diff options
Diffstat (limited to 'drivers/md')
31 files changed, 767 insertions, 2021 deletions
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig index 3ff87cb4dc49..a743e2c572fc 100644 --- a/drivers/md/Kconfig +++ b/drivers/md/Kconfig @@ -61,19 +61,6 @@ config MD_BITMAP_FILE various kernel APIs and can only work with files on a file system not actually sitting on the MD device. -config MD_LINEAR - tristate "Linear (append) mode (deprecated)" - depends on BLK_DEV_MD - help - If you say Y here, then your multiple devices driver will be able to - use the so-called linear mode, i.e. it will combine the hard disk - partitions by simply appending one to the other. - - To compile this as a module, choose M here: the module - will be called linear. - - If unsure, say Y. - config MD_RAID0 tristate "RAID-0 (striping) mode" depends on BLK_DEV_MD @@ -172,27 +159,6 @@ config MD_RAID456 If unsure, say Y. -config MD_MULTIPATH - tristate "Multipath I/O support (deprecated)" - depends on BLK_DEV_MD - help - MD_MULTIPATH provides a simple multi-path personality for use - the MD framework. It is not under active development. New - projects should consider using DM_MULTIPATH which has more - features and more testing. - - If unsure, say N. - -config MD_FAULTY - tristate "Faulty test module for MD (deprecated)" - depends on BLK_DEV_MD - help - The "faulty" module allows for a block device that occasionally returns - read or write errors. It is useful for testing. - - In unsure, say N. - - config MD_CLUSTER tristate "Cluster Support for MD" depends on BLK_DEV_MD diff --git a/drivers/md/Makefile b/drivers/md/Makefile index 84291e38dca8..027d7cfeca3f 100644 --- a/drivers/md/Makefile +++ b/drivers/md/Makefile @@ -29,22 +29,16 @@ dm-zoned-y += dm-zoned-target.o dm-zoned-metadata.o dm-zoned-reclaim.o md-mod-y += md.o md-bitmap.o raid456-y += raid5.o raid5-cache.o raid5-ppl.o -linear-y += md-linear.o -multipath-y += md-multipath.o -faulty-y += md-faulty.o # Note: link order is important. All raid personalities -# and must come before md.o, as they each initialise -# themselves, and md.o may use the personalities when it +# and must come before md.o, as they each initialise +# themselves, and md.o may use the personalities when it # auto-initialised. -obj-$(CONFIG_MD_LINEAR) += linear.o obj-$(CONFIG_MD_RAID0) += raid0.o obj-$(CONFIG_MD_RAID1) += raid1.o obj-$(CONFIG_MD_RAID10) += raid10.o obj-$(CONFIG_MD_RAID456) += raid456.o -obj-$(CONFIG_MD_MULTIPATH) += multipath.o -obj-$(CONFIG_MD_FAULTY) += faulty.o obj-$(CONFIG_MD_CLUSTER) += md-cluster.o obj-$(CONFIG_BCACHE) += bcache/ obj-$(CONFIG_BLK_DEV_MD) += md-mod.o diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index 1402096b8076..dc3f50f69714 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -954,7 +954,6 @@ static int bcache_device_init(struct bcache_device *d, unsigned int block_size, q->limits.max_segment_size = UINT_MAX; q->limits.max_segments = BIO_MAX_VECS; blk_queue_max_discard_sectors(q, UINT_MAX); - q->limits.discard_granularity = 512; q->limits.io_min = block_size; q->limits.logical_block_size = block_size; q->limits.physical_block_size = block_size; diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c index f03d7dba270c..13c65b7e1ed6 100644 --- a/drivers/md/dm-bufio.c +++ b/drivers/md/dm-bufio.c @@ -1170,7 +1170,7 @@ static void __cache_size_refresh(void) * If the allocation may fail we use __get_free_pages. Memory fragmentation * won't have a fatal effect here, but it just causes flushes of some other * buffers and more I/O will be performed. Don't use __get_free_pages if it - * always fails (i.e. order > MAX_ORDER). + * always fails (i.e. order > MAX_PAGE_ORDER). * * If the allocation shouldn't fail we use __vmalloc. This is only for the * initial reserve allocation, so there's no risk of wasting all vmalloc diff --git a/drivers/md/dm-core.h b/drivers/md/dm-core.h index 095b9b49aa82..e6757a30dcca 100644 --- a/drivers/md/dm-core.h +++ b/drivers/md/dm-core.h @@ -22,6 +22,8 @@ #include "dm-ima.h" #define DM_RESERVED_MAX_IOS 1024 +#define DM_MAX_TARGETS 1048576 +#define DM_MAX_TARGET_PARAMS 1024 struct dm_io; diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 2ae8560b6a14..59445763e55a 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -53,15 +53,17 @@ struct convert_context { struct completion restart; struct bio *bio_in; - struct bio *bio_out; struct bvec_iter iter_in; + struct bio *bio_out; struct bvec_iter iter_out; - u64 cc_sector; atomic_t cc_pending; + u64 cc_sector; union { struct skcipher_request *req; struct aead_request *req_aead; } r; + bool aead_recheck; + bool aead_failed; }; @@ -73,10 +75,8 @@ struct dm_crypt_io { struct bio *base_bio; u8 *integrity_metadata; bool integrity_metadata_from_pool:1; - bool in_tasklet:1; struct work_struct work; - struct tasklet_struct tasklet; struct convert_context ctx; @@ -84,6 +84,8 @@ struct dm_crypt_io { blk_status_t error; sector_t sector; + struct bvec_iter saved_bi_iter; + struct rb_node rb_node; } CRYPTO_MINALIGN_ATTR; @@ -1372,10 +1374,13 @@ static int crypt_convert_block_aead(struct crypt_config *cc, if (r == -EBADMSG) { sector_t s = le64_to_cpu(*sector); - DMERR_LIMIT("%pg: INTEGRITY AEAD ERROR, sector %llu", - ctx->bio_in->bi_bdev, s); - dm_audit_log_bio(DM_MSG_PREFIX, "integrity-aead", - ctx->bio_in, s, 0); + ctx->aead_failed = true; + if (ctx->aead_recheck) { + DMERR_LIMIT("%pg: INTEGRITY AEAD ERROR, sector %llu", + ctx->bio_in->bi_bdev, s); + dm_audit_log_bio(DM_MSG_PREFIX, "integrity-aead", + ctx->bio_in, s, 0); + } } if (!r && cc->iv_gen_ops && cc->iv_gen_ops->post) @@ -1673,7 +1678,7 @@ static struct bio *crypt_alloc_buffer(struct dm_crypt_io *io, unsigned int size) unsigned int nr_iovecs = (size + PAGE_SIZE - 1) >> PAGE_SHIFT; gfp_t gfp_mask = GFP_NOWAIT | __GFP_HIGHMEM; unsigned int remaining_size; - unsigned int order = MAX_ORDER; + unsigned int order = MAX_PAGE_ORDER; retry: if (unlikely(gfp_mask & __GFP_DIRECT_RECLAIM)) @@ -1759,10 +1764,11 @@ static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc, io->base_bio = bio; io->sector = sector; io->error = 0; + io->ctx.aead_recheck = false; + io->ctx.aead_failed = false; io->ctx.r.req = NULL; io->integrity_metadata = NULL; io->integrity_metadata_from_pool = false; - io->in_tasklet = false; atomic_set(&io->io_pending, 0); } @@ -1771,12 +1777,7 @@ static void crypt_inc_pending(struct dm_crypt_io *io) atomic_inc(&io->io_pending); } -static void kcryptd_io_bio_endio(struct work_struct *work) -{ - struct dm_crypt_io *io = container_of(work, struct dm_crypt_io, work); - - bio_endio(io->base_bio); -} +static void kcryptd_queue_read(struct dm_crypt_io *io); /* * One of the bios was finished. Check for completion of @@ -1791,6 +1792,15 @@ static void crypt_dec_pending(struct dm_crypt_io *io) if (!atomic_dec_and_test(&io->io_pending)) return; + if (likely(!io->ctx.aead_recheck) && unlikely(io->ctx.aead_failed) && + cc->on_disk_tag_size && bio_data_dir(base_bio) == READ) { + io->ctx.aead_recheck = true; + io->ctx.aead_failed = false; + io->error = 0; + kcryptd_queue_read(io); + return; + } + if (io->ctx.r.req) crypt_free_req(cc, io->ctx.r.req, base_bio); @@ -1801,20 +1811,6 @@ static void crypt_dec_pending(struct dm_crypt_io *io) base_bio->bi_status = error; - /* - * If we are running this function from our tasklet, - * we can't call bio_endio() here, because it will call - * clone_endio() from dm.c, which in turn will - * free the current struct dm_crypt_io structure with - * our tasklet. In this case we need to delay bio_endio() - * execution to after the tasklet is done and dequeued. - */ - if (io->in_tasklet) { - INIT_WORK(&io->work, kcryptd_io_bio_endio); - queue_work(cc->io_queue, &io->work); - return; - } - bio_endio(base_bio); } @@ -1840,15 +1836,19 @@ static void crypt_endio(struct bio *clone) struct dm_crypt_io *io = clone->bi_private; struct crypt_config *cc = io->cc; unsigned int rw = bio_data_dir(clone); - blk_status_t error; + blk_status_t error = clone->bi_status; + + if (io->ctx.aead_recheck && !error) { + kcryptd_queue_crypt(io); + return; + } /* * free the processed pages */ - if (rw == WRITE) + if (rw == WRITE || io->ctx.aead_recheck) crypt_free_buffer_pages(cc, clone); - error = clone->bi_status; bio_put(clone); if (rw == READ && !error) { @@ -1869,6 +1869,22 @@ static int kcryptd_io_read(struct dm_crypt_io *io, gfp_t gfp) struct crypt_config *cc = io->cc; struct bio *clone; + if (io->ctx.aead_recheck) { + if (!(gfp & __GFP_DIRECT_RECLAIM)) + return 1; + crypt_inc_pending(io); + clone = crypt_alloc_buffer(io, io->base_bio->bi_iter.bi_size); + if (unlikely(!clone)) { + crypt_dec_pending(io); + return 1; + } + clone->bi_iter.bi_sector = cc->start + io->sector; + crypt_convert_init(cc, &io->ctx, clone, clone, io->sector); + io->saved_bi_iter = clone->bi_iter; + dm_submit_bio_remap(io->base_bio, clone); + return 0; + } + /* * We need the original biovec array in order to decrypt the whole bio * data *afterwards* -- thanks to immutable biovecs we don't need to @@ -2095,6 +2111,12 @@ static void kcryptd_crypt_write_convert(struct dm_crypt_io *io) io->ctx.bio_out = clone; io->ctx.iter_out = clone->bi_iter; + if (crypt_integrity_aead(cc)) { + bio_copy_data(clone, io->base_bio); + io->ctx.bio_in = clone; + io->ctx.iter_in = clone->bi_iter; + } + sector += bio_sectors(clone); crypt_inc_pending(io); @@ -2131,6 +2153,14 @@ dec: static void kcryptd_crypt_read_done(struct dm_crypt_io *io) { + if (io->ctx.aead_recheck) { + if (!io->error) { + io->ctx.bio_in->bi_iter = io->saved_bi_iter; + bio_copy_data(io->base_bio, io->ctx.bio_in); + } + crypt_free_buffer_pages(io->cc, io->ctx.bio_in); + bio_put(io->ctx.bio_in); + } crypt_dec_pending(io); } @@ -2160,11 +2190,17 @@ static void kcryptd_crypt_read_convert(struct dm_crypt_io *io) crypt_inc_pending(io); - crypt_convert_init(cc, &io->ctx, io->base_bio, io->base_bio, - io->sector); + if (io->ctx.aead_recheck) { + io->ctx.cc_sector = io->sector + cc->iv_offset; + r = crypt_convert(cc, &io->ctx, + test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags), true); + } else { + crypt_convert_init(cc, &io->ctx, io->base_bio, io->base_bio, + io->sector); - r = crypt_convert(cc, &io->ctx, - test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags), true); + r = crypt_convert(cc, &io->ctx, + test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags), true); + } /* * Crypto API backlogged the request, because its queue was full * and we're in softirq context, so continue from a workqueue @@ -2206,10 +2242,13 @@ static void kcryptd_async_done(void *data, int error) if (error == -EBADMSG) { sector_t s = le64_to_cpu(*org_sector_of_dmreq(cc, dmreq)); - DMERR_LIMIT("%pg: INTEGRITY AEAD ERROR, sector %llu", - ctx->bio_in->bi_bdev, s); - dm_audit_log_bio(DM_MSG_PREFIX, "integrity-aead", - ctx->bio_in, s, 0); + ctx->aead_failed = true; + if (ctx->aead_recheck) { + DMERR_LIMIT("%pg: INTEGRITY AEAD ERROR, sector %llu", + ctx->bio_in->bi_bdev, s); + dm_audit_log_bio(DM_MSG_PREFIX, "integrity-aead", + ctx->bio_in, s, 0); + } io->error = BLK_STS_PROTECTION; } else if (error < 0) io->error = BLK_STS_IOERR; @@ -2246,11 +2285,6 @@ static void kcryptd_crypt(struct work_struct *work) kcryptd_crypt_write_convert(io); } -static void kcryptd_crypt_tasklet(unsigned long work) -{ - kcryptd_crypt((struct work_struct *)work); -} - static void kcryptd_queue_crypt(struct dm_crypt_io *io) { struct crypt_config *cc = io->cc; @@ -2262,15 +2296,10 @@ static void kcryptd_queue_crypt(struct dm_crypt_io *io) * irqs_disabled(): the kernel may run some IO completion from the idle thread, but * it is being executed with irqs disabled. */ - if (in_hardirq() || irqs_disabled()) { - io->in_tasklet = true; - tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work); - tasklet_schedule(&io->tasklet); + if (!(in_hardirq() || irqs_disabled())) { + kcryptd_crypt(&io->work); return; } - - kcryptd_crypt(&io->work); - return; } INIT_WORK(&io->work, kcryptd_crypt); @@ -3144,7 +3173,7 @@ static int crypt_ctr_optional(struct dm_target *ti, unsigned int argc, char **ar sval = strchr(opt_string + strlen("integrity:"), ':') + 1; if (!strcasecmp(sval, "aead")) { set_bit(CRYPT_MODE_INTEGRITY_AEAD, &cc->cipher_flags); - } else if (strcasecmp(sval, "none")) { + } else if (strcasecmp(sval, "none")) { ti->error = "Unknown integrity profile"; return -EINVAL; } @@ -3673,7 +3702,7 @@ static void crypt_io_hints(struct dm_target *ti, struct queue_limits *limits) static struct target_type crypt_target = { .name = "crypt", - .version = {1, 24, 0}, + .version = {1, 25, 0}, .module = THIS_MODULE, .ctr = crypt_ctr, .dtr = crypt_dtr, diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c index f57fb821528d..7916ed9f10e8 100644 --- a/drivers/md/dm-flakey.c +++ b/drivers/md/dm-flakey.c @@ -434,7 +434,7 @@ static struct bio *clone_bio(struct dm_target *ti, struct flakey_c *fc, struct b remaining_size = size; - order = MAX_ORDER; + order = MAX_PAGE_ORDER; while (remaining_size) { struct page *pages; unsigned size_to_add, to_copy; diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c index c5f03aab4552..1fc901df84eb 100644 --- a/drivers/md/dm-integrity.c +++ b/drivers/md/dm-integrity.c @@ -278,6 +278,8 @@ struct dm_integrity_c { atomic64_t number_of_mismatches; + mempool_t recheck_pool; + struct notifier_block reboot_notifier; }; @@ -1689,6 +1691,77 @@ failed: get_random_bytes(result, ic->tag_size); } +static noinline void integrity_recheck(struct dm_integrity_io *dio, char *checksum) +{ + struct bio *bio = dm_bio_from_per_bio_data(dio, sizeof(struct dm_integrity_io)); + struct dm_integrity_c *ic = dio->ic; + struct bvec_iter iter; + struct bio_vec bv; + sector_t sector, logical_sector, area, offset; + struct page *page; + void *buffer; + + get_area_and_offset(ic, dio->range.logical_sector, &area, &offset); + dio->metadata_block = get_metadata_sector_and_offset(ic, area, offset, + &dio->metadata_offset); + sector = get_data_sector(ic, area, offset); + logical_sector = dio->range.logical_sector; + + page = mempool_alloc(&ic->recheck_pool, GFP_NOIO); + buffer = page_to_virt(page); + + __bio_for_each_segment(bv, bio, iter, dio->bio_details.bi_iter) { + unsigned pos = 0; + + do { + char *mem; + int r; + struct dm_io_request io_req; + struct dm_io_region io_loc; + io_req.bi_opf = REQ_OP_READ; + io_req.mem.type = DM_IO_KMEM; + io_req.mem.ptr.addr = buffer; + io_req.notify.fn = NULL; + io_req.client = ic->io; + io_loc.bdev = ic->dev->bdev; + io_loc.sector = sector; + io_loc.count = ic->sectors_per_block; + + r = dm_io(&io_req, 1, &io_loc, NULL); + if (unlikely(r)) { + dio->bi_status = errno_to_blk_status(r); + goto free_ret; + } + + integrity_sector_checksum(ic, logical_sector, buffer, checksum); + r = dm_integrity_rw_tag(ic, checksum, &dio->metadata_block, + &dio->metadata_offset, ic->tag_size, TAG_CMP); + if (r) { + if (r > 0) { + DMERR_LIMIT("%pg: Checksum failed at sector 0x%llx", + bio->bi_bdev, logical_sector); + atomic64_inc(&ic->number_of_mismatches); + dm_audit_log_bio(DM_MSG_PREFIX, "integrity-checksum", + bio, logical_sector, 0); + r = -EILSEQ; + } + dio->bi_status = errno_to_blk_status(r); + goto free_ret; + } + + mem = bvec_kmap_local(&bv); + memcpy(mem + pos, buffer, ic->sectors_per_block << SECTOR_SHIFT); + kunmap_local(mem); + + pos += ic->sectors_per_block << SECTOR_SHIFT; + sector += ic->sectors_per_block; + logical_sector += ic->sectors_per_block; + } while (pos < bv.bv_len); + } +free_ret: + mempool_free(page, &ic->recheck_pool); +} + static void integrity_metadata(struct work_struct *w) { struct dm_integrity_io *dio = container_of(w, struct dm_integrity_io, work); @@ -1776,15 +1849,8 @@ again: checksums_ptr - checksums, dio->op == REQ_OP_READ ? TAG_CMP : TAG_WRITE); if (unlikely(r)) { if (r > 0) { - sector_t s; - - s = sector - ((r + ic->tag_size - 1) / ic->tag_size); - DMERR_LIMIT("%pg: Checksum failed at sector 0x%llx", - bio->bi_bdev, s); - r = -EILSEQ; - atomic64_inc(&ic->number_of_mismatches); - dm_audit_log_bio(DM_MSG_PREFIX, "integrity-checksum", - bio, s, 0); + integrity_recheck(dio, checksums); + goto skip_io; } if (likely(checksums != checksums_onstack)) kfree(checksums); @@ -4261,6 +4327,12 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned int argc, char **argv goto bad; } + r = mempool_init_page_pool(&ic->recheck_pool, 1, 0); + if (r) { + ti->error = "Cannot allocate mempool"; + goto bad; + } + ic->metadata_wq = alloc_workqueue("dm-integrity-metadata", WQ_MEM_RECLAIM, METADATA_WORKQUEUE_MAX_ACTIVE); if (!ic->metadata_wq) { @@ -4609,6 +4681,7 @@ static void dm_integrity_dtr(struct dm_target *ti) kvfree(ic->bbs); if (ic->bufio) dm_bufio_client_destroy(ic->bufio); + mempool_exit(&ic->recheck_pool); mempool_exit(&ic->journal_io_mempool); if (ic->io) dm_io_client_destroy(ic->io); @@ -4661,7 +4734,7 @@ static void dm_integrity_dtr(struct dm_target *ti) static struct target_type integrity_target = { .name = "integrity", - .version = {1, 10, 0}, + .version = {1, 11, 0}, .module = THIS_MODULE, .features = DM_TARGET_SINGLETON | DM_TARGET_INTEGRITY, .ctr = dm_integrity_ctr, diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index e65058e0ed06..3b1ad7127cb8 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1941,7 +1941,8 @@ static int copy_params(struct dm_ioctl __user *user, struct dm_ioctl *param_kern minimum_data_size - sizeof(param_kernel->version))) return -EFAULT; - if (param_kernel->data_size < minimum_data_size) { + if (unlikely(param_kernel->data_size < minimum_data_size) || + unlikely(param_kernel->data_size > DM_MAX_TARGETS * DM_MAX_TARGET_PARAMS)) { DMERR("Invalid data size in the ioctl structure: %u", param_kernel->data_size); return -EINVAL; diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c index d01807c50f20..36bcfdccae04 100644 --- a/drivers/md/dm-kcopyd.c +++ b/drivers/md/dm-kcopyd.c @@ -807,7 +807,7 @@ void dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from, */ if (!(job->flags & BIT(DM_KCOPYD_WRITE_SEQ))) { for (i = 0; i < job->num_dests; i++) { - if (bdev_zoned_model(dests[i].bdev) == BLK_ZONED_HM) { + if (bdev_is_zoned(dests[i].bdev)) { job->flags |= BIT(DM_KCOPYD_WRITE_SEQ); break; } diff --git a/drivers/md/dm-stats.c b/drivers/md/dm-stats.c index bdc14ec99814..1e5d988f44da 100644 --- a/drivers/md/dm-stats.c +++ b/drivers/md/dm-stats.c @@ -66,6 +66,9 @@ struct dm_stats_last_position { unsigned int last_rw; }; +#define DM_STAT_MAX_ENTRIES 8388608 +#define DM_STAT_MAX_HISTOGRAM_ENTRIES 134217728 + /* * A typo on the command line could possibly make the kernel run out of memory * and crash. To prevent the crash we account all used memory. We fail if we @@ -285,6 +288,9 @@ static int dm_stats_create(struct dm_stats *stats, sector_t start, sector_t end, if (n_entries != (size_t)n_entries || !(size_t)(n_entries + 1)) return -EOVERFLOW; + if (n_entries > DM_STAT_MAX_ENTRIES) + return -EOVERFLOW; + shared_alloc_size = struct_size(s, stat_shared, n_entries); if ((shared_alloc_size - sizeof(struct dm_stat)) / sizeof(struct dm_stat_shared) != n_entries) return -EOVERFLOW; @@ -297,6 +303,9 @@ static int dm_stats_create(struct dm_stats *stats, sector_t start, sector_t end, if (histogram_alloc_size / (n_histogram_entries + 1) != (size_t)n_entries * sizeof(unsigned long long)) return -EOVERFLOW; + if ((n_histogram_entries + 1) * (size_t)n_entries > DM_STAT_MAX_HISTOGRAM_ENTRIES) + return -EOVERFLOW; + if (!check_shared_memory(shared_alloc_size + histogram_alloc_size + num_possible_cpus() * (percpu_alloc_size + histogram_alloc_size))) return -ENOMEM; diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 198d38b53322..41f1d731ae5a 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -129,7 +129,12 @@ static int alloc_targets(struct dm_table *t, unsigned int num) int dm_table_create(struct dm_table **result, blk_mode_t mode, unsigned int num_targets, struct mapped_device *md) { - struct dm_table *t = kzalloc(sizeof(*t), GFP_KERNEL); + struct dm_table *t; + + if (num_targets > DM_MAX_TARGETS) + return -EOVERFLOW; + + t = kzalloc(sizeof(*t), GFP_KERNEL); if (!t) return -ENOMEM; @@ -144,7 +149,7 @@ int dm_table_create(struct dm_table **result, blk_mode_t mode, if (!num_targets) { kfree(t); - return -ENOMEM; + return -EOVERFLOW; } if (alloc_targets(t, num_targets)) { @@ -1579,21 +1584,18 @@ bool dm_table_has_no_data_devices(struct dm_table *t) return true; } -static int device_not_zoned_model(struct dm_target *ti, struct dm_dev *dev, - sector_t start, sector_t len, void *data) +static int device_not_zoned(struct dm_target *ti, struct dm_dev *dev, + sector_t start, sector_t len, void *data) { - struct request_queue *q = bdev_get_queue(dev->bdev); - enum blk_zoned_model *zoned_model = data; + bool *zoned = data; - return blk_queue_zoned_model(q) != *zoned_model; + return bdev_is_zoned(dev->bdev) != *zoned; } static int device_is_zoned_model(struct dm_target *ti, struct dm_dev *dev, sector_t start, sector_t len, void *data) { - struct request_queue *q = bdev_get_queue(dev->bdev); - - return blk_queue_zoned_model(q) != BLK_ZONED_NONE; + return bdev_is_zoned(dev->bdev); } /* @@ -1603,8 +1605,7 @@ static int device_is_zoned_model(struct dm_target *ti, struct dm_dev *dev, * has the DM_TARGET_MIXED_ZONED_MODEL feature set, the devices can have any * zoned model with all zoned devices having the same zone size. */ -static bool dm_table_supports_zoned_model(struct dm_table *t, - enum blk_zoned_model zoned_model) +static bool dm_table_supports_zoned(struct dm_table *t, bool zoned) { for (unsigned int i = 0; i < t->num_targets; i++) { struct dm_target *ti = dm_table_get_target(t, i); @@ -1623,11 +1624,11 @@ static bool dm_table_supports_zoned_model(struct dm_table *t, if (dm_target_supports_zoned_hm(ti->type)) { if (!ti->type->iterate_devices || - ti->type->iterate_devices(ti, device_not_zoned_model, - &zoned_model)) + ti->type->iterate_devices(ti, device_not_zoned, + &zoned)) return false; } else if (!dm_target_supports_mixed_zoned_model(ti->type)) { - if (zoned_model == BLK_ZONED_HM) + if (zoned) return false; } } @@ -1650,14 +1651,13 @@ static int device_not_matches_zone_sectors(struct dm_target *ti, struct dm_dev * * zone sectors, if the destination device is a zoned block device, it shall * have the specified zone_sectors. */ -static int validate_hardware_zoned_model(struct dm_table *t, - enum blk_zoned_model zoned_model, - unsigned int zone_sectors) +static int validate_hardware_zoned(struct dm_table *t, bool zoned, + unsigned int zone_sectors) { - if (zoned_model == BLK_ZONED_NONE) + if (!zoned) return 0; - if (!dm_table_supports_zoned_model(t, zoned_model)) { + if (!dm_table_supports_zoned(t, zoned)) { DMERR("%s: zoned model is not consistent across all devices", dm_device_name(t->md)); return -EINVAL; @@ -1683,8 +1683,8 @@ int dm_calculate_queue_limits(struct dm_table *t, struct queue_limits *limits) { struct queue_limits ti_limits; - enum blk_zoned_model zoned_model = BLK_ZONED_NONE; unsigned int zone_sectors = 0; + bool zoned = false; blk_set_stacking_limits(limits); @@ -1706,12 +1706,12 @@ int dm_calculate_queue_limits(struct dm_table *t, ti->type->iterate_devices(ti, dm_set_device_limits, &ti_limits); - if (zoned_model == BLK_ZONED_NONE && ti_limits.zoned != BLK_ZONED_NONE) { + if (!zoned && ti_limits.zoned) { /* * After stacking all limits, validate all devices * in table support this zoned model and zone sectors. */ - zoned_model = ti_limits.zoned; + zoned = ti_limits.zoned; zone_sectors = ti_limits.chunk_sectors; } @@ -1744,18 +1744,18 @@ combine_limits: * Verify that the zoned model and zone sectors, as determined before * any .io_hints override, are the same across all devices in the table. * - this is especially relevant if .io_hints is emulating a disk-managed - * zoned model (aka BLK_ZONED_NONE) on host-managed zoned block devices. + * zoned model on host-managed zoned block devices. * BUT... */ - if (limits->zoned != BLK_ZONED_NONE) { + if (limits->zoned) { /* * ...IF the above limits stacking determined a zoned model * validate that all of the table's devices conform to it. */ - zoned_model = limits->zoned; + zoned = limits->zoned; zone_sectors = limits->chunk_sectors; } - if (validate_hardware_zoned_model(t, zoned_model, zone_sectors)) + if (validate_hardware_zoned(t, zoned, zone_sectors)) return -EINVAL; return validate_hardware_logical_block_alignment(t, limits); diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c index 14e58ae70521..1b591bfa90d5 100644 --- a/drivers/md/dm-verity-target.c +++ b/drivers/md/dm-verity-target.c @@ -482,6 +482,63 @@ int verity_for_bv_block(struct dm_verity *v, struct dm_verity_io *io, return 0; } +static int verity_recheck_copy(struct dm_verity *v, struct dm_verity_io *io, + u8 *data, size_t len) +{ + memcpy(data, io->recheck_buffer, len); + io->recheck_buffer += len; + + return 0; +} + +static noinline int verity_recheck(struct dm_verity *v, struct dm_verity_io *io, + struct bvec_iter start, sector_t cur_block) +{ + struct page *page; + void *buffer; + int r; + struct dm_io_request io_req; + struct dm_io_region io_loc; + + page = mempool_alloc(&v->recheck_pool, GFP_NOIO); + buffer = page_to_virt(page); + + io_req.bi_opf = REQ_OP_READ; + io_req.mem.type = DM_IO_KMEM; + io_req.mem.ptr.addr = buffer; + io_req.notify.fn = NULL; + io_req.client = v->io; + io_loc.bdev = v->data_dev->bdev; + io_loc.sector = cur_block << (v->data_dev_block_bits - SECTOR_SHIFT); + io_loc.count = 1 << (v->data_dev_block_bits - SECTOR_SHIFT); + r = dm_io(&io_req, 1, &io_loc, NULL); + if (unlikely(r)) + goto free_ret; + + r = verity_hash(v, verity_io_hash_req(v, io), buffer, + 1 << v->data_dev_block_bits, + verity_io_real_digest(v, io), true); + if (unlikely(r)) + goto free_ret; + + if (memcmp(verity_io_real_digest(v, io), + verity_io_want_digest(v, io), v->digest_size)) { + r = -EIO; + goto free_ret; + } + + io->recheck_buffer = buffer; + r = verity_for_bv_block(v, io, &start, verity_recheck_copy); + if (unlikely(r)) + goto free_ret; + + r = 0; +free_ret: + mempool_free(page, &v->recheck_pool); + + return r; +} + static int verity_bv_zero(struct dm_verity *v, struct dm_verity_io *io, u8 *data, size_t len) { @@ -508,9 +565,7 @@ static int verity_verify_io(struct dm_verity_io *io) { bool is_zero; struct dm_verity *v = io->v; -#if defined(CONFIG_DM_VERITY_FEC) struct bvec_iter start; -#endif struct bvec_iter iter_copy; struct bvec_iter *iter; struct crypto_wait wait; @@ -561,10 +616,7 @@ static int verity_verify_io(struct dm_verity_io *io) if (unlikely(r < 0)) return r; -#if defined(CONFIG_DM_VERITY_FEC) - if (verity_fec_is_enabled(v)) - start = *iter; -#endif + start = *iter; r = verity_for_io_block(v, io, iter, &wait); if (unlikely(r < 0)) return r; @@ -586,6 +638,10 @@ static int verity_verify_io(struct dm_verity_io *io) * tasklet since it may sleep, so fallback to work-queue. */ return -EAGAIN; + } else if (verity_recheck(v, io, start, cur_block) == 0) { + if (v->validated_blocks) + set_bit(cur_block, v->validated_blocks); + continue; #if defined(CONFIG_DM_VERITY_FEC) } else if (verity_fec_decode(v, io, DM_VERITY_BLOCK_TYPE_DATA, cur_block, NULL, &start) == 0) { @@ -645,23 +701,6 @@ static void verity_work(struct work_struct *w) verity_finish_io(io, errno_to_blk_status(verity_verify_io(io))); } -static void verity_tasklet(unsigned long data) -{ - struct dm_verity_io *io = (struct dm_verity_io *)data; - int err; - - io->in_tasklet = true; - err = verity_verify_io(io); - if (err == -EAGAIN || err == -ENOMEM) { - /* fallback to retrying with work-queue */ - INIT_WORK(&io->work, verity_work); - queue_work(io->v->verify_wq, &io->work); - return; - } - - verity_finish_io(io, errno_to_blk_status(err)); -} - static void verity_end_io(struct bio *bio) { struct dm_verity_io *io = bio->bi_private; @@ -674,13 +713,8 @@ static void verity_end_io(struct bio *bio) return; } - if (static_branch_unlikely(&use_tasklet_enabled) && io->v->use_tasklet) { - tasklet_init(&io->tasklet, verity_tasklet, (unsigned long)io); - tasklet_schedule(&io->tasklet); - } else { - INIT_WORK(&io->work, verity_work); - queue_work(io->v->verify_wq, &io->work); - } + INIT_WORK(&io->work, verity_work); + queue_work(io->v->verify_wq, &io->work); } /* @@ -963,6 +997,10 @@ static void verity_dtr(struct dm_target *ti) if (v->verify_wq) destroy_workqueue(v->verify_wq); + mempool_exit(&v->recheck_pool); + if (v->io) + dm_io_client_destroy(v->io); + if (v->bufio) dm_bufio_client_destroy(v->bufio); @@ -1401,6 +1439,20 @@ static int verity_ctr(struct dm_target *ti, unsigned int argc, char **argv) } v->hash_blocks = hash_position; + r = mempool_init_page_pool(&v->recheck_pool, 1, 0); + if (unlikely(r)) { + ti->error = "Cannot allocate mempool"; + goto bad; + } + + v->io = dm_io_client_create(); + if (IS_ERR(v->io)) { + r = PTR_ERR(v->io); + v->io = NULL; + ti->error = "Cannot allocate dm io"; + goto bad; + } + v->bufio = dm_bufio_client_create(v->hash_dev->bdev, 1 << v->hash_dev_block_bits, 1, sizeof(struct buffer_aux), dm_bufio_alloc_callback, NULL, @@ -1508,7 +1560,7 @@ int dm_verity_get_root_digest(struct dm_target *ti, u8 **root_digest, unsigned i static struct target_type verity_target = { .name = "verity", .features = DM_TARGET_IMMUTABLE, - .version = {1, 9, 0}, + .version = {1, 10, 0}, .module = THIS_MODULE, .ctr = verity_ctr, .dtr = verity_dtr, diff --git a/drivers/md/dm-verity.h b/drivers/md/dm-verity.h index f9d522c870e6..db93a91169d5 100644 --- a/drivers/md/dm-verity.h +++ b/drivers/md/dm-verity.h @@ -11,6 +11,7 @@ #ifndef DM_VERITY_H #define DM_VERITY_H +#include <linux/dm-io.h> #include <linux/dm-bufio.h> #include <linux/device-mapper.h> #include <linux/interrupt.h> @@ -68,6 +69,9 @@ struct dm_verity { unsigned long *validated_blocks; /* bitset blocks validated */ char *signature_key_desc; /* signature keyring reference */ + + struct dm_io_client *io; + mempool_t recheck_pool; }; struct dm_verity_io { @@ -76,14 +80,15 @@ struct dm_verity_io { /* original value of bio->bi_end_io */ bio_end_io_t *orig_bi_end_io; + struct bvec_iter iter; + sector_t block; unsigned int n_blocks; bool in_tasklet; - struct bvec_iter iter; - struct work_struct work; - struct tasklet_struct tasklet; + + char *recheck_buffer; /* * Three variably-size fields follow this struct: diff --git a/drivers/md/dm-writecache.c b/drivers/md/dm-writecache.c index 074cb785eafc..b463c28c39ad 100644 --- a/drivers/md/dm-writecache.c +++ b/drivers/md/dm-writecache.c @@ -299,7 +299,7 @@ static int persistent_memory_claim(struct dm_writecache *wc) long i; wc->memory_map = NULL; - pages = kvmalloc_array(p, sizeof(struct page *), GFP_KERNEL); + pages = vmalloc_array(p, sizeof(struct page *)); if (!pages) { r = -ENOMEM; goto err2; @@ -330,7 +330,7 @@ static int persistent_memory_claim(struct dm_writecache *wc) r = -ENOMEM; goto err3; } - kvfree(pages); + vfree(pages); wc->memory_vmapped = true; } @@ -341,7 +341,7 @@ static int persistent_memory_claim(struct dm_writecache *wc) return 0; err3: - kvfree(pages); + vfree(pages); err2: dax_read_unlock(id); err1: @@ -962,7 +962,7 @@ static int writecache_alloc_entries(struct dm_writecache *wc) if (wc->entries) return 0; - wc->entries = vmalloc(array_size(sizeof(struct wc_entry), wc->n_blocks)); + wc->entries = vmalloc_array(wc->n_blocks, sizeof(struct wc_entry)); if (!wc->entries) return -ENOMEM; for (b = 0; b < wc->n_blocks; b++) { diff --git a/drivers/md/dm-zoned-metadata.c b/drivers/md/dm-zoned-metadata.c index 60a4dc01ea18..fdfe30f7b697 100644 --- a/drivers/md/dm-zoned-metadata.c +++ b/drivers/md/dm-zoned-metadata.c @@ -2836,12 +2836,11 @@ static void dmz_print_dev(struct dmz_metadata *zmd, int num) { struct dmz_dev *dev = &zmd->dev[num]; - if (bdev_zoned_model(dev->bdev) == BLK_ZONED_NONE) + if (!bdev_is_zoned(dev->bdev)) dmz_dev_info(dev, "Regular block device"); else - dmz_dev_info(dev, "Host-%s zoned block device", - bdev_zoned_model(dev->bdev) == BLK_ZONED_HA ? - "aware" : "managed"); + dmz_dev_info(dev, "Host-managed zoned block device"); + if (zmd->sb_version > 1) { sector_t sector_offset = dev->zone_offset << zmd->zone_nr_sectors_shift; diff --git a/drivers/md/dm-zoned-target.c b/drivers/md/dm-zoned-target.c index b487f7acc860..621794a9edd6 100644 --- a/drivers/md/dm-zoned-target.c +++ b/drivers/md/dm-zoned-target.c @@ -702,7 +702,7 @@ static int dmz_get_zoned_device(struct dm_target *ti, char *path, } bdev = ddev->bdev; - if (bdev_zoned_model(bdev) == BLK_ZONED_NONE) { + if (!bdev_is_zoned(bdev)) { if (nr_devs == 1) { ti->error = "Invalid regular device"; goto err; @@ -1010,7 +1010,7 @@ static void dmz_io_hints(struct dm_target *ti, struct queue_limits *limits) limits->max_sectors = chunk_sectors; /* We are exposing a drive-managed zoned block device */ - limits->zoned = BLK_ZONED_NONE; + limits->zoned = false; } /* diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 23c32cd1f1d8..8dcabf84d866 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -2675,7 +2675,7 @@ static int lock_fs(struct mapped_device *md) WARN_ON(test_bit(DMF_FROZEN, &md->flags)); - r = freeze_bdev(md->disk->part0); + r = bdev_freeze(md->disk->part0); if (!r) set_bit(DMF_FROZEN, &md->flags); return r; @@ -2685,7 +2685,7 @@ static void unlock_fs(struct mapped_device *md) { if (!test_bit(DMF_FROZEN, &md->flags)) return; - thaw_bdev(md->disk->part0); + bdev_thaw(md->disk->part0); clear_bit(DMF_FROZEN, &md->flags); } diff --git a/drivers/md/md-autodetect.c b/drivers/md/md-autodetect.c index 4b80165afd23..b2a00f213c2c 100644 --- a/drivers/md/md-autodetect.c +++ b/drivers/md/md-autodetect.c @@ -49,7 +49,6 @@ static int md_setup_ents __initdata; * instead of just one. -- KTK * 18May2000: Added support for persistent-superblock arrays: * md=n,0,factor,fault,device-list uses RAID0 for device n - * md=n,-1,factor,fault,device-list uses LINEAR for device n * md=n,device-list reads a RAID superblock from the devices * elements in device-list are read by name_to_kdev_t so can be * a hex number or something like /dev/hda1 /dev/sdb @@ -88,7 +87,7 @@ static int __init md_setup(char *str) md_setup_ents++; switch (get_option(&str, &level)) { /* RAID level */ case 2: /* could be 0 or -1.. */ - if (level == 0 || level == LEVEL_LINEAR) { + if (level == 0) { if (get_option(&str, &factor) != 2 || /* Chunk Size */ get_option(&str, &fault) != 2) { printk(KERN_WARNING "md: Too few arguments supplied to md=.\n"); @@ -96,10 +95,7 @@ static int __init md_setup(char *str) } md_setup_args[ent].level = level; md_setup_args[ent].chunk = 1 << (factor+12); - if (level == LEVEL_LINEAR) - pername = "linear"; - else - pername = "raid0"; + pername = "raid0"; break; } fallthrough; diff --git a/drivers/md/md-faulty.c b/drivers/md/md-faulty.c deleted file mode 100644 index a039e8e20f55..000000000000 --- a/drivers/md/md-faulty.c +++ /dev/null @@ -1,365 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* - * faulty.c : Multiple Devices driver for Linux - * - * Copyright (C) 2004 Neil Brown - * - * fautly-device-simulator personality for md - */ - - -/* - * The "faulty" personality causes some requests to fail. - * - * Possible failure modes are: - * reads fail "randomly" but succeed on retry - * writes fail "randomly" but succeed on retry - * reads for some address fail and then persist until a write - * reads for some address fail and then persist irrespective of write - * writes for some address fail and persist - * all writes fail - * - * Different modes can be active at a time, but only - * one can be set at array creation. Others can be added later. - * A mode can be one-shot or recurrent with the recurrence being - * once in every N requests. - * The bottom 5 bits of the "layout" indicate the mode. The - * remainder indicate a period, or 0 for one-shot. - * - * There is an implementation limit on the number of concurrently - * persisting-faulty blocks. When a new fault is requested that would - * exceed the limit, it is ignored. - * All current faults can be clear using a layout of "0". - * - * Requests are always sent to the device. If they are to fail, - * we clone the bio and insert a new b_end_io into the chain. - */ - -#define WriteTransient 0 -#define ReadTransient 1 -#define WritePersistent 2 -#define ReadPersistent 3 -#define WriteAll 4 /* doesn't go to device */ -#define ReadFixable 5 -#define Modes 6 - -#define ClearErrors 31 -#define ClearFaults 30 - -#define AllPersist 100 /* internal use only */ -#define NoPersist 101 - -#define ModeMask 0x1f -#define ModeShift 5 - -#define MaxFault 50 -#include <linux/blkdev.h> -#include <linux/module.h> -#include <linux/raid/md_u.h> -#include <linux/slab.h> -#include "md.h" -#include <linux/seq_file.h> - - -static void faulty_fail(struct bio *bio) -{ - struct bio *b = bio->bi_private; - - b->bi_iter.bi_size = bio->bi_iter.bi_size; - b->bi_iter.bi_sector = bio->bi_iter.bi_sector; - - bio_put(bio); - - bio_io_error(b); -} - -struct faulty_conf { - int period[Modes]; - atomic_t counters[Modes]; - sector_t faults[MaxFault]; - int modes[MaxFault]; - int nfaults; - struct md_rdev *rdev; -}; - -static int check_mode(struct faulty_conf *conf, int mode) -{ - if (conf->period[mode] == 0 && - atomic_read(&conf->counters[mode]) <= 0) - return 0; /* no failure, no decrement */ - - - if (atomic_dec_and_test(&conf->counters[mode])) { - if (conf->period[mode]) - atomic_set(&conf->counters[mode], conf->period[mode]); - return 1; - } - return 0; -} - -static int check_sector(struct faulty_conf *conf, sector_t start, sector_t end, int dir) -{ - /* If we find a ReadFixable sector, we fix it ... */ - int i; - for (i=0; i<conf->nfaults; i++) - if (conf->faults[i] >= start && - conf->faults[i] < end) { - /* found it ... */ - switch (conf->modes[i] * 2 + dir) { - case WritePersistent*2+WRITE: return 1; - case ReadPersistent*2+READ: return 1; - case ReadFixable*2+READ: return 1; - case ReadFixable*2+WRITE: - conf->modes[i] = NoPersist; - return 0; - case AllPersist*2+READ: - case AllPersist*2+WRITE: return 1; - default: - return 0; - } - } - return 0; -} - -static void add_sector(struct faulty_conf *conf, sector_t start, int mode) -{ - int i; - int n = conf->nfaults; - for (i=0; i<conf->nfaults; i++) - if (conf->faults[i] == start) { - switch(mode) { - case NoPersist: conf->modes[i] = mode; return; - case WritePersistent: - if (conf->modes[i] == ReadPersistent || - conf->modes[i] == ReadFixable) - conf->modes[i] = AllPersist; - else - conf->modes[i] = WritePersistent; - return; - case ReadPersistent: - if (conf->modes[i] == WritePersistent) - conf->modes[i] = AllPersist; - else - conf->modes[i] = ReadPersistent; - return; - case ReadFixable: - if (conf->modes[i] == WritePersistent || - conf->modes[i] == ReadPersistent) - conf->modes[i] = AllPersist; - else - conf->modes[i] = ReadFixable; - return; - } - } else if (conf->modes[i] == NoPersist) - n = i; - - if (n >= MaxFault) - return; - conf->faults[n] = start; - conf->modes[n] = mode; - if (conf->nfaults == n) - conf->nfaults = n+1; -} - -static bool faulty_make_request(struct mddev *mddev, struct bio *bio) -{ - struct faulty_conf *conf = mddev->private; - int failit = 0; - - if (bio_data_dir(bio) == WRITE) { - /* write request */ - if (atomic_read(&conf->counters[WriteAll])) { - /* special case - don't decrement, don't submit_bio_noacct, - * just fail immediately - */ - bio_io_error(bio); - return true; - } - - if (check_sector(conf, bio->bi_iter.bi_sector, - bio_end_sector(bio), WRITE)) - failit = 1; - if (check_mode(conf, WritePersistent)) { - add_sector(conf, bio->bi_iter.bi_sector, - WritePersistent); - failit = 1; - } - if (check_mode(conf, WriteTransient)) - failit = 1; - } else { - /* read request */ - if (check_sector(conf, bio->bi_iter.bi_sector, - bio_end_sector(bio), READ)) - failit = 1; - if (check_mode(conf, ReadTransient)) - failit = 1; - if (check_mode(conf, ReadPersistent)) { - add_sector(conf, bio->bi_iter.bi_sector, - ReadPersistent); - failit = 1; - } - if (check_mode(conf, ReadFixable)) { - add_sector(conf, bio->bi_iter.bi_sector, - ReadFixable); - failit = 1; - } - } - - md_account_bio(mddev, &bio); - if (failit) { - struct bio *b = bio_alloc_clone(conf->rdev->bdev, bio, GFP_NOIO, - &mddev->bio_set); - - b->bi_private = bio; - b->bi_end_io = faulty_fail; - bio = b; - } else - bio_set_dev(bio, conf->rdev->bdev); - - submit_bio_noacct(bio); - return true; -} - -static void faulty_status(struct seq_file *seq, struct mddev *mddev) -{ - struct faulty_conf *conf = mddev->private; - int n; - - if ((n=atomic_read(&conf->counters[WriteTransient])) != 0) - seq_printf(seq, " WriteTransient=%d(%d)", - n, conf->period[WriteTransient]); - - if ((n=atomic_read(&conf->counters[ReadTransient])) != 0) - seq_printf(seq, " ReadTransient=%d(%d)", - n, conf->period[ReadTransient]); - - if ((n=atomic_read(&conf->counters[WritePersistent])) != 0) - seq_printf(seq, " WritePersistent=%d(%d)", - n, conf->period[WritePersistent]); - - if ((n=atomic_read(&conf->counters[ReadPersistent])) != 0) - seq_printf(seq, " ReadPersistent=%d(%d)", - n, conf->period[ReadPersistent]); - - - if ((n=atomic_read(&conf->counters[ReadFixable])) != 0) - seq_printf(seq, " ReadFixable=%d(%d)", - n, conf->period[ReadFixable]); - - if ((n=atomic_read(&conf->counters[WriteAll])) != 0) - seq_printf(seq, " WriteAll"); - - seq_printf(seq, " nfaults=%d", conf->nfaults); -} - - -static int faulty_reshape(struct mddev *mddev) -{ - int mode = mddev->new_layout & ModeMask; - int count = mddev->new_layout >> ModeShift; - struct faulty_conf *conf = mddev->private; - - if (mddev->new_layout < 0) - return 0; - - /* new layout */ - if (mode == ClearFaults) - conf->nfaults = 0; - else if (mode == ClearErrors) { - int i; - for (i=0 ; i < Modes ; i++) { - conf->period[i] = 0; - atomic_set(&conf->counters[i], 0); - } - } else if (mode < Modes) { - conf->period[mode] = count; - if (!count) count++; - atomic_set(&conf->counters[mode], count); - } else - return -EINVAL; - mddev->new_layout = -1; - mddev->layout = -1; /* makes sure further changes come through */ - return 0; -} - -static sector_t faulty_size(struct mddev *mddev, sector_t sectors, int raid_disks) -{ - WARN_ONCE(raid_disks, - "%s does not support generic reshape\n", __func__); - - if (sectors == 0) - return mddev->dev_sectors; - - return sectors; -} - -static int faulty_run(struct mddev *mddev) -{ - struct md_rdev *rdev; - int i; - struct faulty_conf *conf; - - if (md_check_no_bitmap(mddev)) - return -EINVAL; - - conf = kmalloc(sizeof(*conf), GFP_KERNEL); - if (!conf) - return -ENOMEM; - - for (i=0; i<Modes; i++) { - atomic_set(&conf->counters[i], 0); - conf->period[i] = 0; - } - conf->nfaults = 0; - - rdev_for_each(rdev, mddev) { - conf->rdev = rdev; - disk_stack_limits(mddev->gendisk, rdev->bdev, - rdev->data_offset << 9); - } - - md_set_array_sectors(mddev, faulty_size(mddev, 0, 0)); - mddev->private = conf; - - faulty_reshape(mddev); - - return 0; -} - -static void faulty_free(struct mddev *mddev, void *priv) -{ - struct faulty_conf *conf = priv; - - kfree(conf); -} - -static struct md_personality faulty_personality = -{ - .name = "faulty", - .level = LEVEL_FAULTY, - .owner = THIS_MODULE, - .make_request = faulty_make_request, - .run = faulty_run, - .free = faulty_free, - .status = faulty_status, - .check_reshape = faulty_reshape, - .size = faulty_size, -}; - -static int __init raid_init(void) -{ - return register_md_personality(&faulty_personality); -} - -static void raid_exit(void) -{ - unregister_md_personality(&faulty_personality); -} - -module_init(raid_init); -module_exit(raid_exit); -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("Fault injection personality for MD (deprecated)"); -MODULE_ALIAS("md-personality-10"); /* faulty */ -MODULE_ALIAS("md-faulty"); -MODULE_ALIAS("md-level--5"); diff --git a/drivers/md/md-linear.c b/drivers/md/md-linear.c deleted file mode 100644 index 8eca7693b793..000000000000 --- a/drivers/md/md-linear.c +++ /dev/null @@ -1,318 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* - linear.c : Multiple Devices driver for Linux - Copyright (C) 1994-96 Marc ZYNGIER - <zyngier@ufr-info-p7.ibp.fr> or - <maz@gloups.fdn.fr> - - Linear mode management functions. - -*/ - -#include <linux/blkdev.h> -#include <linux/raid/md_u.h> -#include <linux/seq_file.h> -#include <linux/module.h> -#include <linux/slab.h> -#include <trace/events/block.h> -#include "md.h" -#include "md-linear.h" - -/* - * find which device holds a particular offset - */ -static inline struct dev_info *which_dev(struct mddev *mddev, sector_t sector) -{ - int lo, mid, hi; - struct linear_conf *conf; - - lo = 0; - hi = mddev->raid_disks - 1; - conf = mddev->private; - - /* - * Binary Search - */ - - while (hi > lo) { - - mid = (hi + lo) / 2; - if (sector < conf->disks[mid].end_sector) - hi = mid; - else - lo = mid + 1; - } - - return conf->disks + lo; -} - -static sector_t linear_size(struct mddev *mddev, sector_t sectors, int raid_disks) -{ - struct linear_conf *conf; - sector_t array_sectors; - - conf = mddev->private; - WARN_ONCE(sectors || raid_disks, - "%s does not support generic reshape\n", __func__); - array_sectors = conf->array_sectors; - - return array_sectors; -} - -static struct linear_conf *linear_conf(struct mddev *mddev, int raid_disks) -{ - struct linear_conf *conf; - struct md_rdev *rdev; - int i, cnt; - - conf = kzalloc(struct_size(conf, disks, raid_disks), GFP_KERNEL); - if (!conf) - return NULL; - - /* - * conf->raid_disks is copy of mddev->raid_disks. The reason to - * keep a copy of mddev->raid_disks in struct linear_conf is, - * mddev->raid_disks may not be consistent with pointers number of - * conf->disks[] when it is updated in linear_add() and used to - * iterate old conf->disks[] earray in linear_congested(). - * Here conf->raid_disks is always consitent with number of - * pointers in conf->disks[] array, and mddev->private is updated - * with rcu_assign_pointer() in linear_addr(), such race can be - * avoided. - */ - conf->raid_disks = raid_disks; - - cnt = 0; - conf->array_sectors = 0; - - rdev_for_each(rdev, mddev) { - int j = rdev->raid_disk; - struct dev_info *disk = conf->disks + j; - sector_t sectors; - - if (j < 0 || j >= raid_disks || disk->rdev) { - pr_warn("md/linear:%s: disk numbering problem. Aborting!\n", - mdname(mddev)); - goto out; - } - - disk->rdev = rdev; - if (mddev->chunk_sectors) { - sectors = rdev->sectors; - sector_div(sectors, mddev->chunk_sectors); - rdev->sectors = sectors * mddev->chunk_sectors; - } - - disk_stack_limits(mddev->gendisk, rdev->bdev, - rdev->data_offset << 9); - - conf->array_sectors += rdev->sectors; - cnt++; - } - if (cnt != raid_disks) { - pr_warn("md/linear:%s: not enough drives present. Aborting!\n", - mdname(mddev)); - goto out; - } - - /* - * Here we calculate the device offsets. - */ - conf->disks[0].end_sector = conf->disks[0].rdev->sectors; - - for (i = 1; i < raid_disks; i++) - conf->disks[i].end_sector = - conf->disks[i-1].end_sector + - conf->disks[i].rdev->sectors; - - return conf; - -out: - kfree(conf); - return NULL; -} - -static int linear_run (struct mddev *mddev) -{ - struct linear_conf *conf; - int ret; - - if (md_check_no_bitmap(mddev)) - return -EINVAL; - conf = linear_conf(mddev, mddev->raid_disks); - - if (!conf) - return 1; - mddev->private = conf; - md_set_array_sectors(mddev, linear_size(mddev, 0, 0)); - - ret = md_integrity_register(mddev); - if (ret) { - kfree(conf); - mddev->private = NULL; - } - return ret; -} - -static int linear_add(struct mddev *mddev, struct md_rdev *rdev) -{ - /* Adding a drive to a linear array allows the array to grow. - * It is permitted if the new drive has a matching superblock - * already on it, with raid_disk equal to raid_disks. - * It is achieved by creating a new linear_private_data structure - * and swapping it in in-place of the current one. - * The current one is never freed until the array is stopped. - * This avoids races. - */ - struct linear_conf *newconf, *oldconf; - - if (rdev->saved_raid_disk != mddev->raid_disks) - return -EINVAL; - - rdev->raid_disk = rdev->saved_raid_disk; - rdev->saved_raid_disk = -1; - - newconf = linear_conf(mddev,mddev->raid_disks+1); - - if (!newconf) - return -ENOMEM; - - /* newconf->raid_disks already keeps a copy of * the increased - * value of mddev->raid_disks, WARN_ONCE() is just used to make - * sure of this. It is possible that oldconf is still referenced - * in linear_congested(), therefore kfree_rcu() is used to free - * oldconf until no one uses it anymore. - */ - oldconf = rcu_dereference_protected(mddev->private, - lockdep_is_held(&mddev->reconfig_mutex)); - mddev->raid_disks++; - WARN_ONCE(mddev->raid_disks != newconf->raid_disks, - "copied raid_disks doesn't match mddev->raid_disks"); - rcu_assign_pointer(mddev->private, newconf); - md_set_array_sectors(mddev, linear_size(mddev, 0, 0)); - set_capacity_and_notify(mddev->gendisk, mddev->array_sectors); - kfree_rcu(oldconf, rcu); - return 0; -} - -static void linear_free(struct mddev *mddev, void *priv) -{ - struct linear_conf *conf = priv; - - kfree(conf); -} - -static bool linear_make_request(struct mddev *mddev, struct bio *bio) -{ - struct dev_info *tmp_dev; - sector_t start_sector, end_sector, data_offset; - sector_t bio_sector = bio->bi_iter.bi_sector; - - if (unlikely(bio->bi_opf & REQ_PREFLUSH) - && md_flush_request(mddev, bio)) - return true; - - tmp_dev = which_dev(mddev, bio_sector); - start_sector = tmp_dev->end_sector - tmp_dev->rdev->sectors; - end_sector = tmp_dev->end_sector; - data_offset = tmp_dev->rdev->data_offset; - - if (unlikely(bio_sector >= end_sector || - bio_sector < start_sector)) - goto out_of_bounds; - - if (unlikely(is_rdev_broken(tmp_dev->rdev))) { - md_error(mddev, tmp_dev->rdev); - bio_io_error(bio); - return true; - } - - if (unlikely(bio_end_sector(bio) > end_sector)) { - /* This bio crosses a device boundary, so we have to split it */ - struct bio *split = bio_split(bio, end_sector - bio_sector, - GFP_NOIO, &mddev->bio_set); - bio_chain(split, bio); - submit_bio_noacct(bio); - bio = split; - } - - md_account_bio(mddev, &bio); - bio_set_dev(bio, tmp_dev->rdev->bdev); - bio->bi_iter.bi_sector = bio->bi_iter.bi_sector - - start_sector + data_offset; - - if (unlikely((bio_op(bio) == REQ_OP_DISCARD) && - !bdev_max_discard_sectors(bio->bi_bdev))) { - /* Just ignore it */ - bio_endio(bio); - } else { - if (mddev->gendisk) - trace_block_bio_remap(bio, disk_devt(mddev->gendisk), - bio_sector); - mddev_check_write_zeroes(mddev, bio); - submit_bio_noacct(bio); - } - return true; - -out_of_bounds: - pr_err("md/linear:%s: make_request: Sector %llu out of bounds on dev %pg: %llu sectors, offset %llu\n", - mdname(mddev), - (unsigned long long)bio->bi_iter.bi_sector, - tmp_dev->rdev->bdev, - (unsigned long long)tmp_dev->rdev->sectors, - (unsigned long long)start_sector); - bio_io_error(bio); - return true; -} - -static void linear_status (struct seq_file *seq, struct mddev *mddev) -{ - seq_printf(seq, " %dk rounding", mddev->chunk_sectors / 2); -} - -static void linear_error(struct mddev *mddev, struct md_rdev *rdev) -{ - if (!test_and_set_bit(MD_BROKEN, &mddev->flags)) { - char *md_name = mdname(mddev); - - pr_crit("md/linear%s: Disk failure on %pg detected, failing array.\n", - md_name, rdev->bdev); - } -} - -static void linear_quiesce(struct mddev *mddev, int state) -{ -} - -static struct md_personality linear_personality = -{ - .name = "linear", - .level = LEVEL_LINEAR, - .owner = THIS_MODULE, - .make_request = linear_make_request, - .run = linear_run, - .free = linear_free, - .status = linear_status, - .hot_add_disk = linear_add, - .size = linear_size, - .quiesce = linear_quiesce, - .error_handler = linear_error, -}; - -static int __init linear_init (void) -{ - return register_md_personality (&linear_personality); -} - -static void linear_exit (void) -{ - unregister_md_personality (&linear_personality); -} - -module_init(linear_init); -module_exit(linear_exit); -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("Linear device concatenation personality for MD (deprecated)"); -MODULE_ALIAS("md-personality-1"); /* LINEAR - deprecated*/ -MODULE_ALIAS("md-linear"); -MODULE_ALIAS("md-level--1"); diff --git a/drivers/md/md-multipath.c b/drivers/md/md-multipath.c deleted file mode 100644 index d22276870283..000000000000 --- a/drivers/md/md-multipath.c +++ /dev/null @@ -1,471 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* - * multipath.c : Multiple Devices driver for Linux - * - * Copyright (C) 1999, 2000, 2001 Ingo Molnar, Red Hat - * - * Copyright (C) 1996, 1997, 1998 Ingo Molnar, Miguel de Icaza, Gadi Oxman - * - * MULTIPATH management functions. - * - * derived from raid1.c. - */ - -#include <linux/blkdev.h> -#include <linux/module.h> -#include <linux/raid/md_u.h> -#include <linux/seq_file.h> -#include <linux/slab.h> -#include "md.h" -#include "md-multipath.h" - -#define MAX_WORK_PER_DISK 128 - -#define NR_RESERVED_BUFS 32 - -static int multipath_map (struct mpconf *conf) -{ - int i, disks = conf->raid_disks; - - /* - * Later we do read balancing on the read side - * now we use the first available disk. - */ - - rcu_read_lock(); - for (i = 0; i < disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->multipaths[i].rdev); - if (rdev && test_bit(In_sync, &rdev->flags) && - !test_bit(Faulty, &rdev->flags)) { - atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); - return i; - } - } - rcu_read_unlock(); - - pr_crit_ratelimited("multipath_map(): no more operational IO paths?\n"); - return (-1); -} - -static void multipath_reschedule_retry (struct multipath_bh *mp_bh) -{ - unsigned long flags; - struct mddev *mddev = mp_bh->mddev; - struct mpconf *conf = mddev->private; - - spin_lock_irqsave(&conf->device_lock, flags); - list_add(&mp_bh->retry_list, &conf->retry_list); - spin_unlock_irqrestore(&conf->device_lock, flags); - md_wakeup_thread(mddev->thread); -} - -/* - * multipath_end_bh_io() is called when we have finished servicing a multipathed - * operation and are ready to return a success/failure code to the buffer - * cache layer. - */ -static void multipath_end_bh_io(struct multipath_bh *mp_bh, blk_status_t status) -{ - struct bio *bio = mp_bh->master_bio; - struct mpconf *conf = mp_bh->mddev->private; - - bio->bi_status = status; - bio_endio(bio); - mempool_free(mp_bh, &conf->pool); -} - -static void multipath_end_request(struct bio *bio) -{ - struct multipath_bh *mp_bh = bio->bi_private; - struct mpconf *conf = mp_bh->mddev->private; - struct md_rdev *rdev = conf->multipaths[mp_bh->path].rdev; - - if (!bio->bi_status) - multipath_end_bh_io(mp_bh, 0); - else if (!(bio->bi_opf & REQ_RAHEAD)) { - /* - * oops, IO error: - */ - md_error (mp_bh->mddev, rdev); - pr_info("multipath: %pg: rescheduling sector %llu\n", - rdev->bdev, - (unsigned long long)bio->bi_iter.bi_sector); - multipath_reschedule_retry(mp_bh); - } else - multipath_end_bh_io(mp_bh, bio->bi_status); - rdev_dec_pending(rdev, conf->mddev); -} - -static bool multipath_make_request(struct mddev *mddev, struct bio * bio) -{ - struct mpconf *conf = mddev->private; - struct multipath_bh * mp_bh; - struct multipath_info *multipath; - - if (unlikely(bio->bi_opf & REQ_PREFLUSH) - && md_flush_request(mddev, bio)) - return true; - - md_account_bio(mddev, &bio); - mp_bh = mempool_alloc(&conf->pool, GFP_NOIO); - - mp_bh->master_bio = bio; - mp_bh->mddev = mddev; - - mp_bh->path = multipath_map(conf); - if (mp_bh->path < 0) { - bio_io_error(bio); - mempool_free(mp_bh, &conf->pool); - return true; - } - multipath = conf->multipaths + mp_bh->path; - - bio_init_clone(multipath->rdev->bdev, &mp_bh->bio, bio, GFP_NOIO); - - mp_bh->bio.bi_iter.bi_sector += multipath->rdev->data_offset; - mp_bh->bio.bi_opf |= REQ_FAILFAST_TRANSPORT; - mp_bh->bio.bi_end_io = multipath_end_request; - mp_bh->bio.bi_private = mp_bh; - mddev_check_write_zeroes(mddev, &mp_bh->bio); - submit_bio_noacct(&mp_bh->bio); - return true; -} - -static void multipath_status(struct seq_file *seq, struct mddev *mddev) -{ - struct mpconf *conf = mddev->private; - int i; - - seq_printf (seq, " [%d/%d] [", conf->raid_disks, - conf->raid_disks - mddev->degraded); - rcu_read_lock(); - for (i = 0; i < conf->raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->multipaths[i].rdev); - seq_printf (seq, "%s", rdev && test_bit(In_sync, &rdev->flags) ? "U" : "_"); - } - rcu_read_unlock(); - seq_putc(seq, ']'); -} - -/* - * Careful, this can execute in IRQ contexts as well! - */ -static void multipath_error (struct mddev *mddev, struct md_rdev *rdev) -{ - struct mpconf *conf = mddev->private; - - if (conf->raid_disks - mddev->degraded <= 1) { - /* - * Uh oh, we can do nothing if this is our last path, but - * first check if this is a queued request for a device - * which has just failed. - */ - pr_warn("multipath: only one IO path left and IO error.\n"); - /* leave it active... it's all we have */ - return; - } - /* - * Mark disk as unusable - */ - if (test_and_clear_bit(In_sync, &rdev->flags)) { - unsigned long flags; - spin_lock_irqsave(&conf->device_lock, flags); - mddev->degraded++; - spin_unlock_irqrestore(&conf->device_lock, flags); - } - set_bit(Faulty, &rdev->flags); - set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags); - pr_err("multipath: IO failure on %pg, disabling IO path.\n" - "multipath: Operation continuing on %d IO paths.\n", - rdev->bdev, - conf->raid_disks - mddev->degraded); -} - -static void print_multipath_conf (struct mpconf *conf) -{ - int i; - struct multipath_info *tmp; - - pr_debug("MULTIPATH conf printout:\n"); - if (!conf) { - pr_debug("(conf==NULL)\n"); - return; - } - pr_debug(" --- wd:%d rd:%d\n", conf->raid_disks - conf->mddev->degraded, - conf->raid_disks); - - for (i = 0; i < conf->raid_disks; i++) { - tmp = conf->multipaths + i; - if (tmp->rdev) - pr_debug(" disk%d, o:%d, dev:%pg\n", - i,!test_bit(Faulty, &tmp->rdev->flags), - tmp->rdev->bdev); - } -} - -static int multipath_add_disk(struct mddev *mddev, struct md_rdev *rdev) -{ - struct mpconf *conf = mddev->private; - int err = -EEXIST; - int path; - struct multipath_info *p; - int first = 0; - int last = mddev->raid_disks - 1; - - if (rdev->raid_disk >= 0) - first = last = rdev->raid_disk; - - print_multipath_conf(conf); - - for (path = first; path <= last; path++) - if ((p=conf->multipaths+path)->rdev == NULL) { - disk_stack_limits(mddev->gendisk, rdev->bdev, - rdev->data_offset << 9); - - err = md_integrity_add_rdev(rdev, mddev); - if (err) - break; - spin_lock_irq(&conf->device_lock); - mddev->degraded--; - rdev->raid_disk = path; - set_bit(In_sync, &rdev->flags); - spin_unlock_irq(&conf->device_lock); - rcu_assign_pointer(p->rdev, rdev); - err = 0; - break; - } - - print_multipath_conf(conf); - - return err; -} - -static int multipath_remove_disk(struct mddev *mddev, struct md_rdev *rdev) -{ - struct mpconf *conf = mddev->private; - int err = 0; - int number = rdev->raid_disk; - struct multipath_info *p = conf->multipaths + number; - - print_multipath_conf(conf); - - if (rdev == p->rdev) { - if (test_bit(In_sync, &rdev->flags) || - atomic_read(&rdev->nr_pending)) { - pr_warn("hot-remove-disk, slot %d is identified but is still operational!\n", number); - err = -EBUSY; - goto abort; - } - p->rdev = NULL; - if (!test_bit(RemoveSynchronized, &rdev->flags)) { - synchronize_rcu(); - if (atomic_read(&rdev->nr_pending)) { - /* lost the race, try later */ - err = -EBUSY; - p->rdev = rdev; - goto abort; - } - } - err = md_integrity_register(mddev); - } -abort: - - print_multipath_conf(conf); - return err; -} - -/* - * This is a kernel thread which: - * - * 1. Retries failed read operations on working multipaths. - * 2. Updates the raid superblock when problems encounter. - * 3. Performs writes following reads for array syncronising. - */ - -static void multipathd(struct md_thread *thread) -{ - struct mddev *mddev = thread->mddev; - struct multipath_bh *mp_bh; - struct bio *bio; - unsigned long flags; - struct mpconf *conf = mddev->private; - struct list_head *head = &conf->retry_list; - - md_check_recovery(mddev); - for (;;) { - spin_lock_irqsave(&conf->device_lock, flags); - if (list_empty(head)) - break; - mp_bh = list_entry(head->prev, struct multipath_bh, retry_list); - list_del(head->prev); - spin_unlock_irqrestore(&conf->device_lock, flags); - - bio = &mp_bh->bio; - bio->bi_iter.bi_sector = mp_bh->master_bio->bi_iter.bi_sector; - - if ((mp_bh->path = multipath_map (conf))<0) { - pr_err("multipath: %pg: unrecoverable IO read error for block %llu\n", - bio->bi_bdev, - (unsigned long long)bio->bi_iter.bi_sector); - multipath_end_bh_io(mp_bh, BLK_STS_IOERR); - } else { - pr_err("multipath: %pg: redirecting sector %llu to another IO path\n", - bio->bi_bdev, - (unsigned long long)bio->bi_iter.bi_sector); - *bio = *(mp_bh->master_bio); - bio->bi_iter.bi_sector += - conf->multipaths[mp_bh->path].rdev->data_offset; - bio_set_dev(bio, conf->multipaths[mp_bh->path].rdev->bdev); - bio->bi_opf |= REQ_FAILFAST_TRANSPORT; - bio->bi_end_io = multipath_end_request; - bio->bi_private = mp_bh; - submit_bio_noacct(bio); - } - } - spin_unlock_irqrestore(&conf->device_lock, flags); -} - -static sector_t multipath_size(struct mddev *mddev, sector_t sectors, int raid_disks) -{ - WARN_ONCE(sectors || raid_disks, - "%s does not support generic reshape\n", __func__); - - return mddev->dev_sectors; -} - -static int multipath_run (struct mddev *mddev) -{ - struct mpconf *conf; - int disk_idx; - struct multipath_info *disk; - struct md_rdev *rdev; - int working_disks; - int ret; - - if (md_check_no_bitmap(mddev)) - return -EINVAL; - - if (mddev->level != LEVEL_MULTIPATH) { - pr_warn("multipath: %s: raid level not set to multipath IO (%d)\n", - mdname(mddev), mddev->level); - goto out; - } - /* - * copy the already verified devices into our private MULTIPATH - * bookkeeping area. [whatever we allocate in multipath_run(), - * should be freed in multipath_free()] - */ - - conf = kzalloc(sizeof(struct mpconf), GFP_KERNEL); - mddev->private = conf; - if (!conf) - goto out; - - conf->multipaths = kcalloc(mddev->raid_disks, - sizeof(struct multipath_info), - GFP_KERNEL); - if (!conf->multipaths) - goto out_free_conf; - - working_disks = 0; - rdev_for_each(rdev, mddev) { - disk_idx = rdev->raid_disk; - if (disk_idx < 0 || - disk_idx >= mddev->raid_disks) - continue; - - disk = conf->multipaths + disk_idx; - disk->rdev = rdev; - disk_stack_limits(mddev->gendisk, rdev->bdev, - rdev->data_offset << 9); - - if (!test_bit(Faulty, &rdev->flags)) - working_disks++; - } - - conf->raid_disks = mddev->raid_disks; - conf->mddev = mddev; - spin_lock_init(&conf->device_lock); - INIT_LIST_HEAD(&conf->retry_list); - - if (!working_disks) { - pr_warn("multipath: no operational IO paths for %s\n", - mdname(mddev)); - goto out_free_conf; - } - mddev->degraded = conf->raid_disks - working_disks; - - ret = mempool_init_kmalloc_pool(&conf->pool, NR_RESERVED_BUFS, - sizeof(struct multipath_bh)); - if (ret) - goto out_free_conf; - - rcu_assign_pointer(mddev->thread, - md_register_thread(multipathd, mddev, "multipath")); - if (!mddev->thread) - goto out_free_conf; - - pr_info("multipath: array %s active with %d out of %d IO paths\n", - mdname(mddev), conf->raid_disks - mddev->degraded, - mddev->raid_disks); - /* - * Ok, everything is just fine now - */ - md_set_array_sectors(mddev, multipath_size(mddev, 0, 0)); - - if (md_integrity_register(mddev)) - goto out_free_conf; - - return 0; - -out_free_conf: - mempool_exit(&conf->pool); - kfree(conf->multipaths); - kfree(conf); - mddev->private = NULL; -out: - return -EIO; -} - -static void multipath_free(struct mddev *mddev, void *priv) -{ - struct mpconf *conf = priv; - - mempool_exit(&conf->pool); - kfree(conf->multipaths); - kfree(conf); -} - -static struct md_personality multipath_personality = -{ - .name = "multipath", - .level = LEVEL_MULTIPATH, - .owner = THIS_MODULE, - .make_request = multipath_make_request, - .run = multipath_run, - .free = multipath_free, - .status = multipath_status, - .error_handler = multipath_error, - .hot_add_disk = multipath_add_disk, - .hot_remove_disk= multipath_remove_disk, - .size = multipath_size, -}; - -static int __init multipath_init (void) -{ - return register_md_personality (&multipath_personality); -} - -static void __exit multipath_exit (void) -{ - unregister_md_personality (&multipath_personality); -} - -module_init(multipath_init); -module_exit(multipath_exit); -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("simple multi-path personality for MD (deprecated)"); -MODULE_ALIAS("md-personality-7"); /* MULTIPATH */ -MODULE_ALIAS("md-multipath"); -MODULE_ALIAS("md-level--4"); diff --git a/drivers/md/md.c b/drivers/md/md.c index 9bdd57324c37..9e41a9aaba8b 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -543,6 +543,9 @@ static void md_end_flush(struct bio *bio) rdev_dec_pending(rdev, mddev); if (atomic_dec_and_test(&mddev->flush_pending)) { + /* The pair is percpu_ref_get() from md_flush_request() */ + percpu_ref_put(&mddev->active_io); + /* The pre-request flush has finished */ queue_work(md_wq, &mddev->flush_work); } @@ -562,12 +565,8 @@ static void submit_flushes(struct work_struct *ws) rdev_for_each_rcu(rdev, mddev) if (rdev->raid_disk >= 0 && !test_bit(Faulty, &rdev->flags)) { - /* Take two references, one is dropped - * when request finishes, one after - * we reclaim rcu_read_lock - */ struct bio *bi; - atomic_inc(&rdev->nr_pending); + atomic_inc(&rdev->nr_pending); rcu_read_unlock(); bi = bio_alloc_bioset(rdev->bdev, 0, @@ -578,11 +577,14 @@ static void submit_flushes(struct work_struct *ws) atomic_inc(&mddev->flush_pending); submit_bio(bi); rcu_read_lock(); - rdev_dec_pending(rdev, mddev); } rcu_read_unlock(); - if (atomic_dec_and_test(&mddev->flush_pending)) + if (atomic_dec_and_test(&mddev->flush_pending)) { + /* The pair is percpu_ref_get() from md_flush_request() */ + percpu_ref_put(&mddev->active_io); + queue_work(md_wq, &mddev->flush_work); + } } static void md_submit_flush_data(struct work_struct *ws) @@ -631,6 +633,18 @@ bool md_flush_request(struct mddev *mddev, struct bio *bio) /* new request after previous flush is completed */ if (ktime_after(req_start, mddev->prev_flush_start)) { WARN_ON(mddev->flush_bio); + /* + * Grab a reference to make sure mddev_suspend() will wait for + * this flush to be done. + * + * md_flush_reqeust() is called under md_handle_request() and + * 'active_io' is already grabbed, hence percpu_ref_is_zero() + * won't pass, percpu_ref_tryget_live() can't be used because + * percpu_ref_kill() can be called by mddev_suspend() + * concurrently. + */ + WARN_ON(percpu_ref_is_zero(&mddev->active_io)); + percpu_ref_get(&mddev->active_io); mddev->flush_bio = bio; bio = NULL; } @@ -1027,9 +1041,10 @@ void md_super_write(struct mddev *mddev, struct md_rdev *rdev, return; bio = bio_alloc_bioset(rdev->meta_bdev ? rdev->meta_bdev : rdev->bdev, - 1, - REQ_OP_WRITE | REQ_SYNC | REQ_PREFLUSH | REQ_FUA, - GFP_NOIO, &mddev->sync_set); + 1, + REQ_OP_WRITE | REQ_SYNC | REQ_IDLE | REQ_META + | REQ_PREFLUSH | REQ_FUA, + GFP_NOIO, &mddev->sync_set); atomic_inc(&rdev->nr_pending); @@ -1209,6 +1224,7 @@ struct super_type { struct md_rdev *refdev, int minor_version); int (*validate_super)(struct mddev *mddev, + struct md_rdev *freshest, struct md_rdev *rdev); void (*sync_super)(struct mddev *mddev, struct md_rdev *rdev); @@ -1289,17 +1305,11 @@ static int super_90_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor rdev->sb_size = MD_SB_BYTES; rdev->badblocks.shift = -1; - if (sb->level == LEVEL_MULTIPATH) - rdev->desc_nr = -1; - else - rdev->desc_nr = sb->this_disk.number; - - /* not spare disk, or LEVEL_MULTIPATH */ - if (sb->level == LEVEL_MULTIPATH || - (rdev->desc_nr >= 0 && - rdev->desc_nr < MD_SB_DISKS && - sb->disks[rdev->desc_nr].state & - ((1<<MD_DISK_SYNC) | (1 << MD_DISK_ACTIVE)))) + rdev->desc_nr = sb->this_disk.number; + + /* not spare disk */ + if (rdev->desc_nr >= 0 && rdev->desc_nr < MD_SB_DISKS && + sb->disks[rdev->desc_nr].state & ((1<<MD_DISK_SYNC) | (1 << MD_DISK_ACTIVE))) spare_disk = false; if (!refdev) { @@ -1346,8 +1356,9 @@ static int super_90_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor /* * validate_super for 0.90.0 + * note: we are not using "freshest" for 0.9 superblock */ -static int super_90_validate(struct mddev *mddev, struct md_rdev *rdev) +static int super_90_validate(struct mddev *mddev, struct md_rdev *freshest, struct md_rdev *rdev) { mdp_disk_t *desc; mdp_super_t *sb = page_address(rdev->sb_page); @@ -1445,31 +1456,28 @@ static int super_90_validate(struct mddev *mddev, struct md_rdev *rdev) return 0; } - if (mddev->level != LEVEL_MULTIPATH) { - desc = sb->disks + rdev->desc_nr; + desc = sb->disks + rdev->desc_nr; - if (desc->state & (1<<MD_DISK_FAULTY)) - set_bit(Faulty, &rdev->flags); - else if (desc->state & (1<<MD_DISK_SYNC) /* && - desc->raid_disk < mddev->raid_disks */) { - set_bit(In_sync, &rdev->flags); + if (desc->state & (1<<MD_DISK_FAULTY)) + set_bit(Faulty, &rdev->flags); + else if (desc->state & (1<<MD_DISK_SYNC)) { + set_bit(In_sync, &rdev->flags); + rdev->raid_disk = desc->raid_disk; + rdev->saved_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; - rdev->saved_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); - if (desc->state & (1<<MD_DISK_FAILFAST)) - set_bit(FailFast, &rdev->flags); - } else /* MULTIPATH are always insync */ - set_bit(In_sync, &rdev->flags); + } + if (desc->state & (1<<MD_DISK_WRITEMOSTLY)) + set_bit(WriteMostly, &rdev->flags); + if (desc->state & (1<<MD_DISK_FAILFAST)) + set_bit(FailFast, &rdev->flags); return 0; } @@ -1759,10 +1767,7 @@ static int super_1_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor_ && rdev->new_data_offset < sb_start + (rdev->sb_size/512)) return -EINVAL; - if (sb->level == cpu_to_le32(LEVEL_MULTIPATH)) - rdev->desc_nr = -1; - else - rdev->desc_nr = le32_to_cpu(sb->dev_number); + rdev->desc_nr = le32_to_cpu(sb->dev_number); if (!rdev->bb_page) { rdev->bb_page = alloc_page(GFP_KERNEL); @@ -1815,12 +1820,10 @@ static int super_1_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor_ sb->level != 0) return -EINVAL; - /* not spare disk, or LEVEL_MULTIPATH */ - if (sb->level == cpu_to_le32(LEVEL_MULTIPATH) || - (rdev->desc_nr >= 0 && - rdev->desc_nr < le32_to_cpu(sb->max_dev) && - (le16_to_cpu(sb->dev_roles[rdev->desc_nr]) < MD_DISK_ROLE_MAX || - le16_to_cpu(sb->dev_roles[rdev->desc_nr]) == MD_DISK_ROLE_JOURNAL))) + /* not spare disk */ + if (rdev->desc_nr >= 0 && rdev->desc_nr < le32_to_cpu(sb->max_dev) && + (le16_to_cpu(sb->dev_roles[rdev->desc_nr]) < MD_DISK_ROLE_MAX || + le16_to_cpu(sb->dev_roles[rdev->desc_nr]) == MD_DISK_ROLE_JOURNAL)) spare_disk = false; if (!refdev) { @@ -1859,10 +1862,11 @@ static int super_1_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor_ return ret; } -static int super_1_validate(struct mddev *mddev, struct md_rdev *rdev) +static int super_1_validate(struct mddev *mddev, struct md_rdev *freshest, struct md_rdev *rdev) { struct mdp_superblock_1 *sb = page_address(rdev->sb_page); __u64 ev1 = le64_to_cpu(sb->events); + int role; rdev->raid_disk = -1; clear_bit(Faulty, &rdev->flags); @@ -1955,13 +1959,15 @@ static int super_1_validate(struct mddev *mddev, struct md_rdev *rdev) } } else if (mddev->pers == NULL) { /* Insist of good event counter while assembling, except for - * spares (which don't need an event count) */ - ++ev1; + * spares (which don't need an event count). + * Similar to mdadm, we allow event counter difference of 1 + * from the freshest device. + */ if (rdev->desc_nr >= 0 && rdev->desc_nr < le32_to_cpu(sb->max_dev) && (le16_to_cpu(sb->dev_roles[rdev->desc_nr]) < MD_DISK_ROLE_MAX || le16_to_cpu(sb->dev_roles[rdev->desc_nr]) == MD_DISK_ROLE_JOURNAL)) - if (ev1 < mddev->events) + if (ev1 + 1 < mddev->events) return -EINVAL; } else if (mddev->bitmap) { /* If adding to array with a bitmap, then we can accept an @@ -1976,58 +1982,85 @@ static int super_1_validate(struct mddev *mddev, struct md_rdev *rdev) /* just a hot-add of a new device, leave raid_disk at -1 */ return 0; } - if (mddev->level != LEVEL_MULTIPATH) { - int role; - if (rdev->desc_nr < 0 || - rdev->desc_nr >= le32_to_cpu(sb->max_dev)) { - role = MD_DISK_ROLE_SPARE; - rdev->desc_nr = -1; - } else - role = le16_to_cpu(sb->dev_roles[rdev->desc_nr]); - switch(role) { - case MD_DISK_ROLE_SPARE: /* spare */ - break; - case MD_DISK_ROLE_FAULTY: /* faulty */ - set_bit(Faulty, &rdev->flags); - break; - case MD_DISK_ROLE_JOURNAL: /* journal device */ - if (!(le32_to_cpu(sb->feature_map) & MD_FEATURE_JOURNAL)) { - /* journal device without journal feature */ - pr_warn("md: journal device provided without journal feature, ignoring the device\n"); - return -EINVAL; - } - set_bit(Journal, &rdev->flags); - rdev->journal_tail = le64_to_cpu(sb->journal_tail); - rdev->raid_disk = 0; - break; - default: - rdev->saved_raid_disk = role; - if ((le32_to_cpu(sb->feature_map) & - MD_FEATURE_RECOVERY_OFFSET)) { - rdev->recovery_offset = le64_to_cpu(sb->recovery_offset); - if (!(le32_to_cpu(sb->feature_map) & - MD_FEATURE_RECOVERY_BITMAP)) - rdev->saved_raid_disk = -1; - } else { - /* - * If the array is FROZEN, then the device can't - * be in_sync with rest of array. - */ - if (!test_bit(MD_RECOVERY_FROZEN, - &mddev->recovery)) - set_bit(In_sync, &rdev->flags); - } - rdev->raid_disk = role; - break; + + if (rdev->desc_nr < 0 || + rdev->desc_nr >= le32_to_cpu(sb->max_dev)) { + role = MD_DISK_ROLE_SPARE; + rdev->desc_nr = -1; + } else if (mddev->pers == NULL && freshest && ev1 < mddev->events) { + /* + * If we are assembling, and our event counter is smaller than the + * highest event counter, we cannot trust our superblock about the role. + * It could happen that our rdev was marked as Faulty, and all other + * superblocks were updated with +1 event counter. + * Then, before the next superblock update, which typically happens when + * remove_and_add_spares() removes the device from the array, there was + * a crash or reboot. + * If we allow current rdev without consulting the freshest superblock, + * we could cause data corruption. + * Note that in this case our event counter is smaller by 1 than the + * highest, otherwise, this rdev would not be allowed into array; + * both kernel and mdadm allow event counter difference of 1. + */ + struct mdp_superblock_1 *freshest_sb = page_address(freshest->sb_page); + u32 freshest_max_dev = le32_to_cpu(freshest_sb->max_dev); + + if (rdev->desc_nr >= freshest_max_dev) { + /* this is unexpected, better not proceed */ + pr_warn("md: %s: rdev[%pg]: desc_nr(%d) >= freshest(%pg)->sb->max_dev(%u)\n", + mdname(mddev), rdev->bdev, rdev->desc_nr, + freshest->bdev, freshest_max_dev); + return -EUCLEAN; } - if (sb->devflags & WriteMostly1) - set_bit(WriteMostly, &rdev->flags); - if (sb->devflags & FailFast1) - set_bit(FailFast, &rdev->flags); - if (le32_to_cpu(sb->feature_map) & MD_FEATURE_REPLACEMENT) - set_bit(Replacement, &rdev->flags); - } else /* MULTIPATH are always insync */ - set_bit(In_sync, &rdev->flags); + + role = le16_to_cpu(freshest_sb->dev_roles[rdev->desc_nr]); + pr_debug("md: %s: rdev[%pg]: role=%d(0x%x) according to freshest %pg\n", + mdname(mddev), rdev->bdev, role, role, freshest->bdev); + } else { + role = le16_to_cpu(sb->dev_roles[rdev->desc_nr]); + } + switch (role) { + case MD_DISK_ROLE_SPARE: /* spare */ + break; + case MD_DISK_ROLE_FAULTY: /* faulty */ + set_bit(Faulty, &rdev->flags); + break; + case MD_DISK_ROLE_JOURNAL: /* journal device */ + if (!(le32_to_cpu(sb->feature_map) & MD_FEATURE_JOURNAL)) { + /* journal device without journal feature */ + pr_warn("md: journal device provided without journal feature, ignoring the device\n"); + return -EINVAL; + } + set_bit(Journal, &rdev->flags); + rdev->journal_tail = le64_to_cpu(sb->journal_tail); + rdev->raid_disk = 0; + break; + default: + rdev->saved_raid_disk = role; + if ((le32_to_cpu(sb->feature_map) & + MD_FEATURE_RECOVERY_OFFSET)) { + rdev->recovery_offset = le64_to_cpu(sb->recovery_offset); + if (!(le32_to_cpu(sb->feature_map) & + MD_FEATURE_RECOVERY_BITMAP)) + rdev->saved_raid_disk = -1; + } else { + /* + * If the array is FROZEN, then the device can't + * be in_sync with rest of array. + */ + if (!test_bit(MD_RECOVERY_FROZEN, + &mddev->recovery)) + set_bit(In_sync, &rdev->flags); + } + rdev->raid_disk = role; + break; + } + if (sb->devflags & WriteMostly1) + set_bit(WriteMostly, &rdev->flags); + if (sb->devflags & FailFast1) + set_bit(FailFast, &rdev->flags); + if (le32_to_cpu(sb->feature_map) & MD_FEATURE_REPLACEMENT) + set_bit(Replacement, &rdev->flags); return 0; } @@ -2845,10 +2878,6 @@ rewrite: } else pr_debug("md: %pg (skipping faulty)\n", rdev->bdev); - - if (mddev->level == LEVEL_MULTIPATH) - /* only need to write one superblock... */ - break; } if (md_super_wait(mddev) < 0) goto rewrite; @@ -2890,7 +2919,7 @@ static int add_bound_rdev(struct md_rdev *rdev) * and should be added immediately. */ super_types[mddev->major_version]. - validate_super(mddev, rdev); + validate_super(mddev, NULL/*freshest*/, rdev); err = mddev->pers->hot_add_disk(mddev, rdev); if (err) { md_kick_rdev_from_array(rdev); @@ -3827,7 +3856,7 @@ static int analyze_sbs(struct mddev *mddev) } super_types[mddev->major_version]. - validate_super(mddev, freshest); + validate_super(mddev, NULL/*freshest*/, freshest); i = 0; rdev_for_each_safe(rdev, tmp, mddev) { @@ -3842,20 +3871,15 @@ static int analyze_sbs(struct mddev *mddev) } if (rdev != freshest) { if (super_types[mddev->major_version]. - validate_super(mddev, rdev)) { + validate_super(mddev, freshest, rdev)) { pr_warn("md: kicking non-fresh %pg from array!\n", rdev->bdev); md_kick_rdev_from_array(rdev); continue; } } - if (mddev->level == LEVEL_MULTIPATH) { - rdev->desc_nr = i++; - rdev->raid_disk = rdev->desc_nr; - set_bit(In_sync, &rdev->flags); - } else if (rdev->raid_disk >= - (mddev->raid_disks - min(0, mddev->delta_disks)) && - !test_bit(Journal, &rdev->flags)) { + if (rdev->raid_disk >= (mddev->raid_disks - min(0, mddev->delta_disks)) && + !test_bit(Journal, &rdev->flags)) { rdev->raid_disk = -1; clear_bit(In_sync, &rdev->flags); } @@ -6833,7 +6857,7 @@ int md_add_new_disk(struct mddev *mddev, struct mdu_disk_info_s *info) rdev->saved_raid_disk = rdev->raid_disk; } else super_types[mddev->major_version]. - validate_super(mddev, rdev); + validate_super(mddev, NULL/*freshest*/, rdev); if ((info->state & (1<<MD_DISK_SYNC)) && rdev->raid_disk != info->raid_disk) { /* This was a hot-add request, but events doesn't @@ -8076,7 +8100,7 @@ void md_error(struct mddev *mddev, struct md_rdev *rdev) return; mddev->pers->error_handler(mddev, rdev); - if (mddev->pers->level == 0 || mddev->pers->level == LEVEL_LINEAR) + if (mddev->pers->level == 0) return; if (mddev->degraded && !test_bit(MD_BROKEN, &mddev->flags)) @@ -8112,6 +8136,19 @@ static void status_unused(struct seq_file *seq) seq_printf(seq, "\n"); } +static void status_personalities(struct seq_file *seq) +{ + struct md_personality *pers; + + seq_puts(seq, "Personalities : "); + spin_lock(&pers_lock); + list_for_each_entry(pers, &pers_list, list) + seq_printf(seq, "[%s] ", pers->name); + + spin_unlock(&pers_lock); + seq_puts(seq, "\n"); +} + static int status_resync(struct seq_file *seq, struct mddev *mddev) { sector_t max_sectors, resync, res; @@ -8253,20 +8290,10 @@ static int status_resync(struct seq_file *seq, struct mddev *mddev) static void *md_seq_start(struct seq_file *seq, loff_t *pos) __acquires(&all_mddevs_lock) { - struct md_personality *pers; - - seq_puts(seq, "Personalities : "); - spin_lock(&pers_lock); - list_for_each_entry(pers, &pers_list, list) - seq_printf(seq, "[%s] ", pers->name); - - spin_unlock(&pers_lock); - seq_puts(seq, "\n"); seq->poll_event = atomic_read(&md_event_count); - spin_lock(&all_mddevs_lock); - return seq_list_start(&all_mddevs, *pos); + return seq_list_start_head(&all_mddevs, *pos); } static void *md_seq_next(struct seq_file *seq, void *v, loff_t *pos) @@ -8277,16 +8304,23 @@ static void *md_seq_next(struct seq_file *seq, void *v, loff_t *pos) static void md_seq_stop(struct seq_file *seq, void *v) __releases(&all_mddevs_lock) { - status_unused(seq); spin_unlock(&all_mddevs_lock); } static int md_seq_show(struct seq_file *seq, void *v) { - struct mddev *mddev = list_entry(v, struct mddev, all_mddevs); + struct mddev *mddev; sector_t sectors; struct md_rdev *rdev; + if (v == &all_mddevs) { + status_personalities(seq); + if (list_empty(&all_mddevs)) + status_unused(seq); + return 0; + } + + mddev = list_entry(v, struct mddev, all_mddevs); if (!mddev_get(mddev)) return 0; @@ -8362,6 +8396,10 @@ static int md_seq_show(struct seq_file *seq, void *v) } spin_unlock(&mddev->lock); spin_lock(&all_mddevs_lock); + + if (mddev == list_last_entry(&all_mddevs, struct mddev, all_mddevs)) + status_unused(seq); + if (atomic_dec_and_test(&mddev->active)) __mddev_put(mddev); @@ -8754,12 +8792,16 @@ void md_do_sync(struct md_thread *thread) int ret; /* just incase thread restarts... */ - if (test_bit(MD_RECOVERY_DONE, &mddev->recovery) || - test_bit(MD_RECOVERY_WAIT, &mddev->recovery)) + if (test_bit(MD_RECOVERY_DONE, &mddev->recovery)) return; - if (!md_is_rdwr(mddev)) {/* never try to sync a read-only array */ + + if (test_bit(MD_RECOVERY_INTR, &mddev->recovery)) + goto skip; + + if (test_bit(MD_RECOVERY_WAIT, &mddev->recovery) || + !md_is_rdwr(mddev)) {/* never try to sync a read-only array */ set_bit(MD_RECOVERY_INTR, &mddev->recovery); - return; + goto skip; } if (mddev_is_clustered(mddev)) { @@ -9240,44 +9282,19 @@ static int remove_and_add_spares(struct mddev *mddev, struct md_rdev *rdev; int spares = 0; int removed = 0; - bool remove_some = false; if (this && test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)) /* Mustn't remove devices when resync thread is running */ return 0; rdev_for_each(rdev, mddev) { - if ((this == NULL || rdev == this) && - rdev->raid_disk >= 0 && - !test_bit(Blocked, &rdev->flags) && - test_bit(Faulty, &rdev->flags) && - atomic_read(&rdev->nr_pending)==0) { - /* Faulty non-Blocked devices with nr_pending == 0 - * never get nr_pending incremented, - * never get Faulty cleared, and never get Blocked set. - * So we can synchronize_rcu now rather than once per device - */ - remove_some = true; - set_bit(RemoveSynchronized, &rdev->flags); - } - } - - if (remove_some) - synchronize_rcu(); - rdev_for_each(rdev, mddev) { - if ((this == NULL || rdev == this) && - (test_bit(RemoveSynchronized, &rdev->flags) || - rdev_removeable(rdev))) { - if (mddev->pers->hot_remove_disk( - mddev, rdev) == 0) { - sysfs_unlink_rdev(mddev, rdev); - rdev->saved_raid_disk = rdev->raid_disk; - rdev->raid_disk = -1; - removed++; - } + if ((this == NULL || rdev == this) && rdev_removeable(rdev) && + !mddev->pers->hot_remove_disk(mddev, rdev)) { + sysfs_unlink_rdev(mddev, rdev); + rdev->saved_raid_disk = rdev->raid_disk; + rdev->raid_disk = -1; + removed++; } - if (remove_some && test_bit(RemoveSynchronized, &rdev->flags)) - clear_bit(RemoveSynchronized, &rdev->flags); } if (removed && mddev->kobj.sd) @@ -9359,13 +9376,19 @@ static void md_start_sync(struct work_struct *ws) struct mddev *mddev = container_of(ws, struct mddev, sync_work); int spares = 0; bool suspend = false; + char *name; - if (md_spares_need_change(mddev)) + /* + * If reshape is still in progress, spares won't be added or removed + * from conf until reshape is done. + */ + if (mddev->reshape_position == MaxSector && + md_spares_need_change(mddev)) { suspend = true; + mddev_suspend(mddev, false); + } - suspend ? mddev_suspend_and_lock_nointr(mddev) : - mddev_lock_nointr(mddev); - + mddev_lock_nointr(mddev); if (!md_is_rdwr(mddev)) { /* * On a read-only array we can: @@ -9391,8 +9414,10 @@ static void md_start_sync(struct work_struct *ws) if (spares) md_bitmap_write_all(mddev->bitmap); + name = test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) ? + "reshape" : "resync"; rcu_assign_pointer(mddev->sync_thread, - md_register_thread(md_do_sync, mddev, "resync")); + md_register_thread(md_do_sync, mddev, name)); if (!mddev->sync_thread) { pr_warn("%s: could not start resync thread...\n", mdname(mddev)); @@ -9436,6 +9461,20 @@ not_running: sysfs_notify_dirent_safe(mddev->sysfs_action); } +static void unregister_sync_thread(struct mddev *mddev) +{ + if (!test_bit(MD_RECOVERY_DONE, &mddev->recovery)) { + /* resync/recovery still happening */ + clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery); + return; + } + + if (WARN_ON_ONCE(!mddev->sync_thread)) + return; + + md_reap_sync_thread(mddev); +} + /* * This routine is regularly called by all per-raid-array threads to * deal with generic issues like resync and super-block update. @@ -9460,9 +9499,6 @@ not_running: */ void md_check_recovery(struct mddev *mddev) { - if (READ_ONCE(mddev->suspended)) - return; - if (mddev->bitmap) md_bitmap_daemon_work(mddev); @@ -9476,7 +9512,8 @@ void md_check_recovery(struct mddev *mddev) } if (!md_is_rdwr(mddev) && - !test_bit(MD_RECOVERY_NEEDED, &mddev->recovery)) + !test_bit(MD_RECOVERY_NEEDED, &mddev->recovery) && + !test_bit(MD_RECOVERY_DONE, &mddev->recovery)) return; if ( ! ( (mddev->sb_flags & ~ (1<<MD_SB_CHANGE_PENDING)) || @@ -9498,8 +9535,7 @@ void md_check_recovery(struct mddev *mddev) struct md_rdev *rdev; if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)) { - /* sync_work already queued. */ - clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery); + unregister_sync_thread(mddev); goto unlock; } @@ -9562,16 +9598,7 @@ void md_check_recovery(struct mddev *mddev) * still set. */ if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)) { - if (!test_bit(MD_RECOVERY_DONE, &mddev->recovery)) { - /* resync/recovery still happening */ - clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery); - goto unlock; - } - - if (WARN_ON_ONCE(!mddev->sync_thread)) - goto unlock; - - md_reap_sync_thread(mddev); + unregister_sync_thread(mddev); goto unlock; } diff --git a/drivers/md/md.h b/drivers/md/md.h index ade83af123a2..8d881cc59799 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -190,11 +190,6 @@ enum flag_bits { * than other devices in the array */ ClusterRemove, - RemoveSynchronized, /* synchronize_rcu() was called after - * this device was known to be faulty, - * so it is safe to remove without - * another synchronize_rcu() call. - */ ExternalBbl, /* External metadata provides bad * block management for a disk */ diff --git a/drivers/md/raid1-10.c b/drivers/md/raid1-10.c index 3f22edec70e7..512746551f36 100644 --- a/drivers/md/raid1-10.c +++ b/drivers/md/raid1-10.c @@ -173,3 +173,57 @@ static inline void raid1_prepare_flush_writes(struct bitmap *bitmap) else md_bitmap_unplug(bitmap); } + +/* + * Used by fix_read_error() to decay the per rdev read_errors. + * We halve the read error count for every hour that has elapsed + * since the last recorded read error. + */ +static inline void check_decay_read_errors(struct mddev *mddev, struct md_rdev *rdev) +{ + long cur_time_mon; + unsigned long hours_since_last; + unsigned int read_errors = atomic_read(&rdev->read_errors); + + cur_time_mon = ktime_get_seconds(); + + if (rdev->last_read_error == 0) { + /* first time we've seen a read error */ + rdev->last_read_error = cur_time_mon; + return; + } + + hours_since_last = (long)(cur_time_mon - + rdev->last_read_error) / 3600; + + rdev->last_read_error = cur_time_mon; + + /* + * if hours_since_last is > the number of bits in read_errors + * just set read errors to 0. We do this to avoid + * overflowing the shift of read_errors by hours_since_last. + */ + if (hours_since_last >= 8 * sizeof(read_errors)) + atomic_set(&rdev->read_errors, 0); + else + atomic_set(&rdev->read_errors, read_errors >> hours_since_last); +} + +static inline bool exceed_read_errors(struct mddev *mddev, struct md_rdev *rdev) +{ + int max_read_errors = atomic_read(&mddev->max_corr_read_errors); + int read_errors; + + check_decay_read_errors(mddev, rdev); + read_errors = atomic_inc_return(&rdev->read_errors); + if (read_errors > max_read_errors) { + pr_notice("md/"RAID_1_10_NAME":%s: %pg: Raid device exceeded read_error threshold [cur %d:max %d]\n", + mdname(mddev), rdev->bdev, read_errors, max_read_errors); + pr_notice("md/"RAID_1_10_NAME":%s: %pg: Failing raid device\n", + mdname(mddev), rdev->bdev); + md_error(mddev, rdev); + return true; + } + + return false; +} diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 35d12948e0a9..286f8b16c7bd 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -49,6 +49,7 @@ static void lower_barrier(struct r1conf *conf, sector_t sector_nr); #define raid1_log(md, fmt, args...) \ do { if ((md)->queue) blk_add_trace_msg((md)->queue, "raid1 " fmt, ##args); } while (0) +#define RAID_1_10_NAME "raid1" #include "raid1-10.c" #define START(node) ((node)->start) @@ -609,7 +610,6 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect int choose_first; int choose_next_idle; - rcu_read_lock(); /* * Check if we can balance. We can balance on the whole * device if no resync is going on, or below the resync window. @@ -642,7 +642,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect unsigned int pending; bool nonrot; - rdev = rcu_dereference(conf->mirrors[disk].rdev); + rdev = conf->mirrors[disk].rdev; if (r1_bio->bios[disk] == IO_BLOCKED || rdev == NULL || test_bit(Faulty, &rdev->flags)) @@ -773,7 +773,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect } if (best_disk >= 0) { - rdev = rcu_dereference(conf->mirrors[best_disk].rdev); + rdev = conf->mirrors[best_disk].rdev; if (!rdev) goto retry; atomic_inc(&rdev->nr_pending); @@ -784,7 +784,6 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect conf->mirrors[best_disk].next_seq_sect = this_sector + sectors; } - rcu_read_unlock(); *max_sectors = sectors; return best_disk; @@ -1126,8 +1125,6 @@ static void alloc_behind_master_bio(struct r1bio *r1_bio, behind_bio = bio_alloc_bioset(NULL, vcnt, 0, GFP_NOIO, &r1_bio->mddev->bio_set); - if (!behind_bio) - return; /* discard op, we don't support writezero/writesame yet */ if (!bio_has_data(bio)) { @@ -1235,14 +1232,12 @@ static void raid1_read_request(struct mddev *mddev, struct bio *bio, if (r1bio_existed) { /* Need to get the block device name carefully */ - struct md_rdev *rdev; - rcu_read_lock(); - rdev = rcu_dereference(conf->mirrors[r1_bio->read_disk].rdev); + struct md_rdev *rdev = conf->mirrors[r1_bio->read_disk].rdev; + if (rdev) snprintf(b, sizeof(b), "%pg", rdev->bdev); else strcpy(b, "???"); - rcu_read_unlock(); } /* @@ -1396,10 +1391,9 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio, disks = conf->raid_disks * 2; blocked_rdev = NULL; - rcu_read_lock(); max_sectors = r1_bio->sectors; for (i = 0; i < disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->mirrors[i].rdev); + struct md_rdev *rdev = conf->mirrors[i].rdev; /* * The write-behind io is only attempted on drives marked as @@ -1465,7 +1459,6 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio, } r1_bio->bios[i] = bio; } - rcu_read_unlock(); if (unlikely(blocked_rdev)) { /* Wait for this device to become unblocked */ @@ -1617,15 +1610,16 @@ static void raid1_status(struct seq_file *seq, struct mddev *mddev) struct r1conf *conf = mddev->private; int i; + lockdep_assert_held(&mddev->lock); + seq_printf(seq, " [%d/%d] [", conf->raid_disks, conf->raid_disks - mddev->degraded); - rcu_read_lock(); for (i = 0; i < conf->raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->mirrors[i].rdev); + struct md_rdev *rdev = READ_ONCE(conf->mirrors[i].rdev); + seq_printf(seq, "%s", rdev && test_bit(In_sync, &rdev->flags) ? "U" : "_"); } - rcu_read_unlock(); seq_printf(seq, "]"); } @@ -1691,16 +1685,15 @@ static void print_conf(struct r1conf *conf) pr_debug(" --- wd:%d rd:%d\n", conf->raid_disks - conf->mddev->degraded, conf->raid_disks); - rcu_read_lock(); + lockdep_assert_held(&conf->mddev->reconfig_mutex); for (i = 0; i < conf->raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->mirrors[i].rdev); + struct md_rdev *rdev = conf->mirrors[i].rdev; if (rdev) pr_debug(" disk %d, wo:%d, o:%d, dev:%pg\n", i, !test_bit(In_sync, &rdev->flags), !test_bit(Faulty, &rdev->flags), rdev->bdev); } - rcu_read_unlock(); } static void close_sync(struct r1conf *conf) @@ -1810,7 +1803,7 @@ static int raid1_add_disk(struct mddev *mddev, struct md_rdev *rdev) */ if (rdev->saved_raid_disk < 0) conf->fullsync = 1; - rcu_assign_pointer(p->rdev, rdev); + WRITE_ONCE(p->rdev, rdev); break; } if (test_bit(WantReplacement, &p->rdev->flags) && @@ -1826,7 +1819,7 @@ static int raid1_add_disk(struct mddev *mddev, struct md_rdev *rdev) rdev->raid_disk = repl_slot; err = 0; conf->fullsync = 1; - rcu_assign_pointer(p[conf->raid_disks].rdev, rdev); + WRITE_ONCE(p[conf->raid_disks].rdev, rdev); } print_conf(conf); @@ -1862,16 +1855,7 @@ static int raid1_remove_disk(struct mddev *mddev, struct md_rdev *rdev) err = -EBUSY; goto abort; } - p->rdev = NULL; - if (!test_bit(RemoveSynchronized, &rdev->flags)) { - synchronize_rcu(); - if (atomic_read(&rdev->nr_pending)) { - /* lost the race, try later */ - err = -EBUSY; - p->rdev = rdev; - goto abort; - } - } + WRITE_ONCE(p->rdev, NULL); if (conf->mirrors[conf->raid_disks + number].rdev) { /* We just removed a device that is being replaced. * Move down the replacement. We drain all IO before @@ -1892,7 +1876,7 @@ static int raid1_remove_disk(struct mddev *mddev, struct md_rdev *rdev) goto abort; } clear_bit(Replacement, &repl->flags); - p->rdev = repl; + WRITE_ONCE(p->rdev, repl); conf->mirrors[conf->raid_disks + number].rdev = NULL; unfreeze_array(conf); } @@ -1984,12 +1968,12 @@ static void end_sync_write(struct bio *bio) } static int r1_sync_page_io(struct md_rdev *rdev, sector_t sector, - int sectors, struct page *page, int rw) + int sectors, struct page *page, blk_opf_t rw) { if (sync_page_io(rdev, sector, sectors << 9, page, rw, false)) /* success */ return 1; - if (rw == WRITE) { + if (rw == REQ_OP_WRITE) { set_bit(WriteErrorSeen, &rdev->flags); if (!test_and_set_bit(WantReplacement, &rdev->flags)) @@ -2106,7 +2090,7 @@ static int fix_sync_read_error(struct r1bio *r1_bio) rdev = conf->mirrors[d].rdev; if (r1_sync_page_io(rdev, sect, s, pages[idx], - WRITE) == 0) { + REQ_OP_WRITE) == 0) { r1_bio->bios[d]->bi_end_io = NULL; rdev_dec_pending(rdev, mddev); } @@ -2121,7 +2105,7 @@ static int fix_sync_read_error(struct r1bio *r1_bio) rdev = conf->mirrors[d].rdev; if (r1_sync_page_io(rdev, sect, s, pages[idx], - READ) != 0) + REQ_OP_READ) != 0) atomic_add(s, &rdev->corrected_errors); } sectors -= s; @@ -2272,16 +2256,24 @@ static void sync_request_write(struct mddev *mddev, struct r1bio *r1_bio) * 3. Performs writes following reads for array synchronising. */ -static void fix_read_error(struct r1conf *conf, int read_disk, - sector_t sect, int sectors) +static void fix_read_error(struct r1conf *conf, struct r1bio *r1_bio) { + sector_t sect = r1_bio->sector; + int sectors = r1_bio->sectors; + int read_disk = r1_bio->read_disk; struct mddev *mddev = conf->mddev; + struct md_rdev *rdev = conf->mirrors[read_disk].rdev; + + if (exceed_read_errors(mddev, rdev)) { + r1_bio->bios[r1_bio->read_disk] = IO_BLOCKED; + return; + } + while(sectors) { int s = sectors; int d = read_disk; int success = 0; int start; - struct md_rdev *rdev; if (s > (PAGE_SIZE>>9)) s = PAGE_SIZE >> 9; @@ -2290,8 +2282,7 @@ static void fix_read_error(struct r1conf *conf, int read_disk, sector_t first_bad; int bad_sectors; - rcu_read_lock(); - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; if (rdev && (test_bit(In_sync, &rdev->flags) || (!test_bit(Faulty, &rdev->flags) && @@ -2299,15 +2290,14 @@ static void fix_read_error(struct r1conf *conf, int read_disk, is_badblock(rdev, sect, s, &first_bad, &bad_sectors) == 0) { atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); if (sync_page_io(rdev, sect, s<<9, conf->tmppage, REQ_OP_READ, false)) success = 1; rdev_dec_pending(rdev, mddev); if (success) break; - } else - rcu_read_unlock(); + } + d++; if (d == conf->raid_disks * 2) d = 0; @@ -2326,31 +2316,26 @@ static void fix_read_error(struct r1conf *conf, int read_disk, if (d==0) d = conf->raid_disks * 2; d--; - rcu_read_lock(); - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; if (rdev && !test_bit(Faulty, &rdev->flags)) { atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); r1_sync_page_io(rdev, sect, s, - conf->tmppage, WRITE); + conf->tmppage, REQ_OP_WRITE); rdev_dec_pending(rdev, mddev); - } else - rcu_read_unlock(); + } } d = start; while (d != read_disk) { if (d==0) d = conf->raid_disks * 2; d--; - rcu_read_lock(); - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; if (rdev && !test_bit(Faulty, &rdev->flags)) { atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); if (r1_sync_page_io(rdev, sect, s, - conf->tmppage, READ)) { + conf->tmppage, REQ_OP_READ)) { atomic_add(s, &rdev->corrected_errors); pr_info("md/raid1:%s: read error corrected (%d sectors at %llu on %pg)\n", mdname(mddev), s, @@ -2359,8 +2344,7 @@ static void fix_read_error(struct r1conf *conf, int read_disk, rdev->bdev); } rdev_dec_pending(rdev, mddev); - } else - rcu_read_unlock(); + } } sectors -= s; sect += s; @@ -2530,8 +2514,7 @@ static void handle_read_error(struct r1conf *conf, struct r1bio *r1_bio) if (mddev->ro == 0 && !test_bit(FailFast, &rdev->flags)) { freeze_array(conf, 1); - fix_read_error(conf, r1_bio->read_disk, - r1_bio->sector, r1_bio->sectors); + fix_read_error(conf, r1_bio); unfreeze_array(conf); } else if (mddev->ro == 0 && test_bit(FailFast, &rdev->flags)) { md_error(mddev, rdev); @@ -2741,7 +2724,6 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr, r1_bio = raid1_alloc_init_r1buf(conf); - rcu_read_lock(); /* * If we get a correctably read error during resync or recovery, * we might want to read from a different device. So we @@ -2762,7 +2744,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr, struct md_rdev *rdev; bio = r1_bio->bios[i]; - rdev = rcu_dereference(conf->mirrors[i].rdev); + rdev = conf->mirrors[i].rdev; if (rdev == NULL || test_bit(Faulty, &rdev->flags)) { if (i < conf->raid_disks) @@ -2820,7 +2802,6 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr, bio->bi_opf |= MD_FAILFAST; } } - rcu_read_unlock(); if (disk < 0) disk = wonly; r1_bio->read_disk = disk; diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index a5927e98dc67..a5f8419e2df1 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -19,6 +19,8 @@ #include <linux/raid/md_p.h> #include <trace/events/block.h> #include "md.h" + +#define RAID_1_10_NAME "raid10" #include "raid10.h" #include "raid0.h" #include "md-bitmap.h" @@ -743,7 +745,6 @@ static struct md_rdev *read_balance(struct r10conf *conf, struct geom *geo = &conf->geo; raid10_find_phys(conf, r10_bio); - rcu_read_lock(); best_dist_slot = -1; min_pending = UINT_MAX; best_dist_rdev = NULL; @@ -775,18 +776,11 @@ static struct md_rdev *read_balance(struct r10conf *conf, if (r10_bio->devs[slot].bio == IO_BLOCKED) continue; disk = r10_bio->devs[slot].devnum; - rdev = rcu_dereference(conf->mirrors[disk].replacement); + rdev = conf->mirrors[disk].replacement; if (rdev == NULL || test_bit(Faulty, &rdev->flags) || r10_bio->devs[slot].addr + sectors > - rdev->recovery_offset) { - /* - * Read replacement first to prevent reading both rdev - * and replacement as NULL during replacement replace - * rdev. - */ - smp_mb(); - rdev = rcu_dereference(conf->mirrors[disk].rdev); - } + rdev->recovery_offset) + rdev = conf->mirrors[disk].rdev; if (rdev == NULL || test_bit(Faulty, &rdev->flags)) continue; @@ -876,7 +870,6 @@ static struct md_rdev *read_balance(struct r10conf *conf, r10_bio->read_slot = slot; } else rdev = NULL; - rcu_read_unlock(); *max_sectors = best_good_sectors; return rdev; @@ -1198,9 +1191,8 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio, */ gfp = GFP_NOIO | __GFP_HIGH; - rcu_read_lock(); disk = r10_bio->devs[slot].devnum; - err_rdev = rcu_dereference(conf->mirrors[disk].rdev); + err_rdev = conf->mirrors[disk].rdev; if (err_rdev) snprintf(b, sizeof(b), "%pg", err_rdev->bdev); else { @@ -1208,7 +1200,6 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio, /* This never gets dereferenced */ err_rdev = r10_bio->devs[slot].rdev; } - rcu_read_unlock(); } if (!regular_request_wait(mddev, conf, bio, r10_bio->sectors)) @@ -1279,15 +1270,8 @@ static void raid10_write_one_disk(struct mddev *mddev, struct r10bio *r10_bio, int devnum = r10_bio->devs[n_copy].devnum; struct bio *mbio; - if (replacement) { - rdev = conf->mirrors[devnum].replacement; - if (rdev == NULL) { - /* Replacement just got moved to main 'rdev' */ - smp_mb(); - rdev = conf->mirrors[devnum].rdev; - } - } else - rdev = conf->mirrors[devnum].rdev; + rdev = replacement ? conf->mirrors[devnum].replacement : + conf->mirrors[devnum].rdev; mbio = bio_alloc_clone(rdev->bdev, bio, GFP_NOIO, &mddev->bio_set); if (replacement) @@ -1321,25 +1305,6 @@ static void raid10_write_one_disk(struct mddev *mddev, struct r10bio *r10_bio, } } -static struct md_rdev *dereference_rdev_and_rrdev(struct raid10_info *mirror, - struct md_rdev **prrdev) -{ - struct md_rdev *rdev, *rrdev; - - rrdev = rcu_dereference(mirror->replacement); - /* - * Read replacement first to prevent reading both rdev and - * replacement as NULL during replacement replace rdev. - */ - smp_mb(); - rdev = rcu_dereference(mirror->rdev); - if (rdev == rrdev) - rrdev = NULL; - - *prrdev = rrdev; - return rdev; -} - static void wait_blocked_dev(struct mddev *mddev, struct r10bio *r10_bio) { int i; @@ -1348,11 +1313,11 @@ static void wait_blocked_dev(struct mddev *mddev, struct r10bio *r10_bio) retry_wait: blocked_rdev = NULL; - rcu_read_lock(); for (i = 0; i < conf->copies; i++) { struct md_rdev *rdev, *rrdev; - rdev = dereference_rdev_and_rrdev(&conf->mirrors[i], &rrdev); + rdev = conf->mirrors[i].rdev; + rrdev = conf->mirrors[i].replacement; if (rdev && unlikely(test_bit(Blocked, &rdev->flags))) { atomic_inc(&rdev->nr_pending); blocked_rdev = rdev; @@ -1391,7 +1356,6 @@ retry_wait: } } } - rcu_read_unlock(); if (unlikely(blocked_rdev)) { /* Have to wait for this device to get unblocked, then retry */ @@ -1474,14 +1438,14 @@ static void raid10_write_request(struct mddev *mddev, struct bio *bio, wait_blocked_dev(mddev, r10_bio); - rcu_read_lock(); max_sectors = r10_bio->sectors; for (i = 0; i < conf->copies; i++) { int d = r10_bio->devs[i].devnum; struct md_rdev *rdev, *rrdev; - rdev = dereference_rdev_and_rrdev(&conf->mirrors[d], &rrdev); + rdev = conf->mirrors[d].rdev; + rrdev = conf->mirrors[d].replacement; if (rdev && (test_bit(Faulty, &rdev->flags))) rdev = NULL; if (rrdev && (test_bit(Faulty, &rrdev->flags))) @@ -1535,7 +1499,6 @@ static void raid10_write_request(struct mddev *mddev, struct bio *bio, atomic_inc(&rrdev->nr_pending); } } - rcu_read_unlock(); if (max_sectors < r10_bio->sectors) r10_bio->sectors = max_sectors; @@ -1625,17 +1588,8 @@ static void raid10_end_discard_request(struct bio *bio) set_bit(R10BIO_Uptodate, &r10_bio->state); dev = find_bio_disk(conf, r10_bio, bio, &slot, &repl); - if (repl) - rdev = conf->mirrors[dev].replacement; - if (!rdev) { - /* - * raid10_remove_disk uses smp_mb to make sure rdev is set to - * replacement before setting replacement to NULL. It can read - * rdev first without barrier protect even replacement is NULL - */ - smp_rmb(); - rdev = conf->mirrors[dev].rdev; - } + rdev = repl ? conf->mirrors[dev].replacement : + conf->mirrors[dev].rdev; raid_end_discard_bio(r10_bio); rdev_dec_pending(rdev, conf->mddev); @@ -1785,11 +1739,11 @@ retry_discard: * inc refcount on their rdev. Record them by setting * bios[x] to bio */ - rcu_read_lock(); for (disk = 0; disk < geo->raid_disks; disk++) { struct md_rdev *rdev, *rrdev; - rdev = dereference_rdev_and_rrdev(&conf->mirrors[disk], &rrdev); + rdev = conf->mirrors[disk].rdev; + rrdev = conf->mirrors[disk].replacement; r10_bio->devs[disk].bio = NULL; r10_bio->devs[disk].repl_bio = NULL; @@ -1809,7 +1763,6 @@ retry_discard: atomic_inc(&rrdev->nr_pending); } } - rcu_read_unlock(); atomic_set(&r10_bio->remaining, 1); for (disk = 0; disk < geo->raid_disks; disk++) { @@ -1939,6 +1892,8 @@ static void raid10_status(struct seq_file *seq, struct mddev *mddev) struct r10conf *conf = mddev->private; int i; + lockdep_assert_held(&mddev->lock); + if (conf->geo.near_copies < conf->geo.raid_disks) seq_printf(seq, " %dK chunks", mddev->chunk_sectors / 2); if (conf->geo.near_copies > 1) @@ -1953,12 +1908,11 @@ static void raid10_status(struct seq_file *seq, struct mddev *mddev) } seq_printf(seq, " [%d/%d] [", conf->geo.raid_disks, conf->geo.raid_disks - mddev->degraded); - rcu_read_lock(); for (i = 0; i < conf->geo.raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->mirrors[i].rdev); + struct md_rdev *rdev = READ_ONCE(conf->mirrors[i].rdev); + seq_printf(seq, "%s", rdev && test_bit(In_sync, &rdev->flags) ? "U" : "_"); } - rcu_read_unlock(); seq_printf(seq, "]"); } @@ -1980,7 +1934,6 @@ static int _enough(struct r10conf *conf, int previous, int ignore) ncopies = conf->geo.near_copies; } - rcu_read_lock(); do { int n = conf->copies; int cnt = 0; @@ -1988,7 +1941,7 @@ static int _enough(struct r10conf *conf, int previous, int ignore) while (n--) { struct md_rdev *rdev; if (this != ignore && - (rdev = rcu_dereference(conf->mirrors[this].rdev)) && + (rdev = conf->mirrors[this].rdev) && test_bit(In_sync, &rdev->flags)) cnt++; this = (this+1) % disks; @@ -1999,7 +1952,6 @@ static int _enough(struct r10conf *conf, int previous, int ignore) } while (first != 0); has_enough = 1; out: - rcu_read_unlock(); return has_enough; } @@ -2072,8 +2024,7 @@ static void print_conf(struct r10conf *conf) pr_debug(" --- wd:%d rd:%d\n", conf->geo.raid_disks - conf->mddev->degraded, conf->geo.raid_disks); - /* This is only called with ->reconfix_mutex held, so - * rcu protection of rdev is not needed */ + lockdep_assert_held(&conf->mddev->reconfig_mutex); for (i = 0; i < conf->geo.raid_disks; i++) { rdev = conf->mirrors[i].rdev; if (rdev) @@ -2190,7 +2141,7 @@ static int raid10_add_disk(struct mddev *mddev, struct md_rdev *rdev) err = 0; if (rdev->saved_raid_disk != mirror) conf->fullsync = 1; - rcu_assign_pointer(p->rdev, rdev); + WRITE_ONCE(p->rdev, rdev); break; } @@ -2204,7 +2155,7 @@ static int raid10_add_disk(struct mddev *mddev, struct md_rdev *rdev) disk_stack_limits(mddev->gendisk, rdev->bdev, rdev->data_offset << 9); conf->fullsync = 1; - rcu_assign_pointer(p->replacement, rdev); + WRITE_ONCE(p->replacement, rdev); } print_conf(conf); @@ -2246,24 +2197,12 @@ static int raid10_remove_disk(struct mddev *mddev, struct md_rdev *rdev) err = -EBUSY; goto abort; } - *rdevp = NULL; - if (!test_bit(RemoveSynchronized, &rdev->flags)) { - synchronize_rcu(); - if (atomic_read(&rdev->nr_pending)) { - /* lost the race, try later */ - err = -EBUSY; - *rdevp = rdev; - goto abort; - } - } + WRITE_ONCE(*rdevp, NULL); if (p->replacement) { /* We must have just cleared 'rdev' */ - p->rdev = p->replacement; + WRITE_ONCE(p->rdev, p->replacement); clear_bit(Replacement, &p->replacement->flags); - smp_mb(); /* Make sure other CPUs may see both as identical - * but will never see neither -- if they are careful. - */ - p->replacement = NULL; + WRITE_ONCE(p->replacement, NULL); } clear_bit(WantReplacement, &rdev->flags); @@ -2655,42 +2594,6 @@ static void recovery_request_write(struct mddev *mddev, struct r10bio *r10_bio) } } -/* - * Used by fix_read_error() to decay the per rdev read_errors. - * We halve the read error count for every hour that has elapsed - * since the last recorded read error. - * - */ -static void check_decay_read_errors(struct mddev *mddev, struct md_rdev *rdev) -{ - long cur_time_mon; - unsigned long hours_since_last; - unsigned int read_errors = atomic_read(&rdev->read_errors); - - cur_time_mon = ktime_get_seconds(); - - if (rdev->last_read_error == 0) { - /* first time we've seen a read error */ - rdev->last_read_error = cur_time_mon; - return; - } - - hours_since_last = (long)(cur_time_mon - - rdev->last_read_error) / 3600; - - rdev->last_read_error = cur_time_mon; - - /* - * if hours_since_last is > the number of bits in read_errors - * just set read errors to 0. We do this to avoid - * overflowing the shift of read_errors by hours_since_last. - */ - if (hours_since_last >= 8 * sizeof(read_errors)) - atomic_set(&rdev->read_errors, 0); - else - atomic_set(&rdev->read_errors, read_errors >> hours_since_last); -} - static int r10_sync_page_io(struct md_rdev *rdev, sector_t sector, int sectors, struct page *page, enum req_op op) { @@ -2728,7 +2631,6 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 int sect = 0; /* Offset from r10_bio->sector */ int sectors = r10_bio->sectors, slot = r10_bio->read_slot; struct md_rdev *rdev; - int max_read_errors = atomic_read(&mddev->max_corr_read_errors); int d = r10_bio->devs[slot].devnum; /* still own a reference to this rdev, so it cannot @@ -2741,15 +2643,7 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 more fix_read_error() attempts */ return; - check_decay_read_errors(mddev, rdev); - atomic_inc(&rdev->read_errors); - if (atomic_read(&rdev->read_errors) > max_read_errors) { - pr_notice("md/raid10:%s: %pg: Raid device exceeded read_error threshold [cur %d:max %d]\n", - mdname(mddev), rdev->bdev, - atomic_read(&rdev->read_errors), max_read_errors); - pr_notice("md/raid10:%s: %pg: Failing raid device\n", - mdname(mddev), rdev->bdev); - md_error(mddev, rdev); + if (exceed_read_errors(mddev, rdev)) { r10_bio->devs[slot].bio = IO_BLOCKED; return; } @@ -2763,20 +2657,18 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 if (s > (PAGE_SIZE>>9)) s = PAGE_SIZE >> 9; - rcu_read_lock(); do { sector_t first_bad; int bad_sectors; d = r10_bio->devs[sl].devnum; - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; if (rdev && test_bit(In_sync, &rdev->flags) && !test_bit(Faulty, &rdev->flags) && is_badblock(rdev, r10_bio->devs[sl].addr + sect, s, &first_bad, &bad_sectors) == 0) { atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); success = sync_page_io(rdev, r10_bio->devs[sl].addr + sect, @@ -2784,7 +2676,6 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 conf->tmppage, REQ_OP_READ, false); rdev_dec_pending(rdev, mddev); - rcu_read_lock(); if (success) break; } @@ -2792,7 +2683,6 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 if (sl == conf->copies) sl = 0; } while (sl != slot); - rcu_read_unlock(); if (!success) { /* Cannot read from anywhere, just mark the block @@ -2816,20 +2706,18 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 start = sl; /* write it back and re-read */ - rcu_read_lock(); while (sl != slot) { if (sl==0) sl = conf->copies; sl--; d = r10_bio->devs[sl].devnum; - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; if (!rdev || test_bit(Faulty, &rdev->flags) || !test_bit(In_sync, &rdev->flags)) continue; atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); if (r10_sync_page_io(rdev, r10_bio->devs[sl].addr + sect, @@ -2848,7 +2736,6 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 rdev->bdev); } rdev_dec_pending(rdev, mddev); - rcu_read_lock(); } sl = start; while (sl != slot) { @@ -2856,14 +2743,13 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 sl = conf->copies; sl--; d = r10_bio->devs[sl].devnum; - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; if (!rdev || test_bit(Faulty, &rdev->flags) || !test_bit(In_sync, &rdev->flags)) continue; atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); switch (r10_sync_page_io(rdev, r10_bio->devs[sl].addr + sect, @@ -2891,9 +2777,7 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 } rdev_dec_pending(rdev, mddev); - rcu_read_lock(); } - rcu_read_unlock(); sectors -= s; sect += s; @@ -3367,14 +3251,13 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, /* Completed a full sync so the replacements * are now fully recovered. */ - rcu_read_lock(); for (i = 0; i < conf->geo.raid_disks; i++) { struct md_rdev *rdev = - rcu_dereference(conf->mirrors[i].replacement); + conf->mirrors[i].replacement; + if (rdev) rdev->recovery_offset = MaxSector; } - rcu_read_unlock(); } conf->fullsync = 0; } @@ -3455,9 +3338,8 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, struct raid10_info *mirror = &conf->mirrors[i]; struct md_rdev *mrdev, *mreplace; - rcu_read_lock(); - mrdev = rcu_dereference(mirror->rdev); - mreplace = rcu_dereference(mirror->replacement); + mrdev = mirror->rdev; + mreplace = mirror->replacement; if (mrdev && (test_bit(Faulty, &mrdev->flags) || test_bit(In_sync, &mrdev->flags))) @@ -3465,22 +3347,18 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, if (mreplace && test_bit(Faulty, &mreplace->flags)) mreplace = NULL; - if (!mrdev && !mreplace) { - rcu_read_unlock(); + if (!mrdev && !mreplace) continue; - } still_degraded = 0; /* want to reconstruct this device */ rb2 = r10_bio; sect = raid10_find_virt(conf, sector_nr, i); - if (sect >= mddev->resync_max_sectors) { + if (sect >= mddev->resync_max_sectors) /* last stripe is not complete - don't * try to recover this sector. */ - rcu_read_unlock(); continue; - } /* Unless we are doing a full sync, or a replacement * we only need to recover the block if it is set in * the bitmap @@ -3496,14 +3374,12 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, * that there will never be anything to do here */ chunks_skipped = -1; - rcu_read_unlock(); continue; } if (mrdev) atomic_inc(&mrdev->nr_pending); if (mreplace) atomic_inc(&mreplace->nr_pending); - rcu_read_unlock(); r10_bio = raid10_alloc_init_r10buf(conf); r10_bio->state = 0; @@ -3522,10 +3398,9 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, /* Need to check if the array will still be * degraded */ - rcu_read_lock(); for (j = 0; j < conf->geo.raid_disks; j++) { - struct md_rdev *rdev = rcu_dereference( - conf->mirrors[j].rdev); + struct md_rdev *rdev = conf->mirrors[j].rdev; + if (rdev == NULL || test_bit(Faulty, &rdev->flags)) { still_degraded = 1; break; @@ -3540,8 +3415,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, int k; int d = r10_bio->devs[j].devnum; sector_t from_addr, to_addr; - struct md_rdev *rdev = - rcu_dereference(conf->mirrors[d].rdev); + struct md_rdev *rdev = conf->mirrors[d].rdev; sector_t sector, first_bad; int bad_sectors; if (!rdev || @@ -3620,7 +3494,6 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, atomic_inc(&r10_bio->remaining); break; } - rcu_read_unlock(); if (j == conf->copies) { /* Cannot recover, so abort the recovery or * record a bad block */ @@ -3747,12 +3620,10 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, bio = r10_bio->devs[i].bio; bio->bi_status = BLK_STS_IOERR; - rcu_read_lock(); - rdev = rcu_dereference(conf->mirrors[d].rdev); - if (rdev == NULL || test_bit(Faulty, &rdev->flags)) { - rcu_read_unlock(); + rdev = conf->mirrors[d].rdev; + if (rdev == NULL || test_bit(Faulty, &rdev->flags)) continue; - } + sector = r10_bio->devs[i].addr; if (is_badblock(rdev, sector, max_sync, &first_bad, &bad_sectors)) { @@ -3762,7 +3633,6 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, bad_sectors -= (sector - first_bad); if (max_sync > bad_sectors) max_sync = bad_sectors; - rcu_read_unlock(); continue; } } @@ -3778,11 +3648,10 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, bio_set_dev(bio, rdev->bdev); count++; - rdev = rcu_dereference(conf->mirrors[d].replacement); - if (rdev == NULL || test_bit(Faulty, &rdev->flags)) { - rcu_read_unlock(); + rdev = conf->mirrors[d].replacement; + if (rdev == NULL || test_bit(Faulty, &rdev->flags)) continue; - } + atomic_inc(&rdev->nr_pending); /* Need to set up for writing to the replacement */ @@ -3799,7 +3668,6 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, bio->bi_iter.bi_sector = sector + rdev->data_offset; bio_set_dev(bio, rdev->bdev); count++; - rcu_read_unlock(); } if (count < 2) { @@ -4307,11 +4175,7 @@ static int raid10_run(struct mddev *mddev) clear_bit(MD_RECOVERY_SYNC, &mddev->recovery); clear_bit(MD_RECOVERY_CHECK, &mddev->recovery); set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); - set_bit(MD_RECOVERY_RUNNING, &mddev->recovery); - rcu_assign_pointer(mddev->sync_thread, - md_register_thread(md_do_sync, mddev, "reshape")); - if (!mddev->sync_thread) - goto out_free_conf; + set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); } return 0; @@ -4509,11 +4373,11 @@ static int calc_degraded(struct r10conf *conf) int degraded, degraded2; int i; - rcu_read_lock(); degraded = 0; /* 'prev' section first */ for (i = 0; i < conf->prev.raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->mirrors[i].rdev); + struct md_rdev *rdev = conf->mirrors[i].rdev; + if (!rdev || test_bit(Faulty, &rdev->flags)) degraded++; else if (!test_bit(In_sync, &rdev->flags)) @@ -4523,13 +4387,12 @@ static int calc_degraded(struct r10conf *conf) */ degraded++; } - rcu_read_unlock(); if (conf->geo.raid_disks == conf->prev.raid_disks) return degraded; - rcu_read_lock(); degraded2 = 0; for (i = 0; i < conf->geo.raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->mirrors[i].rdev); + struct md_rdev *rdev = conf->mirrors[i].rdev; + if (!rdev || test_bit(Faulty, &rdev->flags)) degraded2++; else if (!test_bit(In_sync, &rdev->flags)) { @@ -4542,7 +4405,6 @@ static int calc_degraded(struct r10conf *conf) degraded2++; } } - rcu_read_unlock(); if (degraded2 > degraded) return degraded2; return degraded; @@ -4707,16 +4569,8 @@ out: clear_bit(MD_RECOVERY_CHECK, &mddev->recovery); clear_bit(MD_RECOVERY_DONE, &mddev->recovery); set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); - set_bit(MD_RECOVERY_RUNNING, &mddev->recovery); - - rcu_assign_pointer(mddev->sync_thread, - md_register_thread(md_do_sync, mddev, "reshape")); - if (!mddev->sync_thread) { - ret = -EAGAIN; - goto abort; - } + set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); conf->reshape_checkpoint = jiffies; - md_wakeup_thread(mddev->sync_thread); md_new_event(); return 0; @@ -4974,16 +4828,15 @@ read_more: blist = read_bio; read_bio->bi_next = NULL; - rcu_read_lock(); for (s = 0; s < conf->copies*2; s++) { struct bio *b; int d = r10_bio->devs[s/2].devnum; struct md_rdev *rdev2; if (s&1) { - rdev2 = rcu_dereference(conf->mirrors[d].replacement); + rdev2 = conf->mirrors[d].replacement; b = r10_bio->devs[s/2].repl_bio; } else { - rdev2 = rcu_dereference(conf->mirrors[d].rdev); + rdev2 = conf->mirrors[d].rdev; b = r10_bio->devs[s/2].bio; } if (!rdev2 || test_bit(Faulty, &rdev2->flags)) @@ -5017,7 +4870,6 @@ read_more: sector_nr += len >> 9; nr_sectors += len >> 9; } - rcu_read_unlock(); r10_bio->sectors = nr_sectors; /* Now submit the read */ @@ -5070,20 +4922,17 @@ static void reshape_request_write(struct mddev *mddev, struct r10bio *r10_bio) struct bio *b; int d = r10_bio->devs[s/2].devnum; struct md_rdev *rdev; - rcu_read_lock(); if (s&1) { - rdev = rcu_dereference(conf->mirrors[d].replacement); + rdev = conf->mirrors[d].replacement; b = r10_bio->devs[s/2].repl_bio; } else { - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; b = r10_bio->devs[s/2].bio; } - if (!rdev || test_bit(Faulty, &rdev->flags)) { - rcu_read_unlock(); + if (!rdev || test_bit(Faulty, &rdev->flags)) continue; - } + atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); md_sync_acct_bio(b, r10_bio->sectors); atomic_inc(&r10_bio->remaining); b->bi_next = NULL; @@ -5154,10 +5003,9 @@ static int handle_reshape_read_error(struct mddev *mddev, if (s > (PAGE_SIZE >> 9)) s = PAGE_SIZE >> 9; - rcu_read_lock(); while (!success) { int d = r10b->devs[slot].devnum; - struct md_rdev *rdev = rcu_dereference(conf->mirrors[d].rdev); + struct md_rdev *rdev = conf->mirrors[d].rdev; sector_t addr; if (rdev == NULL || test_bit(Faulty, &rdev->flags) || @@ -5166,14 +5014,12 @@ static int handle_reshape_read_error(struct mddev *mddev, addr = r10b->devs[slot].addr + idx * PAGE_SIZE; atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); success = sync_page_io(rdev, addr, s << 9, pages[idx], REQ_OP_READ, false); rdev_dec_pending(rdev, mddev); - rcu_read_lock(); if (success) break; failed: @@ -5183,7 +5029,6 @@ static int handle_reshape_read_error(struct mddev *mddev, if (slot == first_slot) break; } - rcu_read_unlock(); if (!success) { /* couldn't read this block, must give up */ set_bit(MD_RECOVERY_INTR, @@ -5209,12 +5054,8 @@ static void end_reshape_write(struct bio *bio) struct md_rdev *rdev = NULL; d = find_bio_disk(conf, r10_bio, bio, &slot, &repl); - if (repl) - rdev = conf->mirrors[d].replacement; - if (!rdev) { - smp_mb(); - rdev = conf->mirrors[d].rdev; - } + rdev = repl ? conf->mirrors[d].replacement : + conf->mirrors[d].rdev; if (bio->bi_status) { /* FIXME should record badblock */ @@ -5249,18 +5090,16 @@ static void raid10_finish_reshape(struct mddev *mddev) mddev->resync_max_sectors = mddev->array_sectors; } else { int d; - rcu_read_lock(); for (d = conf->geo.raid_disks ; d < conf->geo.raid_disks - mddev->delta_disks; d++) { - struct md_rdev *rdev = rcu_dereference(conf->mirrors[d].rdev); + struct md_rdev *rdev = conf->mirrors[d].rdev; if (rdev) clear_bit(In_sync, &rdev->flags); - rdev = rcu_dereference(conf->mirrors[d].replacement); + rdev = conf->mirrors[d].replacement; if (rdev) clear_bit(In_sync, &rdev->flags); } - rcu_read_unlock(); } mddev->layout = mddev->new_layout; mddev->chunk_sectors = 1 << conf->geo.chunk_shift; diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c index 6157f5beb9fe..874874fe4fa1 100644 --- a/drivers/md/raid5-cache.c +++ b/drivers/md/raid5-cache.c @@ -1890,28 +1890,22 @@ r5l_recovery_replay_one_stripe(struct r5conf *conf, continue; /* in case device is broken */ - rcu_read_lock(); - rdev = rcu_dereference(conf->disks[disk_index].rdev); + rdev = conf->disks[disk_index].rdev; if (rdev) { atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); sync_page_io(rdev, sh->sector, PAGE_SIZE, sh->dev[disk_index].page, REQ_OP_WRITE, false); rdev_dec_pending(rdev, rdev->mddev); - rcu_read_lock(); } - rrdev = rcu_dereference(conf->disks[disk_index].replacement); + rrdev = conf->disks[disk_index].replacement; if (rrdev) { atomic_inc(&rrdev->nr_pending); - rcu_read_unlock(); sync_page_io(rrdev, sh->sector, PAGE_SIZE, sh->dev[disk_index].page, REQ_OP_WRITE, false); rdev_dec_pending(rrdev, rrdev->mddev); - rcu_read_lock(); } - rcu_read_unlock(); } ctx->data_parity_stripes++; out: @@ -2948,7 +2942,6 @@ bool r5c_big_stripe_cached(struct r5conf *conf, sector_t sect) if (!log) return false; - WARN_ON_ONCE(!rcu_read_lock_held()); tree_index = r5c_tree_index(conf, sect); slot = radix_tree_lookup(&log->big_stripe_tree, tree_index); return slot != NULL; diff --git a/drivers/md/raid5-ppl.c b/drivers/md/raid5-ppl.c index eaea57aee602..da4ba736c4f0 100644 --- a/drivers/md/raid5-ppl.c +++ b/drivers/md/raid5-ppl.c @@ -620,11 +620,9 @@ static void ppl_do_flush(struct ppl_io_unit *io) struct md_rdev *rdev; struct block_device *bdev = NULL; - rcu_read_lock(); - rdev = rcu_dereference(conf->disks[i].rdev); + rdev = conf->disks[i].rdev; if (rdev && !test_bit(Faulty, &rdev->flags)) bdev = rdev->bdev; - rcu_read_unlock(); if (bdev) { struct bio *bio; @@ -882,9 +880,7 @@ static int ppl_recover_entry(struct ppl_log *log, struct ppl_header_entry *e, (unsigned long long)r_sector, dd_idx, (unsigned long long)sector); - /* Array has not started so rcu dereference is safe */ - rdev = rcu_dereference_protected( - conf->disks[dd_idx].rdev, 1); + rdev = conf->disks[dd_idx].rdev; if (!rdev || (!test_bit(In_sync, &rdev->flags) && sector >= rdev->recovery_offset)) { pr_debug("%s:%*s data member disk %d missing\n", @@ -936,9 +932,7 @@ static int ppl_recover_entry(struct ppl_log *log, struct ppl_header_entry *e, 0, &disk, &sh); BUG_ON(sh.pd_idx != le32_to_cpu(e->parity_disk)); - /* Array has not started so rcu dereference is safe */ - parity_rdev = rcu_dereference_protected( - conf->disks[sh.pd_idx].rdev, 1); + parity_rdev = conf->disks[sh.pd_idx].rdev; BUG_ON(parity_rdev->bdev->bd_dev != log->rdev->bdev->bd_dev); pr_debug("%s:%*s write parity at sector %llu, disk %pg\n", @@ -1404,9 +1398,7 @@ int ppl_init_log(struct r5conf *conf) for (i = 0; i < ppl_conf->count; i++) { struct ppl_log *log = &ppl_conf->child_logs[i]; - /* Array has not started so rcu dereference is safe */ - struct md_rdev *rdev = - rcu_dereference_protected(conf->disks[i].rdev, 1); + struct md_rdev *rdev = conf->disks[i].rdev; mutex_init(&log->io_mutex); spin_lock_init(&log->io_list_lock); diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 26e1e8a5e941..6a7a32f7fb91 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -36,7 +36,6 @@ */ #include <linux/blkdev.h> -#include <linux/delay.h> #include <linux/kthread.h> #include <linux/raid/pq.h> #include <linux/async_tx.h> @@ -694,12 +693,12 @@ int raid5_calc_degraded(struct r5conf *conf) int degraded, degraded2; int i; - rcu_read_lock(); degraded = 0; for (i = 0; i < conf->previous_raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->disks[i].rdev); + struct md_rdev *rdev = READ_ONCE(conf->disks[i].rdev); + if (rdev && test_bit(Faulty, &rdev->flags)) - rdev = rcu_dereference(conf->disks[i].replacement); + rdev = READ_ONCE(conf->disks[i].replacement); if (!rdev || test_bit(Faulty, &rdev->flags)) degraded++; else if (test_bit(In_sync, &rdev->flags)) @@ -717,15 +716,14 @@ int raid5_calc_degraded(struct r5conf *conf) if (conf->raid_disks >= conf->previous_raid_disks) degraded++; } - rcu_read_unlock(); if (conf->raid_disks == conf->previous_raid_disks) return degraded; - rcu_read_lock(); degraded2 = 0; for (i = 0; i < conf->raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->disks[i].rdev); + struct md_rdev *rdev = READ_ONCE(conf->disks[i].rdev); + if (rdev && test_bit(Faulty, &rdev->flags)) - rdev = rcu_dereference(conf->disks[i].replacement); + rdev = READ_ONCE(conf->disks[i].replacement); if (!rdev || test_bit(Faulty, &rdev->flags)) degraded2++; else if (test_bit(In_sync, &rdev->flags)) @@ -739,7 +737,6 @@ int raid5_calc_degraded(struct r5conf *conf) if (conf->raid_disks <= conf->previous_raid_disks) degraded2++; } - rcu_read_unlock(); if (degraded2 > degraded) return degraded2; return degraded; @@ -1184,14 +1181,8 @@ again: bi = &dev->req; rbi = &dev->rreq; /* For writing to replacement */ - rcu_read_lock(); - rrdev = rcu_dereference(conf->disks[i].replacement); - smp_mb(); /* Ensure that if rrdev is NULL, rdev won't be */ - rdev = rcu_dereference(conf->disks[i].rdev); - if (!rdev) { - rdev = rrdev; - rrdev = NULL; - } + rdev = conf->disks[i].rdev; + rrdev = conf->disks[i].replacement; if (op_is_write(op)) { if (replace_only) rdev = NULL; @@ -1212,7 +1203,6 @@ again: rrdev = NULL; if (rrdev) atomic_inc(&rrdev->nr_pending); - rcu_read_unlock(); /* We have already checked bad blocks for reads. Now * need to check for writes. We never accept write errors @@ -2731,28 +2721,6 @@ static void shrink_stripes(struct r5conf *conf) conf->slab_cache = NULL; } -/* - * This helper wraps rcu_dereference_protected() and can be used when - * it is known that the nr_pending of the rdev is elevated. - */ -static struct md_rdev *rdev_pend_deref(struct md_rdev __rcu *rdev) -{ - return rcu_dereference_protected(rdev, - atomic_read(&rcu_access_pointer(rdev)->nr_pending)); -} - -/* - * This helper wraps rcu_dereference_protected() and should be used - * when it is known that the mddev_lock() is held. This is safe - * seeing raid5_remove_disk() has the same lock held. - */ -static struct md_rdev *rdev_mdlock_deref(struct mddev *mddev, - struct md_rdev __rcu *rdev) -{ - return rcu_dereference_protected(rdev, - lockdep_is_held(&mddev->reconfig_mutex)); -} - static void raid5_end_read_request(struct bio * bi) { struct stripe_head *sh = bi->bi_private; @@ -2778,9 +2746,9 @@ static void raid5_end_read_request(struct bio * bi) * In that case it moved down to 'rdev'. * rdev is not removed until all requests are finished. */ - rdev = rdev_pend_deref(conf->disks[i].replacement); + rdev = conf->disks[i].replacement; if (!rdev) - rdev = rdev_pend_deref(conf->disks[i].rdev); + rdev = conf->disks[i].rdev; if (use_new_offset(conf, sh)) s = sh->sector + rdev->new_data_offset; @@ -2893,11 +2861,11 @@ static void raid5_end_write_request(struct bio *bi) for (i = 0 ; i < disks; i++) { if (bi == &sh->dev[i].req) { - rdev = rdev_pend_deref(conf->disks[i].rdev); + rdev = conf->disks[i].rdev; break; } if (bi == &sh->dev[i].rreq) { - rdev = rdev_pend_deref(conf->disks[i].replacement); + rdev = conf->disks[i].replacement; if (rdev) replacement = 1; else @@ -2905,7 +2873,7 @@ static void raid5_end_write_request(struct bio *bi) * replaced it. rdev is not removed * until all requests are finished. */ - rdev = rdev_pend_deref(conf->disks[i].rdev); + rdev = conf->disks[i].rdev; break; } } @@ -3667,15 +3635,13 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh, int bitmap_end = 0; if (test_bit(R5_ReadError, &sh->dev[i].flags)) { - struct md_rdev *rdev; - rcu_read_lock(); - rdev = rcu_dereference(conf->disks[i].rdev); + struct md_rdev *rdev = conf->disks[i].rdev; + if (rdev && test_bit(In_sync, &rdev->flags) && !test_bit(Faulty, &rdev->flags)) atomic_inc(&rdev->nr_pending); else rdev = NULL; - rcu_read_unlock(); if (rdev) { if (!rdev_set_badblocks( rdev, @@ -3793,16 +3759,17 @@ handle_failed_sync(struct r5conf *conf, struct stripe_head *sh, /* During recovery devices cannot be removed, so * locking and refcounting of rdevs is not needed */ - rcu_read_lock(); for (i = 0; i < conf->raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->disks[i].rdev); + struct md_rdev *rdev = conf->disks[i].rdev; + if (rdev && !test_bit(Faulty, &rdev->flags) && !test_bit(In_sync, &rdev->flags) && !rdev_set_badblocks(rdev, sh->sector, RAID5_STRIPE_SECTORS(conf), 0)) abort = 1; - rdev = rcu_dereference(conf->disks[i].replacement); + rdev = conf->disks[i].replacement; + if (rdev && !test_bit(Faulty, &rdev->flags) && !test_bit(In_sync, &rdev->flags) @@ -3810,7 +3777,6 @@ handle_failed_sync(struct r5conf *conf, struct stripe_head *sh, RAID5_STRIPE_SECTORS(conf), 0)) abort = 1; } - rcu_read_unlock(); if (abort) conf->recovery_disabled = conf->mddev->recovery_disabled; @@ -3823,15 +3789,13 @@ static int want_replace(struct stripe_head *sh, int disk_idx) struct md_rdev *rdev; int rv = 0; - rcu_read_lock(); - rdev = rcu_dereference(sh->raid_conf->disks[disk_idx].replacement); + rdev = sh->raid_conf->disks[disk_idx].replacement; if (rdev && !test_bit(Faulty, &rdev->flags) && !test_bit(In_sync, &rdev->flags) && (rdev->recovery_offset <= sh->sector || rdev->mddev->recovery_cp <= sh->sector)) rv = 1; - rcu_read_unlock(); return rv; } @@ -4708,7 +4672,6 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s) s->log_failed = r5l_log_disk_error(conf); /* Now to look around and see what can be done */ - rcu_read_lock(); for (i=disks; i--; ) { struct md_rdev *rdev; sector_t first_bad; @@ -4753,7 +4716,7 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s) /* Prefer to use the replacement for reads, but only * if it is recovered enough and has no bad blocks. */ - rdev = rcu_dereference(conf->disks[i].replacement); + rdev = conf->disks[i].replacement; if (rdev && !test_bit(Faulty, &rdev->flags) && rdev->recovery_offset >= sh->sector + RAID5_STRIPE_SECTORS(conf) && !is_badblock(rdev, sh->sector, RAID5_STRIPE_SECTORS(conf), @@ -4764,7 +4727,7 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s) set_bit(R5_NeedReplace, &dev->flags); else clear_bit(R5_NeedReplace, &dev->flags); - rdev = rcu_dereference(conf->disks[i].rdev); + rdev = conf->disks[i].rdev; clear_bit(R5_ReadRepl, &dev->flags); } if (rdev && test_bit(Faulty, &rdev->flags)) @@ -4811,8 +4774,8 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s) if (test_bit(R5_WriteError, &dev->flags)) { /* This flag does not apply to '.replacement' * only to .rdev, so make sure to check that*/ - struct md_rdev *rdev2 = rcu_dereference( - conf->disks[i].rdev); + struct md_rdev *rdev2 = conf->disks[i].rdev; + if (rdev2 == rdev) clear_bit(R5_Insync, &dev->flags); if (rdev2 && !test_bit(Faulty, &rdev2->flags)) { @@ -4824,8 +4787,8 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s) if (test_bit(R5_MadeGood, &dev->flags)) { /* This flag does not apply to '.replacement' * only to .rdev, so make sure to check that*/ - struct md_rdev *rdev2 = rcu_dereference( - conf->disks[i].rdev); + struct md_rdev *rdev2 = conf->disks[i].rdev; + if (rdev2 && !test_bit(Faulty, &rdev2->flags)) { s->handle_bad_blocks = 1; atomic_inc(&rdev2->nr_pending); @@ -4833,8 +4796,8 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s) clear_bit(R5_MadeGood, &dev->flags); } if (test_bit(R5_MadeGoodRepl, &dev->flags)) { - struct md_rdev *rdev2 = rcu_dereference( - conf->disks[i].replacement); + struct md_rdev *rdev2 = conf->disks[i].replacement; + if (rdev2 && !test_bit(Faulty, &rdev2->flags)) { s->handle_bad_blocks = 1; atomic_inc(&rdev2->nr_pending); @@ -4855,8 +4818,7 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s) if (rdev && !test_bit(Faulty, &rdev->flags)) do_recovery = 1; else if (!rdev) { - rdev = rcu_dereference( - conf->disks[i].replacement); + rdev = conf->disks[i].replacement; if (rdev && !test_bit(Faulty, &rdev->flags)) do_recovery = 1; } @@ -4883,7 +4845,6 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s) else s->replacing = 1; } - rcu_read_unlock(); } /* @@ -5340,23 +5301,23 @@ finish: struct r5dev *dev = &sh->dev[i]; if (test_and_clear_bit(R5_WriteError, &dev->flags)) { /* We own a safe reference to the rdev */ - rdev = rdev_pend_deref(conf->disks[i].rdev); + rdev = conf->disks[i].rdev; if (!rdev_set_badblocks(rdev, sh->sector, RAID5_STRIPE_SECTORS(conf), 0)) md_error(conf->mddev, rdev); rdev_dec_pending(rdev, conf->mddev); } if (test_and_clear_bit(R5_MadeGood, &dev->flags)) { - rdev = rdev_pend_deref(conf->disks[i].rdev); + rdev = conf->disks[i].rdev; rdev_clear_badblocks(rdev, sh->sector, RAID5_STRIPE_SECTORS(conf), 0); rdev_dec_pending(rdev, conf->mddev); } if (test_and_clear_bit(R5_MadeGoodRepl, &dev->flags)) { - rdev = rdev_pend_deref(conf->disks[i].replacement); + rdev = conf->disks[i].replacement; if (!rdev) /* rdev have been moved down */ - rdev = rdev_pend_deref(conf->disks[i].rdev); + rdev = conf->disks[i].rdev; rdev_clear_badblocks(rdev, sh->sector, RAID5_STRIPE_SECTORS(conf), 0); rdev_dec_pending(rdev, conf->mddev); @@ -5515,24 +5476,22 @@ static int raid5_read_one_chunk(struct mddev *mddev, struct bio *raid_bio) &dd_idx, NULL); end_sector = sector + bio_sectors(raid_bio); - rcu_read_lock(); if (r5c_big_stripe_cached(conf, sector)) - goto out_rcu_unlock; + return 0; - rdev = rcu_dereference(conf->disks[dd_idx].replacement); + rdev = conf->disks[dd_idx].replacement; if (!rdev || test_bit(Faulty, &rdev->flags) || rdev->recovery_offset < end_sector) { - rdev = rcu_dereference(conf->disks[dd_idx].rdev); + rdev = conf->disks[dd_idx].rdev; if (!rdev) - goto out_rcu_unlock; + return 0; if (test_bit(Faulty, &rdev->flags) || !(test_bit(In_sync, &rdev->flags) || rdev->recovery_offset >= end_sector)) - goto out_rcu_unlock; + return 0; } atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); if (is_badblock(rdev, sector, bio_sectors(raid_bio), &first_bad, &bad_sectors)) { @@ -5576,10 +5535,6 @@ static int raid5_read_one_chunk(struct mddev *mddev, struct bio *raid_bio) raid_bio->bi_iter.bi_sector); submit_bio_noacct(align_bio); return 1; - -out_rcu_unlock: - rcu_read_unlock(); - return 0; } static struct bio *chunk_aligned_read(struct mddev *mddev, struct bio *raid_bio) @@ -6582,14 +6537,12 @@ static inline sector_t raid5_sync_request(struct mddev *mddev, sector_t sector_n * Note in case of > 1 drive failures it's possible we're rebuilding * one drive while leaving another faulty drive in array. */ - rcu_read_lock(); for (i = 0; i < conf->raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->disks[i].rdev); + struct md_rdev *rdev = conf->disks[i].rdev; if (rdev == NULL || test_bit(Faulty, &rdev->flags)) still_degraded = 1; } - rcu_read_unlock(); md_bitmap_start_sync(mddev->bitmap, sector_nr, &sync_blocks, still_degraded); @@ -6820,18 +6773,7 @@ static void raid5d(struct md_thread *thread) spin_unlock_irq(&conf->device_lock); md_check_recovery(mddev); spin_lock_irq(&conf->device_lock); - - /* - * Waiting on MD_SB_CHANGE_PENDING below may deadlock - * seeing md_check_recovery() is needed to clear - * the flag when using mdmon. - */ - continue; } - - wait_event_lock_irq(mddev->sb_wait, - !test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags), - conf->device_lock); } pr_debug("%d stripes handled\n", handled); @@ -7911,18 +7853,10 @@ static int raid5_run(struct mddev *mddev) for (i = 0; i < conf->raid_disks && conf->previous_raid_disks; i++) { - rdev = rdev_mdlock_deref(mddev, conf->disks[i].rdev); - if (!rdev && conf->disks[i].replacement) { - /* The replacement is all we have yet */ - rdev = rdev_mdlock_deref(mddev, - conf->disks[i].replacement); - conf->disks[i].replacement = NULL; - clear_bit(Replacement, &rdev->flags); - rcu_assign_pointer(conf->disks[i].rdev, rdev); - } + rdev = conf->disks[i].rdev; if (!rdev) continue; - if (rcu_access_pointer(conf->disks[i].replacement) && + if (conf->disks[i].replacement && conf->reshape_progress != MaxSector) { /* replacements and reshape simply do not mix. */ pr_warn("md: cannot handle concurrent replacement and reshape.\n"); @@ -8002,11 +7936,7 @@ static int raid5_run(struct mddev *mddev) clear_bit(MD_RECOVERY_SYNC, &mddev->recovery); clear_bit(MD_RECOVERY_CHECK, &mddev->recovery); set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); - set_bit(MD_RECOVERY_RUNNING, &mddev->recovery); - rcu_assign_pointer(mddev->sync_thread, - md_register_thread(md_do_sync, mddev, "reshape")); - if (!mddev->sync_thread) - goto abort; + set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); } /* Ok, everything is just fine now */ @@ -8106,15 +8036,16 @@ static void raid5_status(struct seq_file *seq, struct mddev *mddev) struct r5conf *conf = mddev->private; int i; + lockdep_assert_held(&mddev->lock); + seq_printf(seq, " level %d, %dk chunk, algorithm %d", mddev->level, conf->chunk_sectors / 2, mddev->layout); seq_printf (seq, " [%d/%d] [", conf->raid_disks, conf->raid_disks - mddev->degraded); - rcu_read_lock(); for (i = 0; i < conf->raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->disks[i].rdev); + struct md_rdev *rdev = READ_ONCE(conf->disks[i].rdev); + seq_printf (seq, "%s", rdev && test_bit(In_sync, &rdev->flags) ? "U" : "_"); } - rcu_read_unlock(); seq_printf (seq, "]"); } @@ -8152,9 +8083,8 @@ static int raid5_spare_active(struct mddev *mddev) unsigned long flags; for (i = 0; i < conf->raid_disks; i++) { - rdev = rdev_mdlock_deref(mddev, conf->disks[i].rdev); - replacement = rdev_mdlock_deref(mddev, - conf->disks[i].replacement); + rdev = conf->disks[i].rdev; + replacement = conf->disks[i].replacement; if (replacement && replacement->recovery_offset == MaxSector && !test_bit(Faulty, &replacement->flags) @@ -8193,7 +8123,7 @@ static int raid5_remove_disk(struct mddev *mddev, struct md_rdev *rdev) struct r5conf *conf = mddev->private; int err = 0; int number = rdev->raid_disk; - struct md_rdev __rcu **rdevp; + struct md_rdev **rdevp; struct disk_info *p; struct md_rdev *tmp; @@ -8216,9 +8146,9 @@ static int raid5_remove_disk(struct mddev *mddev, struct md_rdev *rdev) if (unlikely(number >= conf->pool_size)) return 0; p = conf->disks + number; - if (rdev == rcu_access_pointer(p->rdev)) + if (rdev == p->rdev) rdevp = &p->rdev; - else if (rdev == rcu_access_pointer(p->replacement)) + else if (rdev == p->replacement) rdevp = &p->replacement; else return 0; @@ -8238,37 +8168,24 @@ static int raid5_remove_disk(struct mddev *mddev, struct md_rdev *rdev) if (!test_bit(Faulty, &rdev->flags) && mddev->recovery_disabled != conf->recovery_disabled && !has_failed(conf) && - (!rcu_access_pointer(p->replacement) || - rcu_access_pointer(p->replacement) == rdev) && + (!p->replacement || p->replacement == rdev) && number < conf->raid_disks) { err = -EBUSY; goto abort; } - *rdevp = NULL; - if (!test_bit(RemoveSynchronized, &rdev->flags)) { - lockdep_assert_held(&mddev->reconfig_mutex); - synchronize_rcu(); - if (atomic_read(&rdev->nr_pending)) { - /* lost the race, try later */ - err = -EBUSY; - rcu_assign_pointer(*rdevp, rdev); - } - } + WRITE_ONCE(*rdevp, NULL); if (!err) { err = log_modify(conf, rdev, false); if (err) goto abort; } - tmp = rcu_access_pointer(p->replacement); + tmp = p->replacement; if (tmp) { /* We must have just cleared 'rdev' */ - rcu_assign_pointer(p->rdev, tmp); + WRITE_ONCE(p->rdev, tmp); clear_bit(Replacement, &tmp->flags); - smp_mb(); /* Make sure other CPUs may see both as identical - * but will never see neither - if they are careful - */ - rcu_assign_pointer(p->replacement, NULL); + WRITE_ONCE(p->replacement, NULL); if (!err) err = log_modify(conf, tmp, true); @@ -8336,7 +8253,7 @@ static int raid5_add_disk(struct mddev *mddev, struct md_rdev *rdev) rdev->raid_disk = disk; if (rdev->saved_raid_disk != disk) conf->fullsync = 1; - rcu_assign_pointer(p->rdev, rdev); + WRITE_ONCE(p->rdev, rdev); err = log_modify(conf, rdev, true); @@ -8345,7 +8262,7 @@ static int raid5_add_disk(struct mddev *mddev, struct md_rdev *rdev) } for (disk = first; disk <= last; disk++) { p = conf->disks + disk; - tmp = rdev_mdlock_deref(mddev, p->rdev); + tmp = p->rdev; if (test_bit(WantReplacement, &tmp->flags) && mddev->reshape_position == MaxSector && p->replacement == NULL) { @@ -8354,7 +8271,7 @@ static int raid5_add_disk(struct mddev *mddev, struct md_rdev *rdev) rdev->raid_disk = disk; err = 0; conf->fullsync = 1; - rcu_assign_pointer(p->replacement, rdev); + WRITE_ONCE(p->replacement, rdev); break; } } @@ -8487,7 +8404,7 @@ static int raid5_start_reshape(struct mddev *mddev) if (mddev->recovery_cp < MaxSector) return -EBUSY; for (i = 0; i < conf->raid_disks; i++) - if (rdev_mdlock_deref(mddev, conf->disks[i].replacement)) + if (conf->disks[i].replacement) return -EBUSY; rdev_for_each(rdev, mddev) { @@ -8585,29 +8502,8 @@ static int raid5_start_reshape(struct mddev *mddev) clear_bit(MD_RECOVERY_CHECK, &mddev->recovery); clear_bit(MD_RECOVERY_DONE, &mddev->recovery); set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); - set_bit(MD_RECOVERY_RUNNING, &mddev->recovery); - rcu_assign_pointer(mddev->sync_thread, - md_register_thread(md_do_sync, mddev, "reshape")); - if (!mddev->sync_thread) { - mddev->recovery = 0; - spin_lock_irq(&conf->device_lock); - write_seqcount_begin(&conf->gen_lock); - mddev->raid_disks = conf->raid_disks = conf->previous_raid_disks; - mddev->new_chunk_sectors = - conf->chunk_sectors = conf->prev_chunk_sectors; - mddev->new_layout = conf->algorithm = conf->prev_algo; - rdev_for_each(rdev, mddev) - rdev->new_data_offset = rdev->data_offset; - smp_wmb(); - conf->generation --; - conf->reshape_progress = MaxSector; - mddev->reshape_position = MaxSector; - write_seqcount_end(&conf->gen_lock); - spin_unlock_irq(&conf->device_lock); - return -EAGAIN; - } + set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); conf->reshape_checkpoint = jiffies; - md_wakeup_thread(mddev->sync_thread); md_new_event(); return 0; } @@ -8658,12 +8554,10 @@ static void raid5_finish_reshape(struct mddev *mddev) for (d = conf->raid_disks ; d < conf->raid_disks - mddev->delta_disks; d++) { - rdev = rdev_mdlock_deref(mddev, - conf->disks[d].rdev); + rdev = conf->disks[d].rdev; if (rdev) clear_bit(In_sync, &rdev->flags); - rdev = rdev_mdlock_deref(mddev, - conf->disks[d].replacement); + rdev = conf->disks[d].replacement; if (rdev) clear_bit(In_sync, &rdev->flags); } diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h index 22bea20eccbd..9b5a7dc3f2a0 100644 --- a/drivers/md/raid5.h +++ b/drivers/md/raid5.h @@ -473,8 +473,8 @@ enum { */ struct disk_info { - struct md_rdev __rcu *rdev; - struct md_rdev __rcu *replacement; + struct md_rdev *rdev; + struct md_rdev *replacement; struct page *extra_page; /* extra page to use in prexor */ }; |