[LLVMdev] AVX2 in 3.2

[dag at cray.com]

Nadav Rotem <nrotem at apple.com> writes:

> Hi David, 
>
> There were many changes. For example, efficient lowering of vector
> casts (trunc, zext, sext), simplified vblend patterns, shuffle
> patterns, AVX2 gathers, just to name a few.

Thanks Nadav, that's very helpful!

                          -David