[PATCH] D116879: [llvm] Allow auto-vectorization of sincos() using libmvec

[Tim Schmielau via Phabricator via llvm-commits]

tim.schmielau added a comment.

Of the two reproducers linked in https://bugs.llvm.org/show_bug.cgi?id=51530, the `sincos_simd.cpp` now auto-vectorizes when compiling with `clang++ -fopenmp -fveclib=libmvec -O2 -march=core-avx2 sincos_simd.cpp`, even when the `#pragma omp simd` annotation is left out in the source.

On the templated example `sincos_simd_template.cpp` a helping hand from the user is needed by annotating the loop with `#pragma omp simd` and adding `-fopenmp` to the compiler flags, but at least vectorization is now possible without the user having to manually substitute function calls.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D116879/new/

https://reviews.llvm.org/D116879