* [PATCH] drm/sched: Declare entity idle only after HW submission
@ 2021-06-24 14:08 Boris Brezillon
2021-06-24 15:24 ` Steven Price
2021-06-28 9:46 ` Lucas Stach
0 siblings, 2 replies; 4+ messages in thread
From: Boris Brezillon @ 2021-06-24 14:08 UTC (permalink / raw)
To: dri-devel
Cc: Tomeu Vizoso, Rob Herring, Steven Price, Boris Brezillon,
Alyssa Rosenzweig, Robin Murphy
The panfrost driver tries to kill in-flight jobs on FD close after
destroying the FD scheduler entities. For this to work properly, we
need to make sure the jobs popped from the scheduler entities have
been queued at the HW level before declaring the entity idle, otherwise
we might iterate over a list that doesn't contain those jobs.
Suggested-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Cc: Lucas Stach <l.stach@pengutronix.de>
---
drivers/gpu/drm/scheduler/sched_main.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index 81496ae2602e..aa776ebe326a 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -811,10 +811,10 @@ static int drm_sched_main(void *param)
sched_job = drm_sched_entity_pop_job(entity);
- complete(&entity->entity_idle);
-
- if (!sched_job)
+ if (!sched_job) {
+ complete(&entity->entity_idle);
continue;
+ }
s_fence = sched_job->s_fence;
@@ -823,6 +823,7 @@ static int drm_sched_main(void *param)
trace_drm_run_job(sched_job, entity);
fence = sched->ops->run_job(sched_job);
+ complete(&entity->entity_idle);
drm_sched_fence_scheduled(s_fence);
if (!IS_ERR_OR_NULL(fence)) {
--
2.31.1
^ permalink raw reply related [flat|nested] 4+ messages in thread* Re: [PATCH] drm/sched: Declare entity idle only after HW submission
2021-06-24 14:08 [PATCH] drm/sched: Declare entity idle only after HW submission Boris Brezillon
@ 2021-06-24 15:24 ` Steven Price
2021-06-28 9:46 ` Lucas Stach
1 sibling, 0 replies; 4+ messages in thread
From: Steven Price @ 2021-06-24 15:24 UTC (permalink / raw)
To: Boris Brezillon, dri-devel
Cc: Rob Herring, Robin Murphy, Alyssa Rosenzweig, Tomeu Vizoso
On 24/06/2021 15:08, Boris Brezillon wrote:
> The panfrost driver tries to kill in-flight jobs on FD close after
> destroying the FD scheduler entities. For this to work properly, we
> need to make sure the jobs popped from the scheduler entities have
> been queued at the HW level before declaring the entity idle, otherwise
> we might iterate over a list that doesn't contain those jobs.
>
> Suggested-by: Lucas Stach <l.stach@pengutronix.de>
> Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
> Cc: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Steven Price <steven.price@arm.com>
> ---
> drivers/gpu/drm/scheduler/sched_main.c | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
> index 81496ae2602e..aa776ebe326a 100644
> --- a/drivers/gpu/drm/scheduler/sched_main.c
> +++ b/drivers/gpu/drm/scheduler/sched_main.c
> @@ -811,10 +811,10 @@ static int drm_sched_main(void *param)
>
> sched_job = drm_sched_entity_pop_job(entity);
>
> - complete(&entity->entity_idle);
> -
> - if (!sched_job)
> + if (!sched_job) {
> + complete(&entity->entity_idle);
> continue;
> + }
>
> s_fence = sched_job->s_fence;
>
> @@ -823,6 +823,7 @@ static int drm_sched_main(void *param)
>
> trace_drm_run_job(sched_job, entity);
> fence = sched->ops->run_job(sched_job);
> + complete(&entity->entity_idle);
> drm_sched_fence_scheduled(s_fence);
>
> if (!IS_ERR_OR_NULL(fence)) {
>
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] drm/sched: Declare entity idle only after HW submission
2021-06-24 14:08 [PATCH] drm/sched: Declare entity idle only after HW submission Boris Brezillon
2021-06-24 15:24 ` Steven Price
@ 2021-06-28 9:46 ` Lucas Stach
2021-06-28 11:18 ` Boris Brezillon
1 sibling, 1 reply; 4+ messages in thread
From: Lucas Stach @ 2021-06-28 9:46 UTC (permalink / raw)
To: Boris Brezillon, dri-devel
Cc: Rob Herring, Robin Murphy, Alyssa Rosenzweig, Tomeu Vizoso, Steven Price
Am Donnerstag, dem 24.06.2021 um 16:08 +0200 schrieb Boris Brezillon:
> The panfrost driver tries to kill in-flight jobs on FD close after
> destroying the FD scheduler entities. For this to work properly, we
> need to make sure the jobs popped from the scheduler entities have
> been queued at the HW level before declaring the entity idle, otherwise
> we might iterate over a list that doesn't contain those jobs.
>
> Suggested-by: Lucas Stach <l.stach@pengutronix.de>
> Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
> Cc: Lucas Stach <l.stach@pengutronix.de>
Not sure how much it's worth to review my own suggestion, but the
implementation looks correct to me.
I don't see any downsides for the existing drivers and it solves the
race window for drivers that want to cancel jobs on the HW submission
queue, without introducing yet another synchronization point.
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
> ---
> drivers/gpu/drm/scheduler/sched_main.c | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
> index 81496ae2602e..aa776ebe326a 100644
> --- a/drivers/gpu/drm/scheduler/sched_main.c
> +++ b/drivers/gpu/drm/scheduler/sched_main.c
> @@ -811,10 +811,10 @@ static int drm_sched_main(void *param)
>
> sched_job = drm_sched_entity_pop_job(entity);
>
> - complete(&entity->entity_idle);
> -
> - if (!sched_job)
> + if (!sched_job) {
> + complete(&entity->entity_idle);
> continue;
> + }
>
> s_fence = sched_job->s_fence;
>
> @@ -823,6 +823,7 @@ static int drm_sched_main(void *param)
>
> trace_drm_run_job(sched_job, entity);
> fence = sched->ops->run_job(sched_job);
> + complete(&entity->entity_idle);
> drm_sched_fence_scheduled(s_fence);
>
> if (!IS_ERR_OR_NULL(fence)) {
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] drm/sched: Declare entity idle only after HW submission
2021-06-28 9:46 ` Lucas Stach
@ 2021-06-28 11:18 ` Boris Brezillon
0 siblings, 0 replies; 4+ messages in thread
From: Boris Brezillon @ 2021-06-28 11:18 UTC (permalink / raw)
To: Lucas Stach
Cc: Tomeu Vizoso, dri-devel, Steven Price, Rob Herring,
Alyssa Rosenzweig, Robin Murphy
On Mon, 28 Jun 2021 11:46:08 +0200
Lucas Stach <l.stach@pengutronix.de> wrote:
> Am Donnerstag, dem 24.06.2021 um 16:08 +0200 schrieb Boris Brezillon:
> > The panfrost driver tries to kill in-flight jobs on FD close after
> > destroying the FD scheduler entities. For this to work properly, we
> > need to make sure the jobs popped from the scheduler entities have
> > been queued at the HW level before declaring the entity idle, otherwise
> > we might iterate over a list that doesn't contain those jobs.
> >
> > Suggested-by: Lucas Stach <l.stach@pengutronix.de>
> > Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
> > Cc: Lucas Stach <l.stach@pengutronix.de>
>
> Not sure how much it's worth to review my own suggestion, but the
> implementation looks correct to me.
> I don't see any downsides for the existing drivers and it solves the
> race window for drivers that want to cancel jobs on the HW submission
> queue, without introducing yet another synchronization point.
>
> Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Queued to drm-misc-next.
Thanks,
Boris
>
> > ---
> > drivers/gpu/drm/scheduler/sched_main.c | 7 ++++---
> > 1 file changed, 4 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
> > index 81496ae2602e..aa776ebe326a 100644
> > --- a/drivers/gpu/drm/scheduler/sched_main.c
> > +++ b/drivers/gpu/drm/scheduler/sched_main.c
> > @@ -811,10 +811,10 @@ static int drm_sched_main(void *param)
> >
> > sched_job = drm_sched_entity_pop_job(entity);
> >
> > - complete(&entity->entity_idle);
> > -
> > - if (!sched_job)
> > + if (!sched_job) {
> > + complete(&entity->entity_idle);
> > continue;
> > + }
> >
> > s_fence = sched_job->s_fence;
> >
> > @@ -823,6 +823,7 @@ static int drm_sched_main(void *param)
> >
> > trace_drm_run_job(sched_job, entity);
> > fence = sched->ops->run_job(sched_job);
> > + complete(&entity->entity_idle);
> > drm_sched_fence_scheduled(s_fence);
> >
> > if (!IS_ERR_OR_NULL(fence)) {
>
>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-06-28 11:18 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-06-24 14:08 [PATCH] drm/sched: Declare entity idle only after HW submission Boris Brezillon
2021-06-24 15:24 ` Steven Price
2021-06-28 9:46 ` Lucas Stach
2021-06-28 11:18 ` Boris Brezillon
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.