I haven't tested it with an emulator. Really, I don't see the point of doing such a thing, unless you want to test the emulator. Why do you want to run HwTest in an emulator?

I always use emulators to test code that I cannot debug otherwise.
The problem is that then I would need full debugging to know what's wrong and I can't simply reboot my machine for every byte that I change in the code.
Some things like IPI/SIPI/SMP and alike are nearly impossible to debug. I need to be able to change the count of CPUs or their properties. I can't have all the configs of a PC.

I'm just saying that if Windows can boot in both QEMU and on a real machine, then the proper way to make an OS or any boot-able program is to test them on both: PC and an emulator.

That is why I asked

OK, still I'm not convinced it is useful to run a USELESS program in an emulator (I think you have to adapt it even to emulator bugs and this is pointless with a program which is just an example), anyway when I have time I'll try. In the meantime, can you tell me more precisely your configuration and the problem that you have, maybe via e-mail?

I will first test it on some real machines before I bother to look what's wrong with the emulator.

As I see from Dex4u's post, the hwtest works, but I know that he has an older machine so this might be the problem.

Tested on Bochs 2.4.1, runs fine except for APM which doesn't shutdown. I set up only disks (no pci, no usb), both hard disk and floppy work, but you have to set up correctly CHS for the first, and floppy image must be of correct size (take those generated by makecd).
Note that 1589 call was implemented only recently (without it hwtest doesn't start) and that microcode MSR are still not implemented (they shouldn't either, I think), that's why I don't care for APM not working and I don't test more extensively.

updated to beta2, some small bug corrections and documentation changes. Visit
pfranz73.googlepages.com