| 1 |
On 11/14/2014 06:58 AM, Micha³ Górny wrote: |
| 2 |
> Dnia 2014-11-14, o godz. 09:08:17 |
| 3 |
> Michael Orlitzky <mjo@g.o> napisa³(a): |
| 4 |
> |
| 5 |
>> On 11/13/2014 10:17 AM, Ian Stakenvicius wrote: |
| 6 |
>> |
| 7 |
>>>> Isn't it possible to disable C++ in GCC with USE="-cxx"? |
| 8 |
>>> |
| 9 |
>>> It is.. but unfortunately there's no way in DEPEND to ensure it's |
| 10 |
>>> satisfied, as you can have a gcc installed with that flag enabled but |
| 11 |
>>> have a second one (that's actually selected in gcc-config) with it |
| 12 |
>>> disabled. A pkg_pretend check or a pkg_setup check (if you don't want |
| 13 |
>>> it to just fail in src_configure) is probably the best way to enforce |
| 14 |
>>> that one at this time. Unless there are other ways I'm not aware of?? |
| 15 |
>> |
| 16 |
>> Is this a case (as was recently suggested) where we're doing something |
| 17 |
>> stupid rather than asking for help from the PMS? This problem shows up |
| 18 |
>> in a few places -- off the top of my head: |
| 19 |
>> |
| 20 |
>> * GCC (see sys-apps/systemd-217.ebuild) |
| 21 |
>> * PHP (see comment in app-text/XML-Schema-learner-1.0.0.ebuild) |
| 22 |
>> * Python (all over the place) |
| 23 |
>> * Ruby (all over the place) |
| 24 |
>> |
| 25 |
>> Since all of the above are slotted, we can DEPEND on them, but we can't |
| 26 |
>> actually be sure that we're using the right slot at build time. The |
| 27 |
>> package manager knows that the right version is there, but it's not at |
| 28 |
>> the moment prepared to find and use it. |
| 29 |
>> |
| 30 |
>> Question 1: is it desirable to e.g. switch compilers, compile systemd, |
| 31 |
>> and then switch back? At first I thought the PM should respect my |
| 32 |
>> selected compiler, but after thinking about it for a few minutes, I've |
| 33 |
>> changed my mind. The compiler deps are just like anything else: if I ask |
| 34 |
>> portage to install systemd, it should do what it takes to install |
| 35 |
>> systemd assuming I approve the build plan. |
| 36 |
> |
| 37 |
> Relying on stuff that can be switched outside the PM scope is always |
| 38 |
> a bad idea. Think of eselect-opengl -- for a long time, people had to |
| 39 |
> switch the OpenGL implementation to xorg to build xorg server, and then |
| 40 |
> back to be able to use anything using OpenGL. Then we fixed xorg-server |
| 41 |
> to use the underlying headers directly. |
| 42 |
|
| 43 |
Yeah, having ebuilds switch the global compiler configuration would be |
| 44 |
completely insane. If they need to select a different local compiler via |
| 45 |
a mechanism like CC and CXX, as you suggest below, then that's fine. |
| 46 |
|
| 47 |
> I think we should use the same solution for the gcc issues. We already |
| 48 |
> expect packages to respect CC & CXX -- why not set them then? I'm |
| 49 |
> keeping my CC & CXX at a specific gcc version for a long time now. |
| 50 |
> |
| 51 |
|
| 52 |
|
| 53 |
-- |
| 54 |
Thanks, |
| 55 |
Zac |
Archives