Received: from PACIFIC-CARRIER-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA10248; Wed, 28 Feb 96 11:43:45 EST
Received: from mcs.anl.gov by MIT.EDU with SMTP
	id AA26539; Wed, 28 Feb 96 11:38:55 EST
Received: from dalek (dalek.mcs.anl.gov [140.221.1.2]) by antares.mcs.anl.gov (8.6.10/8.6.10)  with ESMTP
	id KAA29104; Wed, 28 Feb 1996 10:39:27 -0600
From: Jonathan Geisler <geisler@mcs.anl.gov>
Message-Id: <199602281639.KAA16759@dalek>
Subject: Re: pthreads and fork()
To: sam@inf.enst.fr
Date: Wed, 28 Feb 1996 10:39:23 -0600 (CST)
Cc: proven@MIT.EDU, pthreads@MIT.EDU
In-Reply-To: <qw6zqa3n8fg.fsf@gargantua.enst.fr> from "Samuel Tardieu" at Feb 28, 96 09:47:15 am
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 629       

> At least it helped us to point
> out the original design flaw in our code (according to Solaris 2.5
> documentation, fork() duplicates all the threads. Seems to be
> non-posix from your comments).

If you only want to fork 1 thread on Solaris, check out the fork1()
system call.  It should do what you want.

+--------------------------------------------------+
| Jonathan Geisler                                 |
| Mathematics & Computer Science Division          |
| Argonne National Laboratory                      |
| http://www.mcs.anl.gov/people/geisler/index.html |
+--------------------------------------------------+