This is the mail archive of the gdb@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [opensuse-programming] lib failure?


On Mon, Nov 27, 2006 at 03:57:43PM +0100, Oliver Block wrote:
> Hello,
> 
> I was facing a problem with a functio I wrote, checking if a file exists 
> withing the fs. When I was debuggin that , I set 2 breakpoints, the first to 
> the function file_exists and the second to the function call of file_exists 
> inside the program. I got the following output
> 
> **************************** gdb ************************************
> Breakpoint 2, main (argc=1, argv=0xbffff1c4) at ddnscli.c:712
> 712        if(file_exists(lockfile))
> (gdb) s
> 
> Breakpoint 1, file_exists (fname=0xbffff110 "disabled") at ddnscli.c:689
> 689        if( (rval = stat( fname, &attribut)) == -1)
> (gdb) s
> __stat (file=0xbffff070 "\200", buf=0xbffff070) at stat.c:51
> 51      stat.c: Datei oder Verzeichnis nicht gefunden.
>         in stat.c
> (gdb) bt full
> #0  __stat (file=0xbffff070 "\200", buf=0xbffff070) at stat.c:51
> No locals.
> #1  0x0804a56b in file_exists (fname=0xbffff110 "disabled") at ddnscli.c:689
>         attribut = {st_dev = 128, __pad1 = 34760, st_ino = 7, st_mode = 53,
>   st_nlink = 1075277812, st_uid = 1075283840, st_gid = 134522768,
>   st_rdev = 4615200421588234412, __pad2 = 34688, st_size = 40,
>   st_blksize = 1075277812, st_blocks = 1073835200, st_atime = 134522768,
>   st_atimensec = 3221221564, st_mtime = 1074560382, st_mtimensec = 40,
>   st_ctime = 1075277812, st_ctimensec = 3221221592, __unused4 = 1074559989,
>   __unused5 = 40}
>         rval = 1075277812
> *************************************************************************************
> 
> __stat (file=0xbffff070 "\200", buf=0xbffff070) at stat.c:51
> 
> both arguments of stat hold the same address. file should be AFAIK 
> fname=0xbffff110.
> 
> I isolated the function to a test_program and it works fine there.
> 
> Any ideas?

Showing us the actual source of the function would perhaps help.

Ciao, Marcus


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]