx; + struct drm_atomic_state *state; + int ret; + + drm_modeset_acquire_init(&ctx, 0); + state = drm_atomic_state_alloc(&dev_priv->drm); + if (WARN_ON(!state)) + return; + + state->acquire_ctx = &ctx; + +retry: + to_intel_atomic_state(state)->cdclk.force_min_cdclk_changed = true; + to_intel_atomic_state(state)->cdclk.force_min_cdclk = + enable ? 2 * 96000 : 0; + + /* + * Protects dev_priv->cdclk.force_min_cdclk + * Need to lock this here in case we have no active pipes + * and thus wouldn't lock it during the commit otherwise. + */ + ret = drm_modeset_lock(&dev_priv->drm.mode_config.connection_mutex, + &ctx); + if (!ret) + ret = drm_atomic_commit(state); + + if (ret == -EDEADLK) { + drm_atomic_state_clear(state); + drm_modeset_backoff(&ctx); + goto retry; + } + + WARN_ON(ret); + + drm_atomic_state_put(state); + + drm_modeset_drop_locks(&ctx); + drm_modeset_acquire_fini(&ctx); +} + static unsigned long i915_audio_component_get_power(struct device *kdev) { + struct drm_i915_private *dev_priv = kdev_to_i915(kdev); + intel_wakeref_t ret; + /* Catch potential impedance mismatches before they occur! */ BUILD_BUG_ON(sizeof(intel_wakeref_t) > sizeof(unsigned long)); - return intel_display_power_get(kdev_to_i915(kdev), POWER_DOMAIN_AUDIO); + ret = intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO); + + /* Force CDCLK to 2*BCLK as long as we need audio to be powered. */ + if (dev_priv->audio_power_refcount++ == 0) + if (IS_CANNONLAKE(dev_priv) || IS_GEMINILAKE(dev_priv)) + glk_force_audio_cdclk(dev_priv, true); + + return ret; } static void i915_audio_component_put_power(struct device *kdev, unsigned long cookie) { - intel_display_power_put(kdev_to_i915(kdev), POWER_DOMAIN_AUDIO, cookie); + struct drm_i915_private *dev_priv = kdev_to_i915(kdev); + + /* Stop forcing CDCLK to 2*BCLK if no need for audio to be powered. */ + if (--dev_priv->audio_power_refcount == 0) + if (IS_CANNONLAKE(dev_priv) || IS_GEMINILAKE(dev_priv)) + glk_force_audio_cdclk(dev_priv, false); + + intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO, cookie); } static void i915_audio_component_codec_wake_override(struct device *kdev, @@ -985,7 +1047,7 @@ static const struct component_ops i915_audio_component_bind_ops = { * We ignore any error during registration and continue with reduced * functionality (i.e. without HDMI audio). */ -void i915_audio_component_init(struct drm_i915_private *dev_priv) +static void i915_audio_component_init(struct drm_i915_private *dev_priv) { int ret; @@ -1008,7 +1070,7 @@ void i915_audio_component_init(struct drm_i915_private *dev_priv) * Deregisters the audio component, breaking any existing binding to the * corresponding snd_hda_intel driver's master component. */ -void i915_audio_component_cleanup(struct drm_i915_private *dev_priv) +static void i915_audio_component_cleanup(struct drm_i915_private *dev_priv) { if (!dev_priv->audio_component_registered) return; diff --git a/drivers/gpu/drm/i915/intel_audio.h b/drivers/gpu/drm/i915/intel_audio.h new file mode 100644 index 000000000000..a3657c7a7ba2 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_audio.h @@ -0,0 +1,24 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_AUDIO_H__ +#define __INTEL_AUDIO_H__ + +struct drm_connector_state; +struct drm_i915_private; +struct intel_crtc_state; +struct intel_encoder; + +void intel_init_audio_hooks(struct drm_i915_private *dev_priv); +void intel_audio_codec_enable(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + const struct drm_connector_state *conn_state); +void intel_audio_codec_disable(struct intel_encoder *encoder, + const struct intel_crtc_state *old_crtc_state, + const struct drm_connector_state *old_conn_state); +void intel_audio_init(struct drm_i915_private *dev_priv); +void intel_audio_deinit(struct drm_i915_private *dev_priv); + +#endif /* __INTEL_AUDIO_H__ */ diff --git a/drivers/gpu/drm/i915/intel_breadcrumbs.c b/drivers/gpu/drm/i915/intel_breadcrumbs.c index 09ed90c0ba00..3cbffd400b1b 100644 --- a/drivers/gpu/drm/i915/intel_breadcrumbs.c +++ b/drivers/gpu/drm/i915/intel_breadcrumbs.c @@ -27,8 +27,6 @@ #include "i915_drv.h" -#define task_asleep(tsk) ((tsk)->state & TASK_NORMAL && !(tsk)->on_rq) - static void irq_enable(struct intel_engine_cs *engine) { if (!engine->irq_enable) @@ -82,7 +80,7 @@ static inline bool __request_completed(const struct i915_request *rq) return i915_seqno_passed(__hwsp_seqno(rq), rq->fence.seqno); } -bool intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine) +void intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine) { struct intel_breadcrumbs *b = &engine->breadcrumbs; struct intel_context *ce, *cn; @@ -146,19 +144,13 @@ bool intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine) dma_fence_signal(&rq->fence); i915_request_put(rq); } - - return !list_empty(&signal); } -bool intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine) +void intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine) { - bool result; - local_irq_disable(); - result = intel_engine_breadcrumbs_irq(engine); + intel_engine_breadcrumbs_irq(engine); local_irq_enable(); - - return result; } static void signal_irq_work(struct irq_work *work) diff --git a/drivers/gpu/drm/i915/intel_cdclk.c b/drivers/gpu/drm/i915/intel_cdclk.c index d40f8793718c..ae40a8679314 100644 --- a/drivers/gpu/drm/i915/intel_cdclk.c +++ b/drivers/gpu/drm/i915/intel_cdclk.c @@ -21,6 +21,7 @@ * DEALINGS IN THE SOFTWARE. */ +#include "intel_cdclk.h" #include "intel_drv.h" /** @@ -517,7 +518,8 @@ static void vlv_program_pfi_credits(struct drm_i915_private *dev_priv) } static void vlv_set_cdclk(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state) + const struct intel_cdclk_state *cdclk_state, + enum pipe pipe) { int cdclk = cdclk_state->cdclk; u32 val, cmd = cdclk_state->voltage_level; @@ -599,7 +601,8 @@ static void vlv_set_cdclk(struct drm_i915_private *dev_priv, } static void chv_set_cdclk(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state) + const struct intel_cdclk_state *cdclk_state, + enum pipe pipe) { int cdclk = cdclk_state->cdclk; u32 val, cmd = cdclk_state->voltage_level; @@ -698,7 +701,8 @@ static void bdw_get_cdclk(struct drm_i915_private *dev_priv, } static void bdw_set_cdclk(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state) + const struct intel_cdclk_state *cdclk_state, + enum pipe pipe) { int cdclk = cdclk_state->cdclk; u32 val; @@ -988,7 +992,8 @@ static void skl_dpll0_disable(struct drm_i915_private *dev_priv) } static void skl_set_cdclk(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state) + const struct intel_cdclk_state *cdclk_state, + enum pipe pipe) { int cdclk = cdclk_state->cdclk; int vco = cdclk_state->vco; @@ -1124,16 +1129,7 @@ sanitize: dev_priv->cdclk.hw.vco = -1; } -/** - * skl_init_cdclk - Initialize CDCLK on SKL - * @dev_priv: i915 device - * - * Initialize CDCLK for SKL and derivatives. This is generally - * done only during the display core initialization sequence, - * after which the DMC will take care of turning CDCLK off/on - * as needed. - */ -void skl_init_cdclk(struct drm_i915_private *dev_priv) +static void skl_init_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state; @@ -1159,17 +1155,10 @@ void skl_init_cdclk(struct drm_i915_private *dev_priv) cdclk_state.cdclk = skl_calc_cdclk(0, cdclk_state.vco); cdclk_state.voltage_level = skl_calc_voltage_level(cdclk_state.cdclk); - skl_set_cdclk(dev_priv, &cdclk_state); + skl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE); } -/** - * skl_uninit_cdclk - Uninitialize CDCLK on SKL - * @dev_priv: i915 device - * - * Uninitialize CDCLK for SKL and derivatives. This is done only - * during the display core uninitialization sequence. - */ -void skl_uninit_cdclk(struct drm_i915_private *dev_priv) +static void skl_uninit_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw; @@ -1177,7 +1166,7 @@ void skl_uninit_cdclk(struct drm_i915_private *dev_priv) cdclk_state.vco = 0; cdclk_state.voltage_level = skl_calc_voltage_level(cdclk_state.cdclk); - skl_set_cdclk(dev_priv, &cdclk_state); + skl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE); } static int bxt_calc_cdclk(int min_cdclk) @@ -1356,7 +1345,8 @@ static void bxt_de_pll_enable(struct drm_i915_private *dev_priv, int vco) } static void bxt_set_cdclk(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state) + const struct intel_cdclk_state *cdclk_state, + enum pipe pipe) { int cdclk = cdclk_state->cdclk; int vco = cdclk_state->vco; @@ -1409,11 +1399,10 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, bxt_de_pll_enable(dev_priv, vco); val = divider | skl_cdclk_decimal(cdclk); - /* - * FIXME if only the cd2x divider needs changing, it could be done - * without shutting off the pipe (if only one pipe is active). - */ - val |= BXT_CDCLK_CD2X_PIPE_NONE; + if (pipe == INVALID_PIPE) + val |= BXT_CDCLK_CD2X_PIPE_NONE; + else + val |= BXT_CDCLK_CD2X_PIPE(pipe); /* * Disable SSA Precharge when CD clock frequency < 500 MHz, * enable otherwise. @@ -1422,6 +1411,9 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, val |= BXT_CDCLK_SSA_PRECHARGE_ENABLE; I915_WRITE(CDCLK_CTL, val); + if (pipe != INVALID_PIPE) + intel_wait_for_vblank(dev_priv, pipe); + mutex_lock(&dev_priv->pcu_lock); /* * The timeout isn't specified, the 2ms used here is based on @@ -1491,16 +1483,7 @@ sanitize: dev_priv->cdclk.hw.vco = -1; } -/** - * bxt_init_cdclk - Initialize CDCLK on BXT - * @dev_priv: i915 device - * - * Initialize CDCLK for BXT and derivatives. This is generally - * done only during the display core initialization sequence, - * after which the DMC will take care of turning CDCLK off/on - * as needed. - */ -void bxt_init_cdclk(struct drm_i915_private *dev_priv) +static void bxt_init_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state; @@ -1526,17 +1509,10 @@ void bxt_init_cdclk(struct drm_i915_private *dev_priv) } cdclk_state.voltage_level = bxt_calc_voltage_level(cdclk_state.cdclk); - bxt_set_cdclk(dev_priv, &cdclk_state); + bxt_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE); } -/** - * bxt_uninit_cdclk - Uninitialize CDCLK on BXT - * @dev_priv: i915 device - * - * Uninitialize CDCLK for BXT and derivatives. This is done only - * during the display core uninitialization sequence. - */ -void bxt_uninit_cdclk(struct drm_i915_private *dev_priv) +static void bxt_uninit_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw; @@ -1544,7 +1520,7 @@ void bxt_uninit_cdclk(struct drm_i915_private *dev_priv) cdclk_state.vco = 0; cdclk_state.voltage_level = bxt_calc_voltage_level(cdclk_state.cdclk); - bxt_set_cdclk(dev_priv, &cdclk_state); + bxt_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE); } static int cnl_calc_cdclk(int min_cdclk) @@ -1664,7 +1640,8 @@ static void cnl_cdclk_pll_enable(struct drm_i915_private *dev_priv, int vco) } static void cnl_set_cdclk(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state) + const struct intel_cdclk_state *cdclk_state, + enum pipe pipe) { int cdclk = cdclk_state->cdclk; int vco = cdclk_state->vco; @@ -1705,13 +1682,15 @@ static void cnl_set_cdclk(struct drm_i915_private *dev_priv, cnl_cdclk_pll_enable(dev_priv, vco); val = divider | skl_cdclk_decimal(cdclk); - /* - * FIXME if only the cd2x divider needs changing, it could be done - * without shutting off the pipe (if only one pipe is active). - */ - val |= BXT_CDCLK_CD2X_PIPE_NONE; + if (pipe == INVALID_PIPE) + val |= BXT_CDCLK_CD2X_PIPE_NONE; + else + val |= BXT_CDCLK_CD2X_PIPE(pipe); I915_WRITE(CDCLK_CTL, val); + if (pipe != INVALID_PIPE) + intel_wait_for_vblank(dev_priv, pipe); + /* inform PCU of the change */ mutex_lock(&dev_priv->pcu_lock); sandybridge_pcode_write(dev_priv, SKL_PCODE_CDCLK_CONTROL, @@ -1848,7 +1827,8 @@ static int icl_calc_cdclk_pll_vco(struct drm_i915_private *dev_priv, int cdclk) } static void icl_set_cdclk(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state) + const struct intel_cdclk_state *cdclk_state, + enum pipe pipe) { unsigned int cdclk = cdclk_state->cdclk; unsigned int vco = cdclk_state->vco; @@ -1873,6 +1853,11 @@ static void icl_set_cdclk(struct drm_i915_private *dev_priv, if (dev_priv->cdclk.hw.vco != vco) cnl_cdclk_pll_enable(dev_priv, vco); + /* + * On ICL CD2X_DIV can only be 1, so we'll never end up changing the + * divider here synchronized to a pipe while CDCLK is on, nor will we + * need the corresponding vblank wait. + */ I915_WRITE(CDCLK_CTL, ICL_CDCLK_CD2X_PIPE_NONE | skl_cdclk_decimal(cdclk)); @@ -1960,16 +1945,7 @@ out: icl_calc_voltage_level(cdclk_state->cdclk); } -/** - * icl_init_cdclk - Initialize CDCLK on ICL - * @dev_priv: i915 device - * - * Initialize CDCLK for ICL. This consists mainly of initializing - * dev_priv->cdclk.hw and sanitizing the state of the hardware if needed. This - * is generally done only during the display core initialization sequence, after - * which the DMC will take care of turning CDCLK off/on as needed. - */ -void icl_init_cdclk(struct drm_i915_private *dev_priv) +static void icl_init_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state sanitized_state; u32 val; @@ -2003,17 +1979,10 @@ sanitize: sanitized_state.voltage_level = icl_calc_voltage_level(sanitized_state.cdclk); - icl_set_cdclk(dev_priv, &sanitized_state); + icl_set_cdclk(dev_priv, &sanitized_state, INVALID_PIPE); } -/** - * icl_uninit_cdclk - Uninitialize CDCLK on ICL - * @dev_priv: i915 device - * - * Uninitialize CDCLK for ICL. This is done only during the display core - * uninitialization sequence. - */ -void icl_uninit_cdclk(struct drm_i915_private *dev_priv) +static void icl_uninit_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw; @@ -2021,19 +1990,10 @@ void icl_uninit_cdclk(struct drm_i915_private *dev_priv) cdclk_state.vco = 0; cdclk_state.voltage_level = icl_calc_voltage_level(cdclk_state.cdclk); - icl_set_cdclk(dev_priv, &cdclk_state); + icl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE); } -/** - * cnl_init_cdclk - Initialize CDCLK on CNL - * @dev_priv: i915 device - * - * Initialize CDCLK for CNL. This is generally - * done only during the display core initialization sequence, - * after which the DMC will take care of turning CDCLK off/on - * as needed. - */ -void cnl_init_cdclk(struct drm_i915_private *dev_priv) +static void cnl_init_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state; @@ -2049,17 +2009,10 @@ void cnl_init_cdclk(struct drm_i915_private *dev_priv) cdclk_state.vco = cnl_cdclk_pll_vco(dev_priv, cdclk_state.cdclk); cdclk_state.voltage_level = cnl_calc_voltage_level(cdclk_state.cdclk); - cnl_set_cdclk(dev_priv, &cdclk_state); + cnl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE); } -/** - * cnl_uninit_cdclk - Uninitialize CDCLK on CNL - * @dev_priv: i915 device - * - * Uninitialize CDCLK for CNL. This is done only - * during the display core uninitialization sequence. - */ -void cnl_uninit_cdclk(struct drm_i915_private *dev_priv) +static void cnl_uninit_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw; @@ -2067,7 +2020,47 @@ void cnl_uninit_cdclk(struct drm_i915_private *dev_priv) cdclk_state.vco = 0; cdclk_state.voltage_level = cnl_calc_voltage_level(cdclk_state.cdclk); - cnl_set_cdclk(dev_priv, &cdclk_state); + cnl_set_cdclk(dev_priv, &cdclk_state, INVALID_PIPE); +} + +/** + * intel_cdclk_init - Initialize CDCLK + * @i915: i915 device + * + * Initialize CDCLK. This consists mainly of initializing dev_priv->cdclk.hw and + * sanitizing the state of the hardware if needed. This is generally done only + * during the display core initialization sequence, after which the DMC will + * take care of turning CDCLK off/on as needed. + */ +void intel_cdclk_init(struct drm_i915_private *i915) +{ + if (INTEL_GEN(i915) >= 11) + icl_init_cdclk(i915); + else if (IS_CANNONLAKE(i915)) + cnl_init_cdclk(i915); + else if (IS_GEN9_BC(i915)) + skl_init_cdclk(i915); + else if (IS_GEN9_LP(i915)) + bxt_init_cdclk(i915); +} + +/** + * intel_cdclk_uninit - Uninitialize CDCLK + * @i915: i915 device + * + * Uninitialize CDCLK. This is done only during the display core + * uninitialization sequence. + */ +void intel_cdclk_uninit(struct drm_i915_private *i915) +{ + if (INTEL_GEN(i915) >= 11) + icl_uninit_cdclk(i915); + else if (IS_CANNONLAKE(i915)) + cnl_uninit_cdclk(i915); + else if (IS_GEN9_BC(i915)) + skl_uninit_cdclk(i915); + else if (IS_GEN9_LP(i915)) + bxt_uninit_cdclk(i915); } /** @@ -2087,6 +2080,28 @@ bool intel_cdclk_needs_modeset(const struct intel_cdclk_state *a, } /** + * intel_cdclk_needs_cd2x_update - Determine if two CDCLK states require a cd2x divider update + * @dev_priv: Not a CDCLK state, it's the drm_i915_private! + * @a: first CDCLK state + * @b: second CDCLK state + * + * Returns: + * True if the CDCLK states require just a cd2x divider update, false if not. + */ +bool intel_cdclk_needs_cd2x_update(struct drm_i915_private *dev_priv, + const struct intel_cdclk_state *a, + const struct intel_cdclk_state *b) +{ + /* Older hw doesn't have the capability */ + if (INTEL_GEN(dev_priv) < 10 && !IS_GEN9_LP(dev_priv)) + return false; + + return a->cdclk != b->cdclk && + a->vco == b->vco && + a->ref == b->ref; +} + +/** * intel_cdclk_changed - Determine if two CDCLK states are different * @a: first CDCLK state * @b: second CDCLK state @@ -2101,6 +2116,26 @@ bool intel_cdclk_changed(const struct intel_cdclk_state *a, a->voltage_level != b->voltage_level; } +/** + * intel_cdclk_swap_state - make atomic CDCLK configuration effective + * @state: atomic state + * + * This is the CDCLK version of drm_atomic_helper_swap_state() since the + * helper does not handle driver-specific global state. + * + * Similarly to the atomic helpers this function does a complete swap, + * i.e. it also puts the old state into @state. This is used by the commit + * code to determine how CDCLK has changed (for instance did it increase or + * decrease). + */ +void intel_cdclk_swap_state(struct intel_atomic_state *state) +{ + struct drm_i915_private *dev_priv = to_i915(state->base.dev); + + swap(state->cdclk.logical, dev_priv->cdclk.logical); + swap(state->cdclk.actual, dev_priv->cdclk.actual); +} + void intel_dump_cdclk_state(const struct intel_cdclk_state *cdclk_state, const char *context) { @@ -2114,12 +2149,14 @@ void intel_dump_cdclk_state(const struct intel_cdclk_state *cdclk_state, * intel_set_cdclk - Push the CDCLK state to the hardware * @dev_priv: i915 device * @cdclk_state: new CDCLK state + * @pipe: pipe with which to synchronize the update * * Program the hardware based on the passed in CDCLK state, * if necessary. */ -void intel_set_cdclk(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state) +static void intel_set_cdclk(struct drm_i915_private *dev_priv, + const struct intel_cdclk_state *cdclk_state, + enum pipe pipe) { if (!intel_cdclk_changed(&dev_priv->cdclk.hw, cdclk_state)) return; @@ -2129,7 +2166,7 @@ void intel_set_cdclk(struct drm_i915_private *dev_priv, intel_dump_cdclk_state(cdclk_state, "Changing CDCLK to"); - dev_priv->display.set_cdclk(dev_priv, cdclk_state); + dev_priv->display.set_cdclk(dev_priv, cdclk_state, pipe); if (WARN(intel_cdclk_changed(&dev_priv->cdclk.hw, cdclk_state), "cdclk state doesn't match!\n")) { @@ -2138,6 +2175,46 @@ void intel_set_cdclk(struct drm_i915_private *dev_priv, } } +/** + * intel_set_cdclk_pre_plane_update - Push the CDCLK state to the hardware + * @dev_priv: i915 device + * @old_state: old CDCLK state + * @new_state: new CDCLK state + * @pipe: pipe with which to synchronize the update + * + * Program the hardware before updating the HW plane state based on the passed + * in CDCLK state, if necessary. + */ +void +intel_set_cdclk_pre_plane_update(struct drm_i915_private *dev_priv, + const struct intel_cdclk_state *old_state, + const struct intel_cdclk_state *new_state, + enum pipe pipe) +{ + if (pipe == INVALID_PIPE || old_state->cdclk <= new_state->cdclk) + intel_set_cdclk(dev_priv, new_state, pipe); +} + +/** + * intel_set_cdclk_post_plane_update - Push the CDCLK state to the hardware + * @dev_priv: i915 device + * @old_state: old CDCLK state + * @new_state: new CDCLK state + * @pipe: pipe with which to synchronize the update + * + * Program the hardware after updating the HW plane state based on the passed + * in CDCLK state, if necessary. + */ +void +intel_set_cdclk_post_plane_update(struct drm_i915_private *dev_priv, + const struct intel_cdclk_state *old_state, + const struct intel_cdclk_state *new_state, + enum pipe pipe) +{ + if (pipe != INVALID_PIPE && old_state->cdclk > new_state->cdclk) + intel_set_cdclk(dev_priv, new_state, pipe); +} + static int intel_pixel_rate_to_cdclk(struct drm_i915_private *dev_priv, int pixel_rate) { @@ -2188,19 +2265,8 @@ int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_state) /* * According to BSpec, "The CD clock frequency must be at least twice * the frequency of the Azalia BCLK." and BCLK is 96 MHz by default. - * - * FIXME: Check the actual, not default, BCLK being used. - * - * FIXME: This does not depend on ->has_audio because the higher CDCLK - * is required for audio probe, also when there are no audio capable - * displays connected at probe time. This leads to unnecessarily high - * CDCLK when audio is not required. - * - * FIXME: This limit is only applied when there are displays connected - * at probe time. If we probe without displays, we'll still end up using - * the platform minimum CDCLK, failing audio probe. */ - if (INTEL_GEN(dev_priv) >= 9) + if (crtc_state->has_audio && INTEL_GEN(dev_priv) >= 9) min_cdclk = max(2 * 96000, min_cdclk); /* @@ -2240,7 +2306,7 @@ static int intel_compute_min_cdclk(struct drm_atomic_state *state) intel_state->min_cdclk[i] = min_cdclk; } - min_cdclk = 0; + min_cdclk = intel_state->cdclk.force_min_cdclk; for_each_pipe(dev_priv, pipe) min_cdclk = max(intel_state->min_cdclk[pipe], min_cdclk); @@ -2301,7 +2367,8 @@ static int vlv_modeset_calc_cdclk(struct drm_atomic_state *state) vlv_calc_voltage_level(dev_priv, cdclk); if (!intel_state->active_crtcs) { - cdclk = vlv_calc_cdclk(dev_priv, 0); + cdclk = vlv_calc_cdclk(dev_priv, + intel_state->cdclk.force_min_cdclk); intel_state->cdclk.actual.cdclk = cdclk; intel_state->cdclk.actual.voltage_level = @@ -2334,7 +2401,7 @@ static int bdw_modeset_calc_cdclk(struct drm_atomic_state *state) bdw_calc_voltage_level(cdclk); if (!intel_state->active_crtcs) { - cdclk = bdw_calc_cdclk(0); + cdclk = bdw_calc_cdclk(intel_state->cdclk.force_min_cdclk); intel_state->cdclk.actual.cdclk = cdclk; intel_state->cdclk.actual.voltage_level = @@ -2406,7 +2473,7 @@ static int skl_modeset_calc_cdclk(struct drm_atomic_state *state) skl_calc_voltage_level(cdclk); if (!intel_state->active_crtcs) { - cdclk = skl_calc_cdclk(0, vco); + cdclk = skl_calc_cdclk(intel_state->cdclk.force_min_cdclk, vco); intel_state->cdclk.actual.vco = vco; intel_state->cdclk.actual.cdclk = cdclk; @@ -2445,10 +2512,10 @@ static int bxt_modeset_calc_cdclk(struct drm_atomic_state *state) if (!intel_state->active_crtcs) { if (IS_GEMINILAKE(dev_priv)) { - cdclk = glk_calc_cdclk(0); + cdclk = glk_calc_cdclk(intel_state->cdclk.force_min_cdclk); vco = glk_de_pll_vco(dev_priv, cdclk); } else { - cdclk = bxt_calc_cdclk(0); + cdclk = bxt_calc_cdclk(intel_state->cdclk.force_min_cdclk); vco = bxt_de_pll_vco(dev_priv, cdclk); } @@ -2484,7 +2551,7 @@ static int cnl_modeset_calc_cdclk(struct drm_atomic_state *state) cnl_compute_min_voltage_level(intel_state)); if (!intel_state->active_crtcs) { - cdclk = cnl_calc_cdclk(0); + cdclk = cnl_calc_cdclk(intel_state->cdclk.force_min_cdclk); vco = cnl_cdclk_pll_vco(dev_priv, cdclk); intel_state->cdclk.actual.vco = vco; @@ -2520,7 +2587,7 @@ static int icl_modeset_calc_cdclk(struct drm_atomic_state *state) cnl_compute_min_voltage_level(intel_state)); if (!intel_state->active_crtcs) { - cdclk = icl_calc_cdclk(0, ref); + cdclk = icl_calc_cdclk(intel_state->cdclk.force_min_cdclk, ref); vco = icl_calc_cdclk_pll_vco(dev_priv, cdclk); intel_state->cdclk.actual.vco = vco; diff --git a/drivers/gpu/drm/i915/intel_cdclk.h b/drivers/gpu/drm/i915/intel_cdclk.h new file mode 100644 index 000000000000..4d6f7f5f8930 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_cdclk.h @@ -0,0 +1,46 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_CDCLK_H__ +#define __INTEL_CDCLK_H__ + +#include <linux/types.h> + +#include "intel_display.h" + +struct drm_i915_private; +struct intel_atomic_state; +struct intel_cdclk_state; +struct intel_crtc_state; + +int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_state); +void intel_cdclk_init(struct drm_i915_private *i915); +void intel_cdclk_uninit(struct drm_i915_private *i915); +void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv); +void intel_update_max_cdclk(struct drm_i915_private *dev_priv); +void intel_update_cdclk(struct drm_i915_private *dev_priv); +void intel_update_rawclk(struct drm_i915_private *dev_priv); +bool intel_cdclk_needs_cd2x_update(struct drm_i915_private *dev_priv, + const struct intel_cdclk_state *a, + const struct intel_cdclk_state *b); +bool intel_cdclk_needs_modeset(const struct intel_cdclk_state *a, + const struct intel_cdclk_state *b); +bool intel_cdclk_changed(const struct intel_cdclk_state *a, + const struct intel_cdclk_state *b); +void intel_cdclk_swap_state(struct intel_atomic_state *state); +void +intel_set_cdclk_pre_plane_update(struct drm_i915_private *dev_priv, + const struct intel_cdclk_state *old_state, + const struct intel_cdclk_state *new_state, + enum pipe pipe); +void +intel_set_cdclk_post_plane_update(struct drm_i915_private *dev_priv, + const struct intel_cdclk_state *old_state, + const struct intel_cdclk_state *new_state, + enum pipe pipe); +void intel_dump_cdclk_state(const struct intel_cdclk_state *cdclk_state, + const char *context); + +#endif /* __INTEL_CDCLK_H__ */ diff --git a/drivers/gpu/drm/i915/intel_color.c b/drivers/gpu/drm/i915/intel_color.c index 467fd1a1630c..ca341a9e47e6 100644 --- a/drivers/gpu/drm/i915/intel_color.c +++ b/drivers/gpu/drm/i915/intel_color.c @@ -22,6 +22,7 @@ * */ +#include "intel_color.h" #include "intel_drv.h" #define CTM_COEFF_SIGN (1ULL << 63) @@ -273,6 +274,14 @@ static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state) ilk_csc_coeff_limited_range, ilk_csc_postoff_limited_range); } else if (crtc_state->csc_enable) { + /* + * On GLK+ both pipe CSC and degamma LUT are controlled + * by csc_enable. Hence for the cases where the degama + * LUT is needed but CSC is not we need to load an + * identity matrix. + */ + WARN_ON(!IS_CANNONLAKE(dev_priv) && !IS_GEMINILAKE(dev_priv)); + ilk_update_pipe_csc(crtc, ilk_csc_off_zero, ilk_csc_coeff_identity, ilk_csc_off_zero); @@ -351,6 +360,29 @@ static void cherryview_load_csc_matrix(const struct intel_crtc_state *crtc_state I915_WRITE(CGM_PIPE_MODE(pipe), crtc_state->cgm_mode); } +/* i965+ "10.6" bit interpolated format "even DW" (low 8 bits) */ +static u32 i965_lut_10p6_ldw(const struct drm_color_lut *color) +{ + return (color->red & 0xff) << 16 | + (color->green & 0xff) << 8 | + (color->blue & 0xff); +} + +/* i965+ "10.6" interpolated format "odd DW" (high 8 bits) */ +static u32 i965_lut_10p6_udw(const struct drm_color_lut *color) +{ + return (color->red >> 8) << 16 | + (color->green >> 8) << 8 | + (color->blue >> 8); +} + +static u32 ilk_lut_10(const struct drm_color_lut *color) +{ + return drm_color_lut_extract(color->red, 10) << 20 | + drm_color_lut_extract(color->green, 10) << 10 | + drm_color_lut_extract(color->blue, 10); +} + /* Loads the legacy palette/gamma unit for the CRTC. */ static void i9xx_load_luts_internal(const struct intel_crtc_state *crtc_state, const struct drm_property_blob *blob) @@ -381,15 +413,6 @@ static void i9xx_load_luts_internal(const struct intel_crtc_state *crtc_state, else I915_WRITE(LGC_PALETTE(pipe, i), word); } - } else { - for (i = 0; i < 256; i++) { - u32 word = (i << 16) | (i << 8) | i; - - if (HAS_GMCH(dev_priv)) - I915_WRITE(PALETTE(pipe, i), word); - else - I915_WRITE(LGC_PALETTE(pipe, i), word); - } } } @@ -422,6 +445,8 @@ static void ilk_color_commit(const struct intel_crtc_state *crtc_state) val &= ~PIPECONF_GAMMA_MODE_MASK_ILK; val |= PIPECONF_GAMMA_MODE(crtc_state->gamma_mode); I915_WRITE(PIPECONF(pipe), val); + + ilk_load_csc_matrix(crtc_state); } static void hsw_color_commit(const struct intel_crtc_state *crtc_state) @@ -460,84 +485,119 @@ static void skl_color_commit(const struct intel_crtc_state *crtc_state) ilk_load_csc_matrix(crtc_state); } -static void bdw_load_degamma_lut(const struct intel_crtc_state *crtc_state) +static void i965_load_lut_10p6(struct intel_crtc *crtc, + const struct drm_property_blob *blob) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - const struct drm_property_blob *degamma_lut = crtc_state->base.degamma_lut; - u32 i, lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size; + const struct drm_color_lut *lut = blob->data; + int i, lut_size = drm_color_lut_size(blob); enum pipe pipe = crtc->pipe; - I915_WRITE(PREC_PAL_INDEX(pipe), - PAL_PREC_SPLIT_MODE | PAL_PREC_AUTO_INCREMENT); + for (i = 0; i < lut_size - 1; i++) { + I915_WRITE(PALETTE(pipe, 2 * i + 0), + i965_lut_10p6_ldw(&lut[i])); + I915_WRITE(PALETTE(pipe, 2 * i + 1), + i965_lut_10p6_udw(&lut[i])); + } - if (degamma_lut) { - const struct drm_color_lut *lut = degamma_lut->data; + I915_WRITE(PIPEGCMAX(pipe, 0), lut[i].red); + I915_WRITE(PIPEGCMAX(pipe, 1), lut[i].green); + I915_WRITE(PIPEGCMAX(pipe, 2), lut[i].blue); +} - for (i = 0; i < lut_size; i++) { - u32 word = - drm_color_lut_extract(lut[i].red, 10) << 20 | - drm_color_lut_extract(lut[i].green, 10) << 10 | - drm_color_lut_extract(lut[i].blue, 10); +static void i965_load_luts(const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); + const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut; - I915_WRITE(PREC_PAL_DATA(pipe), word); - } - } else { - for (i = 0; i < lut_size; i++) { - u32 v = (i * ((1 << 10) - 1)) / (lut_size - 1); + if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) + i9xx_load_luts(crtc_state); + else + i965_load_lut_10p6(crtc, gamma_lut); +} - I915_WRITE(PREC_PAL_DATA(pipe), - (v << 20) | (v << 10) | v); - } - } +static void ilk_load_lut_10(struct intel_crtc *crtc, + const struct drm_property_blob *blob) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + const struct drm_color_lut *lut = blob->data; + int i, lut_size = drm_color_lut_size(blob); + enum pipe pipe = crtc->pipe; + + for (i = 0; i < lut_size; i++) + I915_WRITE(PREC_PALETTE(pipe, i), ilk_lut_10(&lut[i])); } -static void bdw_load_gamma_lut(const struct intel_crtc_state *crtc_state, u32 offset) +static void ilk_load_luts(const struct intel_crtc_state *crtc_state) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut; - u32 i, lut_size = INTEL_INFO(dev_priv)->color.gamma_lut_size; - enum pipe pipe = crtc->pipe; - WARN_ON(offset & ~PAL_PREC_INDEX_VALUE_MASK); + if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) + i9xx_load_luts(crtc_state); + else + ilk_load_lut_10(crtc, gamma_lut); +} - I915_WRITE(PREC_PAL_INDEX(pipe), - (offset ? PAL_PREC_SPLIT_MODE : 0) | - PAL_PREC_AUTO_INCREMENT | - offset); +static int ivb_lut_10_size(u32 prec_index) +{ + if (prec_index & PAL_PREC_SPLIT_MODE) + return 512; + else + return 1024; +} - if (gamma_lut) { - const struct drm_color_lut *lut = gamma_lut->data; +/* + * IVB/HSW Bspec / PAL_PREC_INDEX: + * "Restriction : Index auto increment mode is not + * supported and must not be enabled." + */ +static void ivb_load_lut_10(struct intel_crtc *crtc, + const struct drm_property_blob *blob, + u32 prec_index) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + int hw_lut_size = ivb_lut_10_size(prec_index); + const struct drm_color_lut *lut = blob->data; + int i, lut_size = drm_color_lut_size(blob); + enum pipe pipe = crtc->pipe; - for (i = 0; i < lut_size; i++) { - u32 word = - (drm_color_lut_extract(lut[i].red, 10) << 20) | - (drm_color_lut_extract(lut[i].green, 10) << 10) | - drm_color_lut_extract(lut[i].blue, 10); + for (i = 0; i < hw_lut_size; i++) { + /* We discard half the user entries in split gamma mode */ + const struct drm_color_lut *entry = + &lut[i * (lut_size - 1) / (hw_lut_size - 1)]; - I915_WRITE(PREC_PAL_DATA(pipe), word); - } + I915_WRITE(PREC_PAL_INDEX(pipe), prec_index++); + I915_WRITE(PREC_PAL_DATA(pipe), ilk_lut_10(entry)); + } - /* Program the max register to clamp values > 1.0. */ - i = lut_size - 1; - I915_WRITE(PREC_PAL_GC_MAX(pipe, 0), - drm_color_lut_extract(lut[i].red, 16)); - I915_WRITE(PREC_PAL_GC_MAX(pipe, 1), - drm_color_lut_extract(lut[i].green, 16)); - I915_WRITE(PREC_PAL_GC_MAX(pipe, 2), - drm_color_lut_extract(lut[i].blue, 16)); - } else { - for (i = 0; i < lut_size; i++) { - u32 v = (i * ((1 << 10) - 1)) / (lut_size - 1); + /* + * Reset the index, otherwise it prevents the legacy palette to be + * written properly. + */ + I915_WRITE(PREC_PAL_INDEX(pipe), 0); +} - I915_WRITE(PREC_PAL_DATA(pipe), - (v << 20) | (v << 10) | v); - } +/* On BDW+ the index auto increment mode actually works */ +static void bdw_load_lut_10(struct intel_crtc *crtc, + const struct drm_property_blob *blob, + u32 prec_index) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + int hw_lut_size = ivb_lut_10_size(prec_index); + const struct drm_color_lut *lut = blob->data; + int i, lut_size = drm_color_lut_size(blob); + enum pipe pipe = crtc->pipe; + + I915_WRITE(PREC_PAL_INDEX(pipe), prec_index | + PAL_PREC_AUTO_INCREMENT); - I915_WRITE(PREC_PAL_GC_MAX(pipe, 0), (1 << 16) - 1); - I915_WRITE(PREC_PAL_GC_MAX(pipe, 1), (1 << 16) - 1); - I915_WRITE(PREC_PAL_GC_MAX(pipe, 2), (1 << 16) - 1); + for (i = 0; i < hw_lut_size; i++) { + /* We discard half the user entries in split gamma mode */ + const struct drm_color_lut *entry = + &lut[i * (lut_size - 1) / (hw_lut_size - 1)]; + + I915_WRITE(PREC_PAL_DATA(pipe), ilk_lut_10(entry)); } /* @@ -547,18 +607,71 @@ static void bdw_load_gamma_lut(const struct intel_crtc_state *crtc_state, u32 of I915_WRITE(PREC_PAL_INDEX(pipe), 0); } -/* Loads the palette/gamma unit for the CRTC on Broadwell+. */ -static void broadwell_load_luts(const struct intel_crtc_state *crtc_state) +static void ivb_load_lut_10_max(struct intel_crtc *crtc) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; - if (crtc_state_is_legacy_gamma(crtc_state)) { + /* Program the max register to clamp values > 1.0. */ + I915_WRITE(PREC_PAL_EXT_GC_MAX(pipe, 0), 1 << 16); + I915_WRITE(PREC_PAL_EXT_GC_MAX(pipe, 1), 1 << 16); + I915_WRITE(PREC_PAL_EXT_GC_MAX(pipe, 2), 1 << 16); + + /* + * Program the gc max 2 register to clamp values > 1.0. + * ToDo: Extend the ABI to be able to program values + * from 3.0 to 7.0 + */ + if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) { + I915_WRITE(PREC_PAL_EXT2_GC_MAX(pipe, 0), 1 << 16); + I915_WRITE(PREC_PAL_EXT2_GC_MAX(pipe, 1), 1 << 16); + I915_WRITE(PREC_PAL_EXT2_GC_MAX(pipe, 2), 1 << 16); + } +} + +static void ivb_load_luts(const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); + const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut; + const struct drm_property_blob *degamma_lut = crtc_state->base.degamma_lut; + + if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) { i9xx_load_luts(crtc_state); + } else if (crtc_state->gamma_mode == GAMMA_MODE_MODE_SPLIT) { + ivb_load_lut_10(crtc, degamma_lut, PAL_PREC_SPLIT_MODE | + PAL_PREC_INDEX_VALUE(0)); + ivb_load_lut_10_max(crtc); + ivb_load_lut_10(crtc, gamma_lut, PAL_PREC_SPLIT_MODE | + PAL_PREC_INDEX_VALUE(512)); } else { - bdw_load_degamma_lut(crtc_state); - bdw_load_gamma_lut(crtc_state, - INTEL_INFO(dev_priv)->color.degamma_lut_size); + const struct drm_property_blob *blob = gamma_lut ?: degamma_lut; + + ivb_load_lut_10(crtc, blob, + PAL_PREC_INDEX_VALUE(0)); + ivb_load_lut_10_max(crtc); + } +} + +static void bdw_load_luts(const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); + const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut; + const struct drm_property_blob *degamma_lut = crtc_state->base.degamma_lut; + + if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) { + i9xx_load_luts(crtc_state); + } else if (crtc_state->gamma_mode == GAMMA_MODE_MODE_SPLIT) { + bdw_load_lut_10(crtc, degamma_lut, PAL_PREC_SPLIT_MODE | + PAL_PREC_INDEX_VALUE(0)); + ivb_load_lut_10_max(crtc); + bdw_load_lut_10(crtc, gamma_lut, PAL_PREC_SPLIT_MODE | + PAL_PREC_INDEX_VALUE(512)); + } else { + const struct drm_property_blob *blob = gamma_lut ?: degamma_lut; + + bdw_load_lut_10(crtc, blob, + PAL_PREC_INDEX_VALUE(0)); + ivb_load_lut_10_max(crtc); } } @@ -568,6 +681,7 @@ static void glk_load_degamma_lut(const struct intel_crtc_state *crtc_state) struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); enum pipe pipe = crtc->pipe; const u32 lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size; + const struct drm_color_lut *lut = crtc_state->base.degamma_lut->data; u32 i; /* @@ -578,58 +692,96 @@ static void glk_load_degamma_lut(const struct intel_crtc_state *crtc_state) I915_WRITE(PRE_CSC_GAMC_INDEX(pipe), 0); I915_WRITE(PRE_CSC_GAMC_INDEX(pipe), PRE_CSC_GAMC_AUTO_INCREMENT); - if (crtc_state->base.degamma_lut) { - struct drm_color_lut *lut = crtc_state->base.degamma_lut->data; + for (i = 0; i < lut_size; i++) { + /* + * First 33 entries represent range from 0 to 1.0 + * 34th and 35th entry will represent extended range + * inputs 3.0 and 7.0 respectively, currently clamped + * at 1.0. Since the precision is 16bit, the user + * value can be directly filled to register. + * The pipe degamma table in GLK+ onwards doesn't + * support different values per channel, so this just + * programs green value which will be equal to Red and + * Blue into the lut registers. + * ToDo: Extend to max 7.0. Enable 32 bit input value + * as compared to just 16 to achieve this. + */ + I915_WRITE(PRE_CSC_GAMC_DATA(pipe), lut[i].green); + } - for (i = 0; i < lut_size; i++) { - /* - * First 33 entries represent range from 0 to 1.0 - * 34th and 35th entry will represent extended range - * inputs 3.0 and 7.0 respectively, currently clamped - * at 1.0. Since the precision is 16bit, the user - * value can be directly filled to register. - * The pipe degamma table in GLK+ onwards doesn't - * support different values per channel, so this just - * programs green value which will be equal to Red and - * Blue into the lut registers. - * ToDo: Extend to max 7.0. Enable 32 bit input value - * as compared to just 16 to achieve this. - */ - I915_WRITE(PRE_CSC_GAMC_DATA(pipe), lut[i].green); - } - } else { - /* load a linear table. */ - for (i = 0; i < lut_size; i++) { - u32 v = (i * (1 << 16)) / (lut_size - 1); + /* Clamp values > 1.0. */ + while (i++ < 35) + I915_WRITE(PRE_CSC_GAMC_DATA(pipe), 1 << 16); +} - I915_WRITE(PRE_CSC_GAMC_DATA(pipe), v); - } +static void glk_load_degamma_lut_linear(const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + const u32 lut_size = INTEL_INFO(dev_priv)->color.degamma_lut_size; + u32 i; + + /* + * When setting the auto-increment bit, the hardware seems to + * ignore the index bits, so we need to reset it to index 0 + * separately. + */ + I915_WRITE(PRE_CSC_GAMC_INDEX(pipe), 0); + I915_WRITE(PRE_CSC_GAMC_INDEX(pipe), PRE_CSC_GAMC_AUTO_INCREMENT); + + for (i = 0; i < lut_size; i++) { + u32 v = (i << 16) / (lut_size - 1); + + I915_WRITE(PRE_CSC_GAMC_DATA(pipe), v); } /* Clamp values > 1.0. */ while (i++ < 35) - I915_WRITE(PRE_CSC_GAMC_DATA(pipe), (1 << 16)); + I915_WRITE(PRE_CSC_GAMC_DATA(pipe), 1 << 16); } static void glk_load_luts(const struct intel_crtc_state *crtc_state) { - glk_load_degamma_lut(crtc_state); + const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut; + struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); - if (crtc_state_is_legacy_gamma(crtc_state)) - i9xx_load_luts(crtc_state); + /* + * On GLK+ both pipe CSC and degamma LUT are controlled + * by csc_enable. Hence for the cases where the CSC is + * needed but degamma LUT is not we need to load a + * linear degamma LUT. In fact we'll just always load + * the degama LUT so that we don't have to reload + * it every time the pipe CSC is being enabled. + */ + if (crtc_state->base.degamma_lut) + glk_load_degamma_lut(crtc_state); else - bdw_load_gamma_lut(crtc_state, 0); + glk_load_degamma_lut_linear(crtc_state); + + if (crtc_state->gamma_mode == GAMMA_MODE_MODE_8BIT) { + i9xx_load_luts(crtc_state); + } else { + bdw_load_lut_10(crtc, gamma_lut, PAL_PREC_INDEX_VALUE(0)); + ivb_load_lut_10_max(crtc); + } } static void icl_load_luts(const struct intel_crtc_state *crtc_state) { - glk_load_degamma_lut(crtc_state); + const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut; + struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); - if (crtc_state_is_legacy_gamma(crtc_state)) + if (crtc_state->base.degamma_lut) + glk_load_degamma_lut(crtc_state); + + if ((crtc_state->gamma_mode & GAMMA_MODE_MODE_MASK) == + GAMMA_MODE_MODE_8BIT) { i9xx_load_luts(crtc_state); - else - /* ToDo: Add support for multi segment gamma LUT */ - bdw_load_gamma_lut(crtc_state, 0); + } else { + bdw_load_lut_10(crtc, gamma_lut, PAL_PREC_INDEX_VALUE(0)); + ivb_load_lut_10_max(crtc); + } } static void cherryview_load_luts(const struct intel_crtc_state *crtc_state) @@ -643,7 +795,7 @@ static void cherryview_load_luts(const struct intel_crtc_state *crtc_state) cherryview_load_csc_matrix(crtc_state); if (crtc_state_is_legacy_gamma(crtc_state)) { - i9xx_load_luts_internal(crtc_state, gamma_lut); + i9xx_load_luts(crtc_state); return; } @@ -682,12 +834,6 @@ static void cherryview_load_luts(const struct intel_crtc_state *crtc_state) I915_WRITE(CGM_PIPE_GAMMA(pipe, i, 1), word1); } } - - /* - * Also program a linear LUT in the legacy block (behind the - * CGM block). - */ - i9xx_load_luts_internal(crtc_state, NULL); } void intel_color_load_luts(const struct intel_crtc_state *crtc_state) @@ -704,6 +850,13 @@ void intel_color_commit(const struct intel_crtc_state *crtc_state) dev_priv->display.color_commit(crtc_state); } +int intel_color_check(struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev); + + return dev_priv->display.color_check(crtc_state); +} + static bool need_plane_update(struct intel_plane *plane, const struct intel_crtc_state *crtc_state) { @@ -771,6 +924,68 @@ static int check_lut_size(const struct drm_property_blob *lut, int expected) return 0; } +static int check_luts(const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev); + const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut; + const struct drm_property_blob *degamma_lut = crtc_state->base.degamma_lut; + int gamma_length, degamma_length; + u32 gamma_tests, degamma_tests; + + /* Always allow legacy gamma LUT with no further checking. */ + if (crtc_state_is_legacy_gamma(crtc_state)) + return 0; + + /* C8 relies on its palette being stored in the legacy LUT */ + if (crtc_state->c8_planes) + return -EINVAL; + + degamma_length = INTEL_INFO(dev_priv)->color.degamma_lut_size; + gamma_length = INTEL_INFO(dev_priv)->color.gamma_lut_size; + degamma_tests = INTEL_INFO(dev_priv)->color.degamma_lut_tests; + gamma_tests = INTEL_INFO(dev_priv)->color.gamma_lut_tests; + + if (check_lut_size(degamma_lut, degamma_length) || + check_lut_size(gamma_lut, gamma_length)) + return -EINVAL; + + if (drm_color_lut_check(degamma_lut, degamma_tests) || + drm_color_lut_check(gamma_lut, gamma_tests)) + return -EINVAL; + + return 0; +} + +static u32 i9xx_gamma_mode(struct intel_crtc_state *crtc_state) +{ + if (!crtc_state->gamma_enable || + crtc_state_is_legacy_gamma(crtc_state)) + return GAMMA_MODE_MODE_8BIT; + else + return GAMMA_MODE_MODE_10BIT; /* i965+ only */ +} + +static int i9xx_color_check(struct intel_crtc_state *crtc_state) +{ + int ret; + + ret = check_luts(crtc_state); + if (ret) + return ret; + + crtc_state->gamma_enable = + crtc_state->base.gamma_lut && + !crtc_state->c8_planes; + + crtc_state->gamma_mode = i9xx_gamma_mode(crtc_state); + + ret = intel_color_add_affected_planes(crtc_state); + if (ret) + return ret; + + return 0; +} + static u32 chv_cgm_mode(const struct intel_crtc_state *crtc_state) { u32 cgm_mode = 0; @@ -788,83 +1003,220 @@ static u32 chv_cgm_mode(const struct intel_crtc_state *crtc_state) return cgm_mode; } -int intel_color_check(struct intel_crtc_state *crtc_state) +/* + * CHV color pipeline: + * u0.10 -> CGM degamma -> u0.14 -> CGM csc -> u0.14 -> CGM gamma -> + * u0.10 -> WGC csc -> u0.10 -> pipe gamma -> u0.10 + * + * We always bypass the WGC csc and use the CGM csc + * instead since it has degamma and better precision. + */ +static int chv_color_check(struct intel_crtc_state *crtc_state) { - struct drm_i915_private *dev_priv = to_i915(crtc_state->base.crtc->dev); - const struct drm_property_blob *gamma_lut = crtc_state->base.gamma_lut; - const struct drm_property_blob *degamma_lut = crtc_state->base.degamma_lut; - bool limited_color_range = false; - int gamma_length, degamma_length; - u32 gamma_tests, degamma_tests; int ret; - degamma_length = INTEL_INFO(dev_priv)->color.degamma_lut_size; - gamma_length = INTEL_INFO(dev_priv)->color.gamma_lut_size; - degamma_tests = INTEL_INFO(dev_priv)->color.degamma_lut_tests; - gamma_tests = INTEL_INFO(dev_priv)->color.gamma_lut_tests; + ret = check_luts(crtc_state); + if (ret) + return ret; - /* C8 needs the legacy LUT all to itself */ - if (crtc_state->c8_planes && - !crtc_state_is_legacy_gamma(crtc_state)) - return -EINVAL; + /* + * Pipe gamma will be used only for the legacy LUT. + * Otherwise we bypass it and use the CGM gamma instead. + */ + crtc_state->gamma_enable = + crtc_state_is_legacy_gamma(crtc_state) && + !crtc_state->c8_planes; + + crtc_state->gamma_mode = GAMMA_MODE_MODE_8BIT; + + crtc_state->cgm_mode = chv_cgm_mode(crtc_state); - crtc_state->gamma_enable = (gamma_lut || degamma_lut) && + ret = intel_color_add_affected_planes(crtc_state); + if (ret) + return ret; + + return 0; +} + +static u32 ilk_gamma_mode(const struct intel_crtc_state *crtc_state) +{ + if (!crtc_state->gamma_enable || + crtc_state_is_legacy_gamma(crtc_state)) + return GAMMA_MODE_MODE_8BIT; + else + return GAMMA_MODE_MODE_10BIT; +} + +static int ilk_color_check(struct intel_crtc_state *crtc_state) +{ + int ret; + + ret = check_luts(crtc_state); + if (ret) + return ret; + + crtc_state->gamma_enable = + crtc_state->base.gamma_lut && !crtc_state->c8_planes; - if (INTEL_GEN(dev_priv) >= 9 || - IS_BROADWELL(dev_priv) || IS_HASWELL(dev_priv)) - limited_color_range = crtc_state->limited_color_range; + /* + * We don't expose the ctm on ilk/snb currently, + * nor do we enable YCbCr output. Also RGB limited + * range output is handled by the hw automagically. + */ + crtc_state->csc_enable = false; + + crtc_state->gamma_mode = ilk_gamma_mode(crtc_state); + + crtc_state->csc_mode = 0; + + ret = intel_color_add_affected_planes(crtc_state); + if (ret) + return ret; + + return 0; +} + +static u32 ivb_gamma_mode(const struct intel_crtc_state *crtc_state) +{ + if (!crtc_state->gamma_enable || + crtc_state_is_legacy_gamma(crtc_state)) + return GAMMA_MODE_MODE_8BIT; + else if (crtc_state->base.gamma_lut && + crtc_state->base.degamma_lut) + return GAMMA_MODE_MODE_SPLIT; + else + return GAMMA_MODE_MODE_10BIT; +} + +static u32 ivb_csc_mode(const struct intel_crtc_state *crtc_state) +{ + bool limited_color_range = ilk_csc_limited_range(crtc_state); + + /* + * CSC comes after the LUT in degamma, RGB->YCbCr, + * and RGB full->limited range mode. + */ + if (crtc_state->base.degamma_lut || + crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB || + limited_color_range) + return 0; + + return CSC_POSITION_BEFORE_GAMMA; +} + +static int ivb_color_check(struct intel_crtc_state *crtc_state) +{ + bool limited_color_range = ilk_csc_limited_range(crtc_state); + int ret; + + ret = check_luts(crtc_state); + if (ret) + return ret; + + crtc_state->gamma_enable = + (crtc_state->base.gamma_lut || + crtc_state->base.degamma_lut) && + !crtc_state->c8_planes; crtc_state->csc_enable = crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB || crtc_state->base.ctm || limited_color_range; + crtc_state->gamma_mode = ivb_gamma_mode(crtc_state); + + crtc_state->csc_mode = ivb_csc_mode(crtc_state); + ret = intel_color_add_affected_planes(crtc_state); if (ret) return ret; + return 0; +} + +static u32 glk_gamma_mode(const struct intel_crtc_state *crtc_state) +{ + if (!crtc_state->gamma_enable || + crtc_state_is_legacy_gamma(crtc_state)) + return GAMMA_MODE_MODE_8BIT; + else + return GAMMA_MODE_MODE_10BIT; +} + +static int glk_color_check(struct intel_crtc_state *crtc_state) +{ + int ret; + + ret = check_luts(crtc_state); + if (ret) + return ret; + + crtc_state->gamma_enable = + crtc_state->base.gamma_lut && + !crtc_state->c8_planes; + + /* On GLK+ degamma LUT is controlled by csc_enable */ + crtc_state->csc_enable = + crtc_state->base.degamma_lut || + crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB || + crtc_state->base.ctm || crtc_state->limited_color_range; + + crtc_state->gamma_mode = glk_gamma_mode(crtc_state); + crtc_state->csc_mode = 0; - if (IS_CHERRYVIEW(dev_priv)) - crtc_state->cgm_mode = chv_cgm_mode(crtc_state); + ret = intel_color_add_affected_planes(crtc_state); + if (ret) + return ret; - /* Always allow legacy gamma LUT with no further checking. */ - if (!crtc_state->gamma_enable || - crtc_state_is_legacy_gamma(crtc_state)) { - crtc_state->gamma_mode = GAMMA_MODE_MODE_8BIT; - if (INTEL_GEN(dev_priv) >= 11 && - crtc_state->gamma_enable) - crtc_state->gamma_mode |= POST_CSC_GAMMA_ENABLE; - return 0; - } + return 0; +} - if (check_lut_size(degamma_lut, degamma_length) || - check_lut_size(gamma_lut, gamma_length)) - return -EINVAL; +static u32 icl_gamma_mode(const struct intel_crtc_state *crtc_state) +{ + u32 gamma_mode = 0; - if (drm_color_lut_check(degamma_lut, degamma_tests) || - drm_color_lut_check(gamma_lut, gamma_tests)) - return -EINVAL; + if (crtc_state->base.degamma_lut) + gamma_mode |= PRE_CSC_GAMMA_ENABLE; - if (INTEL_GEN(dev_priv) >= 11) - crtc_state->gamma_mode = GAMMA_MODE_MODE_10BIT | - PRE_CSC_GAMMA_ENABLE | - POST_CSC_GAMMA_ENABLE; - else if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) - crtc_state->gamma_mode = GAMMA_MODE_MODE_10BIT; - else if (INTEL_GEN(dev_priv) >= 9 || IS_BROADWELL(dev_priv)) - crtc_state->gamma_mode = GAMMA_MODE_MODE_SPLIT; + if (crtc_state->base.gamma_lut && + !crtc_state->c8_planes) + gamma_mode |= POST_CSC_GAMMA_ENABLE; + + if (!crtc_state->base.gamma_lut || + crtc_state_is_legacy_gamma(crtc_state)) + gamma_mode |= GAMMA_MODE_MODE_8BIT; else - crtc_state->gamma_mode = GAMMA_MODE_MODE_8BIT; + gamma_mode |= GAMMA_MODE_MODE_10BIT; - if (INTEL_GEN(dev_priv) >= 11) { - if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB || - crtc_state->limited_color_range) - crtc_state->csc_mode |= ICL_OUTPUT_CSC_ENABLE; + return gamma_mode; +} - if (crtc_state->base.ctm) - crtc_state->csc_mode |= ICL_CSC_ENABLE; - } +static u32 icl_csc_mode(const struct intel_crtc_state *crtc_state) +{ + u32 csc_mode = 0; + + if (crtc_state->base.ctm) + csc_mode |= ICL_CSC_ENABLE; + + if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB || + crtc_state->limited_color_range) + csc_mode |= ICL_OUTPUT_CSC_ENABLE; + + return csc_mode; +} + +static int icl_color_check(struct intel_crtc_state *crtc_state) +{ + int ret; + + ret = check_luts(crtc_state); + if (ret) + return ret; + + crtc_state->gamma_mode = icl_gamma_mode(crtc_state); + + crtc_state->csc_mode = icl_csc_mode(crtc_state); return 0; } @@ -872,25 +1224,33 @@ int intel_color_check(struct intel_crtc_state *crtc_state) void intel_color_init(struct intel_crtc *crtc) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + bool has_ctm = INTEL_INFO(dev_priv)->color.degamma_lut_size != 0; drm_mode_crtc_set_gamma_size(&crtc->base, 256); if (HAS_GMCH(dev_priv)) { - if (IS_CHERRYVIEW(dev_priv)) + if (IS_CHERRYVIEW(dev_priv)) { + dev_priv->display.color_check = chv_color_check; + dev_priv->display.color_commit = i9xx_color_commit; dev_priv->display.load_luts = cherryview_load_luts; - else + } else if (INTEL_GEN(dev_priv) >= 4) { + dev_priv->display.color_check = i9xx_color_check; + dev_priv->display.color_commit = i9xx_color_commit; + dev_priv->display.load_luts = i965_load_luts; + } else { + dev_priv->display.color_check = i9xx_color_check; + dev_priv->display.color_commit = i9xx_color_commit; dev_priv->display.load_luts = i9xx_load_luts; - - dev_priv->display.color_commit = i9xx_color_commit; + } } else { if (INTEL_GEN(dev_priv) >= 11) - dev_priv->display.load_luts = icl_load_luts; - else if (IS_CANNONLAKE(dev_priv) || IS_GEMINILAKE(dev_priv)) - dev_priv->display.load_luts = glk_load_luts; - else if (INTEL_GEN(dev_priv) >= 9 || IS_BROADWELL(dev_priv)) - dev_priv->display.load_luts = broadwell_load_luts; + dev_priv->display.color_check = icl_color_check; + else if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) + dev_priv->display.color_check = glk_color_check; + else if (INTEL_GEN(dev_priv) >= 7) + dev_priv->display.color_check = ivb_color_check; else - dev_priv->display.load_luts = i9xx_load_luts; + dev_priv->display.color_check = ilk_color_check; if (INTEL_GEN(dev_priv) >= 9) dev_priv->display.color_commit = skl_color_commit; @@ -898,13 +1258,21 @@ void intel_color_init(struct intel_crtc *crtc) dev_priv->display.color_commit = hsw_color_commit; else dev_priv->display.color_commit = ilk_color_commit; + + if (INTEL_GEN(dev_priv) >= 11) + dev_priv->display.load_luts = icl_load_luts; + else if (IS_CANNONLAKE(dev_priv) || IS_GEMINILAKE(dev_priv)) + dev_priv->display.load_luts = glk_load_luts; + else if (INTEL_GEN(dev_priv) >= 8) + dev_priv->display.load_luts = bdw_load_luts; + else if (INTEL_GEN(dev_priv) >= 7) + dev_priv->display.load_luts = ivb_load_luts; + else + dev_priv->display.load_luts = ilk_load_luts; } - /* Enable color management support when we have degamma & gamma LUTs. */ - if (INTEL_INFO(dev_priv)->color.degamma_lut_size != 0 && - INTEL_INFO(dev_priv)->color.gamma_lut_size != 0) - drm_crtc_enable_color_mgmt(&crtc->base, - INTEL_INFO(dev_priv)->color.degamma_lut_size, - true, - INTEL_INFO(dev_priv)->color.gamma_lut_size); + drm_crtc_enable_color_mgmt(&crtc->base, + INTEL_INFO(dev_priv)->color.degamma_lut_size, + has_ctm, + INTEL_INFO(dev_priv)->color.gamma_lut_size); } diff --git a/drivers/gpu/drm/i915/intel_color.h b/drivers/gpu/drm/i915/intel_color.h new file mode 100644 index 000000000000..b8a3ce609587 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_color.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_COLOR_H__ +#define __INTEL_COLOR_H__ + +struct intel_crtc_state; +struct intel_crtc; + +void intel_color_init(struct intel_crtc *crtc); +int intel_color_check(struct intel_crtc_state *crtc_state); +void intel_color_commit(const struct intel_crtc_state *crtc_state); +void intel_color_load_luts(const struct intel_crtc_state *crtc_state); + +#endif /* __INTEL_COLOR_H__ */ diff --git a/drivers/gpu/drm/i915/intel_combo_phy.c b/drivers/gpu/drm/i915/intel_combo_phy.c index 3d0271cebf99..2bf4359d7e41 100644 --- a/drivers/gpu/drm/i915/intel_combo_phy.c +++ b/drivers/gpu/drm/i915/intel_combo_phy.c @@ -239,7 +239,8 @@ void icl_combo_phys_uninit(struct drm_i915_private *dev_priv) for_each_combo_port_reverse(dev_priv, port) { u32 val; - if (!icl_combo_phy_verify_state(dev_priv, port)) + if (port == PORT_A && + !icl_combo_phy_verify_state(dev_priv, port)) DRM_WARN("Port %c combo PHY HW state changed unexpectedly\n", port_name(port)); diff --git a/drivers/gpu/drm/i915/intel_connector.c b/drivers/gpu/drm/i915/intel_connector.c index 848dd9e728d8..073b6c3ab7cc 100644 --- a/drivers/gpu/drm/i915/intel_connector.c +++ b/drivers/gpu/drm/i915/intel_connector.c @@ -23,12 +23,17 @@ * DEALINGS IN THE SOFTWARE. */ -#include <linux/slab.h> #include <linux/i2c.h> +#include <linux/slab.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_edid.h> -#include "intel_drv.h" + #include "i915_drv.h" +#include "intel_connector.h" +#include "intel_drv.h" +#include "intel_hdcp.h" +#include "intel_panel.h" int intel_connector_init(struct intel_connector *connector) { diff --git a/drivers/gpu/drm/i915/intel_connector.h b/drivers/gpu/drm/i915/intel_connector.h new file mode 100644 index 000000000000..93a7375c8196 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_connector.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_CONNECTOR_H__ +#define __INTEL_CONNECTOR_H__ + +#include "intel_display.h" + +struct drm_connector; +struct edid; +struct i2c_adapter; +struct intel_connector; +struct intel_encoder; + +int intel_connector_init(struct intel_connector *connector); +struct intel_connector *intel_connector_alloc(void); +void intel_connector_free(struct intel_connector *connector); +void intel_connector_destroy(struct drm_connector *connector); +int intel_connector_register(struct drm_connector *connector); +void intel_connector_unregister(struct drm_connector *connector); +void intel_connector_attach_encoder(struct intel_connector *connector, + struct intel_encoder *encoder); +bool intel_connector_get_hw_state(struct intel_connector *connector); +enum pipe intel_connector_get_pipe(struct intel_connector *connector); +int intel_connector_update_modes(struct drm_connector *connector, + struct edid *edid); +int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter); +void intel_attach_force_audio_property(struct drm_connector *connector); +void intel_attach_broadcast_rgb_property(struct drm_connector *connector); +void intel_attach_aspect_ratio_property(struct drm_connector *connector); +void intel_attach_colorspace_property(struct drm_connector *connector); + +#endif /* __INTEL_CONNECTOR_H__ */ diff --git a/drivers/gpu/drm/i915/intel_context_types.h b/drivers/gpu/drm/i915/intel_context_types.h index 624729a35875..68b4ca1611e0 100644 --- a/drivers/gpu/drm/i915/intel_context_types.h +++ b/drivers/gpu/drm/i915/intel_context_types.h @@ -24,6 +24,7 @@ struct intel_context_ops { int (*pin)(struct intel_context *ce); void (*unpin)(struct intel_context *ce); + void (*reset)(struct intel_context *ce); void (*destroy)(struct kref *kref); }; diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c index 50530e49982c..b665c370111b 100644 --- a/drivers/gpu/drm/i915/intel_crt.c +++ b/drivers/gpu/drm/i915/intel_crt.c @@ -27,13 +27,18 @@ #include <linux/dmi.h> #include <linux/i2c.h> #include <linux/slab.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_edid.h> #include <drm/drm_probe_helper.h> -#include "intel_drv.h" #include <drm/i915_drm.h> + #include "i915_drv.h" +#include "intel_connector.h" +#include "intel_crt.h" +#include "intel_ddi.h" +#include "intel_drv.h" /* Here's the desired hotplug mode */ #define ADPA_HOTPLUG_BITS (ADPA_CRT_HOTPLUG_PERIOD_128 | \ diff --git a/drivers/gpu/drm/i915/intel_crt.h b/drivers/gpu/drm/i915/intel_crt.h new file mode 100644 index 000000000000..1b3fba359efc --- /dev/null +++ b/drivers/gpu/drm/i915/intel_crt.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_CRT_H__ +#define __INTEL_CRT_H__ + +#include "i915_reg.h" + +enum pipe; +struct drm_encoder; +struct drm_i915_private; +struct drm_i915_private; + +bool intel_crt_port_enabled(struct drm_i915_private *dev_priv, + i915_reg_t adpa_reg, enum pipe *pipe); +void intel_crt_init(struct drm_i915_private *dev_priv); +void intel_crt_reset(struct drm_encoder *encoder); + +#endif /* __INTEL_CRT_H__ */ diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c index 862a8f686ef5..f43c2a2563a5 100644 --- a/drivers/gpu/drm/i915/intel_csr.c +++ b/drivers/gpu/drm/i915/intel_csr.c @@ -21,9 +21,12 @@ * IN THE SOFTWARE. * */ + #include <linux/firmware.h> + #include "i915_drv.h" #include "i915_reg.h" +#include "intel_csr.h" /** * DOC: csr support for dmc diff --git a/drivers/gpu/drm/i915/intel_csr.h b/drivers/gpu/drm/i915/intel_csr.h new file mode 100644 index 000000000000..17a32c1e8a35 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_csr.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_CSR_H__ +#define __INTEL_CSR_H__ + +struct drm_i915_private; + +void intel_csr_ucode_init(struct drm_i915_private *i915); +void intel_csr_load_program(struct drm_i915_private *i915); +void intel_csr_ucode_fini(struct drm_i915_private *i915); +void intel_csr_ucode_suspend(struct drm_i915_private *i915); +void intel_csr_ucode_resume(struct drm_i915_private *i915); + +#endif /* __INTEL_CSR_H__ */ diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index 3f1e491bd0c0..24f9106efcc6 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -26,9 +26,19 @@ */ #include <drm/drm_scdc_helper.h> + #include "i915_drv.h" +#include "intel_audio.h" +#include "intel_connector.h" +#include "intel_ddi.h" +#include "intel_dp.h" #include "intel_drv.h" #include "intel_dsi.h" +#include "intel_hdcp.h" +#include "intel_hdmi.h" +#include "intel_lspcon.h" +#include "intel_panel.h" +#include "intel_psr.h" struct ddi_buf_trans { u32 trans1; /* balance leg enable, de-emph level */ @@ -2043,12 +2053,11 @@ intel_ddi_main_link_aux_domain(struct intel_digital_port *dig_port) intel_aux_power_domain(dig_port); } -static u64 intel_ddi_get_power_domains(struct intel_encoder *encoder, - struct intel_crtc_state *crtc_state) +static void intel_ddi_get_power_domains(struct intel_encoder *encoder, + struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); struct intel_digital_port *dig_port; - u64 domains; /* * TODO: Add support for MST encoders. Atm, the following should never @@ -2056,10 +2065,10 @@ static u64 intel_ddi_get_power_domains(struct intel_encoder *encoder, * hook. */ if (WARN_ON(intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST))) - return 0; + return; dig_port = enc_to_dig_port(&encoder->base); - domains = BIT_ULL(dig_port->ddi_io_power_domain); + intel_display_power_get(dev_priv, dig_port->ddi_io_power_domain); /* * AUX power is only needed for (e)DP mode, and for HDMI mode on TC @@ -2067,15 +2076,15 @@ static u64 intel_ddi_get_power_domains(struct intel_encoder *encoder, */ if (intel_crtc_has_dp_encoder(crtc_state) || intel_port_is_tc(dev_priv, encoder->port)) - domains |= BIT_ULL(intel_ddi_main_link_aux_domain(dig_port)); + intel_display_power_get(dev_priv, + intel_ddi_main_link_aux_domain(dig_port)); /* * VDSC power is needed when DSC is enabled */ if (crtc_state->dsc_params.compression_enable) - domains |= BIT_ULL(intel_dsc_power_domain(crtc_state)); - - return domains; + intel_display_power_get(dev_priv, + intel_dsc_power_domain(crtc_state)); } void intel_ddi_enable_pipe_clock(const struct intel_crtc_state *crtc_state) @@ -2793,10 +2802,10 @@ void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder) return; } /* - * DSI ports should have their DDI clock ungated when disabled - * and gated when enabled. + * For DSI we keep the ddi clocks gated + * except during enable/disable sequence. */ - ddi_clk_needed = !encoder->base.crtc; + ddi_clk_needed = false; } val = I915_READ(DPCLKA_CFGCR0_ICL); @@ -3848,14 +3857,16 @@ static int intel_ddi_compute_config(struct intel_encoder *encoder, ret = intel_hdmi_compute_config(encoder, pipe_config, conn_state); else ret = intel_dp_compute_config(encoder, pipe_config, conn_state); + if (ret) + return ret; - if (IS_GEN9_LP(dev_priv) && ret) + if (IS_GEN9_LP(dev_priv)) pipe_config->lane_lat_optim_mask = bxt_ddi_phy_calc_lane_lat_optim_mask(pipe_config->lane_count); intel_ddi_compute_min_voltage_level(dev_priv, pipe_config); - return ret; + return 0; } diff --git a/drivers/gpu/drm/i915/intel_ddi.h b/drivers/gpu/drm/i915/intel_ddi.h new file mode 100644 index 000000000000..9cf69175942e --- /dev/null +++ b/drivers/gpu/drm/i915/intel_ddi.h @@ -0,0 +1,53 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_DDI_H__ +#define __INTEL_DDI_H__ + +#include <drm/i915_drm.h> + +#include "intel_display.h" + +struct drm_connector_state; +struct drm_i915_private; +struct intel_connector; +struct intel_crtc; +struct intel_crtc_state; +struct intel_dp; +struct intel_dpll_hw_state; +struct intel_encoder; + +void intel_ddi_fdi_post_disable(struct intel_encoder *intel_encoder, + const struct intel_crtc_state *old_crtc_state, + const struct drm_connector_state *old_conn_state); +void hsw_fdi_link_train(struct intel_crtc *crtc, + const struct intel_crtc_state *crtc_state); +void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port); +bool intel_ddi_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe); +void intel_ddi_enable_transcoder_func(const struct intel_crtc_state *crtc_state); +void intel_ddi_disable_transcoder_func(const struct intel_crtc_state *crtc_state); +void intel_ddi_enable_pipe_clock(const struct intel_crtc_state *crtc_state); +void intel_ddi_disable_pipe_clock(const struct intel_crtc_state *crtc_state); +void intel_ddi_set_pipe_settings(const struct intel_crtc_state *crtc_state); +void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp); +bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector); +void intel_ddi_get_config(struct intel_encoder *encoder, + struct intel_crtc_state *pipe_config); +void intel_ddi_set_vc_payload_alloc(const struct intel_crtc_state *crtc_state, + bool state); +void intel_ddi_compute_min_voltage_level(struct drm_i915_private *dev_priv, + struct intel_crtc_state *crtc_state); +u32 bxt_signal_levels(struct intel_dp *intel_dp); +u32 ddi_signal_levels(struct intel_dp *intel_dp); +u8 intel_ddi_dp_voltage_max(struct intel_encoder *encoder); +u8 intel_ddi_dp_pre_emphasis_max(struct intel_encoder *encoder, + u8 voltage_swing); +int intel_ddi_toggle_hdcp_signalling(struct intel_encoder *intel_encoder, + bool enable); +void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder); +int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv, + struct intel_dpll_hw_state *state); + +#endif /* __INTEL_DDI_H__ */ diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c index e0f5e0231d04..6af480b95bc6 100644 --- a/drivers/gpu/drm/i915/intel_device_info.c +++ b/drivers/gpu/drm/i915/intel_device_info.c @@ -714,6 +714,99 @@ static u32 read_timestamp_frequency(struct drm_i915_private *dev_priv) return 0; } +#undef INTEL_VGA_DEVICE +#define INTEL_VGA_DEVICE(id, info) (id) + +static const u16 subplatform_ult_ids[] = { + INTEL_HSW_ULT_GT1_IDS(0), + INTEL_HSW_ULT_GT2_IDS(0), + INTEL_HSW_ULT_GT3_IDS(0), + INTEL_BDW_ULT_GT1_IDS(0), + INTEL_BDW_ULT_GT2_IDS(0), + INTEL_BDW_ULT_GT3_IDS(0), + INTEL_BDW_ULT_RSVD_IDS(0), + INTEL_SKL_ULT_GT1_IDS(0), + INTEL_SKL_ULT_GT2_IDS(0), + INTEL_SKL_ULT_GT3_IDS(0), + INTEL_KBL_ULT_GT1_IDS(0), + INTEL_KBL_ULT_GT2_IDS(0), + INTEL_KBL_ULT_GT3_IDS(0), + INTEL_CFL_U_GT2_IDS(0), + INTEL_CFL_U_GT3_IDS(0), + INTEL_WHL_U_GT1_IDS(0), + INTEL_WHL_U_GT2_IDS(0), + INTEL_WHL_U_GT3_IDS(0) +}; + +static const u16 subplatform_ulx_ids[] = { + INTEL_HSW_ULX_GT1_IDS(0), + INTEL_HSW_ULX_GT2_IDS(0), + INTEL_BDW_ULX_GT1_IDS(0), + INTEL_BDW_ULX_GT2_IDS(0), + INTEL_BDW_ULX_GT3_IDS(0), + INTEL_BDW_ULX_RSVD_IDS(0), + INTEL_SKL_ULX_GT1_IDS(0), + INTEL_SKL_ULX_GT2_IDS(0), + INTEL_KBL_ULX_GT1_IDS(0), + INTEL_KBL_ULX_GT2_IDS(0) +}; + +static const u16 subplatform_aml_ids[] = { + INTEL_AML_KBL_GT2_IDS(0), + INTEL_AML_CFL_GT2_IDS(0) +}; + +static const u16 subplatform_portf_ids[] = { + INTEL_CNL_PORT_F_IDS(0), + INTEL_ICL_PORT_F_IDS(0) +}; + +static bool find_devid(u16 id, const u16 *p, unsigned int num) +{ + for (; num; num--, p++) { + if (*p == id) + return true; + } + + return false; +} + +void intel_device_info_subplatform_init(struct drm_i915_private *i915) +{ + const struct intel_device_info *info = INTEL_INFO(i915); + const struct intel_runtime_info *rinfo = RUNTIME_INFO(i915); + const unsigned int pi = __platform_mask_index(rinfo, info->platform); + const unsigned int pb = __platform_mask_bit(rinfo, info->platform); + u16 devid = INTEL_DEVID(i915); + u32 mask = 0; + + /* Make sure IS_<platform> checks are working. */ + RUNTIME_INFO(i915)->platform_mask[pi] = BIT(pb); + + /* Find and mark subplatform bits based on the PCI device id. */ + if (find_devid(devid, subplatform_ult_ids, + ARRAY_SIZE(subplatform_ult_ids))) { + mask = BIT(INTEL_SUBPLATFORM_ULT); + } else if (find_devid(devid, subplatform_ulx_ids, + ARRAY_SIZE(subplatform_ulx_ids))) { + mask = BIT(INTEL_SUBPLATFORM_ULX); + if (IS_HASWELL(i915) || IS_BROADWELL(i915)) { + /* ULX machines are also considered ULT. */ + mask |= BIT(INTEL_SUBPLATFORM_ULT); + } + } else if (find_devid(devid, subplatform_aml_ids, + ARRAY_SIZE(subplatform_aml_ids))) { + mask = BIT(INTEL_SUBPLATFORM_AML); + } else if (find_devid(devid, subplatform_portf_ids, + ARRAY_SIZE(subplatform_portf_ids))) { + mask = BIT(INTEL_SUBPLATFORM_PORTF); + } + + GEM_BUG_ON(mask & ~INTEL_SUBPLATFORM_BITS); + + RUNTIME_INFO(i915)->platform_mask[pi] |= mask; +} + /** * intel_device_info_runtime_init - initialize runtime info * @dev_priv: the i915 device diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h index 7e04b4829aba..0e579f158016 100644 --- a/drivers/gpu/drm/i915/intel_device_info.h +++ b/drivers/gpu/drm/i915/intel_device_info.h @@ -27,6 +27,7 @@ #include <uapi/drm/i915_drm.h> +#include "intel_engine_types.h" #include "intel_display.h" struct drm_printer; @@ -77,6 +78,21 @@ enum intel_platform { INTEL_MAX_PLATFORMS }; +/* + * Subplatform bits share the same namespace per parent platform. In other words + * it is fine for the same bit to be used on multiple parent platforms. + */ + +#define INTEL_SUBPLATFORM_BITS (3) + +/* HSW/BDW/SKL/KBL/CFL */ +#define INTEL_SUBPLATFORM_ULT (0) +#define INTEL_SUBPLATFORM_ULX (1) +#define INTEL_SUBPLATFORM_AML (2) + +/* CNL/ICL */ +#define INTEL_SUBPLATFORM_PORTF (0) + enum intel_ppgtt_type { INTEL_PPGTT_NONE = I915_GEM_PPGTT_NONE, INTEL_PPGTT_ALIASING = I915_GEM_PPGTT_ALIASING, @@ -150,8 +166,6 @@ struct sseu_dev_info { u8 eu_mask[GEN_MAX_SLICES * GEN_MAX_SUBSLICES]; }; -typedef u8 intel_engine_mask_t; - struct intel_device_info { u16 gen_mask; @@ -160,7 +174,6 @@ struct intel_device_info { intel_engine_mask_t engine_mask; /* Engines supported by the HW */ enum intel_platform platform; - u32 platform_mask; enum intel_ppgtt_type ppgtt_type; unsigned int ppgtt_size; /* log2, e.g. 31/32/48 bits */ @@ -197,6 +210,16 @@ struct intel_device_info { }; struct intel_runtime_info { + /* + * Platform mask is used for optimizing or-ed IS_PLATFORM calls into + * into single runtime conditionals, and also to provide groundwork + * for future per platform, or per SKU build optimizations. + * + * Array can be extended when necessary if the corresponding + * BUILD_BUG_ON is hit. + */ + u32 platform_mask[2]; + u16 device_id; u8 num_sprites[I915_MAX_PIPES]; @@ -267,6 +290,7 @@ static inline void sseu_set_eus(struct sseu_dev_info *sseu, const char *intel_platform_name(enum intel_platform platform); +void intel_device_info_subplatform_init(struct drm_i915_private *dev_priv); void intel_device_info_runtime_init(struct drm_i915_private *dev_priv); void intel_device_info_dump_flags(const struct intel_device_info *info, struct drm_printer *p); diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 8576a7f799f2..3bd40a4a6739 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -46,19 +46,29 @@ #include "i915_drv.h" #include "i915_gem_clflush.h" +#include "i915_reset.h" #include "i915_trace.h" +#include "intel_atomic_plane.h" +#include "intel_color.h" +#include "intel_cdclk.h" +#include "intel_crt.h" +#include "intel_ddi.h" +#include "intel_dp.h" #include "intel_drv.h" #include "intel_dsi.h" +#include "intel_dvo.h" +#include "intel_fbc.h" +#include "intel_fbdev.h" #include "intel_frontbuffer.h" - -#include "intel_drv.h" -#include "intel_dsi.h" -#include "intel_frontbuffer.h" - -#include "i915_drv.h" -#include "i915_gem_clflush.h" -#include "i915_reset.h" -#include "i915_trace.h" +#include "intel_hdcp.h" +#include "intel_hdmi.h" +#include "intel_lvds.h" +#include "intel_pipe_crc.h" +#include "intel_pm.h" +#include "intel_psr.h" +#include "intel_sdvo.h" +#include "intel_sprite.h" +#include "intel_tv.h" /* Primary plane formats for gen <= 3 */ static const u32 i8xx_primary_formats[] = { @@ -115,8 +125,8 @@ static void vlv_prepare_pll(struct intel_crtc *crtc, const struct intel_crtc_state *pipe_config); static void chv_prepare_pll(struct intel_crtc *crtc, const struct intel_crtc_state *pipe_config); -static void intel_begin_crtc_commit(struct drm_crtc *, struct drm_crtc_state *); -static void intel_finish_crtc_commit(struct drm_crtc *, struct drm_crtc_state *); +static void intel_begin_crtc_commit(struct intel_atomic_state *, struct intel_crtc *); +static void intel_finish_crtc_commit(struct intel_atomic_state *, struct intel_crtc *); static void intel_crtc_init_scalers(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state); static void skylake_pfit_enable(const struct intel_crtc_state *crtc_state); @@ -467,10 +477,11 @@ static const struct intel_limit intel_limits_bxt = { }; static void -skl_wa_clkgate(struct drm_i915_private *dev_priv, int pipe, bool enable) +skl_wa_827(struct drm_i915_private *dev_priv, int pipe, bool enable) { if (enable) I915_WRITE(CLKGATE_DIS_PSL(pipe), + I915_READ(CLKGATE_DIS_PSL(pipe)) | DUPS1_GATING_DIS | DUPS2_GATING_DIS); else I915_WRITE(CLKGATE_DIS_PSL(pipe), @@ -5530,7 +5541,7 @@ static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state) /* Display WA 827 */ if (needs_nv12_wa(dev_priv, old_crtc_state) && !needs_nv12_wa(dev_priv, pipe_config)) { - skl_wa_clkgate(dev_priv, crtc->pipe, false); + skl_wa_827(dev_priv, crtc->pipe, false); } } @@ -5569,7 +5580,7 @@ static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state, /* Display WA 827 */ if (!needs_nv12_wa(dev_priv, old_crtc_state) && needs_nv12_wa(dev_priv, pipe_config)) { - skl_wa_clkgate(dev_priv, crtc->pipe, true); + skl_wa_827(dev_priv, crtc->pipe, true); } /* @@ -6180,6 +6191,9 @@ bool intel_port_is_combophy(struct drm_i915_private *dev_priv, enum port port) if (port == PORT_NONE) return false; + if (IS_ELKHARTLAKE(dev_priv)) + return port <= PORT_C; + if (INTEL_GEN(dev_priv) >= 11) return port <= PORT_B; @@ -6188,7 +6202,7 @@ bool intel_port_is_combophy(struct drm_i915_private *dev_priv, enum port port) bool intel_port_is_tc(struct drm_i915_private *dev_priv, enum port port) { - if (INTEL_GEN(dev_priv) >= 11) + if (INTEL_GEN(dev_priv) >= 11 && !IS_ELKHARTLAKE(dev_priv)) return port >= PORT_C && port <= PORT_F; return false; @@ -9751,7 +9765,8 @@ static void haswell_get_ddi_pll(struct drm_i915_private *dev_priv, static bool hsw_get_transcoder_state(struct intel_crtc *crtc, struct intel_crtc_state *pipe_config, - u64 *power_domain_mask) + u64 *power_domain_mask, + intel_wakeref_t *wakerefs) { struct drm_device *dev = crtc->base.dev; struct drm_i915_private *dev_priv = to_i915(dev); @@ -9759,6 +9774,7 @@ static bool hsw_get_transcoder_state(struct intel_crtc *crtc, unsigned long panel_transcoder_mask = 0; unsigned long enabled_panel_transcoders = 0; enum transcoder panel_transcoder; + intel_wakeref_t wf; u32 tmp; if (INTEL_GEN(dev_priv) >= 11) @@ -9824,10 +9840,13 @@ static bool hsw_get_transcoder_state(struct intel_crtc *crtc, enabled_panel_transcoders != BIT(TRANSCODER_EDP)); power_domain = POWER_DOMAIN_TRANSCODER(pipe_config->cpu_transcoder); - if (!intel_display_power_get_if_enabled(dev_priv, power_domain)) + WARN_ON(*power_domain_mask & BIT_ULL(power_domain)); + + wf = intel_display_power_get_if_enabled(dev_priv, power_domain); + if (!wf) return false; - WARN_ON(*power_domain_mask & BIT_ULL(power_domain)); + wakerefs[power_domain] = wf; *power_domain_mask |= BIT_ULL(power_domain); tmp = I915_READ(PIPECONF(pipe_config->cpu_transcoder)); @@ -9837,13 +9856,15 @@ static bool hsw_get_transcoder_state(struct intel_crtc *crtc, static bool bxt_get_dsi_transcoder_state(struct intel_crtc *crtc, struct intel_crtc_state *pipe_config, - u64 *power_domain_mask) + u64 *power_domain_mask, + intel_wakeref_t *wakerefs) { struct drm_device *dev = crtc->base.dev; struct drm_i915_private *dev_priv = to_i915(dev); enum intel_display_power_domain power_domain; - enum port port; enum transcoder cpu_transcoder; + intel_wakeref_t wf; + enum port port; u32 tmp; for_each_port_masked(port, BIT(PORT_A) | BIT(PORT_C)) { @@ -9853,10 +9874,13 @@ static bool bxt_get_dsi_transcoder_state(struct intel_crtc *crtc, cpu_transcoder = TRANSCODER_DSI_C; power_domain = POWER_DOMAIN_TRANSCODER(cpu_transcoder); - if (!intel_display_power_get_if_enabled(dev_priv, power_domain)) + WARN_ON(*power_domain_mask & BIT_ULL(power_domain)); + + wf = intel_display_power_get_if_enabled(dev_priv, power_domain); + if (!wf) continue; - WARN_ON(*power_domain_mask & BIT_ULL(power_domain)); + wakerefs[power_domain] = wf; *power_domain_mask |= BIT_ULL(power_domain); /* @@ -9935,6 +9959,7 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc, struct intel_crtc_state *pipe_config) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + intel_wakeref_t wakerefs[POWER_DOMAIN_NUM], wf; enum intel_display_power_domain power_domain; u64 power_domain_mask; bool active; @@ -9942,16 +9967,21 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc, intel_crtc_init_scalers(crtc, pipe_config); power_domain = POWER_DOMAIN_PIPE(crtc->pipe); - if (!intel_display_power_get_if_enabled(dev_priv, power_domain)) + wf = intel_display_power_get_if_enabled(dev_priv, power_domain); + if (!wf) return false; + + wakerefs[power_domain] = wf; power_domain_mask = BIT_ULL(power_domain); pipe_config->shared_dpll = NULL; - active = hsw_get_transcoder_state(crtc, pipe_config, &power_domain_mask); + active = hsw_get_transcoder_state(crtc, pipe_config, + &power_domain_mask, wakerefs); if (IS_GEN9_LP(dev_priv) && - bxt_get_dsi_transcoder_state(crtc, pipe_config, &power_domain_mask)) { + bxt_get_dsi_transcoder_state(crtc, pipe_config, + &power_domain_mask, wakerefs)) { WARN_ON(active); active = true; } @@ -9985,8 +10015,11 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc, } power_domain = POWER_DOMAIN_PIPE_PANEL_FITTER(crtc->pipe); - if (intel_display_power_get_if_enabled(dev_priv, power_domain)) { - WARN_ON(power_domain_mask & BIT_ULL(power_domain)); + WARN_ON(power_domain_mask & BIT_ULL(power_domain)); + + wf = intel_display_power_get_if_enabled(dev_priv, power_domain); + if (wf) { + wakerefs[power_domain] = wf; power_domain_mask |= BIT_ULL(power_domain); if (INTEL_GEN(dev_priv) >= 9) @@ -10018,7 +10051,8 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc, out: for_each_power_domain(power_domain, power_domain_mask) - intel_display_power_put_unchecked(dev_priv, power_domain); + intel_display_power_put(dev_priv, + power_domain, wakerefs[power_domain]); return active; } @@ -12990,10 +13024,16 @@ static int intel_modeset_checks(struct drm_atomic_state *state) return -EINVAL; } + /* keep the current setting */ + if (!intel_state->cdclk.force_min_cdclk_changed) + intel_state->cdclk.force_min_cdclk = + dev_priv->cdclk.force_min_cdclk; + intel_state->modeset = true; intel_state->active_crtcs = dev_priv->active_crtcs; intel_state->cdclk.logical = dev_priv->cdclk.logical; intel_state->cdclk.actual = dev_priv->cdclk.actual; + intel_state->cdclk.pipe = INVALID_PIPE; for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { if (new_crtc_state->active) @@ -13013,6 +13053,8 @@ static int intel_modeset_checks(struct drm_atomic_state *state) * adjusted_mode bits in the crtc directly. */ if (dev_priv->display.modeset_calc_cdclk) { + enum pipe pipe; + ret = dev_priv->display.modeset_calc_cdclk(state); if (ret < 0) return ret; @@ -13029,12 +13071,36 @@ static int intel_modeset_checks(struct drm_atomic_state *state) return ret; } + if (is_power_of_2(intel_state->active_crtcs)) { + struct drm_crtc *crtc; + struct drm_crtc_state *crtc_state; + + pipe = ilog2(intel_state->active_crtcs); + crtc = &intel_get_crtc_for_pipe(dev_priv, pipe)->base; + crtc_state = drm_atomic_get_new_crtc_state(state, crtc); + if (crtc_state && needs_modeset(crtc_state)) + pipe = INVALID_PIPE; + } else { + pipe = INVALID_PIPE; + } + /* All pipes must be switched off while we change the cdclk. */ - if (intel_cdclk_needs_modeset(&dev_priv->cdclk.actual, - &intel_state->cdclk.actual)) { + if (pipe != INVALID_PIPE && + intel_cdclk_needs_cd2x_update(dev_priv, + &dev_priv->cdclk.actual, + &intel_state->cdclk.actual)) { + ret = intel_lock_all_pipes(state); + if (ret < 0) + return ret; + + intel_state->cdclk.pipe = pipe; + } else if (intel_cdclk_needs_modeset(&dev_priv->cdclk.actual, + &intel_state->cdclk.actual)) { ret = intel_modeset_all_pipes(state); if (ret < 0) return ret; + + intel_state->cdclk.pipe = INVALID_PIPE; } DRM_DEBUG_KMS("New cdclk calculated to be logical %u kHz, actual %u kHz\n", @@ -13043,8 +13109,6 @@ static int intel_modeset_checks(struct drm_atomic_state *state) DRM_DEBUG_KMS("New voltage level calculated to be logical %u, actual %u\n", intel_state->cdclk.logical.voltage_level, intel_state->cdclk.actual.voltage_level); - } else { - to_intel_atomic_state(state)->cdclk.logical = dev_priv->cdclk.logical; } intel_modeset_clear_plls(state); @@ -13085,7 +13149,7 @@ static int intel_atomic_check(struct drm_device *dev, struct drm_crtc *crtc; struct drm_crtc_state *old_crtc_state, *crtc_state; int ret, i; - bool any_ms = false; + bool any_ms = intel_state->cdclk.force_min_cdclk_changed; /* Catch I915_MODE_FLAG_INHERITED */ for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, @@ -13210,14 +13274,14 @@ static void intel_update_crtc(struct drm_crtc *crtc, else if (new_plane_state) intel_fbc_enable(intel_crtc, pipe_config, new_plane_state); - intel_begin_crtc_commit(crtc, old_crtc_state); + intel_begin_crtc_commit(to_intel_atomic_state(state), intel_crtc); if (INTEL_GEN(dev_priv) >= 9) skl_update_planes_on_crtc(to_intel_atomic_state(state), intel_crtc); else i9xx_update_planes_on_crtc(to_intel_atomic_state(state), intel_crtc); - intel_finish_crtc_commit(crtc, old_crtc_state); + intel_finish_crtc_commit(to_intel_atomic_state(state), intel_crtc); } static void intel_update_crtcs(struct drm_atomic_state *state) @@ -13445,7 +13509,10 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state) if (intel_state->modeset) { drm_atomic_helper_update_legacy_modeset_state(state->dev, state); - intel_set_cdclk(dev_priv, &dev_priv->cdclk.actual); + intel_set_cdclk_pre_plane_update(dev_priv, + &intel_state->cdclk.actual, + &dev_priv->cdclk.actual, + intel_state->cdclk.pipe); /* * SKL workaround: bspec recommends we disable the SAGV when we @@ -13474,6 +13541,12 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state) /* Now enable the clocks, plane, pipe, and connectors that we set up. */ dev_priv->display.update_crtcs(state); + if (intel_state->modeset) + intel_set_cdclk_post_plane_update(dev_priv, + &intel_state->cdclk.actual, + &dev_priv->cdclk.actual, + intel_state->cdclk.pipe); + /* FIXME: We should call drm_atomic_helper_commit_hw_done() here * already, but still need the state for the delayed optimization. To * fix this: @@ -13675,8 +13748,10 @@ static int intel_atomic_commit(struct drm_device *dev, intel_state->min_voltage_level, sizeof(intel_state->min_voltage_level)); dev_priv->active_crtcs = intel_state->active_crtcs; - dev_priv->cdclk.logical = intel_state->cdclk.logical; - dev_priv->cdclk.actual = intel_state->cdclk.actual; + dev_priv->cdclk.force_min_cdclk = + intel_state->cdclk.force_min_cdclk; + + intel_cdclk_swap_state(intel_state); } drm_atomic_state_get(state); @@ -13996,39 +14071,35 @@ skl_max_scale(const struct intel_crtc_state *crtc_state, return max_scale; } -static void intel_begin_crtc_commit(struct drm_crtc *crtc, - struct drm_crtc_state *old_crtc_state) +static void intel_begin_crtc_commit(struct intel_atomic_state *state, + struct intel_crtc *crtc) { - struct drm_device *dev = crtc->dev; - struct drm_i915_private *dev_priv = to_i915(dev); - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - struct intel_crtc_state *old_intel_cstate = - to_intel_crtc_state(old_crtc_state); - struct intel_atomic_state *old_intel_state = - to_intel_atomic_state(old_crtc_state->state); - struct intel_crtc_state *intel_cstate = - intel_atomic_get_new_crtc_state(old_intel_state, intel_crtc); - bool modeset = needs_modeset(&intel_cstate->base); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + struct intel_crtc_state *old_crtc_state = + intel_atomic_get_old_crtc_state(state, crtc); + struct intel_crtc_state *new_crtc_state = + intel_atomic_get_new_crtc_state(state, crtc); + bool modeset = needs_modeset(&new_crtc_state->base); /* Perform vblank evasion around commit operation */ - intel_pipe_update_start(intel_cstate); + intel_pipe_update_start(new_crtc_state); if (modeset) goto out; - if (intel_cstate->base.color_mgmt_changed || - intel_cstate->update_pipe) - intel_color_commit(intel_cstate); + if (new_crtc_state->base.color_mgmt_changed || + new_crtc_state->update_pipe) + intel_color_commit(new_crtc_state); - if (intel_cstate->update_pipe) - intel_update_pipe_config(old_intel_cstate, intel_cstate); + if (new_crtc_state->update_pipe) + intel_update_pipe_config(old_crtc_state, new_crtc_state); else if (INTEL_GEN(dev_priv) >= 9) - skl_detach_scalers(intel_cstate); + skl_detach_scalers(new_crtc_state); out: if (dev_priv->display.atomic_update_watermarks) - dev_priv->display.atomic_update_watermarks(old_intel_state, - intel_cstate); + dev_priv->display.atomic_update_watermarks(state, + new_crtc_state); } void intel_crtc_arm_fifo_underrun(struct intel_crtc *crtc, @@ -14047,21 +14118,20 @@ void intel_crtc_arm_fifo_underrun(struct intel_crtc *crtc, } } -static void intel_finish_crtc_commit(struct drm_crtc *crtc, - struct drm_crtc_state *old_crtc_state) +static void intel_finish_crtc_commit(struct intel_atomic_state *state, + struct intel_crtc *crtc) { - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - struct intel_atomic_state *old_intel_state = - to_intel_atomic_state(old_crtc_state->state); + struct intel_crtc_state *old_crtc_state = + intel_atomic_get_old_crtc_state(state, crtc); struct intel_crtc_state *new_crtc_state = - intel_atomic_get_new_crtc_state(old_intel_state, intel_crtc); + intel_atomic_get_new_crtc_state(state, crtc); intel_pipe_update_end(new_crtc_state); if (new_crtc_state->update_pipe && !needs_modeset(&new_crtc_state->base) && - old_crtc_state->mode.private_flags & I915_MODE_FLAG_INHERITED) - intel_crtc_arm_fifo_underrun(intel_crtc, new_crtc_state); + old_crtc_state->base.mode.private_flags & I915_MODE_FLAG_INHERITED) + intel_crtc_arm_fifo_underrun(crtc, new_crtc_state); } /** @@ -16211,8 +16281,6 @@ get_encoder_power_domains(struct drm_i915_private *dev_priv) struct intel_encoder *encoder; for_each_intel_encoder(&dev_priv->drm, encoder) { - u64 get_domains; - enum intel_display_power_domain domain; struct intel_crtc_state *crtc_state; if (!encoder->get_power_domains) @@ -16226,9 +16294,7 @@ get_encoder_power_domains(struct drm_i915_private *dev_priv) continue; crtc_state = to_intel_crtc_state(encoder->base.crtc->state); - get_domains = encoder->get_power_domains(encoder, crtc_state); - for_each_power_domain(domain, get_domains) - intel_display_power_get(dev_priv, domain); + encoder->get_power_domains(encoder, crtc_state); } } diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 72c49070ed14..560274d1c50b 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -25,22 +25,34 @@ * */ -#include <linux/i2c.h> -#include <linux/slab.h> #include <linux/export.h> -#include <linux/types.h> +#include <linux/i2c.h> #include <linux/notifier.h> #include <linux/reboot.h> +#include <linux/slab.h> +#include <linux/types.h> #include <asm/byteorder.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_hdcp.h> #include <drm/drm_probe_helper.h> -#include "intel_drv.h" #include <drm/i915_drm.h> + #include "i915_drv.h" +#include "intel_audio.h" +#include "intel_connector.h" +#include "intel_ddi.h" +#include "intel_dp.h" +#include "intel_drv.h" +#include "intel_hdcp.h" +#include "intel_hdmi.h" +#include "intel_lspcon.h" +#include "intel_lvds.h" +#include "intel_panel.h" +#include "intel_psr.h" #define DP_DPRX_ESI_LEN 14 @@ -1856,42 +1868,6 @@ intel_dp_compute_link_config_wide(struct intel_dp *intel_dp, return -EINVAL; } -/* Optimize link config in order: max bpp, min lanes, min clock */ -static int -intel_dp_compute_link_config_fast(struct intel_dp *intel_dp, - struct intel_crtc_state *pipe_config, - const struct link_config_limits *limits) -{ - struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode; - int bpp, clock, lane_count; - int mode_rate, link_clock, link_avail; - - for (bpp = limits->max_bpp; bpp >= limits->min_bpp; bpp -= 2 * 3) { - mode_rate = intel_dp_link_required(adjusted_mode->crtc_clock, - bpp); - - for (lane_count = limits->min_lane_count; - lane_count <= limits->max_lane_count; - lane_count <<= 1) { - for (clock = limits->min_clock; clock <= limits->max_clock; clock++) { - link_clock = intel_dp->common_rates[clock]; - link_avail = intel_dp_max_data_rate(link_clock, - lane_count); - - if (mode_rate <= link_avail) { - pipe_config->lane_count = lane_count; - pipe_config->pipe_bpp = bpp; - pipe_config->port_clock = link_clock; - - return 0; - } - } - } - } - - return -EINVAL; -} - static int intel_dp_dsc_compute_bpp(struct intel_dp *intel_dp, u8 dsc_max_bpc) { int i, num_bpc; @@ -1919,6 +1895,9 @@ static int intel_dp_dsc_compute_config(struct intel_dp *intel_dp, int pipe_bpp; int ret; + pipe_config->fec_enable = !intel_dp_is_edp(intel_dp) && + intel_dp_supports_fec(intel_dp, pipe_config); + if (!intel_dp_supports_dsc(intel_dp, pipe_config)) return -EINVAL; @@ -2002,6 +1981,14 @@ static int intel_dp_dsc_compute_config(struct intel_dp *intel_dp, return 0; } +int intel_dp_min_bpp(const struct intel_crtc_state *crtc_state) +{ + if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_RGB) + return 6 * 3; + else + return 8 * 3; +} + static int intel_dp_compute_link_config(struct intel_encoder *encoder, struct intel_crtc_state *pipe_config, @@ -2025,18 +2012,16 @@ intel_dp_compute_link_config(struct intel_encoder *encoder, limits.min_lane_count = 1; limits.max_lane_count = intel_dp_max_lane_count(intel_dp); - limits.min_bpp = 6 * 3; + limits.min_bpp = intel_dp_min_bpp(pipe_config); limits.max_bpp = intel_dp_compute_bpp(intel_dp, pipe_config); - if (intel_dp_is_edp(intel_dp) && intel_dp->edp_dpcd[0] < DP_EDP_14) { + if (intel_dp_is_edp(intel_dp)) { /* * Use the maximum clock and number of lanes the eDP panel - * advertizes being capable of. The eDP 1.3 and earlier panels - * are generally designed to support only a single clock and - * lane configuration, and typically these values correspond to - * the native resolution of the panel. With eDP 1.4 rate select - * and DSC, this is decreasingly the case, and we need to be - * able to select less than maximum link config. + * advertizes being capable of. The panels are generally + * designed to support only a single clock and lane + * configuration, and typically these values correspond to the + * native resolution of the panel. */ limits.min_lane_count = limits.max_lane_count; limits.min_clock = limits.max_clock; @@ -2050,22 +2035,11 @@ intel_dp_compute_link_config(struct intel_encoder *encoder, intel_dp->common_rates[limits.max_clock], limits.max_bpp, adjusted_mode->crtc_clock); - if (intel_dp_is_edp(intel_dp)) - /* - * Optimize for fast and narrow. eDP 1.3 section 3.3 and eDP 1.4 - * section A.1: "It is recommended that the minimum number of - * lanes be used, using the minimum link rate allowed for that - * lane configuration." - * - * Note that we use the max clock and lane count for eDP 1.3 and - * earlier, and fast vs. wide is irrelevant. - */ - ret = intel_dp_compute_link_config_fast(intel_dp, pipe_config, - &limits); - else - /* Optimize for slow and wide. */ - ret = intel_dp_compute_link_config_wide(intel_dp, pipe_config, - &limits); + /* + * Optimize for slow and wide. This is the place to add alternative + * optimization policy. + */ + ret = intel_dp_compute_link_config_wide(intel_dp, pipe_config, &limits); /* enable compression if the mode doesn't fit available BW */ DRM_DEBUG_KMS("Force DSC en = %d\n", intel_dp->force_dsc_en); @@ -2140,7 +2114,7 @@ intel_dp_compute_config(struct intel_encoder *encoder, to_intel_digital_connector_state(conn_state); bool constant_n = drm_dp_has_quirk(&intel_dp->desc, DP_DPCD_QUIRK_CONSTANT_N); - int ret; + int ret, output_bpp; if (HAS_PCH_SPLIT(dev_priv) && !HAS_DDI(dev_priv) && port != PORT_A) pipe_config->has_pch_encoder = true; @@ -2185,9 +2159,6 @@ intel_dp_compute_config(struct intel_encoder *encoder, if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK) return -EINVAL; - pipe_config->fec_enable = !intel_dp_is_edp(intel_dp) && - intel_dp_supports_fec(intel_dp, pipe_config); - ret = intel_dp_compute_link_config(encoder, pipe_config, conn_state); if (ret < 0) return ret; @@ -2195,25 +2166,22 @@ intel_dp_compute_config(struct intel_encoder *encoder, pipe_config->limited_color_range = intel_dp_limited_color_range(pipe_config, conn_state); - if (!pipe_config->dsc_params.compression_enable) - intel_link_compute_m_n(pipe_config->pipe_bpp, - pipe_config->lane_count, - adjusted_mode->crtc_clock, - pipe_config->port_clock, - &pipe_config->dp_m_n, - constant_n); + if (pipe_config->dsc_params.compression_enable) + output_bpp = pipe_config->dsc_params.compressed_bpp; else - intel_link_compute_m_n(pipe_config->dsc_params.compressed_bpp, - pipe_config->lane_count, - adjusted_mode->crtc_clock, - pipe_config->port_clock, - &pipe_config->dp_m_n, - constant_n); + output_bpp = pipe_config->pipe_bpp; + + intel_link_compute_m_n(output_bpp, + pipe_config->lane_count, + adjusted_mode->crtc_clock, + pipe_config->port_clock, + &pipe_config->dp_m_n, + constant_n); if (intel_connector->panel.downclock_mode != NULL && dev_priv->drrs.type == SEAMLESS_DRRS_SUPPORT) { pipe_config->has_drrs = true; - intel_link_compute_m_n(pipe_config->pipe_bpp, + intel_link_compute_m_n(output_bpp, pipe_config->lane_count, intel_connector->panel.downclock_mode->clock, pipe_config->port_clock, diff --git a/drivers/gpu/drm/i915/intel_dp.h b/drivers/gpu/drm/i915/intel_dp.h new file mode 100644 index 000000000000..5e9e8d13de6e --- /dev/null +++ b/drivers/gpu/drm/i915/intel_dp.h @@ -0,0 +1,122 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_DP_H__ +#define __INTEL_DP_H__ + +#include <linux/types.h> + +#include <drm/i915_drm.h> + +#include "i915_reg.h" + +enum pipe; +struct drm_connector_state; +struct drm_encoder; +struct drm_i915_private; +struct drm_modeset_acquire_ctx; +struct intel_connector; +struct intel_crtc_state; +struct intel_digital_port; +struct intel_dp; +struct intel_encoder; + +struct link_config_limits { + int min_clock, max_clock; + int min_lane_count, max_lane_count; + int min_bpp, max_bpp; +}; + +void intel_dp_adjust_compliance_config(struct intel_dp *intel_dp, + struct intel_crtc_state *pipe_config, + struct link_config_limits *limits); +bool intel_dp_limited_color_range(const struct intel_crtc_state *crtc_state, + const struct drm_connector_state *conn_state); +int intel_dp_min_bpp(const struct intel_crtc_state *crtc_state); +bool intel_dp_port_enabled(struct drm_i915_private *dev_priv, + i915_reg_t dp_reg, enum port port, + enum pipe *pipe); +bool intel_dp_init(struct drm_i915_private *dev_priv, i915_reg_t output_reg, + enum port port); +bool intel_dp_init_connector(struct intel_digital_port *intel_dig_port, + struct intel_connector *intel_connector); +void intel_dp_set_link_params(struct intel_dp *intel_dp, + int link_rate, u8 lane_count, + bool link_mst); +int intel_dp_get_link_train_fallback_values(struct intel_dp *intel_dp, + int link_rate, u8 lane_count); +int intel_dp_retrain_link(struct intel_encoder *encoder, + struct drm_modeset_acquire_ctx *ctx); +void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode); +void intel_dp_sink_set_decompression_state(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state, + bool enable); +void intel_dp_encoder_reset(struct drm_encoder *encoder); +void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder); +void intel_dp_encoder_flush_work(struct drm_encoder *encoder); +int intel_dp_compute_config(struct intel_encoder *encoder, + struct intel_crtc_state *pipe_config, + struct drm_connector_state *conn_state); +bool intel_dp_is_edp(struct intel_dp *intel_dp); +bool intel_dp_is_port_edp(struct drm_i915_private *dev_priv, enum port port); +enum irqreturn intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, + bool long_hpd); +void intel_edp_backlight_on(const struct intel_crtc_state *crtc_state, + const struct drm_connector_state *conn_state); +void intel_edp_backlight_off(const struct drm_connector_state *conn_state); +void intel_edp_panel_vdd_on(struct intel_dp *intel_dp); +void intel_edp_panel_on(struct intel_dp *intel_dp); +void intel_edp_panel_off(struct intel_dp *intel_dp); +void intel_dp_mst_suspend(struct drm_i915_private *dev_priv); +void intel_dp_mst_resume(struct drm_i915_private *dev_priv); +int intel_dp_max_link_rate(struct intel_dp *intel_dp); +int intel_dp_max_lane_count(struct intel_dp *intel_dp); +int intel_dp_rate_select(struct intel_dp *intel_dp, int rate); +void intel_power_sequencer_reset(struct drm_i915_private *dev_priv); +u32 intel_dp_pack_aux(const u8 *src, int src_bytes); + +void intel_edp_drrs_enable(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state); +void intel_edp_drrs_disable(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state); +void intel_edp_drrs_invalidate(struct drm_i915_private *dev_priv, + unsigned int frontbuffer_bits); +void intel_edp_drrs_flush(struct drm_i915_private *dev_priv, + unsigned int frontbuffer_bits); + +void +intel_dp_program_link_training_pattern(struct intel_dp *intel_dp, + u8 dp_train_pat); +void +intel_dp_set_signal_levels(struct intel_dp *intel_dp); +void intel_dp_set_idle_link_train(struct intel_dp *intel_dp); +u8 +intel_dp_voltage_max(struct intel_dp *intel_dp); +u8 +intel_dp_pre_emphasis_max(struct intel_dp *intel_dp, u8 voltage_swing); +void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock, + u8 *link_bw, u8 *rate_select); +bool intel_dp_source_supports_hbr2(struct intel_dp *intel_dp); +bool intel_dp_source_supports_hbr3(struct intel_dp *intel_dp); +bool +intel_dp_get_link_status(struct intel_dp *intel_dp, u8 *link_status); +u16 intel_dp_dsc_get_output_bpp(int link_clock, u8 lane_count, + int mode_clock, int mode_hdisplay); +u8 intel_dp_dsc_get_slice_count(struct intel_dp *intel_dp, int mode_clock, + int mode_hdisplay); + +bool intel_dp_read_dpcd(struct intel_dp *intel_dp); +int intel_dp_link_required(int pixel_clock, int bpp); +int intel_dp_max_data_rate(int max_link_clock, int max_lanes); +bool intel_digital_port_connected(struct intel_encoder *encoder); +void icl_tc_phy_disconnect(struct drm_i915_private *dev_priv, + struct intel_digital_port *dig_port); + +static inline unsigned int intel_dp_unused_lane_mask(int lane_count) +{ + return ~((1 << lane_count) - 1) & 0xf; +} + +#endif /* __INTEL_DP_H__ */ diff --git a/drivers/gpu/drm/i915/intel_dp_link_training.c b/drivers/gpu/drm/i915/intel_dp_link_training.c index b59c87daa4f7..54b069333e2f 100644 --- a/drivers/gpu/drm/i915/intel_dp_link_training.c +++ b/drivers/gpu/drm/i915/intel_dp_link_training.c @@ -21,6 +21,7 @@ * IN THE SOFTWARE. */ +#include "intel_dp.h" #include "intel_drv.h" static void diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c index 19d81cef2ab6..8839eaea8371 100644 --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -23,12 +23,17 @@ * */ -#include "i915_drv.h" -#include "intel_drv.h" #include <drm/drm_atomic_helper.h> #include <drm/drm_edid.h> #include <drm/drm_probe_helper.h> +#include "i915_drv.h" +#include "intel_audio.h" +#include "intel_connector.h" +#include "intel_ddi.h" +#include "intel_dp.h" +#include "intel_drv.h" + static int intel_dp_mst_compute_link_config(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, struct drm_connector_state *conn_state, @@ -119,7 +124,7 @@ static int intel_dp_mst_compute_config(struct intel_encoder *encoder, limits.min_lane_count = limits.max_lane_count = intel_dp_max_lane_count(intel_dp); - limits.min_bpp = 6 * 3; + limits.min_bpp = intel_dp_min_bpp(pipe_config); limits.max_bpp = pipe_config->pipe_bpp; intel_dp_adjust_compliance_config(intel_dp, pipe_config, &limits); diff --git a/drivers/gpu/drm/i915/intel_dpio_phy.c b/drivers/gpu/drm/i915/intel_dpio_phy.c index db295c77ff0d..ab4ac7158b79 100644 --- a/drivers/gpu/drm/i915/intel_dpio_phy.c +++ b/drivers/gpu/drm/i915/intel_dpio_phy.c @@ -21,6 +21,7 @@ * DEALINGS IN THE SOFTWARE. */ +#include "intel_dp.h" #include "intel_drv.h" /** diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index f8c7b291fdc3..a38b9cff5cd0 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -27,23 +27,24 @@ #include <linux/async.h> #include <linux/i2c.h> -#include <linux/hdmi.h> #include <linux/sched/clock.h> #include <linux/stackdepot.h> -#include <drm/i915_drm.h> -#include "i915_drv.h" + +#include <drm/drm_atomic.h> #include <drm/drm_crtc.h> -#include <drm/drm_encoder.h> -#include <drm/drm_fb_helper.h> #include <drm/drm_dp_dual_mode_helper.h> #include <drm/drm_dp_mst_helper.h> +#include <drm/drm_encoder.h> +#include <drm/drm_fb_helper.h> #include <drm/drm_probe_helper.h> #include <drm/drm_rect.h> #include <drm/drm_vblank.h> -#include <drm/drm_atomic.h> +#include <drm/i915_drm.h> #include <drm/i915_mei_hdcp_interface.h> #include <media/cec-notifier.h> +#include "i915_drv.h" + struct drm_printer; /** @@ -271,10 +272,12 @@ struct intel_encoder { * be set correctly before calling this function. */ void (*get_config)(struct intel_encoder *, struct intel_crtc_state *pipe_config); - /* Returns a mask of power domains that need to be referenced as part - * of the hardware state readout code. */ - u64 (*get_power_domains)(struct intel_encoder *encoder, - struct intel_crtc_state *crtc_state); + /* + * Acquires the power domains needed for an active encoder during + * hardware state readout. + */ + void (*get_power_domains)(struct intel_encoder *encoder, + struct intel_crtc_state *crtc_state); /* * Called during system suspend after all pending requests for the * encoder are flushed (for example for DP AUX transactions) and @@ -556,6 +559,11 @@ struct intel_atomic_state { * state only when all crtc's are DPMS off. */ struct intel_cdclk_state actual; + + int force_min_cdclk; + bool force_min_cdclk_changed; + /* pipe to which cd2x update is synchronized */ + enum pipe pipe; } cdclk; bool dpll_set, modeset; @@ -1595,6 +1603,7 @@ void gen11_reset_rps_interrupts(struct drm_i915_private *dev_priv); void gen6_reset_rps_interrupts(struct drm_i915_private *dev_priv); void gen6_enable_rps_interrupts(struct drm_i915_private *dev_priv); void gen6_disable_rps_interrupts(struct drm_i915_private *dev_priv); +void gen6_rps_reset_ei(struct drm_i915_private *dev_priv); static inline u32 gen6_sanitize_rps_pm_mask(const struct drm_i915_private *i915, u32 mask) @@ -1622,85 +1631,8 @@ void gen9_reset_guc_interrupts(struct drm_i915_private *dev_priv); void gen9_enable_guc_interrupts(struct drm_i915_private *dev_priv); void gen9_disable_guc_interrupts(struct drm_i915_private *dev_priv); -/* intel_crt.c */ -bool intel_crt_port_enabled(struct drm_i915_private *dev_priv, - i915_reg_t adpa_reg, enum pipe *pipe); -void intel_crt_init(struct drm_i915_private *dev_priv); -void intel_crt_reset(struct drm_encoder *encoder); - -/* intel_ddi.c */ -void intel_ddi_fdi_post_disable(struct intel_encoder *intel_encoder, - const struct intel_crtc_state *old_crtc_state, - const struct drm_connector_state *old_conn_state); -void hsw_fdi_link_train(struct intel_crtc *crtc, - const struct intel_crtc_state *crtc_state); -void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port); -bool intel_ddi_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe); -void intel_ddi_enable_transcoder_func(const struct intel_crtc_state *crtc_state); -void intel_ddi_disable_transcoder_func(const struct intel_crtc_state *crtc_state); -void intel_ddi_enable_pipe_clock(const struct intel_crtc_state *crtc_state); -void intel_ddi_disable_pipe_clock(const struct intel_crtc_state *crtc_state); -void intel_ddi_set_pipe_settings(const struct intel_crtc_state *crtc_state); -void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp); -bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector); -void intel_ddi_get_config(struct intel_encoder *encoder, - struct intel_crtc_state *pipe_config); - -void intel_ddi_set_vc_payload_alloc(const struct intel_crtc_state *crtc_state, - bool state); -void intel_ddi_compute_min_voltage_level(struct drm_i915_private *dev_priv, - struct intel_crtc_state *crtc_state); -u32 bxt_signal_levels(struct intel_dp *intel_dp); -u32 ddi_signal_levels(struct intel_dp *intel_dp); -u8 intel_ddi_dp_voltage_max(struct intel_encoder *encoder); -u8 intel_ddi_dp_pre_emphasis_max(struct intel_encoder *encoder, - u8 voltage_swing); -int intel_ddi_toggle_hdcp_signalling(struct intel_encoder *intel_encoder, - bool enable); -void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder); -int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv, - struct intel_dpll_hw_state *state); - -unsigned int intel_fb_align_height(const struct drm_framebuffer *fb, - int color_plane, unsigned int height); - -/* intel_audio.c */ -void intel_init_audio_hooks(struct drm_i915_private *dev_priv); -void intel_audio_codec_enable(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state); -void intel_audio_codec_disable(struct intel_encoder *encoder, - const struct intel_crtc_state *old_crtc_state, - const struct drm_connector_state *old_conn_state); -void i915_audio_component_init(struct drm_i915_private *dev_priv); -void i915_audio_component_cleanup(struct drm_i915_private *dev_priv); -void intel_audio_init(struct drm_i915_private *dev_priv); -void intel_audio_deinit(struct drm_i915_private *dev_priv); - -/* intel_cdclk.c */ -int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_state); -void skl_init_cdclk(struct drm_i915_private *dev_priv); -void skl_uninit_cdclk(struct drm_i915_private *dev_priv); -void cnl_init_cdclk(struct drm_i915_private *dev_priv); -void cnl_uninit_cdclk(struct drm_i915_private *dev_priv); -void bxt_init_cdclk(struct drm_i915_private *dev_priv); -void bxt_uninit_cdclk(struct drm_i915_private *dev_priv); -void icl_init_cdclk(struct drm_i915_private *dev_priv); -void icl_uninit_cdclk(struct drm_i915_private *dev_priv); -void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv); -void intel_update_max_cdclk(struct drm_i915_private *dev_priv); -void intel_update_cdclk(struct drm_i915_private *dev_priv); -void intel_update_rawclk(struct drm_i915_private *dev_priv); -bool intel_cdclk_needs_modeset(const struct intel_cdclk_state *a, - const struct intel_cdclk_state *b); -bool intel_cdclk_changed(const struct intel_cdclk_state *a, - const struct intel_cdclk_state *b); -void intel_set_cdclk(struct drm_i915_private *dev_priv, - const struct intel_cdclk_state *cdclk_state); -void intel_dump_cdclk_state(const struct intel_cdclk_state *cdclk_state, - const char *context); - /* intel_display.c */ +void intel_plane_destroy(struct drm_plane *plane); void i830_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe); void i830_disable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe); enum pipe intel_crtc_pch_transcoder(struct intel_crtc *crtc); @@ -1715,6 +1647,8 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv); unsigned int intel_fb_xy_to_linear(int x, int y, const struct intel_plane_state *state, int plane); +unsigned int intel_fb_align_height(const struct drm_framebuffer *fb, + int color_plane, unsigned int height); void intel_add_fb_offsets(int *x, int *y, const struct intel_plane_state *state, int plane); unsigned int intel_rotation_info_size(const struct intel_rotation_info *rot_info); @@ -1882,117 +1816,9 @@ unsigned int i9xx_plane_max_stride(struct intel_plane *plane, u32 pixel_format, u64 modifier, unsigned int rotation); -/* intel_connector.c */ -int intel_connector_init(struct intel_connector *connector); -struct intel_connector *intel_connector_alloc(void); -void intel_connector_free(struct intel_connector *connector); -void intel_connector_destroy(struct drm_connector *connector); -int intel_connector_register(struct drm_connector *connector); -void intel_connector_unregister(struct drm_connector *connector); -void intel_connector_attach_encoder(struct intel_connector *connector, - struct intel_encoder *encoder); -bool intel_connector_get_hw_state(struct intel_connector *connector); -enum pipe intel_connector_get_pipe(struct intel_connector *connector); -int intel_connector_update_modes(struct drm_connector *connector, - struct edid *edid); -int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter); -void intel_attach_force_audio_property(struct drm_connector *connector); -void intel_attach_broadcast_rgb_property(struct drm_connector *connector); -void intel_attach_aspect_ratio_property(struct drm_connector *connector); -void intel_attach_colorspace_property(struct drm_connector *connector); - -/* intel_csr.c */ -void intel_csr_ucode_init(struct drm_i915_private *); -void intel_csr_load_program(struct drm_i915_private *); -void intel_csr_ucode_fini(struct drm_i915_private *); -void intel_csr_ucode_suspend(struct drm_i915_private *); -void intel_csr_ucode_resume(struct drm_i915_private *); - -/* intel_dp.c */ -struct link_config_limits { - int min_clock, max_clock; - int min_lane_count, max_lane_count; - int min_bpp, max_bpp; -}; -void intel_dp_adjust_compliance_config(struct intel_dp *intel_dp, - struct intel_crtc_state *pipe_config, - struct link_config_limits *limits); -bool intel_dp_limited_color_range(const struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state); -bool intel_dp_port_enabled(struct drm_i915_private *dev_priv, - i915_reg_t dp_reg, enum port port, - enum pipe *pipe); -bool intel_dp_init(struct drm_i915_private *dev_priv, i915_reg_t output_reg, - enum port port); -bool intel_dp_init_connector(struct intel_digital_port *intel_dig_port, - struct intel_connector *intel_connector); -void intel_dp_set_link_params(struct intel_dp *intel_dp, - int link_rate, u8 lane_count, - bool link_mst); -int intel_dp_get_link_train_fallback_values(struct intel_dp *intel_dp, - int link_rate, u8 lane_count); +/* intel_dp_link_training.c */ void intel_dp_start_link_train(struct intel_dp *intel_dp); void intel_dp_stop_link_train(struct intel_dp *intel_dp); -int intel_dp_retrain_link(struct intel_encoder *encoder, - struct drm_modeset_acquire_ctx *ctx); -void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode); -void intel_dp_sink_set_decompression_state(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state, - bool enable); -void intel_dp_encoder_reset(struct drm_encoder *encoder); -void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder); -void intel_dp_encoder_flush_work(struct drm_encoder *encoder); -int intel_dp_compute_config(struct intel_encoder *encoder, - struct intel_crtc_state *pipe_config, - struct drm_connector_state *conn_state); -bool intel_dp_is_edp(struct intel_dp *intel_dp); -bool intel_dp_is_port_edp(struct drm_i915_private *dev_priv, enum port port); -enum irqreturn intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, - bool long_hpd); -void intel_edp_backlight_on(const struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state); -void intel_edp_backlight_off(const struct drm_connector_state *conn_state); -void intel_edp_panel_vdd_on(struct intel_dp *intel_dp); -void intel_edp_panel_on(struct intel_dp *intel_dp); -void intel_edp_panel_off(struct intel_dp *intel_dp); -void intel_dp_mst_suspend(struct drm_i915_private *dev_priv); -void intel_dp_mst_resume(struct drm_i915_private *dev_priv); -int intel_dp_max_link_rate(struct intel_dp *intel_dp); -int intel_dp_max_lane_count(struct intel_dp *intel_dp); -int intel_dp_rate_select(struct intel_dp *intel_dp, int rate); -void intel_dp_hot_plug(struct intel_encoder *intel_encoder); -void intel_power_sequencer_reset(struct drm_i915_private *dev_priv); -u32 intel_dp_pack_aux(const u8 *src, int src_bytes); -void intel_plane_destroy(struct drm_plane *plane); -void intel_edp_drrs_enable(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state); -void intel_edp_drrs_disable(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state); -void intel_edp_drrs_invalidate(struct drm_i915_private *dev_priv, - unsigned int frontbuffer_bits); -void intel_edp_drrs_flush(struct drm_i915_private *dev_priv, - unsigned int frontbuffer_bits); - -void -intel_dp_program_link_training_pattern(struct intel_dp *intel_dp, - u8 dp_train_pat); -void -intel_dp_set_signal_levels(struct intel_dp *intel_dp); -void intel_dp_set_idle_link_train(struct intel_dp *intel_dp); -u8 -intel_dp_voltage_max(struct intel_dp *intel_dp); -u8 -intel_dp_pre_emphasis_max(struct intel_dp *intel_dp, u8 voltage_swing); -void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock, - u8 *link_bw, u8 *rate_select); -bool intel_dp_source_supports_hbr2(struct intel_dp *intel_dp); -bool intel_dp_source_supports_hbr3(struct intel_dp *intel_dp); -bool -intel_dp_get_link_status(struct intel_dp *intel_dp, u8 link_status[DP_LINK_STATUS_SIZE]); -u16 intel_dp_dsc_get_output_bpp(int link_clock, u8 lane_count, - int mode_clock, int mode_hdisplay); -u8 intel_dp_dsc_get_slice_count(struct intel_dp *intel_dp, int mode_clock, - int mode_hdisplay); /* intel_vdsc.c */ int intel_dp_compute_dsc_params(struct intel_dp *intel_dp, @@ -2000,18 +1826,6 @@ int intel_dp_compute_dsc_params(struct intel_dp *intel_dp, enum intel_display_power_domain intel_dsc_power_domain(const struct intel_crtc_state *crtc_state); -static inline unsigned int intel_dp_unused_lane_mask(int lane_count) -{ - return ~((1 << lane_count) - 1) & 0xf; -} - -bool intel_dp_read_dpcd(struct intel_dp *intel_dp); -int intel_dp_link_required(int pixel_clock, int bpp); -int intel_dp_max_data_rate(int max_link_clock, int max_lanes); -bool intel_digital_port_connected(struct intel_encoder *encoder); -void icl_tc_phy_disconnect(struct drm_i915_private *dev_priv, - struct intel_digital_port *dig_port); - /* intel_dp_aux_backlight.c */ int intel_dp_aux_init_backlight_funcs(struct intel_connector *intel_connector); @@ -2027,109 +1841,11 @@ void icl_dsi_init(struct drm_i915_private *dev_priv); /* intel_dsi_dcs_backlight.c */ int intel_dsi_dcs_init_backlight_funcs(struct intel_connector *intel_connector); -/* intel_dvo.c */ -void intel_dvo_init(struct drm_i915_private *dev_priv); /* intel_hotplug.c */ void intel_hpd_poll_init(struct drm_i915_private *dev_priv); bool intel_encoder_hotplug(struct intel_encoder *encoder, struct intel_connector *connector); -/* legacy fbdev emulation in intel_fbdev.c */ -#ifdef CONFIG_DRM_FBDEV_EMULATION -extern int intel_fbdev_init(struct drm_device *dev); -extern void intel_fbdev_initial_config_async(struct drm_device *dev); -extern void intel_fbdev_unregister(struct drm_i915_private *dev_priv); -extern void intel_fbdev_fini(struct drm_i915_private *dev_priv); -extern void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous); -extern void intel_fbdev_output_poll_changed(struct drm_device *dev); -extern void intel_fbdev_restore_mode(struct drm_device *dev); -#else -static inline int intel_fbdev_init(struct drm_device *dev) -{ - return 0; -} - -static inline void intel_fbdev_initial_config_async(struct drm_device *dev) -{ -} - -static inline void intel_fbdev_unregister(struct drm_i915_private *dev_priv) -{ -} - -static inline void intel_fbdev_fini(struct drm_i915_private *dev_priv) -{ -} - -static inline void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous) -{ -} - -static inline void intel_fbdev_output_poll_changed(struct drm_device *dev) -{ -} - -static inline void intel_fbdev_restore_mode(struct drm_device *dev) -{ -} -#endif - -/* intel_fbc.c */ -void intel_fbc_choose_crtc(struct drm_i915_private *dev_priv, - struct intel_atomic_state *state); -bool intel_fbc_is_active(struct drm_i915_private *dev_priv); -void intel_fbc_pre_update(struct intel_crtc *crtc, - struct intel_crtc_state *crtc_state, - struct intel_plane_state *plane_state); -void intel_fbc_post_update(struct intel_crtc *crtc); -void intel_fbc_init(struct drm_i915_private *dev_priv); -void intel_fbc_init_pipe_state(struct drm_i915_private *dev_priv); -void intel_fbc_enable(struct intel_crtc *crtc, - struct intel_crtc_state *crtc_state, - struct intel_plane_state *plane_state); -void intel_fbc_disable(struct intel_crtc *crtc); -void intel_fbc_global_disable(struct drm_i915_private *dev_priv); -void intel_fbc_invalidate(struct drm_i915_private *dev_priv, - unsigned int frontbuffer_bits, - enum fb_op_origin origin); -void intel_fbc_flush(struct drm_i915_private *dev_priv, - unsigned int frontbuffer_bits, enum fb_op_origin origin); -void intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv); -void intel_fbc_handle_fifo_underrun_irq(struct drm_i915_private *dev_priv); -int intel_fbc_reset_underrun(struct drm_i915_private *dev_priv); - -/* intel_hdmi.c */ -void intel_hdmi_init(struct drm_i915_private *dev_priv, i915_reg_t hdmi_reg, - enum port port); -void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port, - struct intel_connector *intel_connector); -struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder); -int intel_hdmi_compute_config(struct intel_encoder *encoder, - struct intel_crtc_state *pipe_config, - struct drm_connector_state *conn_state); -bool intel_hdmi_handle_sink_scrambling(struct intel_encoder *encoder, - struct drm_connector *connector, - bool high_tmds_clock_ratio, - bool scrambling); -void intel_dp_dual_mode_set_tmds_output(struct intel_hdmi *hdmi, bool enable); -void intel_infoframe_init(struct intel_digital_port *intel_dig_port); -u32 intel_hdmi_infoframes_enabled(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state); -u32 intel_hdmi_infoframe_enable(unsigned int type); -void intel_hdmi_read_gcp_infoframe(struct intel_encoder *encoder, - struct intel_crtc_state *crtc_state); -void intel_read_infoframe(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state, - enum hdmi_infoframe_type type, - union hdmi_infoframe *frame); - -/* intel_lvds.c */ -bool intel_lvds_port_enabled(struct drm_i915_private *dev_priv, - i915_reg_t lvds_reg, enum pipe *pipe); -void intel_lvds_init(struct drm_i915_private *dev_priv); -struct intel_encoder *intel_get_lvds_encoder(struct drm_i915_private *dev_priv); -bool intel_is_dual_link_lvds(struct drm_i915_private *dev_priv); - /* intel_overlay.c */ void intel_overlay_setup(struct drm_i915_private *dev_priv); void intel_overlay_cleanup(struct drm_i915_private *dev_priv); @@ -2140,92 +1856,6 @@ int intel_overlay_attrs_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); void intel_overlay_reset(struct drm_i915_private *dev_priv); - -/* intel_panel.c */ -int intel_panel_init(struct intel_panel *panel, - struct drm_display_mode *fixed_mode, - struct drm_display_mode *downclock_mode); -void intel_panel_fini(struct intel_panel *panel); -void intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode, - struct drm_display_mode *adjusted_mode); -void intel_pch_panel_fitting(struct intel_crtc *crtc, - struct intel_crtc_state *pipe_config, - int fitting_mode); -void intel_gmch_panel_fitting(struct intel_crtc *crtc, - struct intel_crtc_state *pipe_config, - int fitting_mode); -void intel_panel_set_backlight_acpi(const struct drm_connector_state *conn_state, - u32 level, u32 max); -int intel_panel_setup_backlight(struct drm_connector *connector, - enum pipe pipe); -void intel_panel_enable_backlight(const struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state); -void intel_panel_update_backlight(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state); -void intel_panel_disable_backlight(const struct drm_connector_state *old_conn_state); -struct drm_display_mode * -intel_panel_edid_downclock_mode(struct intel_connector *connector, - const struct drm_display_mode *fixed_mode); -struct drm_display_mode * -intel_panel_edid_fixed_mode(struct intel_connector *connector); -struct drm_display_mode * -intel_panel_vbt_fixed_mode(struct intel_connector *connector); - -#if IS_ENABLED(CONFIG_BACKLIGHT_CLASS_DEVICE) -int intel_backlight_device_register(struct intel_connector *connector); -void intel_backlight_device_unregister(struct intel_connector *connector); -#else /* CONFIG_BACKLIGHT_CLASS_DEVICE */ -static inline int intel_backlight_device_register(struct intel_connector *connector) -{ - return 0; -} -static inline void intel_backlight_device_unregister(struct intel_connector *connector) -{ -} -#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */ - -/* intel_hdcp.c */ -void intel_hdcp_atomic_check(struct drm_connector *connector, - struct drm_connector_state *old_state, - struct drm_connector_state *new_state); -int intel_hdcp_init(struct intel_connector *connector, - const struct intel_hdcp_shim *hdcp_shim); -int intel_hdcp_enable(struct intel_connector *connector); -int intel_hdcp_disable(struct intel_connector *connector); -bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port); -bool intel_hdcp_capable(struct intel_connector *connector); -void intel_hdcp_component_init(struct drm_i915_private *dev_priv); -void intel_hdcp_component_fini(struct drm_i915_private *dev_priv); -void intel_hdcp_cleanup(struct intel_connector *connector); -void intel_hdcp_handle_cp_irq(struct intel_connector *connector); - -/* intel_psr.c */ -#define CAN_PSR(dev_priv) (HAS_PSR(dev_priv) && dev_priv->psr.sink_support) -void intel_psr_init_dpcd(struct intel_dp *intel_dp); -void intel_psr_enable(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state); -void intel_psr_disable(struct intel_dp *intel_dp, - const struct intel_crtc_state *old_crtc_state); -void intel_psr_update(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state); -int intel_psr_debug_set(struct drm_i915_private *dev_priv, u64 value); -void intel_psr_invalidate(struct drm_i915_private *dev_priv, - unsigned frontbuffer_bits, - enum fb_op_origin origin); -void intel_psr_flush(struct drm_i915_private *dev_priv, - unsigned frontbuffer_bits, - enum fb_op_origin origin); -void intel_psr_init(struct drm_i915_private *dev_priv); -void intel_psr_compute_config(struct intel_dp *intel_dp, - struct intel_crtc_state *crtc_state); -void intel_psr_irq_control(struct drm_i915_private *dev_priv, u32 debug); -void intel_psr_irq_handler(struct drm_i915_private *dev_priv, u32 psr_iir); -void intel_psr_short_pulse(struct intel_dp *intel_dp); -int intel_psr_wait_for_idle(const struct intel_crtc_state *new_crtc_state, - u32 *out_value); -bool intel_psr_enabled(struct intel_dp *intel_dp); - /* intel_quirks.c */ void intel_init_quirks(struct drm_i915_private *dev_priv); @@ -2375,102 +2005,6 @@ void chv_phy_powergate_lanes(struct intel_encoder *encoder, bool chv_phy_powergate_ch(struct drm_i915_private *dev_priv, enum dpio_phy phy, enum dpio_channel ch, bool override); - -/* intel_pm.c */ -void intel_init_clock_gating(struct drm_i915_private *dev_priv); -void intel_suspend_hw(struct drm_i915_private *dev_priv); -int ilk_wm_max_level(const struct drm_i915_private *dev_priv); -void intel_update_watermarks(struct intel_crtc *crtc); -void intel_init_pm(struct drm_i915_private *dev_priv); -void intel_init_clock_gating_hooks(struct drm_i915_private *dev_priv); -void intel_pm_setup(struct drm_i915_private *dev_priv); -void intel_gpu_ips_init(struct drm_i915_private *dev_priv); -void intel_gpu_ips_teardown(void); -void intel_init_gt_powersave(struct drm_i915_private *dev_priv); -void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv); -void intel_sanitize_gt_powersave(struct drm_i915_private *dev_priv); -void intel_enable_gt_powersave(struct drm_i915_private *dev_priv); -void intel_disable_gt_powersave(struct drm_i915_private *dev_priv); -void gen6_rps_busy(struct drm_i915_private *dev_priv); -void gen6_rps_reset_ei(struct drm_i915_private *dev_priv); -void gen6_rps_idle(struct drm_i915_private *dev_priv); -void gen6_rps_boost(struct i915_request *rq); -void g4x_wm_get_hw_state(struct drm_i915_private *dev_priv); -void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv); -void ilk_wm_get_hw_state(struct drm_i915_private *dev_priv); -void skl_wm_get_hw_state(struct drm_i915_private *dev_priv); -void skl_pipe_ddb_get_hw_state(struct intel_crtc *crtc, - struct skl_ddb_entry *ddb_y, - struct skl_ddb_entry *ddb_uv); -void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, - struct skl_ddb_allocation *ddb /* out */); -void skl_pipe_wm_get_hw_state(struct intel_crtc *crtc, - struct skl_pipe_wm *out); -void g4x_wm_sanitize(struct drm_i915_private *dev_priv); -void vlv_wm_sanitize(struct drm_i915_private *dev_priv); -bool intel_can_enable_sagv(struct drm_atomic_state *state); -int intel_enable_sagv(struct drm_i915_private *dev_priv); -int intel_disable_sagv(struct drm_i915_private *dev_priv); -bool skl_wm_level_equals(const struct skl_wm_level *l1, - const struct skl_wm_level *l2); -bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb, - const struct skl_ddb_entry entries[], - int num_entries, int ignore_idx); -void skl_write_plane_wm(struct intel_plane *plane, - const struct intel_crtc_state *crtc_state); -void skl_write_cursor_wm(struct intel_plane *plane, - const struct intel_crtc_state *crtc_state); -bool ilk_disable_lp_wm(struct drm_device *dev); -int skl_check_pipe_max_pixel_rate(struct intel_crtc *intel_crtc, - struct intel_crtc_state *cstate); -void intel_init_ipc(struct drm_i915_private *dev_priv); -void intel_enable_ipc(struct drm_i915_private *dev_priv); - -/* intel_sdvo.c */ -bool intel_sdvo_port_enabled(struct drm_i915_private *dev_priv, - i915_reg_t sdvo_reg, enum pipe *pipe); -bool intel_sdvo_init(struct drm_i915_private *dev_priv, - i915_reg_t reg, enum port port); - - -/* intel_sprite.c */ -bool is_planar_yuv_format(u32 pixelformat); -int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode, - int usecs); -struct intel_plane *intel_sprite_plane_create(struct drm_i915_private *dev_priv, - enum pipe pipe, int plane); -int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data, - struct drm_file *file_priv); -void intel_pipe_update_start(const struct intel_crtc_state *new_crtc_state); -void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state); -int intel_plane_check_stride(const struct intel_plane_state *plane_state); -int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state); -int chv_plane_check_rotation(const struct intel_plane_state *plane_state); -struct intel_plane * -skl_universal_plane_create(struct drm_i915_private *dev_priv, - enum pipe pipe, enum plane_id plane_id); - -static inline bool icl_is_nv12_y_plane(enum plane_id id) -{ - /* Don't need to do a gen check, these planes are only available on gen11 */ - if (id == PLANE_SPRITE4 || id == PLANE_SPRITE5) - return true; - - return false; -} - -static inline bool icl_is_hdr_plane(struct drm_i915_private *dev_priv, - enum plane_id plane_id) -{ - if (INTEL_GEN(dev_priv) < 11) - return false; - - return plane_id < PLANE_SPRITE2; -} - -/* intel_tv.c */ -void intel_tv_init(struct drm_i915_private *dev_priv); - /* intel_atomic.c */ int intel_digital_connector_atomic_get_property(struct drm_connector *connector, const struct drm_connector_state *state, @@ -2507,76 +2041,4 @@ int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv, struct intel_crtc *intel_crtc, struct intel_crtc_state *crtc_state); -/* intel_atomic_plane.c */ -void intel_update_plane(struct intel_plane *plane, - const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state); -void intel_update_slave(struct intel_plane *plane, - const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state); -void intel_disable_plane(struct intel_plane *plane, - const struct intel_crtc_state *crtc_state); -struct intel_plane *intel_plane_alloc(void); -void intel_plane_free(struct intel_plane *plane); -struct drm_plane_state *intel_plane_duplicate_state(struct drm_plane *plane); -void intel_plane_destroy_state(struct drm_plane *plane, - struct drm_plane_state *state); -extern const struct drm_plane_helper_funcs intel_plane_helper_funcs; -void skl_update_planes_on_crtc(struct intel_atomic_state *state, - struct intel_crtc *crtc); -void i9xx_update_planes_on_crtc(struct intel_atomic_state *state, - struct intel_crtc *crtc); -int intel_plane_atomic_check_with_state(const struct intel_crtc_state *old_crtc_state, - struct intel_crtc_state *crtc_state, - const struct intel_plane_state *old_plane_state, - struct intel_plane_state *intel_state); - -/* intel_color.c */ -void intel_color_init(struct intel_crtc *crtc); -int intel_color_check(struct intel_crtc_state *crtc_state); -void intel_color_commit(const struct intel_crtc_state *crtc_state); -void intel_color_load_luts(const struct intel_crtc_state *crtc_state); - -/* intel_lspcon.c */ -bool lspcon_init(struct intel_digital_port *intel_dig_port); -void lspcon_resume(struct intel_lspcon *lspcon); -void lspcon_wait_pcon_mode(struct intel_lspcon *lspcon); -void lspcon_write_infoframe(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state, - unsigned int type, - const void *buf, ssize_t len); -void lspcon_read_infoframe(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state, - unsigned int type, - void *frame, ssize_t len); -void lspcon_set_infoframes(struct intel_encoder *encoder, - bool enable, - const struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state); -u32 lspcon_infoframes_enabled(struct intel_encoder *encoder, - const struct intel_crtc_state *pipe_config); -void lspcon_ycbcr420_config(struct drm_connector *connector, - struct intel_crtc_state *crtc_state); - -/* intel_pipe_crc.c */ -#ifdef CONFIG_DEBUG_FS -int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name); -int intel_crtc_verify_crc_source(struct drm_crtc *crtc, - const char *source_name, size_t *values_cnt); -const char *const *intel_crtc_get_crc_sources(struct drm_crtc *crtc, - size_t *count); -void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc); -void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc); -#else -#define intel_crtc_set_crc_source NULL -#define intel_crtc_verify_crc_source NULL -#define intel_crtc_get_crc_sources NULL -static inline void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc) -{ -} - -static inline void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc) -{ -} -#endif #endif /* __INTEL_DRV_H__ */ diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c index a6c82482a841..adef81c8cccb 100644 --- a/drivers/gpu/drm/i915/intel_dvo.c +++ b/drivers/gpu/drm/i915/intel_dvo.c @@ -24,14 +24,20 @@ * Authors: * Eric Anholt <eric@anholt.net> */ + #include <linux/i2c.h> #include <linux/slab.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> -#include "intel_drv.h" #include <drm/i915_drm.h> -#include "i915_drv.h" + #include "dvo.h" +#include "i915_drv.h" +#include "intel_connector.h" +#include "intel_drv.h" +#include "intel_dvo.h" +#include "intel_panel.h" #define SIL164_ADDR 0x38 #define CH7xxx_ADDR 0x76 diff --git a/drivers/gpu/drm/i915/intel_dvo.h b/drivers/gpu/drm/i915/intel_dvo.h new file mode 100644 index 000000000000..3ed0fdf8efff --- /dev/null +++ b/drivers/gpu/drm/i915/intel_dvo.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_DVO_H__ +#define __INTEL_DVO_H__ + +struct drm_i915_private; + +void intel_dvo_init(struct drm_i915_private *dev_priv); + +#endif /* __INTEL_DVO_H__ */ diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c index d0427c2e3997..eea9bec04f1b 100644 --- a/drivers/gpu/drm/i915/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/intel_engine_cs.c @@ -753,6 +753,30 @@ err_unpin: return ret; } +void intel_gt_resume(struct drm_i915_private *i915) +{ + struct intel_engine_cs *engine; + enum intel_engine_id id; + + /* + * After resume, we may need to poke into the pinned kernel + * contexts to paper over any damage caused by the sudden suspend. + * Only the kernel contexts should remain pinned over suspend, + * allowing us to fixup the user contexts on their first pin. + */ + for_each_engine(engine, i915, id) { + struct intel_context *ce; + + ce = engine->kernel_context; + if (ce) + ce->ops->reset(ce); + + ce = engine->preempt_context; + if (ce) + ce->ops->reset(ce); + } +} + /** * intel_engines_cleanup_common - cleans up the engine state created by * the common initiailizers. @@ -1381,40 +1405,33 @@ static void intel_engine_print_registers(const struct intel_engine_cs *engine, if (HAS_EXECLISTS(dev_priv)) { const u32 *hws = &engine->status_page.addr[I915_HWS_CSB_BUF0_INDEX]; + const u8 num_entries = execlists->csb_size; unsigned int idx; u8 read, write; - drm_printf(m, "\tExeclist status: 0x%08x %08x\n", + drm_printf(m, "\tExeclist status: 0x%08x %08x, entries %u\n", ENGINE_READ(engine, RING_EXECLIST_STATUS_LO), - ENGINE_READ(engine, RING_EXECLIST_STATUS_HI)); + ENGINE_READ(engine, RING_EXECLIST_STATUS_HI), + num_entries); read = execlists->csb_head; write = READ_ONCE(*execlists->csb_write); - drm_printf(m, "\tExeclist CSB read %d, write %d [mmio:%d], tasklet queued? %s (%s)\n", + drm_printf(m, "\tExeclist CSB read %d, write %d, tasklet queued? %s (%s)\n", read, write, - GEN8_CSB_WRITE_PTR(ENGINE_READ(engine, RING_CONTEXT_STATUS_PTR)), yesno(test_bit(TASKLET_STATE_SCHED, &engine->execlists.tasklet.state)), enableddisabled(!atomic_read(&engine->execlists.tasklet.count))); - if (read >= GEN8_CSB_ENTRIES) + if (read >= num_entries) read = 0; - if (write >= GEN8_CSB_ENTRIES) + if (write >= num_entries) write = 0; if (read > write) - write += GEN8_CSB_ENTRIES; + write += num_entries; while (read < write) { - idx = ++read % GEN8_CSB_ENTRIES; - drm_printf(m, "\tExeclist CSB[%d]: 0x%08x [mmio:0x%08x], context: %d [mmio:%d]\n", - idx, - hws[idx * 2], - ENGINE_READ_IDX(engine, - RING_CONTEXT_STATUS_BUF_LO, - idx), - hws[idx * 2 + 1], - ENGINE_READ_IDX(engine, - RING_CONTEXT_STATUS_BUF_HI, - idx)); + idx = ++read % num_entries; + drm_printf(m, "\tExeclist CSB[%d]: 0x%08x, context: %d\n", + idx, hws[idx * 2], hws[idx * 2 + 1]); } rcu_read_lock(); diff --git a/drivers/gpu/drm/i915/intel_engine_types.h b/drivers/gpu/drm/i915/intel_engine_types.h index b3249bf6a65f..1f970c76b6a6 100644 --- a/drivers/gpu/drm/i915/intel_engine_types.h +++ b/drivers/gpu/drm/i915/intel_engine_types.h @@ -13,8 +13,10 @@ #include <linux/list.h> #include <linux/types.h> +#include "i915_gem.h" +#include "i915_priolist_types.h" +#include "i915_selftest.h" #include "i915_timeline_types.h" -#include "intel_device_info.h" #include "intel_workarounds_types.h" #include "i915_gem_batch_pool.h" @@ -25,12 +27,16 @@ #define I915_CMD_HASH_ORDER 9 +struct dma_fence; struct drm_i915_reg_table; struct i915_gem_context; struct i915_request; struct i915_sched_attr; struct intel_uncore; +typedef u8 intel_engine_mask_t; +#define ALL_ENGINES ((intel_engine_mask_t)~0ul) + struct intel_hw_status_page { struct i915_vma *vma; u32 *addr; @@ -105,8 +111,9 @@ enum intel_engine_id { VCS3, #define _VCS(n) (VCS0 + (n)) VECS0, - VECS1 + VECS1, #define _VECS(n) (VECS0 + (n)) + I915_NUM_ENGINES }; struct st_preempt_hang { @@ -240,6 +247,11 @@ struct intel_engine_execlists { u32 preempt_complete_status; /** + * @csb_size: context status buffer FIFO size + */ + u8 csb_size; + + /** * @csb_head: context status buffer head */ u8 csb_head; @@ -425,6 +437,7 @@ struct intel_engine_cs { #define I915_ENGINE_SUPPORTS_STATS BIT(1) #define I915_ENGINE_HAS_PREEMPTION BIT(2) #define I915_ENGINE_HAS_SEMAPHORES BIT(3) +#define I915_ENGINE_NEEDS_BREADCRUMB_TASKLET BIT(4) unsigned int flags; /* @@ -508,6 +521,12 @@ intel_engine_has_semaphores(const struct intel_engine_cs *engine) return engine->flags & I915_ENGINE_HAS_SEMAPHORES; } +static inline bool +intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) +{ + return engine->flags & I915_ENGINE_NEEDS_BREADCRUMB_TASKLET; +} + #define instdone_slice_mask(dev_priv__) \ (IS_GEN(dev_priv__, 7) ? \ 1 : RUNTIME_INFO(dev_priv__)->sseu.slice_mask) diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c index 43fe08be3b7d..c805a0966395 100644 --- a/drivers/gpu/drm/i915/intel_fbc.c +++ b/drivers/gpu/drm/i915/intel_fbc.c @@ -40,8 +40,10 @@ #include <drm/drm_fourcc.h> -#include "intel_drv.h" #include "i915_drv.h" +#include "intel_drv.h" +#include "intel_fbc.h" +#include "intel_frontbuffer.h" static inline bool fbc_supported(struct drm_i915_private *dev_priv) { diff --git a/drivers/gpu/drm/i915/intel_fbc.h b/drivers/gpu/drm/i915/intel_fbc.h new file mode 100644 index 000000000000..50272eda8d43 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_fbc.h @@ -0,0 +1,42 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_FBC_H__ +#define __INTEL_FBC_H__ + +#include <linux/types.h> + +#include "intel_frontbuffer.h" + +struct drm_i915_private; +struct intel_atomic_state; +struct intel_crtc; +struct intel_crtc_state; +struct intel_plane_state; + +void intel_fbc_choose_crtc(struct drm_i915_private *dev_priv, + struct intel_atomic_state *state); +bool intel_fbc_is_active(struct drm_i915_private *dev_priv); +void intel_fbc_pre_update(struct intel_crtc *crtc, + struct intel_crtc_state *crtc_state, + struct intel_plane_state *plane_state); +void intel_fbc_post_update(struct intel_crtc *crtc); +void intel_fbc_init(struct drm_i915_private *dev_priv); +void intel_fbc_init_pipe_state(struct drm_i915_private *dev_priv); +void intel_fbc_enable(struct intel_crtc *crtc, + struct intel_crtc_state *crtc_state, + struct intel_plane_state *plane_state); +void intel_fbc_disable(struct intel_crtc *crtc); +void intel_fbc_global_disable(struct drm_i915_private *dev_priv); +void intel_fbc_invalidate(struct drm_i915_private *dev_priv, + unsigned int frontbuffer_bits, + enum fb_op_origin origin); +void intel_fbc_flush(struct drm_i915_private *dev_priv, + unsigned int frontbuffer_bits, enum fb_op_origin origin); +void intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv); +void intel_fbc_handle_fifo_underrun_irq(struct drm_i915_private *dev_priv); +int intel_fbc_reset_underrun(struct drm_i915_private *dev_priv); + +#endif /* __INTEL_FBC_H__ */ diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c index ef93c27e60b4..89db71996148 100644 --- a/drivers/gpu/drm/i915/intel_fbdev.c +++ b/drivers/gpu/drm/i915/intel_fbdev.c @@ -25,26 +25,27 @@ */ #include <linux/async.h> -#include <linux/module.h> -#include <linux/kernel.h> #include <linux/console.h> +#include <linux/delay.h> #include <linux/errno.h> -#include <linux/string.h> +#include <linux/init.h> +#include <linux/kernel.h> #include <linux/mm.h> -#include <linux/tty.h> +#include <linux/module.h> +#include <linux/string.h> #include <linux/sysrq.h> -#include <linux/delay.h> -#include <linux/init.h> +#include <linux/tty.h> #include <linux/vga_switcheroo.h> #include <drm/drm_crtc.h> #include <drm/drm_fb_helper.h> #include <drm/drm_fourcc.h> +#include <drm/i915_drm.h> +#include "i915_drv.h" #include "intel_drv.h" +#include "intel_fbdev.h" #include "intel_frontbuffer.h" -#include <drm/i915_drm.h> -#include "i915_drv.h" static void intel_fbdev_invalidate(struct intel_fbdev *ifbdev) { @@ -284,225 +285,7 @@ out_unlock: return ret; } -static struct drm_fb_helper_crtc * -intel_fb_helper_crtc(struct drm_fb_helper *fb_helper, struct drm_crtc *crtc) -{ - int i; - - for (i = 0; i < fb_helper->crtc_count; i++) - if (fb_helper->crtc_info[i].mode_set.crtc == crtc) - return &fb_helper->crtc_info[i]; - - return NULL; -} - -/* - * Try to read the BIOS display configuration and use it for the initial - * fb configuration. - * - * The BIOS or boot loader will generally create an initial display - * configuration for us that includes some set of active pipes and displays. - * This routine tries to figure out which pipes and connectors are active - * and stuffs them into the crtcs and modes array given to us by the - * drm_fb_helper code. - * - * The overall sequence is: - * intel_fbdev_init - from driver load - * intel_fbdev_init_bios - initialize the intel_fbdev using BIOS data - * drm_fb_helper_init - build fb helper structs - * drm_fb_helper_single_add_all_connectors - more fb helper structs - * intel_fbdev_initial_config - apply the config - * drm_fb_helper_initial_config - call ->probe then register_framebuffer() - * drm_setup_crtcs - build crtc config for fbdev - * intel_fb_initial_config - find active connectors etc - * drm_fb_helper_single_fb_probe - set up fbdev - * intelfb_create - re-use or alloc fb, build out fbdev structs - * - * Note that we don't make special consideration whether we could actually - * switch to the selected modes without a full modeset. E.g. when the display - * is in VGA mode we need to recalculate watermarks and set a new high-res - * framebuffer anyway. - */ -static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper, - struct drm_fb_helper_crtc **crtcs, - struct drm_display_mode **modes, - struct drm_fb_offset *offsets, - bool *enabled, int width, int height) -{ - struct drm_i915_private *dev_priv = to_i915(fb_helper->dev); - unsigned int count = min(fb_helper->connector_count, BITS_PER_LONG); - unsigned long conn_configured, conn_seq; - int i, j; - bool *save_enabled; - bool fallback = true, ret = true; - int num_connectors_enabled = 0; - int num_connectors_detected = 0; - struct drm_modeset_acquire_ctx ctx; - - save_enabled = kcalloc(count, sizeof(bool), GFP_KERNEL); - if (!save_enabled) - return false; - - drm_modeset_acquire_init(&ctx, 0); - - while (drm_modeset_lock_all_ctx(fb_helper->dev, &ctx) != 0) - drm_modeset_backoff(&ctx); - - memcpy(save_enabled, enabled, count); - conn_seq = GENMASK(count - 1, 0); - conn_configured = 0; -retry: - for (i = 0; i < count; i++) { - struct drm_fb_helper_connector *fb_conn; - struct drm_connector *connector; - struct drm_encoder *encoder; - struct drm_fb_helper_crtc *new_crtc; - - fb_conn = fb_helper->connector_info[i]; - connector = fb_conn->connector; - - if (conn_configured & BIT(i)) - continue; - - /* First pass, only consider tiled connectors */ - if (conn_seq == GENMASK(count - 1, 0) && !connector->has_tile) - continue; - - if (connector->status == connector_status_connected) - num_connectors_detected++; - - if (!enabled[i]) { - DRM_DEBUG_KMS("connector %s not enabled, skipping\n", - connector->name); - conn_configured |= BIT(i); - continue; - } - - if (connector->force == DRM_FORCE_OFF) { - DRM_DEBUG_KMS("connector %s is disabled by user, skipping\n", - connector->name); - enabled[i] = false; - continue; - } - - encoder = connector->state->best_encoder; - if (!encoder || WARN_ON(!connector->state->crtc)) { - if (connector->force > DRM_FORCE_OFF) - goto bail; - - DRM_DEBUG_KMS("connector %s has no encoder or crtc, skipping\n", - connector->name); - enabled[i] = false; - conn_configured |= BIT(i); - continue; - } - - num_connectors_enabled++; - - new_crtc = intel_fb_helper_crtc(fb_helper, - connector->state->crtc); - - /* - * Make sure we're not trying to drive multiple connectors - * with a single CRTC, since our cloning support may not - * match the BIOS. - */ - for (j = 0; j < count; j++) { - if (crtcs[j] == new_crtc) { - DRM_DEBUG_KMS("fallback: cloned configuration\n"); - goto bail; - } - } - - DRM_DEBUG_KMS("looking for cmdline mode on connector %s\n", - connector->name); - - /* go for command line mode first */ - modes[i] = drm_pick_cmdline_mode(fb_conn); - - /* try for preferred next */ - if (!modes[i]) { - DRM_DEBUG_KMS("looking for preferred mode on connector %s %d\n", - connector->name, connector->has_tile); - modes[i] = drm_has_preferred_mode(fb_conn, width, - height); - } - - /* No preferred mode marked by the EDID? Are there any modes? */ - if (!modes[i] && !list_empty(&connector->modes)) { - DRM_DEBUG_KMS("using first mode listed on connector %s\n", - connector->name); - modes[i] = list_first_entry(&connector->modes, - struct drm_display_mode, - head); - } - - /* last resort: use current mode */ - if (!modes[i]) { - /* - * IMPORTANT: We want to use the adjusted mode (i.e. - * after the panel fitter upscaling) as the initial - * config, not the input mode, which is what crtc->mode - * usually contains. But since our current - * code puts a mode derived from the post-pfit timings - * into crtc->mode this works out correctly. - * - * This is crtc->mode and not crtc->state->mode for the - * fastboot check to work correctly. crtc_state->mode has - * I915_MODE_FLAG_INHERITED, which we clear to force check - * state. - */ - DRM_DEBUG_KMS("looking for current mode on connector %s\n", - connector->name); - modes[i] = &connector->state->crtc->mode; - } - crtcs[i] = new_crtc; - - DRM_DEBUG_KMS("connector %s on [CRTC:%d:%s]: %dx%d%s\n", - connector->name, - connector->state->crtc->base.id, - connector->state->crtc->name, - modes[i]->hdisplay, modes[i]->vdisplay, - modes[i]->flags & DRM_MODE_FLAG_INTERLACE ? "i" :""); - - fallback = false; - conn_configured |= BIT(i); - } - - if (conn_configured != conn_seq) { /* repeat until no more are found */ - conn_seq = conn_configured; - goto retry; - } - - /* - * If the BIOS didn't enable everything it could, fall back to have the - * same user experiencing of lighting up as much as possible like the - * fbdev helper library. - */ - if (num_connectors_enabled != num_connectors_detected && - num_connectors_enabled < INTEL_INFO(dev_priv)->num_pipes) { - DRM_DEBUG_KMS("fallback: Not all outputs enabled\n"); - DRM_DEBUG_KMS("Enabled: %i, detected: %i\n", num_connectors_enabled, - num_connectors_detected); - fallback = true; - } - - if (fallback) { -bail: - DRM_DEBUG_KMS("Not using firmware configuration\n"); - memcpy(enabled, save_enabled, count); - ret = false; - } - - drm_modeset_drop_locks(&ctx); - drm_modeset_acquire_fini(&ctx); - - kfree(save_enabled); - return ret; -} - static const struct drm_fb_helper_funcs intel_fb_helper_funcs = { - .initial_config = intel_fb_initial_config, .fb_probe = intelfb_create, }; diff --git a/drivers/gpu/drm/i915/intel_fbdev.h b/drivers/gpu/drm/i915/intel_fbdev.h new file mode 100644 index 000000000000..de7c84250eb5 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_fbdev.h @@ -0,0 +1,53 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_FBDEV_H__ +#define __INTEL_FBDEV_H__ + +#include <linux/types.h> + +struct drm_device; +struct drm_i915_private; + +#ifdef CONFIG_DRM_FBDEV_EMULATION +int intel_fbdev_init(struct drm_device *dev); +void intel_fbdev_initial_config_async(struct drm_device *dev); +void intel_fbdev_unregister(struct drm_i915_private *dev_priv); +void intel_fbdev_fini(struct drm_i915_private *dev_priv); +void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous); +void intel_fbdev_output_poll_changed(struct drm_device *dev); +void intel_fbdev_restore_mode(struct drm_device *dev); +#else +static inline int intel_fbdev_init(struct drm_device *dev) +{ + return 0; +} + +static inline void intel_fbdev_initial_config_async(struct drm_device *dev) +{ +} + +static inline void intel_fbdev_unregister(struct drm_i915_private *dev_priv) +{ +} + +static inline void intel_fbdev_fini(struct drm_i915_private *dev_priv) +{ +} + +static inline void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous) +{ +} + +static inline void intel_fbdev_output_poll_changed(struct drm_device *dev) +{ +} + +static inline void intel_fbdev_restore_mode(struct drm_device *dev) +{ +} +#endif + +#endif /* __INTEL_FBDEV_H__ */ diff --git a/drivers/gpu/drm/i915/intel_fifo_underrun.c b/drivers/gpu/drm/i915/intel_fifo_underrun.c index f33de4be4b89..74c8b0528294 100644 --- a/drivers/gpu/drm/i915/intel_fifo_underrun.c +++ b/drivers/gpu/drm/i915/intel_fifo_underrun.c @@ -27,6 +27,7 @@ #include "i915_drv.h" #include "intel_drv.h" +#include "intel_fbc.h" /** * DOC: fifo underrun handling diff --git a/drivers/gpu/drm/i915/intel_frontbuffer.c b/drivers/gpu/drm/i915/intel_frontbuffer.c index 16f253deaf8d..aa34e33b6087 100644 --- a/drivers/gpu/drm/i915/intel_frontbuffer.c +++ b/drivers/gpu/drm/i915/intel_frontbuffer.c @@ -61,9 +61,12 @@ */ +#include "i915_drv.h" +#include "intel_dp.h" #include "intel_drv.h" +#include "intel_fbc.h" #include "intel_frontbuffer.h" -#include "i915_drv.h" +#include "intel_psr.h" void __intel_fb_obj_invalidate(struct drm_i915_gem_object *obj, enum fb_op_origin origin, diff --git a/drivers/gpu/drm/i915/intel_frontbuffer.h b/drivers/gpu/drm/i915/intel_frontbuffer.h index 63cd9a753a72..d5894666f658 100644 --- a/drivers/gpu/drm/i915/intel_frontbuffer.h +++ b/drivers/gpu/drm/i915/intel_frontbuffer.h @@ -24,9 +24,19 @@ #ifndef __INTEL_FRONTBUFFER_H__ #define __INTEL_FRONTBUFFER_H__ +#include "i915_gem_object.h" + struct drm_i915_private; struct drm_i915_gem_object; +enum fb_op_origin { + ORIGIN_GTT, + ORIGIN_CPU, + ORIGIN_CS, + ORIGIN_FLIP, + ORIGIN_DIRTYFB, +}; + void intel_frontbuffer_flip_prepare(struct drm_i915_private *dev_priv, unsigned frontbuffer_bits); void intel_frontbuffer_flip_complete(struct drm_i915_private *dev_priv, diff --git a/drivers/gpu/drm/i915/intel_guc_submission.c b/drivers/gpu/drm/i915/intel_guc_submission.c index c4ad73980988..37f60cb8e9e1 100644 --- a/drivers/gpu/drm/i915/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/intel_guc_submission.c @@ -567,7 +567,7 @@ static void inject_preempt_context(struct work_struct *work) preempt_work[engine->id]); struct intel_guc_client *client = guc->preempt_client; struct guc_stage_desc *stage_desc = __get_stage_desc(client); - struct intel_context *ce = intel_context_lookup(client->owner, engine); + struct intel_context *ce = engine->preempt_context; u32 data[7]; if (!ce->ring->emit) { /* recreate upon load/resume */ @@ -650,9 +650,10 @@ static void wait_for_guc_preempt_report(struct intel_engine_cs *engine) struct guc_ctx_report *report = &data->preempt_ctx_report[engine->guc_id]; - WARN_ON(wait_for_atomic(report->report_return_status == - INTEL_GUC_REPORT_STATUS_COMPLETE, - GUC_PREEMPT_POSTPROCESS_DELAY_MS)); + if (wait_for_atomic(report->report_return_status == + INTEL_GUC_REPORT_STATUS_COMPLETE, + GUC_PREEMPT_POSTPROCESS_DELAY_MS)) + DRM_ERROR("Timed out waiting for GuC preemption report\n"); /* * GuC is expecting that we're also going to clear the affected context * counter, let's also reset the return status to not depend on GuC @@ -871,6 +872,104 @@ static void guc_reset_prepare(struct intel_engine_cs *engine) flush_workqueue(engine->i915->guc.preempt_wq); } +static void guc_reset(struct intel_engine_cs *engine, bool stalled) +{ + struct intel_engine_execlists * const execlists = &engine->execlists; + struct i915_request *rq; + unsigned long flags; + + spin_lock_irqsave(&engine->timeline.lock, flags); + + execlists_cancel_port_requests(execlists); + + /* Push back any incomplete requests for replay after the reset. */ + rq = execlists_unwind_incomplete_requests(execlists); + if (!rq) + goto out_unlock; + + if (!i915_request_started(rq)) + stalled = false; + + i915_reset_request(rq, stalled); + intel_lr_context_reset(engine, rq->hw_context, rq->head, stalled); + +out_unlock: + spin_unlock_irqrestore(&engine->timeline.lock, flags); +} + +static void guc_cancel_requests(struct intel_engine_cs *engine) +{ + struct intel_engine_execlists * const execlists = &engine->execlists; + struct i915_request *rq, *rn; + struct rb_node *rb; + unsigned long flags; + + GEM_TRACE("%s\n", engine->name); + + /* + * Before we call engine->cancel_requests(), we should have exclusive + * access to the submission state. This is arranged for us by the + * caller disabling the interrupt generation, the tasklet and other + * threads that may then access the same state, giving us a free hand + * to reset state. However, we still need to let lockdep be aware that + * we know this state may be accessed in hardirq context, so we + * disable the irq around this manipulation and we want to keep + * the spinlock focused on its duties and not accidentally conflate + * coverage to the submission's irq state. (Similarly, although we + * shouldn't need to disable irq around the manipulation of the + * submission's irq state, we also wish to remind ourselves that + * it is irq state.) + */ + spin_lock_irqsave(&engine->timeline.lock, flags); + + /* Cancel the requests on the HW and clear the ELSP tracker. */ + execlists_cancel_port_requests(execlists); + + /* Mark all executing requests as skipped. */ + list_for_each_entry(rq, &engine->timeline.requests, link) { + if (!i915_request_signaled(rq)) + dma_fence_set_error(&rq->fence, -EIO); + + i915_request_mark_complete(rq); + } + + /* Flush the queued requests to the timeline list (for retiring). */ + while ((rb = rb_first_cached(&execlists->queue))) { + struct i915_priolist *p = to_priolist(rb); + int i; + + priolist_for_each_request_consume(rq, rn, p, i) { + list_del_init(&rq->sched.link); + __i915_request_submit(rq); + dma_fence_set_error(&rq->fence, -EIO); + i915_request_mark_complete(rq); + } + + rb_erase_cached(&p->node, &execlists->queue); + i915_priolist_free(p); + } + + /* Remaining _unready_ requests will be nop'ed when submitted */ + + execlists->queue_priority_hint = INT_MIN; + execlists->queue = RB_ROOT_CACHED; + GEM_BUG_ON(port_isset(execlists->port)); + + spin_unlock_irqrestore(&engine->timeline.lock, flags); +} + +static void guc_reset_finish(struct intel_engine_cs *engine) +{ + struct intel_engine_execlists * const execlists = &engine->execlists; + + if (__tasklet_enable(&execlists->tasklet)) + /* And kick in case we missed a new request submission. */ + tasklet_hi_schedule(&execlists->tasklet); + + GEM_TRACE("%s: depth->%d\n", engine->name, + atomic_read(&execlists->tasklet.count)); +} + /* * Everything below here is concerned with setup & teardown, and is * therefore not part of the somewhat time-critical batch-submission @@ -1262,10 +1361,12 @@ static void guc_interrupts_release(struct drm_i915_private *dev_priv) static void guc_submission_park(struct intel_engine_cs *engine) { intel_engine_unpin_breadcrumbs_irq(engine); + engine->flags &= ~I915_ENGINE_NEEDS_BREADCRUMB_TASKLET; } static void guc_submission_unpark(struct intel_engine_cs *engine) { + engine->flags |= I915_ENGINE_NEEDS_BREADCRUMB_TASKLET; intel_engine_pin_breadcrumbs_irq(engine); } @@ -1290,6 +1391,10 @@ static void guc_set_default_submission(struct intel_engine_cs *engine) engine->unpark = guc_submission_unpark; engine->reset.prepare = guc_reset_prepare; + engine->reset.reset = guc_reset; + engine->reset.finish = guc_reset_finish; + + engine->cancel_requests = guc_cancel_requests; engine->flags &= ~I915_ENGINE_SUPPORTS_STATS; } diff --git a/drivers/gpu/drm/i915/intel_guc_submission.h b/drivers/gpu/drm/i915/intel_guc_submission.h index 169c54568340..aa5e6749c925 100644 --- a/drivers/gpu/drm/i915/intel_guc_submission.h +++ b/drivers/gpu/drm/i915/intel_guc_submission.h @@ -29,6 +29,7 @@ #include "i915_gem.h" #include "i915_selftest.h" +#include "intel_engine_types.h" struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/intel_hangcheck.c b/drivers/gpu/drm/i915/intel_hangcheck.c index 59232df11ada..3d51ed1428d4 100644 --- a/drivers/gpu/drm/i915/intel_hangcheck.c +++ b/drivers/gpu/drm/i915/intel_hangcheck.c @@ -221,8 +221,8 @@ static void hangcheck_declare_hang(struct drm_i915_private *i915, unsigned int stuck) { struct intel_engine_cs *engine; + intel_engine_mask_t tmp; char msg[80]; - unsigned int tmp; int len; /* If some rings hung but others were still busy, only diff --git a/drivers/gpu/drm/i915/intel_hdcp.c b/drivers/gpu/drm/i915/intel_hdcp.c index 86965fa37739..99b007169c49 100644 --- a/drivers/gpu/drm/i915/intel_hdcp.c +++ b/drivers/gpu/drm/i915/intel_hdcp.c @@ -6,14 +6,16 @@ * Sean Paul <seanpaul@chromium.org> */ -#include <drm/drm_hdcp.h> -#include <drm/i915_component.h> +#include <linux/component.h> #include <linux/i2c.h> #include <linux/random.h> -#include <linux/component.h> -#include "intel_drv.h" +#include <drm/drm_hdcp.h> +#include <drm/i915_component.h> + #include "i915_reg.h" +#include "intel_drv.h" +#include "intel_hdcp.h" #define KEY_LOAD_TRIES 5 #define ENCRYPT_STATUS_CHANGE_TIMEOUT_MS 50 diff --git a/drivers/gpu/drm/i915/intel_hdcp.h b/drivers/gpu/drm/i915/intel_hdcp.h new file mode 100644 index 000000000000..a75f25f09d39 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_hdcp.h @@ -0,0 +1,33 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_HDCP_H__ +#define __INTEL_HDCP_H__ + +#include <linux/types.h> + +#include <drm/i915_drm.h> + +struct drm_connector; +struct drm_connector_state; +struct drm_i915_private; +struct intel_connector; +struct intel_hdcp_shim; + +void intel_hdcp_atomic_check(struct drm_connector *connector, + struct drm_connector_state *old_state, + struct drm_connector_state *new_state); +int intel_hdcp_init(struct intel_connector *connector, + const struct intel_hdcp_shim *hdcp_shim); +int intel_hdcp_enable(struct intel_connector *connector); +int intel_hdcp_disable(struct intel_connector *connector); +bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port); +bool intel_hdcp_capable(struct intel_connector *connector); +void intel_hdcp_component_init(struct drm_i915_private *dev_priv); +void intel_hdcp_component_fini(struct drm_i915_private *dev_priv); +void intel_hdcp_cleanup(struct intel_connector *connector); +void intel_hdcp_handle_cp_irq(struct intel_connector *connector); + +#endif /* __INTEL_HDCP_H__ */ diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index 26767785f14a..e1005d7b75fd 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -26,19 +26,30 @@ * Jesse Barnes <jesse.barnes@intel.com> */ -#include <linux/i2c.h> -#include <linux/slab.h> #include <linux/delay.h> #include <linux/hdmi.h> +#include <linux/i2c.h> +#include <linux/slab.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_edid.h> #include <drm/drm_hdcp.h> #include <drm/drm_scdc_helper.h> -#include "intel_drv.h" #include <drm/i915_drm.h> #include <drm/intel_lpe_audio.h> + #include "i915_drv.h" +#include "intel_audio.h" +#include "intel_connector.h" +#include "intel_ddi.h" +#include "intel_dp.h" +#include "intel_drv.h" +#include "intel_hdcp.h" +#include "intel_hdmi.h" +#include "intel_lspcon.h" +#include "intel_sdvo.h" +#include "intel_panel.h" static struct drm_device *intel_hdmi_to_dev(struct intel_hdmi *intel_hdmi) { diff --git a/drivers/gpu/drm/i915/intel_hdmi.h b/drivers/gpu/drm/i915/intel_hdmi.h new file mode 100644 index 000000000000..106c2e0bc3c9 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_hdmi.h @@ -0,0 +1,51 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_HDMI_H__ +#define __INTEL_HDMI_H__ + +#include <linux/hdmi.h> +#include <linux/types.h> + +#include <drm/i915_drm.h> + +#include "i915_reg.h" + +struct drm_connector; +struct drm_encoder; +struct drm_i915_private; +struct intel_connector; +struct intel_digital_port; +struct intel_encoder; +struct intel_crtc_state; +struct intel_hdmi; +struct drm_connector_state; +union hdmi_infoframe; + +void intel_hdmi_init(struct drm_i915_private *dev_priv, i915_reg_t hdmi_reg, + enum port port); +void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port, + struct intel_connector *intel_connector); +struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder); +int intel_hdmi_compute_config(struct intel_encoder *encoder, + struct intel_crtc_state *pipe_config, + struct drm_connector_state *conn_state); +bool intel_hdmi_handle_sink_scrambling(struct intel_encoder *encoder, + struct drm_connector *connector, + bool high_tmds_clock_ratio, + bool scrambling); +void intel_dp_dual_mode_set_tmds_output(struct intel_hdmi *hdmi, bool enable); +void intel_infoframe_init(struct intel_digital_port *intel_dig_port); +u32 intel_hdmi_infoframes_enabled(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state); +u32 intel_hdmi_infoframe_enable(unsigned int type); +void intel_hdmi_read_gcp_infoframe(struct intel_encoder *encoder, + struct intel_crtc_state *crtc_state); +void intel_read_infoframe(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + enum hdmi_infoframe_type type, + union hdmi_infoframe *frame); + +#endif /* __INTEL_HDMI_H__ */ diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index bec232acc8d7..4e0a351bfbca 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -233,7 +233,7 @@ static inline bool need_preempt(const struct intel_engine_cs *engine, { int last_prio; - if (!intel_engine_has_preemption(engine)) + if (!engine->preempt_context) return false; if (i915_request_completed(rq)) @@ -429,13 +429,13 @@ __unwind_incomplete_requests(struct intel_engine_cs *engine) return active; } -void +struct i915_request * execlists_unwind_incomplete_requests(struct intel_engine_execlists *execlists) { struct intel_engine_cs *engine = container_of(execlists, typeof(*engine), execlists); - __unwind_incomplete_requests(engine); + return __unwind_incomplete_requests(engine); } static inline void @@ -893,96 +893,6 @@ invalidate_csb_entries(const u32 *first, const u32 *last) clflush((void *)last); } -static void reset_csb_pointers(struct intel_engine_execlists *execlists) -{ - const unsigned int reset_value = GEN8_CSB_ENTRIES - 1; - - /* - * After a reset, the HW starts writing into CSB entry [0]. We - * therefore have to set our HEAD pointer back one entry so that - * the *first* entry we check is entry 0. To complicate this further, - * as we don't wait for the first interrupt after reset, we have to - * fake the HW write to point back to the last entry so that our - * inline comparison of our cached head position against the last HW - * write works even before the first interrupt. - */ - execlists->csb_head = reset_value; - WRITE_ONCE(*execlists->csb_write, reset_value); - - invalidate_csb_entries(&execlists->csb_status[0], - &execlists->csb_status[GEN8_CSB_ENTRIES - 1]); -} - -static void nop_submission_tasklet(unsigned long data) -{ - /* The driver is wedged; don't process any more events. */ -} - -static void execlists_cancel_requests(struct intel_engine_cs *engine) -{ - struct intel_engine_execlists * const execlists = &engine->execlists; - struct i915_request *rq, *rn; - struct rb_node *rb; - unsigned long flags; - - GEM_TRACE("%s\n", engine->name); - - /* - * Before we call engine->cancel_requests(), we should have exclusive - * access to the submission state. This is arranged for us by the - * caller disabling the interrupt generation, the tasklet and other - * threads that may then access the same state, giving us a free hand - * to reset state. However, we still need to let lockdep be aware that - * we know this state may be accessed in hardirq context, so we - * disable the irq around this manipulation and we want to keep - * the spinlock focused on its duties and not accidentally conflate - * coverage to the submission's irq state. (Similarly, although we - * shouldn't need to disable irq around the manipulation of the - * submission's irq state, we also wish to remind ourselves that - * it is irq state.) - */ - spin_lock_irqsave(&engine->timeline.lock, flags); - - /* Cancel the requests on the HW and clear the ELSP tracker. */ - execlists_cancel_port_requests(execlists); - execlists_user_end(execlists); - - /* Mark all executing requests as skipped. */ - list_for_each_entry(rq, &engine->timeline.requests, link) { - if (!i915_request_signaled(rq)) - dma_fence_set_error(&rq->fence, -EIO); - - i915_request_mark_complete(rq); - } - - /* Flush the queued requests to the timeline list (for retiring). */ - while ((rb = rb_first_cached(&execlists->queue))) { - struct i915_priolist *p = to_priolist(rb); - int i; - - priolist_for_each_request_consume(rq, rn, p, i) { - list_del_init(&rq->sched.link); - __i915_request_submit(rq); - dma_fence_set_error(&rq->fence, -EIO); - i915_request_mark_complete(rq); - } - - rb_erase_cached(&p->node, &execlists->queue); - i915_priolist_free(p); - } - - /* Remaining _unready_ requests will be nop'ed when submitted */ - - execlists->queue_priority_hint = INT_MIN; - execlists->queue = RB_ROOT_CACHED; - GEM_BUG_ON(port_isset(execlists->port)); - - GEM_BUG_ON(__tasklet_is_enabled(&execlists->tasklet)); - execlists->tasklet.func = nop_submission_tasklet; - - spin_unlock_irqrestore(&engine->timeline.lock, flags); -} - static inline bool reset_in_progress(const struct intel_engine_execlists *execlists) { @@ -994,6 +904,7 @@ static void process_csb(struct intel_engine_cs *engine) struct intel_engine_execlists * const execlists = &engine->execlists; struct execlist_port *port = execlists->port; const u32 * const buf = execlists->csb_status; + const u8 num_entries = execlists->csb_size; u8 head, tail; lockdep_assert_held(&engine->timeline.lock); @@ -1029,7 +940,7 @@ static void process_csb(struct intel_engine_cs *engine) unsigned int status; unsigned int count; - if (++head == GEN8_CSB_ENTRIES) + if (++head == num_entries) head = 0; /* @@ -1151,7 +1062,7 @@ static void process_csb(struct intel_engine_cs *engine) * the wash as hardware, working or not, will need to do the * invalidation before. */ - invalidate_csb_entries(&buf[0], &buf[GEN8_CSB_ENTRIES - 1]); + invalidate_csb_entries(&buf[0], &buf[num_entries - 1]); } static void __execlists_submission_tasklet(struct intel_engine_cs *const engine) @@ -1379,9 +1290,33 @@ static int execlists_context_pin(struct intel_context *ce) return __execlists_context_pin(ce, ce->engine); } +static void execlists_context_reset(struct intel_context *ce) +{ + /* + * Because we emit WA_TAIL_DWORDS there may be a disparity + * between our bookkeeping in ce->ring->head and ce->ring->tail and + * that stored in context. As we only write new commands from + * ce->ring->tail onwards, everything before that is junk. If the GPU + * starts reading from its RING_HEAD from the context, it may try to + * execute that junk and die. + * + * The contexts that are stilled pinned on resume belong to the + * kernel, and are local to each engine. All other contexts will + * have their head/tail sanitized upon pinning before use, so they + * will never see garbage, + * + * So to avoid that we reset the context images upon resume. For + * simplicity, we just zero everything out. + */ + intel_ring_reset(ce->ring, 0); + __execlists_update_reg_state(ce, ce->engine); +} + static const struct intel_context_ops execlists_context_ops = { .pin = execlists_context_pin, .unpin = execlists_context_unpin, + + .reset = execlists_context_reset, .destroy = execlists_context_destroy, }; @@ -1451,10 +1386,11 @@ static int emit_pdps(struct i915_request *rq) *cs++ = MI_LOAD_REGISTER_IMM(2 * GEN8_3LVL_PDPES) | MI_LRI_FORCE_POSTED; for (i = GEN8_3LVL_PDPES; i--; ) { const dma_addr_t pd_daddr = i915_page_dir_dma_addr(ppgtt, i); + u32 base = engine->mmio_base; - *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(engine, i)); + *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_UDW(base, i)); *cs++ = upper_32_bits(pd_daddr); - *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(engine, i)); + *cs++ = i915_mmio_reg_offset(GEN8_RING_PDP_LDW(base, i)); *cs++ = lower_32_bits(pd_daddr); } *cs++ = MI_NOOP; @@ -1823,17 +1759,9 @@ static void enable_execlists(struct intel_engine_cs *engine) intel_engine_set_hwsp_writemask(engine, ~0u); /* HWSTAM */ - /* - * Make sure we're not enabling the new 12-deep CSB - * FIFO as that requires a slightly updated handling - * in the ctx switch irq. Since we're currently only - * using only 2 elements of the enhanced execlists the - * deeper FIFO it's not needed and it's not worth adding - * more statements to the irq handler to support it. - */ if (INTEL_GEN(dev_priv) >= 11) I915_WRITE(RING_MODE_GEN7(engine), - _MASKED_BIT_DISABLE(GEN11_GFX_DISABLE_LEGACY_MODE)); + _MASKED_BIT_ENABLE(GEN11_GFX_DISABLE_LEGACY_MODE)); else I915_WRITE(RING_MODE_GEN7(engine), _MASKED_BIT_ENABLE(GFX_RUN_LIST_ENABLE)); @@ -1903,7 +1831,6 @@ static void execlists_reset_prepare(struct intel_engine_cs *engine) /* And flush any current direct submission. */ spin_lock_irqsave(&engine->timeline.lock, flags); - process_csb(engine); /* drain preemption events */ spin_unlock_irqrestore(&engine->timeline.lock, flags); } @@ -1924,14 +1851,48 @@ static bool lrc_regs_ok(const struct i915_request *rq) return true; } -static void execlists_reset(struct intel_engine_cs *engine, bool stalled) +static void reset_csb_pointers(struct intel_engine_execlists *execlists) +{ + const unsigned int reset_value = execlists->csb_size - 1; + + /* + * After a reset, the HW starts writing into CSB entry [0]. We + * therefore have to set our HEAD pointer back one entry so that + * the *first* entry we check is entry 0. To complicate this further, + * as we don't wait for the first interrupt after reset, we have to + * fake the HW write to point back to the last entry so that our + * inline comparison of our cached head position against the last HW + * write works even before the first interrupt. + */ + execlists->csb_head = reset_value; + WRITE_ONCE(*execlists->csb_write, reset_value); + wmb(); /* Make sure this is visible to HW (paranoia?) */ + + invalidate_csb_entries(&execlists->csb_status[0], + &execlists->csb_status[reset_value]); +} + +static void __execlists_reset(struct intel_engine_cs *engine, bool stalled) { struct intel_engine_execlists * const execlists = &engine->execlists; + struct intel_context *ce; struct i915_request *rq; - unsigned long flags; u32 *regs; - spin_lock_irqsave(&engine->timeline.lock, flags); + process_csb(engine); /* drain preemption events */ + + /* Following the reset, we need to reload the CSB read/write pointers */ + reset_csb_pointers(&engine->execlists); + + /* + * Save the currently executing context, even if we completed + * its request, it was still running at the time of the + * reset and will have been clobbered. + */ + if (!port_isset(execlists->port)) + goto out_clear; + + ce = port_request(execlists->port)->hw_context; /* * Catch up with any missed context-switch interrupts. @@ -1946,12 +1907,13 @@ static void execlists_reset(struct intel_engine_cs *engine, bool stalled) /* Push back any incomplete requests for replay after the reset. */ rq = __unwind_incomplete_requests(engine); - - /* Following the reset, we need to reload the CSB read/write pointers */ - reset_csb_pointers(&engine->execlists); - if (!rq) - goto out_unlock; + goto out_replay; + + if (rq->hw_context != ce) { /* caught just before a CS event */ + rq = NULL; + goto out_replay; + } /* * If this request hasn't started yet, e.g. it is waiting on a @@ -1966,7 +1928,7 @@ static void execlists_reset(struct intel_engine_cs *engine, bool stalled) * perfectly and we do not need to flag the result as being erroneous. */ if (!i915_request_started(rq) && lrc_regs_ok(rq)) - goto out_unlock; + goto out_replay; /* * If the request was innocent, we leave the request in the ELSP @@ -1981,7 +1943,7 @@ static void execlists_reset(struct intel_engine_cs *engine, bool stalled) */ i915_reset_request(rq, stalled); if (!stalled && lrc_regs_ok(rq)) - goto out_unlock; + goto out_replay; /* * We want a simple context + ring to execute the breadcrumb update. @@ -1991,21 +1953,103 @@ static void execlists_reset(struct intel_engine_cs *engine, bool stalled) * future request will be after userspace has had the opportunity * to recreate its own state. */ - regs = rq->hw_context->lrc_reg_state; + regs = ce->lrc_reg_state; if (engine->pinned_default_state) { memcpy(regs, /* skip restoring the vanilla PPHWSP */ engine->pinned_default_state + LRC_STATE_PN * PAGE_SIZE, engine->context_size - PAGE_SIZE); } + execlists_init_reg_state(regs, ce, engine, ce->ring); /* Rerun the request; its payload has been neutered (if guilty). */ - rq->ring->head = intel_ring_wrap(rq->ring, rq->head); - intel_ring_update_space(rq->ring); +out_replay: + ce->ring->head = + rq ? intel_ring_wrap(ce->ring, rq->head) : ce->ring->tail; + intel_ring_update_space(ce->ring); + __execlists_update_reg_state(ce, engine); - execlists_init_reg_state(regs, rq->hw_context, engine, rq->ring); - __execlists_update_reg_state(rq->hw_context, engine); +out_clear: + execlists_clear_all_active(execlists); +} + +static void execlists_reset(struct intel_engine_cs *engine, bool stalled) +{ + unsigned long flags; + + GEM_TRACE("%s\n", engine->name); + + spin_lock_irqsave(&engine->timeline.lock, flags); + + __execlists_reset(engine, stalled); + + spin_unlock_irqrestore(&engine->timeline.lock, flags); +} + +static void nop_submission_tasklet(unsigned long data) +{ + /* The driver is wedged; don't process any more events. */ +} + +static void execlists_cancel_requests(struct intel_engine_cs *engine) +{ + struct intel_engine_execlists * const execlists = &engine->execlists; + struct i915_request *rq, *rn; + struct rb_node *rb; + unsigned long flags; + + GEM_TRACE("%s\n", engine->name); + + /* + * Before we call engine->cancel_requests(), we should have exclusive + * access to the submission state. This is arranged for us by the + * caller disabling the interrupt generation, the tasklet and other + * threads that may then access the same state, giving us a free hand + * to reset state. However, we still need to let lockdep be aware that + * we know this state may be accessed in hardirq context, so we + * disable the irq around this manipulation and we want to keep + * the spinlock focused on its duties and not accidentally conflate + * coverage to the submission's irq state. (Similarly, although we + * shouldn't need to disable irq around the manipulation of the + * submission's irq state, we also wish to remind ourselves that + * it is irq state.) + */ + spin_lock_irqsave(&engine->timeline.lock, flags); + + __execlists_reset(engine, true); + + /* Mark all executing requests as skipped. */ + list_for_each_entry(rq, &engine->timeline.requests, link) { + if (!i915_request_signaled(rq)) + dma_fence_set_error(&rq->fence, -EIO); + + i915_request_mark_complete(rq); + } + + /* Flush the queued requests to the timeline list (for retiring). */ + while ((rb = rb_first_cached(&execlists->queue))) { + struct i915_priolist *p = to_priolist(rb); + int i; + + priolist_for_each_request_consume(rq, rn, p, i) { + list_del_init(&rq->sched.link); + __i915_request_submit(rq); + dma_fence_set_error(&rq->fence, -EIO); + i915_request_mark_complete(rq); + } + + rb_erase_cached(&p->node, &execlists->queue); + i915_priolist_free(p); + } + + /* Remaining _unready_ requests will be nop'ed when submitted */ + + execlists->queue_priority_hint = INT_MIN; + execlists->queue = RB_ROOT_CACHED; + GEM_BUG_ON(port_isset(execlists->port)); + + GEM_BUG_ON(__tasklet_is_enabled(&execlists->tasklet)); + execlists->tasklet.func = nop_submission_tasklet; -out_unlock: spin_unlock_irqrestore(&engine->timeline.lock, flags); } @@ -2035,7 +2079,7 @@ static int gen8_emit_bb_start(struct i915_request *rq, { u32 *cs; - cs = intel_ring_begin(rq, 6); + cs = intel_ring_begin(rq, 4); if (IS_ERR(cs)) return PTR_ERR(cs); @@ -2046,19 +2090,37 @@ static int gen8_emit_bb_start(struct i915_request *rq, * particular all the gen that do not need the w/a at all!), if we * took care to make sure that on every switch into this context * (both ordinary and for preemption) that arbitrartion was enabled - * we would be fine. However, there doesn't seem to be a downside to - * being paranoid and making sure it is set before each batch and - * every context-switch. - * - * Note that if we fail to enable arbitration before the request - * is complete, then we do not see the context-switch interrupt and - * the engine hangs (with RING_HEAD == RING_TAIL). - * - * That satisfies both the GPGPU w/a and our heavy-handed paranoia. + * we would be fine. However, for gen8 there is another w/a that + * requires us to not preempt inside GPGPU execution, so we keep + * arbitration disabled for gen8 batches. Arbitration will be + * re-enabled before we close the request + * (engine->emit_fini_breadcrumb). */ + *cs++ = MI_ARB_ON_OFF | MI_ARB_DISABLE; + + /* FIXME(BDW+): Address space and security selectors. */ + *cs++ = MI_BATCH_BUFFER_START_GEN8 | + (flags & I915_DISPATCH_SECURE ? 0 : BIT(8)); + *cs++ = lower_32_bits(offset); + *cs++ = upper_32_bits(offset); + + intel_ring_advance(rq, cs); + + return 0; +} + +static int gen9_emit_bb_start(struct i915_request *rq, + u64 offset, u32 len, + const unsigned int flags) +{ + u32 *cs; + + cs = intel_ring_begin(rq, 6); + if (IS_ERR(cs)) + return PTR_ERR(cs); + *cs++ = MI_ARB_ON_OFF | MI_ARB_ENABLE; - /* FIXME(BDW): Address space and security selectors. */ *cs++ = MI_BATCH_BUFFER_START_GEN8 | (flags & I915_DISPATCH_SECURE ? 0 : BIT(8)); *cs++ = lower_32_bits(offset); @@ -2309,6 +2371,8 @@ void intel_execlists_set_default_submission(struct intel_engine_cs *engine) engine->execlists.tasklet.func = execlists_submission_tasklet; engine->reset.prepare = execlists_reset_prepare; + engine->reset.reset = execlists_reset; + engine->reset.finish = execlists_reset_finish; engine->park = NULL; engine->unpark = NULL; @@ -2316,7 +2380,8 @@ void intel_execlists_set_default_submission(struct intel_engine_cs *engine) engine->flags |= I915_ENGINE_SUPPORTS_STATS; if (!intel_vgpu_active(engine->i915)) engine->flags |= I915_ENGINE_HAS_SEMAPHORES; - if (engine->preempt_context) + if (engine->preempt_context && + HAS_LOGICAL_RING_PREEMPTION(engine->i915)) engine->flags |= I915_ENGINE_HAS_PREEMPTION; } @@ -2350,7 +2415,10 @@ logical_ring_default_vfuncs(struct intel_engine_cs *engine) * until a more refined solution exists. */ } - engine->emit_bb_start = gen8_emit_bb_start; + if (IS_GEN(engine->i915, 8)) + engine->emit_bb_start = gen8_emit_bb_start; + else + engine->emit_bb_start = gen9_emit_bb_start; } static inline void @@ -2429,6 +2497,11 @@ static int logical_ring_init(struct intel_engine_cs *engine) execlists->csb_write = &engine->status_page.addr[intel_hws_csb_write_index(i915)]; + if (INTEL_GEN(engine->i915) < 11) + execlists->csb_size = GEN8_CSB_ENTRIES; + else + execlists->csb_size = GEN11_CSB_ENTRIES; + reset_csb_pointers(execlists); return 0; @@ -2707,14 +2780,14 @@ static void execlists_init_reg_state(u32 *regs, CTX_REG(regs, CTX_CTX_TIMESTAMP, RING_CTX_TIMESTAMP(base), 0); /* PDP values well be assigned later if needed */ - CTX_REG(regs, CTX_PDP3_UDW, GEN8_RING_PDP_UDW(engine, 3), 0); - CTX_REG(regs, CTX_PDP3_LDW, GEN8_RING_PDP_LDW(engine, 3), 0); - CTX_REG(regs, CTX_PDP2_UDW, GEN8_RING_PDP_UDW(engine, 2), 0); - CTX_REG(regs, CTX_PDP2_LDW, GEN8_RING_PDP_LDW(engine, 2), 0); - CTX_REG(regs, CTX_PDP1_UDW, GEN8_RING_PDP_UDW(engine, 1), 0); - CTX_REG(regs, CTX_PDP1_LDW, GEN8_RING_PDP_LDW(engine, 1), 0); - CTX_REG(regs, CTX_PDP0_UDW, GEN8_RING_PDP_UDW(engine, 0), 0); - CTX_REG(regs, CTX_PDP0_LDW, GEN8_RING_PDP_LDW(engine, 0), 0); + CTX_REG(regs, CTX_PDP3_UDW, GEN8_RING_PDP_UDW(base, 3), 0); + CTX_REG(regs, CTX_PDP3_LDW, GEN8_RING_PDP_LDW(base, 3), 0); + CTX_REG(regs, CTX_PDP2_UDW, GEN8_RING_PDP_UDW(base, 2), 0); + CTX_REG(regs, CTX_PDP2_LDW, GEN8_RING_PDP_LDW(base, 2), 0); + CTX_REG(regs, CTX_PDP1_UDW, GEN8_RING_PDP_UDW(base, 1), 0); + CTX_REG(regs, CTX_PDP1_LDW, GEN8_RING_PDP_LDW(base, 1), 0); + CTX_REG(regs, CTX_PDP0_UDW, GEN8_RING_PDP_UDW(base, 0), 0); + CTX_REG(regs, CTX_PDP0_LDW, GEN8_RING_PDP_LDW(base, 0), 0); if (i915_vm_is_4lvl(&ppgtt->vm)) { /* 64b PPGTT (48bit canonical) @@ -2872,31 +2945,6 @@ error_deref_obj: return ret; } -void intel_lr_context_resume(struct drm_i915_private *i915) -{ - struct i915_gem_context *ctx; - struct intel_context *ce; - - /* - * Because we emit WA_TAIL_DWORDS there may be a disparity - * between our bookkeeping in ce->ring->head and ce->ring->tail and - * that stored in context. As we only write new commands from - * ce->ring->tail onwards, everything before that is junk. If the GPU - * starts reading from its RING_HEAD from the context, it may try to - * execute that junk and die. - * - * So to avoid that we reset the context images upon resume. For - * simplicity, we just zero everything out. - */ - list_for_each_entry(ctx, &i915->contexts.list, link) { - list_for_each_entry(ce, &ctx->active_engines, active_link) { - GEM_BUG_ON(!ce->ring); - intel_ring_reset(ce->ring, 0); - __execlists_update_reg_state(ce, ce->engine); - } - } -} - void intel_execlists_show_requests(struct intel_engine_cs *engine, struct drm_printer *m, void (*show_request)(struct drm_printer *m, @@ -2957,6 +3005,37 @@ void intel_execlists_show_requests(struct intel_engine_cs *engine, spin_unlock_irqrestore(&engine->timeline.lock, flags); } +void intel_lr_context_reset(struct intel_engine_cs *engine, + struct intel_context *ce, + u32 head, + bool scrub) +{ + /* + * We want a simple context + ring to execute the breadcrumb update. + * We cannot rely on the context being intact across the GPU hang, + * so clear it and rebuild just what we need for the breadcrumb. + * All pending requests for this context will be zapped, and any + * future request will be after userspace has had the opportunity + * to recreate its own state. + */ + if (scrub) { + u32 *regs = ce->lrc_reg_state; + + if (engine->pinned_default_state) { + memcpy(regs, /* skip restoring the vanilla PPHWSP */ + engine->pinned_default_state + LRC_STATE_PN * PAGE_SIZE, + engine->context_size - PAGE_SIZE); + } + execlists_init_reg_state(regs, ce, engine, ce->ring); + } + + /* Rerun the request; its payload has been neutered (if guilty). */ + ce->ring->head = head; + intel_ring_update_space(ce->ring); + + __execlists_update_reg_state(ce, engine); +} + #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) #include "selftests/intel_lrc.c" #endif diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h index 92642ab91472..84aa230ea27b 100644 --- a/drivers/gpu/drm/i915/intel_lrc.h +++ b/drivers/gpu/drm/i915/intel_lrc.h @@ -36,12 +36,10 @@ #define CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT (1 << 0) #define CTX_CTRL_RS_CTX_ENABLE (1 << 1) #define CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT (1 << 2) -#define RING_CONTEXT_STATUS_BUF_BASE(base) _MMIO((base) + 0x370) -#define RING_CONTEXT_STATUS_BUF_LO(base, i) _MMIO((base) + 0x370 + (i) * 8) -#define RING_CONTEXT_STATUS_BUF_HI(base, i) _MMIO((base) + 0x370 + (i) * 8 + 4) #define RING_CONTEXT_STATUS_PTR(base) _MMIO((base) + 0x3a0) #define RING_EXECLIST_SQ_CONTENTS(base) _MMIO((base) + 0x510) #define RING_EXECLIST_CONTROL(base) _MMIO((base) + 0x550) + #define EL_CTRL_LOAD (1 << 0) /* The docs specify that the write pointer wraps around after 5h, "After status @@ -55,10 +53,11 @@ #define GEN8_CSB_PTR_MASK 0x7 #define GEN8_CSB_READ_PTR_MASK (GEN8_CSB_PTR_MASK << 8) #define GEN8_CSB_WRITE_PTR_MASK (GEN8_CSB_PTR_MASK << 0) -#define GEN8_CSB_WRITE_PTR(csb_status) \ - (((csb_status) & GEN8_CSB_WRITE_PTR_MASK) >> 0) -#define GEN8_CSB_READ_PTR(csb_status) \ - (((csb_status) & GEN8_CSB_READ_PTR_MASK) >> 8) + +#define GEN11_CSB_ENTRIES 12 +#define GEN11_CSB_PTR_MASK 0xf +#define GEN11_CSB_READ_PTR_MASK (GEN11_CSB_PTR_MASK << 8) +#define GEN11_CSB_WRITE_PTR_MASK (GEN11_CSB_PTR_MASK << 0) enum { INTEL_CONTEXT_SCHEDULE_IN = 0, @@ -102,9 +101,13 @@ struct drm_printer; struct drm_i915_private; struct i915_gem_context; -void intel_lr_context_resume(struct drm_i915_private *dev_priv); void intel_execlists_set_default_submission(struct intel_engine_cs *engine); +void intel_lr_context_reset(struct intel_engine_cs *engine, + struct intel_context *ce, + u32 head, + bool scrub); + void intel_execlists_show_requests(struct intel_engine_cs *engine, struct drm_printer *m, void (*show_request)(struct drm_printer *m, diff --git a/drivers/gpu/drm/i915/intel_lspcon.c b/drivers/gpu/drm/i915/intel_lspcon.c index 8d202b13e24f..7028d0cf3bb1 100644 --- a/drivers/gpu/drm/i915/intel_lspcon.c +++ b/drivers/gpu/drm/i915/intel_lspcon.c @@ -22,10 +22,14 @@ * * */ -#include <drm/drm_edid.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_dp_dual_mode_helper.h> +#include <drm/drm_edid.h> + +#include "intel_dp.h" #include "intel_drv.h" +#include "intel_lspcon.h" /* LSPCON OUI Vendor ID(signatures) */ #define LSPCON_VENDOR_PARADE_OUI 0x001CF8 diff --git a/drivers/gpu/drm/i915/intel_lspcon.h b/drivers/gpu/drm/i915/intel_lspcon.h new file mode 100644 index 000000000000..37cfddf8a9c5 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_lspcon.h @@ -0,0 +1,38 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_LSPCON_H__ +#define __INTEL_LSPCON_H__ + +#include <linux/types.h> + +struct drm_connector; +struct drm_connector_state; +struct intel_crtc_state; +struct intel_digital_port; +struct intel_encoder; +struct intel_lspcon; + +bool lspcon_init(struct intel_digital_port *intel_dig_port); +void lspcon_resume(struct intel_lspcon *lspcon); +void lspcon_wait_pcon_mode(struct intel_lspcon *lspcon); +void lspcon_write_infoframe(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + unsigned int type, + const void *buf, ssize_t len); +void lspcon_read_infoframe(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + unsigned int type, + void *frame, ssize_t len); +void lspcon_set_infoframes(struct intel_encoder *encoder, + bool enable, + const struct intel_crtc_state *crtc_state, + const struct drm_connector_state *conn_state); +u32 lspcon_infoframes_enabled(struct intel_encoder *encoder, + const struct intel_crtc_state *pipe_config); +void lspcon_ycbcr420_config(struct drm_connector *connector, + struct intel_crtc_state *crtc_state); + +#endif /* __INTEL_LSPCON_H__ */ diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index 34dd2d71814b..51d1d59c1619 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -28,17 +28,22 @@ */ #include <acpi/button.h> +#include <linux/acpi.h> #include <linux/dmi.h> #include <linux/i2c.h> #include <linux/slab.h> #include <linux/vga_switcheroo.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_edid.h> -#include "intel_drv.h" #include <drm/i915_drm.h> + #include "i915_drv.h" -#include <linux/acpi.h> +#include "intel_connector.h" +#include "intel_drv.h" +#include "intel_lvds.h" +#include "intel_panel.h" /* Private structure for the integrated LVDS support */ struct intel_lvds_pps { diff --git a/drivers/gpu/drm/i915/intel_lvds.h b/drivers/gpu/drm/i915/intel_lvds.h new file mode 100644 index 000000000000..bc9c8b84ba2f --- /dev/null +++ b/drivers/gpu/drm/i915/intel_lvds.h @@ -0,0 +1,22 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_LVDS_H__ +#define __INTEL_LVDS_H__ + +#include <linux/types.h> + +#include "i915_reg.h" + +enum pipe; +struct drm_i915_private; + +bool intel_lvds_port_enabled(struct drm_i915_private *dev_priv, + i915_reg_t lvds_reg, enum pipe *pipe); +void intel_lvds_init(struct drm_i915_private *dev_priv); +struct intel_encoder *intel_get_lvds_encoder(struct drm_i915_private *dev_priv); +bool intel_is_dual_link_lvds(struct drm_i915_private *dev_priv); + +#endif /* __INTEL_LVDS_H__ */ diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c index 5e00ee9270b5..8fa1159d097f 100644 --- a/drivers/gpu/drm/i915/intel_opregion.c +++ b/drivers/gpu/drm/i915/intel_opregion.c @@ -32,9 +32,10 @@ #include <drm/i915_drm.h> -#include "intel_opregion.h" #include "i915_drv.h" #include "intel_drv.h" +#include "intel_opregion.h" +#include "intel_panel.h" #define OPREGION_HEADER_OFFSET 0 #define OPREGION_ACPI_OFFSET 0x100 diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c index a882b8d42bd9..eb317759b5d3 100644 --- a/drivers/gpu/drm/i915/intel_overlay.c +++ b/drivers/gpu/drm/i915/intel_overlay.c @@ -446,7 +446,7 @@ static int intel_overlay_release_old_vid(struct intel_overlay *overlay) if (!overlay->old_vma) return 0; - if (I915_READ(ISR) & I915_OVERLAY_PLANE_FLIP_PENDING_INTERRUPT) { + if (I915_READ(GEN2_ISR) & I915_OVERLAY_PLANE_FLIP_PENDING_INTERRUPT) { /* synchronous slowpath */ struct i915_request *rq; @@ -1430,7 +1430,7 @@ intel_overlay_capture_error_state(struct drm_i915_private *dev_priv) return NULL; error->dovsta = I915_READ(DOVSTA); - error->isr = I915_READ(ISR); + error->isr = I915_READ(GEN2_ISR); error->base = overlay->flip_addr; memcpy_fromio(&error->regs, overlay->regs, sizeof(error->regs)); diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c index 47cd4a338db6..4ab4ce6569e7 100644 --- a/drivers/gpu/drm/i915/intel_panel.c +++ b/drivers/gpu/drm/i915/intel_panel.c @@ -33,7 +33,10 @@ #include <linux/kernel.h> #include <linux/moduleparam.h> #include <linux/pwm.h> + +#include "intel_connector.h" #include "intel_drv.h" +#include "intel_panel.h" #define CRC_PMIC_PWM_PERIOD_NS 21333 diff --git a/drivers/gpu/drm/i915/intel_panel.h b/drivers/gpu/drm/i915/intel_panel.h new file mode 100644 index 000000000000..cedeea443336 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_panel.h @@ -0,0 +1,65 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_PANEL_H__ +#define __INTEL_PANEL_H__ + +#include <linux/types.h> + +#include "intel_display.h" + +struct drm_connector; +struct drm_connector_state; +struct drm_display_mode; +struct intel_connector; +struct intel_crtc; +struct intel_crtc_state; +struct intel_encoder; +struct intel_panel; + +int intel_panel_init(struct intel_panel *panel, + struct drm_display_mode *fixed_mode, + struct drm_display_mode *downclock_mode); +void intel_panel_fini(struct intel_panel *panel); +void intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode, + struct drm_display_mode *adjusted_mode); +void intel_pch_panel_fitting(struct intel_crtc *crtc, + struct intel_crtc_state *pipe_config, + int fitting_mode); +void intel_gmch_panel_fitting(struct intel_crtc *crtc, + struct intel_crtc_state *pipe_config, + int fitting_mode); +void intel_panel_set_backlight_acpi(const struct drm_connector_state *conn_state, + u32 level, u32 max); +int intel_panel_setup_backlight(struct drm_connector *connector, + enum pipe pipe); +void intel_panel_enable_backlight(const struct intel_crtc_state *crtc_state, + const struct drm_connector_state *conn_state); +void intel_panel_update_backlight(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + const struct drm_connector_state *conn_state); +void intel_panel_disable_backlight(const struct drm_connector_state *old_conn_state); +struct drm_display_mode * +intel_panel_edid_downclock_mode(struct intel_connector *connector, + const struct drm_display_mode *fixed_mode); +struct drm_display_mode * +intel_panel_edid_fixed_mode(struct intel_connector *connector); +struct drm_display_mode * +intel_panel_vbt_fixed_mode(struct intel_connector *connector); + +#if IS_ENABLED(CONFIG_BACKLIGHT_CLASS_DEVICE) +int intel_backlight_device_register(struct intel_connector *connector); +void intel_backlight_device_unregister(struct intel_connector *connector); +#else /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +static inline int intel_backlight_device_register(struct intel_connector *connector) +{ + return 0; +} +static inline void intel_backlight_device_unregister(struct intel_connector *connector) +{ +} +#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */ + +#endif /* __INTEL_PANEL_H__ */ diff --git a/drivers/gpu/drm/i915/intel_pipe_crc.c b/drivers/gpu/drm/i915/intel_pipe_crc.c index 0b1378f0bff7..e94b5b1bc1b7 100644 --- a/drivers/gpu/drm/i915/intel_pipe_crc.c +++ b/drivers/gpu/drm/i915/intel_pipe_crc.c @@ -24,11 +24,13 @@ * */ -#include <linux/seq_file.h> #include <linux/circ_buf.h> #include <linux/ctype.h> #include <linux/debugfs.h> +#include <linux/seq_file.h> + #include "intel_drv.h" +#include "intel_pipe_crc.h" static const char * const pipe_crc_sources[] = { [INTEL_PIPE_CRC_SOURCE_NONE] = "none", diff --git a/drivers/gpu/drm/i915/intel_pipe_crc.h b/drivers/gpu/drm/i915/intel_pipe_crc.h new file mode 100644 index 000000000000..81eaf1854788 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_pipe_crc.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_PIPE_CRC_H__ +#define __INTEL_PIPE_CRC_H__ + +#include <linux/types.h> + +struct drm_crtc; +struct intel_crtc; + +#ifdef CONFIG_DEBUG_FS +int intel_crtc_set_crc_source(struct drm_crtc *crtc, const char *source_name); +int intel_crtc_verify_crc_source(struct drm_crtc *crtc, + const char *source_name, size_t *values_cnt); +const char *const *intel_crtc_get_crc_sources(struct drm_crtc *crtc, + size_t *count); +void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc); +void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc); +#else +#define intel_crtc_set_crc_source NULL +#define intel_crtc_verify_crc_source NULL +#define intel_crtc_get_crc_sources NULL +static inline void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc) +{ +} + +static inline void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc) +{ +} +#endif + +#endif /* __INTEL_PIPE_CRC_H__ */ diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 9a6eb2ef5f48..44be676fabd6 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -35,6 +35,9 @@ #include "i915_drv.h" #include "intel_drv.h" +#include "intel_fbc.h" +#include "intel_pm.h" +#include "intel_sprite.h" #include "../../../platform/x86/intel_ips.h" /** @@ -850,7 +853,7 @@ static void pineview_update_wm(struct intel_crtc *unused_crtc) u32 reg; unsigned int wm; - latency = intel_get_cxsr_latency(IS_PINEVIEW_G(dev_priv), + latency = intel_get_cxsr_latency(!IS_MOBILE(dev_priv), dev_priv->is_ddr3, dev_priv->fsb_freq, dev_priv->mem_freq); @@ -4527,10 +4530,10 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate, memset(&wm->wm[level], 0, sizeof(wm->wm[level])); /* - * Wa_1408961008:icl + * Wa_1408961008:icl, ehl * Underruns with WM1+ disabled */ - if (IS_ICELAKE(dev_priv) && + if (IS_GEN(dev_priv, 11) && level == 1 && wm->wm[0].plane_en) { wm->wm[level].plane_res_b = wm->wm[0].plane_res_b; wm->wm[level].plane_res_l = wm->wm[0].plane_res_l; @@ -5251,7 +5254,7 @@ static inline bool skl_ddb_entries_overlap(const struct skl_ddb_entry *a, } bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb, - const struct skl_ddb_entry entries[], + const struct skl_ddb_entry *entries, int num_entries, int ignore_idx) { int i; @@ -6603,7 +6606,7 @@ static void rps_set_power(struct drm_i915_private *dev_priv, int new_power) ei_down * threshold_down / 100)); I915_WRITE(GEN6_RP_CONTROL, - GEN6_RP_MEDIA_TURBO | + (INTEL_GEN(dev_priv) > 9 ? 0 : GEN6_RP_MEDIA_TURBO) | GEN6_RP_MEDIA_HW_NORMAL_MODE | GEN6_RP_MEDIA_IS_GFX | GEN6_RP_ENABLE | @@ -7117,6 +7120,78 @@ static void gen9_enable_rps(struct drm_i915_private *dev_priv) intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL); } +static void gen11_enable_rc6(struct drm_i915_private *dev_priv) +{ + struct intel_engine_cs *engine; + enum intel_engine_id id; + + /* 1a: Software RC state - RC0 */ + I915_WRITE(GEN6_RC_STATE, 0); + + /* + * 1b: Get forcewake during program sequence. Although the driver + * hasn't enabled a state yet where we need forcewake, BIOS may have. + */ + intel_uncore_forcewake_get(&dev_priv->uncore, FORCEWAKE_ALL); + + /* 2a: Disable RC states. */ + I915_WRITE(GEN6_RC_CONTROL, 0); + + /* 2b: Program RC6 thresholds.*/ + I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16 | 85); + I915_WRITE(GEN10_MEDIA_WAKE_RATE_LIMIT, 150); + + I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */ + I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */ + for_each_engine(engine, dev_priv, id) + I915_WRITE(RING_MAX_IDLE(engine->mmio_base), 10); + + if (HAS_GUC(dev_priv)) + I915_WRITE(GUC_MAX_IDLE_COUNT, 0xA); + + I915_WRITE(GEN6_RC_SLEEP, 0); + + I915_WRITE(GEN6_RC6_THRESHOLD, 50000); /* 50/125ms per EI */ + + /* + * 2c: Program Coarse Power Gating Policies. + * + * Bspec's guidance is to use 25us (really 25 * 1280ns) here. What we + * use instead is a more conservative estimate for the maximum time + * it takes us to service a CS interrupt and submit a new ELSP - that + * is the time which the GPU is idle waiting for the CPU to select the + * next request to execute. If the idle hysteresis is less than that + * interrupt service latency, the hardware will automatically gate + * the power well and we will then incur the wake up cost on top of + * the service latency. A similar guide from intel_pstate is that we + * do not want the enable hysteresis to less than the wakeup latency. + * + * igt/gem_exec_nop/sequential provides a rough estimate for the + * service latency, and puts it around 10us for Broadwell (and other + * big core) and around 40us for Broxton (and other low power cores). + * [Note that for legacy ringbuffer submission, this is less than 1us!] + * However, the wakeup latency on Broxton is closer to 100us. To be + * conservative, we have to factor in a context switch on top (due + * to ksoftirqd). + */ + I915_WRITE(GEN9_MEDIA_PG_IDLE_HYSTERESIS, 250); + I915_WRITE(GEN9_RENDER_PG_IDLE_HYSTERESIS, 250); + + /* 3a: Enable RC6 */ + I915_WRITE(GEN6_RC_CONTROL, + GEN6_RC_CTL_HW_ENABLE | + GEN6_RC_CTL_RC6_ENABLE | + GEN6_RC_CTL_EI_MODE(1)); + + /* 3b: Enable Coarse Power Gating only when RC6 is enabled. */ + I915_WRITE(GEN9_PG_ENABLE, + GEN9_RENDER_PG_ENABLE | + GEN9_MEDIA_PG_ENABLE | + GEN11_MEDIA_SAMPLER_PG_ENABLE); + + intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL); +} + static void gen9_enable_rc6(struct drm_i915_private *dev_priv) { struct intel_engine_cs *engine; @@ -8188,14 +8263,14 @@ unsigned long i915_gfx_val(struct drm_i915_private *dev_priv) return val; } -static struct drm_i915_private *i915_mch_dev; +static struct drm_i915_private __rcu *i915_mch_dev; static struct drm_i915_private *mchdev_get(void) { struct drm_i915_private *i915; rcu_read_lock(); - i915 = i915_mch_dev; + i915 = rcu_dereference(i915_mch_dev); if (!kref_get_unless_zero(&i915->drm.ref)) i915 = NULL; rcu_read_unlock(); @@ -8593,6 +8668,8 @@ static void intel_enable_rc6(struct drm_i915_private *dev_priv) cherryview_enable_rc6(dev_priv); else if (IS_VALLEYVIEW(dev_priv)) valleyview_enable_rc6(dev_priv); + else if (INTEL_GEN(dev_priv) >= 11) + gen11_enable_rc6(dev_priv); else if (INTEL_GEN(dev_priv) >= 9) gen9_enable_rc6(dev_priv); else if (IS_BROADWELL(dev_priv)) @@ -9496,7 +9573,7 @@ static void nop_init_clock_gating(struct drm_i915_private *dev_priv) */ void intel_init_clock_gating_hooks(struct drm_i915_private *dev_priv) { - if (IS_ICELAKE(dev_priv)) + if (IS_GEN(dev_priv, 11)) dev_priv->display.init_clock_gating = icl_init_clock_gating; else if (IS_CANNONLAKE(dev_priv)) dev_priv->display.init_clock_gating = cnl_init_clock_gating; @@ -9589,7 +9666,7 @@ void intel_init_pm(struct drm_i915_private *dev_priv) dev_priv->display.initial_watermarks = g4x_initial_watermarks; dev_priv->display.optimize_watermarks = g4x_optimize_watermarks; } else if (IS_PINEVIEW(dev_priv)) { - if (!intel_get_cxsr_latency(IS_PINEVIEW_G(dev_priv), + if (!intel_get_cxsr_latency(!IS_MOBILE(dev_priv), dev_priv->is_ddr3, dev_priv->fsb_freq, dev_priv->mem_freq)) { diff --git a/drivers/gpu/drm/i915/intel_pm.h b/drivers/gpu/drm/i915/intel_pm.h new file mode 100644 index 000000000000..674a3f0f16a7 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_pm.h @@ -0,0 +1,71 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_PM_H__ +#define __INTEL_PM_H__ + +#include <linux/types.h> + +struct drm_atomic_state; +struct drm_device; +struct drm_i915_private; +struct i915_request; +struct intel_crtc; +struct intel_crtc_state; +struct intel_plane; +struct skl_ddb_allocation; +struct skl_ddb_entry; +struct skl_pipe_wm; +struct skl_wm_level; + +void intel_init_clock_gating(struct drm_i915_private *dev_priv); +void intel_suspend_hw(struct drm_i915_private *dev_priv); +int ilk_wm_max_level(const struct drm_i915_private *dev_priv); +void intel_update_watermarks(struct intel_crtc *crtc); +void intel_init_pm(struct drm_i915_private *dev_priv); +void intel_init_clock_gating_hooks(struct drm_i915_private *dev_priv); +void intel_pm_setup(struct drm_i915_private *dev_priv); +void intel_gpu_ips_init(struct drm_i915_private *dev_priv); +void intel_gpu_ips_teardown(void); +void intel_init_gt_powersave(struct drm_i915_private *dev_priv); +void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv); +void intel_sanitize_gt_powersave(struct drm_i915_private *dev_priv); +void intel_enable_gt_powersave(struct drm_i915_private *dev_priv); +void intel_disable_gt_powersave(struct drm_i915_private *dev_priv); +void gen6_rps_busy(struct drm_i915_private *dev_priv); +void gen6_rps_idle(struct drm_i915_private *dev_priv); +void gen6_rps_boost(struct i915_request *rq); +void g4x_wm_get_hw_state(struct drm_i915_private *dev_priv); +void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv); +void ilk_wm_get_hw_state(struct drm_i915_private *dev_priv); +void skl_wm_get_hw_state(struct drm_i915_private *dev_priv); +void skl_pipe_ddb_get_hw_state(struct intel_crtc *crtc, + struct skl_ddb_entry *ddb_y, + struct skl_ddb_entry *ddb_uv); +void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, + struct skl_ddb_allocation *ddb /* out */); +void skl_pipe_wm_get_hw_state(struct intel_crtc *crtc, + struct skl_pipe_wm *out); +void g4x_wm_sanitize(struct drm_i915_private *dev_priv); +void vlv_wm_sanitize(struct drm_i915_private *dev_priv); +bool intel_can_enable_sagv(struct drm_atomic_state *state); +int intel_enable_sagv(struct drm_i915_private *dev_priv); +int intel_disable_sagv(struct drm_i915_private *dev_priv); +bool skl_wm_level_equals(const struct skl_wm_level *l1, + const struct skl_wm_level *l2); +bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb, + const struct skl_ddb_entry *entries, + int num_entries, int ignore_idx); +void skl_write_plane_wm(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state); +void skl_write_cursor_wm(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state); +bool ilk_disable_lp_wm(struct drm_device *dev); +int skl_check_pipe_max_pixel_rate(struct intel_crtc *intel_crtc, + struct intel_crtc_state *cstate); +void intel_init_ipc(struct drm_i915_private *dev_priv); +void intel_enable_ipc(struct drm_i915_private *dev_priv); + +#endif /* __INTEL_PM_H__ */ diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c index ec874d802d48..963663ba0edf 100644 --- a/drivers/gpu/drm/i915/intel_psr.c +++ b/drivers/gpu/drm/i915/intel_psr.c @@ -21,6 +21,14 @@ * DEALINGS IN THE SOFTWARE. */ +#include <drm/drm_atomic_helper.h> + +#include "i915_drv.h" +#include "intel_dp.h" +#include "intel_drv.h" +#include "intel_psr.h" +#include "intel_sprite.h" + /** * DOC: Panel Self Refresh (PSR/SRD) * @@ -51,11 +59,6 @@ * must be correctly synchronized/cancelled when shutting down the pipe." */ -#include <drm/drm_atomic_helper.h> - -#include "intel_drv.h" -#include "i915_drv.h" - static bool psr_global_enabled(u32 debug) { switch (debug & I915_PSR_DEBUG_MODE_MASK) { @@ -531,10 +534,8 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp) val |= EDP_PSR2_TP2_TIME_2500us; /* - * FIXME: There is probably a issue in DMC firmwares(icl_dmc_ver1_07.bin - * and kbl_dmc_ver1_04.bin at least) that causes PSR2 SU to fail after - * exiting DC6 if EDP_PSR_TP1_TP3_SEL is kept in PSR_CTL, so for now - * lets workaround the issue by cleaning PSR_CTL before enable PSR2. + * PSR2 HW is incorrectly using EDP_PSR_TP1_TP3_SEL and BSpec is + * recommending keep this bit unset while PSR2 is enabled. */ I915_WRITE(EDP_PSR_CTL, 0); @@ -629,9 +630,8 @@ void intel_psr_compute_config(struct intel_dp *intel_dp, return; } - if (IS_HASWELL(dev_priv) && - adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) { - DRM_DEBUG_KMS("PSR condition failed: Interlaced is Enabled\n"); + if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) { + DRM_DEBUG_KMS("PSR condition failed: Interlaced mode enabled\n"); return; } @@ -1229,7 +1229,6 @@ void intel_psr_init(struct drm_i915_private *dev_priv) if (val) { DRM_DEBUG_KMS("PSR interruption error set\n"); dev_priv->psr.sink_not_reliable = true; - return; } /* Set link_standby x link_off defaults */ diff --git a/drivers/gpu/drm/i915/intel_psr.h b/drivers/gpu/drm/i915/intel_psr.h new file mode 100644 index 000000000000..dc818826f36d --- /dev/null +++ b/drivers/gpu/drm/i915/intel_psr.h @@ -0,0 +1,40 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_PSR_H__ +#define __INTEL_PSR_H__ + +#include "intel_frontbuffer.h" + +struct drm_i915_private; +struct intel_crtc_state; +struct intel_dp; + +#define CAN_PSR(dev_priv) (HAS_PSR(dev_priv) && dev_priv->psr.sink_support) +void intel_psr_init_dpcd(struct intel_dp *intel_dp); +void intel_psr_enable(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state); +void intel_psr_disable(struct intel_dp *intel_dp, + const struct intel_crtc_state *old_crtc_state); +void intel_psr_update(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state); +int intel_psr_debug_set(struct drm_i915_private *dev_priv, u64 value); +void intel_psr_invalidate(struct drm_i915_private *dev_priv, + unsigned frontbuffer_bits, + enum fb_op_origin origin); +void intel_psr_flush(struct drm_i915_private *dev_priv, + unsigned frontbuffer_bits, + enum fb_op_origin origin); +void intel_psr_init(struct drm_i915_private *dev_priv); +void intel_psr_compute_config(struct intel_dp *intel_dp, + struct intel_crtc_state *crtc_state); +void intel_psr_irq_control(struct drm_i915_private *dev_priv, u32 debug); +void intel_psr_irq_handler(struct drm_i915_private *dev_priv, u32 psr_iir); +void intel_psr_short_pulse(struct intel_dp *intel_dp); +int intel_psr_wait_for_idle(const struct intel_crtc_state *new_crtc_state, + u32 *out_value); +bool intel_psr_enabled(struct intel_dp *intel_dp); + +#endif /* __INTEL_PSR_H__ */ diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 48ba4d61a4ae..029fd8ec1857 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -976,20 +976,16 @@ gen5_irq_disable(struct intel_engine_cs *engine) static void i9xx_irq_enable(struct intel_engine_cs *engine) { - GEM_BUG_ON(engine->id != RCS0); - engine->i915->irq_mask &= ~engine->irq_enable_mask; - ENGINE_WRITE(engine, RING_IMR, engine->i915->irq_mask); - ENGINE_POSTING_READ(engine, RING_IMR); + intel_uncore_write(engine->uncore, GEN2_IMR, engine->i915->irq_mask); + intel_uncore_posting_read_fw(engine->uncore, GEN2_IMR); } static void i9xx_irq_disable(struct intel_engine_cs *engine) { - GEM_BUG_ON(engine->id != RCS0); - engine->i915->irq_mask |= engine->irq_enable_mask; - ENGINE_WRITE(engine, RING_IMR, engine->i915->irq_mask); + intel_uncore_write(engine->uncore, GEN2_IMR, engine->i915->irq_mask); } static void @@ -998,7 +994,7 @@ i8xx_irq_enable(struct intel_engine_cs *engine) struct drm_i915_private *dev_priv = engine->i915; dev_priv->irq_mask &= ~engine->irq_enable_mask; - I915_WRITE16(IMR, dev_priv->irq_mask); + I915_WRITE16(GEN2_IMR, dev_priv->irq_mask); POSTING_READ16(RING_IMR(engine->mmio_base)); } @@ -1008,7 +1004,7 @@ i8xx_irq_disable(struct intel_engine_cs *engine) struct drm_i915_private *dev_priv = engine->i915; dev_priv->irq_mask |= engine->irq_enable_mask; - I915_WRITE16(IMR, dev_priv->irq_mask); + I915_WRITE16(GEN2_IMR, dev_priv->irq_mask); } static int @@ -1512,9 +1508,16 @@ err_unpin: return err; } +static void ring_context_reset(struct intel_context *ce) +{ + intel_ring_reset(ce->ring, 0); +} + static const struct intel_context_ops ring_context_ops = { .pin = ring_context_pin, .unpin = ring_context_unpin, + + .reset = ring_context_reset, .destroy = ring_context_destroy, }; @@ -1585,16 +1588,6 @@ void intel_engine_cleanup(struct intel_engine_cs *engine) kfree(engine); } -void intel_legacy_submission_resume(struct drm_i915_private *dev_priv) -{ - struct intel_engine_cs *engine; - enum intel_engine_id id; - - /* Restart from the beginning of the rings for convenience */ - for_each_engine(engine, dev_priv, id) - intel_ring_reset(engine->buffer, 0); -} - static int load_pd_dir(struct i915_request *rq, const struct i915_hw_ppgtt *ppgtt) { diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index e58d6f04177b..72c7c337ace9 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -165,7 +165,7 @@ void execlists_user_end(struct intel_engine_execlists *execlists); void execlists_cancel_port_requests(struct intel_engine_execlists * const execlists); -void +struct i915_request * execlists_unwind_incomplete_requests(struct intel_engine_execlists *execlists); static inline unsigned int @@ -268,8 +268,6 @@ static inline void intel_ring_put(struct intel_ring *ring) void intel_engine_stop(struct intel_engine_cs *engine); void intel_engine_cleanup(struct intel_engine_cs *engine); -void intel_legacy_submission_resume(struct drm_i915_private *dev_priv); - int __must_check intel_ring_cacheline_align(struct i915_request *rq); u32 __must_check *intel_ring_begin(struct i915_request *rq, unsigned int n); @@ -390,7 +388,7 @@ void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine); void intel_engine_pin_breadcrumbs_irq(struct intel_engine_cs *engine); void intel_engine_unpin_breadcrumbs_irq(struct intel_engine_cs *engine); -bool intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine); +void intel_engine_signal_breadcrumbs(struct intel_engine_cs *engine); void intel_engine_disarm_breadcrumbs(struct intel_engine_cs *engine); static inline void @@ -399,7 +397,7 @@ intel_engine_queue_breadcrumbs(struct intel_engine_cs *engine) irq_work_queue(&engine->breadcrumbs.irq_work); } -bool intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine); +void intel_engine_breadcrumbs_irq(struct intel_engine_cs *engine); void intel_engine_reset_breadcrumbs(struct intel_engine_cs *engine); void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine); @@ -463,6 +461,7 @@ static inline void intel_engine_reset(struct intel_engine_cs *engine, } void intel_engines_sanitize(struct drm_i915_private *i915, bool force); +void intel_gt_resume(struct drm_i915_private *i915); bool intel_engine_is_idle(struct intel_engine_cs *engine); bool intel_engines_are_idle(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index 40ddfbb97acb..d4f4262d0fee 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -32,6 +32,10 @@ #include <drm/drm_print.h> #include "i915_drv.h" +#include "intel_cdclk.h" +#include "intel_crt.h" +#include "intel_csr.h" +#include "intel_dp.h" #include "intel_drv.h" /** @@ -158,7 +162,7 @@ static void cancel_intel_runtime_pm_wakeref(struct drm_i915_private *i915, rpm->debug.count, atomic_read(&rpm->wakeref_count))) { char *buf; - buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + buf = kmalloc(PAGE_SIZE, GFP_NOWAIT | __GFP_NOWARN); if (!buf) return; @@ -194,7 +198,7 @@ __print_intel_runtime_pm_wakeref(struct drm_printer *p, unsigned long i; char *buf; - buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + buf = kmalloc(PAGE_SIZE, GFP_NOWAIT | __GFP_NOWARN); if (!buf) return; @@ -278,7 +282,9 @@ void print_intel_runtime_pm_wakeref(struct drm_i915_private *i915, if (dbg.count <= alloc) break; - s = krealloc(dbg.owners, dbg.count * sizeof(*s), GFP_KERNEL); + s = krealloc(dbg.owners, + dbg.count * sizeof(*s), + GFP_NOWAIT | __GFP_NOWARN); if (!s) goto out; @@ -3442,7 +3448,7 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv) * The enabling order will be from lower to higher indexed wells, * the disabling order is reversed. */ - if (IS_ICELAKE(dev_priv)) { + if (IS_GEN(dev_priv, 11)) { err = set_power_wells(power_domains, icl_power_wells); } else if (IS_CANNONLAKE(dev_priv)) { err = set_power_wells(power_domains, cnl_power_wells); @@ -3660,7 +3666,7 @@ static void skl_display_core_init(struct drm_i915_private *dev_priv, mutex_unlock(&power_domains->lock); - skl_init_cdclk(dev_priv); + intel_cdclk_init(dev_priv); gen9_dbuf_enable(dev_priv); @@ -3677,7 +3683,7 @@ static void skl_display_core_uninit(struct drm_i915_private *dev_priv) gen9_dbuf_disable(dev_priv); - skl_uninit_cdclk(dev_priv); + intel_cdclk_uninit(dev_priv); /* The spec doesn't call for removing the reset handshake flag */ /* disable PG1 and Misc I/O */ @@ -3722,7 +3728,7 @@ void bxt_display_core_init(struct drm_i915_private *dev_priv, mutex_unlock(&power_domains->lock); - bxt_init_cdclk(dev_priv); + intel_cdclk_init(dev_priv); gen9_dbuf_enable(dev_priv); @@ -3739,7 +3745,7 @@ void bxt_display_core_uninit(struct drm_i915_private *dev_priv) gen9_dbuf_disable(dev_priv); - bxt_uninit_cdclk(dev_priv); + intel_cdclk_uninit(dev_priv); /* The spec doesn't call for removing the reset handshake flag */ @@ -3781,7 +3787,7 @@ static void cnl_display_core_init(struct drm_i915_private *dev_priv, bool resume mutex_unlock(&power_domains->lock); /* 5. Enable CD clock */ - cnl_init_cdclk(dev_priv); + intel_cdclk_init(dev_priv); /* 6. Enable DBUF */ gen9_dbuf_enable(dev_priv); @@ -3803,7 +3809,7 @@ static void cnl_display_core_uninit(struct drm_i915_private *dev_priv) gen9_dbuf_disable(dev_priv); /* 3. Disable CD clock */ - cnl_uninit_cdclk(dev_priv); + intel_cdclk_uninit(dev_priv); /* * 4. Disable Power Well 1 (PG1). @@ -3845,7 +3851,7 @@ void icl_display_core_init(struct drm_i915_private *dev_priv, mutex_unlock(&power_domains->lock); /* 5. Enable CDCLK. */ - icl_init_cdclk(dev_priv); + intel_cdclk_init(dev_priv); /* 6. Enable DBUF. */ icl_dbuf_enable(dev_priv); @@ -3870,7 +3876,7 @@ void icl_display_core_uninit(struct drm_i915_private *dev_priv) icl_dbuf_disable(dev_priv); /* 3. Disable CD clock */ - icl_uninit_cdclk(dev_priv); + intel_cdclk_uninit(dev_priv); /* * 4. Disable Power Well 1 (PG1). @@ -4055,7 +4061,7 @@ void intel_power_domains_init_hw(struct drm_i915_private *i915, bool resume) power_domains->initializing = true; - if (IS_ICELAKE(i915)) { + if (INTEL_GEN(i915) >= 11) { icl_display_core_init(i915, resume); } else if (IS_CANNONLAKE(i915)) { cnl_display_core_init(i915, resume); @@ -4203,7 +4209,7 @@ void intel_power_domains_suspend(struct drm_i915_private *i915, intel_power_domains_verify_state(i915); } - if (IS_ICELAKE(i915)) + if (INTEL_GEN(i915) >= 11) icl_display_core_uninit(i915); else if (IS_CANNONLAKE(i915)) cnl_display_core_uninit(i915); diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 68f497493d43..0e3d91d9ef13 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -25,16 +25,23 @@ * Authors: * Eric Anholt <eric@anholt.net> */ -#include <linux/i2c.h> -#include <linux/slab.h> + #include <linux/delay.h> #include <linux/export.h> +#include <linux/i2c.h> +#include <linux/slab.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_edid.h> -#include "intel_drv.h" #include <drm/i915_drm.h> + #include "i915_drv.h" +#include "intel_connector.h" +#include "intel_drv.h" +#include "intel_hdmi.h" +#include "intel_panel.h" +#include "intel_sdvo.h" #include "intel_sdvo_regs.h" #define SDVO_TMDS_MASK (SDVO_OUTPUT_TMDS0 | SDVO_OUTPUT_TMDS1) diff --git a/drivers/gpu/drm/i915/intel_sdvo.h b/drivers/gpu/drm/i915/intel_sdvo.h new file mode 100644 index 000000000000..c9e05bcdd141 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_sdvo.h @@ -0,0 +1,23 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_SDVO_H__ +#define __INTEL_SDVO_H__ + +#include <linux/types.h> + +#include <drm/i915_drm.h> + +#include "i915_reg.h" + +struct drm_i915_private; +enum pipe; + +bool intel_sdvo_port_enabled(struct drm_i915_private *dev_priv, + i915_reg_t sdvo_reg, enum pipe *pipe); +bool intel_sdvo_init(struct drm_i915_private *dev_priv, + i915_reg_t reg, enum port port); + +#endif /* __INTEL_SDVO_H__ */ diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c index 65de7387bf1b..2913e89280d7 100644 --- a/drivers/gpu/drm/i915/intel_sprite.c +++ b/drivers/gpu/drm/i915/intel_sprite.c @@ -29,17 +29,23 @@ * registers; newer ones are much simpler and we can use the new DRM plane * support. */ + +#include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> +#include <drm/drm_color_mgmt.h> #include <drm/drm_crtc.h> #include <drm/drm_fourcc.h> -#include <drm/drm_rect.h> -#include <drm/drm_atomic.h> #include <drm/drm_plane_helper.h> -#include "intel_drv.h" -#include "intel_frontbuffer.h" +#include <drm/drm_rect.h> #include <drm/i915_drm.h> + #include "i915_drv.h" -#include <drm/drm_color_mgmt.h> +#include "intel_atomic_plane.h" +#include "intel_drv.h" +#include "intel_frontbuffer.h" +#include "intel_pm.h" +#include "intel_psr.h" +#include "intel_sprite.h" bool is_planar_yuv_format(u32 pixelformat) { diff --git a/drivers/gpu/drm/i915/intel_sprite.h b/drivers/gpu/drm/i915/intel_sprite.h new file mode 100644 index 000000000000..84be8686be16 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_sprite.h @@ -0,0 +1,55 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_SPRITE_H__ +#define __INTEL_SPRITE_H__ + +#include <linux/types.h> + +#include "i915_drv.h" +#include "intel_display.h" + +struct drm_device; +struct drm_display_mode; +struct drm_file; +struct drm_i915_private; +struct intel_crtc_state; +struct intel_plane_state; + +bool is_planar_yuv_format(u32 pixelformat); +int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode, + int usecs); +struct intel_plane *intel_sprite_plane_create(struct drm_i915_private *dev_priv, + enum pipe pipe, int plane); +int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); +void intel_pipe_update_start(const struct intel_crtc_state *new_crtc_state); +void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state); +int intel_plane_check_stride(const struct intel_plane_state *plane_state); +int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state); +int chv_plane_check_rotation(const struct intel_plane_state *plane_state); +struct intel_plane * +skl_universal_plane_create(struct drm_i915_private *dev_priv, + enum pipe pipe, enum plane_id plane_id); + +static inline bool icl_is_nv12_y_plane(enum plane_id id) +{ + /* Don't need to do a gen check, these planes are only available on gen11 */ + if (id == PLANE_SPRITE4 || id == PLANE_SPRITE5) + return true; + + return false; +} + +static inline bool icl_is_hdr_plane(struct drm_i915_private *dev_priv, + enum plane_id plane_id) +{ + if (INTEL_GEN(dev_priv) < 11) + return false; + + return plane_id < PLANE_SPRITE2; +} + +#endif /* __INTEL_SPRITE_H__ */ diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c index 3924c4944e1f..5dbba33f4202 100644 --- a/drivers/gpu/drm/i915/intel_tv.c +++ b/drivers/gpu/drm/i915/intel_tv.c @@ -33,9 +33,12 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_edid.h> -#include "intel_drv.h" #include <drm/i915_drm.h> + #include "i915_drv.h" +#include "intel_connector.h" +#include "intel_drv.h" +#include "intel_tv.h" enum tv_margin { TV_MARGIN_LEFT, TV_MARGIN_TOP, diff --git a/drivers/gpu/drm/i915/intel_tv.h b/drivers/gpu/drm/i915/intel_tv.h new file mode 100644 index 000000000000..44518575ec5c --- /dev/null +++ b/drivers/gpu/drm/i915/intel_tv.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_TV_H__ +#define __INTEL_TV_H__ + +struct drm_i915_private; + +void intel_tv_init(struct drm_i915_private *dev_priv); + +#endif /* __INTEL_TV_H__ */ diff --git a/drivers/gpu/drm/i915/intel_uc.c b/drivers/gpu/drm/i915/intel_uc.c index 2d360d53757f..25b80ffe71ad 100644 --- a/drivers/gpu/drm/i915/intel_uc.c +++ b/drivers/gpu/drm/i915/intel_uc.c @@ -375,7 +375,7 @@ int intel_uc_init_hw(struct drm_i915_private *i915) intel_guc_init_params(guc); ret = intel_guc_fw_upload(guc); - if (ret == 0 || ret != -ETIMEDOUT) + if (ret == 0) break; DRM_DEBUG_DRIVER("GuC fw load failed: %d; will reset and " diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 5c80704bf283..d1d51e1121e2 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -21,12 +21,13 @@ * IN THE SOFTWARE. */ +#include <linux/pm_runtime.h> +#include <asm/iosf_mbi.h> + #include "i915_drv.h" -#include "intel_drv.h" #include "i915_vgpu.h" - -#include <asm/iosf_mbi.h> -#include <linux/pm_runtime.h> +#include "intel_drv.h" +#include "intel_pm.h" #define FORCEWAKE_ACK_TIMEOUT_MS 50 #define GT_FIFO_TIMEOUT_MS 10 @@ -420,51 +421,6 @@ intel_uncore_forcewake_reset(struct intel_uncore *uncore) return fw; /* track the lost user forcewake domains */ } -static u64 gen9_edram_size(struct drm_i915_private *dev_priv) -{ - const unsigned int ways[8] = { 4, 8, 12, 16, 16, 16, 16, 16 }; - const unsigned int sets[4] = { 1, 1, 2, 2 }; - const u32 cap = dev_priv->edram_cap; - - return EDRAM_NUM_BANKS(cap) * - ways[EDRAM_WAYS_IDX(cap)] * - sets[EDRAM_SETS_IDX(cap)] * - 1024 * 1024; -} - -u64 intel_uncore_edram_size(struct drm_i915_private *dev_priv) -{ - if (!HAS_EDRAM(dev_priv)) - return 0; - - /* The needed capability bits for size calculation - * are not there with pre gen9 so return 128MB always. - */ - if (INTEL_GEN(dev_priv) < 9) - return 128 * 1024 * 1024; - - return gen9_edram_size(dev_priv); -} - -static void intel_uncore_edram_detect(struct drm_i915_private *dev_priv) -{ - if (IS_HASWELL(dev_priv) || - IS_BROADWELL(dev_priv) || - INTEL_GEN(dev_priv) >= 9) { - dev_priv->edram_cap = __raw_uncore_read32(&dev_priv->uncore, - HSW_EDRAM_CAP); - - /* NB: We can't write IDICR yet because we do not have gt funcs - * set up */ - } else { - dev_priv->edram_cap = 0; - } - - if (HAS_EDRAM(dev_priv)) - DRM_INFO("Found %lluMB of eDRAM\n", - intel_uncore_edram_size(dev_priv) / (1024 * 1024)); -} - static bool fpga_check_for_unclaimed_mmio(struct intel_uncore *uncore) { @@ -1569,8 +1525,12 @@ static void uncore_mmio_cleanup(struct intel_uncore *uncore) pci_iounmap(pdev, uncore->regs); } +void intel_uncore_init_early(struct intel_uncore *uncore) +{ + spin_lock_init(&uncore->lock); +} -int intel_uncore_init(struct intel_uncore *uncore) +int intel_uncore_init_mmio(struct intel_uncore *uncore) { struct drm_i915_private *i915 = uncore_to_i915(uncore); int ret; @@ -1584,7 +1544,6 @@ int intel_uncore_init(struct intel_uncore *uncore) if (INTEL_GEN(i915) > 5 && !intel_vgpu_active(i915)) uncore->flags |= UNCORE_HAS_FORCEWAKE; - intel_uncore_edram_detect(i915); intel_uncore_fw_domains_init(uncore); __intel_uncore_early_sanitize(uncore, 0); @@ -1650,7 +1609,7 @@ int intel_uncore_init(struct intel_uncore *uncore) * the forcewake domains. Prune them, to make sure they only reference existing * engines. */ -void intel_uncore_prune(struct intel_uncore *uncore) +void intel_uncore_prune_mmio_domains(struct intel_uncore *uncore) { struct drm_i915_private *i915 = uncore_to_i915(uncore); @@ -1681,7 +1640,7 @@ void intel_uncore_prune(struct intel_uncore *uncore) } } -void intel_uncore_fini(struct intel_uncore *uncore) +void intel_uncore_fini_mmio(struct intel_uncore *uncore) { /* Paranoia: make sure we have disabled everything before we exit. */ intel_uncore_sanitize(uncore_to_i915(uncore)); diff --git a/drivers/gpu/drm/i915/intel_uncore.h b/drivers/gpu/drm/i915/intel_uncore.h index 50d226f68753..d6af3de70121 100644 --- a/drivers/gpu/drm/i915/intel_uncore.h +++ b/drivers/gpu/drm/i915/intel_uncore.h @@ -178,16 +178,16 @@ intel_uncore_has_fifo(const struct intel_uncore *uncore) } void intel_uncore_sanitize(struct drm_i915_private *dev_priv); -int intel_uncore_init(struct intel_uncore *uncore); -void intel_uncore_prune(struct intel_uncore *uncore); +void intel_uncore_init_early(struct intel_uncore *uncore); +int intel_uncore_init_mmio(struct intel_uncore *uncore); +void intel_uncore_prune_mmio_domains(struct intel_uncore *uncore); bool intel_uncore_unclaimed_mmio(struct intel_uncore *uncore); bool intel_uncore_arm_unclaimed_mmio_detection(struct intel_uncore *uncore); -void intel_uncore_fini(struct intel_uncore *uncore); +void intel_uncore_fini_mmio(struct intel_uncore *uncore); void intel_uncore_suspend(struct intel_uncore *uncore); void intel_uncore_resume_early(struct intel_uncore *uncore); void intel_uncore_runtime_resume(struct intel_uncore *uncore); -u64 intel_uncore_edram_size(struct drm_i915_private *dev_priv); void assert_forcewakes_inactive(struct intel_uncore *uncore); void assert_forcewakes_active(struct intel_uncore *uncore, enum forcewake_domains fw_domains); @@ -369,11 +369,26 @@ intel_uncore_read64_2x32(struct intel_uncore *uncore, #define intel_uncore_write64_fw(...) __raw_uncore_write64(__VA_ARGS__) #define intel_uncore_posting_read_fw(...) ((void)intel_uncore_read_fw(__VA_ARGS__)) -static inline void intel_uncore_rmw_or_fw(struct intel_uncore *uncore, - i915_reg_t reg, u32 or_val) +static inline void intel_uncore_rmw(struct intel_uncore *uncore, + i915_reg_t reg, u32 clear, u32 set) { - intel_uncore_write_fw(uncore, reg, - intel_uncore_read_fw(uncore, reg) | or_val); + u32 val; + + val = intel_uncore_read(uncore, reg); + val &= ~clear; + val |= set; + intel_uncore_write(uncore, reg, val); +} + +static inline void intel_uncore_rmw_fw(struct intel_uncore *uncore, + i915_reg_t reg, u32 clear, u32 set) +{ + u32 val; + + val = intel_uncore_read_fw(uncore, reg); + val &= ~clear; + val |= set; + intel_uncore_write_fw(uncore, reg, val); } #define raw_reg_read(base, reg) \ diff --git a/drivers/gpu/drm/i915/intel_workarounds.c b/drivers/gpu/drm/i915/intel_workarounds.c index a04dbc58ec1c..ccaf63679435 100644 --- a/drivers/gpu/drm/i915/intel_workarounds.c +++ b/drivers/gpu/drm/i915/intel_workarounds.c @@ -569,7 +569,7 @@ void intel_engine_init_ctx_wa(struct intel_engine_cs *engine) wa_init_start(wal, "context"); - if (IS_ICELAKE(i915)) + if (IS_GEN(i915, 11)) icl_ctx_workarounds_init(engine); else if (IS_CANNONLAKE(i915)) cnl_ctx_workarounds_init(engine); @@ -729,9 +729,9 @@ cfl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal) } static void -wa_init_mcr(struct drm_i915_private *dev_priv, struct i915_wa_list *wal) +wa_init_mcr(struct drm_i915_private *i915, struct i915_wa_list *wal) { - const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu; + const struct sseu_dev_info *sseu = &RUNTIME_INFO(i915)->sseu; u32 mcr_slice_subslice_mask; /* @@ -747,14 +747,15 @@ wa_init_mcr(struct drm_i915_private *dev_priv, struct i915_wa_list *wal) * something more complex that requires checking the range of every * MMIO read). */ - if (INTEL_GEN(dev_priv) >= 10 && + if (INTEL_GEN(i915) >= 10 && is_power_of_2(sseu->slice_mask)) { /* * read FUSE3 for enabled L3 Bank IDs, if L3 Bank matches * enabled subslice, no need to redirect MCR packet */ u32 slice = fls(sseu->slice_mask); - u32 fuse3 = I915_READ(GEN10_MIRROR_FUSE3); + u32 fuse3 = + intel_uncore_read(&i915->uncore, GEN10_MIRROR_FUSE3); u8 ss_mask = sseu->subslice_mask[slice]; u8 enabled_mask = (ss_mask | ss_mask >> @@ -768,7 +769,7 @@ wa_init_mcr(struct drm_i915_private *dev_priv, struct i915_wa_list *wal) WARN_ON((enabled_mask & disabled_mask) != enabled_mask); } - if (INTEL_GEN(dev_priv) >= 11) + if (INTEL_GEN(i915) >= 11) mcr_slice_subslice_mask = GEN11_MCR_SLICE_MASK | GEN11_MCR_SUBSLICE_MASK; else @@ -788,7 +789,7 @@ wa_init_mcr(struct drm_i915_private *dev_priv, struct i915_wa_list *wal) wa_write_masked_or(wal, GEN8_MCR_SELECTOR, mcr_slice_subslice_mask, - intel_calculate_mcr_s_ss_select(dev_priv)); + intel_calculate_mcr_s_ss_select(i915)); } static void @@ -867,7 +868,7 @@ icl_gt_workarounds_init(struct drm_i915_private *i915, struct i915_wa_list *wal) static void gt_init_workarounds(struct drm_i915_private *i915, struct i915_wa_list *wal) { - if (IS_ICELAKE(i915)) + if (IS_GEN(i915, 11)) icl_gt_workarounds_init(i915, wal); else if (IS_CANNONLAKE(i915)) cnl_gt_workarounds_init(i915, wal); @@ -897,15 +898,14 @@ void intel_gt_init_workarounds(struct drm_i915_private *i915) } static enum forcewake_domains -wal_get_fw_for_rmw(struct drm_i915_private *dev_priv, - const struct i915_wa_list *wal) +wal_get_fw_for_rmw(struct intel_uncore *uncore, const struct i915_wa_list *wal) { enum forcewake_domains fw = 0; struct i915_wa *wa; unsigned int i; for (i = 0, wa = wal->list; i < wal->count; i++, wa++) - fw |= intel_uncore_forcewake_for_reg(&dev_priv->uncore, + fw |= intel_uncore_forcewake_for_reg(uncore, wa->reg, FW_REG_READ | FW_REG_WRITE); @@ -914,7 +914,7 @@ wal_get_fw_for_rmw(struct drm_i915_private *dev_priv, } static void -wa_list_apply(struct drm_i915_private *dev_priv, const struct i915_wa_list *wal) +wa_list_apply(struct intel_uncore *uncore, const struct i915_wa_list *wal) { enum forcewake_domains fw; unsigned long flags; @@ -924,27 +924,22 @@ wa_list_apply(struct drm_i915_private *dev_priv, const struct i915_wa_list *wal) if (!wal->count) return; - fw = wal_get_fw_for_rmw(dev_priv, wal); + fw = wal_get_fw_for_rmw(uncore, wal); - spin_lock_irqsave(&dev_priv->uncore.lock, flags); - intel_uncore_forcewake_get__locked(&dev_priv->uncore, fw); + spin_lock_irqsave(&uncore->lock, flags); + intel_uncore_forcewake_get__locked(uncore, fw); for (i = 0, wa = wal->list; i < wal->count; i++, wa++) { - u32 val = I915_READ_FW(wa->reg); - - val &= ~wa->mask; - val |= wa->val; - - I915_WRITE_FW(wa->reg, val); + intel_uncore_rmw_fw(uncore, wa->reg, wa->mask, wa->val); } - intel_uncore_forcewake_put__locked(&dev_priv->uncore, fw); - spin_unlock_irqrestore(&dev_priv->uncore.lock, flags); + intel_uncore_forcewake_put__locked(uncore, fw); + spin_unlock_irqrestore(&uncore->lock, flags); } -void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv) +void intel_gt_apply_workarounds(struct drm_i915_private *i915) { - wa_list_apply(dev_priv, &dev_priv->gt_wa_list); + wa_list_apply(&i915->uncore, &i915->gt_wa_list); } static bool @@ -961,7 +956,7 @@ wa_verify(const struct i915_wa *wa, u32 cur, const char *name, const char *from) return true; } -static bool wa_list_verify(struct drm_i915_private *dev_priv, +static bool wa_list_verify(struct intel_uncore *uncore, const struct i915_wa_list *wal, const char *from) { @@ -970,15 +965,17 @@ static bool wa_list_verify(struct drm_i915_private *dev_priv, bool ok = true; for (i = 0, wa = wal->list; i < wal->count; i++, wa++) - ok &= wa_verify(wa, I915_READ(wa->reg), wal->name, from); + ok &= wa_verify(wa, + intel_uncore_read(uncore, wa->reg), + wal->name, from); return ok; } -bool intel_gt_verify_workarounds(struct drm_i915_private *dev_priv, +bool intel_gt_verify_workarounds(struct drm_i915_private *i915, const char *from) { - return wa_list_verify(dev_priv, &dev_priv->gt_wa_list, from); + return wa_list_verify(&i915->uncore, &i915->gt_wa_list, from); } static void @@ -1064,7 +1061,7 @@ void intel_engine_init_whitelist(struct intel_engine_cs *engine) wa_init_start(w, "whitelist"); - if (IS_ICELAKE(i915)) + if (IS_GEN(i915, 11)) icl_whitelist_build(w); else if (IS_CANNONLAKE(i915)) cnl_whitelist_build(w); @@ -1088,8 +1085,8 @@ void intel_engine_init_whitelist(struct intel_engine_cs *engine) void intel_engine_apply_whitelist(struct intel_engine_cs *engine) { - struct drm_i915_private *dev_priv = engine->i915; const struct i915_wa_list *wal = &engine->whitelist; + struct intel_uncore *uncore = engine->uncore; const u32 base = engine->mmio_base; struct i915_wa *wa; unsigned int i; @@ -1098,13 +1095,15 @@ void intel_engine_apply_whitelist(struct intel_engine_cs *engine) return; for (i = 0, wa = wal->list; i < wal->count; i++, wa++) - I915_WRITE(RING_FORCE_TO_NONPRIV(base, i), - i915_mmio_reg_offset(wa->reg)); + intel_uncore_write(uncore, + RING_FORCE_TO_NONPRIV(base, i), + i915_mmio_reg_offset(wa->reg)); /* And clear the rest just in case of garbage */ for (; i < RING_MAX_NONPRIV_SLOTS; i++) - I915_WRITE(RING_FORCE_TO_NONPRIV(base, i), - i915_mmio_reg_offset(RING_NOPID(base))); + intel_uncore_write(uncore, + RING_FORCE_TO_NONPRIV(base, i), + i915_mmio_reg_offset(RING_NOPID(base))); } static void @@ -1112,7 +1111,7 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) { struct drm_i915_private *i915 = engine->i915; - if (IS_ICELAKE(i915)) { + if (IS_GEN(i915, 11)) { /* This is not an Wa. Enable for better image quality */ wa_masked_en(wal, _3D_CHICKEN3, @@ -1253,7 +1252,7 @@ void intel_engine_init_workarounds(struct intel_engine_cs *engine) void intel_engine_apply_workarounds(struct intel_engine_cs *engine) { - wa_list_apply(engine->i915, &engine->wa_list); + wa_list_apply(engine->uncore, &engine->wa_list); } #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) diff --git a/drivers/gpu/drm/i915/intel_workarounds.h b/drivers/gpu/drm/i915/intel_workarounds.h index a1bf51c611a9..34eee5ec511e 100644 --- a/drivers/gpu/drm/i915/intel_workarounds.h +++ b/drivers/gpu/drm/i915/intel_workarounds.h @@ -20,9 +20,9 @@ static inline void intel_wa_list_free(struct i915_wa_list *wal) void intel_engine_init_ctx_wa(struct intel_engine_cs *engine); int intel_engine_emit_ctx_wa(struct i915_request *rq); -void intel_gt_init_workarounds(struct drm_i915_private *dev_priv); -void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv); -bool intel_gt_verify_workarounds(struct drm_i915_private *dev_priv, +void intel_gt_init_workarounds(struct drm_i915_private *i915); +void intel_gt_apply_workarounds(struct drm_i915_private *i915); +bool intel_gt_verify_workarounds(struct drm_i915_private *i915, const char *from); void intel_engine_init_whitelist(struct intel_engine_cs *engine); diff --git a/drivers/gpu/drm/i915/selftests/i915_gem.c b/drivers/gpu/drm/i915/selftests/i915_gem.c index 50bb7bbd26d3..6fd70d326468 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem.c @@ -215,5 +215,8 @@ int i915_gem_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_gem_hibernate), }; + if (i915_terminally_wedged(i915)) + return 0; + return i915_subtests(tests, i915); } diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/selftests/i915_gem_context.c index 45f73b8b4e6d..4e1b6efc6b22 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_context.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_context.c @@ -1594,10 +1594,10 @@ out_unlock: } static __maybe_unused const char * -__engine_name(struct drm_i915_private *i915, unsigned int engines) +__engine_name(struct drm_i915_private *i915, intel_engine_mask_t engines) { struct intel_engine_cs *engine; - unsigned int tmp; + intel_engine_mask_t tmp; if (engines == ALL_ENGINES) return "all"; @@ -1610,10 +1610,10 @@ __engine_name(struct drm_i915_private *i915, unsigned int engines) static int __igt_switch_to_kernel_context(struct drm_i915_private *i915, struct i915_gem_context *ctx, - unsigned int engines) + intel_engine_mask_t engines) { struct intel_engine_cs *engine; - unsigned int tmp; + intel_engine_mask_t tmp; int pass; GEM_TRACE("Testing %s\n", __engine_name(i915, engines)); diff --git a/drivers/gpu/drm/i915/selftests/i915_timeline.c b/drivers/gpu/drm/i915/selftests/i915_timeline.c index 8e7bcaa1eb66..bd96afcadfe7 100644 --- a/drivers/gpu/drm/i915/selftests/i915_timeline.c +++ b/drivers/gpu/drm/i915/selftests/i915_timeline.c @@ -838,5 +838,8 @@ int i915_timeline_live_selftests(struct drm_i915_private *i915) SUBTEST(live_hwsp_wrap), }; + if (i915_terminally_wedged(i915)) + return 0; + return i915_subtests(tests, i915); } diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c index 76b4fa150f2e..050bd1e19e02 100644 --- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c +++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c @@ -1124,7 +1124,8 @@ static int igt_reset_engines(void *arg) return 0; } -static u32 fake_hangcheck(struct drm_i915_private *i915, u32 mask) +static u32 fake_hangcheck(struct drm_i915_private *i915, + intel_engine_mask_t mask) { u32 count = i915_reset_count(&i915->gpu_error); diff --git a/drivers/gpu/drm/i915/selftests/intel_lrc.c b/drivers/gpu/drm/i915/selftests/intel_lrc.c index 0d3cae564db8..fbee030db940 100644 --- a/drivers/gpu/drm/i915/selftests/intel_lrc.c +++ b/drivers/gpu/drm/i915/selftests/intel_lrc.c @@ -76,6 +76,185 @@ err_unlock: return err; } +static int live_busywait_preempt(void *arg) +{ + struct drm_i915_private *i915 = arg; + struct i915_gem_context *ctx_hi, *ctx_lo; + struct intel_engine_cs *engine; + struct drm_i915_gem_object *obj; + struct i915_vma *vma; + enum intel_engine_id id; + intel_wakeref_t wakeref; + int err = -ENOMEM; + u32 *map; + + /* + * Verify that even without HAS_LOGICAL_RING_PREEMPTION, we can + * preempt the busywaits used to synchronise between rings. + */ + + mutex_lock(&i915->drm.struct_mutex); + wakeref = intel_runtime_pm_get(i915); + + ctx_hi = kernel_context(i915); + if (!ctx_hi) + goto err_unlock; + ctx_hi->sched.priority = INT_MAX; + + ctx_lo = kernel_context(i915); + if (!ctx_lo) + goto err_ctx_hi; + ctx_lo->sched.priority = INT_MIN; + + obj = i915_gem_object_create_internal(i915, PAGE_SIZE); + if (IS_ERR(obj)) { + err = PTR_ERR(obj); + goto err_ctx_lo; + } + + map = i915_gem_object_pin_map(obj, I915_MAP_WC); + if (IS_ERR(map)) { + err = PTR_ERR(map); + goto err_obj; + } + + vma = i915_vma_instance(obj, &i915->ggtt.vm, NULL); + if (IS_ERR(vma)) { + err = PTR_ERR(vma); + goto err_map; + } + + err = i915_vma_pin(vma, 0, 0, PIN_GLOBAL); + if (err) + goto err_map; + + for_each_engine(engine, i915, id) { + struct i915_request *lo, *hi; + struct igt_live_test t; + u32 *cs; + + if (!intel_engine_can_store_dword(engine)) + continue; + + if (igt_live_test_begin(&t, i915, __func__, engine->name)) { + err = -EIO; + goto err_vma; + } + + /* + * We create two requests. The low priority request + * busywaits on a semaphore (inside the ringbuffer where + * is should be preemptible) and the high priority requests + * uses a MI_STORE_DWORD_IMM to update the semaphore value + * allowing the first request to complete. If preemption + * fails, we hang instead. + */ + + lo = i915_request_alloc(engine, ctx_lo); + if (IS_ERR(lo)) { + err = PTR_ERR(lo); + goto err_vma; + } + + cs = intel_ring_begin(lo, 8); + if (IS_ERR(cs)) { + err = PTR_ERR(cs); + i915_request_add(lo); + goto err_vma; + } + + *cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT; + *cs++ = i915_ggtt_offset(vma); + *cs++ = 0; + *cs++ = 1; + + /* XXX Do we need a flush + invalidate here? */ + + *cs++ = MI_SEMAPHORE_WAIT | + MI_SEMAPHORE_GLOBAL_GTT | + MI_SEMAPHORE_POLL | + MI_SEMAPHORE_SAD_EQ_SDD; + *cs++ = 0; + *cs++ = i915_ggtt_offset(vma); + *cs++ = 0; + + intel_ring_advance(lo, cs); + i915_request_add(lo); + + if (wait_for(READ_ONCE(*map), 10)) { + err = -ETIMEDOUT; + goto err_vma; + } + + /* Low priority request should be busywaiting now */ + if (i915_request_wait(lo, I915_WAIT_LOCKED, 1) != -ETIME) { + pr_err("%s: Busywaiting request did not!\n", + engine->name); + err = -EIO; + goto err_vma; + } + + hi = i915_request_alloc(engine, ctx_hi); + if (IS_ERR(hi)) { + err = PTR_ERR(hi); + goto err_vma; + } + + cs = intel_ring_begin(hi, 4); + if (IS_ERR(cs)) { + err = PTR_ERR(cs); + i915_request_add(hi); + goto err_vma; + } + + *cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT; + *cs++ = i915_ggtt_offset(vma); + *cs++ = 0; + *cs++ = 0; + + intel_ring_advance(hi, cs); + i915_request_add(hi); + + if (i915_request_wait(lo, I915_WAIT_LOCKED, HZ / 5) < 0) { + struct drm_printer p = drm_info_printer(i915->drm.dev); + + pr_err("%s: Failed to preempt semaphore busywait!\n", + engine->name); + + intel_engine_dump(engine, &p, "%s\n", engine->name); + GEM_TRACE_DUMP(); + + i915_gem_set_wedged(i915); + err = -EIO; + goto err_vma; + } + GEM_BUG_ON(READ_ONCE(*map)); + + if (igt_live_test_end(&t)) { + err = -EIO; + goto err_vma; + } + } + + err = 0; +err_vma: + i915_vma_unpin(vma); +err_map: + i915_gem_object_unpin_map(obj); +err_obj: + i915_gem_object_put(obj); +err_ctx_lo: + kernel_context_close(ctx_lo); +err_ctx_hi: + kernel_context_close(ctx_hi); +err_unlock: + if (igt_flush_test(i915, I915_WAIT_LOCKED)) + err = -EIO; + intel_runtime_pm_put(i915, wakeref); + mutex_unlock(&i915->drm.struct_mutex); + return err; +} + static int live_preempt(void *arg) { struct drm_i915_private *i915 = arg; @@ -1127,6 +1306,7 @@ int intel_execlists_live_selftests(struct drm_i915_private *i915) { static const struct i915_subtest tests[] = { SUBTEST(live_sanitycheck), + SUBTEST(live_busywait_preempt), SUBTEST(live_preempt), SUBTEST(live_late_preempt), SUBTEST(live_suppress_self_preempt), diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c index ee0bc91f7664..e0d7ebecb215 100644 --- a/drivers/gpu/drm/i915/selftests/intel_uncore.c +++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c @@ -155,6 +155,17 @@ static int live_forcewake_ops(void *arg) return 0; } + /* + * Not quite as reliable across the gen as one would hope. + * + * Either our theory of operation is incorrect, or there remain + * external parties interfering with the powerwells. + * + * https://bugs.freedesktop.org/show_bug.cgi?id=110210 + */ + if (!IS_ENABLED(CONFIG_DRM_I915_SELFTEST_BROKEN)) + return 0; + /* We have to pick carefully to get the exact behaviour we need */ for (r = registers; r->name; r++) if (r->platforms & INTEL_INFO(i915)->gen_mask) diff --git a/drivers/gpu/drm/i915/selftests/intel_workarounds.c b/drivers/gpu/drm/i915/selftests/intel_workarounds.c index 3baed59008d7..567b6f8dae86 100644 --- a/drivers/gpu/drm/i915/selftests/intel_workarounds.c +++ b/drivers/gpu/drm/i915/selftests/intel_workarounds.c @@ -750,10 +750,11 @@ static bool verify_gt_engine_wa(struct drm_i915_private *i915, enum intel_engine_id id; bool ok = true; - ok &= wa_list_verify(i915, &lists->gt_wa_list, str); + ok &= wa_list_verify(&i915->uncore, &lists->gt_wa_list, str); for_each_engine(engine, i915, id) - ok &= wa_list_verify(i915, &lists->engine[id].wa_list, str); + ok &= wa_list_verify(engine->uncore, + &lists->engine[id].wa_list, str); return ok; } diff --git a/drivers/gpu/drm/i915/selftests/mock_timeline.c b/drivers/gpu/drm/i915/selftests/mock_timeline.c index 416d85233263..e084476469ef 100644 --- a/drivers/gpu/drm/i915/selftests/mock_timeline.c +++ b/drivers/gpu/drm/i915/selftests/mock_timeline.c @@ -16,7 +16,6 @@ void mock_timeline_init(struct i915_timeline *timeline, u64 context) spin_lock_init(&timeline->lock); mutex_init(&timeline->mutex); - INIT_ACTIVE_REQUEST(&timeline->barrier); INIT_ACTIVE_REQUEST(&timeline->last_request); INIT_LIST_HEAD(&timeline->requests); diff --git a/drivers/gpu/drm/i915/test_i915_active_types_standalone.c b/drivers/gpu/drm/i915/test_i915_active_types_standalone.c deleted file mode 100644 index 144ebd153e57..000000000000 --- a/drivers/gpu/drm/i915/test_i915_active_types_standalone.c +++ /dev/null @@ -1,7 +0,0 @@ -/* - * SPDX-License-Identifier: MIT - * - * Copyright © 2019 Intel Corporation - */ - -#include "i915_active_types.h" diff --git a/drivers/gpu/drm/i915/test_i915_gem_context_types_standalone.c b/drivers/gpu/drm/i915/test_i915_gem_context_types_standalone.c deleted file mode 100644 index 4e4da4860bc2..000000000000 --- a/drivers/gpu/drm/i915/test_i915_gem_context_types_standalone.c +++ /dev/null @@ -1,7 +0,0 @@ -/* - * SPDX-License-Identifier: MIT - * - * Copyright © 2019 Intel Corporation - */ - -#include "i915_gem_context_types.h" diff --git a/drivers/gpu/drm/i915/test_i915_timeline_types_standalone.c b/drivers/gpu/drm/i915/test_i915_timeline_types_standalone.c deleted file mode 100644 index f58e148e8946..000000000000 --- a/drivers/gpu/drm/i915/test_i915_timeline_types_standalone.c +++ /dev/null @@ -1,7 +0,0 @@ -/* - * SPDX-License-Identifier: MIT - * - * Copyright © 2019 Intel Corporation - */ - -#include "i915_timeline_types.h" diff --git a/drivers/gpu/drm/i915/test_intel_context_types_standalone.c b/drivers/gpu/drm/i915/test_intel_context_types_standalone.c deleted file mode 100644 index b39e3c4e6551..000000000000 --- a/drivers/gpu/drm/i915/test_intel_context_types_standalone.c +++ /dev/null @@ -1,7 +0,0 @@ -/* - * SPDX-License-Identifier: MIT - * - * Copyright © 2019 Intel Corporation - */ - -#include "intel_context_types.h" diff --git a/drivers/gpu/drm/i915/test_intel_engine_types_standalone.c b/drivers/gpu/drm/i915/test_intel_engine_types_standalone.c deleted file mode 100644 index d05e4cdcbcf9..000000000000 --- a/drivers/gpu/drm/i915/test_intel_engine_types_standalone.c +++ /dev/null @@ -1,7 +0,0 @@ -/* - * SPDX-License-Identifier: MIT - * - * Copyright © 2019 Intel Corporation - */ - -#include "intel_engine_types.h" diff --git a/drivers/gpu/drm/i915/test_intel_workarounds_types_standalone.c b/drivers/gpu/drm/i915/test_intel_workarounds_types_standalone.c deleted file mode 100644 index 4f658bb00825..000000000000 --- a/drivers/gpu/drm/i915/test_intel_workarounds_types_standalone.c +++ /dev/null @@ -1,7 +0,0 @@ -/* - * SPDX-License-Identifier: MIT - * - * Copyright © 2019 Intel Corporation - */ - -#include "intel_workarounds_types.h" diff --git a/drivers/gpu/drm/i915/vlv_dsi.c b/drivers/gpu/drm/i915/vlv_dsi.c index 0a950c976bbb..e0b1ec821960 100644 --- a/drivers/gpu/drm/i915/vlv_dsi.c +++ b/drivers/gpu/drm/i915/vlv_dsi.c @@ -23,16 +23,20 @@ * Author: Jani Nikula <jani.nikula@intel.com> */ +#include <linux/gpio/consumer.h> +#include <linux/slab.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_edid.h> -#include <drm/i915_drm.h> #include <drm/drm_mipi_dsi.h> -#include <linux/slab.h> -#include <linux/gpio/consumer.h> +#include <drm/i915_drm.h> + #include "i915_drv.h" +#include "intel_connector.h" #include "intel_drv.h" #include "intel_dsi.h" +#include "intel_panel.h" /* return pixels in terms of txbyteclkhs */ static u16 txbyteclkhs(u16 pixels, int bpp, int lane_count, @@ -256,6 +260,28 @@ static void band_gap_reset(struct drm_i915_private *dev_priv) mutex_unlock(&dev_priv->sb_lock); } +static int bdw_get_pipemisc_bpp(struct intel_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + u32 tmp; + + tmp = I915_READ(PIPEMISC(crtc->pipe)); + + switch (tmp & PIPEMISC_DITHER_BPC_MASK) { + case PIPEMISC_DITHER_6_BPC: + return 18; + case PIPEMISC_DITHER_8_BPC: + return 24; + case PIPEMISC_DITHER_10_BPC: + return 30; + case PIPEMISC_DITHER_12_BPC: + return 36; + default: + MISSING_CASE(tmp); + return 0; + } +} + static int intel_dsi_compute_config(struct intel_encoder *encoder, struct intel_crtc_state *pipe_config, struct drm_connector_state *conn_state) @@ -1082,6 +1108,8 @@ static void bxt_dsi_get_pipe_config(struct intel_encoder *encoder, bpp = mipi_dsi_pixel_format_to_bpp( pixel_format_from_register_bits(fmt)); + pipe_config->pipe_bpp = bdw_get_pipemisc_bpp(crtc); + /* Enable Frame time stamo based scanline reporting */ adjusted_mode->private_flags |= I915_MODE_FLAG_GET_SCANLINE_FROM_TIMESTAMP; diff --git a/drivers/gpu/drm/lima/lima_gem.c b/drivers/gpu/drm/lima/lima_gem.c index 1d69498bc17e..477c0f766663 100644 --- a/drivers/gpu/drm/lima/lima_gem.c +++ b/drivers/gpu/drm/lima/lima_gem.c @@ -145,40 +145,7 @@ static int lima_gem_sync_bo(struct lima_sched_task *task, struct lima_bo *bo, if (explicit) return 0; - /* implicit sync use bo fence in resv obj */ - if (write) { - unsigned nr_fences; - struct dma_fence **fences; - int i; - - err = reservation_object_get_fences_rcu( - bo->gem.resv, NULL, &nr_fences, &fences); - if (err || !nr_fences) - return err; - - for (i = 0; i < nr_fences; i++) { - err = lima_sched_task_add_dep(task, fences[i]); - if (err) - break; - } - - /* for error case free remaining fences */ - for ( ; i < nr_fences; i++) - dma_fence_put(fences[i]); - - kfree(fences); - } else { - struct dma_fence *fence; - - fence = reservation_object_get_excl_rcu(bo->gem.resv); - if (fence) { - err = lima_sched_task_add_dep(task, fence); - if (err) - dma_fence_put(fence); - } - } - - return err; + return drm_gem_fence_array_add_implicit(&task->deps, &bo->gem, write); } static int lima_gem_lock_bos(struct lima_bo **bos, u32 nr_bos, @@ -251,7 +218,7 @@ static int lima_gem_add_deps(struct drm_file *file, struct lima_submit *submit) if (err) return err; - err = lima_sched_task_add_dep(submit->task, fence); + err = drm_gem_fence_array_add(&submit->task->deps, fence); if (err) { dma_fence_put(fence); return err; diff --git a/drivers/gpu/drm/lima/lima_sched.c b/drivers/gpu/drm/lima/lima_sched.c index 97bd9c1deb87..d53bd45f8d96 100644 --- a/drivers/gpu/drm/lima/lima_sched.c +++ b/drivers/gpu/drm/lima/lima_sched.c @@ -3,6 +3,7 @@ #include <linux/kthread.h> #include <linux/slab.h> +#include <linux/xarray.h> #include "lima_drv.h" #include "lima_sched.h" @@ -126,19 +127,24 @@ int lima_sched_task_init(struct lima_sched_task *task, task->num_bos = num_bos; task->vm = lima_vm_get(vm); + + xa_init_flags(&task->deps, XA_FLAGS_ALLOC); + return 0; } void lima_sched_task_fini(struct lima_sched_task *task) { + struct dma_fence *fence; + unsigned long index; int i; drm_sched_job_cleanup(&task->base); - for (i = 0; i < task->num_dep; i++) - dma_fence_put(task->dep[i]); - - kfree(task->dep); + xa_for_each(&task->deps, index, fence) { + dma_fence_put(fence); + } + xa_destroy(&task->deps); if (task->bos) { for (i = 0; i < task->num_bos; i++) @@ -149,42 +155,6 @@ void lima_sched_task_fini(struct lima_sched_task *task) lima_vm_put(task->vm); } -int lima_sched_task_add_dep(struct lima_sched_task *task, struct dma_fence *fence) -{ - int i, new_dep = 4; - - /* same context's fence is definitly earlier then this task */ - if (fence->context == task->base.s_fence->finished.context) { - dma_fence_put(fence); - return 0; - } - - if (task->dep && task->num_dep == task->max_dep) - new_dep = task->max_dep * 2; - - if (task->max_dep < new_dep) { - void *dep = krealloc(task->dep, sizeof(*task->dep) * new_dep, GFP_KERNEL); - - if (!dep) - return -ENOMEM; - - task->max_dep = new_dep; - task->dep = dep; - } - - for (i = 0; i < task->num_dep; i++) { - if (task->dep[i]->context == fence->context && - dma_fence_is_later(fence, task->dep[i])) { - dma_fence_put(task->dep[i]); - task->dep[i] = fence; - return 0; - } - } - - task->dep[task->num_dep++] = fence; - return 0; -} - int lima_sched_context_init(struct lima_sched_pipe *pipe, struct lima_sched_context *context, atomic_t *guilty) @@ -213,21 +183,9 @@ static struct dma_fence *lima_sched_dependency(struct drm_sched_job *job, struct drm_sched_entity *entity) { struct lima_sched_task *task = to_lima_task(job); - int i; - - for (i = 0; i < task->num_dep; i++) { - struct dma_fence *fence = task->dep[i]; - - if (!task->dep[i]) - continue; - - task->dep[i] = NULL; - if (!dma_fence_is_signaled(fence)) - return fence; - - dma_fence_put(fence); - } + if (!xa_empty(&task->deps)) + return xa_erase(&task->deps, task->last_dep++); return NULL; } @@ -353,7 +311,7 @@ static void lima_sched_free_job(struct drm_sched_job *job) kmem_cache_free(pipe->task_slab, task); } -const struct drm_sched_backend_ops lima_sched_ops = { +static const struct drm_sched_backend_ops lima_sched_ops = { .dependency = lima_sched_dependency, .run_job = lima_sched_run_job, .timedout_job = lima_sched_timedout_job, diff --git a/drivers/gpu/drm/lima/lima_sched.h b/drivers/gpu/drm/lima/lima_sched.h index b017cfa7e327..928af91c1118 100644 --- a/drivers/gpu/drm/lima/lima_sched.h +++ b/drivers/gpu/drm/lima/lima_sched.h @@ -14,9 +14,8 @@ struct lima_sched_task { struct lima_vm *vm; void *frame; - struct dma_fence **dep; - int num_dep; - int max_dep; + struct xarray deps; + unsigned long last_dep; struct lima_bo **bos; int num_bos; @@ -78,7 +77,6 @@ int lima_sched_task_init(struct lima_sched_task *task, struct lima_bo **bos, int num_bos, struct lima_vm *vm); void lima_sched_task_fini(struct lima_sched_task *task); -int lima_sched_task_add_dep(struct lima_sched_task *task, struct dma_fence *fence); int lima_sched_context_init(struct lima_sched_pipe *pipe, struct lima_sched_context *context, diff --git a/drivers/gpu/drm/mediatek/mtk_dpi.c b/drivers/gpu/drm/mediatek/mtk_dpi.c index 22e68a100e7b..5d333138f913 100644 --- a/drivers/gpu/drm/mediatek/mtk_dpi.c +++ b/drivers/gpu/drm/mediatek/mtk_dpi.c @@ -662,13 +662,11 @@ static unsigned int mt8173_calculate_factor(int clock) static unsigned int mt2701_calculate_factor(int clock) { if (clock <= 64000) - return 16; - else if (clock <= 128000) - return 8; - else if (clock <= 256000) return 4; - else + else if (clock <= 128000) return 2; + else + return 1; } static const struct mtk_dpi_conf mt8173_conf = { diff --git a/drivers/gpu/drm/mediatek/mtk_drm_drv.c b/drivers/gpu/drm/mediatek/mtk_drm_drv.c index cf59ea9bccfd..57ce4708ef1b 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_drv.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_drv.c @@ -15,6 +15,7 @@ #include <drm/drmP.h> #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> +#include <drm/drm_fb_helper.h> #include <drm/drm_gem.h> #include <drm/drm_gem_cma_helper.h> #include <drm/drm_of.h> @@ -341,6 +342,8 @@ static struct drm_driver mtk_drm_driver = { .gem_prime_get_sg_table = mtk_gem_prime_get_sg_table, .gem_prime_import_sg_table = mtk_gem_prime_import_sg_table, .gem_prime_mmap = mtk_drm_gem_mmap_buf, + .gem_prime_vmap = mtk_drm_gem_prime_vmap, + .gem_prime_vunmap = mtk_drm_gem_prime_vunmap, .fops = &mtk_drm_fops, .name = DRIVER_NAME, @@ -376,6 +379,10 @@ static int mtk_drm_bind(struct device *dev) if (ret < 0) goto err_deinit; + ret = drm_fbdev_generic_setup(drm, 32); + if (ret) + DRM_ERROR("Failed to initialize fbdev: %d\n", ret); + return 0; err_deinit: diff --git a/drivers/gpu/drm/mediatek/mtk_drm_gem.c b/drivers/gpu/drm/mediatek/mtk_drm_gem.c index 259b7b0de1d2..38483e9ee071 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_gem.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_gem.c @@ -241,3 +241,49 @@ err_gem_free: kfree(mtk_gem); return ERR_PTR(ret); } + +void *mtk_drm_gem_prime_vmap(struct drm_gem_object *obj) +{ + struct mtk_drm_gem_obj *mtk_gem = to_mtk_gem_obj(obj); + struct sg_table *sgt; + struct sg_page_iter iter; + unsigned int npages; + unsigned int i = 0; + + if (mtk_gem->kvaddr) + return mtk_gem->kvaddr; + + sgt = mtk_gem_prime_get_sg_table(obj); + if (IS_ERR(sgt)) + return NULL; + + npages = obj->size >> PAGE_SHIFT; + mtk_gem->pages = kcalloc(npages, sizeof(*mtk_gem->pages), GFP_KERNEL); + if (!mtk_gem->pages) + goto out; + + for_each_sg_page(sgt->sgl, &iter, sgt->orig_nents, 0) { + mtk_gem->pages[i++] = sg_page_iter_page(&iter); + if (i > npages) + break; + } + mtk_gem->kvaddr = vmap(mtk_gem->pages, npages, VM_MAP, + pgprot_writecombine(PAGE_KERNEL)); + +out: + kfree((void *)sgt); + + return mtk_gem->kvaddr; +} + +void mtk_drm_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr) +{ + struct mtk_drm_gem_obj *mtk_gem = to_mtk_gem_obj(obj); + + if (!mtk_gem->pages) + return; + + vunmap(vaddr); + mtk_gem->kvaddr = 0; + kfree((void *)mtk_gem->pages); +} diff --git a/drivers/gpu/drm/mediatek/mtk_drm_gem.h b/drivers/gpu/drm/mediatek/mtk_drm_gem.h index 534639b43a1c..c047a7ef294f 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_gem.h +++ b/drivers/gpu/drm/mediatek/mtk_drm_gem.h @@ -37,6 +37,7 @@ struct mtk_drm_gem_obj { dma_addr_t dma_addr; unsigned long dma_attrs; struct sg_table *sg; + struct page **pages; }; #define to_mtk_gem_obj(x) container_of(x, struct mtk_drm_gem_obj, base) @@ -52,5 +53,7 @@ int mtk_drm_gem_mmap_buf(struct drm_gem_object *obj, struct sg_table *mtk_gem_prime_get_sg_table(struct drm_gem_object *obj); struct drm_gem_object *mtk_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sg); +void *mtk_drm_gem_prime_vmap(struct drm_gem_object *obj); +void mtk_drm_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr); #endif diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi.c b/drivers/gpu/drm/mediatek/mtk_hdmi.c index 915cc84621ae..e04e6c293d39 100644 --- a/drivers/gpu/drm/mediatek/mtk_hdmi.c +++ b/drivers/gpu/drm/mediatek/mtk_hdmi.c @@ -1480,7 +1480,6 @@ static int mtk_hdmi_dt_parse_pdata(struct mtk_hdmi *hdmi, if (IS_ERR(regmap)) ret = PTR_ERR(regmap); if (ret) { - ret = PTR_ERR(regmap); dev_err(dev, "Failed to get system configuration registers: %d\n", ret); @@ -1516,6 +1515,7 @@ static int mtk_hdmi_dt_parse_pdata(struct mtk_hdmi *hdmi, of_node_put(remote); hdmi->ddc_adpt = of_find_i2c_adapter_by_node(i2c_np); + of_node_put(i2c_np); if (!hdmi->ddc_adpt) { dev_err(dev, "Failed to get ddc i2c adapter by node\n"); return -EINVAL; diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi_phy.c b/drivers/gpu/drm/mediatek/mtk_hdmi_phy.c index 4ef9c57ffd44..5223498502c4 100644 --- a/drivers/gpu/drm/mediatek/mtk_hdmi_phy.c +++ b/drivers/gpu/drm/mediatek/mtk_hdmi_phy.c @@ -15,28 +15,6 @@ static const struct phy_ops mtk_hdmi_phy_dev_ops = { .owner = THIS_MODULE, }; -long mtk_hdmi_pll_round_rate(struct clk_hw *hw, unsigned long rate, - unsigned long *parent_rate) -{ - struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw); - - hdmi_phy->pll_rate = rate; - if (rate <= 74250000) - *parent_rate = rate; - else - *parent_rate = rate / 2; - - return rate; -} - -unsigned long mtk_hdmi_pll_recalc_rate(struct clk_hw *hw, - unsigned long parent_rate) -{ - struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw); - - return hdmi_phy->pll_rate; -} - void mtk_hdmi_phy_clear_bits(struct mtk_hdmi_phy *hdmi_phy, u32 offset, u32 bits) { @@ -110,13 +88,11 @@ mtk_hdmi_phy_dev_get_ops(const struct mtk_hdmi_phy *hdmi_phy) return NULL; } -static void mtk_hdmi_phy_clk_get_ops(struct mtk_hdmi_phy *hdmi_phy, - const struct clk_ops **ops) +static void mtk_hdmi_phy_clk_get_data(struct mtk_hdmi_phy *hdmi_phy, + struct clk_init_data *clk_init) { - if (hdmi_phy && hdmi_phy->conf && hdmi_phy->conf->hdmi_phy_clk_ops) - *ops = hdmi_phy->conf->hdmi_phy_clk_ops; - else - dev_err(hdmi_phy->dev, "Failed to get clk ops of phy\n"); + clk_init->flags = hdmi_phy->conf->flags; + clk_init->ops = hdmi_phy->conf->hdmi_phy_clk_ops; } static int mtk_hdmi_phy_probe(struct platform_device *pdev) @@ -129,7 +105,6 @@ static int mtk_hdmi_phy_probe(struct platform_device *pdev) struct clk_init_data clk_init = { .num_parents = 1, .parent_names = (const char * const *)&ref_clk_name, - .flags = CLK_SET_RATE_PARENT | CLK_SET_RATE_GATE, }; struct phy *phy; @@ -167,7 +142,7 @@ static int mtk_hdmi_phy_probe(struct platform_device *pdev) hdmi_phy->dev = dev; hdmi_phy->conf = (struct mtk_hdmi_phy_conf *)of_device_get_match_data(dev); - mtk_hdmi_phy_clk_get_ops(hdmi_phy, &clk_init.ops); + mtk_hdmi_phy_clk_get_data(hdmi_phy, &clk_init); hdmi_phy->pll_hw.init = &clk_init; hdmi_phy->pll = devm_clk_register(dev, &hdmi_phy->pll_hw); if (IS_ERR(hdmi_phy->pll)) { diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi_phy.h b/drivers/gpu/drm/mediatek/mtk_hdmi_phy.h index f39b1fc66612..2d8b3182470d 100644 --- a/drivers/gpu/drm/mediatek/mtk_hdmi_phy.h +++ b/drivers/gpu/drm/mediatek/mtk_hdmi_phy.h @@ -21,6 +21,7 @@ struct mtk_hdmi_phy; struct mtk_hdmi_phy_conf { bool tz_disabled; + unsigned long flags; const struct clk_ops *hdmi_phy_clk_ops; void (*hdmi_phy_enable_tmds)(struct mtk_hdmi_phy *hdmi_phy); void (*hdmi_phy_disable_tmds)(struct mtk_hdmi_phy *hdmi_phy); @@ -48,10 +49,6 @@ void mtk_hdmi_phy_set_bits(struct mtk_hdmi_phy *hdmi_phy, u32 offset, void mtk_hdmi_phy_mask(struct mtk_hdmi_phy *hdmi_phy, u32 offset, u32 val, u32 mask); struct mtk_hdmi_phy *to_mtk_hdmi_phy(struct clk_hw *hw); -long mtk_hdmi_pll_round_rate(struct clk_hw *hw, unsigned long rate, - unsigned long *parent_rate); -unsigned long mtk_hdmi_pll_recalc_rate(struct clk_hw *hw, - unsigned long parent_rate); extern struct platform_driver mtk_hdmi_phy_driver; extern struct mtk_hdmi_phy_conf mtk_hdmi_phy_8173_conf; diff --git a/drivers/gpu/drm/mediatek/mtk_mt2701_hdmi_phy.c b/drivers/gpu/drm/mediatek/mtk_mt2701_hdmi_phy.c index fcc42dc6ea7f..d3cc4022e988 100644 --- a/drivers/gpu/drm/mediatek/mtk_mt2701_hdmi_phy.c +++ b/drivers/gpu/drm/mediatek/mtk_mt2701_hdmi_phy.c @@ -79,7 +79,6 @@ static int mtk_hdmi_pll_prepare(struct clk_hw *hw) mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SLDO_MASK); usleep_range(80, 100); mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_MBIAS_LPF_EN); - mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_POSDIV); mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SER_MASK); mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_PRED_MASK); mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_DRV_MASK); @@ -94,7 +93,6 @@ static void mtk_hdmi_pll_unprepare(struct clk_hw *hw) mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_DRV_MASK); mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_PRED_MASK); mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SER_MASK); - mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_POSDIV); mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_MBIAS_LPF_EN); usleep_range(80, 100); mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SLDO_MASK); @@ -108,6 +106,12 @@ static void mtk_hdmi_pll_unprepare(struct clk_hw *hw) usleep_range(80, 100); } +static long mtk_hdmi_pll_round_rate(struct clk_hw *hw, unsigned long rate, + unsigned long *parent_rate) +{ + return rate; +} + static int mtk_hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate, unsigned long parent_rate) { @@ -116,13 +120,14 @@ static int mtk_hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate, if (rate <= 64000000) pos_div = 3; - else if (rate <= 12800000) - pos_div = 1; + else if (rate <= 128000000) + pos_div = 2; else pos_div = 1; mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_PREDIV_MASK); mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON6, RG_HTPLL_POSDIV_MASK); + mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_POSDIV); mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6, (0x1 << RG_HTPLL_IC), RG_HTPLL_IC_MASK); mtk_hdmi_phy_mask(hdmi_phy, HDMI_CON6, (0x1 << RG_HTPLL_IR), @@ -154,6 +159,39 @@ static int mtk_hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate, return 0; } +static unsigned long mtk_hdmi_pll_recalc_rate(struct clk_hw *hw, + unsigned long parent_rate) +{ + struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw); + unsigned long out_rate, val; + + val = (readl(hdmi_phy->regs + HDMI_CON6) + & RG_HTPLL_PREDIV_MASK) >> RG_HTPLL_PREDIV; + switch (val) { + case 0x00: + out_rate = parent_rate; + break; + case 0x01: + out_rate = parent_rate / 2; + break; + default: + out_rate = parent_rate / 4; + break; + } + + val = (readl(hdmi_phy->regs + HDMI_CON6) + & RG_HTPLL_FBKDIV_MASK) >> RG_HTPLL_FBKDIV; + out_rate *= (val + 1) * 2; + val = (readl(hdmi_phy->regs + HDMI_CON2) + & RG_HDMITX_TX_POSDIV_MASK); + out_rate >>= (val >> RG_HDMITX_TX_POSDIV); + + if (readl(hdmi_phy->regs + HDMI_CON2) & RG_HDMITX_EN_TX_POSDIV) + out_rate /= 5; + + return out_rate; +} + static const struct clk_ops mtk_hdmi_phy_pll_ops = { .prepare = mtk_hdmi_pll_prepare, .unprepare = mtk_hdmi_pll_unprepare, @@ -174,7 +212,6 @@ static void mtk_hdmi_phy_enable_tmds(struct mtk_hdmi_phy *hdmi_phy) mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SLDO_MASK); usleep_range(80, 100); mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_MBIAS_LPF_EN); - mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_POSDIV); mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SER_MASK); mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_PRED_MASK); mtk_hdmi_phy_set_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_DRV_MASK); @@ -186,7 +223,6 @@ static void mtk_hdmi_phy_disable_tmds(struct mtk_hdmi_phy *hdmi_phy) mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_DRV_MASK); mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_PRED_MASK); mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SER_MASK); - mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_EN_TX_POSDIV); mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON2, RG_HDMITX_MBIAS_LPF_EN); usleep_range(80, 100); mtk_hdmi_phy_clear_bits(hdmi_phy, HDMI_CON0, RG_HDMITX_EN_SLDO_MASK); @@ -202,6 +238,7 @@ static void mtk_hdmi_phy_disable_tmds(struct mtk_hdmi_phy *hdmi_phy) struct mtk_hdmi_phy_conf mtk_hdmi_phy_2701_conf = { .tz_disabled = true, + .flags = CLK_SET_RATE_GATE, .hdmi_phy_clk_ops = &mtk_hdmi_phy_pll_ops, .hdmi_phy_enable_tmds = mtk_hdmi_phy_enable_tmds, .hdmi_phy_disable_tmds = mtk_hdmi_phy_disable_tmds, diff --git a/drivers/gpu/drm/mediatek/mtk_mt8173_hdmi_phy.c b/drivers/gpu/drm/mediatek/mtk_mt8173_hdmi_phy.c index ed5916b27658..47f8a2951682 100644 --- a/drivers/gpu/drm/mediatek/mtk_mt8173_hdmi_phy.c +++ b/drivers/gpu/drm/mediatek/mtk_mt8173_hdmi_phy.c @@ -199,6 +199,20 @@ static void mtk_hdmi_pll_unprepare(struct clk_hw *hw) usleep_range(100, 150); } +static long mtk_hdmi_pll_round_rate(struct clk_hw *hw, unsigned long rate, + unsigned long *parent_rate) +{ + struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw); + + hdmi_phy->pll_rate = rate; + if (rate <= 74250000) + *parent_rate = rate; + else + *parent_rate = rate / 2; + + return rate; +} + static int mtk_hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate, unsigned long parent_rate) { @@ -285,6 +299,14 @@ static int mtk_hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate, return 0; } +static unsigned long mtk_hdmi_pll_recalc_rate(struct clk_hw *hw, + unsigned long parent_rate) +{ + struct mtk_hdmi_phy *hdmi_phy = to_mtk_hdmi_phy(hw); + + return hdmi_phy->pll_rate; +} + static const struct clk_ops mtk_hdmi_phy_pll_ops = { .prepare = mtk_hdmi_pll_prepare, .unprepare = mtk_hdmi_pll_unprepare, @@ -309,6 +331,7 @@ static void mtk_hdmi_phy_disable_tmds(struct mtk_hdmi_phy *hdmi_phy) } struct mtk_hdmi_phy_conf mtk_hdmi_phy_8173_conf = { + .flags = CLK_SET_RATE_PARENT | CLK_SET_RATE_GATE, .hdmi_phy_clk_ops = &mtk_hdmi_phy_pll_ops, .hdmi_phy_enable_tmds = mtk_hdmi_phy_enable_tmds, .hdmi_phy_disable_tmds = mtk_hdmi_phy_disable_tmds, diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c index 937cfabb95df..72b01e6be0d9 100644 --- a/drivers/gpu/drm/meson/meson_drv.c +++ b/drivers/gpu/drm/meson/meson_drv.c @@ -90,6 +90,18 @@ static irqreturn_t meson_irq(int irq, void *arg) return IRQ_HANDLED; } +static int meson_dumb_create(struct drm_file *file, struct drm_device *dev, + struct drm_mode_create_dumb *args) +{ + /* + * We need 64bytes aligned stride, and PAGE aligned size + */ + args->pitch = ALIGN(DIV_ROUND_UP(args->width * args->bpp, 8), SZ_64); + args->size = PAGE_ALIGN(args->pitch * args->height); + + return drm_gem_cma_dumb_create_internal(file, dev, args); +} + DEFINE_DRM_GEM_CMA_FOPS(fops); static struct drm_driver meson_driver = { @@ -112,7 +124,7 @@ static struct drm_driver meson_driver = { .gem_prime_mmap = drm_gem_cma_prime_mmap, /* GEM Ops */ - .dumb_create = drm_gem_cma_dumb_create, + .dumb_create = meson_dumb_create, .gem_free_object_unlocked = drm_gem_cma_free_object, .gem_vm_ops = &drm_gem_cma_vm_ops, @@ -317,12 +329,14 @@ static int meson_drv_bind_master(struct device *dev, bool has_components) ret = drm_dev_register(drm, 0); if (ret) - goto free_drm; + goto uninstall_irq; drm_fbdev_generic_setup(drm, 32); return 0; +uninstall_irq: + drm_irq_uninstall(drm); free_drm: drm_dev_put(drm); @@ -336,8 +350,8 @@ static int meson_drv_bind(struct device *dev) static void meson_drv_unbind(struct device *dev) { - struct drm_device *drm = dev_get_drvdata(dev); - struct meson_drm *priv = drm->dev_private; + struct meson_drm *priv = dev_get_drvdata(dev); + struct drm_device *drm = priv->drm; if (priv->canvas) { meson_canvas_free(priv->canvas, priv->canvas_id_osd1); @@ -347,6 +361,7 @@ static void meson_drv_unbind(struct device *dev) } drm_dev_unregister(drm); + drm_irq_uninstall(drm); drm_kms_helper_poll_fini(drm); drm_mode_config_cleanup(drm); drm_dev_put(drm); diff --git a/drivers/gpu/drm/meson/meson_dw_hdmi.c b/drivers/gpu/drm/meson/meson_dw_hdmi.c index 2a860d158f4f..779da21143b9 100644 --- a/drivers/gpu/drm/meson/meson_dw_hdmi.c +++ b/drivers/gpu/drm/meson/meson_dw_hdmi.c @@ -631,7 +631,8 @@ dw_hdmi_mode_valid(struct drm_connector *connector, DRM_DEBUG_DRIVER("Modeline " DRM_MODE_FMT "\n", DRM_MODE_ARG(mode)); /* If sink max TMDS clock, we reject the mode */ - if (mode->clock > connector->display_info.max_tmds_clock) + if (connector->display_info.max_tmds_clock && + mode->clock > connector->display_info.max_tmds_clock) return MODE_BAD; /* Check against non-VIC supported modes */ diff --git a/drivers/gpu/drm/meson/meson_viu.c b/drivers/gpu/drm/meson/meson_viu.c index 0169c98b01c9..b59072342cae 100644 --- a/drivers/gpu/drm/meson/meson_viu.c +++ b/drivers/gpu/drm/meson/meson_viu.c @@ -90,8 +90,8 @@ static int eotf_bypass_coeff[EOTF_COEFF_SIZE] = { EOTF_COEFF_RIGHTSHIFT /* right shift */ }; -void meson_viu_set_g12a_osd1_matrix(struct meson_drm *priv, int *m, - bool csc_on) +static void meson_viu_set_g12a_osd1_matrix(struct meson_drm *priv, + int *m, bool csc_on) { /* VPP WRAP OSD1 matrix */ writel(((m[0] & 0xfff) << 16) | (m[1] & 0xfff), @@ -118,8 +118,8 @@ void meson_viu_set_g12a_osd1_matrix(struct meson_drm *priv, int *m, priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_EN_CTRL)); } -void meson_viu_set_osd_matrix(struct meson_drm *priv, - enum viu_matrix_sel_e m_select, +static void meson_viu_set_osd_matrix(struct meson_drm *priv, + enum viu_matrix_sel_e m_select, int *m, bool csc_on) { if (m_select == VIU_MATRIX_OSD) { @@ -187,10 +187,10 @@ void meson_viu_set_osd_matrix(struct meson_drm *priv, #define OSD_EOTF_LUT_SIZE 33 #define OSD_OETF_LUT_SIZE 41 -void meson_viu_set_osd_lut(struct meson_drm *priv, enum viu_lut_sel_e lut_sel, - unsigned int *r_map, unsigned int *g_map, - unsigned int *b_map, - bool csc_on) +static void +meson_viu_set_osd_lut(struct meson_drm *priv, enum viu_lut_sel_e lut_sel, + unsigned int *r_map, unsigned int *g_map, + unsigned int *b_map, bool csc_on) { unsigned int addr_port; unsigned int data_port; diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4_cec.c b/drivers/gpu/drm/omapdrm/dss/hdmi4_cec.c index 340383150fb9..ebf9c96d43ee 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi4_cec.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi4_cec.c @@ -175,6 +175,7 @@ static int hdmi_cec_adap_enable(struct cec_adapter *adap, bool enable) REG_FLD_MOD(core->base, HDMI_CORE_SYS_INTR_UNMASK4, 0, 3, 3); hdmi_wp_clear_irqenable(core->wp, HDMI_IRQ_CORE); hdmi_wp_set_irqstatus(core->wp, HDMI_IRQ_CORE); + REG_FLD_MOD(core->wp->base, HDMI_WP_CLK, 0, 5, 0); hdmi4_core_disable(core); return 0; } @@ -182,16 +183,24 @@ static int hdmi_cec_adap_enable(struct cec_adapter *adap, bool enable) if (err) return err; + /* + * Initialize CEC clock divider: CEC needs 2MHz clock hence + * set the divider to 24 to get 48/24=2MHz clock + */ + REG_FLD_MOD(core->wp->base, HDMI_WP_CLK, 0x18, 5, 0); + /* Clear TX FIFO */ if (!hdmi_cec_clear_tx_fifo(adap)) { pr_err("cec-%s: could not clear TX FIFO\n", adap->name); - return -EIO; + err = -EIO; + goto err_disable_clk; } /* Clear RX FIFO */ if (!hdmi_cec_clear_rx_fifo(adap)) { pr_err("cec-%s: could not clear RX FIFO\n", adap->name); - return -EIO; + err = -EIO; + goto err_disable_clk; } /* Clear CEC interrupts */ @@ -236,6 +245,12 @@ static int hdmi_cec_adap_enable(struct cec_adapter *adap, bool enable) hdmi_write_reg(core->base, HDMI_CEC_INT_STATUS_1, temp); } return 0; + +err_disable_clk: + REG_FLD_MOD(core->wp->base, HDMI_WP_CLK, 0, 5, 0); + hdmi4_core_disable(core); + + return err; } static int hdmi_cec_adap_log_addr(struct cec_adapter *adap, u8 log_addr) @@ -333,11 +348,8 @@ int hdmi4_cec_init(struct platform_device *pdev, struct hdmi_core_data *core, return ret; core->wp = wp; - /* - * Initialize CEC clock divider: CEC needs 2MHz clock hence - * set the devider to 24 to get 48/24=2MHz clock - */ - REG_FLD_MOD(core->wp->base, HDMI_WP_CLK, 0x18, 5, 0); + /* Disable clock initially, hdmi_cec_adap_enable() manages it */ + REG_FLD_MOD(core->wp->base, HDMI_WP_CLK, 0, 5, 0); ret = cec_register_adapter(core->adap, &pdev->dev); if (ret < 0) { diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c index 813ba42f2753..e384b95ad857 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi4_core.c @@ -708,7 +708,7 @@ int hdmi4_audio_config(struct hdmi_core_data *core, struct hdmi_wp_data *wp, else acore.i2s_cfg.justification = HDMI_AUDIO_JUSTIFY_RIGHT; /* - * The I2S input word length is twice the lenght given in the IEC-60958 + * The I2S input word length is twice the length given in the IEC-60958 * status word. If the word size is greater than * 20 bits, increment by one. */ diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c index 8fee7a8b29d9..569be4efd8d1 100644 --- a/drivers/gpu/drm/panel/panel-simple.c +++ b/drivers/gpu/drm/panel/panel-simple.c @@ -3025,6 +3025,34 @@ static const struct panel_desc_dsi panasonic_vvx10f004b00 = { .lanes = 4, }; +static const struct drm_display_mode lg_acx467akm_7_mode = { + .clock = 150000, + .hdisplay = 1080, + .hsync_start = 1080 + 2, + .hsync_end = 1080 + 2 + 2, + .htotal = 1080 + 2 + 2 + 2, + .vdisplay = 1920, + .vsync_start = 1920 + 2, + .vsync_end = 1920 + 2 + 2, + .vtotal = 1920 + 2 + 2 + 2, + .vrefresh = 60, +}; + +static const struct panel_desc_dsi lg_acx467akm_7 = { + .desc = { + .modes = &lg_acx467akm_7_mode, + .num_modes = 1, + .bpc = 8, + .size = { + .width = 62, + .height = 110, + }, + }, + .flags = 0, + .format = MIPI_DSI_FMT_RGB888, + .lanes = 4, +}; + static const struct of_device_id dsi_of_match[] = { { .compatible = "auo,b080uan01", @@ -3042,6 +3070,9 @@ static const struct of_device_id dsi_of_match[] = { .compatible = "panasonic,vvx10f004b00", .data = &panasonic_vvx10f004b00 }, { + .compatible = "lg,acx467akm-7", + .data = &lg_acx467akm_7 + }, { /* sentinel */ } }; diff --git a/drivers/gpu/drm/panfrost/Kconfig b/drivers/gpu/drm/panfrost/Kconfig new file mode 100644 index 000000000000..7f5e572daa2d --- /dev/null +++ b/drivers/gpu/drm/panfrost/Kconfig @@ -0,0 +1,14 @@ +# SPDX-License-Identifier: GPL-2.0 + +config DRM_PANFROST + tristate "Panfrost (DRM support for ARM Mali Midgard/Bifrost GPUs)" + depends on DRM + depends on ARM || ARM64 || COMPILE_TEST + depends on MMU + select DRM_SCHED + select IOMMU_SUPPORT + select IOMMU_IO_PGTABLE_LPAE + select DRM_GEM_SHMEM_HELPER + help + DRM driver for ARM Mali Midgard (T6xx, T7xx, T8xx) and + Bifrost (G3x, G5x, G7x) GPUs. diff --git a/drivers/gpu/drm/panfrost/Makefile b/drivers/gpu/drm/panfrost/Makefile new file mode 100644 index 000000000000..6de72d13c58f --- /dev/null +++ b/drivers/gpu/drm/panfrost/Makefile @@ -0,0 +1,12 @@ +# SPDX-License-Identifier: GPL-2.0 + +panfrost-y := \ + panfrost_drv.o \ + panfrost_device.o \ + panfrost_devfreq.o \ + panfrost_gem.o \ + panfrost_gpu.o \ + panfrost_job.o \ + panfrost_mmu.o + +obj-$(CONFIG_DRM_PANFROST) += panfrost.o diff --git a/drivers/gpu/drm/panfrost/TODO b/drivers/gpu/drm/panfrost/TODO new file mode 100644 index 000000000000..c2e44add37d8 --- /dev/null +++ b/drivers/gpu/drm/panfrost/TODO @@ -0,0 +1,27 @@ +- Thermal support. + +- Bifrost support: + - DT bindings (Neil, WIP) + - MMU page table format and address space setup + - Bifrost specific feature and issue handling + - Coherent DMA support + +- Support for 2MB pages. The io-pgtable code already supports this. Finishing + support involves either copying or adapting the iommu API to handle passing + aligned addresses and sizes to the io-pgtable code. + +- Per FD address space support. The h/w supports multiple addresses spaces. + The hard part is handling when more address spaces are needed than what + the h/w provides. + +- Support pinning pages on demand (GPU page faults). + +- Support userspace controlled GPU virtual addresses. Needed for Vulkan. (Tomeu) + +- Support for madvise and a shrinker. + +- Compute job support. So called 'compute only' jobs need to be plumbed up to + userspace. + +- Performance counter support. (Boris) + diff --git a/drivers/gpu/drm/panfrost/panfrost_devfreq.c b/drivers/gpu/drm/panfrost/panfrost_devfreq.c new file mode 100644 index 000000000000..a8121ae67ee3 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_devfreq.c @@ -0,0 +1,218 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright 2019 Collabora ltd. */ +#include <linux/devfreq.h> +#include <linux/platform_device.h> +#include <linux/pm_opp.h> +#include <linux/clk.h> +#include <linux/regulator/consumer.h> + +#include "panfrost_device.h" +#include "panfrost_features.h" +#include "panfrost_issues.h" +#include "panfrost_gpu.h" +#include "panfrost_regs.h" + +static void panfrost_devfreq_update_utilization(struct panfrost_device *pfdev, int slot); + +static int panfrost_devfreq_target(struct device *dev, unsigned long *freq, + u32 flags) +{ + struct panfrost_device *pfdev = platform_get_drvdata(to_platform_device(dev)); + struct dev_pm_opp *opp; + unsigned long old_clk_rate = pfdev->devfreq.cur_freq; + unsigned long target_volt, target_rate; + int err; + + opp = devfreq_recommended_opp(dev, freq, flags); + if (IS_ERR(opp)) + return PTR_ERR(opp); + + target_rate = dev_pm_opp_get_freq(opp); + target_volt = dev_pm_opp_get_voltage(opp); + dev_pm_opp_put(opp); + + if (old_clk_rate == target_rate) + return 0; + + /* + * If frequency scaling from low to high, adjust voltage first. + * If frequency scaling from high to low, adjust frequency first. + */ + if (old_clk_rate < target_rate) { + err = regulator_set_voltage(pfdev->regulator, target_volt, + target_volt); + if (err) { + dev_err(dev, "Cannot set voltage %lu uV\n", + target_volt); + return err; + } + } + + err = clk_set_rate(pfdev->clock, target_rate); + if (err) { + dev_err(dev, "Cannot set frequency %lu (%d)\n", target_rate, + err); + regulator_set_voltage(pfdev->regulator, pfdev->devfreq.cur_volt, + pfdev->devfreq.cur_volt); + return err; + } + + if (old_clk_rate > target_rate) { + err = regulator_set_voltage(pfdev->regulator, target_volt, + target_volt); + if (err) + dev_err(dev, "Cannot set voltage %lu uV\n", target_volt); + } + + pfdev->devfreq.cur_freq = target_rate; + pfdev->devfreq.cur_volt = target_volt; + + return 0; +} + +static void panfrost_devfreq_reset(struct panfrost_device *pfdev) +{ + ktime_t now = ktime_get(); + int i; + + for (i = 0; i < NUM_JOB_SLOTS; i++) { + pfdev->devfreq.slot[i].busy_time = 0; + pfdev->devfreq.slot[i].idle_time = 0; + pfdev->devfreq.slot[i].time_last_update = now; + } +} + +static int panfrost_devfreq_get_dev_status(struct device *dev, + struct devfreq_dev_status *status) +{ + struct panfrost_device *pfdev = platform_get_drvdata(to_platform_device(dev)); + int i; + + for (i = 0; i < NUM_JOB_SLOTS; i++) { + panfrost_devfreq_update_utilization(pfdev, i); + } + + status->current_frequency = clk_get_rate(pfdev->clock); + status->total_time = ktime_to_ns(ktime_add(pfdev->devfreq.slot[0].busy_time, + pfdev->devfreq.slot[0].idle_time)); + + status->busy_time = 0; + for (i = 0; i < NUM_JOB_SLOTS; i++) { + status->busy_time += ktime_to_ns(pfdev->devfreq.slot[i].busy_time); + } + + /* We're scheduling only to one core atm, so don't divide for now */ + /* status->busy_time /= NUM_JOB_SLOTS; */ + + panfrost_devfreq_reset(pfdev); + + dev_dbg(pfdev->dev, "busy %lu total %lu %lu %% freq %lu MHz\n", status->busy_time, + status->total_time, + status->busy_time / (status->total_time / 100), + status->current_frequency / 1000 / 1000); + + return 0; +} + +static int panfrost_devfreq_get_cur_freq(struct device *dev, unsigned long *freq) +{ + struct panfrost_device *pfdev = platform_get_drvdata(to_platform_device(dev)); + + *freq = pfdev->devfreq.cur_freq; + + return 0; +} + +static struct devfreq_dev_profile panfrost_devfreq_profile = { + .polling_ms = 50, /* ~3 frames */ + .target = panfrost_devfreq_target, + .get_dev_status = panfrost_devfreq_get_dev_status, + .get_cur_freq = panfrost_devfreq_get_cur_freq, +}; + +int panfrost_devfreq_init(struct panfrost_device *pfdev) +{ + int ret; + struct dev_pm_opp *opp; + + if (!pfdev->regulator) + return 0; + + ret = dev_pm_opp_of_add_table(&pfdev->pdev->dev); + if (ret == -ENODEV) /* Optional, continue without devfreq */ + return 0; + + panfrost_devfreq_reset(pfdev); + + pfdev->devfreq.cur_freq = clk_get_rate(pfdev->clock); + + opp = devfreq_recommended_opp(&pfdev->pdev->dev, &pfdev->devfreq.cur_freq, 0); + if (IS_ERR(opp)) + return PTR_ERR(opp); + + panfrost_devfreq_profile.initial_freq = pfdev->devfreq.cur_freq; + dev_pm_opp_put(opp); + + pfdev->devfreq.devfreq = devm_devfreq_add_device(&pfdev->pdev->dev, + &panfrost_devfreq_profile, "simple_ondemand", NULL); + if (IS_ERR(pfdev->devfreq.devfreq)) { + DRM_DEV_ERROR(&pfdev->pdev->dev, "Couldn't initialize GPU devfreq\n"); + ret = PTR_ERR(pfdev->devfreq.devfreq); + pfdev->devfreq.devfreq = NULL; + return ret; + } + + return 0; +} + +void panfrost_devfreq_resume(struct panfrost_device *pfdev) +{ + int i; + + if (!pfdev->devfreq.devfreq) + return; + + panfrost_devfreq_reset(pfdev); + for (i = 0; i < NUM_JOB_SLOTS; i++) + pfdev->devfreq.slot[i].busy = false; + + devfreq_resume_device(pfdev->devfreq.devfreq); +} + +void panfrost_devfreq_suspend(struct panfrost_device *pfdev) +{ + if (!pfdev->devfreq.devfreq) + return; + + devfreq_suspend_device(pfdev->devfreq.devfreq); +} + +static void panfrost_devfreq_update_utilization(struct panfrost_device *pfdev, int slot) +{ + struct panfrost_devfreq_slot *devfreq_slot = &pfdev->devfreq.slot[slot]; + ktime_t now; + ktime_t last; + + if (!pfdev->devfreq.devfreq) + return; + + now = ktime_get(); + last = pfdev->devfreq.slot[slot].time_last_update; + + /* If we last recorded a transition to busy, we have been idle since */ + if (devfreq_slot->busy) + pfdev->devfreq.slot[slot].busy_time += ktime_sub(now, last); + else + pfdev->devfreq.slot[slot].idle_time += ktime_sub(now, last); + + pfdev->devfreq.slot[slot].time_last_update = now; +} + +/* The job scheduler is expected to call this at every transition busy <-> idle */ +void panfrost_devfreq_record_transition(struct panfrost_device *pfdev, int slot) +{ + struct panfrost_devfreq_slot *devfreq_slot = &pfdev->devfreq.slot[slot]; + + panfrost_devfreq_update_utilization(pfdev, slot); + devfreq_slot->busy = !devfreq_slot->busy; +} diff --git a/drivers/gpu/drm/panfrost/panfrost_devfreq.h b/drivers/gpu/drm/panfrost/panfrost_devfreq.h new file mode 100644 index 000000000000..eb999531ed90 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_devfreq.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright 2019 Collabora ltd. */ + +#ifndef __PANFROST_DEVFREQ_H__ +#define __PANFROST_DEVFREQ_H__ + +int panfrost_devfreq_init(struct panfrost_device *pfdev); + +void panfrost_devfreq_resume(struct panfrost_device *pfdev); +void panfrost_devfreq_suspend(struct panfrost_device *pfdev); + +void panfrost_devfreq_record_transition(struct panfrost_device *pfdev, int slot); + +#endif /* __PANFROST_DEVFREQ_H__ */ diff --git a/drivers/gpu/drm/panfrost/panfrost_device.c b/drivers/gpu/drm/panfrost/panfrost_device.c new file mode 100644 index 000000000000..91e8fb0f2b25 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_device.c @@ -0,0 +1,252 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */ +/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */ + +#include <linux/clk.h> +#include <linux/reset.h> +#include <linux/platform_device.h> +#include <linux/pm_runtime.h> +#include <linux/regulator/consumer.h> + +#include "panfrost_device.h" +#include "panfrost_devfreq.h" +#include "panfrost_features.h" +#include "panfrost_gpu.h" +#include "panfrost_job.h" +#include "panfrost_mmu.h" + +static int panfrost_reset_init(struct panfrost_device *pfdev) +{ + int err; + + pfdev->rstc = devm_reset_control_array_get(pfdev->dev, false, true); + if (IS_ERR(pfdev->rstc)) { + dev_err(pfdev->dev, "get reset failed %ld\n", PTR_ERR(pfdev->rstc)); + return PTR_ERR(pfdev->rstc); + } + + err = reset_control_deassert(pfdev->rstc); + if (err) + return err; + + return 0; +} + +static void panfrost_reset_fini(struct panfrost_device *pfdev) +{ + reset_control_assert(pfdev->rstc); +} + +static int panfrost_clk_init(struct panfrost_device *pfdev) +{ + int err; + unsigned long rate; + + pfdev->clock = devm_clk_get(pfdev->dev, NULL); + if (IS_ERR(pfdev->clock)) { + dev_err(pfdev->dev, "get clock failed %ld\n", PTR_ERR(pfdev->clock)); + return PTR_ERR(pfdev->clock); + } + + rate = clk_get_rate(pfdev->clock); + dev_info(pfdev->dev, "clock rate = %lu\n", rate); + + err = clk_prepare_enable(pfdev->clock); + if (err) + return err; + + return 0; +} + +static void panfrost_clk_fini(struct panfrost_device *pfdev) +{ + clk_disable_unprepare(pfdev->clock); +} + +static int panfrost_regulator_init(struct panfrost_device *pfdev) +{ + int ret; + + pfdev->regulator = devm_regulator_get_optional(pfdev->dev, "mali"); + if (IS_ERR(pfdev->regulator)) { + ret = PTR_ERR(pfdev->regulator); + pfdev->regulator = NULL; + if (ret == -ENODEV) + return 0; + dev_err(pfdev->dev, "failed to get regulator: %d\n", ret); + return ret; + } + + ret = regulator_enable(pfdev->regulator); + if (ret < 0) { + dev_err(pfdev->dev, "failed to enable regulator: %d\n", ret); + return ret; + } + + return 0; +} + +static void panfrost_regulator_fini(struct panfrost_device *pfdev) +{ + if (pfdev->regulator) + regulator_disable(pfdev->regulator); +} + +int panfrost_device_init(struct panfrost_device *pfdev) +{ + int err; + struct resource *res; + + mutex_init(&pfdev->sched_lock); + INIT_LIST_HEAD(&pfdev->scheduled_jobs); + + spin_lock_init(&pfdev->hwaccess_lock); + + err = panfrost_clk_init(pfdev); + if (err) { + dev_err(pfdev->dev, "clk init failed %d\n", err); + return err; + } + + err = panfrost_regulator_init(pfdev); + if (err) { + dev_err(pfdev->dev, "regulator init failed %d\n", err); + goto err_out0; + } + + err = panfrost_reset_init(pfdev); + if (err) { + dev_err(pfdev->dev, "reset init failed %d\n", err); + goto err_out1; + } + + res = platform_get_resource(pfdev->pdev, IORESOURCE_MEM, 0); + pfdev->iomem = devm_ioremap_resource(pfdev->dev, res); + if (IS_ERR(pfdev->iomem)) { + dev_err(pfdev->dev, "failed to ioremap iomem\n"); + err = PTR_ERR(pfdev->iomem); + goto err_out2; + } + + err = panfrost_gpu_init(pfdev); + if (err) + goto err_out2; + + err = panfrost_mmu_init(pfdev); + if (err) + goto err_out3; + + err = panfrost_job_init(pfdev); + if (err) + goto err_out4; + + /* runtime PM will wake us up later */ + panfrost_gpu_power_off(pfdev); + + pm_runtime_set_active(pfdev->dev); + pm_runtime_get_sync(pfdev->dev); + pm_runtime_mark_last_busy(pfdev->dev); + pm_runtime_put_autosuspend(pfdev->dev); + + return 0; +err_out4: + panfrost_mmu_fini(pfdev); +err_out3: + panfrost_gpu_fini(pfdev); +err_out2: + panfrost_reset_fini(pfdev); +err_out1: + panfrost_regulator_fini(pfdev); +err_out0: + panfrost_clk_fini(pfdev); + return err; +} + +void panfrost_device_fini(struct panfrost_device *pfdev) +{ + panfrost_regulator_fini(pfdev); + panfrost_clk_fini(pfdev); +} + +const char *panfrost_exception_name(struct panfrost_device *pfdev, u32 exception_code) +{ + switch (exception_code) { + /* Non-Fault Status code */ + case 0x00: return "NOT_STARTED/IDLE/OK"; + case 0x01: return "DONE"; + case 0x02: return "INTERRUPTED"; + case 0x03: return "STOPPED"; + case 0x04: return "TERMINATED"; + case 0x08: return "ACTIVE"; + /* Job exceptions */ + case 0x40: return "JOB_CONFIG_FAULT"; + case 0x41: return "JOB_POWER_FAULT"; + case 0x42: return "JOB_READ_FAULT"; + case 0x43: return "JOB_WRITE_FAULT"; + case 0x44: return "JOB_AFFINITY_FAULT"; + case 0x48: return "JOB_BUS_FAULT"; + case 0x50: return "INSTR_INVALID_PC"; + case 0x51: return "INSTR_INVALID_ENC"; + case 0x52: return "INSTR_TYPE_MISMATCH"; + case 0x53: return "INSTR_OPERAND_FAULT"; + case 0x54: return "INSTR_TLS_FAULT"; + case 0x55: return "INSTR_BARRIER_FAULT"; + case 0x56: return "INSTR_ALIGN_FAULT"; + case 0x58: return "DATA_INVALID_FAULT"; + case 0x59: return "TILE_RANGE_FAULT"; + case 0x5A: return "ADDR_RANGE_FAULT"; + case 0x60: return "OUT_OF_MEMORY"; + /* GPU exceptions */ + case 0x80: return "DELAYED_BUS_FAULT"; + case 0x88: return "SHAREABILITY_FAULT"; + /* MMU exceptions */ + case 0xC1: return "TRANSLATION_FAULT_LEVEL1"; + case 0xC2: return "TRANSLATION_FAULT_LEVEL2"; + case 0xC3: return "TRANSLATION_FAULT_LEVEL3"; + case 0xC4: return "TRANSLATION_FAULT_LEVEL4"; + case 0xC8: return "PERMISSION_FAULT"; + case 0xC9 ... 0xCF: return "PERMISSION_FAULT"; + case 0xD1: return "TRANSTAB_BUS_FAULT_LEVEL1"; + case 0xD2: return "TRANSTAB_BUS_FAULT_LEVEL2"; + case 0xD3: return "TRANSTAB_BUS_FAULT_LEVEL3"; + case 0xD4: return "TRANSTAB_BUS_FAULT_LEVEL4"; + case 0xD8: return "ACCESS_FLAG"; + case 0xD9 ... 0xDF: return "ACCESS_FLAG"; + case 0xE0 ... 0xE7: return "ADDRESS_SIZE_FAULT"; + case 0xE8 ... 0xEF: return "MEMORY_ATTRIBUTES_FAULT"; + } + + return "UNKNOWN"; +} + +#ifdef CONFIG_PM +int panfrost_device_resume(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct panfrost_device *pfdev = platform_get_drvdata(pdev); + + panfrost_gpu_soft_reset(pfdev); + + /* TODO: Re-enable all other address spaces */ + panfrost_gpu_power_on(pfdev); + panfrost_mmu_enable(pfdev, 0); + panfrost_job_enable_interrupts(pfdev); + panfrost_devfreq_resume(pfdev); + + return 0; +} + +int panfrost_device_suspend(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct panfrost_device *pfdev = platform_get_drvdata(pdev); + + if (!panfrost_job_is_idle(pfdev)) + return -EBUSY; + + panfrost_devfreq_suspend(pfdev); + panfrost_gpu_power_off(pfdev); + + return 0; +} +#endif diff --git a/drivers/gpu/drm/panfrost/panfrost_device.h b/drivers/gpu/drm/panfrost/panfrost_device.h new file mode 100644 index 000000000000..1ba48d105763 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_device.h @@ -0,0 +1,124 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */ +/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */ + +#ifndef __PANFROST_DEVICE_H__ +#define __PANFROST_DEVICE_H__ + +#include <linux/spinlock.h> +#include <drm/drm_device.h> +#include <drm/drm_mm.h> +#include <drm/gpu_scheduler.h> + +struct panfrost_device; +struct panfrost_mmu; +struct panfrost_job_slot; +struct panfrost_job; + +#define NUM_JOB_SLOTS 3 + +struct panfrost_features { + u16 id; + u16 revision; + + u64 shader_present; + u64 tiler_present; + u64 l2_present; + u64 stack_present; + u32 as_present; + u32 js_present; + + u32 l2_features; + u32 core_features; + u32 tiler_features; + u32 mem_features; + u32 mmu_features; + u32 thread_features; + u32 max_threads; + u32 thread_max_workgroup_sz; + u32 thread_max_barrier_sz; + u32 coherency_features; + u32 texture_features[4]; + u32 js_features[16]; + + u32 nr_core_groups; + + unsigned long hw_features[64 / BITS_PER_LONG]; + unsigned long hw_issues[64 / BITS_PER_LONG]; +}; + +struct panfrost_devfreq_slot { + ktime_t busy_time; + ktime_t idle_time; + ktime_t time_last_update; + bool busy; +}; + +struct panfrost_device { + struct device *dev; + struct drm_device *ddev; + struct platform_device *pdev; + + spinlock_t hwaccess_lock; + + struct drm_mm mm; + spinlock_t mm_lock; + + void __iomem *iomem; + struct clk *clock; + struct regulator *regulator; + struct reset_control *rstc; + + struct panfrost_features features; + + struct panfrost_mmu *mmu; + struct panfrost_job_slot *js; + + struct panfrost_job *jobs[NUM_JOB_SLOTS]; + struct list_head scheduled_jobs; + + struct mutex sched_lock; + + struct { + struct devfreq *devfreq; + struct thermal_cooling_device *cooling; + unsigned long cur_freq; + unsigned long cur_volt; + struct panfrost_devfreq_slot slot[NUM_JOB_SLOTS]; + } devfreq; +}; + +struct panfrost_file_priv { + struct panfrost_device *pfdev; + + struct drm_sched_entity sched_entity[NUM_JOB_SLOTS]; +}; + +static inline struct panfrost_device *to_panfrost_device(struct drm_device *ddev) +{ + return ddev->dev_private; +} + +static inline int panfrost_model_cmp(struct panfrost_device *pfdev, s32 id) +{ + s32 match_id = pfdev->features.id; + + if (match_id & 0xf000) + match_id &= 0xf00f; + return match_id - id; +} + +static inline bool panfrost_model_eq(struct panfrost_device *pfdev, s32 id) +{ + return !panfrost_model_cmp(pfdev, id); +} + +int panfrost_device_init(struct panfrost_device *pfdev); +void panfrost_device_fini(struct panfrost_device *pfdev); + +int panfrost_device_resume(struct device *dev); +int panfrost_device_suspend(struct device *dev); + +const char *panfrost_exception_name(struct panfrost_device *pfdev, u32 exception_code); + +#endif diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c new file mode 100644 index 000000000000..c06af78ab833 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_drv.c @@ -0,0 +1,463 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */ +/* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */ +/* Copyright 2019 Collabora ltd. */ + +#include <linux/bitfield.h> +#include <linux/dma-mapping.h> +#include <linux/module.h> +#include <linux/of_platform.h> +#include <linux/pagemap.h> +#include <linux/pm_runtime.h> +#include <drm/panfrost_drm.h> +#include <drm/drm_drv.h> +#include <drm/drm_ioctl.h> +#include <drm/drm_syncobj.h> +#include <drm/drm_utils.h> + +#include "panfrost_device.h" +#include "panfrost_devfreq.h" +#include "panfrost_gem.h" +#include "panfrost_mmu.h" +#include "panfrost_job.h" +#include "panfrost_gpu.h" + +static int panfrost_ioctl_get_param(struct drm_device *ddev, void *data, struct drm_file *file) +{ + struct drm_panfrost_get_param *param = data; + struct panfrost_device *pfdev = ddev->dev_private; + + if (param->pad != 0) + return -EINVAL; + + switch (param->param) { + case DRM_PANFROST_PARAM_GPU_PROD_ID: + param->value = pfdev->features.id; + break; + default: + return -EINVAL; + } + + return 0; +} + +static int panfrost_ioctl_create_bo(struct drm_device *dev, void *data, + struct drm_file *file) +{ + int ret; + struct drm_gem_shmem_object *shmem; + struct drm_panfrost_create_bo *args = data; + + if (!args->size || args->flags || args->pad) + return -EINVAL; + + shmem = drm_gem_shmem_create_with_handle(file, dev, args->size, + &args->handle); + if (IS_ERR(shmem)) + return PTR_ERR(shmem); + + ret = panfrost_mmu_map(to_panfrost_bo(&shmem->base)); + if (ret) + goto err_free; + + args->offset = to_panfrost_bo(&shmem->base)->node.start << PAGE_SHIFT; + + return 0; + +err_free: + drm_gem_object_put_unlocked(&shmem->base); + return ret; +} + +/** + * panfrost_lookup_bos() - Sets up job->bo[] with the GEM objects + * referenced by the job. + * @dev: DRM device + * @file_priv: DRM file for this fd + * @args: IOCTL args + * @job: job being set up + * + * Resolve handles from userspace to BOs and attach them to job. + * + * Note that this function doesn't need to unreference the BOs on + * failure, because that will happen at panfrost_job_cleanup() time. + */ +static int +panfrost_lookup_bos(struct drm_device *dev, + struct drm_file *file_priv, + struct drm_panfrost_submit *args, + struct panfrost_job *job) +{ + job->bo_count = args->bo_handle_count; + + if (!job->bo_count) + return 0; + + job->implicit_fences = kvmalloc_array(job->bo_count, + sizeof(struct dma_fence *), + GFP_KERNEL | __GFP_ZERO); + if (!job->implicit_fences) + return -ENOMEM; + + return drm_gem_objects_lookup(file_priv, + (void __user *)(uintptr_t)args->bo_handles, + job->bo_count, &job->bos); +} + +/** + * panfrost_copy_in_sync() - Sets up job->in_fences[] with the sync objects + * referenced by the job. + * @dev: DRM device + * @file_priv: DRM file for this fd + * @args: IOCTL args + * @job: job being set up + * + * Resolve syncobjs from userspace to fences and attach them to job. + * + * Note that this function doesn't need to unreference the fences on + * failure, because that will happen at panfrost_job_cleanup() time. + */ +static int +panfrost_copy_in_sync(struct drm_device *dev, + struct drm_file *file_priv, + struct drm_panfrost_submit *args, + struct panfrost_job *job) +{ + u32 *handles; + int ret = 0; + int i; + + job->in_fence_count = args->in_sync_count; + + if (!job->in_fence_count) + return 0; + + job->in_fences = kvmalloc_array(job->in_fence_count, + sizeof(struct dma_fence *), + GFP_KERNEL | __GFP_ZERO); + if (!job->in_fences) { + DRM_DEBUG("Failed to allocate job in fences\n"); + return -ENOMEM; + } + + handles = kvmalloc_array(job->in_fence_count, sizeof(u32), GFP_KERNEL); + if (!handles) { + ret = -ENOMEM; + DRM_DEBUG("Failed to allocate incoming syncobj handles\n"); + goto fail; + } + + if (copy_from_user(handles, + (void __user *)(uintptr_t)args->in_syncs, + job->in_fence_count * sizeof(u32))) { + ret = -EFAULT; + DRM_DEBUG("Failed to copy in syncobj handles\n"); + goto fail; + } + + for (i = 0; i < job->in_fence_count; i++) { + ret = drm_syncobj_find_fence(file_priv, handles[i], 0, 0, + &job->in_fences[i]); + if (ret == -EINVAL) + goto fail; + } + +fail: + kvfree(handles); + return ret; +} + +static int panfrost_ioctl_submit(struct drm_device *dev, void *data, + struct drm_file *file) +{ + struct panfrost_device *pfdev = dev->dev_private; + struct drm_panfrost_submit *args = data; + struct drm_syncobj *sync_out; + struct panfrost_job *job; + int ret = 0; + + job = kzalloc(sizeof(*job), GFP_KERNEL); + if (!job) + return -ENOMEM; + + kref_init(&job->refcount); + + job->pfdev = pfdev; + job->jc = args->jc; + job->requirements = args->requirements; + job->flush_id = panfrost_gpu_get_latest_flush_id(pfdev); + job->file_priv = file->driver_priv; + + ret = panfrost_copy_in_sync(dev, file, args, job); + if (ret) + goto fail; + + ret = panfrost_lookup_bos(dev, file, args, job); + if (ret) + goto fail; + + ret = panfrost_job_push(job); + if (ret) + goto fail; + + /* Update the return sync object for the job */ + sync_out = drm_syncobj_find(file, args->out_sync); + if (sync_out) { + drm_syncobj_replace_fence(sync_out, job->render_done_fence); + drm_syncobj_put(sync_out); + } + +fail: + panfrost_job_put(job); + + return ret; +} + +static int +panfrost_ioctl_wait_bo(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + long ret; + struct drm_panfrost_wait_bo *args = data; + struct drm_gem_object *gem_obj; + unsigned long timeout = drm_timeout_abs_to_jiffies(args->timeout_ns); + + if (args->pad) + return -EINVAL; + + gem_obj = drm_gem_object_lookup(file_priv, args->handle); + if (!gem_obj) + return -ENOENT; + + ret = reservation_object_wait_timeout_rcu(gem_obj->resv, true, + true, timeout); + if (!ret) + ret = timeout ? -ETIMEDOUT : -EBUSY; + + drm_gem_object_put_unlocked(gem_obj); + + return ret; +} + +static int panfrost_ioctl_mmap_bo(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + struct drm_panfrost_mmap_bo *args = data; + struct drm_gem_object *gem_obj; + int ret; + + if (args->flags != 0) { + DRM_INFO("unknown mmap_bo flags: %d\n", args->flags); + return -EINVAL; + } + + gem_obj = drm_gem_object_lookup(file_priv, args->handle); + if (!gem_obj) { + DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle); + return -ENOENT; + } + + ret = drm_gem_create_mmap_offset(gem_obj); + if (ret == 0) + args->offset = drm_vma_node_offset_addr(&gem_obj->vma_node); + drm_gem_object_put_unlocked(gem_obj); + + return ret; +} + +static int panfrost_ioctl_get_bo_offset(struct drm_device *dev, void *data, + struct drm_file *file_priv) +{ + struct drm_panfrost_get_bo_offset *args = data; + struct drm_gem_object *gem_obj; + struct panfrost_gem_object *bo; + + gem_obj = drm_gem_object_lookup(file_priv, args->handle); + if (!gem_obj) { + DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle); + return -ENOENT; + } + bo = to_panfrost_bo(gem_obj); + + args->offset = bo->node.start << PAGE_SHIFT; + + drm_gem_object_put_unlocked(gem_obj); + return 0; +} + +static int +panfrost_open(struct drm_device *dev, struct drm_file *file) +{ + struct panfrost_device *pfdev = dev->dev_private; + struct panfrost_file_priv *panfrost_priv; + + panfrost_priv = kzalloc(sizeof(*panfrost_priv), GFP_KERNEL); + if (!panfrost_priv) + return -ENOMEM; + + panfrost_priv->pfdev = pfdev; + file->driver_priv = panfrost_priv; + + return panfrost_job_open(panfrost_priv); +} + +static void +panfrost_postclose(struct drm_device *dev, struct drm_file *file) +{ + struct panfrost_file_priv *panfrost_priv = file->driver_priv; + + panfrost_job_close(panfrost_priv); + + kfree(panfrost_priv); +} + +/* DRM_AUTH is required on SUBMIT for now, while all clients share a single + * address space. Note that render nodes would be able to submit jobs that + * could access BOs from clients authenticated with the master node. + */ +static const struct drm_ioctl_desc panfrost_drm_driver_ioctls[] = { +#define PANFROST_IOCTL(n, func, flags) \ + DRM_IOCTL_DEF_DRV(PANFROST_##n, panfrost_ioctl_##func, flags) + + PANFROST_IOCTL(SUBMIT, submit, DRM_RENDER_ALLOW | DRM_AUTH), + PANFROST_IOCTL(WAIT_BO, wait_bo, DRM_RENDER_ALLOW), + PANFROST_IOCTL(CREATE_BO, create_bo, DRM_RENDER_ALLOW), + PANFROST_IOCTL(MMAP_BO, mmap_bo, DRM_RENDER_ALLOW), + PANFROST_IOCTL(GET_PARAM, get_param, DRM_RENDER_ALLOW), + PANFROST_IOCTL(GET_BO_OFFSET, get_bo_offset, DRM_RENDER_ALLOW), +}; + +DEFINE_DRM_GEM_SHMEM_FOPS(panfrost_drm_driver_fops); + +static struct drm_driver panfrost_drm_driver = { + .driver_features = DRIVER_RENDER | DRIVER_GEM | DRIVER_PRIME | + DRIVER_SYNCOBJ, + .open = panfrost_open, + .postclose = panfrost_postclose, + .ioctls = panfrost_drm_driver_ioctls, + .num_ioctls = ARRAY_SIZE(panfrost_drm_driver_ioctls), + .fops = &panfrost_drm_driver_fops, + .name = "panfrost", + .desc = "panfrost DRM", + .date = "20180908", + .major = 1, + .minor = 0, + + .gem_create_object = panfrost_gem_create_object, + .prime_handle_to_fd = drm_gem_prime_handle_to_fd, + .prime_fd_to_handle = drm_gem_prime_fd_to_handle, + .gem_prime_import_sg_table = panfrost_gem_prime_import_sg_table, + .gem_prime_mmap = drm_gem_prime_mmap, +}; + +static int panfrost_probe(struct platform_device *pdev) +{ + struct panfrost_device *pfdev; + struct drm_device *ddev; + int err; + + pfdev = devm_kzalloc(&pdev->dev, sizeof(*pfdev), GFP_KERNEL); + if (!pfdev) + return -ENOMEM; + + pfdev->pdev = pdev; + pfdev->dev = &pdev->dev; + + platform_set_drvdata(pdev, pfdev); + + /* Allocate and initialze the DRM device. */ + ddev = drm_dev_alloc(&panfrost_drm_driver, &pdev->dev); + if (IS_ERR(ddev)) + return PTR_ERR(ddev); + + ddev->dev_private = pfdev; + pfdev->ddev = ddev; + + spin_lock_init(&pfdev->mm_lock); + + /* 4G enough for now. can be 48-bit */ + drm_mm_init(&pfdev->mm, SZ_32M >> PAGE_SHIFT, (SZ_4G - SZ_32M) >> PAGE_SHIFT); + + pm_runtime_use_autosuspend(pfdev->dev); + pm_runtime_set_autosuspend_delay(pfdev->dev, 50); /* ~3 frames */ + pm_runtime_enable(pfdev->dev); + + err = panfrost_device_init(pfdev); + if (err) { + dev_err(&pdev->dev, "Fatal error during GPU init\n"); + goto err_out0; + } + + dma_set_mask_and_coherent(pfdev->dev, + DMA_BIT_MASK(FIELD_GET(0xff00, pfdev->features.mmu_features))); + + err = panfrost_devfreq_init(pfdev); + if (err) { + dev_err(&pdev->dev, "Fatal error during devfreq init\n"); + goto err_out1; + } + + /* + * Register the DRM device with the core and the connectors with + * sysfs + */ + err = drm_dev_register(ddev, 0); + if (err < 0) + goto err_out1; + + return 0; + +err_out1: + panfrost_device_fini(pfdev); +err_out0: + drm_dev_put(ddev); + return err; +} + +static int panfrost_remove(struct platform_device *pdev) +{ + struct panfrost_device *pfdev = platform_get_drvdata(pdev); + struct drm_device *ddev = pfdev->ddev; + + drm_dev_unregister(ddev); + pm_runtime_get_sync(pfdev->dev); + pm_runtime_put_sync_autosuspend(pfdev->dev); + pm_runtime_disable(pfdev->dev); + panfrost_device_fini(pfdev); + drm_dev_put(ddev); + return 0; +} + +static const struct of_device_id dt_match[] = { + { .compatible = "arm,mali-t604" }, + { .compatible = "arm,mali-t624" }, + { .compatible = "arm,mali-t628" }, + { .compatible = "arm,mali-t720" }, + { .compatible = "arm,mali-t760" }, + { .compatible = "arm,mali-t820" }, + { .compatible = "arm,mali-t830" }, + { .compatible = "arm,mali-t860" }, + { .compatible = "arm,mali-t880" }, + {} +}; +MODULE_DEVICE_TABLE(of, dt_match); + +static const struct dev_pm_ops panfrost_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, pm_runtime_force_resume) + SET_RUNTIME_PM_OPS(panfrost_device_suspend, panfrost_device_resume, NULL) +}; + +static struct platform_driver panfrost_driver = { + .probe = panfrost_probe, + .remove = panfrost_remove, + .driver = { + .name = "panfrost", + .pm = &panfrost_pm_ops, + .of_match_table = dt_match, + }, +}; +module_platform_driver(panfrost_driver); + +MODULE_AUTHOR("Panfrost Project Developers"); +MODULE_DESCRIPTION("Panfrost DRM Driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/panfrost/panfrost_features.h b/drivers/gpu/drm/panfrost/panfrost_features.h new file mode 100644 index 000000000000..5056777c7744 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_features.h @@ -0,0 +1,309 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* (C) COPYRIGHT 2014-2018 ARM Limited. All rights reserved. */ +/* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */ +#ifndef __PANFROST_FEATURES_H__ +#define __PANFROST_FEATURES_H__ + +#include <linux/bitops.h> + +#include "panfrost_device.h" + +enum panfrost_hw_feature { + HW_FEATURE_JOBCHAIN_DISAMBIGUATION, + HW_FEATURE_PWRON_DURING_PWROFF_TRANS, + HW_FEATURE_XAFFINITY, + HW_FEATURE_OUT_OF_ORDER_EXEC, + HW_FEATURE_MRT, + HW_FEATURE_BRNDOUT_CC, + HW_FEATURE_INTERPIPE_REG_ALIASING, + HW_FEATURE_LD_ST_TILEBUFFER, + HW_FEATURE_MSAA_16X, + HW_FEATURE_32_BIT_UNIFORM_ADDRESS, + HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL, + HW_FEATURE_OPTIMIZED_COVERAGE_MASK, + HW_FEATURE_T7XX_PAIRING_RULES, + HW_FEATURE_LD_ST_LEA_TEX, + HW_FEATURE_LINEAR_FILTER_FLOAT, + HW_FEATURE_WORKGROUP_ROUND_MULTIPLE_OF_4, + HW_FEATURE_IMAGES_IN_FRAGMENT_SHADERS, + HW_FEATURE_TEST4_DATUM_MODE, + HW_FEATURE_NEXT_INSTRUCTION_TYPE, + HW_FEATURE_BRNDOUT_KILL, + HW_FEATURE_WARPING, + HW_FEATURE_V4, + HW_FEATURE_FLUSH_REDUCTION, + HW_FEATURE_PROTECTED_MODE, + HW_FEATURE_COHERENCY_REG, + HW_FEATURE_PROTECTED_DEBUG_MODE, + HW_FEATURE_AARCH64_MMU, + HW_FEATURE_TLS_HASHING, + HW_FEATURE_THREAD_GROUP_SPLIT, + HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG, +}; + +#define hw_features_t600 (\ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ + BIT_ULL(HW_FEATURE_V4)) + +#define hw_features_t620 (\ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ + BIT_ULL(HW_FEATURE_V4)) + +#define hw_features_t720 (\ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ + BIT_ULL(HW_FEATURE_OPTIMIZED_COVERAGE_MASK) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ + BIT_ULL(HW_FEATURE_WORKGROUP_ROUND_MULTIPLE_OF_4) | \ + BIT_ULL(HW_FEATURE_WARPING) | \ + BIT_ULL(HW_FEATURE_V4)) + + +#define hw_features_t760 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_MSAA_16X) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT)) + +// T860 +#define hw_features_t860 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_MSAA_16X) | \ + BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT)) + +#define hw_features_t880 hw_features_t860 + +#define hw_features_t830 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_WARPING) | \ + BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT)) + +#define hw_features_t820 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_WARPING) | \ + BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT)) + +#define hw_features_g71 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_WARPING) | \ + BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_MSAA_16X) | \ + BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ + BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ + BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ + BIT_ULL(HW_FEATURE_COHERENCY_REG)) + +#define hw_features_g72 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_WARPING) | \ + BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_MSAA_16X) | \ + BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ + BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ + BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ + BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \ + BIT_ULL(HW_FEATURE_COHERENCY_REG)) + +#define hw_features_g51 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_WARPING) | \ + BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_MSAA_16X) | \ + BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ + BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ + BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ + BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \ + BIT_ULL(HW_FEATURE_COHERENCY_REG)) + +#define hw_features_g52 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_WARPING) | \ + BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_MSAA_16X) | \ + BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ + BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ + BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ + BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \ + BIT_ULL(HW_FEATURE_COHERENCY_REG)) + +#define hw_features_g76 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_WARPING) | \ + BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_MSAA_16X) | \ + BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ + BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ + BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ + BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \ + BIT_ULL(HW_FEATURE_COHERENCY_REG) | \ + BIT_ULL(HW_FEATURE_AARCH64_MMU) | \ + BIT_ULL(HW_FEATURE_TLS_HASHING) | \ + BIT_ULL(HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG)) + +#define hw_features_g31 (\ + BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ + BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ + BIT_ULL(HW_FEATURE_XAFFINITY) | \ + BIT_ULL(HW_FEATURE_WARPING) | \ + BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ + BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ + BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ + BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ + BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ + BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ + BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ + BIT_ULL(HW_FEATURE_MRT) | \ + BIT_ULL(HW_FEATURE_MSAA_16X) | \ + BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ + BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ + BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ + BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ + BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ + BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ + BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ + BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \ + BIT_ULL(HW_FEATURE_COHERENCY_REG) | \ + BIT_ULL(HW_FEATURE_AARCH64_MMU) | \ + BIT_ULL(HW_FEATURE_TLS_HASHING) | \ + BIT_ULL(HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG)) + +static inline bool panfrost_has_hw_feature(struct panfrost_device *pfdev, + enum panfrost_hw_feature feat) +{ + return test_bit(feat, pfdev->features.hw_features); +} + +#endif diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.c b/drivers/gpu/drm/panfrost/panfrost_gem.c new file mode 100644 index 000000000000..8a0376283a21 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_gem.c @@ -0,0 +1,95 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */ + +#include <linux/err.h> +#include <linux/slab.h> +#include <linux/dma-buf.h> +#include <linux/dma-mapping.h> + +#include <drm/panfrost_drm.h> +#include "panfrost_device.h" +#include "panfrost_gem.h" +#include "panfrost_mmu.h" + +/* Called DRM core on the last userspace/kernel unreference of the + * BO. + */ +void panfrost_gem_free_object(struct drm_gem_object *obj) +{ + struct panfrost_gem_object *bo = to_panfrost_bo(obj); + struct panfrost_device *pfdev = obj->dev->dev_private; + + panfrost_mmu_unmap(bo); + + spin_lock(&pfdev->mm_lock); + drm_mm_remove_node(&bo->node); + spin_unlock(&pfdev->mm_lock); + + drm_gem_shmem_free_object(obj); +} + +static const struct drm_gem_object_funcs panfrost_gem_funcs = { + .free = panfrost_gem_free_object, + .print_info = drm_gem_shmem_print_info, + .pin = drm_gem_shmem_pin, + .unpin = drm_gem_shmem_unpin, + .get_sg_table = drm_gem_shmem_get_sg_table, + .vmap = drm_gem_shmem_vmap, + .vunmap = drm_gem_shmem_vunmap, + .vm_ops = &drm_gem_shmem_vm_ops, +}; + +/** + * panfrost_gem_create_object - Implementation of driver->gem_create_object. + * @dev: DRM device + * @size: Size in bytes of the memory the object will reference + * + * This lets the GEM helpers allocate object structs for us, and keep + * our BO stats correct. + */ +struct drm_gem_object *panfrost_gem_create_object(struct drm_device *dev, size_t size) +{ + int ret; + struct panfrost_device *pfdev = dev->dev_private; + struct panfrost_gem_object *obj; + + obj = kzalloc(sizeof(*obj), GFP_KERNEL); + if (!obj) + return NULL; + + obj->base.base.funcs = &panfrost_gem_funcs; + + spin_lock(&pfdev->mm_lock); + ret = drm_mm_insert_node(&pfdev->mm, &obj->node, + roundup(size, PAGE_SIZE) >> PAGE_SHIFT); + spin_unlock(&pfdev->mm_lock); + if (ret) + goto free_obj; + + return &obj->base.base; + +free_obj: + kfree(obj); + return ERR_PTR(ret); +} + +struct drm_gem_object * +panfrost_gem_prime_import_sg_table(struct drm_device *dev, + struct dma_buf_attachment *attach, + struct sg_table *sgt) +{ + struct drm_gem_object *obj; + struct panfrost_gem_object *pobj; + + obj = drm_gem_shmem_prime_import_sg_table(dev, attach, sgt); + if (IS_ERR(obj)) + return ERR_CAST(obj); + + pobj = to_panfrost_bo(obj); + + obj->resv = attach->dmabuf->resv; + + panfrost_mmu_map(pobj); + + return obj; +} diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.h b/drivers/gpu/drm/panfrost/panfrost_gem.h new file mode 100644 index 000000000000..045000eb5fcf --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_gem.h @@ -0,0 +1,29 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */ + +#ifndef __PANFROST_GEM_H__ +#define __PANFROST_GEM_H__ + +#include <drm/drm_gem_shmem_helper.h> +#include <drm/drm_mm.h> + +struct panfrost_gem_object { + struct drm_gem_shmem_object base; + + struct drm_mm_node node; +}; + +static inline +struct panfrost_gem_object *to_panfrost_bo(struct drm_gem_object *obj) +{ + return container_of(to_drm_gem_shmem_obj(obj), struct panfrost_gem_object, base); +} + +struct drm_gem_object *panfrost_gem_create_object(struct drm_device *dev, size_t size); + +struct drm_gem_object * +panfrost_gem_prime_import_sg_table(struct drm_device *dev, + struct dma_buf_attachment *attach, + struct sg_table *sgt); + +#endif /* __PANFROST_GEM_H__ */ diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.c b/drivers/gpu/drm/panfrost/panfrost_gpu.c new file mode 100644 index 000000000000..aceaf6e44a09 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_gpu.c @@ -0,0 +1,362 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */ +/* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */ +/* Copyright 2019 Collabora ltd. */ +#include <linux/bitmap.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/io.h> +#include <linux/iopoll.h> +#include <linux/platform_device.h> + +#include "panfrost_device.h" +#include "panfrost_features.h" +#include "panfrost_issues.h" +#include "panfrost_gpu.h" +#include "panfrost_regs.h" + +#define gpu_write(dev, reg, data) writel(data, dev->iomem + reg) +#define gpu_read(dev, reg) readl(dev->iomem + reg) + +static irqreturn_t panfrost_gpu_irq_handler(int irq, void *data) +{ + struct panfrost_device *pfdev = data; + u32 state = gpu_read(pfdev, GPU_INT_STAT); + u32 fault_status = gpu_read(pfdev, GPU_FAULT_STATUS); + + if (!state) + return IRQ_NONE; + + if (state & GPU_IRQ_MASK_ERROR) { + u64 address = (u64) gpu_read(pfdev, GPU_FAULT_ADDRESS_HI) << 32; + address |= gpu_read(pfdev, GPU_FAULT_ADDRESS_LO); + + dev_warn(pfdev->dev, "GPU Fault 0x%08x (%s) at 0x%016llx\n", + fault_status & 0xFF, panfrost_exception_name(pfdev, fault_status), + address); + + if (state & GPU_IRQ_MULTIPLE_FAULT) + dev_warn(pfdev->dev, "There were multiple GPU faults - some have not been reported\n"); + + gpu_write(pfdev, GPU_INT_MASK, 0); + } + + gpu_write(pfdev, GPU_INT_CLEAR, state); + + return IRQ_HANDLED; +} + +int panfrost_gpu_soft_reset(struct panfrost_device *pfdev) +{ + int ret; + u32 val; + + gpu_write(pfdev, GPU_INT_MASK, 0); + gpu_write(pfdev, GPU_INT_CLEAR, GPU_IRQ_RESET_COMPLETED); + gpu_write(pfdev, GPU_CMD, GPU_CMD_SOFT_RESET); + + ret = readl_relaxed_poll_timeout(pfdev->iomem + GPU_INT_RAWSTAT, + val, val & GPU_IRQ_RESET_COMPLETED, 100, 10000); + + if (ret) { + dev_err(pfdev->dev, "gpu soft reset timed out\n"); + return ret; + } + + gpu_write(pfdev, GPU_INT_CLEAR, GPU_IRQ_MASK_ALL); + gpu_write(pfdev, GPU_INT_MASK, GPU_IRQ_MASK_ALL); + + return 0; +} + +static void panfrost_gpu_init_quirks(struct panfrost_device *pfdev) +{ + u32 quirks = 0; + + if (panfrost_has_hw_issue(pfdev, HW_ISSUE_8443) || + panfrost_has_hw_issue(pfdev, HW_ISSUE_11035)) + quirks |= SC_LS_PAUSEBUFFER_DISABLE; + + if (panfrost_has_hw_issue(pfdev, HW_ISSUE_10327)) + quirks |= SC_SDC_DISABLE_OQ_DISCARD; + + if (panfrost_has_hw_issue(pfdev, HW_ISSUE_10797)) + quirks |= SC_ENABLE_TEXGRD_FLAGS; + + if (!panfrost_has_hw_issue(pfdev, GPUCORE_1619)) { + if (panfrost_model_cmp(pfdev, 0x750) < 0) /* T60x, T62x, T72x */ + quirks |= SC_LS_ATTR_CHECK_DISABLE; + else if (panfrost_model_cmp(pfdev, 0x880) <= 0) /* T76x, T8xx */ + quirks |= SC_LS_ALLOW_ATTR_TYPES; + } + + if (panfrost_has_hw_feature(pfdev, HW_FEATURE_TLS_HASHING)) + quirks |= SC_TLS_HASH_ENABLE; + + if (quirks) + gpu_write(pfdev, GPU_SHADER_CONFIG, quirks); + + + quirks = gpu_read(pfdev, GPU_TILER_CONFIG); + + /* Set tiler clock gate override if required */ + if (panfrost_has_hw_issue(pfdev, HW_ISSUE_T76X_3953)) + quirks |= TC_CLOCK_GATE_OVERRIDE; + + gpu_write(pfdev, GPU_TILER_CONFIG, quirks); + + + quirks = gpu_read(pfdev, GPU_L2_MMU_CONFIG); + + /* Limit read & write ID width for AXI */ + if (panfrost_has_hw_feature(pfdev, HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG)) + quirks &= ~(L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS | + L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES); + else + quirks &= ~(L2_MMU_CONFIG_LIMIT_EXTERNAL_READS | + L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES); + + gpu_write(pfdev, GPU_L2_MMU_CONFIG, quirks); + + quirks = 0; + if ((panfrost_model_eq(pfdev, 0x860) || panfrost_model_eq(pfdev, 0x880)) && + pfdev->features.revision >= 0x2000) + quirks |= JM_MAX_JOB_THROTTLE_LIMIT << JM_JOB_THROTTLE_LIMIT_SHIFT; + else if (panfrost_model_eq(pfdev, 0x6000) && + pfdev->features.coherency_features == COHERENCY_ACE) + quirks |= (COHERENCY_ACE_LITE | COHERENCY_ACE) << + JM_FORCE_COHERENCY_FEATURES_SHIFT; + + if (quirks) + gpu_write(pfdev, GPU_JM_CONFIG, quirks); +} + +#define MAX_HW_REVS 6 + +struct panfrost_model { + const char *name; + u32 id; + u32 id_mask; + u64 features; + u64 issues; + struct { + u32 revision; + u64 issues; + } revs[MAX_HW_REVS]; +}; + +#define GPU_MODEL(_name, _id, ...) \ +{\ + .name = __stringify(_name), \ + .id = _id, \ + .features = hw_features_##_name, \ + .issues = hw_issues_##_name, \ + .revs = { __VA_ARGS__ }, \ +} + +#define GPU_REV_EXT(name, _rev, _p, _s, stat) \ +{\ + .revision = (_rev) << 12 | (_p) << 4 | (_s), \ + .issues = hw_issues_##name##_r##_rev##p##_p##stat, \ +} +#define GPU_REV(name, r, p) GPU_REV_EXT(name, r, p, 0, ) + +static const struct panfrost_model gpu_models[] = { + /* T60x has an oddball version */ + GPU_MODEL(t600, 0x600, + GPU_REV_EXT(t600, 0, 0, 1, _15dev0)), + GPU_MODEL(t620, 0x620, + GPU_REV(t620, 0, 1), GPU_REV(t620, 1, 0)), + GPU_MODEL(t720, 0x720), + GPU_MODEL(t760, 0x750, + GPU_REV(t760, 0, 0), GPU_REV(t760, 0, 1), + GPU_REV_EXT(t760, 0, 1, 0, _50rel0), + GPU_REV(t760, 0, 2), GPU_REV(t760, 0, 3)), + GPU_MODEL(t820, 0x820), + GPU_MODEL(t830, 0x830), + GPU_MODEL(t860, 0x860), + GPU_MODEL(t880, 0x880), + + GPU_MODEL(g71, 0x6000, + GPU_REV_EXT(g71, 0, 0, 1, _05dev0)), + GPU_MODEL(g72, 0x6001), + GPU_MODEL(g51, 0x7000), + GPU_MODEL(g76, 0x7001), + GPU_MODEL(g52, 0x7002), + GPU_MODEL(g31, 0x7003, + GPU_REV(g31, 1, 0)), +}; + +static void panfrost_gpu_init_features(struct panfrost_device *pfdev) +{ + u32 gpu_id, num_js, major, minor, status, rev; + const char *name = "unknown"; + u64 hw_feat = 0; + u64 hw_issues = hw_issues_all; + const struct panfrost_model *model; + int i; + + pfdev->features.l2_features = gpu_read(pfdev, GPU_L2_FEATURES); + pfdev->features.core_features = gpu_read(pfdev, GPU_CORE_FEATURES); + pfdev->features.tiler_features = gpu_read(pfdev, GPU_TILER_FEATURES); + pfdev->features.mem_features = gpu_read(pfdev, GPU_MEM_FEATURES); + pfdev->features.mmu_features = gpu_read(pfdev, GPU_MMU_FEATURES); + pfdev->features.thread_features = gpu_read(pfdev, GPU_THREAD_FEATURES); + pfdev->features.coherency_features = gpu_read(pfdev, GPU_COHERENCY_FEATURES); + for (i = 0; i < 4; i++) + pfdev->features.texture_features[i] = gpu_read(pfdev, GPU_TEXTURE_FEATURES(i)); + + pfdev->features.as_present = gpu_read(pfdev, GPU_AS_PRESENT); + + pfdev->features.js_present = gpu_read(pfdev, GPU_JS_PRESENT); + num_js = hweight32(pfdev->features.js_present); + for (i = 0; i < num_js; i++) + pfdev->features.js_features[i] = gpu_read(pfdev, GPU_JS_FEATURES(i)); + + pfdev->features.shader_present = gpu_read(pfdev, GPU_SHADER_PRESENT_LO); + pfdev->features.shader_present |= (u64)gpu_read(pfdev, GPU_SHADER_PRESENT_HI) << 32; + + pfdev->features.tiler_present = gpu_read(pfdev, GPU_TILER_PRESENT_LO); + pfdev->features.tiler_present |= (u64)gpu_read(pfdev, GPU_TILER_PRESENT_HI) << 32; + + pfdev->features.l2_present = gpu_read(pfdev, GPU_L2_PRESENT_LO); + pfdev->features.l2_present |= (u64)gpu_read(pfdev, GPU_L2_PRESENT_HI) << 32; + pfdev->features.nr_core_groups = hweight64(pfdev->features.l2_present); + + pfdev->features.stack_present = gpu_read(pfdev, GPU_STACK_PRESENT_LO); + pfdev->features.stack_present |= (u64)gpu_read(pfdev, GPU_STACK_PRESENT_HI) << 32; + + gpu_id = gpu_read(pfdev, GPU_ID); + pfdev->features.revision = gpu_id & 0xffff; + pfdev->features.id = gpu_id >> 16; + + /* The T60x has an oddball ID value. Fix it up to the standard Midgard + * format so we (and userspace) don't have to special case it. + */ + if (pfdev->features.id == 0x6956) + pfdev->features.id = 0x0600; + + major = (pfdev->features.revision >> 12) & 0xf; + minor = (pfdev->features.revision >> 4) & 0xff; + status = pfdev->features.revision & 0xf; + rev = pfdev->features.revision; + + gpu_id = pfdev->features.id; + + for (model = gpu_models; model->name; model++) { + int best = -1; + + if (!panfrost_model_eq(pfdev, model->id)) + continue; + + name = model->name; + hw_feat = model->features; + hw_issues |= model->issues; + for (i = 0; i < MAX_HW_REVS; i++) { + if (model->revs[i].revision == rev) { + best = i; + break; + } else if (model->revs[i].revision == (rev & ~0xf)) + best = i; + } + + if (best >= 0) + hw_issues |= model->revs[best].issues; + + break; + } + + bitmap_from_u64(pfdev->features.hw_features, hw_feat); + bitmap_from_u64(pfdev->features.hw_issues, hw_issues); + + dev_info(pfdev->dev, "mali-%s id 0x%x major 0x%x minor 0x%x status 0x%x", + name, gpu_id, major, minor, status); + dev_info(pfdev->dev, "features: %64pb, issues: %64pb", + pfdev->features.hw_features, + pfdev->features.hw_issues); + + dev_info(pfdev->dev, "Features: L2:0x%08x Shader:0x%08x Tiler:0x%08x Mem:0x%0x MMU:0x%08x AS:0x%x JS:0x%x", + gpu_read(pfdev, GPU_L2_FEATURES), + gpu_read(pfdev, GPU_CORE_FEATURES), + gpu_read(pfdev, GPU_TILER_FEATURES), + gpu_read(pfdev, GPU_MEM_FEATURES), + gpu_read(pfdev, GPU_MMU_FEATURES), + gpu_read(pfdev, GPU_AS_PRESENT), + gpu_read(pfdev, GPU_JS_PRESENT)); + + dev_info(pfdev->dev, "shader_present=0x%0llx l2_present=0x%0llx", + pfdev->features.shader_present, pfdev->features.l2_present); +} + +void panfrost_gpu_power_on(struct panfrost_device *pfdev) +{ + int ret; + u32 val; + + /* Just turn on everything for now */ + gpu_write(pfdev, L2_PWRON_LO, pfdev->features.l2_present); + ret = readl_relaxed_poll_timeout(pfdev->iomem + L2_READY_LO, + val, val == pfdev->features.l2_present, 100, 1000); + + gpu_write(pfdev, STACK_PWRON_LO, pfdev->features.stack_present); + ret |= readl_relaxed_poll_timeout(pfdev->iomem + STACK_READY_LO, + val, val == pfdev->features.stack_present, 100, 1000); + + gpu_write(pfdev, SHADER_PWRON_LO, pfdev->features.shader_present); + ret |= readl_relaxed_poll_timeout(pfdev->iomem + SHADER_READY_LO, + val, val == pfdev->features.shader_present, 100, 1000); + + gpu_write(pfdev, TILER_PWRON_LO, pfdev->features.tiler_present); + ret |= readl_relaxed_poll_timeout(pfdev->iomem + TILER_READY_LO, + val, val == pfdev->features.tiler_present, 100, 1000); + + if (ret) + dev_err(pfdev->dev, "error powering up gpu"); +} + +void panfrost_gpu_power_off(struct panfrost_device *pfdev) +{ + gpu_write(pfdev, TILER_PWROFF_LO, 0); + gpu_write(pfdev, SHADER_PWROFF_LO, 0); + gpu_write(pfdev, STACK_PWROFF_LO, 0); + gpu_write(pfdev, L2_PWROFF_LO, 0); +} + +int panfrost_gpu_init(struct panfrost_device *pfdev) +{ + int err, irq; + + err = panfrost_gpu_soft_reset(pfdev); + if (err) + return err; + + panfrost_gpu_init_features(pfdev); + + irq = platform_get_irq_byname(to_platform_device(pfdev->dev), "gpu"); + if (irq <= 0) + return -ENODEV; + + err = devm_request_irq(pfdev->dev, irq, panfrost_gpu_irq_handler, + IRQF_SHARED, "gpu", pfdev); + if (err) { + dev_err(pfdev->dev, "failed to request gpu irq"); + return err; + } + + panfrost_gpu_init_quirks(pfdev); + panfrost_gpu_power_on(pfdev); + + return 0; +} + +void panfrost_gpu_fini(struct panfrost_device *pfdev) +{ + panfrost_gpu_power_off(pfdev); +} + +u32 panfrost_gpu_get_latest_flush_id(struct panfrost_device *pfdev) +{ + if (panfrost_has_hw_feature(pfdev, HW_FEATURE_FLUSH_REDUCTION)) + return gpu_read(pfdev, GPU_LATEST_FLUSH_ID); + return 0; +} diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.h b/drivers/gpu/drm/panfrost/panfrost_gpu.h new file mode 100644 index 000000000000..4112412087b2 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_gpu.h @@ -0,0 +1,19 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */ +/* Copyright 2019 Collabora ltd. */ + +#ifndef __PANFROST_GPU_H__ +#define __PANFROST_GPU_H__ + +struct panfrost_device; + +int panfrost_gpu_init(struct panfrost_device *pfdev); +void panfrost_gpu_fini(struct panfrost_device *pfdev); + +u32 panfrost_gpu_get_latest_flush_id(struct panfrost_device *pfdev); + +int panfrost_gpu_soft_reset(struct panfrost_device *pfdev); +void panfrost_gpu_power_on(struct panfrost_device *pfdev); +void panfrost_gpu_power_off(struct panfrost_device *pfdev); + +#endif diff --git a/drivers/gpu/drm/panfrost/panfrost_issues.h b/drivers/gpu/drm/panfrost/panfrost_issues.h new file mode 100644 index 000000000000..cec6dcdadb5c --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_issues.h @@ -0,0 +1,176 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* (C) COPYRIGHT 2014-2018 ARM Limited. All rights reserved. */ +/* Copyright 2019 Linaro, Ltd., Rob Herring <robh@kernel.org> */ +#ifndef __PANFROST_ISSUES_H__ +#define __PANFROST_ISSUES_H__ + +#include <linux/bitops.h> + +#include "panfrost_device.h" + +/* + * This is not a complete list of issues, but only the ones the driver needs + * to care about. + */ +enum panfrost_hw_issue { + HW_ISSUE_6367, + HW_ISSUE_6787, + HW_ISSUE_8186, + HW_ISSUE_8245, + HW_ISSUE_8316, + HW_ISSUE_8394, + HW_ISSUE_8401, + HW_ISSUE_8408, + HW_ISSUE_8443, + HW_ISSUE_8987, + HW_ISSUE_9435, + HW_ISSUE_9510, + HW_ISSUE_9630, + HW_ISSUE_10327, + HW_ISSUE_10649, + HW_ISSUE_10676, + HW_ISSUE_10797, + HW_ISSUE_10817, + HW_ISSUE_10883, + HW_ISSUE_10959, + HW_ISSUE_10969, + HW_ISSUE_11020, + HW_ISSUE_11024, + HW_ISSUE_11035, + HW_ISSUE_11056, + HW_ISSUE_T76X_3542, + HW_ISSUE_T76X_3953, + HW_ISSUE_TMIX_8463, + GPUCORE_1619, + HW_ISSUE_TMIX_8438, + HW_ISSUE_TGOX_R1_1234, + HW_ISSUE_END +}; + +#define hw_issues_all (\ + BIT_ULL(HW_ISSUE_9435)) + +#define hw_issues_t600 (\ + BIT_ULL(HW_ISSUE_6367) | \ + BIT_ULL(HW_ISSUE_6787) | \ + BIT_ULL(HW_ISSUE_8408) | \ + BIT_ULL(HW_ISSUE_9510) | \ + BIT_ULL(HW_ISSUE_10649) | \ + BIT_ULL(HW_ISSUE_10676) | \ + BIT_ULL(HW_ISSUE_10883) | \ + BIT_ULL(HW_ISSUE_11020) | \ + BIT_ULL(HW_ISSUE_11035) | \ + BIT_ULL(HW_ISSUE_11056) | \ + BIT_ULL(HW_ISSUE_TMIX_8438)) + +#define hw_issues_t600_r0p0_15dev0 (\ + BIT_ULL(HW_ISSUE_8186) | \ + BIT_ULL(HW_ISSUE_8245) | \ + BIT_ULL(HW_ISSUE_8316) | \ + BIT_ULL(HW_ISSUE_8394) | \ + BIT_ULL(HW_ISSUE_8401) | \ + BIT_ULL(HW_ISSUE_8443) | \ + BIT_ULL(HW_ISSUE_8987) | \ + BIT_ULL(HW_ISSUE_9630) | \ + BIT_ULL(HW_ISSUE_10969) | \ + BIT_ULL(GPUCORE_1619)) + +#define hw_issues_t620 (\ + BIT_ULL(HW_ISSUE_10649) | \ + BIT_ULL(HW_ISSUE_10883) | \ + BIT_ULL(HW_ISSUE_10959) | \ + BIT_ULL(HW_ISSUE_11056) | \ + BIT_ULL(HW_ISSUE_TMIX_8438)) + +#define hw_issues_t620_r0p1 (\ + BIT_ULL(HW_ISSUE_10327) | \ + BIT_ULL(HW_ISSUE_10676) | \ + BIT_ULL(HW_ISSUE_10817) | \ + BIT_ULL(HW_ISSUE_11020) | \ + BIT_ULL(HW_ISSUE_11024) | \ + BIT_ULL(HW_ISSUE_11035)) + +#define hw_issues_t620_r1p0 (\ + BIT_ULL(HW_ISSUE_11020) | \ + BIT_ULL(HW_ISSUE_11024)) + +#define hw_issues_t720 (\ + BIT_ULL(HW_ISSUE_10649) | \ + BIT_ULL(HW_ISSUE_10797) | \ + BIT_ULL(HW_ISSUE_10883) | \ + BIT_ULL(HW_ISSUE_11056) | \ + BIT_ULL(HW_ISSUE_TMIX_8438)) + +#define hw_issues_t760 (\ + BIT_ULL(HW_ISSUE_10883) | \ + BIT_ULL(HW_ISSUE_T76X_3953) | \ + BIT_ULL(HW_ISSUE_TMIX_8438)) + +#define hw_issues_t760_r0p0 (\ + BIT_ULL(HW_ISSUE_11020) | \ + BIT_ULL(HW_ISSUE_11024) | \ + BIT_ULL(HW_ISSUE_T76X_3542)) + +#define hw_issues_t760_r0p1 (\ + BIT_ULL(HW_ISSUE_11020) | \ + BIT_ULL(HW_ISSUE_11024) | \ + BIT_ULL(HW_ISSUE_T76X_3542)) + +#define hw_issues_t760_r0p1_50rel0 (\ + BIT_ULL(HW_ISSUE_T76X_3542)) + +#define hw_issues_t760_r0p2 (\ + BIT_ULL(HW_ISSUE_11020) | \ + BIT_ULL(HW_ISSUE_11024) | \ + BIT_ULL(HW_ISSUE_T76X_3542)) + +#define hw_issues_t760_r0p3 (\ + BIT_ULL(HW_ISSUE_T76X_3542)) + +#define hw_issues_t820 (\ + BIT_ULL(HW_ISSUE_10883) | \ + BIT_ULL(HW_ISSUE_T76X_3953) | \ + BIT_ULL(HW_ISSUE_TMIX_8438)) + +#define hw_issues_t830 (\ + BIT_ULL(HW_ISSUE_10883) | \ + BIT_ULL(HW_ISSUE_T76X_3953) | \ + BIT_ULL(HW_ISSUE_TMIX_8438)) + +#define hw_issues_t860 (\ + BIT_ULL(HW_ISSUE_10883) | \ + BIT_ULL(HW_ISSUE_T76X_3953) | \ + BIT_ULL(HW_ISSUE_TMIX_8438)) + +#define hw_issues_t880 (\ + BIT_ULL(HW_ISSUE_10883) | \ + BIT_ULL(HW_ISSUE_T76X_3953) | \ + BIT_ULL(HW_ISSUE_TMIX_8438)) + +#define hw_issues_g31 0 + +#define hw_issues_g31_r1p0 (\ + BIT_ULL(HW_ISSUE_TGOX_R1_1234)) + +#define hw_issues_g51 0 + +#define hw_issues_g52 0 + +#define hw_issues_g71 (\ + BIT_ULL(HW_ISSUE_TMIX_8463) | \ + BIT_ULL(HW_ISSUE_TMIX_8438)) + +#define hw_issues_g71_r0p0_05dev0 (\ + BIT_ULL(HW_ISSUE_T76X_3953)) + +#define hw_issues_g72 0 + +#define hw_issues_g76 0 + +static inline bool panfrost_has_hw_issue(struct panfrost_device *pfdev, + enum panfrost_hw_issue issue) +{ + return test_bit(issue, pfdev->features.hw_issues); +} + +#endif /* __PANFROST_ISSUES_H__ */ diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c new file mode 100644 index 000000000000..0a7ed04f7d52 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_job.c @@ -0,0 +1,560 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */ +/* Copyright 2019 Collabora ltd. */ +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/io.h> +#include <linux/platform_device.h> +#include <linux/pm_runtime.h> +#include <linux/reservation.h> +#include <drm/gpu_scheduler.h> +#include <drm/panfrost_drm.h> + +#include "panfrost_device.h" +#include "panfrost_devfreq.h" +#include "panfrost_job.h" +#include "panfrost_features.h" +#include "panfrost_issues.h" +#include "panfrost_gem.h" +#include "panfrost_regs.h" +#include "panfrost_gpu.h" +#include "panfrost_mmu.h" + +#define job_write(dev, reg, data) writel(data, dev->iomem + (reg)) +#define job_read(dev, reg) readl(dev->iomem + (reg)) + +struct panfrost_queue_state { + struct drm_gpu_scheduler sched; + + u64 fence_context; + u64 emit_seqno; +}; + +struct panfrost_job_slot { + struct panfrost_queue_state queue[NUM_JOB_SLOTS]; + spinlock_t job_lock; +}; + +static struct panfrost_job * +to_panfrost_job(struct drm_sched_job *sched_job) +{ + return container_of(sched_job, struct panfrost_job, base); +} + +struct panfrost_fence { + struct dma_fence base; + struct drm_device *dev; + /* panfrost seqno for signaled() test */ + u64 seqno; + int queue; +}; + +static inline struct panfrost_fence * +to_panfrost_fence(struct dma_fence *fence) +{ + return (struct panfrost_fence *)fence; +} + +static const char *panfrost_fence_get_driver_name(struct dma_fence *fence) +{ + return "panfrost"; +} + +static const char *panfrost_fence_get_timeline_name(struct dma_fence *fence) +{ + struct panfrost_fence *f = to_panfrost_fence(fence); + + switch (f->queue) { + case 0: + return "panfrost-js-0"; + case 1: + return "panfrost-js-1"; + case 2: + return "panfrost-js-2"; + default: + return NULL; + } +} + +static const struct dma_fence_ops panfrost_fence_ops = { + .get_driver_name = panfrost_fence_get_driver_name, + .get_timeline_name = panfrost_fence_get_timeline_name, +}; + +static struct dma_fence *panfrost_fence_create(struct panfrost_device *pfdev, int js_num) +{ + struct panfrost_fence *fence; + struct panfrost_job_slot *js = pfdev->js; + + fence = kzalloc(sizeof(*fence), GFP_KERNEL); + if (!fence) + return ERR_PTR(-ENOMEM); + + fence->dev = pfdev->ddev; + fence->queue = js_num; + fence->seqno = ++js->queue[js_num].emit_seqno; + dma_fence_init(&fence->base, &panfrost_fence_ops, &js->job_lock, + js->queue[js_num].fence_context, fence->seqno); + + return &fence->base; +} + +static int panfrost_job_get_slot(struct panfrost_job *job) +{ + /* JS0: fragment jobs. + * JS1: vertex/tiler jobs + * JS2: compute jobs + */ + if (job->requirements & PANFROST_JD_REQ_FS) + return 0; + +/* Not exposed to userspace yet */ +#if 0 + if (job->requirements & PANFROST_JD_REQ_ONLY_COMPUTE) { + if ((job->requirements & PANFROST_JD_REQ_CORE_GRP_MASK) && + (job->pfdev->features.nr_core_groups == 2)) + return 2; + if (panfrost_has_hw_issue(job->pfdev, HW_ISSUE_8987)) + return 2; + } +#endif + return 1; +} + +static void panfrost_job_write_affinity(struct panfrost_device *pfdev, + u32 requirements, + int js) +{ + u64 affinity; + + /* + * Use all cores for now. + * Eventually we may need to support tiler only jobs and h/w with + * multiple (2) coherent core groups + */ + affinity = pfdev->features.shader_present; + + job_write(pfdev, JS_AFFINITY_NEXT_LO(js), affinity & 0xFFFFFFFF); + job_write(pfdev, JS_AFFINITY_NEXT_HI(js), affinity >> 32); +} + +static void panfrost_job_hw_submit(struct panfrost_job *job, int js) +{ + struct panfrost_device *pfdev = job->pfdev; + unsigned long flags; + u32 cfg; + u64 jc_head = job->jc; + int ret; + + ret = pm_runtime_get_sync(pfdev->dev); + if (ret < 0) + return; + + if (WARN_ON(job_read(pfdev, JS_COMMAND_NEXT(js)))) + goto end; + + panfrost_devfreq_record_transition(pfdev, js); + spin_lock_irqsave(&pfdev->hwaccess_lock, flags); + + job_write(pfdev, JS_HEAD_NEXT_LO(js), jc_head & 0xFFFFFFFF); + job_write(pfdev, JS_HEAD_NEXT_HI(js), jc_head >> 32); + + panfrost_job_write_affinity(pfdev, job->requirements, js); + + /* start MMU, medium priority, cache clean/flush on end, clean/flush on + * start */ + /* TODO: different address spaces */ + cfg = JS_CONFIG_THREAD_PRI(8) | + JS_CONFIG_START_FLUSH_CLEAN_INVALIDATE | + JS_CONFIG_END_FLUSH_CLEAN_INVALIDATE; + + if (panfrost_has_hw_feature(pfdev, HW_FEATURE_FLUSH_REDUCTION)) + cfg |= JS_CONFIG_ENABLE_FLUSH_REDUCTION; + + if (panfrost_has_hw_issue(pfdev, HW_ISSUE_10649)) + cfg |= JS_CONFIG_START_MMU; + + job_write(pfdev, JS_CONFIG_NEXT(js), cfg); + + if (panfrost_has_hw_feature(pfdev, HW_FEATURE_FLUSH_REDUCTION)) + job_write(pfdev, JS_FLUSH_ID_NEXT(js), job->flush_id); + + /* GO ! */ + dev_dbg(pfdev->dev, "JS: Submitting atom %p to js[%d] with head=0x%llx", + job, js, jc_head); + + job_write(pfdev, JS_COMMAND_NEXT(js), JS_COMMAND_START); + + spin_unlock_irqrestore(&pfdev->hwaccess_lock, flags); + +end: + pm_runtime_mark_last_busy(pfdev->dev); + pm_runtime_put_autosuspend(pfdev->dev); +} + +static void panfrost_acquire_object_fences(struct drm_gem_object **bos, + int bo_count, + struct dma_fence **implicit_fences) +{ + int i; + + for (i = 0; i < bo_count; i++) + implicit_fences[i] = reservation_object_get_excl_rcu(bos[i]->resv); +} + +static void panfrost_attach_object_fences(struct drm_gem_object **bos, + int bo_count, + struct dma_fence *fence) +{ + int i; + + for (i = 0; i < bo_count; i++) + reservation_object_add_excl_fence(bos[i]->resv, fence); +} + +int panfrost_job_push(struct panfrost_job *job) +{ + struct panfrost_device *pfdev = job->pfdev; + int slot = panfrost_job_get_slot(job); + struct drm_sched_entity *entity = &job->file_priv->sched_entity[slot]; + struct ww_acquire_ctx acquire_ctx; + int ret = 0; + + mutex_lock(&pfdev->sched_lock); + + ret = drm_gem_lock_reservations(job->bos, job->bo_count, + &acquire_ctx); + if (ret) { + mutex_unlock(&pfdev->sched_lock); + return ret; + } + + ret = drm_sched_job_init(&job->base, entity, NULL); + if (ret) { + mutex_unlock(&pfdev->sched_lock); + goto unlock; + } + + job->render_done_fence = dma_fence_get(&job->base.s_fence->finished); + + kref_get(&job->refcount); /* put by scheduler job completion */ + + panfrost_acquire_object_fences(job->bos, job->bo_count, + job->implicit_fences); + + drm_sched_entity_push_job(&job->base, entity); + + mutex_unlock(&pfdev->sched_lock); + + panfrost_attach_object_fences(job->bos, job->bo_count, + job->render_done_fence); + +unlock: + drm_gem_unlock_reservations(job->bos, job->bo_count, &acquire_ctx); + + return ret; +} + +static void panfrost_job_cleanup(struct kref *ref) +{ + struct panfrost_job *job = container_of(ref, struct panfrost_job, + refcount); + unsigned int i; + + if (job->in_fences) { + for (i = 0; i < job->in_fence_count; i++) + dma_fence_put(job->in_fences[i]); + kvfree(job->in_fences); + } + if (job->implicit_fences) { + for (i = 0; i < job->bo_count; i++) + dma_fence_put(job->implicit_fences[i]); + kvfree(job->implicit_fences); + } + dma_fence_put(job->done_fence); + dma_fence_put(job->render_done_fence); + + if (job->bos) { + for (i = 0; i < job->bo_count; i++) + drm_gem_object_put_unlocked(job->bos[i]); + kvfree(job->bos); + } + + kfree(job); +} + +void panfrost_job_put(struct panfrost_job *job) +{ + kref_put(&job->refcount, panfrost_job_cleanup); +} + +static void panfrost_job_free(struct drm_sched_job *sched_job) +{ + struct panfrost_job *job = to_panfrost_job(sched_job); + + drm_sched_job_cleanup(sched_job); + + panfrost_job_put(job); +} + +static struct dma_fence *panfrost_job_dependency(struct drm_sched_job *sched_job, + struct drm_sched_entity *s_entity) +{ + struct panfrost_job *job = to_panfrost_job(sched_job); + struct dma_fence *fence; + unsigned int i; + + /* Explicit fences */ + for (i = 0; i < job->in_fence_count; i++) { + if (job->in_fences[i]) { + fence = job->in_fences[i]; + job->in_fences[i] = NULL; + return fence; + } + } + + /* Implicit fences, max. one per BO */ + for (i = 0; i < job->bo_count; i++) { + if (job->implicit_fences[i]) { + fence = job->implicit_fences[i]; + job->implicit_fences[i] = NULL; + return fence; + } + } + + return NULL; +} + +static struct dma_fence *panfrost_job_run(struct drm_sched_job *sched_job) +{ + struct panfrost_job *job = to_panfrost_job(sched_job); + struct panfrost_device *pfdev = job->pfdev; + int slot = panfrost_job_get_slot(job); + struct dma_fence *fence = NULL; + + if (unlikely(job->base.s_fence->finished.error)) + return NULL; + + pfdev->jobs[slot] = job; + + fence = panfrost_fence_create(pfdev, slot); + if (IS_ERR(fence)) + return NULL; + + if (job->done_fence) + dma_fence_put(job->done_fence); + job->done_fence = dma_fence_get(fence); + + panfrost_job_hw_submit(job, slot); + + return fence; +} + +void panfrost_job_enable_interrupts(struct panfrost_device *pfdev) +{ + int j; + u32 irq_mask = 0; + + for (j = 0; j < NUM_JOB_SLOTS; j++) { + irq_mask |= MK_JS_MASK(j); + } + + job_write(pfdev, JOB_INT_CLEAR, irq_mask); + job_write(pfdev, JOB_INT_MASK, irq_mask); +} + +static void panfrost_job_timedout(struct drm_sched_job *sched_job) +{ + struct panfrost_job *job = to_panfrost_job(sched_job); + struct panfrost_device *pfdev = job->pfdev; + int js = panfrost_job_get_slot(job); + int i; + + /* + * If the GPU managed to complete this jobs fence, the timeout is + * spurious. Bail out. + */ + if (dma_fence_is_signaled(job->done_fence)) + return; + + dev_err(pfdev->dev, "gpu sched timeout, js=%d, status=0x%x, head=0x%x, tail=0x%x, sched_job=%p", + js, + job_read(pfdev, JS_STATUS(js)), + job_read(pfdev, JS_HEAD_LO(js)), + job_read(pfdev, JS_TAIL_LO(js)), + sched_job); + + for (i = 0; i < NUM_JOB_SLOTS; i++) + drm_sched_stop(&pfdev->js->queue[i].sched); + + if (sched_job) + drm_sched_increase_karma(sched_job); + + /* panfrost_core_dump(pfdev); */ + + panfrost_devfreq_record_transition(pfdev, js); + panfrost_gpu_soft_reset(pfdev); + + /* TODO: Re-enable all other address spaces */ + panfrost_mmu_enable(pfdev, 0); + panfrost_gpu_power_on(pfdev); + panfrost_job_enable_interrupts(pfdev); + + for (i = 0; i < NUM_JOB_SLOTS; i++) + drm_sched_resubmit_jobs(&pfdev->js->queue[i].sched); + + /* restart scheduler after GPU is usable again */ + for (i = 0; i < NUM_JOB_SLOTS; i++) + drm_sched_start(&pfdev->js->queue[i].sched, true); +} + +static const struct drm_sched_backend_ops panfrost_sched_ops = { + .dependency = panfrost_job_dependency, + .run_job = panfrost_job_run, + .timedout_job = panfrost_job_timedout, + .free_job = panfrost_job_free +}; + +static irqreturn_t panfrost_job_irq_handler(int irq, void *data) +{ + struct panfrost_device *pfdev = data; + u32 status = job_read(pfdev, JOB_INT_STAT); + int j; + + dev_dbg(pfdev->dev, "jobslot irq status=%x\n", status); + + if (!status) + return IRQ_NONE; + + pm_runtime_mark_last_busy(pfdev->dev); + + for (j = 0; status; j++) { + u32 mask = MK_JS_MASK(j); + + if (!(status & mask)) + continue; + + job_write(pfdev, JOB_INT_CLEAR, mask); + + if (status & JOB_INT_MASK_ERR(j)) { + job_write(pfdev, JS_COMMAND_NEXT(j), JS_COMMAND_NOP); + + dev_err(pfdev->dev, "js fault, js=%d, status=%s, head=0x%x, tail=0x%x", + j, + panfrost_exception_name(pfdev, job_read(pfdev, JS_STATUS(j))), + job_read(pfdev, JS_HEAD_LO(j)), + job_read(pfdev, JS_TAIL_LO(j))); + + drm_sched_fault(&pfdev->js->queue[j].sched); + } + + if (status & JOB_INT_MASK_DONE(j)) { + panfrost_devfreq_record_transition(pfdev, j); + dma_fence_signal(pfdev->jobs[j]->done_fence); + } + + status &= ~mask; + } + + return IRQ_HANDLED; +} + +int panfrost_job_init(struct panfrost_device *pfdev) +{ + struct panfrost_job_slot *js; + int ret, j, irq; + + pfdev->js = js = devm_kzalloc(pfdev->dev, sizeof(*js), GFP_KERNEL); + if (!js) + return -ENOMEM; + + spin_lock_init(&js->job_lock); + + irq = platform_get_irq_byname(to_platform_device(pfdev->dev), "job"); + if (irq <= 0) + return -ENODEV; + + ret = devm_request_irq(pfdev->dev, irq, panfrost_job_irq_handler, + IRQF_SHARED, "job", pfdev); + if (ret) { + dev_err(pfdev->dev, "failed to request job irq"); + return ret; + } + + for (j = 0; j < NUM_JOB_SLOTS; j++) { + js->queue[j].fence_context = dma_fence_context_alloc(1); + + ret = drm_sched_init(&js->queue[j].sched, + &panfrost_sched_ops, + 1, 0, msecs_to_jiffies(500), + "pan_js"); + if (ret) { + dev_err(pfdev->dev, "Failed to create scheduler: %d.", ret); + goto err_sched; + } + } + + panfrost_job_enable_interrupts(pfdev); + + return 0; + +err_sched: + for (j--; j >= 0; j--) + drm_sched_fini(&js->queue[j].sched); + + return ret; +} + +void panfrost_job_fini(struct panfrost_device *pfdev) +{ + struct panfrost_job_slot *js = pfdev->js; + int j; + + job_write(pfdev, JOB_INT_MASK, 0); + + for (j = 0; j < NUM_JOB_SLOTS; j++) + drm_sched_fini(&js->queue[j].sched); + +} + +int panfrost_job_open(struct panfrost_file_priv *panfrost_priv) +{ + struct panfrost_device *pfdev = panfrost_priv->pfdev; + struct panfrost_job_slot *js = pfdev->js; + struct drm_sched_rq *rq; + int ret, i; + + for (i = 0; i < NUM_JOB_SLOTS; i++) { + rq = &js->queue[i].sched.sched_rq[DRM_SCHED_PRIORITY_NORMAL]; + ret = drm_sched_entity_init(&panfrost_priv->sched_entity[i], &rq, 1, NULL); + if (WARN_ON(ret)) + return ret; + } + return 0; +} + +void panfrost_job_close(struct panfrost_file_priv *panfrost_priv) +{ + int i; + + for (i = 0; i < NUM_JOB_SLOTS; i++) + drm_sched_entity_destroy(&panfrost_priv->sched_entity[i]); +} + +int panfrost_job_is_idle(struct panfrost_device *pfdev) +{ + struct panfrost_job_slot *js = pfdev->js; + int i; + + for (i = 0; i < NUM_JOB_SLOTS; i++) { + /* If there are any jobs in the HW queue, we're not idle */ + if (atomic_read(&js->queue[i].sched.hw_rq_count)) + return false; + + /* Check whether the hardware is idle */ + if (pfdev->devfreq.slot[i].busy) + return false; + } + + return true; +} diff --git a/drivers/gpu/drm/panfrost/panfrost_job.h b/drivers/gpu/drm/panfrost/panfrost_job.h new file mode 100644 index 000000000000..62454128a792 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_job.h @@ -0,0 +1,51 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright 2019 Collabora ltd. */ + +#ifndef __PANFROST_JOB_H__ +#define __PANFROST_JOB_H__ + +#include <uapi/drm/panfrost_drm.h> +#include <drm/gpu_scheduler.h> + +struct panfrost_device; +struct panfrost_gem_object; +struct panfrost_file_priv; + +struct panfrost_job { + struct drm_sched_job base; + + struct kref refcount; + + struct panfrost_device *pfdev; + struct panfrost_file_priv *file_priv; + + /* Optional fences userspace can pass in for the job to depend on. */ + struct dma_fence **in_fences; + u32 in_fence_count; + + /* Fence to be signaled by IRQ handler when the job is complete. */ + struct dma_fence *done_fence; + + __u64 jc; + __u32 requirements; + __u32 flush_id; + + /* Exclusive fences we have taken from the BOs to wait for */ + struct dma_fence **implicit_fences; + struct drm_gem_object **bos; + u32 bo_count; + + /* Fence to be signaled by drm-sched once its done with the job */ + struct dma_fence *render_done_fence; +}; + +int panfrost_job_init(struct panfrost_device *pfdev); +void panfrost_job_fini(struct panfrost_device *pfdev); +int panfrost_job_open(struct panfrost_file_priv *panfrost_priv); +void panfrost_job_close(struct panfrost_file_priv *panfrost_priv); +int panfrost_job_push(struct panfrost_job *job); +void panfrost_job_put(struct panfrost_job *job); +void panfrost_job_enable_interrupts(struct panfrost_device *pfdev); +int panfrost_job_is_idle(struct panfrost_device *pfdev); + +#endif diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c new file mode 100644 index 000000000000..762b1bd2a8c2 --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c @@ -0,0 +1,386 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */ +#include <linux/bitfield.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/io.h> +#include <linux/iopoll.h> +#include <linux/io-pgtable.h> +#include <linux/iommu.h> +#include <linux/platform_device.h> +#include <linux/pm_runtime.h> +#include <linux/sizes.h> + +#include "panfrost_device.h" +#include "panfrost_mmu.h" +#include "panfrost_gem.h" +#include "panfrost_features.h" +#include "panfrost_regs.h" + +#define mmu_write(dev, reg, data) writel(data, dev->iomem + reg) +#define mmu_read(dev, reg) readl(dev->iomem + reg) + +struct panfrost_mmu { + struct io_pgtable_cfg pgtbl_cfg; + struct io_pgtable_ops *pgtbl_ops; + struct mutex lock; +}; + +static int wait_ready(struct panfrost_device *pfdev, u32 as_nr) +{ + int ret; + u32 val; + + /* Wait for the MMU status to indicate there is no active command, in + * case one is pending. */ + ret = readl_relaxed_poll_timeout_atomic(pfdev->iomem + AS_STATUS(as_nr), + val, !(val & AS_STATUS_AS_ACTIVE), 10, 1000); + + if (ret) + dev_err(pfdev->dev, "AS_ACTIVE bit stuck\n"); + + return ret; +} + +static int write_cmd(struct panfrost_device *pfdev, u32 as_nr, u32 cmd) +{ + int status; + + /* write AS_COMMAND when MMU is ready to accept another command */ + status = wait_ready(pfdev, as_nr); + if (!status) + mmu_write(pfdev, AS_COMMAND(as_nr), cmd); + + return status; +} + +static void lock_region(struct panfrost_device *pfdev, u32 as_nr, + u64 iova, size_t size) +{ + u8 region_width; + u64 region = iova & PAGE_MASK; + /* + * fls returns: + * 1 .. 32 + * + * 10 + fls(num_pages) + * results in the range (11 .. 42) + */ + + size = round_up(size, PAGE_SIZE); + + region_width = 10 + fls(size >> PAGE_SHIFT); + if ((size >> PAGE_SHIFT) != (1ul << (region_width - 11))) { + /* not pow2, so must go up to the next pow2 */ + region_width += 1; + } + region |= region_width; + + /* Lock the region that needs to be updated */ + mmu_write(pfdev, AS_LOCKADDR_LO(as_nr), region & 0xFFFFFFFFUL); + mmu_write(pfdev, AS_LOCKADDR_HI(as_nr), (region >> 32) & 0xFFFFFFFFUL); + write_cmd(pfdev, as_nr, AS_COMMAND_LOCK); +} + + +static int mmu_hw_do_operation(struct panfrost_device *pfdev, u32 as_nr, + u64 iova, size_t size, u32 op) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&pfdev->hwaccess_lock, flags); + + if (op != AS_COMMAND_UNLOCK) + lock_region(pfdev, as_nr, iova, size); + + /* Run the MMU operation */ + write_cmd(pfdev, as_nr, op); + + /* Wait for the flush to complete */ + ret = wait_ready(pfdev, as_nr); + + spin_unlock_irqrestore(&pfdev->hwaccess_lock, flags); + + return ret; +} + +void panfrost_mmu_enable(struct panfrost_device *pfdev, u32 as_nr) +{ + struct io_pgtable_cfg *cfg = &pfdev->mmu->pgtbl_cfg; + u64 transtab = cfg->arm_mali_lpae_cfg.transtab; + u64 memattr = cfg->arm_mali_lpae_cfg.memattr; + + mmu_write(pfdev, MMU_INT_CLEAR, ~0); + mmu_write(pfdev, MMU_INT_MASK, ~0); + + mmu_write(pfdev, AS_TRANSTAB_LO(as_nr), transtab & 0xffffffffUL); + mmu_write(pfdev, AS_TRANSTAB_HI(as_nr), transtab >> 32); + + /* Need to revisit mem attrs. + * NC is the default, Mali driver is inner WT. + */ + mmu_write(pfdev, AS_MEMATTR_LO(as_nr), memattr & 0xffffffffUL); + mmu_write(pfdev, AS_MEMATTR_HI(as_nr), memattr >> 32); + + write_cmd(pfdev, as_nr, AS_COMMAND_UPDATE); +} + +static void mmu_disable(struct panfrost_device *pfdev, u32 as_nr) +{ + mmu_write(pfdev, AS_TRANSTAB_LO(as_nr), 0); + mmu_write(pfdev, AS_TRANSTAB_HI(as_nr), 0); + + mmu_write(pfdev, AS_MEMATTR_LO(as_nr), 0); + mmu_write(pfdev, AS_MEMATTR_HI(as_nr), 0); + + write_cmd(pfdev, as_nr, AS_COMMAND_UPDATE); +} + +static size_t get_pgsize(u64 addr, size_t size) +{ + if (addr & (SZ_2M - 1) || size < SZ_2M) + return SZ_4K; + + return SZ_2M; +} + +int panfrost_mmu_map(struct panfrost_gem_object *bo) +{ + struct drm_gem_object *obj = &bo->base.base; + struct panfrost_device *pfdev = to_panfrost_device(obj->dev); + struct io_pgtable_ops *ops = pfdev->mmu->pgtbl_ops; + u64 iova = bo->node.start << PAGE_SHIFT; + unsigned int count; + struct scatterlist *sgl; + struct sg_table *sgt; + int ret; + + sgt = drm_gem_shmem_get_pages_sgt(obj); + if (WARN_ON(IS_ERR(sgt))) + return PTR_ERR(sgt); + + ret = pm_runtime_get_sync(pfdev->dev); + if (ret < 0) + return ret; + + mutex_lock(&pfdev->mmu->lock); + + for_each_sg(sgt->sgl, sgl, sgt->nents, count) { + unsigned long paddr = sg_dma_address(sgl); + size_t len = sg_dma_len(sgl); + + dev_dbg(pfdev->dev, "map: iova=%llx, paddr=%lx, len=%zx", iova, paddr, len); + + while (len) { + size_t pgsize = get_pgsize(iova | paddr, len); + + ops->map(ops, iova, paddr, pgsize, IOMMU_WRITE | IOMMU_READ); + iova += pgsize; + paddr += pgsize; + len -= pgsize; + } + } + + mmu_hw_do_operation(pfdev, 0, bo->node.start << PAGE_SHIFT, + bo->node.size << PAGE_SHIFT, AS_COMMAND_FLUSH_PT); + + mutex_unlock(&pfdev->mmu->lock); + + pm_runtime_mark_last_busy(pfdev->dev); + pm_runtime_put_autosuspend(pfdev->dev); + + return 0; +} + +void panfrost_mmu_unmap(struct panfrost_gem_object *bo) +{ + struct drm_gem_object *obj = &bo->base.base; + struct panfrost_device *pfdev = to_panfrost_device(obj->dev); + struct io_pgtable_ops *ops = pfdev->mmu->pgtbl_ops; + u64 iova = bo->node.start << PAGE_SHIFT; + size_t len = bo->node.size << PAGE_SHIFT; + size_t unmapped_len = 0; + int ret; + + dev_dbg(pfdev->dev, "unmap: iova=%llx, len=%zx", iova, len); + + ret = pm_runtime_get_sync(pfdev->dev); + if (ret < 0) + return; + + mutex_lock(&pfdev->mmu->lock); + + while (unmapped_len < len) { + size_t unmapped_page; + size_t pgsize = get_pgsize(iova, len - unmapped_len); + + unmapped_page = ops->unmap(ops, iova, pgsize); + if (!unmapped_page) + break; + + iova += unmapped_page; + unmapped_len += unmapped_page; + } + + mmu_hw_do_operation(pfdev, 0, bo->node.start << PAGE_SHIFT, + bo->node.size << PAGE_SHIFT, AS_COMMAND_FLUSH_PT); + + mutex_unlock(&pfdev->mmu->lock); + + pm_runtime_mark_last_busy(pfdev->dev); + pm_runtime_put_autosuspend(pfdev->dev); +} + +static void mmu_tlb_inv_context_s1(void *cookie) +{ + struct panfrost_device *pfdev = cookie; + + mmu_hw_do_operation(pfdev, 0, 0, ~0UL, AS_COMMAND_FLUSH_MEM); +} + +static void mmu_tlb_inv_range_nosync(unsigned long iova, size_t size, + size_t granule, bool leaf, void *cookie) +{} + +static void mmu_tlb_sync_context(void *cookie) +{ + //struct panfrost_device *pfdev = cookie; + // TODO: Wait 1000 GPU cycles for HW_ISSUE_6367/T60X +} + +static const struct iommu_gather_ops mmu_tlb_ops = { + .tlb_flush_all = mmu_tlb_inv_context_s1, + .tlb_add_flush = mmu_tlb_inv_range_nosync, + .tlb_sync = mmu_tlb_sync_context, +}; + +static const char *access_type_name(struct panfrost_device *pfdev, + u32 fault_status) +{ + switch (fault_status & AS_FAULTSTATUS_ACCESS_TYPE_MASK) { + case AS_FAULTSTATUS_ACCESS_TYPE_ATOMIC: + if (panfrost_has_hw_feature(pfdev, HW_FEATURE_AARCH64_MMU)) + return "ATOMIC"; + else + return "UNKNOWN"; + case AS_FAULTSTATUS_ACCESS_TYPE_READ: + return "READ"; + case AS_FAULTSTATUS_ACCESS_TYPE_WRITE: + return "WRITE"; + case AS_FAULTSTATUS_ACCESS_TYPE_EX: + return "EXECUTE"; + default: + WARN_ON(1); + return NULL; + } +} + +static irqreturn_t panfrost_mmu_irq_handler(int irq, void *data) +{ + struct panfrost_device *pfdev = data; + u32 status = mmu_read(pfdev, MMU_INT_STAT); + int i; + + if (!status) + return IRQ_NONE; + + dev_err(pfdev->dev, "mmu irq status=%x\n", status); + + for (i = 0; status; i++) { + u32 mask = BIT(i) | BIT(i + 16); + u64 addr; + u32 fault_status; + u32 exception_type; + u32 access_type; + u32 source_id; + + if (!(status & mask)) + continue; + + fault_status = mmu_read(pfdev, AS_FAULTSTATUS(i)); + addr = mmu_read(pfdev, AS_FAULTADDRESS_LO(i)); + addr |= (u64)mmu_read(pfdev, AS_FAULTADDRESS_HI(i)) << 32; + + /* decode the fault status */ + exception_type = fault_status & 0xFF; + access_type = (fault_status >> 8) & 0x3; + source_id = (fault_status >> 16); + + /* terminal fault, print info about the fault */ + dev_err(pfdev->dev, + "Unhandled Page fault in AS%d at VA 0x%016llX\n" + "Reason: %s\n" + "raw fault status: 0x%X\n" + "decoded fault status: %s\n" + "exception type 0x%X: %s\n" + "access type 0x%X: %s\n" + "source id 0x%X\n", + i, addr, + "TODO", + fault_status, + (fault_status & (1 << 10) ? "DECODER FAULT" : "SLAVE FAULT"), + exception_type, panfrost_exception_name(pfdev, exception_type), + access_type, access_type_name(pfdev, fault_status), + source_id); + + mmu_write(pfdev, MMU_INT_CLEAR, mask); + + status &= ~mask; + } + + return IRQ_HANDLED; +}; + +int panfrost_mmu_init(struct panfrost_device *pfdev) +{ + struct io_pgtable_ops *pgtbl_ops; + int err, irq; + + pfdev->mmu = devm_kzalloc(pfdev->dev, sizeof(*pfdev->mmu), GFP_KERNEL); + if (!pfdev->mmu) + return -ENOMEM; + + mutex_init(&pfdev->mmu->lock); + + irq = platform_get_irq_byname(to_platform_device(pfdev->dev), "mmu"); + if (irq <= 0) + return -ENODEV; + + err = devm_request_irq(pfdev->dev, irq, panfrost_mmu_irq_handler, + IRQF_SHARED, "mmu", pfdev); + + if (err) { + dev_err(pfdev->dev, "failed to request mmu irq"); + return err; + } + mmu_write(pfdev, MMU_INT_CLEAR, ~0); + mmu_write(pfdev, MMU_INT_MASK, ~0); + + pfdev->mmu->pgtbl_cfg = (struct io_pgtable_cfg) { + .pgsize_bitmap = SZ_4K | SZ_2M, + .ias = FIELD_GET(0xff, pfdev->features.mmu_features), + .oas = FIELD_GET(0xff00, pfdev->features.mmu_features), + .tlb = &mmu_tlb_ops, + .iommu_dev = pfdev->dev, + }; + + pgtbl_ops = alloc_io_pgtable_ops(ARM_MALI_LPAE, &pfdev->mmu->pgtbl_cfg, + pfdev); + if (!pgtbl_ops) + return -ENOMEM; + + pfdev->mmu->pgtbl_ops = pgtbl_ops; + + panfrost_mmu_enable(pfdev, 0); + + return 0; +} + +void panfrost_mmu_fini(struct panfrost_device *pfdev) +{ + mmu_write(pfdev, MMU_INT_MASK, 0); + mmu_disable(pfdev, 0); + + free_io_pgtable_ops(pfdev->mmu->pgtbl_ops); +} diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.h b/drivers/gpu/drm/panfrost/panfrost_mmu.h new file mode 100644 index 000000000000..f5878d86a5ce --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */ + +#ifndef __PANFROST_MMU_H__ +#define __PANFROST_MMU_H__ + +struct panfrost_gem_object; + +int panfrost_mmu_map(struct panfrost_gem_object *bo); +void panfrost_mmu_unmap(struct panfrost_gem_object *bo); + +int panfrost_mmu_init(struct panfrost_device *pfdev); +void panfrost_mmu_fini(struct panfrost_device *pfdev); + +void panfrost_mmu_enable(struct panfrost_device *pfdev, u32 as_nr); + +#endif diff --git a/drivers/gpu/drm/panfrost/panfrost_regs.h b/drivers/gpu/drm/panfrost/panfrost_regs.h new file mode 100644 index 000000000000..578c5fc2188b --- /dev/null +++ b/drivers/gpu/drm/panfrost/panfrost_regs.h @@ -0,0 +1,298 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright 2018 Marty E. Plummer <hanetzer@startmail.com> */ +/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */ +/* + * Register definitions based on mali_midg_regmap.h + * (C) COPYRIGHT 2010-2018 ARM Limited. All rights reserved. + */ +#ifndef __PANFROST_REGS_H__ +#define __PANFROST_REGS_H__ + +#define GPU_ID 0x00 +#define GPU_L2_FEATURES 0x004 /* (RO) Level 2 cache features */ +#define GPU_CORE_FEATURES 0x008 /* (RO) Shader Core Features */ +#define GPU_TILER_FEATURES 0x00C /* (RO) Tiler Features */ +#define GPU_MEM_FEATURES 0x010 /* (RO) Memory system features */ +#define GROUPS_L2_COHERENT BIT(0) /* Cores groups are l2 coherent */ + +#define GPU_MMU_FEATURES 0x014 /* (RO) MMU features */ +#define GPU_AS_PRESENT 0x018 /* (RO) Address space slots present */ +#define GPU_JS_PRESENT 0x01C /* (RO) Job slots present */ + +#define GPU_INT_RAWSTAT 0x20 +#define GPU_INT_CLEAR 0x24 +#define GPU_INT_MASK 0x28 +#define GPU_INT_STAT 0x2c +#define GPU_IRQ_FAULT BIT(0) +#define GPU_IRQ_MULTIPLE_FAULT BIT(7) +#define GPU_IRQ_RESET_COMPLETED BIT(8) +#define GPU_IRQ_POWER_CHANGED BIT(9) +#define GPU_IRQ_POWER_CHANGED_ALL BIT(10) +#define GPU_IRQ_PERFCNT_SAMPLE_COMPLETED BIT(16) +#define GPU_IRQ_CLEAN_CACHES_COMPLETED BIT(17) +#define GPU_IRQ_MASK_ALL \ + (GPU_IRQ_FAULT |\ + GPU_IRQ_MULTIPLE_FAULT |\ + GPU_IRQ_RESET_COMPLETED |\ + GPU_IRQ_POWER_CHANGED |\ + GPU_IRQ_POWER_CHANGED_ALL |\ + GPU_IRQ_PERFCNT_SAMPLE_COMPLETED |\ + GPU_IRQ_CLEAN_CACHES_COMPLETED) +#define GPU_IRQ_MASK_ERROR \ + ( \ + GPU_IRQ_FAULT |\ + GPU_IRQ_MULTIPLE_FAULT) +#define GPU_CMD 0x30 +#define GPU_CMD_SOFT_RESET 0x01 +#define GPU_STATUS 0x34 +#define GPU_LATEST_FLUSH_ID 0x38 +#define GPU_FAULT_STATUS 0x3C +#define GPU_FAULT_ADDRESS_LO 0x40 +#define GPU_FAULT_ADDRESS_HI 0x44 + +#define GPU_THREAD_MAX_THREADS 0x0A0 /* (RO) Maximum number of threads per core */ +#define GPU_THREAD_MAX_WORKGROUP_SIZE 0x0A4 /* (RO) Maximum workgroup size */ +#define GPU_THREAD_MAX_BARRIER_SIZE 0x0A8 /* (RO) Maximum threads waiting at a barrier */ +#define GPU_THREAD_FEATURES 0x0AC /* (RO) Thread features */ +#define GPU_THREAD_TLS_ALLOC 0x310 /* (RO) Number of threads per core that + * TLS must be allocated for */ + +#define GPU_TEXTURE_FEATURES(n) (0x0B0 + ((n) * 4)) +#define GPU_JS_FEATURES(n) (0x0C0 + ((n) * 4)) + +#define GPU_SHADER_PRESENT_LO 0x100 /* (RO) Shader core present bitmap, low word */ +#define GPU_SHADER_PRESENT_HI 0x104 /* (RO) Shader core present bitmap, high word */ +#define GPU_TILER_PRESENT_LO 0x110 /* (RO) Tiler core present bitmap, low word */ +#define GPU_TILER_PRESENT_HI 0x114 /* (RO) Tiler core present bitmap, high word */ + +#define GPU_L2_PRESENT_LO 0x120 /* (RO) Level 2 cache present bitmap, low word */ +#define GPU_L2_PRESENT_HI 0x124 /* (RO) Level 2 cache present bitmap, high word */ + +#define GPU_COHERENCY_FEATURES 0x300 /* (RO) Coherency features present */ +#define COHERENCY_ACE_LITE BIT(0) +#define COHERENCY_ACE BIT(1) + +#define GPU_STACK_PRESENT_LO 0xE00 /* (RO) Core stack present bitmap, low word */ +#define GPU_STACK_PRESENT_HI 0xE04 /* (RO) Core stack present bitmap, high word */ + +#define SHADER_READY_LO 0x140 /* (RO) Shader core ready bitmap, low word */ +#define SHADER_READY_HI 0x144 /* (RO) Shader core ready bitmap, high word */ + +#define TILER_READY_LO 0x150 /* (RO) Tiler core ready bitmap, low word */ +#define TILER_READY_HI 0x154 /* (RO) Tiler core ready bitmap, high word */ + +#define L2_READY_LO 0x160 /* (RO) Level 2 cache ready bitmap, low word */ +#define L2_READY_HI 0x164 /* (RO) Level 2 cache ready bitmap, high word */ + +#define STACK_READY_LO 0xE10 /* (RO) Core stack ready bitmap, low word */ +#define STACK_READY_HI 0xE14 /* (RO) Core stack ready bitmap, high word */ + + +#define SHADER_PWRON_LO 0x180 /* (WO) Shader core power on bitmap, low word */ +#define SHADER_PWRON_HI 0x184 /* (WO) Shader core power on bitmap, high word */ + +#define TILER_PWRON_LO 0x190 /* (WO) Tiler core power on bitmap, low word */ +#define TILER_PWRON_HI 0x194 /* (WO) Tiler core power on bitmap, high word */ + +#define L2_PWRON_LO 0x1A0 /* (WO) Level 2 cache power on bitmap, low word */ +#define L2_PWRON_HI 0x1A4 /* (WO) Level 2 cache power on bitmap, high word */ + +#define STACK_PWRON_LO 0xE20 /* (RO) Core stack power on bitmap, low word */ +#define STACK_PWRON_HI 0xE24 /* (RO) Core stack power on bitmap, high word */ + + +#define SHADER_PWROFF_LO 0x1C0 /* (WO) Shader core power off bitmap, low word */ +#define SHADER_PWROFF_HI 0x1C4 /* (WO) Shader core power off bitmap, high word */ + +#define TILER_PWROFF_LO 0x1D0 /* (WO) Tiler core power off bitmap, low word */ +#define TILER_PWROFF_HI 0x1D4 /* (WO) Tiler core power off bitmap, high word */ + +#define L2_PWROFF_LO 0x1E0 /* (WO) Level 2 cache power off bitmap, low word */ +#define L2_PWROFF_HI 0x1E4 /* (WO) Level 2 cache power off bitmap, high word */ + +#define STACK_PWROFF_LO 0xE30 /* (RO) Core stack power off bitmap, low word */ +#define STACK_PWROFF_HI 0xE34 /* (RO) Core stack power off bitmap, high word */ + + +#define SHADER_PWRTRANS_LO 0x200 /* (RO) Shader core power transition bitmap, low word */ +#define SHADER_PWRTRANS_HI 0x204 /* (RO) Shader core power transition bitmap, high word */ + +#define TILER_PWRTRANS_LO 0x210 /* (RO) Tiler core power transition bitmap, low word */ +#define TILER_PWRTRANS_HI 0x214 /* (RO) Tiler core power transition bitmap, high word */ + +#define L2_PWRTRANS_LO 0x220 /* (RO) Level 2 cache power transition bitmap, low word */ +#define L2_PWRTRANS_HI 0x224 /* (RO) Level 2 cache power transition bitmap, high word */ + +#define STACK_PWRTRANS_LO 0xE40 /* (RO) Core stack power transition bitmap, low word */ +#define STACK_PWRTRANS_HI 0xE44 /* (RO) Core stack power transition bitmap, high word */ + + +#define SHADER_PWRACTIVE_LO 0x240 /* (RO) Shader core active bitmap, low word */ +#define SHADER_PWRACTIVE_HI 0x244 /* (RO) Shader core active bitmap, high word */ + +#define TILER_PWRACTIVE_LO 0x250 /* (RO) Tiler core active bitmap, low word */ +#define TILER_PWRACTIVE_HI 0x254 /* (RO) Tiler core active bitmap, high word */ + +#define L2_PWRACTIVE_LO 0x260 /* (RO) Level 2 cache active bitmap, low word */ +#define L2_PWRACTIVE_HI 0x264 /* (RO) Level 2 cache active bitmap, high word */ + +#define GPU_JM_CONFIG 0xF00 /* (RW) Job Manager configuration register (Implementation specific register) */ +#define GPU_SHADER_CONFIG 0xF04 /* (RW) Shader core configuration settings (Implementation specific register) */ +#define GPU_TILER_CONFIG 0xF08 /* (RW) Tiler core configuration settings (Implementation specific register) */ +#define GPU_L2_MMU_CONFIG 0xF0C /* (RW) Configuration of the L2 cache and MMU (Implementation specific register) */ + +/* L2_MMU_CONFIG register */ +#define L2_MMU_CONFIG_ALLOW_SNOOP_DISPARITY_SHIFT 23 +#define L2_MMU_CONFIG_ALLOW_SNOOP_DISPARITY (0x1 << L2_MMU_CONFIG_ALLOW_SNOOP_DISPARITY_SHIFT) +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT 24 +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS (0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_OCTANT (0x1 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_QUARTER (0x2 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_HALF (0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) + +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT 26 +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES (0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_OCTANT (0x1 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_QUARTER (0x2 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) +#define L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_HALF (0x3 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) + +#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS_SHIFT 12 +#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS (0x7 << L2_MMU_CONFIG_LIMIT_EXTERNAL_READS_SHIFT) + +#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES_SHIFT 15 +#define L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES (0x7 << L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES_SHIFT) + +/* SHADER_CONFIG register */ +#define SC_ALT_COUNTERS BIT(3) +#define SC_OVERRIDE_FWD_PIXEL_KILL BIT(4) +#define SC_SDC_DISABLE_OQ_DISCARD BIT(6) +#define SC_LS_ALLOW_ATTR_TYPES BIT(16) +#define SC_LS_PAUSEBUFFER_DISABLE BIT(16) +#define SC_TLS_HASH_ENABLE BIT(17) +#define SC_LS_ATTR_CHECK_DISABLE BIT(18) +#define SC_ENABLE_TEXGRD_FLAGS BIT(25) +/* End SHADER_CONFIG register */ + +/* TILER_CONFIG register */ +#define TC_CLOCK_GATE_OVERRIDE BIT(0) + +/* JM_CONFIG register */ +#define JM_TIMESTAMP_OVERRIDE BIT(0) +#define JM_CLOCK_GATE_OVERRIDE BIT(1) +#define JM_JOB_THROTTLE_ENABLE BIT(2) +#define JM_JOB_THROTTLE_LIMIT_SHIFT 3 +#define JM_MAX_JOB_THROTTLE_LIMIT 0x3F +#define JM_FORCE_COHERENCY_FEATURES_SHIFT 2 +#define JM_IDVS_GROUP_SIZE_SHIFT 16 +#define JM_MAX_IDVS_GROUP_SIZE 0x3F + + +/* Job Control regs */ +#define JOB_INT_RAWSTAT 0x1000 +#define JOB_INT_CLEAR 0x1004 +#define JOB_INT_MASK 0x1008 +#define JOB_INT_STAT 0x100c +#define JOB_INT_JS_STATE 0x1010 +#define JOB_INT_THROTTLE 0x1014 + +#define MK_JS_MASK(j) (0x10001 << (j)) +#define JOB_INT_MASK_ERR(j) BIT((j) + 16) +#define JOB_INT_MASK_DONE(j) BIT(j) + +#define JS_BASE 0x1800 +#define JS_HEAD_LO(n) (JS_BASE + ((n) * 0x80) + 0x00) +#define JS_HEAD_HI(n) (JS_BASE + ((n) * 0x80) + 0x04) +#define JS_TAIL_LO(n) (JS_BASE + ((n) * 0x80) + 0x08) +#define JS_TAIL_HI(n) (JS_BASE + ((n) * 0x80) + 0x0c) +#define JS_AFFINITY_LO(n) (JS_BASE + ((n) * 0x80) + 0x10) +#define JS_AFFINITY_HI(n) (JS_BASE + ((n) * 0x80) + 0x14) +#define JS_CONFIG(n) (JS_BASE + ((n) * 0x80) + 0x18) +#define JS_XAFFINITY(n) (JS_BASE + ((n) * 0x80) + 0x1c) +#define JS_COMMAND(n) (JS_BASE + ((n) * 0x80) + 0x20) +#define JS_STATUS(n) (JS_BASE + ((n) * 0x80) + 0x24) +#define JS_HEAD_NEXT_LO(n) (JS_BASE + ((n) * 0x80) + 0x40) +#define JS_HEAD_NEXT_HI(n) (JS_BASE + ((n) * 0x80) + 0x44) +#define JS_AFFINITY_NEXT_LO(n) (JS_BASE + ((n) * 0x80) + 0x50) +#define JS_AFFINITY_NEXT_HI(n) (JS_BASE + ((n) * 0x80) + 0x54) +#define JS_CONFIG_NEXT(n) (JS_BASE + ((n) * 0x80) + 0x58) +#define JS_COMMAND_NEXT(n) (JS_BASE + ((n) * 0x80) + 0x60) +#define JS_FLUSH_ID_NEXT(n) (JS_BASE + ((n) * 0x80) + 0x70) + +/* Possible values of JS_CONFIG and JS_CONFIG_NEXT registers */ +#define JS_CONFIG_START_FLUSH_CLEAN BIT(8) +#define JS_CONFIG_START_FLUSH_CLEAN_INVALIDATE (3u << 8) +#define JS_CONFIG_START_MMU BIT(10) +#define JS_CONFIG_JOB_CHAIN_FLAG BIT(11) +#define JS_CONFIG_END_FLUSH_CLEAN BIT(12) +#define JS_CONFIG_END_FLUSH_CLEAN_INVALIDATE (3u << 12) +#define JS_CONFIG_ENABLE_FLUSH_REDUCTION BIT(14) +#define JS_CONFIG_DISABLE_DESCRIPTOR_WR_BK BIT(15) +#define JS_CONFIG_THREAD_PRI(n) ((n) << 16) + +#define JS_COMMAND_NOP 0x00 +#define JS_COMMAND_START 0x01 +#define JS_COMMAND_SOFT_STOP 0x02 /* Gently stop processing a job chain */ +#define JS_COMMAND_HARD_STOP 0x03 /* Rudely stop processing a job chain */ +#define JS_COMMAND_SOFT_STOP_0 0x04 /* Execute SOFT_STOP if JOB_CHAIN_FLAG is 0 */ +#define JS_COMMAND_HARD_STOP_0 0x05 /* Execute HARD_STOP if JOB_CHAIN_FLAG is 0 */ +#define JS_COMMAND_SOFT_STOP_1 0x06 /* Execute SOFT_STOP if JOB_CHAIN_FLAG is 1 */ +#define JS_COMMAND_HARD_STOP_1 0x07 /* Execute HARD_STOP if JOB_CHAIN_FLAG is 1 */ + +#define JS_STATUS_EVENT_ACTIVE 0x08 + + +/* MMU regs */ +#define MMU_INT_RAWSTAT 0x2000 +#define MMU_INT_CLEAR 0x2004 +#define MMU_INT_MASK 0x2008 +#define MMU_INT_STAT 0x200c + +/* AS_COMMAND register commands */ +#define AS_COMMAND_NOP 0x00 /* NOP Operation */ +#define AS_COMMAND_UPDATE 0x01 /* Broadcasts the values in AS_TRANSTAB and ASn_MEMATTR to all MMUs */ +#define AS_COMMAND_LOCK 0x02 /* Issue a lock region command to all MMUs */ +#define AS_COMMAND_UNLOCK 0x03 /* Issue a flush region command to all MMUs */ +#define AS_COMMAND_FLUSH 0x04 /* Flush all L2 caches then issue a flush region command to all MMUs + (deprecated - only for use with T60x) */ +#define AS_COMMAND_FLUSH_PT 0x04 /* Flush all L2 caches then issue a flush region command to all MMUs */ +#define AS_COMMAND_FLUSH_MEM 0x05 /* Wait for memory accesses to complete, flush all the L1s cache then + flush all L2 caches then issue a flush region command to all MMUs */ + +#define MMU_AS(as) (0x2400 + ((as) << 6)) + +#define AS_TRANSTAB_LO(as) (MMU_AS(as) + 0x00) /* (RW) Translation Table Base Address for address space n, low word */ +#define AS_TRANSTAB_HI(as) (MMU_AS(as) + 0x04) /* (RW) Translation Table Base Address for address space n, high word */ +#define AS_MEMATTR_LO(as) (MMU_AS(as) + 0x08) /* (RW) Memory attributes for address space n, low word. */ +#define AS_MEMATTR_HI(as) (MMU_AS(as) + 0x0C) /* (RW) Memory attributes for address space n, high word. */ +#define AS_LOCKADDR_LO(as) (MMU_AS(as) + 0x10) /* (RW) Lock region address for address space n, low word */ +#define AS_LOCKADDR_HI(as) (MMU_AS(as) + 0x14) /* (RW) Lock region address for address space n, high word */ +#define AS_COMMAND(as) (MMU_AS(as) + 0x18) /* (WO) MMU command register for address space n */ +#define AS_FAULTSTATUS(as) (MMU_AS(as) + 0x1C) /* (RO) MMU fault status register for address space n */ +#define AS_FAULTADDRESS_LO(as) (MMU_AS(as) + 0x20) /* (RO) Fault Address for address space n, low word */ +#define AS_FAULTADDRESS_HI(as) (MMU_AS(as) + 0x24) /* (RO) Fault Address for address space n, high word */ +#define AS_STATUS(as) (MMU_AS(as) + 0x28) /* (RO) Status flags for address space n */ +/* Additional Bifrost AS regsiters */ +#define AS_TRANSCFG_LO(as) (MMU_AS(as) + 0x30) /* (RW) Translation table configuration for address space n, low word */ +#define AS_TRANSCFG_HI(as) (MMU_AS(as) + 0x34) /* (RW) Translation table configuration for address space n, high word */ +#define AS_FAULTEXTRA_LO(as) (MMU_AS(as) + 0x38) /* (RO) Secondary fault address for address space n, low word */ +#define AS_FAULTEXTRA_HI(as) (MMU_AS(as) + 0x3C) /* (RO) Secondary fault address for address space n, high word */ + +/* + * Begin LPAE MMU TRANSTAB register values + */ +#define AS_TRANSTAB_LPAE_ADDR_SPACE_MASK 0xfffffffffffff000 +#define AS_TRANSTAB_LPAE_ADRMODE_IDENTITY 0x2 +#define AS_TRANSTAB_LPAE_ADRMODE_TABLE 0x3 +#define AS_TRANSTAB_LPAE_ADRMODE_MASK 0x3 +#define AS_TRANSTAB_LPAE_READ_INNER BIT(2) +#define AS_TRANSTAB_LPAE_SHARE_OUTER BIT(4) + +#define AS_STATUS_AS_ACTIVE 0x01 + +#define AS_FAULTSTATUS_ACCESS_TYPE_MASK (0x3 << 8) +#define AS_FAULTSTATUS_ACCESS_TYPE_ATOMIC (0x0 << 8) +#define AS_FAULTSTATUS_ACCESS_TYPE_EX (0x1 << 8) +#define AS_FAULTSTATUS_ACCESS_TYPE_READ (0x2 << 8) +#define AS_FAULTSTATUS_ACCESS_TYPE_WRITE (0x3 << 8) + +#endif diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c index a7cbf6c9a153..20a9c296d027 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c @@ -541,6 +541,18 @@ static void vop_core_clks_disable(struct vop *vop) clk_disable(vop->hclk); } +static void vop_win_disable(struct vop *vop, const struct vop_win_data *win) +{ + if (win->phy->scl && win->phy->scl->ext) { + VOP_SCL_SET_EXT(vop, win, yrgb_hor_scl_mode, SCALE_NONE); + VOP_SCL_SET_EXT(vop, win, yrgb_ver_scl_mode, SCALE_NONE); + VOP_SCL_SET_EXT(vop, win, cbcr_hor_scl_mode, SCALE_NONE); + VOP_SCL_SET_EXT(vop, win, cbcr_ver_scl_mode, SCALE_NONE); + } + + VOP_WIN_SET(vop, win, enable, 0); +} + static int vop_enable(struct drm_crtc *crtc) { struct vop *vop = to_vop(crtc); @@ -586,7 +598,7 @@ static int vop_enable(struct drm_crtc *crtc) struct vop_win *vop_win = &vop->win[i]; const struct vop_win_data *win = vop_win->data; - VOP_WIN_SET(vop, win, enable, 0); + vop_win_disable(vop, win); } spin_unlock(&vop->reg_lock); @@ -735,7 +747,7 @@ static void vop_plane_atomic_disable(struct drm_plane *plane, spin_lock(&vop->reg_lock); - VOP_WIN_SET(vop, win, enable, 0); + vop_win_disable(vop, win); spin_unlock(&vop->reg_lock); } @@ -1631,7 +1643,7 @@ static int vop_initial(struct vop *vop) int channel = i * 2 + 1; VOP_WIN_SET(vop, win, channel, (channel + 1) << 4 | channel); - VOP_WIN_SET(vop, win, enable, 0); + vop_win_disable(vop, win); VOP_WIN_SET(vop, win, gate, 1); } diff --git a/drivers/gpu/drm/sun4i/sun4i_backend.c b/drivers/gpu/drm/sun4i/sun4i_backend.c index ee59da4a0172..4e5922c89d7b 100644 --- a/drivers/gpu/drm/sun4i/sun4i_backend.c +++ b/drivers/gpu/drm/sun4i/sun4i_backend.c @@ -361,13 +361,6 @@ int sun4i_backend_update_layer_buffer(struct sun4i_backend *backend, paddr = drm_fb_cma_get_gem_addr(fb, state, 0); DRM_DEBUG_DRIVER("Setting buffer address to %pad\n", &paddr); - /* - * backend DMA accesses DRAM directly, bypassing the system - * bus. As such, the address range is different and the buffer - * address needs to be corrected. - */ - paddr -= PHYS_OFFSET; - if (fb->format->is_yuv) return sun4i_backend_update_yuv_buffer(backend, fb, paddr); @@ -803,6 +796,27 @@ static int sun4i_backend_bind(struct device *dev, struct device *master, dev_set_drvdata(dev, backend); spin_lock_init(&backend->frontend_lock); + if (of_find_property(dev->of_node, "interconnects", NULL)) { + /* + * This assume we have the same DMA constraints for all our the + * devices in our pipeline (all the backends, but also the + * frontends). This sounds bad, but it has always been the case + * for us, and DRM doesn't do per-device allocation either, so + * we would need to fix DRM first... + */ + ret = of_dma_configure(drm->dev, dev->of_node, true); + if (ret) + return ret; + } else { + /* + * If we don't have the interconnect property, most likely + * because of an old DT, we need to set the DMA offset by hand + * on our device since the RAM mapping is at 0 for the DMA bus, + * unlike the CPU. + */ + drm->dev->dma_pfn_offset = PHYS_PFN_OFFSET; + } + backend->engine.node = dev->of_node; backend->engine.ops = &sun4i_backend_engine_ops; backend->engine.id = sun4i_backend_of_get_id(dev->of_node); diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c index fa92e992a282..9d8d8124b1f6 100644 --- a/drivers/gpu/drm/sun4i/sun4i_tcon.c +++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c @@ -236,8 +236,8 @@ static struct sun4i_tcon *sun4i_get_tcon0(struct drm_device *drm) return NULL; } -void sun4i_tcon_set_mux(struct sun4i_tcon *tcon, int channel, - const struct drm_encoder *encoder) +static void sun4i_tcon_set_mux(struct sun4i_tcon *tcon, int channel, + const struct drm_encoder *encoder) { int ret = -ENOTSUPP; diff --git a/drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c b/drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c index dc47720c99ba..39d8509d96a0 100644 --- a/drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c +++ b/drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c @@ -48,8 +48,13 @@ static enum drm_mode_status sun8i_dw_hdmi_mode_valid_h6(struct drm_connector *connector, const struct drm_display_mode *mode) { - /* This is max for HDMI 2.0b (4K@60Hz) */ - if (mode->clock > 594000) + /* + * Controller support maximum of 594 MHz, which correlates to + * 4K@60Hz 4:4:4 or RGB. However, for frequencies greater than + * 340 MHz scrambling has to be enabled. Because scrambling is + * not yet implemented, just limit to 340 MHz for now. + */ + if (mode->clock > 340000) return MODE_CLOCK_HIGH; return MODE_OK; diff --git a/drivers/gpu/drm/sun4i/sun8i_tcon_top.c b/drivers/gpu/drm/sun4i/sun8i_tcon_top.c index fc36e0c10a37..3267d0f9b9b2 100644 --- a/drivers/gpu/drm/sun4i/sun8i_tcon_top.c +++ b/drivers/gpu/drm/sun4i/sun8i_tcon_top.c @@ -227,7 +227,7 @@ static int sun8i_tcon_top_bind(struct device *dev, struct device *master, err_unregister_gates: for (i = 0; i < CLK_NUM; i++) - if (clk_data->hws[i]) + if (!IS_ERR_OR_NULL(clk_data->hws[i])) clk_hw_unregister_gate(clk_data->hws[i]); clk_disable_unprepare(tcon_top->bus); err_assert_reset: @@ -245,7 +245,8 @@ static void sun8i_tcon_top_unbind(struct device *dev, struct device *master, of_clk_del_provider(dev->of_node); for (i = 0; i < CLK_NUM; i++) - clk_hw_unregister_gate(clk_data->hws[i]); + if (clk_data->hws[i]) + clk_hw_unregister_gate(clk_data->hws[i]); clk_disable_unprepare(tcon_top->bus); reset_control_assert(tcon_top->rst); @@ -268,12 +269,12 @@ static int sun8i_tcon_top_remove(struct platform_device *pdev) return 0; } -const struct sun8i_tcon_top_quirks sun8i_r40_tcon_top_quirks = { +static const struct sun8i_tcon_top_quirks sun8i_r40_tcon_top_quirks = { .has_tcon_tv1 = true, .has_dsi = true, }; -const struct sun8i_tcon_top_quirks sun50i_h6_tcon_top_quirks = { +static const struct sun8i_tcon_top_quirks sun50i_h6_tcon_top_quirks = { /* Nothing special */ }; diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c index 4f80100ff5f3..4cce11fd8836 100644 --- a/drivers/gpu/drm/tegra/gem.c +++ b/drivers/gpu/drm/tegra/gem.c @@ -204,7 +204,7 @@ static void tegra_bo_free(struct drm_device *drm, struct tegra_bo *bo) { if (bo->pages) { dma_unmap_sg(drm->dev, bo->sgt->sgl, bo->sgt->nents, - DMA_BIDIRECTIONAL); + DMA_FROM_DEVICE); drm_gem_put_pages(&bo->gem, bo->pages, true, true); sg_free_table(bo->sgt); kfree(bo->sgt); @@ -230,7 +230,7 @@ static int tegra_bo_get_pages(struct drm_device *drm, struct tegra_bo *bo) } err = dma_map_sg(drm->dev, bo->sgt->sgl, bo->sgt->nents, - DMA_BIDIRECTIONAL); + DMA_FROM_DEVICE); if (err == 0) { err = -EFAULT; goto free_sgt; diff --git a/drivers/gpu/drm/tegra/hub.c b/drivers/gpu/drm/tegra/hub.c index ba9b3cfb8c3d..b3436c2aed68 100644 --- a/drivers/gpu/drm/tegra/hub.c +++ b/drivers/gpu/drm/tegra/hub.c @@ -378,14 +378,16 @@ static int tegra_shared_plane_atomic_check(struct drm_plane *plane, static void tegra_shared_plane_atomic_disable(struct drm_plane *plane, struct drm_plane_state *old_state) { - struct tegra_dc *dc = to_tegra_dc(old_state->crtc); struct tegra_plane *p = to_tegra_plane(plane); + struct tegra_dc *dc; u32 value; /* rien ne va plus */ if (!old_state || !old_state->crtc) return; + dc = to_tegra_dc(old_state->crtc); + /* * XXX Legacy helpers seem to sometimes call ->atomic_disable() even * on planes that are already disabled. Make sure we fallback to the diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c index 40057106f5f3..5be5a0817dfe 100644 --- a/drivers/gpu/drm/tegra/sor.c +++ b/drivers/gpu/drm/tegra/sor.c @@ -2871,6 +2871,13 @@ static int tegra_sor_init(struct host1x_client *client) * kernel is possible. */ if (sor->rst) { + err = reset_control_acquire(sor->rst); + if (err < 0) { + dev_err(sor->dev, "failed to acquire SOR reset: %d\n", + err); + return err; + } + err = reset_control_assert(sor->rst); if (err < 0) { dev_err(sor->dev, "failed to assert SOR reset: %d\n", @@ -2894,6 +2901,8 @@ static int tegra_sor_init(struct host1x_client *client) err); return err; } + + reset_control_release(sor->rst); } err = clk_prepare_enable(sor->clk_safe); @@ -3331,7 +3340,7 @@ static int tegra_sor_probe(struct platform_device *pdev) goto remove; } - sor->rst = devm_reset_control_get(&pdev->dev, "sor"); + sor->rst = devm_reset_control_get_exclusive_released(&pdev->dev, "sor"); if (IS_ERR(sor->rst)) { err = PTR_ERR(sor->rst); @@ -3519,6 +3528,8 @@ static int tegra_sor_suspend(struct device *dev) dev_err(dev, "failed to assert reset: %d\n", err); return err; } + + reset_control_release(sor->rst); } usleep_range(1000, 2000); @@ -3542,9 +3553,17 @@ static int tegra_sor_resume(struct device *dev) usleep_range(1000, 2000); if (sor->rst) { + err = reset_control_acquire(sor->rst); + if (err < 0) { + dev_err(dev, "failed to acquire reset: %d\n", err); + clk_disable_unprepare(sor->clk); + return err; + } + err = reset_control_deassert(sor->rst); if (err < 0) { dev_err(dev, "failed to deassert reset: %d\n", err); + reset_control_release(sor->rst); clk_disable_unprepare(sor->clk); return err; } diff --git a/drivers/gpu/drm/tegra/vic.c b/drivers/gpu/drm/tegra/vic.c index 39bfed9623de..982ce37ecde1 100644 --- a/drivers/gpu/drm/tegra/vic.c +++ b/drivers/gpu/drm/tegra/vic.c @@ -106,6 +106,7 @@ static int vic_boot(struct vic *vic) if (vic->booted) return 0; +#ifdef CONFIG_IOMMU_API if (vic->config->supports_sid) { struct iommu_fwspec *spec = dev_iommu_fwspec_get(vic->dev); u32 value; @@ -121,6 +122,7 @@ static int vic_boot(struct vic *vic) vic_writel(vic, value, VIC_THI_STREAMID1); } } +#endif /* setup clockgating registers */ vic_writel(vic, CG_IDLE_CG_DLY_CNT(4) | diff --git a/drivers/gpu/drm/tinydrm/hx8357d.c b/drivers/gpu/drm/tinydrm/hx8357d.c index fab961dded87..5773d0fb6ca1 100644 --- a/drivers/gpu/drm/tinydrm/hx8357d.c +++ b/drivers/gpu/drm/tinydrm/hx8357d.c @@ -267,7 +267,7 @@ static int hx8357d_probe(struct spi_device *spi) spi_set_drvdata(spi, drm); - drm_fbdev_generic_setup(drm, 32); + drm_fbdev_generic_setup(drm, 0); return 0; } diff --git a/drivers/gpu/drm/tinydrm/ili9225.c b/drivers/gpu/drm/tinydrm/ili9225.c index e9116ef4b5bc..4b1a587c0134 100644 --- a/drivers/gpu/drm/tinydrm/ili9225.c +++ b/drivers/gpu/drm/tinydrm/ili9225.c @@ -433,7 +433,7 @@ static int ili9225_probe(struct spi_device *spi) spi_set_drvdata(spi, drm); - drm_fbdev_generic_setup(drm, 32); + drm_fbdev_generic_setup(drm, 0); return 0; } diff --git a/drivers/gpu/drm/tinydrm/ili9341.c b/drivers/gpu/drm/tinydrm/ili9341.c index d15f85e837ae..4ade9e4b924f 100644 --- a/drivers/gpu/drm/tinydrm/ili9341.c +++ b/drivers/gpu/drm/tinydrm/ili9341.c @@ -229,7 +229,7 @@ static int ili9341_probe(struct spi_device *spi) spi_set_drvdata(spi, drm); - drm_fbdev_generic_setup(drm, 32); + drm_fbdev_generic_setup(drm, 0); return 0; } diff --git a/drivers/gpu/drm/tinydrm/mi0283qt.c b/drivers/gpu/drm/tinydrm/mi0283qt.c index c6dc31084a4e..8e169846fbd8 100644 --- a/drivers/gpu/drm/tinydrm/mi0283qt.c +++ b/drivers/gpu/drm/tinydrm/mi0283qt.c @@ -242,7 +242,7 @@ static int mi0283qt_probe(struct spi_device *spi) spi_set_drvdata(spi, drm); - drm_fbdev_generic_setup(drm, 32); + drm_fbdev_generic_setup(drm, 0); return 0; } diff --git a/drivers/gpu/drm/tinydrm/repaper.c b/drivers/gpu/drm/tinydrm/repaper.c index a29b8278324b..370629e2de94 100644 --- a/drivers/gpu/drm/tinydrm/repaper.c +++ b/drivers/gpu/drm/tinydrm/repaper.c @@ -1131,7 +1131,7 @@ static int repaper_probe(struct spi_device *spi) DRM_DEBUG_DRIVER("SPI speed: %uMHz\n", spi->max_speed_hz / 1000000); - drm_fbdev_generic_setup(drm, 32); + drm_fbdev_generic_setup(drm, 0); return 0; } diff --git a/drivers/gpu/drm/tinydrm/st7586.c b/drivers/gpu/drm/tinydrm/st7586.c index 560d7ac0cadc..36bb16a15f7e 100644 --- a/drivers/gpu/drm/tinydrm/st7586.c +++ b/drivers/gpu/drm/tinydrm/st7586.c @@ -408,7 +408,7 @@ static int st7586_probe(struct spi_device *spi) DRM_DEBUG_KMS("preferred_depth=%u, rotation = %u\n", drm->mode_config.preferred_depth, rotation); - drm_fbdev_generic_setup(drm, 32); + drm_fbdev_generic_setup(drm, 0); return 0; } diff --git a/drivers/gpu/drm/tinydrm/st7735r.c b/drivers/gpu/drm/tinydrm/st7735r.c index 022e9849b95b..ce9109e613e0 100644 --- a/drivers/gpu/drm/tinydrm/st7735r.c +++ b/drivers/gpu/drm/tinydrm/st7735r.c @@ -207,7 +207,7 @@ static int st7735r_probe(struct spi_device *spi) spi_set_drvdata(spi, drm); - drm_fbdev_generic_setup(drm, 32); + drm_fbdev_generic_setup(drm, 0); return 0; } diff --git a/drivers/gpu/drm/udl/udl_connector.c b/drivers/gpu/drm/udl/udl_connector.c index 66885c24590f..c1bd5e3d9e4a 100644 --- a/drivers/gpu/drm/udl/udl_connector.c +++ b/drivers/gpu/drm/udl/udl_connector.c @@ -18,18 +18,19 @@ #include "udl_connector.h" #include "udl_drv.h" -static bool udl_get_edid_block(struct udl_device *udl, int block_idx, - u8 *buff) +static int udl_get_edid_block(void *data, u8 *buf, unsigned int block, + size_t len) { int ret, i; u8 *read_buff; + struct udl_device *udl = data; read_buff = kmalloc(2, GFP_KERNEL); if (!read_buff) - return false; + return -1; - for (i = 0; i < EDID_LENGTH; i++) { - int bval = (i + block_idx * EDID_LENGTH) << 8; + for (i = 0; i < len; i++) { + int bval = (i + block * EDID_LENGTH) << 8; ret = usb_control_msg(udl->udev, usb_rcvctrlpipe(udl->udev, 0), (0x02), (0x80 | (0x02 << 5)), bval, @@ -37,60 +38,13 @@ static bool udl_get_edid_block(struct udl_device *udl, int block_idx, if (ret < 1) { DRM_ERROR("Read EDID byte %d failed err %x\n", i, ret); kfree(read_buff); - return false; + return -1; } - buff[i] = read_buff[1]; + buf[i] = read_buff[1]; } kfree(read_buff); - return true; -} - -static bool udl_get_edid(struct udl_device *udl, u8 **result_buff, - int *result_buff_size) -{ - int i, extensions; - u8 *block_buff = NULL, *buff_ptr; - - block_buff = kmalloc(EDID_LENGTH, GFP_KERNEL); - if (block_buff == NULL) - return false; - - if (udl_get_edid_block(udl, 0, block_buff) && - memchr_inv(block_buff, 0, EDID_LENGTH)) { - extensions = ((struct edid *)block_buff)->extensions; - if (extensions > 0) { - /* we have to read all extensions one by one */ - *result_buff_size = EDID_LENGTH * (extensions + 1); - *result_buff = kmalloc(*result_buff_size, GFP_KERNEL); - buff_ptr = *result_buff; - if (buff_ptr == NULL) { - kfree(block_buff); - return false; - } - memcpy(buff_ptr, block_buff, EDID_LENGTH); - kfree(block_buff); - buff_ptr += EDID_LENGTH; - for (i = 1; i < extensions; ++i) { - if (udl_get_edid_block(udl, i, buff_ptr)) { - buff_ptr += EDID_LENGTH; - } else { - kfree(*result_buff); - *result_buff = NULL; - return false; - } - } - return true; - } - /* we have only base edid block */ - *result_buff = block_buff; - *result_buff_size = EDID_LENGTH; - return true; - } - - kfree(block_buff); - - return false; + return 0; } static int udl_get_modes(struct drm_connector *connector) @@ -122,8 +76,6 @@ static enum drm_mode_status udl_mode_valid(struct drm_connector *connector, static enum drm_connector_status udl_detect(struct drm_connector *connector, bool force) { - u8 *edid_buff = NULL; - int edid_buff_size = 0; struct udl_device *udl = connector->dev->dev_private; struct udl_drm_connector *udl_connector = container_of(connector, @@ -136,12 +88,10 @@ udl_detect(struct drm_connector *connector, bool force) udl_connector->edid = NULL; } - - if (!udl_get_edid(udl, &edid_buff, &edid_buff_size)) + udl_connector->edid = drm_do_get_edid(connector, udl_get_edid_block, udl); + if (!udl_connector->edid) return connector_status_disconnected; - udl_connector->edid = (struct edid *)edid_buff; - return connector_status_connected; } diff --git a/drivers/gpu/drm/udl/udl_drv.c b/drivers/gpu/drm/udl/udl_drv.c index 53b7b8c04bc6..b7d01b3664cb 100644 --- a/drivers/gpu/drm/udl/udl_drv.c +++ b/drivers/gpu/drm/udl/udl_drv.c @@ -52,6 +52,7 @@ static struct drm_driver driver = { .driver_features = DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME, .load = udl_driver_load, .unload = udl_driver_unload, + .release = udl_driver_release, /* gem hooks */ .gem_free_object_unlocked = udl_gem_free_object, diff --git a/drivers/gpu/drm/udl/udl_drv.h b/drivers/gpu/drm/udl/udl_drv.h index e9e9b1ff678e..4ae67d882eae 100644 --- a/drivers/gpu/drm/udl/udl_drv.h +++ b/drivers/gpu/drm/udl/udl_drv.h @@ -104,6 +104,7 @@ void udl_urb_completion(struct urb *urb); int udl_driver_load(struct drm_device *dev, unsigned long flags); void udl_driver_unload(struct drm_device *dev); +void udl_driver_release(struct drm_device *dev); int udl_fbdev_init(struct drm_device *dev); void udl_fbdev_cleanup(struct drm_device *dev); diff --git a/drivers/gpu/drm/udl/udl_main.c b/drivers/gpu/drm/udl/udl_main.c index 9086d0d1b880..1f8ef34ade24 100644 --- a/drivers/gpu/drm/udl/udl_main.c +++ b/drivers/gpu/drm/udl/udl_main.c @@ -379,6 +379,12 @@ void udl_driver_unload(struct drm_device *dev) udl_free_urb_list(dev); udl_fbdev_cleanup(dev); - udl_modeset_cleanup(dev); kfree(udl); } + +void udl_driver_release(struct drm_device *dev) +{ + udl_modeset_cleanup(dev); + drm_dev_fini(dev); + kfree(dev); +} diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vgem_drv.c index 5930facd6d2d..11a8f99ba18c 100644 --- a/drivers/gpu/drm/vgem/vgem_drv.c +++ b/drivers/gpu/drm/vgem/vgem_drv.c @@ -191,13 +191,9 @@ static struct drm_gem_object *vgem_gem_create(struct drm_device *dev, ret = drm_gem_handle_create(file, &obj->base, handle); drm_gem_object_put_unlocked(&obj->base); if (ret) - goto err; + return ERR_PTR(ret); return &obj->base; - -err: - __vgem_gem_destroy(obj); - return ERR_PTR(ret); } static int vgem_gem_dumb_create(struct drm_file *file, struct drm_device *dev, diff --git a/drivers/gpu/drm/vkms/vkms_gem.c b/drivers/gpu/drm/vkms/vkms_gem.c index 138b0bb325cf..69048e73377d 100644 --- a/drivers/gpu/drm/vkms/vkms_gem.c +++ b/drivers/gpu/drm/vkms/vkms_gem.c @@ -111,11 +111,8 @@ struct drm_gem_object *vkms_gem_create(struct drm_device *dev, ret = drm_gem_handle_create(file, &obj->gem, handle); drm_gem_object_put_unlocked(&obj->gem); - if (ret) { - drm_gem_object_release(&obj->gem); - kfree(obj); + if (ret) return ERR_PTR(ret); - } return &obj->gem; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c index 0b9ee7fb45d6..66e14e38d5e8 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.c @@ -499,12 +499,9 @@ static int vmw_binding_scrub_shader(struct vmw_ctx_bindinfo *bi, bool rebind) SVGA3dCmdSetShader body; } *cmd; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for shader " - "unbinding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_SET_SHADER; cmd->header.size = sizeof(cmd->body); @@ -534,12 +531,9 @@ static int vmw_binding_scrub_render_target(struct vmw_ctx_bindinfo *bi, SVGA3dCmdSetRenderTarget body; } *cmd; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for render target " - "unbinding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_SETRENDERTARGET; cmd->header.size = sizeof(cmd->body); @@ -576,12 +570,9 @@ static int vmw_binding_scrub_texture(struct vmw_ctx_bindinfo *bi, } body; } *cmd; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for texture " - "unbinding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_SETTEXTURESTATE; cmd->header.size = sizeof(cmd->body); @@ -610,12 +601,10 @@ static int vmw_binding_scrub_dx_shader(struct vmw_ctx_bindinfo *bi, bool rebind) SVGA3dCmdDXSetShader body; } *cmd; - cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), bi->ctx->id); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for DX shader " - "unbinding.\n"); + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), bi->ctx->id); + if (unlikely(cmd == NULL)) return -ENOMEM; - } + cmd->header.id = SVGA_3D_CMD_DX_SET_SHADER; cmd->header.size = sizeof(cmd->body); cmd->body.type = binding->shader_slot + SVGA3D_SHADERTYPE_MIN; @@ -641,12 +630,9 @@ static int vmw_binding_scrub_cb(struct vmw_ctx_bindinfo *bi, bool rebind) SVGA3dCmdDXSetSingleConstantBuffer body; } *cmd; - cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), bi->ctx->id); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for DX shader " - "unbinding.\n"); + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), bi->ctx->id); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_SET_SINGLE_CONSTANT_BUFFER; cmd->header.size = sizeof(cmd->body); @@ -768,12 +754,9 @@ static int vmw_emit_set_sr(struct vmw_ctx_binding_state *cbs, view_id_size = cbs->bind_cmd_count*sizeof(uint32); cmd_size = sizeof(*cmd) + view_id_size; - cmd = vmw_fifo_reserve_dx(ctx->dev_priv, cmd_size, ctx->id); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for DX shader" - " resource binding.\n"); + cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_SET_SHADER_RESOURCES; cmd->header.size = sizeof(cmd->body) + view_id_size; @@ -807,12 +790,9 @@ static int vmw_emit_set_rt(struct vmw_ctx_binding_state *cbs) vmw_collect_view_ids(cbs, loc, SVGA3D_MAX_SIMULTANEOUS_RENDER_TARGETS); view_id_size = cbs->bind_cmd_count*sizeof(uint32); cmd_size = sizeof(*cmd) + view_id_size; - cmd = vmw_fifo_reserve_dx(ctx->dev_priv, cmd_size, ctx->id); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for DX render-target" - " binding.\n"); + cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_SET_RENDERTARGETS; cmd->header.size = sizeof(cmd->body) + view_id_size; @@ -894,12 +874,9 @@ static int vmw_emit_set_so(struct vmw_ctx_binding_state *cbs) so_target_size = cbs->bind_cmd_count*sizeof(SVGA3dSoTarget); cmd_size = sizeof(*cmd) + so_target_size; - cmd = vmw_fifo_reserve_dx(ctx->dev_priv, cmd_size, ctx->id); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for DX SO target" - " binding.\n"); + cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_SET_SOTARGETS; cmd->header.size = sizeof(cmd->body) + so_target_size; @@ -1011,12 +988,9 @@ static int vmw_emit_set_vb(struct vmw_ctx_binding_state *cbs) set_vb_size = cbs->bind_cmd_count*sizeof(SVGA3dVertexBuffer); cmd_size = sizeof(*cmd) + set_vb_size; - cmd = vmw_fifo_reserve_dx(ctx->dev_priv, cmd_size, ctx->id); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for DX vertex buffer" - " binding.\n"); + cmd = VMW_FIFO_RESERVE_DX(ctx->dev_priv, cmd_size, ctx->id); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_SET_VERTEX_BUFFERS; cmd->header.size = sizeof(cmd->body) + set_vb_size; @@ -1167,12 +1141,10 @@ static int vmw_binding_scrub_ib(struct vmw_ctx_bindinfo *bi, bool rebind) SVGA3dCmdDXSetIndexBuffer body; } *cmd; - cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), bi->ctx->id); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for DX index buffer " - "binding.\n"); + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), bi->ctx->id); + if (unlikely(cmd == NULL)) return -ENOMEM; - } + cmd->header.id = SVGA_3D_CMD_DX_SET_INDEX_BUFFER; cmd->header.size = sizeof(cmd->body); if (rebind) { @@ -1269,6 +1241,32 @@ void vmw_binding_state_reset(struct vmw_ctx_binding_state *cbs) vmw_binding_drop(entry); } +/** + * vmw_binding_dirtying - Return whether a binding type is dirtying its resource + * @binding_type: The binding type + * + * Each time a resource is put on the validation list as the result of a + * context binding referencing it, we need to determine whether that resource + * will be dirtied (written to by the GPU) as a result of the corresponding + * GPU operation. Currently rendertarget-, depth-stencil-, and + * stream-output-target bindings are capable of dirtying its resource. + * + * Return: Whether the binding type dirties the resource its binding points to. + */ +u32 vmw_binding_dirtying(enum vmw_ctx_binding_type binding_type) +{ + static u32 is_binding_dirtying[vmw_ctx_binding_max] = { + [vmw_ctx_binding_rt] = VMW_RES_DIRTY_SET, + [vmw_ctx_binding_dx_rt] = VMW_RES_DIRTY_SET, + [vmw_ctx_binding_ds] = VMW_RES_DIRTY_SET, + [vmw_ctx_binding_so] = VMW_RES_DIRTY_SET, + }; + + /* Review this function as new bindings are added. */ + BUILD_BUG_ON(vmw_ctx_binding_max != 11); + return is_binding_dirtying[binding_type]; +} + /* * This function is unused at run-time, and only used to hold various build * asserts important for code optimization assumptions. diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h index 6a2a9d69043b..f6ab79d23923 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_binding.h @@ -205,5 +205,7 @@ extern void vmw_binding_state_free(struct vmw_ctx_binding_state *cbs); extern struct list_head * vmw_binding_state_list(struct vmw_ctx_binding_state *cbs); extern void vmw_binding_state_reset(struct vmw_ctx_binding_state *cbs); +extern u32 vmw_binding_dirtying(enum vmw_ctx_binding_type binding_type); + #endif diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c index 70dab55e7888..56979e412ca8 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c @@ -393,6 +393,7 @@ static void vmw_cmdbuf_ctx_process(struct vmw_cmdbuf_man *man, __vmw_cmdbuf_header_free(entry); break; case SVGA_CB_STATUS_COMMAND_ERROR: + WARN_ONCE(true, "Command buffer error.\n"); entry->cb_header->status = SVGA_CB_STATUS_NONE; list_add_tail(&entry->list, &man->error); schedule_work(&man->work); @@ -511,17 +512,14 @@ static void vmw_cmdbuf_work_func(struct work_struct *work) container_of(work, struct vmw_cmdbuf_man, work); struct vmw_cmdbuf_header *entry, *next; uint32_t dummy; - bool restart[SVGA_CB_CONTEXT_MAX]; bool send_fence = false; struct list_head restart_head[SVGA_CB_CONTEXT_MAX]; int i; struct vmw_cmdbuf_context *ctx; bool global_block = false; - for_each_cmdbuf_ctx(man, i, ctx) { + for_each_cmdbuf_ctx(man, i, ctx) INIT_LIST_HEAD(&restart_head[i]); - restart[i] = false; - } mutex_lock(&man->error_mutex); spin_lock(&man->lock); @@ -533,23 +531,23 @@ static void vmw_cmdbuf_work_func(struct work_struct *work) const char *cmd_name; list_del_init(&entry->list); - restart[entry->cb_context] = true; global_block = true; if (!vmw_cmd_describe(header, &error_cmd_size, &cmd_name)) { - DRM_ERROR("Unknown command causing device error.\n"); - DRM_ERROR("Command buffer offset is %lu\n", - (unsigned long) cb_hdr->errorOffset); + VMW_DEBUG_USER("Unknown command causing device error.\n"); + VMW_DEBUG_USER("Command buffer offset is %lu\n", + (unsigned long) cb_hdr->errorOffset); __vmw_cmdbuf_header_free(entry); send_fence = true; continue; } - DRM_ERROR("Command \"%s\" causing device error.\n", cmd_name); - DRM_ERROR("Command buffer offset is %lu\n", - (unsigned long) cb_hdr->errorOffset); - DRM_ERROR("Command size is %lu\n", - (unsigned long) error_cmd_size); + VMW_DEBUG_USER("Command \"%s\" causing device error.\n", + cmd_name); + VMW_DEBUG_USER("Command buffer offset is %lu\n", + (unsigned long) cb_hdr->errorOffset); + VMW_DEBUG_USER("Command size is %lu\n", + (unsigned long) error_cmd_size); new_start_offset = cb_hdr->errorOffset + error_cmd_size; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c index 14bd760a62fd..63f111068a44 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c @@ -156,12 +156,9 @@ static void vmw_hw_context_destroy(struct vmw_resource *res) } vmw_execbuf_release_pinned_bo(dev_priv); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for surface " - "destruction.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return; - } cmd->header.id = SVGA_3D_CMD_CONTEXT_DESTROY; cmd->header.size = sizeof(cmd->body); @@ -210,7 +207,7 @@ static int vmw_gb_context_init(struct vmw_private *dev_priv, for (i = 0; i < SVGA_COTABLE_DX10_MAX; ++i) { uctx->cotables[i] = vmw_cotable_alloc(dev_priv, &uctx->res, i); - if (unlikely(IS_ERR(uctx->cotables[i]))) { + if (IS_ERR(uctx->cotables[i])) { ret = PTR_ERR(uctx->cotables[i]); goto out_cotables; } @@ -259,9 +256,8 @@ static int vmw_context_init(struct vmw_private *dev_priv, return -ENOMEM; } - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); if (unlikely(cmd == NULL)) { - DRM_ERROR("Fifo reserve failed.\n"); vmw_resource_unreference(&res); return -ENOMEM; } @@ -311,10 +307,8 @@ static int vmw_gb_context_create(struct vmw_resource *res) goto out_no_fifo; } - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for context " - "creation.\n"); ret = -ENOMEM; goto out_no_fifo; } @@ -345,12 +339,10 @@ static int vmw_gb_context_bind(struct vmw_resource *res, BUG_ON(bo->mem.mem_type != VMW_PL_MOB); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for context " - "binding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } + cmd->header.id = SVGA_3D_CMD_BIND_GB_CONTEXT; cmd->header.size = sizeof(cmd->body); cmd->body.cid = res->id; @@ -391,10 +383,8 @@ static int vmw_gb_context_unbind(struct vmw_resource *res, submit_size = sizeof(*cmd2) + (readback ? sizeof(*cmd1) : 0); - cmd = vmw_fifo_reserve(dev_priv, submit_size); + cmd = VMW_FIFO_RESERVE(dev_priv, submit_size); if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for context " - "unbinding.\n"); mutex_unlock(&dev_priv->binding_mutex); return -ENOMEM; } @@ -441,12 +431,9 @@ static int vmw_gb_context_destroy(struct vmw_resource *res) if (likely(res->id == -1)) return 0; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for context " - "destruction.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DESTROY_GB_CONTEXT; cmd->header.size = sizeof(cmd->body); @@ -487,10 +474,8 @@ static int vmw_dx_context_create(struct vmw_resource *res) goto out_no_fifo; } - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for context " - "creation.\n"); ret = -ENOMEM; goto out_no_fifo; } @@ -521,12 +506,9 @@ static int vmw_dx_context_bind(struct vmw_resource *res, BUG_ON(bo->mem.mem_type != VMW_PL_MOB); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for context " - "binding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_BIND_CONTEXT; cmd->header.size = sizeof(cmd->body); @@ -615,10 +597,8 @@ static int vmw_dx_context_unbind(struct vmw_resource *res, submit_size = sizeof(*cmd2) + (readback ? sizeof(*cmd1) : 0); - cmd = vmw_fifo_reserve(dev_priv, submit_size); + cmd = VMW_FIFO_RESERVE(dev_priv, submit_size); if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for context " - "unbinding.\n"); mutex_unlock(&dev_priv->binding_mutex); return -ENOMEM; } @@ -665,12 +645,9 @@ static int vmw_dx_context_destroy(struct vmw_resource *res) if (likely(res->id == -1)) return 0; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for context " - "destruction.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_DESTROY_CONTEXT; cmd->header.size = sizeof(cmd->body); @@ -751,7 +728,7 @@ static int vmw_context_define(struct drm_device *dev, void *data, int ret; if (!dev_priv->has_dx && dx) { - DRM_ERROR("DX contexts not supported by device.\n"); + VMW_DEBUG_USER("DX contexts not supported by device.\n"); return -EINVAL; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c index 44f3f6f107d3..b4f6e1217c9d 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c @@ -171,12 +171,9 @@ static int vmw_cotable_unscrub(struct vmw_resource *res) WARN_ON_ONCE(bo->mem.mem_type != VMW_PL_MOB); lockdep_assert_held(&bo->resv->lock.base); - cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), SVGA3D_INVALID_ID); - if (!cmd) { - DRM_ERROR("Failed reserving FIFO space for cotable " - "binding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (!cmd) return -ENOMEM; - } WARN_ON(vcotbl->ctx->id == SVGA3D_INVALID_ID); WARN_ON(bo->mem.mem_type != VMW_PL_MOB); @@ -262,12 +259,9 @@ int vmw_cotable_scrub(struct vmw_resource *res, bool readback) if (readback) submit_size += sizeof(*cmd0); - cmd1 = vmw_fifo_reserve_dx(dev_priv, submit_size, SVGA3D_INVALID_ID); - if (!cmd1) { - DRM_ERROR("Failed reserving FIFO space for cotable " - "unbinding.\n"); + cmd1 = VMW_FIFO_RESERVE(dev_priv, submit_size); + if (!cmd1) return -ENOMEM; - } vcotbl->size_read_back = 0; if (readback) { @@ -351,13 +345,10 @@ static int vmw_cotable_readback(struct vmw_resource *res) struct vmw_fence_obj *fence; if (!vcotbl->scrubbed) { - cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), - SVGA3D_INVALID_ID); - if (!cmd) { - DRM_ERROR("Failed reserving FIFO space for cotable " - "readback.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (!cmd) return -ENOMEM; - } + cmd->header.id = SVGA_3D_CMD_DX_READBACK_COTABLE; cmd->header.size = sizeof(cmd->body); cmd->body.cid = vcotbl->ctx->id; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 6302c12c2298..96983c47fb40 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -699,6 +699,8 @@ extern int vmw_user_stream_lookup(struct vmw_private *dev_priv, uint32_t *inout_id, struct vmw_resource **out); extern void vmw_resource_unreserve(struct vmw_resource *res, + bool dirty_set, + bool dirty, bool switch_backup, struct vmw_buffer_object *new_backup, unsigned long new_backup_offset); @@ -811,7 +813,6 @@ extern int vmw_fifo_init(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo); extern void vmw_fifo_release(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo); -extern void *vmw_fifo_reserve(struct vmw_private *dev_priv, uint32_t bytes); extern void * vmw_fifo_reserve_dx(struct vmw_private *dev_priv, uint32_t bytes, int ctx_id); extern void vmw_fifo_commit(struct vmw_private *dev_priv, uint32_t bytes); @@ -827,6 +828,18 @@ extern int vmw_fifo_emit_dummy_query(struct vmw_private *dev_priv, extern int vmw_fifo_flush(struct vmw_private *dev_priv, bool interruptible); +#define VMW_FIFO_RESERVE_DX(__priv, __bytes, __ctx_id) \ +({ \ + vmw_fifo_reserve_dx(__priv, __bytes, __ctx_id) ? : ({ \ + DRM_ERROR("FIFO reserve failed at %s for %u bytes\n", \ + __func__, (unsigned int) __bytes); \ + NULL; \ + }); \ +}) + +#define VMW_FIFO_RESERVE(__priv, __bytes) \ + VMW_FIFO_RESERVE_DX(__priv, __bytes, SVGA3D_INVALID_ID) + /** * TTM glue - vmwgfx_ttm_glue.c */ @@ -1311,6 +1324,20 @@ int vmw_host_get_guestinfo(const char *guest_info_param, char *buffer, size_t *length); int vmw_host_log(const char *log); +/* VMW logging */ + +/** + * VMW_DEBUG_USER - Debug output for user-space debugging. + * + * @fmt: printf() like format string. + * + * This macro is for logging user-space error and debugging messages for e.g. + * command buffer execution errors due to malformed commands, invalid context, + * etc. + */ +#define VMW_DEBUG_USER(fmt, ...) \ + DRM_DEBUG_DRIVER(fmt, ##__VA_ARGS__) + /** * Inline helper functions */ diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 88b8178d4687..2ff7ba04d8c8 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -36,6 +36,25 @@ #define VMW_RES_HT_ORDER 12 /* + * Helper macro to get dx_ctx_node if available otherwise print an error + * message. This is for use in command verifier function where if dx_ctx_node + * is not set then command is invalid. + */ +#define VMW_GET_CTX_NODE(__sw_context) \ +({ \ + __sw_context->dx_ctx_node ? __sw_context->dx_ctx_node : ({ \ + VMW_DEBUG_USER("SM context is not set at %s\n", __func__); \ + __sw_context->dx_ctx_node; \ + }); \ +}) + +#define VMW_DECLARE_CMD_VAR(__var, __type) \ + struct { \ + SVGA3dCmdHeader header; \ + __type body; \ + } __var + +/** * struct vmw_relocation - Buffer object relocation * * @head: List head for the command submission context's relocation list @@ -59,9 +78,8 @@ struct vmw_relocation { * command stream is replaced with the actual id after validation. * @vmw_res_rel_nop: NOP relocation. The command is unconditionally replaced * with a NOP. - * @vmw_res_rel_cond_nop: Conditional NOP relocation. If the resource id - * after validation is -1, the command is replaced with a NOP. Otherwise no - * action. + * @vmw_res_rel_cond_nop: Conditional NOP relocation. If the resource id after + * validation is -1, the command is replaced with a NOP. Otherwise no action. */ enum vmw_resource_relocation_type { vmw_res_rel_normal, @@ -75,8 +93,8 @@ enum vmw_resource_relocation_type { * * @head: List head for the software context's relocation list. * @res: Non-ref-counted pointer to the resource. - * @offset: Offset of single byte entries into the command buffer where the - * id that needs fixup is located. + * @offset: Offset of single byte entries into the command buffer where the id + * that needs fixup is located. * @rel_type: Type of relocation. */ struct vmw_resource_relocation { @@ -86,8 +104,9 @@ struct vmw_resource_relocation { enum vmw_resource_relocation_type rel_type:3; }; -/* +/** * struct vmw_ctx_validation_info - Extra validation metadata for contexts + * * @head: List head of context list * @ctx: The context resource * @cur: The context's persistent binding state @@ -142,9 +161,10 @@ static size_t vmw_ptr_diff(void *a, void *b) /** * vmw_execbuf_bindings_commit - Commit modified binding state + * * @sw_context: The command submission context - * @backoff: Whether this is part of the error path and binding state - * changes should be ignored + * @backoff: Whether this is part of the error path and binding state changes + * should be ignored */ static void vmw_execbuf_bindings_commit(struct vmw_sw_context *sw_context, bool backoff) @@ -154,6 +174,7 @@ static void vmw_execbuf_bindings_commit(struct vmw_sw_context *sw_context, list_for_each_entry(entry, &sw_context->ctx_list, head) { if (!backoff) vmw_binding_state_commit(entry->cur, entry->staged); + if (entry->staged != sw_context->staged_bindings) vmw_binding_state_free(entry->staged); else @@ -166,6 +187,7 @@ static void vmw_execbuf_bindings_commit(struct vmw_sw_context *sw_context, /** * vmw_bind_dx_query_mob - Bind the DX query MOB if referenced + * * @sw_context: The command submission context */ static void vmw_bind_dx_query_mob(struct vmw_sw_context *sw_context) @@ -176,8 +198,8 @@ static void vmw_bind_dx_query_mob(struct vmw_sw_context *sw_context) } /** - * vmw_cmd_ctx_first_setup - Perform the setup needed when a context is - * added to the validate list. + * vmw_cmd_ctx_first_setup - Perform the setup needed when a context is added to + * the validate list. * * @dev_priv: Pointer to the device private: * @sw_context: The command submission context @@ -195,11 +217,8 @@ static int vmw_cmd_ctx_first_setup(struct vmw_private *dev_priv, goto out_err; if (!sw_context->staged_bindings) { - sw_context->staged_bindings = - vmw_binding_state_alloc(dev_priv); + sw_context->staged_bindings = vmw_binding_state_alloc(dev_priv); if (IS_ERR(sw_context->staged_bindings)) { - DRM_ERROR("Failed to allocate context binding " - "information.\n"); ret = PTR_ERR(sw_context->staged_bindings); sw_context->staged_bindings = NULL; goto out_err; @@ -209,8 +228,6 @@ static int vmw_cmd_ctx_first_setup(struct vmw_private *dev_priv, if (sw_context->staged_bindings_inuse) { node->staged = vmw_binding_state_alloc(dev_priv); if (IS_ERR(node->staged)) { - DRM_ERROR("Failed to allocate context binding " - "information.\n"); ret = PTR_ERR(node->staged); node->staged = NULL; goto out_err; @@ -225,19 +242,20 @@ static int vmw_cmd_ctx_first_setup(struct vmw_private *dev_priv, list_add_tail(&node->head, &sw_context->ctx_list); return 0; + out_err: return ret; } /** - * vmw_execbuf_res_size - calculate extra size fore the resource validation - * node + * vmw_execbuf_res_size - calculate extra size fore the resource validation node + * * @dev_priv: Pointer to the device private struct. * @res_type: The resource type. * - * Guest-backed contexts and DX contexts require extra size to store - * execbuf private information in the validation node. Typically the - * binding manager associated data structures. + * Guest-backed contexts and DX contexts require extra size to store execbuf + * private information in the validation node. Typically the binding manager + * associated data structures. * * Returns: The extra size requirement based on resource type. */ @@ -254,8 +272,8 @@ static unsigned int vmw_execbuf_res_size(struct vmw_private *dev_priv, * * @rcache: Pointer to the entry to update. * @res: Pointer to the resource. - * @private: Pointer to the execbuf-private space in the resource - * validation node. + * @private: Pointer to the execbuf-private space in the resource validation + * node. */ static void vmw_execbuf_rcache_update(struct vmw_res_cache_entry *rcache, struct vmw_resource *res, @@ -268,17 +286,19 @@ static void vmw_execbuf_rcache_update(struct vmw_res_cache_entry *rcache, } /** - * vmw_execbuf_res_noref_val_add - Add a resource described by an - * unreferenced rcu-protected pointer to the validation list. + * vmw_execbuf_res_noref_val_add - Add a resource described by an unreferenced + * rcu-protected pointer to the validation list. + * * @sw_context: Pointer to the software context. * @res: Unreferenced rcu-protected pointer to the resource. + * @dirty: Whether to change dirty status. * - * Returns: 0 on success. Negative error code on failure. Typical error - * codes are %-EINVAL on inconsistency and %-ESRCH if the resource was - * doomed. + * Returns: 0 on success. Negative error code on failure. Typical error codes + * are %-EINVAL on inconsistency and %-ESRCH if the resource was doomed. */ static int vmw_execbuf_res_noref_val_add(struct vmw_sw_context *sw_context, - struct vmw_resource *res) + struct vmw_resource *res, + u32 dirty) { struct vmw_private *dev_priv = res->dev_priv; int ret; @@ -290,13 +310,17 @@ static int vmw_execbuf_res_noref_val_add(struct vmw_sw_context *sw_context, rcache = &sw_context->res_cache[res_type]; if (likely(rcache->valid && rcache->res == res)) { + if (dirty) + vmw_validation_res_set_dirty(sw_context->ctx, + rcache->private, dirty); vmw_user_resource_noref_release(); return 0; } priv_size = vmw_execbuf_res_size(dev_priv, res_type); ret = vmw_validation_add_resource(sw_context->ctx, res, priv_size, - (void **)&ctx_info, &first_usage); + dirty, (void **)&ctx_info, + &first_usage); vmw_user_resource_noref_release(); if (ret) return ret; @@ -304,8 +328,10 @@ static int vmw_execbuf_res_noref_val_add(struct vmw_sw_context *sw_context, if (priv_size && first_usage) { ret = vmw_cmd_ctx_first_setup(dev_priv, sw_context, res, ctx_info); - if (ret) + if (ret) { + VMW_DEBUG_USER("Failed first usage context setup.\n"); return ret; + } } vmw_execbuf_rcache_update(rcache, res, ctx_info); @@ -315,13 +341,16 @@ static int vmw_execbuf_res_noref_val_add(struct vmw_sw_context *sw_context, /** * vmw_execbuf_res_noctx_val_add - Add a non-context resource to the resource * validation list if it's not already on it + * * @sw_context: Pointer to the software context. * @res: Pointer to the resource. + * @dirty: Whether to change dirty status. * * Returns: Zero on success. Negative error code on failure. */ static int vmw_execbuf_res_noctx_val_add(struct vmw_sw_context *sw_context, - struct vmw_resource *res) + struct vmw_resource *res, + u32 dirty) { struct vmw_res_cache_entry *rcache; enum vmw_res_type res_type = vmw_res_type(res); @@ -329,10 +358,15 @@ static int vmw_execbuf_res_noctx_val_add(struct vmw_sw_context *sw_context, int ret; rcache = &sw_context->res_cache[res_type]; - if (likely(rcache->valid && rcache->res == res)) + if (likely(rcache->valid && rcache->res == res)) { + if (dirty) + vmw_validation_res_set_dirty(sw_context->ctx, + rcache->private, dirty); return 0; + } - ret = vmw_validation_add_resource(sw_context->ctx, res, 0, &ptr, NULL); + ret = vmw_validation_add_resource(sw_context->ctx, res, 0, dirty, + &ptr, NULL); if (ret) return ret; @@ -342,8 +376,8 @@ static int vmw_execbuf_res_noctx_val_add(struct vmw_sw_context *sw_context, } /** - * vmw_view_res_val_add - Add a view and the surface it's pointing to - * to the validation list + * vmw_view_res_val_add - Add a view and the surface it's pointing to to the + * validation list * * @sw_context: The software context holding the validation list. * @view: Pointer to the view resource. @@ -356,27 +390,29 @@ static int vmw_view_res_val_add(struct vmw_sw_context *sw_context, int ret; /* - * First add the resource the view is pointing to, otherwise - * it may be swapped out when the view is validated. + * First add the resource the view is pointing to, otherwise it may be + * swapped out when the view is validated. */ - ret = vmw_execbuf_res_noctx_val_add(sw_context, vmw_view_srf(view)); + ret = vmw_execbuf_res_noctx_val_add(sw_context, vmw_view_srf(view), + vmw_view_dirtying(view)); if (ret) return ret; - return vmw_execbuf_res_noctx_val_add(sw_context, view); + return vmw_execbuf_res_noctx_val_add(sw_context, view, + VMW_RES_DIRTY_NONE); } /** - * vmw_view_id_val_add - Look up a view and add it and the surface it's - * pointing to to the validation list. + * vmw_view_id_val_add - Look up a view and add it and the surface it's pointing + * to to the validation list. * * @sw_context: The software context holding the validation list. * @view_type: The view type to look up. * @id: view id of the view. * - * The view is represented by a view id and the DX context it's created on, - * or scheduled for creation on. If there is no DX context set, the function - * will return an -EINVAL error pointer. + * The view is represented by a view id and the DX context it's created on, or + * scheduled for creation on. If there is no DX context set, the function will + * return an -EINVAL error pointer. * * Returns: Unreferenced pointer to the resource on success, negative error * pointer on failure. @@ -389,10 +425,8 @@ vmw_view_id_val_add(struct vmw_sw_context *sw_context, struct vmw_resource *view; int ret; - if (!ctx_node) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return ERR_PTR(-EINVAL); - } view = vmw_view_lookup(sw_context->man, view_type, id); if (IS_ERR(view)) @@ -413,8 +447,8 @@ vmw_view_id_val_add(struct vmw_sw_context *sw_context, * @sw_context: Pointer to a software context used for this command submission * @ctx: Pointer to the context resource * - * This function puts all resources that were previously bound to @ctx on - * the resource validation list. This is part of the context state reemission + * This function puts all resources that were previously bound to @ctx on the + * resource validation list. This is part of the context state reemission */ static int vmw_resource_context_res_add(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, @@ -433,13 +467,13 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv, if (IS_ERR(res)) continue; - ret = vmw_execbuf_res_noctx_val_add(sw_context, res); + ret = vmw_execbuf_res_noctx_val_add(sw_context, res, + VMW_RES_DIRTY_SET); if (unlikely(ret != 0)) return ret; } } - /* Add all resources bound to the context to the validation list */ mutex_lock(&dev_priv->binding_mutex); binding_list = vmw_context_binding_list(ctx); @@ -448,8 +482,9 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv, if (vmw_res_type(entry->res) == vmw_res_view) ret = vmw_view_res_val_add(sw_context, entry->res); else - ret = vmw_execbuf_res_noctx_val_add(sw_context, - entry->res); + ret = vmw_execbuf_res_noctx_val_add + (sw_context, entry->res, + vmw_binding_dirtying(entry->bt)); if (unlikely(ret != 0)) break; } @@ -472,8 +507,8 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv, * * @list: Pointer to head of relocation list. * @res: The resource. - * @offset: Offset into the command buffer currently being parsed where the - * id that needs fixup is located. Granularity is one byte. + * @offset: Offset into the command buffer currently being parsed where the id + * that needs fixup is located. Granularity is one byte. * @rel_type: Relocation type. */ static int vmw_resource_relocation_add(struct vmw_sw_context *sw_context, @@ -486,7 +521,7 @@ static int vmw_resource_relocation_add(struct vmw_sw_context *sw_context, rel = vmw_validation_mem_alloc(sw_context->ctx, sizeof(*rel)); if (unlikely(!rel)) { - DRM_ERROR("Failed to allocate a resource relocation.\n"); + VMW_DEBUG_USER("Failed to allocate a resource relocation.\n"); return -ENOMEM; } @@ -506,17 +541,15 @@ static int vmw_resource_relocation_add(struct vmw_sw_context *sw_context, static void vmw_resource_relocations_free(struct list_head *list) { /* Memory is validation context memory, so no need to free it */ - INIT_LIST_HEAD(list); } /** * vmw_resource_relocations_apply - Apply all relocations on a list * - * @cb: Pointer to the start of the command buffer bein patch. This need - * not be the same buffer as the one being parsed when the relocation - * list was built, but the contents must be the same modulo the - * resource ids. + * @cb: Pointer to the start of the command buffer bein patch. This need not be + * the same buffer as the one being parsed when the relocation list was built, + * but the contents must be the same modulo the resource ids. * @list: Pointer to the head of the relocation list. */ static void vmw_resource_relocations_apply(uint32_t *cb, @@ -560,14 +593,14 @@ static int vmw_cmd_ok(struct vmw_private *dev_priv, } /** - * vmw_resources_reserve - Reserve all resources on the sw_context's - * resource list. + * vmw_resources_reserve - Reserve all resources on the sw_context's resource + * list. * * @sw_context: Pointer to the software context. * - * Note that since vmware's command submission currently is protected by - * the cmdbuf mutex, no fancy deadlock avoidance is required for resources, - * since only a single thread at once will attempt this. + * Note that since vmware's command submission currently is protected by the + * cmdbuf mutex, no fancy deadlock avoidance is required for resources, since + * only a single thread at once will attempt this. */ static int vmw_resources_reserve(struct vmw_sw_context *sw_context) { @@ -592,22 +625,24 @@ static int vmw_resources_reserve(struct vmw_sw_context *sw_context) } /** - * vmw_cmd_res_check - Check that a resource is present and if so, put it - * on the resource validate list unless it's already there. + * vmw_cmd_res_check - Check that a resource is present and if so, put it on the + * resource validate list unless it's already there. * * @dev_priv: Pointer to a device private structure. * @sw_context: Pointer to the software context. * @res_type: Resource type. + * @dirty: Whether to change dirty status. * @converter: User-space visisble type specific information. - * @id_loc: Pointer to the location in the command buffer currently being - * parsed from where the user-space resource id handle is located. - * @p_val: Pointer to pointer to resource validalidation node. Populated - * on exit. + * @id_loc: Pointer to the location in the command buffer currently being parsed + * from where the user-space resource id handle is located. + * @p_val: Pointer to pointer to resource validalidation node. Populated on + * exit. */ static int vmw_cmd_res_check(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, enum vmw_res_type res_type, + u32 dirty, const struct vmw_user_resource_conv *converter, uint32_t *id_loc, struct vmw_resource **p_res) @@ -621,7 +656,7 @@ vmw_cmd_res_check(struct vmw_private *dev_priv, if (*id_loc == SVGA3D_INVALID_ID) { if (res_type == vmw_res_context) { - DRM_ERROR("Illegal context invalid id.\n"); + VMW_DEBUG_USER("Illegal context invalid id.\n"); return -EINVAL; } return 0; @@ -629,6 +664,9 @@ vmw_cmd_res_check(struct vmw_private *dev_priv, if (likely(rcache->valid_handle && *id_loc == rcache->handle)) { res = rcache->res; + if (dirty) + vmw_validation_res_set_dirty(sw_context->ctx, + rcache->private, dirty); } else { unsigned int size = vmw_execbuf_res_size(dev_priv, res_type); @@ -638,13 +676,13 @@ vmw_cmd_res_check(struct vmw_private *dev_priv, res = vmw_user_resource_noref_lookup_handle (dev_priv, sw_context->fp->tfile, *id_loc, converter); - if (unlikely(IS_ERR(res))) { - DRM_ERROR("Could not find or use resource 0x%08x.\n", - (unsigned int) *id_loc); + if (IS_ERR(res)) { + VMW_DEBUG_USER("Could not find/use resource 0x%08x.\n", + (unsigned int) *id_loc); return PTR_ERR(res); } - ret = vmw_execbuf_res_noref_val_add(sw_context, res); + ret = vmw_execbuf_res_noref_val_add(sw_context, res, dirty); if (unlikely(ret != 0)) return ret; @@ -675,23 +713,16 @@ static int vmw_rebind_all_dx_query(struct vmw_resource *ctx_res) { struct vmw_private *dev_priv = ctx_res->dev_priv; struct vmw_buffer_object *dx_query_mob; - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXBindAllQuery body; - } *cmd; - + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXBindAllQuery); dx_query_mob = vmw_context_get_dx_query_mob(ctx_res); if (!dx_query_mob || dx_query_mob->dx_query_ctx) return 0; - cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), ctx_res->id); - - if (cmd == NULL) { - DRM_ERROR("Failed to rebind queries.\n"); + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), ctx_res->id); + if (cmd == NULL) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_BIND_ALL_QUERY; cmd->header.size = sizeof(cmd->body); @@ -705,8 +736,8 @@ static int vmw_rebind_all_dx_query(struct vmw_resource *ctx_res) } /** - * vmw_rebind_contexts - Rebind all resources previously bound to - * referenced contexts. + * vmw_rebind_contexts - Rebind all resources previously bound to referenced + * contexts. * * @sw_context: Pointer to the software context. * @@ -721,21 +752,23 @@ static int vmw_rebind_contexts(struct vmw_sw_context *sw_context) ret = vmw_binding_rebind_all(val->cur); if (unlikely(ret != 0)) { if (ret != -ERESTARTSYS) - DRM_ERROR("Failed to rebind context.\n"); + VMW_DEBUG_USER("Failed to rebind context.\n"); return ret; } ret = vmw_rebind_all_dx_query(val->ctx); - if (ret != 0) + if (ret != 0) { + VMW_DEBUG_USER("Failed to rebind queries.\n"); return ret; + } } return 0; } /** - * vmw_view_bindings_add - Add an array of view bindings to a context - * binding state tracker. + * vmw_view_bindings_add - Add an array of view bindings to a context binding + * state tracker. * * @sw_context: The execbuf state used for this command. * @view_type: View type for the bindings. @@ -752,13 +785,11 @@ static int vmw_view_bindings_add(struct vmw_sw_context *sw_context, uint32 view_ids[], u32 num_views, u32 first_slot) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); u32 i; - if (!ctx_node) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } for (i = 0; i < num_views; ++i) { struct vmw_ctx_bindinfo_view binding; @@ -768,7 +799,7 @@ static int vmw_view_bindings_add(struct vmw_sw_context *sw_context, view = vmw_view_id_val_add(sw_context, view_type, view_ids[i]); if (IS_ERR(view)) { - DRM_ERROR("View not found.\n"); + VMW_DEBUG_USER("View not found.\n"); return PTR_ERR(view); } } @@ -798,19 +829,18 @@ static int vmw_cmd_cid_check(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_cid_cmd { - SVGA3dCmdHeader header; - uint32_t cid; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, uint32_t) = + container_of(header, typeof(*cmd), header); - cmd = container_of(header, struct vmw_cid_cmd, header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, &cmd->cid, NULL); + VMW_RES_DIRTY_SET, user_context_converter, + &cmd->body, NULL); } /** * vmw_execbuf_info_from_res - Get the private validation metadata for a * recently validated resource + * * @sw_context: Pointer to the command submission context * @res: The resource * @@ -818,8 +848,8 @@ static int vmw_cmd_cid_check(struct vmw_private *dev_priv, * context's resource cache and hence the last resource of that type to be * processed by the validation code. * - * Return: a pointer to the private metadata of the resource, or NULL - * if it wasn't found + * Return: a pointer to the private metadata of the resource, or NULL if it + * wasn't found */ static struct vmw_ctx_validation_info * vmw_execbuf_info_from_res(struct vmw_sw_context *sw_context, @@ -835,36 +865,32 @@ vmw_execbuf_info_from_res(struct vmw_sw_context *sw_context, return NULL; } - static int vmw_cmd_set_render_target_check(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_sid_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdSetRenderTarget body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSetRenderTarget); struct vmw_resource *ctx; struct vmw_resource *res; int ret; - cmd = container_of(header, struct vmw_sid_cmd, header); + cmd = container_of(header, typeof(*cmd), header); if (cmd->body.type >= SVGA3D_RT_MAX) { - DRM_ERROR("Illegal render target type %u.\n", - (unsigned) cmd->body.type); + VMW_DEBUG_USER("Illegal render target type %u.\n", + (unsigned int) cmd->body.type); return -EINVAL; } ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, &cmd->body.cid, - &ctx); + VMW_RES_DIRTY_SET, user_context_converter, + &cmd->body.cid, &ctx); if (unlikely(ret != 0)) return ret; ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, &cmd->body.target.sid, - &res); + VMW_RES_DIRTY_SET, user_surface_converter, + &cmd->body.target.sid, &res); if (unlikely(ret)) return ret; @@ -890,44 +916,38 @@ static int vmw_cmd_surface_copy_check(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_sid_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdSurfaceCopy body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSurfaceCopy); int ret; - cmd = container_of(header, struct vmw_sid_cmd, header); + cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, - &cmd->body.src.sid, NULL); + VMW_RES_DIRTY_NONE, user_surface_converter, + &cmd->body.src.sid, NULL); if (ret) return ret; return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_SET, user_surface_converter, &cmd->body.dest.sid, NULL); } static int vmw_cmd_buffer_copy_check(struct vmw_private *dev_priv, - struct vmw_sw_context *sw_context, - SVGA3dCmdHeader *header) + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXBufferCopy body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXBufferCopy); int ret; cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.src, NULL); if (ret != 0) return ret; return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_SET, user_surface_converter, &cmd->body.dest, NULL); } @@ -935,21 +955,18 @@ static int vmw_cmd_pred_copy_check(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXPredCopyRegion body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXPredCopyRegion); int ret; cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.srcSid, NULL); if (ret != 0) return ret; return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_SET, user_surface_converter, &cmd->body.dstSid, NULL); } @@ -957,20 +974,18 @@ static int vmw_cmd_stretch_blt_check(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_sid_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdSurfaceStretchBlt body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSurfaceStretchBlt); int ret; - cmd = container_of(header, struct vmw_sid_cmd, header); + cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.src.sid, NULL); if (unlikely(ret != 0)) return ret; + return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_SET, user_surface_converter, &cmd->body.dest.sid, NULL); } @@ -978,15 +993,11 @@ static int vmw_cmd_blt_surf_screen_check(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_sid_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdBlitSurfaceToScreen body; - } *cmd; - - cmd = container_of(header, struct vmw_sid_cmd, header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBlitSurfaceToScreen) = + container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.srcImage.sid, NULL); } @@ -994,17 +1005,12 @@ static int vmw_cmd_present_check(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_sid_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdPresent body; - } *cmd; - - - cmd = container_of(header, struct vmw_sid_cmd, header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdPresent) = + container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, &cmd->body.sid, - NULL); + VMW_RES_DIRTY_NONE, user_surface_converter, + &cmd->body.sid, NULL); } /** @@ -1014,11 +1020,10 @@ static int vmw_cmd_present_check(struct vmw_private *dev_priv, * @new_query_bo: The new buffer holding query results. * @sw_context: The software context used for this command submission. * - * This function checks whether @new_query_bo is suitable for holding - * query results, and if another buffer currently is pinned for query - * results. If so, the function prepares the state of @sw_context for - * switching pinned buffers after successful submission of the current - * command batch. + * This function checks whether @new_query_bo is suitable for holding query + * results, and if another buffer currently is pinned for query results. If so, + * the function prepares the state of @sw_context for switching pinned buffers + * after successful submission of the current command batch. */ static int vmw_query_bo_switch_prepare(struct vmw_private *dev_priv, struct vmw_buffer_object *new_query_bo, @@ -1034,7 +1039,7 @@ static int vmw_query_bo_switch_prepare(struct vmw_private *dev_priv, if (unlikely(new_query_bo != sw_context->cur_query_bo)) { if (unlikely(new_query_bo->base.num_pages > 4)) { - DRM_ERROR("Query buffer too large.\n"); + VMW_DEBUG_USER("Query buffer too large.\n"); return -EINVAL; } @@ -1053,13 +1058,11 @@ static int vmw_query_bo_switch_prepare(struct vmw_private *dev_priv, dev_priv->has_mob, false); if (unlikely(ret != 0)) return ret; - } return 0; } - /** * vmw_query_bo_switch_commit - Finalize switching pinned query buffer * @@ -1068,11 +1071,11 @@ static int vmw_query_bo_switch_prepare(struct vmw_private *dev_priv, * * This function will check if we're switching query buffers, and will then, * issue a dummy occlusion query wait used as a query barrier. When the fence - * object following that query wait has signaled, we are sure that all - * preceding queries have finished, and the old query buffer can be unpinned. - * However, since both the new query buffer and the old one are fenced with - * that fence, we can do an asynchronus unpin now, and be sure that the - * old query buffer won't be moved until the fence has signaled. + * object following that query wait has signaled, we are sure that all preceding + * queries have finished, and the old query buffer can be unpinned. However, + * since both the new query buffer and the old one are fenced with that fence, + * we can do an asynchronus unpin now, and be sure that the old query buffer + * won't be moved until the fence has signaled. * * As mentioned above, both the new - and old query buffers need to be fenced * using a sequence emitted *after* calling this function. @@ -1084,7 +1087,6 @@ static void vmw_query_bo_switch_commit(struct vmw_private *dev_priv, * The validate list should still hold references to all * contexts here. */ - if (sw_context->needs_post_query_barrier) { struct vmw_res_cache_entry *ctx_entry = &sw_context->res_cache[vmw_res_context]; @@ -1097,7 +1099,7 @@ static void vmw_query_bo_switch_commit(struct vmw_private *dev_priv, ret = vmw_fifo_emit_dummy_query(dev_priv, ctx->id); if (unlikely(ret != 0)) - DRM_ERROR("Out of fifo space for dummy query.\n"); + VMW_DEBUG_USER("Out of fifo space for dummy query.\n"); } if (dev_priv->pinned_bo != sw_context->cur_query_bo) { @@ -1111,10 +1113,9 @@ static void vmw_query_bo_switch_commit(struct vmw_private *dev_priv, /* * We pin also the dummy_query_bo buffer so that we - * don't need to validate it when emitting - * dummy queries in context destroy paths. + * don't need to validate it when emitting dummy queries + * in context destroy paths. */ - if (!dev_priv->dummy_query_bo_pinned) { vmw_bo_pin_reserved(dev_priv->dummy_query_bo, true); @@ -1131,22 +1132,24 @@ static void vmw_query_bo_switch_commit(struct vmw_private *dev_priv, } /** - * vmw_translate_mob_pointer - Prepare to translate a user-space buffer - * handle to a MOB id. + * vmw_translate_mob_pointer - Prepare to translate a user-space buffer handle + * to a MOB id. * * @dev_priv: Pointer to a device private structure. * @sw_context: The software context used for this command batch validation. * @id: Pointer to the user-space handle to be translated. - * @vmw_bo_p: Points to a location that, on successful return will carry - * a non-reference-counted pointer to the buffer object identified by the + * @vmw_bo_p: Points to a location that, on successful return will carry a + * non-reference-counted pointer to the buffer object identified by the * user-space handle in @id. * * This function saves information needed to translate a user-space buffer * handle to a MOB id. The translation does not take place immediately, but - * during a call to vmw_apply_relocations(). This function builds a relocation - * list and a list of buffers to validate. The former needs to be freed using - * either vmw_apply_relocations() or vmw_free_relocations(). The latter - * needs to be freed using vmw_clear_validations. + * during a call to vmw_apply_relocations(). + * + * This function builds a relocation list and a list of buffers to validate. The + * former needs to be freed using either vmw_apply_relocations() or + * vmw_free_relocations(). The latter needs to be freed using + * vmw_clear_validations. */ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, @@ -1161,7 +1164,7 @@ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv, vmw_validation_preload_bo(sw_context->ctx); vmw_bo = vmw_user_bo_noref_lookup(sw_context->fp->tfile, handle); if (IS_ERR(vmw_bo)) { - DRM_ERROR("Could not find or use MOB buffer.\n"); + VMW_DEBUG_USER("Could not find or use MOB buffer.\n"); return PTR_ERR(vmw_bo); } @@ -1184,19 +1187,20 @@ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv, } /** - * vmw_translate_guest_pointer - Prepare to translate a user-space buffer - * handle to a valid SVGAGuestPtr + * vmw_translate_guest_pointer - Prepare to translate a user-space buffer handle + * to a valid SVGAGuestPtr * * @dev_priv: Pointer to a device private structure. * @sw_context: The software context used for this command batch validation. * @ptr: Pointer to the user-space handle to be translated. - * @vmw_bo_p: Points to a location that, on successful return will carry - * a non-reference-counted pointer to the DMA buffer identified by the - * user-space handle in @id. + * @vmw_bo_p: Points to a location that, on successful return will carry a + * non-reference-counted pointer to the DMA buffer identified by the user-space + * handle in @id. * * This function saves information needed to translate a user-space buffer * handle to a valid SVGAGuestPtr. The translation does not take place * immediately, but during a call to vmw_apply_relocations(). + * * This function builds a relocation list and a list of buffers to validate. * The former needs to be freed using either vmw_apply_relocations() or * vmw_free_relocations(). The latter needs to be freed using @@ -1215,7 +1219,7 @@ static int vmw_translate_guest_ptr(struct vmw_private *dev_priv, vmw_validation_preload_bo(sw_context->ctx); vmw_bo = vmw_user_bo_noref_lookup(sw_context->fp->tfile, handle); if (IS_ERR(vmw_bo)) { - DRM_ERROR("Could not find or use GMR region.\n"); + VMW_DEBUG_USER("Could not find or use GMR region.\n"); return PTR_ERR(vmw_bo); } @@ -1236,10 +1240,8 @@ static int vmw_translate_guest_ptr(struct vmw_private *dev_priv, return 0; } - - /** - * vmw_cmd_dx_define_query - validate a SVGA_3D_CMD_DX_DEFINE_QUERY command. + * vmw_cmd_dx_define_query - validate SVGA_3D_CMD_DX_DEFINE_QUERY command. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context used for this command submission. @@ -1251,67 +1253,52 @@ static int vmw_cmd_dx_define_query(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_dx_define_query_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdDXDefineQuery q; - } *cmd; - - int ret; - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXDefineQuery); + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_resource *cotable_res; + int ret; - - if (ctx_node == NULL) { - DRM_ERROR("DX Context not set for query.\n"); + if (!ctx_node) return -EINVAL; - } - cmd = container_of(header, struct vmw_dx_define_query_cmd, header); + cmd = container_of(header, typeof(*cmd), header); - if (cmd->q.type < SVGA3D_QUERYTYPE_MIN || - cmd->q.type >= SVGA3D_QUERYTYPE_MAX) + if (cmd->body.type < SVGA3D_QUERYTYPE_MIN || + cmd->body.type >= SVGA3D_QUERYTYPE_MAX) return -EINVAL; cotable_res = vmw_context_cotable(ctx_node->ctx, SVGA_COTABLE_DXQUERY); - ret = vmw_cotable_notify(cotable_res, cmd->q.queryId); + ret = vmw_cotable_notify(cotable_res, cmd->body.queryId); return ret; } - - /** - * vmw_cmd_dx_bind_query - validate a SVGA_3D_CMD_DX_BIND_QUERY command. + * vmw_cmd_dx_bind_query - validate SVGA_3D_CMD_DX_BIND_QUERY command. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context used for this command submission. * @header: Pointer to the command header in the command stream. * - * The query bind operation will eventually associate the query ID - * with its backing MOB. In this function, we take the user mode - * MOB ID and use vmw_translate_mob_ptr() to translate it to its - * kernel mode equivalent. + * The query bind operation will eventually associate the query ID with its + * backing MOB. In this function, we take the user mode MOB ID and use + * vmw_translate_mob_ptr() to translate it to its kernel mode equivalent. */ static int vmw_cmd_dx_bind_query(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_dx_bind_query_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdDXBindQuery q; - } *cmd; - + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXBindQuery); struct vmw_buffer_object *vmw_bo; - int ret; - + int ret; - cmd = container_of(header, struct vmw_dx_bind_query_cmd, header); + cmd = container_of(header, typeof(*cmd), header); /* * Look up the buffer pointed to by q.mobid, put it on the relocation * list so its kernel mode MOB ID can be filled in later */ - ret = vmw_translate_mob_ptr(dev_priv, sw_context, &cmd->q.mobid, + ret = vmw_translate_mob_ptr(dev_priv, sw_context, &cmd->body.mobid, &vmw_bo); if (ret != 0) @@ -1322,10 +1309,8 @@ static int vmw_cmd_dx_bind_query(struct vmw_private *dev_priv, return 0; } - - /** - * vmw_cmd_begin_gb_query - validate a SVGA_3D_CMD_BEGIN_GB_QUERY command. + * vmw_cmd_begin_gb_query - validate SVGA_3D_CMD_BEGIN_GB_QUERY command. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context used for this command submission. @@ -1335,21 +1320,16 @@ static int vmw_cmd_begin_gb_query(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_begin_gb_query_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdBeginGBQuery q; - } *cmd; - - cmd = container_of(header, struct vmw_begin_gb_query_cmd, - header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBeginGBQuery) = + container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, &cmd->q.cid, - NULL); + VMW_RES_DIRTY_SET, user_context_converter, + &cmd->body.cid, NULL); } /** - * vmw_cmd_begin_query - validate a SVGA_3D_CMD_BEGIN_QUERY command. + * vmw_cmd_begin_query - validate SVGA_3D_CMD_BEGIN_QUERY command. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context used for this command submission. @@ -1359,38 +1339,30 @@ static int vmw_cmd_begin_query(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_begin_query_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdBeginQuery q; - } *cmd; - - cmd = container_of(header, struct vmw_begin_query_cmd, - header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBeginQuery) = + container_of(header, typeof(*cmd), header); if (unlikely(dev_priv->has_mob)) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdBeginGBQuery q; - } gb_cmd; + VMW_DECLARE_CMD_VAR(gb_cmd, SVGA3dCmdBeginGBQuery); BUG_ON(sizeof(gb_cmd) != sizeof(*cmd)); gb_cmd.header.id = SVGA_3D_CMD_BEGIN_GB_QUERY; gb_cmd.header.size = cmd->header.size; - gb_cmd.q.cid = cmd->q.cid; - gb_cmd.q.type = cmd->q.type; + gb_cmd.body.cid = cmd->body.cid; + gb_cmd.body.type = cmd->body.type; memcpy(cmd, &gb_cmd, sizeof(*cmd)); return vmw_cmd_begin_gb_query(dev_priv, sw_context, header); } return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, &cmd->q.cid, - NULL); + VMW_RES_DIRTY_SET, user_context_converter, + &cmd->body.cid, NULL); } /** - * vmw_cmd_end_gb_query - validate a SVGA_3D_CMD_END_GB_QUERY command. + * vmw_cmd_end_gb_query - validate SVGA_3D_CMD_END_GB_QUERY command. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context used for this command submission. @@ -1401,19 +1373,15 @@ static int vmw_cmd_end_gb_query(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { struct vmw_buffer_object *vmw_bo; - struct vmw_query_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdEndGBQuery q; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdEndGBQuery); int ret; - cmd = container_of(header, struct vmw_query_cmd, header); + cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_cid_check(dev_priv, sw_context, header); if (unlikely(ret != 0)) return ret; - ret = vmw_translate_mob_ptr(dev_priv, sw_context, - &cmd->q.mobid, + ret = vmw_translate_mob_ptr(dev_priv, sw_context, &cmd->body.mobid, &vmw_bo); if (unlikely(ret != 0)) return ret; @@ -1424,7 +1392,7 @@ static int vmw_cmd_end_gb_query(struct vmw_private *dev_priv, } /** - * vmw_cmd_end_query - validate a SVGA_3D_CMD_END_QUERY command. + * vmw_cmd_end_query - validate SVGA_3D_CMD_END_QUERY command. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context used for this command submission. @@ -1435,27 +1403,21 @@ static int vmw_cmd_end_query(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { struct vmw_buffer_object *vmw_bo; - struct vmw_query_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdEndQuery q; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdEndQuery); int ret; - cmd = container_of(header, struct vmw_query_cmd, header); + cmd = container_of(header, typeof(*cmd), header); if (dev_priv->has_mob) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdEndGBQuery q; - } gb_cmd; + VMW_DECLARE_CMD_VAR(gb_cmd, SVGA3dCmdEndGBQuery); BUG_ON(sizeof(gb_cmd) != sizeof(*cmd)); gb_cmd.header.id = SVGA_3D_CMD_END_GB_QUERY; gb_cmd.header.size = cmd->header.size; - gb_cmd.q.cid = cmd->q.cid; - gb_cmd.q.type = cmd->q.type; - gb_cmd.q.mobid = cmd->q.guestResult.gmrId; - gb_cmd.q.offset = cmd->q.guestResult.offset; + gb_cmd.body.cid = cmd->body.cid; + gb_cmd.body.type = cmd->body.type; + gb_cmd.body.mobid = cmd->body.guestResult.gmrId; + gb_cmd.body.offset = cmd->body.guestResult.offset; memcpy(cmd, &gb_cmd, sizeof(*cmd)); return vmw_cmd_end_gb_query(dev_priv, sw_context, header); @@ -1466,8 +1428,7 @@ static int vmw_cmd_end_query(struct vmw_private *dev_priv, return ret; ret = vmw_translate_guest_ptr(dev_priv, sw_context, - &cmd->q.guestResult, - &vmw_bo); + &cmd->body.guestResult, &vmw_bo); if (unlikely(ret != 0)) return ret; @@ -1477,7 +1438,7 @@ static int vmw_cmd_end_query(struct vmw_private *dev_priv, } /** - * vmw_cmd_wait_gb_query - validate a SVGA_3D_CMD_WAIT_GB_QUERY command. + * vmw_cmd_wait_gb_query - validate SVGA_3D_CMD_WAIT_GB_QUERY command. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context used for this command submission. @@ -1488,19 +1449,15 @@ static int vmw_cmd_wait_gb_query(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { struct vmw_buffer_object *vmw_bo; - struct vmw_query_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdWaitForGBQuery q; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdWaitForGBQuery); int ret; - cmd = container_of(header, struct vmw_query_cmd, header); + cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_cid_check(dev_priv, sw_context, header); if (unlikely(ret != 0)) return ret; - ret = vmw_translate_mob_ptr(dev_priv, sw_context, - &cmd->q.mobid, + ret = vmw_translate_mob_ptr(dev_priv, sw_context, &cmd->body.mobid, &vmw_bo); if (unlikely(ret != 0)) return ret; @@ -1509,7 +1466,7 @@ static int vmw_cmd_wait_gb_query(struct vmw_private *dev_priv, } /** - * vmw_cmd_wait_query - validate a SVGA_3D_CMD_WAIT_QUERY command. + * vmw_cmd_wait_query - validate SVGA_3D_CMD_WAIT_QUERY command. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context used for this command submission. @@ -1520,27 +1477,21 @@ static int vmw_cmd_wait_query(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { struct vmw_buffer_object *vmw_bo; - struct vmw_query_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdWaitForQuery q; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdWaitForQuery); int ret; - cmd = container_of(header, struct vmw_query_cmd, header); + cmd = container_of(header, typeof(*cmd), header); if (dev_priv->has_mob) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdWaitForGBQuery q; - } gb_cmd; + VMW_DECLARE_CMD_VAR(gb_cmd, SVGA3dCmdWaitForGBQuery); BUG_ON(sizeof(gb_cmd) != sizeof(*cmd)); gb_cmd.header.id = SVGA_3D_CMD_WAIT_FOR_GB_QUERY; gb_cmd.header.size = cmd->header.size; - gb_cmd.q.cid = cmd->q.cid; - gb_cmd.q.type = cmd->q.type; - gb_cmd.q.mobid = cmd->q.guestResult.gmrId; - gb_cmd.q.offset = cmd->q.guestResult.offset; + gb_cmd.body.cid = cmd->body.cid; + gb_cmd.body.type = cmd->body.type; + gb_cmd.body.mobid = cmd->body.guestResult.gmrId; + gb_cmd.body.offset = cmd->body.guestResult.offset; memcpy(cmd, &gb_cmd, sizeof(*cmd)); return vmw_cmd_wait_gb_query(dev_priv, sw_context, header); @@ -1551,8 +1502,7 @@ static int vmw_cmd_wait_query(struct vmw_private *dev_priv, return ret; ret = vmw_translate_guest_ptr(dev_priv, sw_context, - &cmd->q.guestResult, - &vmw_bo); + &cmd->body.guestResult, &vmw_bo); if (unlikely(ret != 0)) return ret; @@ -1565,54 +1515,52 @@ static int vmw_cmd_dma(struct vmw_private *dev_priv, { struct vmw_buffer_object *vmw_bo = NULL; struct vmw_surface *srf = NULL; - struct vmw_dma_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdSurfaceDMA dma; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSurfaceDMA); int ret; SVGA3dCmdSurfaceDMASuffix *suffix; uint32_t bo_size; + bool dirty; - cmd = container_of(header, struct vmw_dma_cmd, header); - suffix = (SVGA3dCmdSurfaceDMASuffix *)((unsigned long) &cmd->dma + + cmd = container_of(header, typeof(*cmd), header); + suffix = (SVGA3dCmdSurfaceDMASuffix *)((unsigned long) &cmd->body + header->size - sizeof(*suffix)); /* Make sure device and verifier stays in sync. */ if (unlikely(suffix->suffixSize != sizeof(*suffix))) { - DRM_ERROR("Invalid DMA suffix size.\n"); + VMW_DEBUG_USER("Invalid DMA suffix size.\n"); return -EINVAL; } ret = vmw_translate_guest_ptr(dev_priv, sw_context, - &cmd->dma.guest.ptr, - &vmw_bo); + &cmd->body.guest.ptr, &vmw_bo); if (unlikely(ret != 0)) return ret; /* Make sure DMA doesn't cross BO boundaries. */ bo_size = vmw_bo->base.num_pages * PAGE_SIZE; - if (unlikely(cmd->dma.guest.ptr.offset > bo_size)) { - DRM_ERROR("Invalid DMA offset.\n"); + if (unlikely(cmd->body.guest.ptr.offset > bo_size)) { + VMW_DEBUG_USER("Invalid DMA offset.\n"); return -EINVAL; } - bo_size -= cmd->dma.guest.ptr.offset; + bo_size -= cmd->body.guest.ptr.offset; if (unlikely(suffix->maximumOffset > bo_size)) suffix->maximumOffset = bo_size; + dirty = (cmd->body.transfer == SVGA3D_WRITE_HOST_VRAM) ? + VMW_RES_DIRTY_SET : 0; ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, &cmd->dma.host.sid, - NULL); + dirty, user_surface_converter, + &cmd->body.host.sid, NULL); if (unlikely(ret != 0)) { if (unlikely(ret != -ERESTARTSYS)) - DRM_ERROR("could not find surface for DMA.\n"); + VMW_DEBUG_USER("could not find surface for DMA.\n"); return ret; } srf = vmw_res_to_srf(sw_context->res_cache[vmw_res_surface].res); - vmw_kms_cursor_snoop(srf, sw_context->fp->tfile, &vmw_bo->base, - header); + vmw_kms_cursor_snoop(srf, sw_context->fp->tfile, &vmw_bo->base, header); return 0; } @@ -1621,10 +1569,7 @@ static int vmw_cmd_draw(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_draw_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdDrawPrimitives body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDrawPrimitives); SVGA3dVertexDecl *decl = (SVGA3dVertexDecl *)( (unsigned long)header + sizeof(*cmd)); SVGA3dPrimitiveRange *range; @@ -1636,16 +1581,17 @@ static int vmw_cmd_draw(struct vmw_private *dev_priv, if (unlikely(ret != 0)) return ret; - cmd = container_of(header, struct vmw_draw_cmd, header); + cmd = container_of(header, typeof(*cmd), header); maxnum = (header->size - sizeof(cmd->body)) / sizeof(*decl); if (unlikely(cmd->body.numVertexDecls > maxnum)) { - DRM_ERROR("Illegal number of vertex declarations.\n"); + VMW_DEBUG_USER("Illegal number of vertex declarations.\n"); return -EINVAL; } for (i = 0; i < cmd->body.numVertexDecls; ++i, ++decl) { ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, &decl->array.surfaceId, NULL); if (unlikely(ret != 0)) @@ -1655,13 +1601,14 @@ static int vmw_cmd_draw(struct vmw_private *dev_priv, maxnum = (header->size - sizeof(cmd->body) - cmd->body.numVertexDecls * sizeof(*decl)) / sizeof(*range); if (unlikely(cmd->body.numRanges > maxnum)) { - DRM_ERROR("Illegal number of index ranges.\n"); + VMW_DEBUG_USER("Illegal number of index ranges.\n"); return -EINVAL; } range = (SVGA3dPrimitiveRange *) decl; for (i = 0; i < cmd->body.numRanges; ++i, ++range) { ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, &range->indexArray.surfaceId, NULL); if (unlikely(ret != 0)) @@ -1670,30 +1617,24 @@ static int vmw_cmd_draw(struct vmw_private *dev_priv, return 0; } - static int vmw_cmd_tex_state(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_tex_state_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdSetTextureState state; - } *cmd; - + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSetTextureState); SVGA3dTextureState *last_state = (SVGA3dTextureState *) ((unsigned long) header + header->size + sizeof(header)); SVGA3dTextureState *cur_state = (SVGA3dTextureState *) - ((unsigned long) header + sizeof(struct vmw_tex_state_cmd)); + ((unsigned long) header + sizeof(*cmd)); struct vmw_resource *ctx; struct vmw_resource *res; int ret; - cmd = container_of(header, struct vmw_tex_state_cmd, - header); + cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, &cmd->state.cid, - &ctx); + VMW_RES_DIRTY_SET, user_context_converter, + &cmd->body.cid, &ctx); if (unlikely(ret != 0)) return ret; @@ -1702,12 +1643,13 @@ static int vmw_cmd_tex_state(struct vmw_private *dev_priv, continue; if (cur_state->stage >= SVGA3D_NUM_TEXTURE_UNITS) { - DRM_ERROR("Illegal texture/sampler unit %u.\n", - (unsigned) cur_state->stage); + VMW_DEBUG_USER("Illegal texture/sampler unit %u.\n", + (unsigned int) cur_state->stage); return -EINVAL; } ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, &cur_state->value, &res); if (unlikely(ret != 0)) @@ -1744,12 +1686,10 @@ static int vmw_cmd_check_define_gmrfb(struct vmw_private *dev_priv, SVGAFifoCmdDefineGMRFB body; } *cmd = buf; - return vmw_translate_guest_ptr(dev_priv, sw_context, - &cmd->body.ptr, + return vmw_translate_guest_ptr(dev_priv, sw_context, &cmd->body.ptr, &vmw_bo); } - /** * vmw_cmd_res_switch_backup - Utility function to handle backup buffer * switching @@ -1761,14 +1701,13 @@ static int vmw_cmd_check_define_gmrfb(struct vmw_private *dev_priv, * stream. * @backup_offset: Offset of backup into MOB. * - * This function prepares for registering a switch of backup buffers - * in the resource metadata just prior to unreserving. It's basically a wrapper - * around vmw_cmd_res_switch_backup with a different interface. + * This function prepares for registering a switch of backup buffers in the + * resource metadata just prior to unreserving. It's basically a wrapper around + * vmw_cmd_res_switch_backup with a different interface. */ static int vmw_cmd_res_switch_backup(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, - struct vmw_resource *res, - uint32_t *buf_id, + struct vmw_resource *res, uint32_t *buf_id, unsigned long backup_offset) { struct vmw_buffer_object *vbo; @@ -1788,7 +1727,6 @@ static int vmw_cmd_res_switch_backup(struct vmw_private *dev_priv, return 0; } - /** * vmw_cmd_switch_backup - Utility function to handle backup buffer switching * @@ -1801,34 +1739,31 @@ static int vmw_cmd_res_switch_backup(struct vmw_private *dev_priv, * stream. * @backup_offset: Offset of backup into MOB. * - * This function prepares for registering a switch of backup buffers - * in the resource metadata just prior to unreserving. It's basically a wrapper - * around vmw_cmd_res_switch_backup with a different interface. + * This function prepares for registering a switch of backup buffers in the + * resource metadata just prior to unreserving. It's basically a wrapper around + * vmw_cmd_res_switch_backup with a different interface. */ static int vmw_cmd_switch_backup(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, enum vmw_res_type res_type, const struct vmw_user_resource_conv - *converter, - uint32_t *res_id, - uint32_t *buf_id, + *converter, uint32_t *res_id, uint32_t *buf_id, unsigned long backup_offset) { struct vmw_resource *res; int ret; ret = vmw_cmd_res_check(dev_priv, sw_context, res_type, - converter, res_id, &res); + VMW_RES_DIRTY_NONE, converter, res_id, &res); if (ret) return ret; - return vmw_cmd_res_switch_backup(dev_priv, sw_context, res, - buf_id, backup_offset); + return vmw_cmd_res_switch_backup(dev_priv, sw_context, res, buf_id, + backup_offset); } /** - * vmw_cmd_bind_gb_surface - Validate an SVGA_3D_CMD_BIND_GB_SURFACE - * command + * vmw_cmd_bind_gb_surface - Validate SVGA_3D_CMD_BIND_GB_SURFACE command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -1838,22 +1773,16 @@ static int vmw_cmd_bind_gb_surface(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_bind_gb_surface_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdBindGBSurface body; - } *cmd; - - cmd = container_of(header, struct vmw_bind_gb_surface_cmd, header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBindGBSurface) = + container_of(header, typeof(*cmd), header); return vmw_cmd_switch_backup(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, - &cmd->body.sid, &cmd->body.mobid, - 0); + user_surface_converter, &cmd->body.sid, + &cmd->body.mobid, 0); } /** - * vmw_cmd_update_gb_image - Validate an SVGA_3D_CMD_UPDATE_GB_IMAGE - * command + * vmw_cmd_update_gb_image - Validate SVGA_3D_CMD_UPDATE_GB_IMAGE command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -1863,21 +1792,16 @@ static int vmw_cmd_update_gb_image(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_gb_surface_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdUpdateGBImage body; - } *cmd; - - cmd = container_of(header, struct vmw_gb_surface_cmd, header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdUpdateGBImage) = + container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.image.sid, NULL); } /** - * vmw_cmd_update_gb_surface - Validate an SVGA_3D_CMD_UPDATE_GB_SURFACE - * command + * vmw_cmd_update_gb_surface - Validate SVGA_3D_CMD_UPDATE_GB_SURFACE command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -1887,21 +1811,16 @@ static int vmw_cmd_update_gb_surface(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_gb_surface_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdUpdateGBSurface body; - } *cmd; - - cmd = container_of(header, struct vmw_gb_surface_cmd, header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdUpdateGBSurface) = + container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_CLEAR, user_surface_converter, &cmd->body.sid, NULL); } /** - * vmw_cmd_readback_gb_image - Validate an SVGA_3D_CMD_READBACK_GB_IMAGE - * command + * vmw_cmd_readback_gb_image - Validate SVGA_3D_CMD_READBACK_GB_IMAGE command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -1911,20 +1830,16 @@ static int vmw_cmd_readback_gb_image(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_gb_surface_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdReadbackGBImage body; - } *cmd; - - cmd = container_of(header, struct vmw_gb_surface_cmd, header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdReadbackGBImage) = + container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.image.sid, NULL); } /** - * vmw_cmd_readback_gb_surface - Validate an SVGA_3D_CMD_READBACK_GB_SURFACE + * vmw_cmd_readback_gb_surface - Validate SVGA_3D_CMD_READBACK_GB_SURFACE * command * * @dev_priv: Pointer to a device private struct. @@ -1935,20 +1850,16 @@ static int vmw_cmd_readback_gb_surface(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_gb_surface_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdReadbackGBSurface body; - } *cmd; - - cmd = container_of(header, struct vmw_gb_surface_cmd, header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdReadbackGBSurface) = + container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_CLEAR, user_surface_converter, &cmd->body.sid, NULL); } /** - * vmw_cmd_invalidate_gb_image - Validate an SVGA_3D_CMD_INVALIDATE_GB_IMAGE + * vmw_cmd_invalidate_gb_image - Validate SVGA_3D_CMD_INVALIDATE_GB_IMAGE * command * * @dev_priv: Pointer to a device private struct. @@ -1959,21 +1870,17 @@ static int vmw_cmd_invalidate_gb_image(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_gb_surface_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdInvalidateGBImage body; - } *cmd; - - cmd = container_of(header, struct vmw_gb_surface_cmd, header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdInvalidateGBImage) = + container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.image.sid, NULL); } /** - * vmw_cmd_invalidate_gb_surface - Validate an - * SVGA_3D_CMD_INVALIDATE_GB_SURFACE command + * vmw_cmd_invalidate_gb_surface - Validate SVGA_3D_CMD_INVALIDATE_GB_SURFACE + * command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -1983,22 +1890,16 @@ static int vmw_cmd_invalidate_gb_surface(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_gb_surface_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdInvalidateGBSurface body; - } *cmd; - - cmd = container_of(header, struct vmw_gb_surface_cmd, header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdInvalidateGBSurface) = + container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_CLEAR, user_surface_converter, &cmd->body.sid, NULL); } - /** - * vmw_cmd_shader_define - Validate an SVGA_3D_CMD_SHADER_DEFINE - * command + * vmw_cmd_shader_define - Validate SVGA_3D_CMD_SHADER_DEFINE command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2008,20 +1909,16 @@ static int vmw_cmd_shader_define(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_shader_define_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdDefineShader body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDefineShader); int ret; size_t size; struct vmw_resource *ctx; - cmd = container_of(header, struct vmw_shader_define_cmd, - header); + cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, &cmd->body.cid, - &ctx); + VMW_RES_DIRTY_SET, user_context_converter, + &cmd->body.cid, &ctx); if (unlikely(ret != 0)) return ret; @@ -2029,24 +1926,20 @@ static int vmw_cmd_shader_define(struct vmw_private *dev_priv, return 0; size = cmd->header.size - sizeof(cmd->body); - ret = vmw_compat_shader_add(dev_priv, - vmw_context_res_man(ctx), - cmd->body.shid, cmd + 1, - cmd->body.type, size, - &sw_context->staged_cmd_res); + ret = vmw_compat_shader_add(dev_priv, vmw_context_res_man(ctx), + cmd->body.shid, cmd + 1, cmd->body.type, + size, &sw_context->staged_cmd_res); if (unlikely(ret != 0)) return ret; - return vmw_resource_relocation_add(sw_context, - NULL, + return vmw_resource_relocation_add(sw_context, NULL, vmw_ptr_diff(sw_context->buf_start, &cmd->header.id), vmw_res_rel_nop); } /** - * vmw_cmd_shader_destroy - Validate an SVGA_3D_CMD_SHADER_DESTROY - * command + * vmw_cmd_shader_destroy - Validate SVGA_3D_CMD_SHADER_DESTROY command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2056,42 +1949,34 @@ static int vmw_cmd_shader_destroy(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_shader_destroy_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdDestroyShader body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDestroyShader); int ret; struct vmw_resource *ctx; - cmd = container_of(header, struct vmw_shader_destroy_cmd, - header); + cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, &cmd->body.cid, - &ctx); + VMW_RES_DIRTY_SET, user_context_converter, + &cmd->body.cid, &ctx); if (unlikely(ret != 0)) return ret; if (unlikely(!dev_priv->has_mob)) return 0; - ret = vmw_shader_remove(vmw_context_res_man(ctx), - cmd->body.shid, - cmd->body.type, - &sw_context->staged_cmd_res); + ret = vmw_shader_remove(vmw_context_res_man(ctx), cmd->body.shid, + cmd->body.type, &sw_context->staged_cmd_res); if (unlikely(ret != 0)) return ret; - return vmw_resource_relocation_add(sw_context, - NULL, + return vmw_resource_relocation_add(sw_context, NULL, vmw_ptr_diff(sw_context->buf_start, &cmd->header.id), vmw_res_rel_nop); } /** - * vmw_cmd_set_shader - Validate an SVGA_3D_CMD_SET_SHADER - * command + * vmw_cmd_set_shader - Validate SVGA_3D_CMD_SET_SHADER command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2101,27 +1986,23 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_set_shader_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdSetShader body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSetShader); struct vmw_ctx_bindinfo_shader binding; struct vmw_resource *ctx, *res = NULL; struct vmw_ctx_validation_info *ctx_info; int ret; - cmd = container_of(header, struct vmw_set_shader_cmd, - header); + cmd = container_of(header, typeof(*cmd), header); if (cmd->body.type >= SVGA3D_SHADERTYPE_PREDX_MAX) { - DRM_ERROR("Illegal shader type %u.\n", - (unsigned) cmd->body.type); + VMW_DEBUG_USER("Illegal shader type %u.\n", + (unsigned int) cmd->body.type); return -EINVAL; } ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, &cmd->body.cid, - &ctx); + VMW_RES_DIRTY_SET, user_context_converter, + &cmd->body.cid, &ctx); if (unlikely(ret != 0)) return ret; @@ -2130,21 +2011,20 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv, if (cmd->body.shid != SVGA3D_INVALID_ID) { res = vmw_shader_lookup(vmw_context_res_man(ctx), - cmd->body.shid, - cmd->body.type); - + cmd->body.shid, cmd->body.type); if (!IS_ERR(res)) { - ret = vmw_execbuf_res_noctx_val_add(sw_context, res); + ret = vmw_execbuf_res_noctx_val_add(sw_context, res, + VMW_RES_DIRTY_NONE); if (unlikely(ret != 0)) return ret; } } if (IS_ERR_OR_NULL(res)) { - ret = vmw_cmd_res_check(dev_priv, sw_context, - vmw_res_shader, - user_shader_converter, - &cmd->body.shid, &res); + ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_shader, + VMW_RES_DIRTY_NONE, + user_shader_converter, &cmd->body.shid, + &res); if (unlikely(ret != 0)) return ret; } @@ -2157,14 +2037,13 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv, binding.bi.res = res; binding.bi.bt = vmw_ctx_binding_shader; binding.shader_slot = cmd->body.type - SVGA3D_SHADERTYPE_MIN; - vmw_binding_add(ctx_info->staged, &binding.bi, - binding.shader_slot, 0); + vmw_binding_add(ctx_info->staged, &binding.bi, binding.shader_slot, 0); + return 0; } /** - * vmw_cmd_set_shader_const - Validate an SVGA_3D_CMD_SET_SHADER_CONST - * command + * vmw_cmd_set_shader_const - Validate SVGA_3D_CMD_SET_SHADER_CONST command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2174,18 +2053,14 @@ static int vmw_cmd_set_shader_const(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_set_shader_const_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdSetShaderConst body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdSetShaderConst); int ret; - cmd = container_of(header, struct vmw_set_shader_const_cmd, - header); + cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, &cmd->body.cid, - NULL); + VMW_RES_DIRTY_SET, user_context_converter, + &cmd->body.cid, NULL); if (unlikely(ret != 0)) return ret; @@ -2196,8 +2071,7 @@ static int vmw_cmd_set_shader_const(struct vmw_private *dev_priv, } /** - * vmw_cmd_bind_gb_shader - Validate an SVGA_3D_CMD_BIND_GB_SHADER - * command + * vmw_cmd_bind_gb_shader - Validate SVGA_3D_CMD_BIND_GB_SHADER command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2207,22 +2081,16 @@ static int vmw_cmd_bind_gb_shader(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_bind_gb_shader_cmd { - SVGA3dCmdHeader header; - SVGA3dCmdBindGBShader body; - } *cmd; - - cmd = container_of(header, struct vmw_bind_gb_shader_cmd, - header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdBindGBShader) = + container_of(header, typeof(*cmd), header); return vmw_cmd_switch_backup(dev_priv, sw_context, vmw_res_shader, - user_shader_converter, - &cmd->body.shid, &cmd->body.mobid, - cmd->body.offsetInBytes); + user_shader_converter, &cmd->body.shid, + &cmd->body.mobid, cmd->body.offsetInBytes); } /** - * vmw_cmd_dx_set_single_constant_buffer - Validate an + * vmw_cmd_dx_set_single_constant_buffer - Validate * SVGA_3D_CMD_DX_SET_SINGLE_CONSTANT_BUFFER command. * * @dev_priv: Pointer to a device private struct. @@ -2234,23 +2102,18 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXSetSingleConstantBuffer body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetSingleConstantBuffer); struct vmw_resource *res = NULL; - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_cb binding; int ret; - if (unlikely(ctx_node == NULL)) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.sid, &res); if (unlikely(ret != 0)) return ret; @@ -2265,21 +2128,21 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv, if (binding.shader_slot >= SVGA3D_NUM_SHADERTYPE_DX10 || binding.slot >= SVGA3D_DX_MAX_CONSTBUFFERS) { - DRM_ERROR("Illegal const buffer shader %u slot %u.\n", - (unsigned) cmd->body.type, - (unsigned) binding.slot); + VMW_DEBUG_USER("Illegal const buffer shader %u slot %u.\n", + (unsigned int) cmd->body.type, + (unsigned int) binding.slot); return -EINVAL; } - vmw_binding_add(ctx_node->staged, &binding.bi, - binding.shader_slot, binding.slot); + vmw_binding_add(ctx_node->staged, &binding.bi, binding.shader_slot, + binding.slot); return 0; } /** - * vmw_cmd_dx_set_shader_res - Validate an - * SVGA_3D_CMD_DX_SET_SHADER_RESOURCES command + * vmw_cmd_dx_set_shader_res - Validate SVGA_3D_CMD_DX_SET_SHADER_RESOURCES + * command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2289,17 +2152,15 @@ static int vmw_cmd_dx_set_shader_res(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXSetShaderResources body; - } *cmd = container_of(header, typeof(*cmd), header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShaderResources) = + container_of(header, typeof(*cmd), header); u32 num_sr_view = (cmd->header.size - sizeof(cmd->body)) / sizeof(SVGA3dShaderResourceViewId); if ((u64) cmd->body.startView + (u64) num_sr_view > (u64) SVGA3D_DX_MAX_SRVIEWS || cmd->body.type >= SVGA3D_SHADERTYPE_DX10_MAX) { - DRM_ERROR("Invalid shader binding.\n"); + VMW_DEBUG_USER("Invalid shader binding.\n"); return -EINVAL; } @@ -2311,8 +2172,7 @@ static int vmw_cmd_dx_set_shader_res(struct vmw_private *dev_priv, } /** - * vmw_cmd_dx_set_shader - Validate an SVGA_3D_CMD_DX_SET_SHADER - * command + * vmw_cmd_dx_set_shader - Validate SVGA_3D_CMD_DX_SET_SHADER command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2322,36 +2182,32 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXSetShader body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShader); struct vmw_resource *res = NULL; - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_shader binding; int ret = 0; - if (unlikely(ctx_node == NULL)) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } cmd = container_of(header, typeof(*cmd), header); if (cmd->body.type >= SVGA3D_SHADERTYPE_DX10_MAX) { - DRM_ERROR("Illegal shader type %u.\n", - (unsigned) cmd->body.type); + VMW_DEBUG_USER("Illegal shader type %u.\n", + (unsigned int) cmd->body.type); return -EINVAL; } if (cmd->body.shaderId != SVGA3D_INVALID_ID) { res = vmw_shader_lookup(sw_context->man, cmd->body.shaderId, 0); if (IS_ERR(res)) { - DRM_ERROR("Could not find shader for binding.\n"); + VMW_DEBUG_USER("Could not find shader for binding.\n"); return PTR_ERR(res); } - ret = vmw_execbuf_res_noctx_val_add(sw_context, res); + ret = vmw_execbuf_res_noctx_val_add(sw_context, res, + VMW_RES_DIRTY_NONE); if (ret) return ret; } @@ -2361,15 +2217,14 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv, binding.bi.bt = vmw_ctx_binding_dx_shader; binding.shader_slot = cmd->body.type - SVGA3D_SHADERTYPE_MIN; - vmw_binding_add(ctx_node->staged, &binding.bi, - binding.shader_slot, 0); + vmw_binding_add(ctx_node->staged, &binding.bi, binding.shader_slot, 0); return 0; } /** - * vmw_cmd_dx_set_vertex_buffers - Validates an - * SVGA_3D_CMD_DX_SET_VERTEX_BUFFERS command + * vmw_cmd_dx_set_vertex_buffers - Validates SVGA_3D_CMD_DX_SET_VERTEX_BUFFERS + * command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2379,7 +2234,7 @@ static int vmw_cmd_dx_set_vertex_buffers(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_vb binding; struct vmw_resource *res; struct { @@ -2389,22 +2244,21 @@ static int vmw_cmd_dx_set_vertex_buffers(struct vmw_private *dev_priv, } *cmd; int i, ret, num; - if (unlikely(ctx_node == NULL)) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } cmd = container_of(header, typeof(*cmd), header); num = (cmd->header.size - sizeof(cmd->body)) / sizeof(SVGA3dVertexBuffer); if ((u64)num + (u64)cmd->body.startBuffer > (u64)SVGA3D_DX_MAX_VERTEXBUFFERS) { - DRM_ERROR("Invalid number of vertex buffers.\n"); + VMW_DEBUG_USER("Invalid number of vertex buffers.\n"); return -EINVAL; } for (i = 0; i < num; i++) { ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->buf[i].sid, &res); if (unlikely(ret != 0)) @@ -2417,15 +2271,14 @@ static int vmw_cmd_dx_set_vertex_buffers(struct vmw_private *dev_priv, binding.stride = cmd->buf[i].stride; binding.slot = i + cmd->body.startBuffer; - vmw_binding_add(ctx_node->staged, &binding.bi, - 0, binding.slot); + vmw_binding_add(ctx_node->staged, &binding.bi, 0, binding.slot); } return 0; } /** - * vmw_cmd_dx_ia_set_vertex_buffers - Validate an + * vmw_cmd_dx_ia_set_vertex_buffers - Validate * SVGA_3D_CMD_DX_IA_SET_INDEX_BUFFER command. * * @dev_priv: Pointer to a device private struct. @@ -2436,23 +2289,18 @@ static int vmw_cmd_dx_set_index_buffer(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_ib binding; struct vmw_resource *res; - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXSetIndexBuffer body; - } *cmd; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetIndexBuffer); int ret; - if (unlikely(ctx_node == NULL)) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.sid, &res); if (unlikely(ret != 0)) return ret; @@ -2469,8 +2317,8 @@ static int vmw_cmd_dx_set_index_buffer(struct vmw_private *dev_priv, } /** - * vmw_cmd_dx_set_rendertarget - Validate an - * SVGA_3D_CMD_DX_SET_RENDERTARGETS command + * vmw_cmd_dx_set_rendertarget - Validate SVGA_3D_CMD_DX_SET_RENDERTARGETS + * command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2480,32 +2328,29 @@ static int vmw_cmd_dx_set_rendertargets(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXSetRenderTargets body; - } *cmd = container_of(header, typeof(*cmd), header); - int ret; + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetRenderTargets) = + container_of(header, typeof(*cmd), header); u32 num_rt_view = (cmd->header.size - sizeof(cmd->body)) / sizeof(SVGA3dRenderTargetViewId); + int ret; if (num_rt_view > SVGA3D_MAX_SIMULTANEOUS_RENDER_TARGETS) { - DRM_ERROR("Invalid DX Rendertarget binding.\n"); + VMW_DEBUG_USER("Invalid DX Rendertarget binding.\n"); return -EINVAL; } - ret = vmw_view_bindings_add(sw_context, vmw_view_ds, - vmw_ctx_binding_ds, 0, - &cmd->body.depthStencilViewId, 1, 0); + ret = vmw_view_bindings_add(sw_context, vmw_view_ds, vmw_ctx_binding_ds, + 0, &cmd->body.depthStencilViewId, 1, 0); if (ret) return ret; return vmw_view_bindings_add(sw_context, vmw_view_rt, - vmw_ctx_binding_dx_rt, 0, - (void *)&cmd[1], num_rt_view, 0); + vmw_ctx_binding_dx_rt, 0, (void *)&cmd[1], + num_rt_view, 0); } /** - * vmw_cmd_dx_clear_rendertarget_view - Validate an + * vmw_cmd_dx_clear_rendertarget_view - Validate * SVGA_3D_CMD_DX_CLEAR_RENDERTARGET_VIEW command * * @dev_priv: Pointer to a device private struct. @@ -2516,17 +2361,15 @@ static int vmw_cmd_dx_clear_rendertarget_view(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXClearRenderTargetView body; - } *cmd = container_of(header, typeof(*cmd), header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXClearRenderTargetView) = + container_of(header, typeof(*cmd), header); return PTR_RET(vmw_view_id_val_add(sw_context, vmw_view_rt, cmd->body.renderTargetViewId)); } /** - * vmw_cmd_dx_clear_rendertarget_view - Validate an + * vmw_cmd_dx_clear_rendertarget_view - Validate * SVGA_3D_CMD_DX_CLEAR_DEPTHSTENCIL_VIEW command * * @dev_priv: Pointer to a device private struct. @@ -2537,10 +2380,8 @@ static int vmw_cmd_dx_clear_depthstencil_view(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXClearDepthStencilView body; - } *cmd = container_of(header, typeof(*cmd), header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXClearDepthStencilView) = + container_of(header, typeof(*cmd), header); return PTR_RET(vmw_view_id_val_add(sw_context, vmw_view_ds, cmd->body.depthStencilViewId)); @@ -2550,14 +2391,14 @@ static int vmw_cmd_dx_view_define(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_resource *srf; struct vmw_resource *res; enum vmw_view_type view_type; int ret; /* - * This is based on the fact that all affected define commands have - * the same initial command body layout. + * This is based on the fact that all affected define commands have the + * same initial command body layout. */ struct { SVGA3dCmdHeader header; @@ -2565,17 +2406,16 @@ static int vmw_cmd_dx_view_define(struct vmw_private *dev_priv, uint32 sid; } *cmd; - if (unlikely(ctx_node == NULL)) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } view_type = vmw_view_cmd_to_type(header->id); if (view_type == vmw_view_max) return -EINVAL; + cmd = container_of(header, typeof(*cmd), header); ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->sid, &srf); if (unlikely(ret != 0)) return ret; @@ -2585,19 +2425,14 @@ static int vmw_cmd_dx_view_define(struct vmw_private *dev_priv, if (unlikely(ret != 0)) return ret; - return vmw_view_add(sw_context->man, - ctx_node->ctx, - srf, - view_type, - cmd->defined_id, - header, + return vmw_view_add(sw_context->man, ctx_node->ctx, srf, view_type, + cmd->defined_id, header, header->size + sizeof(*header), &sw_context->staged_cmd_res); } /** - * vmw_cmd_dx_set_so_targets - Validate an - * SVGA_3D_CMD_DX_SET_SOTARGETS command. + * vmw_cmd_dx_set_so_targets - Validate SVGA_3D_CMD_DX_SET_SOTARGETS command. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2607,7 +2442,7 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_so binding; struct vmw_resource *res; struct { @@ -2617,22 +2452,20 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv, } *cmd; int i, ret, num; - if (unlikely(ctx_node == NULL)) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } cmd = container_of(header, typeof(*cmd), header); - num = (cmd->header.size - sizeof(cmd->body)) / - sizeof(SVGA3dSoTarget); + num = (cmd->header.size - sizeof(cmd->body)) / sizeof(SVGA3dSoTarget); if (num > SVGA3D_DX_MAX_SOTARGETS) { - DRM_ERROR("Invalid DX SO binding.\n"); + VMW_DEBUG_USER("Invalid DX SO binding.\n"); return -EINVAL; } for (i = 0; i < num; i++) { ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + VMW_RES_DIRTY_SET, user_surface_converter, &cmd->targets[i].sid, &res); if (unlikely(ret != 0)) @@ -2645,8 +2478,7 @@ static int vmw_cmd_dx_set_so_targets(struct vmw_private *dev_priv, binding.size = cmd->targets[i].sizeInBytes; binding.slot = i; - vmw_binding_add(ctx_node->staged, &binding.bi, - 0, binding.slot); + vmw_binding_add(ctx_node->staged, &binding.bi, 0, binding.slot); } return 0; @@ -2656,7 +2488,7 @@ static int vmw_cmd_dx_so_define(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_resource *res; /* * This is based on the fact that all affected define commands have @@ -2669,10 +2501,8 @@ static int vmw_cmd_dx_so_define(struct vmw_private *dev_priv, enum vmw_so_type so_type; int ret; - if (unlikely(ctx_node == NULL)) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } so_type = vmw_so_cmd_to_type(header->id); res = vmw_context_cotable(ctx_node->ctx, vmw_so_cotables[so_type]); @@ -2683,8 +2513,8 @@ static int vmw_cmd_dx_so_define(struct vmw_private *dev_priv, } /** - * vmw_cmd_dx_check_subresource - Validate an - * SVGA_3D_CMD_DX_[X]_SUBRESOURCE command + * vmw_cmd_dx_check_subresource - Validate SVGA_3D_CMD_DX_[X]_SUBRESOURCE + * command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2714,7 +2544,7 @@ static int vmw_cmd_dx_check_subresource(struct vmw_private *dev_priv, cmd = container_of(header, typeof(*cmd), header); return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->sid, NULL); } @@ -2722,32 +2552,30 @@ static int vmw_cmd_dx_cid_check(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); - if (unlikely(ctx_node == NULL)) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } return 0; } /** - * vmw_cmd_dx_view_remove - validate a view remove command and - * schedule the view resource for removal. + * vmw_cmd_dx_view_remove - validate a view remove command and schedule the view + * resource for removal. * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. * @header: Pointer to the command header in the command stream. * - * Check that the view exists, and if it was not created using this - * command batch, conditionally make this command a NOP. + * Check that the view exists, and if it was not created using this command + * batch, conditionally make this command a NOP. */ static int vmw_cmd_dx_view_remove(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct { SVGA3dCmdHeader header; union vmw_view_destroy body; @@ -2756,15 +2584,11 @@ static int vmw_cmd_dx_view_remove(struct vmw_private *dev_priv, struct vmw_resource *view; int ret; - if (!ctx_node) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } - ret = vmw_view_remove(sw_context->man, - cmd->body.view_id, view_type, - &sw_context->staged_cmd_res, - &view); + ret = vmw_view_remove(sw_context->man, cmd->body.view_id, view_type, + &sw_context->staged_cmd_res, &view); if (ret || !view) return ret; @@ -2774,16 +2598,14 @@ static int vmw_cmd_dx_view_remove(struct vmw_private *dev_priv, * relocation to conditionally make this command a NOP to avoid * device errors. */ - return vmw_resource_relocation_add(sw_context, - view, + return vmw_resource_relocation_add(sw_context, view, vmw_ptr_diff(sw_context->buf_start, &cmd->header.id), vmw_res_rel_cond_nop); } /** - * vmw_cmd_dx_define_shader - Validate an SVGA_3D_CMD_DX_DEFINE_SHADER - * command + * vmw_cmd_dx_define_shader - Validate SVGA_3D_CMD_DX_DEFINE_SHADER command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2793,18 +2615,14 @@ static int vmw_cmd_dx_define_shader(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_resource *res; - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXDefineShader body; - } *cmd = container_of(header, typeof(*cmd), header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXDefineShader) = + container_of(header, typeof(*cmd), header); int ret; - if (!ctx_node) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } res = vmw_context_cotable(ctx_node->ctx, SVGA_COTABLE_DXSHADER); ret = vmw_cotable_notify(res, cmd->body.shaderId); @@ -2817,8 +2635,7 @@ static int vmw_cmd_dx_define_shader(struct vmw_private *dev_priv, } /** - * vmw_cmd_dx_destroy_shader - Validate an SVGA_3D_CMD_DX_DESTROY_SHADER - * command + * vmw_cmd_dx_destroy_shader - Validate SVGA_3D_CMD_DX_DESTROY_SHADER command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2828,29 +2645,22 @@ static int vmw_cmd_dx_destroy_shader(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct vmw_ctx_validation_info *ctx_node = sw_context->dx_ctx_node; - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXDestroyShader body; - } *cmd = container_of(header, typeof(*cmd), header); + struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXDestroyShader) = + container_of(header, typeof(*cmd), header); int ret; - if (!ctx_node) { - DRM_ERROR("DX Context not set.\n"); + if (!ctx_node) return -EINVAL; - } ret = vmw_shader_remove(sw_context->man, cmd->body.shaderId, 0, &sw_context->staged_cmd_res); - if (ret) - DRM_ERROR("Could not find shader to remove.\n"); return ret; } /** - * vmw_cmd_dx_bind_shader - Validate an SVGA_3D_CMD_DX_BIND_SHADER - * command + * vmw_cmd_dx_bind_shader - Validate SVGA_3D_CMD_DX_BIND_SHADER command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2862,36 +2672,37 @@ static int vmw_cmd_dx_bind_shader(struct vmw_private *dev_priv, { struct vmw_resource *ctx; struct vmw_resource *res; - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXBindShader body; - } *cmd = container_of(header, typeof(*cmd), header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXBindShader) = + container_of(header, typeof(*cmd), header); int ret; if (cmd->body.cid != SVGA3D_INVALID_ID) { ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_context, - user_context_converter, - &cmd->body.cid, &ctx); + VMW_RES_DIRTY_SET, + user_context_converter, &cmd->body.cid, + &ctx); if (ret) return ret; } else { - if (!sw_context->dx_ctx_node) { - DRM_ERROR("DX Context not set.\n"); + struct vmw_ctx_validation_info *ctx_node = + VMW_GET_CTX_NODE(sw_context); + + if (!ctx_node) return -EINVAL; - } - ctx = sw_context->dx_ctx_node->ctx; + + ctx = ctx_node->ctx; } - res = vmw_shader_lookup(vmw_context_res_man(ctx), - cmd->body.shid, 0); + res = vmw_shader_lookup(vmw_context_res_man(ctx), cmd->body.shid, 0); if (IS_ERR(res)) { - DRM_ERROR("Could not find shader to bind.\n"); + VMW_DEBUG_USER("Could not find shader to bind.\n"); return PTR_ERR(res); } - ret = vmw_execbuf_res_noctx_val_add(sw_context, res); + ret = vmw_execbuf_res_noctx_val_add(sw_context, res, + VMW_RES_DIRTY_NONE); if (ret) { - DRM_ERROR("Error creating resource validation node.\n"); + VMW_DEBUG_USER("Error creating resource validation node.\n"); return ret; } @@ -2901,7 +2712,7 @@ static int vmw_cmd_dx_bind_shader(struct vmw_private *dev_priv, } /** - * vmw_cmd_dx_genmips - Validate an SVGA_3D_CMD_DX_GENMIPS command + * vmw_cmd_dx_genmips - Validate SVGA_3D_CMD_DX_GENMIPS command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2911,18 +2722,16 @@ static int vmw_cmd_dx_genmips(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXGenMips body; - } *cmd = container_of(header, typeof(*cmd), header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXGenMips) = + container_of(header, typeof(*cmd), header); return PTR_RET(vmw_view_id_val_add(sw_context, vmw_view_sr, cmd->body.shaderResourceViewId)); } /** - * vmw_cmd_dx_transfer_from_buffer - - * Validate an SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER command + * vmw_cmd_dx_transfer_from_buffer - Validate + * SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2932,26 +2741,23 @@ static int vmw_cmd_dx_transfer_from_buffer(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdDXTransferFromBuffer body; - } *cmd = container_of(header, typeof(*cmd), header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXTransferFromBuffer) = + container_of(header, typeof(*cmd), header); int ret; ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_NONE, user_surface_converter, &cmd->body.srcSid, NULL); if (ret != 0) return ret; return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, + VMW_RES_DIRTY_SET, user_surface_converter, &cmd->body.destSid, NULL); } /** - * vmw_cmd_intra_surface_copy - - * Validate an SVGA_3D_CMD_INTRA_SURFACE_COPY command + * vmw_cmd_intra_surface_copy - Validate SVGA_3D_CMD_INTRA_SURFACE_COPY command * * @dev_priv: Pointer to a device private struct. * @sw_context: The software context being used for this batch. @@ -2961,20 +2767,17 @@ static int vmw_cmd_intra_surface_copy(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, SVGA3dCmdHeader *header) { - struct { - SVGA3dCmdHeader header; - SVGA3dCmdIntraSurfaceCopy body; - } *cmd = container_of(header, typeof(*cmd), header); + VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdIntraSurfaceCopy) = + container_of(header, typeof(*cmd), header); if (!(dev_priv->capabilities2 & SVGA_CAP2_INTRA_SURFACE_COPY)) return -EINVAL; return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, - user_surface_converter, - &cmd->body.surface.sid, NULL); + VMW_RES_DIRTY_SET, user_surface_converter, + &cmd->body.surface.sid, NULL); } - static int vmw_cmd_check_not_3d(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, void *buf, uint32_t *size) @@ -2997,18 +2800,18 @@ static int vmw_cmd_check_not_3d(struct vmw_private *dev_priv, *size = sizeof(uint32_t) + sizeof(SVGAFifoCmdBlitGMRFBToScreen); break; default: - DRM_ERROR("Unsupported SVGA command: %u.\n", cmd_id); + VMW_DEBUG_USER("Unsupported SVGA command: %u.\n", cmd_id); return -EINVAL; } if (*size > size_remaining) { - DRM_ERROR("Invalid SVGA command (size mismatch):" - " %u.\n", cmd_id); + VMW_DEBUG_USER("Invalid SVGA command (size mismatch): %u.\n", + cmd_id); return -EINVAL; } if (unlikely(!sw_context->kernel)) { - DRM_ERROR("Kernel only SVGA command: %u.\n", cmd_id); + VMW_DEBUG_USER("Kernel only SVGA command: %u.\n", cmd_id); return -EPERM; } @@ -3196,9 +2999,7 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = { VMW_CMD_DEF(SVGA_3D_CMD_DEFINE_GB_SURFACE_V2, &vmw_cmd_invalid, false, false, true), - /* - * DX commands - */ + /* SM commands */ VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_CONTEXT, &vmw_cmd_invalid, false, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_DESTROY_CONTEXT, &vmw_cmd_invalid, @@ -3380,8 +3181,8 @@ bool vmw_cmd_describe(const void *buf, u32 *size, char const **cmd) } static int vmw_cmd_check(struct vmw_private *dev_priv, - struct vmw_sw_context *sw_context, - void *buf, uint32_t *size) + struct vmw_sw_context *sw_context, void *buf, + uint32_t *size) { uint32_t cmd_id; uint32_t size_remaining = *size; @@ -3420,31 +3221,33 @@ static int vmw_cmd_check(struct vmw_private *dev_priv, goto out_new; ret = entry->func(dev_priv, sw_context, header); - if (unlikely(ret != 0)) - goto out_invalid; + if (unlikely(ret != 0)) { + VMW_DEBUG_USER("SVGA3D command: %d failed with error %d\n", + cmd_id + SVGA_3D_CMD_BASE, ret); + return ret; + } return 0; out_invalid: - DRM_ERROR("Invalid SVGA3D command: %d\n", - cmd_id + SVGA_3D_CMD_BASE); + VMW_DEBUG_USER("Invalid SVGA3D command: %d\n", + cmd_id + SVGA_3D_CMD_BASE); return -EINVAL; out_privileged: - DRM_ERROR("Privileged SVGA3D command: %d\n", - cmd_id + SVGA_3D_CMD_BASE); + VMW_DEBUG_USER("Privileged SVGA3D command: %d\n", + cmd_id + SVGA_3D_CMD_BASE); return -EPERM; out_old: - DRM_ERROR("Deprecated (disallowed) SVGA3D command: %d\n", - cmd_id + SVGA_3D_CMD_BASE); + VMW_DEBUG_USER("Deprecated (disallowed) SVGA3D command: %d\n", + cmd_id + SVGA_3D_CMD_BASE); return -EINVAL; out_new: - DRM_ERROR("SVGA3D command: %d not supported by virtual hardware.\n", - cmd_id + SVGA_3D_CMD_BASE); + VMW_DEBUG_USER("SVGA3D command: %d not supported by virtual device.\n", + cmd_id + SVGA_3D_CMD_BASE); return -EINVAL; } static int vmw_cmd_check_all(struct vmw_private *dev_priv, - struct vmw_sw_context *sw_context, - void *buf, + struct vmw_sw_context *sw_context, void *buf, uint32_t size) { int32_t cur_size = size; @@ -3462,7 +3265,7 @@ static int vmw_cmd_check_all(struct vmw_private *dev_priv, } if (unlikely(cur_size != 0)) { - DRM_ERROR("Command verifier out of sync.\n"); + VMW_DEBUG_USER("Command verifier out of sync.\n"); return -EINVAL; } @@ -3472,7 +3275,6 @@ static int vmw_cmd_check_all(struct vmw_private *dev_priv, static void vmw_free_relocations(struct vmw_sw_context *sw_context) { /* Memory is validation context memory, so no need to free it */ - INIT_LIST_HEAD(&sw_context->bo_relocations); } @@ -3520,7 +3322,7 @@ static int vmw_resize_cmd_bounce(struct vmw_sw_context *sw_context, sw_context->cmd_bounce = vmalloc(sw_context->cmd_bounce_size); if (sw_context->cmd_bounce == NULL) { - DRM_ERROR("Failed to allocate command bounce buffer.\n"); + VMW_DEBUG_USER("Failed to allocate command bounce buffer.\n"); sw_context->cmd_bounce_size = 0; return -ENOMEM; } @@ -3535,8 +3337,8 @@ static int vmw_resize_cmd_bounce(struct vmw_sw_context *sw_context, * If this fails for some reason, We sync the fifo and return NULL. * It is then safe to fence buffers with a NULL pointer. * - * If @p_handle is not NULL @file_priv must also not be NULL. Creates - * a userspace handle if @p_handle is not NULL, otherwise not. + * If @p_handle is not NULL @file_priv must also not be NULL. Creates a + * userspace handle if @p_handle is not NULL, otherwise not. */ int vmw_execbuf_fence_commands(struct drm_file *file_priv, @@ -3553,7 +3355,7 @@ int vmw_execbuf_fence_commands(struct drm_file *file_priv, ret = vmw_fifo_send_fence(dev_priv, &sequence); if (unlikely(ret != 0)) { - DRM_ERROR("Fence submission error. Syncing.\n"); + VMW_DEBUG_USER("Fence submission error. Syncing.\n"); synced = true; } @@ -3564,9 +3366,8 @@ int vmw_execbuf_fence_commands(struct drm_file *file_priv, ret = vmw_fence_create(dev_priv->fman, sequence, p_fence); if (unlikely(ret != 0 && !synced)) { - (void) vmw_fallback_wait(dev_priv, false, false, - sequence, false, - VMW_FENCE_WAIT_TIMEOUT); + (void) vmw_fallback_wait(dev_priv, false, false, sequence, + false, VMW_FENCE_WAIT_TIMEOUT); *p_fence = NULL; } @@ -3574,36 +3375,32 @@ int vmw_execbuf_fence_commands(struct drm_file *file_priv, } /** - * vmw_execbuf_copy_fence_user - copy fence object information to - * user-space. + * vmw_execbuf_copy_fence_user - copy fence object information to user-space. * * @dev_priv: Pointer to a vmw_private struct. * @vmw_fp: Pointer to the struct vmw_fpriv representing the calling file. * @ret: Return value from fence object creation. - * @user_fence_rep: User space address of a struct drm_vmw_fence_rep to - * which the information should be copied. + * @user_fence_rep: User space address of a struct drm_vmw_fence_rep to which + * the information should be copied. * @fence: Pointer to the fenc object. * @fence_handle: User-space fence handle. * @out_fence_fd: exported file descriptor for the fence. -1 if not used * @sync_file: Only used to clean up in case of an error in this function. * - * This function copies fence information to user-space. If copying fails, - * The user-space struct drm_vmw_fence_rep::error member is hopefully - * left untouched, and if it's preloaded with an -EFAULT by user-space, - * the error will hopefully be detected. - * Also if copying fails, user-space will be unable to signal the fence - * object so we wait for it immediately, and then unreference the - * user-space reference. + * This function copies fence information to user-space. If copying fails, the + * user-space struct drm_vmw_fence_rep::error member is hopefully left + * untouched, and if it's preloaded with an -EFAULT by user-space, the error + * will hopefully be detected. + * + * Also if copying fails, user-space will be unable to signal the fence object + * so we wait for it immediately, and then unreference the user-space reference. */ void vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv, - struct vmw_fpriv *vmw_fp, - int ret, + struct vmw_fpriv *vmw_fp, int ret, struct drm_vmw_fence_rep __user *user_fence_rep, - struct vmw_fence_obj *fence, - uint32_t fence_handle, - int32_t out_fence_fd, - struct sync_file *sync_file) + struct vmw_fence_obj *fence, uint32_t fence_handle, + int32_t out_fence_fd, struct sync_file *sync_file) { struct drm_vmw_fence_rep fence_rep; @@ -3624,16 +3421,16 @@ vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv, } /* - * copy_to_user errors will be detected by user space not - * seeing fence_rep::error filled in. Typically - * user-space would have pre-set that member to -EFAULT. + * copy_to_user errors will be detected by user space not seeing + * fence_rep::error filled in. Typically user-space would have pre-set + * that member to -EFAULT. */ ret = copy_to_user(user_fence_rep, &fence_rep, sizeof(fence_rep)); /* - * User-space lost the fence object. We need to sync - * and unreference the handle. + * User-space lost the fence object. We need to sync and unreference the + * handle. */ if (unlikely(ret != 0) && (fence_rep.error == 0)) { if (sync_file) @@ -3644,42 +3441,39 @@ vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv, fence_rep.fd = -1; } - ttm_ref_object_base_unref(vmw_fp->tfile, - fence_handle, TTM_REF_USAGE); - DRM_ERROR("Fence copy error. Syncing.\n"); + ttm_ref_object_base_unref(vmw_fp->tfile, fence_handle, + TTM_REF_USAGE); + VMW_DEBUG_USER("Fence copy error. Syncing.\n"); (void) vmw_fence_obj_wait(fence, false, false, VMW_FENCE_WAIT_TIMEOUT); } } /** - * vmw_execbuf_submit_fifo - Patch a command batch and submit it using - * the fifo. + * vmw_execbuf_submit_fifo - Patch a command batch and submit it using the fifo. * * @dev_priv: Pointer to a device private structure. * @kernel_commands: Pointer to the unpatched command batch. * @command_size: Size of the unpatched command batch. * @sw_context: Structure holding the relocation lists. * - * Side effects: If this function returns 0, then the command batch - * pointed to by @kernel_commands will have been modified. + * Side effects: If this function returns 0, then the command batch pointed to + * by @kernel_commands will have been modified. */ static int vmw_execbuf_submit_fifo(struct vmw_private *dev_priv, - void *kernel_commands, - u32 command_size, + void *kernel_commands, u32 command_size, struct vmw_sw_context *sw_context) { void *cmd; if (sw_context->dx_ctx_node) - cmd = vmw_fifo_reserve_dx(dev_priv, command_size, + cmd = VMW_FIFO_RESERVE_DX(dev_priv, command_size, sw_context->dx_ctx_node->ctx->id); else - cmd = vmw_fifo_reserve(dev_priv, command_size); - if (!cmd) { - DRM_ERROR("Failed reserving fifo space for commands.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, command_size); + + if (!cmd) return -ENOMEM; - } vmw_apply_relocations(sw_context); memcpy(cmd, kernel_commands, command_size); @@ -3691,16 +3485,16 @@ static int vmw_execbuf_submit_fifo(struct vmw_private *dev_priv, } /** - * vmw_execbuf_submit_cmdbuf - Patch a command batch and submit it using - * the command buffer manager. + * vmw_execbuf_submit_cmdbuf - Patch a command batch and submit it using the + * command buffer manager. * * @dev_priv: Pointer to a device private structure. * @header: Opaque handle to the command buffer allocation. * @command_size: Size of the unpatched command batch. * @sw_context: Structure holding the relocation lists. * - * Side effects: If this function returns 0, then the command buffer - * represented by @header will have been modified. + * Side effects: If this function returns 0, then the command buffer represented + * by @header will have been modified. */ static int vmw_execbuf_submit_cmdbuf(struct vmw_private *dev_priv, struct vmw_cmdbuf_header *header, @@ -3709,8 +3503,8 @@ static int vmw_execbuf_submit_cmdbuf(struct vmw_private *dev_priv, { u32 id = ((sw_context->dx_ctx_node) ? sw_context->dx_ctx_node->ctx->id : SVGA3D_INVALID_ID); - void *cmd = vmw_cmdbuf_reserve(dev_priv->cman, command_size, - id, false, header); + void *cmd = vmw_cmdbuf_reserve(dev_priv->cman, command_size, id, false, + header); vmw_apply_relocations(sw_context); vmw_resource_relocations_apply(cmd, &sw_context->res_relocations); @@ -3730,22 +3524,23 @@ static int vmw_execbuf_submit_cmdbuf(struct vmw_private *dev_priv, * @header: Out parameter returning the opaque pointer to the command buffer. * * This function checks whether we can use the command buffer manager for - * submission and if so, creates a command buffer of suitable size and - * copies the user data into that buffer. + * submission and if so, creates a command buffer of suitable size and copies + * the user data into that buffer. * * On successful return, the function returns a pointer to the data in the * command buffer and *@header is set to non-NULL. - * If command buffers could not be used, the function will return the value - * of @kernel_commands on function call. That value may be NULL. In that case, - * the value of *@header will be set to NULL. + * + * If command buffers could not be used, the function will return the value of + * @kernel_commands on function call. That value may be NULL. In that case, the + * value of *@header will be set to NULL. + * * If an error is encountered, the function will return a pointer error value. * If the function is interrupted by a signal while sleeping, it will return * -ERESTARTSYS casted to a pointer error value. */ static void *vmw_execbuf_cmdbuf(struct vmw_private *dev_priv, void __user *user_commands, - void *kernel_commands, - u32 command_size, + void *kernel_commands, u32 command_size, struct vmw_cmdbuf_header **header) { size_t cmdbuf_size; @@ -3753,7 +3548,7 @@ static void *vmw_execbuf_cmdbuf(struct vmw_private *dev_priv, *header = NULL; if (command_size > SVGA_CB_MAX_SIZE) { - DRM_ERROR("Command buffer is too large.\n"); + VMW_DEBUG_USER("Command buffer is too large.\n"); return ERR_PTR(-EINVAL); } @@ -3763,15 +3558,14 @@ static void *vmw_execbuf_cmdbuf(struct vmw_private *dev_priv, /* If possible, add a little space for fencing. */ cmdbuf_size = command_size + 512; cmdbuf_size = min_t(size_t, cmdbuf_size, SVGA_CB_MAX_SIZE); - kernel_commands = vmw_cmdbuf_alloc(dev_priv->cman, cmdbuf_size, - true, header); + kernel_commands = vmw_cmdbuf_alloc(dev_priv->cman, cmdbuf_size, true, + header); if (IS_ERR(kernel_commands)) return kernel_commands; - ret = copy_from_user(kernel_commands, user_commands, - command_size); + ret = copy_from_user(kernel_commands, user_commands, command_size); if (ret) { - DRM_ERROR("Failed copying commands.\n"); + VMW_DEBUG_USER("Failed copying commands.\n"); vmw_cmdbuf_header_free(*header); *header = NULL; return ERR_PTR(-EFAULT); @@ -3799,13 +3593,13 @@ static int vmw_execbuf_tie_context(struct vmw_private *dev_priv, res = vmw_user_resource_noref_lookup_handle (dev_priv, sw_context->fp->tfile, handle, user_context_converter); - if (unlikely(IS_ERR(res))) { - DRM_ERROR("Could not find or user DX context 0x%08x.\n", - (unsigned) handle); + if (IS_ERR(res)) { + VMW_DEBUG_USER("Could not find or user DX context 0x%08x.\n", + (unsigned int) handle); return PTR_ERR(res); } - ret = vmw_execbuf_res_noref_val_add(sw_context, res); + ret = vmw_execbuf_res_noref_val_add(sw_context, res, VMW_RES_DIRTY_SET); if (unlikely(ret != 0)) return ret; @@ -3817,19 +3611,16 @@ static int vmw_execbuf_tie_context(struct vmw_private *dev_priv, int vmw_execbuf_process(struct drm_file *file_priv, struct vmw_private *dev_priv, - void __user *user_commands, - void *kernel_commands, - uint32_t command_size, - uint64_t throttle_us, + void __user *user_commands, void *kernel_commands, + uint32_t command_size, uint64_t throttle_us, uint32_t dx_context_handle, struct drm_vmw_fence_rep __user *user_fence_rep, - struct vmw_fence_obj **out_fence, - uint32_t flags) + struct vmw_fence_obj **out_fence, uint32_t flags) { struct vmw_sw_context *sw_context = &dev_priv->ctx; struct vmw_fence_obj *fence = NULL; struct vmw_cmdbuf_header *header; - uint32_t handle; + uint32_t handle = 0; int ret; int32_t out_fence_fd = -1; struct sync_file *sync_file = NULL; @@ -3840,7 +3631,7 @@ int vmw_execbuf_process(struct drm_file *file_priv, if (flags & DRM_VMW_EXECBUF_FLAG_EXPORT_FENCE_FD) { out_fence_fd = get_unused_fd_flags(O_CLOEXEC); if (out_fence_fd < 0) { - DRM_ERROR("Failed to get a fence file descriptor.\n"); + VMW_DEBUG_USER("Failed to get a fence fd.\n"); return out_fence_fd; } } @@ -3873,18 +3664,18 @@ int vmw_execbuf_process(struct drm_file *file_priv, if (unlikely(ret != 0)) goto out_unlock; - - ret = copy_from_user(sw_context->cmd_bounce, - user_commands, command_size); - + ret = copy_from_user(sw_context->cmd_bounce, user_commands, + command_size); if (unlikely(ret != 0)) { ret = -EFAULT; - DRM_ERROR("Failed copying commands.\n"); + VMW_DEBUG_USER("Failed copying commands.\n"); goto out_unlock; } + kernel_commands = sw_context->cmd_bounce; - } else if (!header) + } else if (!header) { sw_context->kernel = true; + } sw_context->fp = vmw_fpriv(file_priv); INIT_LIST_HEAD(&sw_context->ctx_list); @@ -3897,6 +3688,7 @@ int vmw_execbuf_process(struct drm_file *file_priv, memset(sw_context->res_cache, 0, sizeof(sw_context->res_cache)); INIT_LIST_HEAD(&sw_context->res_relocations); INIT_LIST_HEAD(&sw_context->bo_relocations); + if (sw_context->staged_bindings) vmw_binding_state_reset(sw_context->staged_bindings); @@ -3904,8 +3696,10 @@ int vmw_execbuf_process(struct drm_file *file_priv, ret = drm_ht_create(&sw_context->res_ht, VMW_RES_HT_ORDER); if (unlikely(ret != 0)) goto out_unlock; + sw_context->res_ht_initialized = true; } + INIT_LIST_HEAD(&sw_context->staged_cmd_res); sw_context->ctx = &val_ctx; ret = vmw_execbuf_tie_context(dev_priv, sw_context, dx_context_handle); @@ -3932,6 +3726,7 @@ int vmw_execbuf_process(struct drm_file *file_priv, ret = vmw_validation_res_validate(&val_ctx, true); if (unlikely(ret != 0)) goto out_err; + vmw_validation_drop_ht(&val_ctx); ret = mutex_lock_interruptible(&dev_priv->binding_mutex); @@ -3959,17 +3754,15 @@ int vmw_execbuf_process(struct drm_file *file_priv, goto out_err; vmw_query_bo_switch_commit(dev_priv, sw_context); - ret = vmw_execbuf_fence_commands(file_priv, dev_priv, - &fence, + ret = vmw_execbuf_fence_commands(file_priv, dev_priv, &fence, (user_fence_rep) ? &handle : NULL); /* * This error is harmless, because if fence submission fails, * vmw_fifo_send_fence will sync. The error will be propagated to * user-space in @fence_rep */ - if (ret != 0) - DRM_ERROR("Fence submission error. Syncing.\n"); + VMW_DEBUG_USER("Fence submission error. Syncing.\n"); vmw_execbuf_bindings_commit(sw_context, false); vmw_bind_dx_query_mob(sw_context); @@ -3977,21 +3770,19 @@ int vmw_execbuf_process(struct drm_file *file_priv, vmw_validation_bo_fence(sw_context->ctx, fence); - if (unlikely(dev_priv->pinned_bo != NULL && - !dev_priv->query_cid_valid)) + if (unlikely(dev_priv->pinned_bo != NULL && !dev_priv->query_cid_valid)) __vmw_execbuf_release_pinned_bo(dev_priv, fence); /* - * If anything fails here, give up trying to export the fence - * and do a sync since the user mode will not be able to sync - * the fence itself. This ensures we are still functionally - * correct. + * If anything fails here, give up trying to export the fence and do a + * sync since the user mode will not be able to sync the fence itself. + * This ensures we are still functionally correct. */ if (flags & DRM_VMW_EXECBUF_FLAG_EXPORT_FENCE_FD) { sync_file = sync_file_create(&fence->base); if (!sync_file) { - DRM_ERROR("Unable to create sync file for fence\n"); + VMW_DEBUG_USER("Sync file create failed for fence\n"); put_unused_fd(out_fence_fd); out_fence_fd = -1; @@ -4004,8 +3795,8 @@ int vmw_execbuf_process(struct drm_file *file_priv, } vmw_execbuf_copy_fence_user(dev_priv, vmw_fpriv(file_priv), ret, - user_fence_rep, fence, handle, - out_fence_fd, sync_file); + user_fence_rep, fence, handle, out_fence_fd, + sync_file); /* Don't unreference when handing fence out */ if (unlikely(out_fence != NULL)) { @@ -4019,8 +3810,8 @@ int vmw_execbuf_process(struct drm_file *file_priv, mutex_unlock(&dev_priv->cmdbuf_mutex); /* - * Unreference resources outside of the cmdbuf_mutex to - * avoid deadlocks in resource destruction paths. + * Unreference resources outside of the cmdbuf_mutex to avoid deadlocks + * in resource destruction paths. */ vmw_validation_unref_lists(&val_ctx); @@ -4035,8 +3826,7 @@ out_err_nores: vmw_validation_res_unreserve(&val_ctx, true); vmw_resource_relocations_free(&sw_context->res_relocations); vmw_free_relocations(sw_context); - if (unlikely(dev_priv->pinned_bo != NULL && - !dev_priv->query_cid_valid)) + if (unlikely(dev_priv->pinned_bo != NULL && !dev_priv->query_cid_valid)) __vmw_execbuf_release_pinned_bo(dev_priv, NULL); out_unlock: vmw_cmdbuf_res_revert(&sw_context->staged_cmd_res); @@ -4045,8 +3835,8 @@ out_unlock: mutex_unlock(&dev_priv->cmdbuf_mutex); /* - * Unreference resources outside of the cmdbuf_mutex to - * avoid deadlocks in resource destruction paths. + * Unreference resources outside of the cmdbuf_mutex to avoid deadlocks + * in resource destruction paths. */ vmw_validation_unref_lists(&val_ctx); out_free_header: @@ -4064,13 +3854,13 @@ out_free_fence_fd: * * @dev_priv: The device private structure. * - * This function is called to idle the fifo and unpin the query buffer - * if the normal way to do this hits an error, which should typically be - * extremely rare. + * This function is called to idle the fifo and unpin the query buffer if the + * normal way to do this hits an error, which should typically be extremely + * rare. */ static void vmw_execbuf_unpin_panic(struct vmw_private *dev_priv) { - DRM_ERROR("Can't unpin query buffer. Trying to recover.\n"); + VMW_DEBUG_USER("Can't unpin query buffer. Trying to recover.\n"); (void) vmw_fallback_wait(dev_priv, false, true, 0, false, 10*HZ); vmw_bo_pin_reserved(dev_priv->pinned_bo, false); @@ -4082,28 +3872,27 @@ static void vmw_execbuf_unpin_panic(struct vmw_private *dev_priv) /** - * __vmw_execbuf_release_pinned_bo - Flush queries and unpin the pinned - * query bo. + * __vmw_execbuf_release_pinned_bo - Flush queries and unpin the pinned query + * bo. * * @dev_priv: The device private structure. - * @fence: If non-NULL should point to a struct vmw_fence_obj issued - * _after_ a query barrier that flushes all queries touching the current - * buffer pointed to by @dev_priv->pinned_bo + * @fence: If non-NULL should point to a struct vmw_fence_obj issued _after_ a + * query barrier that flushes all queries touching the current buffer pointed to + * by @dev_priv->pinned_bo * - * This function should be used to unpin the pinned query bo, or - * as a query barrier when we need to make sure that all queries have - * finished before the next fifo command. (For example on hardware - * context destructions where the hardware may otherwise leak unfinished - * queries). + * This function should be used to unpin the pinned query bo, or as a query + * barrier when we need to make sure that all queries have finished before the + * next fifo command. (For example on hardware context destructions where the + * hardware may otherwise leak unfinished queries). * - * This function does not return any failure codes, but make attempts - * to do safe unpinning in case of errors. + * This function does not return any failure codes, but make attempts to do safe + * unpinning in case of errors. * - * The function will synchronize on the previous query barrier, and will - * thus not finish until that barrier has executed. + * The function will synchronize on the previous query barrier, and will thus + * not finish until that barrier has executed. * - * the @dev_priv->cmdbuf_mutex needs to be held by the current thread - * before calling this function. + * the @dev_priv->cmdbuf_mutex needs to be held by the current thread before + * calling this function. */ void __vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv, struct vmw_fence_obj *fence) @@ -4153,35 +3942,32 @@ void __vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv, vmw_validation_unref_lists(&val_ctx); vmw_bo_unreference(&dev_priv->pinned_bo); + out_unlock: return; - out_no_emit: vmw_validation_bo_backoff(&val_ctx); out_no_reserve: vmw_validation_unref_lists(&val_ctx); vmw_execbuf_unpin_panic(dev_priv); vmw_bo_unreference(&dev_priv->pinned_bo); - } /** - * vmw_execbuf_release_pinned_bo - Flush queries and unpin the pinned - * query bo. + * vmw_execbuf_release_pinned_bo - Flush queries and unpin the pinned query bo. * * @dev_priv: The device private structure. * - * This function should be used to unpin the pinned query bo, or - * as a query barrier when we need to make sure that all queries have - * finished before the next fifo command. (For example on hardware - * context destructions where the hardware may otherwise leak unfinished - * queries). + * This function should be used to unpin the pinned query bo, or as a query + * barrier when we need to make sure that all queries have finished before the + * next fifo command. (For example on hardware context destructions where the + * hardware may otherwise leak unfinished queries). * - * This function does not return any failure codes, but make attempts - * to do safe unpinning in case of errors. + * This function does not return any failure codes, but make attempts to do safe + * unpinning in case of errors. * - * The function will synchronize on the previous query barrier, and will - * thus not finish until that barrier has executed. + * The function will synchronize on the previous query barrier, and will thus + * not finish until that barrier has executed. */ void vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv) { @@ -4203,8 +3989,8 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data, struct dma_fence *in_fence = NULL; if (unlikely(size < copy_offset[0])) { - DRM_ERROR("Invalid command size, ioctl %d\n", - DRM_VMW_EXECBUF); + VMW_DEBUG_USER("Invalid command size, ioctl %d\n", + DRM_VMW_EXECBUF); return -EINVAL; } @@ -4212,23 +3998,19 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data, return -EFAULT; /* - * Extend the ioctl argument while - * maintaining backwards compatibility: - * We take different code paths depending on the value of - * arg.version. + * Extend the ioctl argument while maintaining backwards compatibility: + * We take different code paths depending on the value of arg.version. */ - if (unlikely(arg.version > DRM_VMW_EXECBUF_VERSION || arg.version == 0)) { - DRM_ERROR("Incorrect execbuf version.\n"); + VMW_DEBUG_USER("Incorrect execbuf version.\n"); return -EINVAL; } if (arg.version > 1 && copy_from_user(&arg.context_handle, (void __user *) (data + copy_offset[0]), - copy_offset[arg.version - 1] - - copy_offset[0]) != 0) + copy_offset[arg.version - 1] - copy_offset[0]) != 0) return -EFAULT; switch (arg.version) { @@ -4240,13 +4022,12 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data, break; } - /* If imported a fence FD from elsewhere, then wait on it */ if (arg.flags & DRM_VMW_EXECBUF_FLAG_IMPORT_FENCE_FD) { in_fence = sync_file_get_fence(arg.imported_fence_fd); if (!in_fence) { - DRM_ERROR("Cannot get imported fence\n"); + VMW_DEBUG_USER("Cannot get imported fence\n"); return -EINVAL; } @@ -4264,8 +4045,8 @@ int vmw_execbuf_ioctl(struct drm_device *dev, unsigned long data, NULL, arg.command_size, arg.throttle_us, arg.context_handle, (void __user *)(unsigned long)arg.fence_rep, - NULL, - arg.flags); + NULL, arg.flags); + ttm_read_unlock(&dev_priv->reservation_sem); if (unlikely(ret != 0)) goto out; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c index 2a9112515f46..972e8fda6d35 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c @@ -642,12 +642,11 @@ int vmw_fb_init(struct vmw_private *vmw_priv) struct vmw_fb_par *par; struct fb_info *info; unsigned fb_width, fb_height; - unsigned fb_bpp, fb_depth, fb_offset, fb_pitch, fb_size; + unsigned int fb_bpp, fb_pitch, fb_size; struct drm_display_mode *init_mode; int ret; fb_bpp = 32; - fb_depth = 24; /* XXX As shouldn't these be as well. */ fb_width = min(vmw_priv->fb_max_width, (unsigned)2048); @@ -655,7 +654,6 @@ int vmw_fb_init(struct vmw_private *vmw_priv) fb_pitch = fb_width * fb_bpp / 8; fb_size = fb_pitch * fb_height; - fb_offset = vmw_read(vmw_priv, SVGA_REG_FB_OFFSET); info = framebuffer_alloc(sizeof(*par), device); if (!info) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c index d0fd147ef75f..ff3586cb6851 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c @@ -395,12 +395,8 @@ void *vmw_fifo_reserve_dx(struct vmw_private *dev_priv, uint32_t bytes, WARN(1, "Command buffer has not been allocated.\n"); ret = NULL; } - if (IS_ERR_OR_NULL(ret)) { - DRM_ERROR("Fifo reserve failure of %u bytes.\n", - (unsigned) bytes); - dump_stack(); + if (IS_ERR_OR_NULL(ret)) return NULL; - } return ret; } @@ -544,7 +540,7 @@ int vmw_fifo_send_fence(struct vmw_private *dev_priv, uint32_t *seqno) int ret = 0; uint32_t bytes = sizeof(u32) + sizeof(*cmd_fence); - fm = vmw_fifo_reserve(dev_priv, bytes); + fm = VMW_FIFO_RESERVE(dev_priv, bytes); if (unlikely(fm == NULL)) { *seqno = atomic_read(&dev_priv->marker_seq); ret = -ENOMEM; @@ -603,12 +599,9 @@ static int vmw_fifo_emit_dummy_legacy_query(struct vmw_private *dev_priv, SVGA3dCmdWaitForQuery body; } *cmd; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - - if (unlikely(cmd == NULL)) { - DRM_ERROR("Out of fifo space for dummy query.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_WAIT_FOR_QUERY; cmd->header.size = sizeof(cmd->body); @@ -652,12 +645,9 @@ static int vmw_fifo_emit_dummy_gb_query(struct vmw_private *dev_priv, SVGA3dCmdWaitForGBQuery body; } *cmd; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - - if (unlikely(cmd == NULL)) { - DRM_ERROR("Out of fifo space for dummy query.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_WAIT_FOR_GB_QUERY; cmd->header.size = sizeof(cmd->body); @@ -699,8 +689,3 @@ int vmw_fifo_emit_dummy_query(struct vmw_private *dev_priv, return vmw_fifo_emit_dummy_legacy_query(dev_priv, cid); } - -void *vmw_fifo_reserve(struct vmw_private *dev_priv, uint32_t bytes) -{ - return vmw_fifo_reserve_dx(dev_priv, bytes, SVGA3D_INVALID_ID); -} diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c index 007a0cc7f232..ae7acc6f3dda 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c @@ -51,7 +51,7 @@ static int vmw_gmr2_bind(struct vmw_private *dev_priv, uint32_t cmd_size = define_size + remap_size; uint32_t i; - cmd_orig = cmd = vmw_fifo_reserve(dev_priv, cmd_size); + cmd_orig = cmd = VMW_FIFO_RESERVE(dev_priv, cmd_size); if (unlikely(cmd == NULL)) return -ENOMEM; @@ -110,11 +110,10 @@ static void vmw_gmr2_unbind(struct vmw_private *dev_priv, uint32_t define_size = sizeof(define_cmd) + 4; uint32_t *cmd; - cmd = vmw_fifo_reserve(dev_priv, define_size); - if (unlikely(cmd == NULL)) { - DRM_ERROR("GMR2 unbind failed.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, define_size); + if (unlikely(cmd == NULL)) return; - } + define_cmd.gmrId = gmr_id; define_cmd.numPages = 0; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c index 172a6ba6539c..a15375eb476e 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c @@ -188,7 +188,7 @@ int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data, struct vmw_fpriv *vmw_fp = vmw_fpriv(file_priv); if (unlikely(arg->pad64 != 0 || arg->max_size == 0)) { - DRM_ERROR("Illegal GET_3D_CAP argument.\n"); + VMW_DEBUG_USER("Illegal GET_3D_CAP argument.\n"); return -EINVAL; } @@ -268,7 +268,7 @@ int vmw_present_ioctl(struct drm_device *dev, void *data, return 0; if (clips_ptr == NULL) { - DRM_ERROR("Variable clips_ptr must be specified.\n"); + VMW_DEBUG_USER("Variable clips_ptr must be specified.\n"); ret = -EINVAL; goto out_clips; } @@ -291,7 +291,7 @@ int vmw_present_ioctl(struct drm_device *dev, void *data, fb = drm_framebuffer_lookup(dev, file_priv, arg->fb_id); if (!fb) { - DRM_ERROR("Invalid framebuffer id.\n"); + VMW_DEBUG_USER("Invalid framebuffer id.\n"); ret = -ENOENT; goto out_no_fb; } @@ -351,7 +351,7 @@ int vmw_present_readback_ioctl(struct drm_device *dev, void *data, return 0; if (clips_ptr == NULL) { - DRM_ERROR("Argument clips_ptr must be specified.\n"); + VMW_DEBUG_USER("Argument clips_ptr must be specified.\n"); ret = -EINVAL; goto out_clips; } @@ -374,14 +374,14 @@ int vmw_present_readback_ioctl(struct drm_device *dev, void *data, fb = drm_framebuffer_lookup(dev, file_priv, arg->fb_id); if (!fb) { - DRM_ERROR("Invalid framebuffer id.\n"); + VMW_DEBUG_USER("Invalid framebuffer id.\n"); ret = -ENOENT; goto out_no_fb; } vfb = vmw_framebuffer_to_vfb(fb); if (!vfb->bo) { - DRM_ERROR("Framebuffer not buffer backed.\n"); + VMW_DEBUG_USER("Framebuffer not buffer backed.\n"); ret = -EINVAL; goto out_no_ttm_lock; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index ed2f67822f45..b97bc8e5944b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -64,11 +64,9 @@ static int vmw_cursor_update_image(struct vmw_private *dev_priv, if (!image) return -EINVAL; - cmd = vmw_fifo_reserve(dev_priv, cmd_size); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Fifo reserve failed.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, cmd_size); + if (unlikely(cmd == NULL)) return -ENOMEM; - } memset(cmd, 0, sizeof(*cmd)); @@ -1202,7 +1200,7 @@ static int vmw_create_bo_proxy(struct drm_device *dev, vmw_bo_unreference(&res->backup); res->backup = vmw_bo_reference(bo_mob); res->backup_offset = 0; - vmw_resource_unreserve(res, false, NULL, 0); + vmw_resource_unreserve(res, false, false, false, NULL, 0); mutex_unlock(&res->dev_priv->cmdbuf_mutex); return 0; @@ -2468,13 +2466,11 @@ int vmw_kms_helper_dirty(struct vmw_private *dev_priv, dirty->unit = unit; if (dirty->fifo_reserve_size > 0) { - dirty->cmd = vmw_fifo_reserve(dev_priv, + dirty->cmd = VMW_FIFO_RESERVE(dev_priv, dirty->fifo_reserve_size); - if (!dirty->cmd) { - DRM_ERROR("Couldn't reserve fifo space " - "for dirty blits.\n"); + if (!dirty->cmd) return -ENOMEM; - } + memset(dirty->cmd, 0, dirty->fifo_reserve_size); } dirty->num_hits = 0; @@ -2604,12 +2600,9 @@ int vmw_kms_update_proxy(struct vmw_resource *res, if (!clips) return 0; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd) * num_clips); - if (!cmd) { - DRM_ERROR("Couldn't reserve fifo space for proxy surface " - "update.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd) * num_clips); + if (!cmd) return -ENOMEM; - } for (i = 0; i < num_clips; ++i, clips += increment, ++cmd) { box = &cmd->body.box; @@ -2827,7 +2820,8 @@ int vmw_du_helper_plane_update(struct vmw_du_update_plane *update) container_of(update->vfb, typeof(*vfbs), base); ret = vmw_validation_add_resource(&val_ctx, &vfbs->surface->res, - 0, NULL, NULL); + 0, VMW_RES_DIRTY_NONE, NULL, + NULL); } if (ret) @@ -2838,7 +2832,7 @@ int vmw_du_helper_plane_update(struct vmw_du_update_plane *update) goto out_unref; reserved_size = update->calc_fifo_size(update, num_hits); - cmd_start = vmw_fifo_reserve(update->dev_priv, reserved_size); + cmd_start = VMW_FIFO_RESERVE(update->dev_priv, reserved_size); if (!cmd_start) { ret = -ENOMEM; goto out_revert; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c index 16be515c4c0f..25e6343bcf21 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c @@ -554,11 +554,9 @@ int vmw_kms_ldu_do_bo_dirty(struct vmw_private *dev_priv, } *cmd; fifo_size = sizeof(*cmd) * num_clips; - cmd = vmw_fifo_reserve(dev_priv, fifo_size); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Fifo reserve failed.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, fifo_size); + if (unlikely(cmd == NULL)) return -ENOMEM; - } memset(cmd, 0, fifo_size); for (i = 0; i < num_clips; i++, clips += increment) { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c index d83cc66e1210..406edc8cef35 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c @@ -146,9 +146,8 @@ static int vmw_setup_otable_base(struct vmw_private *dev_priv, mob->pt_level += VMW_MOBFMT_PTDEPTH_1 - SVGA3D_MOBFMT_PTDEPTH_1; } - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for OTable setup.\n"); ret = -ENOMEM; goto out_no_fifo; } @@ -202,12 +201,9 @@ static void vmw_takedown_otable_base(struct vmw_private *dev_priv, return; bo = otable->page_table->pt_bo; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for OTable " - "takedown.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return; - } memset(cmd, 0, sizeof(*cmd)); cmd->header.id = SVGA_3D_CMD_SET_OTABLE_BASE; @@ -614,16 +610,14 @@ void vmw_mob_unbind(struct vmw_private *dev_priv, BUG_ON(ret != 0); } - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for Memory " - "Object unbinding.\n"); - } else { + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (cmd) { cmd->header.id = SVGA_3D_CMD_DESTROY_GB_MOB; cmd->header.size = sizeof(cmd->body); cmd->body.mobid = mob->id; vmw_fifo_commit(dev_priv, sizeof(*cmd)); } + if (bo) { vmw_bo_fence_single(bo, NULL); ttm_bo_unreserve(bo); @@ -683,12 +677,9 @@ int vmw_mob_bind(struct vmw_private *dev_priv, vmw_fifo_resource_inc(dev_priv); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for Memory " - "Object binding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) goto out_no_cmd_space; - } cmd->header.id = SVGA_3D_CMD_DEFINE_GB_MOB64; cmd->header.size = sizeof(cmd->body); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c index 9f1b9d289bec..d5ef8cf802de 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c @@ -124,7 +124,7 @@ static int vmw_overlay_send_put(struct vmw_private *dev_priv, fifo_size = sizeof(*cmds) + sizeof(*flush) + sizeof(*items) * num_items; - cmds = vmw_fifo_reserve(dev_priv, fifo_size); + cmds = VMW_FIFO_RESERVE(dev_priv, fifo_size); /* hardware has hung, can't do anything here */ if (!cmds) return -ENOMEM; @@ -194,7 +194,7 @@ static int vmw_overlay_send_stop(struct vmw_private *dev_priv, int ret; for (;;) { - cmds = vmw_fifo_reserve(dev_priv, sizeof(*cmds)); + cmds = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmds)); if (cmds) break; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c index a7c30e567f09..711f8fd0dd45 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c @@ -365,14 +365,6 @@ static int vmw_resource_do_validate(struct vmw_resource *res, list_add_tail(&res->mob_head, &res->backup->res_list); } - /* - * Only do this on write operations, and move to - * vmw_resource_unreserve if it can be called after - * backup buffers have been unreserved. Otherwise - * sort out locking. - */ - res->res_dirty = true; - return 0; out_bind_failed: @@ -386,6 +378,8 @@ out_bind_failed: * command submission. * * @res: Pointer to the struct vmw_resource to unreserve. + * @dirty_set: Change dirty status of the resource. + * @dirty: When changing dirty status indicates the new status. * @switch_backup: Backup buffer has been switched. * @new_backup: Pointer to new backup buffer if command submission * switched. May be NULL. @@ -395,6 +389,8 @@ out_bind_failed: * resource lru list, so that it can be evicted if necessary. */ void vmw_resource_unreserve(struct vmw_resource *res, + bool dirty_set, + bool dirty, bool switch_backup, struct vmw_buffer_object *new_backup, unsigned long new_backup_offset) @@ -422,6 +418,9 @@ void vmw_resource_unreserve(struct vmw_resource *res, if (switch_backup) res->backup_offset = new_backup_offset; + if (dirty_set) + res->res_dirty = dirty; + if (!res->func->may_evict || res->id == -1 || res->pin_count) return; @@ -696,7 +695,7 @@ void vmw_resource_unbind_list(struct vmw_buffer_object *vbo) if (!res->func->unbind) continue; - (void) res->func->unbind(res, true, &val_buf); + (void) res->func->unbind(res, res->res_dirty, &val_buf); res->backup_dirty = true; res->res_dirty = false; list_del_init(&res->mob_head); @@ -731,12 +730,9 @@ int vmw_query_readback_all(struct vmw_buffer_object *dx_query_mob) dx_query_ctx = dx_query_mob->dx_query_ctx; dev_priv = dx_query_ctx->dev_priv; - cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), dx_query_ctx->id); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for " - "query MOB read back.\n"); + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), dx_query_ctx->id); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_READBACK_ALL_QUERY; cmd->header.size = sizeof(cmd->body); @@ -932,7 +928,7 @@ int vmw_resource_pin(struct vmw_resource *res, bool interruptible) res->pin_count++; out_no_validate: - vmw_resource_unreserve(res, false, NULL, 0UL); + vmw_resource_unreserve(res, false, false, false, NULL, 0UL); out_no_reserve: mutex_unlock(&dev_priv->cmdbuf_mutex); ttm_write_unlock(&dev_priv->reservation_sem); @@ -968,7 +964,7 @@ void vmw_resource_unpin(struct vmw_resource *res) ttm_bo_unreserve(&vbo->base); } - vmw_resource_unreserve(res, false, NULL, 0UL); + vmw_resource_unreserve(res, false, false, false, NULL, 0UL); mutex_unlock(&dev_priv->cmdbuf_mutex); ttm_read_unlock(&dev_priv->reservation_sem); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c index cd586c52af7e..9a2a3836d89a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c @@ -130,12 +130,9 @@ static int vmw_sou_fifo_create(struct vmw_private *dev_priv, BUG_ON(!sou->buffer); fifo_size = sizeof(*cmd); - cmd = vmw_fifo_reserve(dev_priv, fifo_size); - /* The hardware has hung, nothing we can do about it here. */ - if (unlikely(cmd == NULL)) { - DRM_ERROR("Fifo reserve failed.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, fifo_size); + if (unlikely(cmd == NULL)) return -ENOMEM; - } memset(cmd, 0, fifo_size); cmd->header.cmdType = SVGA_CMD_DEFINE_SCREEN; @@ -182,12 +179,9 @@ static int vmw_sou_fifo_destroy(struct vmw_private *dev_priv, return 0; fifo_size = sizeof(*cmd); - cmd = vmw_fifo_reserve(dev_priv, fifo_size); - /* the hardware has hung, nothing we can do about it here */ - if (unlikely(cmd == NULL)) { - DRM_ERROR("Fifo reserve failed.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, fifo_size); + if (unlikely(cmd == NULL)) return -ENOMEM; - } memset(cmd, 0, fifo_size); cmd->header.cmdType = SVGA_CMD_DESTROY_SCREEN; @@ -998,11 +992,9 @@ static int do_bo_define_gmrfb(struct vmw_private *dev_priv, if (depth == 32) depth = 24; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (!cmd) { - DRM_ERROR("Out of fifo space for dirty framebuffer command.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (!cmd) return -ENOMEM; - } cmd->header = SVGA_CMD_DEFINE_GMRFB; cmd->body.format.bitsPerPixel = framebuffer->base.format->cpp[0] * 8; @@ -1148,7 +1140,8 @@ int vmw_kms_sou_do_surface_dirty(struct vmw_private *dev_priv, if (!srf) srf = &vfbs->surface->res; - ret = vmw_validation_add_resource(&val_ctx, srf, 0, NULL, NULL); + ret = vmw_validation_add_resource(&val_ctx, srf, 0, VMW_RES_DIRTY_NONE, + NULL, NULL); if (ret) return ret; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c index bf32fe446219..d310d21f0d54 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c @@ -218,10 +218,8 @@ static int vmw_gb_shader_create(struct vmw_resource *res) goto out_no_fifo; } - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for shader " - "creation.\n"); ret = -ENOMEM; goto out_no_fifo; } @@ -254,12 +252,9 @@ static int vmw_gb_shader_bind(struct vmw_resource *res, BUG_ON(bo->mem.mem_type != VMW_PL_MOB); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for shader " - "binding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_BIND_GB_SHADER; cmd->header.size = sizeof(cmd->body); @@ -285,12 +280,9 @@ static int vmw_gb_shader_unbind(struct vmw_resource *res, BUG_ON(res->backup->base.mem.mem_type != VMW_PL_MOB); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for shader " - "unbinding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_BIND_GB_SHADER; cmd->header.size = sizeof(cmd->body); @@ -328,10 +320,8 @@ static int vmw_gb_shader_destroy(struct vmw_resource *res) mutex_lock(&dev_priv->binding_mutex); vmw_binding_res_list_scrub(&res->binding_head); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for shader " - "destruction.\n"); mutex_unlock(&dev_priv->binding_mutex); return -ENOMEM; } @@ -400,13 +390,9 @@ static int vmw_dx_shader_unscrub(struct vmw_resource *res) if (!list_empty(&shader->cotable_head) || !shader->committed) return 0; - cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), - shader->ctx->id); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for shader " - "scrubbing.\n"); + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), shader->ctx->id); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_BIND_SHADER; cmd->header.size = sizeof(cmd->body); @@ -491,12 +477,9 @@ static int vmw_dx_shader_scrub(struct vmw_resource *res) return 0; WARN_ON_ONCE(!shader->committed); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { - DRM_ERROR("Failed reserving FIFO space for shader " - "scrubbing.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DX_BIND_SHADER; cmd->header.size = sizeof(cmd->body); @@ -865,14 +848,13 @@ static int vmw_shader_define(struct drm_device *dev, struct drm_file *file_priv, ret = vmw_user_bo_lookup(tfile, buffer_handle, &buffer, NULL); if (unlikely(ret != 0)) { - DRM_ERROR("Could not find buffer for shader " - "creation.\n"); + VMW_DEBUG_USER("Couldn't find buffer for shader creation.\n"); return ret; } if ((u64)buffer->base.num_pages * PAGE_SIZE < (u64)size + (u64)offset) { - DRM_ERROR("Illegal buffer- or shader size.\n"); + VMW_DEBUG_USER("Illegal buffer- or shader size.\n"); ret = -EINVAL; goto out_bad_arg; } @@ -886,7 +868,7 @@ static int vmw_shader_define(struct drm_device *dev, struct drm_file *file_priv, shader_type = SVGA3D_SHADERTYPE_PS; break; default: - DRM_ERROR("Illegal shader type.\n"); + VMW_DEBUG_USER("Illegal shader type.\n"); ret = -EINVAL; goto out_bad_arg; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_simple_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_simple_resource.c index 6a6865384e91..73e9a487e659 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_simple_resource.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_simple_resource.c @@ -239,17 +239,17 @@ vmw_simple_resource_lookup(struct ttm_object_file *tfile, base = ttm_base_object_lookup(tfile, handle); if (!base) { - DRM_ERROR("Invalid %s handle 0x%08lx.\n", - func->res_func.type_name, - (unsigned long) handle); + VMW_DEBUG_USER("Invalid %s handle 0x%08lx.\n", + func->res_func.type_name, + (unsigned long) handle); return ERR_PTR(-ESRCH); } if (ttm_base_object_type(base) != func->ttm_res_type) { ttm_base_object_unref(&base); - DRM_ERROR("Invalid type of %s handle 0x%08lx.\n", - func->res_func.type_name, - (unsigned long) handle); + VMW_DEBUG_USER("Invalid type of %s handle 0x%08lx.\n", + func->res_func.type_name, + (unsigned long) handle); return ERR_PTR(-EINVAL); } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_so.c b/drivers/gpu/drm/vmwgfx/vmwgfx_so.c index bc8bb690f1ea..63807361e16f 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_so.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_so.c @@ -170,13 +170,12 @@ static int vmw_view_create(struct vmw_resource *res) return 0; } - cmd = vmw_fifo_reserve_dx(res->dev_priv, view->cmd_size, - view->ctx->id); + cmd = VMW_FIFO_RESERVE_DX(res->dev_priv, view->cmd_size, view->ctx->id); if (!cmd) { - DRM_ERROR("Failed reserving FIFO space for view creation.\n"); mutex_unlock(&dev_priv->binding_mutex); return -ENOMEM; } + memcpy(cmd, &view->cmd, view->cmd_size); WARN_ON(cmd->body.view_id != view->view_id); /* Sid may have changed due to surface eviction. */ @@ -214,12 +213,9 @@ static int vmw_view_destroy(struct vmw_resource *res) if (!view->committed || res->id == -1) return 0; - cmd = vmw_fifo_reserve_dx(dev_priv, sizeof(*cmd), view->ctx->id); - if (!cmd) { - DRM_ERROR("Failed reserving FIFO space for view " - "destruction.\n"); + cmd = VMW_FIFO_RESERVE_DX(dev_priv, sizeof(*cmd), view->ctx->id); + if (!cmd) return -ENOMEM; - } cmd->header.id = vmw_view_destroy_cmds[view->view_type]; cmd->header.size = sizeof(cmd->body); @@ -338,12 +334,12 @@ int vmw_view_add(struct vmw_cmdbuf_res_manager *man, if (cmd_size != vmw_view_define_sizes[view_type] + sizeof(SVGA3dCmdHeader)) { - DRM_ERROR("Illegal view create command size.\n"); + VMW_DEBUG_USER("Illegal view create command size.\n"); return -EINVAL; } if (!vmw_view_id_ok(user_key, view_type)) { - DRM_ERROR("Illegal view add view id.\n"); + VMW_DEBUG_USER("Illegal view add view id.\n"); return -EINVAL; } @@ -352,8 +348,7 @@ int vmw_view_add(struct vmw_cmdbuf_res_manager *man, ret = ttm_mem_global_alloc(vmw_mem_glob(dev_priv), size, &ttm_opt_ctx); if (ret) { if (ret != -ERESTARTSYS) - DRM_ERROR("Out of graphics memory for view" - " creation.\n"); + DRM_ERROR("Out of graphics memory for view creation\n"); return ret; } @@ -413,7 +408,7 @@ int vmw_view_remove(struct vmw_cmdbuf_res_manager *man, struct vmw_resource **res_p) { if (!vmw_view_id_ok(user_key, view_type)) { - DRM_ERROR("Illegal view remove view id.\n"); + VMW_DEBUG_USER("Illegal view remove view id.\n"); return -EINVAL; } @@ -497,6 +492,30 @@ struct vmw_resource *vmw_view_lookup(struct vmw_cmdbuf_res_manager *man, vmw_view_key(user_key, view_type)); } +/** + * vmw_view_dirtying - Return whether a view type is dirtying its resource + * @res: Pointer to the view + * + * Each time a resource is put on the validation list as the result of a + * view pointing to it, we need to determine whether that resource will + * be dirtied (written to by the GPU) as a result of the corresponding + * GPU operation. Currently only rendertarget- and depth-stencil views are + * capable of dirtying its resource. + * + * Return: Whether the view type of @res dirties the resource it points to. + */ +u32 vmw_view_dirtying(struct vmw_resource *res) +{ + static u32 view_is_dirtying[vmw_view_max] = { + [vmw_view_rt] = VMW_RES_DIRTY_SET, + [vmw_view_ds] = VMW_RES_DIRTY_SET, + }; + + /* Update this function as we add more view types */ + BUILD_BUG_ON(vmw_view_max != 3); + return view_is_dirtying[vmw_view(res)->view_type]; +} + const u32 vmw_view_destroy_cmds[] = { [vmw_view_sr] = SVGA_3D_CMD_DX_DESTROY_SHADERRESOURCE_VIEW, [vmw_view_rt] = SVGA_3D_CMD_DX_DESTROY_RENDERTARGET_VIEW, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_so.h b/drivers/gpu/drm/vmwgfx/vmwgfx_so.h index b80c7252f2fd..12565047bc55 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_so.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_so.h @@ -157,4 +157,5 @@ extern struct vmw_resource *vmw_view_srf(struct vmw_resource *res); extern struct vmw_resource *vmw_view_lookup(struct vmw_cmdbuf_res_manager *man, enum vmw_view_type view_type, u32 user_key); +extern u32 vmw_view_dirtying(struct vmw_resource *res); #endif diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c index 096c2941a8e4..f803bb5e782b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c @@ -111,7 +111,7 @@ struct vmw_stdu_update_gb_image { */ struct vmw_screen_target_display_unit { struct vmw_display_unit base; - const struct vmw_surface *display_srf; + struct vmw_surface *display_srf; enum stdu_content_type content_fb_type; s32 display_width, display_height; @@ -167,12 +167,9 @@ static int vmw_stdu_define_st(struct vmw_private *dev_priv, SVGA3dCmdDefineGBScreenTarget body; } *cmd; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - - if (unlikely(cmd == NULL)) { - DRM_ERROR("Out of FIFO space defining Screen Target\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DEFINE_GB_SCREENTARGET; cmd->header.size = sizeof(cmd->body); @@ -229,12 +226,9 @@ static int vmw_stdu_bind_st(struct vmw_private *dev_priv, memset(&image, 0, sizeof(image)); image.sid = res ? res->id : SVGA3D_INVALID_ID; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - - if (unlikely(cmd == NULL)) { - DRM_ERROR("Out of FIFO space binding a screen target\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_BIND_GB_SCREENTARGET; cmd->header.size = sizeof(cmd->body); @@ -296,12 +290,9 @@ static int vmw_stdu_update_st(struct vmw_private *dev_priv, return -EINVAL; } - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - - if (unlikely(cmd == NULL)) { - DRM_ERROR("Out of FIFO space updating a Screen Target\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } vmw_stdu_populate_update(cmd, stdu->base.unit, 0, stdu->display_width, @@ -335,12 +326,9 @@ static int vmw_stdu_destroy_st(struct vmw_private *dev_priv, if (unlikely(!stdu->defined)) return 0; - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - - if (unlikely(cmd == NULL)) { - DRM_ERROR("Out of FIFO space, screen target not destroyed\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (unlikely(cmd == NULL)) return -ENOMEM; - } cmd->header.id = SVGA_3D_CMD_DESTROY_GB_SCREENTARGET; cmd->header.size = sizeof(cmd->body); @@ -533,6 +521,7 @@ static void vmw_stdu_bo_fifo_commit(struct vmw_kms_dirty *dirty) vmw_fifo_commit(dirty->dev_priv, sizeof(*cmd) + blit_size); + stdu->display_srf->res.res_dirty = true; ddirty->left = ddirty->top = S32_MAX; ddirty->right = ddirty->bottom = S32_MIN; } @@ -629,20 +618,16 @@ static void vmw_stdu_bo_cpu_commit(struct vmw_kms_dirty *dirty) region.x2 = diff.rect.x2; region.y1 = diff.rect.y1; region.y2 = diff.rect.y2; - ret = vmw_kms_update_proxy( - (struct vmw_resource *) &stdu->display_srf->res, - (const struct drm_clip_rect *) ®ion, 1, 1); + ret = vmw_kms_update_proxy(&stdu->display_srf->res, ®ion, + 1, 1); if (ret) goto out_cleanup; dev_priv = vmw_priv(stdu->base.crtc.dev); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - - if (!cmd) { - DRM_ERROR("Cannot reserve FIFO space to update STDU"); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); + if (!cmd) goto out_cleanup; - } vmw_stdu_populate_update(cmd, stdu->base.unit, region.x1, region.x2, @@ -820,6 +805,7 @@ static void vmw_kms_stdu_surface_fifo_commit(struct vmw_kms_dirty *dirty) cmd->body.dest.sid = stdu->display_srf->res.id; update = (struct vmw_stdu_update *) &blit[dirty->num_hits]; commit_size = sizeof(*cmd) + blit_size + sizeof(*update); + stdu->display_srf->res.res_dirty = true; } else { update = dirty->cmd; commit_size = sizeof(*update); @@ -876,7 +862,8 @@ int vmw_kms_stdu_surface_dirty(struct vmw_private *dev_priv, if (!srf) srf = &vfbs->surface->res; - ret = vmw_validation_add_resource(&val_ctx, srf, 0, NULL, NULL); + ret = vmw_validation_add_resource(&val_ctx, srf, 0, VMW_RES_DIRTY_NONE, + NULL, NULL); if (ret) return ret; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c index ef09f7edf931..219471903bc1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c @@ -342,12 +342,9 @@ static void vmw_hw_surface_destroy(struct vmw_resource *res) if (res->id != -1) { - cmd = vmw_fifo_reserve(dev_priv, vmw_surface_destroy_size()); - if (unlikely(!cmd)) { - DRM_ERROR("Failed reserving FIFO space for surface " - "destruction.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, vmw_surface_destroy_size()); + if (unlikely(!cmd)) return; - } vmw_surface_destroy_encode(res->id, cmd); vmw_fifo_commit(dev_priv, vmw_surface_destroy_size()); @@ -414,10 +411,8 @@ static int vmw_legacy_srf_create(struct vmw_resource *res) */ submit_size = vmw_surface_define_size(srf); - cmd = vmw_fifo_reserve(dev_priv, submit_size); + cmd = VMW_FIFO_RESERVE(dev_priv, submit_size); if (unlikely(!cmd)) { - DRM_ERROR("Failed reserving FIFO space for surface " - "creation.\n"); ret = -ENOMEM; goto out_no_fifo; } @@ -468,12 +463,10 @@ static int vmw_legacy_srf_dma(struct vmw_resource *res, BUG_ON(!val_buf->bo); submit_size = vmw_surface_dma_size(srf); - cmd = vmw_fifo_reserve(dev_priv, submit_size); - if (unlikely(!cmd)) { - DRM_ERROR("Failed reserving FIFO space for surface " - "DMA.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, submit_size); + if (unlikely(!cmd)) return -ENOMEM; - } + vmw_bo_get_guest_ptr(val_buf->bo, &ptr); vmw_surface_dma_encode(srf, cmd, &ptr, bind); @@ -556,12 +549,9 @@ static int vmw_legacy_srf_destroy(struct vmw_resource *res) */ submit_size = vmw_surface_destroy_size(); - cmd = vmw_fifo_reserve(dev_priv, submit_size); - if (unlikely(!cmd)) { - DRM_ERROR("Failed reserving FIFO space for surface " - "eviction.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, submit_size); + if (unlikely(!cmd)) return -ENOMEM; - } vmw_surface_destroy_encode(res->id, cmd); vmw_fifo_commit(dev_priv, submit_size); @@ -748,11 +738,10 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, ttm_round_pot(num_sizes * sizeof(struct drm_vmw_size)) + ttm_round_pot(num_sizes * sizeof(struct vmw_surface_offset)); - desc = svga3dsurface_get_desc(req->format); if (unlikely(desc->block_desc == SVGA3DBLOCKDESC_NONE)) { - DRM_ERROR("Invalid surface format for surface creation.\n"); - DRM_ERROR("Format requested is: %d\n", req->format); + VMW_DEBUG_USER("Invalid format %d for surface creation.\n", + req->format); return -EINVAL; } @@ -764,8 +753,7 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, size, &ctx); if (unlikely(ret != 0)) { if (ret != -ERESTARTSYS) - DRM_ERROR("Out of graphics memory for surface" - " creation.\n"); + DRM_ERROR("Out of graphics memory for surface.\n"); goto out_unlock; } @@ -939,12 +927,12 @@ vmw_surface_handle_reference(struct vmw_private *dev_priv, ret = -EINVAL; base = ttm_base_object_lookup_for_ref(dev_priv->tdev, handle); if (unlikely(!base)) { - DRM_ERROR("Could not find surface to reference.\n"); + VMW_DEBUG_USER("Could not find surface to reference.\n"); goto out_no_lookup; } if (unlikely(ttm_base_object_type(base) != VMW_RES_SURFACE)) { - DRM_ERROR("Referenced object is not a surface.\n"); + VMW_DEBUG_USER("Referenced object is not a surface.\n"); goto out_bad_resource; } @@ -1022,8 +1010,8 @@ int vmw_surface_reference_ioctl(struct drm_device *dev, void *data, ret = copy_to_user(user_sizes, &srf->base_size, sizeof(srf->base_size)); if (unlikely(ret != 0)) { - DRM_ERROR("copy_to_user failed %p %u\n", - user_sizes, srf->num_sizes); + VMW_DEBUG_USER("copy_to_user failed %p %u\n", user_sizes, + srf->num_sizes); ttm_ref_object_base_unref(tfile, base->handle, TTM_REF_USAGE); ret = -EFAULT; } @@ -1088,12 +1076,10 @@ static int vmw_gb_surface_create(struct vmw_resource *res) submit_len = sizeof(*cmd); } - cmd = vmw_fifo_reserve(dev_priv, submit_len); + cmd = VMW_FIFO_RESERVE(dev_priv, submit_len); cmd2 = (typeof(cmd2))cmd; cmd3 = (typeof(cmd3))cmd; if (unlikely(!cmd)) { - DRM_ERROR("Failed reserving FIFO space for surface " - "creation.\n"); ret = -ENOMEM; goto out_no_fifo; } @@ -1171,12 +1157,9 @@ static int vmw_gb_surface_bind(struct vmw_resource *res, submit_size = sizeof(*cmd1) + (res->backup_dirty ? sizeof(*cmd2) : 0); - cmd1 = vmw_fifo_reserve(dev_priv, submit_size); - if (unlikely(!cmd1)) { - DRM_ERROR("Failed reserving FIFO space for surface " - "binding.\n"); + cmd1 = VMW_FIFO_RESERVE(dev_priv, submit_size); + if (unlikely(!cmd1)) return -ENOMEM; - } cmd1->header.id = SVGA_3D_CMD_BIND_GB_SURFACE; cmd1->header.size = sizeof(cmd1->body); @@ -1221,12 +1204,9 @@ static int vmw_gb_surface_unbind(struct vmw_resource *res, BUG_ON(bo->mem.mem_type != VMW_PL_MOB); submit_size = sizeof(*cmd3) + (readback ? sizeof(*cmd1) : sizeof(*cmd2)); - cmd = vmw_fifo_reserve(dev_priv, submit_size); - if (unlikely(!cmd)) { - DRM_ERROR("Failed reserving FIFO space for surface " - "unbinding.\n"); + cmd = VMW_FIFO_RESERVE(dev_priv, submit_size); + if (unlikely(!cmd)) return -ENOMEM; - } if (readback) { cmd1 = (void *) cmd; @@ -1280,10 +1260,8 @@ static int vmw_gb_surface_destroy(struct vmw_resource *res) vmw_view_surface_list_destroy(dev_priv, &srf->view_list); vmw_binding_res_list_scrub(&res->binding_head); - cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); + cmd = VMW_FIFO_RESERVE(dev_priv, sizeof(*cmd)); if (unlikely(!cmd)) { - DRM_ERROR("Failed reserving FIFO space for surface " - "destruction.\n"); mutex_unlock(&dev_priv->binding_mutex); return -ENOMEM; } @@ -1405,16 +1383,16 @@ int vmw_surface_gb_priv_define(struct drm_device *dev, if (for_scanout) { if (!svga3dsurface_is_screen_target_format(format)) { - DRM_ERROR("Invalid Screen Target surface format."); + VMW_DEBUG_USER("Invalid Screen Target surface format."); return -EINVAL; } if (size.width > dev_priv->texture_max_width || size.height > dev_priv->texture_max_height) { - DRM_ERROR("%ux%u\n, exceeds max surface size %ux%u", - size.width, size.height, - dev_priv->texture_max_width, - dev_priv->texture_max_height); + VMW_DEBUG_USER("%ux%u\n, exceeds max surface size %ux%u", + size.width, size.height, + dev_priv->texture_max_width, + dev_priv->texture_max_height); return -EINVAL; } } else { @@ -1422,14 +1400,14 @@ int vmw_surface_gb_priv_define(struct drm_device *dev, desc = svga3dsurface_get_desc(format); if (unlikely(desc->block_desc == SVGA3DBLOCKDESC_NONE)) { - DRM_ERROR("Invalid surface format.\n"); + VMW_DEBUG_USER("Invalid surface format.\n"); return -EINVAL; } } /* array_size must be null for non-GL3 host. */ if (array_size > 0 && !dev_priv->has_dx) { - DRM_ERROR("Tried to create DX surface on non-DX host.\n"); + VMW_DEBUG_USER("Tried to create DX surface on non-DX host.\n"); return -EINVAL; } @@ -1651,7 +1629,7 @@ vmw_gb_surface_define_internal(struct drm_device *dev, if (ret == 0) { if (res->backup->base.num_pages * PAGE_SIZE < res->backup_size) { - DRM_ERROR("Surface backup buffer too small.\n"); + VMW_DEBUG_USER("Surface backup buffer too small.\n"); vmw_bo_unreference(&res->backup); ret = -EINVAL; goto out_unlock; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c index e9944ac2e057..f611b2290a1b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c @@ -76,6 +76,8 @@ struct vmw_validation_res_node { u32 switching_backup : 1; u32 first_usage : 1; u32 reserved : 1; + u32 dirty : 1; + u32 dirty_set : 1; unsigned long private[0]; }; @@ -299,6 +301,7 @@ int vmw_validation_add_bo(struct vmw_validation_context *ctx, * @ctx: The validation context. * @res: The resource. * @priv_size: Size of private, additional metadata. + * @dirty: Whether to change dirty status. * @p_node: Output pointer of additional metadata address. * @first_usage: Whether this was the first time this resource was seen. * @@ -307,6 +310,7 @@ int vmw_validation_add_bo(struct vmw_validation_context *ctx, int vmw_validation_add_resource(struct vmw_validation_context *ctx, struct vmw_resource *res, size_t priv_size, + u32 dirty, void **p_node, bool *first_usage) { @@ -321,8 +325,7 @@ int vmw_validation_add_resource(struct vmw_validation_context *ctx, node = vmw_validation_mem_alloc(ctx, sizeof(*node) + priv_size); if (!node) { - DRM_ERROR("Failed to allocate a resource validation " - "entry.\n"); + VMW_DEBUG_USER("Failed to allocate a resource validation entry.\n"); return -ENOMEM; } @@ -358,6 +361,11 @@ int vmw_validation_add_resource(struct vmw_validation_context *ctx, } out_fill: + if (dirty) { + node->dirty_set = 1; + /* Overwriting previous information here is intentional! */ + node->dirty = (dirty & VMW_RES_DIRTY_SET) ? 1 : 0; + } if (first_usage) *first_usage = node->first_usage; if (p_node) @@ -367,6 +375,29 @@ out_fill: } /** + * vmw_validation_res_set_dirty - Register a resource dirty set or clear during + * validation. + * @ctx: The validation context. + * @val_private: The additional meta-data pointer returned when the + * resource was registered with the validation context. Used to identify + * the resource. + * @dirty: Dirty information VMW_RES_DIRTY_XX + */ +void vmw_validation_res_set_dirty(struct vmw_validation_context *ctx, + void *val_private, u32 dirty) +{ + struct vmw_validation_res_node *val; + + if (!dirty) + return; + + val = container_of(val_private, typeof(*val), private); + val->dirty_set = 1; + /* Overwriting previous information here is intentional! */ + val->dirty = (dirty & VMW_RES_DIRTY_SET) ? 1 : 0; +} + +/** * vmw_validation_res_switch_backup - Register a backup MOB switch during * validation. * @ctx: The validation context. @@ -450,15 +481,23 @@ void vmw_validation_res_unreserve(struct vmw_validation_context *ctx, struct vmw_validation_res_node *val; list_splice_init(&ctx->resource_ctx_list, &ctx->resource_list); - - list_for_each_entry(val, &ctx->resource_list, head) { - if (val->reserved) - vmw_resource_unreserve(val->res, - !backoff && - val->switching_backup, - val->new_backup, - val->new_backup_offset); - } + if (backoff) + list_for_each_entry(val, &ctx->resource_list, head) { + if (val->reserved) + vmw_resource_unreserve(val->res, + false, false, false, + NULL, 0); + } + else + list_for_each_entry(val, &ctx->resource_list, head) { + if (val->reserved) + vmw_resource_unreserve(val->res, + val->dirty_set, + val->dirty, + val->switching_backup, + val->new_backup, + val->new_backup_offset); + } } /** diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h index 3b396fea40d7..523f6ac5c335 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h @@ -33,6 +33,10 @@ #include <linux/ww_mutex.h> #include <drm/ttm/ttm_execbuf_util.h> +#define VMW_RES_DIRTY_NONE 0 +#define VMW_RES_DIRTY_SET BIT(0) +#define VMW_RES_DIRTY_CLEAR BIT(1) + /** * struct vmw_validation_mem - Custom interface to provide memory reservations * for the validation code. @@ -237,6 +241,7 @@ void vmw_validation_unref_lists(struct vmw_validation_context *ctx); int vmw_validation_add_resource(struct vmw_validation_context *ctx, struct vmw_resource *res, size_t priv_size, + u32 dirty, void **p_node, bool *first_usage); void vmw_validation_drop_ht(struct vmw_validation_context *ctx); @@ -261,4 +266,6 @@ void *vmw_validation_mem_alloc(struct vmw_validation_context *ctx, int vmw_validation_preload_bo(struct vmw_validation_context *ctx); int vmw_validation_preload_res(struct vmw_validation_context *ctx, unsigned int size); +void vmw_validation_res_set_dirty(struct vmw_validation_context *ctx, + void *val_private, u32 dirty); #endif diff --git a/drivers/gpu/host1x/hw/channel_hw.c b/drivers/gpu/host1x/hw/channel_hw.c index 27101c04a827..4030d64916f0 100644 --- a/drivers/gpu/host1x/hw/channel_hw.c +++ b/drivers/gpu/host1x/hw/channel_hw.c @@ -114,7 +114,7 @@ static inline void synchronize_syncpt_base(struct host1x_job *job) static void host1x_channel_set_streamid(struct host1x_channel *channel) { -#if HOST1X_HW >= 6 +#if IS_ENABLED(CONFIG_IOMMU_API) && HOST1X_HW >= 6 struct iommu_fwspec *spec = dev_iommu_fwspec_get(channel->dev->parent); u32 sid = spec ? spec->ids[0] & 0xffff : 0x7f; |
