From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751284AbXCRFVA (ORCPT ); Sun, 18 Mar 2007 01:21:00 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751065AbXCRFVA (ORCPT ); Sun, 18 Mar 2007 01:21:00 -0400 Received: from mtaout5.012.net.il ([84.95.2.13]:13313 "EHLO mtaout5.012.net.il" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751218AbXCRFVA (ORCPT ); Sun, 18 Mar 2007 01:21:00 -0400 Date: Sun, 18 Mar 2007 07:20:57 +0200 From: Avi Kivity Subject: Re: [kvm-devel] [PATCH 0/15] KVM userspace interface updates In-reply-to: <20070316083650.GA8525@osiris.boeblingen.de.ibm.com> To: Heiko Carstens Cc: kvm-devel@lists.sourceforge.net, Andrew Morton , linux-kernel@vger.kernel.org Message-id: <45FCCC39.7090104@qumranet.com> MIME-version: 1.0 Content-type: text/plain; charset=ISO-8859-1; format=flowed Content-transfer-encoding: 7BIT X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-2.1.12 (firebolt.argo.co.il [0.0.0.0]); Sun, 18 Mar 2007 07:20:58 +0200 (IST) References: <11736212072915-git-send-email-avi@qumranet.com> <20070316083650.GA8525@osiris.boeblingen.de.ibm.com> User-Agent: Thunderbird 1.5.0.9 (X11/20070212) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Heiko Carstens wrote: > On Sun, Mar 11, 2007 at 03:53:12PM +0200, Avi Kivity wrote: > >> This patchset updates the kvm userspace interface to what I hope will >> be the long-term stable interface. Provisions are included for extending >> the interface later. The patches address performance and cleanliness >> concerns. >> > > Searching the mailing list I figured that as soons as the interface seems > to be stable, kvm should/would switch to a system call based interface. > I assume the userspace interface might still change a lot, especially if > kvm is ported to new architectures. > But the general question is: do you still plan to switch to a syscall > interface? > I don't have any present plans for that. Maybe when the interface starts to evolve at a slower pace, or if it is shown to be significantly faster. Not that interface stabilization here doesn't mean a freeze; it means that backwards compatibility starts when this gets merged. -- Do not meddle in the internals of kernels, for they are subtle and quick to panic.