block: use a per disk workqueue for zone write plugging

A zone write plug BIO work function blk_zone_wplug_bio_work() calls
submit_bio_noacct_nocheck() to execute the next unplugged BIO. This
function may block. So executing zone plugs BIO works using the block
layer global kblockd workqueue can potentially lead to preformance or
latency issues as the number of concurrent work for a workqueue is
limited to WQ_DFL_ACTIVE (256).
1) For a system with a large number of zoned disks, issuing write
   requests to otherwise unused zones may be delayed wiating for a work
   thread to become available.
2) Requeue operations which use kblockd but are independent of zone
   write plugging may alsoi end up being delayed.

To avoid these potential performance issues, create a workqueue per
zoned device to execute zone plugs BIO work. The workqueue max active
parameter is set to the maximum number of zone write plugs allocated
with the zone write plug mempool. This limit is equal to the maximum
number of open zones of the disk and defaults to 128 for disks that do
not have a limit on the number of open zones.

Fixes: dd291d77cc90 ("block: Introduce zone write plugging")
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20240420075811.1276893-3-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

1 files changed, 24 insertions, 8 deletions

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 3befebe6b319..3a796420f240 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -1129,7 +1129,7 @@ static void disk_zone_wplug_unplug_bio(struct gendisk *disk,
 	/* Schedule submission of the next plugged BIO if we have one. */
 	if (!bio_list_empty(&zwplug->bio_list)) {
 		spin_unlock_irqrestore(&zwplug->lock, flags);
-		kblockd_schedule_work(&zwplug->bio_work);
+		queue_work(disk->zone_wplugs_wq, &zwplug->bio_work);
 		return;
 	}
 
@@ -1332,7 +1332,7 @@ static void disk_zone_wplug_handle_error(struct gendisk *disk,
 	/* Restart BIO submission if we still have any BIO left. */
 	if (!bio_list_empty(&zwplug->bio_list)) {
 		WARN_ON_ONCE(!(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED));
-		kblockd_schedule_work(&zwplug->bio_work);
+		queue_work(disk->zone_wplugs_wq, &zwplug->bio_work);
 		goto unlock;
 	}
 
@@ -1409,14 +1409,25 @@ static int disk_alloc_zone_resources(struct gendisk *disk,
 
 	disk->zone_wplugs_pool = mempool_create_kmalloc_pool(pool_size,
 						sizeof(struct blk_zone_wplug));
-	if (!disk->zone_wplugs_pool) {
-		kfree(disk->zone_wplugs_hash);
-		disk->zone_wplugs_hash = NULL;
-		disk->zone_wplugs_hash_bits = 0;
-		return -ENOMEM;
-	}
+	if (!disk->zone_wplugs_pool)
+		goto free_hash;
+
+	disk->zone_wplugs_wq =
+		alloc_workqueue("%s_zwplugs", WQ_MEM_RECLAIM | WQ_HIGHPRI,
+				pool_size, disk->disk_name);
+	if (!disk->zone_wplugs_wq)
+		goto destroy_pool;
 
 	return 0;
+
+destroy_pool:
+	mempool_destroy(disk->zone_wplugs_pool);
+	disk->zone_wplugs_pool = NULL;
+free_hash:
+	kfree(disk->zone_wplugs_hash);
+	disk->zone_wplugs_hash = NULL;
+	disk->zone_wplugs_hash_bits = 0;
+	return -ENOMEM;
 }
 
 static void disk_destroy_zone_wplugs_hash_table(struct gendisk *disk)
@@ -1447,6 +1458,11 @@ void disk_free_zone_resources(struct gendisk *disk)
 {
 	cancel_work_sync(&disk->zone_wplugs_work);
 
+	if (disk->zone_wplugs_wq) {
+		destroy_workqueue(disk->zone_wplugs_wq);
+		disk->zone_wplugs_wq = NULL;
+	}
+
 	disk_destroy_zone_wplugs_hash_table(disk);
 
 	/*