http://bugs.winehq.org/show_bug.cgi?id=20236
Summary: The game Sacred crashes Product: Wine Version: 1.1.30 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: daniel.skorka@stud.uni-karlsruhe.de
Created an attachment (id=23876) --> (http://bugs.winehq.org/attachment.cgi?id=23876) Console output from invocation to crash
The game proceeds normally, until the in game menu is used to a) start a new game or b) load an already saved game. The loading screen also displays the usual, but at the very moment loading is done, Sacred crashes, and wine displays an info message. Wine version 1.0.1 is known to work.
http://bugs.winehq.org/show_bug.cgi?id=20236
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #23876|application/octet-stream |text/plain mime type| |
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #1 from Jeff Zaroyko jeffz@jeffz.name 2009-10-02 05:11:47 --- Have you tested 1.0.1? If it works for you then you could probably run a regression test http://wiki.winehq.org/RegressionTesting
Also, how have you installed this game? It looks like you're running it from 'Sacred' in your home directory, rather than from an installed location in your Wine prefix.
http://bugs.winehq.org/show_bug.cgi?id=20236
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |wylda@volny.cz
--- Comment #2 from Wylda wylda@volny.cz 2009-10-02 07:28:18 --- Hi Daniel, to be sure...
a] What version: Sacred 1 or Sacred 2? b] Patches? c] Why don't you make the link between the game in AppDB and this bugreport?
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #3 from Daniel Skorka daniel.skorka@stud.uni-karlsruhe.de 2009-10-03 20:52:10 --- Yes, I have tested wine 1.0.1 and it works. Game was installed using the games installer, I just wanted to install it to this place. The game is Sacred 1, NOT Sacred Underworld, patch version 1.8.2.8, as far as I cant tell. I have linked to the version in AppDB I believe to be the right one. I'll see about the regression test, don't hold your breath.
http://bugs.winehq.org/show_bug.cgi?id=20236
Jeff Zaroyko jeffz@jeffz.name changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression Summary|The game Sacred crashes |Sacred 1 crashes when | |loading a game
http://bugs.winehq.org/show_bug.cgi?id=20236
A Wine user RandomAccountName@mail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |RandomAccountName@mail.com, | |stefan@codeweavers.com
--- Comment #4 from A Wine user RandomAccountName@mail.com 2010-05-12 11:11:17 --- I see the same behavior in the demo. Regression testing indicated:
228f2cf3c22124958a0f049a4463718a96267a7d is first bad commit commit 228f2cf3c22124958a0f049a4463718a96267a7d Author: Stefan Dösinger stefan@codeweavers.com Date: Fri Aug 1 11:47:24 2008 -0500
wined3d: Initialize the texture op function properly.
:040000 040000 ec0004b51f045c3ae13dc9a68f50a38a1a22ebe0 a17f112ec555fa2c076f5dc6dfc731aef5d8872b M dlls
My video hardware: Intel GMA 950, driver version 2.9.0
http://bugs.winehq.org/show_bug.cgi?id=20236
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download URL| |http://download.cnet.com/Sa | |cred-Demo/3000-7537_4-10264 | |156.html
--- Comment #5 from Austin English austinenglish@gmail.com 2010-05-12 12:05:46 --- Please post urls when there is a demo/download available.
http://bugs.winehq.org/show_bug.cgi?id=20236
Stefan Dösinger stefandoesinger@gmx.at changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |stefandoesinger@gmx.at
--- Comment #6 from Stefan Dösinger stefandoesinger@gmx.at 2010-05-14 12:24:23 --- I'm not quite sure how the patch found by the bisect would cause a NULL pointer exception in the game. Are you sure you found the right patch? Does reverting it manually(on the current tree if possible, or an older tree otherwise) fix the problem?
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #7 from A Wine user RandomAccountName@mail.com 2010-05-14 13:13:31 --- Hi Stefan,
Now that you mention it, my crash says "assertion failed..." instead of "unhandled page fault..." even in 1.1.30. Does that mean it's a different crash from the one originally reported here? The symptoms are identical otherwise.
(In reply to comment #5)
Please post urls when there is a demo/download available.
Sorry about that.
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #8 from A Wine user RandomAccountName@mail.com 2010-05-14 14:05:24 --- Looking at my log a little more carefully, this looks rather suspicious:
i915_fragprog.c:973: i915BindProgram: Assertion `p->on_hardware == 0' failed.
Am I correct in thinking that looks like a driver bug? Maybe it's just a coincidence that I see a crash at the same point due to a commit within the same time frame...
For the record, reverting that patch from 1.1.3 did prevent the crash for me. (Reverting from git failed, and I'm not sure how to apply the failed hunk manually to the modern code.)
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #9 from Stefan Dösinger stefandoesinger@gmx.at 2010-05-16 03:55:28 --- Ah, that makes more sense now. I guess the game has an exception handler that sees the failed assertion in the driver, tries to handle it, and fails. Thus the crash inside the game.
I suspect the driver is running out of fragment program constants. Can you make a +d3d_shader log? Maybe it is a genuine driver bug, but maybe we can optimize shaders a little bit.
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #10 from Wylda wylda@volny.cz 2010-05-16 04:16:03 ---
I'm not affected - works fine here.
Debian Lenny 32bit, nvidia 8600GT v195.36.24
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #11 from A Wine user RandomAccountName@mail.com 2010-05-16 07:56:08 --- Created an attachment (id=28011) --> (http://bugs.winehq.org/attachment.cgi?id=28011) +d3d_shader log
Can you make a +d3d_shader log?
Sure, no problem. (4MB decompressed)
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #12 from Stefan Dösinger stefandoesinger@gmx.at 2010-05-16 17:09:56 --- The shaders don't look complicated, they use only 2-3 constants each(24 must be supported, most cards support 256), and they use only 2D textures. I suspect you are tripping over a driver bug. Can you try to update your Mesa installation?
I suspect that before the patch that "broke" the game one of the shaders contained incorrect code, or referenced an incorrect texture type, and thus failed earlier and avoided to trigger the driver bug.
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #13 from A Wine user RandomAccountName@mail.com 2010-05-17 10:53:59 --- Well, I tested it using the Kubuntu 10.04 live CD (with Mesa 7.7.1 and Intel driver 2.9.1) and it still crashes the same way there. I also tried using the nightly build of Mesa from the Xorg-edgers PPA, but that didn't help either...
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #14 from Stefan Dösinger stefandoesinger@gmx.at 2010-05-17 13:51:40 --- Still the same exception? I guess in this case it would be helpful to find failing shader and ask the Mesa developers for their opinion. However, from the log I guess that this isn't an easy job - I think there are 5 different shaders(not many), but all of them are successfully used at least once, so I guess is is some other aspect of the GL config that causes the issue.
Maybe try to file a Mesa bug stating your exact configuration, the failing assertion, and a link to a game demo. It is likely that they'll say that they can't look at the Windows app and need some stand-alone testcase, in which case I can try to assist with more debugging(I don't have this card, so it will be tricky). However, maybe they know what's wrong from the assertion alone.
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #15 from A Wine user RandomAccountName@mail.com 2010-05-19 12:33:48 --- Yeah, it was still the same exception. Searching the FreeDesktop Bugzilla brought me to a very similar-looking bug report:
https://bugs.freedesktop.org/show_bug.cgi?id=22309
http://bugs.winehq.org/show_bug.cgi?id=20236
Xavier Vachon xvachon@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |xvachon@gmail.com
--- Comment #16 from Xavier Vachon xvachon@gmail.com 2010-07-10 22:20:44 --- In current git (1.2-rc7), I am able to load my saved games.
Using Sacred Gold, Gentoo Linux X64 and nvidia-drivers 256.35
Can someone else test and confirm?
http://bugs.winehq.org/show_bug.cgi?id=20236
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jeremielapuree@yahoo.fr
--- Comment #17 from joaopa jeremielapuree@yahoo.fr 2010-10-10 02:07:42 CDT --- For me it does not crash nvidia geforce 7600go driver 256.53
http://bugs.winehq.org/show_bug.cgi?id=20236
--- Comment #18 from joaopa jeremielapuree@yahoo.fr 2010-11-13 15:35:26 CST --- Surely a driver bug.
This bug could be closed as INVALID
http://bugs.winehq.org/show_bug.cgi?id=20236
Dmitry Timoshkov dmitry@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |INVALID
--- Comment #19 from Dmitry Timoshkov dmitry@codeweavers.com 2010-11-14 04:46:41 CST --- Marking as invalid.
http://bugs.winehq.org/show_bug.cgi?id=20236
Dmitry Timoshkov dmitry@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #20 from Dmitry Timoshkov dmitry@codeweavers.com 2010-11-14 04:46:56 CST --- Closing invalid.
-
wine-bugs＠winehq.org