diff mbox

[2/8] ocfs2: flock: drop cross-node lock when failed locally

Message ID 20140319211000.A597C31C25A@corp2gmr1-1.hot.corp.google.com (mailing list archive)
State New, archived
Headers show

Commit Message

Andrew Morton March 19, 2014, 9:10 p.m. UTC 
From: Wengang Wang <wen.gang.wang@oracle.com>
Subject: ocfs2: flock: drop cross-node lock when failed locally

ocfs2_do_flock() calls ocfs2_file_lock() to get the cross-node clock and
then call flock_lock_file_wait() to compete with local processes.  In case
flock_lock_file_wait() failed, say -ENOMEM, clean up work is not done. 
This patch adds the cleanup --drop the cross-node lock which was just
granted.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Wengang Wang <wen.gang.wang@oracle.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 fs/ocfs2/locks.c |    2 ++
 1 file changed, 2 insertions(+)

Comments

Mark Fasheh March 31, 2014, 1:30 a.m. UTC | #1 
On Wed, Mar 19, 2014 at 02:10:00PM -0700, Andrew Morton wrote:
> From: Wengang Wang <wen.gang.wang@oracle.com>
> Subject: ocfs2: flock: drop cross-node lock when failed locally
> 
> ocfs2_do_flock() calls ocfs2_file_lock() to get the cross-node clock and
> then call flock_lock_file_wait() to compete with local processes.  In case
> flock_lock_file_wait() failed, say -ENOMEM, clean up work is not done. 
> This patch adds the cleanup --drop the cross-node lock which was just
> granted.

Out of curiousity was this a bug someone hit, or did you catch this
via code review.

Reviewed-by: Mark Fasheh <mfasheh@suse.de>
	--Mark

--
Mark Fasheh
Wengang Wang March 31, 2014, 1:39 a.m. UTC | #2 
Hi Mark,

I find it by code review.

thanks,
wengang

? 2014?03?31? 09:30, Mark Fasheh ??:
> On Wed, Mar 19, 2014 at 02:10:00PM -0700, Andrew Morton wrote:
>> From: Wengang Wang <wen.gang.wang@oracle.com>
>> Subject: ocfs2: flock: drop cross-node lock when failed locally
>>
>> ocfs2_do_flock() calls ocfs2_file_lock() to get the cross-node clock and
>> then call flock_lock_file_wait() to compete with local processes.  In case
>> flock_lock_file_wait() failed, say -ENOMEM, clean up work is not done.
>> This patch adds the cleanup --drop the cross-node lock which was just
>> granted.
> Out of curiousity was this a bug someone hit, or did you catch this
> via code review.
>
> Reviewed-by: Mark Fasheh <mfasheh@suse.de>
> 	--Mark
>
> --
> Mark Fasheh
>
> _______________________________________________
> Ocfs2-devel mailing list
> Ocfs2-devel@oss.oracle.com
> https://oss.oracle.com/mailman/listinfo/ocfs2-devel
diff mbox

Patch

diff -puN fs/ocfs2/locks.c~ocfs2-flock-drop-cross-node-lock-when-failed-locally fs/ocfs2/locks.c
--- a/fs/ocfs2/locks.c~ocfs2-flock-drop-cross-node-lock-when-failed-locally
+++ a/fs/ocfs2/locks.c
@@ -82,6 +82,8 @@  static int ocfs2_do_flock(struct file *f
 	}
 
 	ret = flock_lock_file_wait(file, fl);
+	if (ret)
+		ocfs2_file_unlock(file);
 
 out:
 	mutex_unlock(&fp->fp_mutex);