From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steven Whitehouse Date: Tue, 4 Jul 2017 10:37:19 +0100 Subject: [Cluster-devel] [GFS2 PATCH] GFS2: Prevent double brelse in gfs2_meta_indirect_buffer In-Reply-To: <1874168289.28426650.1499100419665.JavaMail.zimbra@redhat.com> References: <1874168289.28426650.1499100419665.JavaMail.zimbra@redhat.com> Message-ID: List-Id: To: cluster-devel.redhat.com MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Acked-by: Steven Whitehouse Steve. On 03/07/17 17:46, Bob Peterson wrote: > Hi, > > Before this patch, problems reading in indirect buffers would send > an IO error back to the caller, and release the buffer_head with > brelse() in function gfs2_meta_indirect_buffer, however, it would > still return the address of the buffer_head it released. After the > error was discovered, function gfs2_block_map would call function > release_metapath to free all buffers. That checked: > if (mp->mp_bh[i] == NULL) but since the value was set after the > error, it was non-zero, so brelse was called a second time. This > resulted in the following error: > > kernel: WARNING: at fs/buffer.c:1224 __brelse+0x3a/0x40() > kernel: VFS: brelse: Trying to free free buffer > > This patch changes gfs2_meta_indirect_buffer so it only sets > the buffer_head pointer in cases where it isn't released. > > Signed-off-by: Bob Peterson > --- > diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c > index 663ffc1..c7d2c76 100644 > --- a/fs/gfs2/meta_io.c > +++ b/fs/gfs2/meta_io.c > @@ -419,8 +419,9 @@ int gfs2_meta_indirect_buffer(struct gfs2_inode *ip, int height, u64 num, > if (ret == 0 && gfs2_metatype_check(sdp, bh, mtype)) { > brelse(bh); > ret = -EIO; > + } else { > + *bhp = bh; > } > - *bhp = bh; > return ret; > } > >