Instead of magic numbers (0,1,2,3) in sb_dirty, we have
some flags instead:
MD_CHANGE_DEVS
Some device state has changed requiring superblock update
on all devices.
MD_CHANGE_CLEAN
The array has transitions from 'clean' to 'dirty' or back,
requiring a superblock update on active devices, but possibly
not on spares
MD_CHANGE_PENDING
A superblock update is underway.
We wait for an update to complete by waiting for all flags to be clear. A
flag can be set at any time, even during an update, without risk that the
change will be lost.
Stop exporting md_update_sb - isn't needed.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
-void md_update_sb(mddev_t * mddev)
+static void md_update_sb(mddev_t * mddev, int force_change)
{
int err;
struct list_head *tmp;
{
int err;
struct list_head *tmp;
repeat:
spin_lock_irq(&mddev->write_lock);
repeat:
spin_lock_irq(&mddev->write_lock);
- if (mddev->degraded && mddev->sb_dirty == 3)
+ set_bit(MD_CHANGE_PENDING, &mddev->flags);
+ if (test_and_clear_bit(MD_CHANGE_DEVS, &mddev->flags))
+ force_change = 1;
+ if (test_and_clear_bit(MD_CHANGE_CLEAN, &mddev->flags))
+ /* just a clean<-> dirty transition, possibly leave spares alone,
+ * though if events isn't the right even/odd, we will have to do
+ * spares after all
+ */
+ nospares = 1;
+ if (force_change)
+ nospares = 0;
+ if (mddev->degraded)
/* If the array is degraded, then skipping spares is both
* dangerous and fairly pointless.
* Dangerous because a device that was removed from the array
/* If the array is degraded, then skipping spares is both
* dangerous and fairly pointless.
* Dangerous because a device that was removed from the array
* then a recovery will happen and soon that array won't
* be degraded any more and the spare can go back to sleep then.
*/
* then a recovery will happen and soon that array won't
* be degraded any more and the spare can go back to sleep then.
*/
sync_req = mddev->in_sync;
mddev->utime = get_seconds();
sync_req = mddev->in_sync;
mddev->utime = get_seconds();
- if (mddev->sb_dirty == 3)
- /* just a clean<-> dirty transition, possibly leave spares alone,
- * though if events isn't the right even/odd, we will have to do
- * spares after all
- */
- nospares = 1;
/* If this is just a dirty<->clean transition, and the array is clean
* and 'events' is odd, we can roll back to the previous clean state */
/* If this is just a dirty<->clean transition, and the array is clean
* and 'events' is odd, we can roll back to the previous clean state */
- if (mddev->sb_dirty == 3
&& (mddev->in_sync && mddev->recovery_cp == MaxSector)
&& (mddev->events & 1))
mddev->events--;
&& (mddev->in_sync && mddev->recovery_cp == MaxSector)
&& (mddev->events & 1))
mddev->events--;
MD_BUG();
mddev->events --;
}
MD_BUG();
mddev->events --;
}
sync_sbs(mddev, nospares);
/*
sync_sbs(mddev, nospares);
/*
* nonpersistent superblocks
*/
if (!mddev->persistent) {
* nonpersistent superblocks
*/
if (!mddev->persistent) {
+ clear_bit(MD_CHANGE_PENDING, &mddev->flags);
spin_unlock_irq(&mddev->write_lock);
wake_up(&mddev->sb_wait);
return;
spin_unlock_irq(&mddev->write_lock);
wake_up(&mddev->sb_wait);
return;
break;
}
md_super_wait(mddev);
break;
}
md_super_wait(mddev);
- /* if there was a failure, sb_dirty was set to 1, and we re-write super */
+ /* if there was a failure, MD_CHANGE_DEVS was set, and we re-write super */
spin_lock_irq(&mddev->write_lock);
spin_lock_irq(&mddev->write_lock);
- if (mddev->in_sync != sync_req|| mddev->sb_dirty == 1) {
+ if (mddev->in_sync != sync_req ||
+ test_bit(MD_CHANGE_DEVS, &mddev->flags)) {
/* have to write it out again */
spin_unlock_irq(&mddev->write_lock);
goto repeat;
}
/* have to write it out again */
spin_unlock_irq(&mddev->write_lock);
goto repeat;
}
+ clear_bit(MD_CHANGE_PENDING, &mddev->flags);
spin_unlock_irq(&mddev->write_lock);
wake_up(&mddev->sb_wait);
}
spin_unlock_irq(&mddev->write_lock);
wake_up(&mddev->sb_wait);
}
-EXPORT_SYMBOL_GPL(md_update_sb);
/* words written to sysfs files may, or my not, be \n terminated.
* We want to accept with case. For this we use cmd_match.
/* words written to sysfs files may, or my not, be \n terminated.
* We want to accept with case. For this we use cmd_match.
else {
mddev_t *mddev = rdev->mddev;
kick_rdev_from_array(rdev);
else {
mddev_t *mddev = rdev->mddev;
kick_rdev_from_array(rdev);
+ md_update_sb(mddev, 1);
md_new_event(mddev);
err = 0;
}
md_new_event(mddev);
err = 0;
}
spin_lock_irq(&mddev->write_lock);
if (atomic_read(&mddev->writes_pending) == 0) {
mddev->in_sync = 1;
spin_lock_irq(&mddev->write_lock);
if (atomic_read(&mddev->writes_pending) == 0) {
mddev->in_sync = 1;
+ set_bit(MD_CHANGE_CLEAN, &mddev->flags);
}
spin_unlock_irq(&mddev->write_lock);
} else {
}
spin_unlock_irq(&mddev->write_lock);
} else {
case active:
if (mddev->pers) {
restart_array(mddev);
case active:
if (mddev->pers) {
restart_array(mddev);
+ clear_bit(MD_CHANGE_CLEAN, &mddev->flags);
wake_up(&mddev->sb_wait);
err = 0;
} else {
wake_up(&mddev->sb_wait);
err = 0;
} else {
if (mddev->pers) {
err = update_size(mddev, size);
if (mddev->pers) {
err = update_size(mddev, size);
+ md_update_sb(mddev, 1);
} else {
if (mddev->size == 0 ||
mddev->size > size)
} else {
if (mddev->size == 0 ||
mddev->size > size)
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
- if (mddev->sb_dirty)
- md_update_sb(mddev);
+ if (mddev->flags)
+ md_update_sb(mddev, 0);
set_capacity(disk, mddev->array_size<<1);
set_capacity(disk, mddev->array_size<<1);
if (mddev->ro)
mddev->ro = 0;
}
if (mddev->ro)
mddev->ro = 0;
}
- if (!mddev->in_sync || mddev->sb_dirty) {
+ if (!mddev->in_sync || mddev->flags) {
/* mark array as shutdown cleanly */
mddev->in_sync = 1;
/* mark array as shutdown cleanly */
mddev->in_sync = 1;
+ md_update_sb(mddev, 1);
}
if (mode == 1)
set_disk_ro(disk, 1);
}
if (mode == 1)
set_disk_ro(disk, 1);
goto busy;
kick_rdev_from_array(rdev);
goto busy;
kick_rdev_from_array(rdev);
+ md_update_sb(mddev, 1);
md_new_event(mddev);
return 0;
md_new_event(mddev);
return 0;
+ md_update_sb(mddev, 1);
/*
* Kick recovery, maybe this spare has to be added to the
/*
* Kick recovery, maybe this spare has to be added to the
mddev->max_disks = MD_SB_DISKS;
mddev->max_disks = MD_SB_DISKS;
+ mddev->flags = 0;
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
mddev->default_bitmap_offset = MD_SB_BYTES >> 9;
mddev->bitmap_offset = 0;
mddev->default_bitmap_offset = MD_SB_BYTES >> 9;
mddev->bitmap_offset = 0;
mddev->bitmap_offset = 0;
}
}
mddev->bitmap_offset = 0;
}
}
+ md_update_sb(mddev, 1);
spin_lock_irq(&mddev->write_lock);
if (mddev->in_sync) {
mddev->in_sync = 0;
spin_lock_irq(&mddev->write_lock);
if (mddev->in_sync) {
mddev->in_sync = 0;
+ set_bit(MD_CHANGE_CLEAN, &mddev->flags);
md_wakeup_thread(mddev->thread);
}
spin_unlock_irq(&mddev->write_lock);
}
md_wakeup_thread(mddev->thread);
}
spin_unlock_irq(&mddev->write_lock);
}
- wait_event(mddev->sb_wait, mddev->sb_dirty==0);
+ wait_event(mddev->sb_wait, mddev->flags==0);
}
void md_write_end(mddev_t *mddev)
}
void md_write_end(mddev_t *mddev)
!test_bit(In_sync, &rdev->flags) &&
rdev->recovery_offset < mddev->curr_resync)
rdev->recovery_offset = mddev->curr_resync;
!test_bit(In_sync, &rdev->flags) &&
rdev->recovery_offset < mddev->curr_resync)
rdev->recovery_offset = mddev->curr_resync;
test_bit(MD_RECOVERY_NEEDED, &mddev->recovery) ||
test_bit(MD_RECOVERY_DONE, &mddev->recovery) ||
(mddev->safemode == 1) ||
test_bit(MD_RECOVERY_NEEDED, &mddev->recovery) ||
test_bit(MD_RECOVERY_DONE, &mddev->recovery) ||
(mddev->safemode == 1) ||
if (mddev->safemode && !atomic_read(&mddev->writes_pending) &&
!mddev->in_sync && mddev->recovery_cp == MaxSector) {
mddev->in_sync = 1;
if (mddev->safemode && !atomic_read(&mddev->writes_pending) &&
!mddev->in_sync && mddev->recovery_cp == MaxSector) {
mddev->in_sync = 1;
+ set_bit(MD_CHANGE_CLEAN, &mddev->flags);
}
if (mddev->safemode == 1)
mddev->safemode = 0;
spin_unlock_irq(&mddev->write_lock);
}
if (mddev->safemode == 1)
mddev->safemode = 0;
spin_unlock_irq(&mddev->write_lock);
- if (mddev->sb_dirty)
- md_update_sb(mddev);
+ if (mddev->flags)
+ md_update_sb(mddev, 0);
if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) &&
if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) &&
/* activate any spares */
mddev->pers->spare_active(mddev);
}
/* activate any spares */
mddev->pers->spare_active(mddev);
}
+ md_update_sb(mddev, 1);
/* if array is no-longer degraded, then any saved_raid_disk
* information must be scrapped
/* if array is no-longer degraded, then any saved_raid_disk
* information must be scrapped
char b[BDEVNAME_SIZE];
clear_bit(In_sync, &rdev->flags);
set_bit(Faulty, &rdev->flags);
char b[BDEVNAME_SIZE];
clear_bit(In_sync, &rdev->flags);
set_bit(Faulty, &rdev->flags);
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
conf->working_disks--;
printk(KERN_ALERT "multipath: IO failure on %s,"
" disabling IO path. \n Operation continuing"
conf->working_disks--;
printk(KERN_ALERT "multipath: IO failure on %s,"
" disabling IO path. \n Operation continuing"
}
conf->raid_disks = mddev->raid_disks;
}
conf->raid_disks = mddev->raid_disks;
conf->mddev = mddev;
spin_lock_init(&conf->device_lock);
INIT_LIST_HEAD(&conf->retry_list);
conf->mddev = mddev;
spin_lock_init(&conf->device_lock);
INIT_LIST_HEAD(&conf->retry_list);
}
clear_bit(In_sync, &rdev->flags);
set_bit(Faulty, &rdev->flags);
}
clear_bit(In_sync, &rdev->flags);
set_bit(Faulty, &rdev->flags);
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
printk(KERN_ALERT "raid1: Disk failure on %s, disabling device. \n"
" Operation continuing on %d devices\n",
bdevname(rdev->bdev,b), conf->working_disks);
printk(KERN_ALERT "raid1: Disk failure on %s, disabling device. \n"
" Operation continuing on %d devices\n",
bdevname(rdev->bdev,b), conf->working_disks);
}
clear_bit(In_sync, &rdev->flags);
set_bit(Faulty, &rdev->flags);
}
clear_bit(In_sync, &rdev->flags);
set_bit(Faulty, &rdev->flags);
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
printk(KERN_ALERT "raid10: Disk failure on %s, disabling device. \n"
" Operation continuing on %d devices\n",
bdevname(rdev->bdev,b), conf->working_disks);
printk(KERN_ALERT "raid10: Disk failure on %s, disabling device. \n"
" Operation continuing on %d devices\n",
bdevname(rdev->bdev,b), conf->working_disks);
PRINTK("raid5: error called\n");
if (!test_bit(Faulty, &rdev->flags)) {
PRINTK("raid5: error called\n");
if (!test_bit(Faulty, &rdev->flags)) {
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
if (test_bit(In_sync, &rdev->flags)) {
conf->working_disks--;
mddev->degraded++;
if (test_bit(In_sync, &rdev->flags)) {
conf->working_disks--;
mddev->degraded++;
wait_event(conf->wait_for_overlap,
atomic_read(&conf->reshape_stripes)==0);
mddev->reshape_position = conf->expand_progress;
wait_event(conf->wait_for_overlap,
atomic_read(&conf->reshape_stripes)==0);
mddev->reshape_position = conf->expand_progress;
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
md_wakeup_thread(mddev->thread);
md_wakeup_thread(mddev->thread);
- wait_event(mddev->sb_wait, mddev->sb_dirty == 0 ||
+ wait_event(mddev->sb_wait, mddev->flags == 0 ||
kthread_should_stop());
spin_lock_irq(&conf->device_lock);
conf->expand_lo = mddev->reshape_position;
kthread_should_stop());
spin_lock_irq(&conf->device_lock);
conf->expand_lo = mddev->reshape_position;
mddev->degraded = (conf->raid_disks - conf->previous_raid_disks) - added_devices;
mddev->raid_disks = conf->raid_disks;
mddev->reshape_position = 0;
mddev->degraded = (conf->raid_disks - conf->previous_raid_disks) - added_devices;
mddev->raid_disks = conf->raid_disks;
mddev->reshape_position = 0;
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
clear_bit(MD_RECOVERY_CHECK, &mddev->recovery);
clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
clear_bit(MD_RECOVERY_CHECK, &mddev->recovery);
extern void md_do_sync(mddev_t *mddev);
extern void md_new_event(mddev_t *mddev);
extern void md_do_sync(mddev_t *mddev);
extern void md_new_event(mddev_t *mddev);
-extern void md_update_sb(mddev_t * mddev);
#endif /* CONFIG_MD */
#endif
#endif /* CONFIG_MD */
#endif
dev_t unit;
int md_minor;
struct list_head disks;
dev_t unit;
int md_minor;
struct list_head disks;
+ unsigned long flags;
+#define MD_CHANGE_DEVS 0 /* Some device status has changed */
+#define MD_CHANGE_CLEAN 1 /* transition to or from 'clean' */
+#define MD_CHANGE_PENDING 2 /* superblock update in progress */
+
int ro;
struct gendisk *gendisk;
int ro;
struct gendisk *gendisk;