http://bugs.winehq.org/show_bug.cgi?id=18989
Summary: Trackmania Nations Forever does not run after upgrading to Wine 1.1.23 Product: Wine Version: 1.1.23 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: DIWWYWDYVHQR@spammotel.com
Created an attachment (id=21877) --> (http://bugs.winehq.org/attachment.cgi?id=21877) Console output while starting the game unsuccessfully
I'm running Kubuntu 9.04 and the current Wine development release from "deb http://wine.budgetdedicated.com/apt jaunty". Trackmania Nations Forever was running fine until I upgraded to 1.1.23. With 1.1.23, instead of the game starting in full-screen mode, a small black window pops up and stays there. The game does not start. After I downgraded back to 1.1.22, the game runs again.
I've attached the console output and the console output after setting WINEDEBUG=warn+all.
I also went through the Regression Testing procedure as described in the wiki. I identified the following patch as the cause for the regression:
0571ebcee3d0fdb524d8609ea0ec4703cdfad71d is first bad commit commit 0571ebcee3d0fdb524d8609ea0ec4703cdfad71d Author: Henri Verbeet hverbeet@codeweavers.com Date: Fri Jun 5 10:10:47 2009 +0200
wined3d: Add extension detection to the GL format template table.
:040000 040000 6432b0f02cca54c3725b64037b698cb113321fc9 4f138de8f20ca87cccabfb01621b5bbb3a9fd5a1 M dlls
Please tell me if you want me to provide any further information.
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #1 from Christian DIWWYWDYVHQR@spammotel.com 2009-06-18 16:47:21 --- Created an attachment (id=21878) --> (http://bugs.winehq.org/attachment.cgi?id=21878) Console output while starting the game unsuccessfully with WINEDEBUG=warn+all
http://bugs.winehq.org/show_bug.cgi?id=18989
Christian DIWWYWDYVHQR@spammotel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |DIWWYWDYVHQR@spammotel.com
http://bugs.winehq.org/show_bug.cgi?id=18989
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #21877|application/octet-stream |text/plain mime type| |
http://bugs.winehq.org/show_bug.cgi?id=18989
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #21878|application/octet-stream |text/plain mime type| |
http://bugs.winehq.org/show_bug.cgi?id=18989
Christian DIWWYWDYVHQR@spammotel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |hverbeet@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=18989
Christian DIWWYWDYVHQR@spammotel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |directx-d3d
http://bugs.winehq.org/show_bug.cgi?id=18989
Ken Sharp kennybobs@o2.co.uk changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #2 from Christian DIWWYWDYVHQR@spammotel.com 2009-06-23 07:29:54 --- I'm still affected by this bug after updating to Wine 1.1.24.
http://bugs.winehq.org/show_bug.cgi?id=18989
Laissus Yoann yoann.laissus@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |yoann.laissus@gmail.com
--- Comment #3 from Laissus Yoann yoann.laissus@gmail.com 2009-07-15 03:52:54 --- I don't have this problem with Trackmania. But I have already run a regression test because of a similar problem in wine 1.1.25. (solved with a patch in the git) : http://bugs.winehq.org/show_bug.cgi?id=19187
Can you describe your system and the settings of Direct3D? For me I have tested with a nvidia 6600gt et 9800gs and with all DirectDrawRenderer.
As I describe in this bug there are graphical bugs since wine 1.1.19 : http://bugs.winehq.org/show_bug.cgi?id=19306 So you should test with UseGLSL disabled.
Can you also test with wine 1.1.25 or the latest git?
http://bugs.winehq.org/show_bug.cgi?id=18989
Tobias Kienzler tobias.kienzler@gmx.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |tobias.kienzler@gmx.de
--- Comment #4 from Tobias Kienzler tobias.kienzler@gmx.de 2009-07-16 13:46:02 --- I experience the same regression, however not with trackmania but with all direct3d-apps i tried (e.g. world of goo). i'm new to bugzilla, is this a reason to post this as another bug?
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #5 from Tobias Kienzler tobias.kienzler@gmx.de 2009-07-16 14:11:50 --- simply reverting the patch using git show 0571ebcee3d0fdb524d8609ea0ec4703cdfad71d | patch -p1 -R circumvents this bug, but there was one reject:
patching file dlls/wined3d/utils.c Hunk #1 succeeded at 188 (offset 18 lines). Hunk #2 succeeded at 199 (offset 18 lines). Hunk #3 FAILED at 207. Hunk #4 succeeded at 646 (offset 176 lines). Hunk #5 succeeded at 667 (offset 176 lines). Hunk #6 succeeded at 751 (offset 176 lines). Hunk #7 succeeded at 772 (offset 176 lines). 1 out of 7 hunks FAILED -- saving rejects to file dlls/wined3d/utils.c.rej
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #6 from Tobias Kienzler tobias.kienzler@gmx.de 2009-07-16 14:13:05 --- sorry i forgot to mention, i have an ati radeon mobility 9200
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #7 from Laissus Yoann yoann.laissus@gmail.com 2009-07-16 16:14:08 --- For your bugs with all d3d apps, you must determine if it's the same patch which produces the bug. In this case you have to explain your problem here otherwise the bug that you will open would be marked as duplicated.
For the reject I don't understand this problem because it appears for me with all patch I want to revert. So I revert manually in the source of the problematic version of wine.
Have you tested with the 1.1.25 version or the latest git? Can you also write the games you have tested?
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #8 from Tobias Kienzler tobias.kienzler@gmx.de 2009-07-17 09:47:01 --- (In reply to comment #7)
i never reverted before (in fact up till now i used the ubuntu-beta-repo) so i didn't know whether that could cause trouble. now since reverting does solve the problem, is there a chance this will be officially done too? or is there any way i can use the repo and only replace the wined3d-part from git with reverted patch in a way as automated as possible?
i tested the latest version in git this tuesday. i did the regression test with world of goo and mafia. while mafia didn't output anything besides a fixme and had to be killed blindly (a init error dialog which would only show up in virtual desktop mode), world of goo complained depending on the wine-version about "no direct3d adapter found" (or similar) and segfaulted or did so without message. i didn't test any other apps since then i tried downdating to 1.1.21 (or now reverting said patch) which "fixed" the problem.
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #9 from Tobias Kienzler tobias.kienzler@gmx.de 2009-07-18 06:24:10 --- I tried again with both today's ubuntu and git 1.1.26, bug is still there and can still be fixed by reverting the patch
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #10 from Christian DIWWYWDYVHQR@spammotel.com 2009-07-22 05:29:29 --- I'm still affected by this bug after updating to Wine 1.1.26. Downgrading to 1.1.22 still allows me to run the game.
The computer I'm trying to run the game on is a equipped with a Geforce2 MX. I'm using the proprietary NVIDIA driver, OpenGL version 1.5.8 NVIDIA 96.43.10. Laissus, what exactly do you want to know about my Direct3D settings? Settings from the Game or settings from Wine?
I had no Direct3D key under HKEY_CURRENT_USER > Software > Wine in my Wine registry. Adding the key Direct3D and the String value called UseGLSL with the value disabled does not change anything. The game still runs with Wine 1.1.22, but does not run with Wine 1.1.26.
Like Tobias, I was able to revert the patch against the sources of Wine 1.1.26, with Hunk #3 failing. After compiling Wine 1.1.26 with the patch reverted (except for Hunk #3), the game is running fine.
As Tobias is affected by the same regression as I am, I reckon the status should not be UNCONFIRMED any more. Can somebody change it to NEW?
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #11 from Henri Verbeet hverbeet@gmail.com 2009-07-22 05:41:34 --- Which of the changes in that patch causes this bug?
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #12 from Christian DIWWYWDYVHQR@spammotel.com 2009-07-22 06:08:11 --- Created an attachment (id=22525) --> (http://bugs.winehq.org/attachment.cgi?id=22525) Workaround to make the game run (reverse patch)
I was able to run the game successfully after reverting only hunk #4 of the patch after checking out wine-1.1.26 using git.
I've attached the patch file I used. To revert the patch, I executed:
cat regression-hunk-4.patch | sudo patch -p1 -R
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #13 from Christian DIWWYWDYVHQR@spammotel.com 2009-07-22 08:48:07 --- I don't know if this is helpful at all, but I was able to narrow down the cause of the bug on my computer to a few D3D formats.
If I change the original line of code: if (!GL_SUPPORT(gl_formats_template[i].extension)) continue;
to: if (!GL_SUPPORT(gl_formats_template[i].extension) && ( gl_formats_template[i].fmt == WINED3DFMT_D16_LOCKABLE || gl_formats_template[i].fmt == WINED3DFMT_D24S8 || gl_formats_template[i].fmt == WINED3DFMT_D24X8 || gl_formats_template[i].fmt == WINED3DFMT_D16_UNORM)) continue;
the game still does not start.
However, if I change the line e.g. to: if (!GL_SUPPORT(gl_formats_template[i].extension) && gl_formats_template[i].fmt != WINED3DFMT_D16_LOCKABLE) continue;
the game runs fine again. Instead of WINED3DFMT_D16_LOCKABLE, I can also use WINED3DFMT_D24S8, WINED3DFMT_D24X8, or WINED3DFMT_D16_UNORM.
So the function init_format_texture_info seems to provide wrong data, in my case at least for the D3D formats WINED3DFMT_D16_LOCKABLE, WINED3DFMT_D24S8, WINED3DFMT_D24X8, and WINED3DFMT_D16_UNORM, if executes "continue" for these formats.
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #14 from Henri Verbeet hverbeet@gmail.com 2009-07-22 09:11:57 --- What does your glxinfo output look like?
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #15 from Christian DIWWYWDYVHQR@spammotel.com 2009-07-22 12:20:54 --- Created an attachment (id=22535) --> (http://bugs.winehq.org/attachment.cgi?id=22535) Output of my glxinfo
http://bugs.winehq.org/show_bug.cgi?id=18989
Christian DIWWYWDYVHQR@spammotel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #22535|application/octet-stream |text/plain mime type| |
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #16 from Tobias Kienzler tobias.kienzler@gmx.de 2009-07-22 14:04:39 --- Created an attachment (id=22541) --> (http://bugs.winehq.org/attachment.cgi?id=22541) glxinfo ati mobility radeon 9200
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #17 from Henri Verbeet hverbeet@gmail.com 2009-07-24 18:50:59 --- Created an attachment (id=22592) --> (http://bugs.winehq.org/attachment.cgi?id=22592) patch
Looks like those cards don't actually support ARB_depth_texture, which is required to use those formats for textures. That shouldn't prevent the game from using it as regular onscreen depthstencil, of course. Does the attached patch make it any better?
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #18 from Tobias Kienzler tobias.kienzler@gmx.de 2009-07-25 08:25:05 --- Created an attachment (id=22613) --> (http://bugs.winehq.org/attachment.cgi?id=22613) Mafia Log with patched wine
(using today's git-version) the patch works with mafia untill i quit, when an (again only via alt+tab visible) programm error appears
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #19 from Tobias Kienzler tobias.kienzler@gmx.de 2009-07-25 08:27:00 --- Created an attachment (id=22614) --> (http://bugs.winehq.org/attachment.cgi?id=22614) world of goo log with patch
...world of goo however still doesn't start
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #20 from Christian DIWWYWDYVHQR@spammotel.com 2009-07-26 11:59:34 --- (In reply to comment #17)
Does the attached patch make it any better?
Yes. On my computer, Trackmania starts with wine-1.1.26 and Henri's patch (attachment id=22592) applied. I can play the game and exit it without any problems.
http://bugs.winehq.org/show_bug.cgi?id=18989
Chris chrischmbrs5@googlemail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |chrischmbrs5@googlemail.com
--- Comment #21 from Chris chrischmbrs5@googlemail.com 2009-07-28 14:57:26 --- Same bug and same solution (downgrade wine to earlier version} with Star Wars Racer on Radeon 9200.
Arch Linux 2009.2 Radeon 9200se on open ati source driver.
http://bugs.winehq.org/show_bug.cgi?id=18989
Chris chrischmbrs5@googlemail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #22 from Chris chrischmbrs5@googlemail.com 2009-07-28 15:02:07 --- *** This bug has been confirmed by popular vote. ***
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #23 from Tobias Kienzler tobias.kienzler@gmx.de 2009-08-09 06:45:27 --- I updated to wine-1.1.27, now even with the patch reverted the same errors as I posted after testing Henri's patch. So probably the patch works but a new regression occured between 1.1.26 and 1.1.27 - I'll do the regression test again sometime and post the result...
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #24 from Tobias Kienzler tobias.kienzler@gmx.de 2009-08-09 08:39:01 --- 4ab7af65750bbf0fc9ce16b15ef57eb84a8ce977 is first bad commit commit 4ab7af65750bbf0fc9ce16b15ef57eb84a8ce977 Author: Henri Verbeet hverbeet@codeweavers.com Date: Fri Jul 24 10:44:13 2009 +0200
wined3d: Manage occlusion queries in the context.
:040000 040000 8a5adc7615c227579aa930bf71ef4ef13b7378c9 0584e071412e6c4f9a1bfdf479e8692d42da4bdf M dlls
seems to cause the trouble, is this a new bug then? reverting this and the old one (0571ebcee3d0fdb524d8609ea0ec4703cdfad71d) compiles with errors in device.c »IWineD3DDeviceImpl_CreateQuery« I compiled directly after regression testing and now have quite runable wine, but I guess this is not supposed to work this way...
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #24 from Tobias Kienzler tobias.kienzler@gmx.de 2009-08-09 08:39:01 --- 4ab7af65750bbf0fc9ce16b15ef57eb84a8ce977 is first bad commit commit 4ab7af65750bbf0fc9ce16b15ef57eb84a8ce977 Author: Henri Verbeet hverbeet@codeweavers.com Date: Fri Jul 24 10:44:13 2009 +0200
wined3d: Manage occlusion queries in the context.
:040000 040000 8a5adc7615c227579aa930bf71ef4ef13b7378c9 0584e071412e6c4f9a1bfdf479e8692d42da4bdf M dlls
seems to cause the trouble, is this a new bug then? reverting this and the old one (0571ebcee3d0fdb524d8609ea0ec4703cdfad71d) compiles with errors in device.c »IWineD3DDeviceImpl_CreateQuery« I compiled directly after regression testing and now have quite runable wine, but I guess this is not supposed to work this way...
--- Comment #25 from Tobias Kienzler tobias.kienzler@gmx.de 2009-08-24 01:28:05 --- I just tested wine-1.1.28 (ubuntu repo update) and now everything works fine - thanks a lot :-)
http://bugs.winehq.org/show_bug.cgi?id=18989
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED Summary|Trackmania Nations Forever |Trackmania Nations Forever |does not run after |does not run |upgrading to Wine 1.1.23 |
--- Comment #26 from Vitaliy Margolen vitaliy@kievinfo.com 2009-08-24 07:16:46 --- Reported fixed.
http://bugs.winehq.org/show_bug.cgi?id=18989
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #27 from Alexandre Julliard julliard@winehq.org 2009-09-02 14:27:15 --- Closing bugs fixed in 1.1.29.
http://bugs.winehq.org/show_bug.cgi?id=18989
--- Comment #28 from Christian DIWWYWDYVHQR@spammotel.com 2009-09-17 06:02:43 --- Trackmania also runs with wine v1.1.29 on my machine.
Thanks for fixing this bug.