On Fri, May 25, 2001 at 02:06:06PM -0400, Bill Sommerfeld wrote:
... 
> Somehow I think gdb isn't quite doing the right thing with tracebacks
> through trap handlers.
> 
> that said, clearly something in pcmcia_scan_cis stepped on a bad page.
> 
> The in-kernel DDB should give you a better idea of where it faulted.

Thing is, I don't know much about DDB:

kernel: page fault trap, code=0
Stopped in pid 3(cardslot0) at pcmcia_scan_cis+0x25a: movb 0(%edx,%ex,1),%dl

pcmcia_scan_cis(c0b0a200,c03a9d9c,cb31aec8) at pcmcia_scan_cis+0x25a
pcmcia_read_cis(c0b0a200) at pcmcia_read_cis+0xcd
pcmcia_card_attach(c0b0a200) at pmcia_card_attach+0x39
cardslot_event_thread(c0b14280) at cardslot_event_thread+0x364

Now what? BTW I compiled it with PCMCIACISDEBUG & DIAGNOSTIC/DEBUG and
nothing is printed before the panic..

Cheers,

Patrick