From owner-freebsd-hackers  Thu Apr 22 16:30: 6 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 14D3514C22
	for <hackers@FreeBSD.ORG>; Thu, 22 Apr 1999 16:30:03 -0700 (PDT)
	(envelope-from eischen@vigrid.com)
Received: (from eischen@localhost)
	by pcnet1.pcnet.com (8.8.7/PCNet) id TAA21462;
	Thu, 22 Apr 1999 19:26:54 -0400 (EDT)
Date: Thu, 22 Apr 1999 19:26:54 -0400 (EDT)
From: Daniel Eischen <eischen@vigrid.com>
Message-Id: <199904222326.TAA21462@pcnet1.pcnet.com>
To: eischen@vigrid.com, newton@internode.com.au
Subject: Re: flock + kernel threads bug
Cc: dick@tar.com, hackers@FreeBSD.ORG, luoqi@watermarkgroup.com,
	peter.jeremy@auss2.alcatel.com.au
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

Mark Newton wrote:

>  > 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...
> 
> If you make the BSD API the same as the SysVR4 API it'll make emulating
> it really easy :-)
> 
> lwp_info
> lwp_sema_wait
> lwp_sema_post
> lwp_sema_trywait
> lwp_create
> lwp_exit
> lwp_suspend
> lwp_continue
> lwp_kill
> lwp_self
> lwp_getprivate
> lwp_setprivate
> lwp_wait
> lwp_mutex_unlock
> lwp_mutex_lock
> lwp_cond_wait
> lwp_cond_signal
> lwp_cond_broadcast
> lwp_sigredirect
> lwp_alarm

Seems like a good idea to me.  But are these all system calls, or are some
library routines?  Having lwp_self be a system call doesn't seem optimal.

Dan Eischen
eischen@vigrid.com


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