RFR: 8259948: Aarch64: Add cast nodes for Aarch64 Neon backend [v5]

[Andrew Haley]

On Wed, 22 Sep 2021 03:10:14 GMT, Ningsheng Jian <njian at openjdk.org> wrote:

>> One is _Advanced SIMD copy_, the other _Advanced SIMD scalar copy_.  The _Advanced SIMD copy_ version is missing a comment to that effect. Maybe one day someone could go through assembler_aarch64.hpp and add the missing titles from the A64 instruction set encoding section of the ARM.
>
>> One is _Advanced SIMD copy_, the other _Advanced SIMD scalar copy_. The _Advanced SIMD copy_ version is missing a comment to that effect. Maybe one day someone could go through assembler_aarch64.hpp and add the missing titles from the A64 instruction set encoding section of the ARM.
> 
> Yes, a comment will be helpful. But for the scalar copy, we should use SIMD_RegVariant instead of SIMD_Arrangement?

It looks like the official doc uses _SIMD_RegVariant_ for the scalar copy, so we should do the same. I don't think you'll ever find me suggesting we should deviate from what the Arm reference manuals do, unless we can't avoid it.

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

PR: https://git.openjdk.java.net/jdk/pull/4839