RFR: 8322174: RISC-V: C2 VectorizedHashCode RVV Version [v22]

[Yuri Gaevsky]

On Mon, 11 Aug 2025 06:05:15 GMT, Fei Yang <fyang at openjdk.org> wrote:

> > Based on above experiments it looks reasonable to use `m2` grouping.
> 
> Thanks for the extra JMH numbers. Yes, I agree that `m2` is more reasonable here. That means we won't need to reserve so many vector registers for `instruct varrays_hashcode` in src/hotspot/cpu/riscv/riscv_v.ad. So can you free the unused vector registers? Will take a more closer look after that.

Heh, I completely missed that, thanks a lot for catching this!

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

PR Comment: https://git.openjdk.org/jdk/pull/17413#issuecomment-3173611865