From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH 05/16] KVM-HDR: Implement wallclock over KVM - KVM Virtual Memory Date: Thu, 27 Jan 2011 14:17:26 +0200 Message-ID: <4D416256.1040507@redhat.com> References: <1295892397-11354-1-git-send-email-glommer@redhat.com> <1295892397-11354-6-git-send-email-glommer@redhat.com> <4D4001D6.9030909@redhat.com> <1296044403.15920.54.camel@mothafucka.localdomain> <4D403AEE.6010208@redhat.com> <1296056704.3591.28.camel@mothafucka.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, aliguori@us.ibm.com To: Glauber Costa Return-path: In-Reply-To: <1296056704.3591.28.camel@mothafucka.localdomain> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On 01/26/2011 05:45 PM, Glauber Costa wrote: > On Wed, 2011-01-26 at 17:17 +0200, Avi Kivity wrote: > > On 01/26/2011 02:20 PM, Glauber Costa wrote: > > > On Wed, 2011-01-26 at 13:13 +0200, Avi Kivity wrote: > > > > On 01/24/2011 08:06 PM, Glauber Costa wrote: > > > > > As a proof of concept to KVM - Kernel Virtual Memory, this patch > > > > > implements wallclock grabbing on top of it. At first, it may seem > > > > > as a waste of work to just redo it, since it is working well. But over the > > > > > time, other MSRs were added - think ASYNC_PF - and more will probably come. > > > > > After this patch, we won't need to ever add another virtual MSR to KVM. > > > > > > > > > > > > > So instead of adding MSRs, we're adding area identifiers. What did we gain? > > > > > > * No risk of namespace clashes of any kind, > > > * less need for userspace coordination for feature enablement, > > > > That's a bug, not a feature. > > I don't see why. > I's about feature enablement, not feature discovery. Well, "zero userspace coordination" would be a bug, since it would remove userspace-controlled discovery. Since the userspace patches for these types of features are usually very small, "less coordination" doesn't buy us much. > > > > > * size information goes together with base, allowing for extending > > > structures (well, maybe I should add versioning explicitly?) > > > > > > > We could do that as well with wrmsr, by having the size as the first > > field of the structure. Usually the size isn't really interesting, > > though, since you need to discover/enable the new features independently. > > Which structure? For msrs, we're usually going for just an u64, but of > course we could change that when needed. It's usually a physical address of a structure (together with an enable bit). -- error compiling committee.c: too many arguments to function