On Tue, 7 Sep 2021 22:31:30 GMT, Smita Kamath <svkam...@openjdk.org> 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. This pull request has now been integrated. Changeset: 13e9ea9e Author: Smita Kamath <svkam...@openjdk.org> Committer: Anthony Scarpino <ascarp...@openjdk.org> URL: https://git.openjdk.java.net/jdk/commit/13e9ea9e922030927775345b1abde1313a6ec03f Stats: 116 lines in 16 files changed: 60 ins; 2 del; 54 mod 8273297: AES/GCM non-AVX512+VAES CPUs suffer after 8267125 Reviewed-by: ascarpino, sviswanathan, aph ------------- PR: https://git.openjdk.java.net/jdk/pull/5402
