aboutsummaryrefslogtreecommitdiff
path: root/drivers/md
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/md')
-rw-r--r--drivers/md/Makefile22
-rw-r--r--drivers/md/bitmap.c9
-rw-r--r--drivers/md/dm-exception-store.c38
-rw-r--r--drivers/md/dm-exception-store.h8
-rw-r--r--drivers/md/dm-log-userspace-base.c2
-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.c48
-rw-r--r--drivers/md/raid1.c13
-rw-r--r--drivers/md/raid10.c5
-rw-r--r--drivers/md/raid5.c269
-rw-r--r--drivers/md/raid5.h14
-rw-r--r--drivers/md/raid6altivec.uc2
-rw-r--r--drivers/md/raid6int.uc2
-rw-r--r--drivers/md/raid6test/Makefile42
-rw-r--r--drivers/md/unroll.awk20
-rw-r--r--drivers/md/unroll.pl24
19 files changed, 351 insertions, 225 deletions
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index 1dc4185bd78..e355e7f6a53 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -46,7 +46,7 @@ obj-$(CONFIG_DM_LOG_USERSPACE) += dm-log-userspace.o
obj-$(CONFIG_DM_ZERO) += dm-zero.o
quiet_cmd_unroll = UNROLL $@
- cmd_unroll = $(PERL) $(srctree)/$(src)/unroll.pl $(UNROLL) \
+ cmd_unroll = $(AWK) -f$(srctree)/$(src)/unroll.awk -vN=$(UNROLL) \
< $< > $@ || ( rm -f $@ && exit 1 )
ifeq ($(CONFIG_ALTIVEC),y)
@@ -59,56 +59,56 @@ endif
targets += raid6int1.c
$(obj)/raid6int1.c: UNROLL := 1
-$(obj)/raid6int1.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int1.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int2.c
$(obj)/raid6int2.c: UNROLL := 2
-$(obj)/raid6int2.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int2.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int4.c
$(obj)/raid6int4.c: UNROLL := 4
-$(obj)/raid6int4.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int4.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int8.c
$(obj)/raid6int8.c: UNROLL := 8
-$(obj)/raid6int8.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int8.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int16.c
$(obj)/raid6int16.c: UNROLL := 16
-$(obj)/raid6int16.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int16.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int32.c
$(obj)/raid6int32.c: UNROLL := 32
-$(obj)/raid6int32.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int32.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
CFLAGS_raid6altivec1.o += $(altivec_flags)
targets += raid6altivec1.c
$(obj)/raid6altivec1.c: UNROLL := 1
-$(obj)/raid6altivec1.c: $(src)/raid6altivec.uc $(src)/unroll.pl FORCE
+$(obj)/raid6altivec1.c: $(src)/raid6altivec.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
CFLAGS_raid6altivec2.o += $(altivec_flags)
targets += raid6altivec2.c
$(obj)/raid6altivec2.c: UNROLL := 2
-$(obj)/raid6altivec2.c: $(src)/raid6altivec.uc $(src)/unroll.pl FORCE
+$(obj)/raid6altivec2.c: $(src)/raid6altivec.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
CFLAGS_raid6altivec4.o += $(altivec_flags)
targets += raid6altivec4.c
$(obj)/raid6altivec4.c: UNROLL := 4
-$(obj)/raid6altivec4.c: $(src)/raid6altivec.uc $(src)/unroll.pl FORCE
+$(obj)/raid6altivec4.c: $(src)/raid6altivec.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
CFLAGS_raid6altivec8.o += $(altivec_flags)
targets += raid6altivec8.c
$(obj)/raid6altivec8.c: UNROLL := 8
-$(obj)/raid6altivec8.c: $(src)/raid6altivec.uc $(src)/unroll.pl FORCE
+$(obj)/raid6altivec8.c: $(src)/raid6altivec.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
quiet_cmd_mktable = TABLE $@
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 6986b0059d2..60e2b322db1 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -1624,10 +1624,11 @@ int bitmap_create(mddev_t *mddev)
bitmap->offset = mddev->bitmap_offset;
if (file) {
get_file(file);
- do_sync_mapping_range(file->f_mapping, 0, LLONG_MAX,
- SYNC_FILE_RANGE_WAIT_BEFORE |
- SYNC_FILE_RANGE_WRITE |
- SYNC_FILE_RANGE_WAIT_AFTER);
+ /* As future accesses to this file will use bmap,
+ * and bypass the page cache, we must sync the file
+ * first.
+ */
+ vfs_fsync(file, file->f_dentry, 1);
}
/* read superblock from bitmap file (this sets bitmap->chunksize) */
err = bitmap_read_sb(bitmap);
diff --git a/drivers/md/dm-exception-store.c b/drivers/md/dm-exception-store.c
index 556acff3952..7dbe652efb5 100644
--- a/drivers/md/dm-exception-store.c
+++ b/drivers/md/dm-exception-store.c
@@ -138,16 +138,6 @@ int dm_exception_store_type_unregister(struct dm_exception_store_type *type)
}
EXPORT_SYMBOL(dm_exception_store_type_unregister);
-/*
- * Round a number up to the nearest 'size' boundary. size must
- * be a power of 2.
- */
-static ulong round_up(ulong n, ulong size)
-{
- size--;
- return (n + size) & ~size;
-}
-
static int set_chunk_size(struct dm_exception_store *store,
const char *chunk_size_arg, char **error)
{
@@ -155,7 +145,8 @@ 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;
}
@@ -165,40 +156,35 @@ static int set_chunk_size(struct dm_exception_store *store,
return 0;
}
- /*
- * Chunk size must be multiple of page size. Silently
- * round up if it's not.
- */
- 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 +237,7 @@ 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 812c71872ba..8a223a48802 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 652bd33109e..7ac2c1450d1 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;
diff --git a/drivers/md/dm-log-userspace-transfer.c b/drivers/md/dm-log-userspace-transfer.c
index ba0edad2d04..54abf9e303b 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 d5b2e08750d..0c746420c00 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 57f1bf7f3b7..3a3ba46e6d4 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 376f1ab48a2..724efc63904 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;
};
/*
@@ -578,8 +579,12 @@ 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) {
@@ -1226,6 +1231,7 @@ 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)))
@@ -1822,6 +1828,7 @@ 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 26ba42a7912..b182f86a19d 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -944,6 +944,14 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
desc->raid_disk < mddev->raid_disks */) {
set_bit(In_sync, &rdev->flags);
rdev->raid_disk = desc->raid_disk;
+ } else if (desc->state & (1<<MD_DISK_ACTIVE)) {
+ /* active but not in sync implies recovery up to
+ * reshape position. We don't know exactly where
+ * that is, so set to zero for now */
+ if (mddev->minor_version >= 91) {
+ rdev->recovery_offset = 0;
+ rdev->raid_disk = desc->raid_disk;
+ }
}
if (desc->state & (1<<MD_DISK_WRITEMOSTLY))
set_bit(WriteMostly, &rdev->flags);
@@ -1032,8 +1040,19 @@ static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev)
list_for_each_entry(rdev2, &mddev->disks, same_set) {
mdp_disk_t *d;
int desc_nr;
- if (rdev2->raid_disk >= 0 && test_bit(In_sync, &rdev2->flags)
- && !test_bit(Faulty, &rdev2->flags))
+ int is_active = test_bit(In_sync, &rdev2->flags);
+
+ if (rdev2->raid_disk >= 0 &&
+ sb->minor_version >= 91)
+ /* we have nowhere to store the recovery_offset,
+ * but if it is not below the reshape_position,
+ * we can piggy-back on that.
+ */
+ is_active = 1;
+ if (rdev2->raid_disk < 0 ||
+ test_bit(Faulty, &rdev2->flags))
+ is_active = 0;
+ if (is_active)
desc_nr = rdev2->raid_disk;
else
desc_nr = next_spare++;
@@ -1043,16 +1062,16 @@ static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev)
d->number = rdev2->desc_nr;
d->major = MAJOR(rdev2->bdev->bd_dev);
d->minor = MINOR(rdev2->bdev->bd_dev);
- if (rdev2->raid_disk >= 0 && test_bit(In_sync, &rdev2->flags)
- && !test_bit(Faulty, &rdev2->flags))
+ if (is_active)
d->raid_disk = rdev2->raid_disk;
else
d->raid_disk = rdev2->desc_nr; /* compatibility */
if (test_bit(Faulty, &rdev2->flags))
d->state = (1<<MD_DISK_FAULTY);
- else if (test_bit(In_sync, &rdev2->flags)) {
+ else if (is_active) {
d->state = (1<<MD_DISK_ACTIVE);
- d->state |= (1<<MD_DISK_SYNC);
+ if (test_bit(In_sync, &rdev2->flags))
+ d->state |= (1<<MD_DISK_SYNC);
active++;
working++;
} else {
@@ -1382,8 +1401,6 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
if (rdev->raid_disk >= 0 &&
!test_bit(In_sync, &rdev->flags)) {
- if (mddev->curr_resync_completed > rdev->recovery_offset)
- rdev->recovery_offset = mddev->curr_resync_completed;
if (rdev->recovery_offset > 0) {
sb->feature_map |=
cpu_to_le32(MD_FEATURE_RECOVERY_OFFSET);
@@ -1917,6 +1934,14 @@ static void sync_sbs(mddev_t * mddev, int nospares)
*/
mdk_rdev_t *rdev;
+ /* First make sure individual recovery_offsets are correct */
+ list_for_each_entry(rdev, &mddev->disks, same_set) {
+ if (rdev->raid_disk >= 0 &&
+ !test_bit(In_sync, &rdev->flags) &&
+ mddev->curr_resync_completed > rdev->recovery_offset)
+ rdev->recovery_offset = mddev->curr_resync_completed;
+
+ }
list_for_each_entry(rdev, &mddev->disks, same_set) {
if (rdev->sb_events == mddev->events ||
(nospares &&
@@ -2631,7 +2656,7 @@ static void analyze_sbs(mddev_t * mddev)
rdev->desc_nr = i++;
rdev->raid_disk = rdev->desc_nr;
set_bit(In_sync, &rdev->flags);
- } else if (rdev->raid_disk >= mddev->raid_disks) {
+ } else if (rdev->raid_disk >= (mddev->raid_disks - min(0, mddev->delta_disks))) {
rdev->raid_disk = -1;
clear_bit(In_sync, &rdev->flags);
}
@@ -6504,8 +6529,9 @@ 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);
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index d1b9bd5fd4f..e07ce2e033a 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -64,7 +64,7 @@ static void * r1bio_pool_alloc(gfp_t gfp_flags, void *data)
/* allocate a r1bio with room for raid_disks entries in the bios array */
r1_bio = kzalloc(size, gfp_flags);
- if (!r1_bio)
+ if (!r1_bio && pi->mddev)
unplug_slaves(pi->mddev);
return r1_bio;
@@ -1650,11 +1650,12 @@ 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),
@@ -1683,6 +1684,7 @@ static void raid1d(mddev_t *mddev)
generic_make_request(bio);
}
}
+ cond_resched();
}
if (unplug)
unplug_slaves(mddev);
@@ -1978,13 +1980,14 @@ static int run(mddev_t *mddev)
conf->poolinfo = kmalloc(sizeof(*conf->poolinfo), GFP_KERNEL);
if (!conf->poolinfo)
goto out_no_mem;
- conf->poolinfo->mddev = mddev;
+ conf->poolinfo->mddev = NULL;
conf->poolinfo->raid_disks = mddev->raid_disks;
conf->r1bio_pool = mempool_create(NR_RAID1_BIOS, r1bio_pool_alloc,
r1bio_pool_free,
conf->poolinfo);
if (!conf->r1bio_pool)
goto out_no_mem;
+ conf->poolinfo->mddev = mddev;
spin_lock_init(&conf->device_lock);
mddev->queue->queue_lock = &conf->device_lock;
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 51c4c5c4d87..c2cb7b87b44 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -68,7 +68,7 @@ static void * r10bio_pool_alloc(gfp_t gfp_flags, void *data)
/* allocate a r10bio with room for raid_disks entries in the bios array */
r10_bio = kzalloc(size, gfp_flags);
- if (!r10_bio)
+ if (!r10_bio && conf->mddev)
unplug_slaves(conf->mddev);
return r10_bio;
@@ -1632,6 +1632,7 @@ static void raid10d(mddev_t *mddev)
generic_make_request(bio);
}
}
+ cond_resched();
}
if (unplug)
unplug_slaves(mddev);
@@ -2095,7 +2096,6 @@ static int run(mddev_t *mddev)
if (!conf->tmppage)
goto out_free_conf;
- conf->mddev = mddev;
conf->raid_disks = mddev->raid_disks;
conf->near_copies = nc;
conf->far_copies = fc;
@@ -2132,6 +2132,7 @@ static int run(mddev_t *mddev)
goto out_free_conf;
}
+ conf->mddev = mddev;
spin_lock_init(&conf->device_lock);
mddev->queue->queue_lock = &conf->device_lock;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 94829804ab7..d29215d966d 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -156,13 +156,16 @@ static inline int raid6_next_disk(int disk, int raid_disks)
static int raid6_idx_to_slot(int idx, struct stripe_head *sh,
int *count, int syndrome_disks)
{
- int slot;
+ int slot = *count;
+ if (sh->ddf_layout)
+ (*count)++;
if (idx == sh->pd_idx)
return syndrome_disks;
if (idx == sh->qd_idx)
return syndrome_disks + 1;
- slot = (*count)++;
+ if (!sh->ddf_layout)
+ (*count)++;
return slot;
}
@@ -717,7 +720,7 @@ static int set_syndrome_sources(struct page **srcs, struct stripe_head *sh)
int i;
for (i = 0; i < disks; i++)
- srcs[i] = (void *)raid6_empty_zero_page;
+ srcs[i] = NULL;
count = 0;
i = d0_idx;
@@ -727,9 +730,8 @@ static int set_syndrome_sources(struct page **srcs, struct stripe_head *sh)
srcs[slot] = sh->dev[i].page;
i = raid6_next_disk(i, disks);
} while (i != d0_idx);
- BUG_ON(count != syndrome_disks);
- return count;
+ return syndrome_disks;
}
static struct dma_async_tx_descriptor *
@@ -814,7 +816,7 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu)
* slot number conversion for 'faila' and 'failb'
*/
for (i = 0; i < disks ; i++)
- blocks[i] = (void *)raid6_empty_zero_page;
+ blocks[i] = NULL;
count = 0;
i = d0_idx;
do {
@@ -828,7 +830,6 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu)
failb = slot;
i = raid6_next_disk(i, disks);
} while (i != d0_idx);
- BUG_ON(count != syndrome_disks);
BUG_ON(faila == failb);
if (failb < faila)
@@ -845,7 +846,7 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu)
init_async_submit(&submit, ASYNC_TX_FENCE, NULL,
ops_complete_compute, sh,
to_addr_conv(sh, percpu));
- return async_gen_syndrome(blocks, 0, count+2,
+ return async_gen_syndrome(blocks, 0, syndrome_disks+2,
STRIPE_SIZE, &submit);
} else {
struct page *dest;
@@ -1139,7 +1140,7 @@ static void ops_run_check_pq(struct stripe_head *sh, struct raid5_percpu *percpu
&sh->ops.zero_sum_result, percpu->spare_page, &submit);
}
-static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
+static void __raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
{
int overlap_clear = 0, i, disks = sh->disks;
struct dma_async_tx_descriptor *tx = NULL;
@@ -1204,22 +1205,55 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
put_cpu();
}
+#ifdef CONFIG_MULTICORE_RAID456
+static void async_run_ops(void *param, async_cookie_t cookie)
+{
+ struct stripe_head *sh = param;
+ unsigned long ops_request = sh->ops.request;
+
+ clear_bit_unlock(STRIPE_OPS_REQ_PENDING, &sh->state);
+ wake_up(&sh->ops.wait_for_ops);
+
+ __raid_run_ops(sh, ops_request);
+ release_stripe(sh);
+}
+
+static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
+{
+ /* since handle_stripe can be called outside of raid5d context
+ * we need to ensure sh->ops.request is de-staged before another
+ * request arrives
+ */
+ wait_event(sh->ops.wait_for_ops,
+ !test_and_set_bit_lock(STRIPE_OPS_REQ_PENDING, &sh->state));
+ sh->ops.request = ops_request;
+
+ atomic_inc(&sh->count);
+ async_schedule(async_run_ops, sh);
+}
+#else
+#define raid_run_ops __raid_run_ops
+#endif
+
static int grow_one_stripe(raid5_conf_t *conf)
{
struct stripe_head *sh;
+ int disks = max(conf->raid_disks, conf->previous_raid_disks);
sh = kmem_cache_alloc(conf->slab_cache, GFP_KERNEL);
if (!sh)
return 0;
- memset(sh, 0, sizeof(*sh) + (conf->raid_disks-1)*sizeof(struct r5dev));
+ memset(sh, 0, sizeof(*sh) + (disks-1)*sizeof(struct r5dev));
sh->raid_conf = conf;
spin_lock_init(&sh->lock);
+ #ifdef CONFIG_MULTICORE_RAID456
+ init_waitqueue_head(&sh->ops.wait_for_ops);
+ #endif
- if (grow_buffers(sh, conf->raid_disks)) {
- shrink_buffers(sh, conf->raid_disks);
+ if (grow_buffers(sh, disks)) {
+ shrink_buffers(sh, disks);
kmem_cache_free(conf->slab_cache, sh);
return 0;
}
- sh->disks = conf->raid_disks;
/* we just created an active stripe so... */
atomic_set(&sh->count, 1);
atomic_inc(&conf->active_stripes);
@@ -1231,7 +1265,7 @@ static int grow_one_stripe(raid5_conf_t *conf)
static int grow_stripes(raid5_conf_t *conf, int num)
{
struct kmem_cache *sc;
- int devs = conf->raid_disks;
+ int devs = max(conf->raid_disks, conf->previous_raid_disks);
sprintf(conf->cache_name[0],
"raid%d-%s", conf->level, mdname(conf->mddev));
@@ -1329,6 +1363,9 @@ static int resize_stripes(raid5_conf_t *conf, int newsize)
nsh->raid_conf = conf;
spin_lock_init(&nsh->lock);
+ #ifdef CONFIG_MULTICORE_RAID456
+ init_waitqueue_head(&nsh->ops.wait_for_ops);
+ #endif
list_add(&nsh->lru, &newstripes);
}
@@ -1899,10 +1936,15 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous)
case ALGORITHM_PARITY_N:
break;
case ALGORITHM_ROTATING_N_CONTINUE:
+ /* Like left_symmetric, but P is before Q */
if (sh->pd_idx == 0)
i--; /* P D D D Q */
- else if (i > sh->pd_idx)
- i -= 2; /* D D Q P D */
+ else {
+ /* D D Q P D */
+ if (i < sh->pd_idx)
+ i += raid_disks;
+ i -= (sh->pd_idx + 1);
+ }
break;
case ALGORITHM_LEFT_ASYMMETRIC_6:
case ALGORITHM_RIGHT_ASYMMETRIC_6:
@@ -2896,7 +2938,7 @@ static void handle_stripe_expansion(raid5_conf_t *conf, struct stripe_head *sh,
*
*/
-static bool handle_stripe5(struct stripe_head *sh)
+static void handle_stripe5(struct stripe_head *sh)
{
raid5_conf_t *conf = sh->raid_conf;
int disks = sh->disks, i;
@@ -3167,11 +3209,9 @@ static bool handle_stripe5(struct stripe_head *sh)
ops_run_io(sh, &s);
return_io(return_bi);
-
- return blocked_rdev == NULL;
}
-static bool handle_stripe6(struct stripe_head *sh)
+static void handle_stripe6(struct stripe_head *sh)
{
raid5_conf_t *conf = sh->raid_conf;
int disks = sh->disks;
@@ -3455,17 +3495,14 @@ static bool handle_stripe6(struct stripe_head *sh)
ops_run_io(sh, &s);
return_io(return_bi);
-
- return blocked_rdev == NULL;
}
-/* returns true if the stripe was handled */
-static bool handle_stripe(struct stripe_head *sh)
+static void handle_stripe(struct stripe_head *sh)
{
if (sh->raid_conf->level == 6)
- return handle_stripe6(sh);
+ handle_stripe6(sh);
else
- return handle_stripe5(sh);
+ handle_stripe5(sh);
}
static void raid5_activate_delayed(raid5_conf_t *conf)
@@ -3503,9 +3540,10 @@ static void unplug_slaves(mddev_t *mddev)
{
raid5_conf_t *conf = mddev->private;
int i;
+ int devs = max(conf->raid_disks, conf->previous_raid_disks);
rcu_read_lock();
- for (i = 0; i < conf->raid_disks; i++) {
+ for (i = 0; i < devs; i++) {
mdk_rdev_t *rdev = rcu_dereference(conf->disks[i].rdev);
if (rdev && !test_bit(Faulty, &rdev->flags) && atomic_read(&rdev->nr_pending)) {
struct request_queue *r_queue = bdev_get_queue(rdev->bdev);
@@ -4011,6 +4049,8 @@ 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;
}
@@ -4277,9 +4317,7 @@ static inline sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *ski
clear_bit(STRIPE_INSYNC, &sh->state);
spin_unlock(&sh->lock);
- /* wait for any blocked device to be handled */
- while (unlikely(!handle_stripe(sh)))
- ;
+ handle_stripe(sh);
release_stripe(sh);
return STRIPE_SECTORS;