1 |
On 07/02/2017 16:01, Walter Dnes wrote: |
2 |
> On Mon, Feb 06, 2017 at 10:32:16PM -0800, Ian Zimmerman wrote |
3 |
> |
4 |
>> It depends on how good the Pale build system is. What you would be |
5 |
>> doing is similar to cross-compiling, and you need the build system be |
6 |
>> ready for that kind of thing. |
7 |
> |
8 |
> [...deletia...] |
9 |
> |
10 |
>> Unfortunately I'm not intimate with the Mozilla/Pale build system so I |
11 |
>> can't give you yea or nay, but maybe you can search for information |
12 |
>> whether it can be (really) cross-compiled. If yes, then your scenario |
13 |
>> should work too. |
14 |
> |
15 |
> https://developer.mozilla.org/en-US/docs/Mozilla/Developer_guide/Build_Instructions/Cross-compiling_Mozilla |
16 |
> says it can be cross-compiled. Actually, what I'm trying to do is to |
17 |
> *NOT* make it look like a cross compile. That's why I want to rsync "/" |
18 |
> (all of /dev/sda1) from the VM to a directory on the host. This would |
19 |
> give an entire 32-bit CentOS install, including toolchain, system libs, |
20 |
> etc, without splattering 32-bit libs all over my host system. It would |
21 |
> all be contained in the chroot directory. And, oh yeah, I run the |
22 |
> "default/linux/amd64/13.0/no-multilib" profile. |
23 |
> |
24 |
|
25 |
|
26 |
Why not just do it and see what happens? |
27 |
|
28 |
You won't break anything on the host, the VM stays as it was and if the |
29 |
chroot breaks - well it was a copy that you can recreate with ease |
30 |
|
31 |
|
32 |
-- |
33 |
Alan McKinnon |
34 |
alan.mckinnon@×××××.com |