Buildroot compile problem on x86_64

[Rob Landley]

On Tuesday 06 December 2005 22:15, Mike Frysinger wrote:
> On Tue, Dec 06, 2005 at 09:42:40PM -0600, Rob Landley wrote:
> > On Monday 05 December 2005 18:51, Mike Frysinger wrote:
> > > On Tue, Dec 06, 2005 at 12:20:10AM +0100, Kaspar Schleiser wrote:
> > > > I tried compiling buildroot from svn with uclibc latest snapshot.
> > >
> > > x86_64 isnt usuable atm, so unless you plan on helping develop the
> > > x86_64/uClibc port, just use x86/uClibc
> >
> > What's missing?
>
> lazy relocation (although jocke sent me a patch that should fix it) and
> in general, just testing the port to make sure everything works
> -mike

Testing I can do.  Once I get my firmware 0.9 release out (which is now two 
days overdue, and in an hour will be three), I'll be trying to get it to work 
on x86-64.

(I spent a week and change bothering Jeff Dike to get User Mode Linux to work 
on x86-64 already.  Resulted in four separate patches to UML before it all 
worked.  Prepare to be bothered.)

Let me know when current -svn of uClibc is in a "nothing actually _known_ to 
be wrong with it" state, and I'll thump.  But not before saturday...

P.S.  Anybody know where I can get root filesystem images for other 
architectures with a working copy of gcc on them?  I want to play with qemu's 
support for arm, ppc, x86-64, and perhaps mips.  (Screw sparc, I don't care, 
and don't know anybody who does.)

I have a to-do item to figure out how to make cross-compiling work, but it's 
pretty far down on the list...

Rob
-- 
Steve Ballmer: Innovation!  Inigo Montoya: You keep using that word.
I do not think it means what you think it means.