Subject: uvm_fault in 15ZC
To: Current Users <current-users@NetBSD.ORG>
From: Jukka Marin <jmarin@pyy.jmp.fi>
List: current-users
Date: 05/12/2002 10:56:21
I have been experiencing random crashes since I upgraded to 15ZC on i386.
I thought my hardware was falling apart, but after building a new system
from scratch, it still seems to happen:

uvm_fault(0xe563aa98, 0x523b000, 0, 1) -> e
fatal page fault in supervisor mode
trap type 6 code 0 eip c020ddbe cs 8 eflags 10206 cr2 523b538 cpl 0
panic: trap
syncing disks... 37 26 21 10 7 1 done

I don't have the kernel debugger enabled now (I did before, but I'm running
X and I never could get to the console display after a crash, so it was of
no use).  Before, I never got a crash dump, so I'm not sure if the panics
have always been the same.

How can I find more information about the crash?  Running

nm <kernel> | sort | more

I get

...
c020dba4 T pool_set_drain_hook
c020dbb8 T pool_get
c020dde8 t pool_do_put
c020e070 T pool_put
c020e088 T pool_prime
c020e188 t pool_prime_page
c020e2c8 t pool_catchup
...

so it seems the crash occurred in pool_get.

What else can I do to help debugging this?

Thanks,

  -jm