Note to self: Emulation works better if you take the CPU out of halt after an NMI.

All of a sudden the programs I test actually run.

Running a test suite to verify my #z80 emulator does what it's supposed to. When testing the documented flags, all tests pass except one.

```
024 ALO A,A FAILED CRC:1EC25C69 Expected:C8056BB1
025 ALO A,[B,C] OK
```

The `ALO` test runs some instructions from the Arithmetic and Logical Group of instructions. What I don't understand is why the `ALO A,A` test fails, while `ALO A,[B,C]` is OK. It's running the exact same code!

What am I doing wrong?
What have I missed?
Am I going mad?

My emulator project is coming along nicely. I started coding a few weeks ago and is pretty happy with the progress.

The ABC80 was a Swedish Z80 based micro computer introduced in 1978.

I am no longer with Windows.

Had not booted into it more than a handful of times the last year anyway, but when the SSD with my Linux installation started to warn about imminent failure, I took the executive decision to wipe my Windows disk and replace with a new Linux install.

There's only a few games I'll be missing out on, but that's a small sacrifice. I just wish there was a really good replacement for VSCode, and I could go MS free.

The (relatively new) power supply for my Atari 600 gave up yesterday. Quite annoying, but I pulled out my C64 instead for the first time in ages.

Maybe Smol Snek 2 should be a C64 game. 🤔

As an exercise (in patience) I'm implementing a TNFS server in Rust.

Today I reached the point where I can mount a disk image (Read only) in #fujinet and boot from it on my 600XL. Now I just need to add write support before I start the refactoring. 😅