btrfs: don't hold dev_replace rwsem over whole of btrfs_map_block

Don't hold the dev_replace rwsem for the entirety of btrfs_map_block().

It is only needed to protect:

a) calls to find_live_mirror() and
b) calling into handle_ops_on_dev_replace().

But there is no need to hold the rwsem for any kind of set_io_stripe()
calls.

So relax taking the dev_replace rwsem to only protect both cases and check
if the device replace status has changed in the meantime, for which we have
to re-do the find_live_mirror() calls.

This fixes a deadlock on raid-stripe-tree where device replace performs a
scrub operation, which in turn calls into btrfs_map_block() to find the
physical location of the block.

Reviewed-by: Josef Bacik <[email protected]>
Reviewed-by: Qu Wenruo <[email protected]>
Reviewed-by: Filipe Manana <[email protected]>
Signed-off-by: Johannes Thumshirn <[email protected]>
Signed-off-by: David Sterba <[email protected]>

fs/btrfs/volumes.c

@@ -6650,14 +6650,9 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 	max_len = btrfs_max_io_len(map, map_offset, &io_geom);
 	*length = min_t(u64, map->chunk_len - map_offset, max_len);
 
+again:
 	down_read(&dev_replace->rwsem);
 	dev_replace_is_ongoing = btrfs_dev_replace_is_ongoing(dev_replace);
-	/*
-	 * Hold the semaphore for read during the whole operation, write is
-	 * requested at commit time but must wait.
-	 */
-	if (!dev_replace_is_ongoing)
-		up_read(&dev_replace->rwsem);
 
 	switch (map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
 	case BTRFS_BLOCK_GROUP_RAID0:
@@ -6695,6 +6690,7 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 			   "stripe index math went horribly wrong, got stripe_index=%u, num_stripes=%u",
 			   io_geom.stripe_index, map->num_stripes);
 		ret = -EINVAL;
+		up_read(&dev_replace->rwsem);
 		goto out;
 	}
 
@@ -6710,6 +6706,8 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 		 */
 		num_alloc_stripes += 2;
 
+	up_read(&dev_replace->rwsem);
+
 	/*
 	 * If this I/O maps to a single device, try to return the device and
 	 * physical block information on the stack instead of allocating an
@@ -6782,25 +6780,34 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 		goto out;
 	}
 
+	/*
+	 * Check if something changed the dev_replace state since
+	 * we've checked it for the last time and if redo the whole
+	 * mapping operation.
+	 */
+	down_read(&dev_replace->rwsem);
+	if (dev_replace_is_ongoing !=
+	    btrfs_dev_replace_is_ongoing(dev_replace)) {
+		btrfs_put_bioc(bioc);
+		up_read(&dev_replace->rwsem);
+		goto again;
+	}
+
 	if (op != BTRFS_MAP_READ)
 		io_geom.max_errors = btrfs_chunk_max_errors(map);
 
 	if (dev_replace_is_ongoing && dev_replace->tgtdev != NULL &&
 	    op != BTRFS_MAP_READ) {
 		handle_ops_on_dev_replace(bioc, dev_replace, logical, &io_geom);
 	}
+	up_read(&dev_replace->rwsem);
 
 	*bioc_ret = bioc;
 	bioc->num_stripes = io_geom.num_stripes;
 	bioc->max_errors = io_geom.max_errors;
 	bioc->mirror_num = io_geom.mirror_num;
 
 out:
-	if (dev_replace_is_ongoing) {
-		lockdep_assert_held(&dev_replace->rwsem);
-		/* Unlock and let waiting writers proceed */
-		up_read(&dev_replace->rwsem);
-	}
 	btrfs_free_chunk_map(map);
 	return ret;
 }