diff options
Diffstat (limited to 'block')
-rw-r--r-- | block/Kconfig | 26 | ||||
-rw-r--r-- | block/Makefile | 3 | ||||
-rw-r--r-- | block/badblocks.c | 29 | ||||
-rw-r--r-- | block/bio.c | 19 | ||||
-rw-r--r-- | block/blk-cgroup.c | 9 | ||||
-rw-r--r-- | block/blk-core.c | 132 | ||||
-rw-r--r-- | block/blk-flush.c | 37 | ||||
-rw-r--r-- | block/blk-lib.c | 175 | ||||
-rw-r--r-- | block/blk-map.c | 4 | ||||
-rw-r--r-- | block/blk-merge.c | 81 | ||||
-rw-r--r-- | block/blk-mq-cpumap.c | 1 | ||||
-rw-r--r-- | block/blk-mq-sysfs.c | 47 | ||||
-rw-r--r-- | block/blk-mq.c | 350 | ||||
-rw-r--r-- | block/blk-mq.h | 5 | ||||
-rw-r--r-- | block/blk-settings.c | 34 | ||||
-rw-r--r-- | block/blk-stat.c | 256 | ||||
-rw-r--r-- | block/blk-stat.h | 42 | ||||
-rw-r--r-- | block/blk-sysfs.c | 162 | ||||
-rw-r--r-- | block/blk-wbt.c | 750 | ||||
-rw-r--r-- | block/blk-wbt.h | 171 | ||||
-rw-r--r-- | block/blk.h | 1 | ||||
-rw-r--r-- | block/bsg-lib.c | 25 | ||||
-rw-r--r-- | block/cfq-iosched.c | 16 | ||||
-rw-r--r-- | block/elevator.c | 8 | ||||
-rw-r--r-- | block/partition-generic.c | 65 |
25 files changed, 2189 insertions, 259 deletions
diff --git a/block/Kconfig b/block/Kconfig index 3a024440a669..8bf114a3858a 100644 --- a/block/Kconfig +++ b/block/Kconfig @@ -121,6 +121,32 @@ config BLK_CMDLINE_PARSER See Documentation/block/cmdline-partition.txt for more information. +config BLK_WBT + bool "Enable support for block device writeback throttling" + default n + ---help--- + Enabling this option enables the block layer to throttle buffered + background writeback from the VM, making it more smooth and having + less impact on foreground operations. The throttling is done + dynamically on an algorithm loosely based on CoDel, factoring in + the realtime performance of the disk. + +config BLK_WBT_SQ + bool "Single queue writeback throttling" + default n + depends on BLK_WBT + ---help--- + Enable writeback throttling by default on legacy single queue devices + +config BLK_WBT_MQ + bool "Multiqueue writeback throttling" + default y + depends on BLK_WBT + ---help--- + Enable writeback throttling by default on multiqueue devices. + Multiqueue currently doesn't have support for IO scheduling, + enabling this option is recommended. + menu "Partition Types" source "block/partitions/Kconfig" diff --git a/block/Makefile b/block/Makefile index 934dac73fb37..a827f988c4e6 100644 --- a/block/Makefile +++ b/block/Makefile @@ -5,7 +5,7 @@ obj-$(CONFIG_BLOCK) := bio.o elevator.o blk-core.o blk-tag.o blk-sysfs.o \ blk-flush.o blk-settings.o blk-ioc.o blk-map.o \ blk-exec.o blk-merge.o blk-softirq.o blk-timeout.o \ - blk-lib.o blk-mq.o blk-mq-tag.o \ + blk-lib.o blk-mq.o blk-mq-tag.o blk-stat.o \ blk-mq-sysfs.o blk-mq-cpumap.o ioctl.o \ genhd.o scsi_ioctl.o partition-generic.o ioprio.o \ badblocks.o partitions/ @@ -24,3 +24,4 @@ obj-$(CONFIG_BLK_CMDLINE_PARSER) += cmdline-parser.o obj-$(CONFIG_BLK_DEV_INTEGRITY) += bio-integrity.o blk-integrity.o t10-pi.o obj-$(CONFIG_BLK_MQ_PCI) += blk-mq-pci.o obj-$(CONFIG_BLK_DEV_ZONED) += blk-zoned.o +obj-$(CONFIG_BLK_WBT) += blk-wbt.o diff --git a/block/badblocks.c b/block/badblocks.c index 7be53cb1cc3c..6ebcef282314 100644 --- a/block/badblocks.c +++ b/block/badblocks.c @@ -133,6 +133,26 @@ retry: } EXPORT_SYMBOL_GPL(badblocks_check); +static void badblocks_update_acked(struct badblocks *bb) +{ + u64 *p = bb->page; + int i; + bool unacked = false; + + if (!bb->unacked_exist) + return; + + for (i = 0; i < bb->count ; i++) { + if (!BB_ACK(p[i])) { + unacked = true; + break; + } + } + + if (!unacked) + bb->unacked_exist = 0; +} + /** * badblocks_set() - Add a range of bad blocks to the table. * @bb: the badblocks structure that holds all badblock information @@ -294,6 +314,8 @@ int badblocks_set(struct badblocks *bb, sector_t s, int sectors, bb->changed = 1; if (!acknowledged) bb->unacked_exist = 1; + else + badblocks_update_acked(bb); write_sequnlock_irqrestore(&bb->lock, flags); return rv; @@ -354,7 +376,8 @@ int badblocks_clear(struct badblocks *bb, sector_t s, int sectors) * current range. Earlier ranges could also overlap, * but only this one can overlap the end of the range. */ - if (BB_OFFSET(p[lo]) + BB_LEN(p[lo]) > target) { + if ((BB_OFFSET(p[lo]) + BB_LEN(p[lo]) > target) && + (BB_OFFSET(p[lo]) < target)) { /* Partial overlap, leave the tail of this range */ int ack = BB_ACK(p[lo]); sector_t a = BB_OFFSET(p[lo]); @@ -377,7 +400,8 @@ int badblocks_clear(struct badblocks *bb, sector_t s, int sectors) lo--; } while (lo >= 0 && - BB_OFFSET(p[lo]) + BB_LEN(p[lo]) > s) { + (BB_OFFSET(p[lo]) + BB_LEN(p[lo]) > s) && + (BB_OFFSET(p[lo]) < target)) { /* This range does overlap */ if (BB_OFFSET(p[lo]) < s) { /* Keep the early parts of this range. */ @@ -399,6 +423,7 @@ int badblocks_clear(struct badblocks *bb, sector_t s, int sectors) } } + badblocks_update_acked(bb); bb->changed = 1; out: write_sequnlock_irq(&bb->lock); diff --git a/block/bio.c b/block/bio.c index 2cf6ebabc68c..2b375020fc49 100644 --- a/block/bio.c +++ b/block/bio.c @@ -270,11 +270,15 @@ static void bio_free(struct bio *bio) } } -void bio_init(struct bio *bio) +void bio_init(struct bio *bio, struct bio_vec *table, + unsigned short max_vecs) { memset(bio, 0, sizeof(*bio)); atomic_set(&bio->__bi_remaining, 1); atomic_set(&bio->__bi_cnt, 1); + + bio->bi_io_vec = table; + bio->bi_max_vecs = max_vecs; } EXPORT_SYMBOL(bio_init); @@ -480,7 +484,7 @@ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) return NULL; bio = p + front_pad; - bio_init(bio); + bio_init(bio, NULL, 0); if (nr_iovecs > inline_vecs) { unsigned long idx = 0; @@ -670,6 +674,7 @@ struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, switch (bio_op(bio)) { case REQ_OP_DISCARD: case REQ_OP_SECURE_ERASE: + case REQ_OP_WRITE_ZEROES: break; case REQ_OP_WRITE_SAME: bio->bi_io_vec[bio->bi_vcnt++] = bio_src->bi_io_vec[0]; @@ -1835,15 +1840,7 @@ struct bio *bio_split(struct bio *bio, int sectors, BUG_ON(sectors <= 0); BUG_ON(sectors >= bio_sectors(bio)); - /* - * Discards need a mutable bio_vec to accommodate the payload - * required by the DSM TRIM and UNMAP commands. - */ - if (bio_op(bio) == REQ_OP_DISCARD || bio_op(bio) == REQ_OP_SECURE_ERASE) - split = bio_clone_bioset(bio, gfp, bs); - else - split = bio_clone_fast(bio, gfp, bs); - + split = bio_clone_fast(bio, gfp, bs); if (!split) return NULL; diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index b08ccbb9393a..8ba0af780e88 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -185,7 +185,8 @@ static struct blkcg_gq *blkg_create(struct blkcg *blkcg, } wb_congested = wb_congested_get_create(&q->backing_dev_info, - blkcg->css.id, GFP_NOWAIT); + blkcg->css.id, + GFP_NOWAIT | __GFP_NOWARN); if (!wb_congested) { ret = -ENOMEM; goto err_put_css; @@ -193,7 +194,7 @@ static struct blkcg_gq *blkg_create(struct blkcg *blkcg, /* allocate */ if (!new_blkg) { - new_blkg = blkg_alloc(blkcg, q, GFP_NOWAIT); + new_blkg = blkg_alloc(blkcg, q, GFP_NOWAIT | __GFP_NOWARN); if (unlikely(!new_blkg)) { ret = -ENOMEM; goto err_put_congested; @@ -1022,7 +1023,7 @@ blkcg_css_alloc(struct cgroup_subsys_state *parent_css) } spin_lock_init(&blkcg->lock); - INIT_RADIX_TREE(&blkcg->blkg_tree, GFP_NOWAIT); + INIT_RADIX_TREE(&blkcg->blkg_tree, GFP_NOWAIT | __GFP_NOWARN); INIT_HLIST_HEAD(&blkcg->blkg_list); #ifdef CONFIG_CGROUP_WRITEBACK INIT_LIST_HEAD(&blkcg->cgwb_list); @@ -1240,7 +1241,7 @@ pd_prealloc: if (blkg->pd[pol->plid]) continue; - pd = pol->pd_alloc_fn(GFP_NOWAIT, q->node); + pd = pol->pd_alloc_fn(GFP_NOWAIT | __GFP_NOWARN, q->node); if (!pd) swap(pd, pd_prealloc); if (!pd) { diff --git a/block/blk-core.c b/block/blk-core.c index 2deca48a4a05..61ba08c58b64 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -39,6 +39,7 @@ #include "blk.h" #include "blk-mq.h" +#include "blk-wbt.h" EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_remap); EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_remap); @@ -882,6 +883,7 @@ blk_init_allocated_queue(struct request_queue *q, request_fn_proc *rfn, fail: blk_free_flush_queue(q->fq); + wbt_exit(q); return NULL; } EXPORT_SYMBOL(blk_init_allocated_queue); @@ -1152,6 +1154,7 @@ static struct request *__get_request(struct request_list *rl, unsigned int op, blk_rq_init(q, rq); blk_rq_set_rl(rq, rl); + blk_rq_set_prio(rq, ioc); rq->cmd_flags = op; rq->rq_flags = rq_flags; @@ -1344,6 +1347,7 @@ void blk_requeue_request(struct request_queue *q, struct request *rq) blk_delete_timer(rq); blk_clear_rq_complete(rq); trace_block_rq_requeue(q, rq); + wbt_requeue(q->rq_wb, &rq->issue_stat); if (rq->rq_flags & RQF_QUEUED) blk_queue_end_tag(q, rq); @@ -1436,6 +1440,8 @@ void __blk_put_request(struct request_queue *q, struct request *req) /* this is a bio leak */ WARN_ON(req->bio != NULL); + wbt_done(q->rq_wb, &req->issue_stat); + /* * Request may not have originated from ll_rw_blk. if not, * it didn't come out of our reserved rq pools @@ -1470,38 +1476,6 @@ void blk_put_request(struct request *req) } EXPORT_SYMBOL(blk_put_request); -/** - * blk_add_request_payload - add a payload to a request - * @rq: request to update - * @page: page backing the payload - * @offset: offset in page - * @len: length of the payload. - * - * This allows to later add a payload to an already submitted request by - * a block driver. The driver needs to take care of freeing the payload - * itself. - * - * Note that this is a quite horrible hack and nothing but handling of - * discard requests should ever use it. - */ -void blk_add_request_payload(struct request *rq, struct page *page, - int offset, unsigned int len) -{ - struct bio *bio = rq->bio; - - bio->bi_io_vec->bv_page = page; - bio->bi_io_vec->bv_offset = offset; - bio->bi_io_vec->bv_len = len; - - bio->bi_iter.bi_size = len; - bio->bi_vcnt = 1; - bio->bi_phys_segments = 1; - - rq->__data_len = rq->resid_len = len; - rq->nr_phys_segments = 1; -} -EXPORT_SYMBOL_GPL(blk_add_request_payload); - bool bio_attempt_back_merge(struct request_queue *q, struct request *req, struct bio *bio) { @@ -1653,7 +1627,8 @@ void init_request_from_bio(struct request *req, struct bio *bio) req->errors = 0; req->__sector = bio->bi_iter.bi_sector; - req->ioprio = bio_prio(bio); + if (ioprio_valid(bio_prio(bio))) + req->ioprio = bio_prio(bio); blk_rq_bio_prep(req->q, req, bio); } @@ -1663,6 +1638,7 @@ static blk_qc_t blk_queue_bio(struct request_queue *q, struct bio *bio) int el_ret, where = ELEVATOR_INSERT_SORT; struct request *req; unsigned int request_count = 0; + unsigned int wb_acct; /* * low level driver can indicate that it wants pages above a @@ -1715,17 +1691,22 @@ static blk_qc_t blk_queue_bio(struct request_queue *q, struct bio *bio) } get_rq: + wb_acct = wbt_wait(q->rq_wb, bio, q->queue_lock); + /* * Grab a free request. This is might sleep but can not fail. * Returns with the queue unlocked. */ req = get_request(q, bio->bi_opf, bio, GFP_NOIO); if (IS_ERR(req)) { + __wbt_done(q->rq_wb, wb_acct); bio->bi_error = PTR_ERR(req); bio_endio(bio); goto out_unlock; } + wbt_track(&req->issue_stat, wb_acct); + /* * After dropping the lock and possibly sleeping here, our request * may now be mergeable after it had proven unmergeable (above). @@ -1742,8 +1723,11 @@ get_rq: /* * If this is the first request added after a plug, fire * of a plug trace. + * + * @request_count may become stale because of schedule + * out, so check plug list again. */ - if (!request_count) + if (!request_count || list_empty(&plug->list)) trace_block_plug(q); else { struct request *last = list_entry_rq(plug->list.prev); @@ -1773,7 +1757,12 @@ static inline void blk_partition_remap(struct bio *bio) { struct block_device *bdev = bio->bi_bdev; - if (bio_sectors(bio) && bdev != bdev->bd_contains) { + /* + * Zone reset does not include bi_size so bio_sectors() is always 0. + * Include a test for the reset op code and perform the remap if needed. + */ + if (bdev != bdev->bd_contains && + (bio_sectors(bio) || bio_op(bio) == REQ_OP_ZONE_RESET)) { struct hd_struct *p = bdev->bd_part; bio->bi_iter.bi_sector += p->start_sect; @@ -1926,11 +1915,16 @@ generic_make_request_checks(struct bio *bio) case REQ_OP_WRITE_SAME: if (!bdev_write_same(bio->bi_bdev)) goto not_supported; + break; case REQ_OP_ZONE_REPORT: case REQ_OP_ZONE_RESET: if (!bdev_is_zoned(bio->bi_bdev)) goto not_supported; break; + case REQ_OP_WRITE_ZEROES: + if (!bdev_write_zeroes_sectors(bio->bi_bdev)) + goto not_supported; + break; default: break; } @@ -2464,6 +2458,12 @@ void blk_start_request(struct request *req) { blk_dequeue_request(req); + if (test_bit(QUEUE_FLAG_STATS, &req->q->queue_flags)) { + blk_stat_set_issue_time(&req->issue_stat); + req->rq_flags |= RQF_STATS; + wbt_issue(req->q->rq_wb, &req->issue_stat); + } + /* * We are now handing the request to the hardware, initialize * resid_len to full count and add the timeout handler. @@ -2612,6 +2612,8 @@ bool blk_update_request(struct request *req, int error, unsigned int nr_bytes) return false; } + WARN_ON_ONCE(req->rq_flags & RQF_SPECIAL_PAYLOAD); + req->__data_len -= total_bytes; /* update sector only for requests with clear definition of sector */ @@ -2683,8 +2685,13 @@ EXPORT_SYMBOL_GPL(blk_unprep_request); */ void blk_finish_request(struct request *req, int error) { + struct request_queue *q = req->q; + + if (req->rq_flags & RQF_STATS) + blk_stat_add(&q->rq_stats[rq_data_dir(req)], req); + if (req->rq_flags & RQF_QUEUED) - blk_queue_end_tag(req->q, req); + blk_queue_end_tag(q, req); BUG_ON(blk_queued_rq(req)); @@ -2698,13 +2705,14 @@ void blk_finish_request(struct request *req, int error) blk_account_io_done(req); - if (req->end_io) + if (req->end_io) { + wbt_done(req->q->rq_wb, &req->issue_stat); req->end_io(req, error); - else { + } else { if (blk_bidi_rq(req)) __blk_put_request(req->next_rq->q, req->next_rq); - __blk_put_request(req->q, req); + __blk_put_request(q, req); } } EXPORT_SYMBOL(blk_finish_request); @@ -3289,52 +3297,6 @@ void blk_finish_plug(struct blk_plug *plug) } EXPORT_SYMBOL(blk_finish_plug); -bool blk_poll(struct request_queue *q, blk_qc_t cookie) -{ - struct blk_plug *plug; - long state; - unsigned int queue_num; - struct blk_mq_hw_ctx *hctx; - - if (!q->mq_ops || !q->mq_ops->poll || !blk_qc_t_valid(cookie) || - !test_bit(QUEUE_FLAG_POLL, &q->queue_flags)) - return false; - - queue_num = blk_qc_t_to_queue_num(cookie); - hctx = q->queue_hw_ctx[queue_num]; - hctx->poll_considered++; - - plug = current->plug; - if (plug) - blk_flush_plug_list(plug, false); - - state = current->state; - while (!need_resched()) { - int ret; - - hctx->poll_invoked++; - - ret = q->mq_ops->poll(hctx, blk_qc_t_to_tag(cookie)); - if (ret > 0) { - hctx->poll_success++; - set_current_state(TASK_RUNNING); - return true; - } - - if (signal_pending_state(state, current)) - set_current_state(TASK_RUNNING); - - if (current->state == TASK_RUNNING) - return true; - if (ret < 0) - break; - cpu_relax(); - } - - return false; -} -EXPORT_SYMBOL_GPL(blk_poll); - #ifdef CONFIG_PM /** * blk_pm_runtime_init - Block layer runtime PM initialization routine diff --git a/block/blk-flush.c b/block/blk-flush.c index c486b7aa62ee..20b7c7a02f1c 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -341,6 +341,34 @@ static void flush_data_end_io(struct request *rq, int error) struct blk_flush_queue *fq = blk_get_flush_queue(q, NULL); /* + * Updating q->in_flight[] here for making this tag usable + * early. Because in blk_queue_start_tag(), + * q->in_flight[BLK_RW_ASYNC] is used to limit async I/O and + * reserve tags for sync I/O. + * + * More importantly this way can avoid the following I/O + * deadlock: + * + * - suppose there are 40 fua requests comming to flush queue + * and queue depth is 31 + * - 30 rqs are scheduled then blk_queue_start_tag() can't alloc + * tag for async I/O any more + * - all the 30 rqs are completed before FLUSH_PENDING_TIMEOUT + * and flush_data_end_io() is called + * - the other rqs still can't go ahead if not updating + * q->in_flight[BLK_RW_ASYNC] here, meantime these rqs + * are held in flush data queue and make no progress of + * handling post flush rq + * - only after the post flush rq is handled, all these rqs + * can be completed + */ + + elv_completed_request(q, rq); + + /* for avoiding double accounting */ + rq->rq_flags &= ~RQF_STARTED; + + /* * After populating an empty queue, kick it to avoid stall. Read * the comment in flush_end_io(). */ @@ -396,6 +424,13 @@ void blk_insert_flush(struct request *rq) rq->cmd_flags &= ~REQ_FUA; /* + * REQ_PREFLUSH|REQ_FUA implies REQ_SYNC, so if we clear any + * of those flags, we have to set REQ_SYNC to avoid skewing + * the request accounting. + */ + rq->cmd_flags |= REQ_SYNC; + + /* * An empty flush handed down from a stacking driver may * translate into nothing if the underlying device does not * advertise a write-back cache. In this case, simply @@ -419,7 +454,7 @@ void blk_insert_flush(struct request *rq) if ((policy & REQ_FSEQ_DATA) && !(policy & (REQ_FSEQ_PREFLUSH | REQ_FSEQ_POSTFLUSH))) { if (q->mq_ops) { - blk_mq_insert_request(rq, false, false, true); + blk_mq_insert_request(rq, false, true, false); } else list_add_tail(&rq->queuelist, &q->queue_head); return; diff --git a/block/blk-lib.c b/block/blk-lib.c index 18abda862915..ed89c8f4b2a0 100644 --- a/block/blk-lib.c +++ b/block/blk-lib.c @@ -80,7 +80,7 @@ int __blkdev_issue_discard(struct block_device *bdev, sector_t sector, req_sects = end_sect - sector; } - bio = next_bio(bio, 1, gfp_mask); + bio = next_bio(bio, 0, gfp_mask); bio->bi_iter.bi_sector = sector; bio->bi_bdev = bdev; bio_set_op_attrs(bio, op, 0); @@ -137,24 +137,24 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector, EXPORT_SYMBOL(blkdev_issue_discard); /** - * blkdev_issue_write_same - queue a write same operation + * __blkdev_issue_write_same - generate number of bios with same page * @bdev: target blockdev * @sector: start sector * @nr_sects: number of sectors to write * @gfp_mask: memory allocation flags (for bio_alloc) * @page: page containing data to write + * @biop: pointer to anchor bio * * Description: - * Issue a write same request for the sectors in question. + * Generate and issue number of bios(REQ_OP_WRITE_SAME) with same page. */ -int blkdev_issue_write_same(struct block_device *bdev, sector_t sector, - sector_t nr_sects, gfp_t gfp_mask, - struct page *page) +static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector, + sector_t nr_sects, gfp_t gfp_mask, struct page *page, + struct bio **biop) { struct request_queue *q = bdev_get_queue(bdev); unsigned int max_write_same_sectors; - struct bio *bio = NULL; - int ret = 0; + struct bio *bio = *biop; sector_t bs_mask; if (!q) @@ -164,6 +164,9 @@ int blkdev_issue_write_same(struct block_device *bdev, sector_t sector, if ((sector | nr_sects) & bs_mask) return -EINVAL; + if (!bdev_write_same(bdev)) + return -EOPNOTSUPP; + /* Ensure that max_write_same_sectors doesn't overflow bi_size */ max_write_same_sectors = UINT_MAX >> 9; @@ -185,32 +188,112 @@ int blkdev_issue_write_same(struct block_device *bdev, sector_t sector, bio->bi_iter.bi_size = nr_sects << 9; nr_sects = 0; } + cond_resched(); } - if (bio) { + *biop = bio; + return 0; +} + +/** + * blkdev_issue_write_same - queue a write same operation + * @bdev: target blockdev + * @sector: start sector + * @nr_sects: number of sectors to write + * @gfp_mask: memory allocation flags (for bio_alloc) + * @page: page containing data + * + * Description: + * Issue a write same request for the sectors in question. + */ +int blkdev_issue_write_same(struct block_device *bdev, sector_t sector, + sector_t nr_sects, gfp_t gfp_mask, + struct page *page) +{ + struct bio *bio = NULL; + struct blk_plug plug; + int ret; + + blk_start_plug(&plug); + ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, page, + &bio); + if (ret == 0 && bio) { ret = submit_bio_wait(bio); bio_put(bio); } + blk_finish_plug(&plug); return ret; } EXPORT_SYMBOL(blkdev_issue_write_same); /** - * blkdev_issue_zeroout - generate number of zero filed write bios + * __blkdev_issue_write_zeroes - generate number of bios with WRITE ZEROES * @bdev: blockdev to issue * @sector: start sector * @nr_sects: number of sectors to write * @gfp_mask: memory allocation flags (for bio_alloc) + * @biop: pointer to anchor bio * * Description: - * Generate and issue number of bios with zerofiled pages. + * Generate and issue number of bios(REQ_OP_WRITE_ZEROES) with zerofiled pages. */ +static int __blkdev_issue_write_zeroes(struct block_device *bdev, + sector_t sector, sector_t nr_sects, gfp_t gfp_mask, + struct bio **biop) +{ + struct bio *bio = *biop; + unsigned int max_write_zeroes_sectors; + struct request_queue *q = bdev_get_queue(bdev); + + if (!q) + return -ENXIO; + + /* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */ + max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev); + + if (max_write_zeroes_sectors == 0) + return -EOPNOTSUPP; + + while (nr_sects) { + bio = next_bio(bio, 0, gfp_mask); + bio->bi_iter.bi_sector = sector; + bio->bi_bdev = bdev; + bio_set_op_attrs(bio, REQ_OP_WRITE_ZEROES, 0); -static int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector, - sector_t nr_sects, gfp_t gfp_mask) + if (nr_sects > max_write_zeroes_sectors) { + bio->bi_iter.bi_size = max_write_zeroes_sectors << 9; + nr_sects -= max_write_zeroes_sectors; + sector += max_write_zeroes_sectors; + } else { + bio->bi_iter.bi_size = nr_sects << 9; + nr_sects = 0; + } + cond_resched(); + } + + *biop = bio; + return 0; +} + +/** + * __blkdev_issue_zeroout - generate number of zero filed write bios + * @bdev: blockdev to issue + * @sector: start sector + * @nr_sects: number of sectors to write + * @gfp_mask: memory allocation flags (for bio_alloc) + * @biop: pointer to anchor bio + * @discard: discard flag + * + * Description: + * Generate and issue number of bios with zerofiled pages. + */ +int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector, + sector_t nr_sects, gfp_t gfp_mask, struct bio **biop, + bool discard) { int ret; - struct bio *bio = NULL; + int bi_size = 0; + struct bio *bio = *biop; unsigned int sz; sector_t bs_mask; @@ -218,6 +301,24 @@ static int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector, if ((sector | nr_sects) & bs_mask) return -EINVAL; + if (discard) { + ret = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, + BLKDEV_DISCARD_ZERO, biop); + if (ret == 0 || (ret && ret != -EOPNOTSUPP)) + goto out; + } + + ret = __blkdev_issue_write_zeroes(bdev, sector, nr_sects, gfp_mask, + biop); + if (ret == 0 || (ret && ret != -EOPNOTSUPP)) + goto out; + + ret = __blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, + ZERO_PAGE(0), biop); + if (ret == 0 || (ret && ret != -EOPNOTSUPP)) + goto out; + + ret = 0; while (nr_sects != 0) { bio = next_bio(bio, min(nr_sects, (sector_t)BIO_MAX_PAGES), gfp_mask); @@ -227,21 +328,20 @@ static int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector, while (nr_sects != 0) { sz = min((sector_t) PAGE_SIZE >> 9 , nr_sects); - ret = bio_add_page(bio, ZERO_PAGE(0), sz << 9, 0); - nr_sects -= ret >> 9; - sector += ret >> 9; - if (ret < (sz << 9)) + bi_size = bio_add_page(bio, ZERO_PAGE(0), sz << 9, 0); + nr_sects -= bi_size >> 9; + sector += bi_size >> 9; + if (bi_size < (sz << 9)) break; } + cond_resched(); } - if (bio) { - ret = submit_bio_wait(bio); - bio_put(bio); - return ret; - } - return 0; + *biop = bio; +out: + return ret; } +EXPORT_SYMBOL(__blkdev_issue_zeroout); /** * blkdev_issue_zeroout - zero-fill a block range @@ -258,26 +358,27 @@ static int __blkdev_issue_zeroout(struct block_device *bdev, sector_t sector, * the discard request fail, if the discard flag is not set, or if * discard_zeroes_data is not supported, this function will resort to * zeroing the blocks manually, thus provisioning (allocating, - * anchoring) them. If the block device supports the WRITE SAME command - * blkdev_issue_zeroout() will use it to optimize the process of + * anchoring) them. If the block device supports WRITE ZEROES or WRITE SAME + * command(s), blkdev_issue_zeroout() will use it to optimize the process of * clearing the block range. Otherwise the zeroing will be performed * using regular WRITE calls. */ - int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector, sector_t nr_sects, gfp_t gfp_mask, bool discard) { - if (discard) { - if (!blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, - BLKDEV_DISCARD_ZERO)) - return 0; - } + int ret; + struct bio *bio = NULL; + struct blk_plug plug; - if (bdev_write_same(bdev) && - blkdev_issue_write_same(bdev, sector, nr_sects, gfp_mask, - ZERO_PAGE(0)) == 0) - return 0; + blk_start_plug(&plug); + ret = __blkdev_issue_zeroout(bdev, sector, nr_sects, gfp_mask, + &bio, discard); + if (ret == 0 && bio) { + ret = submit_bio_wait(bio); + bio_put(bio); + } + blk_finish_plug(&plug); - return __blkdev_issue_zeroout(bdev, sector, nr_sects, gfp_mask); + return ret; } EXPORT_SYMBOL(blkdev_issue_zeroout); diff --git a/block/blk-map.c b/block/blk-map.c index 0173a72a8aa9..0acb6640ead7 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -120,6 +120,9 @@ int blk_rq_map_user_iov(struct request_queue *q, struct request *rq, struct iov_iter i; int ret; + if (!iter_is_iovec(iter)) + goto fail; + if (map_data) copy = true; else if (iov_iter_alignment(iter) & align) @@ -142,6 +145,7 @@ int blk_rq_map_user_iov(struct request_queue *q, struct request *rq, unmap_rq: __blk_rq_unmap_user(bio); +fail: rq->bio = NULL; return -EINVAL; } diff --git a/block/blk-merge.c b/block/blk-merge.c index fda6a12fc776..182398cb1524 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -199,6 +199,10 @@ void blk_queue_split(struct request_queue *q, struct bio **bio, case REQ_OP_SECURE_ERASE: split = blk_bio_discard_split(q, *bio, bs, &nsegs); break; + case REQ_OP_WRITE_ZEROES: + split = NULL; + nsegs = (*bio)->bi_phys_segments; + break; case REQ_OP_WRITE_SAME: split = blk_bio_write_same_split(q, *bio, bs, &nsegs); break; @@ -237,15 +241,14 @@ static unsigned int __blk_recalc_rq_segments(struct request_queue *q, if (!bio) return 0; - /* - * This should probably be returning 0, but blk_add_request_payload() - * (Christoph!!!!) - */ - if (bio_op(bio) == REQ_OP_DISCARD || bio_op(bio) == REQ_OP_SECURE_ERASE) - return 1; - - if (bio_op(bio) == REQ_OP_WRITE_SAME) + switch (bio_op(bio)) { + case REQ_OP_DISCARD: + case REQ_OP_SECURE_ERASE: + case REQ_OP_WRITE_ZEROES: + return 0; + case REQ_OP_WRITE_SAME: return 1; + } fbio = bio; cluster = blk_queue_cluster(q); @@ -402,38 +405,21 @@ new_segment: *bvprv = *bvec; } +static inline int __blk_bvec_map_sg(struct request_queue *q, struct bio_vec bv, + struct scatterlist *sglist, struct scatterlist **sg) +{ + *sg = sglist; + sg_set_page(*sg, bv.bv_page, bv.bv_len, bv.bv_offset); + return 1; +} + static int __blk_bios_map_sg(struct request_queue *q, struct bio *bio, struct scatterlist *sglist, struct scatterlist **sg) { struct bio_vec bvec, bvprv = { NULL }; struct bvec_iter iter; - int nsegs, cluster; - - nsegs = 0; - cluster = blk_queue_cluster(q); - - switch (bio_op(bio)) { - case REQ_OP_DISCARD: - case REQ_OP_SECURE_ERASE: - /* - * This is a hack - drivers should be neither modifying the - * biovec, nor relying on bi_vcnt - but because of - * blk_add_request_payload(), a discard bio may or may not have - * a payload we need to set up here (thank you Christoph) and - * bi_vcnt is really the only way of telling if we need to. - */ - if (!bio->bi_vcnt) - return 0; - /* Fall through */ - case REQ_OP_WRITE_SAME: - *sg = sglist; - bvec = bio_iovec(bio); - sg_set_page(*sg, bvec.bv_page, bvec.bv_len, bvec.bv_offset); - return 1; - default: - break; - } + int cluster = blk_queue_cluster(q), nsegs = 0; for_each_bio(bio) bio_for_each_segment(bvec, bio, iter) @@ -453,7 +439,11 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq, struct scatterlist *sg = NULL; int nsegs = 0; - if (rq->bio) + if (rq->rq_flags & RQF_SPECIAL_PAYLOAD) + nsegs = __blk_bvec_map_sg(q, rq->special_vec, sglist, &sg); + else if (rq->bio && bio_op(rq->bio) == REQ_OP_WRITE_SAME) + nsegs = __blk_bvec_map_sg(q, bio_iovec(rq->bio), sglist, &sg); + else if (rq->bio) nsegs = __blk_bios_map_sg(q, rq->bio, sglist, &sg); if (unlikely(rq->rq_flags & RQF_COPY_USER) && @@ -486,12 +476,19 @@ int blk_rq_map_sg(struct request_queue *q, struct request *rq, * Something must have been wrong if the figured number of * segment is bigger than number of req's physical segments */ - WARN_ON(nsegs > rq->nr_phys_segments); + WARN_ON(nsegs > blk_rq_nr_phys_segments(rq)); return nsegs; } EXPORT_SYMBOL(blk_rq_map_sg); +static void req_set_nomerge(struct request_queue *q, struct request *req) +{ + req->cmd_flags |= REQ_NOMERGE; + if (req == q->last_merge) + q->last_merge = NULL; +} + static inline int ll_new_hw_segment(struct request_queue *q, struct request *req, struct bio *bio) @@ -512,9 +509,7 @@ static inline int ll_new_hw_segment(struct request_queue *q, return 1; no_merge: - req->cmd_flags |= REQ_NOMERGE; - if (req == q->last_merge) - q->last_merge = NULL; + req_set_nomerge(q, req); return 0; } @@ -528,9 +523,7 @@ int ll_back_merge_fn(struct request_queue *q, struct request *req, return 0; if (blk_rq_sectors(req) + bio_sectors(bio) > blk_rq_get_max_sectors(req, blk_rq_pos(req))) { - req->cmd_flags |= REQ_NOMERGE; - if (req == q->last_merge) - q->last_merge = NULL; + req_set_nomerge(q, req); return 0; } if (!bio_flagged(req->biotail, BIO_SEG_VALID)) @@ -552,9 +545,7 @@ int ll_front_merge_fn(struct request_queue *q, struct request *req, return 0; if (blk_rq_sectors(req) + bio_sectors(bio) > blk_rq_get_max_sectors(req, bio->bi_iter.bi_sector)) { - req->cmd_flags |= REQ_NOMERGE; - if (req == q->last_merge) - q->last_merge = NULL; + req_set_nomerge(q, req); return 0; } if (!bio_flagged(bio, BIO_SEG_VALID)) diff --git a/block/blk-mq-cpumap.c b/block/blk-mq-cpumap.c index 19b1d9c5f07e..8e61e8640e17 100644 --- a/block/blk-mq-cpumap.c +++ b/block/blk-mq-cpumap.c @@ -87,6 +87,7 @@ int blk_mq_map_queues(struct blk_mq_tag_set *set) free_cpumask_var(cpus); return 0; } +EXPORT_SYMBOL_GPL(blk_mq_map_queues); /* * We have no quick way of doing reverse lookups. This is only used at diff --git a/block/blk-mq-sysfs.c b/block/blk-mq-sysfs.c index 01fb455d3377..eacd3af72099 100644 --- a/block/blk-mq-sysfs.c +++ b/block/blk-mq-sysfs.c @@ -259,6 +259,47 @@ static ssize_t blk_mq_hw_sysfs_cpus_show(struct blk_mq_hw_ctx *hctx, char *page) return ret; } +static void blk_mq_stat_clear(struct blk_mq_hw_ctx *hctx) +{ + struct blk_mq_ctx *ctx; + unsigned int i; + + hctx_for_each_ctx(hctx, ctx, i) { + blk_stat_init(&ctx->stat[BLK_STAT_READ]); + blk_stat_init(&ctx->stat[BLK_STAT_WRITE]); + } +} + +static ssize_t blk_mq_hw_sysfs_stat_store(struct blk_mq_hw_ctx *hctx, + const char *page, size_t count) +{ + blk_mq_stat_clear(hctx); + return count; +} + +static ssize_t print_stat(char *page, struct blk_rq_stat *stat, const char *pre) +{ + return sprintf(page, "%s samples=%llu, mean=%lld, min=%lld, max=%lld\n", + pre, (long long) stat->nr_samples, + (long long) stat->mean, (long long) stat->min, + (long long) stat->max); +} + +static ssize_t blk_mq_hw_sysfs_stat_show(struct blk_mq_hw_ctx *hctx, char *page) +{ + struct blk_rq_stat stat[2]; + ssize_t ret; + + blk_stat_init(&stat[BLK_STAT_READ]); + blk_stat_init(&stat[BLK_STAT_WRITE]); + + blk_hctx_stat_get(hctx, stat); + + ret = print_stat(page, &stat[BLK_STAT_READ], "read :"); + ret += print_stat(page + ret, &stat[BLK_STAT_WRITE], "write:"); + return ret; +} + static struct blk_mq_ctx_sysfs_entry blk_mq_sysfs_dispatched = { .attr = {.name = "dispatched", .mode = S_IRUGO }, .show = blk_mq_sysfs_dispatched_show, @@ -317,6 +358,11 @@ static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_poll = { .show = blk_mq_hw_sysfs_poll_show, .store = blk_mq_hw_sysfs_poll_store, }; +static struct blk_mq_hw_ctx_sysfs_entry blk_mq_hw_sysfs_stat = { + .attr = {.name = "stats", .mode = S_IRUGO | S_IWUSR }, + .show = blk_mq_hw_sysfs_stat_show, + .store = blk_mq_hw_sysfs_stat_store, +}; static struct attribute *default_hw_ctx_attrs[] = { &blk_mq_hw_sysfs_queued.attr, @@ -327,6 +373,7 @@ static struct attribute *default_hw_ctx_attrs[] = { &blk_mq_hw_sysfs_cpus.attr, &blk_mq_hw_sysfs_active.attr, &blk_mq_hw_sysfs_poll.attr, + &blk_mq_hw_sysfs_stat.attr, NULL, }; diff --git a/block/blk-mq.c b/block/blk-mq.c index 077c2416a955..d79fdc11b1ee 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -30,6 +30,8 @@ #include "blk.h" #include "blk-mq.h" #include "blk-mq-tag.h" +#include "blk-stat.h" +#include "blk-wbt.h" static DEFINE_MUTEX(all_q_mutex); static LIST_HEAD(all_q_list); @@ -325,9 +327,12 @@ static void __blk_mq_free_request(struct blk_mq_hw_ctx *hctx, if (rq->rq_flags & RQF_MQ_INFLIGHT) atomic_dec(&hctx->nr_active); + + wbt_done(q->rq_wb, &rq->issue_stat); rq->rq_flags = 0; clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags); + clear_bit(REQ_ATOM_POLL_SLEPT, &rq->atomic_flags); blk_mq_put_tag(hctx, ctx, tag); blk_queue_exit(q); } @@ -353,6 +358,7 @@ inline void __blk_mq_end_request(struct request *rq, int error) blk_account_io_done(rq); if (rq->end_io) { + wbt_done(rq->q->rq_wb, &rq->issue_stat); rq->end_io(rq, error); } else { if (unlikely(blk_bidi_rq(rq))) @@ -403,10 +409,27 @@ static void blk_mq_ipi_complete_request(struct request *rq) put_cpu(); } +static void blk_mq_stat_add(struct request *rq) +{ + if (rq->rq_flags & RQF_STATS) { + /* + * We could rq->mq_ctx here, but there's less of a risk + * of races if we have the completion event add the stats + * to the local software queue. + */ + struct blk_mq_ctx *ctx; + + ctx = __blk_mq_get_ctx(rq->q, raw_smp_processor_id()); + blk_stat_add(&ctx->stat[rq_data_dir(rq)], rq); + } +} + static void __blk_mq_complete_request(struct request *rq) { struct request_queue *q = rq->q; + blk_mq_stat_add(rq); + if (!q->softirq_done_fn) blk_mq_end_request(rq, rq->errors); else @@ -450,6 +473,12 @@ void blk_mq_start_request(struct request *rq) if (unlikely(blk_bidi_rq(rq))) rq->next_rq->resid_len = blk_rq_bytes(rq->next_rq); + if (test_bit(QUEUE_FLAG_STATS, &q->queue_flags)) { + blk_stat_set_issue_time(&rq->issue_stat); + rq->rq_flags |= RQF_STATS; + wbt_issue(q->rq_wb, &rq->issue_stat); + } + blk_add_timer(rq); /* @@ -485,6 +514,7 @@ static void __blk_mq_requeue_request(struct request *rq) struct request_queue *q = rq->q; trace_block_rq_requeue(q, rq); + wbt_requeue(q->rq_wb, &rq->issue_stat); if (test_and_clear_bit(REQ_ATOM_STARTED, &rq->atomic_flags)) { if (q->dma_drain_size && blk_rq_bytes(rq)) @@ -791,41 +821,13 @@ static inline unsigned int queued_to_index(unsigned int queued) return min(BLK_MQ_MAX_DISPATCH_ORDER - 1, ilog2(queued) + 1); } -/* - * Run this hardware queue, pulling any software queues mapped to it in. - * Note that this function currently has various problems around ordering - * of IO. In particular, we'd like FIFO behaviour on handling existing - * items on the hctx->dispatch list. Ignore that for now. - */ -static void blk_mq_process_rq_list(struct blk_mq_hw_ctx *hctx) +bool blk_mq_dispatch_rq_list(struct blk_mq_hw_ctx *hctx, struct list_head *list) { struct request_queue *q = hctx->queue; struct request *rq; - LIST_HEAD(rq_list); LIST_HEAD(driver_list); struct list_head *dptr; - int queued; - - if (unlikely(blk_mq_hctx_stopped(hctx))) - return; - - hctx->run++; - - /* - * Touch any software queue that has pending entries. - */ - flush_busy_ctxs(hctx, &rq_list); - - /* - * If we have previous entries on our dispatch list, grab them - * and stuff them at the front for more fair dispatch. - */ - if (!list_empty_careful(&hctx->dispatch)) { - spin_lock(&hctx->lock); - if (!list_empty(&hctx->dispatch)) - list_splice_init(&hctx->dispatch, &rq_list); - spin_unlock(&hctx->lock); - } + int queued, ret = BLK_MQ_RQ_QUEUE_OK; /* * Start off with dptr being NULL, so we start the first request @@ -837,16 +839,15 @@ static void blk_mq_process_rq_list(struct blk_mq_hw_ctx *hctx) * Now process all the entries, sending them to the driver. */ queued = 0; - while (!list_empty(&rq_list)) { + while (!list_empty(list)) { struct blk_mq_queue_data bd; - int ret; - rq = list_first_entry(&rq_list, struct request, queuelist); + rq = list_first_entry(list, struct request, queuelist); list_del_init(&rq->queuelist); bd.rq = rq; bd.list = dptr; - bd.last = list_empty(&rq_list); + bd.last = list_empty(list); ret = q->mq_ops->queue_rq(hctx, &bd); switch (ret) { @@ -854,7 +855,7 @@ static void blk_mq_process_rq_list(struct blk_mq_hw_ctx *hctx) queued++; break; case BLK_MQ_RQ_QUEUE_BUSY: - list_add(&rq->queuelist, &rq_list); + list_add(&rq->queuelist, list); __blk_mq_requeue_request(rq); break; default: @@ -872,7 +873,7 @@ static void blk_mq_process_rq_list(struct blk_mq_hw_ctx *hctx) * We've done the first request. If we have more than 1 * left in the list, set dptr to defer issue. */ - if (!dptr && rq_list.next != rq_list.prev) + if (!dptr && list->next != list->prev) dptr = &driver_list; } @@ -882,10 +883,11 @@ static void blk_mq_process_rq_list(struct blk_mq_hw_ctx *hctx) * Any items that need requeuing? Stuff them into hctx->dispatch, * that is where we will continue on next queue run. */ - if (!list_empty(&rq_list)) { + if (!list_empty(list)) { spin_lock(&hctx->lock); - list_splice(&rq_list, &hctx->dispatch); + list_splice(list, &hctx->dispatch); spin_unlock(&hctx->lock); + /* * the queue is expected stopped with BLK_MQ_RQ_QUEUE_BUSY, but * it's possible the queue is stopped and restarted again @@ -897,6 +899,43 @@ static void blk_mq_process_rq_list(struct blk_mq_hw_ctx *hctx) **/ blk_mq_run_hw_queue(hctx, true); } + + return ret != BLK_MQ_RQ_QUEUE_BUSY; +} + +/* + * Run this hardware queue, pulling any software queues mapped to it in. + * Note that this function currently has various problems around ordering + * of IO. In particular, we'd like FIFO behaviour on handling existing + * items on the hctx->dispatch list. Ignore that for now. + */ +static void blk_mq_process_rq_list(struct blk_mq_hw_ctx *hctx) +{ + LIST_HEAD(rq_list); + LIST_HEAD(driver_list); + + if (unlikely(blk_mq_hctx_stopped(hctx))) + return; + + hctx->run++; + + /* + * Touch any software queue that has pending entries. + */ + flush_busy_ctxs(hctx, &rq_list); + + /* + * If we have previous entries on our dispatch list, grab them + * and stuff them at the front for more fair dispatch. + */ + if (!list_empty_careful(&hctx->dispatch)) { + spin_lock(&hctx->lock); + if (!list_empty(&hctx->dispatch)) + list_splice_init(&hctx->dispatch, &rq_list); + spin_unlock(&hctx->lock); + } + + blk_mq_dispatch_rq_list(hctx, &rq_list); } static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx) @@ -929,7 +968,7 @@ static int blk_mq_hctx_next_cpu(struct blk_mq_hw_ctx *hctx) return WORK_CPU_UNBOUND; if (--hctx->next_cpu_batch <= 0) { - int cpu = hctx->next_cpu, next_cpu; + int next_cpu; next_cpu = cpumask_next(hctx->next_cpu, hctx->cpumask); if (next_cpu >= nr_cpu_ids) @@ -937,8 +976,6 @@ static int blk_mq_hctx_next_cpu(struct blk_mq_hw_ctx *hctx) hctx->next_cpu = next_cpu; hctx->next_cpu_batch = BLK_MQ_CPU_WORK_BATCH; - - return cpu; } return hctx->next_cpu; @@ -1036,18 +1073,23 @@ void blk_mq_start_hw_queues(struct request_queue *q) } EXPORT_SYMBOL(blk_mq_start_hw_queues); +void blk_mq_start_stopped_hw_queue(struct blk_mq_hw_ctx *hctx, bool async) +{ + if (!blk_mq_hctx_stopped(hctx)) + return; + + clear_bit(BLK_MQ_S_STOPPED, &hctx->state); + blk_mq_run_hw_queue(hctx, async); +} +EXPORT_SYMBOL_GPL(blk_mq_start_stopped_hw_queue); + void blk_mq_start_stopped_hw_queues(struct request_queue *q, bool async) { struct blk_mq_hw_ctx *hctx; int i; - queue_for_each_hw_ctx(q, hctx, i) { - if (!blk_mq_hctx_stopped(hctx)) - continue; - - clear_bit(BLK_MQ_S_STOPPED, &hctx->state); - blk_mq_run_hw_queue(hctx, async); - } + queue_for_each_hw_ctx(q, hctx, i) + blk_mq_start_stopped_hw_queue(hctx, async); } EXPORT_SYMBOL(blk_mq_start_stopped_hw_queues); @@ -1209,7 +1251,7 @@ static void blk_mq_bio_to_request(struct request *rq, struct bio *bio) { init_request_from_bio(rq, bio); - blk_account_io_start(rq, 1); + blk_account_io_start(rq, true); } static inline bool hctx_allow_merges(struct blk_mq_hw_ctx *hctx) @@ -1264,11 +1306,11 @@ static struct request *blk_mq_map_request(struct request_queue *q, return rq; } -static void blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, - struct request *rq, blk_qc_t *cookie) +static void blk_mq_try_issue_directly(struct request *rq, blk_qc_t *cookie) { int ret; struct request_queue *q = rq->q; + struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(q, rq->mq_ctx->cpu); struct blk_mq_queue_data bd = { .rq = rq, .list = NULL, @@ -1318,6 +1360,7 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio) struct blk_plug *plug; struct request *same_queue_rq = NULL; blk_qc_t cookie; + unsigned int wb_acct; blk_queue_bounce(q, &bio); @@ -1332,9 +1375,15 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio) blk_attempt_plug_merge(q, bio, &request_count, &same_queue_rq)) return BLK_QC_T_NONE; + wb_acct = wbt_wait(q->rq_wb, bio, NULL); + rq = blk_mq_map_request(q, bio, &data); - if (unlikely(!rq)) + if (unlikely(!rq)) { + __wbt_done(q->rq_wb, wb_acct); return BLK_QC_T_NONE; + } + + wbt_track(&rq->issue_stat, wb_acct); cookie = blk_tag_to_qc_t(rq->tag, data.hctx->queue_num); @@ -1380,11 +1429,11 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio) if (!(data.hctx->flags & BLK_MQ_F_BLOCKING)) { rcu_read_lock(); - blk_mq_try_issue_directly(data.hctx, old_rq, &cookie); + blk_mq_try_issue_directly(old_rq, &cookie); rcu_read_unlock(); } else { srcu_idx = srcu_read_lock(&data.hctx->queue_rq_srcu); - blk_mq_try_issue_directly(data.hctx, old_rq, &cookie); + blk_mq_try_issue_directly(old_rq, &cookie); srcu_read_unlock(&data.hctx->queue_rq_srcu, srcu_idx); } goto done; @@ -1418,6 +1467,7 @@ static blk_qc_t blk_sq_make_request(struct request_queue *q, struct bio *bio) struct blk_mq_alloc_data data; struct request *rq; blk_qc_t cookie; + unsigned int wb_acct; blk_queue_bounce(q, &bio); @@ -1434,9 +1484,15 @@ static blk_qc_t blk_sq_make_request(struct request_queue *q, struct bio *bio) } else request_count = blk_plug_queued_count(q); + wb_acct = wbt_wait(q->rq_wb, bio, NULL); + rq = blk_mq_map_request(q, bio, &data); - if (unlikely(!rq)) + if (unlikely(!rq)) { + __wbt_done(q->rq_wb, wb_acct); return BLK_QC_T_NONE; + } + + wbt_track(&rq->issue_stat, wb_acct); cookie = blk_tag_to_qc_t(rq->tag, data.hctx->queue_num); @@ -1456,6 +1512,13 @@ static blk_qc_t blk_sq_make_request(struct request_queue *q, struct bio *bio) struct request *last = NULL; blk_mq_bio_to_request(rq, bio); + + /* + * @request_count may become stale because of schedule + * out, so check the list again. + */ + if (list_empty(&plug->mq_list)) + request_count = 0; if (!request_count) trace_block_plug(q); else @@ -1786,6 +1849,8 @@ static void blk_mq_init_cpu_queues(struct request_queue *q, spin_lock_init(&__ctx->lock); INIT_LIST_HEAD(&__ctx->rq_list); __ctx->queue = q; + blk_stat_init(&__ctx->stat[BLK_STAT_READ]); + blk_stat_init(&__ctx->stat[BLK_STAT_WRITE]); /* If the cpu isn't online, the cpu is mapped to first hctx */ if (!cpu_online(i)) @@ -2081,6 +2146,11 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, */ q->nr_requests = set->queue_depth; + /* + * Default to classic polling + */ + q->poll_nsec = -1; + if (set->ops->complete) blk_queue_softirq_done(q, set->ops->complete); @@ -2116,6 +2186,8 @@ void blk_mq_free_queue(struct request_queue *q) list_del_init(&q->all_q_node); mutex_unlock(&all_q_mutex); + wbt_exit(q); + blk_mq_del_queue_tag_set(q); blk_mq_exit_hw_queues(q, set, set->nr_hw_queues); @@ -2162,16 +2234,9 @@ static void blk_mq_queue_reinit_work(void) */ list_for_each_entry(q, &all_q_list, all_q_node) blk_mq_freeze_queue_start(q); - list_for_each_entry(q, &all_q_list, all_q_node) { + list_for_each_entry(q, &all_q_list, all_q_node) blk_mq_freeze_queue_wait(q); - /* - * timeout handler can't touch hw queue during the - * reinitialization - */ - del_timer_sync(&q->timeout); - } - list_for_each_entry(q, &all_q_list, all_q_node) blk_mq_queue_reinit(q, &cpuhp_online_new); @@ -2416,6 +2481,165 @@ void blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set, int nr_hw_queues) } EXPORT_SYMBOL_GPL(blk_mq_update_nr_hw_queues); +static unsigned long blk_mq_poll_nsecs(struct request_queue *q, + struct blk_mq_hw_ctx *hctx, + struct request *rq) +{ + struct blk_rq_stat stat[2]; + unsigned long ret = 0; + + /* + * If stats collection isn't on, don't sleep but turn it on for + * future users + */ + if (!blk_stat_enable(q)) + return 0; + + /* + * We don't have to do this once per IO, should optimize this + * to just use the current window of stats until it changes + */ + memset(&stat, 0, sizeof(stat)); + blk_hctx_stat_get(hctx, stat); + + /* + * As an optimistic guess, use half of the mean service time + * for this type of request. We can (and should) make this smarter. + * For instance, if the completion latencies are tight, we can + * get closer than just half the mean. This is especially + * important on devices where the completion latencies are longer + * than ~10 usec. + */ + if (req_op(rq) == REQ_OP_READ && stat[BLK_STAT_READ].nr_samples) + ret = (stat[BLK_STAT_READ].mean + 1) / 2; + else if (req_op(rq) == REQ_OP_WRITE && stat[BLK_STAT_WRITE].nr_samples) + ret = (stat[BLK_STAT_WRITE].mean + 1) / 2; + + return ret; +} + +static bool blk_mq_poll_hybrid_sleep(struct request_queue *q, + struct blk_mq_hw_ctx *hctx, + struct request *rq) +{ + struct hrtimer_sleeper hs; + enum hrtimer_mode mode; + unsigned int nsecs; + ktime_t kt; + + if (test_bit(REQ_ATOM_POLL_SLEPT, &rq->atomic_flags)) + return false; + + /* + * poll_nsec can be: + * + * -1: don't ever hybrid sleep + * 0: use half of prev avg + * >0: use this specific value + */ + if (q->poll_nsec == -1) + return false; + else if (q->poll_nsec > 0) + nsecs = q->poll_nsec; + else + nsecs = blk_mq_poll_nsecs(q, hctx, rq); + + if (!nsecs) + return false; + + set_bit(REQ_ATOM_POLL_SLEPT, &rq->atomic_flags); + + /* + * This will be replaced with the stats tracking code, using + * 'avg_completion_time / 2' as the pre-sleep target. + */ + kt = ktime_set(0, nsecs); + + mode = HRTIMER_MODE_REL; + hrtimer_init_on_stack(&hs.timer, CLOCK_MONOTONIC, mode); + hrtimer_set_expires(&hs.timer, kt); + + hrtimer_init_sleeper(&hs, current); + do { + if (test_bit(REQ_ATOM_COMPLETE, &rq->atomic_flags)) + break; + set_current_state(TASK_UNINTERRUPTIBLE); + hrtimer_start_expires(&hs.timer, mode); + if (hs.task) + io_schedule(); + hrtimer_cancel(&hs.timer); + mode = HRTIMER_MODE_ABS; + } while (hs.task && !signal_pending(current)); + + __set_current_state(TASK_RUNNING); + destroy_hrtimer_on_stack(&hs.timer); + return true; +} + +static bool __blk_mq_poll(struct blk_mq_hw_ctx *hctx, struct request *rq) +{ + struct request_queue *q = hctx->queue; + long state; + + /* + * If we sleep, have the caller restart the poll loop to reset + * the state. Like for the other success return cases, the + * caller is responsible for checking if the IO completed. If + * the IO isn't complete, we'll get called again and will go + * straight to the busy poll loop. + */ + if (blk_mq_poll_hybrid_sleep(q, hctx, rq)) + return true; + + hctx->poll_considered++; + + state = current->state; + while (!need_resched()) { + int ret; + + hctx->poll_invoked++; + + ret = q->mq_ops->poll(hctx, rq->tag); + if (ret > 0) { + hctx->poll_success++; + set_current_state(TASK_RUNNING); + return true; + } + + if (signal_pending_state(state, current)) + set_current_state(TASK_RUNNING); + + if (current->state == TASK_RUNNING) + return true; + if (ret < 0) + break; + cpu_relax(); + } + + return false; +} + +bool blk_mq_poll(struct request_queue *q, blk_qc_t cookie) +{ + struct blk_mq_hw_ctx *hctx; + struct blk_plug *plug; + struct request *rq; + + if (!q->mq_ops || !q->mq_ops->poll || !blk_qc_t_valid(cookie) || + !test_bit(QUEUE_FLAG_POLL, &q->queue_flags)) + return false; + + plug = current->plug; + if (plug) + blk_flush_plug_list(plug, false); + + hctx = q->queue_hw_ctx[blk_qc_t_to_queue_num(cookie)]; + rq = blk_mq_tag_to_rq(hctx->tags, blk_qc_t_to_tag(cookie)); + + return __blk_mq_poll(hctx, rq); +} +EXPORT_SYMBOL_GPL(blk_mq_poll); + void blk_mq_disable_hotplug(void) { mutex_lock(&all_q_mutex); diff --git a/block/blk-mq.h b/block/blk-mq.h index ac772dac7ce8..63e9116cddbd 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -1,6 +1,8 @@ #ifndef INT_BLK_MQ_H #define INT_BLK_MQ_H +#include "blk-stat.h" + struct blk_mq_tag_set; struct blk_mq_ctx { @@ -18,6 +20,7 @@ struct blk_mq_ctx { /* incremented at completion time */ unsigned long ____cacheline_aligned_in_smp rq_completed[2]; + struct blk_rq_stat stat[2]; struct request_queue *queue; struct kobject kobj; @@ -28,6 +31,7 @@ void blk_mq_freeze_queue(struct request_queue *q); void blk_mq_free_queue(struct request_queue *q); int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr); void blk_mq_wake_waiters(struct request_queue *q); +bool blk_mq_dispatch_rq_list(struct blk_mq_hw_ctx *, struct list_head *); /* * CPU hotplug helpers @@ -38,7 +42,6 @@ void blk_mq_disable_hotplug(void); /* * CPU -> queue mappings */ -int blk_mq_map_queues(struct blk_mq_tag_set *set); extern int blk_mq_hw_queue_to_node(unsigned int *map, unsigned int); static inline struct blk_mq_hw_ctx *blk_mq_map_queue(struct request_queue *q, diff --git a/block/blk-settings.c b/block/blk-settings.c index 55369a65dea2..529e55f52a03 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -13,6 +13,7 @@ #include <linux/gfp.h> #include "blk.h" +#include "blk-wbt.h" unsigned long blk_max_low_pfn; EXPORT_SYMBOL(blk_max_low_pfn); @@ -95,6 +96,7 @@ void blk_set_default_limits(struct queue_limits *lim) lim->max_dev_sectors = 0; lim->chunk_sectors = 0; lim->max_write_same_sectors = 0; + lim->max_write_zeroes_sectors = 0; lim->max_discard_sectors = 0; lim->max_hw_discard_sectors = 0; lim->discard_granularity = 0; @@ -131,6 +133,7 @@ void blk_set_stacking_limits(struct queue_limits *lim) lim->max_sectors = UINT_MAX; lim->max_dev_sectors = UINT_MAX; lim->max_write_same_sectors = UINT_MAX; + lim->max_write_zeroes_sectors = UINT_MAX; } EXPORT_SYMBOL(blk_set_stacking_limits); @@ -250,6 +253,7 @@ void blk_queue_max_hw_sectors(struct request_queue *q, unsigned int max_hw_secto max_sectors = min_not_zero(max_hw_sectors, limits->max_dev_sectors); max_sectors = min_t(unsigned int, max_sectors, BLK_DEF_MAX_SECTORS); limits->max_sectors = max_sectors; + q->backing_dev_info.io_pages = max_sectors >> (PAGE_SHIFT - 9); } EXPORT_SYMBOL(blk_queue_max_hw_sectors); @@ -299,6 +303,19 @@ void blk_queue_max_write_same_sectors(struct request_queue *q, EXPORT_SYMBOL(blk_queue_max_write_same_sectors); /** + * blk_queue_max_write_zeroes_sectors - set max sectors for a single + * write zeroes + * @q: the request queue for the device + * @max_write_zeroes_sectors: maximum number of sectors to write per command + **/ +void blk_queue_max_write_zeroes_sectors(struct request_queue *q, + unsigned int max_write_zeroes_sectors) +{ + q->limits.max_write_zeroes_sectors = max_write_zeroes_sectors; +} +EXPORT_SYMBOL(blk_queue_max_write_zeroes_sectors); + +/** * blk_queue_max_segments - set max hw segments for a request for this queue * @q: the request queue for the device * @max_segments: max number of segments @@ -526,6 +543,8 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b, t->max_dev_sectors = min_not_zero(t->max_dev_sectors, b->max_dev_sectors); t->max_write_same_sectors = min(t->max_write_same_sectors, b->max_write_same_sectors); + t->max_write_zeroes_sectors = min(t->max_write_zeroes_sectors, + b->max_write_zeroes_sectors); t->bounce_pfn = min_not_zero(t->bounce_pfn, b->bounce_pfn); t->seg_boundary_mask = min_not_zero(t->seg_boundary_mask, @@ -837,6 +856,19 @@ void blk_queue_flush_queueable(struct request_queue *q, bool queueable) EXPORT_SYMBOL_GPL(blk_queue_flush_queueable); /** + * blk_set_queue_depth - tell the block layer about the device queue depth + * @q: the request queue for the device + * @depth: queue depth + * + */ +void blk_set_queue_depth(struct request_queue *q, unsigned int depth) +{ + q->queue_depth = depth; + wbt_set_queue_depth(q->rq_wb, depth); +} +EXPORT_SYMBOL(blk_set_queue_depth); + +/** * blk_queue_write_cache - configure queue's write cache * @q: the request queue for the device * @wc: write back cache on or off @@ -856,6 +888,8 @@ void blk_queue_write_cache(struct request_queue *q, bool wc, bool fua) else queue_flag_clear(QUEUE_FLAG_FUA, q); spin_unlock_irq(q->queue_lock); + + wbt_set_write_cache(q->rq_wb, test_bit(QUEUE_FLAG_WC, &q->queue_flags)); } EXPORT_SYMBOL_GPL(blk_queue_write_cache); diff --git a/block/blk-stat.c b/block/blk-stat.c new file mode 100644 index 000000000000..9b43efb8933f --- /dev/null +++ b/block/blk-stat.c @@ -0,0 +1,256 @@ +/* + * Block stat tracking code + * + * Copyright (C) 2016 Jens Axboe + */ +#include <linux/kernel.h> +#include <linux/blk-mq.h> + +#include "blk-stat.h" +#include "blk-mq.h" + +static void blk_stat_flush_batch(struct blk_rq_stat *stat) +{ + const s32 nr_batch = READ_ONCE(stat->nr_batch); + const s32 nr_samples = READ_ONCE(stat->nr_samples); + + if (!nr_batch) + return; + if (!nr_samples) + stat->mean = div64_s64(stat->batch, nr_batch); + else { + stat->mean = div64_s64((stat->mean * nr_samples) + + stat->batch, + nr_batch + nr_samples); + } + + stat->nr_samples += nr_batch; + stat->nr_batch = stat->batch = 0; +} + +static void blk_stat_sum(struct blk_rq_stat *dst, struct blk_rq_stat *src) +{ + if (!src->nr_samples) + return; + + blk_stat_flush_batch(src); + + dst->min = min(dst->min, src->min); + dst->max = max(dst->max, src->max); + + if (!dst->nr_samples) + dst->mean = src->mean; + else { + dst->mean = div64_s64((src->mean * src->nr_samples) + + (dst->mean * dst->nr_samples), + dst->nr_samples + src->nr_samples); + } + dst->nr_samples += src->nr_samples; +} + +static void blk_mq_stat_get(struct request_queue *q, struct blk_rq_stat *dst) +{ + struct blk_mq_hw_ctx *hctx; + struct blk_mq_ctx *ctx; + uint64_t latest = 0; + int i, j, nr; + + blk_stat_init(&dst[BLK_STAT_READ]); + blk_stat_init(&dst[BLK_STAT_WRITE]); + + nr = 0; + do { + uint64_t newest = 0; + + queue_for_each_hw_ctx(q, hctx, i) { + hctx_for_each_ctx(hctx, ctx, j) { + blk_stat_flush_batch(&ctx->stat[BLK_STAT_READ]); + blk_stat_flush_batch(&ctx->stat[BLK_STAT_WRITE]); + + if (!ctx->stat[BLK_STAT_READ].nr_samples && + !ctx->stat[BLK_STAT_WRITE].nr_samples) + continue; + if (ctx->stat[BLK_STAT_READ].time > newest) + newest = ctx->stat[BLK_STAT_READ].time; + if (ctx->stat[BLK_STAT_WRITE].time > newest) + newest = ctx->stat[BLK_STAT_WRITE].time; + } + } + + /* + * No samples + */ + if (!newest) + break; + + if (newest > latest) + latest = newest; + + queue_for_each_hw_ctx(q, hctx, i) { + hctx_for_each_ctx(hctx, ctx, j) { + if (ctx->stat[BLK_STAT_READ].time == newest) { + blk_stat_sum(&dst[BLK_STAT_READ], + &ctx->stat[BLK_STAT_READ]); + nr++; + } + if (ctx->stat[BLK_STAT_WRITE].time == newest) { + blk_stat_sum(&dst[BLK_STAT_WRITE], + &ctx->stat[BLK_STAT_WRITE]); + nr++; + } + } + } + /* + * If we race on finding an entry, just loop back again. + * Should be very rare. + */ + } while (!nr); + + dst[BLK_STAT_READ].time = dst[BLK_STAT_WRITE].time = latest; +} + +void blk_queue_stat_get(struct request_queue *q, struct blk_rq_stat *dst) +{ + if (q->mq_ops) + blk_mq_stat_get(q, dst); + else { + blk_stat_flush_batch(&q->rq_stats[BLK_STAT_READ]); + blk_stat_flush_batch(&q->rq_stats[BLK_STAT_WRITE]); + memcpy(&dst[BLK_STAT_READ], &q->rq_stats[BLK_STAT_READ], + sizeof(struct blk_rq_stat)); + memcpy(&dst[BLK_STAT_WRITE], &q->rq_stats[BLK_STAT_WRITE], + sizeof(struct blk_rq_stat)); + } +} + +void blk_hctx_stat_get(struct blk_mq_hw_ctx *hctx, struct blk_rq_stat *dst) +{ + struct blk_mq_ctx *ctx; + unsigned int i, nr; + + nr = 0; + do { + uint64_t newest = 0; + + hctx_for_each_ctx(hctx, ctx, i) { + blk_stat_flush_batch(&ctx->stat[BLK_STAT_READ]); + blk_stat_flush_batch(&ctx->stat[BLK_STAT_WRITE]); + + if (!ctx->stat[BLK_STAT_READ].nr_samples && + !ctx->stat[BLK_STAT_WRITE].nr_samples) + continue; + + if (ctx->stat[BLK_STAT_READ].time > newest) + newest = ctx->stat[BLK_STAT_READ].time; + if (ctx->stat[BLK_STAT_WRITE].time > newest) + newest = ctx->stat[BLK_STAT_WRITE].time; + } + + if (!newest) + break; + + hctx_for_each_ctx(hctx, ctx, i) { + if (ctx->stat[BLK_STAT_READ].time == newest) { + blk_stat_sum(&dst[BLK_STAT_READ], + &ctx->stat[BLK_STAT_READ]); + nr++; + } + if (ctx->stat[BLK_STAT_WRITE].time == newest) { + blk_stat_sum(&dst[BLK_STAT_WRITE], + &ctx->stat[BLK_STAT_WRITE]); + nr++; + } + } + /* + * If we race on finding an entry, just loop back again. + * Should be very rare, as the window is only updated + * occasionally + */ + } while (!nr); +} + +static void __blk_stat_init(struct blk_rq_stat *stat, s64 time_now) +{ + stat->min = -1ULL; + stat->max = stat->nr_samples = stat->mean = 0; + stat->batch = stat->nr_batch = 0; + stat->time = time_now & BLK_STAT_NSEC_MASK; +} + +void blk_stat_init(struct blk_rq_stat *stat) +{ + __blk_stat_init(stat, ktime_to_ns(ktime_get())); +} + +static bool __blk_stat_is_current(struct blk_rq_stat *stat, s64 now) +{ + return (now & BLK_STAT_NSEC_MASK) == (stat->time & BLK_STAT_NSEC_MASK); +} + +bool blk_stat_is_current(struct blk_rq_stat *stat) +{ + return __blk_stat_is_current(stat, ktime_to_ns(ktime_get())); +} + +void blk_stat_add(struct blk_rq_stat *stat, struct request *rq) +{ + s64 now, value; + + now = __blk_stat_time(ktime_to_ns(ktime_get())); + if (now < blk_stat_time(&rq->issue_stat)) + return; + + if (!__blk_stat_is_current(stat, now)) + __blk_stat_init(stat, now); + + value = now - blk_stat_time(&rq->issue_stat); + if (value > stat->max) + stat->max = value; + if (value < stat->min) + stat->min = value; + + if (stat->batch + value < stat->batch || + stat->nr_batch + 1 == BLK_RQ_STAT_BATCH) + blk_stat_flush_batch(stat); + + stat->batch += value; + stat->nr_batch++; +} + +void blk_stat_clear(struct request_queue *q) +{ + if (q->mq_ops) { + struct blk_mq_hw_ctx *hctx; + struct blk_mq_ctx *ctx; + int i, j; + + queue_for_each_hw_ctx(q, hctx, i) { + hctx_for_each_ctx(hctx, ctx, j) { + blk_stat_init(&ctx->stat[BLK_STAT_READ]); + blk_stat_init(&ctx->stat[BLK_STAT_WRITE]); + } + } + } else { + blk_stat_init(&q->rq_stats[BLK_STAT_READ]); + blk_stat_init(&q->rq_stats[BLK_STAT_WRITE]); + } +} + +void blk_stat_set_issue_time(struct blk_issue_stat *stat) +{ + stat->time = (stat->time & BLK_STAT_MASK) | + (ktime_to_ns(ktime_get()) & BLK_STAT_TIME_MASK); +} + +/* + * Enable stat tracking, return whether it was enabled + */ +bool blk_stat_enable(struct request_queue *q) +{ + if (!test_bit(QUEUE_FLAG_STATS, &q->queue_flags)) { + set_bit(QUEUE_FLAG_STATS, &q->queue_flags); + return false; + } + + return true; +} diff --git a/block/blk-stat.h b/block/blk-stat.h new file mode 100644 index 000000000000..a2050a0a5314 --- /dev/null +++ b/block/blk-stat.h @@ -0,0 +1,42 @@ +#ifndef BLK_STAT_H +#define BLK_STAT_H + +/* + * ~0.13s window as a power-of-2 (2^27 nsecs) + */ +#define BLK_STAT_NSEC 134217728ULL +#define BLK_STAT_NSEC_MASK ~(BLK_STAT_NSEC - 1) + +/* + * Upper 3 bits can be used elsewhere + */ +#define BLK_STAT_RES_BITS 3 +#define BLK_STAT_SHIFT (64 - BLK_STAT_RES_BITS) +#define BLK_STAT_TIME_MASK ((1ULL << BLK_STAT_SHIFT) - 1) +#define BLK_STAT_MASK ~BLK_STAT_TIME_MASK + +enum { + BLK_STAT_READ = 0, + BLK_STAT_WRITE, +}; + +void blk_stat_add(struct blk_rq_stat *, struct request *); +void blk_hctx_stat_get(struct blk_mq_hw_ctx *, struct blk_rq_stat *); +void blk_queue_stat_get(struct request_queue *, struct blk_rq_stat *); +void blk_stat_clear(struct request_queue *); +void blk_stat_init(struct blk_rq_stat *); +bool blk_stat_is_current(struct blk_rq_stat *); +void blk_stat_set_issue_time(struct blk_issue_stat *); +bool blk_stat_enable(struct request_queue *); + +static inline u64 __blk_stat_time(u64 time) +{ + return time & BLK_STAT_TIME_MASK; +} + +static inline u64 blk_stat_time(struct blk_issue_stat *stat) +{ + return __blk_stat_time(stat->time); +} + +#endif diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 488c2e28feb8..1dbce057592d 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -13,6 +13,7 @@ #include "blk.h" #include "blk-mq.h" +#include "blk-wbt.h" struct queue_sysfs_entry { struct attribute attr; @@ -41,6 +42,19 @@ queue_var_store(unsigned long *var, const char *page, size_t count) return count; } +static ssize_t queue_var_store64(s64 *var, const char *page) +{ + int err; + s64 v; + + err = kstrtos64(page, 10, &v); + if (err < 0) + return err; + + *var = v; + return 0; +} + static ssize_t queue_requests_show(struct request_queue *q, char *page) { return queue_var_show(q->nr_requests, (page)); @@ -197,6 +211,11 @@ static ssize_t queue_write_same_max_show(struct request_queue *q, char *page) (unsigned long long)q->limits.max_write_same_sectors << 9); } +static ssize_t queue_write_zeroes_max_show(struct request_queue *q, char *page) +{ + return sprintf(page, "%llu\n", + (unsigned long long)q->limits.max_write_zeroes_sectors << 9); +} static ssize_t queue_max_sectors_store(struct request_queue *q, const char *page, size_t count) @@ -217,6 +236,7 @@ queue_max_sectors_store(struct request_queue *q, const char *page, size_t count) spin_lock_irq(q->queue_lock); q->limits.max_sectors = max_sectors_kb << 1; + q->backing_dev_info.io_pages = max_sectors_kb >> (PAGE_SHIFT - 10); spin_unlock_irq(q->queue_lock); return ret; @@ -336,6 +356,38 @@ queue_rq_affinity_store(struct request_queue *q, const char *page, size_t count) return ret; } +static ssize_t queue_poll_delay_show(struct request_queue *q, char *page) +{ + int val; + + if (q->poll_nsec == -1) + val = -1; + else + val = q->poll_nsec / 1000; + + return sprintf(page, "%d\n", val); +} + +static ssize_t queue_poll_delay_store(struct request_queue *q, const char *page, + size_t count) +{ + int err, val; + + if (!q->mq_ops || !q->mq_ops->poll) + return -EINVAL; + + err = kstrtoint(page, 10, &val); + if (err < 0) + return err; + + if (val == -1) + q->poll_nsec = -1; + else + q->poll_nsec = val * 1000; + + return count; +} + static ssize_t queue_poll_show(struct request_queue *q, char *page) { return queue_var_show(test_bit(QUEUE_FLAG_POLL, &q->queue_flags), page); @@ -364,6 +416,50 @@ static ssize_t queue_poll_store(struct request_queue *q, const char *page, return ret; } +static ssize_t queue_wb_lat_show(struct request_queue *q, char *page) +{ + if (!q->rq_wb) + return -EINVAL; + + return sprintf(page, "%llu\n", div_u64(q->rq_wb->min_lat_nsec, 1000)); +} + +static ssize_t queue_wb_lat_store(struct request_queue *q, const char *page, + size_t count) +{ + struct rq_wb *rwb; + ssize_t ret; + s64 val; + + ret = queue_var_store64(&val, page); + if (ret < 0) + return ret; + if (val < -1) + return -EINVAL; + + rwb = q->rq_wb; + if (!rwb) { + ret = wbt_init(q); + if (ret) + return ret; + + rwb = q->rq_wb; + if (!rwb) + return -EINVAL; + } + + if (val == -1) + rwb->min_lat_nsec = wbt_default_latency_nsec(q); + else if (val >= 0) + rwb->min_lat_nsec = val * 1000ULL; + + if (rwb->enable_state == WBT_STATE_ON_DEFAULT) + rwb->enable_state = WBT_STATE_ON_MANUAL; + + wbt_update_limits(rwb); + return count; +} + static ssize_t queue_wc_show(struct request_queue *q, char *page) { if (test_bit(QUEUE_FLAG_WC, &q->queue_flags)) @@ -401,6 +497,26 @@ static ssize_t queue_dax_show(struct request_queue *q, char *page) return queue_var_show(blk_queue_dax(q), page); } +static ssize_t print_stat(char *page, struct blk_rq_stat *stat, const char *pre) +{ + return sprintf(page, "%s samples=%llu, mean=%lld, min=%lld, max=%lld\n", + pre, (long long) stat->nr_samples, + (long long) stat->mean, (long long) stat->min, + (long long) stat->max); +} + +static ssize_t queue_stats_show(struct request_queue *q, char *page) +{ + struct blk_rq_stat stat[2]; + ssize_t ret; + + blk_queue_stat_get(q, stat); + + ret = print_stat(page, &stat[BLK_STAT_READ], "read :"); + ret += print_stat(page + ret, &stat[BLK_STAT_WRITE], "write:"); + return ret; +} + static struct queue_sysfs_entry queue_requests_entry = { .attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR }, .show = queue_requests_show, @@ -501,6 +617,11 @@ static struct queue_sysfs_entry queue_write_same_max_entry = { .show = queue_write_same_max_show, }; +static struct queue_sysfs_entry queue_write_zeroes_max_entry = { + .attr = {.name = "write_zeroes_max_bytes", .mode = S_IRUGO }, + .show = queue_write_zeroes_max_show, +}; + static struct queue_sysfs_entry queue_nonrot_entry = { .attr = {.name = "rotational", .mode = S_IRUGO | S_IWUSR }, .show = queue_show_nonrot, @@ -542,6 +663,12 @@ static struct queue_sysfs_entry queue_poll_entry = { .store = queue_poll_store, }; +static struct queue_sysfs_entry queue_poll_delay_entry = { + .attr = {.name = "io_poll_delay", .mode = S_IRUGO | S_IWUSR }, + .show = queue_poll_delay_show, + .store = queue_poll_delay_store, +}; + static struct queue_sysfs_entry queue_wc_entry = { .attr = {.name = "write_cache", .mode = S_IRUGO | S_IWUSR }, .show = queue_wc_show, @@ -553,6 +680,17 @@ static struct queue_sysfs_entry queue_dax_entry = { .show = queue_dax_show, }; +static struct queue_sysfs_entry queue_stats_entry = { + .attr = {.name = "stats", .mode = S_IRUGO }, + .show = queue_stats_show, +}; + +static struct queue_sysfs_entry queue_wb_lat_entry = { + .attr = {.name = "wbt_lat_usec", .mode = S_IRUGO | S_IWUSR }, + .show = queue_wb_lat_show, + .store = queue_wb_lat_store, +}; + static struct attribute *default_attrs[] = { &queue_requests_entry.attr, &queue_ra_entry.attr, @@ -573,6 +711,7 @@ static struct attribute *default_attrs[] = { &queue_discard_max_hw_entry.attr, &queue_discard_zeroes_data_entry.attr, &queue_write_same_max_entry.attr, + &queue_write_zeroes_max_entry.attr, &queue_nonrot_entry.attr, &queue_zoned_entry.attr, &queue_nomerges_entry.attr, @@ -582,6 +721,9 @@ static struct attribute *default_attrs[] = { &queue_poll_entry.attr, &queue_wc_entry.attr, &queue_dax_entry.attr, + &queue_stats_entry.attr, + &queue_wb_lat_entry.attr, + &queue_poll_delay_entry.attr, NULL, }; @@ -656,6 +798,7 @@ static void blk_release_queue(struct kobject *kobj) struct request_queue *q = container_of(kobj, struct request_queue, kobj); + wbt_exit(q); bdi_exit(&q->backing_dev_info); blkcg_exit_queue(q); @@ -696,6 +839,23 @@ struct kobj_type blk_queue_ktype = { .release = blk_release_queue, }; +static void blk_wb_init(struct request_queue *q) +{ +#ifndef CONFIG_BLK_WBT_MQ + if (q->mq_ops) + return; +#endif +#ifndef CONFIG_BLK_WBT_SQ + if (q->request_fn) + return; +#endif + + /* + * If this fails, we don't get throttling + */ + wbt_init(q); +} + int blk_register_queue(struct gendisk *disk) { int ret; @@ -735,6 +895,8 @@ int blk_register_queue(struct gendisk *disk) if (q->mq_ops) blk_mq_register_dev(dev, q); + blk_wb_init(q); + if (!q->request_fn) return 0; diff --git a/block/blk-wbt.c b/block/blk-wbt.c new file mode 100644 index 000000000000..6e82769f4042 --- /dev/null +++ b/block/blk-wbt.c @@ -0,0 +1,750 @@ +/* + * buffered writeback throttling. loosely based on CoDel. We can't drop + * packets for IO scheduling, so the logic is something like this: + * + * - Monitor latencies in a defined window of time. + * - If the minimum latency in the above window exceeds some target, increment + * scaling step and scale down queue depth by a factor of 2x. The monitoring + * window is then shrunk to 100 / sqrt(scaling step + 1). + * - For any window where we don't have solid data on what the latencies + * look like, retain status quo. + * - If latencies look good, decrement scaling step. + * - If we're only doing writes, allow the scaling step to go negative. This + * will temporarily boost write performance, snapping back to a stable + * scaling step of 0 if reads show up or the heavy writers finish. Unlike + * positive scaling steps where we shrink the monitoring window, a negative + * scaling step retains the default step==0 window size. + * + * Copyright (C) 2016 Jens Axboe + * + */ +#include <linux/kernel.h> +#include <linux/blk_types.h> +#include <linux/slab.h> +#include <linux/backing-dev.h> +#include <linux/swap.h> + +#include "blk-wbt.h" + +#define CREATE_TRACE_POINTS +#include <trace/events/wbt.h> + +enum { + /* + * Default setting, we'll scale up (to 75% of QD max) or down (min 1) + * from here depending on device stats + */ + RWB_DEF_DEPTH = 16, + + /* + * 100msec window + */ + RWB_WINDOW_NSEC = 100 * 1000 * 1000ULL, + + /* + * Disregard stats, if we don't meet this minimum + */ + RWB_MIN_WRITE_SAMPLES = 3, + + /* + * If we have this number of consecutive windows with not enough + * information to scale up or down, scale up. + */ + RWB_UNKNOWN_BUMP = 5, +}; + +static inline bool rwb_enabled(struct rq_wb *rwb) +{ + return rwb && rwb->wb_normal != 0; +} + +/* + * Increment 'v', if 'v' is below 'below'. Returns true if we succeeded, + * false if 'v' + 1 would be bigger than 'below'. + */ +static bool atomic_inc_below(atomic_t *v, int below) +{ + int cur = atomic_read(v); + + for (;;) { + int old; + + if (cur >= below) + return false; + old = atomic_cmpxchg(v, cur, cur + 1); + if (old == cur) + break; + cur = old; + } + + return true; +} + +static void wb_timestamp(struct rq_wb *rwb, unsigned long *var) +{ + if (rwb_enabled(rwb)) { + const unsigned long cur = jiffies; + + if (cur != *var) + *var = cur; + } +} + +/* + * If a task was rate throttled in balance_dirty_pages() within the last + * second or so, use that to indicate a higher cleaning rate. + */ +static bool wb_recent_wait(struct rq_wb *rwb) +{ + struct bdi_writeback *wb = &rwb->queue->backing_dev_info.wb; + + return time_before(jiffies, wb->dirty_sleep + HZ); +} + +static inline struct rq_wait *get_rq_wait(struct rq_wb *rwb, bool is_kswapd) +{ + return &rwb->rq_wait[is_kswapd]; +} + +static void rwb_wake_all(struct rq_wb *rwb) +{ + int i; + + for (i = 0; i < WBT_NUM_RWQ; i++) { + struct rq_wait *rqw = &rwb->rq_wait[i]; + + if (waitqueue_active(&rqw->wait)) + wake_up_all(&rqw->wait); + } +} + +void __wbt_done(struct rq_wb *rwb, enum wbt_flags wb_acct) +{ + struct rq_wait *rqw; + int inflight, limit; + + if (!(wb_acct & WBT_TRACKED)) + return; + + rqw = get_rq_wait(rwb, wb_acct & WBT_KSWAPD); + inflight = atomic_dec_return(&rqw->inflight); + + /* + * wbt got disabled with IO in flight. Wake up any potential + * waiters, we don't have to do more than that. + */ + if (unlikely(!rwb_enabled(rwb))) { + rwb_wake_all(rwb); + return; + } + + /* + * If the device does write back caching, drop further down + * before we wake people up. + */ + if (rwb->wc && !wb_recent_wait(rwb)) + limit = 0; + else + limit = rwb->wb_normal; + + /* + * Don't wake anyone up if we are above the normal limit. + */ + if (inflight && inflight >= limit) + return; + + if (waitqueue_active(&rqw->wait)) { + int diff = limit - inflight; + + if (!inflight || diff >= rwb->wb_background / 2) + wake_up_all(&rqw->wait); + } +} + +/* + * Called on completion of a request. Note that it's also called when + * a request is merged, when the request gets freed. + */ +void wbt_done(struct rq_wb *rwb, struct blk_issue_stat *stat) +{ + if (!rwb) + return; + + if (!wbt_is_tracked(stat)) { + if (rwb->sync_cookie == stat) { + rwb->sync_issue = 0; + rwb->sync_cookie = NULL; + } + + if (wbt_is_read(stat)) + wb_timestamp(rwb, &rwb->last_comp); + wbt_clear_state(stat); + } else { + WARN_ON_ONCE(stat == rwb->sync_cookie); + __wbt_done(rwb, wbt_stat_to_mask(stat)); + wbt_clear_state(stat); + } +} + +/* + * Return true, if we can't increase the depth further by scaling + */ +static bool calc_wb_limits(struct rq_wb *rwb) +{ + unsigned int depth; + bool ret = false; + + if (!rwb->min_lat_nsec) { + rwb->wb_max = rwb->wb_normal = rwb->wb_background = 0; + return false; + } + + /* + * For QD=1 devices, this is a special case. It's important for those + * to have one request ready when one completes, so force a depth of + * 2 for those devices. On the backend, it'll be a depth of 1 anyway, + * since the device can't have more than that in flight. If we're + * scaling down, then keep a setting of 1/1/1. + */ + if (rwb->queue_depth == 1) { + if (rwb->scale_step > 0) + rwb->wb_max = rwb->wb_normal = 1; + else { + rwb->wb_max = rwb->wb_normal = 2; + ret = true; + } + rwb->wb_background = 1; + } else { + /* + * scale_step == 0 is our default state. If we have suffered + * latency spikes, step will be > 0, and we shrink the + * allowed write depths. If step is < 0, we're only doing + * writes, and we allow a temporarily higher depth to + * increase performance. + */ + depth = min_t(unsigned int, RWB_DEF_DEPTH, rwb->queue_depth); + if (rwb->scale_step > 0) + depth = 1 + ((depth - 1) >> min(31, rwb->scale_step)); + else if (rwb->scale_step < 0) { + unsigned int maxd = 3 * rwb->queue_depth / 4; + + depth = 1 + ((depth - 1) << -rwb->scale_step); + if (depth > maxd) { + depth = maxd; + ret = true; + } + } + + /* + * Set our max/normal/bg queue depths based on how far + * we have scaled down (->scale_step). + */ + rwb->wb_max = depth; + rwb->wb_normal = (rwb->wb_max + 1) / 2; + rwb->wb_background = (rwb->wb_max + 3) / 4; + } + + return ret; +} + +static inline bool stat_sample_valid(struct blk_rq_stat *stat) +{ + /* + * We need at least one read sample, and a minimum of + * RWB_MIN_WRITE_SAMPLES. We require some write samples to know + * that it's writes impacting us, and not just some sole read on + * a device that is in a lower power state. + */ + return stat[BLK_STAT_READ].nr_samples >= 1 && + stat[BLK_STAT_WRITE].nr_samples >= RWB_MIN_WRITE_SAMPLES; +} + +static u64 rwb_sync_issue_lat(struct rq_wb *rwb) +{ + u64 now, issue = ACCESS_ONCE(rwb->sync_issue); + + if (!issue || !rwb->sync_cookie) + return 0; + + now = ktime_to_ns(ktime_get()); + return now - issue; +} + +enum { + LAT_OK = 1, + LAT_UNKNOWN, + LAT_UNKNOWN_WRITES, + LAT_EXCEEDED, +}; + +static int __latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat) +{ + struct backing_dev_info *bdi = &rwb->queue->backing_dev_info; + u64 thislat; + + /* + * If our stored sync issue exceeds the window size, or it + * exceeds our min target AND we haven't logged any entries, + * flag the latency as exceeded. wbt works off completion latencies, + * but for a flooded device, a single sync IO can take a long time + * to complete after being issued. If this time exceeds our + * monitoring window AND we didn't see any other completions in that + * window, then count that sync IO as a violation of the latency. + */ + thislat = rwb_sync_issue_lat(rwb); + if (thislat > rwb->cur_win_nsec || + (thislat > rwb->min_lat_nsec && !stat[BLK_STAT_READ].nr_samples)) { + trace_wbt_lat(bdi, thislat); + return LAT_EXCEEDED; + } + + /* + * No read/write mix, if stat isn't valid + */ + if (!stat_sample_valid(stat)) { + /* + * If we had writes in this stat window and the window is + * current, we're only doing writes. If a task recently + * waited or still has writes in flights, consider us doing + * just writes as well. + */ + if ((stat[BLK_STAT_WRITE].nr_samples && blk_stat_is_current(stat)) || + wb_recent_wait(rwb) || wbt_inflight(rwb)) + return LAT_UNKNOWN_WRITES; + return LAT_UNKNOWN; + } + + /* + * If the 'min' latency exceeds our target, step down. + */ + if (stat[BLK_STAT_READ].min > rwb->min_lat_nsec) { + trace_wbt_lat(bdi, stat[BLK_STAT_READ].min); + trace_wbt_stat(bdi, stat); + return LAT_EXCEEDED; + } + + if (rwb->scale_step) + trace_wbt_stat(bdi, stat); + + return LAT_OK; +} + +static int latency_exceeded(struct rq_wb *rwb) +{ + struct blk_rq_stat stat[2]; + + blk_queue_stat_get(rwb->queue, stat); + return __latency_exceeded(rwb, stat); +} + +static void rwb_trace_step(struct rq_wb *rwb, const char *msg) +{ + struct backing_dev_info *bdi = &rwb->queue->backing_dev_info; + + trace_wbt_step(bdi, msg, rwb->scale_step, rwb->cur_win_nsec, + rwb->wb_background, rwb->wb_normal, rwb->wb_max); +} + +static void scale_up(struct rq_wb *rwb) +{ + /* + * Hit max in previous round, stop here + */ + if (rwb->scaled_max) + return; + + rwb->scale_step--; + rwb->unknown_cnt = 0; + blk_stat_clear(rwb->queue); + + rwb->scaled_max = calc_wb_limits(rwb); + + rwb_wake_all(rwb); + + rwb_trace_step(rwb, "step up"); +} + +/* + * Scale rwb down. If 'hard_throttle' is set, do it quicker, since we + * had a latency violation. + */ +static void scale_down(struct rq_wb *rwb, bool hard_throttle) +{ + /* + * Stop scaling down when we've hit the limit. This also prevents + * ->scale_step from going to crazy values, if the device can't + * keep up. + */ + if (rwb->wb_max == 1) + return; + + if (rwb->scale_step < 0 && hard_throttle) + rwb->scale_step = 0; + else + rwb->scale_step++; + + rwb->scaled_max = false; + rwb->unknown_cnt = 0; + blk_stat_clear(rwb->queue); + calc_wb_limits(rwb); + rwb_trace_step(rwb, "step down"); +} + +static void rwb_arm_timer(struct rq_wb *rwb) +{ + unsigned long expires; + + if (rwb->scale_step > 0) { + /* + * We should speed this up, using some variant of a fast + * integer inverse square root calculation. Since we only do + * this for every window expiration, it's not a huge deal, + * though. + */ + rwb->cur_win_nsec = div_u64(rwb->win_nsec << 4, + int_sqrt((rwb->scale_step + 1) << 8)); + } else { + /* + * For step < 0, we don't want to increase/decrease the + * window size. + */ + rwb->cur_win_nsec = rwb->win_nsec; + } + + expires = jiffies + nsecs_to_jiffies(rwb->cur_win_nsec); + mod_timer(&rwb->window_timer, expires); +} + +static void wb_timer_fn(unsigned long data) +{ + struct rq_wb *rwb = (struct rq_wb *) data; + unsigned int inflight = wbt_inflight(rwb); + int status; + + status = latency_exceeded(rwb); + + trace_wbt_timer(&rwb->queue->backing_dev_info, status, rwb->scale_step, + inflight); + + /* + * If we exceeded the latency target, step down. If we did not, + * step one level up. If we don't know enough to say either exceeded + * or ok, then don't do anything. + */ + switch (status) { + case LAT_EXCEEDED: + scale_down(rwb, true); + break; + case LAT_OK: + scale_up(rwb); + break; + case LAT_UNKNOWN_WRITES: + /* + * We started a the center step, but don't have a valid + * read/write sample, but we do have writes going on. + * Allow step to go negative, to increase write perf. + */ + scale_up(rwb); + break; + case LAT_UNKNOWN: + if (++rwb->unknown_cnt < RWB_UNKNOWN_BUMP) + break; + /* + * We get here when previously scaled reduced depth, and we + * currently don't have a valid read/write sample. For that + * case, slowly return to center state (step == 0). + */ + if (rwb->scale_step > 0) + scale_up(rwb); + else if (rwb->scale_step < 0) + scale_down(rwb, false); + break; + default: + break; + } + + /* + * Re-arm timer, if we have IO in flight + */ + if (rwb->scale_step || inflight) + rwb_arm_timer(rwb); +} + +void wbt_update_limits(struct rq_wb *rwb) +{ + rwb->scale_step = 0; + rwb->scaled_max = false; + calc_wb_limits(rwb); + + rwb_wake_all(rwb); +} + +static bool close_io(struct rq_wb *rwb) +{ + const unsigned long now = jiffies; + + return time_before(now, rwb->last_issue + HZ / 10) || + time_before(now, rwb->last_comp + HZ / 10); +} + +#define REQ_HIPRIO (REQ_SYNC | REQ_META | REQ_PRIO) + +static inline unsigned int get_limit(struct rq_wb *rwb, unsigned long rw) +{ + unsigned int limit; + + /* + * At this point we know it's a buffered write. If this is + * kswapd trying to free memory, or REQ_SYNC is set, set, then + * it's WB_SYNC_ALL writeback, and we'll use the max limit for + * that. If the write is marked as a background write, then use + * the idle limit, or go to normal if we haven't had competing + * IO for a bit. + */ + if ((rw & REQ_HIPRIO) || wb_recent_wait(rwb) || current_is_kswapd()) + limit = rwb->wb_max; + else if ((rw & REQ_BACKGROUND) || close_io(rwb)) { + /* + * If less than 100ms since we completed unrelated IO, + * limit us to half the depth for background writeback. + */ + limit = rwb->wb_background; + } else + limit = rwb->wb_normal; + + return limit; +} + +static inline bool may_queue(struct rq_wb *rwb, struct rq_wait *rqw, + wait_queue_t *wait, unsigned long rw) +{ + /* + * inc it here even if disabled, since we'll dec it at completion. + * this only happens if the task was sleeping in __wbt_wait(), + * and someone turned it off at the same time. + */ + if (!rwb_enabled(rwb)) { + atomic_inc(&rqw->inflight); + return true; + } + + /* + * If the waitqueue is already active and we are not the next + * in line to be woken up, wait for our turn. + */ + if (waitqueue_active(&rqw->wait) && + rqw->wait.task_list.next != &wait->task_list) + return false; + + return atomic_inc_below(&rqw->inflight, get_limit(rwb, rw)); +} + +/* + * Block if we will exceed our limit, or if we are currently waiting for + * the timer to kick off queuing again. + */ +static void __wbt_wait(struct rq_wb *rwb, unsigned long rw, spinlock_t *lock) +{ + struct rq_wait *rqw = get_rq_wait(rwb, current_is_kswapd()); + DEFINE_WAIT(wait); + + if (may_queue(rwb, rqw, &wait, rw)) + return; + + do { + prepare_to_wait_exclusive(&rqw->wait, &wait, + TASK_UNINTERRUPTIBLE); + + if (may_queue(rwb, rqw, &wait, rw)) + break; + + if (lock) + spin_unlock_irq(lock); + + io_schedule(); + + if (lock) + spin_lock_irq(lock); + } while (1); + + finish_wait(&rqw->wait, &wait); +} + +static inline bool wbt_should_throttle(struct rq_wb *rwb, struct bio *bio) +{ + const int op = bio_op(bio); + + /* + * If not a WRITE, do nothing + */ + if (op != REQ_OP_WRITE) + return false; + + /* + * Don't throttle WRITE_ODIRECT + */ + if ((bio->bi_opf & (REQ_SYNC | REQ_IDLE)) == (REQ_SYNC | REQ_IDLE)) + return false; + + return true; +} + +/* + * Returns true if the IO request should be accounted, false if not. + * May sleep, if we have exceeded the writeback limits. Caller can pass + * in an irq held spinlock, if it holds one when calling this function. + * If we do sleep, we'll release and re-grab it. + */ +unsigned int wbt_wait(struct rq_wb *rwb, struct bio *bio, spinlock_t *lock) +{ + unsigned int ret = 0; + + if (!rwb_enabled(rwb)) + return 0; + + if (bio_op(bio) == REQ_OP_READ) + ret = WBT_READ; + + if (!wbt_should_throttle(rwb, bio)) { + if (ret & WBT_READ) + wb_timestamp(rwb, &rwb->last_issue); + return ret; + } + + __wbt_wait(rwb, bio->bi_opf, lock); + + if (!timer_pending(&rwb->window_timer)) + rwb_arm_timer(rwb); + + if (current_is_kswapd()) + ret |= WBT_KSWAPD; + + return ret | WBT_TRACKED; +} + +void wbt_issue(struct rq_wb *rwb, struct blk_issue_stat *stat) +{ + if (!rwb_enabled(rwb)) + return; + + /* + * Track sync issue, in case it takes a long time to complete. Allows + * us to react quicker, if a sync IO takes a long time to complete. + * Note that this is just a hint. 'stat' can go away when the + * request completes, so it's important we never dereference it. We + * only use the address to compare with, which is why we store the + * sync_issue time locally. + */ + if (wbt_is_read(stat) && !rwb->sync_issue) { + rwb->sync_cookie = stat; + rwb->sync_issue = blk_stat_time(stat); + } +} + +void wbt_requeue(struct rq_wb *rwb, struct blk_issue_stat *stat) +{ + if (!rwb_enabled(rwb)) + return; + if (stat == rwb->sync_cookie) { + rwb->sync_issue = 0; + rwb->sync_cookie = NULL; + } +} + +void wbt_set_queue_depth(struct rq_wb *rwb, unsigned int depth) +{ + if (rwb) { + rwb->queue_depth = depth; + wbt_update_limits(rwb); + } +} + +void wbt_set_write_cache(struct rq_wb *rwb, bool write_cache_on) +{ + if (rwb) + rwb->wc = write_cache_on; +} + + /* + * Disable wbt, if enabled by default. Only called from CFQ, if we have + * cgroups enabled + */ +void wbt_disable_default(struct request_queue *q) +{ + struct rq_wb *rwb = q->rq_wb; + + if (rwb && rwb->enable_state == WBT_STATE_ON_DEFAULT) { + del_timer_sync(&rwb->window_timer); + rwb->win_nsec = rwb->min_lat_nsec = 0; + wbt_update_limits(rwb); + } +} +EXPORT_SYMBOL_GPL(wbt_disable_default); + +u64 wbt_default_latency_nsec(struct request_queue *q) +{ + /* + * We default to 2msec for non-rotational storage, and 75msec + * for rotational storage. + */ + if (blk_queue_nonrot(q)) + return 2000000ULL; + else + return 75000000ULL; +} + +int wbt_init(struct request_queue *q) +{ + struct rq_wb *rwb; + int i; + + /* + * For now, we depend on the stats window being larger than + * our monitoring window. Ensure that this isn't inadvertently + * violated. + */ + BUILD_BUG_ON(RWB_WINDOW_NSEC > BLK_STAT_NSEC); + BUILD_BUG_ON(WBT_NR_BITS > BLK_STAT_RES_BITS); + + rwb = kzalloc(sizeof(*rwb), GFP_KERNEL); + if (!rwb) + return -ENOMEM; + + for (i = 0; i < WBT_NUM_RWQ; i++) { + atomic_set(&rwb->rq_wait[i].inflight, 0); + init_waitqueue_head(&rwb->rq_wait[i].wait); + } + + setup_timer(&rwb->window_timer, wb_timer_fn, (unsigned long) rwb); + rwb->wc = 1; + rwb->queue_depth = RWB_DEF_DEPTH; + rwb->last_comp = rwb->last_issue = jiffies; + rwb->queue = q; + rwb->win_nsec = RWB_WINDOW_NSEC; + rwb->enable_state = WBT_STATE_ON_DEFAULT; + wbt_update_limits(rwb); + + /* + * Assign rwb, and turn on stats tracking for this queue + */ + q->rq_wb = rwb; + blk_stat_enable(q); + + rwb->min_lat_nsec = wbt_default_latency_nsec(q); + + wbt_set_queue_depth(rwb, blk_queue_depth(q)); + wbt_set_write_cache(rwb, test_bit(QUEUE_FLAG_WC, &q->queue_flags)); + + return 0; +} + +void wbt_exit(struct request_queue *q) +{ + struct rq_wb *rwb = q->rq_wb; + + if (rwb) { + del_timer_sync(&rwb->window_timer); + q->rq_wb = NULL; + kfree(rwb); + } +} diff --git a/block/blk-wbt.h b/block/blk-wbt.h new file mode 100644 index 000000000000..65f1de519f67 --- /dev/null +++ b/block/blk-wbt.h @@ -0,0 +1,171 @@ +#ifndef WB_THROTTLE_H +#define WB_THROTTLE_H + +#include <linux/kernel.h> +#include <linux/atomic.h> +#include <linux/wait.h> +#include <linux/timer.h> +#include <linux/ktime.h> + +#include "blk-stat.h" + +enum wbt_flags { + WBT_TRACKED = 1, /* write, tracked for throttling */ + WBT_READ = 2, /* read */ + WBT_KSWAPD = 4, /* write, from kswapd */ + + WBT_NR_BITS = 3, /* number of bits */ +}; + +enum { + WBT_NUM_RWQ = 2, +}; + +/* + * Enable states. Either off, or on by default (done at init time), + * or on through manual setup in sysfs. + */ +enum { + WBT_STATE_ON_DEFAULT = 1, + WBT_STATE_ON_MANUAL = 2, +}; + +static inline void wbt_clear_state(struct blk_issue_stat *stat) +{ + stat->time &= BLK_STAT_TIME_MASK; +} + +static inline enum wbt_flags wbt_stat_to_mask(struct blk_issue_stat *stat) +{ + return (stat->time & BLK_STAT_MASK) >> BLK_STAT_SHIFT; +} + +static inline void wbt_track(struct blk_issue_stat *stat, enum wbt_flags wb_acct) +{ + stat->time |= ((u64) wb_acct) << BLK_STAT_SHIFT; +} + +static inline bool wbt_is_tracked(struct blk_issue_stat *stat) +{ + return (stat->time >> BLK_STAT_SHIFT) & WBT_TRACKED; +} + +static inline bool wbt_is_read(struct blk_issue_stat *stat) +{ + return (stat->time >> BLK_STAT_SHIFT) & WBT_READ; +} + +struct rq_wait { + wait_queue_head_t wait; + atomic_t inflight; +}; + +struct rq_wb { + /* + * Settings that govern how we throttle + */ + unsigned int wb_background; /* background writeback */ + unsigned int wb_normal; /* normal writeback */ + unsigned int wb_max; /* max throughput writeback */ + int scale_step; + bool scaled_max; + + short enable_state; /* WBT_STATE_* */ + + /* + * Number of consecutive periods where we don't have enough + * information to make a firm scale up/down decision. + */ + unsigned int unknown_cnt; + + u64 win_nsec; /* default window size */ + u64 cur_win_nsec; /* current window size */ + + struct timer_list window_timer; + + s64 sync_issue; + void *sync_cookie; + + unsigned int wc; + unsigned int queue_depth; + + unsigned long last_issue; /* last non-throttled issue */ + unsigned long last_comp; /* last non-throttled comp */ + unsigned long min_lat_nsec; + struct request_queue *queue; + struct rq_wait rq_wait[WBT_NUM_RWQ]; +}; + +static inline unsigned int wbt_inflight(struct rq_wb *rwb) +{ + unsigned int i, ret = 0; + + for (i = 0; i < WBT_NUM_RWQ; i++) + ret += atomic_read(&rwb->rq_wait[i].inflight); + + return ret; +} + +#ifdef CONFIG_BLK_WBT + +void __wbt_done(struct rq_wb *, enum wbt_flags); +void wbt_done(struct rq_wb *, struct blk_issue_stat *); +enum wbt_flags wbt_wait(struct rq_wb *, struct bio *, spinlock_t *); +int wbt_init(struct request_queue *); +void wbt_exit(struct request_queue *); +void wbt_update_limits(struct rq_wb *); +void wbt_requeue(struct rq_wb *, struct blk_issue_stat *); +void wbt_issue(struct rq_wb *, struct blk_issue_stat *); +void wbt_disable_default(struct request_queue *); + +void wbt_set_queue_depth(struct rq_wb *, unsigned int); +void wbt_set_write_cache(struct rq_wb *, bool); + +u64 wbt_default_latency_nsec(struct request_queue *); + +#else + +static inline void __wbt_done(struct rq_wb *rwb, enum wbt_flags flags) +{ +} +static inline void wbt_done(struct rq_wb *rwb, struct blk_issue_stat *stat) +{ +} +static inline enum wbt_flags wbt_wait(struct rq_wb *rwb, struct bio *bio, + spinlock_t *lock) +{ + return 0; +} +static inline int wbt_init(struct request_queue *q) +{ + return -EINVAL; +} +static inline void wbt_exit(struct request_queue *q) +{ +} +static inline void wbt_update_limits(struct rq_wb *rwb) +{ +} +static inline void wbt_requeue(struct rq_wb *rwb, struct blk_issue_stat *stat) +{ +} +static inline void wbt_issue(struct rq_wb *rwb, struct blk_issue_stat *stat) +{ +} +static inline void wbt_disable_default(struct request_queue *q) +{ +} +static inline void wbt_set_queue_depth(struct rq_wb *rwb, unsigned int depth) +{ +} +static inline void wbt_set_write_cache(struct rq_wb *rwb, bool wc) +{ +} +static inline u64 wbt_default_latency_nsec(struct request_queue *q) +{ + return 0; +} + +#endif /* CONFIG_BLK_WBT */ + +#endif diff --git a/block/blk.h b/block/blk.h index aa132dea598c..041185e5f129 100644 --- a/block/blk.h +++ b/block/blk.h @@ -111,6 +111,7 @@ void blk_account_io_done(struct request *req); enum rq_atomic_flags { REQ_ATOM_COMPLETE = 0, REQ_ATOM_STARTED, + REQ_ATOM_POLL_SLEPT, }; /* diff --git a/block/bsg-lib.c b/block/bsg-lib.c index 650f427d915b..9d652a992316 100644 --- a/block/bsg-lib.c +++ b/block/bsg-lib.c @@ -32,8 +32,13 @@ * bsg_destroy_job - routine to teardown/delete a bsg job * @job: bsg_job that is to be torn down */ -static void bsg_destroy_job(struct bsg_job *job) +static void bsg_destroy_job(struct kref *kref) { + struct bsg_job *job = container_of(kref, struct bsg_job, kref); + struct request *rq = job->req; + + blk_end_request_all(rq, rq->errors); + put_device(job->dev); /* release reference for the request */ kfree(job->request_payload.sg_list); @@ -41,6 +46,18 @@ static void bsg_destroy_job(struct bsg_job *job) kfree(job); } +void bsg_job_put(struct bsg_job *job) +{ + kref_put(&job->kref, bsg_destroy_job); +} +EXPORT_SYMBOL_GPL(bsg_job_put); + +int bsg_job_get(struct bsg_job *job) +{ + return kref_get_unless_zero(&job->kref); +} +EXPORT_SYMBOL_GPL(bsg_job_get); + /** * bsg_job_done - completion routine for bsg requests * @job: bsg_job that is complete @@ -83,8 +100,7 @@ static void bsg_softirq_done(struct request *rq) { struct bsg_job *job = rq->special; - blk_end_request_all(rq, rq->errors); - bsg_destroy_job(job); + bsg_job_put(job); } static int bsg_map_buffer(struct bsg_buffer *buf, struct request *req) @@ -142,6 +158,7 @@ static int bsg_create_job(struct device *dev, struct request *req) job->dev = dev; /* take a reference for the request */ get_device(job->dev); + kref_init(&job->kref); return 0; failjob_rls_rqst_payload: @@ -161,6 +178,8 @@ failjob_rls_job: * Drivers/subsys should pass this to the queue init function. */ void bsg_request_fn(struct request_queue *q) + __releases(q->queue_lock) + __acquires(q->queue_lock) { struct device *dev = q->queuedata; struct request *req; diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index dcbed8c9c82c..c73a6fcaeb9d 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -16,6 +16,7 @@ #include <linux/blktrace_api.h> #include <linux/blk-cgroup.h> #include "blk.h" +#include "blk-wbt.h" /* * tunables @@ -1586,7 +1587,7 @@ static struct blkcg_policy_data *cfq_cpd_alloc(gfp_t gfp) { struct cfq_group_data *cgd; - cgd = kzalloc(sizeof(*cgd), GFP_KERNEL); + cgd = kzalloc(sizeof(*cgd), gfp); if (!cgd) return NULL; return &cgd->cpd; @@ -3762,9 +3763,11 @@ static void check_blkcg_changed(struct cfq_io_cq *cic, struct bio *bio) struct cfq_data *cfqd = cic_to_cfqd(cic); struct cfq_queue *cfqq; uint64_t serial_nr; + bool nonroot_cg; rcu_read_lock(); serial_nr = bio_blkcg(bio)->css.serial_nr; + nonroot_cg = bio_blkcg(bio) != &blkcg_root; rcu_read_unlock(); /* @@ -3775,6 +3778,14 @@ static void check_blkcg_changed(struct cfq_io_cq *cic, struct bio *bio) return; /* + * If we have a non-root cgroup, we can depend on that to + * do proper throttling of writes. Turn off wbt for that + * case, if it was enabled by default. + */ + if (nonroot_cg) + wbt_disable_default(cfqd->queue); + + /* * Drop reference to queues. New queues will be assigned in new * group upon arrival of fresh requests. */ @@ -3845,7 +3856,8 @@ cfq_get_queue(struct cfq_data *cfqd, bool is_sync, struct cfq_io_cq *cic, goto out; } - cfqq = kmem_cache_alloc_node(cfq_pool, GFP_NOWAIT | __GFP_ZERO, + cfqq = kmem_cache_alloc_node(cfq_pool, + GFP_NOWAIT | __GFP_ZERO | __GFP_NOWARN, cfqd->queue->node); if (!cfqq) { cfqq = &cfqd->oom_cfqq; diff --git a/block/elevator.c b/block/elevator.c index a18a5db274e4..40f0c04e5ad3 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -248,13 +248,13 @@ static inline void __elv_rqhash_del(struct request *rq) rq->rq_flags &= ~RQF_HASHED; } -static void elv_rqhash_del(struct request_queue *q, struct request *rq) +void elv_rqhash_del(struct request_queue *q, struct request *rq) { if (ELV_ON_HASH(rq)) __elv_rqhash_del(rq); } -static void elv_rqhash_add(struct request_queue *q, struct request *rq) +void elv_rqhash_add(struct request_queue *q, struct request *rq) { struct elevator_queue *e = q->elevator; @@ -263,13 +263,13 @@ static void elv_rqhash_add(struct request_queue *q, struct request *rq) rq->rq_flags |= RQF_HASHED; } -static void elv_rqhash_reposition(struct request_queue *q, struct request *rq) +void elv_rqhash_reposition(struct request_queue *q, struct request *rq) { __elv_rqhash_del(rq); elv_rqhash_add(q, rq); } -static struct request *elv_rqhash_find(struct request_queue *q, sector_t offset) +struct request *elv_rqhash_find(struct request_queue *q, sector_t offset) { struct elevator_queue *e = q->elevator; struct hlist_node *next; diff --git a/block/partition-generic.c b/block/partition-generic.c index 71d9ed9df8da..d7beb6bbbf66 100644 --- a/block/partition-generic.c +++ b/block/partition-generic.c @@ -430,6 +430,56 @@ static int drop_partitions(struct gendisk *disk, struct block_device *bdev) return 0; } +static bool part_zone_aligned(struct gendisk *disk, + struct block_device *bdev, + sector_t from, sector_t size) +{ + unsigned int zone_size = bdev_zone_size(bdev); + + /* + * If this function is called, then the disk is a zoned block device + * (host-aware or host-managed). This can be detected even if the + * zoned block device support is disabled (CONFIG_BLK_DEV_ZONED not + * set). In this case, however, only host-aware devices will be seen + * as a block device is not created for host-managed devices. Without + * zoned block device support, host-aware drives can still be used as + * regular block devices (no zone operation) and their zone size will + * be reported as 0. Allow this case. + */ + if (!zone_size) + return true; + + /* + * Check partition start and size alignement. If the drive has a + * smaller last runt zone, ignore it and allow the partition to + * use it. Check the zone size too: it should be a power of 2 number + * of sectors. + */ + if (WARN_ON_ONCE(!is_power_of_2(zone_size))) { + u32 rem; + + div_u64_rem(from, zone_size, &rem); + if (rem) + return false; + if ((from + size) < get_capacity(disk)) { + div_u64_rem(size, zone_size, &rem); + if (rem) + return false; + } + + } else { + + if (from & (zone_size - 1)) + return false; + if ((from + size) < get_capacity(disk) && + (size & (zone_size - 1))) + return false; + + } + + return true; +} + int rescan_partitions(struct gendisk *disk, struct block_device *bdev) { struct parsed_partitions *state = NULL; @@ -529,6 +579,21 @@ rescan: } } + /* + * On a zoned block device, partitions should be aligned on the + * device zone size (i.e. zone boundary crossing not allowed). + * Otherwise, resetting the write pointer of the last zone of + * one partition may impact the following partition. + */ + if (bdev_is_zoned(bdev) && + !part_zone_aligned(disk, bdev, from, size)) { + printk(KERN_WARNING + "%s: p%d start %llu+%llu is not zone aligned\n", + disk->disk_name, p, (unsigned long long) from, + (unsigned long long) size); + continue; + } + part = add_partition(disk, p, from, size, state->parts[p].flags, &state->parts[p].info); |