diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2019-07-09 10:45:06 -0700 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2019-07-09 10:45:06 -0700 |
| commit | 3b99107f0e0298e6fe0787f75b8f3d8306dfb230 (patch) | |
| tree | 30536dbc9ca176470a2ae2938f952381e33f5deb /drivers | |
| parent | 0415052db4f92b7e272fc15802ad8b8be672deea (diff) | |
| parent | c9b3007feca018d3f7061f5d5a14cb00766ffe9b (diff) | |
Merge tag 'for-5.3/block-20190708' of git://git.kernel.dk/linux-block
Pull block updates from Jens Axboe:
"This is the main block updates for 5.3. Nothing earth shattering or
major in here, just fixes, additions, and improvements all over the
map. This contains:
- Series of documentation fixes (Bart)
- Optimization of the blk-mq ctx get/put (Bart)
- null_blk removal race condition fix (Bob)
- req/bio_op() cleanups (Chaitanya)
- Series cleaning up the segment accounting, and request/bio mapping
(Christoph)
- Series cleaning up the page getting/putting for bios (Christoph)
- block cgroup cleanups and moving it to where it is used (Christoph)
- block cgroup fixes (Tejun)
- Series of fixes and improvements to bcache, most notably a write
deadlock fix (Coly)
- blk-iolatency STS_AGAIN and accounting fixes (Dennis)
- Series of improvements and fixes to BFQ (Douglas, Paolo)
- debugfs_create() return value check removal for drbd (Greg)
- Use struct_size(), where appropriate (Gustavo)
- Two lighnvm fixes (Heiner, Geert)
- MD fixes, including a read balance and corruption fix (Guoqing,
Marcos, Xiao, Yufen)
- block opal shadow mbr additions (Jonas, Revanth)
- sbitmap compare-and-exhange improvemnts (Pavel)
- Fix for potential bio->bi_size overflow (Ming)
- NVMe pull requests:
- improved PCIe suspent support (Keith Busch)
- error injection support for the admin queue (Akinobu Mita)
- Fibre Channel discovery improvements (James Smart)
- tracing improvements including nvmetc tracing support (Minwoo Im)
- misc fixes and cleanups (Anton Eidelman, Minwoo Im, Chaitanya
Kulkarni)"
- Various little fixes and improvements to drivers and core"
* tag 'for-5.3/block-20190708' of git://git.kernel.dk/linux-block: (153 commits)
blk-iolatency: fix STS_AGAIN handling
block: nr_phys_segments needs to be zero for REQ_OP_WRITE_ZEROES
blk-mq: simplify blk_mq_make_request()
blk-mq: remove blk_mq_put_ctx()
sbitmap: Replace cmpxchg with xchg
block: fix .bi_size overflow
block: sed-opal: check size of shadow mbr
block: sed-opal: ioctl for writing to shadow mbr
block: sed-opal: add ioctl for done-mark of shadow mbr
block: never take page references for ITER_BVEC
direct-io: use bio_release_pages in dio_bio_complete
block_dev: use bio_release_pages in bio_unmap_user
block_dev: use bio_release_pages in blkdev_bio_end_io
iomap: use bio_release_pages in iomap_dio_bio_end_io
block: use bio_release_pages in bio_map_user_iov
block: use bio_release_pages in bio_unmap_user
block: optionally mark pages dirty in bio_release_pages
block: move the BIO_NO_PAGE_REF check into bio_release_pages
block: skd_main.c: Remove call to memset after dma_alloc_coherent
block: mtip32xx: Remove call to memset after dma_alloc_coherent
...
Diffstat (limited to 'drivers')
56 files changed, 1753 insertions, 505 deletions
diff --git a/drivers/block/drbd/drbd_debugfs.c b/drivers/block/drbd/drbd_debugfs.c index f13b48ff5f43..b3b9cd5628fd 100644 --- a/drivers/block/drbd/drbd_debugfs.c +++ b/drivers/block/drbd/drbd_debugfs.c @@ -465,35 +465,20 @@ static const struct file_operations in_flight_summary_fops = { void drbd_debugfs_resource_add(struct drbd_resource *resource) { struct dentry *dentry; - if (!drbd_debugfs_resources) - return; dentry = debugfs_create_dir(resource->name, drbd_debugfs_resources); - if (IS_ERR_OR_NULL(dentry)) - goto fail; resource->debugfs_res = dentry; dentry = debugfs_create_dir("volumes", resource->debugfs_res); - if (IS_ERR_OR_NULL(dentry)) - goto fail; resource->debugfs_res_volumes = dentry; dentry = debugfs_create_dir("connections", resource->debugfs_res); - if (IS_ERR_OR_NULL(dentry)) - goto fail; resource->debugfs_res_connections = dentry; dentry = debugfs_create_file("in_flight_summary", 0440, resource->debugfs_res, resource, &in_flight_summary_fops); - if (IS_ERR_OR_NULL(dentry)) - goto fail; resource->debugfs_res_in_flight_summary = dentry; - return; - -fail: - drbd_debugfs_resource_cleanup(resource); - drbd_err(resource, "failed to create debugfs dentry\n"); } static void drbd_debugfs_remove(struct dentry **dp) @@ -636,35 +621,22 @@ void drbd_debugfs_connection_add(struct drbd_connection *connection) { struct dentry *conns_dir = connection->resource->debugfs_res_connections; struct dentry *dentry; - if (!conns_dir) - return; /* Once we enable mutliple peers, * these connections will have descriptive names. * For now, it is just the one connection to the (only) "peer". */ dentry = debugfs_create_dir("peer", conns_dir); - if (IS_ERR_OR_NULL(dentry)) - goto fail; connection->debugfs_conn = dentry; dentry = debugfs_create_file("callback_history", 0440, connection->debugfs_conn, connection, &connection_callback_history_fops); - if (IS_ERR_OR_NULL(dentry)) - goto fail; connection->debugfs_conn_callback_history = dentry; dentry = debugfs_create_file("oldest_requests", 0440, connection->debugfs_conn, connection, &connection_oldest_requests_fops); - if (IS_ERR_OR_NULL(dentry)) - goto fail; connection->debugfs_conn_oldest_requests = dentry; - return; - -fail: - drbd_debugfs_connection_cleanup(connection); - drbd_err(connection, "failed to create debugfs dentry\n"); } void drbd_debugfs_connection_cleanup(struct drbd_connection *connection) @@ -809,8 +781,6 @@ void drbd_debugfs_device_add(struct drbd_device *device) snprintf(vnr_buf, sizeof(vnr_buf), "%u", device->vnr); dentry = debugfs_create_dir(vnr_buf, vols_dir); - if (IS_ERR_OR_NULL(dentry)) - goto fail; device->debugfs_vol = dentry; snprintf(minor_buf, sizeof(minor_buf), "%u", device->minor); @@ -819,18 +789,14 @@ void drbd_debugfs_device_add(struct drbd_device *device) if (!slink_name) goto fail; dentry = debugfs_create_symlink(minor_buf, drbd_debugfs_minors, slink_name); + device->debugfs_minor = dentry; kfree(slink_name); slink_name = NULL; - if (IS_ERR_OR_NULL(dentry)) - goto fail; - device->debugfs_minor = dentry; #define DCF(name) do { \ dentry = debugfs_create_file(#name, 0440, \ device->debugfs_vol, device, \ &device_ ## name ## _fops); \ - if (IS_ERR_OR_NULL(dentry)) \ - goto fail; \ device->debugfs_vol_ ## name = dentry; \ } while (0) @@ -864,19 +830,9 @@ void drbd_debugfs_peer_device_add(struct drbd_peer_device *peer_device) struct dentry *dentry; char vnr_buf[8]; - if (!conn_dir) - return; - snprintf(vnr_buf, sizeof(vnr_buf), "%u", peer_device->device->vnr); dentry = debugfs_create_dir(vnr_buf, conn_dir); - if (IS_ERR_OR_NULL(dentry)) - goto fail; peer_device->debugfs_peer_dev = dentry; - return; - -fail: - drbd_debugfs_peer_device_cleanup(peer_device); - drbd_err(peer_device, "failed to create debugfs entries\n"); } void drbd_debugfs_peer_device_cleanup(struct drbd_peer_device *peer_device) @@ -917,35 +873,19 @@ void drbd_debugfs_cleanup(void) drbd_debugfs_remove(&drbd_debugfs_root); } -int __init drbd_debugfs_init(void) +void __init drbd_debugfs_init(void) { struct dentry *dentry; dentry = debugfs_create_dir("drbd", NULL); - if (IS_ERR_OR_NULL(dentry)) - goto fail; drbd_debugfs_root = dentry; dentry = debugfs_create_file("version", 0444, drbd_debugfs_root, NULL, &drbd_version_fops); - if (IS_ERR_OR_NULL(dentry)) - goto fail; drbd_debugfs_version = dentry; dentry = debugfs_create_dir("resources", drbd_debugfs_root); - if (IS_ERR_OR_NULL(dentry)) - goto fail; drbd_debugfs_resources = dentry; dentry = debugfs_create_dir("minors", drbd_debugfs_root); - if (IS_ERR_OR_NULL(dentry)) - goto fail; drbd_debugfs_minors = dentry; - return 0; - -fail: - drbd_debugfs_cleanup(); - if (dentry) - return PTR_ERR(dentry); - else - return -EINVAL; } diff --git a/drivers/block/drbd/drbd_debugfs.h b/drivers/block/drbd/drbd_debugfs.h index 4ecfbb3358d7..58e31cef0844 100644 --- a/drivers/block/drbd/drbd_debugfs.h +++ b/drivers/block/drbd/drbd_debugfs.h @@ -6,7 +6,7 @@ #include "drbd_int.h" #ifdef CONFIG_DEBUG_FS -int __init drbd_debugfs_init(void); +void __init drbd_debugfs_init(void); void drbd_debugfs_cleanup(void); void drbd_debugfs_resource_add(struct drbd_resource *resource); @@ -22,7 +22,7 @@ void drbd_debugfs_peer_device_add(struct drbd_peer_device *peer_device); void drbd_debugfs_peer_device_cleanup(struct drbd_peer_device *peer_device); #else -static inline int __init drbd_debugfs_init(void) { return -ENODEV; } +static inline void __init drbd_debugfs_init(void) { } static inline void drbd_debugfs_cleanup(void) { } static inline void drbd_debugfs_resource_add(struct drbd_resource *resource) { } diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c index 9604127dc839..9bd4ddd12b25 100644 --- a/drivers/block/drbd/drbd_main.c +++ b/drivers/block/drbd/drbd_main.c @@ -3009,8 +3009,7 @@ static int __init drbd_init(void) spin_lock_init(&retry.lock); INIT_LIST_HEAD(&retry.writes); - if (drbd_debugfs_init()) - pr_notice("failed to initialize debugfs -- will not be available\n"); + drbd_debugfs_init(); pr_info("initialized. " "Version: " REL_VERSION " (api:%d/proto:%d-%d)\n", diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 9fb9b312ab6b..b933a7eea52b 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -3900,7 +3900,7 @@ static void __init config_types(void) if (!UDP->cmos) UDP->cmos = FLOPPY0_TYPE; drive = 1; - if (!UDP->cmos && FLOPPY1_TYPE) + if (!UDP->cmos) UDP->cmos = FLOPPY1_TYPE; /* FIXME: additional physical CMOS drive detection should go here */ diff --git a/drivers/block/loop.c b/drivers/block/loop.c index f11b7dc16e9d..44c9985f352a 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -264,20 +264,12 @@ lo_do_transfer(struct loop_device *lo, int cmd, return ret; } -static inline void loop_iov_iter_bvec(struct iov_iter *i, - unsigned int direction, const struct bio_vec *bvec, - unsigned long nr_segs, size_t count) -{ - iov_iter_bvec(i, direction, bvec, nr_segs, count); - i->type |= ITER_BVEC_FLAG_NO_REF; -} - static int lo_write_bvec(struct file *file, struct bio_vec *bvec, loff_t *ppos) { struct iov_iter i; ssize_t bw; - loop_iov_iter_bvec(&i, WRITE, bvec, 1, bvec->bv_len); + iov_iter_bvec(&i, WRITE, bvec, 1, bvec->bv_len); file_start_write(file); bw = vfs_iter_write(file, &i, ppos, 0); @@ -355,7 +347,7 @@ static int lo_read_simple(struct loop_device *lo, struct request *rq, ssize_t len; rq_for_each_segment(bvec, rq, iter) { - loop_iov_iter_bvec(&i, READ, &bvec, 1, bvec.bv_len); + iov_iter_bvec(&i, READ, &bvec, 1, bvec.bv_len); len = vfs_iter_read(lo->lo_backing_file, &i, &pos, 0); if (len < 0) return len; @@ -396,7 +388,7 @@ static int lo_read_transfer(struct loop_device *lo, struct request *rq, b.bv_offset = 0; b.bv_len = bvec.bv_len; - loop_iov_iter_bvec(&i, READ, &b, 1, b.bv_len); + iov_iter_bvec(&i, READ, &b, 1, b.bv_len); len = vfs_iter_read(lo->lo_backing_file, &i, &pos, 0); if (len < 0) { ret = len; @@ -563,7 +555,7 @@ static int lo_rw_aio(struct loop_device *lo, struct loop_cmd *cmd, } atomic_set(&cmd->ref, 2); - loop_iov_iter_bvec(&iter, rw, bvec, nr_bvec, blk_rq_bytes(rq)); + iov_iter_bvec(&iter, rw, bvec, nr_bvec, blk_rq_bytes(rq)); iter.iov_offset = offset; cmd->iocb.ki_pos = pos; diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c index a14b09ab3a41..964f78cfffa0 100644 --- a/drivers/block/mtip32xx/mtip32xx.c +++ b/drivers/block/mtip32xx/mtip32xx.c @@ -1577,7 +1577,6 @@ static int exec_drive_command(struct mtip_port *port, u8 *command, ATA_SECT_SIZE * xfer_sz); return -ENOMEM; } - memset(buf, 0, ATA_SECT_SIZE * xfer_sz); } /* Build the FIS. */ @@ -2776,7 +2775,6 @@ static int mtip_dma_alloc(struct driver_data *dd) &port->block1_dma, GFP_KERNEL); if (!port->block1) return -ENOMEM; - memset(port->block1, 0, BLOCK_DMA_ALLOC_SZ); /* Allocate dma memory for command list */ port->command_list = @@ -2789,7 +2787,6 @@ static int mtip_dma_alloc(struct driver_data *dd) port->block1_dma = 0; return -ENOMEM; } - memset(port->command_list, 0, AHCI_CMD_TBL_SZ); /* Setup all pointers into first DMA region */ port->rxfis = port->block1 + AHCI_RX_FIS_OFFSET; @@ -3529,8 +3526,6 @@ static int mtip_init_cmd(struct blk_mq_tag_set *set, struct request *rq, if (!cmd->command) return -ENOMEM; - memset(cmd->command, 0, CMD_DMA_ALLOC_SZ); - sg_init_table(cmd->sg, MTIP_MAX_SG); return 0; } diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c index 447d635c79a2..99328ded60d1 100644 --- a/drivers/block/null_blk_main.c +++ b/drivers/block/null_blk_main.c @@ -327,11 +327,12 @@ static ssize_t nullb_device_power_store(struct config_item *item, set_bit(NULLB_DEV_FL_CONFIGURED, &dev->flags); dev->power = newp; } else if (dev->power && !newp) { - mutex_lock(&lock); - dev->power = newp; - null_del_dev(dev->nullb); - mutex_unlock(&lock); - clear_bit(NULLB_DEV_FL_UP, &dev->flags); + if (test_and_clear_bit(NULLB_DEV_FL_UP, &dev->flags)) { + mutex_lock(&lock); + dev->power = newp; + null_del_dev(dev->nullb); + mutex_unlock(&lock); + } clear_bit(NULLB_DEV_FL_CONFIGURED, &dev->flags); } @@ -1197,7 +1198,7 @@ static blk_status_t null_handle_cmd(struct nullb_cmd *cmd) if (!cmd->error && dev->zoned) { sector_t sector; unsigned int nr_sectors; - int op; + enum req_opf op; if (dev->queue_mode == NULL_Q_BIO) { op = bio_op(cmd->bio); @@ -1488,7 +1489,6 @@ static int setup_queues(struct nullb *nullb) if (!nullb->queues) return -ENOMEM; - nullb->nr_queues = 0; nullb->queue_depth = nullb->dev->hw_queue_depth; return 0; diff --git a/drivers/block/skd_main.c b/drivers/block/skd_main.c index c479235862e5..51569c199a6c 100644 --- a/drivers/block/skd_main.c +++ b/drivers/block/skd_main.c @@ -2694,7 +2694,6 @@ static int skd_cons_skmsg(struct skd_device *skdev) (FIT_QCMD_ALIGN - 1), "not aligned: msg_buf %p mb_dma_address %pad\n", skmsg->msg_buf, &skmsg->mb_dma_address); - memset(skmsg->msg_buf, 0, SKD_N_FITMSG_BYTES); } err_out: diff --git a/drivers/lightnvm/core.c b/drivers/lightnvm/core.c index 7d555b110ecd..a600934fdd9c 100644 --- a/drivers/lightnvm/core.c +++ b/drivers/lightnvm/core.c @@ -478,7 +478,7 @@ static void __nvm_remove_target(struct nvm_target *t, bool graceful) */ static int nvm_remove_tgt(struct nvm_ioctl_remove *remove) { - struct nvm_target *t; + struct nvm_target *t = NULL; struct nvm_dev *dev; down_read(&nvm_lock); diff --git a/drivers/lightnvm/pblk-core.c b/drivers/lightnvm/pblk-core.c index 773537804319..f546e6f28b8a 100644 --- a/drivers/lightnvm/pblk-core.c +++ b/drivers/lightnvm/pblk-core.c @@ -323,14 +323,16 @@ void pblk_free_rqd(struct pblk *pblk, struct nvm_rq *rqd, int type) void pblk_bio_free_pages(struct pblk *pblk, struct bio *bio, int off, int nr_pages) { - struct bio_vec bv; - int i; - - WARN_ON(off + nr_pages != bio->bi_vcnt); - - for (i = off; i < nr_pages + off; i++) { - bv = bio->bi_io_vec[i]; - mempool_free(bv.bv_page, &pblk->page_bio_pool); + struct bio_vec *bv; + struct page *page; + int i, e, nbv = 0; + + for (i = 0; i < bio->bi_vcnt; i++) { + bv = &bio->bi_io_vec[i]; + page = bv->bv_page; + for (e = 0; e < bv->bv_len; e += PBLK_EXPOSED_PAGE_SIZE, nbv++) + if (nbv >= off) + mempool_free(page++, &pblk->page_bio_pool); } } diff --git a/drivers/md/bcache/alloc.c b/drivers/md/bcache/alloc.c index f8986effcb50..6f776823b9ba 100644 --- a/drivers/md/bcache/alloc.c +++ b/drivers/md/bcache/alloc.c @@ -393,6 +393,11 @@ long bch_bucket_alloc(struct cache *ca, unsigned int reserve, bool wait) struct bucket *b; long r; + + /* No allocation if CACHE_SET_IO_DISABLE bit is set */ + if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &ca->set->flags))) + return -1; + /* fastpath */ if (fifo_pop(&ca->free[RESERVE_NONE], r) || fifo_pop(&ca->free[reserve], r)) @@ -484,6 +489,10 @@ int __bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve, { int i; + /* No allocation if CACHE_SET_IO_DISABLE bit is set */ + if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &c->flags))) + return -1; + lockdep_assert_held(&c->bucket_lock); BUG_ON(!n || n > c->caches_loaded || n > MAX_CACHES_PER_SET); diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h index fdf75352e16a..013e35a9e317 100644 --- a/drivers/md/bcache/bcache.h +++ b/drivers/md/bcache/bcache.h @@ -705,8 +705,8 @@ struct cache_set { atomic_long_t writeback_keys_failed; atomic_long_t reclaim; + atomic_long_t reclaimed_journal_buckets; atomic_long_t flush_write; - atomic_long_t retry_flush_write; enum { ON_ERROR_UNREGISTER, @@ -726,8 +726,6 @@ struct cache_set { #define BUCKET_HASH_BITS 12 struct hlist_head bucket_hash[1 << BUCKET_HASH_BITS]; - - DECLARE_HEAP(struct btree *, flush_btree); }; struct bbio { @@ -1006,7 +1004,7 @@ int bch_flash_dev_create(struct cache_set *c, uint64_t size); int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, uint8_t *set_uuid); void bch_cached_dev_detach(struct cached_dev *dc); -void bch_cached_dev_run(struct cached_dev *dc); +int bch_cached_dev_run(struct cached_dev *dc); void bcache_device_stop(struct bcache_device *d); void bch_cache_set_unregister(struct cache_set *c); diff --git a/drivers/md/bcache/bset.c b/drivers/md/bcache/bset.c index 268f1b685084..08768796b543 100644 --- a/drivers/md/bcache/bset.c +++ b/drivers/md/bcache/bset.c @@ -347,22 +347,19 @@ EXPORT_SYMBOL(bch_btree_keys_alloc); void bch_btree_keys_init(struct btree_keys *b, const struct btree_keys_ops *ops, bool *expensive_debug_checks) { - unsigned int i; - b->ops = ops; b->expensive_debug_checks = expensive_debug_checks; b->nsets = 0; b->last_set_unwritten = 0; - /* XXX: shouldn't be needed */ - for (i = 0; i < MAX_BSETS; i++) - b->set[i].size = 0; /* - * Second loop starts at 1 because b->keys[0]->data is the memory we - * allocated + * struct btree_keys in embedded in struct btree, and struct + * bset_tree is embedded into struct btree_keys. They are all + * initialized as 0 by kzalloc() in mca_bucket_alloc(), and + * b->set[0].data is allocated in bch_btree_keys_alloc(), so we + * don't have to initiate b->set[].size and b->set[].data here + * any more. */ - for (i = 1; i < MAX_BSETS; i++) - b->set[i].data = NULL; } EXPORT_SYMBOL(bch_btree_keys_init); @@ -970,45 +967,25 @@ static struct bset_search_iter bset_search_tree(struct bset_tree *t, unsigned int inorder, j, n = 1; do { - /* - * A bit trick here. - * If p < t->size, (int)(p - t->size) is a minus value and - * the most significant bit is set, right shifting 31 bits - * gets 1. If p >= t->size, the most significant bit is - * not set, right shifting 31 bits gets 0. - * So the following 2 lines equals to - * if (p >= t->size) - * p = 0; - * but a branch instruction is avoided. - */ unsigned int p = n << 4; - p &= ((int) (p - t->size)) >> 31; - - prefetch(&t->tree[p]); + if (p < t->size) + prefetch(&t->tree[p]); j = n; f = &t->tree[j]; - /* - * Similar bit trick, use subtract operation to avoid a branch - * instruction. - * - * n = (f->mantissa > bfloat_mantissa()) - * ? j * 2 - * : j * 2 + 1; - * - * We need to subtract 1 from f->mantissa for the sign bit trick - * to work - that's done in make_bfloat() - */ - if (likely(f->exponent != 127)) - n = j * 2 + (((unsigned int) - (f->mantissa - - bfloat_mantissa(search, f))) >> 31); - else - n = (bkey_cmp(tree_to_bkey(t, j), search) > 0) - ? j * 2 - : j * 2 + 1; + if (likely(f->exponent != 127)) { + if (f->mantissa >= bfloat_mantissa(search, f)) + n = j * 2; + else + n = j * 2 + 1; + } else { + if (bkey_cmp(tree_to_bkey(t, j), search) > 0) + n = j * 2; + else + n = j * 2 + 1; + } } while (n < t->size); inorder = to_inorder(j, t); diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c index 773f5fdad25f..ba434d9ac720 100644 --- a/drivers/md/bcache/btree.c +++ b/drivers/md/bcache/btree.c @@ -35,7 +35,7 @@ #include <linux/rcupdate.h> #include <linux/sched/clock.h> #include <linux/rculist.h> - +#include <linux/delay.h> #include <trace/events/bcache.h> /* @@ -613,6 +613,10 @@ static void mca_data_alloc(struct btree *b, struct bkey *k, gfp_t gfp) static struct btree *mca_bucket_alloc(struct cache_set *c, struct bkey *k, gfp_t gfp) { + /* + * kzalloc() is necessary here for initialization, + * see code comments in bch_btree_keys_init(). + */ struct btree *b = kzalloc(sizeof(struct btree), gfp); if (!b) @@ -655,7 +659,25 @@ static int mca_reap(struct btree *b, unsigned int min_order, bool flush) up(&b->io_mutex); } +retry: + /* + * BTREE_NODE_dirty might be cleared in btree_flush_btree() by + * __bch_btree_node_write(). To avoid an extra flush, acquire + * b->write_lock before checking BTREE_NODE_dirty bit. + */ mutex_lock(&b->write_lock); + /* + * If this btree node is selected in btree_flush_write() by journal + * code, delay and retry until the node is flushed by journal code + * and BTREE_NODE_journal_flush bit cleared by btree_flush_write(). + */ + if (btree_node_journal_flush(b)) { + pr_debug("bnode %p is flushing by journal, retry", b); + mutex_unlock(&b->write_lock); + udelay(1); + goto retry; + } + if (btree_node_dirty(b)) __bch_btree_node_write(b, &cl); mutex_unlock(&b->write_lock); @@ -778,10 +800,15 @@ void bch_btree_cache_free(struct cache_set *c) while (!list_empty(&c->btree_cache)) { b = list_first_entry(&c->btree_cache, struct btree, list); - if (btree_node_dirty(b)) + /* + * This function is called by cache_set_free(), no I/O + * request on cache now, it is unnecessary to acquire + * b->write_lock before clearing BTREE_NODE_dirty anymore. + */ + if (btree_node_dirty(b)) { btree_complete_write(b, btree_current_write(b)); - clear_bit(BTREE_NODE_dirty, &b->flags); - + clear_bit(BTREE_NODE_dirty, &b->flags); + } mca_data_free(b); } @@ -1067,11 +1094,25 @@ static void btree_node_free(struct btree *b) BUG_ON(b == b->c->root); +retry: mutex_lock(&b->write_lock); + /* + * If the btree node is selected and flushing in btree_flush_write(), + * delay and retry until the BTREE_NODE_journal_flush bit cleared, + * then it is safe to free the btree node here. Otherwise this btree + * node will be in race condition. + */ + if (btree_node_journal_flush(b)) { + mutex_unlock(&b->write_lock); + pr_debug("bnode %p journal_flush set, retry", b); + udelay(1); + goto retry; + } - if (btree_node_dirty(b)) + if (btree_node_dirty(b)) { btree_complete_write(b, btree_current_write(b)); - clear_bit(BTREE_NODE_dirty, &b->flags); + clear_bit(BTREE_NODE_dirty, &b->flags); + } mutex_unlock(&b->write_lock); diff --git a/drivers/md/bcache/btree.h b/drivers/md/bcache/btree.h index d1c72ef64edf..76cfd121a486 100644 --- a/drivers/md/bcache/btree.h +++ b/drivers/md/bcache/btree.h @@ -158,11 +158,13 @@ enum btree_flags { BTREE_NODE_io_error, BTREE_NODE_dirty, BTREE_NODE_write_idx, + BTREE_NODE_journal_flush, }; BTREE_FLAG(io_error); BTREE_FLAG(dirty); BTREE_FLAG(write_idx); +BTREE_FLAG(journal_flush); static inline struct btree_write *btree_current_write(struct btree *b) { diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c index c25097968319..4d93f07f63e5 100644 --- a/drivers/md/bcache/io.c +++ b/drivers/md/bcache/io.c @@ -58,6 +58,18 @@ void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio) WARN_ONCE(!dc, "NULL pointer of struct cached_dev"); + /* + * Read-ahead requests on a degrading and recovering md raid + * (e.g. raid6) device might be failured immediately by md + * raid code, which is not a real hardware media failure. So + * we shouldn't count failed REQ_RAHEAD bio to dc->io_errors. + */ + if (bio->bi_opf & REQ_RAHEAD) { + pr_warn_ratelimited("%s: Read-ahead I/O failed on backing device, ignore", + dc->backing_dev_name); + return; + } + errors = atomic_add_return(1, &dc->io_errors); if (errors < dc->error_limit) pr_err("%s: IO error on backing device, unrecoverable", diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c index 12dae9348147..be2a2a201603 100644 --- a/drivers/md/bcache/journal.c +++ b/drivers/md/bcache/journal.c @@ -100,6 +100,20 @@ reread: left = ca->sb.bucket_size - offset; blocks = set_blocks(j, block_bytes(ca->set)); + /* + * Nodes in 'list' are in linear increasing order of + * i->j.seq, the node on head has the smallest (oldest) + * journal seq, the node on tail has the biggest + * (latest) journal seq. + */ + + /* + * Check from the oldest jset for last_seq. If + * i->j.seq < j->last_seq, it means the oldest jset + * in list is expired and useless, remove it from + * this list. Otherwise, j is a condidate jset for + * further following checks. + */ while (!list_empty(list)) { i = list_first_entry(list, struct journal_replay, list); @@ -109,13 +123,22 @@ reread: left = ca->sb.bucket_size - offset; kfree(i); } + /* iterate list in reverse order (from latest jset) */ list_for_each_entry_reverse(i, list, list) { if (j->seq == i->j.seq) goto next_set; + /* + * if j->seq is less than any i->j.last_seq + * in list, j is an expired and useless jset. + */ if (j->seq < i->j.last_seq) goto next_set; + /* + * 'where' points to first jset in list which + * is elder then j. + */ if (j->seq > i->j.seq) { where = &i->list; goto add; @@ -129,10 +152,12 @@ add: if (!i) return -ENOMEM; memcpy(&i->j, j, bytes); + /* Add to the location after 'where' points to */ list_add(&i->list, where); ret = 1; - ja->seq[bucket_index] = j->seq; + if (j->seq > ja->seq[bucket_index]) + ja->seq[bucket_index] = j->seq; next_set: offset += blocks * ca->sb.block_size; len -= blocks * ca->sb.block_size; @@ -268,7 +293,7 @@ bsearch: struct journal_replay, list)->j.seq; - return ret; + return 0; #undef read_bucket } @@ -391,60 +416,90 @@ err: } /* Journalling */ -#define journal_max_cmp(l, r) \ - (fifo_idx(&c->journal.pin, btree_current_write(l)->journal) < \ - fifo_idx(&(c)->journal.pin, btree_current_write(r)->journal)) -#define journal_min_cmp(l, r) \ - (fifo_idx(&c->journal.pin, btree_current_write(l)->journal) > \ - fifo_idx(&(c)->journal.pin, btree_current_write(r)->journal)) static void btree_flush_write(struct cache_set *c) { - /* - * Try to find the btree node with that references the oldest journal - * entry, best is our current candidate and is locked if non NULL: - */ - struct btree *b; - int i; + struct btree *b, *t, *btree_nodes[BTREE_FLUSH_NR]; + unsigned int i, n; + + if (c->journal.btree_flushing) + return; + + spin_lock(&c->journal.flush_write_lock); + if (c->journal.btree_flushing) { + spin_unlock(&c->journal.flush_write_lock); + return; + } + c->journal.btree_flushing = true; + spin_unlock(&c->journal.flush_write_lock); atomic_long_inc(&c->flush_write); + memset(btree_nodes, 0, sizeof(btree_nodes)); + n = 0; -retry: - spin_lock(&c->journal.lock); - if (heap_empty(&c->flush_btree)) { - for_each_cached_btree(b, c, i) - if (btree_current_write(b)->journal) { - if (!heap_full(&c->flush_btree)) - heap_add(&c->flush_btree, b, - journal_max_cmp); - else if (journal_max_cmp(b, - heap_peek(&c->flush_btree))) { - c->flush_btree.data[0] = b; - heap_sift(&c->flush_btree, 0, - journal_max_cmp); - } - } + mutex_lock(&c->bucket_lock); + list_for_each_entry_safe_reverse(b, t, &c->btree_cache, list) { + if (btree_node_journal_flush(b)) + pr_err("BUG: flush_write bit should not be set here!"); + + mutex_lock(&b->write_lock); - for (i = c->flush_btree.used / 2 - 1; i >= 0; --i) - heap_sift(&c->flush_btree, i, journal_min_cmp); + if (!btree_node_dirty(b)) { + mutex_unlock(&b->write_lock); + continue; + } + + if (!btree_current_write(b)->journal) { + mutex_unlock(&b->write_lock); + continue; + } + + set_btree_node_journal_flush(b); + + mutex_unlock(&b->write_lock); + + btree_nodes[n++] = b; + if (n == BTREE_FLUSH_NR) + break; } + mutex_unlock(&c->bucket_lock); - b = NULL; - heap_pop(&c->flush_btree, b, journal_min_cmp); - spin_unlock(&c->journal.lock); + for (i = 0; i < n; i++) { + b = btree_nodes[i]; + if (!b) { + pr_err("BUG: btree_nodes[%d] is NULL", i); + continue; + } + + /* safe to check without holding b->write_lock */ + if (!btree_node_journal_flush(b)) { + pr_err("BUG: bnode %p: journal_flush bit cleaned", b); + continue; + } - if (b) { mutex_lock(&b->write_lock); if (!btree_current_write(b)->journal) { + clear_bit(BTREE_NODE_journal_flush, &b->flags); + mutex_unlock(&b->write_lock); + pr_debug("bnode %p: written by others", b); + continue; + } + + if (!btree_node_dirty(b)) { + clear_bit(BTREE_NODE_journal_flush, &b->flags); mutex_unlock(&b->write_lock); - /* We raced */ - atomic_long_inc(&c->retry_flush_write); - goto retry; + pr_debug("bnode %p: dirty bit cleaned by others", b); + continue; } __bch_btree_node_write(b, NULL); + clear_bit(BTREE_NODE_journal_flush, &b->flags); mutex_unlock(&b->write_lock); } + + spin_lock(&c->journal.flush_write_lock); + c->journal.btree_flushing = false; + spin_unlock(&c->journal.flush_write_lock); } #define last_seq(j) ((j)->seq - fifo_used(&(j)->pin) + 1) @@ -559,6 +614,7 @@ static void journal_reclaim(struct cache_set *c) k->ptr[n++] = MAKE_PTR(0, bucket_to_sector(c, ca->sb.d[ja->cur_idx]), ca->sb.nr_this_dev); + atomic_long_inc(&c->reclaimed_journal_buckets); } if (n) { @@ -811,6 +867,10 @@ atomic_t *bch_journal(struct cache_set *c, struct journal_write *w; atomic_t *ret; + /* No journaling if CACHE_SET_IO_DISABLE set already */ + if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &c->flags))) + return NULL; + if (!CACHE_SYNC(&c->sb)) return NULL; @@ -855,7 +915,6 @@ void bch_journal_free(struct cache_set *c) free_pages((unsigned long) c->journal.w[1].data, JSET_BITS); free_pages((unsigned long) c->journal.w[0].data, JSET_BITS); free_fifo(&c->journal.pin); - free_heap(&c->flush_btree); } int bch_journal_alloc(struct cache_set *c) @@ -863,6 +922,7 @@ int bch_journal_alloc(struct cache_set *c) struct journal *j = &c->journal; spin_lock_init(&j->lock); + spin_lock_init(&j->flush_write_lock); INIT_DELAYED_WORK(&j->work, journal_write_work); c->journal_delay_ms = 100; @@ -870,8 +930,7 @@ int bch_journal_alloc(struct cache_set *c) j->w[0].c = c; j->w[1].c = c; - if (!(init_heap(&c->flush_btree, 128, GFP_KERNEL)) || - !(init_fifo(&j->pin, JOURNAL_PIN, GFP_KERNEL)) || + if (!(init_fifo(&j->pin, JOURNAL_PIN, GFP_KERNEL)) || !(j->w[0].data = (void *) __get_free_pages(GFP_KERNEL, JSET_BITS)) || !(j->w[1].data = (void *) __get_free_pages(GFP_KERNEL, JSET_BITS))) return -ENOMEM; diff --git a/drivers/md/bcache/journal.h b/drivers/md/bcache/journal.h index 66f0facff84b..f2ea34d5f431 100644 --- a/drivers/md/bcache/journal.h +++ b/drivers/md/bcache/journal.h @@ -103,6 +103,8 @@ struct journal_write { /* Embedded in struct cache_set */ struct journal { spinlock_t lock; + spinlock_t flush_write_lock; + bool btree_flushing; /* used when waiting because the journal was full */ struct closure_waitlist wait; struct closure io; @@ -154,6 +156,8 @@ struct journal_device { struct bio_vec bv[8]; }; +#define BTREE_FLUSH_NR 8 + #define journal_pin_cmp(c, l, r) \ (fifo_idx(&(c)->journal.pin, (l)) > fifo_idx(&(c)->journal.pin, (r))) diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index 1b63ac876169..26e374fbf57c 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -40,6 +40,7 @@ static const char invalid_uuid[] = { static struct kobject *bcache_kobj; struct mutex bch_register_lock; +bool bcache_is_reboot; LIST_HEAD(bch_cache_sets); static LIST_HEAD(uncached_devices); @@ -49,6 +50,7 @@ static wait_queue_head_t unregister_wait; struct workqueue_struct *bcache_wq; struct workqueue_struct *bch_journal_wq; + #define BTREE_MAX_PAGES (256 * 1024 / PAGE_SIZE) /* limitation of partitions number on single bcache device */ #define BCACHE_MINORS 128 @@ -197,7 +199,9 @@ err: static void write_bdev_super_endio(struct bio *bio) { struct cached_dev *dc = bio->bi_private; - /* XXX: error checking */ + + if (bio->bi_status) + bch_count_backing_io_errors(dc, bio); closure_put(&dc->sb_write); } @@ -691,6 +695,7 @@ static void bcache_device_link(struct bcache_device *d, struct cache_set *c, { unsigned int i; struct cache *ca; + int ret; for_each_cache(ca, d->c, i) bd_link_disk_holder(ca->bdev, d->disk); @@ -698,9 +703,13 @@ static void bcache_device_link(struct bcache_device *d, struct cache_set *c, snprintf(d->name, BCACHEDEVNAME_SIZE, "%s%u", name, d->id); - WARN(sysfs_create_link(&d->kobj, &c->kobj, "cache") || - sysfs_create_link(&c->kobj, &d->kobj, d->name), - "Couldn't create device <-> cache set symlinks"); + ret = sysfs_create_link(&d->kobj, &c->kobj, "cache"); + if (ret < 0) + pr_err("Couldn't create device -> cache set symlink"); + + ret = sysfs_create_link(&c->kobj, &d->kobj, d->name); + if (ret < 0) + pr_err("Couldn't create cache set -> device symlink"); clear_bit(BCACHE_DEV_UNLINK_DONE, &d->flags); } @@ -908,7 +917,7 @@ static int cached_dev_status_update(void *arg) } -void bch_cached_dev_run(struct cached_dev *dc) +int bch_cached_dev_run(struct cached_dev *dc) { struct bcache_device *d = &dc->disk; char *buf = kmemdup_nul(dc->sb.label, SB_LABEL_SIZE, GFP_KERNEL); @@ -919,11 +928,19 @@ void bch_cached_dev_run(struct cached_dev *dc) NULL, }; + if (dc->io_disable) { + pr_err("I/O disabled on cached dev %s", + dc->backing_dev_name); + return -EIO; + } + if (atomic_xchg(&dc->running, 1)) { kfree(env[1]); kfree(env[2]); kfree(buf); - return; + pr_info("cached dev %s is running already", + dc->backing_dev_name); + return -EBUSY; } if (!d->c && @@ -949,8 +966,11 @@ void bch_cached_dev_run(struct cached_dev *dc) kfree(buf); 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"); + sysfs_create_link(&disk_to_dev(d->disk)->kobj, + &d->kobj, "bcache")) { + pr_err("Couldn't create bcache dev <-> disk sysfs symlinks"); + return -ENOMEM; + } dc->status_update_thread = kthread_run(cached_dev_status_update, dc, "bcache_status_update"); @@ -959,6 +979,8 @@ void bch_cached_dev_run(struct cached_dev *dc) "continue to run without monitoring backing " "device status"); } + + return 0; } /* @@ -996,7 +1018,6 @@ static void cached_dev_detach_finish(struct work_struct *w) BUG_ON(!test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags)); BUG_ON(refcount_read(&dc->count)); - mutex_lock(&bch_register_lock); if (test_and_clear_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags)) cancel_writeback_rate_update_dwork(dc); @@ -1012,6 +1033,8 @@ static void cached_dev_detach_finish(struct work_struct *w) bch_write_bdev_super(dc, &cl); closure_sync(&cl); + mutex_lock(&bch_register_lock); + calc_cached_dev_sectors(dc->disk.c); bcache_device_detach(&dc->disk); list_move(&dc->list, &uncached_devices); @@ -1054,6 +1077,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, uint32_t rtime = cpu_to_le32((u32)ktime_get_real_seconds()); struct uuid_entry *u; struct cached_dev *exist_dc, *t; + int ret = 0; if ((set_uuid && memcmp(set_uuid, c->sb.set_uuid, 16)) || (!set_uuid && memcmp(dc->sb.set_uuid, c->sb.set_uuid, 16))) @@ -1153,6 +1177,8 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, down_write(&dc->writeback_lock); if (bch_cached_dev_writeback_start(dc)) { up_write(&dc->writeback_lock); + pr_err("Couldn't start writeback facilities for %s", + dc->disk.disk->disk_name); return -ENOMEM; } @@ -1163,7 +1189,22 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, bch_sectors_dirty_init(&dc->disk); - bch_cached_dev_run(dc); + ret = bch_cached_dev_run(dc); + if (ret && (ret != -EBUSY)) { + up_write(&dc->writeback_lock); + /* + * bch_register_lock is held, bcache_device_stop() is not + * able to be directly called. The kthread and kworker + * created previously in bch_cached_dev_writeback_start() + * have to be stopped manually here. + */ + kthread_stop(dc->writeback_thread); + cancel_writeback_rate_update_dwork(dc); + pr_err("Couldn't run cached device %s", + dc->backing_dev_name); + return ret; + } + bcache_device_link(&dc->disk, c, "bdev"); atomic_inc(&c->attached_dev_nr); @@ -1190,18 +1231,16 @@ static void cached_dev_free(struct closure *cl) { struct cached_dev *dc = container_of(cl, struct cached_dev, disk.cl); - mutex_lock(&bch_register_lock); - if (test_and_clear_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags)) cancel_writeback_rate_update_dwork(dc); if (!IS_ERR_OR_NULL(dc->writeback_thread)) 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); + mutex_lock(&bch_register_lock); + if (atomic_read(&dc->running)) bd_unlink_disk_holder(dc->bdev, dc->disk.disk); bcache_device_free(&dc->disk); @@ -1290,6 +1329,7 @@ static int register_bdev(struct cache_sb *sb, struct page *sb_page, { const char *err = "cannot allocate memory"; struct cache_set *c; + int ret = -ENOMEM; bdevname(bdev, dc->backing_dev_name); memcpy(&dc->sb, sb, sizeof(struct cache_sb)); @@ -1319,14 +1359,18 @@ static int register_bdev(struct cache_sb *sb, struct page *sb_page, bch_cached_dev_attach(dc, c, NULL); if (BDEV_STATE(&dc->sb) == BDEV_STATE_NONE || - BDEV_STATE(&dc->sb) == BDEV_STATE_STALE) - bch_cached_dev_run(dc); + BDEV_STATE(&dc->sb) == BDEV_STATE_STALE) { + err = "failed to run cached device"; + ret = bch_cached_dev_run(dc); + if (ret) + goto err; + } return 0; err: pr_notice("error %s: %s", dc->backing_dev_name, err); bcache_device_stop(&dc->disk); - return -EIO; + return ret; } /* Flash only volumes */ @@ -1437,8 +1481,6 @@ int bch_flash_dev_create(struct cache_set *c, uint64_t size) bool bch_cached_dev_error(struct cached_dev *dc) { - struct cache_set *c; - if (!dc || test_bit(BCACHE_DEV_CLOSING, &dc->disk.flags)) return false; @@ -1449,21 +1491,6 @@ bool bch_cached_dev_error(struct cached_dev *dc) pr_err("stop %s: too many IO errors on backing device %s\n", 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; } @@ -1564,19 +1591,23 @@ static void cache_set_flush(struct closure *cl) kobject_put(&c->internal); kobject_del(&c->kobj); - if (c->gc_thread) + if (!IS_ERR_OR_NULL(c->gc_thread)) kthread_stop(c->gc_thread); if (!IS_ERR_OR_NULL(c->root)) list_add(&c->root->list, &c->btree_cache); - /* Should skip this if we're unregistering because of an error */ - list_for_each_entry(b, &c->btree_cache, list) { - mutex_lock(&b->write_lock); - if (btree_node_dirty(b)) - __bch_btree_node_write(b, NULL); - mutex_unlock(&b->write_lock); - } + /* + * Avoid flushing cached nodes if cache set is retiring + * due to too many I/O errors detected. + */ + if (!test_bit(CACHE_SET_IO_DISABLE, &c->flags)) + list_for_each_entry(b, &c->btree_cache, list) { + mutex_lock(&b->write_lock); + if (btree_node_dirty(b)) + __bch_btree_node_write(b, NULL); + mutex_unlock(&b->write_lock); + } for_each_cache(ca, c, i) if (ca->alloc_thread) @@ -1849,6 +1880,23 @@ static int run_cache_set(struct cache_set *c) if (bch_btree_check(c)) goto err; + /* + * bch_btree_check() may occupy too much system memory which + * has negative effects to user space application (e.g. data + * base) performance. Shrink the mca cache memory proactively + * here to avoid competing memory with user space workloads.. + */ + if (!c->shrinker_disabled) { + struct shrink_control sc; + + sc.gfp_mask = GFP_KERNEL; + sc.nr_to_scan = c->btree_cache_used * c->btree_pages; + /* first run to clear b->accessed tag */ + c->shrink.scan_objects(&c->shrink, &sc); + /* second run to reap non-accessed nodes */ + c->shrink.scan_objects(&c->shrink, &sc); + } + bch_journal_mark(c, &journal); bch_initial_gc_finish(c); pr_debug("btree_check() done"); @@ -1957,7 +2005,7 @@ err: } closure_sync(&cl); - /* XXX: test this, it's broken */ + bch_cache_set_error(c, "%s", err); return -EIO; @@ -2251,9 +2299,13 @@ err: static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr, const char *buffer, size_t size); +static ssize_t bch_pending_bdevs_cleanup(struct kobject *k, + struct kobj_attribute *attr, + const char *buffer, size_t size); kobj_attribute_write(register, register_bcache); kobj_attribute_write(register_quiet, register_bcache); +kobj_attribute_write(pendings_cleanup, bch_pending_bdevs_cleanup); static bool bch_is_open_backing(struct block_device *bdev) { @@ -2301,6 +2353,11 @@ static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr, if (!try_module_get(THIS_MODULE)) return -EBUSY; + /* For latest state of bcache_is_reboot */ + smp_mb(); + if (bcache_is_reboot) + return -EBUSY; + path = kstrndup(buffer, size, GFP_KERNEL); if (!path) goto err; @@ -2378,8 +2435,61 @@ err: goto out; } + +struct pdev { + struct list_head list; + struct cached_dev *dc; +}; + +static ssize_t bch_pending_bdevs_cleanup(struct kobject *k, + struct kobj_attribute *attr, + const char *buffer, + size_t size) +{ + LIST_HEAD(pending_devs); + ssize_t ret = size; + struct cached_dev *dc, *tdc; + struct pdev *pdev, *tpdev; + struct cache_set *c, *tc; + + mutex_lock(&bch_register_lock); + list_for_each_entry_safe(dc, tdc, &uncached_devices, list) { + pdev = kmalloc(sizeof(struct pdev), GFP_KERNEL); + if (!pdev) + break; + pdev->dc = dc; + list_add(&pdev->list, &pending_devs); + } + + list_for_each_entry_safe(pdev, tpdev, &pending_devs, list) { + list_for_each_entry_safe(c, tc, &bch_cache_sets, list) { + char *pdev_set_uuid = pdev->dc->sb.set_uuid; + char *set_uuid = c->sb.uuid; + + if (!memcmp(pdev_set_uuid, set_uuid, 16)) { + list_del(&pdev->list); + kfree(pdev); + break; + } + } + } + mutex_unlock(&bch_register_lock); + + list_for_each_entry_safe(pdev, tpdev, &pending_devs, list) { + pr_info("delete pdev %p", pdev); + list_del(&pdev->list); + bcache_device_stop(&pdev->dc->disk); + kfree(pdev); + } + + return ret; +} + static int bcache_reboot(struct notifier_block *n, unsigned long code, void *x) { + if (bcache_is_reboot) + return NOTIFY_DONE; + if (code == SYS_DOWN || code == SYS_HALT || code == SYS_POWER_OFF) { @@ -2392,19 +2502,45 @@ static int bcache_reboot(struct notifier_block *n, unsigned long code, void *x) mutex_lock(&bch_register_lock); + if (bcache_is_reboot) + goto out; + + /* New registration is rejected since now */ + bcache_is_reboot = true; + /* + * Make registering caller (if there is) on other CPU + * core know bcache_is_reboot set to true earlier + */ + smp_mb(); + if (list_empty(&bch_cache_sets) && list_empty(&uncached_devices)) goto out; + mutex_unlock(&bch_register_lock); + pr_info("Stopping all devices:"); + /* + * The reason bch_register_lock is not held to call + * bch_cache_set_stop() and bcache_device_stop() is to + * avoid potential deadlock during reboot, because cache + * set or bcache device stopping process will acqurie + * bch_register_lock too. + * + * We are safe here because bcache_is_reboot sets to + * true already, register_bcache() will reject new + * registration now. bcache_is_reboot also makes sure + * bcache_reboot() won't be re-entered on by other thread, + * so there is no race in following list iteration by + * list_for_each_entry_safe(). + */ list_for_each_entry_safe(c, tc, &bch_cache_sets, list) bch_cache_set_stop(c); list_for_each_entry_safe(dc, tdc, &uncached_devices, list) bcache_device_stop(&dc->disk); - mutex_unlock(&bch_register_lock); /* * Give an early chance for other kthreads and @@ -2496,6 +2632,7 @@ static int __init bcache_init(void) static const struct attribute *files[] = { &ksysfs_register.attr, &ksysfs_register_quiet.attr, + &ksysfs_pendings_cleanup.attr, NULL }; @@ -2531,6 +2668,8 @@ static int __init bcache_init(void) bch_debug_init(); closure_debug_init(); + bcache_is_reboot = false; + return 0; err: bcache_exit(); diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c index bfb437ffb13c..9f0826712845 100644 --- a/drivers/md/bcache/sysfs.c +++ b/drivers/md/bcache/sysfs.c @@ -16,33 +16,31 @@ #include <linux/sort.h> #include <linux/sched/clock.h> +extern bool bcache_is_reboot; + /* Default is 0 ("writethrough") */ static const char * const bch_cache_modes[] = { "writethrough", "writeback", "writearound", - "none", - NULL + "none" }; /* Default is 0 ("auto") */ static const char * const bch_stop_on_failure_modes[] = { "auto", - "always", - NULL + "always" }; static const char * const cache_replacement_policies[] = { "lru", "fifo", - "random", - NULL + "random" }; static const char * const error_actions[] = { "unregister", - "panic", - NULL + "panic" }; write_attribute(attach); @@ -84,8 +82,8 @@ read_attribute(bset_tree_stats); read_attribute(state); read_attribute(cache_read_races); read_attribute(reclaim); +read_attribute(reclaimed_journal_buckets); read_attribute(flush_write); -read_attribute(retry_flush_write); read_attribute(writeback_keys_done); read_attribute(writeback_keys_failed); read_attribute(io_errors); @@ -180,7 +178,7 @@ SHOW(__bch_cached_dev) var_print(writeback_percent); sysfs_hprint(writeback_rate, wb ? atomic_long_read(&dc->writeback_rate.rate) << 9 : 0); - sysfs_hprint(io_errors, atomic_read(&dc->io_errors)); + sysfs_printf(io_errors, "%i", atomic_read(&dc->io_errors)); sysfs_printf(io_error_limit, "%i", dc->error_limit); sysfs_printf(io_disable, "%i", dc->io_disable); var_print(writeback_rate_update_seconds); @@ -271,6 +269,10 @@ STORE(__cached_dev) struct cache_set *c; struct kobj_uevent_env *env; + /* no user space access if system is rebooting */ + if (bcache_is_reboot) + return -EBUSY; + #define d_strtoul(var) sysfs_strtoul(var, dc->var) #define d_strtoul_nonzero(var) sysfs_strtoul_clamp(var, dc->var, 1, INT_MAX) #define d_strtoi_h(var) sysfs_hatoi(var, dc->var) @@ -329,11 +331,14 @@ STORE(__cached_dev) bch_cache_accounting_clear(&dc->accounting); if (attr == &sysfs_running && - strtoul_or_return(buf)) - bch_cached_dev_run(dc); + strtoul_or_return(buf)) { + v = bch_cached_dev_run(dc); + if (v) + return v; + } if (attr == &sysfs_cache_mode) { - v = __sysfs_match_string(bch_cache_modes, -1, buf); + v = sysfs_match_string(bch_cache_modes, buf); if (v < 0) return v; @@ -344,7 +349,7 @@ STORE(__cached_dev) } if (attr == &sysfs_stop_when_cache_set_failed) { - v = __sysfs_match_string(bch_stop_on_failure_modes, -1, buf); + v = sysfs_match_string(bch_stop_on_failure_modes, buf); if (v < 0) return v; @@ -408,6 +413,10 @@ STORE(bch_cached_dev) struct cached_dev *dc = container_of(kobj, struct cached_dev, disk.kobj); + /* no user space access if system is rebooting */ + if (bcache_is_reboot) + return -EBUSY; + mutex_lock(&bch_register_lock); size = __cached_dev_store(kobj, attr, buf, size); @@ -464,7 +473,7 @@ static struct attribute *bch_cached_dev_files[] = { &sysfs_writeback_rate_p_term_inverse, &sysfs_writeback_rate_minimum, &sysfs_writeback_rate_debug, - &sysfs_errors, + &sysfs_io_errors, &sysfs_io_error_limit, &sysfs_io_disable, &sysfs_dirty_data, @@ -511,6 +520,10 @@ STORE(__bch_flash_dev) kobj); struct uuid_entry *u = &d->c->uuids[d->id]; + /* no user space access if system is rebooting */ + if (bcache_is_reboot) + return -EBUSY; + sysfs_strtoul(data_csum, d->data_csum); if (attr == &sysfs_size) { @@ -693,12 +706,12 @@ SHOW(__bch_cache_set) sysfs_print(reclaim, atomic_long_read(&c->reclaim)); + sysfs_print(reclaimed_journal_buckets, + atomic_long_read(&c->reclaimed_journal_buckets)); + sysfs_print(flush_write, atomic_long_read(&c->flush_write)); - sysfs_print(retry_flush_write, - atomic_long_read(&c->retry_flush_write)); - sysfs_print(writeback_keys_done, atomic_long_read(&c->writeback_keys_done)); sysfs_print(writeback_keys_failed, @@ -746,6 +759,10 @@ STORE(__bch_cache_set) struct cache_set *c = container_of(kobj, struct cache_set, kobj); ssize_t v; + /* no user space access if system is rebooting */ + if (bcache_is_reboot) + return -EBUSY; + if (attr == &sysfs_unregister) bch_cache_set_unregister(c); @@ -799,7 +816,7 @@ STORE(__bch_cache_set) 0, UINT_MAX); if (attr == &sysfs_errors) { - v = __sysfs_match_string(error_actions, -1, buf); + v = sysfs_match_string(error_actions, buf); if (v < 0) return v; @@ -865,6 +882,10 @@ STORE(bch_cache_set_internal) { struct cache_set *c = container_of(kobj, struct cache_set, internal); + /* no user space access if system is rebooting */ + if (bcache_is_reboot) + return -EBUSY; + return bch_cache_set_store(&c->kobj, attr, buf, size); } @@ -914,8 +935,8 @@ static struct attribute *bch_cache_set_internal_files[] = { &sysfs_bset_tree_stats, &sysfs_cache_read_races, &sysfs_reclaim, + &sysfs_reclaimed_journal_buckets, &sysfs_flush_write, - &sysfs_retry_flush_write, &sysfs_writeback_keys_done, &sysfs_writeback_keys_failed, @@ -1050,6 +1071,10 @@ STORE(__bch_cache) struct cache *ca = container_of(kobj, struct cache, kobj); ssize_t v; + /* no user space access if system is rebooting */ + if (bcache_is_reboot) + return -EBUSY; + if (attr == &sysfs_discard) { bool v = strtoul_or_return(buf); @@ -1063,7 +1088,7 @@ STORE(__bch_cache) } if (attr == &sysfs_cache_replacement_policy) { - v = __sysfs_match_string(cache_replacement_policies, -1, buf); + v = sysfs_match_string(cache_replacement_policies, buf); if (v < 0) return v; diff --git a/drivers/md/bcache/util.h b/drivers/md/bcache/util.h index 1fbced94e4cc..c029f7443190 100644 --- a/drivers/md/bcache/util.h +++ b/drivers/md/bcache/util.h @@ -113,8 +113,6 @@ do { \ #define heap_full(h) ((h)->used == (h)->size) -#define heap_empty(h) ((h)->used == 0) - #define DECLARE_FIFO(type, name) \ struct { \ size_t front, back, size, mask; \ diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c index 73f0efac2b9f..d60268fe49e1 100644 --- a/drivers/md/bcache/writeback.c +++ b/drivers/md/bcache/writeback.c @@ -122,6 +122,9 @@ static void __update_writeback_rate(struct cached_dev *dc) static bool set_at_max_writeback_rate(struct cache_set *c, struct cached_dev *dc) { + /* Don't set max writeback rate if gc is running */ + if (!c->gc_mark_valid) + return false; /* * Idle_counter is increased everytime when update_writeback_rate() is * called. If all backing devices attached to the same cache set have @@ -735,6 +738,10 @@ static int bch_writeback_thread(void *arg) } } + if (dc->writeback_write_wq) { + flush_workqueue(dc->writeback_write_wq); + destroy_workqueue(dc->writeback_write_wq); + } cached_dev_put(dc); wait_for_kthread_stop(); @@ -830,6 +837,7 @@ int bch_cached_dev_writeback_start(struct cached_dev *dc) "bcache_writeback"); if (IS_ERR(dc->writeback_thread)) { cached_dev_put(dc); + destroy_workqueue(dc->writeback_write_wq); return PTR_ERR(dc->writeback_thread); } dc->writeback_running = true; diff --git a/drivers/md/md-bitmap.c b/drivers/md/md-bitmap.c index c01d41198f5e..b092c7b5282f 100644 --- a/drivers/md/md-bitmap.c +++ b/drivers/md/md-bitmap.c @@ -1790,6 +1790,8 @@ void md_bitmap_destroy(struct mddev *mddev) return; md_bitmap_wait_behind_writes(mddev); + mempool_destroy(mddev->wb_info_pool); + mddev->wb_info_pool = NULL; mutex_lock(&mddev->bitmap_info.mutex); spin_lock(&mddev->lock); @@ -1900,10 +1902,14 @@ int md_bitmap_load(struct mddev *mddev) sector_t start = 0; sector_t sector = 0; struct bitmap *bitmap = mddev->bitmap; + struct md_rdev *rdev; if (!bitmap) goto out; + rdev_for_each(rdev, mddev) + mddev_create_wb_pool(mddev, rdev, true); + if (mddev_is_clustered(mddev)) md_cluster_ops->load_bitmaps(mddev, mddev->bitmap_info.nodes); @@ -2462,12 +2468,26 @@ static ssize_t backlog_store(struct mddev *mddev, const char *buf, size_t len) { unsigned long backlog; + unsigned long old_mwb = mddev->bitmap_info.max_write_behind; int rv = kstrtoul(buf, 10, &backlog); if (rv) return rv; if (backlog > COUNTER_MAX) return -EINVAL; mddev->bitmap_info.max_write_behind = backlog; + if (!backlog && mddev->wb_info_pool) { + /* wb_info_pool is not needed if backlog is zero */ + mempool_destroy(mddev->wb_info_pool); + mddev->wb_info_pool = NULL; + } else if (backlog && !mddev->wb_info_pool) { + /* wb_info_pool is needed since backlog is not zero */ + struct md_rdev *rdev; + + rdev_for_each(rdev, mddev) + mddev_create_wb_pool(mddev, rdev, false); + } + if (old_mwb != backlog) + md_bitmap_update_sb(mddev->bitmap); return len; } diff --git a/drivers/md/md.c b/drivers/md/md.c index 9801d540fea1..a114b05e3db4 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -37,6 +37,7 @@ */ +#include <linux/sched/mm.h> #include <linux/sched/signal.h> #include <linux/kthread.h> #include <linux/blkdev.h> @@ -124,6 +125,77 @@ static inline int speed_max(struct mddev *mddev) mddev->sync_speed_max : sysctl_speed_limit_max; } +static int rdev_init_wb(struct md_rdev *rdev) +{ + if (rdev->bdev->bd_queue->nr_hw_queues == 1) + return 0; + + spin_lock_init(&rdev->wb_list_lock); + INIT_LIST_HEAD(&rdev->wb_list); + init_waitqueue_head(&rdev->wb_io_wait); + set_bit(WBCollisionCheck, &rdev->flags); + + return 1; +} + +/* + * Create wb_info_pool if rdev is the first multi-queue device flaged + * with writemostly, also write-behind mode is enabled. + */ +void mddev_create_wb_pool(struct mddev *mddev, struct md_rdev *rdev, + bool is_suspend) +{ + if (mddev->bitmap_info.max_write_behind == 0) + return; + + if (!test_bit(WriteMostly, &rdev->flags) || !rdev_init_wb(rdev)) + return; + + if (mddev->wb_info_pool == NULL) { + unsigned int noio_flag; + + if (!is_suspend) + mddev_suspend(mddev); + noio_flag = memalloc_noio_save(); + mddev->wb_info_pool = mempool_create_kmalloc_pool(NR_WB_INFOS, + sizeof(struct wb_info)); + memalloc_noio_restore(noio_flag); + if (!mddev->wb_info_pool) + pr_err("can't alloc memory pool for writemostly\n"); + if (!is_suspend) + mddev_resume(mddev); + } +} +EXPORT_SYMBOL_GPL(mddev_create_wb_pool); + +/* + * destroy wb_info_pool if rdev is the last device flaged with WBCollisionCheck. + */ +static void mddev_destroy_wb_pool(struct mddev *mddev, struct md_rdev *rdev) +{ + if (!test_and_clear_bit(WBCollisionCheck, &rdev->flags)) + return; + + if (mddev->wb_info_pool) { + struct md_rdev *temp; + int num = 0; + + /* + * Check if other rdevs need wb_info_pool. + */ + rdev_for_each(temp, mddev) + if (temp != rdev && + test_bit(WBCollisionCheck, &temp->flags)) + num++; + if (!num) { + mddev_suspend(rdev->mddev); + mempool_destroy(mddev->wb_info_pool); + mddev->wb_info_pool = NULL; + mddev_resume(rdev->mddev); + } + } +} + static struct ctl_table_header *raid_table_header; static struct ctl_table raid_table[] = { @@ -2210,6 +2282,9 @@ static int bind_rdev_to_array(struct md_rdev *rdev, struct mddev *mddev) rdev->mddev = mddev; pr_debug("md: bind<%s>\n", b); + if (mddev->raid_disks) + mddev_create_wb_pool(mddev, rdev, false); + if ((err = kobject_add(&rdev->kobj, &mddev->kobj, "dev-%s", b))) goto fail; @@ -2246,6 +2321,7 @@ static void unbind_rdev_from_array(struct md_rdev *rdev) bd_unlink_disk_holder(rdev->bdev, rdev->mddev->gendisk); list_del_rcu(&rdev->same_set); pr_debug("md: unbind<%s>\n", bdevname(rdev->bdev,b)); + mddev_destroy_wb_pool(rdev->mddev, rdev); rdev->mddev = NULL; sysfs_remove_link(&rdev->kobj, "block"); sysfs_put(rdev->sysfs_state); @@ -2758,8 +2834,10 @@ state_store(struct md_rdev *rdev, const char *buf, size_t len) } } else if (cmd_match(buf, "writemostly")) { set_bit(WriteMostly, &rdev->flags); + mddev_create_wb_pool(rdev->mddev, rdev, false); err = 0; } else if (cmd_match(buf, "-writemostly")) { + mddev_destroy_wb_pool(rdev->mddev, rdev); clear_bit(WriteMostly, &rdev->flags); err = 0; } else if (cmd_match(buf, "blocked")) { @@ -3356,7 +3434,7 @@ rdev_attr_show(struct kobject *kobj, struct attribute *attr, char *page) if (!entry->show) return -EIO; if (!rdev->mddev) - return -EBUSY; + return -ENODEV; return entry->show(rdev, page); } @@ -5588,15 +5666,28 @@ int md_run(struct mddev *mddev) mddev->bitmap = bitmap; } - if (err) { - mddev_detach(mddev); - if (mddev->private) - pers->free(mddev, mddev->private); - mddev->private = NULL; - module_put(pers->owner); - md_bitmap_destroy(mddev); - goto abort; + if (err) + goto bitmap_abort; + + if (mddev->bitmap_info.max_write_behind > 0) { + bool creat_pool = false; + + rdev_for_each(rdev, mddev) { + if (test_bit(WriteMostly, &rdev->flags) && + rdev_init_wb(rdev)) + creat_pool = true; + } + if (creat_pool && mddev->wb_info_pool == NULL) { + mddev->wb_info_pool = + mempool_create_kmalloc_pool(NR_WB_INFOS, + sizeof(struct wb_info)); + if (!mddev->wb_info_pool) { + err = -ENOMEM; + goto bitmap_abort; + } + } } + if (mddev->queue) { bool nonrot = true; @@ -5639,8 +5730,7 @@ int md_run(struct mddev *mddev) spin_unlock(&mddev->lock); rdev_for_each(rdev, mddev) if (rdev->raid_disk >= 0) - if (sysfs_link_rdev(mddev, rdev)) - /* failure here is OK */; + sysfs_link_rdev(mddev, rdev); /* failure here is OK */ if (mddev->degraded && !mddev->ro) /* This ensures that recovering status is reported immediately @@ -5658,6 +5748,13 @@ int md_run(struct mddev *mddev) sysfs_notify(&mddev->kobj, NULL, "degraded"); return 0; +bitmap_abort: + mddev_detach(mddev); + if (mddev->private) + pers->free(mddev, mddev->private); + mddev->private = NULL; + module_put(pers->owner); + md_bitmap_destroy(mddev); abort: bioset_exit(&mddev->bio_set); bioset_exit(&mddev->sync_set); @@ -5826,6 +5923,8 @@ static void __md_stop_writes(struct mddev *mddev) mddev->in_sync = 1; md_update_sb(mddev, 1); } + mempool_destroy(mddev->wb_info_pool); + mddev->wb_info_pool = NULL; } void md_stop_writes(struct mddev *mddev) @@ -8198,8 +8297,7 @@ void md_do_sync(struct md_thread *thread) { struct mddev *mddev = thread->mddev; struct mddev *mddev2; - unsigned int currspeed = 0, - window; + unsigned int currspeed = 0, window; sector_t max_sectors,j, io_sectors, recovery_done; unsigned long mark[SYNC_MARKS]; unsigned long update_time; @@ -8256,7 +8354,7 @@ void md_do_sync(struct md_thread *thread) * 0 == not engaged in resync at all * 2 == checking that there is no conflict with another sync * 1 == like 2, but have yielded to allow conflicting resync to - * commense + * commence * other == active in resync - this many blocks * * Before starting a resync we must have set curr_resync to @@ -8387,7 +8485,7 @@ void md_do_sync(struct md_thread *thread) /* * Tune reconstruction: */ - window = 32*(PAGE_SIZE/512); + window = 32 * (PAGE_SIZE / 512); pr_debug("md: using %dk window, over a total of %lluk.\n", window/2, (unsigned long long)max_sectors/2); @@ -9200,7 +9298,6 @@ static void check_sb_changes(struct mddev *mddev, struct md_rdev *rdev) * perform resync with the new activated disk */ set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); md_wakeup_thread(mddev->thread); - } /* device faulty * We just want to do the minimum to mark the disk diff --git a/drivers/md/md.h b/drivers/md/md.h index 7c930c091193..10f98200e2f8 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -109,6 +109,14 @@ struct md_rdev { * for reporting to userspace and storing * in superblock. */ + + /* + * The members for check collision of write behind IOs. + */ + struct list_head wb_list; + spinlock_t wb_list_lock; + wait_queue_head_t wb_io_wait; + struct work_struct del_work; /* used for delayed sysfs removal */ struct kernfs_node *sysfs_state; /* handle for 'state' @@ -193,6 +201,10 @@ enum flag_bits { * it didn't fail, so don't use FailFast * any more for metadata */ + WBCollisionCheck, /* + * multiqueue device should check if there + * is collision between write behind bios. + */ }; static inline int is_badblock(struct md_rdev *rdev, sector_t s, int sectors, @@ -245,6 +257,14 @@ enum mddev_sb_flags { MD_SB_NEED_REWRITE, /* metadata write needs to be repeated */ }; +#define NR_WB_INFOS 8 +/* record current range of write behind IOs */ +struct wb_info { + sector_t lo; + sector_t hi; + struct list_head list; +}; + struct mddev { void *private; struct md_personality *pers; @@ -461,6 +481,7 @@ struct mddev { */ struct work_struct flush_work; struct work_struct event_work; /* used by dm to report failure event */ + mempool_t *wb_info_pool; void (*sync_super)(struct mddev *mddev, struct md_rdev *rdev); struct md_cluster_info *cluster_info; unsigned int good_device_nr; /* good device num within cluster raid */ @@ -709,6 +730,8 @@ extern struct bio *bio_alloc_mddev(gfp_t gfp_mask, int nr_iovecs, extern void md_reload_sb(struct mddev *mddev, int raid_disk); extern void md_update_sb(struct mddev *mddev, int force); extern void md_kick_rdev_from_array(struct md_rdev * rdev); +extern void mddev_create_wb_pool(struct mddev *mddev, struct md_rdev *rdev, + bool is_suspend); struct md_rdev *md_find_rdev_nr_rcu(struct mddev *mddev, int nr); struct md_rdev *md_find_rdev_rcu(struct mddev *mddev, dev_t dev); diff --git a/drivers/md/raid1-10.c b/drivers/md/raid1-10.c index 400001b815db..54db34163968 100644 --- a/drivers/md/raid1-10.c +++ b/drivers/md/raid1-10.c @@ -3,12 +3,42 @@ #define RESYNC_BLOCK_SIZE (64*1024) #define RESYNC_PAGES ((RESYNC_BLOCK_SIZE + PAGE_SIZE-1) / PAGE_SIZE) +/* + * Number of guaranteed raid bios in case of extreme VM load: + */ +#define NR_RAID_BIOS 256 + +/* when we get a read error on a read-only array, we redirect to another + * device without failing the first device, or trying to over-write to + * correct the read error. To keep track of bad blocks on a per-bio + * level, we store IO_BLOCKED in the appropriate 'bios' pointer + */ +#define IO_BLOCKED ((struct bio *)1) +/* When we successfully write to a known bad-block, we need to remove the + * bad-block marking which must be done from process context. So we record + * the success by setting devs[n].bio to IO_MADE_GOOD + */ +#define IO_MADE_GOOD ((struct bio *)2) + +#define BIO_SPECIAL(bio) ((unsigned long)bio <= 2) + +/* When there are this many requests queue to be written by + * the raid thread, we become 'congested' to provide back-pressure + * for writeback. + */ +static int max_queued_requests = 1024; + /* for managing resync I/O pages */ struct resync_pages { void *raid_bio; struct page *pages[RESYNC_PAGES]; }; +static void rbio_pool_free(void *rbio, void *data) +{ + kfree(rbio); +} + static inline int resync_alloc_pages(struct resync_pages *rp, gfp_t gfp_flags) { diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 2aa36e570e04..34e26834ad28 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -42,31 +42,6 @@ (1L << MD_HAS_PPL) | \ (1L << MD_HAS_MULTIPLE_PPLS)) -/* - * Number of guaranteed r1bios in case of extreme VM load: - */ -#define NR_RAID1_BIOS 256 - -/* when we get a read error on a read-only array, we redirect to another - * device without failing the first device, or trying to over-write to - * correct the read error. To keep track of bad blocks on a per-bio - * level, we store IO_BLOCKED in the appropriate 'bios' pointer - */ -#define IO_BLOCKED ((struct bio *)1) -/* When we successfully write to a known bad-block, we need to remove the - * bad-block marking which must be done from process context. So we record - * the success by setting devs[n].bio to IO_MADE_GOOD - */ -#define IO_MADE_GOOD ((struct bio *)2) - -#define BIO_SPECIAL(bio) ((unsigned long)bio <= 2) - -/* When there are this many requests queue to be written by - * the raid1 thread, we become 'congested' to provide back-pressure - * for writeback. - */ -static int max_queued_requests = 1024; - static void allow_barrier(struct r1conf *conf, sector_t sector_nr); static void lower_barrier(struct r1conf *conf, sector_t sector_nr); @@ -75,6 +50,57 @@ static void lower_barrier(struct r1conf *conf, sector_t sector_nr); #include "raid1-10.c" +static int check_and_add_wb(struct md_rdev *rdev, sector_t lo, sector_t hi) +{ + struct wb_info *wi, *temp_wi; + unsigned long flags; + int ret = 0; + struct mddev *mddev = rdev->mddev; + + wi = mempool_alloc(mddev->wb_info_pool, GFP_NOIO); + + spin_lock_irqsave(&rdev->wb_list_lock, flags); + list_for_each_entry(temp_wi, &rdev->wb_list, list) { + /* collision happened */ + if (hi > temp_wi->lo && lo < temp_wi->hi) { + ret = -EBUSY; + break; + } + } + + if (!ret) { + wi->lo = lo; + wi->hi = hi; + list_add(&wi->list, &rdev->wb_list); + } else + mempool_free(wi, mddev->wb_info_pool); + spin_unlock_irqrestore(&rdev->wb_list_lock, flags); + + return ret; +} + +static void remove_wb(struct md_rdev *rdev, sector_t lo, sector_t hi) +{ + struct wb_info *wi; + unsigned long flags; + int found = 0; + struct mddev *mddev = rdev->mddev; + + spin_lock_irqsave(&rdev->wb_list_lock, flags); + list_for_each_entry(wi, &rdev->wb_list, list) + if (hi == wi->hi && lo == wi->lo) { + list_del(&wi->list); + mempool_free(wi, mddev->wb_info_pool); + found = 1; + break; + } + + if (!found) + WARN(1, "The write behind IO is not recorded\n"); + spin_unlock_irqrestore(&rdev->wb_list_lock, flags); + wake_up(&rdev->wb_io_wait); +} + /* * for resync bio, r1bio pointer can be retrieved from the per-bio * 'struct resync_pages'. @@ -93,11 +119,6 @@ static void * r1bio_pool_alloc(gfp_t gfp_flags, void *data) return kzalloc(size, gfp_flags); } -static void r1bio_pool_free(void *r1_bio, void *data) -{ - kfree(r1_bio); -} - #define RESYNC_DEPTH 32 #define RESYNC_SECTORS (RESYNC_BLOCK_SIZE >> 9) #define RESYNC_WINDOW (RESYNC_BLOCK_SIZE * RESYNC_DEPTH) @@ -173,7 +194,7 @@ out_free_bio: kfree(rps); out_free_r1bio: - r1bio_pool_free(r1_bio, data); + rbio_pool_free(r1_bio, data); return NULL; } @@ -193,7 +214,7 @@ static void r1buf_pool_free(void *__r1_bio, void *data) /* resync pages array stored in the 1st bio's .bi_private */ kfree(rp); - r1bio_pool_free(r1bio, data); + rbio_pool_free(r1bio, data); } static void put_all_bios(struct r1conf *conf, struct r1bio *r1_bio) @@ -476,6 +497,12 @@ static void raid1_end_write_request(struct bio *bio) } if (behind) { + if (test_bit(WBCollisionCheck, &rdev->flags)) { + sector_t lo = r1_bio->sector; + sector_t hi = r1_bio->sector + r1_bio->sectors; + + remove_wb(rdev, lo, hi); + } if (test_bit(WriteMostly, &rdev->flags)) atomic_dec(&r1_bio->behind_remaining); @@ -1449,7 +1476,6 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio, if (!r1_bio->bios[i]) continue; - if (first_clone) { /* do behind I/O ? * Not if there are too many, or cannot @@ -1474,7 +1500,16 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio, 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)) + struct md_rdev *rdev = conf->mirrors[i].rdev; + + if (test_bit(WBCollisionCheck, &rdev->flags)) { + sector_t lo = r1_bio->sector; + sector_t hi = r1_bio->sector + r1_bio->sectors; + + wait_event(rdev->wb_io_wait, + check_and_add_wb(rdev, lo, hi) == 0); + } + if (test_bit(WriteMostly, &rdev->flags)) atomic_inc(&r1_bio->behind_remaining); } @@ -1729,9 +1764,8 @@ static int raid1_add_disk(struct mddev *mddev, struct md_rdev *rdev) first = last = rdev->saved_raid_disk; for (mirror = first; mirror <= last; mirror++) { - p = conf->mirrors+mirror; + p = conf->mirrors + mirror; if (!p->rdev) { - if (mddev->gendisk) disk_stack_limits(mddev->gendisk, rdev->bdev, rdev->data_offset << 9); @@ -2888,7 +2922,6 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr, if (read_targets == 1) bio->bi_opf &= ~MD_FAILFAST; generic_make_request(bio); - } return nr_sectors; } @@ -2947,8 +2980,8 @@ static struct r1conf *setup_conf(struct mddev *mddev) if (!conf->poolinfo) goto abort; conf->poolinfo->raid_disks = mddev->raid_disks * 2; - err = mempool_init(&conf->r1bio_pool, NR_RAID1_BIOS, r1bio_pool_alloc, - r1bio_pool_free, conf->poolinfo); + err = mempool_init(&conf->r1bio_pool, NR_RAID_BIOS, r1bio_pool_alloc, + rbio_pool_free, conf->poolinfo); if (err) goto abort; @@ -3089,7 +3122,7 @@ static int raid1_run(struct mddev *mddev) } mddev->degraded = 0; - for (i=0; i < conf->raid_disks; i++) + for (i = 0; i < conf->raid_disks; i++) if (conf->mirrors[i].rdev == NULL || !test_bit(In_sync, &conf->mirrors[i].rdev->flags) || test_bit(Faulty, &conf->mirrors[i].rdev->flags)) @@ -3124,7 +3157,7 @@ static int raid1_run(struct mddev *mddev) mddev->queue); } - ret = md_integrity_register(mddev); + ret = md_integrity_register(mddev); if (ret) { md_unregister_thread(&mddev->thread); raid1_free(mddev, conf); @@ -3232,8 +3265,8 @@ static int raid1_reshape(struct mddev *mddev) newpoolinfo->mddev = mddev; newpoolinfo->raid_disks = raid_disks * 2; - ret = mempool_init(&newpool, NR_RAID1_BIOS, r1bio_pool_alloc, - r1bio_pool_free, newpoolinfo); + ret = mempool_init(&newpool, NR_RAID_BIOS, r1bio_pool_alloc, + rbio_pool_free, newpoolinfo); if (ret) { kfree(newpoolinfo); return ret; diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index aea11476fee6..8a1354a08a1a 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -64,31 +64,6 @@ * [B A] [D C] [B A] [E C D] */ -/* - * Number of guaranteed r10bios in case of extreme VM load: - */ -#define NR_RAID10_BIOS 256 - -/* when we get a read error on a read-only array, we redirect to another - * device without failing the first device, or trying to over-write to - * correct the read error. To keep track of bad blocks on a per-bio - * level, we store IO_BLOCKED in the appropriate 'bios' pointer - */ -#define IO_BLOCKED ((struct bio *)1) -/* When we successfully write to a known bad-block, we need to remove the - * bad-block marking which must be done from process context. So we record - * the success by setting devs[n].bio to IO_MADE_GOOD - */ -#define IO_MADE_GOOD ((struct bio *)2) - -#define BIO_SPECIAL(bio) ((unsigned long)bio <= 2) - -/* When there are this many requests queued to be written by - * the raid10 thread, we become 'congested' to provide back-pressure - * for writeback. - */ -static int max_queued_requests = 1024; - static void allow_barrier(struct r10conf *conf); static void lower_barrier(struct r10conf *conf); static int _enough(struct r10conf *conf, int previous, int ignore); @@ -123,11 +98,6 @@ static void * r10bio_pool_alloc(gfp_t gfp_flags, void *data) return kzalloc(size, gfp_flags); } -static void r10bio_pool_free(void *r10_bio, void *data) -{ - kfree(r10_bio); -} - #define RESYNC_SECTORS (RESYNC_BLOCK_SIZE >> 9) /* amount of memory to reserve for resync requests */ #define RESYNC_WINDOW (1024*1024) @@ -233,7 +203,7 @@ out_free_bio: } kfree(rps); out_free_r10bio: - r10bio_pool_free(r10_bio, conf); + rbio_pool_free(r10_bio, conf); return NULL; } @@ -261,7 +231,7 @@ static void r10buf_pool_free(void *__r10_bio, void *data) /* resync pages array stored in the 1st bio's .bi_private */ kfree(rp); - r10bio_pool_free(r10bio, conf); + rbio_pool_free(r10bio, conf); } static void put_all_bios(struct r10conf *conf, struct r10bio *r10_bio) @@ -737,15 +707,19 @@ static struct md_rdev *read_balance(struct r10conf *conf, int sectors = r10_bio->sectors; int best_good_sectors; sector_t new_distance, best_dist; - struct md_rdev *best_rdev, *rdev = NULL; + struct md_rdev *best_dist_rdev, *best_pending_rdev, *rdev = NULL; int do_balance; - int best_slot; + int best_dist_slot, best_pending_slot; + bool has_nonrot_disk = false; + unsigned int min_pending; struct geom *geo = &conf->geo; raid10_find_phys(conf, r10_bio); rcu_read_lock(); - best_slot = -1; - best_rdev = NULL; + best_dist_slot = -1; + min_pending = UINT_MAX; + best_dist_rdev = NULL; + best_pending_rdev = NULL; best_dist = MaxSector; best_good_sectors = 0; do_balance = 1; @@ -767,6 +741,8 @@ static struct md_rdev *read_balance(struct r10conf *conf, sector_t first_bad; int bad_sectors; sector_t dev_sector; + unsigned int pending; + bool nonrot; if (r10_bio->devs[slot].bio == IO_BLOCKED) continue; @@ -803,8 +779,8 @@ static struct md_rdev *read_balance(struct r10conf *conf, first_bad - dev_sector; if (good_sectors > best_good_sectors) { best_good_sectors = good_sectors; - best_slot = slot; - best_rdev = rdev; + best_dist_slot = slot; + best_dist_rdev = rdev; } if (!do_balance) /* Must read from here */ @@ -817,14 +793,23 @@ static struct md_rdev *read_balance(struct r10conf *conf, if (!do_balance) break; - if (best_slot >= 0) + nonrot = blk_queue_nonrot(bdev_get_queue(rdev->bdev)); + has_nonrot_disk |= nonrot; + pending = atomic_read(&rdev->nr_pending); + if (min_pending > pending && nonrot) { + min_pending = pending; + best_pending_slot = slot; + best_pending_rdev = rdev; + } + + if (best_dist_slot >= 0) /* At least 2 disks to choose from so failfast is OK */ set_bit(R10BIO_FailFast, &r10_bio->state); /* This optimisation is debatable, and completely destroys * sequential read speed for 'far copies' arrays. So only * keep it for 'near' arrays, and review those later. */ - if (geo->near_copies > 1 && !atomic_read(&rdev->nr_pending)) + if (geo->near_copies > 1 && !pending) new_distance = 0; /* for far > 1 always use the lowest address */ @@ -833,15 +818,21 @@ static struct md_rdev *read_balance(struct r10conf *conf, else new_distance = abs(r10_bio->devs[slot].addr - conf->mirrors[disk].head_position); + if (new_distance < best_dist) { best_dist = new_distance; - best_slot = slot; - best_rdev = rdev; + best_dist_slot = slot; + best_dist_rdev = rdev; } } if (slot >= conf->copies) { - slot = best_slot; - rdev = best_rdev; + if (has_nonrot_disk) { + slot = best_pending_slot; + rdev = best_pending_rdev; + } else { + slot = best_dist_slot; + rdev = best_dist_rdev; + } } if (slot >= 0) { @@ -3675,8 +3666,8 @@ static struct r10conf *setup_conf(struct mddev *mddev) conf->geo = geo; conf->copies = copies; - err = mempool_init(&conf->r10bio_pool, NR_RAID10_BIOS, r10bio_pool_alloc, - r10bio_pool_free, conf); + err = mempool_init(&conf->r10bio_pool, NR_RAID_BIOS, r10bio_pool_alloc, + rbio_pool_free, conf); if (err) goto out; @@ -4780,8 +4771,7 @@ static int handle_reshape_read_error(struct mddev *mddev, int idx = 0; struct page **pages; - r10b = kmalloc(sizeof(*r10b) + - sizeof(struct r10dev) * conf->copies, GFP_NOIO); + r10b = kmalloc(struct_size(r10b, devs, conf->copies), GFP_NOIO); if (!r10b) { set_bit(MD_RECOVERY_INTR, &mddev->recovery); return -ENOMEM; diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index b83bce2beb66..3de4e13bde98 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -5251,7 +5251,6 @@ static int raid5_read_one_chunk(struct mddev *mddev, struct bio *raid_bio) rcu_read_unlock(); raid_bio->bi_next = (void*)rdev; bio_set_dev(align_bi, rdev->bdev); - bio_clear_flag(align_bi, BIO_SEG_VALID); if (is_badblock(rdev, align_bi->bi_iter.bi_sector, bio_sectors(align_bi), @@ -7672,7 +7671,7 @@ abort: static int raid5_add_disk(struct mddev *mddev, struct md_rdev *rdev) { struct r5conf *conf = mddev->private; - int err = -EEXIST; + int ret, err = -EEXIST; int disk; struct disk_info *p; int first = 0; @@ -7687,7 +7686,14 @@ static int raid5_add_disk(struct mddev *mddev, struct md_rdev *rdev) * The array is in readonly mode if journal is missing, so no * write requests running. We should be safe */ - log_init(conf, rdev, false); + ret = log_init(conf, rdev, false); + if (ret) + return ret; + + ret = r5l_start(conf->log); + if (ret) + return ret; + return 0; } if (mddev->recovery_disabled == conf->recovery_disabled) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 120fb593d1da..b2dd4e391f5c 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -1113,15 +1113,15 @@ static struct nvme_id_ns *nvme_identify_ns(struct nvme_ctrl *ctrl, return id; } -static int nvme_set_features(struct nvme_ctrl *dev, unsigned fid, unsigned dword11, - void *buffer, size_t buflen, u32 *result) +static int nvme_features(struct nvme_ctrl *dev, u8 op, unsigned int fid, + unsigned int dword11, void *buffer, size_t buflen, u32 *result) { struct nvme_command c; union nvme_result res; int ret; memset(&c, 0, sizeof(c)); - c.features.opcode = nvme_admin_set_features; + c.features.opcode = op; c.features.fid = cpu_to_le32(fid); c.features.dword11 = cpu_to_le32(dword11); @@ -1132,6 +1132,24 @@ static int nvme_set_features(struct nvme_ctrl *dev, unsigned fid, unsigned dword return ret; } +int nvme_set_features(struct nvme_ctrl *dev, unsigned int fid, + unsigned int dword11, void *buffer, size_t buflen, + u32 *result) +{ + return nvme_features(dev, nvme_admin_set_features, fid, dword11, buffer, + buflen, result); +} +EXPORT_SYMBOL_GPL(nvme_set_features); + +int nvme_get_features(struct nvme_ctrl *dev, unsigned int fid, + unsigned int dword11, void *buffer, size_t buflen, + u32 *result) +{ + return nvme_features(dev, nvme_admin_get_features, fid, dword11, buffer, + buflen, result); +} +EXPORT_SYMBOL_GPL(nvme_get_features); + int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count) { u32 q_count = (*count - 1) | ((*count - 1) << 16); @@ -3318,7 +3336,7 @@ static int nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid) device_add_disk(ctrl->device, ns->disk, nvme_ns_id_attr_groups); nvme_mpath_add_disk(ns, id); - nvme_fault_inject_init(ns); + nvme_fault_inject_init(&ns->fault_inject, ns->disk->disk_name); kfree(id); return 0; @@ -3343,7 +3361,15 @@ static void nvme_ns_remove(struct nvme_ns *ns) if (test_and_set_bit(NVME_NS_REMOVING, &ns->flags)) return; - nvme_fault_inject_fini(ns); + nvme_fault_inject_fini(&ns->fault_inject); + + mutex_lock(&ns->ctrl->subsys->lock); + list_del_rcu(&ns->siblings); + mutex_unlock(&ns->ctrl->subsys->lock); + synchronize_rcu(); /* guarantee not available in head->list */ + nvme_mpath_clear_current_path(ns); + synchronize_srcu(&ns->head->srcu); /* wait for concurrent submissions */ + if (ns->disk && ns->disk->flags & GENHD_FL_UP) { del_gendisk(ns->disk); blk_cleanup_queue(ns->queue); @@ -3351,16 +3377,10 @@ static void nvme_ns_remove(struct nvme_ns *ns) blk_integrity_unregister(ns->disk); } - mutex_lock(&ns->ctrl->subsys->lock); - list_del_rcu(&ns->siblings); - nvme_mpath_clear_current_path(ns); - mutex_unlock(&ns->ctrl->subsys->lock); - down_write(&ns->ctrl->namespaces_rwsem); list_del_init(&ns->list); up_write(&ns->ctrl->namespaces_rwsem); - synchronize_srcu(&ns->head->srcu); nvme_mpath_check_last_path(ns); nvme_put_ns(ns); } @@ -3702,6 +3722,7 @@ EXPORT_SYMBOL_GPL(nvme_start_ctrl); void nvme_uninit_ctrl(struct nvme_ctrl *ctrl) { + nvme_fault_inject_fini(&ctrl->fault_inject); dev_pm_qos_hide_latency_tolerance(ctrl->device); cdev_device_del(&ctrl->cdev, ctrl->device); } @@ -3797,6 +3818,8 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, dev_pm_qos_update_user_latency_tolerance(ctrl->device, min(default_ps_max_latency_us, (unsigned long)S32_MAX)); + nvme_fault_inject_init(&ctrl->fault_inject, dev_name(ctrl->device)); + return 0; out_free_name: kfree_const(ctrl->device->kobj.name); diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c index 5838f7cd53ac..1994d5b42f94 100644 --- a/drivers/nvme/host/fabrics.c +++ b/drivers/nvme/host/fabrics.c @@ -578,7 +578,7 @@ bool __nvmf_check_ready(struct nvme_ctrl *ctrl, struct request *rq, switch (ctrl->state) { case NVME_CTRL_NEW: case NVME_CTRL_CONNECTING: - if (req->cmd->common.opcode == nvme_fabrics_command && + if (nvme_is_fabrics(req->cmd) && req->cmd->fabrics.fctype == nvme_fabrics_type_connect) return true; break; diff --git a/drivers/nvme/host/fault_inject.c b/drivers/nvme/host/fault_inject.c index 4cfd2c9222d4..1352159733b0 100644 --- a/drivers/nvme/host/fault_inject.c +++ b/drivers/nvme/host/fault_inject.c @@ -15,11 +15,10 @@ static DECLARE_FAULT_ATTR(fail_default_attr); static char *fail_request; module_param(fail_request, charp, 0000); -void nvme_fault_inject_init(struct nvme_ns *ns) +void nvme_fault_inject_init(struct nvme_fault_inject *fault_inj, + const char *dev_name) { struct dentry *dir, *parent; - char *name = ns->disk->disk_name; - struct nvme_fault_inject *fault_inj = &ns->fault_inject; struct fault_attr *attr = &fault_inj->attr; /* set default fault injection attribute */ @@ -27,20 +26,20 @@ void nvme_fault_inject_init(struct nvme_ns *ns) setup_fault_attr(&fail_default_attr, fail_request); /* create debugfs directory and attribute */ - parent = debugfs_create_dir(name, NULL); + parent = debugfs_create_dir(dev_name, NULL); if (!parent) { - pr_warn("%s: failed to create debugfs directory\n", name); + pr_warn("%s: failed to create debugfs directory\n", dev_name); return; } *attr = fail_default_attr; dir = fault_create_debugfs_attr("fault_inject", parent, attr); if (IS_ERR(dir)) { - pr_warn("%s: failed to create debugfs attr\n", name); + pr_warn("%s: failed to create debugfs attr\n", dev_name); debugfs_remove_recursive(parent); return; } - ns->fault_inject.parent = parent; + fault_inj->parent = parent; /* create debugfs for status code and dont_retry */ fault_inj->status = NVME_SC_INVALID_OPCODE; @@ -49,29 +48,33 @@ void nvme_fault_inject_init(struct nvme_ns *ns) debugfs_create_bool("dont_retry", 0600, dir, &fault_inj->dont_retry); } -void nvme_fault_inject_fini(struct nvme_ns *ns) +void nvme_fault_inject_fini(struct nvme_fault_inject *fault_inject) { /* remove debugfs directories */ - debugfs_remove_recursive(ns->fault_inject.parent); + debugfs_remove_recursive(fault_inject->parent); } void nvme_should_fail(struct request *req) { struct gendisk *disk = req->rq_disk; - struct nvme_ns *ns = NULL; + struct nvme_fault_inject *fault_inject = NULL; u16 status; - /* - * make sure this request is coming from a valid namespace - */ - if (!disk) - return; + if (disk) { + struct nvme_ns *ns = disk->private_data; + + if (ns) + fault_inject = &ns->fault_inject; + else + WARN_ONCE(1, "No namespace found for request\n"); + } else { + fault_inject = &nvme_req(req)->ctrl->fault_inject; + } - ns = disk->private_data; - if (ns && should_fail(&ns->fault_inject.attr, 1)) { + if (fault_inject && should_fail(&fault_inject->attr, 1)) { /* inject status code and DNR bit */ - status = ns->fault_inject.status; - if (ns->fault_inject.dont_retry) + status = fault_inject->status; + if (fault_inject->dont_retry) status |= NVME_SC_DNR; nvme_req(req)->status = status; } diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index dd8169bbf0d2..9b497d785ed7 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -2607,6 +2607,12 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) if (nvme_fc_ctlr_active_on_rport(ctrl)) return -ENOTUNIQ; + dev_info(ctrl->ctrl.device, + "NVME-FC{%d}: create association : host wwpn 0x%016llx " + " rport wwpn 0x%016llx: NQN \"%s\"\n", + ctrl->cnum, ctrl->lport->localport.port_name, + ctrl->rport->remoteport.port_name, ctrl->ctrl.opts->subsysnqn); + /* * Create the admin queue */ diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c index 4f20a10b39d3..ba009d4c9dfa 100644 --- a/drivers/nvme/host/lightnvm.c +++ b/drivers/nvme/host/lightnvm.c @@ -660,7 +660,7 @@ static struct request *nvme_nvm_alloc_request(struct request_queue *q, rq->cmd_flags &= ~REQ_FAILFAST_DRIVER; if (rqd->bio) - blk_init_request_from_bio(rq, rqd->bio); + blk_rq_append_bio(rq, &rqd->bio); else rq->ioprio = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, IOPRIO_NORM); diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h index 55553d293a98..ea45d7d393ad 100644 --- a/drivers/nvme/host/nvme.h +++ b/drivers/nvme/host/nvme.h @@ -146,6 +146,15 @@ enum nvme_ctrl_state { NVME_CTRL_DEAD, }; +struct nvme_fault_inject { +#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS + struct fault_attr attr; + struct dentry *parent; + bool dont_retry; /* DNR, do not retry */ + u16 status; /* status code */ +#endif +}; + struct nvme_ctrl { bool comp_seen; enum nvme_ctrl_state state; @@ -247,6 +256,8 @@ struct nvme_ctrl { struct page *discard_page; unsigned long discard_page_busy; + + struct nvme_fault_inject fault_inject; }; enum nvme_iopolicy { @@ -313,15 +324,6 @@ struct nvme_ns_head { #endif }; -#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS -struct nvme_fault_inject { - struct fault_attr attr; - struct dentry *parent; - bool dont_retry; /* DNR, do not retry */ - u16 status; /* status code */ -}; -#endif - struct nvme_ns { struct list_head list; @@ -349,9 +351,7 @@ struct nvme_ns { #define NVME_NS_ANA_PENDING 2 u16 noiob; -#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS struct nvme_fault_inject fault_inject; -#endif }; @@ -372,12 +372,18 @@ struct nvme_ctrl_ops { }; #ifdef CONFIG_FAULT_INJECTION_DEBUG_FS -void nvme_fault_inject_init(struct nvme_ns *ns); -void nvme_fault_inject_fini(struct nvme_ns *ns); +void nvme_fault_inject_init(struct nvme_fault_inject *fault_inj, + const char *dev_name); +void nvme_fault_inject_fini(struct nvme_fault_inject *fault_inject); void nvme_should_fail(struct request *req); #else -static inline void nvme_fault_inject_init(struct nvme_ns *ns) {} -static inline void nvme_fault_inject_fini(struct nvme_ns *ns) {} +static inline void nvme_fault_inject_init(struct nvme_fault_inject *fault_inj, + const char *dev_name) +{ +} +static inline void nvme_fault_inject_fini(struct nvme_fault_inject *fault_inj) +{ +} static inline void nvme_should_fail(struct request *req) {} #endif @@ -459,6 +465,12 @@ int __nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd, union nvme_result *result, void *buffer, unsigned bufflen, unsigned timeout, int qid, int at_head, blk_mq_req_flags_t flags, bool poll); +int nvme_set_features(struct nvme_ctrl *dev, unsigned int fid, + unsigned int dword11, void *buffer, size_t buflen, + u32 *result); +int nvme_get_features(struct nvme_ctrl *dev, unsigned int fid, + unsigned int dword11, void *buffer, size_t buflen, + u32 *result); int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count); void nvme_stop_keep_alive(struct nvme_ctrl *ctrl); int nvme_reset_ctrl(struct nvme_ctrl *ctrl); diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 524d6bd6d095..189352081994 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -18,6 +18,7 @@ #include <linux/mutex.h> #include <linux/once.h> #include <linux/pci.h> +#include <linux/suspend.h> #include <linux/t10-pi.h> #include <linux/types.h> #include <linux/io-64-nonatomic-lo-hi.h> @@ -67,20 +68,14 @@ static int io_queue_depth = 1024; module_param_cb(io_queue_depth, &io_queue_depth_ops, &io_queue_depth, 0644); MODULE_PARM_DESC(io_queue_depth, "set io queue depth, should >= 2"); -static int queue_count_set(const char *val, const struct kernel_param *kp); -static const struct kernel_param_ops queue_count_ops = { - .set = queue_count_set, - .get = param_get_int, -}; - static int write_queues; -module_param_cb(write_queues, &queue_count_ops, &write_queues, 0644); +module_param(write_queues, int, 0644); MODULE_PARM_DESC(write_queues, "Number of queues to use for writes. If not set, reads and writes " "will share a queue set."); -static int poll_queues = 0; -module_param_cb(poll_queues, &queue_count_ops, &poll_queues, 0644); +static int poll_queues; +module_param(poll_queues, int, 0644); MODULE_PARM_DESC(poll_queues, "Number of queues to use for polled IO."); struct nvme_dev; @@ -116,6 +111,7 @@ struct nvme_dev { u32 cmbsz; u32 cmbloc; struct nvme_ctrl ctrl; + u32 last_ps; mempool_t *iod_mempool; @@ -144,19 +140,6 @@ static int io_queue_depth_set(const char *val, const struct kernel_param *kp) return param_set_int(val, kp); } -static int queue_count_set(const char *val, const struct kernel_param *kp) -{ - int n, ret; - - ret = kstrtoint(val, 10, &n); - if (ret) - return ret; - if (n > num_possible_cpus()) - n = num_possible_cpus(); - - return param_set_int(val, kp); -} - static inline unsigned int sq_idx(unsigned int qid, u32 stride) { return qid * 2 * stride; @@ -2068,6 +2051,7 @@ static int nvme_setup_irqs(struct nvme_dev *dev, unsigned int nr_io_queues) .priv = dev, }; unsigned int irq_queues, this_p_queues; + unsigned int nr_cpus = num_possible_cpus(); /* * Poll queues don't need interrupts, but we need at least one IO @@ -2078,7 +2062,10 @@ static int nvme_setup_irqs(struct nvme_dev *dev, unsigned int nr_io_queues) this_p_queues = nr_io_queues - 1; irq_queues = 1; } else { - irq_queues = nr_io_queues - this_p_queues + 1; + if (nr_cpus < nr_io_queues - this_p_queues) + irq_queues = nr_cpus + 1; + else + irq_queues = nr_io_queues - this_p_queues + 1; } dev->io_queues[HCTX_TYPE_POLL] = this_p_queues; @@ -2464,10 +2451,8 @@ static void nvme_pci_free_ctrl(struct nvme_ctrl *ctrl) kfree(dev); } -static void nvme_remove_dead_ctrl(struct nvme_dev *dev, int status) +static void nvme_remove_dead_ctrl(struct nvme_dev *dev) { - dev_warn(dev->ctrl.device, "Removing after probe failure status: %d\n", status); - nvme_get_ctrl(&dev->ctrl); nvme_dev_disable(dev, false); nvme_kill_queues(&dev->ctrl); @@ -2480,11 +2465,13 @@ static void nvme_reset_work(struct work_struct *work) struct nvme_dev *dev = container_of(work, struct nvme_dev, ctrl.reset_work); bool was_suspend = !!(dev->ctrl.ctrl_config & NVME_CC_SHN_NORMAL); - int result = -ENODEV; + int result; enum nvme_ctrl_state new_state = NVME_CTRL_LIVE; - if (WARN_ON(dev->ctrl.state != NVME_CTRL_RESETTING)) + if (WARN_ON(dev->ctrl.state != NVME_CTRL_RESETTING)) { + result = -ENODEV; goto out; + } /* * If we're called to reset a live controller first shut it down before @@ -2528,6 +2515,7 @@ static void nvme_reset_work(struct work_struct *work) if (!nvme_change_ctrl_state(&dev->ctrl, NVME_CTRL_CONNECTING)) { dev_warn(dev->ctrl.device, "failed to mark controller CONNECTING\n"); + result = -EBUSY; goto out; } @@ -2588,6 +2576,7 @@ static void nvme_reset_work(struct work_struct *work) if (!nvme_change_ctrl_state(&dev->ctrl, new_state)) { dev_warn(dev->ctrl.device, "failed to mark controller state %d\n", new_state); + result = -ENODEV; goto out; } @@ -2597,7 +2586,10 @@ static void nvme_reset_work(struct work_struct *work) out_unlock: mutex_unlock(&dev->shutdown_lock); out: - nvme_remove_dead_ctrl(dev, result); + if (result) + dev_warn(dev->ctrl.device, + "Removing after probe failure status: %d\n", result); + nvme_remove_dead_ctrl(dev); } static void nvme_remove_dead_ctrl_work(struct work_struct *work) @@ -2835,16 +2827,94 @@ static void nvme_remove(struct pci_dev *pdev) } #ifdef CONFIG_PM_SLEEP +static int nvme_get_power_state(struct nvme_ctrl *ctrl, u32 *ps) +{ + return nvme_get_features(ctrl, NVME_FEAT_POWER_MGMT, 0, NULL, 0, ps); +} + +static int nvme_set_power_state(struct nvme_ctrl *ctrl, u32 ps) +{ + return nvme_set_features(ctrl, NVME_FEAT_POWER_MGMT, ps, NULL, 0, NULL); +} + +static int nvme_resume(struct device *dev) +{ + struct nvme_dev *ndev = pci_get_drvdata(to_pci_dev(dev)); + struct nvme_ctrl *ctrl = &ndev->ctrl; + + if (pm_resume_via_firmware() || !ctrl->npss || + nvme_set_power_state(ctrl, ndev->last_ps) != 0) + nvme_reset_ctrl(ctrl); + return 0; +} + static int nvme_suspend(struct device *dev) { struct pci_dev *pdev = to_pci_dev(dev); struct nvme_dev *ndev = pci_get_drvdata(pdev); + struct nvme_ctrl *ctrl = &ndev->ctrl; + int ret = -EBUSY; + + /* + * The platform does not remove power for a kernel managed suspend so + * use host managed nvme power settings for lowest idle power if + * possible. This should have quicker resume latency than a full device + * shutdown. But if the firmware is involved after the suspend or the + * device does not support any non-default power states, shut down the + * device fully. + */ + if (pm_suspend_via_firmware() || !ctrl->npss) { + nvme_dev_disable(ndev, true); + return 0; + } + + nvme_start_freeze(ctrl); + nvme_wait_freeze(ctrl); + nvme_sync_queues(ctrl); + + if (ctrl->state != NVME_CTRL_LIVE && + ctrl->state != NVME_CTRL_ADMIN_ONLY) + goto unfreeze; + + ndev->last_ps = 0; + ret = nvme_get_power_state(ctrl, &ndev->last_ps); + if (ret < 0) + goto unfreeze; + + ret = nvme_set_power_state(ctrl, ctrl->npss); + if (ret < 0) + goto unfreeze; + + if (ret) { + /* + * Clearing npss forces a controller reset on resume. The + * correct value will be resdicovered then. + */ + nvme_dev_disable(ndev, true); + ctrl->npss = 0; + ret = 0; + goto unfreeze; + } + /* + * A saved state prevents pci pm from generically controlling the + * device's power. If we're using protocol specific settings, we don't + * want pci interfering. + */ + pci_save_state(pdev); +unfreeze: + nvme_unfreeze(ctrl); + return ret; +} + +static int nvme_simple_suspend(struct device *dev) +{ + struct nvme_dev *ndev = pci_get_drvdata(to_pci_dev(dev)); nvme_dev_disable(ndev, true); return 0; } -static int nvme_resume(struct device *dev) +static int nvme_simple_resume(struct device *dev) { struct pci_dev *pdev = to_pci_dev(dev); struct nvme_dev *ndev = pci_get_drvdata(pdev); @@ -2852,9 +2922,16 @@ static int nvme_resume(struct device *dev) nvme_reset_ctrl(&ndev->ctrl); return 0; } -#endif -static SIMPLE_DEV_PM_OPS(nvme_dev_pm_ops, nvme_suspend, nvme_resume); +const struct dev_pm_ops nvme_dev_pm_ops = { + .suspend = nvme_suspend, + .resume = nvme_resume, + .freeze = nvme_simple_suspend, + .thaw = nvme_simple_resume, + .poweroff = nvme_simple_suspend, + .restore = nvme_simple_resume, +}; +#endif /* CONFIG_PM_SLEEP */ static pci_ers_result_t nvme_error_detected(struct pci_dev *pdev, pci_channel_state_t state) @@ -2959,9 +3036,11 @@ static struct pci_driver nvme_driver = { .probe = nvme_probe, .remove = nvme_remove, .shutdown = nvme_shutdown, +#ifdef CONFIG_PM_SLEEP .driver = { .pm = &nvme_dev_pm_ops, }, +#endif .sriov_configure = pci_sriov_configure_simple, .err_handler = &nvme_err_handler, }; diff --git a/drivers/nvme/host/trace.c b/drivers/nvme/host/trace.c index 5f24ea7a28eb..f01ad0fd60bb 100644 --- a/drivers/nvme/host/trace.c +++ b/drivers/nvme/host/trace.c @@ -135,6 +135,69 @@ const char *nvme_trace_parse_nvm_cmd(struct trace_seq *p, } } +static const char *nvme_trace_fabrics_property_set(struct trace_seq *p, u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + u8 attrib = spc[0]; + u32 ofst = get_unaligned_le32(spc + 4); + u64 value = get_unaligned_le64(spc + 8); + + trace_seq_printf(p, "attrib=%u, ofst=0x%x, value=0x%llx", + attrib, ofst, value); + trace_seq_putc(p, 0); + return ret; +} + +static const char *nvme_trace_fabrics_connect(struct trace_seq *p, u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + u16 recfmt = get_unaligned_le16(spc); + u16 qid = get_unaligned_le16(spc + 2); + u16 sqsize = get_unaligned_le16(spc + 4); + u8 cattr = spc[6]; + u32 kato = get_unaligned_le32(spc + 8); + + trace_seq_printf(p, "recfmt=%u, qid=%u, sqsize=%u, cattr=%u, kato=%u", + recfmt, qid, sqsize, cattr, kato); + trace_seq_putc(p, 0); + return ret; +} + +static const char *nvme_trace_fabrics_property_get(struct trace_seq *p, u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + u8 attrib = spc[0]; + u32 ofst = get_unaligned_le32(spc + 4); + + trace_seq_printf(p, "attrib=%u, ofst=0x%x", attrib, ofst); + trace_seq_putc(p, 0); + return ret; +} + +static const char *nvme_trace_fabrics_common(struct trace_seq *p, u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + + trace_seq_printf(p, "spcecific=%*ph", 24, spc); + trace_seq_putc(p, 0); + return ret; +} + +const char *nvme_trace_parse_fabrics_cmd(struct trace_seq *p, + u8 fctype, u8 *spc) +{ + switch (fctype) { + case nvme_fabrics_type_property_set: + return nvme_trace_fabrics_property_set(p, spc); + case nvme_fabrics_type_connect: + return nvme_trace_fabrics_connect(p, spc); + case nvme_fabrics_type_property_get: + return nvme_trace_fabrics_property_get(p, spc); + default: + return nvme_trace_fabrics_common(p, spc); + } +} + const char *nvme_trace_disk_name(struct trace_seq *p, char *name) { const char *ret = trace_seq_buffer_ptr(p); @@ -145,6 +208,5 @@ const char *nvme_trace_disk_name(struct trace_seq *p, char *name) return ret; } -EXPORT_SYMBOL_GPL(nvme_trace_disk_name); EXPORT_TRACEPOINT_SYMBOL_GPL(nvme_sq); diff --git a/drivers/nvme/host/trace.h b/drivers/nvme/host/trace.h index e71502d141ed..daaf700eae79 100644 --- a/drivers/nvme/host/trace.h +++ b/drivers/nvme/host/trace.h @@ -16,59 +16,19 @@ #include "nvme.h" -#define nvme_admin_opcode_name(opcode) { opcode, #opcode } -#define show_admin_opcode_name(val) \ - __print_symbolic(val, \ - nvme_admin_opcode_name(nvme_admin_delete_sq), \ - nvme_admin_opcode_name(nvme_admin_create_sq), \ - nvme_admin_opcode_name(nvme_admin_get_log_page), \ - nvme_admin_opcode_name(nvme_admin_delete_cq), \ - nvme_admin_opcode_name(nvme_admin_create_cq), \ - nvme_admin_opcode_name(nvme_admin_identify), \ - nvme_admin_opcode_name(nvme_admin_abort_cmd), \ - nvme_admin_opcode_name(nvme_admin_set_features), \ - nvme_admin_opcode_name(nvme_admin_get_features), \ - nvme_admin_opcode_name(nvme_admin_async_event), \ - nvme_admin_opcode_name(nvme_admin_ns_mgmt), \ - nvme_admin_opcode_name(nvme_admin_activate_fw), \ - nvme_admin_opcode_name(nvme_admin_download_fw), \ - nvme_admin_opcode_name(nvme_admin_ns_attach), \ - nvme_admin_opcode_name(nvme_admin_keep_alive), \ - nvme_admin_opcode_name(nvme_admin_directive_send), \ - nvme_admin_opcode_name(nvme_admin_directive_recv), \ - nvme_admin_opcode_name(nvme_admin_dbbuf), \ - nvme_admin_opcode_name(nvme_admin_format_nvm), \ - nvme_admin_opcode_name(nvme_admin_security_send), \ - nvme_admin_opcode_name(nvme_admin_security_recv), \ - nvme_admin_opcode_name(nvme_admin_sanitize_nvm)) - -#define nvme_opcode_name(opcode) { opcode, #opcode } -#define show_nvm_opcode_name(val) \ - __print_symbolic(val, \ - nvme_opcode_name(nvme_cmd_flush), \ - nvme_opcode_name(nvme_cmd_write), \ - nvme_opcode_name(nvme_cmd_read), \ - nvme_opcode_name(nvme_cmd_write_uncor), \ - nvme_opcode_name(nvme_cmd_compare), \ - nvme_opcode_name(nvme_cmd_write_zeroes), \ - nvme_opcode_name(nvme_cmd_dsm), \ - nvme_opcode_name(nvme_cmd_resv_register), \ - nvme_opcode_name(nvme_cmd_resv_report), \ - nvme_opcode_name(nvme_cmd_resv_acquire), \ - nvme_opcode_name(nvme_cmd_resv_release)) - -#define show_opcode_name(qid, opcode) \ - (qid ? show_nvm_opcode_name(opcode) : show_admin_opcode_name(opcode)) - const char *nvme_trace_parse_admin_cmd(struct trace_seq *p, u8 opcode, u8 *cdw10); const char *nvme_trace_parse_nvm_cmd(struct trace_seq *p, u8 opcode, u8 *cdw10); +const char *nvme_trace_parse_fabrics_cmd(struct trace_seq *p, u8 fctype, + u8 *spc); -#define parse_nvme_cmd(qid, opcode, cdw10) \ - (qid ? \ - nvme_trace_parse_nvm_cmd(p, opcode, cdw10) : \ - nvme_trace_parse_admin_cmd(p, opcode, cdw10)) +#define parse_nvme_cmd(qid, opcode, fctype, cdw10) \ + ((opcode) == nvme_fabrics_command ? \ + nvme_trace_parse_fabrics_cmd(p, fctype, cdw10) : \ + ((qid) ? \ + nvme_trace_parse_nvm_cmd(p, opcode, cdw10) : \ + nvme_trace_parse_admin_cmd(p, opcode, cdw10))) const char *nvme_trace_disk_name(struct trace_seq *p, char *name); #define __print_disk_name(name) \ @@ -93,6 +53,7 @@ TRACE_EVENT(nvme_setup_cmd, __field(int, qid) __field(u8, opcode) __field(u8, flags) + __field(u8, fctype) __field(u16, cid) __field(u32, nsid) __field(u64, metadata) @@ -106,6 +67,7 @@ TRACE_EVENT(nvme_setup_cmd, __entry->cid = cmd->common.command_id; __entry->nsid = le32_to_cpu(cmd->common.nsid); __entry->metadata = le64_to_cpu(cmd->common.metadata); + __entry->fctype = cmd->fabrics.fctype; __assign_disk_name(__entry->disk, req->rq_disk); memcpy(__entry->cdw10, &cmd->common.cdw10, sizeof(__entry->cdw10)); @@ -114,8 +76,10 @@ TRACE_EVENT(nvme_setup_cmd, __entry->ctrl_id, __print_disk_name(__entry->disk), __entry->qid, __entry->cid, __entry->nsid, __entry->flags, __entry->metadata, - show_opcode_name(__entry->qid, __entry->opcode), - parse_nvme_cmd(__entry->qid, __entry->opcode, __entry->cdw10)) + show_opcode_name(__entry->qid, __entry->opcode, + __entry->fctype), + parse_nvme_cmd(__entry->qid, __entry->opcode, + __entry->fctype, __entry->cdw10)) ); TRACE_EVENT(nvme_complete_rq, @@ -141,7 +105,7 @@ TRACE_EVENT(nvme_complete_rq, __entry->status = nvme_req(req)->status; __assign_disk_name(__entry->disk, req->rq_disk); ), - TP_printk("nvme%d: %sqid=%d, cmdid=%u, res=%llu, retries=%u, flags=0x%x, status=%u", + TP_printk("nvme%d: %sqid=%d, cmdid=%u, res=%#llx, retries=%u, flags=0x%x, status=%#x", __entry->ctrl_id, __print_disk_name(__entry->disk), __entry->qid, __entry->cid, __entry->result, __entry->retries, __entry->flags, __entry->status) diff --git a/drivers/nvme/target/Makefile b/drivers/nvme/target/Makefile index 8c3ad0fb6860..2b33836f3d3e 100644 --- a/drivers/nvme/target/Makefile +++ b/drivers/nvme/target/Makefile @@ -1,5 +1,7 @@ # SPDX-License-Identifier: GPL-2.0 +ccflags-y += -I$(src) + obj-$(CONFIG_NVME_TARGET) += nvmet.o obj-$(CONFIG_NVME_TARGET_LOOP) += nvme-loop.o obj-$(CONFIG_NVME_TARGET_RDMA) += nvmet-rdma.o @@ -14,3 +16,4 @@ nvmet-rdma-y += rdma.o nvmet-fc-y += fc.o nvme-fcloop-y += fcloop.o nvmet-tcp-y += tcp.o +nvmet-$(CONFIG_TRACING) += trace.o diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index 7734a6acff85..dad0243c7c96 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -10,6 +10,9 @@ #include <linux/pci-p2pdma.h> #include <linux/scatterlist.h> +#define CREATE_TRACE_POINTS +#include "trace.h" + #include "nvmet.h" struct workqueue_struct *buffered_io_wq; @@ -311,6 +314,7 @@ int nvmet_enable_port(struct nvmet_port *port) port->inline_data_size = 0; port->enabled = true; + port->tr_ops = ops; return 0; } @@ -321,6 +325,7 @@ void nvmet_disable_port(struct nvmet_port *port) lockdep_assert_held(&nvmet_config_sem); port->enabled = false; + port->tr_ops = NULL; ops = nvmet_transports[port->disc_addr.trtype]; ops->remove_port(port); @@ -689,6 +694,9 @@ static void __nvmet_req_complete(struct nvmet_req *req, u16 status) if (unlikely(status)) nvmet_set_error(req, status); + + trace_nvmet_req_complete(req); + if (req->ns) nvmet_put_namespace(req->ns); req->ops->queue_response(req); @@ -848,6 +856,8 @@ bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq, req->error_loc = NVMET_NO_ERROR_LOC; req->error_slba = 0; + trace_nvmet_req_init(req, req->cmd); + /* no support for fused commands yet */ if (unlikely(flags & (NVME_CMD_FUSE_FIRST | NVME_CMD_FUSE_SECOND))) { req->error_loc = offsetof(struct nvme_common_command, flags); @@ -871,7 +881,7 @@ bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq, status = nvmet_parse_connect_cmd(req); else if (likely(req->sq->qid != 0)) status = nvmet_parse_io_cmd(req); - else if (req->cmd->common.opcode == nvme_fabrics_command) + else if (nvme_is_fabrics(req->cmd)) status = nvmet_parse_fabrics_cmd(req); else if (req->sq->ctrl->subsys->type == NVME_NQN_DISC) status = nvmet_parse_discovery_cmd(req); diff --git a/drivers/nvme/target/discovery.c b/drivers/nvme/target/discovery.c index 5baf269f3f8a..8efca26b4776 100644 --- a/drivers/nvme/target/discovery.c +++ b/drivers/nvme/target/discovery.c @@ -41,6 +41,10 @@ void nvmet_port_disc_changed(struct nvmet_port *port, __nvmet_disc_changed(port, ctrl); } mutex_unlock(&nvmet_disc_subsys->lock); + + /* If transport can signal change, notify transport */ + if (port->tr_ops && port->tr_ops->discovery_chg) + port->tr_ops->discovery_chg(port); } static void __nvmet_subsys_disc_changed(struct nvmet_port *port, diff --git a/drivers/nvme/target/fabrics-cmd.c b/drivers/nvme/target/fabrics-cmd.c index 3b9f79aba98f..d16b55ffe79f 100644 --- a/drivers/nvme/target/fabrics-cmd.c +++ b/drivers/nvme/target/fabrics-cmd.c @@ -268,7 +268,7 @@ u16 nvmet_parse_connect_cmd(struct nvmet_req *req) { struct nvme_command *cmd = req->cmd; - if (cmd->common.opcode != nvme_fabrics_command) { + if (!nvme_is_fabrics(cmd)) { pr_err("invalid command 0x%x on unconnected queue.\n", cmd->fabrics.opcode); req->error_loc = offsetof(struct nvme_common_command, opcode); diff --git a/drivers/nvme/target/fc.c b/drivers/nvme/target/fc.c index 508661af0f50..ce8d819f86cc 100644 --- a/drivers/nvme/target/fc.c +++ b/drivers/nvme/target/fc.c @@ -1806,7 +1806,7 @@ nvmet_fc_prep_fcp_rsp(struct nvmet_fc_tgtport *tgtport, */ rspcnt = atomic_inc_return(&fod->queue->zrspcnt); if (!(rspcnt % fod->queue->ersp_ratio) || - sqe->opcode == nvme_fabrics_command || + nvme_is_fabrics((struct nvme_command *) sqe) || xfr_length != fod->req.transfer_len || (le16_to_cpu(cqe->status) & 0xFFFE) || cqewd[0] || cqewd[1] || (sqe->flags & (NVME_CMD_FUSE_FIRST | NVME_CMD_FUSE_SECOND)) || @@ -2549,6 +2549,16 @@ nvmet_fc_remove_port(struct nvmet_port *port) kfree(pe); } +static void +nvmet_fc_discovery_chg(struct nvmet_port *port) +{ + struct nvmet_fc_port_entry *pe = port->priv; + struct nvmet_fc_tgtport *tgtport = pe->tgtport; + + if (tgtport && tgtport->ops->discovery_event) + tgtport->ops->discovery_event(&tgtport->fc_target_port); +} + static const struct nvmet_fabrics_ops nvmet_fc_tgt_fcp_ops = { .owner = THIS_MODULE, .type = NVMF_TRTYPE_FC, @@ -2557,6 +2567,7 @@ static const struct nvmet_fabrics_ops nvmet_fc_tgt_fcp_ops = { .remove_port = nvmet_fc_remove_port, .queue_response = nvmet_fc_fcp_nvme_cmd_done, .delete_ctrl = nvmet_fc_delete_ctrl, + .discovery_chg = nvmet_fc_discovery_chg, }; static int __init nvmet_fc_init_module(void) diff --git a/drivers/nvme/target/fcloop.c b/drivers/nvme/target/fcloop.c index 381b5a90c48b..b8c1cc54a0db 100644 --- a/drivers/nvme/target/fcloop.c +++ b/drivers/nvme/target/fcloop.c @@ -231,6 +231,11 @@ struct fcloop_lsreq { int status; }; +struct fcloop_rscn { + struct fcloop_tport *tport; + struct work_struct work; +}; + enum { INI_IO_START = 0, INI_IO_ACTIVE = 1, @@ -348,6 +353,37 @@ fcloop_xmt_ls_rsp(struct nvmet_fc_target_port *tport, return 0; } +/* + * Simulate reception of RSCN and converting it to a initiator transport + * call to rescan a remote port. + */ +static void +fcloop_tgt_rscn_work(struct work_struct *work) +{ + struct fcloop_rscn *tgt_rscn = + container_of(work, struct fcloop_rscn, work); + struct fcloop_tport *tport = tgt_rscn->tport; + + if (tport->remoteport) + nvme_fc_rescan_remoteport(tport->remoteport); + kfree(tgt_rscn); +} + +static void +fcloop_tgt_discovery_evt(struct nvmet_fc_target_port *tgtport) +{ + struct fcloop_rscn *tgt_rscn; + + tgt_rscn = kzalloc(sizeof(*tgt_rscn), GFP_KERNEL); + if (!tgt_rscn) + return; + + tgt_rscn->tport = tgtport->private; + INIT_WORK(&tgt_rscn->work, fcloop_tgt_rscn_work); + + schedule_work(&tgt_rscn->work); +} + static void fcloop_tfcp_req_free(struct kref *ref) { @@ -839,6 +875,7 @@ static struct nvmet_fc_target_template tgttemplate = { .fcp_op = fcloop_fcp_op, .fcp_abort = fcloop_tgt_fcp_abort, .fcp_req_release = fcloop_fcp_req_release, + .discovery_event = fcloop_tgt_discovery_evt, .max_hw_queues = FCLOOP_HW_QUEUES, .max_sgl_segments = FCLOOP_SGL_SEGS, .max_dif_sgl_segments = FCLOOP_SGL_SEGS, diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h index c25d88fc9dec..dc270944bb25 100644 --- a/drivers/nvme/target/nvmet.h +++ b/drivers/nvme/target/nvmet.h @@ -140,6 +140,7 @@ struct nvmet_port { void *priv; bool enabled; int inline_data_size; + const struct nvmet_fabrics_ops *tr_ops; }; static inline struct nvmet_port *to_nvmet_port(struct config_item *item) @@ -277,6 +278,7 @@ struct nvmet_fabrics_ops { void (*disc_traddr)(struct nvmet_req *req, struct nvmet_port *port, char *traddr); u16 (*install_queue)(struct nvmet_sq *nvme_sq); + void (*discovery_chg)(struct nvmet_port *port); }; #define NVMET_MAX_INLINE_BIOVEC 8 diff --git a/drivers/nvme/target/trace.c b/drivers/nvme/target/trace.c new file mode 100644 index 000000000000..cdcdd14c6408 --- /dev/null +++ b/drivers/nvme/target/trace.c @@ -0,0 +1,201 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * NVM Express target device driver tracepoints + * Copyright (c) 2018 Johannes Thumshirn, SUSE Linux GmbH + */ + +#include <asm/unaligned.h> +#include "trace.h" + +static const char *nvmet_trace_admin_identify(struct trace_seq *p, u8 *cdw10) +{ + const char *ret = trace_seq_buffer_ptr(p); + u8 cns = cdw10[0]; + u16 ctrlid = get_unaligned_le16(cdw10 + 2); + + trace_seq_printf(p, "cns=%u, ctrlid=%u", cns, ctrlid); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvmet_trace_admin_get_features(struct trace_seq *p, + u8 *cdw10) +{ + const char *ret = trace_seq_buffer_ptr(p); + u8 fid = cdw10[0]; + u8 sel = cdw10[1] & 0x7; + u32 cdw11 = get_unaligned_le32(cdw10 + 4); + + trace_seq_printf(p, "fid=0x%x sel=0x%x cdw11=0x%x", fid, sel, cdw11); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvmet_trace_read_write(struct trace_seq *p, u8 *cdw10) +{ + const char *ret = trace_seq_buffer_ptr(p); + u64 slba = get_unaligned_le64(cdw10); + u16 length = get_unaligned_le16(cdw10 + 8); + u16 control = get_unaligned_le16(cdw10 + 10); + u32 dsmgmt = get_unaligned_le32(cdw10 + 12); + u32 reftag = get_unaligned_le32(cdw10 + 16); + + trace_seq_printf(p, + "slba=%llu, len=%u, ctrl=0x%x, dsmgmt=%u, reftag=%u", + slba, length, control, dsmgmt, reftag); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvmet_trace_dsm(struct trace_seq *p, u8 *cdw10) +{ + const char *ret = trace_seq_buffer_ptr(p); + + trace_seq_printf(p, "nr=%u, attributes=%u", + get_unaligned_le32(cdw10), + get_unaligned_le32(cdw10 + 4)); + trace_seq_putc(p, 0); + + return ret; +} + +static const char *nvmet_trace_common(struct trace_seq *p, u8 *cdw10) +{ + const char *ret = trace_seq_buffer_ptr(p); + + trace_seq_printf(p, "cdw10=%*ph", 24, cdw10); + trace_seq_putc(p, 0); + + return ret; +} + +const char *nvmet_trace_parse_admin_cmd(struct trace_seq *p, + u8 opcode, u8 *cdw10) +{ + switch (opcode) { + case nvme_admin_identify: + return nvmet_trace_admin_identify(p, cdw10); + case nvme_admin_get_features: + return nvmet_trace_admin_get_features(p, cdw10); + default: + return nvmet_trace_common(p, cdw10); + } +} + +const char *nvmet_trace_parse_nvm_cmd(struct trace_seq *p, + u8 opcode, u8 *cdw10) +{ + switch (opcode) { + case nvme_cmd_read: + case nvme_cmd_write: + case nvme_cmd_write_zeroes: + return nvmet_trace_read_write(p, cdw10); + case nvme_cmd_dsm: + return nvmet_trace_dsm(p, cdw10); + default: + return nvmet_trace_common(p, cdw10); + } +} + +static const char *nvmet_trace_fabrics_property_set(struct trace_seq *p, + u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + u8 attrib = spc[0]; + u32 ofst = get_unaligned_le32(spc + 4); + u64 value = get_unaligned_le64(spc + 8); + + trace_seq_printf(p, "attrib=%u, ofst=0x%x, value=0x%llx", + attrib, ofst, value); + trace_seq_putc(p, 0); + return ret; +} + +static const char *nvmet_trace_fabrics_connect(struct trace_seq *p, + u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + u16 recfmt = get_unaligned_le16(spc); + u16 qid = get_unaligned_le16(spc + 2); + u16 sqsize = get_unaligned_le16(spc + 4); + u8 cattr = spc[6]; + u32 kato = get_unaligned_le32(spc + 8); + + trace_seq_printf(p, "recfmt=%u, qid=%u, sqsize=%u, cattr=%u, kato=%u", + recfmt, qid, sqsize, cattr, kato); + trace_seq_putc(p, 0); + return ret; +} + +static const char *nvmet_trace_fabrics_property_get(struct trace_seq *p, + u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + u8 attrib = spc[0]; + u32 ofst = get_unaligned_le32(spc + 4); + + trace_seq_printf(p, "attrib=%u, ofst=0x%x", attrib, ofst); + trace_seq_putc(p, 0); + return ret; +} + +static const char *nvmet_trace_fabrics_common(struct trace_seq *p, u8 *spc) +{ + const char *ret = trace_seq_buffer_ptr(p); + + trace_seq_printf(p, "spcecific=%*ph", 24, spc); + trace_seq_putc(p, 0); + return ret; +} + +const char *nvmet_trace_parse_fabrics_cmd(struct trace_seq *p, + u8 fctype, u8 *spc) +{ + switch (fctype) { + case nvme_fabrics_type_property_set: + return nvmet_trace_fabrics_property_set(p, spc); + case nvme_fabrics_type_connect: + return nvmet_trace_fabrics_connect(p, spc); + case nvme_fabrics_type_property_get: + return nvmet_trace_fabrics_property_get(p, spc); + default: + return nvmet_trace_fabrics_common(p, spc); + } +} + +const char *nvmet_trace_disk_name(struct trace_seq *p, char *name) +{ + const char *ret = trace_seq_buffer_ptr(p); + + if (*name) + trace_seq_printf(p, "disk=%s, ", name); + trace_seq_putc(p, 0); + + return ret; +} + +const char *nvmet_trace_ctrl_name(struct trace_seq *p, struct nvmet_ctrl *ctrl) +{ + const char *ret = trace_seq_buffer_ptr(p); + + /* + * XXX: We don't know the controller instance before executing the + * connect command itself because the connect command for the admin + * queue will not provide the cntlid which will be allocated in this + * command. In case of io queues, the controller instance will be + * mapped by the extra data of the connect command. + * If we can know the extra data of the connect command in this stage, + * we can update this print statement later. + */ + if (ctrl) + trace_seq_printf(p, "%d", ctrl->cntlid); + else + trace_seq_printf(p, "_"); + trace_seq_putc(p, 0); + + return ret; +} + diff --git a/drivers/nvme/target/trace.h b/drivers/nvme/target/trace.h new file mode 100644 index 000000000000..e645caa882dd --- /dev/null +++ b/drivers/nvme/target/trace.h @@ -0,0 +1,141 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * NVM Express target device driver tracepoints + * Copyright (c) 2018 Johannes Thumshirn, SUSE Linux GmbH + * + * This is entirely based on drivers/nvme/host/trace.h + */ + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM nvmet + +#if !defined(_TRACE_NVMET_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_NVMET_H + +#include <linux/nvme.h> +#include <linux/tracepoint.h> +#include <linux/trace_seq.h> + +#include "nvmet.h" + +const char *nvmet_trace_parse_admin_cmd(struct trace_seq *p, u8 opcode, + u8 *cdw10); +const char *nvmet_trace_parse_nvm_cmd(struct trace_seq *p, u8 opcode, + u8 *cdw10); +const char *nvmet_trace_parse_fabrics_cmd(struct trace_seq *p, u8 fctype, + u8 *spc); + +#define parse_nvme_cmd(qid, opcode, fctype, cdw10) \ + ((opcode) == nvme_fabrics_command ? \ + nvmet_trace_parse_fabrics_cmd(p, fctype, cdw10) : \ + (qid ? \ + nvmet_trace_parse_nvm_cmd(p, opcode, cdw10) : \ + nvmet_trace_parse_admin_cmd(p, opcode, cdw10))) + +const char *nvmet_trace_ctrl_name(struct trace_seq *p, struct nvmet_ctrl *ctrl); +#define __print_ctrl_name(ctrl) \ + nvmet_trace_ctrl_name(p, ctrl) + +const char *nvmet_trace_disk_name(struct trace_seq *p, char *name); +#define __print_disk_name(name) \ + nvmet_trace_disk_name(p, name) + +#ifndef TRACE_HEADER_MULTI_READ +static inline struct nvmet_ctrl *nvmet_req_to_ctrl(struct nvmet_req *req) +{ + return req->sq->ctrl; +} + +static inline void __assign_disk_name(char *name, struct nvmet_req *req, + bool init) +{ + struct nvmet_ctrl *ctrl = nvmet_req_to_ctrl(req); + struct nvmet_ns *ns; + + if ((init && req->sq->qid) || (!init && req->cq->qid)) { + ns = nvmet_find_namespace(ctrl, req->cmd->rw.nsid); + strncpy(name, ns->device_path, DISK_NAME_LEN); + return; + } + + memset(name, 0, DISK_NAME_LEN); +} +#endif + +TRACE_EVENT(nvmet_req_init, + TP_PROTO(struct nvmet_req *req, struct nvme_command *cmd), + TP_ARGS(req, cmd), + TP_STRUCT__entry( + __field(struct nvme_command *, cmd) + __field(struct nvmet_ctrl *, ctrl) + __array(char, disk, DISK_NAME_LEN) + __field(int, qid) + __field(u16, cid) + __field(u8, opcode) + __field(u8, fctype) + __field(u8, flags) + __field(u32, nsid) + __field(u64, metadata) + __array(u8, cdw10, 24) + ), + TP_fast_assign( + __entry->cmd = cmd; + __entry->ctrl = nvmet_req_to_ctrl(req); + __assign_disk_name(__entry->disk, req, true); + __entry->qid = req->sq->qid; + __entry->cid = cmd->common.command_id; + __entry->opcode = cmd->common.opcode; + __entry->fctype = cmd->fabrics.fctype; + __entry->flags = cmd->common.flags; + __entry->nsid = le32_to_cpu(cmd->common.nsid); + __entry->metadata = le64_to_cpu(cmd->common.metadata); + memcpy(__entry->cdw10, &cmd->common.cdw10, + sizeof(__entry->cdw10)); + ), + TP_printk("nvmet%s: %sqid=%d, cmdid=%u, nsid=%u, flags=%#x, " + "meta=%#llx, cmd=(%s, %s)", + __print_ctrl_name(__entry->ctrl), + __print_disk_name(__entry->disk), + __entry->qid, __entry->cid, __entry->nsid, + __entry->flags, __entry->metadata, + show_opcode_name(__entry->qid, __entry->opcode, + __entry->fctype), + parse_nvme_cmd(__entry->qid, __entry->opcode, + __entry->fctype, __entry->cdw10)) +); + +TRACE_EVENT(nvmet_req_complete, + TP_PROTO(struct nvmet_req *req), + TP_ARGS(req), + TP_STRUCT__entry( + __field(struct nvmet_ctrl *, ctrl) + __array(char, disk, DISK_NAME_LEN) + __field(int, qid) + __field(int, cid) + __field(u64, result) + __field(u16, status) + ), + TP_fast_assign( + __entry->ctrl = nvmet_req_to_ctrl(req); + __entry->qid = req->cq->qid; + __entry->cid = req->cqe->command_id; + __entry->result = le64_to_cpu(req->cqe->result.u64); + __entry->status = le16_to_cpu(req->cqe->status) >> 1; + __assign_disk_name(__entry->disk, req, false); + ), + TP_printk("nvmet%s: %sqid=%d, cmdid=%u, res=%#llx, status=%#x", + __print_ctrl_name(__entry->ctrl), + __print_disk_name(__entry->disk), + __entry->qid, __entry->cid, __entry->result, __entry->status) + +); + +#endif /* _TRACE_NVMET_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE trace + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/drivers/scsi/lpfc/lpfc.h b/drivers/scsi/lpfc/lpfc.h index aafcffaa25f7..2c3bb8a966e5 100644 --- a/drivers/scsi/lpfc/lpfc.h +++ b/drivers/scsi/lpfc/lpfc.h @@ -274,6 +274,7 @@ struct lpfc_stats { uint32_t elsXmitADISC; uint32_t elsXmitLOGO; uint32_t elsXmitSCR; + uint32_t elsXmitRSCN; uint32_t elsXmitRNID; uint32_t elsXmitFARP; uint32_t elsXmitFARPR; @@ -819,6 +820,7 @@ struct lpfc_hba { uint32_t cfg_use_msi; uint32_t cfg_auto_imax; uint32_t cfg_fcp_imax; + uint32_t cfg_force_rscn; uint32_t cfg_cq_poll_threshold; uint32_t cfg_cq_max_proc_limit; uint32_t cfg_fcp_cpu_map; diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c index d4c65e2109e2..2bd1e014103b 100644 --- a/drivers/scsi/lpfc/lpfc_attr.c +++ b/drivers/scsi/lpfc/lpfc_attr.c @@ -4959,6 +4959,64 @@ static DEVICE_ATTR(lpfc_req_fw_upgrade, S_IRUGO | S_IWUSR, lpfc_request_firmware_upgrade_store); /** + * lpfc_force_rscn_store + * + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: unused string + * @count: unused variable. + * + * Description: + * Force the switch to send a RSCN to all other NPorts in our zone + * If we are direct connect pt2pt, build the RSCN command ourself + * and send to the other NPort. Not supported for private loop. + * + * Returns: + * 0 - on success + * -EIO - if command is not sent + **/ +static ssize_t +lpfc_force_rscn_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct Scsi_Host *shost = class_to_shost(dev); + struct lpfc_vport *vport = (struct lpfc_vport *)shost->hostdata; + int i; + + i = lpfc_issue_els_rscn(vport, 0); + if (i) + return -EIO; + return strlen(buf); +} + +/* + * lpfc_force_rscn: Force an RSCN to be sent to all remote NPorts + * connected to the HBA. + * + * Value range is any ascii value + */ +static int lpfc_force_rscn; +module_param(lpfc_force_rscn, int, 0644); +MODULE_PARM_DESC(lpfc_force_rscn, + "Force an RSCN to be sent to all remote NPorts"); +lpfc_param_show(force_rscn) + +/** + * lpfc_force_rscn_init - Force an RSCN to be sent to all remote NPorts + * @phba: lpfc_hba pointer. + * @val: unused value. + * + * Returns: + * zero if val saved. + **/ +static int +lpfc_force_rscn_init(struct lpfc_hba *phba, int val) +{ + return 0; +} +static DEVICE_ATTR_RW(lpfc_force_rscn); + +/** * lpfc_fcp_imax_store * * @dev: class device that is converted into a Scsi_host. @@ -5958,6 +6016,7 @@ struct device_attribute *lpfc_hba_attrs[] = { &dev_attr_lpfc_nvme_oas, &dev_attr_lpfc_nvme_embed_cmd, &dev_attr_lpfc_fcp_imax, + &dev_attr_lpfc_force_rscn, &dev_attr_lpfc_cq_poll_threshold, &dev_attr_lpfc_cq_max_proc_limit, &dev_attr_lpfc_fcp_cpu_map, @@ -7005,6 +7064,7 @@ lpfc_get_cfgparam(struct lpfc_hba *phba) lpfc_nvme_oas_init(phba, lpfc_nvme_oas); lpfc_nvme_embed_cmd_init(phba, lpfc_nvme_embed_cmd); lpfc_fcp_imax_init(phba, lpfc_fcp_imax); + lpfc_force_rscn_init(phba, lpfc_force_rscn); lpfc_cq_poll_threshold_init(phba, lpfc_cq_poll_threshold); lpfc_cq_max_proc_limit_init(phba, lpfc_cq_max_proc_limit); lpfc_fcp_cpu_map_init(phba, lpfc_fcp_cpu_map); diff --git a/drivers/scsi/lpfc/lpfc_crtn.h b/drivers/scsi/lpfc/lpfc_crtn.h index e0b14d791b8c..866374801140 100644 --- a/drivers/scsi/lpfc/lpfc_crtn.h +++ b/drivers/scsi/lpfc/lpfc_crtn.h @@ -141,6 +141,7 @@ int lpfc_issue_els_adisc(struct lpfc_vport *, struct lpfc_nodelist *, uint8_t); int lpfc_issue_els_logo(struct lpfc_vport *, struct lpfc_nodelist *, uint8_t); int lpfc_issue_els_npiv_logo(struct lpfc_vport *, struct lpfc_nodelist *); int lpfc_issue_els_scr(struct lpfc_vport *, uint32_t, uint8_t); +int lpfc_issue_els_rscn(struct lpfc_vport *vport, uint8_t retry); int lpfc_issue_fabric_reglogin(struct lpfc_vport *); int lpfc_els_free_iocb(struct lpfc_hba *, struct lpfc_iocbq *); int lpfc_ct_free_iocb(struct lpfc_hba *, struct lpfc_iocbq *); @@ -355,6 +356,7 @@ void lpfc_mbox_timeout_handler(struct lpfc_hba *); struct lpfc_nodelist *lpfc_findnode_did(struct lpfc_vport *, uint32_t); struct lpfc_nodelist *lpfc_findnode_wwpn(struct lpfc_vport *, struct lpfc_name *); +struct lpfc_nodelist *lpfc_findnode_mapped(struct lpfc_vport *vport); int lpfc_sli_issue_mbox_wait(struct lpfc_hba *, LPFC_MBOXQ_t *, uint32_t); @@ -555,6 +557,8 @@ void lpfc_ras_stop_fwlog(struct lpfc_hba *phba); int lpfc_check_fwlog_support(struct lpfc_hba *phba); /* NVME interfaces. */ +void lpfc_nvme_rescan_port(struct lpfc_vport *vport, + struct lpfc_nodelist *ndlp); void lpfc_nvme_unregister_port(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp); int lpfc_nvme_register_port(struct lpfc_vport *vport, diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c index 5ac4f8d76b91..968ed0fd37f7 100644 --- a/drivers/scsi/lpfc/lpfc_els.c +++ b/drivers/scsi/lpfc/lpfc_els.c @@ -30,6 +30,8 @@ #include <scsi/scsi_device.h> #include <scsi/scsi_host.h> #include <scsi/scsi_transport_fc.h> +#include <uapi/scsi/fc/fc_fs.h> +#include <uapi/scsi/fc/fc_els.h> #include "lpfc_hw4.h" #include "lpfc_hw.h" @@ -3079,6 +3081,116 @@ lpfc_issue_els_scr(struct lpfc_vport *vport, uint32_t nportid, uint8_t retry) } /** + * lpfc_issue_els_rscn - Issue an RSCN to the Fabric Controller (Fabric) + * or the other nport (pt2pt). + * @vport: pointer to a host virtual N_Port data structure. + * @retry: number of retries to the command IOCB. + * + * This routine issues a RSCN to the Fabric Controller (DID 0xFFFFFD) + * when connected to a fabric, or to the remote port when connected + * in point-to-point mode. When sent to the Fabric Controller, it will + * replay the RSCN to registered recipients. + * + * Note that, in lpfc_prep_els_iocb() routine, the reference count of ndlp + * will be incremented by 1 for holding the ndlp and the reference to ndlp + * will be stored into the context1 field of the IOCB for the completion + * callback function to the RSCN ELS command. + * + * Return code + * 0 - Successfully issued RSCN command + * 1 - Failed to issue RSCN command + **/ +int +lpfc_issue_els_rscn(struct lpfc_vport *vport, uint8_t retry) +{ + struct lpfc_hba *phba = vport->phba; + struct lpfc_iocbq *elsiocb; + struct lpfc_nodelist *ndlp; + struct { + struct fc_els_rscn rscn; + struct fc_els_rscn_page portid; + } *event; + uint32_t nportid; + uint16_t cmdsize = sizeof(*event); + + /* Not supported for private loop */ + if (phba->fc_topology == LPFC_TOPOLOGY_LOOP && + !(vport->fc_flag & FC_PUBLIC_LOOP)) + return 1; + + if (vport->fc_flag & FC_PT2PT) { + /* find any mapped nport - that would be the other nport */ + ndlp = lpfc_findnode_mapped(vport); + if (!ndlp) + return 1; + } else { + nportid = FC_FID_FCTRL; + /* find the fabric controller node */ + ndlp = lpfc_findnode_did(vport, nportid); + if (!ndlp) { + /* if one didn't exist, make one */ + ndlp = lpfc_nlp_init(vport, nportid); + if (!ndlp) + return 1; + lpfc_enqueue_node(vport, ndlp); + } else if (!NLP_CHK_NODE_ACT(ndlp)) { + ndlp = lpfc_enable_node(vport, ndlp, + NLP_STE_UNUSED_NODE); + if (!ndlp) + return 1; + } + } + + elsiocb = lpfc_prep_els_iocb(vport, 1, cmdsize, retry, ndlp, + ndlp->nlp_DID, ELS_CMD_RSCN_XMT); + + if (!elsiocb) { + /* This will trigger the release of the node just + * allocated + */ + lpfc_nlp_put(ndlp); + return 1; + } + + event = ((struct lpfc_dmabuf *)elsiocb->context2)->virt; + + event->rscn.rscn_cmd = ELS_RSCN; + event->rscn.rscn_page_len = sizeof(struct fc_els_rscn_page); + event->rscn.rscn_plen = cpu_to_be16(cmdsize); + + nportid = vport->fc_myDID; + /* appears that page flags must be 0 for fabric to broadcast RSCN */ + event->portid.rscn_page_flags = 0; + event->portid.rscn_fid[0] = (nportid & 0x00FF0000) >> 16; + event->portid.rscn_fid[1] = (nportid & 0x0000FF00) >> 8; + event->portid.rscn_fid[2] = nportid & 0x000000FF; + + lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_ELS_CMD, + "Issue RSCN: did:x%x", + ndlp->nlp_DID, 0, 0); + + phba->fc_stat.elsXmitRSCN++; + elsiocb->iocb_cmpl = lpfc_cmpl_els_cmd; + if (lpfc_sli_issue_iocb(phba, LPFC_ELS_RING, elsiocb, 0) == + IOCB_ERROR) { + /* The additional lpfc_nlp_put will cause the following + * lpfc_els_free_iocb routine to trigger the rlease of + * the node. + */ + lpfc_nlp_put(ndlp); + lpfc_els_free_iocb(phba, elsiocb); + return 1; + } + /* This will cause the callback-function lpfc_cmpl_els_cmd to + * trigger the release of node. + */ + if (!(vport->fc_flag & FC_PT2PT)) + lpfc_nlp_put(ndlp); + + return 0; +} + +/** * lpfc_issue_els_farpr - Issue a farp to an node on a vport * @vport: pointer to a host virtual N_Port data structure. * @nportid: N_Port identifier to the remote node. @@ -6214,6 +6326,8 @@ lpfc_rscn_recovery_check(struct lpfc_vport *vport) continue; } + if (ndlp->nlp_fc4_type & NLP_FC4_NVME) + lpfc_nvme_rescan_port(vport, ndlp); lpfc_disc_state_machine(vport, ndlp, NULL, NLP_EVT_DEVICE_RECOVERY); @@ -6318,6 +6432,19 @@ lpfc_els_rcv_rscn(struct lpfc_vport *vport, struct lpfc_iocbq *cmdiocb, fc_host_post_event(shost, fc_get_event_number(), FCH_EVT_RSCN, lp[i]); + /* Check if RSCN is coming from a direct-connected remote NPort */ + if (vport->fc_flag & FC_PT2PT) { + /* If so, just ACC it, no other action needed for now */ + lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, + "2024 pt2pt RSCN %08x Data: x%x x%x\n", + *lp, vport->fc_flag, payload_len); + lpfc_els_rsp_acc(vport, ELS_CMD_ACC, cmdiocb, ndlp, NULL); + + if (ndlp->nlp_fc4_type & NLP_FC4_NVME) + lpfc_nvme_rescan_port(vport, ndlp); + return 0; + } + /* If we are about to begin discovery, just ACC the RSCN. * Discovery processing will satisfy it. */ diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c index c43852f97f25..28ecaa7fc715 100644 --- a/drivers/scsi/lpfc/lpfc_hbadisc.c +++ b/drivers/scsi/lpfc/lpfc_hbadisc.c @@ -5277,6 +5277,41 @@ lpfc_findnode_did(struct lpfc_vport *vport, uint32_t did) } struct lpfc_nodelist * +lpfc_findnode_mapped(struct lpfc_vport *vport) +{ + struct Scsi_Host *shost = lpfc_shost_from_vport(vport); + struct lpfc_nodelist *ndlp; + uint32_t data1; + unsigned long iflags; + + spin_lock_irqsave(shost->host_lock, iflags); + + list_for_each_entry(ndlp, &vport->fc_nodes, nlp_listp) { + if (ndlp->nlp_state == NLP_STE_UNMAPPED_NODE || + ndlp->nlp_state == NLP_STE_MAPPED_NODE) { + data1 = (((uint32_t)ndlp->nlp_state << 24) | + ((uint32_t)ndlp->nlp_xri << 16) | + ((uint32_t)ndlp->nlp_type << 8) | + ((uint32_t)ndlp->nlp_rpi & 0xff)); + spin_unlock_irqrestore(shost->host_lock, iflags); + lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE, + "2025 FIND node DID " + "Data: x%p x%x x%x x%x %p\n", + ndlp, ndlp->nlp_DID, + ndlp->nlp_flag, data1, + ndlp->active_rrqs_xri_bitmap); + return ndlp; + } + } + spin_unlock_irqrestore(shost->host_lock, iflags); + + /* FIND node did <did> NOT FOUND */ + lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE, + "2026 FIND mapped did NOT FOUND.\n"); + return NULL; +} + +struct lpfc_nodelist * lpfc_setup_disc_node(struct lpfc_vport *vport, uint32_t did) { struct Scsi_Host *shost = lpfc_shost_from_vport(vport); diff --git a/drivers/scsi/lpfc/lpfc_hw.h b/drivers/scsi/lpfc/lpfc_hw.h index edd8f3982023..5b439a6dcde1 100644 --- a/drivers/scsi/lpfc/lpfc_hw.h +++ b/drivers/scsi/lpfc/lpfc_hw.h @@ -601,6 +601,7 @@ struct fc_vft_header { #define ELS_CMD_RPL 0x57000000 #define ELS_CMD_FAN 0x60000000 #define ELS_CMD_RSCN 0x61040000 +#define ELS_CMD_RSCN_XMT 0x61040008 #define ELS_CMD_SCR 0x62000000 #define ELS_CMD_RNID 0x78000000 #define ELS_CMD_LIRR 0x7A000000 @@ -642,6 +643,7 @@ struct fc_vft_header { #define ELS_CMD_RPL 0x57 #define ELS_CMD_FAN 0x60 #define ELS_CMD_RSCN 0x0461 +#define ELS_CMD_RSCN_XMT 0x08000461 #define ELS_CMD_SCR 0x62 #define ELS_CMD_RNID 0x78 #define ELS_CMD_LIRR 0x7A diff --git a/drivers/scsi/lpfc/lpfc_nvme.c b/drivers/scsi/lpfc/lpfc_nvme.c index 9d99cb915390..fdd16d9f55a1 100644 --- a/drivers/scsi/lpfc/lpfc_nvme.c +++ b/drivers/scsi/lpfc/lpfc_nvme.c @@ -2402,6 +2402,50 @@ lpfc_nvme_register_port(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp) #endif } +/** + * lpfc_nvme_rescan_port - Check to see if we should rescan this remoteport + * + * If the ndlp represents an NVME Target, that we are logged into, + * ping the NVME FC Transport layer to initiate a device rescan + * on this remote NPort. + */ +void +lpfc_nvme_rescan_port(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp) +{ +#if (IS_ENABLED(CONFIG_NVME_FC)) + struct lpfc_nvme_rport *rport; + struct nvme_fc_remote_port *remoteport; + + rport = ndlp->nrport; + + lpfc_printf_vlog(vport, KERN_INFO, LOG_NVME_DISC, + "6170 Rescan NPort DID x%06x type x%x " + "state x%x rport %p\n", + ndlp->nlp_DID, ndlp->nlp_type, ndlp->nlp_state, rport); + if (!rport) + goto input_err; + remoteport = rport->remoteport; + if (!remoteport) + goto input_err; + + /* Only rescan if we are an NVME target in the MAPPED state */ + if (remoteport->port_role & FC_PORT_ROLE_NVME_DISCOVERY && + ndlp->nlp_state == NLP_STE_MAPPED_NODE) { + nvme_fc_rescan_remoteport(remoteport); + + lpfc_printf_vlog(vport, KERN_ERR, LOG_NVME_DISC, + "6172 NVME rescanned DID x%06x " + "port_state x%x\n", + ndlp->nlp_DID, remoteport->port_state); + } + return; +input_err: + lpfc_printf_vlog(vport, KERN_ERR, LOG_NVME_DISC, + "6169 State error: lport %p, rport%p FCID x%06x\n", + vport->localport, ndlp->rport, ndlp->nlp_DID); +#endif +} + /* lpfc_nvme_unregister_port - unbind the DID and port_role from this rport. * * There is no notion of Devloss or rport recovery from the current diff --git a/drivers/scsi/lpfc/lpfc_nvmet.c b/drivers/scsi/lpfc/lpfc_nvmet.c index d74bfd264495..06170824a69b 100644 --- a/drivers/scsi/lpfc/lpfc_nvmet.c +++ b/drivers/scsi/lpfc/lpfc_nvmet.c @@ -1139,6 +1139,22 @@ lpfc_nvmet_defer_rcv(struct nvmet_fc_target_port *tgtport, spin_unlock_irqrestore(&ctxp->ctxlock, iflag); } +static void +lpfc_nvmet_discovery_event(struct nvmet_fc_target_port *tgtport) +{ + struct lpfc_nvmet_tgtport *tgtp; + struct lpfc_hba *phba; + uint32_t rc; + + tgtp = tgtport->private; + phba = tgtp->phba; + + rc = lpfc_issue_els_rscn(phba->pport, 0); + lpfc_printf_log(phba, KERN_ERR, LOG_NVME, + "6420 NVMET subsystem change: Notification %s\n", + (rc) ? "Failed" : "Sent"); +} + static struct nvmet_fc_target_template lpfc_tgttemplate = { .targetport_delete = lpfc_nvmet_targetport_delete, .xmt_ls_rsp = lpfc_nvmet_xmt_ls_rsp, @@ -1146,6 +1162,7 @@ static struct nvmet_fc_target_template lpfc_tgttemplate = { .fcp_abort = lpfc_nvmet_xmt_fcp_abort, .fcp_req_release = lpfc_nvmet_xmt_fcp_release, .defer_rcv = lpfc_nvmet_defer_rcv, + .discovery_event = lpfc_nvmet_discovery_event, .max_hw_queues = 1, .max_sgl_segments = LPFC_NVMET_DEFAULT_SEGS, diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c index d1512e4f9791..4329cc44bb55 100644 --- a/drivers/scsi/lpfc/lpfc_sli.c +++ b/drivers/scsi/lpfc/lpfc_sli.c @@ -9398,6 +9398,7 @@ lpfc_sli4_iocb2wqe(struct lpfc_hba *phba, struct lpfc_iocbq *iocbq, if (if_type >= LPFC_SLI_INTF_IF_TYPE_2) { if (pcmd && (*pcmd == ELS_CMD_FLOGI || *pcmd == ELS_CMD_SCR || + *pcmd == ELS_CMD_RSCN_XMT || *pcmd == ELS_CMD_FDISC || *pcmd == ELS_CMD_LOGO || *pcmd == ELS_CMD_PLOGI)) { |