diff options
Diffstat (limited to 'drivers/md')
66 files changed, 1113 insertions, 1073 deletions
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig index 2c8ac3688815..edff083f7c4e 100644 --- a/drivers/md/Kconfig +++ b/drivers/md/Kconfig @@ -201,7 +201,7 @@ config BLK_DEV_DM_BUILTIN config BLK_DEV_DM tristate "Device mapper support" select BLK_DEV_DM_BUILTIN - select DAX + depends on DAX || DAX=n ---help--- Device-mapper is a low level volume manager. It works by allowing people to specify mappings for ranges of logical sectors. Various diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c index 004cc3cc6123..7fa2631b422c 100644 --- a/drivers/md/bcache/alloc.c +++ b/drivers/md/bcache/alloc.c @@ -290,7 +290,7 @@ do { \ if (kthread_should_stop() || \ test_bit(CACHE_SET_IO_DISABLE, &ca->set->flags)) { \ set_current_state(TASK_RUNNING); \ - return 0; \ + goto out; \ } \ \ schedule(); \ @@ -378,6 +378,9 @@ retry_invalidate: bch_prio_write(ca); } } +out: + wait_for_kthread_stop(); + return 0; } /* Allocation */ diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h index d338b7086013..d6bf294f3907 100644 --- a/drivers/md/bcache/bcache.h +++ b/drivers/md/bcache/bcache.h @@ -269,7 +269,7 @@ struct bcache_device { atomic_t *stripe_sectors_dirty; unsigned long *full_dirty_stripes; - struct bio_set *bio_split; + struct bio_set bio_split; unsigned data_csum:1; @@ -345,6 +345,7 @@ struct cached_dev { struct keybuf writeback_keys; + struct task_struct *status_update_thread; /* * Order the write-half of writeback operations strongly in dispatch * order. (Maintain LBA order; don't allow reads completing out of @@ -392,6 +393,9 @@ struct cached_dev { #define DEFAULT_CACHED_DEV_ERROR_LIMIT 64 atomic_t io_errors; unsigned error_limit; + unsigned offline_seconds; + + char backing_dev_name[BDEVNAME_SIZE]; }; enum alloc_reserve { @@ -464,6 +468,8 @@ struct cache { atomic_long_t meta_sectors_written; atomic_long_t btree_sectors_written; atomic_long_t sectors_written; + + char cache_dev_name[BDEVNAME_SIZE]; }; struct gc_stat { @@ -524,9 +530,9 @@ struct cache_set { struct closure sb_write; struct semaphore sb_write_mutex; - mempool_t *search; - mempool_t *bio_meta; - struct bio_set *bio_split; + mempool_t search; + mempool_t bio_meta; + struct bio_set bio_split; /* For the btree cache */ struct shrinker shrink; @@ -651,7 +657,7 @@ struct cache_set { * A btree node on disk could have too many bsets for an iterator to fit * on the stack - have to dynamically allocate them */ - mempool_t *fill_iter; + mempool_t fill_iter; struct bset_sort_state sort; @@ -952,8 +958,6 @@ void bch_prio_write(struct cache *); void bch_write_bdev_super(struct cached_dev *, struct closure *); extern struct workqueue_struct *bcache_wq; -extern const char * const bch_cache_modes[]; -extern const char * const bch_stop_on_failure_modes[]; extern struct mutex bch_register_lock; extern struct list_head bch_cache_sets; diff --git a/drivers/md/bcache/bset.c b/drivers/md/bcache/bset.c index 579c696a5fe0..f3403b45bc28 100644 --- a/drivers/md/bcache/bset.c +++ b/drivers/md/bcache/bset.c @@ -1118,8 +1118,7 @@ struct bkey *bch_btree_iter_next_filter(struct btree_iter *iter, void bch_bset_sort_state_free(struct bset_sort_state *state) { - if (state->pool) - mempool_destroy(state->pool); + mempool_exit(&state->pool); } int bch_bset_sort_state_init(struct bset_sort_state *state, unsigned page_order) @@ -1129,11 +1128,7 @@ int bch_bset_sort_state_init(struct bset_sort_state *state, unsigned page_order) state->page_order = page_order; state->crit_factor = int_sqrt(1 << page_order); - state->pool = mempool_create_page_pool(1, page_order); - if (!state->pool) - return -ENOMEM; - - return 0; + return mempool_init_page_pool(&state->pool, 1, page_order); } EXPORT_SYMBOL(bch_bset_sort_state_init); @@ -1191,7 +1186,7 @@ static void __btree_sort(struct btree_keys *b, struct btree_iter *iter, BUG_ON(order > state->page_order); - outp = mempool_alloc(state->pool, GFP_NOIO); + outp = mempool_alloc(&state->pool, GFP_NOIO); out = page_address(outp); used_mempool = true; order = state->page_order; @@ -1220,7 +1215,7 @@ static void __btree_sort(struct btree_keys *b, struct btree_iter *iter, } if (used_mempool) - mempool_free(virt_to_page(out), state->pool); + mempool_free(virt_to_page(out), &state->pool); else free_pages((unsigned long) out, order); diff --git a/drivers/md/bcache/bset.h b/drivers/md/bcache/bset.h index 0c24280f3b98..b867f2200495 100644 --- a/drivers/md/bcache/bset.h +++ b/drivers/md/bcache/bset.h @@ -347,7 +347,7 @@ static inline struct bkey *bch_bset_search(struct btree_keys *b, /* Sorting */ struct bset_sort_state { - mempool_t *pool; + mempool_t pool; unsigned page_order; unsigned crit_factor; diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c index 17936b2dc7d6..2a0968c04e21 100644 --- a/drivers/md/bcache/btree.c +++ b/drivers/md/bcache/btree.c @@ -204,7 +204,7 @@ void bch_btree_node_read_done(struct btree *b) struct bset *i = btree_bset_first(b); struct btree_iter *iter; - iter = mempool_alloc(b->c->fill_iter, GFP_NOIO); + iter = mempool_alloc(&b->c->fill_iter, GFP_NOIO); iter->size = b->c->sb.bucket_size / b->c->sb.block_size; iter->used = 0; @@ -271,7 +271,7 @@ void bch_btree_node_read_done(struct btree *b) bch_bset_init_next(&b->keys, write_block(b), bset_magic(&b->c->sb)); out: - mempool_free(iter, b->c->fill_iter); + mempool_free(iter, &b->c->fill_iter); return; err: set_btree_node_io_error(b); diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c index 028f7b386e01..d030ce3025a6 100644 --- a/drivers/md/bcache/debug.c +++ b/drivers/md/bcache/debug.c @@ -106,7 +106,6 @@ void bch_btree_verify(struct btree *b) void bch_data_verify(struct cached_dev *dc, struct bio *bio) { - char name[BDEVNAME_SIZE]; struct bio *check; struct bio_vec bv, cbv; struct bvec_iter iter, citer = { 0 }; @@ -134,7 +133,7 @@ void bch_data_verify(struct cached_dev *dc, struct bio *bio) bv.bv_len), dc->disk.c, "verify failed at dev %s sector %llu", - bdevname(dc->bdev, name), + dc->backing_dev_name, (uint64_t) bio->bi_iter.bi_sector); kunmap_atomic(p1); @@ -251,7 +250,9 @@ void bch_debug_exit(void) int __init bch_debug_init(struct kobject *kobj) { - bcache_debug = debugfs_create_dir("bcache", NULL); + if (!IS_ENABLED(CONFIG_DEBUG_FS)) + return 0; + bcache_debug = debugfs_create_dir("bcache", NULL); return IS_ERR_OR_NULL(bcache_debug); } diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c index 7fac97ae036e..9612873afee2 100644 --- a/drivers/md/bcache/io.c +++ b/drivers/md/bcache/io.c @@ -17,12 +17,12 @@ void bch_bbio_free(struct bio *bio, struct cache_set *c) { struct bbio *b = container_of(bio, struct bbio, bio); - mempool_free(b, c->bio_meta); + mempool_free(b, &c->bio_meta); } struct bio *bch_bbio_alloc(struct cache_set *c) { - struct bbio *b = mempool_alloc(c->bio_meta, GFP_NOIO); + struct bbio *b = mempool_alloc(&c->bio_meta, GFP_NOIO); struct bio *bio = &b->bio; bio_init(bio, bio->bi_inline_vecs, bucket_pages(c)); @@ -52,7 +52,6 @@ void bch_submit_bbio(struct bio *bio, struct cache_set *c, /* IO errors */ void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio) { - char buf[BDEVNAME_SIZE]; unsigned errors; WARN_ONCE(!dc, "NULL pointer of struct cached_dev"); @@ -60,7 +59,7 @@ void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio) errors = atomic_add_return(1, &dc->io_errors); if (errors < dc->error_limit) pr_err("%s: IO error on backing device, unrecoverable", - bio_devname(bio, buf)); + dc->backing_dev_name); else bch_cached_dev_error(dc); } @@ -105,19 +104,18 @@ void bch_count_io_errors(struct cache *ca, } if (error) { - char buf[BDEVNAME_SIZE]; unsigned errors = atomic_add_return(1 << IO_ERROR_SHIFT, &ca->io_errors); errors >>= IO_ERROR_SHIFT; if (errors < ca->set->error_limit) pr_err("%s: IO error on %s%s", - bdevname(ca->bdev, buf), m, + ca->cache_dev_name, m, is_read ? ", recovering." : "."); else bch_cache_set_error(ca->set, "%s: too many IO errors %s", - bdevname(ca->bdev, buf), m); + ca->cache_dev_name, m); } } diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index a65e3365eeb9..ae67f5fa8047 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -213,7 +213,7 @@ static void bch_data_insert_start(struct closure *cl) do { unsigned i; struct bkey *k; - struct bio_set *split = op->c->bio_split; + struct bio_set *split = &op->c->bio_split; /* 1 for the device pointer and 1 for the chksum */ if (bch_keylist_realloc(&op->insert_keys, @@ -548,7 +548,7 @@ static int cache_lookup_fn(struct btree_op *op, struct btree *b, struct bkey *k) n = bio_next_split(bio, min_t(uint64_t, INT_MAX, KEY_OFFSET(k) - bio->bi_iter.bi_sector), - GFP_NOIO, s->d->bio_split); + GFP_NOIO, &s->d->bio_split); bio_key = &container_of(n, struct bbio, bio)->key; bch_bkey_copy_single_ptr(bio_key, k, ptr); @@ -649,11 +649,8 @@ static void backing_request_endio(struct bio *bio) */ if (unlikely(s->iop.writeback && bio->bi_opf & REQ_PREFLUSH)) { - char buf[BDEVNAME_SIZE]; - - bio_devname(bio, buf); pr_err("Can't flush %s: returned bi_status %i", - buf, bio->bi_status); + dc->backing_dev_name, bio->bi_status); } else { /* set to orig_bio->bi_status in bio_complete() */ s->iop.status = bio->bi_status; @@ -710,7 +707,7 @@ static void search_free(struct closure *cl) bio_complete(s); closure_debug_destroy(cl); - mempool_free(s, s->d->c->search); + mempool_free(s, &s->d->c->search); } static inline struct search *search_alloc(struct bio *bio, @@ -718,7 +715,7 @@ static inline struct search *search_alloc(struct bio *bio, { struct search *s; - s = mempool_alloc(d->c->search, GFP_NOIO); + s = mempool_alloc(&d->c->search, GFP_NOIO); closure_init(&s->cl, NULL); do_bio_hook(s, bio, request_endio); @@ -867,7 +864,7 @@ static int cached_dev_cache_miss(struct btree *b, struct search *s, s->cache_missed = 1; if (s->cache_miss || s->iop.bypass) { - miss = bio_next_split(bio, sectors, GFP_NOIO, s->d->bio_split); + miss = bio_next_split(bio, sectors, GFP_NOIO, &s->d->bio_split); ret = miss == bio ? MAP_DONE : MAP_CONTINUE; goto out_submit; } @@ -890,14 +887,14 @@ static int cached_dev_cache_miss(struct btree *b, struct search *s, s->iop.replace = true; - miss = bio_next_split(bio, sectors, GFP_NOIO, s->d->bio_split); + miss = bio_next_split(bio, sectors, GFP_NOIO, &s->d->bio_split); /* btree_search_recurse()'s btree iterator is no good anymore */ ret = miss == bio ? MAP_DONE : -EINTR; cache_bio = bio_alloc_bioset(GFP_NOWAIT, DIV_ROUND_UP(s->insert_bio_sectors, PAGE_SECTORS), - dc->disk.bio_split); + &dc->disk.bio_split); if (!cache_bio) goto out_submit; @@ -1011,7 +1008,7 @@ static void cached_dev_write(struct cached_dev *dc, struct search *s) struct bio *flush; flush = bio_alloc_bioset(GFP_NOIO, 0, - dc->disk.bio_split); + &dc->disk.bio_split); if (!flush) { s->iop.status = BLK_STS_RESOURCE; goto insert_data; @@ -1024,7 +1021,7 @@ static void cached_dev_write(struct cached_dev *dc, struct search *s) closure_bio_submit(s->iop.c, flush, cl); } } else { - s->iop.bio = bio_clone_fast(bio, GFP_NOIO, dc->disk.bio_split); + s->iop.bio = bio_clone_fast(bio, GFP_NOIO, &dc->disk.bio_split); /* I/O request sent to backing device */ bio->bi_end_io = backing_request_endio; closure_bio_submit(s->iop.c, bio, cl); diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index d90d9e59ca00..a31e55bcc4e5 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -37,24 +37,6 @@ static const char invalid_uuid[] = { 0xc8, 0x50, 0xfc, 0x5e, 0xcb, 0x16, 0xcd, 0x99 }; -/* Default is -1; we skip past it for struct cached_dev's cache mode */ -const char * const bch_cache_modes[] = { - "default", - "writethrough", - "writeback", - "writearound", - "none", - NULL -}; - -/* Default is -1; we skip past it for stop_when_cache_set_failed */ -const char * const bch_stop_on_failure_modes[] = { - "default", - "auto", - "always", - NULL -}; - static struct kobject *bcache_kobj; struct mutex bch_register_lock; LIST_HEAD(bch_cache_sets); @@ -654,6 +636,11 @@ static int ioctl_dev(struct block_device *b, fmode_t mode, unsigned int cmd, unsigned long arg) { struct bcache_device *d = b->bd_disk->private_data; + struct cached_dev *dc = container_of(d, struct cached_dev, disk); + + if (dc->io_disable) + return -EIO; + return d->ioctl(d, mode, cmd, arg); } @@ -766,8 +753,7 @@ static void bcache_device_free(struct bcache_device *d) put_disk(d->disk); } - if (d->bio_split) - bioset_free(d->bio_split); + bioset_exit(&d->bio_split); kvfree(d->full_dirty_stripes); kvfree(d->stripe_sectors_dirty); @@ -809,9 +795,8 @@ static int bcache_device_init(struct bcache_device *d, unsigned block_size, if (idx < 0) return idx; - if (!(d->bio_split = bioset_create(4, offsetof(struct bbio, bio), - BIOSET_NEED_BVECS | - BIOSET_NEED_RESCUER)) || + if (bioset_init(&d->bio_split, 4, offsetof(struct bbio, bio), + BIOSET_NEED_BVECS|BIOSET_NEED_RESCUER) || !(d->disk = alloc_disk(BCACHE_MINORS))) { ida_simple_remove(&bcache_device_idx, idx); return -ENOMEM; @@ -864,6 +849,44 @@ static void calc_cached_dev_sectors(struct cache_set *c) c->cached_dev_sectors = sectors; } +#define BACKING_DEV_OFFLINE_TIMEOUT 5 +static int cached_dev_status_update(void *arg) +{ + struct cached_dev *dc = arg; + struct request_queue *q; + + /* + * If this delayed worker is stopping outside, directly quit here. + * dc->io_disable might be set via sysfs interface, so check it + * here too. + */ + while (!kthread_should_stop() && !dc->io_disable) { + q = bdev_get_queue(dc->bdev); + if (blk_queue_dying(q)) + dc->offline_seconds++; + else + dc->offline_seconds = 0; + + if (dc->offline_seconds >= BACKING_DEV_OFFLINE_TIMEOUT) { + pr_err("%s: device offline for %d seconds", + dc->backing_dev_name, + BACKING_DEV_OFFLINE_TIMEOUT); + pr_err("%s: disable I/O request due to backing " + "device offline", dc->disk.name); + dc->io_disable = true; + /* let others know earlier that io_disable is true */ + smp_mb(); + bcache_device_stop(&dc->disk); + break; + } + schedule_timeout_interruptible(HZ); + } + + wait_for_kthread_stop(); + return 0; +} + + void bch_cached_dev_run(struct cached_dev *dc) { struct bcache_device *d = &dc->disk; @@ -906,6 +929,14 @@ void bch_cached_dev_run(struct cached_dev *dc) if (sysfs_create_link(&d->kobj, &disk_to_dev(d->disk)->kobj, "dev") || sysfs_create_link(&disk_to_dev(d->disk)->kobj, &d->kobj, "bcache")) pr_debug("error creating sysfs link"); + + dc->status_update_thread = kthread_run(cached_dev_status_update, + dc, "bcache_status_update"); + if (IS_ERR(dc->status_update_thread)) { + pr_warn("failed to create bcache_status_update kthread, " + "continue to run without monitoring backing " + "device status"); + } } /* @@ -936,7 +967,6 @@ static void cancel_writeback_rate_update_dwork(struct cached_dev *dc) static void cached_dev_detach_finish(struct work_struct *w) { struct cached_dev *dc = container_of(w, struct cached_dev, detach); - char buf[BDEVNAME_SIZE]; struct closure cl; closure_init_stack(&cl); @@ -967,7 +997,7 @@ static void cached_dev_detach_finish(struct work_struct *w) mutex_unlock(&bch_register_lock); - pr_info("Caching disabled for %s", bdevname(dc->bdev, buf)); + pr_info("Caching disabled for %s", dc->backing_dev_name); /* Drop ref we took in cached_dev_detach() */ closure_put(&dc->disk.cl); @@ -999,29 +1029,28 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, { uint32_t rtime = cpu_to_le32(get_seconds()); struct uuid_entry *u; - char buf[BDEVNAME_SIZE]; struct cached_dev *exist_dc, *t; - bdevname(dc->bdev, buf); - if ((set_uuid && memcmp(set_uuid, c->sb.set_uuid, 16)) || (!set_uuid && memcmp(dc->sb.set_uuid, c->sb.set_uuid, 16))) return -ENOENT; if (dc->disk.c) { - pr_err("Can't attach %s: already attached", buf); + pr_err("Can't attach %s: already attached", + dc->backing_dev_name); return -EINVAL; } if (test_bit(CACHE_SET_STOPPING, &c->flags)) { - pr_err("Can't attach %s: shutting down", buf); + pr_err("Can't attach %s: shutting down", + dc->backing_dev_name); return -EINVAL; } if (dc->sb.block_size < c->sb.block_size) { /* Will die */ pr_err("Couldn't attach %s: block size less than set's block size", - buf); + dc->backing_dev_name); return -EINVAL; } @@ -1029,7 +1058,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, list_for_each_entry_safe(exist_dc, t, &c->cached_devs, list) { if (!memcmp(dc->sb.uuid, exist_dc->sb.uuid, 16)) { pr_err("Tried to attach %s but duplicate UUID already attached", - buf); + dc->backing_dev_name); return -EINVAL; } @@ -1047,13 +1076,15 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, if (!u) { if (BDEV_STATE(&dc->sb) == BDEV_STATE_DIRTY) { - pr_err("Couldn't find uuid for %s in set", buf); + pr_err("Couldn't find uuid for %s in set", + dc->backing_dev_name); return -ENOENT; } u = uuid_find_empty(c); if (!u) { - pr_err("Not caching %s, no room for UUID", buf); + pr_err("Not caching %s, no room for UUID", + dc->backing_dev_name); return -EINVAL; } } @@ -1112,7 +1143,8 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, up_write(&dc->writeback_lock); pr_info("Caching %s as %s on set %pU", - bdevname(dc->bdev, buf), dc->disk.disk->disk_name, + dc->backing_dev_name, + dc->disk.disk->disk_name, dc->disk.c->sb.set_uuid); return 0; } @@ -1138,6 +1170,8 @@ static void cached_dev_free(struct closure *cl) kthread_stop(dc->writeback_thread); if (dc->writeback_write_wq) destroy_workqueue(dc->writeback_write_wq); + if (!IS_ERR_OR_NULL(dc->status_update_thread)) + kthread_stop(dc->status_update_thread); if (atomic_read(&dc->running)) bd_unlink_disk_holder(dc->bdev, dc->disk.disk); @@ -1225,10 +1259,10 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page, struct block_device *bdev, struct cached_dev *dc) { - char name[BDEVNAME_SIZE]; const char *err = "cannot allocate memory"; struct cache_set *c; + bdevname(bdev, dc->backing_dev_name); memcpy(&dc->sb, sb, sizeof(struct cache_sb)); dc->bdev = bdev; dc->bdev->bd_holder = dc; @@ -1237,6 +1271,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page, bio_first_bvec_all(&dc->sb_bio)->bv_page = sb_page; get_page(sb_page); + if (cached_dev_init(dc, sb->block_size << 9)) goto err; @@ -1247,7 +1282,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page, if (bch_cache_accounting_add_kobjs(&dc->accounting, &dc->disk.kobj)) goto err; - pr_info("registered backing device %s", bdevname(bdev, name)); + pr_info("registered backing device %s", dc->backing_dev_name); list_add(&dc->list, &uncached_devices); list_for_each_entry(c, &bch_cache_sets, list) @@ -1259,7 +1294,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page, return; err: - pr_notice("error %s: %s", bdevname(bdev, name), err); + pr_notice("error %s: %s", dc->backing_dev_name, err); bcache_device_stop(&dc->disk); } @@ -1367,7 +1402,7 @@ int bch_flash_dev_create(struct cache_set *c, uint64_t size) bool bch_cached_dev_error(struct cached_dev *dc) { - char name[BDEVNAME_SIZE]; + struct cache_set *c; if (!dc || test_bit(BCACHE_DEV_CLOSING, &dc->disk.flags)) return false; @@ -1377,7 +1412,22 @@ bool bch_cached_dev_error(struct cached_dev *dc) smp_mb(); pr_err("stop %s: too many IO errors on backing device %s\n", - dc->disk.disk->disk_name, bdevname(dc->bdev, name)); + dc->disk.disk->disk_name, dc->backing_dev_name); + + /* + * If the cached device is still attached to a cache set, + * even dc->io_disable is true and no more I/O requests + * accepted, cache device internal I/O (writeback scan or + * garbage collection) may still prevent bcache device from + * being stopped. So here CACHE_SET_IO_DISABLE should be + * set to c->flags too, to make the internal I/O to cache + * device rejected and stopped immediately. + * If c is NULL, that means the bcache device is not attached + * to any cache set, then no CACHE_SET_IO_DISABLE bit to set. + */ + c = dc->disk.c; + if (c && test_and_set_bit(CACHE_SET_IO_DISABLE, &c->flags)) + pr_info("CACHE_SET_IO_DISABLE already set"); bcache_device_stop(&dc->disk); return true; @@ -1395,7 +1445,7 @@ bool bch_cache_set_error(struct cache_set *c, const char *fmt, ...) return false; if (test_and_set_bit(CACHE_SET_IO_DISABLE, &c->flags)) - pr_warn("CACHE_SET_IO_DISABLE already set"); + pr_info("CACHE_SET_IO_DISABLE already set"); /* XXX: we can be called from atomic context acquire_console_sem(); @@ -1448,14 +1498,10 @@ static void cache_set_free(struct closure *cl) if (c->moving_gc_wq) destroy_workqueue(c->moving_gc_wq); - if (c->bio_split) - bioset_free(c->bio_split); - if (c->fill_iter) - mempool_destroy(c->fill_iter); - if (c->bio_meta) - mempool_destroy(c->bio_meta); - if (c->search) - mempool_destroy(c->search); + bioset_exit(&c->bio_split); + mempool_exit(&c->fill_iter); + mempool_exit(&c->bio_meta); + mempool_exit(&c->search); kfree(c->devices); mutex_lock(&bch_register_lock); @@ -1539,6 +1585,20 @@ static void conditional_stop_bcache_device(struct cache_set *c, */ pr_warn("stop_when_cache_set_failed of %s is \"auto\" and cache is dirty, stop it to avoid potential data corruption.", d->disk->disk_name); + /* + * There might be a small time gap that cache set is + * released but bcache device is not. Inside this time + * gap, regular I/O requests will directly go into + * backing device as no cache set attached to. This + * behavior may also introduce potential inconsistence + * data in writeback mode while cache is dirty. + * Therefore before calling bcache_device_stop() due + * to a broken cache device, dc->io_disable should be + * explicitly set to true. + */ + dc->io_disable = true; + /* make others know io_disable is true earlier */ + smp_mb(); bcache_device_stop(d); } else { /* @@ -1652,21 +1712,17 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb) INIT_LIST_HEAD(&c->btree_cache_freed); INIT_LIST_HEAD(&c->data_buckets); - c->search = mempool_create_slab_pool(32, bch_search_cache); - if (!c->search) - goto err; - iter_size = (sb->bucket_size / sb->block_size + 1) * sizeof(struct btree_iter_set); if (!(c->devices = kzalloc(c->nr_uuids * sizeof(void *), GFP_KERNEL)) || - !(c->bio_meta = mempool_create_kmalloc_pool(2, - sizeof(struct bbio) + sizeof(struct bio_vec) * - bucket_pages(c))) || - !(c->fill_iter = mempool_create_kmalloc_pool(1, iter_size)) || - !(c->bio_split = bioset_create(4, offsetof(struct bbio, bio), - BIOSET_NEED_BVECS | - BIOSET_NEED_RESCUER)) || + mempool_init_slab_pool(&c->search, 32, bch_search_cache) || + mempool_init_kmalloc_pool(&c->bio_meta, 2, + sizeof(struct bbio) + sizeof(struct bio_vec) * + bucket_pages(c)) || + mempool_init_kmalloc_pool(&c->fill_iter, 1, iter_size) || + bioset_init(&c->bio_split, 4, offsetof(struct bbio, bio), + BIOSET_NEED_BVECS|BIOSET_NEED_RESCUER) || !(c->uuids = alloc_bucket_pages(GFP_KERNEL, c)) || !(c->moving_gc_wq = alloc_workqueue("bcache_gc", WQ_MEM_RECLAIM, 0)) || @@ -2003,12 +2059,10 @@ static int cache_alloc(struct cache *ca) static int register_cache(struct cache_sb *sb, struct page *sb_page, struct block_device *bdev, struct cache *ca) { - char name[BDEVNAME_SIZE]; const char *err = NULL; /* must be set for any error case */ int ret = 0; - bdevname(bdev, name); - + bdevname(bdev, ca->cache_dev_name); memcpy(&ca->sb, sb, sizeof(struct cache_sb)); ca->bdev = bdev; ca->bdev->bd_holder = ca; @@ -2045,14 +2099,14 @@ static int register_cache(struct cache_sb *sb, struct page *sb_page, goto out; } - pr_info("registered cache device %s", name); + pr_info("registered cache device %s", ca->cache_dev_name); out: kobject_put(&ca->kobj); err: if (err) - pr_notice("error %s: %s", name, err); + pr_notice("error %s: %s", ca->cache_dev_name, err); return ret; } diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c index dfeef583ee50..8ccbc8f3b3af 100644 --- a/drivers/md/bcache/sysfs.c +++ b/drivers/md/bcache/sysfs.c @@ -16,6 +16,22 @@ #include <linux/sort.h> #include <linux/sched/clock.h> +/* Default is -1; we skip past it for struct cached_dev's cache mode */ +static const char * const bch_cache_modes[] = { + "writethrough", + "writeback", + "writearound", + "none", + NULL +}; + +/* Default is -1; we skip past it for stop_when_cache_set_failed */ +static const char * const bch_stop_on_failure_modes[] = { + "auto", + "always", + NULL +}; + static const char * const cache_replacement_policies[] = { "lru", "fifo", @@ -114,6 +130,20 @@ rw_attribute(btree_shrinker_disabled); rw_attribute(copy_gc_enabled); rw_attribute(size); +static ssize_t bch_snprint_string_list(char *buf, size_t size, const char * const list[], + size_t selected) +{ + char *out = buf; + size_t i; + + for (i = 0; list[i]; i++) + out += snprintf(out, buf + size - out, + i == selected ? "[%s] " : "%s ", list[i]); + + out[-1] = '\n'; + return out - buf; +} + SHOW(__bch_cached_dev) { struct cached_dev *dc = container_of(kobj, struct cached_dev, @@ -124,12 +154,12 @@ SHOW(__bch_cached_dev) if (attr == &sysfs_cache_mode) return bch_snprint_string_list(buf, PAGE_SIZE, - bch_cache_modes + 1, + bch_cache_modes, BDEV_CACHE_MODE(&dc->sb)); if (attr == &sysfs_stop_when_cache_set_failed) return bch_snprint_string_list(buf, PAGE_SIZE, - bch_stop_on_failure_modes + 1, + bch_stop_on_failure_modes, dc->stop_when_cache_set_failed); @@ -253,8 +283,7 @@ STORE(__cached_dev) bch_cached_dev_run(dc); if (attr == &sysfs_cache_mode) { - v = bch_read_string_list(buf, bch_cache_modes + 1); - + v = __sysfs_match_string(bch_cache_modes, -1, buf); if (v < 0) return v; @@ -265,8 +294,7 @@ STORE(__cached_dev) } if (attr == &sysfs_stop_when_cache_set_failed) { - v = bch_read_string_list(buf, bch_stop_on_failure_modes + 1); - + v = __sysfs_match_string(bch_stop_on_failure_modes, -1, buf); if (v < 0) return v; @@ -635,6 +663,7 @@ SHOW_LOCKED(bch_cache_set) STORE(__bch_cache_set) { struct cache_set *c = container_of(kobj, struct cache_set, kobj); + ssize_t v; if (attr == &sysfs_unregister) bch_cache_set_unregister(c); @@ -698,8 +727,7 @@ STORE(__bch_cache_set) c->congested_write_threshold_us); if (attr == &sysfs_errors) { - ssize_t v = bch_read_string_list(buf, error_actions); - + v = __sysfs_match_string(error_actions, -1, buf); if (v < 0) return v; @@ -714,8 +742,7 @@ STORE(__bch_cache_set) c->error_decay = strtoul_or_return(buf) / 88; if (attr == &sysfs_io_disable) { - int v = strtoul_or_return(buf); - + v = strtoul_or_return(buf); if (v) { if (test_and_set_bit(CACHE_SET_IO_DISABLE, &c->flags)) @@ -929,6 +956,7 @@ SHOW_LOCKED(bch_cache) STORE(__bch_cache) { struct cache *ca = container_of(kobj, struct cache, kobj); + ssize_t v; if (attr == &sysfs_discard) { bool v = strtoul_or_return(buf); @@ -943,8 +971,7 @@ STORE(__bch_cache) } if (attr == &sysfs_cache_replacement_policy) { - ssize_t v = bch_read_string_list(buf, cache_replacement_policies); - + v = __sysfs_match_string(cache_replacement_policies, -1, buf); if (v < 0) return v; diff --git a/drivers/md/bcache/util.c b/drivers/md/bcache/util.c index 74febd5230df..fc479b026d6d 100644 --- a/drivers/md/bcache/util.c +++ b/drivers/md/bcache/util.c @@ -120,41 +120,6 @@ ssize_t bch_hprint(char *buf, int64_t v) return sprintf(buf, "%llu.%i%c", q, t * 10 / 1024, units[u]); } -ssize_t bch_snprint_string_list(char *buf, size_t size, const char * const list[], - size_t selected) -{ - char *out = buf; - size_t i; - - for (i = 0; list[i]; i++) - out += snprintf(out, buf + size - out, - i == selected ? "[%s] " : "%s ", list[i]); - - out[-1] = '\n'; - return out - buf; -} - -ssize_t bch_read_string_list(const char *buf, const char * const list[]) -{ - size_t i; - char *s, *d = kstrndup(buf, PAGE_SIZE - 1, GFP_KERNEL); - if (!d) - return -ENOMEM; - - s = strim(d); - - for (i = 0; list[i]; i++) - if (!strcmp(list[i], s)) - break; - - kfree(d); - - if (!list[i]) - return -EINVAL; - - return i; -} - bool bch_is_zero(const char *p, size_t n) { size_t i; diff --git a/drivers/md/bcache/util.h b/drivers/md/bcache/util.h index 268024529edd..cced87f8eb27 100644 --- a/drivers/md/bcache/util.h +++ b/drivers/md/bcache/util.h @@ -365,11 +365,6 @@ ssize_t bch_hprint(char *buf, int64_t v); bool bch_is_zero(const char *p, size_t n); int bch_parse_uuid(const char *s, char *uuid); -ssize_t bch_snprint_string_list(char *buf, size_t size, const char * const list[], - size_t selected); - -ssize_t bch_read_string_list(const char *buf, const char * const list[]); - struct time_stats { spinlock_t lock; /* diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c index 4a9547cdcdc5..ad45ebe1a74b 100644 --- a/drivers/md/bcache/writeback.c +++ b/drivers/md/bcache/writeback.c @@ -244,8 +244,10 @@ static void dirty_endio(struct bio *bio) struct keybuf_key *w = bio->bi_private; struct dirty_io *io = w->private; - if (bio->bi_status) + if (bio->bi_status) { SET_KEY_DIRTY(&w->key, false); + bch_count_backing_io_errors(io->dc, bio); + } closure_put(&io->cl); } diff --git a/drivers/md/dm-bio-prison-v1.c b/drivers/md/dm-bio-prison-v1.c index 874841f0fc83..e794e3662fdd 100644 --- a/drivers/md/dm-bio-prison-v1.c +++ b/drivers/md/dm-bio-prison-v1.c @@ -19,7 +19,7 @@ struct dm_bio_prison { spinlock_t lock; - mempool_t *cell_pool; + mempool_t cell_pool; struct rb_root cells; }; @@ -33,15 +33,16 @@ static struct kmem_cache *_cell_cache; */ struct dm_bio_prison *dm_bio_prison_create(void) { - struct dm_bio_prison *prison = kmalloc(sizeof(*prison), GFP_KERNEL); + struct dm_bio_prison *prison = kzalloc(sizeof(*prison), GFP_KERNEL); + int ret; if (!prison) return NULL; spin_lock_init(&prison->lock); - prison->cell_pool = mempool_create_slab_pool(MIN_CELLS, _cell_cache); - if (!prison->cell_pool) { + ret = mempool_init_slab_pool(&prison->cell_pool, MIN_CELLS, _cell_cache); + if (ret) { kfree(prison); return NULL; } @@ -54,21 +55,21 @@ EXPORT_SYMBOL_GPL(dm_bio_prison_create); void dm_bio_prison_destroy(struct dm_bio_prison *prison) { - mempool_destroy(prison->cell_pool); + mempool_exit(&prison->cell_pool); kfree(prison); } EXPORT_SYMBOL_GPL(dm_bio_prison_destroy); struct dm_bio_prison_cell *dm_bio_prison_alloc_cell(struct dm_bio_prison *prison, gfp_t gfp) { - return mempool_alloc(prison->cell_pool, gfp); + return mempool_alloc(&prison->cell_pool, gfp); } EXPORT_SYMBOL_GPL(dm_bio_prison_alloc_cell); void dm_bio_prison_free_cell(struct dm_bio_prison *prison, struct dm_bio_prison_cell *cell) { - mempool_free(cell, prison->cell_pool); + mempool_free(cell, &prison->cell_pool); } EXPORT_SYMBOL_GPL(dm_bio_prison_free_cell); diff --git a/drivers/md/dm-bio-prison-v2.c b/drivers/md/dm-bio-prison-v2.c index 8ce3a1a588cf..f866bc97b032 100644 --- a/drivers/md/dm-bio-prison-v2.c +++ b/drivers/md/dm-bio-prison-v2.c @@ -21,7 +21,7 @@ struct dm_bio_prison_v2 { struct workqueue_struct *wq; spinlock_t lock; - mempool_t *cell_pool; + mempool_t cell_pool; struct rb_root cells; }; @@ -35,7 +35,8 @@ static struct kmem_cache *_cell_cache; */ struct dm_bio_prison_v2 *dm_bio_prison_create_v2(struct workqueue_struct *wq) { - struct dm_bio_prison_v2 *prison = kmalloc(sizeof(*prison), GFP_KERNEL); + struct dm_bio_prison_v2 *prison = kzalloc(sizeof(*prison), GFP_KERNEL); + int ret; if (!prison) return NULL; @@ -43,8 +44,8 @@ struct dm_bio_prison_v2 *dm_bio_prison_create_v2(struct workqueue_struct *wq) prison->wq = wq; spin_lock_init(&prison->lock); - prison->cell_pool = mempool_create_slab_pool(MIN_CELLS, _cell_cache); - if (!prison->cell_pool) { + ret = mempool_init_slab_pool(&prison->cell_pool, MIN_CELLS, _cell_cache); + if (ret) { kfree(prison); return NULL; } @@ -57,21 +58,21 @@ EXPORT_SYMBOL_GPL(dm_bio_prison_create_v2); void dm_bio_prison_destroy_v2(struct dm_bio_prison_v2 *prison) { - mempool_destroy(prison->cell_pool); + mempool_exit(&prison->cell_pool); kfree(prison); } EXPORT_SYMBOL_GPL(dm_bio_prison_destroy_v2); struct dm_bio_prison_cell_v2 *dm_bio_prison_alloc_cell_v2(struct dm_bio_prison_v2 *prison, gfp_t gfp) { - return mempool_alloc(prison->cell_pool, gfp); + return mempool_alloc(&prison->cell_pool, gfp); } EXPORT_SYMBOL_GPL(dm_bio_prison_alloc_cell_v2); void dm_bio_prison_free_cell_v2(struct dm_bio_prison_v2 *prison, struct dm_bio_prison_cell_v2 *cell) { - mempool_free(cell, prison->cell_pool); + mempool_free(cell, &prison->cell_pool); } EXPORT_SYMBOL_GPL(dm_bio_prison_free_cell_v2); diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c index aa2032fa80d4..dc385b70e4c3 100644 --- a/drivers/md/dm-bufio.c +++ b/drivers/md/dm-bufio.c @@ -6,7 +6,7 @@ * This file is released under the GPL. */ -#include "dm-bufio.h" +#include <linux/dm-bufio.h> #include <linux/device-mapper.h> #include <linux/dm-io.h> @@ -51,19 +51,6 @@ #define DM_BUFIO_DEFAULT_RETAIN_BYTES (256 * 1024) /* - * The number of bvec entries that are embedded directly in the buffer. - * If the chunk size is larger, dm-io is used to do the io. - */ -#define DM_BUFIO_INLINE_VECS 16 - -/* - * Don't try to use kmem_cache_alloc for blocks larger than this. - * For explanation, see alloc_buffer_data below. - */ -#define DM_BUFIO_BLOCK_SIZE_SLAB_LIMIT (PAGE_SIZE >> 1) -#define DM_BUFIO_BLOCK_SIZE_GFP_LIMIT (PAGE_SIZE << (MAX_ORDER - 1)) - -/* * Align buffer writes to this boundary. * Tests show that SSDs have the highest IOPS when using 4k writes. */ @@ -99,13 +86,12 @@ struct dm_bufio_client { struct block_device *bdev; unsigned block_size; - unsigned char sectors_per_block_bits; - unsigned char pages_per_block_bits; - unsigned char blocks_per_page_bits; - unsigned aux_size; + s8 sectors_per_block_bits; void (*alloc_callback)(struct dm_buffer *); void (*write_callback)(struct dm_buffer *); + struct kmem_cache *slab_buffer; + struct kmem_cache *slab_cache; struct dm_io_client *dm_io; struct list_head reserved_buffers; @@ -148,11 +134,11 @@ struct dm_buffer { struct list_head lru_list; sector_t block; void *data; - enum data_mode data_mode; + unsigned char data_mode; /* DATA_MODE_* */ unsigned char list_mode; /* LIST_* */ - unsigned hold_count; blk_status_t read_error; blk_status_t write_error; + unsigned hold_count; unsigned long state; unsigned long last_accessed; unsigned dirty_start; @@ -161,8 +147,7 @@ struct dm_buffer { unsigned write_end; struct dm_bufio_client *c; struct list_head write_list; - struct bio bio; - struct bio_vec bio_vec[DM_BUFIO_INLINE_VECS]; + void (*end_io)(struct dm_buffer *, blk_status_t); #ifdef CONFIG_DM_DEBUG_BLOCK_STACK_TRACING #define MAX_STACK 10 struct stack_trace stack_trace; @@ -172,21 +157,6 @@ struct dm_buffer { /*----------------------------------------------------------------*/ -static struct kmem_cache *dm_bufio_caches[PAGE_SHIFT - SECTOR_SHIFT]; -static char *dm_bufio_cache_names[PAGE_SHIFT - SECTOR_SHIFT]; - -static inline int dm_bufio_cache_index(struct dm_bufio_client *c) -{ - unsigned ret = c->blocks_per_page_bits - 1; - - BUG_ON(ret >= ARRAY_SIZE(dm_bufio_caches)); - - return ret; -} - -#define DM_BUFIO_CACHE(c) (dm_bufio_caches[dm_bufio_cache_index(c)]) -#define DM_BUFIO_CACHE_NAME(c) (dm_bufio_cache_names[dm_bufio_cache_index(c)]) - #define dm_bufio_in_request() (!!current->bio_list) static void dm_bufio_lock(struct dm_bufio_client *c) @@ -319,7 +289,7 @@ static void __remove(struct dm_bufio_client *c, struct dm_buffer *b) /*----------------------------------------------------------------*/ -static void adjust_total_allocated(enum data_mode data_mode, long diff) +static void adjust_total_allocated(unsigned char data_mode, long diff) { static unsigned long * const class_ptr[DATA_MODE_LIMIT] = { &dm_bufio_allocated_kmem_cache, @@ -384,18 +354,18 @@ static void __cache_size_refresh(void) * space. */ static void *alloc_buffer_data(struct dm_bufio_client *c, gfp_t gfp_mask, - enum data_mode *data_mode) + unsigned char *data_mode) { - if (c->block_size <= DM_BUFIO_BLOCK_SIZE_SLAB_LIMIT) { + if (unlikely(c->slab_cache != NULL)) { *data_mode = DATA_MODE_SLAB; - return kmem_cache_alloc(DM_BUFIO_CACHE(c), gfp_mask); + return kmem_cache_alloc(c->slab_cache, gfp_mask); } - if (c->block_size <= DM_BUFIO_BLOCK_SIZE_GFP_LIMIT && + if (c->block_size <= KMALLOC_MAX_SIZE && gfp_mask & __GFP_NORETRY) { *data_mode = DATA_MODE_GET_FREE_PAGES; return (void *)__get_free_pages(gfp_mask, - c->pages_per_block_bits); + c->sectors_per_block_bits - (PAGE_SHIFT - SECTOR_SHIFT)); } *data_mode = DATA_MODE_VMALLOC; @@ -424,15 +394,16 @@ static void *alloc_buffer_data(struct dm_bufio_client *c, gfp_t gfp_mask, * Free buffer's data. */ static void free_buffer_data(struct dm_bufio_client *c, - void *data, enum data_mode data_mode) + void *data, unsigned char data_mode) { switch (data_mode) { case DATA_MODE_SLAB: - kmem_cache_free(DM_BUFIO_CACHE(c), data); + kmem_cache_free(c->slab_cache, data); break; case DATA_MODE_GET_FREE_PAGES: - free_pages((unsigned long)data, c->pages_per_block_bits); + free_pages((unsigned long)data, + c->sectors_per_block_bits - (PAGE_SHIFT - SECTOR_SHIFT)); break; case DATA_MODE_VMALLOC: @@ -451,8 +422,7 @@ static void free_buffer_data(struct dm_bufio_client *c, */ static struct dm_buffer *alloc_buffer(struct dm_bufio_client *c, gfp_t gfp_mask) { - struct dm_buffer *b = kmalloc(sizeof(struct dm_buffer) + c->aux_size, - gfp_mask); + struct dm_buffer *b = kmem_cache_alloc(c->slab_buffer, gfp_mask); if (!b) return NULL; @@ -461,7 +431,7 @@ static struct dm_buffer *alloc_buffer(struct dm_bufio_client *c, gfp_t gfp_mask) b->data = alloc_buffer_data(c, gfp_mask, &b->data_mode); if (!b->data) { - kfree(b); + kmem_cache_free(c->slab_buffer, b); return NULL; } @@ -483,7 +453,7 @@ static void free_buffer(struct dm_buffer *b) adjust_total_allocated(b->data_mode, -(long)c->block_size); free_buffer_data(c, b->data, b->data_mode); - kfree(b); + kmem_cache_free(c->slab_buffer, b); } /* @@ -540,10 +510,6 @@ static void __relink_lru(struct dm_buffer *b, int dirty) * * the memory must be direct-mapped, not vmalloced; * - * the I/O driver can reject requests spuriously if it thinks that - * the requests are too big for the device or if they cross a - * controller-defined memory boundary. - * * If the buffer is small enough (up to DM_BUFIO_INLINE_VECS pages) and * it is not vmalloced, try using the bio interface. * @@ -561,12 +527,11 @@ static void dmio_complete(unsigned long error, void *context) { struct dm_buffer *b = context; - b->bio.bi_status = error ? BLK_STS_IOERR : 0; - b->bio.bi_end_io(&b->bio); + b->end_io(b, unlikely(error != 0) ? BLK_STS_IOERR : 0); } static void use_dmio(struct dm_buffer *b, int rw, sector_t sector, - unsigned n_sectors, unsigned offset, bio_end_io_t *end_io) + unsigned n_sectors, unsigned offset) { int r; struct dm_io_request io_req = { @@ -590,76 +555,77 @@ static void use_dmio(struct dm_buffer *b, int rw, sector_t sector, io_req.mem.ptr.vma = (char *)b->data + offset; } - b->bio.bi_end_io = end_io; - r = dm_io(&io_req, 1, ®ion, NULL); - if (r) { - b->bio.bi_status = errno_to_blk_status(r); - end_io(&b->bio); - } + if (unlikely(r)) + b->end_io(b, errno_to_blk_status(r)); } -static void inline_endio(struct bio *bio) +static void bio_complete(struct bio *bio) { - bio_end_io_t *end_fn = bio->bi_private; + struct dm_buffer *b = bio->bi_private; blk_status_t status = bio->bi_status; - - /* - * Reset the bio to free any attached resources - * (e.g. bio integrity profiles). - */ - bio_reset(bio); - - bio->bi_status = status; - end_fn(bio); + bio_put(bio); + b->end_io(b, status); } -static void use_inline_bio(struct dm_buffer *b, int rw, sector_t sector, - unsigned n_sectors, unsigned offset, bio_end_io_t *end_io) +static void use_bio(struct dm_buffer *b, int rw, sector_t sector, + unsigned n_sectors, unsigned offset) { + struct bio *bio; char *ptr; - unsigned len; + unsigned vec_size, len; - bio_init(&b->bio, b->bio_vec, DM_BUFIO_INLINE_VECS); - b->bio.bi_iter.bi_sector = sector; - bio_set_dev(&b->bio, b->c->bdev); - b->bio.bi_end_io = inline_endio; - /* - * Use of .bi_private isn't a problem here because - * the dm_buffer's inline bio is local to bufio. - */ - b->bio.bi_private = end_io; - bio_set_op_attrs(&b->bio, rw, 0); + vec_size = b->c->block_size >> PAGE_SHIFT; + if (unlikely(b->c->sectors_per_block_bits < PAGE_SHIFT - SECTOR_SHIFT)) + vec_size += 2; + + bio = bio_kmalloc(GFP_NOWAIT | __GFP_NORETRY | __GFP_NOWARN, vec_size); + if (!bio) { +dmio: + use_dmio(b, rw, sector, n_sectors, offset); + return; + } + + bio->bi_iter.bi_sector = sector; + bio_set_dev(bio, b->c->bdev); + bio_set_op_attrs(bio, rw, 0); + bio->bi_end_io = bio_complete; + bio->bi_private = b; ptr = (char *)b->data + offset; len = n_sectors << SECTOR_SHIFT; do { unsigned this_step = min((unsigned)(PAGE_SIZE - offset_in_page(ptr)), len); - if (!bio_add_page(&b->bio, virt_to_page(ptr), this_step, + if (!bio_add_page(bio, virt_to_page(ptr), this_step, offset_in_page(ptr))) { - BUG_ON(b->c->block_size <= PAGE_SIZE); - use_dmio(b, rw, sector, n_sectors, offset, end_io); - return; + bio_put(bio); + goto dmio; } len -= this_step; ptr += this_step; } while (len > 0); - submit_bio(&b->bio); + submit_bio(bio); } -static void submit_io(struct dm_buffer *b, int rw, bio_end_io_t *end_io) +static void submit_io(struct dm_buffer *b, int rw, void (*end_io)(struct dm_buffer *, blk_status_t)) { unsigned n_sectors; sector_t sector; unsigned offset, end; - sector = (b->block << b->c->sectors_per_block_bits) + b->c->start; + b->end_io = end_io; + + if (likely(b->c->sectors_per_block_bits >= 0)) + sector = b->block << b->c->sectors_per_block_bits; + else + sector = b->block * (b->c->block_size >> SECTOR_SHIFT); + sector += b->c->start; if (rw != REQ_OP_WRITE) { - n_sectors = 1 << b->c->sectors_per_block_bits; + n_sectors = b->c->block_size >> SECTOR_SHIFT; offset = 0; } else { if (b->c->write_callback) @@ -676,11 +642,10 @@ static void submit_io(struct dm_buffer *b, int rw, bio_end_io_t *end_io) n_sectors = (end - offset) >> SECTOR_SHIFT; } - if (n_sectors <= ((DM_BUFIO_INLINE_VECS * PAGE_SIZE) >> SECTOR_SHIFT) && - b->data_mode != DATA_MODE_VMALLOC) - use_inline_bio(b, rw, sector, n_sectors, offset, end_io); + if (b->data_mode != DATA_MODE_VMALLOC) + use_bio(b, rw, sector, n_sectors, offset); else - use_dmio(b, rw, sector, n_sectors, offset, end_io); + use_dmio(b, rw, sector, n_sectors, offset); } /*---------------------------------------------------------------- @@ -693,16 +658,14 @@ static void submit_io(struct dm_buffer *b, int rw, bio_end_io_t *end_io) * Set the error, clear B_WRITING bit and wake anyone who was waiting on * it. */ -static void write_endio(struct bio *bio) +static void write_endio(struct dm_buffer *b, blk_status_t status) { - struct dm_buffer *b = container_of(bio, struct dm_buffer, bio); - - b->write_error = bio->bi_status; - if (unlikely(bio->bi_status)) { + b->write_error = status; + if (unlikely(status)) { struct dm_bufio_client *c = b->c; (void)cmpxchg(&c->async_write_error, 0, - blk_status_to_errno(bio->bi_status)); + blk_status_to_errno(status)); } BUG_ON(!test_bit(B_WRITING, &b->state)); @@ -963,8 +926,11 @@ static void __get_memory_limit(struct dm_bufio_client *c, } } - buffers = dm_bufio_cache_size_per_client >> - (c->sectors_per_block_bits + SECTOR_SHIFT); + buffers = dm_bufio_cache_size_per_client; + if (likely(c->sectors_per_block_bits >= 0)) + buffers >>= c->sectors_per_block_bits + SECTOR_SHIFT; + else + buffers /= c->block_size; if (buffers < c->minimum_buffers) buffers = c->minimum_buffers; @@ -1076,11 +1042,9 @@ found_buffer: * The endio routine for reading: set the error, clear the bit and wake up * anyone waiting on the buffer. */ -static void read_endio(struct bio *bio) +static void read_endio(struct dm_buffer *b, blk_status_t status) { - struct dm_buffer *b = container_of(bio, struct dm_buffer, bio); - - b->read_error = bio->bi_status; + b->read_error = status; BUG_ON(!test_bit(B_READING, &b->state)); @@ -1482,13 +1446,13 @@ void dm_bufio_forget(struct dm_bufio_client *c, sector_t block) dm_bufio_unlock(c); } -EXPORT_SYMBOL(dm_bufio_forget); +EXPORT_SYMBOL_GPL(dm_bufio_forget); void dm_bufio_set_minimum_buffers(struct dm_bufio_client *c, unsigned n) { c->minimum_buffers = n; } -EXPORT_SYMBOL(dm_bufio_set_minimum_buffers); +EXPORT_SYMBOL_GPL(dm_bufio_set_minimum_buffers); unsigned dm_bufio_get_block_size(struct dm_bufio_client *c) { @@ -1498,8 +1462,12 @@ EXPORT_SYMBOL_GPL(dm_bufio_get_block_size); sector_t dm_bufio_get_device_size(struct dm_bufio_client *c) { - return i_size_read(c->bdev->bd_inode) >> - (SECTOR_SHIFT + c->sectors_per_block_bits); + sector_t s = i_size_read(c->bdev->bd_inode) >> SECTOR_SHIFT; + if (likely(c->sectors_per_block_bits >= 0)) + s >>= c->sectors_per_block_bits; + else + sector_div(s, c->block_size >> SECTOR_SHIFT); + return s; } EXPORT_SYMBOL_GPL(dm_bufio_get_device_size); @@ -1597,8 +1565,12 @@ static bool __try_evict_buffer(struct dm_buffer *b, gfp_t gfp) static unsigned long get_retain_buffers(struct dm_bufio_client *c) { - unsigned long retain_bytes = READ_ONCE(dm_bufio_retain_bytes); - return retain_bytes >> (c->sectors_per_block_bits + SECTOR_SHIFT); + unsigned long retain_bytes = READ_ONCE(dm_bufio_retain_bytes); + if (likely(c->sectors_per_block_bits >= 0)) + retain_bytes >>= c->sectors_per_block_bits + SECTOR_SHIFT; + else + retain_bytes /= c->block_size; + return retain_bytes; } static unsigned long __scan(struct dm_bufio_client *c, unsigned long nr_to_scan, @@ -1662,9 +1634,13 @@ struct dm_bufio_client *dm_bufio_client_create(struct block_device *bdev, unsign int r; struct dm_bufio_client *c; unsigned i; + char slab_name[27]; - BUG_ON(block_size < 1 << SECTOR_SHIFT || - (block_size & (block_size - 1))); + if (!block_size || block_size & ((1 << SECTOR_SHIFT) - 1)) { + DMERR("%s: block size not specified or is not multiple of 512b", __func__); + r = -EINVAL; + goto bad_client; + } c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) { @@ -1675,13 +1651,11 @@ struct dm_bufio_client *dm_bufio_client_create(struct block_device *bdev, unsign c->bdev = bdev; c->block_size = block_size; - c->sectors_per_block_bits = __ffs(block_size) - SECTOR_SHIFT; - c->pages_per_block_bits = (__ffs(block_size) >= PAGE_SHIFT) ? - __ffs(block_size) - PAGE_SHIFT : 0; - c->blocks_per_page_bits = (__ffs(block_size) < PAGE_SHIFT ? - PAGE_SHIFT - __ffs(block_size) : 0); + if (is_power_of_2(block_size)) + c->sectors_per_block_bits = __ffs(block_size) - SECTOR_SHIFT; + else + c->sectors_per_block_bits = -1; - c->aux_size = aux_size; c->alloc_callback = alloc_callback; c->write_callback = write_callback; @@ -1694,7 +1668,7 @@ struct dm_bufio_client *dm_bufio_client_create(struct block_device *bdev, unsign INIT_LIST_HEAD(&c->reserved_buffers); c->need_reserved_buffers = reserved_buffers; - c->minimum_buffers = DM_BUFIO_MIN_BUFFERS; + dm_bufio_set_minimum_buffers(c, DM_BUFIO_MIN_BUFFERS); init_waitqueue_head(&c->free_buffer_wait); c->async_write_error = 0; @@ -1705,29 +1679,27 @@ struct dm_bufio_client *dm_bufio_client_create(struct block_device *bdev, unsign goto bad_dm_io; } - mutex_lock(&dm_bufio_clients_lock); - if (c->blocks_per_page_bits) { - if (!DM_BUFIO_CACHE_NAME(c)) { - DM_BUFIO_CACHE_NAME(c) = kasprintf(GFP_KERNEL, "dm_bufio_cache-%u", c->block_size); - if (!DM_BUFIO_CACHE_NAME(c)) { - r = -ENOMEM; - mutex_unlock(&dm_bufio_clients_lock); - goto bad; - } - } - - if (!DM_BUFIO_CACHE(c)) { - DM_BUFIO_CACHE(c) = kmem_cache_create(DM_BUFIO_CACHE_NAME(c), - c->block_size, - c->block_size, 0, NULL); - if (!DM_BUFIO_CACHE(c)) { - r = -ENOMEM; - mutex_unlock(&dm_bufio_clients_lock); - goto bad; - } + if (block_size <= KMALLOC_MAX_SIZE && + (block_size < PAGE_SIZE || !is_power_of_2(block_size))) { + unsigned align = min(1U << __ffs(block_size), (unsigned)PAGE_SIZE); + snprintf(slab_name, sizeof slab_name, "dm_bufio_cache-%u", block_size); + c->slab_cache = kmem_cache_create(slab_name, block_size, align, + SLAB_RECLAIM_ACCOUNT, NULL); + if (!c->slab_cache) { + r = -ENOMEM; + goto bad; } } - mutex_unlock(&dm_bufio_clients_lock); + if (aux_size) + snprintf(slab_name, sizeof slab_name, "dm_bufio_buffer-%u", aux_size); + else + snprintf(slab_name, sizeof slab_name, "dm_bufio_buffer"); + c->slab_buffer = kmem_cache_create(slab_name, sizeof(struct dm_buffer) + aux_size, + 0, SLAB_RECLAIM_ACCOUNT, NULL); + if (!c->slab_buffer) { + r = -ENOMEM; + goto bad; + } while (c->need_reserved_buffers) { struct dm_buffer *b = alloc_buffer(c, GFP_KERNEL); @@ -1762,6 +1734,8 @@ bad: list_del(&b->lru_list); free_buffer(b); } + kmem_cache_destroy(c->slab_cache); + kmem_cache_destroy(c->slab_buffer); dm_io_client_destroy(c->dm_io); bad_dm_io: mutex_destroy(&c->lock); @@ -1808,6 +1782,8 @@ void dm_bufio_client_destroy(struct dm_bufio_client *c) for (i = 0; i < LIST_SIZE; i++) BUG_ON(c->n_buffers[i]); + kmem_cache_destroy(c->slab_cache); + kmem_cache_destroy(c->slab_buffer); dm_io_client_destroy(c->dm_io); mutex_destroy(&c->lock); kfree(c); @@ -1911,9 +1887,6 @@ static int __init dm_bufio_init(void) dm_bufio_allocated_vmalloc = 0; dm_bufio_current_allocated = 0; - memset(&dm_bufio_caches, 0, sizeof dm_bufio_caches); - memset(&dm_bufio_cache_names, 0, sizeof dm_bufio_cache_names); - mem = (__u64)mult_frac(totalram_pages - totalhigh_pages, DM_BUFIO_MEMORY_PERCENT, 100) << PAGE_SHIFT; @@ -1948,17 +1921,10 @@ static int __init dm_bufio_init(void) static void __exit dm_bufio_exit(void) { int bug = 0; - int i; cancel_delayed_work_sync(&dm_bufio_work); destroy_workqueue(dm_bufio_wq); - for (i = 0; i < ARRAY_SIZE(dm_bufio_caches); i++) - kmem_cache_destroy(dm_bufio_caches[i]); - - for (i = 0; i < ARRAY_SIZE(dm_bufio_cache_names); i++) - kfree(dm_bufio_cache_names[i]); - if (dm_bufio_client_count) { DMCRIT("%s: dm_bufio_client_count leaked: %d", __func__, dm_bufio_client_count); diff --git a/drivers/md/dm-bufio.h b/drivers/md/dm-bufio.h deleted file mode 100644 index be732d3f8611..000000000000 --- a/drivers/md/dm-bufio.h +++ /dev/null @@ -1,148 +0,0 @@ -/* - * Copyright (C) 2009-2011 Red Hat, Inc. - * - * Author: Mikulas Patocka <mpatocka@redhat.com> - * - * This file is released under the GPL. - */ - -#ifndef DM_BUFIO_H -#define DM_BUFIO_H - -#include <linux/blkdev.h> -#include <linux/types.h> - -/*----------------------------------------------------------------*/ - -struct dm_bufio_client; -struct dm_buffer; - -/* - * Create a buffered IO cache on a given device - */ -struct dm_bufio_client * -dm_bufio_client_create(struct block_device *bdev, unsigned block_size, - unsigned reserved_buffers, unsigned aux_size, - void (*alloc_callback)(struct dm_buffer *), - void (*write_callback)(struct dm_buffer *)); - -/* - * Release a buffered IO cache. - */ -void dm_bufio_client_destroy(struct dm_bufio_client *c); - -/* - * Set the sector range. - * When this function is called, there must be no I/O in progress on the bufio - * client. - */ -void dm_bufio_set_sector_offset(struct dm_bufio_client *c, sector_t start); - -/* - * WARNING: to avoid deadlocks, these conditions are observed: - * - * - At most one thread can hold at most "reserved_buffers" simultaneously. - * - Each other threads can hold at most one buffer. - * - Threads which call only dm_bufio_get can hold unlimited number of - * buffers. - */ - -/* - * Read a given block from disk. Returns pointer to data. Returns a - * pointer to dm_buffer that can be used to release the buffer or to make - * it dirty. - */ -void *dm_bufio_read(struct dm_bufio_client *c, sector_t block, - struct dm_buffer **bp); - -/* - * Like dm_bufio_read, but return buffer from cache, don't read - * it. If the buffer is not in the cache, return NULL. - */ -void *dm_bufio_get(struct dm_bufio_client *c, sector_t block, - struct dm_buffer **bp); - -/* - * Like dm_bufio_read, but don't read anything from the disk. It is - * expected that the caller initializes the buffer and marks it dirty. - */ -void *dm_bufio_new(struct dm_bufio_client *c, sector_t block, - struct dm_buffer **bp); - -/* - * Prefetch the specified blocks to the cache. - * The function starts to read the blocks and returns without waiting for - * I/O to finish. - */ -void dm_bufio_prefetch(struct dm_bufio_client *c, - sector_t block, unsigned n_blocks); - -/* - * Release a reference obtained with dm_bufio_{read,get,new}. The data - * pointer and dm_buffer pointer is no longer valid after this call. - */ -void dm_bufio_release(struct dm_buffer *b); - -/* - * Mark a buffer dirty. It should be called after the buffer is modified. - * - * In case of memory pressure, the buffer may be written after - * dm_bufio_mark_buffer_dirty, but before dm_bufio_write_dirty_buffers. So - * dm_bufio_write_dirty_buffers guarantees that the buffer is on-disk but - * the actual writing may occur earlier. - */ -void dm_bufio_mark_buffer_dirty(struct dm_buffer *b); - -/* - * Mark a part of the buffer dirty. - * - * The specified part of the buffer is scheduled to be written. dm-bufio may - * write the specified part of the buffer or it may write a larger superset. - */ -void dm_bufio_mark_partial_buffer_dirty(struct dm_buffer *b, - unsigned start, unsigned end); - -/* - * Initiate writing of dirty buffers, without waiting for completion. - */ -void dm_bufio_write_dirty_buffers_async(struct dm_bufio_client *c); - -/* - * Write all dirty buffers. Guarantees that all dirty buffers created prior - * to this call are on disk when this call exits. - */ -int dm_bufio_write_dirty_buffers(struct dm_bufio_client *c); - -/* - * Send an empty write barrier to the device to flush hardware disk cache. - */ -int dm_bufio_issue_flush(struct dm_bufio_client *c); - -/* - * Like dm_bufio_release but also move the buffer to the new - * block. dm_bufio_write_dirty_buffers is needed to commit the new block. - */ -void dm_bufio_release_move(struct dm_buffer *b, sector_t new_block); - -/* - * Free the given buffer. - * This is just a hint, if the buffer is in use or dirty, this function - * does nothing. - */ -void dm_bufio_forget(struct dm_bufio_client *c, sector_t block); - -/* - * Set the minimum number of buffers before cleanup happens. - */ -void dm_bufio_set_minimum_buffers(struct dm_bufio_client *c, unsigned n); - -unsigned dm_bufio_get_block_size(struct dm_bufio_client *c); -sector_t dm_bufio_get_device_size(struct dm_bufio_client *c); -sector_t dm_bufio_get_block_number(struct dm_buffer *b); -void *dm_bufio_get_block_data(struct dm_buffer *b); -void *dm_bufio_get_aux_data(struct dm_buffer *b); -struct dm_bufio_client *dm_bufio_get_client(struct dm_buffer *b); - -/*----------------------------------------------------------------*/ - -#endif diff --git a/drivers/md/dm-cache-background-tracker.c b/drivers/md/dm-cache-background-tracker.c index 1d0af0a21fc7..84814e819e4c 100644 --- a/drivers/md/dm-cache-background-tracker.c +++ b/drivers/md/dm-cache-background-tracker.c @@ -166,7 +166,7 @@ static bool max_work_reached(struct background_tracker *b) atomic_read(&b->pending_demotes) >= b->max_work; } -struct bt_work *alloc_work(struct background_tracker *b) +static struct bt_work *alloc_work(struct background_tracker *b) { if (max_work_reached(b)) return NULL; diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c index 47407e43b96a..001c71248246 100644 --- a/drivers/md/dm-cache-target.c +++ b/drivers/md/dm-cache-target.c @@ -447,9 +447,9 @@ struct cache { struct work_struct migration_worker; struct delayed_work waker; struct dm_bio_prison_v2 *prison; - struct bio_set *bs; + struct bio_set bs; - mempool_t *migration_pool; + mempool_t migration_pool; struct dm_cache_policy *policy; unsigned policy_nr_args; @@ -550,7 +550,7 @@ static struct dm_cache_migration *alloc_migration(struct cache *cache) { struct dm_cache_migration *mg; - mg = mempool_alloc(cache->migration_pool, GFP_NOWAIT); + mg = mempool_alloc(&cache->migration_pool, GFP_NOWAIT); if (!mg) return NULL; @@ -569,7 +569,7 @@ static void free_migration(struct dm_cache_migration *mg) if (atomic_dec_and_test(&cache->nr_allocated_migrations)) wake_up(&cache->migration_wait); - mempool_free(mg, cache->migration_pool); + mempool_free(mg, &cache->migration_pool); } /*----------------------------------------------------------------*/ @@ -924,7 +924,7 @@ static void issue_op(struct bio *bio, void *context) static void remap_to_origin_and_cache(struct cache *cache, struct bio *bio, dm_oblock_t oblock, dm_cblock_t cblock) { - struct bio *origin_bio = bio_clone_fast(bio, GFP_NOIO, cache->bs); + struct bio *origin_bio = bio_clone_fast(bio, GFP_NOIO, &cache->bs); BUG_ON(!origin_bio); @@ -2011,7 +2011,7 @@ static void destroy(struct cache *cache) { unsigned i; - mempool_destroy(cache->migration_pool); + mempool_exit(&cache->migration_pool); if (cache->prison) dm_bio_prison_destroy_v2(cache->prison); @@ -2047,8 +2047,7 @@ static void destroy(struct cache *cache) kfree(cache->ctr_args[i]); kfree(cache->ctr_args); - if (cache->bs) - bioset_free(cache->bs); + bioset_exit(&cache->bs); kfree(cache); } @@ -2498,8 +2497,8 @@ static int cache_create(struct cache_args *ca, struct cache **result) cache->features = ca->features; if (writethrough_mode(cache)) { /* Create bioset for writethrough bios issued to origin */ - cache->bs = bioset_create(BIO_POOL_SIZE, 0, 0); - if (!cache->bs) + r = bioset_init(&cache->bs, BIO_POOL_SIZE, 0, 0); + if (r) goto bad; } @@ -2630,9 +2629,9 @@ static int cache_create(struct cache_args *ca, struct cache **result) goto bad; } - cache->migration_pool = mempool_create_slab_pool(MIGRATION_POOL_SIZE, - migration_cache); - if (!cache->migration_pool) { + r = mempool_init_slab_pool(&cache->migration_pool, MIGRATION_POOL_SIZE, + migration_cache); + if (r) { *error = "Error creating cache's migration mempool"; goto bad; } @@ -3387,7 +3386,8 @@ static int process_invalidate_cblocks_message(struct cache *cache, unsigned coun * * The key migration_threshold is supported by the cache target core. */ -static int cache_message(struct dm_target *ti, unsigned argc, char **argv) +static int cache_message(struct dm_target *ti, unsigned argc, char **argv, + char *result, unsigned maxlen) { struct cache *cache = ti->private; diff --git a/drivers/md/dm-core.h b/drivers/md/dm-core.h index 3222e21cbbf8..f21c5d21bf1b 100644 --- a/drivers/md/dm-core.h +++ b/drivers/md/dm-core.h @@ -91,8 +91,8 @@ struct mapped_device { /* * io objects are allocated from here. */ - struct bio_set *io_bs; - struct bio_set *bs; + struct bio_set io_bs; + struct bio_set bs; /* * freeze/thaw support require holding onto a super block diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 8168f737590e..da02f4d8e4b9 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -143,12 +143,14 @@ struct crypt_config { * pool for per bio private data, crypto requests, * encryption requeusts/buffer pages and integrity tags */ - mempool_t *req_pool; - mempool_t *page_pool; - mempool_t *tag_pool; + mempool_t req_pool; + mempool_t page_pool; + mempool_t tag_pool; unsigned tag_pool_max_sectors; - struct bio_set *bs; + struct percpu_counter n_allocated_pages; + + struct bio_set bs; struct mutex bio_alloc_lock; struct workqueue_struct *io_queue; @@ -219,6 +221,12 @@ struct crypt_config { #define MAX_TAG_SIZE 480 #define POOL_ENTRY_SIZE 512 +static DEFINE_SPINLOCK(dm_crypt_clients_lock); +static unsigned dm_crypt_clients_n = 0; +static volatile unsigned long dm_crypt_pages_per_client; +#define DM_CRYPT_MEMORY_PERCENT 2 +#define DM_CRYPT_MIN_PAGES_PER_CLIENT (BIO_MAX_PAGES * 16) + static void clone_init(struct dm_crypt_io *, struct bio *); static void kcryptd_queue_crypt(struct dm_crypt_io *io); static struct scatterlist *crypt_get_sg_data(struct crypt_config *cc, @@ -1237,7 +1245,7 @@ static void crypt_alloc_req_skcipher(struct crypt_config *cc, unsigned key_index = ctx->cc_sector & (cc->tfms_count - 1); if (!ctx->r.req) - ctx->r.req = mempool_alloc(cc->req_pool, GFP_NOIO); + ctx->r.req = mempool_alloc(&cc->req_pool, GFP_NOIO); skcipher_request_set_tfm(ctx->r.req, cc->cipher_tfm.tfms[key_index]); @@ -1254,7 +1262,7 @@ static void crypt_alloc_req_aead(struct crypt_config *cc, struct convert_context *ctx) { if (!ctx->r.req_aead) - ctx->r.req_aead = mempool_alloc(cc->req_pool, GFP_NOIO); + ctx->r.req_aead = mempool_alloc(&cc->req_pool, GFP_NOIO); aead_request_set_tfm(ctx->r.req_aead, cc->cipher_tfm.tfms_aead[0]); @@ -1282,7 +1290,7 @@ static void crypt_free_req_skcipher(struct crypt_config *cc, struct dm_crypt_io *io = dm_per_bio_data(base_bio, cc->per_bio_data_size); if ((struct skcipher_request *)(io + 1) != req) - mempool_free(req, cc->req_pool); + mempool_free(req, &cc->req_pool); } static void crypt_free_req_aead(struct crypt_config *cc, @@ -1291,7 +1299,7 @@ static void crypt_free_req_aead(struct crypt_config *cc, struct dm_crypt_io *io = dm_per_bio_data(base_bio, cc->per_bio_data_size); if ((struct aead_request *)(io + 1) != req) - mempool_free(req, cc->req_pool); + mempool_free(req, &cc->req_pool); } static void crypt_free_req(struct crypt_config *cc, void *req, struct bio *base_bio) @@ -1401,7 +1409,7 @@ retry: if (unlikely(gfp_mask & __GFP_DIRECT_RECLAIM)) mutex_lock(&cc->bio_alloc_lock); - clone = bio_alloc_bioset(GFP_NOIO, nr_iovecs, cc->bs); + clone = bio_alloc_bioset(GFP_NOIO, nr_iovecs, &cc->bs); if (!clone) goto out; @@ -1410,7 +1418,7 @@ retry: remaining_size = size; for (i = 0; i < nr_iovecs; i++) { - page = mempool_alloc(cc->page_pool, gfp_mask); + page = mempool_alloc(&cc->page_pool, gfp_mask); if (!page) { crypt_free_buffer_pages(cc, clone); bio_put(clone); @@ -1445,7 +1453,7 @@ static void crypt_free_buffer_pages(struct crypt_config *cc, struct bio *clone) bio_for_each_segment_all(bv, clone, i) { BUG_ON(!bv->bv_page); - mempool_free(bv->bv_page, cc->page_pool); + mempool_free(bv->bv_page, &cc->page_pool); } } @@ -1484,7 +1492,7 @@ static void crypt_dec_pending(struct dm_crypt_io *io) crypt_free_req(cc, io->ctx.r.req, base_bio); if (unlikely(io->integrity_metadata_from_pool)) - mempool_free(io->integrity_metadata, io->cc->tag_pool); + mempool_free(io->integrity_metadata, &io->cc->tag_pool); else kfree(io->integrity_metadata); @@ -1557,7 +1565,7 @@ static int kcryptd_io_read(struct dm_crypt_io *io, gfp_t gfp) * biovecs we don't need to worry about the block layer * modifying the biovec array; so leverage bio_clone_fast(). */ - clone = bio_clone_fast(io->base_bio, gfp, cc->bs); + clone = bio_clone_fast(io->base_bio, gfp, &cc->bs); if (!clone) return 1; @@ -2155,6 +2163,43 @@ static int crypt_wipe_key(struct crypt_config *cc) return r; } +static void crypt_calculate_pages_per_client(void) +{ + unsigned long pages = (totalram_pages - totalhigh_pages) * DM_CRYPT_MEMORY_PERCENT / 100; + + if (!dm_crypt_clients_n) + return; + + pages /= dm_crypt_clients_n; + if (pages < DM_CRYPT_MIN_PAGES_PER_CLIENT) + pages = DM_CRYPT_MIN_PAGES_PER_CLIENT; + dm_crypt_pages_per_client = pages; +} + +static void *crypt_page_alloc(gfp_t gfp_mask, void *pool_data) +{ + struct crypt_config *cc = pool_data; + struct page *page; + + if (unlikely(percpu_counter_compare(&cc->n_allocated_pages, dm_crypt_pages_per_client) >= 0) && + likely(gfp_mask & __GFP_NORETRY)) + return NULL; + + page = alloc_page(gfp_mask); + if (likely(page != NULL)) + percpu_counter_add(&cc->n_allocated_pages, 1); + + return page; +} + +static void crypt_page_free(void *page, void *pool_data) +{ + struct crypt_config *cc = pool_data; + + __free_page(page); + percpu_counter_sub(&cc->n_allocated_pages, 1); +} + static void crypt_dtr(struct dm_target *ti) { struct crypt_config *cc = ti->private; @@ -2174,12 +2219,14 @@ static void crypt_dtr(struct dm_target *ti) crypt_free_tfms(cc); - if (cc->bs) - bioset_free(cc->bs); + bioset_exit(&cc->bs); - mempool_destroy(cc->page_pool); - mempool_destroy(cc->req_pool); - mempool_destroy(cc->tag_pool); + mempool_exit(&cc->page_pool); + mempool_exit(&cc->req_pool); + mempool_exit(&cc->tag_pool); + + WARN_ON(percpu_counter_sum(&cc->n_allocated_pages) != 0); + percpu_counter_destroy(&cc->n_allocated_pages); if (cc->iv_gen_ops && cc->iv_gen_ops->dtr) cc->iv_gen_ops->dtr(cc); @@ -2197,6 +2244,12 @@ static void crypt_dtr(struct dm_target *ti) /* Must zero key material before freeing */ kzfree(cc); + + spin_lock(&dm_crypt_clients_lock); + WARN_ON(!dm_crypt_clients_n); + dm_crypt_clients_n--; + crypt_calculate_pages_per_client(); + spin_unlock(&dm_crypt_clients_lock); } static int crypt_ctr_ivmode(struct dm_target *ti, const char *ivmode) @@ -2644,6 +2697,15 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) ti->private = cc; + spin_lock(&dm_crypt_clients_lock); + dm_crypt_clients_n++; + crypt_calculate_pages_per_client(); + spin_unlock(&dm_crypt_clients_lock); + + ret = percpu_counter_init(&cc->n_allocated_pages, 0, GFP_KERNEL); + if (ret < 0) + goto bad; + /* Optional parameters need to be read before cipher constructor */ if (argc > 5) { ret = crypt_ctr_optional(ti, argc - 5, &argv[5]); @@ -2679,8 +2741,6 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) iv_size_padding = align_mask; } - ret = -ENOMEM; - /* ...| IV + padding | original IV | original sec. number | bio tag offset | */ additional_req_size = sizeof(struct dm_crypt_request) + iv_size_padding + cc->iv_size + @@ -2688,8 +2748,8 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) sizeof(uint64_t) + sizeof(unsigned int); - cc->req_pool = mempool_create_kmalloc_pool(MIN_IOS, cc->dmreq_start + additional_req_size); - if (!cc->req_pool) { + ret = mempool_init_kmalloc_pool(&cc->req_pool, MIN_IOS, cc->dmreq_start + additional_req_size); + if (ret) { ti->error = "Cannot allocate crypt request mempool"; goto bad; } @@ -2698,14 +2758,14 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) ALIGN(sizeof(struct dm_crypt_io) + cc->dmreq_start + additional_req_size, ARCH_KMALLOC_MINALIGN); - cc->page_pool = mempool_create_page_pool(BIO_MAX_PAGES, 0); - if (!cc->page_pool) { + ret = mempool_init(&cc->page_pool, BIO_MAX_PAGES, crypt_page_alloc, crypt_page_free, cc); + if (ret) { ti->error = "Cannot allocate page mempool"; goto bad; } - cc->bs = bioset_create(MIN_IOS, 0, BIOSET_NEED_BVECS); - if (!cc->bs) { + ret = bioset_init(&cc->bs, MIN_IOS, 0, BIOSET_NEED_BVECS); + if (ret) { ti->error = "Cannot allocate crypt bioset"; goto bad; } @@ -2742,11 +2802,10 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) if (!cc->tag_pool_max_sectors) cc->tag_pool_max_sectors = 1; - cc->tag_pool = mempool_create_kmalloc_pool(MIN_IOS, + ret = mempool_init_kmalloc_pool(&cc->tag_pool, MIN_IOS, cc->tag_pool_max_sectors * cc->on_disk_tag_size); - if (!cc->tag_pool) { + if (ret) { ti->error = "Cannot allocate integrity tags mempool"; - ret = -ENOMEM; goto bad; } @@ -2839,7 +2898,7 @@ static int crypt_map(struct dm_target *ti, struct bio *bio) GFP_NOIO | __GFP_NORETRY | __GFP_NOMEMALLOC | __GFP_NOWARN)))) { if (bio_sectors(bio) > cc->tag_pool_max_sectors) dm_accept_partial_bio(bio, cc->tag_pool_max_sectors); - io->integrity_metadata = mempool_alloc(cc->tag_pool, GFP_NOIO); + io->integrity_metadata = mempool_alloc(&cc->tag_pool, GFP_NOIO); io->integrity_metadata_from_pool = true; } } @@ -2942,7 +3001,8 @@ static void crypt_resume(struct dm_target *ti) * key set <key> * key wipe */ -static int crypt_message(struct dm_target *ti, unsigned argc, char **argv) +static int crypt_message(struct dm_target *ti, unsigned argc, char **argv, + char *result, unsigned maxlen) { struct crypt_config *cc = ti->private; int key_size, ret = -EINVAL; diff --git a/drivers/md/dm-era-target.c b/drivers/md/dm-era-target.c index 73a5c198113a..8e48920a3ffa 100644 --- a/drivers/md/dm-era-target.c +++ b/drivers/md/dm-era-target.c @@ -1635,7 +1635,8 @@ err: DMEMIT("Error"); } -static int era_message(struct dm_target *ti, unsigned argc, char **argv) +static int era_message(struct dm_target *ti, unsigned argc, char **argv, + char *result, unsigned maxlen) { struct era *era = ti->private; diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c index 1b907b15f5c3..21d126a5078c 100644 --- a/drivers/md/dm-flakey.c +++ b/drivers/md/dm-flakey.c @@ -442,8 +442,7 @@ static void flakey_status(struct dm_target *ti, status_type_t type, } } -static int flakey_prepare_ioctl(struct dm_target *ti, - struct block_device **bdev, fmode_t *mode) +static int flakey_prepare_ioctl(struct dm_target *ti, struct block_device **bdev) { struct flakey_c *fc = ti->private; diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c index 46d7c8749222..fc68c7aaef8e 100644 --- a/drivers/md/dm-integrity.c +++ b/drivers/md/dm-integrity.c @@ -18,7 +18,7 @@ #include <crypto/hash.h> #include <crypto/skcipher.h> #include <linux/async_tx.h> -#include "dm-bufio.h" +#include <linux/dm-bufio.h> #define DM_MSG_PREFIX "integrity" @@ -142,7 +142,7 @@ struct dm_integrity_c { unsigned tag_size; __s8 log2_tag_size; sector_t start; - mempool_t *journal_io_mempool; + mempool_t journal_io_mempool; struct dm_io_client *io; struct dm_bufio_client *bufio; struct workqueue_struct *metadata_wq; @@ -1817,7 +1817,7 @@ static void complete_copy_from_journal(unsigned long error, void *context) struct journal_completion *comp = io->comp; struct dm_integrity_c *ic = comp->ic; remove_range(ic, &io->range); - mempool_free(io, ic->journal_io_mempool); + mempool_free(io, &ic->journal_io_mempool); if (unlikely(error != 0)) dm_integrity_io_error(ic, "copying from journal", -EIO); complete_journal_op(comp); @@ -1886,7 +1886,7 @@ static void do_journal_write(struct dm_integrity_c *ic, unsigned write_start, } next_loop = k - 1; - io = mempool_alloc(ic->journal_io_mempool, GFP_NOIO); + io = mempool_alloc(&ic->journal_io_mempool, GFP_NOIO); io->comp = ∁ io->range.logical_sector = sec; io->range.n_sectors = (k - j) << ic->sb->log2_sectors_per_block; @@ -1918,7 +1918,7 @@ static void do_journal_write(struct dm_integrity_c *ic, unsigned write_start, if (j == k) { remove_range_unlocked(ic, &io->range); spin_unlock_irq(&ic->endio_wait.lock); - mempool_free(io, ic->journal_io_mempool); + mempool_free(io, &ic->journal_io_mempool); goto skip_io; } for (l = j; l < k; l++) { @@ -2440,7 +2440,7 @@ static void dm_integrity_free_journal_scatterlist(struct dm_integrity_c *ic, str unsigned i; for (i = 0; i < ic->journal_sections; i++) kvfree(sl[i]); - kfree(sl); + kvfree(sl); } static struct scatterlist **dm_integrity_alloc_journal_scatterlist(struct dm_integrity_c *ic, struct page_list *pl) @@ -2548,6 +2548,9 @@ static int get_mac(struct crypto_shash **hash, struct alg_spec *a, char **error, *error = error_key; return r; } + } else if (crypto_shash_get_flags(*hash) & CRYPTO_TFM_NEED_KEY) { + *error = error_key; + return -ENOKEY; } } @@ -2977,9 +2980,8 @@ static int dm_integrity_ctr(struct dm_target *ti, unsigned argc, char **argv) goto bad; } - ic->journal_io_mempool = mempool_create_slab_pool(JOURNAL_IO_MEMPOOL, journal_io_cache); - if (!ic->journal_io_mempool) { - r = -ENOMEM; + r = mempool_init_slab_pool(&ic->journal_io_mempool, JOURNAL_IO_MEMPOOL, journal_io_cache); + if (r) { ti->error = "Cannot allocate mempool"; goto bad; } @@ -3193,7 +3195,7 @@ static void dm_integrity_dtr(struct dm_target *ti) destroy_workqueue(ic->writer_wq); if (ic->bufio) dm_bufio_client_destroy(ic->bufio); - mempool_destroy(ic->journal_io_mempool); + mempool_exit(&ic->journal_io_mempool); if (ic->io) dm_io_client_destroy(ic->io); if (ic->dev) diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c index a8d914d5abbe..81ffc59d05c9 100644 --- a/drivers/md/dm-io.c +++ b/drivers/md/dm-io.c @@ -22,8 +22,8 @@ #define DM_IO_MAX_REGIONS BITS_PER_LONG struct dm_io_client { - mempool_t *pool; - struct bio_set *bios; + mempool_t pool; + struct bio_set bios; }; /* @@ -49,32 +49,33 @@ struct dm_io_client *dm_io_client_create(void) { struct dm_io_client *client; unsigned min_ios = dm_get_reserved_bio_based_ios(); + int ret; - client = kmalloc(sizeof(*client), GFP_KERNEL); + client = kzalloc(sizeof(*client), GFP_KERNEL); if (!client) return ERR_PTR(-ENOMEM); - client->pool = mempool_create_slab_pool(min_ios, _dm_io_cache); - if (!client->pool) + ret = mempool_init_slab_pool(&client->pool, min_ios, _dm_io_cache); + if (ret) goto bad; - client->bios = bioset_create(min_ios, 0, BIOSET_NEED_BVECS); - if (!client->bios) + ret = bioset_init(&client->bios, min_ios, 0, BIOSET_NEED_BVECS); + if (ret) goto bad; return client; bad: - mempool_destroy(client->pool); + mempool_exit(&client->pool); kfree(client); - return ERR_PTR(-ENOMEM); + return ERR_PTR(ret); } EXPORT_SYMBOL(dm_io_client_create); void dm_io_client_destroy(struct dm_io_client *client) { - mempool_destroy(client->pool); - bioset_free(client->bios); + mempool_exit(&client->pool); + bioset_exit(&client->bios); kfree(client); } EXPORT_SYMBOL(dm_io_client_destroy); @@ -120,7 +121,7 @@ static void complete_io(struct io *io) invalidate_kernel_vmap_range(io->vma_invalidate_address, io->vma_invalidate_size); - mempool_free(io, io->client->pool); + mempool_free(io, &io->client->pool); fn(error_bits, context); } @@ -344,7 +345,7 @@ static void do_region(int op, int op_flags, unsigned region, dm_sector_div_up(remaining, (PAGE_SIZE >> SECTOR_SHIFT))); } - bio = bio_alloc_bioset(GFP_NOIO, num_bvecs, io->client->bios); + bio = bio_alloc_bioset(GFP_NOIO, num_bvecs, &io->client->bios); bio->bi_iter.bi_sector = where->sector + (where->count - remaining); bio_set_dev(bio, where->bdev); bio->bi_end_io = endio; @@ -442,7 +443,7 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions, init_completion(&sio.wait); - io = mempool_alloc(client->pool, GFP_NOIO); + io = mempool_alloc(&client->pool, GFP_NOIO); io->error_bits = 0; atomic_set(&io->count, 1); /* see dispatch_io() */ io->client = client; @@ -474,7 +475,7 @@ static int async_io(struct dm_io_client *client, unsigned int num_regions, return -EIO; } - io = mempool_alloc(client->pool, GFP_NOIO); + io = mempool_alloc(&client->pool, GFP_NOIO); io->error_bits = 0; atomic_set(&io->count, 1); /* see dispatch_io() */ io->client = client; diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index a89fd8f44453..5acf77de5945 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1595,7 +1595,7 @@ static int target_message(struct file *filp, struct dm_ioctl *param, size_t para DMWARN("Target message sector outside device."); r = -EINVAL; } else if (ti->type->message) - r = ti->type->message(ti, argc, argv); + r = ti->type->message(ti, argc, argv, result, maxlen); else { DMWARN("Target type does not support messages"); r = -EINVAL; diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c index e6e7c686646d..ce7efc7434be 100644 --- a/drivers/md/dm-kcopyd.c +++ b/drivers/md/dm-kcopyd.c @@ -47,7 +47,7 @@ struct dm_kcopyd_client { wait_queue_head_t destroyq; atomic_t nr_jobs; - mempool_t *job_pool; + mempool_t job_pool; struct workqueue_struct *kcopyd_wq; struct work_struct kcopyd_work; @@ -479,7 +479,7 @@ static int run_complete_job(struct kcopyd_job *job) */ if (job->master_job == job) { mutex_destroy(&job->lock); - mempool_free(job, kc->job_pool); + mempool_free(job, &kc->job_pool); } fn(read_err, write_err, context); @@ -751,7 +751,7 @@ int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from, * Allocate an array of jobs consisting of one master job * followed by SPLIT_COUNT sub jobs. */ - job = mempool_alloc(kc->job_pool, GFP_NOIO); + job = mempool_alloc(&kc->job_pool, GFP_NOIO); mutex_init(&job->lock); /* @@ -835,7 +835,7 @@ void *dm_kcopyd_prepare_callback(struct dm_kcopyd_client *kc, { struct kcopyd_job *job; - job = mempool_alloc(kc->job_pool, GFP_NOIO); + job = mempool_alloc(&kc->job_pool, GFP_NOIO); memset(job, 0, sizeof(struct kcopyd_job)); job->kc = kc; @@ -879,10 +879,10 @@ int kcopyd_cancel(struct kcopyd_job *job, int block) *---------------------------------------------------------------*/ struct dm_kcopyd_client *dm_kcopyd_client_create(struct dm_kcopyd_throttle *throttle) { - int r = -ENOMEM; + int r; struct dm_kcopyd_client *kc; - kc = kmalloc(sizeof(*kc), GFP_KERNEL); + kc = kzalloc(sizeof(*kc), GFP_KERNEL); if (!kc) return ERR_PTR(-ENOMEM); @@ -892,14 +892,16 @@ struct dm_kcopyd_client *dm_kcopyd_client_create(struct dm_kcopyd_throttle *thro INIT_LIST_HEAD(&kc->pages_jobs); kc->throttle = throttle; - kc->job_pool = mempool_create_slab_pool(MIN_JOBS, _job_cache); - if (!kc->job_pool) + r = mempool_init_slab_pool(&kc->job_pool, MIN_JOBS, _job_cache); + if (r) goto bad_slab; INIT_WORK(&kc->kcopyd_work, do_work); kc->kcopyd_wq = alloc_workqueue("kcopyd", WQ_MEM_RECLAIM, 0); - if (!kc->kcopyd_wq) + if (!kc->kcopyd_wq) { + r = -ENOMEM; goto bad_workqueue; + } kc->pages = NULL; kc->nr_reserved_pages = kc->nr_free_pages = 0; @@ -923,7 +925,7 @@ bad_io_client: bad_client_pages: destroy_workqueue(kc->kcopyd_wq); bad_workqueue: - mempool_destroy(kc->job_pool); + mempool_exit(&kc->job_pool); bad_slab: kfree(kc); @@ -942,7 +944,7 @@ void dm_kcopyd_client_destroy(struct dm_kcopyd_client *kc) destroy_workqueue(kc->kcopyd_wq); dm_io_client_destroy(kc->io_client); client_free_pages(kc); - mempool_destroy(kc->job_pool); + mempool_exit(&kc->job_pool); kfree(kc); } EXPORT_SYMBOL(dm_kcopyd_client_destroy); diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c index d5f8eff7c11d..775c06d953b7 100644 --- a/drivers/md/dm-linear.c +++ b/drivers/md/dm-linear.c @@ -59,6 +59,7 @@ static int linear_ctr(struct dm_target *ti, unsigned int argc, char **argv) ti->num_flush_bios = 1; ti->num_discard_bios = 1; + ti->num_secure_erase_bios = 1; ti->num_write_same_bios = 1; ti->num_write_zeroes_bios = 1; ti->private = lc; @@ -129,8 +130,7 @@ static void linear_status(struct dm_target *ti, status_type_t type, } } -static int linear_prepare_ioctl(struct dm_target *ti, - struct block_device **bdev, fmode_t *mode) +static int linear_prepare_ioctl(struct dm_target *ti, struct block_device **bdev) { struct linear_c *lc = (struct linear_c *) ti->private; struct dm_dev *dev = lc->dev; @@ -154,6 +154,7 @@ static int linear_iterate_devices(struct dm_target *ti, return fn(ti, lc->dev, lc->start, ti->len, data); } +#if IS_ENABLED(CONFIG_DAX_DRIVER) static long linear_dax_direct_access(struct dm_target *ti, pgoff_t pgoff, long nr_pages, void **kaddr, pfn_t *pfn) { @@ -184,6 +185,11 @@ static size_t linear_dax_copy_from_iter(struct dm_target *ti, pgoff_t pgoff, return dax_copy_from_iter(dax_dev, pgoff, addr, bytes, i); } +#else +#define linear_dax_direct_access NULL +#define linear_dax_copy_from_iter NULL +#endif + static struct target_type linear_target = { .name = "linear", .version = {1, 4, 0}, diff --git a/drivers/md/dm-log-userspace-base.c b/drivers/md/dm-log-userspace-base.c index 53b7b06d0aa8..52090bee17c2 100644 --- a/drivers/md/dm-log-userspace-base.c +++ b/drivers/md/dm-log-userspace-base.c @@ -76,7 +76,7 @@ struct log_c { */ uint32_t integrated_flush; - mempool_t *flush_entry_pool; + mempool_t flush_entry_pool; }; static struct kmem_cache *_flush_entry_cache; @@ -249,11 +249,10 @@ static int userspace_ctr(struct dm_dirty_log *log, struct dm_target *ti, goto out; } - lc->flush_entry_pool = mempool_create_slab_pool(FLUSH_ENTRY_POOL_SIZE, - _flush_entry_cache); - if (!lc->flush_entry_pool) { + r = mempool_init_slab_pool(&lc->flush_entry_pool, FLUSH_ENTRY_POOL_SIZE, + _flush_entry_cache); + if (r) { DMERR("Failed to create flush_entry_pool"); - r = -ENOMEM; goto out; } @@ -313,7 +312,7 @@ static int userspace_ctr(struct dm_dirty_log *log, struct dm_target *ti, out: kfree(devices_rdata); if (r) { - mempool_destroy(lc->flush_entry_pool); + mempool_exit(&lc->flush_entry_pool); kfree(lc); kfree(ctr_str); } else { @@ -342,7 +341,7 @@ static void userspace_dtr(struct dm_dirty_log *log) if (lc->log_dev) dm_put_device(lc->ti, lc->log_dev); - mempool_destroy(lc->flush_entry_pool); + mempool_exit(&lc->flush_entry_pool); kfree(lc->usr_argv_str); kfree(lc); @@ -570,7 +569,7 @@ static int userspace_flush(struct dm_dirty_log *log) int mark_list_is_empty; int clear_list_is_empty; struct dm_dirty_log_flush_entry *fe, *tmp_fe; - mempool_t *flush_entry_pool = lc->flush_entry_pool; + mempool_t *flush_entry_pool = &lc->flush_entry_pool; spin_lock_irqsave(&lc->flush_lock, flags); list_splice_init(&lc->mark_list, &mark_list); @@ -653,7 +652,7 @@ static void userspace_mark_region(struct dm_dirty_log *log, region_t region) struct dm_dirty_log_flush_entry *fe; /* Wait for an allocation, but _never_ fail */ - fe = mempool_alloc(lc->flush_entry_pool, GFP_NOIO); + fe = mempool_alloc(&lc->flush_entry_pool, GFP_NOIO); BUG_ON(!fe); spin_lock_irqsave(&lc->flush_lock, flags); @@ -687,7 +686,7 @@ static void userspace_clear_region(struct dm_dirty_log *log, region_t region) * to cause the region to be resync'ed when the * device is activated next time. */ - fe = mempool_alloc(lc->flush_entry_pool, GFP_ATOMIC); + fe = mempool_alloc(&lc->flush_entry_pool, GFP_ATOMIC); if (!fe) { DMERR("Failed to allocate memory to clear region."); return; diff --git a/drivers/md/dm-log-writes.c b/drivers/md/dm-log-writes.c index 3362d866793b..c90c7c08a77f 100644 --- a/drivers/md/dm-log-writes.c +++ b/drivers/md/dm-log-writes.c @@ -52,10 +52,11 @@ * in fact we want to do the data and the discard in the order that they * completed. */ -#define LOG_FLUSH_FLAG (1 << 0) -#define LOG_FUA_FLAG (1 << 1) -#define LOG_DISCARD_FLAG (1 << 2) -#define LOG_MARK_FLAG (1 << 3) +#define LOG_FLUSH_FLAG (1 << 0) +#define LOG_FUA_FLAG (1 << 1) +#define LOG_DISCARD_FLAG (1 << 2) +#define LOG_MARK_FLAG (1 << 3) +#define LOG_METADATA_FLAG (1 << 4) #define WRITE_LOG_VERSION 1ULL #define WRITE_LOG_MAGIC 0x6a736677736872ULL @@ -610,51 +611,6 @@ static int log_mark(struct log_writes_c *lc, char *data) return 0; } -static int log_dax(struct log_writes_c *lc, sector_t sector, size_t bytes, - struct iov_iter *i) -{ - struct pending_block *block; - - if (!bytes) - return 0; - - block = kzalloc(sizeof(struct pending_block), GFP_KERNEL); - if (!block) { - DMERR("Error allocating dax pending block"); - return -ENOMEM; - } - - block->data = kzalloc(bytes, GFP_KERNEL); - if (!block->data) { - DMERR("Error allocating dax data space"); - kfree(block); - return -ENOMEM; - } - - /* write data provided via the iterator */ - if (!copy_from_iter(block->data, bytes, i)) { - DMERR("Error copying dax data"); - kfree(block->data); - kfree(block); - return -EIO; - } - - /* rewind the iterator so that the block driver can use it */ - iov_iter_revert(i, bytes); - - block->datalen = bytes; - block->sector = bio_to_dev_sectors(lc, sector); - block->nr_sectors = ALIGN(bytes, lc->sectorsize) >> lc->sectorshift; - - atomic_inc(&lc->pending_blocks); - spin_lock_irq(&lc->blocks_lock); - list_add_tail(&block->list, &lc->unflushed_blocks); - spin_unlock_irq(&lc->blocks_lock); - wake_up_process(lc->log_kthread); - - return 0; -} - static void log_writes_dtr(struct dm_target *ti) { struct log_writes_c *lc = ti->private; @@ -699,6 +655,7 @@ static int log_writes_map(struct dm_target *ti, struct bio *bio) bool flush_bio = (bio->bi_opf & REQ_PREFLUSH); bool fua_bio = (bio->bi_opf & REQ_FUA); bool discard_bio = (bio_op(bio) == REQ_OP_DISCARD); + bool meta_bio = (bio->bi_opf & REQ_META); pb->block = NULL; @@ -743,6 +700,8 @@ static int log_writes_map(struct dm_target *ti, struct bio *bio) block->flags |= LOG_FUA_FLAG; if (discard_bio) block->flags |= LOG_DISCARD_FLAG; + if (meta_bio) + block->flags |= LOG_METADATA_FLAG; block->sector = bio_to_dev_sectors(lc, bio->bi_iter.bi_sector); block->nr_sectors = bio_to_dev_sectors(lc, bio_sectors(bio)); @@ -860,7 +819,7 @@ static void log_writes_status(struct dm_target *ti, status_type_t type, } static int log_writes_prepare_ioctl(struct dm_target *ti, - struct block_device **bdev, fmode_t *mode) + struct block_device **bdev) { struct log_writes_c *lc = ti->private; struct dm_dev *dev = lc->dev; @@ -887,7 +846,8 @@ static int log_writes_iterate_devices(struct dm_target *ti, * Messages supported: * mark <mark data> - specify the marked data. */ -static int log_writes_message(struct dm_target *ti, unsigned argc, char **argv) +static int log_writes_message(struct dm_target *ti, unsigned argc, char **argv, + char *result, unsigned maxlen) { int r = -EINVAL; struct log_writes_c *lc = ti->private; @@ -920,6 +880,52 @@ static void log_writes_io_hints(struct dm_target *ti, struct queue_limits *limit limits->io_min = limits->physical_block_size; } +#if IS_ENABLED(CONFIG_DAX_DRIVER) +static int log_dax(struct log_writes_c *lc, sector_t sector, size_t bytes, + struct iov_iter *i) +{ + struct pending_block *block; + + if (!bytes) + return 0; + + block = kzalloc(sizeof(struct pending_block), GFP_KERNEL); + if (!block) { + DMERR("Error allocating dax pending block"); + return -ENOMEM; + } + + block->data = kzalloc(bytes, GFP_KERNEL); + if (!block->data) { + DMERR("Error allocating dax data space"); + kfree(block); + return -ENOMEM; + } + + /* write data provided via the iterator */ + if (!copy_from_iter(block->data, bytes, i)) { + DMERR("Error copying dax data"); + kfree(block->data); + kfree(block); + return -EIO; + } + + /* rewind the iterator so that the block driver can use it */ + iov_iter_revert(i, bytes); + + block->datalen = bytes; + block->sector = bio_to_dev_sectors(lc, sector); + block->nr_sectors = ALIGN(bytes, lc->sectorsize) >> lc->sectorshift; + + atomic_inc(&lc->pending_blocks); + spin_lock_irq(&lc->blocks_lock); + list_add_tail(&block->list, &lc->unflushed_blocks); + spin_unlock_irq(&lc->blocks_lock); + wake_up_process(lc->log_kthread); + + return 0; +} + static long log_writes_dax_direct_access(struct dm_target *ti, pgoff_t pgoff, long nr_pages, void **kaddr, pfn_t *pfn) { @@ -956,6 +962,10 @@ static size_t log_writes_dax_copy_from_iter(struct dm_target *ti, dax_copy: return dax_copy_from_iter(lc->dev->dax_dev, pgoff, addr, bytes, i); } +#else +#define log_writes_dax_direct_access NULL +#define log_writes_dax_copy_from_iter NULL +#endif static struct target_type log_writes_target = { .name = "log-writes", diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index a6b7baf31cdd..d94ba6f72ff5 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -520,7 +520,8 @@ static int multipath_clone_and_map(struct dm_target *ti, struct request *rq, bdev = pgpath->path.dev->bdev; q = bdev_get_queue(bdev); - clone = blk_get_request(q, rq->cmd_flags | REQ_NOMERGE, GFP_ATOMIC); + clone = blk_get_request(q, rq->cmd_flags | REQ_NOMERGE, + BLK_MQ_REQ_NOWAIT); if (IS_ERR(clone)) { /* EBUSY, ENODEV or EWOULDBLOCK: requeue */ if (blk_queue_dying(q)) { @@ -714,7 +715,7 @@ static void process_queued_bios(struct work_struct *work) case DM_MAPIO_REMAPPED: generic_make_request(bio); break; - case 0: + case DM_MAPIO_SUBMITTED: break; default: WARN_ONCE(true, "__multipath_map_bio() returned %d\n", r); @@ -1811,7 +1812,8 @@ static void multipath_status(struct dm_target *ti, status_type_t type, spin_unlock_irqrestore(&m->lock, flags); } -static int multipath_message(struct dm_target *ti, unsigned argc, char **argv) +static int multipath_message(struct dm_target *ti, unsigned argc, char **argv, + char *result, unsigned maxlen) { int r = -EINVAL; struct dm_dev *dev; @@ -1875,7 +1877,7 @@ out: } static int multipath_prepare_ioctl(struct dm_target *ti, - struct block_device **bdev, fmode_t *mode) + struct block_device **bdev) { struct multipath *m = ti->private; struct pgpath *current_pgpath; @@ -1888,7 +1890,6 @@ static int multipath_prepare_ioctl(struct dm_target *ti, if (current_pgpath) { if (!test_bit(MPATHF_QUEUE_IO, &m->flags)) { *bdev = current_pgpath->path.dev->bdev; - *mode = current_pgpath->path.dev->mode; r = 0; } else { /* pg_init has not started or completed */ diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c index c1d1034ff7b7..6f823f44b4aa 100644 --- a/drivers/md/dm-raid.c +++ b/drivers/md/dm-raid.c @@ -1370,19 +1370,18 @@ static int parse_raid_params(struct raid_set *rs, struct dm_arg_set *as, * In device-mapper, we specify things in sectors, but * MD records this value in kB */ - value /= 2; - if (value > COUNTER_MAX) { + if (value < 0 || value / 2 > COUNTER_MAX) { rs->ti->error = "Max write-behind limit out of range"; return -EINVAL; } - rs->md.bitmap_info.max_write_behind = value; + rs->md.bitmap_info.max_write_behind = value / 2; } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_DAEMON_SLEEP))) { if (test_and_set_bit(__CTR_FLAG_DAEMON_SLEEP, &rs->ctr_flags)) { rs->ti->error = "Only one daemon_sleep argument pair allowed"; return -EINVAL; } - if (!value || (value > MAX_SCHEDULE_TIMEOUT)) { + if (value < 0) { rs->ti->error = "daemon sleep period out of range"; return -EINVAL; } @@ -1424,27 +1423,33 @@ static int parse_raid_params(struct raid_set *rs, struct dm_arg_set *as, return -EINVAL; } + if (value < 0) { + rs->ti->error = "Bogus stripe cache entries value"; + return -EINVAL; + } rs->stripe_cache_entries = value; } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_MIN_RECOVERY_RATE))) { if (test_and_set_bit(__CTR_FLAG_MIN_RECOVERY_RATE, &rs->ctr_flags)) { rs->ti->error = "Only one min_recovery_rate argument pair allowed"; return -EINVAL; } - if (value > INT_MAX) { + + if (value < 0) { rs->ti->error = "min_recovery_rate out of range"; return -EINVAL; } - rs->md.sync_speed_min = (int)value; + rs->md.sync_speed_min = value; } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_MAX_RECOVERY_RATE))) { if (test_and_set_bit(__CTR_FLAG_MAX_RECOVERY_RATE, &rs->ctr_flags)) { rs->ti->error = "Only one max_recovery_rate argument pair allowed"; return -EINVAL; } - if (value > INT_MAX) { + + if (value < 0) { rs->ti->error = "max_recovery_rate out of range"; return -EINVAL; } - rs->md.sync_speed_max = (int)value; + rs->md.sync_speed_max = value; } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_REGION_SIZE))) { if (test_and_set_bit(__CTR_FLAG_REGION_SIZE, &rs->ctr_flags)) { rs->ti->error = "Only one region_size argument pair allowed"; @@ -1490,6 +1495,12 @@ static int parse_raid_params(struct raid_set *rs, struct dm_arg_set *as, return -EINVAL; } + if (rs->md.sync_speed_max && + rs->md.sync_speed_min > rs->md.sync_speed_max) { + rs->ti->error = "Bogus recovery rates"; + return -EINVAL; + } + if (validate_region_size(rs, region_size)) return -EINVAL; @@ -3408,7 +3419,8 @@ static sector_t rs_get_progress(struct raid_set *rs, unsigned long recovery, set_bit(RT_FLAG_RS_IN_SYNC, &rs->runtime_flags); } else { - if (!test_bit(MD_RECOVERY_INTR, &recovery) && + if (!test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags) && + !test_bit(MD_RECOVERY_INTR, &recovery) && (test_bit(MD_RECOVERY_NEEDED, &recovery) || test_bit(MD_RECOVERY_RESHAPE, &recovery) || test_bit(MD_RECOVERY_RUNNING, &recovery))) @@ -3663,7 +3675,8 @@ static void raid_status(struct dm_target *ti, status_type_t type, } } -static int raid_message(struct dm_target *ti, unsigned int argc, char **argv) +static int raid_message(struct dm_target *ti, unsigned int argc, char **argv, + char *result, unsigned maxlen) { struct raid_set *rs = ti->private; struct mddev *mddev = &rs->md; diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index 580c49cc8079..5903e492bb34 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -23,6 +23,8 @@ #define MAX_RECOVERY 1 /* Maximum number of regions recovered in parallel. */ +#define MAX_NR_MIRRORS (DM_KCOPYD_MAX_REGIONS + 1) + #define DM_RAID1_HANDLE_ERRORS 0x01 #define DM_RAID1_KEEP_LOG 0x02 #define errors_handled(p) ((p)->features & DM_RAID1_HANDLE_ERRORS) @@ -255,7 +257,7 @@ static int mirror_flush(struct dm_target *ti) unsigned long error_bits; unsigned int i; - struct dm_io_region io[ms->nr_mirrors]; + struct dm_io_region io[MAX_NR_MIRRORS]; struct mirror *m; struct dm_io_request io_req = { .bi_op = REQ_OP_WRITE, @@ -651,7 +653,7 @@ static void write_callback(unsigned long error, void *context) static void do_write(struct mirror_set *ms, struct bio *bio) { unsigned int i; - struct dm_io_region io[ms->nr_mirrors], *dest = io; + struct dm_io_region io[MAX_NR_MIRRORS], *dest = io; struct mirror *m; struct dm_io_request io_req = { .bi_op = REQ_OP_WRITE, @@ -1083,7 +1085,7 @@ static int mirror_ctr(struct dm_target *ti, unsigned int argc, char **argv) argc -= args_used; if (!argc || sscanf(argv[0], "%u%c", &nr_mirrors, &dummy) != 1 || - nr_mirrors < 2 || nr_mirrors > DM_KCOPYD_MAX_REGIONS + 1) { + nr_mirrors < 2 || nr_mirrors > MAX_NR_MIRRORS) { ti->error = "Invalid number of mirrors"; dm_dirty_log_destroy(dl); return -EINVAL; @@ -1404,7 +1406,7 @@ static void mirror_status(struct dm_target *ti, status_type_t type, int num_feature_args = 0; struct mirror_set *ms = (struct mirror_set *) ti->private; struct dm_dirty_log *log = dm_rh_dirty_log(ms->rh); - char buffer[ms->nr_mirrors + 1]; + char buffer[MAX_NR_MIRRORS + 1]; switch (type) { case STATUSTYPE_INFO: diff --git a/drivers/md/dm-region-hash.c b/drivers/md/dm-region-hash.c index 85c32b22a420..abf3521b80a8 100644 --- a/drivers/md/dm-region-hash.c +++ b/drivers/md/dm-region-hash.c @@ -63,7 +63,7 @@ struct dm_region_hash { /* hash table */ rwlock_t hash_lock; - mempool_t *region_pool; + mempool_t region_pool; unsigned mask; unsigned nr_buckets; unsigned prime; @@ -169,6 +169,7 @@ struct dm_region_hash *dm_region_hash_create( struct dm_region_hash *rh; unsigned nr_buckets, max_buckets; size_t i; + int ret; /* * Calculate a suitable number of buckets for our hash @@ -179,7 +180,7 @@ struct dm_region_hash *dm_region_hash_create( ; nr_buckets >>= 1; - rh = kmalloc(sizeof(*rh), GFP_KERNEL); + rh = kzalloc(sizeof(*rh), GFP_KERNEL); if (!rh) { DMERR("unable to allocate region hash memory"); return ERR_PTR(-ENOMEM); @@ -220,9 +221,9 @@ struct dm_region_hash *dm_region_hash_create( INIT_LIST_HEAD(&rh->failed_recovered_regions); rh->flush_failure = 0; - rh->region_pool = mempool_create_kmalloc_pool(MIN_REGIONS, - sizeof(struct dm_region)); - if (!rh->region_pool) { + ret = mempool_init_kmalloc_pool(&rh->region_pool, MIN_REGIONS, + sizeof(struct dm_region)); + if (ret) { vfree(rh->buckets); kfree(rh); rh = ERR_PTR(-ENOMEM); @@ -242,14 +243,14 @@ void dm_region_hash_destroy(struct dm_region_hash *rh) list_for_each_entry_safe(reg, nreg, rh->buckets + h, hash_list) { BUG_ON(atomic_read(®->pending)); - mempool_free(reg, rh->region_pool); + mempool_free(reg, &rh->region_pool); } } if (rh->log) dm_dirty_log_destroy(rh->log); - mempool_destroy(rh->region_pool); + mempool_exit(&rh->region_pool); vfree(rh->buckets); kfree(rh); } @@ -287,7 +288,7 @@ static struct dm_region *__rh_alloc(struct dm_region_hash *rh, region_t region) { struct dm_region *reg, *nreg; - nreg = mempool_alloc(rh->region_pool, GFP_ATOMIC); + nreg = mempool_alloc(&rh->region_pool, GFP_ATOMIC); if (unlikely(!nreg)) nreg = kmalloc(sizeof(*nreg), GFP_NOIO | __GFP_NOFAIL); @@ -303,7 +304,7 @@ static struct dm_region *__rh_alloc(struct dm_region_hash *rh, region_t region) reg = __rh_lookup(rh, region); if (reg) /* We lost the race. */ - mempool_free(nreg, rh->region_pool); + mempool_free(nreg, &rh->region_pool); else { __rh_insert(rh, nreg); if (nreg->state == DM_RH_CLEAN) { @@ -481,17 +482,17 @@ void dm_rh_update_states(struct dm_region_hash *rh, int errors_handled) list_for_each_entry_safe(reg, next, &recovered, list) { rh->log->type->clear_region(rh->log, reg->key); complete_resync_work(reg, 1); - mempool_free(reg, rh->region_pool); + mempool_free(reg, &rh->region_pool); } list_for_each_entry_safe(reg, next, &failed_recovered, list) { complete_resync_work(reg, errors_handled ? 0 : 1); - mempool_free(reg, rh->region_pool); + mempool_free(reg, &rh->region_pool); } list_for_each_entry_safe(reg, next, &clean, list) { rh->log->type->clear_region(rh->log, reg->key); - mempool_free(reg, rh->region_pool); + mempool_free(reg, &rh->region_pool); } rh->log->type->flush(rh->log); diff --git a/drivers/md/dm-rq.c b/drivers/md/dm-rq.c index bf0b840645cc..6e547b8dd298 100644 --- a/drivers/md/dm-rq.c +++ b/drivers/md/dm-rq.c @@ -406,7 +406,7 @@ static blk_status_t dm_dispatch_clone_request(struct request *clone, struct requ if (blk_queue_io_stat(clone->q)) clone->rq_flags |= RQF_IO_STAT; - clone->start_time = jiffies; + clone->start_time_ns = ktime_get_ns(); r = blk_insert_cloned_request(clone->q, clone); if (r != BLK_STS_OK && r != BLK_STS_RESOURCE && r != BLK_STS_DEV_RESOURCE) /* must complete clone in terms of original request */ @@ -433,7 +433,7 @@ static int setup_clone(struct request *clone, struct request *rq, { int r; - r = blk_rq_prep_clone(clone, rq, tio->md->bs, gfp_mask, + r = blk_rq_prep_clone(clone, rq, &tio->md->bs, gfp_mask, dm_rq_bio_constructor, tio); if (r) return r; diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c index c5534d294773..3c50c4e4da8f 100644 --- a/drivers/md/dm-snap-persistent.c +++ b/drivers/md/dm-snap-persistent.c @@ -14,7 +14,7 @@ #include <linux/export.h> #include <linux/slab.h> #include <linux/dm-io.h> -#include "dm-bufio.h" +#include <linux/dm-bufio.h> #define DM_MSG_PREFIX "persistent snapshot" #define DM_CHUNK_SIZE_DEFAULT_SECTORS 32 /* 16KB */ diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 216035be5661..f745404da721 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -87,7 +87,7 @@ struct dm_snapshot { */ struct list_head out_of_order_list; - mempool_t *pending_pool; + mempool_t pending_pool; struct dm_exception_table pending; struct dm_exception_table complete; @@ -682,7 +682,7 @@ static void free_completed_exception(struct dm_exception *e) static struct dm_snap_pending_exception *alloc_pending_exception(struct dm_snapshot *s) { - struct dm_snap_pending_exception *pe = mempool_alloc(s->pending_pool, + struct dm_snap_pending_exception *pe = mempool_alloc(&s->pending_pool, GFP_NOIO); atomic_inc(&s->pending_exceptions_count); @@ -695,7 +695,7 @@ static void free_pending_exception(struct dm_snap_pending_exception *pe) { struct dm_snapshot *s = pe->snap; - mempool_free(pe, s->pending_pool); + mempool_free(pe, &s->pending_pool); smp_mb__before_atomic(); atomic_dec(&s->pending_exceptions_count); } @@ -1120,7 +1120,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) origin_mode = FMODE_WRITE; } - s = kmalloc(sizeof(*s), GFP_KERNEL); + s = kzalloc(sizeof(*s), GFP_KERNEL); if (!s) { ti->error = "Cannot allocate private snapshot structure"; r = -ENOMEM; @@ -1196,10 +1196,9 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) goto bad_kcopyd; } - s->pending_pool = mempool_create_slab_pool(MIN_IOS, pending_cache); - if (!s->pending_pool) { + r = mempool_init_slab_pool(&s->pending_pool, MIN_IOS, pending_cache); + if (r) { ti->error = "Could not allocate mempool for pending exceptions"; - r = -ENOMEM; goto bad_pending_pool; } @@ -1259,7 +1258,7 @@ bad_read_metadata: unregister_snapshot(s); bad_load_and_register: - mempool_destroy(s->pending_pool); + mempool_exit(&s->pending_pool); bad_pending_pool: dm_kcopyd_client_destroy(s->kcopyd_client); @@ -1355,7 +1354,7 @@ static void snapshot_dtr(struct dm_target *ti) while (atomic_read(&s->pending_exceptions_count)) msleep(1); /* - * Ensure instructions in mempool_destroy aren't reordered + * Ensure instructions in mempool_exit aren't reordered * before atomic_read. */ smp_mb(); @@ -1367,7 +1366,7 @@ static void snapshot_dtr(struct dm_target *ti) __free_exceptions(s); - mempool_destroy(s->pending_pool); + mempool_exit(&s->pending_pool); dm_exception_store_destroy(s->store); diff --git a/drivers/md/dm-stripe.c b/drivers/md/dm-stripe.c index b5e892149c54..fe7fb9b1aec3 100644 --- a/drivers/md/dm-stripe.c +++ b/drivers/md/dm-stripe.c @@ -169,6 +169,7 @@ static int stripe_ctr(struct dm_target *ti, unsigned int argc, char **argv) ti->num_flush_bios = stripes; ti->num_discard_bios = stripes; + ti->num_secure_erase_bios = stripes; ti->num_write_same_bios = stripes; ti->num_write_zeroes_bios = stripes; @@ -295,6 +296,7 @@ static int stripe_map(struct dm_target *ti, struct bio *bio) return DM_MAPIO_REMAPPED; } if (unlikely(bio_op(bio) == REQ_OP_DISCARD) || + unlikely(bio_op(bio) == REQ_OP_SECURE_ERASE) || unlikely(bio_op(bio) == REQ_OP_WRITE_ZEROES) || unlikely(bio_op(bio) == REQ_OP_WRITE_SAME)) { target_bio_nr = dm_bio_get_target_bio_nr(bio); @@ -311,6 +313,7 @@ static int stripe_map(struct dm_target *ti, struct bio *bio) return DM_MAPIO_REMAPPED; } +#if IS_ENABLED(CONFIG_DAX_DRIVER) static long stripe_dax_direct_access(struct dm_target *ti, pgoff_t pgoff, long nr_pages, void **kaddr, pfn_t *pfn) { @@ -351,6 +354,11 @@ static size_t stripe_dax_copy_from_iter(struct dm_target *ti, pgoff_t pgoff, return dax_copy_from_iter(dax_dev, pgoff, addr, bytes, i); } +#else +#define stripe_dax_direct_access NULL +#define stripe_dax_copy_from_iter NULL +#endif + /* * Stripe status: * @@ -368,7 +376,6 @@ static void stripe_status(struct dm_target *ti, status_type_t type, unsigned status_flags, char *result, unsigned maxlen) { struct stripe_c *sc = (struct stripe_c *) ti->private; - char buffer[sc->stripes + 1]; unsigned int sz = 0; unsigned int i; @@ -377,11 +384,12 @@ static void stripe_status(struct dm_target *ti, status_type_t type, DMEMIT("%d ", sc->stripes); for (i = 0; i < sc->stripes; i++) { DMEMIT("%s ", sc->stripe[i].dev->name); - buffer[i] = atomic_read(&(sc->stripe[i].error_count)) ? - 'D' : 'A'; } - buffer[i] = '\0'; - DMEMIT("1 %s", buffer); + DMEMIT("1 "); + for (i = 0; i < sc->stripes; i++) { + DMEMIT("%c", atomic_read(&(sc->stripe[i].error_count)) ? + 'D' : 'A'); + } break; case STATUSTYPE_TABLE: diff --git a/drivers/md/dm-switch.c b/drivers/md/dm-switch.c index 8d0ba879777e..7924a6a33ddc 100644 --- a/drivers/md/dm-switch.c +++ b/drivers/md/dm-switch.c @@ -466,7 +466,8 @@ static int process_set_region_mappings(struct switch_ctx *sctx, * * Only set_region_mappings is supported. */ -static int switch_message(struct dm_target *ti, unsigned argc, char **argv) +static int switch_message(struct dm_target *ti, unsigned argc, char **argv, + char *result, unsigned maxlen) { static DEFINE_MUTEX(message_mutex); @@ -511,8 +512,7 @@ static void switch_status(struct dm_target *ti, status_type_t type, * * Passthrough all ioctls to the path for sector 0 */ -static int switch_prepare_ioctl(struct dm_target *ti, - struct block_device **bdev, fmode_t *mode) +static int switch_prepare_ioctl(struct dm_target *ti, struct block_device **bdev) { struct switch_ctx *sctx = ti->private; unsigned path_nr; @@ -520,7 +520,6 @@ static int switch_prepare_ioctl(struct dm_target *ti, path_nr = switch_get_path_nr(sctx, 0); *bdev = sctx->path_list[path_nr].dmdev->bdev; - *mode = sctx->path_list[path_nr].dmdev->mode; /* * Only pass ioctls through if the device sizes match exactly. diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 954f4e3b68ac..0589a4da12bb 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -1846,6 +1846,34 @@ static bool dm_table_supports_discards(struct dm_table *t) return true; } +static int device_not_secure_erase_capable(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 q && !blk_queue_secure_erase(q); +} + +static bool dm_table_supports_secure_erase(struct dm_table *t) +{ + struct dm_target *ti; + unsigned int i; + + for (i = 0; i < dm_table_get_num_targets(t); i++) { + ti = dm_table_get_target(t, i); + + if (!ti->num_secure_erase_bios) + return false; + + if (!ti->type->iterate_devices || + ti->type->iterate_devices(ti, device_not_secure_erase_capable, NULL)) + return false; + } + + return true; +} + void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q, struct queue_limits *limits) { @@ -1867,6 +1895,9 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q, } else blk_queue_flag_set(QUEUE_FLAG_DISCARD, q); + if (dm_table_supports_secure_erase(t)) + blk_queue_flag_set(QUEUE_FLAG_SECERASE, q); + if (dm_table_supports_flush(t, (1UL << QUEUE_FLAG_WC))) { wc = true; if (dm_table_supports_flush(t, (1UL << QUEUE_FLAG_FUA))) diff --git a/drivers/md/dm-target.c b/drivers/md/dm-target.c index c0d7e60820c4..314d17ca6466 100644 --- a/drivers/md/dm-target.c +++ b/drivers/md/dm-target.c @@ -16,8 +16,6 @@ static LIST_HEAD(_targets); static DECLARE_RWSEM(_lock); -#define DM_MOD_NAME_SIZE 32 - static inline struct target_type *__find_target_type(const char *name) { struct target_type *tt; diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c index 629c555890c1..5772756c63c1 100644 --- a/drivers/md/dm-thin.c +++ b/drivers/md/dm-thin.c @@ -260,7 +260,7 @@ struct pool { struct dm_deferred_set *all_io_ds; struct dm_thin_new_mapping *next_mapping; - mempool_t *mapping_pool; + mempool_t mapping_pool; process_bio_fn process_bio; process_bio_fn process_discard; @@ -917,7 +917,7 @@ static void process_prepared_mapping_fail(struct dm_thin_new_mapping *m) { cell_error(m->tc->pool, m->cell); list_del(&m->list); - mempool_free(m, m->tc->pool->mapping_pool); + mempool_free(m, &m->tc->pool->mapping_pool); } static void process_prepared_mapping(struct dm_thin_new_mapping *m) @@ -961,7 +961,7 @@ static void process_prepared_mapping(struct dm_thin_new_mapping *m) out: list_del(&m->list); - mempool_free(m, pool->mapping_pool); + mempool_free(m, &pool->mapping_pool); } /*----------------------------------------------------------------*/ @@ -971,7 +971,7 @@ static void free_discard_mapping(struct dm_thin_new_mapping *m) struct thin_c *tc = m->tc; if (m->cell) cell_defer_no_holder(tc, m->cell); - mempool_free(m, tc->pool->mapping_pool); + mempool_free(m, &tc->pool->mapping_pool); } static void process_prepared_discard_fail(struct dm_thin_new_mapping *m) @@ -999,7 +999,7 @@ static void process_prepared_discard_no_passdown(struct dm_thin_new_mapping *m) bio_endio(m->bio); cell_defer_no_holder(tc, m->cell); - mempool_free(m, tc->pool->mapping_pool); + mempool_free(m, &tc->pool->mapping_pool); } /*----------------------------------------------------------------*/ @@ -1092,7 +1092,7 @@ static void process_prepared_discard_passdown_pt1(struct dm_thin_new_mapping *m) metadata_operation_failed(pool, "dm_thin_remove_range", r); bio_io_error(m->bio); cell_defer_no_holder(tc, m->cell); - mempool_free(m, pool->mapping_pool); + mempool_free(m, &pool->mapping_pool); return; } @@ -1105,7 +1105,7 @@ static void process_prepared_discard_passdown_pt1(struct dm_thin_new_mapping *m) metadata_operation_failed(pool, "dm_pool_inc_data_range", r); bio_io_error(m->bio); cell_defer_no_holder(tc, m->cell); - mempool_free(m, pool->mapping_pool); + mempool_free(m, &pool->mapping_pool); return; } @@ -1150,7 +1150,7 @@ static void process_prepared_discard_passdown_pt2(struct dm_thin_new_mapping *m) bio_endio(m->bio); cell_defer_no_holder(tc, m->cell); - mempool_free(m, pool->mapping_pool); + mempool_free(m, &pool->mapping_pool); } static void process_prepared(struct pool *pool, struct list_head *head, @@ -1196,7 +1196,7 @@ static int ensure_next_mapping(struct pool *pool) if (pool->next_mapping) return 0; - pool->next_mapping = mempool_alloc(pool->mapping_pool, GFP_ATOMIC); + pool->next_mapping = mempool_alloc(&pool->mapping_pool, GFP_ATOMIC); return pool->next_mapping ? 0 : -ENOMEM; } @@ -2835,8 +2835,8 @@ static void __pool_destroy(struct pool *pool) destroy_workqueue(pool->wq); if (pool->next_mapping) - mempool_free(pool->next_mapping, pool->mapping_pool); - mempool_destroy(pool->mapping_pool); + mempool_free(pool->next_mapping, &pool->mapping_pool); + mempool_exit(&pool->mapping_pool); dm_deferred_set_destroy(pool->shared_read_ds); dm_deferred_set_destroy(pool->all_io_ds); kfree(pool); @@ -2861,7 +2861,7 @@ static struct pool *pool_create(struct mapped_device *pool_md, return (struct pool *)pmd; } - pool = kmalloc(sizeof(*pool), GFP_KERNEL); + pool = kzalloc(sizeof(*pool), GFP_KERNEL); if (!pool) { *error = "Error allocating memory for pool"; err_p = ERR_PTR(-ENOMEM); @@ -2931,11 +2931,11 @@ static struct pool *pool_create(struct mapped_device *pool_md, } pool->next_mapping = NULL; - pool->mapping_pool = mempool_create_slab_pool(MAPPING_POOL_SIZE, - _new_mapping_cache); - if (!pool->mapping_pool) { + r = mempool_init_slab_pool(&pool->mapping_pool, MAPPING_POOL_SIZE, + _new_mapping_cache); + if (r) { *error = "Error creating pool's mapping mempool"; - err_p = ERR_PTR(-ENOMEM); + err_p = ERR_PTR(r); goto bad_mapping_pool; } @@ -2955,7 +2955,7 @@ static struct pool *pool_create(struct mapped_device *pool_md, return pool; bad_sort_array: - mempool_destroy(pool->mapping_pool); + mempool_exit(&pool->mapping_pool); bad_mapping_pool: dm_deferred_set_destroy(pool->all_io_ds); bad_all_io_ds: @@ -3705,7 +3705,8 @@ static int process_release_metadata_snap_mesg(unsigned argc, char **argv, struct * reserve_metadata_snap * release_metadata_snap */ -static int pool_message(struct dm_target *ti, unsigned argc, char **argv) +static int pool_message(struct dm_target *ti, unsigned argc, char **argv, + char *result, unsigned maxlen) { int r = -EINVAL; struct pool_c *pt = ti->private; diff --git a/drivers/md/dm-unstripe.c b/drivers/md/dm-unstripe.c index 65f838fa2e99..954b7ab4e684 100644 --- a/drivers/md/dm-unstripe.c +++ b/drivers/md/dm-unstripe.c @@ -7,12 +7,6 @@ #include "dm.h" #include <linux/module.h> -#include <linux/init.h> -#include <linux/blkdev.h> -#include <linux/bio.h> -#include <linux/slab.h> -#include <linux/bitops.h> -#include <linux/device-mapper.h> struct unstripe_c { struct dm_dev *dev; @@ -69,12 +63,6 @@ static int unstripe_ctr(struct dm_target *ti, unsigned int argc, char **argv) goto err; } - // FIXME: must support non power of 2 chunk_size, dm-stripe.c does - if (!is_power_of_2(uc->chunk_size)) { - ti->error = "Non power of 2 chunk_size is not supported yet"; - goto err; - } - if (kstrtouint(argv[2], 10, &uc->unstripe)) { ti->error = "Invalid stripe number"; goto err; @@ -98,7 +86,7 @@ static int unstripe_ctr(struct dm_target *ti, unsigned int argc, char **argv) uc->unstripe_offset = uc->unstripe * uc->chunk_size; uc->unstripe_width = (uc->stripes - 1) * uc->chunk_size; - uc->chunk_shift = fls(uc->chunk_size) - 1; + uc->chunk_shift = is_power_of_2(uc->chunk_size) ? fls(uc->chunk_size) - 1 : 0; tmp_len = ti->len; if (sector_div(tmp_len, uc->chunk_size)) { @@ -129,14 +117,18 @@ static sector_t map_to_core(struct dm_target *ti, struct bio *bio) { struct unstripe_c *uc = ti->private; sector_t sector = bio->bi_iter.bi_sector; + sector_t tmp_sector = sector; /* Shift us up to the right "row" on the stripe */ - sector += uc->unstripe_width * (sector >> uc->chunk_shift); + if (uc->chunk_shift) + tmp_sector >>= uc->chunk_shift; + else + sector_div(tmp_sector, uc->chunk_size); - /* Account for what stripe we're operating on */ - sector += uc->unstripe_offset; + sector += uc->unstripe_width * tmp_sector; - return sector; + /* Account for what stripe we're operating on */ + return sector + uc->unstripe_offset; } static int unstripe_map(struct dm_target *ti, struct bio *bio) @@ -185,7 +177,7 @@ static void unstripe_io_hints(struct dm_target *ti, static struct target_type unstripe_target = { .name = "unstriped", - .version = {1, 0, 0}, + .version = {1, 1, 0}, .module = THIS_MODULE, .ctr = unstripe_ctr, .dtr = unstripe_dtr, @@ -197,13 +189,7 @@ static struct target_type unstripe_target = { static int __init dm_unstripe_init(void) { - int r; - - r = dm_register_target(&unstripe_target); - if (r < 0) - DMERR("target registration failed"); - - return r; + return dm_register_target(&unstripe_target); } static void __exit dm_unstripe_exit(void) @@ -215,5 +201,6 @@ module_init(dm_unstripe_init); module_exit(dm_unstripe_exit); MODULE_DESCRIPTION(DM_NAME " unstriped target"); +MODULE_ALIAS("dm-unstriped"); MODULE_AUTHOR("Scott Bauer <scott.bauer@intel.com>"); MODULE_LICENSE("GPL"); diff --git a/drivers/md/dm-verity-fec.c b/drivers/md/dm-verity-fec.c index e13f90832b6b..684af08d0747 100644 --- a/drivers/md/dm-verity-fec.c +++ b/drivers/md/dm-verity-fec.c @@ -309,13 +309,13 @@ static int fec_alloc_bufs(struct dm_verity *v, struct dm_verity_fec_io *fio) unsigned n; if (!fio->rs) - fio->rs = mempool_alloc(v->fec->rs_pool, GFP_NOIO); + fio->rs = mempool_alloc(&v->fec->rs_pool, GFP_NOIO); fec_for_each_prealloc_buffer(n) { if (fio->bufs[n]) continue; - fio->bufs[n] = mempool_alloc(v->fec->prealloc_pool, GFP_NOWAIT); + fio->bufs[n] = mempool_alloc(&v->fec->prealloc_pool, GFP_NOWAIT); if (unlikely(!fio->bufs[n])) { DMERR("failed to allocate FEC buffer"); return -ENOMEM; @@ -327,7 +327,7 @@ static int fec_alloc_bufs(struct dm_verity *v, struct dm_verity_fec_io *fio) if (fio->bufs[n]) continue; - fio->bufs[n] = mempool_alloc(v->fec->extra_pool, GFP_NOWAIT); + fio->bufs[n] = mempool_alloc(&v->fec->extra_pool, GFP_NOWAIT); /* we can manage with even one buffer if necessary */ if (unlikely(!fio->bufs[n])) break; @@ -335,7 +335,7 @@ static int fec_alloc_bufs(struct dm_verity *v, struct dm_verity_fec_io *fio) fio->nbufs = n; if (!fio->output) - fio->output = mempool_alloc(v->fec->output_pool, GFP_NOIO); + fio->output = mempool_alloc(&v->fec->output_pool, GFP_NOIO); return 0; } @@ -493,15 +493,15 @@ void verity_fec_finish_io(struct dm_verity_io *io) if (!verity_fec_is_enabled(io->v)) return; - mempool_free(fio->rs, f->rs_pool); + mempool_free(fio->rs, &f->rs_pool); fec_for_each_prealloc_buffer(n) - mempool_free(fio->bufs[n], f->prealloc_pool); + mempool_free(fio->bufs[n], &f->prealloc_pool); fec_for_each_extra_buffer(fio, n) - mempool_free(fio->bufs[n], f->extra_pool); + mempool_free(fio->bufs[n], &f->extra_pool); - mempool_free(fio->output, f->output_pool); + mempool_free(fio->output, &f->output_pool); } /* @@ -549,9 +549,9 @@ void verity_fec_dtr(struct dm_verity *v) if (!verity_fec_is_enabled(v)) goto out; - mempool_destroy(f->rs_pool); - mempool_destroy(f->prealloc_pool); - mempool_destroy(f->extra_pool); + mempool_exit(&f->rs_pool); + mempool_exit(&f->prealloc_pool); + mempool_exit(&f->extra_pool); kmem_cache_destroy(f->cache); if (f->data_bufio) @@ -570,7 +570,7 @@ static void *fec_rs_alloc(gfp_t gfp_mask, void *pool_data) { struct dm_verity *v = (struct dm_verity *)pool_data; - return init_rs(8, 0x11d, 0, 1, v->fec->roots); + return init_rs_gfp(8, 0x11d, 0, 1, v->fec->roots, gfp_mask); } static void fec_rs_free(void *element, void *pool_data) @@ -675,6 +675,7 @@ int verity_fec_ctr(struct dm_verity *v) struct dm_verity_fec *f = v->fec; struct dm_target *ti = v->ti; u64 hash_blocks; + int ret; if (!verity_fec_is_enabled(v)) { verity_fec_dtr(v); @@ -770,11 +771,11 @@ int verity_fec_ctr(struct dm_verity *v) } /* Preallocate an rs_control structure for each worker thread */ - f->rs_pool = mempool_create(num_online_cpus(), fec_rs_alloc, - fec_rs_free, (void *) v); - if (!f->rs_pool) { + ret = mempool_init(&f->rs_pool, num_online_cpus(), fec_rs_alloc, + fec_rs_free, (void *) v); + if (ret) { ti->error = "Cannot allocate RS pool"; - return -ENOMEM; + return ret; } f->cache = kmem_cache_create("dm_verity_fec_buffers", @@ -786,26 +787,26 @@ int verity_fec_ctr(struct dm_verity *v) } /* Preallocate DM_VERITY_FEC_BUF_PREALLOC buffers for each thread */ - f->prealloc_pool = mempool_create_slab_pool(num_online_cpus() * - DM_VERITY_FEC_BUF_PREALLOC, - f->cache); - if (!f->prealloc_pool) { + ret = mempool_init_slab_pool(&f->prealloc_pool, num_online_cpus() * + DM_VERITY_FEC_BUF_PREALLOC, + f->cache); + if (ret) { ti->error = "Cannot allocate FEC buffer prealloc pool"; - return -ENOMEM; + return ret; } - f->extra_pool = mempool_create_slab_pool(0, f->cache); - if (!f->extra_pool) { + ret = mempool_init_slab_pool(&f->extra_pool, 0, f->cache); + if (ret) { ti->error = "Cannot allocate FEC buffer extra pool"; - return -ENOMEM; + return ret; } /* Preallocate an output buffer for each thread */ - f->output_pool = mempool_create_kmalloc_pool(num_online_cpus(), - 1 << v->data_dev_block_bits); - if (!f->output_pool) { + ret = mempool_init_kmalloc_pool(&f->output_pool, num_online_cpus(), + 1 << v->data_dev_block_bits); + if (ret) { ti->error = "Cannot allocate FEC output pool"; - return -ENOMEM; + return ret; } /* Reserve space for our per-bio data */ diff --git a/drivers/md/dm-verity-fec.h b/drivers/md/dm-verity-fec.h index bb31ce87a933..6ad803b2b36c 100644 --- a/drivers/md/dm-verity-fec.h +++ b/drivers/md/dm-verity-fec.h @@ -46,10 +46,10 @@ struct dm_verity_fec { sector_t hash_blocks; /* blocks covered after v->hash_start */ unsigned char roots; /* number of parity bytes, M-N of RS(M, N) */ unsigned char rsn; /* N of RS(M, N) */ - mempool_t *rs_pool; /* mempool for fio->rs */ - mempool_t *prealloc_pool; /* mempool for preallocated buffers */ - mempool_t *extra_pool; /* mempool for extra buffers */ - mempool_t *output_pool; /* mempool for output */ + mempool_t rs_pool; /* mempool for fio->rs */ + mempool_t prealloc_pool; /* mempool for preallocated buffers */ + mempool_t extra_pool; /* mempool for extra buffers */ + mempool_t output_pool; /* mempool for output */ struct kmem_cache *cache; /* cache for buffers */ }; diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c index aedb8222836b..fc893f636a98 100644 --- a/drivers/md/dm-verity-target.c +++ b/drivers/md/dm-verity-target.c @@ -32,6 +32,7 @@ #define DM_VERITY_OPT_LOGGING "ignore_corruption" #define DM_VERITY_OPT_RESTART "restart_on_corruption" #define DM_VERITY_OPT_IGN_ZEROES "ignore_zero_blocks" +#define DM_VERITY_OPT_AT_MOST_ONCE "check_at_most_once" #define DM_VERITY_OPTS_MAX (2 + DM_VERITY_OPTS_FEC) @@ -347,8 +348,8 @@ out: /* * Calculates the digest for the given bio */ -int verity_for_io_block(struct dm_verity *v, struct dm_verity_io *io, - struct bvec_iter *iter, struct crypto_wait *wait) +static int verity_for_io_block(struct dm_verity *v, struct dm_verity_io *io, + struct bvec_iter *iter, struct crypto_wait *wait) { unsigned int todo = 1 << v->data_dev_block_bits; struct bio *bio = dm_bio_from_per_bio_data(io, v->ti->per_io_data_size); @@ -433,6 +434,18 @@ static int verity_bv_zero(struct dm_verity *v, struct dm_verity_io *io, } /* + * Moves the bio iter one data block forward. + */ +static inline void verity_bv_skip_block(struct dm_verity *v, + struct dm_verity_io *io, + struct bvec_iter *iter) +{ + struct bio *bio = dm_bio_from_per_bio_data(io, v->ti->per_io_data_size); + + bio_advance_iter(bio, iter, 1 << v->data_dev_block_bits); +} + +/* * Verify one "dm_verity_io" structure. */ static int verity_verify_io(struct dm_verity_io *io) @@ -445,9 +458,16 @@ static int verity_verify_io(struct dm_verity_io *io) for (b = 0; b < io->n_blocks; b++) { int r; + sector_t cur_block = io->block + b; struct ahash_request *req = verity_io_hash_req(v, io); - r = verity_hash_for_block(v, io, io->block + b, + if (v->validated_blocks && + likely(test_bit(cur_block, v->validated_blocks))) { + verity_bv_skip_block(v, io, &io->iter); + continue; + } + + r = verity_hash_for_block(v, io, cur_block, verity_io_want_digest(v, io), &is_zero); if (unlikely(r < 0)) @@ -481,13 +501,16 @@ static int verity_verify_io(struct dm_verity_io *io) return r; if (likely(memcmp(verity_io_real_digest(v, io), - verity_io_want_digest(v, io), v->digest_size) == 0)) + verity_io_want_digest(v, io), v->digest_size) == 0)) { + if (v->validated_blocks) + set_bit(cur_block, v->validated_blocks); continue; + } else if (verity_fec_decode(v, io, DM_VERITY_BLOCK_TYPE_DATA, - io->block + b, NULL, &start) == 0) + cur_block, NULL, &start) == 0) continue; else if (verity_handle_err(v, DM_VERITY_BLOCK_TYPE_DATA, - io->block + b)) + cur_block)) return -EIO; } @@ -673,6 +696,8 @@ static void verity_status(struct dm_target *ti, status_type_t type, args += DM_VERITY_OPTS_FEC; if (v->zero_digest) args++; + if (v->validated_blocks) + args++; if (!args) return; DMEMIT(" %u", args); @@ -691,13 +716,14 @@ static void verity_status(struct dm_target *ti, status_type_t type, } if (v->zero_digest) DMEMIT(" " DM_VERITY_OPT_IGN_ZEROES); + if (v->validated_blocks) + DMEMIT(" " DM_VERITY_OPT_AT_MOST_ONCE); sz = verity_fec_status_table(v, sz, result, maxlen); break; } } -static int verity_prepare_ioctl(struct dm_target *ti, - struct block_device **bdev, fmode_t *mode) +static int verity_prepare_ioctl(struct dm_target *ti, struct block_device **bdev) { struct dm_verity *v = ti->private; @@ -740,6 +766,7 @@ static void verity_dtr(struct dm_target *ti) if (v->bufio) dm_bufio_client_destroy(v->bufio); + kvfree(v->validated_blocks); kfree(v->salt); kfree(v->root_digest); kfree(v->zero_digest); @@ -760,6 +787,26 @@ static void verity_dtr(struct dm_target *ti) kfree(v); } +static int verity_alloc_most_once(struct dm_verity *v) +{ + struct dm_target *ti = v->ti; + + /* the bitset can only handle INT_MAX blocks */ + if (v->data_blocks > INT_MAX) { + ti->error = "device too large to use check_at_most_once"; + return -E2BIG; + } + + v->validated_blocks = kvzalloc(BITS_TO_LONGS(v->data_blocks) * + sizeof(unsigned long), GFP_KERNEL); + if (!v->validated_blocks) { + ti->error = "failed to allocate bitset for check_at_most_once"; + return -ENOMEM; + } + + return 0; +} + static int verity_alloc_zero_digest(struct dm_verity *v) { int r = -ENOMEM; @@ -829,6 +876,12 @@ static int verity_parse_opt_args(struct dm_arg_set *as, struct dm_verity *v) } continue; + } else if (!strcasecmp(arg_name, DM_VERITY_OPT_AT_MOST_ONCE)) { + r = verity_alloc_most_once(v); + if (r) + return r; + continue; + } else if (verity_is_fec_opt_arg(arg_name)) { r = verity_fec_parse_opt_args(as, v, &argc, arg_name); if (r) @@ -1096,7 +1149,7 @@ bad: static struct target_type verity_target = { .name = "verity", - .version = {1, 3, 0}, + .version = {1, 4, 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 b675bc015512..3441c10b840c 100644 --- a/drivers/md/dm-verity.h +++ b/drivers/md/dm-verity.h @@ -12,7 +12,7 @@ #ifndef DM_VERITY_H #define DM_VERITY_H -#include "dm-bufio.h" +#include <linux/dm-bufio.h> #include <linux/device-mapper.h> #include <crypto/hash.h> @@ -63,6 +63,7 @@ struct dm_verity { sector_t hash_level_block[DM_VERITY_MAX_LEVELS]; struct dm_verity_fec *fec; /* forward error correction */ + unsigned long *validated_blocks; /* bitset blocks validated */ }; struct dm_verity_io { diff --git a/drivers/md/dm-zoned-target.c b/drivers/md/dm-zoned-target.c index caff02caf083..30602d15ad9a 100644 --- a/drivers/md/dm-zoned-target.c +++ b/drivers/md/dm-zoned-target.c @@ -57,7 +57,7 @@ struct dmz_target { struct workqueue_struct *chunk_wq; /* For cloned BIOs to zones */ - struct bio_set *bio_set; + struct bio_set bio_set; /* For flush */ spinlock_t flush_lock; @@ -121,7 +121,7 @@ static int dmz_submit_read_bio(struct dmz_target *dmz, struct dm_zone *zone, } /* Partial BIO: we need to clone the BIO */ - clone = bio_clone_fast(bio, GFP_NOIO, dmz->bio_set); + clone = bio_clone_fast(bio, GFP_NOIO, &dmz->bio_set); if (!clone) return -ENOMEM; @@ -779,10 +779,9 @@ static int dmz_ctr(struct dm_target *ti, unsigned int argc, char **argv) ti->len = (sector_t)dmz_nr_chunks(dmz->metadata) << dev->zone_nr_sectors_shift; /* Zone BIO */ - dmz->bio_set = bioset_create(DMZ_MIN_BIOS, 0, 0); - if (!dmz->bio_set) { + ret = bioset_init(&dmz->bio_set, DMZ_MIN_BIOS, 0, 0); + if (ret) { ti->error = "Create BIO set failed"; - ret = -ENOMEM; goto err_meta; } @@ -828,7 +827,7 @@ err_cwq: destroy_workqueue(dmz->chunk_wq); err_bio: mutex_destroy(&dmz->chunk_lock); - bioset_free(dmz->bio_set); + bioset_exit(&dmz->bio_set); err_meta: dmz_dtr_metadata(dmz->metadata); err_dev: @@ -858,7 +857,7 @@ static void dmz_dtr(struct dm_target *ti) dmz_dtr_metadata(dmz->metadata); - bioset_free(dmz->bio_set); + bioset_exit(&dmz->bio_set); dmz_put_zoned_device(ti); @@ -898,8 +897,7 @@ static void dmz_io_hints(struct dm_target *ti, struct queue_limits *limits) /* * Pass on ioctl to the backend device. */ -static int dmz_prepare_ioctl(struct dm_target *ti, - struct block_device **bdev, fmode_t *mode) +static int dmz_prepare_ioctl(struct dm_target *ti, struct block_device **bdev) { struct dmz_target *dmz = ti->private; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index ded74e1eb0d1..98dff36b89a3 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -148,8 +148,8 @@ static int dm_numa_node = DM_NUMA_NODE; * For mempools pre-allocation at the table loading time. */ struct dm_md_mempools { - struct bio_set *bs; - struct bio_set *io_bs; + struct bio_set bs; + struct bio_set io_bs; }; struct table_device { @@ -458,67 +458,56 @@ static int dm_blk_getgeo(struct block_device *bdev, struct hd_geometry *geo) return dm_get_geometry(md, geo); } -static char *_dm_claim_ptr = "I belong to device-mapper"; - -static int dm_get_bdev_for_ioctl(struct mapped_device *md, - struct block_device **bdev, - fmode_t *mode) +static int dm_prepare_ioctl(struct mapped_device *md, int *srcu_idx, + struct block_device **bdev) + __acquires(md->io_barrier) { struct dm_target *tgt; struct dm_table *map; - int srcu_idx, r, r2; + int r; retry: r = -ENOTTY; - map = dm_get_live_table(md, &srcu_idx); + map = dm_get_live_table(md, srcu_idx); if (!map || !dm_table_get_size(map)) - goto out; + return r; /* We only support devices that have a single target */ if (dm_table_get_num_targets(map) != 1) - goto out; + return r; tgt = dm_table_get_target(map, 0); if (!tgt->type->prepare_ioctl) - goto out; - - if (dm_suspended_md(md)) { - r = -EAGAIN; - goto out; - } - - r = tgt->type->prepare_ioctl(tgt, bdev, mode); - if (r < 0) - goto out; - - bdgrab(*bdev); - r2 = blkdev_get(*bdev, *mode, _dm_claim_ptr); - if (r2 < 0) { - r = r2; - goto out; - } + return r; - dm_put_live_table(md, srcu_idx); - return r; + if (dm_suspended_md(md)) + return -EAGAIN; -out: - dm_put_live_table(md, srcu_idx); + r = tgt->type->prepare_ioctl(tgt, bdev); if (r == -ENOTCONN && !fatal_signal_pending(current)) { + dm_put_live_table(md, *srcu_idx); msleep(10); goto retry; } + return r; } +static void dm_unprepare_ioctl(struct mapped_device *md, int srcu_idx) + __releases(md->io_barrier) +{ + dm_put_live_table(md, srcu_idx); +} + static int dm_blk_ioctl(struct block_device *bdev, fmode_t mode, unsigned int cmd, unsigned long arg) { struct mapped_device *md = bdev->bd_disk->private_data; - int r; + int r, srcu_idx; - r = dm_get_bdev_for_ioctl(md, &bdev, &mode); + r = dm_prepare_ioctl(md, &srcu_idx, &bdev); if (r < 0) - return r; + goto out; if (r > 0) { /* @@ -536,7 +525,7 @@ static int dm_blk_ioctl(struct block_device *bdev, fmode_t mode, r = __blkdev_driver_ioctl(bdev, mode, cmd, arg); out: - blkdev_put(bdev, mode); + dm_unprepare_ioctl(md, srcu_idx); return r; } @@ -548,7 +537,7 @@ static struct dm_io *alloc_io(struct mapped_device *md, struct bio *bio) struct dm_target_io *tio; struct bio *clone; - clone = bio_alloc_bioset(GFP_NOIO, 0, md->io_bs); + clone = bio_alloc_bioset(GFP_NOIO, 0, &md->io_bs); if (!clone) return NULL; @@ -583,7 +572,7 @@ static struct dm_target_io *alloc_tio(struct clone_info *ci, struct dm_target *t /* the dm_target_io embedded in ci->io is available */ tio = &ci->io->tio; } else { - struct bio *clone = bio_alloc_bioset(gfp_mask, 0, ci->io->md->bs); + struct bio *clone = bio_alloc_bioset(gfp_mask, 0, &ci->io->md->bs); if (!clone) return NULL; @@ -710,6 +699,8 @@ static void dm_put_live_table_fast(struct mapped_device *md) __releases(RCU) rcu_read_unlock(); } +static char *_dm_claim_ptr = "I belong to device-mapper"; + /* * Open a table device so we can use it as a map destination. */ @@ -1029,7 +1020,8 @@ int dm_set_target_max_io_len(struct dm_target *ti, sector_t len) EXPORT_SYMBOL_GPL(dm_set_target_max_io_len); static struct dm_target *dm_dax_get_live_target(struct mapped_device *md, - sector_t sector, int *srcu_idx) + sector_t sector, int *srcu_idx) + __acquires(md->io_barrier) { struct dm_table *map; struct dm_target *ti; @@ -1046,7 +1038,7 @@ static struct dm_target *dm_dax_get_live_target(struct mapped_device *md, } static long dm_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, - long nr_pages, void **kaddr, pfn_t *pfn) + long nr_pages, void **kaddr, pfn_t *pfn) { struct mapped_device *md = dax_get_private(dax_dev); sector_t sector = pgoff * PAGE_SECTORS; @@ -1074,7 +1066,7 @@ static long dm_dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, } static size_t dm_dax_copy_from_iter(struct dax_device *dax_dev, pgoff_t pgoff, - void *addr, size_t bytes, struct iov_iter *i) + void *addr, size_t bytes, struct iov_iter *i) { struct mapped_device *md = dax_get_private(dax_dev); sector_t sector = pgoff * PAGE_SECTORS; @@ -1414,6 +1406,11 @@ static unsigned get_num_discard_bios(struct dm_target *ti) return ti->num_discard_bios; } +static unsigned get_num_secure_erase_bios(struct dm_target *ti) +{ + return ti->num_secure_erase_bios; +} + static unsigned get_num_write_same_bios(struct dm_target *ti) { return ti->num_write_same_bios; @@ -1467,6 +1464,11 @@ static int __send_discard(struct clone_info *ci, struct dm_target *ti) is_split_required_for_discard); } +static int __send_secure_erase(struct clone_info *ci, struct dm_target *ti) +{ + return __send_changing_extent_only(ci, ti, get_num_secure_erase_bios, NULL); +} + static int __send_write_same(struct clone_info *ci, struct dm_target *ti) { return __send_changing_extent_only(ci, ti, get_num_write_same_bios, NULL); @@ -1477,6 +1479,25 @@ static int __send_write_zeroes(struct clone_info *ci, struct dm_target *ti) return __send_changing_extent_only(ci, ti, get_num_write_zeroes_bios, NULL); } +static bool __process_abnormal_io(struct clone_info *ci, struct dm_target *ti, + int *result) +{ + struct bio *bio = ci->bio; + + if (bio_op(bio) == REQ_OP_DISCARD) + *result = __send_discard(ci, ti); + else if (bio_op(bio) == REQ_OP_SECURE_ERASE) + *result = __send_secure_erase(ci, ti); + else if (bio_op(bio) == REQ_OP_WRITE_SAME) + *result = __send_write_same(ci, ti); + else if (bio_op(bio) == REQ_OP_WRITE_ZEROES) + *result = __send_write_zeroes(ci, ti); + else + return false; + + return true; +} + /* * Select the correct strategy for processing a non-flush bio. */ @@ -1491,12 +1512,8 @@ static int __split_and_process_non_flush(struct clone_info *ci) if (!dm_target_is_valid(ti)) return -EIO; - if (unlikely(bio_op(bio) == REQ_OP_DISCARD)) - return __send_discard(ci, ti); - else if (unlikely(bio_op(bio) == REQ_OP_WRITE_SAME)) - return __send_write_same(ci, ti); - else if (unlikely(bio_op(bio) == REQ_OP_WRITE_ZEROES)) - return __send_write_zeroes(ci, ti); + if (unlikely(__process_abnormal_io(ci, ti, &r))) + return r; if (bio_op(bio) == REQ_OP_ZONE_REPORT) len = ci->sector_count; @@ -1566,7 +1583,7 @@ static blk_qc_t __split_and_process_bio(struct mapped_device *md, * won't be affected by this reassignment. */ struct bio *b = bio_clone_bioset(bio, GFP_NOIO, - md->queue->bio_split); + &md->queue->bio_split); ci.io->orig_bio = b; bio_advance(bio, (bio_sectors(bio) - ci.sector_count) << 9); bio_chain(b, bio); @@ -1617,9 +1634,12 @@ static blk_qc_t __process_bio(struct mapped_device *md, goto out; } - tio = alloc_tio(&ci, ti, 0, GFP_NOIO); ci.bio = bio; ci.sector_count = bio_sectors(bio); + if (unlikely(__process_abnormal_io(&ci, ti, &error))) + goto out; + + tio = alloc_tio(&ci, ti, 0, GFP_NOIO); ret = __clone_and_map_simple_bio(&ci, tio, NULL); } out: @@ -1765,10 +1785,8 @@ static void cleanup_mapped_device(struct mapped_device *md) destroy_workqueue(md->wq); if (md->kworker_task) kthread_stop(md->kworker_task); - if (md->bs) - bioset_free(md->bs); - if (md->io_bs) - bioset_free(md->io_bs); + bioset_exit(&md->bs); + bioset_exit(&md->io_bs); if (md->dax_dev) { kill_dax(md->dax_dev); @@ -1807,7 +1825,7 @@ static void cleanup_mapped_device(struct mapped_device *md) static struct mapped_device *alloc_dev(int minor) { int r, numa_node_id = dm_get_numa_node(); - struct dax_device *dax_dev; + struct dax_device *dax_dev = NULL; struct mapped_device *md; void *old_md; @@ -1873,9 +1891,11 @@ static struct mapped_device *alloc_dev(int minor) md->disk->private_data = md; sprintf(md->disk->disk_name, "dm-%d", minor); - dax_dev = alloc_dax(md, md->disk->disk_name, &dm_dax_ops); - if (!dax_dev) - goto bad; + if (IS_ENABLED(CONFIG_DAX_DRIVER)) { + dax_dev = alloc_dax(md, md->disk->disk_name, &dm_dax_ops); + if (!dax_dev) + goto bad; + } md->dax_dev = dax_dev; add_disk_no_queue_reg(md->disk); @@ -1943,16 +1963,10 @@ static void __bind_mempools(struct mapped_device *md, struct dm_table *t) * If so, reload bioset because front_pad may have changed * because a different table was loaded. */ - if (md->bs) { - bioset_free(md->bs); - md->bs = NULL; - } - if (md->io_bs) { - bioset_free(md->io_bs); - md->io_bs = NULL; - } + bioset_exit(&md->bs); + bioset_exit(&md->io_bs); - } else if (md->bs) { + } else if (bioset_initialized(&md->bs)) { /* * There's no need to reload with request-based dm * because the size of front_pad doesn't change. @@ -1964,12 +1978,14 @@ static void __bind_mempools(struct mapped_device *md, struct dm_table *t) goto out; } - BUG_ON(!p || md->bs || md->io_bs); + BUG_ON(!p || + bioset_initialized(&md->bs) || + bioset_initialized(&md->io_bs)); md->bs = p->bs; - p->bs = NULL; + memset(&p->bs, 0, sizeof(p->bs)); md->io_bs = p->io_bs; - p->io_bs = NULL; + memset(&p->io_bs, 0, sizeof(p->io_bs)); out: /* mempool bind completed, no longer need any mempools in the table */ dm_table_free_md_mempools(t); @@ -2883,6 +2899,7 @@ struct dm_md_mempools *dm_alloc_md_mempools(struct mapped_device *md, enum dm_qu struct dm_md_mempools *pools = kzalloc_node(sizeof(*pools), GFP_KERNEL, md->numa_node_id); unsigned int pool_size = 0; unsigned int front_pad, io_front_pad; + int ret; if (!pools) return NULL; @@ -2894,10 +2911,10 @@ struct dm_md_mempools *dm_alloc_md_mempools(struct mapped_device *md, enum dm_qu pool_size = max(dm_get_reserved_bio_based_ios(), min_pool_size); front_pad = roundup(per_io_data_size, __alignof__(struct dm_target_io)) + offsetof(struct dm_target_io, clone); io_front_pad = roundup(front_pad, __alignof__(struct dm_io)) + offsetof(struct dm_io, tio); - pools->io_bs = bioset_create(pool_size, io_front_pad, 0); - if (!pools->io_bs) + ret = bioset_init(&pools->io_bs, pool_size, io_front_pad, 0); + if (ret) goto out; - if (integrity && bioset_integrity_create(pools->io_bs, pool_size)) + if (integrity && bioset_integrity_create(&pools->io_bs, pool_size)) goto out; break; case DM_TYPE_REQUEST_BASED: @@ -2910,11 +2927,11 @@ struct dm_md_mempools *dm_alloc_md_mempools(struct mapped_device *md, enum dm_qu BUG(); } - pools->bs = bioset_create(pool_size, front_pad, 0); - if (!pools->bs) + ret = bioset_init(&pools->bs, pool_size, front_pad, 0); + if (ret) goto out; - if (integrity && bioset_integrity_create(pools->bs, pool_size)) + if (integrity && bioset_integrity_create(&pools->bs, pool_size)) goto out; return pools; @@ -2930,10 +2947,8 @@ void dm_free_md_mempools(struct dm_md_mempools *pools) if (!pools) return; - if (pools->bs) - bioset_free(pools->bs); - if (pools->io_bs) - bioset_free(pools->io_bs); + bioset_exit(&pools->bs); + bioset_exit(&pools->io_bs); kfree(pools); } @@ -3015,20 +3030,19 @@ static int dm_pr_reserve(struct block_device *bdev, u64 key, enum pr_type type, { struct mapped_device *md = bdev->bd_disk->private_data; const struct pr_ops *ops; - fmode_t mode; - int r; + int r, srcu_idx; - r = dm_get_bdev_for_ioctl(md, &bdev, &mode); + r = dm_prepare_ioctl(md, &srcu_idx, &bdev); if (r < 0) - return r; + goto out; ops = bdev->bd_disk->fops->pr_ops; if (ops && ops->pr_reserve) r = ops->pr_reserve(bdev, key, type, flags); else r = -EOPNOTSUPP; - - blkdev_put(bdev, mode); +out: + dm_unprepare_ioctl(md, srcu_idx); return r; } @@ -3036,20 +3050,19 @@ static int dm_pr_release(struct block_device *bdev, u64 key, enum pr_type type) { struct mapped_device *md = bdev->bd_disk->private_data; const struct pr_ops *ops; - fmode_t mode; - int r; + int r, srcu_idx; - r = dm_get_bdev_for_ioctl(md, &bdev, &mode); + r = dm_prepare_ioctl(md, &srcu_idx, &bdev); if (r < 0) - return r; + goto out; ops = bdev->bd_disk->fops->pr_ops; if (ops && ops->pr_release) r = ops->pr_release(bdev, key, type); else r = -EOPNOTSUPP; - - blkdev_put(bdev, mode); +out: + dm_unprepare_ioctl(md, srcu_idx); return r; } @@ -3058,20 +3071,19 @@ static int dm_pr_preempt(struct block_device *bdev, u64 old_key, u64 new_key, { struct mapped_device *md = bdev->bd_disk->private_data; const struct pr_ops *ops; - fmode_t mode; - int r; + int r, srcu_idx; - r = dm_get_bdev_for_ioctl(md, &bdev, &mode); + r = dm_prepare_ioctl(md, &srcu_idx, &bdev); if (r < 0) - return r; + goto out; ops = bdev->bd_disk->fops->pr_ops; if (ops && ops->pr_preempt) r = ops->pr_preempt(bdev, old_key, new_key, type, abort); else r = -EOPNOTSUPP; - - blkdev_put(bdev, mode); +out: + dm_unprepare_ioctl(md, srcu_idx); return r; } @@ -3079,20 +3091,19 @@ static int dm_pr_clear(struct block_device *bdev, u64 key) { struct mapped_device *md = bdev->bd_disk->private_data; const struct pr_ops *ops; - fmode_t mode; - int r; + int r, srcu_idx; - r = dm_get_bdev_for_ioctl(md, &bdev, &mode); + r = dm_prepare_ioctl(md, &srcu_idx, &bdev); if (r < 0) - return r; + goto out; ops = bdev->bd_disk->fops->pr_ops; if (ops && ops->pr_clear) r = ops->pr_clear(bdev, key); else r = -EOPNOTSUPP; - - blkdev_put(bdev, mode); +out: + dm_unprepare_ioctl(md, srcu_idx); return r; } diff --git a/drivers/md/md-faulty.c b/drivers/md/md-faulty.c index 38264b38420f..c2fdf899de14 100644 --- a/drivers/md/md-faulty.c +++ b/drivers/md/md-faulty.c @@ -214,7 +214,7 @@ static bool faulty_make_request(struct mddev *mddev, struct bio *bio) } } if (failit) { - struct bio *b = bio_clone_fast(bio, GFP_NOIO, mddev->bio_set); + struct bio *b = bio_clone_fast(bio, GFP_NOIO, &mddev->bio_set); bio_set_dev(b, conf->rdev->bdev); b->bi_private = bio; diff --git a/drivers/md/md-linear.c b/drivers/md/md-linear.c index 4964323d936b..d45c697c0ebe 100644 --- a/drivers/md/md-linear.c +++ b/drivers/md/md-linear.c @@ -269,7 +269,7 @@ static bool linear_make_request(struct mddev *mddev, struct bio *bio) 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); + GFP_NOIO, &mddev->bio_set); bio_chain(split, bio); generic_make_request(bio); bio = split; diff --git a/drivers/md/md-multipath.c b/drivers/md/md-multipath.c index 0a7e99d62c69..f71fcdb9b39c 100644 --- a/drivers/md/md-multipath.c +++ b/drivers/md/md-multipath.c @@ -80,7 +80,7 @@ static void multipath_end_bh_io(struct multipath_bh *mp_bh, blk_status_t status) bio->bi_status = status; bio_endio(bio); - mempool_free(mp_bh, conf->pool); + mempool_free(mp_bh, &conf->pool); } static void multipath_end_request(struct bio *bio) @@ -117,7 +117,7 @@ static bool multipath_make_request(struct mddev *mddev, struct bio * bio) return true; } - mp_bh = mempool_alloc(conf->pool, GFP_NOIO); + mp_bh = mempool_alloc(&conf->pool, GFP_NOIO); mp_bh->master_bio = bio; mp_bh->mddev = mddev; @@ -125,7 +125,7 @@ static bool multipath_make_request(struct mddev *mddev, struct bio * bio) mp_bh->path = multipath_map(conf); if (mp_bh->path < 0) { bio_io_error(bio); - mempool_free(mp_bh, conf->pool); + mempool_free(mp_bh, &conf->pool); return true; } multipath = conf->multipaths + mp_bh->path; @@ -378,6 +378,7 @@ static int multipath_run (struct mddev *mddev) struct multipath_info *disk; struct md_rdev *rdev; int working_disks; + int ret; if (md_check_no_bitmap(mddev)) return -EINVAL; @@ -431,9 +432,9 @@ static int multipath_run (struct mddev *mddev) } mddev->degraded = conf->raid_disks - working_disks; - conf->pool = mempool_create_kmalloc_pool(NR_RESERVED_BUFS, - sizeof(struct multipath_bh)); - if (conf->pool == NULL) + ret = mempool_init_kmalloc_pool(&conf->pool, NR_RESERVED_BUFS, + sizeof(struct multipath_bh)); + if (ret) goto out_free_conf; mddev->thread = md_register_thread(multipathd, mddev, @@ -455,7 +456,7 @@ static int multipath_run (struct mddev *mddev) return 0; out_free_conf: - mempool_destroy(conf->pool); + mempool_exit(&conf->pool); kfree(conf->multipaths); kfree(conf); mddev->private = NULL; @@ -467,7 +468,7 @@ static void multipath_free(struct mddev *mddev, void *priv) { struct mpconf *conf = priv; - mempool_destroy(conf->pool); + mempool_exit(&conf->pool); kfree(conf->multipaths); kfree(conf); } diff --git a/drivers/md/md-multipath.h b/drivers/md/md-multipath.h index 0adb941f485a..b3099e5fc4d7 100644 --- a/drivers/md/md-multipath.h +++ b/drivers/md/md-multipath.h @@ -13,7 +13,7 @@ struct mpconf { spinlock_t device_lock; struct list_head retry_list; - mempool_t *pool; + mempool_t pool; }; /* diff --git a/drivers/md/md.c b/drivers/md/md.c index 3bea45e8ccff..fc692b7128bb 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -193,10 +193,10 @@ struct bio *bio_alloc_mddev(gfp_t gfp_mask, int nr_iovecs, { struct bio *b; - if (!mddev || !mddev->bio_set) + if (!mddev || !bioset_initialized(&mddev->bio_set)) return bio_alloc(gfp_mask, nr_iovecs); - b = bio_alloc_bioset(gfp_mask, nr_iovecs, mddev->bio_set); + b = bio_alloc_bioset(gfp_mask, nr_iovecs, &mddev->bio_set); if (!b) return NULL; return b; @@ -205,10 +205,10 @@ EXPORT_SYMBOL_GPL(bio_alloc_mddev); static struct bio *md_bio_alloc_sync(struct mddev *mddev) { - if (!mddev || !mddev->sync_set) + if (!mddev || !bioset_initialized(&mddev->sync_set)) return bio_alloc(GFP_NOIO, 1); - return bio_alloc_bioset(GFP_NOIO, 1, mddev->sync_set); + return bio_alloc_bioset(GFP_NOIO, 1, &mddev->sync_set); } /* @@ -510,7 +510,10 @@ static void mddev_delayed_delete(struct work_struct *ws); static void mddev_put(struct mddev *mddev) { - struct bio_set *bs = NULL, *sync_bs = NULL; + struct bio_set bs, sync_bs; + + memset(&bs, 0, sizeof(bs)); + memset(&sync_bs, 0, sizeof(sync_bs)); if (!atomic_dec_and_lock(&mddev->active, &all_mddevs_lock)) return; @@ -521,8 +524,8 @@ static void mddev_put(struct mddev *mddev) list_del_init(&mddev->all_mddevs); bs = mddev->bio_set; sync_bs = mddev->sync_set; - mddev->bio_set = NULL; - mddev->sync_set = NULL; + memset(&mddev->bio_set, 0, sizeof(mddev->bio_set)); + memset(&mddev->sync_set, 0, sizeof(mddev->sync_set)); if (mddev->gendisk) { /* We did a probe so need to clean up. Call * queue_work inside the spinlock so that @@ -535,10 +538,8 @@ static void mddev_put(struct mddev *mddev) kfree(mddev); } spin_unlock(&all_mddevs_lock); - if (bs) - bioset_free(bs); - if (sync_bs) - bioset_free(sync_bs); + bioset_exit(&bs); + bioset_exit(&sync_bs); } static void md_safemode_timeout(struct timer_list *t); @@ -2123,7 +2124,7 @@ int md_integrity_register(struct mddev *mddev) bdev_get_integrity(reference->bdev)); pr_debug("md: data integrity enabled on %s\n", mdname(mddev)); - if (bioset_integrity_create(mddev->bio_set, BIO_POOL_SIZE)) { + if (bioset_integrity_create(&mddev->bio_set, BIO_POOL_SIZE)) { pr_err("md: failed to create integrity pool for %s\n", mdname(mddev)); return -EINVAL; @@ -5497,17 +5498,15 @@ int md_run(struct mddev *mddev) sysfs_notify_dirent_safe(rdev->sysfs_state); } - if (mddev->bio_set == NULL) { - mddev->bio_set = bioset_create(BIO_POOL_SIZE, 0, BIOSET_NEED_BVECS); - if (!mddev->bio_set) - return -ENOMEM; + if (!bioset_initialized(&mddev->bio_set)) { + err = bioset_init(&mddev->bio_set, BIO_POOL_SIZE, 0, BIOSET_NEED_BVECS); + if (err) + return err; } - if (mddev->sync_set == NULL) { - mddev->sync_set = bioset_create(BIO_POOL_SIZE, 0, BIOSET_NEED_BVECS); - if (!mddev->sync_set) { - err = -ENOMEM; + if (!bioset_initialized(&mddev->sync_set)) { + err = bioset_init(&mddev->sync_set, BIO_POOL_SIZE, 0, BIOSET_NEED_BVECS); + if (err) goto abort; - } } spin_lock(&pers_lock); @@ -5668,14 +5667,8 @@ int md_run(struct mddev *mddev) return 0; abort: - if (mddev->bio_set) { - bioset_free(mddev->bio_set); - mddev->bio_set = NULL; - } - if (mddev->sync_set) { - bioset_free(mddev->sync_set); - mddev->sync_set = NULL; - } + bioset_exit(&mddev->bio_set); + bioset_exit(&mddev->sync_set); return err; } @@ -5888,14 +5881,8 @@ void md_stop(struct mddev *mddev) * This is called from dm-raid */ __md_stop(mddev); - if (mddev->bio_set) { - bioset_free(mddev->bio_set); - mddev->bio_set = NULL; - } - if (mddev->sync_set) { - bioset_free(mddev->sync_set); - mddev->sync_set = NULL; - } + bioset_exit(&mddev->bio_set); + bioset_exit(&mddev->sync_set); } EXPORT_SYMBOL_GPL(md_stop); @@ -9256,8 +9243,10 @@ void md_reload_sb(struct mddev *mddev, int nr) check_sb_changes(mddev, rdev); /* Read all rdev's to update recovery_offset */ - rdev_for_each_rcu(rdev, mddev) - read_rdev(mddev, rdev); + rdev_for_each_rcu(rdev, mddev) { + if (!test_bit(Faulty, &rdev->flags)) + read_rdev(mddev, rdev); + } } EXPORT_SYMBOL(md_reload_sb); diff --git a/drivers/md/md.h b/drivers/md/md.h index fbc925cce810..3507cab22cb6 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -452,8 +452,8 @@ struct mddev { struct attribute_group *to_remove; - struct bio_set *bio_set; - struct bio_set *sync_set; /* for sync operations like + struct bio_set bio_set; + struct bio_set sync_set; /* for sync operations like * metadata and bitmap writes */ diff --git a/drivers/md/persistent-data/dm-block-manager.c b/drivers/md/persistent-data/dm-block-manager.c index ea15d220ced7..492a3f8ac119 100644 --- a/drivers/md/persistent-data/dm-block-manager.c +++ b/drivers/md/persistent-data/dm-block-manager.c @@ -5,8 +5,8 @@ */ #include "dm-block-manager.h" #include "dm-persistent-data-internal.h" -#include "../dm-bufio.h" +#include <linux/dm-bufio.h> #include <linux/crc32c.h> #include <linux/module.h> #include <linux/slab.h> diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c index 584c10347267..65ae47a02218 100644 --- a/drivers/md/raid0.c +++ b/drivers/md/raid0.c @@ -479,7 +479,7 @@ static void raid0_handle_discard(struct mddev *mddev, struct bio *bio) if (bio_end_sector(bio) > zone->zone_end) { struct bio *split = bio_split(bio, zone->zone_end - bio->bi_iter.bi_sector, GFP_NOIO, - mddev->bio_set); + &mddev->bio_set); bio_chain(split, bio); generic_make_request(bio); bio = split; @@ -582,7 +582,8 @@ static bool raid0_make_request(struct mddev *mddev, struct bio *bio) sector = bio_sector; if (sectors < bio_sectors(bio)) { - struct bio *split = bio_split(bio, sectors, GFP_NOIO, mddev->bio_set); + struct bio *split = bio_split(bio, sectors, GFP_NOIO, + &mddev->bio_set); bio_chain(split, bio); generic_make_request(bio); bio = split; diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index e2943fb74056..bad28520719b 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -221,7 +221,7 @@ static void free_r1bio(struct r1bio *r1_bio) struct r1conf *conf = r1_bio->mddev->private; put_all_bios(conf, r1_bio); - mempool_free(r1_bio, conf->r1bio_pool); + mempool_free(r1_bio, &conf->r1bio_pool); } static void put_buf(struct r1bio *r1_bio) @@ -236,7 +236,7 @@ static void put_buf(struct r1bio *r1_bio) rdev_dec_pending(conf->mirrors[i].rdev, r1_bio->mddev); } - mempool_free(r1_bio, conf->r1buf_pool); + mempool_free(r1_bio, &conf->r1buf_pool); lower_barrier(conf, sect); } @@ -854,7 +854,7 @@ static void flush_pending_writes(struct r1conf *conf) * there is no normal IO happeing. It must arrange to call * lower_barrier when the particular background IO completes. */ -static void raise_barrier(struct r1conf *conf, sector_t sector_nr) +static sector_t raise_barrier(struct r1conf *conf, sector_t sector_nr) { int idx = sector_to_idx(sector_nr); @@ -885,13 +885,23 @@ static void raise_barrier(struct r1conf *conf, sector_t sector_nr) * max resync count which allowed on current I/O barrier bucket. */ wait_event_lock_irq(conf->wait_barrier, - !conf->array_frozen && + (!conf->array_frozen && !atomic_read(&conf->nr_pending[idx]) && - atomic_read(&conf->barrier[idx]) < RESYNC_DEPTH, + atomic_read(&conf->barrier[idx]) < RESYNC_DEPTH) || + test_bit(MD_RECOVERY_INTR, &conf->mddev->recovery), conf->resync_lock); + if (test_bit(MD_RECOVERY_INTR, &conf->mddev->recovery)) { + atomic_dec(&conf->barrier[idx]); + spin_unlock_irq(&conf->resync_lock); + wake_up(&conf->wait_barrier); + return -EINTR; + } + atomic_inc(&conf->nr_sync_pending); spin_unlock_irq(&conf->resync_lock); + + return 0; } static void lower_barrier(struct r1conf *conf, sector_t sector_nr) @@ -1092,6 +1102,8 @@ static void alloc_behind_master_bio(struct r1bio *r1_bio, goto skip_copy; } + behind_bio->bi_write_hint = bio->bi_write_hint; + while (i < vcnt && size) { struct page *page; int len = min_t(int, PAGE_SIZE, size); @@ -1166,7 +1178,7 @@ alloc_r1bio(struct mddev *mddev, struct bio *bio) struct r1conf *conf = mddev->private; struct r1bio *r1_bio; - r1_bio = mempool_alloc(conf->r1bio_pool, GFP_NOIO); + r1_bio = mempool_alloc(&conf->r1bio_pool, GFP_NOIO); /* Ensure no bio records IO_BLOCKED */ memset(r1_bio->bios, 0, conf->raid_disks * sizeof(r1_bio->bios[0])); init_r1bio(r1_bio, mddev, bio); @@ -1256,7 +1268,7 @@ static void raid1_read_request(struct mddev *mddev, struct bio *bio, if (max_sectors < bio_sectors(bio)) { struct bio *split = bio_split(bio, max_sectors, - gfp, conf->bio_split); + gfp, &conf->bio_split); bio_chain(split, bio); generic_make_request(bio); bio = split; @@ -1266,7 +1278,7 @@ static void raid1_read_request(struct mddev *mddev, struct bio *bio, r1_bio->read_disk = rdisk; - read_bio = bio_clone_fast(bio, gfp, mddev->bio_set); + read_bio = bio_clone_fast(bio, gfp, &mddev->bio_set); r1_bio->bios[rdisk] = read_bio; @@ -1427,7 +1439,7 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio, if (max_sectors < bio_sectors(bio)) { struct bio *split = bio_split(bio, max_sectors, - GFP_NOIO, conf->bio_split); + GFP_NOIO, &conf->bio_split); bio_chain(split, bio); generic_make_request(bio); bio = split; @@ -1467,9 +1479,9 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio, if (r1_bio->behind_master_bio) mbio = bio_clone_fast(r1_bio->behind_master_bio, - GFP_NOIO, mddev->bio_set); + GFP_NOIO, &mddev->bio_set); else - mbio = bio_clone_fast(bio, GFP_NOIO, mddev->bio_set); + mbio = bio_clone_fast(bio, GFP_NOIO, &mddev->bio_set); if (r1_bio->behind_master_bio) { if (test_bit(WriteMostly, &conf->mirrors[i].rdev->flags)) @@ -1645,8 +1657,7 @@ static void close_sync(struct r1conf *conf) _allow_barrier(conf, idx); } - mempool_destroy(conf->r1buf_pool); - conf->r1buf_pool = NULL; + mempool_exit(&conf->r1buf_pool); } static int raid1_spare_active(struct mddev *mddev) @@ -2336,10 +2347,10 @@ static int narrow_write_error(struct r1bio *r1_bio, int i) if (test_bit(R1BIO_BehindIO, &r1_bio->state)) { wbio = bio_clone_fast(r1_bio->behind_master_bio, GFP_NOIO, - mddev->bio_set); + &mddev->bio_set); } else { wbio = bio_clone_fast(r1_bio->master_bio, GFP_NOIO, - mddev->bio_set); + &mddev->bio_set); } bio_set_op_attrs(wbio, REQ_OP_WRITE, 0); @@ -2552,17 +2563,15 @@ static int init_resync(struct r1conf *conf) int buffs; buffs = RESYNC_WINDOW / RESYNC_BLOCK_SIZE; - BUG_ON(conf->r1buf_pool); - conf->r1buf_pool = mempool_create(buffs, r1buf_pool_alloc, r1buf_pool_free, - conf->poolinfo); - if (!conf->r1buf_pool) - return -ENOMEM; - return 0; + BUG_ON(mempool_initialized(&conf->r1buf_pool)); + + return mempool_init(&conf->r1buf_pool, buffs, r1buf_pool_alloc, + r1buf_pool_free, conf->poolinfo); } static struct r1bio *raid1_alloc_init_r1buf(struct r1conf *conf) { - struct r1bio *r1bio = mempool_alloc(conf->r1buf_pool, GFP_NOIO); + struct r1bio *r1bio = mempool_alloc(&conf->r1buf_pool, GFP_NOIO); struct resync_pages *rps; struct bio *bio; int i; @@ -2605,7 +2614,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr, int idx = sector_to_idx(sector_nr); int page_idx = 0; - if (!conf->r1buf_pool) + if (!mempool_initialized(&conf->r1buf_pool)) if (init_resync(conf)) return 0; @@ -2662,9 +2671,12 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr, bitmap_cond_end_sync(mddev->bitmap, sector_nr, mddev_is_clustered(mddev) && (sector_nr + 2 * RESYNC_SECTORS > conf->cluster_sync_high)); - r1_bio = raid1_alloc_init_r1buf(conf); - raise_barrier(conf, sector_nr); + + if (raise_barrier(conf, sector_nr)) + return 0; + + r1_bio = raid1_alloc_init_r1buf(conf); rcu_read_lock(); /* @@ -2938,14 +2950,13 @@ static struct r1conf *setup_conf(struct mddev *mddev) if (!conf->poolinfo) goto abort; conf->poolinfo->raid_disks = mddev->raid_disks * 2; - conf->r1bio_pool = mempool_create(NR_RAID1_BIOS, r1bio_pool_alloc, - r1bio_pool_free, - conf->poolinfo); - if (!conf->r1bio_pool) + err = mempool_init(&conf->r1bio_pool, NR_RAID1_BIOS, r1bio_pool_alloc, + r1bio_pool_free, conf->poolinfo); + if (err) goto abort; - conf->bio_split = bioset_create(BIO_POOL_SIZE, 0, 0); - if (!conf->bio_split) + err = bioset_init(&conf->bio_split, BIO_POOL_SIZE, 0, 0); + if (err) goto abort; conf->poolinfo->mddev = mddev; @@ -3018,7 +3029,7 @@ static struct r1conf *setup_conf(struct mddev *mddev) abort: if (conf) { - mempool_destroy(conf->r1bio_pool); + mempool_exit(&conf->r1bio_pool); kfree(conf->mirrors); safe_put_page(conf->tmppage); kfree(conf->poolinfo); @@ -3026,8 +3037,7 @@ static struct r1conf *setup_conf(struct mddev *mddev) kfree(conf->nr_waiting); kfree(conf->nr_queued); kfree(conf->barrier); - if (conf->bio_split) - bioset_free(conf->bio_split); + bioset_exit(&conf->bio_split); kfree(conf); } return ERR_PTR(err); @@ -3129,7 +3139,7 @@ static void raid1_free(struct mddev *mddev, void *priv) { struct r1conf *conf = priv; - mempool_destroy(conf->r1bio_pool); + mempool_exit(&conf->r1bio_pool); kfree(conf->mirrors); safe_put_page(conf->tmppage); kfree(conf->poolinfo); @@ -3137,8 +3147,7 @@ static void raid1_free(struct mddev *mddev, void *priv) kfree(conf->nr_waiting); kfree(conf->nr_queued); kfree(conf->barrier); - if (conf->bio_split) - bioset_free(conf->bio_split); + bioset_exit(&conf->bio_split); kfree(conf); } @@ -3184,13 +3193,17 @@ static int raid1_reshape(struct mddev *mddev) * At the same time, we "pack" the devices so that all the missing * devices have the higher raid_disk numbers. */ - mempool_t *newpool, *oldpool; + mempool_t newpool, oldpool; struct pool_info *newpoolinfo; struct raid1_info *newmirrors; struct r1conf *conf = mddev->private; int cnt, raid_disks; unsigned long flags; int d, d2; + int ret; + + memset(&newpool, 0, sizeof(newpool)); + memset(&oldpool, 0, sizeof(oldpool)); /* Cannot change chunk_size, layout, or level */ if (mddev->chunk_sectors != mddev->new_chunk_sectors || @@ -3222,17 +3235,17 @@ static int raid1_reshape(struct mddev *mddev) newpoolinfo->mddev = mddev; newpoolinfo->raid_disks = raid_disks * 2; - newpool = mempool_create(NR_RAID1_BIOS, r1bio_pool_alloc, - r1bio_pool_free, newpoolinfo); - if (!newpool) { + ret = mempool_init(&newpool, NR_RAID1_BIOS, r1bio_pool_alloc, + r1bio_pool_free, newpoolinfo); + if (ret) { kfree(newpoolinfo); - return -ENOMEM; + return ret; } newmirrors = kzalloc(sizeof(struct raid1_info) * raid_disks * 2, GFP_KERNEL); if (!newmirrors) { kfree(newpoolinfo); - mempool_destroy(newpool); + mempool_exit(&newpool); return -ENOMEM; } @@ -3272,7 +3285,7 @@ static int raid1_reshape(struct mddev *mddev) set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); md_wakeup_thread(mddev->thread); - mempool_destroy(oldpool); + mempool_exit(&oldpool); return 0; } diff --git a/drivers/md/raid1.h b/drivers/md/raid1.h index eb84bc68e2fd..e7ccad898736 100644 --- a/drivers/md/raid1.h +++ b/drivers/md/raid1.h @@ -118,10 +118,10 @@ struct r1conf { * mempools - it changes when the array grows or shrinks */ struct pool_info *poolinfo; - mempool_t *r1bio_pool; - mempool_t *r1buf_pool; + mempool_t r1bio_pool; + mempool_t r1buf_pool; - struct bio_set *bio_split; + struct bio_set bio_split; /* temporary buffer to synchronous IO when attempting to repair * a read error. diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 3c60774c8430..37d4b236b81b 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -291,14 +291,14 @@ static void free_r10bio(struct r10bio *r10_bio) struct r10conf *conf = r10_bio->mddev->private; put_all_bios(conf, r10_bio); - mempool_free(r10_bio, conf->r10bio_pool); + mempool_free(r10_bio, &conf->r10bio_pool); } static void put_buf(struct r10bio *r10_bio) { struct r10conf *conf = r10_bio->mddev->private; - mempool_free(r10_bio, conf->r10buf_pool); + mempool_free(r10_bio, &conf->r10buf_pool); lower_barrier(conf); } @@ -1204,7 +1204,7 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio, (unsigned long long)r10_bio->sector); if (max_sectors < bio_sectors(bio)) { struct bio *split = bio_split(bio, max_sectors, - gfp, conf->bio_split); + gfp, &conf->bio_split); bio_chain(split, bio); generic_make_request(bio); bio = split; @@ -1213,7 +1213,7 @@ static void raid10_read_request(struct mddev *mddev, struct bio *bio, } slot = r10_bio->read_slot; - read_bio = bio_clone_fast(bio, gfp, mddev->bio_set); + read_bio = bio_clone_fast(bio, gfp, &mddev->bio_set); r10_bio->devs[slot].bio = read_bio; r10_bio->devs[slot].rdev = rdev; @@ -1261,7 +1261,7 @@ static void raid10_write_one_disk(struct mddev *mddev, struct r10bio *r10_bio, } else rdev = conf->mirrors[devnum].rdev; - mbio = bio_clone_fast(bio, GFP_NOIO, mddev->bio_set); + mbio = bio_clone_fast(bio, GFP_NOIO, &mddev->bio_set); if (replacement) r10_bio->devs[n_copy].repl_bio = mbio; else @@ -1509,7 +1509,7 @@ retry_write: if (r10_bio->sectors < bio_sectors(bio)) { struct bio *split = bio_split(bio, r10_bio->sectors, - GFP_NOIO, conf->bio_split); + GFP_NOIO, &conf->bio_split); bio_chain(split, bio); generic_make_request(bio); bio = split; @@ -1533,7 +1533,7 @@ static void __make_request(struct mddev *mddev, struct bio *bio, int sectors) struct r10conf *conf = mddev->private; struct r10bio *r10_bio; - r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO); + r10_bio = mempool_alloc(&conf->r10bio_pool, GFP_NOIO); r10_bio->master_bio = bio; r10_bio->sectors = sectors; @@ -1732,8 +1732,7 @@ static void close_sync(struct r10conf *conf) wait_barrier(conf); allow_barrier(conf); - mempool_destroy(conf->r10buf_pool); - conf->r10buf_pool = NULL; + mempool_exit(&conf->r10buf_pool); } static int raid10_spare_active(struct mddev *mddev) @@ -2583,7 +2582,7 @@ static int narrow_write_error(struct r10bio *r10_bio, int i) if (sectors > sect_to_write) sectors = sect_to_write; /* Write at 'sector' for 'sectors' */ - wbio = bio_clone_fast(bio, GFP_NOIO, mddev->bio_set); + wbio = bio_clone_fast(bio, GFP_NOIO, &mddev->bio_set); bio_trim(wbio, sector - bio->bi_iter.bi_sector, sectors); wsector = r10_bio->devs[i].addr + (sector - r10_bio->sector); wbio->bi_iter.bi_sector = wsector + @@ -2816,25 +2815,25 @@ static void raid10d(struct md_thread *thread) static int init_resync(struct r10conf *conf) { - int buffs; - int i; + int ret, buffs, i; buffs = RESYNC_WINDOW / RESYNC_BLOCK_SIZE; - BUG_ON(conf->r10buf_pool); + BUG_ON(mempool_initialized(&conf->r10buf_pool)); conf->have_replacement = 0; for (i = 0; i < conf->geo.raid_disks; i++) if (conf->mirrors[i].replacement) conf->have_replacement = 1; - conf->r10buf_pool = mempool_create(buffs, r10buf_pool_alloc, r10buf_pool_free, conf); - if (!conf->r10buf_pool) - return -ENOMEM; + ret = mempool_init(&conf->r10buf_pool, buffs, + r10buf_pool_alloc, r10buf_pool_free, conf); + if (ret) + return ret; conf->next_resync = 0; return 0; } static struct r10bio *raid10_alloc_init_r10buf(struct r10conf *conf) { - struct r10bio *r10bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO); + struct r10bio *r10bio = mempool_alloc(&conf->r10buf_pool, GFP_NOIO); struct rsync_pages *rp; struct bio *bio; int nalloc; @@ -2945,7 +2944,7 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, sector_t chunk_mask = conf->geo.chunk_mask; int page_idx = 0; - if (!conf->r10buf_pool) + if (!mempool_initialized(&conf->r10buf_pool)) if (init_resync(conf)) return 0; @@ -3699,13 +3698,13 @@ static struct r10conf *setup_conf(struct mddev *mddev) conf->geo = geo; conf->copies = copies; - conf->r10bio_pool = mempool_create(NR_RAID10_BIOS, r10bio_pool_alloc, - r10bio_pool_free, conf); - if (!conf->r10bio_pool) + err = mempool_init(&conf->r10bio_pool, NR_RAID10_BIOS, r10bio_pool_alloc, + r10bio_pool_free, conf); + if (err) goto out; - conf->bio_split = bioset_create(BIO_POOL_SIZE, 0, 0); - if (!conf->bio_split) + err = bioset_init(&conf->bio_split, BIO_POOL_SIZE, 0, 0); + if (err) goto out; calc_sectors(conf, mddev->dev_sectors); @@ -3733,6 +3732,7 @@ static struct r10conf *setup_conf(struct mddev *mddev) init_waitqueue_head(&conf->wait_barrier); atomic_set(&conf->nr_pending, 0); + err = -ENOMEM; conf->thread = md_register_thread(raid10d, mddev, "raid10"); if (!conf->thread) goto out; @@ -3742,11 +3742,10 @@ static struct r10conf *setup_conf(struct mddev *mddev) out: if (conf) { - mempool_destroy(conf->r10bio_pool); + mempool_exit(&conf->r10bio_pool); kfree(conf->mirrors); safe_put_page(conf->tmppage); - if (conf->bio_split) - bioset_free(conf->bio_split); + bioset_exit(&conf->bio_split); kfree(conf); } return ERR_PTR(err); @@ -3953,7 +3952,7 @@ static int raid10_run(struct mddev *mddev) out_free_conf: md_unregister_thread(&mddev->thread); - mempool_destroy(conf->r10bio_pool); + mempool_exit(&conf->r10bio_pool); safe_put_page(conf->tmppage); kfree(conf->mirrors); kfree(conf); @@ -3966,13 +3965,12 @@ static void raid10_free(struct mddev *mddev, void *priv) { struct r10conf *conf = priv; - mempool_destroy(conf->r10bio_pool); + mempool_exit(&conf->r10bio_pool); safe_put_page(conf->tmppage); kfree(conf->mirrors); kfree(conf->mirrors_old); kfree(conf->mirrors_new); - if (conf->bio_split) - bioset_free(conf->bio_split); + bioset_exit(&conf->bio_split); kfree(conf); } @@ -4543,7 +4541,7 @@ read_more: * on all the target devices. */ // FIXME - mempool_free(r10_bio, conf->r10buf_pool); + mempool_free(r10_bio, &conf->r10buf_pool); set_bit(MD_RECOVERY_INTR, &mddev->recovery); return sectors_done; } diff --git a/drivers/md/raid10.h b/drivers/md/raid10.h index e2e8840de9bf..d3eaaf3eb1bc 100644 --- a/drivers/md/raid10.h +++ b/drivers/md/raid10.h @@ -93,10 +93,10 @@ struct r10conf { */ wait_queue_head_t wait_barrier; - mempool_t *r10bio_pool; - mempool_t *r10buf_pool; + mempool_t r10bio_pool; + mempool_t r10buf_pool; struct page *tmppage; - struct bio_set *bio_split; + struct bio_set bio_split; /* When taking over an array from a different personality, we store * the new thread here until we fully activate the array. diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c index 3c65f52b68f5..2b775abf377b 100644 --- a/drivers/md/raid5-cache.c +++ b/drivers/md/raid5-cache.c @@ -125,9 +125,9 @@ struct r5l_log { struct list_head no_mem_stripes; /* pending stripes, -ENOMEM */ struct kmem_cache *io_kc; - mempool_t *io_pool; - struct bio_set *bs; - mempool_t *meta_pool; + mempool_t io_pool; + struct bio_set bs; + mempool_t meta_pool; struct md_thread *reclaim_thread; unsigned long reclaim_target; /* number of space that need to be @@ -579,7 +579,7 @@ static void r5l_log_endio(struct bio *bio) md_error(log->rdev->mddev, log->rdev); bio_put(bio); - mempool_free(io->meta_page, log->meta_pool); + mempool_free(io->meta_page, &log->meta_pool); spin_lock_irqsave(&log->io_list_lock, flags); __r5l_set_io_unit_state(io, IO_UNIT_IO_END); @@ -748,7 +748,7 @@ static void r5l_submit_current_io(struct r5l_log *log) static struct bio *r5l_bio_alloc(struct r5l_log *log) { - struct bio *bio = bio_alloc_bioset(GFP_NOIO, BIO_MAX_PAGES, log->bs); + struct bio *bio = bio_alloc_bioset(GFP_NOIO, BIO_MAX_PAGES, &log->bs); bio_set_op_attrs(bio, REQ_OP_WRITE, 0); bio_set_dev(bio, log->rdev->bdev); @@ -780,7 +780,7 @@ static struct r5l_io_unit *r5l_new_meta(struct r5l_log *log) struct r5l_io_unit *io; struct r5l_meta_block *block; - io = mempool_alloc(log->io_pool, GFP_ATOMIC); + io = mempool_alloc(&log->io_pool, GFP_ATOMIC); if (!io) return NULL; memset(io, 0, sizeof(*io)); @@ -791,7 +791,7 @@ static struct r5l_io_unit *r5l_new_meta(struct r5l_log *log) bio_list_init(&io->flush_barriers); io->state = IO_UNIT_RUNNING; - io->meta_page = mempool_alloc(log->meta_pool, GFP_NOIO); + io->meta_page = mempool_alloc(&log->meta_pool, GFP_NOIO); block = page_address(io->meta_page); clear_page(block); block->magic = cpu_to_le32(R5LOG_MAGIC); @@ -1223,7 +1223,7 @@ static bool r5l_complete_finished_ios(struct r5l_log *log) log->next_checkpoint = io->log_start; list_del(&io->log_sibling); - mempool_free(io, log->io_pool); + mempool_free(io, &log->io_pool); r5l_run_no_mem_stripe(log); found = true; @@ -1647,7 +1647,7 @@ static int r5l_recovery_allocate_ra_pool(struct r5l_log *log, { struct page *page; - ctx->ra_bio = bio_alloc_bioset(GFP_KERNEL, BIO_MAX_PAGES, log->bs); + ctx->ra_bio = bio_alloc_bioset(GFP_KERNEL, BIO_MAX_PAGES, &log->bs); if (!ctx->ra_bio) return -ENOMEM; @@ -3066,6 +3066,7 @@ int r5l_init_log(struct r5conf *conf, struct md_rdev *rdev) struct request_queue *q = bdev_get_queue(rdev->bdev); struct r5l_log *log; char b[BDEVNAME_SIZE]; + int ret; pr_debug("md/raid:%s: using device %s as journal\n", mdname(conf->mddev), bdevname(rdev->bdev, b)); @@ -3111,16 +3112,16 @@ int r5l_init_log(struct r5conf *conf, struct md_rdev *rdev) if (!log->io_kc) goto io_kc; - log->io_pool = mempool_create_slab_pool(R5L_POOL_SIZE, log->io_kc); - if (!log->io_pool) + ret = mempool_init_slab_pool(&log->io_pool, R5L_POOL_SIZE, log->io_kc); + if (ret) goto io_pool; - log->bs = bioset_create(R5L_POOL_SIZE, 0, BIOSET_NEED_BVECS); - if (!log->bs) + ret = bioset_init(&log->bs, R5L_POOL_SIZE, 0, BIOSET_NEED_BVECS); + if (ret) goto io_bs; - log->meta_pool = mempool_create_page_pool(R5L_POOL_SIZE, 0); - if (!log->meta_pool) + ret = mempool_init_page_pool(&log->meta_pool, R5L_POOL_SIZE, 0); + if (ret) goto out_mempool; spin_lock_init(&log->tree_lock); @@ -3155,11 +3156,11 @@ int r5l_init_log(struct r5conf *conf, struct md_rdev *rdev) rcu_assign_pointer(conf->log, NULL); md_unregister_thread(&log->reclaim_thread); reclaim_thread: - mempool_destroy(log->meta_pool); + mempool_exit(&log->meta_pool); out_mempool: - bioset_free(log->bs); + bioset_exit(&log->bs); io_bs: - mempool_destroy(log->io_pool); + mempool_exit(&log->io_pool); io_pool: kmem_cache_destroy(log->io_kc); io_kc: @@ -3178,9 +3179,9 @@ void r5l_exit_log(struct r5conf *conf) wake_up(&conf->mddev->sb_wait); flush_work(&log->disable_writeback_work); md_unregister_thread(&log->reclaim_thread); - mempool_destroy(log->meta_pool); - bioset_free(log->bs); - mempool_destroy(log->io_pool); + mempool_exit(&log->meta_pool); + bioset_exit(&log->bs); + mempool_exit(&log->io_pool); kmem_cache_destroy(log->io_kc); kfree(log); } diff --git a/drivers/md/raid5-ppl.c b/drivers/md/raid5-ppl.c index 42890a08375b..3a7c36326589 100644 --- a/drivers/md/raid5-ppl.c +++ b/drivers/md/raid5-ppl.c @@ -105,9 +105,9 @@ struct ppl_conf { atomic64_t seq; /* current log write sequence number */ struct kmem_cache *io_kc; - mempool_t *io_pool; - struct bio_set *bs; - struct bio_set *flush_bs; + mempool_t io_pool; + struct bio_set bs; + struct bio_set flush_bs; /* used only for recovery */ int recovered_entries; @@ -244,7 +244,7 @@ static struct ppl_io_unit *ppl_new_iounit(struct ppl_log *log, struct ppl_header *pplhdr; struct page *header_page; - io = mempool_alloc(ppl_conf->io_pool, GFP_NOWAIT); + io = mempool_alloc(&ppl_conf->io_pool, GFP_NOWAIT); if (!io) return NULL; @@ -503,7 +503,7 @@ static void ppl_submit_iounit(struct ppl_io_unit *io) struct bio *prev = bio; bio = bio_alloc_bioset(GFP_NOIO, BIO_MAX_PAGES, - ppl_conf->bs); + &ppl_conf->bs); bio->bi_opf = prev->bi_opf; bio_copy_dev(bio, prev); bio->bi_iter.bi_sector = bio_end_sector(prev); @@ -570,7 +570,7 @@ static void ppl_io_unit_finished(struct ppl_io_unit *io) list_del(&io->log_sibling); spin_unlock(&log->io_list_lock); - mempool_free(io, ppl_conf->io_pool); + mempool_free(io, &ppl_conf->io_pool); spin_lock(&ppl_conf->no_mem_stripes_lock); if (!list_empty(&ppl_conf->no_mem_stripes)) { @@ -642,7 +642,7 @@ static void ppl_do_flush(struct ppl_io_unit *io) struct bio *bio; char b[BDEVNAME_SIZE]; - bio = bio_alloc_bioset(GFP_NOIO, 0, ppl_conf->flush_bs); + bio = bio_alloc_bioset(GFP_NOIO, 0, &ppl_conf->flush_bs); bio_set_dev(bio, bdev); bio->bi_private = io; bio->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH; @@ -1246,11 +1246,9 @@ static void __ppl_exit_log(struct ppl_conf *ppl_conf) kfree(ppl_conf->child_logs); - if (ppl_conf->bs) - bioset_free(ppl_conf->bs); - if (ppl_conf->flush_bs) - bioset_free(ppl_conf->flush_bs); - mempool_destroy(ppl_conf->io_pool); + bioset_exit(&ppl_conf->bs); + bioset_exit(&ppl_conf->flush_bs); + mempool_exit(&ppl_conf->io_pool); kmem_cache_destroy(ppl_conf->io_kc); kfree(ppl_conf); @@ -1387,24 +1385,18 @@ int ppl_init_log(struct r5conf *conf) goto err; } - ppl_conf->io_pool = mempool_create(conf->raid_disks, ppl_io_pool_alloc, - ppl_io_pool_free, ppl_conf->io_kc); - if (!ppl_conf->io_pool) { - ret = -ENOMEM; + ret = mempool_init(&ppl_conf->io_pool, conf->raid_disks, ppl_io_pool_alloc, + ppl_io_pool_free, ppl_conf->io_kc); + if (ret) goto err; - } - ppl_conf->bs = bioset_create(conf->raid_disks, 0, BIOSET_NEED_BVECS); - if (!ppl_conf->bs) { - ret = -ENOMEM; + ret = bioset_init(&ppl_conf->bs, conf->raid_disks, 0, BIOSET_NEED_BVECS); + if (ret) goto err; - } - ppl_conf->flush_bs = bioset_create(conf->raid_disks, 0, 0); - if (!ppl_conf->flush_bs) { - ret = -ENOMEM; + ret = bioset_init(&ppl_conf->flush_bs, conf->raid_disks, 0, 0); + if (ret) goto err; - } ppl_conf->count = conf->raid_disks; ppl_conf->child_logs = kcalloc(ppl_conf->count, sizeof(struct ppl_log), diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index be117d0a65a8..a2e64989b01f 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -5192,7 +5192,7 @@ static int raid5_read_one_chunk(struct mddev *mddev, struct bio *raid_bio) /* * use bio_clone_fast to make a copy of the bio */ - align_bi = bio_clone_fast(raid_bio, GFP_NOIO, mddev->bio_set); + align_bi = bio_clone_fast(raid_bio, GFP_NOIO, &mddev->bio_set); if (!align_bi) return 0; /* @@ -5277,7 +5277,7 @@ static struct bio *chunk_aligned_read(struct mddev *mddev, struct bio *raid_bio) if (sectors < bio_sectors(raid_bio)) { struct r5conf *conf = mddev->private; - split = bio_split(raid_bio, sectors, GFP_NOIO, conf->bio_split); + split = bio_split(raid_bio, sectors, GFP_NOIO, &conf->bio_split); bio_chain(split, raid_bio); generic_make_request(raid_bio); raid_bio = split; @@ -6773,8 +6773,7 @@ static void free_conf(struct r5conf *conf) if (conf->disks[i].extra_page) put_page(conf->disks[i].extra_page); kfree(conf->disks); - if (conf->bio_split) - bioset_free(conf->bio_split); + bioset_exit(&conf->bio_split); kfree(conf->stripe_hashtbl); kfree(conf->pending_data); kfree(conf); @@ -6853,6 +6852,7 @@ static struct r5conf *setup_conf(struct mddev *mddev) int i; int group_cnt, worker_cnt_per_group; struct r5worker_group *new_group; + int ret; if (mddev->new_level != 5 && mddev->new_level != 4 @@ -6950,8 +6950,8 @@ static struct r5conf *setup_conf(struct mddev *mddev) goto abort; } - conf->bio_split = bioset_create(BIO_POOL_SIZE, 0, 0); - if (!conf->bio_split) + ret = bioset_init(&conf->bio_split, BIO_POOL_SIZE, 0, 0); + if (ret) goto abort; conf->mddev = mddev; diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h index 3f8da26032ac..72e75ba6abf0 100644 --- a/drivers/md/raid5.h +++ b/drivers/md/raid5.h @@ -669,7 +669,7 @@ struct r5conf { int pool_size; /* number of disks in stripeheads in pool */ spinlock_t device_lock; struct disk_info *disks; - struct bio_set *bio_split; + struct bio_set bio_split; /* When taking over an array from a different personality, we store * the new thread here until we fully activate the array. |