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

[Diana Picus via llvm-commits]

rovka added a comment.

I have a Cortex-A8 and I think Renato might have a Cortex-A7. I'm going to be away for the next 2 weeks, so I won't be able to help much now, but maybe when I get back we can sort out what hardware we have between us and run a few tests.


https://reviews.llvm.org/D25020