aboutsummaryrefslogtreecommitdiff
path: root/drivers/md
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/md')
-rw-r--r--drivers/md/Kconfig26
-rw-r--r--drivers/md/Makefile22
-rw-r--r--drivers/md/bitmap.c14
-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/linear.c3
-rw-r--r--drivers/md/md.c73
-rw-r--r--drivers/md/md.h1
-rw-r--r--drivers/md/multipath.c6
-rw-r--r--drivers/md/raid0.c8
-rw-r--r--drivers/md/raid1.c21
-rw-r--r--drivers/md/raid10.c17
-rw-r--r--drivers/md/raid5.c1666
-rw-r--r--drivers/md/raid5.h42
-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
24 files changed, 1323 insertions, 772 deletions
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 020f9573fd8..2158377a135 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -124,6 +124,8 @@ config MD_RAID456
select MD_RAID6_PQ
select ASYNC_MEMCPY
select ASYNC_XOR
+ select ASYNC_PQ
+ select ASYNC_RAID6_RECOV
---help---
A RAID-5 set of N drives with a capacity of C MB per drive provides
the capacity of C * (N - 1) MB, and protects against a failure
@@ -152,9 +154,33 @@ config MD_RAID456
If unsure, say Y.
+config MULTICORE_RAID456
+ bool "RAID-4/RAID-5/RAID-6 Multicore processing (EXPERIMENTAL)"
+ depends on MD_RAID456
+ depends on SMP
+ depends on EXPERIMENTAL
+ ---help---
+ Enable the raid456 module to dispatch per-stripe raid operations to a
+ thread pool.
+
+ If unsure, say N.
+
config MD_RAID6_PQ
tristate
+config ASYNC_RAID6_TEST
+ tristate "Self test for hardware accelerated raid6 recovery"
+ depends on MD_RAID6_PQ
+ select ASYNC_RAID6_RECOV
+ ---help---
+ This is a one-shot self test that permutes through the
+ recovery of all the possible two disk failure scenarios for a
+ N-disk array. Recovery is performed with the asynchronous
+ raid6 recovery routines, and will optionally use an offload
+ engine if one is available.
+
+ If unsure, say N.
+
config MD_MULTIPATH
tristate "Multipath I/O support"
depends on BLK_DEV_MD
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 3319c2fec28..60e2b322db1 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -108,6 +108,8 @@ static void bitmap_free_page(struct bitmap *bitmap, unsigned char *page)
* allocated while we're using it
*/
static int bitmap_checkpage(struct bitmap *bitmap, unsigned long page, int create)
+__releases(bitmap->lock)
+__acquires(bitmap->lock)
{
unsigned char *mappage;
@@ -325,7 +327,6 @@ static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
return 0;
bad_alignment:
- rcu_read_unlock();
return -EINVAL;
}
@@ -1207,6 +1208,8 @@ void bitmap_daemon_work(struct bitmap *bitmap)
static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
sector_t offset, int *blocks,
int create)
+__releases(bitmap->lock)
+__acquires(bitmap->lock)
{
/* If 'create', we might release the lock and reclaim it.
* The lock must have been taken with interrupts enabled.
@@ -1621,10 +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/linear.c b/drivers/md/linear.c
index ea484290544..1ceceb334d5 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -108,6 +108,9 @@ static int linear_congested(void *data, int bits)
linear_conf_t *conf;
int i, ret = 0;
+ if (mddev_congested(mddev, bits))
+ return 1;
+
rcu_read_lock();
conf = rcu_dereference(mddev->private);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 6aa497e4baf..b182f86a19d 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -262,6 +262,12 @@ static void mddev_resume(mddev_t *mddev)
mddev->pers->quiesce(mddev, 0);
}
+int mddev_congested(mddev_t *mddev, int bits)
+{
+ return mddev->suspended;
+}
+EXPORT_SYMBOL(mddev_congested);
+
static inline mddev_t *mddev_get(mddev_t *mddev)
{
@@ -938,6 +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);
@@ -1026,8 +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++;
@@ -1037,16 +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 {
@@ -1376,8 +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);
@@ -1911,6 +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 &&
@@ -2625,7 +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);
}
@@ -4218,7 +4249,7 @@ static int do_md_run(mddev_t * mddev)
set_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
mddev->sync_thread = md_register_thread(md_do_sync,
mddev,
- "%s_resync");
+ "resync");
if (!mddev->sync_thread) {
printk(KERN_ERR "%s: could not start resync"
" thread...\n",
@@ -4575,10 +4606,10 @@ static int get_version(void __user * arg)
static int get_array_info(mddev_t * mddev, void __user * arg)
{
mdu_array_info_t info;
- int nr,working,active,failed,spare;
+ int nr,working,insync,failed,spare;
mdk_rdev_t *rdev;
- nr=working=active=failed=spare=0;
+ nr=working=insync=failed=spare=0;
list_for_each_entry(rdev, &mddev->disks, same_set) {
nr++;
if (test_bit(Faulty, &rdev->flags))
@@ -4586,7 +4617,7 @@ static int get_array_info(mddev_t * mddev, void __user * arg)
else {
working++;
if (test_bit(In_sync, &rdev->flags))
- active++;
+ insync++;
else
spare++;
}
@@ -4611,7 +4642,7 @@ static int get_array_info(mddev_t * mddev, void __user * arg)
info.state = (1<<MD_SB_CLEAN);
if (mddev->bitmap && mddev->bitmap_offset)
info.state = (1<<MD_SB_BITMAP_PRESENT);
- info.active_disks = active;
+ info.active_disks = insync;
info.working_disks = working;
info.failed_disks = failed;
info.spare_disks = spare;
@@ -4721,7 +4752,7 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info)
if (!list_empty(&mddev->disks)) {
mdk_rdev_t *rdev0 = list_entry(mddev->disks.next,
mdk_rdev_t, same_set);
- int err = super_types[mddev->major_version]
+ err = super_types[mddev->major_version]
.load_super(rdev, rdev0, mddev->minor_version);
if (err < 0) {
printk(KERN_WARNING
@@ -5631,7 +5662,10 @@ mdk_thread_t *md_register_thread(void (*run) (mddev_t *), mddev_t *mddev,
thread->run = run;
thread->mddev = mddev;
thread->timeout = MAX_SCHEDULE_TIMEOUT;
- thread->tsk = kthread_run(md_thread, thread, name, mdname(thread->mddev));
+ thread->tsk = kthread_run(md_thread, thread,
+ "%s_%s",
+ mdname(thread->mddev),
+ name ?: mddev->pers->name);
if (IS_ERR(thread->tsk)) {
kfree(thread);
return NULL;
@@ -6495,8 +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);
@@ -6745,7 +6780,7 @@ void md_check_recovery(mddev_t *mddev)
}
mddev->sync_thread = md_register_thread(md_do_sync,
mddev,
- "%s_resync");
+ "resync");
if (!mddev->sync_thread) {
printk(KERN_ERR "%s: could not start resync"
" thread...\n",
diff --git a/drivers/md/md.h b/drivers/md/md.h
index f55d2ff9513..f184b69ef33 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -430,6 +430,7 @@ extern void md_write_end(mddev_t *mddev);
extern void md_done_sync(mddev_t *mddev, int blocks, int ok);
extern void md_error(mddev_t *mddev, mdk_rdev_t *rdev);
+extern int mddev_congested(mddev_t *mddev, int bits);
extern void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev,
sector_t sector, int size, struct page *page);
extern void md_super_wait(mddev_t *mddev);
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index d2d3fd54cc6..ee7646f974a 100644
--- a/drivers/md/multipath.c
+++ b/drivers/md/multipath.c
@@ -150,7 +150,6 @@ static int multipath_make_request (struct request_queue *q, struct bio * bio)
}
mp_bh = mempool_alloc(conf->pool, GFP_NOIO);
- memset(mp_bh, 0, sizeof(*mp_bh));
mp_bh->master_bio = bio;
mp_bh->mddev = mddev;
@@ -199,6 +198,9 @@ static int multipath_congested(void *data, int bits)
multipath_conf_t *conf = mddev->private;
int i, ret = 0;
+ if (mddev_congested(mddev, bits))
+ return 1;
+
rcu_read_lock();
for (i = 0; i < mddev->raid_disks ; i++) {
mdk_rdev_t *rdev = rcu_dereference(conf->multipaths[i].rdev);
@@ -504,7 +506,7 @@ static int multipath_run (mddev_t *mddev)
}
{
- mddev->thread = md_register_thread(multipathd, mddev, "%s_multipath");
+ mddev->thread = md_register_thread(multipathd, mddev, NULL);
if (!mddev->thread) {
printk(KERN_ERR "multipath: couldn't allocate thread"
" for %s\n", mdname(mddev));
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index f845ed98fec..d3a4ce06015 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -44,6 +44,9 @@ static int raid0_congested(void *data, int bits)
mdk_rdev_t **devlist = conf->devlist;
int i, ret = 0;
+ if (mddev_congested(mddev, bits))
+ return 1;
+
for (i = 0; i < mddev->raid_disks && !ret ; i++) {
struct request_queue *q = bdev_get_queue(devlist[i]->bdev);
@@ -86,7 +89,7 @@ static void dump_zones(mddev_t *mddev)
static int create_strip_zones(mddev_t *mddev)
{
- int i, c, j, err;
+ int i, c, err;
sector_t curr_zone_end, sectors;
mdk_rdev_t *smallest, *rdev1, *rdev2, *rdev, **dev;
struct strip_zone *zone;
@@ -198,6 +201,8 @@ static int create_strip_zones(mddev_t *mddev)
/* now do the other zones */
for (i = 1; i < conf->nr_strip_zones; i++)
{
+ int j;
+
zone = conf->strip_zone + i;
dev = conf->devlist + i * mddev->raid_disks;
@@ -207,7 +212,6 @@ static int create_strip_zones(mddev_t *mddev)
c = 0;
for (j=0; j<cnt; j++) {
- char b[BDEVNAME_SIZE];
rdev = conf->devlist[j];
printk(KERN_INFO "raid0: checking %s ...",
bdevname(rdev->bdev, b));
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index ff7ed333599..a053423785c 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;
@@ -576,6 +576,9 @@ static int raid1_congested(void *data, int bits)
conf_t *conf = mddev->private;
int i, ret = 0;
+ if (mddev_congested(mddev, bits))
+ return 1;
+
rcu_read_lock();
for (i = 0; i < mddev->raid_disks; i++) {
mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
@@ -851,7 +854,7 @@ static int make_request(struct request_queue *q, struct bio * bio)
read_bio->bi_sector = r1_bio->sector + mirror->rdev->data_offset;
read_bio->bi_bdev = mirror->rdev->bdev;
read_bio->bi_end_io = raid1_end_read_request;
- read_bio->bi_rw = READ | do_sync;
+ read_bio->bi_rw = READ | (do_sync << BIO_RW_SYNCIO);
read_bio->bi_private = r1_bio;
generic_make_request(read_bio);
@@ -943,7 +946,8 @@ static int make_request(struct request_queue *q, struct bio * bio)
mbio->bi_sector = r1_bio->sector + conf->mirrors[i].rdev->data_offset;
mbio->bi_bdev = conf->mirrors[i].rdev->bdev;
mbio->bi_end_io = raid1_end_write_request;
- mbio->bi_rw = WRITE | do_barriers | do_sync;
+ mbio->bi_rw = WRITE | (do_barriers << BIO_RW_BARRIER) |
+ (do_sync << BIO_RW_SYNCIO);
mbio->bi_private = r1_bio;
if (behind_pages) {
@@ -1623,7 +1627,8 @@ static void raid1d(mddev_t *mddev)
conf->mirrors[i].rdev->data_offset;
bio->bi_bdev = conf->mirrors[i].rdev->bdev;
bio->bi_end_io = raid1_end_write_request;
- bio->bi_rw = WRITE | do_sync;
+ bio->bi_rw = WRITE |
+ (do_sync << BIO_RW_SYNCIO);
bio->bi_private = r1_bio;
r1_bio->bios[i] = bio;
generic_make_request(bio);
@@ -1672,12 +1677,13 @@ static void raid1d(mddev_t *mddev)
bio->bi_sector = r1_bio->sector + rdev->data_offset;
bio->bi_bdev = rdev->bdev;
bio->bi_end_io = raid1_end_read_request;
- bio->bi_rw = READ | do_sync;
+ bio->bi_rw = READ | (do_sync << BIO_RW_SYNCIO);
bio->bi_private = r1_bio;
unplug = 1;
generic_make_request(bio);
}
}
+ cond_resched();
}
if (unplug)
unplug_slaves(mddev);
@@ -1973,13 +1979,14 @@ 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;
@@ -2047,7 +2054,7 @@ static int run(mddev_t *mddev)
conf->last_used = j;
- mddev->thread = md_register_thread(raid1d, mddev, "%s_raid1");
+ mddev->thread = md_register_thread(raid1d, mddev, NULL);
if (!mddev->thread) {
printk(KERN_ERR
"raid1: couldn't allocate thread for %s\n",
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index d0a2152e064..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;
@@ -631,6 +631,8 @@ static int raid10_congested(void *data, int bits)
conf_t *conf = mddev->private;
int i, ret = 0;
+ if (mddev_congested(mddev, bits))
+ return 1;
rcu_read_lock();
for (i = 0; i < mddev->raid_disks && ret == 0; i++) {
mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
@@ -882,7 +884,7 @@ static int make_request(struct request_queue *q, struct bio * bio)
mirror->rdev->data_offset;
read_bio->bi_bdev = mirror->rdev->bdev;
read_bio->bi_end_io = raid10_end_read_request;
- read_bio->bi_rw = READ | do_sync;
+ read_bio->bi_rw = READ | (do_sync << BIO_RW_SYNCIO);
read_bio->bi_private = r10_bio;
generic_make_request(read_bio);
@@ -950,7 +952,7 @@ static int make_request(struct request_queue *q, struct bio * bio)
conf->mirrors[d].rdev->data_offset;
mbio->bi_bdev = conf->mirrors[d].rdev->bdev;
mbio->bi_end_io = raid10_end_write_request;
- mbio->bi_rw = WRITE | do_sync;
+ mbio->bi_rw = WRITE | (do_sync << BIO_RW_SYNCIO);
mbio->bi_private = r10_bio;
atomic_inc(&r10_bio->remaining);
@@ -1623,13 +1625,14 @@ static void raid10d(mddev_t *mddev)
bio->bi_sector = r10_bio->devs[r10_bio->read_slot].addr
+ rdev->data_offset;
bio->bi_bdev = rdev->bdev;
- bio->bi_rw = READ | do_sync;
+ bio->bi_rw = READ | (do_sync << BIO_RW_SYNCIO);
bio->bi_private = r10_bio;
bio->bi_end_io = raid10_end_read_request;
unplug = 1;
generic_make_request(bio);
}
}
+ cond_resched();
}
if (unplug)
unplug_slaves(mddev);
@@ -1773,7 +1776,7 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
max_sync = RESYNC_PAGES << (PAGE_SHIFT-9);
if (!test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
/* recovery... the complicated one */
- int i, j, k;
+ int j, k;
r10_bio = NULL;
for (i=0 ; i<conf->raid_disks; i++)
@@ -2093,7 +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;
@@ -2130,6 +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;
@@ -2188,7 +2191,7 @@ static int run(mddev_t *mddev)
}
- mddev->thread = md_register_thread(raid10d, mddev, "%s_raid10");
+ mddev->thread = md_register_thread(raid10d, mddev, NULL);
if (!mddev->thread) {
printk(KERN_ERR
"raid10: couldn't allocate thread for %s\n",
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 826eb346735..d29215d966d 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -47,7 +47,9 @@
#include <linux/kthread.h>
#include <linux/raid/pq.h>
#include <linux/async_tx.h>
+#include <linux/async.h>
#include <linux/seq_file.h>
+#include <linux/cpu.h>
#include "md.h"
#include "raid5.h"
#include "bitmap.h"
@@ -154,13 +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;
}
@@ -499,11 +504,18 @@ async_copy_data(int frombio, struct bio *bio, struct page *page,
struct page *bio_page;
int i;
int page_offset;
+ struct async_submit_ctl submit;
+ enum async_tx_flags flags = 0;
if (bio->bi_sector >= sector)
page_offset = (signed)(bio->bi_sector - sector) * 512;
else
page_offset = (signed)(sector - bio->bi_sector) * -512;
+
+ if (frombio)
+ flags |= ASYNC_TX_FENCE;
+ init_async_submit(&submit, flags, tx, NULL, NULL, NULL);
+
bio_for_each_segment(bvl, bio, i) {
int len = bio_iovec_idx(bio, i)->bv_len;
int clen;
@@ -525,15 +537,14 @@ async_copy_data(int frombio, struct bio *bio, struct page *page,
bio_page = bio_iovec_idx(bio, i)->bv_page;
if (frombio)
tx = async_memcpy(page, bio_page, page_offset,
- b_offset, clen,
- ASYNC_TX_DEP_ACK,
- tx, NULL, NULL);
+ b_offset, clen, &submit);
else
tx = async_memcpy(bio_page, page, b_offset,
- page_offset, clen,
- ASYNC_TX_DEP_ACK,
- tx, NULL, NULL);
+ page_offset, clen, &submit);
}
+ /* chain the operations */
+ submit.depend_tx = tx;
+
if (clen < len) /* hit end of page */
break;
page_offset += len;
@@ -592,6 +603,7 @@ static void ops_run_biofill(struct stripe_head *sh)
{
struct dma_async_tx_descriptor *tx = NULL;
raid5_conf_t *conf = sh->raid_conf;
+ struct async_submit_ctl submit;
int i;
pr_debug("%s: stripe %llu\n", __func__,
@@ -615,22 +627,34 @@ static void ops_run_biofill(struct stripe_head *sh)
}
atomic_inc(&sh->count);
- async_trigger_callback(ASYNC_TX_DEP_ACK | ASYNC_TX_ACK, tx,
- ops_complete_biofill, sh);
+ init_async_submit(&submit, ASYNC_TX_ACK, tx, ops_complete_biofill, sh, NULL);
+ async_trigger_callback(&submit);
}
-static void ops_complete_compute5(void *stripe_head_ref)
+static void mark_target_uptodate(struct stripe_head *sh, int target)
{
- struct stripe_head *sh = stripe_head_ref;
- int target = sh->ops.target;
- struct r5dev *tgt = &sh->dev[target];
+ struct r5dev *tgt;
- pr_debug("%s: stripe %llu\n", __func__,
- (unsigned long long)sh->sector);
+ if (target < 0)
+ return;
+ tgt = &sh->dev[target];
set_bit(R5_UPTODATE, &tgt->flags);
BUG_ON(!test_bit(R5_Wantcompute, &tgt->flags));
clear_bit(R5_Wantcompute, &tgt->flags);
+}
+
+static void ops_complete_compute(void *stripe_head_ref)
+{
+ struct stripe_head *sh = stripe_head_ref;
+
+ pr_debug("%s: stripe %llu\n", __func__,
+ (unsigned long long)sh->sector);
+
+ /* mark the computed target(s) as uptodate */
+ mark_target_uptodate(sh, sh->ops.target);
+ mark_target_uptodate(sh, sh->ops.target2);
+
clear_bit(STRIPE_COMPUTE_RUN, &sh->state);
if (sh->check_state == check_state_compute_run)
sh->check_state = check_state_compute_result;
@@ -638,16 +662,24 @@ static void ops_complete_compute5(void *stripe_head_ref)
release_stripe(sh);
}
-static struct dma_async_tx_descriptor *ops_run_compute5(struct stripe_head *sh)
+/* return a pointer to the address conversion region of the scribble buffer */
+static addr_conv_t *to_addr_conv(struct stripe_head *sh,
+ struct raid5_percpu *percpu)
+{
+ return percpu->scribble + sizeof(struct page *) * (sh->disks + 2);
+}
+
+static struct dma_async_tx_descriptor *
+ops_run_compute5(struct stripe_head *sh, struct raid5_percpu *percpu)
{
- /* kernel stack size limits the total number of disks */
int disks = sh->disks;
- struct page *xor_srcs[disks];
+ struct page **xor_srcs = percpu->scribble;
int target = sh->ops.target;
struct r5dev *tgt = &sh->dev[target];
struct page *xor_dest = tgt->page;
int count = 0;
struct dma_async_tx_descriptor *tx;
+ struct async_submit_ctl submit;
int i;
pr_debug("%s: stripe %llu block: %d\n",
@@ -660,17 +692,213 @@ static struct dma_async_tx_descriptor *ops_run_compute5(struct stripe_head *sh)
atomic_inc(&sh->count);
+ init_async_submit(&submit, ASYNC_TX_FENCE|ASYNC_TX_XOR_ZERO_DST, NULL,
+ ops_complete_compute, sh, to_addr_conv(sh, percpu));
if (unlikely(count == 1))
- tx = async_memcpy(xor_dest, xor_srcs[0], 0, 0, STRIPE_SIZE,
- 0, NULL, ops_complete_compute5, sh);
+ tx = async_memcpy(xor_dest, xor_srcs[0], 0, 0, STRIPE_SIZE, &submit);
+ else
+ tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE, &submit);
+
+ return tx;
+}
+
+/* set_syndrome_sources - populate source buffers for gen_syndrome
+ * @srcs - (struct page *) array of size sh->disks
+ * @sh - stripe_head to parse
+ *
+ * Populates srcs in proper layout order for the stripe and returns the
+ * 'count' of sources to be used in a call to async_gen_syndrome. The P
+ * destination buffer is recorded in srcs[count] and the Q destination
+ * is recorded in srcs[count+1]].
+ */
+static int set_syndrome_sources(struct page **srcs, struct stripe_head *sh)
+{
+ int disks = sh->disks;
+ int syndrome_disks = sh->ddf_layout ? disks : (disks - 2);
+ int d0_idx = raid6_d0(sh);
+ int count;
+ int i;
+
+ for (i = 0; i < disks; i++)
+ srcs[i] = NULL;
+
+ count = 0;
+ i = d0_idx;
+ do {
+ int slot = raid6_idx_to_slot(i, sh, &count, syndrome_disks);
+
+ srcs[slot] = sh->dev[i].page;
+ i = raid6_next_disk(i, disks);
+ } while (i != d0_idx);
+
+ return syndrome_disks;
+}
+
+static struct dma_async_tx_descriptor *
+ops_run_compute6_1(struct stripe_head *sh, struct raid5_percpu *percpu)
+{
+ int disks = sh->disks;
+ struct page **blocks = percpu->scribble;
+ int target;
+ int qd_idx = sh->qd_idx;
+ struct dma_async_tx_descriptor *tx;
+ struct async_submit_ctl submit;
+ struct r5dev *tgt;
+ struct page *dest;
+ int i;
+ int count;
+
+ if (sh->ops.target < 0)
+ target = sh->ops.target2;
+ else if (sh->ops.target2 < 0)
+ target = sh->ops.target;
else
- tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
- ASYNC_TX_XOR_ZERO_DST, NULL,
- ops_complete_compute5, sh);
+ /* we should only have one valid target */
+ BUG();
+ BUG_ON(target < 0);
+ pr_debug("%s: stripe %llu block: %d\n",
+ __func__, (unsigned long long)sh->sector, target);
+
+ tgt = &sh->dev[target];
+ BUG_ON(!test_bit(R5_Wantcompute, &tgt->flags));
+ dest = tgt->page;
+
+ atomic_inc(&sh->count);
+
+ if (target == qd_idx) {
+ count = set_syndrome_sources(blocks, sh);
+ blocks[count] = NULL; /* regenerating p is not necessary */
+ BUG_ON(blocks[count+1] != dest); /* q should already be set */
+ init_async_submit(&submit, ASYNC_TX_FENCE, NULL,
+ ops_complete_compute, sh,
+ to_addr_conv(sh, percpu));
+ tx = async_gen_syndrome(blocks, 0, count+2, STRIPE_SIZE, &submit);
+ } else {
+ /* Compute any data- or p-drive using XOR */
+ count = 0;
+ for (i = disks; i-- ; ) {
+ if (i == target || i == qd_idx)
+ continue;
+ blocks[count++] = sh->dev[i].page;
+ }
+
+ init_async_submit(&submit, ASYNC_TX_FENCE|ASYNC_TX_XOR_ZERO_DST,
+ NULL, ops_complete_compute, sh,
+ to_addr_conv(sh, percpu));
+ tx = async_xor(dest, blocks, 0, count, STRIPE_SIZE, &submit);
+ }
return tx;
}
+static struct dma_async_tx_descriptor *
+ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu)
+{
+ int i, count, disks = sh->disks;
+ int syndrome_disks = sh->ddf_layout ? disks : disks-2;
+ int d0_idx = raid6_d0(sh);
+ int faila = -1, failb = -1;
+ int target = sh->ops.target;
+ int target2 = sh->ops.target2;
+ struct r5dev *tgt = &sh->dev[target];
+ struct r5dev *tgt2 = &sh->dev[target2];
+ struct dma_async_tx_descriptor *tx;
+ struct page **blocks = percpu->scribble;
+ struct async_submit_ctl submit;
+
+ pr_debug("%s: stripe %llu block1: %d block2: %d\n",
+ __func__, (unsigned long long)sh->sector, target, target2);
+ BUG_ON(target < 0 || target2 < 0);
+ BUG_ON(!test_bit(R5_Wantcompute, &tgt->flags));
+ BUG_ON(!test_bit(R5_Wantcompute, &tgt2->flags));
+
+ /* we need to open-code set_syndrome_sources to handle the
+ * slot number conversion for 'faila' and 'failb'
+ */
+ for (i = 0; i < disks ; i++)
+ blocks[i] = NULL;
+ count = 0;
+ i = d0_idx;
+ do {
+ int slot = raid6_idx_to_slot(i, sh, &count, syndrome_disks);
+
+ blocks[slot] = sh->dev[i].page;
+
+ if (i == target)
+ faila = slot;
+ if (i == target2)
+ failb = slot;
+ i = raid6_next_disk(i, disks);
+ } while (i != d0_idx);
+
+ BUG_ON(faila == failb);
+ if (failb < faila)
+ swap(faila, failb);
+ pr_debug("%s: stripe: %llu faila: %d failb: %d\n",
+ __func__, (unsigned long long)sh->sector, faila, failb);
+
+ atomic_inc(&sh->count);
+
+ if (failb == syndrome_disks+1) {
+ /* Q disk is one of the missing disks */
+ if (faila == syndrome_disks) {
+ /* Missing P+Q, just recompute */
+ init_async_submit(&submit, ASYNC_TX_FENCE, NULL,
+ ops_complete_compute, sh,
+ to_addr_conv(sh, percpu));
+ return async_gen_syndrome(blocks, 0, syndrome_disks+2,
+ STRIPE_SIZE, &submit);
+ } else {
+ struct page *dest;
+ int data_target;
+ int qd_idx = sh->qd_idx;
+
+ /* Missing D+Q: recompute D from P, then recompute Q */
+ if (target == qd_idx)
+ data_target = target2;
+ else
+ data_target = target;
+
+ count = 0;
+ for (i = disks; i-- ; ) {
+ if (i == data_target || i == qd_idx)
+ continue;
+ blocks[count++] = sh->dev[i].page;
+ }
+ dest = sh->dev[data_target].page;
+ init_async_submit(&submit,
+ ASYNC_TX_FENCE|ASYNC_TX_XOR_ZERO_DST,
+ NULL, NULL, NULL,
+ to_addr_conv(sh, percpu));
+ tx = async_xor(dest, blocks, 0, count, STRIPE_SIZE,
+ &submit);
+
+ count = set_syndrome_sources(blocks, sh);
+ init_async_submit(&submit, ASYNC_TX_FENCE, tx,
+ ops_complete_compute, sh,
+ to_addr_conv(sh, percpu));
+ return async_gen_syndrome(blocks, 0, count+2,
+ STRIPE_SIZE, &submit);
+ }
+ } else {
+ init_async_submit(&submit, ASYNC_TX_FENCE, NULL,
+ ops_complete_compute, sh,
+ to_addr_conv(sh, percpu));
+ if (failb == syndrome_disks) {
+ /* We're missing D+P. */
+ return async_raid6_datap_recov(syndrome_disks+2,
+ STRIPE_SIZE, faila,
+ blocks, &submit);
+ } else {
+ /* We're missing D+D. */
+ return async_raid6_2data_recov(syndrome_disks+2,
+ STRIPE_SIZE, faila, failb,
+ blocks, &submit);
+ }
+ }
+}
+
+
static void ops_complete_prexor(void *stripe_head_ref)
{
struct stripe_head *sh = stripe_head_ref;
@@ -680,12 +908,13 @@ static void ops_complete_prexor(void *stripe_head_ref)
}
static struct dma_async_tx_descriptor *
-ops_run_prexor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
+ops_run_prexor(struct stripe_head *sh, struct raid5_percpu *percpu,
+ struct dma_async_tx_descriptor *tx)
{
- /* kernel stack size limits the total number of disks */
int disks = sh->disks;
- struct page *xor_srcs[disks];
+ struct page **xor_srcs = percpu->scribble;
int count = 0, pd_idx = sh->pd_idx, i;
+ struct async_submit_ctl submit;
/* existing parity data subtracted */
struct page *xor_dest = xor_srcs[count++] = sh->dev[pd_idx].page;
@@ -700,9 +929,9 @@ ops_run_prexor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
xor_srcs[count++] = dev->page;
}
- tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
- ASYNC_TX_DEP_ACK | ASYNC_TX_XOR_DROP_DST, tx,
- ops_complete_prexor, sh);
+ init_async_submit(&submit, ASYNC_TX_FENCE|ASYNC_TX_XOR_DROP_DST, tx,
+ ops_complete_prexor, sh, to_addr_conv(sh, percpu));
+ tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE, &submit);
return tx;
}
@@ -742,17 +971,21 @@ ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
return tx;
}
-static void ops_complete_postxor(void *stripe_head_ref)
+static void ops_complete_reconstruct(void *stripe_head_ref)
{
struct stripe_head *sh = stripe_head_ref;
- int disks = sh->disks, i, pd_idx = sh->pd_idx;
+ int disks = sh->disks;
+ int pd_idx = sh->pd_idx;
+ int qd_idx = sh->qd_idx;
+ int i;
pr_debug("%s: stripe %llu\n", __func__,
(unsigned long long)sh->sector);
for (i = disks; i--; ) {
struct r5dev *dev = &sh->dev[i];
- if (dev->written || i == pd_idx)
+
+ if (dev->written || i == pd_idx || i == qd_idx)
set_bit(R5_UPTODATE, &dev->flags);
}
@@ -770,12 +1003,12 @@ static void ops_complete_postxor(void *stripe_head_ref)
}
static void
-ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
+ops_run_reconstruct5(struct stripe_head *sh, struct raid5_percpu *percpu,
+ struct dma_async_tx_descriptor *tx)
{
- /* kernel stack size limits the total number of disks */
int disks = sh->disks;
- struct page *xor_srcs[disks];
-
+ struct page **xor_srcs = percpu->scribble;
+ struct async_submit_ctl submit;
int count = 0, pd_idx = sh->pd_idx, i;
struct page *xor_dest;
int prexor = 0;
@@ -809,18 +1042,36 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
* set ASYNC_TX_XOR_DROP_DST and ASYNC_TX_XOR_ZERO_DST
* for the synchronous xor case
*/
- flags = ASYNC_TX_DEP_ACK | ASYNC_TX_ACK |
+ flags = ASYNC_TX_ACK |
(prexor ? ASYNC_TX_XOR_DROP_DST : ASYNC_TX_XOR_ZERO_DST);
atomic_inc(&sh->count);
- if (unlikely(count == 1)) {
- flags &= ~(ASYNC_TX_XOR_DROP_DST | ASYNC_TX_XOR_ZERO_DST);
- tx = async_memcpy(xor_dest, xor_srcs[0], 0, 0, STRIPE_SIZE,
- flags, tx, ops_complete_postxor, sh);
- } else
- tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
- flags, tx, ops_complete_postxor, sh);
+ init_async_submit(&submit, flags, tx, ops_complete_reconstruct, sh,
+ to_addr_conv(sh, percpu));
+ if (unlikely(count == 1))
+ tx = async_memcpy(xor_dest, xor_srcs[0], 0, 0, STRIPE_SIZE, &submit);
+ else
+ tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE, &submit);
+}
+
+static void
+ops_run_reconstruct6(struct stripe_head *sh, struct raid5_percpu *percpu,
+ struct dma_async_tx_descriptor *tx)
+{
+ struct async_submit_ctl submit;
+ struct page **blocks = percpu->scribble;
+ int count;
+
+ pr_debug("%s: stripe %llu\n", __func__, (unsigned long long)sh->sector);
+
+ count = set_syndrome_sources(blocks, sh);
+
+ atomic_inc(&sh->count);
+
+ init_async_submit(&submit, ASYNC_TX_ACK, tx, ops_complete_reconstruct,
+ sh, to_addr_conv(sh, percpu));
+ async_gen_syndrome(blocks, 0, count+2, STRIPE_SIZE, &submit);
}
static void ops_complete_check(void *stripe_head_ref)
@@ -835,63 +1086,115 @@ static void ops_complete_check(void *stripe_head_ref)
release_stripe(sh);
}
-static void ops_run_check(struct stripe_head *sh)
+static void ops_run_check_p(struct stripe_head *sh, struct raid5_percpu *percpu)
{
- /* kernel stack size limits the total number of disks */
int disks = sh->disks;
- struct page *xor_srcs[disks];
+ int pd_idx = sh->pd_idx;
+ int qd_idx = sh->qd_idx;
+ struct page *xor_dest;
+ struct page **xor_srcs = percpu->scribble;
struct dma_async_tx_descriptor *tx;
-
- int count = 0, pd_idx = sh->pd_idx, i;
- struct page *xor_dest = xor_srcs[count++] = sh->dev[pd_idx].page;
+ struct async_submit_ctl submit;
+ int count;
+ int i;
pr_debug("%s: stripe %llu\n", __func__,
(unsigned long long)sh->sector);
+ count = 0;
+ xor_dest = sh->dev[pd_idx].page;
+ xor_srcs[count++] = xor_dest;
for (i = disks; i--; ) {
- struct r5dev *dev = &sh->dev[i];
- if (i != pd_idx)
- xor_srcs[count++] = dev->page;
+ if (i == pd_idx || i == qd_idx)
+ continue;
+ xor_srcs[count++] = sh->dev[i].page;
}
- tx = async_xor_zero_sum(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
- &sh->ops.zero_sum_result, 0, NULL, NULL, NULL);
+ init_async_submit(&submit, 0, NULL, NULL, NULL,
+ to_addr_conv(sh, percpu));
+ tx = async_xor_val(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
+ &sh->ops.zero_sum_result, &submit);
atomic_inc(&sh->count);
- tx = async_trigger_callback(ASYNC_TX_DEP_ACK | ASYNC_TX_ACK, tx,
- ops_complete_check, sh);
+ init_async_submit(&submit, ASYNC_TX_ACK, tx, ops_complete_check, sh, NULL);
+ tx = async_trigger_callback(&submit);
}
-static void raid5_run_ops(struct stripe_head *sh, unsigned long ops_request)
+static void ops_run_check_pq(struct stripe_head *sh, struct raid5_percpu *percpu, int checkp)
+{
+ struct page **srcs = percpu->scribble;
+ struct async_submit_ctl submit;
+ int count;
+
+ pr_debug("%s: stripe %llu checkp: %d\n", __func__,
+ (unsigned long long)sh->sector, checkp);
+
+ count = set_syndrome_sources(srcs, sh);
+ if (!checkp)
+ srcs[count] = NULL;
+
+ atomic_inc(&sh->count);
+ init_async_submit(&submit, ASYNC_TX_ACK, NULL, ops_complete_check,
+ sh, to_addr_conv(sh, percpu));
+ async_syndrome_val(srcs, 0, count+2, STRIPE_SIZE,
+ &sh->ops.zero_sum_result, percpu->spare_page, &submit);
+}
+
+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;
+ raid5_conf_t *conf = sh->raid_conf;
+ int level = conf->level;
+ struct raid5_percpu *percpu;
+ unsigned long cpu;
+ cpu = get_cpu();
+ percpu = per_cpu_ptr(conf->percpu, cpu);
if (test_bit(STRIPE_OP_BIOFILL, &ops_request)) {
ops_run_biofill(sh);
overlap_clear++;
}
if (test_bit(STRIPE_OP_COMPUTE_BLK, &ops_request)) {
- tx = ops_run_compute5(sh);
- /* terminate the chain if postxor is not set to be run */
- if (tx && !test_bit(STRIPE_OP_POSTXOR, &ops_request))
+ if (level < 6)
+ tx = ops_run_compute5(sh, percpu);
+ else {
+ if (sh->ops.target2 < 0 || sh->ops.target < 0)
+ tx = ops_run_compute6_1(sh, percpu);
+ else
+ tx = ops_run_compute6_2(sh, percpu);
+ }
+ /* terminate the chain if reconstruct is not set to be run */
+ if (tx && !test_bit(STRIPE_OP_RECONSTRUCT, &ops_request))
async_tx_ack(tx);
}
if (test_bit(STRIPE_OP_PREXOR, &ops_request))
- tx = ops_run_prexor(sh, tx);
+ tx = ops_run_prexor(sh, percpu, tx);
if (test_bit(STRIPE_OP_BIODRAIN, &ops_request)) {
tx = ops_run_biodrain(sh, tx);
overlap_clear++;
}
- if (test_bit(STRIPE_OP_POSTXOR, &ops_request))
- ops_run_postxor(sh, tx);
+ if (test_bit(STRIPE_OP_RECONSTRUCT, &ops_request)) {
+ if (level < 6)
+ ops_run_reconstruct5(sh, percpu, tx);
+ else
+ ops_run_reconstruct6(sh, percpu, tx);
+ }
- if (test_bit(STRIPE_OP_CHECK, &ops_request))
- ops_run_check(sh);
+ if (test_bit(STRIPE_OP_CHECK, &ops_request)) {
+ if (sh->check_state == check_state_run)
+ ops_run_check_p(sh, percpu);
+ else if (sh->check_state == check_state_run_q)
+ ops_run_check_pq(sh, percpu, 0);
+ else if (sh->check_state == check_state_run_pq)
+ ops_run_check_pq(sh, percpu, 1);
+ else
+ BUG();
+ }
if (overlap_clear)
for (i = disks; i--; ) {
@@ -899,24 +1202,58 @@ static void raid5_run_ops(struct stripe_head *sh, unsigned long ops_request)
if (test_and_clear_bit(R5_Overlap, &dev->flags))
wake_up(&sh->raid_conf->wait_for_overlap);
}
+ 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);
@@ -928,7 +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));
@@ -948,6 +1285,28 @@ static int grow_stripes(raid5_conf_t *conf, int num)
return 0;
}
+/**
+ * scribble_len - return the required size of the scribble region
+ * @num - total number of disks in the array
+ *
+ * The size must be enough to contain:
+ * 1/ a struct page pointer for each device in the array +2
+ * 2/ room to convert each entry in (1) to its corresponding dma
+ * (dma_map_page()) or page (page_address()) address.
+ *
+ * Note: the +2 is for the destination buffers of the ddf/raid6 case where we
+ * calculate over all devices (not just the data blocks), using zeros in place
+ * of the P and Q blocks.
+ */
+static size_t scribble_len(int num)
+{
+ size_t len;
+
+ len = sizeof(struct page *) * (num+2) + sizeof(addr_conv_t) * (num+2);
+
+ return len;
+}
+
static int resize_stripes(raid5_conf_t *conf, int newsize)
{
/* Make all the stripes able to hold 'newsize' devices.
@@ -976,6 +1335,7 @@ static int resize_stripes(raid5_conf_t *conf, int newsize)
struct stripe_head *osh, *nsh;
LIST_HEAD(newstripes);
struct disk_info *ndisks;
+ unsigned long cpu;
int err;
struct kmem_cache *sc;
int i;
@@ -1003,6 +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);
}
@@ -1041,7 +1404,7 @@ static int resize_stripes(raid5_conf_t *conf, int newsize)
/* Step 3.
* At this point, we are holding all the stripes so the array
* is completely stalled, so now is a good time to resize
- * conf->disks.
+ * conf->disks and the scribble region
*/
ndisks = kzalloc(newsize * sizeof(struct disk_info), GFP_NOIO);
if (ndisks) {
@@ -1052,10 +1415,30 @@ static int resize_stripes(raid5_conf_t *conf, int newsize)
} else
err = -ENOMEM;
+ get_online_cpus();
+ conf->scribble_len = scribble_len(newsize);
+ for_each_present_cpu(cpu) {
+ struct raid5_percpu *percpu;
+ void *scribble;
+
+ percpu = per_cpu_ptr(conf->percpu, cpu);
+ scribble = kmalloc(conf->scribble_len, GFP_NOIO);
+
+ if (scribble) {
+ kfree(percpu->scribble);
+ percpu->scribble = scribble;
+ } else {
+ err = -ENOMEM;
+ break;
+ }
+ }
+ put_online_cpus();
+
/* Step 4, return new stripes to service */
while(!list_empty(&newstripes)) {
nsh = list_entry(newstripes.next, struct stripe_head, lru);
list_del_init(&nsh->lru);
+
for (i=conf->raid_disks; i < newsize; i++)
if (nsh->dev[i].page == NULL) {
struct page *p = alloc_page(GFP_NOIO);
@@ -1553,10 +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:
@@ -1594,258 +1982,13 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous)
}
-
-/*
- * Copy data between a page in the stripe cache, and one or more bion
- * The page could align with the middle of the bio, or there could be
- * several bion, each with several bio_vecs, which cover part of the page
- * Multiple bion are linked together on bi_next. There may be extras
- * at the end of this list. We ignore them.
- */
-static void copy_data(int frombio, struct bio *bio,
- struct page *page,
- sector_t sector)
-{
- char *pa = page_address(page);
- struct bio_vec *bvl;
- int i;
- int page_offset;
-
- if (bio->bi_sector >= sector)
- page_offset = (signed)(bio->bi_sector - sector) * 512;
- else
- page_offset = (signed)(sector - bio->bi_sector) * -512;
- bio_for_each_segment(bvl, bio, i) {
- int len = bio_iovec_idx(bio,i)->bv_len;
- int clen;
- int b_offset = 0;
-
- if (page_offset < 0) {
- b_offset = -page_offset;
- page_offset += b_offset;
- len -= b_offset;
- }
-
- if (len > 0 && page_offset + len > STRIPE_SIZE)
- clen = STRIPE_SIZE - page_offset;
- else clen = len;
-
- if (clen > 0) {
- char *ba = __bio_kmap_atomic(bio, i, KM_USER0);
- if (frombio)
- memcpy(pa+page_offset, ba+b_offset, clen);
- else
- memcpy(ba+b_offset, pa+page_offset, clen);
- __bio_kunmap_atomic(ba, KM_USER0);
- }
- if (clen < len) /* hit end of page */
- break;
- page_offset += len;
- }
-}
-
-#define check_xor() do { \
- if (count == MAX_XOR_BLOCKS) { \
- xor_blocks(count, STRIPE_SIZE, dest, ptr);\
- count = 0; \
- } \
- } while(0)
-
-static void compute_parity6(struct stripe_head *sh, int method)
-{
- raid5_conf_t *conf = sh->raid_conf;
- int i, pd_idx, qd_idx, d0_idx, disks = sh->disks, count;
- int syndrome_disks = sh->ddf_layout ? disks : (disks - 2);
- struct bio *chosen;
- /**** FIX THIS: This could be very bad if disks is close to 256 ****/
- void *ptrs[syndrome_disks+2];
-
- pd_idx = sh->pd_idx;
- qd_idx = sh->qd_idx;
- d0_idx = raid6_d0(sh);
-
- pr_debug("compute_parity, stripe %llu, method %d\n",
- (unsigned long long)sh->sector, method);
-
- switch(method) {
- case READ_MODIFY_WRITE:
- BUG(); /* READ_MODIFY_WRITE N/A for RAID-6 */
- case RECONSTRUCT_WRITE:
- for (i= disks; i-- ;)
- if ( i != pd_idx && i != qd_idx && sh->dev[i].towrite ) {
- chosen = sh->dev[i].towrite;
- sh->dev[i].towrite = NULL;
-
- if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
- wake_up(&conf->wait_for_overlap);
-
- BUG_ON(sh->dev[i].written);
- sh->dev[i].written = chosen;
- }
- break;
- case CHECK_PARITY:
- BUG(); /* Not implemented yet */
- }
-
- for (i = disks; i--;)
- if (sh->dev[i].written) {
- sector_t sector = sh->dev[i].sector;
- struct bio *wbi = sh->dev[i].written;
- while (wbi && wbi->bi_sector < sector + STRIPE_SECTORS) {
- copy_data(1, wbi, sh->dev[i].page, sector);
- wbi = r5_next_bio(wbi, sector);
- }
-
- set_bit(R5_LOCKED, &sh->dev[i].flags);
- set_bit(R5_UPTODATE, &sh->dev[i].flags);
- }
-
- /* Note that unlike RAID-5, the ordering of the disks matters greatly.*/
-
- for (i = 0; i < disks; i++)
- ptrs[i] = (void *)raid6_empty_zero_page;
-
- count = 0;
- i = d0_idx;
- do {
- int slot = raid6_idx_to_slot(i, sh, &count, syndrome_disks);
-
- ptrs[slot] = page_address(sh->dev[i].page);
- if (slot < syndrome_disks &&
- !test_bit(R5_UPTODATE, &sh->dev[i].flags)) {
- printk(KERN_ERR "block %d/%d not uptodate "
- "on parity calc\n", i, count);
- BUG();
- }
-
- i = raid6_next_disk(i, disks);
- } while (i != d0_idx);
- BUG_ON(count != syndrome_disks);
-
- raid6_call.gen_syndrome(syndrome_disks+2, STRIPE_SIZE, ptrs);
-
- switch(method) {
- case RECONSTRUCT_WRITE:
- set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
- set_bit(R5_UPTODATE, &sh->dev[qd_idx].flags);
- set_bit(R5_LOCKED, &sh->dev[pd_idx].flags);
- set_bit(R5_LOCKED, &sh->dev[qd_idx].flags);
- break;
- case UPDATE_PARITY:
- set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
- set_bit(R5_UPTODATE, &sh->dev[qd_idx].flags);
- break;
- }
-}
-
-
-/* Compute one missing block */
-static void compute_block_1(struct stripe_head *sh, int dd_idx, int nozero)
-{
- int i, count, disks = sh->disks;
- void *ptr[MAX_XOR_BLOCKS], *dest, *p;
- int qd_idx = sh->qd_idx;
-
- pr_debug("compute_block_1, stripe %llu, idx %d\n",
- (unsigned long long)sh->sector, dd_idx);
-
- if ( dd_idx == qd_idx ) {
- /* We're actually computing the Q drive */
- compute_parity6(sh, UPDATE_PARITY);
- } else {
- dest = page_address(sh->dev[dd_idx].page);
- if (!nozero) memset(dest, 0, STRIPE_SIZE);
- count = 0;
- for (i = disks ; i--; ) {
- if (i == dd_idx || i == qd_idx)
- continue;
- p = page_address(sh->dev[i].page);
- if (test_bit(R5_UPTODATE, &sh->dev[i].flags))
- ptr[count++] = p;
- else
- printk("compute_block() %d, stripe %llu, %d"
- " not present\n", dd_idx,
- (unsigned long long)sh->sector, i);
-
- check_xor();
- }
- if (count)
- xor_blocks(count, STRIPE_SIZE, dest, ptr);
- if (!nozero) set_bit(R5_UPTODATE, &sh->dev[dd_idx].flags);
- else clear_bit(R5_UPTODATE, &sh->dev[dd_idx].flags);
- }
-}
-
-/* Compute two missing blocks */
-static void compute_block_2(struct stripe_head *sh, int dd_idx1, int dd_idx2)
-{
- int i, count, disks = sh->disks;
- int syndrome_disks = sh->ddf_layout ? disks : disks-2;
- int d0_idx = raid6_d0(sh);
- int faila = -1, failb = -1;
- /**** FIX THIS: This could be very bad if disks is close to 256 ****/
- void *ptrs[syndrome_disks+2];
-
- for (i = 0; i < disks ; i++)
- ptrs[i] = (void *)raid6_empty_zero_page;
- count = 0;
- i = d0_idx;
- do {
- int slot = raid6_idx_to_slot(i, sh, &count, syndrome_disks);
-
- ptrs[slot] = page_address(sh->dev[i].page);
-
- if (i == dd_idx1)
- faila = slot;
- if (i == dd_idx2)
- failb = slot;
- i = raid6_next_disk(i, disks);
- } while (i != d0_idx);
- BUG_ON(count != syndrome_disks);
-
- BUG_ON(faila == failb);
- if ( failb < faila ) { int tmp = faila; faila = failb; failb = tmp; }
-
- pr_debug("compute_block_2, stripe %llu, idx %d,%d (%d,%d)\n",
- (unsigned long long)sh->sector, dd_idx1, dd_idx2,
- faila, failb);
-
- if (failb == syndrome_disks+1) {
- /* Q disk is one of the missing disks */
- if (faila == syndrome_disks) {
- /* Missing P+Q, just recompute */
- compute_parity6(sh, UPDATE_PARITY);
- return;
- } else {
- /* We're missing D+Q; recompute D from P */
- compute_block_1(sh, ((dd_idx1 == sh->qd_idx) ?
- dd_idx2 : dd_idx1),
- 0);
- compute_parity6(sh, UPDATE_PARITY); /* Is this necessary? */
- return;
- }
- }
-
- /* We're missing D+P or D+D; */
- if (failb == syndrome_disks) {
- /* We're missing D+P. */
- raid6_datap_recov(syndrome_disks+2, STRIPE_SIZE, faila, ptrs);
- } else {
- /* We're missing D+D. */
- raid6_2data_recov(syndrome_disks+2, STRIPE_SIZE, faila, failb,
- ptrs);
- }
-
- /* Both the above update both missing blocks */
- set_bit(R5_UPTODATE, &sh->dev[dd_idx1].flags);
- set_bit(R5_UPTODATE, &sh->dev[dd_idx2].flags);
-}
-
static void
-schedule_reconstruction5(struct stripe_head *sh, struct stripe_head_state *s,
+schedule_reconstruction(struct stripe_head *sh, struct stripe_head_state *s,
int rcw, int expand)
{
int i, pd_idx = sh->pd_idx, disks = sh->disks;
+ raid5_conf_t *conf = sh->raid_conf;
+ int level = conf->level;
if (rcw) {
/* if we are not expanding this is a proper write request, and
@@ -1858,7 +2001,7 @@ schedule_reconstruction5(struct stripe_head *sh, struct stripe_head_state *s,
} else
sh->reconstruct_state = reconstruct_state_run;
- set_bit(STRIPE_OP_POSTXOR, &s->ops_request);
+ set_bit(STRIPE_OP_RECONSTRUCT, &s->ops_request);
for (i = disks; i--; ) {
struct r5dev *dev = &sh->dev[i];
@@ -1871,17 +2014,18 @@ schedule_reconstruction5(struct stripe_head *sh, struct stripe_head_state *s,
s->locked++;
}
}
- if (s->locked + 1 == disks)
+ if (s->locked + conf->max_degraded == disks)
if (!test_and_set_bit(STRIPE_FULL_WRITE, &sh->state))
- atomic_inc(&sh->raid_conf->pending_full_writes);
+ atomic_inc(&conf->pending_full_writes);
} else {
+ BUG_ON(level == 6);
BUG_ON(!(test_bit(R5_UPTODATE, &sh->dev[pd_idx].flags) ||
test_bit(R5_Wantcompute, &sh->dev[pd_idx].flags)));
sh->reconstruct_state = reconstruct_state_prexor_drain_run;
set_bit(STRIPE_OP_PREXOR, &s->ops_request);
set_bit(STRIPE_OP_BIODRAIN, &s->ops_request);
- set_bit(STRIPE_OP_POSTXOR, &s->ops_request);
+ set_bit(STRIPE_OP_RECONSTRUCT, &s->ops_request);
for (i = disks; i--; ) {
struct r5dev *dev = &sh->dev[i];
@@ -1899,13 +2043,22 @@ schedule_reconstruction5(struct stripe_head *sh, struct stripe_head_state *s,
}
}
- /* keep the parity disk locked while asynchronous operations
+ /* keep the parity disk(s) locked while asynchronous operations
* are in flight
*/
set_bit(R5_LOCKED, &sh->dev[pd_idx].flags);
clear_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
s->locked++;
+ if (level == 6) {
+ int qd_idx = sh->qd_idx;
+ struct r5dev *dev = &sh->dev[qd_idx];
+
+ set_bit(R5_LOCKED, &dev->flags);
+ clear_bit(R5_UPTODATE, &dev->flags);
+ s->locked++;
+ }
+
pr_debug("%s: stripe %llu locked: %d ops_request: %lx\n",
__func__, (unsigned long long)sh->sector,
s->locked, s->ops_request);
@@ -1986,13 +2139,6 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
static void end_reshape(raid5_conf_t *conf);
-static int page_is_zero(struct page *p)
-{
- char *a = page_address(p);
- return ((*(u32*)a) == 0 &&
- memcmp(a, a+4, STRIPE_SIZE-4)==0);
-}
-
static void stripe_set_idx(sector_t stripe, raid5_conf_t *conf, int previous,
struct stripe_head *sh)
{
@@ -2132,9 +2278,10 @@ static int fetch_block5(struct stripe_head *sh, struct stripe_head_state *s,
set_bit(STRIPE_OP_COMPUTE_BLK, &s->ops_request);
set_bit(R5_Wantcompute, &dev->flags);
sh->ops.target = disk_idx;
+ sh->ops.target2 = -1;
s->req_compute = 1;
/* Careful: from this point on 'uptodate' is in the eye
- * of raid5_run_ops which services 'compute' operations
+ * of raid_run_ops which services 'compute' operations
* before writes. R5_Wantcompute flags a block that will
* be R5_UPTODATE by the time it is needed for a
* subsequent operation.
@@ -2173,61 +2320,104 @@ static void handle_stripe_fill5(struct stripe_head *sh,
set_bit(STRIPE_HANDLE, &sh->state);
}
-static void handle_stripe_fill6(struct stripe_head *sh,
- struct stripe_head_state *s, struct r6_state *r6s,
- int disks)
+/* fetch_block6 - checks the given member device to see if its data needs
+ * to be read or computed to satisfy a request.
+ *
+ * Returns 1 when no more member devices need to be checked, otherwise returns
+ * 0 to tell the loop in handle_stripe_fill6 to continue
+ */
+static int fetch_block6(struct stripe_head *sh, struct stripe_head_state *s,
+ struct r6_state *r6s, int disk_idx, int disks)
{
- int i;
- for (i = disks; i--; ) {
- struct r5dev *dev = &sh->dev[i];
- if (!test_bit(R5_LOCKED, &dev->flags) &&
- !test_bit(R5_UPTODATE, &dev->flags) &&
- (dev->toread || (dev->towrite &&
- !test_bit(R5_OVERWRITE, &dev->flags)) ||
- s->syncing || s->expanding ||
- (s->failed >= 1 &&
- (sh->dev[r6s->failed_num[0]].toread ||
- s->to_write)) ||
- (s->failed >= 2 &&
- (sh->dev[r6s->failed_num[1]].toread ||
- s->to_write)))) {
- /* we would like to get this block, possibly
- * by computing it, but we might not be able to
+ struct r5dev *dev = &sh->dev[disk_idx];
+ struct r5dev *fdev[2] = { &sh->dev[r6s->failed_num[0]],
+ &sh->dev[r6s->failed_num[1]] };
+
+ if (!test_bit(R5_LOCKED, &dev->flags) &&
+ !test_bit(R5_UPTODATE, &dev->flags) &&
+ (dev->toread ||
+ (dev->towrite && !test_bit(R5_OVERWRITE, &dev->flags)) ||
+ s->syncing || s->expanding ||
+ (s->failed >= 1 &&
+ (fdev[0]->toread || s->to_write)) ||
+ (s->failed >= 2 &&
+ (fdev[1]->toread || s->to_write)))) {
+ /* we would like to get this block, possibly by computing it,
+ * otherwise read it if the backing disk is insync
+ */
+ BUG_ON(test_bit(R5_Wantcompute, &dev->flags));
+ BUG_ON(test_bit(R5_Wantread, &dev->flags));
+ if ((s->uptodate == disks - 1) &&
+ (s->failed && (disk_idx == r6s->failed_num[0] ||
+ disk_idx == r6s->failed_num[1]))) {
+ /* have disk failed, and we're requested to fetch it;
+ * do compute it
*/
- if ((s->uptodate == disks - 1) &&
- (s->failed && (i == r6s->failed_num[0] ||
- i == r6s->failed_num[1]))) {
- pr_debug("Computing stripe %llu block %d\n",
- (unsigned long long)sh->sector, i);
- compute_block_1(sh, i, 0);
- s->uptodate++;
- } else if ( s->uptodate == disks-2 && s->failed >= 2 ) {
- /* Computing 2-failure is *very* expensive; only
- * do it if failed >= 2
- */
- int other;
- for (other = disks; other--; ) {
- if (other == i)
- continue;
- if (!test_bit(R5_UPTODATE,
- &sh->dev[other].flags))
- break;
- }
- BUG_ON(other < 0);
- pr_debug("Computing stripe %llu blocks %d,%d\n",
- (unsigned long long)sh->sector,
- i, other);
- compute_block_2(sh, i, other);
- s->uptodate += 2;
- } else if (test_bit(R5_Insync, &dev->flags)) {
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantread, &dev->flags);
- s->locked++;
- pr_debug("Reading block %d (sync=%d)\n",
- i, s->syncing);
+ pr_debug("Computing stripe %llu block %d\n",
+ (unsigned long long)sh->sector, disk_idx);
+ set_bit(STRIPE_COMPUTE_RUN, &sh->state);
+ set_bit(STRIPE_OP_COMPUTE_BLK, &s->ops_request);
+ set_bit(R5_Wantcompute, &dev->flags);
+ sh->ops.target = disk_idx;
+ sh->ops.target2 = -1; /* no 2nd target */
+ s->req_compute = 1;
+ s->uptodate++;
+ return 1;
+ } else if (s->uptodate == disks-2 && s->failed >= 2) {
+ /* Computing 2-failure is *very* expensive; only
+ * do it if failed >= 2
+ */
+ int other;
+ for (other = disks; other--; ) {
+ if (other == disk_idx)
+ continue;
+ if (!test_bit(R5_UPTODATE,
+ &sh->dev[other].flags))
+ break;
}
+ BUG_ON(other < 0);
+ pr_debug("Computing stripe %llu blocks %d,%d\n",
+ (unsigned long long)sh->sector,
+ disk_idx, other);
+ set_bit(STRIPE_COMPUTE_RUN, &sh->state);
+ set_bit(STRIPE_OP_COMPUTE_BLK, &s->ops_request);
+ set_bit(R5_Wantcompute, &sh->dev[disk_idx].flags);
+ set_bit(R5_Wantcompute, &sh->dev[other].flags);
+ sh->ops.target = disk_idx;
+ sh->ops.target2 = other;
+ s->uptodate += 2;
+ s->req_compute = 1;
+ return 1;
+ } else if (test_bit(R5_Insync, &dev->flags)) {
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantread, &dev->flags);
+ s->locked++;
+ pr_debug("Reading block %d (sync=%d)\n",
+ disk_idx, s->syncing);
}
}
+
+ return 0;
+}
+
+/**
+ * handle_stripe_fill6 - read or compute data to satisfy pending requests.
+ */
+static void handle_stripe_fill6(struct stripe_head *sh,
+ struct stripe_head_state *s, struct r6_state *r6s,
+ int disks)
+{
+ int i;
+
+ /* look for blocks to read/compute, skip this if a compute
+ * is already in flight, or if the stripe contents are in the
+ * midst of changing due to a write
+ */
+ if (!test_bit(STRIPE_COMPUTE_RUN, &sh->state) && !sh->check_state &&
+ !sh->reconstruct_state)
+ for (i = disks; i--; )
+ if (fetch_block6(sh, s, r6s, i, disks))
+ break;
set_bit(STRIPE_HANDLE, &sh->state);
}
@@ -2361,114 +2551,61 @@ static void handle_stripe_dirtying5(raid5_conf_t *conf,
*/
/* since handle_stripe can be called at any time we need to handle the
* case where a compute block operation has been submitted and then a
- * subsequent call wants to start a write request. raid5_run_ops only
- * handles the case where compute block and postxor are requested
+ * subsequent call wants to start a write request. raid_run_ops only
+ * handles the case where compute block and reconstruct are requested
* simultaneously. If this is not the case then new writes need to be
* held off until the compute completes.
*/
if ((s->req_compute || !test_bit(STRIPE_COMPUTE_RUN, &sh->state)) &&
(s->locked == 0 && (rcw == 0 || rmw == 0) &&
!test_bit(STRIPE_BIT_DELAY, &sh->state)))
- schedule_reconstruction5(sh, s, rcw == 0, 0);
+ schedule_reconstruction(sh, s, rcw == 0, 0);
}
static void handle_stripe_dirtying6(raid5_conf_t *conf,
struct stripe_head *sh, struct stripe_head_state *s,
struct r6_state *r6s, int disks)
{
- int rcw = 0, must_compute = 0, pd_idx = sh->pd_idx, i;
+ int rcw = 0, pd_idx = sh->pd_idx, i;
int qd_idx = sh->qd_idx;
+
+ set_bit(STRIPE_HANDLE, &sh->state);
for (i = disks; i--; ) {
struct r5dev *dev = &sh->dev[i];
- /* Would I have to read this buffer for reconstruct_write */
- if (!test_bit(R5_OVERWRITE, &dev->flags)
- && i != pd_idx && i != qd_idx
- && (!test_bit(R5_LOCKED, &dev->flags)
- ) &&
- !test_bit(R5_UPTODATE, &dev->flags)) {
- if (test_bit(R5_Insync, &dev->flags)) rcw++;
- else {
- pr_debug("raid6: must_compute: "
- "disk %d flags=%#lx\n", i, dev->flags);
- must_compute++;
+ /* check if we haven't enough data */
+ if (!test_bit(R5_OVERWRITE, &dev->flags) &&
+ i != pd_idx && i != qd_idx &&
+ !test_bit(R5_LOCKED, &dev->flags) &&
+ !(test_bit(R5_UPTODATE, &dev->flags) ||
+ test_bit(R5_Wantcompute, &dev->flags))) {
+ rcw++;
+ if (!test_bit(R5_Insync, &dev->flags))
+ continue; /* it's a failed drive */
+
+ if (
+ test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
+ pr_debug("Read_old stripe %llu "
+ "block %d for Reconstruct\n",
+ (unsigned long long)sh->sector, i);
+ set_bit(R5_LOCKED, &dev->flags);
+ set_bit(R5_Wantread, &dev->flags);
+ s->locked++;
+ } else {
+ pr_debug("Request delayed stripe %llu "
+ "block %d for Reconstruct\n",
+ (unsigned long long)sh->sector, i);
+ set_bit(STRIPE_DELAYED, &sh->state);
+ set_bit(STRIPE_HANDLE, &sh->state);
}
}
}
- pr_debug("for sector %llu, rcw=%d, must_compute=%d\n",
- (unsigned long long)sh->sector, rcw, must_compute);
- set_bit(STRIPE_HANDLE, &sh->state);
-
- if (rcw > 0)
- /* want reconstruct write, but need to get some data */
- for (i = disks; i--; ) {
- struct r5dev *dev = &sh->dev[i];
- if (!test_bit(R5_OVERWRITE, &dev->flags)
- && !(s->failed == 0 && (i == pd_idx || i == qd_idx))
- && !test_bit(R5_LOCKED, &dev->flags) &&
- !test_bit(R5_UPTODATE, &dev->flags) &&
- test_bit(R5_Insync, &dev->flags)) {
- if (
- test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
- pr_debug("Read_old stripe %llu "
- "block %d for Reconstruct\n",
- (unsigned long long)sh->sector, i);
- set_bit(R5_LOCKED, &dev->flags);
- set_bit(R5_Wantread, &dev->flags);
- s->locked++;
- } else {
- pr_debug("Request delayed stripe %llu "
- "block %d for Reconstruct\n",
- (unsigned long long)sh->sector, i);
- set_bit(STRIPE_DELAYED, &sh->state);
- set_bit(STRIPE_HANDLE, &sh->state);
- }
- }
- }
/* now if nothing is locked, and if we have enough data, we can start a
* write request
*/
- if (s->locked == 0 && rcw == 0 &&
+ if ((s->req_compute || !test_bit(STRIPE_COMPUTE_RUN, &sh->state)) &&
+ s->locked == 0 && rcw == 0 &&
!test_bit(STRIPE_BIT_DELAY, &sh->state)) {
- if (must_compute > 0) {
- /* We have failed blocks and need to compute them */
- switch (s->failed) {
- case 0:
- BUG();
- case 1:
- compute_block_1(sh, r6s->failed_num[0], 0);
- break;
- case 2:
- compute_block_2(sh, r6s->failed_num[0],
- r6s->failed_num[1]);
- break;
- default: /* This request should have been failed? */
- BUG();
- }
- }
-
- pr_debug("Computing parity for stripe %llu\n",
- (unsigned long long)sh->sector);
- compute_parity6(sh, RECONSTRUCT_WRITE);
- /* now every locked buffer is ready to be written */
- for (i = disks; i--; )
- if (test_bit(R5_LOCKED, &sh->dev[i].flags)) {
- pr_debug("Writing stripe %llu block %d\n",
- (unsigned long long)sh->sector, i);
- s->locked++;
- set_bit(R5_Wantwrite, &sh->dev[i].flags);
- }
- if (s->locked == disks)
- if (!test_and_set_bit(STRIPE_FULL_WRITE, &sh->state))
- atomic_inc(&conf->pending_full_writes);
- /* after a RECONSTRUCT_WRITE, the stripe MUST be in-sync */
- set_bit(STRIPE_INSYNC, &sh->state);
-
- if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
- atomic_dec(&conf->preread_active_stripes);
- if (atomic_read(&conf->preread_active_stripes) <
- IO_THRESHOLD)
- md_wakeup_thread(conf->mddev->thread);
- }
+ schedule_reconstruction(sh, s, 1, 0);
}
}
@@ -2527,7 +2664,7 @@ static void handle_parity_checks5(raid5_conf_t *conf, struct stripe_head *sh,
* we are done. Otherwise update the mismatch count and repair
* parity if !MD_RECOVERY_CHECK
*/
- if (sh->ops.zero_sum_result == 0)
+ if ((sh->ops.zero_sum_result & SUM_CHECK_P_RESULT) == 0)
/* parity is correct (on disc,
* not in buffer any more)
*/
@@ -2544,6 +2681,7 @@ static void handle_parity_checks5(raid5_conf_t *conf, struct stripe_head *sh,
set_bit(R5_Wantcompute,
&sh->dev[sh->pd_idx].flags);
sh->ops.target = sh->pd_idx;
+ sh->ops.target2 = -1;
s->uptodate++;
}
}
@@ -2560,67 +2698,74 @@ static void handle_parity_checks5(raid5_conf_t *conf, struct stripe_head *sh,
static void handle_parity_checks6(raid5_conf_t *conf, struct stripe_head *sh,
- struct stripe_head_state *s,
- struct r6_state *r6s, struct page *tmp_page,
- int disks)
+ struct stripe_head_state *s,
+ struct r6_state *r6s, int disks)
{
- int update_p = 0, update_q = 0;
- struct r5dev *dev;
int pd_idx = sh->pd_idx;
int qd_idx = sh->qd_idx;
+ struct r5dev *dev;
set_bit(STRIPE_HANDLE, &sh->state);
BUG_ON(s->failed > 2);
- BUG_ON(s->uptodate < disks);
+
/* Want to check and possibly repair P and Q.
* However there could be one 'failed' device, in which
* case we can only check one of them, possibly using the
* other to generate missing data
*/
- /* If !tmp_page, we cannot do the calculations,
- * but as we have set STRIPE_HANDLE, we will soon be called
- * by stripe_handle with a tmp_page - just wait until then.
- */
- if (tmp_page) {
+ switch (sh->check_state) {
+ case check_state_idle:
+ /* start a new check operation if there are < 2 failures */
if (s->failed == r6s->q_failed) {
- /* The only possible failed device holds 'Q', so it
+ /* The only possible failed device holds Q, so it
* makes sense to check P (If anything else were failed,
* we would have used P to recreate it).
*/
- compute_block_1(sh, pd_idx, 1);
- if (!page_is_zero(sh->dev[pd_idx].page)) {
- compute_block_1(sh, pd_idx, 0);
- update_p = 1;
- }
+ sh->check_state = check_state_run;
}
if (!r6s->q_failed && s->failed < 2) {
- /* q is not failed, and we didn't use it to generate
+ /* Q is not failed, and we didn't use it to generate
* anything, so it makes sense to check it
*/
- memcpy(page_address(tmp_page),
- page_address(sh->dev[qd_idx].page),
- STRIPE_SIZE);
- compute_parity6(sh, UPDATE_PARITY);
- if (memcmp(page_address(tmp_page),
- page_address(sh->dev[qd_idx].page),
- STRIPE_SIZE) != 0) {
- clear_bit(STRIPE_INSYNC, &sh->state);
- update_q = 1;
- }
+ if (sh->check_state == check_state_run)
+ sh->check_state = check_state_run_pq;
+ else
+ sh->check_state = check_state_run_q;
}
- if (update_p || update_q) {
- conf->mddev->resync_mismatches += STRIPE_SECTORS;
- if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery))
- /* don't try to repair!! */
- update_p = update_q = 0;
+
+ /* discard potentially stale zero_sum_result */
+ sh->ops.zero_sum_result = 0;
+
+ if (sh->check_state == check_state_run) {
+ /* async_xor_zero_sum destroys the contents of P */
+ clear_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
+ s->uptodate--;
}
+ if (sh->check_state >= check_state_run &&
+ sh->check_state <= check_state_run_pq) {
+ /* async_syndrome_zero_sum preserves P and Q, so
+ * no need to mark them !uptodate here
+ */
+ set_bit(STRIPE_OP_CHECK, &s->ops_request);
+ break;
+ }
+
+ /* we have 2-disk failure */
+ BUG_ON(s->failed != 2);
+ /* fall through */
+ case check_state_compute_result:
+ sh->check_state = check_state_idle;
+
+ /* check that a write has not made the stripe insync */
+ if (test_bit(STRIPE_INSYNC, &sh->state))
+ break;
/* now write out any block on a failed drive,
- * or P or Q if they need it
+ * or P or Q if they were recomputed
*/
-
+ BUG_ON(s->uptodate < disks - 1); /* We don't need Q to recover */
if (s->failed == 2) {
dev = &sh->dev[r6s->failed_num[1]];
s->locked++;
@@ -2633,14 +2778,13 @@ static void handle_parity_checks6(raid5_conf_t *conf, struct stripe_head *sh,
set_bit(R5_LOCKED, &dev->flags);
set_bit(R5_Wantwrite, &dev->flags);
}
-
- if (update_p) {
+ if (sh->ops.zero_sum_result & SUM_CHECK_P_RESULT) {
dev = &sh->dev[pd_idx];
s->locked++;
set_bit(R5_LOCKED, &dev->flags);
set_bit(R5_Wantwrite, &dev->flags);
}
- if (update_q) {
+ if (sh->ops.zero_sum_result & SUM_CHECK_Q_RESULT) {
dev = &sh->dev[qd_idx];
s->locked++;
set_bit(R5_LOCKED, &dev->flags);
@@ -2649,6 +2793,70 @@ static void handle_parity_checks6(raid5_conf_t *conf, struct stripe_head *sh,
clear_bit(STRIPE_DEGRADED, &sh->state);
set_bit(STRIPE_INSYNC, &sh->state);
+ break;
+ case check_state_run:
+ case check_state_run_q:
+ case check_state_run_pq:
+ break; /* we will be called again upon completion */
+ case check_state_check_result:
+ sh->check_state = check_state_idle;
+
+ /* handle a successful check operation, if parity is correct
+ * we are done. Otherwise update the mismatch count and repair
+ * parity if !MD_RECOVERY_CHECK
+ */
+ if (sh->ops.zero_sum_result == 0) {
+ /* both parities are correct */
+ if (!s->failed)
+ set_bit(STRIPE_INSYNC, &sh->state);
+ else {
+ /* in contrast to the raid5 case we can validate
+ * parity, but still have a failure to write
+ * back
+ */
+ sh->check_state = check_state_compute_result;
+ /* Returning at this point means that we may go
+ * off and bring p and/or q uptodate again so
+ * we make sure to check zero_sum_result again
+ * to verify if p or q need writeback
+ */
+ }
+ } else {
+ conf->mddev->resync_mismatches += STRIPE_SECTORS;
+ if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery))
+ /* don't try to repair!! */
+ set_bit(STRIPE_INSYNC, &sh->state);
+ else {
+ int *target = &sh->ops.target;
+
+ sh->ops.target = -1;
+ sh->ops.target2 = -1;
+ sh->check_state = check_state_compute_run;
+ set_bit(STRIPE_COMPUTE_RUN, &sh->state);
+ set_bit(STRIPE_OP_COMPUTE_BLK, &s->ops_request);
+ if (sh->ops.zero_sum_result & SUM_CHECK_P_RESULT) {
+ set_bit(R5_Wantcompute,
+ &sh->dev[pd_idx].flags);
+ *target = pd_idx;
+ target = &sh->ops.target2;
+ s->uptodate++;
+ }
+ if (sh->ops.zero_sum_result & SUM_CHECK_Q_RESULT) {
+ set_bit(R5_Wantcompute,
+ &sh->dev[qd_idx].flags);
+ *target = qd_idx;
+ s->uptodate++;
+ }
+ }
+ }
+ break;
+ case check_state_compute_run:
+ break;
+ default:
+ printk(KERN_ERR "%s: unknown check_state: %d sector: %llu\n",
+ __func__, sh->check_state,
+ (unsigned long long) sh->sector);
+ BUG();
}
}
@@ -2666,6 +2874,7 @@ static void handle_stripe_expansion(raid5_conf_t *conf, struct stripe_head *sh,
if (i != sh->pd_idx && i != sh->qd_idx) {
int dd_idx, j;
struct stripe_head *sh2;
+ struct async_submit_ctl submit;
sector_t bn = compute_blocknr(sh, i, 1);
sector_t s = raid5_compute_sector(conf, bn, 0,
@@ -2685,9 +2894,10 @@ static void handle_stripe_expansion(raid5_conf_t *conf, struct stripe_head *sh,
}
/* place all the copies on one channel */
+ init_async_submit(&submit, 0, tx, NULL, NULL, NULL);
tx = async_memcpy(sh2->dev[dd_idx].page,
- sh->dev[i].page, 0, 0, STRIPE_SIZE,
- ASYNC_TX_DEP_ACK, tx, NULL, NULL);
+ sh->dev[i].page, 0, 0, STRIPE_SIZE,
+ &submit);
set_bit(R5_Expanded, &sh2->dev[dd_idx].flags);
set_bit(R5_UPTODATE, &sh2->dev[dd_idx].flags);
@@ -2728,7 +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;
@@ -2756,7 +2966,8 @@ static bool handle_stripe5(struct stripe_head *sh)
rcu_read_lock();
for (i=disks; i--; ) {
mdk_rdev_t *rdev;
- struct r5dev *dev = &sh->dev[i];
+
+ dev = &sh->dev[i];
clear_bit(R5_Insync, &dev->flags);
pr_debug("check %d: state 0x%lx toread %p read %p write %p "
@@ -2973,7 +3184,7 @@ static bool handle_stripe5(struct stripe_head *sh)
/* Need to write out all blocks after computing parity */
sh->disks = conf->raid_disks;
stripe_set_idx(sh->sector, conf, 0, sh);
- schedule_reconstruction5(sh, &s, 1, 1);
+ schedule_reconstruction(sh, &s, 1, 1);
} else if (s.expanded && !sh->reconstruct_state && s.locked == 0) {
clear_bit(STRIPE_EXPAND_READY, &sh->state);
atomic_dec(&conf->reshape_stripes);
@@ -2993,16 +3204,14 @@ static bool handle_stripe5(struct stripe_head *sh)
md_wait_for_blocked_rdev(blocked_rdev, conf->mddev);
if (s.ops_request)
- raid5_run_ops(sh, s.ops_request);
+ raid_run_ops(sh, s.ops_request);
ops_run_io(sh, &s);
return_io(return_bi);
-
- return blocked_rdev == NULL;
}
-static bool handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
+static void handle_stripe6(struct stripe_head *sh)
{
raid5_conf_t *conf = sh->raid_conf;
int disks = sh->disks;
@@ -3014,9 +3223,10 @@ static bool handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
mdk_rdev_t *blocked_rdev = NULL;
pr_debug("handling stripe %llu, state=%#lx cnt=%d, "
- "pd_idx=%d, qd_idx=%d\n",
+ "pd_idx=%d, qd_idx=%d\n, check:%d, reconstruct:%d\n",
(unsigned long long)sh->sector, sh->state,
- atomic_read(&sh->count), pd_idx, qd_idx);
+ atomic_read(&sh->count), pd_idx, qd_idx,
+ sh->check_state, sh->reconstruct_state);
memset(&s, 0, sizeof(s));
spin_lock(&sh->lock);
@@ -3036,35 +3246,26 @@ static bool handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
pr_debug("check %d: state 0x%lx read %p write %p written %p\n",
i, dev->flags, dev->toread, dev->towrite, dev->written);
- /* maybe we can reply to a read */
- if (test_bit(R5_UPTODATE, &dev->flags) && dev->toread) {
- struct bio *rbi, *rbi2;
- pr_debug("Return read for disc %d\n", i);
- spin_lock_irq(&conf->device_lock);
- rbi = dev->toread;
- dev->toread = NULL;
- if (test_and_clear_bit(R5_Overlap, &dev->flags))
- wake_up(&conf->wait_for_overlap);
- spin_unlock_irq(&conf->device_lock);
- while (rbi && rbi->bi_sector < dev->sector + STRIPE_SECTORS) {
- copy_data(0, rbi, dev->page, dev->sector);
- rbi2 = r5_next_bio(rbi, dev->sector);
- spin_lock_irq(&conf->device_lock);
- if (!raid5_dec_bi_phys_segments(rbi)) {
- rbi->bi_next = return_bi;
- return_bi = rbi;
- }
- spin_unlock_irq(&conf->device_lock);
- rbi = rbi2;
- }
- }
+ /* maybe we can reply to a read
+ *
+ * new wantfill requests are only permitted while
+ * ops_complete_biofill is guaranteed to be inactive
+ */
+ if (test_bit(R5_UPTODATE, &dev->flags) && dev->toread &&
+ !test_bit(STRIPE_BIOFILL_RUN, &sh->state))
+ set_bit(R5_Wantfill, &dev->flags);
/* now count some things */
if (test_bit(R5_LOCKED, &dev->flags)) s.locked++;
if (test_bit(R5_UPTODATE, &dev->flags)) s.uptodate++;
+ if (test_bit(R5_Wantcompute, &dev->flags)) {
+ s.compute++;
+ BUG_ON(s.compute > 2);
+ }
-
- if (dev->toread)
+ if (test_bit(R5_Wantfill, &dev->flags)) {
+ s.to_fill++;
+ } else if (dev->toread)
s.to_read++;
if (dev->towrite) {
s.to_write++;
@@ -3105,6 +3306,11 @@ static bool handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
blocked_rdev = NULL;
}
+ if (s.to_fill && !test_bit(STRIPE_BIOFILL_RUN, &sh->state)) {
+ set_bit(STRIPE_OP_BIOFILL, &s.ops_request);
+ set_bit(STRIPE_BIOFILL_RUN, &sh->state);
+ }
+
pr_debug("locked=%d uptodate=%d to_read=%d"
" to_write=%d failed=%d failed_num=%d,%d\n",
s.locked, s.uptodate, s.to_read, s.to_write, s.failed,
@@ -3145,19 +3351,62 @@ static bool handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
* or to load a block that is being partially written.
*/
if (s.to_read || s.non_overwrite || (s.to_write && s.failed) ||
- (s.syncing && (s.uptodate < disks)) || s.expanding)
+ (s.syncing && (s.uptodate + s.compute < disks)) || s.expanding)
handle_stripe_fill6(sh, &s, &r6s, disks);
- /* now to consider writing and what else, if anything should be read */
- if (s.to_write)
+ /* Now we check to see if any write operations have recently
+ * completed
+ */
+ if (sh->reconstruct_state == reconstruct_state_drain_result) {
+ int qd_idx = sh->qd_idx;
+
+ sh->reconstruct_state = reconstruct_state_idle;
+ /* All the 'written' buffers and the parity blocks are ready to
+ * be written back to disk
+ */
+ BUG_ON(!test_bit(R5_UPTODATE, &sh->dev[sh->pd_idx].flags));
+ BUG_ON(!test_bit(R5_UPTODATE, &sh->dev[qd_idx].flags));
+ for (i = disks; i--; ) {
+ dev = &sh->dev[i];
+ if (test_bit(R5_LOCKED, &dev->flags) &&
+ (i == sh->pd_idx || i == qd_idx ||
+ dev->written)) {
+ pr_debug("Writing block %d\n", i);
+ BUG_ON(!test_bit(R5_UPTODATE, &dev->flags));
+ set_bit(R5_Wantwrite, &dev->flags);
+ if (!test_bit(R5_Insync, &dev->flags) ||
+ ((i == sh->pd_idx || i == qd_idx) &&
+ s.failed == 0))
+ set_bit(STRIPE_INSYNC, &sh->state);
+ }
+ }
+ if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
+ atomic_dec(&conf->preread_active_stripes);
+ if (atomic_read(&conf->preread_active_stripes) <
+ IO_THRESHOLD)
+ md_wakeup_thread(conf->mddev->thread);
+ }
+ }
+
+ /* Now to consider new write requests and what else, if anything
+ * should be read. We do not handle new writes when:
+ * 1/ A 'write' operation (copy+gen_syndrome) is already in flight.
+ * 2/ A 'check' operation is in flight, as it may clobber the parity
+ * block.
+ */
+ if (s.to_write && !sh->reconstruct_state && !sh->check_state)
handle_stripe_dirtying6(conf, sh, &s, &r6s, disks);
/* maybe we need to check and possibly fix the parity for this stripe
* Any reads will already have been scheduled, so we just see if enough
- * data is available
+ * data is available. The parity check is held off while parity
+ * dependent operations are in flight.
*/
- if (s.syncing && s.locked == 0 && !test_bit(STRIPE_INSYNC, &sh->state))
- handle_parity_checks6(conf, sh, &s, &r6s, tmp_page, disks);
+ if (sh->check_state ||
+ (s.syncing && s.locked == 0 &&
+ !test_bit(STRIPE_COMPUTE_RUN, &sh->state) &&
+ !test_bit(STRIPE_INSYNC, &sh->state)))
+ handle_parity_checks6(conf, sh, &s, &r6s, disks);
if (s.syncing && s.locked == 0 && test_bit(STRIPE_INSYNC, &sh->state)) {
md_done_sync(conf->mddev, STRIPE_SECTORS,1);
@@ -3178,15 +3427,29 @@ static bool handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
set_bit(R5_Wantwrite, &dev->flags);
set_bit(R5_ReWrite, &dev->flags);
set_bit(R5_LOCKED, &dev->flags);
+ s.locked++;
} else {
/* let's read it back */
set_bit(R5_Wantread, &dev->flags);
set_bit(R5_LOCKED, &dev->flags);
+ s.locked++;
}
}
}
- if (s.expanded && test_bit(STRIPE_EXPANDING, &sh->state)) {
+ /* Finish reconstruct operations initiated by the expansion process */
+ if (sh->reconstruct_state == reconstruct_state_result) {
+ sh->reconstruct_state = reconstruct_state_idle;
+ clear_bit(STRIPE_EXPANDING, &sh->state);
+ for (i = conf->raid_disks; i--; ) {
+ set_bit(R5_Wantwrite, &sh->dev[i].flags);
+ set_bit(R5_LOCKED, &sh->dev[i].flags);
+ s.locked++;
+ }
+ }
+
+ if (s.expanded && test_bit(STRIPE_EXPANDING, &sh->state) &&
+ !sh->reconstruct_state) {
struct stripe_head *sh2
= get_active_stripe(conf, sh->sector, 1, 1, 1);
if (sh2 && test_bit(STRIPE_EXPAND_SOURCE, &sh2->state)) {
@@ -3207,14 +3470,8 @@ static bool handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
/* Need to write out all blocks after computing P&Q */
sh->disks = conf->raid_disks;
stripe_set_idx(sh->sector, conf, 0, sh);
- compute_parity6(sh, RECONSTRUCT_WRITE);
- for (i = conf->raid_disks ; i-- ; ) {
- set_bit(R5_LOCKED, &sh->dev[i].flags);
- s.locked++;
- set_bit(R5_Wantwrite, &sh->dev[i].flags);
- }
- clear_bit(STRIPE_EXPANDING, &sh->state);
- } else if (s.expanded) {
+ schedule_reconstruction(sh, &s, 1, 1);
+ } else if (s.expanded && !sh->reconstruct_state && s.locked == 0) {
clear_bit(STRIPE_EXPAND_READY, &sh->state);
atomic_dec(&conf->reshape_stripes);
wake_up(&conf->wait_for_overlap);
@@ -3232,24 +3489,22 @@ static bool handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
if (unlikely(blocked_rdev))
md_wait_for_blocked_rdev(blocked_rdev, conf->mddev);
+ if (s.ops_request)
+ raid_run_ops(sh, s.ops_request);
+
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, struct page *tmp_page)
+static void handle_stripe(struct stripe_head *sh)
{
if (sh->raid_conf->level == 6)
- return handle_stripe6(sh, tmp_page);
+ handle_stripe6(sh);
else
- return handle_stripe5(sh);
+ handle_stripe5(sh);
}
-
-
static void raid5_activate_delayed(raid5_conf_t *conf)
{
if (atomic_read(&conf->preread_active_stripes) < IO_THRESHOLD) {
@@ -3285,9 +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);
@@ -3331,6 +3587,9 @@ static int raid5_congested(void *data, int bits)
/* No difference between reads and writes. Just check
* how busy the stripe_cache is
*/
+
+ if (mddev_congested(mddev, bits))
+ return 1;
if (conf->inactive_blocked)
return 1;
if (conf->quiesce)
@@ -3790,6 +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;
}
@@ -3880,7 +4141,7 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
INIT_LIST_HEAD(&stripes);
for (i = 0; i < reshape_sectors; i += STRIPE_SECTORS) {
int j;
- int skipped = 0;
+ int skipped_disk = 0;
sh = get_active_stripe(conf, stripe_addr+i, 0, 0, 1);
set_bit(STRIPE_EXPANDING, &sh->state);
atomic_inc(&conf->reshape_stripes);
@@ -3896,14 +4157,14 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
continue;
s = compute_blocknr(sh, j, 0);
if (s < raid5_size(mddev, 0, 0)) {
- skipped = 1;
+ skipped_disk = 1;
continue;
}
memset(page_address(sh->dev[j].page), 0, STRIPE_SIZE);
set_bit(R5_Expanded, &sh->dev[j].flags);
set_bit(R5_UPTODATE, &sh->dev[j].flags);
}
- if (!skipped) {
+ if (!skipped_disk) {
set_bit(STRIPE_EXPAND_READY, &sh->state);
set_bit(STRIPE_HANDLE, &sh->state);
}
@@ -4056,9 +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, NULL)))
- ;
+ handle_stripe(sh);
release_stripe(sh);
return STRIPE_SECTORS;
@@ -4114,7 +4373,7 @@ static int retry_aligned_read(raid5_conf_t *conf, struct bio *raid_bio)
return handled;
}
- handle_stripe(sh, NULL);
+ handle_stripe(sh);
release_stripe(sh);
handled++;
}
@@ -4129,7 +4388,6 @@ static int retry_aligned_read(raid5_conf_t *conf, struct bio *raid_bio)
}
-
/*
* This is our raid5 kernel thread.
*
@@ -4178,8 +4436,9 @@ static void raid5d(mddev_t *mddev)
spin_unlock_irq(&conf->device_lock);
handled++;
- handle_stripe(sh, conf->spare_page);
+ handle_stripe(sh);
release_stripe(sh);
+ cond_resched();
spin_lock_irq(&conf->device_lock);
}
@@ -4306,32 +4565,131 @@ raid5_size(mddev_t *mddev, sector_t sectors, int raid_disks)
if (!sectors)
sectors = mddev->dev_sectors;
- if (!raid_disks) {
+ if (!raid_disks)
/* size is defined by the smallest of previous and new size */
- if (conf->raid_disks < conf->previous_raid_disks)
- raid_disks = conf->raid_disks;
- else
- raid_disks = conf->previous_raid_disks;
- }
+ raid_disks = min(conf->raid_disks, conf->previous_raid_disks);
sectors &= ~((sector_t)mddev->chunk_sectors - 1);
sectors &= ~((sector_t)mddev->new_chunk_sectors - 1);
return sectors * (raid_disks - conf->max_degraded);
}
+static void raid5_free_percpu(raid5_conf_t *conf)
+{
+ struct raid5_percpu *percpu;
+ unsigned long cpu;
+
+ if (!conf->percpu)
+ return;
+
+ get_online_cpus();
+ for_each_possible_cpu(cpu) {
+ percpu = per_cpu_ptr(conf->percpu, cpu);
+ safe_put_page(percpu->spare_page);
+ kfree(percpu->scribble);
+ }
+#ifdef CONFIG_HOTPLUG_CPU
+ unregister_cpu_notifier(&conf->cpu_notify);
+#endif
+ put_online_cpus();
+
+ free_percpu(conf->percpu);
+}
+
static void free_conf(raid5_conf_t *conf)
{
shrink_stripes(conf);
- safe_put_page(conf->spare_page);
+ raid5_free_percpu(conf);
kfree(conf->disks);
kfree(conf->stripe_hashtbl);
kfree(conf);
}
+#ifdef CONFIG_HOTPLUG_CPU
+static int raid456_cpu_notify(struct notifier_block *nfb, unsigned long action,
+ void *hcpu)
+{
+ raid5_conf_t *conf = container_of(nfb, raid5_conf_t, cpu_notify);
+ long cpu = (long)hcpu;
+ struct raid5_percpu *percpu = per_cpu_ptr(conf->percpu, cpu);
+
+ switch (action) {
+ case CPU_UP_PREPARE:
+ case CPU_UP_PREPARE_FROZEN:
+ if (conf->level == 6 && !percpu->spare_page)
+ percpu->spare_page = alloc_page(GFP_KERNEL);
+ if (!percpu->scribble)
+ percpu->scribble = kmalloc(conf->scribble_len, GFP_KERNEL);
+
+ if (!percpu->scribble ||
+ (conf->level == 6 && !percpu->spare_page)) {
+ safe_put_page(percpu->spare_page);
+ kfree(percpu->scribble);
+ pr_err("%s: failed memory allocation for cpu%ld\n",
+ __func__, cpu);
+ return NOTIFY_BAD;
+ }
+ break;
+ case CPU_DEAD:
+ case CPU_DEAD_FROZEN:
+ safe_put_page(percpu->spare_page);
+ kfree(percpu->scribble);
+ percpu->spare_page = NULL;
+ percpu->scribble = NULL;
+ break;
+ default:
+ break;
+ }
+ return NOTIFY_OK;
+}
+#endif
+
+static int raid5_alloc_percpu(raid5_conf_t *conf)
+{
+ unsigned long cpu;
+ struct page *spare_page;
+ struct raid5_percpu *allcpus;
+ void *scribble;
+ int err;
+
+ allcpus = alloc_percpu(struct raid5_percpu);
+ if (!allcpus)
+ return -ENOMEM;
+ conf->percpu = allcpus;
+
+ get_online_cpus();
+ err = 0;
+ for_each_present_cpu(cpu) {
+ if (conf->level == 6) {
+ spare_page = alloc_page(GFP_KERNEL);
+ if (!spare_page) {
+ err = -ENOMEM;
+ break;
+ }
+ per_cpu_ptr(conf->percpu, cpu)->spare_page = spare_page;
+ }
+ scribble = kmalloc(conf->scribble_len, GFP_KERNEL);
+ if (!scribble) {
+ err = -ENOMEM;
+ break;
+ }
+ per_cpu_ptr(conf->percpu, cpu)->scribble = scribble;
+ }
+#ifdef CONFIG_HOTPLUG_CPU
+ conf->cpu_notify.notifier_call = raid456_cpu_notify;
+ conf->cpu_notify.priority = 0;
+ if (err == 0)
+ err = register_cpu_notifier(&conf->cpu_notify);
+#endif
+ put_online_cpus();
+
+ return err;
+}
+
static raid5_conf_t *setup_conf(mddev_t *mddev)
{
raid5_conf_t *conf;
- int raid_disk, memory;
+ int raid_disk, memory, max_disks;
mdk_rdev_t *rdev;
struct disk_info *disk;
@@ -4367,14 +4725,28 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
conf = kzalloc(sizeof(raid5_conf_t), GFP_KERNEL);
if (conf == NULL)
goto abort;
+ spin_lock_init(&conf->device_lock);
+ init_waitqueue_head(&conf->wait_for_stripe);
+ init_waitqueue_head(&conf->wait_for_overlap);
+ INIT_LIST_HEAD(&conf->handle_list);
+ INIT_LIST_HEAD(&conf->hold_list);
+ INIT_LIST_HEAD(&conf->delayed_list);
+ INIT_LIST_HEAD(&conf->bitmap_list);
+ INIT_LIST_HEAD(&conf->inactive_list);
+ atomic_set(&conf->active_stripes, 0);
+ atomic_set(&conf->preread_active_stripes, 0);
+ atomic_set(&conf->active_aligned_reads, 0);
+ conf->bypass_threshold = BYPASS_THRESHOLD;
conf->raid_disks = mddev->raid_disks;
if (mddev->reshape_position == MaxSector)
conf->previous_raid_disks = mddev->raid_disks;
else
conf->previous_raid_disks = mddev->raid_disks - mddev->delta_disks;
+ max_disks = max(conf->raid_disks, conf->previous_raid_disks);
+ conf->scribble_len = scribble_len(max_disks);
- conf->disks = kzalloc(conf->raid_disks * sizeof(struct disk_info),
+ conf->disks = kzalloc(max_disks * sizeof(struct disk_info),
GFP_KERNEL);
if (!conf->disks)
goto abort;
@@ -4384,29 +4756,15 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
if ((conf->stripe_hashtbl = kzalloc(PAGE_SIZE, GFP_KERNEL)) == NULL)
goto abort;
- if (mddev->new_level == 6) {
- conf->spare_page = alloc_page(GFP_KERNEL);
- if (!conf->spare_page)
- goto abort;
- }
- spin_lock_init(&conf->device_lock);
- init_waitqueue_head(&conf->wait_for_stripe);
- init_waitqueue_head(&conf->wait_for_overlap);
- INIT_LIST_HEAD(&conf->handle_list);
- INIT_LIST_HEAD(&conf->hold_list);
- INIT_LIST_HEAD(&conf->delayed_list);
- INIT_LIST_HEAD(&conf->bitmap_list);
- INIT_LIST_HEAD(&conf->inactive_list);
- atomic_set(&conf->active_stripes, 0);
- atomic_set(&conf->preread_active_stripes, 0);
- atomic_set(&conf->active_aligned_reads, 0);
- conf->bypass_threshold = BYPASS_THRESHOLD;
+ conf->level = mddev->new_level;
+ if (raid5_alloc_percpu(conf) != 0)
+ goto abort;
pr_debug("raid5: run(%s) called.\n", mdname(mddev));
list_for_each_entry(rdev, &mddev->disks, same_set) {
raid_disk = rdev->raid_disk;
- if (raid_disk >= conf->raid_disks
+ if (raid_disk >= max_disks
|| raid_disk < 0)
continue;
disk = conf->disks + raid_disk;
@@ -4438,7 +4796,7 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
}
memory = conf->max_nr_stripes * (sizeof(struct stripe_head) +
- conf->raid_disks * ((sizeof(struct bio) + PAGE_SIZE))) / 1024;
+ max_disks * ((sizeof(struct bio) + PAGE_SIZE))) / 1024;
if (grow_stripes(conf, conf->max_nr_stripes)) {
printk(KERN_ERR
"raid5: couldn't allocate %dkB for buffers\n", memory);
@@ -4447,7 +4805,7 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
printk(KERN_INFO "raid5: allocated %dkB for %s\n",
memory, mdname(mddev));
- conf->thread = md_register_thread(raid5d, mddev, "%s_raid5");
+ conf->thread = md_register_thread(raid5d, mddev, NULL);
if (!conf->thread) {
printk(KERN_ERR
"raid5: couldn't allocate thread for %s\n",
@@ -4465,11 +4823,40 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
return ERR_PTR(-ENOMEM);
}
+
+static int only_parity(int raid_disk, int algo, int raid_disks, int max_degraded)
+{
+ switch (algo) {
+ case ALGORITHM_PARITY_0:
+ if (raid_disk < max_degraded)
+ return 1;
+ break;
+ case ALGORITHM_PARITY_N:
+ if (raid_disk >= raid_disks - max_degraded)
+ return 1;
+ break;
+ case ALGORITHM_PARITY_0_6:
+ if (raid_disk == 0 ||
+ raid_disk == raid_disks - 1)
+ return 1;
+ break;
+ case ALGORITHM_LEFT_ASYMMETRIC_6:
+ case ALGORITHM_RIGHT_ASYMMETRIC_6:
+ case ALGORITHM_LEFT_SYMMETRIC_6:
+ case ALGORITHM_RIGHT_SYMMETRIC_6:
+ if (raid_disk == raid_disks - 1)
+ return 1;
+ }
+ return 0;
+}
+
static int run(mddev_t *mddev)
{
raid5_conf_t *conf;
int working_disks = 0, chunk_size;
+ int dirty_parity_disks = 0;
mdk_rdev_t *rdev;
+ sector_t reshape_offset = 0;
if (mddev->recovery_cp != MaxSector)
printk(KERN_NOTICE "raid5: %s is not clean"
@@ -4503,6 +4890,7 @@ static int run(mddev_t *mddev)
"on a stripe boundary\n");
return -EINVAL;
}
+ reshape_offset = here_new * mddev->new_chunk_sectors;
/* here_new is the stripe we will write to */
here_old = mddev->reshape_position;
sector_div(here_old, mddev->chunk_sectors *
@@ -4558,12 +4946,54 @@ static int run(mddev_t *mddev)
/*
* 0 for a fully functional array, 1 or 2 for a degraded array.
*/
- list_for_each_entry(rdev, &mddev->disks, same_set)
- if (rdev->raid_disk >= 0 &&
- test_bit(In_sync, &rdev->flags))
+ list_for_each_entry(rdev, &mddev->disks, same_set) {
+ if (rdev->raid_disk < 0)
+ continue;
+ if (test_bit(In_sync, &rdev->flags))
working_disks++;
+ /* This disc is not fully in-sync. However if it
+ * just stored parity (beyond the recovery_offset),
+ * when we don't need to be concerned about the
+ * array being dirty.
+ * When reshape goes 'backwards', we never have
+ * partially completed devices, so we only need
+ * to worry about reshape going forwards.
+ */
+ /* Hack because v0.91 doesn't store recovery_offset properly. */
+ if (mddev->major_version == 0 &&
+ mddev->minor_version > 90)
+ rdev->recovery_offset = reshape_offset;
+
+ printk("%d: w=%d pa=%d pr=%d m=%d a=%d r=%d op1=%d op2=%d\n",
+ rdev->raid_disk, working_disks, conf->prev_algo,
+ conf->previous_raid_disks, conf->max_degraded,
+ conf->algorithm, conf->raid_disks,
+ only_parity(rdev->raid_disk,
+ conf->prev_algo,
+ conf->previous_raid_disks,
+ conf->max_degraded),
+ only_parity(rdev->raid_disk,
+ conf->algorithm,
+ conf->raid_disks,
+ conf->max_degraded));
+ if (rdev->recovery_offset < reshape_offset) {
+ /* We need to check old and new layout */
+ if (!only_parity(rdev->raid_disk,
+ conf->algorithm,
+ conf->raid_disks,
+ conf->max_degraded))
+ continue;
+ }
+ if (!only_parity(rdev->raid_disk,
+ conf->prev_algo,
+ conf->previous_raid_disks,
+ conf->max_degraded))
+ continue;
+ dirty_parity_disks++;
+ }
- mddev->degraded = conf->raid_disks - working_disks;
+ mddev->degraded = (max(conf->raid_disks, conf->previous_raid_disks)
+ - working_disks);
if (mddev->degraded > conf->max_degraded) {
printk(KERN_ERR "raid5: not enough operational devices for %s"
@@ -4576,7 +5006,7 @@ static int run(mddev_t *mddev)
mddev->dev_sectors &= ~(mddev->chunk_sectors - 1);
mddev->resync_max_sectors = mddev->dev_sectors;
- if (mddev->degraded > 0 &&
+ if (mddev->degraded > dirty_parity_disks &&
mddev->recovery_cp != MaxSector) {
if (mddev->ok_start_degraded)
printk(KERN_WARNING
@@ -4613,7 +5043,7 @@ static int run(mddev_t *mddev)
set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
set_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
mddev->sync_thread = md_register_thread(md_do_sync, mddev,
- "%s_reshape");
+ "reshape");
}
/* read-ahead size must cover two whole stripes, which is
@@ -5002,9 +5432,11 @@ static int raid5_start_reshape(mddev_t *mddev)
!test_bit(Faulty, &rdev->flags)) {
if (raid5_add_disk(mddev, rdev) == 0) {
char nm[20];
- set_bit(In_sync, &rdev->flags);
+ if (rdev->raid_disk >= conf->previous_raid_disks)
+ set_bit(In_sync, &rdev->flags);
+ else
+ rdev->recovery_offset = 0;
added_devices++;
- rdev->recovery_offset = 0;
sprintf(nm, "rd%d", rdev->raid_disk);
if (sysfs_create_link(&mddev->kobj,
&rdev->kobj, nm))
@@ -5031,7 +5463,7 @@ static int raid5_start_reshape(mddev_t *mddev)
set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
set_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
mddev->sync_thread = md_register_thread(md_do_sync, mddev,
- "%s_reshape");
+ "reshape");
if (!mddev->sync_thread) {
mddev->recovery = 0;
spin_lock_irq(&conf->device_lock);
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index 9459689c4ea..dd708359b45 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -2,6 +2,7 @@
#define _RAID5_H
#include <linux/raid/xor.h>
+#include <linux/dmaengine.h>
/*
*
@@ -175,7 +176,9 @@
*/
enum check_states {
check_state_idle = 0,
- check_state_run, /* parity check */
+ check_state_run, /* xor parity check */
+ check_state_run_q, /* q-parity check */
+ check_state_run_pq, /* pq dual parity check */
check_state_check_result,
check_state_compute_run, /* parity repair */
check_state_compute_result,
@@ -211,12 +214,20 @@ struct stripe_head {
int disks; /* disks in stripe */
enum check_states check_state;
enum reconstruct_states reconstruct_state;
- /* stripe_operations
+ /**
+ * struct stripe_operations
* @target - STRIPE_OP_COMPUTE_BLK target
+ * @target2 - 2nd compute target in the raid6 case
+ * @zero_sum_result - P and Q verification flags
+ * @request - async service request flags for raid_run_ops
*/
struct stripe_operations {
- int target;
- u32 zero_sum_result;
+ int target, target2;
+ enum sum_check_flags zero_sum_result;
+ #ifdef CONFIG_MULTICORE_RAID456
+ unsigned long request;
+ wait_queue_head_t wait_for_ops;
+ #endif
} ops;
struct r5dev {
struct bio req;
@@ -291,6 +302,8 @@ struct r6_state {
#define STRIPE_FULL_WRITE 13 /* all blocks are set to be overwritten */
#define STRIPE_BIOFILL_RUN 14
#define STRIPE_COMPUTE_RUN 15
+#define STRIPE_OPS_REQ_PENDING 16
+
/*
* Operation request flags
*/
@@ -298,7 +311,7 @@ struct r6_state {
#define STRIPE_OP_COMPUTE_BLK 1
#define STRIPE_OP_PREXOR 2
#define STRIPE_OP_BIODRAIN 3
-#define STRIPE_OP_POSTXOR 4
+#define STRIPE_OP_RECONSTRUCT 4
#define STRIPE_OP_CHECK 5
/*
@@ -385,8 +398,21 @@ struct raid5_private_data {
* (fresh device added).
* Cleared when a sync completes.
*/
-
- struct page *spare_page; /* Used when checking P/Q in raid6 */
+ /* per cpu variables */
+ struct raid5_percpu {
+ struct page *spare_page; /* Used when checking P/Q in raid6 */
+ void *scribble; /* space for constructing buffer
+ * lists and performing address
+ * conversions
+ */
+ } *percpu;
+ size_t scribble_len; /* size of scribble region must be
+ * associated with conf to handle
+ * cpu hotplug while reshaping
+ */
+#ifdef CONFIG_HOTPLUG_CPU
+ struct notifier_block cpu_notify;
+#endif
/*
* Free stripes pool
@@ -462,7 +488,7 @@ static inline int algorithm_valid_raid6(int layout)
{
return (layout >= 0 && layout <= 5)
||
- (layout == 8 || layout == 10)
+ (layout >= 8 && layout <= 10)
||
(layout >= 16 && layout <= 20);
}
diff --git a/drivers/md/raid6altivec.uc b/drivers/md/raid6altivec.uc
index 699dfeee494..2654d5c854b 100644
--- a/drivers/md/raid6altivec.uc
+++ b/drivers/md/raid6altivec.uc
@@ -15,7 +15,7 @@
*
* $#-way unrolled portable integer math RAID-6 instruction set
*
- * This file is postprocessed using unroll.pl
+ * This file is postprocessed using unroll.awk
*
* <benh> hpa: in process,
* you can just "steal" the vec unit with enable_kernel_altivec() (but
diff --git a/drivers/md/raid6int.uc b/drivers/md/raid6int.uc
index f9bf9cba357..d1e276a14fa 100644
--- a/drivers/md/raid6int.uc
+++ b/drivers/md/raid6int.uc
@@ -15,7 +15,7 @@
*
* $#-way unrolled portable integer math RAID-6 instruction set
*
- * This file is postprocessed using unroll.pl
+ * This file is postprocessed using unroll.awk
*/
#include <linux/raid/pq.h>
diff --git a/drivers/md/raid6test/Makefile b/drivers/md/raid6test/Makefile
index 58ffdf4f516..2874cbef529 100644
--- a/drivers/md/raid6test/Makefile
+++ b/drivers/md/raid6test/Makefile
@@ -7,7 +7,7 @@ CC = gcc
OPTFLAGS = -O2 # Adjust as desired
CFLAGS = -I.. -I ../../../include -g $(OPTFLAGS)
LD = ld
-PERL = perl
+AWK = awk
AR = ar
RANLIB = ranlib
@@ -35,35 +35,35 @@ raid6.a: raid6int1.o raid6int2.o raid6int4.o raid6int8.o raid6int16.o \
raid6test: test.c raid6.a
$(CC) $(CFLAGS) -o raid6test $^
-raid6altivec1.c: raid6altivec.uc ../unroll.pl
- $(PERL) ../unroll.pl 1 < raid6altivec.uc > $@
+raid6altivec1.c: raid6altivec.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=1 < raid6altivec.uc > $@
-raid6altivec2.c: raid6altivec.uc ../unroll.pl
- $(PERL) ../unroll.pl 2 < raid6altivec.uc > $@
+raid6altivec2.c: raid6altivec.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=2 < raid6altivec.uc > $@
-raid6altivec4.c: raid6altivec.uc ../unroll.pl
- $(PERL) ../unroll.pl 4 < raid6altivec.uc > $@
+raid6altivec4.c: raid6altivec.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=4 < raid6altivec.uc > $@
-raid6altivec8.c: raid6altivec.uc ../unroll.pl
- $(PERL) ../unroll.pl 8 < raid6altivec.uc > $@
+raid6altivec8.c: raid6altivec.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=8 < raid6altivec.uc > $@
-raid6int1.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 1 < raid6int.uc > $@
+raid6int1.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=1 < raid6int.uc > $@
-raid6int2.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 2 < raid6int.uc > $@
+raid6int2.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=2 < raid6int.uc > $@
-raid6int4.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 4 < raid6int.uc > $@
+raid6int4.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=4 < raid6int.uc > $@
-raid6int8.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 8 < raid6int.uc > $@
+raid6int8.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=8 < raid6int.uc > $@
-raid6int16.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 16 < raid6int.uc > $@
+raid6int16.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=16 < raid6int.uc > $@
-raid6int32.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 32 < raid6int.uc > $@
+raid6int32.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=32 < raid6int.uc > $@
raid6tables.c: mktables
./mktables > raid6tables.c
diff --git a/drivers/md/unroll.awk b/drivers/md/unroll.awk
new file mode 100644
index 00000000000..c6aa03631df
--- /dev/null
+++ b/drivers/md/unroll.awk
@@ -0,0 +1,20 @@
+
+# This filter requires one command line option of form -vN=n
+# where n must be a decimal number.
+#
+# Repeat each input line containing $$ n times, replacing $$ with 0...n-1.
+# Replace each $# with n, and each $* with a single $.
+
+BEGIN {
+ n = N + 0
+}
+{
+ if (/\$\$/) { rep = n } else { rep = 1 }
+ for (i = 0; i < rep; ++i) {
+ tmp = $0
+ gsub(/\$\$/, i, tmp)
+ gsub(/\$\#/, n, tmp)
+ gsub(/\$\*/, "$", tmp)
+ print tmp
+ }
+}
diff --git a/drivers/md/unroll.pl b/drivers/md/unroll.pl
deleted file mode 100644
index 3acc710a20e..00000000000
--- a/drivers/md/unroll.pl
+++ /dev/null
@@ -1,24 +0,0 @@
-#!/usr/bin/perl
-#
-# Take a piece of C code and for each line which contains the sequence $$
-# repeat n times with $ replaced by 0...n-1; the sequence $# is replaced
-# by the unrolling factor, and $* with a single $
-#
-
-($n) = @ARGV;
-$n += 0;
-
-while ( defined($line = <STDIN>) ) {
- if ( $line =~ /\$\$/ ) {
- $rep = $n;
- } else {
- $rep = 1;
- }
- for ( $i = 0 ; $i < $rep ; $i++ ) {
- $tmp = $line;
- $tmp =~ s/\$\$/$i/g;
- $tmp =~ s/\$\#/$n/g;
- $tmp =~ s/\$\*/\$/g;
- print $tmp;
- }
-}
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-01 07:10:50 +0000