| 1 |
On Monday 10 July 2006 15:20, Patrick McLean wrote: |
| 2 |
> Denis Dupeyron wrote: |
| 3 |
> > This, for me, triggers 3 questions that are gentoo-dev@ material : |
| 4 |
> > |
| 5 |
> > 1) Should all ebuilds that currently filter --fast-math die on its |
| 6 |
> > presence instead of filtering it ? |
| 7 |
> |
| 8 |
> I don't think we should die on anything, if a user wants a particular |
| 9 |
> CFLAG, generally the default should be to let them use it. A warning |
| 10 |
> with a pause may also suffice. |
| 11 |
|
| 12 |
So what do we do for openoffice, which breaks when python was compiled with |
| 13 |
fast-math (probably breaking python's math functions). Do we really want to |
| 14 |
add filters for stupid flags? For me the choise has only two viable options. |
| 15 |
1) Completely ignoring the flag; 2) Dieing when it occurs. |
| 16 |
|
| 17 |
I'm affraid that the usage of -ffast-math is that common that option 1 creates |
| 18 |
too many false bug reports. As such we choose to die on it to protect |
| 19 |
ourselves. It prevents us from diagnosing PEBKAC bugs manually. |
| 20 |
|
| 21 |
Paul |
| 22 |
|
| 23 |
-- |
| 24 |
Paul de Vrieze |
| 25 |
Gentoo Developer |
| 26 |
Mail: pauldv@g.o |
| 27 |
Homepage: http://www.devrieze.net |
Archives