From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon@freedesktop.org Subject: [Bug 109498] Game enabling VBO Core but disabling VAO Core causing reproducible, predictable, spontaneous powerdown Date: Tue, 29 Jan 2019 16:56:00 +0000 Message-ID: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0860138956==" Return-path: Received: from culpepper.freedesktop.org (culpepper.freedesktop.org [IPv6:2610:10:20:722:a800:ff:fe98:4b55]) by gabe.freedesktop.org (Postfix) with ESMTP id 2F99E6E8D8 for ; Tue, 29 Jan 2019 16:56:04 +0000 (UTC) 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 --===============0860138956== Content-Type: multipart/alternative; boundary="15487809640.6AABdc.6711" Content-Transfer-Encoding: 7bit --15487809640.6AABdc.6711 Date: Tue, 29 Jan 2019 16:56:04 +0000 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated https://bugs.freedesktop.org/show_bug.cgi?id=3D109498 Bug ID: 109498 Summary: Game enabling VBO Core but disabling VAO Core causing reproducible, predictable, spontaneous powerdown Product: DRI Version: XOrg git Hardware: x86-64 (AMD64) OS: Linux (All) Status: NEW Severity: normal Priority: medium Component: DRM/AMDgpu Assignee: dri-devel@lists.freedesktop.org Reporter: awenocur@gmail.com The game "openAstroMenace", for some reason defaults to using VBO Core in t= he absence of VAO core, which predictably causes the game to crash on a modern OpenGL implementation. The game itself crashing is not a problem; it's the = PSU switching off that's very concerning. I'm running an ASRock E3V5 WS motherboard with a 4 GB Sapphire R9 Fury X, a= nd an E3-1235L CPU. I'm early on in diagnosing it so I don't have a stack trace, but I think I'= ve ruled out hardware problems, as this system doesn't break a sweat maxing out the Heaven benchmark or the Phoronix JuliaGPU (via ROCm) benchmark, and pla= ys every other game flawlessly. It can run with all CPU cores at 100% for hour= s on end. openAstroMenace runs flawlessly with max settings when VAO Core is enabled. Mesa may be malfunctioning here, but a userspace library wouldn't affect PSU behavior. This is pointing me in the direction of suspecting AMDGPU to be t= he culprit. It exhibits this behavior on both my Ubuntu and Gentoo installatio= ns, and on 4.15 and 4.18 series kernels. The system poweroff is predictable dow= n to less than half a second, and is triggered exactly the same way at exactly t= he same time in every instance. It occurs around 3/4 of a second after the gam= e's menu appears. --=20 You are receiving this mail because: You are the assignee for the bug.= --15487809640.6AABdc.6711 Date: Tue, 29 Jan 2019 16:56:04 +0000 MIME-Version: 1.0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated
Bug ID 109498
Summary Game enabling VBO Core but disabling VAO Core causing reprodu= cible, predictable, spontaneous powerdown
Product DRI
Version XOrg git
Hardware x86-64 (AMD64)
OS Linux (All)
Status NEW
Severity normal
Priority medium
Component DRM/AMDgpu
Assignee dri-devel@lists.freedesktop.org
Reporter awenocur@gmail.com

The game "openAstroMenace", for some reason defaults=
 to using VBO Core in the
absence of VAO core, which predictably causes the game to crash on a modern
OpenGL implementation. The game itself crashing is not a problem; it's the =
PSU
switching off that's very concerning.

I'm running an ASRock E3V5 WS motherboard with a 4 GB Sapphire R9 Fury X, a=
nd
an E3-1235L CPU.

I'm early on in diagnosing it so I don't have a stack trace, but I think I'=
ve
ruled out hardware problems, as this system doesn't break a sweat maxing out
the Heaven benchmark or the Phoronix JuliaGPU (via ROCm) benchmark, and pla=
ys
every other game flawlessly. It can run with all CPU cores at 100% for hour=
s on
end. openAstroMenace runs flawlessly with max settings when VAO Core is
enabled.

Mesa may be malfunctioning here, but a userspace library wouldn't affect PSU
behavior. This is pointing me in the direction of suspecting AMDGPU to be t=
he
culprit. It exhibits this behavior on both my Ubuntu and Gentoo installatio=
ns,
and on 4.15 and 4.18 series kernels. The system poweroff is predictable dow=
n to
less than half a second, and is triggered exactly the same way at exactly t=
he
same time in every instance. It occurs around 3/4 of a second after the gam=
e's
menu appears.


You are receiving this mail because:
  • You are the assignee for the bug.
= --15487809640.6AABdc.6711-- --===============0860138956== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KZHJpLWRldmVs IG1haWxpbmcgbGlzdApkcmktZGV2ZWxAbGlzdHMuZnJlZWRlc2t0b3Aub3JnCmh0dHBzOi8vbGlz dHMuZnJlZWRlc2t0b3Aub3JnL21haWxtYW4vbGlzdGluZm8vZHJpLWRldmVsCg== --===============0860138956==--