From: "Alexander van Heukelum" <heukelum@fastmail.fm>
To: "Joe Perches" <joe@perches.com>,
"Matti Aarnio" <matti.aarnio@zmailer.org>
Cc: "Harvey Harrison" <harvey.harrison@gmail.com>,
"Alexander van Heukelum" <heukelum@mailshack.com>,
"LKML" <linux-kernel@vger.kernel.org>
Subject: Re: Alternative implementation of the generic __ffs
Date: Sun, 20 Apr 2008 10:42:21 +0200 [thread overview]
Message-ID: <1208680941.5276.1248837773@webmail.messagingengine.com> (raw)
In-Reply-To: <1208660817.12388.40.camel@localhost>
On Sat, 19 Apr 2008 20:06:57 -0700, "Joe Perches" <joe@perches.com>
said:
> On Sun, 2008-04-20 at 01:29 +0300, Matti Aarnio wrote:
> > I am curious, why not take the code already in glibc ffs() for ARM ?
> > That is, if the ffs() is all that important detail in kernel ?
Hi,
The glibc version is based on a table-lookup. This makes it
behave differently in hot and cold cache situations. That's
fine if __ffs is used in tight loops, but in the kernel such
use of __ffs is avoided because it might be slow. I added it
to the benchmark, but it would need testing for the cold
cache case too.
As for the importance of __ffs in the kernel: as far as I
know the hot-spots in the kernel using __ffs are the
schedular (sched_find_first_bit) and the cpu mask walking
code (for_each_cpu_mask).
Greetings,
Alexander
> Here's test results with the glibc ffs implementation.
> (small const is still using slower add rather than or)
Added, thanks.
> $ gcc -Os -fomit-frame-pointer ffs.c
> $ ./a.out
> Original: 3155 tics, 8331 tics
> New: 4211 tics, 8793 tics
> Smallest: 4019 tics, 7754 tics
> Small const: 3552 tics, 6308 tics
> glibc: 2816 tics, 6911 tics
> Empty loop: 1516 tics, 2244 tics
>
> $ gcc -O2 -fomit-frame-pointer ffs.c
> $ ./a.out
> Original: 3155 tics, 7828 tics
> New: 4792 tics, 8825 tics
> Smallest: 4401 tics, 7155 tics
> Small const: 3539 tics, 5805 tics
> glibc: 2720 tics, 7061 tics
> Empty loop: 1516 tics, 2148 tics
>
> $ gcc -O3 -fomit-frame-pointer ffs.c
> $ ./a.out
> Original: 3080 tics, 7706 tics
> New: 4721 tics, 8663 tics
> Smallest: 4334 tics, 7116 tics
> Small const: 3466 tics, 5672 tics
> glibc: 2649 tics, 6939 tics
> Empty loop: 1444 tics, 2012 tics
>
>
--
Alexander van Heukelum
heukelum@fastmail.fm
--
http://www.fastmail.fm - A no graphics, no pop-ups email service
next prev parent reply other threads:[~2008-04-20 8:42 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-03-31 17:15 [PATCH] x86: generic versions of find_first_(zero_)bit, convert i386 Alexander van Heukelum
2008-03-31 17:22 ` Stephen Hemminger
2008-03-31 19:38 ` Alexander van Heukelum
2008-03-31 21:58 ` Andi Kleen
2008-04-01 8:47 ` Ingo Molnar
2008-04-01 9:46 ` Alexander van Heukelum
2008-04-01 15:41 ` [PATCH] x86: switch x86_64 to generic find_first_bit Alexander van Heukelum
2008-04-01 15:42 ` [PATCH] x86: optimize find_first_bit for small bitmaps Alexander van Heukelum
2008-04-01 15:47 ` [PATCH] x86: remove x86-specific implementations of find_first_bit Alexander van Heukelum
2008-04-03 9:34 ` Alexander van Heukelum
2008-04-04 8:47 ` Ingo Molnar
2008-04-06 17:03 ` [PATCH] x86: generic versions of find_first_(zero_)bit, convert i386 dean gaudet
2008-04-06 18:51 ` Alexander van Heukelum
2008-04-06 20:22 ` dean gaudet
2008-04-07 8:43 ` Ingo Molnar
2008-04-07 10:25 ` Alexander van Heukelum
2008-04-18 20:18 ` Alternative implementation of the generic __ffs Alexander van Heukelum
2008-04-18 23:46 ` dean gaudet
2008-04-19 0:09 ` Harvey Harrison
2008-04-19 0:20 ` dean gaudet
2008-04-19 0:58 ` Joe Perches
2008-04-19 1:04 ` Harvey Harrison
2008-04-19 1:11 ` dean gaudet
2008-04-19 2:55 ` Joe Perches
2008-04-19 4:13 ` dean gaudet
2008-04-19 10:05 ` Mikael Pettersson
2008-04-19 12:10 ` Alexander van Heukelum
2008-04-19 18:17 ` Joe Perches
2008-04-19 20:26 ` Alexander van Heukelum
2008-04-19 22:29 ` Matti Aarnio
2008-04-20 3:06 ` Joe Perches
2008-04-20 8:42 ` Alexander van Heukelum [this message]
2008-04-20 12:31 ` Matti Aarnio
2008-04-21 11:43 ` Alexander van Heukelum
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=1208680941.5276.1248837773@webmail.messagingengine.com \
--to=heukelum@fastmail.fm \
--cc=harvey.harrison@gmail.com \
--cc=heukelum@mailshack.com \
--cc=joe@perches.com \
--cc=linux-kernel@vger.kernel.org \
--cc=matti.aarnio@zmailer.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
Powered by JetHome