summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2017-09-07 12:41:48 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2017-09-07 12:41:48 -0700
commit3645e6d0dc80be4376f87acc9ee527768387c909 (patch)
tree78cc68cb09c9d24a6bcfaf842a3bca671ed53ee0
parent15d8ffc96464f6571ecf22043c45fad659f11bdd (diff)
parente8a27f836f165c26f867ece7f31eb5c811692319 (diff)
Merge tag 'md/4.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md
Pull MD updates from Shaohua Li: "This update mainly fixes bugs: - Make raid5 ppl support several ppl from Pawel - Several raid5-cache bug fixes from Song - Bitmap fixes from Neil and Me - One raid1/10 regression fix since 4.12 from Me - Other small fixes and cleanup" * tag 'md/4.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/shli/md: md/bitmap: disable bitmap_resize for file-backed bitmaps. raid5-ppl: Recovery support for multiple partial parity logs md: Runtime support for multiple ppls md/raid0: attach correct cgroup info in bio lib/raid6: align AVX512 constants to 512 bits, not bytes raid5: remove raid5_build_block md/r5cache: call mddev_lock/unlock() in r5c_journal_mode_show md: replace seq_release_private with seq_release md: notify about new spare disk in the container md/raid1/10: reset bio allocated from mempool md/raid5: release/flush io in raid5_do_work() md/bitmap: copy correct data for bitmap super
-rw-r--r--block/bio.c2
-rw-r--r--drivers/md/bitmap.c9
-rw-r--r--drivers/md/md.c20
-rw-r--r--drivers/md/md.h1
-rw-r--r--drivers/md/raid0.c4
-rw-r--r--drivers/md/raid1.c22
-rw-r--r--drivers/md/raid10.c35
-rw-r--r--drivers/md/raid5-cache.c12
-rw-r--r--drivers/md/raid5-ppl.c171
-rw-r--r--drivers/md/raid5.c16
-rw-r--r--include/uapi/linux/raid/md_p.h4
-rw-r--r--lib/raid6/avx512.c2
12 files changed, 230 insertions, 68 deletions
diff --git a/block/bio.c b/block/bio.c
index 6745759028da..b38e962fa83e 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -2089,7 +2089,7 @@ void bio_clone_blkcg_association(struct bio *dst, struct bio *src)
if (src->bi_css)
WARN_ON(bio_associate_blkcg(dst, src->bi_css));
}
-
+EXPORT_SYMBOL_GPL(bio_clone_blkcg_association);
#endif /* CONFIG_BLK_CGROUP */
static void __init biovec_init_slabs(void)
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 40f3cd7eab0f..d2121637b4ab 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -625,7 +625,7 @@ re_read:
err = read_sb_page(bitmap->mddev,
offset,
sb_page,
- 0, sizeof(bitmap_super_t));
+ 0, PAGE_SIZE);
}
if (err)
return err;
@@ -2058,6 +2058,11 @@ int bitmap_resize(struct bitmap *bitmap, sector_t blocks,
long pages;
struct bitmap_page *new_bp;
+ if (bitmap->storage.file && !init) {
+ pr_info("md: cannot resize file-based bitmap\n");
+ return -EINVAL;
+ }
+
if (chunksize == 0) {
/* If there is enough space, leave the chunk size unchanged,
* else increase by factor of two until there is enough space.
@@ -2118,7 +2123,7 @@ int bitmap_resize(struct bitmap *bitmap, sector_t blocks,
if (store.sb_page && bitmap->storage.sb_page)
memcpy(page_address(store.sb_page),
page_address(bitmap->storage.sb_page),
- sizeof(bitmap_super_t));
+ PAGE_SIZE);
bitmap_file_unmap(&bitmap->storage);
bitmap->storage = store;
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 078c6f3b7e53..08fcaebc61bd 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1538,7 +1538,8 @@ static int super_1_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor_
} else if (sb->bblog_offset != 0)
rdev->badblocks.shift = 0;
- if (le32_to_cpu(sb->feature_map) & MD_FEATURE_PPL) {
+ if ((le32_to_cpu(sb->feature_map) &
+ (MD_FEATURE_PPL | MD_FEATURE_MULTIPLE_PPLS))) {
rdev->ppl.offset = (__s16)le16_to_cpu(sb->ppl.offset);
rdev->ppl.size = le16_to_cpu(sb->ppl.size);
rdev->ppl.sector = rdev->sb_start + rdev->ppl.offset;
@@ -1657,10 +1658,15 @@ static int super_1_validate(struct mddev *mddev, struct md_rdev *rdev)
if (le32_to_cpu(sb->feature_map) & MD_FEATURE_JOURNAL)
set_bit(MD_HAS_JOURNAL, &mddev->flags);
- if (le32_to_cpu(sb->feature_map) & MD_FEATURE_PPL) {
+ if (le32_to_cpu(sb->feature_map) &
+ (MD_FEATURE_PPL | MD_FEATURE_MULTIPLE_PPLS)) {
if (le32_to_cpu(sb->feature_map) &
(MD_FEATURE_BITMAP_OFFSET | MD_FEATURE_JOURNAL))
return -EINVAL;
+ if ((le32_to_cpu(sb->feature_map) & MD_FEATURE_PPL) &&
+ (le32_to_cpu(sb->feature_map) &
+ MD_FEATURE_MULTIPLE_PPLS))
+ return -EINVAL;
set_bit(MD_HAS_PPL, &mddev->flags);
}
} else if (mddev->pers == NULL) {
@@ -1877,7 +1883,11 @@ retry:
sb->feature_map |= cpu_to_le32(MD_FEATURE_JOURNAL);
if (test_bit(MD_HAS_PPL, &mddev->flags)) {
- sb->feature_map |= cpu_to_le32(MD_FEATURE_PPL);
+ if (test_bit(MD_HAS_MULTIPLE_PPLS, &mddev->flags))
+ sb->feature_map |=
+ cpu_to_le32(MD_FEATURE_MULTIPLE_PPLS);
+ else
+ sb->feature_map |= cpu_to_le32(MD_FEATURE_PPL);
sb->ppl.offset = cpu_to_le16(rdev->ppl.offset);
sb->ppl.size = cpu_to_le16(rdev->ppl.size);
}
@@ -4285,6 +4295,8 @@ new_dev_store(struct mddev *mddev, const char *buf, size_t len)
if (err)
export_rdev(rdev);
mddev_unlock(mddev);
+ if (!err)
+ md_new_event(mddev);
return err ? err : len;
}
@@ -7838,7 +7850,7 @@ static const struct file_operations md_seq_fops = {
.open = md_seq_open,
.read = seq_read,
.llseek = seq_lseek,
- .release = seq_release_private,
+ .release = seq_release,
.poll = mdstat_poll,
};
diff --git a/drivers/md/md.h b/drivers/md/md.h
index c0d436fb88f0..561d22b9a9a8 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -236,6 +236,7 @@ enum mddev_flags {
* never cause the array to become failed.
*/
MD_HAS_PPL, /* The raid array has PPL feature set */
+ MD_HAS_MULTIPLE_PPLS, /* The raid array has multiple PPLs feature set */
};
enum mddev_sb_flags {
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index 05a4521b832f..5a00fc118470 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -30,7 +30,8 @@
((1L << MD_HAS_JOURNAL) | \
(1L << MD_JOURNAL_CLEAN) | \
(1L << MD_FAILFAST_SUPPORTED) |\
- (1L << MD_HAS_PPL))
+ (1L << MD_HAS_PPL) | \
+ (1L << MD_HAS_MULTIPLE_PPLS))
static int raid0_congested(struct mddev *mddev, int bits)
{
@@ -539,6 +540,7 @@ static void raid0_handle_discard(struct mddev *mddev, struct bio *bio)
!discard_bio)
continue;
bio_chain(discard_bio, bio);
+ bio_clone_blkcg_association(discard_bio, bio);
if (mddev->gendisk)
trace_block_bio_remap(bdev_get_queue(rdev->bdev),
discard_bio, disk_devt(mddev->gendisk),
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index baf5e358d22a..f3f3e40dc9d8 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -48,7 +48,8 @@
#define UNSUPPORTED_MDDEV_FLAGS \
((1L << MD_HAS_JOURNAL) | \
(1L << MD_JOURNAL_CLEAN) | \
- (1L << MD_HAS_PPL))
+ (1L << MD_HAS_PPL) | \
+ (1L << MD_HAS_MULTIPLE_PPLS))
/*
* Number of guaranteed r1bios in case of extreme VM load:
@@ -2560,6 +2561,23 @@ static int init_resync(struct r1conf *conf)
return 0;
}
+static struct r1bio *raid1_alloc_init_r1buf(struct r1conf *conf)
+{
+ struct r1bio *r1bio = mempool_alloc(conf->r1buf_pool, GFP_NOIO);
+ struct resync_pages *rps;
+ struct bio *bio;
+ int i;
+
+ for (i = conf->poolinfo->raid_disks; i--; ) {
+ bio = r1bio->bios[i];
+ rps = bio->bi_private;
+ bio_reset(bio);
+ bio->bi_private = rps;
+ }
+ r1bio->master_bio = NULL;
+ return r1bio;
+}
+
/*
* perform a "sync" on one "block"
*
@@ -2645,7 +2663,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr,
bitmap_cond_end_sync(mddev->bitmap, sector_nr,
mddev_is_clustered(mddev) && (sector_nr + 2 * RESYNC_SECTORS > conf->cluster_sync_high));
- r1_bio = mempool_alloc(conf->r1buf_pool, GFP_NOIO);
+ r1_bio = raid1_alloc_init_r1buf(conf);
raise_barrier(conf, sector_nr);
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index d1f948e371e0..374df5796649 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -2796,6 +2796,35 @@ static int init_resync(struct r10conf *conf)
return 0;
}
+static struct r10bio *raid10_alloc_init_r10buf(struct r10conf *conf)
+{
+ struct r10bio *r10bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
+ struct rsync_pages *rp;
+ struct bio *bio;
+ int nalloc;
+ int i;
+
+ if (test_bit(MD_RECOVERY_SYNC, &conf->mddev->recovery) ||
+ test_bit(MD_RECOVERY_RESHAPE, &conf->mddev->recovery))
+ nalloc = conf->copies; /* resync */
+ else
+ nalloc = 2; /* recovery */
+
+ for (i = 0; i < nalloc; i++) {
+ bio = r10bio->devs[i].bio;
+ rp = bio->bi_private;
+ bio_reset(bio);
+ bio->bi_private = rp;
+ bio = r10bio->devs[i].repl_bio;
+ if (bio) {
+ rp = bio->bi_private;
+ bio_reset(bio);
+ bio->bi_private = rp;
+ }
+ }
+ return r10bio;
+}
+
/*
* perform a "sync" on one "block"
*
@@ -3025,7 +3054,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
atomic_inc(&mreplace->nr_pending);
rcu_read_unlock();
- r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
+ r10_bio = raid10_alloc_init_r10buf(conf);
r10_bio->state = 0;
raise_barrier(conf, rb2 != NULL);
atomic_set(&r10_bio->remaining, 0);
@@ -3234,7 +3263,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
}
if (sync_blocks < max_sync)
max_sync = sync_blocks;
- r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
+ r10_bio = raid10_alloc_init_r10buf(conf);
r10_bio->state = 0;
r10_bio->mddev = mddev;
@@ -4358,7 +4387,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr,
read_more:
/* Now schedule reads for blocks from sector_nr to last */
- r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
+ r10_bio = raid10_alloc_init_r10buf(conf);
r10_bio->state = 0;
raise_barrier(conf, sectors_done != 0);
atomic_set(&r10_bio->remaining, 0);
diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c
index 5d7da1c700aa..0b7406ac8ce1 100644
--- a/drivers/md/raid5-cache.c
+++ b/drivers/md/raid5-cache.c
@@ -2529,11 +2529,18 @@ static void r5l_write_super(struct r5l_log *log, sector_t cp)
static ssize_t r5c_journal_mode_show(struct mddev *mddev, char *page)
{
- struct r5conf *conf = mddev->private;
+ struct r5conf *conf;
int ret;
- if (!conf->log)
+ ret = mddev_lock(mddev);
+ if (ret)
+ return ret;
+
+ conf = mddev->private;
+ if (!conf || !conf->log) {
+ mddev_unlock(mddev);
return 0;
+ }
switch (conf->log->r5c_journal_mode) {
case R5C_JOURNAL_MODE_WRITE_THROUGH:
@@ -2551,6 +2558,7 @@ static ssize_t r5c_journal_mode_show(struct mddev *mddev, char *page)
default:
ret = 0;
}
+ mddev_unlock(mddev);
return ret;
}
diff --git a/drivers/md/raid5-ppl.c b/drivers/md/raid5-ppl.c
index 1e237c40d6fa..cd026c88f7ef 100644
--- a/drivers/md/raid5-ppl.c
+++ b/drivers/md/raid5-ppl.c
@@ -87,6 +87,8 @@
* The current io_unit accepting new stripes is always at the end of the list.
*/
+#define PPL_SPACE_SIZE (128 * 1024)
+
struct ppl_conf {
struct mddev *mddev;
@@ -122,6 +124,10 @@ struct ppl_log {
* always at the end of io_list */
spinlock_t io_list_lock;
struct list_head io_list; /* all io_units of this log */
+
+ sector_t next_io_sector;
+ unsigned int entry_space;
+ bool use_multippl;
};
#define PPL_IO_INLINE_BVECS 32
@@ -264,13 +270,12 @@ static int ppl_log_stripe(struct ppl_log *log, struct stripe_head *sh)
int i;
sector_t data_sector = 0;
int data_disks = 0;
- unsigned int entry_space = (log->rdev->ppl.size << 9) - PPL_HEADER_SIZE;
struct r5conf *conf = sh->raid_conf;
pr_debug("%s: stripe: %llu\n", __func__, (unsigned long long)sh->sector);
/* check if current io_unit is full */
- if (io && (io->pp_size == entry_space ||
+ if (io && (io->pp_size == log->entry_space ||
io->entries_count == PPL_HDR_MAX_ENTRIES)) {
pr_debug("%s: add io_unit blocked by seq: %llu\n",
__func__, io->seq);
@@ -451,12 +456,25 @@ static void ppl_submit_iounit(struct ppl_io_unit *io)
pplhdr->entries_count = cpu_to_le32(io->entries_count);
pplhdr->checksum = cpu_to_le32(~crc32c_le(~0, pplhdr, PPL_HEADER_SIZE));
+ /* Rewind the buffer if current PPL is larger then remaining space */
+ if (log->use_multippl &&
+ log->rdev->ppl.sector + log->rdev->ppl.size - log->next_io_sector <
+ (PPL_HEADER_SIZE + io->pp_size) >> 9)
+ log->next_io_sector = log->rdev->ppl.sector;
+
+
bio->bi_end_io = ppl_log_endio;
bio->bi_opf = REQ_OP_WRITE | REQ_FUA;
bio_set_dev(bio, log->rdev->bdev);
- bio->bi_iter.bi_sector = log->rdev->ppl.sector;
+ bio->bi_iter.bi_sector = log->next_io_sector;
bio_add_page(bio, io->header_page, PAGE_SIZE, 0);
+ pr_debug("%s: log->current_io_sector: %llu\n", __func__,
+ (unsigned long long)log->next_io_sector);
+
+ if (log->use_multippl)
+ log->next_io_sector += (PPL_HEADER_SIZE + io->pp_size) >> 9;
+
list_for_each_entry(sh, &io->stripe_list, log_list) {
/* entries for full stripe writes have no partial parity */
if (test_bit(STRIPE_FULL_WRITE, &sh->state))
@@ -813,12 +831,14 @@ out:
return ret;
}
-static int ppl_recover(struct ppl_log *log, struct ppl_header *pplhdr)
+static int ppl_recover(struct ppl_log *log, struct ppl_header *pplhdr,
+ sector_t offset)
{
struct ppl_conf *ppl_conf = log->ppl_conf;
struct md_rdev *rdev = log->rdev;
struct mddev *mddev = rdev->mddev;
- sector_t ppl_sector = rdev->ppl.sector + (PPL_HEADER_SIZE >> 9);
+ sector_t ppl_sector = rdev->ppl.sector + offset +
+ (PPL_HEADER_SIZE >> 9);
struct page *page;
int i;
int ret = 0;
@@ -902,6 +922,9 @@ static int ppl_write_empty_header(struct ppl_log *log)
return -ENOMEM;
pplhdr = page_address(page);
+ /* zero out PPL space to avoid collision with old PPLs */
+ blkdev_issue_zeroout(rdev->bdev, rdev->ppl.sector,
+ log->rdev->ppl.size, GFP_NOIO, 0);
memset(pplhdr->reserved, 0xff, PPL_HDR_RESERVED);
pplhdr->signature = cpu_to_le32(log->ppl_conf->signature);
pplhdr->checksum = cpu_to_le32(~crc32c_le(~0, pplhdr, PAGE_SIZE));
@@ -922,63 +945,110 @@ static int ppl_load_distributed(struct ppl_log *log)
struct ppl_conf *ppl_conf = log->ppl_conf;
struct md_rdev *rdev = log->rdev;
struct mddev *mddev = rdev->mddev;
- struct page *page;
- struct ppl_header *pplhdr;
+ struct page *page, *page2, *tmp;
+ struct ppl_header *pplhdr = NULL, *prev_pplhdr = NULL;
u32 crc, crc_stored;
u32 signature;
- int ret = 0;
+ int ret = 0, i;
+ sector_t pplhdr_offset = 0, prev_pplhdr_offset = 0;
pr_debug("%s: disk: %d\n", __func__, rdev->raid_disk);
-
- /* read PPL header */
+ /* read PPL headers, find the recent one */
page = alloc_page(GFP_KERNEL);
if (!page)
return -ENOMEM;
- if (!sync_page_io(rdev, rdev->ppl.sector - rdev->data_offset,
- PAGE_SIZE, page, REQ_OP_READ, 0, false)) {
- md_error(mddev, rdev);
- ret = -EIO;
- goto out;
+ page2 = alloc_page(GFP_KERNEL);
+ if (!page2) {
+ __free_page(page);
+ return -ENOMEM;
}
- pplhdr = page_address(page);
- /* check header validity */
- crc_stored = le32_to_cpu(pplhdr->checksum);
- pplhdr->checksum = 0;
- crc = ~crc32c_le(~0, pplhdr, PAGE_SIZE);
+ /* searching ppl area for latest ppl */
+ while (pplhdr_offset < rdev->ppl.size - (PPL_HEADER_SIZE >> 9)) {
+ if (!sync_page_io(rdev,
+ rdev->ppl.sector - rdev->data_offset +
+ pplhdr_offset, PAGE_SIZE, page, REQ_OP_READ,
+ 0, false)) {
+ md_error(mddev, rdev);
+ ret = -EIO;
+ /* if not able to read - don't recover any PPL */
+ pplhdr = NULL;
+ break;
+ }
+ pplhdr = page_address(page);
+
+ /* check header validity */
+ crc_stored = le32_to_cpu(pplhdr->checksum);
+ pplhdr->checksum = 0;
+ crc = ~crc32c_le(~0, pplhdr, PAGE_SIZE);
+
+ if (crc_stored != crc) {
+ pr_debug("%s: ppl header crc does not match: stored: 0x%x calculated: 0x%x (offset: %llu)\n",
+ __func__, crc_stored, crc,
+ (unsigned long long)pplhdr_offset);
+ pplhdr = prev_pplhdr;
+ pplhdr_offset = prev_pplhdr_offset;
+ break;
+ }
- if (crc_stored != crc) {
- pr_debug("%s: ppl header crc does not match: stored: 0x%x calculated: 0x%x\n",
- __func__, crc_stored, crc);
- ppl_conf->mismatch_count++;
- goto out;
- }
+ signature = le32_to_cpu(pplhdr->signature);
- signature = le32_to_cpu(pplhdr->signature);
+ if (mddev->external) {
+ /*
+ * For external metadata the header signature is set and
+ * validated in userspace.
+ */
+ ppl_conf->signature = signature;
+ } else if (ppl_conf->signature != signature) {
+ pr_debug("%s: ppl header signature does not match: stored: 0x%x configured: 0x%x (offset: %llu)\n",
+ __func__, signature, ppl_conf->signature,
+ (unsigned long long)pplhdr_offset);
+ pplhdr = prev_pplhdr;
+ pplhdr_offset = prev_pplhdr_offset;
+ break;
+ }
- if (mddev->external) {
- /*
- * For external metadata the header signature is set and
- * validated in userspace.
- */
- ppl_conf->signature = signature;
- } else if (ppl_conf->signature != signature) {
- pr_debug("%s: ppl header signature does not match: stored: 0x%x configured: 0x%x\n",
- __func__, signature, ppl_conf->signature);
- ppl_conf->mismatch_count++;
- goto out;
+ if (prev_pplhdr && le64_to_cpu(prev_pplhdr->generation) >
+ le64_to_cpu(pplhdr->generation)) {
+ /* previous was newest */
+ pplhdr = prev_pplhdr;
+ pplhdr_offset = prev_pplhdr_offset;
+ break;
+ }
+
+ prev_pplhdr_offset = pplhdr_offset;
+ prev_pplhdr = pplhdr;
+
+ tmp = page;
+ page = page2;
+ page2 = tmp;
+
+ /* calculate next potential ppl offset */
+ for (i = 0; i < le32_to_cpu(pplhdr->entries_count); i++)
+ pplhdr_offset +=
+ le32_to_cpu(pplhdr->entries[i].pp_size) >> 9;
+ pplhdr_offset += PPL_HEADER_SIZE >> 9;
}
+ /* no valid ppl found */
+ if (!pplhdr)
+ ppl_conf->mismatch_count++;
+ else
+ pr_debug("%s: latest PPL found at offset: %llu, with generation: %llu\n",
+ __func__, (unsigned long long)pplhdr_offset,
+ le64_to_cpu(pplhdr->generation));
+
/* attempt to recover from log if we are starting a dirty array */
- if (!mddev->pers && mddev->recovery_cp != MaxSector)
- ret = ppl_recover(log, pplhdr);
-out:
+ if (pplhdr && !mddev->pers && mddev->recovery_cp != MaxSector)
+ ret = ppl_recover(log, pplhdr, pplhdr_offset);
+
/* write empty header if we are starting the array */
if (!ret && !mddev->pers)
ret = ppl_write_empty_header(log);
__free_page(page);
+ __free_page(page2);
pr_debug("%s: return: %d mismatch_count: %d recovered_entries: %d\n",
__func__, ret, ppl_conf->mismatch_count,
@@ -1031,6 +1101,7 @@ static int ppl_load(struct ppl_conf *ppl_conf)
static void __ppl_exit_log(struct ppl_conf *ppl_conf)
{
clear_bit(MD_HAS_PPL, &ppl_conf->mddev->flags);
+ clear_bit(MD_HAS_MULTIPLE_PPLS, &ppl_conf->mddev->flags);
kfree(ppl_conf->child_logs);
@@ -1099,6 +1170,22 @@ static int ppl_validate_rdev(struct md_rdev *rdev)
return 0;
}
+static void ppl_init_child_log(struct ppl_log *log, struct md_rdev *rdev)
+{
+ if ((rdev->ppl.size << 9) >= (PPL_SPACE_SIZE +
+ PPL_HEADER_SIZE) * 2) {
+ log->use_multippl = true;
+ set_bit(MD_HAS_MULTIPLE_PPLS,
+ &log->ppl_conf->mddev->flags);
+ log->entry_space = PPL_SPACE_SIZE;
+ } else {
+ log->use_multippl = false;
+ log->entry_space = (log->rdev->ppl.size << 9) -
+ PPL_HEADER_SIZE;
+ }
+ log->next_io_sector = rdev->ppl.sector;
+}
+
int ppl_init_log(struct r5conf *conf)
{
struct ppl_conf *ppl_conf;
@@ -1196,6 +1283,7 @@ int ppl_init_log(struct r5conf *conf)
q = bdev_get_queue(rdev->bdev);
if (test_bit(QUEUE_FLAG_WC, &q->queue_flags))
need_cache_flush = true;
+ ppl_init_child_log(log, rdev);
}
}
@@ -1261,6 +1349,7 @@ int ppl_modify_log(struct r5conf *conf, struct md_rdev *rdev, bool add)
if (!ret) {
log->rdev = rdev;
ret = ppl_write_empty_header(log);
+ ppl_init_child_log(log, rdev);
}
} else {
log->rdev = NULL;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 3ae8bbceb6c4..4188a4881148 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -494,7 +494,6 @@ static int grow_buffers(struct stripe_head *sh, gfp_t gfp)
return 0;
}
-static void raid5_build_block(struct stripe_head *sh, int i, int previous);
static void stripe_set_idx(sector_t stripe, struct r5conf *conf, int previous,
struct stripe_head *sh);
@@ -530,7 +529,7 @@ retry:
WARN_ON(1);
}
dev->flags = 0;
- raid5_build_block(sh, i, previous);
+ dev->sector = raid5_compute_blocknr(sh, i, previous);
}
if (read_seqcount_retry(&conf->gen_lock, seq))
goto retry;
@@ -2662,14 +2661,6 @@ static void raid5_end_write_request(struct bio *bi)
raid5_release_stripe(sh->batch_head);
}
-static void raid5_build_block(struct stripe_head *sh, int i, int previous)
-{
- struct r5dev *dev = &sh->dev[i];
-
- dev->flags = 0;
- dev->sector = raid5_compute_blocknr(sh, i, previous);
-}
-
static void raid5_error(struct mddev *mddev, struct md_rdev *rdev)
{
char b[BDEVNAME_SIZE];
@@ -6237,6 +6228,10 @@ static void raid5_do_work(struct work_struct *work)
spin_unlock_irq(&conf->device_lock);
+ flush_deferred_bios(conf);
+
+ r5l_flush_stripe_to_raid(conf->log);
+
async_tx_issue_pending_all();
blk_finish_plug(&plug);
@@ -7243,6 +7238,7 @@ static int raid5_run(struct mddev *mddev)
pr_warn("md/raid:%s: using journal device and PPL not allowed - disabling PPL\n",
mdname(mddev));
clear_bit(MD_HAS_PPL, &mddev->flags);
+ clear_bit(MD_HAS_MULTIPLE_PPLS, &mddev->flags);
}
if (mddev->private == NULL)
diff --git a/include/uapi/linux/raid/md_p.h b/include/uapi/linux/raid/md_p.h
index d500bd224979..b9197976b660 100644
--- a/include/uapi/linux/raid/md_p.h
+++ b/include/uapi/linux/raid/md_p.h
@@ -324,9 +324,10 @@ struct mdp_superblock_1 {
#define MD_FEATURE_RECOVERY_BITMAP 128 /* recovery that is happening
* is guided by bitmap.
*/
-#define MD_FEATURE_CLUSTERED 256 /* clustered MD */
+#define MD_FEATURE_CLUSTERED 256 /* clustered MD */
#define MD_FEATURE_JOURNAL 512 /* support write cache */
#define MD_FEATURE_PPL 1024 /* support PPL */
+#define MD_FEATURE_MULTIPLE_PPLS 2048 /* support for multiple PPLs */
#define MD_FEATURE_ALL (MD_FEATURE_BITMAP_OFFSET \
|MD_FEATURE_RECOVERY_OFFSET \
|MD_FEATURE_RESHAPE_ACTIVE \
@@ -338,6 +339,7 @@ struct mdp_superblock_1 {
|MD_FEATURE_CLUSTERED \
|MD_FEATURE_JOURNAL \
|MD_FEATURE_PPL \
+ |MD_FEATURE_MULTIPLE_PPLS \
)
struct r5l_payload_header {
diff --git a/lib/raid6/avx512.c b/lib/raid6/avx512.c
index f524a7972006..46df7977b971 100644
--- a/lib/raid6/avx512.c
+++ b/lib/raid6/avx512.c
@@ -29,7 +29,7 @@
static const struct raid6_avx512_constants {
u64 x1d[8];
-} raid6_avx512_constants __aligned(512) = {
+} raid6_avx512_constants __aligned(512/8) = {
{ 0x1d1d1d1d1d1d1d1dULL, 0x1d1d1d1d1d1d1d1dULL,
0x1d1d1d1d1d1d1d1dULL, 0x1d1d1d1d1d1d1d1dULL,
0x1d1d1d1d1d1d1d1dULL, 0x1d1d1d1d1d1d1d1dULL,