This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: Minor reload cleanup/speedup

From: Jan Hubicka <hubicka at ucw dot cz>
To: Richard Henderson <rth at redhat dot com>, Jan Hubicka <jh at suse dot cz>,Paolo Carlini <pcarlini at suse dot de>, gcc-patches at gcc dot gnu dot org
Date: Wed, 4 Feb 2004 23:29:13 +0100
Subject: Re: Minor reload cleanup/speedup
References: <20040204145630.GE21148@kam.mff.cuni.cz> <40210DD7.2020800@suse.de> <20040204160226.GF21148@kam.mff.cuni.cz> <20040204195849.GA22379@redhat.com>

> On Wed, Feb 04, 2004 at 05:02:26PM +0100, Jan Hubicka wrote:
> > Actually the code is faster just because ISDIGIT is implemented in
> > locale's way ....
> 
> No, it absolutely is not.  That's why we use ISDIGIT from libiberty's
> safe-ctype.h instead of isdigit from the system's ctype.h.
Hmm, so then it must be faster just because switch statement expansion
does better job than sequence of jumps :)
All what I observed is large number of oprofile samples near ISDIGIT
test that went away after putting switch in place (and the amount of
samples for whole function decreased)

Honza
> 
> 
> r~

References:
- Minor reload cleanup/speedup
  - From: Jan Hubicka
- Re: Minor reload cleanup/speedup
  - From: Paolo Carlini
- Re: Minor reload cleanup/speedup
  - From: Jan Hubicka
- Re: Minor reload cleanup/speedup
  - From: Richard Henderson

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]