RFR: 8354471: Assertion failure with -XX:-EnableX86ECoreOpts [v2]
Jatin Bhateja
jbhateja at openjdk.org
Wed Apr 16 16:19:46 UTC 2025
On Tue, 15 Apr 2025 15:48:02 GMT, Volodymyr Paprotski <vpaprotski at openjdk.org> wrote:
>> The check to choose between AVX2 and AVX512 implementation was relying on `EnableX86ECoreOpts`. It should be relying on `supports_avxifma` and mirror the `UseIntPolyIntrinsics` check in `vm_version_x86.cpp`.
>>
>> Note, in `stubGenerator_x86_64.cpp`, entry to the patched function is protected already:
>>
>> if (UseIntPolyIntrinsics) {
>> StubRoutines::_intpoly_montgomeryMult_P256 = generate_intpoly_montgomeryMult_P256();
>> StubRoutines::_intpoly_assign = generate_intpoly_assign();
>> }
>
> Volodymyr Paprotski has updated the pull request incrementally with one additional commit since the last revision:
>
> flip the direction of if to prefer AVX512
Minor comment modification suggestion.
Otherwise looks good.
Marked as reviewed by jbhateja (Reviewer).
src/hotspot/cpu/x86/stubGenerator_x86_64_poly_mont.cpp line 576:
> 574: montgomeryMultiply(aLimbs, bLimbs, rLimbs, tmp, _masm);
> 575: } else {
> 576: assert(VM_Version::supports_avxifma(), "Require AVXIFMA support");
Suggestion:
assert(VM_Version::supports_avxifma(), "Require AVX_IFMA support");
-------------
PR Review: https://git.openjdk.org/jdk/pull/24644#pullrequestreview-2773099125
PR Review: https://git.openjdk.org/jdk/pull/24644#pullrequestreview-2773107873
PR Review Comment: https://git.openjdk.org/jdk/pull/24644#discussion_r2047277287
More information about the hotspot-compiler-dev
mailing list