Skip to contentSkip to conversation

s390/cio/chp : Remove deprecated create_singlethread_workqueue

PATCH Applied Bhaktipriya Shridhar 2 participants 2 messages Jul 16, 2016

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 s390/cio/chp : Remove deprecated create_singlethread_workqueue

The workqueue "chp_wq" is involved in performing pending configure tasks for channel paths.

It has a single work item(&cfg_work) and hence doesn't require ordering. Also, it is not being used on a memory reclaim path. Hence, the singlethreaded workqueue has been replaced with the use of system_wq.

System workqueues have been able to handle high level of concurrency for a long time now and hence it's not required to have a singlethreaded workqueue just to gain concurrency. Unlike a dedicated per-cpu workqueue created with create_singlethread_workqueue(), system_wq allows multiple work items to overlap executions even on the same CPU; however, a per-cpu workqueue doesn't have any CPU locality or global ordering guarantee unless the target CPU is explicitly specified and thus the increase of local concurrency shouldn't make any difference.

---
 drivers/s390/cio/chp.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)
drivers/s390/cio/chp.c +2 -9
index 50597f9..92413c4 100644
@@ -47,8 +47,6 @@ static DEFINE_MUTEX(info_lock);
 /* Time after which channel-path status may be outdated. */
 static unsigned long chp_info_expires;
 
-/* Workqueue to perform pending configure tasks. */
-static struct workqueue_struct *chp_wq;
 static struct work_struct cfg_work;
 
 /* Wait queue for configure completion events. */
@@ -714,7 +712,7 @@ static void cfg_func(struct work_struct *work)
 		wake_up_interruptible(&cfg_wait_queue);
 		return;
 	}
-	queue_work(chp_wq, &cfg_work);
+	schedule_work(&cfg_work);
 }
 
 /**
@@ -732,7 +730,7 @@ void chp_cfg_schedule(struct chp_id chpid, int configure)
 	cfg_set_task(chpid, configure ? cfg_configure : cfg_deconfigure);
 	cfg_busy = 1;
 	mutex_unlock(&cfg_lock);
-	queue_work(chp_wq, &cfg_work);
+	schedule_work(&cfg_work);
 }
 
 /**
@@ -766,11 +764,6 @@ static int __init chp_init(void)
 	ret = crw_register_handler(CRW_RSC_CPATH, chp_process_crw);
 	if (ret)
 		return ret;
-	chp_wq = create_singlethread_workqueue("cio_chp");
-	if (!chp_wq) {
-		crw_unregister_handler(CRW_RSC_CPATH);
-		return -ENOMEM;
-	}
 	INIT_WORK(&cfg_work, cfg_func);
 	init_waitqueue_head(&cfg_wait_queue);
 	if (info_update())

--
2.1.4

#1 lore raw
Quoted 17 lines · Bhaktipriya Shridhar

The workqueue "chp_wq" is involved in performing pending configure tasks for channel paths.

It has a single work item(&cfg_work) and hence doesn't require ordering. Also, it is not being used on a memory reclaim path. Hence, the singlethreaded workqueue has been replaced with the use of system_wq.

System workqueues have been able to handle high level of concurrency for a long time now and hence it's not required to have a singlethreaded workqueue just to gain concurrency. Unlike a dedicated per-cpu workqueue created with create_singlethread_workqueue(), system_wq allows multiple work items to overlap executions even on the same CPU; however, a per-cpu workqueue doesn't have any CPU locality or global ordering guarantee unless the target CPU is explicitly specified and thus the increase of local concurrency shouldn't make any difference.

Quoted 1 line

Applied, thanks!

Sebastian