md/raid5: use wait_on_bit() for R5_Overlap

Convert uses of wait_for_overlap wait queue with R5_Overlap bit to
wait_on_bit() / wake_up_bit().

Signed-off-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Link: https://lore.kernel.org/r/20240827153536.6743-2-artur.paszkiewicz@intel.com
Signed-off-by: Song Liu <song@kernel.org>

2 files changed, 30 insertions, 36 deletions

diff --git a/drivers/md/raid5-cache.c b/drivers/md/raid5-cache.c
index 23f2cbcf1a6c..b4f7b79fd187 100644
--- a/drivers/md/raid5-cache.c
+++ b/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);
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index c84a7e0263cd..de7d9959b3dc 100644
--- a/drivers/md/raid5.c
+++ b/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
@@ -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);
 
@@ -5870,14 +5864,19 @@ 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)