[libre-riscv-dev] [isa-dev] Re: FP transcendentals (trigonometry, root/exp/log) proposal

[Jacob Lifshay]

On Wed, Aug 14, 2019, 06:56 lkcl <luke.leighton at gmail.com> wrote:

>
>
> On Monday, August 12, 2019 at 6:52:16 PM UTC+1, MitchAlsup wrote:
>>
>>
>>
>> On Sunday, August 11, 2019 at 10:20:28 PM UTC-5, lkcl wrote:
>>>
>>> https://libre-riscv.org/ztrans_proposal/#khronos_equiv
>>>
>>
>> I would like to point out that the general implementations of ATAN2 do a
>> bunch of special case checks and then simply call ATAN.
>>
>
> if there's no objections, i'll put ATAN back in [where it was formerly an
> alias ATAN(x)=ATAN2(1.0, x)].  that will require implementors to microcode
> the algorithm that Mitch described, and allow them to keep both 3D and UNIX
> Profiles optimal without compromise of either profile.
>

if atan is going back in, it would be a good idea to include atanpi.

We may want to put frecip back in as well, for those who implement division
as multiply by reciprocal.

Jacob