[vectorIntrinsics] Integrated: 8259604: Change the implementation of loadV_partial and storeV_partial in Aarch64-SVE backend

[Wang Huang]

On Mon, 18 Jan 2021 06:55:48 GMT, Wang Huang <whuang at openjdk.org> wrote:

> Now, the implementation of `loadV_partial` is :
> mov $tmp1, 0
> mov $tmp2, vector_length
> sve_whilelo $pTmp, $tmp1, $tmp2
> sve_ldr $dst, $pTmp, $mem
> However, we can encode register `zr` in instruction `sve_whilelo` instead of getting from the first mov instruction.
> 
> The new implementation is :
> mov $tmp, vector_length
> sve_whilelo $pTmp, zr, $tmp
> sve_ldr $dst, $pTmp, $mem
> From this changing, we reduce a mov instruction and a temporary variable($tmp2 ).

This pull request has now been integrated.

Changeset: ca9fad51
Author:    Wang Huang <whuang at openjdk.org>
Committer: Ningsheng Jian <njian at openjdk.org>
URL:       https://git.openjdk.java.net/panama-vector/commit/ca9fad51
Stats:     36 lines in 2 files changed: 0 ins; 12 del; 24 mod

8259604: Change the implementation of loadV_partial and storeV_partial in Aarch64-SVE backend

Co-authored-by: He Xuejin <hexuejin2 at huawei.com>
Reviewed-by: njian

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

PR: https://git.openjdk.java.net/panama-vector/pull/32