On Tue, 7 Sep 2021 22:31:30 GMT, Smita Kamath <[email protected]> wrote:
> Performance dropped up to 10% for 1k data after 8267125 for CPUs that do not > support the new intrinsic. Tests run were crypto.full.AESGCMBench and > crypto.full.AESGCMByteBuffer from the jmh micro benchmarks. > > The problem is each instance of GHASH allocates 96 extra longs for the > AVX512+VAES intrinsic regardless if the intrinsic is used. This extra table > space should be allocated differently so that non-supporting CPUs do not > suffer this penalty. This issue also affects non-Intel CPUs too. src/hotspot/share/opto/library_call.cpp line 6796: > 6794: > 6795: Node* avx512_subkeyHtbl = new_array(klass_node, intcon(96), 0); > 6796: if (avx512_subkeyHtbl == NULL) return false; This looks very Intel-specific, but it's in generic code. Please make this constant 96 a symbol and push it into a header file in the x86 back end. ------------- PR: https://git.openjdk.java.net/jdk/pull/5402
