RFR: 8343689: AArch64: Optimize MulReduction implementation

Wed Feb 5 11:30:18 UTC 2025

On Sat, 18 Jan 2025 09:03:19 GMT, Andrew Haley <aph at openjdk.org> wrote:

> Please provide info about whuch CPUs are benchmarked. How does this compare with Graviton 4?

Hi @theRealAph , I've updated the description to reflect the former.

As for the latter, nothing changes for Graviton 4, as for 128b long vectors we keep using the existing Neon implementation.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/23181#issuecomment-2636468714