>>>>> "Rob" == rhealey  <rhealey@aggregate.com> writes:

Niklas>  NO!  We don't want separate make binaries for cross-compilation.
Niklas> That would make writing multitarget makefiles very hard.  The
Niklas> predefined defines are target attributes and thus belongs in the
Niklas> compiler.

Rob> 	Hmmm, I wasn't aware the primary goal of NetBSD was to make
Rob> it's tools taylored to a cross compilation environment...

Of course it's not, but then again, why make it harder than needed.

Rob> 	By the way, it's too late to change this behaviour in NetBSD
Rob> make, it's been in there for 2 months or so... It's only because
Rob> the non-supp'd Amiga trees are so far behind sun-lamp that it
Rob> hasn't been noticed till now.  I believe you can redefine MACHINE
Rob> and MACHINE_ARCH in the makefile if you wish, or override them
Rob> with environmental/command-line variables.

I did not argue against the MACHINE(_ARCH)? make variables, they are
nice to have in order to specify the host environment.  Such attributes
are interesting as well.  Now, I don't really care too much about the
BSD make, I just use it for BSD sources.  Otherwise I use GNU make.
I'm only argueing for the compiler defining __amiga__, instead of
trusting the environment to supply the define.  Well, I can live with
it either way, just wanted to state my opinion.

Rob> The compiler should defined 3 things:

Rob> 	1) __NetBSD__ <- OS
Rob>    2) __m68k__ <- Architecture
Rob>    3) __amiga__ <- specific machine {__mac__,__sun3__, etc}

Exactly my opinion!  As long as these reflect the target, not the host.



------------------------------------------------------------------------------