RFR: 8332670: C1 clone intrinsic needs memory barriers [v2]

Galder Zamarreño galder at openjdk.org
Thu Jun 6 14:31:45 UTC 2024


On Thu, 6 Jun 2024 14:11:35 GMT, Martin Doerr <mdoerr at openjdk.org> wrote:

>> Galder Zamarreño has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   Keep storestore barrier for array allocation
>>   
>>   * Having c1 array clone use 2 storestore barriers
>>   has no performance impact, so it's safer to keep it in place.
>
> Seems like the `/integrate` got lost.  I can see it in my email.

@TheRealMDoerr I removed it because I had just noticed the test failure in CI, looking into it.

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

PR Comment: https://git.openjdk.org/jdk/pull/19538#issuecomment-2152685855


More information about the hotspot-compiler-dev mailing list