Merge tag 'amd-drm-next-6.12-2024-09-13' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.12-2024-09-13:

amdgpu:
- GPUVM sync fixes
- kdoc fixes
- Misc spelling mistakes
- Add some raven GFXOFF quirks
- Use clamp helper
- DC fixes
- JPEG fixes
- Process isolation fix
- Queue reset fix
- W=1 cleanup
- SMU14 fixes
- JPEG fixes

amdkfd:
- Fetch cacheline info from IP discovery
- Queue reset fix
- RAS fix
- Document SVM events
- CRIU fixes
- Race fix in dma-buf handling

drm:
- dma-buf fd race fixes

Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240913134139.2861073-1-alexander.deucher@amd.com

50 files changed, 678 insertions, 366 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 38408e4e158e..c7b18c52825d 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -39,23 +39,7 @@ ccflags-y := -I$(FULL_AMD_PATH)/include/asic_reg \
 	-I$(FULL_AMD_DISPLAY_PATH)/amdgpu_dm \
 	-I$(FULL_AMD_PATH)/amdkfd
 
-subdir-ccflags-y := -Wextra
-subdir-ccflags-y += -Wunused
-subdir-ccflags-y += -Wmissing-prototypes
-subdir-ccflags-y += -Wmissing-declarations
-subdir-ccflags-y += -Wmissing-include-dirs
-subdir-ccflags-y += -Wold-style-definition
-subdir-ccflags-y += -Wmissing-format-attribute
-# Need this to avoid recursive variable evaluation issues
-cond-flags := $(call cc-option, -Wunused-but-set-variable) \
-	$(call cc-option, -Wunused-const-variable) \
-	$(call cc-option, -Wstringop-truncation) \
-	$(call cc-option, -Wpacked-not-aligned)
-subdir-ccflags-y += $(cond-flags)
-subdir-ccflags-y += -Wno-unused-parameter
-subdir-ccflags-y += -Wno-type-limits
-subdir-ccflags-y += -Wno-sign-compare
-subdir-ccflags-y += -Wno-missing-field-initializers
+# Locally disable W=1 warnings enabled in drm subsystem Makefile
 subdir-ccflags-y += -Wno-override-init
 subdir-ccflags-$(CONFIG_DRM_AMDGPU_WERROR) += -Werror
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index e095572458cd..dcd59040c449 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -237,6 +237,7 @@ extern int sched_policy;
 extern bool debug_evictions;
 extern bool no_system_mem_limit;
 extern int halt_if_hws_hang;
+extern uint amdgpu_svm_default_granularity;
 #else
 static const int __maybe_unused sched_policy = KFD_SCHED_POLICY_HWS;
 static const bool __maybe_unused debug_evictions; /* = false */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index 73b2b401b450..9435af2e6bdc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -20,7 +20,6 @@
  * OTHER DEALINGS IN THE SOFTWARE.
  */
 #include <linux/module.h>
-#include <linux/fdtable.h>
 #include <linux/uaccess.h>
 #include <linux/firmware.h>
 #include "amdgpu.h"
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 6d5fd371d5ce..4afef5b46c7d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -25,7 +25,6 @@
 #include <linux/pagemap.h>
 #include <linux/sched/mm.h>
 #include <linux/sched/task.h>
-#include <linux/fdtable.h>
 #include <drm/ttm/ttm_tt.h>
 
 #include <drm/drm_exec.h>
@@ -818,18 +817,13 @@ static int kfd_mem_export_dmabuf(struct kgd_mem *mem)
 	if (!mem->dmabuf) {
 		struct amdgpu_device *bo_adev;
 		struct dma_buf *dmabuf;
-		int r, fd;
 
 		bo_adev = amdgpu_ttm_adev(mem->bo->tbo.bdev);
-		r = drm_gem_prime_handle_to_fd(&bo_adev->ddev, bo_adev->kfd.client.file,
+		dmabuf = drm_gem_prime_handle_to_dmabuf(&bo_adev->ddev, bo_adev->kfd.client.file,
 					       mem->gem_handle,
 			mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE ?
-					       DRM_RDWR : 0, &fd);
-		if (r)
-			return r;
-		dmabuf = dma_buf_get(fd);
-		close_fd(fd);
-		if (WARN_ON_ONCE(IS_ERR(dmabuf)))
+					       DRM_RDWR : 0);
+		if (IS_ERR(dmabuf))
 			return PTR_ERR(dmabuf);
 		mem->dmabuf = dmabuf;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 8dee7c62c801..86cff30d5c4e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -169,6 +169,16 @@ uint amdgpu_sdma_phase_quantum = 32;
 char *amdgpu_disable_cu;
 char *amdgpu_virtual_display;
 bool enforce_isolation;
+
+/* Specifies the default granularity for SVM, used in buffer
+ * migration and restoration of backing memory when handling
+ * recoverable page faults.
+ *
+ * The value is given as log(numPages(buffer)); for a 2 MiB
+ * buffer it computes to be 9
+ */
+uint amdgpu_svm_default_granularity = 9;
+
 /*
  * OverDrive(bit 14) disabled by default
  * GFX DCS(bit 19) disabled by default
@@ -321,6 +331,13 @@ MODULE_PARM_DESC(msi, "MSI support (1 = enable, 0 = disable, -1 = auto)");
 module_param_named(msi, amdgpu_msi, int, 0444);
 
 /**
+ * DOC: svm_default_granularity (uint)
+ * Used in buffer migration and handling of recoverable page faults
+ */
+MODULE_PARM_DESC(svm_default_granularity, "SVM's default granularity in log(2^Pages), default 9 = 2^9 = 2 MiB");
+module_param_named(svm_default_granularity, amdgpu_svm_default_granularity, uint, 0644);
+
+/**
  * DOC: lockup_timeout (string)
  * Set GPU scheduler timeout value in ms.
  *
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index b779d47a546a..83e54697f0ee 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1397,14 +1397,23 @@ static ssize_t amdgpu_gfx_get_available_compute_partition(struct device *dev,
 static int amdgpu_gfx_run_cleaner_shader_job(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	long timeout = msecs_to_jiffies(1000);
-	struct dma_fence *f = NULL;
+	struct drm_gpu_scheduler *sched = &ring->sched;
+	struct drm_sched_entity entity;
+	struct dma_fence *f;
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
 	int i, r;
 
-	r = amdgpu_job_alloc_with_ib(adev, NULL, NULL,
-				     64, AMDGPU_IB_POOL_DIRECT,
+	/* Initialize the scheduler entity */
+	r = drm_sched_entity_init(&entity, DRM_SCHED_PRIORITY_NORMAL,
+				  &sched, 1, NULL);
+	if (r) {
+		dev_err(adev->dev, "Failed setting up GFX kernel entity.\n");
+		goto err;
+	}
+
+	r = amdgpu_job_alloc_with_ib(ring->adev, &entity, NULL,
+				     64, 0,
 				     &job);
 	if (r)
 		goto err;
@@ -1416,24 +1425,18 @@ static int amdgpu_gfx_run_cleaner_shader_job(struct amdgpu_ring *ring)
 		ib->ptr[i] = ring->funcs->nop;
 	ib->length_dw = ring->funcs->align_mask + 1;
 
-	r = amdgpu_job_submit_direct(job, ring, &f);
-	if (r)
-		goto err_free;
+	f = amdgpu_job_submit(job);
 
-	r = dma_fence_wait_timeout(f, false, timeout);
-	if (r == 0)
-		r = -ETIMEDOUT;
-	else if (r > 0)
-		r = 0;
+	r = dma_fence_wait(f, false);
+	if (r)
+		goto err;
 
-	amdgpu_ib_free(adev, ib, f);
 	dma_fence_put(f);
 
+	/* Clean up the scheduler entity */
+	drm_sched_entity_destroy(&entity);
 	return 0;
 
-err_free:
-	amdgpu_job_free(job);
-	amdgpu_ib_free(adev, ib, f);
 err:
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index f7d5d4f08a53..10b61ff63802 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -832,6 +832,7 @@ int amdgpu_mes_reset_hw_queue_mmio(struct amdgpu_device *adev, int queue_type,
 	struct mes_reset_queue_input queue_input;
 	int r;
 
+	queue_input.queue_type = queue_type;
 	queue_input.use_mmio = true;
 	queue_input.me_id = me_id;
 	queue_input.pipe_id = pipe_id;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pll.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pll.c
index 0bb2466d539a..675aa138ea11 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pll.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pll.c
@@ -94,7 +94,7 @@ static void amdgpu_pll_get_fb_ref_div(struct amdgpu_device *adev, unsigned int n
 		ref_div_max = min(128 / post_div, ref_div_max);
 
 	/* get matching reference and feedback divider */
-	*ref_div = min(max(DIV_ROUND_CLOSEST(den, post_div), 1u), ref_div_max);
+	*ref_div = clamp(DIV_ROUND_CLOSEST(den, post_div), 1u, ref_div_max);
 	*fb_div = DIV_ROUND_CLOSEST(nom * *ref_div * post_div, den);
 
 	/* limit fb divider to its maximum */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
index 863b2a34b2d6..f9ff493c100e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
@@ -22,7 +22,6 @@
  * Authors: Andres Rodriguez <andresx7@gmail.com>
  */
 
-#include <linux/fdtable.h>
 #include <linux/file.h>
 #include <linux/pid.h>
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index e20d19ae01b2..2452dfa6314f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -844,7 +844,7 @@ int amdgpu_vm_update_pdes(struct amdgpu_device *adev,
 	params.vm = vm;
 	params.immediate = immediate;
 
-	r = vm->update_funcs->prepare(&params, NULL, AMDGPU_SYNC_EXPLICIT);
+	r = vm->update_funcs->prepare(&params, NULL);
 	if (r)
 		goto error;
 
@@ -908,10 +908,12 @@ amdgpu_vm_tlb_flush(struct amdgpu_vm_update_params *params,
 {
 	struct amdgpu_vm *vm = params->vm;
 
-	if (!fence || !*fence)
+	tlb_cb->vm = vm;
+	if (!fence || !*fence) {
+		amdgpu_vm_tlb_seq_cb(NULL, &tlb_cb->cb);
 		return;
+	}
 
-	tlb_cb->vm = vm;
 	if (!dma_fence_add_callback(*fence, &tlb_cb->cb,
 				    amdgpu_vm_tlb_seq_cb)) {
 		dma_fence_put(vm->last_tlb_flush);
@@ -939,7 +941,7 @@ amdgpu_vm_tlb_flush(struct amdgpu_vm_update_params *params,
  * @unlocked: unlocked invalidation during MM callback
  * @flush_tlb: trigger tlb invalidation after update completed
  * @allow_override: change MTYPE for local NUMA nodes
- * @resv: fences we need to sync to
+ * @sync: fences we need to sync to
  * @start: start of mapped range
  * @last: last mapped entry
  * @flags: flags for the entries
@@ -955,16 +957,16 @@ amdgpu_vm_tlb_flush(struct amdgpu_vm_update_params *params,
  * 0 for success, negative erro code for failure.
  */
 int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
-			   bool immediate, bool unlocked, bool flush_tlb, bool allow_override,
-			   struct dma_resv *resv, uint64_t start, uint64_t last,
-			   uint64_t flags, uint64_t offset, uint64_t vram_base,
+			   bool immediate, bool unlocked, bool flush_tlb,
+			   bool allow_override, struct amdgpu_sync *sync,
+			   uint64_t start, uint64_t last, uint64_t flags,
+			   uint64_t offset, uint64_t vram_base,
 			   struct ttm_resource *res, dma_addr_t *pages_addr,
 			   struct dma_fence **fence)
 {
 	struct amdgpu_vm_tlb_seq_struct *tlb_cb;
 	struct amdgpu_vm_update_params params;
 	struct amdgpu_res_cursor cursor;
-	enum amdgpu_sync_mode sync_mode;
 	int r, idx;
 
 	if (!drm_dev_enter(adev_to_drm(adev), &idx))
@@ -997,14 +999,6 @@ int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	params.allow_override = allow_override;
 	INIT_LIST_HEAD(&params.tlb_flush_waitlist);
 
-	/* Implicitly sync to command submissions in the same VM before
-	 * unmapping. Sync to moving fences before mapping.
-	 */
-	if (!(flags & AMDGPU_PTE_VALID))
-		sync_mode = AMDGPU_SYNC_EQ_OWNER;
-	else
-		sync_mode = AMDGPU_SYNC_EXPLICIT;
-
 	amdgpu_vm_eviction_lock(vm);
 	if (vm->evicting) {
 		r = -EBUSY;
@@ -1019,7 +1013,7 @@ int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		dma_fence_put(tmp);
 	}
 
-	r = vm->update_funcs->prepare(&params, resv, sync_mode);
+	r = vm->update_funcs->prepare(&params, sync);
 	if (r)
 		goto error_free;
 
@@ -1161,23 +1155,30 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 	struct amdgpu_bo *bo = bo_va->base.bo;
 	struct amdgpu_vm *vm = bo_va->base.vm;
 	struct amdgpu_bo_va_mapping *mapping;
+	struct dma_fence **last_update;
 	dma_addr_t *pages_addr = NULL;
 	struct ttm_resource *mem;
-	struct dma_fence **last_update;
+	struct amdgpu_sync sync;
 	bool flush_tlb = clear;
-	bool uncached;
-	struct dma_resv *resv;
 	uint64_t vram_base;
 	uint64_t flags;
+	bool uncached;
 	int r;
 
+	amdgpu_sync_create(&sync);
 	if (clear || !bo) {
 		mem = NULL;
-		resv = vm->root.bo->tbo.base.resv;
+
+		/* Implicitly sync to command submissions in the same VM before
+		 * unmapping.
+		 */
+		r = amdgpu_sync_resv(adev, &sync, vm->root.bo->tbo.base.resv,
+				     AMDGPU_SYNC_EQ_OWNER, vm);
+		if (r)
+			goto error_free;
 	} else {
 		struct drm_gem_object *obj = &bo->tbo.base;
 
-		resv = bo->tbo.base.resv;
 		if (obj->import_attach && bo_va->is_xgmi) {
 			struct dma_buf *dma_buf = obj->import_attach->dmabuf;
 			struct drm_gem_object *gobj = dma_buf->priv;
@@ -1191,6 +1192,12 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 		if (mem && (mem->mem_type == TTM_PL_TT ||
 			    mem->mem_type == AMDGPU_PL_PREEMPT))
 			pages_addr = bo->tbo.ttm->dma_address;
+
+		/* Implicitly sync to moving fences before mapping anything */
+		r = amdgpu_sync_resv(adev, &sync, bo->tbo.base.resv,
+				     AMDGPU_SYNC_EXPLICIT, vm);
+		if (r)
+			goto error_free;
 	}
 
 	if (bo) {
@@ -1240,12 +1247,12 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 		trace_amdgpu_vm_bo_update(mapping);
 
 		r = amdgpu_vm_update_range(adev, vm, false, false, flush_tlb,
-					   !uncached, resv, mapping->start, mapping->last,
-					   update_flags, mapping->offset,
-					   vram_base, mem, pages_addr,
-					   last_update);
+					   !uncached, &sync, mapping->start,
+					   mapping->last, update_flags,
+					   mapping->offset, vram_base, mem,
+					   pages_addr, last_update);
 		if (r)
-			return r;
+			goto error_free;
 	}
 
 	/* If the BO is not in its preferred location add it back to
@@ -1273,7 +1280,9 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 			trace_amdgpu_vm_bo_mapping(mapping);
 	}
 
-	return 0;
+error_free:
+	amdgpu_sync_free(&sync);
+	return r;
 }
 
 /**
@@ -1420,25 +1429,34 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 			  struct amdgpu_vm *vm,
 			  struct dma_fence **fence)
 {
-	struct dma_resv *resv = vm->root.bo->tbo.base.resv;
 	struct amdgpu_bo_va_mapping *mapping;
-	uint64_t init_pte_value = 0;
 	struct dma_fence *f = NULL;
+	struct amdgpu_sync sync;
 	int r;
 
+
+	/*
+	 * Implicitly sync to command submissions in the same VM before
+	 * unmapping.
+	 */
+	amdgpu_sync_create(&sync);
+	r = amdgpu_sync_resv(adev, &sync, vm->root.bo->tbo.base.resv,
+			     AMDGPU_SYNC_EQ_OWNER, vm);
+	if (r)
+		goto error_free;
+
 	while (!list_empty(&vm->freed)) {
 		mapping = list_first_entry(&vm->freed,
 			struct amdgpu_bo_va_mapping, list);
 		list_del(&mapping->list);
 
 		r = amdgpu_vm_update_range(adev, vm, false, false, true, false,
-					   resv, mapping->start, mapping->last,
-					   init_pte_value, 0, 0, NULL, NULL,
-					   &f);
+					   &sync, mapping->start, mapping->last,
+					   0, 0, 0, NULL, NULL, &f);
 		amdgpu_vm_free_mapping(adev, vm, mapping, f);
 		if (r) {
 			dma_fence_put(f);
-			return r;
+			goto error_free;
 		}
 	}
 
@@ -1449,7 +1467,9 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 		dma_fence_put(f);
 	}
 
-	return 0;
+error_free:
+	amdgpu_sync_free(&sync);
+	return r;
 
 }
 
@@ -2224,7 +2244,7 @@ void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint32_t min_vm_size,
 		phys_ram_gb = ((uint64_t)si.totalram * si.mem_unit +
 			       (1 << 30) - 1) >> 30;
 		vm_size = roundup_pow_of_two(
-			min(max(phys_ram_gb * 3, min_vm_size), max_size));
+			clamp(phys_ram_gb * 3, min_vm_size, max_size));
 	}
 
 	adev->vm_manager.max_pfn = (uint64_t)vm_size << 18;
@@ -2403,7 +2423,6 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm)
 int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		   int32_t xcp_id)
 {
-	struct amdgpu_ip_block *ip_block;
 	struct amdgpu_bo *root_bo;
 	struct amdgpu_bo_vm *root;
 	int r, i;
@@ -2435,11 +2454,6 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode &
 				    AMDGPU_VM_USE_CPU_FOR_GFX);
 
-	/* use CPU for page table update if SDMA is unavailable */
-	ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_SDMA);
-	if (!ip_block || ip_block->status.valid == false)
-		vm->use_cpu_for_update = true;
-
 	DRM_DEBUG_DRIVER("VM update mode is %s\n",
 			 vm->use_cpu_for_update ? "CPU" : "SDMA");
 	WARN_ONCE((vm->use_cpu_for_update &&
@@ -2770,6 +2784,7 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
  * amdgpu_vm_handle_fault - graceful handling of VM faults.
  * @adev: amdgpu device pointer
  * @pasid: PASID of the VM
+ * @ts: Timestamp of the fault
  * @vmid: VMID, only used for GFX 9.4.3.
  * @node_id: Node_id received in IH cookie. Only applicable for
  *           GFX 9.4.3.
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index d12d66dca8e9..52dd7cdfdc81 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -304,8 +304,8 @@ struct amdgpu_vm_update_params {
 
 struct amdgpu_vm_update_funcs {
 	int (*map_table)(struct amdgpu_bo_vm *bo);
-	int (*prepare)(struct amdgpu_vm_update_params *p, struct dma_resv *resv,
-		       enum amdgpu_sync_mode sync_mode);
+	int (*prepare)(struct amdgpu_vm_update_params *p,
+		       struct amdgpu_sync *sync);
 	int (*update)(struct amdgpu_vm_update_params *p,
 		      struct amdgpu_bo_vm *bo, uint64_t pe, uint64_t addr,
 		      unsigned count, uint32_t incr, uint64_t flags);
@@ -505,9 +505,10 @@ int amdgpu_vm_flush_compute_tlb(struct amdgpu_device *adev,
 void amdgpu_vm_bo_base_init(struct amdgpu_vm_bo_base *base,
 			    struct amdgpu_vm *vm, struct amdgpu_bo *bo);
 int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
-			   bool immediate, bool unlocked, bool flush_tlb, bool allow_override,
-			   struct dma_resv *resv, uint64_t start, uint64_t last,
-			   uint64_t flags, uint64_t offset, uint64_t vram_base,
+			   bool immediate, bool unlocked, bool flush_tlb,
+			   bool allow_override, struct amdgpu_sync *sync,
+			   uint64_t start, uint64_t last, uint64_t flags,
+			   uint64_t offset, uint64_t vram_base,
 			   struct ttm_resource *res, dma_addr_t *pages_addr,
 			   struct dma_fence **fence);
 int amdgpu_vm_bo_update(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
index 3895bd7d176a..0c1ef5850a5e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
@@ -39,20 +39,18 @@ static int amdgpu_vm_cpu_map_table(struct amdgpu_bo_vm *table)
  * amdgpu_vm_cpu_prepare - prepare page table update with the CPU
  *
  * @p: see amdgpu_vm_update_params definition
- * @resv: reservation object with embedded fence
- * @sync_mode: synchronization mode
+ * @sync: sync obj with fences to wait on
  *
  * Returns:
  * Negativ errno, 0 for success.
  */
 static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p,
-				 struct dma_resv *resv,
-				 enum amdgpu_sync_mode sync_mode)
+				 struct amdgpu_sync *sync)
 {
-	if (!resv)
+	if (!sync)
 		return 0;
 
-	return amdgpu_bo_sync_wait_resv(p->adev, resv, sync_mode, p->vm, true);
+	return amdgpu_sync_wait(sync, true);
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
index e39d6e7643bf..a076f43097e4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@@ -403,7 +403,7 @@ int amdgpu_vm_pt_clear(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	params.vm = vm;
 	params.immediate = immediate;
 
-	r = vm->update_funcs->prepare(&params, NULL, AMDGPU_SYNC_EXPLICIT);
+	r = vm->update_funcs->prepare(&params, NULL);
 	if (r)
 		goto exit;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
index 9b748d7058b5..4772fba33285 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
@@ -77,32 +77,24 @@ static int amdgpu_vm_sdma_alloc_job(struct amdgpu_vm_update_params *p,
  * amdgpu_vm_sdma_prepare - prepare SDMA command submission
  *
  * @p: see amdgpu_vm_update_params definition
- * @resv: reservation object with embedded fence
- * @sync_mode: synchronization mode
+ * @sync: amdgpu_sync object with fences to wait for
  *
  * Returns:
  * Negativ errno, 0 for success.
  */
 static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p,
-				  struct dma_resv *resv,
-				  enum amdgpu_sync_mode sync_mode)
+				  struct amdgpu_sync *sync)
 {
-	struct amdgpu_sync sync;
 	int r;
 
 	r = amdgpu_vm_sdma_alloc_job(p, 0);
 	if (r)
 		return r;
 
-	if (!resv)
+	if (!sync)
 		return 0;
 
-	amdgpu_sync_create(&sync);
-	r = amdgpu_sync_resv(p->adev, &sync, resv, sync_mode, p->vm);
-	if (!r)
-		r = amdgpu_sync_push_to_job(&sync, p->job);
-	amdgpu_sync_free(&sync);
-
+	r = amdgpu_sync_push_to_job(sync, p->job);
 	if (r) {
 		p->num_dw_left = 0;
 		amdgpu_job_free(p->job);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 8cf5d7925b51..23f0573ae47b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1345,6 +1345,10 @@ static const struct amdgpu_gfxoff_quirk amdgpu_gfxoff_quirk_list[] = {
 	{ 0x1002, 0x15dd, 0x1002, 0x15dd, 0xc6 },
 	/* Apple MacBook Pro (15-inch, 2019) Radeon Pro Vega 20 4 GB */
 	{ 0x1002, 0x69af, 0x106b, 0x019a, 0xc0 },
+	/* https://bbs.openkylin.top/t/topic/171497 */
+	{ 0x1002, 0x15d8, 0x19e5, 0x3e14, 0xc2 },
+	/* HP 705G4 DM with R5 2400G */
+	{ 0x1002, 0x15dd, 0x103c, 0x8464, 0xd6 },
 	{ 0, 0, 0, 0, 0 },
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
index 71f43a5c7f72..6e0e88076224 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
@@ -23,6 +23,7 @@
 
 #include "amdgpu.h"
 #include "amdgpu_jpeg.h"
+#include "amdgpu_cs.h"
 #include "soc15.h"
 #include "soc15d.h"
 #include "vcn_v1_0.h"
@@ -34,6 +35,9 @@
 static void jpeg_v1_0_set_dec_ring_funcs(struct amdgpu_device *adev);
 static void jpeg_v1_0_set_irq_funcs(struct amdgpu_device *adev);
 static void jpeg_v1_0_ring_begin_use(struct amdgpu_ring *ring);
+static int jpeg_v1_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
+				     struct amdgpu_job *job,
+				     struct amdgpu_ib *ib);
 
 static void jpeg_v1_0_decode_ring_patch_wreg(struct amdgpu_ring *ring, uint32_t *ptr, uint32_t reg_offset, uint32_t val)
 {
@@ -300,7 +304,10 @@ static void jpeg_v1_0_decode_ring_emit_ib(struct amdgpu_ring *ring,
 
 	amdgpu_ring_write(ring,
 		PACKETJ(SOC15_REG_OFFSET(JPEG, 0, mmUVD_LMI_JRBC_IB_VMID), 0, 0, PACKETJ_TYPE0));
-	amdgpu_ring_write(ring, (vmid | (vmid << 4)));
+	if (ring->funcs->parse_cs)
+		amdgpu_ring_write(ring, 0);
+	else
+		amdgpu_ring_write(ring, (vmid | (vmid << 4)));
 
 	amdgpu_ring_write(ring,
 		PACKETJ(SOC15_REG_OFFSET(JPEG, 0, mmUVD_LMI_JPEG_VMID), 0, 0, PACKETJ_TYPE0));
@@ -554,6 +561,7 @@ static const struct amdgpu_ring_funcs jpeg_v1_0_decode_ring_vm_funcs = {
 	.get_rptr = jpeg_v1_0_decode_ring_get_rptr,
 	.get_wptr = jpeg_v1_0_decode_ring_get_wptr,
 	.set_wptr = jpeg_v1_0_decode_ring_set_wptr,
+	.parse_cs = jpeg_v1_dec_ring_parse_cs,
 	.emit_frame_size =
 		6 + 6 + /* hdp invalidate / flush */
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
@@ -611,3 +619,69 @@ static void jpeg_v1_0_ring_begin_use(struct amdgpu_ring *ring)
 
 	vcn_v1_0_set_pg_for_begin_use(ring, set_clocks);
 }
+
+/**
+ * jpeg_v1_dec_ring_parse_cs - command submission parser
+ *
+ * @parser: Command submission parser context
+ * @job: the job to parse
+ * @ib: the IB to parse
+ *
+ * Parse the command stream, return -EINVAL for invalid packet,
+ * 0 otherwise
+ */
+static int jpeg_v1_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
+				     struct amdgpu_job *job,
+				     struct amdgpu_ib *ib)
+{
+	u32 i, reg, res, cond, type;
+	int ret = 0;
+	struct amdgpu_device *adev = parser->adev;
+
+	for (i = 0; i < ib->length_dw ; i += 2) {
+		reg  = CP_PACKETJ_GET_REG(ib->ptr[i]);
+		res  = CP_PACKETJ_GET_RES(ib->ptr[i]);
+		cond = CP_PACKETJ_GET_COND(ib->ptr[i]);
+		type = CP_PACKETJ_GET_TYPE(ib->ptr[i]);
+
+		if (res || cond != PACKETJ_CONDITION_CHECK0) /* only allow 0 for now */
+			return -EINVAL;
+
+		if (reg >= JPEG_V1_REG_RANGE_START && reg <= JPEG_V1_REG_RANGE_END)
+			continue;
+
+		switch (type) {
+		case PACKETJ_TYPE0:
+			if (reg != JPEG_V1_LMI_JPEG_WRITE_64BIT_BAR_HIGH &&
+			    reg != JPEG_V1_LMI_JPEG_WRITE_64BIT_BAR_LOW &&
+			    reg != JPEG_V1_LMI_JPEG_READ_64BIT_BAR_HIGH &&
+			    reg != JPEG_V1_LMI_JPEG_READ_64BIT_BAR_LOW &&
+			    reg != JPEG_V1_REG_CTX_INDEX &&
+			    reg != JPEG_V1_REG_CTX_DATA) {
+				ret = -EINVAL;
+			}
+			break;
+		case PACKETJ_TYPE1:
+			if (reg != JPEG_V1_REG_CTX_DATA)
+				ret = -EINVAL;
+			break;
+		case PACKETJ_TYPE3:
+			if (reg != JPEG_V1_REG_SOFT_RESET)
+				ret = -EINVAL;
+			break;
+		case PACKETJ_TYPE6:
+			if (ib->ptr[i] != CP_PACKETJ_NOP)
+				ret = -EINVAL;
+			break;
+		default:
+			ret = -EINVAL;
+		}
+
+		if (ret) {
+			dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+			break;
+		}
+	}
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.h
index bbf33a6a3972..9654d22e0376 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.h
@@ -29,4 +29,15 @@ int jpeg_v1_0_sw_init(void *handle);
 void jpeg_v1_0_sw_fini(void *handle);
 void jpeg_v1_0_start(struct amdgpu_device *adev, int mode);
 
+#define JPEG_V1_REG_RANGE_START	0x8000
+#define JPEG_V1_REG_RANGE_END	0x803f
+
+#define JPEG_V1_LMI_JPEG_WRITE_64BIT_BAR_HIGH	0x8238
+#define JPEG_V1_LMI_JPEG_WRITE_64BIT_BAR_LOW	0x8239
+#define JPEG_V1_LMI_JPEG_READ_64BIT_BAR_HIGH	0x825a
+#define JPEG_V1_LMI_JPEG_READ_64BIT_BAR_LOW	0x825b
+#define JPEG_V1_REG_CTX_INDEX			0x8328
+#define JPEG_V1_REG_CTX_DATA			0x8329
+#define JPEG_V1_REG_SOFT_RESET			0x83a0
+
 #endif /*__JPEG_V1_0_H__*/
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
index 98aa3ccd0d20..41c0f8750dc1 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
@@ -23,6 +23,7 @@
 
 #include "amdgpu.h"
 #include "amdgpu_jpeg.h"
+#include "amdgpu_cs.h"
 #include "amdgpu_pm.h"
 #include "soc15.h"
 #include "soc15d.h"
@@ -538,7 +539,11 @@ void jpeg_v2_0_dec_ring_emit_ib(struct amdgpu_ring *ring,
 
 	amdgpu_ring_write(ring, PACKETJ(mmUVD_LMI_JRBC_IB_VMID_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
-	amdgpu_ring_write(ring, (vmid | (vmid << 4) | (vmid << 8)));
+
+	if (ring->funcs->parse_cs)
+		amdgpu_ring_write(ring, 0);
+	else
+		amdgpu_ring_write(ring, (vmid | (vmid << 4) | (vmid << 8)));
 
 	amdgpu_ring_write(ring, PACKETJ(mmUVD_LMI_JPEG_VMID_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
@@ -764,6 +769,7 @@ static const struct amdgpu_ring_funcs jpeg_v2_0_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v2_0_dec_ring_get_rptr,
 	.get_wptr = jpeg_v2_0_dec_ring_get_wptr,
 	.set_wptr = jpeg_v2_0_dec_ring_set_wptr,
+	.parse_cs = jpeg_v2_dec_ring_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
@@ -810,3 +816,58 @@ const struct amdgpu_ip_block_version jpeg_v2_0_ip_block = {
 		.rev = 0,
 		.funcs = &jpeg_v2_0_ip_funcs,
 };
+
+/**
+ * jpeg_v2_dec_ring_parse_cs - command submission parser
+ *
+ * @parser: Command submission parser context
+ * @job: the job to parse
+ * @ib: the IB to parse
+ *
+ * Parse the command stream, return -EINVAL for invalid packet,
+ * 0 otherwise
+ */
+int jpeg_v2_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
+			      struct amdgpu_job *job,
+			      struct amdgpu_ib *ib)
+{
+	u32 i, reg, res, cond, type;
+	struct amdgpu_device *adev = parser->adev;
+
+	for (i = 0; i < ib->length_dw ; i += 2) {
+		reg  = CP_PACKETJ_GET_REG(ib->ptr[i]);
+		res  = CP_PACKETJ_GET_RES(ib->ptr[i]);
+		cond = CP_PACKETJ_GET_COND(ib->ptr[i]);
+		type = CP_PACKETJ_GET_TYPE(ib->ptr[i]);
+
+		if (res) /* only support 0 at the moment */
+			return -EINVAL;
+
+		switch (type) {
+		case PACKETJ_TYPE0:
+			if (cond != PACKETJ_CONDITION_CHECK0 || reg < JPEG_REG_RANGE_START ||
+			    reg > JPEG_REG_RANGE_END) {
+				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+				return -EINVAL;
+			}
+			break;
+		case PACKETJ_TYPE3:
+			if (cond != PACKETJ_CONDITION_CHECK3 || reg < JPEG_REG_RANGE_START ||
+			    reg > JPEG_REG_RANGE_END) {
+				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+				return -EINVAL;
+			}
+			break;
+		case PACKETJ_TYPE6:
+			if (ib->ptr[i] == CP_PACKETJ_NOP)
+				continue;
+			dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+			return -EINVAL;
+		default:
+			dev_err(adev->dev, "Unknown packet type %d !\n", type);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.h
index 654e43e83e2c..63fadda7a673 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.h
@@ -45,6 +45,9 @@
 
 #define JRBC_DEC_EXTERNAL_REG_WRITE_ADDR				0x18000
 
+#define JPEG_REG_RANGE_START						0x4000
+#define JPEG_REG_RANGE_END						0x41c2
+
 void jpeg_v2_0_dec_ring_insert_start(struct amdgpu_ring *ring);
 void jpeg_v2_0_dec_ring_insert_end(struct amdgpu_ring *ring);
 void jpeg_v2_0_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq,
@@ -57,6 +60,9 @@ void jpeg_v2_0_dec_ring_emit_vm_flush(struct amdgpu_ring *ring,
 				unsigned vmid, uint64_t pd_addr);
 void jpeg_v2_0_dec_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t val);
 void jpeg_v2_0_dec_ring_nop(struct amdgpu_ring *ring, uint32_t count);
+int jpeg_v2_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
+			      struct amdgpu_job *job,
+			      struct amdgpu_ib *ib);
 
 extern const struct amdgpu_ip_block_version jpeg_v2_0_ip_block;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
index d8ef95c847c2..eedb9a829d95 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
@@ -662,6 +662,7 @@ static const struct amdgpu_ring_funcs jpeg_v2_5_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v2_5_dec_ring_get_rptr,
 	.get_wptr = jpeg_v2_5_dec_ring_get_wptr,
 	.set_wptr = jpeg_v2_5_dec_ring_set_wptr,
+	.parse_cs = jpeg_v2_dec_ring_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
@@ -691,6 +692,7 @@ static const struct amdgpu_ring_funcs jpeg_v2_6_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v2_5_dec_ring_get_rptr,
 	.get_wptr = jpeg_v2_5_dec_ring_get_wptr,
 	.set_wptr = jpeg_v2_5_dec_ring_set_wptr,
+	.parse_cs = jpeg_v2_dec_ring_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
index 31cfa3ce6528..b1e7fd25afbc 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
@@ -560,6 +560,7 @@ static const struct amdgpu_ring_funcs jpeg_v3_0_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v3_0_dec_ring_get_rptr,
 	.get_wptr = jpeg_v3_0_dec_ring_get_wptr,
 	.set_wptr = jpeg_v3_0_dec_ring_set_wptr,
+	.parse_cs = jpeg_v2_dec_ring_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
index 3dac8f259d7f..6c5c1a68a9b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
@@ -727,6 +727,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v4_0_dec_ring_get_rptr,
 	.get_wptr = jpeg_v4_0_dec_ring_get_wptr,
 	.set_wptr = jpeg_v4_0_dec_ring_set_wptr,
+	.parse_cs = jpeg_v2_dec_ring_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.h
index 07d36c2abd6b..47638fd4d4e2 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.h
@@ -32,5 +32,4 @@ enum amdgpu_jpeg_v4_0_sub_block {
 };
 
 extern const struct amdgpu_ip_block_version jpeg_v4_0_ip_block;
-
 #endif /* __JPEG_V4_0_H__ */
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index 6ae5a784e187..86958cb2c2ab 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -23,9 +23,9 @@
 
 #include "amdgpu.h"
 #include "amdgpu_jpeg.h"
-#include "amdgpu_cs.h"
 #include "soc15.h"
 #include "soc15d.h"
+#include "jpeg_v2_0.h"
 #include "jpeg_v4_0_3.h"
 #include "mmsch_v4_0_3.h"
 
@@ -59,6 +59,12 @@ static int amdgpu_ih_srcid_jpeg[] = {
 	VCN_4_0__SRCID__JPEG7_DECODE
 };
 
+static inline bool jpeg_v4_0_3_normalizn_reqd(struct amdgpu_device *adev)
+{
+	return amdgpu_sriov_vf(adev) ||
+	       (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4));
+}
+
 /**
  * jpeg_v4_0_3_early_init - set function pointers
  *
@@ -734,32 +740,20 @@ void jpeg_v4_0_3_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq
 		0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE4));
 	amdgpu_ring_write(ring, 0);
 
-	if (ring->adev->jpeg.inst[ring->me].aid_id) {
-		amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_MCM_ADDR_INTERNAL_OFFSET,
-			0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE0));
-		amdgpu_ring_write(ring, 0x4);
-	} else {
-		amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6));
-		amdgpu_ring_write(ring, 0);
-	}
+	amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6));
+	amdgpu_ring_write(ring, 0);
 
 	amdgpu_ring_write(ring,	PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
 	amdgpu_ring_write(ring, 0x3fbc);
 
-	if (ring->adev->jpeg.inst[ring->me].aid_id) {
-		amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_MCM_ADDR_INTERNAL_OFFSET,
-			0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE0));
-		amdgpu_ring_write(ring, 0x0);
-	} else {
-		amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6));
-		amdgpu_ring_write(ring, 0);
-	}
-
 	amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
 		0, 0, PACKETJ_TYPE0));
 	amdgpu_ring_write(ring, 0x1);
 
+	amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6));
+	amdgpu_ring_write(ring, 0);
+
 	amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE7));
 	amdgpu_ring_write(ring, 0);
 }
@@ -834,8 +828,8 @@ void jpeg_v4_0_3_dec_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg,
 {
 	uint32_t reg_offset;
 
-	/* For VF, only local offsets should be used */
-	if (amdgpu_sriov_vf(ring->adev))
+	/* Use normalized offsets if required */
+	if (jpeg_v4_0_3_normalizn_reqd(ring->adev))
 		reg = NORMALIZE_JPEG_REG_OFFSET(reg);
 
 	reg_offset = (reg << 2);
@@ -881,8 +875,8 @@ void jpeg_v4_0_3_dec_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint
 {
 	uint32_t reg_offset;
 
-	/* For VF, only local offsets should be used */
-	if (amdgpu_sriov_vf(ring->adev))
+	/* Use normalized offsets if required */
+	if (jpeg_v4_0_3_normalizn_reqd(ring->adev))
 		reg = NORMALIZE_JPEG_REG_OFFSET(reg);
 
 	reg_offset = (reg << 2);
@@ -1089,7 +1083,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v4_0_3_dec_ring_get_rptr,
 	.get_wptr = jpeg_v4_0_3_dec_ring_get_wptr,
 	.set_wptr = jpeg_v4_0_3_dec_ring_set_wptr,
-	.parse_cs = jpeg_v4_0_3_dec_ring_parse_cs,
+	.parse_cs = jpeg_v2_dec_ring_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
@@ -1254,56 +1248,3 @@ static void jpeg_v4_0_3_set_ras_funcs(struct amdgpu_device *adev)
 {
 	adev->jpeg.ras = &jpeg_v4_0_3_ras;
 }
-
-/**
- * jpeg_v4_0_3_dec_ring_parse_cs - command submission parser
- *
- * @parser: Command submission parser context
- * @job: the job to parse
- * @ib: the IB to parse
- *
- * Parse the command stream, return -EINVAL for invalid packet,
- * 0 otherwise
- */
-int jpeg_v4_0_3_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
-			     struct amdgpu_job *job,
-			     struct amdgpu_ib *ib)
-{
-	uint32_t i, reg, res, cond, type;
-	struct amdgpu_device *adev = parser->adev;
-
-	for (i = 0; i < ib->length_dw ; i += 2) {
-		reg  = CP_PACKETJ_GET_REG(ib->ptr[i]);
-		res  = CP_PACKETJ_GET_RES(ib->ptr[i]);
-		cond = CP_PACKETJ_GET_COND(ib->ptr[i]);
-		type = CP_PACKETJ_GET_TYPE(ib->ptr[i]);
-
-		if (res) /* only support 0 at the moment */
-			return -EINVAL;
-
-		switch (type) {
-		case PACKETJ_TYPE0:
-			if (cond != PACKETJ_CONDITION_CHECK0 || reg < JPEG_REG_RANGE_START || reg > JPEG_REG_RANGE_END) {
-				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
-				return -EINVAL;
-			}
-			break;
-		case PACKETJ_TYPE3:
-			if (cond != PACKETJ_CONDITION_CHECK3 || reg < JPEG_REG_RANGE_START || reg > JPEG_REG_RANGE_END) {
-				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
-				return -EINVAL;
-			}
-			break;
-		case PACKETJ_TYPE6:
-			if (ib->ptr[i] == CP_PACKETJ_NOP)
-				continue;
-			dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
-			return -EINVAL;
-		default:
-			dev_err(adev->dev, "Unknown packet type %d !\n", type);
-			return -EINVAL;
-		}
-	}
-
-	return 0;
-}
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
index 71c54b294e15..747a3e5f6856 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
@@ -46,9 +46,6 @@
 
 #define JRBC_DEC_EXTERNAL_REG_WRITE_ADDR				0x18000
 
-#define JPEG_REG_RANGE_START						0x4000
-#define JPEG_REG_RANGE_END						0x41c2
-
 extern const struct amdgpu_ip_block_version jpeg_v4_0_3_ip_block;
 
 void jpeg_v4_0_3_dec_ring_emit_ib(struct amdgpu_ring *ring,
@@ -65,7 +62,5 @@ void jpeg_v4_0_3_dec_ring_insert_end(struct amdgpu_ring *ring);
 void jpeg_v4_0_3_dec_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t val);
 void jpeg_v4_0_3_dec_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg,
 					uint32_t val, uint32_t mask);
-int jpeg_v4_0_3_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
-				  struct amdgpu_job *job,
-				  struct amdgpu_ib *ib);
+
 #endif /* __JPEG_V4_0_3_H__ */
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c
index f96ac6bce526..44eeed445ea9 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c
@@ -768,6 +768,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_5_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v4_0_5_dec_ring_get_rptr,
 	.get_wptr = jpeg_v4_0_5_dec_ring_get_wptr,
 	.set_wptr = jpeg_v4_0_5_dec_ring_set_wptr,
+	.parse_cs = jpeg_v2_dec_ring_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c
index f4daff90c770..d662aa841f97 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c
@@ -26,6 +26,7 @@
 #include "amdgpu_pm.h"
 #include "soc15.h"
 #include "soc15d.h"
+#include "jpeg_v2_0.h"
 #include "jpeg_v4_0_3.h"
 
 #include "vcn/vcn_5_0_0_offset.h"
@@ -646,7 +647,7 @@ static const struct amdgpu_ring_funcs jpeg_v5_0_0_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v5_0_0_dec_ring_get_rptr,
 	.get_wptr = jpeg_v5_0_0_dec_ring_get_wptr,
 	.set_wptr = jpeg_v5_0_0_dec_ring_set_wptr,
-	.parse_cs = jpeg_v4_0_3_dec_ring_parse_cs,
+	.parse_cs = jpeg_v2_dec_ring_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 0f055d1b1da6..ee91ff9e52a2 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -415,7 +415,7 @@ static int mes_v11_0_reset_queue_mmio(struct amdgpu_mes *mes, uint32_t queue_typ
 		/* wait till dequeue take effects */
 		for (i = 0; i < adev->usec_timeout; i++) {
 			if (!(RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1))
-			break;
+				break;
 			udelay(1);
 		}
 		if (i >= adev->usec_timeout) {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 00350eccd571..9044bdb38cf4 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -36,7 +36,6 @@
 #include <linux/mman.h>
 #include <linux/ptrace.h>
 #include <linux/dma-buf.h>
-#include <linux/fdtable.h>
 #include <linux/processor.h>
 #include "kfd_priv.h"
 #include "kfd_device_queue_manager.h"
@@ -1835,7 +1834,8 @@ static uint32_t get_process_num_bos(struct kfd_process *p)
 }
 
 static int criu_get_prime_handle(struct kgd_mem *mem,
-				 int flags, u32 *shared_fd)
+				 int flags, u32 *shared_fd,
+				 struct file **file)
 {
 	struct dma_buf *dmabuf;
 	int ret;
@@ -1846,13 +1846,14 @@ static int criu_get_prime_handle(struct kgd_mem *mem,
 		return ret;
 	}
 
-	ret = dma_buf_fd(dmabuf, flags);
+	ret = get_unused_fd_flags(flags);
 	if (ret < 0) {
 		pr_err("dmabuf create fd failed, ret:%d\n", ret);
 		goto out_free_dmabuf;
 	}
 
 	*shared_fd = ret;
+	*file = dmabuf->file;
 	return 0;
 
 out_free_dmabuf:
@@ -1860,6 +1861,25 @@ out_free_dmabuf:
 	return ret;
 }
 
+static void commit_files(struct file **files,
+			 struct kfd_criu_bo_bucket *bo_buckets,
+			 unsigned int count,
+			 int err)
+{
+	while (count--) {
+		struct file *file = files[count];
+
+		if (!file)
+			continue;
+		if (err) {
+			fput(file);
+			put_unused_fd(bo_buckets[count].dmabuf_fd);
+		} else {
+			fd_install(bo_buckets[count].dmabuf_fd, file);
+		}
+	}
+}
+
 static int criu_checkpoint_bos(struct kfd_process *p,
 			       uint32_t num_bos,
 			       uint8_t __user *user_bos,
@@ -1868,6 +1888,7 @@ static int criu_checkpoint_bos(struct kfd_process *p,
 {
 	struct kfd_criu_bo_bucket *bo_buckets;
 	struct kfd_criu_bo_priv_data *bo_privs;
+	struct file **files = NULL;
 	int ret = 0, pdd_index, bo_index = 0, id;
 	void *mem;
 
@@ -1881,6 +1902,12 @@ static int criu_checkpoint_bos(struct kfd_process *p,
 		goto exit;
 	}
 
+	files = kvzalloc(num_bos * sizeof(struct file *), GFP_KERNEL);
+	if (!files) {
+		ret = -ENOMEM;
+		goto exit;
+	}
+
 	for (pdd_index = 0; pdd_index < p->n_pdds; pdd_index++) {
 		struct kfd_process_device *pdd = p->pdds[pdd_index];
 		struct amdgpu_bo *dumper_bo;
@@ -1923,7 +1950,7 @@ static int criu_checkpoint_bos(struct kfd_process *p,
 				ret = criu_get_prime_handle(kgd_mem,
 						bo_bucket->alloc_flags &
 						KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE ? DRM_RDWR : 0,
-						&bo_bucket->dmabuf_fd);
+						&bo_bucket->dmabuf_fd, &files[bo_index]);
 				if (ret)
 					goto exit;
 			} else {
@@ -1974,12 +2001,8 @@ static int criu_checkpoint_bos(struct kfd_process *p,
 	*priv_offset += num_bos * sizeof(*bo_privs);
 
 exit:
-	while (ret && bo_index--) {
-		if (bo_buckets[bo_index].alloc_flags
-		    & (KFD_IOC_ALLOC_MEM_FLAGS_VRAM | KFD_IOC_ALLOC_MEM_FLAGS_GTT))
-			close_fd(bo_buckets[bo_index].dmabuf_fd);
-	}
-
+	commit_files(files, bo_buckets, bo_index, ret);
+	kvfree(files);
 	kvfree(bo_buckets);
 	kvfree(bo_privs);
 	return ret;
@@ -2331,7 +2354,8 @@ static int criu_restore_memory_of_gpu(struct kfd_process_device *pdd,
 
 static int criu_restore_bo(struct kfd_process *p,
 			   struct kfd_criu_bo_bucket *bo_bucket,
-			   struct kfd_criu_bo_priv_data *bo_priv)
+			   struct kfd_criu_bo_priv_data *bo_priv,
+			   struct file **file)
 {
 	struct kfd_process_device *pdd;
 	struct kgd_mem *kgd_mem;
@@ -2383,7 +2407,7 @@ static int criu_restore_bo(struct kfd_process *p,
 	if (bo_bucket->alloc_flags
 	    & (KFD_IOC_ALLOC_MEM_FLAGS_VRAM | KFD_IOC_ALLOC_MEM_FLAGS_GTT)) {
 		ret = criu_get_prime_handle(kgd_mem, DRM_RDWR,
-					    &bo_bucket->dmabuf_fd);
+					    &bo_bucket->dmabuf_fd, file);
 		if (ret)
 			return ret;
 	} else {
@@ -2400,6 +2424,7 @@ static int criu_restore_bos(struct kfd_process *p,
 {
 	struct kfd_criu_bo_bucket *bo_buckets = NULL;
 	struct kfd_criu_bo_priv_data *bo_privs = NULL;
+	struct file **files = NULL;
 	int ret = 0;
 	uint32_t i = 0;
 
@@ -2413,6 +2438,12 @@ static int criu_restore_bos(struct kfd_process *p,
 	if (!bo_buckets)
 		return -ENOMEM;
 
+	files = kvzalloc(args->num_bos * sizeof(struct file *), GFP_KERNEL);
+	if (!files) {
+		ret = -ENOMEM;
+		goto exit;
+	}
+
 	ret = copy_from_user(bo_buckets, (void __user *)args->bos,
 			     args->num_bos * sizeof(*bo_buckets));
 	if (ret) {
@@ -2438,7 +2469,7 @@ static int criu_restore_bos(struct kfd_process *p,
 
 	/* Create and map new BOs */
 	for (; i < args->num_bos; i++) {
-		ret = criu_restore_bo(p, &bo_buckets[i], &bo_privs[i]);
+		ret = criu_restore_bo(p, &bo_buckets[i], &bo_privs[i], &files[i]);
 		if (ret) {
 			pr_debug("Failed to restore BO[%d] ret%d\n", i, ret);
 			goto exit;
@@ -2453,11 +2484,8 @@ static int criu_restore_bos(struct kfd_process *p,
 		ret = -EFAULT;
 
 exit:
-	while (ret && i--) {
-		if (bo_buckets[i].alloc_flags
-		   & (KFD_IOC_ALLOC_MEM_FLAGS_VRAM | KFD_IOC_ALLOC_MEM_FLAGS_GTT))
-			close_fd(bo_buckets[i].dmabuf_fd);
-	}
+	commit_files(files, bo_buckets, i, ret);
+	kvfree(files);
 	kvfree(bo_buckets);
 	kvfree(bo_privs);
 	return ret;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index cd7b81b7b939..48caecf7e72e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -1434,7 +1434,8 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 		pcache_info[i].flags = (CRAT_CACHE_FLAGS_ENABLED |
 					CRAT_CACHE_FLAGS_DATA_CACHE |
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
-		pcache_info[0].num_cu_shared = adev->gfx.config.gc_num_tcp_per_wpg / 2;
+		pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_tcp_per_wpg / 2;
+		pcache_info[i].cache_line_size = adev->gfx.config.gc_tcp_cache_line_size;
 		i++;
 	}
 	/* Scalar L1 Instruction Cache per SQC */
@@ -1446,6 +1447,7 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_INST_CACHE |
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_sqc_per_wgp * 2;
+		pcache_info[i].cache_line_size = adev->gfx.config.gc_instruction_cache_line_size;
 		i++;
 	}
 	/* Scalar L1 Data Cache per SQC */
@@ -1456,6 +1458,7 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_DATA_CACHE |
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_sqc_per_wgp * 2;
+		pcache_info[i].cache_line_size = adev->gfx.config.gc_scalar_data_cache_line_size;
 		i++;
 	}
 	/* GL1 Data Cache per SA */
@@ -1468,6 +1471,7 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_DATA_CACHE |
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
+		pcache_info[i].cache_line_size = 0;
 		i++;
 	}
 	/* L2 Data Cache per GPU (Total Tex Cache) */
@@ -1478,6 +1482,7 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_DATA_CACHE |
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
+		pcache_info[i].cache_line_size = adev->gfx.config.gc_tcc_cache_line_size;
 		i++;
 	}
 	/* L3 Data Cache per GPU */
@@ -1488,6 +1493,7 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_DATA_CACHE |
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
+		pcache_info[i].cache_line_size = 0;
 		i++;
 	}
 	return i;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 577d121cc6d1..71b465f8d83e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -2407,10 +2407,9 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 		pdd->sdma_past_activity_counter += sdma_val;
 	}
 
-	list_del(&q->list);
-	qpd->queue_count--;
 	if (q->properties.is_active) {
 		decrement_queue_count(dqm, qpd, q);
+		q->properties.is_active = false;
 		if (!dqm->dev->kfd->shared_resources.enable_mes) {
 			retval = execute_queues_cpsch(dqm,
 						      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0,
@@ -2421,6 +2420,8 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 			retval = remove_queue_mes(dqm, q, qpd);
 		}
 	}
+	list_del(&q->list);
+	qpd->queue_count--;
 
 	/*
 	 * Unconditionally decrement this counter, regardless of the queue's
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
index fecdbbab9894..d46a13156ee9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -167,11 +167,23 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 	case SOC15_IH_CLIENTID_SE3SH:
 	case SOC15_IH_CLIENTID_UTCL2:
 		block = AMDGPU_RAS_BLOCK__GFX;
-		if (amdgpu_ip_version(dev->adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) ||
-			amdgpu_ip_version(dev->adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4))
-			reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
-		else
+		if (amdgpu_ip_version(dev->adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3)) {
+			/* driver mode-2 for gfx poison is only supported by
+			 * pmfw 0x00557300 and onwards */
+			if (dev->adev->pm.fw_version < 0x00557300)
+				reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
+			else
+				reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+		} else if (amdgpu_ip_version(dev->adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4)) {
+			/* driver mode-2 for gfx poison is only supported by
+			 * pmfw 0x05550C00 and onwards */
+			if (dev->adev->pm.fw_version < 0x05550C00)
+				reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
+			else
+				reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+		} else {
 			reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+		}
 		break;
 	case SOC15_IH_CLIENTID_VMC:
 	case SOC15_IH_CLIENTID_VMC1:
@@ -184,11 +196,23 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 	case SOC15_IH_CLIENTID_SDMA3:
 	case SOC15_IH_CLIENTID_SDMA4:
 		block = AMDGPU_RAS_BLOCK__SDMA;
-		if (amdgpu_ip_version(dev->adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) ||
-			amdgpu_ip_version(dev->adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4))
-			reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
-		else
+		if (amdgpu_ip_version(dev->adev, SDMA0_HWIP, 0) == IP_VERSION(4, 4, 2)) {
+			/* driver mode-2 for gfx poison is only supported by
+			 * pmfw 0x00557300 and onwards */
+			if (dev->adev->pm.fw_version < 0x00557300)
+				reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
+			else
+				reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+		} else if (amdgpu_ip_version(dev->adev, SDMA0_HWIP, 0) == IP_VERSION(4, 4, 5)) {
+			/* driver mode-2 for gfx poison is only supported by
+			 * pmfw 0x05550C00 and onwards */
+			if (dev->adev->pm.fw_version < 0x05550C00)
+				reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
+			else
+				reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+		} else {
 			reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+		}
 		break;
 	default:
 		dev_warn(dev->adev->dev,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 9ae9abc6eb43..d6530febabad 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -868,6 +868,12 @@ struct svm_range_list {
 	struct task_struct		*faulting_task;
 	/* check point ts decides if page fault recovery need be dropped */
 	uint64_t			checkpoint_ts[MAX_GPU_INSTANCE];
+
+	/* Default granularity to use in buffer migration
+	 * and restoration of backing memory while handling
+	 * recoverable page faults
+	 */
+	uint8_t default_granularity;
 };
 
 /* Process data */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 20ea745729ee..b439d4d0bd84 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -1046,6 +1046,7 @@ exit:
 		pr_debug("Queue id %d was restored successfully\n", queue_id);
 
 	kfree(q_data);
+	kfree(q_extra_data);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
index ea6a8e43bd5b..de8b9abf7afc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
@@ -235,17 +235,16 @@ void kfd_smi_event_update_gpu_reset(struct kfd_node *dev, bool post_reset,
 		amdgpu_reset_get_desc(reset_context, reset_cause,
 				      sizeof(reset_cause));
 
-	kfd_smi_event_add(0, dev, event, "%x %s\n",
-			  dev->reset_seq_num,
-			  reset_cause);
+	kfd_smi_event_add(0, dev, event, KFD_EVENT_FMT_UPDATE_GPU_RESET(
+			  dev->reset_seq_num, reset_cause));
 }
 
 void kfd_smi_event_update_thermal_throttling(struct kfd_node *dev,
 					     uint64_t throttle_bitmask)
 {
-	kfd_smi_event_add(0, dev, KFD_SMI_EVENT_THERMAL_THROTTLE, "%llx:%llx\n",
+	kfd_smi_event_add(0, dev, KFD_SMI_EVENT_THERMAL_THROTTLE, KFD_EVENT_FMT_THERMAL_THROTTLING(
 			  throttle_bitmask,
-			  amdgpu_dpm_get_thermal_throttling_counter(dev->adev));
+			  amdgpu_dpm_get_thermal_throttling_counter(dev->adev)));
 }
 
 void kfd_smi_event_update_vmfault(struct kfd_node *dev, uint16_t pasid)
@@ -256,8 +255,8 @@ void kfd_smi_event_update_vmfault(struct kfd_node *dev, uint16_t pasid)
 	if (task_info) {
 		/* Report VM faults from user applications, not retry from kernel */
 		if (task_info->pid)
-			kfd_smi_event_add(0, dev, KFD_SMI_EVENT_VMFAULT, "%x:%s\n",
-					 task_info->pid, task_info->task_name);
+			kfd_smi_event_add(0, dev, KFD_SMI_EVENT_VMFAULT, KFD_EVENT_FMT_VMFAULT(
+					  task_info->pid, task_info->task_name));
 		amdgpu_vm_put_task_info(task_info);
 	}
 }
@@ -267,16 +266,16 @@ void kfd_smi_event_page_fault_start(struct kfd_node *node, pid_t pid,
 				    ktime_t ts)
 {
 	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_PAGE_FAULT_START,
-			  "%lld -%d @%lx(%x) %c\n", ktime_to_ns(ts), pid,
-			  address, node->id, write_fault ? 'W' : 'R');
+			  KFD_EVENT_FMT_PAGEFAULT_START(ktime_to_ns(ts), pid,
+			  address, node->id, write_fault ? 'W' : 'R'));
 }
 
 void kfd_smi_event_page_fault_end(struct kfd_node *node, pid_t pid,
 				  unsigned long address, bool migration)
 {
 	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_PAGE_FAULT_END,
-			  "%lld -%d @%lx(%x) %c\n", ktime_get_boottime_ns(),
-			  pid, address, node->id, migration ? 'M' : 'U');
+			  KFD_EVENT_FMT_PAGEFAULT_END(ktime_get_boottime_ns(),
+			  pid, address, node->id, migration ? 'M' : 'U'));
 }
 
 void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid,
@@ -286,9 +285,9 @@ void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid,
 				   uint32_t trigger)
 {
 	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_MIGRATE_START,
-			  "%lld -%d @%lx(%lx) %x->%x %x:%x %d\n",
+			  KFD_EVENT_FMT_MIGRATE_START(
 			  ktime_get_boottime_ns(), pid, start, end - start,
-			  from, to, prefetch_loc, preferred_loc, trigger);
+			  from, to, prefetch_loc, preferred_loc, trigger));
 }
 
 void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid,
@@ -296,24 +295,24 @@ void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid,
 				 uint32_t from, uint32_t to, uint32_t trigger)
 {
 	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_MIGRATE_END,
-			  "%lld -%d @%lx(%lx) %x->%x %d\n",
+			  KFD_EVENT_FMT_MIGRATE_END(
 			  ktime_get_boottime_ns(), pid, start, end - start,
-			  from, to, trigger);
+			  from, to, trigger));
 }
 
 void kfd_smi_event_queue_eviction(struct kfd_node *node, pid_t pid,
 				  uint32_t trigger)
 {
 	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_QUEUE_EVICTION,
-			  "%lld -%d %x %d\n", ktime_get_boottime_ns(), pid,
-			  node->id, trigger);
+			  KFD_EVENT_FMT_QUEUE_EVICTION(ktime_get_boottime_ns(), pid,
+			  node->id, trigger));
 }
 
 void kfd_smi_event_queue_restore(struct kfd_node *node, pid_t pid)
 {
 	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_QUEUE_RESTORE,
-			  "%lld -%d %x\n", ktime_get_boottime_ns(), pid,
-			  node->id);
+			  KFD_EVENT_FMT_QUEUE_RESTORE(ktime_get_boottime_ns(), pid,
+			  node->id, 0));
 }
 
 void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm)
@@ -330,8 +329,8 @@ void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm)
 
 		kfd_smi_event_add(p->lead_thread->pid, pdd->dev,
 				  KFD_SMI_EVENT_QUEUE_RESTORE,
-				  "%lld -%d %x %c\n", ktime_get_boottime_ns(),
-				  p->lead_thread->pid, pdd->dev->id, 'R');
+				  KFD_EVENT_FMT_QUEUE_RESTORE(ktime_get_boottime_ns(),
+				  p->lead_thread->pid, pdd->dev->id, 'R'));
 	}
 	kfd_unref_process(p);
 }
@@ -341,8 +340,8 @@ void kfd_smi_event_unmap_from_gpu(struct kfd_node *node, pid_t pid,
 				  uint32_t trigger)
 {
 	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_UNMAP_FROM_GPU,
-			  "%lld -%d @%lx(%lx) %x %d\n", ktime_get_boottime_ns(),
-			  pid, address, last - address + 1, node->id, trigger);
+			  KFD_EVENT_FMT_UNMAP_FROM_GPU(ktime_get_boottime_ns(),
+			  pid, address, last - address + 1, node->id, trigger));
 }
 
 int kfd_smi_event_open(struct kfd_node *dev, uint32_t *fd)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 40c94c4cdd96..04e746923697 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -309,12 +309,13 @@ static void svm_range_free(struct svm_range *prange, bool do_unmap)
 }
 
 static void
-svm_range_set_default_attributes(int32_t *location, int32_t *prefetch_loc,
-				 uint8_t *granularity, uint32_t *flags)
+svm_range_set_default_attributes(struct svm_range_list *svms, int32_t *location,
+				 int32_t *prefetch_loc, uint8_t *granularity,
+				 uint32_t *flags)
 {
 	*location = KFD_IOCTL_SVM_LOCATION_UNDEFINED;
 	*prefetch_loc = KFD_IOCTL_SVM_LOCATION_UNDEFINED;
-	*granularity = 9;
+	*granularity = svms->default_granularity;
 	*flags =
 		KFD_IOCTL_SVM_FLAG_HOST_ACCESS | KFD_IOCTL_SVM_FLAG_COHERENT;
 }
@@ -358,7 +359,7 @@ svm_range *svm_range_new(struct svm_range_list *svms, uint64_t start,
 		bitmap_copy(prange->bitmap_access, svms->bitmap_supported,
 			    MAX_GPU_INSTANCE);
 
-	svm_range_set_default_attributes(&prange->preferred_loc,
+	svm_range_set_default_attributes(svms, &prange->preferred_loc,
 					 &prange->prefetch_loc,
 					 &prange->granularity, &prange->flags);
 
@@ -2703,9 +2704,10 @@ svm_range_get_range_boundaries(struct kfd_process *p, int64_t addr,
 	*is_heap_stack = vma_is_initial_heap(vma) || vma_is_initial_stack(vma);
 
 	start_limit = max(vma->vm_start >> PAGE_SHIFT,
-		      (unsigned long)ALIGN_DOWN(addr, 2UL << 8));
+		      (unsigned long)ALIGN_DOWN(addr, 1UL << p->svms.default_granularity));
 	end_limit = min(vma->vm_end >> PAGE_SHIFT,
-		    (unsigned long)ALIGN(addr + 1, 2UL << 8));
+		    (unsigned long)ALIGN(addr + 1, 1UL << p->svms.default_granularity));
+
 	/* First range that starts after the fault address */
 	node = interval_tree_iter_first(&p->svms.objects, addr + 1, ULONG_MAX);
 	if (node) {
@@ -3249,6 +3251,12 @@ int svm_range_list_init(struct kfd_process *p)
 		if (KFD_IS_SVM_API_SUPPORTED(p->pdds[i]->dev->adev))
 			bitmap_set(svms->bitmap_supported, i, 1);
 
+	 /* Value of default granularity cannot exceed 0x1B, the
+	  * number of pages supported by a 4-level paging table
+	  */
+	svms->default_granularity = min_t(u8, amdgpu_svm_default_granularity, 0x1B);
+	pr_debug("Default SVM Granularity to use: %d\n", svms->default_granularity);
+
 	return 0;
 }
 
@@ -3776,7 +3784,7 @@ svm_range_get_attr(struct kfd_process *p, struct mm_struct *mm,
 	node = interval_tree_iter_first(&svms->objects, start, last);
 	if (!node) {
 		pr_debug("range attrs not found return default values\n");
-		svm_range_set_default_attributes(&location, &prefetch_loc,
+		svm_range_set_default_attributes(svms, &location, &prefetch_loc,
 						 &granularity, &flags_and);
 		flags_or = flags_and;
 		if (p->xnack_enabled)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index a8d0d1b71723..0cff66735cfe 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -10610,7 +10610,7 @@ static bool should_reset_plane(struct drm_atomic_state *state,
 	 * TODO: We can likely skip bandwidth validation if the only thing that
 	 * changed about the plane was it'z z-ordering.
 	 */
-	if (new_crtc_state->zpos_changed)
+	if (old_plane_state->normalized_zpos != new_plane_state->normalized_zpos)
 		return true;
 
 	if (drm_atomic_crtc_needs_modeset(new_crtc_state))
@@ -11458,6 +11458,17 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 			drm_dbg(dev, "Failed to determine cursor mode\n");
 			goto fail;
 		}
+
+		/*
+		 * If overlay cursor is needed, DC cannot go through the
+		 * native cursor update path. All enabled planes on the CRTC
+		 * need to be added for DC to not disable a plane by mistake
+		 */
+		if (dm_new_crtc_state->cursor_mode == DM_CURSOR_OVERLAY_MODE) {
+			ret = drm_atomic_add_affected_planes(state, crtc);
+			if (ret)
+				goto fail;
+		}
 	}
 
 	/* Remove exiting planes if they are modified */
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 6b5eed37532b..c0c61c03984c 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -1325,7 +1325,7 @@ static bool is_dsc_need_re_compute(
 		if (new_crtc_state->enable && new_crtc_state->active) {
 			if (new_crtc_state->mode_changed || new_crtc_state->active_changed ||
 					new_crtc_state->connectors_changed) {
-				DRM_DEBUG_DRIVER("%s:%d MST_DSC dsc recompte required."
+				DRM_DEBUG_DRIVER("%s:%d MST_DSC dsc recompute required."
 						 "stream 0x%p in new dc_state\n",
 						 __func__, __LINE__, stream);
 				is_dsc_need_re_compute = true;
diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
index ee02b78e290f..838d72eaa87f 100644
--- a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
@@ -1748,10 +1748,6 @@ void dccg35_init(struct dccg *dccg)
 			dccg35_set_dpstreamclk_root_clock_gating(dccg, otg_inst, false);
 		}
 
-	if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpp)
-		for (otg_inst = 0; otg_inst < 4; otg_inst++)
-			dccg35_set_dppclk_root_clock_gating(dccg, otg_inst, 0);
-
 /*
 	dccg35_enable_global_fgcg_rep(
 		dccg, dccg->ctx->dc->debug.enable_fine_grain_clock_gating.bits
@@ -1932,47 +1928,32 @@ static void dccg35_enable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst,
 }
 
 /*get other front end connected to this backend*/
-static uint8_t dccg35_get_other_enabled_symclk_fe(struct dccg *dccg, uint32_t stream_enc_inst, uint32_t link_enc_inst)
+static uint8_t dccg35_get_number_enabled_symclk_fe_connected_to_be(struct dccg *dccg, uint32_t link_enc_inst)
 {
 	uint8_t num_enabled_symclk_fe = 0;
-	uint32_t be_clk_en = 0, fe_clk_en[5] = {0}, be_clk_sel[5] = {0};
+	uint32_t fe_clk_en[5] = {0}, be_clk_sel[5] = {0};
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
-	switch (link_enc_inst) {
-	case 0:
-		REG_GET_3(SYMCLKA_CLOCK_ENABLE, SYMCLKA_CLOCK_ENABLE, &be_clk_en,
-				SYMCLKA_FE_EN, &fe_clk_en[0],
-				SYMCLKA_FE_SRC_SEL, &be_clk_sel[0]);
-				break;
-	case 1:
-		REG_GET_3(SYMCLKB_CLOCK_ENABLE, SYMCLKB_CLOCK_ENABLE, &be_clk_en,
-				SYMCLKB_FE_EN, &fe_clk_en[1],
-				SYMCLKB_FE_SRC_SEL, &be_clk_sel[1]);
-				break;
-	case 2:
-			REG_GET_3(SYMCLKC_CLOCK_ENABLE, SYMCLKC_CLOCK_ENABLE, &be_clk_en,
-				SYMCLKC_FE_EN, &fe_clk_en[2],
-				SYMCLKC_FE_SRC_SEL, &be_clk_sel[2]);
-				break;
-	case 3:
-			REG_GET_3(SYMCLKD_CLOCK_ENABLE, SYMCLKD_CLOCK_ENABLE, &be_clk_en,
-				SYMCLKD_FE_EN, &fe_clk_en[3],
-				SYMCLKD_FE_SRC_SEL, &be_clk_sel[3]);
-				break;
-	case 4:
-			REG_GET_3(SYMCLKE_CLOCK_ENABLE, SYMCLKE_CLOCK_ENABLE, &be_clk_en,
-				SYMCLKE_FE_EN, &fe_clk_en[4],
-				SYMCLKE_FE_SRC_SEL, &be_clk_sel[4]);
-				break;
-	}
-	if (be_clk_en) {
-	/* for DPMST, this backend could be used by multiple front end.
-	only disable the backend if this stream_enc_ins is the last active stream enc connected to this back_end*/
-		uint8_t i;
-		for (i = 0; i != link_enc_inst && i < ARRAY_SIZE(fe_clk_en); i++) {
-			if (fe_clk_en[i] && be_clk_sel[i] == link_enc_inst)
-				num_enabled_symclk_fe++;
-		}
+	REG_GET_2(SYMCLKA_CLOCK_ENABLE, SYMCLKA_FE_EN, &fe_clk_en[0],
+		SYMCLKA_FE_SRC_SEL, &be_clk_sel[0]);
+
+	REG_GET_2(SYMCLKB_CLOCK_ENABLE, SYMCLKB_FE_EN, &fe_clk_en[1],
+		SYMCLKB_FE_SRC_SEL, &be_clk_sel[1]);
+
+	REG_GET_2(SYMCLKC_CLOCK_ENABLE, SYMCLKC_FE_EN, &fe_clk_en[2],
+		SYMCLKC_FE_SRC_SEL, &be_clk_sel[2]);
+
+	REG_GET_2(SYMCLKD_CLOCK_ENABLE,	SYMCLKD_FE_EN, &fe_clk_en[3],
+		SYMCLKD_FE_SRC_SEL, &be_clk_sel[3]);
+
+	REG_GET_2(SYMCLKE_CLOCK_ENABLE,	SYMCLKE_FE_EN, &fe_clk_en[4],
+		SYMCLKE_FE_SRC_SEL, &be_clk_sel[4]);
+
+	uint8_t i;
+
+	for (i = 0; i < ARRAY_SIZE(fe_clk_en); i++) {
+		if (fe_clk_en[i] && be_clk_sel[i] == link_enc_inst)
+			num_enabled_symclk_fe++;
 	}
 	return num_enabled_symclk_fe;
 }
@@ -2020,9 +2001,9 @@ static void dccg35_disable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst
 		break;
 	}
 
-	/*check other enabled symclk fe */
-	num_enabled_symclk_fe = dccg35_get_other_enabled_symclk_fe(dccg, stream_enc_inst, link_enc_inst);
-	/*only turn off backend clk if other front end attachecd to this backend are all off,
+	/*check other enabled symclk fe connected to this be */
+	num_enabled_symclk_fe = dccg35_get_number_enabled_symclk_fe_connected_to_be(dccg, link_enc_inst);
+	/*only turn off backend clk if other front end attached to this backend are all off,
 	 for mst, only turn off the backend if this is the last front end*/
 	if (num_enabled_symclk_fe == 0) {
 		switch (link_enc_inst) {
@@ -2351,6 +2332,14 @@ static void dccg35_disable_symclk_se_cb(
 	/* DMU PHY sequence switches SYMCLK_BE (link_enc_inst) to ref clock once PHY is turned off */
 }
 
+void dccg35_root_gate_disable_control(struct dccg *dccg, uint32_t pipe_idx, uint32_t disable_clock_gating)
+{
+
+	if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpp) {
+		dccg35_set_dppclk_root_clock_gating(dccg, pipe_idx, disable_clock_gating);
+	}
+}
+
 static const struct dccg_funcs dccg35_funcs_new = {
 	.update_dpp_dto = dccg35_update_dpp_dto_cb,
 	.dpp_root_clock_control = dccg35_dpp_root_clock_control_cb,
@@ -2411,7 +2400,7 @@ static const struct dccg_funcs dccg35_funcs = {
 	.enable_symclk_se = dccg35_enable_symclk_se,
 	.disable_symclk_se = dccg35_disable_symclk_se,
 	.set_dtbclk_p_src = dccg35_set_dtbclk_p_src,
-
+	.dccg_root_gate_disable_control = dccg35_root_gate_disable_control,
 };
 
 struct dccg *dccg35_create(
diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.h b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.h
index 1586a45ca3bd..51f98c5c51c4 100644
--- a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.h
@@ -241,6 +241,7 @@ struct dccg *dccg35_create(
 void dccg35_init(struct dccg *dccg);
 
 void dccg35_enable_global_fgcg_rep(struct dccg *dccg, bool value);
+void dccg35_root_gate_disable_control(struct dccg *dccg, uint32_t pipe_idx, uint32_t disable_clock_gating);
 
 
 #endif //__DCN35_DCCG_H__
diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c
index 01f98139292e..5105fd580017 100644
--- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c
+++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c
@@ -951,6 +951,8 @@ static void dpp401_dscl_set_isharp_filter(
  *
  * @dpp_base: High level DPP struct
  * @scl_data: scalaer_data info
+ * @program_isharp_1dlut: flag to program isharp 1D LUT
+ * @bs_coeffs_updated: Blur and Scale Coefficients update flag
  *
  * This is the primary function to program isharp
  *
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
index 01dffed4d30b..a6a1db5ba8ba 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
@@ -3212,15 +3212,19 @@ void dcn10_set_drr(struct pipe_ctx **pipe_ctx,
 	 * as well.
 	 */
 	for (i = 0; i < num_pipes; i++) {
-		if ((pipe_ctx[i]->stream_res.tg != NULL) && pipe_ctx[i]->stream_res.tg->funcs) {
-			if (pipe_ctx[i]->stream_res.tg->funcs->set_drr)
-				pipe_ctx[i]->stream_res.tg->funcs->set_drr(
-					pipe_ctx[i]->stream_res.tg, &params);
+		/* dc_state_destruct() might null the stream resources, so fetch tg
+		 * here first to avoid a race condition. The lifetime of the pointee
+		 * itself (the timing_generator object) is not a problem here.
+		 */
+		struct timing_generator *tg = pipe_ctx[i]->stream_res.tg;
+
+		if ((tg != NULL) && tg->funcs) {
+			if (tg->funcs->set_drr)
+				tg->funcs->set_drr(tg, &params);
 			if (adjust.v_total_max != 0 && adjust.v_total_min != 0)
-				if (pipe_ctx[i]->stream_res.tg->funcs->set_static_screen_control)
-					pipe_ctx[i]->stream_res.tg->funcs->set_static_screen_control(
-						pipe_ctx[i]->stream_res.tg,
-						event_triggers, num_frames);
+				if (tg->funcs->set_static_screen_control)
+					tg->funcs->set_static_screen_control(
+						tg, event_triggers, num_frames);
 		}
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
index a4c6decee0f8..479fd3e89e5a 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
@@ -240,6 +240,10 @@ void dcn35_init_hw(struct dc *dc)
 			dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub,
 					!dc->res_pool->hubbub->ctx->dc->debug.disable_stutter);
 	}
+	if (res_pool->dccg->funcs->dccg_root_gate_disable_control) {
+		for (i = 0; i < res_pool->pipe_count; i++)
+			res_pool->dccg->funcs->dccg_root_gate_disable_control(res_pool->dccg, i, 0);
+	}
 
 	for (i = 0; i < res_pool->audio_count; i++) {
 		struct audio *audio = res_pool->audios[i];
@@ -1414,7 +1418,13 @@ void dcn35_set_drr(struct pipe_ctx **pipe_ctx,
 	params.vertical_total_mid_frame_num = adjust.v_total_mid_frame_num;
 
 	for (i = 0; i < num_pipes; i++) {
-		if ((pipe_ctx[i]->stream_res.tg != NULL) && pipe_ctx[i]->stream_res.tg->funcs) {
+		/* dc_state_destruct() might null the stream resources, so fetch tg
+		 * here first to avoid a race condition. The lifetime of the pointee
+		 * itself (the timing_generator object) is not a problem here.
+		 */
+		struct timing_generator *tg = pipe_ctx[i]->stream_res.tg;
+
+		if ((tg != NULL) && tg->funcs) {
 			if (pipe_ctx[i]->stream && pipe_ctx[i]->stream->ctx->dc->debug.static_screen_wait_frames) {
 				struct dc_crtc_timing *timing = &pipe_ctx[i]->stream->timing;
 				struct dc *dc = pipe_ctx[i]->stream->ctx->dc;
@@ -1426,14 +1436,12 @@ void dcn35_set_drr(struct pipe_ctx **pipe_ctx,
 					num_frames = 2 * (frame_rate % 60);
 				}
 			}
-			if (pipe_ctx[i]->stream_res.tg->funcs->set_drr)
-				pipe_ctx[i]->stream_res.tg->funcs->set_drr(
-					pipe_ctx[i]->stream_res.tg, &params);
+			if (tg->funcs->set_drr)
+				tg->funcs->set_drr(tg, &params);
 			if (adjust.v_total_max != 0 && adjust.v_total_min != 0)
-				if (pipe_ctx[i]->stream_res.tg->funcs->set_static_screen_control)
-					pipe_ctx[i]->stream_res.tg->funcs->set_static_screen_control(
-						pipe_ctx[i]->stream_res.tg,
-						event_triggers, num_frames);
+				if (tg->funcs->set_static_screen_control)
+					tg->funcs->set_static_screen_control(
+						tg, event_triggers, num_frames);
 		}
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
index d619eb229a62..e94e9ba60f55 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
@@ -213,6 +213,7 @@ struct dccg_funcs {
 			uint32_t otg_inst);
 	void (*set_dto_dscclk)(struct dccg *dccg, uint32_t dsc_inst);
 	void (*set_ref_dscclk)(struct dccg *dccg, uint32_t dsc_inst);
+	void (*dccg_root_gate_disable_control)(struct dccg *dccg, uint32_t pipe_idx, uint32_t disable_clock_gating);
 };
 
 #endif //__DAL_DCCG_H__
diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h
index 09cbc3afd6d8..b0fc22383e28 100644
--- a/drivers/gpu/drm/amd/include/atomfirmware.h
+++ b/drivers/gpu/drm/amd/include/atomfirmware.h
@@ -1038,7 +1038,7 @@ struct display_object_info_table_v1_4
   uint16_t  supporteddevices;
   uint8_t   number_of_path;
   uint8_t   reserved;
-  struct    atom_display_object_path_v2 display_path[8];   //the real number of this included in the structure is calculated by using the (whole structure size - the header size- number_of_path)/size of atom_display_object_path
+  struct    atom_display_object_path_v2 display_path[];   //the real number of this included in the structure is calculated by using the (whole structure size - the header size- number_of_path)/size of atom_display_object_path
 };
 
 struct display_object_info_table_v1_5 {
@@ -1048,7 +1048,7 @@ struct display_object_info_table_v1_5 {
 	uint8_t reserved;
 	// the real number of this included in the structure is calculated by using the
 	// (whole structure size - the header size- number_of_path)/size of atom_display_object_path
-	struct atom_display_object_path_v3 display_path[8];
+	struct atom_display_object_path_v3 display_path[];
 };
 
 /* 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
index ac0dd6b97f8d..e71a721c12b9 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
@@ -439,7 +439,16 @@ enum smu_clk_type {
        __SMU_DUMMY_MAP(BACO_CG),			\
        __SMU_DUMMY_MAP(SOC_CG),    \
        __SMU_DUMMY_MAP(LOW_POWER_DCNCLKS),       \
-       __SMU_DUMMY_MAP(WHISPER_MODE),
+       __SMU_DUMMY_MAP(WHISPER_MODE),			\
+       __SMU_DUMMY_MAP(EDC_PWRBRK),				\
+       __SMU_DUMMY_MAP(SOC_EDC_XVMIN),				\
+       __SMU_DUMMY_MAP(GFX_PSM_DIDT),				\
+       __SMU_DUMMY_MAP(APT_ALL_ENABLE),				\
+       __SMU_DUMMY_MAP(APT_SQ_THROTTLE),				\
+       __SMU_DUMMY_MAP(APT_PF_DCS),				\
+       __SMU_DUMMY_MAP(GFX_EDC_XVMIN),				\
+       __SMU_DUMMY_MAP(GFX_DIDT_XVMIN),				\
+       __SMU_DUMMY_MAP(FAN_ABNORMAL),
 
 #undef __SMU_DUMMY_MAP
 #define __SMU_DUMMY_MAP(feature)	SMU_FEATURE_##feature##_BIT
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
index a31fae5feedf..43820d7d2c54 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
@@ -199,6 +199,15 @@ static struct cmn2asic_mapping smu_v14_0_2_feature_mask_map[SMU_FEATURE_COUNT] =
 	FEA_MAP(MEM_TEMP_READ),
 	FEA_MAP(ATHUB_MMHUB_PG),
 	FEA_MAP(SOC_PCC),
+	FEA_MAP(EDC_PWRBRK),
+	FEA_MAP(SOC_EDC_XVMIN),
+	FEA_MAP(GFX_PSM_DIDT),
+	FEA_MAP(APT_ALL_ENABLE),
+	FEA_MAP(APT_SQ_THROTTLE),
+	FEA_MAP(APT_PF_DCS),
+	FEA_MAP(GFX_EDC_XVMIN),
+	FEA_MAP(GFX_DIDT_XVMIN),
+	FEA_MAP(FAN_ABNORMAL),
 	[SMU_FEATURE_DPM_VCLK_BIT] = {1, FEATURE_MM_DPM_BIT},
 	[SMU_FEATURE_DPM_DCLK_BIT] = {1, FEATURE_MM_DPM_BIT},
 	[SMU_FEATURE_PPT_BIT] = {1, FEATURE_THROTTLERS_BIT},
@@ -687,6 +696,9 @@ static int smu_v14_0_2_set_default_dpm_table(struct smu_context *smu)
 		pcie_table->clk_freq[pcie_table->num_of_link_levels] =
 					skutable->LclkFreq[link_level];
 		pcie_table->num_of_link_levels++;
+
+		if (link_level == 0)
+			link_level++;
 	}
 
 	/* dcefclk dpm table setup */
diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index 03bd3c7bd0dc..0e3f8adf162f 100644
--- a/drivers/gpu/drm/drm_prime.c
+++ b/drivers/gpu/drm/drm_prime.c
@@ -410,22 +410,30 @@ static struct dma_buf *export_and_register_object(struct drm_device *dev,
 }
 
 /**
- * drm_gem_prime_handle_to_fd - PRIME export function for GEM drivers
+ * drm_gem_prime_handle_to_dmabuf - PRIME export function for GEM drivers
  * @dev: dev to export the buffer from
  * @file_priv: drm file-private structure
  * @handle: buffer handle to export
  * @flags: flags like DRM_CLOEXEC
- * @prime_fd: pointer to storage for the fd id of the create dma-buf
  *
  * This is the PRIME export function which must be used mandatorily by GEM
  * drivers to ensure correct lifetime management of the underlying GEM object.
  * The actual exporting from GEM object to a dma-buf is done through the
  * &drm_gem_object_funcs.export callback.
+ *
+ * Unlike drm_gem_prime_handle_to_fd(), it returns the struct dma_buf it
+ * has created, without attaching it to any file descriptors.  The difference
+ * between those two is similar to that between anon_inode_getfile() and
+ * anon_inode_getfd(); insertion into descriptor table is something you
+ * can not revert if any cleanup is needed, so the descriptor-returning
+ * variants should only be used when you are past the last failure exit
+ * and the only thing left is passing the new file descriptor to userland.
+ * When all you need is the object itself or when you need to do something
+ * else that might fail, use that one instead.
  */
-int drm_gem_prime_handle_to_fd(struct drm_device *dev,
+struct dma_buf *drm_gem_prime_handle_to_dmabuf(struct drm_device *dev,
 			       struct drm_file *file_priv, uint32_t handle,
-			       uint32_t flags,
-			       int *prime_fd)
+			       uint32_t flags)
 {
 	struct drm_gem_object *obj;
 	int ret = 0;
@@ -434,14 +442,14 @@ int drm_gem_prime_handle_to_fd(struct drm_device *dev,
 	mutex_lock(&file_priv->prime.lock);
 	obj = drm_gem_object_lookup(file_priv, handle);
 	if (!obj)  {
-		ret = -ENOENT;
+		dmabuf = ERR_PTR(-ENOENT);
 		goto out_unlock;
 	}
 
 	dmabuf = drm_prime_lookup_buf_by_handle(&file_priv->prime, handle);
 	if (dmabuf) {
 		get_dma_buf(dmabuf);
-		goto out_have_handle;
+		goto out;
 	}
 
 	mutex_lock(&dev->object_name_lock);
@@ -463,7 +471,6 @@ int drm_gem_prime_handle_to_fd(struct drm_device *dev,
 		/* normally the created dma-buf takes ownership of the ref,
 		 * but if that fails then drop the ref
 		 */
-		ret = PTR_ERR(dmabuf);
 		mutex_unlock(&dev->object_name_lock);
 		goto out;
 	}
@@ -478,34 +485,51 @@ out_have_obj:
 	ret = drm_prime_add_buf_handle(&file_priv->prime,
 				       dmabuf, handle);
 	mutex_unlock(&dev->object_name_lock);
-	if (ret)
-		goto fail_put_dmabuf;
-
-out_have_handle:
-	ret = dma_buf_fd(dmabuf, flags);
-	/*
-	 * We must _not_ remove the buffer from the handle cache since the newly
-	 * created dma buf is already linked in the global obj->dma_buf pointer,
-	 * and that is invariant as long as a userspace gem handle exists.
-	 * Closing the handle will clean out the cache anyway, so we don't leak.
-	 */
-	if (ret < 0) {
-		goto fail_put_dmabuf;
-	} else {
-		*prime_fd = ret;
-		ret = 0;
+	if (ret) {
+		dma_buf_put(dmabuf);
+		dmabuf = ERR_PTR(ret);
 	}
-
-	goto out;
-
-fail_put_dmabuf:
-	dma_buf_put(dmabuf);
 out:
 	drm_gem_object_put(obj);
 out_unlock:
 	mutex_unlock(&file_priv->prime.lock);
+	return dmabuf;
+}
+EXPORT_SYMBOL(drm_gem_prime_handle_to_dmabuf);
 
-	return ret;
+/**
+ * drm_gem_prime_handle_to_fd - PRIME export function for GEM drivers
+ * @dev: dev to export the buffer from
+ * @file_priv: drm file-private structure
+ * @handle: buffer handle to export
+ * @flags: flags like DRM_CLOEXEC
+ * @prime_fd: pointer to storage for the fd id of the create dma-buf
+ *
+ * This is the PRIME export function which must be used mandatorily by GEM
+ * drivers to ensure correct lifetime management of the underlying GEM object.
+ * The actual exporting from GEM object to a dma-buf is done through the
+ * &drm_gem_object_funcs.export callback.
+ */
+int drm_gem_prime_handle_to_fd(struct drm_device *dev,
+			       struct drm_file *file_priv, uint32_t handle,
+			       uint32_t flags,
+			       int *prime_fd)
+{
+	struct dma_buf *dmabuf;
+	int fd = get_unused_fd_flags(flags);
+
+	if (fd < 0)
+		return fd;
+
+	dmabuf = drm_gem_prime_handle_to_dmabuf(dev, file_priv, handle, flags);
+	if (IS_ERR(dmabuf)) {
+		put_unused_fd(fd);
+		return PTR_ERR(dmabuf);
+	}
+
+	fd_install(fd, dmabuf->file);
+	*prime_fd = fd;
+	return 0;
 }
 EXPORT_SYMBOL(drm_gem_prime_handle_to_fd);
 
diff --git a/include/drm/drm_prime.h b/include/drm/drm_prime.h
index 2a1d01e5b56b..fa085c44d4ca 100644
--- a/include/drm/drm_prime.h
+++ b/include/drm/drm_prime.h
@@ -69,6 +69,9 @@ void drm_gem_dmabuf_release(struct dma_buf *dma_buf);
 
 int drm_gem_prime_fd_to_handle(struct drm_device *dev,
 			       struct drm_file *file_priv, int prime_fd, uint32_t *handle);
+struct dma_buf *drm_gem_prime_handle_to_dmabuf(struct drm_device *dev,
+			       struct drm_file *file_priv, uint32_t handle,
+			       uint32_t flags);
 int drm_gem_prime_handle_to_fd(struct drm_device *dev,
 			       struct drm_file *file_priv, uint32_t handle, uint32_t flags,
 			       int *prime_fd);
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index 71a7ce5f2d4c..717307d6b5b7 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -540,26 +540,29 @@ enum kfd_smi_event {
 	KFD_SMI_EVENT_ALL_PROCESS = 64
 };
 
+/* The reason of the page migration event */
 enum KFD_MIGRATE_TRIGGERS {
-	KFD_MIGRATE_TRIGGER_PREFETCH,
-	KFD_MIGRATE_TRIGGER_PAGEFAULT_GPU,
-	KFD_MIGRATE_TRIGGER_PAGEFAULT_CPU,
-	KFD_MIGRATE_TRIGGER_TTM_EVICTION
+	KFD_MIGRATE_TRIGGER_PREFETCH,		/* Prefetch to GPU VRAM or system memory */
+	KFD_MIGRATE_TRIGGER_PAGEFAULT_GPU,	/* GPU page fault recover */
+	KFD_MIGRATE_TRIGGER_PAGEFAULT_CPU,	/* CPU page fault recover */
+	KFD_MIGRATE_TRIGGER_TTM_EVICTION	/* TTM eviction */
 };
 
+/* The reason of user queue evition event */
 enum KFD_QUEUE_EVICTION_TRIGGERS {
-	KFD_QUEUE_EVICTION_TRIGGER_SVM,
-	KFD_QUEUE_EVICTION_TRIGGER_USERPTR,
-	KFD_QUEUE_EVICTION_TRIGGER_TTM,
-	KFD_QUEUE_EVICTION_TRIGGER_SUSPEND,
-	KFD_QUEUE_EVICTION_CRIU_CHECKPOINT,
-	KFD_QUEUE_EVICTION_CRIU_RESTORE
+	KFD_QUEUE_EVICTION_TRIGGER_SVM,		/* SVM buffer migration */
+	KFD_QUEUE_EVICTION_TRIGGER_USERPTR,	/* userptr movement */
+	KFD_QUEUE_EVICTION_TRIGGER_TTM,		/* TTM move buffer */
+	KFD_QUEUE_EVICTION_TRIGGER_SUSPEND,	/* GPU suspend */
+	KFD_QUEUE_EVICTION_CRIU_CHECKPOINT,	/* CRIU checkpoint */
+	KFD_QUEUE_EVICTION_CRIU_RESTORE		/* CRIU restore */
 };
 
+/* The reason of unmap buffer from GPU event */
 enum KFD_SVM_UNMAP_TRIGGERS {
-	KFD_SVM_UNMAP_TRIGGER_MMU_NOTIFY,
-	KFD_SVM_UNMAP_TRIGGER_MMU_NOTIFY_MIGRATE,
-	KFD_SVM_UNMAP_TRIGGER_UNMAP_FROM_CPU
+	KFD_SVM_UNMAP_TRIGGER_MMU_NOTIFY,	/* MMU notifier CPU buffer movement */
+	KFD_SVM_UNMAP_TRIGGER_MMU_NOTIFY_MIGRATE,/* MMU notifier page migration */
+	KFD_SVM_UNMAP_TRIGGER_UNMAP_FROM_CPU	/* Unmap to free the buffer */
 };
 
 #define KFD_SMI_EVENT_MASK_FROM_INDEX(i) (1ULL << ((i) - 1))
@@ -570,6 +573,77 @@ struct kfd_ioctl_smi_events_args {
 	__u32 anon_fd;	/* from KFD */
 };
 
+/*
+ * SVM event tracing via SMI system management interface
+ *
+ * Open event file descriptor
+ *    use ioctl AMDKFD_IOC_SMI_EVENTS, pass in gpuid and return a anonymous file
+ *    descriptor to receive SMI events.
+ *    If calling with sudo permission, then file descriptor can be used to receive
+ *    SVM events from all processes, otherwise, to only receive SVM events of same
+ *    process.
+ *
+ * To enable the SVM event
+ *    Write event file descriptor with KFD_SMI_EVENT_MASK_FROM_INDEX(event) bitmap
+ *    mask to start record the event to the kfifo, use bitmap mask combination
+ *    for multiple events. New event mask will overwrite the previous event mask.
+ *    KFD_SMI_EVENT_MASK_FROM_INDEX(KFD_SMI_EVENT_ALL_PROCESS) bit requires sudo
+ *    permisson to receive SVM events from all process.
+ *
+ * To receive the event
+ *    Application can poll file descriptor to wait for the events, then read event
+ *    from the file into a buffer. Each event is one line string message, starting
+ *    with the event id, then the event specific information.
+ *
+ * To decode event information
+ *    The following event format string macro can be used with sscanf to decode
+ *    the specific event information.
+ *    event triggers: the reason to generate the event, defined as enum for unmap,
+ *    eviction and migrate events.
+ *    node, from, to, prefetch_loc, preferred_loc: GPU ID, or 0 for system memory.
+ *    addr: user mode address, in pages
+ *    size: in pages
+ *    pid: the process ID to generate the event
+ *    ns: timestamp in nanosecond-resolution, starts at system boot time but
+ *        stops during suspend
+ *    migrate_update: GPU page fault is recovered by 'M' for migrate, 'U' for update
+ *    rw: 'W' for write page fault, 'R' for read page fault
+ *    rescheduled: 'R' if the queue restore failed and rescheduled to try again
+ */
+#define KFD_EVENT_FMT_UPDATE_GPU_RESET(reset_seq_num, reset_cause)\
+		"%x %s\n", (reset_seq_num), (reset_cause)
+
+#define KFD_EVENT_FMT_THERMAL_THROTTLING(bitmask, counter)\
+		"%llx:%llx\n", (bitmask), (counter)
+
+#define KFD_EVENT_FMT_VMFAULT(pid, task_name)\
+		"%x:%s\n", (pid), (task_name)
+
+#define KFD_EVENT_FMT_PAGEFAULT_START(ns, pid, addr, node, rw)\
+		"%lld -%d @%lx(%x) %c\n", (ns), (pid), (addr), (node), (rw)
+
+#define KFD_EVENT_FMT_PAGEFAULT_END(ns, pid, addr, node, migrate_update)\
+		"%lld -%d @%lx(%x) %c\n", (ns), (pid), (addr), (node), (migrate_update)
+
+#define KFD_EVENT_FMT_MIGRATE_START(ns, pid, start, size, from, to, prefetch_loc,\
+		preferred_loc, migrate_trigger)\
+		"%lld -%d @%lx(%lx) %x->%x %x:%x %d\n", (ns), (pid), (start), (size),\
+		(from), (to), (prefetch_loc), (preferred_loc), (migrate_trigger)
+
+#define KFD_EVENT_FMT_MIGRATE_END(ns, pid, start, size, from, to, migrate_trigger)\
+		"%lld -%d @%lx(%lx) %x->%x %d\n", (ns), (pid), (start), (size),\
+		(from), (to), (migrate_trigger)
+
+#define KFD_EVENT_FMT_QUEUE_EVICTION(ns, pid, node, evict_trigger)\
+		"%lld -%d %x %d\n", (ns), (pid), (node), (evict_trigger)
+
+#define KFD_EVENT_FMT_QUEUE_RESTORE(ns, pid, node, rescheduled)\
+		"%lld -%d %x %c\n", (ns), (pid), (node), (rescheduled)
+
+#define KFD_EVENT_FMT_UNMAP_FROM_GPU(ns, pid, addr, size, node, unmap_trigger)\
+		"%lld -%d @%lx(%lx) %x %d\n", (ns), (pid), (addr), (size),\
+		(node), (unmap_trigger)
+
 /**************************************************************************************************
  * CRIU IOCTLs (Checkpoint Restore In Userspace)
  *