John Klos wrote:
> 
> ex* has had issues, but I've now seen two panics on netbsd-2 in less 
> than a month:
> 
> panic: tcp_output REXMT
> Begin traceback...
> 0x0042dec0: at tcp_setpersist+0xc8
> 0x0042dee0: at tcp_timer_persist+0x98
> 0x0042df00: at softclock+0x248
> 0x0042df40: at hardclock+0x314
> 0x0042df70: at decr_intr+0xfc
> 0x0042dfa0: at trapstart+0xaec
> 0xffffe750: at ADBDevTable+0x4bb8b388
> trap: pid 362.1 (setiathome): kernel MCHK trap @ 0x2d6d84 (SRR1=0x49030)
> panic: trap
> Faulted in mid-traceback; aborting...dumpsys: TBD
> rebooting
> 
> It might be that this is a slower system (180 or 200 MHz 603e), so it's 
> more apt to get caught in some place where an interrupt isn't properly 
> handled. Of course, this kind of stuff never happens locally - it only 
> ever happens on my colocated machines! So it's hard to do any real 
> checking.
> 
> Ideas, anyone?
> 
> John
> 

Can you give this a try:

http://mail-index.netbsd.org/port-powerpc/2005/06/28/0000.html

The MPC8272 also has a 603e core, and this is exactly the sort
of nonsense backtrace I was seeing.  Nobody ever offered a second
opinion as to whether this was a fix or a workaround, but the
symptoms have not returned.

-Neil