[PATCH] md - 9 of 12 - Fix bad interaction between sync checkpointing and recovery
From: NeilBrown <hidden>
Date: 2003-03-12 03:59:37
### Comments for ChangeSet
Md devices (raid1/raid5) can resync or recover.
There are similar but importantly different.
resync happens after an unclean shutdown
recovery happens when a failed drive is being replaced by a hot spare.
The sync-checkpoint code confused the two somewhat and this causes
problems.
This patch makes sure "recovery_cp" only relates to resync,
not recovery.
It also fixes a small problem with recording spares in
the superblock.
----------- Diffstat output ------------
./drivers/md/md.c | 14 +++++++++-----
1 files changed, 9 insertions(+), 5 deletions(-)
diff ./drivers/md/md.c~current~ ./drivers/md/md.c--- ./drivers/md/md.c~current~ 2003-03-12 13:46:15.000000000 +1100
+++ ./drivers/md/md.c 2003-03-12 13:46:25.000000000 +1100@@ -713,7 +713,7 @@ static void super_90_sync(mddev_t *mddev sb->disks[0].state = (1<<MD_DISK_REMOVED); ITERATE_RDEV(mddev,rdev2,tmp) { mdp_disk_t *d; - if (rdev2->raid_disk >= 0) + if (rdev2->raid_disk >= 0 && rdev2->in_sync && !rdev2->faulty) rdev2->desc_nr = rdev2->raid_disk; else rdev2->desc_nr = next_spare++;
@@ -722,7 +722,7 @@ static void super_90_sync(mddev_t *mddev d->number = rdev2->desc_nr; d->major = MAJOR(rdev2->bdev->bd_dev); d->minor = MINOR(rdev2->bdev->bd_dev); - if (rdev2->raid_disk >= 0) + if (rdev2->raid_disk >= 0 && rdev->in_sync && !rdev2->faulty) d->raid_disk = rdev2->raid_disk; else d->raid_disk = rdev2->desc_nr; /* compatibility */
@@ -3020,9 +3020,13 @@ static void md_do_sync(mddev_t *mddev) sysctl_speed_limit_max); is_mddev_idle(mddev); /* this also initializes IO event counters */ + if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) + j = mddev->recovery_cp; + else + j = 0; for (m = 0; m < SYNC_MARKS; m++) { mark[m] = jiffies; - mark_cnt[m] = mddev->recovery_cp; + mark_cnt[m] = j; } last_mark = 0; mddev->resync_mark = mark[last_mark];
@@ -3039,10 +3043,10 @@ static void md_do_sync(mddev_t *mddev) init_waitqueue_head(&mddev->recovery_wait); last_check = 0; - if (mddev->recovery_cp) + if (j) printk(KERN_INFO "md: resuming recovery of md%d from checkpoint.\n", mdidx(mddev)); - for (j = mddev->recovery_cp; j < max_sectors;) { + while (j < max_sectors) { int sectors; sectors = mddev->pers->sync_request(mddev, j, currspeed < sysctl_speed_limit_min);