Re: main thread pthread_exit/sys_exit bug!

[Roland McGrath <roland@redhat.com>]

> Yes we have the bug, and wait_task_stopped() should be fixed. But it is
> buggy anyway, even if we delay the death of the main thread. But I also
> think we shouldn't.

Sorry, I'd missed the actual bug report among all the tangential verbiage.
I wrote this test case for it.  Is there any problem other than this one?


Thanks,
Roland

==========

#define _GNU_SOURCE
#include <stdio.h>
#include <stdlib.h>
#include <signal.h>
#include <pthread.h>
#include <assert.h>

static void *
thfunc (void *arg)
{
  sleep (2);
  puts ("stopping");
  raise (SIGSTOP);
  puts ("resumed");
  exit (0);
}

int
main (void)
{
  pthread_t th;
  int rc = pthread_create (&th, NULL, &thfunc, NULL);
  assert_perror (rc);
  pthread_exit (0);
  /*NOTREACHED*/
  return 1;
}