diff options
Diffstat (limited to 'drivers/gpu')
96 files changed, 951 insertions, 527 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c index afe5ca81beec..db7858fe0c3d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c @@ -641,6 +641,7 @@ static void aca_error_fini(struct aca_error *aerr) aca_bank_error_remove(aerr, bank_error); out_unlock: + mutex_unlock(&aerr->lock); mutex_destroy(&aerr->lock); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 06c1913d5a3f..29b400cdd6d5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -1439,7 +1439,10 @@ static int init_kfd_vm(struct amdgpu_vm *vm, void **process_info, *process_info = info; } - vm->process_info = *process_info; + if (cmpxchg(&vm->process_info, NULL, *process_info) != NULL) { + ret = -EINVAL; + goto already_acquired; + } /* Validate page directory and attach eviction fence */ ret = amdgpu_bo_reserve(vm->root.bo, true); @@ -1479,6 +1482,7 @@ validate_pd_fail: amdgpu_bo_unreserve(vm->root.bo); reserve_pd_fail: vm->process_info = NULL; +already_acquired: if (info) { dma_fence_put(&info->eviction_fence->base); *process_info = NULL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index d9789e0b5201..d8296dfc5e8a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2690,8 +2690,10 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) break; default: r = amdgpu_discovery_set_ip_blocks(adev); - if (r) + if (r) { + adev->num_ip_blocks = 0; return r; + } break; } @@ -3247,6 +3249,8 @@ int amdgpu_device_set_cg_state(struct amdgpu_device *adev, i = state == AMD_CG_STATE_GATE ? j : adev->num_ip_blocks - j - 1; if (!adev->ip_blocks[i].status.late_initialized) continue; + if (!adev->ip_blocks[i].version) + continue; /* skip CG for GFX, SDMA on S0ix */ if (adev->in_s0ix && (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX || @@ -3286,6 +3290,8 @@ int amdgpu_device_set_pg_state(struct amdgpu_device *adev, i = state == AMD_PG_STATE_GATE ? j : adev->num_ip_blocks - j - 1; if (!adev->ip_blocks[i].status.late_initialized) continue; + if (!adev->ip_blocks[i].version) + continue; /* skip PG for GFX, SDMA on S0ix */ if (adev->in_s0ix && (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX || @@ -3493,6 +3499,8 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) int i, r; for (i = 0; i < adev->num_ip_blocks; i++) { + if (!adev->ip_blocks[i].version) + continue; if (!adev->ip_blocks[i].version->funcs->early_fini) continue; @@ -3570,6 +3578,8 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) if (!adev->ip_blocks[i].status.sw) continue; + if (!adev->ip_blocks[i].version) + continue; if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC) { amdgpu_ucode_free_bo(adev); amdgpu_free_static_csa(&adev->virt.csa_obj); @@ -3596,6 +3606,8 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) for (i = adev->num_ip_blocks - 1; i >= 0; i--) { if (!adev->ip_blocks[i].status.late_initialized) continue; + if (!adev->ip_blocks[i].version) + continue; if (adev->ip_blocks[i].version->funcs->late_fini) adev->ip_blocks[i].version->funcs->late_fini(&adev->ip_blocks[i]); adev->ip_blocks[i].status.late_initialized = false; @@ -7059,6 +7071,15 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev) dev_info(adev->dev, "PCI error: slot reset callback!!\n"); memset(&reset_context, 0, sizeof(reset_context)); + INIT_LIST_HEAD(&device_list); + hive = amdgpu_get_xgmi_hive(adev); + if (hive) { + mutex_lock(&hive->hive_lock); + list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) + list_add_tail(&tmp_adev->reset_list, &device_list); + } else { + list_add_tail(&adev->reset_list, &device_list); + } if (adev->pcie_reset_ctx.swus) link_dev = adev->pcie_reset_ctx.swus; @@ -7099,19 +7120,13 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev) reset_context.reset_req_dev = adev; set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); set_bit(AMDGPU_SKIP_COREDUMP, &reset_context.flags); - INIT_LIST_HEAD(&device_list); - hive = amdgpu_get_xgmi_hive(adev); if (hive) { - mutex_lock(&hive->hive_lock); reset_context.hive = hive; - list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) { + list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) tmp_adev->pcie_reset_ctx.in_link_reset = true; - list_add_tail(&tmp_adev->reset_list, &device_list); - } } else { set_bit(AMDGPU_SKIP_HW_RESET, &reset_context.flags); - list_add_tail(&adev->reset_list, &device_list); } r = amdgpu_device_asic_reset(adev, &device_list, &reset_context); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 77e2133de5cf..7f19554b9ad1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -83,7 +83,7 @@ void amdgpu_driver_unload_kms(struct drm_device *dev) { struct amdgpu_device *adev = drm_to_adev(dev); - if (adev == NULL) + if (adev == NULL || !adev->num_ip_blocks) return; amdgpu_unregister_gpu_instance(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h index dc8d2f52c7d6..e244c12ceb23 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h @@ -368,15 +368,15 @@ struct amdgpu_mode_info { struct drm_property *plane_ctm_property; /** - * @shaper_lut_property: Plane property to set pre-blending shaper LUT - * that converts color content before 3D LUT. If - * plane_shaper_tf_property != Identity TF, AMD color module will + * @plane_shaper_lut_property: Plane property to set pre-blending + * shaper LUT that converts color content before 3D LUT. + * If plane_shaper_tf_property != Identity TF, AMD color module will * combine the user LUT values with pre-defined TF into the LUT * parameters to be programmed. */ struct drm_property *plane_shaper_lut_property; /** - * @shaper_lut_size_property: Plane property for the size of + * @plane_shaper_lut_size_property: Plane property for the size of * pre-blending shaper LUT as supported by the driver (read-only). */ struct drm_property *plane_shaper_lut_size_property; @@ -400,10 +400,10 @@ struct amdgpu_mode_info { */ struct drm_property *plane_lut3d_property; /** - * @plane_degamma_lut_size_property: Plane property to define the max - * size of 3D LUT as supported by the driver (read-only). The max size - * is the max size of one dimension and, therefore, the max number of - * entries for 3D LUT array is the 3D LUT size cubed; + * @plane_lut3d_size_property: Plane property to define the max size + * of 3D LUT as supported by the driver (read-only). The max size is + * the max size of one dimension and, therefore, the max number of + * entries for 3D LUT array is the 3D LUT size cubed. */ struct drm_property *plane_lut3d_size_property; /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c index 6e8aad91bcd3..0d3c18f04ac3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c @@ -332,13 +332,13 @@ static ssize_t ta_if_invoke_debugfs_write(struct file *fp, const char *buf, size if (!context || !context->initialized) { dev_err(adev->dev, "TA is not initialized\n"); ret = -EINVAL; - goto err_free_shared_buf; + goto free_shared_buf; } if (!psp->ta_funcs || !psp->ta_funcs->fn_ta_invoke) { dev_err(adev->dev, "Unsupported function to invoke TA\n"); ret = -EOPNOTSUPP; - goto err_free_shared_buf; + goto free_shared_buf; } context->session_id = ta_id; @@ -346,7 +346,7 @@ static ssize_t ta_if_invoke_debugfs_write(struct file *fp, const char *buf, size mutex_lock(&psp->ras_context.mutex); ret = prep_ta_mem_context(&context->mem_context, shared_buf, shared_buf_len); if (ret) - goto err_free_shared_buf; + goto unlock; ret = psp_fn_ta_invoke(psp, cmd_id); if (ret || context->resp_status) { @@ -354,15 +354,17 @@ static ssize_t ta_if_invoke_debugfs_write(struct file *fp, const char *buf, size ret, context->resp_status); if (!ret) { ret = -EINVAL; - goto err_free_shared_buf; + goto unlock; } } if (copy_to_user((char *)&buf[copy_pos], context->mem_context.shared_buf, shared_buf_len)) ret = -EFAULT; -err_free_shared_buf: +unlock: mutex_unlock(&psp->ras_context.mutex); + +free_shared_buf: kfree(shared_buf); return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 9d67b770bcc2..7c450350847d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -446,8 +446,7 @@ static int amdgpu_userq_wait_for_last_fence(struct amdgpu_usermode_queue *queue) return ret; } -static void amdgpu_userq_cleanup(struct amdgpu_usermode_queue *queue, - int queue_id) +static void amdgpu_userq_cleanup(struct amdgpu_usermode_queue *queue) { struct amdgpu_userq_mgr *uq_mgr = queue->userq_mgr; struct amdgpu_device *adev = uq_mgr->adev; @@ -461,7 +460,6 @@ static void amdgpu_userq_cleanup(struct amdgpu_usermode_queue *queue, uq_funcs->mqd_destroy(queue); amdgpu_userq_fence_driver_free(queue); /* Use interrupt-safe locking since IRQ handlers may access these XArrays */ - xa_erase_irq(&uq_mgr->userq_xa, (unsigned long)queue_id); xa_erase_irq(&adev->userq_doorbell_xa, queue->doorbell_index); queue->userq_mgr = NULL; list_del(&queue->userq_va_list); @@ -470,12 +468,6 @@ static void amdgpu_userq_cleanup(struct amdgpu_usermode_queue *queue, up_read(&adev->reset_domain->sem); } -static struct amdgpu_usermode_queue * -amdgpu_userq_find(struct amdgpu_userq_mgr *uq_mgr, int qid) -{ - return xa_load(&uq_mgr->userq_xa, qid); -} - void amdgpu_userq_ensure_ev_fence(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_eviction_fence_mgr *evf_mgr) @@ -625,22 +617,13 @@ unref_bo: } static int -amdgpu_userq_destroy(struct drm_file *filp, int queue_id) +amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_queue *queue) { - struct amdgpu_fpriv *fpriv = filp->driver_priv; - struct amdgpu_userq_mgr *uq_mgr = &fpriv->userq_mgr; struct amdgpu_device *adev = uq_mgr->adev; - struct amdgpu_usermode_queue *queue; int r = 0; cancel_delayed_work_sync(&uq_mgr->resume_work); mutex_lock(&uq_mgr->userq_mutex); - queue = amdgpu_userq_find(uq_mgr, queue_id); - if (!queue) { - drm_dbg_driver(adev_to_drm(uq_mgr->adev), "Invalid queue id to destroy\n"); - mutex_unlock(&uq_mgr->userq_mutex); - return -EINVAL; - } amdgpu_userq_wait_for_last_fence(queue); /* Cancel any pending hang detection work and cleanup */ if (queue->hang_detect_fence) { @@ -672,7 +655,7 @@ amdgpu_userq_destroy(struct drm_file *filp, int queue_id) drm_warn(adev_to_drm(uq_mgr->adev), "trying to destroy a HW mapping userq\n"); queue->state = AMDGPU_USERQ_STATE_HUNG; } - amdgpu_userq_cleanup(queue, queue_id); + amdgpu_userq_cleanup(queue); mutex_unlock(&uq_mgr->userq_mutex); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); @@ -680,6 +663,37 @@ amdgpu_userq_destroy(struct drm_file *filp, int queue_id) return r; } +static void amdgpu_userq_kref_destroy(struct kref *kref) +{ + int r; + struct amdgpu_usermode_queue *queue = + container_of(kref, struct amdgpu_usermode_queue, refcount); + struct amdgpu_userq_mgr *uq_mgr = queue->userq_mgr; + + r = amdgpu_userq_destroy(uq_mgr, queue); + if (r) + drm_file_err(uq_mgr->file, "Failed to destroy usermode queue %d\n", r); +} + +struct amdgpu_usermode_queue *amdgpu_userq_get(struct amdgpu_userq_mgr *uq_mgr, u32 qid) +{ + struct amdgpu_usermode_queue *queue; + + xa_lock(&uq_mgr->userq_xa); + queue = xa_load(&uq_mgr->userq_xa, qid); + if (queue) + kref_get(&queue->refcount); + xa_unlock(&uq_mgr->userq_xa); + + return queue; +} + +void amdgpu_userq_put(struct amdgpu_usermode_queue *queue) +{ + if (queue) + kref_put(&queue->refcount, amdgpu_userq_kref_destroy); +} + static int amdgpu_userq_priority_permit(struct drm_file *filp, int priority) { @@ -834,6 +848,9 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) goto unlock; } + /* drop this refcount during queue destroy */ + kref_init(&queue->refcount); + /* Wait for mode-1 reset to complete */ down_read(&adev->reset_domain->sem); r = xa_err(xa_store_irq(&adev->userq_doorbell_xa, index, queue, GFP_KERNEL)); @@ -985,7 +1002,9 @@ int amdgpu_userq_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) { union drm_amdgpu_userq *args = data; - int r; + struct amdgpu_fpriv *fpriv = filp->driver_priv; + struct amdgpu_usermode_queue *queue; + int r = 0; if (!amdgpu_userq_enabled(dev)) return -ENOTSUPP; @@ -1000,11 +1019,16 @@ int amdgpu_userq_ioctl(struct drm_device *dev, void *data, drm_file_err(filp, "Failed to create usermode queue\n"); break; - case AMDGPU_USERQ_OP_FREE: - r = amdgpu_userq_destroy(filp, args->in.queue_id); - if (r) - drm_file_err(filp, "Failed to destroy usermode queue\n"); + case AMDGPU_USERQ_OP_FREE: { + xa_lock(&fpriv->userq_mgr.userq_xa); + queue = __xa_erase(&fpriv->userq_mgr.userq_xa, args->in.queue_id); + xa_unlock(&fpriv->userq_mgr.userq_xa); + if (!queue) + return -ENOENT; + + amdgpu_userq_put(queue); break; + } default: drm_dbg_driver(dev, "Invalid user queue op specified: %d\n", args->in.op); @@ -1023,16 +1047,23 @@ amdgpu_userq_restore_all(struct amdgpu_userq_mgr *uq_mgr) /* Resume all the queues for this process */ xa_for_each(&uq_mgr->userq_xa, queue_id, queue) { + queue = amdgpu_userq_get(uq_mgr, queue_id); + if (!queue) + continue; + if (!amdgpu_userq_buffer_vas_mapped(queue)) { drm_file_err(uq_mgr->file, "trying restore queue without va mapping\n"); queue->state = AMDGPU_USERQ_STATE_INVALID_VA; + amdgpu_userq_put(queue); continue; } r = amdgpu_userq_restore_helper(queue); if (r) ret = r; + + amdgpu_userq_put(queue); } if (ret) @@ -1266,9 +1297,13 @@ amdgpu_userq_evict_all(struct amdgpu_userq_mgr *uq_mgr) amdgpu_userq_detect_and_reset_queues(uq_mgr); /* Try to unmap all the queues in this process ctx */ xa_for_each(&uq_mgr->userq_xa, queue_id, queue) { + queue = amdgpu_userq_get(uq_mgr, queue_id); + if (!queue) + continue; r = amdgpu_userq_preempt_helper(queue); if (r) ret = r; + amdgpu_userq_put(queue); } if (ret) @@ -1301,16 +1336,24 @@ amdgpu_userq_wait_for_signal(struct amdgpu_userq_mgr *uq_mgr) int ret; xa_for_each(&uq_mgr->userq_xa, queue_id, queue) { + queue = amdgpu_userq_get(uq_mgr, queue_id); + if (!queue) + continue; + struct dma_fence *f = queue->last_fence; - if (!f || dma_fence_is_signaled(f)) + if (!f || dma_fence_is_signaled(f)) { + amdgpu_userq_put(queue); continue; + } ret = dma_fence_wait_timeout(f, true, msecs_to_jiffies(100)); if (ret <= 0) { drm_file_err(uq_mgr->file, "Timed out waiting for fence=%llu:%llu\n", f->context, f->seqno); + amdgpu_userq_put(queue); return -ETIMEDOUT; } + amdgpu_userq_put(queue); } return 0; @@ -1361,20 +1404,23 @@ int amdgpu_userq_mgr_init(struct amdgpu_userq_mgr *userq_mgr, struct drm_file *f void amdgpu_userq_mgr_fini(struct amdgpu_userq_mgr *userq_mgr) { struct amdgpu_usermode_queue *queue; - unsigned long queue_id; + unsigned long queue_id = 0; + + for (;;) { + xa_lock(&userq_mgr->userq_xa); + queue = xa_find(&userq_mgr->userq_xa, &queue_id, ULONG_MAX, + XA_PRESENT); + if (queue) + __xa_erase(&userq_mgr->userq_xa, queue_id); + xa_unlock(&userq_mgr->userq_xa); - cancel_delayed_work_sync(&userq_mgr->resume_work); + if (!queue) + break; - mutex_lock(&userq_mgr->userq_mutex); - amdgpu_userq_detect_and_reset_queues(userq_mgr); - xa_for_each(&userq_mgr->userq_xa, queue_id, queue) { - amdgpu_userq_wait_for_last_fence(queue); - amdgpu_userq_unmap_helper(queue); - amdgpu_userq_cleanup(queue, queue_id); + amdgpu_userq_put(queue); } xa_destroy(&userq_mgr->userq_xa); - mutex_unlock(&userq_mgr->userq_mutex); mutex_destroy(&userq_mgr->userq_mutex); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h index 5845d8959034..736c1d38297c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h @@ -74,6 +74,7 @@ struct amdgpu_usermode_queue { struct dentry *debugfs_queue; struct delayed_work hang_detect_work; struct dma_fence *hang_detect_fence; + struct kref refcount; struct list_head userq_va_list; }; @@ -112,6 +113,9 @@ struct amdgpu_db_info { struct amdgpu_userq_obj *db_obj; }; +struct amdgpu_usermode_queue *amdgpu_userq_get(struct amdgpu_userq_mgr *uq_mgr, u32 qid); +void amdgpu_userq_put(struct amdgpu_usermode_queue *queue); + int amdgpu_userq_ioctl(struct drm_device *dev, void *data, struct drm_file *filp); int amdgpu_userq_mgr_init(struct amdgpu_userq_mgr *userq_mgr, struct drm_file *file_priv, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c index 8013260e29dc..5239b06b9ab0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c @@ -35,6 +35,8 @@ static const struct dma_fence_ops amdgpu_userq_fence_ops; static struct kmem_cache *amdgpu_userq_fence_slab; +#define AMDGPU_USERQ_MAX_HANDLES (1U << 16) + int amdgpu_userq_fence_slab_init(void) { amdgpu_userq_fence_slab = kmem_cache_create("amdgpu_userq_fence", @@ -464,7 +466,7 @@ int amdgpu_userq_signal_ioctl(struct drm_device *dev, void *data, struct drm_amdgpu_userq_signal *args = data; struct drm_gem_object **gobj_write = NULL; struct drm_gem_object **gobj_read = NULL; - struct amdgpu_usermode_queue *queue; + struct amdgpu_usermode_queue *queue = NULL; struct amdgpu_userq_fence *userq_fence; struct drm_syncobj **syncobj = NULL; u32 *bo_handles_write, num_write_bo_handles; @@ -478,6 +480,11 @@ int amdgpu_userq_signal_ioctl(struct drm_device *dev, void *data, if (!amdgpu_userq_enabled(dev)) return -ENOTSUPP; + if (args->num_syncobj_handles > AMDGPU_USERQ_MAX_HANDLES || + args->num_bo_write_handles > AMDGPU_USERQ_MAX_HANDLES || + args->num_bo_read_handles > AMDGPU_USERQ_MAX_HANDLES) + return -EINVAL; + num_syncobj_handles = args->num_syncobj_handles; syncobj_handles = memdup_user(u64_to_user_ptr(args->syncobj_handles), size_mul(sizeof(u32), num_syncobj_handles)); @@ -546,7 +553,7 @@ int amdgpu_userq_signal_ioctl(struct drm_device *dev, void *data, } /* Retrieve the user queue */ - queue = xa_load(&userq_mgr->userq_xa, args->queue_id); + queue = amdgpu_userq_get(userq_mgr, args->queue_id); if (!queue) { r = -ENOENT; goto put_gobj_write; @@ -641,6 +648,9 @@ free_syncobj: free_syncobj_handles: kfree(syncobj_handles); + if (queue) + amdgpu_userq_put(queue); + return r; } @@ -653,7 +663,7 @@ int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data, struct drm_amdgpu_userq_wait *wait_info = data; struct amdgpu_fpriv *fpriv = filp->driver_priv; struct amdgpu_userq_mgr *userq_mgr = &fpriv->userq_mgr; - struct amdgpu_usermode_queue *waitq; + struct amdgpu_usermode_queue *waitq = NULL; struct drm_gem_object **gobj_write; struct drm_gem_object **gobj_read; struct dma_fence **fences = NULL; @@ -664,6 +674,11 @@ int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data, if (!amdgpu_userq_enabled(dev)) return -ENOTSUPP; + if (wait_info->num_syncobj_handles > AMDGPU_USERQ_MAX_HANDLES || + wait_info->num_bo_write_handles > AMDGPU_USERQ_MAX_HANDLES || + wait_info->num_bo_read_handles > AMDGPU_USERQ_MAX_HANDLES) + return -EINVAL; + num_read_bo_handles = wait_info->num_bo_read_handles; bo_handles_read = memdup_user(u64_to_user_ptr(wait_info->bo_read_handles), size_mul(sizeof(u32), num_read_bo_handles)); @@ -833,7 +848,7 @@ int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data, dma_resv_for_each_fence(&resv_cursor, gobj_read[i]->resv, DMA_RESV_USAGE_READ, fence) { - if (WARN_ON_ONCE(num_fences >= wait_info->num_fences)) { + if (num_fences >= wait_info->num_fences) { r = -EINVAL; goto free_fences; } @@ -850,7 +865,7 @@ int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data, dma_resv_for_each_fence(&resv_cursor, gobj_write[i]->resv, DMA_RESV_USAGE_WRITE, fence) { - if (WARN_ON_ONCE(num_fences >= wait_info->num_fences)) { + if (num_fences >= wait_info->num_fences) { r = -EINVAL; goto free_fences; } @@ -874,8 +889,9 @@ int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data, goto free_fences; dma_fence_unwrap_for_each(f, &iter, fence) { - if (WARN_ON_ONCE(num_fences >= wait_info->num_fences)) { + if (num_fences >= wait_info->num_fences) { r = -EINVAL; + dma_fence_put(fence); goto free_fences; } @@ -898,8 +914,9 @@ int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data, if (r) goto free_fences; - if (WARN_ON_ONCE(num_fences >= wait_info->num_fences)) { + if (num_fences >= wait_info->num_fences) { r = -EINVAL; + dma_fence_put(fence); goto free_fences; } @@ -912,7 +929,7 @@ int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data, */ num_fences = dma_fence_dedup_array(fences, num_fences); - waitq = xa_load(&userq_mgr->userq_xa, wait_info->waitq_id); + waitq = amdgpu_userq_get(userq_mgr, wait_info->waitq_id); if (!waitq) { r = -EINVAL; goto free_fences; @@ -969,32 +986,14 @@ int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data, r = -EFAULT; goto free_fences; } - - kfree(fences); - kfree(fence_info); } - drm_exec_fini(&exec); - for (i = 0; i < num_read_bo_handles; i++) - drm_gem_object_put(gobj_read[i]); - kfree(gobj_read); - - for (i = 0; i < num_write_bo_handles; i++) - drm_gem_object_put(gobj_write[i]); - kfree(gobj_write); - - kfree(timeline_points); - kfree(timeline_handles); - kfree(syncobj_handles); - kfree(bo_handles_write); - kfree(bo_handles_read); - - return 0; - free_fences: - while (num_fences-- > 0) - dma_fence_put(fences[num_fences]); - kfree(fences); + if (fences) { + while (num_fences-- > 0) + dma_fence_put(fences[num_fences]); + kfree(fences); + } free_fence_info: kfree(fence_info); exec_fini: @@ -1018,5 +1017,8 @@ free_bo_handles_write: free_bo_handles_read: kfree(bo_handles_read); + if (waitq) + amdgpu_userq_put(waitq); + return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 09ebb13ca5e8..a926a330700e 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -720,11 +720,6 @@ static int mes_v11_0_set_hw_resources(struct amdgpu_mes *mes) mes_set_hw_res_pkt.enable_reg_active_poll = 1; mes_set_hw_res_pkt.enable_level_process_quantum_check = 1; mes_set_hw_res_pkt.oversubscription_timer = 50; - if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x7f) - mes_set_hw_res_pkt.enable_lr_compute_wa = 1; - else - dev_info_once(mes->adev->dev, - "MES FW version must be >= 0x7f to enable LR compute workaround.\n"); if (amdgpu_mes_log_enable) { mes_set_hw_res_pkt.enable_mes_event_int_logging = 1; diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c index b1c864dc79a8..023c7345ea54 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c @@ -731,6 +731,9 @@ static int mes_v12_0_set_hw_resources(struct amdgpu_mes *mes, int pipe) int i; struct amdgpu_device *adev = mes->adev; union MESAPI_SET_HW_RESOURCES mes_set_hw_res_pkt; + uint32_t mes_rev = (pipe == AMDGPU_MES_SCHED_PIPE) ? + (mes->sched_version & AMDGPU_MES_VERSION_MASK) : + (mes->kiq_version & AMDGPU_MES_VERSION_MASK); memset(&mes_set_hw_res_pkt, 0, sizeof(mes_set_hw_res_pkt)); @@ -779,18 +782,13 @@ static int mes_v12_0_set_hw_resources(struct amdgpu_mes *mes, int pipe) mes_set_hw_res_pkt.use_different_vmid_compute = 1; mes_set_hw_res_pkt.enable_reg_active_poll = 1; mes_set_hw_res_pkt.enable_level_process_quantum_check = 1; - if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x82) - mes_set_hw_res_pkt.enable_lr_compute_wa = 1; - else - dev_info_once(adev->dev, - "MES FW version must be >= 0x82 to enable LR compute workaround.\n"); /* * Keep oversubscribe timer for sdma . When we have unmapped doorbell * handling support, other queue will not use the oversubscribe timer. * handling mode - 0: disabled; 1: basic version; 2: basic+ version */ - mes_set_hw_res_pkt.oversubscription_timer = 50; + mes_set_hw_res_pkt.oversubscription_timer = mes_rev < 0x8b ? 0 : 50; mes_set_hw_res_pkt.unmapped_doorbell_handling = 1; if (amdgpu_mes_log_enable) { diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v15_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v15_0.c index 723ddae17644..73a709773e85 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v15_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v15_0.c @@ -69,12 +69,12 @@ static int psp_v15_0_0_ring_stop(struct psp_context *psp, 0x80000000, 0x80000000, false); } else { /* Write the ring destroy command*/ - WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_64, + WREG32_SOC15(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_64, GFX_CTRL_CMD_ID_DESTROY_RINGS); /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_64), 0x80000000, 0x80000000, false); } @@ -116,7 +116,7 @@ static int psp_v15_0_0_ring_create(struct psp_context *psp, } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_64), 0x80000000, 0x80000000, false); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); @@ -125,23 +125,23 @@ static int psp_v15_0_0_ring_create(struct psp_context *psp, /* Write low address of the ring to C2PMSG_69 */ psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr); - WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_69, psp_ring_reg); + WREG32_SOC15(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_69, psp_ring_reg); /* Write high address of the ring to C2PMSG_70 */ psp_ring_reg = upper_32_bits(ring->ring_mem_mc_addr); - WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_70, psp_ring_reg); + WREG32_SOC15(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_70, psp_ring_reg); /* Write size of ring to C2PMSG_71 */ psp_ring_reg = ring->ring_size; - WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_71, psp_ring_reg); + WREG32_SOC15(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_71, psp_ring_reg); /* Write the ring initialization command to C2PMSG_64 */ psp_ring_reg = ring_type; psp_ring_reg = psp_ring_reg << 16; - WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_64, psp_ring_reg); + WREG32_SOC15(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_64, psp_ring_reg); /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_64), 0x80000000, 0x8000FFFF, false); } @@ -174,7 +174,7 @@ static uint32_t psp_v15_0_0_ring_get_wptr(struct psp_context *psp) if (amdgpu_sriov_vf(adev)) data = RREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_102); else - data = RREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_67); + data = RREG32_SOC15(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_67); return data; } @@ -188,7 +188,7 @@ static void psp_v15_0_0_ring_set_wptr(struct psp_context *psp, uint32_t value) WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_101, GFX_CTRL_CMD_ID_CONSUME_CMD); } else - WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_67, value); + WREG32_SOC15(MP0, 0, regMPASP_PCRU1_MPASP_C2PMSG_67, value); } static const struct psp_funcs psp_v15_0_0_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c index 8122a5cacf07..a0ad1f8a76f0 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc21.c +++ b/drivers/gpu/drm/amd/amdgpu/soc21.c @@ -858,7 +858,9 @@ static int soc21_common_early_init(struct amdgpu_ip_block *ip_block) AMD_CG_SUPPORT_IH_CG | AMD_CG_SUPPORT_BIF_MGCG | AMD_CG_SUPPORT_BIF_LS; - adev->pg_flags = AMD_PG_SUPPORT_VCN | + adev->pg_flags = AMD_PG_SUPPORT_VCN_DPG | + AMD_PG_SUPPORT_VCN | + AMD_PG_SUPPORT_JPEG_DPG | AMD_PG_SUPPORT_JPEG | AMD_PG_SUPPORT_GFX_PG; adev->external_rev_id = adev->rev_id + 0x1; diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c index 0202df5db1e1..6109124f852e 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c @@ -174,6 +174,10 @@ static int vcn_v5_0_0_sw_init(struct amdgpu_ip_block *ip_block) fw_shared->present_flag_0 = cpu_to_le32(AMDGPU_FW_SHARED_FLAG_0_UNIFIED_QUEUE); fw_shared->sq.is_enabled = 1; + fw_shared->present_flag_0 |= cpu_to_le32(AMDGPU_VCN_SMU_DPM_INTERFACE_FLAG); + fw_shared->smu_dpm_interface.smu_interface_type = (adev->flags & AMD_IS_APU) ? + AMDGPU_VCN_SMU_DPM_INTERFACE_APU : AMDGPU_VCN_SMU_DPM_INTERFACE_DGPU; + if (amdgpu_vcnfw_log) amdgpu_vcn_fwlog_init(&adev->vcn.inst[i]); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index 8ea31699d38b..f5d2847e1cbb 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -593,6 +593,7 @@ int pqm_update_queue_properties(struct process_queue_manager *pqm, p->queue_size)) { pr_debug("ring buf 0x%llx size 0x%llx not mapped on GPU\n", p->queue_address, p->queue_size); + amdgpu_bo_unreserve(vm->root.bo); return -EFAULT; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c index 2ba98f384685..cd1e58b8defc 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c @@ -1706,6 +1706,7 @@ __set_dm_plane_colorop_3dlut(struct drm_plane_state *plane_state, struct dc_transfer_func *tf = &dc_plane_state->in_shaper_func; struct drm_atomic_state *state = plane_state->state; const struct amdgpu_device *adev = drm_to_adev(colorop->dev); + bool has_3dlut = adev->dm.dc->caps.color.dpp.hw_3d_lut || adev->dm.dc->caps.color.mpc.preblend; const struct drm_device *dev = colorop->dev; const struct drm_color_lut32 *lut3d; uint32_t lut3d_size; @@ -1722,7 +1723,7 @@ __set_dm_plane_colorop_3dlut(struct drm_plane_state *plane_state, } if (colorop_state && !colorop_state->bypass && colorop->type == DRM_COLOROP_3D_LUT) { - if (!adev->dm.dc->caps.color.dpp.hw_3d_lut) { + if (!has_3dlut) { drm_dbg(dev, "3D LUT is not supported by hardware\n"); return -EINVAL; } @@ -1875,6 +1876,7 @@ amdgpu_dm_plane_set_colorop_properties(struct drm_plane_state *plane_state, struct drm_colorop *colorop = plane_state->color_pipeline; struct drm_device *dev = plane_state->plane->dev; struct amdgpu_device *adev = drm_to_adev(dev); + bool has_3dlut = adev->dm.dc->caps.color.dpp.hw_3d_lut || adev->dm.dc->caps.color.mpc.preblend; int ret; /* 1D Curve - DEGAM TF */ @@ -1907,7 +1909,7 @@ amdgpu_dm_plane_set_colorop_properties(struct drm_plane_state *plane_state, if (ret) return ret; - if (adev->dm.dc->caps.color.dpp.hw_3d_lut) { + if (has_3dlut) { /* 1D Curve & LUT - SHAPER TF & LUT */ colorop = colorop->next; if (!colorop) { diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c index f25c0ede7199..d59ba82d3d7c 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c @@ -60,6 +60,7 @@ int amdgpu_dm_initialize_default_pipeline(struct drm_plane *plane, struct drm_pr struct drm_colorop *ops[MAX_COLOR_PIPELINE_OPS]; struct drm_device *dev = plane->dev; struct amdgpu_device *adev = drm_to_adev(dev); + bool has_3dlut = adev->dm.dc->caps.color.dpp.hw_3d_lut || adev->dm.dc->caps.color.mpc.preblend; int ret; int i = 0; @@ -112,7 +113,7 @@ int amdgpu_dm_initialize_default_pipeline(struct drm_plane *plane, struct drm_pr i++; - if (adev->dm.dc->caps.color.dpp.hw_3d_lut) { + if (has_3dlut) { /* 1D curve - SHAPER TF */ ops[i] = kzalloc_obj(*ops[0]); if (!ops[i]) { diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c index 130190e8a1b2..304437c2284d 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c @@ -765,15 +765,15 @@ int amdgpu_dm_crtc_init(struct amdgpu_display_manager *dm, dm->adev->mode_info.crtcs[crtc_index] = acrtc; /* Don't enable DRM CRTC degamma property for - * 1. Degamma is replaced by color pipeline. - * 2. DCE since it doesn't support programmable degamma anywhere. - * 3. DCN401 since pre-blending degamma LUT doesn't apply to cursor. + * 1. DCE since it doesn't support programmable degamma anywhere. + * 2. DCN401 since pre-blending degamma LUT doesn't apply to cursor. + * Note: DEGAMMA properties are created even if the primary plane has the + * COLOR_PIPELINE property. User space can use either the DEGAMMA properties + * or the COLOR_PIPELINE property. An atomic commit which attempts to enable + * both is rejected. */ - if (plane->color_pipeline_property) - has_degamma = false; - else - has_degamma = dm->adev->dm.dc->caps.color.dpp.dcn_arch && - dm->adev->dm.dc->ctx->dce_version != DCN_VERSION_4_01; + has_degamma = dm->adev->dm.dc->caps.color.dpp.dcn_arch && + dm->adev->dm.dc->ctx->dce_version != DCN_VERSION_4_01; drm_crtc_enable_color_mgmt(&acrtc->base, has_degamma ? MAX_COLOR_LUT_ENTRIES : 0, true, MAX_COLOR_LUT_ENTRIES); 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 70587e5a8d46..127207e18dcb 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 @@ -1256,6 +1256,14 @@ static int amdgpu_dm_plane_atomic_check(struct drm_plane *plane, if (ret) return ret; + /* Reject commits that attempt to use both COLOR_PIPELINE and CRTC DEGAMMA_LUT */ + if (new_plane_state->color_pipeline && new_crtc_state->degamma_lut) { + drm_dbg_atomic(plane->dev, + "[PLANE:%d:%s] COLOR_PIPELINE and CRTC DEGAMMA_LUT cannot be enabled simultaneously\n", + plane->base.id, plane->name); + return -EINVAL; + } + ret = amdgpu_dm_plane_fill_dc_scaling_info(adev, new_plane_state, &scaling_info); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 246893d80f1f..baf820e6eae8 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -170,11 +170,11 @@ struct dc_stream_state *dc_create_stream_for_sink( if (sink == NULL) goto fail; - stream = kzalloc_obj(struct dc_stream_state); + stream = kzalloc_obj(struct dc_stream_state, GFP_ATOMIC); if (stream == NULL) goto fail; - stream->update_scratch = kzalloc((int32_t) dc_update_scratch_space_size(), GFP_KERNEL); + stream->update_scratch = kzalloc((int32_t) dc_update_scratch_space_size(), GFP_ATOMIC); if (stream->update_scratch == NULL) goto fail; diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn20/dcn20_dccg.h b/drivers/gpu/drm/amd/display/dc/dccg/dcn20/dcn20_dccg.h index 3711d400773a..4c4e61bc91b5 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn20/dcn20_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn20/dcn20_dccg.h @@ -38,7 +38,11 @@ DCCG_SRII(PIXEL_RATE_CNTL, OTG, 0),\ DCCG_SRII(PIXEL_RATE_CNTL, OTG, 1),\ SR(DISPCLK_FREQ_CHANGE_CNTL),\ - SR(DC_MEM_GLOBAL_PWR_REQ_CNTL) + SR(DC_MEM_GLOBAL_PWR_REQ_CNTL),\ + SR(MICROSECOND_TIME_BASE_DIV),\ + SR(MILLISECOND_TIME_BASE_DIV),\ + SR(DCCG_GATE_DISABLE_CNTL),\ + SR(DCCG_GATE_DISABLE_CNTL2) #define DCCG_REG_LIST_DCN2() \ DCCG_COMMON_REG_LIST_DCN_BASE(),\ diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn21/dcn21_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn21/dcn21_dccg.c index 75c69348027e..c4d4eea140f3 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn21/dcn21_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn21/dcn21_dccg.c @@ -96,6 +96,25 @@ static void dccg21_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppcl dccg->pipe_dppclk_khz[dpp_inst] = req_dppclk; } +/* + * On DCN21 S0i3 resume, BIOS programs MICROSECOND_TIME_BASE_DIV to + * 0x00120464 as a marker that golden init has already been done. + * dcn21_s0i3_golden_init_wa() reads this marker later in bios_golden_init() + * to decide whether to skip golden init. + * + * dccg2_init() unconditionally overwrites MICROSECOND_TIME_BASE_DIV to + * 0x00120264, destroying the marker before it can be read. + * + * Guard the call: if the S0i3 marker is present, skip dccg2_init() so the + * WA can function correctly. bios_golden_init() will handle init in that case. + */ +static void dccg21_init(struct dccg *dccg) +{ + if (dccg2_is_s0i3_golden_init_wa_done(dccg)) + return; + + dccg2_init(dccg); +} static const struct dccg_funcs dccg21_funcs = { .update_dpp_dto = dccg21_update_dpp_dto, @@ -103,7 +122,7 @@ static const struct dccg_funcs dccg21_funcs = { .set_fifo_errdet_ovr_en = dccg2_set_fifo_errdet_ovr_en, .otg_add_pixel = dccg2_otg_add_pixel, .otg_drop_pixel = dccg2_otg_drop_pixel, - .dccg_init = dccg2_init, + .dccg_init = dccg21_init, .refclk_setup = dccg2_refclk_setup, /* Deprecated - for backward compatibility only */ .allow_clock_gating = dccg2_allow_clock_gating, .enable_memory_low_power = dccg2_enable_memory_low_power, diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn301/dcn301_dccg.h b/drivers/gpu/drm/amd/display/dc/dccg/dcn301/dcn301_dccg.h index 067e49cb238e..e2381ca0be0b 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn301/dcn301_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn301/dcn301_dccg.h @@ -34,7 +34,13 @@ DCCG_SRII(DTO_PARAM, DPPCLK, 1),\ DCCG_SRII(DTO_PARAM, DPPCLK, 2),\ DCCG_SRII(DTO_PARAM, DPPCLK, 3),\ - SR(REFCLK_CNTL) + SR(REFCLK_CNTL),\ + SR(DISPCLK_FREQ_CHANGE_CNTL),\ + SR(DC_MEM_GLOBAL_PWR_REQ_CNTL),\ + SR(MICROSECOND_TIME_BASE_DIV),\ + SR(MILLISECOND_TIME_BASE_DIV),\ + SR(DCCG_GATE_DISABLE_CNTL),\ + SR(DCCG_GATE_DISABLE_CNTL2) #define DCCG_MASK_SH_LIST_DCN301(mask_sh) \ DCCG_SFI(DPPCLK_DTO_CTRL, DTO_ENABLE, DPPCLK, 0, mask_sh),\ diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn31/dcn31_dccg.h b/drivers/gpu/drm/amd/display/dc/dccg/dcn31/dcn31_dccg.h index bf659920d4cc..b5e3849ef12a 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn31/dcn31_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn31/dcn31_dccg.h @@ -64,9 +64,12 @@ SR(DSCCLK1_DTO_PARAM),\ SR(DSCCLK2_DTO_PARAM),\ SR(DSCCLK_DTO_CTRL),\ + SR(DCCG_GATE_DISABLE_CNTL),\ SR(DCCG_GATE_DISABLE_CNTL2),\ SR(DCCG_GATE_DISABLE_CNTL3),\ - SR(HDMISTREAMCLK0_DTO_PARAM) + SR(HDMISTREAMCLK0_DTO_PARAM),\ + SR(DC_MEM_GLOBAL_PWR_REQ_CNTL),\ + SR(MICROSECOND_TIME_BASE_DIV) #define DCCG_MASK_SH_LIST_DCN31(mask_sh) \ diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn314/dcn314_dccg.h b/drivers/gpu/drm/amd/display/dc/dccg/dcn314/dcn314_dccg.h index a609635f35db..ecbdc05f7c45 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn314/dcn314_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn314/dcn314_dccg.h @@ -70,11 +70,14 @@ SR(DSCCLK2_DTO_PARAM),\ SR(DSCCLK3_DTO_PARAM),\ SR(DSCCLK_DTO_CTRL),\ + SR(DCCG_GATE_DISABLE_CNTL),\ SR(DCCG_GATE_DISABLE_CNTL2),\ SR(DCCG_GATE_DISABLE_CNTL3),\ SR(HDMISTREAMCLK0_DTO_PARAM),\ SR(OTG_PIXEL_RATE_DIV),\ - SR(DTBCLK_P_CNTL) + SR(DTBCLK_P_CNTL),\ + SR(DC_MEM_GLOBAL_PWR_REQ_CNTL),\ + SR(MICROSECOND_TIME_BASE_DIV) #define DCCG_MASK_SH_LIST_DCN314_COMMON(mask_sh) \ DCCG_SFI(DPPCLK_DTO_CTRL, DTO_DB_EN, DPPCLK, 0, mask_sh),\ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c index b91517b9fedc..eb198d52a115 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c @@ -72,7 +72,11 @@ void dcn401_initialize_min_clocks(struct dc *dc) * audio corruption. Read current DISPCLK from DENTIST and request the same * freq to ensure that the timing is valid and unchanged. */ - clocks->dispclk_khz = dc->clk_mgr->funcs->get_dispclk_from_dentist(dc->clk_mgr); + if (dc->clk_mgr->funcs->get_dispclk_from_dentist) { + clocks->dispclk_khz = dc->clk_mgr->funcs->get_dispclk_from_dentist(dc->clk_mgr); + } else { + clocks->dispclk_khz = dc->clk_mgr->boot_snapshot.dispclk * 1000; + } } clocks->ref_dtbclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dtbclk_mhz * 1000; clocks->fclk_p_state_change_support = true; diff --git a/drivers/gpu/drm/amd/include/asic_reg/mp/mp_15_0_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/mp/mp_15_0_0_offset.h index 0e4c195297a4..fe97943b9b97 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/mp/mp_15_0_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/mp/mp_15_0_0_offset.h @@ -82,6 +82,24 @@ #define regMPASP_SMN_IH_SW_INT_CTRL 0x0142 #define regMPASP_SMN_IH_SW_INT_CTRL_BASE_IDX 0 +// addressBlock: mp_SmuMpASPPub_PcruDec +// base address: 0x3800000 +#define regMPASP_PCRU1_MPASP_C2PMSG_64 0x4280 +#define regMPASP_PCRU1_MPASP_C2PMSG_64_BASE_IDX 3 +#define regMPASP_PCRU1_MPASP_C2PMSG_65 0x4281 +#define regMPASP_PCRU1_MPASP_C2PMSG_65_BASE_IDX 3 +#define regMPASP_PCRU1_MPASP_C2PMSG_66 0x4282 +#define regMPASP_PCRU1_MPASP_C2PMSG_66_BASE_IDX 3 +#define regMPASP_PCRU1_MPASP_C2PMSG_67 0x4283 +#define regMPASP_PCRU1_MPASP_C2PMSG_67_BASE_IDX 3 +#define regMPASP_PCRU1_MPASP_C2PMSG_68 0x4284 +#define regMPASP_PCRU1_MPASP_C2PMSG_68_BASE_IDX 3 +#define regMPASP_PCRU1_MPASP_C2PMSG_69 0x4285 +#define regMPASP_PCRU1_MPASP_C2PMSG_69_BASE_IDX 3 +#define regMPASP_PCRU1_MPASP_C2PMSG_70 0x4286 +#define regMPASP_PCRU1_MPASP_C2PMSG_70_BASE_IDX 3 +#define regMPASP_PCRU1_MPASP_C2PMSG_71 0x4287 +#define regMPASP_PCRU1_MPASP_C2PMSG_71_BASE_IDX 3 // addressBlock: mp_SmuMp1_SmnDec // base address: 0x0 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 e030f1e186cb..a8d63d4d1f6e 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 @@ -2034,6 +2034,7 @@ static ssize_t smu_v13_0_0_get_gpu_metrics(struct smu_context *smu, smu, SMU_DRIVER_TABLE_GPU_METRICS); SmuMetricsExternal_t metrics_ext; SmuMetrics_t *metrics = &metrics_ext.SmuMetrics; + uint32_t mp1_ver = amdgpu_ip_version(smu->adev, MP1_HWIP, 0); int ret = 0; ret = smu_cmn_get_metrics_table(smu, @@ -2058,7 +2059,12 @@ static ssize_t smu_v13_0_0_get_gpu_metrics(struct smu_context *smu, metrics->Vcn1ActivityPercentage); gpu_metrics->average_socket_power = metrics->AverageSocketPower; - gpu_metrics->energy_accumulator = metrics->EnergyAccumulator; + + if ((mp1_ver == IP_VERSION(13, 0, 0) && smu->smc_fw_version <= 0x004e1e00) || + (mp1_ver == IP_VERSION(13, 0, 10) && smu->smc_fw_version <= 0x00500800)) + gpu_metrics->energy_accumulator = metrics->EnergyAccumulator; + else + gpu_metrics->energy_accumulator = UINT_MAX; if (metrics->AverageGfxActivity <= SMU_13_0_0_BUSY_THRESHOLD) gpu_metrics->average_gfxclk_frequency = metrics->AverageGfxclkFrequencyPostDs; @@ -2216,7 +2222,8 @@ static int smu_v13_0_0_restore_user_od_settings(struct smu_context *smu) user_od_table->OverDriveTable.FeatureCtrlMask = BIT(PP_OD_FEATURE_GFXCLK_BIT) | BIT(PP_OD_FEATURE_UCLK_BIT) | BIT(PP_OD_FEATURE_GFX_VF_CURVE_BIT) | - BIT(PP_OD_FEATURE_FAN_CURVE_BIT); + BIT(PP_OD_FEATURE_FAN_CURVE_BIT) | + BIT(PP_OD_FEATURE_ZERO_FAN_BIT); res = smu_v13_0_0_upload_overdrive_table(smu, user_od_table); user_od_table->OverDriveTable.FeatureCtrlMask = 0; if (res == 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 af0482c9caa7..5500a0f12f0e 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 @@ -2065,7 +2065,8 @@ static ssize_t smu_v13_0_7_get_gpu_metrics(struct smu_context *smu, metrics->Vcn1ActivityPercentage); gpu_metrics->average_socket_power = metrics->AverageSocketPower; - gpu_metrics->energy_accumulator = metrics->EnergyAccumulator; + gpu_metrics->energy_accumulator = smu->smc_fw_version <= 0x00521400 ? + metrics->EnergyAccumulator : UINT_MAX; if (metrics->AverageGfxActivity <= SMU_13_0_7_BUSY_THRESHOLD) gpu_metrics->average_gfxclk_frequency = metrics->AverageGfxclkFrequencyPostDs; @@ -2223,7 +2224,8 @@ static int smu_v13_0_7_restore_user_od_settings(struct smu_context *smu) user_od_table->OverDriveTable.FeatureCtrlMask = BIT(PP_OD_FEATURE_GFXCLK_BIT) | BIT(PP_OD_FEATURE_UCLK_BIT) | BIT(PP_OD_FEATURE_GFX_VF_CURVE_BIT) | - BIT(PP_OD_FEATURE_FAN_CURVE_BIT); + BIT(PP_OD_FEATURE_FAN_CURVE_BIT) | + BIT(PP_OD_FEATURE_ZERO_FAN_BIT); res = smu_v13_0_7_upload_overdrive_table(smu, user_od_table); user_od_table->OverDriveTable.FeatureCtrlMask = 0; if (res == 0) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c index 9994d4369da8..73762d9b5969 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c @@ -2311,7 +2311,8 @@ static int smu_v14_0_2_restore_user_od_settings(struct smu_context *smu) user_od_table->OverDriveTable.FeatureCtrlMask = BIT(PP_OD_FEATURE_GFXCLK_BIT) | BIT(PP_OD_FEATURE_UCLK_BIT) | BIT(PP_OD_FEATURE_GFX_VF_CURVE_BIT) | - BIT(PP_OD_FEATURE_FAN_CURVE_BIT); + BIT(PP_OD_FEATURE_FAN_CURVE_BIT) | + BIT(PP_OD_FEATURE_ZERO_FAN_BIT); res = smu_v14_0_2_upload_overdrive_table(smu, user_od_table); user_od_table->OverDriveTable.FeatureCtrlMask = 0; if (res == 0) diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c index 930aaa659c97..ec632f268644 100644 --- a/drivers/gpu/drm/bridge/samsung-dsim.c +++ b/drivers/gpu/drm/bridge/samsung-dsim.c @@ -1881,6 +1881,14 @@ static int samsung_dsim_register_te_irq(struct samsung_dsim *dsi, struct device return 0; } +static void samsung_dsim_unregister_te_irq(struct samsung_dsim *dsi) +{ + if (dsi->te_gpio) { + free_irq(gpiod_to_irq(dsi->te_gpio), dsi); + gpiod_put(dsi->te_gpio); + } +} + static int samsung_dsim_host_attach(struct mipi_dsi_host *host, struct mipi_dsi_device *device) { @@ -1961,7 +1969,7 @@ of_find_panel_or_bridge: if (!(device->mode_flags & MIPI_DSI_MODE_VIDEO)) { ret = samsung_dsim_register_te_irq(dsi, &device->dev); if (ret) - return ret; + goto err_remove_bridge; } // The next bridge can be used by host_ops->attach @@ -1982,15 +1990,12 @@ of_find_panel_or_bridge: err_release_next_bridge: drm_bridge_put(dsi->bridge.next_bridge); dsi->bridge.next_bridge = NULL; - return ret; -} -static void samsung_dsim_unregister_te_irq(struct samsung_dsim *dsi) -{ - if (dsi->te_gpio) { - free_irq(gpiod_to_irq(dsi->te_gpio), dsi); - gpiod_put(dsi->te_gpio); - } + if (!(device->mode_flags & MIPI_DSI_MODE_VIDEO)) + samsung_dsim_unregister_te_irq(dsi); +err_remove_bridge: + drm_bridge_remove(&dsi->bridge); + return ret; } static int samsung_dsim_host_detach(struct mipi_dsi_host *host, diff --git a/drivers/gpu/drm/bridge/synopsys/dw-dp.c b/drivers/gpu/drm/bridge/synopsys/dw-dp.c index 4ab6922dd79c..fd23ca2834b0 100644 --- a/drivers/gpu/drm/bridge/synopsys/dw-dp.c +++ b/drivers/gpu/drm/bridge/synopsys/dw-dp.c @@ -2049,7 +2049,9 @@ struct dw_dp *dw_dp_bind(struct device *dev, struct drm_encoder *encoder, bridge->type = DRM_MODE_CONNECTOR_DisplayPort; bridge->ycbcr_420_allowed = true; - devm_drm_bridge_add(dev, bridge); + ret = devm_drm_bridge_add(dev, bridge); + if (ret) + return ERR_PTR(ret); dp->aux.dev = dev; dp->aux.drm_dev = encoder->dev; diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi83.c b/drivers/gpu/drm/bridge/ti-sn65dsi83.c index f6736b4457bb..17a885244e1e 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi83.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi83.c @@ -351,9 +351,9 @@ static u8 sn65dsi83_get_dsi_range(struct sn65dsi83 *ctx, * DSI_CLK = mode clock * bpp / dsi_data_lanes / 2 * the 2 is there because the bus is DDR. */ - return DIV_ROUND_UP(clamp((unsigned int)mode->clock * - mipi_dsi_pixel_format_to_bpp(ctx->dsi->format) / - ctx->dsi->lanes / 2, 40000U, 500000U), 5000U); + return clamp((unsigned int)mode->clock * + mipi_dsi_pixel_format_to_bpp(ctx->dsi->format) / + ctx->dsi->lanes / 2, 40000U, 500000U) / 5000U; } static u8 sn65dsi83_get_dsi_div(struct sn65dsi83 *ctx) @@ -517,6 +517,7 @@ static void sn65dsi83_atomic_pre_enable(struct drm_bridge *bridge, struct drm_atomic_state *state) { struct sn65dsi83 *ctx = bridge_to_sn65dsi83(bridge); + const unsigned int dual_factor = ctx->lvds_dual_link ? 2 : 1; const struct drm_bridge_state *bridge_state; const struct drm_crtc_state *crtc_state; const struct drm_display_mode *mode; @@ -653,18 +654,18 @@ static void sn65dsi83_atomic_pre_enable(struct drm_bridge *bridge, /* 32 + 1 pixel clock to ensure proper operation */ le16val = cpu_to_le16(32 + 1); regmap_bulk_write(ctx->regmap, REG_VID_CHA_SYNC_DELAY_LOW, &le16val, 2); - le16val = cpu_to_le16(mode->hsync_end - mode->hsync_start); + le16val = cpu_to_le16((mode->hsync_end - mode->hsync_start) / dual_factor); regmap_bulk_write(ctx->regmap, REG_VID_CHA_HSYNC_PULSE_WIDTH_LOW, &le16val, 2); le16val = cpu_to_le16(mode->vsync_end - mode->vsync_start); regmap_bulk_write(ctx->regmap, REG_VID_CHA_VSYNC_PULSE_WIDTH_LOW, &le16val, 2); regmap_write(ctx->regmap, REG_VID_CHA_HORIZONTAL_BACK_PORCH, - mode->htotal - mode->hsync_end); + (mode->htotal - mode->hsync_end) / dual_factor); regmap_write(ctx->regmap, REG_VID_CHA_VERTICAL_BACK_PORCH, mode->vtotal - mode->vsync_end); regmap_write(ctx->regmap, REG_VID_CHA_HORIZONTAL_FRONT_PORCH, - mode->hsync_start - mode->hdisplay); + (mode->hsync_start - mode->hdisplay) / dual_factor); regmap_write(ctx->regmap, REG_VID_CHA_VERTICAL_FRONT_PORCH, mode->vsync_start - mode->vdisplay); regmap_write(ctx->regmap, REG_VID_CHA_TEST_PATTERN, 0x00); diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c index 276d05d25ad8..98d64ad791d0 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c @@ -1415,6 +1415,7 @@ static int ti_sn_bridge_probe(struct auxiliary_device *adev, { struct ti_sn65dsi86 *pdata = dev_get_drvdata(adev->dev.parent); struct device_node *np = pdata->dev->of_node; + const struct i2c_client *client = to_i2c_client(pdata->dev); int ret; pdata->next_bridge = devm_drm_of_get_bridge(&adev->dev, np, 1, 0); @@ -1433,8 +1434,9 @@ static int ti_sn_bridge_probe(struct auxiliary_device *adev, ? DRM_MODE_CONNECTOR_DisplayPort : DRM_MODE_CONNECTOR_eDP; if (pdata->bridge.type == DRM_MODE_CONNECTOR_DisplayPort) { - pdata->bridge.ops = DRM_BRIDGE_OP_EDID | DRM_BRIDGE_OP_DETECT | - DRM_BRIDGE_OP_HPD; + pdata->bridge.ops = DRM_BRIDGE_OP_EDID | DRM_BRIDGE_OP_DETECT; + if (client->irq) + pdata->bridge.ops |= DRM_BRIDGE_OP_HPD; /* * If comms were already enabled they would have been enabled * with the wrong value of HPD_DISABLE. Update it now. Comms diff --git a/drivers/gpu/drm/drm_client_modeset.c b/drivers/gpu/drm/drm_client_modeset.c index 262b1b8773c5..bb49b8361271 100644 --- a/drivers/gpu/drm/drm_client_modeset.c +++ b/drivers/gpu/drm/drm_client_modeset.c @@ -930,7 +930,8 @@ int drm_client_modeset_probe(struct drm_client_dev *client, unsigned int width, mutex_unlock(&client->modeset_mutex); out: kfree(crtcs); - modes_destroy(dev, modes, connector_count); + if (modes) + modes_destroy(dev, modes, connector_count); kfree(modes); kfree(offsets); kfree(enabled); diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c index 24180bfdf5a2..9ef9e52c0547 100644 --- a/drivers/gpu/drm/drm_gpusvm.c +++ b/drivers/gpu/drm/drm_gpusvm.c @@ -1338,14 +1338,14 @@ bool drm_gpusvm_range_pages_valid(struct drm_gpusvm *gpusvm, EXPORT_SYMBOL_GPL(drm_gpusvm_range_pages_valid); /** - * drm_gpusvm_range_pages_valid_unlocked() - GPU SVM range pages valid unlocked + * drm_gpusvm_pages_valid_unlocked() - GPU SVM pages valid unlocked * @gpusvm: Pointer to the GPU SVM structure - * @range: Pointer to the GPU SVM range structure + * @svm_pages: Pointer to the GPU SVM pages structure * - * This function determines if a GPU SVM range pages are valid. Expected be - * called without holding gpusvm->notifier_lock. + * This function determines if a GPU SVM pages are valid. Expected be called + * without holding gpusvm->notifier_lock. * - * Return: True if GPU SVM range has valid pages, False otherwise + * Return: True if GPU SVM pages are valid, False otherwise */ static bool drm_gpusvm_pages_valid_unlocked(struct drm_gpusvm *gpusvm, struct drm_gpusvm_pages *svm_pages) diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c index bdc79140875c..862675ac5bb2 100644 --- a/drivers/gpu/drm/drm_pagemap.c +++ b/drivers/gpu/drm/drm_pagemap.c @@ -480,18 +480,8 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, .start = start, .end = end, .pgmap_owner = pagemap->owner, - /* - * FIXME: MIGRATE_VMA_SELECT_DEVICE_PRIVATE intermittently - * causes 'xe_exec_system_allocator --r *race*no*' to trigger aa - * engine reset and a hard hang due to getting stuck on a folio - * lock. This should work and needs to be root-caused. The only - * downside of not selecting MIGRATE_VMA_SELECT_DEVICE_PRIVATE - * is that device-to-device migrations won’t work; instead, - * memory will bounce through system memory. This path should be - * rare and only occur when the madvise attributes of memory are - * changed or atomics are being used. - */ - .flags = MIGRATE_VMA_SELECT_SYSTEM | MIGRATE_VMA_SELECT_DEVICE_COHERENT, + .flags = MIGRATE_VMA_SELECT_SYSTEM | MIGRATE_VMA_SELECT_DEVICE_COHERENT | + MIGRATE_VMA_SELECT_DEVICE_PRIVATE, }; unsigned long i, npages = npages_in_range(start, end); unsigned long own_pages = 0, migrated_pages = 0; diff --git a/drivers/gpu/drm/gud/gud_drv.c b/drivers/gpu/drm/gud/gud_drv.c index d0122d477610..17c2dead2c13 100644 --- a/drivers/gpu/drm/gud/gud_drv.c +++ b/drivers/gpu/drm/gud/gud_drv.c @@ -339,7 +339,9 @@ static int gud_stats_debugfs(struct seq_file *m, void *data) } static const struct drm_crtc_helper_funcs gud_crtc_helper_funcs = { - .atomic_check = drm_crtc_helper_atomic_check + .atomic_check = drm_crtc_helper_atomic_check, + .atomic_enable = gud_crtc_atomic_enable, + .atomic_disable = gud_crtc_atomic_disable, }; static const struct drm_crtc_funcs gud_crtc_funcs = { @@ -364,6 +366,10 @@ static const struct drm_plane_funcs gud_plane_funcs = { DRM_GEM_SHADOW_PLANE_FUNCS, }; +static const struct drm_mode_config_helper_funcs gud_mode_config_helpers = { + .atomic_commit_tail = drm_atomic_helper_commit_tail_rpm, +}; + static const struct drm_mode_config_funcs gud_mode_config_funcs = { .fb_create = drm_gem_fb_create_with_dirty, .atomic_check = drm_atomic_helper_check, @@ -499,6 +505,7 @@ static int gud_probe(struct usb_interface *intf, const struct usb_device_id *id) drm->mode_config.min_height = le32_to_cpu(desc.min_height); drm->mode_config.max_height = le32_to_cpu(desc.max_height); drm->mode_config.funcs = &gud_mode_config_funcs; + drm->mode_config.helper_private = &gud_mode_config_helpers; /* Format init */ formats_dev = devm_kmalloc(dev, GUD_FORMATS_MAX_NUM, GFP_KERNEL); diff --git a/drivers/gpu/drm/gud/gud_internal.h b/drivers/gpu/drm/gud/gud_internal.h index d27c31648341..8eec8335f5f9 100644 --- a/drivers/gpu/drm/gud/gud_internal.h +++ b/drivers/gpu/drm/gud/gud_internal.h @@ -62,6 +62,10 @@ int gud_usb_set_u8(struct gud_device *gdrm, u8 request, u8 val); void gud_clear_damage(struct gud_device *gdrm); void gud_flush_work(struct work_struct *work); +void gud_crtc_atomic_enable(struct drm_crtc *crtc, + struct drm_atomic_state *state); +void gud_crtc_atomic_disable(struct drm_crtc *crtc, + struct drm_atomic_state *state); int gud_plane_atomic_check(struct drm_plane *plane, struct drm_atomic_state *state); void gud_plane_atomic_update(struct drm_plane *plane, diff --git a/drivers/gpu/drm/gud/gud_pipe.c b/drivers/gpu/drm/gud/gud_pipe.c index 4b77be94348d..b355bf4d3389 100644 --- a/drivers/gpu/drm/gud/gud_pipe.c +++ b/drivers/gpu/drm/gud/gud_pipe.c @@ -580,6 +580,39 @@ out: return ret; } +void gud_crtc_atomic_enable(struct drm_crtc *crtc, + struct drm_atomic_state *state) +{ + struct drm_device *drm = crtc->dev; + struct gud_device *gdrm = to_gud_device(drm); + int idx; + + if (!drm_dev_enter(drm, &idx)) + return; + + gud_usb_set_u8(gdrm, GUD_REQ_SET_CONTROLLER_ENABLE, 1); + gud_usb_set(gdrm, GUD_REQ_SET_STATE_COMMIT, 0, NULL, 0); + gud_usb_set_u8(gdrm, GUD_REQ_SET_DISPLAY_ENABLE, 1); + + drm_dev_exit(idx); +} + +void gud_crtc_atomic_disable(struct drm_crtc *crtc, + struct drm_atomic_state *state) +{ + struct drm_device *drm = crtc->dev; + struct gud_device *gdrm = to_gud_device(drm); + int idx; + + if (!drm_dev_enter(drm, &idx)) + return; + + gud_usb_set_u8(gdrm, GUD_REQ_SET_DISPLAY_ENABLE, 0); + gud_usb_set_u8(gdrm, GUD_REQ_SET_CONTROLLER_ENABLE, 0); + + drm_dev_exit(idx); +} + void gud_plane_atomic_update(struct drm_plane *plane, struct drm_atomic_state *atomic_state) { @@ -607,24 +640,12 @@ void gud_plane_atomic_update(struct drm_plane *plane, mutex_unlock(&gdrm->damage_lock); } - if (!drm_dev_enter(drm, &idx)) + if (!crtc || !drm_dev_enter(drm, &idx)) return; - if (!old_state->fb) - gud_usb_set_u8(gdrm, GUD_REQ_SET_CONTROLLER_ENABLE, 1); - - if (fb && (crtc->state->mode_changed || crtc->state->connectors_changed)) - gud_usb_set(gdrm, GUD_REQ_SET_STATE_COMMIT, 0, NULL, 0); - - if (crtc->state->active_changed) - gud_usb_set_u8(gdrm, GUD_REQ_SET_DISPLAY_ENABLE, crtc->state->active); - - if (!fb) - goto ctrl_disable; - ret = drm_gem_fb_begin_cpu_access(fb, DMA_FROM_DEVICE); if (ret) - goto ctrl_disable; + goto out; drm_atomic_helper_damage_iter_init(&iter, old_state, new_state); drm_atomic_for_each_plane_damage(&iter, &damage) @@ -632,9 +653,6 @@ void gud_plane_atomic_update(struct drm_plane *plane, drm_gem_fb_end_cpu_access(fb, DMA_FROM_DEVICE); -ctrl_disable: - if (!crtc->state->enable) - gud_usb_set_u8(gdrm, GUD_REQ_SET_CONTROLLER_ENABLE, 0); - +out: drm_dev_exit(idx); } diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c b/drivers/gpu/drm/i915/display/intel_alpm.c index 7ce8c674bb03..f4f1b68f7543 100644 --- a/drivers/gpu/drm/i915/display/intel_alpm.c +++ b/drivers/gpu/drm/i915/display/intel_alpm.c @@ -43,12 +43,6 @@ bool intel_alpm_is_alpm_aux_less(struct intel_dp *intel_dp, void intel_alpm_init(struct intel_dp *intel_dp) { - u8 dpcd; - - if (drm_dp_dpcd_readb(&intel_dp->aux, DP_RECEIVER_ALPM_CAP, &dpcd) < 0) - return; - - intel_dp->alpm_dpcd = dpcd; mutex_init(&intel_dp->alpm.lock); } @@ -562,12 +556,7 @@ void intel_alpm_disable(struct intel_dp *intel_dp) mutex_lock(&intel_dp->alpm.lock); intel_de_rmw(display, ALPM_CTL(display, cpu_transcoder), - ALPM_CTL_ALPM_ENABLE | ALPM_CTL_LOBF_ENABLE | - ALPM_CTL_ALPM_AUX_LESS_ENABLE, 0); - - intel_de_rmw(display, - PORT_ALPM_CTL(cpu_transcoder), - PORT_ALPM_CTL_ALPM_AUX_LESS_ENABLE, 0); + ALPM_CTL_ALPM_ENABLE | ALPM_CTL_LOBF_ENABLE, 0); drm_dbg_kms(display->drm, "Disabling ALPM\n"); mutex_unlock(&intel_dp->alpm.lock); diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 3b8ba8ab76a1..c4246481fc2f 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -1614,7 +1614,6 @@ static void hsw_configure_cpu_transcoder(const struct intel_crtc_state *crtc_sta } intel_set_transcoder_timings(crtc_state); - intel_vrr_set_transcoder_timings(crtc_state); if (cpu_transcoder != TRANSCODER_EDP) intel_de_write(display, TRANS_MULT(display, cpu_transcoder), diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 559cf3bb23fd..696edf40b243 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -4577,6 +4577,7 @@ static bool intel_edp_init_dpcd(struct intel_dp *intel_dp, struct intel_connector *connector) { struct intel_display *display = to_intel_display(intel_dp); + int ret; /* this function is meant to be called only once */ drm_WARN_ON(display->drm, intel_dp->dpcd[DP_DPCD_REV] != 0); @@ -4616,6 +4617,12 @@ intel_edp_init_dpcd(struct intel_dp *intel_dp, struct intel_connector *connector */ intel_dp_init_source_oui(intel_dp); + /* Read the ALPM DPCD caps */ + ret = drm_dp_dpcd_read_byte(&intel_dp->aux, DP_RECEIVER_ALPM_CAP, + &intel_dp->alpm_dpcd); + if (ret < 0) + return false; + /* * This has to be called after intel_dp->edp_dpcd is filled, PSR checks * for SET_POWER_CAPABLE bit in intel_dp->edp_dpcd[1] diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index 62208ffc5101..b7302a32ded4 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -1307,9 +1307,14 @@ static bool psr2_granularity_check(struct intel_crtc_state *crtc_state, u16 sink_y_granularity = crtc_state->has_panel_replay ? connector->dp.panel_replay_caps.su_y_granularity : connector->dp.psr_caps.su_y_granularity; - u16 sink_w_granularity = crtc_state->has_panel_replay ? - connector->dp.panel_replay_caps.su_w_granularity : - connector->dp.psr_caps.su_w_granularity; + u16 sink_w_granularity; + + if (crtc_state->has_panel_replay) + sink_w_granularity = connector->dp.panel_replay_caps.su_w_granularity == + DP_PANEL_REPLAY_FULL_LINE_GRANULARITY ? + crtc_hdisplay : connector->dp.panel_replay_caps.su_w_granularity; + else + sink_w_granularity = connector->dp.psr_caps.su_w_granularity; /* PSR2 HW only send full lines so we only need to validate the width */ if (crtc_hdisplay % sink_w_granularity) @@ -2614,6 +2619,12 @@ void intel_psr2_program_trans_man_trk_ctl(struct intel_dsb *dsb, intel_de_write_dsb(display, dsb, PIPE_SRCSZ_ERLY_TPT(crtc->pipe), crtc_state->pipe_srcsz_early_tpt); + + if (!crtc_state->dsc.compression_enable) + return; + + intel_dsc_su_et_parameters_configure(dsb, encoder, crtc_state, + drm_rect_height(&crtc_state->psr2_su_area)); } static void psr2_man_trk_ctl_calc(struct intel_crtc_state *crtc_state, @@ -2684,11 +2695,12 @@ static void clip_area_update(struct drm_rect *overlap_damage_area, overlap_damage_area->y2 = damage_area->y2; } -static void intel_psr2_sel_fetch_pipe_alignment(struct intel_crtc_state *crtc_state) +static bool intel_psr2_sel_fetch_pipe_alignment(struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); const struct drm_dsc_config *vdsc_cfg = &crtc_state->dsc.config; u16 y_alignment; + bool su_area_changed = false; /* ADLP aligns the SU region to vdsc slice height in case dsc is enabled */ if (crtc_state->dsc.compression_enable && @@ -2697,10 +2709,18 @@ static void intel_psr2_sel_fetch_pipe_alignment(struct intel_crtc_state *crtc_st else y_alignment = crtc_state->su_y_granularity; - crtc_state->psr2_su_area.y1 -= crtc_state->psr2_su_area.y1 % y_alignment; - if (crtc_state->psr2_su_area.y2 % y_alignment) + if (crtc_state->psr2_su_area.y1 % y_alignment) { + crtc_state->psr2_su_area.y1 -= crtc_state->psr2_su_area.y1 % y_alignment; + su_area_changed = true; + } + + if (crtc_state->psr2_su_area.y2 % y_alignment) { crtc_state->psr2_su_area.y2 = ((crtc_state->psr2_su_area.y2 / y_alignment) + 1) * y_alignment; + su_area_changed = true; + } + + return su_area_changed; } /* @@ -2834,7 +2854,7 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); struct intel_plane_state *new_plane_state, *old_plane_state; struct intel_plane *plane; - bool full_update = false, cursor_in_su_area = false; + bool full_update = false, su_area_changed; int i, ret; if (!crtc_state->enable_psr2_sel_fetch) @@ -2941,15 +2961,32 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, if (ret) return ret; - /* - * Adjust su area to cover cursor fully as necessary (early - * transport). This needs to be done after - * drm_atomic_add_affected_planes to ensure visible cursor is added into - * affected planes even when cursor is not updated by itself. - */ - intel_psr2_sel_fetch_et_alignment(state, crtc, &cursor_in_su_area); + do { + bool cursor_in_su_area; + + /* + * Adjust su area to cover cursor fully as necessary + * (early transport). This needs to be done after + * drm_atomic_add_affected_planes to ensure visible + * cursor is added into affected planes even when + * cursor is not updated by itself. + */ + intel_psr2_sel_fetch_et_alignment(state, crtc, &cursor_in_su_area); - intel_psr2_sel_fetch_pipe_alignment(crtc_state); + su_area_changed = intel_psr2_sel_fetch_pipe_alignment(crtc_state); + + /* + * If the cursor was outside the SU area before + * alignment, the alignment step (which only expands + * SU) may pull the cursor partially inside, so we + * must run ET alignment again to fully cover it. But + * if the cursor was already fully inside before + * alignment, expanding the SU area won't change that, + * so no further work is needed. + */ + if (cursor_in_su_area) + break; + } while (su_area_changed); /* * Now that we have the pipe damaged area check if it intersect with @@ -3009,6 +3046,10 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, } skip_sel_fetch_set_loop: + if (full_update) + clip_area_update(&crtc_state->psr2_su_area, &crtc_state->pipe_src, + &crtc_state->pipe_src); + psr2_man_trk_ctl_calc(crtc_state, full_update); crtc_state->pipe_srcsz_early_tpt = psr2_pipe_srcsz_early_tpt_calc(crtc_state, full_update); diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c index 5493082f30a7..2065dac1e3fd 100644 --- a/drivers/gpu/drm/i915/display/intel_vdsc.c +++ b/drivers/gpu/drm/i915/display/intel_vdsc.c @@ -767,6 +767,29 @@ void intel_dsc_dp_pps_write(struct intel_encoder *encoder, sizeof(dp_dsc_pps_sdp)); } +void intel_dsc_su_et_parameters_configure(struct intel_dsb *dsb, struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, int su_lines) +{ + struct intel_display *display = to_intel_display(crtc_state); + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + const struct drm_dsc_config *vdsc_cfg = &crtc_state->dsc.config; + enum pipe pipe = crtc->pipe; + int vdsc_instances_per_pipe = intel_dsc_get_vdsc_per_pipe(crtc_state); + int slice_row_per_frame = su_lines / vdsc_cfg->slice_height; + u32 val; + + drm_WARN_ON_ONCE(display->drm, su_lines % vdsc_cfg->slice_height); + drm_WARN_ON_ONCE(display->drm, vdsc_instances_per_pipe > 2); + + val = DSC_SUPS0_SU_SLICE_ROW_PER_FRAME(slice_row_per_frame); + val |= DSC_SUPS0_SU_PIC_HEIGHT(su_lines); + + intel_de_write_dsb(display, dsb, LNL_DSC0_SU_PARAMETER_SET_0(pipe), val); + + if (vdsc_instances_per_pipe == 2) + intel_de_write_dsb(display, dsb, LNL_DSC1_SU_PARAMETER_SET_0(pipe), val); +} + static i915_reg_t dss_ctl1_reg(struct intel_crtc *crtc, enum transcoder cpu_transcoder) { return is_pipe_dsc(crtc, cpu_transcoder) ? diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.h b/drivers/gpu/drm/i915/display/intel_vdsc.h index 99f64ac54b27..99bb9042592a 100644 --- a/drivers/gpu/drm/i915/display/intel_vdsc.h +++ b/drivers/gpu/drm/i915/display/intel_vdsc.h @@ -13,6 +13,7 @@ struct drm_printer; enum transcoder; struct intel_crtc; struct intel_crtc_state; +struct intel_dsb; struct intel_encoder; bool intel_dsc_source_support(const struct intel_crtc_state *crtc_state); @@ -31,6 +32,8 @@ void intel_dsc_dsi_pps_write(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); void intel_dsc_dp_pps_write(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); +void intel_dsc_su_et_parameters_configure(struct intel_dsb *dsb, struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, int su_lines); void intel_vdsc_state_dump(struct drm_printer *p, int indent, const struct intel_crtc_state *crtc_state); int intel_vdsc_min_cdclk(const struct intel_crtc_state *crtc_state); diff --git a/drivers/gpu/drm/i915/display/intel_vdsc_regs.h b/drivers/gpu/drm/i915/display/intel_vdsc_regs.h index 2d478a84b07c..2b2e3c1b8138 100644 --- a/drivers/gpu/drm/i915/display/intel_vdsc_regs.h +++ b/drivers/gpu/drm/i915/display/intel_vdsc_regs.h @@ -196,6 +196,18 @@ #define DSC_PPS18_NSL_BPG_OFFSET(offset) REG_FIELD_PREP(DSC_PPS18_NSL_BPG_OFFSET_MASK, offset) #define DSC_PPS18_SL_OFFSET_ADJ(offset) REG_FIELD_PREP(DSC_PPS18_SL_OFFSET_ADJ_MASK, offset) +#define _LNL_DSC0_SU_PARAMETER_SET_0_PA 0x78064 +#define _LNL_DSC1_SU_PARAMETER_SET_0_PA 0x78164 +#define _LNL_DSC0_SU_PARAMETER_SET_0_PB 0x78264 +#define _LNL_DSC1_SU_PARAMETER_SET_0_PB 0x78364 +#define LNL_DSC0_SU_PARAMETER_SET_0(pipe) _MMIO_PIPE((pipe), _LNL_DSC0_SU_PARAMETER_SET_0_PA, _LNL_DSC0_SU_PARAMETER_SET_0_PB) +#define LNL_DSC1_SU_PARAMETER_SET_0(pipe) _MMIO_PIPE((pipe), _LNL_DSC1_SU_PARAMETER_SET_0_PA, _LNL_DSC1_SU_PARAMETER_SET_0_PB) + +#define DSC_SUPS0_SU_SLICE_ROW_PER_FRAME_MASK REG_GENMASK(31, 20) +#define DSC_SUPS0_SU_SLICE_ROW_PER_FRAME(rows) REG_FIELD_PREP(DSC_SUPS0_SU_SLICE_ROW_PER_FRAME_MASK, (rows)) +#define DSC_SUPS0_SU_PIC_HEIGHT_MASK REG_GENMASK(15, 0) +#define DSC_SUPS0_SU_PIC_HEIGHT(h) REG_FIELD_PREP(DSC_SUPS0_SU_PIC_HEIGHT_MASK, (h)) + /* Icelake Rate Control Buffer Threshold Registers */ #define DSCA_RC_BUF_THRESH_0 _MMIO(0x6B230) #define DSCA_RC_BUF_THRESH_0_UDW _MMIO(0x6B230 + 4) diff --git a/drivers/gpu/drm/i915/display/intel_vrr.c b/drivers/gpu/drm/i915/display/intel_vrr.c index db74744ddb31..bea005752327 100644 --- a/drivers/gpu/drm/i915/display/intel_vrr.c +++ b/drivers/gpu/drm/i915/display/intel_vrr.c @@ -598,6 +598,18 @@ void intel_vrr_set_transcoder_timings(const struct intel_crtc_state *crtc_state) return; /* + * Bspec says: + * "(note: VRR needs to be programmed after + * TRANS_DDI_FUNC_CTL and before TRANS_CONF)." + * + * In practice it turns out that ICL can hang if + * TRANS_VRR_VMAX/FLIPLINE are written before + * enabling TRANS_DDI_FUNC_CTL. + */ + drm_WARN_ON(display->drm, + !(intel_de_read(display, TRANS_DDI_FUNC_CTL(display, cpu_transcoder)) & TRANS_DDI_FUNC_ENABLE)); + + /* * This bit seems to have two meanings depending on the platform: * TGL: generate VRR "safe window" for DSB vblank waits * ADL/DG2: make TRANS_SET_CONTEXT_LATENCY effective with VRR @@ -939,6 +951,8 @@ void intel_vrr_transcoder_enable(const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); + intel_vrr_set_transcoder_timings(crtc_state); + if (!intel_vrr_possible(crtc_state)) return; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c index c6c64ba29bc4..720a9ad39aa2 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c @@ -153,8 +153,12 @@ int shmem_sg_alloc_table(struct drm_i915_private *i915, struct sg_table *st, } } while (1); - nr_pages = min_t(unsigned long, - folio_nr_pages(folio), page_count - i); + nr_pages = min_array(((unsigned long[]) { + folio_nr_pages(folio), + page_count - i, + max_segment / PAGE_SIZE, + }), 3); + if (!i || sg->length >= max_segment || folio_pfn(folio) != next_pfn) { @@ -164,7 +168,9 @@ int shmem_sg_alloc_table(struct drm_i915_private *i915, struct sg_table *st, st->nents++; sg_set_folio(sg, folio, nr_pages * PAGE_SIZE, 0); } else { - /* XXX: could overflow? */ + nr_pages = min_t(unsigned long, nr_pages, + (max_segment - sg->length) / PAGE_SIZE); + sg->length += nr_pages * PAGE_SIZE; } next_pfn = folio_pfn(folio) + nr_pages; diff --git a/drivers/gpu/drm/imx/ipuv3/parallel-display.c b/drivers/gpu/drm/imx/ipuv3/parallel-display.c index dea85577513a..4ce772bc3cb3 100644 --- a/drivers/gpu/drm/imx/ipuv3/parallel-display.c +++ b/drivers/gpu/drm/imx/ipuv3/parallel-display.c @@ -256,7 +256,9 @@ static int imx_pd_probe(struct platform_device *pdev) platform_set_drvdata(pdev, imxpd); - devm_drm_bridge_add(dev, &imxpd->bridge); + ret = devm_drm_bridge_add(dev, &imxpd->bridge); + if (ret) + return ret; return component_add(dev, &imx_pd_ops); } diff --git a/drivers/gpu/drm/logicvc/logicvc_drm.c b/drivers/gpu/drm/logicvc/logicvc_drm.c index 204b0fee55d0..bbebf4fc7f51 100644 --- a/drivers/gpu/drm/logicvc/logicvc_drm.c +++ b/drivers/gpu/drm/logicvc/logicvc_drm.c @@ -92,7 +92,6 @@ static int logicvc_drm_config_parse(struct logicvc_drm *logicvc) struct device *dev = drm_dev->dev; struct device_node *of_node = dev->of_node; struct logicvc_drm_config *config = &logicvc->config; - struct device_node *layers_node; int ret; logicvc_of_property_parse_bool(of_node, LOGICVC_OF_PROPERTY_DITHERING, @@ -128,7 +127,8 @@ static int logicvc_drm_config_parse(struct logicvc_drm *logicvc) if (ret) return ret; - layers_node = of_get_child_by_name(of_node, "layers"); + struct device_node *layers_node __free(device_node) = + of_get_child_by_name(of_node, "layers"); if (!layers_node) { drm_err(drm_dev, "Missing non-optional layers node\n"); return -EINVAL; diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c index d77b4774d414..e2225c5ba647 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c @@ -78,7 +78,7 @@ static void a2xx_gpummu_destroy(struct msm_mmu *mmu) { struct a2xx_gpummu *gpummu = to_a2xx_gpummu(mmu); - dma_free_attrs(mmu->dev, TABLE_SIZE, gpummu->table, gpummu->pt_base, + dma_free_attrs(mmu->dev, TABLE_SIZE + 32, gpummu->table, gpummu->pt_base, DMA_ATTR_FORCE_CONTIGUOUS); kfree(gpummu); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c index 550a53a7865e..38561f26837e 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c @@ -1759,7 +1759,7 @@ static const u32 x285_protect_regs[] = { A6XX_PROTECT_NORDWR(0x27c06, 0x0000), }; -DECLARE_ADRENO_PROTECT(x285_protect, 64); +DECLARE_ADRENO_PROTECT(x285_protect, 15); static const struct adreno_reglist_pipe a840_nonctxt_regs[] = { { REG_A8XX_CP_SMMU_STREAM_ID_LPAC, 0x00000101, BIT(PIPE_NONE) }, @@ -1966,5 +1966,4 @@ static inline __always_unused void __build_asserts(void) BUILD_BUG_ON(a660_protect.count > a660_protect.count_max); BUILD_BUG_ON(a690_protect.count > a690_protect.count_max); BUILD_BUG_ON(a730_protect.count > a730_protect.count_max); - BUILD_BUG_ON(a840_protect.count > a840_protect.count_max); } diff --git a/drivers/gpu/drm/msm/adreno/a8xx_gpu.c b/drivers/gpu/drm/msm/adreno/a8xx_gpu.c index 5a320f5bde41..b1887e0cf698 100644 --- a/drivers/gpu/drm/msm/adreno/a8xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a8xx_gpu.c @@ -310,11 +310,21 @@ static void a8xx_set_ubwc_config(struct msm_gpu *gpu) hbb = cfg->highest_bank_bit - 13; hbb_hi = hbb >> 2; hbb_lo = hbb & 3; - a8xx_write_pipe(gpu, PIPE_BV, REG_A8XX_GRAS_NC_MODE_CNTL, hbb << 5); - a8xx_write_pipe(gpu, PIPE_BR, REG_A8XX_GRAS_NC_MODE_CNTL, hbb << 5); + + a8xx_write_pipe(gpu, PIPE_BV, REG_A8XX_GRAS_NC_MODE_CNTL, + hbb << 5 | + level3_swizzling_dis << 4 | + level2_swizzling_dis << 3); + + a8xx_write_pipe(gpu, PIPE_BR, REG_A8XX_GRAS_NC_MODE_CNTL, + hbb << 5 | + level3_swizzling_dis << 4 | + level2_swizzling_dis << 3); a8xx_write_pipe(gpu, PIPE_BR, REG_A8XX_RB_CCU_NC_MODE_CNTL, yuvnotcomptofc << 6 | + level3_swizzling_dis << 5 | + level2_swizzling_dis << 4 | hbb_hi << 3 | hbb_lo << 1); diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c index 554d746f115b..4edfe80c5be7 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_device.c +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -302,6 +302,7 @@ static const struct of_device_id dt_match[] = { { .compatible = "qcom,kgsl-3d0" }, {} }; +MODULE_DEVICE_TABLE(of, dt_match); static int adreno_runtime_resume(struct device *dev) { diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h index 303d33dc7783..9f2bceca1789 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h @@ -133,7 +133,7 @@ static const struct dpu_sspp_cfg sc8280xp_sspp[] = { static const struct dpu_lm_cfg sc8280xp_lm[] = { { .name = "lm_0", .id = LM_0, - .base = 0x44000, .len = 0x320, + .base = 0x44000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, @@ -141,7 +141,7 @@ static const struct dpu_lm_cfg sc8280xp_lm[] = { .dspp = DSPP_0, }, { .name = "lm_1", .id = LM_1, - .base = 0x45000, .len = 0x320, + .base = 0x45000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, @@ -149,7 +149,7 @@ static const struct dpu_lm_cfg sc8280xp_lm[] = { .dspp = DSPP_1, }, { .name = "lm_2", .id = LM_2, - .base = 0x46000, .len = 0x320, + .base = 0x46000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, @@ -157,7 +157,7 @@ static const struct dpu_lm_cfg sc8280xp_lm[] = { .dspp = DSPP_2, }, { .name = "lm_3", .id = LM_3, - .base = 0x47000, .len = 0x320, + .base = 0x47000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, @@ -165,14 +165,14 @@ static const struct dpu_lm_cfg sc8280xp_lm[] = { .dspp = DSPP_3, }, { .name = "lm_4", .id = LM_4, - .base = 0x48000, .len = 0x320, + .base = 0x48000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, - .base = 0x49000, .len = 0x320, + .base = 0x49000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h index b09a6af4c474..04b22167f93d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h @@ -134,7 +134,7 @@ static const struct dpu_sspp_cfg sm8450_sspp[] = { static const struct dpu_lm_cfg sm8450_lm[] = { { .name = "lm_0", .id = LM_0, - .base = 0x44000, .len = 0x320, + .base = 0x44000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, @@ -142,7 +142,7 @@ static const struct dpu_lm_cfg sm8450_lm[] = { .dspp = DSPP_0, }, { .name = "lm_1", .id = LM_1, - .base = 0x45000, .len = 0x320, + .base = 0x45000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, @@ -150,7 +150,7 @@ static const struct dpu_lm_cfg sm8450_lm[] = { .dspp = DSPP_1, }, { .name = "lm_2", .id = LM_2, - .base = 0x46000, .len = 0x320, + .base = 0x46000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, @@ -158,7 +158,7 @@ static const struct dpu_lm_cfg sm8450_lm[] = { .dspp = DSPP_2, }, { .name = "lm_3", .id = LM_3, - .base = 0x47000, .len = 0x320, + .base = 0x47000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, @@ -166,14 +166,14 @@ static const struct dpu_lm_cfg sm8450_lm[] = { .dspp = DSPP_3, }, { .name = "lm_4", .id = LM_4, - .base = 0x48000, .len = 0x320, + .base = 0x48000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, - .base = 0x49000, .len = 0x320, + .base = 0x49000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h index 0f7b4a224e4c..42cf3bd5a12a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h @@ -366,8 +366,8 @@ static const struct dpu_intf_cfg sa8775p_intf[] = { .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ .prog_fetch_lines_worst_case = 24, - .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17), - .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16), + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17), }, { .name = "intf_7", .id = INTF_7, .base = 0x3b000, .len = 0x280, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h index 465b6460f875..4c7eb55d474c 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h @@ -131,7 +131,7 @@ static const struct dpu_sspp_cfg sm8550_sspp[] = { static const struct dpu_lm_cfg sm8550_lm[] = { { .name = "lm_0", .id = LM_0, - .base = 0x44000, .len = 0x320, + .base = 0x44000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, @@ -139,7 +139,7 @@ static const struct dpu_lm_cfg sm8550_lm[] = { .dspp = DSPP_0, }, { .name = "lm_1", .id = LM_1, - .base = 0x45000, .len = 0x320, + .base = 0x45000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, @@ -147,7 +147,7 @@ static const struct dpu_lm_cfg sm8550_lm[] = { .dspp = DSPP_1, }, { .name = "lm_2", .id = LM_2, - .base = 0x46000, .len = 0x320, + .base = 0x46000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, @@ -155,7 +155,7 @@ static const struct dpu_lm_cfg sm8550_lm[] = { .dspp = DSPP_2, }, { .name = "lm_3", .id = LM_3, - .base = 0x47000, .len = 0x320, + .base = 0x47000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, @@ -163,14 +163,14 @@ static const struct dpu_lm_cfg sm8550_lm[] = { .dspp = DSPP_3, }, { .name = "lm_4", .id = LM_4, - .base = 0x48000, .len = 0x320, + .base = 0x48000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, - .base = 0x49000, .len = 0x320, + .base = 0x49000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h index 6caa7d40f368..dec83ea8167d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h @@ -131,7 +131,7 @@ static const struct dpu_sspp_cfg sar2130p_sspp[] = { static const struct dpu_lm_cfg sar2130p_lm[] = { { .name = "lm_0", .id = LM_0, - .base = 0x44000, .len = 0x320, + .base = 0x44000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, @@ -139,7 +139,7 @@ static const struct dpu_lm_cfg sar2130p_lm[] = { .dspp = DSPP_0, }, { .name = "lm_1", .id = LM_1, - .base = 0x45000, .len = 0x320, + .base = 0x45000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, @@ -147,7 +147,7 @@ static const struct dpu_lm_cfg sar2130p_lm[] = { .dspp = DSPP_1, }, { .name = "lm_2", .id = LM_2, - .base = 0x46000, .len = 0x320, + .base = 0x46000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, @@ -155,7 +155,7 @@ static const struct dpu_lm_cfg sar2130p_lm[] = { .dspp = DSPP_2, }, { .name = "lm_3", .id = LM_3, - .base = 0x47000, .len = 0x320, + .base = 0x47000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, @@ -163,14 +163,14 @@ static const struct dpu_lm_cfg sar2130p_lm[] = { .dspp = DSPP_3, }, { .name = "lm_4", .id = LM_4, - .base = 0x48000, .len = 0x320, + .base = 0x48000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, - .base = 0x49000, .len = 0x320, + .base = 0x49000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h index 7243eebb85f3..52ff4baa668a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h @@ -130,7 +130,7 @@ static const struct dpu_sspp_cfg x1e80100_sspp[] = { static const struct dpu_lm_cfg x1e80100_lm[] = { { .name = "lm_0", .id = LM_0, - .base = 0x44000, .len = 0x320, + .base = 0x44000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, @@ -138,7 +138,7 @@ static const struct dpu_lm_cfg x1e80100_lm[] = { .dspp = DSPP_0, }, { .name = "lm_1", .id = LM_1, - .base = 0x45000, .len = 0x320, + .base = 0x45000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, @@ -146,7 +146,7 @@ static const struct dpu_lm_cfg x1e80100_lm[] = { .dspp = DSPP_1, }, { .name = "lm_2", .id = LM_2, - .base = 0x46000, .len = 0x320, + .base = 0x46000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, @@ -154,7 +154,7 @@ static const struct dpu_lm_cfg x1e80100_lm[] = { .dspp = DSPP_2, }, { .name = "lm_3", .id = LM_3, - .base = 0x47000, .len = 0x320, + .base = 0x47000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, @@ -162,14 +162,14 @@ static const struct dpu_lm_cfg x1e80100_lm[] = { .dspp = DSPP_3, }, { .name = "lm_4", .id = LM_4, - .base = 0x48000, .len = 0x320, + .base = 0x48000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, - .base = 0x49000, .len = 0x320, + .base = 0x49000, .len = 0x400, .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c index 188ee0af2c90..23dcbe1ce1b8 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c @@ -89,7 +89,7 @@ static void dpu_setup_dspp_gc(struct dpu_hw_dspp *ctx, base = ctx->cap->sblk->gc.base; if (!base) { - DRM_ERROR("invalid ctx %pK gc base\n", ctx); + DRM_ERROR("invalid ctx %p gc base\n", ctx); return; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp_v13.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp_v13.c index e65f1fc026fd..f8f96ad971d7 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp_v13.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp_v13.c @@ -156,11 +156,13 @@ static void dpu_hw_sspp_setup_pe_config_v13(struct dpu_hw_sspp *ctx, u8 color; u32 lr_pe[4], tb_pe[4]; const u32 bytemask = 0xff; - u32 offset = ctx->cap->sblk->sspp_rec0_blk.base; + u32 offset; if (!ctx || !pe_ext) return; + offset = ctx->cap->sblk->sspp_rec0_blk.base; + c = &ctx->hw; /* program SW pixel extension override for all pipes*/ for (color = 0; color < DPU_MAX_PLANES; color++) { diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c index 451a4fcf3e65..7e77d88f8959 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c @@ -350,26 +350,28 @@ static bool _dpu_rm_check_lm_and_get_connected_blks(struct dpu_rm *rm, return true; } -static bool dpu_rm_find_lms(struct dpu_rm *rm, - struct dpu_global_state *global_state, - uint32_t crtc_id, bool skip_dspp, - struct msm_display_topology *topology, - int *lm_idx, int *pp_idx, int *dspp_idx) +static int _dpu_rm_reserve_lms(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, + struct msm_display_topology *topology) { + int lm_idx[MAX_BLOCKS]; + int pp_idx[MAX_BLOCKS]; + int dspp_idx[MAX_BLOCKS] = {0}; int i, lm_count = 0; + if (!topology->num_lm) { + DPU_ERROR("zero LMs in topology\n"); + return -EINVAL; + } + /* Find a primary mixer */ for (i = 0; i < ARRAY_SIZE(rm->mixer_blks) && lm_count < topology->num_lm; i++) { if (!rm->mixer_blks[i]) continue; - if (skip_dspp && to_dpu_hw_mixer(rm->mixer_blks[i])->cap->dspp) { - DPU_DEBUG("Skipping LM_%d, skipping LMs with DSPPs\n", i); - continue; - } - /* * Reset lm_count to an even index. This will drop the previous * primary mixer if failed to find its peer. @@ -408,38 +410,12 @@ static bool dpu_rm_find_lms(struct dpu_rm *rm, } } - return lm_count == topology->num_lm; -} - -static int _dpu_rm_reserve_lms(struct dpu_rm *rm, - struct dpu_global_state *global_state, - uint32_t crtc_id, - struct msm_display_topology *topology) - -{ - int lm_idx[MAX_BLOCKS]; - int pp_idx[MAX_BLOCKS]; - int dspp_idx[MAX_BLOCKS] = {0}; - int i; - bool found; - - if (!topology->num_lm) { - DPU_ERROR("zero LMs in topology\n"); - return -EINVAL; - } - - /* Try using non-DSPP LM blocks first */ - found = dpu_rm_find_lms(rm, global_state, crtc_id, !topology->num_dspp, - topology, lm_idx, pp_idx, dspp_idx); - if (!found && !topology->num_dspp) - found = dpu_rm_find_lms(rm, global_state, crtc_id, false, - topology, lm_idx, pp_idx, dspp_idx); - if (!found) { + if (lm_count != topology->num_lm) { DPU_DEBUG("unable to find appropriate mixers\n"); return -ENAVAIL; } - for (i = 0; i < topology->num_lm; i++) { + for (i = 0; i < lm_count; i++) { global_state->mixer_to_crtc_id[lm_idx[i]] = crtc_id; global_state->pingpong_to_crtc_id[pp_idx[i]] = crtc_id; global_state->dspp_to_crtc_id[dspp_idx[i]] = diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c index e0de545d4077..db6da99375a1 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_host.c +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -584,13 +584,30 @@ void dsi_link_clk_disable_v2(struct msm_dsi_host *msm_host) * FIXME: Reconsider this if/when CMD mode handling is rewritten to use * transfer time and data overhead as a starting point of the calculations. */ -static unsigned long dsi_adjust_pclk_for_compression(const struct drm_display_mode *mode, - const struct drm_dsc_config *dsc) +static unsigned long +dsi_adjust_pclk_for_compression(const struct drm_display_mode *mode, + const struct drm_dsc_config *dsc, + bool is_bonded_dsi) { - int new_hdisplay = DIV_ROUND_UP(mode->hdisplay * drm_dsc_get_bpp_int(dsc), - dsc->bits_per_component * 3); + int hdisplay, new_hdisplay, new_htotal; - int new_htotal = mode->htotal - mode->hdisplay + new_hdisplay; + /* + * For bonded DSI, split hdisplay across two links and round up each + * half separately, passing the full hdisplay would only round up once. + * This also aligns with the hdisplay we program later in + * dsi_timing_setup() + */ + hdisplay = mode->hdisplay; + if (is_bonded_dsi) + hdisplay /= 2; + + new_hdisplay = DIV_ROUND_UP(hdisplay * drm_dsc_get_bpp_int(dsc), + dsc->bits_per_component * 3); + + if (is_bonded_dsi) + new_hdisplay *= 2; + + new_htotal = mode->htotal - mode->hdisplay + new_hdisplay; return mult_frac(mode->clock * 1000u, new_htotal, mode->htotal); } @@ -603,7 +620,7 @@ static unsigned long dsi_get_pclk_rate(const struct drm_display_mode *mode, pclk_rate = mode->clock * 1000u; if (dsc) - pclk_rate = dsi_adjust_pclk_for_compression(mode, dsc); + pclk_rate = dsi_adjust_pclk_for_compression(mode, dsc, is_bonded_dsi); /* * For bonded DSI mode, the current DRM mode has the complete width of the @@ -993,7 +1010,7 @@ static void dsi_timing_setup(struct msm_dsi_host *msm_host, bool is_bonded_dsi) if (msm_host->dsc) { struct drm_dsc_config *dsc = msm_host->dsc; - u32 bytes_per_pclk; + u32 bits_per_pclk; /* update dsc params with timing params */ if (!dsc || !mode->hdisplay || !mode->vdisplay) { @@ -1015,7 +1032,9 @@ static void dsi_timing_setup(struct msm_dsi_host *msm_host, bool is_bonded_dsi) /* * DPU sends 3 bytes per pclk cycle to DSI. If widebus is - * enabled, bus width is extended to 6 bytes. + * enabled, MDP always sends out 48-bit compressed data per + * pclk and on average, DSI consumes an amount of compressed + * data equivalent to the uncompressed pixel depth per pclk. * * Calculate the number of pclks needed to transmit one line of * the compressed data. @@ -1027,12 +1046,12 @@ static void dsi_timing_setup(struct msm_dsi_host *msm_host, bool is_bonded_dsi) * unused anyway. */ h_total -= hdisplay; - if (wide_bus_enabled && !(msm_host->mode_flags & MIPI_DSI_MODE_VIDEO)) - bytes_per_pclk = 6; + if (wide_bus_enabled) + bits_per_pclk = mipi_dsi_pixel_format_to_bpp(msm_host->format); else - bytes_per_pclk = 3; + bits_per_pclk = 24; - hdisplay = DIV_ROUND_UP(msm_dsc_get_bytes_per_line(msm_host->dsc), bytes_per_pclk); + hdisplay = DIV_ROUND_UP(msm_dsc_get_bytes_per_line(msm_host->dsc) * 8, bits_per_pclk); h_total += hdisplay; ha_end = ha_start + hdisplay; diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c index 8cb0db3a9880..01182442dfd6 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c @@ -51,8 +51,8 @@ #define DSI_PHY_7NM_QUIRK_V4_3 BIT(3) /* Hardware is V5.2 */ #define DSI_PHY_7NM_QUIRK_V5_2 BIT(4) -/* Hardware is V7.0 */ -#define DSI_PHY_7NM_QUIRK_V7_0 BIT(5) +/* Hardware is V7.2 */ +#define DSI_PHY_7NM_QUIRK_V7_2 BIT(5) struct dsi_pll_config { bool enable_ssc; @@ -143,7 +143,7 @@ static void dsi_pll_calc_dec_frac(struct dsi_pll_7nm *pll, struct dsi_pll_config if (pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_PRE_V4_1) { config->pll_clock_inverters = 0x28; - } else if ((pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { + } else if ((pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_2)) { if (pll_freq < 163000000ULL) config->pll_clock_inverters = 0xa0; else if (pll_freq < 175000000ULL) @@ -284,7 +284,7 @@ static void dsi_pll_config_hzindep_reg(struct dsi_pll_7nm *pll) } if ((pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2) || - (pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { + (pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_2)) { if (pll->vco_current_rate < 1557000000ULL) vco_config_1 = 0x08; else @@ -699,7 +699,7 @@ static int dsi_7nm_set_usecase(struct msm_dsi_phy *phy) case MSM_DSI_PHY_MASTER: pll_7nm->slave = pll_7nm_list[(pll_7nm->phy->id + 1) % DSI_MAX]; /* v7.0: Enable ATB_EN0 and alternate clock output to external phy */ - if (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0) + if (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_2) writel(0x07, base + REG_DSI_7nm_PHY_CMN_CTRL_5); break; case MSM_DSI_PHY_SLAVE: @@ -987,7 +987,7 @@ static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy, /* Request for REFGEN READY */ if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V4_3) || (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2) || - (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { + (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_2)) { writel(0x1, phy->base + REG_DSI_7nm_PHY_CMN_GLBL_DIGTOP_SPARE10); udelay(500); } @@ -1021,7 +1021,7 @@ static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy, lane_ctrl0 = 0x1f; } - if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { + if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_2)) { if (phy->cphy_mode) { /* TODO: different for second phy */ vreg_ctrl_0 = 0x57; @@ -1097,7 +1097,7 @@ static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy, /* program CMN_CTRL_4 for minor_ver 2 chipsets*/ if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2) || - (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0) || + (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_2) || (readl(base + REG_DSI_7nm_PHY_CMN_REVISION_ID0) & (0xf0)) == 0x20) writel(0x04, base + REG_DSI_7nm_PHY_CMN_CTRL_4); @@ -1213,7 +1213,7 @@ static void dsi_7nm_phy_disable(struct msm_dsi_phy *phy) /* Turn off REFGEN Vote */ if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V4_3) || (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2) || - (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { + (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_2)) { writel(0x0, base + REG_DSI_7nm_PHY_CMN_GLBL_DIGTOP_SPARE10); wmb(); /* Delay to ensure HW removes vote before PHY shut down */ @@ -1502,7 +1502,7 @@ const struct msm_dsi_phy_cfg dsi_phy_3nm_8750_cfgs = { #endif .io_start = { 0xae95000, 0xae97000 }, .num_dsi_phy = 2, - .quirks = DSI_PHY_7NM_QUIRK_V7_0, + .quirks = DSI_PHY_7NM_QUIRK_V7_2, }; const struct msm_dsi_phy_cfg dsi_phy_3nm_kaanapali_cfgs = { @@ -1525,5 +1525,5 @@ const struct msm_dsi_phy_cfg dsi_phy_3nm_kaanapali_cfgs = { #endif .io_start = { 0x9ac1000, 0x9ac4000 }, .num_dsi_phy = 2, - .quirks = DSI_PHY_7NM_QUIRK_V7_0, + .quirks = DSI_PHY_7NM_QUIRK_V7_2, }; diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 00d4530aea71..cc239492c7f0 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -1230,6 +1230,9 @@ nouveau_connector_aux_xfer(struct drm_dp_aux *obj, struct drm_dp_aux_msg *msg) u8 size = msg->size; int ret; + if (pm_runtime_suspended(nv_connector->base.dev->dev)) + return -EBUSY; + nv_encoder = find_encoder(&nv_connector->base, DCB_OUTPUT_DP); if (!nv_encoder) return -ENODEV; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c index e10192fe1d0f..0121d5639471 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c @@ -737,8 +737,8 @@ r535_gsp_acpi_caps(acpi_handle handle, CAPS_METHOD_DATA *caps) if (!obj) goto done; - if (WARN_ON(obj->type != ACPI_TYPE_BUFFER) || - WARN_ON(obj->buffer.length != 4)) + if (obj->type != ACPI_TYPE_BUFFER || + obj->buffer.length != 4) goto done; caps->status = 0; @@ -773,8 +773,8 @@ r535_gsp_acpi_jt(acpi_handle handle, JT_METHOD_DATA *jt) if (!obj) goto done; - if (WARN_ON(obj->type != ACPI_TYPE_BUFFER) || - WARN_ON(obj->buffer.length != 4)) + if (obj->type != ACPI_TYPE_BUFFER || + obj->buffer.length != 4) goto done; jt->status = 0; @@ -861,8 +861,8 @@ r535_gsp_acpi_dod(acpi_handle handle, DOD_METHOD_DATA *dod) _DOD = output.pointer; - if (WARN_ON(_DOD->type != ACPI_TYPE_PACKAGE) || - WARN_ON(_DOD->package.count > ARRAY_SIZE(dod->acpiIdList))) + if (_DOD->type != ACPI_TYPE_PACKAGE || + _DOD->package.count > ARRAY_SIZE(dod->acpiIdList)) return; for (int i = 0; i < _DOD->package.count; i++) { diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c index bd703a2904a1..a70f1db0764e 100644 --- a/drivers/gpu/drm/panthor/panthor_sched.c +++ b/drivers/gpu/drm/panthor/panthor_sched.c @@ -893,14 +893,15 @@ panthor_queue_get_syncwait_obj(struct panthor_group *group, struct panthor_queue out_sync: /* Make sure the CPU caches are invalidated before the seqno is read. - * drm_gem_shmem_sync() is a NOP if map_wc=true, so no need to check + * panthor_gem_sync() is a NOP if map_wc=true, so no need to check * it here. */ - panthor_gem_sync(&bo->base.base, queue->syncwait.offset, + panthor_gem_sync(&bo->base.base, + DRM_PANTHOR_BO_SYNC_CPU_CACHE_FLUSH_AND_INVALIDATE, + queue->syncwait.offset, queue->syncwait.sync64 ? sizeof(struct panthor_syncobj_64b) : - sizeof(struct panthor_syncobj_32b), - DRM_PANTHOR_BO_SYNC_CPU_CACHE_FLUSH_AND_INVALIDATE); + sizeof(struct panthor_syncobj_32b)); return queue->syncwait.kmap + queue->syncwait.offset; diff --git a/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c b/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c index f74a0aa85ba8..29f2b7d24fe5 100644 --- a/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c +++ b/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c @@ -1122,6 +1122,7 @@ static int rzg2l_mipi_dsi_host_attach(struct mipi_dsi_host *host, struct mipi_dsi_device *device) { struct rzg2l_mipi_dsi *dsi = host_to_rzg2l_mipi_dsi(host); + int bpp; int ret; if (device->lanes > dsi->num_data_lanes) { @@ -1131,7 +1132,8 @@ static int rzg2l_mipi_dsi_host_attach(struct mipi_dsi_host *host, return -EINVAL; } - switch (mipi_dsi_pixel_format_to_bpp(device->format)) { + bpp = mipi_dsi_pixel_format_to_bpp(device->format); + switch (bpp) { case 24: break; case 18: @@ -1162,6 +1164,18 @@ static int rzg2l_mipi_dsi_host_attach(struct mipi_dsi_host *host, drm_bridge_add(&dsi->bridge); + /* + * Report the required division ratio setting for the MIPI clock dividers. + * + * vclk * bpp = hsclk * 8 * num_lanes + * + * vclk * DSI_AB_divider = hsclk * 16 + * + * which simplifies to... + * DSI_AB_divider = bpp * 2 / num_lanes + */ + rzg2l_cpg_dsi_div_set_divider(bpp * 2 / dsi->lanes, PLL5_TARGET_DSI); + return 0; } diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index e6ee35406165..2d5cb21a05b6 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -361,6 +361,7 @@ static void drm_sched_run_free_queue(struct drm_gpu_scheduler *sched) /** * drm_sched_job_done - complete a job * @s_job: pointer to the job which is done + * @result: 0 on success, -ERRNO on error * * Finish the job's fence and resubmit the work items. */ diff --git a/drivers/gpu/drm/sitronix/st7586.c b/drivers/gpu/drm/sitronix/st7586.c index b57ebf37a664..16b6b4e368af 100644 --- a/drivers/gpu/drm/sitronix/st7586.c +++ b/drivers/gpu/drm/sitronix/st7586.c @@ -347,6 +347,12 @@ static int st7586_probe(struct spi_device *spi) if (ret) return ret; + /* + * Override value set by mipi_dbi_spi_init(). This driver is a bit + * non-standard, so best to set it explicitly here. + */ + dbi->write_memory_bpw = 8; + /* Cannot read from this controller via SPI */ dbi->read_commands = NULL; @@ -356,15 +362,6 @@ static int st7586_probe(struct spi_device *spi) if (ret) return ret; - /* - * we are using 8-bit data, so we are not actually swapping anything, - * but setting mipi->swap_bytes makes mipi_dbi_typec3_command() do the - * right thing and not use 16-bit transfers (which results in swapped - * bytes on little-endian systems and causes out of order data to be - * sent to the display). - */ - dbi->swap_bytes = true; - drm_mode_config_reset(drm); ret = drm_dev_register(drm, 0); diff --git a/drivers/gpu/drm/solomon/ssd130x.c b/drivers/gpu/drm/solomon/ssd130x.c index 6ecf9e2ff61b..c77455b1834d 100644 --- a/drivers/gpu/drm/solomon/ssd130x.c +++ b/drivers/gpu/drm/solomon/ssd130x.c @@ -737,6 +737,7 @@ static int ssd130x_update_rect(struct ssd130x_device *ssd130x, unsigned int height = drm_rect_height(rect); unsigned int line_length = DIV_ROUND_UP(width, 8); unsigned int page_height = SSD130X_PAGE_HEIGHT; + u8 page_start = ssd130x->page_offset + y / page_height; unsigned int pages = DIV_ROUND_UP(height, page_height); struct drm_device *drm = &ssd130x->drm; u32 array_idx = 0; @@ -774,14 +775,11 @@ static int ssd130x_update_rect(struct ssd130x_device *ssd130x, */ if (!ssd130x->page_address_mode) { - u8 page_start; - /* Set address range for horizontal addressing mode */ ret = ssd130x_set_col_range(ssd130x, ssd130x->col_offset + x, width); if (ret < 0) return ret; - page_start = ssd130x->page_offset + y / page_height; ret = ssd130x_set_page_range(ssd130x, page_start, pages); if (ret < 0) return ret; @@ -813,7 +811,7 @@ static int ssd130x_update_rect(struct ssd130x_device *ssd130x, */ if (ssd130x->page_address_mode) { ret = ssd130x_set_page_pos(ssd130x, - ssd130x->page_offset + i, + page_start + i, ssd130x->col_offset + x); if (ret < 0) return ret; diff --git a/drivers/gpu/drm/tiny/sharp-memory.c b/drivers/gpu/drm/tiny/sharp-memory.c index 64272cd0f6e2..cbf69460ebf3 100644 --- a/drivers/gpu/drm/tiny/sharp-memory.c +++ b/drivers/gpu/drm/tiny/sharp-memory.c @@ -541,8 +541,8 @@ static int sharp_memory_probe(struct spi_device *spi) smd = devm_drm_dev_alloc(dev, &sharp_memory_drm_driver, struct sharp_memory_device, drm); - if (!smd) - return -ENOMEM; + if (IS_ERR(smd)) + return PTR_ERR(smd); spi_set_drvdata(spi, smd); diff --git a/drivers/gpu/drm/ttm/tests/ttm_bo_test.c b/drivers/gpu/drm/ttm/tests/ttm_bo_test.c index d468f8322072..f3103307b5df 100644 --- a/drivers/gpu/drm/ttm/tests/ttm_bo_test.c +++ b/drivers/gpu/drm/ttm/tests/ttm_bo_test.c @@ -222,13 +222,13 @@ static void ttm_bo_reserve_interrupted(struct kunit *test) KUNIT_FAIL(test, "Couldn't create ttm bo reserve task\n"); /* Take a lock so the threaded reserve has to wait */ - mutex_lock(&bo->base.resv->lock.base); + dma_resv_lock(bo->base.resv, NULL); wake_up_process(task); msleep(20); err = kthread_stop(task); - mutex_unlock(&bo->base.resv->lock.base); + dma_resv_unlock(bo->base.resv); KUNIT_ASSERT_EQ(test, err, -ERESTARTSYS); } diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index acb9197db879..0765d69423d2 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -1107,8 +1107,7 @@ struct ttm_bo_swapout_walk { static s64 ttm_bo_swapout_cb(struct ttm_lru_walk *walk, struct ttm_buffer_object *bo) { - struct ttm_resource *res = bo->resource; - struct ttm_place place = { .mem_type = res->mem_type }; + struct ttm_place place = { .mem_type = bo->resource->mem_type }; struct ttm_bo_swapout_walk *swapout_walk = container_of(walk, typeof(*swapout_walk), walk); struct ttm_operation_ctx *ctx = walk->arg.ctx; @@ -1148,7 +1147,7 @@ ttm_bo_swapout_cb(struct ttm_lru_walk *walk, struct ttm_buffer_object *bo) /* * Move to system cached */ - if (res->mem_type != TTM_PL_SYSTEM) { + if (bo->resource->mem_type != TTM_PL_SYSTEM) { struct ttm_resource *evict_mem; struct ttm_place hop; @@ -1180,15 +1179,15 @@ ttm_bo_swapout_cb(struct ttm_lru_walk *walk, struct ttm_buffer_object *bo) if (ttm_tt_is_populated(tt)) { spin_lock(&bdev->lru_lock); - ttm_resource_del_bulk_move(res, bo); + ttm_resource_del_bulk_move(bo->resource, bo); spin_unlock(&bdev->lru_lock); ret = ttm_tt_swapout(bdev, tt, swapout_walk->gfp_flags); spin_lock(&bdev->lru_lock); if (ret) - ttm_resource_add_bulk_move(res, bo); - ttm_resource_move_to_lru_tail(res); + ttm_resource_add_bulk_move(bo->resource, bo); + ttm_resource_move_to_lru_tail(bo->resource); spin_unlock(&bdev->lru_lock); } diff --git a/drivers/gpu/drm/ttm/ttm_pool_internal.h b/drivers/gpu/drm/ttm/ttm_pool_internal.h index 82c4b7e56a99..24c179fd69d1 100644 --- a/drivers/gpu/drm/ttm/ttm_pool_internal.h +++ b/drivers/gpu/drm/ttm/ttm_pool_internal.h @@ -17,7 +17,7 @@ static inline bool ttm_pool_uses_dma32(struct ttm_pool *pool) return pool->alloc_flags & TTM_ALLOCATION_POOL_USE_DMA32; } -static inline bool ttm_pool_beneficial_order(struct ttm_pool *pool) +static inline unsigned int ttm_pool_beneficial_order(struct ttm_pool *pool) { return pool->alloc_flags & 0xff; } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c index 2170b45c30e9..f45d93e8c1c8 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c @@ -105,6 +105,7 @@ struct vmw_cmdbuf_context { * @handle: DMA address handle for the command buffer space if @using_mob is * false. Immutable. * @size: The size of the command buffer space. Immutable. + * @id: Monotonically increasing ID of the last cmdbuf submitted. * @num_contexts: Number of contexts actually enabled. */ struct vmw_cmdbuf_man { @@ -132,6 +133,7 @@ struct vmw_cmdbuf_man { bool has_pool; dma_addr_t handle; size_t size; + u64 id; u32 num_contexts; }; @@ -303,6 +305,8 @@ static int vmw_cmdbuf_header_submit(struct vmw_cmdbuf_header *header) struct vmw_cmdbuf_man *man = header->man; u32 val; + header->cb_header->id = man->id++; + val = upper_32_bits(header->handle); vmw_write(man->dev_priv, SVGA_REG_COMMAND_HIGH, val); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 3057f8baa7d2..e1f18020170a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -1143,7 +1143,7 @@ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv, ret = vmw_user_bo_lookup(sw_context->filp, handle, &vmw_bo); if (ret != 0) { drm_dbg(&dev_priv->drm, "Could not find or use MOB buffer.\n"); - return PTR_ERR(vmw_bo); + return ret; } vmw_bo_placement_set(vmw_bo, VMW_BO_DOMAIN_MOB, VMW_BO_DOMAIN_MOB); ret = vmw_validation_add_bo(sw_context->ctx, vmw_bo); @@ -1199,7 +1199,7 @@ static int vmw_translate_guest_ptr(struct vmw_private *dev_priv, ret = vmw_user_bo_lookup(sw_context->filp, handle, &vmw_bo); if (ret != 0) { drm_dbg(&dev_priv->drm, "Could not find or use GMR region.\n"); - return PTR_ERR(vmw_bo); + return ret; } vmw_bo_placement_set(vmw_bo, VMW_BO_DOMAIN_GMR | VMW_BO_DOMAIN_VRAM, VMW_BO_DOMAIN_GMR | VMW_BO_DOMAIN_VRAM); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c index fd4e76486f2d..45561bc1c9ef 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c @@ -260,6 +260,13 @@ out_no_dirty: return ret; } +static void vmw_bo_dirty_free(struct kref *kref) +{ + struct vmw_bo_dirty *dirty = container_of(kref, struct vmw_bo_dirty, ref_count); + + kvfree(dirty); +} + /** * vmw_bo_dirty_release - Release a dirty-tracking user from a buffer object * @vbo: The buffer object @@ -274,7 +281,7 @@ void vmw_bo_dirty_release(struct vmw_bo *vbo) { struct vmw_bo_dirty *dirty = vbo->dirty; - if (dirty && kref_put(&dirty->ref_count, (void *)kvfree)) + if (dirty && kref_put(&dirty->ref_count, vmw_bo_dirty_free)) vbo->dirty = NULL; } diff --git a/drivers/gpu/drm/xe/regs/xe_engine_regs.h b/drivers/gpu/drm/xe/regs/xe_engine_regs.h index 68172b0248a6..dc5a4fafa70c 100644 --- a/drivers/gpu/drm/xe/regs/xe_engine_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_engine_regs.h @@ -96,6 +96,12 @@ #define ENABLE_SEMAPHORE_POLL_BIT REG_BIT(13) #define RING_CMD_CCTL(base) XE_REG((base) + 0xc4, XE_REG_OPTION_MASKED) + +#define CS_MMIO_GROUP_INSTANCE_SELECT(base) XE_REG((base) + 0xcc) +#define SELECTIVE_READ_ADDRESSING REG_BIT(30) +#define SELECTIVE_READ_GROUP REG_GENMASK(29, 23) +#define SELECTIVE_READ_INSTANCE REG_GENMASK(22, 16) + /* * CMD_CCTL read/write fields take a MOCS value and _not_ a table index. * The lsb of each can be considered a separate enabling bit for encryption. diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c index c59b1414df22..7fd07d1280bb 100644 --- a/drivers/gpu/drm/xe/xe_configfs.c +++ b/drivers/gpu/drm/xe/xe_configfs.c @@ -830,6 +830,7 @@ static void xe_config_device_release(struct config_item *item) mutex_destroy(&dev->lock); + kfree(dev->config.ctx_restore_mid_bb[0].cs); kfree(dev->config.ctx_restore_post_bb[0].cs); kfree(dev); } diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c index 0ddae7fcfc97..8ecdf949f9e4 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.c +++ b/drivers/gpu/drm/xe/xe_exec_queue.c @@ -266,6 +266,16 @@ static struct xe_exec_queue *__xe_exec_queue_alloc(struct xe_device *xe, return q; } +static void __xe_exec_queue_fini(struct xe_exec_queue *q) +{ + int i; + + q->ops->fini(q); + + for (i = 0; i < q->width; ++i) + xe_lrc_put(q->lrc[i]); +} + static int __xe_exec_queue_init(struct xe_exec_queue *q, u32 exec_queue_flags) { int i, err; @@ -320,21 +330,10 @@ static int __xe_exec_queue_init(struct xe_exec_queue *q, u32 exec_queue_flags) return 0; err_lrc: - for (i = i - 1; i >= 0; --i) - xe_lrc_put(q->lrc[i]); + __xe_exec_queue_fini(q); return err; } -static void __xe_exec_queue_fini(struct xe_exec_queue *q) -{ - int i; - - q->ops->fini(q); - - for (i = 0; i < q->width; ++i) - xe_lrc_put(q->lrc[i]); -} - struct xe_exec_queue *xe_exec_queue_create(struct xe_device *xe, struct xe_vm *vm, u32 logical_mask, u16 width, struct xe_hw_engine *hwe, u32 flags, diff --git a/drivers/gpu/drm/xe/xe_gsc_proxy.c b/drivers/gpu/drm/xe/xe_gsc_proxy.c index 42438b21f235..707db650a2ae 100644 --- a/drivers/gpu/drm/xe/xe_gsc_proxy.c +++ b/drivers/gpu/drm/xe/xe_gsc_proxy.c @@ -435,15 +435,11 @@ static int proxy_channel_alloc(struct xe_gsc *gsc) return 0; } -static void xe_gsc_proxy_remove(void *arg) +static void xe_gsc_proxy_stop(struct xe_gsc *gsc) { - struct xe_gsc *gsc = arg; struct xe_gt *gt = gsc_to_gt(gsc); struct xe_device *xe = gt_to_xe(gt); - if (!gsc->proxy.component_added) - return; - /* disable HECI2 IRQs */ scoped_guard(xe_pm_runtime, xe) { CLASS(xe_force_wake, fw_ref)(gt_to_fw(gt), XE_FW_GSC); @@ -455,6 +451,30 @@ static void xe_gsc_proxy_remove(void *arg) } xe_gsc_wait_for_worker_completion(gsc); + gsc->proxy.started = false; +} + +static void xe_gsc_proxy_remove(void *arg) +{ + struct xe_gsc *gsc = arg; + struct xe_gt *gt = gsc_to_gt(gsc); + struct xe_device *xe = gt_to_xe(gt); + + if (!gsc->proxy.component_added) + return; + + /* + * GSC proxy start is an async process that can be ongoing during + * Xe module load/unload. Using devm managed action to register + * xe_gsc_proxy_stop could cause issues if Xe module unload has + * already started when the action is registered, potentially leading + * to the cleanup being called at the wrong time. Therefore, instead + * of registering a separate devm action to undo what is done in + * proxy start, we call it from here, but only if the start has + * completed successfully (tracked with the 'started' flag). + */ + if (gsc->proxy.started) + xe_gsc_proxy_stop(gsc); component_del(xe->drm.dev, &xe_gsc_proxy_component_ops); gsc->proxy.component_added = false; @@ -510,6 +530,7 @@ int xe_gsc_proxy_init(struct xe_gsc *gsc) */ int xe_gsc_proxy_start(struct xe_gsc *gsc) { + struct xe_gt *gt = gsc_to_gt(gsc); int err; /* enable the proxy interrupt in the GSC shim layer */ @@ -521,12 +542,18 @@ int xe_gsc_proxy_start(struct xe_gsc *gsc) */ err = xe_gsc_proxy_request_handler(gsc); if (err) - return err; + goto err_irq_disable; if (!xe_gsc_proxy_init_done(gsc)) { - xe_gt_err(gsc_to_gt(gsc), "GSC FW reports proxy init not completed\n"); - return -EIO; + xe_gt_err(gt, "GSC FW reports proxy init not completed\n"); + err = -EIO; + goto err_irq_disable; } + gsc->proxy.started = true; return 0; + +err_irq_disable: + gsc_proxy_irq_toggle(gsc, false); + return err; } diff --git a/drivers/gpu/drm/xe/xe_gsc_types.h b/drivers/gpu/drm/xe/xe_gsc_types.h index 97c056656df0..5aaa2a75861f 100644 --- a/drivers/gpu/drm/xe/xe_gsc_types.h +++ b/drivers/gpu/drm/xe/xe_gsc_types.h @@ -58,6 +58,8 @@ struct xe_gsc { struct mutex mutex; /** @proxy.component_added: whether the component has been added */ bool component_added; + /** @proxy.started: whether the proxy has been started */ + bool started; /** @proxy.bo: object to store message to and from the GSC */ struct xe_bo *bo; /** @proxy.to_gsc: map of the memory used to send messages to the GSC */ diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c index 9d090d0f2438..df6d04704823 100644 --- a/drivers/gpu/drm/xe/xe_gt.c +++ b/drivers/gpu/drm/xe/xe_gt.c @@ -210,11 +210,15 @@ static int emit_nop_job(struct xe_gt *gt, struct xe_exec_queue *q) return ret; } +/* Dwords required to emit a RMW of a register */ +#define EMIT_RMW_DW 20 + static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q) { - struct xe_reg_sr *sr = &q->hwe->reg_lrc; + struct xe_hw_engine *hwe = q->hwe; + struct xe_reg_sr *sr = &hwe->reg_lrc; struct xe_reg_sr_entry *entry; - int count_rmw = 0, count = 0, ret; + int count_rmw = 0, count_rmw_mcr = 0, count = 0, ret; unsigned long idx; struct xe_bb *bb; size_t bb_len = 0; @@ -224,6 +228,8 @@ static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q) xa_for_each(&sr->xa, idx, entry) { if (entry->reg.masked || entry->clr_bits == ~0) ++count; + else if (entry->reg.mcr) + ++count_rmw_mcr; else ++count_rmw; } @@ -231,17 +237,35 @@ static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q) if (count) bb_len += count * 2 + 1; - if (count_rmw) - bb_len += count_rmw * 20 + 7; + /* + * RMW of MCR registers is the same as a normal RMW, except an + * additional LRI (3 dwords) is required per register to steer the read + * to a nom-terminated instance. + * + * We could probably shorten the batch slightly by eliding the + * steering for consecutive MCR registers that have the same + * group/instance target, but it's not worth the extra complexity to do + * so. + */ + bb_len += count_rmw * EMIT_RMW_DW; + bb_len += count_rmw_mcr * (EMIT_RMW_DW + 3); + + /* + * After doing all RMW, we need 7 trailing dwords to clean up, + * plus an additional 3 dwords to reset steering if any of the + * registers were MCR. + */ + if (count_rmw || count_rmw_mcr) + bb_len += 7 + (count_rmw_mcr ? 3 : 0); - if (q->hwe->class == XE_ENGINE_CLASS_RENDER) + if (hwe->class == XE_ENGINE_CLASS_RENDER) /* * Big enough to emit all of the context's 3DSTATE via * xe_lrc_emit_hwe_state_instructions() */ - bb_len += xe_gt_lrc_size(gt, q->hwe->class) / sizeof(u32); + bb_len += xe_gt_lrc_size(gt, hwe->class) / sizeof(u32); - xe_gt_dbg(gt, "LRC %s WA job: %zu dwords\n", q->hwe->name, bb_len); + xe_gt_dbg(gt, "LRC %s WA job: %zu dwords\n", hwe->name, bb_len); bb = xe_bb_new(gt, bb_len, false); if (IS_ERR(bb)) @@ -276,13 +300,23 @@ static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q) } } - if (count_rmw) { - /* Emit MI_MATH for each RMW reg: 20dw per reg + 7 trailing dw */ - + if (count_rmw || count_rmw_mcr) { xa_for_each(&sr->xa, idx, entry) { if (entry->reg.masked || entry->clr_bits == ~0) continue; + if (entry->reg.mcr) { + struct xe_reg_mcr reg = { .__reg.raw = entry->reg.raw }; + u8 group, instance; + + xe_gt_mcr_get_nonterminated_steering(gt, reg, &group, &instance); + *cs++ = MI_LOAD_REGISTER_IMM | MI_LRI_NUM_REGS(1); + *cs++ = CS_MMIO_GROUP_INSTANCE_SELECT(hwe->mmio_base).addr; + *cs++ = SELECTIVE_READ_ADDRESSING | + REG_FIELD_PREP(SELECTIVE_READ_GROUP, group) | + REG_FIELD_PREP(SELECTIVE_READ_INSTANCE, instance); + } + *cs++ = MI_LOAD_REGISTER_REG | MI_LRR_DST_CS_MMIO; *cs++ = entry->reg.addr; *cs++ = CS_GPR_REG(0, 0).addr; @@ -308,8 +342,9 @@ static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q) *cs++ = CS_GPR_REG(0, 0).addr; *cs++ = entry->reg.addr; - xe_gt_dbg(gt, "REG[%#x] = ~%#x|%#x\n", - entry->reg.addr, entry->clr_bits, entry->set_bits); + xe_gt_dbg(gt, "REG[%#x] = ~%#x|%#x%s\n", + entry->reg.addr, entry->clr_bits, entry->set_bits, + entry->reg.mcr ? " (MCR)" : ""); } /* reset used GPR */ @@ -321,6 +356,13 @@ static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q) *cs++ = 0; *cs++ = CS_GPR_REG(0, 2).addr; *cs++ = 0; + + /* reset steering */ + if (count_rmw_mcr) { + *cs++ = MI_LOAD_REGISTER_IMM | MI_LRI_NUM_REGS(1); + *cs++ = CS_MMIO_GROUP_INSTANCE_SELECT(q->hwe->mmio_base).addr; + *cs++ = 0; + } } cs = xe_lrc_emit_hwe_state_instructions(q, cs); diff --git a/drivers/gpu/drm/xe/xe_lrc.h b/drivers/gpu/drm/xe/xe_lrc.h index c307a3fd9ea2..c1c615447c85 100644 --- a/drivers/gpu/drm/xe/xe_lrc.h +++ b/drivers/gpu/drm/xe/xe_lrc.h @@ -75,7 +75,8 @@ static inline struct xe_lrc *xe_lrc_get(struct xe_lrc *lrc) */ static inline void xe_lrc_put(struct xe_lrc *lrc) { - kref_put(&lrc->refcount, xe_lrc_destroy); + if (lrc) + kref_put(&lrc->refcount, xe_lrc_destroy); } /** diff --git a/drivers/gpu/drm/xe/xe_reg_sr.c b/drivers/gpu/drm/xe/xe_reg_sr.c index 2e5c78940b41..a07be161cfa2 100644 --- a/drivers/gpu/drm/xe/xe_reg_sr.c +++ b/drivers/gpu/drm/xe/xe_reg_sr.c @@ -98,10 +98,12 @@ int xe_reg_sr_add(struct xe_reg_sr *sr, *pentry = *e; ret = xa_err(xa_store(&sr->xa, idx, pentry, GFP_KERNEL)); if (ret) - goto fail; + goto fail_free; return 0; +fail_free: + kfree(pentry); fail: xe_gt_err(gt, "discarding save-restore reg %04lx (clear: %08x, set: %08x, masked: %s, mcr: %s): ret=%d\n", diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c index 248620b0901d..53d420d72164 100644 --- a/drivers/gpu/drm/xe/xe_ring_ops.c +++ b/drivers/gpu/drm/xe/xe_ring_ops.c @@ -280,6 +280,9 @@ static void __emit_job_gen12_simple(struct xe_sched_job *job, struct xe_lrc *lrc i = emit_bb_start(batch_addr, ppgtt_flag, dw, i); + /* Don't preempt fence signaling */ + dw[i++] = MI_ARB_ON_OFF | MI_ARB_DISABLE; + if (job->user_fence.used) { i = emit_flush_dw(dw, i); i = emit_store_imm_ppgtt_posted(job->user_fence.addr, @@ -345,6 +348,9 @@ static void __emit_job_gen12_video(struct xe_sched_job *job, struct xe_lrc *lrc, i = emit_bb_start(batch_addr, ppgtt_flag, dw, i); + /* Don't preempt fence signaling */ + dw[i++] = MI_ARB_ON_OFF | MI_ARB_DISABLE; + if (job->user_fence.used) { i = emit_flush_dw(dw, i); i = emit_store_imm_ppgtt_posted(job->user_fence.addr, @@ -397,6 +403,9 @@ static void __emit_job_gen12_render_compute(struct xe_sched_job *job, i = emit_bb_start(batch_addr, ppgtt_flag, dw, i); + /* Don't preempt fence signaling */ + dw[i++] = MI_ARB_ON_OFF | MI_ARB_DISABLE; + i = emit_render_cache_flush(job, dw, i); if (job->user_fence.used) diff --git a/drivers/gpu/drm/xe/xe_sync.c b/drivers/gpu/drm/xe/xe_sync.c index eb136390dafd..24d6d9af20d6 100644 --- a/drivers/gpu/drm/xe/xe_sync.c +++ b/drivers/gpu/drm/xe/xe_sync.c @@ -146,8 +146,10 @@ int xe_sync_entry_parse(struct xe_device *xe, struct xe_file *xef, if (!signal) { sync->fence = drm_syncobj_fence_get(sync->syncobj); - if (XE_IOCTL_DBG(xe, !sync->fence)) - return -EINVAL; + if (XE_IOCTL_DBG(xe, !sync->fence)) { + err = -EINVAL; + goto free_sync; + } } break; @@ -167,17 +169,21 @@ int xe_sync_entry_parse(struct xe_device *xe, struct xe_file *xef, if (signal) { sync->chain_fence = dma_fence_chain_alloc(); - if (!sync->chain_fence) - return -ENOMEM; + if (!sync->chain_fence) { + err = -ENOMEM; + goto free_sync; + } } else { sync->fence = drm_syncobj_fence_get(sync->syncobj); - if (XE_IOCTL_DBG(xe, !sync->fence)) - return -EINVAL; + if (XE_IOCTL_DBG(xe, !sync->fence)) { + err = -EINVAL; + goto free_sync; + } err = dma_fence_chain_find_seqno(&sync->fence, sync_in.timeline_value); if (err) - return err; + goto free_sync; } break; @@ -200,8 +206,10 @@ int xe_sync_entry_parse(struct xe_device *xe, struct xe_file *xef, if (XE_IOCTL_DBG(xe, IS_ERR(sync->ufence))) return PTR_ERR(sync->ufence); sync->ufence_chain_fence = dma_fence_chain_alloc(); - if (!sync->ufence_chain_fence) - return -ENOMEM; + if (!sync->ufence_chain_fence) { + err = -ENOMEM; + goto free_sync; + } sync->ufence_syncobj = ufence_syncobj; } @@ -216,6 +224,10 @@ int xe_sync_entry_parse(struct xe_device *xe, struct xe_file *xef, sync->timeline_value = sync_in.timeline_value; return 0; + +free_sync: + xe_sync_entry_cleanup(sync); + return err; } ALLOW_ERROR_INJECTION(xe_sync_entry_parse, ERRNO); diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c index 95bf53cc29e3..bc39a9a9790c 100644 --- a/drivers/gpu/drm/xe/xe_vm_madvise.c +++ b/drivers/gpu/drm/xe/xe_vm_madvise.c @@ -453,7 +453,7 @@ int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *fil madvise_range.num_vmas, args->atomic.val)) { err = -EINVAL; - goto madv_fini; + goto free_vmas; } } @@ -490,6 +490,7 @@ int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *fil err_fini: if (madvise_range.has_bo_vmas) drm_exec_fini(&exec); +free_vmas: kfree(madvise_range.vmas); madvise_range.vmas = NULL; madv_fini: diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c index c7b1bd79ab17..462c2fa712e0 100644 --- a/drivers/gpu/drm/xe/xe_wa.c +++ b/drivers/gpu/drm/xe/xe_wa.c @@ -241,12 +241,13 @@ static const struct xe_rtp_entry_sr gt_was[] = { { XE_RTP_NAME("16025250150"), XE_RTP_RULES(GRAPHICS_VERSION(2001)), - XE_RTP_ACTIONS(SET(LSN_VC_REG2, - LSN_LNI_WGT(1) | - LSN_LNE_WGT(1) | - LSN_DIM_X_WGT(1) | - LSN_DIM_Y_WGT(1) | - LSN_DIM_Z_WGT(1))) + XE_RTP_ACTIONS(FIELD_SET(LSN_VC_REG2, + LSN_LNI_WGT_MASK | LSN_LNE_WGT_MASK | + LSN_DIM_X_WGT_MASK | LSN_DIM_Y_WGT_MASK | + LSN_DIM_Z_WGT_MASK, + LSN_LNI_WGT(1) | LSN_LNE_WGT(1) | + LSN_DIM_X_WGT(1) | LSN_DIM_Y_WGT(1) | + LSN_DIM_Z_WGT(1))) }, /* Xe2_HPM */ diff --git a/drivers/gpu/nova-core/gsp.rs b/drivers/gpu/nova-core/gsp.rs index 174feaca0a6b..c69adaa92bbe 100644 --- a/drivers/gpu/nova-core/gsp.rs +++ b/drivers/gpu/nova-core/gsp.rs @@ -47,16 +47,12 @@ struct PteArray<const NUM_ENTRIES: usize>([u64; NUM_ENTRIES]); unsafe impl<const NUM_ENTRIES: usize> AsBytes for PteArray<NUM_ENTRIES> {} impl<const NUM_PAGES: usize> PteArray<NUM_PAGES> { - /// Creates a new page table array mapping `NUM_PAGES` GSP pages starting at address `start`. - fn new(start: DmaAddress) -> Result<Self> { - let mut ptes = [0u64; NUM_PAGES]; - for (i, pte) in ptes.iter_mut().enumerate() { - *pte = start - .checked_add(num::usize_as_u64(i) << GSP_PAGE_SHIFT) - .ok_or(EOVERFLOW)?; - } - - Ok(Self(ptes)) + /// Returns the page table entry for `index`, for a mapping starting at `start`. + // TODO: Replace with `IoView` projection once available. + fn entry(start: DmaAddress, index: usize) -> Result<u64> { + start + .checked_add(num::usize_as_u64(index) << GSP_PAGE_SHIFT) + .ok_or(EOVERFLOW) } } @@ -86,16 +82,22 @@ impl LogBuffer { NUM_PAGES * GSP_PAGE_SIZE, GFP_KERNEL | __GFP_ZERO, )?); - let ptes = PteArray::<NUM_PAGES>::new(obj.0.dma_handle())?; + + let start_addr = obj.0.dma_handle(); // SAFETY: `obj` has just been created and we are its sole user. - unsafe { - // Copy the self-mapping PTE at the expected location. + let pte_region = unsafe { obj.0 - .as_slice_mut(size_of::<u64>(), size_of_val(&ptes))? - .copy_from_slice(ptes.as_bytes()) + .as_slice_mut(size_of::<u64>(), NUM_PAGES * size_of::<u64>())? }; + // Write values one by one to avoid an on-stack instance of `PteArray`. + for (i, chunk) in pte_region.chunks_exact_mut(size_of::<u64>()).enumerate() { + let pte_value = PteArray::<0>::entry(start_addr, i)?; + + chunk.copy_from_slice(&pte_value.to_ne_bytes()); + } + Ok(obj) } } @@ -143,14 +145,14 @@ impl Gsp { // _kgspInitLibosLoggingStructures (allocates memory for buffers) // kgspSetupLibosInitArgs_IMPL (creates pLibosInitArgs[] array) dma_write!( - libos[0] = LibosMemoryRegionInitArgument::new("LOGINIT", &loginit.0) - )?; + libos, [0]?, LibosMemoryRegionInitArgument::new("LOGINIT", &loginit.0) + ); dma_write!( - libos[1] = LibosMemoryRegionInitArgument::new("LOGINTR", &logintr.0) - )?; - dma_write!(libos[2] = LibosMemoryRegionInitArgument::new("LOGRM", &logrm.0))?; - dma_write!(rmargs[0].inner = fw::GspArgumentsCached::new(cmdq))?; - dma_write!(libos[3] = LibosMemoryRegionInitArgument::new("RMARGS", rmargs))?; + libos, [1]?, LibosMemoryRegionInitArgument::new("LOGINTR", &logintr.0) + ); + dma_write!(libos, [2]?, LibosMemoryRegionInitArgument::new("LOGRM", &logrm.0)); + dma_write!(rmargs, [0]?.inner, fw::GspArgumentsCached::new(cmdq)); + dma_write!(libos, [3]?, LibosMemoryRegionInitArgument::new("RMARGS", rmargs)); }, })) }) diff --git a/drivers/gpu/nova-core/gsp/boot.rs b/drivers/gpu/nova-core/gsp/boot.rs index be427fe26a58..94833f7996e8 100644 --- a/drivers/gpu/nova-core/gsp/boot.rs +++ b/drivers/gpu/nova-core/gsp/boot.rs @@ -157,7 +157,7 @@ impl super::Gsp { let wpr_meta = CoherentAllocation::<GspFwWprMeta>::alloc_coherent(dev, 1, GFP_KERNEL | __GFP_ZERO)?; - dma_write!(wpr_meta[0] = GspFwWprMeta::new(&gsp_fw, &fb_layout))?; + dma_write!(wpr_meta, [0]?, GspFwWprMeta::new(&gsp_fw, &fb_layout)); self.cmdq .send_command(bar, commands::SetSystemInfo::new(pdev))?; diff --git a/drivers/gpu/nova-core/gsp/cmdq.rs b/drivers/gpu/nova-core/gsp/cmdq.rs index 46819a82a51a..03a4f3599849 100644 --- a/drivers/gpu/nova-core/gsp/cmdq.rs +++ b/drivers/gpu/nova-core/gsp/cmdq.rs @@ -2,11 +2,7 @@ use core::{ cmp, - mem, - sync::atomic::{ - fence, - Ordering, // - }, // + mem, // }; use kernel::{ @@ -146,30 +142,36 @@ static_assert!(align_of::<MsgqData>() == GSP_PAGE_SIZE); #[repr(C)] // There is no struct defined for this in the open-gpu-kernel-source headers. // Instead it is defined by code in `GspMsgQueuesInit()`. -struct Msgq { +// TODO: Revert to private once `IoView` projections replace the `gsp_mem` module. +pub(super) struct Msgq { /// Header for sending messages, including the write pointer. - tx: MsgqTxHeader, + pub(super) tx: MsgqTxHeader, /// Header for receiving messages, including the read pointer. - rx: MsgqRxHeader, + pub(super) rx: MsgqRxHeader, /// The message queue proper. msgq: MsgqData, } /// Structure shared between the driver and the GSP and containing the command and message queues. #[repr(C)] -struct GspMem { +// TODO: Revert to private once `IoView` projections replace the `gsp_mem` module. +pub(super) struct GspMem { /// Self-mapping page table entries. - ptes: PteArray<{ GSP_PAGE_SIZE / size_of::<u64>() }>, + ptes: PteArray<{ Self::PTE_ARRAY_SIZE }>, /// CPU queue: the driver writes commands here, and the GSP reads them. It also contains the /// write and read pointers that the CPU updates. /// /// This member is read-only for the GSP. - cpuq: Msgq, + pub(super) cpuq: Msgq, /// GSP queue: the GSP writes messages here, and the driver reads them. It also contains the /// write and read pointers that the GSP updates. /// /// This member is read-only for the driver. - gspq: Msgq, + pub(super) gspq: Msgq, +} + +impl GspMem { + const PTE_ARRAY_SIZE: usize = GSP_PAGE_SIZE / size_of::<u64>(); } // SAFETY: These structs don't meet the no-padding requirements of AsBytes but @@ -201,9 +203,19 @@ impl DmaGspMem { let gsp_mem = CoherentAllocation::<GspMem>::alloc_coherent(dev, 1, GFP_KERNEL | __GFP_ZERO)?; - dma_write!(gsp_mem[0].ptes = PteArray::new(gsp_mem.dma_handle())?)?; - dma_write!(gsp_mem[0].cpuq.tx = MsgqTxHeader::new(MSGQ_SIZE, RX_HDR_OFF, MSGQ_NUM_PAGES))?; - dma_write!(gsp_mem[0].cpuq.rx = MsgqRxHeader::new())?; + + let start = gsp_mem.dma_handle(); + // Write values one by one to avoid an on-stack instance of `PteArray`. + for i in 0..GspMem::PTE_ARRAY_SIZE { + dma_write!(gsp_mem, [0]?.ptes.0[i], PteArray::<0>::entry(start, i)?); + } + + dma_write!( + gsp_mem, + [0]?.cpuq.tx, + MsgqTxHeader::new(MSGQ_SIZE, RX_HDR_OFF, MSGQ_NUM_PAGES) + ); + dma_write!(gsp_mem, [0]?.cpuq.rx, MsgqRxHeader::new()); Ok(Self(gsp_mem)) } @@ -317,12 +329,7 @@ impl DmaGspMem { // // - The returned value is between `0` and `MSGQ_NUM_PAGES`. fn gsp_write_ptr(&self) -> u32 { - let gsp_mem = self.0.start_ptr(); - - // SAFETY: - // - The 'CoherentAllocation' contains at least one object. - // - By the invariants of `CoherentAllocation` the pointer is valid. - (unsafe { (*gsp_mem).gspq.tx.write_ptr() } % MSGQ_NUM_PAGES) + super::fw::gsp_mem::gsp_write_ptr(&self.0) } // Returns the index of the memory page the GSP will read the next command from. @@ -331,12 +338,7 @@ impl DmaGspMem { // // - The returned value is between `0` and `MSGQ_NUM_PAGES`. fn gsp_read_ptr(&self) -> u32 { - let gsp_mem = self.0.start_ptr(); - - // SAFETY: - // - The 'CoherentAllocation' contains at least one object. - // - By the invariants of `CoherentAllocation` the pointer is valid. - (unsafe { (*gsp_mem).gspq.rx.read_ptr() } % MSGQ_NUM_PAGES) + super::fw::gsp_mem::gsp_read_ptr(&self.0) } // Returns the index of the memory page the CPU can read the next message from. @@ -345,27 +347,12 @@ impl DmaGspMem { // // - The returned value is between `0` and `MSGQ_NUM_PAGES`. fn cpu_read_ptr(&self) -> u32 { - let gsp_mem = self.0.start_ptr(); - - // SAFETY: - // - The ['CoherentAllocation'] contains at least one object. - // - By the invariants of CoherentAllocation the pointer is valid. - (unsafe { (*gsp_mem).cpuq.rx.read_ptr() } % MSGQ_NUM_PAGES) + super::fw::gsp_mem::cpu_read_ptr(&self.0) } // Informs the GSP that it can send `elem_count` new pages into the message queue. fn advance_cpu_read_ptr(&mut self, elem_count: u32) { - let rptr = self.cpu_read_ptr().wrapping_add(elem_count) % MSGQ_NUM_PAGES; - - // Ensure read pointer is properly ordered. - fence(Ordering::SeqCst); - - let gsp_mem = self.0.start_ptr_mut(); - - // SAFETY: - // - The 'CoherentAllocation' contains at least one object. - // - By the invariants of `CoherentAllocation` the pointer is valid. - unsafe { (*gsp_mem).cpuq.rx.set_read_ptr(rptr) }; + super::fw::gsp_mem::advance_cpu_read_ptr(&self.0, elem_count) } // Returns the index of the memory page the CPU can write the next command to. @@ -374,26 +361,12 @@ impl DmaGspMem { // // - The returned value is between `0` and `MSGQ_NUM_PAGES`. fn cpu_write_ptr(&self) -> u32 { - let gsp_mem = self.0.start_ptr(); - - // SAFETY: - // - The 'CoherentAllocation' contains at least one object. - // - By the invariants of `CoherentAllocation` the pointer is valid. - (unsafe { (*gsp_mem).cpuq.tx.write_ptr() } % MSGQ_NUM_PAGES) + super::fw::gsp_mem::cpu_write_ptr(&self.0) } // Informs the GSP that it can process `elem_count` new pages from the command queue. fn advance_cpu_write_ptr(&mut self, elem_count: u32) { - let wptr = self.cpu_write_ptr().wrapping_add(elem_count) & MSGQ_NUM_PAGES; - let gsp_mem = self.0.start_ptr_mut(); - - // SAFETY: - // - The 'CoherentAllocation' contains at least one object. - // - By the invariants of `CoherentAllocation` the pointer is valid. - unsafe { (*gsp_mem).cpuq.tx.set_write_ptr(wptr) }; - - // Ensure all command data is visible before triggering the GSP read. - fence(Ordering::SeqCst); + super::fw::gsp_mem::advance_cpu_write_ptr(&self.0, elem_count) } } diff --git a/drivers/gpu/nova-core/gsp/fw.rs b/drivers/gpu/nova-core/gsp/fw.rs index 83ff91614e36..040b30ec3089 100644 --- a/drivers/gpu/nova-core/gsp/fw.rs +++ b/drivers/gpu/nova-core/gsp/fw.rs @@ -40,6 +40,75 @@ use crate::{ }, }; +// TODO: Replace with `IoView` projections once available; the `unwrap()` calls go away once we +// switch to the new `dma::Coherent` API. +pub(super) mod gsp_mem { + use core::sync::atomic::{ + fence, + Ordering, // + }; + + use kernel::{ + dma::CoherentAllocation, + dma_read, + dma_write, + prelude::*, // + }; + + use crate::gsp::cmdq::{ + GspMem, + MSGQ_NUM_PAGES, // + }; + + pub(in crate::gsp) fn gsp_write_ptr(qs: &CoherentAllocation<GspMem>) -> u32 { + // PANIC: A `dma::CoherentAllocation` always contains at least one element. + || -> Result<u32> { Ok(dma_read!(qs, [0]?.gspq.tx.0.writePtr) % MSGQ_NUM_PAGES) }().unwrap() + } + + pub(in crate::gsp) fn gsp_read_ptr(qs: &CoherentAllocation<GspMem>) -> u32 { + // PANIC: A `dma::CoherentAllocation` always contains at least one element. + || -> Result<u32> { Ok(dma_read!(qs, [0]?.gspq.rx.0.readPtr) % MSGQ_NUM_PAGES) }().unwrap() + } + + pub(in crate::gsp) fn cpu_read_ptr(qs: &CoherentAllocation<GspMem>) -> u32 { + // PANIC: A `dma::CoherentAllocation` always contains at least one element. + || -> Result<u32> { Ok(dma_read!(qs, [0]?.cpuq.rx.0.readPtr) % MSGQ_NUM_PAGES) }().unwrap() + } + + pub(in crate::gsp) fn advance_cpu_read_ptr(qs: &CoherentAllocation<GspMem>, count: u32) { + let rptr = cpu_read_ptr(qs).wrapping_add(count) % MSGQ_NUM_PAGES; + + // Ensure read pointer is properly ordered. + fence(Ordering::SeqCst); + + // PANIC: A `dma::CoherentAllocation` always contains at least one element. + || -> Result { + dma_write!(qs, [0]?.cpuq.rx.0.readPtr, rptr); + Ok(()) + }() + .unwrap() + } + + pub(in crate::gsp) fn cpu_write_ptr(qs: &CoherentAllocation<GspMem>) -> u32 { + // PANIC: A `dma::CoherentAllocation` always contains at least one element. + || -> Result<u32> { Ok(dma_read!(qs, [0]?.cpuq.tx.0.writePtr) % MSGQ_NUM_PAGES) }().unwrap() + } + + pub(in crate::gsp) fn advance_cpu_write_ptr(qs: &CoherentAllocation<GspMem>, count: u32) { + let wptr = cpu_write_ptr(qs).wrapping_add(count) % MSGQ_NUM_PAGES; + + // PANIC: A `dma::CoherentAllocation` always contains at least one element. + || -> Result { + dma_write!(qs, [0]?.cpuq.tx.0.writePtr, wptr); + Ok(()) + }() + .unwrap(); + + // Ensure all command data is visible before triggering the GSP read. + fence(Ordering::SeqCst); + } +} + /// Empty type to group methods related to heap parameters for running the GSP firmware. enum GspFwHeapParams {} @@ -708,22 +777,6 @@ impl MsgqTxHeader { entryOff: num::usize_into_u32::<GSP_PAGE_SIZE>(), }) } - - /// Returns the value of the write pointer for this queue. - pub(crate) fn write_ptr(&self) -> u32 { - let ptr = core::ptr::from_ref(&self.0.writePtr); - - // SAFETY: `ptr` is a valid pointer to a `u32`. - unsafe { ptr.read_volatile() } - } - - /// Sets the value of the write pointer for this queue. - pub(crate) fn set_write_ptr(&mut self, val: u32) { - let ptr = core::ptr::from_mut(&mut self.0.writePtr); - - // SAFETY: `ptr` is a valid pointer to a `u32`. - unsafe { ptr.write_volatile(val) } - } } // SAFETY: Padding is explicit and does not contain uninitialized data. @@ -739,22 +792,6 @@ impl MsgqRxHeader { pub(crate) fn new() -> Self { Self(Default::default()) } - - /// Returns the value of the read pointer for this queue. - pub(crate) fn read_ptr(&self) -> u32 { - let ptr = core::ptr::from_ref(&self.0.readPtr); - - // SAFETY: `ptr` is a valid pointer to a `u32`. - unsafe { ptr.read_volatile() } - } - - /// Sets the value of the read pointer for this queue. - pub(crate) fn set_read_ptr(&mut self, val: u32) { - let ptr = core::ptr::from_mut(&mut self.0.readPtr); - - // SAFETY: `ptr` is a valid pointer to a `u32`. - unsafe { ptr.write_volatile(val) } - } } // SAFETY: Padding is explicit and does not contain uninitialized data. |