From: Sebastian Riemer <sebastian.rie...@profitbricks.com>

commit bbfa57c0f2243a7c31fd248d22e9861a2802cad5 upstream.

If an fsync occurs on a read-only array, we need to send a
completion for the IO and may not increment the active IO count.
Otherwise, we hit a bug trace and can't stop the MD array anymore.

By advice of Christoph Hellwig we return success upon a flush
request but we return -EROFS for other writes.
We detect flush requests by checking if the bio has zero sectors.

This patch is suitable to any -stable kernel to which it applies.

Cc: Christoph Hellwig <h...@infradead.org>
Cc: Ben Hutchings <b...@decadent.org.uk>
Cc: NeilBrown <ne...@suse.de>
Signed-off-by: Sebastian Riemer <sebastian.rie...@profitbricks.com>
Reported-by: Ben Hutchings <b...@decadent.org.uk>
Acked-by: Paul Menzel <paulepan...@users.sourceforge.net>
Signed-off-by: NeilBrown <ne...@suse.de>
[bwh: Backported to 3.0: md_make_request() returns 0 on error]
Signed-off-by: Ben Hutchings <b...@decadent.org.uk>
---
I noticed you didn't apply this to 3.0.y.  It looks like a simple enough
change - but this is compile-tested only.

Ben.

 drivers/md/md.c |    4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 3db3d1b..1e634a6 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -299,6 +299,10 @@ static void md_make_request(struct request_queue *q, 
struct bio *bio)
                bio_io_error(bio);
                return 0;
        }
+       if (mddev->ro == 1 && unlikely(rw == WRITE)) {
+               bio_endio(bio, bio_sectors(bio) == 0 ? 0 : -EROFS);
+               return 0;
+       }
        smp_rmb(); /* Ensure implications of  'active' are visible */
        rcu_read_lock();
        if (mddev->suspended) {

-- 
Ben Hutchings
Humans are not rational beings; they are rationalising beings.

Attachment: signature.asc
Description: This is a digitally signed message part

Reply via email to