1 |
Hrrmm... gnuconfig has alreaded added linux-android* upstream for |
2 |
their system types in config.sub but using 'android' as a descriptor |
3 |
for the libc really wouldn't make sense if it wasn't an Android |
4 |
system. |
5 |
|
6 |
It would probably be prudent to add linux-android* | linux-bionic* ... |
7 |
there are a few significant differences. I guess the main |
8 |
differentiators would be FHS (android prefixes everything with |
9 |
/system), support for /etc/passwd, /etc/group, /etc/resolv.conf, |
10 |
crypt(3), getpwnam(3), getgrnam(3), and essentially any other missing |
11 |
feature that people might want to add in the future. |
12 |
|
13 |
They also don't have an entry in config.guess for LIBC=android (or |
14 |
LIBC=bionic). Something like this would work. |
15 |
|
16 |
... |
17 |
# ifdef __BIONIC__ |
18 |
# ifdef __ANDROID__ |
19 |
LIBC=android |
20 |
# else |
21 |
LIBC=bionic |
22 |
# endif |
23 |
# else |
24 |
LIBC=gnu |
25 |
# endif |
26 |
... |
27 |
|
28 |
I'll add a gnuconfig revision with those changes in my overlay. |
29 |
|
30 |
To me, it just makes sense to differentiate this way... of should |
31 |
still be possible to build an Android toolchain, so I'll add |
32 |
IUSE=android to the bionic ebuild. The bionic ebuild is otherwise |
33 |
done, but I still have to add something for |
34 |
crosscompile_opts_headers-only before it works with crossdev. |
35 |
|
36 |
Cheers, |
37 |
|
38 |
C |