| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA1 |
| 3 |
|
| 4 |
While fixing a bug it came to my attention that when an ebuild may wish to |
| 5 |
strip off CFLAGS known not to work, it will have to either strip off |
| 6 |
unnecessarily large amounts of optimisations or run the risk of being |
| 7 |
incompatible with different versions of gcc. |
| 8 |
|
| 9 |
For example, if removing -finline-functions from -O3, the optimal result with |
| 10 |
gcc 3 or greater is -O2 -frename-functions. However, gcc versions less than 3 |
| 11 |
have no -frename-functions and will hence fail. Either dropping all the way |
| 12 |
down to -O2 unnecessarily on some systems or some conditional statements are |
| 13 |
required. The latter is obviously desirable to maximise optimisations. |
| 14 |
|
| 15 |
Rather than have conditional statements for every possible connotation of gcc |
| 16 |
in the ebuild however, I have written this addition to the flag-o-matic |
| 17 |
eclass which basically queries gcc for each of the used optimisations and |
| 18 |
strips the unsupported ones. |
| 19 |
|
| 20 |
Attached is a patch for the flag-o-matic eclass. Any feedback or suggestions |
| 21 |
would be appreciated. |
| 22 |
|
| 23 |
Puggy |
| 24 |
-----BEGIN PGP SIGNATURE----- |
| 25 |
Version: GnuPG v1.2.2 (GNU/Linux) |
| 26 |
|
| 27 |
iD8DBQE/L9RwXYnvgFdTojMRAuruAKCfyBswJqQB50OTVRkLlWH1BMl5ugCfRe8W |
| 28 |
kmUf7S1UuOTJ9yFkqCoMydg= |
| 29 |
=66Lv |
| 30 |
-----END PGP SIGNATURE----- |
Archives