1 |
On Thu, 15 Sep 2011 16:33:48 -0400 |
2 |
Mike Frysinger <vapier@g.o> wrote: |
3 |
|
4 |
> On Thursday, September 15, 2011 16:12:00 Michał Górny wrote: |
5 |
> > On Thu, 15 Sep 2011 15:34:06 -0400 Mike Frysinger wrote: |
6 |
> > > KEYWORDS wise, i'd like to avoid having to add "x32" everywhere. |
7 |
> > > instead, reusing "amd64". only downside is the existing USE=amd64 |
8 |
> > > behavior, but we can address that by making MULTILIB_ABIS a |
9 |
> > > USE_EXPAND (i think this came up before with the portage multilib |
10 |
> > > discussion). |
11 |
> > |
12 |
> > Hrm, wouldn't that be more like x86 keyword? AFAICS the type sizes |
13 |
> > for x86 and x32 would match. |
14 |
> |
15 |
> the sizeof(long) and sizeof(void*) are the same between x86 and x32. |
16 |
> however, that's about the only thing. for example, x32 gets access |
17 |
> to 64bit registers when working with 64bit types (long long) and the |
18 |
> tuple is x86_64-pc-linux- gnu. in general, it seems to be closer to |
19 |
> amd64 than x32. -mike |
20 |
|
21 |
I'm rather thinking about potential issues. But OTOH packages fixed for |
22 |
amd64 should probably work with x32 as well. Excluding asm code which |
23 |
would probably need a third variant then. |
24 |
|
25 |
-- |
26 |
Best regards, |
27 |
Michał Górny |