[PATCH] dma-buf: avoid scheduling on fence status query v2

[PATCH] dma-buf: avoid scheduling on fence status query v2

[Andres Rodriguez]

When a timeout of zero is specified, the caller is only interested in
the fence status.

In the current implementation, dma_fence_default_wait will always call
schedule_timeout() at least once for an unsignaled fence. This adds a
significant overhead to a fence status query.

Avoid this overhead by returning early if a zero timeout is specified.

v2: move early return after enable_signaling

Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
---

 If I'm understanding correctly, I don't think we need to register the
 default wait callback. But if that isn't the case please let me know.

 This patch has the same perf improvements as v1.

 drivers/dma-buf/dma-fence.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c
index 0918d3f..57da14c 100644
--- a/drivers/dma-buf/dma-fence.c
+++ b/drivers/dma-buf/dma-fence.c
@@ -402,6 +402,11 @@ dma_fence_default_wait(struct dma_fence *fence, bool intr, signed long timeout)
 		}
 	}
 
+	if (!timeout) {
+		ret = 0;
+		goto out;
+	}
+
 	cb.base.func = dma_fence_default_wait_cb;
 	cb.task = current;
 	list_add(&cb.base.node, &fence->cb_list);
-- 
2.9.3

Re: [PATCH] dma-buf: avoid scheduling on fence status query v2

[Christian König]

Am 26.04.2017 um 16:46 schrieb Andres Rodriguez:
> When a timeout of zero is specified, the caller is only interested in
> the fence status.
>
> In the current implementation, dma_fence_default_wait will always call
> schedule_timeout() at least once for an unsignaled fence. This adds a
> significant overhead to a fence status query.
>
> Avoid this overhead by returning early if a zero timeout is specified.
>
> v2: move early return after enable_signaling
>
> Signed-off-by: Andres Rodriguez <andresx7@gmail.com>

Reviewed-by: Christian König <christian.koenig@amd.com>

> ---
>
>   If I'm understanding correctly, I don't think we need to register the
>   default wait callback. But if that isn't the case please let me know.
>
>   This patch has the same perf improvements as v1.
>
>   drivers/dma-buf/dma-fence.c | 5 +++++
>   1 file changed, 5 insertions(+)
>
> diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c
> index 0918d3f..57da14c 100644
> --- a/drivers/dma-buf/dma-fence.c
> +++ b/drivers/dma-buf/dma-fence.c
> @@ -402,6 +402,11 @@ dma_fence_default_wait(struct dma_fence *fence, bool intr, signed long timeout)
>   		}
>   	}
>   
> +	if (!timeout) {
> +		ret = 0;
> +		goto out;
> +	}
> +
>   	cb.base.func = dma_fence_default_wait_cb;
>   	cb.task = current;
>   	list_add(&cb.base.node, &fence->cb_list);

Re: [PATCH] dma-buf: avoid scheduling on fence status query v2

[Gustavo Padovan]

2017-04-26 Christian König <deathsimple@vodafone.de>:

> Am 26.04.2017 um 16:46 schrieb Andres Rodriguez:
> > When a timeout of zero is specified, the caller is only interested in
> > the fence status.
> > 
> > In the current implementation, dma_fence_default_wait will always call
> > schedule_timeout() at least once for an unsignaled fence. This adds a
> > significant overhead to a fence status query.
> > 
> > Avoid this overhead by returning early if a zero timeout is specified.
> > 
> > v2: move early return after enable_signaling
> > 
> > Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
> 
> Reviewed-by: Christian König <christian.koenig@amd.com>

pushed to drm-misc-next. Thanks all.

Gustavo

Re: [Linaro-mm-sig] [PATCH] dma-buf: avoid scheduling on fence status query v2

[Dave Airlie]

On 28 April 2017 at 07:27, Gustavo Padovan <gustavo@padovan.org> wrote:
> 2017-04-26 Christian König <deathsimple@vodafone.de>:
>
>> Am 26.04.2017 um 16:46 schrieb Andres Rodriguez:
>> > When a timeout of zero is specified, the caller is only interested in
>> > the fence status.
>> >
>> > In the current implementation, dma_fence_default_wait will always call
>> > schedule_timeout() at least once for an unsignaled fence. This adds a
>> > significant overhead to a fence status query.
>> >
>> > Avoid this overhead by returning early if a zero timeout is specified.
>> >
>> > v2: move early return after enable_signaling
>> >
>> > Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
>>
>> Reviewed-by: Christian König <christian.koenig@amd.com>
>
> pushed to drm-misc-next. Thanks all.

I don't see this patch in -rc2, where did it end up going?

Dave.

Re: [Linaro-mm-sig] [PATCH] dma-buf: avoid scheduling on fence status query v2

[Daniel Vetter]

On Wed, May 24, 2017 at 09:47:49AM +1000, Dave Airlie wrote:
> On 28 April 2017 at 07:27, Gustavo Padovan <gustavo@padovan.org> wrote:
> > 2017-04-26 Christian König <deathsimple@vodafone.de>:
> >
> >> Am 26.04.2017 um 16:46 schrieb Andres Rodriguez:
> >> > When a timeout of zero is specified, the caller is only interested in
> >> > the fence status.
> >> >
> >> > In the current implementation, dma_fence_default_wait will always call
> >> > schedule_timeout() at least once for an unsignaled fence. This adds a
> >> > significant overhead to a fence status query.
> >> >
> >> > Avoid this overhead by returning early if a zero timeout is specified.
> >> >
> >> > v2: move early return after enable_signaling
> >> >
> >> > Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
> >>
> >> Reviewed-by: Christian König <christian.koenig@amd.com>
> >
> > pushed to drm-misc-next. Thanks all.
> 
> I don't see this patch in -rc2, where did it end up going?

Queued for 4.13. Makes imo sense since it's just a performance
improvement, not a clear bugfix. But it's in your drm-next, so if you want
to fast-track you can cherry-pick it over:

commit 03c0c5f6641533f5fc14bf4e76d2304197402552
Author: Andres Rodriguez <andresx7@gmail.com>
Date:   Wed Apr 26 10:46:20 2017 -0400

    dma-buf: avoid scheduling on fence status query v2

Cheers, Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch