Merge tag 'for-6.9/block-20240310' of git://git.kernel.dk/linux

Pull block updates from Jens Axboe:

 - MD pull requests via Song:
      - Cleanup redundant checks (Yu Kuai)
      - Remove deprecated headers (Marc Zyngier, Song Liu)
      - Concurrency fixes (Li Lingfeng)
      - Memory leak fix (Li Nan)
      - Refactor raid1 read_balance (Yu Kuai, Paul Luse)
      - Clean up and fix for md_ioctl (Li Nan)
      - Other small fixes (Gui-Dong Han, Heming Zhao)
      - MD atomic limits (Christoph)

 - NVMe pull request via Keith:
      - RDMA target enhancements (Max)
      - Fabrics fixes (Max, Guixin, Hannes)
      - Atomic queue_limits usage (Christoph)
      - Const use for class_register (Ricardo)
      - Identification error handling fixes (Shin'ichiro, Keith)

 - Improvement and cleanup for cached request handling (Christoph)

 - Moving towards atomic queue limits. Core changes and driver bits so
   far (Christoph)

 - Fix UAF issues in aoeblk (Chun-Yi)

 - Zoned fix and cleanups (Damien)

 - s390 dasd cleanups and fixes (Jan, Miroslav)

 - Block issue timestamp caching (me)

 - noio scope guarding for zoned IO (Johannes)

 - block/nvme PI improvements (Kanchan)

 - Ability to terminate long running discard loop (Keith)

 - bdev revalidation fix (Li)

 - Get rid of old nr_queues hack for kdump kernels (Ming)

 - Support for async deletion of ublk (Ming)

 - Improve IRQ bio recycling (Pavel)

 - Factor in CPU capacity for remote vs local completion (Qais)

 - Add shared_tags configfs entry for null_blk (Shin'ichiro

 - Fix for a regression in page refcounts introduced by the folio
   unification (Tony)

 - Misc fixes and cleanups (Arnd, Colin, John, Kunwu, Li, Navid,
   Ricardo, Roman, Tang, Uwe)

* tag 'for-6.9/block-20240310' of git://git.kernel.dk/linux: (221 commits)
  block: partitions: only define function mac_fix_string for CONFIG_PPC_PMAC
  block/swim: Convert to platform remove callback returning void
  cdrom: gdrom: Convert to platform remove callback returning void
  block: remove disk_stack_limits
  md: remove mddev->queue
  md: don't initialize queue limits
  md/raid10: use the atomic queue limit update APIs
  md/raid5: use the atomic queue limit update APIs
  md/raid1: use the atomic queue limit update APIs
  md/raid0: use the atomic queue limit update APIs
  md: add queue limit helpers
  md: add a mddev_is_dm helper
  md: add a mddev_add_trace_msg helper
  md: add a mddev_trace_remap helper
  bcache: move calculation of stripe_size and io_opt into bcache_device_init
  virtio_blk: Do not use disk_set_max_open/active_zones()
  aoe: fix the potential use-after-free problem in aoecmd_cfg_pkts
  block: move capacity validation to blkpg_do_ioctl()
  block: prevent division by zero in blk_rq_stat_sum()
  drbd: atomically update queue limits in drbd_reconsider_queue_parameters
  ...

29 files changed, 687 insertions, 361 deletions

diff --git a/block/bdev.c b/block/bdev.c
index 140093c99bdc..e7adaaf1c219 100644
--- a/block/bdev.c
+++ b/block/bdev.c
@@ -383,7 +383,7 @@ void __init bdev_cache_init(void)
 
 	bdev_cachep = kmem_cache_create("bdev_cache", sizeof(struct bdev_inode),
 			0, (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT|
-				SLAB_MEM_SPREAD|SLAB_ACCOUNT|SLAB_PANIC),
+				SLAB_ACCOUNT|SLAB_PANIC),
 			init_once);
 	err = register_filesystem(&bd_type);
 	if (err)
diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c
index 2c90e5de0acd..d442ee358fc2 100644
--- a/block/bfq-cgroup.c
+++ b/block/bfq-cgroup.c
@@ -127,7 +127,7 @@ static void bfqg_stats_update_group_wait_time(struct bfqg_stats *stats)
 	if (!bfqg_stats_waiting(stats))
 		return;
 
-	now = ktime_get_ns();
+	now = blk_time_get_ns();
 	if (now > stats->start_group_wait_time)
 		bfq_stat_add(&stats->group_wait_time,
 			      now - stats->start_group_wait_time);
@@ -144,7 +144,7 @@ static void bfqg_stats_set_start_group_wait_time(struct bfq_group *bfqg,
 		return;
 	if (bfqg == curr_bfqg)
 		return;
-	stats->start_group_wait_time = ktime_get_ns();
+	stats->start_group_wait_time = blk_time_get_ns();
 	bfqg_stats_mark_waiting(stats);
 }
 
@@ -156,7 +156,7 @@ static void bfqg_stats_end_empty_time(struct bfqg_stats *stats)
 	if (!bfqg_stats_empty(stats))
 		return;
 
-	now = ktime_get_ns();
+	now = blk_time_get_ns();
 	if (now > stats->start_empty_time)
 		bfq_stat_add(&stats->empty_time,
 			      now - stats->start_empty_time);
@@ -183,7 +183,7 @@ void bfqg_stats_set_start_empty_time(struct bfq_group *bfqg)
 	if (bfqg_stats_empty(stats))
 		return;
 
-	stats->start_empty_time = ktime_get_ns();
+	stats->start_empty_time = blk_time_get_ns();
 	bfqg_stats_mark_empty(stats);
 }
 
@@ -192,7 +192,7 @@ void bfqg_stats_update_idle_time(struct bfq_group *bfqg)
 	struct bfqg_stats *stats = &bfqg->stats;
 
 	if (bfqg_stats_idling(stats)) {
-		u64 now = ktime_get_ns();
+		u64 now = blk_time_get_ns();
 
 		if (now > stats->start_idle_time)
 			bfq_stat_add(&stats->idle_time,
@@ -205,7 +205,7 @@ void bfqg_stats_set_start_idle_time(struct bfq_group *bfqg)
 {
 	struct bfqg_stats *stats = &bfqg->stats;
 
-	stats->start_idle_time = ktime_get_ns();
+	stats->start_idle_time = blk_time_get_ns();
 	bfqg_stats_mark_idling(stats);
 }
 
@@ -242,7 +242,7 @@ void bfqg_stats_update_completion(struct bfq_group *bfqg, u64 start_time_ns,
 				  u64 io_start_time_ns, blk_opf_t opf)
 {
 	struct bfqg_stats *stats = &bfqg->stats;
-	u64 now = ktime_get_ns();
+	u64 now = blk_time_get_ns();
 
 	if (now > io_start_time_ns)
 		blkg_rwstat_add(&stats->service_time, opf,
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 3cce6de464a7..4b88a54a9b76 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -1005,7 +1005,7 @@ static struct request *bfq_check_fifo(struct bfq_queue *bfqq,
 
 	rq = rq_entry_fifo(bfqq->fifo.next);
 
-	if (rq == last || ktime_get_ns() < rq->fifo_time)
+	if (rq == last || blk_time_get_ns() < rq->fifo_time)
 		return NULL;
 
 	bfq_log_bfqq(bfqq->bfqd, bfqq, "check_fifo: returned %p", rq);
@@ -1829,7 +1829,7 @@ static void bfq_bfqq_handle_idle_busy_switch(struct bfq_data *bfqd,
 		 * bfq_bfqq_update_budg_for_activation for
 		 * details on the usage of the next variable.
 		 */
-		arrived_in_time =  ktime_get_ns() <=
+		arrived_in_time =  blk_time_get_ns() <=
 			bfqq->ttime.last_end_request +
 			bfqd->bfq_slice_idle * 3;
 	unsigned int act_idx = bfq_actuator_index(bfqd, rq->bio);
@@ -2208,7 +2208,7 @@ static void bfq_add_request(struct request *rq)
 	struct request *next_rq, *prev;
 	unsigned int old_wr_coeff = bfqq->wr_coeff;
 	bool interactive = false;
-	u64 now_ns = ktime_get_ns();
+	u64 now_ns = blk_time_get_ns();
 
 	bfq_log_bfqq(bfqd, bfqq, "add_request %d", rq_is_sync(rq));
 	bfqq->queued[rq_is_sync(rq)]++;
@@ -2262,7 +2262,7 @@ static void bfq_add_request(struct request *rq)
 		      bfqd->rqs_injected && bfqd->tot_rq_in_driver > 0)) &&
 		    time_is_before_eq_jiffies(bfqq->decrease_time_jif +
 					      msecs_to_jiffies(10))) {
-			bfqd->last_empty_occupied_ns = ktime_get_ns();
+			bfqd->last_empty_occupied_ns = blk_time_get_ns();
 			/*
 			 * Start the state machine for measuring the
 			 * total service time of rq: setting
@@ -3294,7 +3294,7 @@ static void bfq_set_budget_timeout(struct bfq_data *bfqd,
 	else
 		timeout_coeff = bfqq->entity.weight / bfqq->entity.orig_weight;
 
-	bfqd->last_budget_start = ktime_get();
+	bfqd->last_budget_start = blk_time_get();
 
 	bfqq->budget_timeout = jiffies +
 		bfqd->bfq_timeout * timeout_coeff;
@@ -3394,7 +3394,7 @@ static void bfq_arm_slice_timer(struct bfq_data *bfqd)
 	else if (bfqq->wr_coeff > 1)
 		sl = max_t(u32, sl, 20ULL * NSEC_PER_MSEC);
 
-	bfqd->last_idling_start = ktime_get();
+	bfqd->last_idling_start = blk_time_get();
 	bfqd->last_idling_start_jiffies = jiffies;
 
 	hrtimer_start(&bfqd->idle_slice_timer, ns_to_ktime(sl),
@@ -3433,7 +3433,7 @@ static void bfq_reset_rate_computation(struct bfq_data *bfqd,
 				       struct request *rq)
 {
 	if (rq != NULL) { /* new rq dispatch now, reset accordingly */
-		bfqd->last_dispatch = bfqd->first_dispatch = ktime_get_ns();
+		bfqd->last_dispatch = bfqd->first_dispatch = blk_time_get_ns();
 		bfqd->peak_rate_samples = 1;
 		bfqd->sequential_samples = 0;
 		bfqd->tot_sectors_dispatched = bfqd->last_rq_max_size =
@@ -3590,7 +3590,7 @@ reset_computation:
  */
 static void bfq_update_peak_rate(struct bfq_data *bfqd, struct request *rq)
 {
-	u64 now_ns = ktime_get_ns();
+	u64 now_ns = blk_time_get_ns();
 
 	if (bfqd->peak_rate_samples == 0) { /* first dispatch */
 		bfq_log(bfqd, "update_peak_rate: goto reset, samples %d",
@@ -4162,7 +4162,7 @@ static bool bfq_bfqq_is_slow(struct bfq_data *bfqd, struct bfq_queue *bfqq,
 	if (compensate)
 		delta_ktime = bfqd->last_idling_start;
 	else
-		delta_ktime = ktime_get();
+		delta_ktime = blk_time_get();
 	delta_ktime = ktime_sub(delta_ktime, bfqd->last_budget_start);
 	delta_usecs = ktime_to_us(delta_ktime);
 
@@ -5591,7 +5591,7 @@ static void bfq_init_bfqq(struct bfq_data *bfqd, struct bfq_queue *bfqq,
 			  struct bfq_io_cq *bic, pid_t pid, int is_sync,
 			  unsigned int act_idx)
 {
-	u64 now_ns = ktime_get_ns();
+	u64 now_ns = blk_time_get_ns();
 
 	bfqq->actuator_idx = act_idx;
 	RB_CLEAR_NODE(&bfqq->entity.rb_node);
@@ -5903,7 +5903,7 @@ static void bfq_update_io_thinktime(struct bfq_data *bfqd,
 	 */
 	if (bfqq->dispatched || bfq_bfqq_busy(bfqq))
 		return;
-	elapsed = ktime_get_ns() - bfqq->ttime.last_end_request;
+	elapsed = blk_time_get_ns() - bfqq->ttime.last_end_request;
 	elapsed = min_t(u64, elapsed, 2ULL * bfqd->bfq_slice_idle);
 
 	ttime->ttime_samples = (7*ttime->ttime_samples + 256) / 8;
@@ -6194,7 +6194,7 @@ static bool __bfq_insert_request(struct bfq_data *bfqd, struct request *rq)
 	bfq_add_request(rq);
 	idle_timer_disabled = waiting && !bfq_bfqq_wait_request(bfqq);
 
-	rq->fifo_time = ktime_get_ns() + bfqd->bfq_fifo_expire[rq_is_sync(rq)];
+	rq->fifo_time = blk_time_get_ns() + bfqd->bfq_fifo_expire[rq_is_sync(rq)];
 	list_add_tail(&rq->queuelist, &bfqq->fifo);
 
 	bfq_rq_enqueued(bfqd, bfqq, rq);
@@ -6370,7 +6370,7 @@ static void bfq_completed_request(struct bfq_queue *bfqq, struct bfq_data *bfqd)
 		bfq_weights_tree_remove(bfqq);
 	}
 
-	now_ns = ktime_get_ns();
+	now_ns = blk_time_get_ns();
 
 	bfqq->ttime.last_end_request = now_ns;
 
@@ -6585,7 +6585,7 @@ static void bfq_completed_request(struct bfq_queue *bfqq, struct bfq_data *bfqd)
 static void bfq_update_inject_limit(struct bfq_data *bfqd,
 				    struct bfq_queue *bfqq)
 {
-	u64 tot_time_ns = ktime_get_ns() - bfqd->last_empty_occupied_ns;
+	u64 tot_time_ns = blk_time_get_ns() - bfqd->last_empty_occupied_ns;
 	unsigned int old_limit = bfqq->inject_limit;
 
 	if (bfqq->last_serv_time_ns > 0 && bfqd->rqs_injected) {
diff --git a/block/bio-integrity.c b/block/bio-integrity.c
index c9a16fba58b9..2e3e8e04961e 100644
--- a/block/bio-integrity.c
+++ b/block/bio-integrity.c
@@ -395,6 +395,7 @@ static blk_status_t bio_integrity_process(struct bio *bio,
 	iter.tuple_size = bi->tuple_size;
 	iter.seed = proc_iter->bi_sector;
 	iter.prot_buf = bvec_virt(bip->bip_vec);
+	iter.pi_offset = bi->pi_offset;
 
 	__bio_for_each_segment(bv, bio, bviter, *proc_iter) {
 		void *kaddr = bvec_kmap_local(&bv);
diff --git a/block/bio.c b/block/bio.c
index c9223e9d31da..d24420ed1c4c 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -16,7 +16,6 @@
 #include <linux/workqueue.h>
 #include <linux/cgroup.h>
 #include <linux/highmem.h>
-#include <linux/sched/sysctl.h>
 #include <linux/blk-crypto.h>
 #include <linux/xarray.h>
 
@@ -763,29 +762,31 @@ static inline void bio_put_percpu_cache(struct bio *bio)
 	struct bio_alloc_cache *cache;
 
 	cache = per_cpu_ptr(bio->bi_pool->cache, get_cpu());
-	if (READ_ONCE(cache->nr_irq) + cache->nr > ALLOC_CACHE_MAX) {
-		put_cpu();
-		bio_free(bio);
-		return;
-	}
+	if (READ_ONCE(cache->nr_irq) + cache->nr > ALLOC_CACHE_MAX)
+		goto out_free;
 
-	bio_uninit(bio);
-
-	if ((bio->bi_opf & REQ_POLLED) && !WARN_ON_ONCE(in_interrupt())) {
+	if (in_task()) {
+		bio_uninit(bio);
 		bio->bi_next = cache->free_list;
+		/* Not necessary but helps not to iopoll already freed bios */
 		bio->bi_bdev = NULL;
 		cache->free_list = bio;
 		cache->nr++;
-	} else {
-		unsigned long flags;
+	} else if (in_hardirq()) {
+		lockdep_assert_irqs_disabled();
 
-		local_irq_save(flags);
+		bio_uninit(bio);
 		bio->bi_next = cache->free_list_irq;
 		cache->free_list_irq = bio;
 		cache->nr_irq++;
-		local_irq_restore(flags);
+	} else {
+		goto out_free;
 	}
 	put_cpu();
+	return;
+out_free:
+	put_cpu();
+	bio_free(bio);
 }
 
 /**
@@ -1154,7 +1155,7 @@ void __bio_release_pages(struct bio *bio, bool mark_dirty)
 
 	bio_for_each_folio_all(fi, bio) {
 		struct page *page;
-		size_t done = 0;
+		size_t nr_pages;
 
 		if (mark_dirty) {
 			folio_lock(fi.folio);
@@ -1162,10 +1163,11 @@ void __bio_release_pages(struct bio *bio, bool mark_dirty)
 			folio_unlock(fi.folio);
 		}
 		page = folio_page(fi.folio, fi.offset / PAGE_SIZE);
+		nr_pages = (fi.offset + fi.length - 1) / PAGE_SIZE -
+			   fi.offset / PAGE_SIZE + 1;
 		do {
 			bio_release_page(bio, page++);
-			done += PAGE_SIZE;
-		} while (done < fi.length);
+		} while (--nr_pages != 0);
 	}
 }
 EXPORT_SYMBOL_GPL(__bio_release_pages);
@@ -1371,21 +1373,12 @@ int submit_bio_wait(struct bio *bio)
 {
 	DECLARE_COMPLETION_ONSTACK_MAP(done,
 			bio->bi_bdev->bd_disk->lockdep_map);
-	unsigned long hang_check;
 
 	bio->bi_private = &done;
 	bio->bi_end_io = submit_bio_wait_endio;
 	bio->bi_opf |= REQ_SYNC;
 	submit_bio(bio);
-
-	/* Prevent hang_check timer from firing at us during very long I/O */
-	hang_check = sysctl_hung_task_timeout_secs;
-	if (hang_check)
-		while (!wait_for_completion_io_timeout(&done,
-					hang_check * (HZ/2)))
-			;
-	else
-		wait_for_completion_io(&done);
+	blk_wait_io(&done);
 
 	return blk_status_to_errno(bio->bi_status);
 }
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index ff93c385ba5a..bdbb557feb5a 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -1846,7 +1846,7 @@ static void blkcg_maybe_throttle_blkg(struct blkcg_gq *blkg, bool use_memdelay)
 {
 	unsigned long pflags;
 	bool clamp;
-	u64 now = ktime_to_ns(ktime_get());
+	u64 now = blk_time_get_ns();
 	u64 exp;
 	u64 delay_nsec = 0;
 	int tok;
diff --git a/block/blk-cgroup.h b/block/blk-cgroup.h
index b927a4a0ad03..78b74106bf10 100644
--- a/block/blk-cgroup.h
+++ b/block/blk-cgroup.h
@@ -19,6 +19,7 @@
 #include <linux/kthread.h>
 #include <linux/blk-mq.h>
 #include <linux/llist.h>
+#include "blk.h"
 
 struct blkcg_gq;
 struct blkg_policy_data;
diff --git a/block/blk-core.c b/block/blk-core.c
index de771093b526..a16b5abdbbf5 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -394,24 +394,34 @@ static void blk_timeout_work(struct work_struct *work)
 {
 }
 
-struct request_queue *blk_alloc_queue(int node_id)
+struct request_queue *blk_alloc_queue(struct queue_limits *lim, int node_id)
 {
 	struct request_queue *q;
+	int error;
 
 	q = kmem_cache_alloc_node(blk_requestq_cachep, GFP_KERNEL | __GFP_ZERO,
 				  node_id);
 	if (!q)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	q->last_merge = NULL;
 
 	q->id = ida_alloc(&blk_queue_ida, GFP_KERNEL);
-	if (q->id < 0)
+	if (q->id < 0) {
+		error = q->id;
 		goto fail_q;
+	}
 
 	q->stats = blk_alloc_queue_stats();
-	if (!q->stats)
+	if (!q->stats) {
+		error = -ENOMEM;
 		goto fail_id;
+	}
+
+	error = blk_set_default_limits(lim);
+	if (error)
+		goto fail_stats;
+	q->limits = *lim;
 
 	q->node = node_id;
 
@@ -425,6 +435,7 @@ struct request_queue *blk_alloc_queue(int node_id)
 	mutex_init(&q->debugfs_mutex);
 	mutex_init(&q->sysfs_lock);
 	mutex_init(&q->sysfs_dir_lock);
+	mutex_init(&q->limits_lock);
 	mutex_init(&q->rq_qos_mutex);
 	spin_lock_init(&q->queue_lock);
 
@@ -435,12 +446,12 @@ struct request_queue *blk_alloc_queue(int node_id)
 	 * Init percpu_ref in atomic mode so that it's faster to shutdown.
 	 * See blk_register_queue() for details.
 	 */
-	if (percpu_ref_init(&q->q_usage_counter,
+	error = percpu_ref_init(&q->q_usage_counter,
 				blk_queue_usage_counter_release,
-				PERCPU_REF_INIT_ATOMIC, GFP_KERNEL))
+				PERCPU_REF_INIT_ATOMIC, GFP_KERNEL);
+	if (error)
 		goto fail_stats;
 
-	blk_set_default_limits(&q->limits);
 	q->nr_requests = BLKDEV_DEFAULT_RQ;
 
 	return q;
@@ -451,7 +462,7 @@ fail_id:
 	ida_free(&blk_queue_ida, q->id);
 fail_q:
 	kmem_cache_free(blk_requestq_cachep, q);
-	return NULL;
+	return ERR_PTR(error);
 }
 
 /**
@@ -1083,6 +1094,7 @@ void blk_start_plug_nr_ios(struct blk_plug *plug, unsigned short nr_ios)
 	if (tsk->plug)
 		return;
 
+	plug->cur_ktime = 0;
 	plug->mq_list = NULL;
 	plug->cached_rq = NULL;
 	plug->nr_ios = min_t(unsigned short, nr_ios, BLK_MAX_REQUEST_COUNT);
@@ -1182,6 +1194,8 @@ void __blk_flush_plug(struct blk_plug *plug, bool from_schedule)
 	 */
 	if (unlikely(!rq_list_empty(plug->cached_rq)))
 		blk_mq_free_plug_rqs(plug);
+
+	current->flags &= ~PF_BLOCK_TS;
 }
 
 /**
@@ -1229,8 +1243,7 @@ int __init blk_dev_init(void)
 	if (!kblockd_workqueue)
 		panic("Failed to create kblockd\n");
 
-	blk_requestq_cachep = kmem_cache_create("request_queue",
-			sizeof(struct request_queue), 0, SLAB_PANIC, NULL);
+	blk_requestq_cachep = KMEM_CACHE(request_queue, SLAB_PANIC);
 
 	blk_debugfs_root = debugfs_create_dir("block", NULL);
 
diff --git a/block/blk-flush.c b/block/blk-flush.c
index 3f4d41952ef2..b0f314f4bc14 100644
--- a/block/blk-flush.c
+++ b/block/blk-flush.c
@@ -143,7 +143,7 @@ static void blk_account_io_flush(struct request *rq)
 	part_stat_lock();
 	part_stat_inc(part, ios[STAT_FLUSH]);
 	part_stat_add(part, nsecs[STAT_FLUSH],
-		      ktime_get_ns() - rq->start_time_ns);
+		      blk_time_get_ns() - rq->start_time_ns);
 	part_stat_unlock();
 }
 
diff --git a/block/blk-integrity.c b/block/blk-integrity.c
index d4e9b4556d14..ccbeb6dfa87a 100644
--- a/block/blk-integrity.c
+++ b/block/blk-integrity.c
@@ -370,6 +370,7 @@ void blk_integrity_register(struct gendisk *disk, struct blk_integrity *template
 	bi->profile = template->profile ? template->profile : &nop_profile;
 	bi->tuple_size = template->tuple_size;
 	bi->tag_size = template->tag_size;
+	bi->pi_offset = template->pi_offset;
 
 	blk_queue_flag_set(QUEUE_FLAG_STABLE_WRITES, disk->queue);
 
diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index 04d44f0bcbc8..9a85bfbbc45a 100644
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -829,7 +829,7 @@ static int ioc_autop_idx(struct ioc *ioc, struct gendisk *disk)
 
 	/* step up/down based on the vrate */
 	vrate_pct = div64_u64(ioc->vtime_base_rate * 100, VTIME_PER_USEC);
-	now_ns = ktime_get_ns();
+	now_ns = blk_time_get_ns();
 
 	if (p->too_fast_vrate_pct && p->too_fast_vrate_pct <= vrate_pct) {
 		if (!ioc->autop_too_fast_at)
@@ -1044,7 +1044,7 @@ static void ioc_now(struct ioc *ioc, struct ioc_now *now)
 	unsigned seq;
 	u64 vrate;
 
-	now->now_ns = ktime_get();
+	now->now_ns = blk_time_get_ns();
 	now->now = ktime_to_us(now->now_ns);
 	vrate = atomic64_read(&ioc->vtime_rate);
 
@@ -2817,7 +2817,7 @@ static void ioc_rqos_done(struct rq_qos *rqos, struct request *rq)
 		return;
 	}
 
-	on_q_ns = ktime_get_ns() - rq->alloc_time_ns;
+	on_q_ns = blk_time_get_ns() - rq->alloc_time_ns;
 	rq_wait_ns = rq->start_time_ns - rq->alloc_time_ns;
 	size_nsec = div64_u64(calc_size_vtime_cost(rq, ioc), VTIME_PER_NSEC);
 
@@ -2900,7 +2900,7 @@ static int blk_iocost_init(struct gendisk *disk)
 	ioc->vtime_base_rate = VTIME_PER_USEC;
 	atomic64_set(&ioc->vtime_rate, VTIME_PER_USEC);
 	seqcount_spinlock_init(&ioc->period_seqcount, &ioc->lock);
-	ioc->period_at = ktime_to_us(ktime_get());
+	ioc->period_at = ktime_to_us(blk_time_get());
 	atomic64_set(&ioc->cur_period, 0);
 	atomic_set(&ioc->hweight_gen, 0);
 
diff --git a/block/blk-iolatency.c b/block/blk-iolatency.c
index c1a6aba1d59e..ebb522788d97 100644
--- a/block/blk-iolatency.c
+++ b/block/blk-iolatency.c
@@ -609,7 +609,7 @@ static void blkcg_iolatency_done_bio(struct rq_qos *rqos, struct bio *bio)
 	if (!iolat->blkiolat->enabled)
 		return;
 
-	now = ktime_to_ns(ktime_get());
+	now = blk_time_get_ns();
 	while (blkg && blkg->parent) {
 		iolat = blkg_to_lat(blkg);
 		if (!iolat) {
@@ -661,7 +661,7 @@ static void blkiolatency_timer_fn(struct timer_list *t)
 	struct blk_iolatency *blkiolat = from_timer(blkiolat, t, timer);
 	struct blkcg_gq *blkg;
 	struct cgroup_subsys_state *pos_css;
-	u64 now = ktime_to_ns(ktime_get());
+	u64 now = blk_time_get_ns();
 
 	rcu_read_lock();
 	blkg_for_each_descendant_pre(blkg, pos_css,
@@ -985,7 +985,7 @@ static void iolatency_pd_init(struct blkg_policy_data *pd)
 	struct blkcg_gq *blkg = lat_to_blkg(iolat);
 	struct rq_qos *rqos = iolat_rq_qos(blkg->q);
 	struct blk_iolatency *blkiolat = BLKIOLATENCY(rqos);
-	u64 now = ktime_to_ns(ktime_get());
+	u64 now = blk_time_get_ns();
 	int cpu;
 
 	if (blk_queue_nonrot(blkg->q))
diff --git a/block/blk-lib.c b/block/blk-lib.c
index e59c3069e835..dc8e35d0a51d 100644
--- a/block/blk-lib.c
+++ b/block/blk-lib.c
@@ -35,6 +35,26 @@ static sector_t bio_discard_limit(struct block_device *bdev, sector_t sector)
 	return round_down(UINT_MAX, discard_granularity) >> SECTOR_SHIFT;
 }
 
+static void await_bio_endio(struct bio *bio)
+{
+	complete(bio->bi_private);
+	bio_put(bio);
+}
+
+/*
+ * await_bio_chain - ends @bio and waits for every chained bio to complete
+ */
+static void await_bio_chain(struct bio *bio)
+{
+	DECLARE_COMPLETION_ONSTACK_MAP(done,
+			bio->bi_bdev->bd_disk->lockdep_map);
+
+	bio->bi_private = &done;
+	bio->bi_end_io = await_bio_endio;
+	bio_endio(bio);
+	blk_wait_io(&done);
+}
+
 int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 		sector_t nr_sects, gfp_t gfp_mask, struct bio **biop)
 {
@@ -77,6 +97,10 @@ int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 		 * is disabled.
 		 */
 		cond_resched();
+		if (fatal_signal_pending(current)) {
+			await_bio_chain(bio);
+			return -EINTR;
+		}
 	}
 
 	*biop = bio;
@@ -120,32 +144,33 @@ static int __blkdev_issue_write_zeroes(struct block_device *bdev,
 		struct bio **biop, unsigned flags)
 {
 	struct bio *bio = *biop;
-	unsigned int max_write_zeroes_sectors;
+	unsigned int max_sectors;
 
 	if (bdev_read_only(bdev))
 		return -EPERM;
 
-	/* Ensure that max_write_zeroes_sectors doesn't overflow bi_size */
-	max_write_zeroes_sectors = bdev_write_zeroes_sectors(bdev);
+	/* Ensure that max_sectors doesn't overflow bi_size */
+	max_sectors = bdev_write_zeroes_sectors(bdev);
 
-	if (max_write_zeroes_sectors == 0)
+	if (max_sectors == 0)
 		return -EOPNOTSUPP;
 
 	while (nr_sects) {
+		unsigned int len = min_t(sector_t, nr_sects, max_sectors);
+
 		bio = blk_next_bio(bio, bdev, 0, REQ_OP_WRITE_ZEROES, gfp_mask);
 		bio->bi_iter.bi_sector = sector;
 		if (flags & BLKDEV_ZERO_NOUNMAP)
 			bio->bi_opf |= REQ_NOUNMAP;
 
-		if (nr_sects > max_write_zeroes_sectors) {
-			bio->bi_iter.bi_size = max_write_zeroes_sectors << 9;
-			nr_sects -= max_write_zeroes_sectors;
-			sector += max_write_zeroes_sectors;
-		} else {
-			bio->bi_iter.bi_size = nr_sects << 9;
-			nr_sects = 0;
-		}
+		bio->bi_iter.bi_size = len << SECTOR_SHIFT;
+		nr_sects -= len;
+		sector += len;
 		cond_resched();
+		if (fatal_signal_pending(current)) {
+			await_bio_chain(bio);
+			return -EINTR;
+		}
 	}
 
 	*biop = bio;
@@ -190,6 +215,10 @@ static int __blkdev_issue_zero_pages(struct block_device *bdev,
 				break;
 		}
 		cond_resched();
+		if (fatal_signal_pending(current)) {
+			await_bio_chain(bio);
+			return -EINTR;
+		}
 	}
 
 	*biop = bio;
@@ -280,7 +309,7 @@ retry:
 		bio_put(bio);
 	}
 	blk_finish_plug(&plug);
-	if (ret && try_write_zeroes) {
+	if (ret && ret != -EINTR && try_write_zeroes) {
 		if (!(flags & BLKDEV_ZERO_NOFALLBACK)) {
 			try_write_zeroes = false;
 			goto retry;
@@ -322,7 +351,7 @@ int blkdev_issue_secure_erase(struct block_device *bdev, sector_t sector,
 		return -EPERM;
 
 	blk_start_plug(&plug);
-	for (;;) {
+	while (nr_sects) {
 		unsigned int len = min_t(sector_t, nr_sects, max_sectors);
 
 		bio = blk_next_bio(bio, bdev, 0, REQ_OP_SECURE_ERASE, gfp);
@@ -331,12 +360,17 @@ int blkdev_issue_secure_erase(struct block_device *bdev, sector_t sector,
 
 		sector += len;
 		nr_sects -= len;
-		if (!nr_sects) {
-			ret = submit_bio_wait(bio);
-			bio_put(bio);
+		cond_resched();
+		if (fatal_signal_pending(current)) {
+			await_bio_chain(bio);
+			ret = -EINTR;
+			bio = NULL;
 			break;
 		}
-		cond_resched();
+	}
+	if (bio) {
+		ret = submit_bio_wait(bio);
+		bio_put(bio);
 	}
 	blk_finish_plug(&plug);
 
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 9c475a89e3af..555ada922cf0 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -21,7 +21,6 @@
 #include <linux/llist.h>
 #include <linux/cpu.h>
 #include <linux/cache.h>
-#include <linux/sched/sysctl.h>
 #include <linux/sched/topology.h>
 #include <linux/sched/signal.h>
 #include <linux/delay.h>
@@ -322,7 +321,7 @@ void blk_rq_init(struct request_queue *q, struct request *rq)
 	RB_CLEAR_NODE(&rq->rb_node);
 	rq->tag = BLK_MQ_NO_TAG;
 	rq->internal_tag = BLK_MQ_NO_TAG;
-	rq->start_time_ns = ktime_get_ns();
+	rq->start_time_ns = blk_time_get_ns();
 	rq->part = NULL;
 	blk_crypto_rq_set_defaults(rq);
 }
@@ -332,7 +331,7 @@ EXPORT_SYMBOL(blk_rq_init);
 static inline void blk_mq_rq_time_init(struct request *rq, u64 alloc_time_ns)
 {
 	if (blk_mq_need_time_stamp(rq))
-		rq->start_time_ns = ktime_get_ns();
+		rq->start_time_ns = blk_time_get_ns();
 	else
 		rq->start_time_ns = 0;
 
@@ -443,7 +442,7 @@ static struct request *__blk_mq_alloc_requests(struct blk_mq_alloc_data *data)
 
 	/* alloc_time includes depth and tag waits */
 	if (blk_queue_rq_alloc_time(q))
-		alloc_time_ns = ktime_get_ns();
+		alloc_time_ns = blk_time_get_ns();
 
 	if (data->cmd_flags & REQ_NOWAIT)
 		data->flags |= BLK_MQ_REQ_NOWAIT;
@@ -628,7 +627,7 @@ struct request *blk_mq_alloc_request_hctx(struct request_queue *q,
 
 	/* alloc_time includes depth and tag waits */
 	if (blk_queue_rq_alloc_time(q))
-		alloc_time_ns = ktime_get_ns();
+		alloc_time_ns = blk_time_get_ns();
 
 	/*
 	 * If the tag allocator sleeps we could get an allocation for a
@@ -1041,7 +1040,7 @@ static inline void __blk_mq_end_request_acct(struct request *rq, u64 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());
+		__blk_mq_end_request_acct(rq, blk_time_get_ns());
 
 	blk_mq_finish_request(rq);
 
@@ -1084,7 +1083,7 @@ void blk_mq_end_request_batch(struct io_comp_batch *iob)
 	u64 now = 0;
 
 	if (iob->need_ts)
-		now = ktime_get_ns();
+		now = blk_time_get_ns();
 
 	while ((rq = rq_list_pop(&iob->req_list)) != NULL) {
 		prefetch(rq->bio);
@@ -1167,10 +1166,11 @@ static inline bool blk_mq_complete_need_ipi(struct request *rq)
 	if (force_irqthreads())
 		return false;
 
-	/* same CPU or cache domain?  Complete locally */
+	/* same CPU or cache domain and capacity?  Complete locally */
 	if (cpu == rq->mq_ctx->cpu ||
 	    (!test_bit(QUEUE_FLAG_SAME_FORCE, &rq->q->queue_flags) &&
-	     cpus_share_cache(cpu, rq->mq_ctx->cpu)))
+	     cpus_share_cache(cpu, rq->mq_ctx->cpu) &&
+	     cpus_equal_capacity(cpu, rq->mq_ctx->cpu)))
 		return false;
 
 	/* don't try to IPI to an offline CPU */
@@ -1254,7 +1254,7 @@ void blk_mq_start_request(struct request *rq)
 
 	if (test_bit(QUEUE_FLAG_STATS, &q->queue_flags) &&
 	    !blk_rq_is_passthrough(rq)) {
-		rq->io_start_time_ns = ktime_get_ns();
+		rq->io_start_time_ns = blk_time_get_ns();
 		rq->stats_sectors = blk_rq_sectors(rq);
 		rq->rq_flags |= RQF_STATS;
 		rq_qos_issue(q, rq);
@@ -1409,22 +1409,10 @@ blk_status_t blk_execute_rq(struct request *rq, bool at_head)
 	blk_mq_insert_request(rq, at_head ? BLK_MQ_INSERT_AT_HEAD : 0);
 	blk_mq_run_hw_queue(hctx, false);
 
-	if (blk_rq_is_poll(rq)) {
+	if (blk_rq_is_poll(rq))
 		blk_rq_poll_completion(rq, &wait.done);
-	} else {
-		/*
-		 * Prevent hang_check timer from firing at us during very long
-		 * I/O
-		 */
-		unsigned long hang_check = sysctl_hung_task_timeout_secs;
-
-		if (hang_check)
-			while (!wait_for_completion_io_timeout(&wait.done,
-					hang_check * (HZ/2)))
-				;
-		else
-			wait_for_completion_io(&wait.done);
-	}
+	else
+		blk_wait_io(&wait.done);
 
 	return wait.ret;
 }
@@ -2892,9 +2880,6 @@ static struct request *blk_mq_get_new_requests(struct request_queue *q,
 	};
 	struct request *rq;
 
-	if (blk_mq_attempt_bio_merge(q, bio, nsegs))
-		return NULL;
-
 	rq_qos_throttle(q, bio);
 
 	if (plug) {
@@ -2913,22 +2898,31 @@ static struct request *blk_mq_get_new_requests(struct request_queue *q,
 }
 
 /*
- * Check if we can use the passed on request for submitting the passed in bio,
- * and remove it from the request list if it can be used.
+ * Check if there is a suitable cached request and return it.
  */
-static bool blk_mq_use_cached_rq(struct request *rq, struct blk_plug *plug,
-		struct bio *bio)
+static struct request *blk_mq_peek_cached_request(struct blk_plug *plug,
+		struct request_queue *q, blk_opf_t opf)
 {
-	enum hctx_type type = blk_mq_get_hctx_type(bio->bi_opf);
-	enum hctx_type hctx_type = rq->mq_hctx->type;
+	enum hctx_type type = blk_mq_get_hctx_type(opf);
+	struct request *rq;
 
-	WARN_ON_ONCE(rq_list_peek(&plug->cached_rq) != rq);
+	if (!plug)
+		return NULL;
+	rq = rq_list_peek(&plug->cached_rq);
+	if (!rq || rq->q != q)
+		return NULL;
+	if (type != rq->mq_hctx->type &&
+	    (type != HCTX_TYPE_READ || rq->mq_hctx->type != HCTX_TYPE_DEFAULT))
+		return NULL;
+	if (op_is_flush(rq->cmd_flags) != op_is_flush(opf))
+		return NULL;
+	return rq;
+}
 
-	if (type != hctx_type &&
-	    !(type == HCTX_TYPE_READ && hctx_type == HCTX_TYPE_DEFAULT))
-		return false;
-	if (op_is_flush(rq->cmd_flags) != op_is_flush(bio->bi_opf))
-		return false;
+static void blk_mq_use_cached_rq(struct request *rq, struct blk_plug *plug,
+		struct bio *bio)
+{
+	WARN_ON_ONCE(rq_list_peek(&plug->cached_rq) != rq);
 
 	/*
 	 * If any qos ->throttle() end up blocking, we will have flushed the
@@ -2941,7 +2935,6 @@ static bool blk_mq_use_cached_rq(struct request *rq, struct blk_plug *plug,
 	blk_mq_rq_time_init(rq, 0);
 	rq->cmd_flags = bio->bi_opf;
 	INIT_LIST_HEAD(&rq->queuelist);
-	return true;
 }
 
 /**
@@ -2963,50 +2956,43 @@ void blk_mq_submit_bio(struct bio *bio)
 	struct blk_plug *plug = blk_mq_plug(bio);
 	const int is_sync = op_is_sync(bio->bi_opf);
 	struct blk_mq_hw_ctx *hctx;
-	struct request *rq = NULL;
 	unsigned int nr_segs = 1;
+	struct request *rq;
 	blk_status_t ret;
 
 	bio = blk_queue_bounce(bio, q);
 
-	if (plug) {
-		rq = rq_list_peek(&plug->cached_rq);
-		if (rq && rq->q != q)
-			rq = NULL;
-	}
-	if (rq) {
-		if (unlikely(bio_may_exceed_limits(bio, &q->limits))) {
-			bio = __bio_split_to_limits(bio, &q->limits, &nr_segs);
-			if (!bio)
-				return;
-		}
-		if (!bio_integrity_prep(bio))
-			return;
-		if (blk_mq_attempt_bio_merge(q, bio, nr_segs))
-			return;
-		if (blk_mq_use_cached_rq(rq, plug, bio))
-			goto done;
-		percpu_ref_get(&q->q_usage_counter);
-	} else {
+	/*
+	 * If the plug has a cached request for this queue, try use it.
+	 *
+	 * The cached request already holds a q_usage_counter reference and we
+	 * don't have to acquire a new one if we use it.
+	 */
+	rq = blk_mq_peek_cached_request(plug, q, bio->bi_opf);
+	if (!rq) {
 		if (unlikely(bio_queue_enter(bio)))
 			return;
-		if (unlikely(bio_may_exceed_limits(bio, &q->limits))) {
-			bio = __bio_split_to_limits(bio, &q->limits, &nr_segs);
-			if (!bio)
-				goto fail;
-		}
-		if (!bio_integrity_prep(bio))
-			goto fail;
 	}
 
-	rq = blk_mq_get_new_requests(q, plug, bio, nr_segs);
-	if (unlikely(!rq)) {
-fail:
-		blk_queue_exit(q);
-		return;
+	if (unlikely(bio_may_exceed_limits(bio, &q->limits))) {
+		bio = __bio_split_to_limits(bio, &q->limits, &nr_segs);
+		if (!bio)
+			goto queue_exit;
+	}
+	if (!bio_integrity_prep(bio))
+		goto queue_exit;
+
+	if (blk_mq_attempt_bio_merge(q, bio, nr_segs))
+		goto queue_exit;
+
+	if (!rq) {
+		rq = blk_mq_get_new_requests(q, plug, bio, nr_segs);
+		if (unlikely(!rq))
+			goto queue_exit;
+	} else {
+		blk_mq_use_cached_rq(rq, plug, bio);
 	}
 
-done:
 	trace_block_getrq(bio);
 
 	rq_qos_track(q, rq, bio);
@@ -3037,6 +3023,15 @@ done:
 	} else {
 		blk_mq_run_dispatch_ops(q, blk_mq_try_issue_directly(hctx, rq));
 	}
+	return;
+
+queue_exit:
+	/*
+	 * Don't drop the queue reference if we were trying to use a cached
+	 * request and thus didn't acquire one.
+	 */
+	if (!rq)
+		blk_queue_exit(q);
 }
 
 #ifdef CONFIG_BLK_MQ_STACKING
@@ -3098,7 +3093,7 @@ blk_status_t blk_insert_cloned_request(struct request *rq)
 	blk_mq_run_dispatch_ops(q,
 			ret = blk_mq_request_issue_directly(rq, true));
 	if (ret)
-		blk_account_io_done(rq, ktime_get_ns());
+		blk_account_io_done(rq, blk_time_get_ns());
 	return ret;
 }
 EXPORT_SYMBOL_GPL(blk_insert_cloned_request);
@@ -4078,15 +4073,16 @@ void blk_mq_release(struct request_queue *q)
 	blk_mq_sysfs_deinit(q);
 }
 
-static struct request_queue *blk_mq_init_queue_data(struct blk_mq_tag_set *set,
-		void *queuedata)
+struct request_queue *blk_mq_alloc_queue(struct blk_mq_tag_set *set,
+		struct queue_limits *lim, void *queuedata)
 {
+	struct queue_limits default_lim = { };
 	struct request_queue *q;
 	int ret;
 
-	q = blk_alloc_queue(set->numa_node);
-	if (!q)
-		return ERR_PTR(-ENOMEM);
+	q = blk_alloc_queue(lim ? lim : &default_lim, set->numa_node);
+	if (IS_ERR(q))
+		return q;
 	q->queuedata = queuedata;
 	ret = blk_mq_init_allocated_queue(set, q);
 	if (ret) {
@@ -4095,20 +4091,15 @@ static struct request_queue *blk_mq_init_queue_data(struct blk_mq_tag_set *set,
 	}
 	return q;
 }
-
-struct request_queue *blk_mq_init_queue(struct blk_mq_tag_set *set)
-{
-	return blk_mq_init_queue_data(set, NULL);
-}
-EXPORT_SYMBOL(blk_mq_init_queue);
+EXPORT_SYMBOL(blk_mq_alloc_queue);
 
 /**
  * blk_mq_destroy_queue - shutdown a request queue
  * @q: request queue to shutdown
  *
- * This shuts down a request queue allocated by blk_mq_init_queue(). All future
+ * This shuts down a request queue allocated by blk_mq_alloc_queue(). All future
  * requests will be failed with -ENODEV. The caller is responsible for dropping
- * the reference from blk_mq_init_queue() by calling blk_put_queue().
+ * the reference from blk_mq_alloc_queue() by calling blk_put_queue().
  *
  * Context: can sleep
  */
@@ -4129,13 +4120,14 @@ void blk_mq_destroy_queue(struct request_queue *q)
 }
 EXPORT_SYMBOL(blk_mq_destroy_queue);
 
-struct gendisk *__blk_mq_alloc_disk(struct blk_mq_tag_set *set, void *queuedata,
+struct gendisk *__blk_mq_alloc_disk(struct blk_mq_tag_set *set,
+		struct queue_limits *lim, void *queuedata,
 		struct lock_class_key *lkclass)
 {
 	struct request_queue *q;
 	struct gendisk *disk;
 
-	q = blk_mq_init_queue_data(set, queuedata);
+	q = blk_mq_alloc_queue(set, lim, queuedata);
 	if (IS_ERR(q))
 		return ERR_CAST(q);
 
@@ -4389,7 +4381,7 @@ static void blk_mq_update_queue_map(struct blk_mq_tag_set *set)
 	if (set->nr_maps == 1)
 		set->map[HCTX_TYPE_DEFAULT].nr_queues = set->nr_hw_queues;
 
-	if (set->ops->map_queues && !is_kdump_kernel()) {
+	if (set->ops->map_queues) {
 		int i;
 
 		/*
@@ -4488,14 +4480,12 @@ int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set)
 
 	/*
 	 * If a crashdump is active, then we are potentially in a very
-	 * memory constrained environment. Limit us to 1 queue and
-	 * 64 tags to prevent using too much memory.
+	 * memory constrained environment. Limit us to  64 tags to prevent
+	 * using too much memory.
 	 */
-	if (is_kdump_kernel()) {
-		set->nr_hw_queues = 1;
-		set->nr_maps = 1;
+	if (is_kdump_kernel())
 		set->queue_depth = min(64U, set->queue_depth);
-	}
+
 	/*
 	 * There is no use for more h/w queues than cpus if we just have
 	 * a single map
@@ -4525,7 +4515,7 @@ int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set)
 						  GFP_KERNEL, set->numa_node);
 		if (!set->map[i].mq_map)
 			goto out_free_mq_map;
-		set->map[i].nr_queues = is_kdump_kernel() ? 1 : set->nr_hw_queues;
+		set->map[i].nr_queues = set->nr_hw_queues;
 	}
 
 	blk_mq_update_queue_map(set);
diff --git a/block/blk-settings.c b/block/blk-settings.c
index 06ea91e51b8b..3c7d8d638ab5 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -26,52 +26,21 @@ void blk_queue_rq_timeout(struct request_queue *q, unsigned int timeout)
 EXPORT_SYMBOL_GPL(blk_queue_rq_timeout);
 
 /**
- * blk_set_default_limits - reset limits to default values
- * @lim:  the queue_limits structure to reset
- *
- * Description:
- *   Returns a queue_limit struct to its default state.
- */
-void blk_set_default_limits(struct queue_limits *lim)
-{
-	lim->max_segments = BLK_MAX_SEGMENTS;
-	lim->max_discard_segments = 1;
-	lim->max_integrity_segments = 0;
-	lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
-	lim->virt_boundary_mask = 0;
-	lim->max_segment_size = BLK_MAX_SEGMENT_SIZE;
-	lim->max_sectors = lim->max_hw_sectors = BLK_SAFE_MAX_SECTORS;
-	lim->max_user_sectors = lim->max_dev_sectors = 0;
-	lim->chunk_sectors = 0;
-	lim->max_write_zeroes_sectors = 0;
-	lim->max_zone_append_sectors = 0;
-	lim->max_discard_sectors = 0;
-	lim->max_hw_discard_sectors = 0;
-	lim->max_secure_erase_sectors = 0;
-	lim->discard_granularity = 512;
-	lim->discard_alignment = 0;
-	lim->discard_misaligned = 0;
-	lim->logical_block_size = lim->physical_block_size = lim->io_min = 512;
-	lim->bounce = BLK_BOUNCE_NONE;
-	lim->alignment_offset = 0;
-	lim->io_opt = 0;
-	lim->misaligned = 0;
-	lim->zoned = false;
-	lim->zone_write_granularity = 0;
-	lim->dma_alignment = 511;
-}
-
-/**
  * blk_set_stacking_limits - set default limits for stacking devices
  * @lim:  the queue_limits structure to reset
  *
- * Description:
- *   Returns a queue_limit struct to its default state. Should be used
- *   by stacking drivers like DM that have no internal limits.
+ * Prepare queue limits for applying limits from underlying devices using
+ * blk_stack_limits().
  */
 void blk_set_stacking_limits(struct queue_limits *lim)
 {
-	blk_set_default_limits(lim);
+	memset(lim, 0, sizeof(*lim));
+	lim->logical_block_size = SECTOR_SIZE;
+	lim->physical_block_size = SECTOR_SIZE;
+	lim->io_min = SECTOR_SIZE;
+	lim->discard_granularity = SECTOR_SIZE;
+	lim->dma_alignment = SECTOR_SIZE - 1;
+	lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
 
 	/* Inherit limits from component devices */
 	lim->max_segments = USHRT_MAX;
@@ -82,9 +51,239 @@ void blk_set_stacking_limits(struct queue_limits *lim)
 	lim->max_dev_sectors = UINT_MAX;
 	lim->max_write_zeroes_sectors = UINT_MAX;
 	lim->max_zone_append_sectors = UINT_MAX;
+	lim->max_user_discard_sectors = UINT_MAX;
 }
 EXPORT_SYMBOL(blk_set_stacking_limits);
 
+static void blk_apply_bdi_limits(struct backing_dev_info *bdi,
+		struct queue_limits *lim)
+{
+	/*
+	 * For read-ahead of large files to be effective, we need to read ahead
+	 * at least twice the optimal I/O size.
+	 */
+	bdi->ra_pages = max(lim->io_opt * 2 / PAGE_SIZE, VM_READAHEAD_PAGES);
+	bdi->io_pages = lim->max_sectors >> PAGE_SECTORS_SHIFT;
+}
+
+static int blk_validate_zoned_limits(struct queue_limits *lim)
+{
+	if (!lim->zoned) {
+		if (WARN_ON_ONCE(lim->max_open_zones) ||
+		    WARN_ON_ONCE(lim->max_active_zones) ||
+		    WARN_ON_ONCE(lim->zone_write_granularity) ||
+		    WARN_ON_ONCE(lim->max_zone_append_sectors))
+			return -EINVAL;
+		return 0;
+	}
+
+	if (WARN_ON_ONCE(!IS_ENABLED(CONFIG_BLK_DEV_ZONED)))
+		return -EINVAL;
+
+	if (lim->zone_write_granularity < lim->logical_block_size)
+		lim->zone_write_granularity = lim->logical_block_size;
+
+	if (lim->max_zone_append_sectors) {
+		/*
+		 * The Zone Append size is limited by the maximum I/O size
+		 * and the zone size given that it can't span zones.
+		 */
+		lim->max_zone_append_sectors =
+			min3(lim->max_hw_sectors,
+			     lim->max_zone_append_sectors,
+			     lim->chunk_sectors);
+	}
+
+	return 0;
+}
+
+/*
+ * Check that the limits in lim are valid, initialize defaults for unset
+ * values, and cap values based on others where needed.
+ */
+static int blk_validate_limits(struct queue_limits *lim)
+{
+	unsigned int max_hw_sectors;
+
+	/*
+	 * Unless otherwise specified, default to 512 byte logical blocks and a
+	 * physical block size equal to the logical block size.
+	 */
+	if (!lim->logical_block_size)
+		lim->logical_block_size = SECTOR_SIZE;
+	if (lim->physical_block_size < lim->logical_block_size)
+		lim->physical_block_size = lim->logical_block_size;
+
+	/*
+	 * The minimum I/O size defaults to the physical block size unless
+	 * explicitly overridden.
+	 */
+	if (lim->io_min < lim->physical_block_size)
+		lim->io_min = lim->physical_block_size;
+
+	/*
+	 * max_hw_sectors has a somewhat weird default for historical reason,
+	 * but driver really should set their own instead of relying on this
+	 * value.
+	 *
+	 * The block layer relies on the fact that every driver can
+	 * handle at lest a page worth of data per I/O, and needs the value
+	 * aligned to the logical block size.
+	 */
+	if (!lim->max_hw_sectors)
+		lim->max_hw_sectors = BLK_SAFE_MAX_SECTORS;
+	if (WARN_ON_ONCE(lim->max_hw_sectors < PAGE_SECTORS))
+		return -EINVAL;
+	lim->max_hw_sectors = round_down(lim->max_hw_sectors,
+			lim->logical_block_size >> SECTOR_SHIFT);
+
+	/*
+	 * The actual max_sectors value is a complex beast and also takes the
+	 * max_dev_sectors value (set by SCSI ULPs) and a user configurable
+	 * value into account.  The ->max_sectors value is always calculated
+	 * from these, so directly setting it won't have any effect.
+	 */
+	max_hw_sectors = min_not_zero(lim->max_hw_sectors,
+				lim->max_dev_sectors);
+	if (lim->max_user_sectors) {
+		if (lim->max_user_sectors > max_hw_sectors ||
+		    lim->max_user_sectors < PAGE_SIZE / SECTOR_SIZE)
+			return -EINVAL;
+		lim->max_sectors = min(max_hw_sectors, lim->max_user_sectors);
+	} else {
+		lim->max_sectors = min(max_hw_sectors, BLK_DEF_MAX_SECTORS_CAP);
+	}
+	lim->max_sectors = round_down(lim->max_sectors,
+			lim->logical_block_size >> SECTOR_SHIFT);
+
+	/*
+	 * Random default for the maximum number of segments.  Driver should not
+	 * rely on this and set their own.
+	 */
+	if (!lim->max_segments)
+		lim->max_segments = BLK_MAX_SEGMENTS;
+
+	lim->max_discard_sectors =
+		min(lim->max_hw_discard_sectors, lim->max_user_discard_sectors);
+
+	if (!lim->max_discard_segments)
+		lim->max_discard_segments = 1;
+
+	if (lim->discard_granularity < lim->physical_block_size)
+		lim->discard_granularity = lim->physical_block_size;
+
+	/*
+	 * By default there is no limit on the segment boundary alignment,
+	 * but if there is one it can't be smaller than the page size as
+	 * that would break all the normal I/O patterns.
+	 */
+	if (!lim->seg_boundary_mask)
+		lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
+	if (WARN_ON_ONCE(lim->seg_boundary_mask < PAGE_SIZE - 1))
+		return -EINVAL;
+
+	/*
+	 * Devices that require a virtual boundary do not support scatter/gather
+	 * I/O natively, but instead require a descriptor list entry for each
+	 * page (which might not be identical to the Linux PAGE_SIZE).  Because
+	 * of that they are not limited by our notion of "segment size".
+	 */
+	if (lim->virt_boundary_mask) {
+		if (WARN_ON_ONCE(lim->max_segment_size &&
+				 lim->max_segment_size != UINT_MAX))
+			return -EINVAL;
+		lim->max_segment_size = UINT_MAX;
+	} else {
+		/*
+		 * The maximum segment size has an odd historic 64k default that
+		 * drivers probably should override.  Just like the I/O size we
+		 * require drivers to at least handle a full page per segment.
+		 */
+		if (!lim->max_segment_size)
+			lim->max_segment_size = BLK_MAX_SEGMENT_SIZE;
+		if (WARN_ON_ONCE(lim->max_segment_size < PAGE_SIZE))
+			return -EINVAL;
+	}
+
+	/*
+	 * We require drivers to at least do logical block aligned I/O, but
+	 * historically could not check for that due to the separate calls
+	 * to set the limits.  Once the transition is finished the check
+	 * below should be narrowed down to check the logical block size.
+	 */
+	if (!lim->dma_alignment)
+		lim->dma_alignment = SECTOR_SIZE - 1;
+	if (WARN_ON_ONCE(lim->dma_alignment > PAGE_SIZE))
+		return -EINVAL;
+
+	if (lim->alignment_offset) {
+		lim->alignment_offset &= (lim->physical_block_size - 1);
+		lim->misaligned = 0;
+	}
+
+	return blk_validate_zoned_limits(lim);
+}
+
+/*
+ * Set the default limits for a newly allocated queue.  @lim contains the
+ * initial limits set by the driver, which could be no limit in which case
+ * all fields are cleared to zero.
+ */
+int blk_set_default_limits(struct queue_limits *lim)
+{
+	/*
+	 * Most defaults are set by capping the bounds in blk_validate_limits,
+	 * but max_user_discard_sectors is special and needs an explicit
+	 * initialization to the max value here.
+	 */
+	lim->max_user_discard_sectors = UINT_MAX;
+	return blk_validate_limits(lim);
+}
+
+/**
+ * queue_limits_commit_update - commit an atomic update of queue limits
+ * @q:		queue to update
+ * @lim:	limits to apply
+ *
+ * Apply the limits in @lim that were obtained from queue_limits_start_update()
+ * and updated by the caller to @q.
+ *
+ * Returns 0 if successful, else a negative error code.
+ */
+int queue_limits_commit_update(struct request_queue *q,
+		struct queue_limits *lim)
+	__releases(q->limits_lock)
+{
+	int error = blk_validate_limits(lim);
+
+	if (!error) {
+		q->limits = *lim;
+		if (q->disk)
+			blk_apply_bdi_limits(q->disk->bdi, lim);
+	}
+	mutex_unlock(&q->limits_lock);
+	return error;
+}
+EXPORT_SYMBOL_GPL(queue_limits_commit_update);
+
+/**
+ * queue_limits_set - apply queue limits to queue
+ * @q:		queue to update
+ * @lim:	limits to apply
+ *
+ * Apply the limits in @lim that were freshly initialized to @q.
+ * To update existing limits use queue_limits_start_update() and
+ * queue_limits_commit_update() instead.
+ *
+ * Returns 0 if successful, else a negative error code.
+ */
+int queue_limits_set(struct request_queue *q, struct queue_limits *lim)
+{
+	mutex_lock(&q->limits_lock);
+	return queue_limits_commit_update(q, lim);
+}
+EXPORT_SYMBOL_GPL(queue_limits_set);
+
 /**
  * blk_queue_bounce_limit - set bounce buffer limit for queue
  * @q: the request queue for the device
@@ -177,8 +376,11 @@ EXPORT_SYMBOL(blk_queue_chunk_sectors);
 void blk_queue_max_discard_sectors(struct request_queue *q,
 		unsigned int max_discard_sectors)
 {
-	q->limits.max_hw_discard_sectors = max_discard_sectors;
-	q->limits.max_discard_sectors = max_discard_sectors;
+	struct queue_limits *lim = &q->limits;
+
+	lim->max_hw_discard_sectors = max_discard_sectors;
+	lim->max_discard_sectors =
+		min(max_discard_sectors, lim->max_user_discard_sectors);
 }
 EXPORT_SYMBOL(blk_queue_max_discard_sectors);
 
@@ -393,15 +595,7 @@ EXPORT_SYMBOL(blk_queue_alignment_offset);
 
 void disk_update_readahead(struct gendisk *disk)
 {
-	struct request_queue *q = disk->queue;
-
-	/*
-	 * For read-ahead of large files to be effective, we need to read ahead
-	 * at least twice the optimal I/O size.
-	 */
-	disk->bdi->ra_pages =
-		max(queue_io_opt(q) * 2 / PAGE_SIZE, VM_READAHEAD_PAGES);
-	disk->bdi->io_pages = queue_max_sectors(q) >> (PAGE_SHIFT - 9);
+	blk_apply_bdi_limits(disk->bdi, &disk->queue->limits);
 }
 EXPORT_SYMBOL_GPL(disk_update_readahead);
 
@@ -689,33 +883,38 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
 	t->zone_write_granularity = max(t->zone_write_granularity,
 					b->zone_write_granularity);
 	t->zoned = max(t->zoned, b->zoned);
+	if (!t->zoned) {
+		t->zone_write_granularity = 0;
+		t->max_zone_append_sectors = 0;
+	}
 	return ret;
 }
 EXPORT_SYMBOL(blk_stack_limits);
 
 /**
- * disk_stack_limits - adjust queue limits for stacked drivers
- * @disk:  MD/DM gendisk (top)
+ * queue_limits_stack_bdev - adjust queue_limits for stacked devices
+ * @t:	the stacking driver limits (top device)
  * @bdev:  the underlying block device (bottom)
  * @offset:  offset to beginning of data within component device
+ * @pfx: prefix to use for warnings logged
  *
  * Description:
- *    Merges the limits for a top level gendisk and a bottom level
- *    block_device.
+ *    This function is used by stacking drivers like MD and DM to ensure
+ *    that all component devices have compatible block sizes and
+ *    alignments.  The stacking driver must provide a queue_limits
+ *    struct (top) and then iteratively call the stacking function for
+ *    all component (bottom) devices.  The stacking function will
+ *    attempt to combine the values and ensure proper alignment.
  */
-void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
-		       sector_t offset)
+void queue_limits_stack_bdev(struct queue_limits *t, struct block_device *bdev,
+		sector_t offset, const char *pfx)
 {
-	struct request_queue *t = disk->queue;
-
-	if (blk_stack_limits(&t->limits, &bdev_get_queue(bdev)->limits,
-			get_start_sect(bdev) + (offset >> 9)) < 0)
+	if (blk_stack_limits(t, &bdev_get_queue(bdev)->limits,
+			get_start_sect(bdev) + offset))
 		pr_notice("%s: Warning: Device %pg is misaligned\n",
-			disk->disk_name, bdev);
-
-	disk_update_readahead(disk);
+			pfx, bdev);
 }
-EXPORT_SYMBOL(disk_stack_limits);
+EXPORT_SYMBOL_GPL(queue_limits_stack_bdev);
 
 /**
  * blk_queue_update_dma_pad - update pad mask
diff --git a/block/blk-stat.c b/block/blk-stat.c
index 7ff76ae6c76a..e42c263e53fb 100644
--- a/block/blk-stat.c
+++ b/block/blk-stat.c
@@ -27,7 +27,7 @@ void blk_rq_stat_init(struct blk_rq_stat *stat)
 /* src is a per-cpu stat, mean isn't initialized */
 void blk_rq_stat_sum(struct blk_rq_stat *dst, struct blk_rq_stat *src)
 {
-	if (!src->nr_samples)
+	if (dst->nr_samples + src->nr_samples <= dst->nr_samples)
 		return;
 
 	dst->min = min(dst->min, src->min);
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index 6b2429cad81a..8c8f69d8ba48 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -174,23 +174,29 @@ static ssize_t queue_discard_max_show(struct request_queue *q, char *page)
 static ssize_t queue_discard_max_store(struct request_queue *q,
 				       const char *page, size_t count)
 {
-	unsigned long max_discard;
-	ssize_t ret = queue_var_store(&max_discard, page, count);
+	unsigned long max_discard_bytes;
+	struct queue_limits lim;
+	ssize_t ret;
+	int err;
 
+	ret = queue_var_store(&max_discard_bytes, page, count);
 	if (ret < 0)
 		return ret;
 
-	if (max_discard & (q->limits.discard_granularity - 1))
+	if (max_discard_bytes & (q->limits.discard_granularity - 1))
 		return -EINVAL;
 
-	max_discard >>= 9;
-	if (max_discard > UINT_MAX)
+	if ((max_discard_bytes >> SECTOR_SHIFT) > UINT_MAX)
 		return -EINVAL;
 
-	if (max_discard > q->limits.max_hw_discard_sectors)
-		max_discard = q->limits.max_hw_discard_sectors;
+	blk_mq_freeze_queue(q);
+	lim = queue_limits_start_update(q);
+	lim.max_user_discard_sectors = max_discard_bytes >> SECTOR_SHIFT;
+	err = queue_limits_commit_update(q, &lim);
+	blk_mq_unfreeze_queue(q);
 
-	q->limits.max_discard_sectors = max_discard;
+	if (err)
+		return err;
 	return ret;
 }
 
@@ -226,35 +232,22 @@ static ssize_t queue_zone_append_max_show(struct request_queue *q, char *page)
 static ssize_t
 queue_max_sectors_store(struct request_queue *q, const char *page, size_t count)
 {
-	unsigned long var;
-	unsigned int max_sectors_kb,
-		max_hw_sectors_kb = queue_max_hw_sectors(q) >> 1,
-			page_kb = 1 << (PAGE_SHIFT - 10);
-	ssize_t ret = queue_var_store(&var, page, count);
+	unsigned long max_sectors_kb;
+	struct queue_limits lim;
+	ssize_t ret;
+	int err;
 
+	ret = queue_var_store(&max_sectors_kb, page, count);
 	if (ret < 0)
 		return ret;
 
-	max_sectors_kb = (unsigned int)var;
-	max_hw_sectors_kb = min_not_zero(max_hw_sectors_kb,
-					 q->limits.max_dev_sectors >> 1);
-	if (max_sectors_kb == 0) {
-		q->limits.max_user_sectors = 0;
-		max_sectors_kb = min(max_hw_sectors_kb,
-				     BLK_DEF_MAX_SECTORS_CAP >> 1);
-	} else {
-		if (max_sectors_kb > max_hw_sectors_kb ||
-		    max_sectors_kb < page_kb)
-			return -EINVAL;
-		q->limits.max_user_sectors = max_sectors_kb << 1;
-	}
-
-	spin_lock_irq(&q->queue_lock);
-	q->limits.max_sectors = max_sectors_kb << 1;
-	if (q->disk)
-		q->disk->bdi->io_pages = max_sectors_kb >> (PAGE_SHIFT - 10);
-	spin_unlock_irq(&q->queue_lock);
-
+	blk_mq_freeze_queue(q);
+	lim = queue_limits_start_update(q);
+	lim.max_user_sectors = max_sectors_kb << 1;
+	err = queue_limits_commit_update(q, &lim);
+	blk_mq_unfreeze_queue(q);
+	if (err)
+		return err;
 	return ret;
 }
 
diff --git a/block/blk-throttle.c b/block/blk-throttle.c
index 16f5766620a4..f4850a6f860b 100644
--- a/block/blk-throttle.c
+++ b/block/blk-throttle.c
@@ -1098,7 +1098,7 @@ static int throtl_dispatch_tg(struct throtl_grp *tg)
 	while ((bio = throtl_peek_queued(&sq->queued[READ])) &&
 	       tg_may_dispatch(tg, bio, NULL)) {
 
-		tg_dispatch_one_bio(tg, bio_data_dir(bio));
+		tg_dispatch_one_bio(tg, READ);
 		nr_reads++;
 
 		if (nr_reads >= max_nr_reads)
@@ -1108,7 +1108,7 @@ static int throtl_dispatch_tg(struct throtl_grp *tg)
 	while ((bio = throtl_peek_queued(&sq->queued[WRITE])) &&
 	       tg_may_dispatch(tg, bio, NULL)) {
 
-		tg_dispatch_one_bio(tg, bio_data_dir(bio));
+		tg_dispatch_one_bio(tg, WRITE);
 		nr_writes++;
 
 		if (nr_writes >= max_nr_writes)
@@ -1815,7 +1815,7 @@ static bool throtl_tg_is_idle(struct throtl_grp *tg)
 	time = min_t(unsigned long, MAX_IDLE_TIME, 4 * tg->idletime_threshold);
 	ret = tg->latency_target == DFL_LATENCY_TARGET ||
 	      tg->idletime_threshold == DFL_IDLE_THRESHOLD ||
-	      (ktime_get_ns() >> 10) - tg->last_finish_time > time ||
+	      (blk_time_get_ns() >> 10) - tg->last_finish_time > time ||
 	      tg->avg_idletime > tg->idletime_threshold ||
 	      (tg->latency_target && tg->bio_cnt &&
 		tg->bad_bio_cnt * 5 < tg->bio_cnt);
@@ -2060,7 +2060,7 @@ static void blk_throtl_update_idletime(struct throtl_grp *tg)
 	if (last_finish_time == 0)
 		return;
 
-	now = ktime_get_ns() >> 10;
+	now = blk_time_get_ns() >> 10;
 	if (now <= last_finish_time ||
 	    last_finish_time == tg->checked_last_finish_time)
 		return;
@@ -2327,7 +2327,7 @@ void blk_throtl_bio_endio(struct bio *bio)
 	if (!tg->td->limit_valid[LIMIT_LOW])
 		return;
 
-	finish_time_ns = ktime_get_ns();
+	finish_time_ns = blk_time_get_ns();
 	tg->last_finish_time = finish_time_ns >> 10;
 
 	start_time = bio_issue_time(&bio->bi_issue) >> 10;
diff --git a/block/blk-wbt.c b/block/blk-wbt.c
index 0c0e270a8265..64472134dd26 100644
--- a/block/blk-wbt.c
+++ b/block/blk-wbt.c
@@ -29,6 +29,7 @@
 #include "blk-wbt.h"
 #include "blk-rq-qos.h"
 #include "elevator.h"
+#include "blk.h"
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/wbt.h>
@@ -274,13 +275,12 @@ static inline bool stat_sample_valid(struct blk_rq_stat *stat)
 
 static u64 rwb_sync_issue_lat(struct rq_wb *rwb)
 {
-	u64 now, issue = READ_ONCE(rwb->sync_issue);
+	u64 issue = READ_ONCE(rwb->sync_issue);
 
 	if (!issue || !rwb->sync_cookie)
 		return 0;
 
-	now = ktime_to_ns(ktime_get());
-	return now - issue;
+	return blk_time_get_ns() - issue;
 }
 
 static inline unsigned int wbt_inflight(struct rq_wb *rwb)
diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index d343e5756a9c..da0f4b2a8fa0 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -11,7 +11,6 @@
 
 #include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/rbtree.h>
 #include <linux/blkdev.h>
 #include <linux/blk-mq.h>
 #include <linux/mm.h>
@@ -177,8 +176,7 @@ static int blk_zone_need_reset_cb(struct blk_zone *zone, unsigned int idx,
 	}
 }
 
-static int blkdev_zone_reset_all_emulated(struct block_device *bdev,
-					  gfp_t gfp_mask)
+static int blkdev_zone_reset_all_emulated(struct block_device *bdev)
 {
 	struct gendisk *disk = bdev->bd_disk;
 	sector_t capacity = bdev_nr_sectors(bdev);
@@ -205,7 +203,7 @@ static int blkdev_zone_reset_all_emulated(struct block_device *bdev,
 		}
 
 		bio = blk_next_bio(bio, bdev, 0, REQ_OP_ZONE_RESET | REQ_SYNC,
-				   gfp_mask);
+				   GFP_KERNEL);
 		bio->bi_iter.bi_sector = sector;
 		sector += zone_sectors;
 
@@ -223,7 +221,7 @@ out_free_need_reset:
 	return ret;
 }
 
-static int blkdev_zone_reset_all(struct block_device *bdev, gfp_t gfp_mask)
+static int blkdev_zone_reset_all(struct block_device *bdev)
 {
 	struct bio bio;
 
@@ -238,7 +236,6 @@ static int blkdev_zone_reset_all(struct block_device *bdev, gfp_t gfp_mask)
  * @sector:	Start sector of the first zone to operate on
  * @nr_sectors:	Number of sectors, should be at least the length of one zone and
  *		must be zone size aligned.
- * @gfp_mask:	Memory allocation flags (for bio_alloc)
  *
  * Description:
  *    Perform the specified operation on the range of zones specified by
@@ -248,7 +245,7 @@ static int blkdev_zone_reset_all(struct block_device *bdev, gfp_t gfp_mask)
  *    or finish request.
  */
 int blkdev_zone_mgmt(struct block_device *bdev, enum req_op op,
-		     sector_t sector, sector_t nr_sectors, gfp_t gfp_mask)
+		     sector_t sector, sector_t nr_sectors)
 {
 	struct request_queue *q = bdev_get_queue(bdev);
 	sector_t zone_sectors = bdev_zone_sectors(bdev);
@@ -285,12 +282,12 @@ int blkdev_zone_mgmt(struct block_device *bdev, enum req_op op,
 	 */
 	if (op == REQ_OP_ZONE_RESET && sector == 0 && nr_sectors == capacity) {
 		if (!blk_queue_zone_resetall(q))
-			return blkdev_zone_reset_all_emulated(bdev, gfp_mask);
-		return blkdev_zone_reset_all(bdev, gfp_mask);
+			return blkdev_zone_reset_all_emulated(bdev);
+		return blkdev_zone_reset_all(bdev);
 	}
 
 	while (sector < end_sector) {
-		bio = blk_next_bio(bio, bdev, 0, op | REQ_SYNC, gfp_mask);
+		bio = blk_next_bio(bio, bdev, 0, op | REQ_SYNC, GFP_KERNEL);
 		bio->bi_iter.bi_sector = sector;
 		sector += zone_sectors;
 
@@ -419,8 +416,7 @@ int blkdev_zone_mgmt_ioctl(struct block_device *bdev, blk_mode_t mode,
 		return -ENOTTY;
 	}
 
-	ret = blkdev_zone_mgmt(bdev, op, zrange.sector, zrange.nr_sectors,
-			       GFP_KERNEL);
+	ret = blkdev_zone_mgmt(bdev, op, zrange.sector, zrange.nr_sectors);
 
 fail:
 	if (cmd == BLKRESETZONE)
diff --git a/block/blk.h b/block/blk.h
index f02b25f22e8b..a19b7b42e650 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -4,6 +4,8 @@
 
 #include <linux/blk-crypto.h>
 #include <linux/memblock.h>	/* for max_pfn/max_low_pfn */
+#include <linux/sched/sysctl.h>
+#include <linux/timekeeping.h>
 #include <xen/xen.h>
 #include "blk-crypto-internal.h"
 
@@ -70,6 +72,18 @@ static inline int bio_queue_enter(struct bio *bio)
 	return __bio_queue_enter(q, bio);
 }
 
+static inline void blk_wait_io(struct completion *done)
+{
+	/* Prevent hang_check timer from firing at us during very long I/O */
+	unsigned long timeout = sysctl_hung_task_timeout_secs * HZ / 2;
+
+	if (timeout)
+		while (!wait_for_completion_io_timeout(done, timeout))
+			;
+	else
+		wait_for_completion_io(done);
+}
+
 #define BIO_INLINE_VECS 4
 struct bio_vec *bvec_alloc(mempool_t *pool, unsigned short *nr_vecs,
 		gfp_t gfp_mask);
@@ -329,7 +343,7 @@ void blk_rq_set_mixed_merge(struct request *rq);
 bool blk_rq_merge_ok(struct request *rq, struct bio *bio);
 enum elv_merge blk_try_merge(struct request *rq, struct bio *bio);
 
-void blk_set_default_limits(struct queue_limits *lim);
+int blk_set_default_limits(struct queue_limits *lim);
 int blk_dev_init(void);
 
 /*
@@ -447,7 +461,7 @@ static inline void bio_release_page(struct bio *bio, struct page *page)
 		unpin_user_page(page);
 }
 
-struct request_queue *blk_alloc_queue(int node_id);
+struct request_queue *blk_alloc_queue(struct queue_limits *lim, int node_id);
 
 int disk_scan_partitions(struct gendisk *disk, blk_mode_t mode);
 
@@ -516,8 +530,75 @@ static inline int req_ref_read(struct request *req)
 	return atomic_read(&req->ref);
 }
 
+static inline u64 blk_time_get_ns(void)
+{
+	struct blk_plug *plug = current->plug;
+
+	if (!plug)
+		return ktime_get_ns();
+
+	/*
+	 * 0 could very well be a valid time, but rather than flag "this is
+	 * a valid timestamp" separately, just accept that we'll do an extra
+	 * ktime_get_ns() if we just happen to get 0 as the current time.
+	 */
+	if (!plug->cur_ktime) {
+		plug->cur_ktime = ktime_get_ns();
+		current->flags |= PF_BLOCK_TS;
+	}
+	return plug->cur_ktime;
+}
+
+static inline ktime_t blk_time_get(void)
+{
+	return ns_to_ktime(blk_time_get_ns());
+}
+
+/*
+ * From most significant bit:
+ * 1 bit: reserved for other usage, see below
+ * 12 bits: original size of bio
+ * 51 bits: issue time of bio
+ */
+#define BIO_ISSUE_RES_BITS      1
+#define BIO_ISSUE_SIZE_BITS     12
+#define BIO_ISSUE_RES_SHIFT     (64 - BIO_ISSUE_RES_BITS)
+#define BIO_ISSUE_SIZE_SHIFT    (BIO_ISSUE_RES_SHIFT - BIO_ISSUE_SIZE_BITS)
+#define BIO_ISSUE_TIME_MASK     ((1ULL << BIO_ISSUE_SIZE_SHIFT) - 1)
+#define BIO_ISSUE_SIZE_MASK     \
+	(((1ULL << BIO_ISSUE_SIZE_BITS) - 1) << BIO_ISSUE_SIZE_SHIFT)
+#define BIO_ISSUE_RES_MASK      (~((1ULL << BIO_ISSUE_RES_SHIFT) - 1))
+
+/* Reserved bit for blk-throtl */
+#define BIO_ISSUE_THROTL_SKIP_LATENCY (1ULL << 63)
+
+static inline u64 __bio_issue_time(u64 time)
+{
+	return time & BIO_ISSUE_TIME_MASK;
+}
+
+static inline u64 bio_issue_time(struct bio_issue *issue)
+{
+	return __bio_issue_time(issue->value);
+}
+
+static inline sector_t bio_issue_size(struct bio_issue *issue)
+{
+	return ((issue->value & BIO_ISSUE_SIZE_MASK) >> BIO_ISSUE_SIZE_SHIFT);
+}
+
+static inline void bio_issue_init(struct bio_issue *issue,
+				       sector_t size)
+{
+	size &= (1ULL << BIO_ISSUE_SIZE_BITS) - 1;
+	issue->value = ((issue->value & BIO_ISSUE_RES_MASK) |
+			(blk_time_get_ns() & BIO_ISSUE_TIME_MASK) |
+			((u64)size << BIO_ISSUE_SIZE_SHIFT));
+}
+
 void bdev_release(struct file *bdev_file);
 int bdev_open(struct block_device *bdev, blk_mode_t mode, void *holder,
 	      const struct blk_holder_ops *hops, struct file *bdev_file);
 int bdev_permission(dev_t dev, blk_mode_t mode, void *holder);
+
 #endif /* BLK_INTERNAL_H */
diff --git a/block/bsg-lib.c b/block/bsg-lib.c
index b3acdbdb6e7e..bcc7dee6abce 100644
--- a/block/bsg-lib.c
+++ b/block/bsg-lib.c
@@ -383,7 +383,7 @@ struct request_queue *bsg_setup_queue(struct device *dev, const char *name,
 	if (blk_mq_alloc_tag_set(set))
 		goto out_tag_set;
 
-	q = blk_mq_init_queue(set);
+	q = blk_mq_alloc_queue(set, NULL, NULL);
 	if (IS_ERR(q)) {
 		ret = PTR_ERR(q);
 		goto out_queue;
diff --git a/block/genhd.c b/block/genhd.c
index a911d2969c07..bb29a68e1d67 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1201,7 +1201,7 @@ static int block_uevent(const struct device *dev, struct kobj_uevent_env *env)
 	return add_uevent_var(env, "DISKSEQ=%llu", disk->diskseq);
 }
 
-struct class block_class = {
+const struct class block_class = {
 	.name		= "block",
 	.dev_uevent	= block_uevent,
 };
@@ -1391,19 +1391,21 @@ out_free_disk:
 	return NULL;
 }
 
-struct gendisk *__blk_alloc_disk(int node, struct lock_class_key *lkclass)
+struct gendisk *__blk_alloc_disk(struct queue_limits *lim, int node,
+		struct lock_class_key *lkclass)
 {
+	struct queue_limits default_lim = { };
 	struct request_queue *q;
 	struct gendisk *disk;
 
-	q = blk_alloc_queue(node);
-	if (!q)
-		return NULL;
+	q = blk_alloc_queue(lim ? lim : &default_lim, node);
+	if (IS_ERR(q))
+		return ERR_CAST(q);
 
 	disk = __alloc_disk_node(q, node, lkclass);
 	if (!disk) {
 		blk_put_queue(q);
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 	}
 	set_bit(GD_OWNS_QUEUE, &disk->state);
 	return disk;
diff --git a/block/holder.c b/block/holder.c
index 37d18c13d958..791091a7eac2 100644
--- a/block/holder.c
+++ b/block/holder.c
@@ -8,6 +8,8 @@ struct bd_holder_disk {
 	int			refcnt;
 };
 
+static DEFINE_MUTEX(blk_holder_mutex);
+
 static struct bd_holder_disk *bd_find_holder_disk(struct block_device *bdev,
 						  struct gendisk *disk)
 {
@@ -80,7 +82,7 @@ int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk)
 	kobject_get(bdev->bd_holder_dir);
 	mutex_unlock(&bdev->bd_disk->open_mutex);
 
-	mutex_lock(&disk->open_mutex);
+	mutex_lock(&blk_holder_mutex);
 	WARN_ON_ONCE(!bdev->bd_holder);
 
 	holder = bd_find_holder_disk(bdev, disk);
@@ -108,7 +110,7 @@ int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk)
 		goto out_del_symlink;
 	list_add(&holder->list, &disk->slave_bdevs);
 
-	mutex_unlock(&disk->open_mutex);
+	mutex_unlock(&blk_holder_mutex);
 	return 0;
 
 out_del_symlink:
@@ -116,7 +118,7 @@ out_del_symlink:
 out_free_holder:
 	kfree(holder);
 out_unlock:
-	mutex_unlock(&disk->open_mutex);
+	mutex_unlock(&blk_holder_mutex);
 	if (ret)
 		kobject_put(bdev->bd_holder_dir);
 	return ret;
@@ -140,7 +142,7 @@ void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk)
 	if (WARN_ON_ONCE(!disk->slave_dir))
 		return;
 
-	mutex_lock(&disk->open_mutex);
+	mutex_lock(&blk_holder_mutex);
 	holder = bd_find_holder_disk(bdev, disk);
 	if (!WARN_ON_ONCE(holder == NULL) && !--holder->refcnt) {
 		del_symlink(disk->slave_dir, bdev_kobj(bdev));
@@ -149,6 +151,6 @@ void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk)
 		list_del_init(&holder->list);
 		kfree(holder);
 	}
-	mutex_unlock(&disk->open_mutex);
+	mutex_unlock(&blk_holder_mutex);
 }
 EXPORT_SYMBOL_GPL(bd_unlink_disk_holder);
diff --git a/block/ioctl.c b/block/ioctl.c
index 4c8aebee595f..0c76137adcaa 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -18,7 +18,7 @@ static int blkpg_do_ioctl(struct block_device *bdev,
 {
 	struct gendisk *disk = bdev->bd_disk;
 	struct blkpg_partition p;
-	sector_t start, length;
+	sector_t start, length, capacity, end;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
@@ -41,6 +41,13 @@ static int blkpg_do_ioctl(struct block_device *bdev,
 
 	start = p.start >> SECTOR_SHIFT;
 	length = p.length >> SECTOR_SHIFT;
+	capacity = get_capacity(disk);
+
+	if (check_add_overflow(start, length, &end))
+		return -EINVAL;
+
+	if (start >= capacity || end > capacity)
+		return -EINVAL;
 
 	switch (op) {
 	case BLKPG_ADD_PARTITION:
diff --git a/block/partitions/core.c b/block/partitions/core.c
index 5f5ed5c75f04..b11e88c82c8c 100644
--- a/block/partitions/core.c
+++ b/block/partitions/core.c
@@ -419,21 +419,10 @@ static bool partition_overlaps(struct gendisk *disk, sector_t start,
 int bdev_add_partition(struct gendisk *disk, int partno, sector_t start,
 		sector_t length)
 {
-	sector_t capacity = get_capacity(disk), end;
 	struct block_device *part;
 	int ret;
 
 	mutex_lock(&disk->open_mutex);
-	if (check_add_overflow(start, length, &end)) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	if (start >= capacity || end > capacity) {
-		ret = -EINVAL;
-		goto out;
-	}
-
 	if (!disk_live(disk)) {
 		ret = -ENXIO;
 		goto out;
diff --git a/block/partitions/mac.c b/block/partitions/mac.c
index 7b521df00a39..c80183156d68 100644
--- a/block/partitions/mac.c
+++ b/block/partitions/mac.c
@@ -20,6 +20,7 @@ extern void note_bootable_part(dev_t dev, int part, int goodness);
  * Code to understand MacOS partition tables.
  */
 
+#ifdef CONFIG_PPC_PMAC
 static inline void mac_fix_string(char *stg, int len)
 {
 	int i;
@@ -27,6 +28,7 @@ static inline void mac_fix_string(char *stg, int len)
 	for (i = len - 1; i >= 0 && stg[i] == ' '; i--)
 		stg[i] = 0;
 }
+#endif
 
 int mac_partition(struct parsed_partitions *state)
 {
diff --git a/block/sed-opal.c b/block/sed-opal.c
index fa4dba5d8531..14fe0fef811c 100644
--- a/block/sed-opal.c
+++ b/block/sed-opal.c
@@ -1212,7 +1212,7 @@ static int cmd_start(struct opal_dev *dev, const u8 *uid, const u8 *method)
 static int start_opal_session_cont(struct opal_dev *dev)
 {
 	u32 hsn, tsn;
-	int error = 0;
+	int error;
 
 	error = parse_and_check_status(dev);
 	if (error)
@@ -1354,7 +1354,7 @@ static int get_active_key_cont(struct opal_dev *dev)
 {
 	const char *activekey;
 	size_t keylen;
-	int error = 0;
+	int error;
 
 	error = parse_and_check_status(dev);
 	if (error)
@@ -2157,7 +2157,7 @@ static int lock_unlock_locking_range(struct opal_dev *dev, void *data)
 	u8 lr_buffer[OPAL_UID_LENGTH];
 	struct opal_lock_unlock *lkul = data;
 	u8 read_locked = 1, write_locked = 1;
-	int err = 0;
+	int err;
 
 	if (build_locking_range(lr_buffer, sizeof(lr_buffer),
 				lkul->session.opal_key.lr) < 0)
@@ -2580,7 +2580,7 @@ static int opal_get_discv(struct opal_dev *dev, struct opal_discovery *discv)
 	const struct opal_step discovery0_step = {
 		opal_discovery0, discv
 	};
-	int ret = 0;
+	int ret;
 
 	mutex_lock(&dev->dev_lock);
 	setup_opal_dev(dev);
@@ -3069,7 +3069,7 @@ bool opal_unlock_from_suspend(struct opal_dev *dev)
 {
 	struct opal_suspend_data *suspend;
 	bool was_failure = false;
-	int ret = 0;
+	int ret;
 
 	if (!dev)
 		return false;
@@ -3112,10 +3112,9 @@ static int opal_read_table(struct opal_dev *dev,
 		{ read_table_data, rw_tbl },
 		{ end_opal_session, }
 	};
-	int ret = 0;
 
 	if (!rw_tbl->size)
-		return ret;
+		return 0;
 
 	return execute_steps(dev, read_table_steps,
 			     ARRAY_SIZE(read_table_steps));
@@ -3129,10 +3128,9 @@ static int opal_write_table(struct opal_dev *dev,
 		{ write_table_data, rw_tbl },
 		{ end_opal_session, }
 	};
-	int ret = 0;
 
 	if (!rw_tbl->size)
-		return ret;
+		return 0;
 
 	return execute_steps(dev, write_table_steps,
 			     ARRAY_SIZE(write_table_steps));
diff --git a/block/t10-pi.c b/block/t10-pi.c
index 914d8cddd43a..d90892fd6f2a 100644
--- a/block/t10-pi.c
+++ b/block/t10-pi.c
@@ -12,14 +12,14 @@
 #include <net/checksum.h>
 #include <asm/unaligned.h>
 
-typedef __be16 (csum_fn) (void *, unsigned int);
+typedef __be16 (csum_fn) (__be16, void *, unsigned int);
 
-static __be16 t10_pi_crc_fn(void *data, unsigned int len)
+static __be16 t10_pi_crc_fn(__be16 crc, void *data, unsigned int len)
 {
-	return cpu_to_be16(crc_t10dif(data, len));
+	return cpu_to_be16(crc_t10dif_update(be16_to_cpu(crc), data, len));
 }
 
-static __be16 t10_pi_ip_fn(void *data, unsigned int len)
+static __be16 t10_pi_ip_fn(__be16 csum, void *data, unsigned int len)
 {
 	return (__force __be16)ip_compute_csum(data, len);
 }
@@ -32,12 +32,16 @@ static __be16 t10_pi_ip_fn(void *data, unsigned int len)
 static blk_status_t t10_pi_generate(struct blk_integrity_iter *iter,
 		csum_fn *fn, enum t10_dif_type type)
 {
+	u8 offset = iter->pi_offset;
 	unsigned int i;
 
 	for (i = 0 ; i < iter->data_size ; i += iter->interval) {
-		struct t10_pi_tuple *pi = iter->prot_buf;
+		struct t10_pi_tuple *pi = iter->prot_buf + offset;
 
-		pi->guard_tag = fn(iter->data_buf, iter->interval);
+		pi->guard_tag = fn(0, iter->data_buf, iter->interval);
+		if (offset)
+			pi->guard_tag = fn(pi->guard_tag, iter->prot_buf,
+					   offset);
 		pi->app_tag = 0;
 
 		if (type == T10_PI_TYPE1_PROTECTION)
@@ -56,12 +60,13 @@ static blk_status_t t10_pi_generate(struct blk_integrity_iter *iter,
 static blk_status_t t10_pi_verify(struct blk_integrity_iter *iter,
 		csum_fn *fn, enum t10_dif_type type)
 {
+	u8 offset = iter->pi_offset;
 	unsigned int i;
 
 	BUG_ON(type == T10_PI_TYPE0_PROTECTION);
 
 	for (i = 0 ; i < iter->data_size ; i += iter->interval) {
-		struct t10_pi_tuple *pi = iter->prot_buf;
+		struct t10_pi_tuple *pi = iter->prot_buf + offset;
 		__be16 csum;
 
 		if (type == T10_PI_TYPE1_PROTECTION ||
@@ -83,7 +88,9 @@ static blk_status_t t10_pi_verify(struct blk_integrity_iter *iter,
 				goto next;
 		}
 
-		csum = fn(iter->data_buf, iter->interval);
+		csum = fn(0, iter->data_buf, iter->interval);
+		if (offset)
+			csum = fn(csum, iter->prot_buf, offset);
 
 		if (pi->guard_tag != csum) {
 			pr_err("%s: guard tag error at sector %llu " \
@@ -134,8 +141,10 @@ static blk_status_t t10_pi_type1_verify_ip(struct blk_integrity_iter *iter)
  */
 static void t10_pi_type1_prepare(struct request *rq)
 {
-	const int tuple_sz = rq->q->integrity.tuple_size;
+	struct blk_integrity *bi = &rq->q->integrity;
+	const int tuple_sz = bi->tuple_size;
 	u32 ref_tag = t10_pi_ref_tag(rq);
+	u8 offset = bi->pi_offset;
 	struct bio *bio;
 
 	__rq_for_each_bio(bio, rq) {
@@ -154,7 +163,7 @@ static void t10_pi_type1_prepare(struct request *rq)
 
 			p = bvec_kmap_local(&iv);
 			for (j = 0; j < iv.bv_len; j += tuple_sz) {
-				struct t10_pi_tuple *pi = p;
+				struct t10_pi_tuple *pi = p + offset;
 
 				if (be32_to_cpu(pi->ref_tag) == virt)
 					pi->ref_tag = cpu_to_be32(ref_tag);
@@ -183,9 +192,11 @@ static void t10_pi_type1_prepare(struct request *rq)
  */
 static void t10_pi_type1_complete(struct request *rq, unsigned int nr_bytes)
 {
-	unsigned intervals = nr_bytes >> rq->q->integrity.interval_exp;
-	const int tuple_sz = rq->q->integrity.tuple_size;
+	struct blk_integrity *bi = &rq->q->integrity;
+	unsigned intervals = nr_bytes >> bi->interval_exp;
+	const int tuple_sz = bi->tuple_size;
 	u32 ref_tag = t10_pi_ref_tag(rq);
+	u8 offset = bi->pi_offset;
 	struct bio *bio;
 
 	__rq_for_each_bio(bio, rq) {
@@ -200,7 +211,7 @@ static void t10_pi_type1_complete(struct request *rq, unsigned int nr_bytes)
 
 			p = bvec_kmap_local(&iv);
 			for (j = 0; j < iv.bv_len && intervals; j += tuple_sz) {
-				struct t10_pi_tuple *pi = p;
+				struct t10_pi_tuple *pi = p + offset;
 
 				if (be32_to_cpu(pi->ref_tag) == ref_tag)
 					pi->ref_tag = cpu_to_be32(virt);
@@ -280,20 +291,24 @@ const struct blk_integrity_profile t10_pi_type3_ip = {
 };
 EXPORT_SYMBOL(t10_pi_type3_ip);
 
-static __be64 ext_pi_crc64(void *data, unsigned int len)
+static __be64 ext_pi_crc64(u64 crc, void *data, unsigned int len)
 {
-	return cpu_to_be64(crc64_rocksoft(data, len));
+	return cpu_to_be64(crc64_rocksoft_update(crc, data, len));
 }
 
 static blk_status_t ext_pi_crc64_generate(struct blk_integrity_iter *iter,
 					enum t10_dif_type type)
 {
+	u8 offset = iter->pi_offset;
 	unsigned int i;
 
 	for (i = 0 ; i < iter->data_size ; i += iter->interval) {
-		struct crc64_pi_tuple *pi = iter->prot_buf;
+		struct crc64_pi_tuple *pi = iter->prot_buf + offset;
 
-		pi->guard_tag = ext_pi_crc64(iter->data_buf, iter->interval);
+		pi->guard_tag = ext_pi_crc64(0, iter->data_buf, iter->interval);
+		if (offset)
+			pi->guard_tag = ext_pi_crc64(be64_to_cpu(pi->guard_tag),
+					iter->prot_buf, offset);
 		pi->app_tag = 0;
 
 		if (type == T10_PI_TYPE1_PROTECTION)
@@ -319,10 +334,11 @@ static bool ext_pi_ref_escape(u8 *ref_tag)
 static blk_status_t ext_pi_crc64_verify(struct blk_integrity_iter *iter,
 				      enum t10_dif_type type)
 {
+	u8 offset = iter->pi_offset;
 	unsigned int i;
 
 	for (i = 0; i < iter->data_size; i += iter->interval) {
-		struct crc64_pi_tuple *pi = iter->prot_buf;
+		struct crc64_pi_tuple *pi = iter->prot_buf + offset;
 		u64 ref, seed;
 		__be64 csum;
 
@@ -343,7 +359,11 @@ static blk_status_t ext_pi_crc64_verify(struct blk_integrity_iter *iter,
 				goto next;
 		}
 
-		csum = ext_pi_crc64(iter->data_buf, iter->interval);
+		csum = ext_pi_crc64(0, iter->data_buf, iter->interval);
+		if (offset)
+			csum = ext_pi_crc64(be64_to_cpu(csum), iter->prot_buf,
+					    offset);
+
 		if (pi->guard_tag != csum) {
 			pr_err("%s: guard tag error at sector %llu " \
 			       "(rcvd %016llx, want %016llx)\n",
@@ -373,8 +393,10 @@ static blk_status_t ext_pi_type1_generate_crc64(struct blk_integrity_iter *iter)
 
 static void ext_pi_type1_prepare(struct request *rq)
 {
-	const int tuple_sz = rq->q->integrity.tuple_size;
+	struct blk_integrity *bi = &rq->q->integrity;
+	const int tuple_sz = bi->tuple_size;
 	u64 ref_tag = ext_pi_ref_tag(rq);
+	u8 offset = bi->pi_offset;
 	struct bio *bio;
 
 	__rq_for_each_bio(bio, rq) {
@@ -393,7 +415,7 @@ static void ext_pi_type1_prepare(struct request *rq)
 
 			p = bvec_kmap_local(&iv);
 			for (j = 0; j < iv.bv_len; j += tuple_sz) {
-				struct crc64_pi_tuple *pi = p;
+				struct crc64_pi_tuple *pi = p +  offset;
 				u64 ref = get_unaligned_be48(pi->ref_tag);
 
 				if (ref == virt)
@@ -411,9 +433,11 @@ static void ext_pi_type1_prepare(struct request *rq)
 
 static void ext_pi_type1_complete(struct request *rq, unsigned int nr_bytes)
 {
-	unsigned intervals = nr_bytes >> rq->q->integrity.interval_exp;
-	const int tuple_sz = rq->q->integrity.tuple_size;
+	struct blk_integrity *bi = &rq->q->integrity;
+	unsigned intervals = nr_bytes >> bi->interval_exp;
+	const int tuple_sz = bi->tuple_size;
 	u64 ref_tag = ext_pi_ref_tag(rq);
+	u8 offset = bi->pi_offset;
 	struct bio *bio;
 
 	__rq_for_each_bio(bio, rq) {
@@ -428,7 +452,7 @@ static void ext_pi_type1_complete(struct request *rq, unsigned int nr_bytes)
 
 			p = bvec_kmap_local(&iv);
 			for (j = 0; j < iv.bv_len && intervals; j += tuple_sz) {
-				struct crc64_pi_tuple *pi = p;
+				struct crc64_pi_tuple *pi = p + offset;
 				u64 ref = get_unaligned_be48(pi->ref_tag);
 
 				if (ref == ref_tag)