From mboxrd@z Thu Jan  1 00:00:00 1970
From: Christoffer Dall <cdall@linaro.org>
Subject: Re: vtime accounting
Date: Wed, 15 Mar 2017 09:30:48 +0100
Message-ID: <20170315083048.GI1277@cbox>
References: <20170308105700.GA109453@lvm>
 <20170313162259.GE18298@potion>
 <20170314082601.GC1277@cbox>
 <20170314165858.GA5435@potion>
 <1d205be1-50ab-59cf-9908-dbbe8c939309@redhat.com>
 <20170314184133.GG1277@cbox>
 <91c3d503-952a-d376-b5d3-df64bc2b6332@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        Christoffer Dall <christoffer.dall@linaro.org>,
        kvm@vger.kernel.org, Marc Zyngier <marc.zyngier@arm.com>,
        Rik van Riel <riel@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mail-wr0-f174.google.com ([209.85.128.174]:35523 "EHLO
        mail-wr0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751145AbdCOIbC (ORCPT <rfc822;kvm@vger.kernel.org>);
        Wed, 15 Mar 2017 04:31:02 -0400
Received: by mail-wr0-f174.google.com with SMTP id g10so5845288wrg.2
        for <kvm@vger.kernel.org>; Wed, 15 Mar 2017 01:31:01 -0700 (PDT)
Content-Disposition: inline
In-Reply-To: <91c3d503-952a-d376-b5d3-df64bc2b6332@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Wed, Mar 15, 2017 at 09:05:28AM +0100, Paolo Bonzini wrote:
> 
> 
> On 14/03/2017 19:41, Christoffer Dall wrote:
> > So would it be worth considering factoring out vtime accounting from
> > guest_enter/exit, such that we could do the vtime accounting from vcpu
> > load/put and mark the RCU extended quiescent state in the run loop?
> > 
> > Disclaimer: I haven't completely convinced myself that vtime accounting
> > from load/put works as it should.
> 
> Me neither, but if it works, it's worth at least benchmarking it to make
> an informed decision.
> 

I'll try to prototype it and see how it looks.

> > For example, when servicing a VM from
> > KVM, should we really be accounting this as kernel time, or as guest
> > time?
> 
> Servicing the request should be fast enough that it doesn't matter.  For
> the most common x86 vmexits, 50% of the time is spent executing the x86
> world switch microcode, and that is already being accounted as guest
> time already.  On microbenchmarks of course it would make a difference:
> 
> $ time ./x86/run x86/vmexit.flat -append cpuid -smp 4
> [...]
> real	0m1.327s
> user	0m1.877s        << this is really guest time
> sys	0m1.844s        << this is spent mostly in KVM_RUN
> 
> 
Good point, and if it's for statistics, this light change in behavior
can hardly constitute any form of user breakage.

Thanks,
-Christoffer