Executing wine over make segfaults.

[Marcus Meissner]

On Tue, Nov 07, 2006 at 06:44:31AM +0100, Pavel Troller wrote:
> > 
> > We just have to find out _why_ it breaks. Should not be too hard
> > with some backtraces or similar.
> > 
> Hi!
>   Today I tried to generate a backtrace. However, I simply failed.
>   I've recompiled wine with full debug, and reinstalled. Then I
> installed the newest gdb (6.5), because my older one (6.4) seemed
> a bit unstable. Then, I did the following:
> 1)
> patrol at tangens:/mnt/cd$ gdb wine
> GNU gdb 6.5
> Copyright (C) 2006 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you are
> welcome to change it and/or distribute copies of it under certain conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for details.
> This GDB was configured as "i686-pc-linux-gnu"...Using host libthread_db library "/lib/libthread_db.so.1".
> 
> (gdb) run 100_prazskych_zajimavosti.exe
> Starting program: /opt/wine/bin/wine 100_prazskych_zajimavosti.exe
> Failed to read a valid object file image from memory.
> [Thread debugging using libthread_db enabled]
> [New Thread -1208980992 (LWP 19254)]
> [New Thread -1208984688 (LWP 19257)]
> [Thread -1208984688 (LWP 19257) exited]
> Cannot find user-level thread for LWP 19254: generic error
> 
> gdb then hung and had to be killed.
> It should be noted that gdb works with standard Linux binaries normally.
> 
> 2)
> patrol at tangens:~$ wine /mnt/cd/100_prazskych_zajimavosti.exe
> fixme:font:WineEngCreateFontInstance just using first face for now
> fixme:font:WineEngCreateFontInstance just using first face for now
> fixme:font:WineEngCreateFontInstance just using first face for now
> fixme:font:WineEngCreateFontInstance just using first face for now
> fixme:font:WineEngCreateFontInstance just using first face for now
> fixme:font:WineEngCreateFontInstance just using first face for now
> fixme:ole:CoResumeClassObjects stub
> Segmentation fault (core dumped)
> patrol at tangens:~$ gdb -c core.19668 /mnt/cd/100_prazskych_zajimavosti.exe
> GNU gdb 6.5
> Copyright (C) 2006 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you are
> welcome to change it and/or distribute copies of it under certain conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for details.
> This GDB was configured as "i686-pc-linux-gnu"...(no debugging symbols found)
> Using host libthread_db library "/lib/libthread_db.so.1".
> 
> 
> warning: core file may not match specified executable file.
> 
> warning: shared library handler failed to enable breakpoint
> (no debugging symbols found)
> Core was generated by `/mnt/cd/100_prazskych_zajimavosti.exe                                         '.
> Program terminated with signal 11, Segmentation fault.
> #0  0x00876440 in ?? ()
> (gdb) bt
> #0  0x00876440 in ?? ()
> #1  0x0000000b in ?? ()
> #2  0x7ffddc8c in ?? ()
> #3  0x7ffddd0c in ?? ()
> #4  0x0000000b in ?? ()
> #5  0x00000000 in ?? ()
> (gdb)   
> 
> First I tried 'gdb -c <core> wine' but it told me that the core was generated
> by another executable, and stated the windows .exe file (you can see the same
> in the above output too). Backtrace is, however, the same in both cases, and,
> as you can see, totally bogus.
> 
> I really don't know how to get better results now.
> Any hints available ?

It is usuyally generated from wine-pthread

gdb wine-pthread core.19668
bt

If it still does not show a backtrace, try:

x /i $eip
info registers

to show the current instruction + registers.

Ciao, Marcus