1 files changed, 441 insertions, 255 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index df110afa97bf..18246b5b6ee3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -25,18 +25,22 @@
  *          Alex Deucher
  *          Jerome Glisse
  */
+
 #include <linux/dma-fence-array.h>
 #include <linux/interval_tree_generic.h>
 #include <linux/idr.h>
 #include <linux/dma-buf.h>
 
 #include <drm/amdgpu_drm.h>
+#include <drm/drm_drv.h>
 #include "amdgpu.h"
 #include "amdgpu_trace.h"
 #include "amdgpu_amdkfd.h"
 #include "amdgpu_gmc.h"
 #include "amdgpu_xgmi.h"
 #include "amdgpu_dma_buf.h"
+#include "amdgpu_res_cursor.h"
+#include "kfd_svm.h"
 
 /**
  * DOC: GPUVM
@@ -92,13 +96,13 @@ struct amdgpu_prt_cb {
 static inline void amdgpu_vm_eviction_lock(struct amdgpu_vm *vm)
 {
 	mutex_lock(&vm->eviction_lock);
-	vm->saved_flags = memalloc_nofs_save();
+	vm->saved_flags = memalloc_noreclaim_save();
 }
 
 static inline int amdgpu_vm_eviction_trylock(struct amdgpu_vm *vm)
 {
 	if (mutex_trylock(&vm->eviction_lock)) {
-		vm->saved_flags = memalloc_nofs_save();
+		vm->saved_flags = memalloc_noreclaim_save();
 		return 1;
 	}
 	return 0;
@@ -106,7 +110,7 @@ static inline int amdgpu_vm_eviction_trylock(struct amdgpu_vm *vm)
 
 static inline void amdgpu_vm_eviction_unlock(struct amdgpu_vm *vm)
 {
-	memalloc_nofs_restore(vm->saved_flags);
+	memalloc_noreclaim_restore(vm->saved_flags);
 	mutex_unlock(&vm->eviction_lock);
 }
 
@@ -300,7 +304,7 @@ static void amdgpu_vm_bo_relocated(struct amdgpu_vm_bo_base *vm_bo)
 static void amdgpu_vm_bo_done(struct amdgpu_vm_bo_base *vm_bo)
 {
 	spin_lock(&vm_bo->vm->invalidated_lock);
-	list_del_init(&vm_bo->vm_status);
+	list_move(&vm_bo->vm_status, &vm_bo->vm->done);
 	spin_unlock(&vm_bo->vm->invalidated_lock);
 }
 
@@ -338,7 +342,7 @@ static void amdgpu_vm_bo_base_init(struct amdgpu_vm_bo_base *base,
 		amdgpu_vm_bo_idle(base);
 
 	if (bo->preferred_domains &
-	    amdgpu_mem_type_to_domain(bo->tbo.mem.mem_type))
+	    amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type))
 		return;
 
 	/*
@@ -609,7 +613,7 @@ void amdgpu_vm_del_from_lru_notify(struct ttm_buffer_object *bo)
 	if (!amdgpu_bo_is_amdgpu_bo(bo))
 		return;
 
-	if (bo->mem.placement & TTM_PL_FLAG_NO_EVICT)
+	if (bo->pin_count)
 		return;
 
 	abo = ttm_to_amdgpu_bo(bo);
@@ -638,27 +642,30 @@ void amdgpu_vm_move_to_lru_tail(struct amdgpu_device *adev,
 	struct amdgpu_vm_bo_base *bo_base;
 
 	if (vm->bulk_moveable) {
-		spin_lock(&ttm_bo_glob.lru_lock);
+		spin_lock(&adev->mman.bdev.lru_lock);
 		ttm_bo_bulk_move_lru_tail(&vm->lru_bulk_move);
-		spin_unlock(&ttm_bo_glob.lru_lock);
+		spin_unlock(&adev->mman.bdev.lru_lock);
 		return;
 	}
 
 	memset(&vm->lru_bulk_move, 0, sizeof(vm->lru_bulk_move));
 
-	spin_lock(&ttm_bo_glob.lru_lock);
+	spin_lock(&adev->mman.bdev.lru_lock);
 	list_for_each_entry(bo_base, &vm->idle, vm_status) {
 		struct amdgpu_bo *bo = bo_base->bo;
+		struct amdgpu_bo *shadow = amdgpu_bo_shadowed(bo);
 
 		if (!bo->parent)
 			continue;
 
-		ttm_bo_move_to_lru_tail(&bo->tbo, &vm->lru_bulk_move);
-		if (bo->shadow)
-			ttm_bo_move_to_lru_tail(&bo->shadow->tbo,
+		ttm_bo_move_to_lru_tail(&bo->tbo, bo->tbo.resource,
+					&vm->lru_bulk_move);
+		if (shadow)
+			ttm_bo_move_to_lru_tail(&shadow->tbo,
+						shadow->tbo.resource,
 						&vm->lru_bulk_move);
 	}
-	spin_unlock(&ttm_bo_glob.lru_lock);
+	spin_unlock(&adev->mman.bdev.lru_lock);
 
 	vm->bulk_moveable = true;
 }
@@ -687,15 +694,21 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 
 	list_for_each_entry_safe(bo_base, tmp, &vm->evicted, vm_status) {
 		struct amdgpu_bo *bo = bo_base->bo;
+		struct amdgpu_bo *shadow = amdgpu_bo_shadowed(bo);
 
 		r = validate(param, bo);
 		if (r)
 			return r;
+		if (shadow) {
+			r = validate(param, shadow);
+			if (r)
+				return r;
+		}
 
 		if (bo->tbo.type != ttm_bo_type_kernel) {
 			amdgpu_vm_bo_moved(bo_base);
 		} else {
-			vm->update_funcs->map_table(bo);
+			vm->update_funcs->map_table(to_amdgpu_bo_vm(bo));
 			amdgpu_vm_bo_relocated(bo_base);
 		}
 	}
@@ -727,7 +740,7 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm)
  *
  * @adev: amdgpu_device pointer
  * @vm: VM to clear BO from
- * @bo: BO to clear
+ * @vmbo: BO to clear
  * @immediate: use an immediate update
  *
  * Root PD needs to be reserved when calling this.
@@ -737,13 +750,14 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm)
  */
 static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 			      struct amdgpu_vm *vm,
-			      struct amdgpu_bo *bo,
+			      struct amdgpu_bo_vm *vmbo,
 			      bool immediate)
 {
 	struct ttm_operation_ctx ctx = { true, false };
 	unsigned level = adev->vm_manager.root_level;
 	struct amdgpu_vm_update_params params;
-	struct amdgpu_bo *ancestor = bo;
+	struct amdgpu_bo *ancestor = &vmbo->bo;
+	struct amdgpu_bo *bo = &vmbo->bo;
 	unsigned entries, ats_entries;
 	uint64_t addr;
 	int r;
@@ -783,14 +797,15 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 	if (r)
 		return r;
 
-	if (bo->shadow) {
-		r = ttm_bo_validate(&bo->shadow->tbo, &bo->shadow->placement,
-				    &ctx);
+	if (vmbo->shadow) {
+		struct amdgpu_bo *shadow = vmbo->shadow;
+
+		r = ttm_bo_validate(&shadow->tbo, &shadow->placement, &ctx);
 		if (r)
 			return r;
 	}
 
-	r = vm->update_funcs->map_table(bo);
+	r = vm->update_funcs->map_table(vmbo);
 	if (r)
 		return r;
 
@@ -814,7 +829,7 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 			amdgpu_gmc_get_vm_pde(adev, level, &value, &flags);
 		}
 
-		r = vm->update_funcs->update(&params, bo, addr, 0, ats_entries,
+		r = vm->update_funcs->update(&params, vmbo, addr, 0, ats_entries,
 					     value, flags);
 		if (r)
 			return r;
@@ -837,7 +852,7 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 			}
 		}
 
-		r = vm->update_funcs->update(&params, bo, addr, 0, entries,
+		r = vm->update_funcs->update(&params, vmbo, addr, 0, entries,
 					     value, flags);
 		if (r)
 			return r;
@@ -847,34 +862,85 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 }
 
 /**
- * amdgpu_vm_bo_param - fill in parameters for PD/PT allocation
+ * amdgpu_vm_pt_create - create bo for PD/PT
  *
  * @adev: amdgpu_device pointer
  * @vm: requesting vm
  * @level: the page table level
  * @immediate: use a immediate update
- * @bp: resulting BO allocation parameters
+ * @vmbo: pointer to the buffer object pointer
  */
-static void amdgpu_vm_bo_param(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+static int amdgpu_vm_pt_create(struct amdgpu_device *adev,
+			       struct amdgpu_vm *vm,
 			       int level, bool immediate,
-			       struct amdgpu_bo_param *bp)
+			       struct amdgpu_bo_vm **vmbo)
 {
-	memset(bp, 0, sizeof(*bp));
+	struct amdgpu_bo_param bp;
+	struct amdgpu_bo *bo;
+	struct dma_resv *resv;
+	unsigned int num_entries;
+	int r;
 
-	bp->size = amdgpu_vm_bo_size(adev, level);
-	bp->byte_align = AMDGPU_GPU_PAGE_SIZE;
-	bp->domain = AMDGPU_GEM_DOMAIN_VRAM;
-	bp->domain = amdgpu_bo_get_preferred_pin_domain(adev, bp->domain);
-	bp->flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
+	memset(&bp, 0, sizeof(bp));
+
+	bp.size = amdgpu_vm_bo_size(adev, level);
+	bp.byte_align = AMDGPU_GPU_PAGE_SIZE;
+	bp.domain = AMDGPU_GEM_DOMAIN_VRAM;
+	bp.domain = amdgpu_bo_get_preferred_pin_domain(adev, bp.domain);
+	bp.flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
 		AMDGPU_GEM_CREATE_CPU_GTT_USWC;
+
+	if (level < AMDGPU_VM_PTB)
+		num_entries = amdgpu_vm_num_entries(adev, level);
+	else
+		num_entries = 0;
+
+	bp.bo_ptr_size = struct_size((*vmbo), entries, num_entries);
+
 	if (vm->use_cpu_for_update)
-		bp->flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
-	else if (!vm->root.base.bo || vm->root.base.bo->shadow)
-		bp->flags |= AMDGPU_GEM_CREATE_SHADOW;
-	bp->type = ttm_bo_type_kernel;
-	bp->no_wait_gpu = immediate;
+		bp.flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
+
+	bp.type = ttm_bo_type_kernel;
+	bp.no_wait_gpu = immediate;
 	if (vm->root.base.bo)
-		bp->resv = vm->root.base.bo->tbo.base.resv;
+		bp.resv = vm->root.base.bo->tbo.base.resv;
+
+	r = amdgpu_bo_create_vm(adev, &bp, vmbo);
+	if (r)
+		return r;
+
+	bo = &(*vmbo)->bo;
+	if (vm->is_compute_context || (adev->flags & AMD_IS_APU)) {
+		(*vmbo)->shadow = NULL;
+		return 0;
+	}
+
+	if (!bp.resv)
+		WARN_ON(dma_resv_lock(bo->tbo.base.resv,
+				      NULL));
+	resv = bp.resv;
+	memset(&bp, 0, sizeof(bp));
+	bp.size = amdgpu_vm_bo_size(adev, level);
+	bp.domain = AMDGPU_GEM_DOMAIN_GTT;
+	bp.flags = AMDGPU_GEM_CREATE_CPU_GTT_USWC;
+	bp.type = ttm_bo_type_kernel;
+	bp.resv = bo->tbo.base.resv;
+	bp.bo_ptr_size = sizeof(struct amdgpu_bo);
+
+	r = amdgpu_bo_create(adev, &bp, &(*vmbo)->shadow);
+
+	if (!resv)
+		dma_resv_unlock(bo->tbo.base.resv);
+
+	if (r) {
+		amdgpu_bo_unref(&bo);
+		return r;
+	}
+
+	(*vmbo)->shadow->parent = amdgpu_bo_ref(bo);
+	amdgpu_bo_add_to_shadow_list((*vmbo)->shadow);
+
+	return 0;
 }
 
 /**
@@ -897,35 +963,33 @@ static int amdgpu_vm_alloc_pts(struct amdgpu_device *adev,
 			       bool immediate)
 {
 	struct amdgpu_vm_pt *entry = cursor->entry;
-	struct amdgpu_bo_param bp;
-	struct amdgpu_bo *pt;
+	struct amdgpu_bo *pt_bo;
+	struct amdgpu_bo_vm *pt;
 	int r;
 
-	if (cursor->level < AMDGPU_VM_PTB && !entry->entries) {
-		unsigned num_entries;
-
-		num_entries = amdgpu_vm_num_entries(adev, cursor->level);
-		entry->entries = kvmalloc_array(num_entries,
-						sizeof(*entry->entries),
-						GFP_KERNEL | __GFP_ZERO);
-		if (!entry->entries)
-			return -ENOMEM;
-	}
-
-	if (entry->base.bo)
+	if (entry->base.bo) {
+		if (cursor->level < AMDGPU_VM_PTB)
+			entry->entries =
+				to_amdgpu_bo_vm(entry->base.bo)->entries;
+		else
+			entry->entries = NULL;
 		return 0;
+	}
 
-	amdgpu_vm_bo_param(adev, vm, cursor->level, immediate, &bp);
-
-	r = amdgpu_bo_create(adev, &bp, &pt);
+	r = amdgpu_vm_pt_create(adev, vm, cursor->level, immediate, &pt);
 	if (r)
 		return r;
 
 	/* Keep a reference to the root directory to avoid
 	 * freeing them up in the wrong order.
 	 */
-	pt->parent = amdgpu_bo_ref(cursor->parent->base.bo);
-	amdgpu_vm_bo_base_init(&entry->base, vm, pt);
+	pt_bo = &pt->bo;
+	pt_bo->parent = amdgpu_bo_ref(cursor->parent->base.bo);
+	amdgpu_vm_bo_base_init(&entry->base, vm, pt_bo);
+	if (cursor->level < AMDGPU_VM_PTB)
+		entry->entries = pt->entries;
+	else
+		entry->entries = NULL;
 
 	r = amdgpu_vm_clear_bo(adev, vm, pt, immediate);
 	if (r)
@@ -935,7 +999,7 @@ static int amdgpu_vm_alloc_pts(struct amdgpu_device *adev,
 
 error_free_pt:
 	amdgpu_bo_unref(&pt->shadow);
-	amdgpu_bo_unref(&pt);
+	amdgpu_bo_unref(&pt_bo);
 	return r;
 }
 
@@ -946,13 +1010,15 @@ error_free_pt:
  */
 static void amdgpu_vm_free_table(struct amdgpu_vm_pt *entry)
 {
+	struct amdgpu_bo *shadow;
+
 	if (entry->base.bo) {
+		shadow = amdgpu_bo_shadowed(entry->base.bo);
 		entry->base.bo->vm_bo = NULL;
 		list_del(&entry->base.vm_status);
-		amdgpu_bo_unref(&entry->base.bo->shadow);
+		amdgpu_bo_unref(&shadow);
 		amdgpu_bo_unref(&entry->base.bo);
 	}
-	kvfree(entry->entries);
 	entry->entries = NULL;
 }
 
@@ -1251,7 +1317,8 @@ static int amdgpu_vm_update_pde(struct amdgpu_vm_update_params *params,
 	level += params->adev->vm_manager.root_level;
 	amdgpu_gmc_get_pde_for_bo(entry->base.bo, level, &pt, &flags);
 	pde = (entry - parent->entries) * 8;
-	return vm->update_funcs->update(params, bo, pde, pt, 1, 0, flags);
+	return vm->update_funcs->update(params, to_amdgpu_bo_vm(bo), pde, pt,
+					1, 0, flags);
 }
 
 /**
@@ -1331,9 +1398,9 @@ error:
  * Make sure to set the right flags for the PTEs at the desired level.
  */
 static void amdgpu_vm_update_flags(struct amdgpu_vm_update_params *params,
-				   struct amdgpu_bo *bo, unsigned level,
+				   struct amdgpu_bo_vm *pt, unsigned int level,
 				   uint64_t pe, uint64_t addr,
-				   unsigned count, uint32_t incr,
+				   unsigned int count, uint32_t incr,
 				   uint64_t flags)
 
 {
@@ -1349,7 +1416,7 @@ static void amdgpu_vm_update_flags(struct amdgpu_vm_update_params *params,
 		flags |= AMDGPU_PTE_EXECUTABLE;
 	}
 
-	params->vm->update_funcs->update(params, bo, pe, addr, count, incr,
+	params->vm->update_funcs->update(params, pt, pe, addr, count, incr,
 					 flags);
 }
 
@@ -1529,9 +1596,9 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 						    nptes, dst, incr, upd_flags,
 						    vm->task_info.pid,
 						    vm->immediate.fence_context);
-			amdgpu_vm_update_flags(params, pt, cursor.level,
-					       pe_start, dst, nptes, incr,
-					       upd_flags);
+			amdgpu_vm_update_flags(params, to_amdgpu_bo_vm(pt),
+					       cursor.level, pe_start, dst,
+					       nptes, incr, upd_flags);
 
 			pe_start += nptes * 8;
 			dst += nptes * incr;
@@ -1554,7 +1621,11 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 			 * completely covered by the range and so potentially still in use.
 			 */
 			while (cursor.pfn < frag_start) {
-				amdgpu_vm_free_pts(adev, params->vm, &cursor);
+				/* Make sure previous mapping is freed */
+				if (cursor.entry->base.bo) {
+					params->table_freed = true;
+					amdgpu_vm_free_pts(adev, params->vm, &cursor);
+				}
 				amdgpu_vm_pt_next(adev, &cursor);
 			}
 
@@ -1570,7 +1641,8 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 /**
  * amdgpu_vm_bo_update_mapping - update a mapping in the vm page table
  *
- * @adev: amdgpu_device pointer
+ * @adev: amdgpu_device pointer of the VM
+ * @bo_adev: amdgpu_device pointer of the mapped BO
  * @vm: requested vm
  * @immediate: immediate submission in a page fault
  * @unlocked: unlocked invalidation during MM callback
@@ -1578,26 +1650,35 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
  * @start: start of mapped range
  * @last: last mapped entry
  * @flags: flags for the entries
- * @addr: addr to set the area to
+ * @offset: offset into nodes and pages_addr
+ * @res: ttm_resource to map
  * @pages_addr: DMA addresses to use for mapping
  * @fence: optional resulting fence
+ * @table_freed: return true if page table is freed
  *
  * Fill in the page table entries between @start and @last.
  *
  * Returns:
  * 0 for success, -EINVAL for failure.
  */
-static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
-				       struct amdgpu_vm *vm, bool immediate,
-				       bool unlocked, struct dma_resv *resv,
-				       uint64_t start, uint64_t last,
-				       uint64_t flags, uint64_t addr,
-				       dma_addr_t *pages_addr,
-				       struct dma_fence **fence)
+int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
+				struct amdgpu_device *bo_adev,
+				struct amdgpu_vm *vm, bool immediate,
+				bool unlocked, struct dma_resv *resv,
+				uint64_t start, uint64_t last,
+				uint64_t flags, uint64_t offset,
+				struct ttm_resource *res,
+				dma_addr_t *pages_addr,
+				struct dma_fence **fence,
+				bool *table_freed)
 {
 	struct amdgpu_vm_update_params params;
+	struct amdgpu_res_cursor cursor;
 	enum amdgpu_sync_mode sync_mode;
-	int r;
+	int r, idx;
+
+	if (!drm_dev_enter(&adev->ddev, &idx))
+		return -ENODEV;
 
 	memset(&params, 0, sizeof(params));
 	params.adev = adev;
@@ -1632,137 +1713,121 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
 	if (r)
 		goto error_unlock;
 
-	r = amdgpu_vm_update_ptes(&params, start, last + 1, addr, flags);
-	if (r)
-		goto error_unlock;
-
-	r = vm->update_funcs->commit(&params, fence);
-
-error_unlock:
-	amdgpu_vm_eviction_unlock(vm);
-	return r;
-}
-
-/**
- * amdgpu_vm_bo_split_mapping - split a mapping into smaller chunks
- *
- * @adev: amdgpu_device pointer
- * @resv: fences we need to sync to
- * @pages_addr: DMA addresses to use for mapping
- * @vm: requested vm
- * @mapping: mapped range and flags to use for the update
- * @flags: HW flags for the mapping
- * @bo_adev: amdgpu_device pointer that bo actually been allocated
- * @nodes: array of drm_mm_nodes with the MC addresses
- * @fence: optional resulting fence
- *
- * Split the mapping into smaller chunks so that each update fits
- * into a SDMA IB.
- *
- * Returns:
- * 0 for success, -EINVAL for failure.
- */
-static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
-				      struct dma_resv *resv,
-				      dma_addr_t *pages_addr,
-				      struct amdgpu_vm *vm,
-				      struct amdgpu_bo_va_mapping *mapping,
-				      uint64_t flags,
-				      struct amdgpu_device *bo_adev,
-				      struct drm_mm_node *nodes,
-				      struct dma_fence **fence)
-{
-	unsigned min_linear_pages = 1 << adev->vm_manager.fragment_size;
-	uint64_t pfn, start = mapping->start;
-	int r;
-
-	/* normally,bo_va->flags only contians READABLE and WIRTEABLE bit go here
-	 * but in case of something, we filter the flags in first place
-	 */
-	if (!(mapping->flags & AMDGPU_PTE_READABLE))
-		flags &= ~AMDGPU_PTE_READABLE;
-	if (!(mapping->flags & AMDGPU_PTE_WRITEABLE))
-		flags &= ~AMDGPU_PTE_WRITEABLE;
-
-	/* Apply ASIC specific mapping flags */
-	amdgpu_gmc_get_vm_pte(adev, mapping, &flags);
-
-	trace_amdgpu_vm_bo_update(mapping);
-
-	pfn = mapping->offset >> PAGE_SHIFT;
-	if (nodes) {
-		while (pfn >= nodes->size) {
-			pfn -= nodes->size;
-			++nodes;
-		}
-	}
-
-	do {
-		dma_addr_t *dma_addr = NULL;
-		uint64_t max_entries;
-		uint64_t addr, last;
-
-		max_entries = mapping->last - start + 1;
-		if (nodes) {
-			addr = nodes->start << PAGE_SHIFT;
-			max_entries = min((nodes->size - pfn) *
-				AMDGPU_GPU_PAGES_IN_CPU_PAGE, max_entries);
-		} else {
-			addr = 0;
-		}
+	amdgpu_res_first(pages_addr ? NULL : res, offset,
+			 (last - start + 1) * AMDGPU_GPU_PAGE_SIZE, &cursor);
+	while (cursor.remaining) {
+		uint64_t tmp, num_entries, addr;
 
+		num_entries = cursor.size >> AMDGPU_GPU_PAGE_SHIFT;
 		if (pages_addr) {
-			uint64_t count;
+			bool contiguous = true;
 
-			for (count = 1;
-			     count < max_entries / AMDGPU_GPU_PAGES_IN_CPU_PAGE;
-			     ++count) {
-				uint64_t idx = pfn + count;
+			if (num_entries > AMDGPU_GPU_PAGES_IN_CPU_PAGE) {
+				uint64_t pfn = cursor.start >> PAGE_SHIFT;
+				uint64_t count;
 
-				if (pages_addr[idx] !=
-				    (pages_addr[idx - 1] + PAGE_SIZE))
-					break;
+				contiguous = pages_addr[pfn + 1] ==
+					pages_addr[pfn] + PAGE_SIZE;
+
+				tmp = num_entries /
+					AMDGPU_GPU_PAGES_IN_CPU_PAGE;
+				for (count = 2; count < tmp; ++count) {
+					uint64_t idx = pfn + count;
+
+					if (contiguous != (pages_addr[idx] ==
+					    pages_addr[idx - 1] + PAGE_SIZE))
+						break;
+				}
+				num_entries = count *
+					AMDGPU_GPU_PAGES_IN_CPU_PAGE;
 			}
 
-			if (count < min_linear_pages) {
-				addr = pfn << PAGE_SHIFT;
-				dma_addr = pages_addr;
+			if (!contiguous) {
+				addr = cursor.start;
+				params.pages_addr = pages_addr;
 			} else {
-				addr = pages_addr[pfn];
-				max_entries = count *
-					AMDGPU_GPU_PAGES_IN_CPU_PAGE;
+				addr = pages_addr[cursor.start >> PAGE_SHIFT];
+				params.pages_addr = NULL;
 			}
 
 		} else if (flags & (AMDGPU_PTE_VALID | AMDGPU_PTE_PRT)) {
-			addr += bo_adev->vm_manager.vram_base_offset;
-			addr += pfn << PAGE_SHIFT;
+			addr = bo_adev->vm_manager.vram_base_offset +
+				cursor.start;
+		} else {
+			addr = 0;
 		}
 
-		last = start + max_entries - 1;
-		r = amdgpu_vm_bo_update_mapping(adev, vm, false, false, resv,
-						start, last, flags, addr,
-						dma_addr, fence);
+		tmp = start + num_entries;
+		r = amdgpu_vm_update_ptes(&params, start, tmp, addr, flags);
 		if (r)
-			return r;
+			goto error_unlock;
 
-		pfn += (last - start + 1) / AMDGPU_GPU_PAGES_IN_CPU_PAGE;
-		if (nodes && nodes->size == pfn) {
-			pfn = 0;
-			++nodes;
-		}
-		start = last + 1;
+		amdgpu_res_next(&cursor, num_entries * AMDGPU_GPU_PAGE_SIZE);
+		start = tmp;
+	}
 
-	} while (unlikely(start != mapping->last + 1));
+	r = vm->update_funcs->commit(&params, fence);
 
-	return 0;
+	if (table_freed)
+		*table_freed = *table_freed || params.table_freed;
+
+error_unlock:
+	amdgpu_vm_eviction_unlock(vm);
+	drm_dev_exit(idx);
+	return r;
 }
 
+void amdgpu_vm_get_memory(struct amdgpu_vm *vm, uint64_t *vram_mem,
+				uint64_t *gtt_mem, uint64_t *cpu_mem)
+{
+	struct amdgpu_bo_va *bo_va, *tmp;
+
+	list_for_each_entry_safe(bo_va, tmp, &vm->idle, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		amdgpu_bo_get_memory(bo_va->base.bo, vram_mem,
+				gtt_mem, cpu_mem);
+	}
+	list_for_each_entry_safe(bo_va, tmp, &vm->evicted, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		amdgpu_bo_get_memory(bo_va->base.bo, vram_mem,
+				gtt_mem, cpu_mem);
+	}
+	list_for_each_entry_safe(bo_va, tmp, &vm->relocated, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		amdgpu_bo_get_memory(bo_va->base.bo, vram_mem,
+				gtt_mem, cpu_mem);
+	}
+	list_for_each_entry_safe(bo_va, tmp, &vm->moved, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		amdgpu_bo_get_memory(bo_va->base.bo, vram_mem,
+				gtt_mem, cpu_mem);
+	}
+	spin_lock(&vm->invalidated_lock);
+	list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		amdgpu_bo_get_memory(bo_va->base.bo, vram_mem,
+				gtt_mem, cpu_mem);
+	}
+	list_for_each_entry_safe(bo_va, tmp, &vm->done, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		amdgpu_bo_get_memory(bo_va->base.bo, vram_mem,
+				gtt_mem, cpu_mem);
+	}
+	spin_unlock(&vm->invalidated_lock);
+}
 /**
  * amdgpu_vm_bo_update - update all BO mappings in the vm page table
  *
  * @adev: amdgpu_device pointer
  * @bo_va: requested BO and VM object
  * @clear: if true clear the entries
+ * @table_freed: return true if page table is freed
  *
  * Fill in the page table entries for @bo_va.
  *
@@ -1770,14 +1835,13 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
  * 0 for success, -EINVAL for failure.
  */
 int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
-			bool clear)
+			bool clear, bool *table_freed)
 {
 	struct amdgpu_bo *bo = bo_va->base.bo;
 	struct amdgpu_vm *vm = bo_va->base.vm;
 	struct amdgpu_bo_va_mapping *mapping;
 	dma_addr_t *pages_addr = NULL;
 	struct ttm_resource *mem;
-	struct drm_mm_node *nodes;
 	struct dma_fence **last_update;
 	struct dma_resv *resv;
 	uint64_t flags;
@@ -1786,11 +1850,9 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 
 	if (clear || !bo) {
 		mem = NULL;
-		nodes = NULL;
 		resv = vm->root.base.bo->tbo.base.resv;
 	} else {
 		struct drm_gem_object *obj = &bo->tbo.base;
-		struct ttm_dma_tt *ttm;
 
 		resv = bo->tbo.base.resv;
 		if (obj->import_attach && bo_va->is_xgmi) {
@@ -1798,15 +1860,13 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 			struct drm_gem_object *gobj = dma_buf->priv;
 			struct amdgpu_bo *abo = gem_to_amdgpu_bo(gobj);
 
-			if (abo->tbo.mem.mem_type == TTM_PL_VRAM)
+			if (abo->tbo.resource->mem_type == TTM_PL_VRAM)
 				bo = gem_to_amdgpu_bo(gobj);
 		}
-		mem = &bo->tbo.mem;
-		nodes = mem->mm_node;
-		if (mem->mem_type == TTM_PL_TT) {
-			ttm = container_of(bo->tbo.ttm, struct ttm_dma_tt, ttm);
-			pages_addr = ttm->dma_address;
-		}
+		mem = bo->tbo.resource;
+		if (mem->mem_type == TTM_PL_TT ||
+		    mem->mem_type == AMDGPU_PL_PREEMPT)
+			pages_addr = bo->tbo.ttm->dma_address;
 	}
 
 	if (bo) {
@@ -1835,9 +1895,26 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 	}
 
 	list_for_each_entry(mapping, &bo_va->invalids, list) {
-		r = amdgpu_vm_bo_split_mapping(adev, resv, pages_addr, vm,
-					       mapping, flags, bo_adev, nodes,
-					       last_update);
+		uint64_t update_flags = flags;
+
+		/* normally,bo_va->flags only contians READABLE and WIRTEABLE bit go here
+		 * but in case of something, we filter the flags in first place
+		 */
+		if (!(mapping->flags & AMDGPU_PTE_READABLE))
+			update_flags &= ~AMDGPU_PTE_READABLE;
+		if (!(mapping->flags & AMDGPU_PTE_WRITEABLE))
+			update_flags &= ~AMDGPU_PTE_WRITEABLE;
+
+		/* Apply ASIC specific mapping flags */
+		amdgpu_gmc_get_vm_pte(adev, mapping, &update_flags);
+
+		trace_amdgpu_vm_bo_update(mapping);
+
+		r = amdgpu_vm_bo_update_mapping(adev, bo_adev, vm, false, false,
+						resv, mapping->start,
+						mapping->last, update_flags,
+						mapping->offset, mem,
+						pages_addr, last_update, table_freed);
 		if (r)
 			return r;
 	}
@@ -1847,7 +1924,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 	 * next command submission.
 	 */
 	if (bo && bo->tbo.base.resv == vm->root.base.bo->tbo.base.resv) {
-		uint32_t mem_type = bo->tbo.mem.mem_type;
+		uint32_t mem_type = bo->tbo.resource->mem_type;
 
 		if (!(bo->preferred_domains &
 		      amdgpu_mem_type_to_domain(mem_type)))
@@ -1988,13 +2065,12 @@ static void amdgpu_vm_prt_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	unsigned i, shared_count;
 	int r;
 
-	r = dma_resv_get_fences_rcu(resv, &excl,
-					      &shared_count, &shared);
+	r = dma_resv_get_fences(resv, &excl, &shared_count, &shared);
 	if (r) {
 		/* Not enough memory to grab the fence list, as last resort
 		 * block for all the fences to complete.
 		 */
-		dma_resv_wait_timeout_rcu(resv, true, false,
+		dma_resv_wait_timeout(resv, true, false,
 						    MAX_SCHEDULE_TIMEOUT);
 		return;
 	}
@@ -2045,9 +2121,10 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 		    mapping->start < AMDGPU_GMC_HOLE_START)
 			init_pte_value = AMDGPU_PTE_DEFAULT_ATC;
 
-		r = amdgpu_vm_bo_update_mapping(adev, vm, false, false, resv,
-						mapping->start, mapping->last,
-						init_pte_value, 0, NULL, &f);
+		r = amdgpu_vm_bo_update_mapping(adev, adev, vm, false, false,
+						resv, mapping->start,
+						mapping->last, init_pte_value,
+						0, NULL, NULL, &f, NULL);
 		amdgpu_vm_free_mapping(adev, vm, mapping, f);
 		if (r) {
 			dma_fence_put(f);
@@ -2089,7 +2166,7 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev,
 
 	list_for_each_entry_safe(bo_va, tmp, &vm->moved, base.vm_status) {
 		/* Per VM BOs never need to bo cleared in the page tables */
-		r = amdgpu_vm_bo_update(adev, bo_va, false);
+		r = amdgpu_vm_bo_update(adev, bo_va, false, NULL);
 		if (r)
 			return r;
 	}
@@ -2108,7 +2185,7 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev,
 		else
 			clear = true;
 
-		r = amdgpu_vm_bo_update(adev, bo_va, clear);
+		r = amdgpu_vm_bo_update(adev, bo_va, clear, NULL);
 		if (r)
 			return r;
 
@@ -2166,7 +2243,7 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 
 
 /**
- * amdgpu_vm_bo_insert_mapping - insert a new mapping
+ * amdgpu_vm_bo_insert_map - insert a new mapping
  *
  * @adev: amdgpu_device pointer
  * @bo_va: bo_va to store the address
@@ -2223,8 +2300,8 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 	uint64_t eaddr;
 
 	/* validate the parameters */
-	if (saddr & AMDGPU_GPU_PAGE_MASK || offset & AMDGPU_GPU_PAGE_MASK ||
-	    size == 0 || size & AMDGPU_GPU_PAGE_MASK)
+	if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK ||
+	    size == 0 || size & ~PAGE_MASK)
 		return -EINVAL;
 
 	/* make sure object fit at this offset */
@@ -2289,8 +2366,8 @@ int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev,
 	int r;
 
 	/* validate the parameters */
-	if (saddr & AMDGPU_GPU_PAGE_MASK || offset & AMDGPU_GPU_PAGE_MASK ||
-	    size == 0 || size & AMDGPU_GPU_PAGE_MASK)
+	if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK ||
+	    size == 0 || size & ~PAGE_MASK)
 		return -EINVAL;
 
 	/* make sure object fit at this offset */
@@ -2435,7 +2512,7 @@ int amdgpu_vm_bo_clear_mappings(struct amdgpu_device *adev,
 			after->start = eaddr + 1;
 			after->last = tmp->last;
 			after->offset = tmp->offset;
-			after->offset += after->start - tmp->start;
+			after->offset += (after->start - tmp->start) << PAGE_SHIFT;
 			after->flags = tmp->flags;
 			after->bo_va = tmp->bo_va;
 			list_add(&after->list, &tmp->bo_va->invalids);
@@ -2605,7 +2682,7 @@ bool amdgpu_vm_evictable(struct amdgpu_bo *bo)
 		return true;
 
 	/* Don't evict VM page tables while they are busy */
-	if (!dma_resv_test_signaled_rcu(bo->tbo.base.resv, true))
+	if (!dma_resv_test_signaled(bo->tbo.base.resv, true))
 		return false;
 
 	/* Try to block ongoing updates */
@@ -2638,7 +2715,7 @@ void amdgpu_vm_bo_invalidate(struct amdgpu_device *adev,
 	struct amdgpu_vm_bo_base *bo_base;
 
 	/* shadow bo doesn't have bo base, its validation needs its parent */
-	if (bo->parent && bo->parent->shadow == bo)
+	if (bo->parent && (amdgpu_bo_shadowed(bo->parent) == bo))
 		bo = bo->parent;
 
 	for (bo_base = bo->vm_bo; bo_base; bo_base = bo_base->next) {
@@ -2785,8 +2862,8 @@ void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint32_t min_vm_size,
  */
 long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout)
 {
-	timeout = dma_resv_wait_timeout_rcu(vm->root.base.bo->tbo.base.resv,
-					    true, true, timeout);
+	timeout = dma_resv_wait_timeout(vm->root.base.bo->tbo.base.resv, true,
+					true, timeout);
 	if (timeout <= 0)
 		return timeout;
 
@@ -2798,7 +2875,6 @@ long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout)
  *
  * @adev: amdgpu_device pointer
  * @vm: requested vm
- * @vm_context: Indicates if it GFX or Compute context
  * @pasid: Process address space identifier
  *
  * Init @vm fields.
@@ -2806,11 +2882,10 @@ long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout)
  * Returns:
  * 0 for success, error for failure.
  */
-int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
-		   int vm_context, u32 pasid)
+int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, u32 pasid)
 {
-	struct amdgpu_bo_param bp;
-	struct amdgpu_bo *root;
+	struct amdgpu_bo *root_bo;
+	struct amdgpu_bo_vm *root;
 	int r, i;
 
 	vm->va = RB_ROOT_CACHED;
@@ -2823,7 +2898,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	INIT_LIST_HEAD(&vm->invalidated);
 	spin_lock_init(&vm->invalidated_lock);
 	INIT_LIST_HEAD(&vm->freed);
-
+	INIT_LIST_HEAD(&vm->done);
 
 	/* create scheduler entities for page table updates */
 	r = drm_sched_entity_init(&vm->immediate, DRM_SCHED_PRIORITY_NORMAL,
@@ -2841,16 +2916,9 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	vm->pte_support_ats = false;
 	vm->is_compute_context = false;
 
-	if (vm_context == AMDGPU_VM_CONTEXT_COMPUTE) {
-		vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode &
-						AMDGPU_VM_USE_CPU_FOR_COMPUTE);
+	vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode &
+				    AMDGPU_VM_USE_CPU_FOR_GFX);
 
-		if (adev->asic_type == CHIP_RAVEN)
-			vm->pte_support_ats = true;
-	} else {
-		vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode &
-						AMDGPU_VM_USE_CPU_FOR_GFX);
-	}
 	DRM_DEBUG_DRIVER("VM update mode is %s\n",
 			 vm->use_cpu_for_update ? "CPU" : "SDMA");
 	WARN_ONCE((vm->use_cpu_for_update &&
@@ -2867,22 +2935,20 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	mutex_init(&vm->eviction_lock);
 	vm->evicting = false;
 
-	amdgpu_vm_bo_param(adev, vm, adev->vm_manager.root_level, false, &bp);
-	if (vm_context == AMDGPU_VM_CONTEXT_COMPUTE)
-		bp.flags &= ~AMDGPU_GEM_CREATE_SHADOW;
-	r = amdgpu_bo_create(adev, &bp, &root);
+	r = amdgpu_vm_pt_create(adev, vm, adev->vm_manager.root_level,
+				false, &root);
 	if (r)
 		goto error_free_delayed;
-
-	r = amdgpu_bo_reserve(root, true);
+	root_bo = &root->bo;
+	r = amdgpu_bo_reserve(root_bo, true);
 	if (r)
 		goto error_free_root;
 
-	r = dma_resv_reserve_shared(root->tbo.base.resv, 1);
+	r = dma_resv_reserve_shared(root_bo->tbo.base.resv, 1);
 	if (r)
 		goto error_unreserve;
 
-	amdgpu_vm_bo_base_init(&vm->root.base, vm, root);
+	amdgpu_vm_bo_base_init(&vm->root.base, vm, root_bo);
 
 	r = amdgpu_vm_clear_bo(adev, vm, root, false);
 	if (r)
@@ -2911,8 +2977,8 @@ error_unreserve:
 	amdgpu_bo_unreserve(vm->root.base.bo);
 
 error_free_root:
-	amdgpu_bo_unref(&vm->root.base.bo->shadow);
-	amdgpu_bo_unref(&vm->root.base.bo);
+	amdgpu_bo_unref(&root->shadow);
+	amdgpu_bo_unref(&root_bo);
 	vm->root.base.bo = NULL;
 
 error_free_delayed:
@@ -3010,7 +3076,9 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	 */
 	if (pte_support_ats != vm->pte_support_ats) {
 		vm->pte_support_ats = pte_support_ats;
-		r = amdgpu_vm_clear_bo(adev, vm, vm->root.base.bo, false);
+		r = amdgpu_vm_clear_bo(adev, vm,
+				       to_amdgpu_bo_vm(vm->root.base.bo),
+				       false);
 		if (r)
 			goto free_idr;
 	}
@@ -3054,7 +3122,7 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	}
 
 	/* Free the shadow bo for compute VM */
-	amdgpu_bo_unref(&vm->root.base.bo->shadow);
+	amdgpu_bo_unref(&to_amdgpu_bo_vm(vm->root.base.bo)->shadow);
 
 	if (pasid)
 		vm->pasid = pasid;
@@ -3173,6 +3241,12 @@ void amdgpu_vm_manager_init(struct amdgpu_device *adev)
 {
 	unsigned i;
 
+	/* Concurrent flushes are only possible starting with Vega10 and
+	 * are broken on Navi10 and Navi14.
+	 */
+	adev->vm_manager.concurrent_flush = !(adev->asic_type < CHIP_VEGA10 ||
+					      adev->asic_type == CHIP_NAVI10 ||
+					      adev->asic_type == CHIP_NAVI14);
 	amdgpu_vmid_mgr_init(adev);
 
 	adev->vm_manager.fence_context =
@@ -3323,22 +3397,33 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm)
 bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
 			    uint64_t addr)
 {
+	bool is_compute_context = false;
 	struct amdgpu_bo *root;
 	uint64_t value, flags;
 	struct amdgpu_vm *vm;
-	long r;
+	int r;
 
 	spin_lock(&adev->vm_manager.pasid_lock);
 	vm = idr_find(&adev->vm_manager.pasid_idr, pasid);
-	if (vm)
+	if (vm) {
 		root = amdgpu_bo_ref(vm->root.base.bo);
-	else
+		is_compute_context = vm->is_compute_context;
+	} else {
 		root = NULL;
+	}
 	spin_unlock(&adev->vm_manager.pasid_lock);
 
 	if (!root)
 		return false;
 
+	addr /= AMDGPU_GPU_PAGE_SIZE;
+
+	if (is_compute_context &&
+	    !svm_range_restore_pages(adev, pasid, addr)) {
+		amdgpu_bo_unref(&root);
+		return true;
+	}
+
 	r = amdgpu_bo_reserve(root, true);
 	if (r)
 		goto error_unref;
@@ -3352,18 +3437,16 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
 	if (!vm)
 		goto error_unlock;
 
-	addr /= AMDGPU_GPU_PAGE_SIZE;
 	flags = AMDGPU_PTE_VALID | AMDGPU_PTE_SNOOPED |
 		AMDGPU_PTE_SYSTEM;
 
-	if (vm->is_compute_context) {
+	if (is_compute_context) {
 		/* Intentionally setting invalid PTE flag
 		 * combination to force a no-retry-fault
 		 */
 		flags = AMDGPU_PTE_EXECUTABLE | AMDGPU_PDE_PTE |
 			AMDGPU_PTE_TF;
 		value = 0;
-
 	} else if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_NEVER) {
 		/* Redirect the access to the dummy page */
 		value = adev->dummy_page_addr;
@@ -3375,8 +3458,15 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
 		value = 0;
 	}
 
-	r = amdgpu_vm_bo_update_mapping(adev, vm, true, false, NULL, addr,
-					addr + 1, flags, value, NULL, NULL);
+	r = dma_resv_reserve_shared(root->tbo.base.resv, 1);
+	if (r) {
+		pr_debug("failed %d to reserve fence slot\n", r);
+		goto error_unlock;
+	}
+
+	r = amdgpu_vm_bo_update_mapping(adev, adev, vm, true, false, NULL, addr,
+					addr, flags, value, NULL, NULL, NULL,
+					NULL);
 	if (r)
 		goto error_unlock;
 
@@ -3385,10 +3475,106 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
 error_unlock:
 	amdgpu_bo_unreserve(root);
 	if (r < 0)
-		DRM_ERROR("Can't handle page fault (%ld)\n", r);
+		DRM_ERROR("Can't handle page fault (%d)\n", r);
 
 error_unref:
 	amdgpu_bo_unref(&root);
 
 	return false;
 }
+
+#if defined(CONFIG_DEBUG_FS)
+/**
+ * amdgpu_debugfs_vm_bo_info  - print BO info for the VM
+ *
+ * @vm: Requested VM for printing BO info
+ * @m: debugfs file
+ *
+ * Print BO information in debugfs file for the VM
+ */
+void amdgpu_debugfs_vm_bo_info(struct amdgpu_vm *vm, struct seq_file *m)
+{
+	struct amdgpu_bo_va *bo_va, *tmp;
+	u64 total_idle = 0;
+	u64 total_evicted = 0;
+	u64 total_relocated = 0;
+	u64 total_moved = 0;
+	u64 total_invalidated = 0;
+	u64 total_done = 0;
+	unsigned int total_idle_objs = 0;
+	unsigned int total_evicted_objs = 0;
+	unsigned int total_relocated_objs = 0;
+	unsigned int total_moved_objs = 0;
+	unsigned int total_invalidated_objs = 0;
+	unsigned int total_done_objs = 0;
+	unsigned int id = 0;
+
+	seq_puts(m, "\tIdle BOs:\n");
+	list_for_each_entry_safe(bo_va, tmp, &vm->idle, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		total_idle += amdgpu_bo_print_info(id++, bo_va->base.bo, m);
+	}
+	total_idle_objs = id;
+	id = 0;
+
+	seq_puts(m, "\tEvicted BOs:\n");
+	list_for_each_entry_safe(bo_va, tmp, &vm->evicted, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		total_evicted += amdgpu_bo_print_info(id++, bo_va->base.bo, m);
+	}
+	total_evicted_objs = id;
+	id = 0;
+
+	seq_puts(m, "\tRelocated BOs:\n");
+	list_for_each_entry_safe(bo_va, tmp, &vm->relocated, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		total_relocated += amdgpu_bo_print_info(id++, bo_va->base.bo, m);
+	}
+	total_relocated_objs = id;
+	id = 0;
+
+	seq_puts(m, "\tMoved BOs:\n");
+	list_for_each_entry_safe(bo_va, tmp, &vm->moved, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		total_moved += amdgpu_bo_print_info(id++, bo_va->base.bo, m);
+	}
+	total_moved_objs = id;
+	id = 0;
+
+	seq_puts(m, "\tInvalidated BOs:\n");
+	spin_lock(&vm->invalidated_lock);
+	list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		total_invalidated += amdgpu_bo_print_info(id++,	bo_va->base.bo, m);
+	}
+	total_invalidated_objs = id;
+	id = 0;
+
+	seq_puts(m, "\tDone BOs:\n");
+	list_for_each_entry_safe(bo_va, tmp, &vm->done, base.vm_status) {
+		if (!bo_va->base.bo)
+			continue;
+		total_done += amdgpu_bo_print_info(id++, bo_va->base.bo, m);
+	}
+	spin_unlock(&vm->invalidated_lock);
+	total_done_objs = id;
+
+	seq_printf(m, "\tTotal idle size:        %12lld\tobjs:\t%d\n", total_idle,
+		   total_idle_objs);
+	seq_printf(m, "\tTotal evicted size:     %12lld\tobjs:\t%d\n", total_evicted,
+		   total_evicted_objs);
+	seq_printf(m, "\tTotal relocated size:   %12lld\tobjs:\t%d\n", total_relocated,
+		   total_relocated_objs);
+	seq_printf(m, "\tTotal moved size:       %12lld\tobjs:\t%d\n", total_moved,
+		   total_moved_objs);
+	seq_printf(m, "\tTotal invalidated size: %12lld\tobjs:\t%d\n", total_invalidated,
+		   total_invalidated_objs);
+	seq_printf(m, "\tTotal done size:        %12lld\tobjs:\t%d\n", total_done,
+		   total_done_objs);
+}
+#endif