> I hope we can make a more direct test.

Okay, but as I mentioned in my previous message, I could amend this
patch to not change the signature of `load-user-init-file' (even
though the change has no logical impact whatsoever that could change
user-facing behavior). Would you accept the patch if it didn't change
the signature of `load-user-init-file'?

Thanks,
Radon