From owner-freebsd-current  Fri Sep 13 06:40:41 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id GAA20033
          for current-outgoing; Fri, 13 Sep 1996 06:40:41 -0700 (PDT)
Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id GAA20028;
          Fri, 13 Sep 1996 06:40:37 -0700 (PDT)
Received: (from root@localhost) by dyson.iquest.net (8.7.5/8.6.9) id IAA02448; Fri, 13 Sep 1996 08:40:24 -0500 (EST)
From: "John S. Dyson" <toor@dyson.iquest.net>
Message-Id: <199609131340.IAA02448@dyson.iquest.net>
Subject: Re: VM/kernel problems?
To: mrcpu@cdsnet.net (Jaye Mathisen)
Date: Fri, 13 Sep 1996 08:40:24 -0500 (EST)
Cc: jhs@FreeBSD.org, sysseh@devetir.qld.gov.au, current@FreeBSD.org
In-Reply-To: <Pine.NEB.3.95.960912221646.9658r-100000@mail.cdsnet.net> from "Jaye Mathisen" at Sep 12, 96 10:18:54 pm
Reply-To: dyson@FreeBSD.org
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-current@FreeBSD.org
X-Loop: FreeBSD.org
Precedence: bulk

> 
> Hmmm, as of -current supped on the 11th and 12th, my box is working just
> yippee skippee after a make world.  
> 
> Although I'm not using John's cache color mumbo-jumbo voodoo
> evilness.
> 
Well, the bugs haven't been in there anyway...  The pmap code is probably
the most difficult code that I work on (for me.)  Also buffer bug was
due to an oversight in vfs_bio.  I don't think that the page-coloring
has caused any bugs yet.  Additionally, the page-coloring is kind-of
being forced on everyone, because the it is on by default (but running
in a more conservative mode: opt for 64K cache.)  If we never
see a performance hit, we might just enable it for a 256K cache, since
newer machines almost all have it, and it really doesn't hurt systems
with smaller caches very much.  You can run without it enabled, but
I don't think that code has be tested very well (sections of vm_page.c
have been re-done to support the coloring.)

(One thing that I do need to do is to move the page coloring config from
 vm_page.h to a machine dependent file somewhere.)

John