GDB locks up -- Cannot find new threads: generic error

[David Lecomber <david at allinea dot com>]

Hi all

I've been using an Opteron - with a 32-bit app and 32-bit GDB taken from
GDB 6.3.50.20050310, and get the aforementioned error.  I'm hoping
someone can help..

I can reproduce this, easily, every time with a short pthreads program.
The problem does *not* occur with GNU gdb Red Hat Linux
(6.1post-1.20040607.52rh)

The session is typically:

(gdb) b main
Breakpoint 1 at 0x804ed18: file
main.cpp, line 10.
(gdb) run
Starting program: a.out
warning: linux_test_for_tracefork: unexpected result from waitpid
(28261,
status 0x117f)
[Thread debugging using libthread_db enabled]
Error while reading shared library symbols:
Cannot find new threads: generic error

at this point GDB does nothing and is unresponsive to any user input.  

The  system is:
kernel-2.4.21-27.EL
glibc-2.3.2-95.30


GDB's stack is:

(gdb) bt
#0 0x400bd098 in sigsuspend () from /lib/tls/libc.so.6
#1 0x080971e5 in child_wait ()
#2 0x0809359a in thread_db_init ()
#3 0x080e4695 in wait_for_inferior ()
#4 0x080e4425 in proceed ()
#5 0x0807b426 in find_default_create_inferior ()
#6 0x080e18cd in kill_if_already_running ()
#7 0x080e1a0b in kill_if_already_running ()
#8 0x080a9f1e in _initialize_cli_dump ()
#9 0x080ab8ea in cmd_func ()
#10 0x08080311 in execute_command ()
#11 0x080f0fa1 in async_disable_stdin ()
#12 0x080f13dd in async_disable_stdin ()
#13 0x081b4d25 in rl_callback_read_char ()
#14 0x080f095b in delete_timer ()
#15 0x080f0eae in stdin_event_handler ()
#16 0x080f0280 in delete_file_handler ()
#17 0x080efdc8 in standard_macro_lookup ()
#18 0x080efe10 in gdb_do_one_event ()
#19 0x080edd3b in catch_errors ()
#20 0x080efe33 in start_event_loop ()
#21 0x080785db in main ()

- and if it helps,  I turned on debug of  lin-lwp and saw:

Starting program: ./a.out
CW: waitpid 28606 received Trace/breakpoint trap (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Child exited (stopped)
CW: waitpid 28606 received Trace/breakpoint trap (stopped)
warning: linux_test_for_tracefork: unexpected result from waitpid
(28629,
status
0x117f)
CW: waitpid 28606 received Trace/breakpoint trap (stopped)
[Thread debugging using libthread_db enabled]
Error while reading shared library symbols:
Cannot find new threads: generic error
LLR: PTRACE_SINGLESTEP process 28606, 0 (resume event thread)
LLW: waitpid 28606 received Trace/breakpoint trap (stopped)


The example below, gcc -m32 ptest.c -lpthread, generates the issue.


#include <pthread.h>
#include <stdio.h>
#define NUM_THREADS 5

void *PrintHello(void *threadid)
{
printf("\n%d: Hello World!\n", threadid);
pthread_exit(NULL);
}

int main (int argc, char *argv[])
{
pthread_t threads[NUM_THREADS];
int rc, t;
for(t=0;t < NUM_THREADS;t++){
printf("Creating thread %d\n", t);
rc = pthread_create(&threads[t], NULL, PrintHello, (void *)t);
if (rc){
printf("ERROR; return code from pthread_create() is %d\n", rc);
exit(-1);
}
}
pthread_exit(NULL);
}

Thanks
David
-- 
David Lecomber <david@allinea.com>