diff options
| author | NeilBrown <neilb@suse.com> | 2016-06-02 16:19:53 +1000 |
|---|---|---|
| committer | Shaohua Li <shli@fb.com> | 2016-06-13 11:54:22 -0700 |
| commit | d787be4092e27728cb4c012bee9762098ef3c662 (patch) | |
| tree | b8de57ed842d3c01f6fdd4f6ee5be6408763a993 /drivers/md/md.c | |
| parent | f5b67ae86ee317db20c0e10d54f16a0bbbd3207d (diff) | |
md: reduce the number of synchronize_rcu() calls when multiple devices fail.
Every time a device is removed with ->hot_remove_disk() a synchronize_rcu() call is made
which can delay several milliseconds in some case.
If lots of devices fail at once - as could happen with a large RAID10 where one set
of devices are removed all at once - these delays can add up to be very inconcenient.
As failure is not reversible we can check for that first, setting a
separate flag if it is found, and then all synchronize_rcu() once for
all the flagged devices. Then ->hot_remove_disk() function can skip the
synchronize_rcu() step if the flag is set.
fix build error(Shaohua)
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Diffstat (limited to 'drivers/md/md.c')
| -rw-r--r-- | drivers/md/md.c | 29 |
1 files changed, 26 insertions, 3 deletions
diff --git a/drivers/md/md.c b/drivers/md/md.c index 0793754eeffd..2ed547f5c3b6 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -8180,15 +8180,34 @@ static int remove_and_add_spares(struct mddev *mddev, struct md_rdev *rdev; int spares = 0; int removed = 0; + bool remove_some = false; - rdev_for_each(rdev, mddev) + rdev_for_each(rdev, mddev) { + if ((this == NULL || rdev == this) && + rdev->raid_disk >= 0 && + !test_bit(Blocked, &rdev->flags) && + test_bit(Faulty, &rdev->flags) && + atomic_read(&rdev->nr_pending)==0) { + /* Faulty non-Blocked devices with nr_pending == 0 + * never get nr_pending incremented, + * never get Faulty cleared, and never get Blocked set. + * So we can synchronize_rcu now rather than once per device + */ + remove_some = true; + set_bit(RemoveSynchronized, &rdev->flags); + } + } + + if (remove_some) + synchronize_rcu(); + rdev_for_each(rdev, mddev) { if ((this == NULL || rdev == this) && rdev->raid_disk >= 0 && !test_bit(Blocked, &rdev->flags) && - (test_bit(Faulty, &rdev->flags) || + ((test_bit(RemoveSynchronized, &rdev->flags) || (!test_bit(In_sync, &rdev->flags) && !test_bit(Journal, &rdev->flags))) && - atomic_read(&rdev->nr_pending)==0) { + atomic_read(&rdev->nr_pending)==0)) { if (mddev->pers->hot_remove_disk( mddev, rdev) == 0) { sysfs_unlink_rdev(mddev, rdev); @@ -8196,6 +8215,10 @@ static int remove_and_add_spares(struct mddev *mddev, removed++; } } + if (remove_some && test_bit(RemoveSynchronized, &rdev->flags)) + clear_bit(RemoveSynchronized, &rdev->flags); + } + if (removed && mddev->kobj.sd) sysfs_notify(&mddev->kobj, NULL, "degraded"); |