[ENBD] 2.4.21 dual athlon 1GB ram enbd problem
Peter T. Breuer
enbd@lists.community.tummy.com
Thu, 7 Aug 2003 04:56:58 +0200 (MET DST)
"A month of sundays ago Keith Lewis wrote:"
> I wrote:
> > # insmod /tmp/enbd.o
> > /tmp/enbd.o: unresolved symbol kunmap_high
> > /tmp/enbd.o: unresolved symbol highmem_start_page
> > /tmp/enbd.o: unresolved symbol kmap_high
>
> And I received lots of good advice and help.
>
> Suggestion 1 was to put the module in /lib/modules... and depmod.
> That didn't actually solve teh problem.
That's sort of a relief. To me it looks as though the kernel is
compiled without highmem support and the module with, but that's only a
guess.
> Suggestion 2 was to build new kernels without highmem support. Will
> try that on the original box as soon as I can reboot it. In the mean time I
> compiled a 2.4.21 kernel on a smaller box and verified that the enbd module
> installs fine. Great!
>
> Suggestion 3 was to fetch a new copy of 2.4.31 which has stuff removed
> (replaced by fr1). I'd be fetching it now except the network here seems to be
> down... First thing tomorrow.
You should also simply be able to remove the offending calls from your
present source code. They don't impact functionality that you will
probably be using. AFAICS the calls are associated with the raw device
interface, and possibly the functionality that allows writes to be
intercepted and buffered in mmemory instead of passing through to the
remote.
Peter