RFR: 8341544: Restore fence() in Mutex
Anton Artemov
duke at openjdk.org
Thu May 15 11:47:30 UTC 2025
On Tue, 13 May 2025 10:20:46 GMT, David Holmes <dholmes at openjdk.org> wrote:
> As per JBS comment I still think we only need to add the missing storeload part of the full-fence, to complement the existing barriers the platform sync primitives provide.
@dholmes-ora Each and every implementation of storeload() is either calling fence() or is calling the same method as fence(), what exactly do you mean by the missing part?
> src/hotspot/share/runtime/mutex.hpp line 218:
>
>> 216: // section provided by the mutex follow the Roach-Motel semantics. Having a fence does
>> 217: // not have any significant impact on peformance, as this is an internal VM
>> 218: // mutex and is generally not in hot code parts.
>
> Suggestion:
>
> // mutex and is generally not in hot code paths.
Addressed in the latest commit.
-------------
PR Comment: https://git.openjdk.org/jdk/pull/25033#issuecomment-2876359379
PR Review Comment: https://git.openjdk.org/jdk/pull/25033#discussion_r2084093703
More information about the hotspot-runtime-dev
mailing list