On 2007/03/07 at 10:43:01 Michal Sabala <[EMAIL PROTECTED]> wrote > `gdb program PID_of_perl_process` > > 0xf7e75189 in pthread_setcanceltype () from /lib/tls/libc.so.6 > (gdb) backtrace > #0 0xf7e75189 in pthread_setcanceltype () from /lib/tls/libc.so.6 > #1 0x080c7d1e in Perl_sv_clear () > #2 0x080c7e75 in Perl_sv_free () > #3 0x080b41b6 in Perl_sighandler () > #4 <signal handler called>
The signal received was CHLD. The program was able to waitpid(-1,WNOHANG) on the child process successfully. I suspect it either hanged on the successive call to waitpid(-1, WNOHANG), which was looking for more child processes, or upon exiting from CHLD sig handler. > I'm investigating is forcing LinuxThreads instead of NPTL will bypass > the issue. Also, I've had a RHEL4 machine hang in futex. Can you please confirm that perl interpreter with single threaded code does not explicitly call pthreads? If so, should a bug be filed on libc6? Thanks, Michal -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
