RFR: 8367333: C2: Vector math operation intrinsification failure
Emanuel Peter
epeter at openjdk.org
Tue Sep 16 06:49:20 UTC 2025
On Fri, 12 Sep 2025 19:14:18 GMT, Vladimir Ivanov <vlivanov at openjdk.org> wrote:
> As part of [JDK-8353786](https://bugs.openjdk.org/browse/JDK-8353786), C2 support for operations backed by the vector math library was completely removed. On JDK side, there is a special dispatching logic added to avoid intrinsic calls in `jdk.internal.vm.vector.VectorSupport`. But it's still possible to observe such paradoxical situations (intrinsic calls with obsolete operation IDs) when processing effectively dead code.
>
> Consider `FloatVector::lanewiseTemplate`:
>
> FloatVector lanewiseTemplate(VectorOperators.Unary op) {
> if (opKind(op, VO_SPECIAL)) {
> ...
> else if (opKind(op, VO_MATHLIB)) {
> return unaryMathOp(op);
> }
> }
> int opc = opCode(op);
> return VectorSupport.unaryOp(opc, ...);
> }
>
>
> At runtime, `unaryMathOp` is unconditionally invoked, but during compilation it's possible to end up with an intrinsification attempt of `VectorSupport.unaryOp()` before `opKind(op, VO_SPECIAL)` is inlined.
>
> It can be reliably reproduced `-XX:+StressIncrementalInlining` flag.
>
> The fix is to fail-fast intrinsification rather than crashing the VM.
>
> Testing: tier1 - tier4
Changes requested by epeter (Reviewer).
test/hotspot/jtreg/compiler/vectorapi/TestVectorMathLib.java line 33:
> 31: * @test
> 32: * @bug 8367333
> 33: * @requires vm.compiler2.enabled
Do you need this `@requires`? It might be nice to be able to run this with other compilers too.
test/hotspot/jtreg/compiler/vectorapi/TestVectorMathLib.java line 40:
> 38: * -XX:CompileCommand=compileonly,compiler.vectorapi.TestVectorMathLib::test*
> 39: * -XX:+StressIncrementalInlining
> 40: * compiler.vectorapi.TestVectorMathLib
Like @jatin-bhateja mentioned: alignment is off.
I'd also like to see a run without flags, maybe with only `-XX:CompileCommand=compileonly,compiler.vectorapi.TestVectorMathLib::test*`
-------------
PR Review: https://git.openjdk.org/jdk/pull/27263#pullrequestreview-3227803098
PR Review Comment: https://git.openjdk.org/jdk/pull/27263#discussion_r2351063466
PR Review Comment: https://git.openjdk.org/jdk/pull/27263#discussion_r2351069465
More information about the hotspot-compiler-dev
mailing list