Re: [PATCH v3 3/3] b43: Use cordic algorithm from kernel library

[Kalle Valo <kvalo@codeaurora.org>]

Priit Laes <plaes@plaes.org> writes:

> On Wed, Nov 14, 2018 at 07:46:28PM +0100, Michael Büsch wrote:
>> On Wed, 14 Nov 2018 20:27:52 +0200
>> Priit Laes <plaes@plaes.org> wrote:
>> 
>> > Kernel library has a common cordic algorithm which is identical
>> > to internally implementatd one, so use it and drop the duplicate
>> > implementation.
>> 
>> 
>> In v2 of the series it has been said that:
>> 
>> Arend van Spriel <arend.vanspriel@broadcom.com> wrote:
>> > I recall doing a comparison between the algorithms and thought I put 
>> > that in the original commit message. However, it is not there. It is not 
>> > exactly the same as in b43 so there are difference for certain angles, 
>> > most results are the same however. This implementation is slightly more 
>> > accurate on the full scale.
>> 
>> 
>> That's not my definition of "identical".
>> 
>> Please do not apply this patch without doing a thorough regression test
>> on actual b43 LP hardware.
>
> Indeed, there's a big discrepancy in the results of both algorithms.
>
> Here's the test script:
> https://gist.github.com/plaes/284993a4fc65e0926d0628a11f0cf874
>
> So at current state, this is self-NAK from me too and this patch should
> be dropped.

Ok, I'll drop patch 3 but keep patches 1-2 still in review. Thanks for
verifying this!

But of course it would be better if somebody could test this properly on
a real device. When that happens, please resubmit patch 3.

-- 
Kalle Valo