On Sat, Nov 04, 2006 at 09:55:12PM -0800, Jonathan Davis wrote:
> I understand why -m32 and -m64 are used while building the temporary
> system. However, once we've chrooted in (or booted), is the -m64
> really ever needed?
>
> I'm not really concerned so much with CLFS as BCLFS. For the most
> part, I just copy and paste the commands, so if it has ${BUILD64} in
> the book, then I'll use it. But I don't see much point in using it in
> BCLFS. Shouldn't gcc automatically be using the -m64 settings on a
> 64-bit machine (pure or multilib)?
>
Yes. You should only need to specify this if you are on multilib
AND building for the non-default size. On clfs, you only need -m32
( or -mabi=32 and -mabi=n32 for mips64, I suppose ).
On the other hand, I've got a lot of references to BUILD64 in my
own multilib scripts. Every time I build something in both sizes, I
mostly start with the 32-bit version, then copy the instructions when
I come to 64-bit - if I did it properly, I've got lots of references
to '32' in the stamp name, log name, messages of what it is building
or where it failed, and gcc/g++, so I just change these to 64 before
adding libdir and removing any overrides for libtool.
For pure64, BUILD64 doesn't help and it would mean the script would
not work on 32-bit systems.
Ken
--
das eine Mal als Tragödie, das andere Mal als Farce
_______________________________________________
Clfs-support mailing list
[email protected]
http://lists.cross-lfs.org/cgi-bin/mailman/listinfo/clfs-support
