NCPFS patches which should go into before 2.2

Petr Vandrovec Ing. VTEI (
Wed, 4 Mar 1998 12:23:22 MET-1

I'm maintaining couple of patches for ncpfs which could be usable for
other ncpfs peoples. And because of several peoples asked me to post
them to Linus, I asked Volker (current ncpfs maintainer) wheather he
lend me ncpfs maintenance... I'll see how he decide.
In the meantime I entreat peoples using ncpfs, especially on non-Intel
architectures because of neither me nor my current beta-testers have
access to non-Intel machine, wheather they could apply kernel patch
available on
(also patch for 2.0.33 is available)
and updated set of userspace utilities available from
and test wheather they can connect to NDS and wheather they can utilize
NFS namespace on servers. If you enable NDS support in ncpfs, you have
to modify your ncpmount commands to add "-b" option to ncpmount commandline
or change username from bindery to NDS form (without leading dot and without
CN=, O=, OU=, C=).
These kernel patches allows you to configure more precisely kernel ncpfs
support - to get state of code as without patches, set
Packet signatures to OFF
Proprietary file locking to OFF
Clear remove/delete inhibit to OFF
Use NFS namespace if available to OFF
Use LONG (OS/2) namespace if available to ON
Allow mounting of volume subdirectories to OFF

There are following known problems:
- You cannot connect with packet signatures in bindery mode to MOAB
(moab bug, VLM cannot do it too)
- You will not see contents of NSS volumes on MOAB if using NFS namespace
(NSS incompatibility, native NetWare volumes are OK)
- You cannot connect to MOAB through native IP
(nobody wrote that)

I'm especially interested in
- non-Intel machines (there will be (probably) unaligned memory accesses)
- Does NDS support work with NetWare 4.0, 4.01, 4.02, IntraNetware for SB
Thanks for your time & input
Petr Vandrovec

P.S.: If you expect fast reaction, CC me, I read kernel-digest only.
P.P.S.: Linus, if you read it: it is 42KB patch, should I split it down
into smaller pieces? But I think that it is better do it in one step
because of all is configurable and can be easy disabled.

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to