cc @Rexicon226

I've had a message from @Rexicon226 with feedback -- he's currently not able to reply on here, so this comment is just me relaying that feedback :P

He says that this PR is correct and overall fine to merge, but that the printing needs to be disabled (I'm guessing to avoid spamming the CI logs, which I would agree with). Looks like that would just require reverting the true in enable_print or something along those lines. Also, of course, there have been CI failures which indicate a couple more tests also need to be disabled.

Thanks to @mlugg for relaying my message, was away from the keyboard for a couple weeks with only a few chances to send GitHub messages. Just wanted to disable the printing in the test runner since it spams the output logs on the CI and gets annoying to read. In my great rewrite of the backend (soon:tm:, master...Rexicon226:zig:riscv), I have the build server all working, so I'll be switching over to that.

Seems that a recent rebase is running some test that doesn't work yet on the RISC-V backend :). Disable that and the PR looks great. I had already caught this bug long ago and it kinda got stuck in my endless refactor branch.

I still have a few months of busy working before I'll get a bit more free time in my schedule, so we'll see. I hope to upstream some of that work soon!