| 1 |
Pacho Ramos schrieb: |
| 2 |
> El dom, 23-09-2012 a las 11:56 +0200, Michał Górny escribió: |
| 3 |
>> On Sun, 23 Sep 2012 11:07:30 +0200 |
| 4 |
>> Thomas Sachau <tommy@g.o> wrote: |
| 5 |
>> |
| 6 |
>>> Matt Turner schrieb: |
| 7 |
>>>> On Sat, Sep 22, 2012 at 2:24 PM, Michał Górny <mgorny@g.o> wrote: |
| 8 |
>>>>> It is a simple eclass using autotools out-of-source builds to build |
| 9 |
>>>>> packages for multiple ABIs when multilib is supported. |
| 10 |
>>>> |
| 11 |
>>>> Thanks a lot, Michał! This looks good to me. |
| 12 |
>>>> |
| 13 |
>>>>> Use case: xorg packages, ask Matt. |
| 14 |
>>>> |
| 15 |
>>>> So the idea is that users want up-to-date 32-bit drivers for games and |
| 16 |
>>>> WINE. The emul- packages aren't a very good solution for a number of |
| 17 |
>>>> reasons. |
| 18 |
>>>> |
| 19 |
>>>> I'd like to add multilib USE flags to Mesa and thus its dependencies. |
| 20 |
>>>> I realized that almost everything in x11-libs/ could be converted very |
| 21 |
>>>> easily, which would allow us to get rid of emul-linux-x86-xlibs in |
| 22 |
>>>> addition to emul-linux-x86-opengl. |
| 23 |
>>>> |
| 24 |
>>>> |
| 25 |
>>> |
| 26 |
>>> This looks like a shortened duplication of a subset of multilib-portage |
| 27 |
>>> features. While this wont hurt multilib-portage (since it does exclude |
| 28 |
>>> most actions on ebuilds with USE=multilib), it will mean a rewrite for |
| 29 |
>>> many ebuilds, which then again need another rewrite (or more likely |
| 30 |
>>> revert), when multilib-portage is accepted in a future EAPI. |
| 31 |
>> |
| 32 |
>> s/when/if/ |
| 33 |
>> |
| 34 |
>>> So i would prefer some help/support with multilib-portage to get it |
| 35 |
>>> accepted sooner, instead of this additional workaround for a subset of |
| 36 |
>>> packages. |
| 37 |
>> |
| 38 |
>> I prefer the simpler solution. |
| 39 |
>> |
| 40 |
>>> P.S.: I know, that users, who want up-to-date 32bit drivers for games |
| 41 |
>>> and wine do use multilib-portage, so we already have a working solution |
| 42 |
>>> for this issue. |
| 43 |
>> |
| 44 |
>> They will no longer have to do that. |
| 45 |
>> |
| 46 |
> |
| 47 |
> I would prefer if eclass way could be extended to packages not using |
| 48 |
> autotools too, otherwise, we will still need emul packages for, for |
| 49 |
> example, qt libs. If that would be possible via eclass, maybe |
| 50 |
> multilib-portage wouldn't be needed but, if not, we will still need it |
| 51 |
> and, then, would be nice if this inclussion for autotools packages |
| 52 |
> wouldn't cause more problems to get the "strong" solution land in the |
| 53 |
> "near" future :/ |
| 54 |
> |
| 55 |
> The simpler solution (eclass) looks fine to me, but it would need to be |
| 56 |
> extended to more packages than autotools based ones to let it replace |
| 57 |
> portage-multilib/emul packages |
| 58 |
> |
| 59 |
|
| 60 |
you mean something like this one? |
| 61 |
|
| 62 |
https://github.com/sjnewbury/multilib-overlay/blob/80c9fd20cfd05481ac19edcadd56ad5e578a8930/eclass/multilib-native.eclass |
| 63 |
|
| 64 |
That was the original eclass allowing cross-compile support, but |
| 65 |
required ebuilds to inherit it. multilib-portage is based on this, but |
| 66 |
does not require to modify the ebuilds themselves. |
| 67 |
|
| 68 |
-- |
| 69 |
|
| 70 |
Thomas Sachau |
| 71 |
Gentoo Linux Developer |
Archives