rbd: rbd workqueues need a resque worker

Need to use WQ_MEM_RECLAIM for our workqueues to prevent I/O lockups under memory pressure - we sit on the memory reclaim path. Cc: [email protected] # 3.17, needs backporting for 3.16 Signed-off-by: Ilya Dryomov <[email protected]> Tested-by: Micha Krause <[email protected]> Reviewed-by: Sage Weil <[email protected]>
sum-it · Oct 14, 2014 · 792c3a9 · 792c3a9
1 parent f9865f0
commit 792c3a9
Showing 1 changed file with 2 additions and 1 deletion.
diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
@@ -5242,7 +5242,8 @@ static int rbd_dev_device_setup(struct rbd_device *rbd_dev)
 	set_capacity(rbd_dev->disk, rbd_dev->mapping.size / SECTOR_SIZE);
 	set_disk_ro(rbd_dev->disk, rbd_dev->mapping.read_only);
 
-	rbd_dev->rq_wq = alloc_workqueue("%s", 0, 0, rbd_dev->disk->disk_name);
+	rbd_dev->rq_wq = alloc_workqueue("%s", WQ_MEM_RECLAIM, 0,
+					 rbd_dev->disk->disk_name);
 	if (!rbd_dev->rq_wq) {
 		ret = -ENOMEM;
 		goto err_out_mapping;