From owner-freebsd-hackers Thu Nov 26 04:12:01 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id EAA24016 for freebsd-hackers-outgoing; Thu, 26 Nov 1998 04:12:01 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from ns1.yes.no (ns1.yes.no [195.204.136.10]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id EAA24001 for ; Thu, 26 Nov 1998 04:11:58 -0800 (PST) (envelope-from eivind@bitbox.follo.net) Received: from bitbox.follo.net (bitbox.follo.net [195.204.143.218]) by ns1.yes.no (8.9.1a/8.9.1) with ESMTP id NAA18300; Thu, 26 Nov 1998 13:11:47 +0100 (CET) Received: (from eivind@localhost) by bitbox.follo.net (8.8.8/8.8.6) id NAA03347; Thu, 26 Nov 1998 13:11:46 +0100 (MET) Message-ID: <19981126131146.Q17813@follo.net> Date: Thu, 26 Nov 1998 13:11:46 +0100 From: Eivind Eklund To: Peter Jeremy , hackers@FreeBSD.ORG Subject: Re: Kernel threads References: <98Nov26.072820est.40325@border.alcanet.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93.2i In-Reply-To: <98Nov26.072820est.40325@border.alcanet.com.au>; from Peter Jeremy on Thu, Nov 26, 1998 at 07:28:49AM +1100 Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Thu, Nov 26, 1998 at 07:28:49AM +1100, Peter Jeremy wrote: > On Tue, 24 Nov 1998 00:52:18 -0800 (PST), Matthew Dillon wrote: > [Redesign the entire KVM subsystem] > > I just don't know how feasible it all is... it would mean a huge amount of > > rewriting. > And I don't suppose you currently have the time to examine it in detail :-). > > > The disadvantage of this > > scheme is that it limits main memory to around 2GB on a 32 bit machine. > > Whilst I don't believe this is a serious problem at present, I can see > that it might be in a couple of years. Presently, the code is almost ready to handle Intels memory extension stuff - modifications should be needed in pmap.c only, if I understood John correctly (private conversation as a result of wanting to run with more than 4GB of memory...) I think giving up this could really limit us for heavy duty server use. With the RAM prices at the level they are today (and dropping), wanting to run more than 2GB is not that exotic. > > The benefits cascade very quickly.... if one is willing to give up > 2GB > > memory configurations on 32 bit cpus. > In which case it would seem worthwhile doing some more detailed > examination. In particular, how much effort would be involved and > whether the 2GB boundary can be avoided. > > Now 3.0 is nearly stable, we need a new bleeding edge for -current. This > sounds like a good start for FreeBSD 4.x :-). I don't think 4.0 should be the next branch. It is easy to bounce it to 4.0 if we want to, but not easy to decrease from 4.0 to 3.1, so let us start at 3.1 :-) Eivind. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message