| 1 |
On Friday 12 November 2004 00:58, Jörg Schaible wrote: |
| 2 |
> Seesh. You're right. I was talking about CFLAGS. Now, is there a policy |
| 3 |
> for "standard accepted safe CFLAGS" ? I know, it is my problem, if |
| 4 |
> something really breaks. But I have really a lot of apps running with |
| 5 |
> my (not too esoteric) settings, and if one single app fails badly (just |
| 6 |
> core dumps) and I can track it down to a single CFLAG flag, what is the |
| 7 |
> actual policy (if there is one)? Why is a filter for the flag just |
| 8 |
> recected? |
| 9 |
|
| 10 |
One thing that I use as baseline is that if "info gcc" says that it is not |
| 11 |
safe (such as -ffast-math, -fno-rtti, -fnoexceptions and various others). |
| 12 |
Many of these optimization options are application specific and should be |
| 13 |
set by the upstream build scripts. In other words, filtering away to |
| 14 |
avoid compiler bugs is something I'm willing to do, filtering away stupid |
| 15 |
defaults is not. |
| 16 |
|
| 17 |
Paul |
| 18 |
|
| 19 |
-- |
| 20 |
Paul de Vrieze |
| 21 |
Gentoo Developer |
| 22 |
Mail: pauldv@g.o |
| 23 |
Homepage: http://www.devrieze.net |
Archives