Interested to hear from other people running the Daystar kernel(s).

By sheer serendipity, or just plain demonic possession of my fingers, I
managed to generate a faulty binary that under a normal 1.4.2 or 1.5.x
kernel, including -current, will generate a Bus error and kill the
process with no harm done. Fine.

Under Daystar-fortified 1.4.2 or 1.5.2 it hangs up the system completely.
(On -current it seems to be a little better; I get a kernel panic "MMU
fault" and get dumped in the debugger. However, I had to hand-patch
locore.s to get this to work so I might have "messed" something up.)

It seems to me that trap.c needs some work, but I don't know where to begin.
So, I'm asking the following:

* Has anyone experienced this before?

* Would another Daystar user be willing to test this binary (64k) to make
  sure it isn't my hardware at fault? Yes, it could shut your system down.
  However, it has never caused any damage, just inconvenience. I want to
  verify my PowerCache isn't subtly defective.

* I went looking through trap.c but am not really sure what portions of
  it might be cache-dependent. Ideas from whiter hats than mine?

The system in question is a Mac IIci running 1.5.2 with John Valdes' recent
patch issue, with 128MB RAM, 2GB disk, and a Daystar PowerCache 50MHz '030
with FPU, with PowerCache and PowerMath turned on in the control panel.

-- 
----------------------------- personal page: http://www.armory.com/~spectre/ --
 Cameron Kaiser, Point Loma Nazarene University * ckaiser@stockholm.ptloma.edu
-- select unique ("Just another SQL hacker") jash from id_rec order by 1; -----