From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
"H. Peter Anvin" <hpa@zytor.com>,
Andi Kleen <andi@firstfloor.org>, Ingo Molnar <mingo@elte.hu>,
Jiri Slaby <jirislaby@gmail.com>,
David Miller <davem@davemloft.net>,
zdenek.kabelac@gmail.com, rjw@sisk.pl,
paulmck@linux.vnet.ibm.com, akpm@linux-foundation.org,
linux-ext4@vger.kernel.org, herbert@gondor.apana.org.au,
penberg@cs.helsinki.fi, clameter@sgi.com,
linux-kernel@vger.kernel.org, pageexec@freemail.hu
Subject: Re: [PATCH 1/1] x86: fix text_poke
Date: Mon, 28 Apr 2008 14:02:08 -0700 [thread overview]
Message-ID: <48163B50.8010608@goop.org> (raw)
In-Reply-To: <20080428204342.GD27154@Krystal>
Mathieu Desnoyers wrote:
> Markers, with immediate values, only clobbers the eax register and the
> ZF. It does not restrain inlining nor loop unrolling. It also requires
> gcc to leave the variables in which the marker is interested "live".
That in itself is pretty significant. If that value would otherwise be
constant folded or strength-reduced away, you're putting a big
limitation on what the compiler can do. The mere fact that its
necessary to do something to preserve many values shows how much the
compiler transforms the code away from what's in the source, and
specifically referencing otherwise unused intermediates inhibits that.
In other words, if you weren't preventing optimisations, you wouldn't
need to preserve values as much, because the optimiser wouldn't be
getting rid of them. If you need to preserve lots of values, you're
necessarily preventing the optimiser from doing its job.
J
next prev parent reply other threads:[~2008-04-28 21:02 UTC|newest]
Thread overview: 173+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <200804191522.54334.rjw@sisk.pl>
2008-04-20 19:04 ` 2.6.25-git2: BUG: unable to handle kernel paging request at ffffffffffffffff Rafael J. Wysocki
2008-04-20 19:14 ` Rafael J. Wysocki
2008-04-20 21:31 ` Linus Torvalds
2008-04-21 1:18 ` Herbert Xu
2008-04-21 2:08 ` Paul E. McKenney
2008-04-21 4:59 ` Paul E. McKenney
2008-04-21 5:47 ` Paul E. McKenney
2008-04-21 13:00 ` Ingo Molnar
2008-04-21 16:06 ` Linus Torvalds
2008-04-21 16:24 ` Rafael J. Wysocki
2008-04-21 15:49 ` Linus Torvalds
2008-04-21 17:05 ` Paul E. McKenney
2008-04-21 17:30 ` Linus Torvalds
2008-04-21 17:43 ` Paul E. McKenney
2008-04-22 1:03 ` Herbert Xu
2008-04-22 13:36 ` Paul E. McKenney
2008-04-21 16:12 ` Rafael J. Wysocki
2008-04-21 16:54 ` Linus Torvalds
2008-04-21 17:06 ` Jiri Slaby
2008-04-21 17:19 ` Rafael J. Wysocki
2008-04-21 17:48 ` Linus Torvalds
2008-04-21 18:22 ` Rafael J. Wysocki
2008-04-21 19:38 ` Jiri Slaby
2008-04-21 20:39 ` David Miller
2008-04-21 21:18 ` Jiri Slaby
2008-04-21 21:58 ` Jiri Slaby
2008-04-21 22:26 ` Jiri Slaby
2008-04-21 22:54 ` Paul E. McKenney
2008-04-21 23:02 ` Jiri Slaby
2008-04-21 23:11 ` Zdenek Kabelac
2008-04-21 23:17 ` Jiri Slaby
2008-04-22 0:54 ` Rafael J. Wysocki
2008-04-22 1:14 ` Linus Torvalds
2008-04-22 1:30 ` Rafael J. Wysocki
2008-04-22 9:49 ` Jiri Slaby
2008-04-22 9:53 ` Ingo Molnar
2008-04-22 18:35 ` Zdenek Kabelac
2008-04-22 18:48 ` Linus Torvalds
2008-04-23 8:50 ` Zdenek Kabelac
2008-04-23 15:53 ` Linus Torvalds
2008-04-23 16:58 ` Pekka Enberg
2008-04-23 17:28 ` Zdenek Kabelac
2008-04-23 17:40 ` Ingo Molnar
2008-04-23 18:52 ` Pekka Enberg
2008-04-23 19:05 ` Christoph Lameter
2008-04-23 19:19 ` Pekka J Enberg
2008-04-23 19:28 ` Christoph Lameter
2008-04-23 20:27 ` Zdenek Kabelac
2008-04-24 22:26 ` Jiri Slaby
2008-04-24 22:41 ` Linus Torvalds
2008-04-25 0:57 ` Jiri Slaby
2008-04-24 23:45 ` Linus Torvalds
2008-04-25 7:36 ` Jiri Slaby
2008-04-25 14:09 ` Pavel Machek
2008-04-25 15:30 ` Rafael J. Wysocki
2008-04-25 17:10 ` Jiri Slaby
2008-04-25 9:13 ` David Miller
2008-04-25 12:15 ` Zdenek Kabelac
2008-04-25 12:27 ` Zdenek Kabelac
2008-04-25 15:12 ` [PATCH 1/1] x86: fix text_poke Jiri Slaby
2008-04-25 15:03 ` Linus Torvalds
2008-04-25 15:17 ` Andi Kleen
2008-04-25 19:36 ` Christoph Lameter
2008-04-26 9:59 ` Andi Kleen
2008-04-26 11:16 ` Jiri Slaby
2008-04-26 11:34 ` Andi Kleen
2008-04-28 20:24 ` VIRTUAL_BUG_ON() Christoph Lameter
2008-05-01 19:22 ` [RFC 1/1] mm: add virt to phys debug Jiri Slaby
2008-05-01 20:18 ` Christoph Lameter
2008-05-06 21:54 ` Jiri Slaby
2008-05-07 17:30 ` Christoph Lameter
2008-05-13 14:38 ` Jiri Slaby
2008-04-25 15:19 ` [PATCH 1/1] x86: fix text_poke Ingo Molnar
2008-04-25 15:26 ` Ingo Molnar
2008-04-25 15:32 ` Ingo Molnar
2008-04-25 15:33 ` Linus Torvalds
2008-04-25 15:48 ` Andi Kleen
2008-04-25 16:06 ` Linus Torvalds
2008-04-25 16:19 ` Andi Kleen
2008-04-25 16:24 ` Linus Torvalds
2008-04-25 16:33 ` Ingo Molnar
2008-04-25 18:13 ` Jeremy Fitzhardinge
2008-05-05 2:36 ` Nick Piggin
2008-04-25 16:30 ` Mathieu Desnoyers
2008-04-25 16:42 ` H. Peter Anvin
2008-04-25 17:09 ` Mathieu Desnoyers
2008-04-25 18:37 ` Mathieu Desnoyers
2008-04-25 18:47 ` H. Peter Anvin
2008-04-25 19:19 ` H. Peter Anvin
2008-04-25 20:04 ` Mathieu Desnoyers
2008-04-25 20:09 ` H. Peter Anvin
2008-04-25 20:18 ` H. Peter Anvin
2008-04-25 20:37 ` Mathieu Desnoyers
2008-04-25 20:41 ` H. Peter Anvin
2008-04-25 20:51 ` Linus Torvalds
2008-04-25 21:12 ` Mathieu Desnoyers
2008-04-25 21:15 ` H. Peter Anvin
2008-04-25 21:47 ` Mathieu Desnoyers
2008-04-25 22:07 ` H. Peter Anvin
2008-04-25 22:30 ` Mathieu Desnoyers
2008-04-25 22:36 ` Linus Torvalds
2008-04-28 20:21 ` Ingo Molnar
2008-04-28 20:55 ` Jeremy Fitzhardinge
2008-04-28 21:01 ` H. Peter Anvin
2008-04-28 22:42 ` Mathieu Desnoyers
2008-04-28 20:43 ` Mathieu Desnoyers
2008-04-28 21:02 ` Jeremy Fitzhardinge [this message]
2008-05-04 15:03 ` Mathieu Desnoyers
2008-05-04 16:18 ` H. Peter Anvin
2008-04-25 22:38 ` H. Peter Anvin
2008-04-25 22:04 ` Linus Torvalds
2008-04-25 23:00 ` Mathieu Desnoyers
2008-04-25 23:13 ` Jeremy Fitzhardinge
2008-04-25 23:34 ` Masami Hiramatsu
2008-04-26 6:21 ` Jeremy Fitzhardinge
2008-04-26 11:56 ` Arnaldo Carvalho de Melo
2008-04-26 23:38 ` Jeremy Fitzhardinge
2008-04-27 1:00 ` Arnaldo Carvalho de Melo
2008-04-26 2:12 ` Frank Ch. Eigler
2008-06-05 17:44 ` Frank Ch. Eigler
2008-04-26 6:50 ` Jeremy Fitzhardinge
2008-04-28 0:49 ` Masami Hiramatsu
2008-04-25 21:02 ` David Miller
2008-04-25 21:11 ` H. Peter Anvin
2008-04-25 16:22 ` Ingo Molnar
2008-04-25 16:37 ` Linus Torvalds
2008-04-25 16:43 ` Ingo Molnar
2008-04-25 16:45 ` Ingo Molnar
2008-04-25 16:51 ` Linus Torvalds
2008-04-25 17:02 ` Ingo Molnar
2008-04-25 17:13 ` Linus Torvalds
2008-04-25 17:26 ` Andi Kleen
2008-04-25 17:29 ` Linus Torvalds
2008-04-25 17:53 ` Ingo Molnar
2008-04-25 18:04 ` Ingo Molnar
2008-04-25 18:09 ` Linus Torvalds
2008-04-25 18:19 ` Ingo Molnar
2008-04-25 18:56 ` Ingo Molnar
2008-04-25 18:13 ` Ingo Molnar
2008-04-25 16:52 ` Ingo Molnar
2008-04-25 16:56 ` Andi Kleen
2008-04-25 15:50 ` Ingo Molnar
2008-04-25 15:57 ` H. Peter Anvin
2008-04-25 18:53 ` Pavel Machek
2008-04-25 16:11 ` Linus Torvalds
2008-04-25 15:54 ` Mathieu Desnoyers
2008-04-25 15:59 ` Ingo Molnar
2008-04-25 16:11 ` Mathieu Desnoyers
2008-04-25 15:27 ` Andi Kleen
2008-04-25 20:18 ` David Miller
2008-04-25 15:23 ` Jiri Slaby
2008-04-25 1:35 ` 2.6.25-git2: BUG: unable to handle kernel paging request at ffffffffffffffff David Miller
2008-04-25 1:48 ` Linus Torvalds
2008-04-25 1:57 ` David Miller
2008-04-25 7:41 ` Jiri Slaby
2008-04-25 7:45 ` David Miller
2008-04-25 8:02 ` Jiri Slaby
2008-04-25 10:53 ` Craig Schlenter
2008-04-25 7:42 ` Jiri Slaby
2008-04-25 7:49 ` David Miller
2008-04-25 7:56 ` Jiri Slaby
2008-04-25 7:58 ` David Miller
2008-04-25 8:00 ` Jiri Slaby
2008-04-25 15:47 ` Randy Dunlap
2008-04-22 21:46 ` Rafael J. Wysocki
2008-04-22 19:09 ` Ingo Molnar
2008-04-22 1:15 ` Rafael J. Wysocki
2008-04-22 1:25 ` [ProbableSpam]Re: " Paul E. McKenney
2008-04-21 21:19 ` Linus Torvalds
2008-04-21 21:54 ` David Miller
2008-04-21 13:17 ` Ingo Molnar
2008-04-21 13:35 ` Rafael J. Wysocki
2008-04-21 18:56 ` Ingo Molnar
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=48163B50.8010608@goop.org \
--to=jeremy@goop.org \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=clameter@sgi.com \
--cc=davem@davemloft.net \
--cc=herbert@gondor.apana.org.au \
--cc=hpa@zytor.com \
--cc=jirislaby@gmail.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@polymtl.ca \
--cc=mingo@elte.hu \
--cc=pageexec@freemail.hu \
--cc=paulmck@linux.vnet.ibm.com \
--cc=penberg@cs.helsinki.fi \
--cc=rjw@sisk.pl \
--cc=torvalds@linux-foundation.org \
--cc=zdenek.kabelac@gmail.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).