Please test snapshots of the coming R9C

[Stefan Nickl]

On 17 Jul 2003, Kent Boortz wrote:

> snickl@REDACTED writes:
> > The problem seems to be that the part of the build that makes the
> > directories "believes" the --host option, while the part that
> > generates the makefiles goes with autodetection.
>
> Haven't tried it but this may work
>
>   % ./configure --host=i386-pc-linux-gnu
>   % make TARGET=i386-pc-linux-gnu

This solutions works for me, thanks!

It also works when changing gentoo's erlang-ebuild from "make || die"
to "make TARGET=${CHOST} || die" (with CHOST set to i386 from make.conf)

> > This gets you trouble when you want to use binaries on different machines,
> > but compile (as always) on the fastest around.
>
> Are you sure -host will make a difference for vanilla "configure"
> scripts, i.e. that it will set -mARCH to gcc? I know it will if you
> build gcc or maybe recompile a kernel but I thought this was because
> they explicitely handled this in their "configure.in".

No, it doesn't get set.
I think youre on the wrong track here, my issue was about Makefiles
trying to run each other, but looking in the wrong place.
The process didn't even get far enough to run the compiler.


CU, SN

-- 
The UNIX equivalent of the "Blue Screen of Death" would be called
"kernel panic". It obviously exists, since I have heard and read about it,
but I've never been witness to it in my professional career.
        John Kirch,
        Networking Consultant and Microsoft Certified Professional (Windows NT)