3 files changed, 144 insertions, 62 deletions

diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 33d20be87db5..9b07b7d44a58 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -158,7 +158,8 @@ static void ttm_bo_release_list(struct kref *list_kref)
 	ttm_mem_global_free(bdev->glob->mem_glob, acc_size);
 }
 
-int ttm_bo_wait_unreserved(struct ttm_buffer_object *bo, bool interruptible)
+static int ttm_bo_wait_unreserved(struct ttm_buffer_object *bo,
+				  bool interruptible)
 {
 	if (interruptible) {
 		return wait_event_interruptible(bo->event_queue,
@@ -168,7 +169,6 @@ int ttm_bo_wait_unreserved(struct ttm_buffer_object *bo, bool interruptible)
 		return 0;
 	}
 }
-EXPORT_SYMBOL(ttm_bo_wait_unreserved);
 
 void ttm_bo_add_to_lru(struct ttm_buffer_object *bo)
 {
@@ -213,14 +213,13 @@ int ttm_bo_del_from_lru(struct ttm_buffer_object *bo)
 	return put_count;
 }
 
-int ttm_bo_reserve_locked(struct ttm_buffer_object *bo,
+int ttm_bo_reserve_nolru(struct ttm_buffer_object *bo,
 			  bool interruptible,
 			  bool no_wait, bool use_sequence, uint32_t sequence)
 {
-	struct ttm_bo_global *glob = bo->glob;
 	int ret;
 
-	while (unlikely(atomic_read(&bo->reserved) != 0)) {
+	while (unlikely(atomic_xchg(&bo->reserved, 1) != 0)) {
 		/**
 		 * Deadlock avoidance for multi-bo reserving.
 		 */
@@ -241,26 +240,36 @@ int ttm_bo_reserve_locked(struct ttm_buffer_object *bo,
 		if (no_wait)
 			return -EBUSY;
 
-		spin_unlock(&glob->lru_lock);
 		ret = ttm_bo_wait_unreserved(bo, interruptible);
-		spin_lock(&glob->lru_lock);
 
 		if (unlikely(ret))
 			return ret;
 	}
 
-	atomic_set(&bo->reserved, 1);
 	if (use_sequence) {
+		bool wake_up = false;
 		/**
 		 * Wake up waiters that may need to recheck for deadlock,
 		 * if we decreased the sequence number.
 		 */
 		if (unlikely((bo->val_seq - sequence < (1 << 31))
 			     || !bo->seq_valid))
-			wake_up_all(&bo->event_queue);
+			wake_up = true;
 
+		/*
+		 * In the worst case with memory ordering these values can be
+		 * seen in the wrong order. However since we call wake_up_all
+		 * in that case, this will hopefully not pose a problem,
+		 * and the worst case would only cause someone to accidentally
+		 * hit -EAGAIN in ttm_bo_reserve when they see old value of
+		 * val_seq. However this would only happen if seq_valid was
+		 * written before val_seq was, and just means some slightly
+		 * increased cpu usage
+		 */
 		bo->val_seq = sequence;
 		bo->seq_valid = true;
+		if (wake_up)
+			wake_up_all(&bo->event_queue);
 	} else {
 		bo->seq_valid = false;
 	}
@@ -289,17 +298,64 @@ int ttm_bo_reserve(struct ttm_buffer_object *bo,
 	int put_count = 0;
 	int ret;
 
-	spin_lock(&glob->lru_lock);
-	ret = ttm_bo_reserve_locked(bo, interruptible, no_wait, use_sequence,
-				    sequence);
-	if (likely(ret == 0))
+	ret = ttm_bo_reserve_nolru(bo, interruptible, no_wait, use_sequence,
+				   sequence);
+	if (likely(ret == 0)) {
+		spin_lock(&glob->lru_lock);
 		put_count = ttm_bo_del_from_lru(bo);
-	spin_unlock(&glob->lru_lock);
+		spin_unlock(&glob->lru_lock);
+		ttm_bo_list_ref_sub(bo, put_count, true);
+	}
 
-	ttm_bo_list_ref_sub(bo, put_count, true);
+	return ret;
+}
+
+int ttm_bo_reserve_slowpath_nolru(struct ttm_buffer_object *bo,
+				  bool interruptible, uint32_t sequence)
+{
+	bool wake_up = false;
+	int ret;
+
+	while (unlikely(atomic_xchg(&bo->reserved, 1) != 0)) {
+		WARN_ON(bo->seq_valid && sequence == bo->val_seq);
+
+		ret = ttm_bo_wait_unreserved(bo, interruptible);
 
+		if (unlikely(ret))
+			return ret;
+	}
+
+	if ((bo->val_seq - sequence < (1 << 31)) || !bo->seq_valid)
+		wake_up = true;
+
+	/**
+	 * Wake up waiters that may need to recheck for deadlock,
+	 * if we decreased the sequence number.
+	 */
+	bo->val_seq = sequence;
+	bo->seq_valid = true;
+	if (wake_up)
+		wake_up_all(&bo->event_queue);
+
+	return 0;
+}
+
+int ttm_bo_reserve_slowpath(struct ttm_buffer_object *bo,
+			    bool interruptible, uint32_t sequence)
+{
+	struct ttm_bo_global *glob = bo->glob;
+	int put_count, ret;
+
+	ret = ttm_bo_reserve_slowpath_nolru(bo, interruptible, sequence);
+	if (likely(!ret)) {
+		spin_lock(&glob->lru_lock);
+		put_count = ttm_bo_del_from_lru(bo);
+		spin_unlock(&glob->lru_lock);
+		ttm_bo_list_ref_sub(bo, put_count, true);
+	}
 	return ret;
 }
+EXPORT_SYMBOL(ttm_bo_reserve_slowpath);
 
 void ttm_bo_unreserve_locked(struct ttm_buffer_object *bo)
 {
@@ -434,6 +490,7 @@ static int ttm_bo_handle_move_mem(struct ttm_buffer_object *bo,
 			bo->mem = tmp_mem;
 			bdev->driver->move_notify(bo, mem);
 			bo->mem = *mem;
+			*mem = tmp_mem;
 		}
 
 		goto out_err;
@@ -510,7 +567,7 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
 	int ret;
 
 	spin_lock(&glob->lru_lock);
-	ret = ttm_bo_reserve_locked(bo, false, true, false, 0);
+	ret = ttm_bo_reserve_nolru(bo, false, true, false, 0);
 
 	spin_lock(&bdev->fence_lock);
 	(void) ttm_bo_wait(bo, false, false, true);
@@ -603,7 +660,7 @@ static int ttm_bo_cleanup_refs_and_unlock(struct ttm_buffer_object *bo,
 			return ret;
 
 		spin_lock(&glob->lru_lock);
-		ret = ttm_bo_reserve_locked(bo, false, true, false, 0);
+		ret = ttm_bo_reserve_nolru(bo, false, true, false, 0);
 
 		/*
 		 * We raced, and lost, someone else holds the reservation now,
@@ -667,7 +724,14 @@ static int ttm_bo_delayed_delete(struct ttm_bo_device *bdev, bool remove_all)
 			kref_get(&nentry->list_kref);
 		}
 
-		ret = ttm_bo_reserve_locked(entry, false, !remove_all, false, 0);
+		ret = ttm_bo_reserve_nolru(entry, false, true, false, 0);
+		if (remove_all && ret) {
+			spin_unlock(&glob->lru_lock);
+			ret = ttm_bo_reserve_nolru(entry, false, false,
+						   false, 0);
+			spin_lock(&glob->lru_lock);
+		}
+
 		if (!ret)
 			ret = ttm_bo_cleanup_refs_and_unlock(entry, false,
 							     !remove_all);
@@ -815,7 +879,7 @@ static int ttm_mem_evict_first(struct ttm_bo_device *bdev,
 
 	spin_lock(&glob->lru_lock);
 	list_for_each_entry(bo, &man->lru, lru) {
-		ret = ttm_bo_reserve_locked(bo, false, true, false, 0);
+		ret = ttm_bo_reserve_nolru(bo, false, true, false, 0);
 		if (!ret)
 			break;
 	}
@@ -1796,7 +1860,7 @@ static int ttm_bo_swapout(struct ttm_mem_shrink *shrink)
 
 	spin_lock(&glob->lru_lock);
 	list_for_each_entry(bo, &glob->swap_lru, swap) {
-		ret = ttm_bo_reserve_locked(bo, false, true, false, 0);
+		ret = ttm_bo_reserve_nolru(bo, false, true, false, 0);
 		if (!ret)
 			break;
 	}
diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index d73d6e3e17b2..8be35c809c7b 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -344,8 +344,12 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 
 	if (ttm->state == tt_unpopulated) {
 		ret = ttm->bdev->driver->ttm_tt_populate(ttm);
-		if (ret)
+		if (ret) {
+			/* if we fail here don't nuke the mm node
+			 * as the bo still owns it */
+			old_copy.mm_node = NULL;
 			goto out1;
+		}
 	}
 
 	add = 0;
@@ -371,8 +375,11 @@ int ttm_bo_move_memcpy(struct ttm_buffer_object *bo,
 						   prot);
 		} else
 			ret = ttm_copy_io_page(new_iomap, old_iomap, page);
-		if (ret)
+		if (ret) {
+			/* failing here, means keep old copy as-is */
+			old_copy.mm_node = NULL;
 			goto out1;
+		}
 	}
 	mb();
 out2:
@@ -422,7 +429,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo,
 	struct ttm_bo_device *bdev = bo->bdev;
 	struct ttm_bo_driver *driver = bdev->driver;
 
-	fbo = kzalloc(sizeof(*fbo), GFP_KERNEL);
+	fbo = kmalloc(sizeof(*fbo), GFP_KERNEL);
 	if (!fbo)
 		return -ENOMEM;
 
@@ -441,7 +448,12 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo,
 	fbo->vm_node = NULL;
 	atomic_set(&fbo->cpu_writers, 0);
 
-	fbo->sync_obj = driver->sync_obj_ref(bo->sync_obj);
+	spin_lock(&bdev->fence_lock);
+	if (bo->sync_obj)
+		fbo->sync_obj = driver->sync_obj_ref(bo->sync_obj);
+	else
+		fbo->sync_obj = NULL;
+	spin_unlock(&bdev->fence_lock);
 	kref_init(&fbo->list_kref);
 	kref_init(&fbo->kref);
 	fbo->destroy = &ttm_transfered_destroy;
@@ -654,13 +666,11 @@ int ttm_bo_move_accel_cleanup(struct ttm_buffer_object *bo,
 		 */
 
 		set_bit(TTM_BO_PRIV_FLAG_MOVING, &bo->priv_flags);
-
-		/* ttm_buffer_object_transfer accesses bo->sync_obj */
-		ret = ttm_buffer_object_transfer(bo, &ghost_obj);
 		spin_unlock(&bdev->fence_lock);
 		if (tmp_obj)
 			driver->sync_obj_unref(&tmp_obj);
 
+		ret = ttm_buffer_object_transfer(bo, &ghost_obj);
 		if (ret)
 			return ret;
 
diff --git a/drivers/gpu/drm/ttm/ttm_execbuf_util.c b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
index cd9e4523dc56..7b90def15674 100644
--- a/drivers/gpu/drm/ttm/ttm_execbuf_util.c
+++ b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
@@ -82,22 +82,6 @@ static void ttm_eu_list_ref_sub(struct list_head *list)
 	}
 }
 
-static int ttm_eu_wait_unreserved_locked(struct list_head *list,
-					 struct ttm_buffer_object *bo)
-{
-	struct ttm_bo_global *glob = bo->glob;
-	int ret;
-
-	ttm_eu_del_from_lru_locked(list);
-	spin_unlock(&glob->lru_lock);
-	ret = ttm_bo_wait_unreserved(bo, true);
-	spin_lock(&glob->lru_lock);
-	if (unlikely(ret != 0))
-		ttm_eu_backoff_reservation_locked(list);
-	return ret;
-}
-
-
 void ttm_eu_backoff_reservation(struct list_head *list)
 {
 	struct ttm_validate_buffer *entry;
@@ -145,47 +129,65 @@ int ttm_eu_reserve_buffers(struct list_head *list)
 	entry = list_first_entry(list, struct ttm_validate_buffer, head);
 	glob = entry->bo->glob;
 
-retry:
 	spin_lock(&glob->lru_lock);
 	val_seq = entry->bo->bdev->val_seq++;
 
+retry:
 	list_for_each_entry(entry, list, head) {
 		struct ttm_buffer_object *bo = entry->bo;
 
-retry_this_bo:
-		ret = ttm_bo_reserve_locked(bo, true, true, true, val_seq);
+		/* already slowpath reserved? */
+		if (entry->reserved)
+			continue;
+
+		ret = ttm_bo_reserve_nolru(bo, true, true, true, val_seq);
 		switch (ret) {
 		case 0:
 			break;
 		case -EBUSY:
-			ret = ttm_eu_wait_unreserved_locked(list, bo);
-			if (unlikely(ret != 0)) {
-				spin_unlock(&glob->lru_lock);
-				ttm_eu_list_ref_sub(list);
-				return ret;
-			}
-			goto retry_this_bo;
+			ttm_eu_del_from_lru_locked(list);
+			spin_unlock(&glob->lru_lock);
+			ret = ttm_bo_reserve_nolru(bo, true, false,
+						   true, val_seq);
+			spin_lock(&glob->lru_lock);
+			if (!ret)
+				break;
+
+			if (unlikely(ret != -EAGAIN))
+				goto err;
+
+			/* fallthrough */
 		case -EAGAIN:
 			ttm_eu_backoff_reservation_locked(list);
+
+			/*
+			 * temporarily increase sequence number every retry,
+			 * to prevent us from seeing our old reservation
+			 * sequence when someone else reserved the buffer,
+			 * but hasn't updated the seq_valid/seqno members yet.
+			 */
+			val_seq = entry->bo->bdev->val_seq++;
+
 			spin_unlock(&glob->lru_lock);
 			ttm_eu_list_ref_sub(list);
-			ret = ttm_bo_wait_unreserved(bo, true);
+			ret = ttm_bo_reserve_slowpath_nolru(bo, true, val_seq);
 			if (unlikely(ret != 0))
 				return ret;
+			spin_lock(&glob->lru_lock);
+			entry->reserved = true;
+			if (unlikely(atomic_read(&bo->cpu_writers) > 0)) {
+				ret = -EBUSY;
+				goto err;
+			}
 			goto retry;
 		default:
-			ttm_eu_backoff_reservation_locked(list);
-			spin_unlock(&glob->lru_lock);
-			ttm_eu_list_ref_sub(list);
-			return ret;
+			goto err;
 		}
 
 		entry->reserved = true;
 		if (unlikely(atomic_read(&bo->cpu_writers) > 0)) {
-			ttm_eu_backoff_reservation_locked(list);
-			spin_unlock(&glob->lru_lock);
-			ttm_eu_list_ref_sub(list);
-			return -EBUSY;
+			ret = -EBUSY;
+			goto err;
 		}
 	}
 
@@ -194,6 +196,12 @@ retry_this_bo:
 	ttm_eu_list_ref_sub(list);
 
 	return 0;
+
+err:
+	ttm_eu_backoff_reservation_locked(list);
+	spin_unlock(&glob->lru_lock);
+	ttm_eu_list_ref_sub(list);
+	return ret;
 }
 EXPORT_SYMBOL(ttm_eu_reserve_buffers);