| 1 |
On Wed, 2003-08-13 at 16:24, David Holm wrote: |
| 2 |
> If they really wanted to test the speed why didn't they use more aggressive compiler flags? |
| 3 |
> |
| 4 |
> I think it should be investigated which packages could be compiled by icc since |
| 5 |
> intel are now providing it for free for non-commercial use. I did some tests with it and |
| 6 |
> whetstone (classic fpu benchmark from the 1970's) doubled in speed compared to gcc on a P4, and |
| 7 |
> it was about 75% faster on an Athlon-XP. Now float-point isn't everything but from my experience |
| 8 |
> icc generally produces better optimized code than gcc unless the application has been hand-tuned |
| 9 |
> (like mplayer). |
| 10 |
> I tried installing gentoo with CC=icc once but I had problems with many ebuilds so I dropped |
| 11 |
> that idea. At the moment extremely few ebuilds support icc. |
| 12 |
|
| 13 |
|
| 14 |
Just realized that they are using march=pentium3, whereas celeron is a pentium2 (cf /etc/make.conf !). |
| 15 |
the use of march can really slow down the machine I think... |
| 16 |
|
| 17 |
|
| 18 |
-- |
| 19 |
gentoo-dev@g.o mailing list |
Archives