[RFC][PATCH 0/2] aarch64: Add optimized ASIMD versions of sinf/cosf

[Sekhar, Ashwin]

On Tue, 2017-06-13 at 15:49 -0300, Adhemerval Zanella wrote:
> 
> On 13/06/2017 14:49, Wilco Dijkstra wrote:
> > 
> > Adhemerval Zanella wrote:
> > 
> > > 
> > > I think a good starting point I would be if Ashwin in could
> > > provide us with a C skeleton with same implementation done in
> > > assembly.
> > I don't see the point of asking him to do that. It would be a
> > significant amount of
> > work which would be wasted once Szabolcs posts his implementation.
> It was not a demand, but rather a suggestion if it were the case he
> has a
> starting implementation based on C (which I know might not be the
> case).
After your suggestion, I tried implementing a C skeleton (attached with
this mail) of sinf using the same algorithm. I am able to see same kind
of speedup with the C version.

But I am not sure whether to spend time addressing other technical
comments on the current patch and cleaning this C version for
submission as there is likely another patch coming from Szabolcs that
supposedly can supersede my work.

Ashwin
> 
> Also, it would be helpful which kind of algorithm strategies Szabolcs
> is
> planing to do different than Ashwin and current implementation that
> is
> intended to supersede both implementation (like short algorithm
> description
> sent by Ashwin).
I would also like know about the same.

In my work, I only used algorithms that are already in libm in other
architectures' sinf/cosf implementations. So I guess the issue
that Szabolcs raised about math code licensing doesn't really apply to
my patch??

Ashwin
> 
> > 
> > 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: s_sinf.c
Type: text/x-csrc
Size: 6906 bytes
Desc: s_sinf.c
URL: <http://sourceware.org/pipermail/libc-alpha/attachments/20170616/b06a763c/attachment.bin>