summaryrefslogtreecommitdiff
path: root/drivers/md
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/md')
-rw-r--r--drivers/md/dm-raid.c42
-rw-r--r--drivers/md/md-bitmap.c8
-rw-r--r--drivers/md/md-cluster.c16
-rw-r--r--drivers/md/md.c73
-rw-r--r--drivers/md/md.h2
-rw-r--r--drivers/md/raid0.c6
-rw-r--r--drivers/md/raid1-10.c2
-rw-r--r--drivers/md/raid1.c94
-rw-r--r--drivers/md/raid1.h22
-rw-r--r--drivers/md/raid10.c16
-rw-r--r--drivers/md/raid5-ppl.c6
-rw-r--r--drivers/md/raid5.c30
12 files changed, 144 insertions, 173 deletions
diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
index 15c538ee9537..79ea85d18e24 100644
--- a/drivers/md/dm-raid.c
+++ b/drivers/md/dm-raid.c
@@ -438,7 +438,7 @@ static bool rs_is_reshapable(struct raid_set *rs)
/* Return true, if raid set in @rs is recovering */
static bool rs_is_recovering(struct raid_set *rs)
{
- return rs->md.recovery_cp < rs->md.dev_sectors;
+ return rs->md.resync_offset < rs->md.dev_sectors;
}
/* Return true, if raid set in @rs is reshaping */
@@ -768,7 +768,7 @@ static struct raid_set *raid_set_alloc(struct dm_target *ti, struct raid_type *r
rs->md.layout = raid_type->algorithm;
rs->md.new_layout = rs->md.layout;
rs->md.delta_disks = 0;
- rs->md.recovery_cp = MaxSector;
+ rs->md.resync_offset = MaxSector;
for (i = 0; i < raid_devs; i++)
md_rdev_init(&rs->dev[i].rdev);
@@ -912,7 +912,7 @@ static int parse_dev_params(struct raid_set *rs, struct dm_arg_set *as)
rs->md.external = 0;
rs->md.persistent = 1;
rs->md.major_version = 2;
- } else if (rebuild && !rs->md.recovery_cp) {
+ } else if (rebuild && !rs->md.resync_offset) {
/*
* Without metadata, we will not be able to tell if the array
* is in-sync or not - we must assume it is not. Therefore,
@@ -1695,20 +1695,20 @@ static void rs_setup_recovery(struct raid_set *rs, sector_t dev_sectors)
{
/* raid0 does not recover */
if (rs_is_raid0(rs))
- rs->md.recovery_cp = MaxSector;
+ rs->md.resync_offset = MaxSector;
/*
* A raid6 set has to be recovered either
* completely or for the grown part to
* ensure proper parity and Q-Syndrome
*/
else if (rs_is_raid6(rs))
- rs->md.recovery_cp = dev_sectors;
+ rs->md.resync_offset = dev_sectors;
/*
* Other raid set types may skip recovery
* depending on the 'nosync' flag.
*/
else
- rs->md.recovery_cp = test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)
+ rs->md.resync_offset = test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)
? MaxSector : dev_sectors;
}
@@ -2143,7 +2143,7 @@ static void super_sync(struct mddev *mddev, struct md_rdev *rdev)
sb->events = cpu_to_le64(mddev->events);
sb->disk_recovery_offset = cpu_to_le64(rdev->recovery_offset);
- sb->array_resync_offset = cpu_to_le64(mddev->recovery_cp);
+ sb->array_resync_offset = cpu_to_le64(mddev->resync_offset);
sb->level = cpu_to_le32(mddev->level);
sb->layout = cpu_to_le32(mddev->layout);
@@ -2334,18 +2334,18 @@ static int super_init_validation(struct raid_set *rs, struct md_rdev *rdev)
}
if (!test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags))
- mddev->recovery_cp = le64_to_cpu(sb->array_resync_offset);
+ mddev->resync_offset = le64_to_cpu(sb->array_resync_offset);
/*
* During load, we set FirstUse if a new superblock was written.
* There are two reasons we might not have a superblock:
* 1) The raid set is brand new - in which case, all of the
* devices must have their In_sync bit set. Also,
- * recovery_cp must be 0, unless forced.
+ * resync_offset must be 0, unless forced.
* 2) This is a new device being added to an old raid set
* and the new device needs to be rebuilt - in which
* case the In_sync bit will /not/ be set and
- * recovery_cp must be MaxSector.
+ * resync_offset must be MaxSector.
* 3) This is/are a new device(s) being added to an old
* raid set during takeover to a higher raid level
* to provide capacity for redundancy or during reshape
@@ -2390,8 +2390,8 @@ static int super_init_validation(struct raid_set *rs, struct md_rdev *rdev)
new_devs > 1 ? "s" : "");
return -EINVAL;
} else if (!test_bit(__CTR_FLAG_REBUILD, &rs->ctr_flags) && rs_is_recovering(rs)) {
- DMERR("'rebuild' specified while raid set is not in-sync (recovery_cp=%llu)",
- (unsigned long long) mddev->recovery_cp);
+ DMERR("'rebuild' specified while raid set is not in-sync (resync_offset=%llu)",
+ (unsigned long long) mddev->resync_offset);
return -EINVAL;
} else if (rs_is_reshaping(rs)) {
DMERR("'rebuild' specified while raid set is being reshaped (reshape_position=%llu)",
@@ -2700,11 +2700,11 @@ static int rs_adjust_data_offsets(struct raid_set *rs)
}
out:
/*
- * Raise recovery_cp in case data_offset != 0 to
+ * Raise resync_offset in case data_offset != 0 to
* avoid false recovery positives in the constructor.
*/
- if (rs->md.recovery_cp < rs->md.dev_sectors)
- rs->md.recovery_cp += rs->dev[0].rdev.data_offset;
+ if (rs->md.resync_offset < rs->md.dev_sectors)
+ rs->md.resync_offset += rs->dev[0].rdev.data_offset;
/* Adjust data offsets on all rdevs but on any raid4/5/6 journal device */
rdev_for_each(rdev, &rs->md) {
@@ -2759,7 +2759,7 @@ static int rs_setup_takeover(struct raid_set *rs)
}
clear_bit(MD_ARRAY_FIRST_USE, &mddev->flags);
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
while (d--) {
rdev = &rs->dev[d].rdev;
@@ -2767,7 +2767,7 @@ static int rs_setup_takeover(struct raid_set *rs)
if (test_bit(d, (void *) rs->rebuild_disks)) {
clear_bit(In_sync, &rdev->flags);
clear_bit(Faulty, &rdev->flags);
- mddev->recovery_cp = rdev->recovery_offset = 0;
+ mddev->resync_offset = rdev->recovery_offset = 0;
/* Bitmap has to be created when we do an "up" takeover */
set_bit(MD_ARRAY_FIRST_USE, &mddev->flags);
}
@@ -3225,7 +3225,7 @@ size_check:
if (r)
goto bad;
- rs_setup_recovery(rs, rs->md.recovery_cp < rs->md.dev_sectors ? rs->md.recovery_cp : rs->md.dev_sectors);
+ rs_setup_recovery(rs, rs->md.resync_offset < rs->md.dev_sectors ? rs->md.resync_offset : rs->md.dev_sectors);
} else {
/* This is no size change or it is shrinking, update size and record in superblocks */
r = rs_set_dev_and_array_sectors(rs, rs->ti->len, false);
@@ -3449,7 +3449,7 @@ static sector_t rs_get_progress(struct raid_set *rs, unsigned long recovery,
} else {
if (state == st_idle && !test_bit(MD_RECOVERY_INTR, &recovery))
- r = mddev->recovery_cp;
+ r = mddev->resync_offset;
else
r = mddev->curr_resync_completed;
@@ -4077,9 +4077,9 @@ static int raid_preresume(struct dm_target *ti)
}
/* Check for any resize/reshape on @rs and adjust/initiate */
- if (mddev->recovery_cp && mddev->recovery_cp < MaxSector) {
+ if (mddev->resync_offset && mddev->resync_offset < MaxSector) {
set_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
- mddev->resync_min = mddev->recovery_cp;
+ mddev->resync_min = mddev->resync_offset;
if (test_bit(RT_FLAG_RS_GROW, &rs->runtime_flags))
mddev->resync_max_sectors = mddev->dev_sectors;
}
diff --git a/drivers/md/md-bitmap.c b/drivers/md/md-bitmap.c
index 7f524a26cebc..334b71404930 100644
--- a/drivers/md/md-bitmap.c
+++ b/drivers/md/md-bitmap.c
@@ -1987,12 +1987,12 @@ static void bitmap_dirty_bits(struct mddev *mddev, unsigned long s,
md_bitmap_set_memory_bits(bitmap, sec, 1);
md_bitmap_file_set_bit(bitmap, sec);
- if (sec < bitmap->mddev->recovery_cp)
+ if (sec < bitmap->mddev->resync_offset)
/* We are asserting that the array is dirty,
- * so move the recovery_cp address back so
+ * so move the resync_offset address back so
* that it is obvious that it is dirty
*/
- bitmap->mddev->recovery_cp = sec;
+ bitmap->mddev->resync_offset = sec;
}
}
@@ -2258,7 +2258,7 @@ static int bitmap_load(struct mddev *mddev)
|| bitmap->events_cleared == mddev->events)
/* no need to keep dirty bits to optimise a
* re-add of a missing device */
- start = mddev->recovery_cp;
+ start = mddev->resync_offset;
mutex_lock(&mddev->bitmap_info.mutex);
err = md_bitmap_init_from_disk(bitmap, start);
diff --git a/drivers/md/md-cluster.c b/drivers/md/md-cluster.c
index 94221d964d4f..5497eaee96e7 100644
--- a/drivers/md/md-cluster.c
+++ b/drivers/md/md-cluster.c
@@ -337,11 +337,11 @@ static void recover_bitmaps(struct md_thread *thread)
md_wakeup_thread(mddev->sync_thread);
if (hi > 0) {
- if (lo < mddev->recovery_cp)
- mddev->recovery_cp = lo;
+ if (lo < mddev->resync_offset)
+ mddev->resync_offset = lo;
/* wake up thread to continue resync in case resync
* is not finished */
- if (mddev->recovery_cp != MaxSector) {
+ if (mddev->resync_offset != MaxSector) {
/*
* clear the REMOTE flag since we will launch
* resync thread in current node.
@@ -863,9 +863,9 @@ static int gather_all_resync_info(struct mddev *mddev, int total_slots)
lockres_free(bm_lockres);
continue;
}
- if ((hi > 0) && (lo < mddev->recovery_cp)) {
+ if ((hi > 0) && (lo < mddev->resync_offset)) {
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
- mddev->recovery_cp = lo;
+ mddev->resync_offset = lo;
md_check_recovery(mddev);
}
@@ -1027,7 +1027,7 @@ static int leave(struct mddev *mddev)
* Also, we should send BITMAP_NEEDS_SYNC message in
* case reshaping is interrupted.
*/
- if ((cinfo->slot_number > 0 && mddev->recovery_cp != MaxSector) ||
+ if ((cinfo->slot_number > 0 && mddev->resync_offset != MaxSector) ||
(mddev->reshape_position != MaxSector &&
test_bit(MD_CLOSING, &mddev->flags)))
resync_bitmap(mddev);
@@ -1605,8 +1605,8 @@ static int gather_bitmaps(struct md_rdev *rdev)
pr_warn("md-cluster: Could not gather bitmaps from slot %d", sn);
goto out;
}
- if ((hi > 0) && (lo < mddev->recovery_cp))
- mddev->recovery_cp = lo;
+ if ((hi > 0) && (lo < mddev->resync_offset))
+ mddev->resync_offset = lo;
}
out:
return err;
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 046fe85c76fe..ac85ec73a409 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -637,6 +637,12 @@ static void __mddev_put(struct mddev *mddev)
return;
/*
+ * If array is freed by stopping array, MD_DELETED is set by
+ * do_md_stop(), MD_DELETED is still set here in case mddev is freed
+ * directly by closing a mddev that is created by create_on_open.
+ */
+ set_bit(MD_DELETED, &mddev->flags);
+ /*
* Call queue_work inside the spinlock so that flush_workqueue() after
* mddev_find will succeed in waiting for the work to be done.
*/
@@ -1409,13 +1415,13 @@ static int super_90_validate(struct mddev *mddev, struct md_rdev *freshest, stru
mddev->layout = -1;
if (sb->state & (1<<MD_SB_CLEAN))
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
else {
if (sb->events_hi == sb->cp_events_hi &&
sb->events_lo == sb->cp_events_lo) {
- mddev->recovery_cp = sb->recovery_cp;
+ mddev->resync_offset = sb->resync_offset;
} else
- mddev->recovery_cp = 0;
+ mddev->resync_offset = 0;
}
memcpy(mddev->uuid+0, &sb->set_uuid0, 4);
@@ -1541,13 +1547,13 @@ static void super_90_sync(struct mddev *mddev, struct md_rdev *rdev)
mddev->minor_version = sb->minor_version;
if (mddev->in_sync)
{
- sb->recovery_cp = mddev->recovery_cp;
+ sb->resync_offset = mddev->resync_offset;
sb->cp_events_hi = (mddev->events>>32);
sb->cp_events_lo = (u32)mddev->events;
- if (mddev->recovery_cp == MaxSector)
+ if (mddev->resync_offset == MaxSector)
sb->state = (1<< MD_SB_CLEAN);
} else
- sb->recovery_cp = 0;
+ sb->resync_offset = 0;
sb->layout = mddev->layout;
sb->chunk_size = mddev->chunk_sectors << 9;
@@ -1895,7 +1901,7 @@ static int super_1_validate(struct mddev *mddev, struct md_rdev *freshest, struc
mddev->bitmap_info.default_space = (4096-1024) >> 9;
mddev->reshape_backwards = 0;
- mddev->recovery_cp = le64_to_cpu(sb->resync_offset);
+ mddev->resync_offset = le64_to_cpu(sb->resync_offset);
memcpy(mddev->uuid, sb->set_uuid, 16);
mddev->max_disks = (4096-256)/2;
@@ -2081,7 +2087,7 @@ static void super_1_sync(struct mddev *mddev, struct md_rdev *rdev)
sb->utime = cpu_to_le64((__u64)mddev->utime);
sb->events = cpu_to_le64(mddev->events);
if (mddev->in_sync)
- sb->resync_offset = cpu_to_le64(mddev->recovery_cp);
+ sb->resync_offset = cpu_to_le64(mddev->resync_offset);
else if (test_bit(MD_JOURNAL_CLEAN, &mddev->flags))
sb->resync_offset = cpu_to_le64(MaxSector);
else
@@ -2761,7 +2767,7 @@ repeat:
/* 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 (nospares
- && (mddev->in_sync && mddev->recovery_cp == MaxSector)
+ && (mddev->in_sync && mddev->resync_offset == MaxSector)
&& mddev->can_decrease_events
&& mddev->events != 1) {
mddev->events--;
@@ -4297,9 +4303,9 @@ __ATTR(chunk_size, S_IRUGO|S_IWUSR, chunk_size_show, chunk_size_store);
static ssize_t
resync_start_show(struct mddev *mddev, char *page)
{
- if (mddev->recovery_cp == MaxSector)
+ if (mddev->resync_offset == MaxSector)
return sprintf(page, "none\n");
- return sprintf(page, "%llu\n", (unsigned long long)mddev->recovery_cp);
+ return sprintf(page, "%llu\n", (unsigned long long)mddev->resync_offset);
}
static ssize_t
@@ -4325,7 +4331,7 @@ resync_start_store(struct mddev *mddev, const char *buf, size_t len)
err = -EBUSY;
if (!err) {
- mddev->recovery_cp = n;
+ mddev->resync_offset = n;
if (mddev->pers)
set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
}
@@ -6417,7 +6423,7 @@ static void md_clean(struct mddev *mddev)
mddev->external_size = 0;
mddev->dev_sectors = 0;
mddev->raid_disks = 0;
- mddev->recovery_cp = 0;
+ mddev->resync_offset = 0;
mddev->resync_min = 0;
mddev->resync_max = MaxSector;
mddev->reshape_position = MaxSector;
@@ -7362,9 +7368,9 @@ int md_set_array_info(struct mddev *mddev, struct mdu_array_info_s *info)
* openned
*/
if (info->state & (1<<MD_SB_CLEAN))
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
else
- mddev->recovery_cp = 0;
+ mddev->resync_offset = 0;
mddev->persistent = ! info->not_persistent;
mddev->external = 0;
@@ -8303,7 +8309,7 @@ static int status_resync(struct seq_file *seq, struct mddev *mddev)
seq_printf(seq, "\tresync=REMOTE");
return 1;
}
- if (mddev->recovery_cp < MaxSector) {
+ if (mddev->resync_offset < MaxSector) {
seq_printf(seq, "\tresync=PENDING");
return 1;
}
@@ -8946,7 +8952,7 @@ static sector_t md_sync_position(struct mddev *mddev, enum sync_action action)
return mddev->resync_min;
case ACTION_RESYNC:
if (!mddev->bitmap)
- return mddev->recovery_cp;
+ return mddev->resync_offset;
return 0;
case ACTION_RESHAPE:
/*
@@ -9184,8 +9190,8 @@ void md_do_sync(struct md_thread *thread)
atomic_read(&mddev->recovery_active) == 0);
mddev->curr_resync_completed = j;
if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) &&
- j > mddev->recovery_cp)
- mddev->recovery_cp = j;
+ j > mddev->resync_offset)
+ mddev->resync_offset = j;
update_time = jiffies;
set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
sysfs_notify_dirent_safe(mddev->sysfs_completed);
@@ -9305,19 +9311,19 @@ void md_do_sync(struct md_thread *thread)
mddev->curr_resync > MD_RESYNC_ACTIVE) {
if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
if (test_bit(MD_RECOVERY_INTR, &mddev->recovery)) {
- if (mddev->curr_resync >= mddev->recovery_cp) {
+ if (mddev->curr_resync >= mddev->resync_offset) {
pr_debug("md: checkpointing %s of %s.\n",
desc, mdname(mddev));
if (test_bit(MD_RECOVERY_ERROR,
&mddev->recovery))
- mddev->recovery_cp =
+ mddev->resync_offset =
mddev->curr_resync_completed;
else
- mddev->recovery_cp =
+ mddev->resync_offset =
mddev->curr_resync;
}
} else
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
} else {
if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery))
mddev->curr_resync = MaxSector;
@@ -9421,6 +9427,12 @@ static bool rdev_is_spare(struct md_rdev *rdev)
static bool rdev_addable(struct md_rdev *rdev)
{
+ struct mddev *mddev;
+
+ mddev = READ_ONCE(rdev->mddev);
+ if (!mddev)
+ return false;
+
/* rdev is already used, don't add it again. */
if (test_bit(Candidate, &rdev->flags) || rdev->raid_disk >= 0 ||
test_bit(Faulty, &rdev->flags))
@@ -9431,7 +9443,7 @@ static bool rdev_addable(struct md_rdev *rdev)
return true;
/* Allow to add if array is read-write. */
- if (md_is_rdwr(rdev->mddev))
+ if (md_is_rdwr(mddev))
return true;
/*
@@ -9533,7 +9545,7 @@ static bool md_choose_sync_action(struct mddev *mddev, int *spares)
}
/* Check if resync is in progress. */
- if (mddev->recovery_cp < MaxSector) {
+ if (mddev->resync_offset < MaxSector) {
remove_spares(mddev, NULL);
set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
@@ -9714,7 +9726,7 @@ void md_check_recovery(struct mddev *mddev)
test_bit(MD_RECOVERY_DONE, &mddev->recovery) ||
(mddev->external == 0 && mddev->safemode == 1) ||
(mddev->safemode == 2
- && !mddev->in_sync && mddev->recovery_cp == MaxSector)
+ && !mddev->in_sync && mddev->resync_offset == MaxSector)
))
return;
@@ -9771,8 +9783,8 @@ void md_check_recovery(struct mddev *mddev)
* remove disk.
*/
rdev_for_each_safe(rdev, tmp, mddev) {
- if (test_and_clear_bit(ClusterRemove, &rdev->flags) &&
- rdev->raid_disk < 0)
+ if (rdev->raid_disk < 0 &&
+ test_and_clear_bit(ClusterRemove, &rdev->flags))
md_kick_rdev_from_array(rdev);
}
}
@@ -10078,8 +10090,11 @@ static void check_sb_changes(struct mddev *mddev, struct md_rdev *rdev)
/* Check for change of roles in the active devices */
rdev_for_each_safe(rdev2, tmp, mddev) {
- if (test_bit(Faulty, &rdev2->flags))
+ if (test_bit(Faulty, &rdev2->flags)) {
+ if (test_bit(ClusterRemove, &rdev2->flags))
+ set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
continue;
+ }
/* Check if the roles changed */
role = le16_to_cpu(sb->dev_roles[rdev2->desc_nr]);
diff --git a/drivers/md/md.h b/drivers/md/md.h
index 67b365621507..51af29a03079 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -523,7 +523,7 @@ struct mddev {
unsigned long normal_io_events; /* IO event timestamp */
atomic_t recovery_active; /* blocks scheduled, but not written */
wait_queue_head_t recovery_wait;
- sector_t recovery_cp;
+ sector_t resync_offset;
sector_t resync_min; /* user requested sync
* starts here */
sector_t resync_max; /* resync should pause
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index cbe2a9054cb9..f1d8811a542a 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -674,7 +674,7 @@ static void *raid0_takeover_raid45(struct mddev *mddev)
mddev->raid_disks--;
mddev->delta_disks = -1;
/* make sure it will be not marked as dirty */
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
mddev_clear_unsupported_flags(mddev, UNSUPPORTED_MDDEV_FLAGS);
create_strip_zones(mddev, &priv_conf);
@@ -717,7 +717,7 @@ static void *raid0_takeover_raid10(struct mddev *mddev)
mddev->raid_disks += mddev->delta_disks;
mddev->degraded = 0;
/* make sure it will be not marked as dirty */
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
mddev_clear_unsupported_flags(mddev, UNSUPPORTED_MDDEV_FLAGS);
create_strip_zones(mddev, &priv_conf);
@@ -760,7 +760,7 @@ static void *raid0_takeover_raid1(struct mddev *mddev)
mddev->delta_disks = 1 - mddev->raid_disks;
mddev->raid_disks = 1;
/* make sure it will be not marked as dirty */
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
mddev_clear_unsupported_flags(mddev, UNSUPPORTED_MDDEV_FLAGS);
create_strip_zones(mddev, &priv_conf);
diff --git a/drivers/md/raid1-10.c b/drivers/md/raid1-10.c
index b8b3a9069701..52881e6032da 100644
--- a/drivers/md/raid1-10.c
+++ b/drivers/md/raid1-10.c
@@ -283,7 +283,7 @@ static inline int raid1_check_read_range(struct md_rdev *rdev,
static inline bool raid1_should_read_first(struct mddev *mddev,
sector_t this_sector, int len)
{
- if ((mddev->recovery_cp < this_sector + len))
+ if ((mddev->resync_offset < this_sector + len))
return true;
if (mddev_is_clustered(mddev) &&
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 64b8176907a9..408c26398321 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -127,10 +127,9 @@ static inline struct r1bio *get_resync_r1bio(struct bio *bio)
return get_resync_pages(bio)->raid_bio;
}
-static void * r1bio_pool_alloc(gfp_t gfp_flags, void *data)
+static void *r1bio_pool_alloc(gfp_t gfp_flags, struct r1conf *conf)
{
- struct pool_info *pi = data;
- int size = offsetof(struct r1bio, bios[pi->raid_disks]);
+ int size = offsetof(struct r1bio, bios[conf->raid_disks * 2]);
/* allocate a r1bio with room for raid_disks entries in the bios array */
return kzalloc(size, gfp_flags);
@@ -145,18 +144,18 @@ static void * r1bio_pool_alloc(gfp_t gfp_flags, void *data)
static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
{
- struct pool_info *pi = data;
+ struct r1conf *conf = data;
struct r1bio *r1_bio;
struct bio *bio;
int need_pages;
int j;
struct resync_pages *rps;
- r1_bio = r1bio_pool_alloc(gfp_flags, pi);
+ r1_bio = r1bio_pool_alloc(gfp_flags, conf);
if (!r1_bio)
return NULL;
- rps = kmalloc_array(pi->raid_disks, sizeof(struct resync_pages),
+ rps = kmalloc_array(conf->raid_disks * 2, sizeof(struct resync_pages),
gfp_flags);
if (!rps)
goto out_free_r1bio;
@@ -164,7 +163,7 @@ static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
/*
* Allocate bios : 1 for reading, n-1 for writing
*/
- for (j = pi->raid_disks ; j-- ; ) {
+ for (j = conf->raid_disks * 2; j-- ; ) {
bio = bio_kmalloc(RESYNC_PAGES, gfp_flags);
if (!bio)
goto out_free_bio;
@@ -177,11 +176,11 @@ static void * r1buf_pool_alloc(gfp_t gfp_flags, void *data)
* If this is a user-requested check/repair, allocate
* RESYNC_PAGES for each bio.
*/
- if (test_bit(MD_RECOVERY_REQUESTED, &pi->mddev->recovery))
- need_pages = pi->raid_disks;
+ if (test_bit(MD_RECOVERY_REQUESTED, &conf->mddev->recovery))
+ need_pages = conf->raid_disks * 2;
else
need_pages = 1;
- for (j = 0; j < pi->raid_disks; j++) {
+ for (j = 0; j < conf->raid_disks * 2; j++) {
struct resync_pages *rp = &rps[j];
bio = r1_bio->bios[j];
@@ -207,7 +206,7 @@ out_free_pages:
resync_free_pages(&rps[j]);
out_free_bio:
- while (++j < pi->raid_disks) {
+ while (++j < conf->raid_disks * 2) {
bio_uninit(r1_bio->bios[j]);
kfree(r1_bio->bios[j]);
}
@@ -220,12 +219,12 @@ out_free_r1bio:
static void r1buf_pool_free(void *__r1_bio, void *data)
{
- struct pool_info *pi = data;
+ struct r1conf *conf = data;
int i;
struct r1bio *r1bio = __r1_bio;
struct resync_pages *rp = NULL;
- for (i = pi->raid_disks; i--; ) {
+ for (i = conf->raid_disks * 2; i--; ) {
rp = get_resync_pages(r1bio->bios[i]);
resync_free_pages(rp);
bio_uninit(r1bio->bios[i]);
@@ -255,7 +254,7 @@ static void free_r1bio(struct r1bio *r1_bio)
struct r1conf *conf = r1_bio->mddev->private;
put_all_bios(conf, r1_bio);
- mempool_free(r1_bio, &conf->r1bio_pool);
+ mempool_free(r1_bio, conf->r1bio_pool);
}
static void put_buf(struct r1bio *r1_bio)
@@ -1305,9 +1304,8 @@ alloc_r1bio(struct mddev *mddev, struct bio *bio)
struct r1conf *conf = mddev->private;
struct r1bio *r1_bio;
- r1_bio = mempool_alloc(&conf->r1bio_pool, GFP_NOIO);
- /* Ensure no bio records IO_BLOCKED */
- memset(r1_bio->bios, 0, conf->raid_disks * sizeof(r1_bio->bios[0]));
+ r1_bio = mempool_alloc(conf->r1bio_pool, GFP_NOIO);
+ memset(r1_bio, 0, offsetof(struct r1bio, bios[conf->raid_disks * 2]));
init_r1bio(r1_bio, mddev, bio);
return r1_bio;
}
@@ -2747,7 +2745,7 @@ static int init_resync(struct r1conf *conf)
BUG_ON(mempool_initialized(&conf->r1buf_pool));
return mempool_init(&conf->r1buf_pool, buffs, r1buf_pool_alloc,
- r1buf_pool_free, conf->poolinfo);
+ r1buf_pool_free, conf);
}
static struct r1bio *raid1_alloc_init_r1buf(struct r1conf *conf)
@@ -2757,7 +2755,7 @@ static struct r1bio *raid1_alloc_init_r1buf(struct r1conf *conf)
struct bio *bio;
int i;
- for (i = conf->poolinfo->raid_disks; i--; ) {
+ for (i = conf->raid_disks * 2; i--; ) {
bio = r1bio->bios[i];
rps = bio->bi_private;
bio_reset(bio, NULL, 0);
@@ -2822,7 +2820,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr,
}
if (mddev->bitmap == NULL &&
- mddev->recovery_cp == MaxSector &&
+ mddev->resync_offset == MaxSector &&
!test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery) &&
conf->fullsync == 0) {
*skipped = 1;
@@ -3085,6 +3083,7 @@ static struct r1conf *setup_conf(struct mddev *mddev)
int i;
struct raid1_info *disk;
struct md_rdev *rdev;
+ size_t r1bio_size;
int err = -ENOMEM;
conf = kzalloc(sizeof(struct r1conf), GFP_KERNEL);
@@ -3121,21 +3120,15 @@ static struct r1conf *setup_conf(struct mddev *mddev)
if (!conf->tmppage)
goto abort;
- conf->poolinfo = kzalloc(sizeof(*conf->poolinfo), GFP_KERNEL);
- if (!conf->poolinfo)
- goto abort;
- conf->poolinfo->raid_disks = mddev->raid_disks * 2;
- err = mempool_init(&conf->r1bio_pool, NR_RAID_BIOS, r1bio_pool_alloc,
- rbio_pool_free, conf->poolinfo);
- if (err)
+ r1bio_size = offsetof(struct r1bio, bios[mddev->raid_disks * 2]);
+ conf->r1bio_pool = mempool_create_kmalloc_pool(NR_RAID_BIOS, r1bio_size);
+ if (!conf->r1bio_pool)
goto abort;
err = bioset_init(&conf->bio_split, BIO_POOL_SIZE, 0, 0);
if (err)
goto abort;
- conf->poolinfo->mddev = mddev;
-
err = -EINVAL;
spin_lock_init(&conf->device_lock);
conf->raid_disks = mddev->raid_disks;
@@ -3198,10 +3191,9 @@ static struct r1conf *setup_conf(struct mddev *mddev)
abort:
if (conf) {
- mempool_exit(&conf->r1bio_pool);
+ mempool_destroy(conf->r1bio_pool);
kfree(conf->mirrors);
safe_put_page(conf->tmppage);
- kfree(conf->poolinfo);
kfree(conf->nr_pending);
kfree(conf->nr_waiting);
kfree(conf->nr_queued);
@@ -3282,9 +3274,9 @@ static int raid1_run(struct mddev *mddev)
}
if (conf->raid_disks - mddev->degraded == 1)
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
- if (mddev->recovery_cp != MaxSector)
+ if (mddev->resync_offset != MaxSector)
pr_info("md/raid1:%s: not clean -- starting background reconstruction\n",
mdname(mddev));
pr_info("md/raid1:%s: active with %d out of %d mirrors\n",
@@ -3311,10 +3303,9 @@ static void raid1_free(struct mddev *mddev, void *priv)
{
struct r1conf *conf = priv;
- mempool_exit(&conf->r1bio_pool);
+ mempool_destroy(conf->r1bio_pool);
kfree(conf->mirrors);
safe_put_page(conf->tmppage);
- kfree(conf->poolinfo);
kfree(conf->nr_pending);
kfree(conf->nr_waiting);
kfree(conf->nr_queued);
@@ -3345,8 +3336,8 @@ static int raid1_resize(struct mddev *mddev, sector_t sectors)
md_set_array_sectors(mddev, newsize);
if (sectors > mddev->dev_sectors &&
- mddev->recovery_cp > mddev->dev_sectors) {
- mddev->recovery_cp = mddev->dev_sectors;
+ mddev->resync_offset > mddev->dev_sectors) {
+ mddev->resync_offset = mddev->dev_sectors;
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
}
mddev->dev_sectors = sectors;
@@ -3367,17 +3358,13 @@ static int raid1_reshape(struct mddev *mddev)
* At the same time, we "pack" the devices so that all the missing
* devices have the higher raid_disk numbers.
*/
- mempool_t newpool, oldpool;
- struct pool_info *newpoolinfo;
+ mempool_t *newpool, *oldpool;
+ size_t new_r1bio_size;
struct raid1_info *newmirrors;
struct r1conf *conf = mddev->private;
int cnt, raid_disks;
unsigned long flags;
int d, d2;
- int ret;
-
- memset(&newpool, 0, sizeof(newpool));
- memset(&oldpool, 0, sizeof(oldpool));
/* Cannot change chunk_size, layout, or level */
if (mddev->chunk_sectors != mddev->new_chunk_sectors ||
@@ -3403,24 +3390,16 @@ static int raid1_reshape(struct mddev *mddev)
return -EBUSY;
}
- newpoolinfo = kmalloc(sizeof(*newpoolinfo), GFP_KERNEL);
- if (!newpoolinfo)
+ new_r1bio_size = offsetof(struct r1bio, bios[raid_disks * 2]);
+ newpool = mempool_create_kmalloc_pool(NR_RAID_BIOS, new_r1bio_size);
+ if (!newpool) {
return -ENOMEM;
- newpoolinfo->mddev = mddev;
- newpoolinfo->raid_disks = raid_disks * 2;
-
- ret = mempool_init(&newpool, NR_RAID_BIOS, r1bio_pool_alloc,
- rbio_pool_free, newpoolinfo);
- if (ret) {
- kfree(newpoolinfo);
- return ret;
}
newmirrors = kzalloc(array3_size(sizeof(struct raid1_info),
raid_disks, 2),
GFP_KERNEL);
if (!newmirrors) {
- kfree(newpoolinfo);
- mempool_exit(&newpool);
+ mempool_destroy(newpool);
return -ENOMEM;
}
@@ -3429,7 +3408,6 @@ static int raid1_reshape(struct mddev *mddev)
/* ok, everything is stopped */
oldpool = conf->r1bio_pool;
conf->r1bio_pool = newpool;
- init_waitqueue_head(&conf->r1bio_pool.wait);
for (d = d2 = 0; d < conf->raid_disks; d++) {
struct md_rdev *rdev = conf->mirrors[d].rdev;
@@ -3446,8 +3424,6 @@ static int raid1_reshape(struct mddev *mddev)
}
kfree(conf->mirrors);
conf->mirrors = newmirrors;
- kfree(conf->poolinfo);
- conf->poolinfo = newpoolinfo;
spin_lock_irqsave(&conf->device_lock, flags);
mddev->degraded += (raid_disks - conf->raid_disks);
@@ -3461,7 +3437,7 @@ static int raid1_reshape(struct mddev *mddev)
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
md_wakeup_thread(mddev->thread);
- mempool_exit(&oldpool);
+ mempool_destroy(oldpool);
return 0;
}
diff --git a/drivers/md/raid1.h b/drivers/md/raid1.h
index 33f318fcc268..d236ef179cfb 100644
--- a/drivers/md/raid1.h
+++ b/drivers/md/raid1.h
@@ -49,22 +49,6 @@ struct raid1_info {
sector_t seq_start;
};
-/*
- * memory pools need a pointer to the mddev, so they can force an unplug
- * when memory is tight, and a count of the number of drives that the
- * pool was allocated for, so they know how much to allocate and free.
- * mddev->raid_disks cannot be used, as it can change while a pool is active
- * These two datums are stored in a kmalloced struct.
- * The 'raid_disks' here is twice the raid_disks in r1conf.
- * This allows space for each 'real' device can have a replacement in the
- * second half of the array.
- */
-
-struct pool_info {
- struct mddev *mddev;
- int raid_disks;
-};
-
struct r1conf {
struct mddev *mddev;
struct raid1_info *mirrors; /* twice 'raid_disks' to
@@ -114,11 +98,7 @@ struct r1conf {
*/
int recovery_disabled;
- /* poolinfo contains information about the content of the
- * mempools - it changes when the array grows or shrinks
- */
- struct pool_info *poolinfo;
- mempool_t r1bio_pool;
+ mempool_t *r1bio_pool;
mempool_t r1buf_pool;
struct bio_set bio_split;
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 95dc354a86a0..b60c30bfb6c7 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -2117,7 +2117,7 @@ static int raid10_add_disk(struct mddev *mddev, struct md_rdev *rdev)
int last = conf->geo.raid_disks - 1;
struct raid10_info *p;
- if (mddev->recovery_cp < MaxSector)
+ if (mddev->resync_offset < MaxSector)
/* only hot-add to in-sync arrays, as recovery is
* very different from resync
*/
@@ -3185,7 +3185,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
* of a clean array, like RAID1 does.
*/
if (mddev->bitmap == NULL &&
- mddev->recovery_cp == MaxSector &&
+ mddev->resync_offset == MaxSector &&
mddev->reshape_position == MaxSector &&
!test_bit(MD_RECOVERY_SYNC, &mddev->recovery) &&
!test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery) &&
@@ -4145,7 +4145,7 @@ static int raid10_run(struct mddev *mddev)
disk->recovery_disabled = mddev->recovery_disabled - 1;
}
- if (mddev->recovery_cp != MaxSector)
+ if (mddev->resync_offset != MaxSector)
pr_notice("md/raid10:%s: not clean -- starting background reconstruction\n",
mdname(mddev));
pr_info("md/raid10:%s: active with %d out of %d devices\n",
@@ -4245,8 +4245,8 @@ static int raid10_resize(struct mddev *mddev, sector_t sectors)
md_set_array_sectors(mddev, size);
if (sectors > mddev->dev_sectors &&
- mddev->recovery_cp > oldsize) {
- mddev->recovery_cp = oldsize;
+ mddev->resync_offset > oldsize) {
+ mddev->resync_offset = oldsize;
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
}
calc_sectors(conf, sectors);
@@ -4275,7 +4275,7 @@ static void *raid10_takeover_raid0(struct mddev *mddev, sector_t size, int devs)
mddev->delta_disks = mddev->raid_disks;
mddev->raid_disks *= 2;
/* make sure it will be not marked as dirty */
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
mddev->dev_sectors = size;
conf = setup_conf(mddev);
@@ -5087,8 +5087,8 @@ static void raid10_finish_reshape(struct mddev *mddev)
return;
if (mddev->delta_disks > 0) {
- if (mddev->recovery_cp > mddev->resync_max_sectors) {
- mddev->recovery_cp = mddev->resync_max_sectors;
+ if (mddev->resync_offset > mddev->resync_max_sectors) {
+ mddev->resync_offset = mddev->resync_max_sectors;
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
}
mddev->resync_max_sectors = mddev->array_sectors;
diff --git a/drivers/md/raid5-ppl.c b/drivers/md/raid5-ppl.c
index c0fb335311aa..56b234683ee6 100644
--- a/drivers/md/raid5-ppl.c
+++ b/drivers/md/raid5-ppl.c
@@ -1163,7 +1163,7 @@ static int ppl_load_distributed(struct ppl_log *log)
le64_to_cpu(pplhdr->generation));
/* attempt to recover from log if we are starting a dirty array */
- if (pplhdr && !mddev->pers && mddev->recovery_cp != MaxSector)
+ if (pplhdr && !mddev->pers && mddev->resync_offset != MaxSector)
ret = ppl_recover(log, pplhdr, pplhdr_offset);
/* write empty header if we are starting the array */
@@ -1422,14 +1422,14 @@ int ppl_init_log(struct r5conf *conf)
if (ret) {
goto err;
- } else if (!mddev->pers && mddev->recovery_cp == 0 &&
+ } else if (!mddev->pers && mddev->resync_offset == 0 &&
ppl_conf->recovered_entries > 0 &&
ppl_conf->mismatch_count == 0) {
/*
* If we are starting a dirty array and the recovery succeeds
* without any issues, set the array as clean.
*/
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
set_bit(MD_SB_CHANGE_CLEAN, &mddev->sb_flags);
} else if (mddev->pers && ppl_conf->mismatch_count > 0) {
/* no mismatch allowed when enabling PPL for a running array */
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 7ec61ee7b218..023649fe2476 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3740,7 +3740,7 @@ static int want_replace(struct stripe_head *sh, int disk_idx)
&& !test_bit(Faulty, &rdev->flags)
&& !test_bit(In_sync, &rdev->flags)
&& (rdev->recovery_offset <= sh->sector
- || rdev->mddev->recovery_cp <= sh->sector))
+ || rdev->mddev->resync_offset <= sh->sector))
rv = 1;
return rv;
}
@@ -3832,7 +3832,7 @@ static int need_this_block(struct stripe_head *sh, struct stripe_head_state *s,
* is missing/faulty, then we need to read everything we can.
*/
if (!force_rcw &&
- sh->sector < sh->raid_conf->mddev->recovery_cp)
+ sh->sector < sh->raid_conf->mddev->resync_offset)
/* reconstruct-write isn't being forced */
return 0;
for (i = 0; i < s->failed && i < 2; i++) {
@@ -4097,7 +4097,7 @@ static int handle_stripe_dirtying(struct r5conf *conf,
int disks)
{
int rmw = 0, rcw = 0, i;
- sector_t recovery_cp = conf->mddev->recovery_cp;
+ sector_t resync_offset = conf->mddev->resync_offset;
/* Check whether resync is now happening or should start.
* If yes, then the array is dirty (after unclean shutdown or
@@ -4107,14 +4107,14 @@ static int handle_stripe_dirtying(struct r5conf *conf,
* generate correct data from the parity.
*/
if (conf->rmw_level == PARITY_DISABLE_RMW ||
- (recovery_cp < MaxSector && sh->sector >= recovery_cp &&
+ (resync_offset < MaxSector && sh->sector >= resync_offset &&
s->failed == 0)) {
/* Calculate the real rcw later - for now make it
* look like rcw is cheaper
*/
rcw = 1; rmw = 2;
- pr_debug("force RCW rmw_level=%u, recovery_cp=%llu sh->sector=%llu\n",
- conf->rmw_level, (unsigned long long)recovery_cp,
+ pr_debug("force RCW rmw_level=%u, resync_offset=%llu sh->sector=%llu\n",
+ conf->rmw_level, (unsigned long long)resync_offset,
(unsigned long long)sh->sector);
} else for (i = disks; i--; ) {
/* would I have to read this buffer for read_modify_write */
@@ -4770,14 +4770,14 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s)
if (test_bit(STRIPE_SYNCING, &sh->state)) {
/* If there is a failed device being replaced,
* we must be recovering.
- * else if we are after recovery_cp, we must be syncing
+ * else if we are after resync_offset, we must be syncing
* else if MD_RECOVERY_REQUESTED is set, we also are syncing.
* else we can only be replacing
* sync and recovery both need to read all devices, and so
* use the same flag.
*/
if (do_recovery ||
- sh->sector >= conf->mddev->recovery_cp ||
+ sh->sector >= conf->mddev->resync_offset ||
test_bit(MD_RECOVERY_REQUESTED, &(conf->mddev->recovery)))
s->syncing = 1;
else
@@ -7780,7 +7780,7 @@ static int raid5_run(struct mddev *mddev)
int first = 1;
int ret = -EIO;
- if (mddev->recovery_cp != MaxSector)
+ if (mddev->resync_offset != MaxSector)
pr_notice("md/raid:%s: not clean -- starting background reconstruction\n",
mdname(mddev));
@@ -7921,7 +7921,7 @@ static int raid5_run(struct mddev *mddev)
mdname(mddev));
mddev->ro = 1;
set_disk_ro(mddev->gendisk, 1);
- } else if (mddev->recovery_cp == MaxSector)
+ } else if (mddev->resync_offset == MaxSector)
set_bit(MD_JOURNAL_CLEAN, &mddev->flags);
}
@@ -7988,7 +7988,7 @@ static int raid5_run(struct mddev *mddev)
mddev->resync_max_sectors = mddev->dev_sectors;
if (mddev->degraded > dirty_parity_disks &&
- mddev->recovery_cp != MaxSector) {
+ mddev->resync_offset != MaxSector) {
if (test_bit(MD_HAS_PPL, &mddev->flags))
pr_crit("md/raid:%s: starting dirty degraded array with PPL.\n",
mdname(mddev));
@@ -8328,8 +8328,8 @@ static int raid5_resize(struct mddev *mddev, sector_t sectors)
md_set_array_sectors(mddev, newsize);
if (sectors > mddev->dev_sectors &&
- mddev->recovery_cp > mddev->dev_sectors) {
- mddev->recovery_cp = mddev->dev_sectors;
+ mddev->resync_offset > mddev->dev_sectors) {
+ mddev->resync_offset = mddev->dev_sectors;
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
}
mddev->dev_sectors = sectors;
@@ -8423,7 +8423,7 @@ static int raid5_start_reshape(struct mddev *mddev)
return -EINVAL;
/* raid5 can't handle concurrent reshape and recovery */
- if (mddev->recovery_cp < MaxSector)
+ if (mddev->resync_offset < MaxSector)
return -EBUSY;
for (i = 0; i < conf->raid_disks; i++)
if (conf->disks[i].replacement)
@@ -8648,7 +8648,7 @@ static void *raid45_takeover_raid0(struct mddev *mddev, int level)
mddev->raid_disks += 1;
mddev->delta_disks = 1;
/* make sure it will be not marked as dirty */
- mddev->recovery_cp = MaxSector;
+ mddev->resync_offset = MaxSector;
return setup_conf(mddev);
}