All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Martin Schwidefsky <schwidefsky@googlemail.com>
Cc: Avi Kivity <avi@qumranet.com>, Arnd Bergmann <arnd@arndb.de>,
	kvm-devel@lists.sourceforge.net, akpm@osdl.org,
	linux-kernel@vger.kernel.org
Subject: Re: [kvm-devel] [PATCH] KVM: Avoid using vmx instruction directly
Date: Fri, 10 Nov 2006 11:38:32 -0800	[thread overview]
Message-ID: <4554D538.10404@goop.org> (raw)
In-Reply-To: <6e0cfd1d0611100446j77a27b29jc23f76a515451377@mail.gmail.com>

Martin Schwidefsky wrote:
> On 11/10/06, Jeremy Fitzhardinge <jeremy@goop.org> wrote:
>> >> Or gcc
>> >> might move the assignment of phys_addr to after the inline assembly.
>> >>
>> > "asm volatile" prevents that (and I'm not 100% sure it's necessary).
>>
>> No, it won't necessarily.  "asm volatile" simply forces gcc to emit the
>> assembler, even if it thinks its output doesn't get used.  It makes no
>> ordering guarantees with respect to other code (or even other "asm
>> volatiles").   The "memory" clobbers should fix the ordering of the asms
>> though.
>
> The "memory" clobber just tells the compiler that any memory object
> might get access by the inline. 

I just meant that two asms with a "memory" clobber will be generated
with a fixed ordering, which "asm volatile" does not necessarily do.

    J

WARNING: multiple messages have this Message-ID (diff)
From: Jeremy Fitzhardinge <jeremy-TSDbQ3PG+2Y@public.gmane.org>
To: Martin Schwidefsky <schwidefsky-gM/Ye1E23mwN+BqQ9rBEUg@public.gmane.org>
Cc: kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org,
	akpm-3NddpPZAyC0@public.gmane.org,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH] KVM: Avoid using vmx instruction directly
Date: Fri, 10 Nov 2006 11:38:32 -0800	[thread overview]
Message-ID: <4554D538.10404@goop.org> (raw)
In-Reply-To: <6e0cfd1d0611100446j77a27b29jc23f76a515451377-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>

Martin Schwidefsky wrote:
> On 11/10/06, Jeremy Fitzhardinge <jeremy-TSDbQ3PG+2Y@public.gmane.org> wrote:
>> >> Or gcc
>> >> might move the assignment of phys_addr to after the inline assembly.
>> >>
>> > "asm volatile" prevents that (and I'm not 100% sure it's necessary).
>>
>> No, it won't necessarily.  "asm volatile" simply forces gcc to emit the
>> assembler, even if it thinks its output doesn't get used.  It makes no
>> ordering guarantees with respect to other code (or even other "asm
>> volatiles").   The "memory" clobbers should fix the ordering of the asms
>> though.
>
> The "memory" clobber just tells the compiler that any memory object
> might get access by the inline. 

I just meant that two asms with a "memory" clobber will be generated
with a fixed ordering, which "asm volatile" does not necessarily do.

    J

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

  reply	other threads:[~2006-11-10 19:39 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-11-09 11:08 [PATCH] KVM: Avoid using vmx instruction directly Avi Kivity
2006-11-09 11:08 ` Avi Kivity
2006-11-09 13:29 ` [kvm-devel] " Arnd Bergmann
2006-11-09 13:29   ` Arnd Bergmann
2006-11-09 13:36   ` [kvm-devel] " Avi Kivity
2006-11-09 13:36     ` Avi Kivity
2006-11-09 14:42     ` [kvm-devel] " Arnd Bergmann
2006-11-09 14:42       ` Arnd Bergmann
2006-11-09 14:52       ` [kvm-devel] " Avi Kivity
2006-11-09 16:37         ` Arnd Bergmann
2006-11-09 16:51           ` Avi Kivity
2006-11-09 23:39         ` Jeremy Fitzhardinge
2006-11-10 12:46           ` Martin Schwidefsky
2006-11-10 19:38             ` Jeremy Fitzhardinge [this message]
2006-11-10 19:38               ` Jeremy Fitzhardinge
2006-11-21 18:35           ` [kvm-devel] " H. Peter Anvin
2006-11-21 18:35             ` H. Peter Anvin
2006-11-21 19:41             ` [kvm-devel] " Avi Kivity
2006-11-21 19:41               ` Avi Kivity
2006-11-21 20:50             ` [kvm-devel] " Jeremy Fitzhardinge
2006-11-21 20:50               ` Jeremy Fitzhardinge
2006-11-22  6:42               ` [kvm-devel] " Avi Kivity
2006-11-22  6:42                 ` Avi Kivity
2006-11-22  9:10                 ` [kvm-devel] " Jeremy Fitzhardinge
2006-11-22  9:10                   ` Jeremy Fitzhardinge

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4554D538.10404@goop.org \
    --to=jeremy@goop.org \
    --cc=akpm@osdl.org \
    --cc=arnd@arndb.de \
    --cc=avi@qumranet.com \
    --cc=kvm-devel@lists.sourceforge.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=schwidefsky@googlemail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.