[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-gnubg] Benchmarks, experiments, speedups
From: |
Øystein Johansen |
Subject: |
Re: [Bug-gnubg] Benchmarks, experiments, speedups |
Date: |
Tue, 15 Mar 2005 19:28:55 +0100 |
User-agent: |
Mozilla Thunderbird 0.8 (Windows/20040913) |
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
macherius wrote:
| Hi,
|
| one more on gcc. I've just compiled myself a gcc 4.0.0 from the
| snapshot sources as of 20050220. There is a new "-ftree-vectorize"
| method that mimics Intel's vectorizer very closely. Attached is code
| produced by gcc 4.0.0 snapshot, you'll see the BIG difference to
| 4.3.x code.
I got a working snapshot of gcc-4.1 today. was tweaking the optimization
parameters a bit and the ftree-vectorizer option gave me about 5%
performance improvment. I still don't see the speed from your Inten build.
| The binary produced by the gcc 4.0.0 snapshot dumped core, but at
| least things are moving in gcc-land. If the code generator is stable,
| the binaries of gnubg produced by gcc 4.0 should be pretty fast.
Hopefully I have done something wrong. Compared to my previous builds
with gcc-3.4.2 I used to have ~21000 evaluations/second and with gcc-4.1
snapshot I get ~22000 evals/sec.
| So intrinsics would be a quick solution for speed improvement, but
| maybe the GTK 2.6 update is more pressing?
I don't feel anything is pressing.... ;-)
"intrinsics" is my vote, but I don't feel I have the knowledge to
implement that. Anyone?
- -Øystein
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFCNyln6kDTFPhwyqYRAhlUAJ4hSBWM/+oWcH8383xB7tZlfUijigCfdsSD
j2l3zmaTTXTThWw77eaPQ3E=
=gLx0
-----END PGP SIGNATURE-----
- Re: [Bug-gnubg] Benchmarks, experiments, speedups,
Øystein Johansen <=