| 1 |
Michał Górny schrieb: |
| 2 |
> Hello, |
| 3 |
> |
| 4 |
> Currently, the multilib-build eclass uses abi_* constants only for USE |
| 5 |
> flags and only ${ABI} is exported to the function. This is bad since it |
| 6 |
> basically requires a reverse mapping of ABI->abi_* values, often |
| 7 |
> inlined as ${ABI} checks. |
| 8 |
> |
| 9 |
> The patches which I will send in reply to this thread aim to fix it. |
| 10 |
> |
| 11 |
> The first patch changes the eclass logic. The abi_* values, with 'abi_' |
| 12 |
> prefix stripped, are called MULTILIB_ABI now. They are used to run |
| 13 |
> the 'foreach' functions, and now are set in the called functions along |
| 14 |
> with ABI. |
| 15 |
> |
| 16 |
> As a downside, the switch required the MULTILIB_ABI -> ABI mapping to |
| 17 |
> occur inside foreach -- as in, another 'for' loop. It shouldn't cause |
| 18 |
> any noticeable difference. |
| 19 |
> |
| 20 |
> Additionally, the 'default' fallback no longer calls |
| 21 |
> multilib_toolchain_setup. This should improve compatibility with |
| 22 |
> multilib-portage and *maybe* cross-compiling. |
| 23 |
|
| 24 |
You know, that multilib-portage does use MULTILIB_ABI as USE-expanded |
| 25 |
variable? Using exactly the same in the eclass will call for collision |
| 26 |
issues. |
| 27 |
|
| 28 |
|
| 29 |
-- |
| 30 |
|
| 31 |
Thomas Sachau |
| 32 |
Gentoo Linux Developer |
Archives