[lkml]   [2015]   [Feb]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Patch in this message
Subject[PATCH 3.18 18/39] md/raid5: fix another livelock caused by non-aligned writes.
3.18-stable review patch.  If anyone has any objections, please let me know.


From: NeilBrown <>

commit b1b02fe97f75b12ab34b2303bfd4e3526d903a58 upstream.

If a non-page-aligned write is destined for a device which
is missing/faulty, we can deadlock.

As the target device is missing, a read-modify-write cycle
is not possible.
As the write is not for a full-page, a recontruct-write cycle
is not possible.

This should be handled by logic in fetch_block() which notices
there is a non-R5_OVERWRITE write to a missing device, and so
loads all blocks.

However since commit 67f455486d2ea2, that code requires
STRIPE_PREREAD_ACTIVE before it will active, and those circumstances

So: in handle_stripe_dirtying, if neither rmw or rcw was possible,
set STRIPE_DELAYED, which will cause STRIPE_PREREAD_ACTIVE be set
after a suitable delay.

Fixes: 67f455486d2ea20b2d94d6adf5b9b783d079e321
Reported-by: Mikulas Patocka <>
Tested-by: Heinz Mauelshagen <>
Signed-off-by: NeilBrown <>
Signed-off-by: Greg Kroah-Hartman <>

drivers/md/raid5.c | 5 +++++
1 file changed, 5 insertions(+)

--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3195,6 +3195,11 @@ static void handle_stripe_dirtying(struc
(unsigned long long)sh->sector,
rcw, qread, test_bit(STRIPE_DELAYED, &sh->state));
+ if (rcw > disks && rmw > disks &&
+ !test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
+ set_bit(STRIPE_DELAYED, &sh->state);
/* now if nothing is locked, and if we have enough data,
* we can start a write request

 \ /
  Last update: 2015-02-09 09:41    [W:0.122 / U:1.164 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site