RFC: mips target support for Juniper Apollo

[cgd@broadcom.com]

At Wed, 10 Mar 2004 07:59:06 +0000 (UTC), "Jim Wilson" wrote:
> It is also missing branch delay slots.  This is a serious
> incompatibility which requires having separate magic numbers for Apollo,
> as it won't be compatible with anything else.
> 
> I'm concerned about the choice of the magic numbers.  I don't want to
> conflict with number that Red Hat is using, but only a Red Hat employee
> can check that.  This primarily means E_MIPS_MACH_APOLLO in the
> include/elf/mips.h file.

Uh, except that the magic numbers E_MIPS_MACH_* are not actually
*used* for anything on most/all MIPS ELF systems, since they are not
actually standard.  I.e., You've solved this problem from a toolchain
perspective, but not an OS perspective.

>From an OS perspective, it either needs a different E_MIPS_ARCH value
(which i don't think is appropriate, since it's not actually
MIPS-comptable... at least for code which uses branches and jumps 8-),
or a different EM_* (e_machine) value.

It *really* *really* *really* should have a different E_MACHINE value,
since it is *fundamentally* incompatible with MIPS.


(Whether or not it should ever call itself "mips" in binutils is
another question, and one that could get fairly ugly.)



cgd