1 |
Dear Greg, |
2 |
|
3 |
Thanks a lot for such a comprehensive summary of present situation! You |
4 |
rock! |
5 |
|
6 |
I agree with all your points (excuse me for not quoting) except the |
7 |
multilib part. |
8 |
|
9 |
Greg Turner <gmt@×××××.us> writes: |
10 |
|
11 |
> [...] |
12 |
> |
13 |
> Unfortunately, also, at some point, this whole mess is going to start |
14 |
> bumping up against the limitations of prefix portage; namely, that it |
15 |
> has not yet been ported beyond the deprecated no-multilib profile. I |
16 |
> am expecting this to break PREFIX -> multilib cross emerges, at least |
17 |
> until the prefix and non-prefix versions of gcc-config are unified. |
18 |
> To fix that, we could either fix prefix portage to support multilib |
19 |
> (better for gentoo) or give the "ESUBROOT"s (made-up fake variable for |
20 |
> ${EROOT}/usr/${CTARGET}/) their own external ${PORTDIR}, containing |
21 |
> the non-prefix portage tree (easier to implement), so that |
22 |
> cross-emerge built against gx86 even on prefix-portage hosts. |
23 |
|
24 |
From the Prefix team: we are not planning to support multilib in Prefix |
25 |
(there was bugs on that, couldn't recall clearly though). Just use |
26 |
different instances of Prefixes for different arches, e.g., amd64 vs |
27 |
x86. |
28 |
|
29 |
From my own point of view, multilib is not cool compared to multiarch |
30 |
from Debian. lib vs lib32 vs lib64 is crazy to me with inconsistent |
31 |
implications across distros, even among people in the same |
32 |
community. LSB/FHS has not done a good job here. |
33 |
|
34 |
> [...] |
35 |
|
36 |
> I'd be curious to hear more about what your GSOC objectives are, |
37 |
> especially whether it's the "easy" or the "hard" fix, above, you are |
38 |
> shooting for. |
39 |
|
40 |
A long version of proposal is here: |
41 |
|
42 |
http://www.awa.tohoku.ac.jp/~benda/projects/android.html |
43 |
|
44 |
|
45 |
|
46 |
I'd like to start a piece of document to definitely describe what we |
47 |
expect for a portage/toolchain/crossdev combo to do in diferent |
48 |
situations, and what we interpret for different environment variables. I |
49 |
think Ruud already has his own version. |
50 |
|
51 |
Zac Medico and Mike Frysinger did masterful jobs for portage and |
52 |
toolchain/crossdev. As the two pieces are central to our distribution, |
53 |
they are not willing to accept random ideas. Let's make a proposal for |
54 |
cross-triplet and cross-eprefix in Gentoo in a systematic, robust and |
55 |
consent way before getting our hands dirty with the serious |
56 |
implementation. |
57 |
|
58 |
I'll roll out a draft and put it on a git repo/wiki/google drive for |
59 |
cooperative edition. |
60 |
|
61 |
Cheers, |
62 |
Benda |