[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20190619164500.GC13630@quack2.suse.cz>
Date: Wed, 19 Jun 2019 18:45:00 +0200
From: Jan Kara <jack@...e.cz>
To: "Darrick J. Wong" <darrick.wong@...cle.com>
Cc: Chengguang Xu <cgxu519@...o.com.cn>, jack@...e.com,
linux-ext4@...r.kernel.org
Subject: Re: [PATCH] ext2: add missing brelse() in ext2_iget()
On Sun 16-06-19 09:45:21, Darrick J. Wong wrote:
> On Sun, Jun 16, 2019 at 11:08:01PM +0800, Chengguang Xu wrote:
> > Add missing brelse() on error path of ext2_iget().
> >
> > Signed-off-by: Chengguang Xu <cgxu519@...o.com.cn>
>
> /me wonders if the brelse ought to be moved down to bad_inode so that
> each error branch only has to set @ret and then jump (thereby
> eliminating the possibility of making this mistake again), but for a
> oneliner quick fix I guess it's fine:
>
> Reviewed-by: Darrick J. Wong <darrick.wong@...cle.com>
I've applied the fix and also the cleanup Darrick suggested (attached).
Honza
>
> --D
>
> > ---
> > fs/ext2/inode.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
> > index e474127dd255..fb3611f02051 100644
> > --- a/fs/ext2/inode.c
> > +++ b/fs/ext2/inode.c
> > @@ -1473,6 +1473,7 @@ struct inode *ext2_iget (struct super_block *sb, unsigned long ino)
> > else
> > ei->i_dir_acl = le32_to_cpu(raw_inode->i_dir_acl);
> > if (i_size_read(inode) < 0) {
> > + brelse(bh);
> > ret = -EFSCORRUPTED;
> > goto bad_inode;
> > }
> > --
> > 2.21.0
> >
> >
> >
>
--
Jan Kara <jack@...e.com>
SUSE Labs, CR
View attachment "0001-ext2-Always-brelse-bh-on-failure-in-ext2_iget.patch" of type "text/x-patch" (1881 bytes)
Powered by blists - more mailing lists