From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marcelo Tosatti <mtosatti@redhat.com>
Subject: Re: [RFC PATCH v3 1/2] add support for Hyper-V reference time counter
Date: Thu, 2 Jan 2014 18:05:55 -0200
Message-ID: <20140102200555.GA9687@amt.cnet>
References: <1386502419-26614-1-git-send-email-vrozenfe@redhat.com>
 <1386502419-26614-2-git-send-email-vrozenfe@redhat.com>
 <20131211185305.GA28207@amt.cnet>
 <52C56684.7040706@dlhnet.de>
 <20140102135721.GA9578@amt.cnet>
 <52C58EE7.1040004@dlhnet.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: Vadim Rozenfeld <vrozenfe@redhat.com>, kvm@vger.kernel.org,
	pbonzini@redhat.com
To: Peter Lieven <pl@dlhnet.de>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:48515 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751398AbaABUGN (ORCPT <rfc822;kvm@vger.kernel.org>);
	Thu, 2 Jan 2014 15:06:13 -0500
Content-Disposition: inline
In-Reply-To: <52C58EE7.1040004@dlhnet.de>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Thu, Jan 02, 2014 at 05:08:07PM +0100, Peter Lieven wrote:
> > Not using irq disable/enable pairs. The subtraction is not dependant on
> > any particular time.
> > 
> >                 local_irq_disable();
> >                 now_ns = get_kernel_ns();
> >                 local_irq_enable();
> >                 delta = user_ns.clock - now_ns;
> > 
> > Any interrupt that would affect the value of get_kernel_ns(), would
> > have a similar effect before the interrupts are disabled. So the 
> > disable/enable pair achieves nothing in practice. It was copied from
> > kvm_guest_time_update.
> > 
> > 
> 
> Thanks for clarifying. What about get_kernel_ns() can't this be interrupted?
> 
> Peter

It can handle interrupts. 

Disabling interrupts there is used when a
{rdtsc, system_time=get_kernel_ns()} tuple is wanted, to reduce
the physical time difference between execution of rdtsc and get_kernel_ns().