With kernel 2.2.9, I got the following messages:
Nov 23 20:33:07 linuix kernel: try_to_unuse: entry 000e4200 count=1
Nov 23 20:33:07 linuix kernel: rw_swap_page: Trying to read unallocated swap (000e4200)
Nov 23 20:33:07 linuix kernel: swap_duplicate at c01207f6: entry 000e4b00, unused page
Most processes that I then started ("ps", "ls -l /tmp", "ipcs", ...) got
stuck in an interruptible state. Eventually I rebooted.
==> What I had been doing:
Under X, start clisp under gdb and, every 3 seconds, start a new process
doing "ls -l /tmp/foobar". Simultaneously, two gcc crosscompilations.
Let this run for 10 hours. Then the swap space looked like this:
Filename Type Size Used Priority
/dev/bmem partition 24572 24368 -1
/dev/hdb5 partition 50364 9800 -2
/dev/hda6 partition 50048 0 -3
(/dev/bmem is a fast, RAM based swap space.) Then I did "swapoff /dev/bmem".
It took around two minutes, and effectively moved the used swap pages from
/dev/bmem to /dev/hdb5. Then I did "swapon -p 1 /dev/bmem". The fast swap
space was back with highest priority, but the system still fest sluggish.
A few minutes later, the kernel problem appeared.
Btw, my X server uses ca. 3 MB of shared memory, so shared memory swapping
might be involved too.
==> Hardware:
CPU: 486,
Disks: 2 IDE disks from WD (no problems ever, last change one year ago)
RAM: 16 MB (used reliably for 2 years)
/dev/bmem: 24 MB uncachable RAM (used reliably for 2 years too)
Bruno
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/