2021-03-29 09:42:56

by Liu Shixin

Subject: [PATCH -next 2/2] s390/cio: use DECLARE_WAIT_QUEUE_HEAD() for wait_queue

wait_queue_head_t can be initialized automatically with DECLARE_WAIT_QUEUE_HEAD()
rather than explicitly calling init_waitqueue_head().

Signed-off-by: Shixin Liu <[email protected]>
---
drivers/s390/cio/css.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c
index f01ef6325039..a974943c27da 100644
--- a/drivers/s390/cio/css.c
+++ b/drivers/s390/cio/css.c
@@ -652,13 +652,12 @@ static void css_sch_todo(struct work_struct *work)

static struct idset *slow_subchannel_set;
static DEFINE_SPINLOCK(slow_subchannel_lock);
-static wait_queue_head_t css_eval_wq;
+static DECLARE_WAIT_QUEUE_HEAD(css_eval_wq);
static atomic_t css_eval_scheduled;

static int __init slow_subchannel_init(void)
{
atomic_set(&css_eval_scheduled, 0);
- init_waitqueue_head(&css_eval_wq);
slow_subchannel_set = idset_sch_new();
if (!slow_subchannel_set) {
CIO_MSG_EVENT(0, "could not allocate slow subchannel set\n");
--
2.25.1

2021-03-29 12:36:32

by Vineeth Vijayan

Subject: Re: [PATCH -next 2/2] s390/cio: use DECLARE_WAIT_QUEUE_HEAD() for wait_queue

For both the patches,

Acked-by: Vineeth Vijayan <[email protected]>

I will correct the description, and Heiko/Vasily will pick this up and
will be part of the s390-tree patchset.

On 3/29/21 11:40 AM, Shixin Liu wrote:
> wait_queue_head_t can be initialized automatically with DECLARE_WAIT_QUEUE_HEAD()
> rather than explicitly calling init_waitqueue_head().

A minor checkpatch WARNING. I will modify this.

>
> Signed-off-by: Shixin Liu <[email protected]>
> ---
> drivers/s390/cio/css.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c
> index f01ef6325039..a974943c27da 100644
> --- a/drivers/s390/cio/css.c
> +++ b/drivers/s390/cio/css.c
> @@ -652,13 +652,12 @@ static void css_sch_todo(struct work_struct *work)
>
> static struct idset *slow_subchannel_set;
> static DEFINE_SPINLOCK(slow_subchannel_lock);
> -static wait_queue_head_t css_eval_wq;
> +static DECLARE_WAIT_QUEUE_HEAD(css_eval_wq);
> static atomic_t css_eval_scheduled;
>
> static int __init slow_subchannel_init(void)
> {
> atomic_set(&css_eval_scheduled, 0);
> - init_waitqueue_head(&css_eval_wq);
> slow_subchannel_set = idset_sch_new();
> if (!slow_subchannel_set) {
> CIO_MSG_EVENT(0, "could not allocate slow subchannel set\n");