Initial revision

7 files changed, 632 insertions, 0 deletions

diff --git a/linux-nfs/ChangeLog b/linux-nfs/ChangeLog
new file mode 100644
index 0000000..11ffe15
--- /dev/null
+++ b/linux-nfs/ChangeLog
@@ -0,0 +1,78 @@
+
+Release 0.4.11
+
+ *	Added async writes.
+ *	Fixed bug where two rpciod's would be started when insmod'ing
+	both nfs.o and nfsd.o.
+
+Release 0.4.12
+
+ *	Fixed compile problem after renaming some debug macros.
+ *	Improved readdir cache, which can now hold up to 16 (configurable)
+	readdir replies.
+ *	Fixed async write bug(s)
+ *	client file locking now does at least lock/unlock without
+	crashing the machine
+ *	Started to work on NFS swapping
+ *	nfs_get_super no longer requires the file handle passed by
+	mount but does a straight xprt_create_proto().
+ *	TCP reconnect should work now (not yet tested for long disconnect
+	periods, but it does work if you kill and restart nfsd).
+
+Release 0.4.13
+
+ *	More writeback bugs removed.
+ *	Added a modified (and ansified) tirpc rpcgen to get rid of all
+	the warnings in files generated from *.x descriptions. That old Sun
+	code is a real mess.
+ *	Cleaned up nfsd export handling a bit. All syscalls now
+	take dev/ino rather than the pathname.
+ *	Added sysctl interface to set/get debug flags (see tools/rpcdebug).
+ *	Cleaned up Makefiles.
+ *	(experimental) Gathered writes for nfsd (use the wdelay option in
+	/etc/exports).
+ *	Fixed silly bug in nfs_readdir (the in-place decoding of readdir
+	replies requires a temporary buffer).
+ *	Fixed readdir bug in nfsd (long directories were truncated).
+
+Release 0.4.14
+
+ *	Upgraded to kernel 2.0.23
+ *	Fixed bug in rpcdebug
+ *	readdir still didn't work right in nfsd. Argh!
+ *	nfsd would refuse to create symlinks with slashes in them:-)
+ *	nfsd's RPC reply cache should now work again.
+ *	Heavily modified rpc.statd for more robust callback/notify handling
+
+Release 0.4.17
+
+ *	Upgraded to kernel 2.1.14
+ *	Got lockd working with HPUX in most areas.
+
+Release 0.4.19
+
+ *	RPC server UDP sockets now receive the sk_buff directly rather
+	than going through sock->ops->recvmsg.
+	Also got rid of all those cli/sti's and replaced them with
+	disable_bh/enable_bh calls.
+ *	Fixed a bug in nfsd's handling of rename and friends.
+
+Release 0.4.20
+
+ *	Some bugfixes, esp in the writeback code
+ *	Avoid some unnecessary cli/sti pairs
+ *	Added nhfsstone
+
+Release 0.4.21
+
+ *	Minor bugfixes
+ *	Moved to post-2.1.16 module handling code
+
+Release 0.4.22
+
+ *	Fixed a bug that made rpcinfo -u host nlockmgr provoke a kernel
+	oops.
+ *	Upgraded to mount-2.6b
+ *	Added NFSv3 support to mountd and nfsd
+ *	Made sure it compiles with glibc2.
+
diff --git a/linux-nfs/INSTALL b/linux-nfs/INSTALL
new file mode 100644
index 0000000..351f733
--- /dev/null
+++ b/linux-nfs/INSTALL
@@ -0,0 +1,11 @@
+
+Even though the Makefiles offer a `make install' instruction,
+I would suggest against using it yet. I have run nfsd and the nfs
+client without kernel oopses for a while, but the picture may change
+if you start playing with lockd. Automatic installation may not even
+work for the kernel makefiles yet.
+
+I'd therefore advise that you use a separate Linux box for testing
+if you have one. Use the etc/copy script to copy all modules and
+support programs, and run the ins script to start the show. rmm
+will clean up afterwards (provided you didn't trigger an oops).
diff --git a/linux-nfs/KNOWNBUGS b/linux-nfs/KNOWNBUGS
new file mode 100644
index 0000000..b0ecd5c
--- /dev/null
+++ b/linux-nfs/KNOWNBUGS
@@ -0,0 +1,37 @@
+
+nfsd:
+
+ *	We currently keep the inode in the exports struct. This is
+	a bad idea with directories that are intended to be used as
+	a mount point. Must store the file name instead and do a
+	lookup when getfh is called. Yuck!
+
+	Even yuckier: what do we do about exports matching when we
+	can't keep the inode number?
+
+ *	stating a file on remote cdrom returns st_blocks == 0 for some
+	apps.
+
+ *	Should allow multiple exports per dev if one of the directories
+	isn't a subdir of the other.
+
+nfsclnt:
+
+ *	On some occasions, an EAGAIN reported by the transport layer
+	will be propagated to the VFS.
+ *	Some operations do not seem to release the inode properly, so
+	unmounting the device fails.
+
+lockd:
+
+ *	Handle portmap registration in a separate thread. portmap may
+	not be running when we try to mount the first NFS volume (esp.
+	when mounting /usr).
+
+ *	Does not inform rpc.statd when hosts no longer require
+	monitoring; hosts are incorrectly monitored until next system
+	reboot.
+
+exportfs/mountd:
+
+ *	Export handling is reported to do odd things at times.
diff --git a/linux-nfs/NEW b/linux-nfs/NEW
new file mode 100644
index 0000000..43f5c69
--- /dev/null
+++ b/linux-nfs/NEW
@@ -0,0 +1,319 @@
+This is the Linux kernel NFS daemon 1.4.7. It is based on linux-nfs
+0.4.22. It is tested on Linux/alpha and Linux/x86 running glibc 2.1.1.
+
+WARNING: The NFS servers in Linux 2.2 to 2.2.11 are not compatible with
+other NFS client implemenations. If you plan to use Linux 2.2.x as an
+NFS server for non-Linux NFS clients, you should apply the patches
+enlosed here.
+
+linux-2.2.7-sunrpc.patch, nfsd-2.2.7-2.lockd.patch, nfsd-2.2.7-3.patch
+and nfsd-2.2.7-nfsfh.patch are required for Linux 2.2.7 to 2.2.11.
+For other kernel versions, they have to be applied by hand if they are
+still needed.
+
+For Linux 2.2.7 to 2.2.10, nfsd-2.2.7-1.lock.patch is also required.
+
+I made my knfsd package available only because I use it and noone else
+seems to maintain it. But I don't have much time to really work on it.
+I will only fix bugs in the NFS utilities and serious kernel NFS bugs
+which I can duplcate easily. If you have any kernel NFS server problem,
+please report it to the Linux kernel mailing list. If it can be
+reproduced with Linux NFS server and client in less than 5 minutes, you
+can also send me a copy in addition to sending it to the Linux kernel
+mailing list. I may take a look when I have time. However I will collect
+kernel NFS related patches. Contributions are more than welcome.
+
+The NFS lock only works with lockd. Please make sure the portmapper,
+portmap, is started before mounting NFS.
+
+Changes from knfsd 1.4.6:
+
+1. Fix a typo in knfslock.init.
+2. A new kernel patch, nfsd-2.2.7-1.lock.patch, to fix some NFS lock
+   bugs.
+
+Changes from knfsd 1.4.5:
+
+1. Rename /var/lib/nfs/xtab.export to /var/lib/nfs/etab.
+
+Changes from knfsd 1.4.4:
+
+1. Try to fix mountd performance problem by introducing
+   /var/lib/nfs/xtab.export. That is
+   
+  a. "exportfs" reads from /var/lib/nfs/xtab and writes to
+     /var/lib/nfs/xtab.export.
+  b. "mountd" reads from /var/lib/nfs/xtab.export and writes to
+     /var/lib/nfs/xtab.
+
+  The idea is "mountd" doesn't have to read /var/lib/nfs/xtab, which
+  is very expensive.
+
+Changes from knfsd 1.4.3:
+
+1. nfsd-2.2.7-nfsfh.patch, a new kernel patch for filehandle.
+2. nfsd-2.2.7-2.lockd.patch, a new patch for lockd.
+3. Misc bug fixes.
+
+Changes from knfsd 1.4.2:
+
+1. A mountd patch so that the syslog reports unknown requests, and also
+   reports *what* is being (un)mounted, from Piete Brooks
+   <Piete.Brooks@cl.cam.ac.uk>.
+2. Fix knfsd.init for restart.
+3. Add knfslock.init.
+4. knfsd-compat.spec is removed.
+5. nfsd-2.2.7-lockd.patch, a patch to start lockd independent of
+   nfs and nfsd.
+
+Changes from knfsd 1.4.1:
+
+1. Resolve symlink for umount from Piete.Brooks@cl.cam.ac.uk (Piete
+   Brooks)
+2. Fix knfsd.init for statd.
+
+Changes from knfsd 1.4:
+
+1. nfsd-2.2.7-3.patch. This is the only patch you need for Linux 2.2.7
+   to 2.2.10.
+2. Remove
+	nfsd-2.2.5-1.patch
+	nfsd-2.2.5-3.patch
+	nfsd-2.2.8-1.patch
+	nfsd-2.2.7-iget.diff
+	nfsd-2.2.5-nfsfh.diff
+	nfsd-2.2.5-file.patch
+	nfsd-2.2.7-quota.patch
+	nfsd-2.2.7-mknod.patch
+3. Statd update by Jeff Uphoff <juphoff@transmeta.com>.
+4. netgroups patch from Peter Breitenlohner <peb@mppmu.mpg.de>.
+5. Add option checking to exportfs.
+
+Changes from knfsd 1.3.3b:
+
+1. Add linux-2.2.7-sunrpc.patch for a SMP bug in sunrpc.
+2. Add --port/-P to nfsd/mountd, by Jeff Johnson <jbj@redhat.com>.
+3. Add nfsd.8, mountd.8 and statd.man, by Olaf Kirch
+   <okir@monad.swb.de>.
+4. Update nfsstat.man by Olaf Kirch <okir@monad.swb.de>.
+5. Statd fix by Jeff Uphoff <juphoff@transmeta.com>.
+6. Remove knfsd-nok.patch.
+
+Changes from knfsd 1.3.3a:
+
+1. Fix stdin/stdout/stdout handling in mountd.
+2. nfsd-2.2.7-mknod.patch. A patch for mknod.
+3. nfsd-2.2.7-quota.patch. A patch for quota.
+
+Changes from knfsd 1.3.3:
+
+1. Fix hostname matching for wildcard, subnet and netgroup.
+
+Changes from knfsd 1.3.2:
+
+1. Modified mountd to allow clients without IP address to hostname map.
+
+Changes from knfsd 1.3.1a:
+
+1. nfsd-2.2.5-3.patch. This is the only patch you need for Linux 2.2.5.
+
+Changes from knfsd 1.3.1:
+
+1. A patch for knfsd.spec from Markus Linnala <maage@cs.tut.fi>.
+
+Changes from knfsd 1.3a:
+
+1. nfsd-2.2.8-1.patch. This is the only patch you need for Linux 2.2.8.
+2. nfsd-2.2.7-2.patch. This is the only patch you need for Linux 2.2.7.
+
+Changes from knfsd 1.3:
+
+1. Adding "--no-nfs-version 3" to mountd in knfsd.init from RedHat 6.0.
+
+Changes from knfsd 1.2.2a:
+
+1. Updated knfsd.init from RedHat 6.0.
+2. nfsd-2.2.7-1.patch. This is the only patch you need for Linux 2.2.7.
+3. Misc updates from RedHat 6.0.
+
+Changes from knfsd 1.2.2:
+
+1. Make the default NFS server kernel thread to 8 in the rc script.
+
+Changes from knfsd 1.2:
+
+1. Moved knfsd.spec to knfsd-compat.spec.
+2. Update knfsd.spec from knfsd-981204-3.src.rpm.
+3. Fix the squash_[ug]id parsing in /etc/exports from Anders
+   Hammarquist <iko@cd.chalmers.se>.
+4. nfsd-2.2.5-file.patch to clear the bogus bit for MKDIR and SYMLINK.
+5. nfsd-2.2.5-1.patch. A NFS patch based on nfsd-2.2.3-1.patch for
+   Linux 2.2.5.
+
+Changes from knfsd 1.1:
+
+1. Remove
+	cache-2.1.131-1.patch
+	linux-2.1.1xx.diff
+	lock-2.1.131.diff
+	lock-2.1.1xx.diff
+	nfsd-2.1.127-5.patch
+	nullproc-2.1.1xx.diff
+	procfs-2.1.127.patch
+	quota-2.1.1xx.diff
+	root-2.1.1xx.diff
+	socket-2.1.1xx.diff
+	sunrpc-2.1.123-1.patch
+2. locks-2.2.3.diff. A patch for file lock.
+3. nfsd-2.2.3-1.patch. A NFS patch by "G. Allen Morris III"
+(gam3@acm.org).
+
+Changes from knfsd 1.0:
+
+1. Handle broken /var/lib/nfs/rmtab.
+2. Handle lower/upper cases in wildcard hostnames in /etc/exports.
+
+Changes from knfsd-981204:
+
+1. Modify etc/rc.nfsd to check /var/lib/nfs/rmtab during startup.
+2. Add knfsd.spec for RedHat 5. Need nfs-server-2.2beta37-1.1.src.rpm
+   and initscripts-3.78.1-2.src.rpm.
+3. Add support for "make install prefix=...".
+
+Changes from knfsd-981122:
+
+1. Modify etc/rc.nfsd and etc/rc.nfsfs to handle statd during shutdown.
+2. Remove maximum knfsd count checking.
+3. Clean up mountd.
+4. cache-2.1.131-1.patch from G. Allen Morris III (gam3@acm.org).
+5. lock-2.1.131.diff. A nfsd lock patch for Linux 2.1.131.
+
+Changes from knfsd-981113:
+
+1. procfs-2.1.127.patch from G. Allen Morris III (gam3@acm.org).
+2. Modify etc/rc.nfsd and etc/rc.nfsfs to better handle statd.
+3. Fix the sub-mounted directories.
+
+Changes from knfsd-981022:
+
+1. Fix buffer overruns from Peter Benie <pjb1008@cus.cam.ac.uk>.
+2. Fix hostname matching.
+3. Correctly handle dupilcations in /etc/exports.
+4. Add -F flag to statd.
+5. nfsd-2.1.127-5.patch from G. Allen Morris III (gam3@acm.org).
+
+Changes from knfsd-981014:
+
+1. lock-2.1.1xx.diff. A nfsd lock patch.
+2. nullproc-2.1.1xx.diff. Allow any clients to call the nfsd NULL proc.
+3. Add etc/rc.nfsfs to handle statd.
+3. Update etc/rc.nfsd to handle statd.
+4. nfsd-2.1.125-2.patch from G. Allen Morris III (gam3@acm.org).
+5. Fix inet_ntoa usage in statd.
+
+Changes from knfsd-981010:
+
+1. Check client aliases when matching for wildcard client hostnames.
+2. Fix memory leak in mountd.
+3. Fix filename in nfsd-2.1.125-1.patch.
+
+Changes from knfsd-980930:
+
+1. nfsd-2.1.125-1.patch from G. Allen Morris III (gam3@acm.org) and me.
+2. Fix the hostent bugs in mountd and statd.
+3. Remove "kexportfs -au" for "rc.nfsd stop".
+
+Changes from knfsd-980925:
+
+1. socket-2.1.1xx.diff for creating socket on NFS client.
+2. There is a knsfd root_squash patch for Linux 2.1.1xx,
+   root-2.1.1xx.diff. It is only tested on linux 2.1.123.
+   It also fixes the server side 0711 mode bug.
+3. sunrpc-2.1.123-1.patch from Bill Hawes <whawes@transmeta.com>.
+4. nfsd-2.1.122-3.patch from G. Allen Morris III (gam3@acm.org).
+5. Various buffer overrun changes.
+6. Fix mountd to check the duplicated entry in rmtab.
+7. Change exportfs to ignore warnings for "-r".
+8. Fix showmount -e.
+
+Changes from knfsd-980922:
+
+1. nfsd-2.1.121-4.patch from G. Allen Morris III (gam3@acm.org).
+2. Make async as default for export. It matches the user space NFS
+   server.
+
+Changes from knfsd-980920:
+
+1. Add NFS mount version flags to mountd. Change rc.nfsd to disable
+   NFS V3 for mountd.
+2. Fix client hostname.
+3. rc.nfsd runs kexportfs with -r instead of -a for restart and reload.
+
+Changes from knfsd-980915:
+
+1. There is a knsfd quota patch for Linux 2.1.1xx, quota-2.1.1xx.diff.
+   It is only tested on linux 2.1.122.
+2. The submount pathname is removed from the xtab file.
+3. rc.nfsd runs kexportfs with -r instead of -a for start.
+4. Fix kshowmount -e.
+5. Fix hostname matching.
+6. Fix compiling on libc 5.
+
+Changes from knfsd-980910:
+
+1. nfsd-2.1.121-3.patch from G. Allen Morris III (gam3@acm.org).
+2. A new flag, -r, for exportfs.
+3. Don't put an entry in xtab if kernel rejects it.
+4. Use the official hostname when checking if 2 hostnames are the same.
+5. Allow submounts.
+
+It is available at
+
+ftp://ftp.varesearch.com/pub/support/hjl/knfsd/knfsd-1.4.7.tar.gz
+ftp://ftp.varesearch.com/pub/support/hjl/knfsd/knfsd-1.4.6-1.4.7.diff.gz
+ftp://ftp.kernel.org/pub/linux/devel/gcc/knfsd-1.4.7.tar.gz
+ftp://ftp.kernel.org/pub/linux/devel/gcc/knfsd-1.4.6-1.4.7.diff.gz
+
+You have to apply the patch, locks-2.2.3.diff, to the Linux kernel
+first. It fixes quite some file lock bugs. That patch is against Linux
+2.2.3. If your kernel is different, you have to apply it by hand.
+
+nfsd-2.2.3-1.patch is a new knfsd patch against linux 2.2.3 from
+"G. Allen Morris III" (gam3@acm.org). It works for me on x86 and alpha.
+It is needed for the none-Linux NFS clients. You can get Allen's
+current patch from
+
+http://www.CSUA.Berkeley.EDU/~gam3/knfsd/
+
+nfsd-2.2.7-3.patch is based on nfsd-2.2.3-1.patch for Linux 2.2.7. If
+you use Linux 2.2.7, you should apply nfsd-2.2.7-3.patch instead of
+nfsd-2.2.3-1.patch. Please don't use any other patches included here
+for Linux 2.2.7. For other kernel versions, you may have to apply it
+by hand.
+
+nfsd-2.2.7-2.lockd.patch is also necessary to start lockd independent
+of nfs and nfsd.
+
+To compile, just do
+
+# ./configure
+# make
+
+Makefile will try to determine which C library you are using and compile
+this package accordingly.
+
+# make install
+
+will install the knfsd binaries. You have to install a knfs start up
+script by hand. There is a new rc.nfsd in etc. I use it for both the
+user-space nfsd and the kernel nfsd.
+
+There is one RPM spec file, knfs.spec, which is for a Linux system
+based on Linux 2.2 without support for Linux 2.0. You also need
+knfsd-1.4.6.tar.gz to create the knfsd RPMs.
+
+Thanks.
+
+H.J.
+hjl@lucon.org
+08/14/99
diff --git a/linux-nfs/README b/linux-nfs/README
new file mode 100644
index 0000000..b210c61
--- /dev/null
+++ b/linux-nfs/README
@@ -0,0 +1,56 @@
+
+This package contains a greatly revised NFS implementation for Linux
+along with the necessary daemons and utilities. There are still several
+features missing that I'd want to include, and there are some recent
+improvements to the Linux NFS kernel client not reflected here (notably
+the attrtimeo fix).
+
+This thing has become much too large for me to handle all alone anymore.
+Originally, I had planned to have most of the NFS implementation running
+stably by August, so I could start to concentrate more on other jobs that
+are currently in the queue (like updating the NAG). As it turned out, it
+was much more work than I anticipated, and I fell short of my time goal.
+I'm therefore looking for volunteers who would like to work with me on
+finishing this package. Otherwise, this project could end up rusting in
+the corner of some FTP site...
+
+Ideally, I would want to hand over parts of the source tree to other
+hackers to maintain/enhance/etc. But that's not a requirement; if you
+feel you don't have that much time, you can also contribute by picking
+up one of the loose ends and finish what needs to be done (take a look
+at the TODO file...) And then, you can also be plainly a tester.
+
+There's currently a mailing list for lockd development at NRAO
+(lockd-statd@linux.nrao.edu --- mail majordomo@linux.nrao.edu to
+subscribe). If Jeff agress, we could turn this into a general linux-nfs
+mailing list.
+
+
+Hope this covers about what I wanted to say,
+Olaf
+
+------------------------------------------------------------------
+
+
+			SOURCE TREE OVERVIEW
+
+
+
+support/	Support libraries for user-space programs
+
+support/nfs	Generic library for nfsd utilities
+support/export	Manipulation of /etc/exports and /var/lib/nfs/{xtab,rmtab}
+
+utils/		Code for various user-space programs.
+utils/exportfs	Management of nfsd export table.
+utils/mount	Modified mount command to support NFS over TCP.
+utils/mountd	New rpc.mountd for kernel nfsd.
+utils/nfsd	New nfsd (just starts kernel nfsd).
+utils/nfsstat	Pretty-print NFS stats from /proc/net/rpc/nfs*
+utils/rquotad	Marco van Wieringen's rquotad
+utils/showmount	Rick Sladkey's showmount client
+utils/statd	Jeff Uphoff's rpc.statd.
+
+tools/		Support tools for developers/debuggers/testers
+tools/rpcdebug	This one sets/gets the debug flags for each of the kernel
+		modules.
diff --git a/linux-nfs/THANKS b/linux-nfs/THANKS
new file mode 100644
index 0000000..22a80dc
--- /dev/null
+++ b/linux-nfs/THANKS
@@ -0,0 +1,10 @@
+
+	This piece of software owes a lot to all the people who
+	hacked on Linux NFS before me, most notably Rick Sladkey
+	and Donald Becker.
+
+	I also wish to thank Holger Grothe for loaning me a hard
+	disk and a monitor to get my old 486 flying again so I have
+	a decentish test platform.
+
+	Olaf
diff --git a/linux-nfs/TODO b/linux-nfs/TODO
new file mode 100644
index 0000000..3a439d4
--- /dev/null
+++ b/linux-nfs/TODO
@@ -0,0 +1,121 @@
+
+Todo/Status List for Linux-NFS
+
+ *	denotes to be done;
+ o	denotes draft implementation, possibly commented out
+ -	denotes done,
+ +	denotes done and tested
+------------------------------------------------------------------
+
+RPC:
+
+ *	Server-side AUTH_DES authentication
+
+NFS:
+
+ *	stat() calls don't check whether the cached attrs are stil valid
+	(this is a problem in the VFS).
+ -	NFS_ROOT stuff needs fixing.
+ o	Swapping over NFS.
+
+	Issues of swapout:
+	 *	Avoid recursion in low memory situations where
+		kmalloc may call try_to_swap_out etc ad inf.
+	 *	Don't do async I/O on swap files.
+
+	For special-casing related to NFS swap I/O, flag swap file 
+	semantics in inode->i_flags. In swapfile.c, change functions
+	to call readpage/writepage if available, otherwise proceed
+	as usual.
+
+ -	Write-back support.
+ *	Disable page cache invalidation/flushing for locked file
+	regions.
+ -	Directory caching (we now have page-sized dircache entries
+	which could easily be organized into a linked list). These
+	dircache pages come along as a linked list that can be copied
+	almost 1-to-1 into a dirent struct. If this is put into the
+	VFS, other remote fs's will also benefit.
+
+	[Note: I just increased the readdir cache to hold more than
+	one directory. With this, the exclusive lock on readdir goes
+	away, too. With a larger cache, it may also be worth to think
+	about directory readahead...]
+ *	Better lookup caching?
+ *	When a read lock is present, don't time out attr cache or
+	page cache for that region. Likewise, if a write lock is present,
+	be lazy on write-back.
+ *	Implement CTO.
+ -	BUG: Invalidate readdir cache after remove/rename/unlink
+ *	Automatic `mounting' when the server crosses mount points
+	transparently (some IRIX machines seem to do this when
+	using -nohide).
+ *	NFSv3 support. This requires careful design to maximize
+	code sharing between NFSv2 and NFSv3.
+ *	More robust rename handling (see comment before nfs_rename).
+ *	Add Miquel's O_EXCL hack for file creation.
+ *	Performance improvement: When a complete reply is received, and
+	the (async) task is woken up, don't put it on rpciod's scheduling
+	queue, but add it to a `fast scheduler queue.' The fast scheduler
+	could be a special handler that's registered on the tq_scheduler task
+	queue. This queue is fired by the kernel scheduler as soon as
+	the other bottom halves have been run.
+
+	Note that implementing this for sync tasks is even trickier than
+	for async tasks, because you have to make sure you do the right
+	thing in rpc_sleep_on().
+ *	writeback of writable mmaps. Dirty pages are not subject to
+	writeback scheduling. Also, msync should make sure pages are
+	written with O_SYNC on.
+
+
+nfsd:
+
+ *	uid/gid mapping, and rpc.ugidd support
+ -	Don't read/write a file that might have mandatory locks.
+ *	Implement secure/kerberos export options (take care of lockd
+	fopen() calls--most clients seem to use NULL creds for lockd).
+ -	there's a bug in readdir wrt large directories. Try mounting
+	the linux source tree and do an ls on include/linux...
+ *	Support for UNIX socket creation.
+ *	Someone should look over the error return codes. I tend to
+	mix up EPERM and EACCES.
+ *	NFSv3 support.
+ -	Refuse to look up inodes in procfs (security issues).
+ o	Delayed writes (delay syncing of file data when nfsd handles
+	several write requests for the same file concurrently).
+	(Draft - see nfsd_write in fs/nfsd/write.c. Needs benchmarking).
+ *	Faster read operations (single copy): mmap the file region
+	to be read into VM, and pass the VMA to the xdr routines
+	which pass the region's VM address into sock->ops->writemsg.
+	This copies the file data directly from the page cache into
+	the network buffer.
+	Release the vma region after encoding.
+ *	Faster write operations (single copy, with IPv6 net layout):
+	Get the unfragmented UDP datagram, pull the header and
+	do normal processing. Then mmap the file, copy the write data,
+	and release VMA.
+ -	Clear setuid/setgid bit after write().
+ *	Quota support.
+
+lockd:
+
+ *	Server should run on privileged port.
+ *	Testing reclaim support.
+ *	HP lockd accepts our GRANT_MSG callback and passes on the grant
+	to the blocking process, but doesn't reply with a GRANT_RES. 
+	It's not clear to me why it would do this.
+ *	Unregister hosts (SM_UNMON) with rpc.statd when appropriate.
+
+mountd
+
+ *	Unregister service from portmapper upon exit/SIGTERM
+
+mount
+
+ *	If available, use version 3 of the mount protocol and
+	obtain pathconf data (fill in data->bsize).
+
+documentation:
+
+ -	Manpages need to be written