From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH RFC 4/5] KVM: add new members to the memory slot for double buffering of bitmaps Date: Mon, 12 Apr 2010 12:19:41 +0300 Message-ID: <4BC2E5AD.4020207@redhat.com> References: <20100409182732.857de4db.yoshikawa.takuya@oss.ntt.co.jp> <20100409183555.18a64dc7.yoshikawa.takuya@oss.ntt.co.jp> <4BC203C8.9010201@redhat.com> <4BC28257.2080003@oss.ntt.co.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: mtosatti@redhat.com, kvm@vger.kernel.org, fernando@oss.ntt.co.jp To: Takuya Yoshikawa Return-path: Received: from mx1.redhat.com ([209.132.183.28]:25900 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752802Ab0DLJTr (ORCPT ); Mon, 12 Apr 2010 05:19:47 -0400 In-Reply-To: <4BC28257.2080003@oss.ntt.co.jp> Sender: kvm-owner@vger.kernel.org List-ID: On 04/12/2010 05:15 AM, Takuya Yoshikawa wrote: > > OK, but we have one problem: ia64. I checked all architectures' dirty > bitmap > implementations and thought generalizing this work is not so hard > except for > ia64. It's already too different from other parts. > > #ifdef CONFIG_IA64 > unsigned long *dirty_bitmap; > #else > ... > #endif > > is acceptable? I think you can keep the bitmap in userspace, but replace the vmalloc() with get_user_pages() and vmap() (in arch/ia64). 'dirty_bitmap' can then be in kvm->arch. Note: this will likely break ia64 without testing. Please copy the patches to kvm-ia64@vger.kernel.org so they can test and fix them if they want to. The patches should at least build, though. If you don't have an ia64 machine, I can build-test them for you. -- I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain.