RFR: 8342881: RISC-V: secondary_super_cache does not scale well: C1 and interpreter [v5]

Gui Cao gcao at openjdk.org
Tue Dec 10 11:09:37 UTC 2024


On Tue, 10 Dec 2024 10:12:50 GMT, Fei Yang <fyang at openjdk.org> wrote:

> Thanks for the update! Will take another look. BTW: Did you try this JMH to see the benefit?

Yes, I have performed this JMH.

Without this patch:

Benchmark                                   Mode  Cnt     Score     Error  Units
SecondarySuperCacheInterContention.test     avgt   15  5904.968 ±  36.829  ns/op
SecondarySuperCacheInterContention.test:t1  avgt   15  5938.559 ± 201.657  ns/op
SecondarySuperCacheInterContention.test:t2  avgt   15  5871.377 ± 176.948  ns/op
Finished running test 'micro:vm.compiler.SecondarySuperCacheInterContention'


Apply this patch:

Benchmark                                   Mode  Cnt   Score   Error  Units
SecondarySuperCacheInterContention.test     avgt   15  52.563 ± 5.843  ns/op
SecondarySuperCacheInterContention.test:t1  avgt   15  52.781 ± 5.336  ns/op
SecondarySuperCacheInterContention.test:t2  avgt   15  52.344 ± 6.431  ns/op
Finished running test 'micro:vm.compiler.SecondarySuperCacheInterContention'

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

PR Comment: https://git.openjdk.org/jdk/pull/21922#issuecomment-2531257812


More information about the hotspot-dev mailing list