On Wed, 25 Feb 1998 00:50:30 -0500 (EST)
"Charles M. Hannum" <mycroft%mit.edu@localhost> wrote:
> Actually, at least all of:
>
> ansi.h aout_machdep.h asm.h cdefs.h cpu.h db_machdep.h endian.h float.h
> frame.h ieeefp.h kcore.h limits.h param.h pcb.h profile.h psl.h ptrace.h
> reg.h setjmp.h signal.h stdarg.h trap.h types.h varargs.h
>
> should just include the version for arch/powerpc/include. Probably
> others as well.
The BeBox should probably do the same thing, as well...
Which brings up an interesting point that I've been concerned about
for a while...
In m68k-land, we have arch/m68k which has some common code shared by all
m68k platforms (there could be a lot more, but that's not really the
point of this message :-)
We sort of have that in PowerPC-land - there is arch/powerpc, but that
also happens to be the name of a full-fledged (OFW-devices only) port.
What sort of interactions are there associated with doing:
machine powermac powerpc
...I belive that causes config(8) to bring in both:
arch/powermac/conf/files.powermac
and
arch/powerpc/conf/files.powerpc
...which would be wrong in that some unnecessary files would be included.
I wanted to solve this by creating sys/cpu/{m68k,powerpc}, but there were
some interesting issues revolving around exporting header files, etc.
Anyhow, in PowerPC-land, most of locore.S, all of pmap.c, vm_machdep.c,
most of machdep.c, etc.
Any thoughts on this?
Jason R. Thorpe
thorpej%nas.nasa.gov@localhost
NASA Ames Research Center Home: +1 408 866 1912
NAS: M/S 258-5 Work: +1 650 604 0935
Moffett Field, CA 94035 Pager: +1 415 428 6939