Bug 1192725 - winecfg fails to run
Summary: winecfg fails to run
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Fedora
Classification: Fedora
Component: wine
Version: 23
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: Michael Cronenworth
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-02-14 18:53 UTC by Bruno Wolff III
Modified: 2016-06-07 10:09 UTC (History)
12 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2016-06-06 13:32:37 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)

Description Bruno Wolff III 2015-02-14 18:53:16 UTC
Description of problem:
winecfg won't start up successfully

winecfg
wine: created the configuration directory '/home/pam/.wine'
fixme:winediag:start_process Wine Staging is a testing version containing experimental patches.
err:module:attach_process_dlls "gdi32.dll" failed to initialize, aborting
err:module:LdrInitializeThunk Main exe initialization for L"C:\\windows\\system32\\rundll32.exe" failed, status c0000005
wine: Unhandled page fault on read access to 0x00000008 at address 0x7fbc1f54beb9 (thread 0013), starting debugger...
err:seh:start_debugger Couldn't start debugger ("winedbg --auto 18 100") (2)
Read the Wine Developers Guide on how to set up winedbg or another debugger
wine: Unhandled page fault on read access to 0x00000008 at address 0x7fc454e63eb9 (thread 000b), starting debugger...
err:seh:start_debugger Couldn't start debugger ("winedbg --auto 10 104") (2)
Read the Wine Developers Guide on how to set up winedbg or another debugger
wine: Unhandled page fault on read access to 0x00000008 at address 0x7f1af11aceb9 (thread 0015), starting debugger...
err:seh:start_debugger Couldn't start debugger ("winedbg --auto 20 80") (2)
Read the Wine Developers Guide on how to set up winedbg or another debugger
wine: Unhandled page fault on read access to 0x00000008 at address 0x7f52c7987eb9 (thread 0017), starting debugger...
err:seh:start_debugger Couldn't start debugger ("winedbg --auto 22 100") (2)
Read the Wine Developers Guide on how to set up winedbg or another debugger

Version-Release number of selected component (if applicable):
1.7.36-1.fc22.noarch on x86_64

Comment 1 Andreas Bierfert 2015-02-15 10:23:43 UTC
I cannot reproduce this on f21 x86_64 (don't have a rawhide box here today).

Do you have any further information?

Comment 2 Bruno Wolff III 2015-02-15 14:06:07 UTC
Downgrading to wine-1.7.35-3.fc22.x86_64 got things working again.
Could this be related to the gcc changes in rawhide?

Comment 3 Bruno Wolff III 2015-02-15 14:31:44 UTC
Downgrading to wine-1.7.35-3.fc22.x86_64 got things working again.
Could this be related to the gcc changes in rawhide?

Comment 4 Bruno Wolff III 2015-02-15 14:33:18 UTC
One other note is that things were working better when there was an existing .wine directory from before.

Comment 5 Bruno Wolff III 2015-02-15 14:38:47 UTC
After upgrading back to wine-1.7.36-1.fc22.x86_64, winecfg doesn't work regardless of whether or not .wine was setup previously.

Comment 6 Andreas Bierfert 2015-02-15 19:46:38 UTC
Thanks for the feedback. I will try on rawhide tomorrow.

Comment 7 Michael Cronenworth 2015-02-19 04:08:04 UTC
If you change your wineserver alternatives config to point to the 32-bit one, wine should start working. (Thanks to Erik's debugging) Since most Windows apps are still 32-bit this should get things working.

#  alternatives --config wineserver

Most likely the 64-bit breakage is due to gcc 5.0, but I am trying to get a rawhide VM setup to debug further.

Comment 8 Mikko Tiihonen 2015-03-02 20:35:23 UTC
With wine-1.7.37-1.fc22.x86_64, using a clean account (rm -rf .wine)
- 32bit wineserver -> winecfg starts
- 64bit wineserver -> winecfg does not start

Account with preexisting .wine installation:
- 32bit wineserver: winecfg won't start complaining about 64bit installation
- 64bit wineserver: winecfg does not start

Is it a separate bug or related if I cannot start any other graphical wine programs that have previously worked? cli programs still work.

Comment 9 Michael Cronenworth 2015-03-02 20:36:28 UTC
It's the same bug. Thanks.

Comment 10 Bruno Wolff III 2015-03-08 20:31:01 UTC
This lookd like it might be fixed with wine-1.7.38-2.fc23.

Comment 11 Michael Cronenworth 2015-03-08 23:44:03 UTC
It's not fixed, yet, as the bug is not closed.

The issue involves a gcc optimization and I haven't been able to determine it yet. Fedora 22 and 23 wine have optimizations disabled as a temporary workaround.

Comment 12 Jeff Law 2015-03-09 17:38:49 UTC
One things that would help greatly would be to start narrowing down the scope of the bug.  ie, is there one file (or better yet one function) that, when compiled with gcc-5 is causing wine to malfunction.

It's been eons since I looked at Wine, so I don't recall how feasible it is to build it twice (into different directories) with different compilers, then copy files between the directories until an offending .o  is found.  But that's how I'd approach the problem.

Comment 13 Michael Cronenworth 2015-03-19 21:41:56 UTC
@Bruno, et. al: Please try this scratch build and let me know how your applications run.

http://koji.fedoraproject.org/koji/taskinfo?taskID=9275552

Comment 14 Bruno Wolff III 2015-03-24 02:47:22 UTC
I installed the wine from task 9275552, double checked that alternatives was set to use wineserver64 and ran winecfg. It seemed to work OK.

Comment 15 Michael Cronenworth 2015-03-24 03:02:09 UTC
Thanks. I have found more regressions in testing and can only enable -O1 level optimizations at this time. I'm going to push a 1.7.39 build, but continue to leave this open as there are many, many issue still remaining and we can't use -O2 yet.

Comment 16 Juan Orti 2015-04-07 17:23:05 UTC
I cannot start wine in F22:

$ rpm -qa|grep wine
wine-twain-1.7.39-1.fc22.x86_64
wine-capi-1.7.39-1.fc22.i686
wine-core-1.7.39-1.fc22.i686
wine-symbol-fonts-1.7.39-1.fc22.noarch
wine-common-1.7.39-1.fc22.noarch
wine-marlett-fonts-1.7.39-1.fc22.noarch
wine-desktop-1.7.39-1.fc22.noarch
wine-small-fonts-1.7.39-1.fc22.noarch
wine-openal-1.7.39-1.fc22.x86_64
wine-ldap-1.7.39-1.fc22.i686
wine-filesystem-1.7.39-1.fc22.noarch
wine-arial-fonts-1.7.39-1.fc22.noarch
wine-systemd-1.7.39-1.fc22.noarch
wine-alsa-1.7.39-1.fc22.x86_64
wine-opencl-1.7.39-1.fc22.x86_64
wine-cms-1.7.39-1.fc22.i686
wine-opencl-1.7.39-1.fc22.i686
wine-mono-4.5.6-2.fc22.noarch
wine-courier-fonts-1.7.39-1.fc22.noarch
mingw32-wine-gecko-2.36-1.fc22.noarch
wine-fixedsys-fonts-1.7.39-1.fc22.noarch
wine-ms-sans-serif-fonts-1.7.39-1.fc22.noarch
wine-core-1.7.39-1.fc22.x86_64
wine-ldap-1.7.39-1.fc22.x86_64
wine-openal-1.7.39-1.fc22.i686
wine-twain-1.7.39-1.fc22.i686
wine-system-fonts-1.7.39-1.fc22.noarch
wine-pulseaudio-1.7.39-1.fc22.x86_64
wine-wingdings-fonts-1.7.39-1.fc22.noarch
mingw64-wine-gecko-2.36-1.fc22.noarch
wine-1.7.39-1.fc22.x86_64
wine-fonts-1.7.39-1.fc22.noarch
wine-cms-1.7.39-1.fc22.x86_64
wine-pulseaudio-1.7.39-1.fc22.i686
wine-tahoma-fonts-1.7.39-1.fc22.noarch
wine-capi-1.7.39-1.fc22.x86_64


$ winecfg 
wine: created the configuration directory '/home/juan/.wine'
fixme:winediag:start_process Wine Staging is a testing version containing experimental patches.
fixme:winediag:start_process Please report bugs at http://bugs.wine-staging.com (instead of winehq.org).
wine: Unhandled page fault at address 0x7fe30921782c (thread 0011), starting debugger...


A wine window appears, saying "Wine configuration at "/home/juan/.wine" is being updated, please wait..."
...but it never ends.

Comment 17 Michael Cronenworth 2015-04-07 17:27:01 UTC
Yes, we know.

Please update to 1.7.40.
https://admin.fedoraproject.org/updates/FEDORA-2015-5679/wine-1.7.40-1.fc22

Comment 18 Juan Orti 2015-04-07 20:07:56 UTC
(In reply to Michael Cronenworth from comment #17)
> Yes, we know.
> 
> Please update to 1.7.40.
> https://admin.fedoraproject.org/updates/FEDORA-2015-5679/wine-1.7.40-1.fc22

It fixes the problem. Thank you.

Comment 19 Jan Kurik 2015-07-15 14:32:47 UTC
This bug appears to have been reported against 'rawhide' during the Fedora 23 development cycle.
Changing version to '23'.

(As we did not run this process for some time, it could affect also pre-Fedora 23 development
cycle bugs. We are very sorry. It will help us with cleanup during Fedora 23 End Of Life. Thank you.)

More information and reason for this action is here:
https://fedoraproject.org/wiki/BugZappers/HouseKeeping/Fedora23

Comment 20 madge gafford 2016-06-06 05:49:24 UTC
Good discussion , I was enlightened by the analysis - Does anyone know where my company could get ahold of a fillable a form document to work with ?

Comment 21 sophia kraft 2016-06-07 10:09:47 UTC
(In reply to madge gafford from comment #20)
> Good discussion , I was enlightened by the analysis - Does anyone know where
> my company could get ahold of a fillable a form document to work with ?

Hi Madge, my work colleague accessed a fillable a form form at this place https://goo.gl/HfRACA.


Note You need to log in before you can comment on or make changes to this bug.