From: Philipp Stanner <phasta@mailbox.org>
To: "Maíra Canal" <mcanal@igalia.com>,
"Matthew Brost" <matthew.brost@intel.com>,
"Danilo Krummrich" <dakr@kernel.org>,
"Philipp Stanner" <phasta@kernel.org>,
"Christian König" <ckoenig.leichtzumerken@gmail.com>,
"Tvrtko Ursulin" <tvrtko.ursulin@igalia.com>,
"Simona Vetter" <simona@ffwll.ch>,
"David Airlie" <airlied@gmail.com>,
"Melissa Wen" <mwen@igalia.com>,
"Lucas Stach" <l.stach@pengutronix.de>,
"Russell King" <linux+etnaviv@armlinux.org.uk>,
"Christian Gmeiner" <christian.gmeiner@gmail.com>,
"Lucas De Marchi" <lucas.demarchi@intel.com>,
"Thomas Hellström" <thomas.hellstrom@linux.intel.com>,
"Rodrigo Vivi" <rodrigo.vivi@intel.com>,
"Boris Brezillon" <boris.brezillon@collabora.com>,
"Rob Herring" <robh@kernel.org>,
"Steven Price" <steven.price@arm.com>,
"Liviu Dudau" <liviu.dudau@arm.com>
Cc: kernel-dev@igalia.com, dri-devel@lists.freedesktop.org,
etnaviv@lists.freedesktop.org, intel-xe@lists.freedesktop.org
Subject: Re: [PATCH v3 7/8] drm/xe: Use DRM_GPU_SCHED_STAT_NO_HANG to skip the reset
Date: Mon, 23 Jun 2025 16:28:29 +0200 [thread overview]
Message-ID: <066b0328ff184b42a666a94c9ac61eebc5b65083.camel@mailbox.org> (raw)
In-Reply-To: <20250618-sched-skip-reset-v3-7-8be5cca2725d@igalia.com>
On Wed, 2025-06-18 at 11:47 -0300, Maíra Canal wrote:
> Xe can skip the reset if TDR has fired before the free job worker and
> can
> also re-arm the timeout timer in some scenarios. Instead of
> manipulating
> scheduler's internals, inform the scheduler that the job did not
> actually
> timeout and no reset was performed through the new status code
> DRM_GPU_SCHED_STAT_NO_HANG.
>
> Note that, in the first case, there is no need to restart submission
> if it
> hasn't been stopped.
>
> Signed-off-by: Maíra Canal <mcanal@igalia.com>
Did you have the opportunity to test that one?
If not, at least a RB from one of the Intel folks is likely a desirable
thing, since the changes are non-trivial.
P.
> ---
> drivers/gpu/drm/xe/xe_guc_submit.c | 12 +++---------
> 1 file changed, 3 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c
> b/drivers/gpu/drm/xe/xe_guc_submit.c
> index
> 9c7e445b9ea7ce7e3610eadca023e6d810e683e9..f6289eeffd852e40b33d0e455d9
> bcc21a4fb1467 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -1078,12 +1078,8 @@ guc_exec_queue_timedout_job(struct
> drm_sched_job *drm_job)
> * list so job can be freed and kick scheduler ensuring free
> job is not
> * lost.
> */
> - if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &job->fence-
> >flags)) {
> - xe_sched_add_pending_job(sched, job);
> - xe_sched_submission_start(sched);
> -
> - return DRM_GPU_SCHED_STAT_RESET;
> - }
> + if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &job->fence-
> >flags))
> + return DRM_GPU_SCHED_STAT_NO_HANG;
>
> /* Kill the run_job entry point */
> xe_sched_submission_stop(sched);
> @@ -1261,10 +1257,8 @@ guc_exec_queue_timedout_job(struct
> drm_sched_job *drm_job)
> * but there is not currently an easy way to do in DRM
> scheduler. With
> * some thought, do this in a follow up.
> */
> - xe_sched_add_pending_job(sched, job);
> xe_sched_submission_start(sched);
> -
> - return DRM_GPU_SCHED_STAT_RESET;
> + return DRM_GPU_SCHED_STAT_NO_HANG;
> }
>
> static void __guc_exec_queue_fini_async(struct work_struct *w)
>
next prev parent reply other threads:[~2025-06-23 14:28 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-18 14:47 [PATCH v3 0/8] drm/sched: Allow drivers to skip the reset with DRM_GPU_SCHED_STAT_NO_HANG Maíra Canal
2025-06-18 14:47 ` [PATCH v3 1/8] drm/sched: Rename DRM_GPU_SCHED_STAT_NOMINAL to DRM_GPU_SCHED_STAT_RESET Maíra Canal
2025-06-30 11:39 ` Philipp Stanner
2025-06-18 14:47 ` [PATCH v3 2/8] drm/sched: Allow drivers to skip the reset and keep on running Maíra Canal
2025-06-30 11:25 ` Maíra Canal
2025-06-30 11:46 ` Philipp Stanner
2025-06-18 14:47 ` [PATCH v3 3/8] drm/sched: Make timeout KUnit tests faster Maíra Canal
2025-06-30 11:53 ` Philipp Stanner
2025-06-30 12:05 ` Maíra Canal
2025-06-30 12:20 ` Philipp Stanner
2025-06-30 14:04 ` Maíra Canal
2025-07-02 14:41 ` Philipp Stanner
2025-06-18 14:47 ` [PATCH v3 4/8] drm/sched: Add new test for DRM_GPU_SCHED_STAT_NO_HANG Maíra Canal
2025-06-24 9:09 ` Tvrtko Ursulin
2025-06-18 14:47 ` [PATCH v3 5/8] drm/v3d: Use DRM_GPU_SCHED_STAT_NO_HANG to skip the reset Maíra Canal
2025-06-24 9:14 ` Tvrtko Ursulin
2025-06-18 14:47 ` [PATCH v3 6/8] drm/etnaviv: " Maíra Canal
2025-06-24 9:37 ` Lucas Stach
2025-06-18 14:47 ` [PATCH v3 7/8] drm/xe: " Maíra Canal
2025-06-23 14:28 ` Philipp Stanner [this message]
2025-06-18 14:47 ` [PATCH v3 8/8] drm/panfrost: " Maíra Canal
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=066b0328ff184b42a666a94c9ac61eebc5b65083.camel@mailbox.org \
--to=phasta@mailbox.org \
--cc=airlied@gmail.com \
--cc=boris.brezillon@collabora.com \
--cc=christian.gmeiner@gmail.com \
--cc=ckoenig.leichtzumerken@gmail.com \
--cc=dakr@kernel.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=etnaviv@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=kernel-dev@igalia.com \
--cc=l.stach@pengutronix.de \
--cc=linux+etnaviv@armlinux.org.uk \
--cc=liviu.dudau@arm.com \
--cc=lucas.demarchi@intel.com \
--cc=matthew.brost@intel.com \
--cc=mcanal@igalia.com \
--cc=mwen@igalia.com \
--cc=phasta@kernel.org \
--cc=robh@kernel.org \
--cc=rodrigo.vivi@intel.com \
--cc=simona@ffwll.ch \
--cc=steven.price@arm.com \
--cc=thomas.hellstrom@linux.intel.com \
--cc=tvrtko.ursulin@igalia.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;
as well as URLs for NNTP newsgroup(s).