RFR: 8323220: Reassociate loop invariants involved in Cmps and Add/Subs [v5]

[Joshua Cao]

On Fri, 9 Feb 2024 06:18:47 GMT, Emanuel Peter <epeter at openjdk.org> wrote:

>> failure is due to mac infra issues and not related to test. @eme64 do the @check's look ok?
>
> @caojoshua sorry, I hope this is not too annoying. I think you are doing good work. We will get there :)
> 
> About the add/sub matching fragility: do you know where the other add/sub are coming from? I wonder if there is anything that we could improve in the IR framework to help with that in the future.
> 
> I hope that in the future we can drop the `@Check` methods completely, see https://bugs.openjdk.org/browse/JDK-8310533
> But that will take a while, since we also plan to refactor the IR framework a little.

@eme64 thank you for integrating `@setup`. We can create random inputs, but still have enough control to avoid traps. I've added the missing `lt` check, and correctness checks for all tests.

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

PR Comment: https://git.openjdk.org/jdk/pull/17375#issuecomment-1947576299