Re: gdb 6.3 segfault at stack.c:294

[Vincent Rubiolo <vincent.rubiolo@windriver.com>]

Hello,

Something that the gdb maintaniers would welcome is a stack trace of the 
faulting gdb. To do that, rebuild gdb with debugging enabled (very easy, 
juste download, configure and make/make install) and when it crashes, 
issue the following at the other gdb prompt:
(gdb) bt

This will give more information on the faulty piece of code to the 
developers.

HTH,

Vincent

Joost VandeVondele wrote:
> Hi,
> 
> trying to debug a segfault in my program, gdb segfaults as well. I'm 
> wondering if this is a known issue. This is what I'm using:
> 
> GNU gdb 6.3
> configured as "i686-pc-linux-gnu"
> uname -a
> Linux vinegar 2.6.8-24.11-smp #1 SMP Fri Jan 14 13:01:26 UTC 2005 i686 
> i686 i386 GNU/Linux
> gcc -v
> Reading specs from /usr/lib/gcc-lib/i586-suse-linux/3.3.4/specs
> Configured with: ../configure --enable-threads=posix --prefix=/usr 
> --with-local-prefix=/usr/local --infodir=/usr/share/info 
> --mandir=/usr/share/man --enable-languages=c,c++,f77,objc,java,ada 
> --disable-checking --libdir=/usr/lib --enable-libgcj 
> --with-gxx-include-dir=/usr/include/g++ --with-slibdir=/lib 
> --with-system-zlib --enable-shared --enable-__cxa_atexit i586-suse-linux
> Thread model: posix
> gcc version 3.3.4 (pre 3.3.5 20040809)
> 
> The executable is a fortran based one, compiled using NAG's f95 (which 
> uses gcc to generate code AFAIK).
> 
> The only real info I have right now is what I get running gdb through 
> gdb on the segfaulting executable:
> 
> ------------------------------------------
> QS/regtest> gdb gdb
> GNU gdb 6.3
> Copyright 2004 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/tls/libthread_db.so.1".
> 
> (gdb) run ../../../exe/Linux-i686-nag/cp2k.sdbg
> Starting program: /home/vondele/bin/gdb 
> ../../../exe/Linux-i686-nag/cp2k.sdbg
> GNU gdb 6.3
> Copyright 2004 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/tls/libthread_db.so.1".
> 
> (gdb) c
> The program is not being run.
> (gdb) run
> Starting program: /data1/vondele/dev/cp2k/exe/Linux-i686-nag/cp2k.sdbg
> Detaching after fork from child process 31218.
> Detaching after fork from child process 31230.
>  @ has created process number                                        0 0
> 
>   **** **** ******  **  PROGRAM STARTED AT                   20050606 
> 141655.015
>  ***** ** ***  *** **   PROGRAM STARTED ON
>  **    ****   ******    PROGRAM STARTED BY
>  ***** **    ** ** **   PROGRAM PROCESS ID 0
>   **** **  *******  **  PROGRAM STARTED IN ./
> 
>  CP2K| Program compiled at                          Mon Jun  6 09:09:00 
> BST 2005
>  CP2K| Program compiled on vinegar
>  CP2K| Program compiled for Linux-i686-nag
>  CP2K| Last CVS entry            message_passing.F/1.50/Mon Jun  6 
> 06:44:18 2005
> 
>  ===== Routine Calling Stack =====
> 
>             1 CP2K
> 
> Program received signal SIGSEGV, Segmentation fault.
> 0x0917ad8d in input_section_types_MP_section_vals_add_values 
> (section_vals_=0x4088b138,
> 
> _input_section_types_section_vals_add_values_section_vals_map_=0x4088b13c, 
> error_=0xbfffa150,
> 
> Program received signal SIGSEGV, Segmentation fault.
> print_frame_args (func=Variable "func" is not available.
> ) at stack.c:294
> 294                   if (SYMBOL_CLASS (nsym) == LOC_REGISTER)
> (gdb) list
> 289                 {
> 290                   struct symbol *nsym;
> 291                   nsym = lookup_symbol
> 292                     (DEPRECATED_SYMBOL_NAME (sym),
> 293                      b, VAR_DOMAIN, (int *) NULL, (struct symtab **) 
> NULL);
> 294                   if (SYMBOL_CLASS (nsym) == LOC_REGISTER)
> 295                     {
> 296                       /* There is a LOC_ARG/LOC_REGISTER pair.  This 
> means that
> 297                          it was passed on the stack and loaded into 
> a register,
> 298                          or passed in a register and stored in a 
> stack slot.
> 
> ------------------------------------------
> 
> I can provide the executable that causes this, but it's quite large 
> (122Mb). I can also try to get some more info locally first, but I'm not 
> a very experienced gdb user.
> 
> Thanks in advance,
> 
> Joost

-- 
Using Firefox and Thunderbird at Wind River:
http://twiki.wrs.com/do/view/ENGtools/MozillaAtWindriver