RE: [Patch, MIPS] Update MIPS memcpy.S for mips32r6/mips64r6 support

[Matthew Fortune <Matthew dot Fortune at imgtec dot com>]

Hi Steve,

Given Corinna's request for MIPS reviewers with my patches I
expect we should co-review things for MIPS. I've been
through and read the patch. I don't have any objection to
this combined indentation and R6 patch especially given it
is pretty much just making newlib match glibc.

It all looks good (the logic was checked in some detail for
glibc so I haven't been through it again).

Minor issues:

> +L(jtable):
> +        bc      L(aligned)
> +        bc      L(r6_unaligned1)
> +        bc      L(r6_unaligned2)
> +        bc      L(r6_unaligned3)
> +# ifdef USE_DOUBLE
> +        bc      L(r6_unaligned4)
> +        bc      L(r6_unaligned5)
> +        bc      L(r6_unaligned6)
> +        bc      L(r6_unaligned7)

Nit: indentation

> +#ifdef R6_CODE
> +	PREFETCH_FOR_STORE (2, a0)
> +#else
>  	PREFETCH_FOR_STORE (4, a0)
>  	PREFETCH_FOR_STORE (5, a0)
> +#endif

Nit: the order of these is reversed vs other tests on R6_CODE.
I.e. there is at least one #ifndef R6_CODE #else #endif. It
seems the '#ifndef R6_CODE' is the common form in the patch
so perhaps that should be used unless guarding just some R6
code with no R<6 alternative. There are a couple more of these.

Thanks,
Matthew