[PATCH] D25020: [ARM] Fix 26% performance regression on Cortex-A9 caused by not using VMLA/VMLS

[Evgeny Astigeevich via llvm-commits]

eastig added a comment.

I've got results of the LNT run when SIMD VMLx are not expanded. I checked my changes with vmlx-fwd.ll from here. It's strange but there are no performance changes.


https://reviews.llvm.org/D25020