diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2021-11-01 09:19:50 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2021-11-01 09:19:50 -0700 |
commit | 33c8846c814c1c27c6e33af005042d15061f948b (patch) | |
tree | da7c105b61758094d1d55ec1326ff28b521dbe9e /block | |
parent | 9ac211426fb6747c92d570647e2ce889e33cbffd (diff) | |
parent | 9b84c629c90374498ab5825dede74a06ea1c775b (diff) | |
download | linux-33c8846c814c1c27c6e33af005042d15061f948b.tar.bz2 |
Merge tag 'for-5.16/block-2021-10-29' of git://git.kernel.dk/linux-block
Pull block updates from Jens Axboe:
- mq-deadline accounting improvements (Bart)
- blk-wbt timer fix (Andrea)
- Untangle the block layer includes (Christoph)
- Rework the poll support to be bio based, which will enable adding
support for polling for bio based drivers (Christoph)
- Block layer core support for multi-actuator drives (Damien)
- blk-crypto improvements (Eric)
- Batched tag allocation support (me)
- Request completion batching support (me)
- Plugging improvements (me)
- Shared tag set improvements (John)
- Concurrent queue quiesce support (Ming)
- Cache bdev in ->private_data for block devices (Pavel)
- bdev dio improvements (Pavel)
- Block device invalidation and block size improvements (Xie)
- Various cleanups, fixes, and improvements (Christoph, Jackie,
Masahira, Tejun, Yu, Pavel, Zheng, me)
* tag 'for-5.16/block-2021-10-29' of git://git.kernel.dk/linux-block: (174 commits)
blk-mq-debugfs: Show active requests per queue for shared tags
block: improve readability of blk_mq_end_request_batch()
virtio-blk: Use blk_validate_block_size() to validate block size
loop: Use blk_validate_block_size() to validate block size
nbd: Use blk_validate_block_size() to validate block size
block: Add a helper to validate the block size
block: re-flow blk_mq_rq_ctx_init()
block: prefetch request to be initialized
block: pass in blk_mq_tags to blk_mq_rq_ctx_init()
block: add rq_flags to struct blk_mq_alloc_data
block: add async version of bio_set_polled
block: kill DIO_MULTI_BIO
block: kill unused polling bits in __blkdev_direct_IO()
block: avoid extra iter advance with async iocb
block: Add independent access ranges support
blk-mq: don't issue request directly in case that current is to be blocked
sbitmap: silence data race warning
blk-cgroup: synchronize blkg creation against policy deactivation
block: refactor bio_iov_bvec_set()
block: add single bio async direct IO helper
...
Diffstat (limited to 'block')
47 files changed, 3127 insertions, 2272 deletions
diff --git a/block/Kconfig b/block/Kconfig index 8e28ae7718bd..c6ce41a5e5b2 100644 --- a/block/Kconfig +++ b/block/Kconfig @@ -73,7 +73,7 @@ config BLK_DEV_ZONED config BLK_DEV_THROTTLING bool "Block layer bio throttling support" - depends on BLK_CGROUP=y + depends on BLK_CGROUP select BLK_CGROUP_RWSTAT help Block layer bio throttling support. It can be used to limit @@ -112,7 +112,7 @@ config BLK_WBT_MQ config BLK_CGROUP_IOLATENCY bool "Enable support for latency based cgroup IO protection" - depends on BLK_CGROUP=y + depends on BLK_CGROUP help Enabling this option enables the .latency interface for IO throttling. The IO controller will attempt to maintain average IO latencies below @@ -132,7 +132,7 @@ config BLK_CGROUP_FC_APPID config BLK_CGROUP_IOCOST bool "Enable support for cost model based cgroup IO controller" - depends on BLK_CGROUP=y + depends on BLK_CGROUP select BLK_RQ_IO_DATA_LEN select BLK_RQ_ALLOC_TIME help @@ -190,39 +190,31 @@ config BLK_INLINE_ENCRYPTION_FALLBACK by falling back to the kernel crypto API when inline encryption hardware is not present. -menu "Partition Types" - source "block/partitions/Kconfig" -endmenu - -endif # BLOCK - config BLOCK_COMPAT - bool - depends on BLOCK && COMPAT - default y + def_bool COMPAT config BLK_MQ_PCI - bool - depends on BLOCK && PCI - default y + def_bool PCI config BLK_MQ_VIRTIO bool - depends on BLOCK && VIRTIO + depends on VIRTIO default y config BLK_MQ_RDMA bool - depends on BLOCK && INFINIBAND + depends on INFINIBAND default y config BLK_PM - def_bool BLOCK && PM + def_bool PM # do not use in new code config BLOCK_HOLDER_DEPRECATED bool source "block/Kconfig.iosched" + +endif # BLOCK diff --git a/block/Kconfig.iosched b/block/Kconfig.iosched index 2f2158e05a91..885fee86dfca 100644 --- a/block/Kconfig.iosched +++ b/block/Kconfig.iosched @@ -1,6 +1,4 @@ # SPDX-License-Identifier: GPL-2.0 -if BLOCK - menu "IO Schedulers" config MQ_IOSCHED_DEADLINE @@ -45,5 +43,3 @@ config BFQ_CGROUP_DEBUG files in a cgroup which can be useful for debugging. endmenu - -endif diff --git a/block/Makefile b/block/Makefile index 41aa1ba69c90..44df57e562bf 100644 --- a/block/Makefile +++ b/block/Makefile @@ -3,13 +3,13 @@ # Makefile for the kernel block layer # -obj-$(CONFIG_BLOCK) := bdev.o fops.o bio.o elevator.o blk-core.o blk-sysfs.o \ +obj-y := bdev.o fops.o bio.o elevator.o blk-core.o blk-sysfs.o \ blk-flush.o blk-settings.o blk-ioc.o blk-map.o \ blk-exec.o blk-merge.o blk-timeout.o \ blk-lib.o blk-mq.o blk-mq-tag.o blk-stat.o \ blk-mq-sysfs.o blk-mq-cpumap.o blk-mq-sched.o ioctl.o \ genhd.o ioprio.o badblocks.o partitions/ blk-rq-qos.o \ - disk-events.o + disk-events.o blk-ia-ranges.o obj-$(CONFIG_BOUNCE) += bounce.o obj-$(CONFIG_BLK_DEV_BSG_COMMON) += bsg.o @@ -36,6 +36,6 @@ obj-$(CONFIG_BLK_DEBUG_FS) += blk-mq-debugfs.o obj-$(CONFIG_BLK_DEBUG_FS_ZONED)+= blk-mq-debugfs-zoned.o obj-$(CONFIG_BLK_SED_OPAL) += sed-opal.o obj-$(CONFIG_BLK_PM) += blk-pm.o -obj-$(CONFIG_BLK_INLINE_ENCRYPTION) += keyslot-manager.o blk-crypto.o +obj-$(CONFIG_BLK_INLINE_ENCRYPTION) += blk-crypto.o blk-crypto-profile.o obj-$(CONFIG_BLK_INLINE_ENCRYPTION_FALLBACK) += blk-crypto-fallback.o obj-$(CONFIG_BLOCK_HOLDER_DEPRECATED) += holder.o diff --git a/block/bdev.c b/block/bdev.c index 485a258b0ab3..7e6156203a71 100644 --- a/block/bdev.c +++ b/block/bdev.c @@ -12,6 +12,7 @@ #include <linux/major.h> #include <linux/device_cgroup.h> #include <linux/blkdev.h> +#include <linux/blk-integrity.h> #include <linux/backing-dev.h> #include <linux/module.h> #include <linux/blkpg.h> @@ -326,12 +327,12 @@ int bdev_read_page(struct block_device *bdev, sector_t sector, if (!ops->rw_page || bdev_get_integrity(bdev)) return result; - result = blk_queue_enter(bdev->bd_disk->queue, 0); + result = blk_queue_enter(bdev_get_queue(bdev), 0); if (result) return result; result = ops->rw_page(bdev, sector + get_start_sect(bdev), page, REQ_OP_READ); - blk_queue_exit(bdev->bd_disk->queue); + blk_queue_exit(bdev_get_queue(bdev)); return result; } @@ -362,7 +363,7 @@ int bdev_write_page(struct block_device *bdev, sector_t sector, if (!ops->rw_page || bdev_get_integrity(bdev)) return -EOPNOTSUPP; - result = blk_queue_enter(bdev->bd_disk->queue, 0); + result = blk_queue_enter(bdev_get_queue(bdev), 0); if (result) return result; @@ -375,7 +376,7 @@ int bdev_write_page(struct block_device *bdev, sector_t sector, clean_page_buffers(page); unlock_page(page); } - blk_queue_exit(bdev->bd_disk->queue); + blk_queue_exit(bdev_get_queue(bdev)); return result; } @@ -492,6 +493,7 @@ struct block_device *bdev_alloc(struct gendisk *disk, u8 partno) spin_lock_init(&bdev->bd_size_lock); bdev->bd_partno = partno; bdev->bd_inode = inode; + bdev->bd_queue = disk->queue; bdev->bd_stats = alloc_percpu(struct disk_stats); if (!bdev->bd_stats) { iput(inode); @@ -962,9 +964,11 @@ EXPORT_SYMBOL(blkdev_put); * @pathname: special file representing the block device * @dev: return value of the block device's dev_t * - * Get a reference to the blockdevice at @pathname in the current - * namespace if possible and return it. Return ERR_PTR(error) - * otherwise. + * Lookup the block device's dev_t at @pathname in the current + * namespace if possible and return it by @dev. + * + * RETURNS: + * 0 if succeeded, errno otherwise. */ int lookup_bdev(const char *pathname, dev_t *dev) { diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c index 85b8e1c3a762..24a5c5329bcd 100644 --- a/block/bfq-cgroup.c +++ b/block/bfq-cgroup.c @@ -6,13 +6,13 @@ #include <linux/slab.h> #include <linux/blkdev.h> #include <linux/cgroup.h> -#include <linux/elevator.h> #include <linux/ktime.h> #include <linux/rbtree.h> #include <linux/ioprio.h> #include <linux/sbitmap.h> #include <linux/delay.h> +#include "elevator.h" #include "bfq-iosched.h" #ifdef CONFIG_BFQ_CGROUP_DEBUG @@ -463,7 +463,7 @@ static int bfqg_stats_init(struct bfqg_stats *stats, gfp_t gfp) { if (blkg_rwstat_init(&stats->bytes, gfp) || blkg_rwstat_init(&stats->ios, gfp)) - return -ENOMEM; + goto error; #ifdef CONFIG_BFQ_CGROUP_DEBUG if (blkg_rwstat_init(&stats->merged, gfp) || @@ -476,13 +476,15 @@ static int bfqg_stats_init(struct bfqg_stats *stats, gfp_t gfp) bfq_stat_init(&stats->dequeue, gfp) || bfq_stat_init(&stats->group_wait_time, gfp) || bfq_stat_init(&stats->idle_time, gfp) || - bfq_stat_init(&stats->empty_time, gfp)) { - bfqg_stats_exit(stats); - return -ENOMEM; - } + bfq_stat_init(&stats->empty_time, gfp)) + goto error; #endif return 0; + +error: + bfqg_stats_exit(stats); + return -ENOMEM; } static struct bfq_group_data *cpd_to_bfqgd(struct blkcg_policy_data *cpd) diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index 480e1a134859..fec18118dc30 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -117,7 +117,6 @@ #include <linux/slab.h> #include <linux/blkdev.h> #include <linux/cgroup.h> -#include <linux/elevator.h> #include <linux/ktime.h> #include <linux/rbtree.h> #include <linux/ioprio.h> @@ -127,6 +126,7 @@ #include <trace/events/block.h> +#include "elevator.h" #include "blk.h" #include "blk-mq.h" #include "blk-mq-tag.h" @@ -6884,8 +6884,8 @@ static void bfq_depth_updated(struct blk_mq_hw_ctx *hctx) struct blk_mq_tags *tags = hctx->sched_tags; unsigned int min_shallow; - min_shallow = bfq_update_depths(bfqd, tags->bitmap_tags); - sbitmap_queue_min_shallow_depth(tags->bitmap_tags, min_shallow); + min_shallow = bfq_update_depths(bfqd, &tags->bitmap_tags); + sbitmap_queue_min_shallow_depth(&tags->bitmap_tags, min_shallow); } static int bfq_init_hctx(struct blk_mq_hw_ctx *hctx, unsigned int index) diff --git a/block/bio-integrity.c b/block/bio-integrity.c index 6b47cddbbca1..d25114715459 100644 --- a/block/bio-integrity.c +++ b/block/bio-integrity.c @@ -6,7 +6,7 @@ * Written by: Martin K. Petersen <martin.petersen@oracle.com> */ -#include <linux/blkdev.h> +#include <linux/blk-integrity.h> #include <linux/mempool.h> #include <linux/export.h> #include <linux/bio.h> @@ -134,7 +134,7 @@ int bio_integrity_add_page(struct bio *bio, struct page *page, iv = bip->bip_vec + bip->bip_vcnt; if (bip->bip_vcnt && - bvec_gap_to_prev(bio->bi_bdev->bd_disk->queue, + bvec_gap_to_prev(bdev_get_queue(bio->bi_bdev), &bip->bip_vec[bip->bip_vcnt - 1], offset)) return 0; diff --git a/block/bio.c b/block/bio.c index a6fb6a0b4295..15ab0d6d1c06 100644 --- a/block/bio.c +++ b/block/bio.c @@ -87,7 +87,8 @@ static struct bio_slab *create_bio_slab(unsigned int size) snprintf(bslab->name, sizeof(bslab->name), "bio-%d", size); bslab->slab = kmem_cache_create(bslab->name, size, - ARCH_KMALLOC_MINALIGN, SLAB_HWCACHE_ALIGN, NULL); + ARCH_KMALLOC_MINALIGN, + SLAB_HWCACHE_ALIGN | SLAB_TYPESAFE_BY_RCU, NULL); if (!bslab->slab) goto fail_alloc_slab; @@ -156,7 +157,7 @@ out: void bvec_free(mempool_t *pool, struct bio_vec *bv, unsigned short nr_vecs) { - BIO_BUG_ON(nr_vecs > BIO_MAX_VECS); + BUG_ON(nr_vecs > BIO_MAX_VECS); if (nr_vecs == BIO_MAX_VECS) mempool_free(bv, pool); @@ -281,6 +282,7 @@ void bio_init(struct bio *bio, struct bio_vec *table, atomic_set(&bio->__bi_remaining, 1); atomic_set(&bio->__bi_cnt, 1); + bio->bi_cookie = BLK_QC_T_NONE; bio->bi_max_vecs = max_vecs; bio->bi_io_vec = table; @@ -546,7 +548,7 @@ EXPORT_SYMBOL(zero_fill_bio); * REQ_OP_READ, zero the truncated part. This function should only * be used for handling corner cases, such as bio eod. */ -void bio_truncate(struct bio *bio, unsigned new_size) +static void bio_truncate(struct bio *bio, unsigned new_size) { struct bio_vec bv; struct bvec_iter iter; @@ -677,7 +679,7 @@ static void bio_alloc_cache_destroy(struct bio_set *bs) void bio_put(struct bio *bio) { if (unlikely(bio_flagged(bio, BIO_REFFED))) { - BIO_BUG_ON(!atomic_read(&bio->__bi_cnt)); + BUG_ON(!atomic_read(&bio->__bi_cnt)); if (!atomic_dec_and_test(&bio->__bi_cnt)) return; } @@ -772,6 +774,23 @@ const char *bio_devname(struct bio *bio, char *buf) } EXPORT_SYMBOL(bio_devname); +/** + * bio_full - check if the bio is full + * @bio: bio to check + * @len: length of one segment to be added + * + * Return true if @bio is full and one segment with @len bytes can't be + * added to the bio, otherwise return false + */ +static inline bool bio_full(struct bio *bio, unsigned len) +{ + if (bio->bi_vcnt >= bio->bi_max_vecs) + return true; + if (bio->bi_iter.bi_size > UINT_MAX - len) + return true; + return false; +} + static inline bool page_is_mergeable(const struct bio_vec *bv, struct page *page, unsigned int len, unsigned int off, bool *same_page) @@ -791,6 +810,44 @@ static inline bool page_is_mergeable(const struct bio_vec *bv, return (bv->bv_page + bv_end / PAGE_SIZE) == (page + off / PAGE_SIZE); } +/** + * __bio_try_merge_page - try appending data to an existing bvec. + * @bio: destination bio + * @page: start page to add + * @len: length of the data to add + * @off: offset of the data relative to @page + * @same_page: return if the segment has been merged inside the same page + * + * Try to add the data at @page + @off to the last bvec of @bio. This is a + * useful optimisation for file systems with a block size smaller than the + * page size. + * + * Warn if (@len, @off) crosses pages in case that @same_page is true. + * + * Return %true on success or %false on failure. + */ +static bool __bio_try_merge_page(struct bio *bio, struct page *page, + unsigned int len, unsigned int off, bool *same_page) +{ + if (WARN_ON_ONCE(bio_flagged(bio, BIO_CLONED))) + return false; + + if (bio->bi_vcnt > 0) { + struct bio_vec *bv = &bio->bi_io_vec[bio->bi_vcnt - 1]; + + if (page_is_mergeable(bv, page, len, off, same_page)) { + if (bio->bi_iter.bi_size > UINT_MAX - len) { + *same_page = false; + return false; + } + bv->bv_len += len; + bio->bi_iter.bi_size += len; + return true; + } + } + return false; +} + /* * Try to merge a page into a segment, while obeying the hardware segment * size limit. This is not for normal read/write bios, but for passthrough @@ -908,7 +965,7 @@ EXPORT_SYMBOL(bio_add_pc_page); int bio_add_zone_append_page(struct bio *bio, struct page *page, unsigned int len, unsigned int offset) { - struct request_queue *q = bio->bi_bdev->bd_disk->queue; + struct request_queue *q = bdev_get_queue(bio->bi_bdev); bool same_page = false; if (WARN_ON_ONCE(bio_op(bio) != REQ_OP_ZONE_APPEND)) @@ -923,45 +980,6 @@ int bio_add_zone_append_page(struct bio *bio, struct page *page, EXPORT_SYMBOL_GPL(bio_add_zone_append_page); /** - * __bio_try_merge_page - try appending data to an existing bvec. - * @bio: destination bio - * @page: start page to add - * @len: length of the data to add - * @off: offset of the data relative to @page - * @same_page: return if the segment has been merged inside the same page - * - * Try to add the data at @page + @off to the last bvec of @bio. This is a - * useful optimisation for file systems with a block size smaller than the - * page size. - * - * Warn if (@len, @off) crosses pages in case that @same_page is true. - * - * Return %true on success or %false on failure. - */ -bool __bio_try_merge_page(struct bio *bio, struct page *page, - unsigned int len, unsigned int off, bool *same_page) -{ - if (WARN_ON_ONCE(bio_flagged(bio, BIO_CLONED))) - return false; - - if (bio->bi_vcnt > 0) { - struct bio_vec *bv = &bio->bi_io_vec[bio->bi_vcnt - 1]; - - if (page_is_mergeable(bv, page, len, off, same_page)) { - if (bio->bi_iter.bi_size > UINT_MAX - len) { - *same_page = false; - return false; - } - bv->bv_len += len; - bio->bi_iter.bi_size += len; - return true; - } - } - return false; -} -EXPORT_SYMBOL_GPL(__bio_try_merge_page); - -/** * __bio_add_page - add page(s) to a bio in a new segment * @bio: destination bio * @page: start page to add @@ -1015,52 +1033,40 @@ int bio_add_page(struct bio *bio, struct page *page, } EXPORT_SYMBOL(bio_add_page); -void bio_release_pages(struct bio *bio, bool mark_dirty) +void __bio_release_pages(struct bio *bio, bool mark_dirty) { struct bvec_iter_all iter_all; struct bio_vec *bvec; - if (bio_flagged(bio, BIO_NO_PAGE_REF)) - return; - bio_for_each_segment_all(bvec, bio, iter_all) { if (mark_dirty && !PageCompound(bvec->bv_page)) set_page_dirty_lock(bvec->bv_page); put_page(bvec->bv_page); } } -EXPORT_SYMBOL_GPL(bio_release_pages); +EXPORT_SYMBOL_GPL(__bio_release_pages); -static void __bio_iov_bvec_set(struct bio *bio, struct iov_iter *iter) +void bio_iov_bvec_set(struct bio *bio, struct iov_iter *iter) { + size_t size = iov_iter_count(iter); + WARN_ON_ONCE(bio->bi_max_vecs); + if (bio_op(bio) == REQ_OP_ZONE_APPEND) { + struct request_queue *q = bdev_get_queue(bio->bi_bdev); + size_t max_sectors = queue_max_zone_append_sectors(q); + + size = min(size, max_sectors << SECTOR_SHIFT); + } + bio->bi_vcnt = iter->nr_segs; bio->bi_io_vec = (struct bio_vec *)iter->bvec; bio->bi_iter.bi_bvec_done = iter->iov_offset; - bio->bi_iter.bi_size = iter->count; + bio->bi_iter.bi_size = size; bio_set_flag(bio, BIO_NO_PAGE_REF); bio_set_flag(bio, BIO_CLONED); } -static int bio_iov_bvec_set(struct bio *bio, struct iov_iter *iter) -{ - __bio_iov_bvec_set(bio, iter); - iov_iter_advance(iter, iter->count); - return 0; -} - -static int bio_iov_bvec_set_append(struct bio *bio, struct iov_iter *iter) -{ - struct request_queue *q = bio->bi_bdev->bd_disk->queue; - struct iov_iter i = *iter; - - iov_iter_truncate(&i, queue_max_zone_append_sectors(q) << 9); - __bio_iov_bvec_set(bio, &i); - iov_iter_advance(iter, i.count); - return 0; -} - static void bio_put_pages(struct page **pages, size_t size, size_t off) { size_t i, nr = DIV_ROUND_UP(size + (off & ~PAGE_MASK), PAGE_SIZE); @@ -1130,7 +1136,7 @@ static int __bio_iov_append_get_pages(struct bio *bio, struct iov_iter *iter) { unsigned short nr_pages = bio->bi_max_vecs - bio->bi_vcnt; unsigned short entries_left = bio->bi_max_vecs - bio->bi_vcnt; - struct request_queue *q = bio->bi_bdev->bd_disk->queue; + struct request_queue *q = bdev_get_queue(bio->bi_bdev); unsigned int max_append_sectors = queue_max_zone_append_sectors(q); struct bio_vec *bv = bio->bi_io_vec + bio->bi_vcnt; struct page **pages = (struct page **)bv; @@ -1202,9 +1208,9 @@ int bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) int ret = 0; if (iov_iter_is_bvec(iter)) { - if (bio_op(bio) == REQ_OP_ZONE_APPEND) - return bio_iov_bvec_set_append(bio, iter); - return bio_iov_bvec_set(bio, iter); + bio_iov_bvec_set(bio, iter); + iov_iter_advance(iter, bio->bi_iter.bi_size); + return 0; } do { @@ -1260,18 +1266,7 @@ int submit_bio_wait(struct bio *bio) } EXPORT_SYMBOL(submit_bio_wait); -/** - * bio_advance - increment/complete a bio by some number of bytes - * @bio: bio to advance - * @bytes: number of bytes to complete - * - * This updates bi_sector, bi_size and bi_idx; if the number of bytes to - * complete doesn't align with a bvec boundary, then bv_len and bv_offset will - * be updated on the last bvec as well. - * - * @bio will then represent the remaining, uncompleted portion of the io. - */ -void bio_advance(struct bio *bio, unsigned bytes) +void __bio_advance(struct bio *bio, unsigned bytes) { if (bio_integrity(bio)) bio_integrity_advance(bio, bytes); @@ -1279,7 +1274,7 @@ void bio_advance(struct bio *bio, unsigned bytes) bio_crypt_advance(bio, bytes); bio_advance_iter(bio, &bio->bi_iter, bytes); } -EXPORT_SYMBOL(bio_advance); +EXPORT_SYMBOL(__bio_advance); void bio_copy_data_iter(struct bio *dst, struct bvec_iter *dst_iter, struct bio *src, struct bvec_iter *src_iter) @@ -1467,10 +1462,10 @@ again: return; if (bio->bi_bdev && bio_flagged(bio, BIO_TRACKED)) - rq_qos_done_bio(bio->bi_bdev->bd_disk->queue, bio); + rq_qos_done_bio(bdev_get_queue(bio->bi_bdev), bio); if (bio->bi_bdev && bio_flagged(bio, BIO_TRACE_COMPLETION)) { - trace_block_bio_complete(bio->bi_bdev->bd_disk->queue, bio); + trace_block_bio_complete(bdev_get_queue(bio->bi_bdev), bio); bio_clear_flag(bio, BIO_TRACE_COMPLETION); } diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index 9a1c5839dd46..88b1fce90520 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -32,6 +32,7 @@ #include <linux/psi.h> #include "blk.h" #include "blk-ioprio.h" +#include "blk-throttle.h" /* * blkcg_pol_mutex protects blkcg_policy[] and policy [de]activation. @@ -620,7 +621,7 @@ struct block_device *blkcg_conf_open_bdev(char **inputp) */ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, char *input, struct blkg_conf_ctx *ctx) - __acquires(rcu) __acquires(&bdev->bd_disk->queue->queue_lock) + __acquires(rcu) __acquires(&bdev->bd_queue->queue_lock) { struct block_device *bdev; struct request_queue *q; @@ -631,7 +632,15 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, if (IS_ERR(bdev)) return PTR_ERR(bdev); - q = bdev->bd_disk->queue; + q = bdev_get_queue(bdev); + + /* + * blkcg_deactivate_policy() requires queue to be frozen, we can grab + * q_usage_counter to prevent concurrent with blkcg_deactivate_policy(). + */ + ret = blk_queue_enter(q, 0); + if (ret) + return ret; rcu_read_lock(); spin_lock_irq(&q->queue_lock); @@ -702,6 +711,7 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, goto success; } success: + blk_queue_exit(q); ctx->bdev = bdev; ctx->blkg = blkg; ctx->body = input; @@ -714,6 +724,7 @@ fail_unlock: rcu_read_unlock(); fail: blkdev_put_no_open(bdev); + blk_queue_exit(q); /* * If queue was bypassing, we should retry. Do so after a * short msleep(). It isn't strictly necessary but queue @@ -736,9 +747,9 @@ EXPORT_SYMBOL_GPL(blkg_conf_prep); * with blkg_conf_prep(). */ void blkg_conf_finish(struct blkg_conf_ctx *ctx) - __releases(&ctx->bdev->bd_disk->queue->queue_lock) __releases(rcu) + __releases(&ctx->bdev->bd_queue->queue_lock) __releases(rcu) { - spin_unlock_irq(&ctx->bdev->bd_disk->queue->queue_lock); + spin_unlock_irq(&bdev_get_queue(ctx->bdev)->queue_lock); rcu_read_unlock(); blkdev_put_no_open(ctx->bdev); } @@ -841,7 +852,7 @@ static void blkcg_fill_root_iostats(void) while ((dev = class_dev_iter_next(&iter))) { struct block_device *bdev = dev_to_bdev(dev); struct blkcg_gq *blkg = - blk_queue_root_blkg(bdev->bd_disk->queue); + blk_queue_root_blkg(bdev_get_queue(bdev)); struct blkg_iostat tmp; int cpu; @@ -1800,7 +1811,7 @@ static inline struct blkcg_gq *blkg_tryget_closest(struct bio *bio, rcu_read_lock(); blkg = blkg_lookup_create(css_to_blkcg(css), - bio->bi_bdev->bd_disk->queue); + bdev_get_queue(bio->bi_bdev)); while (blkg) { if (blkg_tryget(blkg)) { ret_blkg = blkg; @@ -1836,8 +1847,8 @@ void bio_associate_blkg_from_css(struct bio *bio, if (css && css->parent) { bio->bi_blkg = blkg_tryget_closest(bio, css); } else { - blkg_get(bio->bi_bdev->bd_disk->queue->root_blkg); - bio->bi_blkg = bio->bi_bdev->bd_disk->queue->root_blkg; + blkg_get(bdev_get_queue(bio->bi_bdev)->root_blkg); + bio->bi_blkg = bdev_get_queue(bio->bi_bdev)->root_blkg; } } EXPORT_SYMBOL_GPL(bio_associate_blkg_from_css); diff --git a/block/blk-core.c b/block/blk-core.c index 4d8f5fe91588..fd389a16013c 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -18,6 +18,7 @@ #include <linux/blkdev.h> #include <linux/blk-mq.h> #include <linux/blk-pm.h> +#include <linux/blk-integrity.h> #include <linux/highmem.h> #include <linux/mm.h> #include <linux/pagemap.h> @@ -49,6 +50,7 @@ #include "blk-mq.h" #include "blk-mq-sched.h" #include "blk-pm.h" +#include "blk-throttle.h" struct dentry *blk_debugfs_root; @@ -214,8 +216,7 @@ int blk_status_to_errno(blk_status_t status) } EXPORT_SYMBOL_GPL(blk_status_to_errno); -static void print_req_error(struct request *req, blk_status_t status, - const char *caller) +void blk_print_req_error(struct request *req, blk_status_t status) { int idx = (__force int)status; @@ -223,9 +224,9 @@ static void print_req_error(struct request *req, blk_status_t status, return; printk_ratelimited(KERN_ERR - "%s: %s error, dev %s, sector %llu op 0x%x:(%s) flags 0x%x " + "%s error, dev %s, sector %llu op 0x%x:(%s) flags 0x%x " "phys_seg %u prio class %u\n", - caller, blk_errors[idx].name, + blk_errors[idx].name, req->rq_disk ? req->rq_disk->disk_name : "?", blk_rq_pos(req), req_op(req), blk_op_str(req_op(req)), req->cmd_flags & ~REQ_OP_MASK, @@ -233,33 +234,6 @@ static void print_req_error(struct request *req, blk_status_t status, IOPRIO_PRIO_CLASS(req->ioprio)); } -static void req_bio_endio(struct request *rq, struct bio *bio, - unsigned int nbytes, blk_status_t error) -{ - if (error) - bio->bi_status = error; - - if (unlikely(rq->rq_flags & RQF_QUIET)) - bio_set_flag(bio, BIO_QUIET); - - bio_advance(bio, nbytes); - - if (req_op(rq) == REQ_OP_ZONE_APPEND && error == BLK_STS_OK) { - /* - * Partial zone append completions cannot be supported as the - * BIO fragments may end up not being written sequentially. - */ - if (bio->bi_iter.bi_size) - bio->bi_status = BLK_STS_IOERR; - else - bio->bi_iter.bi_sector = rq->__sector; - } - - /* don't actually finish bio if it's part of flush sequence */ - if (bio->bi_iter.bi_size == 0 && !(rq->rq_flags & RQF_FLUSH_SEQ)) - bio_endio(bio); -} - void blk_dump_rq_flags(struct request *rq, char *msg) { printk(KERN_INFO "%s: dev %s: flags=%llx\n", msg, @@ -402,7 +376,7 @@ void blk_cleanup_queue(struct request_queue *q) */ mutex_lock(&q->sysfs_lock); if (q->elevator) - blk_mq_sched_free_requests(q); + blk_mq_sched_free_rqs(q); mutex_unlock(&q->sysfs_lock); percpu_ref_exit(&q->q_usage_counter); @@ -415,7 +389,7 @@ EXPORT_SYMBOL(blk_cleanup_queue); static bool blk_try_enter_queue(struct request_queue *q, bool pm) { rcu_read_lock(); - if (!percpu_ref_tryget_live(&q->q_usage_counter)) + if (!percpu_ref_tryget_live_rcu(&q->q_usage_counter)) goto fail; /* @@ -430,7 +404,7 @@ static bool blk_try_enter_queue(struct request_queue *q, bool pm) return true; fail_put: - percpu_ref_put(&q->q_usage_counter); + blk_queue_exit(q); fail: rcu_read_unlock(); return false; @@ -470,10 +444,11 @@ int blk_queue_enter(struct request_queue *q, blk_mq_req_flags_t flags) static inline int bio_queue_enter(struct bio *bio) { - struct gendisk *disk = bio->bi_bdev->bd_disk; - struct request_queue *q = disk->queue; + struct request_queue *q = bdev_get_queue(bio->bi_bdev); while (!blk_try_enter_queue(q, false)) { + struct gendisk *disk = bio->bi_bdev->bd_disk; + if (bio->bi_opf & REQ_NOWAIT) { if (test_bit(GD_DEAD, &disk->state)) goto dead; @@ -553,7 +528,7 @@ struct request_queue *blk_alloc_queue(int node_id) q->node = node_id; - atomic_set(&q->nr_active_requests_shared_sbitmap, 0); + atomic_set(&q->nr_active_requests_shared_tags, 0); timer_setup(&q->timeout, blk_rq_timed_out_timer, 0); INIT_WORK(&q->timeout_work, blk_timeout_work); @@ -586,7 +561,7 @@ struct request_queue *blk_alloc_queue(int node_id) blk_queue_dma_alignment(q, 511); blk_set_default_limits(&q->limits); - q->nr_requests = BLKDEV_MAX_RQ; + q->nr_requests = BLKDEV_DEFAULT_RQ; return q; @@ -654,8 +629,9 @@ static void handle_bad_sector(struct bio *bio, sector_t maxsector) { char b[BDEVNAME_SIZE]; - pr_info_ratelimited("attempt to access beyond end of device\n" + pr_info_ratelimited("%s: attempt to access beyond end of device\n" "%s: rw=%d, want=%llu, limit=%llu\n", + current->comm, bio_devname(bio, b), bio->bi_opf, bio_end_sector(bio), maxsector); } @@ -797,7 +773,7 @@ static inline blk_status_t blk_check_zone_append(struct request_queue *q, static noinline_for_stack bool submit_bio_checks(struct bio *bio) { struct block_device *bdev = bio->bi_bdev; - struct request_queue *q = bdev->bd_disk->queue; + struct request_queue *q = bdev_get_queue(bdev); blk_status_t status = BLK_STS_IOERR; struct blk_plug *plug; @@ -839,7 +815,7 @@ static noinline_for_stack bool submit_bio_checks(struct bio *bio) } if (!test_bit(QUEUE_FLAG_POLL, &q->queue_flags)) - bio_clear_hipri(bio); + bio_clear_polled(bio); switch (bio_op(bio)) { case REQ_OP_DISCARD: @@ -912,25 +888,22 @@ end_io: return false; } -static blk_qc_t __submit_bio(struct bio *bio) +static void __submit_bio(struct bio *bio) { struct gendisk *disk = bio->bi_bdev->bd_disk; - blk_qc_t ret = BLK_QC_T_NONE; if (unlikely(bio_queue_enter(bio) != 0)) - return BLK_QC_T_NONE; + return; if (!submit_bio_checks(bio) || !blk_crypto_bio_prep(&bio)) goto queue_exit; - if (disk->fops->submit_bio) { - ret = disk->fops->submit_bio(bio); - goto queue_exit; + if (!disk->fops->submit_bio) { + blk_mq_submit_bio(bio); + return; } - return blk_mq_submit_bio(bio); - + disk->fops->submit_bio(bio); queue_exit: blk_queue_exit(disk->queue); - return ret; } /* @@ -952,10 +925,9 @@ queue_exit: * bio_list_on_stack[1] contains bios that were submitted before the current * ->submit_bio_bio, but that haven't been processed yet. */ -static blk_qc_t __submit_bio_noacct(struct bio *bio) +static void __submit_bio_noacct(struct bio *bio) { struct bio_list bio_list_on_stack[2]; - blk_qc_t ret = BLK_QC_T_NONE; BUG_ON(bio->bi_next); @@ -963,7 +935,7 @@ static blk_qc_t __submit_bio_noacct(struct bio *bio) current->bio_list = bio_list_on_stack; do { - struct request_queue *q = bio->bi_bdev->bd_disk->queue; + struct request_queue *q = bdev_get_queue(bio->bi_bdev); struct bio_list lower, same; /* @@ -972,7 +944,7 @@ static blk_qc_t __submit_bio_noacct(struct bio *bio) bio_list_on_stack[1] = bio_list_on_stack[0]; bio_list_init(&bio_list_on_stack[0]); - ret = __submit_bio(bio); + __submit_bio(bio); /* * Sort new bios into those for a lower level and those for the @@ -981,7 +953,7 @@ static blk_qc_t __submit_bio_noacct(struct bio *bio) bio_list_init(&lower); bio_list_init(&same); while ((bio = bio_list_pop(&bio_list_on_stack[0])) != NULL) - if (q == bio->bi_bdev->bd_disk->queue) + if (q == bdev_get_queue(bio->bi_bdev)) bio_list_add(&same, bio); else bio_list_add(&lower, bio); @@ -995,22 +967,19 @@ static blk_qc_t __submit_bio_noacct(struct bio *bio) } while ((bio = bio_list_pop(&bio_list_on_stack[0]))); current->bio_list = NULL; - return ret; } -static blk_qc_t __submit_bio_noacct_mq(struct bio *bio) +static void __submit_bio_noacct_mq(struct bio *bio) { struct bio_list bio_list[2] = { }; - blk_qc_t ret; current->bio_list = bio_list; do { - ret = __submit_bio(bio); + __submit_bio(bio); } while ((bio = bio_list_pop(&bio_list[0]))); current->bio_list = NULL; - return ret; } /** @@ -1022,7 +991,7 @@ static blk_qc_t __submit_bio_noacct_mq(struct bio *bio) * systems and other upper level users of the block layer should use * submit_bio() instead. */ -blk_qc_t submit_bio_noacct(struct bio *bio) +void submit_bio_noacct(struct bio *bio) { /* * We only want one ->submit_bio to be active at a time, else stack @@ -1030,14 +999,12 @@ blk_qc_t submit_bio_noacct(struct bio *bio) * to collect a list of requests submited by a ->submit_bio method while * it is active, and then process them after it returned. */ - if (current->bio_list) { + if (current->bio_list) bio_list_add(¤t->bio_list[0], bio); - return BLK_QC_T_NONE; - } - - if (!bio->bi_bdev->bd_disk->fops->submit_bio) - return __submit_bio_noacct_mq(bio); - return __submit_bio_noacct(bio); + else if (!bio->bi_bdev->bd_disk->fops->submit_bio) + __submit_bio_noacct_mq(bio); + else + __submit_bio_noacct(bio); } EXPORT_SYMBOL(submit_bio_noacct); @@ -1054,10 +1021,10 @@ EXPORT_SYMBOL(submit_bio_noacct); * in @bio. The bio must NOT be touched by thecaller until ->bi_end_io() has * been called. */ -blk_qc_t submit_bio(struct bio *bio) +void submit_bio(struct bio *bio) { if (blkcg_punt_bio_submit(bio)) - return BLK_QC_T_NONE; + return; /* * If it's a regular read/write or a barrier with data attached, @@ -1068,7 +1035,7 @@ blk_qc_t submit_bio(struct bio *bio) if (unlikely(bio_op(bio) == REQ_OP_WRITE_SAME)) count = queue_logical_block_size( - bio->bi_bdev->bd_disk->queue) >> 9; + bdev_get_queue(bio->bi_bdev)) >> 9; else count = bio_sectors(bio); @@ -1089,20 +1056,93 @@ blk_qc_t submit_bio(struct bio *bio) if (unlikely(bio_op(bio) == REQ_OP_READ && bio_flagged(bio, BIO_WORKINGSET))) { unsigned long pflags; - blk_qc_t ret; psi_memstall_enter(&pflags); - ret = submit_bio_noacct(bio); + submit_bio_noacct(bio); psi_memstall_leave(&pflags); - - return ret; + return; } - return submit_bio_noacct(bio); + submit_bio_noacct(bio); } EXPORT_SYMBOL(submit_bio); /** + * bio_poll - poll for BIO completions + * @bio: bio to poll for + * @flags: BLK_POLL_* flags that control the behavior + * + * Poll for completions on queue associated with the bio. Returns number of + * completed entries found. + * + * Note: the caller must either be the context that submitted @bio, or + * be in a RCU critical section to prevent freeing of @bio. + */ +int bio_poll(struct bio *bio, struct io_comp_batch *iob, unsigned int flags) +{ + struct request_queue *q = bdev_get_queue(bio->bi_bdev); + blk_qc_t cookie = READ_ONCE(bio->bi_cookie); + int ret; + + if (cookie == BLK_QC_T_NONE || + !test_bit(QUEUE_FLAG_POLL, &q->queue_flags)) + return 0; + + if (current->plug) + blk_flush_plug(current->plug, false); + + if (blk_queue_enter(q, BLK_MQ_REQ_NOWAIT)) + return 0; + if (WARN_ON_ONCE(!queue_is_mq(q))) + ret = 0; /* not yet implemented, should not happen */ + else + ret = blk_mq_poll(q, cookie, iob, flags); + blk_queue_exit(q); + return ret; +} +EXPORT_SYMBOL_GPL(bio_poll); + +/* + * Helper to implement file_operations.iopoll. Requires the bio to be stored + * in iocb->private, and cleared before freeing the bio. + */ +int iocb_bio_iopoll(struct kiocb *kiocb, struct io_comp_batch *iob, + unsigned int flags) +{ + struct bio *bio; + int ret = 0; + + /* + * Note: the bio cache only uses SLAB_TYPESAFE_BY_RCU, so bio can + * point to a freshly allocated bio at this point. If that happens + * we have a few cases to consider: + * + * 1) the bio is beeing initialized and bi_bdev is NULL. We can just + * simply nothing in this case + * 2) the bio points to a not poll enabled device. bio_poll will catch + * this and return 0 + * 3) the bio points to a poll capable device, including but not + * limited to the one that the original bio pointed to. In this + * case we will call into the actual poll method and poll for I/O, + * even if we don't need to, but it won't cause harm either. + * + * For cases 2) and 3) above the RCU grace period ensures that bi_bdev + * is still allocated. Because partitions hold a reference to the whole + * device bdev and thus disk, the disk is also still valid. Grabbing + * a reference to the queue in bio_poll() ensures the hctxs and requests + * are still valid as well. + */ + rcu_read_lock(); + bio = READ_ONCE(kiocb->private); + if (bio && bio->bi_bdev) + ret = bio_poll(bio, iob, flags); + rcu_read_unlock(); + + return ret; +} +EXPORT_SYMBOL_GPL(iocb_bio_iopoll); + +/** * blk_cloned_rq_check_limits - Helper function to check a cloned request * for the new queue limits * @q: the queue @@ -1177,8 +1217,7 @@ blk_status_t blk_insert_cloned_request(struct request_queue *q, struct request * if (blk_crypto_insert_cloned_request(rq)) return BLK_STS_IOERR; - if (blk_queue_io_stat(q)) - blk_account_io_start(rq); + blk_account_io_start(rq); /* * Since we have a scheduler attached on the top device, @@ -1246,41 +1285,19 @@ again: } } -static void blk_account_io_completion(struct request *req, unsigned int bytes) +void __blk_account_io_done(struct request *req, u64 now) { - if (req->part && blk_do_io_stat(req)) { - const int sgrp = op_stat_group(req_op(req)); - - part_stat_lock(); - part_stat_add(req->part, sectors[sgrp], bytes >> 9); - part_stat_unlock(); - } -} + const int sgrp = op_stat_group(req_op(req)); -void blk_account_io_done(struct request *req, u64 now) -{ - /* - * Account IO completion. flush_rq isn't accounted as a - * normal IO on queueing nor completion. Accounting the - * containing request is enough. - */ - if (req->part && blk_do_io_stat(req) && - !(req->rq_flags & RQF_FLUSH_SEQ)) { - const int sgrp = op_stat_group(req_op(req)); - - part_stat_lock(); - update_io_ticks(req->part, jiffies, true); - part_stat_inc(req->part, ios[sgrp]); - part_stat_add(req->part, nsecs[sgrp], now - req->start_time_ns); - part_stat_unlock(); - } + part_stat_lock(); + update_io_ticks(req->part, jiffies, true); + part_stat_inc(req->part, ios[sgrp]); + part_stat_add(req->part, nsecs[sgrp], now - req->start_time_ns); + part_stat_unlock(); } -void blk_account_io_start(struct request *rq) +void __blk_account_io_start(struct request *rq) { - if (!blk_do_io_stat(rq)) - return; - /* passthrough requests can hold bios that do not have ->bi_bdev set */ if (rq->bio && rq->bio->bi_bdev) rq->part = rq->bio->bi_bdev; @@ -1376,112 +1393,6 @@ void blk_steal_bios(struct bio_list *list, struct request *rq) } EXPORT_SYMBOL_GPL(blk_steal_bios); -/** - * blk_update_request - Complete multiple bytes without completing the request - * @req: the request being processed - * @error: block status code - * @nr_bytes: number of bytes to complete for @req - * - * Description: - * Ends I/O on a number of bytes attached to @req, but doesn't complete - * the request structure even if @req doesn't have leftover. - * If @req has leftover, sets it up for the next range of segments. - * - * Passing the result of blk_rq_bytes() as @nr_bytes guarantees - * %false return from this function. - * - * Note: - * The RQF_SPECIAL_PAYLOAD flag is ignored on purpose in this function - * except in the consistency check at the end of this function. - * - * Return: - * %false - this request doesn't have any more data - * %true - this request has more data - **/ -bool blk_update_request(struct request *req, blk_status_t error, - unsigned int nr_bytes) -{ - int total_bytes; - - trace_block_rq_complete(req, blk_status_to_errno(error), nr_bytes); - - if (!req->bio) - return false; - -#ifdef CONFIG_BLK_DEV_INTEGRITY - if (blk_integrity_rq(req) && req_op(req) == REQ_OP_READ && - error == BLK_STS_OK) - req->q->integrity.profile->complete_fn(req, nr_bytes); -#endif - - if (unlikely(error && !blk_rq_is_passthrough(req) && - !(req->rq_flags & RQF_QUIET))) - print_req_error(req, error, __func__); - - blk_account_io_completion(req, nr_bytes); - - total_bytes = 0; - while (req->bio) { - struct bio *bio = req->bio; - unsigned bio_bytes = min(bio->bi_iter.bi_size, nr_bytes); - - if (bio_bytes == bio->bi_iter.bi_size) - req->bio = bio->bi_next; - - /* Completion has already been traced */ - bio_clear_flag(bio, BIO_TRACE_COMPLETION); - req_bio_endio(req, bio, bio_bytes, error); - - total_bytes += bio_bytes; - nr_bytes -= bio_bytes; - - if (!nr_bytes) - break; - } - - /* - * completely done - */ - if (!req->bio) { - /* - * Reset counters so that the request stacking driver - * can find how many bytes remain in the request - * later. - */ - req->__data_len = 0; - return false; - } - - req->__data_len -= total_bytes; - - /* update sector only for requests with clear definition of sector */ - if (!blk_rq_is_passthrough(req)) - req->__sector += total_bytes >> 9; - - /* mixed attributes always follow the first bio */ - if (req->rq_flags & RQF_MIXED_MERGE) { - req->cmd_flags &= ~REQ_FAILFAST_MASK; - req->cmd_flags |= req->bio->bi_opf & REQ_FAILFAST_MASK; - } - - if (!(req->rq_flags & RQF_SPECIAL_PAYLOAD)) { - /* - * If total number of sectors is less than the first segment - * size, something has gone terribly wrong. - */ - if (blk_rq_bytes(req) < blk_rq_cur_bytes(req)) { - blk_dump_rq_flags(req, "request botched"); - req->__data_len = blk_rq_cur_bytes(req); - } - - /* recalculate the number of segments */ - req->nr_phys_segments = blk_recalc_rq_segments(req); - } - - return true; -} -EXPORT_SYMBOL_GPL(blk_update_request); - #if ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE /** * rq_flush_dcache_pages - Helper function to flush all pages in a request @@ -1629,6 +1540,32 @@ int kblockd_mod_delayed_work_on(int cpu, struct delayed_work *dwork, } EXPORT_SYMBOL(kblockd_mod_delayed_work_on); +void blk_start_plug_nr_ios(struct blk_plug *plug, unsigned short nr_ios) +{ + struct task_struct *tsk = current; + + /* + * If this is a nested plug, don't actually assign it. + */ + if (tsk->plug) + return; + + plug->mq_list = NULL; + plug->cached_rq = NULL; + plug->nr_ios = min_t(unsigned short, nr_ios, BLK_MAX_REQUEST_COUNT); + plug->rq_count = 0; + plug->multiple_queues = false; + plug->has_elevator = false; + plug->nowait = false; + INIT_LIST_HEAD(&plug->cb_list); + + /* + * Store ordering should not be needed here, since a potential + * preempt will imply a full memory barrier + */ + tsk->plug = plug; +} + /** * blk_start_plug - initialize blk_plug and track it inside the task_struct * @plug: The &struct blk_plug that needs to be initialized @@ -1654,25 +1591,7 @@ EXPORT_SYMBOL(kblockd_mod_delayed_work_on); */ void blk_start_plug(struct blk_plug *plug) { - struct task_struct *tsk = current; - - /* - * If this is a nested plug, don't actually assign it. - */ - if (tsk->plug) - return; - - INIT_LIST_HEAD(&plug->mq_list); - INIT_LIST_HEAD(&plug->cb_list); - plug->rq_count = 0; - plug->multiple_queues = false; - plug->nowait = false; - - /* - * Store ordering should not be needed here, since a potential - * preempt will imply a full memory barrier - */ - tsk->plug = plug; + blk_start_plug_nr_ios(plug, 1); } EXPORT_SYMBOL(blk_start_plug); @@ -1718,12 +1637,14 @@ struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug, void *data, } EXPORT_SYMBOL(blk_check_plugged); -void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule) +void blk_flush_plug(struct blk_plug *plug, bool from_schedule) { - flush_plug_callbacks(plug, from_schedule); - - if (!list_empty(&plug->mq_list)) + if (!list_empty(&plug->cb_list)) + flush_plug_callbacks(plug, from_schedule); + if (!rq_list_empty(plug->mq_list)) blk_mq_flush_plug_list(plug, from_schedule); + if (unlikely(!from_schedule && plug->cached_rq)) + blk_mq_free_plug_rqs(plug); } /** @@ -1738,11 +1659,10 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule) */ void blk_finish_plug(struct blk_plug *plug) { - if (plug != current->plug) - return; - blk_flush_plug_list(plug, false); - - current->plug = NULL; + if (plug == current->plug) { + blk_flush_plug(plug, false); + current->plug = NULL; + } } EXPORT_SYMBOL(blk_finish_plug); diff --git a/block/blk-crypto-fallback.c b/block/blk-crypto-fallback.c index c322176a1e09..c87aba8584c6 100644 --- a/block/blk-crypto-fallback.c +++ b/block/blk-crypto-fallback.c @@ -12,12 +12,13 @@ #include <crypto/skcipher.h> #include <linux/blk-cgroup.h> #include <linux/blk-crypto.h> +#include <linux/blk-crypto-profile.h> #include <linux/blkdev.h> #include <linux/crypto.h> -#include <linux/keyslot-manager.h> #include <linux/mempool.h> #include <linux/module.h> #include <linux/random.h> +#include <linux/scatterlist.h> #include "blk-crypto-internal.h" @@ -72,12 +73,12 @@ static mempool_t *bio_fallback_crypt_ctx_pool; static DEFINE_MUTEX(tfms_init_lock); static bool tfms_inited[BLK_ENCRYPTION_MODE_MAX]; -static struct blk_crypto_keyslot { +static struct blk_crypto_fallback_keyslot { enum blk_crypto_mode_num crypto_mode; struct crypto_skcipher *tfms[BLK_ENCRYPTION_MODE_MAX]; } *blk_crypto_keyslots; -static struct blk_keyslot_manager blk_crypto_ksm; +static struct blk_crypto_profile blk_crypto_fallback_profile; static struct workqueue_struct *blk_crypto_wq; static mempool_t *blk_crypto_bounce_page_pool; static struct bio_set crypto_bio_split; @@ -88,9 +89,9 @@ static struct bio_set crypto_bio_split; */ static u8 blank_key[BLK_CRYPTO_MAX_KEY_SIZE]; -static void blk_crypto_evict_keyslot(unsigned int slot) +static void blk_crypto_fallback_evict_keyslot(unsigned int slot) { - struct blk_crypto_keyslot *slotp = &blk_crypto_keyslots[slot]; + struct blk_crypto_fallback_keyslot *slotp = &blk_crypto_keyslots[slot]; enum blk_crypto_mode_num crypto_mode = slotp->crypto_mode; int err; @@ -103,45 +104,41 @@ static void blk_crypto_evict_keyslot(unsigned int slot) slotp->crypto_mode = BLK_ENCRYPTION_MODE_INVALID; } -static int blk_crypto_keyslot_program(struct blk_keyslot_manager *ksm, - const struct blk_crypto_key *key, - unsigned int slot) +static int +blk_crypto_fallback_keyslot_program(struct blk_crypto_profile *profile, + const struct blk_crypto_key *key, + unsigned int slot) { - struct blk_crypto_keyslot *slotp = &blk_crypto_keyslots[slot]; + struct blk_crypto_fallback_keyslot *slotp = &blk_crypto_keyslots[slot]; const enum blk_crypto_mode_num crypto_mode = key->crypto_cfg.crypto_mode; int err; if (crypto_mode != slotp->crypto_mode && slotp->crypto_mode != BLK_ENCRYPTION_MODE_INVALID) - blk_crypto_evict_keyslot(slot); + blk_crypto_fallback_evict_keyslot(slot); slotp->crypto_mode = crypto_mode; err = crypto_skcipher_setkey(slotp->tfms[crypto_mode], key->raw, key->size); if (err) { - blk_crypto_evict_keyslot(slot); + blk_crypto_fallback_evict_keyslot(slot); return err; } return 0; } -static int blk_crypto_keyslot_evict(struct blk_keyslot_manager *ksm, - const struct blk_crypto_key *key, - unsigned int slot) +static int blk_crypto_fallback_keyslot_evict(struct blk_crypto_profile *profile, + const struct blk_crypto_key *key, + unsigned int slot) { - blk_crypto_evict_keyslot(slot); + blk_crypto_fallback_evict_keyslot(slot); return 0; } -/* - * The crypto API fallback KSM ops - only used for a bio when it specifies a - * blk_crypto_key that was not supported by the device's inline encryption - * hardware. - */ -static const struct blk_ksm_ll_ops blk_crypto_ksm_ll_ops = { - .keyslot_program = blk_crypto_keyslot_program, - .keyslot_evict = blk_crypto_keyslot_evict, +static const struct blk_crypto_ll_ops blk_crypto_fallback_ll_ops = { + .keyslot_program = blk_crypto_fallback_keyslot_program, + .keyslot_evict = blk_crypto_fallback_keyslot_evict, }; static void blk_crypto_fallback_encrypt_endio(struct bio *enc_bio) @@ -159,7 +156,7 @@ static void blk_crypto_fallback_encrypt_endio(struct bio *enc_bio) bio_endio(src_bio); } -static struct bio *blk_crypto_clone_bio(struct bio *bio_src) +static struct bio *blk_crypto_fallback_clone_bio(struct bio *bio_src) { struct bvec_iter iter; struct bio_vec bv; @@ -186,13 +183,14 @@ static struct bio *blk_crypto_clone_bio(struct bio *bio_src) return bio; } -static bool blk_crypto_alloc_cipher_req(struct blk_ksm_keyslot *slot, - struct skcipher_request **ciph_req_ret, - struct crypto_wait *wait) +static bool +blk_crypto_fallback_alloc_cipher_req(struct blk_crypto_keyslot *slot, + struct skcipher_request **ciph_req_ret, + struct crypto_wait *wait) { struct skcipher_request *ciph_req; - const struct blk_crypto_keyslot *slotp; - int keyslot_idx = blk_ksm_get_slot_idx(slot); + const struct blk_crypto_fallback_keyslot *slotp; + int keyslot_idx = blk_crypto_keyslot_index(slot); slotp = &blk_crypto_keyslots[keyslot_idx]; ciph_req = skcipher_request_alloc(slotp->tfms[slotp->crypto_mode], @@ -209,7 +207,7 @@ static bool blk_crypto_alloc_cipher_req(struct blk_ksm_keyslot *slot, return true; } -static bool blk_crypto_split_bio_if_needed(struct bio **bio_ptr) +static bool blk_crypto_fallback_split_bio_if_needed(struct bio **bio_ptr) { struct bio *bio = *bio_ptr; unsigned int i = 0; @@ -264,7 +262,7 @@ static bool blk_crypto_fallback_encrypt_bio(struct bio **bio_ptr) { struct bio *src_bio, *enc_bio; struct bio_crypt_ctx *bc; - struct blk_ksm_keyslot *slot; + struct blk_crypto_keyslot *slot; int data_unit_size; struct skcipher_request *ciph_req = NULL; DECLARE_CRYPTO_WAIT(wait); @@ -276,7 +274,7 @@ static bool blk_crypto_fallback_encrypt_bio(struct bio **bio_ptr) blk_status_t blk_st; /* Split the bio if it's too big for single page bvec */ - if (!blk_crypto_split_bio_if_needed(bio_ptr)) + if (!blk_crypto_fallback_split_bio_if_needed(bio_ptr)) return false; src_bio = *bio_ptr; @@ -284,24 +282,25 @@ static bool blk_crypto_fallback_encrypt_bio(struct bio **bio_ptr) data_unit_size = bc->bc_key->crypto_cfg.data_unit_size; /* Allocate bounce bio for encryption */ - enc_bio = blk_crypto_clone_bio(src_bio); + enc_bio = blk_crypto_fallback_clone_bio(src_bio); if (!enc_bio) { src_bio->bi_status = BLK_STS_RESOURCE; return false; } /* - * Use the crypto API fallback keyslot manager to get a crypto_skcipher - * for the algorithm and key specified for this bio. + * Get a blk-crypto-fallback keyslot that contains a crypto_skcipher for + * this bio's algorithm and key. */ - blk_st = blk_ksm_get_slot_for_key(&blk_crypto_ksm, bc->bc_key, &slot); + blk_st = blk_crypto_get_keyslot(&blk_crypto_fallback_profile, + bc->bc_key, &slot); if (blk_st != BLK_STS_OK) { src_bio->bi_status = blk_st; goto out_put_enc_bio; } /* and then allocate an skcipher_request for it */ - if (!blk_crypto_alloc_cipher_req(slot, &ciph_req, &wait)) { + if (!blk_crypto_fallback_alloc_cipher_req(slot, &ciph_req, &wait)) { src_bio->bi_status = BLK_STS_RESOURCE; goto out_release_keyslot; } @@ -362,7 +361,7 @@ out_free_bounce_pages: out_free_ciph_req: skcipher_request_free(ciph_req); out_release_keyslot: - blk_ksm_put_slot(slot); + blk_crypto_put_keyslot(slot); out_put_enc_bio: if (enc_bio) bio_put(enc_bio); @@ -380,7 +379,7 @@ static void blk_crypto_fallback_decrypt_bio(struct work_struct *work) container_of(work, struct bio_fallback_crypt_ctx, work); struct bio *bio = f_ctx->bio; struct bio_crypt_ctx *bc = &f_ctx->crypt_ctx; - struct blk_ksm_keyslot *slot; + struct blk_crypto_keyslot *slot; struct skcipher_request *ciph_req = NULL; DECLARE_CRYPTO_WAIT(wait); u64 curr_dun[BLK_CRYPTO_DUN_ARRAY_SIZE]; @@ -393,17 +392,18 @@ static void blk_crypto_fallback_decrypt_bio(struct work_struct *work) blk_status_t blk_st; /* - * Use the crypto API fallback keyslot manager to get a crypto_skcipher - * for the algorithm and key specified for this bio. + * Get a blk-crypto-fallback keyslot that contains a crypto_skcipher for + * this bio's algorithm and key. */ - blk_st = blk_ksm_get_slot_for_key(&blk_crypto_ksm, bc->bc_key, &slot); + blk_st = blk_crypto_get_keyslot(&blk_crypto_fallback_profile, + bc->bc_key, &slot); if (blk_st != BLK_STS_OK) { bio->bi_status = blk_st; goto out_no_keyslot; } /* and then allocate an skcipher_request for it */ - if (!blk_crypto_alloc_cipher_req(slot, &ciph_req, &wait)) { + if (!blk_crypto_fallback_alloc_cipher_req(slot, &ciph_req, &wait)) { bio->bi_status = BLK_STS_RESOURCE; goto out; } @@ -434,7 +434,7 @@ static void blk_crypto_fallback_decrypt_bio(struct work_struct *work) out: skcipher_request_free(ciph_req); - blk_ksm_put_slot(slot); + blk_crypto_put_keyslot(slot); out_no_keyslot: mempool_free(f_ctx, bio_fallback_crypt_ctx_pool); bio_endio(bio); @@ -473,9 +473,9 @@ static void blk_crypto_fallback_decrypt_endio(struct bio *bio) * @bio_ptr: pointer to the bio to prepare * * If bio is doing a WRITE operation, this splits the bio into two parts if it's - * too big (see blk_crypto_split_bio_if_needed). It then allocates a bounce bio - * for the first part, encrypts it, and update bio_ptr to point to the bounce - * bio. + * too big (see blk_crypto_fallback_split_bio_if_needed()). It then allocates a + * bounce bio for the first part, encrypts it, and updates bio_ptr to point to + * the bounce bio. * * For a READ operation, we mark the bio for decryption by using bi_private and * bi_end_io. @@ -499,8 +499,8 @@ bool blk_crypto_fallback_bio_prep(struct bio **bio_ptr) return false; } - if (!blk_ksm_crypto_cfg_supported(&blk_crypto_ksm, - &bc->bc_key->crypto_cfg)) { + if (!__blk_crypto_cfg_supported(&blk_crypto_fallback_profile, + &bc->bc_key->crypto_cfg)) { bio->bi_status = BLK_STS_NOTSUPP; return false; } @@ -526,7 +526,7 @@ bool blk_crypto_fallback_bio_prep(struct bio **bio_ptr) int blk_crypto_fallback_evict_key(const struct blk_crypto_key *key) { - return blk_ksm_evict_key(&blk_crypto_ksm, key); + return __blk_crypto_evict_key(&blk_crypto_fallback_profile, key); } static bool blk_crypto_fallback_inited; @@ -534,6 +534,7 @@ static int blk_crypto_fallback_init(void) { int i; int err; + struct blk_crypto_profile *profile = &blk_crypto_fallback_profile; if (blk_crypto_fallback_inited) return 0; @@ -544,24 +545,24 @@ static int blk_crypto_fallback_init(void) if (err) goto out; - err = blk_ksm_init(&blk_crypto_ksm, blk_crypto_num_keyslots); + err = blk_crypto_profile_init(profile, blk_crypto_num_keyslots); if (err) goto fail_free_bioset; err = -ENOMEM; - blk_crypto_ksm.ksm_ll_ops = blk_crypto_ksm_ll_ops; - blk_crypto_ksm.max_dun_bytes_supported = BLK_CRYPTO_MAX_IV_SIZE; + profile->ll_ops = blk_crypto_fallback_ll_ops; + profile->max_dun_bytes_supported = BLK_CRYPTO_MAX_IV_SIZE; /* All blk-crypto modes have a crypto API fallback. */ for (i = 0; i < BLK_ENCRYPTION_MODE_MAX; i++) - blk_crypto_ksm.crypto_modes_supported[i] = 0xFFFFFFFF; - blk_crypto_ksm.crypto_modes_supported[BLK_ENCRYPTION_MODE_INVALID] = 0; + profile->modes_supported[i] = 0xFFFFFFFF; + profile->modes_supported[BLK_ENCRYPTION_MODE_INVALID] = 0; blk_crypto_wq = alloc_workqueue("blk_crypto_wq", WQ_UNBOUND | WQ_HIGHPRI | WQ_MEM_RECLAIM, num_online_cpus()); if (!blk_crypto_wq) - goto fail_free_ksm; + goto fail_destroy_profile; blk_crypto_keyslots = kcalloc(blk_crypto_num_keyslots, sizeof(blk_crypto_keyslots[0]), @@ -595,8 +596,8 @@ fail_free_keyslots: kfree(blk_crypto_keyslots); fail_free_wq: destroy_workqueue(blk_crypto_wq); -fail_free_ksm: - blk_ksm_destroy(&blk_crypto_ksm); +fail_destroy_profile: + blk_crypto_profile_destroy(profile); fail_free_bioset: bioset_exit(&crypto_bio_split); out: @@ -610,7 +611,7 @@ out: int blk_crypto_fallback_start_using_mode(enum blk_crypto_mode_num mode_num) { const char *cipher_str = blk_crypto_modes[mode_num].cipher_str; - struct blk_crypto_keyslot *slotp; + struct blk_crypto_fallback_keyslot *slotp; unsigned int i; int err = 0; diff --git a/block/blk-crypto-internal.h b/block/blk-crypto-internal.h index 0d36aae538d7..2fb0d65a464c 100644 --- a/block/blk-crypto-internal.h +++ b/block/blk-crypto-internal.h @@ -7,7 +7,7 @@ #define __LINUX_BLK_CRYPTO_INTERNAL_H #include <linux/bio.h> -#include <linux/blkdev.h> +#include <linux/blk-mq.h> /* Represents a crypto mode supported by blk-crypto */ struct blk_crypto_mode { diff --git a/block/blk-crypto-profile.c b/block/blk-crypto-profile.c new file mode 100644 index 000000000000..605ba0626a5c --- /dev/null +++ b/block/blk-crypto-profile.c @@ -0,0 +1,565 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright 2019 Google LLC + */ + +/** + * DOC: blk-crypto profiles + * + * 'struct blk_crypto_profile' contains all generic inline encryption-related + * state for a particular inline encryption device. blk_crypto_profile serves + * as the way that drivers for inline encryption hardware expose their crypto + * capabilities and certain functions (e.g., functions to program and evict + * keys) to upper layers. Device drivers that want to support inline encryption + * construct a crypto profile, then associate it with the disk's request_queue. + * + * If the device has keyslots, then its blk_crypto_profile also handles managing + * these keyslots in a device-independent way, using the driver-provided + * functions to program and evict keys as needed. This includes keeping track + * of which key and how many I/O requests are using each keyslot, getting + * keyslots for I/O requests, and handling key eviction requests. + * + * For more information, see Documentation/block/inline-encryption.rst. + */ + +#define pr_fmt(fmt) "blk-crypto: " fmt + +#include <linux/blk-crypto-profile.h> +#include <linux/device.h> +#include <linux/atomic.h> +#include <linux/mutex.h> +#include <linux/pm_runtime.h> +#include <linux/wait.h> +#include <linux/blkdev.h> +#include <linux/blk-integrity.h> + +struct blk_crypto_keyslot { + atomic_t slot_refs; + struct list_head idle_slot_node; + struct hlist_node hash_node; + const struct blk_crypto_key *key; + struct blk_crypto_profile *profile; +}; + +static inline void blk_crypto_hw_enter(struct blk_crypto_profile *profile) +{ + /* + * Calling into the driver requires profile->lock held and the device + * resumed. But we must resume the device first, since that can acquire + * and release profile->lock via blk_crypto_reprogram_all_keys(). + */ + if (profile->dev) + pm_runtime_get_sync(profile->dev); + down_write(&profile->lock); +} + +static inline void blk_crypto_hw_exit(struct blk_crypto_profile *profile) +{ + up_write(&profile->lock); + if (profile->dev) + pm_runtime_put_sync(profile->dev); +} + +/** + * blk_crypto_profile_init() - Initialize a blk_crypto_profile + * @profile: the blk_crypto_profile to initialize + * @num_slots: the number of keyslots + * + * Storage drivers must call this when starting to set up a blk_crypto_profile, + * before filling in additional fields. + * + * Return: 0 on success, or else a negative error code. + */ +int blk_crypto_profile_init(struct blk_crypto_profile *profile, + unsigned int num_slots) +{ + unsigned int slot; + unsigned int i; + unsigned int slot_hashtable_size; + + memset(profile, 0, sizeof(*profile)); + init_rwsem(&profile->lock); + + if (num_slots == 0) + return 0; + + /* Initialize keyslot management data. */ + + profile->slots = kvcalloc(num_slots, sizeof(profile->slots[0]), + GFP_KERNEL); + if (!profile->slots) + return -ENOMEM; + + profile->num_slots = num_slots; + + init_waitqueue_head(&profile->idle_slots_wait_queue); + INIT_LIST_HEAD(&profile->idle_slots); + + for (slot = 0; slot < num_slots; slot++) { + profile->slots[slot].profile = profile; + list_add_tail(&profile->slots[slot].idle_slot_node, + &profile->idle_slots); + } + + spin_lock_init(&profile->idle_slots_lock); + + slot_hashtable_size = roundup_pow_of_two(num_slots); + /* + * hash_ptr() assumes bits != 0, so ensure the hash table has at least 2 + * buckets. This only makes a difference when there is only 1 keyslot. + */ + if (slot_hashtable_size < 2) + slot_hashtable_size = 2; + + profile->log_slot_ht_size = ilog2(slot_hashtable_size); + profile->slot_hashtable = + kvmalloc_array(slot_hashtable_size, + sizeof(profile->slot_hashtable[0]), GFP_KERNEL); + if (!profile->slot_hashtable) + goto err_destroy; + for (i = 0; i < slot_hashtable_size; i++) + INIT_HLIST_HEAD(&profile->slot_hashtable[i]); + + return 0; + +err_destroy: + blk_crypto_profile_destroy(profile); + return -ENOMEM; +} +EXPORT_SYMBOL_GPL(blk_crypto_profile_init); + +static void blk_crypto_profile_destroy_callback(void *profile) +{ + blk_crypto_profile_destroy(profile); +} + +/** + * devm_blk_crypto_profile_init() - Resource-managed blk_crypto_profile_init() + * @dev: the device which owns the blk_crypto_profile + * @profile: the blk_crypto_profile to initialize + * @num_slots: the number of keyslots + * + * Like blk_crypto_profile_init(), but causes blk_crypto_profile_destroy() to be + * called automatically on driver detach. + * + * Return: 0 on success, or else a negative error code. + */ +int devm_blk_crypto_profile_init(struct device *dev, + struct blk_crypto_profile *profile, + unsigned int num_slots) +{ + int err = blk_crypto_profile_init(profile, num_slots); + + if (err) + return err; + + return devm_add_action_or_reset(dev, + blk_crypto_profile_destroy_callback, + profile); +} +EXPORT_SYMBOL_GPL(devm_blk_crypto_profile_init); + +static inline struct hlist_head * +blk_crypto_hash_bucket_for_key(struct blk_crypto_profile *profile, + const struct blk_crypto_key *key) +{ + return &profile->slot_hashtable[ + hash_ptr(key, profile->log_slot_ht_size)]; +} + +static void +blk_crypto_remove_slot_from_lru_list(struct blk_crypto_keyslot *slot) +{ + struct blk_crypto_profile *profile = slot->profile; + unsigned long flags; + + spin_lock_irqsave(&profile->idle_slots_lock, flags); + list_del(&slot->idle_slot_node); + spin_unlock_irqrestore(&profile->idle_slots_lock, flags); +} + +static struct blk_crypto_keyslot * +blk_crypto_find_keyslot(struct blk_crypto_profile *profile, + const struct blk_crypto_key *key) +{ + const struct hlist_head *head = + blk_crypto_hash_bucket_for_key(profile, key); + struct blk_crypto_keyslot *slotp; + + hlist_for_each_entry(slotp, head, hash_node) { + if (slotp->key == key) + return slotp; + } + return NULL; +} + +static struct blk_crypto_keyslot * +blk_crypto_find_and_grab_keyslot(struct blk_crypto_profile *profile, + const struct blk_crypto_key *key) +{ + struct blk_crypto_keyslot *slot; + + slot = blk_crypto_find_keyslot(profile, key); + if (!slot) + return NULL; + if (atomic_inc_return(&slot->slot_refs) == 1) { + /* Took first reference to this slot; remove it from LRU list */ + blk_crypto_remove_slot_from_lru_list(slot); + } + return slot; +} + +/** + * blk_crypto_keyslot_index() - Get the index of a keyslot + * @slot: a keyslot that blk_crypto_get_keyslot() returned + * + * Return: the 0-based index of the keyslot within the device's keyslots. + */ +unsigned int blk_crypto_keyslot_index(struct blk_crypto_keyslot *slot) +{ + return slot - slot->profile->slots; +} +EXPORT_SYMBOL_GPL(blk_crypto_keyslot_index); + +/** + * blk_crypto_get_keyslot() - Get a keyslot for a key, if needed. + * @profile: the crypto profile of the device the key will be used on + * @key: the key that will be used + * @slot_ptr: If a keyslot is allocated, an opaque pointer to the keyslot struct + * will be stored here; otherwise NULL will be stored here. + * + * If the device has keyslots, this gets a keyslot that's been programmed with + * the specified key. If the key is already in a slot, this reuses it; + * otherwise this waits for a slot to become idle and programs the key into it. + * + * This must be paired with a call to blk_crypto_put_keyslot(). + * + * Context: Process context. Takes and releases profile->lock. + * Return: BLK_STS_OK on success, meaning that either a keyslot was allocated or + * one wasn't needed; or a blk_status_t error on failure. + */ +blk_status_t blk_crypto_get_keyslot(struct blk_crypto_profile *profile, + const struct blk_crypto_key *key, + struct blk_crypto_keyslot **slot_ptr) +{ + struct blk_crypto_keyslot *slot; + int slot_idx; + int err; + + *slot_ptr = NULL; + + /* + * If the device has no concept of "keyslots", then there is no need to + * get one. + */ + if (profile->num_slots == 0) + return BLK_STS_OK; + + down_read(&profile->lock); + slot = blk_crypto_find_and_grab_keyslot(profile, key); + up_read(&profile->lock); + if (slot) + goto success; + + for (;;) { + blk_crypto_hw_enter(profile); + slot = blk_crypto_find_and_grab_keyslot(profile, key); + if (slot) { + blk_crypto_hw_exit(profile); + goto success; + } + + /* + * If we're here, that means there wasn't a slot that was + * already programmed with the key. So try to program it. + */ + if (!list_empty(&profile->idle_slots)) + break; + + blk_crypto_hw_exit(profile); + wait_event(profile->idle_slots_wait_queue, + !list_empty(&profile->idle_slots)); + } + + slot = list_first_entry(&profile->idle_slots, struct blk_crypto_keyslot, + idle_slot_node); + slot_idx = blk_crypto_keyslot_index(slot); + + err = profile->ll_ops.keyslot_program(profile, key, slot_idx); + if (err) { + wake_up(&profile->idle_slots_wait_queue); + blk_crypto_hw_exit(profile); + return errno_to_blk_status(err); + } + + /* Move this slot to the hash list for the new key. */ + if (slot->key) + hlist_del(&slot->hash_node); + slot->key = key; + hlist_add_head(&slot->hash_node, + blk_crypto_hash_bucket_for_key(profile, key)); + + atomic_set(&slot->slot_refs, 1); + + blk_crypto_remove_slot_from_lru_list(slot); + + blk_crypto_hw_exit(profile); +success: + *slot_ptr = slot; + return BLK_STS_OK; +} + +/** + * blk_crypto_put_keyslot() - Release a reference to a keyslot + * @slot: The keyslot to release the reference of (may be NULL). + * + * Context: Any context. + */ +void blk_crypto_put_keyslot(struct blk_crypto_keyslot *slot) +{ + struct blk_crypto_profile *profile; + unsigned long flags; + + if (!slot) + return; + + profile = slot->profile; + + if (atomic_dec_and_lock_irqsave(&slot->slot_refs, + &profile->idle_slots_lock, flags)) { + list_add_tail(&slot->idle_slot_node, &profile->idle_slots); + spin_unlock_irqrestore(&profile->idle_slots_lock, flags); + wake_up(&profile->idle_slots_wait_queue); + } +} + +/** + * __blk_crypto_cfg_supported() - Check whether the given crypto profile + * supports the given crypto configuration. + * @profile: the crypto profile to check + * @cfg: the crypto configuration to check for + * + * Return: %true if @profile supports the given @cfg. + */ +bool __blk_crypto_cfg_supported(struct blk_crypto_profile *profile, + const struct blk_crypto_config *cfg) +{ + if (!profile) + return false; + if (!(profile->modes_supported[cfg->crypto_mode] & cfg->data_unit_size)) + return false; + if (profile->max_dun_bytes_supported < cfg->dun_bytes) + return false; + return true; +} + +/** + * __blk_crypto_evict_key() - Evict a key from a device. + * @profile: the crypto profile of the device + * @key: the key to evict. It must not still be used in any I/O. + * + * If the device has keyslots, this finds the keyslot (if any) that contains the + * specified key and calls the driver's keyslot_evict function to evict it. + * + * Otherwise, this just calls the driver's keyslot_evict function if it is + * implemented, passing just the key (without any particular keyslot). This + * allows layered devices to evict the key from their underlying devices. + * + * Context: Process context. Takes and releases profile->lock. + * Return: 0 on success or if there's no keyslot with the specified key, -EBUSY + * if the keyslot is still in use, or another -errno value on other + * error. + */ +int __blk_crypto_evict_key(struct blk_crypto_profile *profile, + const struct blk_crypto_key *key) +{ + struct blk_crypto_keyslot *slot; + int err = 0; + + if (profile->num_slots == 0) { + if (profile->ll_ops.keyslot_evict) { + blk_crypto_hw_enter(profile); + err = profile->ll_ops.keyslot_evict(profile, key, -1); + blk_crypto_hw_exit(profile); + return err; + } + return 0; + } + + blk_crypto_hw_enter(profile); + slot = blk_crypto_find_keyslot(profile, key); + if (!slot) + goto out_unlock; + + if (WARN_ON_ONCE(atomic_read(&slot->slot_refs) != 0)) { + err = -EBUSY; + goto out_unlock; + } + err = profile->ll_ops.keyslot_evict(profile, key, + blk_crypto_keyslot_index(slot)); + if (err) + goto out_unlock; + + hlist_del(&slot->hash_node); + slot->key = NULL; + err = 0; +out_unlock: + blk_crypto_hw_exit(profile); + return err; +} + +/** + * blk_crypto_reprogram_all_keys() - Re-program all keyslots. + * @profile: The crypto profile + * + * Re-program all keyslots that are supposed to have a key programmed. This is + * intended only for use by drivers for hardware that loses its keys on reset. + * + * Context: Process context. Takes and releases profile->lock. + */ +void blk_crypto_reprogram_all_keys(struct blk_crypto_profile *profile) +{ + unsigned int slot; + + if (profile->num_slots == 0) + return; + + /* This is for device initialization, so don't resume the device */ + down_write(&profile->lock); + for (slot = 0; slot < profile->num_slots; slot++) { + const struct blk_crypto_key *key = profile->slots[slot].key; + int err; + + if (!key) + continue; + + err = profile->ll_ops.keyslot_program(profile, key, slot); + WARN_ON(err); + } + up_write(&profile->lock); +} +EXPORT_SYMBOL_GPL(blk_crypto_reprogram_all_keys); + +void blk_crypto_profile_destroy(struct blk_crypto_profile *profile) +{ + if (!profile) + return; + kvfree(profile->slot_hashtable); + kvfree_sensitive(profile->slots, + sizeof(profile->slots[0]) * profile->num_slots); + memzero_explicit(profile, sizeof(*profile)); +} +EXPORT_SYMBOL_GPL(blk_crypto_profile_destroy); + +bool blk_crypto_register(struct blk_crypto_profile *profile, + struct request_queue *q) +{ + if (blk_integrity_queue_supports_integrity(q)) { + pr_warn("Integrity and hardware inline encryption are not supported together. Disabling hardware inline encryption.\n"); + return false; + } + q->crypto_profile = profile; + return true; +} +EXPORT_SYMBOL_GPL(blk_crypto_register); + +void blk_crypto_unregister(struct request_queue *q) +{ + q->crypto_profile = NULL; +} + +/** + * blk_crypto_intersect_capabilities() - restrict supported crypto capabilities + * by child device + * @parent: the crypto profile for the parent device + * @child: the crypto profile for the child device, or NULL + * + * This clears all crypto capabilities in @parent that aren't set in @child. If + * @child is NULL, then this clears all parent capabilities. + * + * Only use this when setting up the crypto profile for a layered device, before + * it's been exposed yet. + */ +void blk_crypto_intersect_capabilities(struct blk_crypto_profile *parent, + const struct blk_crypto_profile *child) +{ + if (child) { + unsigned int i; + + parent->max_dun_bytes_supported = + min(parent->max_dun_bytes_supported, + child->max_dun_bytes_supported); + for (i = 0; i < ARRAY_SIZE(child->modes_supported); i++) + parent->modes_supported[i] &= child->modes_supported[i]; + } else { + parent->max_dun_bytes_supported = 0; + memset(parent->modes_supported, 0, + sizeof(parent->modes_supported)); + } +} +EXPORT_SYMBOL_GPL(blk_crypto_intersect_capabilities); + +/** + * blk_crypto_has_capabilities() - Check whether @target supports at least all + * the crypto capabilities that @reference does. + * @target: the target profile + * @reference: the reference profile + * + * Return: %true if @target supports all the crypto capabilities of @reference. + */ +bool blk_crypto_has_capabilities(const struct blk_crypto_profile *target, + const struct blk_crypto_profile *reference) +{ + int i; + + if (!reference) + return true; + + if (!target) + return false; + + for (i = 0; i < ARRAY_SIZE(target->modes_supported); i++) { + if (reference->modes_supported[i] & ~target->modes_supported[i]) + return false; + } + + if (reference->max_dun_bytes_supported > + target->max_dun_bytes_supported) + return false; + + return true; +} +EXPORT_SYMBOL_GPL(blk_crypto_has_capabilities); + +/** + * blk_crypto_update_capabilities() - Update the capabilities of a crypto + * profile to match those of another crypto + * profile. + * @dst: The crypto profile whose capabilities to update. + * @src: The crypto profile whose capabilities this function will update @dst's + * capabilities to. + * + * Blk-crypto requires that crypto capabilities that were + * advertised when a bio was created continue to be supported by the + * device until that bio is ended. This is turn means that a device cannot + * shrink its advertised crypto capabilities without any explicit + * synchronization with upper layers. So if there's no such explicit + * synchronization, @src must support all the crypto capabilities that + * @dst does (i.e. we need blk_crypto_has_capabilities(@src, @dst)). + * + * Note also that as long as the crypto capabilities are being expanded, the + * order of updates becoming visible is not important because it's alright + * for blk-crypto to see stale values - they only cause blk-crypto to + * believe that a crypto capability isn't supported when it actually is (which + * might result in blk-crypto-fallback being used if available, or the bio being + * failed). + */ +void blk_crypto_update_capabilities(struct blk_crypto_profile *dst, + const struct blk_crypto_profile *src) +{ + memcpy(dst->modes_supported, src->modes_supported, + sizeof(dst->modes_supported)); + + dst->max_dun_bytes_supported = src->max_dun_bytes_supported; +} +EXPORT_SYMBOL_GPL(blk_crypto_update_capabilities); diff --git a/block/blk-crypto.c b/block/blk-crypto.c index 103c2e2d50d6..ec9efeeeca91 100644 --- a/block/blk-crypto.c +++ b/block/blk-crypto.c @@ -11,7 +11,7 @@ #include <linux/bio.h> #include <linux/blkdev.h> -#include <linux/keyslot-manager.h> +#include <linux/blk-crypto-profile.h> #include <linux/module.h> #include <linux/slab.h> @@ -218,8 +218,9 @@ static bool bio_crypt_check_alignment(struct bio *bio) blk_status_t __blk_crypto_init_request(struct request *rq) { - return blk_ksm_get_slot_for_key(rq->q->ksm, rq->crypt_ctx->bc_key, - &rq->crypt_keyslot); + return blk_crypto_get_keyslot(rq->q->crypto_profile, + rq->crypt_ctx->bc_key, + &rq->crypt_keyslot); } /** @@ -233,7 +234,7 @@ blk_status_t __blk_crypto_init_request(struct request *rq) */ void __blk_crypto_free_request(struct request *rq) { - blk_ksm_put_slot(rq->crypt_keyslot); + blk_crypto_put_keyslot(rq->crypt_keyslot); mempool_free(rq->crypt_ctx, bio_crypt_ctx_pool); blk_crypto_rq_set_defaults(rq); } @@ -264,6 +265,7 @@ bool __blk_crypto_bio_prep(struct bio **bio_ptr) { struct bio *bio = *bio_ptr; const struct blk_crypto_key *bc_key = bio->bi_crypt_context->bc_key; + struct blk_crypto_profile *profile; /* Error if bio has no data. */ if (WARN_ON_ONCE(!bio_has_data(bio))) { @@ -280,8 +282,8 @@ bool __blk_crypto_bio_prep(struct bio **bio_ptr) * Success if device supports the encryption context, or if we succeeded * in falling back to the crypto API. */ - if (blk_ksm_crypto_cfg_supported(bio->bi_bdev->bd_disk->queue->ksm, - &bc_key->crypto_cfg)) + profile = bdev_get_queue(bio->bi_bdev)->crypto_profile; + if (__blk_crypto_cfg_supported(profile, &bc_key->crypto_cfg)) return true; if (blk_crypto_fallback_bio_prep(bio_ptr)) @@ -357,7 +359,7 @@ bool blk_crypto_config_supported(struct request_queue *q, const struct blk_crypto_config *cfg) { return IS_ENABLED(CONFIG_BLK_INLINE_ENCRYPTION_FALLBACK) || - blk_ksm_crypto_cfg_supported(q->ksm, cfg); + __blk_crypto_cfg_supported(q->crypto_profile, cfg); } /** @@ -378,7 +380,7 @@ bool blk_crypto_config_supported(struct request_queue *q, int blk_crypto_start_using_key(const struct blk_crypto_key *key, struct request_queue *q) { - if (blk_ksm_crypto_cfg_supported(q->ksm, &key->crypto_cfg)) + if (__blk_crypto_cfg_supported(q->crypto_profile, &key->crypto_cfg)) return 0; return blk_crypto_fallback_start_using_mode(key->crypto_cfg.crypto_mode); } @@ -394,18 +396,17 @@ int blk_crypto_start_using_key(const struct blk_crypto_key *key, * evicted from any hardware that it might have been programmed into. The key * must not be in use by any in-flight IO when this function is called. * - * Return: 0 on success or if key is not present in the q's ksm, -err on error. + * Return: 0 on success or if the key wasn't in any keyslot; -errno on error. */ int blk_crypto_evict_key(struct request_queue *q, const struct blk_crypto_key *key) { - if (blk_ksm_crypto_cfg_supported(q->ksm, &key->crypto_cfg)) - return blk_ksm_evict_key(q->ksm, key); + if (__blk_crypto_cfg_supported(q->crypto_profile, &key->crypto_cfg)) + return __blk_crypto_evict_key(q->crypto_profile, key); /* - * If the request queue's associated inline encryption hardware didn't - * have support for the key, then the key might have been programmed - * into the fallback keyslot manager, so try to evict from there. + * If the request_queue didn't support the key, then blk-crypto-fallback + * may have been used, so try to evict the key from blk-crypto-fallback. */ return blk_crypto_fallback_evict_key(key); } diff --git a/block/blk-exec.c b/block/blk-exec.c index d6cd501c0d34..1b8b47f6e79b 100644 --- a/block/blk-exec.c +++ b/block/blk-exec.c @@ -65,13 +65,19 @@ EXPORT_SYMBOL_GPL(blk_execute_rq_nowait); static bool blk_rq_is_poll(struct request *rq) { - return rq->mq_hctx && rq->mq_hctx->type == HCTX_TYPE_POLL; + if (!rq->mq_hctx) + return false; + if (rq->mq_hctx->type != HCTX_TYPE_POLL) + return false; + if (WARN_ON_ONCE(!rq->bio)) + return false; + return true; } static void blk_rq_poll_completion(struct request *rq, struct completion *wait) { do { - blk_poll(rq->q, request_to_qc_t(rq->mq_hctx, rq), true); + bio_poll(rq->bio, NULL, 0); cond_resched(); } while (!completion_done(wait)); } diff --git a/block/blk-flush.c b/block/blk-flush.c index 4201728bf3a5..8e364bda5166 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -379,7 +379,7 @@ static void mq_flush_data_end_io(struct request *rq, blk_status_t error) * @rq is being submitted. Analyze what needs to be done and put it on the * right queue. */ -void blk_insert_flush(struct request *rq) +bool blk_insert_flush(struct request *rq) { struct request_queue *q = rq->q; unsigned long fflags = q->queue_flags; /* may change, cache */ @@ -409,7 +409,7 @@ void blk_insert_flush(struct request *rq) */ if (!policy) { blk_mq_end_request(rq, 0); - return; + return true; } BUG_ON(rq->bio != rq->biotail); /*assumes zero or single bio rq */ @@ -420,10 +420,8 @@ void blk_insert_flush(struct request *rq) * for normal execution. */ if ((policy & REQ_FSEQ_DATA) && - !(policy & (REQ_FSEQ_PREFLUSH | REQ_FSEQ_POSTFLUSH))) { - blk_mq_request_bypass_insert(rq, false, false); - return; - } + !(policy & (REQ_FSEQ_PREFLUSH | REQ_FSEQ_POSTFLUSH))) + return false; /* * @rq should go through flush machinery. Mark it part of flush @@ -439,6 +437,8 @@ void blk_insert_flush(struct request *rq) spin_lock_irq(&fq->mq_flush_lock); blk_flush_complete_seq(rq, fq, REQ_FSEQ_ACTIONS & ~policy, 0); spin_unlock_irq(&fq->mq_flush_lock); + + return true; } /** diff --git a/block/blk-ia-ranges.c b/block/blk-ia-ranges.c new file mode 100644 index 000000000000..c246c425d0d7 --- /dev/null +++ b/block/blk-ia-ranges.c @@ -0,0 +1,348 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Block device concurrent positioning ranges. + * + * Copyright (C) 2021 Western Digital Corporation or its Affiliates. + */ +#include <linux/kernel.h> +#include <linux/blkdev.h> +#include <linux/slab.h> +#include <linux/init.h> + +#include "blk.h" + +static ssize_t +blk_ia_range_sector_show(struct blk_independent_access_range *iar, + char *buf) +{ + return sprintf(buf, "%llu\n", iar->sector); +} + +static ssize_t +blk_ia_range_nr_sectors_show(struct blk_independent_access_range *iar, + char *buf) +{ + return sprintf(buf, "%llu\n", iar->nr_sectors); +} + +struct blk_ia_range_sysfs_entry { + struct attribute attr; + ssize_t (*show)(struct blk_independent_access_range *iar, char *buf); +}; + +static struct blk_ia_range_sysfs_entry blk_ia_range_sector_entry = { + .attr = { .name = "sector", .mode = 0444 }, + .show = blk_ia_range_sector_show, +}; + +static struct blk_ia_range_sysfs_entry blk_ia_range_nr_sectors_entry = { + .attr = { .name = "nr_sectors", .mode = 0444 }, + .show = blk_ia_range_nr_sectors_show, +}; + +static struct attribute *blk_ia_range_attrs[] = { + &blk_ia_range_sector_entry.attr, + &blk_ia_range_nr_sectors_entry.attr, + NULL, +}; +ATTRIBUTE_GROUPS(blk_ia_range); + +static ssize_t blk_ia_range_sysfs_show(struct kobject *kobj, + struct attribute *attr, char *buf) +{ + struct blk_ia_range_sysfs_entry *entry = + container_of(attr, struct blk_ia_range_sysfs_entry, attr); + struct blk_independent_access_range *iar = + container_of(kobj, struct blk_independent_access_range, kobj); + ssize_t ret; + + mutex_lock(&iar->queue->sysfs_lock); + ret = entry->show(iar, buf); + mutex_unlock(&iar->queue->sysfs_lock); + + return ret; +} + +static const struct sysfs_ops blk_ia_range_sysfs_ops = { + .show = blk_ia_range_sysfs_show, +}; + +/* + * Independent access range entries are not freed individually, but alltogether + * with struct blk_independent_access_ranges and its array of ranges. Since + * kobject_add() takes a reference on the parent kobject contained in + * struct blk_independent_access_ranges, the array of independent access range + * entries cannot be freed until kobject_del() is called for all entries. + * So we do not need to do anything here, but still need this no-op release + * operation to avoid complaints from the kobject code. + */ +static void blk_ia_range_sysfs_nop_release(struct kobject *kobj) +{ +} + +static struct kobj_type blk_ia_range_ktype = { + .sysfs_ops = &blk_ia_range_sysfs_ops, + .default_groups = blk_ia_range_groups, + .release = blk_ia_range_sysfs_nop_release, +}; + +/* + * This will be executed only after all independent access range entries are + * removed with kobject_del(), at which point, it is safe to free everything, + * including the array of ranges. + */ +static void blk_ia_ranges_sysfs_release(struct kobject *kobj) +{ + struct blk_independent_access_ranges *iars = + container_of(kobj, struct blk_independent_access_ranges, kobj); + + kfree(iars); +} + +static struct kobj_type blk_ia_ranges_ktype = { + .release = blk_ia_ranges_sysfs_release, +}; + +/** + * disk_register_ia_ranges - register with sysfs a set of independent + * access ranges + * @disk: Target disk + * @new_iars: New set of independent access ranges + * + * Register with sysfs a set of independent access ranges for @disk. + * If @new_iars is not NULL, this set of ranges is registered and the old set + * specified by q->ia_ranges is unregistered. Otherwise, q->ia_ranges is + * registered if it is not already. + */ +int disk_register_independent_access_ranges(struct gendisk *disk, + struct blk_independent_access_ranges *new_iars) +{ + struct request_queue *q = disk->queue; + struct blk_independent_access_ranges *iars; + int i, ret; + + lockdep_assert_held(&q->sysfs_dir_lock); + lockdep_assert_held(&q->sysfs_lock); + + /* If a new range set is specified, unregister the old one */ + if (new_iars) { + if (q->ia_ranges) + disk_unregister_independent_access_ranges(disk); + q->ia_ranges = new_iars; + } + + iars = q->ia_ranges; + if (!iars) + return 0; + + /* + * At this point, iars is the new set of sector access ranges that needs + * to be registered with sysfs. + */ + WARN_ON(iars->sysfs_registered); + ret = kobject_init_and_add(&iars->kobj, &blk_ia_ranges_ktype, + &q->kobj, "%s", "independent_access_ranges"); + if (ret) { + q->ia_ranges = NULL; + kfree(iars); + return ret; + } + + for (i = 0; i < iars->nr_ia_ranges; i++) { + iars->ia_range[i].queue = q; + ret = kobject_init_and_add(&iars->ia_range[i].kobj, + &blk_ia_range_ktype, &iars->kobj, + "%d", i); + if (ret) { + while (--i >= 0) + kobject_del(&iars->ia_range[i].kobj); + kobject_del(&iars->kobj); + kobject_put(&iars->kobj); + return ret; + } + } + + iars->sysfs_registered = true; + + return 0; +} + +void disk_unregister_independent_access_ranges(struct gendisk *disk) +{ + struct request_queue *q = disk->queue; + struct blk_independent_access_ranges *iars = q->ia_ranges; + int i; + + lockdep_assert_held(&q->sysfs_dir_lock); + lockdep_assert_held(&q->sysfs_lock); + + if (!iars) + return; + + if (iars->sysfs_registered) { + for (i = 0; i < iars->nr_ia_ranges; i++) + kobject_del(&iars->ia_range[i].kobj); + kobject_del(&iars->kobj); + kobject_put(&iars->kobj); + } else { + kfree(iars); + } + + q->ia_ranges = NULL; +} + +static struct blk_independent_access_range * +disk_find_ia_range(struct blk_independent_access_ranges *iars, + sector_t sector) +{ + struct blk_independent_access_range *iar; + int i; + + for (i = 0; i < iars->nr_ia_ranges; i++) { + iar = &iars->ia_range[i]; + if (sector >= iar->sector && + sector < iar->sector + iar->nr_sectors) + return iar; + } + + return NULL; +} + +static bool disk_check_ia_ranges(struct gendisk *disk, + struct blk_independent_access_ranges *iars) +{ + struct blk_independent_access_range *iar, *tmp; + sector_t capacity = get_capacity(disk); + sector_t sector = 0; + int i; + + /* + * While sorting the ranges in increasing LBA order, check that the + * ranges do not overlap, that there are no sector holes and that all + * sectors belong to one range. + */ + for (i = 0; i < iars->nr_ia_ranges; i++) { + tmp = disk_find_ia_range(iars, sector); + if (!tmp || tmp->sector != sector) { + pr_warn("Invalid non-contiguous independent access ranges\n"); + return false; + } + + iar = &iars->ia_range[i]; + if (tmp != iar) { + swap(iar->sector, tmp->sector); + swap(iar->nr_sectors, tmp->nr_sectors); + } + + sector += iar->nr_sectors; + } + + if (sector != capacity) { + pr_warn("Independent access ranges do not match disk capacity\n"); + return false; + } + + return true; +} + +static bool disk_ia_ranges_changed(struct gendisk *disk, + struct blk_independent_access_ranges *new) +{ + struct blk_independent_access_ranges *old = disk->queue->ia_ranges; + int i; + + if (!old) + return true; + + if (old->nr_ia_ranges != new->nr_ia_ranges) + return true; + + for (i = 0; i < old->nr_ia_ranges; i++) { + if (new->ia_range[i].sector != old->ia_range[i].sector || + new->ia_range[i].nr_sectors != old->ia_range[i].nr_sectors) + return true; + } + + return false; +} + +/** + * disk_alloc_independent_access_ranges - Allocate an independent access ranges + * data structure + * @disk: target disk + * @nr_ia_ranges: Number of independent access ranges + * + * Allocate a struct blk_independent_access_ranges structure with @nr_ia_ranges + * access range descriptors. + */ +struct blk_independent_access_ranges * +disk_alloc_independent_access_ranges(struct gendisk *disk, int nr_ia_ranges) +{ + struct blk_independent_access_ranges *iars; + + iars = kzalloc_node(struct_size(iars, ia_range, nr_ia_ranges), + GFP_KERNEL, disk->queue->node); + if (iars) + iars->nr_ia_ranges = nr_ia_ranges; + return iars; +} +EXPORT_SYMBOL_GPL(disk_alloc_independent_access_ranges); + +/** + * disk_set_independent_access_ranges - Set a disk independent access ranges + * @disk: target disk + * @iars: independent access ranges structure + * + * Set the independent access ranges information of the request queue + * of @disk to @iars. If @iars is NULL and the independent access ranges + * structure already set is cleared. If there are no differences between + * @iars and the independent access ranges structure already set, @iars + * is freed. + */ +void disk_set_independent_access_ranges(struct gendisk *disk, + struct blk_independent_access_ranges *iars) +{ + struct request_queue *q = disk->queue; + + if (WARN_ON_ONCE(iars && !iars->nr_ia_ranges)) { + kfree(iars); + iars = NULL; + } + + mutex_lock(&q->sysfs_dir_lock); + mutex_lock(&q->sysfs_lock); + + if (iars) { + if (!disk_check_ia_ranges(disk, iars)) { + kfree(iars); + iars = NULL; + goto reg; + } + + if (!disk_ia_ranges_changed(disk, iars)) { + kfree(iars); + goto unlock; + } + } + + /* + * This may be called for a registered queue. E.g. during a device + * revalidation. If that is the case, we need to unregister the old + * set of independent access ranges and register the new set. If the + * queue is not registered, registration of the device request queue + * will register the independent access ranges, so only swap in the + * new set and free the old one. + */ +reg: + if (blk_queue_registered(q)) { + disk_register_independent_access_ranges(disk, iars); + } else { + swap(q->ia_ranges, iars); + kfree(iars); + } + +unlock: + mutex_unlock(&q->sysfs_lock); + mutex_unlock(&q->sysfs_dir_lock); +} +EXPORT_SYMBOL_GPL(disk_set_independent_access_ranges); diff --git a/block/blk-integrity.c b/block/blk-integrity.c index 16d5d5338392..d670d54e5f7a 100644 --- a/block/blk-integrity.c +++ b/block/blk-integrity.c @@ -6,7 +6,7 @@ * Written by: Martin K. Petersen <martin.petersen@oracle.com> */ -#include <linux/blkdev.h> +#include <linux/blk-integrity.h> #include <linux/backing-dev.h> #include <linux/mempool.h> #include <linux/bio.h> @@ -409,9 +409,9 @@ void blk_integrity_register(struct gendisk *disk, struct blk_integrity *template blk_queue_flag_set(QUEUE_FLAG_STABLE_WRITES, disk->queue); #ifdef CONFIG_BLK_INLINE_ENCRYPTION - if (disk->queue->ksm) { + if (disk->queue->crypto_profile) { pr_warn("blk-integrity: Integrity and hardware inline encryption are not supported together. Disabling hardware inline encryption.\n"); - blk_ksm_unregister(disk->queue); + blk_crypto_unregister(disk->queue); } #endif } diff --git a/block/blk-iocost.c b/block/blk-iocost.c index b3880e4ba22a..a5b37cc65b17 100644 --- a/block/blk-iocost.c +++ b/block/blk-iocost.c @@ -3165,12 +3165,12 @@ static ssize_t ioc_qos_write(struct kernfs_open_file *of, char *input, if (IS_ERR(bdev)) return PTR_ERR(bdev); - ioc = q_to_ioc(bdev->bd_disk->queue); + ioc = q_to_ioc(bdev_get_queue(bdev)); if (!ioc) { - ret = blk_iocost_init(bdev->bd_disk->queue); + ret = blk_iocost_init(bdev_get_queue(bdev)); if (ret) goto err; - ioc = q_to_ioc(bdev->bd_disk->queue); + ioc = q_to_ioc(bdev_get_queue(bdev)); } spin_lock_irq(&ioc->lock); @@ -3332,12 +3332,12 @@ static ssize_t ioc_cost_model_write(struct kernfs_open_file *of, char *input, if (IS_ERR(bdev)) return PTR_ERR(bdev); - ioc = q_to_ioc(bdev->bd_disk->queue); + ioc = q_to_ioc(bdev_get_queue(bdev)); if (!ioc) { - ret = blk_iocost_init(bdev->bd_disk->queue); + ret = blk_iocost_init(bdev_get_queue(bdev)); if (ret) goto err; - ioc = q_to_ioc(bdev->bd_disk->queue); + ioc = q_to_ioc(bdev_get_queue(bdev)); } spin_lock_irq(&ioc->lock); diff --git a/block/blk-iolatency.c b/block/blk-iolatency.c index c0545f9da549..6593c7123b97 100644 --- a/block/blk-iolatency.c +++ b/block/blk-iolatency.c @@ -74,6 +74,7 @@ #include <linux/sched/signal.h> #include <trace/events/block.h> #include <linux/blk-mq.h> +#include <linux/blk-cgroup.h> #include "blk-rq-qos.h" #include "blk-stat.h" #include "blk.h" diff --git a/block/blk-merge.c b/block/blk-merge.c index 7a5c81c02c80..df69f4bb7717 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -6,12 +6,45 @@ #include <linux/module.h> #include <linux/bio.h> #include <linux/blkdev.h> +#include <linux/blk-integrity.h> #include <linux/scatterlist.h> #include <trace/events/block.h> #include "blk.h" #include "blk-rq-qos.h" +#include "blk-throttle.h" + +static inline void bio_get_first_bvec(struct bio *bio, struct bio_vec *bv) +{ + *bv = mp_bvec_iter_bvec(bio->bi_io_vec, bio->bi_iter); +} + +static inline void bio_get_last_bvec(struct bio *bio, struct bio_vec *bv) +{ + struct bvec_iter iter = bio->bi_iter; + int idx; + + bio_get_first_bvec(bio, bv); + if (bv->bv_len == bio->bi_iter.bi_size) + return; /* this bio only has a single bvec */ + + bio_advance_iter(bio, &iter, iter.bi_size); + + if (!iter.bi_bvec_done) + idx = iter.bi_idx - 1; + else /* in the middle of bvec */ + idx = iter.bi_idx; + + *bv = bio->bi_io_vec[idx]; + + /* + * iter.bi_bvec_done records actual length of the last bvec + * if this bio ends in the middle of one io vector + */ + if (iter.bi_bvec_done) + bv->bv_len = iter.bi_bvec_done; +} static inline bool bio_will_gap(struct request_queue *q, struct request *prev_rq, struct bio *prev, struct bio *next) @@ -285,13 +318,13 @@ split: * iopoll in direct IO routine. Given performance gain of iopoll for * big IO can be trival, disable iopoll when split needed. */ - bio_clear_hipri(bio); - + bio_clear_polled(bio); return bio_split(bio, sectors, GFP_NOIO, bs); } /** * __blk_queue_split - split a bio and submit the second half + * @q: [in] request_queue new bio is being queued at * @bio: [in, out] bio to be split * @nr_segs: [out] number of segments in the first bio * @@ -302,9 +335,9 @@ split: * of the caller to ensure that q->bio_split is only released after processing * of the split bio has finished. */ -void __blk_queue_split(struct bio **bio, unsigned int *nr_segs) +void __blk_queue_split(struct request_queue *q, struct bio **bio, + unsigned int *nr_segs) { - struct request_queue *q = (*bio)->bi_bdev->bd_disk->queue; struct bio *split = NULL; switch (bio_op(*bio)) { @@ -321,21 +354,6 @@ void __blk_queue_split(struct bio **bio, unsigned int *nr_segs) nr_segs); break; default: - /* - * All drivers must accept single-segments bios that are <= - * PAGE_SIZE. This is a quick and dirty check that relies on - * the fact that bi_io_vec[0] is always valid if a bio has data. - * The check might lead to occasional false negatives when bios - * are cloned, but compared to the performance impact of cloned - * bios themselves the loop below doesn't matter anyway. - */ - if (!q->limits.chunk_sectors && - (*bio)->bi_vcnt == 1 && - ((*bio)->bi_io_vec[0].bv_len + - (*bio)->bi_io_vec[0].bv_offset) <= PAGE_SIZE) { - *nr_segs = 1; - break; - } split = blk_bio_segment_split(q, *bio, &q->bio_split, nr_segs); break; } @@ -365,9 +383,11 @@ void __blk_queue_split(struct bio **bio, unsigned int *nr_segs) */ void blk_queue_split(struct bio **bio) { + struct request_queue *q = bdev_get_queue((*bio)->bi_bdev); unsigned int nr_segs; - __blk_queue_split(bio, &nr_segs); + if (blk_may_split(q, *bio)) + __blk_queue_split(q, bio, &nr_segs); } EXPORT_SYMBOL(blk_queue_split); @@ -558,6 +578,23 @@ static inline unsigned int blk_rq_get_max_segments(struct request *rq) return queue_max_segments(rq->q); } +static inline unsigned int blk_rq_get_max_sectors(struct request *rq, + sector_t offset) +{ + struct request_queue *q = rq->q; + + if (blk_rq_is_passthrough(rq)) + return q->limits.max_hw_sectors; + + if (!q->limits.chunk_sectors || + req_op(rq) == REQ_OP_DISCARD || + req_op(rq) == REQ_OP_SECURE_ERASE) + return blk_queue_get_max_sectors(q, req_op(rq)); + + return min(blk_max_size_offset(q, offset, 0), + blk_queue_get_max_sectors(q, req_op(rq))); +} + static inline int ll_new_hw_segment(struct request *req, struct bio *bio, unsigned int nr_phys_segs) { @@ -718,6 +755,13 @@ static enum elv_merge blk_try_req_merge(struct request *req, return ELEVATOR_NO_MERGE; } +static inline bool blk_write_same_mergeable(struct bio *a, struct bio *b) +{ + if (bio_page(a) == bio_page(b) && bio_offset(a) == bio_offset(b)) + return true; + return false; +} + /* * For non-mq, this has to be called with the request spinlock acquired. * For mq with scheduling, the appropriate queue wide lock should be held. @@ -1023,12 +1067,11 @@ static enum bio_merge_status blk_attempt_bio_merge(struct request_queue *q, * @q: request_queue new bio is being queued at * @bio: new bio being queued * @nr_segs: number of segments in @bio - * @same_queue_rq: pointer to &struct request that gets filled in when - * another request associated with @q is found on the plug list - * (optional, may be %NULL) + * @same_queue_rq: output value, will be true if there's an existing request + * from the passed in @q already in the plug list * - * Determine whether @bio being queued on @q can be merged with a request - * on %current's plugged list. Returns %true if merge was successful, + * Determine whether @bio being queued on @q can be merged with the previous + * request on %current's plugged list. Returns %true if merge was successful, * otherwise %false. * * Plugging coalesces IOs from the same issuer for the same purpose without @@ -1041,36 +1084,26 @@ static enum bio_merge_status blk_attempt_bio_merge(struct request_queue *q, * Caller must ensure !blk_queue_nomerges(q) beforehand. */ bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio, - unsigned int nr_segs, struct request **same_queue_rq) + unsigned int nr_segs, bool *same_queue_rq) { struct blk_plug *plug; struct request *rq; - struct list_head *plug_list; plug = blk_mq_plug(q, bio); - if (!plug) + if (!plug || rq_list_empty(plug->mq_list)) return false; - plug_list = &plug->mq_list; - - list_for_each_entry_reverse(rq, plug_list, queuelist) { - if (rq->q == q && same_queue_rq) { - /* - * Only blk-mq multiple hardware queues case checks the - * rq in the same queue, there should be only one such - * rq in a queue - **/ - *same_queue_rq = rq; - } - - if (rq->q != q) - continue; - - if (blk_attempt_bio_merge(q, rq, bio, nr_segs, false) == - BIO_MERGE_OK) - return true; + /* check the previously added entry for a quick merge attempt */ + rq = rq_list_peek(&plug->mq_list); + if (rq->q == q) { + /* + * Only blk-mq multiple hardware queues case checks the rq in + * the same queue, there should be only one such rq in a queue + */ + *same_queue_rq = true; } - + if (blk_attempt_bio_merge(q, rq, bio, nr_segs, false) == BIO_MERGE_OK) + return true; return false; } diff --git a/block/blk-mq-debugfs.c b/block/blk-mq-debugfs.c index 3b38d15723de..0f8c60e9c719 100644 --- a/block/blk-mq-debugfs.c +++ b/block/blk-mq-debugfs.c @@ -287,7 +287,7 @@ static const char *const cmd_flag_name[] = { CMD_FLAG_NAME(BACKGROUND), CMD_FLAG_NAME(NOWAIT), CMD_FLAG_NAME(NOUNMAP), - CMD_FLAG_NAME(HIPRI), + CMD_FLAG_NAME(POLLED), }; #undef CMD_FLAG_NAME @@ -453,11 +453,11 @@ static void blk_mq_debugfs_tags_show(struct seq_file *m, atomic_read(&tags->active_queues)); seq_puts(m, "\nbitmap_tags:\n"); - sbitmap_queue_show(tags->bitmap_tags, m); + sbitmap_queue_show(&tags->bitmap_tags, m); if (tags->nr_reserved_tags) { seq_puts(m, "\nbreserved_tags:\n"); - sbitmap_queue_show(tags->breserved_tags, m); + sbitmap_queue_show(&tags->breserved_tags, m); } } @@ -488,7 +488,7 @@ static int hctx_tags_bitmap_show(void *data, struct seq_file *m) if (res) goto out; if (hctx->tags) - sbitmap_bitmap_show(&hctx->tags->bitmap_tags->sb, m); + sbitmap_bitmap_show(&hctx->tags->bitmap_tags.sb, m); mutex_unlock(&q->sysfs_lock); out: @@ -522,77 +522,13 @@ static int hctx_sched_tags_bitmap_show(void *data, struct seq_file *m) if (res) goto out; if (hctx->sched_tags) - sbitmap_bitmap_show(&hctx->sched_tags->bitmap_tags->sb, m); + sbitmap_bitmap_show(&hctx->sched_tags->bitmap_tags.sb, m); mutex_unlock(&q->sysfs_lock); out: return res; } -static int hctx_io_poll_show(void *data, struct seq_file *m) -{ - struct blk_mq_hw_ctx *hctx = data; - - seq_printf(m, "considered=%lu\n", hctx->poll_considered); - seq_printf(m, "invoked=%lu\n", hctx->poll_invoked); - seq_printf(m, "success=%lu\n", hctx->poll_success); - return 0; -} - -static ssize_t hctx_io_poll_write(void *data, const char __user *buf, - size_t count, loff_t *ppos) -{ - struct blk_mq_hw_ctx *hctx = data; - - hctx->poll_considered = hctx->poll_invoked = hctx->poll_success = 0; - return count; -} - -static int hctx_dispatched_show(void *data, struct seq_file *m) -{ - struct blk_mq_hw_ctx *hctx = data; - int i; - - seq_printf(m, "%8u\t%lu\n", 0U, hctx->dispatched[0]); - - for (i = 1; i < BLK_MQ_MAX_DISPATCH_ORDER - 1; i++) { - unsigned int d = 1U << (i - 1); - - seq_printf(m, "%8u\t%lu\n", d, hctx->dispatched[i]); - } - - seq_printf(m, "%8u+\t%lu\n", 1U << (i - 1), hctx->dispatched[i]); - return 0; -} - -static ssize_t hctx_dispatched_write(void *data, const char __user *buf, - size_t count, loff_t *ppos) -{ - struct blk_mq_hw_ctx *hctx = data; - int i; - - for (i = 0; i < BLK_MQ_MAX_DISPATCH_ORDER; i++) - hctx->dispatched[i] = 0; - return count; -} - -static int hctx_queued_show(void *data, struct seq_file *m) -{ - struct blk_mq_hw_ctx *hctx = data; - - seq_printf(m, "%lu\n", hctx->queued); - return 0; -} - -static ssize_t hctx_queued_write(void *data, const char __user *buf, - size_t count, loff_t *ppos) -{ - struct blk_mq_hw_ctx *hctx = data; - - hctx->queued = 0; - return count; -} - static int hctx_run_show(void *data, struct seq_file *m) { struct blk_mq_hw_ctx *hctx = data; @@ -614,7 +550,7 @@ static int hctx_active_show(void *data, struct seq_file *m) { struct blk_mq_hw_ctx *hctx = data; - seq_printf(m, "%d\n", atomic_read(&hctx->nr_active)); + seq_printf(m, "%d\n", __blk_mq_active_requests(hctx)); return 0; } @@ -663,57 +599,6 @@ CTX_RQ_SEQ_OPS(default, HCTX_TYPE_DEFAULT); CTX_RQ_SEQ_OPS(read, HCTX_TYPE_READ); CTX_RQ_SEQ_OPS(poll, HCTX_TYPE_POLL); -static int ctx_dispatched_show(void *data, struct seq_file *m) -{ - struct blk_mq_ctx *ctx = data; - - seq_printf(m, "%lu %lu\n", ctx->rq_dispatched[1], ctx->rq_dispatched[0]); - return 0; -} - -static ssize_t ctx_dispatched_write(void *data, const char __user *buf, - size_t count, loff_t *ppos) -{ - struct blk_mq_ctx *ctx = data; - - ctx->rq_dispatched[0] = ctx->rq_dispatched[1] = 0; - return count; -} - -static int ctx_merged_show(void *data, struct seq_file *m) -{ - struct blk_mq_ctx *ctx = data; - - seq_printf(m, "%lu\n", ctx->rq_merged); - return 0; -} - -static ssize_t ctx_merged_write(void *data, const char __user *buf, - size_t count, loff_t *ppos) -{ - struct blk_mq_ctx *ctx = data; - - ctx->rq_merged = 0; - return count; -} - -static int ctx_completed_show(void *data, struct seq_file *m) -{ - struct blk_mq_ctx *ctx = data; - - seq_printf(m, "%lu %lu\n", ctx->rq_completed[1], ctx->rq_completed[0]); - return 0; -} - -static ssize_t ctx_completed_write(void *data, const char __user *buf, - size_t count, loff_t *ppos) -{ - struct blk_mq_ctx *ctx = data; - - ctx->rq_completed[0] = ctx->rq_completed[1] = 0; - return count; -} - static int blk_mq_debugfs_show(struct seq_file *m, void *v) { const struct blk_mq_debugfs_attr *attr = m->private; @@ -789,9 +674,6 @@ static const struct blk_mq_debugfs_attr blk_mq_debugfs_hctx_attrs[] = { {"tags_bitmap", 0400, hctx_tags_bitmap_show}, {"sched_tags", 0400, hctx_sched_tags_show}, {"sched_tags_bitmap", 0400, hctx_sched_tags_bitmap_show}, - {"io_poll", 0600, hctx_io_poll_show, hctx_io_poll_write}, - {"dispatched", 0600, hctx_dispatched_show, hctx_dispatched_write}, - {"queued", 0600, hctx_queued_show, hctx_queued_write}, {"run", 0600, hctx_run_show, hctx_run_write}, {"active", 0400, hctx_active_show}, {"dispatch_busy", 0400, hctx_dispatch_busy_show}, @@ -803,9 +685,6 @@ static const struct blk_mq_debugfs_attr blk_mq_debugfs_ctx_attrs[] = { {"default_rq_list", 0400, .seq_ops = &ctx_default_rq_list_seq_ops}, {"read_rq_list", 0400, .seq_ops = &ctx_read_rq_list_seq_ops}, {"poll_rq_list", 0400, .seq_ops = &ctx_poll_rq_list_seq_ops}, - {"dispatched", 0600, ctx_dispatched_show, ctx_dispatched_write}, - {"merged", 0600, ctx_merged_show, ctx_merged_write}, - {"completed", 0600, ctx_completed_show, ctx_completed_write}, {}, }; diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c index 0f006cabfd91..c62b966dfaba 100644 --- a/block/blk-mq-sched.c +++ b/block/blk-mq-sched.c @@ -57,10 +57,8 @@ void blk_mq_sched_mark_restart_hctx(struct blk_mq_hw_ctx *hctx) } EXPORT_SYMBOL_GPL(blk_mq_sched_mark_restart_hctx); -void blk_mq_sched_restart(struct blk_mq_hw_ctx *hctx) +void __blk_mq_sched_restart(struct blk_mq_hw_ctx *hctx) { - if (!test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) - return; clear_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state); /* @@ -363,7 +361,7 @@ void blk_mq_sched_dispatch_requests(struct blk_mq_hw_ctx *hctx) } } -bool __blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio, +bool blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio, unsigned int nr_segs) { struct elevator_queue *e = q->elevator; @@ -389,13 +387,10 @@ bool __blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio, * potentially merge with. Currently includes a hand-wavy stop * count of 8, to not spend too much time checking for merges. */ - if (blk_bio_list_merge(q, &ctx->rq_lists[type], bio, nr_segs)) { - ctx->rq_merged++; + if (blk_bio_list_merge(q, &ctx->rq_lists[type], bio, nr_segs)) ret = true; - } spin_unlock(&ctx->lock); - return ret; } @@ -515,83 +510,71 @@ void blk_mq_sched_insert_requests(struct blk_mq_hw_ctx *hctx, percpu_ref_put(&q->q_usage_counter); } -static int blk_mq_sched_alloc_tags(struct request_queue *q, - struct blk_mq_hw_ctx *hctx, - unsigned int hctx_idx) +static int blk_mq_sched_alloc_map_and_rqs(struct request_queue *q, + struct blk_mq_hw_ctx *hctx, + unsigned int hctx_idx) { - struct blk_mq_tag_set *set = q->tag_set; - int ret; + if (blk_mq_is_shared_tags(q->tag_set->flags)) { + hctx->sched_tags = q->sched_shared_tags; + return 0; + } + + hctx->sched_tags = blk_mq_alloc_map_and_rqs(q->tag_set, hctx_idx, + q->nr_requests); - hctx->sched_tags = blk_mq_alloc_rq_map(set, hctx_idx, q->nr_requests, - set->reserved_tags, set->flags); if (!hctx->sched_tags) return -ENOMEM; + return 0; +} - ret = blk_mq_alloc_rqs(set, hctx->sched_tags, hctx_idx, q->nr_requests); - if (ret) { - blk_mq_free_rq_map(hctx->sched_tags, set->flags); - hctx->sched_tags = NULL; - } - - return ret; +static void blk_mq_exit_sched_shared_tags(struct request_queue *queue) +{ + blk_mq_free_rq_map(queue->sched_shared_tags); + queue->sched_shared_tags = NULL; } /* called in queue's release handler, tagset has gone away */ -static void blk_mq_sched_tags_teardown(struct request_queue *q) +static void blk_mq_sched_tags_teardown(struct request_queue *q, unsigned int flags) { struct blk_mq_hw_ctx *hctx; int i; queue_for_each_hw_ctx(q, hctx, i) { if (hctx->sched_tags) { - blk_mq_free_rq_map(hctx->sched_tags, hctx->flags); + if (!blk_mq_is_shared_tags(flags)) + blk_mq_free_rq_map(hctx->sched_tags); hctx->sched_tags = NULL; } } + + if (blk_mq_is_shared_tags(flags)) + blk_mq_exit_sched_shared_tags(q); } -static int blk_mq_init_sched_shared_sbitmap(struct request_queue *queue) +static int blk_mq_init_sched_shared_tags(struct request_queue *queue) { struct blk_mq_tag_set *set = queue->tag_set; - int alloc_policy = BLK_MQ_FLAG_TO_ALLOC_POLICY(set->flags); - struct blk_mq_hw_ctx *hctx; - int ret, i; /* * Set initial depth at max so that we don't need to reallocate for * updating nr_requests. */ - ret = blk_mq_init_bitmaps(&queue->sched_bitmap_tags, - &queue->sched_breserved_tags, - MAX_SCHED_RQ, set->reserved_tags, - set->numa_node, alloc_policy); - if (ret) - return ret; - - queue_for_each_hw_ctx(queue, hctx, i) { - hctx->sched_tags->bitmap_tags = - &queue->sched_bitmap_tags; - hctx->sched_tags->breserved_tags = - &queue->sched_breserved_tags; - } + queue->sched_shared_tags = blk_mq_alloc_map_and_rqs(set, + BLK_MQ_NO_HCTX_IDX, + MAX_SCHED_RQ); + if (!queue->sched_shared_tags) + return -ENOMEM; - sbitmap_queue_resize(&queue->sched_bitmap_tags, - queue->nr_requests - set->reserved_tags); + blk_mq_tag_update_sched_shared_tags(queue); return 0; } -static void blk_mq_exit_sched_shared_sbitmap(struct request_queue *queue) -{ - sbitmap_queue_free(&queue->sched_bitmap_tags); - sbitmap_queue_free(&queue->sched_breserved_tags); -} - int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e) { + unsigned int i, flags = q->tag_set->flags; struct blk_mq_hw_ctx *hctx; struct elevator_queue *eq; - unsigned int i; int ret; if (!e) { @@ -606,23 +589,23 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e) * Additionally, this is a per-hw queue depth. */ q->nr_requests = 2 * min_t(unsigned int, q->tag_set->queue_depth, - BLKDEV_MAX_RQ); + BLKDEV_DEFAULT_RQ); - queue_for_each_hw_ctx(q, hctx, i) { - ret = blk_mq_sched_alloc_tags(q, hctx, i); + if (blk_mq_is_shared_tags(flags)) { + ret = blk_mq_init_sched_shared_tags(q); if (ret) - goto err_free_tags; + return ret; } - if (blk_mq_is_sbitmap_shared(q->tag_set->flags)) { - ret = blk_mq_init_sched_shared_sbitmap(q); + queue_for_each_hw_ctx(q, hctx, i) { + ret = blk_mq_sched_alloc_map_and_rqs(q, hctx, i); if (ret) - goto err_free_tags; + goto err_free_map_and_rqs; } ret = e->ops.init_sched(q, e); if (ret) - goto err_free_sbitmap; + goto err_free_map_and_rqs; blk_mq_debugfs_register_sched(q); @@ -631,7 +614,7 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e) ret = e->ops.init_hctx(hctx, i); if (ret) { eq = q->elevator; - blk_mq_sched_free_requests(q); + blk_mq_sched_free_rqs(q); blk_mq_exit_sched(q, eq); kobject_put(&eq->kobj); return ret; @@ -642,12 +625,10 @@ int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e) return 0; -err_free_sbitmap: - if (blk_mq_is_sbitmap_shared(q->tag_set->flags)) - blk_mq_exit_sched_shared_sbitmap(q); -err_free_tags: - blk_mq_sched_free_requests(q); - blk_mq_sched_tags_teardown(q); +err_free_map_and_rqs: + blk_mq_sched_free_rqs(q); + blk_mq_sched_tags_teardown(q, flags); + q->elevator = NULL; return ret; } @@ -656,14 +637,20 @@ err_free_tags: * called in either blk_queue_cleanup or elevator_switch, tagset * is required for freeing requests */ -void blk_mq_sched_free_requests(struct request_queue *q) +void blk_mq_sched_free_rqs(struct request_queue *q) { struct blk_mq_hw_ctx *hctx; int i; - queue_for_each_hw_ctx(q, hctx, i) { - if (hctx->sched_tags) - blk_mq_free_rqs(q->tag_set, hctx->sched_tags, i); + if (blk_mq_is_shared_tags(q->tag_set->flags)) { + blk_mq_free_rqs(q->tag_set, q->sched_shared_tags, + BLK_MQ_NO_HCTX_IDX); + } else { + queue_for_each_hw_ctx(q, hctx, i) { + if (hctx->sched_tags) + blk_mq_free_rqs(q->tag_set, + hctx->sched_tags, i); + } } } @@ -684,8 +671,6 @@ void blk_mq_exit_sched(struct request_queue *q, struct elevator_queue *e) blk_mq_debugfs_unregister_sched(q); if (e->type->ops.exit_sched) e->type->ops.exit_sched(e); - blk_mq_sched_tags_teardown(q); - if (blk_mq_is_sbitmap_shared(flags)) - blk_mq_exit_sched_shared_sbitmap(q); + blk_mq_sched_tags_teardown(q, flags); q->elevator = NULL; } diff --git a/block/blk-mq-sched.h b/block/blk-mq-sched.h index 5246ae040704..25d1034952b6 100644 --- a/block/blk-mq-sched.h +++ b/block/blk-mq-sched.h @@ -2,21 +2,22 @@ #ifndef BLK_MQ_SCHED_H #define BLK_MQ_SCHED_H +#include "elevator.h" #include "blk-mq.h" #include "blk-mq-tag.h" -#define MAX_SCHED_RQ (16 * BLKDEV_MAX_RQ) +#define MAX_SCHED_RQ (16 * BLKDEV_DEFAULT_RQ) void blk_mq_sched_assign_ioc(struct request *rq); bool blk_mq_sched_try_merge(struct request_queue *q, struct bio *bio, unsigned int nr_segs, struct request **merged_request); -bool __blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio, +bool blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio, unsigned int nr_segs); bool blk_mq_sched_try_insert_merge(struct request_queue *q, struct request *rq, struct list_head *free); void blk_mq_sched_mark_restart_hctx(struct blk_mq_hw_ctx *hctx); -void blk_mq_sched_restart(struct blk_mq_hw_ctx *hctx); +void __blk_mq_sched_restart(struct blk_mq_hw_ctx *hctx); void blk_mq_sched_insert_request(struct request *rq, bool at_head, bool run_queue, bool async); @@ -28,45 +29,51 @@ void blk_mq_sched_dispatch_requests(struct blk_mq_hw_ctx *hctx); int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e); void blk_mq_exit_sched(struct request_queue *q, struct elevator_queue *e); -void blk_mq_sched_free_requests(struct request_queue *q); +void blk_mq_sched_free_rqs(struct request_queue *q); -static inline bool -blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio, - unsigned int nr_segs) +static inline void blk_mq_sched_restart(struct blk_mq_hw_ctx *hctx) { - if (blk_queue_nomerges(q) || !bio_mergeable(bio)) - return false; + if (test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) + __blk_mq_sched_restart(hctx); +} - return __blk_mq_sched_bio_merge(q, bio, nr_segs); +static inline bool bio_mergeable(struct bio *bio) +{ + return !(bio->bi_opf & REQ_NOMERGE_FLAGS); } static inline bool blk_mq_sched_allow_merge(struct request_queue *q, struct request *rq, struct bio *bio) { - struct elevator_queue *e = q->elevator; - - if (e && e->type->ops.allow_merge) - return e->type->ops.allow_merge(q, rq, bio); + if (rq->rq_flags & RQF_ELV) { + struct elevator_queue *e = q->elevator; + if (e->type->ops.allow_merge) + return e->type->ops.allow_merge(q, rq, bio); + } return true; } static inline void blk_mq_sched_completed_request(struct request *rq, u64 now) { - struct elevator_queue *e = rq->q->elevator; + if (rq->rq_flags & RQF_ELV) { + struct elevator_queue *e = rq->q->elevator; - if (e && e->type->ops.completed_request) - e->type->ops.completed_request(rq, now); + if (e->type->ops.completed_request) + e->type->ops.completed_request(rq, now); + } } static inline void blk_mq_sched_requeue_request(struct request *rq) { - struct request_queue *q = rq->q; - struct elevator_queue *e = q->elevator; + if (rq->rq_flags & RQF_ELV) { + struct request_queue *q = rq->q; + struct elevator_queue *e = q->elevator; - if ((rq->rq_flags & RQF_ELVPRIV) && e && e->type->ops.requeue_request) - e->type->ops.requeue_request(rq); + if ((rq->rq_flags & RQF_ELVPRIV) && e->type->ops.requeue_request) + e->type->ops.requeue_request(rq); + } } static inline bool blk_mq_sched_has_work(struct blk_mq_hw_ctx *hctx) diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index ff5caeb82542..995336abee33 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -24,13 +24,12 @@ */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { - if (blk_mq_is_sbitmap_shared(hctx->flags)) { + if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - struct blk_mq_tag_set *set = q->tag_set; if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) && !test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) - atomic_inc(&set->active_queues_shared_sbitmap); + atomic_inc(&hctx->tags->active_queues); } else { if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) @@ -45,9 +44,9 @@ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) */ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) { - sbitmap_queue_wake_all(tags->bitmap_tags); + sbitmap_queue_wake_all(&tags->bitmap_tags); if (include_reserve) - sbitmap_queue_wake_all(tags->breserved_tags); + sbitmap_queue_wake_all(&tags->breserved_tags); } /* @@ -57,20 +56,20 @@ void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) { struct blk_mq_tags *tags = hctx->tags; - struct request_queue *q = hctx->queue; - struct blk_mq_tag_set *set = q->tag_set; - if (blk_mq_is_sbitmap_shared(hctx->flags)) { + if (blk_mq_is_shared_tags(hctx->flags)) { + struct request_queue *q = hctx->queue; + if (!test_and_clear_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) return; - atomic_dec(&set->active_queues_shared_sbitmap); } else { if (!test_and_clear_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) return; - atomic_dec(&tags->active_queues); } + atomic_dec(&tags->active_queues); + blk_mq_tag_wakeup_all(tags, false); } @@ -87,6 +86,21 @@ static int __blk_mq_get_tag(struct blk_mq_alloc_data *data, return __sbitmap_queue_get(bt); } +unsigned long blk_mq_get_tags(struct blk_mq_alloc_data *data, int nr_tags, + unsigned int *offset) +{ + struct blk_mq_tags *tags = blk_mq_tags_from_data(data); + struct sbitmap_queue *bt = &tags->bitmap_tags; + unsigned long ret; + + if (data->shallow_depth ||data->flags & BLK_MQ_REQ_RESERVED || + data->hctx->flags & BLK_MQ_F_TAG_QUEUE_SHARED) + return 0; + ret = __sbitmap_queue_get_batch(bt, nr_tags, offset); + *offset += tags->nr_reserved_tags; + return ret; +} + unsigned int blk_mq_get_tag(struct blk_mq_alloc_data *data) { struct blk_mq_tags *tags = blk_mq_tags_from_data(data); @@ -101,10 +115,10 @@ unsigned int blk_mq_get_tag(struct blk_mq_alloc_data *data) WARN_ON_ONCE(1); return BLK_MQ_NO_TAG; } - bt = tags->breserved_tags; + bt = &tags->breserved_tags; tag_offset = 0; } else { - bt = tags->bitmap_tags; + bt = &tags->bitmap_tags; tag_offset = tags->nr_reserved_tags; } @@ -150,9 +164,9 @@ unsigned int blk_mq_get_tag(struct blk_mq_alloc_data *data) data->ctx); tags = blk_mq_tags_from_data(data); if (data->flags & BLK_MQ_REQ_RESERVED) - bt = tags->breserved_tags; + bt = &tags->breserved_tags; else - bt = tags->bitmap_tags; + bt = &tags->bitmap_tags; /* * If destination hw queue is changed, fake wake up on @@ -186,13 +200,19 @@ void blk_mq_put_tag(struct blk_mq_tags *tags, struct blk_mq_ctx *ctx, const int real_tag = tag - tags->nr_reserved_tags; BUG_ON(real_tag >= tags->nr_tags); - sbitmap_queue_clear(tags->bitmap_tags, real_tag, ctx->cpu); + sbitmap_queue_clear(&tags->bitmap_tags, real_tag, ctx->cpu); } else { BUG_ON(tag >= tags->nr_reserved_tags); - sbitmap_queue_clear(tags->breserved_tags, tag, ctx->cpu); + sbitmap_queue_clear(&tags->breserved_tags, tag, ctx->cpu); } } +void blk_mq_put_tags(struct blk_mq_tags *tags, int *tag_array, int nr_tags) +{ + sbitmap_queue_clear_batch(&tags->bitmap_tags, tags->nr_reserved_tags, + tag_array, nr_tags); +} + struct bt_iter_data { struct blk_mq_hw_ctx *hctx; busy_iter_fn *fn; @@ -340,9 +360,9 @@ static void __blk_mq_all_tag_iter(struct blk_mq_tags *tags, WARN_ON_ONCE(flags & BT_TAG_ITER_RESERVED); if (tags->nr_reserved_tags) - bt_tags_for_each(tags, tags->breserved_tags, fn, priv, + bt_tags_for_each(tags, &tags->breserved_tags, fn, priv, flags | BT_TAG_ITER_RESERVED); - bt_tags_for_each(tags, tags->bitmap_tags, fn, priv, flags); + bt_tags_for_each(tags, &tags->bitmap_tags, fn, priv, flags); } /** @@ -379,9 +399,12 @@ void blk_mq_all_tag_iter(struct blk_mq_tags *tags, busy_tag_iter_fn *fn, void blk_mq_tagset_busy_iter(struct blk_mq_tag_set *tagset, busy_tag_iter_fn *fn, void *priv) { - int i; + unsigned int flags = tagset->flags; + int i, nr_tags; + + nr_tags = blk_mq_is_shared_tags(flags) ? 1 : tagset->nr_hw_queues; - for (i = 0; i < tagset->nr_hw_queues; i++) { + for (i = 0; i < nr_tags; i++) { if (tagset->tags && tagset->tags[i]) __blk_mq_all_tag_iter(tagset->tags[i], fn, priv, BT_TAG_ITER_STARTED); @@ -459,8 +482,8 @@ void blk_mq_queue_tag_busy_iter(struct request_queue *q, busy_iter_fn *fn, continue; if (tags->nr_reserved_tags) - bt_for_each(hctx, tags->breserved_tags, fn, priv, true); - bt_for_each(hctx, tags->bitmap_tags, fn, priv, false); + bt_for_each(hctx, &tags->breserved_tags, fn, priv, true); + bt_for_each(hctx, &tags->bitmap_tags, fn, priv, false); } blk_queue_exit(q); } @@ -492,56 +515,10 @@ free_bitmap_tags: return -ENOMEM; } -static int blk_mq_init_bitmap_tags(struct blk_mq_tags *tags, - int node, int alloc_policy) -{ - int ret; - - ret = blk_mq_init_bitmaps(&tags->__bitmap_tags, - &tags->__breserved_tags, - tags->nr_tags, tags->nr_reserved_tags, - node, alloc_policy); - if (ret) - return ret; - - tags->bitmap_tags = &tags->__bitmap_tags; - tags->breserved_tags = &tags->__breserved_tags; - - return 0; -} - -int blk_mq_init_shared_sbitmap(struct blk_mq_tag_set *set) -{ - int alloc_policy = BLK_MQ_FLAG_TO_ALLOC_POLICY(set->flags); - int i, ret; - - ret = blk_mq_init_bitmaps(&set->__bitmap_tags, &set->__breserved_tags, - set->queue_depth, set->reserved_tags, - set->numa_node, alloc_policy); - if (ret) - return ret; - - for (i = 0; i < set->nr_hw_queues; i++) { - struct blk_mq_tags *tags = set->tags[i]; - - tags->bitmap_tags = &set->__bitmap_tags; - tags->breserved_tags = &set->__breserved_tags; - } - - return 0; -} - -void blk_mq_exit_shared_sbitmap(struct blk_mq_tag_set *set) -{ - sbitmap_queue_free(&set->__bitmap_tags); - sbitmap_queue_free(&set->__breserved_tags); -} - struct blk_mq_tags *blk_mq_init_tags(unsigned int total_tags, unsigned int reserved_tags, - int node, unsigned int flags) + int node, int alloc_policy) { - int alloc_policy = BLK_MQ_FLAG_TO_ALLOC_POLICY(flags); struct blk_mq_tags *tags; if (total_tags > BLK_MQ_TAG_MAX) { @@ -557,22 +534,19 @@ struct blk_mq_tags *blk_mq_init_tags(unsigned int total_tags, tags->nr_reserved_tags = reserved_tags; spin_lock_init(&tags->lock); - if (blk_mq_is_sbitmap_shared(flags)) - return tags; - - if (blk_mq_init_bitmap_tags(tags, node, alloc_policy) < 0) { + if (blk_mq_init_bitmaps(&tags->bitmap_tags, &tags->breserved_tags, + total_tags, reserved_tags, node, + alloc_policy) < 0) { kfree(tags); return NULL; } return tags; } -void blk_mq_free_tags(struct blk_mq_tags *tags, unsigned int flags) +void blk_mq_free_tags(struct blk_mq_tags *tags) { - if (!blk_mq_is_sbitmap_shared(flags)) { - sbitmap_queue_free(tags->bitmap_tags); - sbitmap_queue_free(tags->breserved_tags); - } + sbitmap_queue_free(&tags->bitmap_tags); + sbitmap_queue_free(&tags->breserved_tags); kfree(tags); } @@ -592,7 +566,6 @@ int blk_mq_tag_update_depth(struct blk_mq_hw_ctx *hctx, if (tdepth > tags->nr_tags) { struct blk_mq_tag_set *set = hctx->queue->tag_set; struct blk_mq_tags *new; - bool ret; if (!can_grow) return -EINVAL; @@ -604,34 +577,42 @@ int blk_mq_tag_update_depth(struct blk_mq_hw_ctx *hctx, if (tdepth > MAX_SCHED_RQ) return -EINVAL; - new = blk_mq_alloc_rq_map(set, hctx->queue_num, tdepth, - tags->nr_reserved_tags, set->flags); + /* + * Only the sbitmap needs resizing since we allocated the max + * initially. + */ + if (blk_mq_is_shared_tags(set->flags)) + return 0; + + new = blk_mq_alloc_map_and_rqs(set, hctx->queue_num, tdepth); if (!new) return -ENOMEM; - ret = blk_mq_alloc_rqs(set, new, hctx->queue_num, tdepth); - if (ret) { - blk_mq_free_rq_map(new, set->flags); - return -ENOMEM; - } - blk_mq_free_rqs(set, *tagsptr, hctx->queue_num); - blk_mq_free_rq_map(*tagsptr, set->flags); + blk_mq_free_map_and_rqs(set, *tagsptr, hctx->queue_num); *tagsptr = new; } else { /* * Don't need (or can't) update reserved tags here, they * remain static and should never need resizing. */ - sbitmap_queue_resize(tags->bitmap_tags, + sbitmap_queue_resize(&tags->bitmap_tags, tdepth - tags->nr_reserved_tags); } return 0; } -void blk_mq_tag_resize_shared_sbitmap(struct blk_mq_tag_set *set, unsigned int size) +void blk_mq_tag_resize_shared_tags(struct blk_mq_tag_set *set, unsigned int size) +{ + struct blk_mq_tags *tags = set->shared_tags; + + sbitmap_queue_resize(&tags->bitmap_tags, size - set->reserved_tags); +} + +void blk_mq_tag_update_sched_shared_tags(struct request_queue *q) { - sbitmap_queue_resize(&set->__bitmap_tags, size - set->reserved_tags); + sbitmap_queue_resize(&q->sched_shared_tags->bitmap_tags, + q->nr_requests - q->tag_set->reserved_tags); } /** diff --git a/block/blk-mq-tag.h b/block/blk-mq-tag.h index 8ed55af08427..df787b5a23bd 100644 --- a/block/blk-mq-tag.h +++ b/block/blk-mq-tag.h @@ -2,52 +2,30 @@ #ifndef INT_BLK_MQ_TAG_H #define INT_BLK_MQ_TAG_H -/* - * Tag address space map. - */ -struct blk_mq_tags { - unsigned int nr_tags; - unsigned int nr_reserved_tags; - - atomic_t active_queues; - - struct sbitmap_queue *bitmap_tags; - struct sbitmap_queue *breserved_tags; - - struct sbitmap_queue __bitmap_tags; - struct sbitmap_queue __breserved_tags; - - struct request **rqs; - struct request **static_rqs; - struct list_head page_list; - - /* - * used to clear request reference in rqs[] before freeing one - * request pool - */ - spinlock_t lock; -}; +struct blk_mq_alloc_data; extern struct blk_mq_tags *blk_mq_init_tags(unsigned int nr_tags, unsigned int reserved_tags, - int node, unsigned int flags); -extern void blk_mq_free_tags(struct blk_mq_tags *tags, unsigned int flags); + int node, int alloc_policy); +extern void blk_mq_free_tags(struct blk_mq_tags *tags); extern int blk_mq_init_bitmaps(struct sbitmap_queue *bitmap_tags, struct sbitmap_queue *breserved_tags, unsigned int queue_depth, unsigned int reserved, int node, int alloc_policy); -extern int blk_mq_init_shared_sbitmap(struct blk_mq_tag_set *set); -extern void blk_mq_exit_shared_sbitmap(struct blk_mq_tag_set *set); extern unsigned int blk_mq_get_tag(struct blk_mq_alloc_data *data); +unsigned long blk_mq_get_tags(struct blk_mq_alloc_data *data, int nr_tags, + unsigned int *offset); extern void blk_mq_put_tag(struct blk_mq_tags *tags, struct blk_mq_ctx *ctx, unsigned int tag); +void blk_mq_put_tags(struct blk_mq_tags *tags, int *tag_array, int nr_tags); extern int blk_mq_tag_update_depth(struct blk_mq_hw_ctx *hctx, struct blk_mq_tags **tags, unsigned int depth, bool can_grow); -extern void blk_mq_tag_resize_shared_sbitmap(struct blk_mq_tag_set *set, +extern void blk_mq_tag_resize_shared_tags(struct blk_mq_tag_set *set, unsigned int size); +extern void blk_mq_tag_update_sched_shared_tags(struct request_queue *q); extern void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool); void blk_mq_queue_tag_busy_iter(struct request_queue *q, busy_iter_fn *fn, diff --git a/block/blk-mq.c b/block/blk-mq.c index 652a31fc3bb3..07eb1412760b 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -10,14 +10,15 @@ #include <linux/backing-dev.h> #include <linux/bio.h> #include <linux/blkdev.h> +#include <linux/blk-integrity.h> #include <linux/kmemleak.h> #include <linux/mm.h> #include <linux/init.h> #include <linux/slab.h> #include <linux/workqueue.h> #include <linux/smp.h> +#include <linux/interrupt.h> #include <linux/llist.h> -#include <linux/list_sort.h> #include <linux/cpu.h> #include <linux/cache.h> #include <linux/sched/sysctl.h> @@ -63,6 +64,32 @@ static int blk_mq_poll_stats_bkt(const struct request *rq) return bucket; } +#define BLK_QC_T_SHIFT 16 +#define BLK_QC_T_INTERNAL (1U << 31) + +static inline struct blk_mq_hw_ctx *blk_qc_to_hctx(struct request_queue *q, + blk_qc_t qc) +{ + return q->queue_hw_ctx[(qc & ~BLK_QC_T_INTERNAL) >> BLK_QC_T_SHIFT]; +} + +static inline struct request *blk_qc_to_rq(struct blk_mq_hw_ctx *hctx, + blk_qc_t qc) +{ + unsigned int tag = qc & ((1U << BLK_QC_T_SHIFT) - 1); + + if (qc & BLK_QC_T_INTERNAL) + return blk_mq_tag_to_rq(hctx->sched_tags, tag); + return blk_mq_tag_to_rq(hctx->tags, tag); +} + +static inline blk_qc_t blk_rq_to_qc(struct request *rq) +{ + return (rq->mq_hctx->queue_num << BLK_QC_T_SHIFT) | + (rq->tag != -1 ? + rq->tag : (rq->internal_tag | BLK_QC_T_INTERNAL)); +} + /* * Check if any of the ctx, dispatch list or elevator * have pending work in this hardware queue. @@ -214,7 +241,12 @@ EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue); */ void blk_mq_quiesce_queue_nowait(struct request_queue *q) { - blk_queue_flag_set(QUEUE_FLAG_QUIESCED, q); + unsigned long flags; + + spin_lock_irqsave(&q->queue_lock, flags); + if (!q->quiesce_depth++) + blk_queue_flag_set(QUEUE_FLAG_QUIESCED, q); + spin_unlock_irqrestore(&q->queue_lock, flags); } EXPORT_SYMBOL_GPL(blk_mq_quiesce_queue_nowait); @@ -255,10 +287,21 @@ EXPORT_SYMBOL_GPL(blk_mq_quiesce_queue); */ void blk_mq_unquiesce_queue(struct request_queue *q) { - blk_queue_flag_clear(QUEUE_FLAG_QUIESCED, q); + unsigned long flags; + bool run_queue = false; + + spin_lock_irqsave(&q->queue_lock, flags); + if (WARN_ON_ONCE(q->quiesce_depth <= 0)) { + ; + } else if (!--q->quiesce_depth) { + blk_queue_flag_clear(QUEUE_FLAG_QUIESCED, q); + run_queue = true; + } + spin_unlock_irqrestore(&q->queue_lock, flags); /* dispatch requests which are inserted during quiescing */ - blk_mq_run_hw_queues(q, true); + if (run_queue) + blk_mq_run_hw_queues(q, true); } EXPORT_SYMBOL_GPL(blk_mq_unquiesce_queue); @@ -272,74 +315,67 @@ void blk_mq_wake_waiters(struct request_queue *q) blk_mq_tag_wakeup_all(hctx->tags, true); } -/* - * Only need start/end time stamping if we have iostat or - * blk stats enabled, or using an IO scheduler. - */ -static inline bool blk_mq_need_time_stamp(struct request *rq) -{ - return (rq->rq_flags & (RQF_IO_STAT | RQF_STATS)) || rq->q->elevator; -} - static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data, - unsigned int tag, u64 alloc_time_ns) + struct blk_mq_tags *tags, unsigned int tag, u64 alloc_time_ns) { - struct blk_mq_tags *tags = blk_mq_tags_from_data(data); + struct blk_mq_ctx *ctx = data->ctx; + struct blk_mq_hw_ctx *hctx = data->hctx; + struct request_queue *q = data->q; struct request *rq = tags->static_rqs[tag]; - if (data->q->elevator) { - rq->tag = BLK_MQ_NO_TAG; - rq->internal_tag = tag; - } else { + rq->q = q; + rq->mq_ctx = ctx; + rq->mq_hctx = hctx; + rq->cmd_flags = data->cmd_flags; + + if (data->flags & BLK_MQ_REQ_PM) + data->rq_flags |= RQF_PM; + if (blk_queue_io_stat(q)) + data->rq_flags |= RQF_IO_STAT; + rq->rq_flags = data->rq_flags; + + if (!(data->rq_flags & RQF_ELV)) { rq->tag = tag; rq->internal_tag = BLK_MQ_NO_TAG; + } else { + rq->tag = BLK_MQ_NO_TAG; + rq->internal_tag = tag; } + rq->timeout = 0; - /* csd/requeue_work/fifo_time is initialized before use */ - rq->q = data->q; - rq->mq_ctx = data->ctx; - rq->mq_hctx = data->hctx; - rq->rq_flags = 0; - rq->cmd_flags = data->cmd_flags; - if (data->flags & BLK_MQ_REQ_PM) - rq->rq_flags |= RQF_PM; - if (blk_queue_io_stat(data->q)) - rq->rq_flags |= RQF_IO_STAT; - INIT_LIST_HEAD(&rq->queuelist); - INIT_HLIST_NODE(&rq->hash); - RB_CLEAR_NODE(&rq->rb_node); + if (blk_mq_need_time_stamp(rq)) + rq->start_time_ns = ktime_get_ns(); + else + rq->start_time_ns = 0; rq->rq_disk = NULL; rq->part = NULL; #ifdef CONFIG_BLK_RQ_ALLOC_TIME rq->alloc_time_ns = alloc_time_ns; #endif - if (blk_mq_need_time_stamp(rq)) - rq->start_time_ns = ktime_get_ns(); - else - rq->start_time_ns = 0; rq->io_start_time_ns = 0; rq->stats_sectors = 0; rq->nr_phys_segments = 0; #if defined(CONFIG_BLK_DEV_INTEGRITY) rq->nr_integrity_segments = 0; #endif - blk_crypto_rq_set_defaults(rq); - /* tag was already set */ - WRITE_ONCE(rq->deadline, 0); - - rq->timeout = 0; - rq->end_io = NULL; rq->end_io_data = NULL; - data->ctx->rq_dispatched[op_is_sync(data->cmd_flags)]++; + blk_crypto_rq_set_defaults(rq); + INIT_LIST_HEAD(&rq->queuelist); + /* tag was already set */ + WRITE_ONCE(rq->deadline, 0); refcount_set(&rq->ref, 1); - if (!op_is_flush(data->cmd_flags)) { + if (rq->rq_flags & RQF_ELV) { struct elevator_queue *e = data->q->elevator; rq->elv.icq = NULL; - if (e && e->type->ops.prepare_request) { + INIT_HLIST_NODE(&rq->hash); + RB_CLEAR_NODE(&rq->rb_node); + + if (!op_is_flush(data->cmd_flags) && + e->type->ops.prepare_request) { if (e->type->icq_cache) blk_mq_sched_assign_ioc(rq); @@ -348,15 +384,44 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data, } } - data->hctx->queued++; return rq; } -static struct request *__blk_mq_alloc_request(struct blk_mq_alloc_data *data) +static inline struct request * +__blk_mq_alloc_requests_batch(struct blk_mq_alloc_data *data, + u64 alloc_time_ns) +{ + unsigned int tag, tag_offset; + struct blk_mq_tags *tags; + struct request *rq; + unsigned long tag_mask; + int i, nr = 0; + + tag_mask = blk_mq_get_tags(data, data->nr_tags, &tag_offset); + if (unlikely(!tag_mask)) + return NULL; + + tags = blk_mq_tags_from_data(data); + for (i = 0; tag_mask; i++) { + if (!(tag_mask & (1UL << i))) + continue; + prefetch(tags->static_rqs[tag]); + tag = tag_offset + i; + tag_mask &= ~(1UL << i); + rq = blk_mq_rq_ctx_init(data, tags, tag, alloc_time_ns); + rq_list_add(data->cached_rq, rq); + } + data->nr_tags -= nr; + + return rq_list_pop(data->cached_rq); +} + +static struct request *__blk_mq_alloc_requests(struct blk_mq_alloc_data *data) { struct request_queue *q = data->q; struct elevator_queue *e = q->elevator; u64 alloc_time_ns = 0; + struct request *rq; unsigned int tag; /* alloc_time includes depth and tag waits */ @@ -386,6 +451,16 @@ retry: blk_mq_tag_busy(data->hctx); /* + * Try batched alloc if we want more than 1 tag. + */ + if (data->nr_tags > 1) { + rq = __blk_mq_alloc_requests_batch(data, alloc_time_ns); + if (rq) + return rq; + data->nr_tags = 1; + } + + /* * Waiting allocations only fail because of an inactive hctx. In that * case just retry the hctx assignment and tag allocation as CPU hotplug * should have migrated us to an online CPU by now. @@ -394,16 +469,18 @@ retry: if (tag == BLK_MQ_NO_TAG) { if (data->flags & BLK_MQ_REQ_NOWAIT) return NULL; - /* - * Give up the CPU and sleep for a random short time to ensure - * that thread using a realtime scheduling class are migrated - * off the CPU, and thus off the hctx that is going away. + * Give up the CPU and sleep for a random short time to + * ensure that thread using a realtime scheduling class + * are migrated off the CPU, and thus off the hctx that + * is going away. */ msleep(3); goto retry; } - return blk_mq_rq_ctx_init(data, tag, alloc_time_ns); + + return blk_mq_rq_ctx_init(data, blk_mq_tags_from_data(data), tag, + alloc_time_ns); } struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op, @@ -413,6 +490,8 @@ struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op, .q = q, .flags = flags, .cmd_flags = op, + .rq_flags = q->elevator ? RQF_ELV : 0, + .nr_tags = 1, }; struct request *rq; int ret; @@ -421,7 +500,7 @@ struct request *blk_mq_alloc_request(struct request_queue *q, unsigned int op, if (ret) return ERR_PTR(ret); - rq = __blk_mq_alloc_request(&data); + rq = __blk_mq_alloc_requests(&data); if (!rq) goto out_queue_exit; rq->__data_len = 0; @@ -441,6 +520,8 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q, .q = q, .flags = flags, .cmd_flags = op, + .rq_flags = q->elevator ? RQF_ELV : 0, + .nr_tags = 1, }; u64 alloc_time_ns = 0; unsigned int cpu; @@ -485,7 +566,8 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q, tag = blk_mq_get_tag(&data); if (tag == BLK_MQ_NO_TAG) goto out_queue_exit; - return blk_mq_rq_ctx_init(&data, tag, alloc_time_ns); + return blk_mq_rq_ctx_init(&data, blk_mq_tags_from_data(&data), tag, + alloc_time_ns); out_queue_exit: blk_queue_exit(q); @@ -514,12 +596,12 @@ static void __blk_mq_free_request(struct request *rq) void blk_mq_free_request(struct request *rq) { struct request_queue *q = rq->q; - struct elevator_queue *e = q->elevator; - struct blk_mq_ctx *ctx = rq->mq_ctx; struct blk_mq_hw_ctx *hctx = rq->mq_hctx; if (rq->rq_flags & RQF_ELVPRIV) { - if (e && e->type->ops.finish_request) + struct elevator_queue *e = q->elevator; + + if (e->type->ops.finish_request) e->type->ops.finish_request(rq); if (rq->elv.icq) { put_io_context(rq->elv.icq->ioc); @@ -527,7 +609,6 @@ void blk_mq_free_request(struct request *rq) } } - ctx->rq_completed[rq_is_sync(rq)]++; if (rq->rq_flags & RQF_MQ_INFLIGHT) __blk_mq_dec_active_requests(hctx); @@ -542,21 +623,173 @@ void blk_mq_free_request(struct request *rq) } EXPORT_SYMBOL_GPL(blk_mq_free_request); -inline void __blk_mq_end_request(struct request *rq, blk_status_t error) +void blk_mq_free_plug_rqs(struct blk_plug *plug) { - u64 now = 0; + struct request *rq; - if (blk_mq_need_time_stamp(rq)) - now = ktime_get_ns(); + while ((rq = rq_list_pop(&plug->cached_rq)) != NULL) { + percpu_ref_get(&rq->q->q_usage_counter); + blk_mq_free_request(rq); + } +} +static void req_bio_endio(struct request *rq, struct bio *bio, + unsigned int nbytes, blk_status_t error) +{ + if (unlikely(error)) { + bio->bi_status = error; + } else if (req_op(rq) == REQ_OP_ZONE_APPEND) { + /* + * Partial zone append completions cannot be supported as the + * BIO fragments may end up not being written sequentially. + */ + if (bio->bi_iter.bi_size != nbytes) + bio->bi_status = BLK_STS_IOERR; + else + bio->bi_iter.bi_sector = rq->__sector; + } + + bio_advance(bio, nbytes); + + if (unlikely(rq->rq_flags & RQF_QUIET)) + bio_set_flag(bio, BIO_QUIET); + /* don't actually finish bio if it's part of flush sequence */ + if (bio->bi_iter.bi_size == 0 && !(rq->rq_flags & RQF_FLUSH_SEQ)) + bio_endio(bio); +} + +static void blk_account_io_completion(struct request *req, unsigned int bytes) +{ + if (req->part && blk_do_io_stat(req)) { + const int sgrp = op_stat_group(req_op(req)); + + part_stat_lock(); + part_stat_add(req->part, sectors[sgrp], bytes >> 9); + part_stat_unlock(); + } +} + +/** + * blk_update_request - Complete multiple bytes without completing the request + * @req: the request being processed + * @error: block status code + * @nr_bytes: number of bytes to complete for @req + * + * Description: + * Ends I/O on a number of bytes attached to @req, but doesn't complete + * the request structure even if @req doesn't have leftover. + * If @req has leftover, sets it up for the next range of segments. + * + * Passing the result of blk_rq_bytes() as @nr_bytes guarantees + * %false return from this function. + * + * Note: + * The RQF_SPECIAL_PAYLOAD flag is ignored on purpose in this function + * except in the consistency check at the end of this function. + * + * Return: + * %false - this request doesn't have any more data + * %true - this request has more data + **/ +bool blk_update_request(struct request *req, blk_status_t error, + unsigned int nr_bytes) +{ + int total_bytes; + + trace_block_rq_complete(req, error, nr_bytes); + + if (!req->bio) + return false; + +#ifdef CONFIG_BLK_DEV_INTEGRITY + if (blk_integrity_rq(req) && req_op(req) == REQ_OP_READ && + error == BLK_STS_OK) + req->q->integrity.profile->complete_fn(req, nr_bytes); +#endif + + if (unlikely(error && !blk_rq_is_passthrough(req) && + !(req->rq_flags & RQF_QUIET))) + blk_print_req_error(req, error); + + blk_account_io_completion(req, nr_bytes); + + total_bytes = 0; + while (req->bio) { + struct bio *bio = req->bio; + unsigned bio_bytes = min(bio->bi_iter.bi_size, nr_bytes); + + if (bio_bytes == bio->bi_iter.bi_size) + req->bio = bio->bi_next; + + /* Completion has already been traced */ + bio_clear_flag(bio, BIO_TRACE_COMPLETION); + req_bio_endio(req, bio, bio_bytes, error); + + total_bytes += bio_bytes; + nr_bytes -= bio_bytes; + + if (!nr_bytes) + break; + } + + /* + * completely done + */ + if (!req->bio) { + /* + * Reset counters so that the request stacking driver + * can find how many bytes remain in the request + * later. + */ + req->__data_len = 0; + return false; + } + + req->__data_len -= total_bytes; + + /* update sector only for requests with clear definition of sector */ + if (!blk_rq_is_passthrough(req)) + req->__sector += total_bytes >> 9; + + /* mixed attributes always follow the first bio */ + if (req->rq_flags & RQF_MIXED_MERGE) { + req->cmd_flags &= ~REQ_FAILFAST_MASK; + req->cmd_flags |= req->bio->bi_opf & REQ_FAILFAST_MASK; + } + + if (!(req->rq_flags & RQF_SPECIAL_PAYLOAD)) { + /* + * If total number of sectors is less than the first segment + * size, something has gone terribly wrong. + */ + if (blk_rq_bytes(req) < blk_rq_cur_bytes(req)) { + blk_dump_rq_flags(req, "request botched"); + req->__data_len = blk_rq_cur_bytes(req); + } + + /* recalculate the number of segments */ + req->nr_phys_segments = blk_recalc_rq_segments(req); + } + + return true; +} +EXPORT_SYMBOL_GPL(blk_update_request); + +static inline void __blk_mq_end_request_acct(struct request *rq, u64 now) +{ if (rq->rq_flags & RQF_STATS) { blk_mq_poll_stats_start(rq->q); blk_stat_add(rq, now); } blk_mq_sched_completed_request(rq, now); - blk_account_io_done(rq, now); +} + +inline void __blk_mq_end_request(struct request *rq, blk_status_t error) +{ + if (blk_mq_need_time_stamp(rq)) + __blk_mq_end_request_acct(rq, ktime_get_ns()); if (rq->end_io) { rq_qos_done(rq->q, rq); @@ -575,6 +808,57 @@ void blk_mq_end_request(struct request *rq, blk_status_t error) } EXPORT_SYMBOL(blk_mq_end_request); +#define TAG_COMP_BATCH 32 + +static inline void blk_mq_flush_tag_batch(struct blk_mq_hw_ctx *hctx, + int *tag_array, int nr_tags) +{ + struct request_queue *q = hctx->queue; + + blk_mq_put_tags(hctx->tags, tag_array, nr_tags); + percpu_ref_put_many(&q->q_usage_counter, nr_tags); +} + +void blk_mq_end_request_batch(struct io_comp_batch *iob) +{ + int tags[TAG_COMP_BATCH], nr_tags = 0; + struct blk_mq_hw_ctx *cur_hctx = NULL; + struct request *rq; + u64 now = 0; + + if (iob->need_ts) + now = ktime_get_ns(); + + while ((rq = rq_list_pop(&iob->req_list)) != NULL) { + prefetch(rq->bio); + prefetch(rq->rq_next); + + blk_update_request(rq, BLK_STS_OK, blk_rq_bytes(rq)); + if (iob->need_ts) + __blk_mq_end_request_acct(rq, now); + + WRITE_ONCE(rq->state, MQ_RQ_IDLE); + if (!refcount_dec_and_test(&rq->ref)) + continue; + + blk_crypto_free_request(rq); + blk_pm_mark_last_busy(rq); + rq_qos_done(rq->q, rq); + + if (nr_tags == TAG_COMP_BATCH || cur_hctx != rq->mq_hctx) { + if (cur_hctx) + blk_mq_flush_tag_batch(cur_hctx, tags, nr_tags); + nr_tags = 0; + cur_hctx = rq->mq_hctx; + } + tags[nr_tags++] = rq->tag; + } + + if (nr_tags) + blk_mq_flush_tag_batch(cur_hctx, tags, nr_tags); +} +EXPORT_SYMBOL_GPL(blk_mq_end_request_batch); + static void blk_complete_reqs(struct llist_head *list) { struct llist_node *entry = llist_reverse_order(llist_del_all(list)); @@ -658,7 +942,7 @@ bool blk_mq_complete_request_remote(struct request *rq) * For a polled request, always complete locallly, it's pointless * to redirect the completion. */ - if (rq->cmd_flags & REQ_HIPRI) + if (rq->cmd_flags & REQ_POLLED) return false; if (blk_mq_complete_need_ipi(rq)) { @@ -723,7 +1007,14 @@ void blk_mq_start_request(struct request *rq) trace_block_rq_issue(rq); if (test_bit(QUEUE_FLAG_STATS, &q->queue_flags)) { - rq->io_start_time_ns = ktime_get_ns(); + u64 start_time; +#ifdef CONFIG_BLK_CGROUP + if (rq->bio) + start_time = bio_issue_time(&rq->bio->bi_issue); + else +#endif + start_time = ktime_get_ns(); + rq->io_start_time_ns = start_time; rq->stats_sectors = blk_rq_sectors(rq); rq->rq_flags |= RQF_STATS; rq_qos_issue(q, rq); @@ -738,6 +1029,8 @@ void blk_mq_start_request(struct request *rq) if (blk_integrity_rq(rq) && req_op(rq) == REQ_OP_WRITE) q->integrity.profile->prepare_fn(rq); #endif + if (rq->bio && rq->bio->bi_opf & REQ_POLLED) + WRITE_ONCE(rq->bio->bi_cookie, blk_rq_to_qc(rq)); } EXPORT_SYMBOL(blk_mq_start_request); @@ -763,7 +1056,6 @@ void blk_mq_requeue_request(struct request *rq, bool kick_requeue_list) /* this request will be re-inserted to io scheduler queue */ blk_mq_sched_requeue_request(rq); - BUG_ON(!list_empty(&rq->queuelist)); blk_mq_add_to_requeue_list(rq, true, kick_requeue_list); } EXPORT_SYMBOL(blk_mq_requeue_request); @@ -844,17 +1136,6 @@ void blk_mq_delay_kick_requeue_list(struct request_queue *q, } EXPORT_SYMBOL(blk_mq_delay_kick_requeue_list); -struct request *blk_mq_tag_to_rq(struct blk_mq_tags *tags, unsigned int tag) -{ - if (tag < tags->nr_tags) { - prefetch(tags->rqs[tag]); - return tags->rqs[tag]; - } - - return NULL; -} -EXPORT_SYMBOL(blk_mq_tag_to_rq); - static bool blk_mq_rq_inflight(struct blk_mq_hw_ctx *hctx, struct request *rq, void *priv, bool reserved) { @@ -1059,24 +1340,16 @@ struct request *blk_mq_dequeue_from_ctx(struct blk_mq_hw_ctx *hctx, return data.rq; } -static inline unsigned int queued_to_index(unsigned int queued) -{ - if (!queued) - return 0; - - return min(BLK_MQ_MAX_DISPATCH_ORDER - 1, ilog2(queued) + 1); -} - -static bool __blk_mq_get_driver_tag(struct request *rq) +static bool __blk_mq_alloc_driver_tag(struct request *rq) { - struct sbitmap_queue *bt = rq->mq_hctx->tags->bitmap_tags; + struct sbitmap_queue *bt = &rq->mq_hctx->tags->bitmap_tags; unsigned int tag_offset = rq->mq_hctx->tags->nr_reserved_tags; int tag; blk_mq_tag_busy(rq->mq_hctx); if (blk_mq_tag_is_reserved(rq->mq_hctx->sched_tags, rq->internal_tag)) { - bt = rq->mq_hctx->tags->breserved_tags; + bt = &rq->mq_hctx->tags->breserved_tags; tag_offset = 0; } else { if (!hctx_may_queue(rq->mq_hctx, bt)) @@ -1091,11 +1364,9 @@ static bool __blk_mq_get_driver_tag(struct request *rq) return true; } -bool blk_mq_get_driver_tag(struct request *rq) +bool __blk_mq_get_driver_tag(struct blk_mq_hw_ctx *hctx, struct request *rq) { - struct blk_mq_hw_ctx *hctx = rq->mq_hctx; - - if (rq->tag == BLK_MQ_NO_TAG && !__blk_mq_get_driver_tag(rq)) + if (rq->tag == BLK_MQ_NO_TAG && !__blk_mq_alloc_driver_tag(rq)) return false; if ((hctx->flags & BLK_MQ_F_TAG_QUEUE_SHARED) && @@ -1119,7 +1390,7 @@ static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode, struct sbitmap_queue *sbq; list_del_init(&wait->entry); - sbq = hctx->tags->bitmap_tags; + sbq = &hctx->tags->bitmap_tags; atomic_dec(&sbq->ws_active); } spin_unlock(&hctx->dispatch_wait_lock); @@ -1137,7 +1408,7 @@ static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode, static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx *hctx, struct request *rq) { - struct sbitmap_queue *sbq = hctx->tags->bitmap_tags; + struct sbitmap_queue *sbq = &hctx->tags->bitmap_tags; struct wait_queue_head *wq; wait_queue_entry_t *wait; bool ret; @@ -1394,8 +1665,6 @@ out: if (!list_empty(&zone_list)) list_splice_tail_init(&zone_list, list); - hctx->dispatched[queued_to_index(queued)]++; - /* If we didn't flush the entire list, we could have told the driver * there was more coming, but that turned out to be a lie. */ @@ -1899,54 +2168,106 @@ void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx, spin_unlock(&ctx->lock); } -static int plug_rq_cmp(void *priv, const struct list_head *a, - const struct list_head *b) +static void blk_mq_commit_rqs(struct blk_mq_hw_ctx *hctx, int *queued, + bool from_schedule) { - struct request *rqa = container_of(a, struct request, queuelist); - struct request *rqb = container_of(b, struct request, queuelist); + if (hctx->queue->mq_ops->commit_rqs) { + trace_block_unplug(hctx->queue, *queued, !from_schedule); + hctx->queue->mq_ops->commit_rqs(hctx); + } + *queued = 0; +} - if (rqa->mq_ctx != rqb->mq_ctx) - return rqa->mq_ctx > rqb->mq_ctx; - if (rqa->mq_hctx != rqb->mq_hctx) - return rqa->mq_hctx > rqb->mq_hctx; +static void blk_mq_plug_issue_direct(struct blk_plug *plug, bool from_schedule) +{ + struct blk_mq_hw_ctx *hctx = NULL; + struct request *rq; + int queued = 0; + int errors = 0; - return blk_rq_pos(rqa) > blk_rq_pos(rqb); + while ((rq = rq_list_pop(&plug->mq_list))) { + bool last = rq_list_empty(plug->mq_list); + blk_status_t ret; + + if (hctx != rq->mq_hctx) { + if (hctx) + blk_mq_commit_rqs(hctx, &queued, from_schedule); + hctx = rq->mq_hctx; + } + + ret = blk_mq_request_issue_directly(rq, last); + switch (ret) { + case BLK_STS_OK: + queued++; + break; + case BLK_STS_RESOURCE: + case BLK_STS_DEV_RESOURCE: + blk_mq_request_bypass_insert(rq, false, last); + blk_mq_commit_rqs(hctx, &queued, from_schedule); + return; + default: + blk_mq_end_request(rq, ret); + errors++; + break; + } + } + + /* + * If we didn't flush the entire list, we could have told the driver + * there was more coming, but that turned out to be a lie. + */ + if (errors) + blk_mq_commit_rqs(hctx, &queued, from_schedule); } void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule) { + struct blk_mq_hw_ctx *this_hctx; + struct blk_mq_ctx *this_ctx; + unsigned int depth; LIST_HEAD(list); - if (list_empty(&plug->mq_list)) + if (rq_list_empty(plug->mq_list)) return; - list_splice_init(&plug->mq_list, &list); - - if (plug->rq_count > 2 && plug->multiple_queues) - list_sort(NULL, &list, plug_rq_cmp); - plug->rq_count = 0; + if (!plug->multiple_queues && !plug->has_elevator && !from_schedule) { + blk_mq_plug_issue_direct(plug, from_schedule); + if (rq_list_empty(plug->mq_list)) + return; + } + + this_hctx = NULL; + this_ctx = NULL; + depth = 0; do { - struct list_head rq_list; - struct request *rq, *head_rq = list_entry_rq(list.next); - struct list_head *pos = &head_rq->queuelist; /* skip first */ - struct blk_mq_hw_ctx *this_hctx = head_rq->mq_hctx; - struct blk_mq_ctx *this_ctx = head_rq->mq_ctx; - unsigned int depth = 1; - - list_for_each_continue(pos, &list) { - rq = list_entry_rq(pos); - BUG_ON(!rq->q); - if (rq->mq_hctx != this_hctx || rq->mq_ctx != this_ctx) - break; - depth++; + struct request *rq; + + rq = rq_list_pop(&plug->mq_list); + + if (!this_hctx) { + this_hctx = rq->mq_hctx; + this_ctx = rq->mq_ctx; + } else if (this_hctx != rq->mq_hctx || this_ctx != rq->mq_ctx) { + trace_block_unplug(this_hctx->queue, depth, + !from_schedule); + blk_mq_sched_insert_requests(this_hctx, this_ctx, + &list, from_schedule); + depth = 0; + this_hctx = rq->mq_hctx; + this_ctx = rq->mq_ctx; + } - list_cut_before(&rq_list, &list, pos); - trace_block_unplug(head_rq->q, depth, !from_schedule); - blk_mq_sched_insert_requests(this_hctx, this_ctx, &rq_list, + list_add(&rq->queuelist, &list); + depth++; + } while (!rq_list_empty(plug->mq_list)); + + if (!list_empty(&list)) { + trace_block_unplug(this_hctx->queue, depth, !from_schedule); + blk_mq_sched_insert_requests(this_hctx, this_ctx, &list, from_schedule); - } while(!list_empty(&list)); + } } static void blk_mq_bio_to_request(struct request *rq, struct bio *bio, @@ -1969,19 +2290,15 @@ static void blk_mq_bio_to_request(struct request *rq, struct bio *bio, } static blk_status_t __blk_mq_issue_directly(struct blk_mq_hw_ctx *hctx, - struct request *rq, - blk_qc_t *cookie, bool last) + struct request *rq, bool last) { struct request_queue *q = rq->q; struct blk_mq_queue_data bd = { .rq = rq, .last = last, }; - blk_qc_t new_cookie; blk_status_t ret; - new_cookie = request_to_qc_t(hctx, rq); - /* * For OK queue, we are done. For error, caller may kill it. * Any other error (busy), just add it to our list as we @@ -1991,7 +2308,6 @@ static blk_status_t __blk_mq_issue_directly(struct blk_mq_hw_ctx *hctx, switch (ret) { case BLK_STS_OK: blk_mq_update_dispatch_busy(hctx, false); - *cookie = new_cookie; break; case BLK_STS_RESOURCE: case BLK_STS_DEV_RESOURCE: @@ -2000,7 +2316,6 @@ static blk_status_t __blk_mq_issue_directly(struct blk_mq_hw_ctx *hctx, break; default: blk_mq_update_dispatch_busy(hctx, false); - *cookie = BLK_QC_T_NONE; break; } @@ -2009,7 +2324,6 @@ static blk_status_t __blk_mq_issue_directly(struct blk_mq_hw_ctx *hctx, static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, struct request *rq, - blk_qc_t *cookie, bool bypass_insert, bool last) { struct request_queue *q = rq->q; @@ -2029,7 +2343,7 @@ static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, goto insert; } - if (q->elevator && !bypass_insert) + if ((rq->rq_flags & RQF_ELV) && !bypass_insert) goto insert; budget_token = blk_mq_get_dispatch_budget(q); @@ -2043,7 +2357,7 @@ static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, goto insert; } - return __blk_mq_issue_directly(hctx, rq, cookie, last); + return __blk_mq_issue_directly(hctx, rq, last); insert: if (bypass_insert) return BLK_STS_RESOURCE; @@ -2057,7 +2371,6 @@ insert: * blk_mq_try_issue_directly - Try to send a request directly to device driver. * @hctx: Pointer of the associated hardware queue. * @rq: Pointer to request to be sent. - * @cookie: Request queue cookie. * * If the device has enough resources to accept a new request now, send the * request directly to device driver. Else, insert at hctx->dispatch queue, so @@ -2065,7 +2378,7 @@ insert: * queue have higher priority. */ static void blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, - struct request *rq, blk_qc_t *cookie) + struct request *rq) { blk_status_t ret; int srcu_idx; @@ -2074,7 +2387,7 @@ static void blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, hctx_lock(hctx, &srcu_idx); - ret = __blk_mq_try_issue_directly(hctx, rq, cookie, false, true); + ret = __blk_mq_try_issue_directly(hctx, rq, false, true); if (ret == BLK_STS_RESOURCE || ret == BLK_STS_DEV_RESOURCE) blk_mq_request_bypass_insert(rq, false, true); else if (ret != BLK_STS_OK) @@ -2087,11 +2400,10 @@ blk_status_t blk_mq_request_issue_directly(struct request *rq, bool last) { blk_status_t ret; int srcu_idx; - blk_qc_t unused_cookie; struct blk_mq_hw_ctx *hctx = rq->mq_hctx; hctx_lock(hctx, &srcu_idx); - ret = __blk_mq_try_issue_directly(hctx, rq, &unused_cookie, true, last); + ret = __blk_mq_try_issue_directly(hctx, rq, true, last); hctx_unlock(hctx, srcu_idx); return ret; @@ -2135,27 +2447,28 @@ void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx, static void blk_add_rq_to_plug(struct blk_plug *plug, struct request *rq) { - list_add_tail(&rq->queuelist, &plug->mq_list); - plug->rq_count++; - if (!plug->multiple_queues && !list_is_singular(&plug->mq_list)) { - struct request *tmp; + if (!plug->multiple_queues) { + struct request *nxt = rq_list_peek(&plug->mq_list); - tmp = list_first_entry(&plug->mq_list, struct request, - queuelist); - if (tmp->q != rq->q) + if (nxt && nxt->q != rq->q) plug->multiple_queues = true; } + if (!plug->has_elevator && (rq->rq_flags & RQF_ELV)) + plug->has_elevator = true; + rq->rq_next = NULL; + rq_list_add(&plug->mq_list, rq); + plug->rq_count++; } /* - * Allow 4x BLK_MAX_REQUEST_COUNT requests on plug queue for multiple + * Allow 2x BLK_MAX_REQUEST_COUNT requests on plug queue for multiple * queues. This is important for md arrays to benefit from merging * requests. */ static inline unsigned short blk_plug_max_rq_count(struct blk_plug *plug) { if (plug->multiple_queues) - return BLK_MAX_REQUEST_COUNT * 4; + return BLK_MAX_REQUEST_COUNT * 2; return BLK_MAX_REQUEST_COUNT; } @@ -2171,57 +2484,63 @@ static inline unsigned short blk_plug_max_rq_count(struct blk_plug *plug) * * It will not queue the request if there is an error with the bio, or at the * request creation. - * - * Returns: Request queue cookie. */ -blk_qc_t blk_mq_submit_bio(struct bio *bio) +void blk_mq_submit_bio(struct bio *bio) { - struct request_queue *q = bio->bi_bdev->bd_disk->queue; + struct request_queue *q = bdev_get_queue(bio->bi_bdev); const int is_sync = op_is_sync(bio->bi_opf); - const int is_flush_fua = op_is_flush(bio->bi_opf); - struct blk_mq_alloc_data data = { - .q = q, - }; struct request *rq; struct blk_plug *plug; - struct request *same_queue_rq = NULL; - unsigned int nr_segs; - blk_qc_t cookie; + bool same_queue_rq = false; + unsigned int nr_segs = 1; blk_status_t ret; - bool hipri; blk_queue_bounce(q, &bio); - __blk_queue_split(&bio, &nr_segs); + if (blk_may_split(q, bio)) + __blk_queue_split(q, &bio, &nr_segs); if (!bio_integrity_prep(bio)) goto queue_exit; - if (!is_flush_fua && !blk_queue_nomerges(q) && - blk_attempt_plug_merge(q, bio, nr_segs, &same_queue_rq)) - goto queue_exit; - - if (blk_mq_sched_bio_merge(q, bio, nr_segs)) - goto queue_exit; + if (!blk_queue_nomerges(q) && bio_mergeable(bio)) { + if (blk_attempt_plug_merge(q, bio, nr_segs, &same_queue_rq)) + goto queue_exit; + if (blk_mq_sched_bio_merge(q, bio, nr_segs)) + goto queue_exit; + } rq_qos_throttle(q, bio); - hipri = bio->bi_opf & REQ_HIPRI; - - data.cmd_flags = bio->bi_opf; - rq = __blk_mq_alloc_request(&data); - if (unlikely(!rq)) { - rq_qos_cleanup(q, bio); - if (bio->bi_opf & REQ_NOWAIT) - bio_wouldblock_error(bio); - goto queue_exit; + plug = blk_mq_plug(q, bio); + if (plug && plug->cached_rq) { + rq = rq_list_pop(&plug->cached_rq); + INIT_LIST_HEAD(&rq->queuelist); + } else { + struct blk_mq_alloc_data data = { + .q = q, + .nr_tags = 1, + .cmd_flags = bio->bi_opf, + .rq_flags = q->elevator ? RQF_ELV : 0, + }; + + if (plug) { + data.nr_tags = plug->nr_ios; + plug->nr_ios = 1; + data.cached_rq = &plug->cached_rq; + } + rq = __blk_mq_alloc_requests(&data); + if (unlikely(!rq)) { + rq_qos_cleanup(q, bio); + if (bio->bi_opf & REQ_NOWAIT) + bio_wouldblock_error(bio); + goto queue_exit; + } } trace_block_getrq(bio); rq_qos_track(q, rq, bio); - cookie = request_to_qc_t(data.hctx, rq); - blk_mq_bio_to_request(rq, bio, nr_segs); ret = blk_crypto_init_request(rq); @@ -2229,17 +2548,15 @@ blk_qc_t blk_mq_submit_bio(struct bio *bio) bio->bi_status = ret; bio_endio(bio); blk_mq_free_request(rq); - return BLK_QC_T_NONE; + return; } - plug = blk_mq_plug(q, bio); - if (unlikely(is_flush_fua)) { - /* Bypass scheduler for flush requests */ - blk_insert_flush(rq); - blk_mq_run_hw_queue(data.hctx, true); - } else if (plug && (q->nr_hw_queues == 1 || - blk_mq_is_sbitmap_shared(rq->mq_hctx->flags) || - q->mq_ops->commit_rqs || !blk_queue_nonrot(q))) { + if (op_is_flush(bio->bi_opf) && blk_insert_flush(rq)) + return; + + if (plug && (q->nr_hw_queues == 1 || + blk_mq_is_shared_tags(rq->mq_hctx->flags) || + q->mq_ops->commit_rqs || !blk_queue_nonrot(q))) { /* * Use plugging if we have a ->commit_rqs() hook as well, as * we know the driver uses bd->last in a smart fashion. @@ -2250,22 +2567,26 @@ blk_qc_t blk_mq_submit_bio(struct bio *bio) unsigned int request_count = plug->rq_count; struct request *last = NULL; - if (!request_count) + if (!request_count) { trace_block_plug(q); - else - last = list_entry_rq(plug->mq_list.prev); + } else if (!blk_queue_nomerges(q)) { + last = rq_list_peek(&plug->mq_list); + if (blk_rq_bytes(last) < BLK_PLUG_FLUSH_SIZE) + last = NULL; + } - if (request_count >= blk_plug_max_rq_count(plug) || (last && - blk_rq_bytes(last) >= BLK_PLUG_FLUSH_SIZE)) { - blk_flush_plug_list(plug, false); + if (request_count >= blk_plug_max_rq_count(plug) || last) { + blk_mq_flush_plug_list(plug, false); trace_block_plug(q); } blk_add_rq_to_plug(plug, rq); - } else if (q->elevator) { + } else if (rq->rq_flags & RQF_ELV) { /* Insert the request at the IO scheduler queue */ blk_mq_sched_insert_request(rq, false, true, true); } else if (plug && !blk_queue_nomerges(q)) { + struct request *next_rq = NULL; + /* * We do limited plugging. If the bio can be merged, do that. * Otherwise the existing request in the plug list will be @@ -2273,39 +2594,32 @@ blk_qc_t blk_mq_submit_bio(struct bio *bio) * The plug list might get flushed before this. If that happens, * the plug list is empty, and same_queue_rq is invalid. */ - if (list_empty(&plug->mq_list)) - same_queue_rq = NULL; if (same_queue_rq) { - list_del_init(&same_queue_rq->queuelist); + next_rq = rq_list_pop(&plug->mq_list); plug->rq_count--; } blk_add_rq_to_plug(plug, rq); trace_block_plug(q); - if (same_queue_rq) { - data.hctx = same_queue_rq->mq_hctx; + if (next_rq) { trace_block_unplug(q, 1, true); - blk_mq_try_issue_directly(data.hctx, same_queue_rq, - &cookie); + blk_mq_try_issue_directly(next_rq->mq_hctx, next_rq); } } else if ((q->nr_hw_queues > 1 && is_sync) || - !data.hctx->dispatch_busy) { + !rq->mq_hctx->dispatch_busy) { /* * There is no scheduler and we can try to send directly * to the hardware. */ - blk_mq_try_issue_directly(data.hctx, rq, &cookie); + blk_mq_try_issue_directly(rq->mq_hctx, rq); } else { /* Default case. */ blk_mq_sched_insert_request(rq, false, true, true); } - if (!hipri) - return BLK_QC_T_NONE; - return cookie; + return; queue_exit: blk_queue_exit(q); - return BLK_QC_T_NONE; } static size_t order_to_size(unsigned int order) @@ -2314,19 +2628,22 @@ static size_t order_to_size(unsigned int order) } /* called before freeing request pool in @tags */ -static void blk_mq_clear_rq_mapping(struct blk_mq_tag_set *set, - struct blk_mq_tags *tags, unsigned int hctx_idx) +static void blk_mq_clear_rq_mapping(struct blk_mq_tags *drv_tags, + struct blk_mq_tags *tags) { - struct blk_mq_tags *drv_tags = set->tags[hctx_idx]; struct page *page; unsigned long flags; + /* There is no need to clear a driver tags own mapping */ + if (drv_tags == tags) + return; + list_for_each_entry(page, &tags->page_list, lru) { unsigned long start = (unsigned long)page_address(page); unsigned long end = start + order_to_size(page->private); int i; - for (i = 0; i < set->queue_depth; i++) { + for (i = 0; i < drv_tags->nr_tags; i++) { struct request *rq = drv_tags->rqs[i]; unsigned long rq_addr = (unsigned long)rq; @@ -2350,9 +2667,15 @@ static void blk_mq_clear_rq_mapping(struct blk_mq_tag_set *set, void blk_mq_free_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags, unsigned int hctx_idx) { + struct blk_mq_tags *drv_tags; struct page *page; - if (tags->rqs && set->ops->exit_request) { + if (blk_mq_is_shared_tags(set->flags)) + drv_tags = set->shared_tags; + else + drv_tags = set->tags[hctx_idx]; + + if (tags->static_rqs && set->ops->exit_request) { int i; for (i = 0; i < tags->nr_tags; i++) { @@ -2365,7 +2688,7 @@ void blk_mq_free_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags, } } - blk_mq_clear_rq_mapping(set, tags, hctx_idx); + blk_mq_clear_rq_mapping(drv_tags, tags); while (!list_empty(&tags->page_list)) { page = list_first_entry(&tags->page_list, struct page, lru); @@ -2379,21 +2702,20 @@ void blk_mq_free_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags, } } -void blk_mq_free_rq_map(struct blk_mq_tags *tags, unsigned int flags) +void blk_mq_free_rq_map(struct blk_mq_tags *tags) { kfree(tags->rqs); tags->rqs = NULL; kfree(tags->static_rqs); tags->static_rqs = NULL; - blk_mq_free_tags(tags, flags); + blk_mq_free_tags(tags); } -struct blk_mq_tags *blk_mq_alloc_rq_map(struct blk_mq_tag_set *set, - unsigned int hctx_idx, - unsigned int nr_tags, - unsigned int reserved_tags, - unsigned int flags) +static struct blk_mq_tags *blk_mq_alloc_rq_map(struct blk_mq_tag_set *set, + unsigned int hctx_idx, + unsigned int nr_tags, + unsigned int reserved_tags) { struct blk_mq_tags *tags; int node; @@ -2402,7 +2724,8 @@ struct blk_mq_tags *blk_mq_alloc_rq_map(struct blk_mq_tag_set *set, if (node == NUMA_NO_NODE) node = set->numa_node; - tags = blk_mq_init_tags(nr_tags, reserved_tags, node, flags); + tags = blk_mq_init_tags(nr_tags, reserved_tags, node, + BLK_MQ_FLAG_TO_ALLOC_POLICY(set->flags)); if (!tags) return NULL; @@ -2410,7 +2733,7 @@ struct blk_mq_tags *blk_mq_alloc_rq_map(struct blk_mq_tag_set *set, GFP_NOIO | __GFP_NOWARN | __GFP_NORETRY, node); if (!tags->rqs) { - blk_mq_free_tags(tags, flags); + blk_mq_free_tags(tags); return NULL; } @@ -2419,7 +2742,7 @@ struct blk_mq_tags *blk_mq_alloc_rq_map(struct blk_mq_tag_set *set, node); if (!tags->static_rqs) { kfree(tags->rqs); - blk_mq_free_tags(tags, flags); + blk_mq_free_tags(tags); return NULL; } @@ -2441,8 +2764,9 @@ static int blk_mq_init_request(struct blk_mq_tag_set *set, struct request *rq, return 0; } -int blk_mq_alloc_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags, - unsigned int hctx_idx, unsigned int depth) +static int blk_mq_alloc_rqs(struct blk_mq_tag_set *set, + struct blk_mq_tags *tags, + unsigned int hctx_idx, unsigned int depth) { unsigned int i, j, entries_per_page, max_order = 4; size_t rq_size, left; @@ -2853,37 +3177,58 @@ static void blk_mq_init_cpu_queues(struct request_queue *q, } } -static bool __blk_mq_alloc_map_and_request(struct blk_mq_tag_set *set, - int hctx_idx) +struct blk_mq_tags *blk_mq_alloc_map_and_rqs(struct blk_mq_tag_set *set, + unsigned int hctx_idx, + unsigned int depth) { - unsigned int flags = set->flags; - int ret = 0; + struct blk_mq_tags *tags; + int ret; - set->tags[hctx_idx] = blk_mq_alloc_rq_map(set, hctx_idx, - set->queue_depth, set->reserved_tags, flags); - if (!set->tags[hctx_idx]) - return false; + tags = blk_mq_alloc_rq_map(set, hctx_idx, depth, set->reserved_tags); + if (!tags) + return NULL; - ret = blk_mq_alloc_rqs(set, set->tags[hctx_idx], hctx_idx, - set->queue_depth); - if (!ret) - return true; + ret = blk_mq_alloc_rqs(set, tags, hctx_idx, depth); + if (ret) { + blk_mq_free_rq_map(tags); + return NULL; + } - blk_mq_free_rq_map(set->tags[hctx_idx], flags); - set->tags[hctx_idx] = NULL; - return false; + return tags; } -static void blk_mq_free_map_and_requests(struct blk_mq_tag_set *set, - unsigned int hctx_idx) +static bool __blk_mq_alloc_map_and_rqs(struct blk_mq_tag_set *set, + int hctx_idx) { - unsigned int flags = set->flags; + if (blk_mq_is_shared_tags(set->flags)) { + set->tags[hctx_idx] = set->shared_tags; - if (set->tags && set->tags[hctx_idx]) { - blk_mq_free_rqs(set, set->tags[hctx_idx], hctx_idx); - blk_mq_free_rq_map(set->tags[hctx_idx], flags); - set->tags[hctx_idx] = NULL; + return true; } + + set->tags[hctx_idx] = blk_mq_alloc_map_and_rqs(set, hctx_idx, + set->queue_depth); + + return set->tags[hctx_idx]; +} + +void blk_mq_free_map_and_rqs(struct blk_mq_tag_set *set, + struct blk_mq_tags *tags, + unsigned int hctx_idx) +{ + if (tags) { + blk_mq_free_rqs(set, tags, hctx_idx); + blk_mq_free_rq_map(tags); + } +} + +static void __blk_mq_free_map_and_rqs(struct blk_mq_tag_set *set, + unsigned int hctx_idx) +{ + if (!blk_mq_is_shared_tags(set->flags)) + blk_mq_free_map_and_rqs(set, set->tags[hctx_idx], hctx_idx); + + set->tags[hctx_idx] = NULL; } static void blk_mq_map_swqueue(struct request_queue *q) @@ -2916,7 +3261,7 @@ static void blk_mq_map_swqueue(struct request_queue *q) hctx_idx = set->map[j].mq_map[i]; /* unmapped hw queue can be remapped after CPU topo changed */ if (!set->tags[hctx_idx] && - !__blk_mq_alloc_map_and_request(set, hctx_idx)) { + !__blk_mq_alloc_map_and_rqs(set, hctx_idx)) { /* * If tags initialization fail for some hctx, * that hctx won't be brought online. In this @@ -2963,8 +3308,8 @@ static void blk_mq_map_swqueue(struct request_queue *q) * fallback in case of a new remap fails * allocation */ - if (i && set->tags[i]) - blk_mq_free_map_and_requests(set, i); + if (i) + __blk_mq_free_map_and_rqs(set, i); hctx->tags = NULL; continue; @@ -3260,8 +3605,7 @@ static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set, struct blk_mq_hw_ctx *hctx = hctxs[j]; if (hctx) { - if (hctx->tags) - blk_mq_free_map_and_requests(set, j); + __blk_mq_free_map_and_rqs(set, j); blk_mq_exit_hctx(q, set, hctx, j); hctxs[j] = NULL; } @@ -3348,8 +3692,16 @@ static int __blk_mq_alloc_rq_maps(struct blk_mq_tag_set *set) { int i; + if (blk_mq_is_shared_tags(set->flags)) { + set->shared_tags = blk_mq_alloc_map_and_rqs(set, + BLK_MQ_NO_HCTX_IDX, + set->queue_depth); + if (!set->shared_tags) + return -ENOMEM; + } + for (i = 0; i < set->nr_hw_queues; i++) { - if (!__blk_mq_alloc_map_and_request(set, i)) + if (!__blk_mq_alloc_map_and_rqs(set, i)) goto out_unwind; cond_resched(); } @@ -3358,7 +3710,12 @@ static int __blk_mq_alloc_rq_maps(struct blk_mq_tag_set *set) out_unwind: while (--i >= 0) - blk_mq_free_map_and_requests(set, i); + __blk_mq_free_map_and_rqs(set, i); + + if (blk_mq_is_shared_tags(set->flags)) { + blk_mq_free_map_and_rqs(set, set->shared_tags, + BLK_MQ_NO_HCTX_IDX); + } return -ENOMEM; } @@ -3368,7 +3725,7 @@ out_unwind: * may reduce the depth asked for, if memory is tight. set->queue_depth * will be updated to reflect the allocated depth. */ -static int blk_mq_alloc_map_and_requests(struct blk_mq_tag_set *set) +static int blk_mq_alloc_set_map_and_rqs(struct blk_mq_tag_set *set) { unsigned int depth; int err; @@ -3534,27 +3891,15 @@ int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set) if (ret) goto out_free_mq_map; - ret = blk_mq_alloc_map_and_requests(set); + ret = blk_mq_alloc_set_map_and_rqs(set); if (ret) goto out_free_mq_map; - if (blk_mq_is_sbitmap_shared(set->flags)) { - atomic_set(&set->active_queues_shared_sbitmap, 0); - - if (blk_mq_init_shared_sbitmap(set)) { - ret = -ENOMEM; - goto out_free_mq_rq_maps; - } - } - mutex_init(&set->tag_list_lock); INIT_LIST_HEAD(&set->tag_list); return 0; -out_free_mq_rq_maps: - for (i = 0; i < set->nr_hw_queues; i++) - blk_mq_free_map_and_requests(set, i); out_free_mq_map: for (i = 0; i < set->nr_maps; i++) { kfree(set->map[i].mq_map); @@ -3587,10 +3932,12 @@ void blk_mq_free_tag_set(struct blk_mq_tag_set *set) int i, j; for (i = 0; i < set->nr_hw_queues; i++) - blk_mq_free_map_and_requests(set, i); + __blk_mq_free_map_and_rqs(set, i); - if (blk_mq_is_sbitmap_shared(set->flags)) - blk_mq_exit_shared_sbitmap(set); + if (blk_mq_is_shared_tags(set->flags)) { + blk_mq_free_map_and_rqs(set, set->shared_tags, + BLK_MQ_NO_HCTX_IDX); + } for (j = 0; j < set->nr_maps; j++) { kfree(set->map[j].mq_map); @@ -3625,20 +3972,12 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr) * If we're using an MQ scheduler, just update the scheduler * queue depth. This is similar to what the old code would do. */ - if (!hctx->sched_tags) { - ret = blk_mq_tag_update_depth(hctx, &hctx->tags, nr, - false); - if (!ret && blk_mq_is_sbitmap_shared(set->flags)) - blk_mq_tag_resize_shared_sbitmap(set, nr); - } else { + if (hctx->sched_tags) { ret = blk_mq_tag_update_depth(hctx, &hctx->sched_tags, - nr, true); - if (blk_mq_is_sbitmap_shared(set->flags)) { - hctx->sched_tags->bitmap_tags = - &q->sched_bitmap_tags; - hctx->sched_tags->breserved_tags = - &q->sched_breserved_tags; - } + nr, true); + } else { + ret = blk_mq_tag_update_depth(hctx, &hctx->tags, nr, + false); } if (ret) break; @@ -3647,9 +3986,12 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr) } if (!ret) { q->nr_requests = nr; - if (q->elevator && blk_mq_is_sbitmap_shared(set->flags)) - sbitmap_queue_resize(&q->sched_bitmap_tags, - nr - set->reserved_tags); + if (blk_mq_is_shared_tags(set->flags)) { + if (q->elevator) + blk_mq_tag_update_sched_shared_tags(q); + else + blk_mq_tag_resize_shared_tags(set, nr); + } } blk_mq_unquiesce_queue(q); @@ -3868,15 +4210,20 @@ static unsigned long blk_mq_poll_nsecs(struct request_queue *q, return ret; } -static bool blk_mq_poll_hybrid_sleep(struct request_queue *q, - struct request *rq) +static bool blk_mq_poll_hybrid(struct request_queue *q, blk_qc_t qc) { + struct blk_mq_hw_ctx *hctx = blk_qc_to_hctx(q, qc); + struct request *rq = blk_qc_to_rq(hctx, qc); struct hrtimer_sleeper hs; enum hrtimer_mode mode; unsigned int nsecs; ktime_t kt; - if (rq->rq_flags & RQF_MQ_POLL_SLEPT) + /* + * If a request has completed on queue that uses an I/O scheduler, we + * won't get back a request from blk_qc_to_rq. + */ + if (!rq || (rq->rq_flags & RQF_MQ_POLL_SLEPT)) return false; /* @@ -3918,92 +4265,37 @@ static bool blk_mq_poll_hybrid_sleep(struct request_queue *q, __set_current_state(TASK_RUNNING); destroy_hrtimer_on_stack(&hs.timer); - return true; -} - -static bool blk_mq_poll_hybrid(struct request_queue *q, - struct blk_mq_hw_ctx *hctx, blk_qc_t cookie) -{ - struct request *rq; - - if (q->poll_nsec == BLK_MQ_POLL_CLASSIC) - return false; - - if (!blk_qc_t_is_internal(cookie)) - rq = blk_mq_tag_to_rq(hctx->tags, blk_qc_t_to_tag(cookie)); - else { - rq = blk_mq_tag_to_rq(hctx->sched_tags, blk_qc_t_to_tag(cookie)); - /* - * With scheduling, if the request has completed, we'll - * get a NULL return here, as we clear the sched tag when - * that happens. The request still remains valid, like always, - * so we should be safe with just the NULL check. - */ - if (!rq) - return false; - } - - return blk_mq_poll_hybrid_sleep(q, rq); -} - -/** - * blk_poll - poll for IO completions - * @q: the queue - * @cookie: cookie passed back at IO submission time - * @spin: whether to spin for completions - * - * Description: - * Poll for completions on the passed in queue. Returns number of - * completed entries found. If @spin is true, then blk_poll will continue - * looping until at least one completion is found, unless the task is - * otherwise marked running (or we need to reschedule). - */ -int blk_poll(struct request_queue *q, blk_qc_t cookie, bool spin) -{ - struct blk_mq_hw_ctx *hctx; - unsigned int state; - - if (!blk_qc_t_valid(cookie) || - !test_bit(QUEUE_FLAG_POLL, &q->queue_flags)) - return 0; - - if (current->plug) - blk_flush_plug_list(current->plug, false); - - hctx = q->queue_hw_ctx[blk_qc_t_to_queue_num(cookie)]; /* - * 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 specified not to spin, - * we also should not sleep. + * 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 (spin && blk_mq_poll_hybrid(q, hctx, cookie)) - return 1; + return true; +} - hctx->poll_considered++; +static int blk_mq_poll_classic(struct request_queue *q, blk_qc_t cookie, + struct io_comp_batch *iob, unsigned int flags) +{ + struct blk_mq_hw_ctx *hctx = blk_qc_to_hctx(q, cookie); + long state = get_current_state(); + int ret; - state = get_current_state(); do { - int ret; - - hctx->poll_invoked++; - - ret = q->mq_ops->poll(hctx); + ret = q->mq_ops->poll(hctx, iob); if (ret > 0) { - hctx->poll_success++; __set_current_state(TASK_RUNNING); return ret; } if (signal_pending_state(state, current)) __set_current_state(TASK_RUNNING); - if (task_is_running(current)) return 1; - if (ret < 0 || !spin) + + if (ret < 0 || (flags & BLK_POLL_ONESHOT)) break; cpu_relax(); } while (!need_resched()); @@ -4011,7 +4303,17 @@ int blk_poll(struct request_queue *q, blk_qc_t cookie, bool spin) __set_current_state(TASK_RUNNING); return 0; } -EXPORT_SYMBOL_GPL(blk_poll); + +int blk_mq_poll(struct request_queue *q, blk_qc_t cookie, struct io_comp_batch *iob, + unsigned int flags) +{ + if (!(flags & BLK_POLL_NOSLEEP) && + q->poll_nsec != BLK_MQ_POLL_CLASSIC) { + if (blk_mq_poll_hybrid(q, cookie)) + return 1; + } + return blk_mq_poll_classic(q, cookie, iob, flags); +} unsigned int blk_mq_rq_cpu(struct request *rq) { diff --git a/block/blk-mq.h b/block/blk-mq.h index d08779f77a26..28859fc5faee 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -25,18 +25,14 @@ struct blk_mq_ctx { unsigned short index_hw[HCTX_MAX_TYPES]; struct blk_mq_hw_ctx *hctxs[HCTX_MAX_TYPES]; - /* incremented at dispatch time */ - unsigned long rq_dispatched[2]; - unsigned long rq_merged; - - /* incremented at completion time */ - unsigned long ____cacheline_aligned_in_smp rq_completed[2]; - struct request_queue *queue; struct blk_mq_ctxs *ctxs; struct kobject kobj; } ____cacheline_aligned_in_smp; +void blk_mq_submit_bio(struct bio *bio); +int blk_mq_poll(struct request_queue *q, blk_qc_t cookie, struct io_comp_batch *iob, + unsigned int flags); void blk_mq_exit_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); @@ -54,15 +50,12 @@ void blk_mq_put_rq_ref(struct request *rq); */ void blk_mq_free_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags, unsigned int hctx_idx); -void blk_mq_free_rq_map(struct blk_mq_tags *tags, unsigned int flags); -struct blk_mq_tags *blk_mq_alloc_rq_map(struct blk_mq_tag_set *set, - unsigned int hctx_idx, - unsigned int nr_tags, - unsigned int reserved_tags, - unsigned int flags); -int blk_mq_alloc_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags, - unsigned int hctx_idx, unsigned int depth); - +void blk_mq_free_rq_map(struct blk_mq_tags *tags); +struct blk_mq_tags *blk_mq_alloc_map_and_rqs(struct blk_mq_tag_set *set, + unsigned int hctx_idx, unsigned int depth); +void blk_mq_free_map_and_rqs(struct blk_mq_tag_set *set, + struct blk_mq_tags *tags, + unsigned int hctx_idx); /* * Internal helpers for request insertion into sw queues */ @@ -109,9 +102,9 @@ static inline struct blk_mq_hw_ctx *blk_mq_map_queue(struct request_queue *q, enum hctx_type type = HCTX_TYPE_DEFAULT; /* - * The caller ensure that if REQ_HIPRI, poll must be enabled. + * The caller ensure that if REQ_POLLED, poll must be enabled. */ - if (flags & REQ_HIPRI) + if (flags & REQ_POLLED) type = HCTX_TYPE_POLL; else if ((flags & REQ_OP_MASK) == REQ_OP_READ) type = HCTX_TYPE_READ; @@ -128,6 +121,8 @@ extern int __blk_mq_register_dev(struct device *dev, struct request_queue *q); extern int blk_mq_sysfs_register(struct request_queue *q); extern void blk_mq_sysfs_unregister(struct request_queue *q); extern void blk_mq_hctx_kobj_init(struct blk_mq_hw_ctx *hctx); +void blk_mq_free_plug_rqs(struct blk_plug *plug); +void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule); void blk_mq_release(struct request_queue *q); @@ -154,23 +149,27 @@ struct blk_mq_alloc_data { blk_mq_req_flags_t flags; unsigned int shallow_depth; unsigned int cmd_flags; + unsigned int rq_flags; + + /* allocate multiple requests/tags in one go */ + unsigned int nr_tags; + struct request **cached_rq; /* input & output parameter */ struct blk_mq_ctx *ctx; struct blk_mq_hw_ctx *hctx; }; -static inline bool blk_mq_is_sbitmap_shared(unsigned int flags) +static inline bool blk_mq_is_shared_tags(unsigned int flags) { return flags & BLK_MQ_F_TAG_HCTX_SHARED; } static inline struct blk_mq_tags *blk_mq_tags_from_data(struct blk_mq_alloc_data *data) { - if (data->q->elevator) - return data->hctx->sched_tags; - - return data->hctx->tags; + if (!(data->rq_flags & RQF_ELV)) + return data->hctx->tags; + return data->hctx->sched_tags; } static inline bool blk_mq_hctx_stopped(struct blk_mq_hw_ctx *hctx) @@ -220,24 +219,24 @@ static inline int blk_mq_get_rq_budget_token(struct request *rq) static inline void __blk_mq_inc_active_requests(struct blk_mq_hw_ctx *hctx) { - if (blk_mq_is_sbitmap_shared(hctx->flags)) - atomic_inc(&hctx->queue->nr_active_requests_shared_sbitmap); + if (blk_mq_is_shared_tags(hctx->flags)) + atomic_inc(&hctx->queue->nr_active_requests_shared_tags); else atomic_inc(&hctx->nr_active); } static inline void __blk_mq_dec_active_requests(struct blk_mq_hw_ctx *hctx) { - if (blk_mq_is_sbitmap_shared(hctx->flags)) - atomic_dec(&hctx->queue->nr_active_requests_shared_sbitmap); + if (blk_mq_is_shared_tags(hctx->flags)) + atomic_dec(&hctx->queue->nr_active_requests_shared_tags); else atomic_dec(&hctx->nr_active); } static inline int __blk_mq_active_requests(struct blk_mq_hw_ctx *hctx) { - if (blk_mq_is_sbitmap_shared(hctx->flags)) - return atomic_read(&hctx->queue->nr_active_requests_shared_sbitmap); + if (blk_mq_is_shared_tags(hctx->flags)) + return atomic_read(&hctx->queue->nr_active_requests_shared_tags); return atomic_read(&hctx->nr_active); } static inline void __blk_mq_put_driver_tag(struct blk_mq_hw_ctx *hctx, @@ -260,7 +259,20 @@ static inline void blk_mq_put_driver_tag(struct request *rq) __blk_mq_put_driver_tag(rq->mq_hctx, rq); } -bool blk_mq_get_driver_tag(struct request *rq); +bool __blk_mq_get_driver_tag(struct blk_mq_hw_ctx *hctx, struct request *rq); + +static inline bool blk_mq_get_driver_tag(struct request *rq) +{ + struct blk_mq_hw_ctx *hctx = rq->mq_hctx; + + if (rq->tag != BLK_MQ_NO_TAG && + !(hctx->flags & BLK_MQ_F_TAG_QUEUE_SHARED)) { + hctx->tags->rqs[rq->tag] = rq; + return true; + } + + return __blk_mq_get_driver_tag(hctx, rq); +} static inline void blk_mq_clear_mq_map(struct blk_mq_queue_map *qmap) { @@ -331,19 +343,18 @@ static inline bool hctx_may_queue(struct blk_mq_hw_ctx *hctx, if (bt->sb.depth == 1) return true; - if (blk_mq_is_sbitmap_shared(hctx->flags)) { + if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - struct blk_mq_tag_set *set = q->tag_set; if (!test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) return true; - users = atomic_read(&set->active_queues_shared_sbitmap); } else { if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) return true; - users = atomic_read(&hctx->tags->active_queues); } + users = atomic_read(&hctx->tags->active_queues); + if (!users) return true; diff --git a/block/blk-rq-qos.h b/block/blk-rq-qos.h index f000f83e0621..3cfbc8668cba 100644 --- a/block/blk-rq-qos.h +++ b/block/blk-rq-qos.h @@ -189,9 +189,10 @@ static inline void rq_qos_throttle(struct request_queue *q, struct bio *bio) * BIO_TRACKED lets controllers know that a bio went through the * normal rq_qos path. */ - bio_set_flag(bio, BIO_TRACKED); - if (q->rq_qos) + if (q->rq_qos) { + bio_set_flag(bio, BIO_TRACKED); __rq_qos_throttle(q->rq_qos, bio); + } } static inline void rq_qos_track(struct request_queue *q, struct request *rq, diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 614d9d47de36..cef1f713370b 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -17,6 +17,7 @@ #include "blk-mq.h" #include "blk-mq-debugfs.h" #include "blk-wbt.h" +#include "blk-throttle.h" struct queue_sysfs_entry { struct attribute attr; @@ -432,26 +433,11 @@ static ssize_t queue_poll_show(struct request_queue *q, char *page) static ssize_t queue_poll_store(struct request_queue *q, const char *page, size_t count) { - unsigned long poll_on; - ssize_t ret; - - if (!q->tag_set || q->tag_set->nr_maps <= HCTX_TYPE_POLL || - !q->tag_set->map[HCTX_TYPE_POLL].nr_queues) + if (!test_bit(QUEUE_FLAG_POLL, &q->queue_flags)) return -EINVAL; - - ret = queue_var_store(&poll_on, page, count); - if (ret < 0) - return ret; - - if (poll_on) { - blk_queue_flag_set(QUEUE_FLAG_POLL, q); - } else { - blk_mq_freeze_queue(q); - blk_queue_flag_clear(QUEUE_FLAG_POLL, q); - blk_mq_unfreeze_queue(q); - } - - return ret; + pr_info_ratelimited("writes to the poll attribute are ignored.\n"); + pr_info_ratelimited("please use driver specific parameters instead.\n"); + return count; } static ssize_t queue_io_timeout_show(struct request_queue *q, char *page) @@ -887,16 +873,15 @@ int blk_register_queue(struct gendisk *disk) } mutex_lock(&q->sysfs_lock); + + ret = disk_register_independent_access_ranges(disk, NULL); + if (ret) + goto put_dev; + if (q->elevator) { ret = elv_register_queue(q, false); - if (ret) { - mutex_unlock(&q->sysfs_lock); - mutex_unlock(&q->sysfs_dir_lock); - kobject_del(&q->kobj); - blk_trace_remove_sysfs(dev); - kobject_put(&dev->kobj); - return ret; - } + if (ret) + goto put_dev; } blk_queue_flag_set(QUEUE_FLAG_REGISTERED, q); @@ -928,6 +913,16 @@ unlock: } return ret; + +put_dev: + disk_unregister_independent_access_ranges(disk); + mutex_unlock(&q->sysfs_lock); + mutex_unlock(&q->sysfs_dir_lock); + kobject_del(&q->kobj); + blk_trace_remove_sysfs(dev); + kobject_put(&dev->kobj); + + return ret; } /** @@ -972,6 +967,7 @@ void blk_unregister_queue(struct gendisk *disk) mutex_lock(&q->sysfs_lock); if (q->elevator) elv_unregister_queue(q); + disk_unregister_independent_access_ranges(disk); mutex_unlock(&q->sysfs_lock); mutex_unlock(&q->sysfs_dir_lock); diff --git a/block/blk-throttle.c b/block/blk-throttle.c index 7c4e7993ba97..39bb6e68a9a2 100644 --- a/block/blk-throttle.c +++ b/block/blk-throttle.c @@ -13,6 +13,7 @@ #include <linux/blk-cgroup.h> #include "blk.h" #include "blk-cgroup-rwstat.h" +#include "blk-throttle.h" /* Max dispatch from a group in 1 round */ #define THROTL_GRP_QUANTUM 8 @@ -37,60 +38,9 @@ */ #define LATENCY_FILTERED_HD (1000L) /* 1ms */ -static struct blkcg_policy blkcg_policy_throtl; - /* A workqueue to queue throttle related work */ static struct workqueue_struct *kthrotld_workqueue; -/* - * To implement hierarchical throttling, throtl_grps form a tree and bios - * are dispatched upwards level by level until they reach the top and get - * issued. When dispatching bios from the children and local group at each - * level, if the bios are dispatched into a single bio_list, there's a risk - * of a local or child group which can queue many bios at once filling up - * the list starving others. - * - * To avoid such starvation, dispatched bios are queued separately - * according to where they came from. When they are again dispatched to - * the parent, they're popped in round-robin order so that no single source - * hogs the dispatch window. - * - * throtl_qnode is used to keep the queued bios separated by their sources. - * Bios are queued to throtl_qnode which in turn is queued to - * throtl_service_queue and then dispatched in round-robin order. - * - * It's also used to track the reference counts on blkg's. A qnode always - * belongs to a throtl_grp and gets queued on itself or the parent, so - * incrementing the reference of the associated throtl_grp when a qnode is - * queued and decrementing when dequeued is enough to keep the whole blkg - * tree pinned while bios are in flight. - */ -struct throtl_qnode { - struct list_head node; /* service_queue->queued[] */ - struct bio_list bios; /* queued bios */ - struct throtl_grp *tg; /* tg this qnode belongs to */ -}; - -struct throtl_service_queue { - struct throtl_service_queue *parent_sq; /* the parent service_queue */ - - /* - * Bios queued directly to this service_queue or dispatched from - * children throtl_grp's. - */ - struct list_head queued[2]; /* throtl_qnode [READ/WRITE] */ - unsigned int nr_queued[2]; /* number of queued bios */ - - /* - * RB tree of active children throtl_grp's, which are sorted by - * their ->disptime. - */ - struct rb_root_cached pending_tree; /* RB tree of active tgs */ - unsigned int nr_pending; /* # queued in the tree */ - unsigned long first_pending_disptime; /* disptime of the first tg */ - struct timer_list pending_timer; /* fires on first_pending_disptime */ -}; - enum tg_state_flags { THROTL_TG_PENDING = 1 << 0, /* on parent's pending tree */ THROTL_TG_WAS_EMPTY = 1 << 1, /* bio_lists[] became non-empty */ @@ -98,93 +48,6 @@ enum tg_state_flags { #define rb_entry_tg(node) rb_entry((node), struct throtl_grp, rb_node) -enum { - LIMIT_LOW, - LIMIT_MAX, - LIMIT_CNT, -}; - -struct throtl_grp { - /* must be the first member */ - struct blkg_policy_data pd; - - /* active throtl group service_queue member */ - struct rb_node rb_node; - - /* throtl_data this group belongs to */ - struct throtl_data *td; - - /* this group's service queue */ - struct throtl_service_queue service_queue; - - /* - * qnode_on_self is used when bios are directly queued to this - * throtl_grp so that local bios compete fairly with bios - * dispatched from children. qnode_on_parent is used when bios are - * dispatched from this throtl_grp into its parent and will compete - * with the sibling qnode_on_parents and the parent's - * qnode_on_self. - */ - struct throtl_qnode qnode_on_self[2]; - struct throtl_qnode qnode_on_parent[2]; - - /* - * Dispatch time in jiffies. This is the estimated time when group - * will unthrottle and is ready to dispatch more bio. It is used as - * key to sort active groups in service tree. - */ - unsigned long disptime; - - unsigned int flags; - - /* are there any throtl rules between this group and td? */ - bool has_rules[2]; - - /* internally used bytes per second rate limits */ - uint64_t bps[2][LIMIT_CNT]; - /* user configured bps limits */ - uint64_t bps_conf[2][LIMIT_CNT]; - - /* internally used IOPS limits */ - unsigned int iops[2][LIMIT_CNT]; - /* user configured IOPS limits */ - unsigned int iops_conf[2][LIMIT_CNT]; - - /* Number of bytes dispatched in current slice */ - uint64_t bytes_disp[2]; - /* Number of bio's dispatched in current slice */ - unsigned int io_disp[2]; - - unsigned long last_low_overflow_time[2]; - - uint64_t last_bytes_disp[2]; - unsigned int last_io_disp[2]; - - unsigned long last_check_time; - - unsigned long latency_target; /* us */ - unsigned long latency_target_conf; /* us */ - /* When did we start a new slice */ - unsigned long slice_start[2]; - unsigned long slice_end[2]; - - unsigned long last_finish_time; /* ns / 1024 */ - unsigned long checked_last_finish_time; /* ns / 1024 */ - unsigned long avg_idletime; /* ns / 1024 */ - unsigned long idletime_threshold; /* us */ - unsigned long idletime_threshold_conf; /* us */ - - unsigned int bio_cnt; /* total bios */ - unsigned int bad_bio_cnt; /* bios exceeding latency threshold */ - unsigned long bio_cnt_reset_time; - - atomic_t io_split_cnt[2]; - atomic_t last_io_split_cnt[2]; - - struct blkg_rwstat stat_bytes; - struct blkg_rwstat stat_ios; -}; - /* We measure latency for request size from <= 4k to >= 1M */ #define LATENCY_BUCKET_SIZE 9 @@ -231,16 +94,6 @@ struct throtl_data static void throtl_pending_timer_fn(struct timer_list *t); -static inline struct throtl_grp *pd_to_tg(struct blkg_policy_data *pd) -{ - return pd ? container_of(pd, struct throtl_grp, pd) : NULL; -} - -static inline struct throtl_grp *blkg_to_tg(struct blkcg_gq *blkg) -{ - return pd_to_tg(blkg_to_pd(blkg, &blkcg_policy_throtl)); -} - static inline struct blkcg_gq *tg_to_blkg(struct throtl_grp *tg) { return pd_to_blkg(&tg->pd); @@ -1794,7 +1647,7 @@ static void throtl_shutdown_wq(struct request_queue *q) cancel_work_sync(&td->dispatch_work); } -static struct blkcg_policy blkcg_policy_throtl = { +struct blkcg_policy blkcg_policy_throtl = { .dfl_cftypes = throtl_files, .legacy_cftypes = throtl_legacy_files, @@ -2208,9 +2061,9 @@ void blk_throtl_charge_bio_split(struct bio *bio) } while (parent); } -bool blk_throtl_bio(struct bio *bio) +bool __blk_throtl_bio(struct bio *bio) { - struct request_queue *q = bio->bi_bdev->bd_disk->queue; + struct request_queue *q = bdev_get_queue(bio->bi_bdev); struct blkcg_gq *blkg = bio->bi_blkg; struct throtl_qnode *qn = NULL; struct throtl_grp *tg = blkg_to_tg(blkg); @@ -2221,19 +2074,12 @@ bool blk_throtl_bio(struct bio *bio) rcu_read_lock(); - /* see throtl_charge_bio() */ - if (bio_flagged(bio, BIO_THROTTLED)) - goto out; - if (!cgroup_subsys_on_dfl(io_cgrp_subsys)) { blkg_rwstat_add(&tg->stat_bytes, bio->bi_opf, bio->bi_iter.bi_size); blkg_rwstat_add(&tg->stat_ios, bio->bi_opf, 1); } - if (!tg->has_rules[rw]) - goto out; - spin_lock_irq(&q->queue_lock); throtl_update_latency_buckets(td); @@ -2317,7 +2163,6 @@ again: out_unlock: spin_unlock_irq(&q->queue_lock); -out: bio_set_flag(bio, BIO_THROTTLED); #ifdef CONFIG_BLK_DEV_THROTTLING_LOW diff --git a/block/blk-throttle.h b/block/blk-throttle.h new file mode 100644 index 000000000000..175f03abd9e4 --- /dev/null +++ b/block/blk-throttle.h @@ -0,0 +1,182 @@ +#ifndef BLK_THROTTLE_H +#define BLK_THROTTLE_H + +#include "blk-cgroup-rwstat.h" + +/* + * To implement hierarchical throttling, throtl_grps form a tree and bios + * are dispatched upwards level by level until they reach the top and get + * issued. When dispatching bios from the children and local group at each + * level, if the bios are dispatched into a single bio_list, there's a risk + * of a local or child group which can queue many bios at once filling up + * the list starving others. + * + * To avoid such starvation, dispatched bios are queued separately + * according to where they came from. When they are again dispatched to + * the parent, they're popped in round-robin order so that no single source + * hogs the dispatch window. + * + * throtl_qnode is used to keep the queued bios separated by their sources. + * Bios are queued to throtl_qnode which in turn is queued to + * throtl_service_queue and then dispatched in round-robin order. + * + * It's also used to track the reference counts on blkg's. A qnode always + * belongs to a throtl_grp and gets queued on itself or the parent, so + * incrementing the reference of the associated throtl_grp when a qnode is + * queued and decrementing when dequeued is enough to keep the whole blkg + * tree pinned while bios are in flight. + */ +struct throtl_qnode { + struct list_head node; /* service_queue->queued[] */ + struct bio_list bios; /* queued bios */ + struct throtl_grp *tg; /* tg this qnode belongs to */ +}; + +struct throtl_service_queue { + struct throtl_service_queue *parent_sq; /* the parent service_queue */ + + /* + * Bios queued directly to this service_queue or dispatched from + * children throtl_grp's. + */ + struct list_head queued[2]; /* throtl_qnode [READ/WRITE] */ + unsigned int nr_queued[2]; /* number of queued bios */ + + /* + * RB tree of active children throtl_grp's, which are sorted by + * their ->disptime. + */ + struct rb_root_cached pending_tree; /* RB tree of active tgs */ + unsigned int nr_pending; /* # queued in the tree */ + unsigned long first_pending_disptime; /* disptime of the first tg */ + struct timer_list pending_timer; /* fires on first_pending_disptime */ +}; + +enum { + LIMIT_LOW, + LIMIT_MAX, + LIMIT_CNT, +}; + +struct throtl_grp { + /* must be the first member */ + struct blkg_policy_data pd; + + /* active throtl group service_queue member */ + struct rb_node rb_node; + + /* throtl_data this group belongs to */ + struct throtl_data *td; + + /* this group's service queue */ + struct throtl_service_queue service_queue; + + /* + * qnode_on_self is used when bios are directly queued to this + * throtl_grp so that local bios compete fairly with bios + * dispatched from children. qnode_on_parent is used when bios are + * dispatched from this throtl_grp into its parent and will compete + * with the sibling qnode_on_parents and the parent's + * qnode_on_self. + */ + struct throtl_qnode qnode_on_self[2]; + struct throtl_qnode qnode_on_parent[2]; + + /* + * Dispatch time in jiffies. This is the estimated time when group + * will unthrottle and is ready to dispatch more bio. It is used as + * key to sort active groups in service tree. + */ + unsigned long disptime; + + unsigned int flags; + + /* are there any throtl rules between this group and td? */ + bool has_rules[2]; + + /* internally used bytes per second rate limits */ + uint64_t bps[2][LIMIT_CNT]; + /* user configured bps limits */ + uint64_t bps_conf[2][LIMIT_CNT]; + + /* internally used IOPS limits */ + unsigned int iops[2][LIMIT_CNT]; + /* user configured IOPS limits */ + unsigned int iops_conf[2][LIMIT_CNT]; + + /* Number of bytes dispatched in current slice */ + uint64_t bytes_disp[2]; + /* Number of bio's dispatched in current slice */ + unsigned int io_disp[2]; + + unsigned long last_low_overflow_time[2]; + + uint64_t last_bytes_disp[2]; + unsigned int last_io_disp[2]; + + unsigned long last_check_time; + + unsigned long latency_target; /* us */ + unsigned long latency_target_conf; /* us */ + /* When did we start a new slice */ + unsigned long slice_start[2]; + unsigned long slice_end[2]; + + unsigned long last_finish_time; /* ns / 1024 */ + unsigned long checked_last_finish_time; /* ns / 1024 */ + unsigned long avg_idletime; /* ns / 1024 */ + unsigned long idletime_threshold; /* us */ + unsigned long idletime_threshold_conf; /* us */ + + unsigned int bio_cnt; /* total bios */ + unsigned int bad_bio_cnt; /* bios exceeding latency threshold */ + unsigned long bio_cnt_reset_time; + + atomic_t io_split_cnt[2]; + atomic_t last_io_split_cnt[2]; + + struct blkg_rwstat stat_bytes; + struct blkg_rwstat stat_ios; +}; + +extern struct blkcg_policy blkcg_policy_throtl; + +static inline struct throtl_grp *pd_to_tg(struct blkg_policy_data *pd) +{ + return pd ? container_of(pd, struct throtl_grp, pd) : NULL; +} + +static inline struct throtl_grp *blkg_to_tg(struct blkcg_gq *blkg) +{ + return pd_to_tg(blkg_to_pd(blkg, &blkcg_policy_throtl)); +} + +/* + * Internal throttling interface + */ +#ifndef CONFIG_BLK_DEV_THROTTLING +static inline int blk_throtl_init(struct request_queue *q) { return 0; } +static inline void blk_throtl_exit(struct request_queue *q) { } +static inline void blk_throtl_register_queue(struct request_queue *q) { } +static inline void blk_throtl_charge_bio_split(struct bio *bio) { } +static inline bool blk_throtl_bio(struct bio *bio) { return false; } +#else /* CONFIG_BLK_DEV_THROTTLING */ +int blk_throtl_init(struct request_queue *q); +void blk_throtl_exit(struct request_queue *q); +void blk_throtl_register_queue(struct request_queue *q); +void blk_throtl_charge_bio_split(struct bio *bio); +bool __blk_throtl_bio(struct bio *bio); +static inline bool blk_throtl_bio(struct bio *bio) +{ + struct throtl_grp *tg = blkg_to_tg(bio->bi_blkg); + + if (bio_flagged(bio, BIO_THROTTLED)) + return false; + if (!tg->has_rules[bio_data_dir(bio)]) + return false; + + return __blk_throtl_bio(bio); +} +#endif /* CONFIG_BLK_DEV_THROTTLING */ + +#endif diff --git a/block/blk-wbt.c b/block/blk-wbt.c index 874c1c37bf0c..0c119be0e813 100644 --- a/block/blk-wbt.c +++ b/block/blk-wbt.c @@ -357,6 +357,9 @@ static void wb_timer_fn(struct blk_stat_callback *cb) unsigned int inflight = wbt_inflight(rwb); int status; + if (!rwb->rqos.q->disk) + return; + status = latency_exceeded(rwb, cb->stat); trace_wbt_timer(rwb->rqos.q->disk->bdi, status, rqd->scale_step, diff --git a/block/blk.h b/block/blk.h index 6c3c00a8fe19..7afffd548daf 100644 --- a/block/blk.h +++ b/block/blk.h @@ -12,6 +12,8 @@ #include "blk-mq.h" #include "blk-mq-sched.h" +struct elevator_type; + /* Max future timer expiry for timeouts */ #define BLK_MAX_TIMEOUT (5 * HZ) @@ -94,6 +96,44 @@ static inline bool bvec_gap_to_prev(struct request_queue *q, return __bvec_gap_to_prev(q, bprv, offset); } +static inline bool rq_mergeable(struct request *rq) +{ + if (blk_rq_is_passthrough(rq)) + return false; + + if (req_op(rq) == REQ_OP_FLUSH) + return false; + + if (req_op(rq) == REQ_OP_WRITE_ZEROES) + return false; + + if (req_op(rq) == REQ_OP_ZONE_APPEND) + return false; + + if (rq->cmd_flags & REQ_NOMERGE_FLAGS) + return false; + if (rq->rq_flags & RQF_NOMERGE_FLAGS) + return false; + + return true; +} + +/* + * There are two different ways to handle DISCARD merges: + * 1) If max_discard_segments > 1, the driver treats every bio as a range and + * send the bios to controller together. The ranges don't need to be + * contiguous. + * 2) Otherwise, the request will be normal read/write requests. The ranges + * need to be contiguous. + */ +static inline bool blk_discard_mergable(struct request *req) +{ + if (req_op(req) == REQ_OP_DISCARD && + queue_max_discard_segments(req->q) > 1) + return true; + return false; +} + #ifdef CONFIG_BLK_DEV_INTEGRITY void blk_flush_integrity(void); bool __bio_integrity_endio(struct bio *); @@ -175,21 +215,28 @@ static inline void blk_integrity_del(struct gendisk *disk) unsigned long blk_rq_timeout(unsigned long timeout); void blk_add_timer(struct request *req); +void blk_print_req_error(struct request *req, blk_status_t status); bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio, - unsigned int nr_segs, struct request **same_queue_rq); + unsigned int nr_segs, bool *same_queue_rq); bool blk_bio_list_merge(struct request_queue *q, struct list_head *list, struct bio *bio, unsigned int nr_segs); -void blk_account_io_start(struct request *req); -void blk_account_io_done(struct request *req, u64 now); +void __blk_account_io_start(struct request *req); +void __blk_account_io_done(struct request *req, u64 now); + +/* + * Plug flush limits + */ +#define BLK_MAX_REQUEST_COUNT 32 +#define BLK_PLUG_FLUSH_SIZE (128 * 1024) /* * Internal elevator interface */ #define ELV_ON_HASH(rq) ((rq)->rq_flags & RQF_HASHED) -void blk_insert_flush(struct request *rq); +bool blk_insert_flush(struct request *rq); int elevator_switch_mq(struct request_queue *q, struct elevator_type *new_e); @@ -202,7 +249,7 @@ static inline void elevator_exit(struct request_queue *q, { lockdep_assert_held(&q->sysfs_lock); - blk_mq_sched_free_requests(q); + blk_mq_sched_free_rqs(q); __elevator_exit(q, e); } @@ -220,7 +267,32 @@ ssize_t part_timeout_show(struct device *, struct device_attribute *, char *); ssize_t part_timeout_store(struct device *, struct device_attribute *, const char *, size_t); -void __blk_queue_split(struct bio **bio, unsigned int *nr_segs); +static inline bool blk_may_split(struct request_queue *q, struct bio *bio) +{ + switch (bio_op(bio)) { + case REQ_OP_DISCARD: + case REQ_OP_SECURE_ERASE: + case REQ_OP_WRITE_ZEROES: + case REQ_OP_WRITE_SAME: + return true; /* non-trivial splitting decisions */ + default: + break; + } + + /* + * All drivers must accept single-segments bios that are <= PAGE_SIZE. + * This is a quick and dirty check that relies on the fact that + * bi_io_vec[0] is always valid if a bio has data. The check might + * lead to occasional false negatives when bios are cloned, but compared + * to the performance impact of cloned bios themselves the loop below + * doesn't matter anyway. + */ + return q->limits.chunk_sectors || bio->bi_vcnt != 1 || + bio->bi_io_vec->bv_len + bio->bi_io_vec->bv_offset > PAGE_SIZE; +} + +void __blk_queue_split(struct request_queue *q, struct bio **bio, + unsigned int *nr_segs); int ll_back_merge_fn(struct request *req, struct bio *bio, unsigned int nr_segs); bool blk_attempt_req_merge(struct request_queue *q, struct request *rq, @@ -240,7 +312,25 @@ int blk_dev_init(void); */ static inline bool blk_do_io_stat(struct request *rq) { - return rq->rq_disk && (rq->rq_flags & RQF_IO_STAT); + return (rq->rq_flags & RQF_IO_STAT) && rq->rq_disk; +} + +static inline void blk_account_io_done(struct request *req, u64 now) +{ + /* + * Account IO completion. flush_rq isn't accounted as a + * normal IO on queueing nor completion. Accounting the + * containing request is enough. + */ + if (blk_do_io_stat(req) && req->part && + !(req->rq_flags & RQF_FLUSH_SEQ)) + __blk_account_io_done(req, now); +} + +static inline void blk_account_io_start(struct request *req) +{ + if (blk_do_io_stat(req)) + __blk_account_io_start(req); } static inline void req_set_nomerge(struct request_queue *q, struct request *req) @@ -285,22 +375,6 @@ void ioc_clear_queue(struct request_queue *q); int create_task_io_context(struct task_struct *task, gfp_t gfp_mask, int node); -/* - * Internal throttling interface - */ -#ifdef CONFIG_BLK_DEV_THROTTLING -extern int blk_throtl_init(struct request_queue *q); -extern void blk_throtl_exit(struct request_queue *q); -extern void blk_throtl_register_queue(struct request_queue *q); -extern void blk_throtl_charge_bio_split(struct bio *bio); -bool blk_throtl_bio(struct bio *bio); -#else /* CONFIG_BLK_DEV_THROTTLING */ -static inline int blk_throtl_init(struct request_queue *q) { return 0; } -static inline void blk_throtl_exit(struct request_queue *q) { } -static inline void blk_throtl_register_queue(struct request_queue *q) { } -static inline void blk_throtl_charge_bio_split(struct bio *bio) { } -static inline bool blk_throtl_bio(struct bio *bio) { return false; } -#endif /* CONFIG_BLK_DEV_THROTTLING */ #ifdef CONFIG_BLK_DEV_THROTTLING_LOW extern ssize_t blk_throtl_sample_time_show(struct request_queue *q, char *page); extern ssize_t blk_throtl_sample_time_store(struct request_queue *q, @@ -368,13 +442,20 @@ extern struct device_attribute dev_attr_events; extern struct device_attribute dev_attr_events_async; extern struct device_attribute dev_attr_events_poll_msecs; -static inline void bio_clear_hipri(struct bio *bio) +static inline void bio_clear_polled(struct bio *bio) { /* can't support alloc cache if we turn off polling */ bio_clear_flag(bio, BIO_PERCPU_CACHE); - bio->bi_opf &= ~REQ_HIPRI; + bio->bi_opf &= ~REQ_POLLED; } +long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg); +long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg); + extern const struct address_space_operations def_blk_aops; +int disk_register_independent_access_ranges(struct gendisk *disk, + struct blk_independent_access_ranges *new_iars); +void disk_unregister_independent_access_ranges(struct gendisk *disk); + #endif /* BLK_INTERNAL_H */ diff --git a/block/bounce.c b/block/bounce.c index 05fc7148489d..7af1a72835b9 100644 --- a/block/bounce.c +++ b/block/bounce.c @@ -14,6 +14,7 @@ #include <linux/pagemap.h> #include <linux/mempool.h> #include <linux/blkdev.h> +#include <linux/blk-cgroup.h> #include <linux/backing-dev.h> #include <linux/init.h> #include <linux/hash.h> diff --git a/block/elevator.c b/block/elevator.c index ff45d8388f48..1f39f6e8ebb9 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -26,7 +26,6 @@ #include <linux/kernel.h> #include <linux/fs.h> #include <linux/blkdev.h> -#include <linux/elevator.h> #include <linux/bio.h> #include <linux/module.h> #include <linux/slab.h> @@ -40,6 +39,7 @@ #include <trace/events/block.h> +#include "elevator.h" #include "blk.h" #include "blk-mq-sched.h" #include "blk-pm.h" @@ -637,7 +637,7 @@ static struct elevator_type *elevator_get_default(struct request_queue *q) return NULL; if (q->nr_hw_queues != 1 && - !blk_mq_is_sbitmap_shared(q->tag_set->flags)) + !blk_mq_is_shared_tags(q->tag_set->flags)) return NULL; return elevator_get(q, "mq-deadline", false); diff --git a/block/elevator.h b/block/elevator.h new file mode 100644 index 000000000000..16cd8bdedb7e --- /dev/null +++ b/block/elevator.h @@ -0,0 +1,166 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ELEVATOR_H +#define _ELEVATOR_H + +#include <linux/percpu.h> +#include <linux/hashtable.h> + +struct io_cq; +struct elevator_type; +struct blk_mq_debugfs_attr; + +/* + * Return values from elevator merger + */ +enum elv_merge { + ELEVATOR_NO_MERGE = 0, + ELEVATOR_FRONT_MERGE = 1, + ELEVATOR_BACK_MERGE = 2, + ELEVATOR_DISCARD_MERGE = 3, +}; + +struct blk_mq_alloc_data; +struct blk_mq_hw_ctx; + +struct elevator_mq_ops { + int (*init_sched)(struct request_queue *, struct elevator_type *); + void (*exit_sched)(struct elevator_queue *); + int (*init_hctx)(struct blk_mq_hw_ctx *, unsigned int); + void (*exit_hctx)(struct blk_mq_hw_ctx *, unsigned int); + void (*depth_updated)(struct blk_mq_hw_ctx *); + + bool (*allow_merge)(struct request_queue *, struct request *, struct bio *); + bool (*bio_merge)(struct request_queue *, struct bio *, unsigned int); + int (*request_merge)(struct request_queue *q, struct request **, struct bio *); + void (*request_merged)(struct request_queue *, struct request *, enum elv_merge); + void (*requests_merged)(struct request_queue *, struct request *, struct request *); + void (*limit_depth)(unsigned int, struct blk_mq_alloc_data *); + void (*prepare_request)(struct request *); + void (*finish_request)(struct request *); + void (*insert_requests)(struct blk_mq_hw_ctx *, struct list_head *, bool); + struct request *(*dispatch_request)(struct blk_mq_hw_ctx *); + bool (*has_work)(struct blk_mq_hw_ctx *); + void (*completed_request)(struct request *, u64); + void (*requeue_request)(struct request *); + struct request *(*former_request)(struct request_queue *, struct request *); + struct request *(*next_request)(struct request_queue *, struct request *); + void (*init_icq)(struct io_cq *); + void (*exit_icq)(struct io_cq *); +}; + +#define ELV_NAME_MAX (16) + +struct elv_fs_entry { + struct attribute attr; + ssize_t (*show)(struct elevator_queue *, char *); + ssize_t (*store)(struct elevator_queue *, const char *, size_t); +}; + +/* + * identifies an elevator type, such as AS or deadline + */ +struct elevator_type +{ + /* managed by elevator core */ + struct kmem_cache *icq_cache; + + /* fields provided by elevator implementation */ + struct elevator_mq_ops ops; + + size_t icq_size; /* see iocontext.h */ + size_t icq_align; /* ditto */ + struct elv_fs_entry *elevator_attrs; + const char *elevator_name; + const char *elevator_alias; + const unsigned int elevator_features; + struct module *elevator_owner; +#ifdef CONFIG_BLK_DEBUG_FS + const struct blk_mq_debugfs_attr *queue_debugfs_attrs; + const struct blk_mq_debugfs_attr *hctx_debugfs_attrs; +#endif + + /* managed by elevator core */ + char icq_cache_name[ELV_NAME_MAX + 6]; /* elvname + "_io_cq" */ + struct list_head list; +}; + +#define ELV_HASH_BITS 6 + +void elv_rqhash_del(struct request_queue *q, struct request *rq); +void elv_rqhash_add(struct request_queue *q, struct request *rq); +void elv_rqhash_reposition(struct request_queue *q, struct request *rq); +struct request *elv_rqhash_find(struct request_queue *q, sector_t offset); + +/* + * each queue has an elevator_queue associated with it + */ +struct elevator_queue +{ + struct elevator_type *type; + void *elevator_data; + struct kobject kobj; + struct mutex sysfs_lock; + unsigned int registered:1; + DECLARE_HASHTABLE(hash, ELV_HASH_BITS); +}; + +/* + * block elevator interface + */ +extern enum elv_merge elv_merge(struct request_queue *, struct request **, + struct bio *); +extern void elv_merge_requests(struct request_queue *, struct request *, + struct request *); +extern void elv_merged_request(struct request_queue *, struct request *, + enum elv_merge); +extern bool elv_attempt_insert_merge(struct request_queue *, struct request *, + struct list_head *); +extern struct request *elv_former_request(struct request_queue *, struct request *); +extern struct request *elv_latter_request(struct request_queue *, struct request *); +void elevator_init_mq(struct request_queue *q); + +/* + * io scheduler registration + */ +extern int elv_register(struct elevator_type *); +extern void elv_unregister(struct elevator_type *); + +/* + * io scheduler sysfs switching + */ +extern ssize_t elv_iosched_show(struct request_queue *, char *); +extern ssize_t elv_iosched_store(struct request_queue *, const char *, size_t); + +extern bool elv_bio_merge_ok(struct request *, struct bio *); +extern struct elevator_queue *elevator_alloc(struct request_queue *, + struct elevator_type *); + +/* + * Helper functions. + */ +extern struct request *elv_rb_former_request(struct request_queue *, struct request *); +extern struct request *elv_rb_latter_request(struct request_queue *, struct request *); + +/* + * rb support functions. + */ +extern void elv_rb_add(struct rb_root *, struct request *); +extern void elv_rb_del(struct rb_root *, struct request *); +extern struct request *elv_rb_find(struct rb_root *, sector_t); + +/* + * Insertion selection + */ +#define ELEVATOR_INSERT_FRONT 1 +#define ELEVATOR_INSERT_BACK 2 +#define ELEVATOR_INSERT_SORT 3 +#define ELEVATOR_INSERT_REQUEUE 4 +#define ELEVATOR_INSERT_FLUSH 5 +#define ELEVATOR_INSERT_SORT_MERGE 6 + +#define rb_entry_rq(node) rb_entry((node), struct request, rb_node) + +#define rq_entry_fifo(ptr) list_entry((ptr), struct request, queuelist) +#define rq_fifo_clear(rq) list_del_init(&(rq)->queuelist) + +#endif /* _ELEVATOR_H */ diff --git a/block/fops.c b/block/fops.c index 1e970c247e0e..a2f492e50782 100644 --- a/block/fops.c +++ b/block/fops.c @@ -17,7 +17,7 @@ #include <linux/fs.h> #include "blk.h" -static struct inode *bdev_file_inode(struct file *file) +static inline struct inode *bdev_file_inode(struct file *file) { return file->f_mapping->host; } @@ -54,14 +54,12 @@ static void blkdev_bio_end_io_simple(struct bio *bio) static ssize_t __blkdev_direct_IO_simple(struct kiocb *iocb, struct iov_iter *iter, unsigned int nr_pages) { - struct file *file = iocb->ki_filp; - struct block_device *bdev = I_BDEV(bdev_file_inode(file)); + struct block_device *bdev = iocb->ki_filp->private_data; struct bio_vec inline_vecs[DIO_INLINE_BIO_VECS], *vecs; loff_t pos = iocb->ki_pos; bool should_dirty = false; struct bio bio; ssize_t ret; - blk_qc_t qc; if ((pos | iov_iter_alignment(iter)) & (bdev_logical_block_size(bdev) - 1)) @@ -78,7 +76,7 @@ static ssize_t __blkdev_direct_IO_simple(struct kiocb *iocb, bio_init(&bio, vecs, nr_pages); bio_set_dev(&bio, bdev); - bio.bi_iter.bi_sector = pos >> 9; + bio.bi_iter.bi_sector = pos >> SECTOR_SHIFT; bio.bi_write_hint = iocb->ki_hint; bio.bi_private = current; bio.bi_end_io = blkdev_bio_end_io_simple; @@ -102,13 +100,12 @@ static ssize_t __blkdev_direct_IO_simple(struct kiocb *iocb, if (iocb->ki_flags & IOCB_HIPRI) bio_set_polled(&bio, iocb); - qc = submit_bio(&bio); + submit_bio(&bio); for (;;) { set_current_state(TASK_UNINTERRUPTIBLE); if (!READ_ONCE(bio.bi_private)) break; - if (!(iocb->ki_flags & IOCB_HIPRI) || - !blk_poll(bdev_get_queue(bdev), qc, true)) + if (!(iocb->ki_flags & IOCB_HIPRI) || !bio_poll(&bio, NULL, 0)) blk_io_schedule(); } __set_current_state(TASK_RUNNING); @@ -126,6 +123,11 @@ out: return ret; } +enum { + DIO_SHOULD_DIRTY = 1, + DIO_IS_SYNC = 2, +}; + struct blkdev_dio { union { struct kiocb *iocb; @@ -133,35 +135,27 @@ struct blkdev_dio { }; size_t size; atomic_t ref; - bool multi_bio : 1; - bool should_dirty : 1; - bool is_sync : 1; - struct bio bio; + unsigned int flags; + struct bio bio ____cacheline_aligned_in_smp; }; static struct bio_set blkdev_dio_pool; -static int blkdev_iopoll(struct kiocb *kiocb, bool wait) -{ - struct block_device *bdev = I_BDEV(kiocb->ki_filp->f_mapping->host); - struct request_queue *q = bdev_get_queue(bdev); - - return blk_poll(q, READ_ONCE(kiocb->ki_cookie), wait); -} - static void blkdev_bio_end_io(struct bio *bio) { struct blkdev_dio *dio = bio->bi_private; - bool should_dirty = dio->should_dirty; + bool should_dirty = dio->flags & DIO_SHOULD_DIRTY; if (bio->bi_status && !dio->bio.bi_status) dio->bio.bi_status = bio->bi_status; - if (!dio->multi_bio || atomic_dec_and_test(&dio->ref)) { - if (!dio->is_sync) { + if (atomic_dec_and_test(&dio->ref)) { + if (!(dio->flags & DIO_IS_SYNC)) { struct kiocb *iocb = dio->iocb; ssize_t ret; + WRITE_ONCE(iocb->private, NULL); + if (likely(!dio->bio.bi_status)) { ret = dio->size; iocb->ki_pos += ret; @@ -170,8 +164,7 @@ static void blkdev_bio_end_io(struct bio *bio) } dio->iocb->ki_complete(iocb, ret, 0); - if (dio->multi_bio) - bio_put(&dio->bio); + bio_put(&dio->bio); } else { struct task_struct *waiter = dio->waiter; @@ -191,16 +184,12 @@ static void blkdev_bio_end_io(struct bio *bio) static ssize_t __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, unsigned int nr_pages) { - struct file *file = iocb->ki_filp; - struct inode *inode = bdev_file_inode(file); - struct block_device *bdev = I_BDEV(inode); + struct block_device *bdev = iocb->ki_filp->private_data; struct blk_plug plug; struct blkdev_dio *dio; struct bio *bio; - bool is_poll = (iocb->ki_flags & IOCB_HIPRI) != 0; bool is_read = (iov_iter_rw(iter) == READ), is_sync; loff_t pos = iocb->ki_pos; - blk_qc_t qc = BLK_QC_T_NONE; int ret = 0; if ((pos | iov_iter_alignment(iter)) & @@ -210,28 +199,31 @@ static ssize_t __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, bio = bio_alloc_kiocb(iocb, nr_pages, &blkdev_dio_pool); dio = container_of(bio, struct blkdev_dio, bio); - dio->is_sync = is_sync = is_sync_kiocb(iocb); - if (dio->is_sync) { + atomic_set(&dio->ref, 1); + /* + * Grab an extra reference to ensure the dio structure which is embedded + * into the first bio stays around. + */ + bio_get(bio); + + is_sync = is_sync_kiocb(iocb); + if (is_sync) { + dio->flags = DIO_IS_SYNC; dio->waiter = current; - bio_get(bio); } else { + dio->flags = 0; dio->iocb = iocb; } dio->size = 0; - dio->multi_bio = false; - dio->should_dirty = is_read && iter_is_iovec(iter); + if (is_read && iter_is_iovec(iter)) + dio->flags |= DIO_SHOULD_DIRTY; - /* - * Don't plug for HIPRI/polled IO, as those should go straight - * to issue - */ - if (!is_poll) - blk_start_plug(&plug); + blk_start_plug(&plug); for (;;) { bio_set_dev(bio, bdev); - bio->bi_iter.bi_sector = pos >> 9; + bio->bi_iter.bi_sector = pos >> SECTOR_SHIFT; bio->bi_write_hint = iocb->ki_hint; bio->bi_private = dio; bio->bi_end_io = blkdev_bio_end_io; @@ -246,7 +238,7 @@ static ssize_t __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, if (is_read) { bio->bi_opf = REQ_OP_READ; - if (dio->should_dirty) + if (dio->flags & DIO_SHOULD_DIRTY) bio_set_pages_dirty(bio); } else { bio->bi_opf = dio_bio_write_op(iocb); @@ -260,40 +252,15 @@ static ssize_t __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, nr_pages = bio_iov_vecs_to_alloc(iter, BIO_MAX_VECS); if (!nr_pages) { - bool polled = false; - - if (iocb->ki_flags & IOCB_HIPRI) { - bio_set_polled(bio, iocb); - polled = true; - } - - qc = submit_bio(bio); - - if (polled) - WRITE_ONCE(iocb->ki_cookie, qc); + submit_bio(bio); break; } - - if (!dio->multi_bio) { - /* - * AIO needs an extra reference to ensure the dio - * structure which is embedded into the first bio - * stays around. - */ - if (!is_sync) - bio_get(bio); - dio->multi_bio = true; - atomic_set(&dio->ref, 2); - } else { - atomic_inc(&dio->ref); - } - + atomic_inc(&dio->ref); submit_bio(bio); bio = bio_alloc(GFP_KERNEL, nr_pages); } - if (!is_poll) - blk_finish_plug(&plug); + blk_finish_plug(&plug); if (!is_sync) return -EIOCBQUEUED; @@ -302,10 +269,7 @@ static ssize_t __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, set_current_state(TASK_UNINTERRUPTIBLE); if (!READ_ONCE(dio->waiter)) break; - - if (!(iocb->ki_flags & IOCB_HIPRI) || - !blk_poll(bdev_get_queue(bdev), qc, true)) - blk_io_schedule(); + blk_io_schedule(); } __set_current_state(TASK_RUNNING); @@ -318,6 +282,94 @@ static ssize_t __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, return ret; } +static void blkdev_bio_end_io_async(struct bio *bio) +{ + struct blkdev_dio *dio = container_of(bio, struct blkdev_dio, bio); + struct kiocb *iocb = dio->iocb; + ssize_t ret; + + if (likely(!bio->bi_status)) { + ret = dio->size; + iocb->ki_pos += ret; + } else { + ret = blk_status_to_errno(bio->bi_status); + } + + iocb->ki_complete(iocb, ret, 0); + + if (dio->flags & DIO_SHOULD_DIRTY) { + bio_check_pages_dirty(bio); + } else { + bio_release_pages(bio, false); + bio_put(bio); + } +} + +static ssize_t __blkdev_direct_IO_async(struct kiocb *iocb, + struct iov_iter *iter, + unsigned int nr_pages) +{ + struct block_device *bdev = iocb->ki_filp->private_data; + struct blkdev_dio *dio; + struct bio *bio; + loff_t pos = iocb->ki_pos; + int ret = 0; + + if ((pos | iov_iter_alignment(iter)) & + (bdev_logical_block_size(bdev) - 1)) + return -EINVAL; + + bio = bio_alloc_kiocb(iocb, nr_pages, &blkdev_dio_pool); + dio = container_of(bio, struct blkdev_dio, bio); + dio->flags = 0; + dio->iocb = iocb; + bio_set_dev(bio, bdev); + bio->bi_iter.bi_sector = pos >> SECTOR_SHIFT; + bio->bi_write_hint = iocb->ki_hint; + bio->bi_end_io = blkdev_bio_end_io_async; + bio->bi_ioprio = iocb->ki_ioprio; + + if (iov_iter_is_bvec(iter)) { + /* + * Users don't rely on the iterator being in any particular + * state for async I/O returning -EIOCBQUEUED, hence we can + * avoid expensive iov_iter_advance(). Bypass + * bio_iov_iter_get_pages() and set the bvec directly. + */ + bio_iov_bvec_set(bio, iter); + } else { + ret = bio_iov_iter_get_pages(bio, iter); + if (unlikely(ret)) { + bio->bi_status = BLK_STS_IOERR; + bio_endio(bio); + return ret; + } + } + dio->size = bio->bi_iter.bi_size; + + if (iov_iter_rw(iter) == READ) { + bio->bi_opf = REQ_OP_READ; + if (iter_is_iovec(iter)) { + dio->flags |= DIO_SHOULD_DIRTY; + bio_set_pages_dirty(bio); + } + } else { + bio->bi_opf = dio_bio_write_op(iocb); + task_io_account_write(bio->bi_iter.bi_size); + } + + if (iocb->ki_flags & IOCB_HIPRI) { + bio->bi_opf |= REQ_POLLED | REQ_NOWAIT; + submit_bio(bio); + WRITE_ONCE(iocb->private, bio); + } else { + if (iocb->ki_flags & IOCB_NOWAIT) + bio->bi_opf |= REQ_NOWAIT; + submit_bio(bio); + } + return -EIOCBQUEUED; +} + static ssize_t blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter) { unsigned int nr_pages; @@ -326,9 +378,11 @@ static ssize_t blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter) return 0; nr_pages = bio_iov_vecs_to_alloc(iter, BIO_MAX_VECS + 1); - if (is_sync_kiocb(iocb) && nr_pages <= BIO_MAX_VECS) - return __blkdev_direct_IO_simple(iocb, iter, nr_pages); - + if (likely(nr_pages <= BIO_MAX_VECS)) { + if (is_sync_kiocb(iocb)) + return __blkdev_direct_IO_simple(iocb, iter, nr_pages); + return __blkdev_direct_IO_async(iocb, iter, nr_pages); + } return __blkdev_direct_IO(iocb, iter, bio_max_segs(nr_pages)); } @@ -405,8 +459,7 @@ static loff_t blkdev_llseek(struct file *file, loff_t offset, int whence) static int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync) { - struct inode *bd_inode = bdev_file_inode(filp); - struct block_device *bdev = I_BDEV(bd_inode); + struct block_device *bdev = filp->private_data; int error; error = file_write_and_wait_range(filp, start, end); @@ -448,6 +501,8 @@ static int blkdev_open(struct inode *inode, struct file *filp) bdev = blkdev_get_by_dev(inode->i_rdev, filp->f_mode, filp); if (IS_ERR(bdev)) return PTR_ERR(bdev); + + filp->private_data = bdev; filp->f_mapping = bdev->bd_inode->i_mapping; filp->f_wb_err = filemap_sample_wb_err(filp->f_mapping); return 0; @@ -455,29 +510,12 @@ static int blkdev_open(struct inode *inode, struct file *filp) static int blkdev_close(struct inode *inode, struct file *filp) { - struct block_device *bdev = I_BDEV(bdev_file_inode(filp)); + struct block_device *bdev = filp->private_data; blkdev_put(bdev, filp->f_mode); return 0; } -static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) -{ - struct block_device *bdev = I_BDEV(bdev_file_inode(file)); - fmode_t mode = file->f_mode; - - /* - * O_NDELAY can be altered using fcntl(.., F_SETFL, ..), so we have - * to updated it before every ioctl. - */ - if (file->f_flags & O_NDELAY) - mode |= FMODE_NDELAY; - else - mode &= ~FMODE_NDELAY; - - return blkdev_ioctl(bdev, mode, cmd, arg); -} - /* * Write data to the block device. Only intended for the block device itself * and the raw driver which basically is a fake block device. @@ -487,14 +525,14 @@ static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) */ static ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from) { - struct file *file = iocb->ki_filp; - struct inode *bd_inode = bdev_file_inode(file); + struct block_device *bdev = iocb->ki_filp->private_data; + struct inode *bd_inode = bdev->bd_inode; loff_t size = i_size_read(bd_inode); struct blk_plug plug; size_t shorted = 0; ssize_t ret; - if (bdev_read_only(I_BDEV(bd_inode))) + if (bdev_read_only(bdev)) return -EPERM; if (IS_SWAPFILE(bd_inode) && !is_hibernate_resume_dev(bd_inode->i_rdev)) @@ -526,24 +564,26 @@ static ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from) static ssize_t blkdev_read_iter(struct kiocb *iocb, struct iov_iter *to) { - struct file *file = iocb->ki_filp; - struct inode *bd_inode = bdev_file_inode(file); - loff_t size = i_size_read(bd_inode); + struct block_device *bdev = iocb->ki_filp->private_data; + loff_t size = i_size_read(bdev->bd_inode); loff_t pos = iocb->ki_pos; size_t shorted = 0; ssize_t ret; - if (pos >= size) - return 0; - - size -= pos; - if (iov_iter_count(to) > size) { - shorted = iov_iter_count(to) - size; - iov_iter_truncate(to, size); + if (unlikely(pos + iov_iter_count(to) > size)) { + if (pos >= size) + return 0; + size -= pos; + if (iov_iter_count(to) > size) { + shorted = iov_iter_count(to) - size; + iov_iter_truncate(to, size); + } } ret = generic_file_read_iter(iocb, to); - iov_iter_reexpand(to, iov_iter_count(to) + shorted); + + if (unlikely(shorted)) + iov_iter_reexpand(to, iov_iter_count(to) + shorted); return ret; } @@ -592,16 +632,18 @@ static long blkdev_fallocate(struct file *file, int mode, loff_t start, switch (mode) { case FALLOC_FL_ZERO_RANGE: case FALLOC_FL_ZERO_RANGE | FALLOC_FL_KEEP_SIZE: - error = blkdev_issue_zeroout(bdev, start >> 9, len >> 9, - GFP_KERNEL, BLKDEV_ZERO_NOUNMAP); + error = blkdev_issue_zeroout(bdev, start >> SECTOR_SHIFT, + len >> SECTOR_SHIFT, GFP_KERNEL, + BLKDEV_ZERO_NOUNMAP); break; case FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE: - error = blkdev_issue_zeroout(bdev, start >> 9, len >> 9, - GFP_KERNEL, BLKDEV_ZERO_NOFALLBACK); + error = blkdev_issue_zeroout(bdev, start >> SECTOR_SHIFT, + len >> SECTOR_SHIFT, GFP_KERNEL, + BLKDEV_ZERO_NOFALLBACK); break; case FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE | FALLOC_FL_NO_HIDE_STALE: - error = blkdev_issue_discard(bdev, start >> 9, len >> 9, - GFP_KERNEL, 0); + error = blkdev_issue_discard(bdev, start >> SECTOR_SHIFT, + len >> SECTOR_SHIFT, GFP_KERNEL, 0); break; default: error = -EOPNOTSUPP; @@ -618,10 +660,10 @@ const struct file_operations def_blk_fops = { .llseek = blkdev_llseek, .read_iter = blkdev_read_iter, .write_iter = blkdev_write_iter, - .iopoll = blkdev_iopoll, + .iopoll = iocb_bio_iopoll, .mmap = generic_file_mmap, .fsync = blkdev_fsync, - .unlocked_ioctl = block_ioctl, + .unlocked_ioctl = blkdev_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = compat_blkdev_ioctl, #endif diff --git a/block/genhd.c b/block/genhd.c index ab12ae6e636e..2052aeffa39b 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -19,6 +19,7 @@ #include <linux/seq_file.h> #include <linux/slab.h> #include <linux/kmod.h> +#include <linux/major.h> #include <linux/mutex.h> #include <linux/idr.h> #include <linux/log2.h> @@ -625,6 +626,26 @@ void del_gendisk(struct gendisk *disk) } EXPORT_SYMBOL(del_gendisk); +/** + * invalidate_disk - invalidate the disk + * @disk: the struct gendisk to invalidate + * + * A helper to invalidates the disk. It will clean the disk's associated + * buffer/page caches and reset its internal states so that the disk + * can be reused by the drivers. + * + * Context: can sleep + */ +void invalidate_disk(struct gendisk *disk) +{ + struct block_device *bdev = disk->part0; + + invalidate_bdev(bdev); + bdev->bd_inode->i_mapping->wb_err = 0; + set_capacity(disk, 0); +} +EXPORT_SYMBOL(invalidate_disk); + /* sysfs access to bad-blocks list. */ static ssize_t disk_badblocks_show(struct device *dev, struct device_attribute *attr, @@ -884,7 +905,7 @@ ssize_t part_stat_show(struct device *dev, struct device_attribute *attr, char *buf) { struct block_device *bdev = dev_to_bdev(dev); - struct request_queue *q = bdev->bd_disk->queue; + struct request_queue *q = bdev_get_queue(bdev); struct disk_stats stat; unsigned int inflight; @@ -928,7 +949,7 @@ ssize_t part_inflight_show(struct device *dev, struct device_attribute *attr, char *buf) { struct block_device *bdev = dev_to_bdev(dev); - struct request_queue *q = bdev->bd_disk->queue; + struct request_queue *q = bdev_get_queue(bdev); unsigned int inflight[2]; if (queue_is_mq(q)) @@ -1268,6 +1289,9 @@ struct gendisk *__alloc_disk_node(struct request_queue *q, int node_id, if (!disk->bdi) goto out_free_disk; + /* bdev_alloc() might need the queue, set before the first call */ + disk->queue = q; + disk->part0 = bdev_alloc(disk, 0); if (!disk->part0) goto out_free_bdi; @@ -1283,7 +1307,6 @@ struct gendisk *__alloc_disk_node(struct request_queue *q, int node_id, disk_to_dev(disk)->type = &disk_type; device_initialize(disk_to_dev(disk)); inc_diskseq(disk); - disk->queue = q; q->disk = disk; lockdep_init_map(&disk->lockdep_map, "(bio completion)", lkclass, 0); #ifdef CONFIG_BLOCK_HOLDER_DEPRECATED @@ -1388,12 +1411,6 @@ void set_disk_ro(struct gendisk *disk, bool read_only) } EXPORT_SYMBOL(set_disk_ro); -int bdev_read_only(struct block_device *bdev) -{ - return bdev->bd_read_only || get_disk_ro(bdev->bd_disk); -} -EXPORT_SYMBOL(bdev_read_only); - void inc_diskseq(struct gendisk *disk) { disk->diskseq = atomic64_inc_return(&diskseq); diff --git a/block/holder.c b/block/holder.c index 9dc084182337..27cddce1b446 100644 --- a/block/holder.c +++ b/block/holder.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-only #include <linux/genhd.h> +#include <linux/slab.h> struct bd_holder_disk { struct list_head list; diff --git a/block/ioctl.c b/block/ioctl.c index eb0491e90b9a..77b1b2453f39 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -538,12 +538,22 @@ static int blkdev_common_ioctl(struct block_device *bdev, fmode_t mode, * * New commands must be compatible and go into blkdev_common_ioctl */ -int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd, - unsigned long arg) +long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg) { - int ret; - loff_t size; + struct block_device *bdev = I_BDEV(file->f_mapping->host); void __user *argp = (void __user *)arg; + fmode_t mode = file->f_mode; + loff_t size; + int ret; + + /* + * O_NDELAY can be altered using fcntl(.., F_SETFL, ..), so we have + * to updated it before every ioctl. + */ + if (file->f_flags & O_NDELAY) + mode |= FMODE_NDELAY; + else + mode &= ~FMODE_NDELAY; switch (cmd) { /* These need separate implementations for the data structure */ @@ -588,7 +598,6 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd, return -ENOTTY; return bdev->bd_disk->fops->ioctl(bdev, mode, cmd, arg); } -EXPORT_SYMBOL_GPL(blkdev_ioctl); /* for /dev/raw */ #ifdef CONFIG_COMPAT diff --git a/block/keyslot-manager.c b/block/keyslot-manager.c deleted file mode 100644 index 2c4a55bea6ca..000000000000 --- a/block/keyslot-manager.c +++ /dev/null @@ -1,578 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * Copyright 2019 Google LLC - */ - -/** - * DOC: The Keyslot Manager - * - * Many devices with inline encryption support have a limited number of "slots" - * into which encryption contexts may be programmed, and requests can be tagged - * with a slot number to specify the key to use for en/decryption. - * - * As the number of slots is limited, and programming keys is expensive on - * many inline encryption hardware, we don't want to program the same key into - * multiple slots - if multiple requests are using the same key, we want to - * program just one slot with that key and use that slot for all requests. - * - * The keyslot manager manages these keyslots appropriately, and also acts as - * an abstraction between the inline encryption hardware and the upper layers. - * - * Lower layer devices will set up a keyslot manager in their request queue - * and tell it how to perform device specific operations like programming/ - * evicting keys from keyslots. - * - * Upper layers will call blk_ksm_get_slot_for_key() to program a - * key into some slot in the inline encryption hardware. - */ - -#define pr_fmt(fmt) "blk-crypto: " fmt - -#include <linux/keyslot-manager.h> -#include <linux/device.h> -#include <linux/atomic.h> -#include <linux/mutex.h> -#include <linux/pm_runtime.h> -#include <linux/wait.h> -#include <linux/blkdev.h> - -struct blk_ksm_keyslot { - atomic_t slot_refs; - struct list_head idle_slot_node; - struct hlist_node hash_node; - const struct blk_crypto_key *key; - struct blk_keyslot_manager *ksm; -}; - -static inline void blk_ksm_hw_enter(struct blk_keyslot_manager *ksm) -{ - /* - * Calling into the driver requires ksm->lock held and the device - * resumed. But we must resume the device first, since that can acquire - * and release ksm->lock via blk_ksm_reprogram_all_keys(). - */ - if (ksm->dev) - pm_runtime_get_sync(ksm->dev); - down_write(&ksm->lock); -} - -static inline void blk_ksm_hw_exit(struct blk_keyslot_manager *ksm) -{ - up_write(&ksm->lock); - if (ksm->dev) - pm_runtime_put_sync(ksm->dev); -} - -static inline bool blk_ksm_is_passthrough(struct blk_keyslot_manager *ksm) -{ - return ksm->num_slots == 0; -} - -/** - * blk_ksm_init() - Initialize a keyslot manager - * @ksm: The keyslot_manager to initialize. - * @num_slots: The number of key slots to manage. - * - * Allocate memory for keyslots and initialize a keyslot manager. Called by - * e.g. storage drivers to set up a keyslot manager in their request_queue. - * - * Return: 0 on success, or else a negative error code. - */ -int blk_ksm_init(struct blk_keyslot_manager *ksm, unsigned int num_slots) -{ - unsigned int slot; - unsigned int i; - unsigned int slot_hashtable_size; - - memset(ksm, 0, sizeof(*ksm)); - - if (num_slots == 0) - return -EINVAL; - - ksm->slots = kvcalloc(num_slots, sizeof(ksm->slots[0]), GFP_KERNEL); - if (!ksm->slots) - return -ENOMEM; - - ksm->num_slots = num_slots; - - init_rwsem(&ksm->lock); - - init_waitqueue_head(&ksm->idle_slots_wait_queue); - INIT_LIST_HEAD(&ksm->idle_slots); - - for (slot = 0; slot < num_slots; slot++) { - ksm->slots[slot].ksm = ksm; - list_add_tail(&ksm->slots[slot].idle_slot_node, - &ksm->idle_slots); - } - - spin_lock_init(&ksm->idle_slots_lock); - - slot_hashtable_size = roundup_pow_of_two(num_slots); - /* - * hash_ptr() assumes bits != 0, so ensure the hash table has at least 2 - * buckets. This only makes a difference when there is only 1 keyslot. - */ - if (slot_hashtable_size < 2) - slot_hashtable_size = 2; - - ksm->log_slot_ht_size = ilog2(slot_hashtable_size); - ksm->slot_hashtable = kvmalloc_array(slot_hashtable_size, - sizeof(ksm->slot_hashtable[0]), - GFP_KERNEL); - if (!ksm->slot_hashtable) - goto err_destroy_ksm; - for (i = 0; i < slot_hashtable_size; i++) - INIT_HLIST_HEAD(&ksm->slot_hashtable[i]); - - return 0; - -err_destroy_ksm: - blk_ksm_destroy(ksm); - return -ENOMEM; -} -EXPORT_SYMBOL_GPL(blk_ksm_init); - -static void blk_ksm_destroy_callback(void *ksm) -{ - blk_ksm_destroy(ksm); -} - -/** - * devm_blk_ksm_init() - Resource-managed blk_ksm_init() - * @dev: The device which owns the blk_keyslot_manager. - * @ksm: The blk_keyslot_manager to initialize. - * @num_slots: The number of key slots to manage. - * - * Like blk_ksm_init(), but causes blk_ksm_destroy() to be called automatically - * on driver detach. - * - * Return: 0 on success, or else a negative error code. - */ -int devm_blk_ksm_init(struct device *dev, struct blk_keyslot_manager *ksm, - unsigned int num_slots) -{ - int err = blk_ksm_init(ksm, num_slots); - - if (err) - return err; - - return devm_add_action_or_reset(dev, blk_ksm_destroy_callback, ksm); -} -EXPORT_SYMBOL_GPL(devm_blk_ksm_init); - -static inline struct hlist_head * -blk_ksm_hash_bucket_for_key(struct blk_keyslot_manager *ksm, - const struct blk_crypto_key *key) -{ - return &ksm->slot_hashtable[hash_ptr(key, ksm->log_slot_ht_size)]; -} - -static void blk_ksm_remove_slot_from_lru_list(struct blk_ksm_keyslot *slot) -{ - struct blk_keyslot_manager *ksm = slot->ksm; - unsigned long flags; - - spin_lock_irqsave(&ksm->idle_slots_lock, flags); - list_del(&slot->idle_slot_node); - spin_unlock_irqrestore(&ksm->idle_slots_lock, flags); -} - -static struct blk_ksm_keyslot *blk_ksm_find_keyslot( - struct blk_keyslot_manager *ksm, - const struct blk_crypto_key *key) -{ - const struct hlist_head *head = blk_ksm_hash_bucket_for_key(ksm, key); - struct blk_ksm_keyslot *slotp; - - hlist_for_each_entry(slotp, head, hash_node) { - if (slotp->key == key) - return slotp; - } - return NULL; -} - -static struct blk_ksm_keyslot *blk_ksm_find_and_grab_keyslot( - struct blk_keyslot_manager *ksm, - const struct blk_crypto_key *key) -{ - struct blk_ksm_keyslot *slot; - - slot = blk_ksm_find_keyslot(ksm, key); - if (!slot) - return NULL; - if (atomic_inc_return(&slot->slot_refs) == 1) { - /* Took first reference to this slot; remove it from LRU list */ - blk_ksm_remove_slot_from_lru_list(slot); - } - return slot; -} - -unsigned int blk_ksm_get_slot_idx(struct blk_ksm_keyslot *slot) -{ - return slot - slot->ksm->slots; -} -EXPORT_SYMBOL_GPL(blk_ksm_get_slot_idx); - -/** - * blk_ksm_get_slot_for_key() - Program a key into a keyslot. - * @ksm: The keyslot manager to program the key into. - * @key: Pointer to the key object to program, including the raw key, crypto - * mode, and data unit size. - * @slot_ptr: A pointer to return the pointer of the allocated keyslot. - * - * Get a keyslot that's been programmed with the specified key. If one already - * exists, return it with incremented refcount. Otherwise, wait for a keyslot - * to become idle and program it. - * - * Context: Process context. Takes and releases ksm->lock. - * Return: BLK_STS_OK on success (and keyslot is set to the pointer of the - * allocated keyslot), or some other blk_status_t otherwise (and - * keyslot is set to NULL). - */ -blk_status_t blk_ksm_get_slot_for_key(struct blk_keyslot_manager *ksm, - const struct blk_crypto_key *key, - struct blk_ksm_keyslot **slot_ptr) -{ - struct blk_ksm_keyslot *slot; - int slot_idx; - int err; - - *slot_ptr = NULL; - - if (blk_ksm_is_passthrough(ksm)) - return BLK_STS_OK; - - down_read(&ksm->lock); - slot = blk_ksm_find_and_grab_keyslot(ksm, key); - up_read(&ksm->lock); - if (slot) - goto success; - - for (;;) { - blk_ksm_hw_enter(ksm); - slot = blk_ksm_find_and_grab_keyslot(ksm, key); - if (slot) { - blk_ksm_hw_exit(ksm); - goto success; - } - - /* - * If we're here, that means there wasn't a slot that was - * already programmed with the key. So try to program it. - */ - if (!list_empty(&ksm->idle_slots)) - break; - - blk_ksm_hw_exit(ksm); - wait_event(ksm->idle_slots_wait_queue, - !list_empty(&ksm->idle_slots)); - } - - slot = list_first_entry(&ksm->idle_slots, struct blk_ksm_keyslot, - idle_slot_node); - slot_idx = blk_ksm_get_slot_idx(slot); - - err = ksm->ksm_ll_ops.keyslot_program(ksm, key, slot_idx); - if (err) { - wake_up(&ksm->idle_slots_wait_queue); - blk_ksm_hw_exit(ksm); - return errno_to_blk_status(err); - } - - /* Move this slot to the hash list for the new key. */ - if (slot->key) - hlist_del(&slot->hash_node); - slot->key = key; - hlist_add_head(&slot->hash_node, blk_ksm_hash_bucket_for_key(ksm, key)); - - atomic_set(&slot->slot_refs, 1); - - blk_ksm_remove_slot_from_lru_list(slot); - - blk_ksm_hw_exit(ksm); -success: - *slot_ptr = slot; - return BLK_STS_OK; -} - -/** - * blk_ksm_put_slot() - Release a reference to a slot - * @slot: The keyslot to release the reference of. - * - * Context: Any context. - */ -void blk_ksm_put_slot(struct blk_ksm_keyslot *slot) -{ - struct blk_keyslot_manager *ksm; - unsigned long flags; - - if (!slot) - return; - - ksm = slot->ksm; - - if (atomic_dec_and_lock_irqsave(&slot->slot_refs, - &ksm->idle_slots_lock, flags)) { - list_add_tail(&slot->idle_slot_node, &ksm->idle_slots); - spin_unlock_irqrestore(&ksm->idle_slots_lock, flags); - wake_up(&ksm->idle_slots_wait_queue); - } -} - -/** - * blk_ksm_crypto_cfg_supported() - Find out if a crypto configuration is - * supported by a ksm. - * @ksm: The keyslot manager to check - * @cfg: The crypto configuration to check for. - * - * Checks for crypto_mode/data unit size/dun bytes support. - * - * Return: Whether or not this ksm supports the specified crypto config. - */ -bool blk_ksm_crypto_cfg_supported(struct blk_keyslot_manager *ksm, - const struct blk_crypto_config *cfg) -{ - if (!ksm) - return false; - if (!(ksm->crypto_modes_supported[cfg->crypto_mode] & - cfg->data_unit_size)) - return false; - if (ksm->max_dun_bytes_supported < cfg->dun_bytes) - return false; - return true; -} - -/** - * blk_ksm_evict_key() - Evict a key from the lower layer device. - * @ksm: The keyslot manager to evict from - * @key: The key to evict - * - * Find the keyslot that the specified key was programmed into, and evict that - * slot from the lower layer device. The slot must not be in use by any - * in-flight IO when this function is called. - * - * Context: Process context. Takes and releases ksm->lock. - * Return: 0 on success or if there's no keyslot with the specified key, -EBUSY - * if the keyslot is still in use, or another -errno value on other - * error. - */ -int blk_ksm_evict_key(struct blk_keyslot_manager *ksm, - const struct blk_crypto_key *key) -{ - struct blk_ksm_keyslot *slot; - int err = 0; - - if (blk_ksm_is_passthrough(ksm)) { - if (ksm->ksm_ll_ops.keyslot_evict) { - blk_ksm_hw_enter(ksm); - err = ksm->ksm_ll_ops.keyslot_evict(ksm, key, -1); - blk_ksm_hw_exit(ksm); - return err; - } - return 0; - } - - blk_ksm_hw_enter(ksm); - slot = blk_ksm_find_keyslot(ksm, key); - if (!slot) - goto out_unlock; - - if (WARN_ON_ONCE(atomic_read(&slot->slot_refs) != 0)) { - err = -EBUSY; - goto out_unlock; - } - err = ksm->ksm_ll_ops.keyslot_evict(ksm, key, - blk_ksm_get_slot_idx(slot)); - if (err) - goto out_unlock; - - hlist_del(&slot->hash_node); - slot->key = NULL; - err = 0; -out_unlock: - blk_ksm_hw_exit(ksm); - return err; -} - -/** - * blk_ksm_reprogram_all_keys() - Re-program all keyslots. - * @ksm: The keyslot manager - * - * Re-program all keyslots that are supposed to have a key programmed. This is - * intended only for use by drivers for hardware that loses its keys on reset. - * - * Context: Process context. Takes and releases ksm->lock. - */ -void blk_ksm_reprogram_all_keys(struct blk_keyslot_manager *ksm) -{ - unsigned int slot; - - if (blk_ksm_is_passthrough(ksm)) - return; - - /* This is for device initialization, so don't resume the device */ - down_write(&ksm->lock); - for (slot = 0; slot < ksm->num_slots; slot++) { - const struct blk_crypto_key *key = ksm->slots[slot].key; - int err; - - if (!key) - continue; - - err = ksm->ksm_ll_ops.keyslot_program(ksm, key, slot); - WARN_ON(err); - } - up_write(&ksm->lock); -} -EXPORT_SYMBOL_GPL(blk_ksm_reprogram_all_keys); - -void blk_ksm_destroy(struct blk_keyslot_manager *ksm) -{ - if (!ksm) - return; - kvfree(ksm->slot_hashtable); - kvfree_sensitive(ksm->slots, sizeof(ksm->slots[0]) * ksm->num_slots); - memzero_explicit(ksm, sizeof(*ksm)); -} -EXPORT_SYMBOL_GPL(blk_ksm_destroy); - -bool blk_ksm_register(struct blk_keyslot_manager *ksm, struct request_queue *q) -{ - if (blk_integrity_queue_supports_integrity(q)) { - pr_warn("Integrity and hardware inline encryption are not supported together. Disabling hardware inline encryption.\n"); - return false; - } - q->ksm = ksm; - return true; -} -EXPORT_SYMBOL_GPL(blk_ksm_register); - -void blk_ksm_unregister(struct request_queue *q) -{ - q->ksm = NULL; -} - -/** - * blk_ksm_intersect_modes() - restrict supported modes by child device - * @parent: The keyslot manager for parent device - * @child: The keyslot manager for child device, or NULL - * - * Clear any crypto mode support bits in @parent that aren't set in @child. - * If @child is NULL, then all parent bits are cleared. - * - * Only use this when setting up the keyslot manager for a layered device, - * before it's been exposed yet. - */ -void blk_ksm_intersect_modes(struct blk_keyslot_manager *parent, - const struct blk_keyslot_manager *child) -{ - if (child) { - unsigned int i; - - parent->max_dun_bytes_supported = - min(parent->max_dun_bytes_supported, - child->max_dun_bytes_supported); - for (i = 0; i < ARRAY_SIZE(child->crypto_modes_supported); - i++) { - parent->crypto_modes_supported[i] &= - child->crypto_modes_supported[i]; - } - } else { - parent->max_dun_bytes_supported = 0; - memset(parent->crypto_modes_supported, 0, - sizeof(parent->crypto_modes_supported)); - } -} -EXPORT_SYMBOL_GPL(blk_ksm_intersect_modes); - -/** - * blk_ksm_is_superset() - Check if a KSM supports a superset of crypto modes - * and DUN bytes that another KSM supports. Here, - * "superset" refers to the mathematical meaning of the - * word - i.e. if two KSMs have the *same* capabilities, - * they *are* considered supersets of each other. - * @ksm_superset: The KSM that we want to verify is a superset - * @ksm_subset: The KSM that we want to verify is a subset - * - * Return: True if @ksm_superset supports a superset of the crypto modes and DUN - * bytes that @ksm_subset supports. - */ -bool blk_ksm_is_superset(struct blk_keyslot_manager *ksm_superset, - struct blk_keyslot_manager *ksm_subset) -{ - int i; - - if (!ksm_subset) - return true; - - if (!ksm_superset) - return false; - - for (i = 0; i < ARRAY_SIZE(ksm_superset->crypto_modes_supported); i++) { - if (ksm_subset->crypto_modes_supported[i] & - (~ksm_superset->crypto_modes_supported[i])) { - return false; - } - } - - if (ksm_subset->max_dun_bytes_supported > - ksm_superset->max_dun_bytes_supported) { - return false; - } - - return true; -} -EXPORT_SYMBOL_GPL(blk_ksm_is_superset); - -/** - * blk_ksm_update_capabilities() - Update the restrictions of a KSM to those of - * another KSM - * @target_ksm: The KSM whose restrictions to update. - * @reference_ksm: The KSM to whose restrictions this function will update - * @target_ksm's restrictions to. - * - * Blk-crypto requires that crypto capabilities that were - * advertised when a bio was created continue to be supported by the - * device until that bio is ended. This is turn means that a device cannot - * shrink its advertised crypto capabilities without any explicit - * synchronization with upper layers. So if there's no such explicit - * synchronization, @reference_ksm must support all the crypto capabilities that - * @target_ksm does - * (i.e. we need blk_ksm_is_superset(@reference_ksm, @target_ksm) == true). - * - * Note also that as long as the crypto capabilities are being expanded, the - * order of updates becoming visible is not important because it's alright - * for blk-crypto to see stale values - they only cause blk-crypto to - * believe that a crypto capability isn't supported when it actually is (which - * might result in blk-crypto-fallback being used if available, or the bio being - * failed). - */ -void blk_ksm_update_capabilities(struct blk_keyslot_manager *target_ksm, - struct blk_keyslot_manager *reference_ksm) -{ - memcpy(target_ksm->crypto_modes_supported, - reference_ksm->crypto_modes_supported, - sizeof(target_ksm->crypto_modes_supported)); - - target_ksm->max_dun_bytes_supported = - reference_ksm->max_dun_bytes_supported; -} -EXPORT_SYMBOL_GPL(blk_ksm_update_capabilities); - -/** - * blk_ksm_init_passthrough() - Init a passthrough keyslot manager - * @ksm: The keyslot manager to init - * - * Initialize a passthrough keyslot manager. - * Called by e.g. storage drivers to set up a keyslot manager in their - * request_queue, when the storage driver wants to manage its keys by itself. - * This is useful for inline encryption hardware that doesn't have the concept - * of keyslots, and for layered devices. - */ -void blk_ksm_init_passthrough(struct blk_keyslot_manager *ksm) -{ - memset(ksm, 0, sizeof(*ksm)); - init_rwsem(&ksm->lock); -} -EXPORT_SYMBOL_GPL(blk_ksm_init_passthrough); diff --git a/block/kyber-iosched.c b/block/kyber-iosched.c index a0ffbabfac2c..fdd74a4df56f 100644 --- a/block/kyber-iosched.c +++ b/block/kyber-iosched.c @@ -9,12 +9,12 @@ #include <linux/kernel.h> #include <linux/blkdev.h> #include <linux/blk-mq.h> -#include <linux/elevator.h> #include <linux/module.h> #include <linux/sbitmap.h> #include <trace/events/block.h> +#include "elevator.h" #include "blk.h" #include "blk-mq.h" #include "blk-mq-debugfs.h" @@ -453,11 +453,11 @@ static void kyber_depth_updated(struct blk_mq_hw_ctx *hctx) { struct kyber_queue_data *kqd = hctx->queue->elevator->elevator_data; struct blk_mq_tags *tags = hctx->sched_tags; - unsigned int shift = tags->bitmap_tags->sb.shift; + unsigned int shift = tags->bitmap_tags.sb.shift; kqd->async_depth = (1U << shift) * KYBER_ASYNC_PERCENT / 100U; - sbitmap_queue_min_shallow_depth(tags->bitmap_tags, kqd->async_depth); + sbitmap_queue_min_shallow_depth(&tags->bitmap_tags, kqd->async_depth); } static int kyber_init_hctx(struct blk_mq_hw_ctx *hctx, unsigned int hctx_idx) diff --git a/block/mq-deadline.c b/block/mq-deadline.c index 7f3c3932b723..85d919bf60c7 100644 --- a/block/mq-deadline.c +++ b/block/mq-deadline.c @@ -9,7 +9,6 @@ #include <linux/fs.h> #include <linux/blkdev.h> #include <linux/blk-mq.h> -#include <linux/elevator.h> #include <linux/bio.h> #include <linux/module.h> #include <linux/slab.h> @@ -20,6 +19,7 @@ #include <trace/events/block.h> +#include "elevator.h" #include "blk.h" #include "blk-mq.h" #include "blk-mq-debugfs.h" @@ -31,6 +31,11 @@ */ static const int read_expire = HZ / 2; /* max time before a read is submitted. */ static const int write_expire = 5 * HZ; /* ditto for writes, these limits are SOFT! */ +/* + * Time after which to dispatch lower priority requests even if higher + * priority requests are pending. + */ +static const int prio_aging_expire = 10 * HZ; static const int writes_starved = 2; /* max times reads can starve a write */ static const int fifo_batch = 16; /* # of sequential requests treated as one by the above parameters. For throughput. */ @@ -51,17 +56,16 @@ enum dd_prio { enum { DD_PRIO_COUNT = 3 }; -/* I/O statistics per I/O priority. */ +/* + * I/O statistics per I/O priority. It is fine if these counters overflow. + * What matters is that these counters are at least as wide as + * log2(max_outstanding_requests). + */ struct io_stats_per_prio { - local_t inserted; - local_t merged; - local_t dispatched; - local_t completed; -}; - -/* I/O statistics for all I/O priorities (enum dd_prio). */ -struct io_stats { - struct io_stats_per_prio stats[DD_PRIO_COUNT]; + uint32_t inserted; + uint32_t merged; + uint32_t dispatched; + atomic_t completed; }; /* @@ -74,6 +78,7 @@ struct dd_per_prio { struct list_head fifo_list[DD_DIR_COUNT]; /* Next request in FIFO order. Read, write or both are NULL. */ struct request *next_rq[DD_DIR_COUNT]; + struct io_stats_per_prio stats; }; struct deadline_data { @@ -88,8 +93,6 @@ struct deadline_data { unsigned int batching; /* number of sequential requests made */ unsigned int starved; /* times reads have starved writes */ - struct io_stats __percpu *stats; - /* * settings that change how the i/o scheduler behaves */ @@ -98,38 +101,12 @@ struct deadline_data { int writes_starved; int front_merges; u32 async_depth; + int prio_aging_expire; spinlock_t lock; spinlock_t zone_lock; }; -/* Count one event of type 'event_type' and with I/O priority 'prio' */ -#define dd_count(dd, event_type, prio) do { \ - struct io_stats *io_stats = get_cpu_ptr((dd)->stats); \ - \ - BUILD_BUG_ON(!__same_type((dd), struct deadline_data *)); \ - BUILD_BUG_ON(!__same_type((prio), enum dd_prio)); \ - local_inc(&io_stats->stats[(prio)].event_type); \ - put_cpu_ptr(io_stats); \ -} while (0) - -/* - * Returns the total number of dd_count(dd, event_type, prio) calls across all - * CPUs. No locking or barriers since it is fine if the returned sum is slightly - * outdated. - */ -#define dd_sum(dd, event_type, prio) ({ \ - unsigned int cpu; \ - u32 sum = 0; \ - \ - BUILD_BUG_ON(!__same_type((dd), struct deadline_data *)); \ - BUILD_BUG_ON(!__same_type((prio), enum dd_prio)); \ - for_each_present_cpu(cpu) \ - sum += local_read(&per_cpu_ptr((dd)->stats, cpu)-> \ - stats[(prio)].event_type); \ - sum; \ -}) - /* Maps an I/O priority class to a deadline scheduler priority. */ static const enum dd_prio ioprio_class_to_prio[] = { [IOPRIO_CLASS_NONE] = DD_BE_PRIO, @@ -233,7 +210,9 @@ static void dd_merged_requests(struct request_queue *q, struct request *req, const u8 ioprio_class = dd_rq_ioclass(next); const enum dd_prio prio = ioprio_class_to_prio[ioprio_class]; - dd_count(dd, merged, prio); + lockdep_assert_held(&dd->lock); + + dd->per_prio[prio].stats.merged++; /* * if next expires before rq, assign its expire time to rq @@ -270,6 +249,16 @@ deadline_move_request(struct deadline_data *dd, struct dd_per_prio *per_prio, deadline_remove_request(rq->q, per_prio, rq); } +/* Number of requests queued for a given priority level. */ +static u32 dd_queued(struct deadline_data *dd, enum dd_prio prio) +{ + const struct io_stats_per_prio *stats = &dd->per_prio[prio].stats; + + lockdep_assert_held(&dd->lock); + + return stats->inserted - atomic_read(&stats->completed); +} + /* * deadline_check_fifo returns 0 if there are no expired requests on the fifo, * 1 otherwise. Requires !list_empty(&dd->fifo_list[data_dir]) @@ -356,11 +345,26 @@ deadline_next_request(struct deadline_data *dd, struct dd_per_prio *per_prio, } /* + * Returns true if and only if @rq started after @latest_start where + * @latest_start is in jiffies. + */ +static bool started_after(struct deadline_data *dd, struct request *rq, + unsigned long latest_start) +{ + unsigned long start_time = (unsigned long)rq->fifo_time; + + start_time -= dd->fifo_expire[rq_data_dir(rq)]; + + return time_after(start_time, latest_start); +} + +/* * deadline_dispatch_requests selects the best request according to - * read/write expire, fifo_batch, etc + * read/write expire, fifo_batch, etc and with a start time <= @latest_start. */ static struct request *__dd_dispatch_request(struct deadline_data *dd, - struct dd_per_prio *per_prio) + struct dd_per_prio *per_prio, + unsigned long latest_start) { struct request *rq, *next_rq; enum dd_data_dir data_dir; @@ -372,6 +376,8 @@ static struct request *__dd_dispatch_request(struct deadline_data *dd, if (!list_empty(&per_prio->dispatch)) { rq = list_first_entry(&per_prio->dispatch, struct request, queuelist); + if (started_after(dd, rq, latest_start)) + return NULL; list_del_init(&rq->queuelist); goto done; } @@ -449,6 +455,9 @@ dispatch_find_request: dd->batching = 0; dispatch_request: + if (started_after(dd, rq, latest_start)) + return NULL; + /* * rq is the selected appropriate request. */ @@ -457,7 +466,7 @@ dispatch_request: done: ioprio_class = dd_rq_ioclass(rq); prio = ioprio_class_to_prio[ioprio_class]; - dd_count(dd, dispatched, prio); + dd->per_prio[prio].stats.dispatched++; /* * If the request needs its target zone locked, do it. */ @@ -467,6 +476,34 @@ done: } /* + * Check whether there are any requests with priority other than DD_RT_PRIO + * that were inserted more than prio_aging_expire jiffies ago. + */ +static struct request *dd_dispatch_prio_aged_requests(struct deadline_data *dd, + unsigned long now) +{ + struct request *rq; + enum dd_prio prio; + int prio_cnt; + + lockdep_assert_held(&dd->lock); + + prio_cnt = !!dd_queued(dd, DD_RT_PRIO) + !!dd_queued(dd, DD_BE_PRIO) + + !!dd_queued(dd, DD_IDLE_PRIO); + if (prio_cnt < 2) + return NULL; + + for (prio = DD_BE_PRIO; prio <= DD_PRIO_MAX; prio++) { + rq = __dd_dispatch_request(dd, &dd->per_prio[prio], + now - dd->prio_aging_expire); + if (rq) + return rq; + } + + return NULL; +} + +/* * Called from blk_mq_run_hw_queue() -> __blk_mq_sched_dispatch_requests(). * * One confusing aspect here is that we get called for a specific @@ -477,15 +514,26 @@ done: static struct request *dd_dispatch_request(struct blk_mq_hw_ctx *hctx) { struct deadline_data *dd = hctx->queue->elevator->elevator_data; + const unsigned long now = jiffies; struct request *rq; enum dd_prio prio; spin_lock(&dd->lock); + rq = dd_dispatch_prio_aged_requests(dd, now); + if (rq) + goto unlock; + + /* + * Next, dispatch requests in priority order. Ignore lower priority + * requests if any higher priority requests are pending. + */ for (prio = 0; prio <= DD_PRIO_MAX; prio++) { - rq = __dd_dispatch_request(dd, &dd->per_prio[prio]); - if (rq) + rq = __dd_dispatch_request(dd, &dd->per_prio[prio], now); + if (rq || dd_queued(dd, prio)) break; } + +unlock: spin_unlock(&dd->lock); return rq; @@ -519,7 +567,7 @@ static void dd_depth_updated(struct blk_mq_hw_ctx *hctx) dd->async_depth = max(1UL, 3 * q->nr_requests / 4); - sbitmap_queue_min_shallow_depth(tags->bitmap_tags, dd->async_depth); + sbitmap_queue_min_shallow_depth(&tags->bitmap_tags, dd->async_depth); } /* Called by blk_mq_init_hctx() and blk_mq_init_sched(). */ @@ -536,12 +584,21 @@ static void dd_exit_sched(struct elevator_queue *e) for (prio = 0; prio <= DD_PRIO_MAX; prio++) { struct dd_per_prio *per_prio = &dd->per_prio[prio]; + const struct io_stats_per_prio *stats = &per_prio->stats; + uint32_t queued; WARN_ON_ONCE(!list_empty(&per_prio->fifo_list[DD_READ])); WARN_ON_ONCE(!list_empty(&per_prio->fifo_list[DD_WRITE])); - } - free_percpu(dd->stats); + spin_lock(&dd->lock); + queued = dd_queued(dd, prio); + spin_unlock(&dd->lock); + + WARN_ONCE(queued != 0, + "statistics for priority %d: i %u m %u d %u c %u\n", + prio, stats->inserted, stats->merged, + stats->dispatched, atomic_read(&stats->completed)); + } kfree(dd); } @@ -566,11 +623,6 @@ static int dd_init_sched(struct request_queue *q, struct elevator_type *e) eq->elevator_data = dd; - dd->stats = alloc_percpu_gfp(typeof(*dd->stats), - GFP_KERNEL | __GFP_ZERO); - if (!dd->stats) - goto free_dd; - for (prio = 0; prio <= DD_PRIO_MAX; prio++) { struct dd_per_prio *per_prio = &dd->per_prio[prio]; @@ -586,15 +638,13 @@ static int dd_init_sched(struct request_queue *q, struct elevator_type *e) dd->front_merges = 1; dd->last_dir = DD_WRITE; dd->fifo_batch = fifo_batch; + dd->prio_aging_expire = prio_aging_expire; spin_lock_init(&dd->lock); spin_lock_init(&dd->zone_lock); q->elevator = eq; return 0; -free_dd: - kfree(dd); - put_eq: kobject_put(&eq->kobj); return ret; @@ -677,8 +727,11 @@ static void dd_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, blk_req_zone_write_unlock(rq); prio = ioprio_class_to_prio[ioprio_class]; - dd_count(dd, inserted, prio); - rq->elv.priv[0] = (void *)(uintptr_t)1; + per_prio = &dd->per_prio[prio]; + if (!rq->elv.priv[0]) { + per_prio->stats.inserted++; + rq->elv.priv[0] = (void *)(uintptr_t)1; + } if (blk_mq_sched_try_insert_merge(q, rq, &free)) { blk_mq_free_requests(&free); @@ -687,7 +740,6 @@ static void dd_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, trace_block_rq_insert(rq); - per_prio = &dd->per_prio[prio]; if (at_head) { list_add(&rq->queuelist, &per_prio->dispatch); } else { @@ -759,12 +811,13 @@ static void dd_finish_request(struct request *rq) /* * The block layer core may call dd_finish_request() without having - * called dd_insert_requests(). Hence only update statistics for - * requests for which dd_insert_requests() has been called. See also - * blk_mq_request_bypass_insert(). + * called dd_insert_requests(). Skip requests that bypassed I/O + * scheduling. See also blk_mq_request_bypass_insert(). */ - if (rq->elv.priv[0]) - dd_count(dd, completed, prio); + if (!rq->elv.priv[0]) + return; + + atomic_inc(&per_prio->stats.completed); if (blk_queue_is_zoned(q)) { unsigned long flags; @@ -809,6 +862,7 @@ static ssize_t __FUNC(struct elevator_queue *e, char *page) \ #define SHOW_JIFFIES(__FUNC, __VAR) SHOW_INT(__FUNC, jiffies_to_msecs(__VAR)) SHOW_JIFFIES(deadline_read_expire_show, dd->fifo_expire[DD_READ]); SHOW_JIFFIES(deadline_write_expire_show, dd->fifo_expire[DD_WRITE]); +SHOW_JIFFIES(deadline_prio_aging_expire_show, dd->prio_aging_expire); SHOW_INT(deadline_writes_starved_show, dd->writes_starved); SHOW_INT(deadline_front_merges_show, dd->front_merges); SHOW_INT(deadline_async_depth_show, dd->front_merges); @@ -838,6 +892,7 @@ static ssize_t __FUNC(struct elevator_queue *e, const char *page, size_t count) STORE_FUNCTION(__FUNC, __PTR, MIN, MAX, msecs_to_jiffies) STORE_JIFFIES(deadline_read_expire_store, &dd->fifo_expire[DD_READ], 0, INT_MAX); STORE_JIFFIES(deadline_write_expire_store, &dd->fifo_expire[DD_WRITE], 0, INT_MAX); +STORE_JIFFIES(deadline_prio_aging_expire_store, &dd->prio_aging_expire, 0, INT_MAX); STORE_INT(deadline_writes_starved_store, &dd->writes_starved, INT_MIN, INT_MAX); STORE_INT(deadline_front_merges_store, &dd->front_merges, 0, 1); STORE_INT(deadline_async_depth_store, &dd->front_merges, 1, INT_MAX); @@ -856,6 +911,7 @@ static struct elv_fs_entry deadline_attrs[] = { DD_ATTR(front_merges), DD_ATTR(async_depth), DD_ATTR(fifo_batch), + DD_ATTR(prio_aging_expire), __ATTR_NULL }; @@ -947,38 +1003,48 @@ static int dd_async_depth_show(void *data, struct seq_file *m) return 0; } -/* Number of requests queued for a given priority level. */ -static u32 dd_queued(struct deadline_data *dd, enum dd_prio prio) -{ - return dd_sum(dd, inserted, prio) - dd_sum(dd, completed, prio); -} - static int dd_queued_show(void *data, struct seq_file *m) { struct request_queue *q = data; struct deadline_data *dd = q->elevator->elevator_data; + u32 rt, be, idle; + + spin_lock(&dd->lock); + rt = dd_queued(dd, DD_RT_PRIO); + be = dd_queued(dd, DD_BE_PRIO); + idle = dd_queued(dd, DD_IDLE_PRIO); + spin_unlock(&dd->lock); + + seq_printf(m, "%u %u %u\n", rt, be, idle); - seq_printf(m, "%u %u %u\n", dd_queued(dd, DD_RT_PRIO), - dd_queued(dd, DD_BE_PRIO), - dd_queued(dd, DD_IDLE_PRIO)); return 0; } /* Number of requests owned by the block driver for a given priority. */ static u32 dd_owned_by_driver(struct deadline_data *dd, enum dd_prio prio) { - return dd_sum(dd, dispatched, prio) + dd_sum(dd, merged, prio) - - dd_sum(dd, completed, prio); + const struct io_stats_per_prio *stats = &dd->per_prio[prio].stats; + + lockdep_assert_held(&dd->lock); + + return stats->dispatched + stats->merged - + atomic_read(&stats->completed); } static int dd_owned_by_driver_show(void *data, struct seq_file *m) { struct request_queue *q = data; struct deadline_data *dd = q->elevator->elevator_data; + u32 rt, be, idle; + + spin_lock(&dd->lock); + rt = dd_owned_by_driver(dd, DD_RT_PRIO); + be = dd_owned_by_driver(dd, DD_BE_PRIO); + idle = dd_owned_by_driver(dd, DD_IDLE_PRIO); + spin_unlock(&dd->lock); + + seq_printf(m, "%u %u %u\n", rt, be, idle); - seq_printf(m, "%u %u %u\n", dd_owned_by_driver(dd, DD_RT_PRIO), - dd_owned_by_driver(dd, DD_BE_PRIO), - dd_owned_by_driver(dd, DD_IDLE_PRIO)); return 0; } diff --git a/block/partitions/Kconfig b/block/partitions/Kconfig index 278593b8e4e9..7aff4eb81c60 100644 --- a/block/partitions/Kconfig +++ b/block/partitions/Kconfig @@ -2,6 +2,8 @@ # # Partition configuration # +menu "Partition Types" + config PARTITION_ADVANCED bool "Advanced partition selection" help @@ -267,3 +269,5 @@ config CMDLINE_PARTITION help Say Y here if you want to read the partition table from bootargs. The format for the command line is just like mtdparts. + +endmenu diff --git a/block/partitions/core.c b/block/partitions/core.c index 7bea19dd9458..ed5deef1d7e1 100644 --- a/block/partitions/core.c +++ b/block/partitions/core.c @@ -5,6 +5,7 @@ * Copyright (C) 2020 Christoph Hellwig */ #include <linux/fs.h> +#include <linux/major.h> #include <linux/slab.h> #include <linux/ctype.h> #include <linux/genhd.h> @@ -203,7 +204,7 @@ static ssize_t part_alignment_offset_show(struct device *dev, struct block_device *bdev = dev_to_bdev(dev); return sprintf(buf, "%u\n", - queue_limit_alignment_offset(&bdev->bd_disk->queue->limits, + queue_limit_alignment_offset(&bdev_get_queue(bdev)->limits, bdev->bd_start_sect)); } @@ -213,7 +214,7 @@ static ssize_t part_discard_alignment_show(struct device *dev, struct block_device *bdev = dev_to_bdev(dev); return sprintf(buf, "%u\n", - queue_limit_discard_alignment(&bdev->bd_disk->queue->limits, + queue_limit_discard_alignment(&bdev_get_queue(bdev)->limits, bdev->bd_start_sect)); } diff --git a/block/t10-pi.c b/block/t10-pi.c index 00c203b2a921..25a52a2a09a8 100644 --- a/block/t10-pi.c +++ b/block/t10-pi.c @@ -5,7 +5,7 @@ */ #include <linux/t10-pi.h> -#include <linux/blkdev.h> +#include <linux/blk-integrity.h> #include <linux/crc-t10dif.h> #include <linux/module.h> #include <net/checksum.h> |