RFR: 8293695: Implement isInfinite intrinsic for RISC-V

[Fei Yang]

On Tue, 13 Sep 2022 15:48:38 GMT, Aleksei Voitylov <avoitylov at openjdk.org> wrote:

> RISC-V 64 intrinsic for isInfinite follows the logic of x86 intrinsic (introduced by 8285868). This patch adds C2 match for IsInfinite nodes. Existing test is modified to run on RISC-V and passes on both release and fastdebug builds. Benchmark results are below:
> 
> before:
> Benchmark                              Mode  Cnt   Score   Error Units
> DoubleClassCheck.testIsInfiniteBranch  avgt   15  43.547 ± 6.843 ns/op
> DoubleClassCheck.testIsInfiniteCMov    avgt   15  16.301 ± 1.386 ns/op
> DoubleClassCheck.testIsInfiniteStore   avgt   15  16.230 ± 1.477 ns/op
> FloatClassCheck.testIsInfiniteBranch   avgt   15  38.774 ± 3.572 ns/op
> FloatClassCheck.testIsInfiniteCMov     avgt   15  15.064 ± 1.310 ns/op
> FloatClassCheck.testIsInfiniteStore    avgt   15  14.967 ± 1.298 ns/op
> 
> after:
> Benchmark                              Mode  Cnt   Score    Error Units
> DoubleClassCheck.testIsInfiniteBranch  avgt   15  39.987 ±  6.179 ns/op
> DoubleClassCheck.testIsInfiniteCMov    avgt   15  13.477 ±  1.159 ns/op
> DoubleClassCheck.testIsInfiniteStore   avgt   15   9.607 ±  0.834 ns/op
> FloatClassCheck.testIsInfiniteBranch   avgt   15  36.265 ±  3.168 ns/op
> FloatClassCheck.testIsInfiniteCMov     avgt   15  13.230 ±  1.100 ns/op
> FloatClassCheck.testIsInfiniteStore    avgt   15   9.492 ±  0.807 ns/op
> 
> According to 8285868 discussion, isNaN and isFinite methods intrinsification using the same approach might be not beneficial. I'm going to investigate it for RISC-V and propose methods intrinsification as part of further work in case it's profitable.

Looks OK. Also passed Tier1 test on my SiFive Unmatched board.

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

Marked as reviewed by fyang (Reviewer).

PR: https://git.openjdk.org/jdk/pull/10253