From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?UTF-8?Q?Christian_K=c3=b6nig?= Subject: Re: [PATCH] drm/sched: Always trace the dependencies we wait on, to fix a race. Date: Fri, 8 Feb 2019 09:43:04 +0100 Message-ID: <0d7c09cf-caeb-c706-928e-2a668ae78e53@gmail.com> References: <20181207191653.30118-1-eric@anholt.net> <87tvhf2o0l.fsf@anholt.net> Reply-To: christian.koenig@amd.com Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============1144892552==" Return-path: In-Reply-To: <87tvhf2o0l.fsf@anholt.net> Content-Language: en-US List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: Eric Anholt , "Koenig, Christian" , "dri-devel@lists.freedesktop.org" , "amd-gfx@lists.freedesktop.org" Cc: "linux-kernel@vger.kernel.org" List-Id: amd-gfx.lists.freedesktop.org This is a multi-part message in MIME format. --===============1144892552== Content-Type: multipart/alternative; boundary="------------54F459BADCFDA435A524D0A3" Content-Language: en-US This is a multi-part message in MIME format. --------------54F459BADCFDA435A524D0A3 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Am 08.02.19 um 00:10 schrieb Eric Anholt: > "Koenig, Christian" writes: > >> Am 07.12.18 um 20:16 schrieb Eric Anholt: >>> The entity->dependency can go away completely once we've called >>> drm_sched_entity_add_dependency_cb() (if the cb is called before we >>> get around to tracing). The tracepoint is more useful if we trace >>> every dependency instead of just ones that get callbacks installed, >>> anyway, so just do that. >>> >>> Fixes any easy-to-produce OOPS when tracing the scheduler on V3D with >>> "perf record -a -e gpu_scheduler:.\* glxgears" and DEBUG_SLAB enabled. >>> >>> Signed-off-by: Eric Anholt >> Reviewed-by: Christian König >> >> Going to pick that up for upstream and will add with a CC: stable. > Looks like this got misplaced. My fault, pushed to our internal branch now with a CC: stable tag on it. Christian. > > _______________________________________________ > amd-gfx mailing list > amd-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/amd-gfx --------------54F459BADCFDA435A524D0A3 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 8bit
Am 08.02.19 um 00:10 schrieb Eric Anholt:
"Koenig, Christian" <Christian.Koenig@amd.com> writes:


Am 07.12.18 um 20:16 schrieb Eric Anholt:

The entity->dependency can go away completely once we've called
drm_sched_entity_add_dependency_cb() (if the cb is called before we
get around to tracing).  The tracepoint is more useful if we trace
every dependency instead of just ones that get callbacks installed,
anyway, so just do that.

Fixes any easy-to-produce OOPS when tracing the scheduler on V3D with
"perf record -a -e gpu_scheduler:.\* glxgears" and DEBUG_SLAB enabled.

Signed-off-by: Eric Anholt <eric@anholt.net>

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

Going to pick that up for upstream and will add with a CC: stable.

Looks like this got misplaced.

My fault, pushed to our internal branch now with a CC: stable tag on it.

Christian.


_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

--------------54F459BADCFDA435A524D0A3-- --===============1144892552== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KZHJpLWRldmVs IG1haWxpbmcgbGlzdApkcmktZGV2ZWxAbGlzdHMuZnJlZWRlc2t0b3Aub3JnCmh0dHBzOi8vbGlz dHMuZnJlZWRlc2t0b3Aub3JnL21haWxtYW4vbGlzdGluZm8vZHJpLWRldmVsCg== --===============1144892552==--