Commit 62f30c54 authored Nov 10, 2011 by Miao Xie Committed by Chris Mason Nov 10, 2011

Btrfs: fix deadlock caused by the race between relocation



We can not do flushable reservation for the relocation when we create snapshot,
because it may make the transaction commit task and the flush task wait for
each other and the deadlock happens.

Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>

parent 2f120c05

fs/btrfs/transaction.c

+2 −2

Original line number	Diff line number	Diff line
		@@ -882,7 +882,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
		btrfs_reloc_pre_snapshot(trans, pending, &to_reserve);

		if (to_reserve > 0) {
		ret = btrfs_block_rsv_add(root, &pending->block_rsv,
		ret = btrfs_block_rsv_add_noflush(root, &pending->block_rsv,
		to_reserve);
		if (ret) {
		pending->error = ret;