RFR: 8257436: [aarch64] Regressions in ArrayCopyUnalignedDst.testByte/testChar for 65-78 bytes when UseSIMDForMemoryOps is on

[Evgeny Astigeevich]

On Wed, 2 Dec 2020 10:12:41 GMT, Evgeny Astigeevich <github.com+42899633+eastig at openjdk.org> wrote:

> Copying data in range 65 - 80 bytes with
> ldpq
> ldpq 
> ldp
> stpq
> stpq
> stp
> is faster than with
> ldpq
> ldpq
> ldpq
> stpq
> stpq
> stpq
> This fixes regressions of ArrayCopyUnalignedDst.testByte/testChar when UseSIMDForMemoryOps is on.
> See the attached results for performance data
> [test_byte_results.txt](https://github.com/openjdk/jdk/files/5628447/test_byte_results.txt)
> [test_char_results.txt](https://github.com/openjdk/jdk/files/5628451/test_char_results.txt)
> [test_int_results.txt](https://github.com/openjdk/jdk/files/5628462/test_int_results.txt)
> [test_long_results.txt](https://github.com/openjdk/jdk/files/5628468/test_long_results.txt)

@theRealAph could please check if it causes any issues on ThunderX 2?

Thanks,
Evgeny

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

PR: https://git.openjdk.java.net/jdk/pull/1558