From owner-freebsd-hackers  Thu Apr 22 13: 0:35 1999
Delivered-To: freebsd-hackers@freebsd.org
Received: from pcnet1.pcnet.com (pcnet1.pcnet.com [204.213.232.3])
	by hub.freebsd.org (Postfix) with ESMTP id 6993115189
	for <hackers@FreeBSD.ORG>; Thu, 22 Apr 1999 13:00:30 -0700 (PDT)
	(envelope-from eischen@vigrid.com)
Received: (from eischen@localhost)
	by pcnet1.pcnet.com (8.8.7/PCNet) id PAA24764;
	Thu, 22 Apr 1999 15:57:19 -0400 (EDT)
Date: Thu, 22 Apr 1999 15:57:19 -0400 (EDT)
From: Daniel Eischen <eischen@vigrid.com>
Message-Id: <199904221957.PAA24764@pcnet1.pcnet.com>
To: dick@tar.com, luoqi@watermarkgroup.com
Subject: Re: flock + kernel threads bug
Cc: hackers@FreeBSD.ORG, peter.jeremy@auss2.alcatel.com.au
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

> Maybe.  But, if you want POSIX compliant pthreads, you need to be able
> to send a signal to a thread (pthread_kill()), as well as to the process
> (kill()).  Also, you need to be able to set the priority of a thread
> (pthread_setschedparam()) as well a the priority of the process
> (setpriority() or sched_setscheduler(), etc).  The pthread functions
> take a pthread_t parameter to identify the threads.  If the threads
> are kernel threads, how do you communicate to the kernel which thread
> you're acting on if the threads are "anonymous" in userland? 

You still have thread IDs in userland, but you now add:

  _lwp_kill, _lwp_setschedparam, etc,

system calls to control the kernel threads.  Or maybe one big
_lwp_control...

Dan Eischen
eischen@vigrid.com


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message