You should have received a copy of the GNU General Public License
(for example /usr/src/linux/COPYING); if not, write to the Free
Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+
+ Errors, Warnings, etc.
+ Please use:
+ pr_crit() for error conditions that risk data loss
+ pr_err() for error conditions that are unexpected, like an IO error
+ or internal inconsistency
+ pr_warn() for error conditions that could have been predicated, like
+ adding a device to an array when it has incompatible metadata
+ pr_info() for every interesting, very rare events, like an array starting
+ or stopping, or resync starting or stopping
+ pr_debug() for everything else.
+
*/
#include <linux/kthread.h>
#include <linux/raid/md_p.h>
#include <linux/raid/md_u.h>
#include <linux/slab.h>
+#include <trace/events/block.h>
#include "md.h"
#include "bitmap.h"
#include "md-cluster.h"
bi->bi_end_io = md_end_flush;
bi->bi_private = rdev;
bi->bi_bdev = rdev->bdev;
- bio_set_op_attrs(bi, REQ_OP_WRITE, WRITE_FLUSH);
+ bi->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH;
atomic_inc(&mddev->flush_pending);
submit_bio(bi);
rcu_read_lock();
static int alloc_disk_sb(struct md_rdev *rdev)
{
rdev->sb_page = alloc_page(GFP_KERNEL);
- if (!rdev->sb_page) {
- printk(KERN_ALERT "md: out of memory.\n");
+ if (!rdev->sb_page)
return -ENOMEM;
- }
-
return 0;
}
struct mddev *mddev = rdev->mddev;
if (bio->bi_error) {
- printk("md: super_written gets error=%d\n", bio->bi_error);
+ pr_err("md: super_written gets error=%d\n", bio->bi_error);
md_error(mddev, rdev);
- }
+ if (!test_bit(Faulty, &rdev->flags)
+ && (bio->bi_opf & MD_FAILFAST)) {
+ set_bit(MD_SB_NEED_REWRITE, &mddev->sb_flags);
+ set_bit(LastDev, &rdev->flags);
+ }
+ } else
+ clear_bit(LastDev, &rdev->flags);
if (atomic_dec_and_test(&mddev->pending_writes))
wake_up(&mddev->sb_wait);
* if zero is reached.
* If an error occurred, call md_error
*/
- struct bio *bio = bio_alloc_mddev(GFP_NOIO, 1, mddev);
+ struct bio *bio;
+ int ff = 0;
+
+ if (test_bit(Faulty, &rdev->flags))
+ return;
+
+ bio = bio_alloc_mddev(GFP_NOIO, 1, mddev);
atomic_inc(&rdev->nr_pending);
bio_add_page(bio, page, size, 0);
bio->bi_private = rdev;
bio->bi_end_io = super_written;
- bio_set_op_attrs(bio, REQ_OP_WRITE, WRITE_FLUSH_FUA);
+
+ if (test_bit(MD_FAILFAST_SUPPORTED, &mddev->flags) &&
+ test_bit(FailFast, &rdev->flags) &&
+ !test_bit(LastDev, &rdev->flags))
+ ff = MD_FAILFAST;
+ bio->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH | REQ_FUA | ff;
atomic_inc(&mddev->pending_writes);
submit_bio(bio);
}
-void md_super_wait(struct mddev *mddev)
+int md_super_wait(struct mddev *mddev)
{
/* wait for all superblock writes that were scheduled to complete */
wait_event(mddev->sb_wait, atomic_read(&mddev->pending_writes)==0);
+ if (test_and_clear_bit(MD_SB_NEED_REWRITE, &mddev->sb_flags))
+ return -EAGAIN;
+ return 0;
}
int sync_page_io(struct md_rdev *rdev, sector_t sector, int size,
return 0;
fail:
- printk(KERN_WARNING "md: disabled device %s, could not read superblock.\n",
- bdevname(rdev->bdev,b));
+ pr_err("md: disabled device %s, could not read superblock.\n",
+ bdevname(rdev->bdev,b));
return -EINVAL;
}
if (!tmp1 || !tmp2) {
ret = 0;
- printk(KERN_INFO "md.c sb_equal(): failed to allocate memory!\n");
goto abort;
}
{
if (!mddev->bitmap_info.file && !mddev->bitmap_info.offset)
return 0;
- printk(KERN_ERR "%s: bitmaps are not supported for %s\n",
+ pr_warn("%s: bitmaps are not supported for %s\n",
mdname(mddev), mddev->pers->name);
return 1;
}
rdev->sb_start = calc_dev_sboffset(rdev);
ret = read_disk_sb(rdev, MD_SB_BYTES);
- if (ret) return ret;
+ if (ret)
+ return ret;
ret = -EINVAL;
sb = page_address(rdev->sb_page);
if (sb->md_magic != MD_SB_MAGIC) {
- printk(KERN_ERR "md: invalid raid superblock magic on %s\n",
- b);
+ pr_warn("md: invalid raid superblock magic on %s\n", b);
goto abort;
}
if (sb->major_version != 0 ||
sb->minor_version < 90 ||
sb->minor_version > 91) {
- printk(KERN_WARNING "Bad version number %d.%d on %s\n",
- sb->major_version, sb->minor_version,
- b);
+ pr_warn("Bad version number %d.%d on %s\n",
+ sb->major_version, sb->minor_version, b);
goto abort;
}
goto abort;
if (md_csum_fold(calc_sb_csum(sb)) != md_csum_fold(sb->sb_csum)) {
- printk(KERN_WARNING "md: invalid superblock checksum on %s\n",
- b);
+ pr_warn("md: invalid superblock checksum on %s\n", b);
goto abort;
}
__u64 ev1, ev2;
mdp_super_t *refsb = page_address(refdev->sb_page);
if (!uuid_equal(refsb, sb)) {
- printk(KERN_WARNING "md: %s has different UUID to %s\n",
+ pr_warn("md: %s has different UUID to %s\n",
b, bdevname(refdev->bdev,b2));
goto abort;
}
if (!sb_equal(refsb, sb)) {
- printk(KERN_WARNING "md: %s has same UUID"
- " but different superblock to %s\n",
- b, bdevname(refdev->bdev, b2));
+ pr_warn("md: %s has same UUID but different superblock to %s\n",
+ b, bdevname(refdev->bdev, b2));
goto abort;
}
ev1 = md_event(sb);
}
if (desc->state & (1<<MD_DISK_WRITEMOSTLY))
set_bit(WriteMostly, &rdev->flags);
+ if (desc->state & (1<<MD_DISK_FAILFAST))
+ set_bit(FailFast, &rdev->flags);
} else /* MULTIPATH are always insync */
set_bit(In_sync, &rdev->flags);
return 0;
}
if (test_bit(WriteMostly, &rdev2->flags))
d->state |= (1<<MD_DISK_WRITEMOSTLY);
+ if (test_bit(FailFast, &rdev2->flags))
+ d->state |= (1<<MD_DISK_FAILFAST);
}
/* now set the "removed" and "faulty" bits on any missing devices */
for (i=0 ; i < mddev->raid_disks ; i++) {
if (IS_ENABLED(CONFIG_LBDAF) && (u64)num_sectors >= (2ULL << 32) &&
rdev->mddev->level >= 1)
num_sectors = (sector_t)(2ULL << 32) - 2;
- md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
+ do {
+ md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
rdev->sb_page);
- md_super_wait(rdev->mddev);
+ } while (md_super_wait(rdev->mddev) < 0);
return num_sectors;
}
return -EINVAL;
if (calc_sb_1_csum(sb) != sb->sb_csum) {
- printk("md: invalid superblock checksum on %s\n",
+ pr_warn("md: invalid superblock checksum on %s\n",
bdevname(rdev->bdev,b));
return -EINVAL;
}
if (le64_to_cpu(sb->data_size) < 10) {
- printk("md: data_size too small on %s\n",
- bdevname(rdev->bdev,b));
+ pr_warn("md: data_size too small on %s\n",
+ bdevname(rdev->bdev,b));
return -EINVAL;
}
if (sb->pad0 ||
sb->level != refsb->level ||
sb->layout != refsb->layout ||
sb->chunksize != refsb->chunksize) {
- printk(KERN_WARNING "md: %s has strangely different"
- " superblock to %s\n",
+ pr_warn("md: %s has strangely different superblock to %s\n",
bdevname(rdev->bdev,b),
bdevname(refdev->bdev,b2));
return -EINVAL;
case MD_DISK_ROLE_JOURNAL: /* journal device */
if (!(le32_to_cpu(sb->feature_map) & MD_FEATURE_JOURNAL)) {
/* journal device without journal feature */
- printk(KERN_WARNING
- "md: journal device provided without journal feature, ignoring the device\n");
+ pr_warn("md: journal device provided without journal feature, ignoring the device\n");
return -EINVAL;
}
set_bit(Journal, &rdev->flags);
}
if (sb->devflags & WriteMostly1)
set_bit(WriteMostly, &rdev->flags);
+ if (sb->devflags & FailFast1)
+ set_bit(FailFast, &rdev->flags);
if (le32_to_cpu(sb->feature_map) & MD_FEATURE_REPLACEMENT)
set_bit(Replacement, &rdev->flags);
} else /* MULTIPATH are always insync */
sb->chunksize = cpu_to_le32(mddev->chunk_sectors);
sb->level = cpu_to_le32(mddev->level);
sb->layout = cpu_to_le32(mddev->layout);
+ if (test_bit(FailFast, &rdev->flags))
+ sb->devflags |= FailFast1;
+ else
+ sb->devflags &= ~FailFast1;
if (test_bit(WriteMostly, &rdev->flags))
sb->devflags |= WriteMostly1;
sb->data_size = cpu_to_le64(num_sectors);
sb->super_offset = rdev->sb_start;
sb->sb_csum = calc_sb_1_csum(sb);
- md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
- rdev->sb_page);
- md_super_wait(rdev->mddev);
+ do {
+ md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
+ rdev->sb_page);
+ } while (md_super_wait(rdev->mddev) < 0);
return num_sectors;
}
blk_integrity_register(mddev->gendisk,
bdev_get_integrity(reference->bdev));
- printk(KERN_NOTICE "md: data integrity enabled on %s\n", mdname(mddev));
+ pr_debug("md: data integrity enabled on %s\n", mdname(mddev));
if (bioset_integrity_create(mddev->bio_set, BIO_POOL_SIZE)) {
- printk(KERN_ERR "md: failed to create integrity pool for %s\n",
+ pr_err("md: failed to create integrity pool for %s\n",
mdname(mddev));
return -EINVAL;
}
return 0;
if (blk_integrity_compare(mddev->gendisk, rdev->bdev->bd_disk) != 0) {
- printk(KERN_NOTICE "%s: incompatible integrity profile for %s\n",
- mdname(mddev), bdevname(rdev->bdev, name));
+ pr_err("%s: incompatible integrity profile for %s\n",
+ mdname(mddev), bdevname(rdev->bdev, name));
return -ENXIO;
}
rcu_read_unlock();
if (!test_bit(Journal, &rdev->flags) &&
mddev->max_disks && rdev->desc_nr >= mddev->max_disks) {
- printk(KERN_WARNING "md: %s: array is limited to %d devices\n",
- mdname(mddev), mddev->max_disks);
+ pr_warn("md: %s: array is limited to %d devices\n",
+ mdname(mddev), mddev->max_disks);
return -EBUSY;
}
bdevname(rdev->bdev,b);
strreplace(b, '/', '!');
rdev->mddev = mddev;
- printk(KERN_INFO "md: bind<%s>\n", b);
+ pr_debug("md: bind<%s>\n", b);
if ((err = kobject_add(&rdev->kobj, &mddev->kobj, "dev-%s", b)))
goto fail;
return 0;
fail:
- printk(KERN_WARNING "md: failed to register dev-%s for %s\n",
- b, mdname(mddev));
+ pr_warn("md: failed to register dev-%s for %s\n",
+ b, mdname(mddev));
return err;
}
bd_unlink_disk_holder(rdev->bdev, rdev->mddev->gendisk);
list_del_rcu(&rdev->same_set);
- printk(KERN_INFO "md: unbind<%s>\n", bdevname(rdev->bdev,b));
+ pr_debug("md: unbind<%s>\n", bdevname(rdev->bdev,b));
rdev->mddev = NULL;
sysfs_remove_link(&rdev->kobj, "block");
sysfs_put(rdev->sysfs_state);
bdev = blkdev_get_by_dev(dev, FMODE_READ|FMODE_WRITE|FMODE_EXCL,
shared ? (struct md_rdev *)lock_rdev : rdev);
if (IS_ERR(bdev)) {
- printk(KERN_ERR "md: could not open %s.\n",
- __bdevname(dev, b));
+ pr_warn("md: could not open %s.\n", __bdevname(dev, b));
return PTR_ERR(bdev);
}
rdev->bdev = bdev;
{
char b[BDEVNAME_SIZE];
- printk(KERN_INFO "md: export_rdev(%s)\n",
- bdevname(rdev->bdev,b));
+ pr_debug("md: export_rdev(%s)\n", bdevname(rdev->bdev,b));
md_rdev_clear(rdev);
#ifndef MODULE
if (test_bit(AutoDetected, &rdev->flags))
if (mddev->ro) {
if (force_change)
- set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
return;
}
repeat:
if (mddev_is_clustered(mddev)) {
- if (test_and_clear_bit(MD_CHANGE_DEVS, &mddev->flags))
+ if (test_and_clear_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags))
force_change = 1;
- if (test_and_clear_bit(MD_CHANGE_CLEAN, &mddev->flags))
+ if (test_and_clear_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags))
nospares = 1;
ret = md_cluster_ops->metadata_update_start(mddev);
/* Has someone else has updated the sb */
if (!does_sb_need_changing(mddev)) {
if (ret == 0)
md_cluster_ops->metadata_update_cancel(mddev);
- bit_clear_unless(&mddev->flags, BIT(MD_CHANGE_PENDING),
- BIT(MD_CHANGE_DEVS) |
- BIT(MD_CHANGE_CLEAN));
+ bit_clear_unless(&mddev->sb_flags, BIT(MD_SB_CHANGE_PENDING),
+ BIT(MD_SB_CHANGE_DEVS) |
+ BIT(MD_SB_CHANGE_CLEAN));
return;
}
}
}
if (!mddev->persistent) {
- clear_bit(MD_CHANGE_CLEAN, &mddev->flags);
- clear_bit(MD_CHANGE_DEVS, &mddev->flags);
+ clear_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
+ clear_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
if (!mddev->external) {
- clear_bit(MD_CHANGE_PENDING, &mddev->flags);
+ clear_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags);
rdev_for_each(rdev, mddev) {
if (rdev->badblocks.changed) {
rdev->badblocks.changed = 0;
mddev->utime = ktime_get_real_seconds();
- if (test_and_clear_bit(MD_CHANGE_DEVS, &mddev->flags))
+ if (test_and_clear_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags))
force_change = 1;
- if (test_and_clear_bit(MD_CHANGE_CLEAN, &mddev->flags))
+ if (test_and_clear_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_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
pr_debug("md: updating %s RAID superblock on device (in sync %d)\n",
mdname(mddev), mddev->in_sync);
+ if (mddev->queue)
+ blk_add_trace_msg(mddev->queue, "md md_update_sb");
+rewrite:
bitmap_update_sb(mddev->bitmap);
rdev_for_each(rdev, mddev) {
char b[BDEVNAME_SIZE];
/* only need to write one superblock... */
break;
}
- md_super_wait(mddev);
- /* if there was a failure, MD_CHANGE_DEVS was set, and we re-write super */
+ if (md_super_wait(mddev) < 0)
+ goto rewrite;
+ /* if there was a failure, MD_SB_CHANGE_DEVS was set, and we re-write super */
if (mddev_is_clustered(mddev) && ret == 0)
md_cluster_ops->metadata_update_finish(mddev);
if (mddev->in_sync != sync_req ||
- !bit_clear_unless(&mddev->flags, BIT(MD_CHANGE_PENDING),
- BIT(MD_CHANGE_DEVS) | BIT(MD_CHANGE_CLEAN)))
+ !bit_clear_unless(&mddev->sb_flags, BIT(MD_SB_CHANGE_PENDING),
+ BIT(MD_SB_CHANGE_DEVS) | BIT(MD_SB_CHANGE_CLEAN)))
/* have to write it out again */
goto repeat;
wake_up(&mddev->sb_wait);
}
sysfs_notify_dirent_safe(rdev->sysfs_state);
- set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
if (mddev->degraded)
set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
static ssize_t
state_show(struct md_rdev *rdev, char *page)
{
- char *sep = "";
+ char *sep = ",";
size_t len = 0;
unsigned long flags = ACCESS_ONCE(rdev->flags);
if (test_bit(Faulty, &flags) ||
- rdev->badblocks.unacked_exist) {
- len+= sprintf(page+len, "%sfaulty",sep);
- sep = ",";
- }
- if (test_bit(In_sync, &flags)) {
- len += sprintf(page+len, "%sin_sync",sep);
- sep = ",";
- }
- if (test_bit(Journal, &flags)) {
- len += sprintf(page+len, "%sjournal",sep);
- sep = ",";
- }
- if (test_bit(WriteMostly, &flags)) {
- len += sprintf(page+len, "%swrite_mostly",sep);
- sep = ",";
- }
+ (!test_bit(ExternalBbl, &flags) &&
+ rdev->badblocks.unacked_exist))
+ len += sprintf(page+len, "faulty%s", sep);
+ if (test_bit(In_sync, &flags))
+ len += sprintf(page+len, "in_sync%s", sep);
+ if (test_bit(Journal, &flags))
+ len += sprintf(page+len, "journal%s", sep);
+ if (test_bit(WriteMostly, &flags))
+ len += sprintf(page+len, "write_mostly%s", sep);
if (test_bit(Blocked, &flags) ||
(rdev->badblocks.unacked_exist
- && !test_bit(Faulty, &flags))) {
- len += sprintf(page+len, "%sblocked", sep);
- sep = ",";
- }
+ && !test_bit(Faulty, &flags)))
+ len += sprintf(page+len, "blocked%s", sep);
if (!test_bit(Faulty, &flags) &&
!test_bit(Journal, &flags) &&
- !test_bit(In_sync, &flags)) {
- len += sprintf(page+len, "%sspare", sep);
- sep = ",";
- }
- if (test_bit(WriteErrorSeen, &flags)) {
- len += sprintf(page+len, "%swrite_error", sep);
- sep = ",";
- }
- if (test_bit(WantReplacement, &flags)) {
- len += sprintf(page+len, "%swant_replacement", sep);
- sep = ",";
- }
- if (test_bit(Replacement, &flags)) {
- len += sprintf(page+len, "%sreplacement", sep);
- sep = ",";
- }
+ !test_bit(In_sync, &flags))
+ len += sprintf(page+len, "spare%s", sep);
+ if (test_bit(WriteErrorSeen, &flags))
+ len += sprintf(page+len, "write_error%s", sep);
+ if (test_bit(WantReplacement, &flags))
+ len += sprintf(page+len, "want_replacement%s", sep);
+ if (test_bit(Replacement, &flags))
+ len += sprintf(page+len, "replacement%s", sep);
+ if (test_bit(ExternalBbl, &flags))
+ len += sprintf(page+len, "external_bbl%s", sep);
+ if (test_bit(FailFast, &flags))
+ len += sprintf(page+len, "failfast%s", sep);
+
+ if (len)
+ len -= strlen(sep);
return len+sprintf(page+len, "\n");
}
* so that it gets rebuilt based on bitmap
* write_error - sets WriteErrorSeen
* -write_error - clears WriteErrorSeen
+ * {,-}failfast - set/clear FailFast
*/
int err = -EINVAL;
if (cmd_match(buf, "faulty") && rdev->mddev->pers) {
if (err == 0) {
md_kick_rdev_from_array(rdev);
- if (mddev->pers)
- md_update_sb(mddev, 1);
+ if (mddev->pers) {
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
+ md_wakeup_thread(mddev->thread);
+ }
md_new_event(mddev);
}
}
err = 0;
} else if (cmd_match(buf, "-blocked")) {
if (!test_bit(Faulty, &rdev->flags) &&
+ !test_bit(ExternalBbl, &rdev->flags) &&
rdev->badblocks.unacked_exist) {
/* metadata handler doesn't understand badblocks,
* so we need to fail the device
} else if (cmd_match(buf, "insync") && rdev->raid_disk == -1) {
set_bit(In_sync, &rdev->flags);
err = 0;
+ } else if (cmd_match(buf, "failfast")) {
+ set_bit(FailFast, &rdev->flags);
+ err = 0;
+ } else if (cmd_match(buf, "-failfast")) {
+ clear_bit(FailFast, &rdev->flags);
+ err = 0;
} else if (cmd_match(buf, "-insync") && rdev->raid_disk >= 0 &&
!test_bit(Journal, &rdev->flags)) {
if (rdev->mddev->pers == NULL) {
}
} else
err = -EBUSY;
+ } else if (cmd_match(buf, "external_bbl") && (rdev->mddev->external)) {
+ set_bit(ExternalBbl, &rdev->flags);
+ rdev->badblocks.shift = 0;
+ err = 0;
+ } else if (cmd_match(buf, "-external_bbl") && (rdev->mddev->external)) {
+ clear_bit(ExternalBbl, &rdev->flags);
+ err = 0;
}
if (!err)
sysfs_notify_dirent_safe(rdev->sysfs_state);
sector_t size;
rdev = kzalloc(sizeof(*rdev), GFP_KERNEL);
- if (!rdev) {
- printk(KERN_ERR "md: could not alloc mem for new device!\n");
+ if (!rdev)
return ERR_PTR(-ENOMEM);
- }
err = md_rdev_init(rdev);
if (err)
size = i_size_read(rdev->bdev->bd_inode) >> BLOCK_SIZE_BITS;
if (!size) {
- printk(KERN_WARNING
- "md: %s has zero or unknown size, marking faulty!\n",
+ pr_warn("md: %s has zero or unknown size, marking faulty!\n",
bdevname(rdev->bdev,b));
err = -EINVAL;
goto abort_free;
err = super_types[super_format].
load_super(rdev, NULL, super_minor);
if (err == -EINVAL) {
- printk(KERN_WARNING
- "md: %s does not have a valid v%d.%d "
- "superblock, not importing!\n",
+ pr_warn("md: %s does not have a valid v%d.%d superblock, not importing!\n",
bdevname(rdev->bdev,b),
- super_format, super_minor);
+ super_format, super_minor);
goto abort_free;
}
if (err < 0) {
- printk(KERN_WARNING
- "md: could not read %s's sb, not importing!\n",
+ pr_warn("md: could not read %s's sb, not importing!\n",
bdevname(rdev->bdev,b));
goto abort_free;
}
case 0:
break;
default:
- printk( KERN_ERR \
- "md: fatal superblock inconsistency in %s"
- " -- removing from array\n",
+ pr_warn("md: fatal superblock inconsistency in %s -- removing from array\n",
bdevname(rdev->bdev,b));
md_kick_rdev_from_array(rdev);
}
if (mddev->max_disks &&
(rdev->desc_nr >= mddev->max_disks ||
i > mddev->max_disks)) {
- printk(KERN_WARNING
- "md: %s: %s: only %d devices permitted\n",
- mdname(mddev), bdevname(rdev->bdev, b),
- mddev->max_disks);
+ pr_warn("md: %s: %s: only %d devices permitted\n",
+ mdname(mddev), bdevname(rdev->bdev, b),
+ mddev->max_disks);
md_kick_rdev_from_array(rdev);
continue;
}
if (rdev != freshest) {
if (super_types[mddev->major_version].
validate_super(mddev, rdev)) {
- printk(KERN_WARNING "md: kicking non-fresh %s"
- " from array!\n",
+ pr_warn("md: kicking non-fresh %s from array!\n",
bdevname(rdev->bdev,b));
md_kick_rdev_from_array(rdev);
continue;
unsigned long msec;
if (mddev_is_clustered(mddev)) {
- pr_info("md: Safemode is disabled for clustered mode\n");
+ pr_warn("md: Safemode is disabled for clustered mode\n");
return -EINVAL;
}
rv = -EINVAL;
if (!mddev->pers->quiesce) {
- printk(KERN_WARNING "md: %s: %s does not support online personality change\n",
- mdname(mddev), mddev->pers->name);
+ pr_warn("md: %s: %s does not support online personality change\n",
+ mdname(mddev), mddev->pers->name);
goto out_unlock;
}
pers = find_pers(level, clevel);
if (!pers || !try_module_get(pers->owner)) {
spin_unlock(&pers_lock);
- printk(KERN_WARNING "md: personality %s not loaded\n", clevel);
+ pr_warn("md: personality %s not loaded\n", clevel);
rv = -EINVAL;
goto out_unlock;
}
}
if (!pers->takeover) {
module_put(pers->owner);
- printk(KERN_WARNING "md: %s: %s does not support personality takeover\n",
- mdname(mddev), clevel);
+ pr_warn("md: %s: %s does not support personality takeover\n",
+ mdname(mddev), clevel);
rv = -EINVAL;
goto out_unlock;
}
mddev->delta_disks = 0;
mddev->reshape_backwards = 0;
module_put(pers->owner);
- printk(KERN_WARNING "md: %s: %s would not accept array\n",
- mdname(mddev), clevel);
+ pr_warn("md: %s: %s would not accept array\n",
+ mdname(mddev), clevel);
rv = PTR_ERR(priv);
goto out_unlock;
}
pers->sync_request != NULL) {
/* need to add the md_redundancy_group */
if (sysfs_create_group(&mddev->kobj, &md_redundancy_group))
- printk(KERN_WARNING
- "md: cannot register extra attributes for %s\n",
- mdname(mddev));
+ pr_warn("md: cannot register extra attributes for %s\n",
+ mdname(mddev));
mddev->sysfs_action = sysfs_get_dirent(mddev->kobj.sd, "sync_action");
}
if (oldpers->sync_request != NULL &&
clear_bit(In_sync, &rdev->flags);
else {
if (sysfs_link_rdev(mddev, rdev))
- printk(KERN_WARNING "md: cannot register rd%d"
- " for %s after level change\n",
- rdev->raid_disk, mdname(mddev));
+ pr_warn("md: cannot register rd%d for %s after level change\n",
+ rdev->raid_disk, mdname(mddev));
}
}
}
blk_set_stacking_limits(&mddev->queue->limits);
pers->run(mddev);
- set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
mddev_resume(mddev);
if (!mddev->thread)
md_update_sb(mddev, 1);
if (!err) {
mddev->recovery_cp = n;
if (mddev->pers)
- set_bit(MD_CHANGE_CLEAN, &mddev->flags);
+ set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
}
mddev_unlock(mddev);
return err ?: len;
st = read_auto;
break;
case 0:
- if (test_bit(MD_CHANGE_PENDING, &mddev->flags))
+ if (test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags))
st = write_pending;
else if (mddev->in_sync)
st = clean;
spin_lock(&mddev->lock);
if (st == active) {
restart_array(mddev);
- clear_bit(MD_CHANGE_PENDING, &mddev->flags);
+ clear_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags);
+ md_wakeup_thread(mddev->thread);
wake_up(&mddev->sb_wait);
err = 0;
} else /* st == clean */ {
mddev->in_sync = 1;
if (mddev->safemode == 1)
mddev->safemode = 0;
- set_bit(MD_CHANGE_CLEAN, &mddev->flags);
+ set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
}
err = 0;
} else
mddev->in_sync = 1;
if (mddev->safemode == 1)
mddev->safemode = 0;
- set_bit(MD_CHANGE_CLEAN, &mddev->flags);
+ set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
}
err = 0;
} else
err = restart_array(mddev);
if (err)
break;
- clear_bit(MD_CHANGE_PENDING, &mddev->flags);
+ clear_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags);
wake_up(&mddev->sb_wait);
err = 0;
} else {
/* This isn't possible, but as kobject_init_and_add is marked
* __must_check, we must do something with the result
*/
- printk(KERN_WARNING "md: cannot register %s/md - name in use\n",
- disk->disk_name);
+ pr_debug("md: cannot register %s/md - name in use\n",
+ disk->disk_name);
error = 0;
}
if (mddev->kobj.sd &&
sysfs_create_group(&mddev->kobj, &md_bitmap_group))
- printk(KERN_DEBUG "pointless warning\n");
+ pr_debug("pointless warning\n");
mutex_unlock(&mddev->open_mutex);
abort:
mutex_unlock(&disks_mutex);
if (mddev->dev_sectors &&
rdev->data_offset + mddev->dev_sectors
> rdev->sb_start) {
- printk("md: %s: data overlaps metadata\n",
- mdname(mddev));
+ pr_warn("md: %s: data overlaps metadata\n",
+ mdname(mddev));
return -EINVAL;
}
} else {
if (rdev->sb_start + rdev->sb_size/512
> rdev->data_offset) {
- printk("md: %s: metadata overlaps data\n",
- mdname(mddev));
+ pr_warn("md: %s: metadata overlaps data\n",
+ mdname(mddev));
return -EINVAL;
}
}
if (!pers || !try_module_get(pers->owner)) {
spin_unlock(&pers_lock);
if (mddev->level != LEVEL_NONE)
- printk(KERN_WARNING "md: personality for level %d is not loaded!\n",
- mddev->level);
+ pr_warn("md: personality for level %d is not loaded!\n",
+ mddev->level);
else
- printk(KERN_WARNING "md: personality for level %s is not loaded!\n",
- mddev->clevel);
+ pr_warn("md: personality for level %s is not loaded!\n",
+ mddev->clevel);
return -EINVAL;
}
spin_unlock(&pers_lock);
if (rdev < rdev2 &&
rdev->bdev->bd_contains ==
rdev2->bdev->bd_contains) {
- printk(KERN_WARNING
- "%s: WARNING: %s appears to be"
- " on the same physical disk as"
- " %s.\n",
- mdname(mddev),
- bdevname(rdev->bdev,b),
- bdevname(rdev2->bdev,b2));
+ pr_warn("%s: WARNING: %s appears to be on the same physical disk as %s.\n",
+ mdname(mddev),
+ bdevname(rdev->bdev,b),
+ bdevname(rdev2->bdev,b2));
warned = 1;
}
}
if (warned)
- printk(KERN_WARNING
- "True protection against single-disk"
- " failure might be compromised.\n");
+ pr_warn("True protection against single-disk failure might be compromised.\n");
}
mddev->recovery = 0;
if (start_readonly && mddev->ro == 0)
mddev->ro = 2; /* read-only, but switch on first write */
+ /*
+ * NOTE: some pers->run(), for example r5l_recovery_log(), wakes
+ * up mddev->thread. It is important to initialize critical
+ * resources for mddev->thread BEFORE calling pers->run().
+ */
err = pers->run(mddev);
if (err)
- printk(KERN_ERR "md: pers->run() failed ...\n");
+ pr_warn("md: pers->run() failed ...\n");
else if (pers->size(mddev, 0, 0) < mddev->array_sectors) {
- WARN_ONCE(!mddev->external_size, "%s: default size too small,"
- " but 'external_size' not in effect?\n", __func__);
- printk(KERN_ERR
- "md: invalid array_size %llu > default size %llu\n",
- (unsigned long long)mddev->array_sectors / 2,
- (unsigned long long)pers->size(mddev, 0, 0) / 2);
+ WARN_ONCE(!mddev->external_size,
+ "%s: default size too small, but 'external_size' not in effect?\n",
+ __func__);
+ pr_warn("md: invalid array_size %llu > default size %llu\n",
+ (unsigned long long)mddev->array_sectors / 2,
+ (unsigned long long)pers->size(mddev, 0, 0) / 2);
err = -EINVAL;
}
if (err == 0 && pers->sync_request &&
bitmap = bitmap_create(mddev, -1);
if (IS_ERR(bitmap)) {
err = PTR_ERR(bitmap);
- printk(KERN_ERR "%s: failed to create bitmap (%d)\n",
- mdname(mddev), err);
+ pr_warn("%s: failed to create bitmap (%d)\n",
+ mdname(mddev), err);
} else
mddev->bitmap = bitmap;
if (pers->sync_request) {
if (mddev->kobj.sd &&
sysfs_create_group(&mddev->kobj, &md_redundancy_group))
- printk(KERN_WARNING
- "md: cannot register extra attributes for %s\n",
- mdname(mddev));
+ pr_warn("md: cannot register extra attributes for %s\n",
+ mdname(mddev));
mddev->sysfs_action = sysfs_get_dirent_safe(mddev->kobj.sd, "sync_action");
} else if (mddev->ro == 2) /* auto-readonly not meaningful */
mddev->ro = 0;
set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
- if (mddev->flags & MD_UPDATE_SB_FLAGS)
+ if (mddev->sb_flags)
md_update_sb(mddev, 0);
md_new_event(mddev);
mddev->safemode = 0;
mddev->ro = 0;
set_disk_ro(disk, 0);
- printk(KERN_INFO "md: %s switched to read-write mode.\n",
- mdname(mddev));
+ pr_debug("md: %s switched to read-write mode.\n", mdname(mddev));
/* Kick recovery or resync if necessary */
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
md_wakeup_thread(mddev->thread);
mddev->level = LEVEL_NONE;
mddev->clevel[0] = 0;
mddev->flags = 0;
+ mddev->sb_flags = 0;
mddev->ro = 0;
mddev->metadata_type[0] = 0;
mddev->chunk_sectors = 0;
del_timer_sync(&mddev->safemode_timer);
+ if (mddev->pers && mddev->pers->quiesce) {
+ mddev->pers->quiesce(mddev, 1);
+ mddev->pers->quiesce(mddev, 0);
+ }
bitmap_flush(mddev);
- md_super_wait(mddev);
if (mddev->ro == 0 &&
((!mddev->in_sync && !mddev_is_clustered(mddev)) ||
- (mddev->flags & MD_UPDATE_SB_FLAGS))) {
+ mddev->sb_flags)) {
/* mark array as shutdown cleanly */
if (!mddev_is_clustered(mddev))
mddev->in_sync = 1;
struct bitmap *bitmap = mddev->bitmap;
/* wait for behind writes to complete */
if (bitmap && atomic_read(&bitmap->behind_writes) > 0) {
- printk(KERN_INFO "md:%s: behind writes in progress - waiting to stop.\n",
- mdname(mddev));
+ pr_debug("md:%s: behind writes in progress - waiting to stop.\n",
+ mdname(mddev));
/* need to kick something here to make sure I/O goes? */
wait_event(bitmap->behind_wait,
atomic_read(&bitmap->behind_writes) == 0);
* which will now never happen */
wake_up_process(mddev->sync_thread->tsk);
- if (mddev->external && test_bit(MD_CHANGE_PENDING, &mddev->flags))
+ if (mddev->external && test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags))
return -EBUSY;
mddev_unlock(mddev);
wait_event(resync_wait, !test_bit(MD_RECOVERY_RUNNING,
&mddev->recovery));
wait_event(mddev->sb_wait,
- !test_bit(MD_CHANGE_PENDING, &mddev->flags));
+ !test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags));
mddev_lock_nointr(mddev);
mutex_lock(&mddev->open_mutex);
if ((mddev->pers && atomic_read(&mddev->openers) > !!bdev) ||
mddev->sync_thread ||
test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)) {
- printk("md: %s still in use.\n",mdname(mddev));
+ pr_warn("md: %s still in use.\n",mdname(mddev));
if (did_freeze) {
clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
mddev->sysfs_active ||
mddev->sync_thread ||
test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)) {
- printk("md: %s still in use.\n",mdname(mddev));
+ pr_warn("md: %s still in use.\n",mdname(mddev));
mutex_unlock(&mddev->open_mutex);
if (did_freeze) {
clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
* Free resources if final stop
*/
if (mode == 0) {
- printk(KERN_INFO "md: %s stopped.\n", mdname(mddev));
+ pr_info("md: %s stopped.\n", mdname(mddev));
bitmap_destroy(mddev);
if (mddev->bitmap_info.file) {
if (list_empty(&mddev->disks))
return;
- printk(KERN_INFO "md: running: ");
+ pr_info("md: running: ");
rdev_for_each(rdev, mddev) {
char b[BDEVNAME_SIZE];
- printk("<%s>", bdevname(rdev->bdev,b));
+ pr_cont("<%s>", bdevname(rdev->bdev,b));
}
- printk("\n");
+ pr_cont("\n");
err = do_md_run(mddev);
if (err) {
- printk(KERN_WARNING "md: do_md_run() returned %d\n", err);
+ pr_warn("md: do_md_run() returned %d\n", err);
do_md_stop(mddev, 0, NULL);
}
}
struct mddev *mddev;
char b[BDEVNAME_SIZE];
- printk(KERN_INFO "md: autorun ...\n");
+ pr_info("md: autorun ...\n");
while (!list_empty(&pending_raid_disks)) {
int unit;
dev_t dev;
rdev0 = list_entry(pending_raid_disks.next,
struct md_rdev, same_set);
- printk(KERN_INFO "md: considering %s ...\n",
- bdevname(rdev0->bdev,b));
+ pr_debug("md: considering %s ...\n", bdevname(rdev0->bdev,b));
INIT_LIST_HEAD(&candidates);
rdev_for_each_list(rdev, tmp, &pending_raid_disks)
if (super_90_load(rdev, rdev0, 0) >= 0) {
- printk(KERN_INFO "md: adding %s ...\n",
- bdevname(rdev->bdev,b));
+ pr_debug("md: adding %s ...\n",
+ bdevname(rdev->bdev,b));
list_move(&rdev->same_set, &candidates);
}
/*
unit = MINOR(dev);
}
if (rdev0->preferred_minor != unit) {
- printk(KERN_INFO "md: unit number in %s is bad: %d\n",
- bdevname(rdev0->bdev, b), rdev0->preferred_minor);
+ pr_warn("md: unit number in %s is bad: %d\n",
+ bdevname(rdev0->bdev, b), rdev0->preferred_minor);
break;
}
if (!mddev || !mddev->gendisk) {
if (mddev)
mddev_put(mddev);
- printk(KERN_ERR
- "md: cannot allocate memory for md drive.\n");
break;
}
if (mddev_lock(mddev))
- printk(KERN_WARNING "md: %s locked, cannot run\n",
- mdname(mddev));
+ pr_warn("md: %s locked, cannot run\n", mdname(mddev));
else if (mddev->raid_disks || mddev->major_version
|| !list_empty(&mddev->disks)) {
- printk(KERN_WARNING
- "md: %s already running, cannot run %s\n",
+ pr_warn("md: %s already running, cannot run %s\n",
mdname(mddev), bdevname(rdev0->bdev,b));
mddev_unlock(mddev);
} else {
- printk(KERN_INFO "md: created %s\n", mdname(mddev));
+ pr_debug("md: created %s\n", mdname(mddev));
mddev->persistent = 1;
rdev_for_each_list(rdev, tmp, &candidates) {
list_del_init(&rdev->same_set);
}
mddev_put(mddev);
}
- printk(KERN_INFO "md: ... autorun DONE.\n");
+ pr_info("md: ... autorun DONE.\n");
}
#endif /* !MODULE */
info.state |= (1<<MD_DISK_JOURNAL);
if (test_bit(WriteMostly, &rdev->flags))
info.state |= (1<<MD_DISK_WRITEMOSTLY);
+ if (test_bit(FailFast, &rdev->flags))
+ info.state |= (1<<MD_DISK_FAILFAST);
} else {
info.major = info.minor = 0;
info.raid_disk = -1;
if (mddev_is_clustered(mddev) &&
!(info->state & ((1 << MD_DISK_CLUSTER_ADD) | (1 << MD_DISK_CANDIDATE)))) {
- pr_err("%s: Cannot add to clustered mddev.\n",
- mdname(mddev));
+ pr_warn("%s: Cannot add to clustered mddev.\n",
+ mdname(mddev));
return -EINVAL;
}
/* expecting a device which has a superblock */
rdev = md_import_device(dev, mddev->major_version, mddev->minor_version);
if (IS_ERR(rdev)) {
- printk(KERN_WARNING
- "md: md_import_device returned %ld\n",
+ pr_warn("md: md_import_device returned %ld\n",
PTR_ERR(rdev));
return PTR_ERR(rdev);
}
err = super_types[mddev->major_version]
.load_super(rdev, rdev0, mddev->minor_version);
if (err < 0) {
- printk(KERN_WARNING
- "md: %s has different UUID to %s\n",
+ pr_warn("md: %s has different UUID to %s\n",
bdevname(rdev->bdev,b),
bdevname(rdev0->bdev,b2));
export_rdev(rdev);
if (mddev->pers) {
int err;
if (!mddev->pers->hot_add_disk) {
- printk(KERN_WARNING
- "%s: personality does not support diskops!\n",
- mdname(mddev));
+ pr_warn("%s: personality does not support diskops!\n",
+ mdname(mddev));
return -EINVAL;
}
if (mddev->persistent)
else
rdev = md_import_device(dev, -1, -1);
if (IS_ERR(rdev)) {
- printk(KERN_WARNING
- "md: md_import_device returned %ld\n",
+ pr_warn("md: md_import_device returned %ld\n",
PTR_ERR(rdev));
return PTR_ERR(rdev);
}
set_bit(WriteMostly, &rdev->flags);
else
clear_bit(WriteMostly, &rdev->flags);
+ if (info->state & (1<<MD_DISK_FAILFAST))
+ set_bit(FailFast, &rdev->flags);
+ else
+ clear_bit(FailFast, &rdev->flags);
if (info->state & (1<<MD_DISK_JOURNAL)) {
struct md_rdev *rdev2;
* for major_version==0 superblocks
*/
if (mddev->major_version != 0) {
- printk(KERN_WARNING "%s: ADD_NEW_DISK not supported\n",
- mdname(mddev));
+ pr_warn("%s: ADD_NEW_DISK not supported\n", mdname(mddev));
return -EINVAL;
}
int err;
rdev = md_import_device(dev, -1, 0);
if (IS_ERR(rdev)) {
- printk(KERN_WARNING
- "md: error, md_import_device() returned %ld\n",
+ pr_warn("md: error, md_import_device() returned %ld\n",
PTR_ERR(rdev));
return PTR_ERR(rdev);
}
if (info->state & (1<<MD_DISK_WRITEMOSTLY))
set_bit(WriteMostly, &rdev->flags);
+ if (info->state & (1<<MD_DISK_FAILFAST))
+ set_bit(FailFast, &rdev->flags);
if (!mddev->persistent) {
- printk(KERN_INFO "md: nonpersistent superblock ...\n");
+ pr_debug("md: nonpersistent superblock ...\n");
rdev->sb_start = i_size_read(rdev->bdev->bd_inode) / 512;
} else
rdev->sb_start = calc_dev_sboffset(rdev);
md_cluster_ops->remove_disk(mddev, rdev);
md_kick_rdev_from_array(rdev);
- md_update_sb(mddev, 1);
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
+ if (mddev->thread)
+ md_wakeup_thread(mddev->thread);
+ else
+ md_update_sb(mddev, 1);
md_new_event(mddev);
return 0;
busy:
- printk(KERN_WARNING "md: cannot remove active disk %s from %s ...\n",
- bdevname(rdev->bdev,b), mdname(mddev));
+ pr_debug("md: cannot remove active disk %s from %s ...\n",
+ bdevname(rdev->bdev,b), mdname(mddev));
return -EBUSY;
}
return -ENODEV;
if (mddev->major_version != 0) {
- printk(KERN_WARNING "%s: HOT_ADD may only be used with"
- " version-0 superblocks.\n",
+ pr_warn("%s: HOT_ADD may only be used with version-0 superblocks.\n",
mdname(mddev));
return -EINVAL;
}
if (!mddev->pers->hot_add_disk) {
- printk(KERN_WARNING
- "%s: personality does not support diskops!\n",
+ pr_warn("%s: personality does not support diskops!\n",
mdname(mddev));
return -EINVAL;
}
rdev = md_import_device(dev, -1, 0);
if (IS_ERR(rdev)) {
- printk(KERN_WARNING
- "md: error, md_import_device() returned %ld\n",
+ pr_warn("md: error, md_import_device() returned %ld\n",
PTR_ERR(rdev));
return -EINVAL;
}
rdev->sectors = rdev->sb_start;
if (test_bit(Faulty, &rdev->flags)) {
- printk(KERN_WARNING
- "md: can not hot-add faulty %s disk to %s!\n",
+ pr_warn("md: can not hot-add faulty %s disk to %s!\n",
bdevname(rdev->bdev,b), mdname(mddev));
err = -EINVAL;
goto abort_export;
rdev->raid_disk = -1;
- md_update_sb(mddev, 1);
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
+ if (!mddev->thread)
+ md_update_sb(mddev, 1);
/*
* Kick recovery, maybe this spare has to be added to the
* array immediately.
f = fget(fd);
if (f == NULL) {
- printk(KERN_ERR "%s: error: failed to get bitmap file\n",
- mdname(mddev));
+ pr_warn("%s: error: failed to get bitmap file\n",
+ mdname(mddev));
return -EBADF;
}
inode = f->f_mapping->host;
if (!S_ISREG(inode->i_mode)) {
- printk(KERN_ERR "%s: error: bitmap file must be a regular file\n",
- mdname(mddev));
+ pr_warn("%s: error: bitmap file must be a regular file\n",
+ mdname(mddev));
err = -EBADF;
} else if (!(f->f_mode & FMODE_WRITE)) {
- printk(KERN_ERR "%s: error: bitmap file must open for write\n",
- mdname(mddev));
+ pr_warn("%s: error: bitmap file must open for write\n",
+ mdname(mddev));
err = -EBADF;
} else if (atomic_read(&inode->i_writecount) != 1) {
- printk(KERN_ERR "%s: error: bitmap file is already in use\n",
- mdname(mddev));
+ pr_warn("%s: error: bitmap file is already in use\n",
+ mdname(mddev));
err = -EBUSY;
}
if (err) {
info->major_version >= ARRAY_SIZE(super_types) ||
super_types[info->major_version].name == NULL) {
/* maybe try to auto-load a module? */
- printk(KERN_INFO
- "md: superblock version %d not known\n",
+ pr_warn("md: superblock version %d not known\n",
info->major_version);
return -EINVAL;
}
mddev->max_disks = MD_SB_DISKS;
- if (mddev->persistent)
+ if (mddev->persistent) {
mddev->flags = 0;
- set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ mddev->sb_flags = 0;
+ }
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
mddev->bitmap_info.default_offset = MD_SB_BYTES >> 9;
mddev->bitmap_info.default_space = 64*2 - (MD_SB_BYTES >> 9);
if (mddev->bitmap_info.nodes) {
/* hold PW on all the bitmap lock */
if (md_cluster_ops->lock_all_bitmaps(mddev) <= 0) {
- printk("md: can't change bitmap to none since the"
- " array is in use by more than one node\n");
+ pr_warn("md: can't change bitmap to none since the array is in use by more than one node\n");
rv = -EPERM;
md_cluster_ops->unlock_all_bitmaps(mddev);
goto err;
/* need to ensure recovery thread has run */
wait_event_interruptible_timeout(mddev->sb_wait,
!test_bit(MD_RECOVERY_NEEDED,
- &mddev->flags),
+ &mddev->recovery),
msecs_to_jiffies(5000));
if (cmd == STOP_ARRAY || cmd == STOP_ARRAY_RO) {
/* Need to flush page cache, and ensure no-one else opens
}
err = mddev_lock(mddev);
if (err) {
- printk(KERN_INFO
- "md: ioctl lock interrupted, reason %d, cmd %d\n",
- err, cmd);
+ pr_debug("md: ioctl lock interrupted, reason %d, cmd %d\n",
+ err, cmd);
goto out;
}
if (mddev->pers) {
err = update_array_info(mddev, &info);
if (err) {
- printk(KERN_WARNING "md: couldn't update"
- " array info. %d\n", err);
+ pr_warn("md: couldn't update array info. %d\n", err);
goto unlock;
}
goto unlock;
}
if (!list_empty(&mddev->disks)) {
- printk(KERN_WARNING
- "md: array %s already has disks!\n",
- mdname(mddev));
+ pr_warn("md: array %s already has disks!\n", mdname(mddev));
err = -EBUSY;
goto unlock;
}
if (mddev->raid_disks) {
- printk(KERN_WARNING
- "md: array %s already initialised!\n",
- mdname(mddev));
+ pr_warn("md: array %s already initialised!\n", mdname(mddev));
err = -EBUSY;
goto unlock;
}
err = set_array_info(mddev, &info);
if (err) {
- printk(KERN_WARNING "md: couldn't set"
- " array info. %d\n", err);
+ pr_warn("md: couldn't set array info. %d\n", err);
goto unlock;
}
goto unlock;
/* If a device failed while we were read-only, we
* need to make sure the metadata is updated now.
*/
- if (test_bit(MD_CHANGE_DEVS, &mddev->flags)) {
+ if (test_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags)) {
mddev_unlock(mddev);
wait_event(mddev->sb_wait,
- !test_bit(MD_CHANGE_DEVS, &mddev->flags) &&
- !test_bit(MD_CHANGE_PENDING, &mddev->flags));
+ !test_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags) &&
+ !test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags));
mddev_lock_nointr(mddev);
}
} else {
if (test_bit(MD_CLOSING, &mddev->flags)) {
mutex_unlock(&mddev->open_mutex);
- return -ENODEV;
+ err = -ENODEV;
+ goto out;
}
err = 0;
check_disk_change(bdev);
out:
+ if (err)
+ mddev_put(mddev);
return err;
}
wait_event_interruptible_timeout
(thread->wqueue,
test_bit(THREAD_WAKEUP, &thread->flags)
- || kthread_should_stop(),
+ || kthread_should_stop() || kthread_should_park(),
thread->timeout);
clear_bit(THREAD_WAKEUP, &thread->flags);
+ if (kthread_should_park())
+ kthread_parkme();
if (!kthread_should_stop())
thread->run(thread);
}
int register_md_personality(struct md_personality *p)
{
- printk(KERN_INFO "md: %s personality registered for level %d\n",
- p->name, p->level);
+ pr_debug("md: %s personality registered for level %d\n",
+ p->name, p->level);
spin_lock(&pers_lock);
list_add_tail(&p->list, &pers_list);
spin_unlock(&pers_lock);
int unregister_md_personality(struct md_personality *p)
{
- printk(KERN_INFO "md: %s personality unregistered\n", p->name);
+ pr_debug("md: %s personality unregistered\n", p->name);
spin_lock(&pers_lock);
list_del_init(&p->list);
spin_unlock(&pers_lock);
spin_lock(&pers_lock);
/* ensure module won't be unloaded */
if (!md_cluster_ops || !try_module_get(md_cluster_mod)) {
- pr_err("can't find md-cluster module or get it's reference.\n");
+ pr_warn("can't find md-cluster module or get it's reference.\n");
spin_unlock(&pers_lock);
return -ENOENT;
}
spin_lock(&mddev->lock);
if (mddev->in_sync) {
mddev->in_sync = 0;
- set_bit(MD_CHANGE_CLEAN, &mddev->flags);
- set_bit(MD_CHANGE_PENDING, &mddev->flags);
+ set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
+ set_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags);
md_wakeup_thread(mddev->thread);
did_change = 1;
}
if (did_change)
sysfs_notify_dirent_safe(mddev->sysfs_state);
wait_event(mddev->sb_wait,
- !test_bit(MD_CHANGE_PENDING, &mddev->flags));
+ !test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags));
}
EXPORT_SYMBOL(md_write_start);
* attempting a GFP_KERNEL allocation while holding the mddev lock.
* Must be called with mddev_lock held.
*
- * In the ->external case MD_CHANGE_PENDING can not be cleared until mddev->lock
+ * In the ->external case MD_SB_CHANGE_PENDING can not be cleared until mddev->lock
* is dropped, so return -EAGAIN after notifying userspace.
*/
int md_allow_write(struct mddev *mddev)
spin_lock(&mddev->lock);
if (mddev->in_sync) {
mddev->in_sync = 0;
- set_bit(MD_CHANGE_CLEAN, &mddev->flags);
- set_bit(MD_CHANGE_PENDING, &mddev->flags);
+ set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
+ set_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags);
if (mddev->safemode_delay &&
mddev->safemode == 0)
mddev->safemode = 1;
} else
spin_unlock(&mddev->lock);
- if (test_bit(MD_CHANGE_PENDING, &mddev->flags))
+ if (test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags))
return -EAGAIN;
else
return 0;
mddev2->curr_resync >= mddev->curr_resync) {
if (mddev2_minor != mddev2->md_minor) {
mddev2_minor = mddev2->md_minor;
- printk(KERN_INFO "md: delaying %s of %s"
- " until %s has finished (they"
- " share one or more physical units)\n",
- desc, mdname(mddev),
- mdname(mddev2));
+ pr_info("md: delaying %s of %s until %s has finished (they share one or more physical units)\n",
+ desc, mdname(mddev),
+ mdname(mddev2));
}
mddev_put(mddev2);
if (signal_pending(current))
}
}
- printk(KERN_INFO "md: %s of RAID array %s\n", desc, mdname(mddev));
- printk(KERN_INFO "md: minimum _guaranteed_ speed:"
- " %d KB/sec/disk.\n", speed_min(mddev));
- printk(KERN_INFO "md: using maximum available idle IO bandwidth "
- "(but not more than %d KB/sec) for %s.\n",
- speed_max(mddev), desc);
+ pr_info("md: %s of RAID array %s\n", desc, mdname(mddev));
+ pr_debug("md: minimum _guaranteed_ speed: %d KB/sec/disk.\n", speed_min(mddev));
+ pr_debug("md: using maximum available idle IO bandwidth (but not more than %d KB/sec) for %s.\n",
+ speed_max(mddev), desc);
is_mddev_idle(mddev, 1); /* this initializes IO event counters */
* Tune reconstruction:
*/
window = 32*(PAGE_SIZE/512);
- printk(KERN_INFO "md: using %dk window, over a total of %lluk.\n",
- window/2, (unsigned long long)max_sectors/2);
+ pr_debug("md: using %dk window, over a total of %lluk.\n",
+ window/2, (unsigned long long)max_sectors/2);
atomic_set(&mddev->recovery_active, 0);
last_check = 0;
if (j>2) {
- printk(KERN_INFO
- "md: resuming %s of %s from checkpoint.\n",
- desc, mdname(mddev));
+ pr_debug("md: resuming %s of %s from checkpoint.\n",
+ desc, mdname(mddev));
mddev->curr_resync = j;
} else
mddev->curr_resync = 3; /* no longer delayed */
j > mddev->recovery_cp)
mddev->recovery_cp = j;
update_time = jiffies;
- set_bit(MD_CHANGE_CLEAN, &mddev->flags);
+ set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
sysfs_notify(&mddev->kobj, NULL, "sync_completed");
}
}
}
}
- printk(KERN_INFO "md: %s: %s %s.\n",mdname(mddev), desc,
- test_bit(MD_RECOVERY_INTR, &mddev->recovery)
- ? "interrupted" : "done");
+ pr_info("md: %s: %s %s.\n",mdname(mddev), desc,
+ test_bit(MD_RECOVERY_INTR, &mddev->recovery)
+ ? "interrupted" : "done");
/*
* this also signals 'finished resyncing' to md_stop
*/
if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
if (test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
if (mddev->curr_resync >= mddev->recovery_cp) {
- printk(KERN_INFO
- "md: checkpointing %s of %s.\n",
- desc, mdname(mddev));
+ pr_debug("md: checkpointing %s of %s.\n",
+ desc, mdname(mddev));
if (test_bit(MD_RECOVERY_ERROR,
&mddev->recovery))
mddev->recovery_cp =
/* set CHANGE_PENDING here since maybe another update is needed,
* so other nodes are informed. It should be harmless for normal
* raid */
- set_mask_bits(&mddev->flags, 0,
- BIT(MD_CHANGE_PENDING) | BIT(MD_CHANGE_DEVS));
+ set_mask_bits(&mddev->sb_flags, 0,
+ BIT(MD_SB_CHANGE_PENDING) | BIT(MD_SB_CHANGE_DEVS));
spin_lock(&mddev->lock);
if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
if (!test_bit(Journal, &rdev->flags))
spares++;
md_new_event(mddev);
- set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
}
}
no_add:
if (removed)
- set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
return spares;
}
mddev,
"resync");
if (!mddev->sync_thread) {
- printk(KERN_ERR "%s: could not start resync thread...\n",
- mdname(mddev));
+ pr_warn("%s: could not start resync thread...\n",
+ mdname(mddev));
/* leave the spares where they are, it shouldn't hurt */
clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
clear_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
if (signal_pending(current)) {
if (mddev->pers->sync_request && !mddev->external) {
- printk(KERN_INFO "md: %s in immediate safe mode\n",
- mdname(mddev));
+ pr_debug("md: %s in immediate safe mode\n",
+ mdname(mddev));
mddev->safemode = 2;
}
flush_signals(current);
if (mddev->ro && !test_bit(MD_RECOVERY_NEEDED, &mddev->recovery))
return;
if ( ! (
- (mddev->flags & MD_UPDATE_SB_FLAGS & ~ (1<<MD_CHANGE_PENDING)) ||
+ (mddev->sb_flags & ~ (1<<MD_SB_CHANGE_PENDING)) ||
test_bit(MD_RECOVERY_NEEDED, &mddev->recovery) ||
test_bit(MD_RECOVERY_DONE, &mddev->recovery) ||
test_bit(MD_RELOAD_SB, &mddev->flags) ||
md_reap_sync_thread(mddev);
clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
- clear_bit(MD_CHANGE_PENDING, &mddev->flags);
+ clear_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags);
goto unlock;
}
mddev->recovery_cp == MaxSector) {
mddev->in_sync = 1;
did_change = 1;
- set_bit(MD_CHANGE_CLEAN, &mddev->flags);
+ set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
}
if (mddev->safemode == 1)
mddev->safemode = 0;
sysfs_notify_dirent_safe(mddev->sysfs_state);
}
- if (mddev->flags & MD_UPDATE_SB_FLAGS)
+ if (mddev->sb_flags)
md_update_sb(mddev, 0);
if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) &&
if (mddev->pers->spare_active(mddev)) {
sysfs_notify(&mddev->kobj, NULL,
"degraded");
- set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
}
}
if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) &&
rdev->saved_raid_disk = -1;
md_update_sb(mddev, 1);
- /* MD_CHANGE_PENDING should be cleared by md_update_sb, so we can
+ /* MD_SB_CHANGE_PENDING should be cleared by md_update_sb, so we can
* call resync_finish here if MD_CLUSTER_RESYNC_LOCKED is set by
* clustered raid */
if (test_and_clear_bit(MD_CLUSTER_RESYNC_LOCKED, &mddev->flags))
rv = badblocks_set(&rdev->badblocks, s, sectors, 0);
if (rv == 0) {
/* Make sure they get written out promptly */
+ if (test_bit(ExternalBbl, &rdev->flags))
+ sysfs_notify(&rdev->kobj, NULL,
+ "unacknowledged_bad_blocks");
sysfs_notify_dirent_safe(rdev->sysfs_state);
- set_mask_bits(&mddev->flags, 0,
- BIT(MD_CHANGE_CLEAN) | BIT(MD_CHANGE_PENDING));
+ set_mask_bits(&mddev->sb_flags, 0,
+ BIT(MD_SB_CHANGE_CLEAN) | BIT(MD_SB_CHANGE_PENDING));
md_wakeup_thread(rdev->mddev->thread);
return 1;
} else
int rdev_clear_badblocks(struct md_rdev *rdev, sector_t s, int sectors,
int is_new)
{
+ int rv;
if (is_new)
s += rdev->new_data_offset;
else
s += rdev->data_offset;
- return badblocks_clear(&rdev->badblocks,
- s, sectors);
+ rv = badblocks_clear(&rdev->badblocks, s, sectors);
+ if ((rv == 0) && test_bit(ExternalBbl, &rdev->flags))
+ sysfs_notify(&rdev->kobj, NULL, "bad_blocks");
+ return rv;
}
EXPORT_SYMBOL_GPL(rdev_clear_badblocks);
rdev2->saved_raid_disk = role;
ret = remove_and_add_spares(mddev, rdev2);
pr_info("Activated spare: %s\n",
- bdevname(rdev2->bdev,b));
+ bdevname(rdev2->bdev,b));
/* wakeup mddev->thread here, so array could
* perform resync with the new activated disk */
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
* variable in case we err in the future
*/
rdev->sb_page = NULL;
- alloc_disk_sb(rdev);
- ClearPageUptodate(rdev->sb_page);
- rdev->sb_loaded = 0;
- err = super_types[mddev->major_version].load_super(rdev, NULL, mddev->minor_version);
-
+ err = alloc_disk_sb(rdev);
+ if (err == 0) {
+ ClearPageUptodate(rdev->sb_page);
+ rdev->sb_loaded = 0;
+ err = super_types[mddev->major_version].
+ load_super(rdev, NULL, mddev->minor_version);
+ }
if (err < 0) {
pr_warn("%s: %d Could not reload rdev(%d) err: %d. Restoring old values\n",
__func__, __LINE__, rdev->desc_nr, err);
- put_page(rdev->sb_page);
+ if (rdev->sb_page)
+ put_page(rdev->sb_page);
rdev->sb_page = swapout;
rdev->sb_loaded = 1;
return err;
mutex_lock(&detected_devices_mutex);
list_add_tail(&node_detected_dev->list, &all_detected_devices);
mutex_unlock(&detected_devices_mutex);
- } else {
- printk(KERN_CRIT "md: md_autodetect_dev: kzalloc failed"
- ", skipping dev(%d,%d)\n", MAJOR(dev), MINOR(dev));
}
}
i_scanned = 0;
i_passed = 0;
- printk(KERN_INFO "md: Autodetecting RAID arrays.\n");
+ pr_info("md: Autodetecting RAID arrays.\n");
mutex_lock(&detected_devices_mutex);
while (!list_empty(&all_detected_devices) && i_scanned < INT_MAX) {
}
mutex_unlock(&detected_devices_mutex);
- printk(KERN_INFO "md: Scanned %d and added %d devices.\n",
- i_scanned, i_passed);
+ pr_debug("md: Scanned %d and added %d devices.\n", i_scanned, i_passed);
autorun_devices(part);
}
projects / linux.git / blobdiff