RFR: 8373613: PEXT/PDEP intrinsics cause performance regression on AMD pre-Zen 3 CPUs

Alessandro Autiero duke at openjdk.org
Wed Feb 25 16:36:17 UTC 2026


On Tue, 24 Feb 2026 13:08:18 GMT, Emanuel Peter <epeter at openjdk.org> wrote:

> @Auties00 Is there a benchmark integrated that shows the performance impact of the regression and this patch?

i included a benchmark i wrote for Integer.compress/Long.compress using JMH in the original bug report, but I noticed that they were already included in the JDK so I didn't add them to this PR. About the results, I see that @jaskarth already run the benchmarks so I won't attach them to this message. 

On a separate note, I'm not sure why GitHub Actions is only running jcheck instead of the whole CI suite, maybe it's because this is my first time contributing to the JDK? Do I need to do something manual to make them run?

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

PR Comment: https://git.openjdk.org/jdk/pull/29809#issuecomment-3960503798


More information about the hotspot-dev mailing list