Merge drm/drm-next into drm-misc-next

Daniel needs a few commits from drm-next.

Signed-off-by: Maxime Ripard <maxime@cerno.tech>

52 files changed, 939 insertions, 382 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 2f9da6d182f3..97dd9596d17a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -969,6 +969,7 @@ struct amdgpu_device {
 	int				pstate;
 	/* enable runtime pm on the device */
 	bool                            runpm;
+	bool                            in_runpm;
 
 	bool                            pm_sysfs_en;
 	bool                            ucode_sysfs_en;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 8609287620ea..bc2e72a66db9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -126,7 +126,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 		/* this is going to have a few of the MSBs set that we need to
 		 * clear
 		 */
-		bitmap_complement(gpu_resources.queue_bitmap,
+		bitmap_complement(gpu_resources.cp_queue_bitmap,
 				  adev->gfx.mec.queue_bitmap,
 				  KGD_MAX_QUEUES);
 
@@ -137,7 +137,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 				* adev->gfx.mec.num_pipe_per_mec
 				* adev->gfx.mec.num_queue_per_pipe;
 		for (i = last_valid_bit; i < KGD_MAX_QUEUES; ++i)
-			clear_bit(i, gpu_resources.queue_bitmap);
+			clear_bit(i, gpu_resources.cp_queue_bitmap);
 
 		amdgpu_doorbell_get_kfd_info(adev,
 				&gpu_resources.doorbell_physical_address,
@@ -178,18 +178,18 @@ void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
 		kgd2kfd_interrupt(adev->kfd.dev, ih_ring_entry);
 }
 
-void amdgpu_amdkfd_suspend(struct amdgpu_device *adev)
+void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm)
 {
 	if (adev->kfd.dev)
-		kgd2kfd_suspend(adev->kfd.dev);
+		kgd2kfd_suspend(adev->kfd.dev, run_pm);
 }
 
-int amdgpu_amdkfd_resume(struct amdgpu_device *adev)
+int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm)
 {
 	int r = 0;
 
 	if (adev->kfd.dev)
-		r = kgd2kfd_resume(adev->kfd.dev);
+		r = kgd2kfd_resume(adev->kfd.dev, run_pm);
 
 	return r;
 }
@@ -402,7 +402,7 @@ void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd,
 
 	if (amdgpu_sriov_vf(adev))
 		mem_info->mem_clk_max = adev->clock.default_mclk / 100;
-	else if (adev->powerplay.pp_funcs) {
+	else if (adev->pm.dpm_enabled) {
 		if (amdgpu_emu_mode == 1)
 			mem_info->mem_clk_max = 0;
 		else
@@ -427,7 +427,7 @@ uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
 	/* the sclk is in quantas of 10kHz */
 	if (amdgpu_sriov_vf(adev))
 		return adev->clock.default_sclk / 100;
-	else if (adev->powerplay.pp_funcs)
+	else if (adev->pm.dpm_enabled)
 		return amdgpu_dpm_get_sclk(adev, false) / 100;
 	else
 		return 100;
@@ -525,6 +525,14 @@ uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd)
 
 	return adev->gmc.xgmi.hive_id;
 }
+
+uint64_t amdgpu_amdkfd_get_unique_id(struct kgd_dev *kgd)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+
+	return adev->unique_id;
+}
+
 uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct kgd_dev *dst, struct kgd_dev *src)
 {
 	struct amdgpu_device *peer_adev = (struct amdgpu_device *)src;
@@ -647,13 +655,9 @@ int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct kgd_dev *kgd, uint16_t vmid)
 int amdgpu_amdkfd_flush_gpu_tlb_pasid(struct kgd_dev *kgd, uint16_t pasid)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
-	uint32_t flush_type = 0;
+	const uint32_t flush_type = 0;
 	bool all_hub = false;
 
-	if (adev->gmc.xgmi.num_physical_nodes &&
-		adev->asic_type == CHIP_VEGA20)
-		flush_type = 2;
-
 	if (adev->family == AMDGPU_FAMILY_AI)
 		all_hub = true;
 
@@ -677,6 +681,11 @@ void amdgpu_amdkfd_unreserve_memory_limit(struct amdgpu_bo *bo)
 {
 }
 
+int amdgpu_amdkfd_remove_fence_on_pt_pd_bos(struct amdgpu_bo *bo)
+{
+	return 0;
+}
+
 void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev,
 					struct amdgpu_vm *vm)
 {
@@ -713,11 +722,11 @@ void kgd2kfd_exit(void)
 {
 }
 
-void kgd2kfd_suspend(struct kfd_dev *kfd)
+void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 {
 }
 
-int kgd2kfd_resume(struct kfd_dev *kfd)
+int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 {
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 47b0f2957d1f..b0ad3be0b03f 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 *amdgpu_amdkfd_fence_create(u64 context,
 						       struct mm_struct *mm);
 bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm);
 struct amdgpu_amdkfd_fence *to_amdgpu_amdkfd_fence(struct dma_fence *f);
+int amdgpu_amdkfd_remove_fence_on_pt_pd_bos(struct amdgpu_bo *bo);
 
 struct amdkfd_process_info {
 	/* List head of all VMs that belong to a KFD process */
@@ -122,8 +123,8 @@ struct amdkfd_process_info {
 int amdgpu_amdkfd_init(void);
 void amdgpu_amdkfd_fini(void);
 
-void amdgpu_amdkfd_suspend(struct amdgpu_device *adev);
-int amdgpu_amdkfd_resume(struct amdgpu_device *adev);
+void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm);
+int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm);
 void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
 			const void *ih_ring_entry);
 void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev);
@@ -171,6 +172,7 @@ int amdgpu_amdkfd_get_dmabuf_info(struct kgd_dev *kgd, int dma_buf_fd,
 				  uint32_t *flags);
 uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd);
 uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd);
+uint64_t amdgpu_amdkfd_get_unique_id(struct kgd_dev *kgd);
 uint64_t amdgpu_amdkfd_get_mmio_remap_phys_addr(struct kgd_dev *kgd);
 uint32_t amdgpu_amdkfd_get_num_gws(struct kgd_dev *kgd);
 uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct kgd_dev *dst, struct kgd_dev *src);
@@ -249,8 +251,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 			 struct drm_device *ddev,
 			 const struct kgd2kfd_shared_resources *gpu_resources);
 void kgd2kfd_device_exit(struct kfd_dev *kfd);
-void kgd2kfd_suspend(struct kfd_dev *kfd);
-int kgd2kfd_resume(struct kfd_dev *kfd);
+void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm);
+int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm);
 int kgd2kfd_pre_reset(struct kfd_dev *kfd);
 int kgd2kfd_post_reset(struct kfd_dev *kfd);
 void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index a7b17c8deb00..ca91fffb8a36 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -808,4 +808,5 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
 	.get_tile_config = amdgpu_amdkfd_get_tile_config,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
 	.get_hive_id = amdgpu_amdkfd_get_hive_id,
+	.get_unique_id = amdgpu_amdkfd_get_unique_id,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 8562afe5b761..7f91feff7c4f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -739,4 +739,5 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
 	.get_tile_config = kgd_gfx_v9_get_tile_config,
 	.set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base,
 	.get_hive_id = amdgpu_amdkfd_get_hive_id,
+	.get_unique_id = amdgpu_amdkfd_get_unique_id,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index fa8ac9d19a7a..e1d1eed7a25f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -276,6 +276,42 @@ static int amdgpu_amdkfd_remove_eviction_fence(struct amdgpu_bo *bo,
 	return 0;
 }
 
+int amdgpu_amdkfd_remove_fence_on_pt_pd_bos(struct amdgpu_bo *bo)
+{
+	struct amdgpu_bo *root = bo;
+	struct amdgpu_vm_bo_base *vm_bo;
+	struct amdgpu_vm *vm;
+	struct amdkfd_process_info *info;
+	struct amdgpu_amdkfd_fence *ef;
+	int ret;
+
+	/* we can always get vm_bo from root PD bo.*/
+	while (root->parent)
+		root = root->parent;
+
+	vm_bo = root->vm_bo;
+	if (!vm_bo)
+		return 0;
+
+	vm = vm_bo->vm;
+	if (!vm)
+		return 0;
+
+	info = vm->process_info;
+	if (!info || !info->eviction_fence)
+		return 0;
+
+	ef = container_of(dma_fence_get(&info->eviction_fence->base),
+			struct amdgpu_amdkfd_fence, base);
+
+	BUG_ON(!dma_resv_trylock(bo->tbo.base.resv));
+	ret = amdgpu_amdkfd_remove_eviction_fence(bo, ef);
+	dma_resv_unlock(bo->tbo.base.resv);
+
+	dma_fence_put(&ef->base);
+	return ret;
+}
+
 static int amdgpu_amdkfd_bo_validate(struct amdgpu_bo *bo, uint32_t domain,
 				     bool wait)
 {
@@ -847,9 +883,9 @@ static int process_sync_pds_resv(struct amdkfd_process_info *process_info,
 			    vm_list_node) {
 		struct amdgpu_bo *pd = peer_vm->root.base.bo;
 
-		ret = amdgpu_sync_resv(NULL,
-					sync, pd->tbo.base.resv,
-					AMDGPU_FENCE_OWNER_KFD, false);
+		ret = amdgpu_sync_resv(NULL, sync, pd->tbo.base.resv,
+				       AMDGPU_SYNC_NE_OWNER,
+				       AMDGPU_FENCE_OWNER_KFD);
 		if (ret)
 			return ret;
 	}
@@ -1044,6 +1080,8 @@ void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev,
 	list_del(&vm->vm_list_node);
 	mutex_unlock(&process_info->lock);
 
+	vm->process_info = NULL;
+
 	/* Release per-process resources when last compute VM is destroyed */
 	if (!process_info->n_vms) {
 		WARN_ON(!list_empty(&process_info->kfd_bo_list));
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index a62cbc8199de..f355d9a752d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -1461,6 +1461,20 @@ static enum drm_mode_status amdgpu_connector_dp_mode_valid(struct drm_connector
 	return MODE_OK;
 }
 
+static int
+amdgpu_connector_late_register(struct drm_connector *connector)
+{
+	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
+	int r = 0;
+
+	if (amdgpu_connector->ddc_bus->has_aux) {
+		amdgpu_connector->ddc_bus->aux.dev = amdgpu_connector->base.kdev;
+		r = drm_dp_aux_register(&amdgpu_connector->ddc_bus->aux);
+	}
+
+	return r;
+}
+
 static const struct drm_connector_helper_funcs amdgpu_connector_dp_helper_funcs = {
 	.get_modes = amdgpu_connector_dp_get_modes,
 	.mode_valid = amdgpu_connector_dp_mode_valid,
@@ -1475,6 +1489,7 @@ static const struct drm_connector_funcs amdgpu_connector_dp_funcs = {
 	.early_unregister = amdgpu_connector_unregister,
 	.destroy = amdgpu_connector_destroy,
 	.force = amdgpu_connector_dvi_force,
+	.late_register = amdgpu_connector_late_register,
 };
 
 static const struct drm_connector_funcs amdgpu_connector_edp_funcs = {
@@ -1485,6 +1500,7 @@ static const struct drm_connector_funcs amdgpu_connector_edp_funcs = {
 	.early_unregister = amdgpu_connector_unregister,
 	.destroy = amdgpu_connector_destroy,
 	.force = amdgpu_connector_dvi_force,
+	.late_register = amdgpu_connector_late_register,
 };
 
 void
@@ -1931,7 +1947,6 @@ amdgpu_connector_add(struct amdgpu_device *adev,
 		connector->polled = DRM_CONNECTOR_POLL_HPD;
 
 	connector->display_info.subpixel_order = subpixel_order;
-	drm_connector_register(connector);
 
 	if (has_aux)
 		amdgpu_atombios_dp_aux_init(amdgpu_connector);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 41bd2dad842c..438d10ae343b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -654,16 +654,19 @@ out:
 
 static int amdgpu_cs_sync_rings(struct amdgpu_cs_parser *p)
 {
+	struct amdgpu_fpriv *fpriv = p->filp->driver_priv;
 	struct amdgpu_bo_list_entry *e;
 	int r;
 
 	list_for_each_entry(e, &p->validated, tv.head) {
 		struct amdgpu_bo *bo = ttm_to_amdgpu_bo(e->tv.bo);
 		struct dma_resv *resv = bo->tbo.base.resv;
+		enum amdgpu_sync_mode sync_mode;
 
-		r = amdgpu_sync_resv(p->adev, &p->job->sync, resv, p->filp,
-				     amdgpu_bo_explicit_sync(bo));
-
+		sync_mode = amdgpu_bo_explicit_sync(bo) ?
+			AMDGPU_SYNC_EXPLICIT : AMDGPU_SYNC_NE_OWNER;
+		r = amdgpu_sync_resv(p->adev, &p->job->sync, resv, sync_mode,
+				     &fpriv->vm);
 		if (r)
 			return r;
 	}
@@ -1214,7 +1217,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 	job = p->job;
 	p->job = NULL;
 
-	r = drm_sched_job_init(&job->base, entity, p->filp);
+	r = drm_sched_job_init(&job->base, entity, &fpriv->vm);
 	if (r)
 		goto error_unlock;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index f24ed9a1a3e5..3bb74056b9d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -31,6 +31,8 @@
 #include <drm/drm_debugfs.h>
 
 #include "amdgpu.h"
+#include "amdgpu_pm.h"
+#include "amdgpu_dm_debugfs.h"
 
 /**
  * amdgpu_debugfs_add_files - Add simple debugfs entries
@@ -840,6 +842,55 @@ err:
 	return result;
 }
 
+/**
+ * amdgpu_debugfs_regs_gfxoff_write - Enable/disable GFXOFF
+ *
+ * @f: open file handle
+ * @buf: User buffer to write data from
+ * @size: Number of bytes to write
+ * @pos:  Offset to seek to
+ *
+ * Write a 32-bit zero to disable or a 32-bit non-zero to enable
+ */
+static ssize_t amdgpu_debugfs_gfxoff_write(struct file *f, const char __user *buf,
+					 size_t size, loff_t *pos)
+{
+	struct amdgpu_device *adev = file_inode(f)->i_private;
+	ssize_t result = 0;
+	int r;
+
+	if (size & 0x3 || *pos & 0x3)
+		return -EINVAL;
+
+	r = pm_runtime_get_sync(adev->ddev->dev);
+	if (r < 0)
+		return r;
+
+	while (size) {
+		uint32_t value;
+
+		r = get_user(value, (uint32_t *)buf);
+		if (r) {
+			pm_runtime_mark_last_busy(adev->ddev->dev);
+			pm_runtime_put_autosuspend(adev->ddev->dev);
+			return r;
+		}
+
+		amdgpu_gfx_off_ctrl(adev, value ? true : false);
+
+		result += 4;
+		buf += 4;
+		*pos += 4;
+		size -= 4;
+	}
+
+	pm_runtime_mark_last_busy(adev->ddev->dev);
+	pm_runtime_put_autosuspend(adev->ddev->dev);
+
+	return result;
+}
+
+
 static const struct file_operations amdgpu_debugfs_regs_fops = {
 	.owner = THIS_MODULE,
 	.read = amdgpu_debugfs_regs_read,
@@ -888,6 +939,11 @@ static const struct file_operations amdgpu_debugfs_gpr_fops = {
 	.llseek = default_llseek
 };
 
+static const struct file_operations amdgpu_debugfs_gfxoff_fops = {
+	.owner = THIS_MODULE,
+	.write = amdgpu_debugfs_gfxoff_write,
+};
+
 static const struct file_operations *debugfs_regs[] = {
 	&amdgpu_debugfs_regs_fops,
 	&amdgpu_debugfs_regs_didt_fops,
@@ -897,6 +953,7 @@ static const struct file_operations *debugfs_regs[] = {
 	&amdgpu_debugfs_sensors_fops,
 	&amdgpu_debugfs_wave_fops,
 	&amdgpu_debugfs_gpr_fops,
+	&amdgpu_debugfs_gfxoff_fops,
 };
 
 static const char *debugfs_regs_names[] = {
@@ -908,6 +965,7 @@ static const char *debugfs_regs_names[] = {
 	"amdgpu_sensors",
 	"amdgpu_wave",
 	"amdgpu_gpr",
+	"amdgpu_gfxoff",
 };
 
 /**
@@ -1216,6 +1274,8 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_ib_preempt, NULL,
 
 int amdgpu_debugfs_init(struct amdgpu_device *adev)
 {
+	int r, i;
+
 	adev->debugfs_preempt =
 		debugfs_create_file("amdgpu_preempt_ib", 0600,
 				    adev->ddev->primary->debugfs_root, adev,
@@ -1225,12 +1285,73 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 		return -EIO;
 	}
 
+	/* Register debugfs entries for amdgpu_ttm */
+	r = amdgpu_ttm_debugfs_init(adev);
+	if (r) {
+		DRM_ERROR("Failed to init debugfs\n");
+		return r;
+	}
+
+	r = amdgpu_debugfs_pm_init(adev);
+	if (r) {
+		DRM_ERROR("Failed to register debugfs file for dpm!\n");
+		return r;
+	}
+
+	if (amdgpu_debugfs_sa_init(adev)) {
+		dev_err(adev->dev, "failed to register debugfs file for SA\n");
+	}
+
+	if (amdgpu_debugfs_fence_init(adev))
+		dev_err(adev->dev, "fence debugfs file creation failed\n");
+
+	r = amdgpu_debugfs_gem_init(adev);
+	if (r)
+		DRM_ERROR("registering gem debugfs failed (%d).\n", r);
+
+	r = amdgpu_debugfs_regs_init(adev);
+	if (r)
+		DRM_ERROR("registering register debugfs failed (%d).\n", r);
+
+	r = amdgpu_debugfs_firmware_init(adev);
+	if (r)
+		DRM_ERROR("registering firmware debugfs failed (%d).\n", r);
+
+#if defined(CONFIG_DRM_AMD_DC)
+	if (amdgpu_device_has_dc_support(adev)) {
+		if (dtn_debugfs_init(adev))
+			DRM_ERROR("amdgpu: failed initialize dtn debugfs support.\n");
+	}
+#endif
+
+	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+		struct amdgpu_ring *ring = adev->rings[i];
+
+		if (!ring)
+			continue;
+
+		if (amdgpu_debugfs_ring_init(adev, ring)) {
+			DRM_ERROR("Failed to register debugfs file for rings !\n");
+		}
+	}
+
 	return amdgpu_debugfs_add_files(adev, amdgpu_debugfs_list,
 					ARRAY_SIZE(amdgpu_debugfs_list));
 }
 
-void amdgpu_debugfs_preempt_cleanup(struct amdgpu_device *adev)
+void amdgpu_debugfs_fini(struct amdgpu_device *adev)
 {
+	int i;
+
+	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+		struct amdgpu_ring *ring = adev->rings[i];
+
+		if (!ring)
+			continue;
+
+		amdgpu_debugfs_ring_fini(ring);
+	}
+	amdgpu_ttm_debugfs_fini(adev);
 	debugfs_remove(adev->debugfs_preempt);
 }
 
@@ -1239,7 +1360,7 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 {
 	return 0;
 }
-void amdgpu_debugfs_preempt_cleanup(struct amdgpu_device *adev) { }
+void amdgpu_debugfs_fini(struct amdgpu_device *adev) { }
 int amdgpu_debugfs_regs_init(struct amdgpu_device *adev)
 {
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h
index f289d28ad6b2..b382527e359a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h
@@ -34,7 +34,7 @@ struct amdgpu_debugfs {
 int amdgpu_debugfs_regs_init(struct amdgpu_device *adev);
 void amdgpu_debugfs_regs_cleanup(struct amdgpu_device *adev);
 int amdgpu_debugfs_init(struct amdgpu_device *adev);
-void amdgpu_debugfs_preempt_cleanup(struct amdgpu_device *adev);
+void amdgpu_debugfs_fini(struct amdgpu_device *adev);
 int amdgpu_debugfs_add_files(struct amdgpu_device *adev,
 			     const struct drm_info_list *files,
 			     unsigned nfiles);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 53fdafd85cb8..a3aaf2e3794c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -183,20 +183,51 @@ bool amdgpu_device_supports_baco(struct drm_device *dev)
 void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos,
 			       uint32_t *buf, size_t size, bool write)
 {
-	uint64_t last;
 	unsigned long flags;
+	uint32_t hi = ~0;
+	uint64_t last;
+
+
+#ifdef CONFIG_64BIT
+	last = min(pos + size, adev->gmc.visible_vram_size);
+	if (last > pos) {
+		void __iomem *addr = adev->mman.aper_base_kaddr + pos;
+		size_t count = last - pos;
+
+		if (write) {
+			memcpy_toio(addr, buf, count);
+			mb();
+			amdgpu_asic_flush_hdp(adev, NULL);
+		} else {
+			amdgpu_asic_invalidate_hdp(adev, NULL);
+			mb();
+			memcpy_fromio(buf, addr, count);
+		}
+
+		if (count == size)
+			return;
+
+		pos += count;
+		buf += count / 4;
+		size -= count;
+	}
+#endif
+
+	spin_lock_irqsave(&adev->mmio_idx_lock, flags);
+	for (last = pos + size; pos < last; pos += 4) {
+		uint32_t tmp = pos >> 31;
 
-	last = size - 4;
-	for (last += pos; pos <= last; pos += 4) {
-		spin_lock_irqsave(&adev->mmio_idx_lock, flags);
 		WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)pos) | 0x80000000);
-		WREG32_NO_KIQ(mmMM_INDEX_HI, pos >> 31);
+		if (tmp != hi) {
+			WREG32_NO_KIQ(mmMM_INDEX_HI, tmp);
+			hi = tmp;
+		}
 		if (write)
 			WREG32_NO_KIQ(mmMM_DATA, *buf++);
 		else
 			*buf++ = RREG32_NO_KIQ(mmMM_DATA);
-		spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
 	}
+	spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
 }
 
 /*
@@ -2344,15 +2375,16 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev)
 		}
 		adev->ip_blocks[i].status.hw = false;
 		/* handle putting the SMC in the appropriate state */
-		if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) {
-			r = amdgpu_dpm_set_mp1_state(adev, adev->mp1_state);
-			if (r) {
-				DRM_ERROR("SMC failed to set mp1 state %d, %d\n",
-					  adev->mp1_state, r);
-				return r;
+		if(!amdgpu_sriov_vf(adev)){
+			if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) {
+				r = amdgpu_dpm_set_mp1_state(adev, adev->mp1_state);
+				if (r) {
+					DRM_ERROR("SMC failed to set mp1 state %d, %d\n",
+							adev->mp1_state, r);
+					return r;
+				}
 			}
 		}
-
 		adev->ip_blocks[i].status.hw = false;
 	}
 
@@ -2800,7 +2832,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 
 	adev->usec_timeout = AMDGPU_MAX_USEC_TIMEOUT;
 	if (amdgpu_emu_mode == 1)
-		adev->usec_timeout *= 2;
+		adev->usec_timeout *= 10;
 	adev->gmc.gart_size = 512 * 1024 * 1024;
 	adev->accel_working = false;
 	adev->num_rings = 0;
@@ -3088,22 +3120,6 @@ fence_driver_init:
 	} else
 		adev->ucode_sysfs_en = true;
 
-	r = amdgpu_debugfs_gem_init(adev);
-	if (r)
-		DRM_ERROR("registering gem debugfs failed (%d).\n", r);
-
-	r = amdgpu_debugfs_regs_init(adev);
-	if (r)
-		DRM_ERROR("registering register debugfs failed (%d).\n", r);
-
-	r = amdgpu_debugfs_firmware_init(adev);
-	if (r)
-		DRM_ERROR("registering firmware debugfs failed (%d).\n", r);
-
-	r = amdgpu_debugfs_init(adev);
-	if (r)
-		DRM_ERROR("Creating debugfs files failed (%d).\n", r);
-
 	if ((amdgpu_testing & 1)) {
 		if (adev->accel_working)
 			amdgpu_test_moves(adev);
@@ -3225,7 +3241,6 @@ void amdgpu_device_fini(struct amdgpu_device *adev)
 		amdgpu_ucode_sysfs_fini(adev);
 	if (IS_ENABLED(CONFIG_PERF_EVENTS))
 		amdgpu_pmu_fini(adev);
-	amdgpu_debugfs_preempt_cleanup(adev);
 	if (amdgpu_discovery && adev->asic_type >= CHIP_NAVI10)
 		amdgpu_discovery_fini(adev);
 }
@@ -3309,7 +3324,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 		}
 	}
 
-	amdgpu_amdkfd_suspend(adev);
+	amdgpu_amdkfd_suspend(adev, !fbcon);
 
 	amdgpu_ras_suspend(adev);
 
@@ -3393,7 +3408,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 			}
 		}
 	}
-	r = amdgpu_amdkfd_resume(adev);
+	r = amdgpu_amdkfd_resume(adev, !fbcon);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index f95092741c38..27d8ae19a7a4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -307,7 +307,7 @@ int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 
 			for (hw_ip = 0; hw_ip < MAX_HWIP; hw_ip++) {
 				if (hw_id_map[hw_ip] == le16_to_cpu(ip->hw_id)) {
-					DRM_INFO("set register base offset for %s\n",
+					DRM_DEBUG("set register base offset for %s\n",
 							hw_id_names[le16_to_cpu(ip->hw_id)]);
 					adev->reg_offset[hw_ip][ip->number_instance] =
 						ip->base_address;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c
index a2e8c3dfb4f1..ba1bb95a3cf9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c
@@ -1171,3 +1171,20 @@ int amdgpu_dpm_set_xgmi_pstate(struct amdgpu_device *adev,
 
 	return ret;
 }
+
+int amdgpu_dpm_set_df_cstate(struct amdgpu_device *adev,
+			     uint32_t cstate)
+{
+	int ret = 0;
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+	void *pp_handle = adev->powerplay.pp_handle;
+	struct smu_context *smu = &adev->smu;
+
+	if (is_support_sw_smu(adev))
+		ret = smu_set_df_cstate(smu, cstate);
+	else if (pp_funcs &&
+		 pp_funcs->set_df_cstate)
+		ret = pp_funcs->set_df_cstate(pp_handle, cstate);
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
index 902ca6c00cca..168579492a55 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h
@@ -533,4 +533,7 @@ int amdgpu_dpm_baco_exit(struct amdgpu_device *adev);
 
 int amdgpu_dpm_baco_enter(struct amdgpu_device *adev);
 
+int amdgpu_dpm_set_df_cstate(struct amdgpu_device *adev,
+			     uint32_t cstate);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 4598836c5fa4..e1c4c2df716b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -1021,6 +1021,7 @@ static int amdgpu_pci_probe(struct pci_dev *pdev,
 			    const struct pci_device_id *ent)
 {
 	struct drm_device *dev;
+	struct amdgpu_device *adev;
 	unsigned long flags = ent->driver_data;
 	int ret, retry = 0;
 	bool supports_atomic = false;
@@ -1090,6 +1091,8 @@ static int amdgpu_pci_probe(struct pci_dev *pdev,
 
 	pci_set_drvdata(pdev, dev);
 
+	amdgpu_driver_load_kms(dev, ent->driver_data);
+
 retry_init:
 	ret = drm_dev_register(dev, ent->driver_data);
 	if (ret == -EAGAIN && ++retry <= 3) {
@@ -1100,6 +1103,11 @@ retry_init:
 	} else if (ret)
 		goto err_pci;
 
+	adev = dev->dev_private;
+	ret = amdgpu_debugfs_init(adev);
+	if (ret)
+		DRM_ERROR("Creating debugfs files failed (%d).\n", ret);
+
 	return 0;
 
 err_pci:
@@ -1113,6 +1121,7 @@ static void
 amdgpu_pci_remove(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
+	struct amdgpu_device *adev = dev->dev_private;
 
 #ifdef MODULE
 	if (THIS_MODULE->state != MODULE_STATE_GOING)
@@ -1120,6 +1129,8 @@ amdgpu_pci_remove(struct pci_dev *pdev)
 		DRM_ERROR("Hotplug removal is not supported\n");
 	drm_dev_unplug(dev);
 	drm_dev_put(dev);
+	amdgpu_debugfs_fini(adev);
+	amdgpu_driver_unload_kms(dev);
 	pci_disable_device(pdev);
 	pci_set_drvdata(pdev, NULL);
 }
@@ -1220,11 +1231,15 @@ static int amdgpu_pmops_runtime_suspend(struct device *dev)
 		}
 	}
 
+	adev->in_runpm = true;
 	if (amdgpu_device_supports_boco(drm_dev))
 		drm_dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
 	drm_kms_helper_poll_disable(drm_dev);
 
 	ret = amdgpu_device_suspend(drm_dev, false);
+	if (ret)
+		return ret;
+
 	if (amdgpu_device_supports_boco(drm_dev)) {
 		/* Only need to handle PCI state in the driver for ATPX
 		 * PCI core handles it for _PR3.
@@ -1278,6 +1293,7 @@ static int amdgpu_pmops_runtime_resume(struct device *dev)
 	drm_kms_helper_poll_enable(drm_dev);
 	if (amdgpu_device_supports_boco(drm_dev))
 		drm_dev->switch_power_state = DRM_SWITCH_POWER_ON;
+	adev->in_runpm = false;
 	return 0;
 }
 
@@ -1379,15 +1395,13 @@ int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv)
 
 static struct drm_driver kms_driver = {
 	.driver_features =
-	    DRIVER_USE_AGP | DRIVER_ATOMIC |
+	    DRIVER_ATOMIC |
 	    DRIVER_GEM |
 	    DRIVER_RENDER | DRIVER_MODESET | DRIVER_SYNCOBJ |
 	    DRIVER_SYNCOBJ_TIMELINE,
-	.load = amdgpu_driver_load_kms,
 	.open = amdgpu_driver_open_kms,
 	.postclose = amdgpu_driver_postclose_kms,
 	.lastclose = amdgpu_driver_lastclose_kms,
-	.unload = amdgpu_driver_unload_kms,
 	.irq_handler = amdgpu_irq_handler,
 	.ioctls = amdgpu_ioctls_kms,
 	.gem_free_object_unlocked = amdgpu_gem_object_free,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 3c01252b1e0e..7531527067df 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -503,9 +503,6 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring,
  */
 int amdgpu_fence_driver_init(struct amdgpu_device *adev)
 {
-	if (amdgpu_debugfs_fence_init(adev))
-		dev_err(adev->dev, "fence debugfs file creation failed\n");
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 0f960b498792..7403588684b3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -477,7 +477,7 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev)
 		kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.compute_ring[i],
 					   RESET_QUEUES, 0, 0);
 
-	return amdgpu_ring_test_ring(kiq_ring);
+	return amdgpu_ring_test_helper(kiq_ring);
 }
 
 int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index d3c27a3c43f6..7546da0cc70c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -195,6 +195,7 @@ struct amdgpu_gmc {
 	uint32_t                srbm_soft_reset;
 	bool			prt_warning;
 	uint64_t		stolen_size;
+	uint32_t		sdpif_register;
 	/* apertures */
 	u64			shared_aperture_start;
 	u64			shared_aperture_end;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index 60655834d649..ccbd7acfc4cb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -48,7 +48,6 @@
  * produce command buffers which are send to the kernel and
  * put in IBs for execution by the requested ring.
  */
-static int amdgpu_debugfs_sa_init(struct amdgpu_device *adev);
 
 /**
  * amdgpu_ib_get - request an IB (Indirect Buffer)
@@ -295,9 +294,7 @@ int amdgpu_ib_pool_init(struct amdgpu_device *adev)
 	}
 
 	adev->ib_pool_ready = true;
-	if (amdgpu_debugfs_sa_init(adev)) {
-		dev_err(adev->dev, "failed to register debugfs file for SA\n");
-	}
+
 	return 0;
 }
 
@@ -421,7 +418,7 @@ static const struct drm_info_list amdgpu_debugfs_sa_list[] = {
 
 #endif
 
-static int amdgpu_debugfs_sa_init(struct amdgpu_device *adev)
+int amdgpu_debugfs_sa_init(struct amdgpu_device *adev)
 {
 #if defined(CONFIG_DEBUG_FS)
 	return amdgpu_debugfs_add_files(adev, amdgpu_debugfs_sa_list, 1);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 98c196de27a4..6a1b62bc3dcf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -170,10 +170,16 @@ int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags)
 	}
 
 	if (amdgpu_device_supports_boco(dev) &&
-	    (amdgpu_runtime_pm != 0)) /* enable runpm by default */
+	    (amdgpu_runtime_pm != 0)) /* enable runpm by default for boco */
 		adev->runpm = true;
 	else if (amdgpu_device_supports_baco(dev) &&
-		 (amdgpu_runtime_pm > 0)) /* enable runpm if runpm=1 */
+		 (amdgpu_runtime_pm != 0) &&
+		 (adev->asic_type >= CHIP_TOPAZ) &&
+		 (adev->asic_type != CHIP_VEGA20) &&
+		 (adev->asic_type != CHIP_ARCTURUS)) /* enable runpm on VI+ */
+		adev->runpm = true;
+	else if (amdgpu_device_supports_baco(dev) &&
+		 (amdgpu_runtime_pm > 0))  /* enable runpm if runpm=1 on CI */
 		adev->runpm = true;
 
 	/* Call ACPI methods: require modeset init
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 5fa8f59c4ccf..1791c084787d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -1318,6 +1318,11 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo)
 	if (abo->kfd_bo)
 		amdgpu_amdkfd_unreserve_memory_limit(abo);
 
+	/* We only remove the fence if the resv has individualized. */
+	WARN_ON_ONCE(bo->base.resv != &bo->base._resv);
+	if (bo->base.resv == &bo->base._resv)
+		amdgpu_amdkfd_remove_fence_on_pt_pd_bos(abo);
+
 	if (bo->mem.mem_type != TTM_PL_VRAM || !bo->mem.mm_node ||
 	    !(abo->flags & AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE))
 		return;
@@ -1414,30 +1419,52 @@ void amdgpu_bo_fence(struct amdgpu_bo *bo, struct dma_fence *fence,
 }
 
 /**
- * amdgpu_sync_wait_resv - Wait for BO reservation fences
+ * amdgpu_bo_sync_wait_resv - Wait for BO reservation fences
  *
- * @bo: buffer object
+ * @adev: amdgpu device pointer
+ * @resv: reservation object to sync to
+ * @sync_mode: synchronization mode
  * @owner: fence owner
  * @intr: Whether the wait is interruptible
  *
+ * Extract the fences from the reservation object and waits for them to finish.
+ *
  * Returns:
  * 0 on success, errno otherwise.
  */
-int amdgpu_bo_sync_wait(struct amdgpu_bo *bo, void *owner, bool intr)
+int amdgpu_bo_sync_wait_resv(struct amdgpu_device *adev, struct dma_resv *resv,
+			     enum amdgpu_sync_mode sync_mode, void *owner,
+			     bool intr)
 {
-	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
 	struct amdgpu_sync sync;
 	int r;
 
 	amdgpu_sync_create(&sync);
-	amdgpu_sync_resv(adev, &sync, bo->tbo.base.resv, owner, false);
+	amdgpu_sync_resv(adev, &sync, resv, sync_mode, owner);
 	r = amdgpu_sync_wait(&sync, intr);
 	amdgpu_sync_free(&sync);
-
 	return r;
 }
 
 /**
+ * amdgpu_bo_sync_wait - Wrapper for amdgpu_bo_sync_wait_resv
+ * @bo: buffer object to wait for
+ * @owner: fence owner
+ * @intr: Whether the wait is interruptible
+ *
+ * Wrapper to wait for fences in a BO.
+ * Returns:
+ * 0 on success, errno otherwise.
+ */
+int amdgpu_bo_sync_wait(struct amdgpu_bo *bo, void *owner, bool intr)
+{
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+
+	return amdgpu_bo_sync_wait_resv(adev, bo->tbo.base.resv,
+					AMDGPU_SYNC_NE_OWNER, owner, intr);
+}
+
+/**
  * amdgpu_bo_gpu_offset - return GPU offset of bo
  * @bo:	amdgpu object for which we query the offset
  *
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index 36dec51d1ef1..5e39ecd8cc28 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -277,6 +277,9 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo);
 int amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo);
 void amdgpu_bo_fence(struct amdgpu_bo *bo, struct dma_fence *fence,
 		     bool shared);
+int amdgpu_bo_sync_wait_resv(struct amdgpu_device *adev, struct dma_resv *resv,
+			     enum amdgpu_sync_mode sync_mode, void *owner,
+			     bool intr);
 int amdgpu_bo_sync_wait(struct amdgpu_bo *bo, void *owner, bool intr);
 u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo);
 int amdgpu_bo_validate(struct amdgpu_bo *bo);
@@ -316,6 +319,7 @@ void amdgpu_sa_bo_free(struct amdgpu_device *adev,
 void amdgpu_sa_bo_dump_debug_info(struct amdgpu_sa_manager *sa_manager,
 					 struct seq_file *m);
 #endif
+int amdgpu_debugfs_sa_init(struct amdgpu_device *adev);
 
 bool amdgpu_bo_support_uswc(u64 bo_flags);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
index b03b1eb7ba04..bc3cf04a1a94 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
@@ -41,8 +41,6 @@
 #include "hwmgr.h"
 #define WIDTH_4K 3840
 
-static int amdgpu_debugfs_pm_init(struct amdgpu_device *adev);
-
 static const struct cg_flag_name clocks[] = {
 	{AMD_CG_SUPPORT_GFX_MGCG, "Graphics Medium Grain Clock Gating"},
 	{AMD_CG_SUPPORT_GFX_MGLS, "Graphics Medium Grain memory Light Sleep"},
@@ -3398,11 +3396,6 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 		DRM_ERROR("failed to create device file unique_id\n");
 		return ret;
 	}
-	ret = amdgpu_debugfs_pm_init(adev);
-	if (ret) {
-		DRM_ERROR("Failed to register debugfs file for dpm!\n");
-		return ret;
-	}
 
 	if ((adev->asic_type >= CHIP_VEGA10) &&
 	    !(adev->flags & AMD_IS_APU)) {
@@ -3669,7 +3662,7 @@ static const struct drm_info_list amdgpu_pm_info_list[] = {
 };
 #endif
 
-static int amdgpu_debugfs_pm_init(struct amdgpu_device *adev)
+int amdgpu_debugfs_pm_init(struct amdgpu_device *adev)
 {
 #if defined(CONFIG_DEBUG_FS)
 	return amdgpu_debugfs_add_files(adev, amdgpu_pm_info_list, ARRAY_SIZE(amdgpu_pm_info_list));
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h
index 3da1da277805..5db0ef86e84c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h
@@ -43,4 +43,6 @@ void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable);
 void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable);
 void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable);
 
+int amdgpu_debugfs_pm_init(struct amdgpu_device *adev);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 3a1570dafe34..d33f74100094 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -38,6 +38,39 @@
 
 static void psp_set_funcs(struct amdgpu_device *adev);
 
+/*
+ * Due to DF Cstate management centralized to PMFW, the firmware
+ * loading sequence will be updated as below:
+ *   - Load KDB
+ *   - Load SYS_DRV
+ *   - Load tOS
+ *   - Load PMFW
+ *   - Setup TMR
+ *   - Load other non-psp fw
+ *   - Load ASD
+ *   - Load XGMI/RAS/HDCP/DTM TA if any
+ *
+ * This new sequence is required for
+ *   - Arcturus
+ *   - Navi12 and onwards
+ */
+static void psp_check_pmfw_centralized_cstate_management(struct psp_context *psp)
+{
+	struct amdgpu_device *adev = psp->adev;
+
+	psp->pmfw_centralized_cstate_management = false;
+
+	if (amdgpu_sriov_vf(adev))
+		return;
+
+	if (adev->flags & AMD_IS_APU)
+		return;
+
+	if ((adev->asic_type == CHIP_ARCTURUS) ||
+	    (adev->asic_type >= CHIP_NAVI12))
+		psp->pmfw_centralized_cstate_management = true;
+}
+
 static int psp_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -75,6 +108,8 @@ static int psp_early_init(void *handle)
 
 	psp->adev = adev;
 
+	psp_check_pmfw_centralized_cstate_management(psp);
+
 	return 0;
 }
 
@@ -558,7 +593,7 @@ int psp_xgmi_invoke(struct psp_context *psp, uint32_t ta_cmd_id)
 	return psp_ta_invoke(psp, ta_cmd_id, psp->xgmi_context.session_id);
 }
 
-static int psp_xgmi_terminate(struct psp_context *psp)
+int psp_xgmi_terminate(struct psp_context *psp)
 {
 	int ret;
 
@@ -579,7 +614,7 @@ static int psp_xgmi_terminate(struct psp_context *psp)
 	return 0;
 }
 
-static int psp_xgmi_initialize(struct psp_context *psp)
+int psp_xgmi_initialize(struct psp_context *psp)
 {
 	struct ta_xgmi_shared_memory *xgmi_cmd;
 	int ret;
@@ -1013,6 +1048,30 @@ static int psp_dtm_initialize(struct psp_context *psp)
 	return 0;
 }
 
+static int psp_dtm_unload(struct psp_context *psp)
+{
+	int ret;
+	struct psp_gfx_cmd_resp *cmd;
+
+	/*
+	 * TODO: bypass the unloading in sriov for now
+	 */
+	if (amdgpu_sriov_vf(psp->adev))
+		return 0;
+
+	cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+
+	psp_prep_ta_unload_cmd_buf(cmd, psp->dtm_context.session_id);
+
+	ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr);
+
+	kfree(cmd);
+
+	return ret;
+}
+
 int psp_dtm_invoke(struct psp_context *psp, uint32_t ta_cmd_id)
 {
 	/*
@@ -1037,7 +1096,7 @@ static int psp_dtm_terminate(struct psp_context *psp)
 	if (!psp->dtm_context.dtm_initialized)
 		return 0;
 
-	ret = psp_hdcp_unload(psp);
+	ret = psp_dtm_unload(psp);
 	if (ret)
 		return ret;
 
@@ -1057,7 +1116,7 @@ static int psp_hw_start(struct psp_context *psp)
 	struct amdgpu_device *adev = psp->adev;
 	int ret;
 
-	if (!amdgpu_sriov_vf(adev) || !adev->in_gpu_reset) {
+	if (!amdgpu_sriov_vf(adev)) {
 		if (psp->kdb_bin_size &&
 		    (psp->funcs->bootloader_load_kdb != NULL)) {
 			ret = psp_bootloader_load_kdb(psp);
@@ -1092,10 +1151,17 @@ static int psp_hw_start(struct psp_context *psp)
 		return ret;
 	}
 
-	ret = psp_tmr_load(psp);
-	if (ret) {
-		DRM_ERROR("PSP load tmr failed!\n");
-		return ret;
+	/*
+	 * For those ASICs with DF Cstate management centralized
+	 * to PMFW, TMR setup should be performed after PMFW
+	 * loaded and before other non-psp firmware loaded.
+	 */
+	if (!psp->pmfw_centralized_cstate_management) {
+		ret = psp_tmr_load(psp);
+		if (ret) {
+			DRM_ERROR("PSP load tmr failed!\n");
+			return ret;
+		}
 	}
 
 	return 0;
@@ -1292,9 +1358,10 @@ static int psp_np_fw_load(struct psp_context *psp)
 	struct amdgpu_firmware_info *ucode;
 	struct amdgpu_device* adev = psp->adev;
 
-	if (psp->autoload_supported) {
+	if (psp->autoload_supported ||
+	    psp->pmfw_centralized_cstate_management) {
 		ucode = &adev->firmware.ucode[AMDGPU_UCODE_ID_SMC];
-		if (!ucode->fw)
+		if (!ucode->fw || amdgpu_sriov_vf(adev))
 			goto out;
 
 		ret = psp_execute_np_fw_load(psp, ucode);
@@ -1302,6 +1369,14 @@ static int psp_np_fw_load(struct psp_context *psp)
 			return ret;
 	}
 
+	if (psp->pmfw_centralized_cstate_management) {
+		ret = psp_tmr_load(psp);
+		if (ret) {
+			DRM_ERROR("PSP load tmr failed!\n");
+			return ret;
+		}
+	}
+
 out:
 	for (i = 0; i < adev->firmware.max_ucodes; i++) {
 		ucode = &adev->firmware.ucode[i];
@@ -1309,7 +1384,9 @@ out:
 			continue;
 
 		if (ucode->ucode_id == AMDGPU_UCODE_ID_SMC &&
-		    (psp_smu_reload_quirk(psp) || psp->autoload_supported))
+		    (psp_smu_reload_quirk(psp) ||
+		     psp->autoload_supported ||
+		     psp->pmfw_centralized_cstate_management))
 			continue;
 
 		if (amdgpu_sriov_vf(adev) &&
@@ -1420,16 +1497,6 @@ skip_memalloc:
 		return ret;
 	}
 
-	if (adev->gmc.xgmi.num_physical_nodes > 1) {
-		ret = psp_xgmi_initialize(psp);
-		/* Warning the XGMI seesion initialize failure
-		 * Instead of stop driver initialization
-		 */
-		if (ret)
-			dev_err(psp->adev->dev,
-				"XGMI: Failed to initialize XGMI session\n");
-	}
-
 	if (psp->adev->psp.ta_fw) {
 		ret = psp_ras_initialize(psp);
 		if (ret)
@@ -1494,10 +1561,6 @@ static int psp_hw_fini(void *handle)
 	void *tmr_buf;
 	void **pptr;
 
-	if (adev->gmc.xgmi.num_physical_nodes > 1 &&
-	    psp->xgmi_context.initialized == 1)
-                psp_xgmi_terminate(psp);
-
 	if (psp->adev->psp.ta_fw) {
 		psp_ras_terminate(psp);
 		psp_dtm_terminate(psp);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index 611021514c52..37fa184f27f6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -264,6 +264,8 @@ struct psp_context
 	atomic_t			fence_value;
 	/* flag to mark whether gfx fw autoload is supported or not */
 	bool				autoload_supported;
+	/* flag to mark whether df cstate management centralized to PMFW */
+	bool				pmfw_centralized_cstate_management;
 
 	/* xgmi ta firmware and buffer */
 	const struct firmware		*ta_fw;
@@ -362,6 +364,8 @@ int psp_gpu_reset(struct amdgpu_device *adev);
 int psp_update_vcn_sram(struct amdgpu_device *adev, int inst_idx,
 			uint64_t cmd_gpu_addr, int cmd_size);
 
+int psp_xgmi_initialize(struct psp_context *psp);
+int psp_xgmi_terminate(struct psp_context *psp);
 int psp_xgmi_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
 
 int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index cef94e2169fe..930633a0ed64 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -31,6 +31,7 @@
 #include "amdgpu.h"
 #include "amdgpu_ras.h"
 #include "amdgpu_atomfirmware.h"
+#include "amdgpu_xgmi.h"
 #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h"
 
 const char *ras_error_string[] = {
@@ -742,20 +743,6 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev,
 	return 0;
 }
 
-uint64_t get_xgmi_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr)
-{
-	uint32_t df_inst_id;
-
-	if ((!adev->df.funcs)                 ||
-	    (!adev->df.funcs->get_df_inst_id) ||
-	    (!adev->df.funcs->get_dram_base_addr))
-		return addr;
-
-	df_inst_id = adev->df.funcs->get_df_inst_id(adev);
-
-	return addr + adev->df.funcs->get_dram_base_addr(adev, df_inst_id);
-}
-
 /* wrapper of psp_ras_trigger_error */
 int amdgpu_ras_error_inject(struct amdgpu_device *adev,
 		struct ras_inject_if *info)
@@ -775,8 +762,9 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev,
 
 	/* Calculate XGMI relative offset */
 	if (adev->gmc.xgmi.num_physical_nodes > 1) {
-		block_info.address = get_xgmi_relative_phy_addr(adev,
-								block_info.address);
+		block_info.address =
+			amdgpu_xgmi_get_relative_phy_addr(adev,
+							  block_info.address);
 	}
 
 	switch (info->head.block) {
@@ -1319,6 +1307,33 @@ static int amdgpu_ras_interrupt_remove_all(struct amdgpu_device *adev)
 }
 /* ih end */
 
+/* traversal all IPs except NBIO to query error counter */
+static void amdgpu_ras_log_on_err_counter(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj;
+
+	if (!con)
+		return;
+
+	list_for_each_entry(obj, &con->head, node) {
+		struct ras_query_if info = {
+			.head = obj->head,
+		};
+
+		/*
+		 * PCIE_BIF IP has one different isr by ras controller
+		 * interrupt, the specific ras counter query will be
+		 * done in that isr. So skip such block from common
+		 * sync flood interrupt isr calling.
+		 */
+		if (info.head.block == AMDGPU_RAS_BLOCK__PCIE_BIF)
+			continue;
+
+		amdgpu_ras_error_query(adev, &info);
+	}
+}
+
 /* recovery begin */
 
 /* return 0 on success.
@@ -1373,6 +1388,12 @@ static void amdgpu_ras_do_recovery(struct work_struct *work)
 	struct amdgpu_ras *ras =
 		container_of(work, struct amdgpu_ras, recovery_work);
 
+	/*
+	 * Query and print non zero error counter per IP block for
+	 * awareness before recovering GPU.
+	 */
+	amdgpu_ras_log_on_err_counter(ras->adev);
+
 	if (amdgpu_device_should_recover_gpu(ras->adev))
 		amdgpu_device_gpu_recover(ras->adev, 0);
 	atomic_set(&ras->in_recovery, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index 2a8e04895595..ed15b1fa5e98 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -26,9 +26,11 @@
 #include "amdgpu_ras.h"
 #include <linux/bits.h>
 #include "smu_v11_0_i2c.h"
+#include "atom.h"
 
-#define EEPROM_I2C_TARGET_ADDR_ARCTURUS  0xA8
-#define EEPROM_I2C_TARGET_ADDR_VEGA20    0xA0
+#define EEPROM_I2C_TARGET_ADDR_VEGA20    	0xA0
+#define EEPROM_I2C_TARGET_ADDR_ARCTURUS  	0xA8
+#define EEPROM_I2C_TARGET_ADDR_ARCTURUS_D342  	0xA0
 
 /*
  * The 2 macros bellow represent the actual size in bytes that
@@ -55,6 +57,45 @@
 
 #define to_amdgpu_device(x) (container_of(x, struct amdgpu_ras, eeprom_control))->adev
 
+static bool __get_eeprom_i2c_addr_arct(struct amdgpu_device *adev,
+				       uint16_t *i2c_addr)
+{
+	struct atom_context *atom_ctx = adev->mode_info.atom_context;
+
+	if (!i2c_addr || !atom_ctx)
+		return false;
+
+	if (strnstr(atom_ctx->vbios_version,
+	            "D342",
+		    sizeof(atom_ctx->vbios_version)))
+		*i2c_addr = EEPROM_I2C_TARGET_ADDR_ARCTURUS_D342;
+	else
+		*i2c_addr = EEPROM_I2C_TARGET_ADDR_ARCTURUS;
+
+	return true;
+}
+
+static bool __get_eeprom_i2c_addr(struct amdgpu_device *adev,
+				  uint16_t *i2c_addr)
+{
+	if (!i2c_addr)
+		return false;
+
+	switch (adev->asic_type) {
+	case CHIP_VEGA20:
+		*i2c_addr = EEPROM_I2C_TARGET_ADDR_VEGA20;
+		break;
+
+	case CHIP_ARCTURUS:
+		return __get_eeprom_i2c_addr_arct(adev, i2c_addr);
+
+	default:
+		return false;
+	}
+
+	return true;
+}
+
 static void __encode_table_header_to_buff(struct amdgpu_ras_eeprom_table_header *hdr,
 					  unsigned char *buff)
 {
@@ -103,8 +144,6 @@ static int __update_table_header(struct amdgpu_ras_eeprom_control *control,
 	return ret;
 }
 
-
-
 static uint32_t  __calc_hdr_byte_sum(struct amdgpu_ras_eeprom_control *control)
 {
 	int i;
@@ -212,16 +251,17 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control)
 			.buf	= buff,
 	};
 
+	if (!__get_eeprom_i2c_addr(adev, &control->i2c_address))
+		return -EINVAL;
+
 	mutex_init(&control->tbl_mutex);
 
 	switch (adev->asic_type) {
 	case CHIP_VEGA20:
-		control->i2c_address = EEPROM_I2C_TARGET_ADDR_VEGA20;
 		ret = smu_v11_0_i2c_eeprom_control_init(&control->eeprom_accessor);
 		break;
 
 	case CHIP_ARCTURUS:
-		control->i2c_address = EEPROM_I2C_TARGET_ADDR_ARCTURUS;
 		ret = smu_i2c_eeprom_init(&adev->smu, &control->eeprom_accessor);
 		break;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index e5c83e164d82..18e11b0fdc3e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -48,9 +48,6 @@
  * wptr.  The GPU then starts fetching commands and executes
  * them until the pointers are equal again.
  */
-static int amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
-				    struct amdgpu_ring *ring);
-static void amdgpu_debugfs_ring_fini(struct amdgpu_ring *ring);
 
 /**
  * amdgpu_ring_alloc - allocate space on the ring buffer
@@ -334,10 +331,6 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 	for (i = 0; i < DRM_SCHED_PRIORITY_MAX; ++i)
 		atomic_set(&ring->num_jobs[i], 0);
 
-	if (amdgpu_debugfs_ring_init(adev, ring)) {
-		DRM_ERROR("Failed to register debugfs file for rings !\n");
-	}
-
 	return 0;
 }
 
@@ -351,12 +344,13 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
  */
 void amdgpu_ring_fini(struct amdgpu_ring *ring)
 {
-	ring->sched.ready = false;
 
 	/* Not to finish a ring which is not initialized */
 	if (!(ring->adev) || !(ring->adev->rings[ring->idx]))
 		return;
 
+	ring->sched.ready = false;
+
 	amdgpu_device_wb_free(ring->adev, ring->rptr_offs);
 	amdgpu_device_wb_free(ring->adev, ring->wptr_offs);
 
@@ -367,8 +361,6 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 			      &ring->gpu_addr,
 			      (void **)&ring->ring);
 
-	amdgpu_debugfs_ring_fini(ring);
-
 	dma_fence_put(ring->vmid_wait);
 	ring->vmid_wait = NULL;
 	ring->me = 0;
@@ -485,8 +477,8 @@ static const struct file_operations amdgpu_debugfs_ring_fops = {
 
 #endif
 
-static int amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
-				    struct amdgpu_ring *ring)
+int amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
+			     struct amdgpu_ring *ring)
 {
 #if defined(CONFIG_DEBUG_FS)
 	struct drm_minor *minor = adev->ddev->primary;
@@ -507,7 +499,7 @@ static int amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
 	return 0;
 }
 
-static void amdgpu_debugfs_ring_fini(struct amdgpu_ring *ring)
+void amdgpu_debugfs_ring_fini(struct amdgpu_ring *ring)
 {
 #if defined(CONFIG_DEBUG_FS)
 	debugfs_remove(ring->ent);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 930316e60155..24caff085d00 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -328,4 +328,8 @@ static inline void amdgpu_ring_write_multiple(struct amdgpu_ring *ring,
 
 int amdgpu_ring_test_helper(struct amdgpu_ring *ring);
 
+int amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
+			     struct amdgpu_ring *ring);
+void amdgpu_debugfs_ring_fini(struct amdgpu_ring *ring);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
index a2ee30b16212..7854c053e85d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
@@ -70,7 +70,8 @@ uint64_t amdgpu_sdma_get_csa_mc_addr(struct amdgpu_ring *ring,
 	uint32_t index = 0;
 	int r;
 
-	if (vmid == 0 || !amdgpu_mcbp)
+	/* don't enable OS preemption on SDMA under SRIOV */
+	if (amdgpu_sriov_vf(adev) || vmid == 0 || !amdgpu_mcbp)
 		return 0;
 
 	r = amdgpu_sdma_get_index_from_ring(ring, &index);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
index a09b6b9c27d1..b86392253696 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
@@ -202,18 +202,17 @@ int amdgpu_sync_vm_fence(struct amdgpu_sync *sync, struct dma_fence *fence)
  *
  * @sync: sync object to add fences from reservation object to
  * @resv: reservation object with embedded fence
- * @explicit_sync: true if we should only sync to the exclusive fence
+ * @mode: how owner affects which fences we sync to
+ * @owner: owner of the planned job submission
  *
  * Sync to the fence
  */
-int amdgpu_sync_resv(struct amdgpu_device *adev,
-		     struct amdgpu_sync *sync,
-		     struct dma_resv *resv,
-		     void *owner, bool explicit_sync)
+int amdgpu_sync_resv(struct amdgpu_device *adev, struct amdgpu_sync *sync,
+		     struct dma_resv *resv, enum amdgpu_sync_mode mode,
+		     void *owner)
 {
 	struct dma_resv_list *flist;
 	struct dma_fence *f;
-	void *fence_owner;
 	unsigned i;
 	int r = 0;
 
@@ -229,30 +228,46 @@ int amdgpu_sync_resv(struct amdgpu_device *adev,
 		return r;
 
 	for (i = 0; i < flist->shared_count; ++i) {
+		void *fence_owner;
+
 		f = rcu_dereference_protected(flist->shared[i],
 					      dma_resv_held(resv));
+
+		fence_owner = amdgpu_sync_get_owner(f);
+
+		/* Always sync to moves, no matter what */
+		if (fence_owner == AMDGPU_FENCE_OWNER_UNDEFINED) {
+			r = amdgpu_sync_fence(sync, f, false);
+			if (r)
+				break;
+		}
+
 		/* We only want to trigger KFD eviction fences on
 		 * evict or move jobs. Skip KFD fences otherwise.
 		 */
-		fence_owner = amdgpu_sync_get_owner(f);
 		if (fence_owner == AMDGPU_FENCE_OWNER_KFD &&
 		    owner != AMDGPU_FENCE_OWNER_UNDEFINED)
 			continue;
 
-		if (amdgpu_sync_same_dev(adev, f)) {
-			/* VM updates only sync with moves but not with user
-			 * command submissions or KFD evictions fences
-			 */
-			if (owner == AMDGPU_FENCE_OWNER_VM &&
-			    fence_owner != AMDGPU_FENCE_OWNER_UNDEFINED)
+		/* Ignore fences depending on the sync mode */
+		switch (mode) {
+		case AMDGPU_SYNC_ALWAYS:
+			break;
+
+		case AMDGPU_SYNC_NE_OWNER:
+			if (amdgpu_sync_same_dev(adev, f) &&
+			    fence_owner == owner)
 				continue;
+			break;
 
-			/* Ignore fence from the same owner and explicit one as
-			 * long as it isn't undefined.
-			 */
-			if (owner != AMDGPU_FENCE_OWNER_UNDEFINED &&
-			    (fence_owner == owner || explicit_sync))
+		case AMDGPU_SYNC_EQ_OWNER:
+			if (amdgpu_sync_same_dev(adev, f) &&
+			    fence_owner != owner)
 				continue;
+			break;
+
+		case AMDGPU_SYNC_EXPLICIT:
+			continue;
 		}
 
 		r = amdgpu_sync_fence(sync, f, false);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.h
index d62c2b81d92b..cfbe5788b8b9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.h
@@ -31,6 +31,13 @@ struct dma_resv;
 struct amdgpu_device;
 struct amdgpu_ring;
 
+enum amdgpu_sync_mode {
+	AMDGPU_SYNC_ALWAYS,
+	AMDGPU_SYNC_NE_OWNER,
+	AMDGPU_SYNC_EQ_OWNER,
+	AMDGPU_SYNC_EXPLICIT
+};
+
 /*
  * Container for fences used to sync command submissions.
  */
@@ -43,11 +50,9 @@ void amdgpu_sync_create(struct amdgpu_sync *sync);
 int amdgpu_sync_fence(struct amdgpu_sync *sync, struct dma_fence *f,
 		      bool explicit);
 int amdgpu_sync_vm_fence(struct amdgpu_sync *sync, struct dma_fence *fence);
-int amdgpu_sync_resv(struct amdgpu_device *adev,
-		     struct amdgpu_sync *sync,
-		     struct dma_resv *resv,
-		     void *owner,
-		     bool explicit_sync);
+int amdgpu_sync_resv(struct amdgpu_device *adev, struct amdgpu_sync *sync,
+		     struct dma_resv *resv, enum amdgpu_sync_mode mode,
+		     void *owner);
 struct dma_fence *amdgpu_sync_peek_fence(struct amdgpu_sync *sync,
 				     struct amdgpu_ring *ring);
 struct dma_fence *amdgpu_sync_get_fence(struct amdgpu_sync *sync,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 3ab46d4647e4..fe131c21e8a3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -60,15 +60,14 @@
 #include "amdgpu_ras.h"
 #include "bif/bif_4_1_d.h"
 
+#define AMDGPU_TTM_VRAM_MAX_DW_READ	(size_t)128
+
 static int amdgpu_map_buffer(struct ttm_buffer_object *bo,
 			     struct ttm_mem_reg *mem, unsigned num_pages,
 			     uint64_t offset, unsigned window,
 			     struct amdgpu_ring *ring,
 			     uint64_t *addr);
 
-static int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev);
-static void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev);
-
 /**
  * amdgpu_init_mem_type - Initialize a memory manager for a specific type of
  * memory request.
@@ -1591,7 +1590,7 @@ static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo,
 
 	while (len && pos < adev->gmc.mc_vram_size) {
 		uint64_t aligned_pos = pos & ~(uint64_t)3;
-		uint32_t bytes = 4 - (pos & 3);
+		uint64_t bytes = 4 - (pos & 3);
 		uint32_t shift = (pos & 3) * 8;
 		uint32_t mask = 0xffffffff << shift;
 
@@ -1600,20 +1599,28 @@ static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo,
 			bytes = len;
 		}
 
-		spin_lock_irqsave(&adev->mmio_idx_lock, flags);
-		WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)aligned_pos) | 0x80000000);
-		WREG32_NO_KIQ(mmMM_INDEX_HI, aligned_pos >> 31);
-		if (!write || mask != 0xffffffff)
-			value = RREG32_NO_KIQ(mmMM_DATA);
-		if (write) {
-			value &= ~mask;
-			value |= (*(uint32_t *)buf << shift) & mask;
-			WREG32_NO_KIQ(mmMM_DATA, value);
-		}
-		spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
-		if (!write) {
-			value = (value & mask) >> shift;
-			memcpy(buf, &value, bytes);
+		if (mask != 0xffffffff) {
+			spin_lock_irqsave(&adev->mmio_idx_lock, flags);
+			WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)aligned_pos) | 0x80000000);
+			WREG32_NO_KIQ(mmMM_INDEX_HI, aligned_pos >> 31);
+			if (!write || mask != 0xffffffff)
+				value = RREG32_NO_KIQ(mmMM_DATA);
+			if (write) {
+				value &= ~mask;
+				value |= (*(uint32_t *)buf << shift) & mask;
+				WREG32_NO_KIQ(mmMM_DATA, value);
+			}
+			spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
+			if (!write) {
+				value = (value & mask) >> shift;
+				memcpy(buf, &value, bytes);
+			}
+		} else {
+			bytes = (nodes->start + nodes->size) << PAGE_SHIFT;
+			bytes = min(bytes - pos, (uint64_t)len & ~0x3ull);
+
+			amdgpu_device_vram_access(adev, pos, (uint32_t *)buf,
+						  bytes, write);
 		}
 
 		ret += bytes;
@@ -1905,12 +1912,6 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		return r;
 	}
 
-	/* Register debugfs entries for amdgpu_ttm */
-	r = amdgpu_ttm_debugfs_init(adev);
-	if (r) {
-		DRM_ERROR("Failed to init debugfs\n");
-		return r;
-	}
 	return 0;
 }
 
@@ -1932,7 +1933,6 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 	if (!adev->mman.initialized)
 		return;
 
-	amdgpu_ttm_debugfs_fini(adev);
 	amdgpu_ttm_training_reserve_vram_fini(adev);
 	/* return the IP Discovery TMR memory back to VRAM */
 	amdgpu_bo_free_kernel(&adev->discovery_memory, NULL, NULL);
@@ -2107,8 +2107,8 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset,
 	}
 	if (resv) {
 		r = amdgpu_sync_resv(adev, &job->sync, resv,
-				     AMDGPU_FENCE_OWNER_UNDEFINED,
-				     false);
+				     AMDGPU_SYNC_ALWAYS,
+				     AMDGPU_FENCE_OWNER_UNDEFINED);
 		if (r) {
 			DRM_ERROR("sync failed (%d).\n", r);
 			goto error_free;
@@ -2192,7 +2192,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 
 	if (resv) {
 		r = amdgpu_sync_resv(adev, &job->sync, resv,
-				     AMDGPU_FENCE_OWNER_UNDEFINED, false);
+				     AMDGPU_SYNC_ALWAYS,
+				     AMDGPU_FENCE_OWNER_UNDEFINED);
 		if (r) {
 			DRM_ERROR("sync failed (%d).\n", r);
 			goto error_free;
@@ -2273,7 +2274,6 @@ static ssize_t amdgpu_ttm_vram_read(struct file *f, char __user *buf,
 {
 	struct amdgpu_device *adev = file_inode(f)->i_private;
 	ssize_t result = 0;
-	int r;
 
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
@@ -2281,27 +2281,19 @@ static ssize_t amdgpu_ttm_vram_read(struct file *f, char __user *buf,
 	if (*pos >= adev->gmc.mc_vram_size)
 		return -ENXIO;
 
+	size = min(size, (size_t)(adev->gmc.mc_vram_size - *pos));
 	while (size) {
-		unsigned long flags;
-		uint32_t value;
-
-		if (*pos >= adev->gmc.mc_vram_size)
-			return result;
+		size_t bytes = min(size, AMDGPU_TTM_VRAM_MAX_DW_READ * 4);
+		uint32_t value[AMDGPU_TTM_VRAM_MAX_DW_READ];
 
-		spin_lock_irqsave(&adev->mmio_idx_lock, flags);
-		WREG32_NO_KIQ(mmMM_INDEX, ((uint32_t)*pos) | 0x80000000);
-		WREG32_NO_KIQ(mmMM_INDEX_HI, *pos >> 31);
-		value = RREG32_NO_KIQ(mmMM_DATA);
-		spin_unlock_irqrestore(&adev->mmio_idx_lock, flags);
-
-		r = put_user(value, (uint32_t *)buf);
-		if (r)
-			return r;
+		amdgpu_device_vram_access(adev, *pos, value, bytes, false);
+		if (copy_to_user(buf, value, bytes))
+			return -EFAULT;
 
-		result += 4;
-		buf += 4;
-		*pos += 4;
-		size -= 4;
+		result += bytes;
+		buf += bytes;
+		*pos += bytes;
+		size -= bytes;
 	}
 
 	return result;
@@ -2538,7 +2530,7 @@ static const struct {
 
 #endif
 
-static int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev)
+int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev)
 {
 #if defined(CONFIG_DEBUG_FS)
 	unsigned count;
@@ -2574,7 +2566,7 @@ static int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev)
 #endif
 }
 
-static void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev)
+void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev)
 {
 #if defined(CONFIG_DEBUG_FS)
 	unsigned i;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index 0dddedc06ae3..7551f3729445 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -133,4 +133,7 @@ uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_mem_reg *mem);
 uint64_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm,
 				 struct ttm_mem_reg *mem);
 
+int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev);
+void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
index a92f3b18e657..5fd32ad1c575 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
@@ -1099,7 +1099,8 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo,
 			goto err_free;
 	} else {
 		r = amdgpu_sync_resv(adev, &job->sync, bo->tbo.base.resv,
-				     AMDGPU_FENCE_OWNER_UNDEFINED, false);
+				     AMDGPU_SYNC_ALWAYS,
+				     AMDGPU_FENCE_OWNER_UNDEFINED);
 		if (r)
 			goto err_free;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index d16231d6a790..f44c26c373a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -120,23 +120,17 @@ static inline void amdgpu_vm_eviction_unlock(struct amdgpu_vm *vm)
 static unsigned amdgpu_vm_level_shift(struct amdgpu_device *adev,
 				      unsigned level)
 {
-	unsigned shift = 0xff;
-
 	switch (level) {
 	case AMDGPU_VM_PDB2:
 	case AMDGPU_VM_PDB1:
 	case AMDGPU_VM_PDB0:
-		shift = 9 * (AMDGPU_VM_PDB0 - level) +
+		return 9 * (AMDGPU_VM_PDB0 - level) +
 			adev->vm_manager.block_size;
-		break;
 	case AMDGPU_VM_PTB:
-		shift = 0;
-		break;
+		return 0;
 	default:
-		dev_err(adev->dev, "the level%d isn't supported.\n", level);
+		return ~0;
 	}
-
-	return shift;
 }
 
 /**
@@ -235,19 +229,6 @@ static void amdgpu_vm_bo_evicted(struct amdgpu_vm_bo_base *vm_bo)
 	else
 		list_move_tail(&vm_bo->vm_status, &vm->evicted);
 }
-
-/**
- * amdgpu_vm_bo_relocated - vm_bo is reloacted
- *
- * @vm_bo: vm_bo which is relocated
- *
- * State for PDs/PTs which needs to update their parent PD.
- */
-static void amdgpu_vm_bo_relocated(struct amdgpu_vm_bo_base *vm_bo)
-{
-	list_move(&vm_bo->vm_status, &vm_bo->vm->relocated);
-}
-
 /**
  * amdgpu_vm_bo_moved - vm_bo is moved
  *
@@ -291,6 +272,22 @@ static void amdgpu_vm_bo_invalidated(struct amdgpu_vm_bo_base *vm_bo)
 }
 
 /**
+ * amdgpu_vm_bo_relocated - vm_bo is reloacted
+ *
+ * @vm_bo: vm_bo which is relocated
+ *
+ * State for PDs/PTs which needs to update their parent PD.
+ * For the root PD, just move to idle state.
+ */
+static void amdgpu_vm_bo_relocated(struct amdgpu_vm_bo_base *vm_bo)
+{
+	if (vm_bo->bo->parent)
+		list_move(&vm_bo->vm_status, &vm_bo->vm->relocated);
+	else
+		amdgpu_vm_bo_idle(vm_bo);
+}
+
+/**
  * amdgpu_vm_bo_done - vm_bo is done
  *
  * @vm_bo: vm_bo which is now done
@@ -588,8 +585,8 @@ void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm,
 {
 	entry->priority = 0;
 	entry->tv.bo = &vm->root.base.bo->tbo;
-	/* One for TTM and one for the CS job */
-	entry->tv.num_shared = 2;
+	/* Two for VM updates, one for TTM and one for the CS job */
+	entry->tv.num_shared = 4;
 	entry->user_pages = NULL;
 	list_add(&entry->tv.head, validated);
 }
@@ -697,10 +694,7 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			amdgpu_vm_bo_moved(bo_base);
 		} else {
 			vm->update_funcs->map_table(bo);
-			if (bo->parent)
-				amdgpu_vm_bo_relocated(bo_base);
-			else
-				amdgpu_vm_bo_idle(bo_base);
+			amdgpu_vm_bo_relocated(bo_base);
 		}
 	}
 
@@ -803,7 +797,7 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 	params.vm = vm;
 	params.direct = direct;
 
-	r = vm->update_funcs->prepare(&params, AMDGPU_FENCE_OWNER_KFD, NULL);
+	r = vm->update_funcs->prepare(&params, NULL, AMDGPU_SYNC_EXPLICIT);
 	if (r)
 		return r;
 
@@ -1299,7 +1293,7 @@ int amdgpu_vm_update_pdes(struct amdgpu_device *adev,
 	params.vm = vm;
 	params.direct = direct;
 
-	r = vm->update_funcs->prepare(&params, AMDGPU_FENCE_OWNER_VM, NULL);
+	r = vm->update_funcs->prepare(&params, NULL, AMDGPU_SYNC_EXPLICIT);
 	if (r)
 		return r;
 
@@ -1448,21 +1442,14 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 		uint64_t incr, entry_end, pe_start;
 		struct amdgpu_bo *pt;
 
-		/* make sure that the page tables covering the address range are
-		 * actually allocated
-		 */
-		r = amdgpu_vm_alloc_pts(params->adev, params->vm, &cursor,
-					params->direct);
-		if (r)
-			return r;
-
-		pt = cursor.entry->base.bo;
-
-		/* The root level can't be a huge page */
-		if (cursor.level == adev->vm_manager.root_level) {
-			if (!amdgpu_vm_pt_descendant(adev, &cursor))
-				return -ENOENT;
-			continue;
+		if (flags & AMDGPU_PTE_VALID) {
+			/* make sure that the page tables covering the
+			 * address range are actually allocated
+			 */
+			r = amdgpu_vm_alloc_pts(params->adev, params->vm,
+						&cursor, params->direct);
+			if (r)
+				return r;
 		}
 
 		shift = amdgpu_vm_level_shift(adev, cursor.level);
@@ -1480,25 +1467,38 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 			 * smaller than the address shift. Go to the next
 			 * child entry and try again.
 			 */
-			if (!amdgpu_vm_pt_descendant(adev, &cursor))
-				return -ENOENT;
-			continue;
-		} else if (frag >= parent_shift &&
-			   cursor.level - 1 != adev->vm_manager.root_level) {
+			if (amdgpu_vm_pt_descendant(adev, &cursor))
+				continue;
+		} else if (frag >= parent_shift) {
 			/* If the fragment size is even larger than the parent
-			 * shift we should go up one level and check it again
-			 * unless one level up is the root level.
+			 * shift we should go up one level and check it again.
 			 */
 			if (!amdgpu_vm_pt_ancestor(&cursor))
-				return -ENOENT;
+				return -EINVAL;
 			continue;
 		}
 
+		pt = cursor.entry->base.bo;
+		if (!pt) {
+			/* We need all PDs and PTs for mapping something, */
+			if (flags & AMDGPU_PTE_VALID)
+				return -ENOENT;
+
+			/* but unmapping something can happen at a higher
+			 * level.
+			 */
+			if (!amdgpu_vm_pt_ancestor(&cursor))
+				return -EINVAL;
+
+			pt = cursor.entry->base.bo;
+			shift = parent_shift;
+		}
+
 		/* Looks good so far, calculate parameters for the update */
 		incr = (uint64_t)AMDGPU_GPU_PAGE_SIZE << shift;
 		mask = amdgpu_vm_entries_mask(adev, cursor.level);
 		pe_start = ((cursor.pfn >> shift) & mask) * 8;
-		entry_end = (uint64_t)(mask + 1) << shift;
+		entry_end = ((uint64_t)mask + 1) << shift;
 		entry_end += cursor.pfn & ~(entry_end - 1);
 		entry_end = min(entry_end, end);
 
@@ -1506,6 +1506,10 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 			uint64_t upd_end = min(entry_end, frag_end);
 			unsigned nptes = (upd_end - frag_start) >> shift;
 
+			/* This can happen when we set higher level PDs to
+			 * silent to stop fault floods.
+			 */
+			nptes = max(nptes, 1u);
 			amdgpu_vm_update_flags(params, pt, cursor.level,
 					       pe_start, dst, nptes, incr,
 					       flags | AMDGPU_PTE_FRAG(frag));
@@ -1550,7 +1554,7 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
  * @adev: amdgpu_device pointer
  * @vm: requested vm
  * @direct: direct submission in a page fault
- * @exclusive: fence we need to sync to
+ * @resv: fences we need to sync to
  * @start: start of mapped range
  * @last: last mapped entry
  * @flags: flags for the entries
@@ -1565,14 +1569,14 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
  */
 static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
 				       struct amdgpu_vm *vm, bool direct,
-				       struct dma_fence *exclusive,
+				       struct dma_resv *resv,
 				       uint64_t start, uint64_t last,
 				       uint64_t flags, uint64_t addr,
 				       dma_addr_t *pages_addr,
 				       struct dma_fence **fence)
 {
 	struct amdgpu_vm_update_params params;
-	void *owner = AMDGPU_FENCE_OWNER_VM;
+	enum amdgpu_sync_mode sync_mode;
 	int r;
 
 	memset(&params, 0, sizeof(params));
@@ -1581,9 +1585,13 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
 	params.direct = direct;
 	params.pages_addr = pages_addr;
 
-	/* sync to everything except eviction fences on unmapping */
+	/* Implicitly sync to command submissions in the same VM before
+	 * unmapping. Sync to moving fences before mapping.
+	 */
 	if (!(flags & AMDGPU_PTE_VALID))
-		owner = AMDGPU_FENCE_OWNER_KFD;
+		sync_mode = AMDGPU_SYNC_EQ_OWNER;
+	else
+		sync_mode = AMDGPU_SYNC_EXPLICIT;
 
 	amdgpu_vm_eviction_lock(vm);
 	if (vm->evicting) {
@@ -1591,7 +1599,17 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
 		goto error_unlock;
 	}
 
-	r = vm->update_funcs->prepare(&params, owner, exclusive);
+	if (flags & AMDGPU_PTE_VALID) {
+		struct amdgpu_bo *root = vm->root.base.bo;
+
+		if (!dma_fence_is_signaled(vm->last_direct))
+			amdgpu_bo_fence(root, vm->last_direct, true);
+
+		if (!dma_fence_is_signaled(vm->last_delayed))
+			amdgpu_bo_fence(root, vm->last_delayed, true);
+	}
+
+	r = vm->update_funcs->prepare(&params, resv, sync_mode);
 	if (r)
 		goto error_unlock;
 
@@ -1610,7 +1628,7 @@ error_unlock:
  * amdgpu_vm_bo_split_mapping - split a mapping into smaller chunks
  *
  * @adev: amdgpu_device pointer
- * @exclusive: fence we need to sync to
+ * @resv: fences we need to sync to
  * @pages_addr: DMA addresses to use for mapping
  * @vm: requested vm
  * @mapping: mapped range and flags to use for the update
@@ -1626,7 +1644,7 @@ error_unlock:
  * 0 for success, -EINVAL for failure.
  */
 static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
-				      struct dma_fence *exclusive,
+				      struct dma_resv *resv,
 				      dma_addr_t *pages_addr,
 				      struct amdgpu_vm *vm,
 				      struct amdgpu_bo_va_mapping *mapping,
@@ -1702,7 +1720,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 		}
 
 		last = min((uint64_t)mapping->last, start + max_entries - 1);
-		r = amdgpu_vm_bo_update_mapping(adev, vm, false, exclusive,
+		r = amdgpu_vm_bo_update_mapping(adev, vm, false, resv,
 						start, last, flags, addr,
 						dma_addr, fence);
 		if (r)
@@ -1741,7 +1759,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 	dma_addr_t *pages_addr = NULL;
 	struct ttm_mem_reg *mem;
 	struct drm_mm_node *nodes;
-	struct dma_fence *exclusive, **last_update;
+	struct dma_fence **last_update;
+	struct dma_resv *resv;
 	uint64_t flags;
 	struct amdgpu_device *bo_adev = adev;
 	int r;
@@ -1749,7 +1768,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 	if (clear || !bo) {
 		mem = NULL;
 		nodes = NULL;
-		exclusive = NULL;
+		resv = vm->root.base.bo->tbo.base.resv;
 	} else {
 		struct ttm_dma_tt *ttm;
 
@@ -1759,7 +1778,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 			ttm = container_of(bo->tbo.ttm, struct ttm_dma_tt, ttm);
 			pages_addr = ttm->dma_address;
 		}
-		exclusive = bo->tbo.moving;
+		resv = bo->tbo.base.resv;
 	}
 
 	if (bo) {
@@ -1769,7 +1788,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 		flags = 0x0;
 	}
 
-	if (clear || (bo && bo->tbo.base.resv == vm->root.base.bo->tbo.base.resv))
+	if (clear || (bo && bo->tbo.base.resv ==
+		      vm->root.base.bo->tbo.base.resv))
 		last_update = &vm->last_update;
 	else
 		last_update = &bo_va->last_pt_update;
@@ -1783,7 +1803,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 	}
 
 	list_for_each_entry(mapping, &bo_va->invalids, list) {
-		r = amdgpu_vm_bo_split_mapping(adev, exclusive, pages_addr, vm,
+		r = amdgpu_vm_bo_split_mapping(adev, resv, pages_addr, vm,
 					       mapping, flags, bo_adev, nodes,
 					       last_update);
 		if (r)
@@ -1978,6 +1998,7 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 			  struct amdgpu_vm *vm,
 			  struct dma_fence **fence)
 {
+	struct dma_resv *resv = vm->root.base.bo->tbo.base.resv;
 	struct amdgpu_bo_va_mapping *mapping;
 	uint64_t init_pte_value = 0;
 	struct dma_fence *f = NULL;
@@ -1992,7 +2013,7 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 		    mapping->start < AMDGPU_GMC_HOLE_START)
 			init_pte_value = AMDGPU_PTE_DEFAULT_ATC;
 
-		r = amdgpu_vm_bo_update_mapping(adev, vm, false, NULL,
+		r = amdgpu_vm_bo_update_mapping(adev, vm, false, resv,
 						mapping->start, mapping->last,
 						init_pte_value, 0, NULL, &f);
 		amdgpu_vm_free_mapping(adev, vm, mapping, f);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index b4640ab38c95..d00648ee8d54 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -227,8 +227,8 @@ struct amdgpu_vm_update_params {
 
 struct amdgpu_vm_update_funcs {
 	int (*map_table)(struct amdgpu_bo *bo);
-	int (*prepare)(struct amdgpu_vm_update_params *p, void * owner,
-		       struct dma_fence *exclusive);
+	int (*prepare)(struct amdgpu_vm_update_params *p, struct dma_resv *resv,
+		       enum amdgpu_sync_mode sync_mode);
 	int (*update)(struct amdgpu_vm_update_params *p,
 		      struct amdgpu_bo *bo, uint64_t pe, uint64_t addr,
 		      unsigned count, uint32_t incr, uint64_t flags);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
index 73fec7a0ced5..e38516304070 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
@@ -44,26 +44,14 @@ static int amdgpu_vm_cpu_map_table(struct amdgpu_bo *table)
  * Returns:
  * Negativ errno, 0 for success.
  */
-static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p, void *owner,
-				 struct dma_fence *exclusive)
+static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p,
+				 struct dma_resv *resv,
+				 enum amdgpu_sync_mode sync_mode)
 {
-	int r;
-
-	/* Wait for any BO move to be completed */
-	if (exclusive) {
-		r = dma_fence_wait(exclusive, true);
-		if (unlikely(r))
-			return r;
-	}
-
-	/* Don't wait for submissions during page fault */
-	if (p->direct)
+	if (!resv)
 		return 0;
 
-	/* Wait for PT BOs to be idle. PTs share the same resv. object
-	 * as the root PD BO
-	 */
-	return amdgpu_bo_sync_wait(p->vm->root.base.bo, owner, true);
+	return amdgpu_bo_sync_wait_resv(p->adev, resv, sync_mode, p->vm, true);
 }
 
 /**
@@ -86,6 +74,13 @@ static int amdgpu_vm_cpu_update(struct amdgpu_vm_update_params *p,
 {
 	unsigned int i;
 	uint64_t value;
+	int r;
+
+	if (bo->tbo.moving) {
+		r = dma_fence_wait(bo->tbo.moving, true);
+		if (r)
+			return r;
+	}
 
 	pe += (unsigned long)amdgpu_bo_kptr(bo);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
index 19b7f80758f1..4cc7881f438c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
@@ -58,9 +58,9 @@ static int amdgpu_vm_sdma_map_table(struct amdgpu_bo *table)
  * Negativ errno, 0 for success.
  */
 static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p,
-				  void *owner, struct dma_fence *exclusive)
+				  struct dma_resv *resv,
+				  enum amdgpu_sync_mode sync_mode)
 {
-	struct amdgpu_bo *root = p->vm->root.base.bo;
 	unsigned int ndw = AMDGPU_VM_SDMA_MIN_NUM_DW;
 	int r;
 
@@ -70,17 +70,10 @@ static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p,
 
 	p->num_dw_left = ndw;
 
-	/* Wait for moves to be completed */
-	r = amdgpu_sync_fence(&p->job->sync, exclusive, false);
-	if (r)
-		return r;
-
-	/* Don't wait for any submissions during page fault handling */
-	if (p->direct)
+	if (!resv)
 		return 0;
 
-	return amdgpu_sync_resv(p->adev, &p->job->sync, root->tbo.base.resv,
-				owner, false);
+	return amdgpu_sync_resv(p->adev, &p->job->sync, resv, sync_mode, p->vm);
 }
 
 /**
@@ -147,7 +140,7 @@ static void amdgpu_vm_sdma_copy_ptes(struct amdgpu_vm_update_params *p,
 
 	src += p->num_dw_left * 4;
 
-	pe += amdgpu_bo_gpu_offset(bo);
+	pe += amdgpu_gmc_sign_extend(bo->tbo.offset);
 	trace_amdgpu_vm_copy_ptes(pe, src, count, p->direct);
 
 	amdgpu_vm_copy_pte(p->adev, ib, pe, src, count);
@@ -174,7 +167,7 @@ static void amdgpu_vm_sdma_set_ptes(struct amdgpu_vm_update_params *p,
 {
 	struct amdgpu_ib *ib = p->job->ibs;
 
-	pe += amdgpu_bo_gpu_offset(bo);
+	pe += amdgpu_gmc_sign_extend(bo->tbo.offset);
 	trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags, p->direct);
 	if (count < 3) {
 		amdgpu_vm_write_pte(p->adev, ib, pe, addr | flags,
@@ -208,6 +201,11 @@ static int amdgpu_vm_sdma_update(struct amdgpu_vm_update_params *p,
 	uint64_t *pte;
 	int r;
 
+	/* Wait for PD/PT moves to be completed */
+	r = amdgpu_sync_fence(&p->job->sync, bo->tbo.moving, false);
+	if (r)
+		return r;
+
 	do {
 		ndw = p->num_dw_left;
 		ndw -= p->job->ibs->length_dw;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index a97af422575a..7a89c91f7b80 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -365,6 +365,13 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 		return 0;
 
 	if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_PSP)) {
+		ret = psp_xgmi_initialize(&adev->psp);
+		if (ret) {
+			dev_err(adev->dev,
+				"XGMI: Failed to initialize xgmi session\n");
+			return ret;
+		}
+
 		ret = psp_xgmi_get_hive_id(&adev->psp, &adev->gmc.xgmi.hive_id);
 		if (ret) {
 			dev_err(adev->dev,
@@ -451,16 +458,16 @@ exit:
 	return ret;
 }
 
-void amdgpu_xgmi_remove_device(struct amdgpu_device *adev)
+int amdgpu_xgmi_remove_device(struct amdgpu_device *adev)
 {
 	struct amdgpu_hive_info *hive;
 
 	if (!adev->gmc.xgmi.supported)
-		return;
+		return -EINVAL;
 
 	hive = amdgpu_get_xgmi_hive(adev, 1);
 	if (!hive)
-		return;
+		return -EINVAL;
 
 	if (!(hive->number_devices--)) {
 		amdgpu_xgmi_sysfs_destroy(adev, hive);
@@ -471,6 +478,8 @@ void amdgpu_xgmi_remove_device(struct amdgpu_device *adev)
 		amdgpu_xgmi_sysfs_rem_dev_info(adev, hive);
 		mutex_unlock(&hive->hive_lock);
 	}
+
+	return psp_xgmi_terminate(&adev->psp);
 }
 
 int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev)
@@ -521,3 +530,33 @@ void amdgpu_xgmi_ras_fini(struct amdgpu_device *adev)
 		kfree(ras_if);
 	}
 }
+
+uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev,
+					   uint64_t addr)
+{
+	uint32_t df_inst_id;
+	uint64_t dram_base_addr = 0;
+	const struct amdgpu_df_funcs *df_funcs = adev->df.funcs;
+
+	if ((!df_funcs)                 ||
+	    (!df_funcs->get_df_inst_id) ||
+	    (!df_funcs->get_dram_base_addr)) {
+		dev_warn(adev->dev,
+			 "XGMI: relative phy_addr algorithm is not supported\n");
+		return addr;
+	}
+
+	if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) {
+		dev_warn(adev->dev,
+			 "failed to disable DF-Cstate, DF register may not be accessible\n");
+		return addr;
+	}
+
+	df_inst_id = df_funcs->get_df_inst_id(adev);
+	dram_base_addr = df_funcs->get_dram_base_addr(adev, df_inst_id);
+
+	if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW))
+		dev_warn(adev->dev, "failed to enable DF-Cstate\n");
+
+	return addr + dram_base_addr;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
index 74011fbc2251..2aa61adee459 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
@@ -40,12 +40,14 @@ struct amdgpu_hive_info {
 struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lock);
 int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev);
 int amdgpu_xgmi_add_device(struct amdgpu_device *adev);
-void amdgpu_xgmi_remove_device(struct amdgpu_device *adev);
+int amdgpu_xgmi_remove_device(struct amdgpu_device *adev);
 int amdgpu_xgmi_set_pstate(struct amdgpu_device *adev, int pstate);
 int amdgpu_xgmi_get_hops_count(struct amdgpu_device *adev,
 		struct amdgpu_device *peer_adev);
 int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev);
 void amdgpu_xgmi_ras_fini(struct amdgpu_device *adev);
+uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev,
+					   uint64_t addr);
 
 static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev,
 		struct amdgpu_device *bo_adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
index ea702a64f807..9b74cfdba7b8 100644
--- a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
+++ b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
@@ -186,16 +186,10 @@ amdgpu_atombios_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *m
 
 void amdgpu_atombios_dp_aux_init(struct amdgpu_connector *amdgpu_connector)
 {
-	int ret;
-
 	amdgpu_connector->ddc_bus->rec.hpd = amdgpu_connector->hpd.hpd;
-	amdgpu_connector->ddc_bus->aux.dev = amdgpu_connector->base.kdev;
 	amdgpu_connector->ddc_bus->aux.transfer = amdgpu_atombios_dp_aux_transfer;
-	ret = drm_dp_aux_register(&amdgpu_connector->ddc_bus->aux);
-	if (!ret)
-		amdgpu_connector->ddc_bus->has_aux = true;
-
-	WARN(ret, "drm_dp_aux_register_i2c_bus() failed with error %d\n", ret);
+	drm_dp_aux_init(&amdgpu_connector->ddc_bus->aux);
+	amdgpu_connector->ddc_bus->has_aux = true;
 }
 
 /***** general DP utility functions *****/
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
index 5707e4ad1a62..13e12be667fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c
@@ -614,7 +614,6 @@ static int dce_virtual_connector_encoder_init(struct amdgpu_device *adev,
 	connector->display_info.subpixel_order = SubPixelHorizontalRGB;
 	connector->interlace_allowed = false;
 	connector->doublescan_allowed = false;
-	drm_connector_register(connector);
 
 	/* link them */
 	drm_connector_attach_encoder(connector, encoder);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 1785fdad6ecb..8b17f182d532 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -3164,12 +3164,7 @@ static int gfx_v10_0_kiq_enable_kgq(struct amdgpu_device *adev)
 	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
 		kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.gfx_ring[i]);
 
-	r = amdgpu_ring_test_ring(kiq_ring);
-	if (r) {
-		DRM_ERROR("kfq enable failed\n");
-		kiq_ring->sched.ready = false;
-	}
-	return r;
+	return amdgpu_ring_test_helper(kiq_ring);
 }
 #endif
 
@@ -3785,7 +3780,7 @@ static int gfx_v10_0_kiq_disable_kgq(struct amdgpu_device *adev)
 		kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.gfx_ring[i],
 					   PREEMPT_QUEUES, 0, 0);
 
-	return amdgpu_ring_test_ring(kiq_ring);
+	return amdgpu_ring_test_helper(kiq_ring);
 }
 #endif
 
@@ -3923,11 +3918,13 @@ static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 {
 	uint64_t clock;
 
+	amdgpu_gfx_off_ctrl(adev, false);
 	mutex_lock(&adev->gfx.gpu_clock_mutex);
 	WREG32_SOC15(GC, 0, mmRLC_CAPTURE_GPU_CLOCK_COUNT, 1);
 	clock = (uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_LSB) |
 		((uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
 	mutex_unlock(&adev->gfx.gpu_clock_mutex);
+	amdgpu_gfx_off_ctrl(adev, true);
 	return clock;
 }
 
@@ -4411,7 +4408,7 @@ static void gfx_v10_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 
 	control |= ib->length_dw | (vmid << 24);
 
-	if (amdgpu_mcbp && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) {
+	if ((amdgpu_sriov_vf(ring->adev) || amdgpu_mcbp) && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) {
 		control |= INDIRECT_BUFFER_PRE_ENB(1);
 
 		if (flags & AMDGPU_IB_PREEMPTED)
@@ -4419,7 +4416,7 @@ static void gfx_v10_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 
 		if (!(ib->flags & AMDGPU_IB_FLAG_CE))
 			gfx_v10_0_ring_emit_de_meta(ring,
-				    flags & AMDGPU_IB_PREEMPTED ? true : false);
+				    (!amdgpu_sriov_vf(ring->adev) && flags & AMDGPU_IB_PREEMPTED) ? true : false);
 	}
 
 	amdgpu_ring_write(ring, header);
@@ -4566,9 +4563,9 @@ static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flag
 {
 	uint32_t dw2 = 0;
 
-	if (amdgpu_mcbp)
+	if (amdgpu_mcbp || amdgpu_sriov_vf(ring->adev))
 		gfx_v10_0_ring_emit_ce_meta(ring,
-				    flags & AMDGPU_IB_PREEMPTED ? true : false);
+				    (!amdgpu_sriov_vf(ring->adev) && flags & AMDGPU_IB_PREEMPTED) ? true : false);
 
 	gfx_v10_0_ring_emit_tmz(ring, true);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index b33a4eb39193..6a49fe452f29 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1106,10 +1106,11 @@ static void gfx_v9_0_check_fw_write_wait(struct amdgpu_device *adev)
 	adev->gfx.me_fw_write_wait = false;
 	adev->gfx.mec_fw_write_wait = false;
 
-	if ((adev->gfx.mec_fw_version < 0x000001a5) ||
+	if ((adev->asic_type != CHIP_ARCTURUS) &&
+	    ((adev->gfx.mec_fw_version < 0x000001a5) ||
 	    (adev->gfx.mec_feature_version < 46) ||
 	    (adev->gfx.pfp_fw_version < 0x000000b7) ||
-	    (adev->gfx.pfp_feature_version < 46))
+	    (adev->gfx.pfp_feature_version < 46)))
 		DRM_WARN_ONCE("CP firmware version too old, please update!");
 
 	switch (adev->asic_type) {
@@ -1193,6 +1194,14 @@ static bool gfx_v9_0_should_disable_gfxoff(struct pci_dev *pdev)
 	return false;
 }
 
+static bool is_raven_kicker(struct amdgpu_device *adev)
+{
+	if (adev->pm.fw_version >= 0x41e2b)
+		return true;
+	else
+		return false;
+}
+
 static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev)
 {
 	if (gfx_v9_0_should_disable_gfxoff(adev->pdev))
@@ -1205,9 +1214,8 @@ static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev)
 		break;
 	case CHIP_RAVEN:
 		if (!(adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8) &&
-		    ((adev->gfx.rlc_fw_version != 106 &&
+		    ((!is_raven_kicker(adev) &&
 		      adev->gfx.rlc_fw_version < 531) ||
-		     (adev->gfx.rlc_fw_version == 53815) ||
 		     (adev->gfx.rlc_feature_version < 1) ||
 		     !adev->gfx.rlc.is_rlc_v2_1))
 			adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
@@ -3959,6 +3967,7 @@ static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 {
 	uint64_t clock;
 
+	amdgpu_gfx_off_ctrl(adev, false);
 	mutex_lock(&adev->gfx.gpu_clock_mutex);
 	if (adev->asic_type == CHIP_VEGA10 && amdgpu_sriov_runtime(adev)) {
 		uint32_t tmp, lsb, msb, i = 0;
@@ -3977,6 +3986,7 @@ static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 			((uint64_t)RREG32_SOC15(GC, 0, mmRLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
 	}
 	mutex_unlock(&adev->gfx.gpu_clock_mutex);
+	amdgpu_gfx_off_ctrl(adev, true);
 	return clock;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
index b70c7b483c24..e0654a216ab5 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
@@ -135,6 +135,10 @@ static void gfxhub_v2_0_init_cache_regs(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
 
+	/* These regs are not accessible for VF, PF will program these in SRIOV */
+	if (amdgpu_sriov_vf(adev))
+		return;
+
 	/* Setup L2 cache */
 	tmp = RREG32_SOC15(GC, 0, mmGCVM_L2_CNTL);
 	tmp = REG_SET_FIELD(tmp, GCVM_L2_CNTL, ENABLE_L2_CACHE, 1);
@@ -298,9 +302,11 @@ void gfxhub_v2_0_gart_disable(struct amdgpu_device *adev)
 			    ENABLE_ADVANCED_DRIVER_MODEL, 0);
 	WREG32_SOC15(GC, 0, mmGCMC_VM_MX_L1_TLB_CNTL, tmp);
 
-	/* Setup L2 cache */
-	WREG32_FIELD15(GC, 0, GCVM_L2_CNTL, ENABLE_L2_CACHE, 0);
-	WREG32_SOC15(GC, 0, mmGCVM_L2_CNTL3, 0);
+	if (!amdgpu_sriov_vf(adev)) {
+		/* Setup L2 cache */
+		WREG32_FIELD15(GC, 0, GCVM_L2_CNTL, ENABLE_L2_CACHE, 0);
+		WREG32_SOC15(GC, 0, mmGCVM_L2_CNTL3, 0);
+	}
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 90216abf14a4..10171acbf3e1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -476,13 +476,26 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 {
 	bool use_semaphore = gmc_v9_0_use_invalidate_semaphore(adev, vmhub);
 	const unsigned eng = 17;
-	u32 j, inv_req, tmp;
+	u32 j, inv_req, inv_req2, tmp;
 	struct amdgpu_vmhub *hub;
 
 	BUG_ON(vmhub >= adev->num_vmhubs);
 
 	hub = &adev->vmhub[vmhub];
-	inv_req = gmc_v9_0_get_invalidate_req(vmid, flush_type);
+	if (adev->gmc.xgmi.num_physical_nodes &&
+	    adev->asic_type == CHIP_VEGA20) {
+		/* Vega20+XGMI caches PTEs in TC and TLB. Add a
+		 * heavy-weight TLB flush (type 2), which flushes
+		 * both. Due to a race condition with concurrent
+		 * memory accesses using the same TLB cache line, we
+		 * still need a second TLB flush after this.
+		 */
+		inv_req = gmc_v9_0_get_invalidate_req(vmid, 2);
+		inv_req2 = gmc_v9_0_get_invalidate_req(vmid, flush_type);
+	} else {
+		inv_req = gmc_v9_0_get_invalidate_req(vmid, flush_type);
+		inv_req2 = 0;
+	}
 
 	/* This is necessary for a HW workaround under SRIOV as well
 	 * as GFXOFF under bare metal
@@ -521,21 +534,27 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 			DRM_ERROR("Timeout waiting for sem acquire in VM flush!\n");
 	}
 
-	WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, inv_req);
+	do {
+		WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, inv_req);
 
-	/*
-	 * Issue a dummy read to wait for the ACK register to be cleared
-	 * to avoid a false ACK due to the new fast GRBM interface.
-	 */
-	if (vmhub == AMDGPU_GFXHUB_0)
-		RREG32_NO_KIQ(hub->vm_inv_eng0_req + eng);
+		/*
+		 * Issue a dummy read to wait for the ACK register to
+		 * be cleared to avoid a false ACK due to the new fast
+		 * GRBM interface.
+		 */
+		if (vmhub == AMDGPU_GFXHUB_0)
+			RREG32_NO_KIQ(hub->vm_inv_eng0_req + eng);
 
-	for (j = 0; j < adev->usec_timeout; j++) {
-		tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng);
-		if (tmp & (1 << vmid))
-			break;
-		udelay(1);
-	}
+		for (j = 0; j < adev->usec_timeout; j++) {
+			tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng);
+			if (tmp & (1 << vmid))
+				break;
+			udelay(1);
+		}
+
+		inv_req = inv_req2;
+		inv_req2 = 0;
+	} while (inv_req);
 
 	/* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */
 	if (use_semaphore)
@@ -577,9 +596,26 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		return -EIO;
 
 	if (ring->sched.ready) {
+		/* Vega20+XGMI caches PTEs in TC and TLB. Add a
+		 * heavy-weight TLB flush (type 2), which flushes
+		 * both. Due to a race condition with concurrent
+		 * memory accesses using the same TLB cache line, we
+		 * still need a second TLB flush after this.
+		 */
+		bool vega20_xgmi_wa = (adev->gmc.xgmi.num_physical_nodes &&
+				       adev->asic_type == CHIP_VEGA20);
+		/* 2 dwords flush + 8 dwords fence */
+		unsigned int ndw = kiq->pmf->invalidate_tlbs_size + 8;
+
+		if (vega20_xgmi_wa)
+			ndw += kiq->pmf->invalidate_tlbs_size;
+
 		spin_lock(&adev->gfx.kiq.ring_lock);
 		/* 2 dwords flush + 8 dwords fence */
-		amdgpu_ring_alloc(ring, kiq->pmf->invalidate_tlbs_size + 8);
+		amdgpu_ring_alloc(ring, ndw);
+		if (vega20_xgmi_wa)
+			kiq->pmf->kiq_invalidate_tlbs(ring,
+						      pasid, 2, all_hub);
 		kiq->pmf->kiq_invalidate_tlbs(ring,
 					pasid, flush_type, all_hub);
 		amdgpu_fence_emit_polling(ring, &seq);
@@ -1272,6 +1308,19 @@ static void gmc_v9_0_init_golden_registers(struct amdgpu_device *adev)
 }
 
 /**
+ * gmc_v9_0_restore_registers - restores regs
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * This restores register values, saved at suspend.
+ */
+static void gmc_v9_0_restore_registers(struct amdgpu_device *adev)
+{
+	if (adev->asic_type == CHIP_RAVEN)
+		WREG32(mmDCHUBBUB_SDPIF_MMIO_CNTRL_0, adev->gmc.sdpif_register);
+}
+
+/**
  * gmc_v9_0_gart_enable - gart enable
  *
  * @adev: amdgpu_device pointer
@@ -1377,6 +1426,20 @@ static int gmc_v9_0_hw_init(void *handle)
 }
 
 /**
+ * gmc_v9_0_save_registers - saves regs
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * This saves potential register values that should be
+ * restored upon resume
+ */
+static void gmc_v9_0_save_registers(struct amdgpu_device *adev)
+{
+	if (adev->asic_type == CHIP_RAVEN)
+		adev->gmc.sdpif_register = RREG32(mmDCHUBBUB_SDPIF_MMIO_CNTRL_0);
+}
+
+/**
  * gmc_v9_0_gart_disable - gart disable
  *
  * @adev: amdgpu_device pointer
@@ -1412,9 +1475,16 @@ static int gmc_v9_0_hw_fini(void *handle)
 
 static int gmc_v9_0_suspend(void *handle)
 {
+	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	return gmc_v9_0_hw_fini(adev);
+	r = gmc_v9_0_hw_fini(adev);
+	if (r)
+		return r;
+
+	gmc_v9_0_save_registers(adev);
+
+	return 0;
 }
 
 static int gmc_v9_0_resume(void *handle)
@@ -1422,6 +1492,7 @@ static int gmc_v9_0_resume(void *handle)
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	gmc_v9_0_restore_registers(adev);
 	r = gmc_v9_0_hw_init(adev);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
index 65eb378fa035..149d386590df 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
@@ -318,6 +318,7 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device
 {
 	uint32_t bif_doorbell_intr_cntl;
 	struct ras_manager *obj = amdgpu_ras_find_obj(adev, adev->nbio.ras_if);
+	struct ras_err_data err_data = {0, 0, 0, NULL};
 
 	bif_doorbell_intr_cntl = RREG32_SOC15(NBIO, 0, mmBIF_DOORBELL_INT_CNTL);
 	if (REG_GET_FIELD(bif_doorbell_intr_cntl,
@@ -332,7 +333,19 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device
 		 * clear error status after ras_controller_intr according to
 		 * hw team and count ue number for query
 		 */
-		nbio_v7_4_query_ras_error_count(adev, &obj->err_data);
+		nbio_v7_4_query_ras_error_count(adev, &err_data);
+
+		/* logging on error counter and printing for awareness */
+		obj->err_data.ue_count += err_data.ue_count;
+		obj->err_data.ce_count += err_data.ce_count;
+
+		if (err_data.ce_count)
+			DRM_INFO("%ld correctable errors detected in %s block\n",
+				obj->err_data.ce_count, adev->nbio.ras_if->name);
+
+		if (err_data.ue_count)
+			DRM_INFO("%ld uncorrectable errors detected in %s block\n",
+				obj->err_data.ue_count, adev->nbio.ras_if->name);
 
 		DRM_WARN("RAS controller interrupt triggered by NBIF error\n");
 
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
index 0829188c1a5c..8ab3bf3158a9 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
@@ -420,7 +420,8 @@ static int psp_v11_0_ring_init(struct psp_context *psp,
 	struct psp_ring *ring;
 	struct amdgpu_device *adev = psp->adev;
 
-	psp_v11_0_reroute_ih(psp);
+	if (!amdgpu_sriov_vf(adev))
+		psp_v11_0_reroute_ih(psp);
 
 	ring = &psp->km_ring;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
index 67b9830b7c7e..ebfd2cdf4e65 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
@@ -746,11 +746,9 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev)
 			sdma_v5_0_enable(adev, true);
 		}
 
-		r = amdgpu_ring_test_ring(ring);
-		if (r) {
-			ring->sched.ready = false;
+		r = amdgpu_ring_test_helper(ring);
+		if (r)
 			return r;
-		}
 
 		if (adev->mman.buffer_funcs_ring == ring)
 			amdgpu_ttm_set_buffer_funcs_status(adev, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 15f3424a1ff7..2b488dfb2f21 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -272,7 +272,12 @@ static u32 soc15_get_config_memsize(struct amdgpu_device *adev)
 
 static u32 soc15_get_xclk(struct amdgpu_device *adev)
 {
-	return adev->clock.spll.reference_freq;
+	u32 reference_clock = adev->clock.spll.reference_freq;
+
+	if (adev->asic_type == CHIP_RAVEN)
+		return reference_clock / 4;
+
+	return reference_clock;
 }