diff options
Diffstat (limited to 'drivers/gpu/drm')
50 files changed, 771 insertions, 444 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 104b2e0d893b..b0fc116296cb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -233,7 +233,7 @@ enum amdgpu_kiq_irq { #define MAX_KIQ_REG_WAIT 5000 /* in usecs, 5ms */ #define MAX_KIQ_REG_BAILOUT_INTERVAL 5 /* in msecs, 5ms */ -#define MAX_KIQ_REG_TRY 20 +#define MAX_KIQ_REG_TRY 80 /* 20 -> 80 */ int amdgpu_device_ip_set_clockgating_state(void *dev, enum amd_ip_block_type block_type, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index f9b54236102d..95f4c4139fc6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -39,6 +39,7 @@ const unsigned int amdgpu_ctx_num_entities[AMDGPU_HW_IP_NUM] = { [AMDGPU_HW_IP_UVD_ENC] = 1, [AMDGPU_HW_IP_VCN_DEC] = 1, [AMDGPU_HW_IP_VCN_ENC] = 1, + [AMDGPU_HW_IP_VCN_JPEG] = 1, }; static int amdgput_ctx_total_num_entities(void) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 81732a84c2ab..8f3d44e5e787 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -467,9 +467,6 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file if (!info->return_size || !info->return_pointer) return -EINVAL; - /* Ensure IB tests are run on ring */ - flush_delayed_work(&adev->late_init_work); - switch (info->query) { case AMDGPU_INFO_ACCEL_WORKING: ui32 = adev->accel_working; @@ -950,6 +947,9 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) struct amdgpu_fpriv *fpriv; int r, pasid; + /* Ensure IB tests are run on ring */ + flush_delayed_work(&adev->late_init_work); + file_priv->driver_priv = NULL; r = pm_runtime_get_sync(dev->dev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index dad0e2342df9..0877ff9a9594 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -181,7 +181,7 @@ static unsigned amdgpu_vm_num_entries(struct amdgpu_device *adev, if (level == adev->vm_manager.root_level) /* For the root directory */ - return round_up(adev->vm_manager.max_pfn, 1 << shift) >> shift; + return round_up(adev->vm_manager.max_pfn, 1ULL << shift) >> shift; else if (level != AMDGPU_VM_PTB) /* Everything in between */ return 512; @@ -1656,9 +1656,11 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, if (!amdgpu_vm_pt_descendant(adev, &cursor)) return -ENOENT; continue; - } else if (frag >= parent_shift) { + } else if (frag >= parent_shift && + cursor.level - 1 != adev->vm_manager.root_level) { /* If the fragment size is even larger than the parent - * shift we should go up one level and check it again. + * shift we should go up one level and check it again + * unless one level up is the root level. */ if (!amdgpu_vm_pt_ancestor(&cursor)) return -ENOENT; @@ -1666,10 +1668,10 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, } /* Looks good so far, calculate parameters for the update */ - incr = AMDGPU_GPU_PAGE_SIZE << shift; + incr = (uint64_t)AMDGPU_GPU_PAGE_SIZE << shift; mask = amdgpu_vm_entries_mask(adev, cursor.level); pe_start = ((cursor.pfn >> shift) & mask) * 8; - entry_end = (mask + 1) << shift; + entry_end = (uint64_t)(mask + 1) << shift; entry_end += cursor.pfn & ~(entry_end - 1); entry_end = min(entry_end, end); @@ -1682,7 +1684,7 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, flags | AMDGPU_PTE_FRAG(frag)); pe_start += nptes * 8; - dst += nptes * AMDGPU_GPU_PAGE_SIZE << shift; + dst += (uint64_t)nptes * AMDGPU_GPU_PAGE_SIZE << shift; frag_start = upd_end; if (frag_start >= frag_end) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 6d7baf59d6e1..21363b2b2ee5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -2440,12 +2440,13 @@ static void gfx_v9_0_rlc_start(struct amdgpu_device *adev) #endif WREG32_FIELD15(GC, 0, RLC_CNTL, RLC_ENABLE_F32, 1); + udelay(50); /* carrizo do enable cp interrupt after cp inited */ - if (!(adev->flags & AMD_IS_APU)) + if (!(adev->flags & AMD_IS_APU)) { gfx_v9_0_enable_gui_idle_interrupt(adev, true); - - udelay(50); + udelay(50); + } #ifdef AMDGPU_RLC_DEBUG_RETRY /* RLC_GPM_GENERAL_6 : RLC Ucode version */ diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 1d3265c97b70..747c068379dc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -56,6 +56,9 @@ MODULE_FIRMWARE("amdgpu/tonga_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris11_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris10_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris12_mc.bin"); +MODULE_FIRMWARE("amdgpu/polaris11_k_mc.bin"); +MODULE_FIRMWARE("amdgpu/polaris10_k_mc.bin"); +MODULE_FIRMWARE("amdgpu/polaris12_k_mc.bin"); static const u32 golden_settings_tonga_a11[] = { @@ -224,13 +227,39 @@ static int gmc_v8_0_init_microcode(struct amdgpu_device *adev) chip_name = "tonga"; break; case CHIP_POLARIS11: - chip_name = "polaris11"; + if (((adev->pdev->device == 0x67ef) && + ((adev->pdev->revision == 0xe0) || + (adev->pdev->revision == 0xe5))) || + ((adev->pdev->device == 0x67ff) && + ((adev->pdev->revision == 0xcf) || + (adev->pdev->revision == 0xef) || + (adev->pdev->revision == 0xff)))) + chip_name = "polaris11_k"; + else if ((adev->pdev->device == 0x67ef) && + (adev->pdev->revision == 0xe2)) + chip_name = "polaris11_k"; + else + chip_name = "polaris11"; break; case CHIP_POLARIS10: - chip_name = "polaris10"; + if ((adev->pdev->device == 0x67df) && + ((adev->pdev->revision == 0xe1) || + (adev->pdev->revision == 0xf7))) + chip_name = "polaris10_k"; + else + chip_name = "polaris10"; break; case CHIP_POLARIS12: - chip_name = "polaris12"; + if (((adev->pdev->device == 0x6987) && + ((adev->pdev->revision == 0xc0) || + (adev->pdev->revision == 0xc3))) || + ((adev->pdev->device == 0x6981) && + ((adev->pdev->revision == 0x00) || + (adev->pdev->revision == 0x01) || + (adev->pdev->revision == 0x10)))) + chip_name = "polaris12_k"; + else + chip_name = "polaris12"; break; case CHIP_FIJI: case CHIP_CARRIZO: @@ -337,7 +366,7 @@ static int gmc_v8_0_polaris_mc_load_microcode(struct amdgpu_device *adev) const struct mc_firmware_header_v1_0 *hdr; const __le32 *fw_data = NULL; const __le32 *io_mc_regs = NULL; - u32 data, vbios_version; + u32 data; int i, ucode_size, regs_size; /* Skip MC ucode loading on SR-IOV capable boards. @@ -348,13 +377,6 @@ static int gmc_v8_0_polaris_mc_load_microcode(struct amdgpu_device *adev) if (amdgpu_sriov_bios(adev)) return 0; - WREG32(mmMC_SEQ_IO_DEBUG_INDEX, 0x9F); - data = RREG32(mmMC_SEQ_IO_DEBUG_DATA); - vbios_version = data & 0xf; - - if (vbios_version == 0) - return 0; - if (!adev->gmc.fw) return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index eae90922fdbe..322e09b5b448 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -48,6 +48,7 @@ static void vcn_v1_0_set_enc_ring_funcs(struct amdgpu_device *adev); static void vcn_v1_0_set_jpeg_ring_funcs(struct amdgpu_device *adev); static void vcn_v1_0_set_irq_funcs(struct amdgpu_device *adev); static void vcn_v1_0_jpeg_ring_set_patch_ring(struct amdgpu_ring *ring, uint32_t ptr); +static int vcn_v1_0_set_powergating_state(void *handle, enum amd_powergating_state state); /** * vcn_v1_0_early_init - set function pointers @@ -222,7 +223,7 @@ static int vcn_v1_0_hw_fini(void *handle) struct amdgpu_ring *ring = &adev->vcn.ring_dec; if (RREG32_SOC15(VCN, 0, mmUVD_STATUS)) - vcn_v1_0_stop(adev); + vcn_v1_0_set_powergating_state(adev, AMD_PG_STATE_GATE); ring->ready = false; 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 ca925200fe09..5a6edf65c9ea 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -2554,9 +2554,9 @@ static void fill_audio_info(struct audio_info *audio_info, cea_revision = drm_connector->display_info.cea_rev; - strncpy(audio_info->display_name, + strscpy(audio_info->display_name, edid_caps->display_name, - AUDIO_INFO_DISPLAY_NAME_SIZE_IN_CHARS - 1); + AUDIO_INFO_DISPLAY_NAME_SIZE_IN_CHARS); if (cea_revision >= 3) { audio_info->mode_count = edid_caps->audio_mode_count; @@ -3042,6 +3042,7 @@ void amdgpu_dm_connector_funcs_reset(struct drm_connector *connector) state->underscan_enable = false; state->underscan_hborder = 0; state->underscan_vborder = 0; + state->max_bpc = 8; __drm_atomic_helper_connector_reset(connector, &state->base); } @@ -3063,6 +3064,7 @@ amdgpu_dm_connector_atomic_duplicate_state(struct drm_connector *connector) new_state->freesync_capable = state->freesync_capable; new_state->freesync_enable = state->freesync_enable; + new_state->max_bpc = state->max_bpc; return &new_state->base; } @@ -3650,7 +3652,7 @@ amdgpu_dm_create_common_mode(struct drm_encoder *encoder, mode->hdisplay = hdisplay; mode->vdisplay = vdisplay; mode->type &= ~DRM_MODE_TYPE_PREFERRED; - strncpy(mode->name, name, DRM_DISPLAY_MODE_LEN); + strscpy(mode->name, name, DRM_DISPLAY_MODE_LEN); return mode; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index d02c32a1039c..1b0d209d8367 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -342,10 +342,9 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, master->connector_id); aconnector->mst_encoder = dm_dp_create_fake_mst_encoder(master); + drm_connector_attach_encoder(&aconnector->base, + &aconnector->mst_encoder->base); - /* - * TODO: understand why this one is needed - */ drm_object_attach_property( &connector->base, dev->mode_config.path_property, diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index b459867a05b2..a6bcb90e8419 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -2512,6 +2512,8 @@ static void pplib_apply_display_requirements( dc, context->bw.dce.sclk_khz); + pp_display_cfg->min_dcfclock_khz = pp_display_cfg->min_engine_clock_khz; + pp_display_cfg->min_engine_clock_deep_sleep_khz = context->bw.dce.sclk_deep_sleep_khz; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c index 85119c2bdcc8..a2a7e0e94aa6 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c @@ -80,7 +80,9 @@ int phm_enable_dynamic_state_management(struct pp_hwmgr *hwmgr) PHM_FUNC_CHECK(hwmgr); adev = hwmgr->adev; - if (smum_is_dpm_running(hwmgr) && !amdgpu_passthrough(adev)) { + /* Skip for suspend/resume case */ + if (smum_is_dpm_running(hwmgr) && !amdgpu_passthrough(adev) + && adev->in_suspend) { pr_info("dpm has been enabled\n"); return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c index 47ac92369739..0173d0480024 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c @@ -352,6 +352,9 @@ int hwmgr_handle_task(struct pp_hwmgr *hwmgr, enum amd_pp_task task_id, switch (task_id) { case AMD_PP_TASK_DISPLAY_CONFIG_CHANGE: + ret = phm_pre_display_configuration_changed(hwmgr); + if (ret) + return ret; ret = phm_set_cpu_power_state(hwmgr); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c index 91ffb7bc4ee7..56437866d120 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c @@ -265,8 +265,6 @@ int psm_adjust_power_state_dynamic(struct pp_hwmgr *hwmgr, bool skip, if (skip) return 0; - phm_pre_display_configuration_changed(hwmgr); - phm_display_configuration_changed(hwmgr); if (hwmgr->ps) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c index 88f6b35ea6fe..b61a01f55284 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c @@ -3589,8 +3589,10 @@ static int smu7_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, cons } if (i >= sclk_table->count) { - data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; - sclk_table->dpm_levels[i-1].value = sclk; + if (sclk > sclk_table->dpm_levels[i-1].value) { + data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; + sclk_table->dpm_levels[i-1].value = sclk; + } } else { /* TODO: Check SCLK in DAL's minimum clocks * in case DeepSleep divider update is required. @@ -3607,8 +3609,10 @@ static int smu7_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, cons } if (i >= mclk_table->count) { - data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; - mclk_table->dpm_levels[i-1].value = mclk; + if (mclk > mclk_table->dpm_levels[i-1].value) { + data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; + mclk_table->dpm_levels[i-1].value = mclk; + } } if (data->display_timing.num_existing_displays != hwmgr->display_config->num_display) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c index e2bc6e0c229f..79c86247d0ac 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c @@ -3266,8 +3266,10 @@ static int vega10_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, co } if (i >= sclk_table->count) { - data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; - sclk_table->dpm_levels[i-1].value = sclk; + if (sclk > sclk_table->dpm_levels[i-1].value) { + data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; + sclk_table->dpm_levels[i-1].value = sclk; + } } for (i = 0; i < mclk_table->count; i++) { @@ -3276,8 +3278,10 @@ static int vega10_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, co } if (i >= mclk_table->count) { - data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; - mclk_table->dpm_levels[i-1].value = mclk; + if (mclk > mclk_table->dpm_levels[i-1].value) { + data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; + mclk_table->dpm_levels[i-1].value = mclk; + } } if (data->display_timing.num_existing_displays != hwmgr->display_config->num_display) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c index b4eadd47f3a4..3367dd30cdd0 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c @@ -1660,14 +1660,15 @@ static uint32_t vega20_find_highest_dpm_level( return i; } -static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) +static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_mask) { struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend); uint32_t min_freq; int ret = 0; - if (data->smu_features[GNLD_DPM_GFXCLK].enabled) { + if (data->smu_features[GNLD_DPM_GFXCLK].enabled && + (feature_mask & FEATURE_DPM_GFXCLK_MASK)) { min_freq = data->dpm_table.gfx_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, @@ -1676,7 +1677,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UCLK].enabled) { + if (data->smu_features[GNLD_DPM_UCLK].enabled && + (feature_mask & FEATURE_DPM_UCLK_MASK)) { min_freq = data->dpm_table.mem_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, @@ -1692,7 +1694,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UVD].enabled) { + if (data->smu_features[GNLD_DPM_UVD].enabled && + (feature_mask & FEATURE_DPM_UVD_MASK)) { min_freq = data->dpm_table.vclk_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1710,7 +1713,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_VCE].enabled) { + if (data->smu_features[GNLD_DPM_VCE].enabled && + (feature_mask & FEATURE_DPM_VCE_MASK)) { min_freq = data->dpm_table.eclk_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1720,7 +1724,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_SOCCLK].enabled) { + if (data->smu_features[GNLD_DPM_SOCCLK].enabled && + (feature_mask & FEATURE_DPM_SOCCLK_MASK)) { min_freq = data->dpm_table.soc_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1733,14 +1738,15 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret; } -static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) +static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_mask) { struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend); uint32_t max_freq; int ret = 0; - if (data->smu_features[GNLD_DPM_GFXCLK].enabled) { + if (data->smu_features[GNLD_DPM_GFXCLK].enabled && + (feature_mask & FEATURE_DPM_GFXCLK_MASK)) { max_freq = data->dpm_table.gfx_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1750,7 +1756,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UCLK].enabled) { + if (data->smu_features[GNLD_DPM_UCLK].enabled && + (feature_mask & FEATURE_DPM_UCLK_MASK)) { max_freq = data->dpm_table.mem_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1760,7 +1767,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UVD].enabled) { + if (data->smu_features[GNLD_DPM_UVD].enabled && + (feature_mask & FEATURE_DPM_UVD_MASK)) { max_freq = data->dpm_table.vclk_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1777,7 +1785,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_VCE].enabled) { + if (data->smu_features[GNLD_DPM_VCE].enabled && + (feature_mask & FEATURE_DPM_VCE_MASK)) { max_freq = data->dpm_table.eclk_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1787,7 +1796,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_SOCCLK].enabled) { + if (data->smu_features[GNLD_DPM_SOCCLK].enabled && + (feature_mask & FEATURE_DPM_SOCCLK_MASK)) { max_freq = data->dpm_table.soc_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -2126,12 +2136,12 @@ static int vega20_force_dpm_highest(struct pp_hwmgr *hwmgr) data->dpm_table.mem_table.dpm_state.soft_max_level = data->dpm_table.mem_table.dpm_levels[soft_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to highest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2158,12 +2168,12 @@ static int vega20_force_dpm_lowest(struct pp_hwmgr *hwmgr) data->dpm_table.mem_table.dpm_state.soft_max_level = data->dpm_table.mem_table.dpm_levels[soft_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to highest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2176,12 +2186,12 @@ static int vega20_unforce_dpm_levels(struct pp_hwmgr *hwmgr) { int ret = 0; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload DPM Bootup Levels!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload DPM Max Levels!", return ret); @@ -2239,12 +2249,12 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr, data->dpm_table.gfx_table.dpm_state.soft_max_level = data->dpm_table.gfx_table.dpm_levels[soft_max_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, FEATURE_DPM_GFXCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to lowest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, FEATURE_DPM_GFXCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2259,12 +2269,12 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr, data->dpm_table.mem_table.dpm_state.soft_max_level = data->dpm_table.mem_table.dpm_levels[soft_max_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, FEATURE_DPM_UCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to lowest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, FEATURE_DPM_UCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); diff --git a/drivers/gpu/drm/ast/ast_fb.c b/drivers/gpu/drm/ast/ast_fb.c index 0cd827e11fa2..de26df0c6044 100644 --- a/drivers/gpu/drm/ast/ast_fb.c +++ b/drivers/gpu/drm/ast/ast_fb.c @@ -263,6 +263,7 @@ static void ast_fbdev_destroy(struct drm_device *dev, { struct ast_framebuffer *afb = &afbdev->afb; + drm_crtc_force_disable_all(dev); drm_fb_helper_unregister_fbi(&afbdev->helper); if (afb->obj) { diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index dac355812adc..373700c05a00 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -583,7 +583,8 @@ void ast_driver_unload(struct drm_device *dev) drm_mode_config_cleanup(dev); ast_mm_fini(ast); - pci_iounmap(dev->pdev, ast->ioregs); + if (ast->ioregs != ast->regs + AST_IO_MM_OFFSET) + pci_iounmap(dev->pdev, ast->ioregs); pci_iounmap(dev->pdev, ast->regs); kfree(ast); } diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 7c6ac3cadb6b..8bb355d5d43d 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -973,9 +973,21 @@ static int get_clock(void *i2c_priv) { struct ast_i2c_chan *i2c = i2c_priv; struct ast_private *ast = i2c->dev->dev_private; - uint32_t val; + uint32_t val, val2, count, pass; + + count = 0; + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01; + do { + val2 = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01; + if (val == val2) { + pass++; + } else { + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01; + } + } while ((pass < 5) && (count++ < 0x10000)); - val = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4; return val & 1 ? 1 : 0; } @@ -983,9 +995,21 @@ static int get_data(void *i2c_priv) { struct ast_i2c_chan *i2c = i2c_priv; struct ast_private *ast = i2c->dev->dev_private; - uint32_t val; + uint32_t val, val2, count, pass; + + count = 0; + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01; + do { + val2 = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01; + if (val == val2) { + pass++; + } else { + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01; + } + } while ((pass < 5) && (count++ < 0x10000)); - val = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5; return val & 1 ? 1 : 0; } @@ -998,7 +1022,7 @@ static void set_clock(void *i2c_priv, int clock) for (i = 0; i < 0x10000; i++) { ujcrb7 = ((clock & 0x01) ? 0 : 1); - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xfe, ujcrb7); + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xf4, ujcrb7); jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x01); if (ujcrb7 == jtemp) break; @@ -1014,7 +1038,7 @@ static void set_data(void *i2c_priv, int data) for (i = 0; i < 0x10000; i++) { ujcrb7 = ((data & 0x01) ? 0 : 1) << 2; - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xfb, ujcrb7); + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xf1, ujcrb7); jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x04); if (ujcrb7 == jtemp) break; diff --git a/drivers/gpu/drm/i915/gvt/aperture_gm.c b/drivers/gpu/drm/i915/gvt/aperture_gm.c index fe754022e356..359d37d5c958 100644 --- a/drivers/gpu/drm/i915/gvt/aperture_gm.c +++ b/drivers/gpu/drm/i915/gvt/aperture_gm.c @@ -61,10 +61,12 @@ static int alloc_gm(struct intel_vgpu *vgpu, bool high_gm) } mutex_lock(&dev_priv->drm.struct_mutex); + mmio_hw_access_pre(dev_priv); ret = i915_gem_gtt_insert(&dev_priv->ggtt.vm, node, size, I915_GTT_PAGE_SIZE, I915_COLOR_UNEVICTABLE, start, end, flags); + mmio_hw_access_post(dev_priv); mutex_unlock(&dev_priv->drm.struct_mutex); if (ret) gvt_err("fail to alloc %s gm space from host\n", diff --git a/drivers/gpu/drm/i915/gvt/fb_decoder.c b/drivers/gpu/drm/i915/gvt/fb_decoder.c index 481896fb712a..85e6736f0a32 100644 --- a/drivers/gpu/drm/i915/gvt/fb_decoder.c +++ b/drivers/gpu/drm/i915/gvt/fb_decoder.c @@ -235,7 +235,7 @@ int intel_vgpu_decode_primary_plane(struct intel_vgpu *vgpu, plane->bpp = skl_pixel_formats[fmt].bpp; plane->drm_format = skl_pixel_formats[fmt].drm_format; } else { - plane->tiled = !!(val & DISPPLANE_TILED); + plane->tiled = val & DISPPLANE_TILED; fmt = bdw_format_to_drm(val & DISPPLANE_PIXFORMAT_MASK); plane->bpp = bdw_pixel_formats[fmt].bpp; plane->drm_format = bdw_pixel_formats[fmt].drm_format; diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index 58e166effa45..c7103dd2d8d5 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -2447,10 +2447,11 @@ static void intel_vgpu_destroy_all_ppgtt_mm(struct intel_vgpu *vgpu) static void intel_vgpu_destroy_ggtt_mm(struct intel_vgpu *vgpu) { - struct intel_gvt_partial_pte *pos; + struct intel_gvt_partial_pte *pos, *next; - list_for_each_entry(pos, - &vgpu->gtt.ggtt_mm->ggtt_mm.partial_pte_list, list) { + list_for_each_entry_safe(pos, next, + &vgpu->gtt.ggtt_mm->ggtt_mm.partial_pte_list, + list) { gvt_dbg_mm("partial PTE update on hold 0x%lx : 0x%llx\n", pos->offset, pos->data); kfree(pos); diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c index 36a5147cd01e..d6e02c15ef97 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.c +++ b/drivers/gpu/drm/i915/gvt/mmio_context.c @@ -158,6 +158,8 @@ static void load_render_mocs(struct drm_i915_private *dev_priv) int ring_id, i; for (ring_id = 0; ring_id < ARRAY_SIZE(regs); ring_id++) { + if (!HAS_ENGINE(dev_priv, ring_id)) + continue; offset.reg = regs[ring_id]; for (i = 0; i < GEN9_MOCS_SIZE; i++) { gen9_render_mocs.control_table[ring_id][i] = diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index ffdbbac4400e..47062ee979cf 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -1444,6 +1444,7 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv) intel_uncore_sanitize(dev_priv); + intel_gt_init_workarounds(dev_priv); i915_gem_load_init_fences(dev_priv); /* On the 945G/GM, the chipset reports the MSI capability on the diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 9102571e9692..872a2e159a5f 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -67,6 +67,7 @@ #include "intel_ringbuffer.h" #include "intel_uncore.h" #include "intel_wopcm.h" +#include "intel_workarounds.h" #include "intel_uc.h" #include "i915_gem.h" @@ -1805,6 +1806,7 @@ struct drm_i915_private { int dpio_phy_iosf_port[I915_NUM_PHYS_VLV]; struct i915_workarounds workarounds; + struct i915_wa_list gt_wa_list; struct i915_frontbuffer_tracking fb_tracking; @@ -2148,6 +2150,8 @@ struct drm_i915_private { struct delayed_work idle_work; ktime_t last_init_time; + + struct i915_vma *scratch; } gt; /* perform PHY state sanity checks? */ @@ -3870,4 +3874,9 @@ static inline int intel_hws_csb_write_index(struct drm_i915_private *i915) return I915_HWS_CSB_WRITE_INDEX; } +static inline u32 i915_scratch_offset(const struct drm_i915_private *i915) +{ + return i915_ggtt_offset(i915->gt.scratch); +} + #endif diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 0c8aa57ce83b..6ae9a6080cc8 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -5305,7 +5305,7 @@ int i915_gem_init_hw(struct drm_i915_private *dev_priv) } } - intel_gt_workarounds_apply(dev_priv); + intel_gt_apply_workarounds(dev_priv); i915_gem_init_swizzling(dev_priv); @@ -5500,6 +5500,44 @@ err_active: goto out_ctx; } +static int +i915_gem_init_scratch(struct drm_i915_private *i915, unsigned int size) +{ + struct drm_i915_gem_object *obj; + struct i915_vma *vma; + int ret; + + obj = i915_gem_object_create_stolen(i915, size); + if (!obj) + obj = i915_gem_object_create_internal(i915, size); + if (IS_ERR(obj)) { + DRM_ERROR("Failed to allocate scratch page\n"); + return PTR_ERR(obj); + } + + vma = i915_vma_instance(obj, &i915->ggtt.vm, NULL); + if (IS_ERR(vma)) { + ret = PTR_ERR(vma); + goto err_unref; + } + + ret = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH); + if (ret) + goto err_unref; + + i915->gt.scratch = vma; + return 0; + +err_unref: + i915_gem_object_put(obj); + return ret; +} + +static void i915_gem_fini_scratch(struct drm_i915_private *i915) +{ + i915_vma_unpin_and_release(&i915->gt.scratch, 0); +} + int i915_gem_init(struct drm_i915_private *dev_priv) { int ret; @@ -5546,12 +5584,19 @@ int i915_gem_init(struct drm_i915_private *dev_priv) goto err_unlock; } - ret = i915_gem_contexts_init(dev_priv); + ret = i915_gem_init_scratch(dev_priv, + IS_GEN2(dev_priv) ? SZ_256K : PAGE_SIZE); if (ret) { GEM_BUG_ON(ret == -EIO); goto err_ggtt; } + ret = i915_gem_contexts_init(dev_priv); + if (ret) { + GEM_BUG_ON(ret == -EIO); + goto err_scratch; + } + ret = intel_engines_init(dev_priv); if (ret) { GEM_BUG_ON(ret == -EIO); @@ -5624,6 +5669,8 @@ err_pm: err_context: if (ret != -EIO) i915_gem_contexts_fini(dev_priv); +err_scratch: + i915_gem_fini_scratch(dev_priv); err_ggtt: err_unlock: intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL); @@ -5675,8 +5722,11 @@ void i915_gem_fini(struct drm_i915_private *dev_priv) intel_uc_fini(dev_priv); i915_gem_cleanup_engines(dev_priv); i915_gem_contexts_fini(dev_priv); + i915_gem_fini_scratch(dev_priv); mutex_unlock(&dev_priv->drm.struct_mutex); + intel_wa_list_free(&dev_priv->gt_wa_list); + intel_cleanup_gt_powersave(dev_priv); intel_uc_fini_misc(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index d4fac09095f8..1aaccbe7e1de 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -1268,7 +1268,7 @@ relocate_entry(struct i915_vma *vma, else if (gen >= 4) len = 4; else - len = 6; + len = 3; batch = reloc_gpu(eb, vma, len); if (IS_ERR(batch)) @@ -1309,11 +1309,6 @@ relocate_entry(struct i915_vma *vma, *batch++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL; *batch++ = addr; *batch++ = target_offset; - - /* And again for good measure (blb/pnv) */ - *batch++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL; - *batch++ = addr; - *batch++ = target_offset; } goto out; diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 3eb33e000d6f..db4128d6c09b 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -1495,7 +1495,7 @@ static void gem_record_rings(struct i915_gpu_state *error) if (HAS_BROKEN_CS_TLB(i915)) ee->wa_batchbuffer = i915_error_object_create(i915, - engine->scratch); + i915->gt.scratch); request_record_user_bo(request, ee); ee->ctx = diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c index 217ed3ee1cab..76b5f94ea6cb 100644 --- a/drivers/gpu/drm/i915/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/intel_engine_cs.c @@ -490,46 +490,6 @@ void intel_engine_setup_common(struct intel_engine_cs *engine) intel_engine_init_cmd_parser(engine); } -int intel_engine_create_scratch(struct intel_engine_cs *engine, - unsigned int size) -{ - struct drm_i915_gem_object *obj; - struct i915_vma *vma; - int ret; - - WARN_ON(engine->scratch); - - obj = i915_gem_object_create_stolen(engine->i915, size); - if (!obj) - obj = i915_gem_object_create_internal(engine->i915, size); - if (IS_ERR(obj)) { - DRM_ERROR("Failed to allocate scratch page\n"); - return PTR_ERR(obj); - } - - vma = i915_vma_instance(obj, &engine->i915->ggtt.vm, NULL); - if (IS_ERR(vma)) { - ret = PTR_ERR(vma); - goto err_unref; - } - - ret = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH); - if (ret) - goto err_unref; - - engine->scratch = vma; - return 0; - -err_unref: - i915_gem_object_put(obj); - return ret; -} - -void intel_engine_cleanup_scratch(struct intel_engine_cs *engine) -{ - i915_vma_unpin_and_release(&engine->scratch, 0); -} - static void cleanup_status_page(struct intel_engine_cs *engine) { if (HWS_NEEDS_PHYSICAL(engine->i915)) { @@ -704,8 +664,6 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine) { struct drm_i915_private *i915 = engine->i915; - intel_engine_cleanup_scratch(engine); - cleanup_status_page(engine); intel_engine_fini_breadcrumbs(engine); @@ -720,6 +678,8 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine) __intel_context_unpin(i915->kernel_context, engine); i915_timeline_fini(&engine->timeline); + + intel_wa_list_free(&engine->wa_list); } u64 intel_engine_get_active_head(const struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 37c94a54efcb..58d1d3d47dd3 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -442,8 +442,13 @@ static u64 execlists_update_context(struct i915_request *rq) * may not be visible to the HW prior to the completion of the UC * register write and that we may begin execution from the context * before its image is complete leading to invalid PD chasing. + * + * Furthermore, Braswell, at least, wants a full mb to be sure that + * the writes are coherent in memory (visible to the GPU) prior to + * execution, and not just visible to other CPUs (as is the result of + * wmb). */ - wmb(); + mb(); return ce->lrc_desc; } @@ -1443,9 +1448,10 @@ static int execlists_request_alloc(struct i915_request *request) static u32 * gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *engine, u32 *batch) { + /* NB no one else is allowed to scribble over scratch + 256! */ *batch++ = MI_STORE_REGISTER_MEM_GEN8 | MI_SRM_LRM_GLOBAL_GTT; *batch++ = i915_mmio_reg_offset(GEN8_L3SQCREG4); - *batch++ = i915_ggtt_offset(engine->scratch) + 256; + *batch++ = i915_scratch_offset(engine->i915) + 256; *batch++ = 0; *batch++ = MI_LOAD_REGISTER_IMM(1); @@ -1459,7 +1465,7 @@ gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *engine, u32 *batch) *batch++ = MI_LOAD_REGISTER_MEM_GEN8 | MI_SRM_LRM_GLOBAL_GTT; *batch++ = i915_mmio_reg_offset(GEN8_L3SQCREG4); - *batch++ = i915_ggtt_offset(engine->scratch) + 256; + *batch++ = i915_scratch_offset(engine->i915) + 256; *batch++ = 0; return batch; @@ -1496,7 +1502,7 @@ static u32 *gen8_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch) PIPE_CONTROL_GLOBAL_GTT_IVB | PIPE_CONTROL_CS_STALL | PIPE_CONTROL_QW_WRITE, - i915_ggtt_offset(engine->scratch) + + i915_scratch_offset(engine->i915) + 2 * CACHELINE_BYTES); *batch++ = MI_ARB_ON_OFF | MI_ARB_ENABLE; @@ -1573,7 +1579,7 @@ static u32 *gen9_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch) PIPE_CONTROL_GLOBAL_GTT_IVB | PIPE_CONTROL_CS_STALL | PIPE_CONTROL_QW_WRITE, - i915_ggtt_offset(engine->scratch) + i915_scratch_offset(engine->i915) + 2 * CACHELINE_BYTES); } @@ -1793,6 +1799,8 @@ static bool unexpected_starting_state(struct intel_engine_cs *engine) static int gen8_init_common_ring(struct intel_engine_cs *engine) { + intel_engine_apply_workarounds(engine); + intel_mocs_init_engine(engine); intel_engine_reset_breadcrumbs(engine); @@ -2139,7 +2147,7 @@ static int gen8_emit_flush_render(struct i915_request *request, { struct intel_engine_cs *engine = request->engine; u32 scratch_addr = - i915_ggtt_offset(engine->scratch) + 2 * CACHELINE_BYTES; + i915_scratch_offset(engine->i915) + 2 * CACHELINE_BYTES; bool vf_flush_wa = false, dc_flush_wa = false; u32 *cs, flags = 0; int len; @@ -2476,10 +2484,6 @@ int logical_render_ring_init(struct intel_engine_cs *engine) if (ret) return ret; - ret = intel_engine_create_scratch(engine, PAGE_SIZE); - if (ret) - goto err_cleanup_common; - ret = intel_init_workaround_bb(engine); if (ret) { /* @@ -2491,11 +2495,9 @@ int logical_render_ring_init(struct intel_engine_cs *engine) ret); } - return 0; + intel_engine_init_workarounds(engine); -err_cleanup_common: - intel_engine_cleanup_common(engine); - return ret; + return 0; } int logical_xcs_ring_init(struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 187bb0ceb4ac..1f8d2a66c791 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -69,19 +69,28 @@ unsigned int intel_ring_update_space(struct intel_ring *ring) static int gen2_render_ring_flush(struct i915_request *rq, u32 mode) { + unsigned int num_store_dw; u32 cmd, *cs; cmd = MI_FLUSH; - + num_store_dw = 0; if (mode & EMIT_INVALIDATE) cmd |= MI_READ_FLUSH; + if (mode & EMIT_FLUSH) + num_store_dw = 4; - cs = intel_ring_begin(rq, 2); + cs = intel_ring_begin(rq, 2 + 3 * num_store_dw); if (IS_ERR(cs)) return PTR_ERR(cs); *cs++ = cmd; - *cs++ = MI_NOOP; + while (num_store_dw--) { + *cs++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL; + *cs++ = i915_scratch_offset(rq->i915); + *cs++ = 0; + } + *cs++ = MI_FLUSH | MI_NO_WRITE_FLUSH; + intel_ring_advance(rq, cs); return 0; @@ -150,8 +159,7 @@ gen4_render_ring_flush(struct i915_request *rq, u32 mode) */ if (mode & EMIT_INVALIDATE) { *cs++ = GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE; - *cs++ = i915_ggtt_offset(rq->engine->scratch) | - PIPE_CONTROL_GLOBAL_GTT; + *cs++ = i915_scratch_offset(rq->i915) | PIPE_CONTROL_GLOBAL_GTT; *cs++ = 0; *cs++ = 0; @@ -159,8 +167,7 @@ gen4_render_ring_flush(struct i915_request *rq, u32 mode) *cs++ = MI_FLUSH; *cs++ = GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE; - *cs++ = i915_ggtt_offset(rq->engine->scratch) | - PIPE_CONTROL_GLOBAL_GTT; + *cs++ = i915_scratch_offset(rq->i915) | PIPE_CONTROL_GLOBAL_GTT; *cs++ = 0; *cs++ = 0; } @@ -212,8 +219,7 @@ gen4_render_ring_flush(struct i915_request *rq, u32 mode) static int intel_emit_post_sync_nonzero_flush(struct i915_request *rq) { - u32 scratch_addr = - i915_ggtt_offset(rq->engine->scratch) + 2 * CACHELINE_BYTES; + u32 scratch_addr = i915_scratch_offset(rq->i915) + 2 * CACHELINE_BYTES; u32 *cs; cs = intel_ring_begin(rq, 6); @@ -246,8 +252,7 @@ intel_emit_post_sync_nonzero_flush(struct i915_request *rq) static int gen6_render_ring_flush(struct i915_request *rq, u32 mode) { - u32 scratch_addr = - i915_ggtt_offset(rq->engine->scratch) + 2 * CACHELINE_BYTES; + u32 scratch_addr = i915_scratch_offset(rq->i915) + 2 * CACHELINE_BYTES; u32 *cs, flags = 0; int ret; @@ -316,8 +321,7 @@ gen7_render_ring_cs_stall_wa(struct i915_request *rq) static int gen7_render_ring_flush(struct i915_request *rq, u32 mode) { - u32 scratch_addr = - i915_ggtt_offset(rq->engine->scratch) + 2 * CACHELINE_BYTES; + u32 scratch_addr = i915_scratch_offset(rq->i915) + 2 * CACHELINE_BYTES; u32 *cs, flags = 0; /* @@ -971,7 +975,7 @@ i965_emit_bb_start(struct i915_request *rq, } /* Just userspace ABI convention to limit the wa batch bo to a resonable size */ -#define I830_BATCH_LIMIT (256*1024) +#define I830_BATCH_LIMIT SZ_256K #define I830_TLB_ENTRIES (2) #define I830_WA_SIZE max(I830_TLB_ENTRIES*4096, I830_BATCH_LIMIT) static int @@ -979,7 +983,9 @@ i830_emit_bb_start(struct i915_request *rq, u64 offset, u32 len, unsigned int dispatch_flags) { - u32 *cs, cs_offset = i915_ggtt_offset(rq->engine->scratch); + u32 *cs, cs_offset = i915_scratch_offset(rq->i915); + + GEM_BUG_ON(rq->i915->gt.scratch->size < I830_WA_SIZE); cs = intel_ring_begin(rq, 6); if (IS_ERR(cs)) @@ -1437,7 +1443,6 @@ static int intel_init_ring_buffer(struct intel_engine_cs *engine) { struct i915_timeline *timeline; struct intel_ring *ring; - unsigned int size; int err; intel_engine_setup_common(engine); @@ -1462,21 +1467,12 @@ static int intel_init_ring_buffer(struct intel_engine_cs *engine) GEM_BUG_ON(engine->buffer); engine->buffer = ring; - size = PAGE_SIZE; - if (HAS_BROKEN_CS_TLB(engine->i915)) - size = I830_WA_SIZE; - err = intel_engine_create_scratch(engine, size); - if (err) - goto err_unpin; - err = intel_engine_init_common(engine); if (err) - goto err_scratch; + goto err_unpin; return 0; -err_scratch: - intel_engine_cleanup_scratch(engine); err_unpin: intel_ring_unpin(ring); err_ring: @@ -1550,7 +1546,7 @@ static int flush_pd_dir(struct i915_request *rq) /* Stall until the page table load is complete */ *cs++ = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT; *cs++ = i915_mmio_reg_offset(RING_PP_DIR_BASE(engine)); - *cs++ = i915_ggtt_offset(engine->scratch); + *cs++ = i915_scratch_offset(rq->i915); *cs++ = MI_NOOP; intel_ring_advance(rq, cs); @@ -1659,7 +1655,7 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags) /* Insert a delay before the next switch! */ *cs++ = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT; *cs++ = i915_mmio_reg_offset(last_reg); - *cs++ = i915_ggtt_offset(engine->scratch); + *cs++ = i915_scratch_offset(rq->i915); *cs++ = MI_NOOP; } *cs++ = MI_ARB_ON_OFF | MI_ARB_ENABLE; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 2dfa585712c2..767a7192c969 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -15,6 +15,7 @@ #include "i915_selftest.h" #include "i915_timeline.h" #include "intel_gpu_commands.h" +#include "intel_workarounds.h" struct drm_printer; struct i915_sched_attr; @@ -440,7 +441,7 @@ struct intel_engine_cs { struct intel_hw_status_page status_page; struct i915_ctx_workarounds wa_ctx; - struct i915_vma *scratch; + struct i915_wa_list wa_list; u32 irq_keep_mask; /* always keep these interrupts */ u32 irq_enable_mask; /* bitmask to enable ring interrupt */ @@ -898,10 +899,6 @@ void intel_engine_setup_common(struct intel_engine_cs *engine); int intel_engine_init_common(struct intel_engine_cs *engine); void intel_engine_cleanup_common(struct intel_engine_cs *engine); -int intel_engine_create_scratch(struct intel_engine_cs *engine, - unsigned int size); -void intel_engine_cleanup_scratch(struct intel_engine_cs *engine); - int intel_init_render_ring_buffer(struct intel_engine_cs *engine); int intel_init_bsd_ring_buffer(struct intel_engine_cs *engine); int intel_init_blt_ring_buffer(struct intel_engine_cs *engine); diff --git a/drivers/gpu/drm/i915/intel_workarounds.c b/drivers/gpu/drm/i915/intel_workarounds.c index 4bcdeaf8d98f..6e580891db96 100644 --- a/drivers/gpu/drm/i915/intel_workarounds.c +++ b/drivers/gpu/drm/i915/intel_workarounds.c @@ -48,6 +48,20 @@ * - Public functions to init or apply the given workaround type. */ +static void wa_init_start(struct i915_wa_list *wal, const char *name) +{ + wal->name = name; +} + +static void wa_init_finish(struct i915_wa_list *wal) +{ + if (!wal->count) + return; + + DRM_DEBUG_DRIVER("Initialized %u %s workarounds\n", + wal->count, wal->name); +} + static void wa_add(struct drm_i915_private *i915, i915_reg_t reg, const u32 mask, const u32 val) { @@ -580,160 +594,175 @@ int intel_ctx_workarounds_emit(struct i915_request *rq) return 0; } -static void bdw_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void +wal_add(struct i915_wa_list *wal, const struct i915_wa *wa) +{ + const unsigned int grow = 1 << 4; + + GEM_BUG_ON(!is_power_of_2(grow)); + + if (IS_ALIGNED(wal->count, grow)) { /* Either uninitialized or full. */ + struct i915_wa *list; + + list = kmalloc_array(ALIGN(wal->count + 1, grow), sizeof(*wa), + GFP_KERNEL); + if (!list) { + DRM_ERROR("No space for workaround init!\n"); + return; + } + + if (wal->list) + memcpy(list, wal->list, sizeof(*wa) * wal->count); + + wal->list = list; + } + + wal->list[wal->count++] = *wa; +} + +static void +wa_masked_en(struct i915_wa_list *wal, i915_reg_t reg, u32 val) +{ + struct i915_wa wa = { + .reg = reg, + .mask = val, + .val = _MASKED_BIT_ENABLE(val) + }; + + wal_add(wal, &wa); +} + +static void +wa_write_masked_or(struct i915_wa_list *wal, i915_reg_t reg, u32 mask, + u32 val) { + struct i915_wa wa = { + .reg = reg, + .mask = mask, + .val = val + }; + + wal_add(wal, &wa); } -static void chv_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void +wa_write(struct i915_wa_list *wal, i915_reg_t reg, u32 val) { + wa_write_masked_or(wal, reg, ~0, val); } -static void gen9_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void +wa_write_or(struct i915_wa_list *wal, i915_reg_t reg, u32 val) { - /* WaContextSwitchWithConcurrentTLBInvalidate:skl,bxt,kbl,glk,cfl */ - I915_WRITE(GEN9_CSFE_CHICKEN1_RCS, - _MASKED_BIT_ENABLE(GEN9_PREEMPT_GPGPU_SYNC_SWITCH_DISABLE)); + wa_write_masked_or(wal, reg, val, val); +} - /* WaEnableLbsSlaRetryTimerDecrement:skl,bxt,kbl,glk,cfl */ - I915_WRITE(BDW_SCRATCH1, I915_READ(BDW_SCRATCH1) | - GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE); +static void gen9_gt_workarounds_init(struct drm_i915_private *i915) +{ + struct i915_wa_list *wal = &i915->gt_wa_list; /* WaDisableKillLogic:bxt,skl,kbl */ - if (!IS_COFFEELAKE(dev_priv)) - I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | - ECOCHK_DIS_TLB); + if (!IS_COFFEELAKE(i915)) + wa_write_or(wal, + GAM_ECOCHK, + ECOCHK_DIS_TLB); - if (HAS_LLC(dev_priv)) { + if (HAS_LLC(i915)) { /* WaCompressedResourceSamplerPbeMediaNewHashMode:skl,kbl * * Must match Display Engine. See * WaCompressedResourceDisplayNewHashMode. */ - I915_WRITE(MMCD_MISC_CTRL, - I915_READ(MMCD_MISC_CTRL) | - MMCD_PCLA | - MMCD_HOTSPOT_EN); + wa_write_or(wal, + MMCD_MISC_CTRL, + MMCD_PCLA | MMCD_HOTSPOT_EN); } /* WaDisableHDCInvalidation:skl,bxt,kbl,cfl */ - I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | - BDW_DISABLE_HDC_INVALIDATION); - - /* WaProgramL3SqcReg1DefaultForPerf:bxt,glk */ - if (IS_GEN9_LP(dev_priv)) { - u32 val = I915_READ(GEN8_L3SQCREG1); - - val &= ~L3_PRIO_CREDITS_MASK; - val |= L3_GENERAL_PRIO_CREDITS(62) | L3_HIGH_PRIO_CREDITS(2); - I915_WRITE(GEN8_L3SQCREG1, val); - } - - /* WaOCLCoherentLineFlush:skl,bxt,kbl,cfl */ - I915_WRITE(GEN8_L3SQCREG4, - I915_READ(GEN8_L3SQCREG4) | GEN8_LQSC_FLUSH_COHERENT_LINES); - - /* WaEnablePreemptionGranularityControlByUMD:skl,bxt,kbl,cfl,[cnl] */ - I915_WRITE(GEN7_FF_SLICE_CS_CHICKEN1, - _MASKED_BIT_ENABLE(GEN9_FFSC_PERCTX_PREEMPT_CTRL)); + wa_write_or(wal, + GAM_ECOCHK, + BDW_DISABLE_HDC_INVALIDATION); } -static void skl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void skl_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaEnableGapsTsvCreditFix:skl */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | GEN9_GAPS_TSV_CREDIT_DISABLE); + gen9_gt_workarounds_init(i915); /* WaDisableGafsUnitClkGating:skl */ - I915_WRITE(GEN7_UCGCTL4, - I915_READ(GEN7_UCGCTL4) | GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); + wa_write_or(wal, + GEN7_UCGCTL4, + GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:skl */ - if (IS_SKL_REVID(dev_priv, SKL_REVID_H0, REVID_FOREVER)) - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); + if (IS_SKL_REVID(i915, SKL_REVID_H0, REVID_FOREVER)) + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void bxt_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void bxt_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaDisablePooledEuLoadBalancingFix:bxt */ - I915_WRITE(FF_SLICE_CS_CHICKEN2, - _MASKED_BIT_ENABLE(GEN9_POOLED_EU_LOAD_BALANCING_FIX_DISABLE)); + gen9_gt_workarounds_init(i915); /* WaInPlaceDecompressionHang:bxt */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void kbl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void kbl_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaEnableGapsTsvCreditFix:kbl */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | GEN9_GAPS_TSV_CREDIT_DISABLE); + gen9_gt_workarounds_init(i915); /* WaDisableDynamicCreditSharing:kbl */ - if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_B0)) - I915_WRITE(GAMT_CHKN_BIT_REG, - I915_READ(GAMT_CHKN_BIT_REG) | - GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING); + if (IS_KBL_REVID(i915, 0, KBL_REVID_B0)) + wa_write_or(wal, + GAMT_CHKN_BIT_REG, + GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING); /* WaDisableGafsUnitClkGating:kbl */ - I915_WRITE(GEN7_UCGCTL4, - I915_READ(GEN7_UCGCTL4) | GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); + wa_write_or(wal, + GEN7_UCGCTL4, + GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:kbl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); - - /* WaKBLVECSSemaphoreWaitPoll:kbl */ - if (IS_KBL_REVID(dev_priv, KBL_REVID_A0, KBL_REVID_E0)) { - struct intel_engine_cs *engine; - unsigned int tmp; - - for_each_engine(engine, dev_priv, tmp) { - if (engine->id == RCS) - continue; - - I915_WRITE(RING_SEMA_WAIT_POLL(engine->mmio_base), 1); - } - } + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void glk_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void glk_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + gen9_gt_workarounds_init(i915); } -static void cfl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void cfl_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaEnableGapsTsvCreditFix:cfl */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | GEN9_GAPS_TSV_CREDIT_DISABLE); + gen9_gt_workarounds_init(i915); /* WaDisableGafsUnitClkGating:cfl */ - I915_WRITE(GEN7_UCGCTL4, - I915_READ(GEN7_UCGCTL4) | GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); + wa_write_or(wal, + GEN7_UCGCTL4, + GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:cfl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } static void wa_init_mcr(struct drm_i915_private *dev_priv) { const struct sseu_dev_info *sseu = &(INTEL_INFO(dev_priv)->sseu); - u32 mcr; + struct i915_wa_list *wal = &dev_priv->gt_wa_list; u32 mcr_slice_subslice_mask; /* @@ -770,8 +799,6 @@ static void wa_init_mcr(struct drm_i915_private *dev_priv) WARN_ON((enabled_mask & disabled_mask) != enabled_mask); } - mcr = I915_READ(GEN8_MCR_SELECTOR); - if (INTEL_GEN(dev_priv) >= 11) mcr_slice_subslice_mask = GEN11_MCR_SLICE_MASK | GEN11_MCR_SUBSLICE_MASK; @@ -789,148 +816,170 @@ static void wa_init_mcr(struct drm_i915_private *dev_priv) * occasions, such as INSTDONE, where this value is dependent * on s/ss combo, the read should be done with read_subslice_reg. */ - mcr &= ~mcr_slice_subslice_mask; - mcr |= intel_calculate_mcr_s_ss_select(dev_priv); - I915_WRITE(GEN8_MCR_SELECTOR, mcr); + wa_write_masked_or(wal, + GEN8_MCR_SELECTOR, + mcr_slice_subslice_mask, + intel_calculate_mcr_s_ss_select(dev_priv)); } -static void cnl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void cnl_gt_workarounds_init(struct drm_i915_private *i915) { - wa_init_mcr(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; + + wa_init_mcr(i915); /* WaDisableI2mCycleOnWRPort:cnl (pre-prod) */ - if (IS_CNL_REVID(dev_priv, CNL_REVID_B0, CNL_REVID_B0)) - I915_WRITE(GAMT_CHKN_BIT_REG, - I915_READ(GAMT_CHKN_BIT_REG) | - GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT); + if (IS_CNL_REVID(i915, CNL_REVID_B0, CNL_REVID_B0)) + wa_write_or(wal, + GAMT_CHKN_BIT_REG, + GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT); /* WaInPlaceDecompressionHang:cnl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); - - /* WaEnablePreemptionGranularityControlByUMD:cnl */ - I915_WRITE(GEN7_FF_SLICE_CS_CHICKEN1, - _MASKED_BIT_ENABLE(GEN9_FFSC_PERCTX_PREEMPT_CTRL)); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void icl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void icl_gt_workarounds_init(struct drm_i915_private *i915) { - wa_init_mcr(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* This is not an Wa. Enable for better image quality */ - I915_WRITE(_3D_CHICKEN3, - _MASKED_BIT_ENABLE(_3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE)); + wa_init_mcr(i915); /* WaInPlaceDecompressionHang:icl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); - - /* WaPipelineFlushCoherentLines:icl */ - I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) | - GEN8_LQSC_FLUSH_COHERENT_LINES); - - /* Wa_1405543622:icl - * Formerly known as WaGAPZPriorityScheme - */ - I915_WRITE(GEN8_GARBCNTL, I915_READ(GEN8_GARBCNTL) | - GEN11_ARBITRATION_PRIO_ORDER_MASK); - - /* Wa_1604223664:icl - * Formerly known as WaL3BankAddressHashing - */ - I915_WRITE(GEN8_GARBCNTL, - (I915_READ(GEN8_GARBCNTL) & ~GEN11_HASH_CTRL_EXCL_MASK) | - GEN11_HASH_CTRL_EXCL_BIT0); - I915_WRITE(GEN11_GLBLINVL, - (I915_READ(GEN11_GLBLINVL) & ~GEN11_BANK_HASH_ADDR_EXCL_MASK) | - GEN11_BANK_HASH_ADDR_EXCL_BIT0); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); /* WaModifyGamTlbPartitioning:icl */ - I915_WRITE(GEN11_GACB_PERF_CTRL, - (I915_READ(GEN11_GACB_PERF_CTRL) & ~GEN11_HASH_CTRL_MASK) | - GEN11_HASH_CTRL_BIT0 | GEN11_HASH_CTRL_BIT4); - - /* Wa_1405733216:icl - * Formerly known as WaDisableCleanEvicts - */ - I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) | - GEN11_LQSC_CLEAN_EVICT_DISABLE); + wa_write_masked_or(wal, + GEN11_GACB_PERF_CTRL, + GEN11_HASH_CTRL_MASK, + GEN11_HASH_CTRL_BIT0 | GEN11_HASH_CTRL_BIT4); /* Wa_1405766107:icl * Formerly known as WaCL2SFHalfMaxAlloc */ - I915_WRITE(GEN11_LSN_UNSLCVC, I915_READ(GEN11_LSN_UNSLCVC) | - GEN11_LSN_UNSLCVC_GAFS_HALF_SF_MAXALLOC | - GEN11_LSN_UNSLCVC_GAFS_HALF_CL2_MAXALLOC); + wa_write_or(wal, + GEN11_LSN_UNSLCVC, + GEN11_LSN_UNSLCVC_GAFS_HALF_SF_MAXALLOC | + GEN11_LSN_UNSLCVC_GAFS_HALF_CL2_MAXALLOC); /* Wa_220166154:icl * Formerly known as WaDisCtxReload */ - I915_WRITE(GAMW_ECO_DEV_RW_IA_REG, I915_READ(GAMW_ECO_DEV_RW_IA_REG) | - GAMW_ECO_DEV_CTX_RELOAD_DISABLE); + wa_write_or(wal, + GEN8_GAMW_ECO_DEV_RW_IA, + GAMW_ECO_DEV_CTX_RELOAD_DISABLE); /* Wa_1405779004:icl (pre-prod) */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_A0)) - I915_WRITE(SLICE_UNIT_LEVEL_CLKGATE, - I915_READ(SLICE_UNIT_LEVEL_CLKGATE) | - MSCUNIT_CLKGATE_DIS); + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_A0)) + wa_write_or(wal, + SLICE_UNIT_LEVEL_CLKGATE, + MSCUNIT_CLKGATE_DIS); /* Wa_1406680159:icl */ - I915_WRITE(SUBSLICE_UNIT_LEVEL_CLKGATE, - I915_READ(SUBSLICE_UNIT_LEVEL_CLKGATE) | - GWUNIT_CLKGATE_DIS); - - /* Wa_1604302699:icl */ - I915_WRITE(GEN10_L3_CHICKEN_MODE_REGISTER, - I915_READ(GEN10_L3_CHICKEN_MODE_REGISTER) | - GEN11_I2M_WRITE_DISABLE); + wa_write_or(wal, + SUBSLICE_UNIT_LEVEL_CLKGATE, + GWUNIT_CLKGATE_DIS); /* Wa_1406838659:icl (pre-prod) */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_B0)) - I915_WRITE(INF_UNIT_LEVEL_CLKGATE, - I915_READ(INF_UNIT_LEVEL_CLKGATE) | - CGPSF_CLKGATE_DIS); - - /* WaForwardProgressSoftReset:icl */ - I915_WRITE(GEN10_SCRATCH_LNCF2, - I915_READ(GEN10_SCRATCH_LNCF2) | - PMFLUSHDONE_LNICRSDROP | - PMFLUSH_GAPL3UNBLOCK | - PMFLUSHDONE_LNEBLK); + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_B0)) + wa_write_or(wal, + INF_UNIT_LEVEL_CLKGATE, + CGPSF_CLKGATE_DIS); /* Wa_1406463099:icl * Formerly known as WaGamTlbPendError */ - I915_WRITE(GAMT_CHKN_BIT_REG, - I915_READ(GAMT_CHKN_BIT_REG) | - GAMT_CHKN_DISABLE_L3_COH_PIPE); + wa_write_or(wal, + GAMT_CHKN_BIT_REG, + GAMT_CHKN_DISABLE_L3_COH_PIPE); } -void intel_gt_workarounds_apply(struct drm_i915_private *dev_priv) +void intel_gt_init_workarounds(struct drm_i915_private *i915) { - if (INTEL_GEN(dev_priv) < 8) + struct i915_wa_list *wal = &i915->gt_wa_list; + + wa_init_start(wal, "GT"); + + if (INTEL_GEN(i915) < 8) return; - else if (IS_BROADWELL(dev_priv)) - bdw_gt_workarounds_apply(dev_priv); - else if (IS_CHERRYVIEW(dev_priv)) - chv_gt_workarounds_apply(dev_priv); - else if (IS_SKYLAKE(dev_priv)) - skl_gt_workarounds_apply(dev_priv); - else if (IS_BROXTON(dev_priv)) - bxt_gt_workarounds_apply(dev_priv); - else if (IS_KABYLAKE(dev_priv)) - kbl_gt_workarounds_apply(dev_priv); - else if (IS_GEMINILAKE(dev_priv)) - glk_gt_workarounds_apply(dev_priv); - else if (IS_COFFEELAKE(dev_priv)) - cfl_gt_workarounds_apply(dev_priv); - else if (IS_CANNONLAKE(dev_priv)) - cnl_gt_workarounds_apply(dev_priv); - else if (IS_ICELAKE(dev_priv)) - icl_gt_workarounds_apply(dev_priv); + else if (IS_BROADWELL(i915)) + return; + else if (IS_CHERRYVIEW(i915)) + return; + else if (IS_SKYLAKE(i915)) + skl_gt_workarounds_init(i915); + else if (IS_BROXTON(i915)) + bxt_gt_workarounds_init(i915); + else if (IS_KABYLAKE(i915)) + kbl_gt_workarounds_init(i915); + else if (IS_GEMINILAKE(i915)) + glk_gt_workarounds_init(i915); + else if (IS_COFFEELAKE(i915)) + cfl_gt_workarounds_init(i915); + else if (IS_CANNONLAKE(i915)) + cnl_gt_workarounds_init(i915); + else if (IS_ICELAKE(i915)) + icl_gt_workarounds_init(i915); else - MISSING_CASE(INTEL_GEN(dev_priv)); + MISSING_CASE(INTEL_GEN(i915)); + + wa_init_finish(wal); +} + +static enum forcewake_domains +wal_get_fw_for_rmw(struct drm_i915_private *dev_priv, + 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, + wa->reg, + FW_REG_READ | + FW_REG_WRITE); + + return fw; +} + +static void +wa_list_apply(struct drm_i915_private *dev_priv, const struct i915_wa_list *wal) +{ + enum forcewake_domains fw; + unsigned long flags; + struct i915_wa *wa; + unsigned int i; + + if (!wal->count) + return; + + fw = wal_get_fw_for_rmw(dev_priv, wal); + + spin_lock_irqsave(&dev_priv->uncore.lock, flags); + intel_uncore_forcewake_get__locked(dev_priv, 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_forcewake_put__locked(dev_priv, fw); + spin_unlock_irqrestore(&dev_priv->uncore.lock, flags); + + DRM_DEBUG_DRIVER("Applied %u %s workarounds\n", wal->count, wal->name); +} + +void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv) +{ + wa_list_apply(dev_priv, &dev_priv->gt_wa_list); } struct whitelist { @@ -1077,6 +1126,146 @@ void intel_whitelist_workarounds_apply(struct intel_engine_cs *engine) whitelist_apply(engine, whitelist_build(engine, &w)); } +static void rcs_engine_wa_init(struct intel_engine_cs *engine) +{ + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->wa_list; + + if (IS_ICELAKE(i915)) { + /* This is not an Wa. Enable for better image quality */ + wa_masked_en(wal, + _3D_CHICKEN3, + _3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE); + + /* WaPipelineFlushCoherentLines:icl */ + wa_write_or(wal, + GEN8_L3SQCREG4, + GEN8_LQSC_FLUSH_COHERENT_LINES); + + /* + * Wa_1405543622:icl + * Formerly known as WaGAPZPriorityScheme + */ + wa_write_or(wal, + GEN8_GARBCNTL, + GEN11_ARBITRATION_PRIO_ORDER_MASK); + + /* + * Wa_1604223664:icl + * Formerly known as WaL3BankAddressHashing + */ + wa_write_masked_or(wal, + GEN8_GARBCNTL, + GEN11_HASH_CTRL_EXCL_MASK, + GEN11_HASH_CTRL_EXCL_BIT0); + wa_write_masked_or(wal, + GEN11_GLBLINVL, + GEN11_BANK_HASH_ADDR_EXCL_MASK, + GEN11_BANK_HASH_ADDR_EXCL_BIT0); + + /* + * Wa_1405733216:icl + * Formerly known as WaDisableCleanEvicts + */ + wa_write_or(wal, + GEN8_L3SQCREG4, + GEN11_LQSC_CLEAN_EVICT_DISABLE); + + /* Wa_1604302699:icl */ + wa_write_or(wal, + GEN10_L3_CHICKEN_MODE_REGISTER, + GEN11_I2M_WRITE_DISABLE); + + /* WaForwardProgressSoftReset:icl */ + wa_write_or(wal, + GEN10_SCRATCH_LNCF2, + PMFLUSHDONE_LNICRSDROP | + PMFLUSH_GAPL3UNBLOCK | + PMFLUSHDONE_LNEBLK); + } + + if (IS_GEN9(i915) || IS_CANNONLAKE(i915)) { + /* WaEnablePreemptionGranularityControlByUMD:skl,bxt,kbl,cfl,cnl */ + wa_masked_en(wal, + GEN7_FF_SLICE_CS_CHICKEN1, + GEN9_FFSC_PERCTX_PREEMPT_CTRL); + } + + if (IS_SKYLAKE(i915) || IS_KABYLAKE(i915) || IS_COFFEELAKE(i915)) { + /* WaEnableGapsTsvCreditFix:skl,kbl,cfl */ + wa_write_or(wal, + GEN8_GARBCNTL, + GEN9_GAPS_TSV_CREDIT_DISABLE); + } + + if (IS_BROXTON(i915)) { + /* WaDisablePooledEuLoadBalancingFix:bxt */ + wa_masked_en(wal, + FF_SLICE_CS_CHICKEN2, + GEN9_POOLED_EU_LOAD_BALANCING_FIX_DISABLE); + } + + if (IS_GEN9(i915)) { + /* WaContextSwitchWithConcurrentTLBInvalidate:skl,bxt,kbl,glk,cfl */ + wa_masked_en(wal, + GEN9_CSFE_CHICKEN1_RCS, + GEN9_PREEMPT_GPGPU_SYNC_SWITCH_DISABLE); + + /* WaEnableLbsSlaRetryTimerDecrement:skl,bxt,kbl,glk,cfl */ + wa_write_or(wal, + BDW_SCRATCH1, + GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE); + + /* WaProgramL3SqcReg1DefaultForPerf:bxt,glk */ + if (IS_GEN9_LP(i915)) + wa_write_masked_or(wal, + GEN8_L3SQCREG1, + L3_PRIO_CREDITS_MASK, + L3_GENERAL_PRIO_CREDITS(62) | + L3_HIGH_PRIO_CREDITS(2)); + + /* WaOCLCoherentLineFlush:skl,bxt,kbl,cfl */ + wa_write_or(wal, + GEN8_L3SQCREG4, + GEN8_LQSC_FLUSH_COHERENT_LINES); + } +} + +static void xcs_engine_wa_init(struct intel_engine_cs *engine) +{ + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->wa_list; + + /* WaKBLVECSSemaphoreWaitPoll:kbl */ + if (IS_KBL_REVID(i915, KBL_REVID_A0, KBL_REVID_E0)) { + wa_write(wal, + RING_SEMA_WAIT_POLL(engine->mmio_base), + 1); + } +} + +void intel_engine_init_workarounds(struct intel_engine_cs *engine) +{ + struct i915_wa_list *wal = &engine->wa_list; + + if (GEM_WARN_ON(INTEL_GEN(engine->i915) < 8)) + return; + + wa_init_start(wal, engine->name); + + if (engine->id == RCS) + rcs_engine_wa_init(engine); + else + xcs_engine_wa_init(engine); + + wa_init_finish(wal); +} + +void intel_engine_apply_workarounds(struct intel_engine_cs *engine) +{ + wa_list_apply(engine->i915, &engine->wa_list); +} + #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) #include "selftests/intel_workarounds.c" #endif diff --git a/drivers/gpu/drm/i915/intel_workarounds.h b/drivers/gpu/drm/i915/intel_workarounds.h index b11d0623e626..979695a53964 100644 --- a/drivers/gpu/drm/i915/intel_workarounds.h +++ b/drivers/gpu/drm/i915/intel_workarounds.h @@ -7,11 +7,35 @@ #ifndef _I915_WORKAROUNDS_H_ #define _I915_WORKAROUNDS_H_ +#include <linux/slab.h> + +struct i915_wa { + i915_reg_t reg; + u32 mask; + u32 val; +}; + +struct i915_wa_list { + const char *name; + struct i915_wa *list; + unsigned int count; +}; + +static inline void intel_wa_list_free(struct i915_wa_list *wal) +{ + kfree(wal->list); + memset(wal, 0, sizeof(*wal)); +} + int intel_ctx_workarounds_init(struct drm_i915_private *dev_priv); int intel_ctx_workarounds_emit(struct i915_request *rq); -void intel_gt_workarounds_apply(struct drm_i915_private *dev_priv); +void intel_gt_init_workarounds(struct drm_i915_private *dev_priv); +void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv); void intel_whitelist_workarounds_apply(struct intel_engine_cs *engine); +void intel_engine_init_workarounds(struct intel_engine_cs *engine); +void intel_engine_apply_workarounds(struct intel_engine_cs *engine); + #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c index d4530d60767b..ca169f013a14 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c @@ -1594,7 +1594,6 @@ struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane, NULL); drm_crtc_helper_add(crtc, &dpu_crtc_helper_funcs); - plane->crtc = crtc; /* save user friendly CRTC name for later */ snprintf(dpu_crtc->name, DPU_CRTC_NAME_SIZE, "crtc%u", crtc->base.id); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c index 96cdf06e7da2..d31d8281424e 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -488,8 +488,6 @@ static void dpu_encoder_destroy(struct drm_encoder *drm_enc) drm_encoder_cleanup(drm_enc); mutex_destroy(&dpu_enc->enc_lock); - - kfree(dpu_enc); } void dpu_encoder_helper_split_config( diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c index bfcd165e96df..d743e7ca6a3c 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c @@ -216,7 +216,7 @@ static const struct dpu_format dpu_format_map[] = { INTERLEAVED_RGB_FMT(XBGR8888, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, - true, 4, 0, + false, 4, 0, DPU_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBA8888, diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c index 4c03f0b7343e..41bec570c518 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c @@ -39,6 +39,8 @@ #define DSI_PIXEL_PLL_CLK 1 #define NUM_PROVIDED_CLKS 2 +#define VCO_REF_CLK_RATE 19200000 + struct dsi_pll_regs { u32 pll_prop_gain_rate; u32 pll_lockdet_rate; @@ -316,7 +318,7 @@ static int dsi_pll_10nm_vco_set_rate(struct clk_hw *hw, unsigned long rate, parent_rate); pll_10nm->vco_current_rate = rate; - pll_10nm->vco_ref_clk_rate = parent_rate; + pll_10nm->vco_ref_clk_rate = VCO_REF_CLK_RATE; dsi_pll_setup_config(pll_10nm); diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c index c79659ca5706..adbdce3aeda0 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi.c @@ -332,6 +332,12 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, goto fail; } + ret = msm_hdmi_hpd_enable(hdmi->connector); + if (ret < 0) { + DRM_DEV_ERROR(&hdmi->pdev->dev, "failed to enable HPD: %d\n", ret); + goto fail; + } + encoder->bridge = hdmi->bridge; priv->bridges[priv->num_bridges++] = hdmi->bridge; @@ -571,7 +577,7 @@ static int msm_hdmi_bind(struct device *dev, struct device *master, void *data) { struct drm_device *drm = dev_get_drvdata(master); struct msm_drm_private *priv = drm->dev_private; - static struct hdmi_platform_config *hdmi_cfg; + struct hdmi_platform_config *hdmi_cfg; struct hdmi *hdmi; struct device_node *of_node = dev->of_node; int i, err; diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.h b/drivers/gpu/drm/msm/hdmi/hdmi.h index accc9a61611d..5c5df6ab2a57 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.h +++ b/drivers/gpu/drm/msm/hdmi/hdmi.h @@ -245,6 +245,7 @@ void msm_hdmi_bridge_destroy(struct drm_bridge *bridge); void msm_hdmi_connector_irq(struct drm_connector *connector); struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi); +int msm_hdmi_hpd_enable(struct drm_connector *connector); /* * i2c adapter for ddc: diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c index e9c9a0af508e..30e908dfded7 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c @@ -167,8 +167,9 @@ static void enable_hpd_clocks(struct hdmi *hdmi, bool enable) } } -static int hpd_enable(struct hdmi_connector *hdmi_connector) +int msm_hdmi_hpd_enable(struct drm_connector *connector) { + struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector); struct hdmi *hdmi = hdmi_connector->hdmi; const struct hdmi_platform_config *config = hdmi->config; struct device *dev = &hdmi->pdev->dev; @@ -450,7 +451,6 @@ struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi) { struct drm_connector *connector = NULL; struct hdmi_connector *hdmi_connector; - int ret; hdmi_connector = kzalloc(sizeof(*hdmi_connector), GFP_KERNEL); if (!hdmi_connector) @@ -471,12 +471,6 @@ struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi) connector->interlace_allowed = 0; connector->doublescan_allowed = 0; - ret = hpd_enable(hdmi_connector); - if (ret) { - dev_err(&hdmi->pdev->dev, "failed to enable HPD: %d\n", ret); - return ERR_PTR(ret); - } - drm_connector_attach_encoder(connector, hdmi->encoder); return connector; diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c index 4bcdeca7479d..2088a20eb270 100644 --- a/drivers/gpu/drm/msm/msm_atomic.c +++ b/drivers/gpu/drm/msm/msm_atomic.c @@ -34,7 +34,12 @@ static void msm_atomic_wait_for_commit_done(struct drm_device *dev, if (!new_crtc_state->active) continue; + if (drm_crtc_vblank_get(crtc)) + continue; + kms->funcs->wait_for_crtc_commit_done(kms, crtc); + + drm_crtc_vblank_put(crtc); } } diff --git a/drivers/gpu/drm/msm/msm_debugfs.c b/drivers/gpu/drm/msm/msm_debugfs.c index f0da0d3c8a80..d756436c1fcd 100644 --- a/drivers/gpu/drm/msm/msm_debugfs.c +++ b/drivers/gpu/drm/msm/msm_debugfs.c @@ -84,7 +84,7 @@ static int msm_gpu_open(struct inode *inode, struct file *file) ret = mutex_lock_interruptible(&dev->struct_mutex); if (ret) - return ret; + goto free_priv; pm_runtime_get_sync(&gpu->pdev->dev); show_priv->state = gpu->funcs->gpu_state_get(gpu); @@ -94,13 +94,20 @@ static int msm_gpu_open(struct inode *inode, struct file *file) if (IS_ERR(show_priv->state)) { ret = PTR_ERR(show_priv->state); - kfree(show_priv); - return ret; + goto free_priv; } show_priv->dev = dev; - return single_open(file, msm_gpu_show, show_priv); + ret = single_open(file, msm_gpu_show, show_priv); + if (ret) + goto free_priv; + + return 0; + +free_priv: + kfree(show_priv); + return ret; } static const struct file_operations msm_gpu_fops = { diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 4904d0d41409..dcff812c63d0 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -553,17 +553,18 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) kthread_run(kthread_worker_fn, &priv->disp_thread[i].worker, "crtc_commit:%d", priv->disp_thread[i].crtc_id); - ret = sched_setscheduler(priv->disp_thread[i].thread, - SCHED_FIFO, ¶m); - if (ret) - pr_warn("display thread priority update failed: %d\n", - ret); - if (IS_ERR(priv->disp_thread[i].thread)) { dev_err(dev, "failed to create crtc_commit kthread\n"); priv->disp_thread[i].thread = NULL; + goto err_msm_uninit; } + ret = sched_setscheduler(priv->disp_thread[i].thread, + SCHED_FIFO, ¶m); + if (ret) + dev_warn(dev, "disp_thread set priority failed: %d\n", + ret); + /* initialize event thread */ priv->event_thread[i].crtc_id = priv->crtcs[i]->base.id; kthread_init_worker(&priv->event_thread[i].worker); @@ -572,6 +573,12 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) kthread_run(kthread_worker_fn, &priv->event_thread[i].worker, "crtc_event:%d", priv->event_thread[i].crtc_id); + if (IS_ERR(priv->event_thread[i].thread)) { + dev_err(dev, "failed to create crtc_event kthread\n"); + priv->event_thread[i].thread = NULL; + goto err_msm_uninit; + } + /** * event thread should also run at same priority as disp_thread * because it is handling frame_done events. A lower priority @@ -580,34 +587,10 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) * failure at crtc commit level. */ ret = sched_setscheduler(priv->event_thread[i].thread, - SCHED_FIFO, ¶m); + SCHED_FIFO, ¶m); if (ret) - pr_warn("display event thread priority update failed: %d\n", - ret); - - if (IS_ERR(priv->event_thread[i].thread)) { - dev_err(dev, "failed to create crtc_event kthread\n"); - priv->event_thread[i].thread = NULL; - } - - if ((!priv->disp_thread[i].thread) || - !priv->event_thread[i].thread) { - /* clean up previously created threads if any */ - for ( ; i >= 0; i--) { - if (priv->disp_thread[i].thread) { - kthread_stop( - priv->disp_thread[i].thread); - priv->disp_thread[i].thread = NULL; - } - - if (priv->event_thread[i].thread) { - kthread_stop( - priv->event_thread[i].thread); - priv->event_thread[i].thread = NULL; - } - } - goto err_msm_uninit; - } + dev_warn(dev, "event_thread set priority failed:%d\n", + ret); } ret = drm_vblank_init(ddev, priv->num_crtcs); diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index 7a7923e6220d..6942604ad9a8 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -317,6 +317,9 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob uint32_t *ptr; int ret = 0; + if (!nr_relocs) + return 0; + if (offset % 4) { DRM_ERROR("non-aligned cmdstream buffer: %u\n", offset); return -EINVAL; @@ -410,7 +413,6 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, struct msm_file_private *ctx = file->driver_priv; struct msm_gem_submit *submit; struct msm_gpu *gpu = priv->gpu; - struct dma_fence *in_fence = NULL; struct sync_file *sync_file = NULL; struct msm_gpu_submitqueue *queue; struct msm_ringbuffer *ring; @@ -443,6 +445,8 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, ring = gpu->rb[queue->prio]; if (args->flags & MSM_SUBMIT_FENCE_FD_IN) { + struct dma_fence *in_fence; + in_fence = sync_file_get_fence(args->fence_fd); if (!in_fence) @@ -452,11 +456,13 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, * Wait if the fence is from a foreign context, or if the fence * array contains any fence from a foreign context. */ - if (!dma_fence_match_context(in_fence, ring->fctx->context)) { + ret = 0; + if (!dma_fence_match_context(in_fence, ring->fctx->context)) ret = dma_fence_wait(in_fence, true); - if (ret) - return ret; - } + + dma_fence_put(in_fence); + if (ret) + return ret; } ret = mutex_lock_interruptible(&dev->struct_mutex); @@ -582,8 +588,6 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, } out: - if (in_fence) - dma_fence_put(in_fence); submit_cleanup(submit); if (ret) msm_gem_submit_free(submit); diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 11aac8337066..2b7c8946adba 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -345,6 +345,10 @@ static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, { struct msm_gpu_state *state; + /* Check if the target supports capturing crash state */ + if (!gpu->funcs->gpu_state_get) + return; + /* Only save one crash state at a time */ if (gpu->crashstate) return; @@ -434,10 +438,9 @@ static void recover_worker(struct work_struct *work) if (submit) { struct task_struct *task; - rcu_read_lock(); - task = pid_task(submit->pid, PIDTYPE_PID); + task = get_pid_task(submit->pid, PIDTYPE_PID); if (task) { - comm = kstrdup(task->comm, GFP_ATOMIC); + comm = kstrdup(task->comm, GFP_KERNEL); /* * So slightly annoying, in other paths like @@ -450,10 +453,10 @@ static void recover_worker(struct work_struct *work) * about the submit going away. */ mutex_unlock(&dev->struct_mutex); - cmd = kstrdup_quotable_cmdline(task, GFP_ATOMIC); + cmd = kstrdup_quotable_cmdline(task, GFP_KERNEL); + put_task_struct(task); mutex_lock(&dev->struct_mutex); } - rcu_read_unlock(); if (comm && cmd) { dev_err(dev->dev, "%s: offending task: %s (%s)\n", diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c index b23d33622f37..2a90aa4caec0 100644 --- a/drivers/gpu/drm/msm/msm_iommu.c +++ b/drivers/gpu/drm/msm/msm_iommu.c @@ -66,7 +66,7 @@ static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova, // pm_runtime_get_sync(mmu->dev); ret = iommu_map_sg(iommu->domain, iova, sgt->sgl, sgt->nents, prot); // pm_runtime_put_sync(mmu->dev); - WARN_ON(ret < 0); + WARN_ON(!ret); return (ret == len) ? 0 : -EINVAL; } diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index cca933458439..0c2c8d2c631f 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -316,10 +316,11 @@ static void snapshot_buf(struct msm_rd_state *rd, uint64_t iova, uint32_t size) { struct msm_gem_object *obj = submit->bos[idx].obj; + unsigned offset = 0; const char *buf; if (iova) { - buf += iova - submit->bos[idx].iova; + offset = iova - submit->bos[idx].iova; } else { iova = submit->bos[idx].iova; size = obj->base.size; @@ -340,6 +341,8 @@ static void snapshot_buf(struct msm_rd_state *rd, if (IS_ERR(buf)) return; + buf += offset; + rd_write_section(rd, RD_BUFFER_CONTENTS, buf, size); msm_gem_put_vaddr(&obj->base); diff --git a/drivers/gpu/drm/rcar-du/rcar_du_group.c b/drivers/gpu/drm/rcar-du/rcar_du_group.c index d85f0a1c1581..cebf313c6e1f 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_group.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_group.c @@ -202,10 +202,25 @@ void rcar_du_group_put(struct rcar_du_group *rgrp) static void __rcar_du_group_start_stop(struct rcar_du_group *rgrp, bool start) { - struct rcar_du_crtc *rcrtc = &rgrp->dev->crtcs[rgrp->index * 2]; + struct rcar_du_device *rcdu = rgrp->dev; + + /* + * Group start/stop is controlled by the DRES and DEN bits of DSYSR0 + * for the first group and DSYSR2 for the second group. On most DU + * instances, this maps to the first CRTC of the group, and we can just + * use rcar_du_crtc_dsysr_clr_set() to access the correct DSYSR. On + * M3-N, however, DU2 doesn't exist, but DSYSR2 does. We thus need to + * access the register directly using group read/write. + */ + if (rcdu->info->channels_mask & BIT(rgrp->index * 2)) { + struct rcar_du_crtc *rcrtc = &rgrp->dev->crtcs[rgrp->index * 2]; - rcar_du_crtc_dsysr_clr_set(rcrtc, DSYSR_DRES | DSYSR_DEN, - start ? DSYSR_DEN : DSYSR_DRES); + rcar_du_crtc_dsysr_clr_set(rcrtc, DSYSR_DRES | DSYSR_DEN, + start ? DSYSR_DEN : DSYSR_DRES); + } else { + rcar_du_group_write(rgrp, DSYSR, + start ? DSYSR_DEN : DSYSR_DRES); + } } void rcar_du_group_start_stop(struct rcar_du_group *rgrp, bool start) diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index ba80150d1052..895d77d799e4 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -492,8 +492,10 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, if (!fbo) return -ENOMEM; - ttm_bo_get(bo); fbo->base = *bo; + fbo->base.mem.placement |= TTM_PL_FLAG_NO_EVICT; + + ttm_bo_get(bo); fbo->bo = bo; /** |
