https://bugs.freedesktop.org/show_bug.cgi?id=40034
Summary: E-350 misprocesses shader Product: Mesa Version: git Platform: x86 (IA32) OS/Version: Linux (All) Status: NEW Severity: normal Priority: medium Component: Drivers/Gallium/r600 AssignedTo: dri-devel@lists.freedesktop.org ReportedBy: curaga@operamail.com
Created an attachment (id=50154) --> (https://bugs.freedesktop.org/attachment.cgi?id=50154) Screenshot of the failure
If you pull from my e350bug branch ( http://cgit.freedesktop.org/~cand/mesa/log/?h=e350bug ) and execute "pp_jimenezmlaa=8 glxgears", you should see a fully white window. What actually happens on the E-350 is attached.
The gist is "a long shader; output = white;", which should always create white, since the shader has no kilp/discard. While this would rarely happen in practise (the glsl compiler would eliminate dead code), this is a clear bug, and I believe this is causing me other grief.
While the branch is based on ~4 weeks old master, the bug is present when merged with today's master.
The kernel I'm on is 3.0.1.
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #1 from Lauri Kasanen curaga@operamail.com 2011-08-12 01:14:48 PDT --- This test succeeds on both softpipe and llvmpipe.
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #2 from Lauri Kasanen curaga@operamail.com 2011-08-16 04:21:46 PDT --- I have isolated this to the loops. Commenting out all BGNLOOP, ENDLOOP, and BRK instructions lets the card produce white.
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #3 from Lauri Kasanen curaga@operamail.com 2011-08-16 04:50:15 PDT --- Tracking further, the bug only shows when there's more than one break inside a loop.
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #4 from Lauri Kasanen curaga@operamail.com 2011-08-16 05:04:11 PDT --- Scratch that, it appears with only one break too (depends on which break you comment out :P)
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #5 from Lauri Kasanen curaga@operamail.com 2011-08-16 05:52:55 PDT --- Created an attachment (id=50268) View: https://bugs.freedesktop.org/attachment.cgi?id=50268 Review: https://bugs.freedesktop.org/review?bug=40034&attachment=50268
Proposed patch
The attached patch fixes this bug. It also lets MLAA work, showing that there are no other showstoppers for it on r600g / e-350.
Note I'm not at all sure this is the right approach, but hey, it works.
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #6 from Lauri Kasanen curaga@operamail.com 2011-11-04 09:27:15 PDT --- This bug doesn't seem to affect r700 - at least my HD4350 works perfectly with all effects on current master (7.12-devel (git-f800a29))
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #7 from Michel Dänzer michel@daenzer.net 2011-11-07 09:19:59 UTC --- Please send the patch to the mesa-dev mailing list for review, preferably with git send-email or at least generated by git format-patch.
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #8 from Lauri Kasanen curaga@operamail.com 2011-11-08 05:06:44 PST --- I can't; the same function is used for everything from r600 on - I have no idea whether it would break other hw. As mentioned in the comment above, the bug is not there on a r700 dedicated card on master, which casts further doubt on the patch.
It may simply be an issue in Evergreen loop handling elsewhere, but I really do not understand the hw well enough to say that with any certainty.
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #9 from Alex Deucher agd5f@yahoo.com 2012-01-24 05:58:00 PST --- Is this still an issue with latest git? This commit looks relevant: http://cgit.freedesktop.org/mesa/mesa/commit/?id=d89c96c75dbb9c003e4643942f2...
https://bugs.freedesktop.org/show_bug.cgi?id=40034
--- Comment #10 from Lauri Kasanen curaga@operamail.com 2012-01-24 07:11:39 PST --- I can test on the weekend.
https://bugs.freedesktop.org/show_bug.cgi?id=40034
Lauri Kasanen curaga@operamail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #11 from Lauri Kasanen curaga@operamail.com 2012-01-27 02:56:11 PST --- Yes, that commit fixes it. Doubly sure, since I had to apply it on top of master from Dec 19 (the last version that builds without xcb).
dri-devel@lists.freedesktop.org
-
bugzilla-daemon＠freedesktop.org