1 |
Dice R. Random wrote: |
2 |
> |
3 |
> It is not necessary to tell gcc to use special instruction sets for |
4 |
> your processor. When you specify -march gcc automatically knows |
5 |
> exactly what instructions your processor is capable of utilizing most |
6 |
> effectively. |
7 |
I'm not sure about this. Tonight I'll make a test: I'll compile a |
8 |
package with and without these flags and watch the output and the outcome. |
9 |
|
10 |
> |
11 |
> I am not familiar with PIC, but if portage gives a big red warning |
12 |
> about explicitly enabling it in your CFLAGS I'm sure that there have |
13 |
> been problems with it. If a package requires the flag then it should |
14 |
> enable it in its own local CFLAGS. |
15 |
> |
16 |
Already answered about Position Independent Code and why I use it. |
17 |
Shortly: http://www.gentoo.org/doc/en/prelink-howto.xml |
18 |
> A 3dnow instruction path isn't always the best optimization for every |
19 |
> user and every package. Also, AMD's Athlon and Opteron processors are |
20 |
> not the only ones supported under the amd64 arch, there are also |
21 |
> Intel's x86-64 processors to consider. While I imagine that Intel |
22 |
> processors can execute 3dnow instructions I also imagine that SSE |
23 |
> would be faster. |
24 |
I'm really not sure here. I cant tell which is better -3dnow(ext) or |
25 |
sse{1..3}, but isn't it best if one enables all kinds of optimizations |
26 |
supported by the CPU? |
27 |
|
28 |
-- |
29 |
gentoo-amd64@g.o mailing list |