http://bugs.winehq.org/show_bug.cgi?id=16311
Summary: World of Warcraft crashing on start with nVidia 180.x Product: Wine Version: 1.1.9 Platform: PC-x86-64 URL: http://www.worldofwarcraft.com OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: adys.wh+winehqdotorg@gmail.com
Using nVidia beta drivers 180.11 on 9800 GTX+.
WoW runs incredibly well (90 fps average instead of 15-20 fps average) under those new drivers. On first startup, nothing unusual happens. On exit, everything is fine. However, if I startup WoW a second time:
adys@azura:~/wow$ wine wow X Error of failed request: GLXBadDrawable Major opcode of failed request: 142 (GLX) Minor opcode of failed request: 5 (X_GLXMakeCurrent) Serial number of failed request: 147 Current serial number in output stream: 147 Locking assertion failure. Backtrace: #0 /usr/lib32/libxcb-xlib.so.0 [0xf7da1767] #1 /usr/lib32/libxcb-xlib.so.0(xcb_xlib_lock+0x2e) [0xf7da181e] #2 /usr/lib32/libX11.so.6 [0x7e9f6dc9] #3 /usr/lib32/libX11.so.6(XESetCloseDisplay+0x44) [0x7e9d8e14] #4 /usr/lib32/libGL.so.1 [0x7e960249] #5 [0x7dec0af8]
Restarting X fixes the problem until next startup. Numbers seem to always be the same.
Using latest git. I don't know if this is a bug with wine or nVidia; maybe related to bug 15926.
http://bugs.winehq.org/show_bug.cgi?id=16311
Adys adys.wh+winehqdotorg@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Severity|normal |major Summary|World of Warcraft crashing |3d applications crashing on |on start with nVidia 180.x |second start with nVidia | |180.x
--- Comment #1 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-03 00:10:01 --- After further testing, I cannot run any kind of 3d application under wine. However, I can run linux 3d applications without any kind of trouble.
Example with celestia:
adys@azura:~/.wine/drive_c/Program Files/Celestia$ WINEPREFIX="/home/adys/.wine/prefixes/new" wine celestia.exe Could not load Mozilla. HTML rendering will be disabled. X Error of failed request: GLXBadDrawable Major opcode of failed request: 142 (GLX) Minor opcode of failed request: 5 (X_GLXMakeCurrent) Serial number of failed request: 2882 Current serial number in output stream: 2882 Locking assertion failure. Backtrace: #0 /usr/lib32/libxcb-xlib.so.0 [0xf7dee767] #1 /usr/lib32/libxcb-xlib.so.0(xcb_xlib_lock+0x2e) [0xf7dee81e] #2 /usr/lib32/libX11.so.6 [0x7e86cdc9] #3 /usr/lib32/libX11.so.6(XESetCloseDisplay+0x44) [0x7e84ee14] #4 /usr/lib32/libGL.so.1 [0x7da31249] #5 [0x7df9c6e8]
Updating summary and changing severity to major.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #2 from Jeff Zaroyko jeffz@jeffz.name 2008-12-03 00:48:39 --- If the only thing which you have changed is your video driver, then it's unlikely to be a bug in Wine.
180.06 works fine for me with many 3d applications, I don't see the crash that you're getting.
http://bugs.winehq.org/show_bug.cgi?id=16311
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jeremielapuree@yahoo.fr
--- Comment #3 from joaopa jeremielapuree@yahoo.fr 2008-12-03 01:51:43 --- How do you know that it is not a driver bug? Your driver is still in beta stage. There is more chance that the bug is caused by your driver rather than by Wine. You should fill a report bug to Nvidia. With official Nvidia driver, I do not have such a problem.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #4 from joaopa jeremielapuree@yahoo.fr 2008-12-03 02:06:51 --- Anyway, I have just tested this driver. I launched 4 times the Hulk demo and 4 times the Blood2 demo. Everything is fine. It is probably your setup that is broken...
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #5 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-03 02:41:16 ---
How do you know that it is not a driver bug? Your driver is still in beta
stage.
As I said in the original report, I do not know whether it is a wine bug or a driver bug. The fact I'm able to run 3d native programs and only get crashes under wine leads me to believe it's a Wine bug.
I'm trying different setups to see what works and what does not. I run under x86_64, what about you?
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #6 from joaopa jeremielapuree@yahoo.fr 2008-12-03 02:55:27 --- Mine is x86.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #7 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-03 03:20:37 --- Just tested under 180.06, same thing.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #8 from Jeff Zaroyko jeffz@jeffz.name 2008-12-03 04:58:59 --- (In reply to comment #7)
Just tested under 180.06, same thing.
Maybe it's specific to your card? 180.06 with a geforce 8 works fine for me.
http://bugs.winehq.org/show_bug.cgi?id=16311
Adys adys.wh+winehqdotorg@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- URL|http://www.worldofwarcraft.c%7C |om | Summary|3d applications crashing on |3d apps crashing on second |second start with nVidia |start with nVidia |180.x |180.x/9800gtx
--- Comment #9 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-03 05:30:42 --- Perhaps, I couldn't find anyone else with this card. Updating title.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #10 from Jeff Zaroyko jeffz@jeffz.name 2008-12-03 05:44:24 --- (In reply to comment #9)
Perhaps, I couldn't find anyone else with this card. Updating title.
What happens if you go back to the 17x.xx drivers?
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #11 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-03 05:49:36 --- Created an attachment (id=17605) --> (http://bugs.winehq.org/attachment.cgi?id=17605) Trace snippet
I'm not exactly sure what logs to attach. Here is a relevant snippet of a +all trace (4.1MiB), just say if you need any more.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #12 from Vitaliy Margolen vitaliy@kievinfo.com 2008-12-03 09:45:10 --- 8800 GT with 180.11 works just fine here (~40 minutes long TF2 demo in dxlevel 95).
Please attach debug output as-is (no WINEDEBUG flags).
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #13 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-03 15:00:23 --- (In reply to comment #12)
8800 GT with 180.11 works just fine here (~40 minutes long TF2 demo in dxlevel 95).
Please attach debug output as-is (no WINEDEBUG flags).
There's no output before what I previously attached. The output on first start is completely normal.
adys@azura:~/wow$ wine wow X Error of failed request: GLXBadDrawable Major opcode of failed request: 142 (GLX) Minor opcode of failed request: 5 (X_GLXMakeCurrent) Serial number of failed request: 147 Current serial number in output stream: 147 Locking assertion failure. Backtrace: #0 /usr/lib32/libxcb-xlib.so.0 [0xf7da1767] #1 /usr/lib32/libxcb-xlib.so.0(xcb_xlib_lock+0x2e) [0xf7da181e] #2 /usr/lib32/libX11.so.6 [0x7e9f6dc9] #3 /usr/lib32/libX11.so.6(XESetCloseDisplay+0x44) [0x7e9d8e14] #4 /usr/lib32/libGL.so.1 [0x7e960249] #5 [0x7dec0af8]
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #14 from Vitaliy Margolen vitaliy@kievinfo.com 2008-12-04 08:29:39 --- Did you add yourself to "video" group?
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #15 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-04 14:54:37 --- (In reply to comment #14)
Did you add yourself to "video" group?
I already am, yup.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #16 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-08 04:04:33 --- (In reply to comment #15)
(In reply to comment #14)
Did you add yourself to "video" group?
I already am, yup.
Just a fyi, I run the mmap patch from bug 13335, but I tried without it as well.
Is there anything I can do to help?
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #17 from Jeff Zaroyko jeffz@jeffz.name 2008-12-08 04:48:45 --- (In reply to comment #16)
(In reply to comment #15)
(In reply to comment #14)
Did you add yourself to "video" group?
I already am, yup.
Just a fyi, I run the mmap patch from bug 13335, but I tried without it as well.
Is there anything I can do to help?
Still waiting for you to reply to comment #10.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #18 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-08 05:02:23 --- (In reply to comment #17)
Still waiting for you to reply to comment #10.
Sorry for that. All apps work fine under 177.x, however performances are noticably lower.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #19 from Jeff Zaroyko jeffz@jeffz.name 2008-12-08 05:10:28 --- (In reply to comment #18)
(In reply to comment #17)
Still waiting for you to reply to comment #10.
Sorry for that. All apps work fine under 177.x, however performances are noticably lower.
So you should probably report the bug to nvidia instead of here.
If Wine works fine for you on one driver revision but breaks on another then the issue is unlikely to be with Wine.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #20 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-08 22:54:47 --- (In reply to comment #19)
(In reply to comment #18)
(In reply to comment #17)
Still waiting for you to reply to comment #10.
Sorry for that. All apps work fine under 177.x, however performances are noticably lower.
So you should probably report the bug to nvidia instead of here.
If Wine works fine for you on one driver revision but breaks on another then the issue is unlikely to be with Wine.
I sent a bug report to nvidia. Note that in the same logic, they could also say "If it doesn't break with native apps but only with wine, the issue is unlikely to be with nVidia".
Again, if any help is required I'm here. I'd love seeing this fixed.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #21 from Jeff Zaroyko jeffz@jeffz.name 2008-12-08 23:19:10 --- (In reply to comment #20)
I sent a bug report to nvidia. Note that in the same logic, they could also say "If it doesn't break with native apps but only with wine, the issue is unlikely to be with nVidia".
Wine is a native application. It's an OpenGL client like any other.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #22 from Adys adys.wh+winehqdotorg@gmail.com 2008-12-19 07:19:08 --- For the record, this is also still (also) happening under Ubuntu 8.10 on the same hardware.
http://bugs.winehq.org/show_bug.cgi?id=16311
Nicholas LaRoche nlaroche@vt.edu changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |nlaroche@vt.edu
--- Comment #23 from Nicholas LaRoche nlaroche@vt.edu 2009-01-11 03:02:10 --- The same thing occurs for me using the official 180.22 drivers with EVE. This bug should probably be confirmed since it only occurs when using Wine and not when using native apps.
uname -r 2.6.27.9-159.fc10.x86_64
---- [nick@devbox EVE]$ DISPLAY=:0 wine eve.exe -windowed err:ole:CoGetClassObject class {9a5ea990-3034-4d6f-9128-01f3c61022bc} not registered err:ole:CoGetClassObject no class object {9a5ea990-3034-4d6f-9128-01f3c61022bc} could be created for context 0x1 fixme:heap:HeapSetInformation 0x8d0000 0 0x33fc44 4 fixme:ole:CoInitializeSecurity ((nil),-1,(nil),(nil),1,3,(nil),0,(nil)) - stub! X Error of failed request: GLXBadDrawable Major opcode of failed request: 142 (GLX) Minor opcode of failed request: 5 (X_GLXMakeCurrent) Serial number of failed request: 460 Current serial number in output stream: 460 Locking assertion failure. Backtrace: #0 /usr/lib/libxcb-xlib.so.0 [0x60f18767] #1 /usr/lib/libxcb-xlib.so.0(xcb_xlib_lock+0x2e) [0x60f1890e] #2 /usr/lib/libX11.so.6 [0x60e52079] #3 /usr/lib/libX11.so.6(XESetCloseDisplay+0x44) [0x60e33e74] #4 /usr/lib/libGL.so.1 [0x61497f69] #5 [0x7c9f1be8]
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #24 from Nicholas LaRoche nlaroche@vt.edu 2009-01-11 03:44:09 --- (In reply to comment #23)
The same thing occurs for me using the official 180.22 drivers with EVE. This bug should probably be confirmed since it only occurs when using Wine and not when using native apps.
uname -r 2.6.27.9-159.fc10.x86_64
[nick@devbox EVE]$ DISPLAY=:0 wine eve.exe -windowed err:ole:CoGetClassObject class {9a5ea990-3034-4d6f-9128-01f3c61022bc} not registered err:ole:CoGetClassObject no class object {9a5ea990-3034-4d6f-9128-01f3c61022bc} could be created for context 0x1 fixme:heap:HeapSetInformation 0x8d0000 0 0x33fc44 4 fixme:ole:CoInitializeSecurity ((nil),-1,(nil),(nil),1,3,(nil),0,(nil)) - stub! X Error of failed request: GLXBadDrawable Major opcode of failed request: 142 (GLX) Minor opcode of failed request: 5 (X_GLXMakeCurrent) Serial number of failed request: 460 Current serial number in output stream: 460 Locking assertion failure. Backtrace: #0 /usr/lib/libxcb-xlib.so.0 [0x60f18767] #1 /usr/lib/libxcb-xlib.so.0(xcb_xlib_lock+0x2e) [0x60f1890e] #2 /usr/lib/libX11.so.6 [0x60e52079] #3 /usr/lib/libX11.so.6(XESetCloseDisplay+0x44) [0x60e33e74] #4 /usr/lib/libGL.so.1 [0x61497f69] #5 [0x7c9f1be8]
I tested the NeHe Lesson 1 code and I get the same error as noted above. When using the linux version of this code I don't get the error.
This is looking more like a Wine issue with opengl32 or something along those lines.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #25 from Rico kgbricola@web.de 2009-01-11 06:26:52 --- Working fine here with driver 180.18 and 8800GTS on x86_64.
Could you try to remove all unneeded code from nehe lesson 1 so that only the part which breaks the hole thing is in.
You've always spoken only from 3D apps under wine. Could you run winecfg more than one time without the error?
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #26 from Nicholas LaRoche nlaroche@vt.edu 2009-01-11 07:21:51 --- (In reply to comment #25)
Working fine here with driver 180.18 and 8800GTS on x86_64.
Could you try to remove all unneeded code from nehe lesson 1 so that only the part which breaks the hole thing is in.
You've always spoken only from 3D apps under wine. Could you run winecfg more than one time without the error?
winecfg works without a problem either before or after this crash occurs. If you execute the following app compiled for windows with wine the same GLXBadDrawable error is printed out. (more concise than nehe #1)
[code] #include <windows.h> int main() { LoadLibrary( "opengl32.dll" ); return 0; } [/code]
The app crashes immediately after a call to pwglGetProcAddress inside the gdi32 library (in the function wglGetProcAddress). Given that native apps work I initially thought that it was a wine problem. But, given that the crash occurs only after wine is shutdown and restarted again It might be a combination of behaviour from both wine and the 180.x driver.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #27 from Adys adys.wh+winehqdotorg@gmail.com 2009-01-11 08:20:48 --- We had a talk on IRC recently. I sent a new email, but should have precised here too. This issue only happens with Xinerama enabled.
Still unsure if this is a bug in Wine aswell, though.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #28 from Nicholas LaRoche nlaroche@vt.edu 2009-01-11 08:40:26 --- (In reply to comment #27)
We had a talk on IRC recently. I sent a new email, but should have precised here too. This issue only happens with Xinerama enabled.
Still unsure if this is a bug in Wine aswell, though.
Forgot to mention that. I'm running 3x22" monitors across an 8800GTS and a 9600GT with Xinerama enabled.
The problem disappears when I turn Xinerama off. I can still run apps like xlock across all screens with no problem though with the same driver though without a crash.
http://bugs.winehq.org/show_bug.cgi?id=16311
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #29 from Vitaliy Margolen vitaliy@kievinfo.com 2009-01-11 12:38:10 --- Confirming since multiple people having the problem. Looks more and more like a driver bug to me.
http://bugs.winehq.org/show_bug.cgi?id=16311
Adys adys.wh+winehqdotorg@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Summary|3d apps crashing on second |3d apps crashing on second |start with nVidia |start on nVidia 180.x with |180.x/9800gtx |Xinerama enabled
--- Comment #30 from Adys adys.wh+winehqdotorg@gmail.com 2009-01-11 16:25:04 --- Updating title.
http://bugs.winehq.org/show_bug.cgi?id=16311
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |felix.kuperjans@gmx.de
--- Comment #31 from Vitaliy Margolen vitaliy@kievinfo.com 2009-02-09 20:29:17 --- *** Bug 17317 has been marked as a duplicate of this bug. ***
http://bugs.winehq.org/show_bug.cgi?id=16311
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Severity|major |enhancement
--- Comment #32 from Vitaliy Margolen vitaliy@kievinfo.com 2009-02-09 20:30:41 --- Obviously a Driver bug. Will keep it open to collect other bugs like this.
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #33 from Raymond Chace raychace@gmail.com 2009-02-09 21:55:05 --- I have the same problem while running Civilization: Warlords. I run the application once with no trouble, then after I close it the program will not open again unless I restart the X server.
2.6.27.12-170.2.5.fc10.x86_64 x86_64 GNU/Linux akmod-nvidia.x86_64 180.25-1.fc10 installed 2 monitors on 2 video cards (GeForce 7300GS/GeForce 7100 GS) I'm using Xinerama and no compiz or desktop effects.
X Error of failed request: GLXBadDrawable Major opcode of failed request: 128 (GLX) Minor opcode of failed request: 5 (X_GLXMakeCurrent) Serial number of failed request: 1066 Current serial number in output stream: 1066 Locking assertion failure. Backtrace: #0 /usr/lib/libxcb-xlib.so.0 [0x60e39767] #1 /usr/lib/libxcb-xlib.so.0(xcb_xlib_lock+0x2e) [0x60e3990e] #2 /usr/lib/libX11.so.6 [0x60d73079] #3 /usr/lib/libX11.so.6(XESetCloseDisplay+0x44) [0x60d54e74] #4 /usr/lib/nvidia/libGL.so.1 [0x6134bfa9] #5 [0x7d48c4e8]
http://bugs.winehq.org/show_bug.cgi?id=16311
--- Comment #34 from Felix Kuperjans felix.kuperjans@gmx.de 2009-02-10 09:47:15 --- I agree that this bug is caused by the NVIDIA driver, especially because no user of other drivers (e.g. ATI) reported this problem. I'm having the same problem with all Direct3D applications in wine (at the moment I tested 7 apps), but only if I enable xinerama (this bugs seems to occur only with xinerama, I had no problem using two really seperated Displays or a twin configuration). But I'm still wondering why this bug only occurs with wine and afaik not with any other OpenGL application (native UT2004 and OpenGL XEN-HVM worked fine on my system)... I think it could be possible, that there is probably some piece of unclean code in wine (and in the nvidia driver, too) that causes this bug, but that it could be solved by the nvidia-driver as well as by wine. It says "Locking assertion failure.", so probably there is some resource not cleaned up properly by wine, that causes some stale lock, which normally would be cleaned up by the driver on process termination, but whicht isn't in our case, due to a bug in the nvidia driver (or something like that). Probably it would be better if we do further investigation on this bug and try to solve it on the wine site, because I havn't seen any reaction of the NVIDIA team on http://www.nvnews.net/vbulletin/showthread.php?t=124730 and probably they won't solve this bug soon (and having to restart X or to go without xinerama is not a good situation...).
http://bugs.winehq.org/show_bug.cgi?id=16311
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|jeremielapuree@yahoo.fr |
http://bugs.winehq.org/show_bug.cgi?id=16311
Lukas Schneiderbauer lukas.schneiderbauer@aon.at changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |lukas.schneiderbauer@aon.at
http://bugs.winehq.org/show_bug.cgi?id=16311
Simetrical Simetrical+wine@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |Simetrical+wine@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=16311
Roderick Colenbrander thunderbird2k@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED CC| |thunderbird2k@gmail.com Resolution| |INVALID
--- Comment #35 from Roderick Colenbrander thunderbird2k@gmail.com 2009-05-28 15:10:06 --- This issue has been fixed in Nvidia driver 185.18.14. Aaron Plattner from Nvidia contacted me about it. Basically the issue was that the Nvidia drivers didn't like a glXMakeCurrent on the root_window in Xinerama (the call should pass). We do this during opengl init. Since the bug is not a Wine one, I'm marking it as invalid.
http://bugs.winehq.org/show_bug.cgi?id=16311
Adys adys.wh+winehqdotorg@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #36 from Adys adys.wh+winehqdotorg@gmail.com 2009-05-28 16:09:39 --- Closing.
http://bugs.winehq.org/show_bug.cgi?id=16311
Jerome Leclanche adys.wh@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|INVALID |UPSTREAM
--- Comment #37 from Jerome Leclanche adys.wh@gmail.com 2012-09-14 15:23:31 CDT --- Was fixed upstream.
-
wine-bugs＠winehq.org