[lworld] RFR: 8353182: [lworld] C2: Multiple IR test failures in compiler/gcbarriers/TestZGCBarrierElision.java after JDK-8351569 [v2]

[duke]

On Wed, 2 Jul 2025 13:12:24 GMT, Damon Fenacci <dfenacci at openjdk.org> wrote:

>> This is a very similar issue to [JDK-8343420](https://bugs.openjdk.org/browse/JDK-8343420).
>> 
>> A few `TestZGCBarrierElision` tests fail after [JDK-8351569](https://bugs.openjdk.org/browse/JDK-8351569). These tests check that ZGC barrier elision optimization does not elide necessary barriers and partially do so using `VarHandle::getAndSet` and checking that `GetAndSetP` nodes with barriers are created.
>> As found out in [JDK-8343420](https://bugs.openjdk.org/browse/JDK-8343420), [JDK-8351569](https://bugs.openjdk.org/browse/JDK-8351569) refactored `Unsafe::getAndSet` (which is used by `VarHandle::getAndSet`) to take value-classes into account, which implies that checks must be introduced to see if the target and argument objects are value-classes. This is done with a new overloaded `getAndSet` method that is now called instead of the old one. This method is not intrinsified (only inlined) and doesn't result in a `GetAndSet` node. It instead relies on the `compareAndSet` method (which gets intrinsified) and results in a `CompareAndSwap` node.
>> 
>> Here as well, as the goal of the test is mainly to check for ZGC barrier presence, the sensible fix for now seems to be to modify the `TestZGCBarrierElision` tests to make them check for `CompareAndSwap` instead of `GetAndSet` nodes.
>> 
>> Tests: Tier 1-3+
>
> Damon Fenacci has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains two commits:
> 
>  - Merge branch 'lworld' into JDK-8353182
>  - JDK-8353182: [lworld] C2: Multiple IR test failures in compiler/gcbarriers/TestZGCBarrierElision.java after JDK-8351569

@dafedafe 
Your change (at version b3ab62f942d7ab6e5e12fb72e62696a7de14de2a) is now ready to be sponsored by a Committer.

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

PR Comment: https://git.openjdk.org/valhalla/pull/1497#issuecomment-3028192337