Date: Thu, 23 Oct 1997 00:09:00 -0700 From: Donald Acton <acton@opentext.com> To: freebsd-questions@FreeBSD.ORG Cc: acton@vn.opentext.com Subject: 2.2.5-RELEASE > 64Megabyte generates kernel panic Message-ID: <199710230709.AAA13007@stoner.nsg.bc.ca>
next in thread | raw e-mail | index | archive | help
We got a new computer the other day and also managed to ftp
2.2.5-RELEASE. The specifics of the machine are:
ASUS P2L97 AGP Motherboard with 128Mbytes of 10ns SDRAM DIMMS in 64MB modules
266MHz Pentium II
1 IDE disk, 2940 SCSI controller and 2 SMC 9332bdt Ethernet cards.
The basic problem is that we could not install FreeBSD until we
removed one of the memory modules and got the machine down to
64MBytes. (First we tried taking out all the other hardware before we
took out the memory) No problem, we just figured we would install
FreeBSD, build a kernel with the MAXMEM option and be on our
way. Well, it didn't work. We built the kernel with the MAXMEM option
set to the value specified in the FAQ for 128MByte machines and
rebooted. It worked fine. We installed the addtional 64MB of memory
and rebooted and it just crashed. I turned on the verbose messages and
the crash occurs after all the devices have been probed. The messages
on the screen are:
Device Configuration Finished
Considering FFS root f/s
configure() finished
Fatal trap 12: page fault while in kernel mode
fault virtual address = 0xa0
fault code = supervisor read, page not present
.. instruction, stack and frame pointers
code segement = base 0x0, limit 0xffff, type 0x1b
= DPL 0, pres 1, def 32 1, gran 1
processor flags = interrupt enabled, resume, IOPL = 0
current process = 0
interrupt mask =
panic: page fault
Using the generic kernel the results are the same except the various
pointers are different. Prior to 2.2.5 we had quickly tried the 2.2.2-RELEASE
with similar results. (However, we didn't put a lot of effort into trying
to solve the problem.)
On the chance that one of the DIMM modules was bad we tried swaping
it but that didn't help. We also turned on all the motherboard's
memory diagnostics and they all pass. (The diagnostics appear to be more
than just a simple walk through memory.)
So any ideas on how to solve this problem? Could there be some BIOS
setting we need to change? I noticed that motherboard manual mentions
something called the MIMFD (Management Information Database) that is
stored in flash and has some configuration information in it.
Donald Acton
acton@opentext.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199710230709.AAA13007>