> How to trigger this alternate path which is !ssa && lw > 0? In other
> words, do we have tests for both cases of that else branch.
Now that you mention it, we don't have a test for the already existing
behavior. It got introduced in b83c587f6e68c77d15a9b9335044a643172bc975, but I
don't know why it is the way it is. I tried to leave the original behavior,
but I'm not sure if it's even used.
How should I deal with that?
I can't tell right away if it's a redundant path we currently have. Best way to deal with it is to add new test for existing code, making sure it triggers this path.
Regards,
Fabian Maurer