diff options
Diffstat (limited to 'drivers')
253 files changed, 4664 insertions, 2484 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/aldebaran.c b/drivers/gpu/drm/amd/amdgpu/aldebaran.c index 02f4c6f9d4f6..576067d66bb9 100644 --- a/drivers/gpu/drm/amd/amdgpu/aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/aldebaran.c @@ -330,6 +330,7 @@ aldebaran_mode2_restore_hwcontext(struct amdgpu_reset_control *reset_ctl, { struct list_head *reset_device_list = reset_context->reset_device_list; struct amdgpu_device *tmp_adev = NULL; + struct amdgpu_ras *con; int r; if (reset_device_list == NULL) @@ -355,7 +356,30 @@ aldebaran_mode2_restore_hwcontext(struct amdgpu_reset_control *reset_ctl, */ amdgpu_register_gpu_instance(tmp_adev); - /* Resume RAS */ + /* Resume RAS, ecc_irq */ + con = amdgpu_ras_get_context(tmp_adev); + if (!amdgpu_sriov_vf(tmp_adev) && con) { + if (tmp_adev->sdma.ras && + tmp_adev->sdma.ras->ras_block.ras_late_init) { + r = tmp_adev->sdma.ras->ras_block.ras_late_init(tmp_adev, + &tmp_adev->sdma.ras->ras_block.ras_comm); + if (r) { + dev_err(tmp_adev->dev, "SDMA failed to execute ras_late_init! ret:%d\n", r); + goto end; + } + } + + if (tmp_adev->gfx.ras && + tmp_adev->gfx.ras->ras_block.ras_late_init) { + r = tmp_adev->gfx.ras->ras_block.ras_late_init(tmp_adev, + &tmp_adev->gfx.ras->ras_block.ras_comm); + if (r) { + dev_err(tmp_adev->dev, "GFX failed to execute ras_late_init! ret:%d\n", r); + goto end; + } + } + } + amdgpu_ras_resume(tmp_adev); /* Update PSP FW topology after reset */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 616b6c911767..3d8a48f46b01 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1144,6 +1144,7 @@ struct amdgpu_device { bool debug_vm; bool debug_largebar; bool debug_disable_soft_recovery; + bool debug_use_vram_fw_buf; }; static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 067690ba7bff..77e263660288 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -138,6 +138,9 @@ static void amdgpu_amdkfd_reset_work(struct work_struct *work) amdgpu_device_gpu_recover(adev, NULL, &reset_context); } +static const struct drm_client_funcs kfd_client_funcs = { + .unregister = drm_client_release, +}; void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) { int i; @@ -161,7 +164,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) .enable_mes = adev->enable_mes, }; - ret = drm_client_init(&adev->ddev, &adev->kfd.client, "kfd", NULL); + ret = drm_client_init(&adev->ddev, &adev->kfd.client, "kfd", &kfd_client_funcs); if (ret) { dev_err(adev->dev, "Failed to init DRM client: %d\n", ret); return; @@ -695,10 +698,8 @@ err: void amdgpu_amdkfd_set_compute_idle(struct amdgpu_device *adev, bool idle) { enum amd_powergating_state state = idle ? AMD_PG_STATE_GATE : AMD_PG_STATE_UNGATE; - /* Temporary workaround to fix issues observed in some - * compute applications when GFXOFF is enabled on GFX11. - */ - if (IP_VERSION_MAJ(amdgpu_ip_version(adev, GC_HWIP, 0)) == 11) { + if (IP_VERSION_MAJ(amdgpu_ip_version(adev, GC_HWIP, 0)) == 11 && + ((adev->mes.kiq_version & AMDGPU_MES_VERSION_MASK) <= 64)) { pr_debug("GFXOFF is %s\n", idle ? "enabled" : "disabled"); amdgpu_gfx_off_ctrl(adev, idle); } else if ((IP_VERSION_MAJ(amdgpu_ip_version(adev, GC_HWIP, 0)) == 9) && diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index cf6ed5fce291..f262b9d89541 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -311,7 +311,7 @@ void amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(struct kgd_mem *mem); int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_device *adev, struct amdgpu_bo *bo); int amdgpu_amdkfd_gpuvm_restore_process_bos(void *process_info, - struct dma_fence **ef); + struct dma_fence __rcu **ef); int amdgpu_amdkfd_gpuvm_get_vm_fault_info(struct amdgpu_device *adev, struct kfd_vm_fault_info *info); int amdgpu_amdkfd_gpuvm_import_dmabuf_fd(struct amdgpu_device *adev, int fd, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c index 469785d33791..1ef758ac5076 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c @@ -90,7 +90,7 @@ struct amdgpu_amdkfd_fence *to_amdgpu_amdkfd_fence(struct dma_fence *f) return NULL; fence = container_of(f, struct amdgpu_amdkfd_fence, base); - if (fence && f->ops == &amdkfd_fence_ops) + if (f->ops == &amdkfd_fence_ops) return fence; return NULL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index d17b2452cb1f..f183d7faeeec 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -2802,7 +2802,7 @@ unlock_out: put_task_struct(usertask); } -static void replace_eviction_fence(struct dma_fence **ef, +static void replace_eviction_fence(struct dma_fence __rcu **ef, struct dma_fence *new_ef) { struct dma_fence *old_ef = rcu_replace_pointer(*ef, new_ef, true @@ -2837,7 +2837,7 @@ static void replace_eviction_fence(struct dma_fence **ef, * 7. Add fence to all PD and PT BOs. * 8. Unreserve all BOs */ -int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence **ef) +int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence __rcu **ef) { struct amdkfd_process_info *process_info = info; struct amdgpu_vm *peer_vm; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index 96f63fd39b9e..9caba10315a8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -1103,7 +1103,7 @@ amdgpu_connector_dvi_detect(struct drm_connector *connector, bool force) * DDC line. The latter is more complex because with DVI<->HDMI adapters * you don't really know what's connected to which port as both are digital. */ - amdgpu_connector_shared_ddc(&ret, connector, amdgpu_connector); + amdgpu_connector_shared_ddc(&ret, connector, amdgpu_connector); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 440e9a6786fc..6adeddfb3d56 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -870,9 +870,9 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, struct amdgpu_bo *bo = e->bo; int i; - e->user_pages = kvmalloc_array(bo->tbo.ttm->num_pages, - sizeof(struct page *), - GFP_KERNEL | __GFP_ZERO); + e->user_pages = kvcalloc(bo->tbo.ttm->num_pages, + sizeof(struct page *), + GFP_KERNEL); if (!e->user_pages) { DRM_ERROR("kvmalloc_array failure\n"); r = -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index 2cebf2145d9a..e485dd3357c6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -540,7 +540,11 @@ static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf, while (size) { uint32_t value; - value = RREG32_PCIE(*pos); + if (upper_32_bits(*pos)) + value = RREG32_PCIE_EXT(*pos); + else + value = RREG32_PCIE(*pos); + r = put_user(value, (uint32_t *)buf); if (r) goto out; @@ -600,7 +604,10 @@ static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user if (r) goto out; - WREG32_PCIE(*pos, value); + if (upper_32_bits(*pos)) + WREG32_PCIE_EXT(*pos, value); + else + WREG32_PCIE(*pos, value); result += 4; buf += 4; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 85ed0d66a029..b158d27d0a71 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1544,6 +1544,7 @@ bool amdgpu_device_need_post(struct amdgpu_device *adev) return true; fw_ver = *((uint32_t *)adev->pm.fw->data + 69); + release_firmware(adev->pm.fw); if (fw_ver < 0x00160e00) return true; } @@ -2251,15 +2252,8 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev) adev->firmware.gpu_info_fw = NULL; - if (adev->mman.discovery_bin) { - /* - * FIXME: The bounding box is still needed by Navi12, so - * temporarily read it from gpu_info firmware. Should be dropped - * when DAL no longer needs it. - */ - if (adev->asic_type != CHIP_NAVI12) - return 0; - } + if (adev->mman.discovery_bin) + return 0; switch (adev->asic_type) { default: @@ -5252,7 +5246,6 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle, struct amdgpu_device *tmp_adev = NULL; bool need_full_reset, skip_hw_reset, vram_lost = false; int r = 0; - bool gpu_reset_for_dev_remove = 0; /* Try reset handler method first */ tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device, @@ -5272,10 +5265,6 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle, test_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags); skip_hw_reset = test_bit(AMDGPU_SKIP_HW_RESET, &reset_context->flags); - gpu_reset_for_dev_remove = - test_bit(AMDGPU_RESET_FOR_DEVICE_REMOVE, &reset_context->flags) && - test_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags); - /* * ASIC reset has to be done on all XGMI hive nodes ASAP * to allow proper links negotiation in FW (within 1 sec) @@ -5318,18 +5307,6 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle, amdgpu_ras_intr_cleared(); } - /* Since the mode1 reset affects base ip blocks, the - * phase1 ip blocks need to be resumed. Otherwise there - * will be a BIOS signature error and the psp bootloader - * can't load kdb on the next amdgpu install. - */ - if (gpu_reset_for_dev_remove) { - list_for_each_entry(tmp_adev, device_list_handle, reset_list) - amdgpu_device_ip_resume_phase1(tmp_adev); - - goto end; - } - list_for_each_entry(tmp_adev, device_list_handle, reset_list) { if (need_full_reset) { /* post card */ @@ -5566,11 +5543,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, int i, r = 0; bool need_emergency_restart = false; bool audio_suspended = false; - bool gpu_reset_for_dev_remove = false; - - gpu_reset_for_dev_remove = - test_bit(AMDGPU_RESET_FOR_DEVICE_REMOVE, &reset_context->flags) && - test_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags); /* * Special case: RAS triggered and full reset isn't supported @@ -5608,7 +5580,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, if (!amdgpu_sriov_vf(adev) && (adev->gmc.xgmi.num_physical_nodes > 1)) { list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) { list_add_tail(&tmp_adev->reset_list, &device_list); - if (gpu_reset_for_dev_remove && adev->shutdown) + if (adev->shutdown) tmp_adev->shutdown = true; } if (!list_is_first(&adev->reset_list, &device_list)) @@ -5693,10 +5665,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, retry: /* Rest of adevs pre asic reset from XGMI hive. */ list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - if (gpu_reset_for_dev_remove) { - /* Workaroud for ASICs need to disable SMC first */ - amdgpu_device_smu_fini_early(tmp_adev); - } r = amdgpu_device_pre_asic_reset(tmp_adev, reset_context); /*TODO Should we stop ?*/ if (r) { @@ -5728,9 +5696,6 @@ retry: /* Rest of adevs pre asic reset from XGMI hive. */ r = amdgpu_do_asic_reset(device_list_handle, reset_context); if (r && r == -EAGAIN) goto retry; - - if (!r && gpu_reset_for_dev_remove) - goto recover_end; } skip_hw_reset: @@ -5786,7 +5751,6 @@ skip_sched_resume: amdgpu_ras_set_error_query_ready(tmp_adev, true); } -recover_end: tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device, reset_list); amdgpu_device_unlock_reset_domain(tmp_adev->reset_domain); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 0431eafa86b5..c7d60dd0fb97 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -1963,8 +1963,6 @@ static int amdgpu_discovery_set_gc_ip_blocks(struct amdgpu_device *adev) amdgpu_device_ip_block_add(adev, &gfx_v9_0_ip_block); break; case IP_VERSION(9, 4, 3): - if (!amdgpu_exp_hw_support) - return -EINVAL; amdgpu_device_ip_block_add(adev, &gfx_v9_4_3_ip_block); break; case IP_VERSION(10, 1, 10): diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 880137774b4e..cc69005f5b46 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -128,6 +128,7 @@ enum AMDGPU_DEBUG_MASK { AMDGPU_DEBUG_VM = BIT(0), AMDGPU_DEBUG_LARGEBAR = BIT(1), AMDGPU_DEBUG_DISABLE_GPU_SOFT_RECOVERY = BIT(2), + AMDGPU_DEBUG_USE_VRAM_FW_BUF = BIT(3), }; unsigned int amdgpu_vram_limit = UINT_MAX; @@ -2117,6 +2118,11 @@ static void amdgpu_init_debug_options(struct amdgpu_device *adev) pr_info("debug: soft reset for GPU recovery disabled\n"); adev->debug_disable_soft_recovery = true; } + + if (amdgpu_debug_mask & AMDGPU_DEBUG_USE_VRAM_FW_BUF) { + pr_info("debug: place fw in vram for frontdoor loading\n"); + adev->debug_use_vram_fw_buf = true; + } } static unsigned long amdgpu_fix_asic_type(struct pci_dev *pdev, unsigned long flags) @@ -2228,6 +2234,8 @@ static int amdgpu_pci_probe(struct pci_dev *pdev, pci_set_drvdata(pdev, ddev); + amdgpu_init_debug_options(adev); + ret = amdgpu_driver_load_kms(adev, flags); if (ret) goto err_pci; @@ -2308,8 +2316,6 @@ retry_init: amdgpu_get_secondary_funcs(adev); } - amdgpu_init_debug_options(adev); - return 0; err_pci: @@ -2331,38 +2337,6 @@ amdgpu_pci_remove(struct pci_dev *pdev) pm_runtime_forbid(dev->dev); } - if (amdgpu_ip_version(adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 2) && - !amdgpu_sriov_vf(adev)) { - bool need_to_reset_gpu = false; - - if (adev->gmc.xgmi.num_physical_nodes > 1) { - struct amdgpu_hive_info *hive; - - hive = amdgpu_get_xgmi_hive(adev); - if (hive->device_remove_count == 0) - need_to_reset_gpu = true; - hive->device_remove_count++; - amdgpu_put_xgmi_hive(hive); - } else { - need_to_reset_gpu = true; - } - - /* Workaround for ASICs need to reset SMU. - * Called only when the first device is removed. - */ - if (need_to_reset_gpu) { - struct amdgpu_reset_context reset_context; - - adev->shutdown = true; - memset(&reset_context, 0, sizeof(reset_context)); - reset_context.method = AMD_RESET_METHOD_NONE; - reset_context.reset_req_dev = adev; - set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); - set_bit(AMDGPU_RESET_FOR_DEVICE_REMOVE, &reset_context.flags); - amdgpu_device_gpu_recover(adev, NULL, &reset_context); - } - } - amdgpu_driver_unload_kms(dev); /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index d2f273d77e59..55784a9f26c4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -1045,21 +1045,28 @@ int amdgpu_gmc_vram_checking(struct amdgpu_device *adev) * seconds, so here, we just pick up three parts for emulation. */ ret = memcmp(vram_ptr, cptr, 10); - if (ret) - return ret; + if (ret) { + ret = -EIO; + goto release_buffer; + } ret = memcmp(vram_ptr + (size / 2), cptr, 10); - if (ret) - return ret; + if (ret) { + ret = -EIO; + goto release_buffer; + } ret = memcmp(vram_ptr + size - 10, cptr, 10); - if (ret) - return ret; + if (ret) { + ret = -EIO; + goto release_buffer; + } +release_buffer: amdgpu_bo_free_kernel(&vram_bo, &vram_gpu, &vram_ptr); - return 0; + return ret; } static ssize_t current_memory_partition_show( diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index b5ebafd4a3ad..bf4f48fe438d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -1105,7 +1105,12 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) if (amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_AVG_POWER, (void *)&ui32, &ui32_size)) { - return -EINVAL; + /* fall back to input power for backwards compat */ + if (amdgpu_dpm_read_sensor(adev, + AMDGPU_PP_SENSOR_GPU_INPUT_POWER, + (void *)&ui32, &ui32_size)) { + return -EINVAL; + } } ui32 >>= 8; break; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c index 210aea590a52..59fafb8392e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c @@ -218,6 +218,7 @@ static void amdgpu_mca_smu_mca_bank_dump(struct amdgpu_device *adev, int idx, st int amdgpu_mca_smu_log_ras_error(struct amdgpu_device *adev, enum amdgpu_ras_block blk, enum amdgpu_mca_error_type type, struct ras_err_data *err_data) { struct amdgpu_smuio_mcm_config_info mcm_info; + struct ras_err_addr err_addr = {0}; struct mca_bank_set mca_set; struct mca_bank_node *node; struct mca_bank_entry *entry; @@ -246,10 +247,18 @@ int amdgpu_mca_smu_log_ras_error(struct amdgpu_device *adev, enum amdgpu_ras_blo mcm_info.socket_id = entry->info.socket_id; mcm_info.die_id = entry->info.aid; + if (blk == AMDGPU_RAS_BLOCK__UMC) { + err_addr.err_status = entry->regs[MCA_REG_IDX_STATUS]; + err_addr.err_ipid = entry->regs[MCA_REG_IDX_IPID]; + err_addr.err_addr = entry->regs[MCA_REG_IDX_ADDR]; + } + if (type == AMDGPU_MCA_ERROR_TYPE_UE) - amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, (uint64_t)count); + amdgpu_ras_error_statistic_ue_count(err_data, + &mcm_info, &err_addr, (uint64_t)count); else - amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, (uint64_t)count); + amdgpu_ras_error_statistic_ce_count(err_data, + &mcm_info, &err_addr, (uint64_t)count); } out_mca_release: @@ -351,6 +360,9 @@ int amdgpu_mca_smu_get_mca_entry(struct amdgpu_device *adev, enum amdgpu_mca_err const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs; int count; + if (!mca_funcs || !mca_funcs->mca_get_mca_entry) + return -EOPNOTSUPP; + switch (type) { case AMDGPU_MCA_ERROR_TYPE_UE: count = mca_funcs->max_ue_count; @@ -365,10 +377,7 @@ int amdgpu_mca_smu_get_mca_entry(struct amdgpu_device *adev, enum amdgpu_mca_err if (idx >= count) return -EINVAL; - if (mca_funcs && mca_funcs->mca_get_mca_entry) - return mca_funcs->mca_get_mca_entry(adev, type, idx, entry); - - return -EOPNOTSUPP; + return mca_funcs->mca_get_mca_entry(adev, type, idx, entry); } #if defined(CONFIG_DEBUG_FS) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 1bf975b8d083..0328616473f8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -466,7 +466,7 @@ static int psp_sw_init(void *handle) } ret = amdgpu_bo_create_kernel(adev, PSP_1_MEG, PSP_1_MEG, - amdgpu_sriov_vf(adev) ? + (amdgpu_sriov_vf(adev) || adev->debug_use_vram_fw_buf) ? AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT, &psp->fw_pri_bo, &psp->fw_pri_mc_addr, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index bacb59d8b701..31823a30dea2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -305,11 +305,13 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, return -EINVAL; data->head.block = block_id; - /* only ue and ce errors are supported */ + /* only ue, ce and poison errors are supported */ if (!memcmp("ue", err, 2)) data->head.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE; else if (!memcmp("ce", err, 2)) data->head.type = AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE; + else if (!memcmp("poison", err, 6)) + data->head.type = AMDGPU_RAS_ERROR__POISON; else return -EINVAL; @@ -431,9 +433,10 @@ static void amdgpu_ras_instance_mask_check(struct amdgpu_device *adev, * The block is one of: umc, sdma, gfx, etc. * see ras_block_string[] for details * - * The error type is one of: ue, ce, where, + * The error type is one of: ue, ce and poison where, * ue is multi-uncorrectable * ce is single-correctable + * poison is poison * * The sub-block is a the sub-block index, pass 0 if there is no sub-block. * The address and value are hexadecimal numbers, leading 0x is optional. @@ -1067,8 +1070,7 @@ static void amdgpu_ras_error_print_error_data(struct amdgpu_device *adev, mcm_info = &err_info->mcm_info; if (err_info->ce_count) { dev_info(adev->dev, "socket: %d, die: %d, " - "%lld new correctable hardware errors detected in %s block, " - "no user action is needed\n", + "%lld new correctable hardware errors detected in %s block\n", mcm_info->socket_id, mcm_info->die_id, err_info->ce_count, @@ -1080,8 +1082,7 @@ static void amdgpu_ras_error_print_error_data(struct amdgpu_device *adev, err_info = &err_node->err_info; mcm_info = &err_info->mcm_info; dev_info(adev->dev, "socket: %d, die: %d, " - "%lld correctable hardware errors detected in total in %s block, " - "no user action is needed\n", + "%lld correctable hardware errors detected in total in %s block\n", mcm_info->socket_id, mcm_info->die_id, err_info->ce_count, blk_name); } } @@ -1108,16 +1109,14 @@ static void amdgpu_ras_error_generate_report(struct amdgpu_device *adev, adev->smuio.funcs->get_die_id) { dev_info(adev->dev, "socket: %d, die: %d " "%ld correctable hardware errors " - "detected in %s block, no user " - "action is needed.\n", + "detected in %s block\n", adev->smuio.funcs->get_socket_id(adev), adev->smuio.funcs->get_die_id(adev), ras_mgr->err_data.ce_count, blk_name); } else { dev_info(adev->dev, "%ld correctable hardware errors " - "detected in %s block, no user " - "action is needed.\n", + "detected in %s block\n", ras_mgr->err_data.ce_count, blk_name); } @@ -1156,8 +1155,10 @@ static void amdgpu_rasmgr_error_data_statistic_update(struct ras_manager *obj, s for_each_ras_error(err_node, err_data) { err_info = &err_node->err_info; - amdgpu_ras_error_statistic_ce_count(&obj->err_data, &err_info->mcm_info, err_info->ce_count); - amdgpu_ras_error_statistic_ue_count(&obj->err_data, &err_info->mcm_info, err_info->ue_count); + amdgpu_ras_error_statistic_ce_count(&obj->err_data, + &err_info->mcm_info, NULL, err_info->ce_count); + amdgpu_ras_error_statistic_ue_count(&obj->err_data, + &err_info->mcm_info, NULL, err_info->ue_count); } } else { /* for legacy asic path which doesn't has error source info */ @@ -1174,6 +1175,9 @@ static int amdgpu_ras_query_error_status_helper(struct amdgpu_device *adev, enum amdgpu_ras_block blk = info ? info->head.block : AMDGPU_RAS_BLOCK_COUNT; struct amdgpu_ras_block_object *block_obj = NULL; + if (blk == AMDGPU_RAS_BLOCK_COUNT) + return -EINVAL; + if (error_query_mode == AMDGPU_RAS_INVALID_ERROR_QUERY) return -EINVAL; @@ -1915,7 +1919,7 @@ static void amdgpu_ras_interrupt_poison_creation_handler(struct ras_manager *obj struct amdgpu_iv_entry *entry) { dev_info(obj->adev->dev, - "Poison is created, no user action is needed.\n"); + "Poison is created\n"); } static void amdgpu_ras_interrupt_umc_handler(struct ras_manager *obj, @@ -2538,7 +2542,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev) return 0; data = &con->eh_data; - *data = kmalloc(sizeof(**data), GFP_KERNEL | __GFP_ZERO); + *data = kzalloc(sizeof(**data), GFP_KERNEL); if (!*data) { ret = -ENOMEM; goto out; @@ -2825,10 +2829,10 @@ int amdgpu_ras_init(struct amdgpu_device *adev) if (con) return 0; - con = kmalloc(sizeof(struct amdgpu_ras) + + con = kzalloc(sizeof(*con) + sizeof(struct ras_manager) * AMDGPU_RAS_BLOCK_COUNT + sizeof(struct ras_manager) * AMDGPU_RAS_MCA_BLOCK_COUNT, - GFP_KERNEL|__GFP_ZERO); + GFP_KERNEL); if (!con) return -ENOMEM; @@ -2915,6 +2919,11 @@ int amdgpu_ras_init(struct amdgpu_device *adev) amdgpu_ras_query_poison_mode(adev); + /* Packed socket_id to ras feature mask bits[31:29] */ + if (adev->smuio.funcs && + adev->smuio.funcs->get_socket_id) + con->features |= ((adev->smuio.funcs->get_socket_id(adev)) << 29); + /* Get RAS schema for particular SOC */ con->schema = amdgpu_get_ras_schema(adev); @@ -3133,8 +3142,7 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev)) return 0; - /* enable MCA debug on APU device */ - amdgpu_ras_set_mca_debug_mode(adev, !!(adev->flags & AMD_IS_APU)); + amdgpu_ras_set_mca_debug_mode(adev, false); list_for_each_entry_safe(node, tmp, &adev->ras_list, node) { if (!node->ras_obj) { @@ -3691,7 +3699,8 @@ static int ras_err_info_cmp(void *priv, const struct list_head *a, const struct } static struct ras_err_info *amdgpu_ras_error_get_info(struct ras_err_data *err_data, - struct amdgpu_smuio_mcm_config_info *mcm_info) + struct amdgpu_smuio_mcm_config_info *mcm_info, + struct ras_err_addr *err_addr) { struct ras_err_node *err_node; @@ -3705,6 +3714,9 @@ static struct ras_err_info *amdgpu_ras_error_get_info(struct ras_err_data *err_d memcpy(&err_node->err_info.mcm_info, mcm_info, sizeof(*mcm_info)); + if (err_addr) + memcpy(&err_node->err_info.err_addr, err_addr, sizeof(*err_addr)); + err_data->err_list_count++; list_add_tail(&err_node->node, &err_data->err_node_list); list_sort(NULL, &err_data->err_node_list, ras_err_info_cmp); @@ -3713,7 +3725,8 @@ static struct ras_err_info *amdgpu_ras_error_get_info(struct ras_err_data *err_d } int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data, - struct amdgpu_smuio_mcm_config_info *mcm_info, u64 count) + struct amdgpu_smuio_mcm_config_info *mcm_info, + struct ras_err_addr *err_addr, u64 count) { struct ras_err_info *err_info; @@ -3723,7 +3736,7 @@ int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data, if (!count) return 0; - err_info = amdgpu_ras_error_get_info(err_data, mcm_info); + err_info = amdgpu_ras_error_get_info(err_data, mcm_info, err_addr); if (!err_info) return -EINVAL; @@ -3734,7 +3747,8 @@ int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data, } int amdgpu_ras_error_statistic_ce_count(struct ras_err_data *err_data, - struct amdgpu_smuio_mcm_config_info *mcm_info, u64 count) + struct amdgpu_smuio_mcm_config_info *mcm_info, + struct ras_err_addr *err_addr, u64 count) { struct ras_err_info *err_info; @@ -3744,7 +3758,7 @@ int amdgpu_ras_error_statistic_ce_count(struct ras_err_data *err_data, if (!count) return 0; - err_info = amdgpu_ras_error_get_info(err_data, mcm_info); + err_info = amdgpu_ras_error_get_info(err_data, mcm_info, err_addr); if (!err_info) return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index 6a941eb8fb8f..76fb85628716 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -452,10 +452,17 @@ struct ras_fs_data { char debugfs_name[32]; }; +struct ras_err_addr { + uint64_t err_status; + uint64_t err_ipid; + uint64_t err_addr; +}; + struct ras_err_info { struct amdgpu_smuio_mcm_config_info mcm_info; u64 ce_count; u64 ue_count; + struct ras_err_addr err_addr; }; struct ras_err_node { @@ -806,8 +813,10 @@ void amdgpu_ras_inst_reset_ras_error_count(struct amdgpu_device *adev, int amdgpu_ras_error_data_init(struct ras_err_data *err_data); void amdgpu_ras_error_data_fini(struct ras_err_data *err_data); int amdgpu_ras_error_statistic_ce_count(struct ras_err_data *err_data, - struct amdgpu_smuio_mcm_config_info *mcm_info, u64 count); + struct amdgpu_smuio_mcm_config_info *mcm_info, + struct ras_err_addr *err_addr, u64 count); int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data, - struct amdgpu_smuio_mcm_config_info *mcm_info, u64 count); + struct amdgpu_smuio_mcm_config_info *mcm_info, + struct ras_err_addr *err_addr, u64 count); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h index b0335a1c5e90..19899f6b9b2b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h @@ -32,7 +32,6 @@ enum AMDGPU_RESET_FLAGS { AMDGPU_NEED_FULL_RESET = 0, AMDGPU_SKIP_HW_RESET = 1, - AMDGPU_RESET_FOR_DEVICE_REMOVE = 2, }; struct amdgpu_reset_context { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c index 35e0ae9acadc..2c3675d91614 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c @@ -531,13 +531,12 @@ int amdgpu_gfx_rlc_init_microcode(struct amdgpu_device *adev, if (version_major == 2 && version_minor == 1) adev->gfx.rlc.is_rlc_v2_1 = true; - if (version_minor >= 0) { - err = amdgpu_gfx_rlc_init_microcode_v2_0(adev); - if (err) { - dev_err(adev->dev, "fail to init rlc v2_0 microcode\n"); - return err; - } + err = amdgpu_gfx_rlc_init_microcode_v2_0(adev); + if (err) { + dev_err(adev->dev, "fail to init rlc v2_0 microcode\n"); + return err; } + if (version_minor >= 1) amdgpu_gfx_rlc_init_microcode_v2_1(adev); if (version_minor >= 2) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c index dcd8c066bc1f..1b013a44ca99 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c @@ -191,7 +191,8 @@ static bool amdgpu_sync_test_fence(struct amdgpu_device *adev, /* Never sync to VM updates either. */ if (fence_owner == AMDGPU_FENCE_OWNER_VM && - owner != AMDGPU_FENCE_OWNER_UNDEFINED) + owner != AMDGPU_FENCE_OWNER_UNDEFINED && + owner != AMDGPU_FENCE_OWNER_KFD) return false; /* Ignore fences depending on the sync mode */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index b14127429f30..3e12763e477a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -1062,7 +1062,8 @@ int amdgpu_ucode_create_bo(struct amdgpu_device *adev) { if (adev->firmware.load_type != AMDGPU_FW_LOAD_DIRECT) { amdgpu_bo_create_kernel(adev, adev->firmware.fw_size, PAGE_SIZE, - amdgpu_sriov_vf(adev) ? AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT, + (amdgpu_sriov_vf(adev) || adev->debug_use_vram_fw_buf) ? + AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT, &adev->firmware.fw_buf, &adev->firmware.fw_buf_mc, &adev->firmware.fw_buf_ptr); @@ -1397,9 +1398,13 @@ int amdgpu_ucode_request(struct amdgpu_device *adev, const struct firmware **fw, if (err) return -ENODEV; + err = amdgpu_ucode_validate(*fw); - if (err) + if (err) { dev_dbg(adev->dev, "\"%s\" failed to validate\n", fw_name); + release_firmware(*fw); + *fw = NULL; + } return err; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 7da71b6a9dc6..b8fcb6c55698 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -285,6 +285,7 @@ static void amdgpu_vm_bo_reset_state_machine(struct amdgpu_vm *vm) list_for_each_entry_safe(vm_bo, tmp, &vm->idle, vm_status) { struct amdgpu_bo *bo = vm_bo->bo; + vm_bo->moved = true; if (!bo || bo->tbo.type != ttm_bo_type_kernel) list_move(&vm_bo->vm_status, &vm_bo->vm->moved); else if (bo->parent) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index b6cd565562ad..4740dd65b99d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -116,7 +116,7 @@ struct amdgpu_mem_stats; #define AMDGPU_VM_FAULT_STOP_FIRST 1 #define AMDGPU_VM_FAULT_STOP_ALWAYS 2 -/* Reserve 4MB VRAM for page tables */ +/* How much VRAM be reserved for page tables */ #define AMDGPU_VM_RESERVED_VRAM (8ULL << 20) /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c index 6f149b54d4d3..b9a15d51eb5c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c @@ -59,11 +59,8 @@ static inline uint16_t complete_integer_division_u16( static uint16_t vpe_u1_8_from_fraction(uint16_t numerator, uint16_t denominator) { - bool arg1_negative = numerator < 0; - bool arg2_negative = denominator < 0; - - uint16_t arg1_value = (uint16_t)(arg1_negative ? -numerator : numerator); - uint16_t arg2_value = (uint16_t)(arg2_negative ? -denominator : denominator); + u16 arg1_value = numerator; + u16 arg2_value = denominator; uint16_t remainder; @@ -100,9 +97,6 @@ static uint16_t vpe_u1_8_from_fraction(uint16_t numerator, uint16_t denominator) res_value += summand; } - if (arg1_negative ^ arg2_negative) - res_value = -res_value; - return res_value; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 9a95b9f226b8..a6c88f2fe6e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -1313,10 +1313,10 @@ static void __xgmi_v6_4_0_query_error_count(struct amdgpu_device *adev, struct a switch (xgmi_v6_4_0_pcs_mca_get_error_type(adev, status)) { case AMDGPU_MCA_ERROR_TYPE_UE: - amdgpu_ras_error_statistic_ue_count(err_data, mcm_info, 1ULL); + amdgpu_ras_error_statistic_ue_count(err_data, mcm_info, NULL, 1ULL); break; case AMDGPU_MCA_ERROR_TYPE_CE: - amdgpu_ras_error_statistic_ce_count(err_data, mcm_info, 1ULL); + amdgpu_ras_error_statistic_ce_count(err_data, mcm_info, NULL, 1ULL); break; default: break; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h index 6cab882e8061..1592c63b3099 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -43,7 +43,6 @@ struct amdgpu_hive_info { } pstate; struct amdgpu_reset_domain *reset_domain; - uint32_t device_remove_count; atomic_t ras_recovery; }; diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c index f0737fb3a999..d1bba9c64e16 100644 --- a/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c @@ -30,6 +30,8 @@ #define regATHUB_MISC_CNTL_V3_0_1 0x00d7 #define regATHUB_MISC_CNTL_V3_0_1_BASE_IDX 0 +#define regATHUB_MISC_CNTL_V3_3_0 0x00d8 +#define regATHUB_MISC_CNTL_V3_3_0_BASE_IDX 0 static uint32_t athub_v3_0_get_cg_cntl(struct amdgpu_device *adev) @@ -40,6 +42,9 @@ static uint32_t athub_v3_0_get_cg_cntl(struct amdgpu_device *adev) case IP_VERSION(3, 0, 1): data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL_V3_0_1); break; + case IP_VERSION(3, 3, 0): + data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL_V3_3_0); + break; default: data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL); break; @@ -53,6 +58,9 @@ static void athub_v3_0_set_cg_cntl(struct amdgpu_device *adev, uint32_t data) case IP_VERSION(3, 0, 1): WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL_V3_0_1, data); break; + case IP_VERSION(3, 3, 0): + WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL_V3_3_0, data); + break; default: WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL, data); break; diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c index 2c221000782c..a33e890c70d9 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.c +++ b/drivers/gpu/drm/amd/amdgpu/atom.c @@ -395,7 +395,6 @@ static void atom_skip_src_int(atom_exec_context *ctx, uint8_t attr, int *ptr) (*ptr)++; return; } - return; } } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 73f6d7e72c73..d63cab294883 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -3996,16 +3996,13 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev) if (!amdgpu_sriov_vf(adev)) { snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", ucode_prefix); - err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, fw_name); - /* don't check this. There are apparently firmwares in the wild with - * incorrect size in the header - */ - if (err == -ENODEV) - goto out; + err = request_firmware(&adev->gfx.rlc_fw, fw_name, adev->dev); if (err) - dev_dbg(adev->dev, - "gfx10: amdgpu_ucode_request() failed \"%s\"\n", - fw_name); + goto out; + + /* don't validate this firmware. There are apparently firmwares + * in the wild with incorrect size in the header + */ rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; version_major = le16_to_cpu(rlc_hdr->header.header_version_major); version_minor = le16_to_cpu(rlc_hdr->header.header_version_minor); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index bdcf96df69e6..0ea0866c261f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -115,7 +115,7 @@ static const struct soc15_reg_golden golden_settings_gc_11_5_0[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, regGL2C_ADDR_MATCH_MASK, 0xffffffff, 0xfffffff3), SOC15_REG_GOLDEN_VALUE(GC, 0, regGL2C_CTRL, 0xffffffff, 0xf37fff3f), SOC15_REG_GOLDEN_VALUE(GC, 0, regGL2C_CTRL3, 0xfffffffb, 0x00f40188), - SOC15_REG_GOLDEN_VALUE(GC, 0, regGL2C_CTRL4, 0xf0ffffff, 0x8000b007), + SOC15_REG_GOLDEN_VALUE(GC, 0, regGL2C_CTRL4, 0xf0ffffff, 0x80009007), SOC15_REG_GOLDEN_VALUE(GC, 0, regPA_CL_ENHANCE, 0xf1ffffff, 0x00880007), SOC15_REG_GOLDEN_VALUE(GC, 0, regPC_CONFIG_CNTL_1, 0xffffffff, 0x00010000), SOC15_REG_GOLDEN_VALUE(GC, 0, regTA_CNTL_AUX, 0xf7f7ffff, 0x01030000), @@ -4474,11 +4474,43 @@ static int gfx_v11_0_wait_for_idle(void *handle) return -ETIMEDOUT; } +static int gfx_v11_0_request_gfx_index_mutex(struct amdgpu_device *adev, + int req) +{ + u32 i, tmp, val; + + for (i = 0; i < adev->usec_timeout; i++) { + /* Request with MeId=2, PipeId=0 */ + tmp = REG_SET_FIELD(0, CP_GFX_INDEX_MUTEX, REQUEST, req); + tmp = REG_SET_FIELD(tmp, CP_GFX_INDEX_MUTEX, CLIENTID, 4); + WREG32_SOC15(GC, 0, regCP_GFX_INDEX_MUTEX, tmp); + + val = RREG32_SOC15(GC, 0, regCP_GFX_INDEX_MUTEX); + if (req) { + if (val == tmp) + break; + } else { + tmp = REG_SET_FIELD(tmp, CP_GFX_INDEX_MUTEX, + REQUEST, 1); + + /* unlocked or locked by firmware */ + if (val != tmp) + break; + } + udelay(1); + } + + if (i >= adev->usec_timeout) + return -EINVAL; + + return 0; +} + static int gfx_v11_0_soft_reset(void *handle) { u32 grbm_soft_reset = 0; u32 tmp; - int i, j, k; + int r, i, j, k; struct amdgpu_device *adev = (struct amdgpu_device *)handle; tmp = RREG32_SOC15(GC, 0, regCP_INT_CNTL); @@ -4518,6 +4550,13 @@ static int gfx_v11_0_soft_reset(void *handle) } } + /* Try to acquire the gfx mutex before access to CP_VMID_RESET */ + r = gfx_v11_0_request_gfx_index_mutex(adev, 1); + if (r) { + DRM_ERROR("Failed to acquire the gfx mutex during soft reset\n"); + return r; + } + WREG32_SOC15(GC, 0, regCP_VMID_RESET, 0xfffffffe); // Read CP_VMID_RESET register three times. @@ -4526,6 +4565,13 @@ static int gfx_v11_0_soft_reset(void *handle) RREG32_SOC15(GC, 0, regCP_VMID_RESET); RREG32_SOC15(GC, 0, regCP_VMID_RESET); + /* release the gfx mutex */ + r = gfx_v11_0_request_gfx_index_mutex(adev, 0); + if (r) { + DRM_ERROR("Failed to release the gfx mutex during soft reset\n"); + return r; + } + for (i = 0; i < adev->usec_timeout; i++) { if (!RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) && !RREG32_SOC15(GC, 0, regCP_GFX_HQD_ACTIVE)) @@ -6337,6 +6383,9 @@ static int gfx_v11_0_get_cu_info(struct amdgpu_device *adev, mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { + bitmap = i * adev->gfx.config.max_sh_per_se + j; + if (!((gfx_v11_0_get_sa_active_bitmap(adev) >> bitmap) & 1)) + continue; mask = 1; counter = 0; gfx_v11_0_select_se_sh(adev, i, j, 0xffffffff, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index 00b21ece081f..131cddbdda0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -3828,8 +3828,8 @@ static void gfx_v9_4_3_inst_query_ras_err_count(struct amdgpu_device *adev, /* the caller should make sure initialize value of * err_data->ue_count and err_data->ce_count */ - amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, ue_count); - amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, ce_count); + amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, NULL, ue_count); + amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, NULL, ce_count); } static void gfx_v9_4_3_inst_reset_ras_err_count(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index 53a2ba5fcf4b..22175da0e16a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -102,7 +102,9 @@ static void gfxhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) WREG32_SOC15_RLC(GC, 0, mmMC_VM_SYSTEM_APERTURE_LOW_ADDR, min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); - if (adev->apu_flags & AMD_APU_IS_RAVEN2) + if (adev->apu_flags & (AMD_APU_IS_RAVEN2 | + AMD_APU_IS_RENOIR | + AMD_APU_IS_GREEN_SARDINE)) /* * Raven2 has a HW issue that it is unable to use the * vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c index 55423ff1bb49..49aecdcee006 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c @@ -139,7 +139,9 @@ gfxhub_v1_2_xcc_init_system_aperture_regs(struct amdgpu_device *adev, WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_LOW_ADDR, min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); - if (adev->apu_flags & AMD_APU_IS_RAVEN2) + if (adev->apu_flags & (AMD_APU_IS_RAVEN2 | + AMD_APU_IS_RENOIR | + AMD_APU_IS_GREEN_SARDINE)) /* * Raven2 has a HW issue that it is unable to use the * vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. @@ -454,10 +456,12 @@ static void gfxhub_v1_2_xcc_gart_disable(struct amdgpu_device *adev, WREG32_SOC15_RLC(GC, GET_INST(GC, j), regMC_VM_MX_L1_TLB_CNTL, tmp); /* Setup L2 cache */ - tmp = RREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0); - WREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL, tmp); - WREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL3, 0); + if (!amdgpu_sriov_vf(adev)) { + tmp = RREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0); + WREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL, tmp); + WREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL3, 0); + } } } diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index a5a05c16c10d..6c5185608854 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -1041,6 +1041,10 @@ static int gmc_v10_0_hw_fini(void *handle) amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0); + if (adev->gmc.ecc_irq.funcs && + amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) + amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0); + return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index 23d7b548d13f..c9c653cfc765 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -941,6 +941,11 @@ static int gmc_v11_0_hw_fini(void *handle) } amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0); + + if (adev->gmc.ecc_irq.funcs && + amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) + amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0); + gmc_v11_0_gart_disable(adev); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 473a774294ce..f9039d64ff2d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -2380,6 +2380,10 @@ static int gmc_v9_0_hw_fini(void *handle) amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0); + if (adev->gmc.ecc_irq.funcs && + amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) + amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0); + return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 843219a91736..e3ddd22aa172 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -96,7 +96,9 @@ static void mmhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) WREG32_SOC15(MMHUB, 0, mmMC_VM_SYSTEM_APERTURE_LOW_ADDR, min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); - if (adev->apu_flags & AMD_APU_IS_RAVEN2) + if (adev->apu_flags & (AMD_APU_IS_RAVEN2 | + AMD_APU_IS_RENOIR | + AMD_APU_IS_GREEN_SARDINE)) /* * Raven2 has a HW issue that it is unable to use the vram which * is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. So here is the diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c index 9b0146732e13..fb53aacdcba2 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c @@ -652,8 +652,8 @@ static void mmhub_v1_8_inst_query_ras_error_count(struct amdgpu_device *adev, AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, &ue_count); - amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, ce_count); - amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, ue_count); + amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, NULL, ce_count); + amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, NULL, ue_count); } static void mmhub_v1_8_query_ras_error_count(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c index 6d24c84924cb..19986ff6a48d 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c @@ -401,8 +401,7 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device if (err_data.ce_count) dev_info(adev->dev, "%ld correctable hardware " - "errors detected in %s block, " - "no user action is needed.\n", + "errors detected in %s block\n", obj->err_data.ce_count, get_ras_block_str(adev->nbio.ras_if)); diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c index 25a3da83e0fb..e90f33780803 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c @@ -597,8 +597,7 @@ static void nbio_v7_9_handle_ras_controller_intr_no_bifring(struct amdgpu_device if (err_data.ce_count) dev_info(adev->dev, "%ld correctable hardware " - "errors detected in %s block, " - "no user action is needed.\n", + "errors detected in %s block\n", obj->err_data.ce_count, get_ras_block_str(adev->nbio.ras_if)); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c index 0f24af6f2810..2d688dca26be 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c @@ -2156,7 +2156,7 @@ static void sdma_v4_4_2_inst_query_ras_error_count(struct amdgpu_device *adev, AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, &ue_count); - amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, ue_count); + amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, NULL, ue_count); } static void sdma_v4_4_2_query_ras_error_count(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c index e9c2ff74f0bc..7458a218e89d 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c @@ -26,6 +26,7 @@ #include "amdgpu.h" #include "umc/umc_12_0_0_offset.h" #include "umc/umc_12_0_0_sh_mask.h" +#include "mp/mp_13_0_6_sh_mask.h" const uint32_t umc_v12_0_channel_idx_tbl[] @@ -88,16 +89,26 @@ static void umc_v12_0_reset_error_count(struct amdgpu_device *adev) umc_v12_0_reset_error_count_per_channel, NULL); } -bool umc_v12_0_is_uncorrectable_error(uint64_t mc_umc_status) +bool umc_v12_0_is_uncorrectable_error(struct amdgpu_device *adev, uint64_t mc_umc_status) { + if (amdgpu_ras_is_poison_mode_supported(adev) && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1)) + return true; + return ((REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC) == 1 || REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UC) == 1 || REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, TCC) == 1)); } -bool umc_v12_0_is_correctable_error(uint64_t mc_umc_status) +bool umc_v12_0_is_correctable_error(struct amdgpu_device *adev, uint64_t mc_umc_status) { + if (amdgpu_ras_is_poison_mode_supported(adev) && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1)) + return false; + return (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1 || (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 && @@ -105,7 +116,7 @@ bool umc_v12_0_is_correctable_error(uint64_t mc_umc_status) /* Identify data parity error in replay mode */ ((REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, ErrorCodeExt) == 0x5 || REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, ErrorCodeExt) == 0xb) && - !(umc_v12_0_is_uncorrectable_error(mc_umc_status))))); + !(umc_v12_0_is_uncorrectable_error(adev, mc_umc_status))))); } static void umc_v12_0_query_correctable_error_count(struct amdgpu_device *adev, @@ -124,7 +135,7 @@ static void umc_v12_0_query_correctable_error_count(struct amdgpu_device *adev, mc_umc_status = RREG64_PCIE_EXT((mc_umc_status_addr + umc_reg_offset) * 4); - if (umc_v12_0_is_correctable_error(mc_umc_status)) + if (umc_v12_0_is_correctable_error(adev, mc_umc_status)) *error_count += 1; } @@ -142,7 +153,7 @@ static void umc_v12_0_query_uncorrectable_error_count(struct amdgpu_device *adev mc_umc_status = RREG64_PCIE_EXT((mc_umc_status_addr + umc_reg_offset) * 4); - if (umc_v12_0_is_uncorrectable_error(mc_umc_status)) + if (umc_v12_0_is_uncorrectable_error(adev, mc_umc_status)) *error_count += 1; } @@ -166,8 +177,8 @@ static int umc_v12_0_query_error_count(struct amdgpu_device *adev, umc_v12_0_query_correctable_error_count(adev, umc_reg_offset, &ce_count); umc_v12_0_query_uncorrectable_error_count(adev, umc_reg_offset, &ue_count); - amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, ue_count); - amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, ce_count); + amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, NULL, ue_count); + amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, NULL, ce_count); return 0; } @@ -360,6 +371,59 @@ static int umc_v12_0_err_cnt_init_per_channel(struct amdgpu_device *adev, return 0; } +static void umc_v12_0_ecc_info_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status) +{ + amdgpu_mca_smu_log_ras_error(adev, + AMDGPU_RAS_BLOCK__UMC, AMDGPU_MCA_ERROR_TYPE_CE, ras_error_status); + amdgpu_mca_smu_log_ras_error(adev, + AMDGPU_RAS_BLOCK__UMC, AMDGPU_MCA_ERROR_TYPE_UE, ras_error_status); +} + +static void umc_v12_0_ecc_info_query_ras_error_address(struct amdgpu_device *adev, + void *ras_error_status) +{ + struct ras_err_node *err_node; + uint64_t mc_umc_status; + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + + for_each_ras_error(err_node, err_data) { + mc_umc_status = err_node->err_info.err_addr.err_status; + if (!mc_umc_status) + continue; + + if (umc_v12_0_is_uncorrectable_error(adev, mc_umc_status)) { + uint64_t mca_addr, err_addr, mca_ipid; + uint32_t InstanceIdLo; + struct amdgpu_smuio_mcm_config_info *mcm_info; + + mcm_info = &err_node->err_info.mcm_info; + mca_addr = err_node->err_info.err_addr.err_addr; + mca_ipid = err_node->err_info.err_addr.err_ipid; + + err_addr = REG_GET_FIELD(mca_addr, MCA_UMC_UMC0_MCUMC_ADDRT0, ErrorAddr); + InstanceIdLo = REG_GET_FIELD(mca_ipid, MCMP1_IPIDT0, InstanceIdLo); + + dev_info(adev->dev, "UMC:IPID:0x%llx, aid:%d, inst:%d, ch:%d, err_addr:0x%llx\n", + mca_ipid, + mcm_info->die_id, + MCA_IPID_LO_2_UMC_INST(InstanceIdLo), + MCA_IPID_LO_2_UMC_CH(InstanceIdLo), + err_addr); + + umc_v12_0_convert_error_address(adev, + err_data, err_addr, + MCA_IPID_LO_2_UMC_CH(InstanceIdLo), + MCA_IPID_LO_2_UMC_INST(InstanceIdLo), + mcm_info->die_id); + + /* Clear umc error address content */ + memset(&err_node->err_info.err_addr, + 0, sizeof(err_node->err_info.err_addr)); + } + } +} + static void umc_v12_0_err_cnt_init(struct amdgpu_device *adev) { amdgpu_umc_loop_channels(adev, @@ -386,4 +450,6 @@ struct amdgpu_umc_ras umc_v12_0_ras = { }, .err_cnt_init = umc_v12_0_err_cnt_init, .query_ras_poison_mode = umc_v12_0_query_ras_poison_mode, + .ecc_info_query_ras_error_count = umc_v12_0_ecc_info_query_ras_error_count, + .ecc_info_query_ras_error_address = umc_v12_0_ecc_info_query_ras_error_address, }; diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h index b34b1e358f8b..e8de3a92251a 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h +++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h @@ -117,8 +117,12 @@ (pa) |= (UMC_V12_0_CHANNEL_HASH_CH6(channel_idx, pa) << UMC_V12_0_PA_CH6_BIT); \ } while (0) -bool umc_v12_0_is_uncorrectable_error(uint64_t mc_umc_status); -bool umc_v12_0_is_correctable_error(uint64_t mc_umc_status); +#define MCA_IPID_LO_2_UMC_CH(_ipid_lo) (((((_ipid_lo) >> 20) & 0x1) * 4) + \ + (((_ipid_lo) >> 12) & 0xF)) +#define MCA_IPID_LO_2_UMC_INST(_ipid_lo) (((_ipid_lo) >> 21) & 0x7) + +bool umc_v12_0_is_uncorrectable_error(struct amdgpu_device *adev, uint64_t mc_umc_status); +bool umc_v12_0_is_correctable_error(struct amdgpu_device *adev, uint64_t mc_umc_status); extern const uint32_t umc_v12_0_channel_idx_tbl[] diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c b/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c index 530549314ce4..a3ee3c4c650f 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c @@ -64,7 +64,7 @@ static void umc_v6_7_query_error_status_helper(struct amdgpu_device *adev, uint64_t reg_value; if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1) - dev_info(adev->dev, "Deferred error, no user action is needed.\n"); + dev_info(adev->dev, "Deferred error\n"); if (mc_umc_status) dev_info(adev->dev, "MCA STATUS 0x%llx, umc_reg_offset 0x%x\n", mc_umc_status, umc_reg_offset); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index 62b205dac63a..6604a3f99c5e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -330,12 +330,6 @@ static void kfd_init_apertures_vi(struct kfd_process_device *pdd, uint8_t id) pdd->gpuvm_limit = pdd->dev->kfd->shared_resources.gpuvm_size - 1; - /* dGPUs: the reserved space for kernel - * before SVM - */ - pdd->qpd.cwsr_base = SVM_CWSR_BASE; - pdd->qpd.ib_base = SVM_IB_BASE; - pdd->scratch_base = MAKE_SCRATCH_APP_BASE_VI(); pdd->scratch_limit = MAKE_SCRATCH_APP_LIMIT(pdd->scratch_base); } @@ -345,18 +339,18 @@ static void kfd_init_apertures_v9(struct kfd_process_device *pdd, uint8_t id) pdd->lds_base = MAKE_LDS_APP_BASE_V9(); pdd->lds_limit = MAKE_LDS_APP_LIMIT(pdd->lds_base); - pdd->gpuvm_base = PAGE_SIZE; + /* Raven needs SVM to support graphic handle, etc. Leave the small + * reserved space before SVM on Raven as well, even though we don't + * have to. + * Set gpuvm_base and gpuvm_limit to CANONICAL addresses so that they + * are used in Thunk to reserve SVM. + */ + pdd->gpuvm_base = SVM_USER_BASE; pdd->gpuvm_limit = pdd->dev->kfd->shared_resources.gpuvm_size - 1; pdd->scratch_base = MAKE_SCRATCH_APP_BASE_V9(); pdd->scratch_limit = MAKE_SCRATCH_APP_LIMIT(pdd->scratch_base); - - /* - * Place TBA/TMA on opposite side of VM hole to prevent - * stray faults from triggering SVM on these pages. - */ - pdd->qpd.cwsr_base = pdd->dev->kfd->shared_resources.gpuvm_size; } int kfd_init_apertures(struct kfd_process *process) @@ -413,6 +407,12 @@ int kfd_init_apertures(struct kfd_process *process) return -EINVAL; } } + + /* dGPUs: the reserved space for kernel + * before SVM + */ + pdd->qpd.cwsr_base = SVM_CWSR_BASE; + pdd->qpd.ib_base = SVM_IB_BASE; } dev_dbg(kfd_device, "node id %u\n", id); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c index b854cbf06dce..f856901055d3 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -260,19 +260,6 @@ static void svm_migrate_put_sys_page(unsigned long addr) put_page(page); } -static unsigned long svm_migrate_successful_pages(struct migrate_vma *migrate) -{ - unsigned long cpages = 0; - unsigned long i; - - for (i = 0; i < migrate->npages; i++) { - if (migrate->src[i] & MIGRATE_PFN_VALID && - migrate->src[i] & MIGRATE_PFN_MIGRATE) - cpages++; - } - return cpages; -} - static unsigned long svm_migrate_unsuccessful_pages(struct migrate_vma *migrate) { unsigned long upages = 0; @@ -402,6 +389,7 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange, struct dma_fence *mfence = NULL; struct migrate_vma migrate = { 0 }; unsigned long cpages = 0; + unsigned long mpages = 0; dma_addr_t *scratch; void *buf; int r = -ENOMEM; @@ -450,12 +438,13 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange, r = svm_migrate_copy_to_vram(node, prange, &migrate, &mfence, scratch, ttm_res_offset); migrate_vma_pages(&migrate); - pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n", - svm_migrate_successful_pages(&migrate), cpages, migrate.npages); - svm_migrate_copy_done(adev, mfence); migrate_vma_finalize(&migrate); + mpages = cpages - svm_migrate_unsuccessful_pages(&migrate); + pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n", + mpages, cpages, migrate.npages); + kfd_smi_event_migration_end(node, p->lead_thread->pid, start >> PAGE_SHIFT, end >> PAGE_SHIFT, 0, node->id, trigger); @@ -465,12 +454,12 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange, out_free: kvfree(buf); out: - if (!r && cpages) { + if (!r && mpages) { pdd = svm_range_get_pdd_by_node(prange, node); if (pdd) - WRITE_ONCE(pdd->page_in, pdd->page_in + cpages); + WRITE_ONCE(pdd->page_in, pdd->page_in + mpages); - return cpages; + return mpages; } return r; } @@ -498,7 +487,7 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc, struct vm_area_struct *vma; uint64_t ttm_res_offset; struct kfd_node *node; - unsigned long cpages = 0; + unsigned long mpages = 0; long r = 0; if (start_mgr < prange->start || last_mgr > prange->last) { @@ -540,15 +529,15 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc, pr_debug("failed %ld to migrate\n", r); break; } else { - cpages += r; + mpages += r; } ttm_res_offset += next - addr; addr = next; } - if (cpages) { + if (mpages) { prange->actual_loc = best_loc; - prange->vram_pages = prange->vram_pages + cpages; + prange->vram_pages += mpages; } else if (!prange->actual_loc) { /* if no page migrated and all pages from prange are at * sys ram drop svm_bo got from svm_range_vram_node_new @@ -1037,7 +1026,7 @@ int kgd2kfd_init_zone_device(struct amdgpu_device *adev) } else { res = devm_request_free_mem_region(adev->dev, &iomem_resource, size); if (IS_ERR(res)) - return -ENOMEM; + return PTR_ERR(res); pgmap->range.start = res->start; pgmap->range.end = res->end; pgmap->type = MEMORY_DEVICE_PRIVATE; @@ -1053,10 +1042,10 @@ int kgd2kfd_init_zone_device(struct amdgpu_device *adev) r = devm_memremap_pages(adev->dev, pgmap); if (IS_ERR(r)) { pr_err("failed to register HMM device memory\n"); - /* Disable SVM support capability */ - pgmap->type = 0; if (pgmap->type == MEMORY_DEVICE_PRIVATE) devm_release_mem_region(adev->dev, res->start, resource_size(res)); + /* Disable SVM support capability */ + pgmap->type = 0; return PTR_ERR(r); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index 45366b4ca976..17fbedbf3651 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -917,7 +917,7 @@ struct kfd_process { * fence will be triggered during eviction and new one will be created * during restore */ - struct dma_fence *ef; + struct dma_fence __rcu *ef; /* Work items for evicting and restoring BOs */ struct delayed_work eviction_work; @@ -970,7 +970,7 @@ struct kfd_process { struct work_struct debug_event_workarea; /* Tracks debug per-vmid request for debug flags */ - bool dbg_flags; + u32 dbg_flags; atomic_t poison; /* Queues are in paused stated because we are in the process of doing a CRIU checkpoint */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 71df51fcc1b0..717a60d7a4ea 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -1110,6 +1110,7 @@ static void kfd_process_wq_release(struct work_struct *work) { struct kfd_process *p = container_of(work, struct kfd_process, release_work); + struct dma_fence *ef; kfd_process_dequeue_from_all_devices(p); pqm_uninit(&p->pqm); @@ -1118,7 +1119,9 @@ static void kfd_process_wq_release(struct work_struct *work) * destroyed. This allows any BOs to be freed without * triggering pointless evictions or waiting for fences. */ - dma_fence_signal(p->ef); + synchronize_rcu(); + ef = rcu_access_pointer(p->ef); + dma_fence_signal(ef); kfd_process_remove_sysfs(p); @@ -1127,7 +1130,7 @@ static void kfd_process_wq_release(struct work_struct *work) svm_range_list_fini(p); kfd_process_destroy_pdds(p); - dma_fence_put(p->ef); + dma_fence_put(ef); kfd_event_free_process(p); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index 70ca761bd60b..c50a0dc9c9c0 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -158,13 +158,12 @@ svm_is_valid_dma_mapping_addr(struct device *dev, dma_addr_t dma_addr) static int svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange, unsigned long offset, unsigned long npages, - unsigned long *hmm_pfns, uint32_t gpuidx, uint64_t *vram_pages) + unsigned long *hmm_pfns, uint32_t gpuidx) { enum dma_data_direction dir = DMA_BIDIRECTIONAL; dma_addr_t *addr = prange->dma_addr[gpuidx]; struct device *dev = adev->dev; struct page *page; - uint64_t vram_pages_dev; int i, r; if (!addr) { @@ -174,7 +173,6 @@ svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange, prange->dma_addr[gpuidx] = addr; } - vram_pages_dev = 0; addr += offset; for (i = 0; i < npages; i++) { if (svm_is_valid_dma_mapping_addr(dev, addr[i])) @@ -184,7 +182,6 @@ svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange, if (is_zone_device_page(page)) { struct amdgpu_device *bo_adev = prange->svm_bo->node->adev; - vram_pages_dev++; addr[i] = (hmm_pfns[i] << PAGE_SHIFT) + bo_adev->vm_manager.vram_base_offset - bo_adev->kfd.pgmap.range.start; @@ -201,14 +198,14 @@ svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange, pr_debug_ratelimited("dma mapping 0x%llx for page addr 0x%lx\n", addr[i] >> PAGE_SHIFT, page_to_pfn(page)); } - *vram_pages = vram_pages_dev; + return 0; } static int svm_range_dma_map(struct svm_range *prange, unsigned long *bitmap, unsigned long offset, unsigned long npages, - unsigned long *hmm_pfns, uint64_t *vram_pages) + unsigned long *hmm_pfns) { struct kfd_process *p; uint32_t gpuidx; @@ -227,7 +224,7 @@ svm_range_dma_map(struct svm_range *prange, unsigned long *bitmap, } r = svm_range_dma_map_dev(pdd->dev->adev, prange, offset, npages, - hmm_pfns, gpuidx, vram_pages); + hmm_pfns, gpuidx); if (r) break; } @@ -407,14 +404,9 @@ static void svm_range_bo_release(struct kref *kref) spin_lock(&svm_bo->list_lock); } spin_unlock(&svm_bo->list_lock); - if (!dma_fence_is_signaled(&svm_bo->eviction_fence->base)) { - /* We're not in the eviction worker. - * Signal the fence and synchronize with any - * pending eviction work. - */ + if (!dma_fence_is_signaled(&svm_bo->eviction_fence->base)) + /* We're not in the eviction worker. Signal the fence. */ dma_fence_signal(&svm_bo->eviction_fence->base); - cancel_work_sync(&svm_bo->eviction_work); - } dma_fence_put(&svm_bo->eviction_fence->base); amdgpu_bo_unref(&svm_bo->bo); kfree(svm_bo); @@ -885,14 +877,29 @@ static void svm_range_debug_dump(struct svm_range_list *svms) static void * svm_range_copy_array(void *psrc, size_t size, uint64_t num_elements, - uint64_t offset) + uint64_t offset, uint64_t *vram_pages) { + unsigned char *src = (unsigned char *)psrc + offset; unsigned char *dst; + uint64_t i; dst = kvmalloc_array(num_elements, size, GFP_KERNEL); if (!dst) return NULL; - memcpy(dst, (unsigned char *)psrc + offset, num_elements * size); + + if (!vram_pages) { + memcpy(dst, src, num_elements * size); + return (void *)dst; + } + + *vram_pages = 0; + for (i = 0; i < num_elements; i++) { + dma_addr_t *temp; + temp = (dma_addr_t *)dst + i; + *temp = *((dma_addr_t *)src + i); + if (*temp&SVM_RANGE_VRAM_DOMAIN) + (*vram_pages)++; + } return (void *)dst; } @@ -906,7 +913,7 @@ svm_range_copy_dma_addrs(struct svm_range *dst, struct svm_range *src) if (!src->dma_addr[i]) continue; dst->dma_addr[i] = svm_range_copy_array(src->dma_addr[i], - sizeof(*src->dma_addr[i]), src->npages, 0); + sizeof(*src->dma_addr[i]), src->npages, 0, NULL); if (!dst->dma_addr[i]) return -ENOMEM; } @@ -917,7 +924,7 @@ svm_range_copy_dma_addrs(struct svm_range *dst, struct svm_range *src) static int svm_range_split_array(void *ppnew, void *ppold, size_t size, uint64_t old_start, uint64_t old_n, - uint64_t new_start, uint64_t new_n) + uint64_t new_start, uint64_t new_n, uint64_t *new_vram_pages) { unsigned char *new, *old, *pold; uint64_t d; @@ -929,11 +936,12 @@ svm_range_split_array(void *ppnew, void *ppold, size_t size, return 0; d = (new_start - old_start) * size; - new = svm_range_copy_array(pold, size, new_n, d); + /* get dma addr array for new range and calculte its vram page number */ + new = svm_range_copy_array(pold, size, new_n, d, new_vram_pages); if (!new) return -ENOMEM; d = (new_start == old_start) ? new_n * size : 0; - old = svm_range_copy_array(pold, size, old_n, d); + old = svm_range_copy_array(pold, size, old_n, d, NULL); if (!old) { kvfree(new); return -ENOMEM; @@ -955,10 +963,13 @@ svm_range_split_pages(struct svm_range *new, struct svm_range *old, for (i = 0; i < MAX_GPU_INSTANCE; i++) { r = svm_range_split_array(&new->dma_addr[i], &old->dma_addr[i], sizeof(*old->dma_addr[i]), old->start, - npages, new->start, new->npages); + npages, new->start, new->npages, + old->actual_loc ? &new->vram_pages : NULL); if (r) return r; } + if (old->actual_loc) + old->vram_pages -= new->vram_pages; return 0; } @@ -982,11 +993,6 @@ svm_range_split_nodes(struct svm_range *new, struct svm_range *old, new->svm_bo = svm_range_bo_ref(old->svm_bo); new->ttm_res = old->ttm_res; - /* set new's vram_pages as old range's now, the acurate vram_pages - * will be updated during mapping - */ - new->vram_pages = min(old->vram_pages, new->npages); - spin_lock(&new->svm_bo->list_lock); list_add(&new->svm_bo_list, &new->svm_bo->range_list); spin_unlock(&new->svm_bo->list_lock); @@ -1109,7 +1115,7 @@ static int svm_range_split_tail(struct svm_range *prange, uint64_t new_last, struct list_head *insert_list, struct list_head *remap_list) { - struct svm_range *tail; + struct svm_range *tail = NULL; int r = svm_range_split(prange, prange->start, new_last, &tail); if (!r) { @@ -1124,7 +1130,7 @@ static int svm_range_split_head(struct svm_range *prange, uint64_t new_start, struct list_head *insert_list, struct list_head *remap_list) { - struct svm_range *head; + struct svm_range *head = NULL; int r = svm_range_split(prange, new_start, prange->last, &head); if (!r) { @@ -1573,7 +1579,6 @@ static int svm_range_validate_and_map(struct mm_struct *mm, struct svm_validate_context *ctx; unsigned long start, end, addr; struct kfd_process *p; - uint64_t vram_pages; void *owner; int32_t idx; int r = 0; @@ -1648,15 +1653,13 @@ static int svm_range_validate_and_map(struct mm_struct *mm, } } - vram_pages = 0; - start = prange->start << PAGE_SHIFT; - end = (prange->last + 1) << PAGE_SHIFT; + start = map_start << PAGE_SHIFT; + end = (map_last + 1) << PAGE_SHIFT; for (addr = start; !r && addr < end; ) { struct hmm_range *hmm_range; unsigned long map_start_vma; unsigned long map_last_vma; struct vm_area_struct *vma; - uint64_t vram_pages_vma; unsigned long next = 0; unsigned long offset; unsigned long npages; @@ -1683,13 +1686,11 @@ static int svm_range_validate_and_map(struct mm_struct *mm, } if (!r) { - offset = (addr - start) >> PAGE_SHIFT; + offset = (addr >> PAGE_SHIFT) - prange->start; r = svm_range_dma_map(prange, ctx->bitmap, offset, npages, - hmm_range->hmm_pfns, &vram_pages_vma); + hmm_range->hmm_pfns); if (r) pr_debug("failed %d to dma map range\n", r); - else - vram_pages += vram_pages_vma; } svm_range_lock(prange); @@ -1722,19 +1723,6 @@ static int svm_range_validate_and_map(struct mm_struct *mm, addr = next; } - if (addr == end) { - prange->vram_pages = vram_pages; - - /* if prange does not include any vram page and it - * has not released svm_bo drop its svm_bo reference - * and set its actaul_loc to sys ram - */ - if (!vram_pages && prange->ttm_res) { - prange->actual_loc = 0; - svm_range_vram_node_free(prange); - } - } - svm_range_unreserve_bos(ctx); if (!r) prange->validate_timestamp = ktime_get_boottime(); @@ -2352,8 +2340,10 @@ retry: mutex_unlock(&svms->lock); mmap_write_unlock(mm); - /* Pairs with mmget in svm_range_add_list_work */ - mmput(mm); + /* Pairs with mmget in svm_range_add_list_work. If dropping the + * last mm refcount, schedule release work to avoid circular locking + */ + mmput_async(mm); spin_lock(&svms->deferred_list_lock); } @@ -2664,6 +2654,7 @@ svm_range_get_range_boundaries(struct kfd_process *p, int64_t addr, { struct vm_area_struct *vma; struct interval_tree_node *node; + struct rb_node *rb_node; unsigned long start_limit, end_limit; vma = vma_lookup(p->mm, addr << PAGE_SHIFT); @@ -2683,16 +2674,15 @@ svm_range_get_range_boundaries(struct kfd_process *p, int64_t addr, if (node) { end_limit = min(end_limit, node->start); /* Last range that ends before the fault address */ - node = container_of(rb_prev(&node->rb), - struct interval_tree_node, rb); + rb_node = rb_prev(&node->rb); } else { /* Last range must end before addr because * there was no range after addr */ - node = container_of(rb_last(&p->svms.objects.rb_root), - struct interval_tree_node, rb); + rb_node = rb_last(&p->svms.objects.rb_root); } - if (node) { + if (rb_node) { + node = container_of(rb_node, struct interval_tree_node, rb); if (node->last >= addr) { WARN(1, "Overlap with prev node and page fault addr\n"); return -EFAULT; @@ -3439,13 +3429,14 @@ svm_range_trigger_migration(struct mm_struct *mm, struct svm_range *prange, int svm_range_schedule_evict_svm_bo(struct amdgpu_amdkfd_fence *fence) { - if (!fence) - return -EINVAL; - - if (dma_fence_is_signaled(&fence->base)) - return 0; - - if (fence->svm_bo) { + /* Dereferencing fence->svm_bo is safe here because the fence hasn't + * signaled yet and we're under the protection of the fence->lock. + * After the fence is signaled in svm_range_bo_release, we cannot get + * here any more. + * + * Reference is dropped in svm_range_evict_svm_bo_worker. + */ + if (svm_bo_ref_unless_zero(fence->svm_bo)) { WRITE_ONCE(fence->svm_bo->evicting, 1); schedule_work(&fence->svm_bo->eviction_work); } @@ -3460,8 +3451,6 @@ static void svm_range_evict_svm_bo_worker(struct work_struct *work) int r = 0; svm_bo = container_of(work, struct svm_range_bo, eviction_work); - if (!svm_bo_ref_unless_zero(svm_bo)) - return; /* svm_bo was freed while eviction was pending */ if (mmget_not_zero(svm_bo->eviction_fence->mm)) { mm = svm_bo->eviction_fence->mm; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 057284bf50bb..e5f7c92eebcb 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -1342,10 +1342,11 @@ static int kfd_create_indirect_link_prop(struct kfd_topology_device *kdev, int g num_cpu++; } + if (list_empty(&kdev->io_link_props)) + return -ENODATA; + gpu_link = list_first_entry(&kdev->io_link_props, - struct kfd_iolink_properties, list); - if (!gpu_link) - return -ENOMEM; + struct kfd_iolink_properties, list); for (i = 0; i < num_cpu; i++) { /* CPU <--> GPU */ @@ -1423,15 +1424,17 @@ static int kfd_add_peer_prop(struct kfd_topology_device *kdev, peer->gpu->adev)) return ret; + if (list_empty(&kdev->io_link_props)) + return -ENODATA; + iolink1 = list_first_entry(&kdev->io_link_props, - struct kfd_iolink_properties, list); - if (!iolink1) - return -ENOMEM; + struct kfd_iolink_properties, list); + + if (list_empty(&peer->io_link_props)) + return -ENODATA; iolink2 = list_first_entry(&peer->io_link_props, - struct kfd_iolink_properties, list); - if (!iolink2) - return -ENOMEM; + struct kfd_iolink_properties, list); props = kfd_alloc_struct(props); if (!props) @@ -1449,17 +1452,19 @@ static int kfd_add_peer_prop(struct kfd_topology_device *kdev, /* CPU->CPU link*/ cpu_dev = kfd_topology_device_by_proximity_domain(iolink1->node_to); if (cpu_dev) { - list_for_each_entry(iolink3, &cpu_dev->io_link_props, list) - if (iolink3->node_to == iolink2->node_to) - break; - - props->weight += iolink3->weight; - props->min_latency += iolink3->min_latency; - props->max_latency += iolink3->max_latency; - props->min_bandwidth = min(props->min_bandwidth, - iolink3->min_bandwidth); - props->max_bandwidth = min(props->max_bandwidth, - iolink3->max_bandwidth); + list_for_each_entry(iolink3, &cpu_dev->io_link_props, list) { + if (iolink3->node_to != iolink2->node_to) + continue; + + props->weight += iolink3->weight; + props->min_latency += iolink3->min_latency; + props->max_latency += iolink3->max_latency; + props->min_bandwidth = min(props->min_bandwidth, + iolink3->min_bandwidth); + props->max_bandwidth = min(props->max_bandwidth, + iolink3->max_bandwidth); + break; + } } else { WARN(1, "CPU node not found"); } 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 54861136dafd..d55eeb30ccb2 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -37,6 +37,7 @@ #include "dc/dc_dmub_srv.h" #include "dc/dc_edid_parser.h" #include "dc/dc_stat.h" +#include "dc/dc_state.h" #include "amdgpu_dm_trace.h" #include "dpcd_defs.h" #include "link/protocols/link_dpcd.h" @@ -66,7 +67,6 @@ #include "amdgpu_dm_debugfs.h" #endif #include "amdgpu_dm_psr.h" -#include "amdgpu_dm_replay.h" #include "ivsrcid/ivsrcid_vislands30.h" @@ -1294,7 +1294,9 @@ static void mmhub_read_system_context(struct amdgpu_device *adev, struct dc_phy_ /* AGP aperture is disabled */ if (agp_bot > agp_top) { logical_addr_low = adev->gmc.fb_start >> 18; - if (adev->apu_flags & AMD_APU_IS_RAVEN2) + if (adev->apu_flags & (AMD_APU_IS_RAVEN2 | + AMD_APU_IS_RENOIR | + AMD_APU_IS_GREEN_SARDINE)) /* * Raven2 has a HW issue that it is unable to use the vram which * is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. So here is the @@ -1306,7 +1308,9 @@ static void mmhub_read_system_context(struct amdgpu_device *adev, struct dc_phy_ logical_addr_high = adev->gmc.fb_end >> 18; } else { logical_addr_low = min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18; - if (adev->apu_flags & AMD_APU_IS_RAVEN2) + if (adev->apu_flags & (AMD_APU_IS_RAVEN2 | + AMD_APU_IS_RENOIR | + AMD_APU_IS_GREEN_SARDINE)) /* * Raven2 has a HW issue that it is unable to use the vram which * is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. So here is the @@ -1711,6 +1715,8 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) init_data.nbio_reg_offsets = adev->reg_offset[NBIO_HWIP][0]; init_data.clk_reg_offsets = adev->reg_offset[CLK_HWIP][0]; + init_data.flags.disable_ips = DMUB_IPS_DISABLE_ALL; + /* Enable DWB for tested platforms only */ if (amdgpu_ip_version(adev, DCE_HWIP, 0) >= IP_VERSION(3, 0, 0)) init_data.num_virtual_links = 1; @@ -2607,12 +2613,10 @@ static enum dc_status amdgpu_dm_commit_zero_streams(struct dc *dc) memset(del_streams, 0, sizeof(del_streams)); - context = dc_create_state(dc); + context = dc_state_create_current_copy(dc); if (context == NULL) goto context_alloc_fail; - dc_resource_state_copy_construct_current(dc, context); - /* First remove from context all streams */ for (i = 0; i < context->stream_count; i++) { struct dc_stream_state *stream = context->streams[i]; @@ -2622,12 +2626,12 @@ static enum dc_status amdgpu_dm_commit_zero_streams(struct dc *dc) /* Remove all planes for removed streams and then remove the streams */ for (i = 0; i < del_streams_count; i++) { - if (!dc_rem_all_planes_for_stream(dc, del_streams[i], context)) { + if (!dc_state_rem_all_planes_for_stream(dc, del_streams[i], context)) { res = DC_FAIL_DETACH_SURFACES; goto fail; } - res = dc_remove_stream_from_ctx(dc, context, del_streams[i]); + res = dc_state_remove_stream(dc, context, del_streams[i]); if (res != DC_OK) goto fail; } @@ -2635,7 +2639,7 @@ static enum dc_status amdgpu_dm_commit_zero_streams(struct dc *dc) res = dc_commit_streams(dc, context->streams, context->stream_count); fail: - dc_release_state(context); + dc_state_release(context); context_alloc_fail: return res; @@ -2662,7 +2666,7 @@ static int dm_suspend(void *handle) dc_allow_idle_optimizations(adev->dm.dc, false); - dm->cached_dc_state = dc_copy_state(dm->dc->current_state); + dm->cached_dc_state = dc_state_create_copy(dm->dc->current_state); dm_gpureset_toggle_interrupts(adev, dm->cached_dc_state, false); @@ -2856,7 +2860,7 @@ static int dm_resume(void *handle) bool need_hotplug = false; if (dm->dc->caps.ips_support) { - dc_dmub_srv_exit_low_power_state(dm->dc); + dc_dmub_srv_apply_idle_power_optimizations(dm->dc, false); } if (amdgpu_in_reset(adev)) { @@ -2909,7 +2913,7 @@ static int dm_resume(void *handle) dm_gpureset_toggle_interrupts(adev, dm->cached_dc_state, true); - dc_release_state(dm->cached_dc_state); + dc_state_release(dm->cached_dc_state); dm->cached_dc_state = NULL; amdgpu_dm_irq_resume_late(adev); @@ -2919,10 +2923,9 @@ static int dm_resume(void *handle) return 0; } /* Recreate dc_state - DC invalidates it when setting power state to S3. */ - dc_release_state(dm_state->context); - dm_state->context = dc_create_state(dm->dc); + dc_state_release(dm_state->context); + dm_state->context = dc_state_create(dm->dc); /* TODO: Remove dc_state->dccg, use dc->dccg directly. */ - dc_resource_state_construct(dm->dc, dm_state->context); /* Before powering on DC we need to re-initialize DMUB. */ dm_dmub_hw_resume(adev); @@ -3998,7 +4001,7 @@ dm_atomic_duplicate_state(struct drm_private_obj *obj) old_state = to_dm_atomic_state(obj->state); if (old_state && old_state->context) - new_state->context = dc_copy_state(old_state->context); + new_state->context = dc_state_create_copy(old_state->context); if (!new_state->context) { kfree(new_state); @@ -4014,7 +4017,7 @@ static void dm_atomic_destroy_state(struct drm_private_obj *obj, struct dm_atomic_state *dm_state = to_dm_atomic_state(state); if (dm_state && dm_state->context) - dc_release_state(dm_state->context); + dc_state_release(dm_state->context); kfree(dm_state); } @@ -4050,14 +4053,12 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev) if (!state) return -ENOMEM; - state->context = dc_create_state(adev->dm.dc); + state->context = dc_state_create_current_copy(adev->dm.dc); if (!state->context) { kfree(state); return -ENOMEM; } - dc_resource_state_copy_construct_current(adev->dm.dc, state->context); - drm_atomic_private_obj_init(adev_to_drm(adev), &adev->dm.atomic_obj, &state->base, @@ -4065,7 +4066,7 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev) r = amdgpu_display_modeset_create_props(adev); if (r) { - dc_release_state(state->context); + dc_state_release(state->context); kfree(state); return r; } @@ -4077,7 +4078,7 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev) r = amdgpu_dm_audio_init(adev); if (r) { - dc_release_state(state->context); + dc_state_release(state->context); kfree(state); return r; } @@ -4391,7 +4392,6 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) enum dc_connection_type new_connection_type = dc_connection_none; const struct dc_plane_cap *plane; bool psr_feature_enabled = false; - bool replay_feature_enabled = false; int max_overlay = dm->dc->caps.max_slave_planes; dm->display_indexes_num = dm->dc->caps.max_streams; @@ -4503,20 +4503,6 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) } } - if (!(amdgpu_dc_debug_mask & DC_DISABLE_REPLAY)) { - switch (adev->ip_versions[DCE_HWIP][0]) { - case IP_VERSION(3, 1, 4): - case IP_VERSION(3, 1, 5): - case IP_VERSION(3, 1, 6): - case IP_VERSION(3, 2, 0): - case IP_VERSION(3, 2, 1): - replay_feature_enabled = true; - break; - default: - replay_feature_enabled = amdgpu_dc_feature_mask & DC_REPLAY_MASK; - break; - } - } /* loops over all connectors on the board */ for (i = 0; i < link_cnt; i++) { struct dc_link *link = NULL; @@ -4585,12 +4571,6 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) amdgpu_dm_update_connector_after_detect(aconnector); setup_backlight_device(dm, aconnector); - /* - * Disable psr if replay can be enabled - */ - if (replay_feature_enabled && amdgpu_dm_setup_replay(link, aconnector)) - psr_feature_enabled = false; - if (psr_feature_enabled) amdgpu_dm_set_psr_caps(link); @@ -6260,8 +6240,9 @@ create_stream_for_sink(struct drm_connector *connector, if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) mod_build_hf_vsif_infopacket(stream, &stream->vsp_infopacket); - - if (stream->link->psr_settings.psr_feature_enabled || stream->link->replay_settings.replay_feature_enabled) { + else if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT || + stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST || + stream->signal == SIGNAL_TYPE_EDP) { // // should decide stream support vsc sdp colorimetry capability // before building vsc info packet @@ -6277,8 +6258,9 @@ create_stream_for_sink(struct drm_connector *connector, if (stream->out_transfer_func->tf == TRANSFER_FUNCTION_GAMMA22) tf = TRANSFER_FUNC_GAMMA_22; mod_build_vsc_infopacket(stream, &stream->vsc_infopacket, stream->output_color_space, tf); - aconnector->psr_skip_count = AMDGPU_DM_PSR_ENTRY_DELAY; + if (stream->link->psr_settings.psr_feature_enabled) + aconnector->psr_skip_count = AMDGPU_DM_PSR_ENTRY_DELAY; } finish: dc_sink_release(sink); @@ -6658,7 +6640,7 @@ static enum dc_status dm_validate_stream_and_context(struct dc *dc, if (!dc_plane_state) goto cleanup; - dc_state = dc_create_state(dc); + dc_state = dc_state_create(dc); if (!dc_state) goto cleanup; @@ -6685,9 +6667,9 @@ static enum dc_status dm_validate_stream_and_context(struct dc *dc, dc_result = dc_validate_plane(dc, dc_plane_state); if (dc_result == DC_OK) - dc_result = dc_add_stream_to_ctx(dc, dc_state, stream); + dc_result = dc_state_add_stream(dc, dc_state, stream); - if (dc_result == DC_OK && !dc_add_plane_to_context( + if (dc_result == DC_OK && !dc_state_add_plane( dc, stream, dc_plane_state, @@ -6699,7 +6681,7 @@ static enum dc_status dm_validate_stream_and_context(struct dc *dc, cleanup: if (dc_state) - dc_release_state(dc_state); + dc_state_release(dc_state); if (dc_plane_state) dc_plane_state_release(dc_plane_state); @@ -7007,8 +6989,7 @@ static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder, if (IS_ERR(mst_state)) return PTR_ERR(mst_state); - if (!mst_state->pbn_div.full) - mst_state->pbn_div.full = dfixed_const(dm_mst_get_pbn_divider(aconnector->mst_root->dc_link)); + mst_state->pbn_div.full = dfixed_const(dm_mst_get_pbn_divider(aconnector->mst_root->dc_link)); if (!state->duplicated) { int max_bpc = conn_state->max_requested_bpc; @@ -8858,7 +8839,7 @@ static void amdgpu_dm_commit_streams(struct drm_atomic_state *state, dc_stream_get_status(dm_new_crtc_state->stream); if (!status) - status = dc_stream_get_status_from_state(dc_state, + status = dc_state_get_stream_status(dc_state, dm_new_crtc_state->stream); if (!status) drm_err(dev, @@ -9001,7 +8982,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) if (new_con_state->crtc && new_con_state->crtc->state->active && drm_atomic_crtc_needs_modeset(new_con_state->crtc->state)) { - dc_dmub_srv_exit_low_power_state(dm->dc); + dc_dmub_srv_apply_idle_power_optimizations(dm->dc, false); break; } } @@ -9312,10 +9293,10 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) if (!new_con_state->writeback_job) continue; - new_crtc_state = NULL; + new_crtc_state = drm_atomic_get_new_crtc_state(state, &acrtc->base); - if (acrtc) - new_crtc_state = drm_atomic_get_new_crtc_state(state, &acrtc->base); + if (!new_crtc_state) + continue; if (acrtc->wb_enabled) continue; @@ -9783,7 +9764,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, crtc->base.id); /* i.e. reset mode */ - if (dc_remove_stream_from_ctx( + if (dc_state_remove_stream( dm->dc, dm_state->context, dm_old_crtc_state->stream) != DC_OK) { @@ -9826,7 +9807,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, DRM_DEBUG_ATOMIC("Enabling DRM crtc: %d\n", crtc->base.id); - if (dc_add_stream_to_ctx( + if (dc_state_add_stream( dm->dc, dm_state->context, dm_new_crtc_state->stream) != DC_OK) { @@ -10148,7 +10129,7 @@ static int dm_update_plane_state(struct dc *dc, if (ret) return ret; - if (!dc_remove_plane_from_context( + if (!dc_state_remove_plane( dc, dm_old_crtc_state->stream, dm_old_plane_state->dc_state, @@ -10226,7 +10207,7 @@ static int dm_update_plane_state(struct dc *dc, * state. It'll be released when the atomic state is * cleaned. */ - if (!dc_add_plane_to_context( + if (!dc_state_add_plane( dc, dm_new_crtc_state->stream, dc_new_plane_state, @@ -10910,7 +10891,7 @@ static bool dm_edid_parser_send_cea(struct amdgpu_display_manager *dm, input->cea_total_length = total_length; memcpy(input->payload, data, length); - res = dm_execute_dmub_cmd(dm->dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY); + res = dc_wake_and_execute_dmub_cmd(dm->dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY); if (!res) { DRM_ERROR("EDID CEA parser failed\n"); return false; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 2d5af83d40b5..9c1871b866cc 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -747,7 +747,7 @@ enum amdgpu_transfer_function { AMDGPU_TRANSFER_FUNCTION_GAMMA22_INV_EOTF, AMDGPU_TRANSFER_FUNCTION_GAMMA24_INV_EOTF, AMDGPU_TRANSFER_FUNCTION_GAMMA26_INV_EOTF, - AMDGPU_TRANSFER_FUNCTION_COUNT + AMDGPU_TRANSFER_FUNCTION_COUNT }; struct dm_plane_state { @@ -844,7 +844,7 @@ struct dm_crtc_state { int abm_level; - /** + /** * @regamma_tf: * * Pre-defined transfer function for converting internal FB -> wire diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c index c6ed0d854b01..c87b64e464ed 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c @@ -85,6 +85,18 @@ void amdgpu_dm_init_color_mod(void) setup_x_points_distribution(); } +static inline struct fixed31_32 amdgpu_dm_fixpt_from_s3132(__u64 x) +{ + struct fixed31_32 val; + + /* If negative, convert to 2's complement. */ + if (x & (1ULL << 63)) + x = -(x & ~(1ULL << 63)); + + val.value = x; + return val; +} + #ifdef AMD_PRIVATE_COLOR /* Pre-defined Transfer Functions (TF) * @@ -430,7 +442,7 @@ static void __drm_ctm_to_dc_matrix(const struct drm_color_ctm *ctm, } /* gamut_remap_matrix[i] = ctm[i - floor(i/4)] */ - matrix[i] = dc_fixpt_from_s3132(ctm->matrix[i - (i / 4)]); + matrix[i] = amdgpu_dm_fixpt_from_s3132(ctm->matrix[i - (i / 4)]); } } @@ -452,7 +464,7 @@ static void __drm_ctm_3x4_to_dc_matrix(const struct drm_color_ctm_3x4 *ctm, */ for (i = 0; i < 12; i++) { /* gamut_remap_matrix[i] = ctm[i - floor(i/4)] */ - matrix[i] = dc_fixpt_from_s3132(ctm->matrix[i]); + matrix[i] = amdgpu_dm_fixpt_from_s3132(ctm->matrix[i]); } } @@ -630,8 +642,7 @@ static int __set_input_tf(struct dc_color_caps *caps, struct dc_transfer_func *f static enum dc_transfer_func_predefined amdgpu_tf_to_dc_tf(enum amdgpu_transfer_function tf) { - switch (tf) - { + switch (tf) { default: case AMDGPU_TRANSFER_FUNCTION_DEFAULT: case AMDGPU_TRANSFER_FUNCTION_IDENTITY: @@ -1137,7 +1148,7 @@ amdgpu_dm_plane_set_color_properties(struct drm_plane_state *plane_state, uint32_t shaper_size, lut3d_size, blend_size; int ret; - dc_plane_state->hdr_mult = dc_fixpt_from_s3132(dm_plane_state->hdr_mult); + dc_plane_state->hdr_mult = amdgpu_dm_fixpt_from_s3132(dm_plane_state->hdr_mult); shaper_lut = __extract_blob_lut(dm_plane_state->shaper_lut, &shaper_size); shaper_size = shaper_lut != NULL ? shaper_size : 0; @@ -1225,10 +1236,10 @@ int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc, * plane and CRTC degamma at the same time. Explicitly reject atomic * updates when userspace sets both plane and CRTC degamma properties. */ - if (has_crtc_cm_degamma && ret != -EINVAL){ + if (has_crtc_cm_degamma && ret != -EINVAL) { drm_dbg_kms(crtc->base.crtc->dev, "doesn't support plane and CRTC degamma at the same time\n"); - return -EINVAL; + return -EINVAL; } /* If we are here, it means we don't have plane degamma settings, check diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c index 7545a184e43a..6e715ef3a556 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c @@ -29,7 +29,6 @@ #include "dc.h" #include "amdgpu.h" #include "amdgpu_dm_psr.h" -#include "amdgpu_dm_replay.h" #include "amdgpu_dm_crtc.h" #include "amdgpu_dm_plane.h" #include "amdgpu_dm_trace.h" @@ -124,12 +123,7 @@ static void amdgpu_dm_crtc_vblank_control_worker(struct work_struct *work) * fill_dc_dirty_rects(). */ if (vblank_work->stream && vblank_work->stream->link) { - /* - * Prioritize replay, instead of psr - */ - if (vblank_work->stream->link->replay_settings.replay_feature_enabled) - amdgpu_dm_replay_enable(vblank_work->stream, false); - else if (vblank_work->enable) { + if (vblank_work->enable) { if (vblank_work->stream->link->psr_settings.psr_version < DC_PSR_VERSION_SU_1 && vblank_work->stream->link->psr_settings.psr_allow_active) amdgpu_dm_psr_disable(vblank_work->stream); @@ -138,7 +132,6 @@ static void amdgpu_dm_crtc_vblank_control_worker(struct work_struct *work) #ifdef CONFIG_DRM_AMD_SECURE_DISPLAY !amdgpu_dm_crc_window_is_activated(&vblank_work->acrtc->base) && #endif - vblank_work->stream->link->panel_config.psr.disallow_replay && vblank_work->acrtc->dm_irq_params.allow_psr_entry) { amdgpu_dm_psr_enable(vblank_work->stream); } @@ -312,7 +305,7 @@ dm_crtc_additional_color_mgmt(struct drm_crtc *crtc) { struct amdgpu_device *adev = drm_to_adev(crtc->dev); - if(adev->dm.dc->caps.color.mpc.ogam_ram) + if (adev->dm.dc->caps.color.mpc.ogam_ram) drm_object_attach_property(&crtc->base, adev->mode_info.regamma_tf_property, AMDGPU_TRANSFER_FUNCTION_DEFAULT); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 98b41ec7288e..68a846323912 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -2976,7 +2976,6 @@ static int dmub_trace_mask_set(void *data, u64 val) struct amdgpu_device *adev = data; struct dmub_srv *srv = adev->dm.dc->ctx->dmub_srv->dmub; enum dmub_gpint_command cmd; - enum dmub_status status; u64 mask = 0xffff; u8 shift = 0; u32 res; @@ -3003,13 +3002,7 @@ static int dmub_trace_mask_set(void *data, u64 val) break; } - status = dmub_srv_send_gpint_command(srv, cmd, res, 30); - - if (status == DMUB_STATUS_TIMEOUT) - return -ETIMEDOUT; - else if (status == DMUB_STATUS_INVALID) - return -EINVAL; - else if (status != DMUB_STATUS_OK) + if (!dc_wake_and_execute_gpint(adev->dm.dc->ctx, cmd, res, NULL, DM_DMUB_WAIT_TYPE_WAIT)) return -EIO; usleep_range(100, 1000); @@ -3026,7 +3019,6 @@ static int dmub_trace_mask_show(void *data, u64 *val) enum dmub_gpint_command cmd = DMUB_GPINT__GET_TRACE_BUFFER_MASK_WORD0; struct amdgpu_device *adev = data; struct dmub_srv *srv = adev->dm.dc->ctx->dmub_srv->dmub; - enum dmub_status status; u8 shift = 0; u64 raw = 0; u64 res = 0; @@ -3036,23 +3028,12 @@ static int dmub_trace_mask_show(void *data, u64 *val) return -EINVAL; while (i < 4) { - status = dmub_srv_send_gpint_command(srv, cmd, 0, 30); - - if (status == DMUB_STATUS_OK) { - status = dmub_srv_get_gpint_response(srv, (u32 *) &raw); - - if (status == DMUB_STATUS_INVALID) - return -EINVAL; - else if (status != DMUB_STATUS_OK) - return -EIO; - } else if (status == DMUB_STATUS_TIMEOUT) { - return -ETIMEDOUT; - } else if (status == DMUB_STATUS_INVALID) { - return -EINVAL; - } else { + uint32_t response; + + if (!dc_wake_and_execute_gpint(adev->dm.dc->ctx, cmd, 0, &response, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) return -EIO; - } + raw = response; usleep_range(100, 1000); cmd++; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index eaf8d9f48244..85b7f58a7f35 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -979,6 +979,11 @@ int dm_helper_dmub_aux_transfer_sync( struct aux_payload *payload, enum aux_return_code_type *operation_result) { + if (!link->hpd_status) { + *operation_result = AUX_RET_ERROR_HPD_DISCON; + return -1; + } + return amdgpu_dm_process_dmub_aux_transfer_sync(ctx, link->link_index, payload, operation_result); } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c index 08ce3bb8f640..1f08c6564c3b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c @@ -51,6 +51,9 @@ static bool link_supports_psrsu(struct dc_link *link) !link->dpcd_caps.psr_info.psr2_su_y_granularity_cap) return false; + if (amdgpu_dc_debug_mask & DC_DISABLE_PSR_SU) + return false; + return dc_dmub_check_min_version(dc->ctx->dmub_srv->dmub); } diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile index 390e7a99be54..7991ae468f75 100644 --- a/drivers/gpu/drm/amd/display/dc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/Makefile @@ -34,8 +34,6 @@ DC_LIBS += dcn21 DC_LIBS += dcn201 DC_LIBS += dcn30 DC_LIBS += dcn301 -DC_LIBS += dcn302 -DC_LIBS += dcn303 DC_LIBS += dcn31 DC_LIBS += dcn314 DC_LIBS += dcn32 @@ -62,7 +60,7 @@ AMD_DC = $(addsuffix /Makefile, $(addprefix $(FULL_AMD_DISPLAY_PATH)/dc/,$(DC_LI include $(AMD_DC) DISPLAY_CORE = dc.o dc_stat.o dc_resource.o dc_hw_sequencer.o dc_sink.o \ -dc_surface.o dc_debug.o dc_stream.o dc_link_enc_cfg.o dc_link_exports.o +dc_surface.o dc_debug.o dc_stream.o dc_link_enc_cfg.o dc_link_exports.o dc_state.o DISPLAY_CORE += dc_vm_helper.o diff --git a/drivers/gpu/drm/amd/display/dc/basics/conversion.c b/drivers/gpu/drm/amd/display/dc/basics/conversion.c index e295a839ab47..1090d235086a 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/conversion.c +++ b/drivers/gpu/drm/amd/display/dc/basics/conversion.c @@ -103,7 +103,8 @@ void convert_float_matrix( static uint32_t find_gcd(uint32_t a, uint32_t b) { - uint32_t remainder = 0; + uint32_t remainder; + while (b != 0) { remainder = a % b; a = b; diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index fcd65a2057ad..960c4b4f6ddf 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -1014,13 +1014,20 @@ static enum bp_result get_ss_info_v4_5( DC_LOG_BIOS("AS_SIGNAL_TYPE_HDMI ss_percentage: %d\n", ss_info->spread_spectrum_percentage); break; case AS_SIGNAL_TYPE_DISPLAY_PORT: - ss_info->spread_spectrum_percentage = + if (bp->base.integrated_info) { + DC_LOG_BIOS("gpuclk_ss_percentage (unit of 0.001 percent): %d\n", bp->base.integrated_info->gpuclk_ss_percentage); + ss_info->spread_spectrum_percentage = + bp->base.integrated_info->gpuclk_ss_percentage; + ss_info->type.CENTER_MODE = + bp->base.integrated_info->gpuclk_ss_type; + } else { + ss_info->spread_spectrum_percentage = disp_cntl_tbl->dp_ss_percentage; - ss_info->spread_spectrum_range = + ss_info->spread_spectrum_range = disp_cntl_tbl->dp_ss_rate_10hz * 10; - if (disp_cntl_tbl->dp_ss_mode & ATOM_SS_CENTRE_SPREAD_MODE) - ss_info->type.CENTER_MODE = true; - + if (disp_cntl_tbl->dp_ss_mode & ATOM_SS_CENTRE_SPREAD_MODE) + ss_info->type.CENTER_MODE = true; + } DC_LOG_BIOS("AS_SIGNAL_TYPE_DISPLAY_PORT ss_percentage: %d\n", ss_info->spread_spectrum_percentage); break; case AS_SIGNAL_TYPE_GPU_PLL: @@ -2813,6 +2820,8 @@ static enum bp_result get_integrated_info_v2_2( info->ma_channel_number = info_v2_2->umachannelnumber; info->dp_ss_control = le16_to_cpu(info_v2_2->reserved1); + info->gpuclk_ss_percentage = info_v2_2->gpuclk_ss_percentage; + info->gpuclk_ss_type = info_v2_2->gpuclk_ss_type; for (i = 0; i < NUMBER_OF_UCHAR_FOR_GUID; ++i) { info->ext_disp_conn_info.gu_id[i] = diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index ab0adabf9dd4..293a919d605d 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -123,7 +123,7 @@ static void encoder_control_dmcub( sizeof(cmd.digx_encoder_control.header); cmd.digx_encoder_control.encoder_control.dig.stream_param = *dig; - dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result encoder_control_digx_v1_5( @@ -259,7 +259,7 @@ static void transmitter_control_dmcub( sizeof(cmd.dig1_transmitter_control.header); cmd.dig1_transmitter_control.transmitter_control.dig = *dig; - dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result transmitter_control_v1_6( @@ -321,7 +321,7 @@ static void transmitter_control_dmcub_v1_7( sizeof(cmd.dig1_transmitter_control.header); cmd.dig1_transmitter_control.transmitter_control.dig_v1_7 = *dig; - dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result transmitter_control_v1_7( @@ -429,7 +429,7 @@ static void set_pixel_clock_dmcub( sizeof(cmd.set_pixel_clock.header); cmd.set_pixel_clock.pixel_clock.clk = *clk; - dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result set_pixel_clock_v7( @@ -796,7 +796,7 @@ static void enable_disp_power_gating_dmcub( sizeof(cmd.enable_disp_power_gating.header); cmd.enable_disp_power_gating.power_gating.pwr = *pwr; - dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result enable_disp_power_gating_v2_1( @@ -1006,7 +1006,7 @@ static void enable_lvtma_control_dmcub( pwrseq_instance; cmd.lvtma_control.data.bypass_panel_control_wait = bypass_panel_control_wait; - dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result enable_lvtma_control( diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c index 7dacb0f82d29..28a2a837d2f0 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c @@ -29,6 +29,7 @@ #include "dc_types.h" #include "dccg.h" #include "clk_mgr_internal.h" +#include "dc_state_priv.h" #include "link.h" #include "dce100/dce_clk_mgr.h" @@ -63,7 +64,7 @@ int clk_mgr_helper_get_active_display_cnt( /* Don't count SubVP phantom pipes as part of active * display count */ - if (stream->mall_stream_config.type == SUBVP_PHANTOM) + if (dc_state_get_stream_subvp_type(context, stream) == SUBVP_PHANTOM) continue; /* diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c index 3db4ef564b99..ce1386e22576 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c @@ -253,7 +253,7 @@ void dcn31_update_clocks(struct clk_mgr *clk_mgr_base, cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz; cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c index 7326b7565846..a84f1e376dee 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c @@ -87,6 +87,20 @@ static const struct IP_BASE CLK_BASE = { { { { 0x00016C00, 0x02401800, 0, 0, 0, #define CLK1_CLK_PLL_REQ__PllSpineDiv_MASK 0x0000F000L #define CLK1_CLK_PLL_REQ__FbMult_frac_MASK 0xFFFF0000L +#define regCLK1_CLK2_BYPASS_CNTL 0x029c +#define regCLK1_CLK2_BYPASS_CNTL_BASE_IDX 0 + +#define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_SEL__SHIFT 0x0 +#define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_DIV__SHIFT 0x10 +#define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_SEL_MASK 0x00000007L +#define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_DIV_MASK 0x000F0000L + +#define regCLK6_0_CLK6_spll_field_8 0x464b +#define regCLK6_0_CLK6_spll_field_8_BASE_IDX 0 + +#define CLK6_0_CLK6_spll_field_8__spll_ssc_en__SHIFT 0xd +#define CLK6_0_CLK6_spll_field_8__spll_ssc_en_MASK 0x00002000L + #define REG(reg_name) \ (CLK_BASE.instance[0].segment[reg ## reg_name ## _BASE_IDX] + reg ## reg_name) @@ -131,35 +145,63 @@ static int dcn314_get_active_display_cnt_wa( return display_count; } -static void dcn314_disable_otg_wa(struct clk_mgr *clk_mgr_base, struct dc_state *context, bool disable) +static void dcn314_disable_otg_wa(struct clk_mgr *clk_mgr_base, struct dc_state *context, + bool safe_to_lower, bool disable) { struct dc *dc = clk_mgr_base->ctx->dc; int i; for (i = 0; i < dc->res_pool->pipe_count; ++i) { - struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + struct pipe_ctx *pipe = safe_to_lower + ? &context->res_ctx.pipe_ctx[i] + : &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe->top_pipe || pipe->prev_odm_pipe) continue; if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) { - struct stream_encoder *stream_enc = pipe->stream_res.stream_enc; - if (disable) { - if (stream_enc && stream_enc->funcs->disable_fifo) - pipe->stream_res.stream_enc->funcs->disable_fifo(stream_enc); + if (pipe->stream_res.tg && pipe->stream_res.tg->funcs->immediate_disable_crtc) + pipe->stream_res.tg->funcs->immediate_disable_crtc(pipe->stream_res.tg); - pipe->stream_res.tg->funcs->immediate_disable_crtc(pipe->stream_res.tg); reset_sync_context_for_pipe(dc, context, i); } else { pipe->stream_res.tg->funcs->enable_crtc(pipe->stream_res.tg); - - if (stream_enc && stream_enc->funcs->enable_fifo) - pipe->stream_res.stream_enc->funcs->enable_fifo(stream_enc); } } } } +bool dcn314_is_spll_ssc_enabled(struct clk_mgr *clk_mgr_base) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + uint32_t ssc_enable; + + REG_GET(CLK6_0_CLK6_spll_field_8, spll_ssc_en, &ssc_enable); + + return ssc_enable == 1; +} + +void dcn314_init_clocks(struct clk_mgr *clk_mgr) +{ + struct clk_mgr_internal *clk_mgr_int = TO_CLK_MGR_INTERNAL(clk_mgr); + uint32_t ref_dtbclk = clk_mgr->clks.ref_dtbclk_khz; + + memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); + // Assumption is that boot state always supports pstate + clk_mgr->clks.ref_dtbclk_khz = ref_dtbclk; // restore ref_dtbclk + clk_mgr->clks.p_state_change_support = true; + clk_mgr->clks.prev_p_state_change_support = true; + clk_mgr->clks.pwr_state = DCN_PWR_STATE_UNKNOWN; + clk_mgr->clks.zstate_support = DCN_ZSTATE_SUPPORT_UNKNOWN; + + // to adjust dp_dto reference clock if ssc is enable otherwise to apply dprefclk + if (dcn314_is_spll_ssc_enabled(clk_mgr)) + clk_mgr->dp_dto_source_clock_in_khz = + dce_adjust_dp_ref_freq_for_ss(clk_mgr_int, clk_mgr->dprefclk_khz); + else + clk_mgr->dp_dto_source_clock_in_khz = clk_mgr->dprefclk_khz; +} + void dcn314_update_clocks(struct clk_mgr *clk_mgr_base, struct dc_state *context, bool safe_to_lower) @@ -252,11 +294,11 @@ void dcn314_update_clocks(struct clk_mgr *clk_mgr_base, } if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz)) { - dcn314_disable_otg_wa(clk_mgr_base, context, true); + dcn314_disable_otg_wa(clk_mgr_base, context, safe_to_lower, true); clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz; dcn314_smu_set_dispclk(clk_mgr, clk_mgr_base->clks.dispclk_khz); - dcn314_disable_otg_wa(clk_mgr_base, context, false); + dcn314_disable_otg_wa(clk_mgr_base, context, safe_to_lower, false); update_dispclk = true; } @@ -284,7 +326,7 @@ void dcn314_update_clocks(struct clk_mgr *clk_mgr_base, cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz; cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr) @@ -436,6 +478,11 @@ static DpmClocks314_t dummy_clocks; static struct dcn314_watermarks dummy_wms = { 0 }; +static struct dcn314_ss_info_table ss_info_table = { + .ss_divider = 1000, + .ss_percentage = {0, 0, 375, 375, 375} +}; + static void dcn314_build_watermark_ranges(struct clk_bw_params *bw_params, struct dcn314_watermarks *table) { int i, num_valid_sets; @@ -708,13 +755,31 @@ static struct clk_mgr_funcs dcn314_funcs = { .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, .get_dtb_ref_clk_frequency = dcn31_get_dtb_ref_freq_khz, .update_clocks = dcn314_update_clocks, - .init_clocks = dcn31_init_clocks, + .init_clocks = dcn314_init_clocks, .enable_pme_wa = dcn314_enable_pme_wa, .are_clock_states_equal = dcn314_are_clock_states_equal, .notify_wm_ranges = dcn314_notify_wm_ranges }; extern struct clk_mgr_funcs dcn3_fpga_funcs; +static void dcn314_read_ss_info_from_lut(struct clk_mgr_internal *clk_mgr) +{ + uint32_t clock_source; + //uint32_t ssc_enable; + + REG_GET(CLK1_CLK2_BYPASS_CNTL, CLK2_BYPASS_SEL, &clock_source); + //REG_GET(CLK6_0_CLK6_spll_field_8, spll_ssc_en, &ssc_enable); + + if (dcn314_is_spll_ssc_enabled(&clk_mgr->base) && (clock_source < ARRAY_SIZE(ss_info_table.ss_percentage))) { + clk_mgr->dprefclk_ss_percentage = ss_info_table.ss_percentage[clock_source]; + + if (clk_mgr->dprefclk_ss_percentage != 0) { + clk_mgr->ss_on_dprefclk = true; + clk_mgr->dprefclk_ss_divider = ss_info_table.ss_divider; + } + } +} + void dcn314_clk_mgr_construct( struct dc_context *ctx, struct clk_mgr_dcn314 *clk_mgr, @@ -782,6 +847,7 @@ void dcn314_clk_mgr_construct( clk_mgr->base.base.dprefclk_khz = 600000; clk_mgr->base.base.clks.ref_dtbclk_khz = 600000; dce_clock_read_ss_info(&clk_mgr->base); + dcn314_read_ss_info_from_lut(&clk_mgr->base); /*if bios enabled SS, driver needs to adjust dtb clock, only enable with correct bios*/ clk_mgr->base.base.bw_params = &dcn314_bw_params; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h index 171f84340eb2..002c28e80720 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h @@ -28,6 +28,8 @@ #define __DCN314_CLK_MGR_H__ #include "clk_mgr_internal.h" +#define DCN314_NUM_CLOCK_SOURCES 5 + struct dcn314_watermarks; struct dcn314_smu_watermark_set { @@ -40,9 +42,18 @@ struct clk_mgr_dcn314 { struct dcn314_smu_watermark_set smu_wm_set; }; +struct dcn314_ss_info_table { + uint32_t ss_divider; + uint32_t ss_percentage[DCN314_NUM_CLOCK_SOURCES]; +}; + bool dcn314_are_clock_states_equal(struct dc_clocks *a, struct dc_clocks *b); +bool dcn314_is_spll_ssc_enabled(struct clk_mgr *clk_mgr_base); + +void dcn314_init_clocks(struct clk_mgr *clk_mgr); + void dcn314_update_clocks(struct clk_mgr *clk_mgr_base, struct dc_state *context, bool safe_to_lower); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c index 8776055bbeaa..644da4637320 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c @@ -232,7 +232,7 @@ static void dcn315_update_clocks(struct clk_mgr *clk_mgr_base, cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz; cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static void dcn315_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass, diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c index 09151cc56ce4..12f3e8aa46d8 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c @@ -239,7 +239,7 @@ static void dcn316_update_clocks(struct clk_mgr *clk_mgr_base, cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz; cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static void dcn316_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass, diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c index 95c0b49b531a..aadd07bc68c5 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c @@ -25,7 +25,6 @@ #include "dccg.h" #include "clk_mgr_internal.h" - #include "dcn32/dcn32_clk_mgr_smu_msg.h" #include "dcn20/dcn20_clk_mgr.h" #include "dce100/dce_clk_mgr.h" @@ -34,7 +33,7 @@ #include "core_types.h" #include "dm_helpers.h" #include "link.h" - +#include "dc_state_priv.h" #include "atomfirmware.h" #include "smu13_driver_if.h" @@ -458,13 +457,43 @@ static int dcn32_get_dispclk_from_dentist(struct clk_mgr *clk_mgr_base) return 0; } -static void dcn32_auto_dpm_test_log(struct dc_clocks *new_clocks, struct clk_mgr_internal *clk_mgr) +static bool dcn32_check_native_scaling(struct pipe_ctx *pipe) +{ + bool is_native_scaling = false; + int width = pipe->plane_state->src_rect.width; + int height = pipe->plane_state->src_rect.height; + + if (pipe->stream->timing.h_addressable == width && + pipe->stream->timing.v_addressable == height && + pipe->plane_state->dst_rect.width == width && + pipe->plane_state->dst_rect.height == height) + is_native_scaling = true; + + return is_native_scaling; +} + +static void dcn32_auto_dpm_test_log( + struct dc_clocks *new_clocks, + struct clk_mgr_internal *clk_mgr, + struct dc_state *context) { unsigned int dispclk_khz_reg, dppclk_khz_reg, dprefclk_khz_reg, dcfclk_khz_reg, dtbclk_khz_reg, - fclk_khz_reg; + fclk_khz_reg, mall_ss_size_bytes; int dramclk_khz_override, fclk_khz_override, num_fclk_levels; - msleep(5); + struct pipe_ctx *pipe_ctx_list[MAX_PIPES]; + int active_pipe_count = 0; + + for (int i = 0; i < MAX_PIPES; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (pipe_ctx->stream && dc_state_get_pipe_subvp_type(context, pipe_ctx) != SUBVP_PHANTOM) { + pipe_ctx_list[active_pipe_count] = pipe_ctx; + active_pipe_count++; + } + } + + mall_ss_size_bytes = context->bw_ctx.bw.dcn.mall_ss_size_bytes; dispclk_khz_reg = REG_READ(CLK1_CLK0_CURRENT_CNT); // DISPCLK dppclk_khz_reg = REG_READ(CLK1_CLK1_CURRENT_CNT); // DPPCLK @@ -494,16 +523,49 @@ static void dcn32_auto_dpm_test_log(struct dc_clocks *new_clocks, struct clk_mgr // // AutoDPMTest: clk1:%d - clk2:%d - clk3:%d - clk4:%d\n" //////////////////////////////////////////////////////////////////////////// - if (new_clocks && + if (new_clocks && active_pipe_count > 0 && new_clocks->dramclk_khz > 0 && new_clocks->fclk_khz > 0 && new_clocks->dcfclk_khz > 0 && new_clocks->dppclk_khz > 0) { + uint32_t pix_clk_list[MAX_PIPES] = {0}; + int p_state_list[MAX_PIPES] = {0}; + int disp_src_width_list[MAX_PIPES] = {0}; + int disp_src_height_list[MAX_PIPES] = {0}; + uint64_t disp_src_refresh_list[MAX_PIPES] = {0}; + bool is_scaled_list[MAX_PIPES] = {0}; + + for (int i = 0; i < active_pipe_count; i++) { + struct pipe_ctx *curr_pipe_ctx = pipe_ctx_list[i]; + uint64_t refresh_rate; + + pix_clk_list[i] = curr_pipe_ctx->stream->timing.pix_clk_100hz; + p_state_list[i] = curr_pipe_ctx->p_state_type; + + refresh_rate = (curr_pipe_ctx->stream->timing.pix_clk_100hz * (uint64_t)100 + + curr_pipe_ctx->stream->timing.v_total * curr_pipe_ctx->stream->timing.h_total - (uint64_t)1); + refresh_rate = div_u64(refresh_rate, curr_pipe_ctx->stream->timing.v_total); + refresh_rate = div_u64(refresh_rate, curr_pipe_ctx->stream->timing.h_total); + disp_src_refresh_list[i] = refresh_rate; + + if (curr_pipe_ctx->plane_state) { + is_scaled_list[i] = !(dcn32_check_native_scaling(curr_pipe_ctx)); + disp_src_width_list[i] = curr_pipe_ctx->plane_state->src_rect.width; + disp_src_height_list[i] = curr_pipe_ctx->plane_state->src_rect.height; + } + } + DC_LOG_AUTO_DPM_TEST("AutoDPMTest: dramclk:%d - fclk:%d - " "dcfclk:%d - dppclk:%d - dispclk_hw:%d - " "dppclk_hw:%d - dprefclk_hw:%d - dcfclk_hw:%d - " - "dtbclk_hw:%d - fclk_hw:%d\n", + "dtbclk_hw:%d - fclk_hw:%d - pix_clk_0:%d - pix_clk_1:%d - " + "pix_clk_2:%d - pix_clk_3:%d - mall_ss_size:%d - p_state_type_0:%d - " + "p_state_type_1:%d - p_state_type_2:%d - p_state_type_3:%d - " + "pix_width_0:%d - pix_height_0:%d - refresh_rate_0:%lld - is_scaled_0:%d - " + "pix_width_1:%d - pix_height_1:%d - refresh_rate_1:%lld - is_scaled_1:%d - " + "pix_width_2:%d - pix_height_2:%d - refresh_rate_2:%lld - is_scaled_2:%d - " + "pix_width_3:%d - pix_height_3:%d - refresh_rate_3:%lld - is_scaled_3:%d - LOG_END\n", dramclk_khz_override, fclk_khz_override, new_clocks->dcfclk_khz, @@ -513,7 +575,14 @@ static void dcn32_auto_dpm_test_log(struct dc_clocks *new_clocks, struct clk_mgr dprefclk_khz_reg, dcfclk_khz_reg, dtbclk_khz_reg, - fclk_khz_reg); + fclk_khz_reg, + pix_clk_list[0], pix_clk_list[1], pix_clk_list[3], pix_clk_list[2], + mall_ss_size_bytes, + p_state_list[0], p_state_list[1], p_state_list[2], p_state_list[3], + disp_src_width_list[0], disp_src_height_list[0], disp_src_refresh_list[0], is_scaled_list[0], + disp_src_width_list[1], disp_src_height_list[1], disp_src_refresh_list[1], is_scaled_list[1], + disp_src_width_list[2], disp_src_height_list[2], disp_src_refresh_list[2], is_scaled_list[2], + disp_src_width_list[3], disp_src_height_list[3], disp_src_refresh_list[3], is_scaled_list[3]); } } @@ -686,6 +755,7 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, /* DCCG requires KHz precision for DTBCLK */ clk_mgr_base->clks.ref_dtbclk_khz = dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DTBCLK, khz_to_mhz_ceil(new_clocks->ref_dtbclk_khz)); + dcn32_update_clocks_update_dtb_dto(clk_mgr, context, clk_mgr_base->clks.ref_dtbclk_khz); } @@ -713,8 +783,8 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, dmcu->funcs->set_psr_wait_loop(dmcu, clk_mgr_base->clks.dispclk_khz / 1000 / 7); - if (dc->config.enable_auto_dpm_test_logs && safe_to_lower) { - dcn32_auto_dpm_test_log(new_clocks, clk_mgr); + if (dc->config.enable_auto_dpm_test_logs) { + dcn32_auto_dpm_test_log(new_clocks, clk_mgr, context); } } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c index 8d4c0b209872..9c660d1facc7 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c @@ -50,6 +50,7 @@ #include "dc_dmub_srv.h" #include "link.h" #include "logger_types.h" + #undef DC_LOGGER #define DC_LOGGER \ clk_mgr->base.base.ctx->logger @@ -342,7 +343,7 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base, cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz; cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr) @@ -417,9 +418,8 @@ bool dcn35_are_clock_states_equal(struct dc_clocks *a, } static void dcn35_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass, - struct clk_mgr *clk_mgr_base, struct clk_log_info *log_info) + struct clk_mgr_dcn35 *clk_mgr) { - } static struct clk_bw_params dcn35_bw_params = { @@ -986,7 +986,6 @@ void dcn35_clk_mgr_construct( struct dccg *dccg) { struct dcn35_smu_dpm_clks smu_dpm_clks = { 0 }; - struct clk_log_info log_info = {0}; clk_mgr->base.base.ctx = ctx; clk_mgr->base.base.funcs = &dcn35_funcs; @@ -1039,7 +1038,7 @@ void dcn35_clk_mgr_construct( dcn35_bw_params.wm_table = ddr5_wm_table; } /* Saved clocks configured at boot for debug purposes */ - dcn35_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, &clk_mgr->base.base, &log_info); + dcn35_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, clk_mgr); clk_mgr->base.base.dprefclk_khz = dcn35_smu_get_dprefclk(&clk_mgr->base); clk_mgr->base.base.clks.ref_dtbclk_khz = 600000; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 9d3925603979..aa7c02ba948e 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -34,6 +34,8 @@ #include "dce/dce_hwseq.h" #include "resource.h" +#include "dc_state.h" +#include "dc_state_priv.h" #include "gpio_service_interface.h" #include "clk_mgr.h" @@ -519,7 +521,7 @@ dc_stream_forward_dmub_crc_window(struct dc_dmub_srv *dmub_srv, cmd.secure_display.roi_info.y_end = rect->y + rect->height; } - dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); + dc_wake_and_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); } static inline void @@ -808,7 +810,7 @@ static void dc_destruct(struct dc *dc) link_enc_cfg_init(dc, dc->current_state); if (dc->current_state) { - dc_release_state(dc->current_state); + dc_state_release(dc->current_state); dc->current_state = NULL; } @@ -1020,29 +1022,27 @@ static bool dc_construct(struct dc *dc, } #endif + if (!create_links(dc, init_params->num_virtual_links)) + goto fail; + + /* Create additional DIG link encoder objects if fewer than the platform + * supports were created during link construction. + */ + if (!create_link_encoders(dc)) + goto fail; + /* Creation of current_state must occur after dc->dml * is initialized in dc_create_resource_pool because * on creation it copies the contents of dc->dml */ - dc->current_state = dc_create_state(dc); + dc->current_state = dc_state_create(dc); if (!dc->current_state) { dm_error("%s: failed to create validate ctx\n", __func__); goto fail; } - if (!create_links(dc, init_params->num_virtual_links)) - goto fail; - - /* Create additional DIG link encoder objects if fewer than the platform - * supports were created during link construction. - */ - if (!create_link_encoders(dc)) - goto fail; - - dc_resource_state_construct(dc, dc->current_state); - return true; fail: @@ -1085,7 +1085,7 @@ static void apply_ctx_interdependent_lock(struct dc *dc, } } -static void dc_update_viusal_confirm_color(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx) +static void dc_update_visual_confirm_color(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx) { if (dc->ctx->dce_version >= DCN_VERSION_1_0) { memset(&pipe_ctx->visual_confirm_color, 0, sizeof(struct tg_color)); @@ -1105,9 +1105,9 @@ static void dc_update_viusal_confirm_color(struct dc *dc, struct dc_state *conte if (dc->debug.visual_confirm == VISUAL_CONFIRM_MPCTREE) get_mpctree_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SUBVP) - get_subvp_visual_confirm_color(dc, context, pipe_ctx, &(pipe_ctx->visual_confirm_color)); + get_subvp_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); else if (dc->debug.visual_confirm == VISUAL_CONFIRM_MCLK_SWITCH) - get_mclk_switch_visual_confirm_color(dc, context, pipe_ctx, &(pipe_ctx->visual_confirm_color)); + get_mclk_switch_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); } } } @@ -1115,7 +1115,7 @@ static void dc_update_viusal_confirm_color(struct dc *dc, struct dc_state *conte static void disable_dangling_plane(struct dc *dc, struct dc_state *context) { int i, j; - struct dc_state *dangling_context = dc_create_state(dc); + struct dc_state *dangling_context = dc_state_create_current_copy(dc); struct dc_state *current_ctx; struct pipe_ctx *pipe; struct timing_generator *tg; @@ -1123,8 +1123,6 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) if (dangling_context == NULL) return; - dc_resource_state_copy_construct(dc->current_state, dangling_context); - for (i = 0; i < dc->res_pool->pipe_count; i++) { struct dc_stream_state *old_stream = dc->current_state->res_ctx.pipe_ctx[i].stream; @@ -1161,6 +1159,7 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) } if (should_disable && old_stream) { + bool is_phantom = dc_state_get_stream_subvp_type(dc->current_state, old_stream) == SUBVP_PHANTOM; pipe = &dc->current_state->res_ctx.pipe_ctx[i]; tg = pipe->stream_res.tg; /* When disabling plane for a phantom pipe, we must turn on the @@ -1169,22 +1168,29 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) * state that can result in underflow or hang when enabling it * again for different use. */ - if (old_stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (is_phantom) { if (tg->funcs->enable_crtc) { int main_pipe_width, main_pipe_height; + struct dc_stream_state *old_paired_stream = dc_state_get_paired_subvp_stream(dc->current_state, old_stream); - main_pipe_width = old_stream->mall_stream_config.paired_stream->dst.width; - main_pipe_height = old_stream->mall_stream_config.paired_stream->dst.height; + main_pipe_width = old_paired_stream->dst.width; + main_pipe_height = old_paired_stream->dst.height; if (dc->hwss.blank_phantom) dc->hwss.blank_phantom(dc, tg, main_pipe_width, main_pipe_height); tg->funcs->enable_crtc(tg); } } - dc_rem_all_planes_for_stream(dc, old_stream, dangling_context); + + if (is_phantom) + dc_state_rem_all_phantom_planes_for_stream(dc, old_stream, dangling_context, true); + else + dc_state_rem_all_planes_for_stream(dc, old_stream, dangling_context); disable_all_writeback_pipes_for_stream(dc, old_stream, dangling_context); - if (pipe->stream && pipe->plane_state) - dc_update_viusal_confirm_color(dc, context, pipe); + if (pipe->stream && pipe->plane_state) { + set_p_state_switch_method(dc, context, pipe); + dc_update_visual_confirm_color(dc, context, pipe); + } if (dc->hwss.apply_ctx_for_surface) { apply_ctx_interdependent_lock(dc, dc->current_state, old_stream, true); @@ -1203,7 +1209,7 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) * The OTG is set to disable on falling edge of VUPDATE so the plane disable * will still get it's double buffer update. */ - if (old_stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (is_phantom) { if (tg->funcs->disable_phantom_crtc) tg->funcs->disable_phantom_crtc(tg); } @@ -1212,7 +1218,7 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) current_ctx = dc->current_state; dc->current_state = dangling_context; - dc_release_state(current_ctx); + dc_state_release(current_ctx); } static void disable_vbios_mode_if_required( @@ -1284,7 +1290,7 @@ static void wait_for_no_pipes_pending(struct dc *dc, struct dc_state *context) int count = 0; struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (!pipe->plane_state || pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) + if (!pipe->plane_state || dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) continue; /* Timeout 100 ms */ @@ -1510,7 +1516,7 @@ static void program_timing_sync( } for (k = 0; k < group_size; k++) { - struct dc_stream_status *status = dc_stream_get_status_from_state(ctx, pipe_set[k]->stream); + struct dc_stream_status *status = dc_state_get_stream_status(ctx, pipe_set[k]->stream); status->timing_sync_info.group_id = num_group; status->timing_sync_info.group_size = group_size; @@ -1555,7 +1561,7 @@ static void program_timing_sync( if (group_size > 1) { if (sync_type == TIMING_SYNCHRONIZABLE) { dc->hwss.enable_timing_synchronization( - dc, group_index, group_size, pipe_set); + dc, ctx, group_index, group_size, pipe_set); } else if (sync_type == VBLANK_SYNCHRONIZABLE) { dc->hwss.enable_vblanks_synchronization( @@ -1837,7 +1843,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i]; /* Check old context for SubVP */ - subvp_prev_use |= (old_pipe->stream && old_pipe->stream->mall_stream_config.type == SUBVP_PHANTOM); + subvp_prev_use |= (dc_state_get_pipe_subvp_type(dc->current_state, old_pipe) == SUBVP_PHANTOM); if (subvp_prev_use) break; } @@ -1995,9 +2001,9 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c old_state = dc->current_state; dc->current_state = context; - dc_release_state(old_state); + dc_state_release(old_state); - dc_retain_state(dc->current_state); + dc_state_retain(dc->current_state); return result; } @@ -2068,12 +2074,10 @@ enum dc_status dc_commit_streams(struct dc *dc, if (handle_exit_odm2to1) res = commit_minimal_transition_state(dc, dc->current_state); - context = dc_create_state(dc); + context = dc_state_create_current_copy(dc); if (!context) goto context_alloc_fail; - dc_resource_state_copy_construct_current(dc, context); - res = dc_validate_with_context(dc, set, stream_count, context, false); if (res != DC_OK) { BREAK_TO_DEBUGGER(); @@ -2088,7 +2092,7 @@ enum dc_status dc_commit_streams(struct dc *dc, streams[i]->out.otg_offset = context->stream_status[j].primary_otg_inst; if (dc_is_embedded_signal(streams[i]->signal)) { - struct dc_stream_status *status = dc_stream_get_status_from_state(context, streams[i]); + struct dc_stream_status *status = dc_state_get_stream_status(context, streams[i]); if (dc->hwss.is_abm_supported) status->is_abm_supported = dc->hwss.is_abm_supported(dc, context, streams[i]); @@ -2099,7 +2103,7 @@ enum dc_status dc_commit_streams(struct dc *dc, } fail: - dc_release_state(context); + dc_state_release(context); context_alloc_fail: @@ -2153,7 +2157,7 @@ static bool is_flip_pending_in_pipes(struct dc *dc, struct dc_state *context) pipe = &context->res_ctx.pipe_ctx[i]; // Don't check flip pending on phantom pipes - if (!pipe->plane_state || (pipe->stream && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM)) + if (!pipe->plane_state || (dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM)) continue; /* Must set to false to start with, due to OR in update function */ @@ -2211,7 +2215,7 @@ void dc_post_update_surfaces_to_stream(struct dc *dc) if (context->res_ctx.pipe_ctx[i].stream == NULL || context->res_ctx.pipe_ctx[i].plane_state == NULL) { context->res_ctx.pipe_ctx[i].pipe_idx = i; - dc->hwss.disable_plane(dc, &context->res_ctx.pipe_ctx[i]); + dc->hwss.disable_plane(dc, context, &context->res_ctx.pipe_ctx[i]); } process_deferred_updates(dc); @@ -2226,103 +2230,6 @@ void dc_post_update_surfaces_to_stream(struct dc *dc) dc->wm_optimized_required = false; } -static void init_state(struct dc *dc, struct dc_state *context) -{ - /* Each context must have their own instance of VBA and in order to - * initialize and obtain IP and SOC the base DML instance from DC is - * initially copied into every context - */ - memcpy(&context->bw_ctx.dml, &dc->dml, sizeof(struct display_mode_lib)); -} - -struct dc_state *dc_create_state(struct dc *dc) -{ - struct dc_state *context = kvzalloc(sizeof(struct dc_state), - GFP_KERNEL); - - if (!context) - return NULL; - - init_state(dc, context); - -#ifdef CONFIG_DRM_AMD_DC_FP - if (dc->debug.using_dml2) { - dml2_create(dc, &dc->dml2_options, &context->bw_ctx.dml2); - } -#endif - kref_init(&context->refcount); - - return context; -} - -struct dc_state *dc_copy_state(struct dc_state *src_ctx) -{ - int i, j; - struct dc_state *new_ctx = kvmalloc(sizeof(struct dc_state), GFP_KERNEL); - - if (!new_ctx) - return NULL; - memcpy(new_ctx, src_ctx, sizeof(struct dc_state)); - -#ifdef CONFIG_DRM_AMD_DC_FP - if (new_ctx->bw_ctx.dml2 && !dml2_create_copy(&new_ctx->bw_ctx.dml2, src_ctx->bw_ctx.dml2)) { - dc_release_state(new_ctx); - return NULL; - } -#endif - - for (i = 0; i < MAX_PIPES; i++) { - struct pipe_ctx *cur_pipe = &new_ctx->res_ctx.pipe_ctx[i]; - - if (cur_pipe->top_pipe) - cur_pipe->top_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->top_pipe->pipe_idx]; - - if (cur_pipe->bottom_pipe) - cur_pipe->bottom_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->bottom_pipe->pipe_idx]; - - if (cur_pipe->prev_odm_pipe) - cur_pipe->prev_odm_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->prev_odm_pipe->pipe_idx]; - - if (cur_pipe->next_odm_pipe) - cur_pipe->next_odm_pipe = &new_ctx->res_ctx.pipe_ctx[cur_pipe->next_odm_pipe->pipe_idx]; - - } - - for (i = 0; i < new_ctx->stream_count; i++) { - dc_stream_retain(new_ctx->streams[i]); - for (j = 0; j < new_ctx->stream_status[i].plane_count; j++) - dc_plane_state_retain( - new_ctx->stream_status[i].plane_states[j]); - } - - kref_init(&new_ctx->refcount); - - return new_ctx; -} - -void dc_retain_state(struct dc_state *context) -{ - kref_get(&context->refcount); -} - -static void dc_state_free(struct kref *kref) -{ - struct dc_state *context = container_of(kref, struct dc_state, refcount); - dc_resource_state_destruct(context); - -#ifdef CONFIG_DRM_AMD_DC_FP - dml2_destroy(context->bw_ctx.dml2); - context->bw_ctx.dml2 = 0; -#endif - - kvfree(context); -} - -void dc_release_state(struct dc_state *context) -{ - kref_put(&context->refcount, dc_state_free); -} - bool dc_set_generic_gpio_for_stereo(bool enable, struct gpio_service *gpio_service) { @@ -2995,11 +2902,9 @@ static void copy_stream_update_to_stream(struct dc *dc, update->dsc_config->num_slices_v != 0); /* Use temporarry context for validating new DSC config */ - struct dc_state *dsc_validate_context = dc_create_state(dc); + struct dc_state *dsc_validate_context = dc_state_create_copy(dc->current_state); if (dsc_validate_context) { - dc_resource_state_copy_construct(dc->current_state, dsc_validate_context); - stream->timing.dsc_cfg = *update->dsc_config; stream->timing.flags.DSC = enable_dsc; if (!dc->res_pool->funcs->validate_bandwidth(dc, dsc_validate_context, true)) { @@ -3008,7 +2913,7 @@ static void copy_stream_update_to_stream(struct dc *dc, update->dsc_config = NULL; } - dc_release_state(dsc_validate_context); + dc_state_release(dsc_validate_context); } else { DC_ERROR("Failed to allocate new validate context for DSC change\n"); update->dsc_config = NULL; @@ -3107,30 +3012,27 @@ static bool update_planes_and_stream_state(struct dc *dc, new_planes[i] = srf_updates[i].surface; /* initialize scratch memory for building context */ - context = dc_create_state(dc); + context = dc_state_create_copy(dc->current_state); if (context == NULL) { DC_ERROR("Failed to allocate new validate context!\n"); return false; } - dc_resource_state_copy_construct( - dc->current_state, context); - /* For each full update, remove all existing phantom pipes first. * Ensures that we have enough pipes for newly added MPO planes */ - if (dc->res_pool->funcs->remove_phantom_pipes) - dc->res_pool->funcs->remove_phantom_pipes(dc, context, false); + dc_state_remove_phantom_streams_and_planes(dc, context); + dc_state_release_phantom_streams_and_planes(dc, context); /*remove old surfaces from context */ - if (!dc_rem_all_planes_for_stream(dc, stream, context)) { + if (!dc_state_rem_all_planes_for_stream(dc, stream, context)) { BREAK_TO_DEBUGGER(); goto fail; } /* add surface to context */ - if (!dc_add_all_planes_for_stream(dc, stream, new_planes, surface_count, context)) { + if (!dc_state_add_all_planes_for_stream(dc, stream, new_planes, surface_count, context)) { BREAK_TO_DEBUGGER(); goto fail; @@ -3155,19 +3057,6 @@ static bool update_planes_and_stream_state(struct dc *dc, if (update_type == UPDATE_TYPE_FULL) { if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false)) { - /* For phantom pipes we remove and create a new set of phantom pipes - * for each full update (because we don't know if we'll need phantom - * pipes until after the first round of validation). However, if validation - * fails we need to keep the existing phantom pipes (because we don't update - * the dc->current_state). - * - * The phantom stream/plane refcount is decremented for validation because - * we assume it'll be removed (the free comes when the dc_state is freed), - * but if validation fails we have to increment back the refcount so it's - * consistent. - */ - if (dc->res_pool->funcs->retain_phantom_pipes) - dc->res_pool->funcs->retain_phantom_pipes(dc, dc->current_state); BREAK_TO_DEBUGGER(); goto fail; } @@ -3188,7 +3077,7 @@ static bool update_planes_and_stream_state(struct dc *dc, return true; fail: - dc_release_state(context); + dc_state_release(context); return false; @@ -3384,7 +3273,7 @@ void dc_dmub_update_dirty_rect(struct dc *dc, update_dirty_rect->panel_inst = panel_inst; update_dirty_rect->pipe_idx = j; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); } } } @@ -3486,18 +3375,24 @@ static void commit_planes_for_stream_fast(struct dc *dc, { int i, j; struct pipe_ctx *top_pipe_to_program = NULL; + struct dc_stream_status *stream_status = NULL; dc_z10_restore(dc); top_pipe_to_program = resource_get_otg_master_for_stream( &context->res_ctx, stream); - if (dc->debug.visual_confirm) { - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + if (!top_pipe_to_program) + return; - if (pipe->stream && pipe->plane_state) - dc_update_viusal_confirm_color(dc, context, pipe); + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->stream && pipe->plane_state) { + set_p_state_switch_method(dc, context, pipe); + + if (dc->debug.visual_confirm) + dc_update_visual_confirm_color(dc, context, pipe); } } @@ -3521,6 +3416,8 @@ static void commit_planes_for_stream_fast(struct dc *dc, } } + stream_status = dc_state_get_stream_status(context, stream); + build_dmub_cmd_list(dc, srf_updates, surface_count, @@ -3533,7 +3430,8 @@ static void commit_planes_for_stream_fast(struct dc *dc, context->dmub_cmd_count, context->block_sequence, &(context->block_sequence_steps), - top_pipe_to_program); + top_pipe_to_program, + stream_status); hwss_execute_sequence(dc, context->block_sequence, context->block_sequence_steps); @@ -3624,12 +3522,12 @@ static void commit_planes_for_stream(struct dc *dc, top_pipe_to_program = resource_get_otg_master_for_stream( &context->res_ctx, stream); - + ASSERT(top_pipe_to_program != NULL); for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i]; // Check old context for SubVP - subvp_prev_use |= (old_pipe->stream && old_pipe->stream->mall_stream_config.type == SUBVP_PHANTOM); + subvp_prev_use |= (dc_state_get_pipe_subvp_type(dc->current_state, old_pipe) == SUBVP_PHANTOM); if (subvp_prev_use) break; } @@ -3637,19 +3535,22 @@ static void commit_planes_for_stream(struct dc *dc, for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (pipe->stream && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) { subvp_curr_use = true; break; } } - if (dc->debug.visual_confirm) - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->stream && pipe->plane_state) { + set_p_state_switch_method(dc, context, pipe); - if (pipe->stream && pipe->plane_state) - dc_update_viusal_confirm_color(dc, context, pipe); + if (dc->debug.visual_confirm) + dc_update_visual_confirm_color(dc, context, pipe); } + } if (stream->test_pattern.type != DP_TEST_PATTERN_VIDEO_MODE) { struct pipe_ctx *mpcc_pipe; @@ -4022,7 +3923,7 @@ static bool could_mpcc_tree_change_for_active_pipes(struct dc *dc, for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i]; - if (pipe->stream && pipe->stream->mall_stream_config.type != SUBVP_NONE) { + if (dc_state_get_pipe_subvp_type(dc->current_state, pipe) != SUBVP_NONE) { subvp_active = true; break; } @@ -4059,7 +3960,7 @@ struct pipe_split_policy_backup { static void release_minimal_transition_state(struct dc *dc, struct dc_state *context, struct pipe_split_policy_backup *policy) { - dc_release_state(context); + dc_state_release(context); /* restore previous pipe split and odm policy */ if (!dc->config.is_vmin_only_asic) dc->debug.pipe_split_policy = policy->mpc_policy; @@ -4070,7 +3971,7 @@ static void release_minimal_transition_state(struct dc *dc, static struct dc_state *create_minimal_transition_state(struct dc *dc, struct dc_state *base_context, struct pipe_split_policy_backup *policy) { - struct dc_state *minimal_transition_context = dc_create_state(dc); + struct dc_state *minimal_transition_context = NULL; unsigned int i, j; if (!dc->config.is_vmin_only_asic) { @@ -4082,7 +3983,9 @@ static struct dc_state *create_minimal_transition_state(struct dc *dc, policy->subvp_policy = dc->debug.force_disable_subvp; dc->debug.force_disable_subvp = true; - dc_resource_state_copy_construct(base_context, minimal_transition_context); + minimal_transition_context = dc_state_create_copy(base_context); + if (!minimal_transition_context) + return NULL; /* commit minimal state */ if (dc->res_pool->funcs->validate_bandwidth(dc, minimal_transition_context, false)) { @@ -4114,7 +4017,6 @@ static bool commit_minimal_transition_state_for_windowed_mpo_odm(struct dc *dc, bool success = false; struct dc_state *minimal_transition_context; struct pipe_split_policy_backup policy; - struct mall_temp_config mall_temp_config; /* commit based on new context */ /* Since all phantom pipes are removed in full validation, @@ -4123,8 +4025,6 @@ static bool commit_minimal_transition_state_for_windowed_mpo_odm(struct dc *dc, * pipe as subvp/phantom will be cleared (dc copy constructor * creates a shallow copy). */ - if (dc->res_pool->funcs->save_mall_state) - dc->res_pool->funcs->save_mall_state(dc, context, &mall_temp_config); minimal_transition_context = create_minimal_transition_state(dc, context, &policy); if (minimal_transition_context) { @@ -4137,16 +4037,6 @@ static bool commit_minimal_transition_state_for_windowed_mpo_odm(struct dc *dc, success = dc_commit_state_no_check(dc, minimal_transition_context) == DC_OK; } release_minimal_transition_state(dc, minimal_transition_context, &policy); - if (dc->res_pool->funcs->restore_mall_state) - dc->res_pool->funcs->restore_mall_state(dc, context, &mall_temp_config); - /* If we do a minimal transition with plane removal and the context - * has subvp we also have to retain back the phantom stream / planes - * since the refcount is decremented as part of the min transition - * (we commit a state with no subvp, so the phantom streams / planes - * had to be removed). - */ - if (dc->res_pool->funcs->retain_phantom_pipes) - dc->res_pool->funcs->retain_phantom_pipes(dc, context); } if (!success) { @@ -4214,7 +4104,7 @@ static bool commit_minimal_transition_state(struct dc *dc, for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i]; - if (pipe->stream && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (pipe->stream && dc_state_get_pipe_subvp_type(dc->current_state, pipe) == SUBVP_PHANTOM) { subvp_in_use = true; break; } @@ -4455,6 +4345,8 @@ static bool should_commit_minimal_transition_for_windowed_mpo_odm(struct dc *dc, cur_pipe = resource_get_otg_master_for_stream(&dc->current_state->res_ctx, stream); new_pipe = resource_get_otg_master_for_stream(&context->res_ctx, stream); + if (!cur_pipe || !new_pipe) + return false; cur_is_odm_in_use = resource_get_odm_slice_count(cur_pipe) > 1; new_is_odm_in_use = resource_get_odm_slice_count(new_pipe) > 1; if (cur_is_odm_in_use == new_is_odm_in_use) @@ -4480,7 +4372,6 @@ bool dc_update_planes_and_stream(struct dc *dc, struct dc_state *context; enum surface_update_type update_type; int i; - struct mall_temp_config mall_temp_config; struct dc_fast_update fast_update[MAX_SURFACES] = {0}; /* In cases where MPO and split or ODM are used transitions can @@ -4524,23 +4415,10 @@ bool dc_update_planes_and_stream(struct dc *dc, * pipe as subvp/phantom will be cleared (dc copy constructor * creates a shallow copy). */ - if (dc->res_pool->funcs->save_mall_state) - dc->res_pool->funcs->save_mall_state(dc, context, &mall_temp_config); if (!commit_minimal_transition_state(dc, context)) { - dc_release_state(context); + dc_state_release(context); return false; } - if (dc->res_pool->funcs->restore_mall_state) - dc->res_pool->funcs->restore_mall_state(dc, context, &mall_temp_config); - - /* If we do a minimal transition with plane removal and the context - * has subvp we also have to retain back the phantom stream / planes - * since the refcount is decremented as part of the min transition - * (we commit a state with no subvp, so the phantom streams / planes - * had to be removed). - */ - if (dc->res_pool->funcs->retain_phantom_pipes) - dc->res_pool->funcs->retain_phantom_pipes(dc, context); update_type = UPDATE_TYPE_FULL; } @@ -4597,7 +4475,7 @@ bool dc_update_planes_and_stream(struct dc *dc, struct dc_state *old = dc->current_state; dc->current_state = context; - dc_release_state(old); + dc_state_release(old); // clear any forced full updates for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -4656,14 +4534,12 @@ void dc_commit_updates_for_stream(struct dc *dc, if (update_type >= UPDATE_TYPE_FULL) { /* initialize scratch memory for building context */ - context = dc_create_state(dc); + context = dc_state_create_copy(state); if (context == NULL) { DC_ERROR("Failed to allocate new validate context!\n"); return; } - dc_resource_state_copy_construct(state, context); - for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i]; struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -4702,7 +4578,7 @@ void dc_commit_updates_for_stream(struct dc *dc, if (update_type >= UPDATE_TYPE_FULL) { if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false)) { DC_ERROR("Mode validation failed for stream update!\n"); - dc_release_state(context); + dc_state_release(context); return; } } @@ -4735,7 +4611,7 @@ void dc_commit_updates_for_stream(struct dc *dc, struct dc_state *old = dc->current_state; dc->current_state = context; - dc_release_state(old); + dc_state_release(old); for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; @@ -4808,7 +4684,7 @@ void dc_set_power_state( switch (power_state) { case DC_ACPI_CM_POWER_STATE_D0: - dc_resource_state_construct(dc, dc->current_state); + dc_state_construct(dc, dc->current_state); dc_z10_restore(dc); @@ -4823,7 +4699,7 @@ void dc_set_power_state( default: ASSERT(dc->current_state->stream_count == 0); - dc_resource_state_destruct(dc->current_state); + dc_state_destruct(dc->current_state); break; } @@ -4900,6 +4776,38 @@ bool dc_set_psr_allow_active(struct dc *dc, bool enable) return true; } +/* enable/disable eDP Replay without specify stream for eDP */ +bool dc_set_replay_allow_active(struct dc *dc, bool active) +{ + int i; + bool allow_active; + + for (i = 0; i < dc->current_state->stream_count; i++) { + struct dc_link *link; + struct dc_stream_state *stream = dc->current_state->streams[i]; + + link = stream->link; + if (!link) + continue; + + if (link->replay_settings.replay_feature_enabled) { + if (active && !link->replay_settings.replay_allow_active) { + allow_active = true; + if (!dc_link_set_replay_allow_active(link, &allow_active, + false, false, NULL)) + return false; + } else if (!active && link->replay_settings.replay_allow_active) { + allow_active = false; + if (!dc_link_set_replay_allow_active(link, &allow_active, + true, false, NULL)) + return false; + } + } + } + + return true; +} + void dc_allow_idle_optimizations(struct dc *dc, bool allow) { if (dc->debug.disable_idle_power_optimizations) @@ -5093,18 +5001,28 @@ void dc_mclk_switch_using_fw_based_vblank_stretch_shut_down(struct dc *dc) */ bool dc_is_dmub_outbox_supported(struct dc *dc) { - /* DCN31 B0 USB4 DPIA needs dmub notifications for interrupts */ - if (dc->ctx->asic_id.chip_family == FAMILY_YELLOW_CARP && - dc->ctx->asic_id.hw_internal_rev == YELLOW_CARP_B0 && - !dc->debug.dpia_debug.bits.disable_dpia) - return true; + switch (dc->ctx->asic_id.chip_family) { - if (dc->ctx->asic_id.chip_family == AMDGPU_FAMILY_GC_11_0_1 && - !dc->debug.dpia_debug.bits.disable_dpia) - return true; + case FAMILY_YELLOW_CARP: + /* DCN31 B0 USB4 DPIA needs dmub notifications for interrupts */ + if (dc->ctx->asic_id.hw_internal_rev == YELLOW_CARP_B0 && + !dc->debug.dpia_debug.bits.disable_dpia) + return true; + break; + + case AMDGPU_FAMILY_GC_11_0_1: + case AMDGPU_FAMILY_GC_11_5_0: + if (!dc->debug.dpia_debug.bits.disable_dpia) + return true; + break; + + default: + break; + } /* dmub aux needs dmub notifications to be enabled */ return dc->debug.enable_dmub_aux_for_legacy_ddc; + } /** @@ -5201,7 +5119,7 @@ bool dc_process_dmub_aux_transfer_async(struct dc *dc, ); } - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -5255,7 +5173,7 @@ bool dc_process_dmub_set_config_async(struct dc *dc, cmd.set_config_access.set_config_control.cmd_pkt.msg_type = payload->msg_type; cmd.set_config_access.set_config_control.cmd_pkt.msg_data = payload->msg_data; - if (!dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) { + if (!dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) { /* command is not processed by dmub */ notify->sc_status = SET_CONFIG_UNKNOWN_ERROR; return is_cmd_complete; @@ -5298,7 +5216,7 @@ enum dc_status dc_process_dmub_set_mst_slots(const struct dc *dc, cmd.set_mst_alloc_slots.mst_slots_control.instance = dc->links[link_index]->ddc_hw_inst; cmd.set_mst_alloc_slots.mst_slots_control.mst_alloc_slots = mst_alloc_slots; - if (!dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + if (!dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) /* command is not processed by dmub */ return DC_ERROR_UNEXPECTED; @@ -5336,7 +5254,7 @@ void dc_process_dmub_dpia_hpd_int_enable(const struct dc *dc, cmd.dpia_hpd_int_enable.header.type = DMUB_CMD__DPIA_HPD_INT_ENABLE; cmd.dpia_hpd_int_enable.enable = hpd_int_enable; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); DC_LOG_DEBUG("%s: hpd_int_enable(%d)\n", __func__, hpd_int_enable); } @@ -5435,6 +5353,8 @@ bool dc_abm_save_restore( struct dc_link *link = stream->sink->link; struct dc_link *edp_links[MAX_NUM_EDP]; + if (link->replay_settings.replay_feature_enabled) + return false; /*find primary pipe associated with stream*/ for (i = 0; i < MAX_PIPES; i++) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c index fe07160932d6..9c05b1a07142 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c @@ -31,6 +31,7 @@ #include "basics/dc_common.h" #include "resource.h" #include "dc_dmub_srv.h" +#include "dc_state_priv.h" #define NUM_ELEMENTS(a) (sizeof(a) / sizeof((a)[0])) @@ -425,45 +426,130 @@ void get_hdr_visual_confirm_color( } void get_subvp_visual_confirm_color( - struct dc *dc, - struct dc_state *context, struct pipe_ctx *pipe_ctx, struct tg_color *color) { uint32_t color_value = MAX_TG_COLOR_VALUE; - bool enable_subvp = false; - int i; - - if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx || !context) - return; + if (pipe_ctx) { + switch (pipe_ctx->p_state_type) { + case P_STATE_SUB_VP: + color->color_r_cr = color_value; + color->color_g_y = 0; + color->color_b_cb = 0; + break; + case P_STATE_DRR_SUB_VP: + color->color_r_cr = 0; + color->color_g_y = color_value; + color->color_b_cb = 0; + break; + case P_STATE_V_BLANK_SUB_VP: + color->color_r_cr = 0; + color->color_g_y = 0; + color->color_b_cb = color_value; + break; + default: + break; + } + } +} - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; +void get_mclk_switch_visual_confirm_color( + struct pipe_ctx *pipe_ctx, + struct tg_color *color) +{ + uint32_t color_value = MAX_TG_COLOR_VALUE; - if (pipe->stream && pipe->stream->mall_stream_config.paired_stream && - pipe->stream->mall_stream_config.type == SUBVP_MAIN) { - /* SubVP enable - red */ - color->color_g_y = 0; + if (pipe_ctx) { + switch (pipe_ctx->p_state_type) { + case P_STATE_V_BLANK: + color->color_r_cr = color_value; + color->color_g_y = color_value; color->color_b_cb = 0; + break; + case P_STATE_FPO: + color->color_r_cr = 0; + color->color_g_y = color_value; + color->color_b_cb = color_value; + break; + case P_STATE_V_ACTIVE: color->color_r_cr = color_value; - enable_subvp = true; - - if (pipe_ctx->stream == pipe->stream) - return; + color->color_g_y = 0; + color->color_b_cb = color_value; + break; + case P_STATE_SUB_VP: + color->color_r_cr = color_value; + color->color_g_y = 0; + color->color_b_cb = 0; + break; + case P_STATE_DRR_SUB_VP: + color->color_r_cr = 0; + color->color_g_y = color_value; + color->color_b_cb = 0; + break; + case P_STATE_V_BLANK_SUB_VP: + color->color_r_cr = 0; + color->color_g_y = 0; + color->color_b_cb = color_value; + break; + default: break; } } +} - if (enable_subvp && pipe_ctx->stream->mall_stream_config.type == SUBVP_NONE) { - color->color_r_cr = 0; - if (pipe_ctx->stream->allow_freesync == 1) { - /* SubVP enable and DRR on - green */ - color->color_b_cb = 0; - color->color_g_y = color_value; +void set_p_state_switch_method( + struct dc *dc, + struct dc_state *context, + struct pipe_ctx *pipe_ctx) +{ + struct vba_vars_st *vba = &context->bw_ctx.dml.vba; + bool enable_subvp; + + if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx || !vba || !context) + return; + + if (vba->DRAMClockChangeSupport[vba->VoltageLevel][vba->maxMpcComb] != + dm_dram_clock_change_unsupported) { + /* MCLK switching is supported */ + if (!pipe_ctx->has_vactive_margin) { + /* In Vblank - yellow */ + pipe_ctx->p_state_type = P_STATE_V_BLANK; + + if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) { + /* FPO + Vblank - cyan */ + pipe_ctx->p_state_type = P_STATE_FPO; + } } else { - /* SubVP enable and No DRR - blue */ - color->color_g_y = 0; - color->color_b_cb = color_value; + /* In Vactive - pink */ + pipe_ctx->p_state_type = P_STATE_V_ACTIVE; + } + + /* SubVP */ + enable_subvp = false; + + for (int i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->stream && dc_state_get_paired_subvp_stream(context, pipe->stream) && + dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN) { + /* SubVP enable - red */ + pipe_ctx->p_state_type = P_STATE_SUB_VP; + enable_subvp = true; + + if (pipe_ctx->stream == pipe->stream) + return; + break; + } + } + + if (enable_subvp && dc_state_get_pipe_subvp_type(context, pipe_ctx) == SUBVP_NONE) { + if (pipe_ctx->stream->allow_freesync == 1) { + /* SubVP enable and DRR on - green */ + pipe_ctx->p_state_type = P_STATE_DRR_SUB_VP; + } else { + /* SubVP enable and No DRR - blue */ + pipe_ctx->p_state_type = P_STATE_V_BLANK_SUB_VP; + } } } } @@ -473,7 +559,8 @@ void hwss_build_fast_sequence(struct dc *dc, unsigned int dmub_cmd_count, struct block_sequence block_sequence[], int *num_steps, - struct pipe_ctx *pipe_ctx) + struct pipe_ctx *pipe_ctx, + struct dc_stream_status *stream_status) { struct dc_plane_state *plane = pipe_ctx->plane_state; struct dc_stream_state *stream = pipe_ctx->stream; @@ -490,7 +577,8 @@ void hwss_build_fast_sequence(struct dc *dc, if (dc->hwss.subvp_pipe_control_lock_fast) { block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.dc = dc; block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.lock = true; - block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.pipe_ctx = pipe_ctx; + block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.subvp_immediate_flip = + plane->flip_immediate && stream_status->mall_stream_config.type == SUBVP_MAIN; block_sequence[*num_steps].func = DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST; (*num_steps)++; } @@ -529,7 +617,7 @@ void hwss_build_fast_sequence(struct dc *dc, } if (dc->hwss.update_plane_addr && current_mpc_pipe->plane_state->update_flags.bits.addr_update) { if (resource_is_pipe_type(current_mpc_pipe, OTG_MASTER) && - current_mpc_pipe->stream->mall_stream_config.type == SUBVP_MAIN) { + stream_status->mall_stream_config.type == SUBVP_MAIN) { block_sequence[*num_steps].params.subvp_save_surf_addr.dc_dmub_srv = dc->ctx->dmub_srv; block_sequence[*num_steps].params.subvp_save_surf_addr.addr = ¤t_mpc_pipe->plane_state->address; block_sequence[*num_steps].params.subvp_save_surf_addr.subvp_index = current_mpc_pipe->subvp_index; @@ -612,7 +700,8 @@ void hwss_build_fast_sequence(struct dc *dc, if (dc->hwss.subvp_pipe_control_lock_fast) { block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.dc = dc; block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.lock = false; - block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.pipe_ctx = pipe_ctx; + block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.subvp_immediate_flip = + plane->flip_immediate && stream_status->mall_stream_config.type == SUBVP_MAIN; block_sequence[*num_steps].func = DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST; (*num_steps)++; } @@ -724,7 +813,7 @@ void hwss_send_dmcub_cmd(union block_sequence_params *params) union dmub_rb_cmd *cmd = params->send_dmcub_cmd_params.cmd; enum dm_dmub_wait_type wait_type = params->send_dmcub_cmd_params.wait_type; - dm_execute_dmub_cmd(ctx, cmd, wait_type); + dc_wake_and_execute_dmub_cmd(ctx, cmd, wait_type); } void hwss_program_manual_trigger(union block_sequence_params *params) @@ -812,42 +901,6 @@ void hwss_subvp_save_surf_addr(union block_sequence_params *params) dc_dmub_srv_subvp_save_surf_addr(dc_dmub_srv, addr, subvp_index); } -void get_mclk_switch_visual_confirm_color( - struct dc *dc, - struct dc_state *context, - struct pipe_ctx *pipe_ctx, - struct tg_color *color) -{ - uint32_t color_value = MAX_TG_COLOR_VALUE; - struct vba_vars_st *vba = &context->bw_ctx.dml.vba; - - if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx || !vba || !context) - return; - - if (vba->DRAMClockChangeSupport[vba->VoltageLevel][vba->maxMpcComb] != - dm_dram_clock_change_unsupported) { - /* MCLK switching is supported */ - if (!pipe_ctx->has_vactive_margin) { - /* In Vblank - yellow */ - color->color_r_cr = color_value; - color->color_g_y = color_value; - - if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) { - /* FPO + Vblank - cyan */ - color->color_r_cr = 0; - color->color_g_y = color_value; - color->color_b_cb = color_value; - } - } else { - /* In Vactive - pink */ - color->color_r_cr = color_value; - color->color_b_cb = color_value; - } - /* SubVP */ - get_subvp_visual_confirm_color(dc, context, pipe_ctx, color); - } -} - void get_surface_tile_visual_confirm_color( struct pipe_ctx *pipe_ctx, struct tg_color *color) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c index ed94187c2afa..c6c35037bdb8 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c @@ -467,6 +467,13 @@ bool dc_link_setup_psr(struct dc_link *link, return link->dc->link_srv->edp_setup_psr(link, stream, psr_config, psr_context); } +bool dc_link_set_replay_allow_active(struct dc_link *link, const bool *allow_active, + bool wait, bool force_static, const unsigned int *power_opts) +{ + return link->dc->link_srv->edp_set_replay_allow_active(link, allow_active, wait, + force_static, power_opts); +} + bool dc_link_get_replay_state(const struct dc_link *link, uint64_t *state) { return link->dc->link_srv->edp_get_replay_state(link, state); @@ -497,7 +504,7 @@ void dc_link_enable_hpd_filter(struct dc_link *link, bool enable) link->dc->link_srv->enable_hpd_filter(link, enable); } -bool dc_link_validate(struct dc *dc, const struct dc_stream_state *streams, const unsigned int count) +bool dc_link_dp_dpia_validate(struct dc *dc, const struct dc_stream_state *streams, const unsigned int count) { return dc->link_srv->validate_dpia_bandwidth(streams, count); } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 4382d9ae4292..9fbdb09697fd 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -42,6 +42,7 @@ #include "link_enc_cfg.h" #include "link.h" #include "clk_mgr.h" +#include "dc_state_priv.h" #include "virtual/virtual_link_hwss.h" #include "link/hwss/link_hwss_dio.h" #include "link/hwss/link_hwss_dpia.h" @@ -2193,6 +2194,10 @@ void resource_log_pipe_topology_update(struct dc *dc, struct dc_state *state) for (stream_idx = 0; stream_idx < state->stream_count; stream_idx++) { otg_master = resource_get_otg_master_for_stream( &state->res_ctx, state->streams[stream_idx]); + if (!otg_master || otg_master->stream_res.tg == NULL) { + DC_LOG_DC("topology update: otg_master NULL stream_idx %d!\n", stream_idx); + return; + } slice_count = resource_get_opp_heads_for_otg_master(otg_master, &state->res_ctx, opp_heads); for (slice_idx = 0; slice_idx < slice_count; slice_idx++) { @@ -2459,6 +2464,9 @@ void resource_remove_otg_master_for_stream_output(struct dc_state *context, struct pipe_ctx *otg_master = resource_get_otg_master_for_stream( &context->res_ctx, stream); + if (!otg_master) + return; + ASSERT(resource_get_odm_slice_count(otg_master) == 1); ASSERT(otg_master->plane_state == NULL); ASSERT(otg_master->stream_res.stream_enc); @@ -2993,189 +3001,6 @@ bool resource_update_pipes_for_plane_with_slice_count( return result; } -bool dc_add_plane_to_context( - const struct dc *dc, - struct dc_stream_state *stream, - struct dc_plane_state *plane_state, - struct dc_state *context) -{ - struct resource_pool *pool = dc->res_pool; - struct pipe_ctx *otg_master_pipe; - struct dc_stream_status *stream_status = NULL; - bool added = false; - - stream_status = dc_stream_get_status_from_state(context, stream); - if (stream_status == NULL) { - dm_error("Existing stream not found; failed to attach surface!\n"); - goto out; - } else if (stream_status->plane_count == MAX_SURFACE_NUM) { - dm_error("Surface: can not attach plane_state %p! Maximum is: %d\n", - plane_state, MAX_SURFACE_NUM); - goto out; - } - - otg_master_pipe = resource_get_otg_master_for_stream( - &context->res_ctx, stream); - added = resource_append_dpp_pipes_for_plane_composition(context, - dc->current_state, pool, otg_master_pipe, plane_state); - - if (added) { - stream_status->plane_states[stream_status->plane_count] = - plane_state; - stream_status->plane_count++; - dc_plane_state_retain(plane_state); - } - -out: - return added; -} - -bool dc_remove_plane_from_context( - const struct dc *dc, - struct dc_stream_state *stream, - struct dc_plane_state *plane_state, - struct dc_state *context) -{ - int i; - struct dc_stream_status *stream_status = NULL; - struct resource_pool *pool = dc->res_pool; - - if (!plane_state) - return true; - - for (i = 0; i < context->stream_count; i++) - if (context->streams[i] == stream) { - stream_status = &context->stream_status[i]; - break; - } - - if (stream_status == NULL) { - dm_error("Existing stream not found; failed to remove plane.\n"); - return false; - } - - resource_remove_dpp_pipes_for_plane_composition( - context, pool, plane_state); - - for (i = 0; i < stream_status->plane_count; i++) { - if (stream_status->plane_states[i] == plane_state) { - dc_plane_state_release(stream_status->plane_states[i]); - break; - } - } - - if (i == stream_status->plane_count) { - dm_error("Existing plane_state not found; failed to detach it!\n"); - return false; - } - - stream_status->plane_count--; - - /* Start at the plane we've just released, and move all the planes one index forward to "trim" the array */ - for (; i < stream_status->plane_count; i++) - stream_status->plane_states[i] = stream_status->plane_states[i + 1]; - - stream_status->plane_states[stream_status->plane_count] = NULL; - - if (stream_status->plane_count == 0 && dc->config.enable_windowed_mpo_odm) - /* ODM combine could prevent us from supporting more planes - * we will reset ODM slice count back to 1 when all planes have - * been removed to maximize the amount of planes supported when - * new planes are added. - */ - resource_update_pipes_for_stream_with_slice_count( - context, dc->current_state, dc->res_pool, stream, 1); - - return true; -} - -/** - * dc_rem_all_planes_for_stream - Remove planes attached to the target stream. - * - * @dc: Current dc state. - * @stream: Target stream, which we want to remove the attached plans. - * @context: New context. - * - * Return: - * Return true if DC was able to remove all planes from the target - * stream, otherwise, return false. - */ -bool dc_rem_all_planes_for_stream( - const struct dc *dc, - struct dc_stream_state *stream, - struct dc_state *context) -{ - int i, old_plane_count; - struct dc_stream_status *stream_status = NULL; - struct dc_plane_state *del_planes[MAX_SURFACE_NUM] = { 0 }; - - for (i = 0; i < context->stream_count; i++) - if (context->streams[i] == stream) { - stream_status = &context->stream_status[i]; - break; - } - - if (stream_status == NULL) { - dm_error("Existing stream %p not found!\n", stream); - return false; - } - - old_plane_count = stream_status->plane_count; - - for (i = 0; i < old_plane_count; i++) - del_planes[i] = stream_status->plane_states[i]; - - for (i = 0; i < old_plane_count; i++) - if (!dc_remove_plane_from_context(dc, stream, del_planes[i], context)) - return false; - - return true; -} - -static bool add_all_planes_for_stream( - const struct dc *dc, - struct dc_stream_state *stream, - const struct dc_validation_set set[], - int set_count, - struct dc_state *context) -{ - int i, j; - - for (i = 0; i < set_count; i++) - if (set[i].stream == stream) - break; - - if (i == set_count) { - dm_error("Stream %p not found in set!\n", stream); - return false; - } - - for (j = 0; j < set[i].plane_count; j++) - if (!dc_add_plane_to_context(dc, stream, set[i].plane_states[j], context)) - return false; - - return true; -} - -bool dc_add_all_planes_for_stream( - const struct dc *dc, - struct dc_stream_state *stream, - struct dc_plane_state * const *plane_states, - int plane_count, - struct dc_state *context) -{ - struct dc_validation_set set; - int i; - - set.stream = stream; - set.plane_count = plane_count; - - for (i = 0; i < plane_count; i++) - set.plane_states[i] = plane_states[i]; - - return add_all_planes_for_stream(dc, stream, &set, 1, context); -} - bool dc_is_timing_changed(struct dc_stream_state *cur_stream, struct dc_stream_state *new_stream) { @@ -3327,84 +3152,6 @@ static struct audio *find_first_free_audio( return NULL; } -/* - * dc_add_stream_to_ctx() - Add a new dc_stream_state to a dc_state. - */ -enum dc_status dc_add_stream_to_ctx( - struct dc *dc, - struct dc_state *new_ctx, - struct dc_stream_state *stream) -{ - enum dc_status res; - DC_LOGGER_INIT(dc->ctx->logger); - - if (new_ctx->stream_count >= dc->res_pool->timing_generator_count) { - DC_LOG_WARNING("Max streams reached, can't add stream %p !\n", stream); - return DC_ERROR_UNEXPECTED; - } - - new_ctx->streams[new_ctx->stream_count] = stream; - dc_stream_retain(stream); - new_ctx->stream_count++; - - res = resource_add_otg_master_for_stream_output( - new_ctx, dc->res_pool, stream); - if (res != DC_OK) - DC_LOG_WARNING("Adding stream %p to context failed with err %d!\n", stream, res); - - return res; -} - -/* - * dc_remove_stream_from_ctx() - Remove a stream from a dc_state. - */ -enum dc_status dc_remove_stream_from_ctx( - struct dc *dc, - struct dc_state *new_ctx, - struct dc_stream_state *stream) -{ - int i; - struct dc_context *dc_ctx = dc->ctx; - struct pipe_ctx *del_pipe = resource_get_otg_master_for_stream( - &new_ctx->res_ctx, stream); - - if (!del_pipe) { - DC_ERROR("Pipe not found for stream %p !\n", stream); - return DC_ERROR_UNEXPECTED; - } - - resource_update_pipes_for_stream_with_slice_count(new_ctx, - dc->current_state, dc->res_pool, stream, 1); - resource_remove_otg_master_for_stream_output( - new_ctx, dc->res_pool, stream); - - for (i = 0; i < new_ctx->stream_count; i++) - if (new_ctx->streams[i] == stream) - break; - - if (new_ctx->streams[i] != stream) { - DC_ERROR("Context doesn't have stream %p !\n", stream); - return DC_ERROR_UNEXPECTED; - } - - dc_stream_release(new_ctx->streams[i]); - new_ctx->stream_count--; - - /* Trim back arrays */ - for (; i < new_ctx->stream_count; i++) { - new_ctx->streams[i] = new_ctx->streams[i + 1]; - new_ctx->stream_status[i] = new_ctx->stream_status[i + 1]; - } - - new_ctx->streams[new_ctx->stream_count] = NULL; - memset( - &new_ctx->stream_status[new_ctx->stream_count], - 0, - sizeof(new_ctx->stream_status[0])); - - return DC_OK; -} - static struct dc_stream_state *find_pll_sharable_stream( struct dc_stream_state *stream_needs_pll, struct dc_state *context) @@ -3784,34 +3531,6 @@ enum dc_status resource_map_pool_resources( return DC_ERROR_UNEXPECTED; } -/** - * dc_resource_state_copy_construct_current() - Creates a new dc_state from existing state - * - * @dc: copy out of dc->current_state - * @dst_ctx: copy into this - * - * This function makes a shallow copy of the current DC state and increments - * refcounts on existing streams and planes. - */ -void dc_resource_state_copy_construct_current( - const struct dc *dc, - struct dc_state *dst_ctx) -{ - dc_resource_state_copy_construct(dc->current_state, dst_ctx); -} - - -void dc_resource_state_construct( - const struct dc *dc, - struct dc_state *dst_ctx) -{ - dst_ctx->clk_mgr = dc->clk_mgr; - - /* Initialise DIG link encoder resource tracking variables. */ - link_enc_cfg_init(dc, dst_ctx); -} - - bool dc_resource_is_dsc_encoding_supported(const struct dc *dc) { if (dc->res_pool == NULL) @@ -3855,6 +3574,31 @@ static bool planes_changed_for_existing_stream(struct dc_state *context, return false; } +static bool add_all_planes_for_stream( + const struct dc *dc, + struct dc_stream_state *stream, + const struct dc_validation_set set[], + int set_count, + struct dc_state *state) +{ + int i, j; + + for (i = 0; i < set_count; i++) + if (set[i].stream == stream) + break; + + if (i == set_count) { + dm_error("Stream %p not found in set!\n", stream); + return false; + } + + for (j = 0; j < set[i].plane_count; j++) + if (!dc_state_add_plane(dc, stream, set[i].plane_states[j], state)) + return false; + + return true; +} + /** * dc_validate_with_context - Validate and update the potential new stream in the context object * @@ -3960,7 +3704,8 @@ enum dc_status dc_validate_with_context(struct dc *dc, unchanged_streams[i], set, set_count)) { - if (!dc_rem_all_planes_for_stream(dc, + + if (!dc_state_rem_all_planes_for_stream(dc, unchanged_streams[i], context)) { res = DC_FAIL_DETACH_SURFACES; @@ -3982,12 +3727,24 @@ enum dc_status dc_validate_with_context(struct dc *dc, } } - if (!dc_rem_all_planes_for_stream(dc, del_streams[i], context)) { - res = DC_FAIL_DETACH_SURFACES; - goto fail; + if (dc_state_get_stream_subvp_type(context, del_streams[i]) == SUBVP_PHANTOM) { + /* remove phantoms specifically */ + if (!dc_state_rem_all_phantom_planes_for_stream(dc, del_streams[i], context, true)) { + res = DC_FAIL_DETACH_SURFACES; + goto fail; + } + + res = dc_state_remove_phantom_stream(dc, context, del_streams[i]); + dc_state_release_phantom_stream(dc, context, del_streams[i]); + } else { + if (!dc_state_rem_all_planes_for_stream(dc, del_streams[i], context)) { + res = DC_FAIL_DETACH_SURFACES; + goto fail; + } + + res = dc_state_remove_stream(dc, context, del_streams[i]); } - res = dc_remove_stream_from_ctx(dc, context, del_streams[i]); if (res != DC_OK) goto fail; } @@ -4010,7 +3767,7 @@ enum dc_status dc_validate_with_context(struct dc *dc, /* Add new streams and then add all planes for the new stream */ for (i = 0; i < add_streams_count; i++) { calculate_phy_pix_clks(add_streams[i]); - res = dc_add_stream_to_ctx(dc, context, add_streams[i]); + res = dc_state_add_stream(dc, context, add_streams[i]); if (res != DC_OK) goto fail; @@ -4516,84 +4273,6 @@ static void set_vtem_info_packet( *info_packet = stream->vtem_infopacket; } -void dc_resource_state_destruct(struct dc_state *context) -{ - int i, j; - - for (i = 0; i < context->stream_count; i++) { - for (j = 0; j < context->stream_status[i].plane_count; j++) - dc_plane_state_release( - context->stream_status[i].plane_states[j]); - - context->stream_status[i].plane_count = 0; - dc_stream_release(context->streams[i]); - context->streams[i] = NULL; - } - context->stream_count = 0; - context->stream_mask = 0; - memset(&context->res_ctx, 0, sizeof(context->res_ctx)); - memset(&context->pp_display_cfg, 0, sizeof(context->pp_display_cfg)); - memset(&context->dcn_bw_vars, 0, sizeof(context->dcn_bw_vars)); - context->clk_mgr = NULL; - memset(&context->bw_ctx.bw, 0, sizeof(context->bw_ctx.bw)); - memset(context->block_sequence, 0, sizeof(context->block_sequence)); - context->block_sequence_steps = 0; - memset(context->dc_dmub_cmd, 0, sizeof(context->dc_dmub_cmd)); - context->dmub_cmd_count = 0; - memset(&context->perf_params, 0, sizeof(context->perf_params)); - memset(&context->scratch, 0, sizeof(context->scratch)); -} - -void dc_resource_state_copy_construct( - const struct dc_state *src_ctx, - struct dc_state *dst_ctx) -{ - int i, j; - struct kref refcount = dst_ctx->refcount; -#ifdef CONFIG_DRM_AMD_DC_FP - struct dml2_context *dml2 = NULL; - - // Need to preserve allocated dml2 context - if (src_ctx->clk_mgr && src_ctx->clk_mgr->ctx->dc->debug.using_dml2) - dml2 = dst_ctx->bw_ctx.dml2; -#endif - - *dst_ctx = *src_ctx; - -#ifdef CONFIG_DRM_AMD_DC_FP - // Preserve allocated dml2 context - if (src_ctx->clk_mgr && src_ctx->clk_mgr->ctx->dc->debug.using_dml2) - dst_ctx->bw_ctx.dml2 = dml2; -#endif - - for (i = 0; i < MAX_PIPES; i++) { - struct pipe_ctx *cur_pipe = &dst_ctx->res_ctx.pipe_ctx[i]; - - if (cur_pipe->top_pipe) - cur_pipe->top_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->top_pipe->pipe_idx]; - - if (cur_pipe->bottom_pipe) - cur_pipe->bottom_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->bottom_pipe->pipe_idx]; - - if (cur_pipe->next_odm_pipe) - cur_pipe->next_odm_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->next_odm_pipe->pipe_idx]; - - if (cur_pipe->prev_odm_pipe) - cur_pipe->prev_odm_pipe = &dst_ctx->res_ctx.pipe_ctx[cur_pipe->prev_odm_pipe->pipe_idx]; - } - - for (i = 0; i < dst_ctx->stream_count; i++) { - dc_stream_retain(dst_ctx->streams[i]); - for (j = 0; j < dst_ctx->stream_status[i].plane_count; j++) - dc_plane_state_retain( - dst_ctx->stream_status[i].plane_states[j]); - } - - /* context refcount should not be overridden */ - dst_ctx->refcount = refcount; - -} - struct clock_source *dc_resource_find_first_free_pll( struct resource_context *res_ctx, const struct resource_pool *pool) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_state.c b/drivers/gpu/drm/amd/display/dc/core/dc_state.c new file mode 100644 index 000000000000..88c6436b28b6 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/core/dc_state.c @@ -0,0 +1,867 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "core_types.h" +#include "core_status.h" +#include "dc_state.h" +#include "dc_state_priv.h" +#include "dc_stream_priv.h" +#include "dc_plane_priv.h" + +#include "dm_services.h" +#include "resource.h" +#include "link_enc_cfg.h" + +#include "dml2/dml2_wrapper.h" +#include "dml2/dml2_internal_types.h" + +#define DC_LOGGER \ + dc->ctx->logger +#define DC_LOGGER_INIT(logger) + +/* Private dc_state helper functions */ +static bool dc_state_track_phantom_stream(struct dc_state *state, + struct dc_stream_state *phantom_stream) +{ + if (state->phantom_stream_count >= MAX_PHANTOM_PIPES) + return false; + + state->phantom_streams[state->phantom_stream_count++] = phantom_stream; + + return true; +} + +static bool dc_state_untrack_phantom_stream(struct dc_state *state, struct dc_stream_state *phantom_stream) +{ + bool res = false; + int i; + + /* first find phantom stream in the dc_state */ + for (i = 0; i < state->phantom_stream_count; i++) { + if (state->phantom_streams[i] == phantom_stream) { + state->phantom_streams[i] = NULL; + res = true; + break; + } + } + + /* failed to find stream in state */ + if (!res) + return res; + + /* trim back phantom streams */ + state->phantom_stream_count--; + for (; i < state->phantom_stream_count; i++) + state->phantom_streams[i] = state->phantom_streams[i + 1]; + + return res; +} + +static bool dc_state_is_phantom_stream_tracked(struct dc_state *state, struct dc_stream_state *phantom_stream) +{ + int i; + + for (i = 0; i < state->phantom_stream_count; i++) { + if (state->phantom_streams[i] == phantom_stream) + return true; + } + + return false; +} + +static bool dc_state_track_phantom_plane(struct dc_state *state, + struct dc_plane_state *phantom_plane) +{ + if (state->phantom_plane_count >= MAX_PHANTOM_PIPES) + return false; + + state->phantom_planes[state->phantom_plane_count++] = phantom_plane; + + return true; +} + +static bool dc_state_untrack_phantom_plane(struct dc_state *state, struct dc_plane_state *phantom_plane) +{ + bool res = false; + int i; + + /* first find phantom plane in the dc_state */ + for (i = 0; i < state->phantom_plane_count; i++) { + if (state->phantom_planes[i] == phantom_plane) { + state->phantom_planes[i] = NULL; + res = true; + break; + } + } + + /* failed to find plane in state */ + if (!res) + return res; + + /* trim back phantom planes */ + state->phantom_plane_count--; + for (; i < state->phantom_plane_count; i++) + state->phantom_planes[i] = state->phantom_planes[i + 1]; + + return res; +} + +static bool dc_state_is_phantom_plane_tracked(struct dc_state *state, struct dc_plane_state *phantom_plane) +{ + int i; + + for (i = 0; i < state->phantom_plane_count; i++) { + if (state->phantom_planes[i] == phantom_plane) + return true; + } + + return false; +} + +static void dc_state_copy_internal(struct dc_state *dst_state, struct dc_state *src_state) +{ + int i, j; + + memcpy(dst_state, src_state, sizeof(struct dc_state)); + + for (i = 0; i < MAX_PIPES; i++) { + struct pipe_ctx *cur_pipe = &dst_state->res_ctx.pipe_ctx[i]; + + if (cur_pipe->top_pipe) + cur_pipe->top_pipe = &dst_state->res_ctx.pipe_ctx[cur_pipe->top_pipe->pipe_idx]; + + if (cur_pipe->bottom_pipe) + cur_pipe->bottom_pipe = &dst_state->res_ctx.pipe_ctx[cur_pipe->bottom_pipe->pipe_idx]; + + if (cur_pipe->prev_odm_pipe) + cur_pipe->prev_odm_pipe = &dst_state->res_ctx.pipe_ctx[cur_pipe->prev_odm_pipe->pipe_idx]; + + if (cur_pipe->next_odm_pipe) + cur_pipe->next_odm_pipe = &dst_state->res_ctx.pipe_ctx[cur_pipe->next_odm_pipe->pipe_idx]; + } + + /* retain phantoms */ + for (i = 0; i < dst_state->phantom_stream_count; i++) + dc_stream_retain(dst_state->phantom_streams[i]); + + for (i = 0; i < dst_state->phantom_plane_count; i++) + dc_plane_state_retain(dst_state->phantom_planes[i]); + + /* retain streams and planes */ + for (i = 0; i < dst_state->stream_count; i++) { + dc_stream_retain(dst_state->streams[i]); + for (j = 0; j < dst_state->stream_status[i].plane_count; j++) + dc_plane_state_retain( + dst_state->stream_status[i].plane_states[j]); + } + +} + +static void init_state(struct dc *dc, struct dc_state *state) +{ + /* Each context must have their own instance of VBA and in order to + * initialize and obtain IP and SOC the base DML instance from DC is + * initially copied into every context + */ + memcpy(&state->bw_ctx.dml, &dc->dml, sizeof(struct display_mode_lib)); +} + +/* Public dc_state functions */ +struct dc_state *dc_state_create(struct dc *dc) +{ + struct dc_state *state = kvzalloc(sizeof(struct dc_state), + GFP_KERNEL); + + if (!state) + return NULL; + + init_state(dc, state); + dc_state_construct(dc, state); + +#ifdef CONFIG_DRM_AMD_DC_FP + if (dc->debug.using_dml2) + dml2_create(dc, &dc->dml2_options, &state->bw_ctx.dml2); +#endif + + kref_init(&state->refcount); + + return state; +} + +void dc_state_copy(struct dc_state *dst_state, struct dc_state *src_state) +{ + struct kref refcount = dst_state->refcount; +#ifdef CONFIG_DRM_AMD_DC_FP + struct dml2_context *dst_dml2 = dst_state->bw_ctx.dml2; +#endif + + dc_state_copy_internal(dst_state, src_state); + +#ifdef CONFIG_DRM_AMD_DC_FP + dst_state->bw_ctx.dml2 = dst_dml2; + if (src_state->bw_ctx.dml2) + dml2_copy(dst_state->bw_ctx.dml2, src_state->bw_ctx.dml2); +#endif + + /* context refcount should not be overridden */ + dst_state->refcount = refcount; +} + +struct dc_state *dc_state_create_copy(struct dc_state *src_state) +{ + struct dc_state *new_state; + + new_state = kvmalloc(sizeof(struct dc_state), + GFP_KERNEL); + if (!new_state) + return NULL; + + dc_state_copy_internal(new_state, src_state); + +#ifdef CONFIG_DRM_AMD_DC_FP + if (src_state->bw_ctx.dml2 && + !dml2_create_copy(&new_state->bw_ctx.dml2, src_state->bw_ctx.dml2)) { + dc_state_release(new_state); + return NULL; + } +#endif + + kref_init(&new_state->refcount); + + return new_state; +} + +void dc_state_copy_current(struct dc *dc, struct dc_state *dst_state) +{ + dc_state_copy(dst_state, dc->current_state); +} + +struct dc_state *dc_state_create_current_copy(struct dc *dc) +{ + return dc_state_create_copy(dc->current_state); +} + +void dc_state_construct(struct dc *dc, struct dc_state *state) +{ + state->clk_mgr = dc->clk_mgr; + + /* Initialise DIG link encoder resource tracking variables. */ + if (dc->res_pool) + link_enc_cfg_init(dc, state); +} + +void dc_state_destruct(struct dc_state *state) +{ + int i, j; + + for (i = 0; i < state->stream_count; i++) { + for (j = 0; j < state->stream_status[i].plane_count; j++) + dc_plane_state_release( + state->stream_status[i].plane_states[j]); + + state->stream_status[i].plane_count = 0; + dc_stream_release(state->streams[i]); + state->streams[i] = NULL; + } + state->stream_count = 0; + + /* release tracked phantoms */ + for (i = 0; i < state->phantom_stream_count; i++) { + dc_stream_release(state->phantom_streams[i]); + state->phantom_streams[i] = NULL; + } + + for (i = 0; i < state->phantom_plane_count; i++) { + dc_plane_state_release(state->phantom_planes[i]); + state->phantom_planes[i] = NULL; + } + state->stream_mask = 0; + memset(&state->res_ctx, 0, sizeof(state->res_ctx)); + memset(&state->pp_display_cfg, 0, sizeof(state->pp_display_cfg)); + memset(&state->dcn_bw_vars, 0, sizeof(state->dcn_bw_vars)); + state->clk_mgr = NULL; + memset(&state->bw_ctx.bw, 0, sizeof(state->bw_ctx.bw)); + memset(state->block_sequence, 0, sizeof(state->block_sequence)); + state->block_sequence_steps = 0; + memset(state->dc_dmub_cmd, 0, sizeof(state->dc_dmub_cmd)); + state->dmub_cmd_count = 0; + memset(&state->perf_params, 0, sizeof(state->perf_params)); + memset(&state->scratch, 0, sizeof(state->scratch)); +} + +void dc_state_retain(struct dc_state *state) +{ + kref_get(&state->refcount); +} + +static void dc_state_free(struct kref *kref) +{ + struct dc_state *state = container_of(kref, struct dc_state, refcount); + + dc_state_destruct(state); + +#ifdef CONFIG_DRM_AMD_DC_FP + dml2_destroy(state->bw_ctx.dml2); + state->bw_ctx.dml2 = 0; +#endif + + kvfree(state); +} + +void dc_state_release(struct dc_state *state) +{ + kref_put(&state->refcount, dc_state_free); +} +/* + * dc_state_add_stream() - Add a new dc_stream_state to a dc_state. + */ +enum dc_status dc_state_add_stream( + struct dc *dc, + struct dc_state *state, + struct dc_stream_state *stream) +{ + enum dc_status res; + + DC_LOGGER_INIT(dc->ctx->logger); + + if (state->stream_count >= dc->res_pool->timing_generator_count) { + DC_LOG_WARNING("Max streams reached, can't add stream %p !\n", stream); + return DC_ERROR_UNEXPECTED; + } + + state->streams[state->stream_count] = stream; + dc_stream_retain(stream); + state->stream_count++; + + res = resource_add_otg_master_for_stream_output( + state, dc->res_pool, stream); + if (res != DC_OK) + DC_LOG_WARNING("Adding stream %p to context failed with err %d!\n", stream, res); + + return res; +} + +/* + * dc_state_remove_stream() - Remove a stream from a dc_state. + */ +enum dc_status dc_state_remove_stream( + struct dc *dc, + struct dc_state *state, + struct dc_stream_state *stream) +{ + int i; + struct pipe_ctx *del_pipe = resource_get_otg_master_for_stream( + &state->res_ctx, stream); + + if (!del_pipe) { + dm_error("Pipe not found for stream %p !\n", stream); + return DC_ERROR_UNEXPECTED; + } + + resource_update_pipes_for_stream_with_slice_count(state, + dc->current_state, dc->res_pool, stream, 1); + resource_remove_otg_master_for_stream_output( + state, dc->res_pool, stream); + + for (i = 0; i < state->stream_count; i++) + if (state->streams[i] == stream) + break; + + if (state->streams[i] != stream) { + dm_error("Context doesn't have stream %p !\n", stream); + return DC_ERROR_UNEXPECTED; + } + + dc_stream_release(state->streams[i]); + state->stream_count--; + + /* Trim back arrays */ + for (; i < state->stream_count; i++) { + state->streams[i] = state->streams[i + 1]; + state->stream_status[i] = state->stream_status[i + 1]; + } + + state->streams[state->stream_count] = NULL; + memset( + &state->stream_status[state->stream_count], + 0, + sizeof(state->stream_status[0])); + + return DC_OK; +} + +bool dc_state_add_plane( + const struct dc *dc, + struct dc_stream_state *stream, + struct dc_plane_state *plane_state, + struct dc_state *state) +{ + struct resource_pool *pool = dc->res_pool; + struct pipe_ctx *otg_master_pipe; + struct dc_stream_status *stream_status = NULL; + bool added = false; + + stream_status = dc_state_get_stream_status(state, stream); + if (stream_status == NULL) { + dm_error("Existing stream not found; failed to attach surface!\n"); + goto out; + } else if (stream_status->plane_count == MAX_SURFACE_NUM) { + dm_error("Surface: can not attach plane_state %p! Maximum is: %d\n", + plane_state, MAX_SURFACE_NUM); + goto out; + } + + otg_master_pipe = resource_get_otg_master_for_stream( + &state->res_ctx, stream); + if (otg_master_pipe) + added = resource_append_dpp_pipes_for_plane_composition(state, + dc->current_state, pool, otg_master_pipe, plane_state); + + if (added) { + stream_status->plane_states[stream_status->plane_count] = + plane_state; + stream_status->plane_count++; + dc_plane_state_retain(plane_state); + } + +out: + return added; +} + +bool dc_state_remove_plane( + const struct dc *dc, + struct dc_stream_state *stream, + struct dc_plane_state *plane_state, + struct dc_state *state) +{ + int i; + struct dc_stream_status *stream_status = NULL; + struct resource_pool *pool = dc->res_pool; + + if (!plane_state) + return true; + + for (i = 0; i < state->stream_count; i++) + if (state->streams[i] == stream) { + stream_status = &state->stream_status[i]; + break; + } + + if (stream_status == NULL) { + dm_error("Existing stream not found; failed to remove plane.\n"); + return false; + } + + resource_remove_dpp_pipes_for_plane_composition( + state, pool, plane_state); + + for (i = 0; i < stream_status->plane_count; i++) { + if (stream_status->plane_states[i] == plane_state) { + dc_plane_state_release(stream_status->plane_states[i]); + break; + } + } + + if (i == stream_status->plane_count) { + dm_error("Existing plane_state not found; failed to detach it!\n"); + return false; + } + + stream_status->plane_count--; + + /* Start at the plane we've just released, and move all the planes one index forward to "trim" the array */ + for (; i < stream_status->plane_count; i++) + stream_status->plane_states[i] = stream_status->plane_states[i + 1]; + + stream_status->plane_states[stream_status->plane_count] = NULL; + + if (stream_status->plane_count == 0 && dc->config.enable_windowed_mpo_odm) + /* ODM combine could prevent us from supporting more planes + * we will reset ODM slice count back to 1 when all planes have + * been removed to maximize the amount of planes supported when + * new planes are added. + */ + resource_update_pipes_for_stream_with_slice_count( + state, dc->current_state, dc->res_pool, stream, 1); + + return true; +} + +/** + * dc_state_rem_all_planes_for_stream - Remove planes attached to the target stream. + * + * @dc: Current dc state. + * @stream: Target stream, which we want to remove the attached plans. + * @state: context from which the planes are to be removed. + * + * Return: + * Return true if DC was able to remove all planes from the target + * stream, otherwise, return false. + */ +bool dc_state_rem_all_planes_for_stream( + const struct dc *dc, + struct dc_stream_state *stream, + struct dc_state *state) +{ + int i, old_plane_count; + struct dc_stream_status *stream_status = NULL; + struct dc_plane_state *del_planes[MAX_SURFACE_NUM] = { 0 }; + + for (i = 0; i < state->stream_count; i++) + if (state->streams[i] == stream) { + stream_status = &state->stream_status[i]; + break; + } + + if (stream_status == NULL) { + dm_error("Existing stream %p not found!\n", stream); + return false; + } + + old_plane_count = stream_status->plane_count; + + for (i = 0; i < old_plane_count; i++) + del_planes[i] = stream_status->plane_states[i]; + + for (i = 0; i < old_plane_count; i++) + if (!dc_state_remove_plane(dc, stream, del_planes[i], state)) + return false; + + return true; +} + +bool dc_state_add_all_planes_for_stream( + const struct dc *dc, + struct dc_stream_state *stream, + struct dc_plane_state * const *plane_states, + int plane_count, + struct dc_state *state) +{ + int i; + bool result = true; + + for (i = 0; i < plane_count; i++) + if (!dc_state_add_plane(dc, stream, plane_states[i], state)) { + result = false; + break; + } + + return result; +} + +/* Private dc_state functions */ + +/** + * dc_state_get_stream_status - Get stream status from given dc state + * @state: DC state to find the stream status in + * @stream: The stream to get the stream status for + * + * The given stream is expected to exist in the given dc state. Otherwise, NULL + * will be returned. + */ +struct dc_stream_status *dc_state_get_stream_status( + struct dc_state *state, + struct dc_stream_state *stream) +{ + uint8_t i; + + if (state == NULL) + return NULL; + + for (i = 0; i < state->stream_count; i++) { + if (stream == state->streams[i]) + return &state->stream_status[i]; + } + + return NULL; +} + +enum mall_stream_type dc_state_get_pipe_subvp_type(const struct dc_state *state, + const struct pipe_ctx *pipe_ctx) +{ + return dc_state_get_stream_subvp_type(state, pipe_ctx->stream); +} + +enum mall_stream_type dc_state_get_stream_subvp_type(const struct dc_state *state, + const struct dc_stream_state *stream) +{ + int i; + + enum mall_stream_type type = SUBVP_NONE; + + for (i = 0; i < state->stream_count; i++) { + if (state->streams[i] == stream) { + type = state->stream_status[i].mall_stream_config.type; + break; + } + } + + return type; +} + +struct dc_stream_state *dc_state_get_paired_subvp_stream(const struct dc_state *state, + const struct dc_stream_state *stream) +{ + int i; + + struct dc_stream_state *paired_stream = NULL; + + for (i = 0; i < state->stream_count; i++) { + if (state->streams[i] == stream) { + paired_stream = state->stream_status[i].mall_stream_config.paired_stream; + break; + } + } + + return paired_stream; +} + +struct dc_stream_state *dc_state_create_phantom_stream(const struct dc *dc, + struct dc_state *state, + struct dc_stream_state *main_stream) +{ + struct dc_stream_state *phantom_stream; + + DC_LOGGER_INIT(dc->ctx->logger); + + phantom_stream = dc_create_stream_for_sink(main_stream->sink); + + if (!phantom_stream) { + DC_LOG_ERROR("Failed to allocate phantom stream.\n"); + return NULL; + } + + /* track phantom stream in dc_state */ + dc_state_track_phantom_stream(state, phantom_stream); + + phantom_stream->is_phantom = true; + phantom_stream->signal = SIGNAL_TYPE_VIRTUAL; + phantom_stream->dpms_off = true; + + return phantom_stream; +} + +void dc_state_release_phantom_stream(const struct dc *dc, + struct dc_state *state, + struct dc_stream_state *phantom_stream) +{ + DC_LOGGER_INIT(dc->ctx->logger); + + if (!dc_state_untrack_phantom_stream(state, phantom_stream)) { + DC_LOG_ERROR("Failed to free phantom stream %p in dc state %p.\n", phantom_stream, state); + return; + } + + dc_stream_release(phantom_stream); +} + +struct dc_plane_state *dc_state_create_phantom_plane(struct dc *dc, + struct dc_state *state, + struct dc_plane_state *main_plane) +{ + struct dc_plane_state *phantom_plane = dc_create_plane_state(dc); + + DC_LOGGER_INIT(dc->ctx->logger); + + if (!phantom_plane) { + DC_LOG_ERROR("Failed to allocate phantom plane.\n"); + return NULL; + } + + /* track phantom inside dc_state */ + dc_state_track_phantom_plane(state, phantom_plane); + + phantom_plane->is_phantom = true; + + return phantom_plane; +} + +void dc_state_release_phantom_plane(const struct dc *dc, + struct dc_state *state, + struct dc_plane_state *phantom_plane) +{ + DC_LOGGER_INIT(dc->ctx->logger); + + if (!dc_state_untrack_phantom_plane(state, phantom_plane)) { + DC_LOG_ERROR("Failed to free phantom plane %p in dc state %p.\n", phantom_plane, state); + return; + } + + dc_plane_state_release(phantom_plane); +} + +/* add phantom streams to context and generate correct meta inside dc_state */ +enum dc_status dc_state_add_phantom_stream(struct dc *dc, + struct dc_state *state, + struct dc_stream_state *phantom_stream, + struct dc_stream_state *main_stream) +{ + struct dc_stream_status *main_stream_status; + struct dc_stream_status *phantom_stream_status; + enum dc_status res = dc_state_add_stream(dc, state, phantom_stream); + + /* check if stream is tracked */ + if (res == DC_OK && !dc_state_is_phantom_stream_tracked(state, phantom_stream)) { + /* stream must be tracked if added to state */ + dc_state_track_phantom_stream(state, phantom_stream); + } + + /* setup subvp meta */ + main_stream_status = dc_state_get_stream_status(state, main_stream); + phantom_stream_status = dc_state_get_stream_status(state, phantom_stream); + phantom_stream_status->mall_stream_config.type = SUBVP_PHANTOM; + phantom_stream_status->mall_stream_config.paired_stream = main_stream; + main_stream_status->mall_stream_config.type = SUBVP_MAIN; + main_stream_status->mall_stream_config.paired_stream = phantom_stream; + + return res; +} + +enum dc_status dc_state_remove_phantom_stream(struct dc *dc, + struct dc_state *state, + struct dc_stream_state *phantom_stream) +{ + struct dc_stream_status *main_stream_status; + struct dc_stream_status *phantom_stream_status; + + /* reset subvp meta */ + phantom_stream_status = dc_state_get_stream_status(state, phantom_stream); + main_stream_status = dc_state_get_stream_status(state, phantom_stream_status->mall_stream_config.paired_stream); + phantom_stream_status->mall_stream_config.type = SUBVP_NONE; + phantom_stream_status->mall_stream_config.paired_stream = NULL; + if (main_stream_status) { + main_stream_status->mall_stream_config.type = SUBVP_NONE; + main_stream_status->mall_stream_config.paired_stream = NULL; + } + + /* remove stream from state */ + return dc_state_remove_stream(dc, state, phantom_stream); +} + +bool dc_state_add_phantom_plane( + const struct dc *dc, + struct dc_stream_state *phantom_stream, + struct dc_plane_state *phantom_plane, + struct dc_state *state) +{ + bool res = dc_state_add_plane(dc, phantom_stream, phantom_plane, state); + + /* check if stream is tracked */ + if (res && !dc_state_is_phantom_plane_tracked(state, phantom_plane)) { + /* stream must be tracked if added to state */ + dc_state_track_phantom_plane(state, phantom_plane); + } + + return res; +} + +bool dc_state_remove_phantom_plane( + const struct dc *dc, + struct dc_stream_state *phantom_stream, + struct dc_plane_state *phantom_plane, + struct dc_state *state) +{ + return dc_state_remove_plane(dc, phantom_stream, phantom_plane, state); +} + +bool dc_state_rem_all_phantom_planes_for_stream( + const struct dc *dc, + struct dc_stream_state *phantom_stream, + struct dc_state *state, + bool should_release_planes) +{ + int i, old_plane_count; + struct dc_stream_status *stream_status = NULL; + struct dc_plane_state *del_planes[MAX_SURFACE_NUM] = { 0 }; + + for (i = 0; i < state->stream_count; i++) + if (state->streams[i] == phantom_stream) { + stream_status = &state->stream_status[i]; + break; + } + + if (stream_status == NULL) { + dm_error("Existing stream %p not found!\n", phantom_stream); + return false; + } + + old_plane_count = stream_status->plane_count; + + for (i = 0; i < old_plane_count; i++) + del_planes[i] = stream_status->plane_states[i]; + + for (i = 0; i < old_plane_count; i++) { + if (!dc_state_remove_plane(dc, phantom_stream, del_planes[i], state)) + return false; + if (should_release_planes) + dc_state_release_phantom_plane(dc, state, del_planes[i]); + } + + return true; +} + +bool dc_state_add_all_phantom_planes_for_stream( + const struct dc *dc, + struct dc_stream_state *phantom_stream, + struct dc_plane_state * const *phantom_planes, + int plane_count, + struct dc_state *state) +{ + return dc_state_add_all_planes_for_stream(dc, phantom_stream, phantom_planes, plane_count, state); +} + +bool dc_state_remove_phantom_streams_and_planes( + struct dc *dc, + struct dc_state *state) +{ + int i; + bool removed_phantom = false; + struct dc_stream_state *phantom_stream = NULL; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &state->res_ctx.pipe_ctx[i]; + + if (pipe->plane_state && pipe->stream && dc_state_get_pipe_subvp_type(state, pipe) == SUBVP_PHANTOM) { + phantom_stream = pipe->stream; + + dc_state_rem_all_phantom_planes_for_stream(dc, phantom_stream, state, false); + dc_state_remove_phantom_stream(dc, state, phantom_stream); + removed_phantom = true; + } + } + return removed_phantom; +} + +void dc_state_release_phantom_streams_and_planes( + struct dc *dc, + struct dc_state *state) +{ + int i; + + for (i = 0; i < state->phantom_stream_count; i++) + dc_state_release_phantom_stream(dc, state, state->phantom_streams[i]); + + for (i = 0; i < state->phantom_plane_count; i++) + dc_state_release_phantom_plane(dc, state, state->phantom_planes[i]); +} diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 38cd29b210c0..54670e0b1518 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -31,6 +31,8 @@ #include "ipp.h" #include "timing_generator.h" #include "dc_dmub_srv.h" +#include "dc_state_priv.h" +#include "dc_stream_priv.h" #define DC_LOGGER dc->ctx->logger @@ -54,7 +56,7 @@ void update_stream_signal(struct dc_stream_state *stream, struct dc_sink *sink) } } -static bool dc_stream_construct(struct dc_stream_state *stream, +bool dc_stream_construct(struct dc_stream_state *stream, struct dc_sink *dc_sink_data) { uint32_t i = 0; @@ -121,13 +123,12 @@ static bool dc_stream_construct(struct dc_stream_state *stream, } stream->out_transfer_func->type = TF_TYPE_BYPASS; - stream->stream_id = stream->ctx->dc_stream_id_count; - stream->ctx->dc_stream_id_count++; + dc_stream_assign_stream_id(stream); return true; } -static void dc_stream_destruct(struct dc_stream_state *stream) +void dc_stream_destruct(struct dc_stream_state *stream) { dc_sink_release(stream->sink); if (stream->out_transfer_func != NULL) { @@ -136,6 +137,13 @@ static void dc_stream_destruct(struct dc_stream_state *stream) } } +void dc_stream_assign_stream_id(struct dc_stream_state *stream) +{ + /* MSB is reserved to indicate phantoms */ + stream->stream_id = stream->ctx->dc_stream_id_count; + stream->ctx->dc_stream_id_count++; +} + void dc_stream_retain(struct dc_stream_state *stream) { kref_get(&stream->refcount); @@ -196,8 +204,7 @@ struct dc_stream_state *dc_copy_stream(const struct dc_stream_state *stream) if (new_stream->out_transfer_func) dc_transfer_func_retain(new_stream->out_transfer_func); - new_stream->stream_id = new_stream->ctx->dc_stream_id_count; - new_stream->ctx->dc_stream_id_count++; + dc_stream_assign_stream_id(new_stream); /* If using dynamic encoder assignment, wait till stream committed to assign encoder. */ if (new_stream->ctx->dc->res_pool->funcs->link_encs_assign) @@ -209,31 +216,6 @@ struct dc_stream_state *dc_copy_stream(const struct dc_stream_state *stream) } /** - * dc_stream_get_status_from_state - Get stream status from given dc state - * @state: DC state to find the stream status in - * @stream: The stream to get the stream status for - * - * The given stream is expected to exist in the given dc state. Otherwise, NULL - * will be returned. - */ -struct dc_stream_status *dc_stream_get_status_from_state( - struct dc_state *state, - struct dc_stream_state *stream) -{ - uint8_t i; - - if (state == NULL) - return NULL; - - for (i = 0; i < state->stream_count; i++) { - if (stream == state->streams[i]) - return &state->stream_status[i]; - } - - return NULL; -} - -/** * dc_stream_get_status() - Get current stream status of the given stream state * @stream: The stream to get the stream status for. * @@ -244,7 +226,7 @@ struct dc_stream_status *dc_stream_get_status( struct dc_stream_state *stream) { struct dc *dc = stream->ctx->dc; - return dc_stream_get_status_from_state(dc->current_state, stream); + return dc_state_get_stream_status(dc->current_state, stream); } static void program_cursor_attributes( @@ -465,7 +447,8 @@ bool dc_stream_add_writeback(struct dc *dc, if (dc->hwss.enable_writeback) { struct dc_stream_status *stream_status = dc_stream_get_status(stream); struct dwbc *dwb = dc->res_pool->dwbc[wb_info->dwb_pipe_inst]; - dwb->otg_inst = stream_status->primary_otg_inst; + if (stream_status) + dwb->otg_inst = stream_status->primary_otg_inst; } if (!dc->hwss.update_bandwidth(dc, dc->current_state)) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c index a80e45300783..19a2c7140ae8 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c @@ -32,10 +32,12 @@ #include "transform.h" #include "dpp.h" +#include "dc_plane_priv.h" + /******************************************************************************* * Private functions ******************************************************************************/ -static void dc_plane_construct(struct dc_context *ctx, struct dc_plane_state *plane_state) +void dc_plane_construct(struct dc_context *ctx, struct dc_plane_state *plane_state) { plane_state->ctx = ctx; @@ -63,7 +65,7 @@ static void dc_plane_construct(struct dc_context *ctx, struct dc_plane_state *pl } -static void dc_plane_destruct(struct dc_plane_state *plane_state) +void dc_plane_destruct(struct dc_plane_state *plane_state) { if (plane_state->gamma_correction != NULL) { dc_gamma_release(&plane_state->gamma_correction); diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 2c85f8ee682f..5d7aa882416b 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -27,6 +27,8 @@ #define DC_INTERFACE_H_ #include "dc_types.h" +#include "dc_state.h" +#include "dc_plane.h" #include "grph_object_defs.h" #include "logger_types.h" #include "hdcp_msg_types.h" @@ -49,7 +51,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.264" +#define DC_VER "3.2.266" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -461,6 +463,12 @@ enum dml_hostvm_override_opts { DML_HOSTVM_OVERRIDE_TRUE = 0x2, }; +enum dc_replay_power_opts { + replay_power_opt_invalid = 0x0, + replay_power_opt_smu_opt_static_screen = 0x1, + replay_power_opt_z10_static_screen = 0x10, +}; + enum dcc_option { DCC_ENABLE = 0, DCC_DISABLE = 1, @@ -979,6 +987,8 @@ struct dc_debug_options { unsigned int ips2_eval_delay_us; unsigned int ips2_entry_delay_us; bool disable_timeout; + bool disable_extblankadj; + unsigned int static_screen_wait_frames; }; struct gpu_info_soc_bounding_box_v1_0; @@ -1389,13 +1399,6 @@ struct dc_surface_update { /* * Create a new surface with default parameters; */ -struct dc_plane_state *dc_create_plane_state(struct dc *dc); -const struct dc_plane_status *dc_plane_get_status( - const struct dc_plane_state *plane_state); - -void dc_plane_state_retain(struct dc_plane_state *plane_state); -void dc_plane_state_release(struct dc_plane_state *plane_state); - void dc_gamma_retain(struct dc_gamma *dc_gamma); void dc_gamma_release(struct dc_gamma **dc_gamma); struct dc_gamma *dc_create_gamma(void); @@ -1459,37 +1462,20 @@ enum dc_status dc_validate_global_state( struct dc_state *new_ctx, bool fast_validate); - -void dc_resource_state_construct( - const struct dc *dc, - struct dc_state *dst_ctx); - bool dc_acquire_release_mpc_3dlut( struct dc *dc, bool acquire, struct dc_stream_state *stream, struct dc_3dlut **lut, struct dc_transfer_func **shaper); -void dc_resource_state_copy_construct( - const struct dc_state *src_ctx, - struct dc_state *dst_ctx); - -void dc_resource_state_copy_construct_current( - const struct dc *dc, - struct dc_state *dst_ctx); - -void dc_resource_state_destruct(struct dc_state *context); - bool dc_resource_is_dsc_encoding_supported(const struct dc *dc); +void get_audio_check(struct audio_info *aud_modes, + struct audio_check *aud_chk); enum dc_status dc_commit_streams(struct dc *dc, struct dc_stream_state *streams[], uint8_t stream_count); -struct dc_state *dc_create_state(struct dc *dc); -struct dc_state *dc_copy_state(struct dc_state *src_ctx); -void dc_retain_state(struct dc_state *context); -void dc_release_state(struct dc_state *context); struct dc_plane_state *dc_get_surface_for_mpcc(struct dc *dc, struct dc_stream_state *stream, @@ -2098,6 +2084,20 @@ bool dc_link_setup_psr(struct dc_link *dc_link, const struct dc_stream_state *stream, struct psr_config *psr_config, struct psr_context *psr_context); +/* + * Communicate with DMUB to allow or disallow Panel Replay on the specified link: + * + * @link: pointer to the dc_link struct instance + * @enable: enable(active) or disable(inactive) replay + * @wait: state transition need to wait the active set completed. + * @force_static: force disable(inactive) the replay + * @power_opts: set power optimazation parameters to DMUB. + * + * return: allow Replay active will return true, else will return false. + */ +bool dc_link_set_replay_allow_active(struct dc_link *dc_link, const bool *enable, + bool wait, bool force_static, const unsigned int *power_opts); + bool dc_link_get_replay_state(const struct dc_link *dc_link, uint64_t *state); /* On eDP links this function call will stall until T12 has elapsed. @@ -2193,11 +2193,11 @@ int dc_link_dp_dpia_handle_usb4_bandwidth_allocation_for_link( * * @dc: pointer to dc struct * @stream: pointer to all possible streams - * @num_streams: number of valid DPIA streams + * @count: number of valid DPIA streams * * return: TRUE if bw used by DPIAs doesn't exceed available BW else return FALSE */ -bool dc_link_validate(struct dc *dc, const struct dc_stream_state *streams, +bool dc_link_dp_dpia_validate(struct dc *dc, const struct dc_stream_state *streams, const unsigned int count); /* Sink Interfaces - A sink corresponds to a display output device */ @@ -2342,6 +2342,9 @@ void dc_hardware_release(struct dc *dc); void dc_mclk_switch_using_fw_based_vblank_stretch_shut_down(struct dc *dc); bool dc_set_psr_allow_active(struct dc *dc, bool enable); + +bool dc_set_replay_allow_active(struct dc *dc, bool active); + void dc_z10_restore(const struct dc *dc); void dc_z10_save_init(struct dc *dc); diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c index 1a4d615ccdec..2b79a0e5638e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -33,6 +33,7 @@ #include "cursor_reg_cache.h" #include "resource.h" #include "clk_mgr.h" +#include "dc_state_priv.h" #define CTX dc_dmub_srv->ctx #define DC_LOGGER CTX->logger @@ -140,7 +141,10 @@ bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv, if (status == DMUB_STATUS_QUEUE_FULL) { /* Execute and wait for queue to become empty again. */ - dmub_srv_cmd_execute(dmub); + status = dmub_srv_cmd_execute(dmub); + if (status == DMUB_STATUS_POWER_STATE_D3) + return false; + dmub_srv_wait_for_idle(dmub, 100000); /* Requeue the command. */ @@ -148,16 +152,20 @@ bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv, } if (status != DMUB_STATUS_OK) { - DC_ERROR("Error queueing DMUB command: status=%d\n", status); - dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + if (status != DMUB_STATUS_POWER_STATE_D3) { + DC_ERROR("Error queueing DMUB command: status=%d\n", status); + dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + } return false; } } status = dmub_srv_cmd_execute(dmub); if (status != DMUB_STATUS_OK) { - DC_ERROR("Error starting DMUB execution: status=%d\n", status); - dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + if (status != DMUB_STATUS_POWER_STATE_D3) { + DC_ERROR("Error starting DMUB execution: status=%d\n", status); + dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + } return false; } @@ -218,7 +226,10 @@ bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int coun if (status == DMUB_STATUS_QUEUE_FULL) { /* Execute and wait for queue to become empty again. */ - dmub_srv_cmd_execute(dmub); + status = dmub_srv_cmd_execute(dmub); + if (status == DMUB_STATUS_POWER_STATE_D3) + return false; + dmub_srv_wait_for_idle(dmub, 100000); /* Requeue the command. */ @@ -226,16 +237,20 @@ bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int coun } if (status != DMUB_STATUS_OK) { - DC_ERROR("Error queueing DMUB command: status=%d\n", status); - dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + if (status != DMUB_STATUS_POWER_STATE_D3) { + DC_ERROR("Error queueing DMUB command: status=%d\n", status); + dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + } return false; } } status = dmub_srv_cmd_execute(dmub); if (status != DMUB_STATUS_OK) { - DC_ERROR("Error starting DMUB execution: status=%d\n", status); - dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + if (status != DMUB_STATUS_POWER_STATE_D3) { + DC_ERROR("Error starting DMUB execution: status=%d\n", status); + dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + } return false; } @@ -287,17 +302,11 @@ bool dc_dmub_srv_optimized_init_done(struct dc_dmub_srv *dc_dmub_srv) bool dc_dmub_srv_notify_stream_mask(struct dc_dmub_srv *dc_dmub_srv, unsigned int stream_mask) { - struct dmub_srv *dmub; - const uint32_t timeout = 30; - if (!dc_dmub_srv || !dc_dmub_srv->dmub) return false; - dmub = dc_dmub_srv->dmub; - - return dmub_srv_send_gpint_command( - dmub, DMUB_GPINT__IDLE_OPT_NOTIFY_STREAM_MASK, - stream_mask, timeout) == DMUB_STATUS_OK; + return dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__IDLE_OPT_NOTIFY_STREAM_MASK, + stream_mask, NULL, DM_DMUB_WAIT_TYPE_WAIT); } bool dc_dmub_srv_is_restore_required(struct dc_dmub_srv *dc_dmub_srv) @@ -346,7 +355,7 @@ void dc_dmub_srv_drr_update_cmd(struct dc *dc, uint32_t tg_inst, uint32_t vtotal cmd.drr_update.header.payload_bytes = sizeof(cmd.drr_update) - sizeof(cmd.drr_update.header); // Send the command to the DMCUB. - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst) @@ -360,7 +369,7 @@ void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst) cmd.drr_update.header.payload_bytes = sizeof(cmd.drr_update) - sizeof(cmd.drr_update.header); // Send the command to the DMCUB. - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static uint8_t dc_dmub_srv_get_pipes_for_stream(struct dc *dc, struct dc_stream_state *stream) @@ -453,7 +462,7 @@ bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool should_manage_pstate, stru sizeof(cmd.fw_assisted_mclk_switch) - sizeof(cmd.fw_assisted_mclk_switch.header); // Send the command to the DMCUB. - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -474,7 +483,7 @@ void dc_dmub_srv_query_caps_cmd(struct dc_dmub_srv *dc_dmub_srv) cmd.query_feature_caps.header.payload_bytes = sizeof(struct dmub_cmd_query_feature_caps_data); /* If command was processed, copy feature caps to dmub srv */ - if (dm_execute_dmub_cmd(dc_dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && + if (dc_wake_and_execute_dmub_cmd(dc_dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && cmd.query_feature_caps.header.ret_status == 0) { memcpy(&dc_dmub_srv->dmub->feature_caps, &cmd.query_feature_caps.query_feature_caps_data, @@ -499,7 +508,7 @@ void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pi cmd.visual_confirm_color.visual_confirm_color_data.visual_confirm_color.panel_inst = panel_inst; // If command was processed, copy feature caps to dmub srv - if (dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && + if (dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && cmd.visual_confirm_color.header.ret_status == 0) { memcpy(&dc->ctx->dmub_srv->dmub->visual_confirm_color, &cmd.visual_confirm_color.visual_confirm_color_data, @@ -510,10 +519,11 @@ void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pi /** * populate_subvp_cmd_drr_info - Helper to populate DRR pipe info for the DMCUB subvp command * - * @dc: [in] current dc state + * @dc: [in] pointer to dc object * @subvp_pipe: [in] pipe_ctx for the SubVP pipe * @vblank_pipe: [in] pipe_ctx for the DRR pipe * @pipe_data: [in] Pipe data which stores the VBLANK/DRR info + * @context: [in] DC state for access to phantom stream * * Populate the DMCUB SubVP command with DRR pipe info. All the information * required for calculating the SubVP + DRR microschedule is populated here. @@ -524,12 +534,14 @@ void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pi * 3. Populate the drr_info with the min and max supported vtotal values */ static void populate_subvp_cmd_drr_info(struct dc *dc, + struct dc_state *context, struct pipe_ctx *subvp_pipe, struct pipe_ctx *vblank_pipe, struct dmub_cmd_fw_assisted_mclk_switch_pipe_data_v2 *pipe_data) { + struct dc_stream_state *phantom_stream = dc_state_get_paired_subvp_stream(context, subvp_pipe->stream); struct dc_crtc_timing *main_timing = &subvp_pipe->stream->timing; - struct dc_crtc_timing *phantom_timing = &subvp_pipe->stream->mall_stream_config.paired_stream->timing; + struct dc_crtc_timing *phantom_timing = &phantom_stream->timing; struct dc_crtc_timing *drr_timing = &vblank_pipe->stream->timing; uint16_t drr_frame_us = 0; uint16_t min_drr_supported_us = 0; @@ -617,7 +629,7 @@ static void populate_subvp_cmd_vblank_pipe_info(struct dc *dc, continue; // Find the SubVP pipe - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) + if (dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN) break; } @@ -634,7 +646,7 @@ static void populate_subvp_cmd_vblank_pipe_info(struct dc *dc, if (vblank_pipe->stream->ignore_msa_timing_param && (vblank_pipe->stream->allow_freesync || vblank_pipe->stream->vrr_active_variable || vblank_pipe->stream->vrr_active_fixed)) - populate_subvp_cmd_drr_info(dc, pipe, vblank_pipe, pipe_data); + populate_subvp_cmd_drr_info(dc, context, pipe, vblank_pipe, pipe_data); } /** @@ -659,10 +671,17 @@ static void update_subvp_prefetch_end_to_mall_start(struct dc *dc, uint32_t subvp0_prefetch_us = 0; uint32_t subvp1_prefetch_us = 0; uint32_t prefetch_delta_us = 0; - struct dc_crtc_timing *phantom_timing0 = &subvp_pipes[0]->stream->mall_stream_config.paired_stream->timing; - struct dc_crtc_timing *phantom_timing1 = &subvp_pipes[1]->stream->mall_stream_config.paired_stream->timing; + struct dc_stream_state *phantom_stream0 = NULL; + struct dc_stream_state *phantom_stream1 = NULL; + struct dc_crtc_timing *phantom_timing0 = NULL; + struct dc_crtc_timing *phantom_timing1 = NULL; struct dmub_cmd_fw_assisted_mclk_switch_pipe_data_v2 *pipe_data = NULL; + phantom_stream0 = dc_state_get_paired_subvp_stream(context, subvp_pipes[0]->stream); + phantom_stream1 = dc_state_get_paired_subvp_stream(context, subvp_pipes[1]->stream); + phantom_timing0 = &phantom_stream0->timing; + phantom_timing1 = &phantom_stream1->timing; + subvp0_prefetch_us = div64_u64(((uint64_t)(phantom_timing0->v_total - phantom_timing0->v_front_porch) * (uint64_t)phantom_timing0->h_total * 1000000), (((uint64_t)phantom_timing0->pix_clk_100hz * 100) + dc->caps.subvp_prefetch_end_to_mall_start_us)); @@ -712,8 +731,9 @@ static void populate_subvp_cmd_pipe_info(struct dc *dc, uint32_t j; struct dmub_cmd_fw_assisted_mclk_switch_pipe_data_v2 *pipe_data = &cmd->fw_assisted_mclk_switch_v2.config_data.pipe_data[cmd_pipe_index]; + struct dc_stream_state *phantom_stream = dc_state_get_paired_subvp_stream(context, subvp_pipe->stream); struct dc_crtc_timing *main_timing = &subvp_pipe->stream->timing; - struct dc_crtc_timing *phantom_timing = &subvp_pipe->stream->mall_stream_config.paired_stream->timing; + struct dc_crtc_timing *phantom_timing = &phantom_stream->timing; uint32_t out_num_stream, out_den_stream, out_num_plane, out_den_plane, out_num, out_den; pipe_data->mode = SUBVP; @@ -767,7 +787,7 @@ static void populate_subvp_cmd_pipe_info(struct dc *dc, for (j = 0; j < dc->res_pool->pipe_count; j++) { struct pipe_ctx *phantom_pipe = &context->res_ctx.pipe_ctx[j]; - if (phantom_pipe->stream == subvp_pipe->stream->mall_stream_config.paired_stream) { + if (phantom_pipe->stream == dc_state_get_paired_subvp_stream(context, subvp_pipe->stream)) { pipe_data->pipe_config.subvp_data.phantom_pipe_index = phantom_pipe->stream_res.tg->inst; if (phantom_pipe->bottom_pipe) { pipe_data->pipe_config.subvp_data.phantom_split_pipe_index = phantom_pipe->bottom_pipe->plane_res.hubp->inst; @@ -801,6 +821,7 @@ void dc_dmub_setup_subvp_dmub_command(struct dc *dc, union dmub_rb_cmd cmd; struct pipe_ctx *subvp_pipes[2]; uint32_t wm_val_refclk = 0; + enum mall_stream_type pipe_mall_type; memset(&cmd, 0, sizeof(cmd)); // FW command for SUBVP @@ -816,7 +837,7 @@ void dc_dmub_setup_subvp_dmub_command(struct dc *dc, */ if (resource_is_pipe_type(pipe, OTG_MASTER) && resource_is_pipe_type(pipe, DPP_PIPE) && - pipe->stream->mall_stream_config.type == SUBVP_MAIN) + dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN) subvp_pipes[subvp_count++] = pipe; } @@ -824,6 +845,7 @@ void dc_dmub_setup_subvp_dmub_command(struct dc *dc, // For each pipe that is a "main" SUBVP pipe, fill in pipe data for DMUB SUBVP cmd for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + pipe_mall_type = dc_state_get_pipe_subvp_type(context, pipe); if (!pipe->stream) continue; @@ -834,12 +856,11 @@ void dc_dmub_setup_subvp_dmub_command(struct dc *dc, */ if (resource_is_pipe_type(pipe, OTG_MASTER) && resource_is_pipe_type(pipe, DPP_PIPE) && - pipe->stream->mall_stream_config.paired_stream && - pipe->stream->mall_stream_config.type == SUBVP_MAIN) { + pipe_mall_type == SUBVP_MAIN) { populate_subvp_cmd_pipe_info(dc, context, &cmd, pipe, cmd_pipe_index++); } else if (resource_is_pipe_type(pipe, OTG_MASTER) && resource_is_pipe_type(pipe, DPP_PIPE) && - pipe->stream->mall_stream_config.type == SUBVP_NONE) { + pipe_mall_type == SUBVP_NONE) { // Don't need to check for ActiveDRAMClockChangeMargin < 0, not valid in cases where // we run through DML without calculating "natural" P-state support populate_subvp_cmd_vblank_pipe_info(dc, context, &cmd, pipe, cmd_pipe_index++); @@ -861,7 +882,7 @@ void dc_dmub_setup_subvp_dmub_command(struct dc *dc, cmd.fw_assisted_mclk_switch_v2.config_data.watermark_a_cache = wm_val_refclk < 0xFFFF ? wm_val_refclk : 0xFFFF; } - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } bool dc_dmub_srv_get_diagnostic_data(struct dc_dmub_srv *dc_dmub_srv, struct dmub_diagnostic_data *diag_data) @@ -1098,7 +1119,7 @@ void dc_send_update_cursor_info_to_dmu( pipe_idx, pCtx->plane_res.hubp, pCtx->plane_res.dpp); /* Combine 2nd cmds update_curosr_info to DMU */ - dm_execute_dmub_cmd_list(pCtx->stream->ctx, 2, cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd_list(pCtx->stream->ctx, 2, cmd, DM_DMUB_WAIT_TYPE_WAIT); } } @@ -1112,25 +1133,20 @@ bool dc_dmub_check_min_version(struct dmub_srv *srv) void dc_dmub_srv_enable_dpia_trace(const struct dc *dc) { struct dc_dmub_srv *dc_dmub_srv = dc->ctx->dmub_srv; - struct dmub_srv *dmub; - enum dmub_status status; - static const uint32_t timeout_us = 30; if (!dc_dmub_srv || !dc_dmub_srv->dmub) { DC_LOG_ERROR("%s: invalid parameters.", __func__); return; } - dmub = dc_dmub_srv->dmub; - - status = dmub_srv_send_gpint_command(dmub, DMUB_GPINT__SET_TRACE_BUFFER_MASK_WORD1, 0x0010, timeout_us); - if (status != DMUB_STATUS_OK) { + if (!dc_wake_and_execute_gpint(dc->ctx, DMUB_GPINT__SET_TRACE_BUFFER_MASK_WORD1, + 0x0010, NULL, DM_DMUB_WAIT_TYPE_WAIT)) { DC_LOG_ERROR("timeout updating trace buffer mask word\n"); return; } - status = dmub_srv_send_gpint_command(dmub, DMUB_GPINT__UPDATE_TRACE_BUFFER_MASK, 0x0000, timeout_us); - if (status != DMUB_STATUS_OK) { + if (!dc_wake_and_execute_gpint(dc->ctx, DMUB_GPINT__UPDATE_TRACE_BUFFER_MASK, + 0x0000, NULL, DM_DMUB_WAIT_TYPE_WAIT)) { DC_LOG_ERROR("timeout updating trace buffer mask word\n"); return; } @@ -1148,6 +1164,9 @@ bool dc_dmub_srv_is_hw_pwr_up(struct dc_dmub_srv *dc_dmub_srv, bool wait) struct dc_context *dc_ctx = dc_dmub_srv->ctx; enum dmub_status status; + if (!dc_dmub_srv || !dc_dmub_srv->dmub) + return true; + if (dc_dmub_srv->ctx->dc->debug.dmcub_emulation) return true; @@ -1169,7 +1188,7 @@ bool dc_dmub_srv_is_hw_pwr_up(struct dc_dmub_srv *dc_dmub_srv, bool wait) return true; } -void dc_dmub_srv_notify_idle(const struct dc *dc, bool allow_idle) +static void dc_dmub_srv_notify_idle(const struct dc *dc, bool allow_idle) { union dmub_rb_cmd cmd = {0}; @@ -1190,20 +1209,20 @@ void dc_dmub_srv_notify_idle(const struct dc *dc, bool allow_idle) dc->hwss.set_idle_state(dc, true); } - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + /* NOTE: This does not use the "wake" interface since this is part of the wake path. */ + /* We also do not perform a wait since DMCUB could enter idle after the notification. */ + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); } -void dc_dmub_srv_exit_low_power_state(const struct dc *dc) +static void dc_dmub_srv_exit_low_power_state(const struct dc *dc) { - const uint32_t max_num_polls = 10000; uint32_t allow_state = 0; uint32_t commit_state = 0; - int i; if (dc->debug.dmcub_emulation) return; - if (!dc->idle_optimizations_allowed) + if (!dc->ctx->dmub_srv || !dc->ctx->dmub_srv->dmub) return; if (dc->hwss.get_idle_state && @@ -1215,8 +1234,16 @@ void dc_dmub_srv_exit_low_power_state(const struct dc *dc) if (!(allow_state & DMUB_IPS2_ALLOW_MASK)) { // Wait for evaluation time - udelay(dc->debug.ips2_eval_delay_us); - commit_state = dc->hwss.get_idle_state(dc); + for (;;) { + udelay(dc->debug.ips2_eval_delay_us); + commit_state = dc->hwss.get_idle_state(dc); + if (commit_state & DMUB_IPS2_ALLOW_MASK) + break; + + /* allow was still set, retry eval delay */ + dc->hwss.set_idle_state(dc, false); + } + if (!(commit_state & DMUB_IPS2_COMMIT_MASK)) { // Tell PMFW to exit low power state dc->clk_mgr->funcs->exit_low_power_state(dc->clk_mgr); @@ -1225,17 +1252,13 @@ void dc_dmub_srv_exit_low_power_state(const struct dc *dc) udelay(dc->debug.ips2_entry_delay_us); dc->clk_mgr->funcs->exit_low_power_state(dc->clk_mgr); - for (i = 0; i < max_num_polls; ++i) { + for (;;) { commit_state = dc->hwss.get_idle_state(dc); if (commit_state & DMUB_IPS2_COMMIT_MASK) break; udelay(1); - - if (dc->debug.disable_timeout) - i--; } - ASSERT(i < max_num_polls); if (!dc_dmub_srv_is_hw_pwr_up(dc->ctx->dmub_srv, true)) ASSERT(0); @@ -1250,17 +1273,13 @@ void dc_dmub_srv_exit_low_power_state(const struct dc *dc) dc_dmub_srv_notify_idle(dc, false); if (!(allow_state & DMUB_IPS1_ALLOW_MASK)) { - for (i = 0; i < max_num_polls; ++i) { + for (;;) { commit_state = dc->hwss.get_idle_state(dc); if (commit_state & DMUB_IPS1_COMMIT_MASK) break; udelay(1); - - if (dc->debug.disable_timeout) - i--; } - ASSERT(i < max_num_polls); } } @@ -1282,3 +1301,117 @@ void dc_dmub_srv_set_power_state(struct dc_dmub_srv *dc_dmub_srv, enum dc_acpi_c else dmub_srv_set_power_state(dmub, DMUB_POWER_STATE_D3); } + +void dc_dmub_srv_apply_idle_power_optimizations(const struct dc *dc, bool allow_idle) +{ + struct dc_dmub_srv *dc_dmub_srv = dc->ctx->dmub_srv; + + if (!dc_dmub_srv || !dc_dmub_srv->dmub) + return; + + if (dc_dmub_srv->idle_allowed == allow_idle) + return; + + /* + * Entering a low power state requires a driver notification. + * Powering up the hardware requires notifying PMFW and DMCUB. + * Clearing the driver idle allow requires a DMCUB command. + * DMCUB commands requires the DMCUB to be powered up and restored. + * + * Exit out early to prevent an infinite loop of DMCUB commands + * triggering exit low power - use software state to track this. + */ + dc_dmub_srv->idle_allowed = allow_idle; + + if (!allow_idle) + dc_dmub_srv_exit_low_power_state(dc); + else + dc_dmub_srv_notify_idle(dc, allow_idle); +} + +bool dc_wake_and_execute_dmub_cmd(const struct dc_context *ctx, union dmub_rb_cmd *cmd, + enum dm_dmub_wait_type wait_type) +{ + return dc_wake_and_execute_dmub_cmd_list(ctx, 1, cmd, wait_type); +} + +bool dc_wake_and_execute_dmub_cmd_list(const struct dc_context *ctx, unsigned int count, + union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type) +{ + struct dc_dmub_srv *dc_dmub_srv = ctx->dmub_srv; + bool result = false, reallow_idle = false; + + if (!dc_dmub_srv || !dc_dmub_srv->dmub) + return false; + + if (count == 0) + return true; + + if (dc_dmub_srv->idle_allowed) { + dc_dmub_srv_apply_idle_power_optimizations(ctx->dc, false); + reallow_idle = true; + } + + /* + * These may have different implementations in DM, so ensure + * that we guide it to the expected helper. + */ + if (count > 1) + result = dm_execute_dmub_cmd_list(ctx, count, cmd, wait_type); + else + result = dm_execute_dmub_cmd(ctx, cmd, wait_type); + + if (result && reallow_idle) + dc_dmub_srv_apply_idle_power_optimizations(ctx->dc, true); + + return result; +} + +static bool dc_dmub_execute_gpint(const struct dc_context *ctx, enum dmub_gpint_command command_code, + uint16_t param, uint32_t *response, enum dm_dmub_wait_type wait_type) +{ + struct dc_dmub_srv *dc_dmub_srv = ctx->dmub_srv; + const uint32_t wait_us = wait_type == DM_DMUB_WAIT_TYPE_NO_WAIT ? 0 : 30; + enum dmub_status status; + + if (response) + *response = 0; + + if (!dc_dmub_srv || !dc_dmub_srv->dmub) + return false; + + status = dmub_srv_send_gpint_command(dc_dmub_srv->dmub, command_code, param, wait_us); + if (status != DMUB_STATUS_OK) { + if (status == DMUB_STATUS_TIMEOUT && wait_type == DM_DMUB_WAIT_TYPE_NO_WAIT) + return true; + + return false; + } + + if (response && wait_type == DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) + dmub_srv_get_gpint_response(dc_dmub_srv->dmub, response); + + return true; +} + +bool dc_wake_and_execute_gpint(const struct dc_context *ctx, enum dmub_gpint_command command_code, + uint16_t param, uint32_t *response, enum dm_dmub_wait_type wait_type) +{ + struct dc_dmub_srv *dc_dmub_srv = ctx->dmub_srv; + bool result = false, reallow_idle = false; + + if (!dc_dmub_srv || !dc_dmub_srv->dmub) + return false; + + if (dc_dmub_srv->idle_allowed) { + dc_dmub_srv_apply_idle_power_optimizations(ctx->dc, false); + reallow_idle = true; + } + + result = dc_dmub_execute_gpint(ctx, command_code, param, response, wait_type); + + if (result && reallow_idle) + dc_dmub_srv_apply_idle_power_optimizations(ctx->dc, true); + + return result; +} diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h index c25ce7546f71..952bfb368886 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h @@ -50,6 +50,8 @@ struct dc_dmub_srv { struct dc_context *ctx; void *dm; + + bool idle_allowed; }; void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv); @@ -100,8 +102,59 @@ void dc_dmub_srv_enable_dpia_trace(const struct dc *dc); void dc_dmub_srv_subvp_save_surf_addr(const struct dc_dmub_srv *dc_dmub_srv, const struct dc_plane_address *addr, uint8_t subvp_index); bool dc_dmub_srv_is_hw_pwr_up(struct dc_dmub_srv *dc_dmub_srv, bool wait); -void dc_dmub_srv_notify_idle(const struct dc *dc, bool allow_idle); -void dc_dmub_srv_exit_low_power_state(const struct dc *dc); + +void dc_dmub_srv_apply_idle_power_optimizations(const struct dc *dc, bool allow_idle); void dc_dmub_srv_set_power_state(struct dc_dmub_srv *dc_dmub_srv, enum dc_acpi_cm_power_state powerState); + +/** + * dc_wake_and_execute_dmub_cmd() - Wrapper for DMUB command execution. + * + * Refer to dc_wake_and_execute_dmub_cmd_list() for usage and limitations, + * This function is a convenience wrapper for a single command execution. + * + * @ctx: DC context + * @cmd: The command to send/receive + * @wait_type: The wait behavior for the execution + * + * Return: true on command submission success, false otherwise + */ +bool dc_wake_and_execute_dmub_cmd(const struct dc_context *ctx, union dmub_rb_cmd *cmd, + enum dm_dmub_wait_type wait_type); + +/** + * dc_wake_and_execute_dmub_cmd_list() - Wrapper for DMUB command list execution. + * + * If the DMCUB hardware was asleep then it wakes the DMUB before + * executing the command and attempts to re-enter if the command + * submission was successful. + * + * This should be the preferred command submission interface provided + * the DC lock is acquired. + * + * Entry/exit out of idle power optimizations would need to be + * manually performed otherwise through dc_allow_idle_optimizations(). + * + * @ctx: DC context + * @count: Number of commands to send/receive + * @cmd: Array of commands to send + * @wait_type: The wait behavior for the execution + * + * Return: true on command submission success, false otherwise + */ +bool dc_wake_and_execute_dmub_cmd_list(const struct dc_context *ctx, unsigned int count, + union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type); + +/** + * dc_wake_and_execute_gpint() + * + * @ctx: DC context + * @command_code: The command ID to send to DMCUB + * @param: The parameter to message DMCUB + * @response: Optional response out value - may be NULL. + * @wait_type: The wait behavior for the execution + */ +bool dc_wake_and_execute_gpint(const struct dc_context *ctx, enum dmub_gpint_command command_code, + uint16_t param, uint32_t *response, enum dm_dmub_wait_type wait_type); + #endif /* _DMUB_DC_SRV_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index eeeeeef4d717..1cb7765f593a 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -1377,6 +1377,12 @@ struct dp_trace { #ifndef DP_TUNNELING_STATUS #define DP_TUNNELING_STATUS 0xE0025 /* 1.4a */ #endif +#ifndef DP_TUNNELING_MAX_LINK_RATE +#define DP_TUNNELING_MAX_LINK_RATE 0xE0028 /* 1.4a */ +#endif +#ifndef DP_TUNNELING_MAX_LANE_COUNT +#define DP_TUNNELING_MAX_LANE_COUNT 0xE0029 /* 1.4a */ +#endif #ifndef DPTX_BW_ALLOCATION_MODE_CONTROL #define DPTX_BW_ALLOCATION_MODE_CONTROL 0xE0030 /* 1.4a */ #endif diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c index cb6eaddab720..8f9a67825615 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c @@ -50,7 +50,7 @@ static inline void submit_dmub_read_modify_write( cmd_buf->header.payload_bytes = sizeof(struct dmub_cmd_read_modify_write_sequence) * offload->reg_seq_count; - dm_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT); + dc_wake_and_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT); memset(cmd_buf, 0, sizeof(*cmd_buf)); @@ -67,7 +67,7 @@ static inline void submit_dmub_burst_write( cmd_buf->header.payload_bytes = sizeof(uint32_t) * offload->reg_seq_count; - dm_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT); + dc_wake_and_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT); memset(cmd_buf, 0, sizeof(*cmd_buf)); @@ -80,7 +80,7 @@ static inline void submit_dmub_reg_wait( { struct dmub_rb_cmd_reg_wait *cmd_buf = &offload->cmd_data.reg_wait; - dm_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT); + dc_wake_and_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT); memset(cmd_buf, 0, sizeof(*cmd_buf)); offload->reg_seq_count = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h index e2a3aa8812df..811474f4419b 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -244,7 +244,7 @@ enum pixel_format { #define DC_MAX_DIRTY_RECTS 3 struct dc_flip_addrs { struct dc_plane_address address; - unsigned int flip_timestamp_in_us; + unsigned long long flip_timestamp_in_us; bool flip_immediate; /* TODO: add flip duration for FreeSync */ bool triplebuffer_flips; diff --git a/drivers/gpu/drm/amd/display/dc/dc_plane.h b/drivers/gpu/drm/amd/display/dc/dc_plane.h new file mode 100644 index 000000000000..ef380cae816a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dc_plane.h @@ -0,0 +1,38 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DC_PLANE_H_ +#define _DC_PLANE_H_ + +#include "dc.h" +#include "dc_hw_types.h" + +struct dc_plane_state *dc_create_plane_state(struct dc *dc); +const struct dc_plane_status *dc_plane_get_status( + const struct dc_plane_state *plane_state); +void dc_plane_state_retain(struct dc_plane_state *plane_state); +void dc_plane_state_release(struct dc_plane_state *plane_state); + +#endif /* _DC_PLANE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_plane_priv.h b/drivers/gpu/drm/amd/display/dc/dc_plane_priv.h new file mode 100644 index 000000000000..9ee184c1df00 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dc_plane_priv.h @@ -0,0 +1,34 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DC_PLANE_PRIV_H_ +#define _DC_PLANE_PRIV_H_ + +#include "dc_plane.h" + +void dc_plane_construct(struct dc_context *ctx, struct dc_plane_state *plane_state); +void dc_plane_destruct(struct dc_plane_state *plane_state); + +#endif /* _DC_PLANE_PRIV_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_state.h b/drivers/gpu/drm/amd/display/dc/dc_state.h new file mode 100644 index 000000000000..d167fdbfa8a9 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dc_state.h @@ -0,0 +1,78 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DC_STATE_H_ +#define _DC_STATE_H_ + +#include "dc.h" +#include "inc/core_status.h" + +struct dc_state *dc_state_create(struct dc *dc); +void dc_state_copy(struct dc_state *dst_state, struct dc_state *src_state); +struct dc_state *dc_state_create_copy(struct dc_state *src_state); +void dc_state_copy_current(struct dc *dc, struct dc_state *dst_state); +struct dc_state *dc_state_create_current_copy(struct dc *dc); +void dc_state_construct(struct dc *dc, struct dc_state *state); +void dc_state_destruct(struct dc_state *state); +void dc_state_retain(struct dc_state *state); +void dc_state_release(struct dc_state *state); + +enum dc_status dc_state_add_stream(struct dc *dc, + struct dc_state *state, + struct dc_stream_state *stream); + +enum dc_status dc_state_remove_stream( + struct dc *dc, + struct dc_state *state, + struct dc_stream_state *stream); + +bool dc_state_add_plane( + const struct dc *dc, + struct dc_stream_state *stream, + struct dc_plane_state *plane_state, + struct dc_state *state); + +bool dc_state_remove_plane( + const struct dc *dc, + struct dc_stream_state *stream, + struct dc_plane_state *plane_state, + struct dc_state *state); + +bool dc_state_rem_all_planes_for_stream( + const struct dc *dc, + struct dc_stream_state *stream, + struct dc_state *state); + +bool dc_state_add_all_planes_for_stream( + const struct dc *dc, + struct dc_stream_state *stream, + struct dc_plane_state * const *plane_states, + int plane_count, + struct dc_state *state); + +struct dc_stream_status *dc_state_get_stream_status( + struct dc_state *state, + struct dc_stream_state *stream); +#endif /* _DC_STATE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_state_priv.h b/drivers/gpu/drm/amd/display/dc/dc_state_priv.h new file mode 100644 index 000000000000..c1f44e09a6c1 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dc_state_priv.h @@ -0,0 +1,102 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DC_STATE_PRIV_H_ +#define _DC_STATE_PRIV_H_ + +#include "dc_state.h" +#include "dc_stream.h" + +/* Get the type of the provided resource (none, phantom, main) based on the provided + * context. If the context is unavailable, determine only if phantom or not. + */ +enum mall_stream_type dc_state_get_pipe_subvp_type(const struct dc_state *state, + const struct pipe_ctx *pipe_ctx); +enum mall_stream_type dc_state_get_stream_subvp_type(const struct dc_state *state, + const struct dc_stream_state *stream); + +/* Gets the phantom stream if main is provided, gets the main if phantom is provided.*/ +struct dc_stream_state *dc_state_get_paired_subvp_stream(const struct dc_state *state, + const struct dc_stream_state *stream); + +/* allocate's phantom stream or plane and returns pointer to the object */ +struct dc_stream_state *dc_state_create_phantom_stream(const struct dc *dc, + struct dc_state *state, + struct dc_stream_state *main_stream); +struct dc_plane_state *dc_state_create_phantom_plane(struct dc *dc, + struct dc_state *state, + struct dc_plane_state *main_plane); + +/* deallocate's phantom stream or plane */ +void dc_state_release_phantom_stream(const struct dc *dc, + struct dc_state *state, + struct dc_stream_state *phantom_stream); +void dc_state_release_phantom_plane(const struct dc *dc, + struct dc_state *state, + struct dc_plane_state *phantom_plane); + +/* add/remove phantom stream to context and generate subvp meta data */ +enum dc_status dc_state_add_phantom_stream(struct dc *dc, + struct dc_state *state, + struct dc_stream_state *phantom_stream, + struct dc_stream_state *main_stream); +enum dc_status dc_state_remove_phantom_stream(struct dc *dc, + struct dc_state *state, + struct dc_stream_state *phantom_stream); + +bool dc_state_add_phantom_plane( + const struct dc *dc, + struct dc_stream_state *phantom_stream, + struct dc_plane_state *phantom_plane, + struct dc_state *state); + +bool dc_state_remove_phantom_plane( + const struct dc *dc, + struct dc_stream_state *phantom_stream, + struct dc_plane_state *phantom_plane, + struct dc_state *state); + +bool dc_state_rem_all_phantom_planes_for_stream( + const struct dc *dc, + struct dc_stream_state *phantom_stream, + struct dc_state *state, + bool should_release_planes); + +bool dc_state_add_all_phantom_planes_for_stream( + const struct dc *dc, + struct dc_stream_state *phantom_stream, + struct dc_plane_state * const *phantom_planes, + int plane_count, + struct dc_state *state); + +bool dc_state_remove_phantom_streams_and_planes( + struct dc *dc, + struct dc_state *state); + +void dc_state_release_phantom_streams_and_planes( + struct dc *dc, + struct dc_state *state); + +#endif /* _DC_STATE_PRIV_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index 4ac48c346a33..ee10941caa59 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -38,6 +38,14 @@ struct timing_sync_info { bool master; }; +struct mall_stream_config { + /* MALL stream config to indicate if the stream is phantom or not. + * We will use a phantom stream to indicate that the pipe is phantom. + */ + enum mall_stream_type type; + struct dc_stream_state *paired_stream; // master / slave stream +}; + struct dc_stream_status { int primary_otg_inst; int stream_enc_inst; @@ -50,6 +58,7 @@ struct dc_stream_status { struct timing_sync_info timing_sync_info; struct dc_plane_state *plane_states[MAX_SURFACE_NUM]; bool is_abm_supported; + struct mall_stream_config mall_stream_config; }; enum hubp_dmdata_mode { @@ -147,31 +156,6 @@ struct test_pattern { #define SUBVP_DRR_MARGIN_US 100 // 100us for DRR margin (SubVP + DRR) -enum mall_stream_type { - SUBVP_NONE, // subvp not in use - SUBVP_MAIN, // subvp in use, this stream is main stream - SUBVP_PHANTOM, // subvp in use, this stream is a phantom stream -}; - -struct mall_stream_config { - /* MALL stream config to indicate if the stream is phantom or not. - * We will use a phantom stream to indicate that the pipe is phantom. - */ - enum mall_stream_type type; - struct dc_stream_state *paired_stream; // master / slave stream -}; - -/* Temp struct used to save and restore MALL config - * during validation. - * - * TODO: Move MALL config into dc_state instead of stream struct - * to avoid needing to save/restore. - */ -struct mall_temp_config { - struct mall_stream_config mall_stream_config[MAX_PIPES]; - bool is_phantom_plane[MAX_PIPES]; -}; - struct dc_stream_debug_options { char force_odm_combine_segments; }; @@ -301,7 +285,7 @@ struct dc_stream_state { bool has_non_synchronizable_pclk; bool vblank_synchronized; bool fpo_in_use; - struct mall_stream_config mall_stream_config; + bool is_phantom; }; #define ABM_LEVEL_IMMEDIATE_DISABLE 255 @@ -415,41 +399,6 @@ bool dc_stream_get_scanoutpos(const struct dc_stream_state *stream, uint32_t *h_position, uint32_t *v_position); -enum dc_status dc_add_stream_to_ctx( - struct dc *dc, - struct dc_state *new_ctx, - struct dc_stream_state *stream); - -enum dc_status dc_remove_stream_from_ctx( - struct dc *dc, - struct dc_state *new_ctx, - struct dc_stream_state *stream); - - -bool dc_add_plane_to_context( - const struct dc *dc, - struct dc_stream_state *stream, - struct dc_plane_state *plane_state, - struct dc_state *context); - -bool dc_remove_plane_from_context( - const struct dc *dc, - struct dc_stream_state *stream, - struct dc_plane_state *plane_state, - struct dc_state *context); - -bool dc_rem_all_planes_for_stream( - const struct dc *dc, - struct dc_stream_state *stream, - struct dc_state *context); - -bool dc_add_all_planes_for_stream( - const struct dc *dc, - struct dc_stream_state *stream, - struct dc_plane_state * const *plane_states, - int plane_count, - struct dc_state *context); - bool dc_stream_add_writeback(struct dc *dc, struct dc_stream_state *stream, struct dc_writeback_info *wb_info); @@ -518,9 +467,6 @@ void update_stream_signal(struct dc_stream_state *stream, struct dc_sink *sink); void dc_stream_retain(struct dc_stream_state *dc_stream); void dc_stream_release(struct dc_stream_state *dc_stream); -struct dc_stream_status *dc_stream_get_status_from_state( - struct dc_state *state, - struct dc_stream_state *stream); struct dc_stream_status *dc_stream_get_status( struct dc_stream_state *dc_stream); diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream_priv.h b/drivers/gpu/drm/amd/display/dc/dc_stream_priv.h new file mode 100644 index 000000000000..7476fd52ce2b --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dc_stream_priv.h @@ -0,0 +1,37 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DC_STREAM_PRIV_H_ +#define _DC_STREAM_PRIV_H_ + +#include "dc_stream.h" + +bool dc_stream_construct(struct dc_stream_state *stream, + struct dc_sink *dc_sink_data); +void dc_stream_destruct(struct dc_stream_state *stream); + +void dc_stream_assign_stream_id(struct dc_stream_state *stream); + +#endif // _DC_STREAM_PRIV_H_ diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 7313cfe69498..b08ccb8c68bc 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -1140,25 +1140,34 @@ struct dc_panel_config { } ilr; }; +#define MAX_SINKS_PER_LINK 4 + /* * USB4 DPIA BW ALLOCATION STRUCTS */ struct dc_dpia_bw_alloc { - int sink_verified_bw; // The Verified BW that sink can allocated and use that has been verified already - int sink_allocated_bw; // The Actual Allocated BW that sink currently allocated - int sink_max_bw; // The Max BW that sink can require/support + int remote_sink_req_bw[MAX_SINKS_PER_LINK]; // BW requested by remote sinks + int link_verified_bw; // The Verified BW that link can allocated and use that has been verified already + int link_max_bw; // The Max BW that link can require/support + int allocated_bw; // The Actual Allocated BW for this DPIA int estimated_bw; // The estimated available BW for this DPIA int bw_granularity; // BW Granularity + int dp_overhead; // DP overhead in dp tunneling bool bw_alloc_enabled; // The BW Alloc Mode Support is turned ON for all 3: DP-Tx & Dpia & CM bool response_ready; // Response ready from the CM side + uint8_t nrd_max_lane_count; // Non-reduced max lane count + uint8_t nrd_max_link_rate; // Non-reduced max link rate }; -#define MAX_SINKS_PER_LINK 4 - enum dc_hpd_enable_select { HPD_EN_FOR_ALL_EDP = 0, HPD_EN_FOR_PRIMARY_EDP_ONLY, HPD_EN_FOR_SECONDARY_EDP_ONLY, }; +enum mall_stream_type { + SUBVP_NONE, // subvp not in use + SUBVP_MAIN, // subvp in use, this stream is main stream + SUBVP_PHANTOM, // subvp in use, this stream is a phantom stream +}; #endif /* DC_TYPES_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c index 874b132fe1d7..a6006776333d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c @@ -135,7 +135,7 @@ static void dmcu_set_backlight_level( 0, 1, 80000); } -static void dce_abm_init(struct abm *abm, uint32_t backlight) +static void dce_abm_init(struct abm *abm, uint32_t backlight, uint32_t user_level) { struct dce_abm *abm_dce = TO_DCE_ABM(abm); @@ -162,7 +162,7 @@ static void dce_abm_init(struct abm *abm, uint32_t backlight) BL1_PWM_TARGET_ABM_LEVEL, backlight); REG_UPDATE(BL1_PWM_USER_LEVEL, - BL1_PWM_USER_LEVEL, backlight); + BL1_PWM_USER_LEVEL, user_level); REG_UPDATE_2(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES, ABM1_LS_MIN_PIXEL_VALUE_THRES, 0, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c index 140598f18bbd..f0458b8f00af 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c @@ -782,7 +782,7 @@ static void get_azalia_clock_info_dp( /*audio_dto_module = dpDtoSourceClockInkhz * 10,000; * [khz] ->[100Hz] */ azalia_clock_info->audio_dto_module = - pll_info->dp_dto_source_clock_in_khz * 10; + pll_info->audio_dto_source_clock_in_khz * 10; } void dce_aud_wall_dto_setup( diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c index 5d3f6fa1011e..970644b695cd 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c @@ -975,6 +975,9 @@ static bool dcn31_program_pix_clk( look_up_in_video_optimized_rate_tlb(pix_clk_params->requested_pix_clk_100hz / 10); struct bp_pixel_clock_parameters bp_pc_params = {0}; enum transmitter_color_depth bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_24; + + if (clock_source->ctx->dc->clk_mgr->dp_dto_source_clock_in_khz != 0) + dp_dto_ref_khz = clock_source->ctx->dc->clk_mgr->dp_dto_source_clock_in_khz; // For these signal types Driver to program DP_DTO without calling VBIOS Command table if (dc_is_dp_signal(pix_clk_params->signal_type) || dc_is_virtual_signal(pix_clk_params->signal_type)) { if (e) { @@ -1088,6 +1091,10 @@ static bool get_pixel_clk_frequency_100hz( struct dce110_clk_src *clk_src = TO_DCE110_CLK_SRC(clock_source); unsigned int clock_hz = 0; unsigned int modulo_hz = 0; + unsigned int dp_dto_ref_khz = clock_source->ctx->dc->clk_mgr->dprefclk_khz; + + if (clock_source->ctx->dc->clk_mgr->dp_dto_source_clock_in_khz != 0) + dp_dto_ref_khz = clock_source->ctx->dc->clk_mgr->dp_dto_source_clock_in_khz; if (clock_source->id == CLOCK_SOURCE_ID_DP_DTO) { clock_hz = REG_READ(PHASE[inst]); @@ -1100,7 +1107,7 @@ static bool get_pixel_clk_frequency_100hz( modulo_hz = REG_READ(MODULO[inst]); if (modulo_hz) *pixel_clk_khz = div_u64((uint64_t)clock_hz* - clock_source->ctx->dc->clk_mgr->dprefclk_khz*10, + dp_dto_ref_khz*10, modulo_hz); else *pixel_clk_khz = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c index 8c5e7f858be3..ccc154b0281c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c @@ -57,9 +57,9 @@ static unsigned int abm_feature_support(struct abm *abm, unsigned int panel_inst return ret; } -static void dmub_abm_init_ex(struct abm *abm, uint32_t backlight) +static void dmub_abm_init_ex(struct abm *abm, uint32_t backlight, uint32_t user_level) { - dmub_abm_init(abm, backlight); + dmub_abm_init(abm, backlight, user_level); } static unsigned int dmub_abm_get_current_backlight_ex(struct abm *abm) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c index 42c802afc468..f9d6a181164a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c @@ -76,10 +76,10 @@ static void dmub_abm_enable_fractional_pwm(struct dc_context *dc) cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.panel_mask = panel_mask; cmd.abm_set_pwm_frac.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pwm_frac_data); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } -void dmub_abm_init(struct abm *abm, uint32_t backlight) +void dmub_abm_init(struct abm *abm, uint32_t backlight, uint32_t user_level) { struct dce_abm *dce_abm = TO_DMUB_ABM(abm); @@ -106,7 +106,7 @@ void dmub_abm_init(struct abm *abm, uint32_t backlight) BL1_PWM_TARGET_ABM_LEVEL, backlight); REG_UPDATE(BL1_PWM_USER_LEVEL, - BL1_PWM_USER_LEVEL, backlight); + BL1_PWM_USER_LEVEL, user_level); REG_UPDATE_2(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES, ABM1_LS_MIN_PIXEL_VALUE_THRES, 0, @@ -155,7 +155,7 @@ bool dmub_abm_set_level(struct abm *abm, uint32_t level, uint8_t panel_mask) cmd.abm_set_level.abm_set_level_data.panel_mask = panel_mask; cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_level_data); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -186,7 +186,7 @@ void dmub_abm_init_config(struct abm *abm, cmd.abm_init_config.header.payload_bytes = sizeof(struct dmub_cmd_abm_init_config_data); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } @@ -203,7 +203,7 @@ bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, un cmd.abm_pause.abm_pause_data.panel_mask = panel_mask; cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_pause_data); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -246,7 +246,7 @@ bool dmub_abm_save_restore( cmd.abm_save_restore.header.payload_bytes = sizeof(struct dmub_rb_cmd_abm_save_restore); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); // Copy iramtable data into local structure memcpy((void *)pData, dc->dmub_srv->dmub->scratch_mem_fb.cpu_addr, bytes); @@ -274,7 +274,7 @@ bool dmub_abm_set_pipe(struct abm *abm, cmd.abm_set_pipe.abm_set_pipe_data.ramping_boundary = ramping_boundary; cmd.abm_set_pipe.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pipe_data); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -296,7 +296,7 @@ bool dmub_abm_set_backlight_level(struct abm *abm, cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_inst); cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h index 07ea6c8d414f..761685e5b8c9 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h @@ -30,7 +30,7 @@ struct abm_save_restore; -void dmub_abm_init(struct abm *abm, uint32_t backlight); +void dmub_abm_init(struct abm *abm, uint32_t backlight, uint32_t user_level); bool dmub_abm_set_level(struct abm *abm, uint32_t level, uint8_t panel_mask); unsigned int dmub_abm_get_current_backlight(struct abm *abm); unsigned int dmub_abm_get_target_backlight(struct abm *abm); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c index 2aa0e01a6891..ba1fec3016d5 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c @@ -47,7 +47,7 @@ void dmub_hw_lock_mgr_cmd(struct dc_dmub_srv *dmub_srv, if (!lock) cmd.lock_hw.lock_hw_data.should_release = 1; - dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } void dmub_hw_lock_mgr_inbox0_cmd(struct dc_dmub_srv *dmub_srv, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c index d8009b2dc56a..98a778996e1a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c @@ -48,5 +48,5 @@ void dmub_enable_outbox_notification(struct dc_dmub_srv *dmub_srv) sizeof(cmd.outbox1_enable.header); cmd.outbox1_enable.enable = true; - dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c index 9d4170a356a2..3e243e407bb8 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c @@ -105,23 +105,18 @@ static enum dc_psr_state convert_psr_state(uint32_t raw_state) */ static void dmub_psr_get_state(struct dmub_psr *dmub, enum dc_psr_state *state, uint8_t panel_inst) { - struct dmub_srv *srv = dmub->ctx->dmub_srv->dmub; uint32_t raw_state = 0; uint32_t retry_count = 0; - enum dmub_status status; do { // Send gpint command and wait for ack - status = dmub_srv_send_gpint_command(srv, DMUB_GPINT__GET_PSR_STATE, panel_inst, 30); - - if (status == DMUB_STATUS_OK) { - // GPINT was executed, get response - dmub_srv_get_gpint_response(srv, &raw_state); + if (dc_wake_and_execute_gpint(dmub->ctx, DMUB_GPINT__GET_PSR_STATE, panel_inst, &raw_state, + DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) { *state = convert_psr_state(raw_state); - } else + } else { // Return invalid state when GPINT times out *state = PSR_STATE_INVALID; - + } } while (++retry_count <= 1000 && *state == PSR_STATE_INVALID); // Assert if max retry hit @@ -171,7 +166,7 @@ static bool dmub_psr_set_version(struct dmub_psr *dmub, struct dc_stream_state * cmd.psr_set_version.psr_set_version_data.panel_inst = panel_inst; cmd.psr_set_version.header.payload_bytes = sizeof(struct dmub_cmd_psr_set_version_data); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -199,7 +194,7 @@ static void dmub_psr_enable(struct dmub_psr *dmub, bool enable, bool wait, uint8 cmd.psr_enable.header.payload_bytes = 0; // Send header only - dm_execute_dmub_cmd(dc->dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); /* Below loops 1000 x 500us = 500 ms. * Exit PSR may need to wait 1-2 frames to power up. Timeout after at @@ -248,7 +243,7 @@ static void dmub_psr_set_level(struct dmub_psr *dmub, uint16_t psr_level, uint8_ cmd.psr_set_level.psr_set_level_data.psr_level = psr_level; cmd.psr_set_level.psr_set_level_data.cmd_version = DMUB_CMD_PSR_CONTROL_VERSION_1; cmd.psr_set_level.psr_set_level_data.panel_inst = panel_inst; - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } /* @@ -267,7 +262,7 @@ static void dmub_psr_set_sink_vtotal_in_psr_active(struct dmub_psr *dmub, cmd.psr_set_vtotal.psr_set_vtotal_data.psr_vtotal_idle = psr_vtotal_idle; cmd.psr_set_vtotal.psr_set_vtotal_data.psr_vtotal_su = psr_vtotal_su; - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } /* @@ -286,7 +281,7 @@ static void dmub_psr_set_power_opt(struct dmub_psr *dmub, unsigned int power_opt cmd.psr_set_power_opt.psr_set_power_opt_data.power_opt = power_opt; cmd.psr_set_power_opt.psr_set_power_opt_data.panel_inst = panel_inst; - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } /* @@ -423,7 +418,7 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub, copy_settings_data->relock_delay_frame_cnt = 2; copy_settings_data->dsc_slice_height = psr_context->dsc_slice_height; - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -444,7 +439,7 @@ static void dmub_psr_force_static(struct dmub_psr *dmub, uint8_t panel_inst) cmd.psr_force_static.header.sub_type = DMUB_CMD__PSR_FORCE_STATIC; cmd.psr_enable.header.payload_bytes = 0; - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } /* @@ -452,13 +447,11 @@ static void dmub_psr_force_static(struct dmub_psr *dmub, uint8_t panel_inst) */ static void dmub_psr_get_residency(struct dmub_psr *dmub, uint32_t *residency, uint8_t panel_inst) { - struct dmub_srv *srv = dmub->ctx->dmub_srv->dmub; uint16_t param = (uint16_t)(panel_inst << 8); /* Send gpint command and wait for ack */ - dmub_srv_send_gpint_command(srv, DMUB_GPINT__PSR_RESIDENCY, param, 30); - - dmub_srv_get_gpint_response(srv, residency); + dc_wake_and_execute_gpint(dmub->ctx, DMUB_GPINT__PSR_RESIDENCY, param, residency, + DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY); } static const struct dmub_psr_funcs psr_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c index 28149e53c2a6..38e4797e9476 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c @@ -258,13 +258,97 @@ static void dmub_replay_residency(struct dmub_replay *dmub, uint8_t panel_inst, *residency = 0; } +/** + * Set REPLAY power optimization flags and coasting vtotal. + */ +static void dmub_replay_set_power_opt_and_coasting_vtotal(struct dmub_replay *dmub, + unsigned int power_opt, uint8_t panel_inst, uint16_t coasting_vtotal) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = dmub->ctx; + + memset(&cmd, 0, sizeof(cmd)); + cmd.replay_set_power_opt_and_coasting_vtotal.header.type = DMUB_CMD__REPLAY; + cmd.replay_set_power_opt_and_coasting_vtotal.header.sub_type = + DMUB_CMD__REPLAY_SET_POWER_OPT_AND_COASTING_VTOTAL; + cmd.replay_set_power_opt_and_coasting_vtotal.header.payload_bytes = + sizeof(struct dmub_rb_cmd_replay_set_power_opt_and_coasting_vtotal); + cmd.replay_set_power_opt_and_coasting_vtotal.replay_set_power_opt_data.power_opt = power_opt; + cmd.replay_set_power_opt_and_coasting_vtotal.replay_set_power_opt_data.panel_inst = panel_inst; + cmd.replay_set_power_opt_and_coasting_vtotal.replay_set_coasting_vtotal_data.coasting_vtotal = coasting_vtotal; + + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); +} + +/** + * send Replay general cmd to DMUB. + */ +static void dmub_replay_send_cmd(struct dmub_replay *dmub, + enum replay_FW_Message_type msg, union dmub_replay_cmd_set *cmd_element) +{ + union dmub_rb_cmd cmd; + struct dc_context *ctx = NULL; + + if (dmub == NULL || cmd_element == NULL) + return; + + ctx = dmub->ctx; + if (ctx != NULL) { + + if (msg != Replay_Msg_Not_Support) { + memset(&cmd, 0, sizeof(cmd)); + //Header + cmd.replay_set_timing_sync.header.type = DMUB_CMD__REPLAY; + } else + return; + } else + return; + + switch (msg) { + case Replay_Set_Timing_Sync_Supported: + //Header + cmd.replay_set_timing_sync.header.sub_type = + DMUB_CMD__REPLAY_SET_TIMING_SYNC_SUPPORTED; + cmd.replay_set_timing_sync.header.payload_bytes = + sizeof(struct dmub_rb_cmd_replay_set_timing_sync); + //Cmd Body + cmd.replay_set_timing_sync.replay_set_timing_sync_data.panel_inst = + cmd_element->sync_data.panel_inst; + cmd.replay_set_timing_sync.replay_set_timing_sync_data.timing_sync_supported = + cmd_element->sync_data.timing_sync_supported; + break; + case Replay_Set_Residency_Frameupdate_Timer: + //Header + cmd.replay_set_frameupdate_timer.header.sub_type = + DMUB_CMD__REPLAY_SET_RESIDENCY_FRAMEUPDATE_TIMER; + cmd.replay_set_frameupdate_timer.header.payload_bytes = + sizeof(struct dmub_rb_cmd_replay_set_frameupdate_timer); + //Cmd Body + cmd.replay_set_frameupdate_timer.data.panel_inst = + cmd_element->panel_inst; + cmd.replay_set_frameupdate_timer.data.enable = + cmd_element->timer_data.enable; + cmd.replay_set_frameupdate_timer.data.frameupdate_count = + cmd_element->timer_data.frameupdate_count; + break; + case Replay_Msg_Not_Support: + default: + return; + break; + } + + dc_wake_and_execute_dmub_cmd(ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); +} + static const struct dmub_replay_funcs replay_funcs = { - .replay_copy_settings = dmub_replay_copy_settings, - .replay_enable = dmub_replay_enable, - .replay_get_state = dmub_replay_get_state, - .replay_set_power_opt = dmub_replay_set_power_opt, - .replay_set_coasting_vtotal = dmub_replay_set_coasting_vtotal, - .replay_residency = dmub_replay_residency, + .replay_copy_settings = dmub_replay_copy_settings, + .replay_enable = dmub_replay_enable, + .replay_get_state = dmub_replay_get_state, + .replay_set_power_opt = dmub_replay_set_power_opt, + .replay_set_coasting_vtotal = dmub_replay_set_coasting_vtotal, + .replay_residency = dmub_replay_residency, + .replay_set_power_opt_and_coasting_vtotal = dmub_replay_set_power_opt_and_coasting_vtotal, + .replay_send_cmd = dmub_replay_send_cmd, }; /* diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h index b3ee90a0b8b3..3613aff994d7 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h @@ -51,6 +51,8 @@ struct dmub_replay_funcs { uint8_t panel_inst); void (*replay_residency)(struct dmub_replay *dmub, uint8_t panel_inst, uint32_t *residency, const bool is_start, const bool is_alpm); + void (*replay_set_power_opt_and_coasting_vtotal)(struct dmub_replay *dmub, + unsigned int power_opt, uint8_t panel_inst, uint16_t coasting_vtotal); }; struct dmub_replay *dmub_replay_create(struct dc_context *ctx); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile index 0dd62934a18c..ae6a131be71b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile @@ -22,7 +22,7 @@ # # Makefile for DCN. -DCN10 = dcn10_init.o dcn10_ipp.o \ +DCN10 = dcn10_ipp.o \ dcn10_hw_sequencer_debug.o \ dcn10_dpp.o dcn10_opp.o \ dcn10_hubp.o dcn10_mpc.o \ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile index bd760442ff89..3dae3943b056 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile @@ -2,7 +2,7 @@ # # Makefile for DCN. -DCN20 = dcn20_init.o dcn20_dpp.o dcn20_dpp_cm.o dcn20_hubp.o \ +DCN20 = dcn20_dpp.o dcn20_dpp_cm.o dcn20_hubp.o \ dcn20_mpc.o dcn20_opp.o dcn20_hubbub.o dcn20_mmhubbub.o \ dcn20_stream_encoder.o dcn20_link_encoder.o dcn20_dccg.o \ dcn20_vmid.o dcn20_dwb.o dcn20_dwb_scl.o diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/Makefile b/drivers/gpu/drm/amd/display/dc/dcn201/Makefile index a101e6511555..2b0b4f32e13b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn201/Makefile @@ -1,8 +1,7 @@ # SPDX-License-Identifier: MIT # # Makefile for DCN. -DCN201 = dcn201_init.o \ - dcn201_hubbub.o\ +DCN201 = dcn201_hubbub.o\ dcn201_mpc.o dcn201_hubp.o dcn201_opp.o dcn201_dpp.o \ dcn201_dccg.o dcn201_link_encoder.o diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile index dd1eea7212f4..ca92f5c8e7fb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile @@ -2,7 +2,7 @@ # # Makefile for DCN21. -DCN21 = dcn21_init.o dcn21_hubp.o dcn21_hubbub.o \ +DCN21 = dcn21_hubp.o dcn21_hubbub.o \ dcn21_link_encoder.o dcn21_dccg.o AMD_DAL_DCN21 = $(addprefix $(AMDDALPATH)/dc/dcn21/,$(DCN21)) diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c index 68cad55c72ab..e13d69a22c1c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c @@ -691,7 +691,7 @@ static void dmcub_PLAT_54186_wa(struct hubp *hubp, cmd.PLAT_54186_wa.flip.flip_params.vmid = flip_regs->vmid; PERF_TRACE(); // TODO: remove after performance is stable. - dm_execute_dmub_cmd(hubp->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(hubp->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); PERF_TRACE(); // TODO: remove after performance is stable. } diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/Makefile b/drivers/gpu/drm/amd/display/dc/dcn30/Makefile index cd95f322235e..b5b2aa3b3783 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn30/Makefile @@ -23,9 +23,7 @@ # # -DCN30 := \ - dcn30_init.o \ - dcn30_hubbub.o \ +DCN30 := dcn30_hubbub.o \ dcn30_hubp.o \ dcn30_dpp.o \ dcn30_dccg.o \ diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/Makefile b/drivers/gpu/drm/amd/display/dc/dcn301/Makefile index 090011300dcd..d241f665e40a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn301/Makefile @@ -10,7 +10,7 @@ # # Makefile for dcn30. -DCN301 = dcn301_init.o dcn301_dccg.o \ +DCN301 = dcn301_dccg.o \ dcn301_dio_link_encoder.o dcn301_panel_cntl.o dcn301_hubbub.o AMD_DAL_DCN301 = $(addprefix $(AMDDALPATH)/dc/dcn301/,$(DCN301)) diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/Makefile b/drivers/gpu/drm/amd/display/dc/dcn302/Makefile deleted file mode 100644 index 0fcd03569d74..000000000000 --- a/drivers/gpu/drm/amd/display/dc/dcn302/Makefile +++ /dev/null @@ -1,12 +0,0 @@ -# -# (c) Copyright 2020 Advanced Micro Devices, Inc. All the rights reserved -# -# Authors: AMD -# -# Makefile for dcn302. - -DCN3_02 = dcn302_init.o - -AMD_DAL_DCN3_02 = $(addprefix $(AMDDALPATH)/dc/dcn302/,$(DCN3_02)) - -AMD_DISPLAY_FILES += $(AMD_DAL_DCN3_02) diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/Makefile b/drivers/gpu/drm/amd/display/dc/dcn31/Makefile index 11a2662e58ef..5d93ac16c03a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn31/Makefile @@ -10,7 +10,7 @@ # # Makefile for dcn31. -DCN31 = dcn31_hubbub.o dcn31_init.o dcn31_hubp.o \ +DCN31 = dcn31_hubbub.o dcn31_hubp.o \ dcn31_dccg.o dcn31_dio_link_encoder.o dcn31_panel_cntl.o \ dcn31_apg.o dcn31_hpo_dp_stream_encoder.o dcn31_hpo_dp_link_encoder.o \ dcn31_afmt.o dcn31_vpg.o diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c index 4596f3bac1b4..26be5fee7411 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c @@ -125,7 +125,7 @@ static bool query_dp_alt_from_dmub(struct link_encoder *enc, cmd->query_dp_alt.header.payload_bytes = sizeof(cmd->query_dp_alt.data); cmd->query_dp_alt.data.phy_id = phy_id_from_transmitter(enc10->base.transmitter); - if (!dm_execute_dmub_cmd(enc->ctx, cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + if (!dc_wake_and_execute_dmub_cmd(enc->ctx, cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) return false; return true; @@ -436,7 +436,7 @@ static bool link_dpia_control(struct dc_context *dc_ctx, cmd.dig1_dpia_control.dpia_control = *dpia_control; - dm_execute_dmub_cmd(dc_ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c index d849b1eaa4a5..03248422d6ff 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c @@ -52,7 +52,7 @@ static bool dcn31_query_backlight_info(struct panel_cntl *panel_cntl, union dmub cmd->panel_cntl.header.payload_bytes = sizeof(cmd->panel_cntl.data); cmd->panel_cntl.data.pwrseq_inst = dcn31_panel_cntl->base.pwrseq_inst; - return dm_execute_dmub_cmd(dc_dmub_srv->ctx, cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY); + return dc_wake_and_execute_dmub_cmd(dc_dmub_srv->ctx, cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY); } static uint32_t dcn31_get_16_bit_backlight_from_pwm(struct panel_cntl *panel_cntl) @@ -85,7 +85,7 @@ static uint32_t dcn31_panel_cntl_hw_init(struct panel_cntl *panel_cntl) panel_cntl->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV; cmd.panel_cntl.data.bl_pwm_ref_div2 = panel_cntl->stored_backlight_registers.PANEL_PWRSEQ_REF_DIV2; - if (!dm_execute_dmub_cmd(dc_dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + if (!dc_wake_and_execute_dmub_cmd(dc_dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) return 0; panel_cntl->stored_backlight_registers.BL_PWM_CNTL = cmd.panel_cntl.data.bl_pwm_cntl; diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/Makefile b/drivers/gpu/drm/amd/display/dc/dcn314/Makefile index d5c177346a3b..b134ab05aa71 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn314/Makefile @@ -10,8 +10,7 @@ # # Makefile for dcn314. -DCN314 = dcn314_init.o \ - dcn314_dio_stream_encoder.o dcn314_dccg.o +DCN314 = dcn314_dio_stream_encoder.o dcn314_dccg.o AMD_DAL_DCN314 = $(addprefix $(AMDDALPATH)/dc/dcn314/,$(DCN314)) diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/Makefile b/drivers/gpu/drm/amd/display/dc/dcn32/Makefile index 905b74b53092..5314770fff1c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn32/Makefile @@ -10,7 +10,7 @@ # # Makefile for dcn32. -DCN32 = dcn32_hubbub.o dcn32_init.o dcn32_dccg.o \ +DCN32 = dcn32_hubbub.o dcn32_dccg.o \ dcn32_mmhubbub.o dcn32_dpp.o dcn32_hubp.o dcn32_mpc.o \ dcn32_dio_stream_encoder.o dcn32_dio_link_encoder.o dcn32_resource_helpers.o \ dcn32_hpo_dp_link_encoder.o diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c index e8159a459bce..87760600e154 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c @@ -28,6 +28,7 @@ #include "dcn20/dcn20_resource.h" #include "dml/dcn32/display_mode_vba_util_32.h" #include "dml/dcn32/dcn32_fpu.h" +#include "dc_state_priv.h" static bool is_dual_plane(enum surface_pixel_format format) { @@ -190,7 +191,7 @@ bool dcn32_subvp_in_use(struct dc *dc, for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (pipe->stream && pipe->stream->mall_stream_config.type != SUBVP_NONE) + if (dc_state_get_pipe_subvp_type(context, pipe) != SUBVP_NONE) return true; } return false; @@ -264,18 +265,17 @@ static void override_det_for_subvp(struct dc *dc, struct dc_state *context, uint // Do not override if a stream has multiple planes for (i = 0; i < context->stream_count; i++) { - if (context->stream_status[i].plane_count > 1) { + if (context->stream_status[i].plane_count > 1) return; - } - if (context->streams[i]->mall_stream_config.type != SUBVP_PHANTOM) { + + if (dc_state_get_stream_subvp_type(context, context->streams[i]) != SUBVP_PHANTOM) stream_count++; - } } for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - if (pipe_ctx->stream && pipe_ctx->plane_state && pipe_ctx->stream->mall_stream_config.type != SUBVP_PHANTOM) { + if (pipe_ctx->stream && pipe_ctx->plane_state && dc_state_get_pipe_subvp_type(context, pipe_ctx) != SUBVP_PHANTOM) { if (dcn32_allow_subvp_high_refresh_rate(dc, context, pipe_ctx)) { if (pipe_ctx->stream->timing.v_addressable == 1080 && pipe_ctx->stream->timing.h_addressable == 1920) { @@ -290,7 +290,7 @@ static void override_det_for_subvp(struct dc *dc, struct dc_state *context, uint for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - if (pipe_ctx->stream && pipe_ctx->plane_state && pipe_ctx->stream->mall_stream_config.type != SUBVP_PHANTOM) { + if (pipe_ctx->stream && pipe_ctx->plane_state && dc_state_get_pipe_subvp_type(context, pipe_ctx) != SUBVP_PHANTOM) { if (pipe_ctx->stream->timing.v_addressable == 1080 && pipe_ctx->stream->timing.h_addressable == 1920) { if (pipe_segments[i] > 4) pipe_segments[i] = 4; @@ -337,14 +337,14 @@ void dcn32_determine_det_override(struct dc *dc, for (i = 0; i < context->stream_count; i++) { /* Don't count SubVP streams for DET allocation */ - if (context->streams[i]->mall_stream_config.type != SUBVP_PHANTOM) + if (dc_state_get_stream_subvp_type(context, context->streams[i]) != SUBVP_PHANTOM) stream_count++; } if (stream_count > 0) { stream_segments = 18 / stream_count; for (i = 0; i < context->stream_count; i++) { - if (context->streams[i]->mall_stream_config.type == SUBVP_PHANTOM) + if (dc_state_get_stream_subvp_type(context, context->streams[i]) == SUBVP_PHANTOM) continue; if (context->stream_status[i].plane_count > 0) @@ -430,71 +430,6 @@ void dcn32_set_det_allocations(struct dc *dc, struct dc_state *context, dcn32_determine_det_override(dc, context, pipes); } -/** - * dcn32_save_mall_state(): Save MALL (SubVP) state for fast validation cases - * - * This function saves the MALL (SubVP) case for fast validation cases. For fast validation, - * there are situations where a shallow copy of the dc->current_state is created for the - * validation. In this case we want to save and restore the mall config because we always - * teardown subvp at the beginning of validation (and don't attempt to add it back if it's - * fast validation). If we don't restore the subvp config in cases of fast validation + - * shallow copy of the dc->current_state, the dc->current_state will have a partially - * removed subvp state when we did not intend to remove it. - * - * NOTE: This function ONLY works if the streams are not moved to a different pipe in the - * validation. We don't expect this to happen in fast_validation=1 cases. - * - * @dc: Current DC state - * @context: New DC state to be programmed - * @temp_config: struct used to cache the existing MALL state - * - * Return: void - */ -void dcn32_save_mall_state(struct dc *dc, - struct dc_state *context, - struct mall_temp_config *temp_config) -{ - uint32_t i; - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - - if (pipe->stream) - temp_config->mall_stream_config[i] = pipe->stream->mall_stream_config; - - if (pipe->plane_state) - temp_config->is_phantom_plane[i] = pipe->plane_state->is_phantom; - } -} - -/** - * dcn32_restore_mall_state(): Restore MALL (SubVP) state for fast validation cases - * - * Restore the MALL state based on the previously saved state from dcn32_save_mall_state - * - * @dc: Current DC state - * @context: New DC state to be programmed, restore MALL state into here - * @temp_config: struct that has the cached MALL state - * - * Return: void - */ -void dcn32_restore_mall_state(struct dc *dc, - struct dc_state *context, - struct mall_temp_config *temp_config) -{ - uint32_t i; - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - - if (pipe->stream) - pipe->stream->mall_stream_config = temp_config->mall_stream_config[i]; - - if (pipe->plane_state) - pipe->plane_state->is_phantom = temp_config->is_phantom_plane[i]; - } -} - #define MAX_STRETCHED_V_BLANK 1000 // in micro-seconds (must ensure to match value in FW) /* * Scaling factor for v_blank stretch calculations considering timing in @@ -589,13 +524,14 @@ static int get_refresh_rate(struct dc_stream_state *fpo_candidate_stream) * * Return: Pointer to FPO stream candidate if config can support FPO, otherwise NULL */ -struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch(struct dc *dc, const struct dc_state *context) +struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch(struct dc *dc, struct dc_state *context) { int refresh_rate = 0; const int minimum_refreshrate_supported = 120; struct dc_stream_state *fpo_candidate_stream = NULL; bool is_fpo_vactive = false; uint32_t fpo_vactive_margin_us = 0; + struct dc_stream_status *fpo_stream_status = NULL; if (context == NULL) return NULL; @@ -618,16 +554,28 @@ struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stre DC_FP_START(); dcn32_assign_fpo_vactive_candidate(dc, context, &fpo_candidate_stream); DC_FP_END(); - + if (fpo_candidate_stream) + fpo_stream_status = dc_state_get_stream_status(context, fpo_candidate_stream); DC_FP_START(); is_fpo_vactive = dcn32_find_vactive_pipe(dc, context, dc->debug.fpo_vactive_min_active_margin_us); DC_FP_END(); if (!is_fpo_vactive || dc->debug.disable_fpo_vactive) return NULL; - } else + } else { fpo_candidate_stream = context->streams[0]; + if (fpo_candidate_stream) + fpo_stream_status = dc_state_get_stream_status(context, fpo_candidate_stream); + } - if (!fpo_candidate_stream) + /* In DCN32/321, FPO uses per-pipe P-State force. + * If there's no planes, HUBP is power gated and + * therefore programming UCLK_PSTATE_FORCE does + * nothing (P-State will always be asserted naturally + * on a pipe that has HUBP power gated. Therefore we + * only want to enable FPO if the FPO pipe has both + * a stream and a plane. + */ + if (!fpo_candidate_stream || !fpo_stream_status || fpo_stream_status->plane_count == 0) return NULL; if (fpo_candidate_stream->sink->edid_caps.panel_patch.disable_fams) @@ -716,10 +664,11 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context) for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + enum mall_stream_type pipe_mall_type = dc_state_get_pipe_subvp_type(context, pipe); if (resource_is_pipe_type(pipe, OPP_HEAD) && resource_is_pipe_type(pipe, DPP_PIPE)) { - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) { + if (pipe_mall_type == SUBVP_MAIN) { subvp_count++; subvp_disallow |= disallow_subvp_in_active_plus_blank(pipe); @@ -728,7 +677,7 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context) refresh_rate = div_u64(refresh_rate, pipe->stream->timing.v_total); refresh_rate = div_u64(refresh_rate, pipe->stream->timing.h_total); } - if (pipe->stream->mall_stream_config.type == SUBVP_NONE) { + if (pipe_mall_type == SUBVP_NONE) { non_subvp_pipes++; drr_psr_capable = (drr_psr_capable || dcn32_is_psr_capable(pipe)); if (pipe->stream->ignore_msa_timing_param && @@ -776,10 +725,11 @@ bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + enum mall_stream_type pipe_mall_type = dc_state_get_pipe_subvp_type(context, pipe); if (resource_is_pipe_type(pipe, OPP_HEAD) && resource_is_pipe_type(pipe, DPP_PIPE)) { - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) { + if (pipe_mall_type == SUBVP_MAIN) { subvp_count++; subvp_disallow |= disallow_subvp_in_active_plus_blank(pipe); @@ -788,7 +738,7 @@ bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int refresh_rate = div_u64(refresh_rate, pipe->stream->timing.v_total); refresh_rate = div_u64(refresh_rate, pipe->stream->timing.h_total); } - if (pipe->stream->mall_stream_config.type == SUBVP_NONE) { + if (pipe_mall_type == SUBVP_NONE) { non_subvp_pipes++; vblank_psr_capable = (vblank_psr_capable || dcn32_is_psr_capable(pipe)); if (pipe->stream->ignore_msa_timing_param && diff --git a/drivers/gpu/drm/amd/display/dc/dcn35/Makefile b/drivers/gpu/drm/amd/display/dc/dcn35/Makefile index fa7ec82ae5f5..0e317e0c36a0 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn35/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn35/Makefile @@ -10,7 +10,7 @@ # # Makefile for DCN35. -DCN35 = dcn35_init.o dcn35_dio_stream_encoder.o \ +DCN35 = dcn35_dio_stream_encoder.o \ dcn35_dio_link_encoder.o dcn35_dccg.o \ dcn35_hubp.o dcn35_hubbub.o \ dcn35_mmhubbub.o dcn35_opp.o dcn35_dpp.o dcn35_pg_cntl.o dcn35_dwb.o diff --git a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_dio_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_dio_link_encoder.c index f91e08895275..da94e5309fba 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_dio_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_dio_link_encoder.c @@ -256,6 +256,10 @@ void dcn35_link_encoder_construct( enc10->base.features.flags.bits.IS_UHBR10_CAPABLE = bp_cap_info.DP_UHBR10_EN; enc10->base.features.flags.bits.IS_UHBR13_5_CAPABLE = bp_cap_info.DP_UHBR13_5_EN; enc10->base.features.flags.bits.IS_UHBR20_CAPABLE = bp_cap_info.DP_UHBR20_EN; + if (bp_cap_info.DP_IS_USB_C) { + /*BIOS not switch to use CONNECTOR_ID_USBC = 24 yet*/ + enc10->base.features.flags.bits.DP_IS_USB_C = 1; + } } else { DC_LOG_WARNING("%s: Failed to get encoder_cap_info from VBIOS with error code %d!\n", @@ -264,4 +268,5 @@ void dcn35_link_encoder_construct( } if (enc10->base.ctx->dc->debug.hdmi20_disable) enc10->base.features.flags.bits.HDMI_6GB_EN = 0; + } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c index ec77b2b41ba3..38ab9ad60ef8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c @@ -33,6 +33,7 @@ #include "link.h" #include "dcn20_fpu.h" +#include "dc_state_priv.h" #define DC_LOGGER \ dc->ctx->logger @@ -440,7 +441,115 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv14_soc = { .use_urgent_burst_bw = 0 }; -struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv12_soc = { 0 }; +struct _vcs_dpi_soc_bounding_box_st dcn2_0_nv12_soc = { + .clock_limits = { + { + .state = 0, + .dcfclk_mhz = 560.0, + .fabricclk_mhz = 560.0, + .dispclk_mhz = 513.0, + .dppclk_mhz = 513.0, + .phyclk_mhz = 540.0, + .socclk_mhz = 560.0, + .dscclk_mhz = 171.0, + .dram_speed_mts = 1069.0, + }, + { + .state = 1, + .dcfclk_mhz = 694.0, + .fabricclk_mhz = 694.0, + .dispclk_mhz = 642.0, + .dppclk_mhz = 642.0, + .phyclk_mhz = 600.0, + .socclk_mhz = 694.0, + .dscclk_mhz = 214.0, + .dram_speed_mts = 1324.0, + }, + { + .state = 2, + .dcfclk_mhz = 875.0, + .fabricclk_mhz = 875.0, + .dispclk_mhz = 734.0, + .dppclk_mhz = 734.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 875.0, + .dscclk_mhz = 245.0, + .dram_speed_mts = 1670.0, + }, + { + .state = 3, + .dcfclk_mhz = 1000.0, + .fabricclk_mhz = 1000.0, + .dispclk_mhz = 1100.0, + .dppclk_mhz = 1100.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1000.0, + .dscclk_mhz = 367.0, + .dram_speed_mts = 2000.0, + }, + { + .state = 4, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 2000.0, + }, + { + .state = 5, + .dcfclk_mhz = 1200.0, + .fabricclk_mhz = 1200.0, + .dispclk_mhz = 1284.0, + .dppclk_mhz = 1284.0, + .phyclk_mhz = 810.0, + .socclk_mhz = 1200.0, + .dscclk_mhz = 428.0, + .dram_speed_mts = 2000.0, + }, + }, + + .num_states = 5, + .sr_exit_time_us = 1.9, + .sr_enter_plus_exit_time_us = 4.4, + .urgent_latency_us = 3.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 40.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 40.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .ideal_dram_bw_after_urgent_percent = 40.0, + .max_request_size_bytes = 256, + .dram_channel_width_bytes = 16, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.5, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 131, + .urgent_out_of_order_return_per_channel_bytes = 4096, + .channel_interleave_bytes = 256, + .num_banks = 8, + .num_chans = 16, + .vmm_page_size_bytes = 4096, + .dram_clock_change_latency_us = 45.0, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3850, + .xfc_bus_transport_time_us = 20, + .xfc_xbuf_latency_tolerance_us = 50, + .use_urgent_burst_bw = 0, +}; struct _vcs_dpi_ip_params_st dcn2_1_ip = { .odm_capable = 1, @@ -1074,7 +1183,7 @@ void dcn20_calculate_dlg_params(struct dc *dc, pipes[pipe_idx].pipe.dest.vupdate_width = get_vupdate_width(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); pipes[pipe_idx].pipe.dest.vready_offset = get_vready_offset(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); - if (context->res_ctx.pipe_ctx[i].stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (dc_state_get_pipe_subvp_type(context, &context->res_ctx.pipe_ctx[i]) == SUBVP_PHANTOM) { // Phantom pipe requires that DET_SIZE = 0 and no unbounded requests context->res_ctx.pipe_ctx[i].det_buffer_size_kb = 0; context->res_ctx.pipe_ctx[i].unbounded_req = false; @@ -1424,7 +1533,7 @@ int dcn20_populate_dml_pipes_from_context(struct dc *dc, */ if (res_ctx->pipe_ctx[i].plane_state && (res_ctx->pipe_ctx[i].plane_state->address.type == PLN_ADDR_TYPE_VIDEO_PROGRESSIVE || - res_ctx->pipe_ctx[i].stream->mall_stream_config.type == SUBVP_PHANTOM)) + dc_state_get_pipe_subvp_type(context, &res_ctx->pipe_ctx[i]) == SUBVP_PHANTOM)) pipes[pipe_cnt].pipe.src.num_cursors = 0; else pipes[pipe_cnt].pipe.src.num_cursors = dc->dml.ip.number_of_cursors; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c index de209ca0cf8c..9f37f717a1f8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c @@ -32,6 +32,7 @@ #include "clk_mgr/dcn32/dcn32_smu13_driver_if.h" #include "dcn30/dcn30_resource.h" #include "link.h" +#include "dc_state_priv.h" #define DC_LOGGER_INIT(logger) @@ -341,7 +342,7 @@ void dcn32_helper_populate_phantom_dlg_params(struct dc *dc, if (!pipe->stream) continue; - if (pipe->plane_state && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (pipe->plane_state && dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) { pipes[pipe_idx].pipe.dest.vstartup_start = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); pipes[pipe_idx].pipe.dest.vupdate_offset = @@ -624,7 +625,7 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc, if (pipe->plane_state && !pipe->top_pipe && !dcn32_is_center_timing(pipe) && !(pipe->stream->timing.pix_clk_100hz / 10000 > DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ) && (!dcn32_is_psr_capable(pipe) || (context->stream_count == 1 && dc->caps.dmub_caps.subvp_psr)) && - pipe->stream->mall_stream_config.type == SUBVP_NONE && + dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_NONE && (refresh_rate < 120 || dcn32_allow_subvp_high_refresh_rate(dc, context, pipe)) && !pipe->plane_state->address.tmz_surface && (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] <= 0 || @@ -682,7 +683,7 @@ static bool dcn32_enough_pipes_for_subvp(struct dc *dc, struct dc_state *context // Find the minimum pipe split count for non SubVP pipes if (resource_is_pipe_type(pipe, OPP_HEAD) && - pipe->stream->mall_stream_config.type == SUBVP_NONE) { + dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_NONE) { split_cnt = 0; while (pipe) { split_cnt++; @@ -735,8 +736,8 @@ static bool subvp_subvp_schedulable(struct dc *dc, struct dc_state *context) * and also to store the two main SubVP pipe pointers in subvp_pipes[2]. */ if (pipe->stream && pipe->plane_state && !pipe->top_pipe && - pipe->stream->mall_stream_config.type == SUBVP_MAIN) { - phantom = pipe->stream->mall_stream_config.paired_stream; + dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN) { + phantom = dc_state_get_paired_subvp_stream(context, pipe->stream); microschedule_lines = (phantom->timing.v_total - phantom->timing.v_front_porch) + phantom->timing.v_addressable; @@ -804,6 +805,9 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context) int16_t stretched_drr_us = 0; int16_t drr_stretched_vblank_us = 0; int16_t max_vblank_mallregion = 0; + struct dc_stream_state *phantom_stream; + bool subvp_found = false; + bool drr_found = false; // Find SubVP pipe for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -816,8 +820,10 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context) continue; // Find the SubVP pipe - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) + if (dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN) { + subvp_found = true; break; + } } // Find the DRR pipe @@ -825,32 +831,37 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context) drr_pipe = &context->res_ctx.pipe_ctx[i]; // We check for master pipe only - if (!resource_is_pipe_type(pipe, OTG_MASTER) || - !resource_is_pipe_type(pipe, DPP_PIPE)) + if (!resource_is_pipe_type(drr_pipe, OTG_MASTER) || + !resource_is_pipe_type(drr_pipe, DPP_PIPE)) continue; - if (drr_pipe->stream->mall_stream_config.type == SUBVP_NONE && drr_pipe->stream->ignore_msa_timing_param && - (drr_pipe->stream->allow_freesync || drr_pipe->stream->vrr_active_variable || drr_pipe->stream->vrr_active_fixed)) + if (dc_state_get_pipe_subvp_type(context, drr_pipe) == SUBVP_NONE && drr_pipe->stream->ignore_msa_timing_param && + (drr_pipe->stream->allow_freesync || drr_pipe->stream->vrr_active_variable || drr_pipe->stream->vrr_active_fixed)) { + drr_found = true; break; + } } - main_timing = &pipe->stream->timing; - phantom_timing = &pipe->stream->mall_stream_config.paired_stream->timing; - drr_timing = &drr_pipe->stream->timing; - prefetch_us = (phantom_timing->v_total - phantom_timing->v_front_porch) * phantom_timing->h_total / - (double)(phantom_timing->pix_clk_100hz * 100) * 1000000 + - dc->caps.subvp_prefetch_end_to_mall_start_us; - subvp_active_us = main_timing->v_addressable * main_timing->h_total / - (double)(main_timing->pix_clk_100hz * 100) * 1000000; - drr_frame_us = drr_timing->v_total * drr_timing->h_total / - (double)(drr_timing->pix_clk_100hz * 100) * 1000000; - // P-State allow width and FW delays already included phantom_timing->v_addressable - mall_region_us = phantom_timing->v_addressable * phantom_timing->h_total / - (double)(phantom_timing->pix_clk_100hz * 100) * 1000000; - stretched_drr_us = drr_frame_us + mall_region_us + SUBVP_DRR_MARGIN_US; - drr_stretched_vblank_us = (drr_timing->v_total - drr_timing->v_addressable) * drr_timing->h_total / - (double)(drr_timing->pix_clk_100hz * 100) * 1000000 + (stretched_drr_us - drr_frame_us); - max_vblank_mallregion = drr_stretched_vblank_us > mall_region_us ? drr_stretched_vblank_us : mall_region_us; + if (subvp_found && drr_found) { + phantom_stream = dc_state_get_paired_subvp_stream(context, pipe->stream); + main_timing = &pipe->stream->timing; + phantom_timing = &phantom_stream->timing; + drr_timing = &drr_pipe->stream->timing; + prefetch_us = (phantom_timing->v_total - phantom_timing->v_front_porch) * phantom_timing->h_total / + (double)(phantom_timing->pix_clk_100hz * 100) * 1000000 + + dc->caps.subvp_prefetch_end_to_mall_start_us; + subvp_active_us = main_timing->v_addressable * main_timing->h_total / + (double)(main_timing->pix_clk_100hz * 100) * 1000000; + drr_frame_us = drr_timing->v_total * drr_timing->h_total / + (double)(drr_timing->pix_clk_100hz * 100) * 1000000; + // P-State allow width and FW delays already included phantom_timing->v_addressable + mall_region_us = phantom_timing->v_addressable * phantom_timing->h_total / + (double)(phantom_timing->pix_clk_100hz * 100) * 1000000; + stretched_drr_us = drr_frame_us + mall_region_us + SUBVP_DRR_MARGIN_US; + drr_stretched_vblank_us = (drr_timing->v_total - drr_timing->v_addressable) * drr_timing->h_total / + (double)(drr_timing->pix_clk_100hz * 100) * 1000000 + (stretched_drr_us - drr_frame_us); + max_vblank_mallregion = drr_stretched_vblank_us > mall_region_us ? drr_stretched_vblank_us : mall_region_us; + } /* We consider SubVP + DRR schedulable if the stretched frame duration of the DRR display (i.e. the * highest refresh rate + margin that can support UCLK P-State switch) passes the static analysis @@ -895,6 +906,8 @@ static bool subvp_vblank_schedulable(struct dc *dc, struct dc_state *context) struct dc_crtc_timing *main_timing = NULL; struct dc_crtc_timing *phantom_timing = NULL; struct dc_crtc_timing *vblank_timing = NULL; + struct dc_stream_state *phantom_stream; + enum mall_stream_type pipe_mall_type; /* For SubVP + VBLANK/DRR cases, we assume there can only be * a single VBLANK/DRR display. If DML outputs SubVP + VBLANK @@ -904,6 +917,7 @@ static bool subvp_vblank_schedulable(struct dc *dc, struct dc_state *context) */ for (i = 0; i < dc->res_pool->pipe_count; i++) { pipe = &context->res_ctx.pipe_ctx[i]; + pipe_mall_type = dc_state_get_pipe_subvp_type(context, pipe); // We check for master pipe, but it shouldn't matter since we only need // the pipe for timing info (stream should be same for any pipe splits) @@ -911,18 +925,19 @@ static bool subvp_vblank_schedulable(struct dc *dc, struct dc_state *context) !resource_is_pipe_type(pipe, DPP_PIPE)) continue; - if (!found && pipe->stream->mall_stream_config.type == SUBVP_NONE) { + if (!found && pipe_mall_type == SUBVP_NONE) { // Found pipe which is not SubVP or Phantom (i.e. the VBLANK pipe). vblank_index = i; found = true; } - if (!subvp_pipe && pipe->stream->mall_stream_config.type == SUBVP_MAIN) + if (!subvp_pipe && pipe_mall_type == SUBVP_MAIN) subvp_pipe = pipe; } if (found) { + phantom_stream = dc_state_get_paired_subvp_stream(context, subvp_pipe->stream); main_timing = &subvp_pipe->stream->timing; - phantom_timing = &subvp_pipe->stream->mall_stream_config.paired_stream->timing; + phantom_timing = &phantom_stream->timing; vblank_timing = &context->res_ctx.pipe_ctx[vblank_index].stream->timing; // Prefetch time is equal to VACTIVE + BP + VSYNC of the phantom pipe // Also include the prefetch end to mallstart delay time @@ -977,7 +992,7 @@ static bool subvp_subvp_admissable(struct dc *dc, continue; if (pipe->plane_state && !pipe->top_pipe && - pipe->stream->mall_stream_config.type == SUBVP_MAIN) { + dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN) { refresh_rate = (pipe->stream->timing.pix_clk_100hz * (uint64_t)100 + pipe->stream->timing.v_total * pipe->stream->timing.h_total - (uint64_t)1); refresh_rate = div_u64(refresh_rate, pipe->stream->timing.v_total); @@ -1026,23 +1041,23 @@ static bool subvp_validate_static_schedulability(struct dc *dc, for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + enum mall_stream_type pipe_mall_type = dc_state_get_pipe_subvp_type(context, pipe); if (!pipe->stream) continue; if (pipe->plane_state && !pipe->top_pipe) { - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) + if (pipe_mall_type == SUBVP_MAIN) subvp_count++; - if (pipe->stream->mall_stream_config.type == SUBVP_NONE) { + if (pipe_mall_type == SUBVP_NONE) non_subvp_pipes++; - } } // Count how many planes that aren't SubVP/phantom are capable of VACTIVE // switching (SubVP + VACTIVE unsupported). In situations where we force // SubVP for a VACTIVE plane, we don't want to increment the vactive_count. if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vlevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] > 0 && - pipe->stream->mall_stream_config.type == SUBVP_NONE) { + pipe_mall_type == SUBVP_NONE) { vactive_count++; } pipe_idx++; @@ -1078,7 +1093,7 @@ static void assign_subvp_index(struct dc *dc, struct dc_state *context) struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; if (resource_is_pipe_type(pipe_ctx, OTG_MASTER) && - pipe_ctx->stream->mall_stream_config.type == SUBVP_MAIN) { + dc_state_get_pipe_subvp_type(context, pipe_ctx) == SUBVP_MAIN) { pipe_ctx->subvp_index = index++; } else { pipe_ctx->subvp_index = 0; @@ -1532,7 +1547,8 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, // If SubVP pipe config is unsupported (or cannot be used for UCLK switching) // remove phantom pipes and repopulate dml pipes if (!found_supported_config) { - dc->res_pool->funcs->remove_phantom_pipes(dc, context, false); + dc_state_remove_phantom_streams_and_planes(dc, context); + dc_state_release_phantom_streams_and_planes(dc, context); vba->DRAMClockChangeSupport[*vlevel][vba->maxMpcComb] = dm_dram_clock_change_unsupported; *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, false); @@ -1684,7 +1700,7 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context, pipes[pipe_idx].pipe.dest.vready_offset = get_vready_offset(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); - if (context->res_ctx.pipe_ctx[i].stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (dc_state_get_pipe_subvp_type(context, &context->res_ctx.pipe_ctx[i]) == SUBVP_PHANTOM) { // Phantom pipe requires that DET_SIZE = 0 and no unbounded requests context->res_ctx.pipe_ctx[i].det_buffer_size_kb = 0; context->res_ctx.pipe_ctx[i].unbounded_req = false; @@ -1716,7 +1732,7 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context, context->res_ctx.pipe_ctx[i].plane_state != context->res_ctx.pipe_ctx[i].top_pipe->plane_state) && context->res_ctx.pipe_ctx[i].prev_odm_pipe == NULL) { /* SS: all active surfaces stored in MALL */ - if (context->res_ctx.pipe_ctx[i].stream->mall_stream_config.type != SUBVP_PHANTOM) { + if (dc_state_get_pipe_subvp_type(context, &context->res_ctx.pipe_ctx[i]) != SUBVP_PHANTOM) { context->bw_ctx.bw.dcn.mall_ss_size_bytes += context->res_ctx.pipe_ctx[i].surface_size_in_mall_bytes; if (context->res_ctx.pipe_ctx[i].stream->link->psr_settings.psr_version == DC_PSR_VERSION_UNSUPPORTED) { @@ -1930,7 +1946,8 @@ bool dcn32_internal_validate_bw(struct dc *dc, return false; // For each full update, remove all existing phantom pipes first - dc->res_pool->funcs->remove_phantom_pipes(dc, context, fast_validate); + dc_state_remove_phantom_streams_and_planes(dc, context); + dc_state_release_phantom_streams_and_planes(dc, context); dc->res_pool->funcs->update_soc_for_wm_a(dc, context); @@ -3448,7 +3465,15 @@ void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *co for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { const struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (!pipe->stream) + /* In DCN32/321, FPO uses per-pipe P-State force. + * If there's no planes, HUBP is power gated and + * therefore programming UCLK_PSTATE_FORCE does + * nothing (P-State will always be asserted naturally + * on a pipe that has HUBP power gated. Therefore we + * only want to enable FPO if the FPO pipe has both + * a stream and a plane. + */ + if (!pipe->stream || !pipe->plane_state) continue; if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] <= 0) { diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c index 3d12dabd39e4..475c4ec43c01 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c @@ -166,9 +166,9 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_5_soc = { .num_states = 5, .sr_exit_time_us = 14.0, .sr_enter_plus_exit_time_us = 16.0, - .sr_exit_z8_time_us = 525.0, - .sr_enter_plus_exit_z8_time_us = 715.0, - .fclk_change_latency_us = 20.0, + .sr_exit_z8_time_us = 210.0, + .sr_enter_plus_exit_z8_time_us = 320.0, + .fclk_change_latency_us = 24.0, .usr_retraining_latency_us = 2, .writeback_latency_us = 12.0, diff --git a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c index 180f8a98a361..9be5ebf3a8c0 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c @@ -5420,7 +5420,7 @@ static void CalculateOutputLink( *OutBpp = TruncToValidBPP((1 - Downspreading / 100) * 13500, OutputLinkDPLanes, HTotal, HActive, PixelClockBackEnd, ForcedOutputLinkBPP, LinkDSCEnable, Output, OutputFormat, DSCInputBitPerComponent, NumberOfDSCSlices, (dml_uint_t)AudioSampleRate, AudioSampleLayout, ODMModeNoDSC, ODMModeDSC, RequiredSlots); - if (OutBpp == 0 && PHYCLKD32PerState < 20000 / 32 && DSCEnable == dml_dsc_enable_if_necessary && ForcedOutputLinkBPP == 0) { + if (*OutBpp == 0 && PHYCLKD32PerState < 20000 / 32 && DSCEnable == dml_dsc_enable_if_necessary && ForcedOutputLinkBPP == 0) { *RequiresDSC = true; LinkDSCEnable = true; *OutBpp = TruncToValidBPP((1 - Downspreading / 100) * 13500, OutputLinkDPLanes, HTotal, HActive, PixelClockBackEnd, ForcedOutputLinkBPP, LinkDSCEnable, Output, @@ -6229,7 +6229,7 @@ static void set_calculate_prefetch_schedule_params(struct display_mode_lib_st *m CalculatePrefetchSchedule_params->GPUVMEnable = mode_lib->ms.cache_display_cfg.plane.GPUVMEnable; CalculatePrefetchSchedule_params->HostVMEnable = mode_lib->ms.cache_display_cfg.plane.HostVMEnable; CalculatePrefetchSchedule_params->HostVMMaxNonCachedPageTableLevels = mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels; - CalculatePrefetchSchedule_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes; + CalculatePrefetchSchedule_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024; CalculatePrefetchSchedule_params->DynamicMetadataEnable = mode_lib->ms.cache_display_cfg.plane.DynamicMetadataEnable[k]; CalculatePrefetchSchedule_params->DynamicMetadataVMEnabled = mode_lib->ms.ip.dynamic_metadata_vm_enabled; CalculatePrefetchSchedule_params->DynamicMetadataLinesBeforeActiveRequired = mode_lib->ms.cache_display_cfg.plane.DynamicMetadataLinesBeforeActiveRequired[k]; @@ -6329,7 +6329,7 @@ static void dml_prefetch_check(struct display_mode_lib_st *mode_lib) mode_lib->ms.NoOfDPPThisState, mode_lib->ms.dpte_group_bytes, s->HostVMInefficiencyFactor, - mode_lib->ms.soc.hostvm_min_page_size_kbytes, + mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024, mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels); s->NextMaxVStartup = s->MaxVStartupAllPlanes[j]; @@ -6542,7 +6542,7 @@ static void dml_prefetch_check(struct display_mode_lib_st *mode_lib) mode_lib->ms.cache_display_cfg.plane.HostVMEnable, mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels, mode_lib->ms.cache_display_cfg.plane.GPUVMEnable, - mode_lib->ms.soc.hostvm_min_page_size_kbytes, + mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024, mode_lib->ms.PDEAndMetaPTEBytesPerFrame[j][k], mode_lib->ms.MetaRowBytes[j][k], mode_lib->ms.DPTEBytesPerRow[j][k], @@ -7687,7 +7687,7 @@ dml_bool_t dml_core_mode_support(struct display_mode_lib_st *mode_lib) CalculateVMRowAndSwath_params->HostVMMaxNonCachedPageTableLevels = mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels; CalculateVMRowAndSwath_params->GPUVMMaxPageTableLevels = mode_lib->ms.cache_display_cfg.plane.GPUVMMaxPageTableLevels; CalculateVMRowAndSwath_params->GPUVMMinPageSizeKBytes = mode_lib->ms.cache_display_cfg.plane.GPUVMMinPageSizeKBytes; - CalculateVMRowAndSwath_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes; + CalculateVMRowAndSwath_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024; CalculateVMRowAndSwath_params->PTEBufferModeOverrideEn = mode_lib->ms.cache_display_cfg.plane.PTEBufferModeOverrideEn; CalculateVMRowAndSwath_params->PTEBufferModeOverrideVal = mode_lib->ms.cache_display_cfg.plane.PTEBufferMode; CalculateVMRowAndSwath_params->PTEBufferSizeNotExceeded = mode_lib->ms.PTEBufferSizeNotExceededPerState; @@ -7957,7 +7957,7 @@ dml_bool_t dml_core_mode_support(struct display_mode_lib_st *mode_lib) UseMinimumDCFCLK_params->GPUVMMaxPageTableLevels = mode_lib->ms.cache_display_cfg.plane.GPUVMMaxPageTableLevels; UseMinimumDCFCLK_params->HostVMEnable = mode_lib->ms.cache_display_cfg.plane.HostVMEnable; UseMinimumDCFCLK_params->NumberOfActiveSurfaces = mode_lib->ms.num_active_planes; - UseMinimumDCFCLK_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes; + UseMinimumDCFCLK_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024; UseMinimumDCFCLK_params->HostVMMaxNonCachedPageTableLevels = mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels; UseMinimumDCFCLK_params->DynamicMetadataVMEnabled = mode_lib->ms.ip.dynamic_metadata_vm_enabled; UseMinimumDCFCLK_params->ImmediateFlipRequirement = s->ImmediateFlipRequiredFinal; @@ -8699,7 +8699,7 @@ void dml_core_mode_programming(struct display_mode_lib_st *mode_lib, const struc CalculateVMRowAndSwath_params->HostVMMaxNonCachedPageTableLevels = mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels; CalculateVMRowAndSwath_params->GPUVMMaxPageTableLevels = mode_lib->ms.cache_display_cfg.plane.GPUVMMaxPageTableLevels; CalculateVMRowAndSwath_params->GPUVMMinPageSizeKBytes = mode_lib->ms.cache_display_cfg.plane.GPUVMMinPageSizeKBytes; - CalculateVMRowAndSwath_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes; + CalculateVMRowAndSwath_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024; CalculateVMRowAndSwath_params->PTEBufferModeOverrideEn = mode_lib->ms.cache_display_cfg.plane.PTEBufferModeOverrideEn; CalculateVMRowAndSwath_params->PTEBufferModeOverrideVal = mode_lib->ms.cache_display_cfg.plane.PTEBufferMode; CalculateVMRowAndSwath_params->PTEBufferSizeNotExceeded = s->dummy_boolean_array[0]; @@ -8805,7 +8805,7 @@ void dml_core_mode_programming(struct display_mode_lib_st *mode_lib, const struc mode_lib->ms.cache_display_cfg.hw.DPPPerSurface, locals->dpte_group_bytes, s->HostVMInefficiencyFactor, - mode_lib->ms.soc.hostvm_min_page_size_kbytes, + mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024, mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels); locals->TCalc = 24.0 / locals->DCFCLKDeepSleep; @@ -8995,7 +8995,7 @@ void dml_core_mode_programming(struct display_mode_lib_st *mode_lib, const struc CalculatePrefetchSchedule_params->GPUVMEnable = mode_lib->ms.cache_display_cfg.plane.GPUVMEnable; CalculatePrefetchSchedule_params->HostVMEnable = mode_lib->ms.cache_display_cfg.plane.HostVMEnable; CalculatePrefetchSchedule_params->HostVMMaxNonCachedPageTableLevels = mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels; - CalculatePrefetchSchedule_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes; + CalculatePrefetchSchedule_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024; CalculatePrefetchSchedule_params->DynamicMetadataEnable = mode_lib->ms.cache_display_cfg.plane.DynamicMetadataEnable[k]; CalculatePrefetchSchedule_params->DynamicMetadataVMEnabled = mode_lib->ms.ip.dynamic_metadata_vm_enabled; CalculatePrefetchSchedule_params->DynamicMetadataLinesBeforeActiveRequired = mode_lib->ms.cache_display_cfg.plane.DynamicMetadataLinesBeforeActiveRequired[k]; @@ -9240,7 +9240,7 @@ void dml_core_mode_programming(struct display_mode_lib_st *mode_lib, const struc mode_lib->ms.cache_display_cfg.plane.HostVMEnable, mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels, mode_lib->ms.cache_display_cfg.plane.GPUVMEnable, - mode_lib->ms.soc.hostvm_min_page_size_kbytes, + mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024, locals->PDEAndMetaPTEBytesFrame[k], locals->MetaRowByte[k], locals->PixelPTEBytesPerRow[k], @@ -9446,13 +9446,13 @@ void dml_core_mode_programming(struct display_mode_lib_st *mode_lib, const struc CalculateWatermarks_params->CompressedBufferSizeInkByte = locals->CompressedBufferSizeInkByte; // Output - CalculateWatermarks_params->Watermark = &s->dummy_watermark; // Watermarks *Watermark - CalculateWatermarks_params->DRAMClockChangeSupport = &mode_lib->ms.support.DRAMClockChangeSupport[0]; - CalculateWatermarks_params->MaxActiveDRAMClockChangeLatencySupported = &s->dummy_single_array[0][0]; // dml_float_t *MaxActiveDRAMClockChangeLatencySupported[] - CalculateWatermarks_params->SubViewportLinesNeededInMALL = &mode_lib->ms.SubViewportLinesNeededInMALL[j]; // dml_uint_t SubViewportLinesNeededInMALL[] - CalculateWatermarks_params->FCLKChangeSupport = &mode_lib->ms.support.FCLKChangeSupport[0]; - CalculateWatermarks_params->MaxActiveFCLKChangeLatencySupported = &s->dummy_single[0]; // dml_float_t *MaxActiveFCLKChangeLatencySupported - CalculateWatermarks_params->USRRetrainingSupport = &mode_lib->ms.support.USRRetrainingSupport[0]; + CalculateWatermarks_params->Watermark = &locals->Watermark; // Watermarks *Watermark + CalculateWatermarks_params->DRAMClockChangeSupport = &locals->DRAMClockChangeSupport; + CalculateWatermarks_params->MaxActiveDRAMClockChangeLatencySupported = locals->MaxActiveDRAMClockChangeLatencySupported; // dml_float_t *MaxActiveDRAMClockChangeLatencySupported[] + CalculateWatermarks_params->SubViewportLinesNeededInMALL = locals->SubViewportLinesNeededInMALL; // dml_uint_t SubViewportLinesNeededInMALL[] + CalculateWatermarks_params->FCLKChangeSupport = &locals->FCLKChangeSupport; + CalculateWatermarks_params->MaxActiveFCLKChangeLatencySupported = &locals->MaxActiveFCLKChangeLatencySupported; // dml_float_t *MaxActiveFCLKChangeLatencySupported + CalculateWatermarks_params->USRRetrainingSupport = &locals->USRRetrainingSupport; CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( &mode_lib->scratch, diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_types.h index e85866db80ff..7ca7f2a743c2 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_types.h @@ -38,5 +38,6 @@ #include "core_types.h" #include "dsc.h" #include "clk_mgr.h" +#include "dc_state_priv.h" #endif //__DML2_DC_TYPES_H__ diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c index 32f8a43af3d6..282d70e2b18a 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c @@ -51,7 +51,7 @@ unsigned int dml2_helper_calculate_num_ways_for_subvp(struct dml2_context *ctx, // Find the phantom pipes if (pipe->stream && pipe->plane_state && !pipe->top_pipe && !pipe->prev_odm_pipe && - pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { + ctx->config.svp_pstate.callbacks.get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) { bytes_per_pixel = pipe->plane_state->format >= SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616 ? 8 : 4; mblk_width = ctx->config.mall_cfg.mblk_width_pixels; mblk_height = bytes_per_pixel == 4 ? mblk_width = ctx->config.mall_cfg.mblk_height_4bpe_pixels : ctx->config.mall_cfg.mblk_height_8bpe_pixels; @@ -253,7 +253,7 @@ static bool assign_subvp_pipe(struct dml2_context *ctx, struct dc_state *context * to combine this with SubVP can cause issues with the scheduling). */ if (pipe->plane_state && !pipe->top_pipe && - pipe->stream->mall_stream_config.type == SUBVP_NONE && refresh_rate < 120 && + ctx->config.svp_pstate.callbacks.get_pipe_subvp_type(context, pipe) == SUBVP_NONE && refresh_rate < 120 && vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] <= 0) { while (pipe) { num_pipes++; @@ -317,7 +317,7 @@ static bool enough_pipes_for_subvp(struct dml2_context *ctx, struct dc_state *st // Find the minimum pipe split count for non SubVP pipes if (pipe->stream && !pipe->top_pipe && - pipe->stream->mall_stream_config.type == SUBVP_NONE) { + ctx->config.svp_pstate.callbacks.get_pipe_subvp_type(state, pipe) == SUBVP_NONE) { split_cnt = 0; while (pipe) { split_cnt++; @@ -372,8 +372,8 @@ static bool subvp_subvp_schedulable(struct dml2_context *ctx, struct dc_state *c * and also to store the two main SubVP pipe pointers in subvp_pipes[2]. */ if (pipe->stream && pipe->plane_state && !pipe->top_pipe && - pipe->stream->mall_stream_config.type == SUBVP_MAIN) { - phantom = pipe->stream->mall_stream_config.paired_stream; + ctx->config.svp_pstate.callbacks.get_pipe_subvp_type(context, pipe) == SUBVP_MAIN) { + phantom = ctx->config.svp_pstate.callbacks.get_paired_subvp_stream(context, pipe->stream); microschedule_lines = (phantom->timing.v_total - phantom->timing.v_front_porch) + phantom->timing.v_addressable; @@ -435,6 +435,7 @@ bool dml2_svp_drr_schedulable(struct dml2_context *ctx, struct dc_state *context struct pipe_ctx *pipe = NULL; struct dc_crtc_timing *main_timing = NULL; struct dc_crtc_timing *phantom_timing = NULL; + struct dc_stream_state *phantom_stream; int16_t prefetch_us = 0; int16_t mall_region_us = 0; int16_t drr_frame_us = 0; // nominal frame time @@ -453,12 +454,13 @@ bool dml2_svp_drr_schedulable(struct dml2_context *ctx, struct dc_state *context continue; // Find the SubVP pipe - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) + if (ctx->config.svp_pstate.callbacks.get_pipe_subvp_type(context, pipe) == SUBVP_MAIN) break; } + phantom_stream = ctx->config.svp_pstate.callbacks.get_paired_subvp_stream(context, pipe->stream); main_timing = &pipe->stream->timing; - phantom_timing = &pipe->stream->mall_stream_config.paired_stream->timing; + phantom_timing = &phantom_stream->timing; prefetch_us = (phantom_timing->v_total - phantom_timing->v_front_porch) * phantom_timing->h_total / (double)(phantom_timing->pix_clk_100hz * 100) * 1000000 + ctx->config.svp_pstate.subvp_prefetch_end_to_mall_start_us; @@ -519,6 +521,8 @@ static bool subvp_vblank_schedulable(struct dml2_context *ctx, struct dc_state * struct dc_crtc_timing *main_timing = NULL; struct dc_crtc_timing *phantom_timing = NULL; struct dc_crtc_timing *vblank_timing = NULL; + struct dc_stream_state *phantom_stream; + enum mall_stream_type pipe_mall_type; /* For SubVP + VBLANK/DRR cases, we assume there can only be * a single VBLANK/DRR display. If DML outputs SubVP + VBLANK @@ -528,19 +532,20 @@ static bool subvp_vblank_schedulable(struct dml2_context *ctx, struct dc_state * */ for (i = 0; i < ctx->config.dcn_pipe_count; i++) { pipe = &context->res_ctx.pipe_ctx[i]; + pipe_mall_type = ctx->config.svp_pstate.callbacks.get_pipe_subvp_type(context, pipe); // We check for master pipe, but it shouldn't matter since we only need // the pipe for timing info (stream should be same for any pipe splits) if (!pipe->stream || !pipe->plane_state || pipe->top_pipe || pipe->prev_odm_pipe) continue; - if (!found && pipe->stream->mall_stream_config.type == SUBVP_NONE) { + if (!found && pipe_mall_type == SUBVP_NONE) { // Found pipe which is not SubVP or Phantom (i.e. the VBLANK pipe). vblank_index = i; found = true; } - if (!subvp_pipe && pipe->stream->mall_stream_config.type == SUBVP_MAIN) + if (!subvp_pipe && pipe_mall_type == SUBVP_MAIN) subvp_pipe = pipe; } // Use ignore_msa_timing_param flag to identify as DRR @@ -548,8 +553,9 @@ static bool subvp_vblank_schedulable(struct dml2_context *ctx, struct dc_state * // SUBVP + DRR case schedulable = dml2_svp_drr_schedulable(ctx, context, &context->res_ctx.pipe_ctx[vblank_index].stream->timing); } else if (found) { + phantom_stream = ctx->config.svp_pstate.callbacks.get_paired_subvp_stream(context, subvp_pipe->stream); main_timing = &subvp_pipe->stream->timing; - phantom_timing = &subvp_pipe->stream->mall_stream_config.paired_stream->timing; + phantom_timing = &phantom_stream->timing; vblank_timing = &context->res_ctx.pipe_ctx[vblank_index].stream->timing; // Prefetch time is equal to VACTIVE + BP + VSYNC of the phantom pipe // Also include the prefetch end to mallstart delay time @@ -602,19 +608,20 @@ bool dml2_svp_validate_static_schedulability(struct dml2_context *ctx, struct dc for (i = 0, pipe_idx = 0; i < ctx->config.dcn_pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + enum mall_stream_type pipe_mall_type = ctx->config.svp_pstate.callbacks.get_pipe_subvp_type(context, pipe); if (!pipe->stream) continue; if (pipe->plane_state && !pipe->top_pipe && - pipe->stream->mall_stream_config.type == SUBVP_MAIN) + pipe_mall_type == SUBVP_MAIN) subvp_count++; // Count how many planes that aren't SubVP/phantom are capable of VACTIVE // switching (SubVP + VACTIVE unsupported). In situations where we force // SubVP for a VACTIVE plane, we don't want to increment the vactive_count. if (vba->ActiveDRAMClockChangeLatencyMargin[vba->pipe_plane[pipe_idx]] > 0 && - pipe->stream->mall_stream_config.type == SUBVP_NONE) { + pipe_mall_type == SUBVP_NONE) { vactive_count++; } pipe_idx++; @@ -708,14 +715,10 @@ static void set_phantom_stream_timing(struct dml2_context *ctx, struct dc_state static struct dc_stream_state *enable_phantom_stream(struct dml2_context *ctx, struct dc_state *state, unsigned int dc_pipe_idx, unsigned int svp_height, unsigned int vstartup) { struct pipe_ctx *ref_pipe = &state->res_ctx.pipe_ctx[dc_pipe_idx]; - struct dc_stream_state *phantom_stream = ctx->config.svp_pstate.callbacks.create_stream_for_sink(ref_pipe->stream->sink); - - phantom_stream->signal = SIGNAL_TYPE_VIRTUAL; - phantom_stream->dpms_off = true; - phantom_stream->mall_stream_config.type = SUBVP_PHANTOM; - phantom_stream->mall_stream_config.paired_stream = ref_pipe->stream; - ref_pipe->stream->mall_stream_config.type = SUBVP_MAIN; - ref_pipe->stream->mall_stream_config.paired_stream = phantom_stream; + struct dc_stream_state *phantom_stream = ctx->config.svp_pstate.callbacks.create_phantom_stream( + ctx->config.svp_pstate.callbacks.dc, + state, + ref_pipe->stream); /* stream has limited viewport and small timing */ memcpy(&phantom_stream->timing, &ref_pipe->stream->timing, sizeof(phantom_stream->timing)); @@ -723,7 +726,10 @@ static struct dc_stream_state *enable_phantom_stream(struct dml2_context *ctx, s memcpy(&phantom_stream->dst, &ref_pipe->stream->dst, sizeof(phantom_stream->dst)); set_phantom_stream_timing(ctx, state, ref_pipe, phantom_stream, dc_pipe_idx, svp_height, vstartup); - ctx->config.svp_pstate.callbacks.add_stream_to_ctx(ctx->config.svp_pstate.callbacks.dc, state, phantom_stream); + ctx->config.svp_pstate.callbacks.add_phantom_stream(ctx->config.svp_pstate.callbacks.dc, + state, + phantom_stream, + ref_pipe->stream); return phantom_stream; } @@ -740,7 +746,10 @@ static void enable_phantom_plane(struct dml2_context *ctx, if (curr_pipe->top_pipe && curr_pipe->top_pipe->plane_state == curr_pipe->plane_state) { phantom_plane = prev_phantom_plane; } else { - phantom_plane = ctx->config.svp_pstate.callbacks.create_plane(ctx->config.svp_pstate.callbacks.dc); + phantom_plane = ctx->config.svp_pstate.callbacks.create_phantom_plane( + ctx->config.svp_pstate.callbacks.dc, + state, + curr_pipe->plane_state); } memcpy(&phantom_plane->address, &curr_pipe->plane_state->address, sizeof(phantom_plane->address)); @@ -763,9 +772,7 @@ static void enable_phantom_plane(struct dml2_context *ctx, phantom_plane->clip_rect.y = 0; phantom_plane->clip_rect.height = phantom_stream->timing.v_addressable; - phantom_plane->is_phantom = true; - - ctx->config.svp_pstate.callbacks.add_plane_to_context(ctx->config.svp_pstate.callbacks.dc, phantom_stream, phantom_plane, state); + ctx->config.svp_pstate.callbacks.add_phantom_plane(ctx->config.svp_pstate.callbacks.dc, phantom_stream, phantom_plane, state); curr_pipe = curr_pipe->bottom_pipe; prev_phantom_plane = phantom_plane; @@ -790,7 +797,7 @@ static void add_phantom_pipes_for_main_pipe(struct dml2_context *ctx, struct dc_ // We determine which phantom pipes were added by comparing with // the phantom stream. if (pipe->plane_state && pipe->stream && pipe->stream == phantom_stream && - pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { + ctx->config.svp_pstate.callbacks.get_pipe_subvp_type(state, pipe) == SUBVP_PHANTOM) { pipe->stream->use_dynamic_meta = false; pipe->plane_state->flip_immediate = false; if (!ctx->config.svp_pstate.callbacks.build_scaling_params(pipe)) { @@ -800,7 +807,7 @@ static void add_phantom_pipes_for_main_pipe(struct dml2_context *ctx, struct dc_ } } -static bool remove_all_planes_for_stream(struct dml2_context *ctx, struct dc_stream_state *stream, struct dc_state *context) +static bool remove_all_phantom_planes_for_stream(struct dml2_context *ctx, struct dc_stream_state *stream, struct dc_state *context) { int i, old_plane_count; struct dc_stream_status *stream_status = NULL; @@ -821,9 +828,11 @@ static bool remove_all_planes_for_stream(struct dml2_context *ctx, struct dc_str for (i = 0; i < old_plane_count; i++) del_planes[i] = stream_status->plane_states[i]; - for (i = 0; i < old_plane_count; i++) - if (!ctx->config.svp_pstate.callbacks.remove_plane_from_context(ctx->config.svp_pstate.callbacks.dc, stream, del_planes[i], context)) + for (i = 0; i < old_plane_count; i++) { + if (!ctx->config.svp_pstate.callbacks.remove_phantom_plane(ctx->config.svp_pstate.callbacks.dc, stream, del_planes[i], context)) return false; + ctx->config.svp_pstate.callbacks.release_phantom_plane(ctx->config.svp_pstate.callbacks.dc, context, del_planes[i]); + } return true; } @@ -832,35 +841,21 @@ bool dml2_svp_remove_all_phantom_pipes(struct dml2_context *ctx, struct dc_state { int i; bool removed_pipe = false; - struct dc_plane_state *phantom_plane = NULL; struct dc_stream_state *phantom_stream = NULL; for (i = 0; i < ctx->config.dcn_pipe_count; i++) { struct pipe_ctx *pipe = &state->res_ctx.pipe_ctx[i]; // build scaling params for phantom pipes - if (pipe->plane_state && pipe->stream && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { - phantom_plane = pipe->plane_state; + if (pipe->plane_state && pipe->stream && ctx->config.svp_pstate.callbacks.get_pipe_subvp_type(state, pipe) == SUBVP_PHANTOM) { phantom_stream = pipe->stream; - remove_all_planes_for_stream(ctx, pipe->stream, state); - ctx->config.svp_pstate.callbacks.remove_stream_from_ctx(ctx->config.svp_pstate.callbacks.dc, state, pipe->stream); - - /* Ref count is incremented on allocation and also when added to the context. - * Therefore we must call release for the the phantom plane and stream once - * they are removed from the ctx to finally decrement the refcount to 0 to free. - */ - ctx->config.svp_pstate.callbacks.plane_state_release(phantom_plane); - ctx->config.svp_pstate.callbacks.stream_release(phantom_stream); + remove_all_phantom_planes_for_stream(ctx, phantom_stream, state); + ctx->config.svp_pstate.callbacks.remove_phantom_stream(ctx->config.svp_pstate.callbacks.dc, state, phantom_stream); + ctx->config.svp_pstate.callbacks.release_phantom_stream(ctx->config.svp_pstate.callbacks.dc, state, phantom_stream); removed_pipe = true; } - // Clear all phantom stream info - if (pipe->stream) { - pipe->stream->mall_stream_config.type = SUBVP_NONE; - pipe->stream->mall_stream_config.paired_stream = NULL; - } - if (pipe->plane_state) { pipe->plane_state->is_phantom = false; } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c index 279e7605a0a2..64d01a9cd68c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c @@ -626,8 +626,8 @@ static void populate_dml_output_cfg_from_stream_state(struct dml_output_cfg_st * if (is_dp2p0_output_encoder(pipe)) out->OutputEncoder[location] = dml_dp2p0; break; - out->OutputEncoder[location] = dml_edp; case SIGNAL_TYPE_EDP: + out->OutputEncoder[location] = dml_edp; break; case SIGNAL_TYPE_HDMI_TYPE_A: case SIGNAL_TYPE_DVI_SINGLE_LINK: @@ -1049,8 +1049,10 @@ static void dml2_populate_pipe_to_plane_index_mapping(struct dml2_context *dml2, void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_state *context, struct dml_display_cfg_st *dml_dispcfg) { - int i = 0, j = 0; + int i = 0, j = 0, k = 0; int disp_cfg_stream_location, disp_cfg_plane_location; + enum mall_stream_type stream_mall_type; + struct pipe_ctx *current_pipe_context; for (i = 0; i < __DML2_WRAPPER_MAX_STREAMS_PLANES__; i++) { dml2->v20.scratch.dml_to_dc_pipe_mapping.disp_cfg_to_stream_id_valid[i] = false; @@ -1070,7 +1072,17 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_stat dml2_populate_pipe_to_plane_index_mapping(dml2, context); for (i = 0; i < context->stream_count; i++) { + current_pipe_context = NULL; + for (k = 0; k < MAX_PIPES; k++) { + /* find one pipe allocated to this stream for the purpose of getting + info about the link later */ + if (context->streams[i] == context->res_ctx.pipe_ctx[k].stream) { + current_pipe_context = &context->res_ctx.pipe_ctx[k]; + break; + } + } disp_cfg_stream_location = map_stream_to_dml_display_cfg(dml2, context->streams[i], dml_dispcfg); + stream_mall_type = dc_state_get_stream_subvp_type(context, context->streams[i]); if (disp_cfg_stream_location < 0) disp_cfg_stream_location = dml_dispcfg->num_timings++; @@ -1078,7 +1090,7 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_stat ASSERT(disp_cfg_stream_location >= 0 && disp_cfg_stream_location <= __DML2_WRAPPER_MAX_STREAMS_PLANES__); populate_dml_timing_cfg_from_stream_state(&dml_dispcfg->timing, disp_cfg_stream_location, context->streams[i]); - populate_dml_output_cfg_from_stream_state(&dml_dispcfg->output, disp_cfg_stream_location, context->streams[i], &context->res_ctx.pipe_ctx[i]); + populate_dml_output_cfg_from_stream_state(&dml_dispcfg->output, disp_cfg_stream_location, context->streams[i], current_pipe_context); switch (context->streams[i]->debug.force_odm_combine_segments) { case 2: dml2->v20.dml_core_ctx.policy.ODMUse[disp_cfg_stream_location] = dml_odm_use_policy_combine_2to1; @@ -1115,10 +1127,10 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_stat populate_dml_surface_cfg_from_plane_state(dml2->v20.dml_core_ctx.project, &dml_dispcfg->surface, disp_cfg_plane_location, context->stream_status[i].plane_states[j]); populate_dml_plane_cfg_from_plane_state(&dml_dispcfg->plane, disp_cfg_plane_location, context->stream_status[i].plane_states[j], context); - if (context->streams[i]->mall_stream_config.type == SUBVP_MAIN) { + if (stream_mall_type == SUBVP_MAIN) { dml_dispcfg->plane.UseMALLForPStateChange[disp_cfg_plane_location] = dml_use_mall_pstate_change_sub_viewport; dml_dispcfg->plane.UseMALLForStaticScreen[disp_cfg_plane_location] = dml_use_mall_static_screen_optimize; - } else if (context->streams[i]->mall_stream_config.type == SUBVP_PHANTOM) { + } else if (stream_mall_type == SUBVP_PHANTOM) { dml_dispcfg->plane.UseMALLForPStateChange[disp_cfg_plane_location] = dml_use_mall_pstate_change_phantom_pipe; dml_dispcfg->plane.UseMALLForStaticScreen[disp_cfg_plane_location] = dml_use_mall_static_screen_disable; dml2->v20.dml_core_ctx.policy.ImmediateFlipRequirement[disp_cfg_plane_location] = dml_immediate_flip_not_required; @@ -1135,7 +1147,7 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_stat if (j >= 1) { populate_dml_timing_cfg_from_stream_state(&dml_dispcfg->timing, disp_cfg_plane_location, context->streams[i]); - populate_dml_output_cfg_from_stream_state(&dml_dispcfg->output, disp_cfg_plane_location, context->streams[i], &context->res_ctx.pipe_ctx[i]); + populate_dml_output_cfg_from_stream_state(&dml_dispcfg->output, disp_cfg_plane_location, context->streams[i], current_pipe_context); switch (context->streams[i]->debug.force_odm_combine_segments) { case 2: dml2->v20.dml_core_ctx.policy.ODMUse[disp_cfg_plane_location] = dml_odm_use_policy_combine_2to1; @@ -1147,9 +1159,9 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_stat break; } - if (context->streams[i]->mall_stream_config.type == SUBVP_MAIN) + if (stream_mall_type == SUBVP_MAIN) dml_dispcfg->plane.UseMALLForPStateChange[disp_cfg_plane_location] = dml_use_mall_pstate_change_sub_viewport; - else if (context->streams[i]->mall_stream_config.type == SUBVP_PHANTOM) + else if (stream_mall_type == SUBVP_PHANTOM) dml_dispcfg->plane.UseMALLForPStateChange[disp_cfg_plane_location] = dml_use_mall_pstate_change_phantom_pipe; dml2->v20.scratch.dml_to_dc_pipe_mapping.disp_cfg_to_stream_id[disp_cfg_plane_location] = context->streams[i]->stream_id; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c index 814dbdcf9a78..1068b962d1c1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c @@ -155,11 +155,19 @@ unsigned int dml2_util_get_maximum_odm_combine_for_output(bool force_odm_4to1, e bool is_dp2p0_output_encoder(const struct pipe_ctx *pipe_ctx) { + if (pipe_ctx == NULL || pipe_ctx->stream == NULL) + return false; + /* If this assert is hit then we have a link encoder dynamic management issue */ ASSERT(pipe_ctx->stream_res.hpo_dp_stream_enc ? pipe_ctx->link_res.hpo_dp_link_enc != NULL : true); - if (pipe_ctx->stream == NULL) - return false; + /* Count MST hubs once by treating only 1st remote sink in topology as an encoder */ + if (pipe_ctx->stream->link && pipe_ctx->stream->link->remote_sinks[0]) { + return (pipe_ctx->stream_res.hpo_dp_stream_enc && + pipe_ctx->link_res.hpo_dp_link_enc && + dc_is_dp_signal(pipe_ctx->stream->signal) && + (pipe_ctx->stream->link->remote_sinks[0]->sink_id == pipe_ctx->stream->sink->sink_id)); + } return (pipe_ctx->stream_res.hpo_dp_stream_enc && pipe_ctx->link_res.hpo_dp_link_enc && @@ -279,6 +287,7 @@ static void populate_pipe_ctx_dlg_params_from_dml(struct pipe_ctx *pipe_ctx, str void dml2_calculate_rq_and_dlg_params(const struct dc *dc, struct dc_state *context, struct resource_context *out_new_hw_state, struct dml2_context *in_ctx, unsigned int pipe_cnt) { unsigned int dc_pipe_ctx_index, dml_pipe_idx, plane_id; + enum mall_stream_type pipe_mall_type; bool unbounded_req_enabled = false; struct dml2_calculate_rq_and_dlg_params_scratch *s = &in_ctx->v20.scratch.calculate_rq_and_dlg_params_scratch; @@ -326,7 +335,8 @@ void dml2_calculate_rq_and_dlg_params(const struct dc *dc, struct dc_state *cont */ populate_pipe_ctx_dlg_params_from_dml(&context->res_ctx.pipe_ctx[dc_pipe_ctx_index], &context->bw_ctx.dml2->v20.dml_core_ctx, dml_pipe_idx); - if (context->res_ctx.pipe_ctx[dc_pipe_ctx_index].stream->mall_stream_config.type == SUBVP_PHANTOM) { + pipe_mall_type = dc_state_get_pipe_subvp_type(context, &context->res_ctx.pipe_ctx[dc_pipe_ctx_index]); + if (pipe_mall_type == SUBVP_PHANTOM) { // Phantom pipe requires that DET_SIZE = 0 and no unbounded requests context->res_ctx.pipe_ctx[dc_pipe_ctx_index].det_buffer_size_kb = 0; context->res_ctx.pipe_ctx[dc_pipe_ctx_index].unbounded_req = false; @@ -353,7 +363,7 @@ void dml2_calculate_rq_and_dlg_params(const struct dc *dc, struct dc_state *cont context->res_ctx.pipe_ctx[dc_pipe_ctx_index].plane_state != context->res_ctx.pipe_ctx[dc_pipe_ctx_index].top_pipe->plane_state) && context->res_ctx.pipe_ctx[dc_pipe_ctx_index].prev_odm_pipe == NULL) { /* SS: all active surfaces stored in MALL */ - if (context->res_ctx.pipe_ctx[dc_pipe_ctx_index].stream->mall_stream_config.type != SUBVP_PHANTOM) { + if (pipe_mall_type != SUBVP_PHANTOM) { context->bw_ctx.bw.dcn.mall_ss_size_bytes += context->res_ctx.pipe_ctx[dc_pipe_ctx_index].surface_size_in_mall_bytes; } else { /* SUBVP: phantom surfaces only stored in MALL */ diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c index 9d354fde6908..26307e599614 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c @@ -418,7 +418,7 @@ static int find_drr_eligible_stream(struct dc_state *display_state) int i; for (i = 0; i < display_state->stream_count; i++) { - if (display_state->streams[i]->mall_stream_config.type == SUBVP_NONE + if (dc_state_get_stream_subvp_type(display_state, display_state->streams[i]) == SUBVP_NONE && display_state->streams[i]->ignore_msa_timing_param) { // Use ignore_msa_timing_param flag to identify as DRR return i; @@ -634,6 +634,8 @@ static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_s dml2_extract_watermark_set(&context->bw_ctx.bw.dcn.watermarks.b, &dml2->v20.dml_core_ctx); memcpy(&context->bw_ctx.bw.dcn.watermarks.c, &dml2->v20.g6_temp_read_watermark_set, sizeof(context->bw_ctx.bw.dcn.watermarks.c)); dml2_extract_watermark_set(&context->bw_ctx.bw.dcn.watermarks.d, &dml2->v20.dml_core_ctx); + //copy for deciding zstate use + context->bw_ctx.dml.vba.StutterPeriod = context->bw_ctx.dml2->v20.dml_core_ctx.mp.StutterPeriod; } return result; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h index 0de6886969c6..ee0eb184eb6d 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h @@ -93,15 +93,34 @@ struct dml2_dc_callbacks { struct dml2_dc_svp_callbacks { struct dc *dc; bool (*build_scaling_params)(struct pipe_ctx *pipe_ctx); - struct dc_stream_state* (*create_stream_for_sink)(struct dc_sink *dc_sink_data); - struct dc_plane_state* (*create_plane)(struct dc *dc); - enum dc_status (*add_stream_to_ctx)(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); - bool (*add_plane_to_context)(const struct dc *dc, struct dc_stream_state *stream, struct dc_plane_state *plane_state, struct dc_state *context); - bool (*remove_plane_from_context)(const struct dc *dc, struct dc_stream_state *stream, struct dc_plane_state *plane_state, struct dc_state *context); - enum dc_status (*remove_stream_from_ctx)(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *stream); - void (*plane_state_release)(struct dc_plane_state *plane_state); - void (*stream_release)(struct dc_stream_state *stream); + struct dc_stream_state* (*create_phantom_stream)(const struct dc *dc, + struct dc_state *state, + struct dc_stream_state *main_stream); + struct dc_plane_state* (*create_phantom_plane)(struct dc *dc, + struct dc_state *state, + struct dc_plane_state *main_plane); + enum dc_status (*add_phantom_stream)(struct dc *dc, + struct dc_state *state, + struct dc_stream_state *phantom_stream, + struct dc_stream_state *main_stream); + bool (*add_phantom_plane)(const struct dc *dc, struct dc_stream_state *stream, struct dc_plane_state *plane_state, struct dc_state *context); + bool (*remove_phantom_plane)(const struct dc *dc, + struct dc_stream_state *stream, + struct dc_plane_state *plane_state, + struct dc_state *context); + enum dc_status (*remove_phantom_stream)(struct dc *dc, + struct dc_state *state, + struct dc_stream_state *stream); + void (*release_phantom_plane)(const struct dc *dc, + struct dc_state *state, + struct dc_plane_state *plane); + void (*release_phantom_stream)(const struct dc *dc, + struct dc_state *state, + struct dc_stream_state *stream); void (*release_dsc)(struct resource_context *res_ctx, const struct resource_pool *pool, struct display_stream_compressor **dsc); + enum mall_stream_type (*get_pipe_subvp_type)(const struct dc_state *state, const struct pipe_ctx *pipe_ctx); + enum mall_stream_type (*get_stream_subvp_type)(const struct dc_state *state, const struct dc_stream_state *stream); + struct dc_stream_state *(*get_paired_subvp_stream)(const struct dc_state *state, const struct dc_stream_state *stream); }; struct dml2_clks_table_entry { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/Makefile b/drivers/gpu/drm/amd/display/dc/hwss/Makefile index bccd46bd1815..254136f8e3f9 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/Makefile +++ b/drivers/gpu/drm/amd/display/dc/hwss/Makefile @@ -78,7 +78,7 @@ ifdef CONFIG_DRM_AMD_DC_FP # DCN ############################################################################### -HWSS_DCN10 = dcn10_hwseq.o +HWSS_DCN10 = dcn10_hwseq.o dcn10_init.o AMD_DAL_HWSS_DCN10 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn10/,$(HWSS_DCN10)) @@ -86,7 +86,7 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN10) ############################################################################### -HWSS_DCN20 = dcn20_hwseq.o +HWSS_DCN20 = dcn20_hwseq.o dcn20_init.o AMD_DAL_HWSS_DCN20 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn20/,$(HWSS_DCN20)) @@ -94,7 +94,7 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN20) ############################################################################### -HWSS_DCN201 = dcn201_hwseq.o +HWSS_DCN201 = dcn201_hwseq.o dcn201_init.o AMD_DAL_HWSS_DCN201 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn201/,$(HWSS_DCN201)) @@ -102,7 +102,7 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN201) ############################################################################### -HWSS_DCN21 = dcn21_hwseq.o +HWSS_DCN21 = dcn21_hwseq.o dcn21_init.o AMD_DAL_HWSS_DCN21 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn21/,$(HWSS_DCN21)) @@ -114,7 +114,7 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN21) ############################################################################### -HWSS_DCN30 = dcn30_hwseq.o +HWSS_DCN30 = dcn30_hwseq.o dcn30_init.o AMD_DAL_HWSS_DCN30 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn30/,$(HWSS_DCN30)) @@ -122,7 +122,7 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN30) ############################################################################### -HWSS_DCN301 = dcn301_hwseq.o +HWSS_DCN301 = dcn301_hwseq.o dcn301_init.o AMD_DAL_HWSS_DCN301 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn301/,$(HWSS_DCN301)) @@ -130,15 +130,17 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN301) ############################################################################### -HWSS_DCN302 = dcn302_hwseq.o +HWSS_DCN302 = dcn302_hwseq.o dcn302_init.o AMD_DAL_HWSS_DCN302 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn302/,$(HWSS_DCN302)) AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN302) + + ############################################################################### -HWSS_DCN303 = dcn303_hwseq.o +HWSS_DCN303 = dcn303_hwseq.o dcn303_init.o AMD_DAL_HWSS_DCN303 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn303/,$(HWSS_DCN303)) @@ -146,7 +148,7 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN303) ############################################################################### -HWSS_DCN31 = dcn31_hwseq.o +HWSS_DCN31 = dcn31_hwseq.o dcn31_init.o AMD_DAL_HWSS_DCN31 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn31/,$(HWSS_DCN31)) @@ -154,7 +156,7 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN31) ############################################################################### -HWSS_DCN314 = dcn314_hwseq.o +HWSS_DCN314 = dcn314_hwseq.o dcn314_init.o AMD_DAL_HWSS_DCN314 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn314/,$(HWSS_DCN314)) @@ -162,7 +164,7 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN314) ############################################################################### -HWSS_DCN32 = dcn32_hwseq.o +HWSS_DCN32 = dcn32_hwseq.o dcn32_init.o AMD_DAL_HWSS_DCN32 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn32/,$(HWSS_DCN32)) @@ -170,7 +172,7 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN32) ############################################################################### -HWSS_DCN35 = dcn35_hwseq.o +HWSS_DCN35 = dcn35_hwseq.o dcn35_init.o AMD_DAL_HWSS_DCN35 = $(addprefix $(AMDDALPATH)/dc/hwss/dcn35/,$(HWSS_DCN35)) @@ -180,4 +182,4 @@ AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCN35) ############################################################################### -endif
\ No newline at end of file +endif diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c index c73fe5e9b361..5660f15da291 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c @@ -55,6 +55,7 @@ #include "audio.h" #include "reg_helper.h" #include "panel_cntl.h" +#include "dc_state_priv.h" #include "dpcd_defs.h" /* include DCE11 register header files */ #include "dce/dce_11_0_d.h" @@ -1353,7 +1354,7 @@ static void build_audio_output( if (state->clk_mgr && (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT || pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)) { - audio_output->pll_info.dp_dto_source_clock_in_khz = + audio_output->pll_info.audio_dto_source_clock_in_khz = state->clk_mgr->funcs->get_dp_ref_clk_frequency( state->clk_mgr); } @@ -1596,7 +1597,7 @@ static enum dc_status apply_single_controller_ctx_to_hw( * is constructed with the same sink). Make sure not to override * and link programming on the main. */ - if (pipe_ctx->stream->mall_stream_config.type != SUBVP_PHANTOM) { + if (dc_state_get_pipe_subvp_type(context, pipe_ctx) != SUBVP_PHANTOM) { pipe_ctx->stream->link->psr_settings.psr_feature_enabled = false; pipe_ctx->stream->link->replay_settings.replay_feature_enabled = false; } @@ -1684,7 +1685,7 @@ static void disable_vga_and_power_gate_all_controllers( true); dc->current_state->res_ctx.pipe_ctx[i].pipe_idx = i; - dc->hwss.disable_plane(dc, + dc->hwss.disable_plane(dc, dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i]); } } @@ -2124,7 +2125,8 @@ static void dce110_reset_hw_ctx_wrap( BREAK_TO_DEBUGGER(); } pipe_ctx_old->stream_res.tg->funcs->disable_crtc(pipe_ctx_old->stream_res.tg); - pipe_ctx_old->stream->link->phy_state.symclk_ref_cnts.otg = 0; + if (dc_is_hdmi_tmds_signal(pipe_ctx_old->stream->signal)) + pipe_ctx_old->stream->link->phy_state.symclk_ref_cnts.otg = 0; pipe_ctx_old->plane_res.mi->funcs->free_mem_input( pipe_ctx_old->plane_res.mi, dc->current_state->stream_count); @@ -2133,7 +2135,7 @@ static void dce110_reset_hw_ctx_wrap( old_clk)) old_clk->funcs->cs_power_down(old_clk); - dc->hwss.disable_plane(dc, pipe_ctx_old); + dc->hwss.disable_plane(dc, dc->current_state, pipe_ctx_old); pipe_ctx_old->stream = NULL; } @@ -2497,6 +2499,7 @@ static bool wait_for_reset_trigger_to_occur( /* Enable timing synchronization for a group of Timing Generators. */ static void dce110_enable_timing_synchronization( struct dc *dc, + struct dc_state *state, int group_index, int group_size, struct pipe_ctx *grouped_pipes[]) @@ -2590,6 +2593,7 @@ static void init_hw(struct dc *dc) struct dmcu *dmcu; struct dce_hwseq *hws = dc->hwseq; uint32_t backlight = MAX_BACKLIGHT_LEVEL; + uint32_t user_level = MAX_BACKLIGHT_LEVEL; bp = dc->ctx->dc_bios; for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -2639,13 +2643,15 @@ static void init_hw(struct dc *dc) for (i = 0; i < dc->link_count; i++) { struct dc_link *link = dc->links[i]; - if (link->panel_cntl) + if (link->panel_cntl) { backlight = link->panel_cntl->funcs->hw_init(link->panel_cntl); + user_level = link->panel_cntl->stored_backlight_registers.USER_LEVEL; + } } abm = dc->res_pool->abm; if (abm != NULL) - abm->funcs->abm_init(abm, backlight); + abm->funcs->abm_init(abm, backlight, user_level); dmcu = dc->res_pool->dmcu; if (dmcu != NULL && abm != NULL) @@ -2842,7 +2848,7 @@ static void dce110_post_unlock_program_front_end( { } -static void dce110_power_down_fe(struct dc *dc, struct pipe_ctx *pipe_ctx) +static void dce110_power_down_fe(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx) { struct dce_hwseq *hws = dc->hwseq; int fe_idx = pipe_ctx->plane_res.mi ? diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c index cdb903116eb7..6dd479e8a348 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c @@ -56,6 +56,7 @@ #include "dc_trace.h" #include "dce/dmub_outbox.h" #include "link.h" +#include "dc_state_priv.h" #define DC_LOGGER \ dc_logger @@ -115,7 +116,7 @@ void dcn10_lock_all_pipes(struct dc *dc, !pipe_ctx->stream || (!pipe_ctx->plane_state && !old_pipe_ctx->plane_state) || !tg->funcs->is_tg_enabled(tg) || - pipe_ctx->stream->mall_stream_config.type == SUBVP_PHANTOM) + dc_state_get_pipe_subvp_type(context, pipe_ctx) == SUBVP_PHANTOM) continue; if (lock) @@ -1057,7 +1058,8 @@ static void dcn10_reset_back_end_for_pipe( if (pipe_ctx->stream_res.tg->funcs->set_drr) pipe_ctx->stream_res.tg->funcs->set_drr( pipe_ctx->stream_res.tg, NULL); - pipe_ctx->stream->link->phy_state.symclk_ref_cnts.otg = 0; + if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) + pipe_ctx->stream->link->phy_state.symclk_ref_cnts.otg = 0; } for (i = 0; i < dc->res_pool->pipe_count; i++) @@ -1180,7 +1182,9 @@ void dcn10_verify_allow_pstate_change_high(struct dc *dc) } /* trigger HW to start disconnect plane from stream on the next vsync */ -void dcn10_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn10_plane_atomic_disconnect(struct dc *dc, + struct dc_state *state, + struct pipe_ctx *pipe_ctx) { struct dce_hwseq *hws = dc->hwseq; struct hubp *hubp = pipe_ctx->plane_res.hubp; @@ -1200,7 +1204,7 @@ void dcn10_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) mpc->funcs->remove_mpcc(mpc, mpc_tree_params, mpcc_to_remove); // Phantom pipes have OTG disabled by default, so MPCC_STATUS will never assert idle, // so don't wait for MPCC_IDLE in the programming sequence - if (opp != NULL && !pipe_ctx->plane_state->is_phantom) + if (opp != NULL && dc_state_get_pipe_subvp_type(state, pipe_ctx) != SUBVP_PHANTOM) opp->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; dc->optimized_required = true; @@ -1290,7 +1294,7 @@ void dcn10_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) pipe_ctx->plane_state = NULL; } -void dcn10_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn10_disable_plane(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx) { struct dce_hwseq *hws = dc->hwseq; DC_LOGGER_INIT(dc->ctx->logger); @@ -1416,12 +1420,12 @@ void dcn10_init_pipes(struct dc *dc, struct dc_state *context) dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; - hws->funcs.plane_atomic_disconnect(dc, pipe_ctx); + hws->funcs.plane_atomic_disconnect(dc, context, pipe_ctx); if (tg->funcs->is_tg_enabled(tg)) tg->funcs->unlock(tg); - dc->hwss.disable_plane(dc, pipe_ctx); + dc->hwss.disable_plane(dc, context, pipe_ctx); pipe_ctx->stream_res.tg = NULL; pipe_ctx->plane_res.hubp = NULL; @@ -1486,6 +1490,7 @@ void dcn10_init_hw(struct dc *dc) struct dc_bios *dcb = dc->ctx->dc_bios; struct resource_pool *res_pool = dc->res_pool; uint32_t backlight = MAX_BACKLIGHT_LEVEL; + uint32_t user_level = MAX_BACKLIGHT_LEVEL; bool is_optimized_init_done = false; if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) @@ -1583,12 +1588,14 @@ void dcn10_init_hw(struct dc *dc) for (i = 0; i < dc->link_count; i++) { struct dc_link *link = dc->links[i]; - if (link->panel_cntl) + if (link->panel_cntl) { backlight = link->panel_cntl->funcs->hw_init(link->panel_cntl); + user_level = link->panel_cntl->stored_backlight_registers.USER_LEVEL; + } } if (abm != NULL) - abm->funcs->abm_init(abm, backlight); + abm->funcs->abm_init(abm, backlight, user_level); if (dmcu != NULL && !dmcu->auto_load_dmcu) dmcu->funcs->dmcu_init(dmcu); @@ -2262,6 +2269,7 @@ void dcn10_enable_vblanks_synchronization( void dcn10_enable_timing_synchronization( struct dc *dc, + struct dc_state *state, int group_index, int group_size, struct pipe_ctx *grouped_pipes[]) @@ -2276,7 +2284,7 @@ void dcn10_enable_timing_synchronization( DC_SYNC_INFO("Setting up OTG reset trigger\n"); for (i = 1; i < group_size; i++) { - if (grouped_pipes[i]->stream && grouped_pipes[i]->stream->mall_stream_config.type == SUBVP_PHANTOM) + if (grouped_pipes[i]->stream && dc_state_get_pipe_subvp_type(state, grouped_pipes[i]) == SUBVP_PHANTOM) continue; opp = grouped_pipes[i]->stream_res.opp; @@ -2296,14 +2304,14 @@ void dcn10_enable_timing_synchronization( if (grouped_pipes[i]->stream == NULL) continue; - if (grouped_pipes[i]->stream && grouped_pipes[i]->stream->mall_stream_config.type == SUBVP_PHANTOM) + if (grouped_pipes[i]->stream && dc_state_get_pipe_subvp_type(state, grouped_pipes[i]) == SUBVP_PHANTOM) continue; grouped_pipes[i]->stream->vblank_synchronized = false; } for (i = 1; i < group_size; i++) { - if (grouped_pipes[i]->stream && grouped_pipes[i]->stream->mall_stream_config.type == SUBVP_PHANTOM) + if (grouped_pipes[i]->stream && dc_state_get_pipe_subvp_type(state, grouped_pipes[i]) == SUBVP_PHANTOM) continue; grouped_pipes[i]->stream_res.tg->funcs->enable_reset_trigger( @@ -2317,11 +2325,11 @@ void dcn10_enable_timing_synchronization( * synchronized. Look at last pipe programmed to reset. */ - if (grouped_pipes[1]->stream && grouped_pipes[1]->stream->mall_stream_config.type != SUBVP_PHANTOM) + if (grouped_pipes[1]->stream && dc_state_get_pipe_subvp_type(state, grouped_pipes[1]) != SUBVP_PHANTOM) wait_for_reset_trigger_to_occur(dc_ctx, grouped_pipes[1]->stream_res.tg); for (i = 1; i < group_size; i++) { - if (grouped_pipes[i]->stream && grouped_pipes[i]->stream->mall_stream_config.type == SUBVP_PHANTOM) + if (grouped_pipes[i]->stream && dc_state_get_pipe_subvp_type(state, grouped_pipes[i]) == SUBVP_PHANTOM) continue; grouped_pipes[i]->stream_res.tg->funcs->disable_reset_trigger( @@ -2329,7 +2337,7 @@ void dcn10_enable_timing_synchronization( } for (i = 1; i < group_size; i++) { - if (grouped_pipes[i]->stream && grouped_pipes[i]->stream->mall_stream_config.type == SUBVP_PHANTOM) + if (dc_state_get_pipe_subvp_type(state, grouped_pipes[i]) == SUBVP_PHANTOM) continue; opp = grouped_pipes[i]->stream_res.opp; @@ -3021,7 +3029,7 @@ void dcn10_post_unlock_program_front_end( for (i = 0; i < dc->res_pool->pipe_count; i++) if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable) - dc->hwss.disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); + dc->hwss.disable_plane(dc, dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i]); for (i = 0; i < dc->res_pool->pipe_count; i++) if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable) { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h index ef6d56da417c..bc5dd68a2408 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h @@ -75,7 +75,7 @@ void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx); void dcn10_reset_hw_ctx_wrap( struct dc *dc, struct dc_state *context); -void dcn10_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn10_disable_plane(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx); void dcn10_lock_all_pipes( struct dc *dc, struct dc_state *context, @@ -108,13 +108,16 @@ void dcn10_power_down_on_boot(struct dc *dc); enum dc_status dce110_apply_ctx_to_hw( struct dc *dc, struct dc_state *context); -void dcn10_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn10_plane_atomic_disconnect(struct dc *dc, + struct dc_state *state, + struct pipe_ctx *pipe_ctx); void dcn10_update_dchub(struct dce_hwseq *hws, struct dchub_init_data *dh_data); void dcn10_update_pending_status(struct pipe_ctx *pipe_ctx); void dce110_power_down(struct dc *dc); void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context); void dcn10_enable_timing_synchronization( struct dc *dc, + struct dc_state *state, int group_index, int group_size, struct pipe_ctx *grouped_pipes[]); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_init.c index a5bdac79a744..a5bdac79a744 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_init.h index 8c6fd7b844a4..8c6fd7b844a4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_init.h diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c index c3c83178eb1e..e931342fcf4c 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c @@ -55,6 +55,7 @@ #include "inc/link_enc_cfg.h" #include "link_hwss.h" #include "link.h" +#include "dc_state_priv.h" #define DC_LOGGER \ dc_logger @@ -623,9 +624,9 @@ void dcn20_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) } -void dcn20_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn20_disable_plane(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx) { - bool is_phantom = pipe_ctx->plane_state && pipe_ctx->plane_state->is_phantom; + bool is_phantom = dc_state_get_pipe_subvp_type(state, pipe_ctx) == SUBVP_PHANTOM; struct timing_generator *tg = is_phantom ? pipe_ctx->stream_res.tg : NULL; DC_LOGGER_INIT(dc->ctx->logger); @@ -847,7 +848,7 @@ enum dc_status dcn20_enable_stream_timing( /* TODO enable stream if timing changed */ /* TODO unblank stream if DP */ - if (pipe_ctx->stream && pipe_ctx->stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (pipe_ctx->stream && dc_state_get_pipe_subvp_type(context, pipe_ctx) == SUBVP_PHANTOM) { if (pipe_ctx->stream_res.tg && pipe_ctx->stream_res.tg->funcs->phantom_crtc_post_enable) pipe_ctx->stream_res.tg->funcs->phantom_crtc_post_enable(pipe_ctx->stream_res.tg); } @@ -1368,8 +1369,14 @@ void dcn20_pipe_control_lock( } } -static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx *new_pipe) +static void dcn20_detect_pipe_changes(struct dc_state *old_state, + struct dc_state *new_state, + struct pipe_ctx *old_pipe, + struct pipe_ctx *new_pipe) { + bool old_is_phantom = dc_state_get_pipe_subvp_type(old_state, old_pipe) == SUBVP_PHANTOM; + bool new_is_phantom = dc_state_get_pipe_subvp_type(new_state, new_pipe) == SUBVP_PHANTOM; + new_pipe->update_flags.raw = 0; /* If non-phantom pipe is being transitioned to a phantom pipe, @@ -1379,8 +1386,8 @@ static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx * be different). The post_unlock sequence will set the correct * update flags to enable the phantom pipe. */ - if (old_pipe->plane_state && !old_pipe->plane_state->is_phantom && - new_pipe->plane_state && new_pipe->plane_state->is_phantom) { + if (old_pipe->plane_state && !old_is_phantom && + new_pipe->plane_state && new_is_phantom) { new_pipe->update_flags.bits.disable = 1; return; } @@ -1400,6 +1407,10 @@ static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx new_pipe->update_flags.bits.scaler = 1; new_pipe->update_flags.bits.viewport = 1; new_pipe->update_flags.bits.det_size = 1; + if (new_pipe->stream->test_pattern.type != DP_TEST_PATTERN_VIDEO_MODE && + new_pipe->stream_res.test_pattern_params.width != 0 && + new_pipe->stream_res.test_pattern_params.height != 0) + new_pipe->update_flags.bits.test_pattern_changed = 1; if (!new_pipe->top_pipe && !new_pipe->prev_odm_pipe) { new_pipe->update_flags.bits.odm = 1; new_pipe->update_flags.bits.global_sync = 1; @@ -1412,14 +1423,14 @@ static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx * The remove-add sequence of the phantom pipe always results in the pipe * being blanked in enable_stream_timing (DPG). */ - if (new_pipe->stream && new_pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) + if (new_pipe->stream && dc_state_get_pipe_subvp_type(new_state, new_pipe) == SUBVP_PHANTOM) new_pipe->update_flags.bits.enable = 1; /* Phantom pipes are effectively disabled, if the pipe was previously phantom * we have to enable */ - if (old_pipe->plane_state && old_pipe->plane_state->is_phantom && - new_pipe->plane_state && !new_pipe->plane_state->is_phantom) + if (old_pipe->plane_state && old_is_phantom && + new_pipe->plane_state && !new_is_phantom) new_pipe->update_flags.bits.enable = 1; if (old_pipe->plane_state && !new_pipe->plane_state) { @@ -1556,6 +1567,7 @@ static void dcn20_update_dchubp_dpp( struct dc_plane_state *plane_state = pipe_ctx->plane_state; struct dccg *dccg = dc->res_pool->dccg; bool viewport_changed = false; + enum mall_stream_type pipe_mall_type = dc_state_get_pipe_subvp_type(context, pipe_ctx); if (pipe_ctx->update_flags.bits.dppclk) dpp->funcs->dpp_dppclk_control(dpp, false, true); @@ -1701,7 +1713,7 @@ static void dcn20_update_dchubp_dpp( pipe_ctx->update_flags.bits.plane_changed || plane_state->update_flags.bits.addr_update) { if (resource_is_pipe_type(pipe_ctx, OTG_MASTER) && - pipe_ctx->stream->mall_stream_config.type == SUBVP_MAIN) { + pipe_mall_type == SUBVP_MAIN) { union block_sequence_params params; params.subvp_save_surf_addr.dc_dmub_srv = dc->ctx->dmub_srv; @@ -1715,7 +1727,7 @@ static void dcn20_update_dchubp_dpp( if (pipe_ctx->update_flags.bits.enable) hubp->funcs->set_blank(hubp, false); /* If the stream paired with this plane is phantom, the plane is also phantom */ - if (pipe_ctx->stream && pipe_ctx->stream->mall_stream_config.type == SUBVP_PHANTOM + if (pipe_ctx->stream && pipe_mall_type == SUBVP_PHANTOM && hubp->funcs->phantom_hubp_post_enable) hubp->funcs->phantom_hubp_post_enable(hubp); } @@ -1773,7 +1785,7 @@ static void dcn20_program_pipe( pipe_ctx->pipe_dlg_param.vupdate_offset, pipe_ctx->pipe_dlg_param.vupdate_width); - if (pipe_ctx->stream->mall_stream_config.type != SUBVP_PHANTOM) + if (dc_state_get_pipe_subvp_type(context, pipe_ctx) != SUBVP_PHANTOM) pipe_ctx->stream_res.tg->funcs->wait_for_state(pipe_ctx->stream_res.tg, CRTC_STATE_VACTIVE); pipe_ctx->stream_res.tg->funcs->set_vtg_params( @@ -1912,7 +1924,7 @@ void dcn20_program_front_end_for_ctx( /* Set pipe update flags and lock pipes */ for (i = 0; i < dc->res_pool->pipe_count; i++) - dcn20_detect_pipe_changes(&dc->current_state->res_ctx.pipe_ctx[i], + dcn20_detect_pipe_changes(dc->current_state, context, &dc->current_state->res_ctx.pipe_ctx[i], &context->res_ctx.pipe_ctx[i]); /* When disabling phantom pipes, turn on phantom OTG first (so we can get double @@ -1922,15 +1934,16 @@ void dcn20_program_front_end_for_ctx( struct dc_stream_state *stream = dc->current_state->res_ctx.pipe_ctx[i].stream; if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable && stream && - dc->current_state->res_ctx.pipe_ctx[i].stream->mall_stream_config.type == SUBVP_PHANTOM) { + dc_state_get_pipe_subvp_type(dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i]) == SUBVP_PHANTOM) { struct timing_generator *tg = dc->current_state->res_ctx.pipe_ctx[i].stream_res.tg; if (tg->funcs->enable_crtc) { if (dc->hwss.blank_phantom) { int main_pipe_width, main_pipe_height; + struct dc_stream_state *phantom_stream = dc_state_get_paired_subvp_stream(dc->current_state, dc->current_state->res_ctx.pipe_ctx[i].stream); - main_pipe_width = dc->current_state->res_ctx.pipe_ctx[i].stream->mall_stream_config.paired_stream->dst.width; - main_pipe_height = dc->current_state->res_ctx.pipe_ctx[i].stream->mall_stream_config.paired_stream->dst.height; + main_pipe_width = phantom_stream->dst.width; + main_pipe_height = phantom_stream->dst.height; dc->hwss.blank_phantom(dc, tg, main_pipe_width, main_pipe_height); } tg->funcs->enable_crtc(tg); @@ -1959,9 +1972,9 @@ void dcn20_program_front_end_for_ctx( * DET allocation. */ if (hubbub->funcs->program_det_size && (context->res_ctx.pipe_ctx[i].update_flags.bits.disable || - (context->res_ctx.pipe_ctx[i].plane_state && context->res_ctx.pipe_ctx[i].plane_state->is_phantom))) + (context->res_ctx.pipe_ctx[i].plane_state && dc_state_get_pipe_subvp_type(context, &context->res_ctx.pipe_ctx[i]) == SUBVP_PHANTOM))) hubbub->funcs->program_det_size(hubbub, dc->current_state->res_ctx.pipe_ctx[i].plane_res.hubp->inst, 0); - hws->funcs.plane_atomic_disconnect(dc, &dc->current_state->res_ctx.pipe_ctx[i]); + hws->funcs.plane_atomic_disconnect(dc, dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i]); DC_LOG_DC("Reset mpcc for pipe %d\n", dc->current_state->res_ctx.pipe_ctx[i].pipe_idx); } @@ -1984,7 +1997,7 @@ void dcn20_program_front_end_for_ctx( * but the MPO still exists until the double buffered update of the main pipe so we * will get a frame of underflow if the phantom pipe is programmed here. */ - if (pipe->stream && pipe->stream->mall_stream_config.type != SUBVP_PHANTOM) + if (pipe->stream && dc_state_get_pipe_subvp_type(context, pipe) != SUBVP_PHANTOM) dcn20_program_pipe(dc, pipe, context); } @@ -2034,7 +2047,7 @@ void dcn20_post_unlock_program_front_end( for (i = 0; i < dc->res_pool->pipe_count; i++) if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable) - dc->hwss.disable_plane(dc, &dc->current_state->res_ctx.pipe_ctx[i]); + dc->hwss.disable_plane(dc, dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i]); /* * If we are enabling a pipe, we need to wait for pending clear as this is a critical @@ -2046,7 +2059,7 @@ void dcn20_post_unlock_program_front_end( struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; // Don't check flip pending on phantom pipes if (pipe->plane_state && !pipe->top_pipe && pipe->update_flags.bits.enable && - pipe->stream->mall_stream_config.type != SUBVP_PHANTOM) { + dc_state_get_pipe_subvp_type(context, pipe) != SUBVP_PHANTOM) { struct hubp *hubp = pipe->plane_res.hubp; int j = 0; for (j = 0; j < TIMEOUT_FOR_PIPE_ENABLE_US / polling_interval_us @@ -2069,7 +2082,7 @@ void dcn20_post_unlock_program_front_end( * programming sequence). */ while (pipe) { - if (pipe->stream && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (pipe->stream && dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) { /* When turning on the phantom pipe we want to run through the * entire enable sequence, so apply all the "enable" flags. */ @@ -2139,7 +2152,7 @@ void dcn20_prepare_bandwidth( struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; // At optimize don't restore the original watermark value - if (pipe->stream && pipe->stream->mall_stream_config.type != SUBVP_NONE) { + if (pipe->stream && dc_state_get_pipe_subvp_type(context, pipe) != SUBVP_NONE) { context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = 4U * 1000U * 1000U * 1000U; break; } @@ -2183,7 +2196,7 @@ void dcn20_optimize_bandwidth( struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; // At optimize don't need to restore the original watermark value - if (pipe->stream && pipe->stream->mall_stream_config.type != SUBVP_NONE) { + if (pipe->stream && dc_state_get_pipe_subvp_type(context, pipe) != SUBVP_NONE) { context->bw_ctx.bw.dcn.watermarks.a.cstate_pstate.pstate_change_ns = 4U * 1000U * 1000U * 1000U; break; } @@ -2217,7 +2230,8 @@ void dcn20_optimize_bandwidth( dc->clk_mgr, context, true); - if (context->bw_ctx.bw.dcn.clk.zstate_support == DCN_ZSTATE_SUPPORT_ALLOW) { + if (context->bw_ctx.bw.dcn.clk.zstate_support == DCN_ZSTATE_SUPPORT_ALLOW && + !dc->debug.disable_extblankadj) { for (i = 0; i < dc->res_pool->pipe_count; ++i) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; @@ -2610,7 +2624,8 @@ static void dcn20_reset_back_end_for_pipe( * the case where the same symclk is shared across multiple otg * instances */ - link->phy_state.symclk_ref_cnts.otg = 0; + if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) + link->phy_state.symclk_ref_cnts.otg = 0; if (link->phy_state.symclk_state == SYMCLK_ON_TX_OFF) { link_hwss->disable_link_output(link, &pipe_ctx->link_res, pipe_ctx->stream->signal); @@ -2943,7 +2958,7 @@ void dcn20_fpga_init_hw(struct dc *dc) dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; /*to do*/ - hws->funcs.plane_atomic_disconnect(dc, pipe_ctx); + hws->funcs.plane_atomic_disconnect(dc, context, pipe_ctx); } /* initialize DWB pointer to MCIF_WB */ @@ -2960,7 +2975,7 @@ void dcn20_fpga_init_hw(struct dc *dc) for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - dc->hwss.disable_plane(dc, pipe_ctx); + dc->hwss.disable_plane(dc, context, pipe_ctx); pipe_ctx->stream_res.tg = NULL; pipe_ctx->plane_res.hubp = NULL; diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h index ab02e4e9c8c2..b94c85340abf 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h @@ -52,7 +52,7 @@ void dcn20_program_output_csc(struct dc *dc, void dcn20_enable_stream(struct pipe_ctx *pipe_ctx); void dcn20_unblank_stream(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings); -void dcn20_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_disable_plane(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx); void dcn20_disable_pixel_data( struct dc *dc, struct pipe_ctx *pipe_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_init.c index 884e3e323338..884e3e323338 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_init.h index 12277797cd71..12277797cd71 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_init.h diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.c index d3fe6092f50e..d5769f38874f 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.c @@ -320,7 +320,7 @@ void dcn201_init_hw(struct dc *dc) res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; pipe_ctx->stream_res.opp = res_pool->opps[i]; /*To do: number of MPCC != number of opp*/ - hws->funcs.plane_atomic_disconnect(dc, pipe_ctx); + hws->funcs.plane_atomic_disconnect(dc, context, pipe_ctx); } /* initialize DWB pointer to MCIF_WB */ @@ -337,7 +337,7 @@ void dcn201_init_hw(struct dc *dc) for (i = 0; i < res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - dc->hwss.disable_plane(dc, pipe_ctx); + dc->hwss.disable_plane(dc, context, pipe_ctx); pipe_ctx->stream_res.tg = NULL; pipe_ctx->plane_res.hubp = NULL; @@ -369,7 +369,9 @@ void dcn201_init_hw(struct dc *dc) } /* trigger HW to start disconnect plane from stream on the next vsync */ -void dcn201_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn201_plane_atomic_disconnect(struct dc *dc, + struct dc_state *state, + struct pipe_ctx *pipe_ctx) { struct dce_hwseq *hws = dc->hwseq; struct hubp *hubp = pipe_ctx->plane_res.hubp; diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.h index 26cd62be6418..6a50a9894be6 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_hwseq.h @@ -33,7 +33,7 @@ void dcn201_init_hw(struct dc *dc); void dcn201_unblank_stream(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings); void dcn201_update_plane_addr(const struct dc *dc, struct pipe_ctx *pipe_ctx); -void dcn201_plane_atomic_disconnect(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn201_plane_atomic_disconnect(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx); void dcn201_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx); void dcn201_set_cursor_attribute(struct pipe_ctx *pipe_ctx); void dcn201_pipe_control_lock( diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_init.c index a13bf6c9386e..a13bf6c9386e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_init.h index 1168887b033d..1168887b033d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn201/dcn201_init.h diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c index 08783ad097d2..8e88dcaf88f5 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c @@ -154,7 +154,7 @@ static bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, cmd.abm_set_pipe.abm_set_pipe_data.ramping_boundary = ramping_boundary; cmd.abm_set_pipe.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pipe_data); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -173,7 +173,7 @@ static void dmub_abm_set_backlight(struct dc_context *dc, uint32_t backlight_pwm cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_inst); cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data); - dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } void dcn21_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx) diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_init.c index 18249c6b6d81..18249c6b6d81 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_init.h index 3ed24292648a..3ed24292648a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_init.h diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c index fd8a8c10a201..c34c13e1e0a4 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c @@ -51,7 +51,7 @@ #include "dcn20/dcn20_hwseq.h" #include "dcn30/dcn30_resource.h" #include "link.h" - +#include "dc_state_priv.h" @@ -476,6 +476,7 @@ void dcn30_init_hw(struct dc *dc) int i; int edp_num; uint32_t backlight = MAX_BACKLIGHT_LEVEL; + uint32_t user_level = MAX_BACKLIGHT_LEVEL; if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); @@ -612,13 +613,15 @@ void dcn30_init_hw(struct dc *dc) for (i = 0; i < dc->link_count; i++) { struct dc_link *link = dc->links[i]; - if (link->panel_cntl) + if (link->panel_cntl) { backlight = link->panel_cntl->funcs->hw_init(link->panel_cntl); + user_level = link->panel_cntl->stored_backlight_registers.USER_LEVEL; + } } for (i = 0; i < dc->res_pool->pipe_count; i++) { if (abms[i] != NULL) - abms[i]->funcs->abm_init(abms[i], backlight); + abms[i]->funcs->abm_init(abms[i], backlight, user_level); } /* power AFMT HDMI memory TODO: may move to dis/en output save power*/ @@ -754,7 +757,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.mall.header.sub_type = DMUB_CMD__MALL_ACTION_NO_DF_REQ; cmd.mall.header.payload_bytes = sizeof(cmd.mall) - sizeof(cmd.mall.header); - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); return true; } @@ -876,7 +879,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.mall.cursor_height = cursor_attr.height; cmd.mall.cursor_pitch = cursor_attr.pitch; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); /* Use copied cursor, and it's okay to not switch back */ cursor_attr.address.quad_part = cmd.mall.cursor_copy_dst.quad_part; @@ -892,7 +895,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.mall.tmr_scale = tmr_scale; cmd.mall.debug_bits = dc->debug.mall_error_as_fatal; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); return true; } @@ -909,7 +912,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.mall.header.payload_bytes = sizeof(cmd.mall) - sizeof(cmd.mall.header); - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -966,7 +969,7 @@ void dcn30_hardware_release(struct dc *dc) if (!pipe->stream) continue; - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) { + if (dc_state_get_pipe_subvp_type(dc->current_state, pipe) == SUBVP_MAIN) { subvp_in_use = true; break; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c index 9894caedffed..9894caedffed 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.h index c280ff90bfa3..c280ff90bfa3 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.h diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.c index 6477009ce065..6477009ce065 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.h index 0bca48ccbfa2..0bca48ccbfa2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.h diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn302/dcn302_init.c index 637f9514d37b..637f9514d37b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn302/dcn302_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn302/dcn302_init.h index 899587b93aa1..899587b93aa1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn302/dcn302_init.h diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn303/dcn303_init.c index edb4d68b8187..edb4d68b8187 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn303/dcn303_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn303/dcn303_init.h index 4949981126d7..4949981126d7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn303/dcn303_init.h diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c index 5d62805f3bdf..7423880fabb6 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c @@ -113,6 +113,7 @@ void dcn31_init_hw(struct dc *dc) struct dc_bios *dcb = dc->ctx->dc_bios; struct resource_pool *res_pool = dc->res_pool; uint32_t backlight = MAX_BACKLIGHT_LEVEL; + uint32_t user_level = MAX_BACKLIGHT_LEVEL; int i; if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) @@ -224,13 +225,15 @@ void dcn31_init_hw(struct dc *dc) for (i = 0; i < dc->link_count; i++) { struct dc_link *link = dc->links[i]; - if (link->panel_cntl) + if (link->panel_cntl) { backlight = link->panel_cntl->funcs->hw_init(link->panel_cntl); + user_level = link->panel_cntl->stored_backlight_registers.USER_LEVEL; + } } for (i = 0; i < dc->res_pool->pipe_count; i++) { if (abms[i] != NULL) - abms[i]->funcs->abm_init(abms[i], backlight); + abms[i]->funcs->abm_init(abms[i], backlight, user_level); } /* power AFMT HDMI memory TODO: may move to dis/en output save power*/ @@ -416,7 +419,7 @@ void dcn31_z10_save_init(struct dc *dc) cmd.dcn_restore.header.type = DMUB_CMD__IDLE_OPT; cmd.dcn_restore.header.sub_type = DMUB_CMD__IDLE_OPT_DCN_SAVE_INIT; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } void dcn31_z10_restore(const struct dc *dc) @@ -434,7 +437,7 @@ void dcn31_z10_restore(const struct dc *dc) cmd.dcn_restore.header.type = DMUB_CMD__IDLE_OPT; cmd.dcn_restore.header.sub_type = DMUB_CMD__IDLE_OPT_DCN_RESTORE; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } void dcn31_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on) @@ -524,7 +527,8 @@ static void dcn31_reset_back_end_for_pipe( if (pipe_ctx->stream_res.tg->funcs->set_odm_bypass) pipe_ctx->stream_res.tg->funcs->set_odm_bypass( pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); - pipe_ctx->stream->link->phy_state.symclk_ref_cnts.otg = 0; + if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) + pipe_ctx->stream->link->phy_state.symclk_ref_cnts.otg = 0; if (pipe_ctx->stream_res.tg->funcs->set_drr) pipe_ctx->stream_res.tg->funcs->set_drr( diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c index 669f524bd064..669f524bd064 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.h index a3db08c8bd35..a3db08c8bd35 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.h diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c index ccb7e317e86a..ccb7e317e86a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.h index 8f92e66577cf..8f92e66577cf 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.h diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c index 5c323718ec90..6c9299c7683d 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c @@ -51,6 +51,7 @@ #include "dcn32/dcn32_resource.h" #include "link.h" #include "../dcn20/dcn20_hwseq.h" +#include "dc_state_priv.h" #define DC_LOGGER_INIT(logger) @@ -277,7 +278,7 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.cab.header.sub_type = DMUB_CMD__CAB_NO_DCN_REQ; cmd.cab.header.payload_bytes = sizeof(cmd.cab) - sizeof(cmd.cab.header); - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); return true; } @@ -311,7 +312,7 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.cab.header.payload_bytes = sizeof(cmd.cab) - sizeof(cmd.cab.header); cmd.cab.cab_alloc_ways = (uint8_t)ways; - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); return true; } @@ -327,7 +328,7 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.cab.header.payload_bytes = sizeof(cmd.cab) - sizeof(cmd.cab.header); - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + dc_wake_and_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -348,8 +349,7 @@ void dcn32_commit_subvp_config(struct dc *dc, struct dc_state *context) for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; - if (pipe_ctx->stream && pipe_ctx->stream->mall_stream_config.paired_stream && - pipe_ctx->stream->mall_stream_config.type == SUBVP_MAIN) { + if (pipe_ctx->stream && dc_state_get_pipe_subvp_type(context, pipe_ctx) == SUBVP_MAIN) { // There is at least 1 SubVP pipe, so enable SubVP enable_subvp = true; break; @@ -375,18 +375,20 @@ void dcn32_subvp_pipe_control_lock(struct dc *dc, bool subvp_immediate_flip = false; bool subvp_in_use = false; struct pipe_ctx *pipe; + enum mall_stream_type pipe_mall_type = SUBVP_NONE; for (i = 0; i < dc->res_pool->pipe_count; i++) { pipe = &context->res_ctx.pipe_ctx[i]; + pipe_mall_type = dc_state_get_pipe_subvp_type(context, pipe); - if (pipe->stream && pipe->plane_state && pipe->stream->mall_stream_config.type == SUBVP_MAIN) { + if (pipe->stream && pipe->plane_state && pipe_mall_type == SUBVP_MAIN) { subvp_in_use = true; break; } } if (top_pipe_to_program && top_pipe_to_program->stream && top_pipe_to_program->plane_state) { - if (top_pipe_to_program->stream->mall_stream_config.type == SUBVP_MAIN && + if (dc_state_get_pipe_subvp_type(context, top_pipe_to_program) == SUBVP_MAIN && top_pipe_to_program->plane_state->flip_immediate) subvp_immediate_flip = true; } @@ -398,7 +400,7 @@ void dcn32_subvp_pipe_control_lock(struct dc *dc, if (!lock) { for (i = 0; i < dc->res_pool->pipe_count; i++) { pipe = &context->res_ctx.pipe_ctx[i]; - if (pipe->stream && pipe->plane_state && pipe->stream->mall_stream_config.type == SUBVP_MAIN && + if (pipe->stream && pipe->plane_state && pipe_mall_type == SUBVP_MAIN && should_lock_all_pipes) pipe->stream_res.tg->funcs->wait_for_state(pipe->stream_res.tg, CRTC_STATE_VBLANK); } @@ -416,14 +418,7 @@ void dcn32_subvp_pipe_control_lock_fast(union block_sequence_params *params) { struct dc *dc = params->subvp_pipe_control_lock_fast_params.dc; bool lock = params->subvp_pipe_control_lock_fast_params.lock; - struct pipe_ctx *pipe_ctx = params->subvp_pipe_control_lock_fast_params.pipe_ctx; - bool subvp_immediate_flip = false; - - if (pipe_ctx && pipe_ctx->stream && pipe_ctx->plane_state) { - if (pipe_ctx->stream->mall_stream_config.type == SUBVP_MAIN && - pipe_ctx->plane_state->flip_immediate) - subvp_immediate_flip = true; - } + bool subvp_immediate_flip = params->subvp_pipe_control_lock_fast_params.subvp_immediate_flip; // Don't need to lock for DRR VSYNC flips -- FW will wait for DRR pending update cleared. if (subvp_immediate_flip) { @@ -609,7 +604,7 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context) struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; struct hubp *hubp = pipe->plane_res.hubp; - if (!pipe->stream || !(pipe->stream->mall_stream_config.type == SUBVP_MAIN || + if (!pipe->stream || !(dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN || pipe->stream->fpo_in_use)) { if (hubp && hubp->funcs->hubp_update_force_pstate_disallow) hubp->funcs->hubp_update_force_pstate_disallow(hubp, false); @@ -624,7 +619,7 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context) struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; struct hubp *hubp = pipe->plane_res.hubp; - if (pipe->stream && (pipe->stream->mall_stream_config.type == SUBVP_MAIN || + if (pipe->stream && (dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN || pipe->stream->fpo_in_use)) { if (hubp && hubp->funcs->hubp_update_force_pstate_disallow) hubp->funcs->hubp_update_force_pstate_disallow(hubp, true); @@ -671,8 +666,8 @@ void dcn32_update_mall_sel(struct dc *dc, struct dc_state *context) if (cursor_size > 16384) cache_cursor = true; - if (pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { - hubp->funcs->hubp_update_mall_sel(hubp, 1, false); + if (dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) { + hubp->funcs->hubp_update_mall_sel(hubp, 1, false); } else { // MALL not supported with Stereo3D hubp->funcs->hubp_update_mall_sel(hubp, @@ -714,9 +709,8 @@ void dcn32_program_mall_pipe_config(struct dc *dc, struct dc_state *context) * see if CURSOR_REQ_MODE will be back to 1 for SubVP * when it should be 0 for MPO */ - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) { + if (dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN) hubp->funcs->hubp_prepare_subvp_buffering(hubp, true); - } } } } @@ -759,6 +753,7 @@ void dcn32_init_hw(struct dc *dc) int i; int edp_num; uint32_t backlight = MAX_BACKLIGHT_LEVEL; + uint32_t user_level = MAX_BACKLIGHT_LEVEL; if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); @@ -913,13 +908,15 @@ void dcn32_init_hw(struct dc *dc) for (i = 0; i < dc->link_count; i++) { struct dc_link *link = dc->links[i]; - if (link->panel_cntl) + if (link->panel_cntl) { backlight = link->panel_cntl->funcs->hw_init(link->panel_cntl); + user_level = link->panel_cntl->stored_backlight_registers.USER_LEVEL; + } } for (i = 0; i < dc->res_pool->pipe_count; i++) { if (abms[i] != NULL && abms[i]->funcs != NULL) - abms[i]->funcs->abm_init(abms[i], backlight); + abms[i]->funcs->abm_init(abms[i], backlight, user_level); } /* power AFMT HDMI memory TODO: may move to dis/en output save power*/ @@ -960,6 +957,12 @@ void dcn32_init_hw(struct dc *dc) dc->caps.dmub_caps.subvp_psr = dc->ctx->dmub_srv->dmub->feature_caps.subvp_psr_support; dc->caps.dmub_caps.gecc_enable = dc->ctx->dmub_srv->dmub->feature_caps.gecc_enable; dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch; + + if (dc->ctx->dmub_srv->dmub->fw_version < + DMUB_FW_VERSION(7, 0, 35)) { + dc->debug.force_disable_subvp = true; + dc->debug.disable_fpo_optimizations = true; + } } } @@ -1222,7 +1225,7 @@ void dcn32_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_ continue; if ((pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal)) - && pipe->stream->mall_stream_config.type != SUBVP_PHANTOM) { + && dc_state_get_pipe_subvp_type(dc->current_state, pipe) != SUBVP_PHANTOM) { pipe->stream_res.tg->funcs->disable_crtc(pipe->stream_res.tg); reset_sync_context_for_pipe(dc, context, i); otg_disabled[i] = true; @@ -1373,8 +1376,8 @@ void dcn32_update_phantom_vp_position(struct dc *dc, for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (pipe->stream && pipe->stream->mall_stream_config.type == SUBVP_MAIN && - pipe->stream->mall_stream_config.paired_stream == phantom_pipe->stream) { + if (pipe->stream && dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_MAIN && + dc_state_get_paired_subvp_stream(context, pipe->stream) == phantom_pipe->stream) { if (pipe->plane_state && pipe->plane_state->update_flags.bits.position_change) { phantom_plane->src_rect.x = pipe->plane_state->src_rect.x; @@ -1399,21 +1402,19 @@ void dcn32_update_phantom_vp_position(struct dc *dc, void dcn32_apply_update_flags_for_phantom(struct pipe_ctx *phantom_pipe) { phantom_pipe->update_flags.raw = 0; - if (phantom_pipe->stream && phantom_pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { - if (resource_is_pipe_type(phantom_pipe, DPP_PIPE)) { - phantom_pipe->update_flags.bits.enable = 1; - phantom_pipe->update_flags.bits.mpcc = 1; - phantom_pipe->update_flags.bits.dppclk = 1; - phantom_pipe->update_flags.bits.hubp_interdependent = 1; - phantom_pipe->update_flags.bits.hubp_rq_dlg_ttu = 1; - phantom_pipe->update_flags.bits.gamut_remap = 1; - phantom_pipe->update_flags.bits.scaler = 1; - phantom_pipe->update_flags.bits.viewport = 1; - phantom_pipe->update_flags.bits.det_size = 1; - if (resource_is_pipe_type(phantom_pipe, OTG_MASTER)) { - phantom_pipe->update_flags.bits.odm = 1; - phantom_pipe->update_flags.bits.global_sync = 1; - } + if (resource_is_pipe_type(phantom_pipe, DPP_PIPE)) { + phantom_pipe->update_flags.bits.enable = 1; + phantom_pipe->update_flags.bits.mpcc = 1; + phantom_pipe->update_flags.bits.dppclk = 1; + phantom_pipe->update_flags.bits.hubp_interdependent = 1; + phantom_pipe->update_flags.bits.hubp_rq_dlg_ttu = 1; + phantom_pipe->update_flags.bits.gamut_remap = 1; + phantom_pipe->update_flags.bits.scaler = 1; + phantom_pipe->update_flags.bits.viewport = 1; + phantom_pipe->update_flags.bits.det_size = 1; + if (resource_is_pipe_type(phantom_pipe, OTG_MASTER)) { + phantom_pipe->update_flags.bits.odm = 1; + phantom_pipe->update_flags.bits.global_sync = 1; } } } @@ -1485,8 +1486,8 @@ void dcn32_enable_phantom_streams(struct dc *dc, struct dc_state *context) * pipe, wait for the double buffer update to complete first before we do * ANY phantom pipe programming. */ - if (pipe->stream && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM && - old_pipe->stream && old_pipe->stream->mall_stream_config.type != SUBVP_PHANTOM) { + if (pipe->stream && dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM && + old_pipe->stream && dc_state_get_pipe_subvp_type(dc->current_state, old_pipe) != SUBVP_PHANTOM) { old_pipe->stream_res.tg->funcs->wait_for_state( old_pipe->stream_res.tg, CRTC_STATE_VBLANK); @@ -1498,7 +1499,7 @@ void dcn32_enable_phantom_streams(struct dc *dc, struct dc_state *context) for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i]; - if (new_pipe->stream && new_pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { + if (new_pipe->stream && dc_state_get_pipe_subvp_type(context, new_pipe) == SUBVP_PHANTOM) { // If old context or new context has phantom pipes, apply // the phantom timings now. We can't change the phantom // pipe configuration safely without driver acquiring diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c index 427cfc8c24a4..427cfc8c24a4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.h index 89a591eb2c23..89a591eb2c23 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.h diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c index 9262d3336182..9c806385ecbd 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c @@ -56,6 +56,7 @@ #include "dcn30/dcn30_cm_common.h" #include "dcn31/dcn31_hwseq.h" #include "dcn20/dcn20_hwseq.h" +#include "dc_state_priv.h" #define DC_LOGGER_INIT(logger) \ struct dal_logger *dc_logger = logger @@ -133,6 +134,7 @@ void dcn35_init_hw(struct dc *dc) struct dc_bios *dcb = dc->ctx->dc_bios; struct resource_pool *res_pool = dc->res_pool; uint32_t backlight = MAX_BACKLIGHT_LEVEL; + uint32_t user_level = MAX_BACKLIGHT_LEVEL; int i; if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) @@ -279,13 +281,15 @@ void dcn35_init_hw(struct dc *dc) for (i = 0; i < dc->link_count; i++) { struct dc_link *link = dc->links[i]; - if (link->panel_cntl) + if (link->panel_cntl) { backlight = link->panel_cntl->funcs->hw_init(link->panel_cntl); + user_level = link->panel_cntl->stored_backlight_registers.USER_LEVEL; + } } if (dc->ctx->dmub_srv) { for (i = 0; i < dc->res_pool->pipe_count; i++) { if (abms[i] != NULL && abms[i]->funcs != NULL) - abms[i]->funcs->abm_init(abms[i], backlight); + abms[i]->funcs->abm_init(abms[i], backlight, user_level); } } @@ -687,11 +691,7 @@ bool dcn35_apply_idle_power_optimizations(struct dc *dc, bool enable) } // TODO: review other cases when idle optimization is allowed - - if (!enable) - dc_dmub_srv_exit_low_power_state(dc); - else - dc_dmub_srv_notify_idle(dc, enable); + dc_dmub_srv_apply_idle_power_optimizations(dc, enable); return true; } @@ -701,7 +701,7 @@ void dcn35_z10_restore(const struct dc *dc) if (dc->debug.disable_z10) return; - dc_dmub_srv_exit_low_power_state(dc); + dc_dmub_srv_apply_idle_power_optimizations(dc, false); dcn31_z10_restore(dc); } @@ -817,12 +817,12 @@ void dcn35_init_pipes(struct dc *dc, struct dc_state *context) dc->res_pool->opps[i]->mpcc_disconnect_pending[pipe_ctx->plane_res.mpcc_inst] = true; pipe_ctx->stream_res.opp = dc->res_pool->opps[i]; - hws->funcs.plane_atomic_disconnect(dc, pipe_ctx); + hws->funcs.plane_atomic_disconnect(dc, context, pipe_ctx); if (tg->funcs->is_tg_enabled(tg)) tg->funcs->unlock(tg); - dc->hwss.disable_plane(dc, pipe_ctx); + dc->hwss.disable_plane(dc, context, pipe_ctx); pipe_ctx->stream_res.tg = NULL; pipe_ctx->plane_res.hubp = NULL; @@ -949,10 +949,10 @@ void dcn35_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx) pipe_ctx->plane_state = NULL; } -void dcn35_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) +void dcn35_disable_plane(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx) { struct dce_hwseq *hws = dc->hwseq; - bool is_phantom = pipe_ctx->plane_state && pipe_ctx->plane_state->is_phantom; + bool is_phantom = dc_state_get_pipe_subvp_type(state, pipe_ctx) == SUBVP_PHANTOM; struct timing_generator *tg = is_phantom ? pipe_ctx->stream_res.tg : NULL; DC_LOGGER_INIT(dc->ctx->logger); @@ -1123,21 +1123,28 @@ void dcn35_calc_blocks_to_ungate(struct dc *dc, struct dc_state *context, update_state->pg_res_update[PG_HPO] = true; } + /** - * power down sequence - * ONO Region 3, DCPG 25: hpo - SKIPPED - * ONO Region 4, DCPG 0: dchubp0, dpp0 - * ONO Region 6, DCPG 1: dchubp1, dpp1 - * ONO Region 8, DCPG 2: dchubp2, dpp2 - * ONO Region 10, DCPG 3: dchubp3, dpp3 - * ONO Region 1, DCPG 23: dchubbub dchvm dchubbubmem - SKIPPED. PMFW will pwr dwn at IPS2 entry - * ONO Region 5, DCPG 16: dsc0 - * ONO Region 7, DCPG 17: dsc1 - * ONO Region 9, DCPG 18: dsc2 - * ONO Region 11, DCPG 19: dsc3 - * ONO Region 2, DCPG 24: mpc opp optc dwb - * ONO Region 0, DCPG 22: dccg dio dcio - SKIPPED. will be pwr dwn after lono timer is armed -*/ + * dcn35_hw_block_power_down() - power down sequence + * + * The following sequence describes the ON-OFF (ONO) for power down: + * + * ONO Region 3, DCPG 25: hpo - SKIPPED + * ONO Region 4, DCPG 0: dchubp0, dpp0 + * ONO Region 6, DCPG 1: dchubp1, dpp1 + * ONO Region 8, DCPG 2: dchubp2, dpp2 + * ONO Region 10, DCPG 3: dchubp3, dpp3 + * ONO Region 1, DCPG 23: dchubbub dchvm dchubbubmem - SKIPPED. PMFW will pwr dwn at IPS2 entry + * ONO Region 5, DCPG 16: dsc0 + * ONO Region 7, DCPG 17: dsc1 + * ONO Region 9, DCPG 18: dsc2 + * ONO Region 11, DCPG 19: dsc3 + * ONO Region 2, DCPG 24: mpc opp optc dwb + * ONO Region 0, DCPG 22: dccg dio dcio - SKIPPED. will be pwr dwn after lono timer is armed + * + * @dc: Current DC state + * @update_state: update PG sequence states for HW block + */ void dcn35_hw_block_power_down(struct dc *dc, struct pg_block_update *update_state) { @@ -1175,20 +1182,27 @@ void dcn35_hw_block_power_down(struct dc *dc, //domain22, 23, 25 currently always on. } + /** - * power up sequence - * ONO Region 0, DCPG 22: dccg dio dcio - SKIPPED - * ONO Region 2, DCPG 24: mpc opp optc dwb - * ONO Region 5, DCPG 16: dsc0 - * ONO Region 7, DCPG 17: dsc1 - * ONO Region 9, DCPG 18: dsc2 - * ONO Region 11, DCPG 19: dsc3 - * ONO Region 1, DCPG 23: dchubbub dchvm dchubbubmem - SKIPPED. PMFW will power up at IPS2 exit - * ONO Region 4, DCPG 0: dchubp0, dpp0 - * ONO Region 6, DCPG 1: dchubp1, dpp1 - * ONO Region 8, DCPG 2: dchubp2, dpp2 - * ONO Region 10, DCPG 3: dchubp3, dpp3 - * ONO Region 3, DCPG 25: hpo - SKIPPED + * dcn35_hw_block_power_up() - power up sequence + * + * The following sequence describes the ON-OFF (ONO) for power up: + * + * ONO Region 0, DCPG 22: dccg dio dcio - SKIPPED + * ONO Region 2, DCPG 24: mpc opp optc dwb + * ONO Region 5, DCPG 16: dsc0 + * ONO Region 7, DCPG 17: dsc1 + * ONO Region 9, DCPG 18: dsc2 + * ONO Region 11, DCPG 19: dsc3 + * ONO Region 1, DCPG 23: dchubbub dchvm dchubbubmem - SKIPPED. PMFW will power up at IPS2 exit + * ONO Region 4, DCPG 0: dchubp0, dpp0 + * ONO Region 6, DCPG 1: dchubp1, dpp1 + * ONO Region 8, DCPG 2: dchubp2, dpp2 + * ONO Region 10, DCPG 3: dchubp3, dpp3 + * ONO Region 3, DCPG 25: hpo - SKIPPED + * + * @dc: Current DC state + * @update_state: update PG sequence states for HW block */ void dcn35_hw_block_power_up(struct dc *dc, struct pg_block_update *update_state) @@ -1315,3 +1329,44 @@ uint32_t dcn35_get_idle_state(const struct dc *dc) return 0; } + +void dcn35_set_drr(struct pipe_ctx **pipe_ctx, + int num_pipes, struct dc_crtc_timing_adjust adjust) +{ + int i = 0; + struct drr_params params = {0}; + // DRR set trigger event mapped to OTG_TRIG_A (bit 11) for manual control flow + unsigned int event_triggers = 0x800; + // Note DRR trigger events are generated regardless of whether num frames met. + unsigned int num_frames = 2; + + params.vertical_total_max = adjust.v_total_max; + params.vertical_total_min = adjust.v_total_min; + params.vertical_total_mid = adjust.v_total_mid; + params.vertical_total_mid_frame_num = adjust.v_total_mid_frame_num; + + for (i = 0; i < num_pipes; i++) { + if ((pipe_ctx[i]->stream_res.tg != NULL) && pipe_ctx[i]->stream_res.tg->funcs) { + struct dc_crtc_timing *timing = &pipe_ctx[i]->stream->timing; + struct dc *dc = pipe_ctx[i]->stream->ctx->dc; + + if (dc->debug.static_screen_wait_frames) { + unsigned int frame_rate = timing->pix_clk_100hz / (timing->h_total * timing->v_total); + + if (frame_rate >= 120 && dc->caps.ips_support && + dc->config.disable_ips != DMUB_IPS_DISABLE_ALL) { + /*ips enable case*/ + num_frames = 2 * (frame_rate % 60); + } + } + if (pipe_ctx[i]->stream_res.tg->funcs->set_drr) + pipe_ctx[i]->stream_res.tg->funcs->set_drr( + pipe_ctx[i]->stream_res.tg, ¶ms); + if (adjust.v_total_max != 0 && adjust.v_total_min != 0) + if (pipe_ctx[i]->stream_res.tg->funcs->set_static_screen_control) + pipe_ctx[i]->stream_res.tg->funcs->set_static_screen_control( + pipe_ctx[i]->stream_res.tg, + event_triggers, num_frames); + } + } +} diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.h index 3837038dc4a8..fd66316e33de 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.h @@ -57,7 +57,7 @@ void dcn35_init_pipes(struct dc *dc, struct dc_state *context); void dcn35_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx); void dcn35_enable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx, struct dc_state *context); -void dcn35_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn35_disable_plane(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx); void dcn35_calc_blocks_to_gate(struct dc *dc, struct dc_state *context, struct pg_block_update *update_state); @@ -86,4 +86,8 @@ void dcn35_dsc_pg_control( void dcn35_set_idle_state(const struct dc *dc, bool allow_idle); uint32_t dcn35_get_idle_state(const struct dc *dc); + +void dcn35_set_drr(struct pipe_ctx **pipe_ctx, + int num_pipes, struct dc_crtc_timing_adjust adjust); + #endif /* __DC_HWSS_DCN35_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c index d594905eb246..a630aa77dcec 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c @@ -68,7 +68,7 @@ static const struct hw_sequencer_funcs dcn35_funcs = { .prepare_bandwidth = dcn35_prepare_bandwidth, .optimize_bandwidth = dcn35_optimize_bandwidth, .update_bandwidth = dcn20_update_bandwidth, - .set_drr = dcn10_set_drr, + .set_drr = dcn35_set_drr, .get_position = dcn10_get_position, .set_static_screen_control = dcn30_set_static_screen_control, .setup_stereo = dcn10_setup_stereo, diff --git a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.h index b67015032c35..b67015032c35 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.h diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/CMakeLists.txt b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/CMakeLists.txt new file mode 100644 index 000000000000..951ca2da4486 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/CMakeLists.txt @@ -0,0 +1,4 @@ +dal3_subdirectory_sources( + dcn351_init.c + dcn351_init.h +) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/Makefile b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/Makefile new file mode 100644 index 000000000000..b24ad27fe6ef --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/Makefile @@ -0,0 +1,17 @@ +# +# (c) Copyright 2022 Advanced Micro Devices, Inc. All the rights reserved +# +# All rights reserved. This notice is intended as a precaution against +# inadvertent publication and does not imply publication or any waiver +# of confidentiality. The year included in the foregoing notice is the +# year of creation of the work. +# +# Authors: AMD +# +# Makefile for DCN351. + +DCN351 = dcn351_init.o + +AMD_DAL_DCN351 = $(addprefix $(AMDDALPATH)/dc/dcn351/,$(DCN351)) + +AMD_DISPLAY_FILES += $(AMD_DAL_DCN351) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c new file mode 100644 index 000000000000..143d3fc0221c --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c @@ -0,0 +1,171 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dce110/dce110_hwseq.h" +#include "dcn10/dcn10_hwseq.h" +#include "dcn20/dcn20_hwseq.h" +#include "dcn21/dcn21_hwseq.h" +#include "dcn30/dcn30_hwseq.h" +#include "dcn301/dcn301_hwseq.h" +#include "dcn31/dcn31_hwseq.h" +#include "dcn32/dcn32_hwseq.h" +#include "dcn35/dcn35_hwseq.h" + +#include "dcn351_init.h" + +static const struct hw_sequencer_funcs dcn351_funcs = { + .program_gamut_remap = dcn30_program_gamut_remap, + .init_hw = dcn35_init_hw, + .power_down_on_boot = dcn35_power_down_on_boot, + .apply_ctx_to_hw = dce110_apply_ctx_to_hw, + .apply_ctx_for_surface = NULL, + .program_front_end_for_ctx = dcn20_program_front_end_for_ctx, + .wait_for_pending_cleared = dcn10_wait_for_pending_cleared, + .post_unlock_program_front_end = dcn20_post_unlock_program_front_end, + .update_plane_addr = dcn20_update_plane_addr, + .update_dchub = dcn10_update_dchub, + .update_pending_status = dcn10_update_pending_status, + .program_output_csc = dcn20_program_output_csc, + .enable_accelerated_mode = dce110_enable_accelerated_mode, + .enable_timing_synchronization = dcn10_enable_timing_synchronization, + .enable_per_frame_crtc_position_reset = dcn10_enable_per_frame_crtc_position_reset, + .update_info_frame = dcn31_update_info_frame, + .send_immediate_sdp_message = dcn10_send_immediate_sdp_message, + .enable_stream = dcn20_enable_stream, + .disable_stream = dce110_disable_stream, + .unblank_stream = dcn32_unblank_stream, + .blank_stream = dce110_blank_stream, + .enable_audio_stream = dce110_enable_audio_stream, + .disable_audio_stream = dce110_disable_audio_stream, + .disable_plane = dcn35_disable_plane, + .disable_pixel_data = dcn20_disable_pixel_data, + .pipe_control_lock = dcn20_pipe_control_lock, + .interdependent_update_lock = dcn10_lock_all_pipes, + .cursor_lock = dcn10_cursor_lock, + .prepare_bandwidth = dcn35_prepare_bandwidth, + .optimize_bandwidth = dcn35_optimize_bandwidth, + .update_bandwidth = dcn20_update_bandwidth, + .set_drr = dcn10_set_drr, + .get_position = dcn10_get_position, + .set_static_screen_control = dcn30_set_static_screen_control, + .setup_stereo = dcn10_setup_stereo, + .set_avmute = dcn30_set_avmute, + .log_hw_state = dcn10_log_hw_state, + .get_hw_state = dcn10_get_hw_state, + .clear_status_bits = dcn10_clear_status_bits, + .wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect, + .edp_backlight_control = dce110_edp_backlight_control, + .edp_power_control = dce110_edp_power_control, + .edp_wait_for_T12 = dce110_edp_wait_for_T12, + .edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready, + .set_cursor_position = dcn10_set_cursor_position, + .set_cursor_attribute = dcn10_set_cursor_attribute, + .set_cursor_sdr_white_level = dcn10_set_cursor_sdr_white_level, + .setup_periodic_interrupt = dcn10_setup_periodic_interrupt, + .set_clock = dcn10_set_clock, + .get_clock = dcn10_get_clock, + .program_triplebuffer = dcn20_program_triple_buffer, + .enable_writeback = dcn30_enable_writeback, + .disable_writeback = dcn30_disable_writeback, + .update_writeback = dcn30_update_writeback, + .mmhubbub_warmup = dcn30_mmhubbub_warmup, + .dmdata_status_done = dcn20_dmdata_status_done, + .program_dmdata_engine = dcn30_program_dmdata_engine, + .set_dmdata_attributes = dcn20_set_dmdata_attributes, + .init_sys_ctx = dcn31_init_sys_ctx, + .init_vm_ctx = dcn20_init_vm_ctx, + .set_flip_control_gsl = dcn20_set_flip_control_gsl, + .get_vupdate_offset_from_vsync = dcn10_get_vupdate_offset_from_vsync, + .calc_vupdate_position = dcn10_calc_vupdate_position, + .power_down = dce110_power_down, + .set_backlight_level = dcn21_set_backlight_level, + .set_abm_immediate_disable = dcn21_set_abm_immediate_disable, + .set_pipe = dcn21_set_pipe, + .enable_lvds_link_output = dce110_enable_lvds_link_output, + .enable_tmds_link_output = dce110_enable_tmds_link_output, + .enable_dp_link_output = dce110_enable_dp_link_output, + .disable_link_output = dcn32_disable_link_output, + .z10_restore = dcn35_z10_restore, + .z10_save_init = dcn31_z10_save_init, + .set_disp_pattern_generator = dcn30_set_disp_pattern_generator, + .optimize_pwr_state = dcn21_optimize_pwr_state, + .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, + .update_visual_confirm_color = dcn10_update_visual_confirm_color, + .apply_idle_power_optimizations = dcn35_apply_idle_power_optimizations, + .update_dsc_pg = dcn32_update_dsc_pg, + .calc_blocks_to_gate = dcn35_calc_blocks_to_gate, + .calc_blocks_to_ungate = dcn35_calc_blocks_to_ungate, + .hw_block_power_up = dcn35_hw_block_power_up, + .hw_block_power_down = dcn35_hw_block_power_down, + .root_clock_control = dcn35_root_clock_control, + .set_idle_state = dcn35_set_idle_state, + .get_idle_state = dcn35_get_idle_state +}; + +static const struct hwseq_private_funcs dcn351_private_funcs = { + .init_pipes = dcn35_init_pipes, + .update_plane_addr = dcn20_update_plane_addr, + .plane_atomic_disconnect = dcn10_plane_atomic_disconnect, + .update_mpcc = dcn20_update_mpcc, + .set_input_transfer_func = dcn32_set_input_transfer_func, + .set_output_transfer_func = dcn32_set_output_transfer_func, + .power_down = dce110_power_down, + .enable_display_power_gating = dcn10_dummy_display_power_gating, + .blank_pixel_data = dcn20_blank_pixel_data, + .reset_hw_ctx_wrap = dcn31_reset_hw_ctx_wrap, + .enable_stream_timing = dcn20_enable_stream_timing, + .edp_backlight_control = dce110_edp_backlight_control, + .setup_vupdate_interrupt = dcn20_setup_vupdate_interrupt, + .did_underflow_occur = dcn10_did_underflow_occur, + .init_blank = dcn20_init_blank, + .disable_vga = NULL, + .bios_golden_init = dcn10_bios_golden_init, + .plane_atomic_disable = dcn35_plane_atomic_disable, + //.plane_atomic_disable = dcn20_plane_atomic_disable,/*todo*/ + //.hubp_pg_control = dcn35_hubp_pg_control, + .enable_power_gating_plane = dcn35_enable_power_gating_plane, + .dpp_root_clock_control = dcn35_dpp_root_clock_control, + .program_all_writeback_pipes_in_tree = dcn30_program_all_writeback_pipes_in_tree, + .update_odm = dcn35_update_odm, + .set_hdr_multiplier = dcn10_set_hdr_multiplier, + .verify_allow_pstate_change_high = dcn10_verify_allow_pstate_change_high, + .wait_for_blank_complete = dcn20_wait_for_blank_complete, + .dccg_init = dcn20_dccg_init, + .set_mcm_luts = dcn32_set_mcm_luts, + .setup_hpo_hw_control = dcn35_setup_hpo_hw_control, + .calculate_dccg_k1_k2_values = dcn32_calculate_dccg_k1_k2_values, + .set_pixels_per_cycle = dcn32_set_pixels_per_cycle, + .is_dp_dig_pixel_rate_div_policy = dcn32_is_dp_dig_pixel_rate_div_policy, + .dsc_pg_control = dcn35_dsc_pg_control, + .dsc_pg_status = dcn32_dsc_pg_status, + .enable_plane = dcn35_enable_plane, +}; + +void dcn351_hw_sequencer_construct(struct dc *dc) +{ + dc->hwss = dcn351_funcs; + dc->hwseq->funcs = dcn351_private_funcs; + +} diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.h new file mode 100644 index 000000000000..970b01008b23 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.h @@ -0,0 +1,33 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_DCN351_INIT_H__ +#define __DC_DCN351_INIT_H__ + +struct dc; + +void dcn351_hw_sequencer_construct(struct dc *dc); + +#endif /* __DC_DCN351_INIT_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h index 45dc6d4e9562..a54399383318 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h @@ -50,7 +50,7 @@ struct pg_block_update; struct subvp_pipe_control_lock_fast_params { struct dc *dc; bool lock; - struct pipe_ctx *pipe_ctx; + bool subvp_immediate_flip; }; struct pipe_control_lock_params { @@ -200,7 +200,7 @@ struct hw_sequencer_funcs { struct dc_state *context); enum dc_status (*apply_ctx_to_hw)(struct dc *dc, struct dc_state *context); - void (*disable_plane)(struct dc *dc, struct pipe_ctx *pipe_ctx); + void (*disable_plane)(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx); void (*disable_pixel_data)(struct dc *dc, struct pipe_ctx *pipe_ctx, bool blank); void (*apply_ctx_for_surface)(struct dc *dc, const struct dc_stream_state *stream, @@ -248,6 +248,7 @@ struct hw_sequencer_funcs { void (*enable_per_frame_crtc_position_reset)(struct dc *dc, int group_size, struct pipe_ctx *grouped_pipes[]); void (*enable_timing_synchronization)(struct dc *dc, + struct dc_state *state, int group_index, int group_size, struct pipe_ctx *grouped_pipes[]); void (*enable_vblanks_synchronization)(struct dc *dc, @@ -454,17 +455,18 @@ void get_mpctree_visual_confirm_color( struct tg_color *color); void get_subvp_visual_confirm_color( - struct dc *dc, - struct dc_state *context, struct pipe_ctx *pipe_ctx, struct tg_color *color); void get_mclk_switch_visual_confirm_color( - struct dc *dc, - struct dc_state *context, struct pipe_ctx *pipe_ctx, struct tg_color *color); +void set_p_state_switch_method( + struct dc *dc, + struct dc_state *context, + struct pipe_ctx *pipe_ctx); + void hwss_execute_sequence(struct dc *dc, struct block_sequence block_sequence[], int num_steps); @@ -474,7 +476,8 @@ void hwss_build_fast_sequence(struct dc *dc, unsigned int dmub_cmd_count, struct block_sequence block_sequence[], int *num_steps, - struct pipe_ctx *pipe_ctx); + struct pipe_ctx *pipe_ctx, + struct dc_stream_status *stream_status); void hwss_send_dmcub_cmd(union block_sequence_params *params); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h index 82c592166875..6137cf09aa54 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h @@ -79,6 +79,7 @@ struct hwseq_private_funcs { void (*update_plane_addr)(const struct dc *dc, struct pipe_ctx *pipe_ctx); void (*plane_atomic_disconnect)(struct dc *dc, + struct dc_state *state, struct pipe_ctx *pipe_ctx); void (*update_mpcc)(struct dc *dc, struct pipe_ctx *pipe_ctx); bool (*set_input_transfer_func)(struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index 10397d4dfb07..f74ae0d41d3c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -200,11 +200,7 @@ struct resource_funcs { unsigned int pipe_cnt, unsigned int index); - bool (*remove_phantom_pipes)(struct dc *dc, struct dc_state *context, bool fast_update); - void (*retain_phantom_pipes)(struct dc *dc, struct dc_state *context); void (*get_panel_config_defaults)(struct dc_panel_config *panel_config); - void (*save_mall_state)(struct dc *dc, struct dc_state *context, struct mall_temp_config *temp_config); - void (*restore_mall_state)(struct dc *dc, struct dc_state *context, struct mall_temp_config *temp_config); void (*build_pipe_pix_clk_params)(struct pipe_ctx *pipe_ctx); }; @@ -385,6 +381,16 @@ union pipe_update_flags { uint32_t raw; }; +enum p_state_switch_method { + P_STATE_UNKNOWN = 0, + P_STATE_V_BLANK = 1, + P_STATE_FPO, + P_STATE_V_ACTIVE, + P_STATE_SUB_VP, + P_STATE_DRR_SUB_VP, + P_STATE_V_BLANK_SUB_VP +}; + struct pipe_ctx { struct dc_plane_state *plane_state; struct dc_stream_state *stream; @@ -433,6 +439,7 @@ struct pipe_ctx { struct dwbc *dwbc; struct mcif_wb *mcif_wb; union pipe_update_flags update_flags; + enum p_state_switch_method p_state_type; struct tg_color visual_confirm_color; bool has_vactive_margin; /* subvp_index: only valid if the pipe is a SUBVP_MAIN*/ @@ -526,6 +533,14 @@ struct dc_state { * @stream_status: Planes status on a given stream */ struct dc_stream_status stream_status[MAX_PIPES]; + /** + * @phantom_streams: Stream state properties for phantoms + */ + struct dc_stream_state *phantom_streams[MAX_PHANTOM_PIPES]; + /** + * @phantom_planes: Planes state properties for phantoms + */ + struct dc_plane_state *phantom_planes[MAX_PHANTOM_PIPES]; /** * @stream_count: Total of streams in use @@ -534,6 +549,14 @@ struct dc_state { uint8_t stream_mask; /** + * @stream_count: Total phantom streams in use + */ + uint8_t phantom_stream_count; + /** + * @stream_count: Total phantom planes in use + */ + uint8_t phantom_plane_count; + /** * @res_ctx: Persistent state of resources */ struct resource_context res_ctx; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h index 9f521cf0fc5a..3f0161d64675 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h @@ -36,7 +36,7 @@ struct abm { }; struct abm_funcs { - void (*abm_init)(struct abm *abm, uint32_t back_light); + void (*abm_init)(struct abm *abm, uint32_t back_light, uint32_t user_level); bool (*set_abm_level)(struct abm *abm, unsigned int abm_level); bool (*set_abm_immediate_disable)(struct abm *abm, unsigned int panel_inst); bool (*set_pipe)(struct abm *abm, unsigned int controller_id, unsigned int panel_inst); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h index fa9614bcb160..17e014d3bdc8 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h @@ -62,6 +62,25 @@ struct dcn3_clk_internal { uint32_t CLK4_CLK0_CURRENT_CNT; //fclk }; +struct dcn35_clk_internal { + int dummy; + uint32_t CLK1_CLK0_CURRENT_CNT; //dispclk + uint32_t CLK1_CLK1_CURRENT_CNT; //dppclk + uint32_t CLK1_CLK2_CURRENT_CNT; //dprefclk + uint32_t CLK1_CLK3_CURRENT_CNT; //dcfclk + uint32_t CLK1_CLK4_CURRENT_CNT; //dtbclk + //uint32_t CLK1_CLK5_CURRENT_CNT; //dpiaclk + //uint32_t CLK1_CLK6_CURRENT_CNT; //srdbgclk + uint32_t CLK1_CLK3_DS_CNTL; //dcf_deep_sleep_divider + uint32_t CLK1_CLK3_ALLOW_DS; //dcf_deep_sleep_allow + + uint32_t CLK1_CLK0_BYPASS_CNTL; //dispclk bypass + uint32_t CLK1_CLK1_BYPASS_CNTL; //dppclk bypass + uint32_t CLK1_CLK2_BYPASS_CNTL; //dprefclk bypass + uint32_t CLK1_CLK3_BYPASS_CNTL; //dcfclk bypass + uint32_t CLK1_CLK4_BYPASS_CNTL; //dtbclk bypass +}; + struct dcn301_clk_internal { int dummy; uint32_t CLK1_CLK0_CURRENT_CNT; //dispclk @@ -314,6 +333,7 @@ struct clk_mgr { bool force_smu_not_present; bool dc_mode_softmax_enabled; int dprefclk_khz; // Used by program pixel clock in clock source funcs, need to figureout where this goes + int dp_dto_source_clock_in_khz; // Used to program DP DTO with ss adjustment on DCN314 int dentist_vco_freq_khz; struct clk_state_registers_and_bypass boot_snapshot; struct clk_bw_params *bw_params; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h index b95ae9596c3b..dcae23faeee3 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h @@ -43,6 +43,7 @@ * to be used inside loops and for determining array sizes. */ #define MAX_PIPES 6 +#define MAX_PHANTOM_PIPES (MAX_PIPES / 2) #define MAX_DIG_LINK_ENCODERS 7 #define MAX_DWB_PIPES 1 #define MAX_HPO_DP2_ENCODERS 4 diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h index 248adc1705e3..5dcbaa2db964 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h @@ -40,6 +40,7 @@ struct panel_cntl_backlight_registers { unsigned int BL_PWM_PERIOD_CNTL; unsigned int LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV; unsigned int PANEL_PWRSEQ_REF_DIV2; + unsigned int USER_LEVEL; }; struct panel_cntl_funcs { diff --git a/drivers/gpu/drm/amd/display/dc/inc/link.h b/drivers/gpu/drm/amd/display/dc/inc/link.h index 7439865d1b50..26fe81f213da 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link.h @@ -289,6 +289,8 @@ struct link_service { bool (*edp_replay_residency)(const struct dc_link *link, unsigned int *residency, const bool is_start, const bool is_alpm); + bool (*edp_set_replay_power_opt_and_coasting_vtotal)(struct dc_link *link, + const unsigned int *power_opts, uint16_t coasting_vtotal); bool (*edp_wait_for_t12)(struct dc_link *link); bool (*edp_is_ilr_optimization_required)(struct dc_link *link, diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index 0458d2d749f4..c958ef37b78a 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -573,9 +573,6 @@ void update_audio_usage( unsigned int resource_pixel_format_to_bpp(enum surface_pixel_format format); -void get_audio_check(struct audio_info *aud_modes, - struct audio_check *aud_chk); - bool get_temp_dp_link_res(struct dc_link *link, struct link_resource *link_res, struct dc_link_settings *link_settings); @@ -622,5 +619,4 @@ enum dc_status update_dp_encoder_resources_for_test_harness(const struct dc *dc, struct pipe_ctx *pipe_ctx); bool check_subvp_sw_cursor_fallback_req(const struct dc *dc, struct dc_stream_state *stream); - #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c index 5fe8b4871c77..3cbfbf8d107e 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c @@ -900,11 +900,15 @@ bool link_set_dsc_pps_packet(struct pipe_ctx *pipe_ctx, bool enable, bool immedi { struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; struct dc_stream_state *stream = pipe_ctx->stream; - DC_LOGGER_INIT(dsc->ctx->logger); - if (!pipe_ctx->stream->timing.flags.DSC || !dsc) + if (!pipe_ctx->stream->timing.flags.DSC) return false; + if (!dsc) + return false; + + DC_LOGGER_INIT(dsc->ctx->logger); + if (enable) { struct dsc_config dsc_cfg; uint8_t dsc_packed_pps[128]; @@ -2005,17 +2009,11 @@ static enum dc_status enable_link_dp(struct dc_state *state, } } - /* - * If the link is DP-over-USB4 do the following: - * - Train with fallback when enabling DPIA link. Conventional links are + /* Train with fallback when enabling DPIA link. Conventional links are * trained with fallback during sink detection. - * - Allocate only what the stream needs for bw in Gbps. Inform the CM - * in case stream needs more or less bw from what has been allocated - * earlier at plug time. */ - if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) { + if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) do_fallback = true; - } /* * Temporary w/a to get DP2.0 link rates to work with SST. @@ -2197,6 +2195,32 @@ static enum dc_status enable_link( return status; } +static bool allocate_usb4_bandwidth_for_stream(struct dc_stream_state *stream, int bw) +{ + return true; +} + +static bool allocate_usb4_bandwidth(struct dc_stream_state *stream) +{ + bool ret; + + int bw = dc_bandwidth_in_kbps_from_timing(&stream->timing, + dc_link_get_highest_encoding_format(stream->sink->link)); + + ret = allocate_usb4_bandwidth_for_stream(stream, bw); + + return ret; +} + +static bool deallocate_usb4_bandwidth(struct dc_stream_state *stream) +{ + bool ret; + + ret = allocate_usb4_bandwidth_for_stream(stream, 0); + + return ret; +} + void link_set_dpms_off(struct pipe_ctx *pipe_ctx) { struct dc *dc = pipe_ctx->stream->ctx->dc; @@ -2232,6 +2256,9 @@ void link_set_dpms_off(struct pipe_ctx *pipe_ctx) update_psp_stream_config(pipe_ctx, true); dc->hwss.blank_stream(pipe_ctx); + if (pipe_ctx->stream->link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) + deallocate_usb4_bandwidth(pipe_ctx->stream); + if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) deallocate_mst_payload(pipe_ctx); else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT && @@ -2474,6 +2501,9 @@ void link_set_dpms_on( } } + if (pipe_ctx->stream->link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) + allocate_usb4_bandwidth(pipe_ctx->stream); + if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) allocate_mst_payload(pipe_ctx); else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT && diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c index 5464d8d26bd3..37d3027c32dc 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c @@ -216,6 +216,7 @@ static void construct_link_service_edp_panel_control(struct link_service *link_s link_srv->edp_send_replay_cmd = edp_send_replay_cmd; link_srv->edp_set_coasting_vtotal = edp_set_coasting_vtotal; link_srv->edp_replay_residency = edp_replay_residency; + link_srv->edp_set_replay_power_opt_and_coasting_vtotal = edp_set_replay_power_opt_and_coasting_vtotal; link_srv->edp_wait_for_t12 = edp_wait_for_t12; link_srv->edp_is_ilr_optimization_required = diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.c b/drivers/gpu/drm/amd/display/dc/link/link_validation.c index b45fda96eaf6..8fe66c367850 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.c @@ -346,23 +346,61 @@ enum dc_status link_validate_mode_timing( return DC_OK; } +/* + * This function calculates the bandwidth required for the stream timing + * and aggregates the stream bandwidth for the respective dpia link + * + * @stream: pointer to the dc_stream_state struct instance + * @num_streams: number of streams to be validated + * + * return: true if validation is succeeded + */ bool link_validate_dpia_bandwidth(const struct dc_stream_state *stream, const unsigned int num_streams) { - bool ret = true; - int bw_needed[MAX_DPIA_NUM]; - struct dc_link *link[MAX_DPIA_NUM]; - - if (!num_streams || num_streams > MAX_DPIA_NUM) - return ret; + int bw_needed[MAX_DPIA_NUM] = {0}; + struct dc_link *dpia_link[MAX_DPIA_NUM] = {0}; + int num_dpias = 0; for (uint8_t i = 0; i < num_streams; ++i) { + if (stream[i].signal == SIGNAL_TYPE_DISPLAY_PORT) { + /* new dpia sst stream, check whether it exceeds max dpia */ + if (num_dpias >= MAX_DPIA_NUM) + return false; - link[i] = stream[i].link; - bw_needed[i] = dc_bandwidth_in_kbps_from_timing(&stream[i].timing, - dc_link_get_highest_encoding_format(link[i])); + dpia_link[num_dpias] = stream[i].link; + bw_needed[num_dpias] = dc_bandwidth_in_kbps_from_timing(&stream[i].timing, + dc_link_get_highest_encoding_format(dpia_link[num_dpias])); + num_dpias++; + } else if (stream[i].signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + uint8_t j = 0; + /* check whether its a known dpia link */ + for (; j < num_dpias; ++j) { + if (dpia_link[j] == stream[i].link) + break; + } + + if (j == num_dpias) { + /* new dpia mst stream, check whether it exceeds max dpia */ + if (num_dpias >= MAX_DPIA_NUM) + return false; + else { + dpia_link[j] = stream[i].link; + num_dpias++; + } + } + + bw_needed[j] += dc_bandwidth_in_kbps_from_timing(&stream[i].timing, + dc_link_get_highest_encoding_format(dpia_link[j])); + } } - ret = dpia_validate_usb4_bw(link, bw_needed, num_streams); + /* Include dp overheads */ + for (uint8_t i = 0; i < num_dpias; ++i) { + int dp_overhead = 0; + + dp_overhead = link_dp_dpia_get_dp_overhead_in_dp_tunneling(dpia_link[i]); + bw_needed[i] += dp_overhead; + } - return ret; + return dpia_validate_usb4_bw(dpia_link, bw_needed, num_dpias); } diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.h b/drivers/gpu/drm/amd/display/dc/link/link_validation.h index 4a954317d0da..595fb05946e9 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.h +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.h @@ -25,6 +25,7 @@ #ifndef __LINK_VALIDATION_H__ #define __LINK_VALIDATION_H__ #include "link.h" + enum dc_status link_validate_mode_timing( const struct dc_stream_state *stream, struct dc_link *link, diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c index 3c5334cdb3fb..289f5d133342 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c @@ -1398,7 +1398,7 @@ static bool get_usbc_cable_id(struct dc_link *link, union dp_cable_id *cable_id) cmd.cable_id.header.payload_bytes = sizeof(cmd.cable_id.data); cmd.cable_id.data.input.phy_inst = resource_transmitter_to_phy_idx( link->dc, link->link_enc->transmitter); - if (dm_execute_dmub_cmd(link->dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && + if (dc_wake_and_execute_dmub_cmd(link->dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && cmd.cable_id.header.ret_status == 1) { cable_id->raw = cmd.cable_id.data.output_raw; DC_LOG_DC("usbc_cable_id = %d.\n", cable_id->raw); diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c index 0bb749133909..6af42ba9885c 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c @@ -82,24 +82,33 @@ bool dpia_query_hpd_status(struct dc_link *link) { union dmub_rb_cmd cmd = {0}; struct dc_dmub_srv *dmub_srv = link->ctx->dmub_srv; - bool is_hpd_high = false; /* prepare QUERY_HPD command */ cmd.query_hpd.header.type = DMUB_CMD__QUERY_HPD_STATE; cmd.query_hpd.data.instance = link->link_id.enum_id - ENUM_ID_1; cmd.query_hpd.data.ch_type = AUX_CHANNEL_DPIA; - /* Return HPD status reported by DMUB if query successfully executed. */ - if (dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && cmd.query_hpd.data.status == AUX_RET_SUCCESS) - is_hpd_high = cmd.query_hpd.data.result; - - DC_LOG_DEBUG("%s: link(%d) dpia(%d) cmd_status(%d) result(%d)\n", - __func__, - link->link_index, - link->link_id.enum_id - ENUM_ID_1, - cmd.query_hpd.data.status, - cmd.query_hpd.data.result); - - return is_hpd_high; + /* Query dpia hpd status from dmub */ + if (dc_wake_and_execute_dmub_cmd(dmub_srv->ctx, &cmd, + DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && + cmd.query_hpd.data.status == AUX_RET_SUCCESS) { + DC_LOG_DEBUG("%s: for link(%d) dpia(%d) success, current_hpd_status(%d) new_hpd_status(%d)\n", + __func__, + link->link_index, + link->link_id.enum_id - ENUM_ID_1, + link->hpd_status, + cmd.query_hpd.data.result); + link->hpd_status = cmd.query_hpd.data.result; + } else { + DC_LOG_ERROR("%s: for link(%d) dpia(%d) failed with status(%d), current_hpd_status(%d) new_hpd_status(0)\n", + __func__, + link->link_index, + link->link_id.enum_id - ENUM_ID_1, + cmd.query_hpd.data.status, + link->hpd_status); + link->hpd_status = false; + } + + return link->hpd_status; } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c index 7581023daa47..dd0d2b206462 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c @@ -50,15 +50,28 @@ static bool get_bw_alloc_proceed_flag(struct dc_link *tmp) && tmp->hpd_status && tmp->dpia_bw_alloc_config.bw_alloc_enabled); } + static void reset_bw_alloc_struct(struct dc_link *link) { link->dpia_bw_alloc_config.bw_alloc_enabled = false; - link->dpia_bw_alloc_config.sink_verified_bw = 0; - link->dpia_bw_alloc_config.sink_max_bw = 0; + link->dpia_bw_alloc_config.link_verified_bw = 0; + link->dpia_bw_alloc_config.link_max_bw = 0; + link->dpia_bw_alloc_config.allocated_bw = 0; link->dpia_bw_alloc_config.estimated_bw = 0; link->dpia_bw_alloc_config.bw_granularity = 0; + link->dpia_bw_alloc_config.dp_overhead = 0; link->dpia_bw_alloc_config.response_ready = false; + link->dpia_bw_alloc_config.nrd_max_lane_count = 0; + link->dpia_bw_alloc_config.nrd_max_link_rate = 0; + for (int i = 0; i < MAX_SINKS_PER_LINK; i++) + link->dpia_bw_alloc_config.remote_sink_req_bw[i] = 0; + DC_LOG_DEBUG("reset usb4 bw alloc of link(%d)\n", link->link_index); } + +#define BW_GRANULARITY_0 4 // 0.25 Gbps +#define BW_GRANULARITY_1 2 // 0.5 Gbps +#define BW_GRANULARITY_2 1 // 1 Gbps + static uint8_t get_bw_granularity(struct dc_link *link) { uint8_t bw_granularity = 0; @@ -71,16 +84,20 @@ static uint8_t get_bw_granularity(struct dc_link *link) switch (bw_granularity & 0x3) { case 0: - bw_granularity = 4; + bw_granularity = BW_GRANULARITY_0; break; case 1: + bw_granularity = BW_GRANULARITY_1; + break; + case 2: default: - bw_granularity = 2; + bw_granularity = BW_GRANULARITY_2; break; } return bw_granularity; } + static int get_estimated_bw(struct dc_link *link) { uint8_t bw_estimated_bw = 0; @@ -93,31 +110,33 @@ static int get_estimated_bw(struct dc_link *link) return bw_estimated_bw * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); } -static bool allocate_usb4_bw(int *stream_allocated_bw, int bw_needed, struct dc_link *link) + +static int get_non_reduced_max_link_rate(struct dc_link *link) { - if (bw_needed > 0) - *stream_allocated_bw += bw_needed; + uint8_t nrd_max_link_rate = 0; - return true; + core_link_read_dpcd( + link, + DP_TUNNELING_MAX_LINK_RATE, + &nrd_max_link_rate, + sizeof(uint8_t)); + + return nrd_max_link_rate; } -static bool deallocate_usb4_bw(int *stream_allocated_bw, int bw_to_dealloc, struct dc_link *link) -{ - bool ret = false; - if (*stream_allocated_bw > 0) { - *stream_allocated_bw -= bw_to_dealloc; - ret = true; - } else { - //Do nothing for now - ret = true; - } +static int get_non_reduced_max_lane_count(struct dc_link *link) +{ + uint8_t nrd_max_lane_count = 0; - // Unplug so reset values - if (!link->hpd_status) - reset_bw_alloc_struct(link); + core_link_read_dpcd( + link, + DP_TUNNELING_MAX_LANE_COUNT, + &nrd_max_lane_count, + sizeof(uint8_t)); - return ret; + return nrd_max_lane_count; } + /* * Read all New BW alloc configuration ex: estimated_bw, allocated_bw, * granuality, Driver_ID, CM_Group, & populate the BW allocation structs @@ -125,10 +144,22 @@ static bool deallocate_usb4_bw(int *stream_allocated_bw, int bw_to_dealloc, stru */ static void init_usb4_bw_struct(struct dc_link *link) { - // Init the known values + reset_bw_alloc_struct(link); + + /* init the known values */ link->dpia_bw_alloc_config.bw_granularity = get_bw_granularity(link); link->dpia_bw_alloc_config.estimated_bw = get_estimated_bw(link); + link->dpia_bw_alloc_config.nrd_max_link_rate = get_non_reduced_max_link_rate(link); + link->dpia_bw_alloc_config.nrd_max_lane_count = get_non_reduced_max_lane_count(link); + + DC_LOG_DEBUG("%s: bw_granularity(%d), estimated_bw(%d)\n", + __func__, link->dpia_bw_alloc_config.bw_granularity, + link->dpia_bw_alloc_config.estimated_bw); + DC_LOG_DEBUG("%s: nrd_max_link_rate(%d), nrd_max_lane_count(%d)\n", + __func__, link->dpia_bw_alloc_config.nrd_max_link_rate, + link->dpia_bw_alloc_config.nrd_max_lane_count); } + static uint8_t get_lowest_dpia_index(struct dc_link *link) { const struct dc *dc_struct = link->dc; @@ -141,51 +172,66 @@ static uint8_t get_lowest_dpia_index(struct dc_link *link) dc_struct->links[i]->ep_type != DISPLAY_ENDPOINT_USB4_DPIA) continue; - if (idx > dc_struct->links[i]->link_index) + if (idx > dc_struct->links[i]->link_index) { idx = dc_struct->links[i]->link_index; + break; + } } return idx; } + /* - * Get the Max Available BW or Max Estimated BW for each Host Router + * Get the maximum dp tunnel banwidth of host router * - * @link: pointer to the dc_link struct instance - * @type: ESTIMATD BW or MAX AVAILABLE BW + * @dc: pointer to the dc struct instance + * @hr_index: host router index * - * return: response_ready flag from dc_link struct + * return: host router maximum dp tunnel bandwidth */ -static int get_host_router_total_bw(struct dc_link *link, uint8_t type) +static int get_host_router_total_dp_tunnel_bw(const struct dc *dc, uint8_t hr_index) { - const struct dc *dc_struct = link->dc; - uint8_t lowest_dpia_index = get_lowest_dpia_index(link); - uint8_t idx = (link->link_index - lowest_dpia_index) / 2, idx_temp = 0; - struct dc_link *link_temp; + uint8_t lowest_dpia_index = get_lowest_dpia_index(dc->links[0]); + uint8_t hr_index_temp = 0; + struct dc_link *link_dpia_primary, *link_dpia_secondary; int total_bw = 0; - int i; - - for (i = 0; i < MAX_PIPES * 2; ++i) { - if (!dc_struct->links[i] || dc_struct->links[i]->ep_type != DISPLAY_ENDPOINT_USB4_DPIA) - continue; + for (uint8_t i = 0; i < MAX_PIPES * 2; ++i) { - link_temp = dc_struct->links[i]; - if (!link_temp || !link_temp->hpd_status) + if (!dc->links[i] || dc->links[i]->ep_type != DISPLAY_ENDPOINT_USB4_DPIA) continue; - idx_temp = (link_temp->link_index - lowest_dpia_index) / 2; - - if (idx_temp == idx) { - - if (type == HOST_ROUTER_BW_ESTIMATED) - total_bw += link_temp->dpia_bw_alloc_config.estimated_bw; - else if (type == HOST_ROUTER_BW_ALLOCATED) - total_bw += link_temp->dpia_bw_alloc_config.sink_allocated_bw; + hr_index_temp = (dc->links[i]->link_index - lowest_dpia_index) / 2; + + if (hr_index_temp == hr_index) { + link_dpia_primary = dc->links[i]; + link_dpia_secondary = dc->links[i + 1]; + + /** + * If BW allocation enabled on both DPIAs, then + * HR BW = Estimated(dpia_primary) + Allocated(dpia_secondary) + * otherwise HR BW = Estimated(bw alloc enabled dpia) + */ + if ((link_dpia_primary->hpd_status && + link_dpia_primary->dpia_bw_alloc_config.bw_alloc_enabled) && + (link_dpia_secondary->hpd_status && + link_dpia_secondary->dpia_bw_alloc_config.bw_alloc_enabled)) { + total_bw += link_dpia_primary->dpia_bw_alloc_config.estimated_bw + + link_dpia_secondary->dpia_bw_alloc_config.allocated_bw; + } else if (link_dpia_primary->hpd_status && + link_dpia_primary->dpia_bw_alloc_config.bw_alloc_enabled) { + total_bw = link_dpia_primary->dpia_bw_alloc_config.estimated_bw; + } else if (link_dpia_secondary->hpd_status && + link_dpia_secondary->dpia_bw_alloc_config.bw_alloc_enabled) { + total_bw += link_dpia_secondary->dpia_bw_alloc_config.estimated_bw; + } + break; } } return total_bw; } + /* * Cleanup function for when the dpia is unplugged to reset struct * and perform any required clean up @@ -194,42 +240,49 @@ static int get_host_router_total_bw(struct dc_link *link, uint8_t type) * * return: none */ -static bool dpia_bw_alloc_unplug(struct dc_link *link) +static void dpia_bw_alloc_unplug(struct dc_link *link) { - if (!link) - return true; - - return deallocate_usb4_bw(&link->dpia_bw_alloc_config.sink_allocated_bw, - link->dpia_bw_alloc_config.sink_allocated_bw, link); + if (link) { + DC_LOG_DEBUG("%s: resetting bw alloc config for link(%d)\n", + __func__, link->link_index); + reset_bw_alloc_struct(link); + } } + static void set_usb4_req_bw_req(struct dc_link *link, int req_bw) { uint8_t requested_bw; uint32_t temp; - // 1. Add check for this corner case #1 - if (req_bw > link->dpia_bw_alloc_config.estimated_bw) + /* Error check whether request bw greater than allocated */ + if (req_bw > link->dpia_bw_alloc_config.estimated_bw) { + DC_LOG_ERROR("%s: Request bw greater than estimated bw for link(%d)\n", + __func__, link->link_index); req_bw = link->dpia_bw_alloc_config.estimated_bw; + } temp = req_bw * link->dpia_bw_alloc_config.bw_granularity; requested_bw = temp / Kbps_TO_Gbps; - // Always make sure to add more to account for floating points + /* Always make sure to add more to account for floating points */ if (temp % Kbps_TO_Gbps) ++requested_bw; - // 2. Add check for this corner case #2 + /* Error check whether requested and allocated are equal */ req_bw = requested_bw * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); - if (req_bw == link->dpia_bw_alloc_config.sink_allocated_bw) - return; + if (req_bw == link->dpia_bw_alloc_config.allocated_bw) { + DC_LOG_ERROR("%s: Request bw equals to allocated bw for link(%d)\n", + __func__, link->link_index); + } - if (core_link_write_dpcd( + link->dpia_bw_alloc_config.response_ready = false; // Reset flag + core_link_write_dpcd( link, REQUESTED_BW, &requested_bw, - sizeof(uint8_t)) == DC_OK) - link->dpia_bw_alloc_config.response_ready = false; // Reset flag + sizeof(uint8_t)); } + /* * Return the response_ready flag from dc_link struct * @@ -241,6 +294,7 @@ static bool get_cm_response_ready_flag(struct dc_link *link) { return link->dpia_bw_alloc_config.response_ready; } + // ------------------------------------------------------------------ // PUBLIC FUNCTIONS // ------------------------------------------------------------------ @@ -277,27 +331,27 @@ bool link_dp_dpia_set_dptx_usb4_bw_alloc_support(struct dc_link *link) DPTX_BW_ALLOCATION_MODE_CONTROL, &response, sizeof(uint8_t)) != DC_OK) { - DC_LOG_DEBUG("%s: **** FAILURE Enabling DPtx BW Allocation Mode Support ***\n", - __func__); + DC_LOG_DEBUG("%s: FAILURE Enabling DPtx BW Allocation Mode Support for link(%d)\n", + __func__, link->link_index); } else { // SUCCESS Enabled DPtx BW Allocation Mode Support - link->dpia_bw_alloc_config.bw_alloc_enabled = true; - DC_LOG_DEBUG("%s: **** SUCCESS Enabling DPtx BW Allocation Mode Support ***\n", - __func__); + DC_LOG_DEBUG("%s: SUCCESS Enabling DPtx BW Allocation Mode Support for link(%d)\n", + __func__, link->link_index); ret = true; init_usb4_bw_struct(link); + link->dpia_bw_alloc_config.bw_alloc_enabled = true; } } out: return ret; } + void dpia_handle_bw_alloc_response(struct dc_link *link, uint8_t bw, uint8_t result) { int bw_needed = 0; int estimated = 0; - int host_router_total_estimated_bw = 0; if (!get_bw_alloc_proceed_flag((link))) return; @@ -306,14 +360,22 @@ void dpia_handle_bw_alloc_response(struct dc_link *link, uint8_t bw, uint8_t res case DPIA_BW_REQ_FAILED: - DC_LOG_DEBUG("%s: *** *** BW REQ FAILURE for DP-TX Request *** ***\n", __func__); + /* + * Ideally, we shouldn't run into this case as we always validate available + * bandwidth and request within that limit + */ + estimated = bw * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); + + DC_LOG_ERROR("%s: BW REQ FAILURE for DP-TX Request for link(%d)\n", + __func__, link->link_index); + DC_LOG_ERROR("%s: current estimated_bw(%d), new estimated_bw(%d)\n", + __func__, link->dpia_bw_alloc_config.estimated_bw, estimated); - // Update the new Estimated BW value updated by CM - link->dpia_bw_alloc_config.estimated_bw = - bw * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); + /* Update the new Estimated BW value updated by CM */ + link->dpia_bw_alloc_config.estimated_bw = estimated; + /* Allocate the previously requested bandwidth */ set_usb4_req_bw_req(link, link->dpia_bw_alloc_config.estimated_bw); - link->dpia_bw_alloc_config.response_ready = false; /* * If FAIL then it is either: @@ -326,68 +388,34 @@ void dpia_handle_bw_alloc_response(struct dc_link *link, uint8_t bw, uint8_t res case DPIA_BW_REQ_SUCCESS: - DC_LOG_DEBUG("%s: *** BW REQ SUCCESS for DP-TX Request ***\n", __func__); - - // 1. SUCCESS 1st time before any Pruning is done - // 2. SUCCESS after prev. FAIL before any Pruning is done - // 3. SUCCESS after Pruning is done but before enabling link - bw_needed = bw * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); - // 1. - if (!link->dpia_bw_alloc_config.sink_allocated_bw) { - - allocate_usb4_bw(&link->dpia_bw_alloc_config.sink_allocated_bw, bw_needed, link); - link->dpia_bw_alloc_config.sink_verified_bw = - link->dpia_bw_alloc_config.sink_allocated_bw; - - // SUCCESS from first attempt - if (link->dpia_bw_alloc_config.sink_allocated_bw > - link->dpia_bw_alloc_config.sink_max_bw) - link->dpia_bw_alloc_config.sink_verified_bw = - link->dpia_bw_alloc_config.sink_max_bw; - } - // 3. - else if (link->dpia_bw_alloc_config.sink_allocated_bw) { - - // Find out how much do we need to de-alloc - if (link->dpia_bw_alloc_config.sink_allocated_bw > bw_needed) - deallocate_usb4_bw(&link->dpia_bw_alloc_config.sink_allocated_bw, - link->dpia_bw_alloc_config.sink_allocated_bw - bw_needed, link); - else - allocate_usb4_bw(&link->dpia_bw_alloc_config.sink_allocated_bw, - bw_needed - link->dpia_bw_alloc_config.sink_allocated_bw, link); - } + DC_LOG_DEBUG("%s: BW REQ SUCCESS for DP-TX Request for link(%d)\n", + __func__, link->link_index); + DC_LOG_DEBUG("%s: current allocated_bw(%d), new allocated_bw(%d)\n", + __func__, link->dpia_bw_alloc_config.allocated_bw, bw_needed); - // 4. If this is the 2nd sink then any unused bw will be reallocated to master DPIA - // => check if estimated_bw changed + link->dpia_bw_alloc_config.allocated_bw = bw_needed; link->dpia_bw_alloc_config.response_ready = true; break; case DPIA_EST_BW_CHANGED: - DC_LOG_DEBUG("%s: *** ESTIMATED BW CHANGED for DP-TX Request ***\n", __func__); - estimated = bw * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); - host_router_total_estimated_bw = get_host_router_total_bw(link, HOST_ROUTER_BW_ESTIMATED); - // 1. If due to unplug of other sink - if (estimated == host_router_total_estimated_bw) { - // First update the estimated & max_bw fields - if (link->dpia_bw_alloc_config.estimated_bw < estimated) - link->dpia_bw_alloc_config.estimated_bw = estimated; - } - // 2. If due to realloc bw btw 2 dpia due to plug OR realloc unused Bw - else { - // We lost estimated bw usually due to plug event of other dpia - link->dpia_bw_alloc_config.estimated_bw = estimated; - } + DC_LOG_DEBUG("%s: ESTIMATED BW CHANGED for link(%d)\n", + __func__, link->link_index); + DC_LOG_DEBUG("%s: current estimated_bw(%d), new estimated_bw(%d)\n", + __func__, link->dpia_bw_alloc_config.estimated_bw, estimated); + + link->dpia_bw_alloc_config.estimated_bw = estimated; break; case DPIA_BW_ALLOC_CAPS_CHANGED: - DC_LOG_DEBUG("%s: *** BW ALLOC CAPABILITY CHANGED for DP-TX Request ***\n", __func__); + DC_LOG_ERROR("%s: BW ALLOC CAPABILITY CHANGED to Disabled for link(%d)\n", + __func__, link->link_index); link->dpia_bw_alloc_config.bw_alloc_enabled = false; break; } @@ -405,21 +433,21 @@ int dpia_handle_usb4_bandwidth_allocation_for_link(struct dc_link *link, int pea if (link->hpd_status && peak_bw > 0) { // If DP over USB4 then we need to check BW allocation - link->dpia_bw_alloc_config.sink_max_bw = peak_bw; - set_usb4_req_bw_req(link, link->dpia_bw_alloc_config.sink_max_bw); + link->dpia_bw_alloc_config.link_max_bw = peak_bw; + set_usb4_req_bw_req(link, link->dpia_bw_alloc_config.link_max_bw); do { - if (!(timeout > 0)) + if (timeout > 0) timeout--; else break; - fsleep(10 * 1000); + msleep(10); } while (!get_cm_response_ready_flag(link)); if (!timeout) ret = 0;// ERROR TIMEOUT waiting for response for allocating bw - else if (link->dpia_bw_alloc_config.sink_allocated_bw > 0) - ret = get_host_router_total_bw(link, HOST_ROUTER_BW_ALLOCATED); + else if (link->dpia_bw_alloc_config.allocated_bw > 0) + ret = link->dpia_bw_alloc_config.allocated_bw; } //2. Cold Unplug else if (!link->hpd_status) @@ -428,65 +456,102 @@ int dpia_handle_usb4_bandwidth_allocation_for_link(struct dc_link *link, int pea out: return ret; } -int link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int req_bw) +bool link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int req_bw) { - int ret = 0; + bool ret = false; uint8_t timeout = 10; + DC_LOG_DEBUG("%s: ENTER: link(%d), hpd_status(%d), current allocated_bw(%d), req_bw(%d)\n", + __func__, link->link_index, link->hpd_status, + link->dpia_bw_alloc_config.allocated_bw, req_bw); + if (!get_bw_alloc_proceed_flag(link)) goto out; - /* - * Sometimes stream uses same timing parameters as the already - * allocated max sink bw so no need to re-alloc - */ - if (req_bw != link->dpia_bw_alloc_config.sink_allocated_bw) { - set_usb4_req_bw_req(link, req_bw); - do { - if (!(timeout > 0)) - timeout--; - else - break; - udelay(10 * 1000); - } while (!get_cm_response_ready_flag(link)); + set_usb4_req_bw_req(link, req_bw); + do { + if (timeout > 0) + timeout--; + else + break; + msleep(10); + } while (!get_cm_response_ready_flag(link)); - if (!timeout) - ret = 0;// ERROR TIMEOUT waiting for response for allocating bw - else if (link->dpia_bw_alloc_config.sink_allocated_bw > 0) - ret = get_host_router_total_bw(link, HOST_ROUTER_BW_ALLOCATED); - } + if (timeout) + ret = true; out: + DC_LOG_DEBUG("%s: EXIT: timeout(%d), ret(%d)\n", __func__, timeout, ret); return ret; } + bool dpia_validate_usb4_bw(struct dc_link **link, int *bw_needed_per_dpia, const unsigned int num_dpias) { bool ret = true; - int bw_needed_per_hr[MAX_HR_NUM] = { 0, 0 }; - uint8_t lowest_dpia_index = 0, dpia_index = 0; - uint8_t i; + int bw_needed_per_hr[MAX_HR_NUM] = { 0, 0 }, host_router_total_dp_bw = 0; + uint8_t lowest_dpia_index, i, hr_index; if (!num_dpias || num_dpias > MAX_DPIA_NUM) return ret; - //Get total Host Router BW & Validate against each Host Router max BW + lowest_dpia_index = get_lowest_dpia_index(link[0]); + + /* get total Host Router BW with granularity for the given modes */ for (i = 0; i < num_dpias; ++i) { + int granularity_Gbps = 0; + int bw_granularity = 0; if (!link[i]->dpia_bw_alloc_config.bw_alloc_enabled) continue; - lowest_dpia_index = get_lowest_dpia_index(link[i]); if (link[i]->link_index < lowest_dpia_index) continue; - dpia_index = (link[i]->link_index - lowest_dpia_index) / 2; - bw_needed_per_hr[dpia_index] += bw_needed_per_dpia[i]; - if (bw_needed_per_hr[dpia_index] > get_host_router_total_bw(link[i], HOST_ROUTER_BW_ALLOCATED)) { + granularity_Gbps = (Kbps_TO_Gbps / link[i]->dpia_bw_alloc_config.bw_granularity); + bw_granularity = (bw_needed_per_dpia[i] / granularity_Gbps) * granularity_Gbps + + ((bw_needed_per_dpia[i] % granularity_Gbps) ? granularity_Gbps : 0); - ret = false; - break; + hr_index = (link[i]->link_index - lowest_dpia_index) / 2; + bw_needed_per_hr[hr_index] += bw_granularity; + } + + /* validate against each Host Router max BW */ + for (hr_index = 0; hr_index < MAX_HR_NUM; ++hr_index) { + if (bw_needed_per_hr[hr_index]) { + host_router_total_dp_bw = get_host_router_total_dp_tunnel_bw(link[0]->dc, hr_index); + if (bw_needed_per_hr[hr_index] > host_router_total_dp_bw) { + ret = false; + break; + } } } return ret; } + +int link_dp_dpia_get_dp_overhead_in_dp_tunneling(struct dc_link *link) +{ + int dp_overhead = 0, link_mst_overhead = 0; + + if (!get_bw_alloc_proceed_flag((link))) + return dp_overhead; + + /* if its mst link, add MTPH overhead */ + if ((link->type == dc_connection_mst_branch) && + !link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED) { + /* For 8b/10b encoding: MTP is 64 time slots long, slot 0 is used for MTPH + * MST overhead is 1/64 of link bandwidth (excluding any overhead) + */ + const struct dc_link_settings *link_cap = + dc_link_get_link_cap(link); + uint32_t link_bw_in_kbps = (uint32_t)link_cap->link_rate * + (uint32_t)link_cap->lane_count * + LINK_RATE_REF_FREQ_IN_KHZ * 8; + link_mst_overhead = (link_bw_in_kbps / 64) + ((link_bw_in_kbps % 64) ? 1 : 0); + } + + /* add all the overheads */ + dp_overhead = link_mst_overhead; + + return dp_overhead; +} diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h index 7292690383ae..3b6d8494f9d5 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h @@ -59,9 +59,9 @@ bool link_dp_dpia_set_dptx_usb4_bw_alloc_support(struct dc_link *link); * @link: pointer to the dc_link struct instance * @req_bw: Bw requested by the stream * - * return: allocated bw else return 0 + * return: true if allocated successfully */ -int link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int req_bw); +bool link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int req_bw); /* * Handle the USB4 BW Allocation related functionality here: @@ -99,4 +99,13 @@ void dpia_handle_bw_alloc_response(struct dc_link *link, uint8_t bw, uint8_t res */ bool dpia_validate_usb4_bw(struct dc_link **link, int *bw_needed, const unsigned int num_dpias); +/* + * Obtain all the DP overheads in dp tunneling for the dpia link + * + * @link: pointer to the dc_link struct instance + * + * return: DP overheads in DP tunneling + */ +int link_dp_dpia_get_dp_overhead_in_dp_tunneling(struct dc_link *link); + #endif /* DC_INC_LINK_DP_DPIA_BW_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c index 90339c2dfd84..5a0b04518956 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c @@ -807,7 +807,7 @@ void dp_decide_lane_settings( const struct link_training_settings *lt_settings, const union lane_adjust ln_adjust[LANE_COUNT_DP_MAX], struct dc_lane_settings hw_lane_settings[LANE_COUNT_DP_MAX], - union dpcd_training_lane dpcd_lane_settings[LANE_COUNT_DP_MAX]) + union dpcd_training_lane *dpcd_lane_settings) { uint32_t lane; diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.h index 7d027bac8255..851bd17317a0 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.h @@ -111,7 +111,7 @@ void dp_decide_lane_settings( const struct link_training_settings *lt_settings, const union lane_adjust ln_adjust[LANE_COUNT_DP_MAX], struct dc_lane_settings hw_lane_settings[LANE_COUNT_DP_MAX], - union dpcd_training_lane dpcd_lane_settings[LANE_COUNT_DP_MAX]); + union dpcd_training_lane *dpcd_lane_settings); enum dc_dp_training_pattern decide_cr_training_pattern( const struct dc_link_settings *link_settings); diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c index bf53a86ea817..046d3e205415 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c @@ -529,6 +529,9 @@ bool edp_set_backlight_level(const struct dc_link *link, if (dc_is_embedded_signal(link->connector_signal)) { struct pipe_ctx *pipe_ctx = get_pipe_from_link(link); + if (link->panel_cntl) + link->panel_cntl->stored_backlight_registers.USER_LEVEL = backlight_pwm_u16_16; + if (pipe_ctx) { /* Disable brightness ramping when the display is blanked * as it can hang the DMCU @@ -927,8 +930,8 @@ bool edp_get_replay_state(const struct dc_link *link, uint64_t *state) bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream) { /* To-do: Setup Replay */ - struct dc *dc = link->ctx->dc; - struct dmub_replay *replay = dc->res_pool->replay; + struct dc *dc; + struct dmub_replay *replay; int i; unsigned int panel_inst; struct replay_context replay_context = { 0 }; @@ -944,6 +947,10 @@ bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream if (!link) return false; + dc = link->ctx->dc; + + replay = dc->res_pool->replay; + if (!replay) return false; @@ -972,8 +979,7 @@ bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream replay_context.line_time_in_ns = lineTimeInNs; - if (replay) - link->replay_settings.replay_feature_enabled = + link->replay_settings.replay_feature_enabled = replay->funcs->replay_copy_settings(replay, link, &replay_context, panel_inst); if (link->replay_settings.replay_feature_enabled) { @@ -1065,6 +1071,33 @@ bool edp_replay_residency(const struct dc_link *link, return true; } +bool edp_set_replay_power_opt_and_coasting_vtotal(struct dc_link *link, + const unsigned int *power_opts, uint16_t coasting_vtotal) +{ + struct dc *dc = link->ctx->dc; + struct dmub_replay *replay = dc->res_pool->replay; + unsigned int panel_inst; + + if (!dc_get_edp_link_panel_inst(dc, link, &panel_inst)) + return false; + + /* Only both power and coasting vtotal changed, this func could return true */ + if (power_opts && link->replay_settings.replay_power_opt_active != *power_opts && + coasting_vtotal && link->replay_settings.coasting_vtotal != coasting_vtotal) { + if (link->replay_settings.replay_feature_enabled && + replay->funcs->replay_set_power_opt_and_coasting_vtotal) { + replay->funcs->replay_set_power_opt_and_coasting_vtotal(replay, + *power_opts, panel_inst, coasting_vtotal); + link->replay_settings.replay_power_opt_active = *power_opts; + link->replay_settings.coasting_vtotal = coasting_vtotal; + } else + return false; + } else + return false; + + return true; +} + static struct abm *get_abm_from_stream_res(const struct dc_link *link) { int i; diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h index b7493ff4fcee..34e521af7bb4 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h @@ -63,6 +63,8 @@ bool edp_set_coasting_vtotal(struct dc_link *link, uint16_t coasting_vtotal); bool edp_replay_residency(const struct dc_link *link, unsigned int *residency, const bool is_start, const bool is_alpm); bool edp_get_replay_state(const struct dc_link *link, uint64_t *state); +bool edp_set_replay_power_opt_and_coasting_vtotal(struct dc_link *link, + const unsigned int *power_opts, uint16_t coasting_vtotal); bool edp_wait_for_t12(struct dc_link *link); bool edp_is_ilr_optimization_required(struct dc_link *link, struct dc_crtc_timing *crtc_timing); diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c index a2c4db2cebdd..823493543325 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c @@ -166,6 +166,16 @@ static bool optc32_disable_crtc(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); + REG_UPDATE_5(OPTC_DATA_SOURCE_SELECT, + OPTC_SEG0_SRC_SEL, 0xf, + OPTC_SEG1_SRC_SEL, 0xf, + OPTC_SEG2_SRC_SEL, 0xf, + OPTC_SEG3_SRC_SEL, 0xf, + OPTC_NUM_OF_INPUT_SEGMENT, 0); + + REG_UPDATE(OPTC_MEMORY_CONFIG, + OPTC_MEM_SEL, 0); + /* disable otg request until end of the first line * in the vertical blank region */ @@ -198,6 +208,13 @@ static void optc32_disable_phantom_otg(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); + REG_UPDATE_5(OPTC_DATA_SOURCE_SELECT, + OPTC_SEG0_SRC_SEL, 0xf, + OPTC_SEG1_SRC_SEL, 0xf, + OPTC_SEG2_SRC_SEL, 0xf, + OPTC_SEG3_SRC_SEL, 0xf, + OPTC_NUM_OF_INPUT_SEGMENT, 0); + REG_UPDATE(OTG_CONTROL, OTG_MASTER_EN, 0); } diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c index a4a39f1638cf..5b1547508850 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c @@ -138,6 +138,16 @@ static bool optc35_disable_crtc(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); + REG_UPDATE_5(OPTC_DATA_SOURCE_SELECT, + OPTC_SEG0_SRC_SEL, 0xf, + OPTC_SEG1_SRC_SEL, 0xf, + OPTC_SEG2_SRC_SEL, 0xf, + OPTC_SEG3_SRC_SEL, 0xf, + OPTC_NUM_OF_INPUT_SEGMENT, 0); + + REG_UPDATE(OPTC_MEMORY_CONFIG, + OPTC_MEM_SEL, 0); + /* disable otg request until end of the first line * in the vertical blank region */ diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c index f6cbcc9b4006..c4d71e7f18af 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c @@ -89,6 +89,8 @@ #include "dcn20/dcn20_vmid.h" #include "dml/dcn32/dcn32_fpu.h" +#include "dc_state_priv.h" + #include "dml2/dml2_wrapper.h" #define DC_LOGGER_INIT(logger) @@ -1644,7 +1646,7 @@ static void dcn32_enable_phantom_plane(struct dc *dc, if (curr_pipe->top_pipe && curr_pipe->top_pipe->plane_state == curr_pipe->plane_state) phantom_plane = prev_phantom_plane; else - phantom_plane = dc_create_plane_state(dc); + phantom_plane = dc_state_create_phantom_plane(dc, context, curr_pipe->plane_state); memcpy(&phantom_plane->address, &curr_pipe->plane_state->address, sizeof(phantom_plane->address)); memcpy(&phantom_plane->scaling_quality, &curr_pipe->plane_state->scaling_quality, @@ -1665,9 +1667,7 @@ static void dcn32_enable_phantom_plane(struct dc *dc, phantom_plane->clip_rect.y = 0; phantom_plane->clip_rect.height = phantom_stream->src.height; - phantom_plane->is_phantom = true; - - dc_add_plane_to_context(dc, phantom_stream, phantom_plane, context); + dc_state_add_phantom_plane(dc, phantom_stream, phantom_plane, context); curr_pipe = curr_pipe->bottom_pipe; prev_phantom_plane = phantom_plane; @@ -1683,13 +1683,7 @@ static struct dc_stream_state *dcn32_enable_phantom_stream(struct dc *dc, struct dc_stream_state *phantom_stream = NULL; struct pipe_ctx *ref_pipe = &context->res_ctx.pipe_ctx[dc_pipe_idx]; - phantom_stream = dc_create_stream_for_sink(ref_pipe->stream->sink); - phantom_stream->signal = SIGNAL_TYPE_VIRTUAL; - phantom_stream->dpms_off = true; - phantom_stream->mall_stream_config.type = SUBVP_PHANTOM; - phantom_stream->mall_stream_config.paired_stream = ref_pipe->stream; - ref_pipe->stream->mall_stream_config.type = SUBVP_MAIN; - ref_pipe->stream->mall_stream_config.paired_stream = phantom_stream; + phantom_stream = dc_state_create_phantom_stream(dc, context, ref_pipe->stream); /* stream has limited viewport and small timing */ memcpy(&phantom_stream->timing, &ref_pipe->stream->timing, sizeof(phantom_stream->timing)); @@ -1699,81 +1693,10 @@ static struct dc_stream_state *dcn32_enable_phantom_stream(struct dc *dc, dcn32_set_phantom_stream_timing(dc, context, ref_pipe, phantom_stream, pipes, pipe_cnt, dc_pipe_idx); DC_FP_END(); - dc_add_stream_to_ctx(dc, context, phantom_stream); + dc_state_add_phantom_stream(dc, context, phantom_stream, ref_pipe->stream); return phantom_stream; } -void dcn32_retain_phantom_pipes(struct dc *dc, struct dc_state *context) -{ - int i; - struct dc_plane_state *phantom_plane = NULL; - struct dc_stream_state *phantom_stream = NULL; - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - - if (resource_is_pipe_type(pipe, OTG_MASTER) && - resource_is_pipe_type(pipe, DPP_PIPE) && - pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { - phantom_plane = pipe->plane_state; - phantom_stream = pipe->stream; - - dc_plane_state_retain(phantom_plane); - dc_stream_retain(phantom_stream); - } - } -} - -// return true if removed piped from ctx, false otherwise -bool dcn32_remove_phantom_pipes(struct dc *dc, struct dc_state *context, bool fast_update) -{ - int i; - bool removed_pipe = false; - struct dc_plane_state *phantom_plane = NULL; - struct dc_stream_state *phantom_stream = NULL; - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - // build scaling params for phantom pipes - if (pipe->plane_state && pipe->stream && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { - phantom_plane = pipe->plane_state; - phantom_stream = pipe->stream; - - dc_rem_all_planes_for_stream(dc, pipe->stream, context); - dc_remove_stream_from_ctx(dc, context, pipe->stream); - - /* Ref count is incremented on allocation and also when added to the context. - * Therefore we must call release for the the phantom plane and stream once - * they are removed from the ctx to finally decrement the refcount to 0 to free. - */ - dc_plane_state_release(phantom_plane); - dc_stream_release(phantom_stream); - - removed_pipe = true; - } - - /* For non-full updates, a shallow copy of the current state - * is created. In this case we don't want to erase the current - * state (there can be 2 HIRQL threads, one in flip, and one in - * checkMPO) that can cause a race condition. - * - * This is just a workaround, needs a proper fix. - */ - if (!fast_update) { - // Clear all phantom stream info - if (pipe->stream) { - pipe->stream->mall_stream_config.type = SUBVP_NONE; - pipe->stream->mall_stream_config.paired_stream = NULL; - } - - if (pipe->plane_state) { - pipe->plane_state->is_phantom = false; - } - } - } - return removed_pipe; -} - /* TODO: Input to this function should indicate which pipe indexes (or streams) * require a phantom pipe / stream */ @@ -1798,7 +1721,7 @@ void dcn32_add_phantom_pipes(struct dc *dc, struct dc_state *context, // We determine which phantom pipes were added by comparing with // the phantom stream. if (pipe->plane_state && pipe->stream && pipe->stream == phantom_stream && - pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { + dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) { pipe->stream->use_dynamic_meta = false; pipe->plane_state->flip_immediate = false; if (!resource_build_scaling_params(pipe)) { @@ -1817,7 +1740,6 @@ static bool dml1_validate(struct dc *dc, struct dc_state *context, bool fast_val int vlevel = 0; int pipe_cnt = 0; display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_KERNEL); - struct mall_temp_config mall_temp_config; /* To handle Freesync properly, setting FreeSync DML parameters * to its default state for the first stage of validation @@ -1827,29 +1749,12 @@ static bool dml1_validate(struct dc *dc, struct dc_state *context, bool fast_val DC_LOGGER_INIT(dc->ctx->logger); - /* For fast validation, there are situations where a shallow copy of - * of the dc->current_state is created for the validation. In this case - * we want to save and restore the mall config because we always - * teardown subvp at the beginning of validation (and don't attempt - * to add it back if it's fast validation). If we don't restore the - * subvp config in cases of fast validation + shallow copy of the - * dc->current_state, the dc->current_state will have a partially - * removed subvp state when we did not intend to remove it. - */ - if (fast_validate) { - memset(&mall_temp_config, 0, sizeof(mall_temp_config)); - dcn32_save_mall_state(dc, context, &mall_temp_config); - } - BW_VAL_TRACE_COUNT(); DC_FP_START(); out = dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate); DC_FP_END(); - if (fast_validate) - dcn32_restore_mall_state(dc, context, &mall_temp_config); - if (pipe_cnt == 0) goto validate_out; @@ -1933,7 +1838,7 @@ int dcn32_populate_dml_pipes_from_context( * This is just a workaround -- needs a proper fix. */ if (!fast_validate) { - switch (pipe->stream->mall_stream_config.type) { + switch (dc_state_get_pipe_subvp_type(context, pipe)) { case SUBVP_MAIN: pipes[pipe_cnt].pipe.src.use_mall_for_pstate_change = dm_use_mall_pstate_change_sub_viewport; subvp_in_use = true; @@ -2037,10 +1942,6 @@ static struct resource_funcs dcn32_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .update_soc_for_wm_a = dcn30_update_soc_for_wm_a, .add_phantom_pipes = dcn32_add_phantom_pipes, - .remove_phantom_pipes = dcn32_remove_phantom_pipes, - .retain_phantom_pipes = dcn32_retain_phantom_pipes, - .save_mall_state = dcn32_save_mall_state, - .restore_mall_state = dcn32_restore_mall_state, .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params, }; @@ -2454,16 +2355,19 @@ static bool dcn32_resource_construct( dc->dml2_options.callbacks.get_opp_head = &resource_get_opp_head; dc->dml2_options.svp_pstate.callbacks.dc = dc; - dc->dml2_options.svp_pstate.callbacks.add_plane_to_context = &dc_add_plane_to_context; - dc->dml2_options.svp_pstate.callbacks.add_stream_to_ctx = &dc_add_stream_to_ctx; + dc->dml2_options.svp_pstate.callbacks.add_phantom_plane = &dc_state_add_phantom_plane; + dc->dml2_options.svp_pstate.callbacks.add_phantom_stream = &dc_state_add_phantom_stream; dc->dml2_options.svp_pstate.callbacks.build_scaling_params = &resource_build_scaling_params; - dc->dml2_options.svp_pstate.callbacks.create_plane = &dc_create_plane_state; - dc->dml2_options.svp_pstate.callbacks.remove_plane_from_context = &dc_remove_plane_from_context; - dc->dml2_options.svp_pstate.callbacks.remove_stream_from_ctx = &dc_remove_stream_from_ctx; - dc->dml2_options.svp_pstate.callbacks.create_stream_for_sink = &dc_create_stream_for_sink; - dc->dml2_options.svp_pstate.callbacks.plane_state_release = &dc_plane_state_release; - dc->dml2_options.svp_pstate.callbacks.stream_release = &dc_stream_release; + dc->dml2_options.svp_pstate.callbacks.create_phantom_plane = &dc_state_create_phantom_plane; + dc->dml2_options.svp_pstate.callbacks.remove_phantom_plane = &dc_state_remove_phantom_plane; + dc->dml2_options.svp_pstate.callbacks.remove_phantom_stream = &dc_state_remove_phantom_stream; + dc->dml2_options.svp_pstate.callbacks.create_phantom_stream = &dc_state_create_phantom_stream; + dc->dml2_options.svp_pstate.callbacks.release_phantom_plane = &dc_state_release_phantom_plane; + dc->dml2_options.svp_pstate.callbacks.release_phantom_stream = &dc_state_release_phantom_stream; dc->dml2_options.svp_pstate.callbacks.release_dsc = &dcn20_release_dsc; + dc->dml2_options.svp_pstate.callbacks.get_pipe_subvp_type = &dc_state_get_pipe_subvp_type; + dc->dml2_options.svp_pstate.callbacks.get_stream_subvp_type = &dc_state_get_stream_subvp_type; + dc->dml2_options.svp_pstate.callbacks.get_paired_subvp_stream = &dc_state_get_paired_subvp_stream; dc->dml2_options.svp_pstate.subvp_fw_processing_delay_us = dc->caps.subvp_fw_processing_delay_us; dc->dml2_options.svp_pstate.subvp_prefetch_end_to_mall_start_us = dc->caps.subvp_prefetch_end_to_mall_start_us; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h index 9ca799da1a56..0c87b0fabba7 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h @@ -91,12 +91,6 @@ bool dcn32_release_post_bldn_3dlut( struct dc_3dlut **lut, struct dc_transfer_func **shaper); -bool dcn32_remove_phantom_pipes(struct dc *dc, - struct dc_state *context, bool fast_update); - -void dcn32_retain_phantom_pipes(struct dc *dc, - struct dc_state *context); - void dcn32_add_phantom_pipes(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, @@ -169,15 +163,7 @@ void dcn32_determine_det_override(struct dc *dc, void dcn32_set_det_allocations(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes); -void dcn32_save_mall_state(struct dc *dc, - struct dc_state *context, - struct mall_temp_config *temp_config); - -void dcn32_restore_mall_state(struct dc *dc, - struct dc_state *context, - struct mall_temp_config *temp_config); - -struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch(struct dc *dc, const struct dc_state *context); +struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch(struct dc *dc, struct dc_state *context); bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c index 12986fe0b289..74412e5f03fe 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c @@ -92,6 +92,8 @@ #include "vm_helper.h" #include "dcn20/dcn20_vmid.h" +#include "dc_state_priv.h" + #define DC_LOGGER_INIT(logger) enum dcn321_clk_src_array_id { @@ -1605,10 +1607,6 @@ static struct resource_funcs dcn321_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .update_soc_for_wm_a = dcn30_update_soc_for_wm_a, .add_phantom_pipes = dcn32_add_phantom_pipes, - .remove_phantom_pipes = dcn32_remove_phantom_pipes, - .retain_phantom_pipes = dcn32_retain_phantom_pipes, - .save_mall_state = dcn32_save_mall_state, - .restore_mall_state = dcn32_restore_mall_state, .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params, }; @@ -2008,16 +2006,19 @@ static bool dcn321_resource_construct( dc->dml2_options.callbacks.get_opp_head = &resource_get_opp_head; dc->dml2_options.svp_pstate.callbacks.dc = dc; - dc->dml2_options.svp_pstate.callbacks.add_plane_to_context = &dc_add_plane_to_context; - dc->dml2_options.svp_pstate.callbacks.add_stream_to_ctx = &dc_add_stream_to_ctx; + dc->dml2_options.svp_pstate.callbacks.add_phantom_plane = &dc_state_add_phantom_plane; + dc->dml2_options.svp_pstate.callbacks.add_phantom_stream = &dc_state_add_phantom_stream; dc->dml2_options.svp_pstate.callbacks.build_scaling_params = &resource_build_scaling_params; - dc->dml2_options.svp_pstate.callbacks.create_plane = &dc_create_plane_state; - dc->dml2_options.svp_pstate.callbacks.remove_plane_from_context = &dc_remove_plane_from_context; - dc->dml2_options.svp_pstate.callbacks.remove_stream_from_ctx = &dc_remove_stream_from_ctx; - dc->dml2_options.svp_pstate.callbacks.create_stream_for_sink = &dc_create_stream_for_sink; - dc->dml2_options.svp_pstate.callbacks.plane_state_release = &dc_plane_state_release; - dc->dml2_options.svp_pstate.callbacks.stream_release = &dc_stream_release; + dc->dml2_options.svp_pstate.callbacks.create_phantom_plane = &dc_state_create_phantom_plane; + dc->dml2_options.svp_pstate.callbacks.remove_phantom_plane = &dc_state_remove_phantom_plane; + dc->dml2_options.svp_pstate.callbacks.remove_phantom_stream = &dc_state_remove_phantom_stream; + dc->dml2_options.svp_pstate.callbacks.create_phantom_stream = &dc_state_create_phantom_stream; + dc->dml2_options.svp_pstate.callbacks.release_phantom_plane = &dc_state_release_phantom_plane; + dc->dml2_options.svp_pstate.callbacks.release_phantom_stream = &dc_state_release_phantom_stream; dc->dml2_options.svp_pstate.callbacks.release_dsc = &dcn20_release_dsc; + dc->dml2_options.svp_pstate.callbacks.get_pipe_subvp_type = &dc_state_get_pipe_subvp_type; + dc->dml2_options.svp_pstate.callbacks.get_stream_subvp_type = &dc_state_get_stream_subvp_type; + dc->dml2_options.svp_pstate.callbacks.get_paired_subvp_stream = &dc_state_get_paired_subvp_stream; dc->dml2_options.svp_pstate.subvp_fw_processing_delay_us = dc->caps.subvp_fw_processing_delay_us; dc->dml2_options.svp_pstate.subvp_prefetch_end_to_mall_start_us = dc->caps.subvp_prefetch_end_to_mall_start_us; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c index 4e1db842b98c..761ec9891875 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c @@ -96,12 +96,15 @@ #include "reg_helper.h" #include "dce/dmub_abm.h" #include "dce/dmub_psr.h" +#include "dce/dmub_replay.h" #include "dce/dce_aux.h" #include "dce/dce_i2c.h" #include "dml/dcn31/display_mode_vba_31.h" /*temp*/ #include "vm_helper.h" #include "dcn20/dcn20_vmid.h" +#include "dc_state_priv.h" + #include "link_enc_cfg.h" #define DC_LOGGER_INIT(logger) @@ -766,7 +769,7 @@ static const struct dc_debug_options debug_defaults_drv = { .enable_hpo_pg_support = false, .enable_legacy_fast_update = true, .enable_single_display_2to1_odm_policy = false, - .disable_idle_power_optimizations = true, + .disable_idle_power_optimizations = false, .dmcub_emulation = false, .disable_boot_optimizations = false, .disable_unbounded_requesting = false, @@ -778,13 +781,15 @@ static const struct dc_debug_options debug_defaults_drv = { .ignore_pg = true, .psp_disabled_wa = true, .ips2_eval_delay_us = 200, - .ips2_entry_delay_us = 400 + .ips2_entry_delay_us = 400, + .static_screen_wait_frames = 2, }; static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, .disallow_psrsu = false, + .disallow_replay = false, }, .ilr = { .optimize_edp_link_rate = true, @@ -1543,6 +1548,9 @@ static void dcn35_resource_destruct(struct dcn35_resource_pool *pool) if (pool->base.psr != NULL) dmub_psr_destroy(&pool->base.psr); + if (pool->base.replay != NULL) + dmub_replay_destroy(&pool->base.replay); + if (pool->base.pg_cntl != NULL) dcn_pg_cntl_destroy(&pool->base.pg_cntl); @@ -2027,6 +2035,14 @@ static bool dcn35_resource_construct( goto create_fail; } + /* Replay */ + pool->base.replay = dmub_replay_create(ctx); + if (pool->base.replay == NULL) { + dm_error("DC: failed to create replay obj!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + /* ABM */ for (i = 0; i < pool->base.res_cap->num_timing_generator; i++) { pool->base.multiple_abms[i] = dmub_abm_create(ctx, diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h index d1a4ed6f5916..c78c9224ab60 100644 --- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h @@ -86,6 +86,7 @@ enum dmub_status { DMUB_STATUS_TIMEOUT, DMUB_STATUS_INVALID, DMUB_STATUS_HW_FAILURE, + DMUB_STATUS_POWER_STATE_D3 }; /* enum dmub_asic - dmub asic identifier */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index 3c092064c72e..c64b6c848ef7 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -653,7 +653,7 @@ union dmub_fw_boot_options { uint32_t gpint_scratch8: 1; /* 1 if GPINT is in scratch8*/ uint32_t usb4_cm_version: 1; /**< 1 CM support */ uint32_t dpia_hpd_int_enable_supported: 1; /* 1 if dpia hpd int enable supported */ - uint32_t usb4_dpia_bw_alloc_supported: 1; /* 1 if USB4 dpia BW allocation supported */ + uint32_t reserved0: 1; uint32_t disable_clk_ds: 1; /* 1 if disallow dispclk_ds and dppclk_ds*/ uint32_t disable_timeout_recovery : 1; /* 1 if timeout recovery should be disabled */ uint32_t ips_pg_disable: 1; /* 1 to disable ONO domains power gating*/ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c index 53ac1c66dd86..9ad738805320 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -768,7 +768,7 @@ enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub, return DMUB_STATUS_INVALID; if (dmub->power_state != DMUB_POWER_STATE_D0) - return DMUB_STATUS_INVALID; + return DMUB_STATUS_POWER_STATE_D3; if (dmub->inbox1_rb.rptr > dmub->inbox1_rb.capacity || dmub->inbox1_rb.wrpt > dmub->inbox1_rb.capacity) { @@ -789,7 +789,7 @@ enum dmub_status dmub_srv_cmd_execute(struct dmub_srv *dmub) return DMUB_STATUS_INVALID; if (dmub->power_state != DMUB_POWER_STATE_D0) - return DMUB_STATUS_INVALID; + return DMUB_STATUS_POWER_STATE_D3; /** * Read back all the queued commands to ensure that they've diff --git a/drivers/gpu/drm/amd/display/include/audio_types.h b/drivers/gpu/drm/amd/display/include/audio_types.h index 66a54da0641c..915a031a43cb 100644 --- a/drivers/gpu/drm/amd/display/include/audio_types.h +++ b/drivers/gpu/drm/amd/display/include/audio_types.h @@ -64,7 +64,7 @@ enum audio_dto_source { /* PLL information required for AZALIA DTO calculation */ struct audio_pll_info { - uint32_t dp_dto_source_clock_in_khz; + uint32_t audio_dto_source_clock_in_khz; uint32_t feed_back_divider; enum audio_dto_source dto_source; bool ss_enabled; diff --git a/drivers/gpu/drm/amd/display/include/fixed31_32.h b/drivers/gpu/drm/amd/display/include/fixed31_32.h index 84da1dd34efd..d4cf7ead1d87 100644 --- a/drivers/gpu/drm/amd/display/include/fixed31_32.h +++ b/drivers/gpu/drm/amd/display/include/fixed31_32.h @@ -69,18 +69,6 @@ static const struct fixed31_32 dc_fixpt_epsilon = { 1LL }; static const struct fixed31_32 dc_fixpt_half = { 0x80000000LL }; static const struct fixed31_32 dc_fixpt_one = { 0x100000000LL }; -static inline struct fixed31_32 dc_fixpt_from_s3132(__u64 x) -{ - struct fixed31_32 val; - - /* If negative, convert to 2's complement. */ - if (x & (1ULL << 63)) - x = -(x & ~(1ULL << 63)); - - val.value = x; - return val; -} - /* * @brief * Initialization routines diff --git a/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h b/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h index bc96d0211360..813463ffe15c 100644 --- a/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h +++ b/drivers/gpu/drm/amd/display/include/grph_object_ctrl_defs.h @@ -417,6 +417,8 @@ struct integrated_info { /* V2.1 */ struct edp_info edp1_info; struct edp_info edp2_info; + uint32_t gpuclk_ss_percentage; + uint32_t gpuclk_ss_type; }; /* diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index 47296d155c3a..3955b7e4b2e2 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -81,6 +81,7 @@ fail_alloc_context: void mod_freesync_destroy(struct mod_freesync *mod_freesync) { struct core_freesync *core_freesync = NULL; + if (mod_freesync == NULL) return; core_freesync = MOD_FREESYNC_TO_CORE(mod_freesync); @@ -278,9 +279,8 @@ static void apply_below_the_range(struct core_freesync *core_freesync, } } else if (last_render_time_in_us > (max_render_time_in_us + in_out_vrr->btr.margin_in_us / 2)) { /* Enter Below the Range */ - if (!in_out_vrr->btr.btr_active) { + if (!in_out_vrr->btr.btr_active) in_out_vrr->btr.btr_active = true; - } } /* BTR set to "not active" so disengage */ diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c index 1ddb4f5eac8e..182e7532dda8 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c @@ -63,6 +63,7 @@ static inline enum mod_hdcp_status check_hdcp_capable_dp(struct mod_hdcp *hdcp) static inline enum mod_hdcp_status check_r0p_available_dp(struct mod_hdcp *hdcp) { enum mod_hdcp_status status; + if (is_dp_hdcp(hdcp)) { status = (hdcp->auth.msg.hdcp1.bstatus & DP_BSTATUS_R0_PRIME_READY) ? @@ -131,9 +132,8 @@ static inline uint8_t get_device_count(struct mod_hdcp *hdcp) static inline enum mod_hdcp_status check_device_count(struct mod_hdcp *hdcp) { /* Avoid device count == 0 to do authentication */ - if (0 == get_device_count(hdcp)) { + if (get_device_count(hdcp) == 0) return MOD_HDCP_STATUS_HDCP1_DEVICE_COUNT_MISMATCH_FAILURE; - } /* Some MST display may choose to report the internal panel as an HDCP RX. * To update this condition with 1(because the immediate repeater's internal diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c index 91c22b96ebde..733f22bed021 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c @@ -208,9 +208,8 @@ static inline uint8_t get_device_count(struct mod_hdcp *hdcp) static enum mod_hdcp_status check_device_count(struct mod_hdcp *hdcp) { /* Avoid device count == 0 to do authentication */ - if (0 == get_device_count(hdcp)) { + if (get_device_count(hdcp) == 0) return MOD_HDCP_STATUS_HDCP1_DEVICE_COUNT_MISMATCH_FAILURE; - } /* Some MST display may choose to report the internal panel as an HDCP RX. */ /* To update this condition with 1(because the immediate repeater's internal */ @@ -689,9 +688,8 @@ static enum mod_hdcp_status validate_stream_ready(struct mod_hdcp *hdcp, if (is_hdmi_dvi_sl_hdcp(hdcp)) { if (!process_rxstatus(hdcp, event_ctx, input, &status)) goto out; - if (event_ctx->rx_id_list_ready) { + if (event_ctx->rx_id_list_ready) goto out; - } } if (is_hdmi_dvi_sl_hdcp(hdcp)) if (!mod_hdcp_execute_and_set(check_stream_ready_available, diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h index c62df3bcc7cb..1d83c1b9da10 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.h @@ -86,10 +86,12 @@ #define HDCP_CPIRQ_TRACE(hdcp) \ HDCP_LOG_FSM(hdcp, "[Link %d] --> CPIRQ", hdcp->config.index) #define HDCP_EVENT_TRACE(hdcp, event) \ - if (event == MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) \ - HDCP_TIMEOUT_TRACE(hdcp); \ - else if (event == MOD_HDCP_EVENT_CPIRQ) \ - HDCP_CPIRQ_TRACE(hdcp) + do { \ + if (event == MOD_HDCP_EVENT_WATCHDOG_TIMEOUT) \ + HDCP_TIMEOUT_TRACE(hdcp); \ + else if (event == MOD_HDCP_EVENT_CPIRQ) \ + HDCP_CPIRQ_TRACE(hdcp); \ + } while (0) /* TODO: find some way to tell if logging is off to save time */ #define HDCP_DDC_READ_TRACE(hdcp, msg_name, msg, msg_size) do { \ mod_hdcp_dump_binary_message(msg, msg_size, hdcp->buf, \ diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c index ee67a35c2a8e..8c137d7c032e 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c @@ -443,7 +443,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption(struct mod_hdcp for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { if (hdcp->displays[i].adjust.disable || hdcp->displays[i].state != MOD_HDCP_DISPLAY_ACTIVE) - continue; + continue; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -926,7 +926,7 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { if (hdcp->displays[i].adjust.disable || hdcp->displays[i].state != MOD_HDCP_DISPLAY_ACTIVE) - continue; + continue; hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.display_handle = hdcp->displays[i].index; hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.session_handle = hdcp->auth.id; diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h index 5b71bc96b98c..7844ea91650b 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.h @@ -98,9 +98,9 @@ enum ta_dtm_encoder_type { * This enum defines software value for dio_output_type */ typedef enum { - TA_DTM_DIO_OUTPUT_TYPE__INVALID, - TA_DTM_DIO_OUTPUT_TYPE__DIRECT, - TA_DTM_DIO_OUTPUT_TYPE__DPIA + TA_DTM_DIO_OUTPUT_TYPE__INVALID, + TA_DTM_DIO_OUTPUT_TYPE__DIRECT, + TA_DTM_DIO_OUTPUT_TYPE__DPIA } ta_dtm_dio_output_type; struct ta_dtm_topology_update_input_v3 { @@ -237,11 +237,11 @@ enum ta_hdcp2_hdcp2_msg_id_max_size { #define TA_HDCP__HDCP1_KSV_LIST_MAX_ENTRIES 127 #define TA_HDCP__HDCP1_V_PRIME_SIZE 20 #define TA_HDCP__HDCP2_TX_BUF_MAX_SIZE \ - TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_NO_STORED_KM + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_STORED_KM + 6 + (TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_NO_STORED_KM + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_STORED_KM + 6) // 64 bits boundaries #define TA_HDCP__HDCP2_RX_BUF_MAX_SIZE \ - TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_CERT + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_RECEIVER_INFO + 4 + (TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_SEND_CERT + TA_HDCP_HDCP2_MSG_ID_MAX_SIZE__AKE_RECEIVER_INFO + 4) enum ta_hdcp_status { TA_HDCP_STATUS__SUCCESS = 0x00, diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h index afe1f6cce528..cc3dc9b589f6 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h @@ -23,34 +23,6 @@ * */ - - - -/* - * Copyright 2016 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: AMD - * - */ - #ifndef MOD_FREESYNC_H_ #define MOD_FREESYNC_H_ diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c index 84f9b412a4f1..738ee763f24a 100644 --- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c +++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c @@ -147,12 +147,15 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, } /* VSC packet set to 4 for PSR-SU, or 2 for PSR1 */ - if (stream->link->psr_settings.psr_version == DC_PSR_VERSION_SU_1) - vsc_packet_revision = vsc_packet_rev4; - else if (stream->link->replay_settings.config.replay_supported) + if (stream->link->psr_settings.psr_feature_enabled) { + if (stream->link->psr_settings.psr_version == DC_PSR_VERSION_SU_1) + vsc_packet_revision = vsc_packet_rev4; + else if (stream->link->psr_settings.psr_version == DC_PSR_VERSION_1) + vsc_packet_revision = vsc_packet_rev2; + } + + if (stream->link->replay_settings.config.replay_supported) vsc_packet_revision = vsc_packet_rev4; - else if (stream->link->psr_settings.psr_version == DC_PSR_VERSION_1) - vsc_packet_revision = vsc_packet_rev2; /* Update to revision 5 for extended colorimetry support */ if (stream->use_vsc_sdp_for_colorimetry) diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c index 1675314a3ff2..ad98e504c00d 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c @@ -31,7 +31,7 @@ #define DIV_ROUNDUP(a, b) (((a)+((b)/2))/(b)) #define bswap16_based_on_endian(big_endian, value) \ - (big_endian) ? cpu_to_be16(value) : cpu_to_le16(value) + ((big_endian) ? cpu_to_be16(value) : cpu_to_le16(value)) /* Possible Min Reduction config from least aggressive to most aggressive * 0 1 2 3 4 5 6 7 8 9 10 11 12 @@ -973,6 +973,34 @@ bool psr_su_set_dsc_slice_height(struct dc *dc, struct dc_link *link, return true; } +void set_replay_coasting_vtotal(struct dc_link *link, + enum replay_coasting_vtotal_type type, + uint16_t vtotal) +{ + link->replay_settings.coasting_vtotal_table[type] = vtotal; +} + +void calculate_replay_link_off_frame_count(struct dc_link *link, + uint16_t vtotal, uint16_t htotal) +{ + uint8_t max_link_off_frame_count = 0; + uint16_t max_deviation_line = 0, pixel_deviation_per_line = 0; + + max_deviation_line = link->dpcd_caps.pr_info.max_deviation_line; + pixel_deviation_per_line = link->dpcd_caps.pr_info.pixel_deviation_per_line; + + if (htotal != 0 && vtotal != 0) + max_link_off_frame_count = htotal * max_deviation_line / (pixel_deviation_per_line * vtotal); + else + ASSERT(0); + + link->replay_settings.link_off_frame_count_level = + max_link_off_frame_count >= PR_LINK_OFF_FRAME_COUNT_BEST ? PR_LINK_OFF_FRAME_COUNT_BEST : + max_link_off_frame_count >= PR_LINK_OFF_FRAME_COUNT_GOOD ? PR_LINK_OFF_FRAME_COUNT_GOOD : + PR_LINK_OFF_FRAME_COUNT_FAIL; + +} + bool fill_custom_backlight_caps(unsigned int config_no, struct dm_acpi_atif_backlight_caps *caps) { unsigned int data_points_size; diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h index d9e0d67d67f7..c17bbc6fb38c 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h @@ -54,6 +54,11 @@ bool dmub_init_abm_config(struct resource_pool *res_pool, unsigned int inst); void init_replay_config(struct dc_link *link, struct replay_config *pr_config); +void set_replay_coasting_vtotal(struct dc_link *link, + enum replay_coasting_vtotal_type type, + uint16_t vtotal); +void calculate_replay_link_off_frame_count(struct dc_link *link, + uint16_t vtotal, uint16_t htotal); bool is_psr_su_specific_panel(struct dc_link *link); void mod_power_calc_psr_configs(struct psr_config *psr_config, diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index bf7f258c324a..1dc5dd9b7bf7 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -244,7 +244,6 @@ enum DC_FEATURE_MASK { DC_DISABLE_LTTPR_DP2_0 = (1 << 6), //0x40, disabled by default DC_PSR_ALLOW_SMU_OPT = (1 << 7), //0x80, disabled by default DC_PSR_ALLOW_MULTI_DISP_OPT = (1 << 8), //0x100, disabled by default - DC_REPLAY_MASK = (1 << 9), //0x200, disabled by default for dcn < 3.1.4 }; enum DC_DEBUG_MASK { @@ -255,9 +254,10 @@ enum DC_DEBUG_MASK { DC_DISABLE_PSR = 0x10, DC_FORCE_SUBVP_MCLK_SWITCH = 0x20, DC_DISABLE_MPO = 0x40, - DC_DISABLE_REPLAY = 0x50, DC_ENABLE_DPIA_TRACE = 0x80, DC_ENABLE_DML2 = 0x100, + DC_DISABLE_PSR_SU = 0x200, + DC_DISABLE_REPLAY = 0x400, }; enum amd_dpm_forced_level; diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_11_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_11_0_offset.h index 7ee3d291120d..6f80bfa7e41a 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_11_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_11_0_offset.h @@ -8707,10 +8707,10 @@ #define regBIF_BX1_MM_CFGREGS_CNTL_BASE_IDX 2 #define regBIF_BX1_BX_RESET_CNTL 0x00f0 #define regBIF_BX1_BX_RESET_CNTL_BASE_IDX 2 -#define regBIF_BX1_INTERRUPT_CNTL 0x8e11 -#define regBIF_BX1_INTERRUPT_CNTL_BASE_IDX 5 -#define regBIF_BX1_INTERRUPT_CNTL2 0x8e12 -#define regBIF_BX1_INTERRUPT_CNTL2_BASE_IDX 5 +#define regBIF_BX1_INTERRUPT_CNTL 0x00f1 +#define regBIF_BX1_INTERRUPT_CNTL_BASE_IDX 2 +#define regBIF_BX1_INTERRUPT_CNTL2 0x00f2 +#define regBIF_BX1_INTERRUPT_CNTL2_BASE_IDX 2 #define regBIF_BX1_CLKREQB_PAD_CNTL 0x00f8 #define regBIF_BX1_CLKREQB_PAD_CNTL_BASE_IDX 2 #define regBIF_BX1_BIF_FEATURES_CONTROL_MISC 0x00fb diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index 36a5ad8c00c5..edcb85560ced 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -318,6 +318,7 @@ enum pp_xgmi_plpd_mode { #define MAX_GFX_CLKS 8 #define MAX_CLKS 4 #define NUM_VCN 4 +#define NUM_JPEG_ENG 32 struct seq_file; enum amd_pp_clock_type; @@ -775,6 +776,85 @@ struct gpu_metrics_v1_4 { uint16_t padding; }; +struct gpu_metrics_v1_5 { + struct metrics_table_header common_header; + + /* Temperature (Celsius) */ + uint16_t temperature_hotspot; + uint16_t temperature_mem; + uint16_t temperature_vrsoc; + + /* Power (Watts) */ + uint16_t curr_socket_power; + + /* Utilization (%) */ + uint16_t average_gfx_activity; + uint16_t average_umc_activity; // memory controller + uint16_t vcn_activity[NUM_VCN]; + uint16_t jpeg_activity[NUM_JPEG_ENG]; + + /* Energy (15.259uJ (2^-16) units) */ + uint64_t energy_accumulator; + + /* Driver attached timestamp (in ns) */ + uint64_t system_clock_counter; + + /* Throttle status */ + uint32_t throttle_status; + + /* Clock Lock Status. Each bit corresponds to clock instance */ + uint32_t gfxclk_lock_status; + + /* Link width (number of lanes) and speed (in 0.1 GT/s) */ + uint16_t pcie_link_width; + uint16_t pcie_link_speed; + + /* XGMI bus width and bitrate (in Gbps) */ + uint16_t xgmi_link_width; + uint16_t xgmi_link_speed; + + /* Utilization Accumulated (%) */ + uint32_t gfx_activity_acc; + uint32_t mem_activity_acc; + + /*PCIE accumulated bandwidth (GB/sec) */ + uint64_t pcie_bandwidth_acc; + + /*PCIE instantaneous bandwidth (GB/sec) */ + uint64_t pcie_bandwidth_inst; + + /* PCIE L0 to recovery state transition accumulated count */ + uint64_t pcie_l0_to_recov_count_acc; + + /* PCIE replay accumulated count */ + uint64_t pcie_replay_count_acc; + + /* PCIE replay rollover accumulated count */ + uint64_t pcie_replay_rover_count_acc; + + /* PCIE NAK sent accumulated count */ + uint32_t pcie_nak_sent_count_acc; + + /* PCIE NAK received accumulated count */ + uint32_t pcie_nak_rcvd_count_acc; + + /* XGMI accumulated data transfer size(KiloBytes) */ + uint64_t xgmi_read_data_acc[NUM_XGMI_LINKS]; + uint64_t xgmi_write_data_acc[NUM_XGMI_LINKS]; + + /* PMFW attached timestamp (10ns resolution) */ + uint64_t firmware_timestamp; + + /* Current clocks (Mhz) */ + uint16_t current_gfxclk[MAX_GFX_CLKS]; + uint16_t current_socclk[MAX_CLKS]; + uint16_t current_vclk0[MAX_CLKS]; + uint16_t current_dclk0[MAX_CLKS]; + uint16_t current_uclk; + + uint16_t padding; +}; + /* * gpu_metrics_v2_0 is not recommended as it's not naturally aligned. * Use gpu_metrics_v2_1 or later instead. diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index 2cd995b0ceba..087d57850304 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -2168,7 +2168,9 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_ if (amdgpu_dpm_is_overdrive_supported(adev)) *states = ATTR_STATE_SUPPORTED; } else if (DEVICE_ATTR_IS(mem_busy_percent)) { - if (adev->flags & AMD_IS_APU || gc_ver == IP_VERSION(9, 0, 1)) + if ((adev->flags & AMD_IS_APU && + gc_ver != IP_VERSION(9, 4, 3)) || + gc_ver == IP_VERSION(9, 0, 1)) *states = ATTR_STATE_UNSUPPORTED; } else if (DEVICE_ATTR_IS(pcie_bw)) { /* PCIe Perf counters won't work on APU nodes */ @@ -4347,11 +4349,19 @@ static int amdgpu_debugfs_pm_info_pp(struct seq_file *m, struct amdgpu_device *a if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VDDNB, (void *)&value, &size)) seq_printf(m, "\t%u mV (VDDNB)\n", value); size = sizeof(uint32_t); - if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_AVG_POWER, (void *)&query, &size)) - seq_printf(m, "\t%u.%02u W (average GPU)\n", query >> 8, query & 0xff); + if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_AVG_POWER, (void *)&query, &size)) { + if (adev->flags & AMD_IS_APU) + seq_printf(m, "\t%u.%02u W (average SoC including CPU)\n", query >> 8, query & 0xff); + else + seq_printf(m, "\t%u.%02u W (average SoC)\n", query >> 8, query & 0xff); + } size = sizeof(uint32_t); - if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_INPUT_POWER, (void *)&query, &size)) - seq_printf(m, "\t%u.%02u W (current GPU)\n", query >> 8, query & 0xff); + if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_INPUT_POWER, (void *)&query, &size)) { + if (adev->flags & AMD_IS_APU) + seq_printf(m, "\t%u.%02u W (current SoC including CPU)\n", query >> 8, query & 0xff); + else + seq_printf(m, "\t%u.%02u W (current SoC)\n", query >> 8, query & 0xff); + } size = sizeof(value); seq_printf(m, "\n"); @@ -4377,9 +4387,9 @@ static int amdgpu_debugfs_pm_info_pp(struct seq_file *m, struct amdgpu_device *a /* VCN clocks */ if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VCN_POWER_STATE, (void *)&value, &size)) { if (!value) { - seq_printf(m, "VCN: Disabled\n"); + seq_printf(m, "VCN: Powered down\n"); } else { - seq_printf(m, "VCN: Enabled\n"); + seq_printf(m, "VCN: Powered up\n"); if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_UVD_DCLK, (void *)&value, &size)) seq_printf(m, "\t%u MHz (DCLK)\n", value/100); if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_UVD_VCLK, (void *)&value, &size)) @@ -4391,9 +4401,9 @@ static int amdgpu_debugfs_pm_info_pp(struct seq_file *m, struct amdgpu_device *a /* UVD clocks */ if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_UVD_POWER, (void *)&value, &size)) { if (!value) { - seq_printf(m, "UVD: Disabled\n"); + seq_printf(m, "UVD: Powered down\n"); } else { - seq_printf(m, "UVD: Enabled\n"); + seq_printf(m, "UVD: Powered up\n"); if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_UVD_DCLK, (void *)&value, &size)) seq_printf(m, "\t%u MHz (DCLK)\n", value/100); if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_UVD_VCLK, (void *)&value, &size)) @@ -4405,9 +4415,9 @@ static int amdgpu_debugfs_pm_info_pp(struct seq_file *m, struct amdgpu_device *a /* VCE clocks */ if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VCE_POWER, (void *)&value, &size)) { if (!value) { - seq_printf(m, "VCE: Disabled\n"); + seq_printf(m, "VCE: Powered down\n"); } else { - seq_printf(m, "VCE: Enabled\n"); + seq_printf(m, "VCE: Powered up\n"); if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VCE_ECCLK, (void *)&value, &size)) seq_printf(m, "\t%u MHz (ECCLK)\n", value/100); } diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/process_pptables_v1_0.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/process_pptables_v1_0.c index f2a55c1413f5..17882f8dfdd3 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/process_pptables_v1_0.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/process_pptables_v1_0.c @@ -200,7 +200,7 @@ static int get_platform_power_management_table( struct pp_hwmgr *hwmgr, ATOM_Tonga_PPM_Table *atom_ppm_table) { - struct phm_ppm_table *ptr = kzalloc(sizeof(ATOM_Tonga_PPM_Table), GFP_KERNEL); + struct phm_ppm_table *ptr = kzalloc(sizeof(*ptr), GFP_KERNEL); struct phm_ppt_v1_information *pp_table_information = (struct phm_ppt_v1_information *)(hwmgr->pptable); diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c index 11372fcc59c8..aa91730e4eaf 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c @@ -2974,6 +2974,8 @@ static int smu7_hwmgr_backend_init(struct pp_hwmgr *hwmgr) result = smu7_get_evv_voltages(hwmgr); if (result) { pr_info("Get EVV Voltage Failed. Abort Driver loading!\n"); + kfree(hwmgr->backend); + hwmgr->backend = NULL; return -EINVAL; } } else { @@ -3019,8 +3021,10 @@ static int smu7_hwmgr_backend_init(struct pp_hwmgr *hwmgr) } result = smu7_update_edc_leakage_table(hwmgr); - if (result) + if (result) { + smu7_hwmgr_backend_fini(hwmgr); return result; + } return 0; } @@ -3995,6 +3999,7 @@ static int smu7_read_sensor(struct pp_hwmgr *hwmgr, int idx, uint32_t sclk, mclk, activity_percent; uint32_t offset, val_vid; struct smu7_hwmgr *data = (struct smu7_hwmgr *)(hwmgr->backend); + struct amdgpu_device *adev = hwmgr->adev; /* size must be at least 4 bytes for all sensors */ if (*size < 4) @@ -4038,7 +4043,21 @@ static int smu7_read_sensor(struct pp_hwmgr *hwmgr, int idx, *size = 4; return 0; case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: - return smu7_get_gpu_power(hwmgr, (uint32_t *)value); + if ((adev->asic_type != CHIP_HAWAII) && + (adev->asic_type != CHIP_BONAIRE) && + (adev->asic_type != CHIP_FIJI) && + (adev->asic_type != CHIP_TONGA)) + return smu7_get_gpu_power(hwmgr, (uint32_t *)value); + else + return -EOPNOTSUPP; + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: + if ((adev->asic_type != CHIP_HAWAII) && + (adev->asic_type != CHIP_BONAIRE) && + (adev->asic_type != CHIP_FIJI) && + (adev->asic_type != CHIP_TONGA)) + return -EOPNOTSUPP; + else + return smu7_get_gpu_power(hwmgr, (uint32_t *)value); case AMDGPU_PP_SENSOR_VDDGFX: if ((data->vr_config & VRCONF_VDDGFX_MASK) == (VR_SVI2_PLANE_2 << VRCONF_VDDGFX_SHIFT)) diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c index 9e4228232f02..ad1fd3150d03 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c @@ -2298,6 +2298,7 @@ static uint32_t ci_get_mac_definition(uint32_t value) case SMU_MAX_ENTRIES_SMIO: return SMU7_MAX_ENTRIES_SMIO; case SMU_MAX_LEVELS_VDDC: + case SMU_MAX_LEVELS_VDDGFX: return SMU7_MAX_LEVELS_VDDC; case SMU_MAX_LEVELS_VDDCI: return SMU7_MAX_LEVELS_VDDCI; diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c index 97d9802fe673..17d2f5bff4a7 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c @@ -2263,6 +2263,7 @@ static uint32_t iceland_get_mac_definition(uint32_t value) case SMU_MAX_ENTRIES_SMIO: return SMU71_MAX_ENTRIES_SMIO; case SMU_MAX_LEVELS_VDDC: + case SMU_MAX_LEVELS_VDDGFX: return SMU71_MAX_LEVELS_VDDC; case SMU_MAX_LEVELS_VDDCI: return SMU71_MAX_LEVELS_VDDCI; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h index fef2d290f3f2..7b812b9994d7 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h @@ -123,7 +123,7 @@ typedef enum { VOLTAGE_GUARDBAND_COUNT } GFX_GUARDBAND_e; -#define SMU_METRICS_TABLE_VERSION 0x9 +#define SMU_METRICS_TABLE_VERSION 0xB typedef struct __attribute__((packed, aligned(4))) { uint32_t AccumulationCounter; @@ -219,7 +219,103 @@ typedef struct __attribute__((packed, aligned(4))) { uint32_t PCIenReplayARolloverCountAcc; // The Pcie counter itself is accumulated uint32_t PCIeNAKSentCountAcc; // The Pcie counter itself is accumulated uint32_t PCIeNAKReceivedCountAcc; // The Pcie counter itself is accumulated -} MetricsTable_t; + + // VCN/JPEG ACTIVITY + uint32_t VcnBusy[4]; + uint32_t JpegBusy[32]; +} MetricsTableX_t; + +typedef struct __attribute__((packed, aligned(4))) { + uint32_t AccumulationCounter; + + //TEMPERATURE + uint32_t MaxSocketTemperature; + uint32_t MaxVrTemperature; + uint32_t MaxHbmTemperature; + uint64_t MaxSocketTemperatureAcc; + uint64_t MaxVrTemperatureAcc; + uint64_t MaxHbmTemperatureAcc; + + //POWER + uint32_t SocketPowerLimit; + uint32_t MaxSocketPowerLimit; + uint32_t SocketPower; + + //ENERGY + uint64_t Timestamp; + uint64_t SocketEnergyAcc; + uint64_t CcdEnergyAcc; + uint64_t XcdEnergyAcc; + uint64_t AidEnergyAcc; + uint64_t HbmEnergyAcc; + + //FREQUENCY + uint32_t CclkFrequencyLimit; + uint32_t GfxclkFrequencyLimit; + uint32_t FclkFrequency; + uint32_t UclkFrequency; + uint32_t SocclkFrequency[4]; + uint32_t VclkFrequency[4]; + uint32_t DclkFrequency[4]; + uint32_t LclkFrequency[4]; + uint64_t GfxclkFrequencyAcc[8]; + uint64_t CclkFrequencyAcc[96]; + + //FREQUENCY RANGE + uint32_t MaxCclkFrequency; + uint32_t MinCclkFrequency; + uint32_t MaxGfxclkFrequency; + uint32_t MinGfxclkFrequency; + uint32_t FclkFrequencyTable[4]; + uint32_t UclkFrequencyTable[4]; + uint32_t SocclkFrequencyTable[4]; + uint32_t VclkFrequencyTable[4]; + uint32_t DclkFrequencyTable[4]; + uint32_t LclkFrequencyTable[4]; + uint32_t MaxLclkDpmRange; + uint32_t MinLclkDpmRange; + + //XGMI + uint32_t XgmiWidth; + uint32_t XgmiBitrate; + uint64_t XgmiReadBandwidthAcc[8]; + uint64_t XgmiWriteBandwidthAcc[8]; + + //ACTIVITY + uint32_t SocketC0Residency; + uint32_t SocketGfxBusy; + uint32_t DramBandwidthUtilization; + uint64_t SocketC0ResidencyAcc; + uint64_t SocketGfxBusyAcc; + uint64_t DramBandwidthAcc; + uint32_t MaxDramBandwidth; + uint64_t DramBandwidthUtilizationAcc; + uint64_t PcieBandwidthAcc[4]; + + //THROTTLERS + uint32_t ProchotResidencyAcc; + uint32_t PptResidencyAcc; + uint32_t SocketThmResidencyAcc; + uint32_t VrThmResidencyAcc; + uint32_t HbmThmResidencyAcc; + uint32_t GfxLockXCDMak; + + // New Items at end to maintain driver compatibility + uint32_t GfxclkFrequency[8]; + + //PSNs + uint64_t PublicSerialNumber_AID[4]; + uint64_t PublicSerialNumber_XCD[8]; + uint64_t PublicSerialNumber_CCD[12]; + + //XGMI Data tranfser size + uint64_t XgmiReadDataSizeAcc[8];//in KByte + uint64_t XgmiWriteDataSizeAcc[8];//in KByte + + // VCN/JPEG ACTIVITY + uint32_t VcnBusy[4]; + uint32_t JpegBusy[32]; +} MetricsTableA_t; #define SMU_VF_METRICS_TABLE_VERSION 0x3 diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c index 3998c9b31d07..a28649f21093 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c @@ -248,6 +248,8 @@ struct PPTable_t { #define SMUQ10_TO_UINT(x) ((x) >> 10) #define SMUQ10_FRAC(x) ((x) & 0x3ff) #define SMUQ10_ROUND(x) ((SMUQ10_TO_UINT(x)) + ((SMUQ10_FRAC(x)) >= 0x200)) +#define GET_METRIC_FIELD(field) ((adev->flags & AMD_IS_APU) ?\ + (metrics_a->field) : (metrics_x->field)) struct smu_v13_0_6_dpm_map { enum smu_clk_type clk_type; @@ -330,7 +332,8 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu) SMU_TABLE_INIT(tables, SMU_TABLE_PMSTATUSLOG, SMU13_TOOL_SIZE, PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); - SMU_TABLE_INIT(tables, SMU_TABLE_SMU_METRICS, sizeof(MetricsTable_t), + SMU_TABLE_INIT(tables, SMU_TABLE_SMU_METRICS, + max(sizeof(MetricsTableX_t), sizeof(MetricsTableA_t)), PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT); @@ -338,12 +341,13 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu) PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT); - smu_table->metrics_table = kzalloc(sizeof(MetricsTable_t), GFP_KERNEL); + smu_table->metrics_table = kzalloc(max(sizeof(MetricsTableX_t), + sizeof(MetricsTableA_t)), GFP_KERNEL); if (!smu_table->metrics_table) return -ENOMEM; smu_table->metrics_time = 0; - smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_4); + smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_5); smu_table->gpu_metrics_table = kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL); if (!smu_table->gpu_metrics_table) { @@ -469,9 +473,11 @@ static ssize_t smu_v13_0_6_get_pm_metrics(struct smu_context *smu, static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) { struct smu_table_context *smu_table = &smu->smu_table; - MetricsTable_t *metrics = (MetricsTable_t *)smu_table->metrics_table; + MetricsTableX_t *metrics_x = (MetricsTableX_t *)smu_table->metrics_table; + MetricsTableA_t *metrics_a = (MetricsTableA_t *)smu_table->metrics_table; struct PPTable_t *pptable = (struct PPTable_t *)smu_table->driver_pptable; + struct amdgpu_device *adev = smu->adev; int ret, i, retry = 100; uint32_t table_version; @@ -483,7 +489,7 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) return ret; /* Ensure that metrics have been updated */ - if (metrics->AccumulationCounter) + if (GET_METRIC_FIELD(AccumulationCounter)) break; usleep_range(1000, 1100); @@ -500,29 +506,29 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) table_version; pptable->MaxSocketPowerLimit = - SMUQ10_ROUND(metrics->MaxSocketPowerLimit); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketPowerLimit)); pptable->MaxGfxclkFrequency = - SMUQ10_ROUND(metrics->MaxGfxclkFrequency); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxGfxclkFrequency)); pptable->MinGfxclkFrequency = - SMUQ10_ROUND(metrics->MinGfxclkFrequency); + SMUQ10_ROUND(GET_METRIC_FIELD(MinGfxclkFrequency)); for (i = 0; i < 4; ++i) { pptable->FclkFrequencyTable[i] = - SMUQ10_ROUND(metrics->FclkFrequencyTable[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(FclkFrequencyTable)[i]); pptable->UclkFrequencyTable[i] = - SMUQ10_ROUND(metrics->UclkFrequencyTable[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(UclkFrequencyTable)[i]); pptable->SocclkFrequencyTable[i] = SMUQ10_ROUND( - metrics->SocclkFrequencyTable[i]); + GET_METRIC_FIELD(SocclkFrequencyTable)[i]); pptable->VclkFrequencyTable[i] = - SMUQ10_ROUND(metrics->VclkFrequencyTable[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(VclkFrequencyTable)[i]); pptable->DclkFrequencyTable[i] = - SMUQ10_ROUND(metrics->DclkFrequencyTable[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(DclkFrequencyTable)[i]); pptable->LclkFrequencyTable[i] = - SMUQ10_ROUND(metrics->LclkFrequencyTable[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(LclkFrequencyTable)[i]); } /* use AID0 serial number by default */ - pptable->PublicSerialNumber_AID = metrics->PublicSerialNumber_AID[0]; + pptable->PublicSerialNumber_AID = GET_METRIC_FIELD(PublicSerialNumber_AID)[0]; pptable->Init = true; } @@ -824,7 +830,8 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu, uint32_t *value) { struct smu_table_context *smu_table = &smu->smu_table; - MetricsTable_t *metrics = (MetricsTable_t *)smu_table->metrics_table; + MetricsTableX_t *metrics_x = (MetricsTableX_t *)smu_table->metrics_table; + MetricsTableA_t *metrics_a = (MetricsTableA_t *)smu_table->metrics_table; struct amdgpu_device *adev = smu->adev; int ret = 0; int xcc_id; @@ -839,50 +846,50 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu, case METRICS_AVERAGE_GFXCLK: if (smu->smc_fw_version >= 0x552F00) { xcc_id = GET_INST(GC, 0); - *value = SMUQ10_ROUND(metrics->GfxclkFrequency[xcc_id]); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(GfxclkFrequency)[xcc_id]); } else { *value = 0; } break; case METRICS_CURR_SOCCLK: case METRICS_AVERAGE_SOCCLK: - *value = SMUQ10_ROUND(metrics->SocclkFrequency[0]); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(SocclkFrequency)[0]); break; case METRICS_CURR_UCLK: case METRICS_AVERAGE_UCLK: - *value = SMUQ10_ROUND(metrics->UclkFrequency); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(UclkFrequency)); break; case METRICS_CURR_VCLK: - *value = SMUQ10_ROUND(metrics->VclkFrequency[0]); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(VclkFrequency)[0]); break; case METRICS_CURR_DCLK: - *value = SMUQ10_ROUND(metrics->DclkFrequency[0]); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(DclkFrequency)[0]); break; case METRICS_CURR_FCLK: - *value = SMUQ10_ROUND(metrics->FclkFrequency); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(FclkFrequency)); break; case METRICS_AVERAGE_GFXACTIVITY: - *value = SMUQ10_ROUND(metrics->SocketGfxBusy); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(SocketGfxBusy)); break; case METRICS_AVERAGE_MEMACTIVITY: - *value = SMUQ10_ROUND(metrics->DramBandwidthUtilization); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(DramBandwidthUtilization)); break; case METRICS_CURR_SOCKETPOWER: - *value = SMUQ10_ROUND(metrics->SocketPower) << 8; + *value = SMUQ10_ROUND(GET_METRIC_FIELD(SocketPower)) << 8; break; case METRICS_TEMPERATURE_HOTSPOT: - *value = SMUQ10_ROUND(metrics->MaxSocketTemperature) * + *value = SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketTemperature)) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; case METRICS_TEMPERATURE_MEM: - *value = SMUQ10_ROUND(metrics->MaxHbmTemperature) * + *value = SMUQ10_ROUND(GET_METRIC_FIELD(MaxHbmTemperature)) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; /* This is the max of all VRs and not just SOC VR. * No need to define another data type for the same. */ case METRICS_TEMPERATURE_VRSOC: - *value = SMUQ10_ROUND(metrics->MaxVrTemperature) * + *value = SMUQ10_ROUND(GET_METRIC_FIELD(MaxVrTemperature)) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; default: @@ -963,7 +970,9 @@ static int smu_v13_0_6_print_clks(struct smu_context *smu, char *buf, int size, if (i < (clocks.num_levels - 1)) clk2 = clocks.data[i + 1].clocks_in_khz / 1000; - if (curr_clk >= clk1 && curr_clk < clk2) { + if (curr_clk == clk1) { + level = i; + } else if (curr_clk >= clk1 && curr_clk < clk2) { level = (curr_clk - clk1) <= (clk2 - curr_clk) ? i : i + 1; @@ -2067,67 +2076,70 @@ static int smu_v13_0_6_get_current_pcie_link_speed(struct smu_context *smu) static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table) { struct smu_table_context *smu_table = &smu->smu_table; - struct gpu_metrics_v1_4 *gpu_metrics = - (struct gpu_metrics_v1_4 *)smu_table->gpu_metrics_table; + struct gpu_metrics_v1_5 *gpu_metrics = + (struct gpu_metrics_v1_5 *)smu_table->gpu_metrics_table; struct amdgpu_device *adev = smu->adev; - int ret = 0, xcc_id, inst, i; - MetricsTable_t *metrics; + int ret = 0, xcc_id, inst, i, j; + MetricsTableX_t *metrics_x; + MetricsTableA_t *metrics_a; u16 link_width_level; - metrics = kzalloc(sizeof(MetricsTable_t), GFP_KERNEL); - ret = smu_v13_0_6_get_metrics_table(smu, metrics, true); + metrics_x = kzalloc(max(sizeof(MetricsTableX_t), sizeof(MetricsTableA_t)), GFP_KERNEL); + ret = smu_v13_0_6_get_metrics_table(smu, metrics_x, true); if (ret) { - kfree(metrics); + kfree(metrics_x); return ret; } - smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 4); + metrics_a = (MetricsTableA_t *)metrics_x; + + smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 5); gpu_metrics->temperature_hotspot = - SMUQ10_ROUND(metrics->MaxSocketTemperature); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketTemperature)); /* Individual HBM stack temperature is not reported */ gpu_metrics->temperature_mem = - SMUQ10_ROUND(metrics->MaxHbmTemperature); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxHbmTemperature)); /* Reports max temperature of all voltage rails */ gpu_metrics->temperature_vrsoc = - SMUQ10_ROUND(metrics->MaxVrTemperature); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxVrTemperature)); gpu_metrics->average_gfx_activity = - SMUQ10_ROUND(metrics->SocketGfxBusy); + SMUQ10_ROUND(GET_METRIC_FIELD(SocketGfxBusy)); gpu_metrics->average_umc_activity = - SMUQ10_ROUND(metrics->DramBandwidthUtilization); + SMUQ10_ROUND(GET_METRIC_FIELD(DramBandwidthUtilization)); gpu_metrics->curr_socket_power = - SMUQ10_ROUND(metrics->SocketPower); + SMUQ10_ROUND(GET_METRIC_FIELD(SocketPower)); /* Energy counter reported in 15.259uJ (2^-16) units */ - gpu_metrics->energy_accumulator = metrics->SocketEnergyAcc; + gpu_metrics->energy_accumulator = GET_METRIC_FIELD(SocketEnergyAcc); for (i = 0; i < MAX_GFX_CLKS; i++) { xcc_id = GET_INST(GC, i); if (xcc_id >= 0) gpu_metrics->current_gfxclk[i] = - SMUQ10_ROUND(metrics->GfxclkFrequency[xcc_id]); + SMUQ10_ROUND(GET_METRIC_FIELD(GfxclkFrequency)[xcc_id]); if (i < MAX_CLKS) { gpu_metrics->current_socclk[i] = - SMUQ10_ROUND(metrics->SocclkFrequency[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(SocclkFrequency)[i]); inst = GET_INST(VCN, i); if (inst >= 0) { gpu_metrics->current_vclk0[i] = - SMUQ10_ROUND(metrics->VclkFrequency[inst]); + SMUQ10_ROUND(GET_METRIC_FIELD(VclkFrequency)[inst]); gpu_metrics->current_dclk0[i] = - SMUQ10_ROUND(metrics->DclkFrequency[inst]); + SMUQ10_ROUND(GET_METRIC_FIELD(DclkFrequency)[inst]); } } } - gpu_metrics->current_uclk = SMUQ10_ROUND(metrics->UclkFrequency); + gpu_metrics->current_uclk = SMUQ10_ROUND(GET_METRIC_FIELD(UclkFrequency)); /* Throttle status is not reported through metrics now */ gpu_metrics->throttle_status = 0; /* Clock Lock Status. Each bit corresponds to each GFXCLK instance */ - gpu_metrics->gfxclk_lock_status = metrics->GfxLockXCDMak >> GET_INST(GC, 0); + gpu_metrics->gfxclk_lock_status = GET_METRIC_FIELD(GfxLockXCDMak) >> GET_INST(GC, 0); if (!(adev->flags & AMD_IS_APU)) { link_width_level = smu_v13_0_6_get_current_pcie_link_width_level(smu); @@ -2139,38 +2151,57 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table gpu_metrics->pcie_link_speed = smu_v13_0_6_get_current_pcie_link_speed(smu); gpu_metrics->pcie_bandwidth_acc = - SMUQ10_ROUND(metrics->PcieBandwidthAcc[0]); + SMUQ10_ROUND(metrics_x->PcieBandwidthAcc[0]); gpu_metrics->pcie_bandwidth_inst = - SMUQ10_ROUND(metrics->PcieBandwidth[0]); + SMUQ10_ROUND(metrics_x->PcieBandwidth[0]); gpu_metrics->pcie_l0_to_recov_count_acc = - metrics->PCIeL0ToRecoveryCountAcc; + metrics_x->PCIeL0ToRecoveryCountAcc; gpu_metrics->pcie_replay_count_acc = - metrics->PCIenReplayAAcc; + metrics_x->PCIenReplayAAcc; gpu_metrics->pcie_replay_rover_count_acc = - metrics->PCIenReplayARolloverCountAcc; + metrics_x->PCIenReplayARolloverCountAcc; + gpu_metrics->pcie_nak_sent_count_acc = + metrics_x->PCIeNAKSentCountAcc; + gpu_metrics->pcie_nak_rcvd_count_acc = + metrics_x->PCIeNAKReceivedCountAcc; } gpu_metrics->system_clock_counter = ktime_get_boottime_ns(); gpu_metrics->gfx_activity_acc = - SMUQ10_ROUND(metrics->SocketGfxBusyAcc); + SMUQ10_ROUND(GET_METRIC_FIELD(SocketGfxBusyAcc)); gpu_metrics->mem_activity_acc = - SMUQ10_ROUND(metrics->DramBandwidthUtilizationAcc); + SMUQ10_ROUND(GET_METRIC_FIELD(DramBandwidthUtilizationAcc)); for (i = 0; i < NUM_XGMI_LINKS; i++) { gpu_metrics->xgmi_read_data_acc[i] = - SMUQ10_ROUND(metrics->XgmiReadDataSizeAcc[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(XgmiReadDataSizeAcc)[i]); gpu_metrics->xgmi_write_data_acc[i] = - SMUQ10_ROUND(metrics->XgmiWriteDataSizeAcc[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(XgmiWriteDataSizeAcc)[i]); } - gpu_metrics->xgmi_link_width = SMUQ10_ROUND(metrics->XgmiWidth); - gpu_metrics->xgmi_link_speed = SMUQ10_ROUND(metrics->XgmiBitrate); + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + inst = GET_INST(JPEG, i); + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + gpu_metrics->jpeg_activity[(i * adev->jpeg.num_jpeg_rings) + j] = + SMUQ10_ROUND(GET_METRIC_FIELD(JpegBusy) + [(inst * adev->jpeg.num_jpeg_rings) + j]); + } + } - gpu_metrics->firmware_timestamp = metrics->Timestamp; + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + inst = GET_INST(VCN, i); + gpu_metrics->vcn_activity[i] = + SMUQ10_ROUND(GET_METRIC_FIELD(VcnBusy)[inst]); + } + + gpu_metrics->xgmi_link_width = SMUQ10_ROUND(GET_METRIC_FIELD(XgmiWidth)); + gpu_metrics->xgmi_link_speed = SMUQ10_ROUND(GET_METRIC_FIELD(XgmiBitrate)); + + gpu_metrics->firmware_timestamp = GET_METRIC_FIELD(Timestamp); *table = (void *)gpu_metrics; - kfree(metrics); + kfree(metrics_x); return sizeof(*gpu_metrics); } @@ -2206,17 +2237,18 @@ static int smu_v13_0_6_mode2_reset(struct smu_context *smu) continue; } - if (ret) { - dev_err(adev->dev, - "failed to send mode2 message \tparam: 0x%08x error code %d\n", - SMU_RESET_MODE_2, ret); + if (ret) goto out; - } + } while (ret == -ETIME && timeout); out: mutex_unlock(&smu->message_lock); + if (ret) + dev_err(adev->dev, "failed to send mode2 reset, error code %d", + ret); + return ret; } @@ -2524,9 +2556,9 @@ static int mca_umc_mca_get_err_count(const struct mca_ras_info *mca_ras, struct return 0; } - if (type == AMDGPU_MCA_ERROR_TYPE_UE && umc_v12_0_is_uncorrectable_error(status0)) + if (type == AMDGPU_MCA_ERROR_TYPE_UE && umc_v12_0_is_uncorrectable_error(adev, status0)) *count = 1; - else if (type == AMDGPU_MCA_ERROR_TYPE_CE && umc_v12_0_is_correctable_error(status0)) + else if (type == AMDGPU_MCA_ERROR_TYPE_CE && umc_v12_0_is_correctable_error(adev, status0)) *count = 1; return 0; @@ -2864,6 +2896,13 @@ static int smu_v13_0_6_select_xgmi_plpd_policy(struct smu_context *smu, return ret; } +static ssize_t smu_v13_0_6_get_ecc_info(struct smu_context *smu, + void *table) +{ + /* Support ecc info by default */ + return 0; +} + static const struct pptable_funcs smu_v13_0_6_ppt_funcs = { /* init dpm */ .get_allowed_feature_mask = smu_v13_0_6_get_allowed_feature_mask, @@ -2918,6 +2957,7 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = { .i2c_init = smu_v13_0_6_i2c_control_init, .i2c_fini = smu_v13_0_6_i2c_control_fini, .send_hbm_bad_pages_num = smu_v13_0_6_smu_send_hbm_bad_page_num, + .get_ecc_info = smu_v13_0_6_get_ecc_info, }; void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 001a5cf09657..00cd615bbcdc 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -989,6 +989,9 @@ void smu_cmn_init_soft_gpu_metrics(void *table, uint8_t frev, uint8_t crev) case METRICS_VERSION(1, 4): structure_size = sizeof(struct gpu_metrics_v1_4); break; + case METRICS_VERSION(1, 5): + structure_size = sizeof(struct gpu_metrics_v1_5); + break; case METRICS_VERSION(2, 0): structure_size = sizeof(struct gpu_metrics_v2_0); break; diff --git a/drivers/gpu/drm/drm_framebuffer.c b/drivers/gpu/drm/drm_framebuffer.c index 3cc0ffc28e86..888aadb6a4ac 100644 --- a/drivers/gpu/drm/drm_framebuffer.c +++ b/drivers/gpu/drm/drm_framebuffer.c @@ -461,6 +461,7 @@ int drm_mode_rmfb(struct drm_device *dev, u32 fb_id, INIT_WORK_ONSTACK(&arg.work, drm_mode_rmfb_work_fn); INIT_LIST_HEAD(&arg.fbs); + drm_WARN_ON(dev, !list_empty(&fb->filp_head)); list_add_tail(&fb->filp_head, &arg.fbs); schedule_work(&arg.work); @@ -827,6 +828,8 @@ void drm_framebuffer_free(struct kref *kref) container_of(kref, struct drm_framebuffer, base.refcount); struct drm_device *dev = fb->dev; + drm_WARN_ON(dev, !list_empty(&fb->filp_head)); + /* * The lookup idr holds a weak reference, which has not necessarily been * removed at this point. Check for that. @@ -1119,7 +1122,7 @@ void drm_framebuffer_remove(struct drm_framebuffer *fb) dev = fb->dev; - WARN_ON(!list_empty(&fb->filp_head)); + drm_WARN_ON(dev, !list_empty(&fb->filp_head)); /* * drm ABI mandates that we remove any deleted framebuffers from active diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c index 9e8e4c60983d..672c655c7a8e 100644 --- a/drivers/gpu/drm/drm_plane.c +++ b/drivers/gpu/drm/drm_plane.c @@ -1503,6 +1503,7 @@ retry: out: if (fb) drm_framebuffer_put(fb); + fb = NULL; if (plane->old_fb) drm_framebuffer_put(plane->old_fb); plane->old_fb = NULL; diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c index 884a1da36089..6b25e195232f 100644 --- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c @@ -3067,24 +3067,29 @@ static void intel_c20pll_state_verify(const struct intel_crtc_state *state, { struct drm_i915_private *i915 = to_i915(crtc->base.dev); const struct intel_c20pll_state *mpll_sw_state = &state->cx0pll_state.c20; - bool use_mplla; + bool sw_use_mpllb = mpll_sw_state->tx[0] & C20_PHY_USE_MPLLB; + bool hw_use_mpllb = mpll_hw_state->tx[0] & C20_PHY_USE_MPLLB; int i; - use_mplla = intel_c20_use_mplla(mpll_hw_state->clock); - if (use_mplla) { - for (i = 0; i < ARRAY_SIZE(mpll_sw_state->mplla); i++) { - I915_STATE_WARN(i915, mpll_hw_state->mplla[i] != mpll_sw_state->mplla[i], - "[CRTC:%d:%s] mismatch in C20MPLLA: Register[%d] (expected 0x%04x, found 0x%04x)", - crtc->base.base.id, crtc->base.name, i, - mpll_sw_state->mplla[i], mpll_hw_state->mplla[i]); - } - } else { + I915_STATE_WARN(i915, sw_use_mpllb != hw_use_mpllb, + "[CRTC:%d:%s] mismatch in C20: Register MPLLB selection (expected %d, found %d)", + crtc->base.base.id, crtc->base.name, + sw_use_mpllb, hw_use_mpllb); + + if (hw_use_mpllb) { for (i = 0; i < ARRAY_SIZE(mpll_sw_state->mpllb); i++) { I915_STATE_WARN(i915, mpll_hw_state->mpllb[i] != mpll_sw_state->mpllb[i], "[CRTC:%d:%s] mismatch in C20MPLLB: Register[%d] (expected 0x%04x, found 0x%04x)", crtc->base.base.id, crtc->base.name, i, mpll_sw_state->mpllb[i], mpll_hw_state->mpllb[i]); } + } else { + for (i = 0; i < ARRAY_SIZE(mpll_sw_state->mplla); i++) { + I915_STATE_WARN(i915, mpll_hw_state->mplla[i] != mpll_sw_state->mplla[i], + "[CRTC:%d:%s] mismatch in C20MPLLA: Register[%d] (expected 0x%04x, found 0x%04x)", + crtc->base.base.id, crtc->base.name, i, + mpll_sw_state->mplla[i], mpll_hw_state->mplla[i]); + } } for (i = 0; i < ARRAY_SIZE(mpll_sw_state->tx); i++) { diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c index 5f091502719b..6fd4fa52253a 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.c +++ b/drivers/gpu/drm/i915/display/intel_display_power.c @@ -405,8 +405,8 @@ print_async_put_domains_state(struct i915_power_domains *power_domains) struct drm_i915_private, display.power.domains); - drm_dbg(&i915->drm, "async_put_wakeref %lu\n", - power_domains->async_put_wakeref); + drm_dbg(&i915->drm, "async_put_wakeref: %s\n", + str_yes_no(power_domains->async_put_wakeref)); print_power_domains(power_domains, "async_put_domains[0]", &power_domains->async_put_domains[0]); diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 3b2482bf683f..c3b906ebe542 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -2101,7 +2101,7 @@ static int intel_dp_dsc_compute_pipe_bpp(struct intel_dp *intel_dp, } } - dsc_max_bpc = intel_dp_dsc_min_src_input_bpc(i915); + dsc_max_bpc = intel_dp_dsc_max_src_input_bpc(i915); if (!dsc_max_bpc) return -EINVAL; diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index b6e2e70e1290..8f702c3fc62d 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -3319,11 +3319,11 @@ void intel_psr_connector_debugfs_add(struct intel_connector *connector) struct drm_i915_private *i915 = to_i915(connector->base.dev); struct dentry *root = connector->base.debugfs_entry; - if (connector->base.connector_type != DRM_MODE_CONNECTOR_eDP) { - if (!(HAS_DP20(i915) && - connector->base.connector_type == DRM_MODE_CONNECTOR_DisplayPort)) - return; - } + /* TODO: Add support for MST connectors as well. */ + if ((connector->base.connector_type != DRM_MODE_CONNECTOR_eDP && + connector->base.connector_type != DRM_MODE_CONNECTOR_DisplayPort) || + connector->mst_port) + return; debugfs_create_file("i915_psr_sink_status", 0444, root, connector, &i915_psr_sink_status_fops); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context_types.h b/drivers/gpu/drm/i915/gem/i915_gem_context_types.h index c573c067779f..03bc7f9d191b 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_context_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_context_types.h @@ -412,9 +412,9 @@ struct i915_gem_context { /** @stale: tracks stale engines to be destroyed */ struct { - /** @lock: guards engines */ + /** @stale.lock: guards engines */ spinlock_t lock; - /** @engines: list of stale engines */ + /** @stale.engines: list of stale engines */ struct list_head engines; } stale; }; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index 81a57dd52dfd..555022c0652c 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -1159,7 +1159,7 @@ static void reloc_cache_unmap(struct reloc_cache *cache) vaddr = unmask_page(cache->vaddr); if (cache->vaddr & KMAP) - kunmap_atomic(vaddr); + kunmap_local(vaddr); else io_mapping_unmap_atomic((void __iomem *)vaddr); } @@ -1175,7 +1175,7 @@ static void reloc_cache_remap(struct reloc_cache *cache, if (cache->vaddr & KMAP) { struct page *page = i915_gem_object_get_page(obj, cache->page); - vaddr = kmap_atomic(page); + vaddr = kmap_local_page(page); cache->vaddr = unmask_flags(cache->vaddr) | (unsigned long)vaddr; } else { @@ -1205,7 +1205,7 @@ static void reloc_cache_reset(struct reloc_cache *cache, struct i915_execbuffer if (cache->vaddr & CLFLUSH_AFTER) mb(); - kunmap_atomic(vaddr); + kunmap_local(vaddr); i915_gem_object_finish_access(obj); } else { struct i915_ggtt *ggtt = cache_to_ggtt(cache); @@ -1237,7 +1237,7 @@ static void *reloc_kmap(struct drm_i915_gem_object *obj, struct page *page; if (cache->vaddr) { - kunmap_atomic(unmask_page(cache->vaddr)); + kunmap_local(unmask_page(cache->vaddr)); } else { unsigned int flushes; int err; @@ -1259,7 +1259,7 @@ static void *reloc_kmap(struct drm_i915_gem_object *obj, if (!obj->mm.dirty) set_page_dirty(page); - vaddr = kmap_atomic(page); + vaddr = kmap_local_page(page); cache->vaddr = unmask_flags(cache->vaddr) | (unsigned long)vaddr; cache->page = pageno; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index 25eeeb863209..58e6c680fe0d 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -500,17 +500,15 @@ static void i915_gem_object_read_from_page_kmap(struct drm_i915_gem_object *obj, u64 offset, void *dst, int size) { pgoff_t idx = offset >> PAGE_SHIFT; - void *src_map; void *src_ptr; - src_map = kmap_atomic(i915_gem_object_get_page(obj, idx)); - - src_ptr = src_map + offset_in_page(offset); + src_ptr = kmap_local_page(i915_gem_object_get_page(obj, idx)) + + offset_in_page(offset); if (!(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_READ)) drm_clflush_virt_range(src_ptr, size); memcpy(dst, src_ptr, size); - kunmap_atomic(src_map); + kunmap_local(src_ptr); } static void diff --git a/drivers/gpu/drm/i915/gem/i915_gem_phys.c b/drivers/gpu/drm/i915/gem/i915_gem_phys.c index 5df128e2f4dc..ef85c6dc9fd5 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_phys.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_phys.c @@ -65,16 +65,13 @@ static int i915_gem_object_get_pages_phys(struct drm_i915_gem_object *obj) dst = vaddr; for (i = 0; i < obj->base.size / PAGE_SIZE; i++) { struct page *page; - void *src; page = shmem_read_mapping_page(mapping, i); if (IS_ERR(page)) goto err_st; - src = kmap_atomic(page); - memcpy(dst, src, PAGE_SIZE); + memcpy_from_page(dst, page, 0, PAGE_SIZE); drm_clflush_virt_range(dst, PAGE_SIZE); - kunmap_atomic(src); put_page(page); dst += PAGE_SIZE; @@ -113,16 +110,13 @@ i915_gem_object_put_pages_phys(struct drm_i915_gem_object *obj, for (i = 0; i < obj->base.size / PAGE_SIZE; i++) { struct page *page; - char *dst; page = shmem_read_mapping_page(mapping, i); if (IS_ERR(page)) continue; - dst = kmap_atomic(page); drm_clflush_virt_range(src, PAGE_SIZE); - memcpy(dst, src, PAGE_SIZE); - kunmap_atomic(dst); + memcpy_to_page(page, 0, src, PAGE_SIZE); set_page_dirty(page); if (obj->mm.madv == I915_MADV_WILLNEED) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c index 73a4a4eb29e0..38b72d86560f 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c @@ -485,11 +485,13 @@ shmem_pwrite(struct drm_i915_gem_object *obj, if (err < 0) return err; - vaddr = kmap_atomic(page); + vaddr = kmap_local_page(page); + pagefault_disable(); unwritten = __copy_from_user_inatomic(vaddr + pg, user_data, len); - kunmap_atomic(vaddr); + pagefault_enable(); + kunmap_local(vaddr); err = aops->write_end(obj->base.filp, mapping, offset, len, len - unwritten, page, data); diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c index 6b9f6cf50bf6..c9e6d77abab0 100644 --- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c +++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c @@ -1082,7 +1082,7 @@ __cpu_check_shmem(struct drm_i915_gem_object *obj, u32 dword, u32 val) goto err_unlock; for (n = 0; n < obj->base.size >> PAGE_SHIFT; ++n) { - u32 *ptr = kmap_atomic(i915_gem_object_get_page(obj, n)); + u32 *ptr = kmap_local_page(i915_gem_object_get_page(obj, n)); if (needs_flush & CLFLUSH_BEFORE) drm_clflush_virt_range(ptr, PAGE_SIZE); @@ -1090,12 +1090,12 @@ __cpu_check_shmem(struct drm_i915_gem_object *obj, u32 dword, u32 val) if (ptr[dword] != val) { pr_err("n=%lu ptr[%u]=%u, val=%u\n", n, dword, ptr[dword], val); - kunmap_atomic(ptr); + kunmap_local(ptr); err = -EINVAL; break; } - kunmap_atomic(ptr); + kunmap_local(ptr); } i915_gem_object_finish_access(obj); diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c index 3fd68a099a85..2a0c0634d446 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c @@ -24,7 +24,6 @@ static int cpu_set(struct context *ctx, unsigned long offset, u32 v) { unsigned int needs_clflush; struct page *page; - void *map; u32 *cpu; int err; @@ -34,8 +33,7 @@ static int cpu_set(struct context *ctx, unsigned long offset, u32 v) goto out; page = i915_gem_object_get_page(ctx->obj, offset >> PAGE_SHIFT); - map = kmap_atomic(page); - cpu = map + offset_in_page(offset); + cpu = kmap_local_page(page) + offset_in_page(offset); if (needs_clflush & CLFLUSH_BEFORE) drm_clflush_virt_range(cpu, sizeof(*cpu)); @@ -45,7 +43,7 @@ static int cpu_set(struct context *ctx, unsigned long offset, u32 v) if (needs_clflush & CLFLUSH_AFTER) drm_clflush_virt_range(cpu, sizeof(*cpu)); - kunmap_atomic(map); + kunmap_local(cpu); i915_gem_object_finish_access(ctx->obj); out: @@ -57,7 +55,6 @@ static int cpu_get(struct context *ctx, unsigned long offset, u32 *v) { unsigned int needs_clflush; struct page *page; - void *map; u32 *cpu; int err; @@ -67,15 +64,14 @@ static int cpu_get(struct context *ctx, unsigned long offset, u32 *v) goto out; page = i915_gem_object_get_page(ctx->obj, offset >> PAGE_SHIFT); - map = kmap_atomic(page); - cpu = map + offset_in_page(offset); + cpu = kmap_local_page(page) + offset_in_page(offset); if (needs_clflush & CLFLUSH_BEFORE) drm_clflush_virt_range(cpu, sizeof(*cpu)); *v = *cpu; - kunmap_atomic(map); + kunmap_local(cpu); i915_gem_object_finish_access(ctx->obj); out: diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c index 7021b6e9b219..89d4dc8b60c6 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c @@ -489,12 +489,12 @@ static int cpu_fill(struct drm_i915_gem_object *obj, u32 value) for (n = 0; n < real_page_count(obj); n++) { u32 *map; - map = kmap_atomic(i915_gem_object_get_page(obj, n)); + map = kmap_local_page(i915_gem_object_get_page(obj, n)); for (m = 0; m < DW_PER_PAGE; m++) map[m] = value; if (!has_llc) drm_clflush_virt_range(map, PAGE_SIZE); - kunmap_atomic(map); + kunmap_local(map); } i915_gem_object_finish_access(obj); @@ -520,7 +520,7 @@ static noinline int cpu_check(struct drm_i915_gem_object *obj, for (n = 0; n < real_page_count(obj); n++) { u32 *map, m; - map = kmap_atomic(i915_gem_object_get_page(obj, n)); + map = kmap_local_page(i915_gem_object_get_page(obj, n)); if (needs_flush & CLFLUSH_BEFORE) drm_clflush_virt_range(map, PAGE_SIZE); @@ -546,7 +546,7 @@ static noinline int cpu_check(struct drm_i915_gem_object *obj, } out_unmap: - kunmap_atomic(map); + kunmap_local(map); if (err) break; } diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c index e57f9390076c..d684a70f2c04 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c @@ -504,7 +504,7 @@ static int igt_dmabuf_export_vmap(void *arg) } if (memchr_inv(ptr, 0, dmabuf->size)) { - pr_err("Exported object not initialiased to zero!\n"); + pr_err("Exported object not initialised to zero!\n"); err = -EINVAL; goto out; } diff --git a/drivers/gpu/drm/i915/gt/intel_gsc.h b/drivers/gpu/drm/i915/gt/intel_gsc.h index 7ab3ca0f9f26..013c64251448 100644 --- a/drivers/gpu/drm/i915/gt/intel_gsc.h +++ b/drivers/gpu/drm/i915/gt/intel_gsc.h @@ -21,8 +21,11 @@ struct mei_aux_device; /** * struct intel_gsc - graphics security controller * - * @gem_obj: scratch memory GSC operations - * @intf : gsc interface + * @intf: gsc interface + * @intf.adev: MEI aux. device for this @intf + * @intf.gem_obj: scratch memory GSC operations + * @intf.irq: IRQ for this device (%-1 for no IRQ) + * @intf.id: this interface's id number/index */ struct intel_gsc { struct intel_gsc_intf { diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h index 9de41703fae5..50962cfd1353 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h @@ -469,6 +469,9 @@ #define XEHP_PSS_MODE2 MCR_REG(0x703c) #define SCOREBOARD_STALL_FLUSH_CONTROL REG_BIT(5) +#define XEHP_PSS_CHICKEN MCR_REG(0x7044) +#define FD_END_COLLECT REG_BIT(5) + #define GEN7_SC_INSTDONE _MMIO(0x7100) #define GEN12_SC_INSTDONE_EXTRA _MMIO(0x7104) #define GEN12_SC_INSTDONE_EXTRA2 _MMIO(0x7108) diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index 4cbf9e512645..3eacbc50caf8 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -777,6 +777,9 @@ static void dg2_ctx_workarounds_init(struct intel_engine_cs *engine, /* Wa_18019271663:dg2 */ wa_masked_en(wal, CACHE_MODE_1, MSAA_OPTIMIZATION_REDUC_DISABLE); + + /* Wa_14019877138:dg2 */ + wa_mcr_masked_en(wal, XEHP_PSS_CHICKEN, FD_END_COLLECT); } static void xelpg_ctx_gt_tuning_init(struct intel_engine_cs *engine, diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.h b/drivers/gpu/drm/i915/gt/uc/intel_guc.h index e22c12ce245a..813cc888e6fa 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.h @@ -105,61 +105,67 @@ struct intel_guc { */ struct { /** - * @lock: protects everything in submission_state, - * ce->guc_id.id, and ce->guc_id.ref when transitioning in and - * out of zero + * @submission_state.lock: protects everything in + * submission_state, ce->guc_id.id, and ce->guc_id.ref + * when transitioning in and out of zero */ spinlock_t lock; /** - * @guc_ids: used to allocate new guc_ids, single-lrc + * @submission_state.guc_ids: used to allocate new + * guc_ids, single-lrc */ struct ida guc_ids; /** - * @num_guc_ids: Number of guc_ids, selftest feature to be able - * to reduce this number while testing. + * @submission_state.num_guc_ids: Number of guc_ids, selftest + * feature to be able to reduce this number while testing. */ int num_guc_ids; /** - * @guc_ids_bitmap: used to allocate new guc_ids, multi-lrc + * @submission_state.guc_ids_bitmap: used to allocate + * new guc_ids, multi-lrc */ unsigned long *guc_ids_bitmap; /** - * @guc_id_list: list of intel_context with valid guc_ids but no - * refs + * @submission_state.guc_id_list: list of intel_context + * with valid guc_ids but no refs */ struct list_head guc_id_list; /** - * @guc_ids_in_use: Number single-lrc guc_ids in use + * @submission_state.guc_ids_in_use: Number single-lrc + * guc_ids in use */ unsigned int guc_ids_in_use; /** - * @destroyed_contexts: list of contexts waiting to be destroyed - * (deregistered with the GuC) + * @submission_state.destroyed_contexts: list of contexts + * waiting to be destroyed (deregistered with the GuC) */ struct list_head destroyed_contexts; /** - * @destroyed_worker: worker to deregister contexts, need as we - * need to take a GT PM reference and can't from destroy - * function as it might be in an atomic context (no sleeping) + * @submission_state.destroyed_worker: worker to deregister + * contexts, need as we need to take a GT PM reference and + * can't from destroy function as it might be in an atomic + * context (no sleeping) */ struct work_struct destroyed_worker; /** - * @reset_fail_worker: worker to trigger a GT reset after an - * engine reset fails + * @submission_state.reset_fail_worker: worker to trigger + * a GT reset after an engine reset fails */ struct work_struct reset_fail_worker; /** - * @reset_fail_mask: mask of engines that failed to reset + * @submission_state.reset_fail_mask: mask of engines that + * failed to reset */ intel_engine_mask_t reset_fail_mask; /** - * @sched_disable_delay_ms: schedule disable delay, in ms, for - * contexts + * @submission_state.sched_disable_delay_ms: schedule + * disable delay, in ms, for contexts */ unsigned int sched_disable_delay_ms; /** - * @sched_disable_gucid_threshold: threshold of min remaining available - * guc_ids before we start bypassing the schedule disable delay + * @submission_state.sched_disable_gucid_threshold: + * threshold of min remaining available guc_ids before + * we start bypassing the schedule disable delay */ unsigned int sched_disable_gucid_threshold; } submission_state; @@ -243,37 +249,40 @@ struct intel_guc { */ struct { /** - * @lock: Lock protecting the below fields and the engine stats. + * @timestamp.lock: Lock protecting the below fields and + * the engine stats. */ spinlock_t lock; /** - * @gt_stamp: 64 bit extended value of the GT timestamp. + * @timestamp.gt_stamp: 64-bit extended value of the GT + * timestamp. */ u64 gt_stamp; /** - * @ping_delay: Period for polling the GT timestamp for - * overflow. + * @timestamp.ping_delay: Period for polling the GT + * timestamp for overflow. */ unsigned long ping_delay; /** - * @work: Periodic work to adjust GT timestamp, engine and - * context usage for overflows. + * @timestamp.work: Periodic work to adjust GT timestamp, + * engine and context usage for overflows. */ struct delayed_work work; /** - * @shift: Right shift value for the gpm timestamp + * @timestamp.shift: Right shift value for the gpm timestamp */ u32 shift; /** - * @last_stat_jiffies: jiffies at last actual stats collection time - * We use this timestamp to ensure we don't oversample the - * stats because runtime power management events can trigger - * stats collection at much higher rates than required. + * @timestamp.last_stat_jiffies: jiffies at last actual + * stats collection time. We use this timestamp to ensure + * we don't oversample the stats because runtime power + * management events can trigger stats collection at much + * higher rates than required. */ unsigned long last_stat_jiffies; } timestamp; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c index 362639162ed6..756093eaf2ad 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c @@ -1343,16 +1343,13 @@ size_t intel_uc_fw_copy_rsa(struct intel_uc_fw *uc_fw, void *dst, u32 max_len) for_each_sgt_page(page, iter, uc_fw->obj->mm.pages) { u32 len = min_t(u32, size, PAGE_SIZE - offset); - void *vaddr; if (idx > 0) { idx--; continue; } - vaddr = kmap_atomic(page); - memcpy(dst, vaddr + offset, len); - kunmap_atomic(vaddr); + memcpy_from_page(dst, page, offset, len); offset = 0; dst += len; diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c index ddf49c2dbb91..2905df83e180 100644 --- a/drivers/gpu/drm/i915/i915_cmd_parser.c +++ b/drivers/gpu/drm/i915/i915_cmd_parser.c @@ -1211,11 +1211,11 @@ static u32 *copy_batch(struct drm_i915_gem_object *dst_obj, for (n = offset >> PAGE_SHIFT; remain; n++) { int len = min(remain, PAGE_SIZE - x); - src = kmap_atomic(i915_gem_object_get_page(src_obj, n)); + src = kmap_local_page(i915_gem_object_get_page(src_obj, n)); if (src_needs_clflush) drm_clflush_virt_range(src + x, len); memcpy(ptr, src + x, len); - kunmap_atomic(src); + kunmap_local(src); ptr += len; remain -= len; diff --git a/drivers/gpu/drm/i915/i915_hwmon.c b/drivers/gpu/drm/i915/i915_hwmon.c index 975da8e7f2a9..8c3f443c8347 100644 --- a/drivers/gpu/drm/i915/i915_hwmon.c +++ b/drivers/gpu/drm/i915/i915_hwmon.c @@ -175,7 +175,7 @@ hwm_power1_max_interval_show(struct device *dev, struct device_attribute *attr, * tau4 = (4 | x) << y * but add 2 when doing the final right shift to account for units */ - tau4 = ((1 << x_w) | x) << y; + tau4 = (u64)((1 << x_w) | x) << y; /* val in hwmon interface units (millisec) */ out = mul_u64_u32_shr(tau4, SF_TIME, hwmon->scl_shift_time + x_w); @@ -211,7 +211,7 @@ hwm_power1_max_interval_store(struct device *dev, r = FIELD_PREP(PKG_MAX_WIN, PKG_MAX_WIN_DEFAULT); x = REG_FIELD_GET(PKG_MAX_WIN_X, r); y = REG_FIELD_GET(PKG_MAX_WIN_Y, r); - tau4 = ((1 << x_w) | x) << y; + tau4 = (u64)((1 << x_w) | x) << y; max_win = mul_u64_u32_shr(tau4, SF_TIME, hwmon->scl_shift_time + x_w); if (val > max_win) diff --git a/drivers/gpu/drm/i915/i915_perf_types.h b/drivers/gpu/drm/i915/i915_perf_types.h index 13b1ae9b96c7..46445248d193 100644 --- a/drivers/gpu/drm/i915/i915_perf_types.h +++ b/drivers/gpu/drm/i915/i915_perf_types.h @@ -291,7 +291,8 @@ struct i915_perf_stream { int size_exponent; /** - * @ptr_lock: Locks reads and writes to all head/tail state + * @oa_buffer.ptr_lock: Locks reads and writes to all + * head/tail state * * Consider: the head and tail pointer state needs to be read * consistently from a hrtimer callback (atomic context) and @@ -313,7 +314,8 @@ struct i915_perf_stream { spinlock_t ptr_lock; /** - * @head: Although we can always read back the head pointer register, + * @oa_buffer.head: Although we can always read back + * the head pointer register, * we prefer to avoid trusting the HW state, just to avoid any * risk that some hardware condition could * somehow bump the * head pointer unpredictably and cause us to forward the wrong @@ -322,7 +324,8 @@ struct i915_perf_stream { u32 head; /** - * @tail: The last verified tail that can be read by userspace. + * @oa_buffer.tail: The last verified tail that can be + * read by userspace. */ u32 tail; } oa_buffer; diff --git a/drivers/gpu/drm/imagination/pvr_device.h b/drivers/gpu/drm/imagination/pvr_device.h index 2ca7e535799f..ecdd5767d8ef 100644 --- a/drivers/gpu/drm/imagination/pvr_device.h +++ b/drivers/gpu/drm/imagination/pvr_device.h @@ -193,13 +193,14 @@ struct pvr_device { * @queues: Queue-related fields. */ struct { - /** @active: Active queue list. */ + /** @queues.active: Active queue list. */ struct list_head active; - /** @idle: Idle queue list. */ + /** @queues.idle: Idle queue list. */ struct list_head idle; - /** @lock: Lock protecting access to the active/idle lists. */ + /** @queues.lock: Lock protecting access to the active/idle + * lists. */ struct mutex lock; } queues; @@ -207,18 +208,18 @@ struct pvr_device { * @watchdog: Watchdog for communications with firmware. */ struct { - /** @work: Work item for watchdog callback. */ + /** @watchdog.work: Work item for watchdog callback. */ struct delayed_work work; /** - * @old_kccb_cmds_executed: KCCB command execution count at last - * watchdog poll. + * @watchdog.old_kccb_cmds_executed: KCCB command execution + * count at last watchdog poll. */ u32 old_kccb_cmds_executed; /** - * @kccb_stall_count: Number of watchdog polls KCCB has been - * stalled for. + * @watchdog.kccb_stall_count: Number of watchdog polls + * KCCB has been stalled for. */ u32 kccb_stall_count; } watchdog; @@ -227,43 +228,46 @@ struct pvr_device { * @kccb: Circular buffer for communications with firmware. */ struct { - /** @ccb: Kernel CCB. */ + /** @kccb.ccb: Kernel CCB. */ struct pvr_ccb ccb; - /** @rtn_q: Waitqueue for KCCB command return waiters. */ + /** @kccb.rtn_q: Waitqueue for KCCB command return waiters. */ wait_queue_head_t rtn_q; - /** @rtn_obj: Object representing KCCB return slots. */ + /** @kccb.rtn_obj: Object representing KCCB return slots. */ struct pvr_fw_object *rtn_obj; /** - * @rtn: Pointer to CPU mapping of KCCB return slots. Must be - * accessed by READ_ONCE()/WRITE_ONCE(). + * @kccb.rtn: Pointer to CPU mapping of KCCB return slots. + * Must be accessed by READ_ONCE()/WRITE_ONCE(). */ u32 *rtn; - /** @slot_count: Total number of KCCB slots available. */ + /** @kccb.slot_count: Total number of KCCB slots available. */ u32 slot_count; - /** @reserved_count: Number of KCCB slots reserved for future use. */ + /** @kccb.reserved_count: Number of KCCB slots reserved for + * future use. */ u32 reserved_count; /** - * @waiters: List of KCCB slot waiters. + * @kccb.waiters: List of KCCB slot waiters. */ struct list_head waiters; - /** @fence_ctx: KCCB fence context. */ + /** @kccb.fence_ctx: KCCB fence context. */ struct { - /** @id: KCCB fence context ID allocated with dma_fence_context_alloc(). */ + /** @kccb.fence_ctx.id: KCCB fence context ID + * allocated with dma_fence_context_alloc(). */ u64 id; - /** @seqno: Sequence number incremented each time a fence is created. */ + /** @kccb.fence_ctx.seqno: Sequence number incremented + * each time a fence is created. */ atomic_t seqno; /** - * @lock: Lock used to synchronize access to fences allocated by this - * context. + * @kccb.fence_ctx.lock: Lock used to synchronize + * access to fences allocated by this context. */ spinlock_t lock; } fence_ctx; diff --git a/drivers/gpu/drm/nouveau/nouveau_vmm.c b/drivers/gpu/drm/nouveau/nouveau_vmm.c index a6602c012671..3dda885df5b2 100644 --- a/drivers/gpu/drm/nouveau/nouveau_vmm.c +++ b/drivers/gpu/drm/nouveau/nouveau_vmm.c @@ -108,6 +108,9 @@ nouveau_vma_new(struct nouveau_bo *nvbo, struct nouveau_vmm *vmm, } else { ret = nvif_vmm_get(&vmm->vmm, PTES, false, mem->mem.page, 0, mem->mem.size, &tmp); + if (ret) + goto done; + vma->addr = tmp.addr; } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/ga100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/ga100.c index c8ce7ff18713..e74493a4569e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/ga100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/ga100.c @@ -550,6 +550,10 @@ ga100_fifo_nonstall_ctor(struct nvkm_fifo *fifo) struct nvkm_engn *engn = list_first_entry(&runl->engns, typeof(*engn), head); runl->nonstall.vector = engn->func->nonstall(engn); + + /* if no nonstall vector just keep going */ + if (runl->nonstall.vector == -1) + continue; if (runl->nonstall.vector < 0) { RUNL_ERROR(runl, "nonstall %d", runl->nonstall.vector); return runl->nonstall.vector; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/r535.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/r535.c index d088e636edc3..de2ebe8f2134 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/r535.c @@ -350,7 +350,7 @@ r535_engn_nonstall(struct nvkm_engn *engn) int ret; ret = nvkm_gsp_intr_nonstall(subdev->device->gsp, subdev->type, subdev->inst); - WARN_ON(ret < 0); + WARN_ON(ret == -ENOENT); return ret; } diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c index 04bceaa28a19..da1bebb896f7 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c @@ -25,12 +25,8 @@ int nvkm_gsp_intr_nonstall(struct nvkm_gsp *gsp, enum nvkm_subdev_type type, int inst) { for (int i = 0; i < gsp->intr_nr; i++) { - if (gsp->intr[i].type == type && gsp->intr[i].inst == inst) { - if (gsp->intr[i].nonstall != ~0) - return gsp->intr[i].nonstall; - - return -EINVAL; - } + if (gsp->intr[i].type == type && gsp->intr[i].inst == inst) + return gsp->intr[i].nonstall; } return -ENOENT; diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c index 44508c2dd614..85b3b4871a1d 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c @@ -35,7 +35,6 @@ #include "rockchip_drm_drv.h" #include "rockchip_drm_gem.h" -#include "rockchip_drm_fb.h" #include "rockchip_drm_vop2.h" #include "rockchip_rgb.h" @@ -959,12 +958,6 @@ static void vop2_enable(struct vop2 *vop2) return; } - ret = regmap_reinit_cache(vop2->map, &vop2_regmap_config); - if (ret) { - drm_err(vop2->drm, "failed to reinit cache: %d\n", ret); - return; - } - if (vop2->data->soc_id == 3566) vop2_writel(vop2, RK3568_OTP_WIN_EN, 1); @@ -996,6 +989,8 @@ static void vop2_disable(struct vop2 *vop2) pm_runtime_put_sync(vop2->dev); + regcache_drop_region(vop2->map, 0, vop2_regmap_config.max_register); + clk_disable_unprepare(vop2->pclk); clk_disable_unprepare(vop2->aclk); clk_disable_unprepare(vop2->hclk); @@ -1685,7 +1680,6 @@ static unsigned long rk3588_calc_cru_cfg(struct vop2_video_port *vp, int id, unsigned long dclk_core_rate = v_pixclk >> 2; unsigned long dclk_rate = v_pixclk; unsigned long dclk_out_rate; - unsigned long if_dclk_rate; unsigned long if_pixclk_rate; int K = 1; @@ -1700,13 +1694,13 @@ static unsigned long rk3588_calc_cru_cfg(struct vop2_video_port *vp, int id, } if_pixclk_rate = (dclk_core_rate << 1) / K; - if_dclk_rate = dclk_core_rate / K; /* + * if_dclk_rate = dclk_core_rate / K; * *if_pixclk_div = dclk_rate / if_pixclk_rate; * *if_dclk_div = dclk_rate / if_dclk_rate; */ - *if_pixclk_div = 2; - *if_dclk_div = 4; + *if_pixclk_div = 2; + *if_dclk_div = 4; } else if (vop2_output_if_is_edp(id)) { /* * edp_pixclk = edp_dclk > dclk_core diff --git a/drivers/gpu/drm/v3d/v3d_debugfs.c b/drivers/gpu/drm/v3d/v3d_debugfs.c index f843a50d5dce..94eafcecc65b 100644 --- a/drivers/gpu/drm/v3d/v3d_debugfs.c +++ b/drivers/gpu/drm/v3d/v3d_debugfs.c @@ -62,9 +62,9 @@ static const struct v3d_reg_def v3d_core_reg_defs[] = { REGDEF(33, 71, V3D_PTB_BPCA), REGDEF(33, 71, V3D_PTB_BPCS), - REGDEF(33, 41, V3D_GMP_STATUS(33)), - REGDEF(33, 41, V3D_GMP_CFG(33)), - REGDEF(33, 41, V3D_GMP_VIO_ADDR(33)), + REGDEF(33, 42, V3D_GMP_STATUS(33)), + REGDEF(33, 42, V3D_GMP_CFG(33)), + REGDEF(33, 42, V3D_GMP_VIO_ADDR(33)), REGDEF(33, 71, V3D_ERR_FDBGO), REGDEF(33, 71, V3D_ERR_FDBGB), @@ -74,13 +74,13 @@ static const struct v3d_reg_def v3d_core_reg_defs[] = { static const struct v3d_reg_def v3d_csd_reg_defs[] = { REGDEF(41, 71, V3D_CSD_STATUS), - REGDEF(41, 41, V3D_CSD_CURRENT_CFG0(41)), - REGDEF(41, 41, V3D_CSD_CURRENT_CFG1(41)), - REGDEF(41, 41, V3D_CSD_CURRENT_CFG2(41)), - REGDEF(41, 41, V3D_CSD_CURRENT_CFG3(41)), - REGDEF(41, 41, V3D_CSD_CURRENT_CFG4(41)), - REGDEF(41, 41, V3D_CSD_CURRENT_CFG5(41)), - REGDEF(41, 41, V3D_CSD_CURRENT_CFG6(41)), + REGDEF(41, 42, V3D_CSD_CURRENT_CFG0(41)), + REGDEF(41, 42, V3D_CSD_CURRENT_CFG1(41)), + REGDEF(41, 42, V3D_CSD_CURRENT_CFG2(41)), + REGDEF(41, 42, V3D_CSD_CURRENT_CFG3(41)), + REGDEF(41, 42, V3D_CSD_CURRENT_CFG4(41)), + REGDEF(41, 42, V3D_CSD_CURRENT_CFG5(41)), + REGDEF(41, 42, V3D_CSD_CURRENT_CFG6(41)), REGDEF(71, 71, V3D_CSD_CURRENT_CFG0(71)), REGDEF(71, 71, V3D_CSD_CURRENT_CFG1(71)), REGDEF(71, 71, V3D_CSD_CURRENT_CFG2(71)), |