[BENCHMARK] Povray official benchmark scores for 387 and SSE insn sets

[Paolo Bonzini]

> Definitely, that's IMHO the way to go.
> I already suggested this on gcc-patches in November, just did not get to
> code that up yet (though, it wouldn't be appropriate for 4.0 anyway).

I'm working on this right now.  Given it only applies to -mfpmath=sse, 
it may still be appropriate for 4.0.

Paolo