Subject: [PATCH] ext4 can fail badly when device stops accepting
	BIO_RW_BARRIER requests.
From: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
To: Neil Brown <neilb@suse.de>
Cc: sct@redhat.com, akpm@linux-foundation.org, adilger@clusterfs.com,
       linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org
In-Reply-To: <18346.24466.83745.944149@notabene.brown>
References: <18346.24466.83745.944149@notabene.brown>
Content-Type: text/plain
Date: Wed, 06 Feb 2008 22:25:21 -0600
Message-Id: <1202358321.6497.5.camel@norville.austin.ibm.com>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2689
Lines: 74

Duplicating Neil Brown's jbd patch for jbd2.  I guess this can go
through the ext4 queue rather than straight into -mm.

Neil's text:

Some devices - notably dm and md - can change their behaviour in
response to BIO_RW_BARRIER requests.  They might start out accepting
such requests but on reconfiguration, they find out that they cannot
any more.

ext3 (and other filesystems) deal with this by always testing if
BIO_RW_BARRIER requests fail with EOPNOTSUPP, and retrying the write
requests without the barrier (probably after waiting for any pending
writes to complete).

However there is a bug in the handling for this for ext3.

When ext3 (jbd actually) decides to submit a BIO_RW_BARRIER request,
it sets the buffer_ordered flag on the buffer head.
If the request completes successfully, the flag STAYS SET.

Other code might then write the same buffer_head after the device has
been reconfigured to not accept barriers.  This write will then fail,
but the "other code" is not ready to handle EOPNOTSUPP errors and the
error will be treated as fatal.

This can be seen without having to reconfigure a device at exactly the
wrong time by putting:

                if (buffer_ordered(bh))
                        printk("OH DEAR, and ordered buffer\n");


in the while loop in "commit phase 5" of journal_commit_transaction.

If it ever prints the "OH DEAR ..." message (as it does sometimes for
me), then that request could (in different circumstances) have failed
with EOPNOTSUPP, but that isn't tested for.

My proposed fix is to clear the buffer_ordered flag after it has been
used, as in the following patch.

Thanks,
NeilBrown

Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>

diff -Nurp linux-2.6.24-mm1/fs/jbd2/commit.c linux/fs/jbd2/commit.c
--- linux-2.6.24-mm1/fs/jbd2/commit.c	2008-02-04 09:08:44.000000000 -0600
+++ linux/fs/jbd2/commit.c	2008-02-06 22:11:14.000000000 -0600
@@ -148,6 +148,8 @@ static int journal_submit_commit_record(
 		barrier_done = 1;
 	}
 	ret = submit_bh(WRITE, bh);
+	if (barrier_done)
+		clear_buffer_ordered(bh);
 
 	/* is it possible for another commit to fail at roughly
 	 * the same time as this one?  If so, we don't want to
@@ -166,7 +168,6 @@ static int journal_submit_commit_record(
 		spin_unlock(&journal->j_state_lock);
 
 		/* And try again, without the barrier */
-		clear_buffer_ordered(bh);
 		set_buffer_uptodate(bh);
 		set_buffer_dirty(bh);
 		ret = submit_bh(WRITE, bh);


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/