https://bugs.freedesktop.org/show_bug.cgi?id=38581
Summary: [RADEON:KMS:R600G] wine, counter strike source crashes during gameplay Product: Mesa Version: git Platform: x86-64 (AMD64) OS/Version: Linux (All) Status: NEW Severity: normal Priority: medium Component: Drivers/Gallium/r600 AssignedTo: dri-devel@lists.freedesktop.org ReportedBy: bpaterni@gmail.com
since bug 35051 no longer seems to be a problem for me, I've been able to play the game somewhat. Now the problem is that the game always crashes during a firefight. I'm not sure what exactly triggers the crash, but for me this bug is always reproducible on the first engagement with the enemy (terrorists/counter terrorists)
Attached should be a complete wine log of the crash. I believe the problematic portion is up near the middle where a few shaders fail validation:
fixme:d3d_shader:print_glsl_info_log Error received from GLSL shader #89: fixme:d3d_shader:print_glsl_info_log error: shader uses too many varying components (36 > 32)
Hardware is rv770 (HD 4850) being run on mesa (git)
Let me know if I can provide more info.
https://bugs.freedesktop.org/show_bug.cgi?id=38581
--- Comment #1 from Brian Paterni bpaterni@gmail.com 2011-06-22 17:54:35 PDT --- Created an attachment (id=48311) --> (https://bugs.freedesktop.org/attachment.cgi?id=48311) wine steam/cssource log
https://bugs.freedesktop.org/show_bug.cgi?id=38581
--- Comment #2 from Henri Verbeet hverbeet@gmail.com 2011-06-23 02:47:59 PDT --- That's probably due to the driver reporting too few varyings.
https://bugs.freedesktop.org/show_bug.cgi?id=38581
Sven Arvidsson sa@whiz.se changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sa@whiz.se
--- Comment #3 from Sven Arvidsson sa@whiz.se 2011-06-23 03:55:22 PDT --- Commit de77324d8f14951e4dc17f570e49451a0cd33121introduced this.
Also discussed on mesa-dev: http://lists.freedesktop.org/archives/mesa-dev/2011-June/008344.html
https://bugs.freedesktop.org/show_bug.cgi?id=38581
Roland Scheidegger sroland@vmware.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #4 from Roland Scheidegger sroland@vmware.com 2011-06-23 07:39:38 PDT --- Should be fixed by 1e5cef96d184b00eb588b48ecd02386998077d82
https://bugs.freedesktop.org/show_bug.cgi?id=38581
Brian Paterni bpaterni@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|FIXED |
--- Comment #5 from Brian Paterni bpaterni@gmail.com 2011-06-23 09:06:29 PDT --- Actually, the driver reporting fewer varyings seems to be only a secondary issue and is not the root cause since git-1e5cef9 still experiences the crash.
I would try and upload an apitrace of the crash, but it looks like it has broken that as well. glretrace reports:
60: warning: unknown call glXDestroyContext 5243: warning: unknown call glXDestroyContext warning: incomplete call 5244 ()
warning: incomplete call 5245 ()
Rendered 0 frames in 0.091652 secs, average of 0 fps
https://bugs.freedesktop.org/show_bug.cgi?id=38581
--- Comment #6 from Roland Scheidegger sroland@vmware.com 2011-06-23 10:14:19 PDT --- No idea what's going on then. In any case there certainly should be no more reports of the "too many varying components", though this never should have caused a crash neither, so maybe what you're seeing was really totally unrelated to that.
https://bugs.freedesktop.org/show_bug.cgi?id=38581
Brian Paterni bpaterni@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|REOPENED |RESOLVED Resolution| |NOTOURBUG
--- Comment #7 from Brian Paterni bpaterni@gmail.com 2011-06-25 20:47:09 PDT --- It appears I didn't investigate this issue thoroughly enough before filing a bug report here. What I've been seeing is just http://bugs.winehq.org/show_bug.cgi?id=7698 .
In case anyone else is wondering I ended up working around 7698 by following the instructions in the 'known issues' section of the counter strike appdb entry: http://appdb.winehq.org/objectManager.php?sClass=version&iId=3731 (the 'ClientScheme.res' method)
CSSource now plays reasonably well. There are a few artifacts popping in and out over time. Though these artifacts are a great deal less of obstruction to gameplay as the ETQW artifacts discussing in bug 38452 on fdo
dri-devel@lists.freedesktop.org
-
bugzilla-daemon＠freedesktop.org