From: Pekka Paalanen <ppaalanen@gmail.com>
To: "Christian König" <christian.koenig@amd.com>
Cc: "Rob Clark" <robdclark@gmail.com>,
dri-devel@lists.freedesktop.org, freedreno@lists.freedesktop.org,
"Daniel Vetter" <daniel@ffwll.ch>,
"Christian König" <ckoenig.leichtzumerken@gmail.com>,
"Michel Dänzer" <michel@daenzer.net>,
"Tvrtko Ursulin" <tvrtko.ursulin@intel.com>,
"Rodrigo Vivi" <rodrigo.vivi@intel.com>,
"Alex Deucher" <alexander.deucher@amd.com>,
"Simon Ser" <contact@emersion.fr>,
"Rob Clark" <robdclark@chromium.org>,
"Sumit Semwal" <sumit.semwal@linaro.org>,
"Gustavo Padovan" <gustavo@padovan.org>,
"open list:SYNC FILE FRAMEWORK" <linux-media@vger.kernel.org>,
"moderated list:DMA BUFFER SHARING FRAMEWORK"
<linaro-mm-sig@lists.linaro.org>,
"open list" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v4 06/14] dma-buf/sync_file: Support (E)POLLPRI
Date: Tue, 21 Feb 2023 10:38:41 +0200 [thread overview]
Message-ID: <20230221103841.0d501f01@eldfell> (raw)
In-Reply-To: <2e6e9581-6de8-6aca-3e73-946fbc6ad2a3@amd.com>
[-- Attachment #1: Type: text/plain, Size: 2602 bytes --]
On Mon, 20 Feb 2023 09:31:41 +0100
Christian König <christian.koenig@amd.com> wrote:
> Am 18.02.23 um 22:15 schrieb Rob Clark:
> > From: Rob Clark <robdclark@chromium.org>
> >
> > Allow userspace to use the EPOLLPRI/POLLPRI flag to indicate an urgent
> > wait (as opposed to a "housekeeping" wait to know when to cleanup after
> > some work has completed). Usermode components of GPU driver stacks
> > often poll() on fence fd's to know when it is safe to do things like
> > free or reuse a buffer, but they can also poll() on a fence fd when
> > waiting to read back results from the GPU. The EPOLLPRI/POLLPRI flag
> > lets the kernel differentiate these two cases.
> >
> > Signed-off-by: Rob Clark <robdclark@chromium.org>
>
> The code looks clean, but the different poll flags and their meaning are
> certainly not my field of expertise.
A good question. epoll_ctl manual refers to poll(2) which says:
POLLPRI
There is some exceptional condition on the file descriptor. Possibilities include:
• There is out-of-band data on a TCP socket (see tcp(7)).
• A pseudoterminal master in packet mode has seen a state change on the slave (see ioctl_tty(2)).
• A cgroup.events file has been modified (see cgroups(7)).
It seems to be about selecting what events will trigger the poll,
more than how (fast) to poll. At least it is not documented to be
ignored in 'events', so I guess it should work.
Thanks,
pq
> Feel free to add Acked-by: Christian König <christian.koenig@amd.com>,
> somebody with more background in this should probably take a look as well.
>
> Regards,
> Christian.
>
> > ---
> > drivers/dma-buf/sync_file.c | 8 ++++++++
> > 1 file changed, 8 insertions(+)
> >
> > diff --git a/drivers/dma-buf/sync_file.c b/drivers/dma-buf/sync_file.c
> > index fb6ca1032885..c30b2085ee0a 100644
> > --- a/drivers/dma-buf/sync_file.c
> > +++ b/drivers/dma-buf/sync_file.c
> > @@ -192,6 +192,14 @@ static __poll_t sync_file_poll(struct file *file, poll_table *wait)
> > {
> > struct sync_file *sync_file = file->private_data;
> >
> > + /*
> > + * The POLLPRI/EPOLLPRI flag can be used to signal that
> > + * userspace wants the fence to signal ASAP, express this
> > + * as an immediate deadline.
> > + */
> > + if (poll_requested_events(wait) & EPOLLPRI)
> > + dma_fence_set_deadline(sync_file->fence, ktime_get());
> > +
> > poll_wait(file, &sync_file->wq, wait);
> >
> > if (list_empty(&sync_file->cb.node) &&
>
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2023-02-21 8:39 UTC|newest]
Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-18 21:15 [PATCH v4 00/14] dma-fence: Deadline awareness Rob Clark
2023-02-18 21:15 ` [PATCH v4 01/14] dma-buf/dma-fence: Add deadline awareness Rob Clark
2023-02-22 10:23 ` Tvrtko Ursulin
2023-02-22 15:28 ` Christian König
2023-02-22 17:04 ` Tvrtko Ursulin
2023-02-22 17:16 ` Rob Clark
2023-02-22 17:33 ` Tvrtko Ursulin
2023-02-22 18:57 ` Rob Clark
2023-02-22 11:01 ` Luben Tuikov
2023-02-18 21:15 ` [PATCH v4 02/14] dma-buf/fence-array: Add fence deadline support Rob Clark
2023-02-18 21:15 ` [PATCH v4 03/14] dma-buf/fence-chain: " Rob Clark
2023-02-22 10:27 ` Tvrtko Ursulin
2023-02-22 15:55 ` Rob Clark
2023-02-18 21:15 ` [PATCH v4 04/14] dma-buf/dma-resv: Add a way to set fence deadline Rob Clark
2023-02-20 8:16 ` Christian König
2023-02-18 21:15 ` [PATCH v4 05/14] dma-buf/sync_file: Add SET_DEADLINE ioctl Rob Clark
2023-02-20 8:27 ` Christian König
2023-02-20 16:09 ` Rob Clark
2023-02-21 8:41 ` Pekka Paalanen
2023-02-23 9:19 ` Christian König
2023-02-20 8:48 ` Pekka Paalanen
2023-02-18 21:15 ` [PATCH v4 06/14] dma-buf/sync_file: Support (E)POLLPRI Rob Clark
2023-02-20 8:31 ` Christian König
2023-02-21 8:38 ` Pekka Paalanen [this message]
2023-02-20 8:53 ` Pekka Paalanen
2023-02-20 16:14 ` Rob Clark
2023-02-21 8:37 ` Pekka Paalanen
2023-02-21 16:01 ` Sebastian Wick
2023-02-21 17:55 ` Rob Clark
2023-02-21 16:48 ` Luben Tuikov
2023-02-21 17:53 ` Rob Clark
2023-02-22 9:49 ` Pekka Paalanen
2023-02-22 10:26 ` Luben Tuikov
2023-02-22 15:37 ` Rob Clark
2023-02-23 9:38 ` Pekka Paalanen
2023-02-23 18:51 ` Rob Clark
2023-02-24 9:26 ` Pekka Paalanen
2023-02-24 9:41 ` Tvrtko Ursulin
2023-02-24 10:24 ` Pekka Paalanen
2023-02-24 10:50 ` Tvrtko Ursulin
2023-02-24 11:00 ` Pekka Paalanen
2023-02-24 11:37 ` Tvrtko Ursulin
2023-02-24 15:26 ` Luben Tuikov
2023-02-24 17:59 ` Rob Clark
2023-02-27 21:35 ` Rodrigo Vivi
2023-02-27 22:20 ` Rob Clark
2023-02-27 22:44 ` Sebastian Wick
2023-02-27 23:48 ` Rob Clark
2023-02-28 14:30 ` Sebastian Wick
2023-02-28 22:52 ` Rob Clark
2023-03-01 15:31 ` Sebastian Wick
2023-03-01 16:02 ` Rob Clark
2023-03-01 15:45 ` Rodrigo Vivi
2023-02-24 16:59 ` Rob Clark
2023-02-24 19:44 ` Rob Clark
2023-02-27 9:34 ` Pekka Paalanen
2023-02-27 18:43 ` Rob Clark
2023-02-18 21:15 ` [PATCH v4 07/14] dma-buf/sw_sync: Add fence deadline support Rob Clark
2023-02-20 8:29 ` Christian König
2023-02-18 21:15 ` [PATCH v4 08/14] drm/scheduler: " Rob Clark
2023-02-21 19:40 ` Luben Tuikov
2023-02-18 21:15 ` [PATCH v4 12/14] drm/msm: Add deadline based boost support Rob Clark
2023-02-18 21:15 ` [PATCH v4 14/14] drm/i915: " Rob Clark
2023-02-20 15:46 ` Tvrtko Ursulin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230221103841.0d501f01@eldfell \
--to=ppaalanen@gmail.com \
--cc=alexander.deucher@amd.com \
--cc=christian.koenig@amd.com \
--cc=ckoenig.leichtzumerken@gmail.com \
--cc=contact@emersion.fr \
--cc=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=freedreno@lists.freedesktop.org \
--cc=gustavo@padovan.org \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=michel@daenzer.net \
--cc=robdclark@chromium.org \
--cc=robdclark@gmail.com \
--cc=rodrigo.vivi@intel.com \
--cc=sumit.semwal@linaro.org \
--cc=tvrtko.ursulin@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox