This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Re: [mips] ABI shouldn't depend on ISA
Alexandre Oliva <aoliva at redhat dot com> writes:
> On Mar 25, 2003, Andreas Jaeger <aj at suse dot de> wrote:
>
>>> +#define _MIPS_ISA_MIPS32 6 /* Defined for code that uses o32. */
>>> +#define _MIPS_ISA_MIPS64 7 /* Defined for code that uses n64. */
>
>> Or are my comments wrong? Then you know why they're needed ;-)
>
> They're wrong. MIPS32 and MIPS64 are ISA names just like MIPS1,
> MIPS2, MIPS3, MIPS4, MIPS5 (and the most recent MIPS32R2). They're
> what you get when you specify -mips1, -mips2, -mips3, -mips4, -mips5,
> -mips32 and -mips64 in the GCC command line. Their definitions are
> absolutely similar to those of earlier ISAs and, frankly, they should
> be meaningful to anyone familiar with the various revisions of the
> MIPS ISAs. There's no point in adding comments to these two, unless
> we're adding comments to the others explaining that MIPS1 stands for
> MIPS I ISA, MIPS2 stands for MIPS II ISA, etc. A single comment at
> the top might do it, but then, I still find it would be pointless:
> _MIPS_ISA already implies the meaning. It takes some imagination to
> try to find a different meaning for them :-)
Ok, then leave it as is and commit the patch.
Andreas
--
Andreas Jaeger
SuSE Labs aj at suse dot de
private aj at arthur dot inka dot de
http://www.suse.de/~aj