summaryrefslogtreecommitdiff
path: root/drivers/md
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/md')
-rw-r--r--drivers/md/bitmap.c24
-rw-r--r--drivers/md/bitmap.h2
-rw-r--r--drivers/md/dm-exception-store.c22
-rw-r--r--drivers/md/dm-exception-store.h8
-rw-r--r--drivers/md/dm-log-userspace-base.c4
-rw-r--r--drivers/md/dm-log-userspace-transfer.c6
-rw-r--r--drivers/md/dm-snap-persistent.c16
-rw-r--r--drivers/md/dm-snap.c25
-rw-r--r--drivers/md/dm.c11
-rw-r--r--drivers/md/md.c16
-rw-r--r--drivers/md/md.h1
-rw-r--r--drivers/md/raid1.c8
-rw-r--r--drivers/md/raid10.c1
-rw-r--r--drivers/md/raid5.c2
14 files changed, 100 insertions, 46 deletions
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 3319c2fec28e..0aee97a30b80 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -1077,23 +1077,31 @@ static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
* out to disk
*/
-void bitmap_daemon_work(struct bitmap *bitmap)
+void bitmap_daemon_work(mddev_t *mddev)
{
+ struct bitmap *bitmap;
unsigned long j;
unsigned long flags;
struct page *page = NULL, *lastpage = NULL;
int blocks;
void *paddr;
- if (bitmap == NULL)
+ /* Use a mutex to guard daemon_work against
+ * bitmap_destroy.
+ */
+ mutex_lock(&mddev->bitmap_mutex);
+ bitmap = mddev->bitmap;
+ if (bitmap == NULL) {
+ mutex_unlock(&mddev->bitmap_mutex);
return;
+ }
if (time_before(jiffies, bitmap->daemon_lastrun + bitmap->daemon_sleep*HZ))
goto done;
bitmap->daemon_lastrun = jiffies;
if (bitmap->allclean) {
bitmap->mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
- return;
+ goto done;
}
bitmap->allclean = 1;
@@ -1202,6 +1210,7 @@ void bitmap_daemon_work(struct bitmap *bitmap)
done:
if (bitmap->allclean == 0)
bitmap->mddev->thread->timeout = bitmap->daemon_sleep * HZ;
+ mutex_unlock(&mddev->bitmap_mutex);
}
static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
@@ -1538,9 +1547,9 @@ void bitmap_flush(mddev_t *mddev)
*/
sleep = bitmap->daemon_sleep;
bitmap->daemon_sleep = 0;
- bitmap_daemon_work(bitmap);
- bitmap_daemon_work(bitmap);
- bitmap_daemon_work(bitmap);
+ bitmap_daemon_work(mddev);
+ bitmap_daemon_work(mddev);
+ bitmap_daemon_work(mddev);
bitmap->daemon_sleep = sleep;
bitmap_update_sb(bitmap);
}
@@ -1571,6 +1580,7 @@ static void bitmap_free(struct bitmap *bitmap)
kfree(bp);
kfree(bitmap);
}
+
void bitmap_destroy(mddev_t *mddev)
{
struct bitmap *bitmap = mddev->bitmap;
@@ -1578,7 +1588,9 @@ void bitmap_destroy(mddev_t *mddev)
if (!bitmap) /* there was no bitmap */
return;
+ mutex_lock(&mddev->bitmap_mutex);
mddev->bitmap = NULL; /* disconnect from the md device */
+ mutex_unlock(&mddev->bitmap_mutex);
if (mddev->thread)
mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
diff --git a/drivers/md/bitmap.h b/drivers/md/bitmap.h
index e98900671ca9..7e38d13ddcac 100644
--- a/drivers/md/bitmap.h
+++ b/drivers/md/bitmap.h
@@ -282,7 +282,7 @@ void bitmap_close_sync(struct bitmap *bitmap);
void bitmap_cond_end_sync(struct bitmap *bitmap, sector_t sector);
void bitmap_unplug(struct bitmap *bitmap);
-void bitmap_daemon_work(struct bitmap *bitmap);
+void bitmap_daemon_work(mddev_t *mddev);
#endif
#endif
diff --git a/drivers/md/dm-exception-store.c b/drivers/md/dm-exception-store.c
index 556acff3952f..932d1b123143 100644
--- a/drivers/md/dm-exception-store.c
+++ b/drivers/md/dm-exception-store.c
@@ -155,7 +155,8 @@ static int set_chunk_size(struct dm_exception_store *store,
char *value;
chunk_size_ulong = simple_strtoul(chunk_size_arg, &value, 10);
- if (*chunk_size_arg == '\0' || *value != '\0') {
+ if (*chunk_size_arg == '\0' || *value != '\0' ||
+ chunk_size_ulong > UINT_MAX) {
*error = "Invalid chunk size";
return -EINVAL;
}
@@ -171,34 +172,35 @@ static int set_chunk_size(struct dm_exception_store *store,
*/
chunk_size_ulong = round_up(chunk_size_ulong, PAGE_SIZE >> 9);
- return dm_exception_store_set_chunk_size(store, chunk_size_ulong,
+ return dm_exception_store_set_chunk_size(store,
+ (unsigned) chunk_size_ulong,
error);
}
int dm_exception_store_set_chunk_size(struct dm_exception_store *store,
- unsigned long chunk_size_ulong,
+ unsigned chunk_size,
char **error)
{
/* Check chunk_size is a power of 2 */
- if (!is_power_of_2(chunk_size_ulong)) {
+ if (!is_power_of_2(chunk_size)) {
*error = "Chunk size is not a power of 2";
return -EINVAL;
}
/* Validate the chunk size against the device block size */
- if (chunk_size_ulong % (bdev_logical_block_size(store->cow->bdev) >> 9)) {
+ if (chunk_size % (bdev_logical_block_size(store->cow->bdev) >> 9)) {
*error = "Chunk size is not a multiple of device blocksize";
return -EINVAL;
}
- if (chunk_size_ulong > INT_MAX >> SECTOR_SHIFT) {
+ if (chunk_size > INT_MAX >> SECTOR_SHIFT) {
*error = "Chunk size is too high";
return -EINVAL;
}
- store->chunk_size = chunk_size_ulong;
- store->chunk_mask = chunk_size_ulong - 1;
- store->chunk_shift = ffs(chunk_size_ulong) - 1;
+ store->chunk_size = chunk_size;
+ store->chunk_mask = chunk_size - 1;
+ store->chunk_shift = ffs(chunk_size) - 1;
return 0;
}
@@ -251,7 +253,7 @@ int dm_exception_store_create(struct dm_target *ti, int argc, char **argv,
r = set_chunk_size(tmp_store, argv[2], &ti->error);
if (r)
- goto bad_cow;
+ goto bad_ctr;
r = type->ctr(tmp_store, 0, NULL);
if (r) {
diff --git a/drivers/md/dm-exception-store.h b/drivers/md/dm-exception-store.h
index 812c71872ba0..8a223a48802c 100644
--- a/drivers/md/dm-exception-store.h
+++ b/drivers/md/dm-exception-store.h
@@ -101,9 +101,9 @@ struct dm_exception_store {
struct dm_dev *cow;
/* Size of data blocks saved - must be a power of 2 */
- chunk_t chunk_size;
- chunk_t chunk_mask;
- chunk_t chunk_shift;
+ unsigned chunk_size;
+ unsigned chunk_mask;
+ unsigned chunk_shift;
void *context;
};
@@ -169,7 +169,7 @@ int dm_exception_store_type_register(struct dm_exception_store_type *type);
int dm_exception_store_type_unregister(struct dm_exception_store_type *type);
int dm_exception_store_set_chunk_size(struct dm_exception_store *store,
- unsigned long chunk_size_ulong,
+ unsigned chunk_size,
char **error);
int dm_exception_store_create(struct dm_target *ti, int argc, char **argv,
diff --git a/drivers/md/dm-log-userspace-base.c b/drivers/md/dm-log-userspace-base.c
index 6e186b1a062d..7ac2c1450d10 100644
--- a/drivers/md/dm-log-userspace-base.c
+++ b/drivers/md/dm-log-userspace-base.c
@@ -156,7 +156,7 @@ static int userspace_ctr(struct dm_dirty_log *log, struct dm_target *ti,
}
/* The ptr value is sufficient for local unique id */
- lc->luid = (uint64_t)lc;
+ lc->luid = (unsigned long)lc;
lc->ti = ti;
@@ -582,7 +582,7 @@ static int userspace_status(struct dm_dirty_log *log, status_type_t status_type,
break;
case STATUSTYPE_TABLE:
sz = 0;
- table_args = strstr(lc->usr_argv_str, " ");
+ table_args = strchr(lc->usr_argv_str, ' ');
BUG_ON(!table_args); /* There will always be a ' ' */
table_args++;
diff --git a/drivers/md/dm-log-userspace-transfer.c b/drivers/md/dm-log-userspace-transfer.c
index ba0edad2d048..54abf9e303b7 100644
--- a/drivers/md/dm-log-userspace-transfer.c
+++ b/drivers/md/dm-log-userspace-transfer.c
@@ -129,11 +129,13 @@ static int fill_pkg(struct cn_msg *msg, struct dm_ulog_request *tfr)
* This is the connector callback that delivers data
* that was sent from userspace.
*/
-static void cn_ulog_callback(void *data)
+static void cn_ulog_callback(struct cn_msg *msg, struct netlink_skb_parms *nsp)
{
- struct cn_msg *msg = (struct cn_msg *)data;
struct dm_ulog_request *tfr = (struct dm_ulog_request *)(msg + 1);
+ if (!cap_raised(nsp->eff_cap, CAP_SYS_ADMIN))
+ return;
+
spin_lock(&receiving_list_lock);
if (msg->len == 0)
fill_pkg(msg, NULL);
diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c
index d5b2e08750d5..0c746420c008 100644
--- a/drivers/md/dm-snap-persistent.c
+++ b/drivers/md/dm-snap-persistent.c
@@ -284,12 +284,13 @@ static int read_header(struct pstore *ps, int *new_snapshot)
{
int r;
struct disk_header *dh;
- chunk_t chunk_size;
+ unsigned chunk_size;
int chunk_size_supplied = 1;
char *chunk_err;
/*
- * Use default chunk size (or hardsect_size, if larger) if none supplied
+ * Use default chunk size (or logical_block_size, if larger)
+ * if none supplied
*/
if (!ps->store->chunk_size) {
ps->store->chunk_size = max(DM_CHUNK_SIZE_DEFAULT_SECTORS,
@@ -334,10 +335,9 @@ static int read_header(struct pstore *ps, int *new_snapshot)
return 0;
if (chunk_size_supplied)
- DMWARN("chunk size %llu in device metadata overrides "
- "table chunk size of %llu.",
- (unsigned long long)chunk_size,
- (unsigned long long)ps->store->chunk_size);
+ DMWARN("chunk size %u in device metadata overrides "
+ "table chunk size of %u.",
+ chunk_size, ps->store->chunk_size);
/* We had a bogus chunk_size. Fix stuff up. */
free_area(ps);
@@ -345,8 +345,8 @@ static int read_header(struct pstore *ps, int *new_snapshot)
r = dm_exception_store_set_chunk_size(ps->store, chunk_size,
&chunk_err);
if (r) {
- DMERR("invalid on-disk chunk size %llu: %s.",
- (unsigned long long)chunk_size, chunk_err);
+ DMERR("invalid on-disk chunk size %u: %s.",
+ chunk_size, chunk_err);
return r;
}
diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 57f1bf7f3b7a..3a3ba46e6d4b 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -296,6 +296,7 @@ static void __insert_origin(struct origin *o)
*/
static int register_snapshot(struct dm_snapshot *snap)
{
+ struct dm_snapshot *l;
struct origin *o, *new_o;
struct block_device *bdev = snap->origin->bdev;
@@ -319,7 +320,11 @@ static int register_snapshot(struct dm_snapshot *snap)
__insert_origin(o);
}
- list_add_tail(&snap->list, &o->snapshots);
+ /* Sort the list according to chunk size, largest-first smallest-last */
+ list_for_each_entry(l, &o->snapshots, list)
+ if (l->store->chunk_size < snap->store->chunk_size)
+ break;
+ list_add_tail(&snap->list, &l->list);
up_write(&_origins_lock);
return 0;
@@ -668,6 +673,11 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
bio_list_init(&s->queued_bios);
INIT_WORK(&s->queued_bios_work, flush_queued_bios);
+ if (!s->store->chunk_size) {
+ ti->error = "Chunk size not set";
+ goto bad_load_and_register;
+ }
+
/* Add snapshot to the list of snapshots for this origin */
/* Exceptions aren't triggered till snapshot_resume() is called */
if (register_snapshot(s)) {
@@ -951,7 +961,7 @@ static void start_copy(struct dm_snap_pending_exception *pe)
src.bdev = bdev;
src.sector = chunk_to_sector(s->store, pe->e.old_chunk);
- src.count = min(s->store->chunk_size, dev_size - src.sector);
+ src.count = min((sector_t)s->store->chunk_size, dev_size - src.sector);
dest.bdev = s->store->cow->bdev;
dest.sector = chunk_to_sector(s->store, pe->e.new_chunk);
@@ -1142,6 +1152,8 @@ static int snapshot_status(struct dm_target *ti, status_type_t type,
unsigned sz = 0;
struct dm_snapshot *snap = ti->private;
+ down_write(&snap->lock);
+
switch (type) {
case STATUSTYPE_INFO:
if (!snap->valid)
@@ -1173,6 +1185,8 @@ static int snapshot_status(struct dm_target *ti, status_type_t type,
break;
}
+ up_write(&snap->lock);
+
return 0;
}
@@ -1388,7 +1402,7 @@ static void origin_resume(struct dm_target *ti)
struct dm_dev *dev = ti->private;
struct dm_snapshot *snap;
struct origin *o;
- chunk_t chunk_size = 0;
+ unsigned chunk_size = 0;
down_read(&_origins_lock);
o = __lookup_origin(dev->bdev);
@@ -1465,7 +1479,7 @@ static int __init dm_snapshot_init(void)
r = dm_register_target(&snapshot_target);
if (r) {
DMERR("snapshot target register failed %d", r);
- return r;
+ goto bad_register_snapshot_target;
}
r = dm_register_target(&origin_target);
@@ -1522,6 +1536,9 @@ bad2:
dm_unregister_target(&origin_target);
bad1:
dm_unregister_target(&snapshot_target);
+
+bad_register_snapshot_target:
+ dm_exception_store_exit();
return r;
}
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index b4845b14740d..ae087b0c49f9 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -47,6 +47,7 @@ struct dm_io {
atomic_t io_count;
struct bio *bio;
unsigned long start_time;
+ spinlock_t endio_lock;
};
/*
@@ -576,8 +577,12 @@ static void dec_pending(struct dm_io *io, int error)
struct mapped_device *md = io->md;
/* Push-back supersedes any I/O errors */
- if (error && !(io->error > 0 && __noflush_suspending(md)))
- io->error = error;
+ if (unlikely(error)) {
+ spin_lock_irqsave(&io->endio_lock, flags);
+ if (!(io->error > 0 && __noflush_suspending(md)))
+ io->error = error;
+ spin_unlock_irqrestore(&io->endio_lock, flags);
+ }
if (atomic_dec_and_test(&io->io_count)) {
if (io->error == DM_ENDIO_REQUEUE) {
@@ -1224,6 +1229,7 @@ static void __split_and_process_bio(struct mapped_device *md, struct bio *bio)
atomic_set(&ci.io->io_count, 1);
ci.io->bio = bio;
ci.io->md = md;
+ spin_lock_init(&ci.io->endio_lock);
ci.sector = bio->bi_sector;
ci.sector_count = bio_sectors(bio);
if (unlikely(bio_empty_barrier(bio)))
@@ -1819,6 +1825,7 @@ static struct mapped_device *alloc_dev(int minor)
bad_bdev:
destroy_workqueue(md->wq);
bad_thread:
+ del_gendisk(md->disk);
put_disk(md->disk);
bad_disk:
blk_cleanup_queue(md->queue);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 9dd872000cec..2938e9ca711e 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -276,7 +276,9 @@ static void mddev_put(mddev_t *mddev)
if (!atomic_dec_and_lock(&mddev->active, &all_mddevs_lock))
return;
if (!mddev->raid_disks && list_empty(&mddev->disks) &&
- !mddev->hold_active) {
+ mddev->ctime == 0 && !mddev->hold_active) {
+ /* Array is not configured at all, and not held active,
+ * so destroy it */
list_del(&mddev->all_mddevs);
if (mddev->gendisk) {
/* we did a probe so need to clean up.
@@ -361,6 +363,7 @@ static mddev_t * mddev_find(dev_t unit)
mutex_init(&new->open_mutex);
mutex_init(&new->reconfig_mutex);
+ mutex_init(&new->bitmap_mutex);
INIT_LIST_HEAD(&new->disks);
INIT_LIST_HEAD(&new->all_mddevs);
init_timer(&new->safemode_timer);
@@ -5039,6 +5042,10 @@ static int set_array_info(mddev_t * mddev, mdu_array_info_t *info)
mddev->minor_version = info->minor_version;
mddev->patch_version = info->patch_version;
mddev->persistent = !info->not_persistent;
+ /* ensure mddev_put doesn't delete this now that there
+ * is some minimal configuration.
+ */
+ mddev->ctime = get_seconds();
return 0;
}
mddev->major_version = MD_MAJOR_VERSION;
@@ -6495,8 +6502,9 @@ void md_do_sync(mddev_t *mddev)
skip:
mddev->curr_resync = 0;
mddev->curr_resync_completed = 0;
- mddev->resync_min = 0;
- mddev->resync_max = MaxSector;
+ if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery))
+ /* We completed so max setting can be forgotten. */
+ mddev->resync_max = MaxSector;
sysfs_notify(&mddev->kobj, NULL, "sync_completed");
wake_up(&resync_wait);
set_bit(MD_RECOVERY_DONE, &mddev->recovery);
@@ -6594,7 +6602,7 @@ void md_check_recovery(mddev_t *mddev)
if (mddev->bitmap)
- bitmap_daemon_work(mddev->bitmap);
+ bitmap_daemon_work(mddev);
if (mddev->ro)
return;
diff --git a/drivers/md/md.h b/drivers/md/md.h
index f8fc188bc762..d7aad830018e 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -289,6 +289,7 @@ struct mddev_s
* hot-adding a bitmap. It should
* eventually be settable by sysfs.
*/
+ struct mutex bitmap_mutex;
struct list_head all_mddevs;
};
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 8726fd7ebce5..d3e492e8781c 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1643,11 +1643,12 @@ static void raid1d(mddev_t *mddev)
r1_bio->sector,
r1_bio->sectors);
unfreeze_array(conf);
- }
+ } else
+ md_error(mddev,
+ conf->mirrors[r1_bio->read_disk].rdev);
bio = r1_bio->bios[r1_bio->read_disk];
- if ((disk=read_balance(conf, r1_bio)) == -1 ||
- disk == r1_bio->read_disk) {
+ if ((disk=read_balance(conf, r1_bio)) == -1) {
printk(KERN_ALERT "raid1: %s: unrecoverable I/O"
" read error for block %llu\n",
bdevname(bio->bi_bdev,b),
@@ -1676,6 +1677,7 @@ static void raid1d(mddev_t *mddev)
generic_make_request(bio);
}
}
+ cond_resched();
}
if (unplug)
unplug_slaves(mddev);
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 3d9020cf6f6e..6a5a5fb18b30 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1630,6 +1630,7 @@ static void raid10d(mddev_t *mddev)
generic_make_request(bio);
}
}
+ cond_resched();
}
if (unplug)
unplug_slaves(mddev);
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index b8a2c5dc67ba..c339c8fdcbd5 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3790,6 +3790,8 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
sector_nr = conf->reshape_progress;
sector_div(sector_nr, new_data_disks);
if (sector_nr) {
+ mddev->curr_resync_completed = sector_nr;
+ sysfs_notify(&mddev->kobj, NULL, "sync_completed");
*skipped = 1;
return sector_nr;
}