On Wed, 13 Nov 2024 21:14:58 GMT, Volodymyr Paprotski <vpaprot...@openjdk.org> wrote:
> Measuring throughput with JMH parameters `-f 1 -i 2 -wi 3 -r 20 -w 30 -p > algorithm=AES/CBC/NoPadding -p dataSize=30000000 -p provider=SunJCE -p > keyLength=128 org.openjdk.bench.javax.crypto.full.AESBench` > > Before: > > Benchmark (algorithm) (dataSize) (keyLength) (provider) > Mode Cnt Score Error Units > AESBench.decrypt AES/CBC/NoPadding 30000000 128 SunJCE > thrpt 2 25.383 ops/s > AESBench.decrypt2 AES/CBC/NoPadding 30000000 128 SunJCE > thrpt 2 32.230 ops/s > AESBench.encrypt AES/CBC/NoPadding 30000000 128 SunJCE > thrpt 2 20.489 ops/s > AESBench.encrypt2 AES/CBC/NoPadding 30000000 128 SunJCE > thrpt 2 21.383 ops/s > > > After: > > Benchmark (algorithm) (dataSize) (keyLength) (provider) > Mode Cnt Score Error Units > AESBench.decrypt AES/CBC/NoPadding 30000000 128 SunJCE > thrpt 2 215.144 ops/s > AESBench.decrypt2 AES/CBC/NoPadding 30000000 128 SunJCE > thrpt 2 411.265 ops/s > AESBench.encrypt AES/CBC/NoPadding 30000000 128 SunJCE > thrpt 2 64.341 ops/s > AESBench.encrypt2 AES/CBC/NoPadding 30000000 128 SunJCE > thrpt 2 73.114 ops/s > > > I have not deterministically proven why chunking works: before the change, > the CBC intrinsic is not being used; and after chunking, it is. There is > quite a bit of GC activity in the default AESBench, so `encrypt2/decrypt2` > versions isolate just crypto (see comment below). src/java.base/share/classes/com/sun/crypto/provider/CipherBlockChaining.java line 153: > 151: // chunkSize is multiple of block size, but otherwise fairly > arbitrary > 152: // Should be large enough to provide intrinsic with optimization > 153: // oportunities nit: type-o -> opportunities src/java.base/share/classes/com/sun/crypto/provider/CipherBlockChaining.java line 155: > 153: // oportunities > 154: int processed = 0; > 155: final int chunkSize = 1024*100; should probably be a constant -- maybe even a system property ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/22086#discussion_r1841201401 PR Review Comment: https://git.openjdk.org/jdk/pull/22086#discussion_r1841201127
