drm/amdgpu: extract amdgpu_vm_lock_by_pasid from amdgpu_vm_handle_fault

This is tricky to implement right and we're going to need
it from the devcoredump.

Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

2 files changed, 54 insertions, 29 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 00a532f4e027..115a7b269af3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2950,6 +2950,50 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 }
 
 /**
+ * amdgpu_vm_lock_by_pasid - return an amdgpu_vm and its root bo from a pasid, if possible.
+ * @adev: amdgpu device pointer
+ * @root: root BO of the VM
+ * @pasid: PASID of the VM
+ * The caller needs to unreserve and unref the root bo on success.
+ */
+struct amdgpu_vm *amdgpu_vm_lock_by_pasid(struct amdgpu_device *adev,
+					  struct amdgpu_bo **root, u32 pasid)
+{
+	unsigned long irqflags;
+	struct amdgpu_vm *vm;
+	int r;
+
+	xa_lock_irqsave(&adev->vm_manager.pasids, irqflags);
+	vm = xa_load(&adev->vm_manager.pasids, pasid);
+	*root = vm ? amdgpu_bo_ref(vm->root.bo) : NULL;
+	xa_unlock_irqrestore(&adev->vm_manager.pasids, irqflags);
+
+	if (!*root)
+		return NULL;
+
+	r = amdgpu_bo_reserve(*root, true);
+	if (r)
+		goto error_unref;
+
+	/* Double check that the VM still exists */
+	xa_lock_irqsave(&adev->vm_manager.pasids, irqflags);
+	vm = xa_load(&adev->vm_manager.pasids, pasid);
+	if (vm && vm->root.bo != *root)
+		vm = NULL;
+	xa_unlock_irqrestore(&adev->vm_manager.pasids, irqflags);
+	if (!vm)
+		goto error_unlock;
+
+	return vm;
+error_unlock:
+	amdgpu_bo_unreserve(*root);
+
+error_unref:
+	amdgpu_bo_unref(root);
+	return NULL;
+}
+
+/**
  * amdgpu_vm_handle_fault - graceful handling of VM faults.
  * @adev: amdgpu device pointer
  * @pasid: PASID of the VM
@@ -2964,50 +3008,29 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
  * shouldn't be reported any more.
  */
 bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
-			    u32 vmid, u32 node_id, uint64_t addr, uint64_t ts,
-			    bool write_fault)
+			    u32 vmid, u32 node_id, uint64_t addr,
+			    uint64_t ts, bool write_fault)
 {
 	bool is_compute_context = false;
 	struct amdgpu_bo *root;
-	unsigned long irqflags;
 	uint64_t value, flags;
 	struct amdgpu_vm *vm;
 	int r;
 
-	xa_lock_irqsave(&adev->vm_manager.pasids, irqflags);
-	vm = xa_load(&adev->vm_manager.pasids, pasid);
-	if (vm) {
-		root = amdgpu_bo_ref(vm->root.bo);
-		is_compute_context = vm->is_compute_context;
-	} else {
-		root = NULL;
-	}
-	xa_unlock_irqrestore(&adev->vm_manager.pasids, irqflags);
-
-	if (!root)
+	vm = amdgpu_vm_lock_by_pasid(adev, &root, pasid);
+	if (!vm)
 		return false;
 
+	is_compute_context = vm->is_compute_context;
+
 	if (is_compute_context && !svm_range_restore_pages(adev, pasid, vmid,
 	    node_id, addr >> PAGE_SHIFT, ts, write_fault)) {
+		amdgpu_bo_unreserve(root);
 		amdgpu_bo_unref(&root);
 		return true;
 	}
 
 	addr /= AMDGPU_GPU_PAGE_SIZE;
-
-	r = amdgpu_bo_reserve(root, true);
-	if (r)
-		goto error_unref;
-
-	/* Double check that the VM still exists */
-	xa_lock_irqsave(&adev->vm_manager.pasids, irqflags);
-	vm = xa_load(&adev->vm_manager.pasids, pasid);
-	if (vm && vm->root.bo != root)
-		vm = NULL;
-	xa_unlock_irqrestore(&adev->vm_manager.pasids, irqflags);
-	if (!vm)
-		goto error_unlock;
-
 	flags = AMDGPU_PTE_VALID | AMDGPU_PTE_SNOOPED |
 		AMDGPU_PTE_SYSTEM;
 
@@ -3046,7 +3069,6 @@ error_unlock:
 	if (r < 0)
 		dev_err(adev->dev, "Can't handle page fault (%d)\n", r);
 
-error_unref:
 	amdgpu_bo_unref(&root);
 
 	return false;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 3b32f41c3655..d083d7aab75c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -592,6 +592,9 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
 			    u32 vmid, u32 node_id, uint64_t addr, uint64_t ts,
 			    bool write_fault);
 
+struct amdgpu_vm *amdgpu_vm_lock_by_pasid(struct amdgpu_device *adev,
+					  struct amdgpu_bo **root, u32 pasid);
+
 void amdgpu_vm_set_task_info(struct amdgpu_vm *vm);
 
 void amdgpu_vm_move_to_lru_tail(struct amdgpu_device *adev,