Mailing-List: contact cygwin-developers-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-developers-subscribe AT sources DOT redhat DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin-developers/>
List-Post: <mailto:cygwin-developers AT sources DOT redhat DOT com>
List-Help: <mailto:cygwin-developers-help AT sources DOT redhat DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-developers-owner AT sources DOT redhat DOT com
Delivered-To: mailing list cygwin-developers AT sources DOT redhat DOT com
Date: Fri, 18 May 2001 22:45:35 -0400
From: Christopher Faylor <cgf AT redhat DOT com>
To: cygwin-developers AT cygwin DOT com
Subject: Re: pthread gotchas
Message-ID: <20010518224535.A26035@redhat.com>
Reply-To: cygwin-developers AT cygwin DOT com
Mail-Followup-To: cygwin-developers AT cygwin DOT com
References: <20010518221107 DOT A12195 AT redhat DOT com> <021a01c0e00b$e8d48dc0$0200a8c0 AT lifelesswks>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.11i
In-Reply-To: <021a01c0e00b$e8d48dc0$0200a8c0@lifelesswks>; from robert.collins@itdomain.com.au on Sat, May 19, 2001 at 12:32:10PM +1000

On Sat, May 19, 2001 at 12:32:10PM +1000, Robert Collins wrote:
>----- Original Message -----
>From: "Christopher Faylor" <cgf AT redhat DOT com>
>Subject: pthread gotchas
>
>>Just to be clear up front, the gotchas I'm talking about are my
>>gotchas.  There were apparently some aspects of the way that pthreads
>>work that I was not familiar with.
>
>Same here.  :]
>
>>I am looking into finally getting signals working in a multi-threaded
>>enviroment given Corinna's recent problems with cygrunsrv.  So, I wrote
>>a program which forked, started a separate thread, and then called
>>waitpid().  I wanted to see how a SIGINT with a signal handler would
>>work with pthreads.  I ran the program on linux.
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^
>
>**** MOST IMPORTANT ****
>I think delivering signals to threads is broken.  I cannot
>see where you use getthread2signal in signal.cc.  That would
>break waitpid the way you see it broken.
>**** ****

The whole point of my efforts is to get signals working with threads on
cygwin.  I know that they are broken.  The above is just the tip of the
iceberg.  The signal delivery mechanism doesn't even attempt to take
multiple threads into account.  I've tried to carefully make stuff
thread safe but I still have a ways ago to get the signal mechanism
to be able to send signals to any thread other than the main one.

This email was talking about linux.  I ran the program on linux first so
that I could get a feel for how it was supposed to work and was
surprised to find that linux broke my perceptions.

I have included my test program below.  The waitpid, which is invoked in
the secondary thread fails unless I move the fork into the same thread.

I'd still be interested in hearing if there is a good pthreads reference
book which talks about this kind of stuff.

cgf

#include <stdio.h>
#include <pthread.h>
#include <wait.h>

static int pid = 0;

static void *
tf (void *v)
{
  int status;
  printf ("in tf, waiting for %d\n", pid);
  printf ("%d = waitpid\n", status = waitpid (pid, NULL, 0));
  if (status)
    perror ("waitpid");
  puts ("returning");
  return NULL;
}

int
main (int argc, char **argv)
{
  pthread_t tid;
  pthread_attr_t att = {0};
  void *thread_return;

  setbuf (stdout, NULL);
  if ((pid = fork ()) == 0)
    {
      puts ("in subproc, sleeping");
      sleep (10);
      puts ("in subproc, exiting");
      exit (0);
    }

  printf ("in parent, starting thread, pid %d", pid);
  printf ("%d = pthread_create\n", pthread_create (&tid,  NULL, tf, NULL));
  printf ("%d = pthread_join\n", pthread_join (tid, &thread_return));
  puts ("exiting");
}