From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hollis Blanchard Subject: Re: KVM: MMU: nuke shadowed pgtable pages and pte's on memslot destruction Date: Mon, 21 Jul 2008 17:22:08 -0500 Message-ID: <200807211722.08528.hollisb@us.ibm.com> References: <20080704010618.GA25834@dmt.cnet> <1216674207.22095.18.camel@localhost.localdomain> <20080721213440.GA15961@dmt.cnet> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Cc: Avi Kivity , kvm-devel , "Zhang, Xiantao" To: Marcelo Tosatti Return-path: Received: from e35.co.us.ibm.com ([32.97.110.153]:34459 "EHLO e35.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751014AbYGUWW3 (ORCPT ); Mon, 21 Jul 2008 18:22:29 -0400 Received: from d03relay04.boulder.ibm.com (d03relay04.boulder.ibm.com [9.17.195.106]) by e35.co.us.ibm.com (8.13.8/8.13.8) with ESMTP id m6LMMS7i006681 for ; Mon, 21 Jul 2008 18:22:28 -0400 Received: from d03av02.boulder.ibm.com (d03av02.boulder.ibm.com [9.17.195.168]) by d03relay04.boulder.ibm.com (8.13.8/8.13.8/NCO v9.0) with ESMTP id m6LMMSJG171212 for ; Mon, 21 Jul 2008 16:22:28 -0600 Received: from d03av02.boulder.ibm.com (loopback [127.0.0.1]) by d03av02.boulder.ibm.com (8.12.11.20060308/8.13.3) with ESMTP id m6LMMRSq006026 for ; Mon, 21 Jul 2008 16:22:27 -0600 In-Reply-To: <20080721213440.GA15961@dmt.cnet> Content-Disposition: inline Sender: kvm-owner@vger.kernel.org List-ID: On Monday 21 July 2008 16:34:40 Marcelo Tosatti wrote: > On Mon, Jul 21, 2008 at 04:03:27PM -0500, Hollis Blanchard wrote: > > > > By the way, what is the testcase for this, i.e. how do I remove a > > memslot? > > The testcase I used was RH6.2 graphical install, which changes the > cirrus mode from linear frame buffer to the standard one, thus > destroying the vram memory slot (should be able to do that with the > framebuffer driver). Hmm, we're not using any graphics adapters with PowerPC right now. I've skimmed through cirrus_vga.c but I don't see any obvious "mode change" routine; what is the function that triggers this behavior? Some "(un)register physical memory" call? > All you need on this callback is destroy all shadow mappings, to avoid > the MMU code from referencing a memslot that is no longer existant. In addition to that, I also need a testcase to actually exercise the code... :) -- Hollis Blanchard IBM Linux Technology Center