RFR: 8317763: Follow-up to AVX512 intrinsics for Arrays.sort() PR [v5]

[Paul Sandoz]

On Thu, 12 Oct 2023 03:44:28 GMT, Danny Thomas <duke at openjdk.org> wrote:

> At least on Saphire Rapids the [emulation suggested here](https://github.com/natmaurice/x86-simd-sort/commit/41d03b2d8f3b62a2ee6a3a97a8da7f193a407026) only imposes a 6% penalty for `intSort`, while also mitigating the performance issue on Zen 4.

Thanks for testing this out, this seems a good solution to explore. 

Is there is a way to `ifdef` switching on the CPU vendor? otherwise we could probably set a flag in the build script based of the CPU vendor, which may be more robust if different compilers are used.

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

PR Comment: https://git.openjdk.org/jdk/pull/16124#issuecomment-1759992778