[07/15] ocfs2: fix journal commit deadlock
diff mbox

Message ID 548f65dd.p6iIKNzwu5PVSHMt%akpm@linux-foundation.org
State New, archived
Headers show

Commit Message

Andrew Morton Dec. 15, 2014, 10:51 p.m. UTC
From: Junxiao Bi <junxiao.bi@oracle.com>
Subject: ocfs2: fix journal commit deadlock

For buffer write, page lock will be got in write_begin and released in
write_end, in ocfs2_write_end_nolock(), before it unlock the page in
ocfs2_free_write_ctxt(), it calls ocfs2_run_deallocs(), this will ask for
the read lock of journal->j_trans_barrier.  Holding page lock and ask for
journal->j_trans_barrier breaks the locking order.

This will cause a deadlock with journal commit threads, ocfs2cmt will get
write lock of journal->j_trans_barrier first, then it wakes up kjournald2
to do the commit work, at last it waits until done.  To commit journal,
kjournald2 needs flushing data first, it needs get the cache page lock.

Since some ocfs2 cluster locks are holding by write process, this
deadlock may hung the whole cluster.

unlock pages before ocfs2_run_deallocs() can fix the locking order, also
put unlock before ocfs2_commit_trans() to make page lock is unlocked
before j_trans_barrier to preserve unlocking order.

Signed-off-by: Junxiao Bi <junxiao.bi@oracle.com>
Reviewed-by: Wengang Wang <wen.gang.wang@oracle.com>
Cc: <stable@vger.kernel.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 fs/ocfs2/aops.c |   16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

Comments

Mark Fasheh Dec. 17, 2014, 9:01 p.m. UTC | #1
On Mon, Dec 15, 2014 at 02:51:09PM -0800, Andrew Morton wrote:
> From: Junxiao Bi <junxiao.bi@oracle.com>
> Subject: ocfs2: fix journal commit deadlock
> 
> For buffer write, page lock will be got in write_begin and released in
> write_end, in ocfs2_write_end_nolock(), before it unlock the page in
> ocfs2_free_write_ctxt(), it calls ocfs2_run_deallocs(), this will ask for
> the read lock of journal->j_trans_barrier.  Holding page lock and ask for
> journal->j_trans_barrier breaks the locking order.
> 
> This will cause a deadlock with journal commit threads, ocfs2cmt will get
> write lock of journal->j_trans_barrier first, then it wakes up kjournald2
> to do the commit work, at last it waits until done.  To commit journal,
> kjournald2 needs flushing data first, it needs get the cache page lock.
> 
> Since some ocfs2 cluster locks are holding by write process, this
> deadlock may hung the whole cluster.
> 
> unlock pages before ocfs2_run_deallocs() can fix the locking order, also
> put unlock before ocfs2_commit_trans() to make page lock is unlocked
> before j_trans_barrier to preserve unlocking order.
> 
> Signed-off-by: Junxiao Bi <junxiao.bi@oracle.com>
> Reviewed-by: Wengang Wang <wen.gang.wang@oracle.com>
> Cc: <stable@vger.kernel.org>
> Cc: Mark Fasheh <mfasheh@suse.com>
> Cc: Joel Becker <jlbec@evilplan.org>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

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

--
Mark Fasheh

Patch
diff mbox

diff -puN fs/ocfs2/aops.c~ocfs2-fix-journal-commit-deadlock fs/ocfs2/aops.c
--- a/fs/ocfs2/aops.c~ocfs2-fix-journal-commit-deadlock
+++ a/fs/ocfs2/aops.c
@@ -894,7 +894,7 @@  void ocfs2_unlock_and_free_pages(struct
 	}
 }
 
-static void ocfs2_free_write_ctxt(struct ocfs2_write_ctxt *wc)
+static void ocfs2_unlock_pages(struct ocfs2_write_ctxt *wc)
 {
 	int i;
 
@@ -915,7 +915,11 @@  static void ocfs2_free_write_ctxt(struct
 		page_cache_release(wc->w_target_page);
 	}
 	ocfs2_unlock_and_free_pages(wc->w_pages, wc->w_num_pages);
+}
 
+static void ocfs2_free_write_ctxt(struct ocfs2_write_ctxt *wc)
+{
+	ocfs2_unlock_pages(wc);
 	brelse(wc->w_di_bh);
 	kfree(wc);
 }
@@ -2041,11 +2045,19 @@  out_write_size:
 	ocfs2_journal_dirty(handle, wc->w_di_bh);
 
 out:
+	/* unlock pages before dealloc since it needs acquiring j_trans_barrier
+	 * lock, or it will cause a deadlock since journal commit threads holds
+	 * this lock and will ask for the page lock when flushing the data.
+	 * put it here to preserve the unlock order.
+	 */
+	ocfs2_unlock_pages(wc);
+
 	ocfs2_commit_trans(osb, handle);
 
 	ocfs2_run_deallocs(osb, &wc->w_dealloc);
 
-	ocfs2_free_write_ctxt(wc);
+	brelse(wc->w_di_bh);
+	kfree(wc);
 
 	return copied;
 }