On Mon, Jul 02, 2007 at 12:57:07PM -0700, Matthew Dillon wrote: > > Hmmm. Is your real kernel compiled with any odd options? Like > increasing the amount of KVM or something like that?
No it's just a copy of GENERIC with SMP, DDB_UNATTENDED, CPU_ENABLE_EST, and USB_DEBUG added and unneeded drivers removed. > > Hmm. So far I can't reproduce the problem. Lets see if we can get > some more information off the stack, try this patch. If you get > anything that looks like code addresses from the stack dump try > matching them up with symbols from nm kernel.debug. I did a fresh cvs up to rule out any leftover cruft. I tried three times and the stack is the same each time. Joe Fatal trap 12: page fault while in kernel mode mp_lock = 00000001; cpuid = 1 fault virtual address = 0xc fault code = supervisor read, page not present instruction pointer = 0x1f:0xc stack pointer = 0x10:0x5e3628d4 frame pointer = 0x10:0x5e362bac processor eflags = 5e362b98 0000002f 000002a0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 interrupt enabled, resume, IOPL = 0 current process = 5 (sh) current thread = pri 38 (CRIT) <- SMP: XXX kernel: type 12 trap, code=4 CPU1 stopping CPUs: 0x00000001
