diff options
Diffstat (limited to 'block/blk-mq.c')
| -rw-r--r-- | block/blk-mq.c | 1088 |
1 files changed, 670 insertions, 418 deletions
diff --git a/block/blk-mq.c b/block/blk-mq.c index 32afb87efbd0..4806b867e37d 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -28,6 +28,7 @@ #include <linux/prefetch.h> #include <linux/blk-crypto.h> #include <linux/part_stat.h> +#include <linux/sched/isolation.h> #include <trace/events/block.h> @@ -42,6 +43,7 @@ static DEFINE_PER_CPU(struct llist_head, blk_cpu_done); static DEFINE_PER_CPU(call_single_data_t, blk_cpu_csd); +static DEFINE_MUTEX(blk_mq_cpuhp_lock); static void blk_mq_insert_request(struct request *rq, blk_insert_t flags); static void blk_mq_request_bypass_insert(struct request *rq, @@ -87,41 +89,83 @@ struct mq_inflight { unsigned int inflight[2]; }; -static bool blk_mq_check_inflight(struct request *rq, void *priv) +static bool blk_mq_check_in_driver(struct request *rq, void *priv) { struct mq_inflight *mi = priv; - if (rq->part && blk_do_io_stat(rq) && - (!mi->part->bd_partno || rq->part == mi->part) && + if (rq->rq_flags & RQF_IO_STAT && + (!bdev_is_partition(mi->part) || rq->part == mi->part) && blk_mq_rq_state(rq) == MQ_RQ_IN_FLIGHT) mi->inflight[rq_data_dir(rq)]++; return true; } -unsigned int blk_mq_in_flight(struct request_queue *q, - struct block_device *part) +void blk_mq_in_driver_rw(struct block_device *part, unsigned int inflight[2]) { struct mq_inflight mi = { .part = part }; - blk_mq_queue_tag_busy_iter(q, blk_mq_check_inflight, &mi); + blk_mq_queue_tag_busy_iter(bdev_get_queue(part), blk_mq_check_in_driver, + &mi); + inflight[READ] = mi.inflight[READ]; + inflight[WRITE] = mi.inflight[WRITE]; +} + +#ifdef CONFIG_LOCKDEP +static bool blk_freeze_set_owner(struct request_queue *q, + struct task_struct *owner) +{ + if (!owner) + return false; - return mi.inflight[0] + mi.inflight[1]; + if (!q->mq_freeze_depth) { + q->mq_freeze_owner = owner; + q->mq_freeze_owner_depth = 1; + q->mq_freeze_disk_dead = !q->disk || + test_bit(GD_DEAD, &q->disk->state) || + !blk_queue_registered(q); + q->mq_freeze_queue_dying = blk_queue_dying(q); + return true; + } + + if (owner == q->mq_freeze_owner) + q->mq_freeze_owner_depth += 1; + return false; } -void blk_mq_in_flight_rw(struct request_queue *q, struct block_device *part, - unsigned int inflight[2]) +/* verify the last unfreeze in owner context */ +static bool blk_unfreeze_check_owner(struct request_queue *q) { - struct mq_inflight mi = { .part = part }; + if (q->mq_freeze_owner != current) + return false; + if (--q->mq_freeze_owner_depth == 0) { + q->mq_freeze_owner = NULL; + return true; + } + return false; +} - blk_mq_queue_tag_busy_iter(q, blk_mq_check_inflight, &mi); - inflight[0] = mi.inflight[0]; - inflight[1] = mi.inflight[1]; +#else + +static bool blk_freeze_set_owner(struct request_queue *q, + struct task_struct *owner) +{ + return false; } -void blk_freeze_queue_start(struct request_queue *q) +static bool blk_unfreeze_check_owner(struct request_queue *q) { + return false; +} +#endif + +bool __blk_freeze_queue_start(struct request_queue *q, + struct task_struct *owner) +{ + bool freeze; + mutex_lock(&q->mq_freeze_lock); + freeze = blk_freeze_set_owner(q, owner); if (++q->mq_freeze_depth == 1) { percpu_ref_kill(&q->q_usage_counter); mutex_unlock(&q->mq_freeze_lock); @@ -130,6 +174,14 @@ void blk_freeze_queue_start(struct request_queue *q) } else { mutex_unlock(&q->mq_freeze_lock); } + + return freeze; +} + +void blk_freeze_queue_start(struct request_queue *q) +{ + if (__blk_freeze_queue_start(q, current)) + blk_freeze_acquire_lock(q); } EXPORT_SYMBOL_GPL(blk_freeze_queue_start); @@ -148,35 +200,17 @@ int blk_mq_freeze_queue_wait_timeout(struct request_queue *q, } EXPORT_SYMBOL_GPL(blk_mq_freeze_queue_wait_timeout); -/* - * Guarantee no request is in use, so we can change any data structure of - * the queue afterward. - */ -void blk_freeze_queue(struct request_queue *q) +void blk_mq_freeze_queue_nomemsave(struct request_queue *q) { - /* - * In the !blk_mq case we are only calling this to kill the - * q_usage_counter, otherwise this increases the freeze depth - * and waits for it to return to zero. For this reason there is - * no blk_unfreeze_queue(), and blk_freeze_queue() is not - * exported to drivers as the only user for unfreeze is blk_mq. - */ blk_freeze_queue_start(q); blk_mq_freeze_queue_wait(q); } +EXPORT_SYMBOL_GPL(blk_mq_freeze_queue_nomemsave); -void blk_mq_freeze_queue(struct request_queue *q) +bool __blk_mq_unfreeze_queue(struct request_queue *q, bool force_atomic) { - /* - * ...just an alias to keep freeze and unfreeze actions balanced - * in the blk_mq_* namespace - */ - blk_freeze_queue(q); -} -EXPORT_SYMBOL_GPL(blk_mq_freeze_queue); + bool unfreeze; -void __blk_mq_unfreeze_queue(struct request_queue *q, bool force_atomic) -{ mutex_lock(&q->mq_freeze_lock); if (force_atomic) q->q_usage_counter.data->force_atomic = true; @@ -186,14 +220,38 @@ void __blk_mq_unfreeze_queue(struct request_queue *q, bool force_atomic) percpu_ref_resurrect(&q->q_usage_counter); wake_up_all(&q->mq_freeze_wq); } + unfreeze = blk_unfreeze_check_owner(q); mutex_unlock(&q->mq_freeze_lock); + + return unfreeze; } -void blk_mq_unfreeze_queue(struct request_queue *q) +void blk_mq_unfreeze_queue_nomemrestore(struct request_queue *q) +{ + if (__blk_mq_unfreeze_queue(q, false)) + blk_unfreeze_release_lock(q); +} +EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue_nomemrestore); + +/* + * non_owner variant of blk_freeze_queue_start + * + * Unlike blk_freeze_queue_start, the queue doesn't need to be unfrozen + * by the same task. This is fragile and should not be used if at all + * possible. + */ +void blk_freeze_queue_start_non_owner(struct request_queue *q) +{ + __blk_freeze_queue_start(q, NULL); +} +EXPORT_SYMBOL_GPL(blk_freeze_queue_start_non_owner); + +/* non_owner variant of blk_mq_unfreeze_queue */ +void blk_mq_unfreeze_queue_non_owner(struct request_queue *q) { __blk_mq_unfreeze_queue(q, false); } -EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue); +EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue_non_owner); /* * FIXME: replace the scsi_internal_device_*block_nowait() calls in the @@ -282,8 +340,9 @@ void blk_mq_quiesce_tagset(struct blk_mq_tag_set *set) if (!blk_queue_skip_tagset_quiesce(q)) blk_mq_quiesce_queue_nowait(q); } - blk_mq_wait_quiesce_done(set); mutex_unlock(&set->tag_list_lock); + + blk_mq_wait_quiesce_done(set); } EXPORT_SYMBOL_GPL(blk_mq_quiesce_tagset); @@ -322,7 +381,6 @@ void blk_rq_init(struct request_queue *q, struct request *rq) rq->tag = BLK_MQ_NO_TAG; rq->internal_tag = BLK_MQ_NO_TAG; rq->start_time_ns = blk_time_get_ns(); - rq->part = NULL; blk_crypto_rq_set_defaults(rq); } EXPORT_SYMBOL(blk_rq_init); @@ -330,14 +388,9 @@ EXPORT_SYMBOL(blk_rq_init); /* Set start and alloc time when the allocated request is actually used */ static inline void blk_mq_rq_time_init(struct request *rq, u64 alloc_time_ns) { - if (blk_mq_need_time_stamp(rq)) - rq->start_time_ns = blk_time_get_ns(); - else - rq->start_time_ns = 0; - #ifdef CONFIG_BLK_RQ_ALLOC_TIME if (blk_queue_rq_alloc_time(rq->q)) - rq->alloc_time_ns = alloc_time_ns ?: rq->start_time_ns; + rq->alloc_time_ns = alloc_time_ns; else rq->alloc_time_ns = 0; #endif @@ -358,8 +411,6 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data, if (data->flags & BLK_MQ_REQ_PM) data->rq_flags |= RQF_PM; - if (blk_queue_io_stat(q)) - data->rq_flags |= RQF_IO_STAT; rq->rq_flags = data->rq_flags; if (data->rq_flags & RQF_SCHED_TAGS) { @@ -375,9 +426,7 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data, rq->io_start_time_ns = 0; rq->stats_sectors = 0; rq->nr_phys_segments = 0; -#if defined(CONFIG_BLK_DEV_INTEGRITY) rq->nr_integrity_segments = 0; -#endif rq->end_io = NULL; rq->end_io_data = NULL; @@ -421,7 +470,7 @@ __blk_mq_alloc_requests_batch(struct blk_mq_alloc_data *data) prefetch(tags->static_rqs[tag]); tag_mask &= ~(1UL << i); rq = blk_mq_rq_ctx_init(data, tags, tag); - rq_list_add(data->cached_rq, rq); + rq_list_add_head(data->cached_rqs, rq); nr++; } if (!(data->rq_flags & RQF_SCHED_TAGS)) @@ -430,7 +479,7 @@ __blk_mq_alloc_requests_batch(struct blk_mq_alloc_data *data) percpu_ref_get_many(&data->q->q_usage_counter, nr - 1); data->nr_tags -= nr; - return rq_list_pop(data->cached_rq); + return rq_list_pop(data->cached_rqs); } static struct request *__blk_mq_alloc_requests(struct blk_mq_alloc_data *data) @@ -447,6 +496,10 @@ static struct request *__blk_mq_alloc_requests(struct blk_mq_alloc_data *data) if (data->cmd_flags & REQ_NOWAIT) data->flags |= BLK_MQ_REQ_NOWAIT; +retry: + data->ctx = blk_mq_get_ctx(q); + data->hctx = blk_mq_map_queue(data->cmd_flags, data->ctx); + if (q->elevator) { /* * All requests use scheduler tags when an I/O scheduler is @@ -468,13 +521,9 @@ static struct request *__blk_mq_alloc_requests(struct blk_mq_alloc_data *data) if (ops->limit_depth) ops->limit_depth(data->cmd_flags, data); } - } - -retry: - data->ctx = blk_mq_get_ctx(q); - data->hctx = blk_mq_map_queue(q, data->cmd_flags, data->ctx); - if (!(data->rq_flags & RQF_SCHED_TAGS)) + } else { blk_mq_tag_busy(data->hctx); + } if (data->flags & BLK_MQ_REQ_RESERVED) data->rq_flags |= RQF_RESV; @@ -525,9 +574,13 @@ static struct request *blk_mq_rq_cache_fill(struct request_queue *q, struct blk_mq_alloc_data data = { .q = q, .flags = flags, + .shallow_depth = 0, .cmd_flags = opf, + .rq_flags = 0, .nr_tags = plug->nr_ios, - .cached_rq = &plug->cached_rq, + .cached_rqs = &plug->cached_rqs, + .ctx = NULL, + .hctx = NULL }; struct request *rq; @@ -552,14 +605,14 @@ static struct request *blk_mq_alloc_cached_request(struct request_queue *q, if (!plug) return NULL; - if (rq_list_empty(plug->cached_rq)) { + if (rq_list_empty(&plug->cached_rqs)) { if (plug->nr_ios == 1) return NULL; rq = blk_mq_rq_cache_fill(q, plug, opf, flags); if (!rq) return NULL; } else { - rq = rq_list_peek(&plug->cached_rq); + rq = rq_list_peek(&plug->cached_rqs); if (!rq || rq->q != q) return NULL; @@ -568,8 +621,8 @@ static struct request *blk_mq_alloc_cached_request(struct request_queue *q, if (op_is_flush(rq->cmd_flags) != op_is_flush(opf)) return NULL; - plug->cached_rq = rq_list_next(rq); - blk_mq_rq_time_init(rq, 0); + rq_list_pop(&plug->cached_rqs); + blk_mq_rq_time_init(rq, blk_time_get_ns()); } rq->cmd_flags = opf; @@ -587,8 +640,13 @@ struct request *blk_mq_alloc_request(struct request_queue *q, blk_opf_t opf, struct blk_mq_alloc_data data = { .q = q, .flags = flags, + .shallow_depth = 0, .cmd_flags = opf, + .rq_flags = 0, .nr_tags = 1, + .cached_rqs = NULL, + .ctx = NULL, + .hctx = NULL }; int ret; @@ -616,8 +674,13 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q, struct blk_mq_alloc_data data = { .q = q, .flags = flags, + .shallow_depth = 0, .cmd_flags = opf, + .rq_flags = 0, .nr_tags = 1, + .cached_rqs = NULL, + .ctx = NULL, + .hctx = NULL }; u64 alloc_time_ns = 0; struct request *rq; @@ -690,6 +753,8 @@ static void blk_mq_finish_request(struct request *rq) { struct request_queue *q = rq->q; + blk_zone_finish_request(rq); + if (rq->rq_flags & RQF_USE_SCHED) { q->elevator->type->ops.finish_request(rq); /* @@ -743,7 +808,7 @@ void blk_mq_free_plug_rqs(struct blk_plug *plug) { struct request *rq; - while ((rq = rq_list_pop(&plug->cached_rq)) != NULL) + while ((rq = rq_list_pop(&plug->cached_rqs)) != NULL) blk_mq_free_request(rq); } @@ -761,34 +826,9 @@ void blk_dump_rq_flags(struct request *rq, char *msg) } EXPORT_SYMBOL(blk_dump_rq_flags); -static void req_bio_endio(struct request *rq, struct bio *bio, - unsigned int nbytes, blk_status_t error) -{ - if (unlikely(error)) { - bio->bi_status = error; - } else if (req_op(rq) == REQ_OP_ZONE_APPEND) { - /* - * Partial zone append completions cannot be supported as the - * BIO fragments may end up not being written sequentially. - */ - if (bio->bi_iter.bi_size != nbytes) - bio->bi_status = BLK_STS_IOERR; - else - bio->bi_iter.bi_sector = rq->__sector; - } - - bio_advance(bio, nbytes); - - if (unlikely(rq->rq_flags & RQF_QUIET)) - bio_set_flag(bio, BIO_QUIET); - /* don't actually finish bio if it's part of flush sequence */ - if (bio->bi_iter.bi_size == 0 && !(rq->rq_flags & RQF_FLUSH_SEQ)) - bio_endio(bio); -} - static void blk_account_io_completion(struct request *req, unsigned int bytes) { - if (req->part && blk_do_io_stat(req)) { + if (req->rq_flags & RQF_IO_STAT) { const int sgrp = op_stat_group(req_op(req)); part_stat_lock(); @@ -808,7 +848,7 @@ static void blk_print_req_error(struct request *req, blk_status_t status) blk_op_str(req_op(req)), (__force u32)(req->cmd_flags & ~REQ_OP_MASK), req->nr_phys_segments, - IOPRIO_PRIO_CLASS(req->ioprio)); + IOPRIO_PRIO_CLASS(req_get_ioprio(req))); } /* @@ -826,10 +866,8 @@ static void blk_complete_request(struct request *req) if (!bio) return; -#ifdef CONFIG_BLK_DEV_INTEGRITY if (blk_integrity_rq(req) && req_op(req) == REQ_OP_READ) - req->q->integrity.profile->complete_fn(req, total_bytes); -#endif + blk_integrity_complete(req, total_bytes); /* * Upper layers may call blk_crypto_evict_key() anytime after the last @@ -845,8 +883,7 @@ static void blk_complete_request(struct request *req) /* Completion has already been traced */ bio_clear_flag(bio, BIO_TRACE_COMPLETION); - if (req_op(req) == REQ_OP_ZONE_APPEND) - bio->bi_iter.bi_sector = req->__sector; + blk_zone_update_request_bio(req, bio); if (!is_flush) bio_endio(bio); @@ -889,6 +926,8 @@ static void blk_complete_request(struct request *req) bool blk_update_request(struct request *req, blk_status_t error, unsigned int nr_bytes) { + bool is_flush = req->rq_flags & RQF_FLUSH_SEQ; + bool quiet = req->rq_flags & RQF_QUIET; int total_bytes; trace_block_rq_complete(req, error, nr_bytes); @@ -896,11 +935,9 @@ bool blk_update_request(struct request *req, blk_status_t error, if (!req->bio) return false; -#ifdef CONFIG_BLK_DEV_INTEGRITY if (blk_integrity_rq(req) && req_op(req) == REQ_OP_READ && error == BLK_STS_OK) - req->q->integrity.profile->complete_fn(req, nr_bytes); -#endif + blk_integrity_complete(req, nr_bytes); /* * Upper layers may call blk_crypto_evict_key() anytime after the last @@ -909,9 +946,8 @@ bool blk_update_request(struct request *req, blk_status_t error, if (blk_crypto_rq_has_keyslot(req) && nr_bytes >= blk_rq_bytes(req)) __blk_crypto_rq_put_keyslot(req); - if (unlikely(error && !blk_rq_is_passthrough(req) && - !(req->rq_flags & RQF_QUIET)) && - !test_bit(GD_DEAD, &req->q->disk->state)) { + if (unlikely(error && !blk_rq_is_passthrough(req) && !quiet) && + !test_bit(GD_DEAD, &req->q->disk->state)) { blk_print_req_error(req, error); trace_block_rq_error(req, error, nr_bytes); } @@ -923,12 +959,33 @@ bool blk_update_request(struct request *req, blk_status_t error, struct bio *bio = req->bio; unsigned bio_bytes = min(bio->bi_iter.bi_size, nr_bytes); - if (bio_bytes == bio->bi_iter.bi_size) + if (unlikely(error)) + bio->bi_status = error; + + if (bio_bytes == bio->bi_iter.bi_size) { req->bio = bio->bi_next; + } else if (bio_is_zone_append(bio) && error == BLK_STS_OK) { + /* + * Partial zone append completions cannot be supported + * as the BIO fragments may end up not being written + * sequentially. + */ + bio->bi_status = BLK_STS_IOERR; + } /* Completion has already been traced */ bio_clear_flag(bio, BIO_TRACE_COMPLETION); - req_bio_endio(req, bio, bio_bytes, error); + if (unlikely(quiet)) + bio_set_flag(bio, BIO_QUIET); + + bio_advance(bio, bio_bytes); + + /* Don't actually finish bio if it's part of flush sequence */ + if (!bio->bi_iter.bi_size) { + blk_zone_update_request_bio(req, bio); + if (!is_flush) + bio_endio(bio); + } total_bytes += bio_bytes; nr_bytes -= bio_bytes; @@ -989,38 +1046,76 @@ static inline void blk_account_io_done(struct request *req, u64 now) * normal IO on queueing nor completion. Accounting the * containing request is enough. */ - if (blk_do_io_stat(req) && req->part && - !(req->rq_flags & RQF_FLUSH_SEQ)) { + if ((req->rq_flags & (RQF_IO_STAT|RQF_FLUSH_SEQ)) == RQF_IO_STAT) { const int sgrp = op_stat_group(req_op(req)); part_stat_lock(); update_io_ticks(req->part, jiffies, true); part_stat_inc(req->part, ios[sgrp]); part_stat_add(req->part, nsecs[sgrp], now - req->start_time_ns); + part_stat_local_dec(req->part, + in_flight[op_is_write(req_op(req))]); part_stat_unlock(); } } +static inline bool blk_rq_passthrough_stats(struct request *req) +{ + struct bio *bio = req->bio; + + if (!blk_queue_passthrough_stat(req->q)) + return false; + + /* Requests without a bio do not transfer data. */ + if (!bio) + return false; + + /* + * Stats are accumulated in the bdev, so must have one attached to a + * bio to track stats. Most drivers do not set the bdev for passthrough + * requests, but nvme is one that will set it. + */ + if (!bio->bi_bdev) + return false; + + /* + * We don't know what a passthrough command does, but we know the + * payload size and data direction. Ensuring the size is aligned to the + * block size filters out most commands with payloads that don't + * represent sector access. + */ + if (blk_rq_bytes(req) & (bdev_logical_block_size(bio->bi_bdev) - 1)) + return false; + return true; +} + static inline void blk_account_io_start(struct request *req) { trace_block_io_start(req); - if (blk_do_io_stat(req)) { - /* - * All non-passthrough requests are created from a bio with one - * exception: when a flush command that is part of a flush sequence - * generated by the state machine in blk-flush.c is cloned onto the - * lower device by dm-multipath we can get here without a bio. - */ - if (req->bio) - req->part = req->bio->bi_bdev; - else - req->part = req->q->disk->part0; + if (!blk_queue_io_stat(req->q)) + return; + if (blk_rq_is_passthrough(req) && !blk_rq_passthrough_stats(req)) + return; - part_stat_lock(); - update_io_ticks(req->part, jiffies, false); - part_stat_unlock(); - } + req->rq_flags |= RQF_IO_STAT; + req->start_time_ns = blk_time_get_ns(); + + /* + * All non-passthrough requests are created from a bio with one + * exception: when a flush command that is part of a flush sequence + * generated by the state machine in blk-flush.c is cloned onto the + * lower device by dm-multipath we can get here without a bio. + */ + if (req->bio) + req->part = req->bio->bi_bdev; + else + req->part = req->q->disk->part0; + + part_stat_lock(); + update_io_ticks(req->part, jiffies, false); + part_stat_local_inc(req->part, in_flight[op_is_write(req_op(req))]); + part_stat_unlock(); } static inline void __blk_mq_end_request_acct(struct request *rq, u64 now) @@ -1129,7 +1224,7 @@ static void blk_complete_reqs(struct llist_head *list) rq->q->mq_ops->complete(rq); } -static __latent_entropy void blk_done_softirq(struct softirq_action *h) +static __latent_entropy void blk_done_softirq(void) { blk_complete_reqs(this_cpu_ptr(&blk_cpu_done)); } @@ -1261,10 +1356,9 @@ void blk_mq_start_request(struct request *rq) WRITE_ONCE(rq->state, MQ_RQ_IN_FLIGHT); rq->mq_hctx->tags->rqs[rq->tag] = rq; -#ifdef CONFIG_BLK_DEV_INTEGRITY if (blk_integrity_rq(rq) && req_op(rq) == REQ_OP_WRITE) - q->integrity.profile->prepare_fn(rq); -#endif + blk_integrity_prepare(rq); + if (rq->bio && rq->bio->bi_opf & REQ_POLLED) WRITE_ONCE(rq->bio->bi_cookie, rq->mq_hctx->queue_num); } @@ -1304,8 +1398,7 @@ static void blk_add_rq_to_plug(struct blk_plug *plug, struct request *rq) */ if (!plug->has_elevator && (rq->rq_flags & RQF_SCHED_TAGS)) plug->has_elevator = true; - rq->rq_next = NULL; - rq_list_add(&plug->mq_list, rq); + rq_list_add_tail(&plug->mq_list, rq); plug->rq_count++; } @@ -1330,11 +1423,6 @@ void blk_execute_rq_nowait(struct request *rq, bool at_head) blk_account_io_start(rq); - /* - * As plugging can be enabled for passthrough requests on a zoned - * device, directly accessing the plug instead of using blk_mq_plug() - * should not have any consequences. - */ if (current->plug && !at_head) { blk_add_rq_to_plug(current->plug, rq); return; @@ -1462,19 +1550,17 @@ static void blk_mq_requeue_work(struct work_struct *work) while (!list_empty(&rq_list)) { rq = list_entry(rq_list.next, struct request, queuelist); + list_del_init(&rq->queuelist); /* - * If RQF_DONTPREP ist set, the request has been started by the + * If RQF_DONTPREP is set, the request has been started by the * driver already and might have driver-specific data allocated * already. Insert it into the hctx dispatch list to avoid * block layer merges for the request. */ - if (rq->rq_flags & RQF_DONTPREP) { - list_del_init(&rq->queuelist); + if (rq->rq_flags & RQF_DONTPREP) blk_mq_request_bypass_insert(rq, 0); - } else { - list_del_init(&rq->queuelist); + else blk_mq_insert_request(rq, BLK_MQ_INSERT_AT_HEAD); - } } while (!list_empty(&flush_list)) { @@ -1707,7 +1793,6 @@ void blk_mq_flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list) sbitmap_for_each_set(&hctx->ctx_map, flush_busy_ctx, &data); } -EXPORT_SYMBOL_GPL(blk_mq_flush_busy_ctxs); struct dispatch_rq_data { struct blk_mq_hw_ctx *hctx; @@ -1921,19 +2006,6 @@ static void blk_mq_handle_dev_resource(struct request *rq, __blk_mq_requeue_request(rq); } -static void blk_mq_handle_zone_resource(struct request *rq, - struct list_head *zone_list) -{ - /* - * If we end up here it is because we cannot dispatch a request to a - * specific zone due to LLD level zone-write locking or other zone - * related resource not being available. In this case, set the request - * aside in zone_list for retrying it later. - */ - list_add(&rq->queuelist, zone_list); - __blk_mq_requeue_request(rq); -} - enum prep_dispatch { PREP_DISPATCH_OK, PREP_DISPATCH_NO_TAG, @@ -2012,14 +2084,13 @@ static void blk_mq_commit_rqs(struct blk_mq_hw_ctx *hctx, int queued, * Returns true if we did some work AND can potentially do more. */ bool blk_mq_dispatch_rq_list(struct blk_mq_hw_ctx *hctx, struct list_head *list, - unsigned int nr_budgets) + bool get_budget) { enum prep_dispatch prep; struct request_queue *q = hctx->queue; struct request *rq; int queued; blk_status_t ret = BLK_STS_OK; - LIST_HEAD(zone_list); bool needs_resource = false; if (list_empty(list)) @@ -2035,7 +2106,7 @@ bool blk_mq_dispatch_rq_list(struct blk_mq_hw_ctx *hctx, struct list_head *list, rq = list_first_entry(list, struct request, queuelist); WARN_ON_ONCE(hctx != rq->mq_hctx); - prep = blk_mq_prep_dispatch_rq(rq, !nr_budgets); + prep = blk_mq_prep_dispatch_rq(rq, get_budget); if (prep != PREP_DISPATCH_OK) break; @@ -2044,12 +2115,6 @@ bool blk_mq_dispatch_rq_list(struct blk_mq_hw_ctx *hctx, struct list_head *list, bd.rq = rq; bd.last = list_empty(list); - /* - * once the request is queued to lld, no need to cover the - * budget any more - */ - if (nr_budgets) - nr_budgets--; ret = q->mq_ops->queue_rq(hctx, &bd); switch (ret) { case BLK_STS_OK: @@ -2061,23 +2126,11 @@ bool blk_mq_dispatch_rq_list(struct blk_mq_hw_ctx *hctx, struct list_head *list, case BLK_STS_DEV_RESOURCE: blk_mq_handle_dev_resource(rq, list); goto out; - case BLK_STS_ZONE_RESOURCE: - /* - * Move the request to zone_list and keep going through - * the dispatch list to find more requests the drive can - * accept. - */ - blk_mq_handle_zone_resource(rq, &zone_list); - needs_resource = true; - break; default: blk_mq_end_request(rq, ret); } } while (!list_empty(list)); out: - if (!list_empty(&zone_list)) - list_splice_tail_init(&zone_list, list); - /* If we didn't flush the entire list, we could have told the driver * there was more coming, but that turned out to be a lie. */ @@ -2095,7 +2148,11 @@ out: ((hctx->flags & BLK_MQ_F_TAG_QUEUE_SHARED) || blk_mq_is_shared_tags(hctx->flags)); - if (nr_budgets) + /* + * If the caller allocated budgets, free the budgets of the + * requests that have not yet been passed to the block driver. + */ + if (!get_budget) blk_mq_release_budgets(q, list); spin_lock(&hctx->lock); @@ -2164,6 +2221,15 @@ static inline int blk_mq_first_mapped_cpu(struct blk_mq_hw_ctx *hctx) } /* + * ->next_cpu is always calculated from hctx->cpumask, so simply use + * it for speeding up the check + */ +static bool blk_mq_hctx_empty_cpumask(struct blk_mq_hw_ctx *hctx) +{ + return hctx->next_cpu >= nr_cpu_ids; +} + +/* * It'd be great if the workqueue API had a way to pass * in a mask and had some smarts for more clever placement. * For now we just round-robin here, switching for every @@ -2174,7 +2240,8 @@ static int blk_mq_hctx_next_cpu(struct blk_mq_hw_ctx *hctx) bool tried = false; int next_cpu = hctx->next_cpu; - if (hctx->queue->nr_hw_queues == 1) + /* Switch to unbound if no allowable CPUs in this hctx */ + if (hctx->queue->nr_hw_queues == 1 || blk_mq_hctx_empty_cpumask(hctx)) return WORK_CPU_UNBOUND; if (--hctx->next_cpu_batch <= 0) { @@ -2225,6 +2292,24 @@ void blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, unsigned long msecs) } EXPORT_SYMBOL(blk_mq_delay_run_hw_queue); +static inline bool blk_mq_hw_queue_need_run(struct blk_mq_hw_ctx *hctx) +{ + bool need_run; + + /* + * When queue is quiesced, we may be switching io scheduler, or + * updating nr_hw_queues, or other things, and we can't run queue + * any more, even blk_mq_hctx_has_pending() can't be called safely. + * + * And queue will be rerun in blk_mq_unquiesce_queue() if it is + * quiesced. + */ + __blk_mq_run_dispatch_ops(hctx->queue, false, + need_run = !blk_queue_quiesced(hctx->queue) && + blk_mq_hctx_has_pending(hctx)); + return need_run; +} + /** * blk_mq_run_hw_queue - Start to run a hardware queue. * @hctx: Pointer to the hardware queue to run. @@ -2245,20 +2330,23 @@ void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async) might_sleep_if(!async && hctx->flags & BLK_MQ_F_BLOCKING); - /* - * When queue is quiesced, we may be switching io scheduler, or - * updating nr_hw_queues, or other things, and we can't run queue - * any more, even __blk_mq_hctx_has_pending() can't be called safely. - * - * And queue will be rerun in blk_mq_unquiesce_queue() if it is - * quiesced. - */ - __blk_mq_run_dispatch_ops(hctx->queue, false, - need_run = !blk_queue_quiesced(hctx->queue) && - blk_mq_hctx_has_pending(hctx)); + need_run = blk_mq_hw_queue_need_run(hctx); + if (!need_run) { + unsigned long flags; - if (!need_run) - return; + /* + * Synchronize with blk_mq_unquiesce_queue(), because we check + * if hw queue is quiesced locklessly above, we need the use + * ->queue_lock to make sure we see the up-to-date status to + * not miss rerunning the hw queue. + */ + spin_lock_irqsave(&hctx->queue->queue_lock, flags); + need_run = blk_mq_hw_queue_need_run(hctx); + spin_unlock_irqrestore(&hctx->queue->queue_lock, flags); + + if (!need_run) + return; + } if (async || !cpumask_test_cpu(raw_smp_processor_id(), hctx->cpumask)) { blk_mq_delay_run_hw_queue(hctx, 0); @@ -2415,6 +2503,12 @@ void blk_mq_start_stopped_hw_queue(struct blk_mq_hw_ctx *hctx, bool async) return; clear_bit(BLK_MQ_S_STOPPED, &hctx->state); + /* + * Pairs with the smp_mb() in blk_mq_hctx_stopped() to order the + * clearing of BLK_MQ_S_STOPPED above and the checking of dispatch + * list in the subsequent routine. + */ + smp_mb__after_atomic(); blk_mq_run_hw_queue(hctx, async); } EXPORT_SYMBOL_GPL(blk_mq_start_stopped_hw_queue); @@ -2566,9 +2660,13 @@ static void blk_mq_bio_to_request(struct request *rq, struct bio *bio, if (bio->bi_opf & REQ_RAHEAD) rq->cmd_flags |= REQ_FAILFAST_MASK; + rq->bio = rq->biotail = bio; rq->__sector = bio->bi_iter.bi_sector; - rq->write_hint = bio->bi_write_hint; - blk_rq_bio_prep(rq, bio, nr_segs); + rq->__data_len = bio->bi_iter.bi_size; + rq->nr_phys_segments = nr_segs; + if (bio_integrity(bio)) + rq->nr_integrity_segments = blk_rq_count_integrity_sg(rq->q, + bio); /* This can't fail, since GFP_NOIO includes __GFP_DIRECT_RECLAIM. */ err = blk_crypto_rq_bio_prep(rq, bio, GFP_NOIO); @@ -2642,6 +2740,7 @@ static void blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, if (blk_mq_hctx_stopped(hctx) || blk_queue_quiesced(rq->q)) { blk_mq_insert_request(rq, 0); + blk_mq_run_hw_queue(hctx, false); return; } @@ -2672,6 +2771,7 @@ static blk_status_t blk_mq_request_issue_directly(struct request *rq, bool last) if (blk_mq_hctx_stopped(hctx) || blk_queue_quiesced(rq->q)) { blk_mq_insert_request(rq, 0); + blk_mq_run_hw_queue(hctx, false); return BLK_STS_OK; } @@ -2680,15 +2780,15 @@ static blk_status_t blk_mq_request_issue_directly(struct request *rq, bool last) return __blk_mq_issue_directly(hctx, rq, last); } -static void blk_mq_plug_issue_direct(struct blk_plug *plug) +static void blk_mq_issue_direct(struct rq_list *rqs) { struct blk_mq_hw_ctx *hctx = NULL; struct request *rq; int queued = 0; blk_status_t ret = BLK_STS_OK; - while ((rq = rq_list_pop(&plug->mq_list))) { - bool last = rq_list_empty(plug->mq_list); + while ((rq = rq_list_pop(rqs))) { + bool last = rq_list_empty(rqs); if (hctx != rq->mq_hctx) { if (hctx) { @@ -2719,26 +2819,74 @@ out: blk_mq_commit_rqs(hctx, queued, false); } -static void __blk_mq_flush_plug_list(struct request_queue *q, - struct blk_plug *plug) +static void __blk_mq_flush_list(struct request_queue *q, struct rq_list *rqs) { if (blk_queue_quiesced(q)) return; - q->mq_ops->queue_rqs(&plug->mq_list); + q->mq_ops->queue_rqs(rqs); +} + +static unsigned blk_mq_extract_queue_requests(struct rq_list *rqs, + struct rq_list *queue_rqs) +{ + struct request *rq = rq_list_pop(rqs); + struct request_queue *this_q = rq->q; + struct request **prev = &rqs->head; + struct rq_list matched_rqs = {}; + struct request *last = NULL; + unsigned depth = 1; + + rq_list_add_tail(&matched_rqs, rq); + while ((rq = *prev)) { + if (rq->q == this_q) { + /* move rq from rqs to matched_rqs */ + *prev = rq->rq_next; + rq_list_add_tail(&matched_rqs, rq); + depth++; + } else { + /* leave rq in rqs */ + prev = &rq->rq_next; + last = rq; + } + } + + rqs->tail = last; + *queue_rqs = matched_rqs; + return depth; } -static void blk_mq_dispatch_plug_list(struct blk_plug *plug, bool from_sched) +static void blk_mq_dispatch_queue_requests(struct rq_list *rqs, unsigned depth) +{ + struct request_queue *q = rq_list_peek(rqs)->q; + + trace_block_unplug(q, depth, true); + + /* + * Peek first request and see if we have a ->queue_rqs() hook. + * If we do, we can dispatch the whole list in one go. + * We already know at this point that all requests belong to the + * same queue, caller must ensure that's the case. + */ + if (q->mq_ops->queue_rqs) { + blk_mq_run_dispatch_ops(q, __blk_mq_flush_list(q, rqs)); + if (rq_list_empty(rqs)) + return; + } + + blk_mq_run_dispatch_ops(q, blk_mq_issue_direct(rqs)); +} + +static void blk_mq_dispatch_list(struct rq_list *rqs, bool from_sched) { struct blk_mq_hw_ctx *this_hctx = NULL; struct blk_mq_ctx *this_ctx = NULL; - struct request *requeue_list = NULL; - struct request **requeue_lastp = &requeue_list; + struct rq_list requeue_list = {}; unsigned int depth = 0; bool is_passthrough = false; LIST_HEAD(list); do { - struct request *rq = rq_list_pop(&plug->mq_list); + struct request *rq = rq_list_pop(rqs); if (!this_hctx) { this_hctx = rq->mq_hctx; @@ -2746,14 +2894,14 @@ static void blk_mq_dispatch_plug_list(struct blk_plug *plug, bool from_sched) is_passthrough = blk_rq_is_passthrough(rq); } else if (this_hctx != rq->mq_hctx || this_ctx != rq->mq_ctx || is_passthrough != blk_rq_is_passthrough(rq)) { - rq_list_add_tail(&requeue_lastp, rq); + rq_list_add_tail(&requeue_list, rq); continue; } - list_add(&rq->queuelist, &list); + list_add_tail(&rq->queuelist, &list); depth++; - } while (!rq_list_empty(plug->mq_list)); + } while (!rq_list_empty(rqs)); - plug->mq_list = requeue_list; + *rqs = requeue_list; trace_block_unplug(this_hctx->queue, depth, !from_sched); percpu_ref_get(&this_hctx->queue->q_usage_counter); @@ -2773,9 +2921,22 @@ static void blk_mq_dispatch_plug_list(struct blk_plug *plug, bool from_sched) percpu_ref_put(&this_hctx->queue->q_usage_counter); } +static void blk_mq_dispatch_multiple_queue_requests(struct rq_list *rqs) +{ + do { + struct rq_list queue_rqs; + unsigned depth; + + depth = blk_mq_extract_queue_requests(rqs, &queue_rqs); + blk_mq_dispatch_queue_requests(&queue_rqs, depth); + while (!rq_list_empty(&queue_rqs)) + blk_mq_dispatch_list(&queue_rqs, false); + } while (!rq_list_empty(rqs)); +} + void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule) { - struct request *rq; + unsigned int depth; /* * We may have been called recursively midway through handling @@ -2786,36 +2947,23 @@ void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule) */ if (plug->rq_count == 0) return; + depth = plug->rq_count; plug->rq_count = 0; - if (!plug->multiple_queues && !plug->has_elevator && !from_schedule) { - struct request_queue *q; - - rq = rq_list_peek(&plug->mq_list); - q = rq->q; - - /* - * Peek first request and see if we have a ->queue_rqs() hook. - * If we do, we can dispatch the whole plug list in one go. We - * already know at this point that all requests belong to the - * same queue, caller must ensure that's the case. - */ - if (q->mq_ops->queue_rqs) { - blk_mq_run_dispatch_ops(q, - __blk_mq_flush_plug_list(q, plug)); - if (rq_list_empty(plug->mq_list)) - return; + if (!plug->has_elevator && !from_schedule) { + if (plug->multiple_queues) { + blk_mq_dispatch_multiple_queue_requests(&plug->mq_list); + return; } - blk_mq_run_dispatch_ops(q, - blk_mq_plug_issue_direct(plug)); - if (rq_list_empty(plug->mq_list)) + blk_mq_dispatch_queue_requests(&plug->mq_list, depth); + if (rq_list_empty(&plug->mq_list)) return; } do { - blk_mq_dispatch_plug_list(plug, from_schedule); - } while (!rq_list_empty(plug->mq_list)); + blk_mq_dispatch_list(&plug->mq_list, from_schedule); + } while (!rq_list_empty(&plug->mq_list)); } static void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx, @@ -2865,13 +3013,18 @@ static bool blk_mq_attempt_bio_merge(struct request_queue *q, static struct request *blk_mq_get_new_requests(struct request_queue *q, struct blk_plug *plug, - struct bio *bio, - unsigned int nsegs) + struct bio *bio) { struct blk_mq_alloc_data data = { .q = q, - .nr_tags = 1, + .flags = 0, + .shallow_depth = 0, .cmd_flags = bio->bi_opf, + .rq_flags = 0, + .nr_tags = 1, + .cached_rqs = NULL, + .ctx = NULL, + .hctx = NULL }; struct request *rq; @@ -2880,16 +3033,13 @@ static struct request *blk_mq_get_new_requests(struct request_queue *q, if (plug) { data.nr_tags = plug->nr_ios; plug->nr_ios = 1; - data.cached_rq = &plug->cached_rq; + data.cached_rqs = &plug->cached_rqs; } rq = __blk_mq_alloc_requests(&data); - if (rq) - return rq; - rq_qos_cleanup(q, bio); - if (bio->bi_opf & REQ_NOWAIT) - bio_wouldblock_error(bio); - return NULL; + if (unlikely(!rq)) + rq_qos_cleanup(q, bio); + return rq; } /* @@ -2903,7 +3053,7 @@ static struct request *blk_mq_peek_cached_request(struct blk_plug *plug, if (!plug) return NULL; - rq = rq_list_peek(&plug->cached_rq); + rq = rq_list_peek(&plug->cached_rqs); if (!rq || rq->q != q) return NULL; if (type != rq->mq_hctx->type && @@ -2917,21 +3067,32 @@ static struct request *blk_mq_peek_cached_request(struct blk_plug *plug, static void blk_mq_use_cached_rq(struct request *rq, struct blk_plug *plug, struct bio *bio) { - WARN_ON_ONCE(rq_list_peek(&plug->cached_rq) != rq); + if (rq_list_pop(&plug->cached_rqs) != rq) + WARN_ON_ONCE(1); /* * If any qos ->throttle() end up blocking, we will have flushed the * plug and hence killed the cached_rq list as well. Pop this entry * before we throttle. */ - plug->cached_rq = rq_list_next(rq); rq_qos_throttle(rq->q, bio); - blk_mq_rq_time_init(rq, 0); + blk_mq_rq_time_init(rq, blk_time_get_ns()); rq->cmd_flags = bio->bi_opf; INIT_LIST_HEAD(&rq->queuelist); } +static bool bio_unaligned(const struct bio *bio, struct request_queue *q) +{ + unsigned int bs_mask = queue_logical_block_size(q) - 1; + + /* .bi_sector of any zero sized bio need to be initialized */ + if ((bio->bi_iter.bi_size & bs_mask) || + ((bio->bi_iter.bi_sector << SECTOR_SHIFT) & bs_mask)) + return true; + return false; +} + /** * blk_mq_submit_bio - Create and send a request to block device. * @bio: Bio pointer. @@ -2948,44 +3109,79 @@ static void blk_mq_use_cached_rq(struct request *rq, struct blk_plug *plug, void blk_mq_submit_bio(struct bio *bio) { struct request_queue *q = bdev_get_queue(bio->bi_bdev); - struct blk_plug *plug = blk_mq_plug(bio); + struct blk_plug *plug = current->plug; const int is_sync = op_is_sync(bio->bi_opf); struct blk_mq_hw_ctx *hctx; - unsigned int nr_segs = 1; + unsigned int nr_segs; struct request *rq; blk_status_t ret; - bio = blk_queue_bounce(bio, q); + /* + * If the plug has a cached request for this queue, try to use it. + */ + rq = blk_mq_peek_cached_request(plug, q, bio->bi_opf); + + /* + * A BIO that was released from a zone write plug has already been + * through the preparation in this function, already holds a reference + * on the queue usage counter, and is the only write BIO in-flight for + * the target zone. Go straight to preparing a request for it. + */ + if (bio_zone_write_plugging(bio)) { + nr_segs = bio->__bi_nr_segments; + if (rq) + blk_queue_exit(q); + goto new_request; + } /* - * If the plug has a cached request for this queue, try use it. - * * The cached request already holds a q_usage_counter reference and we * don't have to acquire a new one if we use it. */ - rq = blk_mq_peek_cached_request(plug, q, bio->bi_opf); if (!rq) { if (unlikely(bio_queue_enter(bio))) return; } - if (unlikely(bio_may_exceed_limits(bio, &q->limits))) { - bio = __bio_split_to_limits(bio, &q->limits, &nr_segs); - if (!bio) - goto queue_exit; + /* + * Device reconfiguration may change logical block size or reduce the + * number of poll queues, so the checks for alignment and poll support + * have to be done with queue usage counter held. + */ + if (unlikely(bio_unaligned(bio, q))) { + bio_io_error(bio); + goto queue_exit; } + + if ((bio->bi_opf & REQ_POLLED) && !blk_mq_can_poll(q)) { + bio->bi_status = BLK_STS_NOTSUPP; + bio_endio(bio); + goto queue_exit; + } + + bio = __bio_split_to_limits(bio, &q->limits, &nr_segs); + if (!bio) + goto queue_exit; + if (!bio_integrity_prep(bio)) goto queue_exit; if (blk_mq_attempt_bio_merge(q, bio, nr_segs)) goto queue_exit; - if (!rq) { - rq = blk_mq_get_new_requests(q, plug, bio, nr_segs); - if (unlikely(!rq)) - goto queue_exit; - } else { + if (blk_queue_is_zoned(q) && blk_zone_plug_bio(bio, nr_segs)) + goto queue_exit; + +new_request: + if (rq) { blk_mq_use_cached_rq(rq, plug, bio); + } else { + rq = blk_mq_get_new_requests(q, plug, bio); + if (unlikely(!rq)) { + if (bio->bi_opf & REQ_NOWAIT) + bio_wouldblock_error(bio); + goto queue_exit; + } } trace_block_getrq(bio); @@ -3002,6 +3198,9 @@ void blk_mq_submit_bio(struct bio *bio) return; } + if (bio_zone_write_plugging(bio)) + blk_zone_write_plug_init_request(rq); + if (op_is_flush(bio->bi_opf) && blk_insert_flush(rq)) return; @@ -3037,7 +3236,7 @@ queue_exit: blk_status_t blk_insert_cloned_request(struct request *rq) { struct request_queue *q = rq->q; - unsigned int max_sectors = blk_queue_get_max_sectors(q, req_op(rq)); + unsigned int max_sectors = blk_queue_get_max_sectors(rq); unsigned int max_segments = blk_rq_get_max_segments(rq); blk_status_t ret; @@ -3134,19 +3333,21 @@ int blk_rq_prep_clone(struct request *rq, struct request *rq_src, int (*bio_ctr)(struct bio *, struct bio *, void *), void *data) { - struct bio *bio, *bio_src; + struct bio *bio_src; if (!bs) bs = &fs_bio_set; __rq_for_each_bio(bio_src, rq_src) { - bio = bio_alloc_clone(rq->q->disk->part0, bio_src, gfp_mask, - bs); + struct bio *bio = bio_alloc_clone(rq->q->disk->part0, bio_src, + gfp_mask, bs); if (!bio) goto free_and_out; - if (bio_ctr && bio_ctr(bio, bio_src, data)) + if (bio_ctr && bio_ctr(bio, bio_src, data)) { + bio_put(bio); goto free_and_out; + } if (rq->bio) { rq->biotail->bi_next = bio; @@ -3154,7 +3355,6 @@ int blk_rq_prep_clone(struct request *rq, struct request *rq_src, } else { rq->bio = rq->biotail = bio; } - bio = NULL; } /* Copy attributes of the original request to the clone request. */ @@ -3165,8 +3365,7 @@ int blk_rq_prep_clone(struct request *rq, struct request *rq_src, rq->special_vec = rq_src->special_vec; } rq->nr_phys_segments = rq_src->nr_phys_segments; - rq->ioprio = rq_src->ioprio; - rq->write_hint = rq_src->write_hint; + rq->nr_integrity_segments = rq_src->nr_integrity_segments; if (rq->bio && blk_crypto_rq_bio_prep(rq, rq->bio, gfp_mask) < 0) goto free_and_out; @@ -3174,8 +3373,6 @@ int blk_rq_prep_clone(struct request *rq, struct request *rq_src, return 0; free_and_out: - if (bio) - bio_put(bio); blk_rq_unprep_clone(rq); return -ENOMEM; @@ -3338,8 +3535,7 @@ static struct blk_mq_tags *blk_mq_alloc_rq_map(struct blk_mq_tag_set *set, if (node == NUMA_NO_NODE) node = set->numa_node; - tags = blk_mq_init_tags(nr_tags, reserved_tags, node, - BLK_MQ_FLAG_TO_ALLOC_POLICY(set->flags)); + tags = blk_mq_init_tags(nr_tags, reserved_tags, set->flags, node); if (!tags) return NULL; @@ -3483,14 +3679,30 @@ static bool blk_mq_hctx_has_requests(struct blk_mq_hw_ctx *hctx) return data.has_rq; } -static inline bool blk_mq_last_cpu_in_hctx(unsigned int cpu, - struct blk_mq_hw_ctx *hctx) +static bool blk_mq_hctx_has_online_cpu(struct blk_mq_hw_ctx *hctx, + unsigned int this_cpu) { - if (cpumask_first_and(hctx->cpumask, cpu_online_mask) != cpu) - return false; - if (cpumask_next_and(cpu, hctx->cpumask, cpu_online_mask) < nr_cpu_ids) - return false; - return true; + enum hctx_type type = hctx->type; + int cpu; + + /* + * hctx->cpumask has to rule out isolated CPUs, but userspace still + * might submit IOs on these isolated CPUs, so use the queue map to + * check if all CPUs mapped to this hctx are offline + */ + for_each_online_cpu(cpu) { + struct blk_mq_hw_ctx *h = blk_mq_map_queue_type(hctx->queue, + type, cpu); + + if (h != hctx) + continue; + + /* this hctx has at least one online CPU */ + if (this_cpu != cpu) + return true; + } + + return false; } static int blk_mq_hctx_notify_offline(unsigned int cpu, struct hlist_node *node) @@ -3498,8 +3710,7 @@ static int blk_mq_hctx_notify_offline(unsigned int cpu, struct hlist_node *node) struct blk_mq_hw_ctx *hctx = hlist_entry_safe(node, struct blk_mq_hw_ctx, cpuhp_online); - if (!cpumask_test_cpu(cpu, hctx->cpumask) || - !blk_mq_last_cpu_in_hctx(cpu, hctx)) + if (blk_mq_hctx_has_online_cpu(hctx, cpu)) return 0; /* @@ -3526,12 +3737,28 @@ static int blk_mq_hctx_notify_offline(unsigned int cpu, struct hlist_node *node) return 0; } +/* + * Check if one CPU is mapped to the specified hctx + * + * Isolated CPUs have been ruled out from hctx->cpumask, which is supposed + * to be used for scheduling kworker only. For other usage, please call this + * helper for checking if one CPU belongs to the specified hctx + */ +static bool blk_mq_cpu_mapped_to_hctx(unsigned int cpu, + const struct blk_mq_hw_ctx *hctx) +{ + struct blk_mq_hw_ctx *mapped_hctx = blk_mq_map_queue_type(hctx->queue, + hctx->type, cpu); + + return mapped_hctx == hctx; +} + static int blk_mq_hctx_notify_online(unsigned int cpu, struct hlist_node *node) { struct blk_mq_hw_ctx *hctx = hlist_entry_safe(node, struct blk_mq_hw_ctx, cpuhp_online); - if (cpumask_test_cpu(cpu, hctx->cpumask)) + if (blk_mq_cpu_mapped_to_hctx(cpu, hctx)) clear_bit(BLK_MQ_S_INACTIVE, &hctx->state); return 0; } @@ -3549,7 +3776,7 @@ static int blk_mq_hctx_notify_dead(unsigned int cpu, struct hlist_node *node) enum hctx_type type; hctx = hlist_entry_safe(node, struct blk_mq_hw_ctx, cpuhp_dead); - if (!cpumask_test_cpu(cpu, hctx->cpumask)) + if (!blk_mq_cpu_mapped_to_hctx(cpu, hctx)) return 0; ctx = __blk_mq_get_ctx(hctx->queue, cpu); @@ -3573,13 +3800,91 @@ static int blk_mq_hctx_notify_dead(unsigned int cpu, struct hlist_node *node) return 0; } -static void blk_mq_remove_cpuhp(struct blk_mq_hw_ctx *hctx) +static void __blk_mq_remove_cpuhp(struct blk_mq_hw_ctx *hctx) { - if (!(hctx->flags & BLK_MQ_F_STACKING)) + lockdep_assert_held(&blk_mq_cpuhp_lock); + + if (!(hctx->flags & BLK_MQ_F_STACKING) && + !hlist_unhashed(&hctx->cpuhp_online)) { cpuhp_state_remove_instance_nocalls(CPUHP_AP_BLK_MQ_ONLINE, &hctx->cpuhp_online); - cpuhp_state_remove_instance_nocalls(CPUHP_BLK_MQ_DEAD, - &hctx->cpuhp_dead); + INIT_HLIST_NODE(&hctx->cpuhp_online); + } + + if (!hlist_unhashed(&hctx->cpuhp_dead)) { + cpuhp_state_remove_instance_nocalls(CPUHP_BLK_MQ_DEAD, + &hctx->cpuhp_dead); + INIT_HLIST_NODE(&hctx->cpuhp_dead); + } +} + +static void blk_mq_remove_cpuhp(struct blk_mq_hw_ctx *hctx) +{ + mutex_lock(&blk_mq_cpuhp_lock); + __blk_mq_remove_cpuhp(hctx); + mutex_unlock(&blk_mq_cpuhp_lock); +} + +static void __blk_mq_add_cpuhp(struct blk_mq_hw_ctx *hctx) +{ + lockdep_assert_held(&blk_mq_cpuhp_lock); + + if (!(hctx->flags & BLK_MQ_F_STACKING) && + hlist_unhashed(&hctx->cpuhp_online)) + cpuhp_state_add_instance_nocalls(CPUHP_AP_BLK_MQ_ONLINE, + &hctx->cpuhp_online); + + if (hlist_unhashed(&hctx->cpuhp_dead)) + cpuhp_state_add_instance_nocalls(CPUHP_BLK_MQ_DEAD, + &hctx->cpuhp_dead); +} + +static void __blk_mq_remove_cpuhp_list(struct list_head *head) +{ + struct blk_mq_hw_ctx *hctx; + + lockdep_assert_held(&blk_mq_cpuhp_lock); + + list_for_each_entry(hctx, head, hctx_list) + __blk_mq_remove_cpuhp(hctx); +} + +/* + * Unregister cpuhp callbacks from exited hw queues + * + * Safe to call if this `request_queue` is live + */ +static void blk_mq_remove_hw_queues_cpuhp(struct request_queue *q) +{ + LIST_HEAD(hctx_list); + + spin_lock(&q->unused_hctx_lock); + list_splice_init(&q->unused_hctx_list, &hctx_list); + spin_unlock(&q->unused_hctx_lock); + + mutex_lock(&blk_mq_cpuhp_lock); + __blk_mq_remove_cpuhp_list(&hctx_list); + mutex_unlock(&blk_mq_cpuhp_lock); + + spin_lock(&q->unused_hctx_lock); + list_splice(&hctx_list, &q->unused_hctx_list); + spin_unlock(&q->unused_hctx_lock); +} + +/* + * Register cpuhp callbacks from all hw queues + * + * Safe to call if this `request_queue` is live + */ +static void blk_mq_add_hw_queues_cpuhp(struct request_queue *q) +{ + struct blk_mq_hw_ctx *hctx; + unsigned long i; + + mutex_lock(&blk_mq_cpuhp_lock); + queue_for_each_hw_ctx(q, hctx, i) + __blk_mq_add_cpuhp(hctx); + mutex_unlock(&blk_mq_cpuhp_lock); } /* @@ -3630,8 +3935,6 @@ static void blk_mq_exit_hctx(struct request_queue *q, if (set->ops->exit_hctx) set->ops->exit_hctx(hctx, hctx_idx); - blk_mq_remove_cpuhp(hctx); - xa_erase(&q->hctx_table, hctx_idx); spin_lock(&q->unused_hctx_lock); @@ -3648,6 +3951,7 @@ static void blk_mq_exit_hw_queues(struct request_queue *q, queue_for_each_hw_ctx(q, hctx, i) { if (i == nr_queue) break; + blk_mq_remove_cpuhp(hctx); blk_mq_exit_hctx(q, set, hctx, i); } } @@ -3658,16 +3962,11 @@ static int blk_mq_init_hctx(struct request_queue *q, { hctx->queue_num = hctx_idx; - if (!(hctx->flags & BLK_MQ_F_STACKING)) - cpuhp_state_add_instance_nocalls(CPUHP_AP_BLK_MQ_ONLINE, - &hctx->cpuhp_online); - cpuhp_state_add_instance_nocalls(CPUHP_BLK_MQ_DEAD, &hctx->cpuhp_dead); - hctx->tags = set->tags[hctx_idx]; if (set->ops->init_hctx && set->ops->init_hctx(hctx, set->driver_data, hctx_idx)) - goto unregister_cpu_notifier; + goto fail; if (blk_mq_init_request(set, hctx->fq->flush_rq, hctx_idx, hctx->numa_node)) @@ -3684,8 +3983,7 @@ static int blk_mq_init_hctx(struct request_queue *q, exit_hctx: if (set->ops->exit_hctx) set->ops->exit_hctx(hctx, hctx_idx); - unregister_cpu_notifier: - blk_mq_remove_cpuhp(hctx); + fail: return -1; } @@ -3711,6 +4009,8 @@ blk_mq_alloc_hctx(struct request_queue *q, struct blk_mq_tag_set *set, INIT_DELAYED_WORK(&hctx->run_work, blk_mq_run_work_fn); spin_lock_init(&hctx->lock); INIT_LIST_HEAD(&hctx->dispatch); + INIT_HLIST_NODE(&hctx->cpuhp_dead); + INIT_HLIST_NODE(&hctx->cpuhp_online); hctx->queue = q; hctx->flags = set->flags & ~BLK_MQ_F_TAG_QUEUE_SHARED; @@ -3907,6 +4207,8 @@ static void blk_mq_map_swqueue(struct request_queue *q) } queue_for_each_hw_ctx(q, hctx, i) { + int cpu; + /* * If no software queues are mapped to this hardware queue, * disable it and free the request entries. @@ -3934,6 +4236,15 @@ static void blk_mq_map_swqueue(struct request_queue *q) sbitmap_resize(&hctx->ctx_map, hctx->nr_ctx); /* + * Rule out isolated CPUs from hctx->cpumask to avoid + * running block kworker on isolated CPUs + */ + for_each_cpu(cpu, hctx->cpumask) { + if (cpu_is_isolated(cpu)) + cpumask_clear_cpu(cpu, hctx->cpumask); + } + + /* * Initialize batch roundrobin counts */ hctx->next_cpu = blk_mq_first_mapped_cpu(hctx); @@ -3964,13 +4275,14 @@ static void blk_mq_update_tag_set_shared(struct blk_mq_tag_set *set, bool shared) { struct request_queue *q; + unsigned int memflags; lockdep_assert_held(&set->tag_list_lock); list_for_each_entry(q, &set->tag_list, tag_set_list) { - blk_mq_freeze_queue(q); + memflags = blk_mq_freeze_queue(q); queue_set_hctx_shared(q, shared); - blk_mq_unfreeze_queue(q); + blk_mq_unfreeze_queue(q, memflags); } } @@ -4075,7 +4387,13 @@ struct request_queue *blk_mq_alloc_queue(struct blk_mq_tag_set *set, struct request_queue *q; int ret; - q = blk_alloc_queue(lim ? lim : &default_lim, set->numa_node); + if (!lim) + lim = &default_lim; + lim->features |= BLK_FEAT_IO_STAT | BLK_FEAT_NOWAIT; + if (set->nr_maps > HCTX_TYPE_POLL) + lim->features |= BLK_FEAT_POLL; + + q = blk_alloc_queue(lim, set->numa_node); if (IS_ERR(q)) return q; q->queuedata = queuedata; @@ -4151,6 +4469,15 @@ struct gendisk *blk_mq_alloc_disk_for_queue(struct request_queue *q, } EXPORT_SYMBOL(blk_mq_alloc_disk_for_queue); +/* + * Only hctx removed from cpuhp list can be reused + */ +static bool blk_mq_hctx_is_reusable(struct blk_mq_hw_ctx *hctx) +{ + return hlist_unhashed(&hctx->cpuhp_online) && + hlist_unhashed(&hctx->cpuhp_dead); +} + static struct blk_mq_hw_ctx *blk_mq_alloc_and_init_hctx( struct blk_mq_tag_set *set, struct request_queue *q, int hctx_idx, int node) @@ -4160,7 +4487,7 @@ static struct blk_mq_hw_ctx *blk_mq_alloc_and_init_hctx( /* reuse dead hctx first */ spin_lock(&q->unused_hctx_lock); list_for_each_entry(tmp, &q->unused_hctx_list, hctx_list) { - if (tmp->numa_node == node) { + if (tmp->numa_node == node && blk_mq_hctx_is_reusable(tmp)) { hctx = tmp; break; } @@ -4185,14 +4512,12 @@ static struct blk_mq_hw_ctx *blk_mq_alloc_and_init_hctx( return NULL; } -static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set, - struct request_queue *q) +static void __blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set, + struct request_queue *q) { struct blk_mq_hw_ctx *hctx; unsigned long i, j; - /* protect against switching io scheduler */ - mutex_lock(&q->sysfs_lock); for (i = 0; i < set->nr_hw_queues; i++) { int old_node; int node = blk_mq_get_hctx_node(set, i); @@ -4225,18 +4550,18 @@ static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set, xa_for_each_start(&q->hctx_table, j, hctx, j) blk_mq_exit_hctx(q, set, hctx, j); - mutex_unlock(&q->sysfs_lock); } -static void blk_mq_update_poll_flag(struct request_queue *q) +static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set, + struct request_queue *q) { - struct blk_mq_tag_set *set = q->tag_set; + __blk_mq_realloc_hw_ctxs(set, q); - if (set->nr_maps > HCTX_TYPE_POLL && - set->map[HCTX_TYPE_POLL].nr_queues) - blk_queue_flag_set(QUEUE_FLAG_POLL, q); - else - blk_queue_flag_clear(QUEUE_FLAG_POLL, q); + /* unregister cpuhp callbacks for exited hctxs */ + blk_mq_remove_hw_queues_cpuhp(q); + + /* register cpuhp for new initialized hctxs */ + blk_mq_add_hw_queues_cpuhp(q); } int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, @@ -4245,6 +4570,12 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, /* mark the queue as mq asap */ q->mq_ops = set->ops; + /* + * ->tag_set has to be setup before initialize hctx, which cpuphp + * handler needs it for checking queue mapping + */ + q->tag_set = set; + if (blk_mq_alloc_ctxs(q)) goto err_exit; @@ -4263,10 +4594,7 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, INIT_WORK(&q->timeout_work, blk_mq_timeout_work); blk_queue_rq_timeout(q, set->timeout ? set->timeout : 30 * HZ); - q->tag_set = set; - q->queue_flags |= QUEUE_FLAG_MQ_DEFAULT; - blk_mq_update_poll_flag(q); INIT_DELAYED_WORK(&q->requeue_work, blk_mq_requeue_work); INIT_LIST_HEAD(&q->flush_list); @@ -4276,8 +4604,8 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, q->nr_requests = set->queue_depth; blk_mq_init_cpu_queues(q, set->nr_hw_queues); - blk_mq_add_queue_tag_set(set, q); blk_mq_map_swqueue(q); + blk_mq_add_queue_tag_set(set, q); return 0; err_hctxs: @@ -4497,6 +4825,8 @@ int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set) goto out_free_srcu; } + init_rwsem(&set->update_nr_hwq_lock); + ret = -ENOMEM; set->tags = kcalloc_node(set->nr_hw_queues, sizeof(struct blk_mq_tags *), GFP_KERNEL, @@ -4590,13 +4920,15 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr) int ret; unsigned long i; + if (WARN_ON_ONCE(!q->mq_freeze_depth)) + return -EINVAL; + if (!set) return -EINVAL; if (q->nr_requests == nr) return 0; - blk_mq_freeze_queue(q); blk_mq_quiesce_queue(q); ret = 0; @@ -4630,95 +4962,16 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr) } blk_mq_unquiesce_queue(q); - blk_mq_unfreeze_queue(q); return ret; } -/* - * request_queue and elevator_type pair. - * It is just used by __blk_mq_update_nr_hw_queues to cache - * the elevator_type associated with a request_queue. - */ -struct blk_mq_qe_pair { - struct list_head node; - struct request_queue *q; - struct elevator_type *type; -}; - -/* - * Cache the elevator_type in qe pair list and switch the - * io scheduler to 'none' - */ -static bool blk_mq_elv_switch_none(struct list_head *head, - struct request_queue *q) -{ - struct blk_mq_qe_pair *qe; - - qe = kmalloc(sizeof(*qe), GFP_NOIO | __GFP_NOWARN | __GFP_NORETRY); - if (!qe) - return false; - - /* q->elevator needs protection from ->sysfs_lock */ - mutex_lock(&q->sysfs_lock); - - /* the check has to be done with holding sysfs_lock */ - if (!q->elevator) { - kfree(qe); - goto unlock; - } - - INIT_LIST_HEAD(&qe->node); - qe->q = q; - qe->type = q->elevator->type; - /* keep a reference to the elevator module as we'll switch back */ - __elevator_get(qe->type); - list_add(&qe->node, head); - elevator_disable(q); -unlock: - mutex_unlock(&q->sysfs_lock); - - return true; -} - -static struct blk_mq_qe_pair *blk_lookup_qe_pair(struct list_head *head, - struct request_queue *q) -{ - struct blk_mq_qe_pair *qe; - - list_for_each_entry(qe, head, node) - if (qe->q == q) - return qe; - - return NULL; -} - -static void blk_mq_elv_switch_back(struct list_head *head, - struct request_queue *q) -{ - struct blk_mq_qe_pair *qe; - struct elevator_type *t; - - qe = blk_lookup_qe_pair(head, q); - if (!qe) - return; - t = qe->type; - list_del(&qe->node); - kfree(qe); - - mutex_lock(&q->sysfs_lock); - elevator_switch(q, t); - /* drop the reference acquired in blk_mq_elv_switch_none */ - elevator_put(t); - mutex_unlock(&q->sysfs_lock); -} - static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, int nr_hw_queues) { struct request_queue *q; - LIST_HEAD(head); int prev_nr_hw_queues = set->nr_hw_queues; + unsigned int memflags; int i; lockdep_assert_held(&set->tag_list_lock); @@ -4730,30 +4983,26 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, if (set->nr_maps == 1 && nr_hw_queues == set->nr_hw_queues) return; - list_for_each_entry(q, &set->tag_list, tag_set_list) - blk_mq_freeze_queue(q); - /* - * Switch IO scheduler to 'none', cleaning up the data associated - * with the previous scheduler. We will switch back once we are done - * updating the new sw to hw queue mappings. - */ - list_for_each_entry(q, &set->tag_list, tag_set_list) - if (!blk_mq_elv_switch_none(&head, q)) - goto switch_back; - + memflags = memalloc_noio_save(); list_for_each_entry(q, &set->tag_list, tag_set_list) { blk_mq_debugfs_unregister_hctxs(q); blk_mq_sysfs_unregister_hctxs(q); } - if (blk_mq_realloc_tag_set_tags(set, nr_hw_queues) < 0) + list_for_each_entry(q, &set->tag_list, tag_set_list) + blk_mq_freeze_queue_nomemsave(q); + + if (blk_mq_realloc_tag_set_tags(set, nr_hw_queues) < 0) { + list_for_each_entry(q, &set->tag_list, tag_set_list) + blk_mq_unfreeze_queue_nomemrestore(q); goto reregister; + } fallback: blk_mq_update_queue_map(set); list_for_each_entry(q, &set->tag_list, tag_set_list) { - blk_mq_realloc_hw_ctxs(set, q); - blk_mq_update_poll_flag(q); + __blk_mq_realloc_hw_ctxs(set, q); + if (q->nr_hw_queues != set->nr_hw_queues) { int i = prev_nr_hw_queues; @@ -4768,18 +5017,19 @@ fallback: blk_mq_map_swqueue(q); } + /* elv_update_nr_hw_queues() unfreeze queue for us */ + list_for_each_entry(q, &set->tag_list, tag_set_list) + elv_update_nr_hw_queues(q); + reregister: list_for_each_entry(q, &set->tag_list, tag_set_list) { blk_mq_sysfs_register_hctxs(q); blk_mq_debugfs_register_hctxs(q); - } -switch_back: - list_for_each_entry(q, &set->tag_list, tag_set_list) - blk_mq_elv_switch_back(&head, q); - - list_for_each_entry(q, &set->tag_list, tag_set_list) - blk_mq_unfreeze_queue(q); + blk_mq_remove_hw_queues_cpuhp(q); + blk_mq_add_hw_queues_cpuhp(q); + } + memalloc_noio_restore(memflags); /* Free the excess tags when nr_hw_queues shrink. */ for (i = set->nr_hw_queues; i < prev_nr_hw_queues; i++) @@ -4788,9 +5038,11 @@ switch_back: void blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, int nr_hw_queues) { + down_write(&set->update_nr_hwq_lock); mutex_lock(&set->tag_list_lock); __blk_mq_update_nr_hw_queues(set, nr_hw_queues); mutex_unlock(&set->tag_list_lock); + up_write(&set->update_nr_hwq_lock); } EXPORT_SYMBOL_GPL(blk_mq_update_nr_hw_queues); @@ -4824,9 +5076,9 @@ static int blk_hctx_poll(struct request_queue *q, struct blk_mq_hw_ctx *hctx, int blk_mq_poll(struct request_queue *q, blk_qc_t cookie, struct io_comp_batch *iob, unsigned int flags) { - struct blk_mq_hw_ctx *hctx = xa_load(&q->hctx_table, cookie); - - return blk_hctx_poll(q, hctx, iob, flags); + if (!blk_mq_can_poll(q)) + return 0; + return blk_hctx_poll(q, xa_load(&q->hctx_table, cookie), iob, flags); } int blk_rq_poll(struct request *rq, struct io_comp_batch *iob, |