RFR: JDK-8300808: Accelerate Base64 on x86 for AVX2 [v11]

Scott Gibbons duke at openjdk.org
Tue Feb 7 14:30:53 UTC 2023


On Tue, 7 Feb 2023 00:12:21 GMT, Scott Gibbons <duke at openjdk.org> wrote:

>> Added code for Base64 acceleration (encode and decode) which will accelerate ~4x for AVX2 platforms.
>> 
>> Encode performance:
>> **Old:**
>> 
>> Benchmark                      (maxNumBytes)   Mode  Cnt     Score   Error   Units
>> Base64Encode.testBase64Encode           1024  thrpt    3  4309.439 ± 2.632  ops/ms
>> 
>> 
>> **New:**
>> 
>> Benchmark                      (maxNumBytes)   Mode  Cnt      Score     Error   Units
>> Base64Encode.testBase64Encode           1024  thrpt    3  24211.397 ± 102.026  ops/ms
>> 
>> 
>> Decode performance:
>> **Old:**
>> 
>> Benchmark                      (errorIndex)  (lineSize)  (maxNumBytes)   Mode  Cnt     Score    Error   Units
>> Base64Decode.testBase64Decode           144           4           1024  thrpt    3  3961.768 ± 93.409  ops/ms
>> 
>> **New:**
>> Benchmark                      (errorIndex)  (lineSize)  (maxNumBytes)   Mode  Cnt      Score    Error   Units
>> Base64Decode.testBase64Decode           144           4           1024  thrpt    3  14738.051 ± 24.383  ops/ms
>
> Scott Gibbons has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Add algorithm comments

@sviswa7 Can you please share the test(s) you used to determine the stated failures?  I've run all of the tier1-3 tests in the suite with no failures.  Plus, if what you say is true, then the same logic would apply for the non-URL case as well as it applies to 0x2F ('/') in the same manner as 0x5F ('_').

If your test(s) do indeed show the failure, we should probably add them to the test suite.  Thanks.

-------------

PR: https://git.openjdk.org/jdk/pull/12126


More information about the core-libs-dev mailing list