Malloc bug?

[Joshua_Giles]

Hello All,

I'm using a 2.2.4 based (errata, -34 I believe) glibc that is distributed with Red Hat for x86.  I've got a problem with a multithreaded application I have created; The program stops exectution without terminating.  Running it through gdb reveals:

#0  0x4008cbb5 in __sigsuspend (set=0x4d9fe948)
    at ../sysdeps/unix/sysv/linux/sigsuspend.c:45
#1  0x400521d9 in __pthread_wait_for_restart_signal (self=0x4d9ffbe0)
    at pthread.c:969
#2  0x40053a2c in __pthread_lock (lock=0x4001649c, self=0x4d9ffbe0)
    at spinlock.c:149
#3  0x40050d46 in __pthread_mutex_lock (mutex=0x4001648c) at mutex.c:109
#4  0x40177e50 in _dl_open (file=0x4d9fed5e "/lib/security/pam_unix.so",
    mode=-2147483646, caller=0x4019d8ef) at dl-open.c:401
#5  0x401a3375 in dlopen_doit (a=0x4d9febf8) at dlopen.c:39
#6  0x4000d7c3 in _dl_catch_error (objname=0x8243048, errstring=0x824304c,
    operate=0x401a3340 <dlopen_doit>, args=0x4d9febf8) at dl-error.c:152
#7  0x401a36fb in _dlerror_run (operate=0x401a3340 <dlopen_doit>,
    args=0x4d9febf8) at dlerror.c:130
#8  0x401a3326 in __dlopen_check (file=0x4d9fed5e "/lib/security/pam_unix.so",
    mode=2) at dlopen.c:53
#9  0x4019d8ef in _pam_add_handler () from /lib/libpam.so.0
#10 0x4019d28a in _pam_parse_conf_file () from /lib/libpam.so.0
#11 0x4019d46c in _pam_init_handlers () from /lib/libpam.so.0
#12 0x4019c4ed in pam_start () from /lib/libpam.so.0
#13 0x40034466 in ?? ()
#14 0x400346e5 in ?? ()
#15 0x400347fe in ?? ()
#16 0x4002ae44 in ?? ()

I've noticed some similar problems on other posts in the glibc bugs archive, but I didn't see any resolution; The next place to look would be the GNATS bug database, but it appears to be down.

Read ya,

Josh