RFR: 8355293: [TEST] RISC-V: enable more ir tests [v2]

[Fei Yang]

On Thu, 24 Apr 2025 10:28:37 GMT, Hamlin Li <mli at openjdk.org> wrote:

>> Ah, I see. There is a `ProdRed_Int.java` test there in the same directory. Then should we check for `MUL_REDUCTION_VD` instead of `MUL_REDUCTION_VF`? We are operating on doubles here.
>
> Thanks for catching!
> ~~To confirm, do you mean the line 111 in this file? I'll fix it.~~
> ~~Or you suggest some fix in ProdRed_Int.java too?~~
> I see what you meant, I'll fix it.

I see two occurrences in file ProdRed_Double.java: Both L94 and L111.
Regarding ProdRed_Int.java, seems that we can do a similar cleanup like you do in this PR for other tests.
I mean:

diff --git a/test/hotspot/jtreg/compiler/loopopts/superword/ProdRed_Int.java b/test/hotspot/jtreg/compiler/loopopts/superword/ProdRed_Int.java
index ebc8251e025..9ca670117cf 100644
--- a/test/hotspot/jtreg/compiler/loopopts/superword/ProdRed_Int.java
+++ b/test/hotspot/jtreg/compiler/loopopts/superword/ProdRed_Int.java
@@ -85,8 +85,7 @@ public static void prodReductionInit(int[] a, int[] b) {
     @IR(applyIfCPUFeature = {"sse4.1", "true"},
         applyIfAnd = {"SuperWordReductions", "true", "LoopMaxUnroll", ">= 8"},
         counts = {IRNode.MUL_REDUCTION_VI, ">= 1", IRNode.MUL_REDUCTION_VI, "<= 2"}) // one for main-loop, one for vector-post-loop
-    @IR(applyIfPlatform = {"riscv64", "true"},
-        applyIfCPUFeature = {"rvv", "true"},
+    @IR(applyIfCPUFeature = {"rvv", "true"},
         applyIfAnd = {"SuperWordReductions", "true", "LoopMaxUnroll", ">= 8"},
         counts = {IRNode.MUL_REDUCTION_VI, ">= 1", IRNode.MUL_REDUCTION_VI, "<= 2"}) // one for main-loop, one for vector-post-loop
     public static int prodReductionImplement(int[] a, int[] b, int total) {

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

PR Review Comment: https://git.openjdk.org/jdk/pull/24797#discussion_r2058079363