On Wed, 11 May 2022 15:10:55 GMT, Paul Sandoz <psan...@openjdk.org> wrote:
>> Checking whether the indexes of masked lanes are inside of the valid memory
>> boundary is necessary for masked vector memory access. However, this could
>> be saved if the given offset is inside of the vector range that could make
>> sure no IOOBE (IndexOutOfBoundaryException) happens. The masked load APIs
>> have saved this kind of check for common cases. And this patch did the
>> similar optimization for the masked vector store.
>>
>> The performance for the new added store masked benchmarks improves about
>> `1.83x ~ 2.62x` on a x86 system:
>>
>> Benchmark Before After Gain Units
>> StoreMaskedBenchmark.byteStoreArrayMask 12757.936 23291.118 1.826 ops/ms
>> StoreMaskedBenchmark.doubleStoreArrayMask 1520.932 3921.616 2.578 ops/ms
>> StoreMaskedBenchmark.floatStoreArrayMask 2713.031 7122.535 2.625 ops/ms
>> StoreMaskedBenchmark.intStoreArrayMask 4113.772 8220.206 1.998 ops/ms
>> StoreMaskedBenchmark.longStoreArrayMask 1993.986 4874.148 2.444 ops/ms
>> StoreMaskedBenchmark.shortStoreArrayMask 8543.593 17821.086 2.086 ops/ms
>>
>> Similar performane gain can also be observed on ARM hardware.
>
> src/jdk.incubator.vector/share/classes/jdk/incubator/vector/X-Vector.java.template
> line 4086:
>
>> 4084: } else {
>> 4085: $Type$Species vsp = vspecies();
>> 4086: if (offset < 0 || offset > (a.length - vsp.length())) {
>
> Can we use `VectorIntrinsics.checkFromIndexSize`? e.g.
>
> if (!VectorIntrinsics.checkFromIndexSize(offset, vsp.length(), a.length)) {
> ...
Thanks for the review @PaulSandoz ! For the
`VectorIntrinsics.checkFromIndexSize`, I'm afraid it's not suitable to be used
here because the `outOfBounds` exception will be thrown if the offset is not
inside of the valid array boundary. And for the masked operations, this is not
needed since we only need to check the masked lanes. Please correct me if I
didn't understand correctly. Thanks!
-------------
PR: https://git.openjdk.java.net/jdk/pull/8620
