RFR: 8324751: C2 SuperWord: Aliasing Analysis runtime check [v22]

Emanuel Peter epeter at openjdk.org
Thu Aug 28 05:56:04 UTC 2025


On Wed, 27 Aug 2025 16:02:27 GMT, Vladimir Kozlov <kvn at openjdk.org> wrote:

>> This looks like "rabbit hole" :(
>> 
>> May be file a separate RFE to investigate this behavior later by some other engineer.  Most concerning is that reproduced on different platforms.
>> 
>> I agree that we may accept this regression since it happened in corner case. I assume our benchmarks are not affected by this. Right?
>
>> @vnkozlov You I think the patch is now stable, and you can review again.
>> 
>> For the edge-case regressions [here](https://github.com/openjdk/jdk/pull/24278#issuecomment-3213393035): should I file a bug or RFE? At least that way we already are tracking it, and can say we are aware of it if it should ever come up. And we can also provide some possible work-arounds.
> 
> I think it should be RFE - to improve this implementation for this corner case.

@vnkozlov @mhaessig Thank you very much for reviewing. I know this was a huge change set to work through, so I'm very thankful 😊 

@vnkozlov I did some additional high-tier testing. No related failures. (aborted some remaining tests on macosx-aarch64, since it was the only platform that did not finish tests - preserving some resources there)

I also filed the follow-up RFE:
[JDK-8366274](https://bugs.openjdk.org/browse/JDK-8366274) C2 SuperWord: investigate edge case performance regressions from JDK-8324751

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

PR Comment: https://git.openjdk.org/jdk/pull/24278#issuecomment-3231991774


More information about the hotspot-compiler-dev mailing list