diff options
Diffstat (limited to 'drivers/gpu')
176 files changed, 2514 insertions, 1480 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/aldebaran.c b/drivers/gpu/drm/amd/amdgpu/aldebaran.c index c6cc493a5486..2b97b8a96fb4 100644 --- a/drivers/gpu/drm/amd/amdgpu/aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/aldebaran.c @@ -148,30 +148,22 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, struct amdgpu_reset_context *reset_context) { struct amdgpu_device *adev = (struct amdgpu_device *)reset_ctl->handle; + struct list_head *reset_device_list = reset_context->reset_device_list; struct amdgpu_device *tmp_adev = NULL; - struct list_head reset_device_list; int r = 0; dev_dbg(adev->dev, "aldebaran perform hw reset\n"); + + if (reset_device_list == NULL) + return -EINVAL; + if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2) && reset_context->hive == NULL) { /* Wrong context, return error */ return -EINVAL; } - INIT_LIST_HEAD(&reset_device_list); - if (reset_context->hive) { - list_for_each_entry (tmp_adev, - &reset_context->hive->device_list, - gmc.xgmi.head) - list_add_tail(&tmp_adev->reset_list, - &reset_device_list); - } else { - list_add_tail(&reset_context->reset_req_dev->reset_list, - &reset_device_list); - } - - list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { mutex_lock(&tmp_adev->reset_cntl->reset_lock); tmp_adev->reset_cntl->active_reset = AMD_RESET_METHOD_MODE2; } @@ -179,7 +171,7 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, * Mode2 reset doesn't need any sync between nodes in XGMI hive, instead launch * them together so that they can be completed asynchronously on multiple nodes */ - list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { /* For XGMI run all resets in parallel to speed up the process */ if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { if (!queue_work(system_unbound_wq, @@ -197,7 +189,7 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, /* For XGMI wait for all resets to complete before proceed */ if (!r) { - list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { flush_work(&tmp_adev->reset_cntl->reset_work); r = tmp_adev->asic_reset_res; @@ -207,7 +199,7 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, } } - list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { mutex_unlock(&tmp_adev->reset_cntl->reset_lock); tmp_adev->reset_cntl->active_reset = AMD_RESET_METHOD_NONE; } @@ -339,10 +331,13 @@ static int aldebaran_mode2_restore_hwcontext(struct amdgpu_reset_control *reset_ctl, struct amdgpu_reset_context *reset_context) { + struct list_head *reset_device_list = reset_context->reset_device_list; struct amdgpu_device *tmp_adev = NULL; - struct list_head reset_device_list; int r; + if (reset_device_list == NULL) + return -EINVAL; + if (reset_context->reset_req_dev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2) && reset_context->hive == NULL) { @@ -350,19 +345,7 @@ aldebaran_mode2_restore_hwcontext(struct amdgpu_reset_control *reset_ctl, return -EINVAL; } - INIT_LIST_HEAD(&reset_device_list); - if (reset_context->hive) { - list_for_each_entry (tmp_adev, - &reset_context->hive->device_list, - gmc.xgmi.head) - list_add_tail(&tmp_adev->reset_list, - &reset_device_list); - } else { - list_add_tail(&reset_context->reset_req_dev->reset_list, - &reset_device_list); - } - - list_for_each_entry (tmp_adev, &reset_device_list, reset_list) { + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { dev_info(tmp_adev->dev, "GPU reset succeeded, trying to resume\n"); r = aldebaran_mode2_restore_ip(tmp_adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index e146810c700b..d597e2656c47 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -317,7 +317,7 @@ enum amdgpu_kiq_irq { AMDGPU_CP_KIQ_IRQ_DRIVER0 = 0, AMDGPU_CP_KIQ_IRQ_LAST }; - +#define SRIOV_USEC_TIMEOUT 1200000 /* wait 12 * 100ms for SRIOV */ #define MAX_KIQ_REG_WAIT 5000 /* in usecs, 5ms */ #define MAX_KIQ_REG_BAILOUT_INTERVAL 5 /* in msecs, 5ms */ #define MAX_KIQ_REG_TRY 1000 diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 3c09dcc0986e..647220a8762d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -96,6 +96,7 @@ struct amdgpu_amdkfd_fence { struct amdgpu_kfd_dev { struct kfd_dev *dev; uint64_t vram_used; + uint64_t vram_used_aligned; bool init_complete; struct work_struct reset_work; }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index a699134a1e8c..cbd593f7d553 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -40,10 +40,10 @@ #define AMDGPU_USERPTR_RESTORE_DELAY_MS 1 /* - * Align VRAM allocations to 2MB to avoid fragmentation caused by 4K allocations in the tail 2MB + * Align VRAM availability to 2MB to avoid fragmentation caused by 4K allocations in the tail 2MB * BO chunk */ -#define VRAM_ALLOCATION_ALIGN (1 << 21) +#define VRAM_AVAILABLITY_ALIGN (1 << 21) /* Impose limit on how much memory KFD can use */ static struct { @@ -149,7 +149,7 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev, * to avoid fragmentation caused by 4K allocations in the tail * 2M BO chunk. */ - vram_needed = ALIGN(size, VRAM_ALLOCATION_ALIGN); + vram_needed = size; } else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { system_mem_needed = size; } else if (!(alloc_flag & @@ -182,8 +182,10 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev, */ WARN_ONCE(vram_needed && !adev, "adev reference can't be null when vram is used"); - if (adev) + if (adev) { adev->kfd.vram_used += vram_needed; + adev->kfd.vram_used_aligned += ALIGN(vram_needed, VRAM_AVAILABLITY_ALIGN); + } kfd_mem_limit.system_mem_used += system_mem_needed; kfd_mem_limit.ttm_mem_used += ttm_mem_needed; @@ -203,8 +205,10 @@ void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev, } else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { WARN_ONCE(!adev, "adev reference can't be null when alloc mem flags vram is set"); - if (adev) - adev->kfd.vram_used -= ALIGN(size, VRAM_ALLOCATION_ALIGN); + if (adev) { + adev->kfd.vram_used -= size; + adev->kfd.vram_used_aligned -= ALIGN(size, VRAM_AVAILABLITY_ALIGN); + } } else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { kfd_mem_limit.system_mem_used -= size; } else if (!(alloc_flag & @@ -1608,15 +1612,14 @@ size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev) uint64_t reserved_for_pt = ESTIMATE_PT_SIZE(amdgpu_amdkfd_total_mem_size); size_t available; - spin_lock(&kfd_mem_limit.mem_limit_lock); available = adev->gmc.real_vram_size - - adev->kfd.vram_used + - adev->kfd.vram_used_aligned - atomic64_read(&adev->vram_pin_size) - reserved_for_pt; spin_unlock(&kfd_mem_limit.mem_limit_lock); - return ALIGN_DOWN(available, VRAM_ALLOCATION_ALIGN); + return ALIGN_DOWN(available, VRAM_AVAILABLITY_ALIGN); } int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c index fd8f3731758e..b81b77a9efa6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c @@ -314,7 +314,7 @@ amdgpu_atomfirmware_get_vram_info(struct amdgpu_device *adev, mem_channel_number = vram_info->v30.channel_num; mem_channel_width = vram_info->v30.channel_width; if (vram_width) - *vram_width = mem_channel_number * mem_channel_width; + *vram_width = mem_channel_number * (1 << mem_channel_width); break; default: return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index d8f1335bc68f..b7bae833c804 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -837,16 +837,12 @@ static int amdgpu_cs_vm_handling(struct amdgpu_cs_parser *p) continue; r = amdgpu_vm_bo_update(adev, bo_va, false); - if (r) { - mutex_unlock(&p->bo_list->bo_list_mutex); + if (r) return r; - } r = amdgpu_sync_fence(&p->job->sync, bo_va->last_pt_update); - if (r) { - mutex_unlock(&p->bo_list->bo_list_mutex); + if (r) return r; - } } r = amdgpu_vm_handle_moved(adev, vm); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index e2eec985adb3..cb00c7d6f50b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -1705,7 +1705,7 @@ static ssize_t amdgpu_reset_dump_register_list_write(struct file *f, { struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private; char reg_offset[11]; - uint32_t *new, *tmp = NULL; + uint32_t *new = NULL, *tmp = NULL; int ret, i = 0, len = 0; do { @@ -1747,7 +1747,8 @@ static ssize_t amdgpu_reset_dump_register_list_write(struct file *f, ret = size; error_free: - kfree(tmp); + if (tmp != new) + kfree(tmp); kfree(new); return ret; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index c4a6fe3070b6..1400abee9f40 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2456,12 +2456,14 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) if (!hive->reset_domain || !amdgpu_reset_get_reset_domain(hive->reset_domain)) { r = -ENOENT; + amdgpu_put_xgmi_hive(hive); goto init_failed; } /* Drop the early temporary reset domain we created for device */ amdgpu_reset_put_reset_domain(adev->reset_domain); adev->reset_domain = hive->reset_domain; + amdgpu_put_xgmi_hive(hive); } } @@ -4413,8 +4415,6 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev, retry: amdgpu_amdkfd_pre_reset(adev); - amdgpu_amdkfd_pre_reset(adev); - if (from_hypervisor) r = amdgpu_virt_request_full_gpu(adev, true); else @@ -4742,6 +4742,8 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle, tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device, reset_list); amdgpu_reset_reg_dumps(tmp_adev); + + reset_context->reset_device_list = device_list_handle; r = amdgpu_reset_perform_reset(tmp_adev, reset_context); /* If reset handler not implemented, continue; otherwise return */ if (r == -ENOSYS) @@ -5522,7 +5524,8 @@ bool amdgpu_device_is_peer_accessible(struct amdgpu_device *adev, ~*peer_adev->dev->dma_mask : ~((1ULL << 32) - 1); resource_size_t aper_limit = adev->gmc.aper_base + adev->gmc.aper_size - 1; - bool p2p_access = !(pci_p2pdma_distance_many(adev->pdev, + bool p2p_access = !adev->gmc.xgmi.connected_to_cpu && + !(pci_p2pdma_distance_many(adev->pdev, &peer_adev->dev, 1, true) < 0); return pcie_p2p && p2p_access && (adev->gmc.visible_vram_size && diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c index ecada5eadfe3..e325150879df 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c @@ -66,10 +66,15 @@ static bool is_fru_eeprom_supported(struct amdgpu_device *adev) return true; case CHIP_SIENNA_CICHLID: if (strnstr(atom_ctx->vbios_version, "D603", + sizeof(atom_ctx->vbios_version))) { + if (strnstr(atom_ctx->vbios_version, "D603GLXE", sizeof(atom_ctx->vbios_version))) - return true; - else + return false; + else + return true; + } else { return false; + } default: return false; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index 5071b96be982..c2fd6f3076a6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -159,7 +159,10 @@ void amdgpu_job_free(struct amdgpu_job *job) amdgpu_sync_free(&job->sync); amdgpu_sync_free(&job->sched_sync); - dma_fence_put(&job->hw_fence); + if (!job->hw_fence.ops) + kfree(job); + else + dma_fence_put(&job->hw_fence); } int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity, @@ -272,10 +275,6 @@ void amdgpu_job_stop_all_jobs_on_sched(struct drm_gpu_scheduler *sched) /* Signal all jobs not yet scheduled */ for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) { struct drm_sched_rq *rq = &sched->sched_rq[i]; - - if (!rq) - continue; - spin_lock(&rq->lock); list_for_each_entry(s_entity, &rq->entities, list) { while ((s_job = to_drm_sched_job(spsc_queue_pop(&s_entity->job_queue)))) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index b067ce45d226..9f7a5e393f85 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -2401,7 +2401,7 @@ static int psp_load_smu_fw(struct psp_context *psp) static bool fw_load_skip_check(struct psp_context *psp, struct amdgpu_firmware_info *ucode) { - if (!ucode->fw) + if (!ucode->fw || !ucode->ucode_size) return true; if (ucode->ucode_id == AMDGPU_UCODE_ID_SMC && @@ -2641,6 +2641,9 @@ static int psp_hw_fini(void *handle) psp_rap_terminate(psp); psp_dtm_terminate(psp); psp_hdcp_terminate(psp); + + if (adev->gmc.xgmi.num_physical_nodes > 1) + psp_xgmi_terminate(psp); } psp_asd_terminate(psp); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h index 9e55a5d7a825..ffda1560c648 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h @@ -37,6 +37,7 @@ struct amdgpu_reset_context { struct amdgpu_device *reset_req_dev; struct amdgpu_job *job; struct amdgpu_hive_info *hive; + struct list_head *reset_device_list; unsigned long flags; }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 3b4c19412625..134575a3893c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -637,6 +637,8 @@ struct amdgpu_ttm_tt { #endif }; +#define ttm_to_amdgpu_ttm_tt(ptr) container_of(ptr, struct amdgpu_ttm_tt, ttm) + #ifdef CONFIG_DRM_AMDGPU_USERPTR /* * amdgpu_ttm_tt_get_user_pages - get device accessible pages that back user @@ -648,7 +650,7 @@ struct amdgpu_ttm_tt { int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) { struct ttm_tt *ttm = bo->tbo.ttm; - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); unsigned long start = gtt->userptr; struct vm_area_struct *vma; struct mm_struct *mm; @@ -702,7 +704,7 @@ out_unlock: */ bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm) { - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); bool r = false; if (!gtt || !gtt->userptr) @@ -751,7 +753,7 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_device *bdev, struct ttm_tt *ttm) { struct amdgpu_device *adev = amdgpu_ttm_adev(bdev); - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY); enum dma_data_direction direction = write ? DMA_BIDIRECTIONAL : DMA_TO_DEVICE; @@ -788,7 +790,7 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_device *bdev, struct ttm_tt *ttm) { struct amdgpu_device *adev = amdgpu_ttm_adev(bdev); - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY); enum dma_data_direction direction = write ? DMA_BIDIRECTIONAL : DMA_TO_DEVICE; @@ -822,7 +824,7 @@ static void amdgpu_ttm_gart_bind(struct amdgpu_device *adev, { struct amdgpu_bo *abo = ttm_to_amdgpu_bo(tbo); struct ttm_tt *ttm = tbo->ttm; - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); if (amdgpu_bo_encrypted(abo)) flags |= AMDGPU_PTE_TMZ; @@ -860,7 +862,7 @@ static int amdgpu_ttm_backend_bind(struct ttm_device *bdev, struct ttm_resource *bo_mem) { struct amdgpu_device *adev = amdgpu_ttm_adev(bdev); - struct amdgpu_ttm_tt *gtt = (void*)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); uint64_t flags; int r; @@ -927,7 +929,7 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo) { struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev); struct ttm_operation_ctx ctx = { false, false }; - struct amdgpu_ttm_tt *gtt = (void *)bo->ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(bo->ttm); struct ttm_placement placement; struct ttm_place placements; struct ttm_resource *tmp; @@ -998,7 +1000,7 @@ static void amdgpu_ttm_backend_unbind(struct ttm_device *bdev, struct ttm_tt *ttm) { struct amdgpu_device *adev = amdgpu_ttm_adev(bdev); - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); /* if the pages have userptr pinning then clear that first */ if (gtt->userptr) { @@ -1025,7 +1027,7 @@ static void amdgpu_ttm_backend_unbind(struct ttm_device *bdev, static void amdgpu_ttm_backend_destroy(struct ttm_device *bdev, struct ttm_tt *ttm) { - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); if (gtt->usertask) put_task_struct(gtt->usertask); @@ -1079,7 +1081,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_device *bdev, struct ttm_operation_ctx *ctx) { struct amdgpu_device *adev = amdgpu_ttm_adev(bdev); - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); pgoff_t i; int ret; @@ -1113,7 +1115,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_device *bdev, static void amdgpu_ttm_tt_unpopulate(struct ttm_device *bdev, struct ttm_tt *ttm) { - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); struct amdgpu_device *adev; pgoff_t i; @@ -1182,7 +1184,7 @@ int amdgpu_ttm_tt_set_userptr(struct ttm_buffer_object *bo, /* Set TTM_TT_FLAG_EXTERNAL before populate but after create. */ bo->ttm->page_flags |= TTM_TT_FLAG_EXTERNAL; - gtt = (void *)bo->ttm; + gtt = ttm_to_amdgpu_ttm_tt(bo->ttm); gtt->userptr = addr; gtt->userflags = flags; @@ -1199,7 +1201,7 @@ int amdgpu_ttm_tt_set_userptr(struct ttm_buffer_object *bo, */ struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm) { - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); if (gtt == NULL) return NULL; @@ -1218,7 +1220,7 @@ struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm) bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start, unsigned long end, unsigned long *userptr) { - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); unsigned long size; if (gtt == NULL || !gtt->userptr) @@ -1241,7 +1243,7 @@ bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start, */ bool amdgpu_ttm_tt_is_userptr(struct ttm_tt *ttm) { - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); if (gtt == NULL || !gtt->userptr) return false; @@ -1254,7 +1256,7 @@ bool amdgpu_ttm_tt_is_userptr(struct ttm_tt *ttm) */ bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm) { - struct amdgpu_ttm_tt *gtt = (void *)ttm; + struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); if (gtt == NULL) return false; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c index 108e8e8a1a36..576849e95296 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c @@ -496,8 +496,7 @@ static int amdgpu_vkms_sw_init(void *handle) adev_to_drm(adev)->mode_config.max_height = YRES_MAX; adev_to_drm(adev)->mode_config.preferred_depth = 24; - /* disable prefer shadow for now due to hibernation issues */ - adev_to_drm(adev)->mode_config.prefer_shadow = 0; + adev_to_drm(adev)->mode_config.prefer_shadow = 1; adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 1b108d03e785..f2aebbf3fbe3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -742,7 +742,7 @@ int amdgpu_xgmi_remove_device(struct amdgpu_device *adev) amdgpu_put_xgmi_hive(hive); } - return psp_xgmi_terminate(&adev->psp); + return 0; } static int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c index 33a8a7365aef..f0e235f98afb 100644 --- a/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c @@ -28,13 +28,44 @@ #include "navi10_enum.h" #include "soc15_common.h" +#define regATHUB_MISC_CNTL_V3_0_1 0x00d7 +#define regATHUB_MISC_CNTL_V3_0_1_BASE_IDX 0 + + +static uint32_t athub_v3_0_get_cg_cntl(struct amdgpu_device *adev) +{ + uint32_t data; + + switch (adev->ip_versions[ATHUB_HWIP][0]) { + case IP_VERSION(3, 0, 1): + data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL_V3_0_1); + break; + default: + data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL); + break; + } + return data; +} + +static void athub_v3_0_set_cg_cntl(struct amdgpu_device *adev, uint32_t data) +{ + switch (adev->ip_versions[ATHUB_HWIP][0]) { + case IP_VERSION(3, 0, 1): + WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL_V3_0_1, data); + break; + default: + WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL, data); + break; + } +} + static void athub_v3_0_update_medium_grain_clock_gating(struct amdgpu_device *adev, bool enable) { uint32_t def, data; - def = data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL); + def = data = athub_v3_0_get_cg_cntl(adev); if (enable && (adev->cg_flags & AMD_CG_SUPPORT_ATHUB_MGCG)) data |= ATHUB_MISC_CNTL__CG_ENABLE_MASK; @@ -42,7 +73,7 @@ athub_v3_0_update_medium_grain_clock_gating(struct amdgpu_device *adev, data &= ~ATHUB_MISC_CNTL__CG_ENABLE_MASK; if (def != data) - WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL, data); + athub_v3_0_set_cg_cntl(adev, data); } static void @@ -51,7 +82,7 @@ athub_v3_0_update_medium_grain_light_sleep(struct amdgpu_device *adev, { uint32_t def, data; - def = data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL); + def = data = athub_v3_0_get_cg_cntl(adev); if (enable && (adev->cg_flags & AMD_CG_SUPPORT_ATHUB_LS)) data |= ATHUB_MISC_CNTL__CG_MEM_LS_ENABLE_MASK; @@ -59,7 +90,7 @@ athub_v3_0_update_medium_grain_light_sleep(struct amdgpu_device *adev, data &= ~ATHUB_MISC_CNTL__CG_MEM_LS_ENABLE_MASK; if (def != data) - WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL, data); + athub_v3_0_set_cg_cntl(adev, data); } int athub_v3_0_set_clockgating(struct amdgpu_device *adev, @@ -70,6 +101,7 @@ int athub_v3_0_set_clockgating(struct amdgpu_device *adev, switch (adev->ip_versions[ATHUB_HWIP][0]) { case IP_VERSION(3, 0, 0): + case IP_VERSION(3, 0, 1): case IP_VERSION(3, 0, 2): athub_v3_0_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE); @@ -88,7 +120,7 @@ void athub_v3_0_get_clockgating(struct amdgpu_device *adev, u64 *flags) int data; /* AMD_CG_SUPPORT_ATHUB_MGCG */ - data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL); + data = athub_v3_0_get_cg_cntl(adev); if (data & ATHUB_MISC_CNTL__CG_ENABLE_MASK) *flags |= AMD_CG_SUPPORT_ATHUB_MGCG; diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 9c964cd3b5d4..288fce7dc0ed 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -2796,8 +2796,7 @@ static int dce_v10_0_sw_init(void *handle) adev_to_drm(adev)->mode_config.max_height = 16384; adev_to_drm(adev)->mode_config.preferred_depth = 24; - /* disable prefer shadow for now due to hibernation issues */ - adev_to_drm(adev)->mode_config.prefer_shadow = 0; + adev_to_drm(adev)->mode_config.prefer_shadow = 1; adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true; diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index e0ad9f27dc3f..cbe5250b31cb 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -2914,8 +2914,7 @@ static int dce_v11_0_sw_init(void *handle) adev_to_drm(adev)->mode_config.max_height = 16384; adev_to_drm(adev)->mode_config.preferred_depth = 24; - /* disable prefer shadow for now due to hibernation issues */ - adev_to_drm(adev)->mode_config.prefer_shadow = 0; + adev_to_drm(adev)->mode_config.prefer_shadow = 1; adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true; diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index 77f5e998a120..b1c44fab074f 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c @@ -2673,8 +2673,7 @@ static int dce_v6_0_sw_init(void *handle) adev_to_drm(adev)->mode_config.max_width = 16384; adev_to_drm(adev)->mode_config.max_height = 16384; adev_to_drm(adev)->mode_config.preferred_depth = 24; - /* disable prefer shadow for now due to hibernation issues */ - adev_to_drm(adev)->mode_config.prefer_shadow = 0; + adev_to_drm(adev)->mode_config.prefer_shadow = 1; adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true; adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base; diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index 802e5c753271..a22b45c92792 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -2693,8 +2693,11 @@ static int dce_v8_0_sw_init(void *handle) adev_to_drm(adev)->mode_config.max_height = 16384; adev_to_drm(adev)->mode_config.preferred_depth = 24; - /* disable prefer shadow for now due to hibernation issues */ - adev_to_drm(adev)->mode_config.prefer_shadow = 0; + if (adev->asic_type == CHIP_HAWAII) + /* disable prefer shadow for now due to hibernation issues */ + adev_to_drm(adev)->mode_config.prefer_shadow = 0; + else + adev_to_drm(adev)->mode_config.prefer_shadow = 1; adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index fafbad3cf08d..a3cd5c1e8529 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -4274,35 +4274,45 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev) } - info = &adev->firmware.ucode[AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS]; - info->ucode_id = AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS; - info->fw = adev->gfx.rlc_fw; - adev->firmware.fw_size += - ALIGN(adev->gfx.rlc.global_tap_delays_ucode_size_bytes, PAGE_SIZE); + if (adev->gfx.rlc.global_tap_delays_ucode_size_bytes) { + info = &adev->firmware.ucode[AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS]; + info->ucode_id = AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS; + info->fw = adev->gfx.rlc_fw; + adev->firmware.fw_size += + ALIGN(adev->gfx.rlc.global_tap_delays_ucode_size_bytes, PAGE_SIZE); + } - info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE0_TAP_DELAYS]; - info->ucode_id = AMDGPU_UCODE_ID_SE0_TAP_DELAYS; - info->fw = adev->gfx.rlc_fw; - adev->firmware.fw_size += - ALIGN(adev->gfx.rlc.se0_tap_delays_ucode_size_bytes, PAGE_SIZE); + if (adev->gfx.rlc.se0_tap_delays_ucode_size_bytes) { + info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE0_TAP_DELAYS]; + info->ucode_id = AMDGPU_UCODE_ID_SE0_TAP_DELAYS; + info->fw = adev->gfx.rlc_fw; + adev->firmware.fw_size += + ALIGN(adev->gfx.rlc.se0_tap_delays_ucode_size_bytes, PAGE_SIZE); + } - info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE1_TAP_DELAYS]; - info->ucode_id = AMDGPU_UCODE_ID_SE1_TAP_DELAYS; - info->fw = adev->gfx.rlc_fw; - adev->firmware.fw_size += - ALIGN(adev->gfx.rlc.se1_tap_delays_ucode_size_bytes, PAGE_SIZE); + if (adev->gfx.rlc.se1_tap_delays_ucode_size_bytes) { + info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE1_TAP_DELAYS]; + info->ucode_id = AMDGPU_UCODE_ID_SE1_TAP_DELAYS; + info->fw = adev->gfx.rlc_fw; + adev->firmware.fw_size += + ALIGN(adev->gfx.rlc.se1_tap_delays_ucode_size_bytes, PAGE_SIZE); + } - info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE2_TAP_DELAYS]; - info->ucode_id = AMDGPU_UCODE_ID_SE2_TAP_DELAYS; - info->fw = adev->gfx.rlc_fw; - adev->firmware.fw_size += - ALIGN(adev->gfx.rlc.se2_tap_delays_ucode_size_bytes, PAGE_SIZE); + if (adev->gfx.rlc.se2_tap_delays_ucode_size_bytes) { + info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE2_TAP_DELAYS]; + info->ucode_id = AMDGPU_UCODE_ID_SE2_TAP_DELAYS; + info->fw = adev->gfx.rlc_fw; + adev->firmware.fw_size += + ALIGN(adev->gfx.rlc.se2_tap_delays_ucode_size_bytes, PAGE_SIZE); + } - info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE3_TAP_DELAYS]; - info->ucode_id = AMDGPU_UCODE_ID_SE3_TAP_DELAYS; - info->fw = adev->gfx.rlc_fw; - adev->firmware.fw_size += - ALIGN(adev->gfx.rlc.se3_tap_delays_ucode_size_bytes, PAGE_SIZE); + if (adev->gfx.rlc.se3_tap_delays_ucode_size_bytes) { + info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE3_TAP_DELAYS]; + info->ucode_id = AMDGPU_UCODE_ID_SE3_TAP_DELAYS; + info->fw = adev->gfx.rlc_fw; + adev->firmware.fw_size += + ALIGN(adev->gfx.rlc.se3_tap_delays_ucode_size_bytes, PAGE_SIZE); + } info = &adev->firmware.ucode[AMDGPU_UCODE_ID_CP_MEC1]; info->ucode_id = AMDGPU_UCODE_ID_CP_MEC1; @@ -4846,7 +4856,7 @@ static int gfx_v10_0_sw_init(void *handle) case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): adev->gfx.me.num_me = 1; - adev->gfx.me.num_pipe_per_me = 2; + adev->gfx.me.num_pipe_per_me = 1; adev->gfx.me.num_queue_per_pipe = 1; adev->gfx.mec.num_mec = 2; adev->gfx.mec.num_pipe_per_mec = 4; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index 6fd71cb10e54..f6b1bb40e503 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -53,6 +53,7 @@ #define GFX11_MEC_HPD_SIZE 2048 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L +#define RLC_PG_DELAY_3_DEFAULT_GC_11_0_1 0x1388 #define regCGTT_WD_CLK_CTRL 0x5086 #define regCGTT_WD_CLK_CTRL_BASE_IDX 1 @@ -130,6 +131,8 @@ static void gfx_v11_0_ring_invalidate_tlbs(struct amdgpu_ring *ring, bool all_hub, uint8_t dst_sel); static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev); static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev); +static void gfx_v11_0_update_perf_clk(struct amdgpu_device *adev, + bool enable); static void gfx11_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask) { @@ -1138,6 +1141,7 @@ static const struct amdgpu_gfx_funcs gfx_v11_0_gfx_funcs = { .read_wave_vgprs = &gfx_v11_0_read_wave_vgprs, .select_me_pipe_q = &gfx_v11_0_select_me_pipe_q, .init_spm_golden = &gfx_v11_0_init_spm_golden_registers, + .update_perfmon_mgcg = &gfx_v11_0_update_perf_clk, }; static int gfx_v11_0_gpu_early_init(struct amdgpu_device *adev) @@ -5181,9 +5185,12 @@ static void gfx_v11_0_update_coarse_grain_clock_gating(struct amdgpu_device *ade data = REG_SET_FIELD(data, SDMA0_RLC_CGCG_CTRL, CGCG_INT_ENABLE, 1); WREG32_SOC15(GC, 0, regSDMA0_RLC_CGCG_CTRL, data); - data = RREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL); - data = REG_SET_FIELD(data, SDMA1_RLC_CGCG_CTRL, CGCG_INT_ENABLE, 1); - WREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL, data); + /* Some ASICs only have one SDMA instance, not need to configure SDMA1 */ + if (adev->sdma.num_instances > 1) { + data = RREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL); + data = REG_SET_FIELD(data, SDMA1_RLC_CGCG_CTRL, CGCG_INT_ENABLE, 1); + WREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL, data); + } } else { /* Program RLC_CGCG_CGLS_CTRL */ def = data = RREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL); @@ -5212,9 +5219,12 @@ static void gfx_v11_0_update_coarse_grain_clock_gating(struct amdgpu_device *ade data &= ~SDMA0_RLC_CGCG_CTRL__CGCG_INT_ENABLE_MASK; WREG32_SOC15(GC, 0, regSDMA0_RLC_CGCG_CTRL, data); - data = RREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL); - data &= ~SDMA1_RLC_CGCG_CTRL__CGCG_INT_ENABLE_MASK; - WREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL, data); + /* Some ASICs only have one SDMA instance, not need to configure SDMA1 */ + if (adev->sdma.num_instances > 1) { + data = RREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL); + data &= ~SDMA1_RLC_CGCG_CTRL__CGCG_INT_ENABLE_MASK; + WREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL, data); + } } } @@ -5279,6 +5289,38 @@ static const struct amdgpu_rlc_funcs gfx_v11_0_rlc_funcs = { .update_spm_vmid = gfx_v11_0_update_spm_vmid, }; +static void gfx_v11_cntl_power_gating(struct amdgpu_device *adev, bool enable) +{ + u32 data = RREG32_SOC15(GC, 0, regRLC_PG_CNTL); + + if (enable && (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG)) + data |= RLC_PG_CNTL__GFX_POWER_GATING_ENABLE_MASK; + else + data &= ~RLC_PG_CNTL__GFX_POWER_GATING_ENABLE_MASK; + + WREG32_SOC15(GC, 0, regRLC_PG_CNTL, data); + + // Program RLC_PG_DELAY3 for CGPG hysteresis + if (enable && (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG)) { + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(11, 0, 1): + WREG32_SOC15(GC, 0, regRLC_PG_DELAY_3, RLC_PG_DELAY_3_DEFAULT_GC_11_0_1); + break; + default: + break; + } + } +} + +static void gfx_v11_cntl_pg(struct amdgpu_device *adev, bool enable) +{ + amdgpu_gfx_rlc_enter_safe_mode(adev); + + gfx_v11_cntl_power_gating(adev, enable); + + amdgpu_gfx_rlc_exit_safe_mode(adev); +} + static int gfx_v11_0_set_powergating_state(void *handle, enum amd_powergating_state state) { @@ -5293,6 +5335,10 @@ static int gfx_v11_0_set_powergating_state(void *handle, case IP_VERSION(11, 0, 2): amdgpu_gfx_off_ctrl(adev, enable); break; + case IP_VERSION(11, 0, 1): + gfx_v11_cntl_pg(adev, enable); + amdgpu_gfx_off_ctrl(adev, enable); + break; default: break; } @@ -5310,6 +5356,7 @@ static int gfx_v11_0_set_clockgating_state(void *handle, switch (adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(11, 0, 0): + case IP_VERSION(11, 0, 1): case IP_VERSION(11, 0, 2): gfx_v11_0_update_gfx_clock_gating(adev, state == AMD_CG_STATE_GATE); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index c6e0f9313a7f..fc9c1043244c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -2587,7 +2587,8 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev) gfx_v9_0_tiling_mode_table_init(adev); - gfx_v9_0_setup_rb(adev); + if (adev->gfx.num_gfx_rings) + gfx_v9_0_setup_rb(adev); gfx_v9_0_get_cu_info(adev, &adev->gfx.cu_info); adev->gfx.config.db_debug2 = RREG32_SOC15(GC, 0, mmDB_DEBUG2); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 9ae8cdaa033e..f513e2c2e964 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -419,6 +419,7 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint32_t seq; uint16_t queried_pasid; bool ret; + u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : adev->usec_timeout; struct amdgpu_ring *ring = &adev->gfx.kiq.ring; struct amdgpu_kiq *kiq = &adev->gfx.kiq; @@ -437,7 +438,7 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, amdgpu_ring_commit(ring); spin_unlock(&adev->gfx.kiq.ring_lock); - r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout); + r = amdgpu_fence_wait_polling(ring, seq, usec_timeout); if (r < 1) { dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r); return -ETIME; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 22761a3bb818..4603653916f5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -896,6 +896,7 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint32_t seq; uint16_t queried_pasid; bool ret; + u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : adev->usec_timeout; struct amdgpu_ring *ring = &adev->gfx.kiq.ring; struct amdgpu_kiq *kiq = &adev->gfx.kiq; @@ -935,7 +936,7 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, amdgpu_ring_commit(ring); spin_unlock(&adev->gfx.kiq.ring_lock); - r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout); + r = amdgpu_fence_wait_polling(ring, seq, usec_timeout); if (r < 1) { dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r); up_read(&adev->reset_domain->sem); @@ -1624,12 +1625,15 @@ static int gmc_v9_0_sw_init(void *handle) amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 47); else amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48); + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2)) + adev->gmc.translate_further = adev->vm_manager.num_level > 1; break; case IP_VERSION(9, 4, 1): adev->num_vmhubs = 3; /* Keep the vm size same with Vega20 */ amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48); + adev->gmc.translate_further = adev->vm_manager.num_level > 1; break; default: break; diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c index 39a696cd45b5..29c3484ae1f1 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c @@ -40,6 +40,156 @@ static void hdp_v5_2_flush_hdp(struct amdgpu_device *adev, 0); } +static void hdp_v5_2_update_mem_power_gating(struct amdgpu_device *adev, + bool enable) +{ + uint32_t hdp_clk_cntl; + uint32_t hdp_mem_pwr_cntl; + + if (!(adev->cg_flags & (AMD_CG_SUPPORT_HDP_LS | + AMD_CG_SUPPORT_HDP_DS | + AMD_CG_SUPPORT_HDP_SD))) + return; + + hdp_clk_cntl = RREG32_SOC15(HDP, 0, regHDP_CLK_CNTL); + hdp_mem_pwr_cntl = RREG32_SOC15(HDP, 0, regHDP_MEM_POWER_CTRL); + + /* Before doing clock/power mode switch, forced on MEM clock */ + hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL, + ATOMIC_MEM_CLK_SOFT_OVERRIDE, 1); + hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL, + RC_MEM_CLK_SOFT_OVERRIDE, 1); + WREG32_SOC15(HDP, 0, regHDP_CLK_CNTL, hdp_clk_cntl); + + /* disable clock and power gating before any changing */ + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + ATOMIC_MEM_POWER_CTRL_EN, 0); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + ATOMIC_MEM_POWER_LS_EN, 0); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + ATOMIC_MEM_POWER_DS_EN, 0); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + ATOMIC_MEM_POWER_SD_EN, 0); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + RC_MEM_POWER_CTRL_EN, 0); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + RC_MEM_POWER_LS_EN, 0); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + RC_MEM_POWER_DS_EN, 0); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + RC_MEM_POWER_SD_EN, 0); + WREG32_SOC15(HDP, 0, regHDP_MEM_POWER_CTRL, hdp_mem_pwr_cntl); + + /* Already disabled above. The actions below are for "enabled" only */ + if (enable) { + /* only one clock gating mode (LS/DS/SD) can be enabled */ + if (adev->cg_flags & AMD_CG_SUPPORT_HDP_SD) { + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, + HDP_MEM_POWER_CTRL, + ATOMIC_MEM_POWER_SD_EN, 1); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, + HDP_MEM_POWER_CTRL, + RC_MEM_POWER_SD_EN, 1); + } else if (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS) { + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, + HDP_MEM_POWER_CTRL, + ATOMIC_MEM_POWER_LS_EN, 1); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, + HDP_MEM_POWER_CTRL, + RC_MEM_POWER_LS_EN, 1); + } else if (adev->cg_flags & AMD_CG_SUPPORT_HDP_DS) { + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, + HDP_MEM_POWER_CTRL, + ATOMIC_MEM_POWER_DS_EN, 1); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, + HDP_MEM_POWER_CTRL, + RC_MEM_POWER_DS_EN, 1); + } + + /* confirmed that ATOMIC/RC_MEM_POWER_CTRL_EN have to be set for SRAM LS/DS/SD */ + if (adev->cg_flags & (AMD_CG_SUPPORT_HDP_LS | AMD_CG_SUPPORT_HDP_DS | + AMD_CG_SUPPORT_HDP_SD)) { + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + ATOMIC_MEM_POWER_CTRL_EN, 1); + hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL, + RC_MEM_POWER_CTRL_EN, 1); + WREG32_SOC15(HDP, 0, regHDP_MEM_POWER_CTRL, hdp_mem_pwr_cntl); + } + } + + /* disable MEM clock override after clock/power mode changing */ + hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL, + ATOMIC_MEM_CLK_SOFT_OVERRIDE, 0); + hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL, + RC_MEM_CLK_SOFT_OVERRIDE, 0); + WREG32_SOC15(HDP, 0, regHDP_CLK_CNTL, hdp_clk_cntl); +} + +static void hdp_v5_2_update_medium_grain_clock_gating(struct amdgpu_device *adev, + bool enable) +{ + uint32_t hdp_clk_cntl; + + if (!(adev->cg_flags & AMD_CG_SUPPORT_HDP_MGCG)) + return; + + hdp_clk_cntl = RREG32_SOC15(HDP, 0, regHDP_CLK_CNTL); + + if (enable) { + hdp_clk_cntl &= + ~(uint32_t) + (HDP_CLK_CNTL__ATOMIC_MEM_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__RC_MEM_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__DBUS_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__DYN_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__XDP_REG_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__HDP_REG_CLK_SOFT_OVERRIDE_MASK); + } else { + hdp_clk_cntl |= HDP_CLK_CNTL__ATOMIC_MEM_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__RC_MEM_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__DBUS_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__DYN_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__XDP_REG_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__HDP_REG_CLK_SOFT_OVERRIDE_MASK; + } + + WREG32_SOC15(HDP, 0, regHDP_CLK_CNTL, hdp_clk_cntl); +} + +static void hdp_v5_2_get_clockgating_state(struct amdgpu_device *adev, + u64 *flags) +{ + uint32_t tmp; + + /* AMD_CG_SUPPORT_HDP_MGCG */ + tmp = RREG32_SOC15(HDP, 0, regHDP_CLK_CNTL); + if (!(tmp & (HDP_CLK_CNTL__ATOMIC_MEM_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__RC_MEM_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__DBUS_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__DYN_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__XDP_REG_CLK_SOFT_OVERRIDE_MASK | + HDP_CLK_CNTL__HDP_REG_CLK_SOFT_OVERRIDE_MASK))) + *flags |= AMD_CG_SUPPORT_HDP_MGCG; + + /* AMD_CG_SUPPORT_HDP_LS/DS/SD */ + tmp = RREG32_SOC15(HDP, 0, regHDP_MEM_POWER_CTRL); + if (tmp & HDP_MEM_POWER_CTRL__ATOMIC_MEM_POWER_LS_EN_MASK) + *flags |= AMD_CG_SUPPORT_HDP_LS; + else if (tmp & HDP_MEM_POWER_CTRL__ATOMIC_MEM_POWER_DS_EN_MASK) + *flags |= AMD_CG_SUPPORT_HDP_DS; + else if (tmp & HDP_MEM_POWER_CTRL__ATOMIC_MEM_POWER_SD_EN_MASK) + *flags |= AMD_CG_SUPPORT_HDP_SD; +} + +static void hdp_v5_2_update_clock_gating(struct amdgpu_device *adev, + bool enable) +{ + hdp_v5_2_update_mem_power_gating(adev, enable); + hdp_v5_2_update_medium_grain_clock_gating(adev, enable); +} + const struct amdgpu_hdp_funcs hdp_v5_2_funcs = { .flush_hdp = hdp_v5_2_flush_hdp, + .update_clock_gating = hdp_v5_2_update_clock_gating, + .get_clock_gating_state = hdp_v5_2_get_clockgating_state, }; diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c index 92dc60a9d209..085e613f3646 100644 --- a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c @@ -727,6 +727,7 @@ static const struct amd_ip_funcs ih_v6_0_ip_funcs = { static const struct amdgpu_ih_funcs ih_v6_0_funcs = { .get_wptr = ih_v6_0_get_wptr, .decode_iv = amdgpu_ih_decode_iv_helper, + .decode_iv_ts = amdgpu_ih_decode_iv_ts_helper, .set_rptr = ih_v6_0_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 120ea294abef..cc3fdbbcd314 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -183,6 +183,7 @@ static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes, mes_add_queue_pkt.trap_handler_addr = input->tba_addr; mes_add_queue_pkt.tma_addr = input->tma_addr; mes_add_queue_pkt.is_kfd_process = input->is_kfd_process; + mes_add_queue_pkt.trap_en = 1; return mes_v11_0_submit_pkt_and_poll_completion(mes, &mes_add_queue_pkt, sizeof(mes_add_queue_pkt), diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 3f44a099c52a..3e51e773f92b 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -176,6 +176,7 @@ static void mmhub_v1_0_init_cache_regs(struct amdgpu_device *adev) tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1); WREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL2, tmp); + tmp = mmVM_L2_CNTL3_DEFAULT; if (adev->gmc.translate_further) { tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12); tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c index cac72ced94c8..e8058edc1d10 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c @@ -518,18 +518,41 @@ static u64 mmhub_v3_0_1_get_mc_fb_offset(struct amdgpu_device *adev) static void mmhub_v3_0_1_update_medium_grain_clock_gating(struct amdgpu_device *adev, bool enable) { - //TODO + uint32_t def, data; + + def = data = RREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG); + + if (enable) + data |= MM_ATC_L2_MISC_CG__ENABLE_MASK; + else + data &= ~MM_ATC_L2_MISC_CG__ENABLE_MASK; + + if (def != data) + WREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG, data); } static void mmhub_v3_0_1_update_medium_grain_light_sleep(struct amdgpu_device *adev, bool enable) { - //TODO + uint32_t def, data; + + def = data = RREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG); + + if (enable) + data |= MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK; + else + data &= ~MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK; + + if (def != data) + WREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG, data); } static int mmhub_v3_0_1_set_clockgating(struct amdgpu_device *adev, enum amd_clockgating_state state) { + if (amdgpu_sriov_vf(adev)) + return 0; + mmhub_v3_0_1_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE); mmhub_v3_0_1_update_medium_grain_light_sleep(adev, @@ -539,7 +562,20 @@ static int mmhub_v3_0_1_set_clockgating(struct amdgpu_device *adev, static void mmhub_v3_0_1_get_clockgating(struct amdgpu_device *adev, u64 *flags) { - //TODO + int data; + + if (amdgpu_sriov_vf(adev)) + *flags = 0; + + data = RREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG); + + /* AMD_CG_SUPPORT_MC_MGCG */ + if (data & MM_ATC_L2_MISC_CG__ENABLE_MASK) + *flags |= AMD_CG_SUPPORT_MC_MGCG; + + /* AMD_CG_SUPPORT_MC_LS */ + if (data & MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK) + *flags |= AMD_CG_SUPPORT_MC_LS; } const struct amdgpu_mmhub_funcs mmhub_v3_0_1_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c index 6e0145b2b408..445cb06b9d26 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c @@ -295,9 +295,17 @@ static void mmhub_v9_4_disable_identity_aperture(struct amdgpu_device *adev, static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid) { struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + unsigned int num_level, block_size; uint32_t tmp; int i; + num_level = adev->vm_manager.num_level; + block_size = adev->vm_manager.block_size; + if (adev->gmc.translate_further) + num_level -= 1; + else + block_size -= 9; + for (i = 0; i <= 14; i++) { tmp = RREG32_SOC15_OFFSET(MMHUB, 0, mmVML2VC0_VM_CONTEXT1_CNTL, hubid * MMHUB_INSTANCE_REGISTER_OFFSET + i); @@ -305,7 +313,7 @@ static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid) ENABLE_CONTEXT, 1); tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH, - adev->vm_manager.num_level); + num_level); tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL, RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL, @@ -323,7 +331,7 @@ static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid) EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL, PAGE_TABLE_BLOCK_SIZE, - adev->vm_manager.block_size - 9); + block_size); /* Send no-retry XNACK on fault to suppress VM fault storm. */ tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL, RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c index 4b5396d3e60f..eec13cb5bf75 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c @@ -409,9 +409,11 @@ static u32 navi10_ih_get_wptr(struct amdgpu_device *adev, u32 wptr, tmp; struct amdgpu_ih_regs *ih_regs; - if (ih == &adev->irq.ih) { + if (ih == &adev->irq.ih || ih == &adev->irq.ih_soft) { /* Only ring0 supports writeback. On other rings fall back * to register-based code with overflow checking below. + * ih_soft ring doesn't have any backing hardware registers, + * update wptr and return. */ wptr = le32_to_cpu(*ih->wptr_cpu); @@ -483,6 +485,9 @@ static void navi10_ih_set_rptr(struct amdgpu_device *adev, { struct amdgpu_ih_regs *ih_regs; + if (ih == &adev->irq.ih_soft) + return; + if (ih->use_doorbell) { /* XXX check if swapping is necessary on BE */ *ih->rptr_cpu = ih->rptr; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_7.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_7.c index 01e8288d09a8..1dc95ef21da6 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_7.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_7.c @@ -247,6 +247,81 @@ static void nbio_v7_7_init_registers(struct amdgpu_device *adev) } +static void nbio_v7_7_update_medium_grain_clock_gating(struct amdgpu_device *adev, + bool enable) +{ + uint32_t def, data; + + if (enable && !(adev->cg_flags & AMD_CG_SUPPORT_BIF_MGCG)) + return; + + def = data = RREG32_SOC15(NBIO, 0, regBIF0_CPM_CONTROL); + if (enable) { + data |= (BIF0_CPM_CONTROL__LCLK_DYN_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__TXCLK_DYN_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__TXCLK_LCNT_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__TXCLK_REGS_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__TXCLK_PRBS_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__REFCLK_REGS_GATE_ENABLE_MASK); + } else { + data &= ~(BIF0_CPM_CONTROL__LCLK_DYN_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__TXCLK_DYN_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__TXCLK_LCNT_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__TXCLK_REGS_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__TXCLK_PRBS_GATE_ENABLE_MASK | + BIF0_CPM_CONTROL__REFCLK_REGS_GATE_ENABLE_MASK); + } + + if (def != data) + WREG32_SOC15(NBIO, 0, regBIF0_CPM_CONTROL, data); +} + +static void nbio_v7_7_update_medium_grain_light_sleep(struct amdgpu_device *adev, + bool enable) +{ + uint32_t def, data; + + if (enable && !(adev->cg_flags & AMD_CG_SUPPORT_BIF_LS)) + return; + + def = data = RREG32_SOC15(NBIO, 0, regBIF0_PCIE_CNTL2); + if (enable) + data |= BIF0_PCIE_CNTL2__SLV_MEM_LS_EN_MASK; + else + data &= ~BIF0_PCIE_CNTL2__SLV_MEM_LS_EN_MASK; + + if (def != data) + WREG32_SOC15(NBIO, 0, regBIF0_PCIE_CNTL2, data); + + def = data = RREG32_SOC15(NBIO, 0, regBIF0_PCIE_TX_POWER_CTRL_1); + if (enable) { + data |= (BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_LS_EN_MASK | + BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_LS_EN_MASK); + } else { + data &= ~(BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_LS_EN_MASK | + BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_LS_EN_MASK); + } + + if (def != data) + WREG32_SOC15(NBIO, 0, regBIF0_PCIE_TX_POWER_CTRL_1, data); +} + +static void nbio_v7_7_get_clockgating_state(struct amdgpu_device *adev, + u64 *flags) +{ + uint32_t data; + + /* AMD_CG_SUPPORT_BIF_MGCG */ + data = RREG32_SOC15(NBIO, 0, regBIF0_CPM_CONTROL); + if (data & BIF0_CPM_CONTROL__LCLK_DYN_GATE_ENABLE_MASK) + *flags |= AMD_CG_SUPPORT_BIF_MGCG; + + /* AMD_CG_SUPPORT_BIF_LS */ + data = RREG32_SOC15(NBIO, 0, regBIF0_PCIE_CNTL2); + if (data & BIF0_PCIE_CNTL2__SLV_MEM_LS_EN_MASK) + *flags |= AMD_CG_SUPPORT_BIF_LS; +} + const struct amdgpu_nbio_funcs nbio_v7_7_funcs = { .get_hdp_flush_req_offset = nbio_v7_7_get_hdp_flush_req_offset, .get_hdp_flush_done_offset = nbio_v7_7_get_hdp_flush_done_offset, @@ -262,6 +337,9 @@ const struct amdgpu_nbio_funcs nbio_v7_7_funcs = { .enable_doorbell_aperture = nbio_v7_7_enable_doorbell_aperture, .enable_doorbell_selfring_aperture = nbio_v7_7_enable_doorbell_selfring_aperture, .ih_doorbell_range = nbio_v7_7_ih_doorbell_range, + .update_medium_grain_clock_gating = nbio_v7_7_update_medium_grain_clock_gating, + .update_medium_grain_light_sleep = nbio_v7_7_update_medium_grain_light_sleep, + .get_clockgating_state = nbio_v7_7_get_clockgating_state, .ih_control = nbio_v7_7_ih_control, .init_registers = nbio_v7_7_init_registers, }; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c index a2588200ea58..0b2ac418e4ac 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c @@ -101,6 +101,16 @@ static int psp_v12_0_init_microcode(struct psp_context *psp) adev->psp.dtm_context.context.bin_desc.start_addr = (uint8_t *)adev->psp.hdcp_context.context.bin_desc.start_addr + le32_to_cpu(ta_hdr->dtm.offset_bytes); + + if (adev->apu_flags & AMD_APU_IS_RENOIR) { + adev->psp.securedisplay_context.context.bin_desc.fw_version = + le32_to_cpu(ta_hdr->securedisplay.fw_version); + adev->psp.securedisplay_context.context.bin_desc.size_bytes = + le32_to_cpu(ta_hdr->securedisplay.size_bytes); + adev->psp.securedisplay_context.context.bin_desc.start_addr = + (uint8_t *)adev->psp.hdcp_context.context.bin_desc.start_addr + + le32_to_cpu(ta_hdr->securedisplay.offset_bytes); + } } return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index 726a5bba40b2..a75a286e1ecf 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -20,7 +20,6 @@ * OTHER DEALINGS IN THE SOFTWARE. * */ -#include <linux/dev_printk.h> #include <drm/drm_drv.h> #include <linux/vmalloc.h> #include "amdgpu.h" diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c index 52816de5e17b..55284b24f113 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc21.c +++ b/drivers/gpu/drm/amd/amdgpu/soc21.c @@ -494,6 +494,20 @@ static void soc21_pre_asic_init(struct amdgpu_device *adev) { } +static int soc21_update_umd_stable_pstate(struct amdgpu_device *adev, + bool enter) +{ + if (enter) + amdgpu_gfx_rlc_enter_safe_mode(adev); + else + amdgpu_gfx_rlc_exit_safe_mode(adev); + + if (adev->gfx.funcs->update_perfmon_mgcg) + adev->gfx.funcs->update_perfmon_mgcg(adev, !enter); + + return 0; +} + static const struct amdgpu_asic_funcs soc21_asic_funcs = { .read_disabled_bios = &soc21_read_disabled_bios, @@ -513,6 +527,7 @@ static const struct amdgpu_asic_funcs soc21_asic_funcs = .supports_baco = &amdgpu_dpm_is_baco_supported, .pre_asic_init = &soc21_pre_asic_init, .query_video_codecs = &soc21_query_video_codecs, + .update_umd_stable_pstate = &soc21_update_umd_stable_pstate, }; static int soc21_common_early_init(void *handle) @@ -546,8 +561,10 @@ static int soc21_common_early_init(void *handle) case IP_VERSION(11, 0, 0): adev->cg_flags = AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_CGLS | +#if 0 AMD_CG_SUPPORT_GFX_3D_CGCG | AMD_CG_SUPPORT_GFX_3D_CGLS | +#endif AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_REPEATER_FGCG | AMD_CG_SUPPORT_GFX_FGCG | @@ -575,7 +592,9 @@ static int soc21_common_early_init(void *handle) AMD_CG_SUPPORT_VCN_MGCG | AMD_CG_SUPPORT_JPEG_MGCG | AMD_CG_SUPPORT_ATHUB_MGCG | - AMD_CG_SUPPORT_ATHUB_LS; + AMD_CG_SUPPORT_ATHUB_LS | + AMD_CG_SUPPORT_IH_CG | + AMD_CG_SUPPORT_HDP_SD; adev->pg_flags = AMD_PG_SUPPORT_VCN | AMD_PG_SUPPORT_VCN_DPG | @@ -586,9 +605,25 @@ static int soc21_common_early_init(void *handle) break; case IP_VERSION(11, 0, 1): adev->cg_flags = + AMD_CG_SUPPORT_GFX_CGCG | + AMD_CG_SUPPORT_GFX_CGLS | + AMD_CG_SUPPORT_GFX_MGCG | + AMD_CG_SUPPORT_GFX_FGCG | + AMD_CG_SUPPORT_REPEATER_FGCG | + AMD_CG_SUPPORT_GFX_PERF_CLK | + AMD_CG_SUPPORT_MC_MGCG | + AMD_CG_SUPPORT_MC_LS | + AMD_CG_SUPPORT_HDP_MGCG | + AMD_CG_SUPPORT_HDP_LS | + AMD_CG_SUPPORT_ATHUB_MGCG | + AMD_CG_SUPPORT_ATHUB_LS | + AMD_CG_SUPPORT_IH_CG | + AMD_CG_SUPPORT_BIF_MGCG | + AMD_CG_SUPPORT_BIF_LS | AMD_CG_SUPPORT_VCN_MGCG | AMD_CG_SUPPORT_JPEG_MGCG; adev->pg_flags = + AMD_PG_SUPPORT_GFX_PG | AMD_PG_SUPPORT_JPEG; adev->external_rev_id = adev->rev_id + 0x1; break; @@ -683,6 +718,8 @@ static int soc21_common_set_clockgating_state(void *handle, switch (adev->ip_versions[NBIO_HWIP][0]) { case IP_VERSION(4, 3, 0): + case IP_VERSION(4, 3, 1): + case IP_VERSION(7, 7, 0): adev->nbio.funcs->update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE); adev->nbio.funcs->update_medium_grain_light_sleep(adev, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index ca14c3ef742e..fb2d74f30448 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -1115,7 +1115,7 @@ static int vcn_v4_0_start(struct amdgpu_device *adev) * * Stop VCN block with dpg mode */ -static int vcn_v4_0_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx) +static void vcn_v4_0_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx) { uint32_t tmp; @@ -1133,7 +1133,6 @@ static int vcn_v4_0_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx) /* disable dynamic power gating mode */ WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 0, ~UVD_POWER_STATUS__UVD_PG_MODE_MASK); - return 0; } /** @@ -1154,7 +1153,7 @@ static int vcn_v4_0_stop(struct amdgpu_device *adev) fw_shared->sq.queue_mode |= FW_QUEUE_DPG_HOLD_OFF; if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { - r = vcn_v4_0_stop_dpg_mode(adev, i); + vcn_v4_0_stop_dpg_mode(adev, i); continue; } diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c index cdd599a08125..03b7066471f9 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c @@ -334,9 +334,11 @@ static u32 vega10_ih_get_wptr(struct amdgpu_device *adev, u32 wptr, tmp; struct amdgpu_ih_regs *ih_regs; - if (ih == &adev->irq.ih) { + if (ih == &adev->irq.ih || ih == &adev->irq.ih_soft) { /* Only ring0 supports writeback. On other rings fall back * to register-based code with overflow checking below. + * ih_soft ring doesn't have any backing hardware registers, + * update wptr and return. */ wptr = le32_to_cpu(*ih->wptr_cpu); @@ -409,6 +411,9 @@ static void vega10_ih_set_rptr(struct amdgpu_device *adev, { struct amdgpu_ih_regs *ih_regs; + if (ih == &adev->irq.ih_soft) + return; + if (ih->use_doorbell) { /* XXX check if swapping is necessary on BE */ *ih->rptr_cpu = ih->rptr; diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c index 3b4eb8285943..2022ffbb8dba 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c @@ -385,9 +385,11 @@ static u32 vega20_ih_get_wptr(struct amdgpu_device *adev, u32 wptr, tmp; struct amdgpu_ih_regs *ih_regs; - if (ih == &adev->irq.ih) { + if (ih == &adev->irq.ih || ih == &adev->irq.ih_soft) { /* Only ring0 supports writeback. On other rings fall back * to register-based code with overflow checking below. + * ih_soft ring doesn't have any backing hardware registers, + * update wptr and return. */ wptr = le32_to_cpu(*ih->wptr_cpu); @@ -461,6 +463,9 @@ static void vega20_ih_set_rptr(struct amdgpu_device *adev, { struct amdgpu_ih_regs *ih_regs; + if (ih == &adev->irq.ih_soft) + return; + if (ih->use_doorbell) { /* XXX check if swapping is necessary on BE */ *ih->rptr_cpu = ih->rptr; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 2b3d8bc8f0aa..dc774ddf3445 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -874,7 +874,7 @@ static int kfd_ioctl_wait_events(struct file *filp, struct kfd_process *p, err = kfd_wait_on_events(p, args->num_events, (void __user *)args->events_ptr, (args->wait_for_all != 0), - args->timeout, &args->wait_result); + &args->timeout, &args->wait_result); return err; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index f5853835f03a..22c0929d410b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -102,13 +102,18 @@ static void kfd_device_info_set_sdma_info(struct kfd_dev *kfd) switch (sdma_version) { case IP_VERSION(6, 0, 0): - case IP_VERSION(6, 0, 1): case IP_VERSION(6, 0, 2): /* Reserve 1 for paging and 1 for gfx */ kfd->device_info.num_reserved_sdma_queues_per_engine = 2; /* BIT(0)=engine-0 queue-0; BIT(1)=engine-1 queue-0; BIT(2)=engine-0 queue-1; ... */ kfd->device_info.reserved_sdma_queues_bitmap = 0xFULL; break; + case IP_VERSION(6, 0, 1): + /* Reserve 1 for paging and 1 for gfx */ + kfd->device_info.num_reserved_sdma_queues_per_engine = 2; + /* BIT(0)=engine-0 queue-0; BIT(1)=engine-0 queue-1; ... */ + kfd->device_info.reserved_sdma_queues_bitmap = 0x3ULL; + break; default: break; } @@ -377,12 +382,8 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) f2g = &gfx_v10_3_kfd2kgd; break; case IP_VERSION(10, 3, 6): - gfx_target_version = 100306; - if (!vf) - f2g = &gfx_v10_3_kfd2kgd; - break; case IP_VERSION(10, 3, 7): - gfx_target_version = 100307; + gfx_target_version = 100306; if (!vf) f2g = &gfx_v10_3_kfd2kgd; break; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index 3942a56c28bb..83e3ce9f6049 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -894,7 +894,8 @@ static long user_timeout_to_jiffies(uint32_t user_timeout_ms) return msecs_to_jiffies(user_timeout_ms) + 1; } -static void free_waiters(uint32_t num_events, struct kfd_event_waiter *waiters) +static void free_waiters(uint32_t num_events, struct kfd_event_waiter *waiters, + bool undo_auto_reset) { uint32_t i; @@ -903,6 +904,9 @@ static void free_waiters(uint32_t num_events, struct kfd_event_waiter *waiters) spin_lock(&waiters[i].event->lock); remove_wait_queue(&waiters[i].event->wq, &waiters[i].wait); + if (undo_auto_reset && waiters[i].activated && + waiters[i].event && waiters[i].event->auto_reset) + set_event(waiters[i].event); spin_unlock(&waiters[i].event->lock); } @@ -911,7 +915,7 @@ static void free_waiters(uint32_t num_events, struct kfd_event_waiter *waiters) int kfd_wait_on_events(struct kfd_process *p, uint32_t num_events, void __user *data, - bool all, uint32_t user_timeout_ms, + bool all, uint32_t *user_timeout_ms, uint32_t *wait_result) { struct kfd_event_data __user *events = @@ -920,7 +924,7 @@ int kfd_wait_on_events(struct kfd_process *p, int ret = 0; struct kfd_event_waiter *event_waiters = NULL; - long timeout = user_timeout_to_jiffies(user_timeout_ms); + long timeout = user_timeout_to_jiffies(*user_timeout_ms); event_waiters = alloc_event_waiters(num_events); if (!event_waiters) { @@ -970,15 +974,11 @@ int kfd_wait_on_events(struct kfd_process *p, } if (signal_pending(current)) { - /* - * This is wrong when a nonzero, non-infinite timeout - * is specified. We need to use - * ERESTARTSYS_RESTARTBLOCK, but struct restart_block - * contains a union with data for each user and it's - * in generic kernel code that I don't want to - * touch yet. - */ ret = -ERESTARTSYS; + if (*user_timeout_ms != KFD_EVENT_TIMEOUT_IMMEDIATE && + *user_timeout_ms != KFD_EVENT_TIMEOUT_INFINITE) + *user_timeout_ms = jiffies_to_msecs( + max(0l, timeout-1)); break; } @@ -1019,7 +1019,7 @@ int kfd_wait_on_events(struct kfd_process *p, event_waiters, events); out_unlock: - free_waiters(num_events, event_waiters); + free_waiters(num_events, event_waiters, ret == -ERESTARTSYS); mutex_unlock(&p->event_mutex); out: if (ret) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index d03a3b9c9c5d..bf610e3b683b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -1317,7 +1317,7 @@ void kfd_event_free_process(struct kfd_process *p); int kfd_event_mmap(struct kfd_process *process, struct vm_area_struct *vma); int kfd_wait_on_events(struct kfd_process *p, uint32_t num_events, void __user *data, - bool all, uint32_t user_timeout_ms, + bool all, uint32_t *user_timeout_ms, uint32_t *wait_result); void kfd_signal_event_interrupt(u32 pasid, uint32_t partial_id, uint32_t valid_id_bits); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index a67ba8879a56..11074cc8c333 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -541,7 +541,6 @@ svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange, kfree(svm_bo); return -ESRCH; } - svm_bo->svms = prange->svms; svm_bo->eviction_fence = amdgpu_amdkfd_fence_create(dma_fence_context_alloc(1), mm, @@ -3273,7 +3272,6 @@ int svm_range_schedule_evict_svm_bo(struct amdgpu_amdkfd_fence *fence) static void svm_range_evict_svm_bo_worker(struct work_struct *work) { struct svm_range_bo *svm_bo; - struct kfd_process *p; struct mm_struct *mm; int r = 0; @@ -3281,13 +3279,12 @@ static void svm_range_evict_svm_bo_worker(struct work_struct *work) if (!svm_bo_ref_unless_zero(svm_bo)) return; /* svm_bo was freed while eviction was pending */ - /* svm_range_bo_release destroys this worker thread. So during - * the lifetime of this thread, kfd_process and mm will be valid. - */ - p = container_of(svm_bo->svms, struct kfd_process, svms); - mm = p->mm; - if (!mm) + if (mmget_not_zero(svm_bo->eviction_fence->mm)) { + mm = svm_bo->eviction_fence->mm; + } else { + svm_range_bo_unref(svm_bo); return; + } mmap_read_lock(mm); spin_lock(&svm_bo->list_lock); @@ -3305,8 +3302,7 @@ static void svm_range_evict_svm_bo_worker(struct work_struct *work) mutex_lock(&prange->migrate_mutex); do { - r = svm_migrate_vram_to_ram(prange, - svm_bo->eviction_fence->mm, + r = svm_migrate_vram_to_ram(prange, mm, KFD_MIGRATE_TRIGGER_TTM_EVICTION); } while (!r && prange->actual_loc && --retries); @@ -3324,6 +3320,7 @@ static void svm_range_evict_svm_bo_worker(struct work_struct *work) } spin_unlock(&svm_bo->list_lock); mmap_read_unlock(mm); + mmput(mm); dma_fence_signal(&svm_bo->eviction_fence->base); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h index 9156b041ef17..cfac13ad06ef 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h @@ -46,7 +46,6 @@ struct svm_range_bo { spinlock_t list_lock; struct amdgpu_amdkfd_fence *eviction_fence; struct work_struct eviction_work; - struct svm_range_list *svms; uint32_t evicting; struct work_struct release_work; }; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 25990bec600d..3f0a4a415907 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -1392,8 +1392,8 @@ static int kfd_build_p2p_node_entry(struct kfd_topology_device *dev, static int kfd_create_indirect_link_prop(struct kfd_topology_device *kdev, int gpu_node) { + struct kfd_iolink_properties *gpu_link, *tmp_link, *cpu_link; struct kfd_iolink_properties *props = NULL, *props2 = NULL; - struct kfd_iolink_properties *gpu_link, *cpu_link; struct kfd_topology_device *cpu_dev; int ret = 0; int i, num_cpu; @@ -1416,16 +1416,19 @@ static int kfd_create_indirect_link_prop(struct kfd_topology_device *kdev, int g continue; /* find CPU <--> CPU links */ + cpu_link = NULL; cpu_dev = kfd_topology_device_by_proximity_domain(i); if (cpu_dev) { - list_for_each_entry(cpu_link, + list_for_each_entry(tmp_link, &cpu_dev->io_link_props, list) { - if (cpu_link->node_to == gpu_link->node_to) + if (tmp_link->node_to == gpu_link->node_to) { + cpu_link = tmp_link; break; + } } } - if (cpu_link->node_to != gpu_link->node_to) + if (!cpu_link) return -ENOMEM; /* CPU <--> CPU <--> GPU, GPU node*/ 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 8660d93cc405..5140d9c2bf3b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -3825,8 +3825,11 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev) adev_to_drm(adev)->mode_config.max_height = 16384; adev_to_drm(adev)->mode_config.preferred_depth = 24; - /* disable prefer shadow for now due to hibernation issues */ - adev_to_drm(adev)->mode_config.prefer_shadow = 0; + if (adev->asic_type == CHIP_HAWAII) + /* disable prefer shadow for now due to hibernation issues */ + adev_to_drm(adev)->mode_config.prefer_shadow = 0; + else + adev_to_drm(adev)->mode_config.prefer_shadow = 1; /* indicates support for immediate flip */ adev_to_drm(adev)->mode_config.async_page_flip = true; @@ -4135,6 +4138,7 @@ static void register_backlight_device(struct amdgpu_display_manager *dm, } } +static void amdgpu_set_panel_orientation(struct drm_connector *connector); /* * In this architecture, the association @@ -4326,6 +4330,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) adev_to_drm(adev)->vblank_disable_immediate = false; } } + amdgpu_set_panel_orientation(&aconnector->base); } /* Software is initialized. Now we can register interrupt handlers. */ @@ -6684,6 +6689,10 @@ static void amdgpu_set_panel_orientation(struct drm_connector *connector) connector->connector_type != DRM_MODE_CONNECTOR_LVDS) return; + mutex_lock(&connector->dev->mode_config.mutex); + amdgpu_dm_connector_get_modes(connector); + mutex_unlock(&connector->dev->mode_config.mutex); + encoder = amdgpu_dm_connector_to_encoder(connector); if (!encoder) return; @@ -6728,8 +6737,6 @@ static void amdgpu_dm_connector_ddc_get_modes(struct drm_connector *connector, * restored here. */ amdgpu_dm_update_freesync_caps(connector, edid); - - amdgpu_set_panel_orientation(connector); } else { amdgpu_dm_connector->num_modes = 0; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c index b841b8b0a9d8..987bde4dca3d 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c @@ -34,6 +34,7 @@ #include "dal_asic_id.h" #include "amdgpu_display.h" #include "amdgpu_dm_trace.h" +#include "amdgpu_dm_plane.h" #include "gc/gc_11_0_0_offset.h" #include "gc/gc_11_0_0_sh_mask.h" @@ -149,12 +150,12 @@ static void add_modifier(uint64_t **mods, uint64_t *size, uint64_t *cap, uint64_ *size += 1; } -bool modifier_has_dcc(uint64_t modifier) +static bool modifier_has_dcc(uint64_t modifier) { return IS_AMD_FMT_MOD(modifier) && AMD_FMT_MOD_GET(DCC, modifier); } -unsigned modifier_gfx9_swizzle_mode(uint64_t modifier) +static unsigned modifier_gfx9_swizzle_mode(uint64_t modifier) { if (modifier == DRM_FORMAT_MOD_LINEAR) return 0; @@ -660,7 +661,7 @@ static int get_plane_modifiers(struct amdgpu_device *adev, unsigned int plane_ty add_gfx10_1_modifiers(adev, mods, &size, &capacity); break; case AMDGPU_FAMILY_GC_11_0_0: - case AMDGPU_FAMILY_GC_11_0_2: + case AMDGPU_FAMILY_GC_11_0_1: add_gfx11_modifiers(adev, mods, &size, &capacity); break; } @@ -1412,7 +1413,7 @@ static bool dm_plane_format_mod_supported(struct drm_plane *plane, } break; case AMDGPU_FAMILY_GC_11_0_0: - case AMDGPU_FAMILY_GC_11_0_2: + case AMDGPU_FAMILY_GC_11_0_1: switch (AMD_FMT_MOD_GET(TILE, modifier)) { case AMD_FMT_MOD_TILE_GFX11_256K_R_X: case AMD_FMT_MOD_TILE_GFX9_64K_R_X: diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.h index 95168c2cfa6f..286981a2dd40 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.h @@ -36,17 +36,9 @@ int fill_dc_scaling_info(struct amdgpu_device *adev, const struct drm_plane_state *state, struct dc_scaling_info *scaling_info); -void get_min_max_dc_plane_scaling(struct drm_device *dev, - struct drm_framebuffer *fb, - int *min_downscale, int *max_upscale); - int dm_plane_helper_check_state(struct drm_plane_state *state, struct drm_crtc_state *new_crtc_state); -bool modifier_has_dcc(uint64_t modifier); - -unsigned int modifier_gfx9_swizzle_mode(uint64_t modifier); - int fill_plane_buffer_attributes(struct amdgpu_device *adev, const struct amdgpu_framebuffer *afb, const enum surface_pixel_format format, diff --git a/drivers/gpu/drm/amd/display/dc/basics/conversion.c b/drivers/gpu/drm/amd/display/dc/basics/conversion.c index 6767fab55c26..352e9afb85c6 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/conversion.c +++ b/drivers/gpu/drm/amd/display/dc/basics/conversion.c @@ -100,3 +100,24 @@ void convert_float_matrix( matrix[i] = (uint16_t)reg_value; } } + +static uint32_t find_gcd(uint32_t a, uint32_t b) +{ + uint32_t remainder = 0; + while (b != 0) { + remainder = a % b; + a = b; + b = remainder; + } + return a; +} + +void reduce_fraction(uint32_t num, uint32_t den, + uint32_t *out_num, uint32_t *out_den) +{ + uint32_t gcd = 0; + + gcd = find_gcd(num, den); + *out_num = num / gcd; + *out_den = den / gcd; +} diff --git a/drivers/gpu/drm/amd/display/dc/basics/conversion.h b/drivers/gpu/drm/amd/display/dc/basics/conversion.h index ade785c4fdc7..81da4e6f7a1a 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/conversion.h +++ b/drivers/gpu/drm/amd/display/dc/basics/conversion.h @@ -38,6 +38,9 @@ void convert_float_matrix( struct fixed31_32 *flt, uint32_t buffer_size); +void reduce_fraction(uint32_t num, uint32_t den, + uint32_t *out_num, uint32_t *out_den); + static inline unsigned int log_2(unsigned int num) { return ilog2(num); 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 4c76091fd1f2..f276abb63bcd 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 @@ -337,7 +337,7 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p break; } - case AMDGPU_FAMILY_GC_11_0_2: { + case AMDGPU_FAMILY_GC_11_0_1: { struct clk_mgr_dcn314 *clk_mgr = kzalloc(sizeof(*clk_mgr), GFP_KERNEL); if (clk_mgr == NULL) { @@ -397,7 +397,7 @@ void dc_destroy_clk_mgr(struct clk_mgr *clk_mgr_base) dcn32_clk_mgr_destroy(clk_mgr); break; - case AMDGPU_FAMILY_GC_11_0_2: + case AMDGPU_FAMILY_GC_11_0_1: dcn314_clk_mgr_destroy(clk_mgr); break; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c index 0202dc682682..ca6dfd2d7561 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -24,10 +24,9 @@ */ #include "dccg.h" -#include "clk_mgr_internal.h" +#include "rn_clk_mgr.h" #include "dcn20/dcn20_clk_mgr.h" -#include "rn_clk_mgr.h" #include "dml/dcn20/dcn20_fpu.h" #include "dce100/dce_clk_mgr.h" diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h index 2e088c5171b2..f1319957e400 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.h @@ -28,6 +28,7 @@ #include "clk_mgr.h" #include "dm_pp_smu.h" +#include "clk_mgr_internal.h" extern struct wm_table ddr4_wm_table_gs; extern struct wm_table lpddr4_wm_table_gs; 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 ee99974b3b62..beb025cd3dc2 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 @@ -307,16 +307,6 @@ static void dcn314_enable_pme_wa(struct clk_mgr *clk_mgr_base) dcn314_smu_enable_pme_wa(clk_mgr); } -void dcn314_init_clocks(struct clk_mgr *clk_mgr) -{ - memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); - // Assumption is that boot state always supports pstate - clk_mgr->clks.p_state_change_support = true; - clk_mgr->clks.prev_p_state_change_support = true; - clk_mgr->clks.pwr_state = DCN_PWR_STATE_UNKNOWN; - clk_mgr->clks.zstate_support = DCN_ZSTATE_SUPPORT_UNKNOWN; -} - bool dcn314_are_clock_states_equal(struct dc_clocks *a, struct dc_clocks *b) { @@ -425,7 +415,7 @@ static struct wm_table lpddr5_wm_table = { } }; -static DpmClocks_t dummy_clocks; +static DpmClocks314_t dummy_clocks; static struct dcn314_watermarks dummy_wms = { 0 }; @@ -510,7 +500,7 @@ static void dcn314_notify_wm_ranges(struct clk_mgr *clk_mgr_base) static void dcn314_get_dpm_table_from_smu(struct clk_mgr_internal *clk_mgr, struct dcn314_smu_dpm_clks *smu_dpm_clks) { - DpmClocks_t *table = smu_dpm_clks->dpm_clks; + DpmClocks314_t *table = smu_dpm_clks->dpm_clks; if (!clk_mgr->smu_ver) return; @@ -527,6 +517,26 @@ static void dcn314_get_dpm_table_from_smu(struct clk_mgr_internal *clk_mgr, dcn314_smu_transfer_dpm_table_smu_2_dram(clk_mgr); } +static inline bool is_valid_clock_value(uint32_t clock_value) +{ + return clock_value > 1 && clock_value < 100000; +} + +static unsigned int convert_wck_ratio(uint8_t wck_ratio) +{ + switch (wck_ratio) { + case WCK_RATIO_1_2: + return 2; + + case WCK_RATIO_1_4: + return 4; + + default: + break; + } + return 1; +} + static uint32_t find_max_clk_value(const uint32_t clocks[], uint32_t num_clocks) { uint32_t max = 0; @@ -540,89 +550,127 @@ static uint32_t find_max_clk_value(const uint32_t clocks[], uint32_t num_clocks) return max; } -static unsigned int find_clk_for_voltage( - const DpmClocks_t *clock_table, - const uint32_t clocks[], - unsigned int voltage) -{ - int i; - int max_voltage = 0; - int clock = 0; - - for (i = 0; i < NUM_SOC_VOLTAGE_LEVELS; i++) { - if (clock_table->SocVoltage[i] == voltage) { - return clocks[i]; - } else if (clock_table->SocVoltage[i] >= max_voltage && - clock_table->SocVoltage[i] < voltage) { - max_voltage = clock_table->SocVoltage[i]; - clock = clocks[i]; - } - } - - ASSERT(clock); - return clock; -} - static void dcn314_clk_mgr_helper_populate_bw_params(struct clk_mgr_internal *clk_mgr, struct integrated_info *bios_info, - const DpmClocks_t *clock_table) + const DpmClocks314_t *clock_table) { - int i, j; struct clk_bw_params *bw_params = clk_mgr->base.bw_params; - uint32_t max_dispclk = 0, max_dppclk = 0; - - j = -1; - - ASSERT(NUM_DF_PSTATE_LEVELS <= MAX_NUM_DPM_LVL); - - /* Find lowest DPM, FCLK is filled in reverse order*/ + struct clk_limit_table_entry def_max = bw_params->clk_table.entries[bw_params->clk_table.num_entries - 1]; + uint32_t max_pstate = 0, max_fclk = 0, min_pstate = 0, max_dispclk = 0, max_dppclk = 0; + int i; - for (i = NUM_DF_PSTATE_LEVELS - 1; i >= 0; i--) { - if (clock_table->DfPstateTable[i].FClk != 0) { - j = i; - break; + /* Find highest valid fclk pstate */ + for (i = 0; i < clock_table->NumDfPstatesEnabled; i++) { + if (is_valid_clock_value(clock_table->DfPstateTable[i].FClk) && + clock_table->DfPstateTable[i].FClk > max_fclk) { + max_fclk = clock_table->DfPstateTable[i].FClk; + max_pstate = i; } } - if (j == -1) { - /* clock table is all 0s, just use our own hardcode */ - ASSERT(0); - return; - } - - bw_params->clk_table.num_entries = j + 1; + /* We expect the table to contain at least one valid fclk entry. */ + ASSERT(is_valid_clock_value(max_fclk)); - /* dispclk and dppclk can be max at any voltage, same number of levels for both */ + /* Dispclk and dppclk can be max at any voltage, same number of levels for both */ if (clock_table->NumDispClkLevelsEnabled <= NUM_DISPCLK_DPM_LEVELS && clock_table->NumDispClkLevelsEnabled <= NUM_DPPCLK_DPM_LEVELS) { max_dispclk = find_max_clk_value(clock_table->DispClocks, clock_table->NumDispClkLevelsEnabled); max_dppclk = find_max_clk_value(clock_table->DppClocks, clock_table->NumDispClkLevelsEnabled); } else { + /* Invalid number of entries in the table from PMFW. */ ASSERT(0); } - for (i = 0; i < bw_params->clk_table.num_entries; i++, j--) { - bw_params->clk_table.entries[i].fclk_mhz = clock_table->DfPstateTable[j].FClk; - bw_params->clk_table.entries[i].memclk_mhz = clock_table->DfPstateTable[j].MemClk; - bw_params->clk_table.entries[i].voltage = clock_table->DfPstateTable[j].Voltage; - switch (clock_table->DfPstateTable[j].WckRatio) { - case WCK_RATIO_1_2: - bw_params->clk_table.entries[i].wck_ratio = 2; - break; - case WCK_RATIO_1_4: - bw_params->clk_table.entries[i].wck_ratio = 4; - break; - default: - bw_params->clk_table.entries[i].wck_ratio = 1; + /* Base the clock table on dcfclk, need at least one entry regardless of pmfw table */ + for (i = 0; i < clock_table->NumDcfClkLevelsEnabled; i++) { + uint32_t min_fclk = clock_table->DfPstateTable[0].FClk; + int j; + + for (j = 1; j < clock_table->NumDfPstatesEnabled; j++) { + if (is_valid_clock_value(clock_table->DfPstateTable[j].FClk) && + clock_table->DfPstateTable[j].FClk < min_fclk && + clock_table->DfPstateTable[j].Voltage <= clock_table->SocVoltage[i]) { + min_fclk = clock_table->DfPstateTable[j].FClk; + min_pstate = j; + } } - bw_params->clk_table.entries[i].dcfclk_mhz = find_clk_for_voltage(clock_table, clock_table->DcfClocks, clock_table->DfPstateTable[j].Voltage); - bw_params->clk_table.entries[i].socclk_mhz = find_clk_for_voltage(clock_table, clock_table->SocClocks, clock_table->DfPstateTable[j].Voltage); + + /* First search defaults for the clocks we don't read using closest lower or equal default dcfclk */ + for (j = bw_params->clk_table.num_entries - 1; j > 0; j--) + if (bw_params->clk_table.entries[j].dcfclk_mhz <= clock_table->DcfClocks[i]) + break; + + bw_params->clk_table.entries[i].phyclk_mhz = bw_params->clk_table.entries[j].phyclk_mhz; + bw_params->clk_table.entries[i].phyclk_d18_mhz = bw_params->clk_table.entries[j].phyclk_d18_mhz; + bw_params->clk_table.entries[i].dtbclk_mhz = bw_params->clk_table.entries[j].dtbclk_mhz; + + /* Now update clocks we do read */ + bw_params->clk_table.entries[i].fclk_mhz = min_fclk; + bw_params->clk_table.entries[i].memclk_mhz = clock_table->DfPstateTable[min_pstate].MemClk; + bw_params->clk_table.entries[i].voltage = clock_table->DfPstateTable[min_pstate].Voltage; + bw_params->clk_table.entries[i].dcfclk_mhz = clock_table->DcfClocks[i]; + bw_params->clk_table.entries[i].socclk_mhz = clock_table->SocClocks[i]; bw_params->clk_table.entries[i].dispclk_mhz = max_dispclk; bw_params->clk_table.entries[i].dppclk_mhz = max_dppclk; + bw_params->clk_table.entries[i].wck_ratio = convert_wck_ratio( + clock_table->DfPstateTable[min_pstate].WckRatio); + }; + + /* Make sure to include at least one entry at highest pstate */ + if (max_pstate != min_pstate || i == 0) { + if (i > MAX_NUM_DPM_LVL - 1) + i = MAX_NUM_DPM_LVL - 1; + + bw_params->clk_table.entries[i].fclk_mhz = max_fclk; + bw_params->clk_table.entries[i].memclk_mhz = clock_table->DfPstateTable[max_pstate].MemClk; + bw_params->clk_table.entries[i].voltage = clock_table->DfPstateTable[max_pstate].Voltage; + bw_params->clk_table.entries[i].dcfclk_mhz = find_max_clk_value(clock_table->DcfClocks, NUM_DCFCLK_DPM_LEVELS); + bw_params->clk_table.entries[i].socclk_mhz = find_max_clk_value(clock_table->SocClocks, NUM_SOCCLK_DPM_LEVELS); + bw_params->clk_table.entries[i].dispclk_mhz = max_dispclk; + bw_params->clk_table.entries[i].dppclk_mhz = max_dppclk; + bw_params->clk_table.entries[i].wck_ratio = convert_wck_ratio( + clock_table->DfPstateTable[max_pstate].WckRatio); + i++; } + bw_params->clk_table.num_entries = i--; + + /* Make sure all highest clocks are included*/ + bw_params->clk_table.entries[i].socclk_mhz = find_max_clk_value(clock_table->SocClocks, NUM_SOCCLK_DPM_LEVELS); + bw_params->clk_table.entries[i].dispclk_mhz = find_max_clk_value(clock_table->DispClocks, NUM_DISPCLK_DPM_LEVELS); + bw_params->clk_table.entries[i].dppclk_mhz = find_max_clk_value(clock_table->DppClocks, NUM_DPPCLK_DPM_LEVELS); + ASSERT(clock_table->DcfClocks[i] == find_max_clk_value(clock_table->DcfClocks, NUM_DCFCLK_DPM_LEVELS)); + bw_params->clk_table.entries[i].phyclk_mhz = def_max.phyclk_mhz; + bw_params->clk_table.entries[i].phyclk_d18_mhz = def_max.phyclk_d18_mhz; + bw_params->clk_table.entries[i].dtbclk_mhz = def_max.dtbclk_mhz; + /* + * Set any 0 clocks to max default setting. Not an issue for + * power since we aren't doing switching in such case anyway + */ + for (i = 0; i < bw_params->clk_table.num_entries; i++) { + if (!bw_params->clk_table.entries[i].fclk_mhz) { + bw_params->clk_table.entries[i].fclk_mhz = def_max.fclk_mhz; + bw_params->clk_table.entries[i].memclk_mhz = def_max.memclk_mhz; + bw_params->clk_table.entries[i].voltage = def_max.voltage; + } + if (!bw_params->clk_table.entries[i].dcfclk_mhz) + bw_params->clk_table.entries[i].dcfclk_mhz = def_max.dcfclk_mhz; + if (!bw_params->clk_table.entries[i].socclk_mhz) + bw_params->clk_table.entries[i].socclk_mhz = def_max.socclk_mhz; + if (!bw_params->clk_table.entries[i].dispclk_mhz) + bw_params->clk_table.entries[i].dispclk_mhz = def_max.dispclk_mhz; + if (!bw_params->clk_table.entries[i].dppclk_mhz) + bw_params->clk_table.entries[i].dppclk_mhz = def_max.dppclk_mhz; + if (!bw_params->clk_table.entries[i].phyclk_mhz) + bw_params->clk_table.entries[i].phyclk_mhz = def_max.phyclk_mhz; + if (!bw_params->clk_table.entries[i].phyclk_d18_mhz) + bw_params->clk_table.entries[i].phyclk_d18_mhz = def_max.phyclk_d18_mhz; + if (!bw_params->clk_table.entries[i].dtbclk_mhz) + bw_params->clk_table.entries[i].dtbclk_mhz = def_max.dtbclk_mhz; + } + ASSERT(bw_params->clk_table.entries[i-1].dcfclk_mhz); bw_params->vram_type = bios_info->memory_type; - bw_params->num_channels = bios_info->ma_channel_number; + bw_params->num_channels = bios_info->ma_channel_number ? bios_info->ma_channel_number : 4; for (i = 0; i < WM_SET_COUNT; i++) { bw_params->wm_table.entries[i].wm_inst = i; @@ -641,7 +689,7 @@ 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 = dcn314_init_clocks, + .init_clocks = dcn31_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 @@ -681,10 +729,10 @@ void dcn314_clk_mgr_construct( } ASSERT(clk_mgr->smu_wm_set.wm_set); - smu_dpm_clks.dpm_clks = (DpmClocks_t *)dm_helpers_allocate_gpu_mem( + smu_dpm_clks.dpm_clks = (DpmClocks314_t *)dm_helpers_allocate_gpu_mem( clk_mgr->base.base.ctx, DC_MEM_ALLOC_TYPE_FRAME_BUFFER, - sizeof(DpmClocks_t), + sizeof(DpmClocks314_t), &smu_dpm_clks.mc_address.quad_part); if (smu_dpm_clks.dpm_clks == NULL) { @@ -729,7 +777,7 @@ void dcn314_clk_mgr_construct( if (clk_mgr->base.base.ctx->dc->debug.pstate_enabled) { dcn314_get_dpm_table_from_smu(&clk_mgr->base, &smu_dpm_clks); - if (ctx->dc_bios && ctx->dc_bios->integrated_info) { + if (ctx->dc_bios && ctx->dc_bios->integrated_info && ctx->dc->config.use_default_clock_table == false) { dcn314_clk_mgr_helper_populate_bw_params( &clk_mgr->base, ctx->dc_bios->integrated_info, 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 c695a4498c50..171f84340eb2 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 @@ -42,7 +42,7 @@ struct clk_mgr_dcn314 { bool dcn314_are_clock_states_equal(struct dc_clocks *a, struct dc_clocks *b); -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/dcn314/dcn314_smu.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.h index a7958dc96581..047d19ea919c 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.h @@ -36,6 +36,37 @@ typedef enum { WCK_RATIO_MAX } WCK_RATIO_e; +typedef struct { + uint32_t FClk; + uint32_t MemClk; + uint32_t Voltage; + uint8_t WckRatio; + uint8_t Spare[3]; +} DfPstateTable314_t; + +//Freq in MHz +//Voltage in milli volts with 2 fractional bits +typedef struct { + uint32_t DcfClocks[NUM_DCFCLK_DPM_LEVELS]; + uint32_t DispClocks[NUM_DISPCLK_DPM_LEVELS]; + uint32_t DppClocks[NUM_DPPCLK_DPM_LEVELS]; + uint32_t SocClocks[NUM_SOCCLK_DPM_LEVELS]; + uint32_t VClocks[NUM_VCN_DPM_LEVELS]; + uint32_t DClocks[NUM_VCN_DPM_LEVELS]; + uint32_t SocVoltage[NUM_SOC_VOLTAGE_LEVELS]; + DfPstateTable314_t DfPstateTable[NUM_DF_PSTATE_LEVELS]; + + uint8_t NumDcfClkLevelsEnabled; + uint8_t NumDispClkLevelsEnabled; //Applies to both Dispclk and Dppclk + uint8_t NumSocClkLevelsEnabled; + uint8_t VcnClkLevelsEnabled; //Applies to both Vclk and Dclk + uint8_t NumDfPstatesEnabled; + uint8_t spare[3]; + + uint32_t MinGfxClk; + uint32_t MaxGfxClk; +} DpmClocks314_t; + struct dcn314_watermarks { // Watermarks WatermarkRowGeneric_t WatermarkRow[WM_COUNT][NUM_WM_RANGES]; @@ -43,7 +74,7 @@ struct dcn314_watermarks { }; struct dcn314_smu_dpm_clks { - DpmClocks_t *dpm_clks; + DpmClocks314_t *dpm_clks; union large_integer mc_address; }; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index e42f44fc1c08..fb22c3d70528 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -1074,8 +1074,15 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) struct dc_stream_state *old_stream = dc->current_state->res_ctx.pipe_ctx[i].stream; bool should_disable = true; - bool pipe_split_change = - context->res_ctx.pipe_ctx[i].top_pipe != dc->current_state->res_ctx.pipe_ctx[i].top_pipe; + bool pipe_split_change = false; + + if ((context->res_ctx.pipe_ctx[i].top_pipe) && + (dc->current_state->res_ctx.pipe_ctx[i].top_pipe)) + pipe_split_change = context->res_ctx.pipe_ctx[i].top_pipe->pipe_idx != + dc->current_state->res_ctx.pipe_ctx[i].top_pipe->pipe_idx; + else + pipe_split_change = context->res_ctx.pipe_ctx[i].top_pipe != + dc->current_state->res_ctx.pipe_ctx[i].top_pipe; for (j = 0; j < context->stream_count; j++) { if (old_stream == context->streams[j]) { @@ -1087,7 +1094,8 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) dc->current_state->stream_count != context->stream_count) should_disable = true; - if (old_stream && !dc->current_state->res_ctx.pipe_ctx[i].top_pipe) { + if (old_stream && !dc->current_state->res_ctx.pipe_ctx[i].top_pipe && + !dc->current_state->res_ctx.pipe_ctx[i].prev_odm_pipe) { struct pipe_ctx *old_pipe, *new_pipe; old_pipe = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -3229,7 +3237,7 @@ static void commit_planes_for_stream(struct dc *dc, odm_pipe->ttu_regs.min_ttu_vblank = MAX_TTU; } - if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed) { + if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed) if (top_pipe_to_program && top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable) { if (should_use_dmub_lock(stream->link)) { @@ -3247,7 +3255,6 @@ static void commit_planes_for_stream(struct dc *dc, top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable( top_pipe_to_program->stream_res.tg); } - } if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) { if (dc->hwss.subvp_pipe_control_lock) @@ -3466,7 +3473,7 @@ static void commit_planes_for_stream(struct dc *dc, dc->hwss.pipe_control_lock(dc, top_pipe_to_program, false); } - if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed) { + if ((update_type != UPDATE_TYPE_FAST) && stream->update_flags.bits.dsc_changed) if (top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_enable) { top_pipe_to_program->stream_res.tg->funcs->wait_for_state( top_pipe_to_program->stream_res.tg, @@ -3493,21 +3500,19 @@ static void commit_planes_for_stream(struct dc *dc, top_pipe_to_program->stream_res.tg->funcs->lock_doublebuffer_disable( top_pipe_to_program->stream_res.tg); } - } - if (update_type != UPDATE_TYPE_FAST) { + if (update_type != UPDATE_TYPE_FAST) dc->hwss.post_unlock_program_front_end(dc, context); - /* Since phantom pipe programming is moved to post_unlock_program_front_end, - * move the SubVP lock to after the phantom pipes have been setup - */ - if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) { - if (dc->hwss.subvp_pipe_control_lock) - dc->hwss.subvp_pipe_control_lock(dc, context, false, should_lock_all_pipes, NULL, subvp_prev_use); - } else { - if (dc->hwss.subvp_pipe_control_lock) - dc->hwss.subvp_pipe_control_lock(dc, context, false, should_lock_all_pipes, top_pipe_to_program, subvp_prev_use); - } + /* Since phantom pipe programming is moved to post_unlock_program_front_end, + * move the SubVP lock to after the phantom pipes have been setup + */ + if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) { + if (dc->hwss.subvp_pipe_control_lock) + dc->hwss.subvp_pipe_control_lock(dc, context, false, should_lock_all_pipes, NULL, subvp_prev_use); + } else { + if (dc->hwss.subvp_pipe_control_lock) + dc->hwss.subvp_pipe_control_lock(dc, context, false, should_lock_all_pipes, top_pipe_to_program, subvp_prev_use); } // Fire manual trigger only when bottom plane is flipped @@ -4292,7 +4297,7 @@ bool dc_is_dmub_outbox_supported(struct dc *dc) !dc->debug.dpia_debug.bits.disable_dpia) return true; - if (dc->ctx->asic_id.chip_family == AMDGPU_FAMILY_GC_11_0_2 && + if (dc->ctx->asic_id.chip_family == AMDGPU_FAMILY_GC_11_0_1 && !dc->debug.dpia_debug.bits.disable_dpia) return true; @@ -4340,6 +4345,7 @@ void dc_enable_dmub_outbox(struct dc *dc) struct dc_context *dc_ctx = dc->ctx; dmub_enable_outbox_notification(dc_ctx->dmub_srv); + DC_LOG_DC("%s: dmub outbox notifications enabled\n", __func__); } /** diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index 9e51338441d0..66d2ae7aacf5 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -3372,7 +3372,7 @@ bool dc_link_setup_psr(struct dc_link *link, switch(link->ctx->asic_id.chip_family) { case FAMILY_YELLOW_CARP: case AMDGPU_FAMILY_GC_10_3_6: - case AMDGPU_FAMILY_GC_11_0_2: + case AMDGPU_FAMILY_GC_11_0_1: if(!dc->debug.disable_z10) psr_context->psr_level.bits.SKIP_CRTC_DISABLE = false; break; 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 ffc0f1c0ea93..7dbab15bfa68 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -169,7 +169,7 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id) if (ASICREV_IS_GC_11_0_2(asic_id.hw_internal_rev)) dc_version = DCN_VERSION_3_21; break; - case AMDGPU_FAMILY_GC_11_0_2: + case AMDGPU_FAMILY_GC_11_0_1: dc_version = DCN_VERSION_3_14; break; default: diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 8e1e40083ec8..5908b60db313 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -47,7 +47,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.196" +#define DC_VER "3.2.198" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -213,6 +213,7 @@ struct dc_caps { uint32_t cache_num_ways; uint16_t subvp_fw_processing_delay_us; uint16_t subvp_prefetch_end_to_mall_start_us; + uint8_t subvp_swath_height_margin_lines; // subvp start line must be aligned to 2 x swath height uint16_t subvp_pstate_allow_width_us; uint16_t subvp_vertical_int_margin_us; bool seamless_odm; @@ -352,6 +353,7 @@ struct dc_config { bool use_pipe_ctx_sync_logic; bool ignore_dpref_ss; bool enable_mipi_converter_optimization; + bool use_default_clock_table; }; enum visual_confirm { @@ -609,6 +611,7 @@ struct dc_bounding_box_overrides { int percent_of_ideal_drambw; int dram_clock_change_latency_ns; int dummy_clock_change_latency_ns; + int fclk_clock_change_latency_ns; /* This forces a hard min on the DCFCLK we use * for DML. Unlike the debug option for forcing * DCFCLK, this override affects watermark calculations @@ -751,6 +754,7 @@ struct dc_debug_options { uint32_t mst_start_top_delay; uint8_t psr_power_use_phy_fsm; enum dml_hostvm_override_opts dml_hostvm_override; + bool dml_disallow_alternate_prefetch_modes; bool use_legacy_soc_bb_mechanism; bool exit_idle_opt_for_cursor_updates; bool enable_single_display_2to1_odm_policy; 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 2d61c2a91cee..09b304507bad 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -29,6 +29,7 @@ #include "dm_helpers.h" #include "dc_hw_types.h" #include "core_types.h" +#include "../basics/conversion.h" #define CTX dc_dmub_srv->ctx #define DC_LOGGER CTX->logger @@ -275,8 +276,7 @@ void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst) union dmub_rb_cmd cmd = { 0 }; cmd.drr_update.header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH; - // TODO: Uncomment once FW headers are promoted - //cmd.drr_update.header.sub_type = DMUB_CMD__FAMS_SET_MANUAL_TRIGGER; + cmd.drr_update.header.sub_type = DMUB_CMD__FAMS_SET_MANUAL_TRIGGER; cmd.drr_update.dmub_optc_state_req.tg_inst = tg_inst; cmd.drr_update.header.payload_bytes = sizeof(cmd.drr_update) - sizeof(cmd.drr_update.header); @@ -601,6 +601,7 @@ static void populate_subvp_cmd_pipe_info(struct dc *dc, &cmd->fw_assisted_mclk_switch_v2.config_data.pipe_data[cmd_pipe_index]; 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; + uint32_t out_num, out_den; pipe_data->mode = SUBVP; pipe_data->pipe_config.subvp_data.pix_clk_100hz = subvp_pipe->stream->timing.pix_clk_100hz; @@ -612,6 +613,16 @@ static void populate_subvp_cmd_pipe_info(struct dc *dc, main_timing->v_total - main_timing->v_front_porch - main_timing->v_addressable; pipe_data->pipe_config.subvp_data.mall_region_lines = phantom_timing->v_addressable; pipe_data->pipe_config.subvp_data.main_pipe_index = subvp_pipe->pipe_idx; + pipe_data->pipe_config.subvp_data.is_drr = subvp_pipe->stream->ignore_msa_timing_param; + + /* Calculate the scaling factor from the src and dst height. + * e.g. If 3840x2160 being downscaled to 1920x1080, the scaling factor is 1/2. + * Reduce the fraction 1080/2160 = 1/2 for the "scaling factor" + */ + reduce_fraction(subvp_pipe->stream->src.height, subvp_pipe->stream->dst.height, &out_num, &out_den); + // TODO: Uncomment below lines once DMCUB include headers are promoted + //pipe_data->pipe_config.subvp_data.scale_factor_numerator = out_num; + //pipe_data->pipe_config.subvp_data.scale_factor_denominator = out_den; // Prefetch lines is equal to VACTIVE + BP + VSYNC pipe_data->pipe_config.subvp_data.prefetch_lines = diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index a0af0f6afeef..9544abf75e84 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -344,6 +344,7 @@ enum dc_detect_reason { DETECT_REASON_HPDRX, DETECT_REASON_FALLBACK, DETECT_REASON_RETRAIN, + DETECT_REASON_TDR, }; bool dc_link_detect(struct dc_link *dc_link, enum dc_detect_reason reason); 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 213de8cabfad..165392380842 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 @@ -543,9 +543,11 @@ static void dce112_get_pix_clk_dividers_helper ( switch (pix_clk_params->color_depth) { case COLOR_DEPTH_101010: actual_pixel_clock_100hz = (actual_pixel_clock_100hz * 5) >> 2; + actual_pixel_clock_100hz -= actual_pixel_clock_100hz % 10; break; case COLOR_DEPTH_121212: actual_pixel_clock_100hz = (actual_pixel_clock_100hz * 6) >> 2; + actual_pixel_clock_100hz -= actual_pixel_clock_100hz % 10; break; case COLOR_DEPTH_161616: actual_pixel_clock_100hz = actual_pixel_clock_100hz * 2; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c index d4a6504dfe00..db7ca4b0cdb9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c @@ -361,8 +361,6 @@ void dpp1_cnv_setup ( select = INPUT_CSC_SELECT_ICSC; break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: - pixel_format = 22; - break; case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: pixel_format = 26; /* ARGB16161616_UNORM */ break; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c index b54c12400323..564e061ccb58 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -278,9 +278,6 @@ void hubp1_program_pixel_format( SURFACE_PIXEL_FORMAT, 10); break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: - REG_UPDATE(DCSURF_SURFACE_CONFIG, - SURFACE_PIXEL_FORMAT, 22); - break; case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: /*we use crossbar already*/ REG_UPDATE(DCSURF_SURFACE_CONFIG, SURFACE_PIXEL_FORMAT, 26); /* ARGB16161616_UNORM */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index bed783747f16..5b5d952b2b8c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -110,6 +110,7 @@ void dcn10_lock_all_pipes(struct dc *dc, */ if (pipe_ctx->top_pipe || !pipe_ctx->stream || + !pipe_ctx->plane_state || !tg->funcs->is_tg_enabled(tg)) continue; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c index 769974375b4b..8e9384094f6d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c @@ -131,6 +131,12 @@ struct mpcc *mpc1_get_mpcc_for_dpp(struct mpc_tree *tree, int dpp_id) while (tmp_mpcc != NULL) { if (tmp_mpcc->dpp_id == dpp_id) return tmp_mpcc; + + /* avoid circular linked list */ + ASSERT(tmp_mpcc != tmp_mpcc->mpcc_bot); + if (tmp_mpcc == tmp_mpcc->mpcc_bot) + break; + tmp_mpcc = tmp_mpcc->mpcc_bot; } return NULL; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index e1a9a45b03b6..3fc300cd1ce9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -465,6 +465,11 @@ void optc1_enable_optc_clock(struct timing_generator *optc, bool enable) OTG_CLOCK_ON, 1, 1, 1000); } else { + + //last chance to clear underflow, otherwise, it will always there due to clock is off. + if (optc->funcs->is_optc_underflow_occurred(optc) == true) + optc->funcs->clear_optc_underflow(optc); + REG_UPDATE_2(OTG_CLOCK_CONTROL, OTG_CLOCK_GATE_DIS, 0, OTG_CLOCK_EN, 0); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c index ea1f14af0db7..eaa7032f0f1a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c @@ -166,8 +166,6 @@ static void dpp2_cnv_setup ( select = DCN2_ICSC_SELECT_ICSC_A; break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: - pixel_format = 22; - break; case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: pixel_format = 26; /* ARGB16161616_UNORM */ break; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c index 936af65381ef..9570c2118ccc 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c @@ -463,9 +463,6 @@ void hubp2_program_pixel_format( SURFACE_PIXEL_FORMAT, 10); break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: - REG_UPDATE(DCSURF_SURFACE_CONFIG, - SURFACE_PIXEL_FORMAT, 22); - break; case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: /*we use crossbar already*/ REG_UPDATE(DCSURF_SURFACE_CONFIG, SURFACE_PIXEL_FORMAT, 26); /* ARGB16161616_UNORM */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c index 3d307dd58e9a..116f67a0b989 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c @@ -531,6 +531,12 @@ static struct mpcc *mpc2_get_mpcc_for_dpp(struct mpc_tree *tree, int dpp_id) while (tmp_mpcc != NULL) { if (tmp_mpcc->dpp_id == 0xf || tmp_mpcc->dpp_id == dpp_id) return tmp_mpcc; + + /* avoid circular linked list */ + ASSERT(tmp_mpcc != tmp_mpcc->mpcc_bot); + if (tmp_mpcc == tmp_mpcc->mpcc_bot) + break; + tmp_mpcc = tmp_mpcc->mpcc_bot; } return NULL; diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c index c5e200d09038..5752271f22df 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c @@ -67,9 +67,15 @@ static uint32_t convert_and_clamp( void dcn21_dchvm_init(struct hubbub *hubbub) { struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); - uint32_t riommu_active; + uint32_t riommu_active, prefetch_done; int i; + REG_GET(DCHVM_RIOMMU_STAT0, HOSTVM_PREFETCH_DONE, &prefetch_done); + + if (prefetch_done) { + hubbub->riommu_active = true; + return; + } //Init DCHVM block REG_UPDATE(DCHVM_CTRL0, HOSTVM_INIT_REQ, 1); diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c index 77b00f86c216..4a668d6563df 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c @@ -244,8 +244,6 @@ void dpp3_cnv_setup ( select = INPUT_CSC_SELECT_ICSC; break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: - pixel_format = 22; - break; case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: pixel_format = 26; /* ARGB16161616_UNORM */ break; diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c index 6a4dcafb9bba..dc3e8df706b3 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c @@ -86,7 +86,7 @@ bool hubp3_program_surface_flip_and_addr( VMID, address->vmid); if (address->type == PLN_ADDR_TYPE_GRPH_STEREO) { - REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_MODE_FOR_STEREOSYNC, 0x1); + REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_MODE_FOR_STEREOSYNC, 0); REG_UPDATE(DCSURF_FLIP_CONTROL, SURFACE_FLIP_IN_STEREOSYNC, 0x1); } else { diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c index 0a67f8a5656d..d97076648acb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c @@ -372,7 +372,7 @@ static struct stream_encoder *dcn303_stream_encoder_create(enum engine_id eng_id int afmt_inst; /* Mapping of VPG, AFMT, DME register blocks to DIO block instance */ - if (eng_id <= ENGINE_ID_DIGE) { + if (eng_id <= ENGINE_ID_DIGB) { vpg_inst = eng_id; afmt_inst = eng_id; } else 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 a788d160953b..ab70ebd8f223 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 @@ -104,6 +104,9 @@ static bool has_query_dp_alt(struct link_encoder *enc) { struct dc_dmub_srv *dc_dmub_srv = enc->ctx->dmub_srv; + if (enc->ctx->dce_version >= DCN_VERSION_3_15) + return true; + /* Supports development firmware and firmware >= 4.0.11 */ return dc_dmub_srv && !(dc_dmub_srv->dmub->fw_version >= DMUB_FW_VERSION(4, 0, 0) && diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.h index 7c77c71591a0..82c3b3ac1f0d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.h @@ -162,7 +162,8 @@ SE_SF(DP_SYM32_ENC0_DP_SYM32_ENC_SDP_AUDIO_CONTROL0, AIP_ENABLE, mask_sh),\ SE_SF(DP_SYM32_ENC0_DP_SYM32_ENC_SDP_AUDIO_CONTROL0, ACM_ENABLE, mask_sh),\ SE_SF(DP_SYM32_ENC0_DP_SYM32_ENC_VID_CRC_CONTROL, CRC_ENABLE, mask_sh),\ - SE_SF(DP_SYM32_ENC0_DP_SYM32_ENC_VID_CRC_CONTROL, CRC_CONT_MODE_ENABLE, mask_sh) + SE_SF(DP_SYM32_ENC0_DP_SYM32_ENC_VID_CRC_CONTROL, CRC_CONT_MODE_ENABLE, mask_sh),\ + SE_SF(DP_SYM32_ENC0_DP_SYM32_ENC_HBLANK_CONTROL, HBLANK_MINIMUM_SYMBOL_WIDTH, mask_sh) #define DCN3_1_HPO_DP_STREAM_ENC_REG_FIELD_LIST(type) \ diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c index 468a893ff785..aedff18aff56 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c @@ -2153,7 +2153,7 @@ static bool dcn31_resource_construct( pool->base.usb4_dpia_count = 4; } - if (dc->ctx->asic_id.chip_family == AMDGPU_FAMILY_GC_11_0_2) + if (dc->ctx->asic_id.chip_family == AMDGPU_FAMILY_GC_11_0_1) pool->base.usb4_dpia_count = 4; /* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.h b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.h index 41f8ec99da6b..901436591ed4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.h @@ -32,7 +32,6 @@ container_of(pool, struct dcn31_resource_pool, base) extern struct _vcs_dpi_ip_params_st dcn3_1_ip; -extern struct _vcs_dpi_soc_bounding_box_st dcn3_1_soc; struct dcn31_resource_pool { struct resource_pool base; diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/Makefile b/drivers/gpu/drm/amd/display/dc/dcn314/Makefile index e3b5a95e03b1..702c28c2560e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn314/Makefile @@ -13,31 +13,6 @@ DCN314 = dcn314_resource.o dcn314_hwseq.o dcn314_init.o \ dcn314_dio_stream_encoder.o dcn314_dccg.o dcn314_optc.o -ifdef CONFIG_X86 -CFLAGS_$(AMDDALPATH)/dc/dcn314/dcn314_resource.o := -mhard-float -msse -endif - -ifdef CONFIG_PPC64 -CFLAGS_$(AMDDALPATH)/dc/dcn314/dcn314_resource.o := -mhard-float -maltivec -endif - -ifdef CONFIG_CC_IS_GCC -ifeq ($(call cc-ifversion, -lt, 0701, y), y) -IS_OLD_GCC = 1 -endif -endif - -ifdef CONFIG_X86 -ifdef IS_OLD_GCC -# Stack alignment mismatch, proceed with caution. -# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 -# (8B stack alignment). -CFLAGS_$(AMDDALPATH)/dc/dcn314/dcn314_resource.o += -mpreferred-stack-boundary=4 -else -CFLAGS_$(AMDDALPATH)/dc/dcn314/dcn314_resource.o += -msse2 -endif -endif - AMD_DAL_DCN314 = $(addprefix $(AMDDALPATH)/dc/dcn314/,$(DCN314)) AMD_DISPLAY_FILES += $(AMD_DAL_DCN314) diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dio_stream_encoder.c index b384f30395d3..e3351ddc566c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dio_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dio_stream_encoder.c @@ -317,6 +317,7 @@ static void enc314_stream_encoder_dp_unblank( /* switch DP encoder to CRTC data, but reset it the fifo first. It may happen * that it overflows during mode transition, and sometimes doesn't recover. */ + REG_UPDATE(DIG_FIFO_CTRL0, DIG_FIFO_READ_START_LEVEL, 0x7); REG_UPDATE(DP_STEER_FIFO, DP_STEER_FIFO_RESET, 1); udelay(10); diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c index 755c715ad8dc..39931d48f385 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c @@ -343,7 +343,10 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig { struct dc_stream_state *stream = pipe_ctx->stream; unsigned int odm_combine_factor = 0; + struct dc *dc = pipe_ctx->stream->ctx->dc; + bool two_pix_per_container = false; + two_pix_per_container = optc2_is_two_pixels_per_containter(&stream->timing); odm_combine_factor = get_odm_config(pipe_ctx, NULL); if (is_dp_128b_132b_signal(pipe_ctx)) { @@ -355,16 +358,13 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig else *k2_div = PIXEL_RATE_DIV_BY_4; } else if (dc_is_dp_signal(pipe_ctx->stream->signal)) { - if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) { + if (two_pix_per_container) { *k1_div = PIXEL_RATE_DIV_BY_1; *k2_div = PIXEL_RATE_DIV_BY_2; - } else if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR422) { - *k1_div = PIXEL_RATE_DIV_BY_2; - *k2_div = PIXEL_RATE_DIV_BY_2; } else { - if (odm_combine_factor == 1) - *k2_div = PIXEL_RATE_DIV_BY_4; - else if (odm_combine_factor == 2) + *k1_div = PIXEL_RATE_DIV_BY_1; + *k2_div = PIXEL_RATE_DIV_BY_4; + if ((odm_combine_factor == 2) || dc->debug.enable_dp_dig_pixel_rate_div_policy) *k2_div = PIXEL_RATE_DIV_BY_2; } } @@ -374,3 +374,31 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig return odm_combine_factor; } + +void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx) +{ + uint32_t pix_per_cycle = 1; + uint32_t odm_combine_factor = 1; + + if (!pipe_ctx || !pipe_ctx->stream || !pipe_ctx->stream_res.stream_enc) + return; + + odm_combine_factor = get_odm_config(pipe_ctx, NULL); + if (optc2_is_two_pixels_per_containter(&pipe_ctx->stream->timing) || odm_combine_factor > 1 + || dcn314_is_dp_dig_pixel_rate_div_policy(pipe_ctx)) + pix_per_cycle = 2; + + if (pipe_ctx->stream_res.stream_enc->funcs->set_input_mode) + pipe_ctx->stream_res.stream_enc->funcs->set_input_mode(pipe_ctx->stream_res.stream_enc, + pix_per_cycle); +} + +bool dcn314_is_dp_dig_pixel_rate_div_policy(struct pipe_ctx *pipe_ctx) +{ + struct dc *dc = pipe_ctx->stream->ctx->dc; + + if (dc_is_dp_signal(pipe_ctx->stream->signal) && !is_dp_128b_132b_signal(pipe_ctx) && + dc->debug.enable_dp_dig_pixel_rate_div_policy) + return true; + return false; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h index be0f5e4d48e1..d014580592ac 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h @@ -39,4 +39,8 @@ void dcn314_enable_power_gating_plane(struct dce_hwseq *hws, bool enable); unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div); +void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx); + +bool dcn314_is_dp_dig_pixel_rate_div_policy(struct pipe_ctx *pipe_ctx); + #endif /* __DC_HWSS_DCN314_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c index b9debeb081fd..fcf67eb3478f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c @@ -145,6 +145,8 @@ static const struct hwseq_private_funcs dcn314_private_funcs = { .set_shaper_3dlut = dcn20_set_shaper_3dlut, .setup_hpo_hw_control = dcn31_setup_hpo_hw_control, .calculate_dccg_k1_k2_values = dcn314_calculate_dccg_k1_k2_values, + .set_pixels_per_cycle = dcn314_set_pixels_per_cycle, + .is_dp_dig_pixel_rate_div_policy = dcn314_is_dp_dig_pixel_rate_div_policy, }; void dcn314_hw_sequencer_construct(struct dc *dc) diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_optc.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_optc.c index 0c7980266b85..38aa28ec6b13 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_optc.c @@ -98,7 +98,8 @@ static void optc314_set_odm_combine(struct timing_generator *optc, int *opp_id, REG_UPDATE(OPTC_WIDTH_CONTROL, OPTC_SEGMENT_WIDTH, mpcc_hactive); - REG_SET(OTG_H_TIMING_CNTL, 0, OTG_H_TIMING_DIV_MODE, opp_cnt - 1); + REG_UPDATE(OTG_H_TIMING_CNTL, + OTG_H_TIMING_DIV_MODE, opp_cnt - 1); optc1->opp_count = opp_cnt; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c index 63861cdfb09f..2a2a4a9cc117 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c @@ -70,6 +70,7 @@ #include "dce110/dce110_resource.h" #include "dml/display_mode_vba.h" #include "dml/dcn31/dcn31_fpu.h" +#include "dml/dcn314/dcn314_fpu.h" #include "dcn314/dcn314_dccg.h" #include "dcn10/dcn10_resource.h" #include "dcn31/dcn31_panel_cntl.h" @@ -132,155 +133,6 @@ static const struct IP_BASE DCN_BASE = { { { { 0x00000012, 0x000000C0, 0x000034C #define DC_LOGGER_INIT(logger) -#define DCN3_14_DEFAULT_DET_SIZE 384 -#define DCN3_14_MAX_DET_SIZE 384 -#define DCN3_14_MIN_COMPBUF_SIZE_KB 128 -#define DCN3_14_CRB_SEGMENT_SIZE_KB 64 -struct _vcs_dpi_ip_params_st dcn3_14_ip = { - .VBlankNomDefaultUS = 668, - .gpuvm_enable = 1, - .gpuvm_max_page_table_levels = 1, - .hostvm_enable = 1, - .hostvm_max_page_table_levels = 2, - .rob_buffer_size_kbytes = 64, - .det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE, - .config_return_buffer_size_in_kbytes = 1792, - .compressed_buffer_segment_size_in_kbytes = 64, - .meta_fifo_size_in_kentries = 32, - .zero_size_buffer_entries = 512, - .compbuf_reserved_space_64b = 256, - .compbuf_reserved_space_zs = 64, - .dpp_output_buffer_pixels = 2560, - .opp_output_buffer_lines = 1, - .pixel_chunk_size_kbytes = 8, - .meta_chunk_size_kbytes = 2, - .min_meta_chunk_size_bytes = 256, - .writeback_chunk_size_kbytes = 8, - .ptoi_supported = false, - .num_dsc = 4, - .maximum_dsc_bits_per_component = 10, - .dsc422_native_support = false, - .is_line_buffer_bpp_fixed = true, - .line_buffer_fixed_bpp = 48, - .line_buffer_size_bits = 789504, - .max_line_buffer_lines = 12, - .writeback_interface_buffer_size_kbytes = 90, - .max_num_dpp = 4, - .max_num_otg = 4, - .max_num_hdmi_frl_outputs = 1, - .max_num_wb = 1, - .max_dchub_pscl_bw_pix_per_clk = 4, - .max_pscl_lb_bw_pix_per_clk = 2, - .max_lb_vscl_bw_pix_per_clk = 4, - .max_vscl_hscl_bw_pix_per_clk = 4, - .max_hscl_ratio = 6, - .max_vscl_ratio = 6, - .max_hscl_taps = 8, - .max_vscl_taps = 8, - .dpte_buffer_size_in_pte_reqs_luma = 64, - .dpte_buffer_size_in_pte_reqs_chroma = 34, - .dispclk_ramp_margin_percent = 1, - .max_inter_dcn_tile_repeaters = 8, - .cursor_buffer_size = 16, - .cursor_chunk_size = 2, - .writeback_line_buffer_buffer_size = 0, - .writeback_min_hscl_ratio = 1, - .writeback_min_vscl_ratio = 1, - .writeback_max_hscl_ratio = 1, - .writeback_max_vscl_ratio = 1, - .writeback_max_hscl_taps = 1, - .writeback_max_vscl_taps = 1, - .dppclk_delay_subtotal = 46, - .dppclk_delay_scl = 50, - .dppclk_delay_scl_lb_only = 16, - .dppclk_delay_cnvc_formatter = 27, - .dppclk_delay_cnvc_cursor = 6, - .dispclk_delay_subtotal = 119, - .dynamic_metadata_vm_enabled = false, - .odm_combine_4to1_supported = false, - .dcc_supported = true, -}; - -struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = { - /*TODO: correct dispclk/dppclk voltage level determination*/ - .clock_limits = { - { - .state = 0, - .dispclk_mhz = 1200.0, - .dppclk_mhz = 1200.0, - .phyclk_mhz = 600.0, - .phyclk_d18_mhz = 667.0, - .dscclk_mhz = 186.0, - .dtbclk_mhz = 625.0, - }, - { - .state = 1, - .dispclk_mhz = 1200.0, - .dppclk_mhz = 1200.0, - .phyclk_mhz = 810.0, - .phyclk_d18_mhz = 667.0, - .dscclk_mhz = 209.0, - .dtbclk_mhz = 625.0, - }, - { - .state = 2, - .dispclk_mhz = 1200.0, - .dppclk_mhz = 1200.0, - .phyclk_mhz = 810.0, - .phyclk_d18_mhz = 667.0, - .dscclk_mhz = 209.0, - .dtbclk_mhz = 625.0, - }, - { - .state = 3, - .dispclk_mhz = 1200.0, - .dppclk_mhz = 1200.0, - .phyclk_mhz = 810.0, - .phyclk_d18_mhz = 667.0, - .dscclk_mhz = 371.0, - .dtbclk_mhz = 625.0, - }, - { - .state = 4, - .dispclk_mhz = 1200.0, - .dppclk_mhz = 1200.0, - .phyclk_mhz = 810.0, - .phyclk_d18_mhz = 667.0, - .dscclk_mhz = 417.0, - .dtbclk_mhz = 625.0, - }, - }, - .num_states = 5, - .sr_exit_time_us = 9.0, - .sr_enter_plus_exit_time_us = 11.0, - .sr_exit_z8_time_us = 442.0, - .sr_enter_plus_exit_z8_time_us = 560.0, - .writeback_latency_us = 12.0, - .dram_channel_width_bytes = 4, - .round_trip_ping_latency_dcfclk_cycles = 106, - .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_sdp_bw_after_urgent = 80.0, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, - .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0, - .max_avg_sdp_bw_use_normal_percent = 60.0, - .max_avg_dram_bw_use_normal_percent = 60.0, - .fabric_datapath_to_dcn_data_return_bytes = 32, - .return_bus_width_bytes = 64, - .downspread_percent = 0.38, - .dcn_downspread_percent = 0.5, - .gpuvm_min_page_size_bytes = 4096, - .hostvm_min_page_size_bytes = 4096, - .do_urgent_latency_adjustment = false, - .urgent_latency_adjustment_fabric_clock_component_us = 0, - .urgent_latency_adjustment_fabric_clock_reference_mhz = 0, -}; - enum dcn31_clk_src_array_id { DCN31_CLK_SRC_PLL0, DCN31_CLK_SRC_PLL1, @@ -602,6 +454,7 @@ static const struct dcn31_hpo_dp_stream_encoder_registers hpo_dp_stream_enc_regs hpo_dp_stream_encoder_reg_list(0), hpo_dp_stream_encoder_reg_list(1), hpo_dp_stream_encoder_reg_list(2), + hpo_dp_stream_encoder_reg_list(3) }; static const struct dcn31_hpo_dp_stream_encoder_shift hpo_dp_se_shift = { @@ -1402,7 +1255,7 @@ static struct stream_encoder *dcn314_stream_encoder_create( int afmt_inst; /* Mapping of VPG, AFMT, DME register blocks to DIO block instance */ - if (eng_id <= ENGINE_ID_DIGF) { + if (eng_id < ENGINE_ID_DIGF) { vpg_inst = eng_id; afmt_inst = eng_id; } else @@ -1447,7 +1300,8 @@ static struct hpo_dp_stream_encoder *dcn31_hpo_dp_stream_encoder_create( * VPG[8] -> HPO_DP[2] * VPG[9] -> HPO_DP[3] */ - vpg_inst = hpo_dp_inst + 6; + //Uses offset index 5-8, but actually maps to vpg_inst 6-9 + vpg_inst = hpo_dp_inst + 5; /* Mapping of APG register blocks to HPO DP block instance: * APG[0] -> HPO_DP[0] @@ -1793,109 +1647,16 @@ static struct clock_source *dcn31_clock_source_create( return NULL; } -static bool is_dual_plane(enum surface_pixel_format format) -{ - return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN || format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA; -} - static int dcn314_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, bool fast_validate) { - int i, pipe_cnt; - struct resource_context *res_ctx = &context->res_ctx; - struct pipe_ctx *pipe; - bool upscaled = false; - - dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); - - for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { - struct dc_crtc_timing *timing; - - if (!res_ctx->pipe_ctx[i].stream) - continue; - pipe = &res_ctx->pipe_ctx[i]; - timing = &pipe->stream->timing; - - if (dc_extended_blank_supported(dc) && pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min - && pipe->stream->adjust.v_total_min > timing->v_total) - pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min; - - if (pipe->plane_state && - (pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height || - pipe->plane_state->src_rect.width < pipe->plane_state->dst_rect.width)) - upscaled = true; - - /* - * Immediate flip can be set dynamically after enabling the plane. - * We need to require support for immediate flip or underflow can be - * intermittently experienced depending on peak b/w requirements. - */ - pipes[pipe_cnt].pipe.src.immediate_flip = true; - - pipes[pipe_cnt].pipe.src.unbounded_req_mode = false; - pipes[pipe_cnt].pipe.src.hostvm = dc->res_pool->hubbub->riommu_active; - pipes[pipe_cnt].pipe.src.gpuvm = true; - pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0; - pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0; - pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch; - pipes[pipe_cnt].pipe.src.dcc_rate = 3; - pipes[pipe_cnt].dout.dsc_input_bpc = 0; - - if (pipes[pipe_cnt].dout.dsc_enable) { - switch (timing->display_color_depth) { - case COLOR_DEPTH_888: - pipes[pipe_cnt].dout.dsc_input_bpc = 8; - break; - case COLOR_DEPTH_101010: - pipes[pipe_cnt].dout.dsc_input_bpc = 10; - break; - case COLOR_DEPTH_121212: - pipes[pipe_cnt].dout.dsc_input_bpc = 12; - break; - default: - ASSERT(0); - break; - } - } - - pipe_cnt++; - } - context->bw_ctx.dml.ip.det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE; - - dc->config.enable_4to1MPC = false; - if (pipe_cnt == 1 && pipe->plane_state && !dc->debug.disable_z9_mpc) { - if (is_dual_plane(pipe->plane_state->format) - && pipe->plane_state->src_rect.width <= 1920 && pipe->plane_state->src_rect.height <= 1080) { - dc->config.enable_4to1MPC = true; - } else if (!is_dual_plane(pipe->plane_state->format) && pipe->plane_state->src_rect.width <= 5120) { - /* Limit to 5k max to avoid forced pipe split when there is not enough detile for swath */ - context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192; - pipes[0].pipe.src.unbounded_req_mode = true; - } - } else if (context->stream_count >= dc->debug.crb_alloc_policy_min_disp_count - && dc->debug.crb_alloc_policy > DET_SIZE_DEFAULT) { - context->bw_ctx.dml.ip.det_buffer_size_kbytes = dc->debug.crb_alloc_policy * 64; - } else if (context->stream_count >= 3 && upscaled) { - context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192; - } - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - - if (!pipe->stream) - continue; + int pipe_cnt; - if (pipe->stream->signal == SIGNAL_TYPE_EDP && dc->debug.seamless_boot_odm_combine && - pipe->stream->apply_seamless_boot_optimization) { - - if (pipe->stream->apply_boot_odm_mode == dm_odm_combine_policy_2to1) { - context->bw_ctx.dml.vba.ODMCombinePolicy = dm_odm_combine_policy_2to1; - break; - } - } - } + DC_FP_START(); + pipe_cnt = dcn314_populate_dml_pipes_from_context_fpu(dc, context, pipes, fast_validate); + DC_FP_END(); return pipe_cnt; } @@ -1906,88 +1667,9 @@ static struct dc_cap_funcs cap_funcs = { static void dcn314_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { - struct clk_limit_table *clk_table = &bw_params->clk_table; - struct _vcs_dpi_voltage_scaling_st *clock_tmp = dcn3_14_soc._clock_tmp; - unsigned int i, closest_clk_lvl; - int max_dispclk_mhz = 0, max_dppclk_mhz = 0; - int j; - - // Default clock levels are used for diags, which may lead to overclocking. - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { - - dcn3_14_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator; - dcn3_14_ip.max_num_dpp = dc->res_pool->pipe_count; - - if (bw_params->num_channels > 0) - dcn3_14_soc.num_chans = bw_params->num_channels; - - ASSERT(dcn3_14_soc.num_chans); - ASSERT(clk_table->num_entries); - - /* Prepass to find max clocks independent of voltage level. */ - for (i = 0; i < clk_table->num_entries; ++i) { - if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz) - max_dispclk_mhz = clk_table->entries[i].dispclk_mhz; - if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz) - max_dppclk_mhz = clk_table->entries[i].dppclk_mhz; - } - - for (i = 0; i < clk_table->num_entries; i++) { - /* loop backwards*/ - for (closest_clk_lvl = 0, j = dcn3_14_soc.num_states - 1; j >= 0; j--) { - if ((unsigned int) dcn3_14_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { - closest_clk_lvl = j; - break; - } - } - if (clk_table->num_entries == 1) { - /*smu gives one DPM level, let's take the highest one*/ - closest_clk_lvl = dcn3_14_soc.num_states - 1; - } - - clock_tmp[i].state = i; - - /* Clocks dependent on voltage level. */ - clock_tmp[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; - if (clk_table->num_entries == 1 && - clock_tmp[i].dcfclk_mhz < dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) { - /*SMU fix not released yet*/ - clock_tmp[i].dcfclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz; - } - clock_tmp[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; - clock_tmp[i].socclk_mhz = clk_table->entries[i].socclk_mhz; - - if (clk_table->entries[i].memclk_mhz && clk_table->entries[i].wck_ratio) - clock_tmp[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2 * clk_table->entries[i].wck_ratio; - - /* Clocks independent of voltage level. */ - clock_tmp[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz : - dcn3_14_soc.clock_limits[closest_clk_lvl].dispclk_mhz; - - clock_tmp[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz : - dcn3_14_soc.clock_limits[closest_clk_lvl].dppclk_mhz; - - clock_tmp[i].dram_bw_per_chan_gbps = dcn3_14_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; - clock_tmp[i].dscclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dscclk_mhz; - clock_tmp[i].dtbclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; - clock_tmp[i].phyclk_d18_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; - clock_tmp[i].phyclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_mhz; - } - for (i = 0; i < clk_table->num_entries; i++) - dcn3_14_soc.clock_limits[i] = clock_tmp[i]; - if (clk_table->num_entries) - dcn3_14_soc.num_states = clk_table->num_entries; - } - - if (max_dispclk_mhz) { - dcn3_14_soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2; - dc->dml.soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2; - } - - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31); - else - dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31_FPGA); + DC_FP_START(); + dcn314_update_bw_bounding_box_fpu(dc, bw_params); + DC_FP_END(); } static struct resource_funcs dcn314_res_pool_funcs = { @@ -2069,6 +1751,7 @@ static bool dcn314_resource_construct( dc->caps.post_blend_color_processing = true; dc->caps.force_dp_tps4_for_cp2520 = true; dc->caps.dp_hpo = true; + dc->caps.dp_hdmi21_pcon_support = true; dc->caps.edp_dsc_support = true; dc->caps.extended_aux_timeout_support = true; dc->caps.dmcub_support = true; diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.h index c41108847ce0..0dd3153aa5c1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.h @@ -29,6 +29,9 @@ #include "core_types.h" +extern struct _vcs_dpi_ip_params_st dcn3_14_ip; +extern struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc; + #define TO_DCN314_RES_POOL(pool)\ container_of(pool, struct dcn314_resource_pool, base) diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.h b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.h index 39929fa67a51..22849eaa6f24 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.h @@ -32,7 +32,6 @@ container_of(pool, struct dcn315_resource_pool, base) extern struct _vcs_dpi_ip_params_st dcn3_15_ip; -extern struct _vcs_dpi_ip_params_st dcn3_15_soc; struct dcn315_resource_pool { struct resource_pool base; diff --git a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.h b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.h index 0dc5a6c13ae7..aba6d634131b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.h @@ -32,7 +32,6 @@ container_of(pool, struct dcn316_resource_pool, base) extern struct _vcs_dpi_ip_params_st dcn3_16_ip; -extern struct _vcs_dpi_ip_params_st dcn3_16_soc; struct dcn316_resource_pool { struct resource_pool base; diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c index a31c64b50410..0d5e8a441512 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c @@ -225,19 +225,19 @@ void dccg32_set_dpstreamclk( case 0: REG_UPDATE_2(DPSTREAMCLK_CNTL, DPSTREAMCLK0_EN, - (src == REFCLK) ? 0 : 1, DPSTREAMCLK0_SRC_SEL, 0); + (src == REFCLK) ? 0 : 1, DPSTREAMCLK0_SRC_SEL, otg_inst); break; case 1: REG_UPDATE_2(DPSTREAMCLK_CNTL, DPSTREAMCLK1_EN, - (src == REFCLK) ? 0 : 1, DPSTREAMCLK1_SRC_SEL, 1); + (src == REFCLK) ? 0 : 1, DPSTREAMCLK1_SRC_SEL, otg_inst); break; case 2: REG_UPDATE_2(DPSTREAMCLK_CNTL, DPSTREAMCLK2_EN, - (src == REFCLK) ? 0 : 1, DPSTREAMCLK2_SRC_SEL, 2); + (src == REFCLK) ? 0 : 1, DPSTREAMCLK2_SRC_SEL, otg_inst); break; case 3: REG_UPDATE_2(DPSTREAMCLK_CNTL, DPSTREAMCLK3_EN, - (src == REFCLK) ? 0 : 1, DPSTREAMCLK3_SRC_SEL, 3); + (src == REFCLK) ? 0 : 1, DPSTREAMCLK3_SRC_SEL, otg_inst); break; default: BREAK_TO_DEBUGGER(); diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_stream_encoder.c index 26648ce772da..38a48983f663 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dio_stream_encoder.c @@ -310,6 +310,11 @@ static void enc32_stream_encoder_dp_unblank( // TODO: Confirm if we need to wait for DIG_SYMCLK_FE_ON REG_WAIT(DIG_FE_CNTL, DIG_SYMCLK_FE_ON, 1, 10, 5000); + /* read start level = 0 will bring underflow / overflow and DIG_FIFO_ERROR = 1 + * so set it to 1/2 full = 7 before reset as suggested by hardware team. + */ + REG_UPDATE(DIG_FIFO_CTRL0, DIG_FIFO_READ_START_LEVEL, 0x7); + REG_UPDATE(DIG_FIFO_CTRL0, DIG_FIFO_RESET, 1); REG_WAIT(DIG_FIFO_CTRL0, DIG_FIFO_RESET_DONE, 1, 10, 5000); diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c index d38341f68b17..8d9d96c39808 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c @@ -250,6 +250,7 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c uint32_t total_lines = 0; uint32_t lines_per_way = 0; uint32_t num_ways = 0; + uint32_t prev_addr_low = 0; for (i = 0; i < ctx->stream_count; i++) { stream = ctx->streams[i]; @@ -267,10 +268,20 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c plane = ctx->stream_status[i].plane_states[j]; // Calculate total surface size - surface_size = plane->plane_size.surface_pitch * + if (prev_addr_low != plane->address.grph.addr.u.low_part) { + /* if plane address are different from prev FB, then userspace allocated separate FBs*/ + surface_size += plane->plane_size.surface_pitch * plane->plane_size.surface_size.height * (plane->format >= SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616 ? 8 : 4); + prev_addr_low = plane->address.grph.addr.u.low_part; + } else { + /* We have the same fb for all the planes. + * Xorg always creates one giant fb that holds all surfaces, + * so allocating it once is sufficient. + * */ + continue; + } // Convert surface size + starting address to number of cache lines required // (alignment accounted for) cache_lines_used += dcn32_cache_lines_for_surface(dc, surface_size, @@ -284,24 +295,38 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c } // Include cursor size for CAB allocation - if (stream->cursor_position.enable && plane->address.grph.cursor_cache_addr.quad_part) { - cursor_size = dc->caps.max_cursor_size * dc->caps.max_cursor_size; - switch (stream->cursor_attributes.color_format) { - case CURSOR_MODE_MONO: - cursor_size /= 2; - break; - case CURSOR_MODE_COLOR_1BIT_AND: - case CURSOR_MODE_COLOR_PRE_MULTIPLIED_ALPHA: - case CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA: - cursor_size *= 4; - break; + for (j = 0; j < dc->res_pool->pipe_count; j++) { + struct pipe_ctx *pipe = &ctx->res_ctx.pipe_ctx[j]; + struct hubp *hubp = pipe->plane_res.hubp; - case CURSOR_MODE_COLOR_64BIT_FP_PRE_MULTIPLIED: - case CURSOR_MODE_COLOR_64BIT_FP_UN_PRE_MULTIPLIED: - cursor_size *= 8; - break; - } - cache_lines_used += dcn32_cache_lines_for_surface(dc, surface_size, + if (pipe->stream && pipe->plane_state && hubp) + /* Find the cursor plane and use the exact size instead of + * using the max for calculation + */ + if (hubp->curs_attr.width > 0) { + cursor_size = hubp->curs_attr.width * hubp->curs_attr.height; + break; + } + } + + switch (stream->cursor_attributes.color_format) { + case CURSOR_MODE_MONO: + cursor_size /= 2; + break; + case CURSOR_MODE_COLOR_1BIT_AND: + case CURSOR_MODE_COLOR_PRE_MULTIPLIED_ALPHA: + case CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA: + cursor_size *= 4; + break; + + case CURSOR_MODE_COLOR_64BIT_FP_PRE_MULTIPLIED: + case CURSOR_MODE_COLOR_64BIT_FP_UN_PRE_MULTIPLIED: + cursor_size *= 8; + break; + } + + if (stream->cursor_position.enable && plane->address.grph.cursor_cache_addr.quad_part) { + cache_lines_used += dcn32_cache_lines_for_surface(dc, cursor_size, plane->address.grph.cursor_cache_addr.quad_part); } } @@ -314,13 +339,36 @@ static uint32_t dcn32_calculate_cab_allocation(struct dc *dc, struct dc_state *c if (cache_lines_used % lines_per_way > 0) num_ways++; + for (i = 0; i < ctx->stream_count; i++) { + stream = ctx->streams[i]; + for (j = 0; j < ctx->stream_status[i].plane_count; j++) { + plane = ctx->stream_status[i].plane_states[j]; + + if (stream->cursor_position.enable && plane && + !plane->address.grph.cursor_cache_addr.quad_part && + cursor_size > 16384) { + /* Cursor caching is not supported since it won't be on the same line. + * So we need an extra line to accommodate it. With large cursors and a single 4k monitor + * this case triggers corruption. If we're at the edge, then dont trigger display refresh + * from MALL. We only need to cache cursor if its greater that 64x64 at 4 bpp. + */ + num_ways++; + /* We only expect one cursor plane */ + break; + } + } + } + return num_ways; } bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable) { union dmub_rb_cmd cmd; - uint8_t ways; + uint8_t ways, i; + int j; + bool stereo_in_use = false; + struct dc_plane_state *plane = NULL; if (!dc->ctx->dmub_srv) return false; @@ -349,7 +397,23 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable) * and configure HUBP's to fetch from MALL */ ways = dcn32_calculate_cab_allocation(dc, dc->current_state); - if (ways <= dc->caps.cache_num_ways) { + + /* MALL not supported with Stereo3D. If any plane is using stereo, + * don't try to enter MALL. + */ + for (i = 0; i < dc->current_state->stream_count; i++) { + for (j = 0; j < dc->current_state->stream_status[i].plane_count; j++) { + plane = dc->current_state->stream_status[i].plane_states[j]; + + if (plane->address.type == PLN_ADDR_TYPE_GRPH_STEREO) { + stereo_in_use = true; + break; + } + } + if (stereo_in_use) + break; + } + if (ways <= dc->caps.cache_num_ways && !stereo_in_use) { memset(&cmd, 0, sizeof(cmd)); cmd.cab.header.type = DMUB_CMD__CAB_FOR_SS; cmd.cab.header.sub_type = DMUB_CMD__CAB_DCN_SS_FIT_IN_CAB; @@ -683,9 +747,11 @@ void dcn32_update_mall_sel(struct dc *dc, struct dc_state *context) if (pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { hubp->funcs->hubp_update_mall_sel(hubp, 1, false); } else { + // MALL not supported with Stereo3D hubp->funcs->hubp_update_mall_sel(hubp, num_ways <= dc->caps.cache_num_ways && - pipe->stream->link->psr_settings.psr_version == DC_PSR_VERSION_UNSUPPORTED ? 2 : 0, + pipe->stream->link->psr_settings.psr_version == DC_PSR_VERSION_UNSUPPORTED && + pipe->plane_state->address.type != PLN_ADDR_TYPE_GRPH_STEREO ? 2 : 0, cache_cursor); } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c index eff1f4e17689..1fad7b48bd5b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c @@ -281,7 +281,7 @@ static struct timing_generator_funcs dcn32_tg_funcs = { .lock_doublebuffer_enable = optc3_lock_doublebuffer_enable, .lock_doublebuffer_disable = optc3_lock_doublebuffer_disable, .enable_optc_clock = optc1_enable_optc_clock, - .set_drr = optc31_set_drr, // TODO: Update to optc32_set_drr once FW headers are promoted + .set_drr = optc32_set_drr, .get_last_used_drr_vtotal = optc2_get_last_used_drr_vtotal, .set_vtotal_min_max = optc3_set_vtotal_min_max, .set_static_screen_control = optc1_set_static_screen_control, diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c index 9a26d24b579f..8b887b552f2c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c @@ -867,7 +867,7 @@ static const struct dc_debug_options debug_defaults_drv = { } }, .use_max_lb = true, - .force_disable_subvp = true, + .force_disable_subvp = false, .exit_idle_opt_for_cursor_updates = true, .enable_single_display_2to1_odm_policy = true, .enable_dp_dig_pixel_rate_div_policy = 1, @@ -2051,6 +2051,7 @@ static bool dcn32_resource_construct( dc->caps.max_cab_allocation_bytes = 67108864; // 64MB = 1024 * 1024 * 64 dc->caps.subvp_fw_processing_delay_us = 15; dc->caps.subvp_prefetch_end_to_mall_start_us = 15; + dc->caps.subvp_swath_height_margin_lines = 16; dc->caps.subvp_pstate_allow_width_us = 20; dc->caps.subvp_vertical_int_margin_us = 30; 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 b3f8503cea9c..ab918fe38f6a 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 @@ -63,7 +63,7 @@ uint32_t dcn32_helper_calculate_num_ways_for_subvp(struct dc *dc, struct dc_stat if (pipe->stream && pipe->plane_state && !pipe->top_pipe && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { bytes_per_pixel = pipe->plane_state->format >= SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616 ? 8 : 4; - mall_region_pixels = pipe->stream->timing.h_addressable * pipe->stream->timing.v_addressable; + mall_region_pixels = pipe->plane_state->plane_size.surface_pitch * pipe->stream->timing.v_addressable; // For bytes required in MALL, calculate based on number of MBlks required num_mblks = (mall_region_pixels * bytes_per_pixel + @@ -144,7 +144,7 @@ bool dcn32_all_pipes_have_stream_and_plane(struct dc *dc, struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; if (!pipe->stream) - continue; + return false; if (!pipe->plane_state) return false; diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c index 8157e40d2c7e..c8b7d6ff38f4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c @@ -868,7 +868,7 @@ static const struct dc_debug_options debug_defaults_drv = { } }, .use_max_lb = true, - .force_disable_subvp = true, + .force_disable_subvp = false, .exit_idle_opt_for_cursor_updates = true, .enable_single_display_2to1_odm_policy = true, .enable_dp_dig_pixel_rate_div_policy = 1, @@ -1662,8 +1662,9 @@ static bool dcn321_resource_construct( dc->caps.max_cab_allocation_bytes = 33554432; // 32MB = 1024 * 1024 * 32 dc->caps.subvp_fw_processing_delay_us = 15; dc->caps.subvp_prefetch_end_to_mall_start_us = 15; + dc->caps.subvp_swath_height_margin_lines = 16; dc->caps.subvp_pstate_allow_width_us = 20; - + dc->caps.subvp_vertical_int_margin_us = 30; dc->caps.max_slave_planes = 1; dc->caps.max_slave_yuv_planes = 1; dc->caps.max_slave_rgb_planes = 1; diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index 359f6e9a1da0..86a3b5bfd699 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -61,7 +61,6 @@ CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn10/dcn10_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/dcn20_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20v2.o := $(dml_ccflags) @@ -71,6 +70,7 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_mode_vba_30.o := $(dml_ccflags) $(fram CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_rq_dlg_calc_30.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_mode_vba_31.o := $(dml_ccflags) $(frame_warn_flag) CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_rq_dlg_calc_31.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn314/dcn314_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/dcn30_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/dcn32_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn32/display_mode_vba_32.o := $(dml_ccflags) $(frame_warn_flag) @@ -82,7 +82,6 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn301/dcn301_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn302/dcn302_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn303/dcn303_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dsc/rc_calc_fpu.o := $(dml_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calcs.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_auto.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_ccflags) -Wno-tautological-compare @@ -131,6 +130,7 @@ DML += dcn321/dcn321_fpu.o DML += dcn301/dcn301_fpu.o DML += dcn302/dcn302_fpu.o DML += dcn303/dcn303_fpu.o +DML += dcn314/dcn314_fpu.o DML += dsc/rc_calc_fpu.o DML += calcs/dcn_calcs.o calcs/dcn_calc_math.o calcs/dcn_calc_auto.o endif 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 ca44df4fca74..d34e0f1314d9 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 @@ -30,6 +30,7 @@ #include "dchubbub.h" #include "dcn20/dcn20_resource.h" #include "dcn21/dcn21_resource.h" +#include "clk_mgr/dcn21/rn_clk_mgr.h" #include "dcn20_fpu.h" diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c index 7ef66e511ec8..d211cf6d234c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c @@ -26,6 +26,7 @@ #include "clk_mgr.h" #include "dcn20/dcn20_resource.h" #include "dcn301/dcn301_resource.h" +#include "clk_mgr/dcn301/vg_clk_mgr.h" #include "dml/dcn20/dcn20_fpu.h" #include "dcn301_fpu.h" diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c index e36cfa5985ea..149a1b17cdf3 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c @@ -25,6 +25,9 @@ #include "resource.h" #include "clk_mgr.h" +#include "dcn31/dcn31_resource.h" +#include "dcn315/dcn315_resource.h" +#include "dcn316/dcn316_resource.h" #include "dml/dcn20/dcn20_fpu.h" #include "dcn31_fpu.h" @@ -114,7 +117,7 @@ struct _vcs_dpi_ip_params_st dcn3_1_ip = { .dcc_supported = true, }; -struct _vcs_dpi_soc_bounding_box_st dcn3_1_soc = { +static struct _vcs_dpi_soc_bounding_box_st dcn3_1_soc = { /*TODO: correct dispclk/dppclk voltage level determination*/ .clock_limits = { { @@ -259,7 +262,7 @@ struct _vcs_dpi_ip_params_st dcn3_15_ip = { .dcc_supported = true, }; -struct _vcs_dpi_soc_bounding_box_st dcn3_15_soc = { +static struct _vcs_dpi_soc_bounding_box_st dcn3_15_soc = { .sr_exit_time_us = 9.0, .sr_enter_plus_exit_time_us = 11.0, .sr_exit_z8_time_us = 50.0, @@ -355,7 +358,7 @@ struct _vcs_dpi_ip_params_st dcn3_16_ip = { .dcc_supported = true, }; -struct _vcs_dpi_soc_bounding_box_st dcn3_16_soc = { +static struct _vcs_dpi_soc_bounding_box_st dcn3_16_soc = { /*TODO: correct dispclk/dppclk voltage level determination*/ .clock_limits = { { diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c index 3fab19134480..d63b4209b14c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c @@ -26,7 +26,7 @@ #include "dc.h" #include "dc_link.h" #include "../display_mode_lib.h" -#include "dml/dcn30/display_mode_vba_30.h" +#include "../dcn30/display_mode_vba_30.h" #include "display_mode_vba_31.h" #include "../dml_inline_defs.h" diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c index 66b82e4f05c6..35d10b4d018b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c @@ -27,7 +27,7 @@ #include "../display_mode_vba.h" #include "../dml_inline_defs.h" #include "display_rq_dlg_calc_31.h" -#include "dml/dcn30/display_mode_vba_30.h" +#include "../dcn30/display_mode_vba_30.h" static bool is_dual_plane(enum source_format_class source_format) { diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c new file mode 100644 index 000000000000..34a5d0f87b5f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c @@ -0,0 +1,376 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright 2022 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 "clk_mgr.h" +#include "resource.h" +#include "dcn31/dcn31_hubbub.h" +#include "dcn314_fpu.h" +#include "dml/dcn20/dcn20_fpu.h" +#include "dml/display_mode_vba.h" + +struct _vcs_dpi_ip_params_st dcn3_14_ip = { + .VBlankNomDefaultUS = 668, + .gpuvm_enable = 1, + .gpuvm_max_page_table_levels = 1, + .hostvm_enable = 1, + .hostvm_max_page_table_levels = 2, + .rob_buffer_size_kbytes = 64, + .det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE, + .config_return_buffer_size_in_kbytes = 1792, + .compressed_buffer_segment_size_in_kbytes = 64, + .meta_fifo_size_in_kentries = 32, + .zero_size_buffer_entries = 512, + .compbuf_reserved_space_64b = 256, + .compbuf_reserved_space_zs = 64, + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .meta_chunk_size_kbytes = 2, + .min_meta_chunk_size_bytes = 256, + .writeback_chunk_size_kbytes = 8, + .ptoi_supported = false, + .num_dsc = 4, + .maximum_dsc_bits_per_component = 10, + .dsc422_native_support = false, + .is_line_buffer_bpp_fixed = true, + .line_buffer_fixed_bpp = 48, + .line_buffer_size_bits = 789504, + .max_line_buffer_lines = 12, + .writeback_interface_buffer_size_kbytes = 90, + .max_num_dpp = 4, + .max_num_otg = 4, + .max_num_hdmi_frl_outputs = 1, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 6, + .max_vscl_ratio = 6, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dpte_buffer_size_in_pte_reqs_luma = 64, + .dpte_buffer_size_in_pte_reqs_chroma = 34, + .dispclk_ramp_margin_percent = 1, + .max_inter_dcn_tile_repeaters = 8, + .cursor_buffer_size = 16, + .cursor_chunk_size = 2, + .writeback_line_buffer_buffer_size = 0, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_max_hscl_taps = 1, + .writeback_max_vscl_taps = 1, + .dppclk_delay_subtotal = 46, + .dppclk_delay_scl = 50, + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_cnvc_formatter = 27, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 119, + .dynamic_metadata_vm_enabled = false, + .odm_combine_4to1_supported = false, + .dcc_supported = true, +}; + +struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = { + /*TODO: correct dispclk/dppclk voltage level determination*/ + .clock_limits = { + { + .state = 0, + .dispclk_mhz = 1200.0, + .dppclk_mhz = 1200.0, + .phyclk_mhz = 600.0, + .phyclk_d18_mhz = 667.0, + .dscclk_mhz = 186.0, + .dtbclk_mhz = 600.0, + }, + { + .state = 1, + .dispclk_mhz = 1200.0, + .dppclk_mhz = 1200.0, + .phyclk_mhz = 810.0, + .phyclk_d18_mhz = 667.0, + .dscclk_mhz = 209.0, + .dtbclk_mhz = 600.0, + }, + { + .state = 2, + .dispclk_mhz = 1200.0, + .dppclk_mhz = 1200.0, + .phyclk_mhz = 810.0, + .phyclk_d18_mhz = 667.0, + .dscclk_mhz = 209.0, + .dtbclk_mhz = 600.0, + }, + { + .state = 3, + .dispclk_mhz = 1200.0, + .dppclk_mhz = 1200.0, + .phyclk_mhz = 810.0, + .phyclk_d18_mhz = 667.0, + .dscclk_mhz = 371.0, + .dtbclk_mhz = 600.0, + }, + { + .state = 4, + .dispclk_mhz = 1200.0, + .dppclk_mhz = 1200.0, + .phyclk_mhz = 810.0, + .phyclk_d18_mhz = 667.0, + .dscclk_mhz = 417.0, + .dtbclk_mhz = 600.0, + }, + }, + .num_states = 5, + .sr_exit_time_us = 9.0, + .sr_enter_plus_exit_time_us = 11.0, + .sr_exit_z8_time_us = 442.0, + .sr_enter_plus_exit_z8_time_us = 560.0, + .writeback_latency_us = 12.0, + .dram_channel_width_bytes = 4, + .round_trip_ping_latency_dcfclk_cycles = 106, + .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_sdp_bw_after_urgent = 80.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 65.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 30.0, + .max_avg_sdp_bw_use_normal_percent = 60.0, + .max_avg_dram_bw_use_normal_percent = 60.0, + .fabric_datapath_to_dcn_data_return_bytes = 32, + .return_bus_width_bytes = 64, + .downspread_percent = 0.38, + .dcn_downspread_percent = 0.5, + .gpuvm_min_page_size_bytes = 4096, + .hostvm_min_page_size_bytes = 4096, + .do_urgent_latency_adjustment = false, + .urgent_latency_adjustment_fabric_clock_component_us = 0, + .urgent_latency_adjustment_fabric_clock_reference_mhz = 0, +}; + + +void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params) +{ + struct clk_limit_table *clk_table = &bw_params->clk_table; + struct _vcs_dpi_voltage_scaling_st *clock_limits = + dcn3_14_soc.clock_limits; + unsigned int i, closest_clk_lvl; + int max_dispclk_mhz = 0, max_dppclk_mhz = 0; + int j; + + dc_assert_fp_enabled(); + + // Default clock levels are used for diags, which may lead to overclocking. + if (!IS_DIAG_DC(dc->ctx->dce_environment) && dc->config.use_default_clock_table == false) { + + dcn3_14_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator; + dcn3_14_ip.max_num_dpp = dc->res_pool->pipe_count; + + if (bw_params->num_channels > 0) + dcn3_14_soc.num_chans = bw_params->num_channels; + + ASSERT(dcn3_14_soc.num_chans); + ASSERT(clk_table->num_entries); + + /* Prepass to find max clocks independent of voltage level. */ + for (i = 0; i < clk_table->num_entries; ++i) { + if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz) + max_dispclk_mhz = clk_table->entries[i].dispclk_mhz; + if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz) + max_dppclk_mhz = clk_table->entries[i].dppclk_mhz; + } + + for (i = 0; i < clk_table->num_entries; i++) { + /* loop backwards*/ + for (closest_clk_lvl = 0, j = dcn3_14_soc.num_states - 1; j >= 0; j--) { + if ((unsigned int) dcn3_14_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { + closest_clk_lvl = j; + break; + } + } + if (clk_table->num_entries == 1) { + /*smu gives one DPM level, let's take the highest one*/ + closest_clk_lvl = dcn3_14_soc.num_states - 1; + } + + clock_limits[i].state = i; + + /* Clocks dependent on voltage level. */ + clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; + if (clk_table->num_entries == 1 && + clock_limits[i].dcfclk_mhz < dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) { + /*SMU fix not released yet*/ + clock_limits[i].dcfclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dcfclk_mhz; + } + clock_limits[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; + clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz; + + if (clk_table->entries[i].memclk_mhz && clk_table->entries[i].wck_ratio) + clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2 * clk_table->entries[i].wck_ratio; + + /* Clocks independent of voltage level. */ + clock_limits[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz : + dcn3_14_soc.clock_limits[closest_clk_lvl].dispclk_mhz; + + clock_limits[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz : + dcn3_14_soc.clock_limits[closest_clk_lvl].dppclk_mhz; + + clock_limits[i].dram_bw_per_chan_gbps = dcn3_14_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; + clock_limits[i].dscclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dscclk_mhz; + clock_limits[i].dtbclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; + clock_limits[i].phyclk_d18_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; + clock_limits[i].phyclk_mhz = dcn3_14_soc.clock_limits[closest_clk_lvl].phyclk_mhz; + } + for (i = 0; i < clk_table->num_entries; i++) + dcn3_14_soc.clock_limits[i] = clock_limits[i]; + if (clk_table->num_entries) { + dcn3_14_soc.num_states = clk_table->num_entries; + } + } + + if (max_dispclk_mhz) { + dcn3_14_soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2; + dc->dml.soc.dispclk_dppclk_vco_speed_mhz = max_dispclk_mhz * 2; + } + + if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) + dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31); + else + dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31_FPGA); +} + +static bool is_dual_plane(enum surface_pixel_format format) +{ + return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN || format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA; +} + +int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + bool fast_validate) +{ + int i, pipe_cnt; + struct resource_context *res_ctx = &context->res_ctx; + struct pipe_ctx *pipe; + bool upscaled = false; + + dc_assert_fp_enabled(); + + dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + + for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { + struct dc_crtc_timing *timing; + + if (!res_ctx->pipe_ctx[i].stream) + continue; + pipe = &res_ctx->pipe_ctx[i]; + timing = &pipe->stream->timing; + + if (dc_extended_blank_supported(dc) && pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min + && pipe->stream->adjust.v_total_min > timing->v_total) + pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min; + + if (pipe->plane_state && + (pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height || + pipe->plane_state->src_rect.width < pipe->plane_state->dst_rect.width)) + upscaled = true; + + /* + * Immediate flip can be set dynamically after enabling the plane. + * We need to require support for immediate flip or underflow can be + * intermittently experienced depending on peak b/w requirements. + */ + pipes[pipe_cnt].pipe.src.immediate_flip = true; + + pipes[pipe_cnt].pipe.src.unbounded_req_mode = false; + pipes[pipe_cnt].pipe.src.hostvm = dc->res_pool->hubbub->riommu_active; + pipes[pipe_cnt].pipe.src.gpuvm = true; + pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0; + pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0; + pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch; + pipes[pipe_cnt].pipe.src.dcc_rate = 3; + pipes[pipe_cnt].dout.dsc_input_bpc = 0; + + if (pipes[pipe_cnt].dout.dsc_enable) { + switch (timing->display_color_depth) { + case COLOR_DEPTH_888: + pipes[pipe_cnt].dout.dsc_input_bpc = 8; + break; + case COLOR_DEPTH_101010: + pipes[pipe_cnt].dout.dsc_input_bpc = 10; + break; + case COLOR_DEPTH_121212: + pipes[pipe_cnt].dout.dsc_input_bpc = 12; + break; + default: + ASSERT(0); + break; + } + } + + pipe_cnt++; + } + context->bw_ctx.dml.ip.det_buffer_size_kbytes = DCN3_14_DEFAULT_DET_SIZE; + + dc->config.enable_4to1MPC = false; + if (pipe_cnt == 1 && pipe->plane_state && !dc->debug.disable_z9_mpc) { + if (is_dual_plane(pipe->plane_state->format) + && pipe->plane_state->src_rect.width <= 1920 && pipe->plane_state->src_rect.height <= 1080) { + dc->config.enable_4to1MPC = true; + } else if (!is_dual_plane(pipe->plane_state->format) && pipe->plane_state->src_rect.width <= 5120) { + /* Limit to 5k max to avoid forced pipe split when there is not enough detile for swath */ + context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192; + pipes[0].pipe.src.unbounded_req_mode = true; + } + } else if (context->stream_count >= dc->debug.crb_alloc_policy_min_disp_count + && dc->debug.crb_alloc_policy > DET_SIZE_DEFAULT) { + context->bw_ctx.dml.ip.det_buffer_size_kbytes = dc->debug.crb_alloc_policy * 64; + } else if (context->stream_count >= 3 && upscaled) { + context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192; + } + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (!pipe->stream) + continue; + + if (pipe->stream->signal == SIGNAL_TYPE_EDP && dc->debug.seamless_boot_odm_combine && + pipe->stream->apply_seamless_boot_optimization) { + + if (pipe->stream->apply_boot_odm_mode == dm_odm_combine_policy_2to1) { + context->bw_ctx.dml.vba.ODMCombinePolicy = dm_odm_combine_policy_2to1; + break; + } + } + } + + return pipe_cnt; +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h new file mode 100644 index 000000000000..d32c5bb99f4c --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h @@ -0,0 +1,40 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright 2022 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 __DCN314_FPU_H__ +#define __DCN314_FPU_H__ + +#define DCN3_14_DEFAULT_DET_SIZE 384 +#define DCN3_14_MAX_DET_SIZE 384 +#define DCN3_14_MIN_COMPBUF_SIZE_KB 128 +#define DCN3_14_CRB_SEGMENT_SIZE_KB 64 + +void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params); +int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, + display_e2e_pipe_params_st *pipes, + bool fast_validate); + +#endif 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 66453546e24f..8e4c9d0887ce 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 @@ -473,8 +473,11 @@ void dcn32_set_phantom_stream_timing(struct dc *dc, // DML calculation for MALL region doesn't take into account FW delay // and required pstate allow width for multi-display cases + /* Add 16 lines margin to the MALL REGION because SUB_VP_START_LINE must be aligned + * to 2 swaths (i.e. 16 lines) + */ phantom_vactive = get_subviewport_lines_needed_in_mall(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx) + - pstate_width_fw_delay_lines; + pstate_width_fw_delay_lines + dc->caps.subvp_swath_height_margin_lines; // For backporch of phantom pipe, use vstartup of the main pipe phantom_bp = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); @@ -490,6 +493,7 @@ void dcn32_set_phantom_stream_timing(struct dc *dc, phantom_stream->timing.v_front_porch + phantom_stream->timing.v_sync_width + phantom_bp; + phantom_stream->timing.flags.DSC = 0; // Don't need DSC for phantom timing } /** @@ -983,9 +987,15 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, * DML favors voltage over p-state, but we're more interested in * supporting p-state over voltage. We can't support p-state in * prefetch mode > 0 so try capping the prefetch mode to start. + * Override present for testing. */ - context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = + if (dc->debug.dml_disallow_alternate_prefetch_modes) + context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = dm_prefetch_support_uclk_fclk_and_stutter; + else + context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = + dm_prefetch_support_uclk_fclk_and_stutter_if_possible; + *vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, *pipe_cnt); /* This may adjust vlevel and maxMpcComb */ if (*vlevel < context->bw_ctx.dml.soc.num_states) @@ -1004,6 +1014,15 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, dc->debug.force_subvp_mclk_switch)) { dcn32_merge_pipes_for_subvp(dc, context); + // to re-initialize viewport after the pipe merge + for (int i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (!pipe_ctx->plane_state || !pipe_ctx->stream) + continue; + + resource_build_scaling_params(pipe_ctx); + } while (!found_supported_config && dcn32_enough_pipes_for_subvp(dc, context) && dcn32_assign_subvp_pipe(dc, context, &dc_pipe_idx)) { @@ -1014,7 +1033,9 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, * will not allow for switch in VBLANK. The DRR display must have it's VBLANK stretched * enough to support MCLK switching. */ - if (*vlevel == context->bw_ctx.dml.soc.num_states) { + if (*vlevel == context->bw_ctx.dml.soc.num_states && + context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final == + dm_prefetch_support_uclk_fclk_and_stutter) { context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = dm_prefetch_support_stutter; /* There are params (such as FabricClock) that need to be recalculated @@ -1344,7 +1365,8 @@ bool dcn32_internal_validate_bw(struct dc *dc, int split[MAX_PIPES] = { 0 }; bool merge[MAX_PIPES] = { false }; bool newly_split[MAX_PIPES] = { false }; - int pipe_cnt, i, pipe_idx, vlevel; + int pipe_cnt, i, pipe_idx; + int vlevel = context->bw_ctx.dml.soc.num_states; struct vba_vars_st *vba = &context->bw_ctx.dml.vba; dc_assert_fp_enabled(); @@ -1373,17 +1395,22 @@ bool dcn32_internal_validate_bw(struct dc *dc, DC_FP_END(); } - if (fast_validate || vlevel == context->bw_ctx.dml.soc.num_states || - vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported) { + if (fast_validate || + (dc->debug.dml_disallow_alternate_prefetch_modes && + (vlevel == context->bw_ctx.dml.soc.num_states || + vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported))) { /* - * If mode is unsupported or there's still no p-state support then - * fall back to favoring voltage. + * If dml_disallow_alternate_prefetch_modes is false, then we have already + * tried alternate prefetch modes during full validation. + * + * If mode is unsupported or there is no p-state support, then + * fall back to favouring voltage. * - * If Prefetch mode 0 failed for this config, or passed with Max UCLK, try if - * supported with Prefetch mode 1 (dm_prefetch_support_fclk_and_stutter == 2) + * If Prefetch mode 0 failed for this config, or passed with Max UCLK, then try + * to support with Prefetch mode 1 (dm_prefetch_support_fclk_and_stutter == 2) */ context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = - dm_prefetch_support_fclk_and_stutter; + dm_prefetch_support_fclk_and_stutter; vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); @@ -2098,6 +2125,13 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; } + if ((int)(dcn3_2_soc.fclk_change_latency_us * 1000) + != dc->bb_overrides.fclk_clock_change_latency_ns + && dc->bb_overrides.fclk_clock_change_latency_ns) { + dcn3_2_soc.fclk_change_latency_us = + dc->bb_overrides.fclk_clock_change_latency_ns / 1000; + } + if ((int)(dcn3_2_soc.dummy_pstate_latency_us * 1000) != dc->bb_overrides.dummy_clock_change_latency_ns && dc->bb_overrides.dummy_clock_change_latency_ns) { diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c index 890612db08dc..cb2025771646 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c @@ -221,7 +221,6 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman // VBA_DELTA // Calculate DET size, swath height dml32_CalculateSwathAndDETConfiguration( - &v->dummy_vars.dml32_CalculateSwathAndDETConfiguration, mode_lib->vba.DETSizeOverride, mode_lib->vba.UsesMALLForPStateChange, mode_lib->vba.ConfigReturnBufferSizeInKByte, @@ -461,7 +460,6 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman { dml32_CalculateVMRowAndSwath( - &v->dummy_vars.dml32_CalculateVMRowAndSwath, mode_lib->vba.NumberOfActiveSurfaces, v->dummy_vars.DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation.SurfaceParameters, v->SurfaceSizeInMALL, @@ -757,9 +755,7 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman v->dummy_vars.DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation.myPipe.BytePerPixelY = v->BytePerPixelY[k]; v->dummy_vars.DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation.myPipe.BytePerPixelC = v->BytePerPixelC[k]; v->dummy_vars.DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation.myPipe.ProgressiveToInterlaceUnitInOPP = mode_lib->vba.ProgressiveToInterlaceUnitInOPP; - v->ErrorResult[k] = dml32_CalculatePrefetchSchedule( - &v->dummy_vars.dml32_CalculatePrefetchSchedule, - v->dummy_vars.DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation.HostVMInefficiencyFactor, + v->ErrorResult[k] = dml32_CalculatePrefetchSchedule(v->dummy_vars.DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation.HostVMInefficiencyFactor, &v->dummy_vars.DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation.myPipe, v->DSCDelay[k], mode_lib->vba.DPPCLKDelaySubtotal + mode_lib->vba.DPPCLKDelayCNVCFormater, mode_lib->vba.DPPCLKDelaySCL, @@ -1167,7 +1163,6 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman v->dummy_vars.DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation.mmSOCParameters.SMNLatency = mode_lib->vba.SMNLatency; dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( - &v->dummy_vars.dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport, mode_lib->vba.USRRetrainingRequiredFinal, mode_lib->vba.UsesMALLForPStateChange, mode_lib->vba.PrefetchModePerState[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb], @@ -1952,7 +1947,6 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l } dml32_CalculateSwathAndDETConfiguration( - &v->dummy_vars.dml32_CalculateSwathAndDETConfiguration, mode_lib->vba.DETSizeOverride, mode_lib->vba.UsesMALLForPStateChange, mode_lib->vba.ConfigReturnBufferSizeInKByte, @@ -2549,7 +2543,6 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l } dml32_CalculateSwathAndDETConfiguration( - &v->dummy_vars.dml32_CalculateSwathAndDETConfiguration, mode_lib->vba.DETSizeOverride, mode_lib->vba.UsesMALLForPStateChange, mode_lib->vba.ConfigReturnBufferSizeInKByte, @@ -2749,7 +2742,6 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l { dml32_CalculateVMRowAndSwath( - &v->dummy_vars.dml32_CalculateVMRowAndSwath, mode_lib->vba.NumberOfActiveSurfaces, v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.SurfParameters, mode_lib->vba.SurfaceSizeInMALL, @@ -3266,7 +3258,6 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.NoTimeForPrefetch[i][j][k] = dml32_CalculatePrefetchSchedule( - &v->dummy_vars.dml32_CalculatePrefetchSchedule, v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.HostVMInefficiencyFactor, &v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.myPipe, mode_lib->vba.DSCDelayPerState[i][k], @@ -3566,7 +3557,6 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l { dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( - &v->dummy_vars.dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport, mode_lib->vba.USRRetrainingRequiredFinal, mode_lib->vba.UsesMALLForPStateChange, mode_lib->vba.PrefetchModePerState[i][j], diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c index 07f8f3b8626b..05fc14a47fba 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c @@ -391,7 +391,6 @@ void dml32_CalculateBytePerPixelAndBlockSizes( } // CalculateBytePerPixelAndBlockSizes void dml32_CalculateSwathAndDETConfiguration( - struct dml32_CalculateSwathAndDETConfiguration *st_vars, unsigned int DETSizeOverride[], enum dm_use_mall_for_pstate_change_mode UseMALLForPStateChange[], unsigned int ConfigReturnBufferSizeInKByte, @@ -456,10 +455,18 @@ void dml32_CalculateSwathAndDETConfiguration( bool ViewportSizeSupportPerSurface[], bool *ViewportSizeSupport) { + unsigned int MaximumSwathHeightY[DC__NUM_DPP__MAX]; + unsigned int MaximumSwathHeightC[DC__NUM_DPP__MAX]; + unsigned int RoundedUpMaxSwathSizeBytesY[DC__NUM_DPP__MAX]; + unsigned int RoundedUpMaxSwathSizeBytesC[DC__NUM_DPP__MAX]; + unsigned int RoundedUpSwathSizeBytesY; + unsigned int RoundedUpSwathSizeBytesC; + double SwathWidthdoubleDPP[DC__NUM_DPP__MAX]; + double SwathWidthdoubleDPPChroma[DC__NUM_DPP__MAX]; unsigned int k; - - st_vars->TotalActiveDPP = 0; - st_vars->NoChromaSurfaces = true; + unsigned int TotalActiveDPP = 0; + bool NoChromaSurfaces = true; + unsigned int DETBufferSizeInKByteForSwathCalculation; #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: ForceSingleDPP = %d\n", __func__, ForceSingleDPP); @@ -494,43 +501,43 @@ void dml32_CalculateSwathAndDETConfiguration( DPPPerSurface, /* Output */ - st_vars->SwathWidthdoubleDPP, - st_vars->SwathWidthdoubleDPPChroma, + SwathWidthdoubleDPP, + SwathWidthdoubleDPPChroma, SwathWidth, SwathWidthChroma, - st_vars->MaximumSwathHeightY, - st_vars->MaximumSwathHeightC, + MaximumSwathHeightY, + MaximumSwathHeightC, swath_width_luma_ub, swath_width_chroma_ub); for (k = 0; k < NumberOfActiveSurfaces; ++k) { - st_vars->RoundedUpMaxSwathSizeBytesY[k] = swath_width_luma_ub[k] * BytePerPixDETY[k] * st_vars->MaximumSwathHeightY[k]; - st_vars->RoundedUpMaxSwathSizeBytesC[k] = swath_width_chroma_ub[k] * BytePerPixDETC[k] * st_vars->MaximumSwathHeightC[k]; + RoundedUpMaxSwathSizeBytesY[k] = swath_width_luma_ub[k] * BytePerPixDETY[k] * MaximumSwathHeightY[k]; + RoundedUpMaxSwathSizeBytesC[k] = swath_width_chroma_ub[k] * BytePerPixDETC[k] * MaximumSwathHeightC[k]; #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: k=%0d DPPPerSurface = %d\n", __func__, k, DPPPerSurface[k]); dml_print("DML::%s: k=%0d swath_width_luma_ub = %d\n", __func__, k, swath_width_luma_ub[k]); dml_print("DML::%s: k=%0d BytePerPixDETY = %f\n", __func__, k, BytePerPixDETY[k]); - dml_print("DML::%s: k=%0d MaximumSwathHeightY = %d\n", __func__, k, st_vars->MaximumSwathHeightY[k]); + dml_print("DML::%s: k=%0d MaximumSwathHeightY = %d\n", __func__, k, MaximumSwathHeightY[k]); dml_print("DML::%s: k=%0d RoundedUpMaxSwathSizeBytesY = %d\n", __func__, k, - st_vars->RoundedUpMaxSwathSizeBytesY[k]); + RoundedUpMaxSwathSizeBytesY[k]); dml_print("DML::%s: k=%0d swath_width_chroma_ub = %d\n", __func__, k, swath_width_chroma_ub[k]); dml_print("DML::%s: k=%0d BytePerPixDETC = %f\n", __func__, k, BytePerPixDETC[k]); - dml_print("DML::%s: k=%0d MaximumSwathHeightC = %d\n", __func__, k, st_vars->MaximumSwathHeightC[k]); + dml_print("DML::%s: k=%0d MaximumSwathHeightC = %d\n", __func__, k, MaximumSwathHeightC[k]); dml_print("DML::%s: k=%0d RoundedUpMaxSwathSizeBytesC = %d\n", __func__, k, - st_vars->RoundedUpMaxSwathSizeBytesC[k]); + RoundedUpMaxSwathSizeBytesC[k]); #endif if (SourcePixelFormat[k] == dm_420_10) { - st_vars->RoundedUpMaxSwathSizeBytesY[k] = dml_ceil((unsigned int) st_vars->RoundedUpMaxSwathSizeBytesY[k], 256); - st_vars->RoundedUpMaxSwathSizeBytesC[k] = dml_ceil((unsigned int) st_vars->RoundedUpMaxSwathSizeBytesC[k], 256); + RoundedUpMaxSwathSizeBytesY[k] = dml_ceil((unsigned int) RoundedUpMaxSwathSizeBytesY[k], 256); + RoundedUpMaxSwathSizeBytesC[k] = dml_ceil((unsigned int) RoundedUpMaxSwathSizeBytesC[k], 256); } } for (k = 0; k < NumberOfActiveSurfaces; ++k) { - st_vars->TotalActiveDPP = st_vars->TotalActiveDPP + (ForceSingleDPP ? 1 : DPPPerSurface[k]); + TotalActiveDPP = TotalActiveDPP + (ForceSingleDPP ? 1 : DPPPerSurface[k]); if (SourcePixelFormat[k] == dm_420_8 || SourcePixelFormat[k] == dm_420_10 || SourcePixelFormat[k] == dm_420_12 || SourcePixelFormat[k] == dm_rgbe_alpha) { - st_vars->NoChromaSurfaces = false; + NoChromaSurfaces = false; } } @@ -540,10 +547,10 @@ void dml32_CalculateSwathAndDETConfiguration( // if unbounded req is enabled, program reserved space such that the ROB will not hold more than 8 swaths worth of data // - assume worst-case compression rate of 4. [ROB size - 8 * swath_size / max_compression ratio] // - assume for "narrow" vp case in which the ROB can fit 8 swaths, the DET should be big enough to do full size req - *CompBufReservedSpaceNeedAdjustment = ((int) ROBSizeKBytes - (int) *CompBufReservedSpaceKBytes) > (int) (st_vars->RoundedUpMaxSwathSizeBytesY[0]/512); + *CompBufReservedSpaceNeedAdjustment = ((int) ROBSizeKBytes - (int) *CompBufReservedSpaceKBytes) > (int) (RoundedUpMaxSwathSizeBytesY[0]/512); if (*CompBufReservedSpaceNeedAdjustment == 1) { - *CompBufReservedSpaceKBytes = ROBSizeKBytes - st_vars->RoundedUpMaxSwathSizeBytesY[0]/512; + *CompBufReservedSpaceKBytes = ROBSizeKBytes - RoundedUpMaxSwathSizeBytesY[0]/512; } #ifdef __DML_VBA_DEBUG__ @@ -551,7 +558,7 @@ void dml32_CalculateSwathAndDETConfiguration( dml_print("DML::%s: CompBufReservedSpaceNeedAdjustment = %d\n", __func__, *CompBufReservedSpaceNeedAdjustment); #endif - *UnboundedRequestEnabled = dml32_UnboundedRequest(UseUnboundedRequestingFinal, st_vars->TotalActiveDPP, st_vars->NoChromaSurfaces, Output[0], SurfaceTiling[0], *CompBufReservedSpaceNeedAdjustment, DisableUnboundRequestIfCompBufReservedSpaceNeedAdjustment); + *UnboundedRequestEnabled = dml32_UnboundedRequest(UseUnboundedRequestingFinal, TotalActiveDPP, NoChromaSurfaces, Output[0], SurfaceTiling[0], *CompBufReservedSpaceNeedAdjustment, DisableUnboundRequestIfCompBufReservedSpaceNeedAdjustment); dml32_CalculateDETBufferSize(DETSizeOverride, UseMALLForPStateChange, @@ -566,8 +573,8 @@ void dml32_CalculateSwathAndDETConfiguration( SourcePixelFormat, ReadBandwidthLuma, ReadBandwidthChroma, - st_vars->RoundedUpMaxSwathSizeBytesY, - st_vars->RoundedUpMaxSwathSizeBytesC, + RoundedUpMaxSwathSizeBytesY, + RoundedUpMaxSwathSizeBytesC, DPPPerSurface, /* Output */ @@ -575,7 +582,7 @@ void dml32_CalculateSwathAndDETConfiguration( CompressedBufferSizeInkByte); #ifdef __DML_VBA_DEBUG__ - dml_print("DML::%s: TotalActiveDPP = %d\n", __func__, st_vars->TotalActiveDPP); + dml_print("DML::%s: TotalActiveDPP = %d\n", __func__, TotalActiveDPP); dml_print("DML::%s: nomDETInKByte = %d\n", __func__, nomDETInKByte); dml_print("DML::%s: ConfigReturnBufferSizeInKByte = %d\n", __func__, ConfigReturnBufferSizeInKByte); dml_print("DML::%s: UseUnboundedRequestingFinal = %d\n", __func__, UseUnboundedRequestingFinal); @@ -586,42 +593,42 @@ void dml32_CalculateSwathAndDETConfiguration( *ViewportSizeSupport = true; for (k = 0; k < NumberOfActiveSurfaces; ++k) { - st_vars->DETBufferSizeInKByteForSwathCalculation = (UseMALLForPStateChange[k] == + DETBufferSizeInKByteForSwathCalculation = (UseMALLForPStateChange[k] == dm_use_mall_pstate_change_phantom_pipe ? 1024 : DETBufferSizeInKByte[k]); #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: k=%0d DETBufferSizeInKByteForSwathCalculation = %d\n", __func__, k, - st_vars->DETBufferSizeInKByteForSwathCalculation); + DETBufferSizeInKByteForSwathCalculation); #endif - if (st_vars->RoundedUpMaxSwathSizeBytesY[k] + st_vars->RoundedUpMaxSwathSizeBytesC[k] <= - st_vars->DETBufferSizeInKByteForSwathCalculation * 1024 / 2) { - SwathHeightY[k] = st_vars->MaximumSwathHeightY[k]; - SwathHeightC[k] = st_vars->MaximumSwathHeightC[k]; - st_vars->RoundedUpSwathSizeBytesY = st_vars->RoundedUpMaxSwathSizeBytesY[k]; - st_vars->RoundedUpSwathSizeBytesC = st_vars->RoundedUpMaxSwathSizeBytesC[k]; - } else if (st_vars->RoundedUpMaxSwathSizeBytesY[k] >= 1.5 * st_vars->RoundedUpMaxSwathSizeBytesC[k] && - st_vars->RoundedUpMaxSwathSizeBytesY[k] / 2 + st_vars->RoundedUpMaxSwathSizeBytesC[k] <= - st_vars->DETBufferSizeInKByteForSwathCalculation * 1024 / 2) { - SwathHeightY[k] = st_vars->MaximumSwathHeightY[k] / 2; - SwathHeightC[k] = st_vars->MaximumSwathHeightC[k]; - st_vars->RoundedUpSwathSizeBytesY = st_vars->RoundedUpMaxSwathSizeBytesY[k] / 2; - st_vars->RoundedUpSwathSizeBytesC = st_vars->RoundedUpMaxSwathSizeBytesC[k]; - } else if (st_vars->RoundedUpMaxSwathSizeBytesY[k] < 1.5 * st_vars->RoundedUpMaxSwathSizeBytesC[k] && - st_vars->RoundedUpMaxSwathSizeBytesY[k] + st_vars->RoundedUpMaxSwathSizeBytesC[k] / 2 <= - st_vars->DETBufferSizeInKByteForSwathCalculation * 1024 / 2) { - SwathHeightY[k] = st_vars->MaximumSwathHeightY[k]; - SwathHeightC[k] = st_vars->MaximumSwathHeightC[k] / 2; - st_vars->RoundedUpSwathSizeBytesY = st_vars->RoundedUpMaxSwathSizeBytesY[k]; - st_vars->RoundedUpSwathSizeBytesC = st_vars->RoundedUpMaxSwathSizeBytesC[k] / 2; + if (RoundedUpMaxSwathSizeBytesY[k] + RoundedUpMaxSwathSizeBytesC[k] <= + DETBufferSizeInKByteForSwathCalculation * 1024 / 2) { + SwathHeightY[k] = MaximumSwathHeightY[k]; + SwathHeightC[k] = MaximumSwathHeightC[k]; + RoundedUpSwathSizeBytesY = RoundedUpMaxSwathSizeBytesY[k]; + RoundedUpSwathSizeBytesC = RoundedUpMaxSwathSizeBytesC[k]; + } else if (RoundedUpMaxSwathSizeBytesY[k] >= 1.5 * RoundedUpMaxSwathSizeBytesC[k] && + RoundedUpMaxSwathSizeBytesY[k] / 2 + RoundedUpMaxSwathSizeBytesC[k] <= + DETBufferSizeInKByteForSwathCalculation * 1024 / 2) { + SwathHeightY[k] = MaximumSwathHeightY[k] / 2; + SwathHeightC[k] = MaximumSwathHeightC[k]; + RoundedUpSwathSizeBytesY = RoundedUpMaxSwathSizeBytesY[k] / 2; + RoundedUpSwathSizeBytesC = RoundedUpMaxSwathSizeBytesC[k]; + } else if (RoundedUpMaxSwathSizeBytesY[k] < 1.5 * RoundedUpMaxSwathSizeBytesC[k] && + RoundedUpMaxSwathSizeBytesY[k] + RoundedUpMaxSwathSizeBytesC[k] / 2 <= + DETBufferSizeInKByteForSwathCalculation * 1024 / 2) { + SwathHeightY[k] = MaximumSwathHeightY[k]; + SwathHeightC[k] = MaximumSwathHeightC[k] / 2; + RoundedUpSwathSizeBytesY = RoundedUpMaxSwathSizeBytesY[k]; + RoundedUpSwathSizeBytesC = RoundedUpMaxSwathSizeBytesC[k] / 2; } else { - SwathHeightY[k] = st_vars->MaximumSwathHeightY[k] / 2; - SwathHeightC[k] = st_vars->MaximumSwathHeightC[k] / 2; - st_vars->RoundedUpSwathSizeBytesY = st_vars->RoundedUpMaxSwathSizeBytesY[k] / 2; - st_vars->RoundedUpSwathSizeBytesC = st_vars->RoundedUpMaxSwathSizeBytesC[k] / 2; + SwathHeightY[k] = MaximumSwathHeightY[k] / 2; + SwathHeightC[k] = MaximumSwathHeightC[k] / 2; + RoundedUpSwathSizeBytesY = RoundedUpMaxSwathSizeBytesY[k] / 2; + RoundedUpSwathSizeBytesC = RoundedUpMaxSwathSizeBytesC[k] / 2; } - if ((st_vars->RoundedUpMaxSwathSizeBytesY[k] / 2 + st_vars->RoundedUpMaxSwathSizeBytesC[k] / 2 > - st_vars->DETBufferSizeInKByteForSwathCalculation * 1024 / 2) + if ((RoundedUpMaxSwathSizeBytesY[k] / 2 + RoundedUpMaxSwathSizeBytesC[k] / 2 > + DETBufferSizeInKByteForSwathCalculation * 1024 / 2) || SwathWidth[k] > MaximumSwathWidthLuma[k] || (SwathHeightC[k] > 0 && SwathWidthChroma[k] > MaximumSwathWidthChroma[k])) { *ViewportSizeSupport = false; @@ -636,7 +643,7 @@ void dml32_CalculateSwathAndDETConfiguration( #endif DETBufferSizeY[k] = DETBufferSizeInKByte[k] * 1024; DETBufferSizeC[k] = 0; - } else if (st_vars->RoundedUpSwathSizeBytesY <= 1.5 * st_vars->RoundedUpSwathSizeBytesC) { + } else if (RoundedUpSwathSizeBytesY <= 1.5 * RoundedUpSwathSizeBytesC) { #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: k=%0d Half DET for plane0, half for plane1\n", __func__, k); #endif @@ -654,11 +661,11 @@ void dml32_CalculateSwathAndDETConfiguration( dml_print("DML::%s: k=%0d SwathHeightY = %d\n", __func__, k, SwathHeightY[k]); dml_print("DML::%s: k=%0d SwathHeightC = %d\n", __func__, k, SwathHeightC[k]); dml_print("DML::%s: k=%0d RoundedUpMaxSwathSizeBytesY = %d\n", __func__, - k, st_vars->RoundedUpMaxSwathSizeBytesY[k]); + k, RoundedUpMaxSwathSizeBytesY[k]); dml_print("DML::%s: k=%0d RoundedUpMaxSwathSizeBytesC = %d\n", __func__, - k, st_vars->RoundedUpMaxSwathSizeBytesC[k]); - dml_print("DML::%s: k=%0d RoundedUpSwathSizeBytesY = %d\n", __func__, k, st_vars->RoundedUpSwathSizeBytesY); - dml_print("DML::%s: k=%0d RoundedUpSwathSizeBytesC = %d\n", __func__, k, st_vars->RoundedUpSwathSizeBytesC); + k, RoundedUpMaxSwathSizeBytesC[k]); + dml_print("DML::%s: k=%0d RoundedUpSwathSizeBytesY = %d\n", __func__, k, RoundedUpSwathSizeBytesY); + dml_print("DML::%s: k=%0d RoundedUpSwathSizeBytesC = %d\n", __func__, k, RoundedUpSwathSizeBytesC); dml_print("DML::%s: k=%0d DETBufferSizeInKByte = %d\n", __func__, k, DETBufferSizeInKByte[k]); dml_print("DML::%s: k=%0d DETBufferSizeY = %d\n", __func__, k, DETBufferSizeY[k]); dml_print("DML::%s: k=%0d DETBufferSizeC = %d\n", __func__, k, DETBufferSizeC[k]); @@ -1867,7 +1874,6 @@ void dml32_CalculateSurfaceSizeInMall( } // CalculateSurfaceSizeInMall void dml32_CalculateVMRowAndSwath( - struct dml32_CalculateVMRowAndSwath *st_vars, unsigned int NumberOfActiveSurfaces, DmlPipe myPipe[], unsigned int SurfaceSizeInMALL[], @@ -1933,6 +1939,21 @@ void dml32_CalculateVMRowAndSwath( unsigned int BIGK_FRAGMENT_SIZE[]) { unsigned int k; + unsigned int PTEBufferSizeInRequestsForLuma[DC__NUM_DPP__MAX]; + unsigned int PTEBufferSizeInRequestsForChroma[DC__NUM_DPP__MAX]; + unsigned int PDEAndMetaPTEBytesFrameY; + unsigned int PDEAndMetaPTEBytesFrameC; + unsigned int MetaRowByteY[DC__NUM_DPP__MAX]; + unsigned int MetaRowByteC[DC__NUM_DPP__MAX]; + unsigned int PixelPTEBytesPerRowY[DC__NUM_DPP__MAX]; + unsigned int PixelPTEBytesPerRowC[DC__NUM_DPP__MAX]; + unsigned int PixelPTEBytesPerRowY_one_row_per_frame[DC__NUM_DPP__MAX]; + unsigned int PixelPTEBytesPerRowC_one_row_per_frame[DC__NUM_DPP__MAX]; + unsigned int dpte_row_width_luma_ub_one_row_per_frame[DC__NUM_DPP__MAX]; + unsigned int dpte_row_height_luma_one_row_per_frame[DC__NUM_DPP__MAX]; + unsigned int dpte_row_width_chroma_ub_one_row_per_frame[DC__NUM_DPP__MAX]; + unsigned int dpte_row_height_chroma_one_row_per_frame[DC__NUM_DPP__MAX]; + bool one_row_per_frame_fits_in_buffer[DC__NUM_DPP__MAX]; for (k = 0; k < NumberOfActiveSurfaces; ++k) { if (HostVMEnable == true) { @@ -1954,15 +1975,15 @@ void dml32_CalculateVMRowAndSwath( myPipe[k].SourcePixelFormat == dm_rgbe_alpha) { if ((myPipe[k].SourcePixelFormat == dm_420_10 || myPipe[k].SourcePixelFormat == dm_420_12) && !IsVertical(myPipe[k].SourceRotation)) { - st_vars->PTEBufferSizeInRequestsForLuma[k] = + PTEBufferSizeInRequestsForLuma[k] = (PTEBufferSizeInRequestsLuma + PTEBufferSizeInRequestsChroma) / 2; - st_vars->PTEBufferSizeInRequestsForChroma[k] = st_vars->PTEBufferSizeInRequestsForLuma[k]; + PTEBufferSizeInRequestsForChroma[k] = PTEBufferSizeInRequestsForLuma[k]; } else { - st_vars->PTEBufferSizeInRequestsForLuma[k] = PTEBufferSizeInRequestsLuma; - st_vars->PTEBufferSizeInRequestsForChroma[k] = PTEBufferSizeInRequestsChroma; + PTEBufferSizeInRequestsForLuma[k] = PTEBufferSizeInRequestsLuma; + PTEBufferSizeInRequestsForChroma[k] = PTEBufferSizeInRequestsChroma; } - st_vars->PDEAndMetaPTEBytesFrameC = dml32_CalculateVMAndRowBytes( + PDEAndMetaPTEBytesFrameC = dml32_CalculateVMAndRowBytes( myPipe[k].ViewportStationary, myPipe[k].DCCEnable, myPipe[k].DPPPerSurface, @@ -1982,21 +2003,21 @@ void dml32_CalculateVMRowAndSwath( GPUVMMaxPageTableLevels, GPUVMMinPageSizeKBytes[k], HostVMMinPageSize, - st_vars->PTEBufferSizeInRequestsForChroma[k], + PTEBufferSizeInRequestsForChroma[k], myPipe[k].PitchC, myPipe[k].DCCMetaPitchC, myPipe[k].BlockWidthC, myPipe[k].BlockHeightC, /* Output */ - &st_vars->MetaRowByteC[k], - &st_vars->PixelPTEBytesPerRowC[k], + &MetaRowByteC[k], + &PixelPTEBytesPerRowC[k], &dpte_row_width_chroma_ub[k], &dpte_row_height_chroma[k], &dpte_row_height_linear_chroma[k], - &st_vars->PixelPTEBytesPerRowC_one_row_per_frame[k], - &st_vars->dpte_row_width_chroma_ub_one_row_per_frame[k], - &st_vars->dpte_row_height_chroma_one_row_per_frame[k], + &PixelPTEBytesPerRowC_one_row_per_frame[k], + &dpte_row_width_chroma_ub_one_row_per_frame[k], + &dpte_row_height_chroma_one_row_per_frame[k], &meta_req_width_chroma[k], &meta_req_height_chroma[k], &meta_row_width_chroma[k], @@ -2024,19 +2045,19 @@ void dml32_CalculateVMRowAndSwath( &VInitPreFillC[k], &MaxNumSwathC[k]); } else { - st_vars->PTEBufferSizeInRequestsForLuma[k] = PTEBufferSizeInRequestsLuma + PTEBufferSizeInRequestsChroma; - st_vars->PTEBufferSizeInRequestsForChroma[k] = 0; - st_vars->PixelPTEBytesPerRowC[k] = 0; - st_vars->PDEAndMetaPTEBytesFrameC = 0; - st_vars->MetaRowByteC[k] = 0; + PTEBufferSizeInRequestsForLuma[k] = PTEBufferSizeInRequestsLuma + PTEBufferSizeInRequestsChroma; + PTEBufferSizeInRequestsForChroma[k] = 0; + PixelPTEBytesPerRowC[k] = 0; + PDEAndMetaPTEBytesFrameC = 0; + MetaRowByteC[k] = 0; MaxNumSwathC[k] = 0; PrefetchSourceLinesC[k] = 0; - st_vars->dpte_row_height_chroma_one_row_per_frame[k] = 0; - st_vars->dpte_row_width_chroma_ub_one_row_per_frame[k] = 0; - st_vars->PixelPTEBytesPerRowC_one_row_per_frame[k] = 0; + dpte_row_height_chroma_one_row_per_frame[k] = 0; + dpte_row_width_chroma_ub_one_row_per_frame[k] = 0; + PixelPTEBytesPerRowC_one_row_per_frame[k] = 0; } - st_vars->PDEAndMetaPTEBytesFrameY = dml32_CalculateVMAndRowBytes( + PDEAndMetaPTEBytesFrameY = dml32_CalculateVMAndRowBytes( myPipe[k].ViewportStationary, myPipe[k].DCCEnable, myPipe[k].DPPPerSurface, @@ -2056,21 +2077,21 @@ void dml32_CalculateVMRowAndSwath( GPUVMMaxPageTableLevels, GPUVMMinPageSizeKBytes[k], HostVMMinPageSize, - st_vars->PTEBufferSizeInRequestsForLuma[k], + PTEBufferSizeInRequestsForLuma[k], myPipe[k].PitchY, myPipe[k].DCCMetaPitchY, myPipe[k].BlockWidthY, myPipe[k].BlockHeightY, /* Output */ - &st_vars->MetaRowByteY[k], - &st_vars->PixelPTEBytesPerRowY[k], + &MetaRowByteY[k], + &PixelPTEBytesPerRowY[k], &dpte_row_width_luma_ub[k], &dpte_row_height_luma[k], &dpte_row_height_linear_luma[k], - &st_vars->PixelPTEBytesPerRowY_one_row_per_frame[k], - &st_vars->dpte_row_width_luma_ub_one_row_per_frame[k], - &st_vars->dpte_row_height_luma_one_row_per_frame[k], + &PixelPTEBytesPerRowY_one_row_per_frame[k], + &dpte_row_width_luma_ub_one_row_per_frame[k], + &dpte_row_height_luma_one_row_per_frame[k], &meta_req_width[k], &meta_req_height[k], &meta_row_width[k], @@ -2098,19 +2119,19 @@ void dml32_CalculateVMRowAndSwath( &VInitPreFillY[k], &MaxNumSwathY[k]); - PDEAndMetaPTEBytesFrame[k] = st_vars->PDEAndMetaPTEBytesFrameY + st_vars->PDEAndMetaPTEBytesFrameC; - MetaRowByte[k] = st_vars->MetaRowByteY[k] + st_vars->MetaRowByteC[k]; + PDEAndMetaPTEBytesFrame[k] = PDEAndMetaPTEBytesFrameY + PDEAndMetaPTEBytesFrameC; + MetaRowByte[k] = MetaRowByteY[k] + MetaRowByteC[k]; - if (st_vars->PixelPTEBytesPerRowY[k] <= 64 * st_vars->PTEBufferSizeInRequestsForLuma[k] && - st_vars->PixelPTEBytesPerRowC[k] <= 64 * st_vars->PTEBufferSizeInRequestsForChroma[k]) { + if (PixelPTEBytesPerRowY[k] <= 64 * PTEBufferSizeInRequestsForLuma[k] && + PixelPTEBytesPerRowC[k] <= 64 * PTEBufferSizeInRequestsForChroma[k]) { PTEBufferSizeNotExceeded[k] = true; } else { PTEBufferSizeNotExceeded[k] = false; } - st_vars->one_row_per_frame_fits_in_buffer[k] = (st_vars->PixelPTEBytesPerRowY_one_row_per_frame[k] <= 64 * 2 * - st_vars->PTEBufferSizeInRequestsForLuma[k] && - st_vars->PixelPTEBytesPerRowC_one_row_per_frame[k] <= 64 * 2 * st_vars->PTEBufferSizeInRequestsForChroma[k]); + one_row_per_frame_fits_in_buffer[k] = (PixelPTEBytesPerRowY_one_row_per_frame[k] <= 64 * 2 * + PTEBufferSizeInRequestsForLuma[k] && + PixelPTEBytesPerRowC_one_row_per_frame[k] <= 64 * 2 * PTEBufferSizeInRequestsForChroma[k]); } dml32_CalculateMALLUseForStaticScreen( @@ -2118,7 +2139,7 @@ void dml32_CalculateVMRowAndSwath( MALLAllocatedForDCN, UseMALLForStaticScreen, // mode SurfaceSizeInMALL, - st_vars->one_row_per_frame_fits_in_buffer, + one_row_per_frame_fits_in_buffer, /* Output */ UsesMALLForStaticScreen); // boolen @@ -2144,13 +2165,13 @@ void dml32_CalculateVMRowAndSwath( !(UseMALLForPStateChange[k] == dm_use_mall_pstate_change_full_frame); if (use_one_row_for_frame[k]) { - dpte_row_height_luma[k] = st_vars->dpte_row_height_luma_one_row_per_frame[k]; - dpte_row_width_luma_ub[k] = st_vars->dpte_row_width_luma_ub_one_row_per_frame[k]; - st_vars->PixelPTEBytesPerRowY[k] = st_vars->PixelPTEBytesPerRowY_one_row_per_frame[k]; - dpte_row_height_chroma[k] = st_vars->dpte_row_height_chroma_one_row_per_frame[k]; - dpte_row_width_chroma_ub[k] = st_vars->dpte_row_width_chroma_ub_one_row_per_frame[k]; - st_vars->PixelPTEBytesPerRowC[k] = st_vars->PixelPTEBytesPerRowC_one_row_per_frame[k]; - PTEBufferSizeNotExceeded[k] = st_vars->one_row_per_frame_fits_in_buffer[k]; + dpte_row_height_luma[k] = dpte_row_height_luma_one_row_per_frame[k]; + dpte_row_width_luma_ub[k] = dpte_row_width_luma_ub_one_row_per_frame[k]; + PixelPTEBytesPerRowY[k] = PixelPTEBytesPerRowY_one_row_per_frame[k]; + dpte_row_height_chroma[k] = dpte_row_height_chroma_one_row_per_frame[k]; + dpte_row_width_chroma_ub[k] = dpte_row_width_chroma_ub_one_row_per_frame[k]; + PixelPTEBytesPerRowC[k] = PixelPTEBytesPerRowC_one_row_per_frame[k]; + PTEBufferSizeNotExceeded[k] = one_row_per_frame_fits_in_buffer[k]; } if (MetaRowByte[k] <= DCCMetaBufferSizeBytes) @@ -2158,7 +2179,7 @@ void dml32_CalculateVMRowAndSwath( else DCCMetaBufferSizeNotExceeded[k] = false; - PixelPTEBytesPerRow[k] = st_vars->PixelPTEBytesPerRowY[k] + st_vars->PixelPTEBytesPerRowC[k]; + PixelPTEBytesPerRow[k] = PixelPTEBytesPerRowY[k] + PixelPTEBytesPerRowC[k]; if (use_one_row_for_frame[k]) PixelPTEBytesPerRow[k] = PixelPTEBytesPerRow[k] / 2; @@ -2169,11 +2190,11 @@ void dml32_CalculateVMRowAndSwath( myPipe[k].VRatioChroma, myPipe[k].DCCEnable, myPipe[k].HTotal / myPipe[k].PixelClock, - st_vars->MetaRowByteY[k], st_vars->MetaRowByteC[k], + MetaRowByteY[k], MetaRowByteC[k], meta_row_height[k], meta_row_height_chroma[k], - st_vars->PixelPTEBytesPerRowY[k], - st_vars->PixelPTEBytesPerRowC[k], + PixelPTEBytesPerRowY[k], + PixelPTEBytesPerRowC[k], dpte_row_height_luma[k], dpte_row_height_chroma[k], @@ -2189,12 +2210,12 @@ void dml32_CalculateVMRowAndSwath( dml_print("DML::%s: k=%d, dpte_row_height_luma = %d\n", __func__, k, dpte_row_height_luma[k]); dml_print("DML::%s: k=%d, dpte_row_width_luma_ub = %d\n", __func__, k, dpte_row_width_luma_ub[k]); - dml_print("DML::%s: k=%d, PixelPTEBytesPerRowY = %d\n", __func__, k, st_vars->PixelPTEBytesPerRowY[k]); + dml_print("DML::%s: k=%d, PixelPTEBytesPerRowY = %d\n", __func__, k, PixelPTEBytesPerRowY[k]); dml_print("DML::%s: k=%d, dpte_row_height_chroma = %d\n", __func__, k, dpte_row_height_chroma[k]); dml_print("DML::%s: k=%d, dpte_row_width_chroma_ub = %d\n", __func__, k, dpte_row_width_chroma_ub[k]); - dml_print("DML::%s: k=%d, PixelPTEBytesPerRowC = %d\n", __func__, k, st_vars->PixelPTEBytesPerRowC[k]); + dml_print("DML::%s: k=%d, PixelPTEBytesPerRowC = %d\n", __func__, k, PixelPTEBytesPerRowC[k]); dml_print("DML::%s: k=%d, PixelPTEBytesPerRow = %d\n", __func__, k, PixelPTEBytesPerRow[k]); dml_print("DML::%s: k=%d, PTEBufferSizeNotExceeded = %d\n", __func__, k, PTEBufferSizeNotExceeded[k]); @@ -3342,7 +3363,6 @@ double dml32_CalculateExtraLatency( } // CalculateExtraLatency bool dml32_CalculatePrefetchSchedule( - struct dml32_CalculatePrefetchSchedule *st_vars, double HostVMInefficiencyFactor, DmlPipe *myPipe, unsigned int DSCDelay, @@ -3406,18 +3426,45 @@ bool dml32_CalculatePrefetchSchedule( double *VReadyOffsetPix) { bool MyError = false; - - st_vars->TimeForFetchingMetaPTE = 0; - st_vars->TimeForFetchingRowInVBlank = 0; - st_vars->LinesToRequestPrefetchPixelData = 0; - st_vars->max_vratio_pre = __DML_MAX_VRATIO_PRE__; - st_vars->Tsw_est1 = 0; - st_vars->Tsw_est3 = 0; + unsigned int DPPCycles, DISPCLKCycles; + double DSTTotalPixelsAfterScaler; + double LineTime; + double dst_y_prefetch_equ; + double prefetch_bw_oto; + double Tvm_oto; + double Tr0_oto; + double Tvm_oto_lines; + double Tr0_oto_lines; + double dst_y_prefetch_oto; + double TimeForFetchingMetaPTE = 0; + double TimeForFetchingRowInVBlank = 0; + double LinesToRequestPrefetchPixelData = 0; + unsigned int HostVMDynamicLevelsTrips; + double trip_to_mem; + double Tvm_trips; + double Tr0_trips; + double Tvm_trips_rounded; + double Tr0_trips_rounded; + double Lsw_oto; + double Tpre_rounded; + double prefetch_bw_equ; + double Tvm_equ; + double Tr0_equ; + double Tdmbf; + double Tdmec; + double Tdmsks; + double prefetch_sw_bytes; + double bytes_pp; + double dep_bytes; + unsigned int max_vratio_pre = __DML_MAX_VRATIO_PRE__; + double min_Lsw; + double Tsw_est1 = 0; + double Tsw_est3 = 0; if (GPUVMEnable == true && HostVMEnable == true) - st_vars->HostVMDynamicLevelsTrips = HostVMMaxNonCachedPageTableLevels; + HostVMDynamicLevelsTrips = HostVMMaxNonCachedPageTableLevels; else - st_vars->HostVMDynamicLevelsTrips = 0; + HostVMDynamicLevelsTrips = 0; #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: GPUVMEnable = %d\n", __func__, GPUVMEnable); dml_print("DML::%s: GPUVMPageTableLevels = %d\n", __func__, GPUVMPageTableLevels); @@ -3440,19 +3487,19 @@ bool dml32_CalculatePrefetchSchedule( TSetup, /* output */ - &st_vars->Tdmbf, - &st_vars->Tdmec, - &st_vars->Tdmsks, + &Tdmbf, + &Tdmec, + &Tdmsks, VUpdateOffsetPix, VUpdateWidthPix, VReadyOffsetPix); - st_vars->LineTime = myPipe->HTotal / myPipe->PixelClock; - st_vars->trip_to_mem = UrgentLatency; - st_vars->Tvm_trips = UrgentExtraLatency + st_vars->trip_to_mem * (GPUVMPageTableLevels * (st_vars->HostVMDynamicLevelsTrips + 1) - 1); + LineTime = myPipe->HTotal / myPipe->PixelClock; + trip_to_mem = UrgentLatency; + Tvm_trips = UrgentExtraLatency + trip_to_mem * (GPUVMPageTableLevels * (HostVMDynamicLevelsTrips + 1) - 1); if (DynamicMetadataVMEnabled == true) - *Tdmdl = TWait + st_vars->Tvm_trips + st_vars->trip_to_mem; + *Tdmdl = TWait + Tvm_trips + trip_to_mem; else *Tdmdl = TWait + UrgentExtraLatency; @@ -3462,15 +3509,15 @@ bool dml32_CalculatePrefetchSchedule( #endif if (DynamicMetadataEnable == true) { - if (VStartup * st_vars->LineTime < *TSetup + *Tdmdl + st_vars->Tdmbf + st_vars->Tdmec + st_vars->Tdmsks) { + if (VStartup * LineTime < *TSetup + *Tdmdl + Tdmbf + Tdmec + Tdmsks) { *NotEnoughTimeForDynamicMetadata = true; #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: Not Enough Time for Dynamic Meta!\n", __func__); dml_print("DML::%s: Tdmbf: %fus - time for dmd transfer from dchub to dio output buffer\n", - __func__, st_vars->Tdmbf); - dml_print("DML::%s: Tdmec: %fus - time dio takes to transfer dmd\n", __func__, st_vars->Tdmec); + __func__, Tdmbf); + dml_print("DML::%s: Tdmec: %fus - time dio takes to transfer dmd\n", __func__, Tdmec); dml_print("DML::%s: Tdmsks: %fus - time before active dmd must complete transmission at dio\n", - __func__, st_vars->Tdmsks); + __func__, Tdmsks); dml_print("DML::%s: Tdmdl: %fus - time for fabric to become ready and fetch dmd\n", __func__, *Tdmdl); #endif @@ -3482,21 +3529,21 @@ bool dml32_CalculatePrefetchSchedule( } *Tdmdl_vm = (DynamicMetadataEnable == true && DynamicMetadataVMEnabled == true && - GPUVMEnable == true ? TWait + st_vars->Tvm_trips : 0); + GPUVMEnable == true ? TWait + Tvm_trips : 0); if (myPipe->ScalerEnabled) - st_vars->DPPCycles = DPPCLKDelaySubtotalPlusCNVCFormater + DPPCLKDelaySCL; + DPPCycles = DPPCLKDelaySubtotalPlusCNVCFormater + DPPCLKDelaySCL; else - st_vars->DPPCycles = DPPCLKDelaySubtotalPlusCNVCFormater + DPPCLKDelaySCLLBOnly; + DPPCycles = DPPCLKDelaySubtotalPlusCNVCFormater + DPPCLKDelaySCLLBOnly; - st_vars->DPPCycles = st_vars->DPPCycles + myPipe->NumberOfCursors * DPPCLKDelayCNVCCursor; + DPPCycles = DPPCycles + myPipe->NumberOfCursors * DPPCLKDelayCNVCCursor; - st_vars->DISPCLKCycles = DISPCLKDelaySubtotal; + DISPCLKCycles = DISPCLKDelaySubtotal; if (myPipe->Dppclk == 0.0 || myPipe->Dispclk == 0.0) return true; - *DSTXAfterScaler = st_vars->DPPCycles * myPipe->PixelClock / myPipe->Dppclk + st_vars->DISPCLKCycles * + *DSTXAfterScaler = DPPCycles * myPipe->PixelClock / myPipe->Dppclk + DISPCLKCycles * myPipe->PixelClock / myPipe->Dispclk + DSCDelay; *DSTXAfterScaler = *DSTXAfterScaler + (myPipe->ODMMode != dm_odm_combine_mode_disabled ? 18 : 0) @@ -3506,10 +3553,10 @@ bool dml32_CalculatePrefetchSchedule( + ((myPipe->ODMMode == dm_odm_mode_mso_1to4) ? myPipe->HActive * 3 / 4 : 0); #ifdef __DML_VBA_DEBUG__ - dml_print("DML::%s: DPPCycles: %d\n", __func__, st_vars->DPPCycles); + dml_print("DML::%s: DPPCycles: %d\n", __func__, DPPCycles); dml_print("DML::%s: PixelClock: %f\n", __func__, myPipe->PixelClock); dml_print("DML::%s: Dppclk: %f\n", __func__, myPipe->Dppclk); - dml_print("DML::%s: DISPCLKCycles: %d\n", __func__, st_vars->DISPCLKCycles); + dml_print("DML::%s: DISPCLKCycles: %d\n", __func__, DISPCLKCycles); dml_print("DML::%s: DISPCLK: %f\n", __func__, myPipe->Dispclk); dml_print("DML::%s: DSCDelay: %d\n", __func__, DSCDelay); dml_print("DML::%s: ODMMode: %d\n", __func__, myPipe->ODMMode); @@ -3522,9 +3569,9 @@ bool dml32_CalculatePrefetchSchedule( else *DSTYAfterScaler = 0; - st_vars->DSTTotalPixelsAfterScaler = *DSTYAfterScaler * myPipe->HTotal + *DSTXAfterScaler; - *DSTYAfterScaler = dml_floor(st_vars->DSTTotalPixelsAfterScaler / myPipe->HTotal, 1); - *DSTXAfterScaler = st_vars->DSTTotalPixelsAfterScaler - ((double) (*DSTYAfterScaler * myPipe->HTotal)); + DSTTotalPixelsAfterScaler = *DSTYAfterScaler * myPipe->HTotal + *DSTXAfterScaler; + *DSTYAfterScaler = dml_floor(DSTTotalPixelsAfterScaler / myPipe->HTotal, 1); + *DSTXAfterScaler = DSTTotalPixelsAfterScaler - ((double) (*DSTYAfterScaler * myPipe->HTotal)); #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: DSTXAfterScaler: %d (final)\n", __func__, *DSTXAfterScaler); dml_print("DML::%s: DSTYAfterScaler: %d (final)\n", __func__, *DSTYAfterScaler); @@ -3532,132 +3579,132 @@ bool dml32_CalculatePrefetchSchedule( MyError = false; - st_vars->Tr0_trips = st_vars->trip_to_mem * (st_vars->HostVMDynamicLevelsTrips + 1); + Tr0_trips = trip_to_mem * (HostVMDynamicLevelsTrips + 1); if (GPUVMEnable == true) { - st_vars->Tvm_trips_rounded = dml_ceil(4.0 * st_vars->Tvm_trips / st_vars->LineTime, 1.0) / 4.0 * st_vars->LineTime; - st_vars->Tr0_trips_rounded = dml_ceil(4.0 * st_vars->Tr0_trips / st_vars->LineTime, 1.0) / 4.0 * st_vars->LineTime; + Tvm_trips_rounded = dml_ceil(4.0 * Tvm_trips / LineTime, 1.0) / 4.0 * LineTime; + Tr0_trips_rounded = dml_ceil(4.0 * Tr0_trips / LineTime, 1.0) / 4.0 * LineTime; if (GPUVMPageTableLevels >= 3) { - *Tno_bw = UrgentExtraLatency + st_vars->trip_to_mem * - (double) ((GPUVMPageTableLevels - 2) * (st_vars->HostVMDynamicLevelsTrips + 1) - 1); + *Tno_bw = UrgentExtraLatency + trip_to_mem * + (double) ((GPUVMPageTableLevels - 2) * (HostVMDynamicLevelsTrips + 1) - 1); } else if (GPUVMPageTableLevels == 1 && myPipe->DCCEnable != true) { - st_vars->Tr0_trips_rounded = dml_ceil(4.0 * UrgentExtraLatency / st_vars->LineTime, 1.0) / - 4.0 * st_vars->LineTime; // VBA_ERROR + Tr0_trips_rounded = dml_ceil(4.0 * UrgentExtraLatency / LineTime, 1.0) / + 4.0 * LineTime; // VBA_ERROR *Tno_bw = UrgentExtraLatency; } else { *Tno_bw = 0; } } else if (myPipe->DCCEnable == true) { - st_vars->Tvm_trips_rounded = st_vars->LineTime / 4.0; - st_vars->Tr0_trips_rounded = dml_ceil(4.0 * st_vars->Tr0_trips / st_vars->LineTime, 1.0) / 4.0 * st_vars->LineTime; + Tvm_trips_rounded = LineTime / 4.0; + Tr0_trips_rounded = dml_ceil(4.0 * Tr0_trips / LineTime, 1.0) / 4.0 * LineTime; *Tno_bw = 0; } else { - st_vars->Tvm_trips_rounded = st_vars->LineTime / 4.0; - st_vars->Tr0_trips_rounded = st_vars->LineTime / 2.0; + Tvm_trips_rounded = LineTime / 4.0; + Tr0_trips_rounded = LineTime / 2.0; *Tno_bw = 0; } - st_vars->Tvm_trips_rounded = dml_max(st_vars->Tvm_trips_rounded, st_vars->LineTime / 4.0); - st_vars->Tr0_trips_rounded = dml_max(st_vars->Tr0_trips_rounded, st_vars->LineTime / 4.0); + Tvm_trips_rounded = dml_max(Tvm_trips_rounded, LineTime / 4.0); + Tr0_trips_rounded = dml_max(Tr0_trips_rounded, LineTime / 4.0); if (myPipe->SourcePixelFormat == dm_420_8 || myPipe->SourcePixelFormat == dm_420_10 || myPipe->SourcePixelFormat == dm_420_12) { - st_vars->bytes_pp = myPipe->BytePerPixelY + myPipe->BytePerPixelC / 4; + bytes_pp = myPipe->BytePerPixelY + myPipe->BytePerPixelC / 4; } else { - st_vars->bytes_pp = myPipe->BytePerPixelY + myPipe->BytePerPixelC; + bytes_pp = myPipe->BytePerPixelY + myPipe->BytePerPixelC; } - st_vars->prefetch_sw_bytes = PrefetchSourceLinesY * swath_width_luma_ub * myPipe->BytePerPixelY + prefetch_sw_bytes = PrefetchSourceLinesY * swath_width_luma_ub * myPipe->BytePerPixelY + PrefetchSourceLinesC * swath_width_chroma_ub * myPipe->BytePerPixelC; - st_vars->prefetch_bw_oto = dml_max(st_vars->bytes_pp * myPipe->PixelClock / myPipe->DPPPerSurface, - st_vars->prefetch_sw_bytes / (dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * st_vars->LineTime)); + prefetch_bw_oto = dml_max(bytes_pp * myPipe->PixelClock / myPipe->DPPPerSurface, + prefetch_sw_bytes / (dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime)); - st_vars->min_Lsw = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) / st_vars->max_vratio_pre; - st_vars->min_Lsw = dml_max(st_vars->min_Lsw, 1.0); - st_vars->Lsw_oto = dml_ceil(4.0 * dml_max(st_vars->prefetch_sw_bytes / st_vars->prefetch_bw_oto / st_vars->LineTime, st_vars->min_Lsw), 1.0) / 4.0; + min_Lsw = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) / max_vratio_pre; + min_Lsw = dml_max(min_Lsw, 1.0); + Lsw_oto = dml_ceil(4.0 * dml_max(prefetch_sw_bytes / prefetch_bw_oto / LineTime, min_Lsw), 1.0) / 4.0; if (GPUVMEnable == true) { - st_vars->Tvm_oto = dml_max3( - st_vars->Tvm_trips, - *Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / st_vars->prefetch_bw_oto, - st_vars->LineTime / 4.0); + Tvm_oto = dml_max3( + Tvm_trips, + *Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / prefetch_bw_oto, + LineTime / 4.0); } else - st_vars->Tvm_oto = st_vars->LineTime / 4.0; + Tvm_oto = LineTime / 4.0; if ((GPUVMEnable == true || myPipe->DCCEnable == true)) { - st_vars->Tr0_oto = dml_max4( - st_vars->Tr0_trips, - (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / st_vars->prefetch_bw_oto, - (st_vars->LineTime - st_vars->Tvm_oto)/2.0, - st_vars->LineTime / 4.0); + Tr0_oto = dml_max4( + Tr0_trips, + (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / prefetch_bw_oto, + (LineTime - Tvm_oto)/2.0, + LineTime / 4.0); #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: Tr0_oto max0 = %f\n", __func__, - (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / st_vars->prefetch_bw_oto); - dml_print("DML::%s: Tr0_oto max1 = %f\n", __func__, st_vars->Tr0_trips); - dml_print("DML::%s: Tr0_oto max2 = %f\n", __func__, st_vars->LineTime - st_vars->Tvm_oto); - dml_print("DML::%s: Tr0_oto max3 = %f\n", __func__, st_vars->LineTime / 4); + (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / prefetch_bw_oto); + dml_print("DML::%s: Tr0_oto max1 = %f\n", __func__, Tr0_trips); + dml_print("DML::%s: Tr0_oto max2 = %f\n", __func__, LineTime - Tvm_oto); + dml_print("DML::%s: Tr0_oto max3 = %f\n", __func__, LineTime / 4); #endif } else - st_vars->Tr0_oto = (st_vars->LineTime - st_vars->Tvm_oto) / 2.0; + Tr0_oto = (LineTime - Tvm_oto) / 2.0; - st_vars->Tvm_oto_lines = dml_ceil(4.0 * st_vars->Tvm_oto / st_vars->LineTime, 1) / 4.0; - st_vars->Tr0_oto_lines = dml_ceil(4.0 * st_vars->Tr0_oto / st_vars->LineTime, 1) / 4.0; - st_vars->dst_y_prefetch_oto = st_vars->Tvm_oto_lines + 2 * st_vars->Tr0_oto_lines + st_vars->Lsw_oto; + Tvm_oto_lines = dml_ceil(4.0 * Tvm_oto / LineTime, 1) / 4.0; + Tr0_oto_lines = dml_ceil(4.0 * Tr0_oto / LineTime, 1) / 4.0; + dst_y_prefetch_oto = Tvm_oto_lines + 2 * Tr0_oto_lines + Lsw_oto; - st_vars->dst_y_prefetch_equ = VStartup - (*TSetup + dml_max(TWait + TCalc, *Tdmdl)) / st_vars->LineTime - + dst_y_prefetch_equ = VStartup - (*TSetup + dml_max(TWait + TCalc, *Tdmdl)) / LineTime - (*DSTYAfterScaler + (double) *DSTXAfterScaler / (double) myPipe->HTotal); #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: HTotal = %d\n", __func__, myPipe->HTotal); - dml_print("DML::%s: min_Lsw = %f\n", __func__, st_vars->min_Lsw); + dml_print("DML::%s: min_Lsw = %f\n", __func__, min_Lsw); dml_print("DML::%s: *Tno_bw = %f\n", __func__, *Tno_bw); dml_print("DML::%s: UrgentExtraLatency = %f\n", __func__, UrgentExtraLatency); - dml_print("DML::%s: trip_to_mem = %f\n", __func__, st_vars->trip_to_mem); + dml_print("DML::%s: trip_to_mem = %f\n", __func__, trip_to_mem); dml_print("DML::%s: BytePerPixelY = %d\n", __func__, myPipe->BytePerPixelY); dml_print("DML::%s: PrefetchSourceLinesY = %f\n", __func__, PrefetchSourceLinesY); dml_print("DML::%s: swath_width_luma_ub = %d\n", __func__, swath_width_luma_ub); dml_print("DML::%s: BytePerPixelC = %d\n", __func__, myPipe->BytePerPixelC); dml_print("DML::%s: PrefetchSourceLinesC = %f\n", __func__, PrefetchSourceLinesC); dml_print("DML::%s: swath_width_chroma_ub = %d\n", __func__, swath_width_chroma_ub); - dml_print("DML::%s: prefetch_sw_bytes = %f\n", __func__, st_vars->prefetch_sw_bytes); - dml_print("DML::%s: bytes_pp = %f\n", __func__, st_vars->bytes_pp); + dml_print("DML::%s: prefetch_sw_bytes = %f\n", __func__, prefetch_sw_bytes); + dml_print("DML::%s: bytes_pp = %f\n", __func__, bytes_pp); dml_print("DML::%s: PDEAndMetaPTEBytesFrame = %d\n", __func__, PDEAndMetaPTEBytesFrame); dml_print("DML::%s: MetaRowByte = %d\n", __func__, MetaRowByte); dml_print("DML::%s: PixelPTEBytesPerRow = %d\n", __func__, PixelPTEBytesPerRow); dml_print("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, HostVMInefficiencyFactor); - dml_print("DML::%s: Tvm_trips = %f\n", __func__, st_vars->Tvm_trips); - dml_print("DML::%s: Tr0_trips = %f\n", __func__, st_vars->Tr0_trips); - dml_print("DML::%s: prefetch_bw_oto = %f\n", __func__, st_vars->prefetch_bw_oto); - dml_print("DML::%s: Tr0_oto = %f\n", __func__, st_vars->Tr0_oto); - dml_print("DML::%s: Tvm_oto = %f\n", __func__, st_vars->Tvm_oto); - dml_print("DML::%s: Tvm_oto_lines = %f\n", __func__, st_vars->Tvm_oto_lines); - dml_print("DML::%s: Tr0_oto_lines = %f\n", __func__, st_vars->Tr0_oto_lines); - dml_print("DML::%s: Lsw_oto = %f\n", __func__, st_vars->Lsw_oto); - dml_print("DML::%s: dst_y_prefetch_oto = %f\n", __func__, st_vars->dst_y_prefetch_oto); - dml_print("DML::%s: dst_y_prefetch_equ = %f\n", __func__, st_vars->dst_y_prefetch_equ); + dml_print("DML::%s: Tvm_trips = %f\n", __func__, Tvm_trips); + dml_print("DML::%s: Tr0_trips = %f\n", __func__, Tr0_trips); + dml_print("DML::%s: prefetch_bw_oto = %f\n", __func__, prefetch_bw_oto); + dml_print("DML::%s: Tr0_oto = %f\n", __func__, Tr0_oto); + dml_print("DML::%s: Tvm_oto = %f\n", __func__, Tvm_oto); + dml_print("DML::%s: Tvm_oto_lines = %f\n", __func__, Tvm_oto_lines); + dml_print("DML::%s: Tr0_oto_lines = %f\n", __func__, Tr0_oto_lines); + dml_print("DML::%s: Lsw_oto = %f\n", __func__, Lsw_oto); + dml_print("DML::%s: dst_y_prefetch_oto = %f\n", __func__, dst_y_prefetch_oto); + dml_print("DML::%s: dst_y_prefetch_equ = %f\n", __func__, dst_y_prefetch_equ); #endif - st_vars->dst_y_prefetch_equ = dml_floor(4.0 * (st_vars->dst_y_prefetch_equ + 0.125), 1) / 4.0; - st_vars->Tpre_rounded = st_vars->dst_y_prefetch_equ * st_vars->LineTime; + dst_y_prefetch_equ = dml_floor(4.0 * (dst_y_prefetch_equ + 0.125), 1) / 4.0; + Tpre_rounded = dst_y_prefetch_equ * LineTime; #ifdef __DML_VBA_DEBUG__ - dml_print("DML::%s: dst_y_prefetch_equ: %f (after round)\n", __func__, st_vars->dst_y_prefetch_equ); - dml_print("DML::%s: LineTime: %f\n", __func__, st_vars->LineTime); + dml_print("DML::%s: dst_y_prefetch_equ: %f (after round)\n", __func__, dst_y_prefetch_equ); + dml_print("DML::%s: LineTime: %f\n", __func__, LineTime); dml_print("DML::%s: VStartup: %d\n", __func__, VStartup); dml_print("DML::%s: Tvstartup: %fus - time between vstartup and first pixel of active\n", - __func__, VStartup * st_vars->LineTime); + __func__, VStartup * LineTime); dml_print("DML::%s: TSetup: %fus - time from vstartup to vready\n", __func__, *TSetup); dml_print("DML::%s: TCalc: %fus - time for calculations in dchub starting at vready\n", __func__, TCalc); - dml_print("DML::%s: Tdmbf: %fus - time for dmd transfer from dchub to dio output buffer\n", __func__, st_vars->Tdmbf); - dml_print("DML::%s: Tdmec: %fus - time dio takes to transfer dmd\n", __func__, st_vars->Tdmec); + dml_print("DML::%s: Tdmbf: %fus - time for dmd transfer from dchub to dio output buffer\n", __func__, Tdmbf); + dml_print("DML::%s: Tdmec: %fus - time dio takes to transfer dmd\n", __func__, Tdmec); dml_print("DML::%s: Tdmdl_vm: %fus - time for vm stages of dmd\n", __func__, *Tdmdl_vm); dml_print("DML::%s: Tdmdl: %fus - time for fabric to become ready and fetch dmd\n", __func__, *Tdmdl); dml_print("DML::%s: DSTYAfterScaler: %d lines - number of lines of pipeline and buffer delay after scaler\n", __func__, *DSTYAfterScaler); #endif - st_vars->dep_bytes = dml_max(PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor, + dep_bytes = dml_max(PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor, MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor); - if (st_vars->prefetch_sw_bytes < st_vars->dep_bytes) - st_vars->prefetch_sw_bytes = 2 * st_vars->dep_bytes; + if (prefetch_sw_bytes < dep_bytes) + prefetch_sw_bytes = 2 * dep_bytes; *PrefetchBandwidth = 0; *DestinationLinesToRequestVMInVBlank = 0; @@ -3665,61 +3712,61 @@ bool dml32_CalculatePrefetchSchedule( *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBWLuma = 0; - if (st_vars->dst_y_prefetch_equ > 1) { + if (dst_y_prefetch_equ > 1) { double PrefetchBandwidth1; double PrefetchBandwidth2; double PrefetchBandwidth3; double PrefetchBandwidth4; - if (st_vars->Tpre_rounded - *Tno_bw > 0) { + if (Tpre_rounded - *Tno_bw > 0) { PrefetchBandwidth1 = (PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor + 2 * MetaRowByte + 2 * PixelPTEBytesPerRow * HostVMInefficiencyFactor - + st_vars->prefetch_sw_bytes) / (st_vars->Tpre_rounded - *Tno_bw); - st_vars->Tsw_est1 = st_vars->prefetch_sw_bytes / PrefetchBandwidth1; + + prefetch_sw_bytes) / (Tpre_rounded - *Tno_bw); + Tsw_est1 = prefetch_sw_bytes / PrefetchBandwidth1; } else PrefetchBandwidth1 = 0; - if (VStartup == MaxVStartup && (st_vars->Tsw_est1 / st_vars->LineTime < st_vars->min_Lsw) - && st_vars->Tpre_rounded - st_vars->min_Lsw * st_vars->LineTime - 0.75 * st_vars->LineTime - *Tno_bw > 0) { + if (VStartup == MaxVStartup && (Tsw_est1 / LineTime < min_Lsw) + && Tpre_rounded - min_Lsw * LineTime - 0.75 * LineTime - *Tno_bw > 0) { PrefetchBandwidth1 = (PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor + 2 * MetaRowByte + 2 * PixelPTEBytesPerRow * HostVMInefficiencyFactor) - / (st_vars->Tpre_rounded - st_vars->min_Lsw * st_vars->LineTime - 0.75 * st_vars->LineTime - *Tno_bw); + / (Tpre_rounded - min_Lsw * LineTime - 0.75 * LineTime - *Tno_bw); } - if (st_vars->Tpre_rounded - *Tno_bw - 2 * st_vars->Tr0_trips_rounded > 0) - PrefetchBandwidth2 = (PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor + st_vars->prefetch_sw_bytes) / - (st_vars->Tpre_rounded - *Tno_bw - 2 * st_vars->Tr0_trips_rounded); + if (Tpre_rounded - *Tno_bw - 2 * Tr0_trips_rounded > 0) + PrefetchBandwidth2 = (PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor + prefetch_sw_bytes) / + (Tpre_rounded - *Tno_bw - 2 * Tr0_trips_rounded); else PrefetchBandwidth2 = 0; - if (st_vars->Tpre_rounded - st_vars->Tvm_trips_rounded > 0) { + if (Tpre_rounded - Tvm_trips_rounded > 0) { PrefetchBandwidth3 = (2 * MetaRowByte + 2 * PixelPTEBytesPerRow * HostVMInefficiencyFactor - + st_vars->prefetch_sw_bytes) / (st_vars->Tpre_rounded - st_vars->Tvm_trips_rounded); - st_vars->Tsw_est3 = st_vars->prefetch_sw_bytes / PrefetchBandwidth3; + + prefetch_sw_bytes) / (Tpre_rounded - Tvm_trips_rounded); + Tsw_est3 = prefetch_sw_bytes / PrefetchBandwidth3; } else PrefetchBandwidth3 = 0; if (VStartup == MaxVStartup && - (st_vars->Tsw_est3 / st_vars->LineTime < st_vars->min_Lsw) && st_vars->Tpre_rounded - st_vars->min_Lsw * st_vars->LineTime - 0.75 * - st_vars->LineTime - st_vars->Tvm_trips_rounded > 0) { + (Tsw_est3 / LineTime < min_Lsw) && Tpre_rounded - min_Lsw * LineTime - 0.75 * + LineTime - Tvm_trips_rounded > 0) { PrefetchBandwidth3 = (2 * MetaRowByte + 2 * PixelPTEBytesPerRow * HostVMInefficiencyFactor) - / (st_vars->Tpre_rounded - st_vars->min_Lsw * st_vars->LineTime - 0.75 * st_vars->LineTime - st_vars->Tvm_trips_rounded); + / (Tpre_rounded - min_Lsw * LineTime - 0.75 * LineTime - Tvm_trips_rounded); } - if (st_vars->Tpre_rounded - st_vars->Tvm_trips_rounded - 2 * st_vars->Tr0_trips_rounded > 0) { - PrefetchBandwidth4 = st_vars->prefetch_sw_bytes / - (st_vars->Tpre_rounded - st_vars->Tvm_trips_rounded - 2 * st_vars->Tr0_trips_rounded); + if (Tpre_rounded - Tvm_trips_rounded - 2 * Tr0_trips_rounded > 0) { + PrefetchBandwidth4 = prefetch_sw_bytes / + (Tpre_rounded - Tvm_trips_rounded - 2 * Tr0_trips_rounded); } else { PrefetchBandwidth4 = 0; } #ifdef __DML_VBA_DEBUG__ - dml_print("DML::%s: Tpre_rounded: %f\n", __func__, st_vars->Tpre_rounded); + dml_print("DML::%s: Tpre_rounded: %f\n", __func__, Tpre_rounded); dml_print("DML::%s: Tno_bw: %f\n", __func__, *Tno_bw); - dml_print("DML::%s: Tvm_trips_rounded: %f\n", __func__, st_vars->Tvm_trips_rounded); - dml_print("DML::%s: Tsw_est1: %f\n", __func__, st_vars->Tsw_est1); - dml_print("DML::%s: Tsw_est3: %f\n", __func__, st_vars->Tsw_est3); + dml_print("DML::%s: Tvm_trips_rounded: %f\n", __func__, Tvm_trips_rounded); + dml_print("DML::%s: Tsw_est1: %f\n", __func__, Tsw_est1); + dml_print("DML::%s: Tsw_est3: %f\n", __func__, Tsw_est3); dml_print("DML::%s: PrefetchBandwidth1: %f\n", __func__, PrefetchBandwidth1); dml_print("DML::%s: PrefetchBandwidth2: %f\n", __func__, PrefetchBandwidth2); dml_print("DML::%s: PrefetchBandwidth3: %f\n", __func__, PrefetchBandwidth3); @@ -3732,9 +3779,9 @@ bool dml32_CalculatePrefetchSchedule( if (PrefetchBandwidth1 > 0) { if (*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / PrefetchBandwidth1 - >= st_vars->Tvm_trips_rounded + >= Tvm_trips_rounded && (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) - / PrefetchBandwidth1 >= st_vars->Tr0_trips_rounded) { + / PrefetchBandwidth1 >= Tr0_trips_rounded) { Case1OK = true; } else { Case1OK = false; @@ -3745,9 +3792,9 @@ bool dml32_CalculatePrefetchSchedule( if (PrefetchBandwidth2 > 0) { if (*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / PrefetchBandwidth2 - >= st_vars->Tvm_trips_rounded + >= Tvm_trips_rounded && (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) - / PrefetchBandwidth2 < st_vars->Tr0_trips_rounded) { + / PrefetchBandwidth2 < Tr0_trips_rounded) { Case2OK = true; } else { Case2OK = false; @@ -3758,9 +3805,9 @@ bool dml32_CalculatePrefetchSchedule( if (PrefetchBandwidth3 > 0) { if (*Tno_bw + PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / PrefetchBandwidth3 < - st_vars->Tvm_trips_rounded && (MetaRowByte + PixelPTEBytesPerRow * + Tvm_trips_rounded && (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / PrefetchBandwidth3 >= - st_vars->Tr0_trips_rounded) { + Tr0_trips_rounded) { Case3OK = true; } else { Case3OK = false; @@ -3770,80 +3817,80 @@ bool dml32_CalculatePrefetchSchedule( } if (Case1OK) - st_vars->prefetch_bw_equ = PrefetchBandwidth1; + prefetch_bw_equ = PrefetchBandwidth1; else if (Case2OK) - st_vars->prefetch_bw_equ = PrefetchBandwidth2; + prefetch_bw_equ = PrefetchBandwidth2; else if (Case3OK) - st_vars->prefetch_bw_equ = PrefetchBandwidth3; + prefetch_bw_equ = PrefetchBandwidth3; else - st_vars->prefetch_bw_equ = PrefetchBandwidth4; + prefetch_bw_equ = PrefetchBandwidth4; #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: Case1OK: %d\n", __func__, Case1OK); dml_print("DML::%s: Case2OK: %d\n", __func__, Case2OK); dml_print("DML::%s: Case3OK: %d\n", __func__, Case3OK); - dml_print("DML::%s: prefetch_bw_equ: %f\n", __func__, st_vars->prefetch_bw_equ); + dml_print("DML::%s: prefetch_bw_equ: %f\n", __func__, prefetch_bw_equ); #endif - if (st_vars->prefetch_bw_equ > 0) { + if (prefetch_bw_equ > 0) { if (GPUVMEnable == true) { - st_vars->Tvm_equ = dml_max3(*Tno_bw + PDEAndMetaPTEBytesFrame * - HostVMInefficiencyFactor / st_vars->prefetch_bw_equ, - st_vars->Tvm_trips, st_vars->LineTime / 4); + Tvm_equ = dml_max3(*Tno_bw + PDEAndMetaPTEBytesFrame * + HostVMInefficiencyFactor / prefetch_bw_equ, + Tvm_trips, LineTime / 4); } else { - st_vars->Tvm_equ = st_vars->LineTime / 4; + Tvm_equ = LineTime / 4; } if ((GPUVMEnable == true || myPipe->DCCEnable == true)) { - st_vars->Tr0_equ = dml_max4((MetaRowByte + PixelPTEBytesPerRow * - HostVMInefficiencyFactor) / st_vars->prefetch_bw_equ, st_vars->Tr0_trips, - (st_vars->LineTime - st_vars->Tvm_equ) / 2, st_vars->LineTime / 4); + Tr0_equ = dml_max4((MetaRowByte + PixelPTEBytesPerRow * + HostVMInefficiencyFactor) / prefetch_bw_equ, Tr0_trips, + (LineTime - Tvm_equ) / 2, LineTime / 4); } else { - st_vars->Tr0_equ = (st_vars->LineTime - st_vars->Tvm_equ) / 2; + Tr0_equ = (LineTime - Tvm_equ) / 2; } } else { - st_vars->Tvm_equ = 0; - st_vars->Tr0_equ = 0; + Tvm_equ = 0; + Tr0_equ = 0; #ifdef __DML_VBA_DEBUG__ dml_print("DML: prefetch_bw_equ equals 0! %s:%d\n", __FILE__, __LINE__); #endif } } - if (st_vars->dst_y_prefetch_oto < st_vars->dst_y_prefetch_equ) { - *DestinationLinesForPrefetch = st_vars->dst_y_prefetch_oto; - st_vars->TimeForFetchingMetaPTE = st_vars->Tvm_oto; - st_vars->TimeForFetchingRowInVBlank = st_vars->Tr0_oto; - *PrefetchBandwidth = st_vars->prefetch_bw_oto; + if (dst_y_prefetch_oto < dst_y_prefetch_equ) { + *DestinationLinesForPrefetch = dst_y_prefetch_oto; + TimeForFetchingMetaPTE = Tvm_oto; + TimeForFetchingRowInVBlank = Tr0_oto; + *PrefetchBandwidth = prefetch_bw_oto; } else { - *DestinationLinesForPrefetch = st_vars->dst_y_prefetch_equ; - st_vars->TimeForFetchingMetaPTE = st_vars->Tvm_equ; - st_vars->TimeForFetchingRowInVBlank = st_vars->Tr0_equ; - *PrefetchBandwidth = st_vars->prefetch_bw_equ; + *DestinationLinesForPrefetch = dst_y_prefetch_equ; + TimeForFetchingMetaPTE = Tvm_equ; + TimeForFetchingRowInVBlank = Tr0_equ; + *PrefetchBandwidth = prefetch_bw_equ; } - *DestinationLinesToRequestVMInVBlank = dml_ceil(4.0 * st_vars->TimeForFetchingMetaPTE / st_vars->LineTime, 1.0) / 4.0; + *DestinationLinesToRequestVMInVBlank = dml_ceil(4.0 * TimeForFetchingMetaPTE / LineTime, 1.0) / 4.0; *DestinationLinesToRequestRowInVBlank = - dml_ceil(4.0 * st_vars->TimeForFetchingRowInVBlank / st_vars->LineTime, 1.0) / 4.0; + dml_ceil(4.0 * TimeForFetchingRowInVBlank / LineTime, 1.0) / 4.0; - st_vars->LinesToRequestPrefetchPixelData = *DestinationLinesForPrefetch - + LinesToRequestPrefetchPixelData = *DestinationLinesForPrefetch - *DestinationLinesToRequestVMInVBlank - 2 * *DestinationLinesToRequestRowInVBlank; #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: DestinationLinesForPrefetch = %f\n", __func__, *DestinationLinesForPrefetch); dml_print("DML::%s: DestinationLinesToRequestVMInVBlank = %f\n", __func__, *DestinationLinesToRequestVMInVBlank); - dml_print("DML::%s: TimeForFetchingRowInVBlank = %f\n", __func__, st_vars->TimeForFetchingRowInVBlank); - dml_print("DML::%s: LineTime = %f\n", __func__, st_vars->LineTime); + dml_print("DML::%s: TimeForFetchingRowInVBlank = %f\n", __func__, TimeForFetchingRowInVBlank); + dml_print("DML::%s: LineTime = %f\n", __func__, LineTime); dml_print("DML::%s: DestinationLinesToRequestRowInVBlank = %f\n", __func__, *DestinationLinesToRequestRowInVBlank); dml_print("DML::%s: PrefetchSourceLinesY = %f\n", __func__, PrefetchSourceLinesY); - dml_print("DML::%s: LinesToRequestPrefetchPixelData = %f\n", __func__, st_vars->LinesToRequestPrefetchPixelData); + dml_print("DML::%s: LinesToRequestPrefetchPixelData = %f\n", __func__, LinesToRequestPrefetchPixelData); #endif - if (st_vars->LinesToRequestPrefetchPixelData >= 1 && st_vars->prefetch_bw_equ > 0) { - *VRatioPrefetchY = (double) PrefetchSourceLinesY / st_vars->LinesToRequestPrefetchPixelData; + if (LinesToRequestPrefetchPixelData >= 1 && prefetch_bw_equ > 0) { + *VRatioPrefetchY = (double) PrefetchSourceLinesY / LinesToRequestPrefetchPixelData; *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: VRatioPrefetchY = %f\n", __func__, *VRatioPrefetchY); @@ -3851,12 +3898,12 @@ bool dml32_CalculatePrefetchSchedule( dml_print("DML::%s: VInitPreFillY = %d\n", __func__, VInitPreFillY); #endif if ((SwathHeightY > 4) && (VInitPreFillY > 3)) { - if (st_vars->LinesToRequestPrefetchPixelData > (VInitPreFillY - 3.0) / 2.0) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillY - 3.0) / 2.0) { *VRatioPrefetchY = dml_max((double) PrefetchSourceLinesY / - st_vars->LinesToRequestPrefetchPixelData, + LinesToRequestPrefetchPixelData, (double) MaxNumSwathY * SwathHeightY / - (st_vars->LinesToRequestPrefetchPixelData - + (LinesToRequestPrefetchPixelData - (VInitPreFillY - 3.0) / 2.0)); *VRatioPrefetchY = dml_max(*VRatioPrefetchY, 1.0); } else { @@ -3870,7 +3917,7 @@ bool dml32_CalculatePrefetchSchedule( #endif } - *VRatioPrefetchC = (double) PrefetchSourceLinesC / st_vars->LinesToRequestPrefetchPixelData; + *VRatioPrefetchC = (double) PrefetchSourceLinesC / LinesToRequestPrefetchPixelData; *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); #ifdef __DML_VBA_DEBUG__ @@ -3879,11 +3926,11 @@ bool dml32_CalculatePrefetchSchedule( dml_print("DML::%s: VInitPreFillC = %d\n", __func__, VInitPreFillC); #endif if ((SwathHeightC > 4)) { - if (st_vars->LinesToRequestPrefetchPixelData > (VInitPreFillC - 3.0) / 2.0) { + if (LinesToRequestPrefetchPixelData > (VInitPreFillC - 3.0) / 2.0) { *VRatioPrefetchC = dml_max(*VRatioPrefetchC, (double) MaxNumSwathC * SwathHeightC / - (st_vars->LinesToRequestPrefetchPixelData - + (LinesToRequestPrefetchPixelData - (VInitPreFillC - 3.0) / 2.0)); *VRatioPrefetchC = dml_max(*VRatioPrefetchC, 1.0); } else { @@ -3898,25 +3945,25 @@ bool dml32_CalculatePrefetchSchedule( } *RequiredPrefetchPixDataBWLuma = (double) PrefetchSourceLinesY - / st_vars->LinesToRequestPrefetchPixelData * myPipe->BytePerPixelY * swath_width_luma_ub - / st_vars->LineTime; + / LinesToRequestPrefetchPixelData * myPipe->BytePerPixelY * swath_width_luma_ub + / LineTime; #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: BytePerPixelY = %d\n", __func__, myPipe->BytePerPixelY); dml_print("DML::%s: swath_width_luma_ub = %d\n", __func__, swath_width_luma_ub); - dml_print("DML::%s: LineTime = %f\n", __func__, st_vars->LineTime); + dml_print("DML::%s: LineTime = %f\n", __func__, LineTime); dml_print("DML::%s: RequiredPrefetchPixDataBWLuma = %f\n", __func__, *RequiredPrefetchPixDataBWLuma); #endif *RequiredPrefetchPixDataBWChroma = (double) PrefetchSourceLinesC / - st_vars->LinesToRequestPrefetchPixelData + LinesToRequestPrefetchPixelData * myPipe->BytePerPixelC - * swath_width_chroma_ub / st_vars->LineTime; + * swath_width_chroma_ub / LineTime; } else { MyError = true; #ifdef __DML_VBA_DEBUG__ dml_print("DML:%s: MyErr set. LinesToRequestPrefetchPixelData: %f, should be > 0\n", - __func__, st_vars->LinesToRequestPrefetchPixelData); + __func__, LinesToRequestPrefetchPixelData); #endif *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; @@ -3925,15 +3972,15 @@ bool dml32_CalculatePrefetchSchedule( } #ifdef __DML_VBA_DEBUG__ dml_print("DML: Tpre: %fus - sum of time to request meta pte, 2 x data pte + meta data, swaths\n", - (double)st_vars->LinesToRequestPrefetchPixelData * st_vars->LineTime + - 2.0*st_vars->TimeForFetchingRowInVBlank + st_vars->TimeForFetchingMetaPTE); - dml_print("DML: Tvm: %fus - time to fetch page tables for meta surface\n", st_vars->TimeForFetchingMetaPTE); + (double)LinesToRequestPrefetchPixelData * LineTime + + 2.0*TimeForFetchingRowInVBlank + TimeForFetchingMetaPTE); + dml_print("DML: Tvm: %fus - time to fetch page tables for meta surface\n", TimeForFetchingMetaPTE); dml_print("DML: To: %fus - time for propagation from scaler to optc\n", - (*DSTYAfterScaler + ((double) (*DSTXAfterScaler) / (double) myPipe->HTotal)) * st_vars->LineTime); + (*DSTYAfterScaler + ((double) (*DSTXAfterScaler) / (double) myPipe->HTotal)) * LineTime); dml_print("DML: Tvstartup - TSetup - Tcalc - Twait - Tpre - To > 0\n"); - dml_print("DML: Tslack(pre): %fus - time left over in schedule\n", VStartup * st_vars->LineTime - - st_vars->TimeForFetchingMetaPTE - 2*st_vars->TimeForFetchingRowInVBlank - (*DSTYAfterScaler + - ((double) (*DSTXAfterScaler) / (double) myPipe->HTotal)) * st_vars->LineTime - TWait - TCalc - *TSetup); + dml_print("DML: Tslack(pre): %fus - time left over in schedule\n", VStartup * LineTime - + TimeForFetchingMetaPTE - 2*TimeForFetchingRowInVBlank - (*DSTYAfterScaler + + ((double) (*DSTXAfterScaler) / (double) myPipe->HTotal)) * LineTime - TWait - TCalc - *TSetup); dml_print("DML: row_bytes = dpte_row_bytes (per_pipe) = PixelPTEBytesPerRow = : %d\n", PixelPTEBytesPerRow); #endif @@ -3941,7 +3988,7 @@ bool dml32_CalculatePrefetchSchedule( MyError = true; #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: MyErr set, dst_y_prefetch_equ = %f (should be > 1)\n", - __func__, st_vars->dst_y_prefetch_equ); + __func__, dst_y_prefetch_equ); #endif } @@ -3957,10 +4004,10 @@ bool dml32_CalculatePrefetchSchedule( dml_print("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, HostVMInefficiencyFactor); dml_print("DML::%s: DestinationLinesToRequestVMInVBlank = %f\n", __func__, *DestinationLinesToRequestVMInVBlank); - dml_print("DML::%s: LineTime = %f\n", __func__, st_vars->LineTime); + dml_print("DML::%s: LineTime = %f\n", __func__, LineTime); #endif prefetch_vm_bw = PDEAndMetaPTEBytesFrame * HostVMInefficiencyFactor / - (*DestinationLinesToRequestVMInVBlank * st_vars->LineTime); + (*DestinationLinesToRequestVMInVBlank * LineTime); #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: prefetch_vm_bw = %f\n", __func__, prefetch_vm_bw); #endif @@ -3977,7 +4024,7 @@ bool dml32_CalculatePrefetchSchedule( prefetch_row_bw = 0; } else if (*DestinationLinesToRequestRowInVBlank > 0) { prefetch_row_bw = (MetaRowByte + PixelPTEBytesPerRow * HostVMInefficiencyFactor) / - (*DestinationLinesToRequestRowInVBlank * st_vars->LineTime); + (*DestinationLinesToRequestRowInVBlank * LineTime); #ifdef __DML_VBA_DEBUG__ dml_print("DML::%s: MetaRowByte = %d\n", __func__, MetaRowByte); @@ -4000,12 +4047,12 @@ bool dml32_CalculatePrefetchSchedule( if (MyError) { *PrefetchBandwidth = 0; - st_vars->TimeForFetchingMetaPTE = 0; - st_vars->TimeForFetchingRowInVBlank = 0; + TimeForFetchingMetaPTE = 0; + TimeForFetchingRowInVBlank = 0; *DestinationLinesToRequestVMInVBlank = 0; *DestinationLinesToRequestRowInVBlank = 0; *DestinationLinesForPrefetch = 0; - st_vars->LinesToRequestPrefetchPixelData = 0; + LinesToRequestPrefetchPixelData = 0; *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBWLuma = 0; @@ -4159,7 +4206,6 @@ void dml32_CalculateFlipSchedule( } // CalculateFlipSchedule void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( - struct dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport *st_vars, bool USRRetrainingRequiredFinal, enum dm_use_mall_for_pstate_change_mode UseMALLForPStateChange[], unsigned int PrefetchMode, @@ -4221,15 +4267,37 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( double ActiveDRAMClockChangeLatencyMargin[]) { unsigned int i, j, k; - - st_vars->SurfaceWithMinActiveFCLKChangeMargin = 0; - st_vars->DRAMClockChangeSupportNumber = 0; - st_vars->DRAMClockChangeMethod = 0; - st_vars->FoundFirstSurfaceWithMinActiveFCLKChangeMargin = false; - st_vars->MinActiveFCLKChangeMargin = 0.; - st_vars->SecondMinActiveFCLKChangeMarginOneDisplayInVBLank = 0.; - st_vars->TotalPixelBW = 0.0; - st_vars->TotalActiveWriteback = 0; + unsigned int SurfaceWithMinActiveFCLKChangeMargin = 0; + unsigned int DRAMClockChangeSupportNumber = 0; + unsigned int LastSurfaceWithoutMargin; + unsigned int DRAMClockChangeMethod = 0; + bool FoundFirstSurfaceWithMinActiveFCLKChangeMargin = false; + double MinActiveFCLKChangeMargin = 0.; + double SecondMinActiveFCLKChangeMarginOneDisplayInVBLank = 0.; + double ActiveClockChangeLatencyHidingY; + double ActiveClockChangeLatencyHidingC; + double ActiveClockChangeLatencyHiding; + double EffectiveDETBufferSizeY; + double ActiveFCLKChangeLatencyMargin[DC__NUM_DPP__MAX]; + double USRRetrainingLatencyMargin[DC__NUM_DPP__MAX]; + double TotalPixelBW = 0.0; + bool SynchronizedSurfaces[DC__NUM_DPP__MAX][DC__NUM_DPP__MAX]; + double EffectiveLBLatencyHidingY; + double EffectiveLBLatencyHidingC; + double LinesInDETY[DC__NUM_DPP__MAX]; + double LinesInDETC[DC__NUM_DPP__MAX]; + unsigned int LinesInDETYRoundedDownToSwath[DC__NUM_DPP__MAX]; + unsigned int LinesInDETCRoundedDownToSwath[DC__NUM_DPP__MAX]; + double FullDETBufferingTimeY; + double FullDETBufferingTimeC; + double WritebackDRAMClockChangeLatencyMargin; + double WritebackFCLKChangeLatencyMargin; + double WritebackLatencyHiding; + bool SameTimingForFCLKChange; + + unsigned int TotalActiveWriteback = 0; + unsigned int LBLatencyHidingSourceLinesY[DC__NUM_DPP__MAX]; + unsigned int LBLatencyHidingSourceLinesC[DC__NUM_DPP__MAX]; Watermark->UrgentWatermark = mmSOCParameters.UrgentLatency + mmSOCParameters.ExtraLatency; Watermark->USRRetrainingWatermark = mmSOCParameters.UrgentLatency + mmSOCParameters.ExtraLatency @@ -4261,13 +4329,13 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( #endif - st_vars->TotalActiveWriteback = 0; + TotalActiveWriteback = 0; for (k = 0; k < NumberOfActiveSurfaces; ++k) { if (WritebackEnable[k] == true) - st_vars->TotalActiveWriteback = st_vars->TotalActiveWriteback + 1; + TotalActiveWriteback = TotalActiveWriteback + 1; } - if (st_vars->TotalActiveWriteback <= 1) { + if (TotalActiveWriteback <= 1) { Watermark->WritebackUrgentWatermark = mmSOCParameters.WritebackLatency; } else { Watermark->WritebackUrgentWatermark = mmSOCParameters.WritebackLatency @@ -4277,7 +4345,7 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( Watermark->WritebackUrgentWatermark = Watermark->WritebackUrgentWatermark + mmSOCParameters.USRRetrainingLatency; - if (st_vars->TotalActiveWriteback <= 1) { + if (TotalActiveWriteback <= 1) { Watermark->WritebackDRAMClockChangeWatermark = mmSOCParameters.DRAMClockChangeLatency + mmSOCParameters.WritebackLatency; Watermark->WritebackFCLKChangeWatermark = mmSOCParameters.FCLKChangeLatency @@ -4307,14 +4375,14 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( #endif for (k = 0; k < NumberOfActiveSurfaces; ++k) { - st_vars->TotalPixelBW = st_vars->TotalPixelBW + DPPPerSurface[k] * (SwathWidthY[k] * BytePerPixelDETY[k] * VRatio[k] + + TotalPixelBW = TotalPixelBW + DPPPerSurface[k] * (SwathWidthY[k] * BytePerPixelDETY[k] * VRatio[k] + SwathWidthC[k] * BytePerPixelDETC[k] * VRatioChroma[k]) / (HTotal[k] / PixelClock[k]); } for (k = 0; k < NumberOfActiveSurfaces; ++k) { - st_vars->LBLatencyHidingSourceLinesY[k] = dml_min((double) MaxLineBufferLines, dml_floor(LineBufferSize / LBBitPerPixel[k] / (SwathWidthY[k] / dml_max(HRatio[k], 1.0)), 1)) - (VTaps[k] - 1); - st_vars->LBLatencyHidingSourceLinesC[k] = dml_min((double) MaxLineBufferLines, dml_floor(LineBufferSize / LBBitPerPixel[k] / (SwathWidthC[k] / dml_max(HRatioChroma[k], 1.0)), 1)) - (VTapsChroma[k] - 1); + LBLatencyHidingSourceLinesY[k] = dml_min((double) MaxLineBufferLines, dml_floor(LineBufferSize / LBBitPerPixel[k] / (SwathWidthY[k] / dml_max(HRatio[k], 1.0)), 1)) - (VTaps[k] - 1); + LBLatencyHidingSourceLinesC[k] = dml_min((double) MaxLineBufferLines, dml_floor(LineBufferSize / LBBitPerPixel[k] / (SwathWidthC[k] / dml_max(HRatioChroma[k], 1.0)), 1)) - (VTapsChroma[k] - 1); #ifdef __DML_VBA_DEBUG__ @@ -4325,72 +4393,72 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( dml_print("DML::%s: k=%d, VTaps = %d\n", __func__, k, VTaps[k]); #endif - st_vars->EffectiveLBLatencyHidingY = st_vars->LBLatencyHidingSourceLinesY[k] / VRatio[k] * (HTotal[k] / PixelClock[k]); - st_vars->EffectiveLBLatencyHidingC = st_vars->LBLatencyHidingSourceLinesC[k] / VRatioChroma[k] * (HTotal[k] / PixelClock[k]); - st_vars->EffectiveDETBufferSizeY = DETBufferSizeY[k]; + EffectiveLBLatencyHidingY = LBLatencyHidingSourceLinesY[k] / VRatio[k] * (HTotal[k] / PixelClock[k]); + EffectiveLBLatencyHidingC = LBLatencyHidingSourceLinesC[k] / VRatioChroma[k] * (HTotal[k] / PixelClock[k]); + EffectiveDETBufferSizeY = DETBufferSizeY[k]; if (UnboundedRequestEnabled) { - st_vars->EffectiveDETBufferSizeY = st_vars->EffectiveDETBufferSizeY + EffectiveDETBufferSizeY = EffectiveDETBufferSizeY + CompressedBufferSizeInkByte * 1024 * (SwathWidthY[k] * BytePerPixelDETY[k] * VRatio[k]) - / (HTotal[k] / PixelClock[k]) / st_vars->TotalPixelBW; + / (HTotal[k] / PixelClock[k]) / TotalPixelBW; } - st_vars->LinesInDETY[k] = (double) st_vars->EffectiveDETBufferSizeY / BytePerPixelDETY[k] / SwathWidthY[k]; - st_vars->LinesInDETYRoundedDownToSwath[k] = dml_floor(st_vars->LinesInDETY[k], SwathHeightY[k]); - st_vars->FullDETBufferingTimeY = st_vars->LinesInDETYRoundedDownToSwath[k] * (HTotal[k] / PixelClock[k]) / VRatio[k]; + LinesInDETY[k] = (double) EffectiveDETBufferSizeY / BytePerPixelDETY[k] / SwathWidthY[k]; + LinesInDETYRoundedDownToSwath[k] = dml_floor(LinesInDETY[k], SwathHeightY[k]); + FullDETBufferingTimeY = LinesInDETYRoundedDownToSwath[k] * (HTotal[k] / PixelClock[k]) / VRatio[k]; - st_vars->ActiveClockChangeLatencyHidingY = st_vars->EffectiveLBLatencyHidingY + st_vars->FullDETBufferingTimeY + ActiveClockChangeLatencyHidingY = EffectiveLBLatencyHidingY + FullDETBufferingTimeY - (DSTXAfterScaler[k] / HTotal[k] + DSTYAfterScaler[k]) * HTotal[k] / PixelClock[k]; if (NumberOfActiveSurfaces > 1) { - st_vars->ActiveClockChangeLatencyHidingY = st_vars->ActiveClockChangeLatencyHidingY + ActiveClockChangeLatencyHidingY = ActiveClockChangeLatencyHidingY - (1 - 1 / NumberOfActiveSurfaces) * SwathHeightY[k] * HTotal[k] / PixelClock[k] / VRatio[k]; } if (BytePerPixelDETC[k] > 0) { - st_vars->LinesInDETC[k] = DETBufferSizeC[k] / BytePerPixelDETC[k] / SwathWidthC[k]; - st_vars->LinesInDETCRoundedDownToSwath[k] = dml_floor(st_vars->LinesInDETC[k], SwathHeightC[k]); - st_vars->FullDETBufferingTimeC = st_vars->LinesInDETCRoundedDownToSwath[k] * (HTotal[k] / PixelClock[k]) + LinesInDETC[k] = DETBufferSizeC[k] / BytePerPixelDETC[k] / SwathWidthC[k]; + LinesInDETCRoundedDownToSwath[k] = dml_floor(LinesInDETC[k], SwathHeightC[k]); + FullDETBufferingTimeC = LinesInDETCRoundedDownToSwath[k] * (HTotal[k] / PixelClock[k]) / VRatioChroma[k]; - st_vars->ActiveClockChangeLatencyHidingC = st_vars->EffectiveLBLatencyHidingC + st_vars->FullDETBufferingTimeC + ActiveClockChangeLatencyHidingC = EffectiveLBLatencyHidingC + FullDETBufferingTimeC - (DSTXAfterScaler[k] / HTotal[k] + DSTYAfterScaler[k]) * HTotal[k] / PixelClock[k]; if (NumberOfActiveSurfaces > 1) { - st_vars->ActiveClockChangeLatencyHidingC = st_vars->ActiveClockChangeLatencyHidingC + ActiveClockChangeLatencyHidingC = ActiveClockChangeLatencyHidingC - (1 - 1 / NumberOfActiveSurfaces) * SwathHeightC[k] * HTotal[k] / PixelClock[k] / VRatioChroma[k]; } - st_vars->ActiveClockChangeLatencyHiding = dml_min(st_vars->ActiveClockChangeLatencyHidingY, - st_vars->ActiveClockChangeLatencyHidingC); + ActiveClockChangeLatencyHiding = dml_min(ActiveClockChangeLatencyHidingY, + ActiveClockChangeLatencyHidingC); } else { - st_vars->ActiveClockChangeLatencyHiding = st_vars->ActiveClockChangeLatencyHidingY; + ActiveClockChangeLatencyHiding = ActiveClockChangeLatencyHidingY; } - ActiveDRAMClockChangeLatencyMargin[k] = st_vars->ActiveClockChangeLatencyHiding - Watermark->UrgentWatermark + ActiveDRAMClockChangeLatencyMargin[k] = ActiveClockChangeLatencyHiding - Watermark->UrgentWatermark - Watermark->DRAMClockChangeWatermark; - st_vars->ActiveFCLKChangeLatencyMargin[k] = st_vars->ActiveClockChangeLatencyHiding - Watermark->UrgentWatermark + ActiveFCLKChangeLatencyMargin[k] = ActiveClockChangeLatencyHiding - Watermark->UrgentWatermark - Watermark->FCLKChangeWatermark; - st_vars->USRRetrainingLatencyMargin[k] = st_vars->ActiveClockChangeLatencyHiding - Watermark->USRRetrainingWatermark; + USRRetrainingLatencyMargin[k] = ActiveClockChangeLatencyHiding - Watermark->USRRetrainingWatermark; if (WritebackEnable[k]) { - st_vars->WritebackLatencyHiding = WritebackInterfaceBufferSize * 1024 + WritebackLatencyHiding = WritebackInterfaceBufferSize * 1024 / (WritebackDestinationWidth[k] * WritebackDestinationHeight[k] / (WritebackSourceHeight[k] * HTotal[k] / PixelClock[k]) * 4); if (WritebackPixelFormat[k] == dm_444_64) - st_vars->WritebackLatencyHiding = st_vars->WritebackLatencyHiding / 2; + WritebackLatencyHiding = WritebackLatencyHiding / 2; - st_vars->WritebackDRAMClockChangeLatencyMargin = st_vars->WritebackLatencyHiding + WritebackDRAMClockChangeLatencyMargin = WritebackLatencyHiding - Watermark->WritebackDRAMClockChangeWatermark; - st_vars->WritebackFCLKChangeLatencyMargin = st_vars->WritebackLatencyHiding + WritebackFCLKChangeLatencyMargin = WritebackLatencyHiding - Watermark->WritebackFCLKChangeWatermark; ActiveDRAMClockChangeLatencyMargin[k] = dml_min(ActiveDRAMClockChangeLatencyMargin[k], - st_vars->WritebackFCLKChangeLatencyMargin); - st_vars->ActiveFCLKChangeLatencyMargin[k] = dml_min(st_vars->ActiveFCLKChangeLatencyMargin[k], - st_vars->WritebackDRAMClockChangeLatencyMargin); + WritebackFCLKChangeLatencyMargin); + ActiveFCLKChangeLatencyMargin[k] = dml_min(ActiveFCLKChangeLatencyMargin[k], + WritebackDRAMClockChangeLatencyMargin); } MaxActiveDRAMClockChangeLatencySupported[k] = (UseMALLForPStateChange[k] == dm_use_mall_pstate_change_phantom_pipe) ? @@ -4409,41 +4477,41 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( HTotal[i] == HTotal[j] && VTotal[i] == VTotal[j] && VActive[i] == VActive[j]) || (SynchronizeDRRDisplaysForUCLKPStateChangeFinal && (DRRDisplay[i] || DRRDisplay[j]))) { - st_vars->SynchronizedSurfaces[i][j] = true; + SynchronizedSurfaces[i][j] = true; } else { - st_vars->SynchronizedSurfaces[i][j] = false; + SynchronizedSurfaces[i][j] = false; } } } for (k = 0; k < NumberOfActiveSurfaces; ++k) { if ((UseMALLForPStateChange[k] != dm_use_mall_pstate_change_phantom_pipe) && - (!st_vars->FoundFirstSurfaceWithMinActiveFCLKChangeMargin || - st_vars->ActiveFCLKChangeLatencyMargin[k] < st_vars->MinActiveFCLKChangeMargin)) { - st_vars->FoundFirstSurfaceWithMinActiveFCLKChangeMargin = true; - st_vars->MinActiveFCLKChangeMargin = st_vars->ActiveFCLKChangeLatencyMargin[k]; - st_vars->SurfaceWithMinActiveFCLKChangeMargin = k; + (!FoundFirstSurfaceWithMinActiveFCLKChangeMargin || + ActiveFCLKChangeLatencyMargin[k] < MinActiveFCLKChangeMargin)) { + FoundFirstSurfaceWithMinActiveFCLKChangeMargin = true; + MinActiveFCLKChangeMargin = ActiveFCLKChangeLatencyMargin[k]; + SurfaceWithMinActiveFCLKChangeMargin = k; } } - *MinActiveFCLKChangeLatencySupported = st_vars->MinActiveFCLKChangeMargin + mmSOCParameters.FCLKChangeLatency; + *MinActiveFCLKChangeLatencySupported = MinActiveFCLKChangeMargin + mmSOCParameters.FCLKChangeLatency; - st_vars->SameTimingForFCLKChange = true; + SameTimingForFCLKChange = true; for (k = 0; k < NumberOfActiveSurfaces; ++k) { - if (!st_vars->SynchronizedSurfaces[k][st_vars->SurfaceWithMinActiveFCLKChangeMargin]) { + if (!SynchronizedSurfaces[k][SurfaceWithMinActiveFCLKChangeMargin]) { if ((UseMALLForPStateChange[k] != dm_use_mall_pstate_change_phantom_pipe) && - (st_vars->SameTimingForFCLKChange || - st_vars->ActiveFCLKChangeLatencyMargin[k] < - st_vars->SecondMinActiveFCLKChangeMarginOneDisplayInVBLank)) { - st_vars->SecondMinActiveFCLKChangeMarginOneDisplayInVBLank = st_vars->ActiveFCLKChangeLatencyMargin[k]; + (SameTimingForFCLKChange || + ActiveFCLKChangeLatencyMargin[k] < + SecondMinActiveFCLKChangeMarginOneDisplayInVBLank)) { + SecondMinActiveFCLKChangeMarginOneDisplayInVBLank = ActiveFCLKChangeLatencyMargin[k]; } - st_vars->SameTimingForFCLKChange = false; + SameTimingForFCLKChange = false; } } - if (st_vars->MinActiveFCLKChangeMargin > 0) { + if (MinActiveFCLKChangeMargin > 0) { *FCLKChangeSupport = dm_fclock_change_vactive; - } else if ((st_vars->SameTimingForFCLKChange || st_vars->SecondMinActiveFCLKChangeMarginOneDisplayInVBLank > 0) && + } else if ((SameTimingForFCLKChange || SecondMinActiveFCLKChangeMarginOneDisplayInVBLank > 0) && (PrefetchMode <= 1)) { *FCLKChangeSupport = dm_fclock_change_vblank; } else { @@ -4453,7 +4521,7 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( *USRRetrainingSupport = true; for (k = 0; k < NumberOfActiveSurfaces; ++k) { if ((UseMALLForPStateChange[k] != dm_use_mall_pstate_change_phantom_pipe) && - (st_vars->USRRetrainingLatencyMargin[k] < 0)) { + (USRRetrainingLatencyMargin[k] < 0)) { *USRRetrainingSupport = false; } } @@ -4464,42 +4532,42 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( UseMALLForPStateChange[k] != dm_use_mall_pstate_change_phantom_pipe && ActiveDRAMClockChangeLatencyMargin[k] < 0) { if (PrefetchMode > 0) { - st_vars->DRAMClockChangeSupportNumber = 2; - } else if (st_vars->DRAMClockChangeSupportNumber == 0) { - st_vars->DRAMClockChangeSupportNumber = 1; - st_vars->LastSurfaceWithoutMargin = k; - } else if (st_vars->DRAMClockChangeSupportNumber == 1 && - !st_vars->SynchronizedSurfaces[st_vars->LastSurfaceWithoutMargin][k]) { - st_vars->DRAMClockChangeSupportNumber = 2; + DRAMClockChangeSupportNumber = 2; + } else if (DRAMClockChangeSupportNumber == 0) { + DRAMClockChangeSupportNumber = 1; + LastSurfaceWithoutMargin = k; + } else if (DRAMClockChangeSupportNumber == 1 && + !SynchronizedSurfaces[LastSurfaceWithoutMargin][k]) { + DRAMClockChangeSupportNumber = 2; } } } for (k = 0; k < NumberOfActiveSurfaces; ++k) { if (UseMALLForPStateChange[k] == dm_use_mall_pstate_change_full_frame) - st_vars->DRAMClockChangeMethod = 1; + DRAMClockChangeMethod = 1; else if (UseMALLForPStateChange[k] == dm_use_mall_pstate_change_sub_viewport) - st_vars->DRAMClockChangeMethod = 2; + DRAMClockChangeMethod = 2; } - if (st_vars->DRAMClockChangeMethod == 0) { - if (st_vars->DRAMClockChangeSupportNumber == 0) + if (DRAMClockChangeMethod == 0) { + if (DRAMClockChangeSupportNumber == 0) *DRAMClockChangeSupport = dm_dram_clock_change_vactive; - else if (st_vars->DRAMClockChangeSupportNumber == 1) + else if (DRAMClockChangeSupportNumber == 1) *DRAMClockChangeSupport = dm_dram_clock_change_vblank; else *DRAMClockChangeSupport = dm_dram_clock_change_unsupported; - } else if (st_vars->DRAMClockChangeMethod == 1) { - if (st_vars->DRAMClockChangeSupportNumber == 0) + } else if (DRAMClockChangeMethod == 1) { + if (DRAMClockChangeSupportNumber == 0) *DRAMClockChangeSupport = dm_dram_clock_change_vactive_w_mall_full_frame; - else if (st_vars->DRAMClockChangeSupportNumber == 1) + else if (DRAMClockChangeSupportNumber == 1) *DRAMClockChangeSupport = dm_dram_clock_change_vblank_w_mall_full_frame; else *DRAMClockChangeSupport = dm_dram_clock_change_unsupported; } else { - if (st_vars->DRAMClockChangeSupportNumber == 0) + if (DRAMClockChangeSupportNumber == 0) *DRAMClockChangeSupport = dm_dram_clock_change_vactive_w_mall_sub_vp; - else if (st_vars->DRAMClockChangeSupportNumber == 1) + else if (DRAMClockChangeSupportNumber == 1) *DRAMClockChangeSupport = dm_dram_clock_change_vblank_w_mall_sub_vp; else *DRAMClockChangeSupport = dm_dram_clock_change_unsupported; @@ -4513,7 +4581,7 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( dst_y_pstate = dml_ceil((mmSOCParameters.DRAMClockChangeLatency + mmSOCParameters.UrgentLatency) / (HTotal[k] / PixelClock[k]), 1); src_y_pstate_l = dml_ceil(dst_y_pstate * VRatio[k], SwathHeightY[k]); - src_y_ahead_l = dml_floor(DETBufferSizeY[k] / BytePerPixelDETY[k] / SwathWidthY[k], SwathHeightY[k]) + st_vars->LBLatencyHidingSourceLinesY[k]; + src_y_ahead_l = dml_floor(DETBufferSizeY[k] / BytePerPixelDETY[k] / SwathWidthY[k], SwathHeightY[k]) + LBLatencyHidingSourceLinesY[k]; sub_vp_lines_l = src_y_pstate_l + src_y_ahead_l + meta_row_height[k]; #ifdef __DML_VBA_DEBUG__ @@ -4521,7 +4589,7 @@ dml_print("DML::%s: k=%d, DETBufferSizeY = %d\n", __func__, k, DET dml_print("DML::%s: k=%d, BytePerPixelDETY = %f\n", __func__, k, BytePerPixelDETY[k]); dml_print("DML::%s: k=%d, SwathWidthY = %d\n", __func__, k, SwathWidthY[k]); dml_print("DML::%s: k=%d, SwathHeightY = %d\n", __func__, k, SwathHeightY[k]); -dml_print("DML::%s: k=%d, LBLatencyHidingSourceLinesY = %d\n", __func__, k, st_vars->LBLatencyHidingSourceLinesY[k]); +dml_print("DML::%s: k=%d, LBLatencyHidingSourceLinesY = %d\n", __func__, k, LBLatencyHidingSourceLinesY[k]); dml_print("DML::%s: k=%d, dst_y_pstate = %d\n", __func__, k, dst_y_pstate); dml_print("DML::%s: k=%d, src_y_pstate_l = %d\n", __func__, k, src_y_pstate_l); dml_print("DML::%s: k=%d, src_y_ahead_l = %d\n", __func__, k, src_y_ahead_l); @@ -4532,7 +4600,7 @@ dml_print("DML::%s: k=%d, sub_vp_lines_l = %d\n", __func__, k, sub_vp_lines_l if (BytePerPixelDETC[k] > 0) { src_y_pstate_c = dml_ceil(dst_y_pstate * VRatioChroma[k], SwathHeightC[k]); - src_y_ahead_c = dml_floor(DETBufferSizeC[k] / BytePerPixelDETC[k] / SwathWidthC[k], SwathHeightC[k]) + st_vars->LBLatencyHidingSourceLinesC[k]; + src_y_ahead_c = dml_floor(DETBufferSizeC[k] / BytePerPixelDETC[k] / SwathWidthC[k], SwathHeightC[k]) + LBLatencyHidingSourceLinesC[k]; sub_vp_lines_c = src_y_pstate_c + src_y_ahead_c + meta_row_height_chroma[k]; SubViewportLinesNeededInMALL[k] = dml_max(sub_vp_lines_l, sub_vp_lines_c); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h index 37a314ce284b..d293856ba906 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h @@ -30,7 +30,6 @@ #include "os_types.h" #include "../dc_features.h" #include "../display_mode_structs.h" -#include "dml/display_mode_vba.h" unsigned int dml32_dscceComputeDelay( unsigned int bpc, @@ -82,7 +81,6 @@ void dml32_CalculateSinglePipeDPPCLKAndSCLThroughput( double *DPPCLKUsingSingleDPP); void dml32_CalculateSwathAndDETConfiguration( - struct dml32_CalculateSwathAndDETConfiguration *st_vars, unsigned int DETSizeOverride[], enum dm_use_mall_for_pstate_change_mode UseMALLForPStateChange[], unsigned int ConfigReturnBufferSizeInKByte, @@ -362,7 +360,6 @@ void dml32_CalculateSurfaceSizeInMall( bool *ExceededMALLSize); void dml32_CalculateVMRowAndSwath( - struct dml32_CalculateVMRowAndSwath *st_vars, unsigned int NumberOfActiveSurfaces, DmlPipe myPipe[], unsigned int SurfaceSizeInMALL[], @@ -715,7 +712,6 @@ double dml32_CalculateExtraLatency( unsigned int HostVMMaxNonCachedPageTableLevels); bool dml32_CalculatePrefetchSchedule( - struct dml32_CalculatePrefetchSchedule *st_vars, double HostVMInefficiencyFactor, DmlPipe *myPipe, unsigned int DSCDelay, @@ -811,7 +807,6 @@ void dml32_CalculateFlipSchedule( bool *ImmediateFlipSupportedForPipe); void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( - struct dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport *st_vars, bool USRRetrainingRequiredFinal, enum dm_use_mall_for_pstate_change_mode UseMALLForPStateChange[], unsigned int PrefetchMode, diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c index 84b4b00f29cb..c87091683b5d 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c @@ -498,6 +498,13 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; } + if ((int)(dcn3_21_soc.fclk_change_latency_us * 1000) + != dc->bb_overrides.fclk_clock_change_latency_ns + && dc->bb_overrides.fclk_clock_change_latency_ns) { + dcn3_21_soc.fclk_change_latency_us = + dc->bb_overrides.fclk_clock_change_latency_ns / 1000; + } + if ((int)(dcn3_21_soc.dummy_pstate_latency_us * 1000) != dc->bb_overrides.dummy_clock_change_latency_ns && dc->bb_overrides.dummy_clock_change_latency_ns) { diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h index 8460aefe7b6d..492aec634b68 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h @@ -182,108 +182,6 @@ void Calculate256BBlockSizes( unsigned int *BlockWidth256BytesY, unsigned int *BlockWidth256BytesC); -struct dml32_CalculateSwathAndDETConfiguration { - unsigned int MaximumSwathHeightY[DC__NUM_DPP__MAX]; - unsigned int MaximumSwathHeightC[DC__NUM_DPP__MAX]; - unsigned int RoundedUpMaxSwathSizeBytesY[DC__NUM_DPP__MAX]; - unsigned int RoundedUpMaxSwathSizeBytesC[DC__NUM_DPP__MAX]; - unsigned int RoundedUpSwathSizeBytesY; - unsigned int RoundedUpSwathSizeBytesC; - double SwathWidthdoubleDPP[DC__NUM_DPP__MAX]; - double SwathWidthdoubleDPPChroma[DC__NUM_DPP__MAX]; - unsigned int TotalActiveDPP; - bool NoChromaSurfaces; - unsigned int DETBufferSizeInKByteForSwathCalculation; -}; - -struct dml32_CalculateVMRowAndSwath { - unsigned int PTEBufferSizeInRequestsForLuma[DC__NUM_DPP__MAX]; - unsigned int PTEBufferSizeInRequestsForChroma[DC__NUM_DPP__MAX]; - unsigned int PDEAndMetaPTEBytesFrameY; - unsigned int PDEAndMetaPTEBytesFrameC; - unsigned int MetaRowByteY[DC__NUM_DPP__MAX]; - unsigned int MetaRowByteC[DC__NUM_DPP__MAX]; - unsigned int PixelPTEBytesPerRowY[DC__NUM_DPP__MAX]; - unsigned int PixelPTEBytesPerRowC[DC__NUM_DPP__MAX]; - unsigned int PixelPTEBytesPerRowY_one_row_per_frame[DC__NUM_DPP__MAX]; - unsigned int PixelPTEBytesPerRowC_one_row_per_frame[DC__NUM_DPP__MAX]; - unsigned int dpte_row_width_luma_ub_one_row_per_frame[DC__NUM_DPP__MAX]; - unsigned int dpte_row_height_luma_one_row_per_frame[DC__NUM_DPP__MAX]; - unsigned int dpte_row_width_chroma_ub_one_row_per_frame[DC__NUM_DPP__MAX]; - unsigned int dpte_row_height_chroma_one_row_per_frame[DC__NUM_DPP__MAX]; - bool one_row_per_frame_fits_in_buffer[DC__NUM_DPP__MAX]; -}; - -struct dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport { - unsigned int SurfaceWithMinActiveFCLKChangeMargin; - unsigned int DRAMClockChangeSupportNumber; - unsigned int LastSurfaceWithoutMargin; - unsigned int DRAMClockChangeMethod; - bool FoundFirstSurfaceWithMinActiveFCLKChangeMargin; - double MinActiveFCLKChangeMargin; - double SecondMinActiveFCLKChangeMarginOneDisplayInVBLank; - double ActiveClockChangeLatencyHidingY; - double ActiveClockChangeLatencyHidingC; - double ActiveClockChangeLatencyHiding; - double EffectiveDETBufferSizeY; - double ActiveFCLKChangeLatencyMargin[DC__NUM_DPP__MAX]; - double USRRetrainingLatencyMargin[DC__NUM_DPP__MAX]; - double TotalPixelBW; - bool SynchronizedSurfaces[DC__NUM_DPP__MAX][DC__NUM_DPP__MAX]; - double EffectiveLBLatencyHidingY; - double EffectiveLBLatencyHidingC; - double LinesInDETY[DC__NUM_DPP__MAX]; - double LinesInDETC[DC__NUM_DPP__MAX]; - unsigned int LinesInDETYRoundedDownToSwath[DC__NUM_DPP__MAX]; - unsigned int LinesInDETCRoundedDownToSwath[DC__NUM_DPP__MAX]; - double FullDETBufferingTimeY; - double FullDETBufferingTimeC; - double WritebackDRAMClockChangeLatencyMargin; - double WritebackFCLKChangeLatencyMargin; - double WritebackLatencyHiding; - bool SameTimingForFCLKChange; - unsigned int TotalActiveWriteback; - unsigned int LBLatencyHidingSourceLinesY[DC__NUM_DPP__MAX]; - unsigned int LBLatencyHidingSourceLinesC[DC__NUM_DPP__MAX]; -}; - -struct dml32_CalculatePrefetchSchedule { - unsigned int DPPCycles, DISPCLKCycles; - double DSTTotalPixelsAfterScaler; - double LineTime; - double dst_y_prefetch_equ; - double prefetch_bw_oto; - double Tvm_oto; - double Tr0_oto; - double Tvm_oto_lines; - double Tr0_oto_lines; - double dst_y_prefetch_oto; - double TimeForFetchingMetaPTE; - double TimeForFetchingRowInVBlank; - double LinesToRequestPrefetchPixelData; - unsigned int HostVMDynamicLevelsTrips; - double trip_to_mem; - double Tvm_trips; - double Tr0_trips; - double Tvm_trips_rounded; - double Tr0_trips_rounded; - double Lsw_oto; - double Tpre_rounded; - double prefetch_bw_equ; - double Tvm_equ; - double Tr0_equ; - double Tdmbf; - double Tdmec; - double Tdmsks; - double prefetch_sw_bytes; - double bytes_pp; - double dep_bytes; - unsigned int max_vratio_pre; - double min_Lsw; - double Tsw_est1; - double Tsw_est3; -}; - struct DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation { unsigned int dummy_integer_array[2][DC__NUM_DPP__MAX]; double dummy_single_array[2][DC__NUM_DPP__MAX]; @@ -355,10 +253,6 @@ struct dummy_vars { struct DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerformanceCalculation; struct dml32_ModeSupportAndSystemConfigurationFull dml32_ModeSupportAndSystemConfigurationFull; - struct dml32_CalculateSwathAndDETConfiguration dml32_CalculateSwathAndDETConfiguration; - struct dml32_CalculateVMRowAndSwath dml32_CalculateVMRowAndSwath; - struct dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport; - struct dml32_CalculatePrefetchSchedule dml32_CalculatePrefetchSchedule; }; struct vba_vars_st { diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c index db7b0b155374..226af06278ce 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c @@ -116,7 +116,7 @@ static void setup_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) dto_params.timing = &pipe_ctx->stream->timing; dto_params.ref_dtbclk_khz = dc->clk_mgr->funcs->get_dtb_ref_clk_frequency(dc->clk_mgr); - dccg->funcs->set_dpstreamclk(dccg, DTBCLK0, tg->inst, link_enc->inst); + dccg->funcs->set_dpstreamclk(dccg, DTBCLK0, tg->inst, stream_enc->inst); dccg->funcs->enable_symclk32_se(dccg, stream_enc->inst, phyd32clk); dccg->funcs->set_dtbclk_dto(dccg, &dto_params); stream_enc->funcs->enable_stream(stream_enc); @@ -137,7 +137,7 @@ static void reset_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) stream_enc->funcs->disable(stream_enc); dccg->funcs->set_dtbclk_dto(dccg, &dto_params); dccg->funcs->disable_symclk32_se(dccg, stream_enc->inst); - dccg->funcs->set_dpstreamclk(dccg, REFCLK, tg->inst, pipe_ctx->link_res.hpo_dp_link_enc->inst); + dccg->funcs->set_dpstreamclk(dccg, REFCLK, tg->inst, stream_enc->inst); } static void setup_hpo_dp_stream_attribute(struct pipe_ctx *pipe_ctx) diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h index ab06c7fc7452..9f3558c0ef11 100644 --- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h +++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h @@ -244,13 +244,15 @@ enum { #define ASICREV_IS_GC_10_3_7(eChipRev) ((eChipRev >= GC_10_3_7_A0) && (eChipRev < GC_10_3_7_UNKNOWN)) #define AMDGPU_FAMILY_GC_11_0_0 145 -#define AMDGPU_FAMILY_GC_11_0_2 148 +#define AMDGPU_FAMILY_GC_11_0_1 148 #define GC_11_0_0_A0 0x1 #define GC_11_0_2_A0 0x10 +#define GC_11_0_3_A0 0x20 #define GC_11_UNKNOWN 0xFF #define ASICREV_IS_GC_11_0_0(eChipRev) (eChipRev < GC_11_0_2_A0) -#define ASICREV_IS_GC_11_0_2(eChipRev) (eChipRev >= GC_11_0_2_A0 && eChipRev < GC_11_UNKNOWN) +#define ASICREV_IS_GC_11_0_2(eChipRev) (eChipRev >= GC_11_0_2_A0 && eChipRev < GC_11_0_3_A0) +#define ASICREV_IS_GC_11_0_3(eChipRev) (eChipRev >= GC_11_0_3_A0 && eChipRev < GC_11_UNKNOWN) /* * ASIC chip ID diff --git a/drivers/gpu/drm/amd/display/include/logger_types.h b/drivers/gpu/drm/amd/display/include/logger_types.h index f093b49c5e6e..3bf08a60c45c 100644 --- a/drivers/gpu/drm/amd/display/include/logger_types.h +++ b/drivers/gpu/drm/amd/display/include/logger_types.h @@ -119,13 +119,15 @@ enum dc_log_type { LOG_HDMI_RETIMER_REDRIVER, LOG_DSC, LOG_SMU_MSG, + LOG_DC2RESERVED4, + LOG_DC2RESERVED5, LOG_DWB, LOG_GAMMA_DEBUG, LOG_MAX_HW_POINTS, LOG_ALL_TF_CHANNELS, LOG_SAMPLE_1DLUT, LOG_DP2, - LOG_SECTION_TOTAL_COUNT + LOG_DC2RESERVED12, }; #define DC_MIN_LOG_MASK ((1 << LOG_ERROR) | \ diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index da09ba7589f7..0f39ab9dc5b4 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -613,10 +613,6 @@ static void build_vrr_infopacket_data_v1(const struct mod_vrr_params *vrr, * Note: We should never go above the field rate of the mode timing set. */ infopacket->sb[8] = (unsigned char)((vrr->max_refresh_in_uhz + 500000) / 1000000); - - /* FreeSync HDR */ - infopacket->sb[9] = 0; - infopacket->sb[10] = 0; } static void build_vrr_infopacket_data_v3(const struct mod_vrr_params *vrr, @@ -684,10 +680,6 @@ static void build_vrr_infopacket_data_v3(const struct mod_vrr_params *vrr, /* PB16 : Reserved bits 7:1, FixedRate bit 0 */ infopacket->sb[16] = (vrr->state == VRR_STATE_ACTIVE_FIXED) ? 1 : 0; - - //FreeSync HDR - infopacket->sb[9] = 0; - infopacket->sb[10] = 0; } static void build_vrr_infopacket_fs2_data(enum color_transfer_func app_tf, @@ -772,8 +764,7 @@ static void build_vrr_infopacket_header_v2(enum signal_type signal, /* HB2 = [Bits 7:5 = 0] [Bits 4:0 = Length = 0x09] */ infopacket->hb2 = 0x09; - *payload_size = 0x0A; - + *payload_size = 0x09; } else if (dc_is_dp_signal(signal)) { /* HEADER */ @@ -822,9 +813,9 @@ static void build_vrr_infopacket_header_v3(enum signal_type signal, infopacket->hb1 = version; /* HB2 = [Bits 7:5 = 0] [Bits 4:0 = Length] */ - *payload_size = 0x10; - infopacket->hb2 = *payload_size - 1; //-1 for checksum + infopacket->hb2 = 0x10; + *payload_size = 0x10; } else if (dc_is_dp_signal(signal)) { /* HEADER */ diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_7_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_7_0_offset.h index 2ed95790a600..cf8d60c4df1b 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_7_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_7_0_offset.h @@ -15243,6 +15243,8 @@ #define regBIF0_PCIE_TX_TRACKING_ADDR_HI_BASE_IDX 5 #define regBIF0_PCIE_TX_TRACKING_CTRL_STATUS 0x420186 #define regBIF0_PCIE_TX_TRACKING_CTRL_STATUS_BASE_IDX 5 +#define regBIF0_PCIE_TX_POWER_CTRL_1 0x420187 +#define regBIF0_PCIE_TX_POWER_CTRL_1_BASE_IDX 5 #define regBIF0_PCIE_TX_CTRL_4 0x42018b #define regBIF0_PCIE_TX_CTRL_4_BASE_IDX 5 #define regBIF0_PCIE_TX_STATUS 0x420194 diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_7_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_7_0_sh_mask.h index eb62a18fcc48..3d60c9e92548 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_7_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_7_0_sh_mask.h @@ -85627,6 +85627,19 @@ #define BIF0_PCIE_TX_TRACKING_CTRL_STATUS__TX_TRACKING_PORT_MASK 0x0000000EL #define BIF0_PCIE_TX_TRACKING_CTRL_STATUS__TX_TRACKING_UNIT_ID_MASK 0x00007F00L #define BIF0_PCIE_TX_TRACKING_CTRL_STATUS__TX_TRACKING_STATUS_VALID_MASK 0x00008000L +//BIF0_PCIE_TX_POWER_CTRL_1 +#define BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_LS_EN__SHIFT 0x0 +#define BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_DS_EN__SHIFT 0x1 +#define BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_SD_EN__SHIFT 0x2 +#define BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_LS_EN__SHIFT 0x3 +#define BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_DS_EN__SHIFT 0x4 +#define BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_SD_EN__SHIFT 0x5 +#define BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_LS_EN_MASK 0x00000001L +#define BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_DS_EN_MASK 0x00000002L +#define BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_SD_EN_MASK 0x00000004L +#define BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_LS_EN_MASK 0x00000008L +#define BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_DS_EN_MASK 0x00000010L +#define BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_SD_EN_MASK 0x00000020L //BIF0_PCIE_TX_CTRL_4 #define BIF0_PCIE_TX_CTRL_4__TX_PORT_ACCESS_TIMER_SKEW__SHIFT 0x0 #define BIF0_PCIE_TX_CTRL_4__TX_PORT_ACCESS_TIMER_SKEW_MASK 0x0000000FL diff --git a/drivers/gpu/drm/amd/include/mes_v11_api_def.h b/drivers/gpu/drm/amd/include/mes_v11_api_def.h index 80dab1146439..50bfa513cb35 100644 --- a/drivers/gpu/drm/amd/include/mes_v11_api_def.h +++ b/drivers/gpu/drm/amd/include/mes_v11_api_def.h @@ -268,7 +268,8 @@ union MESAPI__ADD_QUEUE { uint32_t is_tmz_queue : 1; uint32_t map_kiq_utility_queue : 1; uint32_t is_kfd_process : 1; - uint32_t reserved : 22; + uint32_t trap_en : 1; + uint32_t reserved : 21; }; struct MES_API_STATUS api_status; uint64_t tma_addr; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h index 78620b0bd279..063f4a737605 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h @@ -24,12 +24,8 @@ #ifndef SMU13_DRIVER_IF_V13_0_0_H #define SMU13_DRIVER_IF_V13_0_0_H -// *** IMPORTANT *** -// PMFW TEAM: Always increment the interface version on any change to this file -#define SMU13_DRIVER_IF_VERSION 0x23 - //Increment this version if SkuTable_t or BoardTable_t change -#define PPTABLE_VERSION 0x1D +#define PPTABLE_VERSION 0x24 #define NUM_GFXCLK_DPM_LEVELS 16 #define NUM_SOCCLK_DPM_LEVELS 8 @@ -1193,8 +1189,17 @@ typedef struct { // SECTION: Advanced Options uint32_t DebugOverrides; + // Section: Total Board Power idle vs active coefficients + uint8_t TotalBoardPowerSupport; + uint8_t TotalBoardPowerPadding[3]; + + int16_t TotalIdleBoardPowerM; + int16_t TotalIdleBoardPowerB; + int16_t TotalBoardPowerM; + int16_t TotalBoardPowerB; + // SECTION: Sku Reserved - uint32_t Spare[64]; + uint32_t Spare[61]; // Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; @@ -1259,7 +1264,8 @@ typedef struct { // SECTION: Clock Spread Spectrum // UCLK Spread Spectrum - uint16_t UclkSpreadPadding; + uint8_t UclkTrainingModeSpreadPercent; + uint8_t UclkSpreadPadding; uint16_t UclkSpreadFreq; // kHz // UCLK Spread Spectrum @@ -1272,11 +1278,7 @@ typedef struct { // Section: Memory Config uint8_t DramWidth; // Width of interface to the channel for each DRAM module. See DRAM_BIT_WIDTH_TYPE_e - uint8_t PaddingMem1[3]; - - // Section: Total Board Power - uint16_t TotalBoardPower; //Only needed for TCP Estimated case, where TCP = TGP+Total Board Power - uint16_t BoardPowerPadding; + uint8_t PaddingMem1[7]; // SECTION: UMC feature flags uint8_t HsrEnabled; @@ -1375,8 +1377,11 @@ typedef struct { uint16_t Vcn1ActivityPercentage ; uint32_t EnergyAccumulator; - uint16_t AverageSocketPower ; + uint16_t AverageSocketPower; + uint16_t AverageTotalBoardPower; + uint16_t AvgTemperature[TEMP_COUNT]; + uint16_t TempPadding; uint8_t PcieRate ; uint8_t PcieWidth ; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h index 76f695a1d065..ae2d337158f3 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h @@ -27,7 +27,7 @@ // *** IMPORTANT *** // SMU TEAM: Always increment the interface version if // any structure is changed in this file -#define PMFW_DRIVER_IF_VERSION 4 +#define PMFW_DRIVER_IF_VERSION 5 typedef struct { int32_t value; @@ -197,6 +197,8 @@ typedef struct { uint16_t SkinTemp; uint16_t DeviceState; + uint16_t CurTemp; //[centi-Celsius] + uint16_t spare2; } SmuMetrics_t; typedef struct { diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h index c02e5e576728..f442bf085a31 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h @@ -28,9 +28,9 @@ #define SMU13_DRIVER_IF_VERSION_INV 0xFFFFFFFF #define SMU13_DRIVER_IF_VERSION_YELLOW_CARP 0x04 #define SMU13_DRIVER_IF_VERSION_ALDE 0x08 -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_4 0x04 +#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_4 0x05 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_5 0x04 -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0 0x2C +#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0 0x30 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_7 0x2C #define SMU13_MODE1_RESET_WAIT_TIME_IN_MS 500 //500ms @@ -291,5 +291,11 @@ int smu_v13_0_set_default_dpm_tables(struct smu_context *smu); void smu_v13_0_set_smu_mailbox_registers(struct smu_context *smu); int smu_v13_0_mode1_reset(struct smu_context *smu); + +int smu_v13_0_get_pptable_from_firmware(struct smu_context *smu, + void **table, + uint32_t *size, + uint32_t pptable_id); + #endif #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index fa520d79ef67..6db67f082d91 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -4283,6 +4283,7 @@ static const struct pptable_funcs sienna_cichlid_ppt_funcs = { .dump_pptable = sienna_cichlid_dump_pptable, .init_microcode = smu_v11_0_init_microcode, .load_microcode = smu_v11_0_load_microcode, + .fini_microcode = smu_v11_0_fini_microcode, .init_smc_tables = sienna_cichlid_init_smc_tables, .fini_smc_tables = smu_v11_0_fini_smc_tables, .init_power = smu_v11_0_init_power, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index e8fe84f806d1..24488f4cb78c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -84,9 +84,6 @@ MODULE_FIRMWARE("amdgpu/smu_13_0_7.bin"); static const int link_width[] = {0, 1, 2, 4, 8, 12, 16}; static const int link_speed[] = {25, 50, 80, 160}; -static int smu_v13_0_get_pptable_from_firmware(struct smu_context *smu, void **table, uint32_t *size, - uint32_t pptable_id); - int smu_v13_0_init_microcode(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; @@ -212,6 +209,9 @@ int smu_v13_0_init_pptable_microcode(struct smu_context *smu) if (!adev->scpm_enabled) return 0; + if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 7)) + return 0; + /* override pptable_id from driver parameter */ if (amdgpu_smu_pptable_id >= 0) { pptable_id = amdgpu_smu_pptable_id; @@ -219,24 +219,17 @@ int smu_v13_0_init_pptable_microcode(struct smu_context *smu) } else { pptable_id = smu->smu_table.boot_values.pp_table_id; - if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 7) && - pptable_id == 3667) - pptable_id = 36671; - - if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 7) && - pptable_id == 3688) - pptable_id = 36881; /* * Temporary solution for SMU V13.0.0 with SCPM enabled: - * - use 36831 signed pptable when pp_table_id is 3683 - * - use 36641 signed pptable when pp_table_id is 3664 or 0 - * TODO: drop these when the pptable carried in vbios is ready. + * - use vbios carried pptable when pptable_id is 3664, 3715 or 3795 + * - use 36831 soft pptable when pptable_id is 3683 */ if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 0)) { switch (pptable_id) { - case 0: case 3664: - pptable_id = 36641; + case 3715: + case 3795: + pptable_id = 0; break; case 3683: pptable_id = 36831; @@ -425,8 +418,10 @@ static int smu_v13_0_get_pptable_from_vbios(struct smu_context *smu, void **tabl return 0; } -static int smu_v13_0_get_pptable_from_firmware(struct smu_context *smu, void **table, uint32_t *size, - uint32_t pptable_id) +int smu_v13_0_get_pptable_from_firmware(struct smu_context *smu, + void **table, + uint32_t *size, + uint32_t pptable_id) { const struct smc_firmware_header_v1_0 *hdr; struct amdgpu_device *adev = smu->adev; @@ -478,7 +473,7 @@ int smu_v13_0_setup_pptable(struct smu_context *smu) /* * Temporary solution for SMU V13.0.0 with SCPM disabled: - * - use 3664 or 3683 on request + * - use 3664, 3683 or 3715 on request * - use 3664 when pptable_id is 0 * TODO: drop these when the pptable carried in vbios is ready. */ @@ -489,6 +484,7 @@ int smu_v13_0_setup_pptable(struct smu_context *smu) break; case 3664: case 3683: + case 3715: break; default: dev_err(adev->dev, "Unsupported pptable id %d\n", pptable_id); @@ -2344,8 +2340,8 @@ int smu_v13_0_set_gfx_power_up_by_imu(struct smu_context *smu) index = smu_cmn_to_asic_specific_index(smu, CMN2ASIC_MAPPING_MSG, SMU_MSG_EnableGfxImu); - - return smu_cmn_send_msg_without_waiting(smu, index, 0); + /* Param 1 to tell PMFW to enable GFXOFF feature */ + return smu_cmn_send_msg_without_waiting(smu, index, 1); } int smu_v13_0_od_edit_dpm_table(struct smu_context *smu, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 1bbeceeb9e3c..7db2fd9ea74a 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -388,11 +388,29 @@ static int smu_v13_0_0_append_powerplay_table(struct smu_context *smu) return 0; } -static int smu_v13_0_0_setup_pptable(struct smu_context *smu) +static int smu_v13_0_0_get_pptable_from_pmfw(struct smu_context *smu, + void **table, + uint32_t *size) { struct smu_table_context *smu_table = &smu->smu_table; void *combo_pptable = smu_table->combo_pptable; + int ret = 0; + + ret = smu_cmn_get_combo_pptable(smu); + if (ret) + return ret; + + *table = combo_pptable; + *size = sizeof(struct smu_13_0_0_powerplay_table); + + return 0; +} + +static int smu_v13_0_0_setup_pptable(struct smu_context *smu) +{ + struct smu_table_context *smu_table = &smu->smu_table; struct amdgpu_device *adev = smu->adev; + uint32_t pptable_id; int ret = 0; /* @@ -401,17 +419,51 @@ static int smu_v13_0_0_setup_pptable(struct smu_context *smu) * rely on the combo pptable(and its revelant SMU message). */ if (adev->scpm_enabled) { - ret = smu_cmn_get_combo_pptable(smu); - if (ret) - return ret; - - smu->smu_table.power_play_table = combo_pptable; - smu->smu_table.power_play_table_size = sizeof(struct smu_13_0_0_powerplay_table); + ret = smu_v13_0_0_get_pptable_from_pmfw(smu, + &smu_table->power_play_table, + &smu_table->power_play_table_size); } else { - ret = smu_v13_0_setup_pptable(smu); - if (ret) - return ret; + /* override pptable_id from driver parameter */ + if (amdgpu_smu_pptable_id >= 0) { + pptable_id = amdgpu_smu_pptable_id; + dev_info(adev->dev, "override pptable id %d\n", pptable_id); + } else { + pptable_id = smu_table->boot_values.pp_table_id; + } + + /* + * Temporary solution for SMU V13.0.0 with SCPM disabled: + * - use vbios carried pptable when pptable_id is 3664, 3715 or 3795 + * - use soft pptable when pptable_id is 3683 + */ + if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 0)) { + switch (pptable_id) { + case 3664: + case 3715: + case 3795: + pptable_id = 0; + break; + case 3683: + break; + default: + dev_err(adev->dev, "Unsupported pptable id %d\n", pptable_id); + return -EINVAL; + } + } + + /* force using vbios pptable in sriov mode */ + if ((amdgpu_sriov_vf(adev) || !pptable_id) && (amdgpu_emu_mode != 1)) + ret = smu_v13_0_0_get_pptable_from_pmfw(smu, + &smu_table->power_play_table, + &smu_table->power_play_table_size); + else + ret = smu_v13_0_get_pptable_from_firmware(smu, + &smu_table->power_play_table, + &smu_table->power_play_table_size, + pptable_id); } + if (ret) + return ret; ret = smu_v13_0_0_store_powerplay_table(smu); if (ret) @@ -1792,7 +1844,9 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .dump_pptable = smu_v13_0_0_dump_pptable, .init_microcode = smu_v13_0_init_microcode, .load_microcode = smu_v13_0_load_microcode, + .fini_microcode = smu_v13_0_fini_microcode, .init_smc_tables = smu_v13_0_0_init_smc_tables, + .fini_smc_tables = smu_v13_0_fini_smc_tables, .init_power = smu_v13_0_init_power, .fini_power = smu_v13_0_fini_power, .check_fw_status = smu_v13_0_check_fw_status, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c index 82d3718d8324..97e1d55dcaad 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c @@ -71,7 +71,6 @@ static struct cmn2asic_msg_mapping smu_v13_0_4_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(TestMessage, PPSMC_MSG_TestMessage, 1), MSG_MAP(GetSmuVersion, PPSMC_MSG_GetPmfwVersion, 1), MSG_MAP(GetDriverIfVersion, PPSMC_MSG_GetDriverIfVersion, 1), - MSG_MAP(EnableGfxOff, PPSMC_MSG_EnableGfxOff, 1), MSG_MAP(AllowGfxOff, PPSMC_MSG_AllowGfxOff, 1), MSG_MAP(DisallowGfxOff, PPSMC_MSG_DisallowGfxOff, 1), MSG_MAP(PowerDownVcn, PPSMC_MSG_PowerDownVcn, 1), @@ -199,6 +198,9 @@ static int smu_v13_0_4_fini_smc_tables(struct smu_context *smu) kfree(smu_table->watermarks_table); smu_table->watermarks_table = NULL; + kfree(smu_table->gpu_metrics_table); + smu_table->gpu_metrics_table = NULL; + return 0; } @@ -226,18 +228,6 @@ static int smu_v13_0_4_system_features_control(struct smu_context *smu, bool en) return ret; } -static int smu_v13_0_4_post_smu_init(struct smu_context *smu) -{ - struct amdgpu_device *adev = smu->adev; - int ret = 0; - - /* allow message will be sent after enable message */ - ret = smu_cmn_send_smc_msg(smu, SMU_MSG_EnableGfxOff, NULL); - if (ret) - dev_err(adev->dev, "Failed to Enable GfxOff!\n"); - return ret; -} - static ssize_t smu_v13_0_4_get_gpu_metrics(struct smu_context *smu, void **table) { @@ -1026,7 +1016,6 @@ static const struct pptable_funcs smu_v13_0_4_ppt_funcs = { .get_pp_feature_mask = smu_cmn_get_pp_feature_mask, .set_driver_table_location = smu_v13_0_set_driver_table_location, .gfx_off_control = smu_v13_0_gfx_off_control, - .post_init = smu_v13_0_4_post_smu_init, .mode2_reset = smu_v13_0_4_mode2_reset, .get_dpm_ultimate_freq = smu_v13_0_4_get_dpm_ultimate_freq, .od_edit_dpm_table = smu_v13_0_od_edit_dpm_table, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c index 47360ef5c175..66445964efbd 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c @@ -176,6 +176,9 @@ static int smu_v13_0_5_fini_smc_tables(struct smu_context *smu) kfree(smu_table->watermarks_table); smu_table->watermarks_table = NULL; + kfree(smu_table->gpu_metrics_table); + smu_table->gpu_metrics_table = NULL; + return 0; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index 9dd56e73218b..fcf24c540859 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -400,11 +400,27 @@ static int smu_v13_0_7_append_powerplay_table(struct smu_context *smu) return 0; } +static int smu_v13_0_7_get_pptable_from_pmfw(struct smu_context *smu, + void **table, + uint32_t *size) +{ + struct smu_table_context *smu_table = &smu->smu_table; + void *combo_pptable = smu_table->combo_pptable; + int ret = 0; + + ret = smu_cmn_get_combo_pptable(smu); + if (ret) + return ret; + + *table = combo_pptable; + *size = sizeof(struct smu_13_0_7_powerplay_table); + + return 0; +} static int smu_v13_0_7_setup_pptable(struct smu_context *smu) { struct smu_table_context *smu_table = &smu->smu_table; - void *combo_pptable = smu_table->combo_pptable; struct amdgpu_device *adev = smu->adev; int ret = 0; @@ -413,18 +429,11 @@ static int smu_v13_0_7_setup_pptable(struct smu_context *smu) * be used directly by driver. To get the raw pptable, we need to * rely on the combo pptable(and its revelant SMU message). */ - if (adev->scpm_enabled) { - ret = smu_cmn_get_combo_pptable(smu); - if (ret) - return ret; - - smu->smu_table.power_play_table = combo_pptable; - smu->smu_table.power_play_table_size = sizeof(struct smu_13_0_7_powerplay_table); - } else { - ret = smu_v13_0_setup_pptable(smu); - if (ret) - return ret; - } + ret = smu_v13_0_7_get_pptable_from_pmfw(smu, + &smu_table->power_play_table, + &smu_table->power_play_table_size); + if (ret) + return ret; ret = smu_v13_0_7_store_powerplay_table(smu); if (ret) @@ -1567,6 +1576,16 @@ static int smu_v13_0_7_set_mp1_state(struct smu_context *smu, return ret; } +static bool smu_v13_0_7_is_mode1_reset_supported(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + + /* SRIOV does not support SMU mode1 reset */ + if (amdgpu_sriov_vf(adev)) + return false; + + return true; +} static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_7_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_7_set_default_dpm_table, @@ -1574,7 +1593,9 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .dump_pptable = smu_v13_0_7_dump_pptable, .init_microcode = smu_v13_0_init_microcode, .load_microcode = smu_v13_0_load_microcode, + .fini_microcode = smu_v13_0_fini_microcode, .init_smc_tables = smu_v13_0_7_init_smc_tables, + .fini_smc_tables = smu_v13_0_fini_smc_tables, .init_power = smu_v13_0_init_power, .fini_power = smu_v13_0_fini_power, .check_fw_status = smu_v13_0_7_check_fw_status, @@ -1624,6 +1645,8 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .baco_set_state = smu_v13_0_baco_set_state, .baco_enter = smu_v13_0_baco_enter, .baco_exit = smu_v13_0_baco_exit, + .mode1_reset_is_support = smu_v13_0_7_is_mode1_reset_supported, + .mode1_reset = smu_v13_0_mode1_reset, .set_mp1_state = smu_v13_0_7_set_mp1_state, }; diff --git a/drivers/gpu/drm/bridge/lvds-codec.c b/drivers/gpu/drm/bridge/lvds-codec.c index 702ea803a743..39e7004de720 100644 --- a/drivers/gpu/drm/bridge/lvds-codec.c +++ b/drivers/gpu/drm/bridge/lvds-codec.c @@ -180,7 +180,7 @@ static int lvds_codec_probe(struct platform_device *pdev) of_node_put(bus_node); if (ret == -ENODEV) { dev_warn(dev, "missing 'data-mapping' DT property\n"); - } else if (ret) { + } else if (ret < 0) { dev_err(dev, "invalid 'data-mapping' DT property\n"); return ret; } else { diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index 86d670c71286..ad068865ba20 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -168,21 +168,6 @@ void drm_gem_private_object_init(struct drm_device *dev, } EXPORT_SYMBOL(drm_gem_private_object_init); -static void -drm_gem_remove_prime_handles(struct drm_gem_object *obj, struct drm_file *filp) -{ - /* - * Note: obj->dma_buf can't disappear as long as we still hold a - * handle reference in obj->handle_count. - */ - mutex_lock(&filp->prime.lock); - if (obj->dma_buf) { - drm_prime_remove_buf_handle_locked(&filp->prime, - obj->dma_buf); - } - mutex_unlock(&filp->prime.lock); -} - /** * drm_gem_object_handle_free - release resources bound to userspace handles * @obj: GEM object to clean up. @@ -253,7 +238,7 @@ drm_gem_object_release_handle(int id, void *ptr, void *data) if (obj->funcs->close) obj->funcs->close(obj, file_priv); - drm_gem_remove_prime_handles(obj, file_priv); + drm_prime_remove_buf_handle(&file_priv->prime, id); drm_vma_node_revoke(&obj->vma_node, file_priv); drm_gem_object_handle_put_unlocked(obj); diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h index 1fbbc19f1ac0..7bb98e6a446d 100644 --- a/drivers/gpu/drm/drm_internal.h +++ b/drivers/gpu/drm/drm_internal.h @@ -74,8 +74,8 @@ int drm_prime_fd_to_handle_ioctl(struct drm_device *dev, void *data, void drm_prime_init_file_private(struct drm_prime_file_private *prime_fpriv); void drm_prime_destroy_file_private(struct drm_prime_file_private *prime_fpriv); -void drm_prime_remove_buf_handle_locked(struct drm_prime_file_private *prime_fpriv, - struct dma_buf *dma_buf); +void drm_prime_remove_buf_handle(struct drm_prime_file_private *prime_fpriv, + uint32_t handle); /* drm_drv.c */ struct drm_minor *drm_minor_acquire(unsigned int minor_id); diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c index a3f180653b8b..eb09e86044c6 100644 --- a/drivers/gpu/drm/drm_prime.c +++ b/drivers/gpu/drm/drm_prime.c @@ -190,29 +190,33 @@ static int drm_prime_lookup_buf_handle(struct drm_prime_file_private *prime_fpri return -ENOENT; } -void drm_prime_remove_buf_handle_locked(struct drm_prime_file_private *prime_fpriv, - struct dma_buf *dma_buf) +void drm_prime_remove_buf_handle(struct drm_prime_file_private *prime_fpriv, + uint32_t handle) { struct rb_node *rb; - rb = prime_fpriv->dmabufs.rb_node; + mutex_lock(&prime_fpriv->lock); + + rb = prime_fpriv->handles.rb_node; while (rb) { struct drm_prime_member *member; - member = rb_entry(rb, struct drm_prime_member, dmabuf_rb); - if (member->dma_buf == dma_buf) { + member = rb_entry(rb, struct drm_prime_member, handle_rb); + if (member->handle == handle) { rb_erase(&member->handle_rb, &prime_fpriv->handles); rb_erase(&member->dmabuf_rb, &prime_fpriv->dmabufs); - dma_buf_put(dma_buf); + dma_buf_put(member->dma_buf); kfree(member); - return; - } else if (member->dma_buf < dma_buf) { + break; + } else if (member->handle < handle) { rb = rb->rb_right; } else { rb = rb->rb_left; } } + + mutex_unlock(&prime_fpriv->lock); } void drm_prime_init_file_private(struct drm_prime_file_private *prime_fpriv) diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c index 5dcfa7feffa9..885c74f60366 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi.c +++ b/drivers/gpu/drm/i915/display/icl_dsi.c @@ -2070,7 +2070,14 @@ void icl_dsi_init(struct drm_i915_private *dev_priv) else intel_dsi->ports = BIT(port); + if (drm_WARN_ON(&dev_priv->drm, intel_connector->panel.vbt.dsi.bl_ports & ~intel_dsi->ports)) + intel_connector->panel.vbt.dsi.bl_ports &= intel_dsi->ports; + intel_dsi->dcs_backlight_ports = intel_connector->panel.vbt.dsi.bl_ports; + + if (drm_WARN_ON(&dev_priv->drm, intel_connector->panel.vbt.dsi.cabc_ports & ~intel_dsi->ports)) + intel_connector->panel.vbt.dsi.cabc_ports &= intel_dsi->ports; + intel_dsi->dcs_cabc_ports = intel_connector->panel.vbt.dsi.cabc_ports; for_each_dsi_port(port, intel_dsi->ports) { diff --git a/drivers/gpu/drm/i915/display/intel_backlight.c b/drivers/gpu/drm/i915/display/intel_backlight.c index 110fc98ec280..f5e1d692976e 100644 --- a/drivers/gpu/drm/i915/display/intel_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_backlight.c @@ -16,6 +16,7 @@ #include "intel_dsi_dcs_backlight.h" #include "intel_panel.h" #include "intel_pci_config.h" +#include "intel_pps.h" /** * scale - scale values from one range to another @@ -971,26 +972,24 @@ int intel_backlight_device_register(struct intel_connector *connector) if (!name) return -ENOMEM; - bd = backlight_device_register(name, connector->base.kdev, connector, - &intel_backlight_device_ops, &props); - - /* - * Using the same name independent of the drm device or connector - * prevents registration of multiple backlight devices in the - * driver. However, we need to use the default name for backward - * compatibility. Use unique names for subsequent backlight devices as a - * fallback when the default name already exists. - */ - if (IS_ERR(bd) && PTR_ERR(bd) == -EEXIST) { + bd = backlight_device_get_by_name(name); + if (bd) { + put_device(&bd->dev); + /* + * Using the same name independent of the drm device or connector + * prevents registration of multiple backlight devices in the + * driver. However, we need to use the default name for backward + * compatibility. Use unique names for subsequent backlight devices as a + * fallback when the default name already exists. + */ kfree(name); name = kasprintf(GFP_KERNEL, "card%d-%s-backlight", i915->drm.primary->index, connector->base.name); if (!name) return -ENOMEM; - - bd = backlight_device_register(name, connector->base.kdev, connector, - &intel_backlight_device_ops, &props); } + bd = backlight_device_register(name, connector->base.kdev, connector, + &intel_backlight_device_ops, &props); if (IS_ERR(bd)) { drm_err(&i915->drm, @@ -1773,9 +1772,13 @@ void intel_backlight_init_funcs(struct intel_panel *panel) panel->backlight.pwm_funcs = &i9xx_pwm_funcs; } - if (connector->base.connector_type == DRM_MODE_CONNECTOR_eDP && - intel_dp_aux_init_backlight_funcs(connector) == 0) - return; + if (connector->base.connector_type == DRM_MODE_CONNECTOR_eDP) { + if (intel_dp_aux_init_backlight_funcs(connector) == 0) + return; + + if (!(dev_priv->quirks & QUIRK_NO_PPS_BACKLIGHT_POWER_HOOK)) + connector->panel.backlight.power = intel_pps_backlight_power; + } /* We're using a standard PWM backlight interface */ panel->backlight.funcs = &pwm_bl_funcs; diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c index 51dde5bfd956..198a2f4920cc 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.c +++ b/drivers/gpu/drm/i915/display/intel_bios.c @@ -1596,6 +1596,8 @@ static void parse_dsi_backlight_ports(struct drm_i915_private *i915, struct intel_panel *panel, enum port port) { + enum port port_bc = DISPLAY_VER(i915) >= 11 ? PORT_B : PORT_C; + if (!panel->vbt.dsi.config->dual_link || i915->vbt.version < 197) { panel->vbt.dsi.bl_ports = BIT(port); if (panel->vbt.dsi.config->cabc_supported) @@ -1609,11 +1611,11 @@ static void parse_dsi_backlight_ports(struct drm_i915_private *i915, panel->vbt.dsi.bl_ports = BIT(PORT_A); break; case DL_DCS_PORT_C: - panel->vbt.dsi.bl_ports = BIT(PORT_C); + panel->vbt.dsi.bl_ports = BIT(port_bc); break; default: case DL_DCS_PORT_A_AND_C: - panel->vbt.dsi.bl_ports = BIT(PORT_A) | BIT(PORT_C); + panel->vbt.dsi.bl_ports = BIT(PORT_A) | BIT(port_bc); break; } @@ -1625,12 +1627,12 @@ static void parse_dsi_backlight_ports(struct drm_i915_private *i915, panel->vbt.dsi.cabc_ports = BIT(PORT_A); break; case DL_DCS_PORT_C: - panel->vbt.dsi.cabc_ports = BIT(PORT_C); + panel->vbt.dsi.cabc_ports = BIT(port_bc); break; default: case DL_DCS_PORT_A_AND_C: panel->vbt.dsi.cabc_ports = - BIT(PORT_A) | BIT(PORT_C); + BIT(PORT_A) | BIT(port_bc); break; } } diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c index 79269d2c476b..3699869ab2db 100644 --- a/drivers/gpu/drm/i915/display/intel_bw.c +++ b/drivers/gpu/drm/i915/display/intel_bw.c @@ -404,15 +404,17 @@ static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel int clpchgroup; int j; - if (i < num_groups - 1) - bi_next = &dev_priv->max_bw[i + 1]; - clpchgroup = (sa->deburst * qi.deinterleave / num_channels) << i; - if (i < num_groups - 1 && clpchgroup < clperchgroup) - bi_next->num_planes = (ipqdepth - clpchgroup) / clpchgroup + 1; - else - bi_next->num_planes = 0; + if (i < num_groups - 1) { + bi_next = &dev_priv->max_bw[i + 1]; + + if (clpchgroup < clperchgroup) + bi_next->num_planes = (ipqdepth - clpchgroup) / + clpchgroup + 1; + else + bi_next->num_planes = 0; + } bi->num_qgv_points = qi.num_points; bi->num_psf_gv_points = qi.num_psf_points; diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 32292c0be2bd..ac90d455a7c7 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -5293,8 +5293,6 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp, intel_panel_init(intel_connector); - if (!(dev_priv->quirks & QUIRK_NO_PPS_BACKLIGHT_POWER_HOOK)) - intel_connector->panel.backlight.power = intel_pps_backlight_power; intel_backlight_setup(intel_connector, pipe); intel_edp_add_properties(intel_dp); diff --git a/drivers/gpu/drm/i915/display/intel_quirks.c b/drivers/gpu/drm/i915/display/intel_quirks.c index c8488f5ebd04..e415cd7c0b84 100644 --- a/drivers/gpu/drm/i915/display/intel_quirks.c +++ b/drivers/gpu/drm/i915/display/intel_quirks.c @@ -191,6 +191,9 @@ static struct intel_quirk intel_quirks[] = { /* ASRock ITX*/ { 0x3185, 0x1849, 0x2212, quirk_increase_ddi_disabled_time }, { 0x3184, 0x1849, 0x2212, quirk_increase_ddi_disabled_time }, + /* ECS Liva Q2 */ + { 0x3185, 0x1019, 0xa94d, quirk_increase_ddi_disabled_time }, + { 0x3184, 0x1019, 0xa94d, quirk_increase_ddi_disabled_time }, }; void intel_init_quirks(struct drm_i915_private *i915) diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c index b9b1fed99874..35136d26e517 100644 --- a/drivers/gpu/drm/i915/display/vlv_dsi.c +++ b/drivers/gpu/drm/i915/display/vlv_dsi.c @@ -1933,7 +1933,14 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv) else intel_dsi->ports = BIT(port); + if (drm_WARN_ON(&dev_priv->drm, intel_connector->panel.vbt.dsi.bl_ports & ~intel_dsi->ports)) + intel_connector->panel.vbt.dsi.bl_ports &= intel_dsi->ports; + intel_dsi->dcs_backlight_ports = intel_connector->panel.vbt.dsi.bl_ports; + + if (drm_WARN_ON(&dev_priv->drm, intel_connector->panel.vbt.dsi.cabc_ports & ~intel_dsi->ports)) + intel_connector->panel.vbt.dsi.cabc_ports &= intel_dsi->ports; + intel_dsi->dcs_cabc_ports = intel_connector->panel.vbt.dsi.cabc_ports; /* Create a DSI host (and a device) for each port. */ diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index ccec4055fde3..389e9f157ca5 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -268,7 +268,7 @@ static void __i915_gem_object_free_mmaps(struct drm_i915_gem_object *obj) */ void __i915_gem_object_pages_fini(struct drm_i915_gem_object *obj) { - assert_object_held(obj); + assert_object_held_shared(obj); if (!list_empty(&obj->vma.list)) { struct i915_vma *vma; @@ -331,15 +331,7 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915, continue; } - if (!i915_gem_object_trylock(obj, NULL)) { - /* busy, toss it back to the pile */ - if (llist_add(&obj->freed, &i915->mm.free_list)) - queue_delayed_work(i915->wq, &i915->mm.free_work, msecs_to_jiffies(10)); - continue; - } - __i915_gem_object_pages_fini(obj); - i915_gem_object_unlock(obj); __i915_gem_free_object(obj); /* But keep the pointer alive for RCU-protected lookups */ @@ -359,7 +351,7 @@ void i915_gem_flush_free_objects(struct drm_i915_private *i915) static void __i915_gem_free_work(struct work_struct *work) { struct drm_i915_private *i915 = - container_of(work, struct drm_i915_private, mm.free_work.work); + container_of(work, struct drm_i915_private, mm.free_work); i915_gem_flush_free_objects(i915); } @@ -391,7 +383,7 @@ static void i915_gem_free_object(struct drm_gem_object *gem_obj) */ if (llist_add(&obj->freed, &i915->mm.free_list)) - queue_delayed_work(i915->wq, &i915->mm.free_work, 0); + queue_work(i915->wq, &i915->mm.free_work); } void __i915_gem_object_flush_frontbuffer(struct drm_i915_gem_object *obj, @@ -745,7 +737,7 @@ bool i915_gem_object_needs_ccs_pages(struct drm_i915_gem_object *obj) void i915_gem_init__objects(struct drm_i915_private *i915) { - INIT_DELAYED_WORK(&i915->mm.free_work, __i915_gem_free_work); + INIT_WORK(&i915->mm.free_work, __i915_gem_free_work); } void i915_objects_module_exit(void) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h index 5cf36a130061..9f6b14ec189a 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h @@ -335,7 +335,6 @@ struct drm_i915_gem_object { #define I915_BO_READONLY BIT(7) #define I915_TILING_QUIRK_BIT 8 /* unknown swizzling; do not release! */ #define I915_BO_PROTECTED BIT(9) -#define I915_BO_WAS_BOUND_BIT 10 /** * @mem_flags - Mutable placement-related flags * @@ -616,6 +615,8 @@ struct drm_i915_gem_object { * pages were last acquired. */ bool dirty:1; + + u32 tlb; } mm; struct { diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c index 97c820eee115..8357dbdcab5c 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c @@ -6,14 +6,15 @@ #include <drm/drm_cache.h> +#include "gt/intel_gt.h" +#include "gt/intel_gt_pm.h" + #include "i915_drv.h" #include "i915_gem_object.h" #include "i915_scatterlist.h" #include "i915_gem_lmem.h" #include "i915_gem_mman.h" -#include "gt/intel_gt.h" - void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj, struct sg_table *pages, unsigned int sg_page_sizes) @@ -190,6 +191,18 @@ static void unmap_object(struct drm_i915_gem_object *obj, void *ptr) vunmap(ptr); } +static void flush_tlb_invalidate(struct drm_i915_gem_object *obj) +{ + struct drm_i915_private *i915 = to_i915(obj->base.dev); + struct intel_gt *gt = to_gt(i915); + + if (!obj->mm.tlb) + return; + + intel_gt_invalidate_tlb(gt, obj->mm.tlb); + obj->mm.tlb = 0; +} + struct sg_table * __i915_gem_object_unset_pages(struct drm_i915_gem_object *obj) { @@ -215,13 +228,7 @@ __i915_gem_object_unset_pages(struct drm_i915_gem_object *obj) __i915_gem_object_reset_page_iter(obj); obj->mm.page_sizes.phys = obj->mm.page_sizes.sg = 0; - if (test_and_clear_bit(I915_BO_WAS_BOUND_BIT, &obj->flags)) { - struct drm_i915_private *i915 = to_i915(obj->base.dev); - intel_wakeref_t wakeref; - - with_intel_runtime_pm_if_active(&i915->runtime_pm, wakeref) - intel_gt_invalidate_tlbs(to_gt(i915)); - } + flush_tlb_invalidate(obj); return pages; } diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c index 68c2b0d8f187..f435e06125aa 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.c +++ b/drivers/gpu/drm/i915/gt/intel_gt.c @@ -11,7 +11,9 @@ #include "pxp/intel_pxp.h" #include "i915_drv.h" +#include "i915_perf_oa_regs.h" #include "intel_context.h" +#include "intel_engine_pm.h" #include "intel_engine_regs.h" #include "intel_ggtt_gmch.h" #include "intel_gt.h" @@ -36,8 +38,6 @@ static void __intel_gt_init_early(struct intel_gt *gt) { spin_lock_init(>->irq_lock); - mutex_init(>->tlb_invalidate_lock); - INIT_LIST_HEAD(>->closed_vma); spin_lock_init(>->closed_lock); @@ -48,6 +48,8 @@ static void __intel_gt_init_early(struct intel_gt *gt) intel_gt_init_reset(gt); intel_gt_init_requests(gt); intel_gt_init_timelines(gt); + mutex_init(>->tlb.invalidate_lock); + seqcount_mutex_init(>->tlb.seqno, >->tlb.invalidate_lock); intel_gt_pm_init_early(gt); intel_uc_init_early(>->uc); @@ -768,6 +770,7 @@ void intel_gt_driver_late_release_all(struct drm_i915_private *i915) intel_gt_fini_requests(gt); intel_gt_fini_reset(gt); intel_gt_fini_timelines(gt); + mutex_destroy(>->tlb.invalidate_lock); intel_engines_free(gt); } } @@ -906,7 +909,7 @@ get_reg_and_bit(const struct intel_engine_cs *engine, const bool gen8, return rb; } -void intel_gt_invalidate_tlbs(struct intel_gt *gt) +static void mmio_invalidate_full(struct intel_gt *gt) { static const i915_reg_t gen8_regs[] = { [RENDER_CLASS] = GEN8_RTCR, @@ -924,13 +927,11 @@ void intel_gt_invalidate_tlbs(struct intel_gt *gt) struct drm_i915_private *i915 = gt->i915; struct intel_uncore *uncore = gt->uncore; struct intel_engine_cs *engine; + intel_engine_mask_t awake, tmp; enum intel_engine_id id; const i915_reg_t *regs; unsigned int num = 0; - if (I915_SELFTEST_ONLY(gt->awake == -ENODEV)) - return; - if (GRAPHICS_VER(i915) == 12) { regs = gen12_regs; num = ARRAY_SIZE(gen12_regs); @@ -945,28 +946,41 @@ void intel_gt_invalidate_tlbs(struct intel_gt *gt) "Platform does not implement TLB invalidation!")) return; - GEM_TRACE("\n"); - - assert_rpm_wakelock_held(&i915->runtime_pm); - - mutex_lock(>->tlb_invalidate_lock); intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL); spin_lock_irq(&uncore->lock); /* serialise invalidate with GT reset */ + awake = 0; for_each_engine(engine, gt, id) { struct reg_and_bit rb; + if (!intel_engine_pm_is_awake(engine)) + continue; + rb = get_reg_and_bit(engine, regs == gen8_regs, regs, num); if (!i915_mmio_reg_offset(rb.reg)) continue; intel_uncore_write_fw(uncore, rb.reg, rb.bit); + awake |= engine->mask; } + GT_TRACE(gt, "invalidated engines %08x\n", awake); + + /* Wa_2207587034:tgl,dg1,rkl,adl-s,adl-p */ + if (awake && + (IS_TIGERLAKE(i915) || + IS_DG1(i915) || + IS_ROCKETLAKE(i915) || + IS_ALDERLAKE_S(i915) || + IS_ALDERLAKE_P(i915))) + intel_uncore_write_fw(uncore, GEN12_OA_TLB_INV_CR, 1); + spin_unlock_irq(&uncore->lock); - for_each_engine(engine, gt, id) { + for_each_engine_masked(engine, gt, awake, tmp) { + struct reg_and_bit rb; + /* * HW architecture suggest typical invalidation time at 40us, * with pessimistic cases up to 100us and a recommendation to @@ -974,12 +988,8 @@ void intel_gt_invalidate_tlbs(struct intel_gt *gt) */ const unsigned int timeout_us = 100; const unsigned int timeout_ms = 4; - struct reg_and_bit rb; rb = get_reg_and_bit(engine, regs == gen8_regs, regs, num); - if (!i915_mmio_reg_offset(rb.reg)) - continue; - if (__intel_wait_for_register_fw(uncore, rb.reg, rb.bit, 0, timeout_us, timeout_ms, @@ -996,5 +1006,38 @@ void intel_gt_invalidate_tlbs(struct intel_gt *gt) * transitions. */ intel_uncore_forcewake_put_delayed(uncore, FORCEWAKE_ALL); - mutex_unlock(>->tlb_invalidate_lock); +} + +static bool tlb_seqno_passed(const struct intel_gt *gt, u32 seqno) +{ + u32 cur = intel_gt_tlb_seqno(gt); + + /* Only skip if a *full* TLB invalidate barrier has passed */ + return (s32)(cur - ALIGN(seqno, 2)) > 0; +} + +void intel_gt_invalidate_tlb(struct intel_gt *gt, u32 seqno) +{ + intel_wakeref_t wakeref; + + if (I915_SELFTEST_ONLY(gt->awake == -ENODEV)) + return; + + if (intel_gt_is_wedged(gt)) + return; + + if (tlb_seqno_passed(gt, seqno)) + return; + + with_intel_gt_pm_if_awake(gt, wakeref) { + mutex_lock(>->tlb.invalidate_lock); + if (tlb_seqno_passed(gt, seqno)) + goto unlock; + + mmio_invalidate_full(gt); + + write_seqcount_invalidate(>->tlb.seqno); +unlock: + mutex_unlock(>->tlb.invalidate_lock); + } } diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h b/drivers/gpu/drm/i915/gt/intel_gt.h index 82d6f248d876..40b06adf509a 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.h +++ b/drivers/gpu/drm/i915/gt/intel_gt.h @@ -101,6 +101,16 @@ void intel_gt_info_print(const struct intel_gt_info *info, void intel_gt_watchdog_work(struct work_struct *work); -void intel_gt_invalidate_tlbs(struct intel_gt *gt); +static inline u32 intel_gt_tlb_seqno(const struct intel_gt *gt) +{ + return seqprop_sequence(>->tlb.seqno); +} + +static inline u32 intel_gt_next_invalidate_tlb_full(const struct intel_gt *gt) +{ + return intel_gt_tlb_seqno(gt) | 1; +} + +void intel_gt_invalidate_tlb(struct intel_gt *gt, u32 seqno); #endif /* __INTEL_GT_H__ */ diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.h b/drivers/gpu/drm/i915/gt/intel_gt_pm.h index bc898df7a48c..a334787a4939 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_pm.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.h @@ -55,6 +55,9 @@ static inline void intel_gt_pm_might_put(struct intel_gt *gt) for (tmp = 1, intel_gt_pm_get(gt); tmp; \ intel_gt_pm_put(gt), tmp = 0) +#define with_intel_gt_pm_if_awake(gt, wf) \ + for (wf = intel_gt_pm_get_if_awake(gt); wf; intel_gt_pm_put_async(gt), wf = 0) + static inline int intel_gt_pm_wait_for_idle(struct intel_gt *gt) { return intel_wakeref_wait_for_idle(>->wakeref); diff --git a/drivers/gpu/drm/i915/gt/intel_gt_types.h b/drivers/gpu/drm/i915/gt/intel_gt_types.h index df708802889d..3804a583382b 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_types.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_types.h @@ -11,6 +11,7 @@ #include <linux/llist.h> #include <linux/mutex.h> #include <linux/notifier.h> +#include <linux/seqlock.h> #include <linux/spinlock.h> #include <linux/types.h> #include <linux/workqueue.h> @@ -83,7 +84,22 @@ struct intel_gt { struct intel_uc uc; struct intel_gsc gsc; - struct mutex tlb_invalidate_lock; + struct { + /* Serialize global tlb invalidations */ + struct mutex invalidate_lock; + + /* + * Batch TLB invalidations + * + * After unbinding the PTE, we need to ensure the TLB + * are invalidated prior to releasing the physical pages. + * But we only need one such invalidation for all unbinds, + * so we track how many TLB invalidations have been + * performed since unbind the PTE and only emit an extra + * invalidate if no full barrier has been passed. + */ + seqcount_mutex_t seqno; + } tlb; struct i915_wa_list wa_list; diff --git a/drivers/gpu/drm/i915/gt/intel_migrate.c b/drivers/gpu/drm/i915/gt/intel_migrate.c index 2c35324b5f68..933648cc90ff 100644 --- a/drivers/gpu/drm/i915/gt/intel_migrate.c +++ b/drivers/gpu/drm/i915/gt/intel_migrate.c @@ -638,9 +638,9 @@ static int emit_copy(struct i915_request *rq, return 0; } -static int scatter_list_length(struct scatterlist *sg) +static u64 scatter_list_length(struct scatterlist *sg) { - int len = 0; + u64 len = 0; while (sg && sg_dma_len(sg)) { len += sg_dma_len(sg); @@ -650,28 +650,26 @@ static int scatter_list_length(struct scatterlist *sg) return len; } -static void +static int calculate_chunk_sz(struct drm_i915_private *i915, bool src_is_lmem, - int *src_sz, u32 bytes_to_cpy, u32 ccs_bytes_to_cpy) + u64 bytes_to_cpy, u64 ccs_bytes_to_cpy) { - if (ccs_bytes_to_cpy) { - if (!src_is_lmem) - /* - * When CHUNK_SZ is passed all the pages upto CHUNK_SZ - * will be taken for the blt. in Flat-ccs supported - * platform Smem obj will have more pages than required - * for main meory hence limit it to the required size - * for main memory - */ - *src_sz = min_t(int, bytes_to_cpy, CHUNK_SZ); - } else { /* ccs handling is not required */ - *src_sz = CHUNK_SZ; - } + if (ccs_bytes_to_cpy && !src_is_lmem) + /* + * When CHUNK_SZ is passed all the pages upto CHUNK_SZ + * will be taken for the blt. in Flat-ccs supported + * platform Smem obj will have more pages than required + * for main meory hence limit it to the required size + * for main memory + */ + return min_t(u64, bytes_to_cpy, CHUNK_SZ); + else + return CHUNK_SZ; } -static void get_ccs_sg_sgt(struct sgt_dma *it, u32 bytes_to_cpy) +static void get_ccs_sg_sgt(struct sgt_dma *it, u64 bytes_to_cpy) { - u32 len; + u64 len; do { GEM_BUG_ON(!it->sg || !sg_dma_len(it->sg)); @@ -702,13 +700,13 @@ intel_context_migrate_copy(struct intel_context *ce, { struct sgt_dma it_src = sg_sgt(src), it_dst = sg_sgt(dst), it_ccs; struct drm_i915_private *i915 = ce->engine->i915; - u32 ccs_bytes_to_cpy = 0, bytes_to_cpy; + u64 ccs_bytes_to_cpy = 0, bytes_to_cpy; enum i915_cache_level ccs_cache_level; u32 src_offset, dst_offset; u8 src_access, dst_access; struct i915_request *rq; - int src_sz, dst_sz; - bool ccs_is_src; + u64 src_sz, dst_sz; + bool ccs_is_src, overwrite_ccs; int err; GEM_BUG_ON(ce->vm != ce->engine->gt->migrate.context->vm); @@ -749,6 +747,8 @@ intel_context_migrate_copy(struct intel_context *ce, get_ccs_sg_sgt(&it_ccs, bytes_to_cpy); } + overwrite_ccs = HAS_FLAT_CCS(i915) && !ccs_bytes_to_cpy && dst_is_lmem; + src_offset = 0; dst_offset = CHUNK_SZ; if (HAS_64K_PAGES(ce->engine->i915)) { @@ -788,8 +788,8 @@ intel_context_migrate_copy(struct intel_context *ce, if (err) goto out_rq; - calculate_chunk_sz(i915, src_is_lmem, &src_sz, - bytes_to_cpy, ccs_bytes_to_cpy); + src_sz = calculate_chunk_sz(i915, src_is_lmem, + bytes_to_cpy, ccs_bytes_to_cpy); len = emit_pte(rq, &it_src, src_cache_level, src_is_lmem, src_offset, src_sz); @@ -852,6 +852,25 @@ intel_context_migrate_copy(struct intel_context *ce, if (err) goto out_rq; ccs_bytes_to_cpy -= ccs_sz; + } else if (overwrite_ccs) { + err = rq->engine->emit_flush(rq, EMIT_INVALIDATE); + if (err) + goto out_rq; + + /* + * While we can't always restore/manage the CCS state, + * we still need to ensure we don't leak the CCS state + * from the previous user, so make sure we overwrite it + * with something. + */ + err = emit_copy_ccs(rq, dst_offset, INDIRECT_ACCESS, + dst_offset, DIRECT_ACCESS, len); + if (err) + goto out_rq; + + err = rq->engine->emit_flush(rq, EMIT_INVALIDATE); + if (err) + goto out_rq; } /* Arbitration is re-enabled between requests. */ diff --git a/drivers/gpu/drm/i915/gt/intel_ppgtt.c b/drivers/gpu/drm/i915/gt/intel_ppgtt.c index d8b94d638559..6ee8d1127016 100644 --- a/drivers/gpu/drm/i915/gt/intel_ppgtt.c +++ b/drivers/gpu/drm/i915/gt/intel_ppgtt.c @@ -206,8 +206,12 @@ void ppgtt_bind_vma(struct i915_address_space *vm, void ppgtt_unbind_vma(struct i915_address_space *vm, struct i915_vma_resource *vma_res) { - if (vma_res->allocated) - vm->clear_range(vm, vma_res->start, vma_res->vma_size); + if (!vma_res->allocated) + return; + + vm->clear_range(vm, vma_res->start, vma_res->vma_size); + if (vma_res->tlb) + vma_invalidate_tlb(vm, vma_res->tlb); } static unsigned long pd_count(u64 size, int shift) diff --git a/drivers/gpu/drm/i915/gt/intel_region_lmem.c b/drivers/gpu/drm/i915/gt/intel_region_lmem.c index 6e90032e12e9..aa6aed837194 100644 --- a/drivers/gpu/drm/i915/gt/intel_region_lmem.c +++ b/drivers/gpu/drm/i915/gt/intel_region_lmem.c @@ -15,6 +15,7 @@ #include "gt/intel_gt_mcr.h" #include "gt/intel_gt_regs.h" +#ifdef CONFIG_64BIT static void _release_bars(struct pci_dev *pdev) { int resno; @@ -111,6 +112,9 @@ static void i915_resize_lmem_bar(struct drm_i915_private *i915, resource_size_t pci_assign_unassigned_bus_resources(pdev->bus); pci_write_config_dword(pdev, PCI_COMMAND, pci_cmd); } +#else +static void i915_resize_lmem_bar(struct drm_i915_private *i915, resource_size_t lmem_size) {} +#endif static int region_lmem_release(struct intel_memory_region *mem) diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index 76916aed897a..834c707d1877 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -4027,6 +4027,13 @@ static inline void guc_init_lrc_mapping(struct intel_guc *guc) xa_destroy(&guc->context_lookup); /* + * A reset might have occurred while we had a pending stalled request, + * so make sure we clean that up. + */ + guc->stalled_request = NULL; + guc->submission_stall_reason = STALL_NONE; + + /* * Some contexts might have been pinned before we enabled GuC * submission, so we need to add them to the GuC bookeeping. * Also, after a reset the of the GuC we want to make sure that the diff --git a/drivers/gpu/drm/i915/gvt/aperture_gm.c b/drivers/gpu/drm/i915/gvt/aperture_gm.c index 557f3314291a..3b81a6d35a7b 100644 --- a/drivers/gpu/drm/i915/gvt/aperture_gm.c +++ b/drivers/gpu/drm/i915/gvt/aperture_gm.c @@ -298,7 +298,7 @@ no_enough_resource: } /** - * inte_gvt_free_vgpu_resource - free HW resource owned by a vGPU + * intel_vgpu_free_resource() - free HW resource owned by a vGPU * @vgpu: a vGPU * * This function is used to free the HW resource owned by a vGPU. @@ -328,7 +328,7 @@ void intel_vgpu_reset_resource(struct intel_vgpu *vgpu) } /** - * intel_alloc_vgpu_resource - allocate HW resource for a vGPU + * intel_vgpu_alloc_resource() - allocate HW resource for a vGPU * @vgpu: vGPU * @param: vGPU creation params * diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index b4f69364f9a1..ce0eb03709c3 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -2341,7 +2341,7 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off, gvt_vgpu_err("fail to populate guest ggtt entry\n"); /* guest driver may read/write the entry when partial * update the entry in this situation p2m will fail - * settting the shadow entry to point to a scratch page + * setting the shadow entry to point to a scratch page */ ops->set_pfn(&m, gvt->gtt.scratch_mfn); } else diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index beea5895e499..61423da36710 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -905,7 +905,7 @@ static int update_fdi_rx_iir_status(struct intel_vgpu *vgpu, else if (FDI_RX_IMR_TO_PIPE(offset) != INVALID_INDEX) index = FDI_RX_IMR_TO_PIPE(offset); else { - gvt_vgpu_err("Unsupport registers %x\n", offset); + gvt_vgpu_err("Unsupported registers %x\n", offset); return -EINVAL; } @@ -3052,7 +3052,7 @@ int intel_vgpu_default_mmio_read(struct intel_vgpu *vgpu, unsigned int offset, } /** - * intel_t_default_mmio_write - default MMIO write handler + * intel_vgpu_default_mmio_write() - default MMIO write handler * @vgpu: a vGPU * @offset: access offset * @p_data: write data buffer diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c index c85bafe7539e..1c6e941c9666 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.c +++ b/drivers/gpu/drm/i915/gvt/mmio_context.c @@ -546,7 +546,7 @@ static void switch_mmio(struct intel_vgpu *pre, } /** - * intel_gvt_switch_render_mmio - switch mmio context of specific engine + * intel_gvt_switch_mmio - switch mmio context of specific engine * @pre: the last vGPU that own the engine * @next: the vGPU to switch to * @engine: the engine diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index d25647be25d1..086bbe8945d6 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -247,7 +247,7 @@ struct i915_gem_mm { * List of objects which are pending destruction. */ struct llist_head free_list; - struct delayed_work free_work; + struct work_struct free_work; /** * Count of objects pending destructions. Used to skip needlessly * waiting on an RCU barrier if no objects are waiting to be freed. @@ -1378,7 +1378,7 @@ static inline void i915_gem_drain_freed_objects(struct drm_i915_private *i915) * armed the work again. */ while (atomic_read(&i915->mm.free_count)) { - flush_delayed_work(&i915->mm.free_work); + flush_work(&i915->mm.free_work); flush_delayed_work(&i915->bdev.wq); rcu_barrier(); } diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index ef3b04c7e153..260371716490 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -538,8 +538,6 @@ int i915_vma_bind(struct i915_vma *vma, bind_flags); } - set_bit(I915_BO_WAS_BOUND_BIT, &vma->obj->flags); - atomic_or(bind_flags, &vma->flags); return 0; } @@ -1310,6 +1308,19 @@ err_unpin: return err; } +void vma_invalidate_tlb(struct i915_address_space *vm, u32 *tlb) +{ + /* + * Before we release the pages that were bound by this vma, we + * must invalidate all the TLBs that may still have a reference + * back to our physical address. It only needs to be done once, + * so after updating the PTE to point away from the pages, record + * the most recent TLB invalidation seqno, and if we have not yet + * flushed the TLBs upon release, perform a full invalidation. + */ + WRITE_ONCE(*tlb, intel_gt_next_invalidate_tlb_full(vm->gt)); +} + static void __vma_put_pages(struct i915_vma *vma, unsigned int count) { /* We allocate under vma_get_pages, so beware the shrinker */ @@ -1941,7 +1952,12 @@ struct dma_fence *__i915_vma_evict(struct i915_vma *vma, bool async) vma->vm->skip_pte_rewrite; trace_i915_vma_unbind(vma); - unbind_fence = i915_vma_resource_unbind(vma_res); + if (async) + unbind_fence = i915_vma_resource_unbind(vma_res, + &vma->obj->mm.tlb); + else + unbind_fence = i915_vma_resource_unbind(vma_res, NULL); + vma->resource = NULL; atomic_and(~(I915_VMA_BIND_MASK | I915_VMA_ERROR | I915_VMA_GGTT_WRITE), @@ -1949,10 +1965,13 @@ struct dma_fence *__i915_vma_evict(struct i915_vma *vma, bool async) i915_vma_detach(vma); - if (!async && unbind_fence) { - dma_fence_wait(unbind_fence, false); - dma_fence_put(unbind_fence); - unbind_fence = NULL; + if (!async) { + if (unbind_fence) { + dma_fence_wait(unbind_fence, false); + dma_fence_put(unbind_fence); + unbind_fence = NULL; + } + vma_invalidate_tlb(vma->vm, &vma->obj->mm.tlb); } /* diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h index 88ca0bd9c900..33a58f605d75 100644 --- a/drivers/gpu/drm/i915/i915_vma.h +++ b/drivers/gpu/drm/i915/i915_vma.h @@ -213,6 +213,7 @@ bool i915_vma_misplaced(const struct i915_vma *vma, u64 size, u64 alignment, u64 flags); void __i915_vma_set_map_and_fenceable(struct i915_vma *vma); void i915_vma_revoke_mmap(struct i915_vma *vma); +void vma_invalidate_tlb(struct i915_address_space *vm, u32 *tlb); struct dma_fence *__i915_vma_evict(struct i915_vma *vma, bool async); int __i915_vma_unbind(struct i915_vma *vma); int __must_check i915_vma_unbind(struct i915_vma *vma); diff --git a/drivers/gpu/drm/i915/i915_vma_resource.c b/drivers/gpu/drm/i915/i915_vma_resource.c index 27c55027387a..5a67995ea5fe 100644 --- a/drivers/gpu/drm/i915/i915_vma_resource.c +++ b/drivers/gpu/drm/i915/i915_vma_resource.c @@ -223,10 +223,13 @@ i915_vma_resource_fence_notify(struct i915_sw_fence *fence, * Return: A refcounted pointer to a dma-fence that signals when unbinding is * complete. */ -struct dma_fence *i915_vma_resource_unbind(struct i915_vma_resource *vma_res) +struct dma_fence *i915_vma_resource_unbind(struct i915_vma_resource *vma_res, + u32 *tlb) { struct i915_address_space *vm = vma_res->vm; + vma_res->tlb = tlb; + /* Reference for the sw fence */ i915_vma_resource_get(vma_res); diff --git a/drivers/gpu/drm/i915/i915_vma_resource.h b/drivers/gpu/drm/i915/i915_vma_resource.h index 5d8427caa2ba..06923d1816e7 100644 --- a/drivers/gpu/drm/i915/i915_vma_resource.h +++ b/drivers/gpu/drm/i915/i915_vma_resource.h @@ -67,6 +67,7 @@ struct i915_page_sizes { * taken when the unbind is scheduled. * @skip_pte_rewrite: During ggtt suspend and vm takedown pte rewriting * needs to be skipped for unbind. + * @tlb: pointer for obj->mm.tlb, if async unbind. Otherwise, NULL * * The lifetime of a struct i915_vma_resource is from a binding request to * the actual possible asynchronous unbind has completed. @@ -119,6 +120,8 @@ struct i915_vma_resource { bool immediate_unbind:1; bool needs_wakeref:1; bool skip_pte_rewrite:1; + + u32 *tlb; }; bool i915_vma_resource_hold(struct i915_vma_resource *vma_res, @@ -131,7 +134,8 @@ struct i915_vma_resource *i915_vma_resource_alloc(void); void i915_vma_resource_free(struct i915_vma_resource *vma_res); -struct dma_fence *i915_vma_resource_unbind(struct i915_vma_resource *vma_res); +struct dma_fence *i915_vma_resource_unbind(struct i915_vma_resource *vma_res, + u32 *tlb); void __i915_vma_resource_init(struct i915_vma_resource *vma_res); diff --git a/drivers/gpu/drm/i915/intel_gvt_mmio_table.c b/drivers/gpu/drm/i915/intel_gvt_mmio_table.c index 157e166672d7..5595639d0033 100644 --- a/drivers/gpu/drm/i915/intel_gvt_mmio_table.c +++ b/drivers/gpu/drm/i915/intel_gvt_mmio_table.c @@ -1076,7 +1076,8 @@ static int iterate_skl_plus_mmio(struct intel_gvt_mmio_table_iter *iter) MMIO_D(GEN8_HDC_CHICKEN1); MMIO_D(GEN9_WM_CHICKEN3); - if (IS_KABYLAKE(dev_priv) || IS_COFFEELAKE(dev_priv)) + if (IS_KABYLAKE(dev_priv) || + IS_COFFEELAKE(dev_priv) || IS_COMETLAKE(dev_priv)) MMIO_D(GAMT_CHKN_BIT_REG); if (!IS_BROXTON(dev_priv)) MMIO_D(GEN9_CTX_PREEMPT_REG); diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index f06babdb3a8c..9fe4b583cc28 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -6561,7 +6561,10 @@ void skl_wm_get_hw_state(struct drm_i915_private *dev_priv) enum plane_id plane_id; u8 slices; - skl_pipe_wm_get_hw_state(crtc, &crtc_state->wm.skl.optimal); + memset(&crtc_state->wm.skl.optimal, 0, + sizeof(crtc_state->wm.skl.optimal)); + if (crtc_state->hw.active) + skl_pipe_wm_get_hw_state(crtc, &crtc_state->wm.skl.optimal); crtc_state->wm.skl.raw = crtc_state->wm.skl.optimal; memset(&dbuf_state->ddb[pipe], 0, sizeof(dbuf_state->ddb[pipe])); @@ -6572,6 +6575,9 @@ void skl_wm_get_hw_state(struct drm_i915_private *dev_priv) struct skl_ddb_entry *ddb_y = &crtc_state->wm.skl.plane_ddb_y[plane_id]; + if (!crtc_state->hw.active) + continue; + skl_ddb_get_hw_plane_state(dev_priv, crtc->pipe, plane_id, ddb, ddb_y); diff --git a/drivers/gpu/drm/imx/dcss/dcss-kms.c b/drivers/gpu/drm/imx/dcss/dcss-kms.c index 9b84df34a6a1..8cf3352d8858 100644 --- a/drivers/gpu/drm/imx/dcss/dcss-kms.c +++ b/drivers/gpu/drm/imx/dcss/dcss-kms.c @@ -142,8 +142,6 @@ struct dcss_kms_dev *dcss_kms_attach(struct dcss_dev *dcss) drm_kms_helper_poll_init(drm); - drm_bridge_connector_enable_hpd(kms->connector); - ret = drm_dev_register(drm, 0); if (ret) goto cleanup_crtc; diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c index 1b70938cfd2c..bd4ca11d3ff5 100644 --- a/drivers/gpu/drm/meson/meson_drv.c +++ b/drivers/gpu/drm/meson/meson_drv.c @@ -115,8 +115,11 @@ static bool meson_vpu_has_available_connectors(struct device *dev) for_each_endpoint_of_node(dev->of_node, ep) { /* If the endpoint node exists, consider it enabled */ remote = of_graph_get_remote_port(ep); - if (remote) + if (remote) { + of_node_put(remote); + of_node_put(ep); return true; + } } return false; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c index c682d4e02d1b..52a626117f70 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -2061,6 +2061,12 @@ void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc) intf_cfg.stream_sel = 0; /* Don't care value for video mode */ intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); + + if (phys_enc->hw_intf) + intf_cfg.intf = phys_enc->hw_intf->idx; + if (phys_enc->hw_wb) + intf_cfg.wb = phys_enc->hw_wb->idx; + if (phys_enc->hw_pp->merge_3d) intf_cfg.merge_3d = phys_enc->hw_pp->merge_3d->idx; diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.c b/drivers/gpu/drm/msm/dp/dp_ctrl.c index ab6aa13b1639..013ca02e17cb 100644 --- a/drivers/gpu/drm/msm/dp/dp_ctrl.c +++ b/drivers/gpu/drm/msm/dp/dp_ctrl.c @@ -1214,7 +1214,7 @@ static int dp_ctrl_link_train_2(struct dp_ctrl_private *ctrl, if (ret) return ret; - dp_ctrl_train_pattern_set(ctrl, pattern | DP_RECOVERED_CLOCK_OUT_EN); + dp_ctrl_train_pattern_set(ctrl, pattern); for (tries = 0; tries <= maximum_retries; tries++) { drm_dp_link_train_channel_eq_delay(ctrl->aux, ctrl->panel->dpcd); diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c index 2c23324a2296..72c018e26f47 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c @@ -109,7 +109,7 @@ static const char * const dsi_8996_bus_clk_names[] = { static const struct msm_dsi_config msm8996_dsi_cfg = { .io_offset = DSI_6G_REG_SHIFT, .reg_cfg = { - .num = 2, + .num = 3, .regs = { {"vdda", 18160, 1 }, /* 1.25 V */ {"vcca", 17000, 32 }, /* 0.925 V */ @@ -148,7 +148,7 @@ static const char * const dsi_sdm660_bus_clk_names[] = { static const struct msm_dsi_config sdm660_dsi_cfg = { .io_offset = DSI_6G_REG_SHIFT, .reg_cfg = { - .num = 2, + .num = 1, .regs = { {"vdda", 12560, 4 }, /* 1.2 V */ }, diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c index a39de3bdc7fa..56dfa2d24be1 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c @@ -347,7 +347,7 @@ int msm_dsi_dphy_timing_calc_v3(struct msm_dsi_dphy_timing *timing, } else { timing->shared_timings.clk_pre = linear_inter(tmax, tmin, pcnt2, 0, false); - timing->shared_timings.clk_pre_inc_by_2 = 0; + timing->shared_timings.clk_pre_inc_by_2 = 0; } timing->ta_go = 3; diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 1ed4cd09dbf8..16884db272de 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -469,6 +469,8 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) } } + drm_helper_move_panel_connectors_to_head(ddev); + ddev->mode_config.funcs = &mode_config_funcs; ddev->mode_config.helper_private = &mode_config_helper_funcs; diff --git a/drivers/gpu/drm/msm/msm_gpu_devfreq.c b/drivers/gpu/drm/msm/msm_gpu_devfreq.c index d1f70426f554..85c443a37e4e 100644 --- a/drivers/gpu/drm/msm/msm_gpu_devfreq.c +++ b/drivers/gpu/drm/msm/msm_gpu_devfreq.c @@ -213,6 +213,8 @@ void msm_devfreq_init(struct msm_gpu *gpu) if (IS_ERR(df->devfreq)) { DRM_DEV_ERROR(&gpu->pdev->dev, "Couldn't initialize GPU devfreq\n"); + dev_pm_qos_remove_request(&df->idle_freq); + dev_pm_qos_remove_request(&df->boost_freq); df->devfreq = NULL; return; } diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index a92ffde53f0b..db2f847c8535 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -196,6 +196,9 @@ static int rd_open(struct inode *inode, struct file *file) file->private_data = rd; rd->open = true; + /* Reset fifo to clear any previously unread data: */ + rd->fifo.head = rd->fifo.tail = 0; + /* the parsing tools need to know gpu-id to know which * register database to load. * diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 05076e530e7d..e29175e4b44c 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -820,6 +820,15 @@ nouveau_bo_move_m2mf(struct ttm_buffer_object *bo, int evict, if (ret == 0) { ret = nouveau_fence_new(chan, false, &fence); if (ret == 0) { + /* TODO: figure out a better solution here + * + * wait on the fence here explicitly as going through + * ttm_bo_move_accel_cleanup somehow doesn't seem to do it. + * + * Without this the operation can timeout and we'll fallback to a + * software copy, which might take several minutes to finish. + */ + nouveau_fence_wait(fence, false, false); ret = ttm_bo_move_accel_cleanup(bo, &fence->base, evict, false, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c index 568182e68dd7..d8cf71fb0512 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c @@ -2605,6 +2605,27 @@ nv172_chipset = { }; static const struct nvkm_device_chip +nv173_chipset = { + .name = "GA103", + .bar = { 0x00000001, tu102_bar_new }, + .bios = { 0x00000001, nvkm_bios_new }, + .devinit = { 0x00000001, ga100_devinit_new }, + .fb = { 0x00000001, ga102_fb_new }, + .gpio = { 0x00000001, ga102_gpio_new }, + .i2c = { 0x00000001, gm200_i2c_new }, + .imem = { 0x00000001, nv50_instmem_new }, + .mc = { 0x00000001, ga100_mc_new }, + .mmu = { 0x00000001, tu102_mmu_new }, + .pci = { 0x00000001, gp100_pci_new }, + .privring = { 0x00000001, gm200_privring_new }, + .timer = { 0x00000001, gk20a_timer_new }, + .top = { 0x00000001, ga100_top_new }, + .disp = { 0x00000001, ga102_disp_new }, + .dma = { 0x00000001, gv100_dma_new }, + .fifo = { 0x00000001, ga102_fifo_new }, +}; + +static const struct nvkm_device_chip nv174_chipset = { .name = "GA104", .bar = { 0x00000001, tu102_bar_new }, @@ -3067,6 +3088,7 @@ nvkm_device_ctor(const struct nvkm_device_func *func, case 0x167: device->chip = &nv167_chipset; break; case 0x168: device->chip = &nv168_chipset; break; case 0x172: device->chip = &nv172_chipset; break; + case 0x173: device->chip = &nv173_chipset; break; case 0x174: device->chip = &nv174_chipset; break; case 0x176: device->chip = &nv176_chipset; break; case 0x177: device->chip = &nv177_chipset; break; diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 2b12389f841a..ee0165687239 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1605,6 +1605,9 @@ int radeon_suspend_kms(struct drm_device *dev, bool suspend, if (r) { /* delay GPU reset to resume */ radeon_fence_driver_force_completion(rdev, i); + } else { + /* finish executing delayed work */ + flush_delayed_work(&rdev->fence_drv[i].lockup_work); } } diff --git a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c index b4dfa166eccd..34234a144e87 100644 --- a/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c +++ b/drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c @@ -531,7 +531,7 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi, struct drm_display_mode *mode) { struct mipi_dsi_device *device = dsi->device; - unsigned int Bpp = mipi_dsi_pixel_format_to_bpp(device->format) / 8; + int Bpp = mipi_dsi_pixel_format_to_bpp(device->format) / 8; u16 hbp = 0, hfp = 0, hsa = 0, hblk = 0, vblk = 0; u32 basic_ctl = 0; size_t bytes; @@ -555,7 +555,7 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi, * (4 bytes). Its minimal size is therefore 10 bytes */ #define HSA_PACKET_OVERHEAD 10 - hsa = max((unsigned int)HSA_PACKET_OVERHEAD, + hsa = max(HSA_PACKET_OVERHEAD, (mode->hsync_end - mode->hsync_start) * Bpp - HSA_PACKET_OVERHEAD); /* @@ -564,7 +564,7 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi, * therefore 6 bytes */ #define HBP_PACKET_OVERHEAD 6 - hbp = max((unsigned int)HBP_PACKET_OVERHEAD, + hbp = max(HBP_PACKET_OVERHEAD, (mode->htotal - mode->hsync_end) * Bpp - HBP_PACKET_OVERHEAD); /* @@ -574,7 +574,7 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi, * 16 bytes */ #define HFP_PACKET_OVERHEAD 16 - hfp = max((unsigned int)HFP_PACKET_OVERHEAD, + hfp = max(HFP_PACKET_OVERHEAD, (mode->hsync_start - mode->hdisplay) * Bpp - HFP_PACKET_OVERHEAD); /* @@ -583,7 +583,7 @@ static void sun6i_dsi_setup_timings(struct sun6i_dsi *dsi, * bytes). Its minimal size is therefore 10 bytes. */ #define HBLK_PACKET_OVERHEAD 10 - hblk = max((unsigned int)HBLK_PACKET_OVERHEAD, + hblk = max(HBLK_PACKET_OVERHEAD, (mode->htotal - (mode->hsync_end - mode->hsync_start)) * Bpp - HBLK_PACKET_OVERHEAD); diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index 0e210df65c30..97184c333526 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -912,7 +912,7 @@ int ttm_bo_validate(struct ttm_buffer_object *bo, /* * We might need to add a TTM. */ - if (bo->resource->mem_type == TTM_PL_SYSTEM) { + if (!bo->resource || bo->resource->mem_type == TTM_PL_SYSTEM) { ret = ttm_tt_create(bo, true); if (ret) return ret; diff --git a/drivers/gpu/drm/vc4/Kconfig b/drivers/gpu/drm/vc4/Kconfig index 061be9a6619d..b0f3117102ca 100644 --- a/drivers/gpu/drm/vc4/Kconfig +++ b/drivers/gpu/drm/vc4/Kconfig @@ -8,6 +8,7 @@ config DRM_VC4 depends on DRM depends on SND && SND_SOC depends on COMMON_CLK + depends on PM select DRM_DISPLAY_HDMI_HELPER select DRM_DISPLAY_HELPER select DRM_KMS_HELPER diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index 592c3b5d03e6..1e5f68704d7d 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -2855,7 +2855,7 @@ static int vc5_hdmi_init_resources(struct vc4_hdmi *vc4_hdmi) return 0; } -static int __maybe_unused vc4_hdmi_runtime_suspend(struct device *dev) +static int vc4_hdmi_runtime_suspend(struct device *dev) { struct vc4_hdmi *vc4_hdmi = dev_get_drvdata(dev); @@ -2972,17 +2972,15 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data) vc4_hdmi->disable_4kp60 = true; } + pm_runtime_enable(dev); + /* - * We need to have the device powered up at this point to call - * our reset hook and for the CEC init. + * We need to have the device powered up at this point to call + * our reset hook and for the CEC init. */ - ret = vc4_hdmi_runtime_resume(dev); + ret = pm_runtime_resume_and_get(dev); if (ret) - goto err_put_ddc; - - pm_runtime_get_noresume(dev); - pm_runtime_set_active(dev); - pm_runtime_enable(dev); + goto err_disable_runtime_pm; if ((of_device_is_compatible(dev->of_node, "brcm,bcm2711-hdmi0") || of_device_is_compatible(dev->of_node, "brcm,bcm2711-hdmi1")) && @@ -3028,6 +3026,7 @@ err_destroy_conn: err_destroy_encoder: drm_encoder_cleanup(encoder); pm_runtime_put_sync(dev); +err_disable_runtime_pm: pm_runtime_disable(dev); err_put_ddc: put_device(&vc4_hdmi->ddc->dev); |