[vectorIntrinsics] RFR: 8265916: Add StoreVectorScatter nodes implementation for Arm SVE [v2]

[Xiaohong Gong]

On Wed, 28 Apr 2021 03:34:19 GMT, Wang Huang <whuang at openjdk.org> wrote:

>> * add `sve_st1w_scatter` and `sve_st1w_scatter`
>> * implement `scatterI` and `scatterL`
>> * `load gather partial` and `store scatter partial` are not contained in this patch, and they will be in next patches
>
> Wang Huang has updated the pull request incrementally with one additional commit since the last revision:
> 
>   fix neon crash

src/hotspot/cpu/aarch64/aarch64_sve.ad line 229:

> 227:       case Op_ExtractUB:
> 228:       // Vector API specific
> 229:       case Op_StoreVectorScatter:

It might need the partial implementation for this opcode. It's ok for me to add them in a separate patch. However, to make sure everything works well, I suggestion to add a block here, like

  case Op_StoreVectorScatter:
        // Currently the implementation for partial vectors are not implemented yet.
        // Will add them in a separate patch.
        return vlen * type2aelembytes(bt) == MaxVectorSize;

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

PR: https://git.openjdk.java.net/panama-vector/pull/73