From 2dffdc0724004f38f5e39907747b53e4b0d80e59 Mon Sep 17 00:00:00 2001 From: Guoqing Jiang Date: Tue, 16 May 2017 14:01:25 +0800 Subject: md-cluster: fix potential lock issue in add_new_disk The add_new_disk returns with communication locked if __sendmsg returns failure, fix it with call unlock_comm before return. Reported-by: Dan Carpenter CC: Goldwyn Rodrigues Signed-off-by: Guoqing Jiang Signed-off-by: Shaohua Li --- drivers/md/md-cluster.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/md') diff --git a/drivers/md/md-cluster.c b/drivers/md/md-cluster.c index 7299ce2f08a8..03082e17c65c 100644 --- a/drivers/md/md-cluster.c +++ b/drivers/md/md-cluster.c @@ -1311,8 +1311,10 @@ static int add_new_disk(struct mddev *mddev, struct md_rdev *rdev) cmsg.raid_slot = cpu_to_le32(rdev->desc_nr); lock_comm(cinfo, 1); ret = __sendmsg(cinfo, &cmsg); - if (ret) + if (ret) { + unlock_comm(cinfo); return ret; + } cinfo->no_new_dev_lockres->flags |= DLM_LKF_NOQUEUE; ret = dlm_lock_sync(cinfo->no_new_dev_lockres, DLM_LOCK_EX); cinfo->no_new_dev_lockres->flags &= ~DLM_LKF_NOQUEUE; -- cgit v1.2.3 From f52236e0b0a0820e938e16a776309e76b7bd6c43 Mon Sep 17 00:00:00 2001 From: Gilad Ben-Yossef Date: Thu, 18 May 2017 13:47:25 +0300 Subject: dm verity: fix no salt use case DM-Verity has an (undocumented) mode where no salt is used. This was never handled directly by the DM-Verity code, instead working due to the fact that calling crypto_shash_update() with a zero length data is an implicit noop. This is no longer the case now that we have switched to crypto_ahash_update(). Fix the issue by introducing explicit handling of the no salt use case to DM-Verity. Signed-off-by: Gilad Ben-Yossef Reported-by: Marian Csontos Fixes: d1ac3ff ("dm verity: switch to using asynchronous hash crypto API") Tested-by: Milan Broz Signed-off-by: Mike Snitzer --- drivers/md/dm-verity-target.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/md') diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c index 97de961a3bfc..1ec9b2c51c07 100644 --- a/drivers/md/dm-verity-target.c +++ b/drivers/md/dm-verity-target.c @@ -166,7 +166,7 @@ static int verity_hash_init(struct dm_verity *v, struct ahash_request *req, return r; } - if (likely(v->version >= 1)) + if (likely(v->salt_size && (v->version >= 1))) r = verity_hash_update(v, req, v->salt, v->salt_size, res); return r; @@ -177,7 +177,7 @@ static int verity_hash_final(struct dm_verity *v, struct ahash_request *req, { int r; - if (unlikely(!v->version)) { + if (unlikely(v->salt_size && (!v->version))) { r = verity_hash_update(v, req, v->salt, v->salt_size, res); if (r < 0) { -- cgit v1.2.3 From 702a6204f804bad946c455e7cd8d50d79c9d1629 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Sat, 20 May 2017 14:56:21 -0400 Subject: dm integrity: use kvmalloc() instead of dm_integrity_kvmalloc() Signed-off-by: Mikulas Patocka Signed-off-by: Mike Snitzer --- drivers/md/dm-integrity.c | 27 ++++++--------------------- 1 file changed, 6 insertions(+), 21 deletions(-) (limited to 'drivers/md') diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c index c7f7c8d76576..1feeb2ccf5a1 100644 --- a/drivers/md/dm-integrity.c +++ b/drivers/md/dm-integrity.c @@ -2374,21 +2374,6 @@ static void dm_integrity_set(struct dm_target *ti, struct dm_integrity_c *ic) blk_queue_max_integrity_segments(disk->queue, UINT_MAX); } -/* FIXME: use new kvmalloc */ -static void *dm_integrity_kvmalloc(size_t size, gfp_t gfp) -{ - void *ptr = NULL; - - if (size <= PAGE_SIZE) - ptr = kmalloc(size, GFP_KERNEL | gfp); - if (!ptr && size <= KMALLOC_MAX_SIZE) - ptr = kmalloc(size, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY | gfp); - if (!ptr) - ptr = __vmalloc(size, GFP_KERNEL | gfp, PAGE_KERNEL); - - return ptr; -} - static void dm_integrity_free_page_list(struct dm_integrity_c *ic, struct page_list *pl) { unsigned i; @@ -2407,7 +2392,7 @@ static struct page_list *dm_integrity_alloc_page_list(struct dm_integrity_c *ic) struct page_list *pl; unsigned i; - pl = dm_integrity_kvmalloc(page_list_desc_size, __GFP_ZERO); + pl = kvmalloc(page_list_desc_size, GFP_KERNEL | __GFP_ZERO); if (!pl) return NULL; @@ -2437,7 +2422,7 @@ static struct scatterlist **dm_integrity_alloc_journal_scatterlist(struct dm_int struct scatterlist **sl; unsigned i; - sl = dm_integrity_kvmalloc(ic->journal_sections * sizeof(struct scatterlist *), __GFP_ZERO); + sl = kvmalloc(ic->journal_sections * sizeof(struct scatterlist *), GFP_KERNEL | __GFP_ZERO); if (!sl) return NULL; @@ -2453,7 +2438,7 @@ static struct scatterlist **dm_integrity_alloc_journal_scatterlist(struct dm_int n_pages = (end_index - start_index + 1); - s = dm_integrity_kvmalloc(n_pages * sizeof(struct scatterlist), 0); + s = kvmalloc(n_pages * sizeof(struct scatterlist), GFP_KERNEL); if (!s) { dm_integrity_free_journal_scatterlist(ic, sl); return NULL; @@ -2617,7 +2602,7 @@ static int create_journal(struct dm_integrity_c *ic, char **error) goto bad; } - sg = dm_integrity_kvmalloc((ic->journal_pages + 1) * sizeof(struct scatterlist), 0); + sg = kvmalloc((ic->journal_pages + 1) * sizeof(struct scatterlist), GFP_KERNEL); if (!sg) { *error = "Unable to allocate sg list"; r = -ENOMEM; @@ -2673,7 +2658,7 @@ static int create_journal(struct dm_integrity_c *ic, char **error) r = -ENOMEM; goto bad; } - ic->sk_requests = dm_integrity_kvmalloc(ic->journal_sections * sizeof(struct skcipher_request *), __GFP_ZERO); + ic->sk_requests = kvmalloc(ic->journal_sections * sizeof(struct skcipher_request *), GFP_KERNEL | __GFP_ZERO); if (!ic->sk_requests) { *error = "Unable to allocate sk requests"; r = -ENOMEM; @@ -2740,7 +2725,7 @@ retest_commit_id: r = -ENOMEM; goto bad; } - ic->journal_tree = dm_integrity_kvmalloc(journal_tree_size, 0); + ic->journal_tree = kvmalloc(journal_tree_size, GFP_KERNEL); if (!ic->journal_tree) { *error = "Could not allocate memory for journal tree"; r = -ENOMEM; -- cgit v1.2.3 From 8c1e2162f27b319da913683143c0c6c09b083ebb Mon Sep 17 00:00:00 2001 From: Junaid Shahid Date: Thu, 18 May 2017 12:00:51 -0700 Subject: dm ioctl: restore __GFP_HIGH in copy_params() Commit d224e9381897 ("drivers/md/dm-ioctl.c: use kvmalloc rather than opencoded variant") left out the __GFP_HIGH flag when converting from __vmalloc to kvmalloc. This can cause the DM ioctl to fail in some low memory situations where it wouldn't have failed earlier. Add __GFP_HIGH back to avoid any potential regression. Fixes: d224e9381897 ("drivers/md/dm-ioctl.c: use kvmalloc rather than opencoded variant") Signed-off-by: Junaid Shahid Signed-off-by: Mikulas Patocka Acked-by: David Rientjes Signed-off-by: Mike Snitzer --- drivers/md/dm-ioctl.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'drivers/md') diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 0555b4410e05..41852ae287a5 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1710,12 +1710,13 @@ static int copy_params(struct dm_ioctl __user *user, struct dm_ioctl *param_kern } /* - * Try to avoid low memory issues when a device is suspended. + * Use __GFP_HIGH to avoid low memory issues when a device is + * suspended and the ioctl is needed to resume it. * Use kmalloc() rather than vmalloc() when we can. */ dmi = NULL; noio_flag = memalloc_noio_save(); - dmi = kvmalloc(param_kernel->data_size, GFP_KERNEL); + dmi = kvmalloc(param_kernel->data_size, GFP_KERNEL | __GFP_HIGH); memalloc_noio_restore(noio_flag); if (!dmi) { -- cgit v1.2.3 From 4179bc30b2fe85f827d93e0ae7ae8f49ad3afc02 Mon Sep 17 00:00:00 2001 From: Kyungchan Koh Date: Wed, 24 May 2017 10:16:27 -0700 Subject: md: uuid debug statement now in processor byte order. Previously, the uuid debug statements were printed in little-endian format, which wasn't consistent in machines that might not be in little-endian byte order. With this change, the output will be consistent for all machines with different byte-ordering. Signed-off-by: Kyungchan Koh Signed-off-by: Shaohua Li --- drivers/md/bitmap.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'drivers/md') diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index bf7419a56454..f4eace5ea184 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -485,10 +485,10 @@ void bitmap_print_sb(struct bitmap *bitmap) pr_debug(" magic: %08x\n", le32_to_cpu(sb->magic)); pr_debug(" version: %d\n", le32_to_cpu(sb->version)); pr_debug(" uuid: %08x.%08x.%08x.%08x\n", - *(__u32 *)(sb->uuid+0), - *(__u32 *)(sb->uuid+4), - *(__u32 *)(sb->uuid+8), - *(__u32 *)(sb->uuid+12)); + le32_to_cpu(*(__u32 *)(sb->uuid+0)), + le32_to_cpu(*(__u32 *)(sb->uuid+4)), + le32_to_cpu(*(__u32 *)(sb->uuid+8)), + le32_to_cpu(*(__u32 *)(sb->uuid+12))); pr_debug(" events: %llu\n", (unsigned long long) le64_to_cpu(sb->events)); pr_debug("events cleared: %llu\n", -- cgit v1.2.3 From e153903686deac9ea8c499b49516126f1a844fdb Mon Sep 17 00:00:00 2001 From: Nix Date: Tue, 16 May 2017 10:13:31 +0100 Subject: md: report sector of stripes with check mismatches This makes it possible, with appropriate filesystem support, for a sysadmin to tell what is affected by the mismatch, and whether it should be ignored (if it's inside a swap partition, for instance). We ratelimit to prevent log flooding: if there are so many mismatches that ratelimiting is necessary, the individual messages are relatively unlikely to be important (either the machine is swapping like crazy or something is very wrong with the disk). Signed-off-by: Nick Alcock Signed-off-by: Shaohua Li --- drivers/md/raid5.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) (limited to 'drivers/md') diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 9c4f7659f8b1..722064689e82 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -4085,10 +4085,15 @@ static void handle_parity_checks5(struct r5conf *conf, struct stripe_head *sh, set_bit(STRIPE_INSYNC, &sh->state); else { atomic64_add(STRIPE_SECTORS, &conf->mddev->resync_mismatches); - if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery)) + if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery)) { /* don't try to repair!! */ set_bit(STRIPE_INSYNC, &sh->state); - else { + pr_warn_ratelimited("%s: mismatch sector in range " + "%llu-%llu\n", mdname(conf->mddev), + (unsigned long long) sh->sector, + (unsigned long long) sh->sector + + STRIPE_SECTORS); + } else { sh->check_state = check_state_compute_run; set_bit(STRIPE_COMPUTE_RUN, &sh->state); set_bit(STRIPE_OP_COMPUTE_BLK, &s->ops_request); @@ -4237,10 +4242,15 @@ static void handle_parity_checks6(struct r5conf *conf, struct stripe_head *sh, } } else { atomic64_add(STRIPE_SECTORS, &conf->mddev->resync_mismatches); - if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery)) + if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery)) { /* don't try to repair!! */ set_bit(STRIPE_INSYNC, &sh->state); - else { + pr_warn_ratelimited("%s: mismatch sector in range " + "%llu-%llu\n", mdname(conf->mddev), + (unsigned long long) sh->sector, + (unsigned long long) sh->sector + + STRIPE_SECTORS); + } else { int *target = &sh->ops.target; sh->ops.target = -1; -- cgit v1.2.3 From ff0361b34ac63ef80c785c32d62e0e9d89a2cf89 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Wed, 31 May 2017 09:44:32 +0200 Subject: dm: make flush bios explicitly sync Commit b685d3d65ac7 ("block: treat REQ_FUA and REQ_PREFLUSH as synchronous") removed REQ_SYNC flag from WRITE_{FUA|PREFLUSH|...} definitions. generic_make_request_checks() however strips REQ_FUA and REQ_PREFLUSH flags from a bio when the storage doesn't report volatile write cache and thus write effectively becomes asynchronous which can lead to performance regressions. Fix the problem by making sure all bios which are synchronous are properly marked with REQ_SYNC. Fixes: b685d3d65ac7 ("block: treat REQ_FUA and REQ_PREFLUSH as synchronous") Cc: stable@vger.kernel.org Signed-off-by: Jan Kara Signed-off-by: Mike Snitzer --- drivers/md/dm-bufio.c | 2 +- drivers/md/dm-integrity.c | 3 ++- drivers/md/dm-raid1.c | 2 +- drivers/md/dm-snap-persistent.c | 3 ++- drivers/md/dm.c | 2 +- 5 files changed, 7 insertions(+), 5 deletions(-) (limited to 'drivers/md') diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c index cd8139593ccd..840c1496b2b1 100644 --- a/drivers/md/dm-bufio.c +++ b/drivers/md/dm-bufio.c @@ -1334,7 +1334,7 @@ int dm_bufio_issue_flush(struct dm_bufio_client *c) { struct dm_io_request io_req = { .bi_op = REQ_OP_WRITE, - .bi_op_flags = REQ_PREFLUSH, + .bi_op_flags = REQ_PREFLUSH | REQ_SYNC, .mem.type = DM_IO_KMEM, .mem.ptr.addr = NULL, .client = c->dm_io, diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c index 1feeb2ccf5a1..7910bfe50da4 100644 --- a/drivers/md/dm-integrity.c +++ b/drivers/md/dm-integrity.c @@ -783,7 +783,8 @@ static void write_journal(struct dm_integrity_c *ic, unsigned commit_start, unsi for (i = 0; i < commit_sections; i++) rw_section_mac(ic, commit_start + i, true); } - rw_journal(ic, REQ_OP_WRITE, REQ_FUA, commit_start, commit_sections, &io_comp); + rw_journal(ic, REQ_OP_WRITE, REQ_FUA | REQ_SYNC, commit_start, + commit_sections, &io_comp); } else { unsigned to_end; io_comp.in_flight = (atomic_t)ATOMIC_INIT(2); diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index a95cbb80fb34..e61c45047c25 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -260,7 +260,7 @@ static int mirror_flush(struct dm_target *ti) struct mirror *m; struct dm_io_request io_req = { .bi_op = REQ_OP_WRITE, - .bi_op_flags = REQ_PREFLUSH, + .bi_op_flags = REQ_PREFLUSH | REQ_SYNC, .mem.type = DM_IO_KMEM, .mem.ptr.addr = NULL, .client = ms->io_client, diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c index b93476c3ba3f..c5534d294773 100644 --- a/drivers/md/dm-snap-persistent.c +++ b/drivers/md/dm-snap-persistent.c @@ -741,7 +741,8 @@ static void persistent_commit_exception(struct dm_exception_store *store, /* * Commit exceptions to disk. */ - if (ps->valid && area_io(ps, REQ_OP_WRITE, REQ_PREFLUSH | REQ_FUA)) + if (ps->valid && area_io(ps, REQ_OP_WRITE, + REQ_PREFLUSH | REQ_FUA | REQ_SYNC)) ps->valid = 0; /* diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 6ef9500226c0..37ccd73c79ec 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1657,7 +1657,7 @@ static struct mapped_device *alloc_dev(int minor) bio_init(&md->flush_bio, NULL, 0); md->flush_bio.bi_bdev = md->bdev; - md->flush_bio.bi_opf = REQ_OP_WRITE | REQ_PREFLUSH; + md->flush_bio.bi_opf = REQ_OP_WRITE | REQ_PREFLUSH | REQ_SYNC; dm_stats_init(&md->stats); -- cgit v1.2.3 From 5a8948f8a32ba56c17b3fb75d318ac98157f3ba5 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Wed, 31 May 2017 09:44:33 +0200 Subject: md: Make flush bios explicitely sync Commit b685d3d65ac7 "block: treat REQ_FUA and REQ_PREFLUSH as synchronous" removed REQ_SYNC flag from WRITE_{FUA|PREFLUSH|...} definitions. generic_make_request_checks() however strips REQ_FUA and REQ_PREFLUSH flags from a bio when the storage doesn't report volatile write cache and thus write effectively becomes asynchronous which can lead to performance regressions Fix the problem by making sure all bios which are synchronous are properly marked with REQ_SYNC. CC: linux-raid@vger.kernel.org CC: Shaohua Li Fixes: b685d3d65ac791406e0dfd8779cc9b3707fea5a3 CC: stable@vger.kernel.org Signed-off-by: Jan Kara Signed-off-by: Shaohua Li --- drivers/md/md.c | 2 +- drivers/md/raid5-cache.c | 4 ++-- drivers/md/raid5-ppl.c | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) (limited to 'drivers/md') diff --git a/drivers/md/md.c b/drivers/md/md.c index 10367ffe92e3..212a6777ff31 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -765,7 +765,7 @@ void md_super_write(struct mddev *mddev, struct md_rdev *rdev, test_bit(FailFast, &rdev->flags) && !test_bit(LastDev, &rdev->flags)) ff = MD_FAILFAST; - bio->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH | REQ_FUA | ff; + bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_PREFLUSH | REQ_FUA | ff; atomic_inc(&mddev->pending_writes); submit_bio(bio); diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c index 4c00bc248287..0a7af8b0a80a 100644 --- a/drivers/md/raid5-cache.c +++ b/drivers/md/raid5-cache.c @@ -1782,7 +1782,7 @@ static int r5l_log_write_empty_meta_block(struct r5l_log *log, sector_t pos, mb->checksum = cpu_to_le32(crc32c_le(log->uuid_checksum, mb, PAGE_SIZE)); if (!sync_page_io(log->rdev, pos, PAGE_SIZE, page, REQ_OP_WRITE, - REQ_FUA, false)) { + REQ_SYNC | REQ_FUA, false)) { __free_page(page); return -EIO; } @@ -2388,7 +2388,7 @@ r5c_recovery_rewrite_data_only_stripes(struct r5l_log *log, mb->checksum = cpu_to_le32(crc32c_le(log->uuid_checksum, mb, PAGE_SIZE)); sync_page_io(log->rdev, ctx->pos, PAGE_SIZE, page, - REQ_OP_WRITE, REQ_FUA, false); + REQ_OP_WRITE, REQ_SYNC | REQ_FUA, false); sh->log_start = ctx->pos; list_add_tail(&sh->r5c, &log->stripe_in_journal_list); atomic_inc(&log->stripe_in_journal_count); diff --git a/drivers/md/raid5-ppl.c b/drivers/md/raid5-ppl.c index 5d25bebf3328..ccce92e68d7f 100644 --- a/drivers/md/raid5-ppl.c +++ b/drivers/md/raid5-ppl.c @@ -907,8 +907,8 @@ static int ppl_write_empty_header(struct ppl_log *log) pplhdr->checksum = cpu_to_le32(~crc32c_le(~0, pplhdr, PAGE_SIZE)); if (!sync_page_io(rdev, rdev->ppl.sector - rdev->data_offset, - PPL_HEADER_SIZE, page, REQ_OP_WRITE | REQ_FUA, 0, - false)) { + PPL_HEADER_SIZE, page, REQ_OP_WRITE | REQ_SYNC | + REQ_FUA, 0, false)) { md_error(rdev->mddev, rdev); ret = -EIO; } -- cgit v1.2.3 From a415c0f10627913793709ddb75add09d2ea334dc Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 5 Jun 2017 16:05:13 +1000 Subject: md: initialise ->writes_pending in personality modules. The new per-cpu counter for writes_pending is initialised in md_alloc(), which is not called by dm-raid. So dm-raid fails when md_write_start() is called. Move the initialization to the personality modules that need it. This way it is always initialised when needed, but isn't unnecessarily initialized (requiring memory allocation) when the personality doesn't use writes_pending. Reported-by: Heinz Mauelshagen Fixes: 4ad23a976413 ("MD: use per-cpu counter for writes_pending") Signed-off-by: NeilBrown Signed-off-by: Shaohua Li --- drivers/md/md.c | 16 ++++++++++++---- drivers/md/md.h | 1 + drivers/md/raid1.c | 2 ++ drivers/md/raid10.c | 3 +++ drivers/md/raid5.c | 3 +++ 5 files changed, 21 insertions(+), 4 deletions(-) (limited to 'drivers/md') diff --git a/drivers/md/md.c b/drivers/md/md.c index 212a6777ff31..87edc342ccb3 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -5174,6 +5174,18 @@ static void mddev_delayed_delete(struct work_struct *ws) static void no_op(struct percpu_ref *r) {} +int mddev_init_writes_pending(struct mddev *mddev) +{ + if (mddev->writes_pending.percpu_count_ptr) + return 0; + if (percpu_ref_init(&mddev->writes_pending, no_op, 0, GFP_KERNEL) < 0) + return -ENOMEM; + /* We want to start with the refcount at zero */ + percpu_ref_put(&mddev->writes_pending); + return 0; +} +EXPORT_SYMBOL_GPL(mddev_init_writes_pending); + static int md_alloc(dev_t dev, char *name) { /* @@ -5239,10 +5251,6 @@ static int md_alloc(dev_t dev, char *name) blk_queue_make_request(mddev->queue, md_make_request); blk_set_stacking_limits(&mddev->queue->limits); - if (percpu_ref_init(&mddev->writes_pending, no_op, 0, GFP_KERNEL) < 0) - goto abort; - /* We want to start with the refcount at zero */ - percpu_ref_put(&mddev->writes_pending); disk = alloc_disk(1 << shift); if (!disk) { blk_cleanup_queue(mddev->queue); diff --git a/drivers/md/md.h b/drivers/md/md.h index 11f15146ce51..0fa1de42c42b 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -648,6 +648,7 @@ extern void md_unregister_thread(struct md_thread **threadp); extern void md_wakeup_thread(struct md_thread *thread); extern void md_check_recovery(struct mddev *mddev); extern void md_reap_sync_thread(struct mddev *mddev); +extern int mddev_init_writes_pending(struct mddev *mddev); extern void md_write_start(struct mddev *mddev, struct bio *bi); extern void md_write_inc(struct mddev *mddev, struct bio *bi); extern void md_write_end(struct mddev *mddev); diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index af5056d56878..e1a7e3d4c5e4 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -3063,6 +3063,8 @@ static int raid1_run(struct mddev *mddev) mdname(mddev)); return -EIO; } + if (mddev_init_writes_pending(mddev) < 0) + return -ENOMEM; /* * copy the already verified devices into our private RAID1 * bookkeeping area. [whatever we allocate in run(), diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 4343d7ff9916..797ed60abd5e 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -3611,6 +3611,9 @@ static int raid10_run(struct mddev *mddev) int first = 1; bool discard_supported = false; + if (mddev_init_writes_pending(mddev) < 0) + return -ENOMEM; + if (mddev->private == NULL) { conf = setup_conf(mddev); if (IS_ERR(conf)) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 722064689e82..ec0f951ae19f 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -7118,6 +7118,9 @@ static int raid5_run(struct mddev *mddev) long long min_offset_diff = 0; int first = 1; + if (mddev_init_writes_pending(mddev) < 0) + return -ENOMEM; + if (mddev->recovery_cp != MaxSector) pr_notice("md/raid:%s: not clean -- starting background reconstruction\n", mdname(mddev)); -- cgit v1.2.3