Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1435753 > unrolled thread

[PATCH] spi: spi-bfin-sport: Remove deprecated create_singlethread_workqueue

Back to article view | Back to linux.kernel

Contents

  [PATCH] spi: spi-bfin-sport: Remove deprecated  create_singlethread_workqueue Bhaktipriya Shridhar <bhaktipriya96@gmail.com> - 2016-07-02 11:00 +0200
    Re: [PATCH] spi: spi-bfin-sport: Remove deprecated  create_singlethread_workqueue Tejun Heo <tj@kernel.org> - 2016-07-02 13:50 +0200

#1435753 — [PATCH] spi: spi-bfin-sport: Remove deprecated create_singlethread_workqueue

The workqueue "workqueue" serves as a driver message queue.
It has a single work item(&drv_data->pump_messages) 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.

Work item has been flushed in bfin_sport_spi_destroy_queue() to ensure
that there are no pending tasks while disconnecting the driver.

Signed-off-by: Bhaktipriya Shridhar <bhaktipriya96@gmail.com>
---
 drivers/spi/spi-bfin-sport.c | 15 ++++-----------
 1 file changed, 4 insertions(+), 11 deletions(-)

diff --git a/drivers/spi/spi-bfin-sport.c b/drivers/spi/spi-bfin-sport.c
index 6c96755..01d0ba9 100644
--- a/drivers/spi/spi-bfin-sport.c
+++ b/drivers/spi/spi-bfin-sport.c
@@ -64,8 +64,6 @@ struct bfin_sport_spi_master_data {
 	/* Pin request list */
 	u16 *pin_req;

-	/* Driver message queue */
-	struct workqueue_struct *workqueue;
 	struct work_struct pump_messages;
 	spinlock_t lock;
 	struct list_head queue;
@@ -300,7 +298,7 @@ bfin_sport_spi_giveback(struct bfin_sport_spi_master_data *drv_data)
 	drv_data->cur_msg = NULL;
 	drv_data->cur_transfer = NULL;
 	drv_data->cur_chip = NULL;
-	queue_work(drv_data->workqueue, &drv_data->pump_messages);
+	schedule_work(&drv_data->pump_messages);
 	spin_unlock_irqrestore(&drv_data->lock, flags);

 	if (!drv_data->cs_change)
@@ -556,7 +554,7 @@ bfin_sport_spi_transfer(struct spi_device *spi, struct spi_message *msg)
 	list_add_tail(&msg->queue, &drv_data->queue);

 	if (drv_data->run && !drv_data->busy)
-		queue_work(drv_data->workqueue, &drv_data->pump_messages);
+		schedule_work(&drv_data->pump_messages);

 	spin_unlock_irqrestore(&drv_data->lock, flags);

@@ -666,12 +664,7 @@ bfin_sport_spi_init_queue(struct bfin_sport_spi_master_data *drv_data)
 	tasklet_init(&drv_data->pump_transfers,
 		     bfin_sport_spi_pump_transfers, (unsigned long)drv_data);

-	/* init messages workqueue */
 	INIT_WORK(&drv_data->pump_messages, bfin_sport_spi_pump_messages);
-	drv_data->workqueue =
-	    create_singlethread_workqueue(dev_name(drv_data->master->dev.parent));
-	if (drv_data->workqueue == NULL)
-		return -EBUSY;

 	return 0;
 }
@@ -694,7 +687,7 @@ bfin_sport_spi_start_queue(struct bfin_sport_spi_master_data *drv_data)
 	drv_data->cur_chip = NULL;
 	spin_unlock_irqrestore(&drv_data->lock, flags);

-	queue_work(drv_data->workqueue, &drv_data->pump_messages);
+	schedule_work(&drv_data->pump_messages);

 	return 0;
 }
@@ -738,7 +731,7 @@ bfin_sport_spi_destroy_queue(struct bfin_sport_spi_master_data *drv_data)
 	if (status)
 		return status;

-	destroy_workqueue(drv_data->workqueue);
+	flush_work(&drv_data->pump_messages);

 	return 0;
 }
--
2.1.4

[toc] | [next] | [standalone]

#1435807

On Sat, Jul 02, 2016 at 02:29:41PM +0530, Bhaktipriya Shridhar wrote:
> The workqueue "workqueue" serves as a driver message queue.
> It has a single work item(&drv_data->pump_messages) 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.
> 
> Work item has been flushed in bfin_sport_spi_destroy_queue() to ensure
> that there are no pending tasks while disconnecting the driver.
> 
> Signed-off-by: Bhaktipriya Shridhar <bhaktipriya96@gmail.com>

Acked-by: Tejun Heo <tj@kernel.org>

Thanks.

-- 
tejun

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web