Received: (from majordom@localhost)
	by hyperreal.org (8.8.5/8.8.5) id UAA18687;
	Sun, 17 Aug 1997 20:59:21 -0700 (PDT)
Received: from valis.worldgate.com (marcs@valis.worldgate.com [198.161.84.2])
	by hyperreal.org (8.8.5/8.8.5) with ESMTP id UAA18645
	for <new-httpd@apache.org>; Sun, 17 Aug 1997 20:59:16 -0700 (PDT)
Received: from localhost (marcs@localhost)
	by valis.worldgate.com (8.8.5/8.8.5) with SMTP id VAA00720
	for <new-httpd@apache.org>; Sun, 17 Aug 1997 21:59:12 -0600 (MDT)
Date: Sun, 17 Aug 1997 21:59:12 -0600 (MDT)
From: Marc Slemko <marcs@worldgate.com>
To: new-httpd@apache.org
Subject: Re: SAFE_UNSERIALIZED_ACCEPT 
In-Reply-To: <Pine.LNX.3.95dg3.970817201133.6685D-100000@twinlark.arctic.org>
Message-ID: <Pine.BSF.3.95q.970817215628.13178a-100000@valis.worldgate.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: new-httpd-owner@apache.org
Precedence: bulk
Reply-To: new-httpd@apache.org

On Sun, 17 Aug 1997, Dean Gaudet wrote:

> Is it on by default?  I'm wondering about the interaction of select() and
> accept() on the same socket, same time, different pids.  That's why linux
> doesn't have it yet, nobody has done the thinkwork.

Yes, it is on by default.  I haven't really looked into the issue above;
it is possible that accept and select end up in different channels and the
wakeup_one only applies to the accept one, but that doesn't sound too
reasonable.

> 
> Dean
> 
> On Sun, 17 Aug 1997, Marc Slemko wrote:
> 
> > On Sun, 17 Aug 1997, Randy Terbush wrote:
> > 
> > > > I'd say try it on a real OS where all the children aren't woken up.  Like
> > > > recent FreeBSD.  <g>
> > > 
> > > How recent? Is that -current code?
> > 
> > 3.0 tree as of rev 1.23 on 1997/03/31, 2.2 tree as of rev 1.16.2.2 on
> > 1997/04/03.
> > 
> > Look for wakeup_one in sys/kern/uipc_socket2.c (in a function like
> > soisconnected or something like that).
> > 
> > 
>