From: Gabriel Paubert <paubert@iram.es>
To: Arjan van de Ven <arjanv@redhat.com>
Cc: linux-kernel@vger.kernel.org, torvalds@osdl.org, akpm@osdl.org
Subject: Re: using gcc built-ins for bitops?
Date: Thu, 24 Jun 2004 13:51:46 +0200 [thread overview]
Message-ID: <20040624115146.GA7513@iram.es> (raw)
In-Reply-To: <20040624070936.GB30057@devserv.devel.redhat.com>
On Thu, Jun 24, 2004 at 09:09:36AM +0200, Arjan van de Ven wrote:
> Hi,
>
> gcc 3.4 gained support for several typical bitops as builtin directives.
> Using these over inline asm has a few advantages:
> * gcc can optimize constants into these better
> * gcc can reorder and schedule the code better
> * gcc can allocate registers etc better for the code
>
> The question is if we consider it desirable to go down this road or not. In
> order to help that discussion I've attached a patch below that switches the
> i386 ffz() function to the gcc builtin version, conditional on gcc having
> support for this. Before I go down the road of converting more functions
> and/or architectures.... is this worth doing?
>
> Greetings,
> Arjan van de Ven
>
>
> --- linux-2.6.7/include/asm-i386/bitops.h~ 2004-06-23 23:45:06.048614387 +0200
> +++ linux-2.6.7/include/asm-i386/bitops.h 2004-06-23 23:45:06.048614387 +0200
> @@ -344,6 +344,8 @@
> *
> * Undefined if no zero exists, so code should check against ~0UL first.
> */
> +
> +#ifndef HAVE_BUILTIN_CTZL
> static inline unsigned long ffz(unsigned long word)
> {
> __asm__("bsfl %1,%0"
> @@ -351,6 +353,12 @@
> :"r" (~word));
> return word;
> }
> +#else
> +static inline unsigned long ffz (unsigned long word)
> +{
> + return __builtin_ctzl (~word);
> +}
> +#endif
>
> /**
> * __ffs - find first bit in word.
> --- linux-2.6.7/include/linux/compiler-gcc3.h~ 2004-06-24 09:26:04.123455290 +0200
> +++ linux-2.6.7/include/linux/compiler-gcc3.h 2004-06-24 09:26:04.123455290 +0200
> @@ -19,6 +19,11 @@
> # define __attribute_used__ __attribute__((__unused__))
> #endif
>
> +#if __GNUC_MINOR__ >= 4
Please do no test only on minor. People have been arguing
whether the next major release of GCC should be called 3.5
or 4.0 since tree-ssa has been merged.
This said, I'd rather use GCC intrinsics when they can
perform the task, especially the ones that expand
to more than one machine instruction.
> +#define HAVE_BUILTIN_CTZL
> +#endif
Gabriel
next prev parent reply other threads:[~2004-06-24 11:57 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-06-24 7:09 Arjan van de Ven
2004-06-24 9:00 ` Andrew Morton
2004-06-24 9:06 ` Arjan van de Ven
2004-06-24 13:41 ` Ben Collins
2004-06-24 9:17 ` Arjan van de Ven
2004-06-24 11:31 ` Arjan van de Ven
2004-06-24 12:05 ` Jakub Jelinek
2004-06-24 12:35 ` Arjan van de Ven
2004-06-24 15:30 ` Linus Torvalds
2004-06-24 9:31 ` Paul Jackson
2004-06-24 10:36 ` Jakub Jelinek
2004-06-24 15:25 ` Linus Torvalds
2004-06-24 11:51 ` Gabriel Paubert [this message]
2004-06-24 12:21 ` Arjan van de Ven
[not found] <2awGH-DF-17@gated-at.bofh.it>
2004-06-24 13:46 ` Pascal Schmidt
2004-06-24 13:49 ` Arjan van de Ven
2004-06-24 13:51 ` Pascal Schmidt
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=20040624115146.GA7513@iram.es \
--to=paubert@iram.es \
--cc=akpm@osdl.org \
--cc=arjanv@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=torvalds@osdl.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
all inboxes | Powered by JetHome®