From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon@freedesktop.org Subject: [Bug 96731] [RADEONSI] [LLVM] [bisected] GPU lockups when running Alien: Isolation Date: Thu, 30 Jun 2016 23:58:57 +0000 Message-ID: References: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0875519266==" 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 CF11D6E911 for ; Thu, 30 Jun 2016 23:58:56 +0000 (UTC) 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 --===============0875519266== Content-Type: multipart/alternative; boundary="14673311360.2539b56e.13568"; charset="UTF-8" --14673311360.2539b56e.13568 Date: Thu, 30 Jun 2016 23:58:56 +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=3D96731 --- Comment #6 from Matt Arsenault --- The only obvious difference I see in the dump diffs without looking at any particular shader is the number of used registers changed. This is probably because previously the implicit uses of the super registers were missing wh= en the AsmPrinter counts them. If the dynamic was out of bounds, it is more li= kely to be out of bounds of the allocated VGPRs, in which case the hardware beha= vior is to return v0. If there are out of bounds accesses, it would now read an undefined register. I don't know if there are any actual out of bounds dyn= amic vector accesses --=20 You are receiving this mail because: You are the assignee for the bug.= --14673311360.2539b56e.13568 Date: Thu, 30 Jun 2016 23:58:56 +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

Comment= # 6 on bug 96731<= /a> from Matt Arsenault 
The only obvious difference I see in the dump diffs without lo=
oking at any
particular shader is the number of used registers changed. This is probably
because previously the implicit uses of the super registers were missing wh=
en
the AsmPrinter counts them. If the dynamic was out of bounds, it is more li=
kely
to be out of bounds of the allocated VGPRs, in which case the hardware beha=
vior
is to return v0. If there are out of bounds accesses, it would now read an
undefined register. I don't know if there are any actual  out of bounds dyn=
amic
vector accesses

You are receiving this mail because:
  • You are the assignee for the bug.
= --14673311360.2539b56e.13568-- --===============0875519266== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KZHJpLWRldmVs IG1haWxpbmcgbGlzdApkcmktZGV2ZWxAbGlzdHMuZnJlZWRlc2t0b3Aub3JnCmh0dHBzOi8vbGlz dHMuZnJlZWRlc2t0b3Aub3JnL21haWxtYW4vbGlzdGluZm8vZHJpLWRldmVsCg== --===============0875519266==--