1 files changed, 72 insertions, 15 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 098dd1ba751a..b17771dd5ce7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -172,6 +172,7 @@ static void amdgpu_vm_bo_base_init(struct amdgpu_vm_bo_base *base,
 	 * is validated on next vm use to avoid fault.
 	 * */
 	list_move_tail(&base->vm_status, &vm->evicted);
+	base->moved = true;
 }
 
 /**
@@ -369,7 +370,6 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 	uint64_t addr;
 	int r;
 
-	addr = amdgpu_bo_gpu_offset(bo);
 	entries = amdgpu_bo_size(bo) / 8;
 
 	if (pte_support_ats) {
@@ -387,7 +387,7 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 		ats_entries = 0;
 	}
 
-	ring = container_of(vm->entity.sched, struct amdgpu_ring, sched);
+	ring = container_of(vm->entity.rq->sched, struct amdgpu_ring, sched);
 
 	r = reservation_object_reserve_shared(bo->tbo.resv);
 	if (r)
@@ -401,6 +401,7 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
 	if (r)
 		goto error;
 
+	addr = amdgpu_bo_gpu_offset(bo);
 	if (ats_entries) {
 		uint64_t ats_value;
 
@@ -495,11 +496,12 @@ static int amdgpu_vm_alloc_levels(struct amdgpu_device *adev,
 	eaddr = eaddr & ((1 << shift) - 1);
 
 	flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
+	if (vm->root.base.bo->shadow)
+		flags |= AMDGPU_GEM_CREATE_SHADOW;
 	if (vm->use_cpu_for_update)
 		flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
 	else
-		flags |= (AMDGPU_GEM_CREATE_NO_CPU_ACCESS |
-				AMDGPU_GEM_CREATE_SHADOW);
+		flags |= AMDGPU_GEM_CREATE_NO_CPU_ACCESS;
 
 	/* walk over the address space and allocate the page tables */
 	for (pt_idx = from; pt_idx <= to; ++pt_idx) {
@@ -1113,7 +1115,7 @@ restart:
 		struct amdgpu_ring *ring;
 		struct dma_fence *fence;
 
-		ring = container_of(vm->entity.sched, struct amdgpu_ring,
+		ring = container_of(vm->entity.rq->sched, struct amdgpu_ring,
 				    sched);
 
 		amdgpu_ring_pad_ib(ring, params.ib);
@@ -1403,7 +1405,7 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev,
 					   addr, flags);
 	}
 
-	ring = container_of(vm->entity.sched, struct amdgpu_ring, sched);
+	ring = container_of(vm->entity.rq->sched, struct amdgpu_ring, sched);
 
 	nptes = last - start + 1;
 
@@ -2344,6 +2346,35 @@ struct amdgpu_bo_va_mapping *amdgpu_vm_bo_lookup_mapping(struct amdgpu_vm *vm,
 }
 
 /**
+ * amdgpu_vm_bo_trace_cs - trace all reserved mappings
+ *
+ * @vm: the requested vm
+ * @ticket: CS ticket
+ *
+ * Trace all mappings of BOs reserved during a command submission.
+ */
+void amdgpu_vm_bo_trace_cs(struct amdgpu_vm *vm, struct ww_acquire_ctx *ticket)
+{
+	struct amdgpu_bo_va_mapping *mapping;
+
+	if (!trace_amdgpu_vm_bo_cs_enabled())
+		return;
+
+	for (mapping = amdgpu_vm_it_iter_first(&vm->va, 0, U64_MAX); mapping;
+	     mapping = amdgpu_vm_it_iter_next(mapping, 0, U64_MAX)) {
+		if (mapping->bo_va && mapping->bo_va->base.bo) {
+			struct amdgpu_bo *bo;
+
+			bo = mapping->bo_va->base.bo;
+			if (READ_ONCE(bo->tbo.resv->lock.ctx) != ticket)
+				continue;
+		}
+
+		trace_amdgpu_vm_bo_cs(mapping);
+	}
+}
+
+/**
  * amdgpu_vm_bo_rmv - remove a bo to a specific vm
  *
  * @adev: amdgpu_device pointer
@@ -2453,28 +2484,52 @@ static uint32_t amdgpu_vm_get_block_size(uint64_t vm_size)
  * amdgpu_vm_adjust_size - adjust vm size, block size and fragment size
  *
  * @adev: amdgpu_device pointer
- * @vm_size: the default vm size if it's set auto
+ * @min_vm_size: the minimum vm size in GB if it's set auto
  * @fragment_size_default: Default PTE fragment size
  * @max_level: max VMPT level
  * @max_bits: max address space size in bits
  *
  */
-void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint32_t vm_size,
+void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint32_t min_vm_size,
 			   uint32_t fragment_size_default, unsigned max_level,
 			   unsigned max_bits)
 {
+	unsigned int max_size = 1 << (max_bits - 30);
+	unsigned int vm_size;
 	uint64_t tmp;
 
 	/* adjust vm size first */
 	if (amdgpu_vm_size != -1) {
-		unsigned max_size = 1 << (max_bits - 30);
-
 		vm_size = amdgpu_vm_size;
 		if (vm_size > max_size) {
 			dev_warn(adev->dev, "VM size (%d) too large, max is %u GB\n",
 				 amdgpu_vm_size, max_size);
 			vm_size = max_size;
 		}
+	} else {
+		struct sysinfo si;
+		unsigned int phys_ram_gb;
+
+		/* Optimal VM size depends on the amount of physical
+		 * RAM available. Underlying requirements and
+		 * assumptions:
+		 *
+		 *  - Need to map system memory and VRAM from all GPUs
+		 *     - VRAM from other GPUs not known here
+		 *     - Assume VRAM <= system memory
+		 *  - On GFX8 and older, VM space can be segmented for
+		 *    different MTYPEs
+		 *  - Need to allow room for fragmentation, guard pages etc.
+		 *
+		 * This adds up to a rough guess of system memory x3.
+		 * Round up to power of two to maximize the available
+		 * VM size with the given page table size.
+		 */
+		si_meminfo(&si);
+		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));
 	}
 
 	adev->vm_manager.max_pfn = (uint64_t)vm_size << 18;
@@ -2587,7 +2642,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
 	if (vm->use_cpu_for_update)
 		flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
-	else
+	else if (vm_context != AMDGPU_VM_CONTEXT_COMPUTE)
 		flags |= AMDGPU_GEM_CREATE_SHADOW;
 
 	size = amdgpu_vm_bo_size(adev, adev->vm_manager.root_level);
@@ -2642,7 +2697,7 @@ error_free_root:
 	vm->root.base.bo = NULL;
 
 error_free_sched_entity:
-	drm_sched_entity_destroy(&ring->sched, &vm->entity);
+	drm_sched_entity_destroy(&vm->entity);
 
 	return r;
 }
@@ -2662,8 +2717,7 @@ error_free_sched_entity:
  * - pasid (old PASID is released, because compute manages its own PASIDs)
  *
  * Reinitializes the page directory to reflect the changed ATS
- * setting. May leave behind an unused shadow BO for the page
- * directory when switching from SDMA updates to CPU updates.
+ * setting.
  *
  * Returns:
  * 0 for success, -errno for errors.
@@ -2713,6 +2767,9 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 		vm->pasid = 0;
 	}
 
+	/* Free the shadow bo for compute VM */
+	amdgpu_bo_unref(&vm->root.base.bo->shadow);
+
 error:
 	amdgpu_bo_unreserve(vm->root.base.bo);
 	return r;
@@ -2779,7 +2836,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 		spin_unlock_irqrestore(&adev->vm_manager.pasid_lock, flags);
 	}
 
-	drm_sched_entity_destroy(vm->entity.sched, &vm->entity);
+	drm_sched_entity_destroy(&vm->entity);
 
 	if (!RB_EMPTY_ROOT(&vm->va.rb_root)) {
 		dev_err(adev->dev, "still active bo inside vm\n");