vmunix: zs0: parity error ignored (summ

[Geoffrey H. Cooper]

OK, not many responses on this one, but I got tired and tried my fix.  16
hours later, everything seems fine, and I don't get these messages
anymore.  Obviously, I am not giving a warranty!

The problem is a plethora of random messages from one of the serial ports.
Sun customer support reports that this condition was always happening, but
that the message wasn't getting logged until 4.1.  The message looks like:

 Jul 28 05:00:14 aurora vmunix: zs0: parity error ignored
 Jul 28 05:00:43 aurora vmunix: zs0: parity error ignored
 Jul 28 05:00:43 aurora last message repeated 2 times
 Jul 28 06:05:37 aurora vmunix: zs0: parity error ignored

and so on.  Zs0 means port A and Zs1 means port B (I get it only on zs0,
but others report zs1).  Well, if the only difference is that the message
is being logged now, this is something I can fix!

Upon perusal of my /VMUNIX with my trusty ADB:

 _zsa_srint+0x70:                sethi   %hi(0xf80f7c00), %o1
 _zsa_srint+0x74:                or      %o1, 0x153, %o1          ! -0x7f082ad
 _zsa_srint+0x78:                and     %o2, 0x7f, %o2
 _zsa_srint+0x7c:                call    _log

 -0x7f082ad?s
 _zsops_async+0x2f:              zs%d: parity error ignored

STRINGS confirms that there is only one string of the above form in the
kernel.

I did this:

   # cp /vmunix /vmunix.save
   # adb -w /vmunix
   _zsa_srint+0x7c?W 0x1000000
	...xxxxxx = 01000000...
   _zsa_srint+0x7c?i
   _zsa_srint+0x7c:     nop
   ^D
   # /etc/reboot

[note: ADB patches right away, so if you blow it, revert to vmunix.save
and repeat the entire procedure.  If the system won't boot, boot sd()-a,
respond with returns to everything except the BOOT: message, where you
type "vmunix.save"]

geof at aurora.com / aurora!geof at decwrl.dec.com / geof%aurora.com at decwrl.dec.com

X-From: sjk at nil.UUCP (Scott J. Kramer)

We've got the same lossage on our Sun-3/180, SunOS 4.1 system with Telebit
Trailblazer+ and Supra 2400 modems on the a & b serial ports.  What's
puzzling to me is why it also happens on zs1:

    Jul 25 04:42:02 pixar vmunix: zs0: parity error ignored
    Jul 25 05:53:47 pixar vmunix: zs1: parity error ignored
    Jul 25 06:32:22 pixar vmunix: zs1: parity error went up

and that "... went up" message is peculiar, too.

Another problem is that the Supra modem on the b port (currently the
first line in a 4-number dialup rotary (we've got two other Supra's on
a CoALM board)) goes into limbo about once a day and requires
power-cycling to clear it; "ATZ"ing it is futile.  This prevents
anyone from dialing in until it's reset since our UUCP incallers only
use that first number.  I've unsuccessfully tried several combinations
of modems and ports; I still can't tell whether it's a hardware and/or
software problem.  A few of the incoming/outgoing calls are at 1200bps
and I suspect that the trouble is related to the modems and/or serial
lines not properly resetting.

At first I thought the "modem in limbo" problem was related to the
parity error stuff, but there doesn't seem to be any correlation.

So I can't tell you what those kernel messages really mean (sure is a
loss not having system source!), tho' I'm anxious to find out!

Scott J. Kramer			inet:	pixar!sjk at ucbvax.Berkeley.EDU
Pixar				uucp:	...!{sun,ucbvax}!pixar!sjk
1001 West Cutting Boulevard	voice:	415-236-4000
Richmond, CA, 94804
USA
-- 
geof at aurora.com / aurora!geof at decwrl.dec.com / geof%aurora.com at decwrl.dec.com