Merge remote-tracking branch 'drm/drm-next' into drm-misc-next

Requested for backmerging airlied's drm-legacy cleanup.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>

40 files changed, 571 insertions, 308 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index f8c58c425eb9..fdd0ca4b0f0b 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -23,7 +23,7 @@
 # Makefile for the drm device driver.  This driver provides support for the
 # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
 
-FULL_AMD_PATH=$(src)/..
+FULL_AMD_PATH=$(srctree)/$(src)/..
 DISPLAY_FOLDER_NAME=display
 FULL_AMD_DISPLAY_PATH = $(FULL_AMD_PATH)/$(DISPLAY_FOLDER_NAME)
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 6e71749cb3bb..14398f55f602 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -436,6 +436,12 @@ struct amdgpu_cs_chunk {
 	void			*kdata;
 };
 
+struct amdgpu_cs_post_dep {
+	struct drm_syncobj *syncobj;
+	struct dma_fence_chain *chain;
+	u64 point;
+};
+
 struct amdgpu_cs_parser {
 	struct amdgpu_device	*adev;
 	struct drm_file		*filp;
@@ -465,8 +471,8 @@ struct amdgpu_cs_parser {
 	/* user fence */
 	struct amdgpu_bo_list_entry	uf_entry;
 
-	unsigned num_post_dep_syncobjs;
-	struct drm_syncobj **post_dep_syncobjs;
+	unsigned			num_post_deps;
+	struct amdgpu_cs_post_dep	*post_deps;
 };
 
 static inline u32 amdgpu_get_ib_value(struct amdgpu_cs_parser *p,
@@ -827,6 +833,7 @@ struct amdgpu_device {
 	/* For pre-DCE11. DCE11 and later are in "struct amdgpu_device->dm" */
 	struct work_struct		hotplug_work;
 	struct amdgpu_irq_src		crtc_irq;
+	struct amdgpu_irq_src		vupdate_irq;
 	struct amdgpu_irq_src		pageflip_irq;
 	struct amdgpu_irq_src		hpd_irq;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index acf8ae0cee9a..aeead072fa79 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -335,6 +335,43 @@ void amdgpu_amdkfd_free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
 	amdgpu_bo_unref(&(bo));
 }
 
+uint32_t amdgpu_amdkfd_get_fw_version(struct kgd_dev *kgd,
+				      enum kgd_engine_type type)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+
+	switch (type) {
+	case KGD_ENGINE_PFP:
+		return adev->gfx.pfp_fw_version;
+
+	case KGD_ENGINE_ME:
+		return adev->gfx.me_fw_version;
+
+	case KGD_ENGINE_CE:
+		return adev->gfx.ce_fw_version;
+
+	case KGD_ENGINE_MEC1:
+		return adev->gfx.mec_fw_version;
+
+	case KGD_ENGINE_MEC2:
+		return adev->gfx.mec2_fw_version;
+
+	case KGD_ENGINE_RLC:
+		return adev->gfx.rlc_fw_version;
+
+	case KGD_ENGINE_SDMA1:
+		return adev->sdma.instance[0].fw_version;
+
+	case KGD_ENGINE_SDMA2:
+		return adev->sdma.instance[1].fw_version;
+
+	default:
+		return 0;
+	}
+
+	return 0;
+}
+
 void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd,
 				      struct kfd_local_mem_info *mem_info)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 775f815f9521..4e37fa7e85b1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -81,6 +81,18 @@ struct amdgpu_kfd_dev {
 	uint64_t vram_used;
 };
 
+enum kgd_engine_type {
+	KGD_ENGINE_PFP = 1,
+	KGD_ENGINE_ME,
+	KGD_ENGINE_CE,
+	KGD_ENGINE_MEC1,
+	KGD_ENGINE_MEC2,
+	KGD_ENGINE_RLC,
+	KGD_ENGINE_SDMA1,
+	KGD_ENGINE_SDMA2,
+	KGD_ENGINE_MAX
+};
+
 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);
@@ -142,6 +154,8 @@ int amdgpu_amdkfd_alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
 				void **mem_obj, uint64_t *gpu_addr,
 				void **cpu_ptr, bool mqd_gfx9);
 void amdgpu_amdkfd_free_gtt_mem(struct kgd_dev *kgd, void *mem_obj);
+uint32_t amdgpu_amdkfd_get_fw_version(struct kgd_dev *kgd,
+				      enum kgd_engine_type type);
 void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd,
 				      struct kfd_local_mem_info *mem_info);
 uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct kgd_dev *kgd);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
index ff7fac7df34b..fa09e11a600c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
@@ -22,14 +22,12 @@
 
 #include <linux/fdtable.h>
 #include <linux/uaccess.h>
-#include <linux/firmware.h>
 #include <linux/mmu_context.h>
 #include <drm/drmP.h>
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
 #include "cikd.h"
 #include "cik_sdma.h"
-#include "amdgpu_ucode.h"
 #include "gfx_v7_0.h"
 #include "gca/gfx_7_2_d.h"
 #include "gca/gfx_7_2_enum.h"
@@ -139,7 +137,6 @@ static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd, uint8_t vmid);
 static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
 							uint8_t vmid);
 
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type);
 static void set_scratch_backing_va(struct kgd_dev *kgd,
 					uint64_t va, uint32_t vmid);
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
@@ -191,7 +188,6 @@ static const struct kfd2kgd_calls kfd2kgd = {
 	.address_watch_get_offset = kgd_address_watch_get_offset,
 	.get_atc_vmid_pasid_mapping_pasid = get_atc_vmid_pasid_mapping_pasid,
 	.get_atc_vmid_pasid_mapping_valid = get_atc_vmid_pasid_mapping_valid,
-	.get_fw_version = get_fw_version,
 	.set_scratch_backing_va = set_scratch_backing_va,
 	.get_tile_config = get_tile_config,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
@@ -792,63 +788,6 @@ static void set_scratch_backing_va(struct kgd_dev *kgd,
 	unlock_srbm(kgd);
 }
 
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
-	const union amdgpu_firmware_header *hdr;
-
-	switch (type) {
-	case KGD_ENGINE_PFP:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.pfp_fw->data;
-		break;
-
-	case KGD_ENGINE_ME:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.me_fw->data;
-		break;
-
-	case KGD_ENGINE_CE:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.ce_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC1:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.mec_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC2:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.mec2_fw->data;
-		break;
-
-	case KGD_ENGINE_RLC:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.rlc_fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA1:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->sdma.instance[0].fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA2:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->sdma.instance[1].fw->data;
-		break;
-
-	default:
-		return 0;
-	}
-
-	if (hdr == NULL)
-		return 0;
-
-	/* Only 12 bit in use*/
-	return hdr->common.ucode_version;
-}
-
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
 			uint64_t page_table_base)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
index 56ea929f524b..fec3a6aa1de6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
@@ -23,12 +23,10 @@
 #include <linux/module.h>
 #include <linux/fdtable.h>
 #include <linux/uaccess.h>
-#include <linux/firmware.h>
 #include <linux/mmu_context.h>
 #include <drm/drmP.h>
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
-#include "amdgpu_ucode.h"
 #include "gfx_v8_0.h"
 #include "gca/gfx_8_0_sh_mask.h"
 #include "gca/gfx_8_0_d.h"
@@ -95,7 +93,6 @@ static bool get_atc_vmid_pasid_mapping_valid(struct kgd_dev *kgd,
 		uint8_t vmid);
 static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
 		uint8_t vmid);
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type);
 static void set_scratch_backing_va(struct kgd_dev *kgd,
 					uint64_t va, uint32_t vmid);
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
@@ -148,7 +145,6 @@ static const struct kfd2kgd_calls kfd2kgd = {
 			get_atc_vmid_pasid_mapping_pasid,
 	.get_atc_vmid_pasid_mapping_valid =
 			get_atc_vmid_pasid_mapping_valid,
-	.get_fw_version = get_fw_version,
 	.set_scratch_backing_va = set_scratch_backing_va,
 	.get_tile_config = get_tile_config,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
@@ -751,63 +747,6 @@ static void set_scratch_backing_va(struct kgd_dev *kgd,
 	unlock_srbm(kgd);
 }
 
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
-	const union amdgpu_firmware_header *hdr;
-
-	switch (type) {
-	case KGD_ENGINE_PFP:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.pfp_fw->data;
-		break;
-
-	case KGD_ENGINE_ME:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.me_fw->data;
-		break;
-
-	case KGD_ENGINE_CE:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.ce_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC1:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.mec_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC2:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.mec2_fw->data;
-		break;
-
-	case KGD_ENGINE_RLC:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->gfx.rlc_fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA1:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->sdma.instance[0].fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA2:
-		hdr = (const union amdgpu_firmware_header *)
-						adev->sdma.instance[1].fw->data;
-		break;
-
-	default:
-		return 0;
-	}
-
-	if (hdr == NULL)
-		return 0;
-
-	/* Only 12 bit in use*/
-	return hdr->common.ucode_version;
-}
-
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
 		uint64_t page_table_base)
 {
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 5c51d4910650..ef3d93b995b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -25,12 +25,10 @@
 #include <linux/module.h>
 #include <linux/fdtable.h>
 #include <linux/uaccess.h>
-#include <linux/firmware.h>
 #include <linux/mmu_context.h>
 #include <drm/drmP.h>
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
-#include "amdgpu_ucode.h"
 #include "soc15_hw_ip.h"
 #include "gc/gc_9_0_offset.h"
 #include "gc/gc_9_0_sh_mask.h"
@@ -111,7 +109,6 @@ static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
 		uint8_t vmid);
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
 		uint64_t page_table_base);
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type);
 static void set_scratch_backing_va(struct kgd_dev *kgd,
 					uint64_t va, uint32_t vmid);
 static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid);
@@ -158,7 +155,6 @@ static const struct kfd2kgd_calls kfd2kgd = {
 			get_atc_vmid_pasid_mapping_pasid,
 	.get_atc_vmid_pasid_mapping_valid =
 			get_atc_vmid_pasid_mapping_valid,
-	.get_fw_version = get_fw_version,
 	.set_scratch_backing_va = set_scratch_backing_va,
 	.get_tile_config = amdgpu_amdkfd_get_tile_config,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
@@ -874,56 +870,6 @@ static void set_scratch_backing_va(struct kgd_dev *kgd,
 	 */
 }
 
-/* FIXME: Does this need to be ASIC-specific code? */
-static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
-{
-	struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
-	const union amdgpu_firmware_header *hdr;
-
-	switch (type) {
-	case KGD_ENGINE_PFP:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.pfp_fw->data;
-		break;
-
-	case KGD_ENGINE_ME:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.me_fw->data;
-		break;
-
-	case KGD_ENGINE_CE:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.ce_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC1:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.mec_fw->data;
-		break;
-
-	case KGD_ENGINE_MEC2:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.mec2_fw->data;
-		break;
-
-	case KGD_ENGINE_RLC:
-		hdr = (const union amdgpu_firmware_header *)adev->gfx.rlc_fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA1:
-		hdr = (const union amdgpu_firmware_header *)adev->sdma.instance[0].fw->data;
-		break;
-
-	case KGD_ENGINE_SDMA2:
-		hdr = (const union amdgpu_firmware_header *)adev->sdma.instance[1].fw->data;
-		break;
-
-	default:
-		return 0;
-	}
-
-	if (hdr == NULL)
-		return 0;
-
-	/* Only 12 bit in use*/
-	return hdr->common.ucode_version;
-}
-
 static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
 		uint64_t page_table_base)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 52a5e4fdc95b..2f6239b6be6f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -215,6 +215,8 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, union drm_amdgpu_cs
 		case AMDGPU_CHUNK_ID_SYNCOBJ_IN:
 		case AMDGPU_CHUNK_ID_SYNCOBJ_OUT:
 		case AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES:
+		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT:
+		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL:
 			break;
 
 		default:
@@ -804,9 +806,11 @@ static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error,
 		ttm_eu_backoff_reservation(&parser->ticket,
 					   &parser->validated);
 
-	for (i = 0; i < parser->num_post_dep_syncobjs; i++)
-		drm_syncobj_put(parser->post_dep_syncobjs[i]);
-	kfree(parser->post_dep_syncobjs);
+	for (i = 0; i < parser->num_post_deps; i++) {
+		drm_syncobj_put(parser->post_deps[i].syncobj);
+		kfree(parser->post_deps[i].chain);
+	}
+	kfree(parser->post_deps);
 
 	dma_fence_put(parser->fence);
 
@@ -1117,13 +1121,18 @@ static int amdgpu_cs_process_fence_dep(struct amdgpu_cs_parser *p,
 }
 
 static int amdgpu_syncobj_lookup_and_add_to_sync(struct amdgpu_cs_parser *p,
-						 uint32_t handle)
+						 uint32_t handle, u64 point,
+						 u64 flags)
 {
-	int r;
 	struct dma_fence *fence;
-	r = drm_syncobj_find_fence(p->filp, handle, 0, 0, &fence);
-	if (r)
+	int r;
+
+	r = drm_syncobj_find_fence(p->filp, handle, point, flags, &fence);
+	if (r) {
+		DRM_ERROR("syncobj %u failed to find fence @ %llu (%d)!\n",
+			  handle, point, r);
 		return r;
+	}
 
 	r = amdgpu_sync_fence(p->adev, &p->job->sync, fence, true);
 	dma_fence_put(fence);
@@ -1134,46 +1143,118 @@ static int amdgpu_syncobj_lookup_and_add_to_sync(struct amdgpu_cs_parser *p,
 static int amdgpu_cs_process_syncobj_in_dep(struct amdgpu_cs_parser *p,
 					    struct amdgpu_cs_chunk *chunk)
 {
+	struct drm_amdgpu_cs_chunk_sem *deps;
 	unsigned num_deps;
 	int i, r;
-	struct drm_amdgpu_cs_chunk_sem *deps;
 
 	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
 	num_deps = chunk->length_dw * 4 /
 		sizeof(struct drm_amdgpu_cs_chunk_sem);
+	for (i = 0; i < num_deps; ++i) {
+		r = amdgpu_syncobj_lookup_and_add_to_sync(p, deps[i].handle,
+							  0, 0);
+		if (r)
+			return r;
+	}
+
+	return 0;
+}
+
 
+static int amdgpu_cs_process_syncobj_timeline_in_dep(struct amdgpu_cs_parser *p,
+						     struct amdgpu_cs_chunk *chunk)
+{
+	struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps;
+	unsigned num_deps;
+	int i, r;
+
+	syncobj_deps = (struct drm_amdgpu_cs_chunk_syncobj *)chunk->kdata;
+	num_deps = chunk->length_dw * 4 /
+		sizeof(struct drm_amdgpu_cs_chunk_syncobj);
 	for (i = 0; i < num_deps; ++i) {
-		r = amdgpu_syncobj_lookup_and_add_to_sync(p, deps[i].handle);
+		r = amdgpu_syncobj_lookup_and_add_to_sync(p,
+							  syncobj_deps[i].handle,
+							  syncobj_deps[i].point,
+							  syncobj_deps[i].flags);
 		if (r)
 			return r;
 	}
+
 	return 0;
 }
 
 static int amdgpu_cs_process_syncobj_out_dep(struct amdgpu_cs_parser *p,
 					     struct amdgpu_cs_chunk *chunk)
 {
+	struct drm_amdgpu_cs_chunk_sem *deps;
 	unsigned num_deps;
 	int i;
-	struct drm_amdgpu_cs_chunk_sem *deps;
+
 	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
 	num_deps = chunk->length_dw * 4 /
 		sizeof(struct drm_amdgpu_cs_chunk_sem);
 
-	p->post_dep_syncobjs = kmalloc_array(num_deps,
-					     sizeof(struct drm_syncobj *),
-					     GFP_KERNEL);
-	p->num_post_dep_syncobjs = 0;
+	p->post_deps = kmalloc_array(num_deps, sizeof(*p->post_deps),
+				     GFP_KERNEL);
+	p->num_post_deps = 0;
 
-	if (!p->post_dep_syncobjs)
+	if (!p->post_deps)
 		return -ENOMEM;
 
+
 	for (i = 0; i < num_deps; ++i) {
-		p->post_dep_syncobjs[i] = drm_syncobj_find(p->filp, deps[i].handle);
-		if (!p->post_dep_syncobjs[i])
+		p->post_deps[i].syncobj =
+			drm_syncobj_find(p->filp, deps[i].handle);
+		if (!p->post_deps[i].syncobj)
 			return -EINVAL;
-		p->num_post_dep_syncobjs++;
+		p->post_deps[i].chain = NULL;
+		p->post_deps[i].point = 0;
+		p->num_post_deps++;
 	}
+
+	return 0;
+}
+
+
+static int amdgpu_cs_process_syncobj_timeline_out_dep(struct amdgpu_cs_parser *p,
+						      struct amdgpu_cs_chunk
+						      *chunk)
+{
+	struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps;
+	unsigned num_deps;
+	int i;
+
+	syncobj_deps = (struct drm_amdgpu_cs_chunk_syncobj *)chunk->kdata;
+	num_deps = chunk->length_dw * 4 /
+		sizeof(struct drm_amdgpu_cs_chunk_syncobj);
+
+	p->post_deps = kmalloc_array(num_deps, sizeof(*p->post_deps),
+				     GFP_KERNEL);
+	p->num_post_deps = 0;
+
+	if (!p->post_deps)
+		return -ENOMEM;
+
+	for (i = 0; i < num_deps; ++i) {
+		struct amdgpu_cs_post_dep *dep = &p->post_deps[i];
+
+		dep->chain = NULL;
+		if (syncobj_deps[i].point) {
+			dep->chain = kmalloc(sizeof(*dep->chain), GFP_KERNEL);
+			if (!dep->chain)
+				return -ENOMEM;
+		}
+
+		dep->syncobj = drm_syncobj_find(p->filp,
+						syncobj_deps[i].handle);
+		if (!dep->syncobj) {
+			kfree(dep->chain);
+			return -EINVAL;
+		}
+		dep->point = syncobj_deps[i].point;
+		p->num_post_deps++;
+	}
+
 	return 0;
 }
 
@@ -1187,19 +1268,33 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
 
 		chunk = &p->chunks[i];
 
-		if (chunk->chunk_id == AMDGPU_CHUNK_ID_DEPENDENCIES ||
-		    chunk->chunk_id == AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES) {
+		switch (chunk->chunk_id) {
+		case AMDGPU_CHUNK_ID_DEPENDENCIES:
+		case AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES:
 			r = amdgpu_cs_process_fence_dep(p, chunk);
 			if (r)
 				return r;
-		} else if (chunk->chunk_id == AMDGPU_CHUNK_ID_SYNCOBJ_IN) {
+			break;
+		case AMDGPU_CHUNK_ID_SYNCOBJ_IN:
 			r = amdgpu_cs_process_syncobj_in_dep(p, chunk);
 			if (r)
 				return r;
-		} else if (chunk->chunk_id == AMDGPU_CHUNK_ID_SYNCOBJ_OUT) {
+			break;
+		case AMDGPU_CHUNK_ID_SYNCOBJ_OUT:
 			r = amdgpu_cs_process_syncobj_out_dep(p, chunk);
 			if (r)
 				return r;
+			break;
+		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT:
+			r = amdgpu_cs_process_syncobj_timeline_in_dep(p, chunk);
+			if (r)
+				return r;
+			break;
+		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL:
+			r = amdgpu_cs_process_syncobj_timeline_out_dep(p, chunk);
+			if (r)
+				return r;
+			break;
 		}
 	}
 
@@ -1210,8 +1305,17 @@ static void amdgpu_cs_post_dependencies(struct amdgpu_cs_parser *p)
 {
 	int i;
 
-	for (i = 0; i < p->num_post_dep_syncobjs; ++i)
-		drm_syncobj_replace_fence(p->post_dep_syncobjs[i], p->fence);
+	for (i = 0; i < p->num_post_deps; ++i) {
+		if (p->post_deps[i].chain && p->post_deps[i].point) {
+			drm_syncobj_add_point(p->post_deps[i].syncobj,
+					      p->post_deps[i].chain,
+					      p->fence, p->post_deps[i].point);
+			p->post_deps[i].chain = NULL;
+		} else {
+			drm_syncobj_replace_fence(p->post_deps[i].syncobj,
+						  p->fence);
+		}
+	}
 }
 
 static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 85f8792c2a5f..b9371ec5e04f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2471,6 +2471,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	mutex_init(&adev->virt.vf_errors.lock);
 	hash_init(adev->mn_hash);
 	mutex_init(&adev->lock_reset);
+	mutex_init(&adev->virt.dpm_mutex);
 
 	amdgpu_device_check_arguments(adev);
 
@@ -3191,11 +3192,16 @@ static int amdgpu_device_recover_vram(struct amdgpu_device *adev)
 			break;
 
 		if (fence) {
-			r = dma_fence_wait_timeout(fence, false, tmo);
+			tmo = dma_fence_wait_timeout(fence, false, tmo);
 			dma_fence_put(fence);
 			fence = next;
-			if (r <= 0)
+			if (tmo == 0) {
+				r = -ETIMEDOUT;
 				break;
+			} else if (tmo < 0) {
+				r = tmo;
+				break;
+			}
 		} else {
 			fence = next;
 		}
@@ -3206,8 +3212,8 @@ static int amdgpu_device_recover_vram(struct amdgpu_device *adev)
 		tmo = dma_fence_wait_timeout(fence, false, tmo);
 	dma_fence_put(fence);
 
-	if (r <= 0 || tmo <= 0) {
-		DRM_ERROR("recover vram bo from shadow failed\n");
+	if (r < 0 || tmo <= 0) {
+		DRM_ERROR("recover vram bo from shadow failed, r is %ld, tmo is %ld\n", r, tmo);
 		return -EIO;
 	}
 
@@ -3428,7 +3434,7 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
 
 				vram_lost = amdgpu_device_check_vram_lost(tmp_adev);
 				if (vram_lost) {
-					DRM_ERROR("VRAM is lost!\n");
+					DRM_INFO("VRAM is lost due to GPU reset!\n");
 					atomic_inc(&tmp_adev->vram_lost_counter);
 				}
 
@@ -3695,6 +3701,7 @@ static void amdgpu_device_get_min_pci_speed_width(struct amdgpu_device *adev,
 	struct pci_dev *pdev = adev->pdev;
 	enum pci_bus_speed cur_speed;
 	enum pcie_link_width cur_width;
+	u32 ret = 1;
 
 	*speed = PCI_SPEED_UNKNOWN;
 	*width = PCIE_LNK_WIDTH_UNKNOWN;
@@ -3702,6 +3709,10 @@ static void amdgpu_device_get_min_pci_speed_width(struct amdgpu_device *adev,
 	while (pdev) {
 		cur_speed = pcie_get_speed_cap(pdev);
 		cur_width = pcie_get_width_cap(pdev);
+		ret = pcie_bandwidth_available(adev->pdev, NULL,
+						       NULL, &cur_width);
+		if (!ret)
+			cur_width = PCIE_LNK_WIDTH_RESRV;
 
 		if (cur_speed != PCI_SPEED_UNKNOWN) {
 			if (*speed == PCI_SPEED_UNKNOWN)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index f34e3ab5a9f3..1e2cc9d68a05 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -75,9 +75,10 @@
  * - 3.29.0 - Add AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID
  * - 3.30.0 - Add AMDGPU_SCHED_OP_CONTEXT_PRIORITY_OVERRIDE.
  * - 3.31.0 - Add support for per-flip tiling attribute changes with DC
+ * - 3.32.0 - Add syncobj timeline support to AMDGPU_CS.
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	31
+#define KMS_DRIVER_MINOR	32
 #define KMS_DRIVER_PATCHLEVEL	0
 
 int amdgpu_vram_limit = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index ee47c11e92ce..4dee2326b29c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -136,8 +136,9 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f,
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_fence *fence;
-	struct dma_fence *old, **ptr;
+	struct dma_fence __rcu **ptr;
 	uint32_t seq;
+	int r;
 
 	fence = kmem_cache_alloc(amdgpu_fence_slab, GFP_KERNEL);
 	if (fence == NULL)
@@ -153,15 +154,24 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f,
 			       seq, flags | AMDGPU_FENCE_FLAG_INT);
 
 	ptr = &ring->fence_drv.fences[seq & ring->fence_drv.num_fences_mask];
+	if (unlikely(rcu_dereference_protected(*ptr, 1))) {
+		struct dma_fence *old;
+
+		rcu_read_lock();
+		old = dma_fence_get_rcu_safe(ptr);
+		rcu_read_unlock();
+
+		if (old) {
+			r = dma_fence_wait(old, false);
+			dma_fence_put(old);
+			if (r)
+				return r;
+		}
+	}
+
 	/* This function can't be called concurrently anyway, otherwise
 	 * emitting the fence would mess up the hardware ring buffer.
 	 */
-	old = rcu_dereference_protected(*ptr, 1);
-	if (old && !dma_fence_is_signaled(old)) {
-		DRM_INFO("rcu slot is busy\n");
-		dma_fence_wait(old, false);
-	}
-
 	rcu_assign_pointer(*ptr, dma_fence_get(&fence->base));
 
 	*f = &fence->base;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 61107cfc9af6..d4fcf5475464 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -31,6 +31,7 @@
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
 #include "amdgpu_display.h"
+#include "amdgpu_xgmi.h"
 
 void amdgpu_gem_object_free(struct drm_gem_object *gobj)
 {
@@ -668,6 +669,7 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 	struct amdgpu_device *adev = dev->dev_private;
 	struct drm_amdgpu_gem_op *args = data;
 	struct drm_gem_object *gobj;
+	struct amdgpu_vm_bo_base *base;
 	struct amdgpu_bo *robj;
 	int r;
 
@@ -706,6 +708,15 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 			amdgpu_bo_unreserve(robj);
 			break;
 		}
+		for (base = robj->vm_bo; base; base = base->next)
+			if (amdgpu_xgmi_same_hive(amdgpu_ttm_adev(robj->tbo.bdev),
+				amdgpu_ttm_adev(base->vm->root.base.bo->tbo.bdev))) {
+				r = -EINVAL;
+				amdgpu_bo_unreserve(robj);
+				goto out;
+			}
+
+
 		robj->preferred_domains = args->value & (AMDGPU_GEM_DOMAIN_VRAM |
 							AMDGPU_GEM_DOMAIN_GTT |
 							AMDGPU_GEM_DOMAIN_CPU);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index 0b8ef2d27d6b..fe393a46f881 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -35,6 +35,7 @@
 #include "amdgpu_trace.h"
 
 #define AMDGPU_IB_TEST_TIMEOUT	msecs_to_jiffies(1000)
+#define AMDGPU_IB_TEST_GFX_XGMI_TIMEOUT	msecs_to_jiffies(2000)
 
 /*
  * IB
@@ -344,6 +345,8 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev)
 		 * cost waiting for it coming back under RUNTIME only
 		*/
 		tmo_gfx = 8 * AMDGPU_IB_TEST_TIMEOUT;
+	} else if (adev->gmc.xgmi.hive_id) {
+		tmo_gfx = AMDGPU_IB_TEST_GFX_XGMI_TIMEOUT;
 	}
 
 	for (i = 0; i < adev->num_rings; ++i) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 2e376064bad8..b17d0545728e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -696,6 +696,10 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		if (adev->pm.dpm_enabled) {
 			dev_info.max_engine_clock = amdgpu_dpm_get_sclk(adev, false) * 10;
 			dev_info.max_memory_clock = amdgpu_dpm_get_mclk(adev, false) * 10;
+		} else if (amdgpu_sriov_vf(adev) && amdgim_is_hwperf(adev) &&
+			   adev->virt.ops->get_pp_clk) {
+			dev_info.max_engine_clock = amdgpu_virt_get_sclk(adev, false) * 10;
+			dev_info.max_memory_clock = amdgpu_virt_get_mclk(adev, false) * 10;
 		} else {
 			dev_info.max_engine_clock = adev->clock.default_sclk * 10;
 			dev_info.max_memory_clock = adev->clock.default_mclk * 10;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
index 889e443eeee7..2e9e3db778c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
@@ -58,7 +58,7 @@ struct amdgpu_hpd;
 #define to_amdgpu_encoder(x) container_of(x, struct amdgpu_encoder, base)
 #define to_amdgpu_framebuffer(x) container_of(x, struct amdgpu_framebuffer, base)
 
-#define to_dm_plane_state(x)	container_of(x, struct dm_plane_state, base);
+#define to_dm_plane_state(x)	container_of(x, struct dm_plane_state, base)
 
 #define AMDGPU_MAX_HPD_PINS 6
 #define AMDGPU_MAX_CRTCS 6
@@ -406,7 +406,7 @@ struct amdgpu_crtc {
 	struct amdgpu_flip_work *pflip_works;
 	enum amdgpu_flip_status pflip_status;
 	int deferred_flip_completion;
-	u64 last_flip_vblank;
+	u32 last_flip_vblank;
 	/* pll sharing */
 	struct amdgpu_atom_ss ss;
 	bool ss_enabled;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index ec9e45004bff..93b2c5a48a71 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -88,12 +88,14 @@ static void amdgpu_bo_destroy(struct ttm_buffer_object *tbo)
 	if (bo->gem_base.import_attach)
 		drm_prime_gem_destroy(&bo->gem_base, bo->tbo.sg);
 	drm_gem_object_release(&bo->gem_base);
-	amdgpu_bo_unref(&bo->parent);
+	/* in case amdgpu_device_recover_vram got NULL of bo->parent */
 	if (!list_empty(&bo->shadow_list)) {
 		mutex_lock(&adev->shadow_list_lock);
 		list_del_init(&bo->shadow_list);
 		mutex_unlock(&adev->shadow_list_lock);
 	}
+	amdgpu_bo_unref(&bo->parent);
+
 	kfree(bo->metadata);
 	kfree(bo);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
index 88362019d1dd..95144e49c7f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
@@ -144,7 +144,7 @@ static ssize_t amdgpu_get_dpm_state(struct device *dev,
 	struct amdgpu_device *adev = ddev->dev_private;
 	enum amd_pm_state_type pm;
 
-	if (adev->smu.ppt_funcs->get_current_power_state)
+	if (is_support_sw_smu(adev) && adev->smu.ppt_funcs->get_current_power_state)
 		pm = amdgpu_smu_get_current_power_state(adev);
 	else if (adev->powerplay.pp_funcs->get_current_power_state)
 		pm = amdgpu_dpm_get_current_power_state(adev);
@@ -327,6 +327,18 @@ static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev,
 		goto fail;
 	}
 
+        if (amdgpu_sriov_vf(adev)) {
+                if (amdgim_is_hwperf(adev) &&
+                    adev->virt.ops->force_dpm_level) {
+                        mutex_lock(&adev->pm.mutex);
+                        adev->virt.ops->force_dpm_level(adev, level);
+                        mutex_unlock(&adev->pm.mutex);
+                        return count;
+                } else {
+                        return -EINVAL;
+		}
+        }
+
 	if (current_level == level)
 		return count;
 
@@ -790,6 +802,10 @@ static ssize_t amdgpu_get_pp_dpm_sclk(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 
+	if (amdgpu_sriov_vf(adev) && amdgim_is_hwperf(adev) &&
+	    adev->virt.ops->get_pp_clk)
+		return adev->virt.ops->get_pp_clk(adev, PP_SCLK, buf);
+
 	if (is_support_sw_smu(adev))
 		return smu_print_clk_levels(&adev->smu, PP_SCLK, buf);
 	else if (adev->powerplay.pp_funcs->print_clock_levels)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 2206bb4b0903..905cce1814f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -187,13 +187,13 @@ static int psp_tmr_init(struct psp_context *psp)
 	int ret;
 
 	/*
-	 * Allocate 3M memory aligned to 1M from Frame Buffer (local
-	 * physical).
+	 * According to HW engineer, they prefer the TMR address be "naturally
+	 * aligned" , e.g. the start address be an integer divide of TMR size.
 	 *
 	 * Note: this memory need be reserved till the driver
 	 * uninitializes.
 	 */
-	ret = amdgpu_bo_create_kernel(psp->adev, PSP_TMR_SIZE, 0x100000,
+	ret = amdgpu_bo_create_kernel(psp->adev, PSP_TMR_SIZE, PSP_TMR_SIZE,
 				      AMDGPU_GEM_DOMAIN_VRAM,
 				      &psp->tmr_bo, &psp->tmr_mc_addr, &psp->tmr_buf);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 469cb6477b8e..22bd21efe6b1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -530,6 +530,33 @@ int amdgpu_ras_feature_enable(struct amdgpu_device *adev,
 	return 0;
 }
 
+/* Only used in device probe stage and called only once. */
+int amdgpu_ras_feature_enable_on_boot(struct amdgpu_device *adev,
+		struct ras_common_if *head, bool enable)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	int ret;
+
+	if (!con)
+		return -EINVAL;
+
+	if (con->flags & AMDGPU_RAS_FLAG_INIT_BY_VBIOS) {
+		/* If ras is enabled by vbios, we set up ras object first in
+		 * both case. For enable, that is all what we need do. For
+		 * disable, we need perform a ras TA disable cmd after that.
+		 */
+		ret = __amdgpu_ras_feature_enable(adev, head, 1);
+		if (ret)
+			return ret;
+
+		if (!enable)
+			ret = amdgpu_ras_feature_enable(adev, head, 0);
+	} else
+		ret = amdgpu_ras_feature_enable(adev, head, enable);
+
+	return ret;
+}
+
 static int amdgpu_ras_disable_all_features(struct amdgpu_device *adev,
 		bool bypass)
 {
@@ -558,11 +585,13 @@ static int amdgpu_ras_enable_all_features(struct amdgpu_device *adev,
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
 	int ras_block_count = AMDGPU_RAS_BLOCK_COUNT;
 	int i;
+	const enum amdgpu_ras_error_type default_ras_type =
+		AMDGPU_RAS_ERROR__NONE;
 
 	for (i = 0; i < ras_block_count; i++) {
 		struct ras_common_if head = {
 			.block = i,
-			.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+			.type = default_ras_type,
 			.sub_block_index = 0,
 		};
 		strcpy(head.name, ras_block_str(i));
@@ -1368,9 +1397,6 @@ int amdgpu_ras_init(struct amdgpu_device *adev)
 
 	amdgpu_ras_mask &= AMDGPU_RAS_BLOCK_MASK;
 
-	if (con->flags & AMDGPU_RAS_FLAG_INIT_BY_VBIOS)
-		amdgpu_ras_enable_all_features(adev, 1);
-
 	if (amdgpu_ras_fs_init(adev))
 		goto fs_out;
 
@@ -1398,18 +1424,25 @@ void amdgpu_ras_post_init(struct amdgpu_device *adev)
 	if (!con)
 		return;
 
-	/* We enable ras on all hw_supported block, but as boot parameter might
-	 * disable some of them and one or more IP has not implemented yet.
-	 * So we disable them on behalf.
-	 */
 	if (con->flags & AMDGPU_RAS_FLAG_INIT_BY_VBIOS) {
+		/* Set up all other IPs which are not implemented. There is a
+		 * tricky thing that IP's actual ras error type should be
+		 * MULTI_UNCORRECTABLE, but as driver does not handle it, so
+		 * ERROR_NONE make sense anyway.
+		 */
+		amdgpu_ras_enable_all_features(adev, 1);
+
+		/* We enable ras on all hw_supported block, but as boot
+		 * parameter might disable some of them and one or more IP has
+		 * not implemented yet. So we disable them on behalf.
+		 */
 		list_for_each_entry_safe(obj, tmp, &con->head, node) {
 			if (!amdgpu_ras_is_supported(adev, obj->head.block)) {
 				amdgpu_ras_feature_enable(adev, &obj->head, 0);
 				/* there should be no any reference. */
 				WARN_ON(alive_obj(obj));
 			}
-		};
+		}
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 682f2be0d68c..eaef5edefc34 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -262,6 +262,9 @@ int amdgpu_ras_pre_fini(struct amdgpu_device *adev);
 int amdgpu_ras_feature_enable(struct amdgpu_device *adev,
 		struct ras_common_if *head, bool enable);
 
+int amdgpu_ras_feature_enable_on_boot(struct amdgpu_device *adev,
+		struct ras_common_if *head, bool enable);
+
 int amdgpu_ras_sysfs_create(struct amdgpu_device *adev,
 		struct ras_fs_if *head);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
index c17af30e758d..1ba9ba3b54f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
@@ -28,11 +28,8 @@
 #define AMDGPU_MAX_SDMA_INSTANCES		2
 
 enum amdgpu_sdma_irq {
-	AMDGPU_SDMA_IRQ_TRAP0 = 0,
-	AMDGPU_SDMA_IRQ_TRAP1,
-	AMDGPU_SDMA_IRQ_ECC0,
-	AMDGPU_SDMA_IRQ_ECC1,
-
+	AMDGPU_SDMA_IRQ_INSTANCE0  = 0,
+	AMDGPU_SDMA_IRQ_INSTANCE1,
 	AMDGPU_SDMA_IRQ_LAST
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 462a04e0f5e6..7d484fad3909 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -36,6 +36,7 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev)
 	/* enable virtual display */
 	adev->mode_info.num_crtc = 1;
 	adev->enable_virtual_display = true;
+	adev->ddev->driver->driver_features &= ~DRIVER_ATOMIC;
 	adev->cg_flags = 0;
 	adev->pg_flags = 0;
 }
@@ -375,4 +376,53 @@ void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev)
 	}
 }
 
+static uint32_t parse_clk(char *buf, bool min)
+{
+        char *ptr = buf;
+        uint32_t clk = 0;
+
+        do {
+                ptr = strchr(ptr, ':');
+                if (!ptr)
+                        break;
+                ptr+=2;
+                clk = simple_strtoul(ptr, NULL, 10);
+        } while (!min);
+
+        return clk * 100;
+}
+
+uint32_t amdgpu_virt_get_sclk(struct amdgpu_device *adev, bool lowest)
+{
+	char *buf = NULL;
+	uint32_t clk = 0;
+
+	buf = kzalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	adev->virt.ops->get_pp_clk(adev, PP_SCLK, buf);
+	clk = parse_clk(buf, lowest);
+
+	kfree(buf);
+
+	return clk;
+}
+
+uint32_t amdgpu_virt_get_mclk(struct amdgpu_device *adev, bool lowest)
+{
+	char *buf = NULL;
+	uint32_t clk = 0;
+
+	buf = kzalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	adev->virt.ops->get_pp_clk(adev, PP_MCLK, buf);
+	clk = parse_clk(buf, lowest);
+
+	kfree(buf);
+
+	return clk;
+}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index 722deefc0a7e..584947b7ccf3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -57,6 +57,8 @@ struct amdgpu_virt_ops {
 	int (*reset_gpu)(struct amdgpu_device *adev);
 	int (*wait_reset)(struct amdgpu_device *adev);
 	void (*trans_msg)(struct amdgpu_device *adev, u32 req, u32 data1, u32 data2, u32 data3);
+	int (*get_pp_clk)(struct amdgpu_device *adev, u32 type, char *buf);
+	int (*force_dpm_level)(struct amdgpu_device *adev, u32 level);
 };
 
 /*
@@ -83,6 +85,8 @@ enum AMDGIM_FEATURE_FLAG {
 	AMDGIM_FEATURE_GIM_LOAD_UCODES   = 0x2,
 	/* VRAM LOST by GIM */
 	AMDGIM_FEATURE_GIM_FLR_VRAMLOST = 0x4,
+	/* HW PERF SIM in GIM */
+	AMDGIM_FEATURE_HW_PERF_SIMULATION = (1 << 3),
 };
 
 struct amd_sriov_msg_pf2vf_info_header {
@@ -252,6 +256,8 @@ struct amdgpu_virt {
 	struct amdgpu_vf_error_buffer   vf_errors;
 	struct amdgpu_virt_fw_reserve	fw_reserve;
 	uint32_t gim_feature;
+	/* protect DPM events to GIM */
+	struct mutex                    dpm_mutex;
 };
 
 #define amdgpu_sriov_enabled(adev) \
@@ -278,6 +284,9 @@ static inline bool is_virtual_machine(void)
 #endif
 }
 
+#define amdgim_is_hwperf(adev) \
+	((adev)->virt.gim_feature & AMDGIM_FEATURE_HW_PERF_SIMULATION)
+
 bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev);
 void amdgpu_virt_init_setting(struct amdgpu_device *adev);
 uint32_t amdgpu_virt_kiq_rreg(struct amdgpu_device *adev, uint32_t reg);
@@ -295,5 +304,7 @@ int amdgpu_virt_fw_reserve_get_checksum(void *obj, unsigned long obj_size,
 					unsigned int key,
 					unsigned int chksum);
 void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev);
+uint32_t amdgpu_virt_get_sclk(struct amdgpu_device *adev, bool lowest);
+uint32_t amdgpu_virt_get_mclk(struct amdgpu_device *adev, bool lowest);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 21c712e34148..a07c85815b7a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -769,14 +769,17 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 
 	addr = 0;
 	if (ats_entries) {
-		uint64_t ats_value;
+		uint64_t value = 0, flags;
 
-		ats_value = AMDGPU_PTE_DEFAULT_ATC;
-		if (level != AMDGPU_VM_PTB)
-			ats_value |= AMDGPU_PDE_PTE;
+		flags = AMDGPU_PTE_DEFAULT_ATC;
+		if (level != AMDGPU_VM_PTB) {
+			/* Handle leaf PDEs as PTEs */
+			flags |= AMDGPU_PDE_PTE;
+			amdgpu_gmc_get_vm_pde(adev, level, &value, &flags);
+		}
 
 		r = vm->update_funcs->update(&params, bo, addr, 0, ats_entries,
-					     0, ats_value);
+					     value, flags);
 		if (r)
 			return r;
 
@@ -784,15 +787,22 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 	}
 
 	if (entries) {
-		uint64_t value = 0;
-
-		/* Workaround for fault priority problem on GMC9 */
-		if (level == AMDGPU_VM_PTB &&
-		    adev->asic_type >= CHIP_VEGA10)
-			value = AMDGPU_PTE_EXECUTABLE;
+		uint64_t value = 0, flags = 0;
+
+		if (adev->asic_type >= CHIP_VEGA10) {
+			if (level != AMDGPU_VM_PTB) {
+				/* Handle leaf PDEs as PTEs */
+				flags |= AMDGPU_PDE_PTE;
+				amdgpu_gmc_get_vm_pde(adev, level,
+						      &value, &flags);
+			} else {
+				/* Workaround for fault priority problem on GMC9 */
+				flags = AMDGPU_PTE_EXECUTABLE;
+			}
+		}
 
 		r = vm->update_funcs->update(&params, bo, addr, 0, entries,
-					     0, value);
+					     value, flags);
 		if (r)
 			return r;
 	}
@@ -2027,7 +2037,8 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 	INIT_LIST_HEAD(&bo_va->valids);
 	INIT_LIST_HEAD(&bo_va->invalids);
 
-	if (bo && amdgpu_xgmi_same_hive(adev, amdgpu_ttm_adev(bo->tbo.bdev))) {
+	if (bo && amdgpu_xgmi_same_hive(adev, amdgpu_ttm_adev(bo->tbo.bdev)) &&
+	    (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM)) {
 		bo_va->is_xgmi = true;
 		mutex_lock(&adev->vm_manager.lock_pstate);
 		/* Power up XGMI if it can be potentially used */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index beac15bca526..91baf95212a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -303,6 +303,7 @@ struct amdgpu_vm_manager {
 	const struct amdgpu_vm_pte_funcs	*vm_pte_funcs;
 	struct drm_sched_rq			*vm_pte_rqs[AMDGPU_MAX_RINGS];
 	unsigned				vm_pte_num_rqs;
+	struct amdgpu_ring			*page_fault;
 
 	/* partial resident texture handling */
 	spinlock_t				prt_lock;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 336834797af3..a48c84c51775 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -24,6 +24,7 @@
 #include <linux/list.h>
 #include "amdgpu.h"
 #include "amdgpu_xgmi.h"
+#include "amdgpu_smu.h"
 
 
 static DEFINE_MUTEX(xgmi_mutex);
@@ -216,7 +217,17 @@ int amdgpu_xgmi_set_pstate(struct amdgpu_device *adev, int pstate)
 
 	if (hive->pstate == pstate)
 		return 0;
-	/* Todo : sent the message to SMU for pstate change */
+
+	dev_dbg(adev->dev, "Set xgmi pstate %d.\n", pstate);
+
+	if (is_support_sw_smu(adev))
+		ret = smu_set_xgmi_pstate(&adev->smu, pstate);
+	if (ret)
+		dev_err(adev->dev,
+			"XGMI: Set pstate failure on device %llx, hive %llx, ret %d",
+			adev->gmc.xgmi.node_id,
+			adev->gmc.xgmi.hive_id, ret);
+
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index 189599b694e8..d42808b05971 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -977,8 +977,8 @@ static int cik_sdma_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 	}
@@ -1114,7 +1114,7 @@ static int cik_sdma_set_trap_irq_state(struct amdgpu_device *adev,
 	u32 sdma_cntl;
 
 	switch (type) {
-	case AMDGPU_SDMA_IRQ_TRAP0:
+	case AMDGPU_SDMA_IRQ_INSTANCE0:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA0_REGISTER_OFFSET);
@@ -1130,7 +1130,7 @@ static int cik_sdma_set_trap_irq_state(struct amdgpu_device *adev,
 			break;
 		}
 		break;
-	case AMDGPU_SDMA_IRQ_TRAP1:
+	case AMDGPU_SDMA_IRQ_INSTANCE1:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA1_REGISTER_OFFSET);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 3765d97b8512..ba67d1023264 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -2455,8 +2455,6 @@ static int gfx_v9_0_rlc_resume(struct amdgpu_device *adev)
 	/* disable CG */
 	WREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL, 0);
 
-	adev->gfx.rlc.funcs->reset(adev);
-
 	gfx_v9_0_init_pg(adev);
 
 	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
@@ -3568,7 +3566,7 @@ static int gfx_v9_0_ecc_late_init(void *handle)
 	int r;
 
 	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) {
-		amdgpu_ras_feature_enable(adev, &ras_block, 0);
+		amdgpu_ras_feature_enable_on_boot(adev, &ras_block, 0);
 		return 0;
 	}
 
@@ -3581,7 +3579,7 @@ static int gfx_v9_0_ecc_late_init(void *handle)
 
 	**ras_if = ras_block;
 
-	r = amdgpu_ras_feature_enable(adev, *ras_if, 1);
+	r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
 	if (r)
 		goto feature;
 
@@ -4840,10 +4838,16 @@ static int gfx_v9_0_cp_ecc_error_irq(struct amdgpu_device *adev,
 				  struct amdgpu_irq_src *source,
 				  struct amdgpu_iv_entry *entry)
 {
+	struct ras_common_if *ras_if = adev->gfx.ras_if;
 	struct ras_dispatch_if ih_data = {
-		.head = *adev->gfx.ras_if,
 		.entry = entry,
 	};
+
+	if (!ras_if)
+		return 0;
+
+	ih_data.head = *ras_if;
+
 	DRM_ERROR("CP ECC ERROR IRQ\n");
 	amdgpu_ras_interrupt_dispatch(adev, &ih_data);
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 404875147ec3..3fd79e07944d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -248,10 +248,16 @@ static int gmc_v9_0_process_ecc_irq(struct amdgpu_device *adev,
 		struct amdgpu_irq_src *source,
 		struct amdgpu_iv_entry *entry)
 {
+	struct ras_common_if *ras_if = adev->gmc.ras_if;
 	struct ras_dispatch_if ih_data = {
-		.head = *adev->gmc.ras_if,
 		.entry = entry,
 	};
+
+	if (!ras_if)
+		return 0;
+
+	ih_data.head = *ras_if;
+
 	amdgpu_ras_interrupt_dispatch(adev, &ih_data);
 	return 0;
 }
@@ -676,7 +682,7 @@ static int gmc_v9_0_ecc_late_init(void *handle)
 	int r;
 
 	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) {
-		amdgpu_ras_feature_enable(adev, &ras_block, 0);
+		amdgpu_ras_feature_enable_on_boot(adev, &ras_block, 0);
 		return 0;
 	}
 	/* handle resume path. */
@@ -689,7 +695,7 @@ static int gmc_v9_0_ecc_late_init(void *handle)
 
 	**ras_if = ras_block;
 
-	r = amdgpu_ras_feature_enable(adev, *ras_if, 1);
+	r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
 	if (r)
 		goto feature;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
index 73851ebb3833..8dbad496b29f 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
@@ -157,6 +157,82 @@ static void xgpu_ai_mailbox_trans_msg (struct amdgpu_device *adev,
 	xgpu_ai_mailbox_set_valid(adev, false);
 }
 
+static int xgpu_ai_get_pp_clk(struct amdgpu_device *adev, u32 type, char *buf)
+{
+        int r = 0;
+        u32 req, val, size;
+
+        if (!amdgim_is_hwperf(adev) || buf == NULL)
+                return -EBADRQC;
+
+        switch(type) {
+        case PP_SCLK:
+                req = IDH_IRQ_GET_PP_SCLK;
+                break;
+        case PP_MCLK:
+                req = IDH_IRQ_GET_PP_MCLK;
+                break;
+        default:
+                return -EBADRQC;
+        }
+
+        mutex_lock(&adev->virt.dpm_mutex);
+
+        xgpu_ai_mailbox_trans_msg(adev, req, 0, 0, 0);
+
+        r = xgpu_ai_poll_msg(adev, IDH_SUCCESS);
+        if (!r && adev->fw_vram_usage.va != NULL) {
+                val = RREG32_NO_KIQ(
+                        SOC15_REG_OFFSET(NBIO, 0,
+                                         mmBIF_BX_PF0_MAILBOX_MSGBUF_RCV_DW1));
+                size = strnlen((((char *)adev->virt.fw_reserve.p_pf2vf) +
+                                val), PAGE_SIZE);
+
+                if (size < PAGE_SIZE)
+                        strcpy(buf,((char *)adev->virt.fw_reserve.p_pf2vf + val));
+                else
+                        size = 0;
+
+                r = size;
+                goto out;
+        }
+
+        r = xgpu_ai_poll_msg(adev, IDH_FAIL);
+        if(r)
+                pr_info("%s DPM request failed",
+                        (type == PP_SCLK)? "SCLK" : "MCLK");
+
+out:
+        mutex_unlock(&adev->virt.dpm_mutex);
+        return r;
+}
+
+static int xgpu_ai_force_dpm_level(struct amdgpu_device *adev, u32 level)
+{
+        int r = 0;
+        u32 req = IDH_IRQ_FORCE_DPM_LEVEL;
+
+        if (!amdgim_is_hwperf(adev))
+                return -EBADRQC;
+
+        mutex_lock(&adev->virt.dpm_mutex);
+        xgpu_ai_mailbox_trans_msg(adev, req, level, 0, 0);
+
+        r = xgpu_ai_poll_msg(adev, IDH_SUCCESS);
+        if (!r)
+                goto out;
+
+        r = xgpu_ai_poll_msg(adev, IDH_FAIL);
+        if (!r)
+                pr_info("DPM request failed");
+        else
+                pr_info("Mailbox is broken");
+
+out:
+        mutex_unlock(&adev->virt.dpm_mutex);
+        return r;
+}
+
 static int xgpu_ai_send_access_requests(struct amdgpu_device *adev,
 					enum idh_request req)
 {
@@ -375,4 +451,6 @@ const struct amdgpu_virt_ops xgpu_ai_virt_ops = {
 	.reset_gpu = xgpu_ai_request_reset,
 	.wait_reset = NULL,
 	.trans_msg = xgpu_ai_mailbox_trans_msg,
+	.get_pp_clk = xgpu_ai_get_pp_clk,
+	.force_dpm_level = xgpu_ai_force_dpm_level,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h
index b4a9ceea334b..39d151b79153 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h
@@ -35,6 +35,10 @@ enum idh_request {
 	IDH_REL_GPU_FINI_ACCESS,
 	IDH_REQ_GPU_RESET_ACCESS,
 
+	IDH_IRQ_FORCE_DPM_LEVEL = 10,
+	IDH_IRQ_GET_PP_SCLK,
+	IDH_IRQ_GET_PP_MCLK,
+
 	IDH_LOG_VF_ERROR       = 200,
 };
 
@@ -43,6 +47,8 @@ enum idh_event {
 	IDH_READY_TO_ACCESS_GPU,
 	IDH_FLR_NOTIFICATION,
 	IDH_FLR_NOTIFICATION_CMPL,
+	IDH_SUCCESS,
+	IDH_FAIL,
 	IDH_EVENT_MAX
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
index 6a0fcd67662a..aef9d059ae52 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
@@ -515,7 +515,7 @@ static void xgpu_vi_mailbox_flr_work(struct work_struct *work)
 
 	/* wait until RCV_MSG become 3 */
 	if (xgpu_vi_poll_msg(adev, IDH_FLR_NOTIFICATION_CMPL)) {
-		pr_err("failed to recieve FLR_CMPL\n");
+		pr_err("failed to receive FLR_CMPL\n");
 		return;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
index 2b3429d90690..b91df7bd1d98 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
@@ -674,7 +674,7 @@ static int psp_v11_0_xgmi_set_topology_info(struct psp_context *psp,
 	for (i = 0; i < topology_info_input->num_nodes; i++) {
 		topology_info_input->nodes[i].node_id = topology->nodes[i].node_id;
 		topology_info_input->nodes[i].num_hops = topology->nodes[i].num_hops;
-		topology_info_input->nodes[i].is_sharing_enabled = topology->nodes[i].is_sharing_enabled;
+		topology_info_input->nodes[i].is_sharing_enabled = 1;
 		topology_info_input->nodes[i].sdma_engine = topology->nodes[i].sdma_engine;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
index cca3552b36ed..36196372e8db 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
@@ -870,8 +870,8 @@ static int sdma_v2_4_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 	}
@@ -1006,7 +1006,7 @@ static int sdma_v2_4_set_trap_irq_state(struct amdgpu_device *adev,
 	u32 sdma_cntl;
 
 	switch (type) {
-	case AMDGPU_SDMA_IRQ_TRAP0:
+	case AMDGPU_SDMA_IRQ_INSTANCE0:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA0_REGISTER_OFFSET);
@@ -1022,7 +1022,7 @@ static int sdma_v2_4_set_trap_irq_state(struct amdgpu_device *adev,
 			break;
 		}
 		break;
-	case AMDGPU_SDMA_IRQ_TRAP1:
+	case AMDGPU_SDMA_IRQ_INSTANCE1:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA1_REGISTER_OFFSET);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index 0ce8331baeb2..6d39544e7829 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -1154,8 +1154,8 @@ static int sdma_v3_0_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 	}
@@ -1340,7 +1340,7 @@ static int sdma_v3_0_set_trap_irq_state(struct amdgpu_device *adev,
 	u32 sdma_cntl;
 
 	switch (type) {
-	case AMDGPU_SDMA_IRQ_TRAP0:
+	case AMDGPU_SDMA_IRQ_INSTANCE0:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA0_REGISTER_OFFSET);
@@ -1356,7 +1356,7 @@ static int sdma_v3_0_set_trap_irq_state(struct amdgpu_device *adev,
 			break;
 		}
 		break;
-	case AMDGPU_SDMA_IRQ_TRAP1:
+	case AMDGPU_SDMA_IRQ_INSTANCE1:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(mmSDMA0_CNTL + SDMA1_REGISTER_OFFSET);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 8691b621148e..9c88ce513d78 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -156,7 +156,6 @@ static const struct soc15_reg_golden golden_settings_sdma0_4_2[] =
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC7_RB_RPTR_ADDR_LO, 0xfffffffd, 0x00000001),
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
 	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_PAGE, 0x000003ff, 0x000003c0),
-	SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_WATERMK, 0xFE000000, 0x00000000),
 };
 
 static const struct soc15_reg_golden golden_settings_sdma1_4_2[] = {
@@ -186,7 +185,6 @@ static const struct soc15_reg_golden golden_settings_sdma1_4_2[] = {
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC7_RB_RPTR_ADDR_LO, 0xfffffffd, 0x00000001),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000),
 	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_PAGE, 0x000003ff, 0x000003c0),
-	SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_WATERMK, 0xFE000000, 0x00000000),
 };
 
 static const struct soc15_reg_golden golden_settings_sdma_rv1[] =
@@ -851,7 +849,7 @@ static void sdma_v4_0_gfx_resume(struct amdgpu_device *adev, unsigned int i)
 	wptr_poll_cntl = RREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL);
 	wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl,
 				       SDMA0_GFX_RB_WPTR_POLL_CNTL,
-				       F32_POLL_ENABLE, amdgpu_sriov_vf(adev));
+				       F32_POLL_ENABLE, amdgpu_sriov_vf(adev)? 1 : 0);
 	WREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, wptr_poll_cntl);
 
 	/* enable DMA RB */
@@ -942,7 +940,7 @@ static void sdma_v4_0_page_resume(struct amdgpu_device *adev, unsigned int i)
 	wptr_poll_cntl = RREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL);
 	wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl,
 				       SDMA0_PAGE_RB_WPTR_POLL_CNTL,
-				       F32_POLL_ENABLE, amdgpu_sriov_vf(adev));
+				       F32_POLL_ENABLE, amdgpu_sriov_vf(adev)? 1 : 0);
 	WREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL, wptr_poll_cntl);
 
 	/* enable DMA RB */
@@ -1518,7 +1516,7 @@ static int sdma_v4_0_late_init(void *handle)
 	int r;
 
 	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) {
-		amdgpu_ras_feature_enable(adev, &ras_block, 0);
+		amdgpu_ras_feature_enable_on_boot(adev, &ras_block, 0);
 		return 0;
 	}
 
@@ -1532,7 +1530,7 @@ static int sdma_v4_0_late_init(void *handle)
 
 	**ras_if = ras_block;
 
-	r = amdgpu_ras_feature_enable(adev, *ras_if, 1);
+	r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
 	if (r)
 		goto feature;
 
@@ -1551,13 +1549,13 @@ static int sdma_v4_0_late_init(void *handle)
 	if (r)
 		goto sysfs;
 resume:
-	r = amdgpu_irq_get(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_ECC0);
+	r = amdgpu_irq_get(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE0);
 	if (r)
 		goto irq;
 
-	r = amdgpu_irq_get(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_ECC1);
+	r = amdgpu_irq_get(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE1);
 	if (r) {
-		amdgpu_irq_put(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_ECC0);
+		amdgpu_irq_put(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE0);
 		goto irq;
 	}
 
@@ -1621,8 +1619,8 @@ static int sdma_v4_0_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 
@@ -1641,8 +1639,8 @@ static int sdma_v4_0_sw_init(void *handle)
 			r = amdgpu_ring_init(adev, ring, 1024,
 					     &adev->sdma.trap_irq,
 					     (i == 0) ?
-					     AMDGPU_SDMA_IRQ_TRAP0 :
-					     AMDGPU_SDMA_IRQ_TRAP1);
+					     AMDGPU_SDMA_IRQ_INSTANCE0 :
+					     AMDGPU_SDMA_IRQ_INSTANCE1);
 			if (r)
 				return r;
 		}
@@ -1709,8 +1707,8 @@ static int sdma_v4_0_hw_fini(void *handle)
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
-	amdgpu_irq_put(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_ECC0);
-	amdgpu_irq_put(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_ECC1);
+	amdgpu_irq_put(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE0);
+	amdgpu_irq_put(adev, &adev->sdma.ecc_irq, AMDGPU_SDMA_IRQ_INSTANCE1);
 
 	sdma_v4_0_ctx_switch_enable(adev, false);
 	sdma_v4_0_enable(adev, false);
@@ -1780,13 +1778,12 @@ static int sdma_v4_0_set_trap_irq_state(struct amdgpu_device *adev,
 					unsigned type,
 					enum amdgpu_interrupt_state state)
 {
-	unsigned int instance = (type == AMDGPU_SDMA_IRQ_TRAP0) ? 0 : 1;
 	u32 sdma_cntl;
 
-	sdma_cntl = RREG32_SDMA(instance, mmSDMA0_CNTL);
+	sdma_cntl = RREG32_SDMA(type, mmSDMA0_CNTL);
 	sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE,
 		       state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
-	WREG32_SDMA(instance, mmSDMA0_CNTL, sdma_cntl);
+	WREG32_SDMA(type, mmSDMA0_CNTL, sdma_cntl);
 
 	return 0;
 }
@@ -1866,10 +1863,16 @@ static int sdma_v4_0_process_ecc_irq(struct amdgpu_device *adev,
 				      struct amdgpu_irq_src *source,
 				      struct amdgpu_iv_entry *entry)
 {
+	struct ras_common_if *ras_if = adev->sdma.ras_if;
 	struct ras_dispatch_if ih_data = {
-		.head = *adev->sdma.ras_if,
 		.entry = entry,
 	};
+
+	if (!ras_if)
+		return 0;
+
+	ih_data.head = *ras_if;
+
 	amdgpu_ras_interrupt_dispatch(adev, &ih_data);
 	return 0;
 }
@@ -1908,7 +1911,7 @@ static int sdma_v4_0_set_ecc_irq_state(struct amdgpu_device *adev,
 {
 	u32 sdma_edc_config;
 
-	u32 reg_offset = (type == AMDGPU_SDMA_IRQ_ECC0) ?
+	u32 reg_offset = (type == AMDGPU_SDMA_IRQ_INSTANCE0) ?
 		sdma_v4_0_get_reg_offset(adev, 0, mmSDMA0_EDC_CONFIG) :
 		sdma_v4_0_get_reg_offset(adev, 1, mmSDMA0_EDC_CONFIG);
 
@@ -2292,6 +2295,7 @@ static void sdma_v4_0_set_vm_pte_funcs(struct amdgpu_device *adev)
 				&sched->sched_rq[DRM_SCHED_PRIORITY_KERNEL];
 		}
 		adev->vm_manager.vm_pte_num_rqs = adev->sdma.num_instances - 1;
+		adev->vm_manager.page_fault = &adev->sdma.instance[0].page;
 	} else {
 		for (i = 0; i < adev->sdma.num_instances; i++) {
 			sched = &adev->sdma.instance[i].ring.sched;
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c
index f15f196684ba..3eeefd40dae0 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dma.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c
@@ -503,8 +503,8 @@ static int si_dma_sw_init(void *handle)
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
 				     (i == 0) ?
-				     AMDGPU_SDMA_IRQ_TRAP0 :
-				     AMDGPU_SDMA_IRQ_TRAP1);
+				     AMDGPU_SDMA_IRQ_INSTANCE0 :
+				     AMDGPU_SDMA_IRQ_INSTANCE1);
 		if (r)
 			return r;
 	}
@@ -591,7 +591,7 @@ static int si_dma_set_trap_irq_state(struct amdgpu_device *adev,
 	u32 sdma_cntl;
 
 	switch (type) {
-	case AMDGPU_SDMA_IRQ_TRAP0:
+	case AMDGPU_SDMA_IRQ_INSTANCE0:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(DMA_CNTL + DMA0_REGISTER_OFFSET);
@@ -607,7 +607,7 @@ static int si_dma_set_trap_irq_state(struct amdgpu_device *adev,
 			break;
 		}
 		break;
-	case AMDGPU_SDMA_IRQ_TRAP1:
+	case AMDGPU_SDMA_IRQ_INSTANCE1:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
 			sdma_cntl = RREG32(DMA_CNTL + DMA1_REGISTER_OFFSET);
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index bdb5ad93990d..4900e4958dec 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -470,6 +470,12 @@ static int soc15_asic_reset(struct amdgpu_device *adev)
 	case CHIP_VEGA12:
 		soc15_asic_get_baco_capability(adev, &baco_reset);
 		break;
+	case CHIP_VEGA20:
+		if (adev->psp.sos_fw_version >= 0x80067)
+			soc15_asic_get_baco_capability(adev, &baco_reset);
+		else
+			baco_reset = false;
+		break;
 	default:
 		baco_reset = false;
 		break;
@@ -895,7 +901,8 @@ static int soc15_common_early_init(void *handle)
 
 			adev->pg_flags = AMD_PG_SUPPORT_SDMA | AMD_PG_SUPPORT_VCN;
 		} else if (adev->pdev->device == 0x15d8) {
-			adev->cg_flags = AMD_CG_SUPPORT_GFX_MGLS |
+			adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG |
+				AMD_CG_SUPPORT_GFX_MGLS |
 				AMD_CG_SUPPORT_GFX_CP_LS |
 				AMD_CG_SUPPORT_GFX_3D_CGCG |
 				AMD_CG_SUPPORT_GFX_3D_CGLS |
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
index bed78a778e3f..40363ca6c5f1 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
@@ -283,7 +283,7 @@ static int vce_v2_0_stop(struct amdgpu_device *adev)
 	}
 
 	if (vce_v2_0_wait_for_idle(adev)) {
-		DRM_INFO("VCE is busy, Can't set clock gateing");
+		DRM_INFO("VCE is busy, Can't set clock gating");
 		return 0;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
index aadc3e66ebd7..f3f5938430d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
@@ -382,6 +382,7 @@ static int vce_v4_0_start(struct amdgpu_device *adev)
 static int vce_v4_0_stop(struct amdgpu_device *adev)
 {
 
+	/* Disable VCPU */
 	WREG32_P(SOC15_REG_OFFSET(VCE, 0, mmVCE_VCPU_CNTL), 0, ~0x200001);
 
 	/* hold on ECPU */
@@ -389,8 +390,8 @@ static int vce_v4_0_stop(struct amdgpu_device *adev)
 			VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK,
 			~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);
 
-	/* clear BUSY flag */
-	WREG32_P(SOC15_REG_OFFSET(VCE, 0, mmVCE_STATUS), 0, ~VCE_STATUS__JOB_BUSY_MASK);
+	/* clear VCE_STATUS */
+	WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_STATUS), 0);
 
 	/* Set Clock-Gating off */
 	/* if (adev->cg_flags & AMD_CG_SUPPORT_VCE_MGCG)
@@ -922,6 +923,7 @@ static int vce_v4_0_set_clockgating_state(void *handle,
 
 	return 0;
 }
+#endif
 
 static int vce_v4_0_set_powergating_state(void *handle,
 					  enum amd_powergating_state state)
@@ -935,16 +937,11 @@ static int vce_v4_0_set_powergating_state(void *handle,
 	 */
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	if (!(adev->pg_flags & AMD_PG_SUPPORT_VCE))
-		return 0;
-
 	if (state == AMD_PG_STATE_GATE)
-		/* XXX do we need a vce_v4_0_stop()? */
-		return 0;
+		return vce_v4_0_stop(adev);
 	else
 		return vce_v4_0_start(adev);
 }
-#endif
 
 static void vce_v4_0_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_job *job,
 					struct amdgpu_ib *ib, uint32_t flags)
@@ -1059,7 +1056,7 @@ const struct amd_ip_funcs vce_v4_0_ip_funcs = {
 	.soft_reset = NULL /* vce_v4_0_soft_reset */,
 	.post_soft_reset = NULL /* vce_v4_0_post_soft_reset */,
 	.set_clockgating_state = vce_v4_0_set_clockgating_state,
-	.set_powergating_state = NULL /* vce_v4_0_set_powergating_state */,
+	.set_powergating_state = vce_v4_0_set_powergating_state,
 };
 
 static const struct amdgpu_ring_funcs vce_v4_0_ring_vm_funcs = {