[vectorIntrinsics] RFR: Improve mask reduction operations on AVX

[Mai Đặng Quân Anh]

On Tue, 2 Nov 2021 15:59:16 GMT, Paul Sandoz <psandoz at openjdk.org> wrote:

>> Hi,
>> This patch improves the logic of vector mask reduction operations on AVX, especially int, float, long, double, by using vmovmskpd and vmovmskps instructions. I also do a little refactoring to reduce duplication in toLong. The patch temporarily disables these operations on Neon, though.
>> Thank you very much.
>
> Thanks, this will require review from engineers more knowledgeable than I on the x86 aspects and ARM aspects. IMO we cannot integrate until there is a plan for ARM either as part of this PR or a follow on PR.

Thank @PaulSandoz for the response, I hope this PR will be useful in the future.

Besides, I found out that C2 fails with `Long128Vector`, even a simple operation of `LONG_128_1.eq(LONG_128_2)`, is this a known behaviour?

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

PR: https://git.openjdk.java.net/panama-vector/pull/158