rtVAX 300.

I wouldn't worry too much about that. I once saw a document about doing bringup on a new architecture that said you should expect it to take at least several hundred failed boot attempts before you first get hello-world booted. In this case, I wouldn't expect it to be quite that bad, because the basic CPU architecture is already supported. But I wouldn't be surprised if it took several dozen attempts. [...] [...] Hm. sp = pc-1, fp = -1, ap = 0x200. My first guess is, like yours, that it isn't actually executing stuff. First thing I'd look at is what's in memory starting at 1800, with a peek also at what's at 200 (because that's where ap points). If I had to hazard a guess with just the information I have, I'd guess that 1800 holds the bootblock you built but with a non-executable header on it. At this point I'd probably try netbooting something that just goes into an infinite loop. Maybe something like .globl start start: nop 1: nop br 1b because that's about the simplest thing you can do that (if actually run) behaves differently from what you have now. I'd also take a detailed look (I'm talking "hexdump -C" level of detailed here) at the file being MOP-booted to make sure its structure is what you think it is. You're making me wish I had one of those cute little machines (well, aside from liking cute little machines, that is); if I had I would be in a position to do this myself. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B