Repeated panics...Suspect USB issues

[Josh Tolbert]

Hi,

My 5.3-RELEASE-p5 file server at home has crashed a few times under
the same conditions. I believe the crashes to be USB-related.

The machine itself is a Tyan Tiger LE motherboard, dual P3/733 CPUs,
512M RAM, Intel Pro/1000 MT NIC (64-bit, 66MHz), 2x 80G WD IDE
drives hanging from the on-board Promise IDE RAID controller and a
52x CD-RW. Attached to the machine is an HP LaserJet 6MP printer,
but it's connected through a USB-to-parallel adapter since the Tiger
LE has no parallel port.

I installed CUPS on the machine after connecting the printer to the
machine via the USB-to-parallel adapter (ulpt0: Prolific Technology
Inc. IEEE-1284 Controller, rev 1.00/2.02, addr 2, iclass 7/1) and
rebuilt Samba with support for CUPS so I could share the printer
to my Windows machines.

Related package versions are:

cups-base-1.1.23.0_3
cups-lpr-1.1.23.0
samba-3.0.11,1

Everything worked great, as I've come to expect from several years
of happy FreeBSD use (barring my complaints about the psm driver),
but the machine panics after a few jobs are sent to the 6MP from a
Windows machine. The first few jobs process just fine, then I get
a panic message like this:

Fatal trap 12: page fault while in kernel mode
cupid = 0; apic id = 00
fault virtual address   = 0x0
fault code              = supervisor write, page not present
instruction pointer     = 0x8:0xc047b0cd
stack pointer           = 0x10:0xd5266c68
frame pointer           = 0x10:0xd5266c68
code segment            = base 0x0, limit 0xfffff, type 0x1b,
                        = DPL 0, pres 1, def32 1, gran 1
process eflags          = interrupt enabled, resume, IOPL = 0
current process         = 43 (swi5: clock sio)
trap number             = 12
panic: page fault
cpuid = 0
boot() called on cpu#0
uptime: 10d2h23m44s

The only things that have changed on the machine directly before
the panics started was the addition of the printer/adapter, CUPS
and a rebuild of Samba.  The kernel config is fairly straightforward,
just cleaned out a bit and /etc/make.conf contains no unusual
CFLAGS/COPTFLAGS beyond "-O -pipe".  CPUTYPE?=p3, NO_LPR=true and
CUPS_OVERWRITE_BASE=yes are also in /etc/make.conf.

Since the current process listed in the panic involves sio, and the
only "serial" device I have attached to the machine is the
USB-to-parallel adapter, I'm guessing the problem is USB-related.

Is there any other information I can provide that would help in diagnosis of
the problem? I don't have my kernel/world built with debugging symbols and the
like if necessary.

What can I do to help get this sorted out?

Thanks,
Josh
-- 
Josh Tolbert
hemi at puresimplicity.net  ||  http://www.puresimplicity.net/~hemi/

If your sysadmin's not being fascist, you're paying him too much.
   --Sam Greenfield