summaryrefslogtreecommitdiffstats
path: root/drivers/md/raid5.c
diff options
context:
space:
mode:
authorNeilBrown <neilb@suse.de>2011-12-23 10:17:50 +1100
committerNeilBrown <neilb@suse.de>2011-12-23 10:17:50 +1100
commit908f4fbd265733310c17ecc906299846b5dac44a (patch)
tree71e364c57887213431143e6fd1296e8433cc6293 /drivers/md/raid5.c
parent2e61ebbcc45438899235d7a39f17291cb24e746c (diff)
downloadop-kernel-dev-908f4fbd265733310c17ecc906299846b5dac44a.zip
op-kernel-dev-908f4fbd265733310c17ecc906299846b5dac44a.tar.gz
md/raid5: be more thorough in calculating 'degraded' value.
When an array is being reshaped to change the number of devices, the two halves can be differently degraded. e.g. one could be missing a device and the other not. So we need to be more careful about calculating the 'degraded' attribute. Instead of just inc/dec at appropriate times, perform a full re-calculation examining both possible cases. This doesn't happen often so it not a big cost, and we already have most of the code to do it. Signed-off-by: NeilBrown <neilb@suse.de>
Diffstat (limited to 'drivers/md/raid5.c')
-rw-r--r--drivers/md/raid5.c63
1 files changed, 33 insertions, 30 deletions
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 858fdbb..a68c492 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -370,12 +370,10 @@ static struct stripe_head *__find_stripe(struct r5conf *conf, sector_t sector,
* of the two sections, and some non-in_sync devices may
* be insync in the section most affected by failed devices.
*/
-static int has_failed(struct r5conf *conf)
+static int calc_degraded(struct r5conf *conf)
{
- int degraded;
+ int degraded, degraded2;
int i;
- if (conf->mddev->reshape_position == MaxSector)
- return conf->mddev->degraded > conf->max_degraded;
rcu_read_lock();
degraded = 0;
@@ -399,14 +397,14 @@ static int has_failed(struct r5conf *conf)
degraded++;
}
rcu_read_unlock();
- if (degraded > conf->max_degraded)
- return 1;
+ if (conf->raid_disks == conf->previous_raid_disks)
+ return degraded;
rcu_read_lock();
- degraded = 0;
+ degraded2 = 0;
for (i = 0; i < conf->raid_disks; i++) {
struct md_rdev *rdev = rcu_dereference(conf->disks[i].rdev);
if (!rdev || test_bit(Faulty, &rdev->flags))
- degraded++;
+ degraded2++;
else if (test_bit(In_sync, &rdev->flags))
;
else
@@ -416,9 +414,22 @@ static int has_failed(struct r5conf *conf)
* almost certainly hasn't.
*/
if (conf->raid_disks <= conf->previous_raid_disks)
- degraded++;
+ degraded2++;
}
rcu_read_unlock();
+ if (degraded2 > degraded)
+ return degraded2;
+ return degraded;
+}
+
+static int has_failed(struct r5conf *conf)
+{
+ int degraded;
+
+ if (conf->mddev->reshape_position == MaxSector)
+ return conf->mddev->degraded > conf->max_degraded;
+
+ degraded = calc_degraded(conf);
if (degraded > conf->max_degraded)
return 1;
return 0;
@@ -1724,18 +1735,15 @@ static void error(struct mddev *mddev, struct md_rdev *rdev)
{
char b[BDEVNAME_SIZE];
struct r5conf *conf = mddev->private;
+ unsigned long flags;
pr_debug("raid456: error called\n");
- if (test_and_clear_bit(In_sync, &rdev->flags)) {
- unsigned long flags;
- spin_lock_irqsave(&conf->device_lock, flags);
- mddev->degraded++;
- spin_unlock_irqrestore(&conf->device_lock, flags);
- /*
- * if recovery was running, make sure it aborts.
- */
- set_bit(MD_RECOVERY_INTR, &mddev->recovery);
- }
+ spin_lock_irqsave(&conf->device_lock, flags);
+ clear_bit(In_sync, &rdev->flags);
+ mddev->degraded = calc_degraded(conf);
+ spin_unlock_irqrestore(&conf->device_lock, flags);
+ set_bit(MD_RECOVERY_INTR, &mddev->recovery);
+
set_bit(Blocked, &rdev->flags);
set_bit(Faulty, &rdev->flags);
set_bit(MD_CHANGE_DEVS, &mddev->flags);
@@ -4852,8 +4860,7 @@ static int run(struct mddev *mddev)
dirty_parity_disks++;
}
- mddev->degraded = (max(conf->raid_disks, conf->previous_raid_disks)
- - working_disks);
+ mddev->degraded = calc_degraded(conf);
if (has_failed(conf)) {
printk(KERN_ERR "md/raid:%s: not enough operational devices"
@@ -5025,7 +5032,7 @@ static int raid5_spare_active(struct mddev *mddev)
}
}
spin_lock_irqsave(&conf->device_lock, flags);
- mddev->degraded -= count;
+ mddev->degraded = calc_degraded(conf);
spin_unlock_irqrestore(&conf->device_lock, flags);
print_raid5_conf(conf);
return count;
@@ -5286,8 +5293,7 @@ static int raid5_start_reshape(struct mddev *mddev)
* pre and post number of devices.
*/
spin_lock_irqsave(&conf->device_lock, flags);
- mddev->degraded += (conf->raid_disks - conf->previous_raid_disks)
- - added_devices;
+ mddev->degraded = calc_degraded(conf);
spin_unlock_irqrestore(&conf->device_lock, flags);
}
mddev->raid_disks = conf->raid_disks;
@@ -5356,12 +5362,9 @@ static void raid5_finish_reshape(struct mddev *mddev)
revalidate_disk(mddev->gendisk);
} else {
int d;
- mddev->degraded = conf->raid_disks;
- for (d = 0; d < conf->raid_disks ; d++)
- if (conf->disks[d].rdev &&
- test_bit(In_sync,
- &conf->disks[d].rdev->flags))
- mddev->degraded--;
+ spin_lock_irq(&conf->device_lock);
+ mddev->degraded = calc_degraded(conf);
+ spin_unlock_irq(&conf->device_lock);
for (d = conf->raid_disks ;
d < conf->raid_disks - mddev->delta_disks;
d++) {
OpenPOWER on IntegriCloud