* [PATCH] dma-buf: avoid scheduling on fence status query v2
@ 2017-04-26 14:46 ` Andres Rodriguez
0 siblings, 0 replies; 7+ messages in thread
From: Andres Rodriguez @ 2017-04-26 14:46 UTC (permalink / raw)
To: dri-devel; +Cc: andresx7, deathsimple, sumit.semwal, linux-media, linaro-mm-sig
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
^ permalink raw reply related [flat|nested] 7+ messages in thread
* [PATCH] dma-buf: avoid scheduling on fence status query v2
@ 2017-04-26 14:46 ` Andres Rodriguez
0 siblings, 0 replies; 7+ messages in thread
From: Andres Rodriguez @ 2017-04-26 14:46 UTC (permalink / raw)
To: dri-devel; +Cc: linaro-mm-sig, linux-media, andresx7
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
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] dma-buf: avoid scheduling on fence status query v2
2017-04-26 14:46 ` Andres Rodriguez
(?)
@ 2017-04-26 14:49 ` Christian König
2017-04-27 21:27 ` Gustavo Padovan
-1 siblings, 1 reply; 7+ messages in thread
From: Christian König @ 2017-04-26 14:49 UTC (permalink / raw)
To: Andres Rodriguez, dri-devel; +Cc: sumit.semwal, linux-media, linaro-mm-sig
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);
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] dma-buf: avoid scheduling on fence status query v2
2017-04-26 14:49 ` Christian König
@ 2017-04-27 21:27 ` Gustavo Padovan
2017-05-23 23:47 ` [Linaro-mm-sig] " Dave Airlie
0 siblings, 1 reply; 7+ messages in thread
From: Gustavo Padovan @ 2017-04-27 21:27 UTC (permalink / raw)
To: Christian König
Cc: Andres Rodriguez, dri-devel, linaro-mm-sig, linux-media
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
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Linaro-mm-sig] [PATCH] dma-buf: avoid scheduling on fence status query v2
2017-04-27 21:27 ` Gustavo Padovan
@ 2017-05-23 23:47 ` Dave Airlie
2017-05-24 12:21 ` Daniel Vetter
0 siblings, 1 reply; 7+ messages in thread
From: Dave Airlie @ 2017-05-23 23:47 UTC (permalink / raw)
To: Gustavo Padovan, Christian König, Andres Rodriguez,
dri-devel, linaro-mm-sig, Linux Media Mailing List
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.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Linaro-mm-sig] [PATCH] dma-buf: avoid scheduling on fence status query v2
2017-05-23 23:47 ` [Linaro-mm-sig] " Dave Airlie
@ 2017-05-24 12:21 ` Daniel Vetter
0 siblings, 0 replies; 7+ messages in thread
From: Daniel Vetter @ 2017-05-24 12:21 UTC (permalink / raw)
To: Dave Airlie
Cc: Gustavo Padovan, Christian König, Andres Rodriguez,
dri-devel, linaro-mm-sig, Linux Media Mailing List
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
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Linaro-mm-sig] [PATCH] dma-buf: avoid scheduling on fence status query v2
@ 2017-05-24 12:21 ` Daniel Vetter
0 siblings, 0 replies; 7+ messages in thread
From: Daniel Vetter @ 2017-05-24 12:21 UTC (permalink / raw)
To: Dave Airlie
Cc: dri-devel, linaro-mm-sig, Andres Rodriguez, Linux Media Mailing List
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
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2017-05-24 12:21 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-04-26 14:46 [PATCH] dma-buf: avoid scheduling on fence status query v2 Andres Rodriguez
2017-04-26 14:46 ` Andres Rodriguez
2017-04-26 14:49 ` Christian König
2017-04-27 21:27 ` Gustavo Padovan
2017-05-23 23:47 ` [Linaro-mm-sig] " Dave Airlie
2017-05-24 12:21 ` Daniel Vetter
2017-05-24 12:21 ` Daniel Vetter
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.