Matthias Kretz

Note that you can always write your own non-member function that implements this on top of Vc 1.4 and `std::experimental::simd`. E.g. https://godbolt.org/z/0rksOw (this is a state close to the master...

since you want this for 8-bit integers I'll put it on Vc 2

I'd like to support simd for user-defined T at some point when the C++ reflection capabilities are good enough. However, what you're really looking for is std::complex. It almost works,...

True, but consider that std::log/exp/sin/... are defined on std::complex and thus would automatically exist for std::complex. Intel certainly did some heroics to support simd in their instruction set. But I...

> Maybe I can try to merge in my branch (branched off from mkretz/datapar) and then after pull when all conflicts will be solved? You should give it a try....

What do you think about organizing our work on https://github.com/VcDevel/Vc/projects/3? Makes it easy for you to put subtasks on my to list - and for me to see where you...

> a merge from mkretz/NEON to mkretz/datapar Wait, did you do `git merge mkretz/NEON` then? That's not what I intended. I wanted to have the NEON commits rebased onto mkretz/datapar....

I agree those two commits should go to mkretz/datapar. They'll not apply without a bit of extra work, but you should give it a shot.

Looks about right, yes. 6cdad3a3fd5ade362fd315951b52b0fc393b8bf5 might not be important, though. I'm trying to fade out the use of the Vc_IMPL macro in `mkretz/datapar`. Short summary: there are two somewhat orthogonal...

@oshadura Oh, and please don't push to my branches. If you think something should be merged to my branch please open a pull request instead.