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

[Diana Picus via llvm-commits]

Thanks. I was actually reviewing it anyway (gotta run now but will
post some comments in about an hour or so).

Diana

On 4 October 2016 at 16:00, Renato Golin <renato.golin at linaro.org> wrote:
> rengolin added a reviewer: rovka.
> rengolin added a comment.
>
> Hi,
>
> Adding Diana, as she was looking into the VMLA hazard a few weeks ago in an attempt to remove a few CPU-specific flags. This review will need a deeper look in relation to the `isLikeA9` check.
>
> cheers,
> --renato
>
>
> https://reviews.llvm.org/D25020
>
>
>