1 |
On 09-11-2012 09:32:47 -0600, William Hubbs wrote: |
2 |
> > Testing it. Testing the /usr move broadly. |
3 |
> > You want gen_usr_ldscript on non-bootable systems (eg prefix, mingw) to |
4 |
> > be (mostly) a noop too which you can't distinguish from only CHOST. |
5 |
> |
6 |
> That has been done already by toolchain a while back. Take a look at the |
7 |
> code in toolchain-funcs.eclass. There are very few platforms where this |
8 |
> function does anything at all these days. It would just be a matter of |
9 |
> removing linux from the platforms it supports. |
10 |
|
11 |
They tested it? Are you referring to the platforms from bug #417451? |
12 |
(excluding the Prefix platforms, because in the Prefix tree |
13 |
gen_usr_ldscript just works, because it breaks existing installs to |
14 |
disable it) |
15 |
|
16 |
> > IMHO this needs a discussion on -dev before going through the council. |
17 |
> |
18 |
> I would send the patch to do this to -dev anyway, so at that time I |
19 |
> guess we could have folks apply it and rebuild their systems and see |
20 |
> what breaks. |
21 |
> |
22 |
> Since applying the patch itself will not force any rebuilds, it should |
23 |
> just end up being a natural migration; as things are rebuilt the |
24 |
> libraries will move from /lib to /usr/lib with no harm being done. |
25 |
|
26 |
I thought so too, until I got dangling symlinks for older zlib for some |
27 |
reason (preserve-libs? ebuild?) which wasn't really a funny experience. |
28 |
So, before this is done, we must test it in all forms, not rely on theory. |
29 |
|
30 |
Alexis' profile suggestion nicely allows existing installs to remain as |
31 |
is, new installs to be without /usr-split, and people to move over when |
32 |
they deem that a good idea. |
33 |
|
34 |
|
35 |
-- |
36 |
Fabian Groffen |
37 |
Gentoo on a different level |