[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20080218141340.GB667@cs181133002.pp.htv.fi>
Date: Mon, 18 Feb 2008 16:13:40 +0200
From: Adrian Bunk <bunk@...nel.org>
To: Geert Uytterhoeven <Geert.Uytterhoeven@...ycom.com>
Cc: Michael Ellerman <michael@...erman.id.au>,
Roel Kluin <12o3l@...cali.nl>,
lkml <linux-kernel@...r.kernel.org>, cbe-oss-dev@...abs.org,
linuxppc-dev@...abs.org, Willy Tarreau <w@....eu>,
Arjan van de Ven <arjan@...radead.org>
Subject: Re: [PATCH 1/3] Fix Unlikely(x) == y
On Mon, Feb 18, 2008 at 03:01:35PM +0100, Geert Uytterhoeven wrote:
> On Mon, 18 Feb 2008, Adrian Bunk wrote:
> >
> > This means it generates faster code with a current gcc for your platform.
> >
> > But a future gcc might e.g. replace the whole loop with a division
> > (gcc SVN head (that will soon become gcc 4.3) already does
> > transformations like replacing loops with divisions [1]).
>
> Hence shouldn't we ask the gcc people what's the purpose of __builtin_expect(),
> if it doesn't live up to its promise?
That's a different issue.
My point here is that we do not know how the latest gcc available in the
year 2010 might transform this code, and how a likely/unlikely placed
there might influence gcc's optimizations then.
If this is in hotpath code with a measurable speedup when using
likely/unlikely with a current gcc then it's worth it.
But otherwise it brings no real advantage today and the longterm effects
are not predictable.
> With kind regards,
>
> Geert Uytterhoeven
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists