nfsd4_process_open2 will currently will get access to the file, and then
call nfsd4_truncate to (possibly) truncate it. If that operation fails
though, then the access references will never be released as the
nfs4_ol_stateid is never initialized.

Fix by moving the nfsd4_truncate call into nfs4_get_vfs_file, ensuring
that the refcounts are properly put if the truncate fails.
Signed-off-by: NJeff Layton <jlayton@primarydata.com>
Signed-off-by: NChristoph Hellwig <hch@infradead.org>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

sparse complains that we're stuffing non-byte-swapped values into
__be32's here. Since they're supposed to be opaque, it doesn't matter
much. Just add __force to make sparse happy.
Signed-off-by: NJeff Layton <jlayton@primarydata.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

If nfsd needs to recall a delegation for some reason it implies that there is
contention on the file, so further delegations should not be handed out.

The current code fails to do so, and the result is effectively a
live-lock under some workloads: a client attempting a conflicting
operation on a read-delegated file receives NFS4ERR_DELAY and retries
the operation, but by the time it retries the server may already have
given out another delegation.

We could simply avoid delegations for (say) 30 seconds after any recall, but
this is probably too heavy handed.

We could keep a list of inodes (or inode numbers or filehandles) for recalled
delegations, but that requires memory allocation and searching.

The approach taken here is to use a bloom filter to record the filehandles
which are currently blocked from delegation, and to accept the cost of a few
false positives.

We have 2 bloom filters, each of which is valid for 30 seconds.   When a
delegation is recalled the filehandle is added to one filter and will remain
disabled for between 30 and 60 seconds.

We keep a count of the number of filehandles that have been added, so when
that count is zero we can bypass all other tests.

The bloom filters have 256 bits and 3 hash functions.  This should allow a
couple of dozen blocked  filehandles with minimal false positives.  If many
more filehandles are all blocked at once, behaviour will degrade towards
rejecting all delegations for between 30 and 60 seconds, then resetting and
allowing new delegations.
Signed-off-by: NNeilBrown <neilb@suse.de>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

27b11428 ("nfsd4: remove lockowner when removing lock stateid")
introduced a memory leak.

Cc: stable@vger.kernel.org
Reported-by: NJeff Layton <jeff.layton@primarydata.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

We don't want the stateid to be found in the hash table before the delegation
is granted.

Currently this is protected by the client_mutex, but we want to break that
up and this is a necessary step toward that goal.
Signed-off-by: NBenny Halevy <bhalevy@primarydata.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

...as the name is a bit more descriptive and we've started using it for
other purposes.
Signed-off-by: NBenny Halevy <bhalevy@primarydata.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This fixes a bug in the handling of the fi_delegations list.

nfs4_setlease does not hold the recall_lock when adding to it. The
client_mutex is held, which prevents against concurrent list changes,
but nfsd_break_deleg_cb does not hold while walking it. New delegations
could theoretically creep onto the list while we're walking it there.
Signed-off-by: NBenny Halevy <bhalevy@primarydata.com>
Signed-off-by: NJeff Layton <jlayton@primarydata.com>
Cc: stable@vger.kernel.org
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

The laundromat uses two variables to calculate when it should next run,
but one is completely ignored at the end of the run. Merge the two and
rename the variable to be more descriptive of what it does.
Signed-off-by: NJeff Layton <jlayton@primarydata.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

RPC_MAX_AUTH_SIZE is scattered around several places.  Better to set it
once in the auth code, where this kind of estimate should be made.  And
while we're at it we can leave it zero when we're not using krb5i or
krb5p.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Once we know the limits the session places on the size of the rpc, we
can also use that information to release any unnecessary reserved reply
buffer space.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

We can simplify session limit enforcement by restricting the xdr buflen
to the session size.

Also fix a preexisting bug: we should really have been taking into
account the auth-required space when comparing against session limits,
which are limits on the size of the entire rpc reply, including any krb5
overhead.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Limits on maxresp_sz mean that we only ever need to replay rpc's that
are contained entirely in the head.

The one exception is very small zero-copy reads.  That's an odd corner
case as clients wouldn't normally ask those to be cached.

in any case, this seems a little more robust.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

xdr_reserve_space should now be calculating the length correctly as we
go, so there's no longer any need to fix it up here.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This is a mechanical transformation with no change in behavior.
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

No need for a kmem_cache_destroy wrapper in nfsd, just do proper
goto based unwinding.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

We're not cleaning up everything we need to on error.  In particular,
we're not removing our lease.  Among other problems this can cause the
struct nfs4_file used as fl_owner to be referenced after it has been
destroyed.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

The current code assumes a one-to-one lockowner<->lock stateid
correspondance.

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

The nfsv4 state code has always assumed a one-to-one correspondance
between lock stateid's and lockowners even if it appears not to in some
places.

We may actually change that, but for now when FREE_STATEID releases a
lock stateid it also needs to release the parent lockowner.

Symptoms were a subsequent LOCK crashing in find_lockowner_str when it
calls same_lockowner_ino on a lockowner that unexpectedly has an empty
so_stateids list.

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

Signed-off-by: NKinglong Mee <kinglongmee@gmail.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Move the state locking and file descriptor reference out from the
callers and into nfs4_preprocess_stateid_op() itself.
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

They do not need to be used outside fs/nfsd/nfs4state.c
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

It is large, it is used in more than one place, and it is not performance
critical. Let gcc figure out whether it should be inlined...
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Mainly to ensure that we don't leave any hanging timers.
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>
Cc: stable@vger.kernel.org
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Aside from making it clearer what is non-trivial in create_client(), it
also fixes a bug whereby we can call free_client() before idr_init()
has been called.
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>
Cc: stable@vger.kernel.org
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

RFC5661 obsoletes NFS4ERR_STALE_STATEID in favour of NFS4ERR_BAD_STATEID.

Note that because nfsd encodes the clientid boot time in the stateid, we
can hit this error case in certain scenarios where the Linux client
state management thread exits early, before it has finished recovering
all state.
Reported-by: NIdan Kedar <idank@primarydata.com>
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Maybe this is comment true, who cares?  Handle this like any other
error.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

This isn't actually used anywhere else.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

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

Connection from alloc_conn must be freed through free_conn,
otherwise, the reference of svc_xprt will never be put.
Signed-off-by: NKinglong Mee <kinglongmee@gmail.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

When stopping nfsd, I got BUG messages, and soft lockup messages,
The problem is cuased by double rb_erase() in nfs4_state_destroy_net()
and destroy_client().

This patch just let nfsd traversing unconfirmed client through
hash-table instead of rbtree.

[ 2325.021995] BUG: unable to handle kernel NULL pointer dereference at
          (null)
[ 2325.022809] IP: [<ffffffff8133c18c>] rb_erase+0x14c/0x390
[ 2325.022982] PGD 7a91b067 PUD 7a33d067 PMD 0
[ 2325.022982] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC
[ 2325.022982] Modules linked in: nfsd(OF) cfg80211 rfkill bridge stp
llc snd_intel8x0 snd_ac97_codec ac97_bus auth_rpcgss nfs_acl serio_raw
e1000 i2c_piix4 ppdev snd_pcm snd_timer lockd pcspkr joydev parport_pc
snd parport i2c_core soundcore microcode sunrpc ata_generic pata_acpi
[last unloaded: nfsd]
[ 2325.022982] CPU: 1 PID: 2123 Comm: nfsd Tainted: GF          O
3.14.0-rc8+ #2
[ 2325.022982] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS
VirtualBox 12/01/2006
[ 2325.022982] task: ffff88007b384800 ti: ffff8800797f6000 task.ti:
ffff8800797f6000
[ 2325.022982] RIP: 0010:[<ffffffff8133c18c>]  [<ffffffff8133c18c>]
rb_erase+0x14c/0x390
[ 2325.022982] RSP: 0018:ffff8800797f7d98  EFLAGS: 00010246
[ 2325.022982] RAX: ffff880079c1f010 RBX: ffff880079f4c828 RCX:
0000000000000000
[ 2325.022982] RDX: 0000000000000000 RSI: ffff880079bcb070 RDI:
ffff880079f4c810
[ 2325.022982] RBP: ffff8800797f7d98 R08: 0000000000000000 R09:
ffff88007964fc70
[ 2325.022982] R10: 0000000000000000 R11: 0000000000000400 R12:
ffff880079f4c800
[ 2325.022982] R13: ffff880079bcb000 R14: ffff8800797f7da8 R15:
ffff880079f4c860
[ 2325.022982] FS:  0000000000000000(0000) GS:ffff88007f900000(0000)
knlGS:0000000000000000
[ 2325.022982] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 2325.022982] CR2: 0000000000000000 CR3: 000000007a3ef000 CR4:
00000000000006e0
[ 2325.022982] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[ 2325.022982] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
0000000000000400
[ 2325.022982] Stack:
[ 2325.022982]  ffff8800797f7de0 ffffffffa0191c6e ffff8800797f7da8
ffff8800797f7da8
[ 2325.022982]  ffff880079f4c810 ffff880079bcb000 ffffffff81cc26c0
ffff880079c1f010
[ 2325.022982]  ffff880079bcb070 ffff8800797f7e28 ffffffffa01977f2
ffff8800797f7df0
[ 2325.022982] Call Trace:
[ 2325.022982]  [<ffffffffa0191c6e>] destroy_client+0x32e/0x3b0 [nfsd]
[ 2325.022982]  [<ffffffffa01977f2>] nfs4_state_shutdown_net+0x1a2/0x220
[nfsd]
[ 2325.022982]  [<ffffffffa01700b8>] nfsd_shutdown_net+0x38/0x70 [nfsd]
[ 2325.022982]  [<ffffffffa017013e>] nfsd_last_thread+0x4e/0x80 [nfsd]
[ 2325.022982]  [<ffffffffa001f1eb>] svc_shutdown_net+0x2b/0x30 [sunrpc]
[ 2325.022982]  [<ffffffffa017064b>] nfsd_destroy+0x5b/0x80 [nfsd]
[ 2325.022982]  [<ffffffffa0170773>] nfsd+0x103/0x130 [nfsd]
[ 2325.022982]  [<ffffffffa0170670>] ? nfsd_destroy+0x80/0x80 [nfsd]
[ 2325.022982]  [<ffffffff810a8232>] kthread+0xd2/0xf0
[ 2325.022982]  [<ffffffff810a8160>] ? insert_kthread_work+0x40/0x40
[ 2325.022982]  [<ffffffff816c493c>] ret_from_fork+0x7c/0xb0
[ 2325.022982]  [<ffffffff810a8160>] ? insert_kthread_work+0x40/0x40
[ 2325.022982] Code: 48 83 e1 fc 48 89 10 0f 84 02 01 00 00 48 3b 41 10
0f 84 08 01 00 00 48 89 51 08 48 89 fa e9 74 ff ff ff 0f 1f 40 00 48 8b
50 10 <f6> 02 01 0f 84 93 00 00 00 48 8b 7a 10 48 85 ff 74 05 f6 07 01
[ 2325.022982] RIP  [<ffffffff8133c18c>] rb_erase+0x14c/0x390
[ 2325.022982]  RSP <ffff8800797f7d98>
[ 2325.022982] CR2: 0000000000000000
[ 2325.022982] ---[ end trace 28c27ed011655e57 ]---

[  228.064071] BUG: soft lockup - CPU#0 stuck for 22s! [nfsd:558]
[  228.064428] Modules linked in: ip6t_rpfilter ip6t_REJECT cfg80211
xt_conntrack rfkill ebtable_nat ebtable_broute bridge stp llc
ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6
nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw
ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4
nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security
iptable_raw nfsd(OF) auth_rpcgss nfs_acl lockd snd_intel8x0
snd_ac97_codec ac97_bus joydev snd_pcm snd_timer e1000 sunrpc snd ppdev
parport_pc serio_raw pcspkr i2c_piix4 microcode parport soundcore
i2c_core ata_generic pata_acpi
[  228.064539] CPU: 0 PID: 558 Comm: nfsd Tainted: GF          O
3.14.0-rc8+ #2
[  228.064539] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS
VirtualBox 12/01/2006
[  228.064539] task: ffff880076adec00 ti: ffff880074616000 task.ti:
ffff880074616000
[  228.064539] RIP: 0010:[<ffffffff8133ba17>]  [<ffffffff8133ba17>]
rb_next+0x27/0x50
[  228.064539] RSP: 0018:ffff880074617de0  EFLAGS: 00000282
[  228.064539] RAX: ffff880074478010 RBX: ffff88007446f860 RCX:
0000000000000014
[  228.064539] RDX: ffff880074478010 RSI: 0000000000000000 RDI:
ffff880074478010
[  228.064539] RBP: ffff880074617de0 R08: 0000000000000000 R09:
0000000000000012
[  228.064539] R10: 0000000000000001 R11: ffffffffffffffec R12:
ffffea0001d11a00
[  228.064539] R13: ffff88007f401400 R14: ffff88007446f800 R15:
ffff880074617d50
[  228.064539] FS:  0000000000000000(0000) GS:ffff88007f800000(0000)
knlGS:0000000000000000
[  228.064539] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[  228.064539] CR2: 00007fe9ac6ec000 CR3: 000000007a5d6000 CR4:
00000000000006f0
[  228.064539] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[  228.064539] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
0000000000000400
[  228.064539] Stack:
[  228.064539]  ffff880074617e28 ffffffffa01ab7db ffff880074617df0
ffff880074617df0
[  228.064539]  ffff880079273000 ffffffff81cc26c0 ffffffff81cc26c0
0000000000000000
[  228.064539]  0000000000000000 ffff880074617e48 ffffffffa01840b8
ffffffff81cc26c0
[  228.064539] Call Trace:
[  228.064539]  [<ffffffffa01ab7db>] nfs4_state_shutdown_net+0x18b/0x220
[nfsd]
[  228.064539]  [<ffffffffa01840b8>] nfsd_shutdown_net+0x38/0x70 [nfsd]
[  228.064539]  [<ffffffffa018413e>] nfsd_last_thread+0x4e/0x80 [nfsd]
[  228.064539]  [<ffffffffa00aa1eb>] svc_shutdown_net+0x2b/0x30 [sunrpc]
[  228.064539]  [<ffffffffa018464b>] nfsd_destroy+0x5b/0x80 [nfsd]
[  228.064539]  [<ffffffffa0184773>] nfsd+0x103/0x130 [nfsd]
[  228.064539]  [<ffffffffa0184670>] ? nfsd_destroy+0x80/0x80 [nfsd]
[  228.064539]  [<ffffffff810a8232>] kthread+0xd2/0xf0
[  228.064539]  [<ffffffff810a8160>] ? insert_kthread_work+0x40/0x40
[  228.064539]  [<ffffffff816c493c>] ret_from_fork+0x7c/0xb0
[  228.064539]  [<ffffffff810a8160>] ? insert_kthread_work+0x40/0x40
[  228.064539] Code: 1f 44 00 00 55 48 8b 17 48 89 e5 48 39 d7 74 3b 48
8b 47 08 48 85 c0 75 0e eb 25 66 0f 1f 84 00 00 00 00 00 48 89 d0 48 8b
50 10 <48> 85 d2 75 f4 5d c3 66 90 48 3b 78 08 75 f6 48 8b 10 48 89 c7

Fixes: ac55fdc4 (nfsd: move the confirmed and unconfirmed hlists...)
Signed-off-by: NKinglong Mee <kinglongmee@gmail.com>
Cc: stable@vger.kernel.org
Reviewed-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

CLAIM_FH was added by NFSv4.1.  It is the same as CLAIM_NULL except that it
uses only current FH to identify the file to be opened.

The NFS client is using CLAIM_FH if the FH is available when opening a file.
Currently, we cannot get any delegation if we stat a file before open it
because the server delegation code does not recognize CLAIM_FH.

We tested this patch and found delegation can be handed out now when claim is
CLAIM_FH.

See http://marc.info/?l=linux-nfs&m=136369847801388&w=2 and
http://www.linux-nfs.org/wiki/index.php/Server_4.0_and_4.1_issues#New_open_claim_typesSigned-off-by: NMing Chen <mchen@cs.stonybrook.edu>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

As far as I can tell, this list is used only under the state lock, so we
may as well do this in the simpler order.
Acked-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

If failed after calling alloc_session but before init_session, nfsd will call __free_session to
free se_slots in session. But, session->se_fchannel.maxreqs is not initialized (value is zero).
So that, the memory malloced for slots will be lost in free_session_slots for maxreqs is zero.

This path sets the information for channel in alloc_session after mallocing slots succeed,
instead in init_session.
Signed-off-by: NKinglong Mee <kinglongmee@gmail.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

the length for backchannel checking should be multiplied by sizeof(__be32).
Signed-off-by: NKinglong Mee <kinglongmee@gmail.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

check_forechannel_attrs gets drc memory, so nfsd must put it when
check_backchannel_attrs fails.

After many requests with bad back channel attrs, nfsd will deny any
client's CREATE_SESSION forever.

A new test case named CSESS29 for pynfs will send in another mail.
Signed-off-by: NKinglong Mee <kinglongmee@gmail.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

commit 5b6feee9 forgot
recording the back channel attrs in nfsd4_session.

nfsd just check the back channel attars by check_backchannel_attrs,
but do not  record it in nfsd4_session in the latest kernel.
Signed-off-by: NKinglong Mee <kinglongmee@gmail.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

For now FL_DELEG is just a synonym for FL_LEASE.  So this patch doesn't
change behavior.

Next we'll modify break_lease to treat FL_DELEG leases differently, to
account for the fact that NFSv4 delegations should be broken in more
situations than Windows oplocks.
Acked-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

This reverts commit 7ebe40f2.  We forgot
the nfs4_put_delegation call in fs/nfsd/nfs4callback.c which should not
be unhashing the stateid.  This lead to warnings from the idr code when
we tried to removed id's twice.
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

A comment claims the caller should take it, but that's not being done.
Note we don't want it around the cancel_delayed_work_sync since that may
wait on work which holds the client lock.
Reported-by: NBenny Halevy <bhalevy@primarydata.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

[use list_splice_init]
Signed-off-by: NBenny Halevy <bhalevy@primarydata.com>
[bfields: no need for recall_lock here]
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>