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/raid10.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/raid10.c')
| -rw-r--r-- | drivers/md/raid10.c | 19 |
1 files changed, 11 insertions, 8 deletions
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 34facda18e72..8ee5d96e6a2d 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -1766,7 +1766,7 @@ static int raid10_remove_disk(struct mddev *mddev, struct md_rdev *rdev) err = -EBUSY; goto abort; } - /* Only remove faulty devices if recovery + /* Only remove non-faulty devices if recovery * is not possible. */ if (!test_bit(Faulty, &rdev->flags) && @@ -1778,13 +1778,16 @@ static int raid10_remove_disk(struct mddev *mddev, struct md_rdev *rdev) goto abort; } *rdevp = NULL; - synchronize_rcu(); - if (atomic_read(&rdev->nr_pending)) { - /* lost the race, try later */ - err = -EBUSY; - *rdevp = rdev; - goto abort; - } else if (p->replacement) { + if (!test_bit(RemoveSynchronized, &rdev->flags)) { + synchronize_rcu(); + if (atomic_read(&rdev->nr_pending)) { + /* lost the race, try later */ + err = -EBUSY; + *rdevp = rdev; + goto abort; + } + } + if (p->replacement) { /* We must have just cleared 'rdev' */ p->rdev = p->replacement; clear_bit(Replacement, &p->replacement->flags); |