Re: [Sbcl-devel] Patch: Smaller code size for some float comparisons on x86-64

[Paul K. <pv...@pv...>]

On 2011-08-29, at 5:04 PM, Lutz Euler wrote:
> in the tradition of trying to squeeze out every last byte from the
> machine code generated by the compiler, I would like to propose a
> patch that does this for some float comparisons for equality. It simply
> uses narrower registers to spare the REX prefix and/or be able to use
> a shorter instruction encoding.

Committed, with a slight aesthetic change.

Thanks,

Paul Khuong