Re: [lworld] RFR: 8385328: [lworld] C2: assert(phi->_idx >= nodes_size()) failed: only new Phi per instance memory slice still happens after JDK-8374742 [v6]

[Daniel D . Daugherty]

On Mon, 8 Jun 2026 13:32:58 GMT, Roland Westrelin <rol...@openjdk.org> wrote:

>> Roland Westrelin has updated the pull request incrementally with one 
>> additional commit since the last revision:
>> 
>>   review
>
> I updated the change again so that during igvn, if a `LoadFlat` or 
> `StoreFlat` that is mismatched is encountered then the memory chain is not 
> optimized further. Given those are not eliminated during EA, they could be 
> for the `t_oop` instance.

@rwestrel - Please note that the ProblemListing for 
tools/javac/platform/CanHandleClassFilesTest.java
has been widened to include all platforms.

I have not ProblemListed the BigApps tests, but we are seeing noise there also,
especially with applications/renaissance/RenaissanceStressTest.java.

Is there an ETA for this fix to land in repo-valhalla?

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

PR Comment: https://git.openjdk.org/valhalla/pull/2489#issuecomment-4723691477