Merge tag 'v6.0' into rdma.git for-next

Trvial merge conflicts against rdma.git for-rc resolved matching
linux-next:
            drivers/infiniband/hw/hns/hns_roce_hw_v2.c
            drivers/infiniband/hw/hns/hns_roce_main.c

https://lore.kernel.org/r/20220929124005.105149-1-broonie@kernel.org

Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

57 files changed, 979 insertions, 340 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/aldebaran.c b/drivers/gpu/drm/amd/amdgpu/aldebaran.c
index c6cc493a5486..2b97b8a96fb4 100644
--- a/drivers/gpu/drm/amd/amdgpu/aldebaran.c
+++ b/drivers/gpu/drm/amd/amdgpu/aldebaran.c
@@ -148,30 +148,22 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl,
 			      struct amdgpu_reset_context *reset_context)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)reset_ctl->handle;
+	struct list_head *reset_device_list = reset_context->reset_device_list;
 	struct amdgpu_device *tmp_adev = NULL;
-	struct list_head reset_device_list;
 	int r = 0;
 
 	dev_dbg(adev->dev, "aldebaran perform hw reset\n");
+
+	if (reset_device_list == NULL)
+		return -EINVAL;
+
 	if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2) &&
 	    reset_context->hive == NULL) {
 		/* Wrong context, return error */
 		return -EINVAL;
 	}
 
-	INIT_LIST_HEAD(&reset_device_list);
-	if (reset_context->hive) {
-		list_for_each_entry (tmp_adev,
-				     &reset_context->hive->device_list,
-				     gmc.xgmi.head)
-			list_add_tail(&tmp_adev->reset_list,
-				      &reset_device_list);
-	} else {
-		list_add_tail(&reset_context->reset_req_dev->reset_list,
-			      &reset_device_list);
-	}
-
-	list_for_each_entry (tmp_adev, &reset_device_list, reset_list) {
+	list_for_each_entry(tmp_adev, reset_device_list, reset_list) {
 		mutex_lock(&tmp_adev->reset_cntl->reset_lock);
 		tmp_adev->reset_cntl->active_reset = AMD_RESET_METHOD_MODE2;
 	}
@@ -179,7 +171,7 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl,
 	 * Mode2 reset doesn't need any sync between nodes in XGMI hive, instead launch
 	 * them together so that they can be completed asynchronously on multiple nodes
 	 */
-	list_for_each_entry (tmp_adev, &reset_device_list, reset_list) {
+	list_for_each_entry(tmp_adev, reset_device_list, reset_list) {
 		/* For XGMI run all resets in parallel to speed up the process */
 		if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) {
 			if (!queue_work(system_unbound_wq,
@@ -197,7 +189,7 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl,
 
 	/* For XGMI wait for all resets to complete before proceed */
 	if (!r) {
-		list_for_each_entry (tmp_adev, &reset_device_list, reset_list) {
+		list_for_each_entry(tmp_adev, reset_device_list, reset_list) {
 			if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) {
 				flush_work(&tmp_adev->reset_cntl->reset_work);
 				r = tmp_adev->asic_reset_res;
@@ -207,7 +199,7 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl,
 		}
 	}
 
-	list_for_each_entry (tmp_adev, &reset_device_list, reset_list) {
+	list_for_each_entry(tmp_adev, reset_device_list, reset_list) {
 		mutex_unlock(&tmp_adev->reset_cntl->reset_lock);
 		tmp_adev->reset_cntl->active_reset = AMD_RESET_METHOD_NONE;
 	}
@@ -339,10 +331,13 @@ static int
 aldebaran_mode2_restore_hwcontext(struct amdgpu_reset_control *reset_ctl,
 				  struct amdgpu_reset_context *reset_context)
 {
+	struct list_head *reset_device_list = reset_context->reset_device_list;
 	struct amdgpu_device *tmp_adev = NULL;
-	struct list_head reset_device_list;
 	int r;
 
+	if (reset_device_list == NULL)
+		return -EINVAL;
+
 	if (reset_context->reset_req_dev->ip_versions[MP1_HWIP][0] ==
 		    IP_VERSION(13, 0, 2) &&
 	    reset_context->hive == NULL) {
@@ -350,19 +345,7 @@ aldebaran_mode2_restore_hwcontext(struct amdgpu_reset_control *reset_ctl,
 		return -EINVAL;
 	}
 
-	INIT_LIST_HEAD(&reset_device_list);
-	if (reset_context->hive) {
-		list_for_each_entry (tmp_adev,
-				     &reset_context->hive->device_list,
-				     gmc.xgmi.head)
-			list_add_tail(&tmp_adev->reset_list,
-				      &reset_device_list);
-	} else {
-		list_add_tail(&reset_context->reset_req_dev->reset_list,
-			      &reset_device_list);
-	}
-
-	list_for_each_entry (tmp_adev, &reset_device_list, reset_list) {
+	list_for_each_entry(tmp_adev, reset_device_list, reset_list) {
 		dev_info(tmp_adev->dev,
 			 "GPU reset succeeded, trying to resume\n");
 		r = aldebaran_mode2_restore_ip(tmp_adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index e146810c700b..d597e2656c47 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -317,7 +317,7 @@ enum amdgpu_kiq_irq {
 	AMDGPU_CP_KIQ_IRQ_DRIVER0 = 0,
 	AMDGPU_CP_KIQ_IRQ_LAST
 };
-
+#define SRIOV_USEC_TIMEOUT  1200000 /* wait 12 * 100ms for SRIOV */
 #define MAX_KIQ_REG_WAIT       5000 /* in usecs, 5ms */
 #define MAX_KIQ_REG_BAILOUT_INTERVAL   5 /* in msecs, 5ms */
 #define MAX_KIQ_REG_TRY 1000
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 130060834b4e..48bd660ddb85 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -1050,6 +1050,10 @@ bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev)
 {
 	if (adev->flags & AMD_IS_APU)
 		return false;
+
+	if (amdgpu_sriov_vf(adev))
+		return false;
+
 	return pm_suspend_target_state != PM_SUSPEND_TO_IDLE;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 3c09dcc0986e..647220a8762d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -96,6 +96,7 @@ struct amdgpu_amdkfd_fence {
 struct amdgpu_kfd_dev {
 	struct kfd_dev *dev;
 	uint64_t vram_used;
+	uint64_t vram_used_aligned;
 	bool init_complete;
 	struct work_struct reset_work;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index a699134a1e8c..2170db83e41d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -40,10 +40,10 @@
 #define AMDGPU_USERPTR_RESTORE_DELAY_MS 1
 
 /*
- * Align VRAM allocations to 2MB to avoid fragmentation caused by 4K allocations in the tail 2MB
+ * Align VRAM availability to 2MB to avoid fragmentation caused by 4K allocations in the tail 2MB
  * BO chunk
  */
-#define VRAM_ALLOCATION_ALIGN (1 << 21)
+#define VRAM_AVAILABLITY_ALIGN (1 << 21)
 
 /* Impose limit on how much memory KFD can use */
 static struct {
@@ -149,7 +149,7 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 		 * to avoid fragmentation caused by 4K allocations in the tail
 		 * 2M BO chunk.
 		 */
-		vram_needed = ALIGN(size, VRAM_ALLOCATION_ALIGN);
+		vram_needed = size;
 	} else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) {
 		system_mem_needed = size;
 	} else if (!(alloc_flag &
@@ -182,8 +182,10 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 	 */
 	WARN_ONCE(vram_needed && !adev,
 		  "adev reference can't be null when vram is used");
-	if (adev)
+	if (adev) {
 		adev->kfd.vram_used += vram_needed;
+		adev->kfd.vram_used_aligned += ALIGN(vram_needed, VRAM_AVAILABLITY_ALIGN);
+	}
 	kfd_mem_limit.system_mem_used += system_mem_needed;
 	kfd_mem_limit.ttm_mem_used += ttm_mem_needed;
 
@@ -203,8 +205,10 @@ void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
 	} else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) {
 		WARN_ONCE(!adev,
 			  "adev reference can't be null when alloc mem flags vram is set");
-		if (adev)
-			adev->kfd.vram_used -= ALIGN(size, VRAM_ALLOCATION_ALIGN);
+		if (adev) {
+			adev->kfd.vram_used -= size;
+			adev->kfd.vram_used_aligned -= ALIGN(size, VRAM_AVAILABLITY_ALIGN);
+		}
 	} else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) {
 		kfd_mem_limit.system_mem_used -= size;
 	} else if (!(alloc_flag &
@@ -1608,15 +1612,14 @@ size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev)
 	uint64_t reserved_for_pt =
 		ESTIMATE_PT_SIZE(amdgpu_amdkfd_total_mem_size);
 	size_t available;
-
 	spin_lock(&kfd_mem_limit.mem_limit_lock);
 	available = adev->gmc.real_vram_size
-		- adev->kfd.vram_used
+		- adev->kfd.vram_used_aligned
 		- atomic64_read(&adev->vram_pin_size)
 		- reserved_for_pt;
 	spin_unlock(&kfd_mem_limit.mem_limit_lock);
 
-	return ALIGN_DOWN(available, VRAM_ALLOCATION_ALIGN);
+	return ALIGN_DOWN(available, VRAM_AVAILABLITY_ALIGN);
 }
 
 int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
@@ -1725,7 +1728,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, user_addr);
 
 	if (user_addr) {
-		pr_debug("creating userptr BO for user_addr = %llu\n", user_addr);
+		pr_debug("creating userptr BO for user_addr = %llx\n", user_addr);
 		ret = init_user_pages(*mem, user_addr, criu_resume);
 		if (ret)
 			goto allocate_init_user_pages_failed;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
index fd8f3731758e..b81b77a9efa6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
@@ -314,7 +314,7 @@ amdgpu_atomfirmware_get_vram_info(struct amdgpu_device *adev,
 					mem_channel_number = vram_info->v30.channel_num;
 					mem_channel_width = vram_info->v30.channel_width;
 					if (vram_width)
-						*vram_width = mem_channel_number * mem_channel_width;
+						*vram_width = mem_channel_number * (1 << mem_channel_width);
 					break;
 				default:
 					return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index d8f1335bc68f..b7bae833c804 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -837,16 +837,12 @@ static int amdgpu_cs_vm_handling(struct amdgpu_cs_parser *p)
 			continue;
 
 		r = amdgpu_vm_bo_update(adev, bo_va, false);
-		if (r) {
-			mutex_unlock(&p->bo_list->bo_list_mutex);
+		if (r)
 			return r;
-		}
 
 		r = amdgpu_sync_fence(&p->job->sync, bo_va->last_pt_update);
-		if (r) {
-			mutex_unlock(&p->bo_list->bo_list_mutex);
+		if (r)
 			return r;
-		}
 	}
 
 	r = amdgpu_vm_handle_moved(adev, vm);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index e2eec985adb3..cb00c7d6f50b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -1705,7 +1705,7 @@ static ssize_t amdgpu_reset_dump_register_list_write(struct file *f,
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
 	char reg_offset[11];
-	uint32_t *new, *tmp = NULL;
+	uint32_t *new = NULL, *tmp = NULL;
 	int ret, i = 0, len = 0;
 
 	do {
@@ -1747,7 +1747,8 @@ static ssize_t amdgpu_reset_dump_register_list_write(struct file *f,
 	ret = size;
 
 error_free:
-	kfree(tmp);
+	if (tmp != new)
+		kfree(tmp);
 	kfree(new);
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index c4a6fe3070b6..25e1f5ed7ead 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2365,8 +2365,16 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 		}
 		adev->ip_blocks[i].status.sw = true;
 
-		/* need to do gmc hw init early so we can allocate gpu mem */
-		if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC) {
+		if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON) {
+			/* need to do common hw init early so everything is set up for gmc */
+			r = adev->ip_blocks[i].version->funcs->hw_init((void *)adev);
+			if (r) {
+				DRM_ERROR("hw_init %d failed %d\n", i, r);
+				goto init_failed;
+			}
+			adev->ip_blocks[i].status.hw = true;
+		} else if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC) {
+			/* need to do gmc hw init early so we can allocate gpu mem */
 			/* Try to reserve bad pages early */
 			if (amdgpu_sriov_vf(adev))
 				amdgpu_virt_exchange_data(adev);
@@ -2456,12 +2464,14 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 			if (!hive->reset_domain ||
 			    !amdgpu_reset_get_reset_domain(hive->reset_domain)) {
 				r = -ENOENT;
+				amdgpu_put_xgmi_hive(hive);
 				goto init_failed;
 			}
 
 			/* Drop the early temporary reset domain we created for device */
 			amdgpu_reset_put_reset_domain(adev->reset_domain);
 			adev->reset_domain = hive->reset_domain;
+			amdgpu_put_xgmi_hive(hive);
 		}
 	}
 
@@ -3050,8 +3060,8 @@ static int amdgpu_device_ip_reinit_early_sriov(struct amdgpu_device *adev)
 	int i, r;
 
 	static enum amd_ip_block_type ip_order[] = {
-		AMD_IP_BLOCK_TYPE_GMC,
 		AMD_IP_BLOCK_TYPE_COMMON,
+		AMD_IP_BLOCK_TYPE_GMC,
 		AMD_IP_BLOCK_TYPE_PSP,
 		AMD_IP_BLOCK_TYPE_IH,
 	};
@@ -3142,7 +3152,8 @@ static int amdgpu_device_ip_resume_phase1(struct amdgpu_device *adev)
 			continue;
 		if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON ||
 		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC ||
-		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH) {
+		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH ||
+		    (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP && amdgpu_sriov_vf(adev))) {
 
 			r = adev->ip_blocks[i].version->funcs->resume(adev);
 			if (r) {
@@ -4054,12 +4065,20 @@ static void amdgpu_device_evict_resources(struct amdgpu_device *adev)
 int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 {
 	struct amdgpu_device *adev = drm_to_adev(dev);
+	int r = 0;
 
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
 
 	adev->in_suspend = true;
 
+	if (amdgpu_sriov_vf(adev)) {
+		amdgpu_virt_fini_data_exchange(adev);
+		r = amdgpu_virt_request_full_gpu(adev, false);
+		if (r)
+			return r;
+	}
+
 	if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D3))
 		DRM_WARN("smart shift update failed\n");
 
@@ -4083,6 +4102,9 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 
 	amdgpu_device_ip_suspend_phase2(adev);
 
+	if (amdgpu_sriov_vf(adev))
+		amdgpu_virt_release_full_gpu(adev, false);
+
 	return 0;
 }
 
@@ -4101,6 +4123,12 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	int r = 0;
 
+	if (amdgpu_sriov_vf(adev)) {
+		r = amdgpu_virt_request_full_gpu(adev, true);
+		if (r)
+			return r;
+	}
+
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
 
@@ -4115,6 +4143,13 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 	}
 
 	r = amdgpu_device_ip_resume(adev);
+
+	/* no matter what r is, always need to properly release full GPU */
+	if (amdgpu_sriov_vf(adev)) {
+		amdgpu_virt_init_data_exchange(adev);
+		amdgpu_virt_release_full_gpu(adev, true);
+	}
+
 	if (r) {
 		dev_err(adev->dev, "amdgpu_device_ip_resume failed (%d).\n", r);
 		return r;
@@ -4413,8 +4448,6 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev,
 retry:
 	amdgpu_amdkfd_pre_reset(adev);
 
-	amdgpu_amdkfd_pre_reset(adev);
-
 	if (from_hypervisor)
 		r = amdgpu_virt_request_full_gpu(adev, true);
 	else
@@ -4742,6 +4775,8 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle,
 	tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device,
 				    reset_list);
 	amdgpu_reset_reg_dumps(tmp_adev);
+
+	reset_context->reset_device_list = device_list_handle;
 	r = amdgpu_reset_perform_reset(tmp_adev, reset_context);
 	/* If reset handler not implemented, continue; otherwise return */
 	if (r == -ENOSYS)
@@ -5522,7 +5557,8 @@ bool amdgpu_device_is_peer_accessible(struct amdgpu_device *adev,
 		~*peer_adev->dev->dma_mask : ~((1ULL << 32) - 1);
 	resource_size_t aper_limit =
 		adev->gmc.aper_base + adev->gmc.aper_size - 1;
-	bool p2p_access = !(pci_p2pdma_distance_many(adev->pdev,
+	bool p2p_access = !adev->gmc.xgmi.connected_to_cpu &&
+			  !(pci_p2pdma_distance_many(adev->pdev,
 					&peer_adev->dev, 1, true) < 0);
 
 	return pcie_p2p && p2p_access && (adev->gmc.visible_vram_size &&
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index c20922a5af9f..23998f727c7f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -38,6 +38,8 @@
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_damage_helper.h>
+#include <drm/drm_drv.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_fb_helper.h>
@@ -498,6 +500,12 @@ static const struct drm_framebuffer_funcs amdgpu_fb_funcs = {
 	.create_handle = drm_gem_fb_create_handle,
 };
 
+static const struct drm_framebuffer_funcs amdgpu_fb_funcs_atomic = {
+	.destroy = drm_gem_fb_destroy,
+	.create_handle = drm_gem_fb_create_handle,
+	.dirty = drm_atomic_helper_dirtyfb,
+};
+
 uint32_t amdgpu_display_supported_domains(struct amdgpu_device *adev,
 					  uint64_t bo_flags)
 {
@@ -1100,7 +1108,10 @@ static int amdgpu_display_gem_fb_verify_and_init(struct drm_device *dev,
 	if (ret)
 		goto err;
 
-	ret = drm_framebuffer_init(dev, &rfb->base, &amdgpu_fb_funcs);
+	if (drm_drv_uses_atomic_modeset(dev))
+		ret = drm_framebuffer_init(dev, &rfb->base, &amdgpu_fb_funcs_atomic);
+	else
+		ret = drm_framebuffer_init(dev, &rfb->base, &amdgpu_fb_funcs);
 	if (ret)
 		goto err;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 8adeb7469f1e..d0d99ed607dd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -400,7 +400,6 @@ unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring)
 	/* We are not protected by ring lock when reading the last sequence
 	 * but it's ok to report slightly wrong fence count here.
 	 */
-	amdgpu_fence_process(ring);
 	emitted = 0x100000000ull;
 	emitted -= atomic_read(&ring->fence_drv.last_seq);
 	emitted += READ_ONCE(ring->fence_drv.sync_seq);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c
index ecada5eadfe3..e325150879df 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c
@@ -66,10 +66,15 @@ static bool is_fru_eeprom_supported(struct amdgpu_device *adev)
 		return true;
 	case CHIP_SIENNA_CICHLID:
 		if (strnstr(atom_ctx->vbios_version, "D603",
+		    sizeof(atom_ctx->vbios_version))) {
+			if (strnstr(atom_ctx->vbios_version, "D603GLXE",
 			    sizeof(atom_ctx->vbios_version)))
-			return true;
-		else
+				return false;
+			else
+				return true;
+		} else {
 			return false;
+		}
 	default:
 		return false;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 23a696d38390..de5b936b016d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -304,6 +304,10 @@ struct amdgpu_gfx {
 	uint32_t			rlc_srlg_feature_version;
 	uint32_t			rlc_srls_fw_version;
 	uint32_t			rlc_srls_feature_version;
+	uint32_t			rlcp_ucode_version;
+	uint32_t			rlcp_ucode_feature_version;
+	uint32_t			rlcv_ucode_version;
+	uint32_t			rlcv_ucode_feature_version;
 	uint32_t			mec_feature_version;
 	uint32_t			mec2_feature_version;
 	bool				mec_fw_write_wait;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index 5071b96be982..c2fd6f3076a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -159,7 +159,10 @@ void amdgpu_job_free(struct amdgpu_job *job)
 	amdgpu_sync_free(&job->sync);
 	amdgpu_sync_free(&job->sched_sync);
 
-	dma_fence_put(&job->hw_fence);
+	if (!job->hw_fence.ops)
+		kfree(job);
+	else
+		dma_fence_put(&job->hw_fence);
 }
 
 int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity,
@@ -272,10 +275,6 @@ void amdgpu_job_stop_all_jobs_on_sched(struct drm_gpu_scheduler *sched)
 	/* Signal all jobs not yet scheduled */
 	for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
 		struct drm_sched_rq *rq = &sched->sched_rq[i];
-
-		if (!rq)
-			continue;
-
 		spin_lock(&rq->lock);
 		list_for_each_entry(s_entity, &rq->entities, list) {
 			while ((s_job = to_drm_sched_job(spsc_queue_pop(&s_entity->job_queue)))) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index fe82b8b19a4e..0c546245793b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -181,6 +181,9 @@ int amdgpu_mes_init(struct amdgpu_device *adev)
 	for (i = 0; i < AMDGPU_MES_MAX_SDMA_PIPES; i++) {
 		if (adev->ip_versions[SDMA0_HWIP][0] < IP_VERSION(6, 0, 0))
 			adev->mes.sdma_hqd_mask[i] = i ? 0 : 0x3fc;
+		/* zero sdma_hqd_mask for non-existent engine */
+		else if (adev->sdma.num_instances == 1)
+			adev->mes.sdma_hqd_mask[i] = i ? 0 : 0xfc;
 		else
 			adev->mes.sdma_hqd_mask[i] = 0xfc;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
index 7b46f6bf4187..ad980f4b66e1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
@@ -222,6 +222,8 @@ struct mes_add_queue_input {
 	uint64_t	tba_addr;
 	uint64_t	tma_addr;
 	uint32_t	is_kfd_process;
+	uint32_t	is_aql_queue;
+	uint32_t	queue_size;
 };
 
 struct mes_remove_queue_input {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index b067ce45d226..c9dec2434f37 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -486,11 +486,14 @@ static int psp_sw_fini(void *handle)
 		release_firmware(psp->ta_fw);
 		psp->ta_fw = NULL;
 	}
-	if (adev->psp.cap_fw) {
+	if (psp->cap_fw) {
 		release_firmware(psp->cap_fw);
 		psp->cap_fw = NULL;
 	}
-
+	if (psp->toc_fw) {
+		release_firmware(psp->toc_fw);
+		psp->toc_fw = NULL;
+	}
 	if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 0) ||
 	    adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 7))
 		psp_sysfs_fini(adev);
@@ -753,7 +756,7 @@ static int psp_tmr_init(struct psp_context *psp)
 	}
 
 	pptr = amdgpu_sriov_vf(psp->adev) ? &tmr_buf : NULL;
-	ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_SIZE(psp->adev),
+	ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_ALIGNMENT,
 				      AMDGPU_GEM_DOMAIN_VRAM,
 				      &psp->tmr_bo, &psp->tmr_mc_addr, pptr);
 
@@ -2401,7 +2404,7 @@ static int psp_load_smu_fw(struct psp_context *psp)
 static bool fw_load_skip_check(struct psp_context *psp,
 			       struct amdgpu_firmware_info *ucode)
 {
-	if (!ucode->fw)
+	if (!ucode->fw || !ucode->ucode_size)
 		return true;
 
 	if (ucode->ucode_id == AMDGPU_UCODE_ID_SMC &&
@@ -2641,6 +2644,9 @@ static int psp_hw_fini(void *handle)
 		psp_rap_terminate(psp);
 		psp_dtm_terminate(psp);
 		psp_hdcp_terminate(psp);
+
+		if (adev->gmc.xgmi.num_physical_nodes > 1)
+			psp_xgmi_terminate(psp);
 	}
 
 	psp_asd_terminate(psp);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index c32b74bd970f..e593e8c2a54d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -36,6 +36,7 @@
 #define PSP_CMD_BUFFER_SIZE	0x1000
 #define PSP_1_MEG		0x100000
 #define PSP_TMR_SIZE(adev)	((adev)->asic_type == CHIP_ALDEBARAN ? 0x800000 : 0x400000)
+#define PSP_TMR_ALIGNMENT	0x100000
 #define PSP_FW_NAME_LEN		0x24
 
 enum psp_shared_mem_size {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index ff5361f5c2d4..12c6f97945a5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1811,7 +1811,8 @@ static void amdgpu_ras_log_on_err_counter(struct amdgpu_device *adev)
 		amdgpu_ras_query_error_status(adev, &info);
 
 		if (adev->ip_versions[MP0_HWIP][0] != IP_VERSION(11, 0, 2) &&
-		    adev->ip_versions[MP0_HWIP][0] != IP_VERSION(11, 0, 4)) {
+		    adev->ip_versions[MP0_HWIP][0] != IP_VERSION(11, 0, 4) &&
+		    adev->ip_versions[MP0_HWIP][0] != IP_VERSION(13, 0, 0)) {
 			if (amdgpu_ras_reset_error_status(adev, info.head.block))
 				dev_warn(adev->dev, "Failed to reset error counter and error status");
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
index 9e55a5d7a825..ffda1560c648 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
@@ -37,6 +37,7 @@ struct amdgpu_reset_context {
 	struct amdgpu_device *reset_req_dev;
 	struct amdgpu_job *job;
 	struct amdgpu_hive_info *hive;
+	struct list_head *reset_device_list;
 	unsigned long flags;
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
index 6373bfb47d55..e23f6192c50e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
@@ -272,3 +272,267 @@ void amdgpu_gfx_rlc_fini(struct amdgpu_device *adev)
 			      &adev->gfx.rlc.cp_table_gpu_addr,
 			      (void **)&adev->gfx.rlc.cp_table_ptr);
 }
+
+static int amdgpu_gfx_rlc_init_microcode_v2_0(struct amdgpu_device *adev)
+{
+	const struct common_firmware_header *common_hdr;
+	const struct rlc_firmware_header_v2_0 *rlc_hdr;
+	struct amdgpu_firmware_info *info;
+	unsigned int *tmp;
+	unsigned int i;
+
+	rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data;
+
+	adev->gfx.rlc_fw_version = le32_to_cpu(rlc_hdr->header.ucode_version);
+	adev->gfx.rlc_feature_version = le32_to_cpu(rlc_hdr->ucode_feature_version);
+	adev->gfx.rlc.save_and_restore_offset =
+		le32_to_cpu(rlc_hdr->save_and_restore_offset);
+	adev->gfx.rlc.clear_state_descriptor_offset =
+		le32_to_cpu(rlc_hdr->clear_state_descriptor_offset);
+	adev->gfx.rlc.avail_scratch_ram_locations =
+		le32_to_cpu(rlc_hdr->avail_scratch_ram_locations);
+	adev->gfx.rlc.reg_restore_list_size =
+		le32_to_cpu(rlc_hdr->reg_restore_list_size);
+	adev->gfx.rlc.reg_list_format_start =
+		le32_to_cpu(rlc_hdr->reg_list_format_start);
+	adev->gfx.rlc.reg_list_format_separate_start =
+		le32_to_cpu(rlc_hdr->reg_list_format_separate_start);
+	adev->gfx.rlc.starting_offsets_start =
+		le32_to_cpu(rlc_hdr->starting_offsets_start);
+	adev->gfx.rlc.reg_list_format_size_bytes =
+		le32_to_cpu(rlc_hdr->reg_list_format_size_bytes);
+	adev->gfx.rlc.reg_list_size_bytes =
+		le32_to_cpu(rlc_hdr->reg_list_size_bytes);
+	adev->gfx.rlc.register_list_format =
+		kmalloc(adev->gfx.rlc.reg_list_format_size_bytes +
+			adev->gfx.rlc.reg_list_size_bytes, GFP_KERNEL);
+	if (!adev->gfx.rlc.register_list_format) {
+		dev_err(adev->dev, "failed to allocate memory for rlc register_list_format\n");
+		return -ENOMEM;
+	}
+
+	tmp = (unsigned int *)((uintptr_t)rlc_hdr +
+			le32_to_cpu(rlc_hdr->reg_list_format_array_offset_bytes));
+	for (i = 0 ; i < (rlc_hdr->reg_list_format_size_bytes >> 2); i++)
+		adev->gfx.rlc.register_list_format[i] = le32_to_cpu(tmp[i]);
+
+	adev->gfx.rlc.register_restore = adev->gfx.rlc.register_list_format + i;
+
+	tmp = (unsigned int *)((uintptr_t)rlc_hdr +
+			le32_to_cpu(rlc_hdr->reg_list_array_offset_bytes));
+	for (i = 0 ; i < (rlc_hdr->reg_list_size_bytes >> 2); i++)
+		adev->gfx.rlc.register_restore[i] = le32_to_cpu(tmp[i]);
+
+	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+		info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_G];
+		info->ucode_id = AMDGPU_UCODE_ID_RLC_G;
+		info->fw = adev->gfx.rlc_fw;
+		if (info->fw) {
+			common_hdr = (const struct common_firmware_header *)info->fw->data;
+			adev->firmware.fw_size +=
+				ALIGN(le32_to_cpu(common_hdr->ucode_size_bytes), PAGE_SIZE);
+		}
+	}
+
+	return 0;
+}
+
+static void amdgpu_gfx_rlc_init_microcode_v2_1(struct amdgpu_device *adev)
+{
+	const struct rlc_firmware_header_v2_1 *rlc_hdr;
+	struct amdgpu_firmware_info *info;
+
+	rlc_hdr = (const struct rlc_firmware_header_v2_1 *)adev->gfx.rlc_fw->data;
+	adev->gfx.rlc_srlc_fw_version = le32_to_cpu(rlc_hdr->save_restore_list_cntl_ucode_ver);
+	adev->gfx.rlc_srlc_feature_version = le32_to_cpu(rlc_hdr->save_restore_list_cntl_feature_ver);
+	adev->gfx.rlc.save_restore_list_cntl_size_bytes = le32_to_cpu(rlc_hdr->save_restore_list_cntl_size_bytes);
+	adev->gfx.rlc.save_restore_list_cntl = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->save_restore_list_cntl_offset_bytes);
+	adev->gfx.rlc_srlg_fw_version = le32_to_cpu(rlc_hdr->save_restore_list_gpm_ucode_ver);
+	adev->gfx.rlc_srlg_feature_version = le32_to_cpu(rlc_hdr->save_restore_list_gpm_feature_ver);
+	adev->gfx.rlc.save_restore_list_gpm_size_bytes = le32_to_cpu(rlc_hdr->save_restore_list_gpm_size_bytes);
+	adev->gfx.rlc.save_restore_list_gpm = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->save_restore_list_gpm_offset_bytes);
+	adev->gfx.rlc_srls_fw_version = le32_to_cpu(rlc_hdr->save_restore_list_srm_ucode_ver);
+	adev->gfx.rlc_srls_feature_version = le32_to_cpu(rlc_hdr->save_restore_list_srm_feature_ver);
+	adev->gfx.rlc.save_restore_list_srm_size_bytes = le32_to_cpu(rlc_hdr->save_restore_list_srm_size_bytes);
+	adev->gfx.rlc.save_restore_list_srm = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->save_restore_list_srm_offset_bytes);
+	adev->gfx.rlc.reg_list_format_direct_reg_list_length =
+		le32_to_cpu(rlc_hdr->reg_list_format_direct_reg_list_length);
+
+	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+		if (adev->gfx.rlc.save_restore_list_gpm_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM];
+			info->ucode_id = AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.save_restore_list_gpm_size_bytes, PAGE_SIZE);
+		}
+
+		if (adev->gfx.rlc.save_restore_list_srm_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM];
+			info->ucode_id = AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.save_restore_list_srm_size_bytes, PAGE_SIZE);
+		}
+	}
+}
+
+static void amdgpu_gfx_rlc_init_microcode_v2_2(struct amdgpu_device *adev)
+{
+	const struct rlc_firmware_header_v2_2 *rlc_hdr;
+	struct amdgpu_firmware_info *info;
+
+	rlc_hdr = (const struct rlc_firmware_header_v2_2 *)adev->gfx.rlc_fw->data;
+	adev->gfx.rlc.rlc_iram_ucode_size_bytes = le32_to_cpu(rlc_hdr->rlc_iram_ucode_size_bytes);
+	adev->gfx.rlc.rlc_iram_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->rlc_iram_ucode_offset_bytes);
+	adev->gfx.rlc.rlc_dram_ucode_size_bytes = le32_to_cpu(rlc_hdr->rlc_dram_ucode_size_bytes);
+	adev->gfx.rlc.rlc_dram_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->rlc_dram_ucode_offset_bytes);
+
+	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+		if (adev->gfx.rlc.rlc_iram_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_IRAM];
+			info->ucode_id = AMDGPU_UCODE_ID_RLC_IRAM;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.rlc_iram_ucode_size_bytes, PAGE_SIZE);
+		}
+
+		if (adev->gfx.rlc.rlc_dram_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_DRAM];
+			info->ucode_id = AMDGPU_UCODE_ID_RLC_DRAM;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.rlc_dram_ucode_size_bytes, PAGE_SIZE);
+		}
+	}
+}
+
+static void amdgpu_gfx_rlc_init_microcode_v2_3(struct amdgpu_device *adev)
+{
+	const struct rlc_firmware_header_v2_3 *rlc_hdr;
+	struct amdgpu_firmware_info *info;
+
+	rlc_hdr = (const struct rlc_firmware_header_v2_3 *)adev->gfx.rlc_fw->data;
+	adev->gfx.rlcp_ucode_version = le32_to_cpu(rlc_hdr->rlcp_ucode_version);
+	adev->gfx.rlcp_ucode_feature_version = le32_to_cpu(rlc_hdr->rlcp_ucode_feature_version);
+	adev->gfx.rlc.rlcp_ucode_size_bytes = le32_to_cpu(rlc_hdr->rlcp_ucode_size_bytes);
+	adev->gfx.rlc.rlcp_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->rlcp_ucode_offset_bytes);
+
+	adev->gfx.rlcv_ucode_version = le32_to_cpu(rlc_hdr->rlcv_ucode_version);
+	adev->gfx.rlcv_ucode_feature_version = le32_to_cpu(rlc_hdr->rlcv_ucode_feature_version);
+	adev->gfx.rlc.rlcv_ucode_size_bytes = le32_to_cpu(rlc_hdr->rlcv_ucode_size_bytes);
+	adev->gfx.rlc.rlcv_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->rlcv_ucode_offset_bytes);
+
+	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+		if (adev->gfx.rlc.rlcp_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_P];
+			info->ucode_id = AMDGPU_UCODE_ID_RLC_P;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.rlcp_ucode_size_bytes, PAGE_SIZE);
+		}
+
+		if (adev->gfx.rlc.rlcv_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_V];
+			info->ucode_id = AMDGPU_UCODE_ID_RLC_V;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.rlcv_ucode_size_bytes, PAGE_SIZE);
+		}
+	}
+}
+
+static void amdgpu_gfx_rlc_init_microcode_v2_4(struct amdgpu_device *adev)
+{
+	const struct rlc_firmware_header_v2_4 *rlc_hdr;
+	struct amdgpu_firmware_info *info;
+
+	rlc_hdr = (const struct rlc_firmware_header_v2_4 *)adev->gfx.rlc_fw->data;
+	adev->gfx.rlc.global_tap_delays_ucode_size_bytes = le32_to_cpu(rlc_hdr->global_tap_delays_ucode_size_bytes);
+	adev->gfx.rlc.global_tap_delays_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->global_tap_delays_ucode_offset_bytes);
+	adev->gfx.rlc.se0_tap_delays_ucode_size_bytes = le32_to_cpu(rlc_hdr->se0_tap_delays_ucode_size_bytes);
+	adev->gfx.rlc.se0_tap_delays_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->se0_tap_delays_ucode_offset_bytes);
+	adev->gfx.rlc.se1_tap_delays_ucode_size_bytes = le32_to_cpu(rlc_hdr->se1_tap_delays_ucode_size_bytes);
+	adev->gfx.rlc.se1_tap_delays_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->se1_tap_delays_ucode_offset_bytes);
+	adev->gfx.rlc.se2_tap_delays_ucode_size_bytes = le32_to_cpu(rlc_hdr->se2_tap_delays_ucode_size_bytes);
+	adev->gfx.rlc.se2_tap_delays_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->se2_tap_delays_ucode_offset_bytes);
+	adev->gfx.rlc.se3_tap_delays_ucode_size_bytes = le32_to_cpu(rlc_hdr->se3_tap_delays_ucode_size_bytes);
+	adev->gfx.rlc.se3_tap_delays_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->se3_tap_delays_ucode_offset_bytes);
+
+	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+		if (adev->gfx.rlc.global_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.global_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
+
+		if (adev->gfx.rlc.se0_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE0_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_SE0_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.se0_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
+
+		if (adev->gfx.rlc.se1_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE1_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_SE1_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.se1_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
+
+		if (adev->gfx.rlc.se2_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE2_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_SE2_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.se2_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
+
+		if (adev->gfx.rlc.se3_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE3_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_SE3_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.se3_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
+	}
+}
+
+int amdgpu_gfx_rlc_init_microcode(struct amdgpu_device *adev,
+				  uint16_t version_major,
+				  uint16_t version_minor)
+{
+	int err;
+
+	if (version_major < 2) {
+		/* only support rlc_hdr v2.x and onwards */
+		dev_err(adev->dev, "unsupported rlc fw hdr\n");
+		return -EINVAL;
+	}
+
+	/* is_rlc_v2_1 is still used in APU code path */
+	if (version_major == 2 && version_minor == 1)
+		adev->gfx.rlc.is_rlc_v2_1 = true;
+
+	if (version_minor >= 0) {
+		err = amdgpu_gfx_rlc_init_microcode_v2_0(adev);
+		if (err) {
+			dev_err(adev->dev, "fail to init rlc v2_0 microcode\n");
+			return err;
+		}
+	}
+	if (version_minor >= 1)
+		amdgpu_gfx_rlc_init_microcode_v2_1(adev);
+	if (version_minor >= 2)
+		amdgpu_gfx_rlc_init_microcode_v2_2(adev);
+	if (version_minor == 3)
+		amdgpu_gfx_rlc_init_microcode_v2_3(adev);
+	if (version_minor == 4)
+		amdgpu_gfx_rlc_init_microcode_v2_4(adev);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
index 03ac36b2c2cf..23f060db9255 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
@@ -267,5 +267,7 @@ int amdgpu_gfx_rlc_init_csb(struct amdgpu_device *adev);
 int amdgpu_gfx_rlc_init_cpt(struct amdgpu_device *adev);
 void amdgpu_gfx_rlc_setup_cp_table(struct amdgpu_device *adev);
 void amdgpu_gfx_rlc_fini(struct amdgpu_device *adev);
-
+int amdgpu_gfx_rlc_init_microcode(struct amdgpu_device *adev,
+				  uint16_t version_major,
+				  uint16_t version_minor);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 3b4c19412625..134575a3893c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -637,6 +637,8 @@ struct amdgpu_ttm_tt {
 #endif
 };
 
+#define ttm_to_amdgpu_ttm_tt(ptr)	container_of(ptr, struct amdgpu_ttm_tt, ttm)
+
 #ifdef CONFIG_DRM_AMDGPU_USERPTR
 /*
  * amdgpu_ttm_tt_get_user_pages - get device accessible pages that back user
@@ -648,7 +650,7 @@ struct amdgpu_ttm_tt {
 int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages)
 {
 	struct ttm_tt *ttm = bo->tbo.ttm;
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 	unsigned long start = gtt->userptr;
 	struct vm_area_struct *vma;
 	struct mm_struct *mm;
@@ -702,7 +704,7 @@ out_unlock:
  */
 bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm)
 {
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 	bool r = false;
 
 	if (!gtt || !gtt->userptr)
@@ -751,7 +753,7 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_device *bdev,
 				     struct ttm_tt *ttm)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 	int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY);
 	enum dma_data_direction direction = write ?
 		DMA_BIDIRECTIONAL : DMA_TO_DEVICE;
@@ -788,7 +790,7 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_device *bdev,
 					struct ttm_tt *ttm)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 	int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY);
 	enum dma_data_direction direction = write ?
 		DMA_BIDIRECTIONAL : DMA_TO_DEVICE;
@@ -822,7 +824,7 @@ static void amdgpu_ttm_gart_bind(struct amdgpu_device *adev,
 {
 	struct amdgpu_bo *abo = ttm_to_amdgpu_bo(tbo);
 	struct ttm_tt *ttm = tbo->ttm;
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 
 	if (amdgpu_bo_encrypted(abo))
 		flags |= AMDGPU_PTE_TMZ;
@@ -860,7 +862,7 @@ static int amdgpu_ttm_backend_bind(struct ttm_device *bdev,
 				   struct ttm_resource *bo_mem)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
-	struct amdgpu_ttm_tt *gtt = (void*)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 	uint64_t flags;
 	int r;
 
@@ -927,7 +929,7 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
 	struct ttm_operation_ctx ctx = { false, false };
-	struct amdgpu_ttm_tt *gtt = (void *)bo->ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(bo->ttm);
 	struct ttm_placement placement;
 	struct ttm_place placements;
 	struct ttm_resource *tmp;
@@ -998,7 +1000,7 @@ static void amdgpu_ttm_backend_unbind(struct ttm_device *bdev,
 				      struct ttm_tt *ttm)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 
 	/* if the pages have userptr pinning then clear that first */
 	if (gtt->userptr) {
@@ -1025,7 +1027,7 @@ static void amdgpu_ttm_backend_unbind(struct ttm_device *bdev,
 static void amdgpu_ttm_backend_destroy(struct ttm_device *bdev,
 				       struct ttm_tt *ttm)
 {
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 
 	if (gtt->usertask)
 		put_task_struct(gtt->usertask);
@@ -1079,7 +1081,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_device *bdev,
 				  struct ttm_operation_ctx *ctx)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 	pgoff_t i;
 	int ret;
 
@@ -1113,7 +1115,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_device *bdev,
 static void amdgpu_ttm_tt_unpopulate(struct ttm_device *bdev,
 				     struct ttm_tt *ttm)
 {
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 	struct amdgpu_device *adev;
 	pgoff_t i;
 
@@ -1182,7 +1184,7 @@ int amdgpu_ttm_tt_set_userptr(struct ttm_buffer_object *bo,
 	/* Set TTM_TT_FLAG_EXTERNAL before populate but after create. */
 	bo->ttm->page_flags |= TTM_TT_FLAG_EXTERNAL;
 
-	gtt = (void *)bo->ttm;
+	gtt = ttm_to_amdgpu_ttm_tt(bo->ttm);
 	gtt->userptr = addr;
 	gtt->userflags = flags;
 
@@ -1199,7 +1201,7 @@ int amdgpu_ttm_tt_set_userptr(struct ttm_buffer_object *bo,
  */
 struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm)
 {
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 
 	if (gtt == NULL)
 		return NULL;
@@ -1218,7 +1220,7 @@ struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm)
 bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start,
 				  unsigned long end, unsigned long *userptr)
 {
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 	unsigned long size;
 
 	if (gtt == NULL || !gtt->userptr)
@@ -1241,7 +1243,7 @@ bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start,
  */
 bool amdgpu_ttm_tt_is_userptr(struct ttm_tt *ttm)
 {
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 
 	if (gtt == NULL || !gtt->userptr)
 		return false;
@@ -1254,7 +1256,7 @@ bool amdgpu_ttm_tt_is_userptr(struct ttm_tt *ttm)
  */
 bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm)
 {
-	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 
 	if (gtt == NULL)
 		return false;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
index ebed3f5226db..59edf32f775e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
@@ -260,8 +260,12 @@ struct rlc_firmware_header_v2_2 {
 /* version_major=2, version_minor=3 */
 struct rlc_firmware_header_v2_3 {
     struct rlc_firmware_header_v2_2 v2_2;
+    uint32_t rlcp_ucode_version;
+    uint32_t rlcp_ucode_feature_version;
     uint32_t rlcp_ucode_size_bytes;
     uint32_t rlcp_ucode_offset_bytes;
+    uint32_t rlcv_ucode_version;
+    uint32_t rlcv_ucode_feature_version;
     uint32_t rlcv_ucode_size_bytes;
     uint32_t rlcv_ucode_offset_bytes;
 };
@@ -390,6 +394,7 @@ union amdgpu_firmware_header {
 	struct rlc_firmware_header_v2_1 rlc_v2_1;
 	struct rlc_firmware_header_v2_2 rlc_v2_2;
 	struct rlc_firmware_header_v2_3 rlc_v2_3;
+	struct rlc_firmware_header_v2_4 rlc_v2_4;
 	struct sdma_firmware_header_v1_0 sdma;
 	struct sdma_firmware_header_v1_1 sdma_v1_1;
 	struct sdma_firmware_header_v2_0 sdma_v2_0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index f36e4f08db6d..0b52af415b28 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -191,7 +191,7 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 		fw_name = FIRMWARE_VCN4_0_2;
 		if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) &&
 			(adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
-			adev->vcn.indirect_sram = false;
+			adev->vcn.indirect_sram = true;
 		break;
 	case IP_VERSION(4, 0, 4):
 		fw_name = FIRMWARE_VCN4_0_4;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
index 108e8e8a1a36..576849e95296 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
@@ -496,8 +496,7 @@ static int amdgpu_vkms_sw_init(void *handle)
 	adev_to_drm(adev)->mode_config.max_height = YRES_MAX;
 
 	adev_to_drm(adev)->mode_config.preferred_depth = 24;
-	/* disable prefer shadow for now due to hibernation issues */
-	adev_to_drm(adev)->mode_config.prefer_shadow = 0;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 
 	adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 59cac347baa3..690fd4f639f1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2484,8 +2484,7 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
 		/* Intentionally setting invalid PTE flag
 		 * combination to force a no-retry-fault
 		 */
-		flags = AMDGPU_PTE_EXECUTABLE | AMDGPU_PDE_PTE |
-			AMDGPU_PTE_TF;
+		flags = AMDGPU_PTE_SNOOPED | AMDGPU_PTE_PRT;
 		value = 0;
 	} else if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_NEVER) {
 		/* Redirect the access to the dummy page */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 1b108d03e785..f2aebbf3fbe3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -742,7 +742,7 @@ int amdgpu_xgmi_remove_device(struct amdgpu_device *adev)
 		amdgpu_put_xgmi_hive(hive);
 	}
 
-	return psp_xgmi_terminate(&adev->psp);
+	return 0;
 }
 
 static int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block)
diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c
index 33a8a7365aef..f0e235f98afb 100644
--- a/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/athub_v3_0.c
@@ -28,13 +28,44 @@
 #include "navi10_enum.h"
 #include "soc15_common.h"
 
+#define regATHUB_MISC_CNTL_V3_0_1			0x00d7
+#define regATHUB_MISC_CNTL_V3_0_1_BASE_IDX		0
+
+
+static uint32_t athub_v3_0_get_cg_cntl(struct amdgpu_device *adev)
+{
+	uint32_t data;
+
+	switch (adev->ip_versions[ATHUB_HWIP][0]) {
+	case IP_VERSION(3, 0, 1):
+		data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL_V3_0_1);
+		break;
+	default:
+		data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL);
+		break;
+	}
+	return data;
+}
+
+static void athub_v3_0_set_cg_cntl(struct amdgpu_device *adev, uint32_t data)
+{
+	switch (adev->ip_versions[ATHUB_HWIP][0]) {
+	case IP_VERSION(3, 0, 1):
+		WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL_V3_0_1, data);
+		break;
+	default:
+		WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL, data);
+		break;
+	}
+}
+
 static void
 athub_v3_0_update_medium_grain_clock_gating(struct amdgpu_device *adev,
 					    bool enable)
 {
 	uint32_t def, data;
 
-	def = data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL);
+	def = data = athub_v3_0_get_cg_cntl(adev);
 
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_ATHUB_MGCG))
 		data |= ATHUB_MISC_CNTL__CG_ENABLE_MASK;
@@ -42,7 +73,7 @@ athub_v3_0_update_medium_grain_clock_gating(struct amdgpu_device *adev,
 		data &= ~ATHUB_MISC_CNTL__CG_ENABLE_MASK;
 
 	if (def != data)
-		WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL, data);
+		athub_v3_0_set_cg_cntl(adev, data);
 }
 
 static void
@@ -51,7 +82,7 @@ athub_v3_0_update_medium_grain_light_sleep(struct amdgpu_device *adev,
 {
 	uint32_t def, data;
 
-	def = data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL);
+	def = data = athub_v3_0_get_cg_cntl(adev);
 
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_ATHUB_LS))
 		data |= ATHUB_MISC_CNTL__CG_MEM_LS_ENABLE_MASK;
@@ -59,7 +90,7 @@ athub_v3_0_update_medium_grain_light_sleep(struct amdgpu_device *adev,
 		data &= ~ATHUB_MISC_CNTL__CG_MEM_LS_ENABLE_MASK;
 
 	if (def != data)
-		WREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL, data);
+		athub_v3_0_set_cg_cntl(adev, data);
 }
 
 int athub_v3_0_set_clockgating(struct amdgpu_device *adev,
@@ -70,6 +101,7 @@ int athub_v3_0_set_clockgating(struct amdgpu_device *adev,
 
 	switch (adev->ip_versions[ATHUB_HWIP][0]) {
 	case IP_VERSION(3, 0, 0):
+	case IP_VERSION(3, 0, 1):
 	case IP_VERSION(3, 0, 2):
 		athub_v3_0_update_medium_grain_clock_gating(adev,
 				state == AMD_CG_STATE_GATE);
@@ -88,7 +120,7 @@ void athub_v3_0_get_clockgating(struct amdgpu_device *adev, u64 *flags)
 	int data;
 
 	/* AMD_CG_SUPPORT_ATHUB_MGCG */
-	data = RREG32_SOC15(ATHUB, 0, regATHUB_MISC_CNTL);
+	data = athub_v3_0_get_cg_cntl(adev);
 	if (data & ATHUB_MISC_CNTL__CG_ENABLE_MASK)
 		*flags |= AMD_CG_SUPPORT_ATHUB_MGCG;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 9c964cd3b5d4..288fce7dc0ed 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -2796,8 +2796,7 @@ static int dce_v10_0_sw_init(void *handle)
 	adev_to_drm(adev)->mode_config.max_height = 16384;
 
 	adev_to_drm(adev)->mode_config.preferred_depth = 24;
-	/* disable prefer shadow for now due to hibernation issues */
-	adev_to_drm(adev)->mode_config.prefer_shadow = 0;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 
 	adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index e0ad9f27dc3f..cbe5250b31cb 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -2914,8 +2914,7 @@ static int dce_v11_0_sw_init(void *handle)
 	adev_to_drm(adev)->mode_config.max_height = 16384;
 
 	adev_to_drm(adev)->mode_config.preferred_depth = 24;
-	/* disable prefer shadow for now due to hibernation issues */
-	adev_to_drm(adev)->mode_config.prefer_shadow = 0;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 
 	adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
index 77f5e998a120..b1c44fab074f 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
@@ -2673,8 +2673,7 @@ static int dce_v6_0_sw_init(void *handle)
 	adev_to_drm(adev)->mode_config.max_width = 16384;
 	adev_to_drm(adev)->mode_config.max_height = 16384;
 	adev_to_drm(adev)->mode_config.preferred_depth = 24;
-	/* disable prefer shadow for now due to hibernation issues */
-	adev_to_drm(adev)->mode_config.prefer_shadow = 0;
+	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 	adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true;
 	adev_to_drm(adev)->mode_config.fb_base = adev->gmc.aper_base;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index 802e5c753271..a22b45c92792 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -2693,8 +2693,11 @@ static int dce_v8_0_sw_init(void *handle)
 	adev_to_drm(adev)->mode_config.max_height = 16384;
 
 	adev_to_drm(adev)->mode_config.preferred_depth = 24;
-	/* disable prefer shadow for now due to hibernation issues */
-	adev_to_drm(adev)->mode_config.prefer_shadow = 0;
+	if (adev->asic_type == CHIP_HAWAII)
+		/* disable prefer shadow for now due to hibernation issues */
+		adev_to_drm(adev)->mode_config.prefer_shadow = 0;
+	else
+		adev_to_drm(adev)->mode_config.prefer_shadow = 1;
 
 	adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index fafbad3cf08d..a3cd5c1e8529 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4274,35 +4274,45 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev)
 
 		}
 
-		info = &adev->firmware.ucode[AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS];
-		info->ucode_id = AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS;
-		info->fw = adev->gfx.rlc_fw;
-		adev->firmware.fw_size +=
-			ALIGN(adev->gfx.rlc.global_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		if (adev->gfx.rlc.global_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_GLOBAL_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.global_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
 
-		info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE0_TAP_DELAYS];
-		info->ucode_id = AMDGPU_UCODE_ID_SE0_TAP_DELAYS;
-		info->fw = adev->gfx.rlc_fw;
-		adev->firmware.fw_size +=
-			ALIGN(adev->gfx.rlc.se0_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		if (adev->gfx.rlc.se0_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE0_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_SE0_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.se0_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
 
-		info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE1_TAP_DELAYS];
-		info->ucode_id = AMDGPU_UCODE_ID_SE1_TAP_DELAYS;
-		info->fw = adev->gfx.rlc_fw;
-		adev->firmware.fw_size +=
-			ALIGN(adev->gfx.rlc.se1_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		if (adev->gfx.rlc.se1_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE1_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_SE1_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.se1_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
 
-		info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE2_TAP_DELAYS];
-		info->ucode_id = AMDGPU_UCODE_ID_SE2_TAP_DELAYS;
-		info->fw = adev->gfx.rlc_fw;
-		adev->firmware.fw_size +=
-			ALIGN(adev->gfx.rlc.se2_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		if (adev->gfx.rlc.se2_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE2_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_SE2_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.se2_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
 
-		info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE3_TAP_DELAYS];
-		info->ucode_id = AMDGPU_UCODE_ID_SE3_TAP_DELAYS;
-		info->fw = adev->gfx.rlc_fw;
-		adev->firmware.fw_size +=
-			ALIGN(adev->gfx.rlc.se3_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		if (adev->gfx.rlc.se3_tap_delays_ucode_size_bytes) {
+			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SE3_TAP_DELAYS];
+			info->ucode_id = AMDGPU_UCODE_ID_SE3_TAP_DELAYS;
+			info->fw = adev->gfx.rlc_fw;
+			adev->firmware.fw_size +=
+				ALIGN(adev->gfx.rlc.se3_tap_delays_ucode_size_bytes, PAGE_SIZE);
+		}
 
 		info = &adev->firmware.ucode[AMDGPU_UCODE_ID_CP_MEC1];
 		info->ucode_id = AMDGPU_UCODE_ID_CP_MEC1;
@@ -4846,7 +4856,7 @@ static int gfx_v10_0_sw_init(void *handle)
 	case IP_VERSION(10, 3, 3):
 	case IP_VERSION(10, 3, 7):
 		adev->gfx.me.num_me = 1;
-		adev->gfx.me.num_pipe_per_me = 2;
+		adev->gfx.me.num_pipe_per_me = 1;
 		adev->gfx.me.num_queue_per_pipe = 1;
 		adev->gfx.mec.num_mec = 2;
 		adev->gfx.mec.num_pipe_per_mec = 4;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 6fd71cb10e54..daf8ba8235cd 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -53,6 +53,7 @@
 #define GFX11_MEC_HPD_SIZE	2048
 
 #define RLCG_UCODE_LOADING_START_ADDRESS	0x00002000L
+#define RLC_PG_DELAY_3_DEFAULT_GC_11_0_1	0x1388
 
 #define regCGTT_WD_CLK_CTRL		0x5086
 #define regCGTT_WD_CLK_CTRL_BASE_IDX	1
@@ -130,6 +131,8 @@ static void gfx_v11_0_ring_invalidate_tlbs(struct amdgpu_ring *ring,
 					   bool all_hub, uint8_t dst_sel);
 static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev);
 static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev);
+static void gfx_v11_0_update_perf_clk(struct amdgpu_device *adev,
+				      bool enable);
 
 static void gfx11_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask)
 {
@@ -471,49 +474,6 @@ static void gfx_v11_0_free_microcode(struct amdgpu_device *adev)
 	kfree(adev->gfx.rlc.register_list_format);
 }
 
-static void gfx_v11_0_init_rlc_ext_microcode(struct amdgpu_device *adev)
-{
-	const struct rlc_firmware_header_v2_1 *rlc_hdr;
-
-	rlc_hdr = (const struct rlc_firmware_header_v2_1 *)adev->gfx.rlc_fw->data;
-	adev->gfx.rlc_srlc_fw_version = le32_to_cpu(rlc_hdr->save_restore_list_cntl_ucode_ver);
-	adev->gfx.rlc_srlc_feature_version = le32_to_cpu(rlc_hdr->save_restore_list_cntl_feature_ver);
-	adev->gfx.rlc.save_restore_list_cntl_size_bytes = le32_to_cpu(rlc_hdr->save_restore_list_cntl_size_bytes);
-	adev->gfx.rlc.save_restore_list_cntl = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->save_restore_list_cntl_offset_bytes);
-	adev->gfx.rlc_srlg_fw_version = le32_to_cpu(rlc_hdr->save_restore_list_gpm_ucode_ver);
-	adev->gfx.rlc_srlg_feature_version = le32_to_cpu(rlc_hdr->save_restore_list_gpm_feature_ver);
-	adev->gfx.rlc.save_restore_list_gpm_size_bytes = le32_to_cpu(rlc_hdr->save_restore_list_gpm_size_bytes);
-	adev->gfx.rlc.save_restore_list_gpm = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->save_restore_list_gpm_offset_bytes);
-	adev->gfx.rlc_srls_fw_version = le32_to_cpu(rlc_hdr->save_restore_list_srm_ucode_ver);
-	adev->gfx.rlc_srls_feature_version = le32_to_cpu(rlc_hdr->save_restore_list_srm_feature_ver);
-	adev->gfx.rlc.save_restore_list_srm_size_bytes = le32_to_cpu(rlc_hdr->save_restore_list_srm_size_bytes);
-	adev->gfx.rlc.save_restore_list_srm = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->save_restore_list_srm_offset_bytes);
-	adev->gfx.rlc.reg_list_format_direct_reg_list_length =
-			le32_to_cpu(rlc_hdr->reg_list_format_direct_reg_list_length);
-}
-
-static void gfx_v11_0_init_rlc_iram_dram_microcode(struct amdgpu_device *adev)
-{
-	const struct rlc_firmware_header_v2_2 *rlc_hdr;
-
-	rlc_hdr = (const struct rlc_firmware_header_v2_2 *)adev->gfx.rlc_fw->data;
-	adev->gfx.rlc.rlc_iram_ucode_size_bytes = le32_to_cpu(rlc_hdr->rlc_iram_ucode_size_bytes);
-	adev->gfx.rlc.rlc_iram_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->rlc_iram_ucode_offset_bytes);
-	adev->gfx.rlc.rlc_dram_ucode_size_bytes = le32_to_cpu(rlc_hdr->rlc_dram_ucode_size_bytes);
-	adev->gfx.rlc.rlc_dram_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->rlc_dram_ucode_offset_bytes);
-}
-
-static void gfx_v11_0_init_rlcp_rlcv_microcode(struct amdgpu_device *adev)
-{
-	const struct rlc_firmware_header_v2_3 *rlc_hdr;
-
-	rlc_hdr = (const struct rlc_firmware_header_v2_3 *)adev->gfx.rlc_fw->data;
-	adev->gfx.rlc.rlcp_ucode_size_bytes = le32_to_cpu(rlc_hdr->rlcp_ucode_size_bytes);
-	adev->gfx.rlc.rlcp_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->rlcp_ucode_offset_bytes);
-	adev->gfx.rlc.rlcv_ucode_size_bytes = le32_to_cpu(rlc_hdr->rlcv_ucode_size_bytes);
-	adev->gfx.rlc.rlcv_ucode = (u8 *)rlc_hdr + le32_to_cpu(rlc_hdr->rlcv_ucode_offset_bytes);
-}
-
 static int gfx_v11_0_init_microcode(struct amdgpu_device *adev)
 {
 	char fw_name[40];
@@ -524,8 +484,6 @@ static int gfx_v11_0_init_microcode(struct amdgpu_device *adev)
 	const struct gfx_firmware_header_v1_0 *cp_hdr;
 	const struct gfx_firmware_header_v2_0 *cp_hdr_v2_0;
 	const struct rlc_firmware_header_v2_0 *rlc_hdr;
-	unsigned int *tmp = NULL;
-	unsigned int i = 0;
 	uint16_t version_major;
 	uint16_t version_minor;
 
@@ -580,58 +538,14 @@ static int gfx_v11_0_init_microcode(struct amdgpu_device *adev)
 		if (err)
 			goto out;
 		err = amdgpu_ucode_validate(adev->gfx.rlc_fw);
+		if (err)
+			goto out;
 		rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data;
 		version_major = le16_to_cpu(rlc_hdr->header.header_version_major);
 		version_minor = le16_to_cpu(rlc_hdr->header.header_version_minor);
-
-		adev->gfx.rlc_fw_version = le32_to_cpu(rlc_hdr->header.ucode_version);
-		adev->gfx.rlc_feature_version = le32_to_cpu(rlc_hdr->ucode_feature_version);
-		adev->gfx.rlc.save_and_restore_offset =
-			le32_to_cpu(rlc_hdr->save_and_restore_offset);
-		adev->gfx.rlc.clear_state_descriptor_offset =
-			le32_to_cpu(rlc_hdr->clear_state_descriptor_offset);
-		adev->gfx.rlc.avail_scratch_ram_locations =
-			le32_to_cpu(rlc_hdr->avail_scratch_ram_locations);
-		adev->gfx.rlc.reg_restore_list_size =
-			le32_to_cpu(rlc_hdr->reg_restore_list_size);
-		adev->gfx.rlc.reg_list_format_start =
-			le32_to_cpu(rlc_hdr->reg_list_format_start);
-		adev->gfx.rlc.reg_list_format_separate_start =
-			le32_to_cpu(rlc_hdr->reg_list_format_separate_start);
-		adev->gfx.rlc.starting_offsets_start =
-			le32_to_cpu(rlc_hdr->starting_offsets_start);
-		adev->gfx.rlc.reg_list_format_size_bytes =
-			le32_to_cpu(rlc_hdr->reg_list_format_size_bytes);
-		adev->gfx.rlc.reg_list_size_bytes =
-			le32_to_cpu(rlc_hdr->reg_list_size_bytes);
-		adev->gfx.rlc.register_list_format =
-			kmalloc(adev->gfx.rlc.reg_list_format_size_bytes +
-					adev->gfx.rlc.reg_list_size_bytes, GFP_KERNEL);
-		if (!adev->gfx.rlc.register_list_format) {
-			err = -ENOMEM;
+		err = amdgpu_gfx_rlc_init_microcode(adev, version_major, version_minor);
+		if (err)
 			goto out;
-		}
-
-		tmp = (unsigned int *)((uintptr_t)rlc_hdr +
-							   le32_to_cpu(rlc_hdr->reg_list_format_array_offset_bytes));
-		for (i = 0 ; i < (rlc_hdr->reg_list_format_size_bytes >> 2); i++)
-			adev->gfx.rlc.register_list_format[i] =	le32_to_cpu(tmp[i]);
-
-		adev->gfx.rlc.register_restore = adev->gfx.rlc.register_list_format + i;
-
-		tmp = (unsigned int *)((uintptr_t)rlc_hdr +
-							   le32_to_cpu(rlc_hdr->reg_list_array_offset_bytes));
-		for (i = 0 ; i < (rlc_hdr->reg_list_size_bytes >> 2); i++)
-			adev->gfx.rlc.register_restore[i] = le32_to_cpu(tmp[i]);
-
-		if (version_major == 2) {
-			if (version_minor >= 1)
-				gfx_v11_0_init_rlc_ext_microcode(adev);
-			if (version_minor >= 2)
-				gfx_v11_0_init_rlc_iram_dram_microcode(adev);
-			if (version_minor == 3)
-				gfx_v11_0_init_rlcp_rlcv_microcode(adev);
-		}
 	}
 
 	snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec.bin", ucode_prefix);
@@ -766,60 +680,6 @@ static int gfx_v11_0_init_microcode(struct amdgpu_device *adev)
 			adev->firmware.fw_size +=
 				ALIGN(le32_to_cpu(cp_hdr->jt_size) * 4, PAGE_SIZE);
 		}
-
-		info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_G];
-		info->ucode_id = AMDGPU_UCODE_ID_RLC_G;
-		info->fw = adev->gfx.rlc_fw;
-		if (info->fw) {
-			header = (const struct common_firmware_header *)info->fw->data;
-			adev->firmware.fw_size +=
-				ALIGN(le32_to_cpu(header->ucode_size_bytes), PAGE_SIZE);
-		}
-		if (adev->gfx.rlc.save_restore_list_gpm_size_bytes &&
-		    adev->gfx.rlc.save_restore_list_srm_size_bytes) {
-			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM];
-			info->ucode_id = AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM;
-			info->fw = adev->gfx.rlc_fw;
-			adev->firmware.fw_size +=
-				ALIGN(adev->gfx.rlc.save_restore_list_gpm_size_bytes, PAGE_SIZE);
-
-			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM];
-			info->ucode_id = AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM;
-			info->fw = adev->gfx.rlc_fw;
-			adev->firmware.fw_size +=
-				ALIGN(adev->gfx.rlc.save_restore_list_srm_size_bytes, PAGE_SIZE);
-		}
-
-		if (adev->gfx.rlc.rlc_iram_ucode_size_bytes &&
-		    adev->gfx.rlc.rlc_dram_ucode_size_bytes) {
-			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_IRAM];
-			info->ucode_id = AMDGPU_UCODE_ID_RLC_IRAM;
-			info->fw = adev->gfx.rlc_fw;
-			adev->firmware.fw_size +=
-				ALIGN(adev->gfx.rlc.rlc_iram_ucode_size_bytes, PAGE_SIZE);
-
-			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_DRAM];
-			info->ucode_id = AMDGPU_UCODE_ID_RLC_DRAM;
-			info->fw = adev->gfx.rlc_fw;
-			adev->firmware.fw_size +=
-				ALIGN(adev->gfx.rlc.rlc_dram_ucode_size_bytes, PAGE_SIZE);
-		}
-
-		if (adev->gfx.rlc.rlcp_ucode_size_bytes) {
-			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_P];
-			info->ucode_id = AMDGPU_UCODE_ID_RLC_P;
-			info->fw = adev->gfx.rlc_fw;
-			adev->firmware.fw_size +=
-				ALIGN(adev->gfx.rlc.rlcp_ucode_size_bytes, PAGE_SIZE);
-		}
-
-		if (adev->gfx.rlc.rlcv_ucode_size_bytes) {
-			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_RLC_V];
-			info->ucode_id = AMDGPU_UCODE_ID_RLC_V;
-			info->fw = adev->gfx.rlc_fw;
-			adev->firmware.fw_size +=
-				ALIGN(adev->gfx.rlc.rlcv_ucode_size_bytes, PAGE_SIZE);
-		}
 	}
 
 out:
@@ -1138,6 +998,7 @@ static const struct amdgpu_gfx_funcs gfx_v11_0_gfx_funcs = {
 	.read_wave_vgprs = &gfx_v11_0_read_wave_vgprs,
 	.select_me_pipe_q = &gfx_v11_0_select_me_pipe_q,
 	.init_spm_golden = &gfx_v11_0_init_spm_golden_registers,
+	.update_perfmon_mgcg = &gfx_v11_0_update_perf_clk,
 };
 
 static int gfx_v11_0_gpu_early_init(struct amdgpu_device *adev)
@@ -5181,9 +5042,12 @@ static void gfx_v11_0_update_coarse_grain_clock_gating(struct amdgpu_device *ade
 		data = REG_SET_FIELD(data, SDMA0_RLC_CGCG_CTRL, CGCG_INT_ENABLE, 1);
 		WREG32_SOC15(GC, 0, regSDMA0_RLC_CGCG_CTRL, data);
 
-		data = RREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL);
-		data = REG_SET_FIELD(data, SDMA1_RLC_CGCG_CTRL, CGCG_INT_ENABLE, 1);
-		WREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL, data);
+		/* Some ASICs only have one SDMA instance, not need to configure SDMA1 */
+		if (adev->sdma.num_instances > 1) {
+			data = RREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL);
+			data = REG_SET_FIELD(data, SDMA1_RLC_CGCG_CTRL, CGCG_INT_ENABLE, 1);
+			WREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL, data);
+		}
 	} else {
 		/* Program RLC_CGCG_CGLS_CTRL */
 		def = data = RREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL);
@@ -5212,9 +5076,12 @@ static void gfx_v11_0_update_coarse_grain_clock_gating(struct amdgpu_device *ade
 		data &= ~SDMA0_RLC_CGCG_CTRL__CGCG_INT_ENABLE_MASK;
 		WREG32_SOC15(GC, 0, regSDMA0_RLC_CGCG_CTRL, data);
 
-		data = RREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL);
-		data &= ~SDMA1_RLC_CGCG_CTRL__CGCG_INT_ENABLE_MASK;
-		WREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL, data);
+		/* Some ASICs only have one SDMA instance, not need to configure SDMA1 */
+		if (adev->sdma.num_instances > 1) {
+			data = RREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL);
+			data &= ~SDMA1_RLC_CGCG_CTRL__CGCG_INT_ENABLE_MASK;
+			WREG32_SOC15(GC, 0, regSDMA1_RLC_CGCG_CTRL, data);
+		}
 	}
 }
 
@@ -5250,6 +5117,8 @@ static void gfx_v11_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid)
 {
 	u32 reg, data;
 
+	amdgpu_gfx_off_ctrl(adev, false);
+
 	reg = SOC15_REG_OFFSET(GC, 0, regRLC_SPM_MC_CNTL);
 	if (amdgpu_sriov_is_pp_one_vf(adev))
 		data = RREG32_NO_KIQ(reg);
@@ -5263,6 +5132,8 @@ static void gfx_v11_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid)
 		WREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL, data);
 	else
 		WREG32_SOC15(GC, 0, regRLC_SPM_MC_CNTL, data);
+
+	amdgpu_gfx_off_ctrl(adev, true);
 }
 
 static const struct amdgpu_rlc_funcs gfx_v11_0_rlc_funcs = {
@@ -5279,6 +5150,38 @@ static const struct amdgpu_rlc_funcs gfx_v11_0_rlc_funcs = {
 	.update_spm_vmid = gfx_v11_0_update_spm_vmid,
 };
 
+static void gfx_v11_cntl_power_gating(struct amdgpu_device *adev, bool enable)
+{
+	u32 data = RREG32_SOC15(GC, 0, regRLC_PG_CNTL);
+
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG))
+		data |= RLC_PG_CNTL__GFX_POWER_GATING_ENABLE_MASK;
+	else
+		data &= ~RLC_PG_CNTL__GFX_POWER_GATING_ENABLE_MASK;
+
+	WREG32_SOC15(GC, 0, regRLC_PG_CNTL, data);
+
+	// Program RLC_PG_DELAY3 for CGPG hysteresis
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG)) {
+		switch (adev->ip_versions[GC_HWIP][0]) {
+		case IP_VERSION(11, 0, 1):
+			WREG32_SOC15(GC, 0, regRLC_PG_DELAY_3, RLC_PG_DELAY_3_DEFAULT_GC_11_0_1);
+			break;
+		default:
+			break;
+		}
+	}
+}
+
+static void gfx_v11_cntl_pg(struct amdgpu_device *adev, bool enable)
+{
+	amdgpu_gfx_rlc_enter_safe_mode(adev);
+
+	gfx_v11_cntl_power_gating(adev, enable);
+
+	amdgpu_gfx_rlc_exit_safe_mode(adev);
+}
+
 static int gfx_v11_0_set_powergating_state(void *handle,
 					   enum amd_powergating_state state)
 {
@@ -5293,6 +5196,10 @@ static int gfx_v11_0_set_powergating_state(void *handle,
 	case IP_VERSION(11, 0, 2):
 		amdgpu_gfx_off_ctrl(adev, enable);
 		break;
+	case IP_VERSION(11, 0, 1):
+		gfx_v11_cntl_pg(adev, enable);
+		amdgpu_gfx_off_ctrl(adev, enable);
+		break;
 	default:
 		break;
 	}
@@ -5310,6 +5217,7 @@ static int gfx_v11_0_set_clockgating_state(void *handle,
 
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(11, 0, 0):
+	case IP_VERSION(11, 0, 1):
 	case IP_VERSION(11, 0, 2):
 	        gfx_v11_0_update_gfx_clock_gating(adev,
 	                        state ==  AMD_CG_STATE_GATE);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index c6e0f9313a7f..037af8352677 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -2587,7 +2587,8 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev)
 
 	gfx_v9_0_tiling_mode_table_init(adev);
 
-	gfx_v9_0_setup_rb(adev);
+	if (adev->gfx.num_gfx_rings)
+		gfx_v9_0_setup_rb(adev);
 	gfx_v9_0_get_cu_info(adev, &adev->gfx.cu_info);
 	adev->gfx.config.db_debug2 = RREG32_SOC15(GC, 0, mmDB_DEBUG2);
 
@@ -5596,7 +5597,7 @@ static void gfx_v9_0_ring_emit_patch_cond_exec(struct amdgpu_ring *ring, unsigne
 	BUG_ON(offset > ring->buf_mask);
 	BUG_ON(ring->ring[offset] != 0x55aa55aa);
 
-	cur = (ring->wptr & ring->buf_mask) - 1;
+	cur = (ring->wptr - 1) & ring->buf_mask;
 	if (likely(cur > offset))
 		ring->ring[offset] = cur - offset;
 	else
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 9ae8cdaa033e..f513e2c2e964 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -419,6 +419,7 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	uint32_t seq;
 	uint16_t queried_pasid;
 	bool ret;
+	u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : adev->usec_timeout;
 	struct amdgpu_ring *ring = &adev->gfx.kiq.ring;
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
 
@@ -437,7 +438,7 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 
 		amdgpu_ring_commit(ring);
 		spin_unlock(&adev->gfx.kiq.ring_lock);
-		r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout);
+		r = amdgpu_fence_wait_polling(ring, seq, usec_timeout);
 		if (r < 1) {
 			dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r);
 			return -ETIME;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 22761a3bb818..67ca16a8027c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -896,6 +896,7 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	uint32_t seq;
 	uint16_t queried_pasid;
 	bool ret;
+	u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : adev->usec_timeout;
 	struct amdgpu_ring *ring = &adev->gfx.kiq.ring;
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
 
@@ -935,7 +936,7 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 
 		amdgpu_ring_commit(ring);
 		spin_unlock(&adev->gfx.kiq.ring_lock);
-		r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout);
+		r = amdgpu_fence_wait_polling(ring, seq, usec_timeout);
 		if (r < 1) {
 			dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r);
 			up_read(&adev->reset_domain->sem);
@@ -1102,10 +1103,13 @@ static void gmc_v9_0_get_vm_pde(struct amdgpu_device *adev, int level,
 			*flags |= AMDGPU_PDE_BFS(0x9);
 
 	} else if (level == AMDGPU_VM_PDB0) {
-		if (*flags & AMDGPU_PDE_PTE)
+		if (*flags & AMDGPU_PDE_PTE) {
 			*flags &= ~AMDGPU_PDE_PTE;
-		else
+			if (!(*flags & AMDGPU_PTE_VALID))
+				*addr |= 1 << PAGE_SHIFT;
+		} else {
 			*flags |= AMDGPU_PTE_TF;
+		}
 	}
 }
 
@@ -1624,12 +1628,15 @@ static int gmc_v9_0_sw_init(void *handle)
 			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 47);
 		else
 			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
+		if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2))
+			adev->gmc.translate_further = adev->vm_manager.num_level > 1;
 		break;
 	case IP_VERSION(9, 4, 1):
 		adev->num_vmhubs = 3;
 
 		/* Keep the vm size same with Vega20 */
 		amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
+		adev->gmc.translate_further = adev->vm_manager.num_level > 1;
 		break;
 	default:
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c
index 39a696cd45b5..29c3484ae1f1 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c
@@ -40,6 +40,156 @@ static void hdp_v5_2_flush_hdp(struct amdgpu_device *adev,
 			0);
 }
 
+static void hdp_v5_2_update_mem_power_gating(struct amdgpu_device *adev,
+					     bool enable)
+{
+	uint32_t hdp_clk_cntl;
+	uint32_t hdp_mem_pwr_cntl;
+
+	if (!(adev->cg_flags & (AMD_CG_SUPPORT_HDP_LS |
+				AMD_CG_SUPPORT_HDP_DS |
+				AMD_CG_SUPPORT_HDP_SD)))
+		return;
+
+	hdp_clk_cntl = RREG32_SOC15(HDP, 0, regHDP_CLK_CNTL);
+	hdp_mem_pwr_cntl = RREG32_SOC15(HDP, 0, regHDP_MEM_POWER_CTRL);
+
+	/* Before doing clock/power mode switch, forced on MEM clock */
+	hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL,
+				     ATOMIC_MEM_CLK_SOFT_OVERRIDE, 1);
+	hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL,
+				     RC_MEM_CLK_SOFT_OVERRIDE, 1);
+	WREG32_SOC15(HDP, 0, regHDP_CLK_CNTL, hdp_clk_cntl);
+
+	/* disable clock and power gating before any changing */
+	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+					 ATOMIC_MEM_POWER_CTRL_EN, 0);
+	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+					 ATOMIC_MEM_POWER_LS_EN, 0);
+	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+					 ATOMIC_MEM_POWER_DS_EN, 0);
+	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+					 ATOMIC_MEM_POWER_SD_EN, 0);
+	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+					 RC_MEM_POWER_CTRL_EN, 0);
+	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+					 RC_MEM_POWER_LS_EN, 0);
+	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+					 RC_MEM_POWER_DS_EN, 0);
+	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+					 RC_MEM_POWER_SD_EN, 0);
+	WREG32_SOC15(HDP, 0, regHDP_MEM_POWER_CTRL, hdp_mem_pwr_cntl);
+
+	/* Already disabled above. The actions below are for "enabled" only */
+	if (enable) {
+		/* only one clock gating mode (LS/DS/SD) can be enabled */
+		if (adev->cg_flags & AMD_CG_SUPPORT_HDP_SD) {
+			hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
+							 HDP_MEM_POWER_CTRL,
+							 ATOMIC_MEM_POWER_SD_EN, 1);
+			hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
+							 HDP_MEM_POWER_CTRL,
+							 RC_MEM_POWER_SD_EN, 1);
+		} else if (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS) {
+			hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
+							 HDP_MEM_POWER_CTRL,
+							 ATOMIC_MEM_POWER_LS_EN, 1);
+			hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
+							 HDP_MEM_POWER_CTRL,
+							 RC_MEM_POWER_LS_EN, 1);
+		} else if (adev->cg_flags & AMD_CG_SUPPORT_HDP_DS) {
+			hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
+							 HDP_MEM_POWER_CTRL,
+							 ATOMIC_MEM_POWER_DS_EN, 1);
+			hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
+							 HDP_MEM_POWER_CTRL,
+							 RC_MEM_POWER_DS_EN, 1);
+		}
+
+		/* confirmed that ATOMIC/RC_MEM_POWER_CTRL_EN have to be set for SRAM LS/DS/SD */
+		if (adev->cg_flags & (AMD_CG_SUPPORT_HDP_LS | AMD_CG_SUPPORT_HDP_DS |
+				      AMD_CG_SUPPORT_HDP_SD)) {
+			hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+							 ATOMIC_MEM_POWER_CTRL_EN, 1);
+			hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
+							 RC_MEM_POWER_CTRL_EN, 1);
+			WREG32_SOC15(HDP, 0, regHDP_MEM_POWER_CTRL, hdp_mem_pwr_cntl);
+		}
+	}
+
+	/* disable MEM clock override after clock/power mode changing */
+	hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL,
+				     ATOMIC_MEM_CLK_SOFT_OVERRIDE, 0);
+	hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL,
+				     RC_MEM_CLK_SOFT_OVERRIDE, 0);
+	WREG32_SOC15(HDP, 0, regHDP_CLK_CNTL, hdp_clk_cntl);
+}
+
+static void hdp_v5_2_update_medium_grain_clock_gating(struct amdgpu_device *adev,
+						      bool enable)
+{
+	uint32_t hdp_clk_cntl;
+
+	if (!(adev->cg_flags & AMD_CG_SUPPORT_HDP_MGCG))
+		return;
+
+	hdp_clk_cntl = RREG32_SOC15(HDP, 0, regHDP_CLK_CNTL);
+
+	if (enable) {
+		hdp_clk_cntl &=
+			~(uint32_t)
+			(HDP_CLK_CNTL__ATOMIC_MEM_CLK_SOFT_OVERRIDE_MASK |
+			 HDP_CLK_CNTL__RC_MEM_CLK_SOFT_OVERRIDE_MASK |
+			 HDP_CLK_CNTL__DBUS_CLK_SOFT_OVERRIDE_MASK |
+			 HDP_CLK_CNTL__DYN_CLK_SOFT_OVERRIDE_MASK |
+			 HDP_CLK_CNTL__XDP_REG_CLK_SOFT_OVERRIDE_MASK |
+			 HDP_CLK_CNTL__HDP_REG_CLK_SOFT_OVERRIDE_MASK);
+	} else {
+		hdp_clk_cntl |= HDP_CLK_CNTL__ATOMIC_MEM_CLK_SOFT_OVERRIDE_MASK |
+			HDP_CLK_CNTL__RC_MEM_CLK_SOFT_OVERRIDE_MASK |
+			HDP_CLK_CNTL__DBUS_CLK_SOFT_OVERRIDE_MASK |
+			HDP_CLK_CNTL__DYN_CLK_SOFT_OVERRIDE_MASK |
+			HDP_CLK_CNTL__XDP_REG_CLK_SOFT_OVERRIDE_MASK |
+			HDP_CLK_CNTL__HDP_REG_CLK_SOFT_OVERRIDE_MASK;
+	}
+
+	WREG32_SOC15(HDP, 0, regHDP_CLK_CNTL, hdp_clk_cntl);
+}
+
+static void hdp_v5_2_get_clockgating_state(struct amdgpu_device *adev,
+					   u64 *flags)
+{
+	uint32_t tmp;
+
+	/* AMD_CG_SUPPORT_HDP_MGCG */
+	tmp = RREG32_SOC15(HDP, 0, regHDP_CLK_CNTL);
+	if (!(tmp & (HDP_CLK_CNTL__ATOMIC_MEM_CLK_SOFT_OVERRIDE_MASK |
+		     HDP_CLK_CNTL__RC_MEM_CLK_SOFT_OVERRIDE_MASK |
+		     HDP_CLK_CNTL__DBUS_CLK_SOFT_OVERRIDE_MASK |
+		     HDP_CLK_CNTL__DYN_CLK_SOFT_OVERRIDE_MASK |
+		     HDP_CLK_CNTL__XDP_REG_CLK_SOFT_OVERRIDE_MASK |
+		     HDP_CLK_CNTL__HDP_REG_CLK_SOFT_OVERRIDE_MASK)))
+		*flags |= AMD_CG_SUPPORT_HDP_MGCG;
+
+	/* AMD_CG_SUPPORT_HDP_LS/DS/SD */
+	tmp = RREG32_SOC15(HDP, 0, regHDP_MEM_POWER_CTRL);
+	if (tmp & HDP_MEM_POWER_CTRL__ATOMIC_MEM_POWER_LS_EN_MASK)
+		*flags |= AMD_CG_SUPPORT_HDP_LS;
+	else if (tmp & HDP_MEM_POWER_CTRL__ATOMIC_MEM_POWER_DS_EN_MASK)
+		*flags |= AMD_CG_SUPPORT_HDP_DS;
+	else if (tmp & HDP_MEM_POWER_CTRL__ATOMIC_MEM_POWER_SD_EN_MASK)
+		*flags |= AMD_CG_SUPPORT_HDP_SD;
+}
+
+static void hdp_v5_2_update_clock_gating(struct amdgpu_device *adev,
+					      bool enable)
+{
+	hdp_v5_2_update_mem_power_gating(adev, enable);
+	hdp_v5_2_update_medium_grain_clock_gating(adev, enable);
+}
+
 const struct amdgpu_hdp_funcs hdp_v5_2_funcs = {
 	.flush_hdp = hdp_v5_2_flush_hdp,
+	.update_clock_gating = hdp_v5_2_update_clock_gating,
+	.get_clock_gating_state = hdp_v5_2_get_clockgating_state,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
index 92dc60a9d209..085e613f3646 100644
--- a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
@@ -727,6 +727,7 @@ static const struct amd_ip_funcs ih_v6_0_ip_funcs = {
 static const struct amdgpu_ih_funcs ih_v6_0_funcs = {
 	.get_wptr = ih_v6_0_get_wptr,
 	.decode_iv = amdgpu_ih_decode_iv_helper,
+	.decode_iv_ts = amdgpu_ih_decode_iv_ts_helper,
 	.set_rptr = ih_v6_0_set_rptr
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 120ea294abef..f92744b8d79d 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -183,6 +183,11 @@ static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes,
 	mes_add_queue_pkt.trap_handler_addr = input->tba_addr;
 	mes_add_queue_pkt.tma_addr = input->tma_addr;
 	mes_add_queue_pkt.is_kfd_process = input->is_kfd_process;
+	mes_add_queue_pkt.trap_en = 1;
+
+	/* For KFD, gds_size is re-used for queue size (needed in MES for AQL queues) */
+	mes_add_queue_pkt.is_aql_queue = input->is_aql_queue;
+	mes_add_queue_pkt.gds_size = input->queue_size;
 
 	return mes_v11_0_submit_pkt_and_poll_completion(mes,
 			&mes_add_queue_pkt, sizeof(mes_add_queue_pkt),
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
index 3f44a099c52a..3e51e773f92b 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
@@ -176,6 +176,7 @@ static void mmhub_v1_0_init_cache_regs(struct amdgpu_device *adev)
 	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
 	WREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL2, tmp);
 
+	tmp = mmVM_L2_CNTL3_DEFAULT;
 	if (adev->gmc.translate_further) {
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c
index cac72ced94c8..e8058edc1d10 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c
@@ -518,18 +518,41 @@ static u64 mmhub_v3_0_1_get_mc_fb_offset(struct amdgpu_device *adev)
 static void mmhub_v3_0_1_update_medium_grain_clock_gating(struct amdgpu_device *adev,
 							  bool enable)
 {
-	//TODO
+	uint32_t def, data;
+
+	def = data = RREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG);
+
+	if (enable)
+		data |= MM_ATC_L2_MISC_CG__ENABLE_MASK;
+	else
+		data &= ~MM_ATC_L2_MISC_CG__ENABLE_MASK;
+
+	if (def != data)
+		WREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG, data);
 }
 
 static void mmhub_v3_0_1_update_medium_grain_light_sleep(struct amdgpu_device *adev,
 							 bool enable)
 {
-	//TODO
+	uint32_t def, data;
+
+	def = data = RREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG);
+
+	if (enable)
+		data |= MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK;
+	else
+		data &= ~MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK;
+
+	if (def != data)
+		WREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG, data);
 }
 
 static int mmhub_v3_0_1_set_clockgating(struct amdgpu_device *adev,
 					enum amd_clockgating_state state)
 {
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
 	mmhub_v3_0_1_update_medium_grain_clock_gating(adev,
 			state == AMD_CG_STATE_GATE);
 	mmhub_v3_0_1_update_medium_grain_light_sleep(adev,
@@ -539,7 +562,20 @@ static int mmhub_v3_0_1_set_clockgating(struct amdgpu_device *adev,
 
 static void mmhub_v3_0_1_get_clockgating(struct amdgpu_device *adev, u64 *flags)
 {
-	//TODO
+	int data;
+
+	if (amdgpu_sriov_vf(adev))
+		*flags = 0;
+
+	data = RREG32_SOC15(MMHUB, 0, regMM_ATC_L2_MISC_CG);
+
+	/* AMD_CG_SUPPORT_MC_MGCG */
+	if (data & MM_ATC_L2_MISC_CG__ENABLE_MASK)
+		*flags |= AMD_CG_SUPPORT_MC_MGCG;
+
+	/* AMD_CG_SUPPORT_MC_LS */
+	if (data & MM_ATC_L2_MISC_CG__MEM_LS_ENABLE_MASK)
+		*flags |= AMD_CG_SUPPORT_MC_LS;
 }
 
 const struct amdgpu_mmhub_funcs mmhub_v3_0_1_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c
index 6e0145b2b408..445cb06b9d26 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c
@@ -295,9 +295,17 @@ static void mmhub_v9_4_disable_identity_aperture(struct amdgpu_device *adev,
 static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	unsigned int num_level, block_size;
 	uint32_t tmp;
 	int i;
 
+	num_level = adev->vm_manager.num_level;
+	block_size = adev->vm_manager.block_size;
+	if (adev->gmc.translate_further)
+		num_level -= 1;
+	else
+		block_size -= 9;
+
 	for (i = 0; i <= 14; i++) {
 		tmp = RREG32_SOC15_OFFSET(MMHUB, 0, mmVML2VC0_VM_CONTEXT1_CNTL,
 				hubid * MMHUB_INSTANCE_REGISTER_OFFSET + i);
@@ -305,7 +313,7 @@ static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid)
 				    ENABLE_CONTEXT, 1);
 		tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL,
 				    PAGE_TABLE_DEPTH,
-				    adev->vm_manager.num_level);
+				    num_level);
 		tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL,
 				    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
 		tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL,
@@ -323,7 +331,7 @@ static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid)
 				    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
 		tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL,
 				    PAGE_TABLE_BLOCK_SIZE,
-				    adev->vm_manager.block_size - 9);
+				    block_size);
 		/* Send no-retry XNACK on fault to suppress VM fault storm. */
 		tmp = REG_SET_FIELD(tmp, VML2VC0_VM_CONTEXT1_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c
index 4b5396d3e60f..eec13cb5bf75 100644
--- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c
@@ -409,9 +409,11 @@ static u32 navi10_ih_get_wptr(struct amdgpu_device *adev,
 	u32 wptr, tmp;
 	struct amdgpu_ih_regs *ih_regs;
 
-	if (ih == &adev->irq.ih) {
+	if (ih == &adev->irq.ih || ih == &adev->irq.ih_soft) {
 		/* Only ring0 supports writeback. On other rings fall back
 		 * to register-based code with overflow checking below.
+		 * ih_soft ring doesn't have any backing hardware registers,
+		 * update wptr and return.
 		 */
 		wptr = le32_to_cpu(*ih->wptr_cpu);
 
@@ -483,6 +485,9 @@ static void navi10_ih_set_rptr(struct amdgpu_device *adev,
 {
 	struct amdgpu_ih_regs *ih_regs;
 
+	if (ih == &adev->irq.ih_soft)
+		return;
+
 	if (ih->use_doorbell) {
 		/* XXX check if swapping is necessary on BE */
 		*ih->rptr_cpu = ih->rptr;
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c b/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c
index b465baa26762..aa761ff3a5fa 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c
@@ -380,6 +380,7 @@ static void nbio_v2_3_enable_aspm(struct amdgpu_device *adev,
 		WREG32_PCIE(smnPCIE_LC_CNTL, data);
 }
 
+#ifdef CONFIG_PCIEASPM
 static void nbio_v2_3_program_ltr(struct amdgpu_device *adev)
 {
 	uint32_t def, data;
@@ -401,9 +402,11 @@ static void nbio_v2_3_program_ltr(struct amdgpu_device *adev)
 	if (def != data)
 		WREG32_PCIE(smnBIF_CFG_DEV0_EPF0_DEVICE_CNTL2, data);
 }
+#endif
 
 static void nbio_v2_3_program_aspm(struct amdgpu_device *adev)
 {
+#ifdef CONFIG_PCIEASPM
 	uint32_t def, data;
 
 	def = data = RREG32_PCIE(smnPCIE_LC_CNTL);
@@ -459,7 +462,10 @@ static void nbio_v2_3_program_aspm(struct amdgpu_device *adev)
 	if (def != data)
 		WREG32_PCIE(smnPCIE_LC_CNTL6, data);
 
-	nbio_v2_3_program_ltr(adev);
+	/* Don't bother about LTR if LTR is not enabled
+	 * in the path */
+	if (adev->pdev->ltr_path)
+		nbio_v2_3_program_ltr(adev);
 
 	def = data = RREG32_SOC15(NBIO, 0, mmRCC_BIF_STRAP3);
 	data |= 0x5DE0 << RCC_BIF_STRAP3__STRAP_VLINK_ASPM_IDLE_TIMER__SHIFT;
@@ -483,6 +489,7 @@ static void nbio_v2_3_program_aspm(struct amdgpu_device *adev)
 	data &= ~PCIE_LC_CNTL3__LC_DSC_DONT_ENTER_L23_AFTER_PME_ACK_MASK;
 	if (def != data)
 		WREG32_PCIE(smnPCIE_LC_CNTL3, data);
+#endif
 }
 
 static void nbio_v2_3_apply_lc_spc_mode_wa(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c
index f7f6ddebd3e4..37615a77287b 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c
@@ -282,6 +282,7 @@ static void nbio_v6_1_init_registers(struct amdgpu_device *adev)
 			mmBIF_BX_DEV0_EPF0_VF0_HDP_MEM_COHERENCY_FLUSH_CNTL) << 2;
 }
 
+#ifdef CONFIG_PCIEASPM
 static void nbio_v6_1_program_ltr(struct amdgpu_device *adev)
 {
 	uint32_t def, data;
@@ -303,9 +304,11 @@ static void nbio_v6_1_program_ltr(struct amdgpu_device *adev)
 	if (def != data)
 		WREG32_PCIE(smnBIF_CFG_DEV0_EPF0_DEVICE_CNTL2, data);
 }
+#endif
 
 static void nbio_v6_1_program_aspm(struct amdgpu_device *adev)
 {
+#ifdef CONFIG_PCIEASPM
 	uint32_t def, data;
 
 	def = data = RREG32_PCIE(smnPCIE_LC_CNTL);
@@ -361,7 +364,10 @@ static void nbio_v6_1_program_aspm(struct amdgpu_device *adev)
 	if (def != data)
 		WREG32_PCIE(smnPCIE_LC_CNTL6, data);
 
-	nbio_v6_1_program_ltr(adev);
+	/* Don't bother about LTR if LTR is not enabled
+	 * in the path */
+	if (adev->pdev->ltr_path)
+		nbio_v6_1_program_ltr(adev);
 
 	def = data = RREG32_PCIE(smnRCC_BIF_STRAP3);
 	data |= 0x5DE0 << RCC_BIF_STRAP3__STRAP_VLINK_ASPM_IDLE_TIMER__SHIFT;
@@ -385,6 +391,7 @@ static void nbio_v6_1_program_aspm(struct amdgpu_device *adev)
 	data &= ~PCIE_LC_CNTL3__LC_DSC_DONT_ENTER_L23_AFTER_PME_ACK_MASK;
 	if (def != data)
 		WREG32_PCIE(smnPCIE_LC_CNTL3, data);
+#endif
 }
 
 const struct amdgpu_nbio_funcs nbio_v6_1_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
index 11848d1e238b..19455a725939 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
@@ -673,6 +673,7 @@ struct amdgpu_nbio_ras nbio_v7_4_ras = {
 };
 
 
+#ifdef CONFIG_PCIEASPM
 static void nbio_v7_4_program_ltr(struct amdgpu_device *adev)
 {
 	uint32_t def, data;
@@ -694,9 +695,11 @@ static void nbio_v7_4_program_ltr(struct amdgpu_device *adev)
 	if (def != data)
 		WREG32_PCIE(smnBIF_CFG_DEV0_EPF0_DEVICE_CNTL2, data);
 }
+#endif
 
 static void nbio_v7_4_program_aspm(struct amdgpu_device *adev)
 {
+#ifdef CONFIG_PCIEASPM
 	uint32_t def, data;
 
 	if (adev->ip_versions[NBIO_HWIP][0] == IP_VERSION(7, 4, 4))
@@ -755,7 +758,10 @@ static void nbio_v7_4_program_aspm(struct amdgpu_device *adev)
 	if (def != data)
 		WREG32_PCIE(smnPCIE_LC_CNTL6, data);
 
-	nbio_v7_4_program_ltr(adev);
+	/* Don't bother about LTR if LTR is not enabled
+	 * in the path */
+	if (adev->pdev->ltr_path)
+		nbio_v7_4_program_ltr(adev);
 
 	def = data = RREG32_PCIE(smnRCC_BIF_STRAP3);
 	data |= 0x5DE0 << RCC_BIF_STRAP3__STRAP_VLINK_ASPM_IDLE_TIMER__SHIFT;
@@ -779,6 +785,7 @@ static void nbio_v7_4_program_aspm(struct amdgpu_device *adev)
 	data &= ~PCIE_LC_CNTL3__LC_DSC_DONT_ENTER_L23_AFTER_PME_ACK_MASK;
 	if (def != data)
 		WREG32_PCIE(smnPCIE_LC_CNTL3, data);
+#endif
 }
 
 const struct amdgpu_nbio_funcs nbio_v7_4_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_7.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_7.c
index 01e8288d09a8..def89379b51a 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_7.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_7.c
@@ -28,6 +28,14 @@
 #include "nbio/nbio_7_7_0_sh_mask.h"
 #include <uapi/linux/kfd_ioctl.h>
 
+static void nbio_v7_7_remap_hdp_registers(struct amdgpu_device *adev)
+{
+	WREG32_SOC15(NBIO, 0, regBIF_BX0_REMAP_HDP_MEM_FLUSH_CNTL,
+		     adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL);
+	WREG32_SOC15(NBIO, 0, regBIF_BX0_REMAP_HDP_REG_FLUSH_CNTL,
+		     adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_REG_FLUSH_CNTL);
+}
+
 static u32 nbio_v7_7_get_rev_id(struct amdgpu_device *adev)
 {
 	u32 tmp;
@@ -68,12 +76,6 @@ static void nbio_v7_7_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 		doorbell_range = REG_SET_FIELD(doorbell_range,
 					       GDC0_BIF_CSDMA_DOORBELL_RANGE,
 					       SIZE, doorbell_size);
-		doorbell_range = REG_SET_FIELD(doorbell_range,
-					       GDC0_BIF_SDMA0_DOORBELL_RANGE,
-					       OFFSET, doorbell_index);
-		doorbell_range = REG_SET_FIELD(doorbell_range,
-					       GDC0_BIF_SDMA0_DOORBELL_RANGE,
-					       SIZE, doorbell_size);
 	} else {
 		doorbell_range = REG_SET_FIELD(doorbell_range,
 					       GDC0_BIF_SDMA0_DOORBELL_RANGE,
@@ -247,6 +249,81 @@ static void nbio_v7_7_init_registers(struct amdgpu_device *adev)
 
 }
 
+static void nbio_v7_7_update_medium_grain_clock_gating(struct amdgpu_device *adev,
+						       bool enable)
+{
+	uint32_t def, data;
+
+	if (enable && !(adev->cg_flags & AMD_CG_SUPPORT_BIF_MGCG))
+		return;
+
+	def = data = RREG32_SOC15(NBIO, 0, regBIF0_CPM_CONTROL);
+	if (enable) {
+		data |= (BIF0_CPM_CONTROL__LCLK_DYN_GATE_ENABLE_MASK |
+			 BIF0_CPM_CONTROL__TXCLK_DYN_GATE_ENABLE_MASK |
+			 BIF0_CPM_CONTROL__TXCLK_LCNT_GATE_ENABLE_MASK |
+			 BIF0_CPM_CONTROL__TXCLK_REGS_GATE_ENABLE_MASK |
+			 BIF0_CPM_CONTROL__TXCLK_PRBS_GATE_ENABLE_MASK |
+			 BIF0_CPM_CONTROL__REFCLK_REGS_GATE_ENABLE_MASK);
+	} else {
+		data &= ~(BIF0_CPM_CONTROL__LCLK_DYN_GATE_ENABLE_MASK |
+			  BIF0_CPM_CONTROL__TXCLK_DYN_GATE_ENABLE_MASK |
+			  BIF0_CPM_CONTROL__TXCLK_LCNT_GATE_ENABLE_MASK |
+			  BIF0_CPM_CONTROL__TXCLK_REGS_GATE_ENABLE_MASK |
+			  BIF0_CPM_CONTROL__TXCLK_PRBS_GATE_ENABLE_MASK |
+			  BIF0_CPM_CONTROL__REFCLK_REGS_GATE_ENABLE_MASK);
+	}
+
+	if (def != data)
+		WREG32_SOC15(NBIO, 0, regBIF0_CPM_CONTROL, data);
+}
+
+static void nbio_v7_7_update_medium_grain_light_sleep(struct amdgpu_device *adev,
+						      bool enable)
+{
+	uint32_t def, data;
+
+	if (enable && !(adev->cg_flags & AMD_CG_SUPPORT_BIF_LS))
+		return;
+
+	def = data = RREG32_SOC15(NBIO, 0, regBIF0_PCIE_CNTL2);
+	if (enable)
+		data |= BIF0_PCIE_CNTL2__SLV_MEM_LS_EN_MASK;
+	else
+		data &= ~BIF0_PCIE_CNTL2__SLV_MEM_LS_EN_MASK;
+
+	if (def != data)
+		WREG32_SOC15(NBIO, 0, regBIF0_PCIE_CNTL2, data);
+
+	def = data = RREG32_SOC15(NBIO, 0, regBIF0_PCIE_TX_POWER_CTRL_1);
+	if (enable) {
+		data |= (BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_LS_EN_MASK |
+			BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_LS_EN_MASK);
+	} else {
+		data &= ~(BIF0_PCIE_TX_POWER_CTRL_1__MST_MEM_LS_EN_MASK |
+			BIF0_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_LS_EN_MASK);
+	}
+
+	if (def != data)
+		WREG32_SOC15(NBIO, 0, regBIF0_PCIE_TX_POWER_CTRL_1, data);
+}
+
+static void nbio_v7_7_get_clockgating_state(struct amdgpu_device *adev,
+					    u64 *flags)
+{
+	uint32_t data;
+
+	/* AMD_CG_SUPPORT_BIF_MGCG */
+	data = RREG32_SOC15(NBIO, 0, regBIF0_CPM_CONTROL);
+	if (data & BIF0_CPM_CONTROL__LCLK_DYN_GATE_ENABLE_MASK)
+		*flags |= AMD_CG_SUPPORT_BIF_MGCG;
+
+	/* AMD_CG_SUPPORT_BIF_LS */
+	data = RREG32_SOC15(NBIO, 0, regBIF0_PCIE_CNTL2);
+	if (data & BIF0_PCIE_CNTL2__SLV_MEM_LS_EN_MASK)
+		*flags |= AMD_CG_SUPPORT_BIF_LS;
+}
+
 const struct amdgpu_nbio_funcs nbio_v7_7_funcs = {
 	.get_hdp_flush_req_offset = nbio_v7_7_get_hdp_flush_req_offset,
 	.get_hdp_flush_done_offset = nbio_v7_7_get_hdp_flush_done_offset,
@@ -262,6 +339,10 @@ const struct amdgpu_nbio_funcs nbio_v7_7_funcs = {
 	.enable_doorbell_aperture = nbio_v7_7_enable_doorbell_aperture,
 	.enable_doorbell_selfring_aperture = nbio_v7_7_enable_doorbell_selfring_aperture,
 	.ih_doorbell_range = nbio_v7_7_ih_doorbell_range,
+	.update_medium_grain_clock_gating = nbio_v7_7_update_medium_grain_clock_gating,
+	.update_medium_grain_light_sleep = nbio_v7_7_update_medium_grain_light_sleep,
+	.get_clockgating_state = nbio_v7_7_get_clockgating_state,
 	.ih_control = nbio_v7_7_ih_control,
 	.init_registers = nbio_v7_7_init_registers,
+	.remap_hdp_registers = nbio_v7_7_remap_hdp_registers,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c
index a2588200ea58..0b2ac418e4ac 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c
@@ -101,6 +101,16 @@ static int psp_v12_0_init_microcode(struct psp_context *psp)
 		adev->psp.dtm_context.context.bin_desc.start_addr =
 			(uint8_t *)adev->psp.hdcp_context.context.bin_desc.start_addr +
 			le32_to_cpu(ta_hdr->dtm.offset_bytes);
+
+		if (adev->apu_flags & AMD_APU_IS_RENOIR) {
+			adev->psp.securedisplay_context.context.bin_desc.fw_version =
+				le32_to_cpu(ta_hdr->securedisplay.fw_version);
+			adev->psp.securedisplay_context.context.bin_desc.size_bytes =
+				le32_to_cpu(ta_hdr->securedisplay.size_bytes);
+			adev->psp.securedisplay_context.context.bin_desc.start_addr =
+				(uint8_t *)adev->psp.hdcp_context.context.bin_desc.start_addr +
+				le32_to_cpu(ta_hdr->securedisplay.offset_bytes);
+		}
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
index 726a5bba40b2..a75a286e1ecf 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
@@ -20,7 +20,6 @@
  * OTHER DEALINGS IN THE SOFTWARE.
  *
  */
-#include <linux/dev_printk.h>
 #include <drm/drm_drv.h>
 #include <linux/vmalloc.h>
 #include "amdgpu.h"
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 65181efba50e..56424f75dd2c 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -1504,6 +1504,11 @@ static int sdma_v4_0_start(struct amdgpu_device *adev)
 		WREG32_SDMA(i, mmSDMA0_CNTL, temp);
 
 		if (!amdgpu_sriov_vf(adev)) {
+			ring = &adev->sdma.instance[i].ring;
+			adev->nbio.funcs->sdma_doorbell_range(adev, i,
+				ring->use_doorbell, ring->doorbell_index,
+				adev->doorbell_index.sdma_doorbell_range);
+
 			/* unhalt engine */
 			temp = RREG32_SDMA(i, mmSDMA0_F32_CNTL);
 			temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index fde6154f2009..183024d7c184 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1211,25 +1211,6 @@ static int soc15_common_sw_fini(void *handle)
 	return 0;
 }
 
-static void soc15_doorbell_range_init(struct amdgpu_device *adev)
-{
-	int i;
-	struct amdgpu_ring *ring;
-
-	/* sdma/ih doorbell range are programed by hypervisor */
-	if (!amdgpu_sriov_vf(adev)) {
-		for (i = 0; i < adev->sdma.num_instances; i++) {
-			ring = &adev->sdma.instance[i].ring;
-			adev->nbio.funcs->sdma_doorbell_range(adev, i,
-				ring->use_doorbell, ring->doorbell_index,
-				adev->doorbell_index.sdma_doorbell_range);
-		}
-
-		adev->nbio.funcs->ih_doorbell_range(adev, adev->irq.ih.use_doorbell,
-						adev->irq.ih.doorbell_index);
-	}
-}
-
 static int soc15_common_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -1249,12 +1230,6 @@ static int soc15_common_hw_init(void *handle)
 
 	/* enable the doorbell aperture */
 	soc15_enable_doorbell_aperture(adev, true);
-	/* HW doorbell routing policy: doorbell writing not
-	 * in SDMA/IH/MM/ACV range will be routed to CP. So
-	 * we need to init SDMA/IH/MM/ACV doorbell range prior
-	 * to CP ip block init and ring test.
-	 */
-	soc15_doorbell_range_init(adev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c
index 52816de5e17b..6e564b549b9f 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc21.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc21.c
@@ -421,6 +421,7 @@ static bool soc21_need_full_reset(struct amdgpu_device *adev)
 {
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(11, 0, 0):
+		return amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC);
 	case IP_VERSION(11, 0, 2):
 		return false;
 	default:
@@ -494,6 +495,20 @@ static void soc21_pre_asic_init(struct amdgpu_device *adev)
 {
 }
 
+static int soc21_update_umd_stable_pstate(struct amdgpu_device *adev,
+					  bool enter)
+{
+	if (enter)
+		amdgpu_gfx_rlc_enter_safe_mode(adev);
+	else
+		amdgpu_gfx_rlc_exit_safe_mode(adev);
+
+	if (adev->gfx.funcs->update_perfmon_mgcg)
+		adev->gfx.funcs->update_perfmon_mgcg(adev, !enter);
+
+	return 0;
+}
+
 static const struct amdgpu_asic_funcs soc21_asic_funcs =
 {
 	.read_disabled_bios = &soc21_read_disabled_bios,
@@ -513,6 +528,7 @@ static const struct amdgpu_asic_funcs soc21_asic_funcs =
 	.supports_baco = &amdgpu_dpm_is_baco_supported,
 	.pre_asic_init = &soc21_pre_asic_init,
 	.query_video_codecs = &soc21_query_video_codecs,
+	.update_umd_stable_pstate = &soc21_update_umd_stable_pstate,
 };
 
 static int soc21_common_early_init(void *handle)
@@ -546,8 +562,10 @@ static int soc21_common_early_init(void *handle)
 	case IP_VERSION(11, 0, 0):
 		adev->cg_flags = AMD_CG_SUPPORT_GFX_CGCG |
 			AMD_CG_SUPPORT_GFX_CGLS |
+#if 0
 			AMD_CG_SUPPORT_GFX_3D_CGCG |
 			AMD_CG_SUPPORT_GFX_3D_CGLS |
+#endif
 			AMD_CG_SUPPORT_GFX_MGCG |
 			AMD_CG_SUPPORT_REPEATER_FGCG |
 			AMD_CG_SUPPORT_GFX_FGCG |
@@ -575,7 +593,9 @@ static int soc21_common_early_init(void *handle)
 			AMD_CG_SUPPORT_VCN_MGCG |
 			AMD_CG_SUPPORT_JPEG_MGCG |
 			AMD_CG_SUPPORT_ATHUB_MGCG |
-			AMD_CG_SUPPORT_ATHUB_LS;
+			AMD_CG_SUPPORT_ATHUB_LS |
+			AMD_CG_SUPPORT_IH_CG |
+			AMD_CG_SUPPORT_HDP_SD;
 		adev->pg_flags =
 			AMD_PG_SUPPORT_VCN |
 			AMD_PG_SUPPORT_VCN_DPG |
@@ -586,9 +606,26 @@ static int soc21_common_early_init(void *handle)
 		break;
 	case IP_VERSION(11, 0, 1):
 		adev->cg_flags =
+			AMD_CG_SUPPORT_GFX_CGCG |
+			AMD_CG_SUPPORT_GFX_CGLS |
+			AMD_CG_SUPPORT_GFX_MGCG |
+			AMD_CG_SUPPORT_GFX_FGCG |
+			AMD_CG_SUPPORT_REPEATER_FGCG |
+			AMD_CG_SUPPORT_GFX_PERF_CLK |
+			AMD_CG_SUPPORT_MC_MGCG |
+			AMD_CG_SUPPORT_MC_LS |
+			AMD_CG_SUPPORT_HDP_MGCG |
+			AMD_CG_SUPPORT_HDP_LS |
+			AMD_CG_SUPPORT_ATHUB_MGCG |
+			AMD_CG_SUPPORT_ATHUB_LS |
+			AMD_CG_SUPPORT_IH_CG |
+			AMD_CG_SUPPORT_BIF_MGCG |
+			AMD_CG_SUPPORT_BIF_LS |
 			AMD_CG_SUPPORT_VCN_MGCG |
 			AMD_CG_SUPPORT_JPEG_MGCG;
 		adev->pg_flags =
+			AMD_PG_SUPPORT_GFX_PG |
+			AMD_PG_SUPPORT_VCN_DPG |
 			AMD_PG_SUPPORT_JPEG;
 		adev->external_rev_id = adev->rev_id + 0x1;
 		break;
@@ -683,6 +720,8 @@ static int soc21_common_set_clockgating_state(void *handle,
 
 	switch (adev->ip_versions[NBIO_HWIP][0]) {
 	case IP_VERSION(4, 3, 0):
+	case IP_VERSION(4, 3, 1):
+	case IP_VERSION(7, 7, 0):
 		adev->nbio.funcs->update_medium_grain_clock_gating(adev,
 				state == AMD_CG_STATE_GATE);
 		adev->nbio.funcs->update_medium_grain_light_sleep(adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index ca14c3ef742e..fb2d74f30448 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -1115,7 +1115,7 @@ static int vcn_v4_0_start(struct amdgpu_device *adev)
  *
  * Stop VCN block with dpg mode
  */
-static int vcn_v4_0_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx)
+static void vcn_v4_0_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t tmp;
 
@@ -1133,7 +1133,6 @@ static int vcn_v4_0_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx)
 	/* disable dynamic power gating mode */
 	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 0,
 		~UVD_POWER_STATUS__UVD_PG_MODE_MASK);
-	return 0;
 }
 
 /**
@@ -1154,7 +1153,7 @@ static int vcn_v4_0_stop(struct amdgpu_device *adev)
 		fw_shared->sq.queue_mode |= FW_QUEUE_DPG_HOLD_OFF;
 
 		if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
-			r = vcn_v4_0_stop_dpg_mode(adev, i);
+			vcn_v4_0_stop_dpg_mode(adev, i);
 			continue;
 		}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
index cdd599a08125..1e83db0c5438 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
@@ -289,6 +289,10 @@ static int vega10_ih_irq_init(struct amdgpu_device *adev)
 		}
 	}
 
+	if (!amdgpu_sriov_vf(adev))
+		adev->nbio.funcs->ih_doorbell_range(adev, adev->irq.ih.use_doorbell,
+						    adev->irq.ih.doorbell_index);
+
 	pci_set_master(adev->pdev);
 
 	/* enable interrupts */
@@ -334,9 +338,11 @@ static u32 vega10_ih_get_wptr(struct amdgpu_device *adev,
 	u32 wptr, tmp;
 	struct amdgpu_ih_regs *ih_regs;
 
-	if (ih == &adev->irq.ih) {
+	if (ih == &adev->irq.ih || ih == &adev->irq.ih_soft) {
 		/* Only ring0 supports writeback. On other rings fall back
 		 * to register-based code with overflow checking below.
+		 * ih_soft ring doesn't have any backing hardware registers,
+		 * update wptr and return.
 		 */
 		wptr = le32_to_cpu(*ih->wptr_cpu);
 
@@ -409,6 +415,9 @@ static void vega10_ih_set_rptr(struct amdgpu_device *adev,
 {
 	struct amdgpu_ih_regs *ih_regs;
 
+	if (ih == &adev->irq.ih_soft)
+		return;
+
 	if (ih->use_doorbell) {
 		/* XXX check if swapping is necessary on BE */
 		*ih->rptr_cpu = ih->rptr;
diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
index 3b4eb8285943..59dfca093155 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
@@ -340,6 +340,10 @@ static int vega20_ih_irq_init(struct amdgpu_device *adev)
 		}
 	}
 
+	if (!amdgpu_sriov_vf(adev))
+		adev->nbio.funcs->ih_doorbell_range(adev, adev->irq.ih.use_doorbell,
+						    adev->irq.ih.doorbell_index);
+
 	pci_set_master(adev->pdev);
 
 	/* enable interrupts */
@@ -385,9 +389,11 @@ static u32 vega20_ih_get_wptr(struct amdgpu_device *adev,
 	u32 wptr, tmp;
 	struct amdgpu_ih_regs *ih_regs;
 
-	if (ih == &adev->irq.ih) {
+	if (ih == &adev->irq.ih || ih == &adev->irq.ih_soft) {
 		/* Only ring0 supports writeback. On other rings fall back
 		 * to register-based code with overflow checking below.
+		 * ih_soft ring doesn't have any backing hardware registers,
+		 * update wptr and return.
 		 */
 		wptr = le32_to_cpu(*ih->wptr_cpu);
 
@@ -461,6 +467,9 @@ static void vega20_ih_set_rptr(struct amdgpu_device *adev,
 {
 	struct amdgpu_ih_regs *ih_regs;
 
+	if (ih == &adev->irq.ih_soft)
+		return;
+
 	if (ih->use_doorbell) {
 		/* XXX check if swapping is necessary on BE */
 		*ih->rptr_cpu = ih->rptr;