Merge branch 'md-6.12-raid5-opt' into md-6.12

From Artur:

The wait_for_overlap wait queue is currently used in two cases, which
are not really related:
 - waiting for actual overlapping bios, which uses R5_Overlap bit,
 - waiting for events related to reshape.

Handling every write request in raid5_make_request() involves adding to
and removing from this wait queue, which uses a spinlock. With fast
storage and multiple submitting threads the contention on this lock is
noticeable.

This patch series aims to resolve this by separating the two cases
mentioned above and using this wait queue only when reshape is in
progress.

The results when testing 4k random writes on raid5 with null_blk
(8 jobs, qd=64, group_thread_cnt=8):
before: 463k IOPS
after:  523k IOPS

The improvement is not huge with this series alone but it is just one of
the bottlenecks. When applied onto some other changes I'm working on, it
allowed to go from 845k IOPS to 975k IOPS on the same test.

* md-6.12-raid5-opt:
  md/raid5: rename wait_for_overlap to wait_for_reshape
  md/raid5: only add to wq if reshape is in progress
  md/raid5: use wait_on_bit() for R5_Overlap
Signed-off-by: Song Liu <song@kernel.org>

drivers/md/raid5-cache.c

@@ -2798,7 +2798,6 @@ void r5c_finish_stripe_write_out(struct r5conf *conf,
 {
 	struct r5l_log *log = READ_ONCE(conf->log);
 	int i;
-	int do_wakeup = 0;
 	sector_t tree_index;
 	void __rcu **pslot;
 	uintptr_t refcount;
@@ -2815,7 +2814,7 @@ void r5c_finish_stripe_write_out(struct r5conf *conf,
 	for (i = sh->disks; i--; ) {
 		clear_bit(R5_InJournal, &sh->dev[i].flags);
 		if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
-			do_wakeup = 1;
+			wake_up_bit(&sh->dev[i].flags, R5_Overlap);
 	}
 
 	/*
@@ -2828,9 +2827,6 @@ void r5c_finish_stripe_write_out(struct r5conf *conf,
 		if (atomic_dec_and_test(&conf->pending_full_writes))
 			md_wakeup_thread(conf->mddev->thread);
 
-	if (do_wakeup)
-		wake_up(&conf->wait_for_overlap);
-
 	spin_lock_irq(&log->stripe_in_journal_lock);
 	list_del_init(&sh->r5c);
 	spin_unlock_irq(&log->stripe_in_journal_lock);

drivers/md/raid5.c

@@ -2337,7 +2337,7 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
 		for (i = disks; i--; ) {
 			struct r5dev *dev = &sh->dev[i];
 			if (test_and_clear_bit(R5_Overlap, &dev->flags))
-				wake_up(&sh->raid_conf->wait_for_overlap);
+				wake_up_bit(&dev->flags, R5_Overlap);
 		}
 	}
 	local_unlock(&conf->percpu->lock);
@@ -3473,7 +3473,7 @@ static bool stripe_bio_overlaps(struct stripe_head *sh, struct bio *bi,
 		 * With PPL only writes to consecutive data chunks within a
 		 * stripe are allowed because for a single stripe_head we can
 		 * only have one PPL entry at a time, which describes one data
-		 * range. Not really an overlap, but wait_for_overlap can be
+		 * range. Not really an overlap, but R5_Overlap can be
 		 * used to handle this.
 		 */
 		sector_t sector;
@@ -3652,7 +3652,7 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
 		log_stripe_write_finished(sh);
 
 		if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
-			wake_up(&conf->wait_for_overlap);
+			wake_up_bit(&sh->dev[i].flags, R5_Overlap);
 
 		while (bi && bi->bi_iter.bi_sector <
 			sh->dev[i].sector + RAID5_STRIPE_SECTORS(conf)) {
@@ -3697,7 +3697,7 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
 			sh->dev[i].toread = NULL;
 			spin_unlock_irq(&sh->stripe_lock);
 			if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
-				wake_up(&conf->wait_for_overlap);
+				wake_up_bit(&sh->dev[i].flags, R5_Overlap);
 			if (bi)
 				s->to_read--;
 			while (bi && bi->bi_iter.bi_sector <
@@ -3736,7 +3736,7 @@ handle_failed_sync(struct r5conf *conf, struct stripe_head *sh,
 	BUG_ON(sh->batch_head);
 	clear_bit(STRIPE_SYNCING, &sh->state);
 	if (test_and_clear_bit(R5_Overlap, &sh->dev[sh->pd_idx].flags))
-		wake_up(&conf->wait_for_overlap);
+		wake_up_bit(&sh->dev[sh->pd_idx].flags, R5_Overlap);
 	s->syncing = 0;
 	s->replacing = 0;
 	/* There is nothing more to do for sync/check/repair.
@@ -4877,7 +4877,6 @@ static void break_stripe_batch_list(struct stripe_head *head_sh,
 {
 	struct stripe_head *sh, *next;
 	int i;
-	int do_wakeup = 0;
 
 	list_for_each_entry_safe(sh, next, &head_sh->batch_list, batch_list) {
 
@@ -4913,7 +4912,7 @@ static void break_stripe_batch_list(struct stripe_head *head_sh,
 		spin_unlock_irq(&sh->stripe_lock);
 		for (i = 0; i < sh->disks; i++) {
 			if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
-				do_wakeup = 1;
+				wake_up_bit(&sh->dev[i].flags, R5_Overlap);
 			sh->dev[i].flags = head_sh->dev[i].flags &
 				(~((1 << R5_WriteError) | (1 << R5_Overlap)));
 		}
@@ -4927,12 +4926,9 @@ static void break_stripe_batch_list(struct stripe_head *head_sh,
 	spin_unlock_irq(&head_sh->stripe_lock);
 	for (i = 0; i < head_sh->disks; i++)
 		if (test_and_clear_bit(R5_Overlap, &head_sh->dev[i].flags))
-			do_wakeup = 1;
+			wake_up_bit(&head_sh->dev[i].flags, R5_Overlap);
 	if (head_sh->state & handle_flags)
 		set_bit(STRIPE_HANDLE, &head_sh->state);
-
-	if (do_wakeup)
-		wake_up(&head_sh->raid_conf->wait_for_overlap);
 }
 
 static void handle_stripe(struct stripe_head *sh)
@@ -5198,7 +5194,7 @@ static void handle_stripe(struct stripe_head *sh)
 		md_done_sync(conf->mddev, RAID5_STRIPE_SECTORS(conf), 1);
 		clear_bit(STRIPE_SYNCING, &sh->state);
 		if (test_and_clear_bit(R5_Overlap, &sh->dev[sh->pd_idx].flags))
-			wake_up(&conf->wait_for_overlap);
+			wake_up_bit(&sh->dev[sh->pd_idx].flags, R5_Overlap);
 	}
 
 	/* If the failed drives are just a ReadError, then we might need
@@ -5261,7 +5257,7 @@ static void handle_stripe(struct stripe_head *sh)
 	} else if (s.expanded && !sh->reconstruct_state && s.locked == 0) {
 		clear_bit(STRIPE_EXPAND_READY, &sh->state);
 		atomic_dec(&conf->reshape_stripes);
-		wake_up(&conf->wait_for_overlap);
+		wake_up(&conf->wait_for_reshape);
 		md_done_sync(conf->mddev, RAID5_STRIPE_SECTORS(conf), 1);
 	}
 
@@ -5755,12 +5751,11 @@ static void make_discard_request(struct mddev *mddev, struct bio *bi)
 		int d;
 	again:
 		sh = raid5_get_active_stripe(conf, NULL, logical_sector, 0);
-		prepare_to_wait(&conf->wait_for_overlap, &w,
-				TASK_UNINTERRUPTIBLE);
 		set_bit(R5_Overlap, &sh->dev[sh->pd_idx].flags);
 		if (test_bit(STRIPE_SYNCING, &sh->state)) {
 			raid5_release_stripe(sh);
-			schedule();
+			wait_on_bit(&sh->dev[sh->pd_idx].flags, R5_Overlap,
+				    TASK_UNINTERRUPTIBLE);
 			goto again;
 		}
 		clear_bit(R5_Overlap, &sh->dev[sh->pd_idx].flags);
@@ -5772,12 +5767,12 @@ static void make_discard_request(struct mddev *mddev, struct bio *bi)
 				set_bit(R5_Overlap, &sh->dev[d].flags);
 				spin_unlock_irq(&sh->stripe_lock);
 				raid5_release_stripe(sh);
-				schedule();
+				wait_on_bit(&sh->dev[d].flags, R5_Overlap,
+					    TASK_UNINTERRUPTIBLE);
 				goto again;
 			}
 		}
 		set_bit(STRIPE_DISCARD, &sh->state);
-		finish_wait(&conf->wait_for_overlap, &w);
 		sh->overwrite_disks = 0;
 		for (d = 0; d < conf->raid_disks; d++) {
 			if (d == sh->pd_idx || d == sh->qd_idx)
@@ -5854,7 +5849,6 @@ static int add_all_stripe_bios(struct r5conf *conf,
 		struct bio *bi, int forwrite, int previous)
 {
 	int dd_idx;
-	int ret = 1;
 
 	spin_lock_irq(&sh->stripe_lock);
 
@@ -5870,13 +5864,18 @@ static int add_all_stripe_bios(struct r5conf *conf,
 
 		if (stripe_bio_overlaps(sh, bi, dd_idx, forwrite)) {
 			set_bit(R5_Overlap, &dev->flags);
-			ret = 0;
-			continue;
+			spin_unlock_irq(&sh->stripe_lock);
+			raid5_release_stripe(sh);
+			/* release batch_last before wait to avoid risk of deadlock */
+			if (ctx->batch_last) {
+				raid5_release_stripe(ctx->batch_last);
+				ctx->batch_last = NULL;
+			}
+			md_wakeup_thread(conf->mddev->thread);
+			wait_on_bit(&dev->flags, R5_Overlap, TASK_UNINTERRUPTIBLE);
+			return 0;
 		}
 	}
-
-	if (!ret)
-		goto out;
 
 	for (dd_idx = 0; dd_idx < sh->disks; dd_idx++) {
 		struct r5dev *dev = &sh->dev[dd_idx];
@@ -5893,9 +5892,8 @@ static int add_all_stripe_bios(struct r5conf *conf,
 			  RAID5_STRIPE_SHIFT(conf), ctx->sectors_to_do);
 	}
 
-out:
 	spin_unlock_irq(&sh->stripe_lock);
-	return ret;
+	return 1;
 }
 
 enum reshape_loc {
@@ -5991,17 +5989,17 @@ static enum stripe_result make_stripe_request(struct mddev *mddev,
 		goto out_release;
 	}
 
-	if (test_bit(STRIPE_EXPANDING, &sh->state) ||
-	    !add_all_stripe_bios(conf, ctx, sh, bi, rw, previous)) {
-		/*
-		 * Stripe is busy expanding or add failed due to
-		 * overlap. Flush everything and wait a while.
-		 */
+	if (test_bit(STRIPE_EXPANDING, &sh->state)) {
 		md_wakeup_thread(mddev->thread);
 		ret = STRIPE_SCHEDULE_AND_RETRY;
 		goto out_release;
 	}
 
+	if (!add_all_stripe_bios(conf, ctx, sh, bi, rw, previous)) {
+		ret = STRIPE_RETRY;
+		goto out;
+	}
+
 	if (stripe_can_batch(sh)) {
 		stripe_add_to_batch_list(conf, sh, ctx->batch_last);
 		if (ctx->batch_last)
@@ -6072,6 +6070,7 @@ static sector_t raid5_bio_lowest_chunk_sector(struct r5conf *conf,
 static bool raid5_make_request(struct mddev *mddev, struct bio * bi)
 {
 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
+	bool on_wq;
 	struct r5conf *conf = mddev->private;
 	sector_t logical_sector;
 	struct stripe_request_ctx ctx = {};
@@ -6145,11 +6144,15 @@ static bool raid5_make_request(struct mddev *mddev, struct bio * bi)
 	 * sequential IO pattern. We don't bother with the optimization when
 	 * reshaping as the performance benefit is not worth the complexity.
 	 */
-	if (likely(conf->reshape_progress == MaxSector))
+	if (likely(conf->reshape_progress == MaxSector)) {
 		logical_sector = raid5_bio_lowest_chunk_sector(conf, bi);
+		on_wq = false;
+	} else {
+		add_wait_queue(&conf->wait_for_reshape, &wait);
+		on_wq = true;
+	}
 	s = (logical_sector - ctx.first_sector) >> RAID5_STRIPE_SHIFT(conf);
 
-	add_wait_queue(&conf->wait_for_overlap, &wait);
 	while (1) {
 		res = make_stripe_request(mddev, conf, &ctx, logical_sector,
 					  bi);
@@ -6160,6 +6163,7 @@ static bool raid5_make_request(struct mddev *mddev, struct bio * bi)
 			continue;
 
 		if (res == STRIPE_SCHEDULE_AND_RETRY) {
+			WARN_ON_ONCE(!on_wq);
 			/*
 			 * Must release the reference to batch_last before
 			 * scheduling and waiting for work to be done,
@@ -6184,7 +6188,8 @@ static bool raid5_make_request(struct mddev *mddev, struct bio * bi)
 		logical_sector = ctx.first_sector +
 			(s << RAID5_STRIPE_SHIFT(conf));
 	}
-	remove_wait_queue(&conf->wait_for_overlap, &wait);
+	if (unlikely(on_wq))
+		remove_wait_queue(&conf->wait_for_reshape, &wait);
 
 	if (ctx.batch_last)
 		raid5_release_stripe(ctx.batch_last);
@@ -6337,7 +6342,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 	     : (safepos < writepos && readpos > writepos)) ||
 	    time_after(jiffies, conf->reshape_checkpoint + 10*HZ)) {
 		/* Cannot proceed until we've updated the superblock... */
-		wait_event(conf->wait_for_overlap,
+		wait_event(conf->wait_for_reshape,
 			   atomic_read(&conf->reshape_stripes)==0
 			   || test_bit(MD_RECOVERY_INTR, &mddev->recovery));
 		if (atomic_read(&conf->reshape_stripes) != 0)
@@ -6363,7 +6368,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 		spin_lock_irq(&conf->device_lock);
 		conf->reshape_safe = mddev->reshape_position;
 		spin_unlock_irq(&conf->device_lock);
-		wake_up(&conf->wait_for_overlap);
+		wake_up(&conf->wait_for_reshape);
 		sysfs_notify_dirent_safe(mddev->sysfs_completed);
 	}
 
@@ -6446,7 +6451,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 	    (sector_nr - mddev->curr_resync_completed) * 2
 	    >= mddev->resync_max - mddev->curr_resync_completed) {
 		/* Cannot proceed until we've updated the superblock... */
-		wait_event(conf->wait_for_overlap,
+		wait_event(conf->wait_for_reshape,
 			   atomic_read(&conf->reshape_stripes) == 0
 			   || test_bit(MD_RECOVERY_INTR, &mddev->recovery));
 		if (atomic_read(&conf->reshape_stripes) != 0)
@@ -6472,7 +6477,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 		spin_lock_irq(&conf->device_lock);
 		conf->reshape_safe = mddev->reshape_position;
 		spin_unlock_irq(&conf->device_lock);
-		wake_up(&conf->wait_for_overlap);
+		wake_up(&conf->wait_for_reshape);
 		sysfs_notify_dirent_safe(mddev->sysfs_completed);
 	}
 ret:
@@ -6507,7 +6512,7 @@ static inline sector_t raid5_sync_request(struct mddev *mddev, sector_t sector_n
 	}
 
 	/* Allow raid5_quiesce to complete */
-	wait_event(conf->wait_for_overlap, conf->quiesce != 2);
+	wait_event(conf->wait_for_reshape, conf->quiesce != 2);
 
 	if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
 		return reshape_request(mddev, sector_nr, skipped);
@@ -7493,7 +7498,7 @@ static struct r5conf *setup_conf(struct mddev *mddev)
 
 	init_waitqueue_head(&conf->wait_for_quiescent);
 	init_waitqueue_head(&conf->wait_for_stripe);
-	init_waitqueue_head(&conf->wait_for_overlap);
+	init_waitqueue_head(&conf->wait_for_reshape);
 	INIT_LIST_HEAD(&conf->handle_list);
 	INIT_LIST_HEAD(&conf->loprio_list);
 	INIT_LIST_HEAD(&conf->hold_list);
@@ -8552,7 +8557,7 @@ static void end_reshape(struct r5conf *conf)
 			    !test_bit(In_sync, &rdev->flags))
 				rdev->recovery_offset = MaxSector;
 		spin_unlock_irq(&conf->device_lock);
-		wake_up(&conf->wait_for_overlap);
+		wake_up(&conf->wait_for_reshape);
 
 		mddev_update_io_opt(conf->mddev,
 			conf->raid_disks - conf->max_degraded);
@@ -8616,13 +8621,13 @@ static void raid5_quiesce(struct mddev *mddev, int quiesce)
 		conf->quiesce = 1;
 		unlock_all_device_hash_locks_irq(conf);
 		/* allow reshape to continue */
-		wake_up(&conf->wait_for_overlap);
+		wake_up(&conf->wait_for_reshape);
 	} else {
 		/* re-enable writes */
 		lock_all_device_hash_locks_irq(conf);
 		conf->quiesce = 0;
 		wake_up(&conf->wait_for_quiescent);
-		wake_up(&conf->wait_for_overlap);
+		wake_up(&conf->wait_for_reshape);
 		unlock_all_device_hash_locks_irq(conf);
 	}
 	log_quiesce(conf, quiesce);
@@ -8941,7 +8946,7 @@ static void raid5_prepare_suspend(struct mddev *mddev)
 {
 	struct r5conf *conf = mddev->private;
 
-	wake_up(&conf->wait_for_overlap);
+	wake_up(&conf->wait_for_reshape);
 }
 
 static struct md_personality raid6_personality =

drivers/md/raid5.h

@@ -668,7 +668,7 @@ struct r5conf {
 	struct llist_head	released_stripes;
 	wait_queue_head_t	wait_for_quiescent;
 	wait_queue_head_t	wait_for_stripe;
-	wait_queue_head_t	wait_for_overlap;
+	wait_queue_head_t	wait_for_reshape;
 	unsigned long		cache_state;
 	struct shrinker		*shrinker;
 	int			pool_size; /* number of disks in stripeheads in pool */