Hi Xuelei, Thanks to you for raising these concerns and providing your feedback.
On 5/24/19 7:47 PM, Xuelei Fan wrote: > Good, I have no further comment for this update. Please go ahead. > > I think there is a possible improvement by calling > Cipher.getInstance(algorithm) only one time for each transformation > algorithm. But may not worthy as the duplicated transformation > algorithm number is still small. I'm fine if you want to leave it as it > is. Yes, I agree on both statements: there can be an improvement there but it won't be significant. Here it's Webrev.02: * http://cr.openjdk.java.net/~mbalao/webrevs/8223482/8223482.webrev.02/ At Webrev.02, Cipher.getInstance results are cached so we don't need to create instances unnecessary. Benchmarks do not show any significant performance increase nor decrease compared to Webrev.01: Benchmark (testMode) Mode Cnt Score Error Units SupportedCiphersuites.test_TLS12Communication FIPS thrpt 10 167.393 ± 35.659 ops/s SupportedCiphersuites.test_TLS12Communication NON_FIPS thrpt 10 593.441 ± 110.044 ops/s FIPS_with_8223482_webrev02.txt average: 396.86 NON_FIPS_with_8223482_webrev02.txt average: 888.05 Are you okay to go with Webrev.02? Kind regards, Martin.-
