From owner-freebsd-hackers  Mon Feb  5 07:26:42 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id HAA02812
          for hackers-outgoing; Mon, 5 Feb 1996 07:26:42 -0800 (PST)
Received: from Root.COM (implode.Root.COM [198.145.90.17])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id HAA02804
          for <hackers@FreeBSD.ORG>; Mon, 5 Feb 1996 07:26:40 -0800 (PST)
Received: from localhost (localhost [127.0.0.1]) by Root.COM (8.6.12/8.6.5) with SMTP id HAA02821; Mon, 5 Feb 1996 07:22:50 -0800
Message-Id: <199602051522.HAA02821@Root.COM>
X-Authentication-Warning: implode.Root.COM: Host localhost didn't use HELO protocol
To: A JOSEPH KOSHY <koshy@india.hp.com>
cc: gjennejohn@frt.dec.com, hackers@FreeBSD.ORG
Subject: Re: Willows 
In-reply-to: Your message of "Mon, 05 Feb 1996 18:15:02 +0530."
             <199602051245.AA155974303@fakir.india.hp.com> 
From: David Greenman <davidg@Root.COM>
Reply-To: davidg@Root.COM
Date: Mon, 05 Feb 1996 07:22:49 -0800
Sender: owner-hackers@FreeBSD.ORG
Precedence: bulk

>On this topic; do we support the P5's 1MB page size?  This could reduce

   No, we don't support variable page sizes and we're not likely ever going to
support them due to architectural constraints in the VM system (and I'm not
interested in changing this because of performance considerations).

>by tlb stepping).  Even the accelarators which use sparse addressing
>of their registers (Mach64, P9000, P9100 etc) this could be a win.

   Perhaps. The general decrease in performance and overall munging of the VM
code to handle variable page sizes would negate any benefit, IMO.

-DG

David Greenman
Core Team/Principal Architect, The FreeBSD Project