Subject: kern/18409: kernel panic during write
To: None <gnats-bugs@gnats.netbsd.org>
From: None <sysadmin@terc.edu>
List: netbsd-bugs
Date: 09/24/2002 18:26:14
>Number:         18409
>Category:       kern
>Synopsis:       kernel panic / uvm_fault during write to ufs
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Sep 24 21:42:00 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator:     Robby Griffin
>Release:        NetBSD 1.6
>Organization:
	TERC
>Environment:
System: NetBSD freshmeat 1.6 NetBSD 1.6 (TERC_SERVER) #0: Fri Sep 20 23:07:48 EDT 2002 root@foiegras:/usr/src/sys/arch/i386/compile/TERC_SERVER i386
Architecture: i386
Machine: i386
>Description:
	While saving a file in emacs:

uvm_fault(0xcaf6fb04, 0x0, 0, 1) -> e
fatal page fault in supervisor mode
trap type 6 code 0 eip c016d155 cs 8 eflags 10283 cr2 38 cpl 0
panic: trap
Begin traceback...
trap() at trap+0x202
--- trap (number 6) ---
VOP_PUTPAGES(0,0,0,0,0) at VOP_PUTPAGES+0x3d
end(cb2dabd4,cb3bfa78,0,0,cb3bfa78) at 0xcb299868
VOP_PUTPAGES(cb3bfa78,0,0,0,0) at VOP_PUTPAGES+0x49
vinvalbuf(cb3bfa78,1,ffffffff,cb2b956c,0,0,caf8e900,caf8ea0c) at vinvalbuf+0x28
vclean(cb3bfa78,8,cb2b956c,caf6a564) at vclean+0x80
vgonel(cb3bfa78,cb2b956c,1,0,c05de500) at vgonel+0x46
vrecycle(cb3bfa78,0,cb2b956c,c0144bcf,817af48) at vrecycle+0x20
end(0,c05ed000,20,cb2b956c,cb2b956c) at 0xcb2918ce
end(0,c05ed000,20,cb2b956c,cb2dae4c) at 0xcb293147
end(1d00,cb2daee0,1,c01715c9,cb284b90) at 0xcb2929ca
spec_write(cb2dae4c,0,4,c01e2480,cb284b90) at spec_write+0x70
ufsspec_write(cb2dae4c,30002,0,c0150fd2,20) at ufsspec_write+0x33
VOP_WRITE(cb284b90,cb2daee0,1,c0522f00,cb2daf80) at VOP_WRITE+0x3b
vn_write(cb2db000,cb2db028,cb2daee0,c0522f00,1) at vn_write+0x9f
dofilewrite(cb2b956c,6,cb2db000,817af48,20) at dofilewrite+0x9b
sys_write(cb2b956c,cb2daf80,cb2daf78,c0548038) at sys_write+0x46
syscall_plain(bfbf001f,80a001f,813001f,814001f,80aa700) at syscall_plain+0xa7
End traceback...

	Kernel, core, config and dmesg available upon request. This is a
scratch machine for testing 1.6, so I'm open to suggestions. Bits of
info requested so far:

	options SOFTDEP? yes.
	possible flaky hardware? possible, but no previous manifestation.
	anything else going on? arla /afs mounted, setting up sshd/krb5.

>How-To-Repeat:
	not sure, only seen this once.
>Fix:
	unknown.
>Release-Note:
>Audit-Trail:
>Unformatted: