Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 4 Feb 1996 07:43:43 +1100 (EST)
From:      Douglas Thomas Crosher  <dtc@scrooge.ee.swin.oz.au>
To:        freebsd-stable@freebsd.org
Subject:   Help: trap fault when swapping
Message-ID:  <199602032043.HAA07606@scrooge.ee.swin.oz.au>

next in thread | raw e-mail | index | archive | help


	Every time I try running large lisp programs on my rather
memory limit PC the freebsd kernels give me trouble. I not sure if it
is my hardware that is perhaps picking up some errors in all the
swapping or a kernel fault.  I doubt it is a memory problem as the
machine had been running well on jobs that worked the memory and
processors hard and long, but without the swapping.

I had been running current, but switched to stable to try and get my
machine stable. So the OS is Freebsd-stable of last nights sup. But
I've now got the following panic from the kernel on the last two
attempts to get a result out of my job.

-=-=-
Fatal trap 12: page fault while in kernel mode
fault virtual address = 0xdeadc10a
			  ^^^^
			quite an appropriate address I thought :)
fault code = supervisor read, page not present
instruction pointer = 0x8: 0xf011287c
code segment = base 0x0, limit 0xfffff, type 0x1b
	= DPL 0, pres 1 def 32 1 gran 1
processor eflags = interrupt enabled, resume IOPL=0
current process = 13192 (top)
interrupt mask = 
kernel: type 12 trap, code = 0
stopped at _sysctl_doproc + 0x64: cmp $0x1, 0x2c(%ebx)
-=-=-=-

I did a 'show map' and got:
Task map 0xf011187c: pmap=0x12c7b80 ref=-1962934272 nentries=251658240 version=5
					^^^^^^^ is this a problem?

	mapentry 0xf0112923: start-0x5376000, end=0xfe95858d, prot=68/0/copy, wired,object-0x5350ffff, offset-0xa8e8.

Then it had a Trap Fault 12...

-=-=-=

>From what I remember of the first panic it stopped with the same fault
at _sysctl_doproc.

The machine is a P90 with P/I-P55TP4 motherboard, 16M ram, and was
swapping off two WD IDE hard disks.

The lisp job was about 40M in size and was using mprotect to help
reduce the garbage collection effort. The lisp program was over nfs if
that is significant.

If it is likely the hardward, do people have machines that really
thrash with SCSI-2 busses with parity checking?

Regards
Douglas Crosher



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199602032043.HAA07606>