On Mon, 12 Oct 2020 22:00:24 GMT, CoreyAshford <github.com+51754783+coreyashf...@openjdk.org> wrote:
>> This latest push passes the intrinsic regression test. I had run the >> intrinsic TestBase64 regression test on the >> previous push, but not the one in utils. Interesting. Somehow it didn't >> occur to me that there could be a problem >> there if the intrinsic TestBase64 test passed. I will check into the other >> regression test. Don't review this latest >> push just yet. > > Ok, all is clear. I just ran `jdk/java/util/Base64/TestBase64.java` which > passes as well. Please review again when > convenient. Hi Corey, thanks for taking some stuff out of the “too short” path. There may be a performance regression when decoding many short arrays because of the stub call overhead and the usage of the slower part of the Java implementation. We could do it a little better in many cases to compute the maximum possible iteration count i: i = (sl - sp) / block_size if (i * block_size > sl - 12) i-- if (i <= 0) return 0 What do you think? I don’t think branch prediction hints are helpful for the “too short” check. And we should better use CCR1 instead of CCR2 which is specified as non-volatile. Did you already find a 2nd reviewer for the PPC64 part? Best regards, Martin From: CoreyAshford <notificati...@github.com> Sent: Dienstag, 13. Oktober 2020 00:01 To: openjdk/jdk <j...@noreply.github.com> Cc: Doerr, Martin <martin.do...@sap.com>; Mention <ment...@noreply.github.com> Subject: Re: [openjdk/jdk] 8248188: Add IntrinsicCandidate and API for Base64 decoding (#293) Ok, all is clear. I just ran jdk/java/util/Base64/TestBase64.java which passes as well. Please review again when convenient. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<https://github.com/openjdk/jdk/pull/293#issuecomment-707367283>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AKR64KGYKHVCHZYEHSHMSD3SKN4ARANCNFSM4RVHNW5Q>. ------------- PR: https://git.openjdk.java.net/jdk/pull/293