On Mon, 1 Jun 2026 07:16:42 GMT, Joel Sikström <[email protected]> wrote:
> Hello, > > The StoreStore membar in the AArch64 implementation of TemplateTable::aaload > is redundant after > [JDK-8378862](https://bugs.openjdk.org/browse/JDK-8378862), which calls > `OrderAccess::storestore()` in the VM instead. > > I suggest we remove it so we don't have a double storestore-membar for this > path. I don't think we need a comment here explaining that the VM handles > memory ordering for us. > > Testing: > * Oracle's tier1-4 on linux-aarch64 release/fastdebug. > > > > > --------- > - [x] I confirm that I make this contribution in accordance with the [OpenJDK > Interim AI Policy](https://openjdk.org/legal/ai). Thank you for the reviews everyone! The failing GHA runs are unrelated and in estNullableInlineTypes.java, which is fixed in https://github.com/openjdk/valhalla/pull/2493. ------------- PR Comment: https://git.openjdk.org/valhalla/pull/2495#issuecomment-4599363716
