Closed marcoeilers closed 6 months ago
I'm surprised that we are still running into non-linear terms here, considering that I am using the silicon branch with the wildcard multiplication optimization
That's probably because I'm an idiot and forgot some cases there (added them now)
I would be happy to merge this as soon as the release is out (assuming the tests are passing)
This avoids having to prove that the predicate can be folded, which is not needed (but can still sometimes fail e.g. if wildcard multiplications lead to non-linear terms, like in an example that @jcp19 found).