RFR: 8288904: Incorrect memory ordering in UL [v2]

Johan Sjölén duke at openjdk.org
Wed Jul 27 13:06:56 UTC 2022


On Wed, 22 Jun 2022 07:45:26 GMT, Johan Sjölén <duke at openjdk.org> wrote:

>> Please review this fix for JDK-8288904, thank you.
>> 
>> Passed tier1-tier3 tests, excluding faulty tests.
>
> Johan Sjölén has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Comment the loadstore

I've now re-run tier1-3 tests and they are fully passing (excluding an issue fixed in 8291289). I propose that this patch is merged as is and a new ticket is added to evaluate and perhaps rewrite this implementation.

@dholmes-ora , I believe that it is meant to be multiple reader, single writer. It seems that this is only altered through `LogConfiguration`, which holds the `ConfigurationLock` during reading and writing. I assume that people have accepted this code as correct as empirically it works.

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

PR: https://git.openjdk.org/jdk/pull/9225


More information about the hotspot-runtime-dev mailing list