From: "H. Peter Anvin" <hpa@kernel.org>
Subject: Re: [PATCH 03 of 36] x86: add memory barriers to wrmsr
Date: Wed, 25 Jun 2008 16:18:31 -0700
Message-ID: <4862D247.2010709@kernel.org>
References: <patchbomb.1214367536@localhost>	<93c7057b1f4acae501b2.1214367539@localhost>	<20080624214441.13202f12@infradead.org>	<4862B3E9.50601@goop.org> <20080625153136.2b3b6737@infradead.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Return-path: <linux-kernel-owner+glk-linux-kernel-3=40m.gmane.org-S1756794AbYFYXzE@vger.kernel.org>
In-Reply-To: <20080625153136.2b3b6737@infradead.org>
Sender: linux-kernel-owner@vger.kernel.org
To: Arjan van de Ven <arjan@infradead.org>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>, Ingo Molnar <mingo@elte.hu>, LKML <linux-kernel@vger.kernel.org>, x86@kernel.org, xen-devel <xen-devel@lists.xensource.com>, Stephen Tweedie <sct@redhat.com>, Eduardo Habkost <ehabkost@redhat.com>, Mark McLoughlin <markmc@redhat.com>
List-Id: xen-devel@lists.xenproject.org

Arjan van de Ven wrote:
>>>   
>> I suppose, though I would be inclined to put the barriers in the
>> wrmsr macro itself to act as documentation. 
> 
> 
> yeah I meant like this:
> 
> static inline void native_write_msr(unsigned int msr,
>  				    unsigned low, unsigned high)
> {
> 	barrier();
> 	asm volatile("wrmsr" : : "c" (msr), "a"(low), "d" (high));
> 	barrier();
> }
> 
> or in the same in the thing that calls this.
> 

Actually, I believe the barrier(); before is actually incorrect, since 
it would affect the wrmsr() register arguments rather than the wrmsr 
instruction itself.

	-hpa