On Sun, Sep 28, 2003 at 05:10:39PM -0700, Chuck Silvers wrote:
> well, I tried a bunch of tests this this afternoon and I couldn't get
> anything to fail with SIGILLs or anything like that.  could you guys
> give some more details on exactly what goes wrong and what the machine
> is doing when it happens?

I finally managed to reproduce this without X running (by using window(1)
from the console and doing a Mozilla build in one sub-window, and a kernel
build in the other). What's more, it's reproducable with a
DIAGNOSTIC/DEBUG enabled kernel, and I got a panic last night - here are
the details (copied by hand, unfortunately, as the machine wedged after
printing the backtrace):

diagnostic assert "pcb->pcb_kmapsr == 0" failed
.../sys/arch/powerpc/powerpc/trap.c" line 490

Traceback:
0xd59addb0: at __assert+28
0xd59addc0: at copyout+134
0xd59ade50: at sys___fstat13+d4
0xd59adee0: at syscall_plain+e0
0xd59adf40: user SC trap #279 by 0x41c3db20: srrl=0xd032
   rl=0xffffdc80 cr=0x44000048 xer=0 ctr=0x41c3db18

(I scribbled all this down in a bit of a hurry last night, but I think
most of it is right)

Ian.