LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Michael Ellerman <michael@ellerman.id.au>
To: Arjan van de Ven <arjan@infradead.org>
Cc: Geoff Levand <geoffrey.levand@am.sony.com>,
	linuxppc-dev@ozlabs.org, Roel Kluin <12o3l@tiscali.nl>,
	Willy Tarreau <w@1wt.eu>, lkml <linux-kernel@vger.kernel.org>,
	cbe-oss-dev@ozlabs.org
Subject: Re: [PATCH 1/3] Fix Unlikely(x) == y
Date: Sun, 17 Feb 2008 22:50:03 +1100	[thread overview]
Message-ID: <1203249003.6718.24.camel@concordia> (raw)
In-Reply-To: <20080216103927.2a02352b@laptopd505.fenrus.org>

[-- Attachment #1: Type: text/plain, Size: 2148 bytes --]

On Sat, 2008-02-16 at 10:39 -0800, Arjan van de Ven wrote:
> > >> > you found a great set of bugs..
> > >> > but to be honest... I suspect it's just best to remove unlikely
> > >> > altogether for these cases; unlikely() is almost a
> > >> > go-faster-stripes thing, and if you don't know how to use it you
> > >> > shouldn't be using it... so just removing it for all wrong cases
> > >> > is actually the best thing to do imo.

Hi Arjan,

In general I agree with you that unlikely() is overused, and often in
inappropriate places.

> for mordern (last 10 years) x86 cpus... the cpu branchpredictor is better
> than the coder in general. Same for most other architectures.
> 
> unlikely() creates bigger code as well.
> 
> Now... we're talking about your super duper hotpath function here right?
> One where you care about 0.5 cycle speed improvement? (less on modern
> systems ;)

The first patch was to platforms/ps3 code, which runs on the Cell, in
particular the PPE ... which is not an x86 :)

eg:

[michael@schoenaich ~]$ cat branch.c
#include <stdio.h>
int main(void)
{
        int i, j;

        for (i = 0, j = 0; i < 1000000000; i++)
                if (i % 4 == 0)
                        j++;

        printf("j = %d\n", j);
        return 0;
}
[michael@schoenaich ~]$ ppu-gcc -Wall -O3 -o branch branch.c
[michael@schoenaich ~]$ time ./branch
real    0m5.172s

[michael@schoenaich ~]$ cat branch.c
..
        for (i = 0, j = 0; i < 1000000000; i++)
                if (__builtin_expect(i % 4 == 0, 0))
                        j++;
..
[michael@schoenaich ~]$ ppu-gcc -Wall -O3 -o branch branch.c
[michael@schoenaich ~]$ time ./branch
real    0m3.762s


Which looks as though unlikely() is helping. Admittedly we don't have a
lot of kernel code that looks like that, but at least unlikely is doing
what we want it to.

cheers

-- 
Michael Ellerman
OzLabs, IBM Australia Development Lab

wwweb: http://michael.ellerman.id.au
phone: +61 2 6212 1183 (tie line 70 21183)

We do not inherit the earth from our ancestors,
we borrow it from our children. - S.M.A.R.T Person

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

  reply	other threads:[~2008-02-17 11:50 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-02-16 16:08 Roel Kluin
2008-02-16 17:25 ` Arjan van de Ven
2008-02-16 17:33   ` Willy Tarreau
2008-02-16 17:42     ` Arjan van de Ven
2008-02-16 17:58       ` Willy Tarreau
2008-02-16 18:29         ` Arjan van de Ven
2008-02-17  9:45         ` [Cbe-oss-dev] " Andrew Pinski
2008-02-17 10:08           ` Willy Tarreau
2008-02-16 18:31       ` Geoff Levand
2008-02-16 18:39         ` Arjan van de Ven
2008-02-17 11:50           ` Michael Ellerman [this message]
2008-02-18 13:56             ` Adrian Bunk
2008-02-18 14:01               ` Geert Uytterhoeven
2008-02-18 14:13                 ` Adrian Bunk
2008-02-18 21:46                   ` Michael Ellerman
2008-02-19  7:43                     ` Adrian Bunk
2008-02-18 19:22                 ` [Cbe-oss-dev] " Andrew Pinski
2008-02-18 14:27               ` David Howells
2008-02-18 14:59                 ` Roel Kluin
2008-02-18 18:11                 ` Valdis.Kletnieks
2008-02-18 18:33                   ` Arjan van de Ven
2008-02-18 14:39       ` Andi Kleen
2008-02-19  2:33         ` Nick Piggin
2008-02-19  2:40           ` Arjan van de Ven
2008-02-19  4:41             ` Nick Piggin
2008-02-19  5:58           ` Willy Tarreau
2008-02-19  6:20             ` Nick Piggin
2008-02-19  9:28             ` Andi Kleen
2008-02-20  7:32               ` Willy Tarreau
2008-02-19  9:25           ` Andi Kleen
2008-02-19  9:46             ` Nick Piggin
2008-02-19  9:57               ` Andi Kleen
2008-02-19 22:25                 ` Nick Piggin
2008-02-16 18:41 ` Geoff Levand

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=1203249003.6718.24.camel@concordia \
    --to=michael@ellerman.id.au \
    --cc=12o3l@tiscali.nl \
    --cc=arjan@infradead.org \
    --cc=cbe-oss-dev@ozlabs.org \
    --cc=geoffrey.levand@am.sony.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxppc-dev@ozlabs.org \
    --cc=w@1wt.eu \
    --subject='Re: [PATCH 1/3] Fix Unlikely(x) == y' \
    /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

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).