Subject: kern/17357: panic in 1.6B/i386
To: None <>
From: None <>
List: netbsd-bugs
Date: 06/22/2002 00:22:55
>Number:         17357
>Category:       kern
>Synopsis:       panic trying to startx on current/i386
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Jun 21 17:23:01 PDT 2002
>Originator:     dive
>Release:        NetBSD 1.6B
System: NetBSD 1.6A NetBSD 1.6A (EROS) #343: Mon Jun 10 08:14:34 EDT 2002 i386
Architecture: i386
Machine: i386
	Trying to startx causes a panic in XFree86. I got a dump, but could
not savecore it (savecore needs to be able to save from a specified swap
device!) because of the fact that my kernel is set for dumps on wd0b, and
the crashed kernel dumped to wd1b, which is where it should since i have
moved my swap.

uvm_fault(0xe138b8,0x0,0,1) -> e
fatal page fault in supervisor mode
trap type 6 code 0 eip c0237630 cs 8 eflags 10292 cr 237 cp 10
panic: trap
Begin traceback...
trap() at trap+0x20b
--- trap (number 6) ---
com_isa_attach(1f,1f,1f,1f,e80) at com_isa_attach+0xc
End traceback...
syncing discs...
* at this point the machine freezes until I hit the debug keys again and
then I typed t to get a trace. Since I had to write it down by hand I didn't
get the full lines, but I think I got enough info:

perhaps it's a bug in LFS, but I don't think it is. Also, I'm using a USB
mouse, and my LFS partition is only used for source, and has *nothing*
related to X on it, so com_isa_attach and all the lfs things shouldn't even
happen when X starts, right?
boot 1.6B, login as dive, startx
revert to 1.6A, but that's only how i worked around it long enough to send
this pr.
 NetBSD 1.6B (EROS) #347: Fri Jun 21 17:18:00 EDT 2002