http://bugs.winehq.org/show_bug.cgi?id=15259
Summary: err:seh:setup_exception_record stack overflow in LineageII Product: Wine Version: 1.1.4 Platform: Other OS/Version: other Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: bazurbat@gmail.com
Created an attachment (id=16049) --> (http://bugs.winehq.org/attachment.cgi?id=16049) console output
The game LineageII Interlude crashes with:
err:seh:setup_exception_record stack overflow 356 bytes in thread 0009 eip 782489a0 esp 002311cc stack 0x230000-0x231000-0x330000
(numbers can vary) in some game locations. 100% reproducable. Also tried with every wine version from 1.0 to 1.1.4. Version 0.9.52 works fine.
With +relay wine crashes even before running the game. For details see attached wine_relay.log. Also attached log with +seh and console output.
Please tell if there anything I can do to help you debug the problem.
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #1 from Oleg Kolosov bazurbat@gmail.com 2008-09-12 15:11:51 --- Created an attachment (id=16051) --> (http://bugs.winehq.org/attachment.cgi?id=16051) output with +seh on
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #2 from Oleg Kolosov bazurbat@gmail.com 2008-09-12 15:19:25 --- See log with +relay on:
http://www.box.net/shared/5649pr55gb
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #3 from Austin English austinenglish@gmail.com 2008-09-12 15:52:22 --- Please run a regression test: http://wiki.winehq.org/RegressionTesting
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #4 from Oleg Kolosov bazurbat@gmail.com 2008-09-25 14:40:04 --- Please don't close this bug yet. I've found a couple of problems introduced somewhere after 0.9.54. There are lot to check and there are somewhat hard to pin exactly this bug. So, I'm still working on it.
http://bugs.winehq.org/show_bug.cgi?id=15259
Bartłomiej Zimoń uzi18@o2.pl changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |uzi18@o2.pl
--- Comment #5 from Bartłomiej Zimoń uzi18@o2.pl 2008-10-11 15:59:14 --- Have the same here:
$ wine Safari\ Kids.EXE +relay fixme:ntoskrnl:KeInitializeTimerEx 0x1110f0 0 err:module:import_dll Library VIDEOPRT.SYS (which is needed by L"C:\windows\system32\drivers\pcisys.sys") not found err:winedevice:ServiceMain driver L"PCISys" failed to load err:alsa:ALSA_CheckSetVolume Could not find 'PCM Playback Volume' element ALSA lib pcm_dmix.c:947:(snd_pcm_dmix_open) The dmix plugin supports only playback stream err:alsa:ALSA_CheckSetVolume Could not find 'PCM Playback Volume' element ALSA lib pcm_dmix.c:947:(snd_pcm_dmix_open) The dmix plugin supports only playback stream err:alsa:ALSA_CheckSetVolume Could not find 'PCM Playback Volume' element fixme:mixer:ALSA_MixerInit No master control found on SAA7134, disabling mixer fixme:bitblt:X11DRV_ClientSideDIBCopy potential optimization: client-side color-index mode DIB copy fixme:bitblt:X11DRV_ClientSideDIBCopy potential optimization: client-side color-index mode DIB copy fixme:bitblt:X11DRV_ClientSideDIBCopy potential optimization: client-side color-index mode DIB copy fixme:bitblt:X11DRV_ClientSideDIBCopy potential optimization: client-side color-index mode DIB copy fixme:bitblt:X11DRV_ClientSideDIBCopy potential optimization: client-side color-index mode DIB copy fixme:bitblt:X11DRV_ClientSideDIBCopy potential optimization: client-side color-index mode DIB copy err:seh:setup_exception_record stack overflow 540 bytes in thread 0009 eip 006dac0e esp 00231114 stack 0x230000-0x231000-0x330000
is there any workaround?
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #6 from Lei Zhang thestig@google.com 2008-10-15 19:03:25 --- The proper way to do a +relay log is: WINEDEBUG=+relay wine foo.exe.
Please do not paste logs, attach them instead. (as seen right above the text input box)
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #7 from Oleg Kolosov bazurbat@gmail.com 2008-10-22 11:55:26 --- With LineageII wine 0.9.52 works absolutely stable for me.
Starting from 0.9.53 there are "stack overflows". But it strongly depends on the location you're in, time of day, etc. Sometimes I can't even login, sometimes this happens after hours of play with different wine versions. I think this might be due to some combination of textures or effects, that's displayed on the screen simultaneously, which wine doesn't support or supports only partially. It's hard to do regression testing, as rebuilding takes time, after several tries bug may go away and manifest itself later, but several 'git bisect good' were done already (sorry for my English).
Running with WINEDEBUG=+relay gives me 'Segmentation Fault' right away.
What else I can try? Is there some way to increase exception stack size?
I've also tried with different values for nofile in limits.conf, this seems to have some effect but disabling limits completely doesn't help either.
Tried to install DirectX 9.0 with winetricks, tried DXWrapper as AppDB suggests, tried different Direct3D registry settings - no difference.
There is still dll substitution left to try, I've not tried everything I can yet.
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #8 from Austin English austinenglish@gmail.com 2008-10-22 12:34:40 --- (In reply to comment #7)
With LineageII wine 0.9.52 works absolutely stable for me.
Starting from 0.9.53 there are "stack overflows". But it strongly depends on the location you're in, time of day, etc. Sometimes I can't even login, sometimes this happens after hours of play with different wine versions. I think this might be due to some combination of textures or effects, that's displayed on the screen simultaneously, which wine doesn't support or supports only partially. It's hard to do regression testing, as rebuilding takes time, after several tries bug may go away and manifest itself later, but several 'git bisect good' were done already (sorry for my English).
The best you can do in this case is try to find a way to reproduce it reliably and test several times for each bisect.
Running with WINEDEBUG=+relay gives me 'Segmentation Fault' right away.
Has it always done this, or only aftyer .9.53? If only after, just use that each time.
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #9 from Austin English austinenglish@gmail.com 2009-04-26 20:52:34 --- Is this still an issue in current (1.1.20 or newer) wine?
http://bugs.winehq.org/show_bug.cgi?id=15259
Felix Saphir felix.saphir@presswatch.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |felix.saphir@presswatch.de
--- Comment #10 from Felix Saphir felix.saphir@presswatch.de 2009-04-27 06:02:43 --- Seems to be an issue in 1.1.19 and 1.1.20: When I start an application from my Delphi 5 IDE, Wine stops with this stack overflow and I have to kill the process manually. This does not happen with 1.1.18, so I thought about running a regression. I'll do this sometimes during the next days. Would another log of +relay or +seh help?
http://bugs.winehq.org/show_bug.cgi?id=15259
Felix Saphir felix.saphir@presswatch.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |julliard@winehq.org
--- Comment #11 from Felix Saphir felix.saphir@presswatch.de 2009-05-03 15:47:32 --- Regression testing identified this commit:
| 5316dd011f15ca488ccfa69b5873b075407fe238 is first bad commit | commit 5316dd011f15ca488ccfa69b5873b075407fe238 | Author: Alexandre Julliard julliard@winehq.org | Date: Wed Apr 8 19:38:02 2009 +0200 | | server: Define a generic context structure instead of using the | platform-specific version. | | :040000 040000 41719fe4a7ce6b86753b36303a5264c6c847fe47 366780690f2486debdb592a57615af7cf1436220 M dlls | :040000 040000 4025fa0fd81087892954b99232674cd2d6903c3b 10d95ef1a25200110f09c0c5309bd11e2a02682c M include | :040000 040000 fc00cc8a7f975bcd3ec13528b934f40a63c88462 c5afa6534d7e0e320655ec76b692d35cb014e1a6 M server
Is there anything else I could do to help?
HTH
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #12 from Austin English austinenglish@gmail.com 2009-05-03 18:50:07 --- (In reply to comment #11)
Regression testing identified this commit:
| 5316dd011f15ca488ccfa69b5873b075407fe238 is first bad commit | commit 5316dd011f15ca488ccfa69b5873b075407fe238 | Author: Alexandre Julliard julliard@winehq.org | Date: Wed Apr 8 19:38:02 2009 +0200 | | server: Define a generic context structure instead of using the | platform-specific version. | | :040000 040000 41719fe4a7ce6b86753b36303a5264c6c847fe47 366780690f2486debdb592a57615af7cf1436220 M dlls | :040000 040000 4025fa0fd81087892954b99232674cd2d6903c3b 10d95ef1a25200110f09c0c5309bd11e2a02682c M include | :040000 040000 fc00cc8a7f975bcd3ec13528b934f40a63c88462 c5afa6534d7e0e320655ec76b692d35cb014e1a6 M server
Is there anything else I could do to help?
HTH
Try reverting that patch and see if it works, e.g.,: $ git reset --hard 5316dd011f15ca488ccfa69b5873b075407fe238 test, should be bad. Then do: $ git show 5316dd011f15ca488ccfa69b5873b075407fe238 | patch -p1 -R test, should work.
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #13 from Felix Saphir felix.saphir@presswatch.de 2009-05-04 13:30:38 ---
Try reverting that patch and see if it works, e.g.,: $ git reset --hard 5316dd011f15ca488ccfa69b5873b075407fe238 test, should be bad. Then do:
Yes, crashes
$ git show 5316dd011f15ca488ccfa69b5873b075407fe238 | patch -p1 -R test, should work.
This one works fine.
Anything else I can do to help?
http://bugs.winehq.org/show_bug.cgi?id=15259
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #14 from Austin English austinenglish@gmail.com 2009-05-04 16:57:12 --- (In reply to comment #13)
Try reverting that patch and see if it works, e.g.,: $ git reset --hard 5316dd011f15ca488ccfa69b5873b075407fe238 test, should be bad. Then do:
Yes, crashes
$ git show 5316dd011f15ca488ccfa69b5873b075407fe238 | patch -p1 -R test, should work.
This one works fine.
Anything else I can do to help?
Short of investigating the patch and trying to fix it, it'd be best to wait until Alexandre says something/posts a patch.
Thanks for investigating!
http://bugs.winehq.org/show_bug.cgi?id=15259
onlyjob@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |onlyjob@gmail.com
--- Comment #15 from onlyjob@gmail.com 2009-05-13 19:52:47 --- Exactly the same problem with every version of wine > 1.0.1 on any portable application created using thinstall/thinapp v4+.
http://bugs.winehq.org/show_bug.cgi?id=15259
tbepdb tbepdb@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |tbepdb@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=15259
--- Comment #16 from Felix Saphir felix.saphir@presswatch.de 2009-08-07 18:56:23 --- (In reply to comment #10)
Seems to be an issue in 1.1.19 and 1.1.20: When I start an application from my Delphi 5 IDE, Wine stops with this stack overflow and I have to kill the process manually. This does not happen with 1.1.18, so I thought about running a regression. I'll do this sometimes during the next days. Would another log of +relay or +seh help?
Delphi 5 IDE does not crash with 1.1.27 anymore, fixed for me!
http://bugs.winehq.org/show_bug.cgi?id=15259
Dmitry Timoshkov dmitry@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #17 from Dmitry Timoshkov dmitry@codeweavers.com 2009-08-08 09:32:17 --- Reported fixed.
http://bugs.winehq.org/show_bug.cgi?id=15259
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #18 from Alexandre Julliard julliard@winehq.org 2009-08-21 12:36:52 --- Closing bugs fixed in 1.1.28.