Integrated: 8345146: [PPC64] Make intrinsic conversions between bit representations of half precision values and floats

[Martin Doerr]

On Thu, 28 Nov 2024 11:21:46 GMT, Martin Doerr <mdoerr at openjdk.org> wrote:

> PPC64 implementation of [JDK-8289552](https://bugs.openjdk.org/browse/JDK-8289552). I've implemented some more instructions which may be useful in the future.
> VectorCastNodes are not yet implemented on PPC64. Power9 is recognized by the availability of the "darn" instruction.
> 
> Performance on Power9:
> Before patch:
> 
> Benchmark                                     (size)   Mode  Cnt      Score     Error   Units
> Fp16ConversionBenchmark.float16ToFloat          2048  thrpt   15     18.995 ?   0.156  ops/ms
> Fp16ConversionBenchmark.floatToFloat16          2048  thrpt   15     18.730 ?   0.331  ops/ms
> 
> 
> After patch:
> 
> Benchmark                                     (size)   Mode  Cnt      Score      Error   Units
> Fp16ConversionBenchmark.float16ToFloat          2048  thrpt   15    522.637 ?   11.274  ops/ms
> Fp16ConversionBenchmark.floatToFloat16          2048  thrpt   15    408.112 ?    9.069  ops/ms

This pull request has now been integrated.

Changeset: b42d79eb
Author:    Martin Doerr <mdoerr at openjdk.org>
URL:       https://git.openjdk.org/jdk/commit/b42d79eb6a6d497dc63718c2854609bebca4498c
Stats:     182 lines in 15 files changed: 179 ins; 2 del; 1 mod

8345146: [PPC64] Make intrinsic conversions between bit representations of half precision values and floats

Reviewed-by: rrich, lucy

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

PR: https://git.openjdk.org/jdk/pull/22433