| 1 |
Richard Freeman <rich0@g.o> posted 4880F05D.4010905@g.o, |
| 2 |
excerpted below, on Fri, 18 Jul 2008 15:34:53 -0400: |
| 3 |
|
| 4 |
> Personally, I use -Os across the board when it doesn't break. As you |
| 5 |
> said I tend to be memory/IO bound, and optimizing for size helps with |
| 6 |
> both (swapping causes IO). I'd probably benefit from using -O3 on the |
| 7 |
> aforementioned CPU-intensive apps. |
| 8 |
|
| 9 |
I did before gcc-4.3, but with 4.3, several of the other flags (-freorder- |
| 10 |
blocks-and-partition, for instance, which increases cache hit rates but |
| 11 |
also increases size) I was using folded into -O2 but not -Os, and I |
| 12 |
decided -O2 was a better choice as a result. |
| 13 |
|
| 14 |
Has anyone done a study of -Os vs -O2 with gcc-4.3.x, similar to the ICC/ |
| 15 |
gcc-4.x study linked upthread? On x86_64 would be an extra bonus, |
| 16 |
particularly if it were on AMD, or if both Intel and AMD were studied, as |
| 17 |
would the effect of such flags as -freorder-blocks-and-partition, as |
| 18 |
mentioned above. I'd love to see links, if so! |
| 19 |
|
| 20 |
-- |
| 21 |
Duncan - List replies preferred. No HTML msgs. |
| 22 |
"Every nonfree program has a lord, a master -- |
| 23 |
and if you use the program, he is your master." Richard Stallman |
| 24 |
|
| 25 |
-- |
| 26 |
gentoo-dev@l.g.o mailing list |
Archives