From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from aserp1040.oracle.com ([141.146.126.69]:30707 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S935429AbcIVSD4 (ORCPT ); Thu, 22 Sep 2016 14:03:56 -0400 Date: Thu, 22 Sep 2016 11:03:45 -0700 From: Liu Bo To: David Sterba Cc: linux-btrfs@vger.kernel.org, Josef Bacik Subject: Re: [PATCH] Btrfs: don't BUG() during drop snapshot Message-ID: <20160922180345.GB12906@localhost.localdomain> Reply-To: bo.li.liu@oracle.com References: <1474556904-22495-1-git-send-email-dsterba@suse.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1474556904-22495-1-git-send-email-dsterba@suse.com> Sender: linux-btrfs-owner@vger.kernel.org List-ID: On Thu, Sep 22, 2016 at 05:08:24PM +0200, David Sterba wrote: > From: Josef Bacik > > Really there's lots of things that can go wrong here, kill all the > BUG_ON()'s and replace the logic ones with ASSERT()'s and return EIO > instead. > > Signed-off-by: Josef Bacik > Signed-off-by: David Sterba > --- > fs/btrfs/extent-tree.c | 27 +++++++++++++++++++++++---- > 1 file changed, 23 insertions(+), 4 deletions(-) > > diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c > index c95f85c292a4..37aba7d00c8f 100644 > --- a/fs/btrfs/extent-tree.c > +++ b/fs/btrfs/extent-tree.c > @@ -8884,12 +8884,15 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, > &wc->flags[level - 1]); > if (ret < 0) { > btrfs_tree_unlock(next); > + free_extent_buffer(next); > return ret; > } > > if (unlikely(wc->refs[level - 1] == 0)) { > btrfs_err(root->fs_info, "Missing references."); > - BUG(); > + btrfs_tree_unlock(next); > + free_extent_buffer(next); > + return -EIO; > } > *lookup_info = 0; > > @@ -8941,7 +8944,13 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, > } > > level--; > - BUG_ON(level != btrfs_header_level(next)); > + ASSERT(level == btrfs_header_level(next)); > + if (level != btrfs_header_level(next)) { > + btrfs_err("mismatched level"); > + btrfs_tree_unlock(next); > + free_extent_buffer(next); > + return -EIO; > + } > path->nodes[level] = next; > path->slots[level] = 0; > path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING; > @@ -8956,8 +8965,14 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, > if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) { > parent = path->nodes[level]->start; > } else { > - BUG_ON(root->root_key.objectid != > + ASSERT(root->root_key.objectid == > btrfs_header_owner(path->nodes[level])); > + if (root->root_key.objectid != > + btrfs_header_owner(path->nodes[level])) { > + btrfs_err("mismatched block owner"); > + btrfs_tree_unlock(next); > + free_extent_buffer(next); We need a 'return' here, otherwise the next 'if (need_account)' may use @next after free. > + } > parent = 0; > } > > @@ -8972,7 +8987,11 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, > } > ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent, > root->root_key.objectid, level - 1, 0); > - BUG_ON(ret); /* -ENOMEM */ > + if (ret) { > + btrfs_tree_unlock(next); > + free_extent_buffer(next); > + return ret; > + } > } Can we put a label of "out" here since most of the above cleanups use the same way to bail out? Thanks, -liubo > btrfs_tree_unlock(next); > free_extent_buffer(next); > -- > 2.7.1 > > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html