From mboxrd@z Thu Jan  1 00:00:00 1970
From: Anthony Liguori <anthony@codemonkey.ws>
Subject: Re: [patch 0/7] force the TSC unreliable by reporting C2 state
Date: Wed, 18 Jun 2008 15:09:41 -0500
Message-ID: <48596B85.7090008@codemonkey.ws>
References: <20080618164205.108219607@localhost.localdomain>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: Avi Kivity <avi@qumranet.com>, kvm@vger.kernel.org
To: Marcelo Tosatti <mtosatti@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from yw-out-2324.google.com ([74.125.46.29]:37868 "EHLO
	yw-out-2324.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1753954AbYFRUKY (ORCPT <rfc822;kvm@vger.kernel.org>);
	Wed, 18 Jun 2008 16:10:24 -0400
Received: by yw-out-2324.google.com with SMTP id 9so234258ywe.1
        for <kvm@vger.kernel.org>; Wed, 18 Jun 2008 13:10:12 -0700 (PDT)
In-Reply-To: <20080618164205.108219607@localhost.localdomain>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

Marcelo Tosatti wrote:
> Avi, I don't think this causes such a huge performance regression. NOHZ
> makes the frequency of timer reads go down significantly.
>   

Have we yet determined why the TSC is so unstable in the first place?  
In theory, it should be relatively stable on single-node Intel and 
Barcelona chips.

Regards,

Anthony Liguori

> As for constant tick guests, well, the impact will be similar to
> changing to SMP, since those fallback to ACPI timer anyway now.
>
> The C2 emulation is required by Ubuntu 7.10 for example, which refuses
> to process the CST notification.
>
> Stock Linux kernels (as old as 2.6.20) do mark C2 invalid upon the CST
> notification.
>
>