From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dkim1.fusionio.com ([66.114.96.53]:57569 "EHLO dkim1.fusionio.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756078Ab3FMRwi (ORCPT ); Thu, 13 Jun 2013 13:52:38 -0400 Received: from mx1.fusionio.com (unknown [10.101.1.160]) by dkim1.fusionio.com (Postfix) with ESMTP id 3ADEF7C069A for ; Thu, 13 Jun 2013 11:52:37 -0600 (MDT) Received: from mail1.int.fusionio.com (mail1.int.fusionio.com [10.101.1.21]) by mx1.fusionio.com with ESMTP id LRDDASsvPEwWiJia (version=TLSv1 cipher=AES128-SHA bits=128 verify=NO) for ; Thu, 13 Jun 2013 11:52:36 -0600 (MDT) From: Josef Bacik To: Subject: [PATCH] Btrfs: fix not being able to find skinny extents during relocate Date: Thu, 13 Jun 2013 13:52:34 -0400 Message-ID: <1371145954-3070-1-git-send-email-jbacik@fusionio.com> MIME-Version: 1.0 Content-Type: text/plain Sender: linux-btrfs-owner@vger.kernel.org List-ID: We unconditionally search for the EXTENT_ITEM_KEY for metadata during balance, and then check the key that we found to see if it is actually a METADATA_ITEM_KEY, but this doesn't work right because METADATA is a higher key value, so if what we are looking for happens to be the first item in the leaf the search will dump us out at the previous leaf, and we won't find our item. So instead do what we do everywhere else, search for the skinny extent first and if we don't find it go back and re-search for the extent item. This patch fixes the panic I was hitting when balancing a large file system with skinny extents. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/relocation.c | 35 +++++++++++++++++++++++++++-------- 1 files changed, 27 insertions(+), 8 deletions(-) diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 6162d90..df3c632 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -3309,6 +3309,8 @@ static int __add_tree_block(struct reloc_control *rc, struct btrfs_path *path; struct btrfs_key key; int ret; + bool skinny = btrfs_fs_incompat(rc->extent_root->fs_info, + SKINNY_METADATA); if (tree_block_processed(bytenr, blocksize, rc)) return 0; @@ -3319,10 +3321,15 @@ static int __add_tree_block(struct reloc_control *rc, path = btrfs_alloc_path(); if (!path) return -ENOMEM; - +again: key.objectid = bytenr; - key.type = BTRFS_EXTENT_ITEM_KEY; - key.offset = blocksize; + if (skinny) { + key.type = BTRFS_METADATA_ITEM_KEY; + key.offset = (u64)-1; + } else { + key.type = BTRFS_EXTENT_ITEM_KEY; + key.offset = blocksize; + } path->search_commit_root = 1; path->skip_locking = 1; @@ -3330,11 +3337,23 @@ static int __add_tree_block(struct reloc_control *rc, if (ret < 0) goto out; - btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); - if (ret > 0) { - if (key.objectid == bytenr && - key.type == BTRFS_METADATA_ITEM_KEY) - ret = 0; + if (ret > 0 && skinny) { + if (path->slots[0]) { + path->slots[0]--; + btrfs_item_key_to_cpu(path->nodes[0], &key, + path->slots[0]); + if (key.objectid == bytenr && + (key.type == BTRFS_METADATA_ITEM_KEY || + (key.type == BTRFS_EXTENT_ITEM_KEY && + key.offset == blocksize))) + ret = 0; + } + + if (ret) { + skinny = false; + btrfs_release_path(path); + goto again; + } } BUG_ON(ret); -- 1.7.7.6