Date: Mon, 17 Jun 1996 21:35:58 +0300 (EET DST) From: Heikki Suonsivu <hsu@clinet.fi> To: davidg@Root.COM Cc: freebsd-stable@freebsd.org Subject: Re: Trap 12/supervisor read, page not present Message-ID: <199606171835.VAA23212@cantina.clinet.fi> In-Reply-To: David Greenman's message of 17 Jun 1996 03:32:59 %2B0300 References: <199606170022.RAA06364@Root.COM>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <199606170022.RAA06364@Root.COM> David Greenman <davidg@Root.COM> writes:
>At 05:34 AM 6/15/96 -0700, you wrote:
>> This appears to be a "known" bug. I didn't notice in your original message;
>>was this with 2.1R or a recent SUP of -stable?
>
>Both, with -stable sup'd on Friday. I resupped yesterday and got
>rev 1.22.4.4 of vm_map.c, and built a new kernel with that (old one
>was 1.22.4.2).
>
>Now the machine reboots every few hours with nothing at all in the
>dmesg and savecore reports "no core dump." Perhaps there is something
>else I am doing wrong? Bad hardware? My machine isn't pounding nearly
>as hard as, say, ftp.cdrom.com.
...so I'm at a loss to explain your instability problems. It would help if
you could describe the hardware you're using, your kernel configuration, and
the kind of load that is on the machine.
I have four different machines (1 pentium, 3 486 at 100 or 120MHz), which
do this, and the only "special" piece of hardware is cyclades serial board.
But cyclades alone does not wake this up, it only happens when there are
large number of busy modem lines connected to cyclades ports. If I connect
mostly dedicated lines and possibly couple of modems there are no apparent
problems, even though the load on serial ports is probably much higher with
dedicated lines (higher data rate). This might be timing dependent, as I
had lots of trouble with one 100MHz 486 and it became completely unstable
when up to 32 ports were connected, even though the software was
identical (same hardware was relatively stable with FreeBSD 1.1.5.1).
These things do not seem to have any effect:
- both current and stable have the same problem
- disks can be anything, both scsi and ide
- the PC box itself does not seem to have effect
Common for people reporting this problem:
- using kernel ppp (pppd)
These things might have effect, but not always:
- cyclades serial boards (problem with handshake/flow control, maybe, as
dedicated lines do not present this problem?)
--
Heikki Suonsivu, T{ysikuu 10 C 83/02210 Espoo/FINLAND, hsu@clinet.fi
mobile +358-40-5519679 work +358-0-4375360 fax -4555276 home -8031121
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199606171835.VAA23212>