From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.12] helo=sc8-sf-mx2.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.24) id 1AZxE1-0008N7-3h for user-mode-linux-devel@lists.sourceforge.net; Fri, 26 Dec 2003 11:02:17 -0800 Received: from smtp001.mail.ukl.yahoo.com ([217.12.11.32]) by sc8-sf-mx2.sourceforge.net with smtp (Exim 4.24) id 1AZxE0-00032Y-Ir for user-mode-linux-devel@lists.sourceforge.net; Fri, 26 Dec 2003 11:02:16 -0800 From: BlaisorBlade Subject: Re: [uml-devel] UML patches against vanilla 2.6.0 References: <200312261905.48277.blaisorblade_spam@yahoo.it> <20031226182737.GK2378@erdos.home> In-Reply-To: <20031226182737.GK2378@erdos.home> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Disposition: inline Message-Id: <200312261956.13626.blaisorblade_spam@yahoo.it> Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: The user-mode Linux development list List-Post: List-Help: List-Subscribe: , List-Archive: Date: Fri, 26 Dec 2003 19:56:13 +0100 Content-Transfer-Encoding: quoted-printable To: user-mode-linux-devel@lists.sourceforge.net Alle 19:27, venerd=EC 26 dicembre 2003, Cameron Patrick ha scritto: > On Fri, Dec 26, 2003 at 07:05:48PM +0100, BlaisorBlade wrote: > | Go here: > | > | http://web.tiscali.it/no-redirect-tiscali/blaisorblade/index.html > > Which says: > > The only problematic patch is A-04-MM-compile-fix.patch, but > maybe i386 can live even with it (I don't remember if I got a > successful compilation with it applied). However, I think it's > not at all safe even if it compiles. > > and in the README: > > but I think you'd better avoid doing this or do a complete > backup, since you'll need it...). > Eeek! I just upgraded my home desktop to a 2.6.0 kernel with that patch > applied (actually as part of Jeff's 2.6.0-test9 patch) and so far it > seems to be working. Are you sure? If you compile a stock 2.6.0 with Jeff's 2.6.0-test9 patch an= d=20 try to compile an i386 kernel, you'll get soon errors about=20 do_mmap_pgoff(except if you applied a host-skas to your 2.6 tree). > Looking at the patch, it seems innocuous enough - > why should it be unsafe on an i386 kernel? I don't know actually, but it doesn't seem innocuous. Instead, FIXADDR_USER_START is replaced by FIXADDR_START(+ some syntactical= =20 changes that do not hurt). And they have a different meaning(see=20 include/asm-i386/fixmap.h). Since I'm not a VM hacker, and I don't understand the change, I just say th= at=20 that shouldn't be used, for prudence's sake. That mention about backup is=20 maybe a bit exaggerated, but it's a form of "no warranty" disclaimer. *NOTE*: the combo patches from Ingo Molnar contain this, too(I don't know i= f=20 he noticed this change at all), and he probably used this without problem. = So=20 I may be wrong at all. Bye --=20 cat <