public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Satyam Sharma <ssatyam@cse.iitk.ac.in>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	David Howells <dhowells@redhat.com>,
	Nick Piggin <nickpiggin@yahoo.com.au>, Andi Kleen <ak@suse.de>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: [PATCH 7/8] i386: bitops: Kill needless usage of __asm__ __volatile__
Date: Mon, 23 Jul 2007 10:39:47 -0700	[thread overview]
Message-ID: <46A4E7E3.8090105@goop.org> (raw)
In-Reply-To: <Pine.LNX.4.64.0707232202580.22155@cselinux1.cse.iitk.ac.in>

Satyam Sharma wrote:
> Hi Jeremy,
>
>
> On Mon, 23 Jul 2007, Jeremy Fitzhardinge wrote:
>
>   
>> Satyam Sharma wrote:
>>     
>>> From: Satyam Sharma <ssatyam@cse.iitk.ac.in>
>>>
>>> [7/8] i386: bitops: Kill needless usage of __asm__ __volatile__
>>>
>>> Another oddity I noticed in this file. The semantics of __volatile__
>>> when used to qualify inline __asm__ are that the compiler will not
>>> (1) elid, or, (2) reorder, or, (3) intersperse, our inline asm with
>>> the rest of the generated code.
>>>   
>>>       
>> "asm volatile" does not mean that at all.  It only guarantees (1),
>>     
>
>
> Actually, you're probably right about (2), but (3)?
>
> From the gcc manual:
>
> <quote>
>
> Similarly, you can't expect a sequence of volatile asm instructions to
> remain perfectly consecutive. If you want consecutive output, use a
> single asm. Also GCC will perform some optimizations across a volatile
> asm instruction, GCC does not "forget everything" when it encounters a
> volatile asm instruction the way some other compilers do.
>
> </quote>
>
> I'm reading "Similarly, you can't expect a sequence of volatile asm
> instructions to remain perfectly consecutive" to mean they're talking
> about something like:
>
> asm volatile(...);
> asm volatile(...);
> asm volatile(...);
>
> But "If you want consecutive output, use a single asm" probably means:
>
> asm volatile(... multiple instructions here ...);
>
> would actually ensure the code written in there would not be
> interspersed ... at least that's how I read it.
>   

I'm not quite sure what your point is.  The paragraph you quoted is
pretty explicit in saying that volatile doesn't prevent an "asm
volatile" from being interspersed with other code, and the example just
before that is explicit in talking about how to use dependencies to
control the ordering of asm volatiles with respect to surrounding code. 
In fact nothing in that section precludes asm volatiles from being
reordered with respect to each other either; you just have to make sure
your dependency constraints are all correct.

    J

  reply	other threads:[~2007-07-23 17:40 UTC|newest]

Thread overview: 92+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-07-23 16:05 [PATCH 0/8] i386: bitops: Cleanup, sanitize, optimize Satyam Sharma
2007-07-23 16:05 ` [PATCH 1/8] i386: bitops: Update/correct comments Satyam Sharma
2007-07-23 16:05 ` [PATCH 2/8] i386: bitops: Rectify bogus "Ir" constraints Satyam Sharma
2007-07-23 16:10   ` Andi Kleen
2007-07-23 16:21     ` Satyam Sharma
2007-07-23 16:30       ` Andi Kleen
2007-07-23 16:36         ` Jan Hubicka
2007-07-23 18:05         ` H. Peter Anvin
2007-07-23 18:28           ` H. Peter Anvin
2007-07-23 17:57   ` Linus Torvalds
2007-07-23 18:14     ` Satyam Sharma
2007-07-23 18:32       ` Andi Kleen
2007-07-23 18:39     ` H. Peter Anvin
2007-07-23 18:52       ` Satyam Sharma
2007-07-23 16:05 ` [PATCH 3/8] i386: bitops: Rectify bogus "+m" constraints Satyam Sharma
2007-07-23 16:37   ` Andi Kleen
2007-07-23 17:15     ` Satyam Sharma
2007-07-23 17:46   ` Linus Torvalds
2007-07-24  9:22     ` David Howells
2007-07-23 16:05 ` [PATCH 4/8] i386: bitops: Kill volatile-casting of memory addresses Satyam Sharma
2007-07-23 17:52   ` Linus Torvalds
2007-07-24  4:19     ` Nick Piggin
2007-07-24  6:23       ` Satyam Sharma
2007-07-24  7:16         ` Nick Piggin
2007-07-24  9:49     ` Benjamin Herrenschmidt
2007-07-24 17:20       ` Linus Torvalds
2007-07-24 17:39         ` Jeff Garzik
2007-07-25  4:54         ` Nick Piggin
2007-07-23 16:05 ` [PATCH 5/8] i386: bitops: Contain warnings fallout from the death of volatiles Satyam Sharma
2007-07-23 16:05 ` [PATCH 6/8] i386: bitops: Don't mark memory as clobbered unnecessarily Satyam Sharma
2007-07-23 16:13   ` Andi Kleen
2007-07-23 16:26     ` Satyam Sharma
2007-07-23 16:33       ` Andi Kleen
2007-07-23 17:12         ` Satyam Sharma
2007-07-23 17:49           ` Jeremy Fitzhardinge
2007-07-23 17:55   ` Linus Torvalds
2007-07-24  9:52     ` Benjamin Herrenschmidt
2007-07-24 17:24       ` Linus Torvalds
2007-07-24 17:42         ` Trond Myklebust
2007-07-24 18:13           ` Linus Torvalds
2007-07-24 18:28             ` Trond Myklebust
2007-07-24 21:37             ` Benjamin Herrenschmidt
2007-07-24 21:55               ` Trond Myklebust
2007-07-24 22:32                 ` Benjamin Herrenschmidt
2007-07-25  4:10                   ` Nick Piggin
2007-07-24 21:36         ` Benjamin Herrenschmidt
2007-07-24  3:57   ` Nick Piggin
2007-07-24  6:38     ` Satyam Sharma
2007-07-24  7:24       ` Nick Piggin
2007-07-24  8:29         ` Satyam Sharma
2007-07-24  8:39           ` Nick Piggin
2007-07-24  8:38         ` Trent Piepho
2007-07-24 19:39           ` Linus Torvalds
2007-07-24 20:37             ` Andi Kleen
2007-07-24 20:08               ` Linus Torvalds
2007-07-24 21:31                 ` Jeremy Fitzhardinge
2007-07-24 21:46                   ` Linus Torvalds
2007-07-26  1:07             ` Trent Piepho
2007-07-26  1:18               ` Linus Torvalds
2007-07-26  1:22                 ` Linus Torvalds
2007-07-24  9:44       ` David Howells
2007-07-24 10:02         ` Satyam Sharma
2007-07-23 16:06 ` [PATCH 7/8] i386: bitops: Kill needless usage of __asm__ __volatile__ Satyam Sharma
2007-07-23 16:18   ` Andi Kleen
2007-07-23 16:22     ` [PATCH 7/8] i386: bitops: Kill needless usage of __asm__ __volatile__ II Andi Kleen
2007-07-23 16:32     ` [PATCH 7/8] i386: bitops: Kill needless usage of __asm__ __volatile__ Satyam Sharma
2007-07-23 16:23   ` Jeremy Fitzhardinge
2007-07-23 16:43     ` Satyam Sharma
2007-07-23 17:39       ` Jeremy Fitzhardinge [this message]
2007-07-23 18:07         ` Satyam Sharma
2007-07-23 18:28           ` Jeremy Fitzhardinge
2007-07-23 20:29             ` Trent Piepho
2007-07-23 20:40               ` Jeremy Fitzhardinge
2007-07-23 21:06                 ` Trent Piepho
2007-07-23 21:30               ` Andi Kleen
2007-07-23 21:48                 ` Nicholas Miell
2007-07-23 16:06 ` [PATCH 8/8] i386: bitops: smp_mb__{before, after}_clear_bit() definitions Satyam Sharma
2007-07-24  3:53   ` Nick Piggin
2007-07-24  7:34     ` Satyam Sharma
2007-07-24  7:48       ` Jeremy Fitzhardinge
2007-07-24  8:31         ` Nick Piggin
2007-07-24  8:20       ` Nick Piggin
2007-07-24  9:21         ` Satyam Sharma
2007-07-24 10:25           ` Nick Piggin
2007-07-24 11:10             ` Satyam Sharma
2007-07-24 11:32               ` Nick Piggin
2007-07-24 11:45                 ` Satyam Sharma
2007-07-24 12:01                   ` Nick Piggin
2007-07-24 17:12                   ` Linus Torvalds
2007-07-24 19:01                     ` Satyam Sharma
2007-07-30 17:57 ` [PATCH 0/8] i386: bitops: Cleanup, sanitize, optimize Denis Vlasenko
2007-07-31  1:07   ` Satyam Sharma

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=46A4E7E3.8090105@goop.org \
    --to=jeremy@goop.org \
    --cc=ak@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=dhowells@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nickpiggin@yahoo.com.au \
    --cc=ssatyam@cse.iitk.ac.in \
    --cc=torvalds@linux-foundation.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox