From mboxrd@z Thu Jan 1 00:00:00 1970 From: tgh Subject: Re: question about the meaning of memory auto-translate and paravirtual and no pseudophysical overlay Date: Fri, 13 Apr 2007 08:34:48 +0800 Message-ID: <461ED028.8060906@ncic.ac.cn> References: <991B62EB36934C4EBD5B605518A724764C421E@pdsmsx404.ccr.corp.intel.com> <200704112228.45793.mark.williamson@cl.cam.ac.uk> <461D8240.20405@ncic.ac.cn> <200704121637.32931.mark.williamson@cl.cam.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=GB2312 Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <200704121637.32931.mark.williamson@cl.cam.ac.uk> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Mark Williamson Cc: xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org Thank you for your reply Thank you Mark Williamson =D0=B4=B5=C0: > Nb. I'm focussing on x86 (and x86_32 where appropriate) here and have b= een in=20 > the rest of the thread (unless otherwise specified)... other architect= ures=20 > deal with pseudophysical addressses differently. > > =20 >> Mark Williamson =D0=B4=B5=C0: >> =20 >>>> if there is no pseudophysical addresses,a physical host computer cou= ld >>>> only paravirtualise one VM,is it right? >>>> =20 >>> No, AFAIK pseudophysical addresses are mostly there for the convenien= ce >>> of the guest. Xen has some support for them so that guests can use t= hem >>> more efficiently but that's not strictly necessary. In principle, th= ey >>> could be eliminated from Xen entirely (would require modifying the PV >>> guests to manage the pseudophys abstraction themselves). >>> =20 >> Is there this kind paravirtual os for xen at present? >> or in future,will this kind paravirtual os come out? >> and what is the advantages for this kind of os? >> =20 > > There's two aspects to this answer, I guess: > > 1) Will guests which maintain their own pseudophysical abstraction rath= er than=20 > using Xen's be written? / Will Xen's pseudophysical support be remove= d? > > This isn't likely for the timebeing; Xen needs to incorporate the=20 > pseudophysical support it has for backwards-compatibility purposes and = given=20 > that there's no reason for guests not to use it. > > 2) Will guests that don't use a pseudophysical abstraction at all be wr= itten? > > I don't know if anyone has written / ported an OS that does this... I=20 > wouldn't expect to see any general purpose OSes using this for a while = - they=20 > typically seem to need the pseudophysical abstraction to keep their gen= eric=20 > memory management code happy. > > More minimal, special purposes OSes (especially if targetted to Xen) mi= ght be=20 > able to do away with pseudophysical addresses entirely and just use vir= tual=20 > and machine addresses. > > Cheers, > Mark > > =20