Re: Segmentation fault after xdm login

[Harold L Hunt II <huntharo at msu dot edu>]

Cem,

I forgot to email you yesterday and tell you to try the 
Test103/XFree86-xserv-4.3.0-16 release.  I backported the final version 
of the WaitFor.c cleanups from XFree86 CVS HEAD to our 4.3 tree.  Your 
crash was happening in WaitFor.c, so it is possible that this newest 
version may randomly fix your problem.  Otherwise we will have to do 
some further debugging.

If you still get a crash, I have posted a debug version of 
Test103/XFree86-xserv-4.3.0-16, which you can run the same as before in gdb:

http://www.msu.edu/~huntharo/xwin/shadow/XWin-Test103-DEBUG.exe.bz2 
(3096 KiB)

Thanks for testing,

Harold

P.S. The information you sent last time was great.  I should only need 
the stackdump and backtrace this time if you have a crash.

Cem Han wrote:
> My first attempt at the xdm login screen causes Xwin(debug version or 
> not) instantly reappear with the same xdm login and the second attempt 
> results in a segmentation fault. The remote box is a Redhat 9. In the 
> xdm-errors file I can't see any messages at all even after increasing 
> the "DisplayManager.debugLevel" parameter located in 
> "/etc/X11/xdm/xdm-config".
> All the following output belongs to an "XWin-Test101-DEBUG.exe" session.
>
> Thanks a lot,
> Cem Han
>
> contents of  XWin-Test101-DEBUG.exe.stackdump:
>
> Exception: STATUS_ACCESS_VIOLATION at eip=61093CB1
> eax=00000000 ebx=00000020 ecx=109D49F0 edx=00000000 esi=109D4A10 
> edi=00000000
> ebp=0022FD98 esp=0022FD78 program=g:\apps\cygwin\home\Cem 
> Han\XWin-Test101-DEBUG.exe
> cs=001B ds=0023 es=0023 fs=003B gs=0000 ss=0023
> Stack trace:
> Frame     Function  Args
> 0022FD98  61093CB1  (61128390, 611283B4, 00000018, 10974148)
> 0022FDB8  61093BC0  (61128390, FFFFFFFF, 00000388, 109360C0)
> 0022FDE8  6103EB65  (10974148, 10109860, 0022FE18, 007E7C4D)
> 0022FDF8  0043DAC4  (10974148, 00000000, 0022FE38, 00000000)
> 0022FE18  007E7C4D  (10A04E98, 00000104, 0022FE38, 007E7EBD)
> 0022FE28  007E7CBB  (10900E8C, 00000000, 0022FE48, 007DB888)
> 0022FE38  007E7EBD  (10900E80, 00000000, 0022FE58, 004168AB)
> 0022FE48  007DB888  (10900A38, 00000000, 0022FE88, 004197F9)
> 0022FE58  004168AB  (10900A38, 00000004, 0022FE78, 0043DAC4)
> 0022FE88  004197F9  (1080E1D8, 00000006, 00000001, 0043DAC4)
> 0022FEA8  0041A3B2  (108083A0, 108120E8, 61600B64, 00000001)
> 0022FEF0  00401A1D  (00000003, 61600B64, 10100330, 0022FF24)
> 0022FF40  61005018  (610CFEE0, FFFFFFFE, 000007D0, 610CFE04)
> 0022FF90  610052ED  (00000000, 00000000, 00000001, 00000000)
> 0022FFB0  008652A1  (00401407, 037F0009, 0022FFF0, 77E814C7)
> 0022FFC0  0040103C  (00000001, 00000032, 7FFDF000, F4CE4CF0)
> End of stack trace (more stack frames may be present)
>
> **************************************************
> gdb backtrace from the recently posted XWin-Test101-DEBUG.exe:
>
> #0  0x61093af5 in strtosigno () from /usr/bin/cygwin1.dll
> #1  0x6103eb65 in free () from /usr/bin/cygwin1.dll
> #2  0x0043dac4 in Xfree (ptr=0x109d59d8) at utils.c:1317
> #3  0x004423ec in TimerInit () at WaitFor.c:528
> #4  0x0043ec12 in OsInit () at osinit.c:215
> #5  0x004014f9 in main (argc=3, argv=0x10101c78, envp=0x10100330)
>    at main.c:305