https://bugs.freedesktop.org/show_bug.cgi?id=28000
Summary: [Mobile GM965/GL96][Wine][3D] Display and textures corrupted in Warhammer 40k Soulstorm Product: DRI Version: XOrg CVS Platform: x86 (IA32) OS/Version: Linux (All) Status: NEW Severity: critical Priority: medium Component: General AssignedTo: dri-devel@lists.freedesktop.org ReportedBy: fdelente@mail.cpod.fr
Created an attachment (id=35465) --> (https://bugs.freedesktop.org/attachment.cgi?id=35465) Xorg.0.log
$ lspci 00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 03) 00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 03) $ uname -m i686 $ uname -a Linux slick 2.6.32.3-smp #7 SMP PREEMPT Tue Apr 6 19:10:03 CEST 2010 i686 Intel(R) Core(TM)2 Duo CPU T5750 @ 2.00GHz GenuineIntel GNU/Linux -- xf86-video-intel/xserver/mesa/libdrm version: xf86-video-intel-2.11.0-i386-1 xorg-server-1.8.0-i386-1 Mesa-7.8.1-i386-1 libdrm-2.4.20-i386-1 -- kernel version: ("uname -r") 2.6.32.3-smp -- Linux distribution: Slackware 13.0 initially, but I recompiled many packages from source -- Machine or mobo model: Packard Bell BG46 fr -- Display connector: (e.g. VGA, DVI, HDMI, S-video, ...) VGA? LCD panel 3) Reproduce steps. Probability if not 100% reproducible. 100% reproducible: when starting Soulstorm, some bitmaps and 3D textures are corrupted. The same game works flawlessly on another computer with an Nvidia card. 4) Attachment: -- Xorg.0.log -- dmesg output (better with boot option "drm.debug=0x06") -- xorg.conf (if you changed any default options) No change there -- screenshot or photo (optional, a picture is worth a thousand words) -- output of "xrandr --verbose" for display mode issue -- intel_reg_dumper output (see the guide) for display issue. -- for GPU hang, get the last batch buffer (see the guide).
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #1 from fdelente@mail.cpod.fr 2010-05-06 11:34:11 PDT --- Created an attachment (id=35466) --> (https://bugs.freedesktop.org/attachment.cgi?id=35466) dmesg
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #2 from fdelente@mail.cpod.fr 2010-05-06 11:34:56 PDT --- Created an attachment (id=35467) --> (https://bugs.freedesktop.org/attachment.cgi?id=35467) xrandr --verbose
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #3 from fdelente@mail.cpod.fr 2010-05-06 11:36:09 PDT --- Created an attachment (id=35468) --> (https://bugs.freedesktop.org/attachment.cgi?id=35468) Wine output with LIBGL and MESA debug mode
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #4 from fdelente@mail.cpod.fr 2010-05-06 11:37:12 PDT --- Created an attachment (id=35469) --> (https://bugs.freedesktop.org/attachment.cgi?id=35469) Corruption of bitmaps
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #5 from fdelente@mail.cpod.fr 2010-05-06 11:38:25 PDT --- Created an attachment (id=35470) --> (https://bugs.freedesktop.org/attachment.cgi?id=35470) Textures corruption
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #6 from Sven Arvidsson sa@whiz.se 2010-05-06 14:13:31 PDT --- Having played both the original game and the Winter Assault expansion without problems using the i965 driver, I gave the demo of Soulstorm a try. It works fine here, but if I disable S3TC support, I get the same corruption as your screenshots.
Could you try either:
- Installing the libtxc_dxtn library (should be available somewhere for your distro, otherwise get it here: http://homepage.hispeed.ch/rscheidegger/dri_experimental/s3tc_index.html )
- Setting "Enable S3TC texture compression" to "Yes" in driconf. (or just force_s3tc_enable=true )
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #7 from fdelente@mail.cpod.fr 2010-05-07 06:28:14 PDT --- (In reply to comment #6)
Could you try either:
- Installing the libtxc_dxtn library (should be available somewhere for your
distro, otherwise get it here: http://homepage.hispeed.ch/rscheidegger/dri_experimental/s3tc_index.html )
Installing that library has solved the textures and bitmaps corruption problem, thanks! However, the GPU still hangs as mentionned in dmesg.
I tried getting a dump, but I couldn't find intel_reg_dumper in intel-gpu-tools-1.0.2, I only found intel_gpu_dump. However, it doesn't work as a user, and it hangs the machine as root, forcing me to reboot (I tried intel_gpu_dump > gpu_dump as root, and the machine becomes completely unresponsive).
- Setting "Enable S3TC texture compression" to "Yes" in driconf. (or just
force_s3tc_enable=true )
I couldn't find driconf, and I don't know where to put force_s3tc_enable=true; however, since installing libtxc_dxtn worked, I guess it isn't mandatory?
Thanks for your help!
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #8 from Sven Arvidsson sa@whiz.se 2010-05-07 07:02:53 PDT --- (In reply to comment #7)
Installing that library has solved the textures and bitmaps corruption problem, thanks! However, the GPU still hangs as mentionned in dmesg.
I tried getting a dump, but I couldn't find intel_reg_dumper in intel-gpu-tools-1.0.2, I only found intel_gpu_dump. However, it doesn't work as a user, and it hangs the machine as root, forcing me to reboot (I tried intel_gpu_dump > gpu_dump as root, and the machine becomes completely unresponsive).
The GPU reset function will probably mean that you won't get a meaningful dump with that kernel anyway. The best way is to retry with git master of Mesa, and the latest rc of 2.6.34.
If the hang is still there, grab a dump with intel_error_decode.
(At least this is what I usually do.)
I couldn't find driconf, and I don't know where to put force_s3tc_enable=true; however, since installing libtxc_dxtn worked, I guess it isn't mandatory?
No, shouldn't be necessary if the lib is used.
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #9 from fdelente@mail.cpod.fr 2010-05-07 09:36:34 PDT --- (In reply to comment #8) Thanks for your suggestions, installing the latest git of Mesa and/or 2.6.34-rc6 solved the GPU hang problem.
Now the game plays, but some models are distorted: the polygons extend out of the characters(?).
Furthermore, about 1 pixel is chopped of the letters (see screenshot).
Thanks again anyway!
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #10 from fdelente@mail.cpod.fr 2010-05-07 09:37:29 PDT --- Created an attachment (id=35500) --> (https://bugs.freedesktop.org/attachment.cgi?id=35500) Polygons stretched / models of characters distorted (not the buildings?)
https://bugs.freedesktop.org/show_bug.cgi?id=28000
--- Comment #11 from Sven Arvidsson sa@whiz.se 2010-05-07 11:24:09 PDT --- (In reply to comment #9)
Now the game plays, but some models are distorted: the polygons extend out of the characters(?).
Furthermore, about 1 pixel is chopped of the letters (see screenshot).
These I can't reproduce, sorry.
I guess either the demo isn't enough to reproduce this, or it's a problem specific to your hardware (I use a X4500HD).
https://bugs.freedesktop.org/show_bug.cgi?id=28000
Martin Peres martin.peres@free.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |INVALID Status|NEW |RESOLVED
--- Comment #12 from Martin Peres martin.peres@free.fr --- Hi,
Freedesktop's Bugzilla instance is EOLed and open bugs are about to be migrated to http://gitlab.freedesktop.org.
To avoid migrating out of date bugs, I am now closing all the bugs that did not see any activity in the past year. If the issue is still happening, please create a new bug in the relevant project at https://gitlab.freedesktop.org/drm (use misc by default).
Sorry about the noise!
dri-devel@lists.freedesktop.org