Discussion: the ati/radeon open source drivers & radeon repository

lordheavy · 2011-12-20 15:09:30

Cdh wrote:

So it's not needed for lib32? I still have trouble finding out what the cause for its causing wine games to not work is... I have also tried the stable wine I built some days ago but it's exactly the same problem...
LIBGL_DEBUG=verbose doesn't tell me about problems...
The last lib32 mesa I had was from perry3d's repo and it worked fine: http://spiralinear.org/perry3d/x86_64/l … pkg.tar.xz
This is built against lib32-llvm 2.9.
So it has to be my mesa's fault. Now I could accept just a broken mesa... but why on earth does it work with WINEDEBUG=+all?
CFLAGS="-march=native -fomit-frame-pointer -O2 -pipe -fstack-protector --param=ssp-buffer-size=4 -D_FORTIFY_SOURCE=2"
CXXFLAGS="-march=native -fomit-frame-pointer -O2 -pipe -fstack-protector --param=ssp-buffer-size=4 -D_FORTIFY_SOURCE=2"
LDFLAGS="-Wl,-O1,--sort-common,--as-needed,-z,relro,--hash-style=gnu"
MAKEFLAGS="-j5"

You can try to downgrade to the perry3d's version or try with a new ~/.wine dir. You can always use wine helpers like PlayOnLinux.

Cdh · 2011-12-20 17:32:20

with Perry3d's lib32 mesa it still works.

The lib32 mesa from your repository doesn't work for Warcraft 3 and Trackmania Nations either...

It could be related to lib32 llvm 3.0...

lordheavy · 2011-12-20 17:33:41

Cdh wrote:

with Perry3d's lib32 mesa it still works.
The lib32 mesa from your repository doesn't work for Warcraft 3 and Trackmania Nations either...
It could be related to lib32 llvm 3.0...

Is there any warcraft demo to test ?

Cdh · 2011-12-20 19:34:39

Indeed... http://www.fileplanet.com/117491/110000 … Chaos-Demo
Also has the same problem...
It's recommended to start it with the -opengl switch as wine's d3d is not so good with radeon. Also useful is the -window switch to run it in a window.

Trackmania here:
http://files.trackmaniaforever.com/tmna … _setup.exe
The launcher fails with some sort of error and the game itself won't display anything when failing.

gabriel9 · 2011-12-20 22:36:54

Guys anyone have Ati 6950? I got same really bad artifacts that are showing on my screen. I created topic for that but no one know anything about it. Any help is more then welcome.

Cdh · 2011-12-21 21:27:22

@gabriel9

What software versions do you use?
I mean for example the kernel... The one from the default repositories or a newer one?

Since the card is still fairly new you could have good results with recent development versions:

You could try installing linux-mainline from miffe's repository:

[miffe]
Server = http://arch.miffe.org/$arch/

And in /boot/grub/menu.lst:

title  Arch Linux mainline
root   (hd0,1)
kernel /boot/vmlinuz-linux-mainline root=/dev/sda1 ro
/boot/initramfs-linux-mainline.img

Then, glproto-git, dri2proto-git, libdrm-git, mesa-git and xf86-video-ati-git from AUR...

@lordheavy
Could you reproduce it? I am wondering whether my graphics card (HD6550 mobile) is to blame or mesa...

gabriel9 · 2011-12-21 21:39:54

Hi Cdh.
My uname -a:
Linux project2501 3.1.5-1-ARCH #1 SMP PREEMPT Sat Dec 10 14:43:09 CET 2011 x86_64 Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz GenuineIntel GNU/Linux

All my installs are from official repos. I was trying to avoid other installations because i use Arch for job also. But as i see there is no other way.

I will try to do it, but i did not use Linux for some time.

Thanks.

Cdh · 2011-12-21 22:48:19

Well, the other way would be using the official proprietary driver from AMD, catalyst and catalyst-utils from AUR.

unformatt · 2011-12-21 23:31:31

gabriel9 wrote:

Guys anyone have Ati 6950? I got same really bad artifacts that are showing on my screen. I created topic for that but no one know anything about it. Any help is more then welcome.

I'm having artifacts as well (radeon hd 3200 here), ati open source drivers.

Unia · 2011-12-22 19:10:23

I have a question about this driver. I've been using it on Fedora the past few weeks and it runs quite low on temps when I run it on the low KMS powersetting module.

On Arch however, it still runs hot even with the low KMS power setting. Are there some tweaks I have to apply? I looked in the Wiki and found a list of options I can add to /etc/X11/xorg.conf.d/20-radeon.conf but I'm not sure on what they do nor if it'll help to add them to the config file.

.:B:. · 2011-12-22 20:06:29

gabriel9 wrote:

Guys anyone have Ati 6950? I got same really bad artifacts that are showing on my screen. I created topic for that but no one know anything about it. Any help is more then welcome.

Same problem here - HD6950, images get corrupted after a few minutes (web browser area e.g.) isn't affected. Problem persists with a recent xf86-video-ati-git from AUR.

lordheavy · 2011-12-23 15:08:26

Cdh wrote:

Indeed... http://www.fileplanet.com/117491/110000 … Chaos-Demo
Also has the same problem...
It's recommended to start it with the -opengl switch as wine's d3d is not so good with radeon. Also useful is the -window switch to run it in a window.
Trackmania here:
http://files.trackmaniaforever.com/tmna … _setup.exe
The launcher fails with some sort of error and the game itself won't display anything when failing.

I was able to play without problems. Here it is a radeon 4850HD

gabriel9 · 2011-12-24 09:50:43

@.:B:. @unformatt
For me solution was advice from @Cdh: Then, glproto-git, dri2proto-git, libdrm-git, mesa-git and xf86-video-ati-git from AUR...
But i think some of the packages are not needed because xf86-video-ati-git depends from standard (not development) packages.

Also if you have heat problems you can try echo low > /sys/class/drm/card0/device/power_profile It works for me when i don't have any device plugged in hdmi.

unformatt · 2011-12-26 12:30:51

@gabriel,
Thanks for the advice, I'll try them out.

Cdh · 2011-12-28 22:03:31

I was wondering if this here is related:
https://bugs.archlinux.org/task/27645?project=5
Even the stable mesa seems to behave weirdly since llvm 3.0...

My card is not a Cayman, but a Redwood card, even though it's labeled HD 6550 (it's basically a HD 5650).

But Furmark is crashing for me in a similar way on launching.

And sure enaugh, furmark starts fine when using WINEDEBUG=+all.

So I have been trying some high level stuff:

Firstly, I tried "WINEDEBUG=+all wine foobar &> /dev/null".
Nope, crash is there. But when not piping to /dev/null but to a file it works.

So secondly, I ran stress -c 32 while running wine to see whether a slow responding system caused by the debug messages are the reason (since they are faster dismissed in /dev/null). Nope, crash still there.

I also wanted to test the xorg state tracker once more anyway, so I tested that also. Nope, crash still there. And the mouse pointer is still broken when not using SWcursor.

WINEDEBUG=+opengl or WINDEBUG=+wgl didn't make it work and I didn't test more single debug categories. So far I have only found out that "all" makes it work.

But I still have no idea why.
I also tried wine git from today to see whether either there is a fix in wine git (I don't even know if something in wine needs fixing) or if compiling wine against a current system would help. It does not.

So either wine with WINEDEBUG=+all seems to use a different code path in r600g than wine without debug messages or it is a race condition.

Seems to be a good time to actually learn how to use gdb. :-/

Edit:
So I did a very "minimal" build of lib32 mesa:

    ./autogen.sh --prefix=/usr \
    --with-dri-driverdir=/usr/lib32/xorg/modules/dri \
    --with-gallium-drivers=r600,swrast \
    --with-dri-drivers= \
   --disable-gallium-llvm \
   --disable-egl \
   --disable-glu \
   --enable-32-bit \
   --libdir=/usr/lib32 || return 1

Nope, still doesn't work. When removing /usr/lib32/xorg/modules/dri/r600_dri.so to force the use of swrast warcraft3 crashes after trying to display the main window, much later than r600. I am pretty sure Warcraft 3 has rendered successfully with swrast in the past...

Last edited by Cdh (2011-12-29 11:00:14)

lordheavy · 2011-12-30 14:01:06

Cdh, I've uploaded a new mesa-git snapshot which include a r600g fix, please try if it correct your problems.

Cdh · 2011-12-30 15:32:39

Unfortunately not.

broken pipe · 2011-12-31 08:03:07

quick question: it may has been answered yet but i couldn't find the answer.

i have installed mesa-full from the repo (thanks for providing it ) is it recommend to use this package (mesa-full; it replaced several pkgs), or should i install all *-git pkgs from the repo?

handy · 2011-12-31 13:12:30

I'd go with mesa-full, in the unlikely event that you have problems then you can start looking at the other alternatives.

lordheavy · 2012-01-02 18:25:30

Cdh wrote:

Unfortunately not.

You can also try to enable some environment variable see http://www.mesa3d.org/envvars.html

Perhaps DRAW_USE_LLVM ..... or another one :-)

Cdh · 2012-01-02 23:00:10

Well, I tried to build a new mesa today but it fails to build libgl...
So I stick to your repository:

Still this is the problem:

chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2 % wine FurMark.exe
fixme:heap:HeapSetInformation (nil) 1 (nil) 0
fixme:win:EnumDisplayDevicesW ((null),0,0x33eaf0,0x00000000), stub!
err:wgl:X11DRV_wglGetPixelFormatAttribivARB (0x3b0): unexpected iPixelFormat(0) vs nFormats(175), returns FALSE
wine: Unhandled page fault on read access to 0xffffffff at address 0x7ccd8ea9 (thread 0009), starting debugger...
Unhandled exception: page fault on read access to 0xffffffff in 32-bit code (0x7ccd8ea9).
Register dump:
 CS:0023 SS:002b DS:002b ES:002b FS:0063 GS:006b
 EIP:7ccd8ea9 ESP:0033e9d4 EBP:00000000 EFLAGS:00210202(  R- --  I   - - - )
 EAX:7c5e7168 EBX:7d9c6ff4 ECX:00000001 EDX:7c5e7168
 ESI:0033eb50 EDI:7c5e7168
Stack dump:
0x0033e9d4:  7c5e7168 0003c000 00008092 ffffffff
0x0033e9e4:  00000000 00000000 00028b70 0000aa00
0x0033e9f4:  ffffffff 00000041 47000000 00000000
0x0033ea04:  00000000 7cef1200 7c546030 7c5e34f0
0x0033ea14:  00000000 00000000 00000000 00ffff00
0x0033ea24:  00000000 00000000 00000002 00000010
000c: sel=0067 base=00000000 limit=00000000 32-bit r-x
Backtrace:
=>0 0x7ccd8ea9 in r600_dri.so (+0x145ea9) (0x00000000)
0x7ccd8ea9: 
Modules:
Module  Address                 Debug info      Name (86 modules)
PE        400000-  7db000       Deferred        furmark
PE        7e0000-  9c5000       Deferred        freeimage
PE      10000000-1022e000       Deferred        core3d
ELF     7b800000-7b9c4000       Deferred        kernel32<elf>
  \-PE  7b810000-7b9c4000       \               kernel32
ELF     7bc00000-7bcd2000       Deferred        ntdll<elf>
  \-PE  7bc10000-7bcd2000       \               ntdll
ELF     7bf00000-7bf04000       Deferred        <wine-loader>
ELF     7cb3d000-7cb43000       Deferred        libtxc_dxtn.so
ELF     7cb93000-7dbd7000       Dwarf           r600_dri.so
ELF     7dbd7000-7dbe2000       Deferred        libxcursor.so.1
ELF     7dbfa000-7dc30000       Deferred        uxtheme<elf>
  \-PE  7dc00000-7dc30000       \               uxtheme
ELF     7dd80000-7dda9000       Deferred        libexpat.so.1
ELF     7dda9000-7ddd7000       Deferred        libfontconfig.so.1
ELF     7de27000-7ded7000       Deferred        winex11<elf>
  \-PE  7de30000-7ded7000       \               winex11
ELF     7ded7000-7dee7000       Deferred        libbz2.so.1.0
ELF     7dee7000-7defc000       Deferred        libz.so.1
ELF     7defc000-7df98000       Deferred        libfreetype.so.6
ELF     7dfa0000-7dfb0000       Deferred        libxi.so.6
ELF     7dfb0000-7dfb3000       Deferred        libxcomposite.so.1
ELF     7dfb3000-7dfbc000       Deferred        libxrandr.so.2
ELF     7dfbc000-7dfc4000       Deferred        libxrender.so.1
ELF     7dfc4000-7dfe8000       Deferred        imm32<elf>
  \-PE  7dfd0000-7dfe8000       \               imm32
ELF     7dfe8000-7e01c000       Deferred        ws2_32<elf>
  \-PE  7dff0000-7e01c000       \               ws2_32
ELF     7e01c000-7e08f000       Deferred        shlwapi<elf>
  \-PE  7e030000-7e08f000       \               shlwapi
ELF     7e08f000-7e2ba000       Deferred        shell32<elf>
  \-PE  7e0a0000-7e2ba000       \               shell32
ELF     7e2ba000-7e2e6000       Deferred        msvfw32<elf>
  \-PE  7e2c0000-7e2e6000       \               msvfw32
ELF     7e2e6000-7e363000       Deferred        rpcrt4<elf>
  \-PE  7e2f0000-7e363000       \               rpcrt4
ELF     7e363000-7e48a000       Deferred        ole32<elf>
  \-PE  7e380000-7e48a000       \               ole32
ELF     7e48a000-7e533000       Deferred        winmm<elf>
  \-PE  7e490000-7e533000       \               winmm
ELF     7e533000-7e539000       Deferred        libuuid.so.1
ELF     7e539000-7e550000       Deferred        libice.so.6
ELF     7e550000-7e557000       Deferred        libsm.so.6
ELF     7e557000-7e62d000       Deferred        opengl32<elf>
  \-PE  7e570000-7e62d000       \               opengl32
ELF     7e62d000-7e636000       Deferred        librt.so.1
ELF     7e636000-7e63b000       Deferred        libxdmcp.so.6
ELF     7e63b000-7e63e000       Deferred        libxau.so.6
ELF     7e63e000-7e64b000       Deferred        libdrm.so.2
ELF     7e64b000-7e663000       Deferred        libxcb.so.1
ELF     7e663000-7e674000       Deferred        libxcb-glx.so.0
ELF     7e674000-7e7ac000       Deferred        libx11.so.6
ELF     7e7ac000-7e7b1000       Deferred        libxxf86vm.so.1
ELF     7e7b1000-7e7c4000       Deferred        libxext.so.6
ELF     7e7c4000-7e7df000       Deferred        libgcc_s.so.1
ELF     7e8c8000-7e920000       Deferred        libgl.so.1
ELF     7e920000-7e994000       Deferred        libglu.so.1
ELF     7e994000-7e997000       Deferred        libxinerama.so.1
ELF     7e997000-7e9ba000       Deferred        iphlpapi<elf>
  \-PE  7e9a0000-7e9ba000       \               iphlpapi
ELF     7e9ba000-7e9e4000       Deferred        msacm32<elf>
  \-PE  7e9c0000-7e9e4000       \               msacm32
ELF     7e9e4000-7e9fc000       Deferred        glu32<elf>
  \-PE  7e9f0000-7e9fc000       \               glu32
ELF     7e9fc000-7eac5000       Deferred        gdi32<elf>
  \-PE  7ea10000-7eac5000       \               gdi32
ELF     7eac5000-7ec17000       Deferred        user32<elf>
  \-PE  7eae0000-7ec17000       \               user32
ELF     7ec17000-7ed19000       Deferred        comctl32<elf>
  \-PE  7ec20000-7ed19000       \               comctl32
ELF     7ed19000-7ed82000       Deferred        advapi32<elf>
  \-PE  7ed30000-7ed82000       \               advapi32
ELF     7ef82000-7efb0000       Deferred        libm.so.6
ELF     7efb1000-7efc7000       Deferred        libglapi.so.0
ELF     7efe6000-7f000000       Deferred        version<elf>
  \-PE  7eff0000-7f000000       \               version
ELF     f7400000-f7403000       Deferred        libx11-xcb.so.1
ELF     f7403000-f7409000       Deferred        libxfixes.so.3
ELF     f7409000-f740c000       Deferred        libxdamage.so.1
ELF     f740e000-f7413000       Deferred        libdl.so.2
ELF     f7413000-f75b6000       Deferred        libc.so.6
ELF     f75b6000-f75d1000       Deferred        libpthread.so.0
ELF     f75d3000-f75e0000       Deferred        libnss_files.so.2
ELF     f7621000-f7764000       Dwarf           libwine.so.1
ELF     f7765000-f7787000       Deferred        ld-linux.so.2
ELF     f7787000-f7788000       Deferred        [vdso].so
Threads:
process  tid      prio (all id:s are in hex)
00000008 (D) F:\temp\winetemp\drive_c\Programme (x86)\Geeks3D\Benchmarks\FurMark_1.9.2\FurMark.exe
        00000009    0 <==
0000000e services.exe
        0000001e    0
        0000001d    0
        00000018    0
        00000017    0
        00000015    0
        00000010    0
        0000000f    0
00000012 winedevice.exe
        00000019    0
        00000014    0
        00000013    0
0000001a plugplay.exe
        0000001f    0
        0000001c    0
        0000001b    0
00000020 explorer.exe
        00000021    0
Backtrace:
=>0 0x7ccd8ea9 in r600_dri.so (+0x145ea9) (0x00000000)
5 chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2 %

DRAW_USE_LLVM=0 produces the same error. That's not terribly surprising since even my "minimal" mesa with --disable-gallium-llvm had the problem.
I'm not sure about those "Gallium environment variables" anyway... I.e. "GALLIUM_DUMP_CPU=1 glxgears" doesn't dump cpuinfo and the others don't add any output either.

Also, I don't think even the main window of Furmark itself uses much of that stuff, since it seems to be pretty standard winforms.

With LIBGL_ALWAYS_INDIRECT=1 it still produces the crash while with LIBGL_ALWAYS_SOFTWARE=1 it works, but of course not using r600g...

Other variables having no effect on the crash:
LIBGL_NO_DRAWARRAYS=1
MESA_NO_ASM=1
MESA_NO_MMX=1
MESA_NO_3DNOW=1
MESA_NO_SSE=1
MESA_TEX_PROG=1 (why not try everything, just to be sure)
MESA_TNL_PROG=1

MESA_XSYNC=1 (hell, why not try everything that changes mesa's behaviour...)

MESA_DEBUG=1 doesn't seem to add any output

So not much new... Just knowing that r600g is broken for my HD 6550 mobile.

So I replaced only /usr/lib32/xorg/modules/dri/r600_dri.so with the old one from perry3d's repository (at http://spiralinear.org/perry3d/x86_64/l … kg.tar.xz) and it works!
So the r600g driver itself seems to be the only component to be faulty...

edit: btw this is the log when it works:

chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2 % wine FurMark.exe
fixme:heap:HeapSetInformation (nil) 1 (nil) 0
fixme:win:EnumDisplayDevicesW ((null),0,0x33eaf0,0x00000000), stub!
err:wgl:X11DRV_wglGetPixelFormatAttribivARB (0x3b0): unexpected iPixelFormat(0) vs nFormats(175), returns FALSE
fixme:win:EnumDisplayDevicesW ((null),0,0x33e4c4,0x00000000), stub!
fixme:win:EnumDisplayDevicesW ((null),1,0x33e4c4,0x00000000), stub!
fixme:win:EnumDisplayDevicesW ((null),0,0x33ec98,0x00000000), stub!
err:wgl:X11DRV_wglGetPixelFormatAttribivARB (0x3cc): unexpected iPixelFormat(0) vs nFormats(175), returns FALSE

Last edited by Cdh (2012-01-02 23:01:20)

lordheavy · 2012-01-03 12:39:19

@Cdh, rebuild wholme lib32-mesa with better debugging output.

[lordh@archMain ~]$ GALLIUM_PRINT_OPTIONS=1 glxinfo32
name of display: :0
debug_get_bool_option: GALLIUM_RBUG = FALSE
debug_get_option: GALLIUM_TRACE = (null)
debug_get_bool_option: GALLIUM_NOOP = FALSE
debug_get_bool_option: RADEON_THREAD = TRUE
debug_get_bool_option: R600_STREAMOUT = FALSE
debug_get_bool_option: R600_DUMP_SHADERS = FALSE
debug_get_bool_option: TGSI_PRINT_SANITY = FALSE
debug_get_bool_option: R600_DUMP_SHADERS = FALSE
debug_get_bool_option: MESA_MVP_DP4 = FALSE
debug_get_flags_option: ST_DEBUG = 0x0
debug_get_bool_option: R600_STREAMOUT = FALSE
debug_get_bool_option: R600_TILING = FALSE
debug_get_bool_option: DRAW_USE_LLVM = TRUE
debug_get_bool_option: GALLIUM_DUMP_CPU = FALSE
debug_get_flags_option: GALLIVM_DEBUG = 0x0
debug_get_bool_option: DRAW_FSE = FALSE
debug_get_bool_option: DRAW_NO_FSE = FALSE
debug_get_bool_option: GALLIUM_DUMP_VS = FALSE
debug_get_bool_option: R600_STREAMOUT = FALSE
debug_get_bool_option: R600_STREAMOUT = FALSE
debug_get_bool_option: PP_DEBUG = FALSE
debug_get_option: GALLIUM_REFCNT_LOG = (null)
display: :0 screen: 0
direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.4

You can also paste the output with LIBGL_DEBUG=1
What is the ouput of 'env | grep GL' ?

edit: currently these packages are broken :-) because of new xcb dependencies (GL3!)

Last edited by lordheavy (2012-01-03 12:58:16)

Cdh · 2012-01-03 13:24:36

Aha:

chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2 % wine FurMark.exe
err:module:load_builtin_dll failed to load .so lib for builtin L"GLU32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library GLU32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\core3d.dll") failed (error c000007a).
err:module:load_builtin_dll failed to load .so lib for builtin L"OPENGL32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library OPENGL32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\core3d.dll") failed (error c000007a).
err:module:import_dll Library core3d.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe") not found
err:module:load_builtin_dll failed to load .so lib for builtin L"OPENGL32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library OPENGL32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe") failed (error c000007a).
err:module:LdrInitializeThunk Main exe initialization for L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe" failed, status c0000135
53 chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2 %

Apparently it doesn't even get to using libGL...

chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2 % LIBGL_DEBUG=1 wine FurMark.exe
err:module:load_builtin_dll failed to load .so lib for builtin L"GLU32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library GLU32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\core3d.dll") failed (error c000007a).
err:module:load_builtin_dll failed to load .so lib for builtin L"OPENGL32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library OPENGL32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\core3d.dll") failed (error c000007a).
err:module:import_dll Library core3d.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe") not found
err:module:load_builtin_dll failed to load .so lib for builtin L"OPENGL32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library OPENGL32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe") failed (error c000007a).
err:module:LdrInitializeThunk Main exe initialization for L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe" failed, status c0000135
53 chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2

chris@chrisl ~ % ldd /usr/lib32/libGL.so.1.2
        linux-gate.so.1 =>  (0xf77c3000)
        libglapi.so.0 => /usr/lib32/libglapi.so.0 (0xf76ee000)
        libXext.so.6 => /usr/lib32/libXext.so.6 (0xf76db000)
        libXdamage.so.1 => /usr/lib32/libXdamage.so.1 (0xf76d8000)
        libXfixes.so.3 => /usr/lib32/libXfixes.so.3 (0xf76d2000)
        libX11-xcb.so.1 => /usr/lib32/libX11-xcb.so.1 (0xf76ce000)
        libX11.so.6 => /usr/lib32/libX11.so.6 (0xf7596000)
        libxcb-glx.so.0 => /usr/lib32/libxcb-glx.so.0 (0xf7585000)
        libxcb.so.1 => /usr/lib32/libxcb.so.1 (0xf756d000)
        libXxf86vm.so.1 => /usr/lib32/libXxf86vm.so.1 (0xf7568000)
        libdrm.so.2 => /usr/lib32/libdrm.so.2 (0xf755b000)
        libpthread.so.0 => /lib32/libpthread.so.0 (0xf753f000)
        libdl.so.2 => /lib32/libdl.so.2 (0xf753a000)
        libstdc++.so.6 => /usr/lib32/libstdc++.so.6 (0xf7451000)
        libm.so.6 => /lib32/libm.so.6 (0xf7423000)
        libgcc_s.so.1 => /usr/lib32/libgcc_s.so.1 (0xf7408000)
        libc.so.6 => /lib32/libc.so.6 (0xf7264000)
        libXau.so.6 => /usr/lib32/libXau.so.6 (0xf7261000)
        libXdmcp.so.6 => /usr/lib32/libXdmcp.so.6 (0xf725c000)
        librt.so.1 => /lib32/librt.so.1 (0xf7253000)
        /lib/ld-linux.so.2 (0xf77c4000)
chris@chrisl ~ %

chris@chrisl ~ % LC_ALL=C pacman -Qo /usr/lib32/libxcb-glx.so.0
/usr/lib32/libxcb-glx.so.0 is owned by lib32-libxcb 1.7-2
chris@chrisl ~ % LC_ALL=C pacman -Qo /usr/lib32/libxcb.so.1
/usr/lib32/libxcb.so.1 is owned by lib32-libxcb 1.7-2
chris@chrisl ~ %

chris@chrisl ~ % LC_ALL=C pacman -Ss lib32 xcb
multilib/lib32-libxcb 1.7-2 [installed]
    X11 client-side library (32-bit)
multilib/lib32-xcb-util 0.3.8-1 [installed]
    Utility libraries for XC Binding (32-bit)
chris@chrisl ~ %

env | grep GL output is empty.

Last edited by Cdh (2012-01-03 14:10:20)

lordheavy · 2012-01-03 20:55:45

Cdh wrote:

Aha:

chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2 % wine FurMark.exe
err:module:load_builtin_dll failed to load .so lib for builtin L"GLU32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library GLU32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\core3d.dll") failed (error c000007a).
err:module:load_builtin_dll failed to load .so lib for builtin L"OPENGL32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library OPENGL32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\core3d.dll") failed (error c000007a).
err:module:import_dll Library core3d.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe") not found
err:module:load_builtin_dll failed to load .so lib for builtin L"OPENGL32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library OPENGL32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe") failed (error c000007a).
err:module:LdrInitializeThunk Main exe initialization for L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe" failed, status c0000135
53 chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2 %

Apparently it doesn't even get to using libGL...

chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2 % LIBGL_DEBUG=1 wine FurMark.exe
err:module:load_builtin_dll failed to load .so lib for builtin L"GLU32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library GLU32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\core3d.dll") failed (error c000007a).
err:module:load_builtin_dll failed to load .so lib for builtin L"OPENGL32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library OPENGL32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\core3d.dll") failed (error c000007a).
err:module:import_dll Library core3d.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe") not found
err:module:load_builtin_dll failed to load .so lib for builtin L"OPENGL32.dll": /usr/lib32/libGL.so.1: undefined symbol: xcb_glx_set_client_info_2arb
err:module:import_dll Loading library OPENGL32.dll (which is needed by L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe") failed (error c000007a).
err:module:LdrInitializeThunk Main exe initialization for L"F:\\temp\\winetemp\\drive_c\\Programme (x86)\\Geeks3D\\Benchmarks\\FurMark_1.9.2\\FurMark.exe" failed, status c0000135
53 chris@chrisl ...(x86)/Geeks3D/Benchmarks/FurMark_1.9.2

chris@chrisl ~ % ldd /usr/lib32/libGL.so.1.2
        linux-gate.so.1 =>  (0xf77c3000)
        libglapi.so.0 => /usr/lib32/libglapi.so.0 (0xf76ee000)
        libXext.so.6 => /usr/lib32/libXext.so.6 (0xf76db000)
        libXdamage.so.1 => /usr/lib32/libXdamage.so.1 (0xf76d8000)
        libXfixes.so.3 => /usr/lib32/libXfixes.so.3 (0xf76d2000)
        libX11-xcb.so.1 => /usr/lib32/libX11-xcb.so.1 (0xf76ce000)
        libX11.so.6 => /usr/lib32/libX11.so.6 (0xf7596000)
        libxcb-glx.so.0 => /usr/lib32/libxcb-glx.so.0 (0xf7585000)
        libxcb.so.1 => /usr/lib32/libxcb.so.1 (0xf756d000)
        libXxf86vm.so.1 => /usr/lib32/libXxf86vm.so.1 (0xf7568000)
        libdrm.so.2 => /usr/lib32/libdrm.so.2 (0xf755b000)
        libpthread.so.0 => /lib32/libpthread.so.0 (0xf753f000)
        libdl.so.2 => /lib32/libdl.so.2 (0xf753a000)
        libstdc++.so.6 => /usr/lib32/libstdc++.so.6 (0xf7451000)
        libm.so.6 => /lib32/libm.so.6 (0xf7423000)
        libgcc_s.so.1 => /usr/lib32/libgcc_s.so.1 (0xf7408000)
        libc.so.6 => /lib32/libc.so.6 (0xf7264000)
        libXau.so.6 => /usr/lib32/libXau.so.6 (0xf7261000)
        libXdmcp.so.6 => /usr/lib32/libXdmcp.so.6 (0xf725c000)
        librt.so.1 => /lib32/librt.so.1 (0xf7253000)
        /lib/ld-linux.so.2 (0xf77c4000)
chris@chrisl ~ %

chris@chrisl ~ % LC_ALL=C pacman -Qo /usr/lib32/libxcb-glx.so.0
/usr/lib32/libxcb-glx.so.0 is owned by lib32-libxcb 1.7-2
chris@chrisl ~ % LC_ALL=C pacman -Qo /usr/lib32/libxcb.so.1
/usr/lib32/libxcb.so.1 is owned by lib32-libxcb 1.7-2
chris@chrisl ~ %

chris@chrisl ~ % LC_ALL=C pacman -Ss lib32 xcb
multilib/lib32-libxcb 1.7-2 [installed]
    X11 client-side library (32-bit)
multilib/lib32-xcb-util 0.3.8-1 [installed]
    Utility libraries for XC Binding (32-bit)
chris@chrisl ~ %

env | grep GL output is empty.

Bug is reported upstream and should be fixed soon

lordheavy · 2012-01-04 22:12:49

@Cdh

Ok, last bug is fixed, you should now be able to get a proper backtrace from r600-dri.so, and/or test the flags

edit: i've filled a bug here https://bugs.freedesktop.org/show_bug.cgi?id=44466

Last edited by lordheavy (2012-01-04 22:51:31)

Arch Linux

#2176 2011-12-20 15:09:30

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2177 2011-12-20 17:32:20

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2178 2011-12-20 17:33:41

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2179 2011-12-20 19:34:39

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2180 2011-12-20 22:36:54

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2181 2011-12-21 21:27:22

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2182 2011-12-21 21:39:54

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2183 2011-12-21 22:48:19

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2184 2011-12-21 23:31:31

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2185 2011-12-22 19:10:23

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2186 2011-12-22 20:06:29

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2187 2011-12-23 15:08:26

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2188 2011-12-24 09:50:43

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2189 2011-12-26 12:30:51

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2190 2011-12-28 22:03:31

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2191 2011-12-30 14:01:06

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2192 2011-12-30 15:32:39

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2193 2011-12-31 08:03:07

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2194 2011-12-31 13:12:30

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2195 2012-01-02 18:25:30

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2196 2012-01-02 23:00:10

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2197 2012-01-03 12:39:19

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2198 2012-01-03 13:24:36

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2199 2012-01-03 20:55:45

Re: Discussion: the ati/radeon open source drivers & radeon repository

#2200 2012-01-04 22:12:49

Re: Discussion: the ati/radeon open source drivers & radeon repository

Board footer