From mboxrd@z Thu Jan  1 00:00:00 1970
From: Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org>
Subject: Re: [PATCH] KVM: Avoid using vmx instruction directly
Date: Wed, 22 Nov 2006 08:42:32 +0200
Message-ID: <4563F158.3060209@qumranet.com>
References: <4563667B.2060209@goop.org>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Cc: kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org, akpm-3NddpPZAyC0@public.gmane.org,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, "H. Peter Anvin" <hpa-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org>
Return-path: <kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org>
To: Jeremy Fitzhardinge <jeremy-TSDbQ3PG+2Y@public.gmane.org>
In-Reply-To: <4563667B.2060209-TSDbQ3PG+2Y@public.gmane.org>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/kvm-devel>,
	<mailto:kvm-devel-request-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=kvm-devel>
List-Post: <mailto:kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org>
List-Help: <mailto:kvm-devel-request-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org?subject=help>
List-Subscribe: <https://lists.sourceforge.net/lists/listinfo/kvm-devel>,
	<mailto:kvm-devel-request-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org?subject=subscribe>
Sender: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
Errors-To: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
List-Id: kvm.vger.kernel.org

Jeremy Fitzhardinge wrote:
>
>
> Like "volatile" variables, I think "asm volatile" is probably overused.
> If you want to guarantee specific ordering of asms, it's probably better
> to add an explicit dependency between them rather than rely on asm
> volatile; this could either be a "memory" clobber, or something more
> fine-grained.  For example:
>
>     /* need never be instansiated; never actually referenced */
>     extern int spin_sequencer;
>
>     /* %0 never referenced */
>     asm("take spinlock" : "+m" (spin_sequencer)...);
>
>     ...
>
>     /* again, %0 never referenced */
>     asm("release spinlock" : "+m" (spin_sequencer)...);
>

Very interesting.

Will it work on load/store architectures?  Since all memory access is 
through a register, won't the constraint generate a useless register 
load (and a use of the variable)?


-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV