1 |
Mike Edenfield wrote: |
2 |
> More importantly, -O2 seems to be the "typical" optimization |
3 |
> setting, and almost all free software packages are built and |
4 |
> tested and generally "supported", for whatever that means in an |
5 |
> open-source world, under -O2. If you report a bug in a package |
6 |
> and you use -Os, the first thing the devs will ask is "recompile |
7 |
> it using normal CFLAGS and try again." |
8 |
|
9 |
Although I agree with your reasoning above, you are contradicting |
10 |
yourself in the following two statements: |
11 |
|
12 |
> At least, it's no more broken under -Os than under -O2. |
13 |
> [...] benefits of using -Os over -O2 are minimal |
14 |
> compared against the possible problems it might cause. |
15 |
|
16 |
If -Os is no more broken than -O2, then it shouldn't cause any extra |
17 |
problems. :) |
18 |
|
19 |
> But given that disk space is dirt cheap |
20 |
|
21 |
It's not about disk space, it's about the amount that needs to be |
22 |
loaded from disk upon first run. |
23 |
|
24 |
> and modern OS |
25 |
> don't need to read an entire binary into memory to execute it, |
26 |
|
27 |
But if the entire binary is larger, each coherent subsection will be |
28 |
larger too, so more will have to be loaded with -O2 than with -Os. |
29 |
Processors are fast enough and getting faster all the time, it is |
30 |
only those disks that don't get any quicker -- not until we drop |
31 |
all those spinning platters and go solid state. |
32 |
|
33 |
Benno |
34 |
-- |
35 |
gentoo-user@g.o mailing list |