From owner-freebsd-current Sun Jul 26 05:16:58 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id FAA18427 for freebsd-current-outgoing; Sun, 26 Jul 1998 05:16:58 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from news1.gtn.com (news1.gtn.com [194.77.0.15]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id FAA18422 for ; Sun, 26 Jul 1998 05:16:55 -0700 (PDT) (envelope-from andreas@klemm.gtn.com) Received: (from uucp@localhost) by news1.gtn.com (8.8.6/8.8.6) with UUCP id OAA23710; Sun, 26 Jul 1998 14:15:07 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.8/8.8.8) id NAA27684; Sun, 26 Jul 1998 13:37:03 +0200 (CEST) (envelope-from andreas) Message-ID: <19980726133703.A27677@klemm.gtn.com> Date: Sun, 26 Jul 1998 13:37:03 +0200 From: Andreas Klemm To: dg@root.com Cc: Brian Feldman , freebsd-current@FreeBSD.ORG Subject: Re: New LINT options: what is VM coloring? References: <19980726112401.A5570@klemm.gtn.com> <199807261020.DAA07327@implode.root.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93.1i In-Reply-To: <199807261020.DAA07327@implode.root.com>; from David Greenman on Sun, Jul 26, 1998 at 03:20:23AM -0700 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT SMP Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Sun, Jul 26, 1998 at 03:20:23AM -0700, David Greenman wrote: > >On Sat, Jul 25, 1998 at 05:10:54PM -0700, David Greenman wrote: > >> Page coloring is the process of allocating pages that have physical > >> alignment that provides optimal utilization of the memory cache. This can > >> be a big win for direct-mapped caches (e.g. most Pentium L2 caches), but > >> loses effectiveness with set-associative caches (e.g. Pentium Pro, which > >> has a set size of 4). > > > >Do you think it would be better to turn PQ off using: > > > > options PQ_NOOPT # No coloring > > You mean in the Pentium Pro case? No, page coloring has basically no > overhead and can only help - even with associative caches. I think the > "PQ_NOOPT" option is only for performance comparison/testing. Ok, thanks ! -- Andreas Klemm http://www.FreeBSD.ORG/~andreas What gives you 90% more speed, for example, in kernel compilation ? http://www.FreeBSD.ORG/~fsmp/SMP/akgraph-a/graph1.html "NT = Not Today" (Maggie Biggs) ``powered by FreeBSD SMP'' To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message