This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Haney's real matrix test regression
- To: Richard Kenner <kenner at vlsi1 dot ultra dot nyu dot edu>
- Subject: Re: Haney's real matrix test regression
- From: Paolo Carlini <pcarlini at unitus dot it>
- Date: Thu, 25 Oct 2001 19:45:08 +0200
- CC: gcc at gcc dot gnu dot org
- Organization: Universita' della Tuscia
- References: <10110251737.AA14862@vlsi1.ultra.nyu.edu>
- Reply-To: pcarlini at unitus dot it
Richard Kenner wrote:
> Suggestions?
>
> Look at the code and see why it's slower.
Ok. Give me a few hours and I will post a comparison between the assembly
produced by 3.0.2 and today's 3.1.
On my system, 3.0.2 has the very same figures of 3.1 20011023 but if you think
that it may be useful, I can also post the assembly produced by 3.1 20011023,
just teach me how to use CVS to recostruct the old tree :-)
Anyway, gcc3.0.2:
Timing Real matrix operations for n = 200
1...2...
Language Answer Check Time Iter Time
O-O C++ (2.41602e+06): 13.95 0.1395
Hand-coded C (2.41602e+06): 14.89 0.1489
Cheers,
Paolo Carlini.