From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon@freedesktop.org Subject: [Bug 85320] [RV620][RV630][RS880] GPU hangs using UVD hardware acceleration Date: Fri, 01 May 2015 16:23:19 +0000 Message-ID: References: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0651949361==" Return-path: Received: from culpepper.freedesktop.org (unknown [131.252.210.165]) by gabe.freedesktop.org (Postfix) with ESMTP id 4ADCB6E916 for ; Fri, 1 May 2015 09:23:19 -0700 (PDT) In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: dri-devel@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org --===============0651949361== Content-Type: multipart/alternative; boundary="1430497399.fFaDdCAA4.24369"; charset="UTF-8" --1430497399.fFaDdCAA4.24369 Date: Fri, 1 May 2015 16:23:19 +0000 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable https://bugs.freedesktop.org/show_bug.cgi?id=3D85320 --- Comment #46 from Christian K=C3=B6nig --- (In reply to Nicola Mori from comment #44) > The new patch works also for me. A couple of questions, Christian: does t= he > patch remove some features? Do you think to mainline it or rather impleme= nt > a different fix now that the problem seems to be better defined? Thanks. Instead of submitting the commands to the hardware directly with semaphores= to sync between the GFX and UVD engines we block until the dependent task is completed. That's rather bad in a couple of different cases, for example doing 3D gami= ng and video playback at the same time. What essentially happens is instead of keeping UVD and GFX busy all the same time (and only occasionally block one engine waiting the other one) you do = it more like this: 1. Run UVD job. 2. Wait for UVD to finish. 3. Run GFX. 4. Wait for GFX to finish. 5. Run UVD 6. Wait for UVD to finish. .... --=20 You are receiving this mail because: You are the assignee for the bug. --1430497399.fFaDdCAA4.24369 Date: Fri, 1 May 2015 16:23:19 +0000 MIME-Version: 1.0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Commen= t # 46 on bug 85320<= /a> from Christian K=C3=B6nig 
(In reply to Nicola Mori from comment #44)
> The new patch works also for me. A couple of que=
stions, Christian: does the
> patch remove some features? Do you think to mainline it or rather impl=
ement
> a different fix now that the problem seems to be better defined? Thank=
s.

Instead of submitting the commands to the hardware directly with semaphores=
 to
sync between the GFX and UVD engines we block until the dependent task is
completed.

That's rather bad in a couple of different cases, for example doing 3D gami=
ng
and video playback at the same time.

What essentially happens is instead of keeping UVD and GFX busy all the same
time (and only occasionally block one engine waiting the other one) you do =
it
more like this:

1. Run UVD job.
2. Wait for UVD to finish.
3. Run GFX.
4. Wait for GFX to finish.
5. Run UVD
6. Wait for UVD to finish.
....

You are receiving this mail because: =20=20=20=20=20=20
  • You are the assignee for the bug.
--1430497399.fFaDdCAA4.24369-- --===============0651949361== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KZHJpLWRldmVs IG1haWxpbmcgbGlzdApkcmktZGV2ZWxAbGlzdHMuZnJlZWRlc2t0b3Aub3JnCmh0dHA6Ly9saXN0 cy5mcmVlZGVza3RvcC5vcmcvbWFpbG1hbi9saXN0aW5mby9kcmktZGV2ZWwK --===============0651949361==--