The full reclen doesn't include the fragment header, but sk_tcplen does.
Fix this to make it an apples-to-apples comparison.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Soon we want to support multiple fragments, in which case it may be
legal for a single fragment to be smaller than 8 bytes, so we'll want to
delay this check till we've reached the last fragment.

Also fix an outdated comment.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Byte-swapping in place is always a little dubious.

Let's instead define this field to always be big-endian, and do the
swapping on demand where we need it.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Write the client's ip address to any state file and all appropriate
state for that client will be forgotten.
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Controlling the read and write functions allows me to add in "forget
client w.x.y.z", since we won't be limited to reading and writing only
u64 values.
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

I also log basic information that I can figure out about the type of
state (such as number of locks for each client IP address).  This can be
useful for checking that state was actually dropped and later for
checking if the client was able to recover.
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

The eventual goal is to forget state based on ip address, so it makes
sense to call this function in a for-each-client loop until the correct
amount of state is forgotten.  I also use this patch as an opportunity
to rename the forget function from "func()" to "forget()".
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Once I have a client, I can easily use its delegation list rather than
searching the file hash table for delegations to remove.
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Using "forget_n_state()" forces me to implement the code needed to
forget a specific client's openowners.
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

I use the new "forget_n_state()" function to iterate through each client
first when searching for locks.  This may slow down forgetting locks a
little bit, but it implements most of the code needed to forget a
specified client's locks.
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

I added in a generic for-each loop that takes a pass over the client_lru
list for the current net namespace and calls some function.  The next few
patches will update other operations to use this function as well.  A value
of 0 still means "forget everything that is found".
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Each function touches state in some way, so getting the lock earlier
can help simplify code.
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

There were only a small number of functions in this file and since they
all affect stored state I think it makes sense to put them in state.h
instead.  I also dropped most static inline declarations since there are
no callers when fault injection is not enabled.
Signed-off-by: NBryan Schumaker <bjschuma@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Grace time is a part of NFSv4 state engine, which is constructed per network
namespace.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Lease time is a part of NFSv4 state engine, which is constructed per network
namespace.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This is a cleanup patch. Functions nfsd_pool_stats_open() and
nfsd_pool_stats_release() are declared in fs/nfsd/nfsd.h.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Flag in_grace is a part of client tracking state, which is network namesapce
aware. So let'a replace global static variable with per-net one.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Opening and closing of this file is done in client tracking init and exit
operations.
Client tracking is done in network namespace context already. So let's make
this file opened and closed per network context - this will simlify it's
management.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Split NFSv4 state init and shutdown into two different calls: per-net one and
generic one.
Per-net cwinit/shutdown pair have to be called for any namespace, generic pair
- only once on NSFd kthreads start and shutdown respectively.

Refresh of diff-nfsd-call-state-init-twice
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This patch renames nfs4_state_start_net() into nfs4_state_create_net(), where
get_net() now performed.
Also it introduces new nfs4_state_start_net(), which is now responsible for
state creation and initializing all per-net data and which is now called from
nfs4_state_start().
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This patch renames __nfs4_state_shutdown_net() into nfs4_state_shutdown_net(),
__nfs4_state_shutdown() into nfs4_state_shutdown_net() and moves all network
related shutdown operations to nfs4_state_shutdown_net().
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

NFSv4 delegations are stored in global list. But they are nfs4_client
dependent, which is network namespace aware already.
State shutdown and laundromat are done per network namespace as well.
So, delegations unhash have to be done in network namespace context.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This lock protects the client lru list and session hash table, which are
allocated per network namespace already.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Protection of __nfs4_state_shutdown() with nfs4_lock_state() looks redundant.

This function is called by the last NFSd thread on it's exit and state lock
protects actually two functions (del_recall_lru is protected by recall_lock):
1) nfsd4_client_tracking_exit
2) __nfs4_state_shutdown_net

"nfsd4_client_tracking_exit" doesn't require state lock protection, because it's
state can be modified only by tracker callbacks.
Here a re they:
1) create: is called only from nfsd4_proc_compound.
2) remove: is called from either nfsd4_proc_compound or nfs4_laundromat.
3) check: is called only from nfsd4_proc_compound.
4) grace_done; called only from nfs4_laundromat.

nfsd4_proc_compound is called onll by NFSd kthread, which is exiting right
now.
nfs4_laundromat is called by laundry_wq. But laundromat_work was canceled
already.

"__nfs4_state_shutdown_net" also doesn't require state lock protection,
because all NFSd kthreads are dead, and no race can happen with NFSd start,
because "nfsd_up" flag is still set.
Moreover, all Nfsd shutdown is protected with global nfsd_mutex.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

That function is only called under nfsd_mutex: we know that because the
only caller is nfsd_svc, via

        nfsd_svc
          nfsd_startup
            nfs4_state_start
              nfsd4_client_tracking_init
                client_tracking_ops->init == nfsd4_load_reboot_recovery_data

The shared state accessed here includes:

        - user_recovery_dirname: used here, modified only by
          nfs4_reset_recoverydir, which can be verified to only be
          called under nfsd_mutex.
        - filesystem state, protected by i_mutex (handwaving slightly
	  here)
        - rec_file, reclaim_str_hashtbl, reclaim_str_hashtbl_size: other
          than here, used only from code called from nfsd or laundromat
          threads, both of which should be started only after this runs
          (see nfsd_svc) and stopped before this could run again (see
          nfsd_shutdown, called from nfsd_last_thread).
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

The spec requires badname, not inval, in these cases.

Some callers want us to return enoent, but I can see no justification
for that.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Linus has pointed out that indiscriminate use of BUG's can make it
harder to diagnose bugs because they can bring a machine down, often
before we manage to get any useful debugging information to the logs.
(Consider, for example, a BUG() that fires in a workqueue, or while
holding a spinlock).

Most of these BUG's won't do much more than kill an nfsd thread, but it
would still probably be safer to get out the warning without dying.

There's still more of this to do in nfsd/.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Our server rejects compounds containing more than one write operation.
It's unclear whether this is really permitted by the spec; with 4.0,
it's possibly OK, with 4.1 (which has clearer limits on compound
parameters), it's probably not OK.  No client that we're aware of has
ever done this, but in theory it could be useful.

The source of the limitation: we need an array of iovecs to pass to the
write operation.  In the worst case that array of iovecs could have
hundreds of elements (the maximum rwsize divided by the page size), so
it's too big to put on the stack, or in each compound op.  So we instead
keep a single such array in the compound argument.

We fill in that array at the time we decode the xdr operation.

But we decode every op in the compound before executing any of them.  So
once we've used that array we can't decode another write.

If we instead delay filling in that array till the time we actually
perform the write, we can reuse it.

Another option might be to switch to decoding compound ops one at a
time.  I considered doing that, but it has a number of other side
effects, and I'd rather fix just this one problem for now.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

In preparation for moving some of this elsewhere.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

In preparation for moving some of it elsewhere.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

The comment here is totally bogus:
	- OP_WRITE + 1 is RELEASE_LOCKOWNER.  Maybe there was some older
	  version of the spec in which that served as a sort of
	  OP_ILLEGAL?  No idea, but it's clearly wrong now.
	- In any case, I can't see that the spec says anything about
	  what to do if the client sends us less ops than promised.
	  It's clearly nutty client behavior, and we should do
	  whatever's easiest: returning an xdr error (even though it
	  won't be consistent with the error on the last op returned)
	  seems fine to me.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Very embarassing: 1091006c "nfsd: turn
on reply cache for NFSv4" missed a line, effectively leaving the reply
cache off in the v4 case.  I thought I'd tested that, but I guess not.

This time, wrote a pynfs test to confirm it works.

Cc: stable@vger.kernel.org
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This patch moves laundromat_work to nfsd per-net context, thus allowing to run
multiple laundries.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Passing net context looks as overkill.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This patch replaces init_net by SVC_NET(), where possible and also passes
proper context to nested functions where required.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This list holds nfs4 clients (open) stateowner queue for last close replay,
which are network namespace aware. So let's make this list per network
namespace too.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This list holds nfs4 clients queue for lease renewal, which are network
namespace aware. So let's make this list per network namespace too.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This hash holds established sessions state and closely associated with
nfs4_clients info, which are network namespace aware. So let's make it
allocated per network namespace too.

Note: this hash can be allocated in per-net operations. But it looks
better to allocate it on nfsd state start and thus don't waste resources
if server is not running.
Signed-off-by: NStanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>