linux-toradex.git/include/linux/nfs_fs_sb.h, branch v3.4.16

NFS: Request fh_expire_type attribute in "server caps" operation

2012-03-02T22:18:10+00:00

The fh_expire_type file attribute is a filesystem wide attribute that
consists of flags that indicate what characteristics file handles
on this FSID have.

Our client doesn't support volatile file handles.  It should find
out early (say, at mount time) whether the server is going to play
shenanighans with file handles during a migration.

Signed-off-by: Chuck Lever 
Signed-off-by: Trond Myklebust

NFSv4: parse and display server implementation ids

2012-03-01T22:10:22+00:00

Shows the implementation ids in /proc/self/mountstats.  This doesn't break
the nfs-utils mountstats tool.

Signed-off-by: Weston Andros Adamson 
Signed-off-by: Trond Myklebust

NFSv4.1: Add a module parameter to set the number of session slots

2012-02-15T05:19:44+00:00

Add the module parameter 'max_session_slots' to set the initial number
of slots that the NFSv4.1 client will attempt to negotiate with the
server.

Signed-off-by: Trond Myklebust

NFSv4.1: Convert slotid from u8 to u32

2012-02-15T05:19:43+00:00

It is perfectly legal to negotiate up to 2^32-1 slots in the protocol,
and with 10GigE, we are already seeing that 255 slots is far too limiting.

Signed-off-by: Trond Myklebust

SUNRPC: Change the default limit to the number of TCP slots

2012-02-06T23:47:35+00:00

Since the scheme of limiting the number of TCP slots to whatever will
fit in the current TCP window seems to be working well (Andy reports
getting within 20% of the 'iperf' send performance on a 10GigE link)
we should just let that be the default mode of operation.

Users may still set their own limits using the tcp_max_slot_table_entries
parameter if they need to.

Signed-off-by: Trond Myklebust

NFSv4: Replace lock_owner->ld_id with an ida based allocator

2012-01-31T23:20:28+00:00

Again, We're unlikely to ever need more than 2^31 simultaneous lock
owners, so let's replace the custom allocator.

Now that there are no more users, we can also get rid of the custom
allocator code.

Signed-off-by: Trond Myklebust

NFSv4: Replace state_owner->so_owner_id with an ida based allocator

2012-01-31T23:20:28+00:00

We're unlikely to ever need more than 2^31 simultaneous open owners,
so let's replace the custom allocator with the generic ida allocator.

Signed-off-by: Trond Myklebust

NFS: make NFS client allocated per network namespace context

2012-01-31T23:20:26+00:00

This patch adds new net variable to nfs_client structure. This variable is set
on NFS client creation and cheched during matching NFS client search.
Initially current->nsproxy->net_ns is used as network namespace owner for new
NFS client to create. This network namespace pointer is set during mount
options parsing and thus can be passed from user-spave utils in future if will
be necessary.

Signed-off-by: Stanislav Kinsbursky 
Signed-off-by: Trond Myklebust

NFS: Cache state owners after files are closed

2012-01-05T16:59:18+00:00

Servers have a finite amount of memory to store NFSv4 open and lock
owners.  Moreover, servers may have a difficult time determining when
they can reap their state owner table, thanks to gray areas in the
NFSv4 protocol specification.  Thus clients should be careful to reuse
state owners when possible.

Currently Linux is not too careful.  When a user has closed all her
files on one mount point, the state owner's reference count goes to
zero, and it is released.  The next OPEN allocates a new one.  A
workload that serially opens and closes files can run through a large
number of open owners this way.

When a state owner's reference count goes to zero, slap it onto a free
list for that nfs_server, with an expiry time.  Garbage collect before
looking for a state owner.  This makes state owners for active users
available for re-use.

Now that there can be unused state owners remaining at umount time,
purge the state owner free list when a server is destroyed.  Also be
sure not to reclaim unused state owners during state recovery.

This change has benefits for the client as well.  For some workloads,
this approach drops the number of OPEN_CONFIRM calls from the same as
the number of OPEN calls, down to just one.  This reduces wire traffic
and thus open(2) latency.  Before this patch, untarring a kernel
source tarball shows the OPEN_CONFIRM call counter steadily increasing
through the test.  With the patch, the OPEN_CONFIRM count remains at 1
throughout the entire untar.

As long as the expiry time is kept short, I don't think garbage
collection should be terribly expensive, although it does bounce the
clp->cl_lock around a bit.

[ At some point we should rationalize the use of the nfs_server
->destroy method. ]

Signed-off-by: Chuck Lever 
[Trond: Fixed a garbage collection race and a few efficiency issues]
Signed-off-by: Trond Myklebust

NFS: Re-enable compilation of nfs with !CONFIG_NFS_V4 || !CONFIG_NFS_V4_1

2011-08-01T00:27:04+00:00

Fix two recently introduced compile problems:

Fix a typo in fs/nfs/pnfs.h

Move the pnfs_blksize declaration outside the CONFIG_NFS_V4 section in
struct nfs_server.

Reported-by: Jens Axboe 
Signed-off-by: Trond Myklebust 
Signed-off-by: Linus Torvalds