[lkml]   [2014]   [Oct]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Patch in this message
Subject[GIT PULL REQUEST] one last-minute update for md/raid5

Hi Linus,
I hope this is in time for 3.17...
It turns out that 'discard_zeroes_data' is not a guarantee, and RAID4/5/6
really depends on it for DISCARD to be handled safely.
So it is safest to default to disabling DISCARD on these arrays until some
sort of whitelist arrangement can be found.


The following changes since commit fe82dcec644244676d55a1384c958d5f67979adb:

Linux 3.17-rc7 (2014-09-28 14:29:07 -0700)

are available in the git repository at:

git:// tags/md/3.17-final-fix

for you to fetch changes up to 8e0e99ba64c7ba46133a7c8a3e3f7de01f23bd93:

md/raid5: disable 'DISCARD' by default due to safety concerns. (2014-10-02 13:45:00 +1000)

One fix for raid5 discard issue.

NeilBrown (1):
md/raid5: disable 'DISCARD' by default due to safety concerns.

drivers/md/raid5.c | 18 +++++++++++++++++-
1 file changed, 17 insertions(+), 1 deletion(-)
commit 8e0e99ba64c7ba46133a7c8a3e3f7de01f23bd93
Author: NeilBrown <>
Date: Thu Oct 2 13:45:00 2014 +1000

md/raid5: disable 'DISCARD' by default due to safety concerns.

It has come to my attention (thanks Martin) that 'discard_zeroes_data'
is only a hint. Some devices in some cases don't do what it
says on the label.

The use of DISCARD in RAID5 depends on reads from discarded regions
being predictably zero. If a write to a previously discarded region
performs a read-modify-write cycle it assumes that the parity block
was consistent with the data blocks. If all were zero, this would
be the case. If some are and some aren't this would not be the case.
This could lead to data corruption after a device failure when
data needs to be reconstructed from the parity.

As we cannot trust 'discard_zeroes_data', ignore it by default
and so disallow DISCARD on all raid4/5/6 arrays.

As many devices are trustworthy, and as there are benefits to using
DISCARD, add a module parameter to over-ride this caution and cause
DISCARD to work if discard_zeroes_data is set.

If a site want to enable DISCARD on some arrays but not on others they
should select DISCARD support at the filesystem level, and set the
raid456 module parameter.

As this is a data-safety issue, I believe this patch is suitable for
DISCARD support for RAID456 was added in 3.7

Cc: Shaohua Li <>
Cc: "Martin K. Petersen" <>
Cc: Mike Snitzer <>
Cc: Heinz Mauelshagen <>
Cc: (3.7+)
Acked-by: Martin K. Petersen <>
Acked-by: Mike Snitzer <>
Fixes: 620125f2bf8ff0c4969b79653b54d7bcc9d40637
Signed-off-by: NeilBrown <>

diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 183588b11fc1..9f0fbecd1eb5 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -64,6 +64,10 @@
#define cpu_to_group(cpu) cpu_to_node(cpu)

+static bool devices_handle_discard_safely = false;
+module_param(devices_handle_discard_safely, bool, 0644);
+ "Set to Y if all devices in each array reliably return zeroes on reads from discarded regions");
static struct workqueue_struct *raid5_wq;
* Stripe cache
@@ -6208,7 +6212,7 @@ static int run(struct mddev *mddev)
mddev->queue->limits.discard_granularity = stripe;
* unaligned part of discard request will be ignored, so can't
- * guarantee discard_zerors_data
+ * guarantee discard_zeroes_data
mddev->queue->limits.discard_zeroes_data = 0;

@@ -6233,6 +6237,18 @@ static int run(struct mddev *mddev)
discard_supported = false;
+ /* Unfortunately, discard_zeroes_data is not currently
+ * a guarantee - just a hint. So we only allow DISCARD
+ * if the sysadmin has confirmed that only safe devices
+ * are in use by setting a module parameter.
+ */
+ if (!devices_handle_discard_safely) {
+ if (discard_supported) {
+ pr_info("md/raid456: discard support disabled due to uncertainty.\n");
+ pr_info("Set raid456.devices_handle_discard_safely=Y to override.\n");
+ }
+ discard_supported = false;
+ }

if (discard_supported &&[unhandled content-type:application/pgp-signature]
 \ /
  Last update: 2014-10-03 08:01    [W:0.047 / U:9.020 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site