1 |
В письме от 23 января 2013 08:03:56 пользователь Alexis Ballier написал: |
2 |
> On Wed, 23 Jan 2013 09:24:26 +0100 |
3 |
> |
4 |
> Michał Górny <mgorny@g.o> wrote: |
5 |
> > On Mon, 21 Jan 2013 10:27:30 -0300 |
6 |
> > |
7 |
> > Alexis Ballier <aballier@g.o> wrote: |
8 |
> > > > To be honest, I don't know if there's other way to hide USE flags |
9 |
> > > > than using USE_EXPAND_HIDDEN. If we want to use that, we'd have |
10 |
> > > > |
11 |
> > > > to split the flags per-arch, i.e. have: |
12 |
> > > > MULTILIB_AMD64="abi1 abi2 abi3" |
13 |
> > > > MULTILIB_PPC64="abi1 abi2 abi3" |
14 |
> > > > |
15 |
> > > > with appropriate USE_EXPAND_HIDDEN set by profiles. |
16 |
> > > |
17 |
> > > I don't like that at all. |
18 |
> > > I'd go for ABI= the union of all the MULTILIB_ABIS variables (if |
19 |
> > > there is no name collision) |
20 |
> > > we certainly want skype to depend on libitneeds[abi_x86], not |
21 |
> > > 'amd64? ( libitneeds[abi_amd64_x86] ) x86? ( libitneeds )' |
22 |
> > |
23 |
> > Just a quick idea. |
24 |
> > |
25 |
> > How would you feel about abi_x86_32? (similarly _64, _x32) |
26 |
> > |
27 |
> > That would be almost natural names with the trick variable being |
28 |
> > ABI_X86, therefore having all the fore-mentioned advantages. |
29 |
> > |
30 |
> > The deps would look like: |
31 |
> > libitneeds[abi_x86_32] |
32 |
> |
33 |
> Sounds good too, I just would want it to be shared between arches that |
34 |
> can: amd64/x86/x32, ppc/ppc64, mips, sparc32/sparc64, etc. |
35 |
> You mean you will hide ABI_X86 USE_EXPAND on non-x86 arches, right ? |
36 |
> This would have all the benefits I think, very good idea :) |
37 |
> |
38 |
> Alexis. |
39 |
|
40 |
Shared abi names are bad idea. For example |
41 |
mips abis : |
42 |
o32 |
43 |
n32 |
44 |
n64 |
45 |
eabi |
46 |
x86: |
47 |
x86_32 |
48 |
x86_x32 |
49 |
x86_64 |
50 |
|
51 |
Actualy first three one are equivalent in their internal behavior. But i dont |
52 |
think that its good idea to have one name for all. Think about multiarch |
53 |
installs where you can have binaries from different architectures in one |
54 |
system. |
55 |
-- |
56 |
Best Regards, |
57 |
Alexey 'Alexxy' Shvetsov |
58 |
Petersburg Nuclear Physics Institute, NRC Kurchatov Institute, |
59 |
Gatchina, Russia |
60 |
Department of Molecular and Radiation Biophysics |
61 |
Gentoo Team Ru |
62 |
Gentoo Linux Dev |
63 |
mailto:alexxyum@×××××.com |
64 |
mailto:alexxy@g.o |
65 |
mailto:alexxy@×××××××××××××.ru |