[PATCH v2 0/67] nfs41 client patch for 2.6.30

[PATCH v2 0/67] nfs41 client patch for 2.6.30

[Benny Halevy]

Trond,

I've integrated the latest client patches from Andy
(with a few minor changes listed below) and updated
git://linux-nfs.org/~bhalevy/linux-pnfs.git
branches:
	nfs41-for-next
	nfs41-for-2.6.30
	nfs41

nfs41-for-{next,2.6.30} contain all non-backchannel patches
rebased onto {v2.6.29,nfs-2.6/for-next} respectively.

nfs41 contains the backchannel patches on top of nfs41-for-2.6.30.

changes since v1:
- fixed bisectability issues
- got rid of rpc_clnt.cl_private, use *session in seq args/res instead
- remove ifdefs around nfs_client_initdata.minorversion
- use kcalloc to allocate slot table
- alloc session with GFP_KERNEL
- moved nfs4_init_slot_table's definition from "sessions client infrastructure"
  to first usage, in "create_session operations".
- embed resp_len in nfs_getaclres
- moved new xdr op args/res definitions away from "client xdr definitions",
  to where they're first used.
- move find_slot and free_slot closer to where they're first used.
- Revert "NFS: fix decode_fs_locations_maxsz"
- nfs41: get fs_locations replen before encoding the GETATTR
- nfs41: get getacl replen before encoding the GETATTR

(changes from Andy's set)
- getaclres moved into right patch
- dropped "nfs41: stubs for nfs41 procedures"
- edited create_session, destroy_session, session setup, and session_reset
  for proc_{create,destroy}_session interface.
- reverse EXCHGID4_INVAL_FLAG_MASK_{A,R}

Here's the list of patches:

[PATCH v2 01/67] nfs41: common protocol definitions
[PATCH v2 02/67] nfs41: Add Kconfig symbols for NFSv4.1
[PATCH v2 03/67] nfs41: define NFS4_MAX_MINOR_VERSION based on CONFIG_NFS_V4_1
[PATCH v2 04/67] nfs41: add mount command option minorversion
[PATCH v2 05/67] nfs41: nfs_client.cl_minorversion
[PATCH v2 06/67] nfs41: Use mount minorversion option
[PATCH v2 07/67] nfs41: translate NFS4ERR_MINOR_VERS_MISMATCH to EPROTONOSUPPORT
[PATCH v2 08/67] nfs41: client xdr definitions
[PATCH v2 09/67] nfs41: sessions client infrastructure
[PATCH v2 10/67] nfs41: use nfs4_server_caps_arg
[PATCH v2 11/67] nfs41: use nfs4_readlink_res
[PATCH v2 12/67] nfs41: use nfs4_statfs_res
[PATCH v2 13/67] nfs41: use nfs4_fsinfo_res
[PATCH v2 14/67] nfs41: use nfs4_pathconf_res
[PATCH v2 15/67] nfs41: use nfs4_getaclres
[PATCH v2 16/67] NFS: get rid of unused xdr decode_setattr(, res) argument
[PATCH v2 17/67] nfs41: use nfs4_setaclres
[PATCH v2 18/67] nfs41: use nfs4_fs_locations_res
[PATCH v2 19/67] nfs41: introduce nfs4_call_sync
[PATCH v2 20/67] nfs41: nfs41: pass *session in seq_args and seq_res
[PATCH v2 21/67] nfs41: set up seq_res.sr_slotid
[PATCH v2 22/67] NFS: use decode_change_info_maxsz for xdr maxsz calculations
[PATCH v2 23/67] NFS: define and initialize compound_hdr.replen
[PATCH v2 24/67] NFS: update hdr->replen for every encode op
[PATCH v2 25/67] NFS: use dynamically computed compound_hdr.replen for xdr_inline_pages offset
[PATCH v2 26/67] nfs41: encode minorversion in compound header
[PATCH v2 27/67] nfs41: xdr {encode,decode}_sequence
[PATCH v2 28/67] nfs41: nfs4_setup_sequence
[PATCH v2 29/67] nfs41: find slot
[PATCH v2 30/67] nfs41: setup_sequence method
[PATCH v2 31/67] nfs41: free slot
[PATCH v2 32/67] nfs41: nfs41_sequence_free_slot
[PATCH v2 33/67] nfs41: nfs41_sequence_done
[PATCH v2 34/67] nfs41: nfs41_call_sync_done
[PATCH v2 35/67] nfs41: close sequence setup/done support
[PATCH v2 36/67] nfs41: open sequence setup/done support
[PATCH v2 37/67] nfs41: lock sequence setup/done support
[PATCH v2 38/67] nfs41: locku sequence setup/done support
[PATCH v2 39/67] nfs41: unlink sequence setup/done support
[PATCH v2 40/67] nfs41: read sequence setup/done support
[PATCH v2 41/67] nfs41 write sequence setup done support
[PATCH v2 42/67] nfs41 commit sequence setup done support
[PATCH v2 43/67] nfs41 delegreturn sequence setup done support
[PATCH v2 44/67] nfs41: exchange_id operation
[PATCH v2 45/67] nfs41: get_lease_time
[PATCH v2 46/67] nfs41: create_session operation
[PATCH v2 47/67] nfs41: verify session channel attribues
[PATCH v2 48/67] nfs41: use session attributes for rsize and wsize
[PATCH v2 49/67] nfs41: destroy_session operation
[PATCH v2 50/67] nfs41: enable nfs_client only nfs4_async_handle_error
[PATCH v2 51/67] nfs41: sequence operation
[PATCH v2 52/67] nfs41: reset the session slot table
[PATCH v2 53/67] nfs41: add session setup to the state manager
[PATCH v2 54/67] nfs41: add session reset to state manager
[PATCH v2 55/67] nfs41: sunrpc: Export the call prepare state for session reset
[PATCH v2 56/67] nfs41: use rpc prepare call state for session reset
[PATCH v2 57/67] nfs41: kick start nfs41 session recovery when handling errors
[PATCH v2 58/67] nfs41: schedule async session reset
[PATCH v2 59/67] nfs41: lease renewal
[PATCH v2 60/67] nfs41: support minorversion 1 for nfs4_check_lease
[PATCH v2 61/67] nfs41: introduce get_state_renewal_cred
[PATCH v2 62/67] nfs41: establish sessions-based clientid
[PATCH v2 63/67] nfs41: add a get_clid_cred function to nfs4_state_recovery_ops
[PATCH v2 64/67] nfs41: get_clid_cred for EXCHANGE_ID
[PATCH v2 65/67] nfs41: recover lease in _nfs4_lookup_root
[PATCH v2 66/67] nfs41: fail mount on EXCHANGE_ID, CREATE_SESSION failure
[PATCH v2 67/67] nfs41: increment_{open,lock}_seqid

Re: [pnfs] [PATCH v2 0/67] nfs41 client patch for 2.6.30

[Benny Halevy]

On Apr. 02, 2009, 16:46 +0300, Benny Halevy <bhalevy@panasas.com> wrote:
> Trond,
> 
> I've integrated the latest client patches from Andy
> (with a few minor changes listed below) and updated
> git://linux-nfs.org/~bhalevy/linux-pnfs.git
> branches:
> 	nfs41-for-next
> 	nfs41-for-2.6.30
> 	nfs41
> 
> nfs41-for-{next,2.6.30} contain all non-backchannel patches
> rebased onto {v2.6.29,nfs-2.6/for-next} respectively.
> 
> nfs41 contains the backchannel patches on top of nfs41-for-2.6.30.
> 
> changes since v1:
> - fixed bisectability issues
> - got rid of rpc_clnt.cl_private, use *session in seq args/res instead
> - remove ifdefs around nfs_client_initdata.minorversion
> - use kcalloc to allocate slot table
> - alloc session with GFP_KERNEL
> - moved nfs4_init_slot_table's definition from "sessions client infrastructure"
>   to first usage, in "create_session operations".
> - embed resp_len in nfs_getaclres
> - moved new xdr op args/res definitions away from "client xdr definitions",
>   to where they're first used.
> - move find_slot and free_slot closer to where they're first used.
> - Revert "NFS: fix decode_fs_locations_maxsz"
> - nfs41: get fs_locations replen before encoding the GETATTR
> - nfs41: get getacl replen before encoding the GETATTR
> 
> (changes from Andy's set)
> - getaclres moved into right patch
> - dropped "nfs41: stubs for nfs41 procedures"
> - edited create_session, destroy_session, session setup, and session_reset
>   for proc_{create,destroy}_session interface.
> - reverse EXCHGID4_INVAL_FLAG_MASK_{A,R}
> 
> Here's the list of patches:

The patches are based off of nfs-2.6/for-next
for your convenience.

Benny

> 
> [PATCH v2 01/67] nfs41: common protocol definitions
> [PATCH v2 02/67] nfs41: Add Kconfig symbols for NFSv4.1
> [PATCH v2 03/67] nfs41: define NFS4_MAX_MINOR_VERSION based on CONFIG_NFS_V4_1
> [PATCH v2 04/67] nfs41: add mount command option minorversion
> [PATCH v2 05/67] nfs41: nfs_client.cl_minorversion
> [PATCH v2 06/67] nfs41: Use mount minorversion option
> [PATCH v2 07/67] nfs41: translate NFS4ERR_MINOR_VERS_MISMATCH to EPROTONOSUPPORT
> [PATCH v2 08/67] nfs41: client xdr definitions
> [PATCH v2 09/67] nfs41: sessions client infrastructure
> [PATCH v2 10/67] nfs41: use nfs4_server_caps_arg
> [PATCH v2 11/67] nfs41: use nfs4_readlink_res
> [PATCH v2 12/67] nfs41: use nfs4_statfs_res
> [PATCH v2 13/67] nfs41: use nfs4_fsinfo_res
> [PATCH v2 14/67] nfs41: use nfs4_pathconf_res
> [PATCH v2 15/67] nfs41: use nfs4_getaclres
> [PATCH v2 16/67] NFS: get rid of unused xdr decode_setattr(, res) argument
> [PATCH v2 17/67] nfs41: use nfs4_setaclres
> [PATCH v2 18/67] nfs41: use nfs4_fs_locations_res
> [PATCH v2 19/67] nfs41: introduce nfs4_call_sync
> [PATCH v2 20/67] nfs41: nfs41: pass *session in seq_args and seq_res
> [PATCH v2 21/67] nfs41: set up seq_res.sr_slotid
> [PATCH v2 22/67] NFS: use decode_change_info_maxsz for xdr maxsz calculations
> [PATCH v2 23/67] NFS: define and initialize compound_hdr.replen
> [PATCH v2 24/67] NFS: update hdr->replen for every encode op
> [PATCH v2 25/67] NFS: use dynamically computed compound_hdr.replen for xdr_inline_pages offset
> [PATCH v2 26/67] nfs41: encode minorversion in compound header
> [PATCH v2 27/67] nfs41: xdr {encode,decode}_sequence
> [PATCH v2 28/67] nfs41: nfs4_setup_sequence
> [PATCH v2 29/67] nfs41: find slot
> [PATCH v2 30/67] nfs41: setup_sequence method
> [PATCH v2 31/67] nfs41: free slot
> [PATCH v2 32/67] nfs41: nfs41_sequence_free_slot
> [PATCH v2 33/67] nfs41: nfs41_sequence_done
> [PATCH v2 34/67] nfs41: nfs41_call_sync_done
> [PATCH v2 35/67] nfs41: close sequence setup/done support
> [PATCH v2 36/67] nfs41: open sequence setup/done support
> [PATCH v2 37/67] nfs41: lock sequence setup/done support
> [PATCH v2 38/67] nfs41: locku sequence setup/done support
> [PATCH v2 39/67] nfs41: unlink sequence setup/done support
> [PATCH v2 40/67] nfs41: read sequence setup/done support
> [PATCH v2 41/67] nfs41 write sequence setup done support
> [PATCH v2 42/67] nfs41 commit sequence setup done support
> [PATCH v2 43/67] nfs41 delegreturn sequence setup done support
> [PATCH v2 44/67] nfs41: exchange_id operation
> [PATCH v2 45/67] nfs41: get_lease_time
> [PATCH v2 46/67] nfs41: create_session operation
> [PATCH v2 47/67] nfs41: verify session channel attribues
> [PATCH v2 48/67] nfs41: use session attributes for rsize and wsize
> [PATCH v2 49/67] nfs41: destroy_session operation
> [PATCH v2 50/67] nfs41: enable nfs_client only nfs4_async_handle_error
> [PATCH v2 51/67] nfs41: sequence operation
> [PATCH v2 52/67] nfs41: reset the session slot table
> [PATCH v2 53/67] nfs41: add session setup to the state manager
> [PATCH v2 54/67] nfs41: add session reset to state manager
> [PATCH v2 55/67] nfs41: sunrpc: Export the call prepare state for session reset
> [PATCH v2 56/67] nfs41: use rpc prepare call state for session reset
> [PATCH v2 57/67] nfs41: kick start nfs41 session recovery when handling errors
> [PATCH v2 58/67] nfs41: schedule async session reset
> [PATCH v2 59/67] nfs41: lease renewal
> [PATCH v2 60/67] nfs41: support minorversion 1 for nfs4_check_lease
> [PATCH v2 61/67] nfs41: introduce get_state_renewal_cred
> [PATCH v2 62/67] nfs41: establish sessions-based clientid
> [PATCH v2 63/67] nfs41: add a get_clid_cred function to nfs4_state_recovery_ops
> [PATCH v2 64/67] nfs41: get_clid_cred for EXCHANGE_ID
> [PATCH v2 65/67] nfs41: recover lease in _nfs4_lookup_root
> [PATCH v2 66/67] nfs41: fail mount on EXCHANGE_ID, CREATE_SESSION failure
> [PATCH v2 67/67] nfs41: increment_{open,lock}_seqid
> _______________________________________________
> pNFS mailing list
> pNFS@linux-nfs.org
> http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs

[PATCH v2 01/67] nfs41: common protocol definitions

[Benny Halevy]

Define all NFSv4.1 common operation and error code constants.

Note that some of the definitions are used by both the nfs41 client
and the server code. This patch is duplicated in the nfs41 and nfsd41
sessions patchset.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: add exchange id flags]
Signed-off-by: Mike Sager <sager@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[removed server-only hunk changing NFSERR_REPLAY_ME]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: add SEQ4_XX to nfs41-common-protocol]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: generic error code update]
[nfs41: reverse EXCHGID4_INVAL_FLAG_MASK_{A,R}]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/nfs4.h |  128 +++++++++++++++++++++++++++++++++++++++++++++++++-
 1 files changed, 127 insertions(+), 1 deletions(-)

diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index b912311..29ed600 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -21,6 +21,7 @@
 #define NFS4_FHSIZE		128
 #define NFS4_MAXPATHLEN		PATH_MAX
 #define NFS4_MAXNAMLEN		NAME_MAX
+#define NFS4_MAX_SESSIONID_LEN	16
 
 #define NFS4_ACCESS_READ        0x0001
 #define NFS4_ACCESS_LOOKUP      0x0002
@@ -38,6 +39,7 @@
 #define NFS4_OPEN_RESULT_CONFIRM 0x0002
 #define NFS4_OPEN_RESULT_LOCKTYPE_POSIX 0x0004
 
+#define NFS4_SHARE_ACCESS_MASK	0x000F
 #define NFS4_SHARE_ACCESS_READ	0x0001
 #define NFS4_SHARE_ACCESS_WRITE	0x0002
 #define NFS4_SHARE_ACCESS_BOTH	0x0003
@@ -45,6 +47,19 @@
 #define NFS4_SHARE_DENY_WRITE	0x0002
 #define NFS4_SHARE_DENY_BOTH	0x0003
 
+/* nfs41 */
+#define NFS4_SHARE_WANT_MASK		0xFF00
+#define NFS4_SHARE_WANT_NO_PREFERENCE	0x0000
+#define NFS4_SHARE_WANT_READ_DELEG	0x0100
+#define NFS4_SHARE_WANT_WRITE_DELEG	0x0200
+#define NFS4_SHARE_WANT_ANY_DELEG	0x0300
+#define NFS4_SHARE_WANT_NO_DELEG	0x0400
+#define NFS4_SHARE_WANT_CANCEL		0x0500
+
+#define NFS4_SHARE_WHEN_MASK		0xF0000
+#define NFS4_SHARE_SIGNAL_DELEG_WHEN_RESRC_AVAIL	0x10000
+#define NFS4_SHARE_PUSH_DELEG_WHEN_UNCONTENDED		0x20000
+
 #define NFS4_SET_TO_SERVER_TIME	0
 #define NFS4_SET_TO_CLIENT_TIME	1
 
@@ -88,6 +103,31 @@
 #define NFS4_ACE_GENERIC_EXECUTE              0x001200A0
 #define NFS4_ACE_MASK_ALL                     0x001F01FF
 
+#define EXCHGID4_FLAG_SUPP_MOVED_REFER		0x00000001
+#define EXCHGID4_FLAG_SUPP_MOVED_MIGR		0x00000002
+#define EXCHGID4_FLAG_USE_NON_PNFS		0x00010000
+#define EXCHGID4_FLAG_USE_PNFS_MDS		0x00020000
+#define EXCHGID4_FLAG_USE_PNFS_DS		0x00040000
+#define EXCHGID4_FLAG_UPD_CONFIRMED_REC_A	0x40000000
+#define EXCHGID4_FLAG_CONFIRMED_R		0x80000000
+/*
+ * Since the validity of these bits depends on whether
+ * they're set in the argument or response, have separate
+ * invalid flag masks for arg (_A) and resp (_R).
+ */
+#define EXCHGID4_FLAG_MASK_A			0x40070003
+#define EXCHGID4_FLAG_MASK_R			0x80070003
+
+#define SEQ4_STATUS_CB_PATH_DOWN		0x00000001
+#define SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRING	0x00000002
+#define SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRED	0x00000004
+#define SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED	0x00000008
+#define SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED	0x00000010
+#define SEQ4_STATUS_ADMIN_STATE_REVOKED		0x00000020
+#define SEQ4_STATUS_RECALLABLE_STATE_REVOKED	0x00000040
+#define SEQ4_STATUS_LEASE_MOVED			0x00000080
+#define SEQ4_STATUS_RESTART_RECLAIM_NEEDED	0x00000100
+
 #define NFS4_MAX_UINT64	(~(u64)0)
 
 enum nfs4_acl_whotype {
@@ -154,6 +194,28 @@ enum nfs_opnum4 {
 	OP_VERIFY = 37,
 	OP_WRITE = 38,
 	OP_RELEASE_LOCKOWNER = 39,
+
+	/* nfs41 */
+	OP_BACKCHANNEL_CTL = 40,
+	OP_BIND_CONN_TO_SESSION = 41,
+	OP_EXCHANGE_ID = 42,
+	OP_CREATE_SESSION = 43,
+	OP_DESTROY_SESSION = 44,
+	OP_FREE_STATEID = 45,
+	OP_GET_DIR_DELEGATION = 46,
+	OP_GETDEVICEINFO = 47,
+	OP_GETDEVICELIST = 48,
+	OP_LAYOUTCOMMIT = 49,
+	OP_LAYOUTGET = 50,
+	OP_LAYOUTRETURN = 51,
+	OP_SECINFO_NO_NAME = 52,
+	OP_SEQUENCE = 53,
+	OP_SET_SSV = 54,
+	OP_TEST_STATEID = 55,
+	OP_WANT_DELEGATION = 56,
+	OP_DESTROY_CLIENTID = 57,
+	OP_RECLAIM_COMPLETE = 58,
+
 	OP_ILLEGAL = 10044,
 };
 
@@ -230,7 +292,48 @@ enum nfsstat4 {
 	NFS4ERR_DEADLOCK = 10045,
 	NFS4ERR_FILE_OPEN = 10046,
 	NFS4ERR_ADMIN_REVOKED = 10047,
-	NFS4ERR_CB_PATH_DOWN = 10048
+	NFS4ERR_CB_PATH_DOWN = 10048,
+
+	/* nfs41 */
+	NFS4ERR_BADIOMODE	= 10049,
+	NFS4ERR_BADLAYOUT	= 10050,
+	NFS4ERR_BAD_SESSION_DIGEST = 10051,
+	NFS4ERR_BADSESSION	= 10052,
+	NFS4ERR_BADSLOT		= 10053,
+	NFS4ERR_COMPLETE_ALREADY = 10054,
+	NFS4ERR_CONN_NOT_BOUND_TO_SESSION = 10055,
+	NFS4ERR_DELEG_ALREADY_WANTED = 10056,
+	NFS4ERR_BACK_CHAN_BUSY	= 10057,	/* backchan reqs outstanding */
+	NFS4ERR_LAYOUTTRYLATER	= 10058,
+	NFS4ERR_LAYOUTUNAVAILABLE = 10059,
+	NFS4ERR_NOMATCHING_LAYOUT = 10060,
+	NFS4ERR_RECALLCONFLICT	= 10061,
+	NFS4ERR_UNKNOWN_LAYOUTTYPE = 10062,
+	NFS4ERR_SEQ_MISORDERED = 10063, 	/* unexpected seq.id in req */
+	NFS4ERR_SEQUENCE_POS	= 10064,	/* [CB_]SEQ. op not 1st op */
+	NFS4ERR_REQ_TOO_BIG	= 10065,	/* request too big */
+	NFS4ERR_REP_TOO_BIG	= 10066,	/* reply too big */
+	NFS4ERR_REP_TOO_BIG_TO_CACHE = 10067,	/* rep. not all cached */
+	NFS4ERR_RETRY_UNCACHED_REP = 10068,	/* retry & rep. uncached */
+	NFS4ERR_UNSAFE_COMPOUND = 10069,	/* retry/recovery too hard */
+	NFS4ERR_TOO_MANY_OPS	= 10070,	/* too many ops in [CB_]COMP */
+	NFS4ERR_OP_NOT_IN_SESSION = 10071,	/* op needs [CB_]SEQ. op */
+	NFS4ERR_HASH_ALG_UNSUPP = 10072,	/* hash alg. not supp. */
+						/* Error 10073 is unused. */
+	NFS4ERR_CLIENTID_BUSY	= 10074,	/* clientid has state */
+	NFS4ERR_PNFS_IO_HOLE	= 10075,	/* IO to _SPARSE file hole */
+	NFS4ERR_SEQ_FALSE_RETRY	= 10076,	/* retry not origional */
+	NFS4ERR_BAD_HIGH_SLOT	= 10077,	/* sequence arg bad */
+	NFS4ERR_DEADSESSION	= 10078,	/* persistent session dead */
+	NFS4ERR_ENCR_ALG_UNSUPP = 10079,	/* SSV alg mismatch */
+	NFS4ERR_PNFS_NO_LAYOUT	= 10080,	/* direct I/O with no layout */
+	NFS4ERR_NOT_ONLY_OP	= 10081,	/* bad compound */
+	NFS4ERR_WRONG_CRED	= 10082,	/* permissions:state change */
+	NFS4ERR_WRONG_TYPE	= 10083,	/* current operation mismatch */
+	NFS4ERR_DIRDELEG_UNAVAIL = 10084,	/* no directory delegation */
+	NFS4ERR_REJECT_DELEG	= 10085,	/* on callback */
+	NFS4ERR_RETURNCONFLICT	= 10086,	/* outstanding layoutreturn */
+	NFS4ERR_DELEG_REVOKED	= 10087,	/* deleg./layout revoked */
 };
 
 /*
@@ -391,6 +494,29 @@ enum {
 	NFSPROC4_CLNT_GETACL,
 	NFSPROC4_CLNT_SETACL,
 	NFSPROC4_CLNT_FS_LOCATIONS,
+
+	/* nfs41 */
+	NFSPROC4_CLNT_EXCHANGE_ID,
+	NFSPROC4_CLNT_CREATE_SESSION,
+	NFSPROC4_CLNT_DESTROY_SESSION,
+	NFSPROC4_CLNT_SEQUENCE,
+	NFSPROC4_CLNT_GET_LEASE_TIME,
+};
+
+/* nfs41 types */
+struct nfs4_sessionid {
+	unsigned char data[NFS4_MAX_SESSIONID_LEN];
+};
+
+/* Create Session Flags */
+#define SESSION4_PERSIST	 0x001
+#define SESSION4_BACK_CHAN 	 0x002
+#define SESSION4_RDMA		 0x004
+
+enum state_protect_how4 {
+	SP4_NONE	= 0,
+	SP4_MACH_CRED	= 1,
+	SP4_SSV		= 2
 };
 
 #endif
-- 
1.6.2.1

[PATCH v2 02/67] nfs41: Add Kconfig symbols for NFSv4.1

[Benny Halevy]

From: Ricardo Labiaga <ricardo.labiaga@netapp.com>

Added CONFIG_NFS_V4_1 and made it depend upon CONFIG_NFS_V4 and EXPERIMENTAL.
Indicate that CONFIG_NFS_V4_1 is for NFS developers at the moment

At the moment we're expecting folks trying out nfs41 to
actively participate in the development process by helping us
debug issues and ideally send patches to fix problems.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/Kconfig |    9 +++++++++
 1 files changed, 9 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index e67f3ec..5d6d6f4 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -74,6 +74,15 @@ config NFS_V4
 
 	  If unsure, say N.
 
+config NFS_V4_1
+	bool "NFS client support for NFSv4.1 (DEVELOPER ONLY)"
+	depends on NFS_V4 && EXPERIMENTAL
+	help
+	  This option enables support for minor version 1 of the NFSv4 protocol
+	  (draft-ietf-nfsv4-minorversion1) in the kernel's NFS client.
+
+	  Unless you're an NFS developer, say N.
+
 config ROOT_NFS
 	bool "Root file system on NFS"
 	depends on NFS_FS=y && IP_PNP
-- 
1.6.2.1

[PATCH v2 03/67] nfs41: define NFS4_MAX_MINOR_VERSION based on CONFIG_NFS_V4_1

[Benny Halevy]

From: Mike Sager <sager@netapp.com>

If 4.1 isn't supported, NFS4_MAX_MINOR_VERSION will be 0.

Signed-off-by: Mike Sager <sager@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/nfs4.h |    7 +++++++
 1 files changed, 7 insertions(+), 0 deletions(-)

diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index 29ed600..c0c1e8e 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -454,6 +454,13 @@ enum lock_type4 {
 #define NFSPROC4_NULL 0
 #define NFSPROC4_COMPOUND 1
 #define NFS4_MINOR_VERSION 0
+
+#if defined(CONFIG_NFS_V4_1)
+#define NFS4_MAX_MINOR_VERSION 1
+#else
+#define NFS4_MAX_MINOR_VERSION 0
+#endif /* CONFIG_NFS_V4_1 */
+
 #define NFS4_DEBUG 1
 
 /* Index of predefined Linux client operations */
-- 
1.6.2.1

[PATCH v2 04/67] nfs41: add mount command option minorversion

[Benny Halevy]

From: Mike Sager <sager@netapp.com>

mount -t nfs4 -o minorversion=[0|1] specifies whether to use 4.0 or 4.1.
By default, the minorversion is set to 0.

Signed-off-by: Mike Sager <sager@netapp.com>
[set default minorversion to 0 as per Trond and SteveD's request]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/internal.h |    1 +
 fs/nfs/super.c    |   10 ++++++++++
 2 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index e4d6a83..ffa6bd5 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -44,6 +44,7 @@ struct nfs_parsed_mount_data {
 	unsigned int		auth_flavor_len;
 	rpc_authflavor_t	auth_flavors[1];
 	char			*client_address;
+	unsigned int		minorversion;
 	char			*fscache_uniq;
 
 	struct {
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index f0f88a5..4d22b18 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -91,6 +91,7 @@ enum {
 	Opt_mountport,
 	Opt_mountvers,
 	Opt_nfsvers,
+	Opt_minorversion,
 
 	/* Mount options that take string arguments */
 	Opt_sec, Opt_proto, Opt_mountproto, Opt_mounthost,
@@ -156,6 +157,7 @@ static const match_table_t nfs_mount_option_tokens = {
 	{ Opt_mountvers, "mountvers=%u" },
 	{ Opt_nfsvers, "nfsvers=%u" },
 	{ Opt_nfsvers, "vers=%u" },
+	{ Opt_minorversion, "minorversion=%u" },
 
 	{ Opt_sec, "sec=%s" },
 	{ Opt_proto, "proto=%s" },
@@ -1227,6 +1229,13 @@ static int nfs_parse_mount_options(char *raw,
 				nfs_parse_invalid_value("nfsvers");
 			}
 			break;
+		case Opt_minorversion:
+			if (match_int(args, &option))
+				return 0;
+			if (option < 0 || option > NFS4_MAX_MINOR_VERSION)
+				return 0;
+			mnt->minorversion = option;
+			break;
 
 		/*
 		 * options that take text values
@@ -2279,6 +2288,7 @@ static int nfs4_validate_mount_data(void *options,
 	args->nfs_server.port	= NFS_PORT; /* 2049 unless user set port= */
 	args->auth_flavors[0]	= RPC_AUTH_UNIX;
 	args->auth_flavor_len	= 0;
+	args->minorversion	= 0;
 
 	switch (data->version) {
 	case 1:
-- 
1.6.2.1

[PATCH v2 05/67] nfs41: nfs_client.cl_minorversion

[Benny Halevy]

This field is set to the nfsv4 minor version for this mount.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>

Note: This patch sets the referral to the same minorversion as the
current mount. Revisit in future patch.

Signed-off-by: Andy Adamson <andros@netapp.com>
[removed cl_minorversion assignment in nfs_set_client]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/client.c           |    9 ++++++---
 include/linux/nfs_fs_sb.h |    3 ++-
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 0f962db..25ea8e4 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -1101,7 +1101,8 @@ static int nfs4_set_client(struct nfs_server *server,
 		const size_t addrlen,
 		const char *ip_addr,
 		rpc_authflavor_t authflavour,
-		int proto, const struct rpc_timeout *timeparms)
+		int proto, const struct rpc_timeout *timeparms,
+		u32 minorversion)
 {
 	struct nfs_client_initdata cl_init = {
 		.hostname = hostname,
@@ -1164,7 +1165,8 @@ static int nfs4_init_server(struct nfs_server *server,
 			data->client_address,
 			data->auth_flavors[0],
 			data->nfs_server.protocol,
-			&timeparms);
+			&timeparms,
+			data->minorversion);
 	if (error < 0)
 		goto error;
 
@@ -1282,7 +1284,8 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 				parent_client->cl_ipaddr,
 				data->authflavor,
 				parent_server->client->cl_xprt->prot,
-				parent_server->client->cl_timeout);
+				parent_server->client->cl_timeout,
+				parent_client->cl_minorversion);
 	if (error < 0)
 		goto error;
 
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 6ad7594..477de6c 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -63,7 +63,8 @@ struct nfs_client {
 	 */
 	char			cl_ipaddr[48];
 	unsigned char		cl_id_uniquifier;
-#endif
+	u32			cl_minorversion;
+#endif /* CONFIG_NFS_V4 */
 
 #ifdef CONFIG_NFS_FSCACHE
 	struct fscache_cookie	*fscache;	/* client index cache cookie */
-- 
1.6.2.1

[PATCH v2 06/67] nfs41: Use mount minorversion option

[Benny Halevy]

Use the mount minorversion option to initialize the nfs_client cl_minorversion
and match it in nfs_match_client() when looking up a nfs_client.

[nfs41: remove ifdefs around nfs_client_initdata.minorversion]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/client.c |    7 ++++++-
 1 files changed, 6 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 25ea8e4..3aabf1b 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -102,6 +102,7 @@ struct nfs_client_initdata {
 	size_t addrlen;
 	const struct nfs_rpc_ops *rpc_ops;
 	int proto;
+	u32 minorversion;
 };
 
 /*
@@ -150,6 +151,7 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client");
 	clp->cl_boot_time = CURRENT_TIME;
 	clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED;
+	clp->cl_minorversion = cl_init->minorversion;
 #endif
 	cred = rpc_lookup_machine_cred();
 	if (!IS_ERR(cred))
@@ -420,7 +422,9 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat
 
 		if (clp->cl_proto != data->proto)
 			continue;
-
+		/* Match nfsv4 minorversion */
+		if (clp->cl_minorversion != data->minorversion)
+			continue;
 		/* Match the full socket address */
 		if (!nfs_sockaddr_cmp(sap, clap))
 			continue;
@@ -1110,6 +1114,7 @@ static int nfs4_set_client(struct nfs_server *server,
 		.addrlen = addrlen,
 		.rpc_ops = &nfs_v4_clientops,
 		.proto = proto,
+		.minorversion = minorversion,
 	};
 	struct nfs_client *clp;
 	int error;
-- 
1.6.2.1

[PATCH v2 07/67] nfs41: translate NFS4ERR_MINOR_VERS_MISMATCH to EPROTONOSUPPORT

[Benny Halevy]

To be returned to the mount command when trying to mount a v4 server
using minorversion 1.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4state.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 0298e90..bc683ed 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1107,6 +1107,8 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 			nfs4_clear_machine_cred(clp);
 			status = -EAGAIN;
 		}
+		if (status == -NFS4ERR_MINOR_VERS_MISMATCH)
+			status = -EPROTONOSUPPORT;
 	}
 	return status;
 }
-- 
1.6.2.1

[PATCH v2 08/67] nfs41: client xdr definitions

[Benny Halevy]

Define stubs for sequence args and res data structures and embed
them in all other nfs4 and nfs41 xdr types.  They are needed for
sending any op in a nfs41 compound rpc.

Signed-off-by: Andy Adamson<andros@netapp.com>
[moved new args/res definitions away, to where they're first used]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/nfs_xdr.h |   52 +++++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 52 insertions(+), 0 deletions(-)

diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 9708e78..00a0a1d 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -145,6 +145,15 @@ struct nfs4_change_info {
 };
 
 struct nfs_seqid;
+
+struct nfs4_sequence_args {
+	/* stub */
+};
+
+struct nfs4_sequence_res {
+	/* stub */
+};
+
 /*
  * Arguments to the open call.
  */
@@ -165,6 +174,7 @@ struct nfs_openargs {
 	const struct nfs_server *server;	 /* Needed for ID mapping */
 	const u32 *		bitmask;
 	__u32			claim;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_openres {
@@ -181,6 +191,7 @@ struct nfs_openres {
 	__u32			do_recall;
 	__u64			maxsize;
 	__u32			attrset[NFS4_BITMAP_SIZE];
+	struct nfs4_sequence_res	seq_res;
 };
 
 /*
@@ -206,6 +217,7 @@ struct nfs_closeargs {
 	struct nfs_seqid *	seqid;
 	fmode_t			fmode;
 	const u32 *		bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_closeres {
@@ -213,6 +225,7 @@ struct nfs_closeres {
 	struct nfs_fattr *	fattr;
 	struct nfs_seqid *	seqid;
 	const struct nfs_server *server;
+	struct nfs4_sequence_res	seq_res;
 };
 /*
  *  * Arguments to the lock,lockt, and locku call.
@@ -233,12 +246,14 @@ struct nfs_lock_args {
 	unsigned char		block : 1;
 	unsigned char		reclaim : 1;
 	unsigned char		new_lock_owner : 1;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_lock_res {
 	nfs4_stateid		stateid;
 	struct nfs_seqid *	lock_seqid;
 	struct nfs_seqid *	open_seqid;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs_locku_args {
@@ -246,32 +261,38 @@ struct nfs_locku_args {
 	struct file_lock *	fl;
 	struct nfs_seqid *	seqid;
 	nfs4_stateid *		stateid;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_locku_res {
 	nfs4_stateid		stateid;
 	struct nfs_seqid *	seqid;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs_lockt_args {
 	struct nfs_fh *		fh;
 	struct file_lock *	fl;
 	struct nfs_lowner	lock_owner;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_lockt_res {
 	struct file_lock *	denied; /* LOCK, LOCKT failed */
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_delegreturnargs {
 	const struct nfs_fh *fhandle;
 	const nfs4_stateid *stateid;
 	const u32 * bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_delegreturnres {
 	struct nfs_fattr * fattr;
 	const struct nfs_server *server;
+	struct nfs4_sequence_res	seq_res;
 };
 
 /*
@@ -284,12 +305,14 @@ struct nfs_readargs {
 	__u32			count;
 	unsigned int		pgbase;
 	struct page **		pages;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_readres {
 	struct nfs_fattr *	fattr;
 	__u32			count;
 	int                     eof;
+	struct nfs4_sequence_res	seq_res;
 };
 
 /*
@@ -304,6 +327,7 @@ struct nfs_writeargs {
 	unsigned int		pgbase;
 	struct page **		pages;
 	const u32 *		bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_writeverf {
@@ -316,6 +340,7 @@ struct nfs_writeres {
 	struct nfs_writeverf *	verf;
 	__u32			count;
 	const struct nfs_server *server;
+	struct nfs4_sequence_res	seq_res;
 };
 
 /*
@@ -325,12 +350,14 @@ struct nfs_removeargs {
 	const struct nfs_fh	*fh;
 	struct qstr		name;
 	const u32 *		bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_removeres {
 	const struct nfs_server *server;
 	struct nfs4_change_info	cinfo;
 	struct nfs_fattr	dir_attr;
+	struct nfs4_sequence_res 	seq_res;
 };
 
 /*
@@ -383,6 +410,7 @@ struct nfs_setattrargs {
 	struct iattr *                  iap;
 	const struct nfs_server *	server; /* Needed for name mapping */
 	const u32 *			bitmask;
+	struct nfs4_sequence_args 	seq_args;
 };
 
 struct nfs_setaclargs {
@@ -390,6 +418,7 @@ struct nfs_setaclargs {
 	size_t				acl_len;
 	unsigned int			acl_pgbase;
 	struct page **			acl_pages;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_getaclargs {
@@ -397,11 +426,13 @@ struct nfs_getaclargs {
 	size_t				acl_len;
 	unsigned int			acl_pgbase;
 	struct page **			acl_pages;
+	struct nfs4_sequence_args 	seq_args;
 };
 
 struct nfs_setattrres {
 	struct nfs_fattr *              fattr;
 	const struct nfs_server *	server;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs_linkargs {
@@ -583,6 +614,7 @@ struct nfs4_accessargs {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
 	u32				access;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_accessres {
@@ -590,6 +622,7 @@ struct nfs4_accessres {
 	struct nfs_fattr *		fattr;
 	u32				supported;
 	u32				access;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_create_arg {
@@ -609,6 +642,7 @@ struct nfs4_create_arg {
 	const struct iattr *		attrs;
 	const struct nfs_fh *		dir_fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args 	seq_args;
 };
 
 struct nfs4_create_res {
@@ -617,21 +651,25 @@ struct nfs4_create_res {
 	struct nfs_fattr *		fattr;
 	struct nfs4_change_info		dir_cinfo;
 	struct nfs_fattr *		dir_fattr;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_fsinfo_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_getattr_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_getattr_res {
 	const struct nfs_server *	server;
 	struct nfs_fattr *		fattr;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_link_arg {
@@ -639,6 +677,7 @@ struct nfs4_link_arg {
 	const struct nfs_fh *		dir_fh;
 	const struct qstr *		name;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args 	seq_args;
 };
 
 struct nfs4_link_res {
@@ -646,6 +685,7 @@ struct nfs4_link_res {
 	struct nfs_fattr *		fattr;
 	struct nfs4_change_info		cinfo;
 	struct nfs_fattr *		dir_attr;
+	struct nfs4_sequence_res	seq_res;
 };
 
 
@@ -653,21 +693,25 @@ struct nfs4_lookup_arg {
 	const struct nfs_fh *		dir_fh;
 	const struct qstr *		name;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_lookup_res {
 	const struct nfs_server *	server;
 	struct nfs_fattr *		fattr;
 	struct nfs_fh *			fh;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_lookup_root_arg {
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_pathconf_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_readdir_arg {
@@ -678,11 +722,13 @@ struct nfs4_readdir_arg {
 	struct page **			pages;	/* zero-copy data */
 	unsigned int			pgbase;	/* zero-copy data */
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_readdir_res {
 	nfs4_verifier			verifier;
 	unsigned int			pgbase;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_readlink {
@@ -690,6 +736,7 @@ struct nfs4_readlink {
 	unsigned int			pgbase;
 	unsigned int			pglen;   /* zero-copy data */
 	struct page **			pages;   /* zero-copy data */
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_rename_arg {
@@ -698,6 +745,7 @@ struct nfs4_rename_arg {
 	const struct qstr *		old_name;
 	const struct qstr *		new_name;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_rename_res {
@@ -706,6 +754,7 @@ struct nfs4_rename_res {
 	struct nfs_fattr *		old_fattr;
 	struct nfs4_change_info		new_cinfo;
 	struct nfs_fattr *		new_fattr;
+	struct nfs4_sequence_res	seq_res;
 };
 
 #define NFS4_SETCLIENTID_NAMELEN	(127)
@@ -724,6 +773,7 @@ struct nfs4_setclientid {
 struct nfs4_statfs_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_server_caps_res {
@@ -731,6 +781,7 @@ struct nfs4_server_caps_res {
 	u32				acl_bitmask;
 	u32				has_links;
 	u32				has_symlinks;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_string {
@@ -765,6 +816,7 @@ struct nfs4_fs_locations_arg {
 	const struct qstr *name;
 	struct page *page;
 	const u32 *bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 #endif /* CONFIG_NFS_V4 */
-- 
1.6.2.1

[PATCH v2 09/67] nfs41: sessions client infrastructure

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

NFSv4.1 Sessions basic data types, initialization, and destruction.

The session is always associated with a struct nfs_client that holds
the exchange_id results.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[remove extraneous rpc_clnt pointer, use the struct nfs_client cl_rpcclient.
remove the rpc_clnt parameter from nfs4 nfs4_init_session]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Use the presence of a session to determine behaviour instead of the
minorversion number.]
Signed-off-by: Andy Adamson <andros@netapp.com>
[constified nfs4_has_session's struct nfs_client parameter]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Rename nfs4_put_session() to nfs4_destroy_session() and call it from nfs4_free_client() not nfs4_free_server().
Also get rid of nfs4_get_session() and the ref_count in nfs4_session struct as keeping track of nfs_client should be sufficient]
Signed-off-by: Alexandros Batsakis <Alexandros.Batsakis@netapp.com>
[nfs41: pass rsize and wsize into nfs4_init_session]
Signed-off-by: Andy Adamson <andros@netapp.com>
[separated out removal of rpc_clnt parameter from nfs4_init_session ot a
 patch of its own]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Pass the nfs_client pointer into nfs4_alloc_session]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: don't assign to session->clp->cl_session in nfs4_destroy_session]
[nfs41: fixup nfs4_clear_client_minor_version]
[introduce nfs4_clear_client_minor_version() in this patch]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Refactor nfs4_init_session]
    Moved session allocation into nfs4_init_client_minor_version, called from
    nfs4_init_client.
    Leave rwise and wsize initialization in nfs4_init_session, called from
    nfs4_init_server.
    Reverted moving of nfs_fsid definition to nfs_fs_sb.h
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Move NFS4_MAX_SLOT_TABLE define from under CONFIG_NFS_V4_1]
[Fix comile error when CONFIG_NFS_V4_1 is not set.]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[moved nfs4_init_slot_table definition to "create_session operation"]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: alloc session with GFP_KERNEL]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/client.c           |   60 +++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/internal.h         |   12 +++++++++
 fs/nfs/nfs4_fs.h          |    4 +++
 fs/nfs/nfs4proc.c         |   34 +++++++++++++++++++++++++
 include/linux/nfs_fs_sb.h |   49 ++++++++++++++++++++++++++++++++++++
 include/linux/nfs_xdr.h   |   15 +++++++++++
 6 files changed, 174 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 3aabf1b..efbaba3 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -184,12 +184,27 @@ static void nfs4_shutdown_client(struct nfs_client *clp)
 }
 
 /*
+ * Clears/puts all minor version specific parts from an nfs_client struct
+ * reverting it to minorversion 0.
+ */
+static void nfs4_clear_client_minor_version(struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(clp)) {
+		nfs4_destroy_session(clp->cl_session);
+		clp->cl_session = NULL;
+	}
+#endif /* CONFIG_NFS_V4_1 */
+}
+
+/*
  * Destroy a shared client record
  */
 static void nfs_free_client(struct nfs_client *clp)
 {
 	dprintk("--> nfs_free_client(%u)\n", clp->rpc_ops->version);
 
+	nfs4_clear_client_minor_version(clp);
 	nfs4_shutdown_client(clp);
 
 	nfs_fscache_release_client_cookie(clp);
@@ -1054,6 +1069,30 @@ error:
 
 #ifdef CONFIG_NFS_V4
 /*
+ * Initialize the minor version specific parts of an NFS4 client record
+ */
+static int nfs4_init_client_minor_version(struct nfs_client *clp)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (clp->cl_minorversion) {
+		struct nfs4_session *session = NULL;
+		/*
+		 * Create the session and mark it expired.
+		 * When a SEQUENCE operation encounters the expired session
+		 * it will do session recovery to initialize it.
+		 */
+		session = nfs4_alloc_session(clp);
+		if (!session)
+			return -ENOMEM;
+
+		clp->cl_session = session;
+	}
+#endif /* CONFIG_NFS_V4_1 */
+
+	return 0;
+}
+
+/*
  * Initialise an NFS4 client record
  */
 static int nfs4_init_client(struct nfs_client *clp,
@@ -1087,6 +1126,10 @@ static int nfs4_init_client(struct nfs_client *clp,
 	}
 	__set_bit(NFS_CS_IDMAP, &clp->cl_res_state);
 
+	error = nfs4_init_client_minor_version(clp);
+	if (error < 0)
+		goto error;
+
 	nfs_mark_client_ready(clp, NFS_CS_READY);
 	return 0;
 
@@ -1144,6 +1187,21 @@ error:
 }
 
 /*
+ * Initialize a session.
+ * Note: save the mount rsize and wsize for create_server negotiation.
+ */
+static void nfs4_init_session(struct nfs_client *clp,
+			      unsigned int wsize, unsigned int rsize)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (nfs4_has_session(clp)) {
+		clp->cl_session->fc_attrs.max_rqst_sz = wsize;
+		clp->cl_session->fc_attrs.max_resp_sz = rsize;
+	}
+#endif /* CONFIG_NFS_V4_1 */
+}
+
+/*
  * Create a version 4 volume record
  */
 static int nfs4_init_server(struct nfs_server *server,
@@ -1221,6 +1279,8 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 	BUG_ON(!server->nfs_client->rpc_ops);
 	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
 
+	nfs4_init_session(server->nfs_client, server->wsize, server->rsize);
+
 	/* Probe the root fh to retrieve its FSID */
 	error = nfs4_path_walk(server, mntfh, data->nfs_server.export_path);
 	if (error < 0)
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index ffa6bd5..7cef45d 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -207,6 +207,18 @@ extern int nfs4_path_walk(struct nfs_server *server,
 #endif
 
 /*
+ * Determine if sessions are in use.
+ */
+static inline int nfs4_has_session(const struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (clp->cl_session)
+		return 1;
+#endif /* CONFIG_NFS_V4_1 */
+	return 0;
+}
+
+/*
  * Determine the device name as a string
  */
 static inline char *nfs_devname(const struct vfsmount *mnt_parent,
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 4e4d332..5d7b036 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -202,6 +202,10 @@ extern int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 
 extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops;
 extern struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops;
+#if defined(CONFIG_NFS_V4_1)
+extern void nfs4_destroy_session(struct nfs4_session *session);
+extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
+#endif /* CONFIG_NFS_V4_1 */
 
 extern const u32 nfs4_fattr_bitmap[2];
 extern const u32 nfs4_statfs_bitmap[2];
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 97baccc..884bbe0 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3726,6 +3726,40 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 	return status;
 }
 
+#ifdef CONFIG_NFS_V4_1
+/* Destroy the slot table */
+static void nfs4_destroy_slot_table(struct nfs4_session *session)
+{
+	if (session->fc_slot_table.slots == NULL)
+		return;
+	kfree(session->fc_slot_table.slots);
+	session->fc_slot_table.slots = NULL;
+	return;
+}
+
+struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
+{
+	struct nfs4_session *session;
+	struct nfs4_slot_table *tbl;
+
+	session = kzalloc(sizeof(struct nfs4_session), GFP_KERNEL);
+	if (!session)
+		return NULL;
+	tbl = &session->fc_slot_table;
+	spin_lock_init(&tbl->slot_tbl_lock);
+	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "Slot table");
+	session->clp = clp;
+	return session;
+}
+
+void nfs4_destroy_session(struct nfs4_session *session)
+{
+	nfs4_destroy_slot_table(session);
+	kfree(session);
+}
+
+#endif /* CONFIG_NFS_V4_1 */
+
 struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
 	.owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT,
 	.state_flag_bit	= NFS_STATE_RECLAIM_REBOOT,
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 477de6c..283ef0a 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -4,9 +4,12 @@
 #include <linux/list.h>
 #include <linux/backing-dev.h>
 #include <linux/wait.h>
+#include <linux/nfs_xdr.h>
+#include <linux/sunrpc/xprt.h>
 
 #include <asm/atomic.h>
 
+struct nfs4_session;
 struct nfs_iostats;
 struct nlm_host;
 
@@ -66,6 +69,10 @@ struct nfs_client {
 	u32			cl_minorversion;
 #endif /* CONFIG_NFS_V4 */
 
+#ifdef CONFIG_NFS_V4_1
+	struct nfs4_session	*cl_session; 	/* sharred session */
+#endif /* CONFIG_NFS_V4_1 */
+
 #ifdef CONFIG_NFS_FSCACHE
 	struct fscache_cookie	*fscache;	/* client index cache cookie */
 #endif
@@ -146,4 +153,46 @@ struct nfs_server {
 #define NFS_CAP_ACLS		(1U << 3)
 #define NFS_CAP_ATOMIC_OPEN	(1U << 4)
 
+
+/* maximum number of slots to use */
+#define NFS4_MAX_SLOT_TABLE RPC_MAX_SLOT_TABLE
+
+#if defined(CONFIG_NFS_V4_1)
+
+/* Sessions */
+#define SLOT_TABLE_SZ (NFS4_MAX_SLOT_TABLE/(8*sizeof(long)))
+struct nfs4_slot_table {
+	struct nfs4_slot *slots;		/* seqid per slot */
+	unsigned long   used_slots[SLOT_TABLE_SZ]; /* used/unused bitmap */
+	spinlock_t	slot_tbl_lock;
+	struct rpc_wait_queue	slot_tbl_waitq;	/* allocators may wait here */
+	int		max_slots;		/* # slots in table */
+	int		highest_used_slotid;	/* sent to server on each SEQ.
+						 * op for dynamic resizing */
+};
+
+static inline int slot_idx(struct nfs4_slot_table *tbl, struct nfs4_slot *sp)
+{
+	return sp - tbl->slots;
+}
+
+/*
+ * Session related parameters
+ */
+struct nfs4_session {
+	struct nfs4_sessionid		sess_id;
+	u32				flags;
+	unsigned long			session_state;
+	u32				hash_alg;
+	u32				ssv_len;
+
+	/* The fore and back channel */
+	struct nfs4_channel_attrs	fc_attrs;
+	struct nfs4_slot_table		fc_slot_table;
+	struct nfs4_channel_attrs	bc_attrs;
+					/* back channel has one slot */
+	struct nfs_client		*clp;
+};
+
+#endif /* CONFIG_NFS_V4_1 */
 #endif
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 00a0a1d..9480a32 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -146,6 +146,21 @@ struct nfs4_change_info {
 
 struct nfs_seqid;
 
+/* nfs41 sessions channel attributes */
+struct nfs4_channel_attrs {
+	u32			headerpadsz;
+	u32			max_rqst_sz;
+	u32			max_resp_sz;
+	u32			max_resp_sz_cached;
+	u32			max_ops;
+	u32			max_reqs;
+};
+
+/* nfs41 sessions slot seqid */
+struct nfs4_slot {
+	u32		 	seq_nr;
+};
+
 struct nfs4_sequence_args {
 	/* stub */
 };
-- 
1.6.2.1

[PATCH v2 10/67] nfs41: use nfs4_server_caps_arg

[Benny Halevy]

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_server_caps_arg]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |    5 ++++-
 fs/nfs/nfs4xdr.c        |    5 +++--
 include/linux/nfs_xdr.h |    5 +++++
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 884bbe0..dcdd35b 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1584,10 +1584,13 @@ void nfs4_close_context(struct nfs_open_context *ctx, int is_sync)
 
 static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle)
 {
+	struct nfs4_server_caps_arg args = {
+		.fhandle = fhandle,
+	};
 	struct nfs4_server_caps_res res = {};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SERVER_CAPS],
-		.rpc_argp = fhandle,
+		.rpc_argp = &args,
 		.rpc_resp = &res,
 	};
 	int status;
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 1690f0e..9130586 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1900,7 +1900,8 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs
 /*
  * GETATTR_BITMAP request
  */
-static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p, const struct nfs_fh *fhandle)
+static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p,
+				    struct nfs4_server_caps_arg *args)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
@@ -1909,7 +1910,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p, const struc
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, &hdr);
-	encode_putfh(&xdr, fhandle, &hdr);
+	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_getattr_one(&xdr, FATTR4_WORD0_SUPPORTED_ATTRS|
 			   FATTR4_WORD0_LINK_SUPPORT|
 			   FATTR4_WORD0_SYMLINK_SUPPORT|
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 9480a32..2d95f98 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -791,6 +791,11 @@ struct nfs4_statfs_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_server_caps_arg {
+	struct nfs_fh		       *fhandle;
+	struct nfs4_sequence_args	seq_args;
+};
+
 struct nfs4_server_caps_res {
 	u32				attr_bitmask[2];
 	u32				acl_bitmask;
-- 
1.6.2.1

[PATCH v2 11/67] nfs41: use nfs4_readlink_res

[Benny Halevy]

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_readlink_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |    3 ++-
 fs/nfs/nfs4xdr.c        |    3 ++-
 include/linux/nfs_xdr.h |    4 ++++
 3 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index dcdd35b..414c03d 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1960,10 +1960,11 @@ static int _nfs4_proc_readlink(struct inode *inode, struct page *page,
 		.pglen    = pglen,
 		.pages    = &page,
 	};
+	struct nfs4_readlink_res res;
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READLINK],
 		.rpc_argp = &args,
-		.rpc_resp = NULL,
+		.rpc_resp = &res,
 	};
 
 	return rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 9130586..1e41420 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4235,7 +4235,8 @@ out:
 /*
  * Decode READLINK response
  */
-static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p, void *res)
+static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p,
+				 struct nfs4_readlink_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 2d95f98..ac01781 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -754,6 +754,10 @@ struct nfs4_readlink {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_readlink_res {
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs4_rename_arg {
 	const struct nfs_fh *		old_dir;
 	const struct nfs_fh *		new_dir;
-- 
1.6.2.1

[PATCH v2 12/67] nfs41: use nfs4_statfs_res

[Benny Halevy]

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_statfs_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |    5 ++++-
 fs/nfs/nfs4xdr.c        |    5 +++--
 include/linux/nfs_xdr.h |    5 +++++
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 414c03d..85e6bc6 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2426,10 +2426,13 @@ static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
 		.fh = fhandle,
 		.bitmask = server->attr_bitmask,
 	};
+	struct nfs4_statfs_res res = {
+		.fsstat = fsstat,
+	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_STATFS],
 		.rpc_argp = &args,
-		.rpc_resp = fsstat,
+		.rpc_resp = &res,
 	};
 
 	nfs_fattr_init(fsstat->fattr);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 1e41420..b7871ad 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4387,7 +4387,8 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pat
 /*
  * STATFS request
  */
-static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p, struct nfs_fsstat *fsstat)
+static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p,
+			       struct nfs4_statfs_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4398,7 +4399,7 @@ static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p, struct nfs_fssta
 	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
-		status = decode_statfs(&xdr, fsstat);
+		status = decode_statfs(&xdr, res->fsstat);
 	return status;
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index ac01781..3f1ff24 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -795,6 +795,11 @@ struct nfs4_statfs_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_statfs_res {
+	struct nfs_fsstat	       *fsstat;
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs4_server_caps_arg {
 	struct nfs_fh		       *fhandle;
 	struct nfs4_sequence_args	seq_args;
-- 
1.6.2.1

[PATCH v2 13/67] nfs41: use nfs4_fsinfo_res

[Benny Halevy]

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_fsinfo_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |    5 ++++-
 fs/nfs/nfs4xdr.c        |    5 +++--
 include/linux/nfs_xdr.h |    5 +++++
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 85e6bc6..0b82a6b 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2458,10 +2458,13 @@ static int _nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
 		.fh = fhandle,
 		.bitmask = server->attr_bitmask,
 	};
+	struct nfs4_fsinfo_res res = {
+		.fsinfo = fsinfo,
+	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FSINFO],
 		.rpc_argp = &args,
-		.rpc_resp = fsinfo,
+		.rpc_resp = &res,
 	};
 
 	return rpc_call_sync(server->client, &msg, 0);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index b7871ad..d9ab820 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4351,7 +4351,8 @@ out:
 /*
  * FSINFO request
  */
-static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *fsinfo)
+static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
+			       struct nfs4_fsinfo_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4362,7 +4363,7 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs_fsinf
 	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
-		status = decode_fsinfo(&xdr, fsinfo);
+		status = decode_fsinfo(&xdr, res->fsinfo);
 	return status;
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 3f1ff24..8343023 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -675,6 +675,11 @@ struct nfs4_fsinfo_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_fsinfo_res {
+	struct nfs_fsinfo	       *fsinfo;
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs4_getattr_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
-- 
1.6.2.1

[PATCH v2 14/67] nfs41: use nfs4_pathconf_res

[Benny Halevy]

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_pathconf_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |    5 ++++-
 fs/nfs/nfs4xdr.c        |    5 +++--
 include/linux/nfs_xdr.h |    5 +++++
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 0b82a6b..ad6acc9 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2496,10 +2496,13 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle
 		.fh = fhandle,
 		.bitmask = server->attr_bitmask,
 	};
+	struct nfs4_pathconf_res res = {
+		.pathconf = pathconf,
+	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_PATHCONF],
 		.rpc_argp = &args,
-		.rpc_resp = pathconf,
+		.rpc_resp = &res,
 	};
 
 	/* None of the pathconf attributes are mandatory to implement */
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index d9ab820..a77ee3d 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4370,7 +4370,8 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
 /*
  * PATHCONF request
  */
-static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pathconf *pathconf)
+static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p,
+				 struct nfs4_pathconf_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4381,7 +4382,7 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pat
 	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
-		status = decode_pathconf(&xdr, pathconf);
+		status = decode_pathconf(&xdr, res->pathconf);
 	return status;
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 8343023..67942e7 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -734,6 +734,11 @@ struct nfs4_pathconf_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_pathconf_res {
+	struct nfs_pathconf	       *pathconf;
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs4_readdir_arg {
 	const struct nfs_fh *		fh;
 	u64				cookie;
-- 
1.6.2.1

[PATCH v2 15/67] nfs41: use nfs4_getaclres

[Benny Halevy]

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: embed resp_len in nfs_getaclres]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |   20 +++++++++++---------
 fs/nfs/nfs4xdr.c        |    5 +++--
 include/linux/nfs_xdr.h |    5 +++++
 3 files changed, 19 insertions(+), 11 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index ad6acc9..4270cb4 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2758,12 +2758,14 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 		.acl_pages = pages,
 		.acl_len = buflen,
 	};
-	size_t resp_len = buflen;
+	struct nfs_getaclres res = {
+		.acl_len = buflen,
+	};
 	void *resp_buf;
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETACL],
 		.rpc_argp = &args,
-		.rpc_resp = &resp_len,
+		.rpc_resp = &res,
 	};
 	struct page *localpage = NULL;
 	int ret;
@@ -2777,7 +2779,7 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 			return -ENOMEM;
 		args.acl_pages[0] = localpage;
 		args.acl_pgbase = 0;
-		resp_len = args.acl_len = PAGE_SIZE;
+		args.acl_len = PAGE_SIZE;
 	} else {
 		resp_buf = buf;
 		buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase);
@@ -2785,18 +2787,18 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 	ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
 	if (ret)
 		goto out_free;
-	if (resp_len > args.acl_len)
-		nfs4_write_cached_acl(inode, NULL, resp_len);
+	if (res.acl_len > args.acl_len)
+		nfs4_write_cached_acl(inode, NULL, res.acl_len);
 	else
-		nfs4_write_cached_acl(inode, resp_buf, resp_len);
+		nfs4_write_cached_acl(inode, resp_buf, res.acl_len);
 	if (buf) {
 		ret = -ERANGE;
-		if (resp_len > buflen)
+		if (res.acl_len > buflen)
 			goto out_free;
 		if (localpage)
-			memcpy(buf, resp_buf, resp_len);
+			memcpy(buf, resp_buf, res.acl_len);
 	}
-	ret = resp_len;
+	ret = res.acl_len;
 out_free:
 	if (localpage)
 		__free_page(localpage);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index a77ee3d..3e77789 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4018,7 +4018,8 @@ out:
  * Decode GETACL response
  */
 static int
-nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p, size_t *acl_len)
+nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p,
+		    struct nfs_getaclres *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4031,7 +4032,7 @@ nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p, size_t *acl_len)
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
-	status = decode_getacl(&xdr, rqstp, acl_len);
+	status = decode_getacl(&xdr, rqstp, &res->acl_len);
 
 out:
 	return status;
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 67942e7..b34a66e 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -444,6 +444,11 @@ struct nfs_getaclargs {
 	struct nfs4_sequence_args 	seq_args;
 };
 
+struct nfs_getaclres {
+	size_t				acl_len;
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs_setattrres {
 	struct nfs_fattr *              fattr;
 	const struct nfs_server *	server;
-- 
1.6.2.1

[PATCH v2 16/67] NFS: get rid of unused xdr decode_setattr(, res) argument

[Benny Halevy]

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4xdr.c |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 3e77789..27dd25d 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -3658,7 +3658,7 @@ decode_savefh(struct xdr_stream *xdr)
 	return decode_op_hdr(xdr, OP_SAVEFH);
 }
 
-static int decode_setattr(struct xdr_stream *xdr, struct nfs_setattrres *res)
+static int decode_setattr(struct xdr_stream *xdr)
 {
 	__be32 *p;
 	uint32_t bmlen;
@@ -4009,7 +4009,7 @@ nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p, void *res)
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
-	status = decode_setattr(&xdr, res);
+	status = decode_setattr(&xdr);
 out:
 	return status;
 }
@@ -4162,7 +4162,7 @@ static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_se
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
-	status = decode_setattr(&xdr, res);
+	status = decode_setattr(&xdr);
 	if (status)
 		goto out;
 	decode_getfattr(&xdr, res->fattr, res->server);
-- 
1.6.2.1

[PATCH v2 17/67] nfs41: use nfs4_setaclres

[Benny Halevy]

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs_setaclres]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |    3 ++-
 fs/nfs/nfs4xdr.c        |    3 ++-
 include/linux/nfs_xdr.h |    4 ++++
 3 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4270cb4..50ce9ed 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2845,10 +2845,11 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl
 		.acl_pages	= pages,
 		.acl_len	= buflen,
 	};
+	struct nfs_setaclres res;
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_SETACL],
 		.rpc_argp	= &arg,
-		.rpc_resp	= NULL,
+		.rpc_resp	= &res,
 	};
 	int ret;
 
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 27dd25d..aa350d5 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -3996,7 +3996,8 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args
  * Decode SETACL response
  */
 static int
-nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p, void *res)
+nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p,
+		    struct nfs_setaclres *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index b34a66e..d6cb194 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -436,6 +436,10 @@ struct nfs_setaclargs {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs_setaclres {
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs_getaclargs {
 	struct nfs_fh *			fh;
 	size_t				acl_len;
-- 
1.6.2.1

[PATCH v2 18/67] nfs41: use nfs4_fs_locations_res

[Benny Halevy]

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[find nfs4_fs_locations_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |    5 ++++-
 fs/nfs/nfs4xdr.c        |    6 ++++--
 include/linux/nfs_xdr.h |    5 +++++
 3 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 50ce9ed..2fa9442 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3725,10 +3725,13 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 		.page = page,
 		.bitmask = bitmask,
 	};
+	struct nfs4_fs_locations_res res = {
+		.fs_locations = fs_locations,
+	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FS_LOCATIONS],
 		.rpc_argp = &args,
-		.rpc_resp = fs_locations,
+		.rpc_resp = &res,
 	};
 	int status;
 
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index aa350d5..e448e33 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4504,7 +4504,8 @@ out:
 /*
  * FS_LOCATIONS request
  */
-static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs4_fs_locations *res)
+static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p,
+				     struct nfs4_fs_locations_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4519,7 +4520,8 @@ static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 	if ((status = decode_lookup(&xdr)) != 0)
 		goto out;
 	xdr_enter_page(&xdr, PAGE_SIZE);
-	status = decode_getfattr(&xdr, &res->fattr, res->server);
+	status = decode_getfattr(&xdr, &res->fs_locations->fattr,
+				 res->fs_locations->server);
 out:
 	return status;
 }
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index d6cb194..d32a112 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -867,6 +867,11 @@ struct nfs4_fs_locations_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_fs_locations_res {
+	struct nfs4_fs_locations       *fs_locations;
+	struct nfs4_sequence_res	seq_res;
+};
+
 #endif /* CONFIG_NFS_V4 */
 
 struct nfs_page;
-- 
1.6.2.1

[PATCH v2 19/67] nfs41: introduce nfs4_call_sync

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Use nfs4_call_sync rather than rpc_call_sync to provide
for a nfs41 sessions-enabled interface for sessions manipulation.

The nfs41 rpc logic uses the rpc_call_prepare method to
recover and create the session, as well as selecting a free slot id
and the rpc_call_done to free the slot and update slot table
related metadata.

In the coming patches we'll add rpc prepare and done routines
for setting up the sequence op and processing the sequence result.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_call_sync]
As per 11-14-08 review.
Squash into "nfs41: introduce nfs4_call_sync" and "nfs41: nfs4_setup_sequence"
Define two functions one for v4 and one for v41
add a pointer to struct nfs4_client to the correct one.
Signed-off-by: Andy Adamson <andros@netapp.com>
[added BUG() in _nfs4_call_sync_session if !CONFIG_NFS_V4_1]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[group minorversion specific stuff together]
Signed-off-by: Alexandros Batsakis <Alexandros.Batsakis@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: fixup nfs4_clear_client_minor_version]
[introduce nfs4_init_client_minor_version() in this patch]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[cleaned-up patch: got rid of nfs_call_sync_t, dprintks, cosmetics, extra server defs]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/client.c           |    5 +++
 fs/nfs/internal.h         |   12 ++++++++
 fs/nfs/nfs4proc.c         |   67 ++++++++++++++++++++++++++++++++-------------
 include/linux/nfs_fs_sb.h |    8 +++++
 4 files changed, 73 insertions(+), 19 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index efbaba3..562f658 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -194,6 +194,8 @@ static void nfs4_clear_client_minor_version(struct nfs_client *clp)
 		nfs4_destroy_session(clp->cl_session);
 		clp->cl_session = NULL;
 	}
+
+	clp->cl_call_sync = _nfs4_call_sync;
 #endif /* CONFIG_NFS_V4_1 */
 }
 
@@ -1073,6 +1075,8 @@ error:
  */
 static int nfs4_init_client_minor_version(struct nfs_client *clp)
 {
+	clp->cl_call_sync = _nfs4_call_sync;
+
 #if defined(CONFIG_NFS_V4_1)
 	if (clp->cl_minorversion) {
 		struct nfs4_session *session = NULL;
@@ -1086,6 +1090,7 @@ static int nfs4_init_client_minor_version(struct nfs_client *clp)
 			return -ENOMEM;
 
 		clp->cl_session = session;
+		clp->cl_call_sync = _nfs4_call_sync_session;
 	}
 #endif /* CONFIG_NFS_V4_1 */
 
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 7cef45d..8d67c28 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -206,6 +206,18 @@ extern int nfs4_path_walk(struct nfs_server *server,
 			  const char *path);
 #endif
 
+/* nfs4proc.c */
+extern int _nfs4_call_sync(struct nfs_server *server,
+			   struct rpc_message *msg,
+			   struct nfs4_sequence_args *args,
+			   struct nfs4_sequence_res *res,
+			   int cache_reply);
+extern int _nfs4_call_sync_session(struct nfs_server *server,
+				   struct rpc_message *msg,
+				   struct nfs4_sequence_args *args,
+				   struct nfs4_sequence_res *res,
+				   int cache_reply);
+
 /*
  * Determine if sessions are in use.
  */
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 2fa9442..95c0ace 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -271,6 +271,33 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
 	spin_unlock(&clp->cl_lock);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+
+int _nfs4_call_sync_session(struct nfs_server *server,
+			    struct rpc_message *msg,
+			    struct nfs4_sequence_args *args,
+			    struct nfs4_sequence_res *res,
+			    int cache_reply)
+{
+	/* in preparation for setting up the sequence op */
+	return rpc_call_sync(server->client, msg, 0);
+}
+
+#endif /* CONFIG_NFS_V4_1 */
+
+int _nfs4_call_sync(struct nfs_server *server,
+		    struct rpc_message *msg,
+		    struct nfs4_sequence_args *args,
+		    struct nfs4_sequence_res *res,
+		    int cache_reply)
+{
+	return rpc_call_sync(server->client, msg, 0);
+}
+
+#define nfs4_call_sync(server, msg, args, res, cache_reply) \
+	(server)->nfs_client->cl_call_sync((server), (msg), &(args)->seq_args, \
+			&(res)->seq_res, (cache_reply))
+
 static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
@@ -1269,7 +1296,7 @@ static int _nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
 	} else
 		memcpy(&arg.stateid, &zero_stateid, sizeof(arg.stateid));
 
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	if (status == 0 && state != NULL)
 		renew_lease(server, timestamp);
 	return status;
@@ -1595,7 +1622,7 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f
 	};
 	int status;
 
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	if (status == 0) {
 		memcpy(server->attr_bitmask, res.attr_bitmask, sizeof(server->attr_bitmask));
 		if (res.attr_bitmask[0] & FATTR4_WORD0_ACL)
@@ -1609,6 +1636,7 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f
 		server->cache_consistency_bitmask[1] &= FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY;
 		server->acl_bitmask = res.acl_bitmask;
 	}
+
 	return status;
 }
 
@@ -1641,7 +1669,7 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
 		.rpc_resp = &res,
 	};
 	nfs_fattr_init(info->fattr);
-	return rpc_call_sync(server->client, &msg, 0);
+	return nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
@@ -1731,7 +1759,7 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
 	};
 	
 	nfs_fattr_init(fattr);
-	return rpc_call_sync(server->client, &msg, 0);
+	return nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr)
@@ -1815,7 +1843,7 @@ static int _nfs4_proc_lookupfh(struct nfs_server *server, const struct nfs_fh *d
 	nfs_fattr_init(fattr);
 
 	dprintk("NFS call  lookupfh %s\n", name->name);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	dprintk("NFS reply lookupfh: %d\n", status);
 	return status;
 }
@@ -1901,7 +1929,7 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry
 			args.access |= NFS4_ACCESS_EXECUTE;
 	}
 	nfs_fattr_init(&fattr);
-	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	if (!status) {
 		entry->mask = 0;
 		if (res.access & NFS4_ACCESS_READ)
@@ -1967,7 +1995,7 @@ static int _nfs4_proc_readlink(struct inode *inode, struct page *page,
 		.rpc_resp = &res,
 	};
 
-	return rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	return nfs4_call_sync(NFS_SERVER(inode), &msg, &args, &res, 0);
 }
 
 static int nfs4_proc_readlink(struct inode *inode, struct page *page,
@@ -2061,7 +2089,7 @@ static int _nfs4_proc_remove(struct inode *dir, struct qstr *name)
 	int			status;
 
 	nfs_fattr_init(&res.dir_attr);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 1);
 	if (status == 0) {
 		update_changeattr(dir, &res.cinfo);
 		nfs_post_op_update_inode(dir, &res.dir_attr);
@@ -2129,7 +2157,7 @@ static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name,
 	
 	nfs_fattr_init(res.old_fattr);
 	nfs_fattr_init(res.new_fattr);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 
 	if (!status) {
 		update_changeattr(old_dir, &res.old_cinfo);
@@ -2178,7 +2206,7 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *
 
 	nfs_fattr_init(res.fattr);
 	nfs_fattr_init(res.dir_attr);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	if (!status) {
 		update_changeattr(dir, &res.cinfo);
 		nfs_post_op_update_inode(dir, res.dir_attr);
@@ -2239,7 +2267,8 @@ static struct nfs4_createdata *nfs4_alloc_createdata(struct inode *dir,
 
 static int nfs4_do_create(struct inode *dir, struct dentry *dentry, struct nfs4_createdata *data)
 {
-	int status = rpc_call_sync(NFS_CLIENT(dir), &data->msg, 0);
+	int status = nfs4_call_sync(NFS_SERVER(dir), &data->msg,
+				    &data->arg, &data->res, 1);
 	if (status == 0) {
 		update_changeattr(dir, &data->res.dir_cinfo);
 		nfs_post_op_update_inode(dir, data->res.dir_fattr);
@@ -2348,7 +2377,7 @@ static int _nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
 			(unsigned long long)cookie);
 	nfs4_setup_readdir(cookie, NFS_COOKIEVERF(dir), dentry, &args);
 	res.pgbase = args.pgbase;
-	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	status = nfs4_call_sync(NFS_SERVER(dir), &msg, &args, &res, 0);
 	if (status == 0)
 		memcpy(NFS_COOKIEVERF(dir), res.verifier.data, NFS4_VERIFIER_SIZE);
 
@@ -2436,7 +2465,7 @@ static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
 	};
 
 	nfs_fattr_init(fsstat->fattr);
-	return rpc_call_sync(server->client, &msg, 0);
+	return  nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsstat *fsstat)
@@ -2467,7 +2496,7 @@ static int _nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
 		.rpc_resp = &res,
 	};
 
-	return rpc_call_sync(server->client, &msg, 0);
+	return nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo)
@@ -2512,7 +2541,7 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle
 	}
 
 	nfs_fattr_init(pathconf->fattr);
-	return rpc_call_sync(server->client, &msg, 0);
+	return nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
@@ -2784,7 +2813,7 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 		resp_buf = buf;
 		buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase);
 	}
-	ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	ret = nfs4_call_sync(NFS_SERVER(inode), &msg, &args, &res, 0);
 	if (ret)
 		goto out_free;
 	if (res.acl_len > args.acl_len)
@@ -2857,7 +2886,7 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl
 		return -EOPNOTSUPP;
 	nfs_inode_return_delegation(inode);
 	buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
-	ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	ret = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	nfs_access_zap_cache(inode);
 	nfs_zap_acl_cache(inode);
 	return ret;
@@ -3146,7 +3175,7 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock
 		goto out;
 	lsp = request->fl_u.nfs4_fl.owner;
 	arg.lock_owner.id = lsp->ls_id.id;
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	switch (status) {
 		case 0:
 			request->fl_type = F_UNLCK;
@@ -3739,7 +3768,7 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 	nfs_fattr_init(&fs_locations->fattr);
 	fs_locations->server = server;
 	fs_locations->nlocations = 0;
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	nfs_fixup_referral_attributes(&fs_locations->fattr);
 	dprintk("%s: returned status = %d\n", __func__, status);
 	return status;
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 283ef0a..d541a49 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -12,6 +12,9 @@
 struct nfs4_session;
 struct nfs_iostats;
 struct nlm_host;
+struct nfs4_sequence_args;
+struct nfs4_sequence_res;
+struct nfs_server;
 
 /*
  * The nfs_client identifies our client state to the server.
@@ -67,6 +70,11 @@ struct nfs_client {
 	char			cl_ipaddr[48];
 	unsigned char		cl_id_uniquifier;
 	u32			cl_minorversion;
+	int		     (* cl_call_sync)(struct nfs_server *server,
+					      struct rpc_message *msg,
+					      struct nfs4_sequence_args *args,
+					      struct nfs4_sequence_res *res,
+					      int cache_reply);
 #endif /* CONFIG_NFS_V4 */
 
 #ifdef CONFIG_NFS_V4_1
-- 
1.6.2.1

[PATCH v2 20/67] nfs41: nfs41: pass *session in seq_args and seq_res

[Benny Halevy]

To be used for getting the rpc's minorversion and for nfs41 xdr
{en,de}coding of the sequence operation.
Reset the seq session ptrs for minorversion=0 rpc calls.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |    1 +
 include/linux/nfs_xdr.h |    4 ++--
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 95c0ace..df2a786 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -291,6 +291,7 @@ int _nfs4_call_sync(struct nfs_server *server,
 		    struct nfs4_sequence_res *res,
 		    int cache_reply)
 {
+	args->sa_session = res->sr_session = NULL;
 	return rpc_call_sync(server->client, msg, 0);
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index d32a112..f933413 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -162,11 +162,11 @@ struct nfs4_slot {
 };
 
 struct nfs4_sequence_args {
-	/* stub */
+	struct nfs4_session	*sa_session;
 };
 
 struct nfs4_sequence_res {
-	/* stub */
+	struct nfs4_session	*sr_session;
 };
 
 /*
-- 
1.6.2.1

[PATCH v2 21/67] nfs41: set up seq_res.sr_slotid

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Initialize nfs4_sequence_res sr_slotid to NFS4_MAX_SLOT_TABLE.

[was nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
[pulled definition of struct nfs4_sequence_res.sr_slotid to here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |    5 +++++
 fs/nfs/read.c           |    1 +
 fs/nfs/unlink.c         |    1 +
 fs/nfs/write.c          |    2 ++
 include/linux/nfs_xdr.h |    1 +
 5 files changed, 10 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index df2a786..2e70c7e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -371,6 +371,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct path *path,
 	p->o_arg.server = server;
 	p->o_arg.bitmask = server->attr_bitmask;
 	p->o_arg.claim = NFS4_OPEN_CLAIM_NULL;
+	p->o_res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	if (flags & O_EXCL) {
 		u32 *s = (u32 *) p->o_arg.u.verifier.data;
 		s[0] = jiffies;
@@ -1463,6 +1464,7 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	calldata->res.fattr = &calldata->fattr;
 	calldata->res.seqid = calldata->arg.seqid;
 	calldata->res.server = server;
+	calldata->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	calldata->path.mnt = mntget(path->mnt);
 	calldata->path.dentry = dget(path->dentry);
 
@@ -3091,6 +3093,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	memcpy(&data->stateid, stateid, sizeof(data->stateid));
 	data->res.fattr = &data->fattr;
 	data->res.server = server;
+	data->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	nfs_fattr_init(data->res.fattr);
 	data->timestamp = jiffies;
 	data->rpc_status = 0;
@@ -3243,6 +3246,7 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl,
 	p->arg.fl = &p->fl;
 	p->arg.seqid = seqid;
 	p->res.seqid = seqid;
+	p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	p->arg.stateid = &lsp->ls_stateid;
 	p->lsp = lsp;
 	atomic_inc(&lsp->ls_count);
@@ -3415,6 +3419,7 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
 	p->arg.lock_owner.clientid = server->nfs_client->cl_clientid;
 	p->arg.lock_owner.id = lsp->ls_id.id;
 	p->res.lock_seqid = p->arg.lock_seqid;
+	p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	p->lsp = lsp;
 	atomic_inc(&lsp->ls_count);
 	p->ctx = get_nfs_open_context(ctx);
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 4ace3c5..70ba2b4 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -46,6 +46,7 @@ struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount)
 		memset(p, 0, sizeof(*p));
 		INIT_LIST_HEAD(&p->pages);
 		p->npages = pagecount;
+		p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 		if (pagecount <= ARRAY_SIZE(p->page_array))
 			p->pagevec = p->page_array;
 		else {
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index ecc2953..83d0ce2 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -241,6 +241,7 @@ nfs_async_unlink(struct inode *dir, struct dentry *dentry)
 		status = PTR_ERR(data->cred);
 		goto out_free;
 	}
+	data->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 
 	status = -EBUSY;
 	spin_lock(&dentry->d_lock);
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index e560a78..035e6fb 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -52,6 +52,7 @@ struct nfs_write_data *nfs_commitdata_alloc(void)
 	if (p) {
 		memset(p, 0, sizeof(*p));
 		INIT_LIST_HEAD(&p->pages);
+		p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	}
 	return p;
 }
@@ -71,6 +72,7 @@ struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount)
 		memset(p, 0, sizeof(*p));
 		INIT_LIST_HEAD(&p->pages);
 		p->npages = pagecount;
+		p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 		if (pagecount <= ARRAY_SIZE(p->page_array))
 			p->pagevec = p->page_array;
 		else {
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index f933413..892fd27 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -167,6 +167,7 @@ struct nfs4_sequence_args {
 
 struct nfs4_sequence_res {
 	struct nfs4_session	*sr_session;
+	u8			sr_slotid;	/* slot used to send request */
 };
 
 /*
-- 
1.6.2.1

[PATCH v2 22/67] NFS: use decode_change_info_maxsz for xdr maxsz calculations

[Benny Halevy]

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4xdr.c |   10 +++++++---
 1 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index e448e33..487a197 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -192,12 +192,16 @@ static int nfs4_stat_to_errno(int);
 				 decode_verifier_maxsz)
 #define encode_remove_maxsz	(op_encode_hdr_maxsz + \
 				nfs4_name_maxsz)
+#define decode_remove_maxsz	(op_decode_hdr_maxsz + \
+				 decode_change_info_maxsz)
 #define encode_rename_maxsz	(op_encode_hdr_maxsz + \
 				2 * nfs4_name_maxsz)
-#define decode_rename_maxsz	(op_decode_hdr_maxsz + 5 + 5)
+#define decode_rename_maxsz	(op_decode_hdr_maxsz + \
+				 decode_change_info_maxsz + \
+				 decode_change_info_maxsz)
 #define encode_link_maxsz	(op_encode_hdr_maxsz + \
 				nfs4_name_maxsz)
-#define decode_link_maxsz	(op_decode_hdr_maxsz + 5)
+#define decode_link_maxsz	(op_decode_hdr_maxsz + decode_change_info_maxsz)
 #define encode_lock_maxsz	(op_encode_hdr_maxsz + \
 				 7 + \
 				 1 + encode_stateid_maxsz + 8)
@@ -414,7 +418,7 @@ static int nfs4_stat_to_errno(int);
 				encode_getattr_maxsz)
 #define NFS4_dec_remove_sz	(compound_decode_hdr_maxsz + \
 				decode_putfh_maxsz + \
-				op_decode_hdr_maxsz + 5 + \
+				decode_remove_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_rename_sz	(compound_encode_hdr_maxsz + \
 				encode_putfh_maxsz + \
-- 
1.6.2.1

[PATCH v2 23/67] NFS: define and initialize compound_hdr.replen

[Benny Halevy]

replen holds the running count of expected reply bytes.
repl will then be used by encoding routines for xdr_inline_pages offset
after which data bytes are to be received directly into the xdr
buffer pages.

NOTE: According to the nfsv4 and v4.1 RFCs, the replied tag SHOULD be the same
is the one sent, but this is not required as a MUST for the server to do so.
The server may screw us if it replies a tag of a different length in the
compound result.

[NFS: cb_compoundhdr.replen is in words not bytes]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4xdr.c |   77 ++++++++++++++++++++++++++++++------------------------
 1 files changed, 43 insertions(+), 34 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 487a197..c85dbee 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -545,6 +545,7 @@ struct compound_hdr {
 	__be32 *	nops_p;
 	uint32_t	taglen;
 	char *		tag;
+	uint32_t	replen;		/* expected reply words */
 };
 
 /*
@@ -580,9 +581,17 @@ static void encode_string(struct xdr_stream *xdr, unsigned int len, const char *
 	xdr_encode_opaque(p, str, len);
 }
 
-static void encode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr)
+static void encode_compound_hdr(struct xdr_stream *xdr,
+				struct rpc_rqst *req,
+				struct compound_hdr *hdr)
 {
 	__be32 *p;
+	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
+
+	/* initialize running count of expected bytes in reply.
+	 * NOTE: the replied tag SHOULD be the same is the one sent,
+	 * but this is not required as a MUST for the server to do so. */
+	hdr->replen = RPC_REPHDRSIZE + auth->au_rslack + 3 + hdr->taglen;
 
 	dprintk("encode_compound: tag=%.*s\n", (int)hdr->taglen, hdr->tag);
 	BUG_ON(hdr->taglen > NFS4_MAXTAGLEN);
@@ -1359,7 +1368,7 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_access(&xdr, args->access, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1378,7 +1387,7 @@ static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
 	encode_getfh(&xdr, &hdr);
@@ -1398,7 +1407,7 @@ static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, __be32 *p, const struc
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putrootfh(&xdr, &hdr);
 	encode_getfh(&xdr, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1417,7 +1426,7 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_remove(&xdr, &args->name, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1436,7 +1445,7 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->old_dir, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_putfh(&xdr, args->new_dir, &hdr);
@@ -1459,7 +1468,7 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, __be32 *p, const struct nfs4_
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
@@ -1482,7 +1491,7 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_create(&xdr, args, &hdr);
@@ -1513,7 +1522,7 @@ static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, __be32 *p, const struct nf
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
 	encode_nops(&hdr);
@@ -1531,7 +1540,7 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, __be32 *p, struct nfs_closea
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_close(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1550,7 +1559,7 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, __be32 *p, struct nfs_openarg
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_open(&xdr, args, &hdr);
@@ -1573,7 +1582,7 @@ static int nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, __be32 *p, struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open_confirm(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1591,7 +1600,7 @@ static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, __be32 *p, struct nfs_
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1610,7 +1619,7 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, __be32 *p, struct n
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open_downgrade(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1629,7 +1638,7 @@ static int nfs4_xdr_enc_lock(struct rpc_rqst *req, __be32 *p, struct nfs_lock_ar
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_lock(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1647,7 +1656,7 @@ static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, __be32 *p, struct nfs_lockt_
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_lockt(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1665,7 +1674,7 @@ static int nfs4_xdr_enc_locku(struct rpc_rqst *req, __be32 *p, struct nfs_locku_
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_locku(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1685,7 +1694,7 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n
 	unsigned int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readlink(&xdr, args, req, &hdr);
 
@@ -1713,7 +1722,7 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
 	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readdir(&xdr, args, req, &hdr);
 
@@ -1744,7 +1753,7 @@ static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readarg
 	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_read(&xdr, args, &hdr);
 
@@ -1771,7 +1780,7 @@ static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, __be32 *p, struct nfs_seta
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_setattr(&xdr, args, args->server, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1794,7 +1803,7 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0, &hdr);
 
@@ -1817,7 +1826,7 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writea
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_write(&xdr, args, &hdr);
 	req->rq_snd_buf.flags |= XDRBUF_WRITE;
@@ -1837,7 +1846,7 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, __be32 *p, struct nfs_write
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_commit(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1856,7 +1865,7 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs4_fsin
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_fsinfo(&xdr, args->bitmask, &hdr);
 	encode_nops(&hdr);
@@ -1874,7 +1883,7 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, __be32 *p, const struct n
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_one(&xdr, args->bitmask[0] & nfs4_pathconf_bitmap[0],
 			   &hdr);
@@ -1893,7 +1902,7 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_two(&xdr, args->bitmask[0] & nfs4_statfs_bitmap[0],
 			   args->bitmask[1] & nfs4_statfs_bitmap[1], &hdr);
@@ -1913,7 +1922,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p,
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_getattr_one(&xdr, FATTR4_WORD0_SUPPORTED_ATTRS|
 			   FATTR4_WORD0_LINK_SUPPORT|
@@ -1934,7 +1943,7 @@ static int nfs4_xdr_enc_renew(struct rpc_rqst *req, __be32 *p, struct nfs_client
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_renew(&xdr, clp, &hdr);
 	encode_nops(&hdr);
 	return 0;
@@ -1951,7 +1960,7 @@ static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, __be32 *p, struct nfs4
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_setclientid(&xdr, sc, &hdr);
 	encode_nops(&hdr);
 	return 0;
@@ -1969,7 +1978,7 @@ static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, __be32 *p, str
 	const u32 lease_bitmap[2] = { FATTR4_WORD0_LEASE_TIME, 0 };
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_setclientid_confirm(&xdr, clp, &hdr);
 	encode_putrootfh(&xdr, &hdr);
 	encode_fsinfo(&xdr, lease_bitmap, &hdr);
@@ -1988,7 +1997,7 @@ static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, __be32 *p, const struc
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_delegreturn(&xdr, args->stateid, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -2009,7 +2018,7 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
 	encode_fs_locations(&xdr, args->bitmask, &hdr);
@@ -3989,7 +3998,7 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args
 	int status;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	status = encode_setacl(&xdr, args, &hdr);
 	encode_nops(&hdr);
-- 
1.6.2.1

[PATCH v2 24/67] NFS: update hdr->replen for every encode op

[Benny Halevy]

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4xdr.c |   31 +++++++++++++++++++++++++++++++
 1 files changed, 31 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index c85dbee..746c6a5 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -749,6 +749,7 @@ static void encode_access(struct xdr_stream *xdr, u32 access, struct compound_hd
 	WRITE32(OP_ACCESS);
 	WRITE32(access);
 	hdr->nops++;
+	hdr->replen += decode_access_maxsz;
 }
 
 static void encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg, struct compound_hdr *hdr)
@@ -760,6 +761,7 @@ static void encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg
 	WRITE32(arg->seqid->sequence->counter);
 	WRITEMEM(arg->stateid->data, NFS4_STATEID_SIZE);
 	hdr->nops++;
+	hdr->replen += decode_close_maxsz;
 }
 
 static void encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *args, struct compound_hdr *hdr)
@@ -771,6 +773,7 @@ static void encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *ar
 	WRITE64(args->offset);
 	WRITE32(args->count);
 	hdr->nops++;
+	hdr->replen += decode_commit_maxsz;
 }
 
 static void encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *create, struct compound_hdr *hdr)
@@ -802,6 +805,7 @@ static void encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *
 	WRITE32(create->name->len);
 	WRITEMEM(create->name->name, create->name->len);
 	hdr->nops++;
+	hdr->replen += decode_create_maxsz;
 
 	encode_attrs(xdr, create->attrs, create->server);
 }
@@ -815,6 +819,7 @@ static void encode_getattr_one(struct xdr_stream *xdr, uint32_t bitmap, struct c
 	WRITE32(1);
 	WRITE32(bitmap);
 	hdr->nops++;
+	hdr->replen += decode_getattr_maxsz;
 }
 
 static void encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm1, struct compound_hdr *hdr)
@@ -827,6 +832,7 @@ static void encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm
 	WRITE32(bm0);
 	WRITE32(bm1);
 	hdr->nops++;
+	hdr->replen += decode_getattr_maxsz;
 }
 
 static void encode_getfattr(struct xdr_stream *xdr, const u32* bitmask, struct compound_hdr *hdr)
@@ -854,6 +860,7 @@ static void encode_getfh(struct xdr_stream *xdr, struct compound_hdr *hdr)
 	RESERVE_SPACE(4);
 	WRITE32(OP_GETFH);
 	hdr->nops++;
+	hdr->replen += decode_getfh_maxsz;
 }
 
 static void encode_link(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr)
@@ -865,6 +872,7 @@ static void encode_link(struct xdr_stream *xdr, const struct qstr *name, struct
 	WRITE32(name->len);
 	WRITEMEM(name->name, name->len);
 	hdr->nops++;
+	hdr->replen += decode_link_maxsz;
 }
 
 static inline int nfs4_lock_type(struct file_lock *fl, int block)
@@ -912,6 +920,7 @@ static void encode_lock(struct xdr_stream *xdr, const struct nfs_lock_args *args
 		WRITE32(args->lock_seqid->sequence->counter);
 	}
 	hdr->nops++;
+	hdr->replen += decode_lock_maxsz;
 }
 
 static void encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *args, struct compound_hdr *hdr)
@@ -928,6 +937,7 @@ static void encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *ar
 	WRITEMEM("lock id:", 8);
 	WRITE64(args->lock_owner.id);
 	hdr->nops++;
+	hdr->replen += decode_lockt_maxsz;
 }
 
 static void encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *args, struct compound_hdr *hdr)
@@ -942,6 +952,7 @@ static void encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *ar
 	WRITE64(args->fl->fl_start);
 	WRITE64(nfs4_lock_length(args->fl));
 	hdr->nops++;
+	hdr->replen += decode_locku_maxsz;
 }
 
 static void encode_lookup(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr)
@@ -954,6 +965,7 @@ static void encode_lookup(struct xdr_stream *xdr, const struct qstr *name, struc
 	WRITE32(len);
 	WRITEMEM(name->name, len);
 	hdr->nops++;
+	hdr->replen += decode_lookup_maxsz;
 }
 
 static void encode_share_access(struct xdr_stream *xdr, fmode_t fmode)
@@ -1093,6 +1105,7 @@ static void encode_open(struct xdr_stream *xdr, const struct nfs_openargs *arg,
 		BUG();
 	}
 	hdr->nops++;
+	hdr->replen += decode_open_maxsz;
 }
 
 static void encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_confirmargs *arg, struct compound_hdr *hdr)
@@ -1104,6 +1117,7 @@ static void encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_co
 	WRITEMEM(arg->stateid->data, NFS4_STATEID_SIZE);
 	WRITE32(arg->seqid->sequence->counter);
 	hdr->nops++;
+	hdr->replen += decode_open_confirm_maxsz;
 }
 
 static void encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_closeargs *arg, struct compound_hdr *hdr)
@@ -1116,6 +1130,7 @@ static void encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_close
 	WRITE32(arg->seqid->sequence->counter);
 	encode_share_access(xdr, arg->fmode);
 	hdr->nops++;
+	hdr->replen += decode_open_downgrade_maxsz;
 }
 
 static void
@@ -1129,6 +1144,7 @@ encode_putfh(struct xdr_stream *xdr, const struct nfs_fh *fh, struct compound_hd
 	WRITE32(len);
 	WRITEMEM(fh->data, len);
 	hdr->nops++;
+	hdr->replen += decode_putfh_maxsz;
 }
 
 static void encode_putrootfh(struct xdr_stream *xdr, struct compound_hdr *hdr)
@@ -1138,6 +1154,7 @@ static void encode_putrootfh(struct xdr_stream *xdr, struct compound_hdr *hdr)
 	RESERVE_SPACE(4);
 	WRITE32(OP_PUTROOTFH);
 	hdr->nops++;
+	hdr->replen += decode_putrootfh_maxsz;
 }
 
 static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context *ctx)
@@ -1166,6 +1183,7 @@ static void encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args,
 	WRITE64(args->offset);
 	WRITE32(args->count);
 	hdr->nops++;
+	hdr->replen += decode_read_maxsz;
 }
 
 static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg *readdir, struct rpc_rqst *req, struct compound_hdr *hdr)
@@ -1191,6 +1209,7 @@ static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg
 	WRITE32(attrs[0] & readdir->bitmask[0]);
 	WRITE32(attrs[1] & readdir->bitmask[1]);
 	hdr->nops++;
+	hdr->replen += decode_readdir_maxsz;
 	dprintk("%s: cookie = %Lu, verifier = %08x:%08x, bitmap = %08x:%08x\n",
 			__func__,
 			(unsigned long long)readdir->cookie,
@@ -1207,6 +1226,7 @@ static void encode_readlink(struct xdr_stream *xdr, const struct nfs4_readlink *
 	RESERVE_SPACE(4);
 	WRITE32(OP_READLINK);
 	hdr->nops++;
+	hdr->replen += decode_readlink_maxsz;
 }
 
 static void encode_remove(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr)
@@ -1218,6 +1238,7 @@ static void encode_remove(struct xdr_stream *xdr, const struct qstr *name, struc
 	WRITE32(name->len);
 	WRITEMEM(name->name, name->len);
 	hdr->nops++;
+	hdr->replen += decode_remove_maxsz;
 }
 
 static void encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, const struct qstr *newname, struct compound_hdr *hdr)
@@ -1233,6 +1254,7 @@ static void encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, co
 	WRITE32(newname->len);
 	WRITEMEM(newname->name, newname->len);
 	hdr->nops++;
+	hdr->replen += decode_rename_maxsz;
 }
 
 static void encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_stateid, struct compound_hdr *hdr)
@@ -1243,6 +1265,7 @@ static void encode_renew(struct xdr_stream *xdr, const struct nfs_client *client
 	WRITE32(OP_RENEW);
 	WRITE64(client_stateid->cl_clientid);
 	hdr->nops++;
+	hdr->replen += decode_renew_maxsz;
 }
 
 static void
@@ -1253,6 +1276,7 @@ encode_restorefh(struct xdr_stream *xdr, struct compound_hdr *hdr)
 	RESERVE_SPACE(4);
 	WRITE32(OP_RESTOREFH);
 	hdr->nops++;
+	hdr->replen += decode_restorefh_maxsz;
 }
 
 static int
@@ -1272,6 +1296,7 @@ encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg, struct compoun
 	WRITE32(arg->acl_len);
 	xdr_write_pages(xdr, arg->acl_pages, arg->acl_pgbase, arg->acl_len);
 	hdr->nops++;
+	hdr->replen += decode_setacl_maxsz;
 	return 0;
 }
 
@@ -1283,6 +1308,7 @@ encode_savefh(struct xdr_stream *xdr, struct compound_hdr *hdr)
 	RESERVE_SPACE(4);
 	WRITE32(OP_SAVEFH);
 	hdr->nops++;
+	hdr->replen += decode_savefh_maxsz;
 }
 
 static void encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs *arg, const struct nfs_server *server, struct compound_hdr *hdr)
@@ -1293,6 +1319,7 @@ static void encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs
 	WRITE32(OP_SETATTR);
 	WRITEMEM(arg->stateid.data, NFS4_STATEID_SIZE);
 	hdr->nops++;
+	hdr->replen += decode_setattr_maxsz;
 	encode_attrs(xdr, arg->iap, server);
 }
 
@@ -1312,6 +1339,7 @@ static void encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclie
 	RESERVE_SPACE(4);
 	WRITE32(setclientid->sc_cb_ident);
 	hdr->nops++;
+	hdr->replen += decode_setclientid_maxsz;
 }
 
 static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_client *client_state, struct compound_hdr *hdr)
@@ -1323,6 +1351,7 @@ static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_
 	WRITE64(client_state->cl_clientid);
 	WRITEMEM(client_state->cl_confirm.data, NFS4_VERIFIER_SIZE);
 	hdr->nops++;
+	hdr->replen += decode_setclientid_confirm_maxsz;
 }
 
 static void encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *args, struct compound_hdr *hdr)
@@ -1341,6 +1370,7 @@ static void encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *arg
 
 	xdr_write_pages(xdr, args->pages, args->pgbase, args->count);
 	hdr->nops++;
+	hdr->replen += decode_write_maxsz;
 }
 
 static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *stateid, struct compound_hdr *hdr)
@@ -1352,6 +1382,7 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state
 	WRITE32(OP_DELEGRETURN);
 	WRITEMEM(stateid->data, NFS4_STATEID_SIZE);
 	hdr->nops++;
+	hdr->replen += decode_delegreturn_maxsz;
 }
 /*
  * END OF "GENERIC" ENCODE ROUTINES.
-- 
1.6.2.1

[PATCH v2 25/67] NFS: use dynamically computed compound_hdr.replen for xdr_inline_pages offset

[Benny Halevy]

As Trond suggested, rather than passing a constant to xdr_inline_pages,
keep a running count of the expected reply bytes.  In preparation for
nfs41, where additional op sequence are expteced when talking to nfs41
servers.

[NFS: cb_compoundhdr.replen is in words not bytes]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: get fs_locations replen before encoding the GETATTR]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: get getacl replen before encoding the GETATTR]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4xdr.c |   48 ++++++++++--------------------------------------
 1 files changed, 10 insertions(+), 38 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 746c6a5..6888946 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1721,20 +1721,13 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n
 	struct compound_hdr hdr = {
 		.nops = 0,
 	};
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
-	unsigned int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readlink(&xdr, args, req, &hdr);
 
-	/* set up reply kvec
-	 *    toplevel_status + taglen + rescount + OP_PUTFH + status
-	 *      + OP_READLINK + status + string length = 8
-	 */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_readlink_sz) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages,
+	xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2, args->pages,
 			args->pgbase, args->pglen);
 	encode_nops(&hdr);
 	return 0;
@@ -1749,23 +1742,16 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
 	struct compound_hdr hdr = {
 		.nops = 0,
 	};
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
-	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readdir(&xdr, args, req, &hdr);
 
-	/* set up reply kvec
-	 *    toplevel_status + taglen + rescount + OP_PUTFH + status
-	 *      + OP_READDIR + status + verifer(2)  = 9
-	 */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_readdir_sz) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages,
+	xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2, args->pages,
 			 args->pgbase, args->count);
 	dprintk("%s: inlined page args = (%u, %p, %u, %u)\n",
-			__func__, replen, args->pages,
+			__func__, hdr.replen << 2, args->pages,
 			args->pgbase, args->count);
 	encode_nops(&hdr);
 	return 0;
@@ -1776,24 +1762,17 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
  */
 static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args)
 {
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
 		.nops = 0,
 	};
-	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_read(&xdr, args, &hdr);
 
-	/* set up reply kvec
-	 *    toplevel status + taglen=0 + rescount + OP_PUTFH + status
-	 *       + OP_READ + status + eof + datalen = 9
-	 */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_read_sz) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen,
+	xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2,
 			 args->pages, args->pgbase, args->count);
 	req->rq_rcv_buf.flags |= XDRBUF_READ;
 	encode_nops(&hdr);
@@ -1827,20 +1806,18 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 		struct nfs_getaclargs *args)
 {
 	struct xdr_stream xdr;
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
 	struct compound_hdr hdr = {
 		.nops   = 0,
 	};
-	int replen;
+	uint32_t replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
+	replen = hdr.replen + nfs4_fattr_bitmap_maxsz + 1;
 	encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0, &hdr);
 
-	/* set up reply buffer: */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_getacl_sz) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen,
+	xdr_inline_pages(&req->rq_rcv_buf, replen << 2,
 		args->acl_pages, args->acl_pgbase, args->acl_len);
 	encode_nops(&hdr);
 	return 0;
@@ -2045,21 +2022,16 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 	struct compound_hdr hdr = {
 		.nops = 0,
 	};
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
-	int replen;
+	uint32_t replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
+	replen = hdr.replen;	/* get the attribute into args->page */
 	encode_fs_locations(&xdr, args->bitmask, &hdr);
 
-	/* set up reply
-	 *   toplevel_status + OP_PUTFH + status
-	 *   + OP_LOOKUP + status + OP_GETATTR + status = 7
-	 */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + 7) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen, &args->page,
+	xdr_inline_pages(&req->rq_rcv_buf, replen << 2, &args->page,
 			0, PAGE_SIZE);
 	encode_nops(&hdr);
 	return 0;
-- 
1.6.2.1

[PATCH v2 26/67] nfs41: encode minorversion in compound header

[Benny Halevy]

Signed-off-by: Andy Adamdon <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4xdr.c |   70 ++++++++++++++++++++++++++++++-----------------------
 1 files changed, 40 insertions(+), 30 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 6888946..19ca0b5 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -546,6 +546,7 @@ struct compound_hdr {
 	uint32_t	taglen;
 	char *		tag;
 	uint32_t	replen;		/* expected reply words */
+	u32		minorversion;
 };
 
 /*
@@ -598,7 +599,7 @@ static void encode_compound_hdr(struct xdr_stream *xdr,
 	RESERVE_SPACE(12+(XDR_QUADLEN(hdr->taglen)<<2));
 	WRITE32(hdr->taglen);
 	WRITEMEM(hdr->tag, hdr->taglen);
-	WRITE32(NFS4_MINOR_VERSION);
+	WRITE32(hdr->minorversion);
 	hdr->nops_p = p;
 	WRITE32(hdr->nops);
 }
@@ -1388,6 +1389,15 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state
  * END OF "GENERIC" ENCODE ROUTINES.
  */
 
+static u32 nfs4_xdr_minorversion(const struct nfs4_sequence_args *args)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (args->sa_session)
+		return args->sa_session->clp->cl_minorversion;
+#endif /* CONFIG_NFS_V4_1 */
+	return 0;
+}
+
 /*
  * Encode an ACCESS request
  */
@@ -1395,7 +1405,7 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1414,7 +1424,7 @@ static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1434,7 +1444,7 @@ static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, __be32 *p, const struc
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1453,7 +1463,7 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1472,7 +1482,7 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1495,7 +1505,7 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, __be32 *p, const struct nfs4_
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1518,7 +1528,7 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1549,7 +1559,7 @@ static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, __be32 *p, const struct nf
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1567,7 +1577,7 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, __be32 *p, struct nfs_closea
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1586,7 +1596,7 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, __be32 *p, struct nfs_openarg
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1627,7 +1637,7 @@ static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, __be32 *p, struct nfs_
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1646,7 +1656,7 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, __be32 *p, struct n
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops	= 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1665,7 +1675,7 @@ static int nfs4_xdr_enc_lock(struct rpc_rqst *req, __be32 *p, struct nfs_lock_ar
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1683,7 +1693,7 @@ static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, __be32 *p, struct nfs_lockt_
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1701,7 +1711,7 @@ static int nfs4_xdr_enc_locku(struct rpc_rqst *req, __be32 *p, struct nfs_locku_
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1719,7 +1729,7 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1740,7 +1750,7 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1764,7 +1774,7 @@ static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readarg
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1786,7 +1796,7 @@ static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, __be32 *p, struct nfs_seta
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1807,7 +1817,7 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 	uint32_t replen;
 
@@ -1830,7 +1840,7 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writea
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1850,7 +1860,7 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, __be32 *p, struct nfs_write
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1869,7 +1879,7 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs4_fsin
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops	= 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1887,7 +1897,7 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, __be32 *p, const struct n
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1906,7 +1916,7 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1926,7 +1936,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p,
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -2001,7 +2011,7 @@ static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, __be32 *p, const struc
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -2020,7 +2030,7 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 	uint32_t replen;
 
@@ -3996,7 +4006,7 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 	int status;
 
-- 
1.6.2.1

[PATCH v2 27/67] nfs41: xdr {encode,decode}_sequence

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Implement stubs for encode and decode sequence, defined as no-ops when
CONFIG_NFS_V4_1 is not defined.
Add the nfsv41 encode and decode sizes. Add encode_sequence to all
nfs4_enc_* routines and decode_sequence to all nfs4_dec_* routines as required
by v41.

[was nfs41: minorversion support for xdr]
[added nfs_client argument to encode_sequence so not to use sequence_args to pass sa_session]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4xdr.c |  242 +++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 files changed, 232 insertions(+), 10 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 19ca0b5..5b944cd 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -244,43 +244,63 @@ static int nfs4_stat_to_errno(int);
 				(encode_getattr_maxsz)
 #define decode_fs_locations_maxsz \
 				(0)
+
+#if defined(CONFIG_NFS_V4_1)
+#define encode_sequence_maxsz	0 /* stub */
+#define decode_sequence_maxsz	0 /* stub */
+#else /* CONFIG_NFS_V4_1 */
+#define encode_sequence_maxsz	0
+#define decode_sequence_maxsz	0
+#endif /* CONFIG_NFS_V4_1 */
+
 #define NFS4_enc_compound_sz	(1024)  /* XXX: large enough? */
 #define NFS4_dec_compound_sz	(1024)  /* XXX: large enough? */
 #define NFS4_enc_read_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_read_maxsz)
 #define NFS4_dec_read_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_read_maxsz)
 #define NFS4_enc_readlink_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_readlink_maxsz)
 #define NFS4_dec_readlink_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_readlink_maxsz)
 #define NFS4_enc_readdir_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_readdir_maxsz)
 #define NFS4_dec_readdir_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_readdir_maxsz)
 #define NFS4_enc_write_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_write_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_write_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_write_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_commit_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_commit_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_commit_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_commit_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_open_sz        (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_savefh_maxsz + \
 				encode_open_maxsz + \
@@ -289,6 +309,7 @@ static int nfs4_stat_to_errno(int);
 				encode_restorefh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_open_sz        (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_savefh_maxsz + \
 				decode_open_maxsz + \
@@ -305,43 +326,53 @@ static int nfs4_stat_to_errno(int);
 				 decode_putfh_maxsz + \
 				 decode_open_confirm_maxsz)
 #define NFS4_enc_open_noattr_sz	(compound_encode_hdr_maxsz + \
+					encode_sequence_maxsz + \
 					encode_putfh_maxsz + \
 					encode_open_maxsz + \
 					encode_getattr_maxsz)
 #define NFS4_dec_open_noattr_sz	(compound_decode_hdr_maxsz + \
+					decode_sequence_maxsz + \
 					decode_putfh_maxsz + \
 					decode_open_maxsz + \
 					decode_getattr_maxsz)
 #define NFS4_enc_open_downgrade_sz \
 				(compound_encode_hdr_maxsz + \
+				 encode_sequence_maxsz + \
 				 encode_putfh_maxsz + \
 				 encode_open_downgrade_maxsz + \
 				 encode_getattr_maxsz)
 #define NFS4_dec_open_downgrade_sz \
 				(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_open_downgrade_maxsz + \
 				 decode_getattr_maxsz)
 #define NFS4_enc_close_sz	(compound_encode_hdr_maxsz + \
+				 encode_sequence_maxsz + \
 				 encode_putfh_maxsz + \
 				 encode_close_maxsz + \
 				 encode_getattr_maxsz)
 #define NFS4_dec_close_sz	(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_close_maxsz + \
 				 decode_getattr_maxsz)
 #define NFS4_enc_setattr_sz	(compound_encode_hdr_maxsz + \
+				 encode_sequence_maxsz + \
 				 encode_putfh_maxsz + \
 				 encode_setattr_maxsz + \
 				 encode_getattr_maxsz)
 #define NFS4_dec_setattr_sz	(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_setattr_maxsz + \
 				 decode_getattr_maxsz)
 #define NFS4_enc_fsinfo_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_fsinfo_maxsz)
 #define NFS4_dec_fsinfo_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_fsinfo_maxsz)
 #define NFS4_enc_renew_sz	(compound_encode_hdr_maxsz + \
@@ -363,64 +394,81 @@ static int nfs4_stat_to_errno(int);
 				decode_putrootfh_maxsz + \
 				decode_fsinfo_maxsz)
 #define NFS4_enc_lock_sz        (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_lock_maxsz)
 #define NFS4_dec_lock_sz        (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_lock_maxsz)
 #define NFS4_enc_lockt_sz       (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_lockt_maxsz)
 #define NFS4_dec_lockt_sz       (compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_lockt_maxsz)
 #define NFS4_enc_locku_sz       (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_locku_maxsz)
 #define NFS4_dec_locku_sz       (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_locku_maxsz)
 #define NFS4_enc_access_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_access_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_access_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_access_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_getattr_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_getattr_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_lookup_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_lookup_maxsz + \
 				encode_getattr_maxsz + \
 				encode_getfh_maxsz)
 #define NFS4_dec_lookup_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_lookup_maxsz + \
 				decode_getattr_maxsz + \
 				decode_getfh_maxsz)
 #define NFS4_enc_lookup_root_sz (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putrootfh_maxsz + \
 				encode_getattr_maxsz + \
 				encode_getfh_maxsz)
 #define NFS4_dec_lookup_root_sz (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putrootfh_maxsz + \
 				decode_getattr_maxsz + \
 				decode_getfh_maxsz)
 #define NFS4_enc_remove_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_remove_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_remove_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_remove_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_rename_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_savefh_maxsz + \
 				encode_putfh_maxsz + \
@@ -429,6 +477,7 @@ static int nfs4_stat_to_errno(int);
 				encode_restorefh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_rename_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_savefh_maxsz + \
 				decode_putfh_maxsz + \
@@ -437,6 +486,7 @@ static int nfs4_stat_to_errno(int);
 				decode_restorefh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_link_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_savefh_maxsz + \
 				encode_putfh_maxsz + \
@@ -445,6 +495,7 @@ static int nfs4_stat_to_errno(int);
 				encode_restorefh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_dec_link_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_savefh_maxsz + \
 				decode_putfh_maxsz + \
@@ -453,16 +504,19 @@ static int nfs4_stat_to_errno(int);
 				decode_restorefh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_symlink_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_symlink_maxsz + \
 				encode_getattr_maxsz + \
 				encode_getfh_maxsz)
 #define NFS4_dec_symlink_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_symlink_maxsz + \
 				decode_getattr_maxsz + \
 				decode_getfh_maxsz)
 #define NFS4_enc_create_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_savefh_maxsz + \
 				encode_create_maxsz + \
@@ -471,6 +525,7 @@ static int nfs4_stat_to_errno(int);
 				encode_restorefh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_create_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_savefh_maxsz + \
 				decode_create_maxsz + \
@@ -479,49 +534,63 @@ static int nfs4_stat_to_errno(int);
 				decode_restorefh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_pathconf_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_pathconf_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_statfs_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_statfs_maxsz)
 #define NFS4_dec_statfs_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_statfs_maxsz)
 #define NFS4_enc_server_caps_sz (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_server_caps_sz (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_delegreturn_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_delegreturn_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_delegreturn_sz (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_delegreturn_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_getacl_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_getacl_maxsz)
 #define NFS4_dec_getacl_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_getacl_maxsz)
 #define NFS4_enc_setacl_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_setacl_maxsz)
 #define NFS4_dec_setacl_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_setacl_maxsz)
 #define NFS4_enc_fs_locations_sz \
 				(compound_encode_hdr_maxsz + \
+				 encode_sequence_maxsz + \
 				 encode_putfh_maxsz + \
 				 encode_lookup_maxsz + \
 				 encode_fs_locations_maxsz)
 #define NFS4_dec_fs_locations_sz \
 				(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_lookup_maxsz + \
 				 decode_fs_locations_maxsz)
@@ -1385,6 +1454,24 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state
 	hdr->nops++;
 	hdr->replen += decode_delegreturn_maxsz;
 }
+
+/* NFSv4.1 operations */
+static void encode_sequence(struct xdr_stream *xdr,
+			    const struct nfs4_sequence_args *args,
+			    struct compound_hdr *hdr)
+{
+#if defined(CONFIG_NFS_V4_1)
+	struct nfs4_session *session = args->sa_session;
+
+	if (!session)
+		return;
+
+	/* stub */
+	hdr->nops++;
+	hdr->replen += decode_sequence_maxsz;
+#endif /* CONFIG_NFS_V4_1 */
+}
+
 /*
  * END OF "GENERIC" ENCODE ROUTINES.
  */
@@ -1410,6 +1497,7 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_access(&xdr, args->access, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1429,6 +1517,7 @@ static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
 	encode_getfh(&xdr, &hdr);
@@ -1449,6 +1538,7 @@ static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, __be32 *p, const struc
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putrootfh(&xdr, &hdr);
 	encode_getfh(&xdr, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1468,6 +1558,7 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_remove(&xdr, &args->name, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1487,6 +1578,7 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->old_dir, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_putfh(&xdr, args->new_dir, &hdr);
@@ -1510,6 +1602,7 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, __be32 *p, const struct nfs4_
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
@@ -1533,6 +1626,7 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_create(&xdr, args, &hdr);
@@ -1564,6 +1658,7 @@ static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, __be32 *p, const struct nf
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
 	encode_nops(&hdr);
@@ -1582,6 +1677,7 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, __be32 *p, struct nfs_closea
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_close(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1601,6 +1697,7 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, __be32 *p, struct nfs_openarg
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_open(&xdr, args, &hdr);
@@ -1642,6 +1739,7 @@ static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, __be32 *p, struct nfs_
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1661,6 +1759,7 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, __be32 *p, struct n
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open_downgrade(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1680,6 +1779,7 @@ static int nfs4_xdr_enc_lock(struct rpc_rqst *req, __be32 *p, struct nfs_lock_ar
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_lock(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1698,6 +1798,7 @@ static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, __be32 *p, struct nfs_lockt_
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_lockt(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1716,6 +1817,7 @@ static int nfs4_xdr_enc_locku(struct rpc_rqst *req, __be32 *p, struct nfs_locku_
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_locku(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1734,6 +1836,7 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readlink(&xdr, args, req, &hdr);
 
@@ -1755,6 +1858,7 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readdir(&xdr, args, req, &hdr);
 
@@ -1779,6 +1883,7 @@ static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readarg
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_read(&xdr, args, &hdr);
 
@@ -1801,6 +1906,7 @@ static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, __be32 *p, struct nfs_seta
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_setattr(&xdr, args, args->server, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1823,6 +1929,7 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	replen = hdr.replen + nfs4_fattr_bitmap_maxsz + 1;
 	encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0, &hdr);
@@ -1845,6 +1952,7 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writea
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_write(&xdr, args, &hdr);
 	req->rq_snd_buf.flags |= XDRBUF_WRITE;
@@ -1865,6 +1973,7 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, __be32 *p, struct nfs_write
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_commit(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1884,6 +1993,7 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs4_fsin
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_fsinfo(&xdr, args->bitmask, &hdr);
 	encode_nops(&hdr);
@@ -1902,6 +2012,7 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, __be32 *p, const struct n
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_one(&xdr, args->bitmask[0] & nfs4_pathconf_bitmap[0],
 			   &hdr);
@@ -1921,6 +2032,7 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_two(&xdr, args->bitmask[0] & nfs4_statfs_bitmap[0],
 			   args->bitmask[1] & nfs4_statfs_bitmap[1], &hdr);
@@ -1941,6 +2053,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p,
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_getattr_one(&xdr, FATTR4_WORD0_SUPPORTED_ATTRS|
 			   FATTR4_WORD0_LINK_SUPPORT|
@@ -2016,6 +2129,7 @@ static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, __be32 *p, const struc
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_delegreturn(&xdr, args->stateid, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -2036,6 +2150,7 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
 	replen = hdr.replen;	/* get the attribute into args->page */
@@ -3762,6 +3877,20 @@ static int decode_delegreturn(struct xdr_stream *xdr)
 	return decode_op_hdr(xdr, OP_DELEGRETURN);
 }
 
+static int decode_sequence(struct xdr_stream *xdr,
+			   struct nfs4_sequence_res *res,
+			   struct rpc_rqst *rqstp)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (!res->sr_session)
+		return 0;
+
+	/* stub */
+#endif /* CONFIG_NFS_V4_1 */
+
+	return 0;
+}
+
 /*
  * END OF "GENERIC" DECODE ROUTINES.
  */
@@ -3779,6 +3908,9 @@ static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, __be32 *p, struct
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -3800,7 +3932,11 @@ static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_ac
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
 	if (status != 0)
@@ -3823,7 +3959,11 @@ static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_lo
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3846,7 +3986,11 @@ static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, __be32 *p, struct nf
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putrootfh(&xdr)) != 0)
 		goto out;
@@ -3866,7 +4010,11 @@ static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, __be32 *p, struct nfs_rem
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3887,7 +4035,11 @@ static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_re
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3917,7 +4069,11 @@ static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_link
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3950,7 +4106,11 @@ static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_cr
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3990,6 +4150,9 @@ static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_g
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4012,6 +4175,7 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	status = encode_setacl(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -4033,6 +4197,9 @@ nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p,
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4056,6 +4223,9 @@ nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p,
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4078,6 +4248,9 @@ static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, __be32 *p, struct nfs_clos
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4108,6 +4281,9 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, __be32 *p, struct nfs_openr
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4162,6 +4338,9 @@ static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, __be32 *p, struct nf
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4186,6 +4365,9 @@ static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_se
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4210,6 +4392,9 @@ static int nfs4_xdr_dec_lock(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock_
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4231,6 +4416,9 @@ static int nfs4_xdr_dec_lockt(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4252,6 +4440,9 @@ static int nfs4_xdr_dec_locku(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4274,6 +4465,9 @@ static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p,
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4295,6 +4489,9 @@ static int nfs4_xdr_dec_readdir(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_r
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4316,6 +4513,9 @@ static int nfs4_xdr_dec_read(struct rpc_rqst *rqstp, __be32 *p, struct nfs_readr
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4339,6 +4539,9 @@ static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, __be32 *p, struct nfs_writ
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4365,6 +4568,9 @@ static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, __be32 *p, struct nfs_wri
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (status)
 		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
+		goto out;
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
@@ -4389,6 +4595,8 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (!status)
+		status = decode_sequence(&xdr, &res->seq_res, req);
+	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
 		status = decode_fsinfo(&xdr, res->fsinfo);
@@ -4408,6 +4616,8 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p,
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (!status)
+		status = decode_sequence(&xdr, &res->seq_res, req);
+	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
 		status = decode_pathconf(&xdr, res->pathconf);
@@ -4427,6 +4637,8 @@ static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p,
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
 	if (!status)
+		status = decode_sequence(&xdr, &res->seq_res, req);
+	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
 		status = decode_statfs(&xdr, res->fsstat);
@@ -4443,7 +4655,11 @@ static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, __be32 *p, struct nfs4
 	int status;
 
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, req);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -4516,7 +4732,10 @@ static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, __be32 *p, struct nf
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
-	if (status != 0)
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
 	if (status != 0)
@@ -4539,7 +4758,10 @@ static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p,
 
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
-	if (status != 0)
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, req);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
-- 
1.6.2.1

[PATCH v2 28/67] nfs41: nfs4_setup_sequence

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Perform the nfs4_setup_sequence in the rpc_call_prepare state.

If a session slot is not available, we will rpc_sleep_on the
slot wait queue leaving the tk_action as rpc_call_prepare.

Once we have a session slot, hang on to it even through rpc_restart_calls.
Ensure the nfs41_sequence_res sr_slot pointer is NULL before rpc_run_task is
called as nfs41_setup_sequence will only find a new slot if it is NULL.

A future patch will call free slot after any rpc_restart_calls, and handle the
rpc restart that result from a sequence operation error.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
[nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: simplify nfs4_call_sync]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_call_sync]
[nfs41: check for session not minorversion]
[nfs41: remove rpc_message from nfs41_call_sync_args]
[moved NFS4_MAX_SLOT_TABLE logic into nfs41_setup_sequence]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs41_call_sync_data use nfs_client not nfs_server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: expose nfs4_call_sync_session for lease renewal]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove unnecessary return check]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |  100 +++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 files changed, 98 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 2e70c7e..4d48bf8 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -273,14 +273,110 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
 
 #if defined(CONFIG_NFS_V4_1)
 
+static int nfs41_setup_sequence(struct nfs4_session *session,
+				struct nfs4_sequence_args *args,
+				struct nfs4_sequence_res *res,
+				int cache_reply,
+				struct rpc_task *task)
+{
+	/* slot already allocated? */
+	if (res->sr_slotid != NFS4_MAX_SLOT_TABLE)
+		return 0;
+
+	memset(res, 0, sizeof(*res));
+	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+	return 0;
+}
+
+int nfs4_setup_sequence(struct nfs_client *clp,
+			struct nfs4_sequence_args *args,
+			struct nfs4_sequence_res *res,
+			int cache_reply,
+			struct rpc_task *task)
+{
+	int ret = 0;
+
+	dprintk("--> %s clp %p session %p sr_slotid %d\n",
+		__func__, clp, clp->cl_session, res->sr_slotid);
+
+	if (!nfs4_has_session(clp))
+		goto out;
+	ret = nfs41_setup_sequence(clp->cl_session, args, res, cache_reply,
+				   task);
+	if (ret != -EAGAIN) {
+		/* terminate rpc task */
+		task->tk_status = ret;
+		task->tk_action = NULL;
+	}
+out:
+	dprintk("<-- %s status=%d\n", __func__, ret);
+	return ret;
+}
+
+struct nfs41_call_sync_data {
+	struct nfs_client *clp;
+	struct nfs4_sequence_args *seq_args;
+	struct nfs4_sequence_res *seq_res;
+	int cache_reply;
+};
+
+static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs41_call_sync_data *data = calldata;
+
+	dprintk("--> %s data->clp->cl_session %p\n", __func__,
+		data->clp->cl_session);
+	if (nfs4_setup_sequence(data->clp, data->seq_args,
+				data->seq_res, data->cache_reply, task))
+		return;
+	rpc_call_start(task);
+}
+
+struct rpc_call_ops nfs41_call_sync_ops = {
+	.rpc_call_prepare = nfs41_call_sync_prepare,
+};
+
+static int nfs4_call_sync_sequence(struct nfs_client *clp,
+				   struct rpc_clnt *clnt,
+				   struct rpc_message *msg,
+				   struct nfs4_sequence_args *args,
+				   struct nfs4_sequence_res *res,
+				   int cache_reply)
+{
+	int ret;
+	struct rpc_task *task;
+	struct nfs41_call_sync_data data = {
+		.clp = clp,
+		.seq_args = args,
+		.seq_res = res,
+		.cache_reply = cache_reply,
+	};
+	struct rpc_task_setup task_setup = {
+		.rpc_client = clnt,
+		.rpc_message = msg,
+		.callback_ops = &nfs41_call_sync_ops,
+		.callback_data = &data
+	};
+
+	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+	task = rpc_run_task(&task_setup);
+	if (IS_ERR(task))
+		ret = PTR_ERR(task);
+	else {
+		ret = task->tk_status;
+		rpc_put_task(task);
+	}
+	return ret;
+}
+
 int _nfs4_call_sync_session(struct nfs_server *server,
 			    struct rpc_message *msg,
 			    struct nfs4_sequence_args *args,
 			    struct nfs4_sequence_res *res,
 			    int cache_reply)
 {
-	/* in preparation for setting up the sequence op */
-	return rpc_call_sync(server->client, msg, 0);
+	return nfs4_call_sync_sequence(server->nfs_client, server->client,
+				       msg, args, res, cache_reply);
 }
 
 #endif /* CONFIG_NFS_V4_1 */
-- 
1.6.2.1

[PATCH v2 29/67] nfs41: find slot

[Benny Halevy]

Find a free slot using bitmap-based allocation.
Use the optimized ffz function to find a zero bit
in the bitmap that indicates a free slot, starting
the search from the 'lowest_free_slotid' position.

If found, mark the slot as used in the bitmap, get
the slot's slotid and seqid, and update max_slotid
to be used by the SEQUENCE operation.

Also, update lowest_free_slotid for next search.

If no free slot was found the caller has to wait
for a free slot (outside the scope of this function)

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: find slot return slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[use find_first_zero_bit for nfs4_find_slot as per review comment 21/85.]
[use NFS4_MAX_SLOT_TABLE rather than NFS4_NO_SLOT]
[nfs41: rpc_sleep_on slot_tbl_waitq must be called under slot_tbl_lock]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   33 +++++++++++++++++++++++++++++++++
 1 files changed, 33 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4d48bf8..582d61e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -273,6 +273,39 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
 
 #if defined(CONFIG_NFS_V4_1)
 
+/*
+ * nfs4_find_slot - efficiently look for a free slot
+ *
+ * nfs4_find_slot looks for an unset bit in the used_slots bitmap.
+ * If found, we mark the slot as used, update the highest_used_slotid,
+ * and respectively set up the sequence operation args.
+ * The slot number is returned if found, or NFS4_MAX_SLOT_TABLE otherwise.
+ */
+static u8
+nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task)
+{
+	int slotid;
+	u8 ret_id = NFS4_MAX_SLOT_TABLE;
+	BUILD_BUG_ON((u8)NFS4_MAX_SLOT_TABLE != (int)NFS4_MAX_SLOT_TABLE);
+
+	spin_lock(&tbl->slot_tbl_lock);
+	dprintk("--> %s used_slots=%04lx highest_used=%d max_slots=%d\n",
+		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
+		tbl->max_slots);
+	slotid = find_first_zero_bit(tbl->used_slots, tbl->max_slots);
+	if (slotid >= tbl->max_slots)
+		goto out;
+	__set_bit(slotid, tbl->used_slots);
+	if (slotid > tbl->highest_used_slotid)
+		tbl->highest_used_slotid = slotid;
+	ret_id = slotid;
+out:
+	dprintk("<-- %s used_slots=%04lx highest_used=%d slotid=%d \n",
+		__func__, tbl->used_slots[0], tbl->highest_used_slotid, ret_id);
+	spin_unlock(&tbl->slot_tbl_lock);
+	return ret_id;
+}
+
 static int nfs41_setup_sequence(struct nfs4_session *session,
 				struct nfs4_sequence_args *args,
 				struct nfs4_sequence_res *res,
-- 
1.6.2.1

[PATCH v2 30/67] nfs41: setup_sequence method

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Allocate a slot in the session slot table and set the sequence op arguments.

Called at the rpc prepare stage.

Add a status to nfs41_sequence_res, initialize it to one so that we catch
rpc level failures which do not go through decode_sequence which sets
the new status field.

Note that upon an rpc level failure, we don't know if the server processed the
sequence operation or not. Proceed as if the server did process the sequence
operation.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
[nfs41: sequence args use slotid]
[nfs41: find slot return slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
As per 11-14-08 review
[move extern declaration from nfs41: sequence setup/done support]
[removed sa_session definition, changed sa_cache_this into a u8 to reduce footprint]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: rpc_sleep_on slot_tbl_waitq must be called under slot_tbl_lock]
    Otherwise there's a race (we've hit) with nfs4_free_slot where
    nfs41_setup_sequence sees a full slot table, unlocks slot_tbl_lock,
    nfs4_free_slots happen concurrently and call rpc_wake_up_next
    where there's nobody to wake up yet, context goes back to
    nfs41_setup_sequence which goes to sleep when the slot table
    is actually empty now and there's no-one to wake it up anymore.
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4_fs.h        |   10 ++++++++++
 fs/nfs/nfs4proc.c       |   34 ++++++++++++++++++++++++++++++++--
 include/linux/nfs_xdr.h |    4 ++++
 3 files changed, 46 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 5d7b036..c2ef040 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -203,8 +203,18 @@ extern int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops;
 extern struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops;
 #if defined(CONFIG_NFS_V4_1)
+extern int nfs4_setup_sequence(struct nfs_client *clp,
+		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
+		int cache_reply, struct rpc_task *task);
 extern void nfs4_destroy_session(struct nfs4_session *session);
 extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
+#else /* CONFIG_NFS_v4_1 */
+static inline int nfs4_setup_sequence(struct nfs_client *clp,
+		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
+		int cache_reply, struct rpc_task *task)
+{
+	return 0;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 extern const u32 nfs4_fattr_bitmap[2];
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 582d61e..b57b50e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -280,6 +280,8 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
  * If found, we mark the slot as used, update the highest_used_slotid,
  * and respectively set up the sequence operation args.
  * The slot number is returned if found, or NFS4_MAX_SLOT_TABLE otherwise.
+ *
+ * Note: must be called with under the slot_tbl_lock.
  */
 static u8
 nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task)
@@ -288,7 +290,6 @@ nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task)
 	u8 ret_id = NFS4_MAX_SLOT_TABLE;
 	BUILD_BUG_ON((u8)NFS4_MAX_SLOT_TABLE != (int)NFS4_MAX_SLOT_TABLE);
 
-	spin_lock(&tbl->slot_tbl_lock);
 	dprintk("--> %s used_slots=%04lx highest_used=%d max_slots=%d\n",
 		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
 		tbl->max_slots);
@@ -302,7 +303,6 @@ nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task)
 out:
 	dprintk("<-- %s used_slots=%04lx highest_used=%d slotid=%d \n",
 		__func__, tbl->used_slots[0], tbl->highest_used_slotid, ret_id);
-	spin_unlock(&tbl->slot_tbl_lock);
 	return ret_id;
 }
 
@@ -312,12 +312,42 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 				int cache_reply,
 				struct rpc_task *task)
 {
+	struct nfs4_slot *slot;
+	struct nfs4_slot_table *tbl;
+	u8 slotid;
+
+	dprintk("--> %s\n", __func__);
 	/* slot already allocated? */
 	if (res->sr_slotid != NFS4_MAX_SLOT_TABLE)
 		return 0;
 
 	memset(res, 0, sizeof(*res));
 	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+	tbl = &session->fc_slot_table;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	slotid = nfs4_find_slot(tbl, task);
+	if (slotid == NFS4_MAX_SLOT_TABLE) {
+		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
+		spin_unlock(&tbl->slot_tbl_lock);
+		dprintk("<-- %s: no free slots\n", __func__);
+		return -EAGAIN;
+	}
+	spin_unlock(&tbl->slot_tbl_lock);
+
+	slot = tbl->slots + slotid;
+	args->sa_slotid = slotid;
+	args->sa_cache_this = cache_reply;
+
+	dprintk("<-- %s slotid=%d seqid=%d\n", __func__, slotid, slot->seq_nr);
+
+	res->sr_slotid = slotid;
+	res->sr_renewal_time = jiffies;
+	/*
+	 * sr_status is only set in decode_sequence, and so will remain
+	 * set to 1 if an rpc level failure occurs.
+	 */
+	res->sr_status = 1;
 	return 0;
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 892fd27..1e00b79 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -163,11 +163,15 @@ struct nfs4_slot {
 
 struct nfs4_sequence_args {
 	struct nfs4_session	*sa_session;
+	u8			sa_slotid;
+	u8			sa_cache_this;
 };
 
 struct nfs4_sequence_res {
 	struct nfs4_session	*sr_session;
 	u8			sr_slotid;	/* slot used to send request */
+	unsigned long		sr_renewal_time;
+	int			sr_status;	/* sequence operation status */
 };
 
 /*
-- 
1.6.2.1

[PATCH v2 31/67] nfs41: free slot

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Free a slot in the slot table.

Mark the slot as free in the bitmap-based allocation table
by clearing a bit corresponding to the slotid.

Update lowest_free_slotid if freed slotid is lower than that.
Update highest_used_slotid.  In the case the freed slotid
equals the highest_used_slotid, scan downwards for the next
highest used slotid using the optimized fls* functions.

Finally, wake up thread waiting on slot_tbl_waitq for a free slot
to become available.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: free slot use slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use find_first_zero_bit for nfs4_find_slot]
    While at it, obliterate lowest_free_slotid and fix-up related comments.
    As per review comment 21/85.
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use __clear_bit for nfs4_free_slot]
    While at it, fix-up function comment.
    Part of review comment 22/85.
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use find_last_bit in nfs4_free_slot to determine highest used slot.]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: rpc_sleep_on slot_tbl_waitq must be called under slot_tbl_lock]
    Otherwise there's a race (we've hit) with nfs4_free_slot where
    nfs41_setup_sequence sees a full slot table, unlocks slot_tbl_lock,
    nfs4_free_slots happen concurrently and call rpc_wake_up_next
    where there's nobody to wake up yet, context goes back to
    nfs41_setup_sequence which goes to sleep when the slot table
    is actually empty now and there's no-one to wake it up anymore.
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   36 ++++++++++++++++++++++++++++++++++++
 1 files changed, 36 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b57b50e..b17c75e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -274,6 +274,42 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
 #if defined(CONFIG_NFS_V4_1)
 
 /*
+ * nfs4_free_slot - free a slot and efficiently update slot table.
+ *
+ * freeing a slot is trivially done by clearing its respective bit
+ * in the bitmap.
+ * If the freed slotid equals highest_used_slotid we want to update it
+ * so that the server would be able to size down the slot table if needed,
+ * otherwise we know that the highest_used_slotid is still in use.
+ * When updating highest_used_slotid there may be "holes" in the bitmap
+ * so we need to scan down from highest_used_slotid to 0 looking for the now
+ * highest slotid in use.
+ * If none found, highest_used_slotid is set to -1.
+ */
+static void
+nfs4_free_slot(struct nfs4_slot_table *tbl, u8 free_slotid)
+{
+	int slotid = free_slotid;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	/* clear used bit in bitmap */
+	__clear_bit(slotid, tbl->used_slots);
+
+	/* update highest_used_slotid when it is freed */
+	if (slotid == tbl->highest_used_slotid) {
+		slotid = find_last_bit(tbl->used_slots, tbl->max_slots);
+		if (slotid >= 0 && slotid < tbl->max_slots)
+			tbl->highest_used_slotid = slotid;
+		else
+			tbl->highest_used_slotid = -1;
+	}
+	rpc_wake_up_next(&tbl->slot_tbl_waitq);
+	spin_unlock(&tbl->slot_tbl_lock);
+	dprintk("%s: free_slotid %u highest_used_slotid %d\n", __func__,
+		free_slotid, tbl->highest_used_slotid);
+}
+
+/*
  * nfs4_find_slot - efficiently look for a free slot
  *
  * nfs4_find_slot looks for an unset bit in the used_slots bitmap.
-- 
1.6.2.1

[PATCH v2 32/67] nfs41: nfs41_sequence_free_slot

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

[from nfs41: separate free slot from sequence done]

Don't free the slot until after all rpc_restart_calls have completed.
Session reset will require more work.

As noted by Trond, since we're using rpc_wake_up_next rather than
rpc_wake_up() we must always wake up the next task in the queue
either by going through nfs4_free_slot, or just calling
rpc_wake_up_next if no slot is to be freed.

[nfs41: sequence res use slotid]
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
[got rid of nfs4_sequence_res.sr_session, use nfs_client.cl_session instead]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: rpc_wake_up_next if sessions slot was not consumed.]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/internal.h |   14 ++++++++++++++
 fs/nfs/nfs4proc.c |   21 +++++++++++++++++++++
 2 files changed, 35 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 8d67c28..12f4c5e 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -230,6 +230,20 @@ static inline int nfs4_has_session(const struct nfs_client *clp)
 	return 0;
 }
 
+#ifdef CONFIG_NFS_V4_1
+extern void nfs41_sequence_free_slot(const struct nfs_client *,
+				     struct nfs4_sequence_res *res);
+#endif /* CONFIG_NFS_V4_1 */
+
+static inline void nfs4_sequence_free_slot(const struct nfs_client *clp,
+					   struct nfs4_sequence_res *res)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(clp))
+		nfs41_sequence_free_slot(clp, res);
+#endif /* CONFIG_NFS_V4_1 */
+}
+
 /*
  * Determine the device name as a string
  */
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b17c75e..5e49696 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -309,6 +309,27 @@ nfs4_free_slot(struct nfs4_slot_table *tbl, u8 free_slotid)
 		free_slotid, tbl->highest_used_slotid);
 }
 
+void nfs41_sequence_free_slot(const struct nfs_client *clp,
+			      struct nfs4_sequence_res *res)
+{
+	struct nfs4_slot_table *tbl;
+
+	if (!nfs4_has_session(clp)) {
+		dprintk("%s: No session\n", __func__);
+		return;
+	}
+	tbl = &clp->cl_session->fc_slot_table;
+	if (res->sr_slotid == NFS4_MAX_SLOT_TABLE) {
+		dprintk("%s: No slot\n", __func__);
+		/* just wake up the next guy waiting since
+		 * we may have not consumed a slot after all */
+		rpc_wake_up_next(&tbl->slot_tbl_waitq);
+		return;
+	}
+	nfs4_free_slot(tbl, res->sr_slotid);
+	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+}
+
 /*
  * nfs4_find_slot - efficiently look for a free slot
  *
-- 
1.6.2.1

[PATCH v2 33/67] nfs41: nfs41_sequence_done

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Handle session level errors, update slot sequence id and
sessions bookeeping, free slot.

[nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: bail out early out of nfs41_sequence_done if !res->sr_session]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move nfs4_sequence_done from nfs41: nfs41_call_sync_done]
Signed-off-by: Andy Adamson <andros@netapp.com>
[move nfs4_sequence_free_slot from nfs41: separate free slot from sequence done]
    Don't free the slot until after all rpc_restart_calls have completed.
    Session reset will require more work.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[moved reset sr_slotid to nfs41_sequence_free_slot]
[free slot also on unexpectecd error]
[remove seq_res.sr_session member, use nfs_client's instead]
[ditch seq_res.sr_flags until used]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[look at sr_slotid for bailing out early from nfs41_sequence_done]
[nfs41: rpc_wake_up_next if sessions slot was not consumed.]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove unused error checking in nfs41_sequence_done]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: remove nfs4_has_session check in nfs41_sequence_done]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: remove nfs_client pointer check]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   58 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 58 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 5e49696..dbc701b 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -330,6 +330,46 @@ void nfs41_sequence_free_slot(const struct nfs_client *clp,
 	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
 }
 
+static void nfs41_sequence_done(struct nfs_client *clp,
+				struct nfs4_sequence_res *res,
+				int rpc_status)
+{
+	unsigned long timestamp;
+	struct nfs4_slot_table *tbl;
+	struct nfs4_slot *slot;
+
+	/*
+	 * sr_status remains 1 if an RPC level error occurred. The server
+	 * may or may not have processed the sequence operation..
+	 * Proceed as if the server received and processed the sequence
+	 * operation.
+	 */
+	if (res->sr_status == 1)
+		res->sr_status = NFS_OK;
+
+	/* -ERESTARTSYS can result in skipping nfs41_sequence_setup */
+	if (res->sr_slotid == NFS4_MAX_SLOT_TABLE)
+		goto out;
+
+	tbl = &clp->cl_session->fc_slot_table;
+	slot = tbl->slots + res->sr_slotid;
+
+	if (res->sr_status == 0) {
+		/* Update the slot's sequence and clientid lease timer */
+		++slot->seq_nr;
+		timestamp = res->sr_renewal_time;
+		spin_lock(&clp->cl_lock);
+		if (time_before(clp->cl_last_renewal, timestamp))
+			clp->cl_last_renewal = timestamp;
+		spin_unlock(&clp->cl_lock);
+		return;
+	}
+out:
+	/* The session may be reset by one of the error handlers. */
+	dprintk("%s: Error %d free the slot \n", __func__, res->sr_status);
+	nfs41_sequence_free_slot(clp, res);
+}
+
 /*
  * nfs4_find_slot - efficiently look for a free slot
  *
@@ -515,6 +555,24 @@ int _nfs4_call_sync(struct nfs_server *server,
 	(server)->nfs_client->cl_call_sync((server), (msg), &(args)->seq_args, \
 			&(res)->seq_res, (cache_reply))
 
+static void nfs4_sequence_done(const struct nfs_server *server,
+			       struct nfs4_sequence_res *res, int rpc_status)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(server->nfs_client))
+		nfs41_sequence_done(server->nfs_client, res, rpc_status);
+#endif /* CONFIG_NFS_V4_1 */
+}
+
+/* no restart, therefore free slot here */
+static void nfs4_sequence_done_free_slot(const struct nfs_server *server,
+					 struct nfs4_sequence_res *res,
+					 int rpc_status)
+{
+	nfs4_sequence_done(server, res, rpc_status);
+	nfs4_sequence_free_slot(server->nfs_client, res);
+}
+
 static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
-- 
1.6.2.1

[PATCH v2 34/67] nfs41: nfs41_call_sync_done

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Implement nfs4.1 synchronous rpc_call_done method
that essentially just calls nfs4_sequence_done, that turns
around and calls nfs41_sequence_done for minorversion1 rpcs.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move adding nfs4_sequence_free_slot from nfs41-separate-free-slot-from-sequence-done]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs41_call_sync_data use nfs_client not nfs_server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |    9 +++++++++
 1 files changed, 9 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index dbc701b..2b481db 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -492,8 +492,17 @@ static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata)
 	rpc_call_start(task);
 }
 
+static void nfs41_call_sync_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs41_call_sync_data *data = calldata;
+
+	nfs41_sequence_done(data->clp, data->seq_res, task->tk_status);
+	nfs41_sequence_free_slot(data->clp, data->seq_res);
+}
+
 struct rpc_call_ops nfs41_call_sync_ops = {
 	.rpc_call_prepare = nfs41_call_sync_prepare,
+	.rpc_call_done = nfs41_call_sync_done,
 };
 
 static int nfs4_call_sync_sequence(struct nfs_client *clp,
-- 
1.6.2.1

[PATCH v2 35/67] nfs41: close sequence setup/done support

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Separate nfs4_close calls from nfs41: sequence setup/done support
Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: separate free slot from sequence done]
[nfs41: sequence res use slotid]
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |    8 +++++++-
 1 files changed, 7 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 2b481db..638dea9 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1630,6 +1630,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 	struct nfs4_state *state = calldata->state;
 	struct nfs_server *server = NFS_SERVER(calldata->inode);
 
+	nfs4_sequence_done(server, &calldata->res.seq_res, task->tk_status);
 	if (RPC_ASSASSINATED(task))
 		return;
         /* hmm. we are done with the inode, and in the process of freeing
@@ -1652,6 +1653,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 				return;
 			}
 	}
+	nfs4_sequence_free_slot(server->nfs_client, &calldata->res.seq_res);
 	nfs_refresh_inode(calldata->inode, calldata->res.fattr);
 }
 
@@ -1692,6 +1694,10 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data)
 		calldata->arg.fmode = FMODE_WRITE;
 	}
 	calldata->timestamp = jiffies;
+	if (nfs4_setup_sequence((NFS_SERVER(calldata->inode))->nfs_client,
+				&calldata->arg.seq_args, &calldata->res.seq_res,
+				1, task))
+		return;
 	rpc_call_start(task);
 }
 
@@ -1731,7 +1737,7 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	};
 	int status = -ENOMEM;
 
-	calldata = kmalloc(sizeof(*calldata), GFP_KERNEL);
+	calldata = kzalloc(sizeof(*calldata), GFP_KERNEL);
 	if (calldata == NULL)
 		goto out;
 	calldata->inode = state->inode;
-- 
1.6.2.1

[PATCH v2 36/67] nfs41: open sequence setup/done support

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Separate nfs4_open calls from nfs41: sequence setup/done support
Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
[use nfs4_sequence_done_free_slot]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 638dea9..8654a96 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1241,6 +1241,10 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata)
 		nfs_copy_fh(&data->o_res.fh, data->o_arg.fh);
 	}
 	data->timestamp = jiffies;
+	if (nfs4_setup_sequence(data->o_arg.server->nfs_client,
+				&data->o_arg.seq_args,
+				&data->o_res.seq_res, 1, task))
+		return;
 	rpc_call_start(task);
 	return;
 out_no_action:
@@ -1253,6 +1257,10 @@ static void nfs4_open_done(struct rpc_task *task, void *calldata)
 	struct nfs4_opendata *data = calldata;
 
 	data->rpc_status = task->tk_status;
+
+	nfs4_sequence_done_free_slot(data->o_arg.server, &data->o_res.seq_res,
+				     task->tk_status);
+
 	if (RPC_ASSASSINATED(task))
 		return;
 	if (task->tk_status == 0) {
-- 
1.6.2.1

[PATCH v2 37/67] nfs41: lock sequence setup/done support

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Separate nfs4_lock calls from nfs41: sequence setup/done support
Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
[use nfs4_sequence_done_free_slot]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 8654a96..9f1da6c 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3691,6 +3691,7 @@ struct nfs4_lockdata {
 	unsigned long timestamp;
 	int rpc_status;
 	int cancelled;
+	struct nfs_server *server;
 };
 
 static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
@@ -3718,6 +3719,7 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
 	p->res.lock_seqid = p->arg.lock_seqid;
 	p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	p->lsp = lsp;
+	p->server = server;
 	atomic_inc(&lsp->ls_count);
 	p->ctx = get_nfs_open_context(ctx);
 	memcpy(&p->fl, fl, sizeof(p->fl));
@@ -3747,6 +3749,9 @@ static void nfs4_lock_prepare(struct rpc_task *task, void *calldata)
 	} else
 		data->arg.new_lock_owner = 0;
 	data->timestamp = jiffies;
+	if (nfs4_setup_sequence(data->server->nfs_client, &data->arg.seq_args,
+				&data->res.seq_res, 1, task))
+		return;
 	rpc_call_start(task);
 	dprintk("%s: done!, ret = %d\n", __func__, data->rpc_status);
 }
@@ -3757,6 +3762,9 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata)
 
 	dprintk("%s: begin!\n", __func__);
 
+	nfs4_sequence_done_free_slot(data->server, &data->res.seq_res,
+				     task->tk_status);
+
 	data->rpc_status = task->tk_status;
 	if (RPC_ASSASSINATED(task))
 		goto out;
-- 
1.6.2.1

[PATCH v2 38/67] nfs41: locku sequence setup/done support

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Separate nfs4_locku calls from nfs41: sequence setup/done support
Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   10 +++++++++-
 1 files changed, 9 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 9f1da6c..6e6bb5b 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3536,7 +3536,7 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl,
 	struct nfs4_unlockdata *p;
 	struct inode *inode = lsp->ls_state->inode;
 
-	p = kmalloc(sizeof(*p), GFP_KERNEL);
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
 	if (p == NULL)
 		return NULL;
 	p->arg.fh = NFS_FH(inode);
@@ -3567,6 +3567,8 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 {
 	struct nfs4_unlockdata *calldata = data;
 
+	nfs4_sequence_done(calldata->server, &calldata->res.seq_res,
+			   task->tk_status);
 	if (RPC_ASSASSINATED(task))
 		return;
 	switch (task->tk_status) {
@@ -3585,6 +3587,8 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 			if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN)
 				rpc_restart_call(task);
 	}
+	nfs4_sequence_free_slot(calldata->server->nfs_client,
+				&calldata->res.seq_res);
 }
 
 static void nfs4_locku_prepare(struct rpc_task *task, void *data)
@@ -3599,6 +3603,10 @@ static void nfs4_locku_prepare(struct rpc_task *task, void *data)
 		return;
 	}
 	calldata->timestamp = jiffies;
+	if (nfs4_setup_sequence(calldata->server->nfs_client,
+				&calldata->arg.seq_args,
+				&calldata->res.seq_res, 1, task))
+		return;
 	rpc_call_start(task);
 }
 
-- 
1.6.2.1

[PATCH v2 39/67] nfs41: unlink sequence setup/done support

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Implement the rpc_call_prepare methods for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: separate free slot from sequence done]
[nfs41: sequence res use slotid]
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |    2 ++
 fs/nfs/unlink.c   |   17 +++++++++++++++++
 2 files changed, 19 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 6e6bb5b..54172c2 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2424,8 +2424,10 @@ static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir)
 {
 	struct nfs_removeres *res = task->tk_msg.rpc_resp;
 
+	nfs4_sequence_done(res->server, &res->seq_res, task->tk_status);
 	if (nfs4_async_handle_error(task, res->server, NULL) == -EAGAIN)
 		return 0;
+	nfs4_sequence_free_slot(res->server->nfs_client, &res->seq_res);
 	update_changeattr(dir, &res->cinfo);
 	nfs_post_op_update_inode(dir, &res->dir_attr);
 	return 1;
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index 83d0ce2..089a21b 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -15,6 +15,7 @@
 #include <linux/wait.h>
 
 #include "internal.h"
+#include "nfs4_fs.h"
 
 struct nfs_unlinkdata {
 	struct hlist_node list;
@@ -102,9 +103,25 @@ static void nfs_async_unlink_release(void *calldata)
 	nfs_sb_deactive(sb);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+void nfs_unlink_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs_unlinkdata *data = calldata;
+	struct nfs_server *server = NFS_SERVER(data->dir);
+
+	if (nfs4_setup_sequence(server->nfs_client, &data->args.seq_args,
+				&data->res.seq_res, 1, task))
+		return;
+	rpc_call_start(task);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static const struct rpc_call_ops nfs_unlink_ops = {
 	.rpc_call_done = nfs_async_unlink_done,
 	.rpc_release = nfs_async_unlink_release,
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_unlink_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 };
 
 static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct nfs_unlinkdata *data)
-- 
1.6.2.1

[PATCH v2 40/67] nfs41: read sequence setup/done support

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Implement the read rpc_call_prepare method for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move the nfs4_sequence_free_slot call in nfs_readpage_retry from]
[nfs41: separate free slot from sequence done]
[remove nfs_readargs.nfs_server, use calldata->inode instead]
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Support sessions with O_DIRECT]
Signed-off-by: Dean Hildebrand <dhildeb@us.ibm.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/direct.c   |    3 +++
 fs/nfs/internal.h |    3 +++
 fs/nfs/nfs4proc.c |    5 +++++
 fs/nfs/read.c     |   30 ++++++++++++++++++++++++++++--
 4 files changed, 39 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 08f6b04..c8c53a0 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -259,6 +259,9 @@ static void nfs_direct_read_release(void *calldata)
 }
 
 static const struct rpc_call_ops nfs_read_direct_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_read_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_direct_read_result,
 	.rpc_release = nfs_direct_read_release,
 };
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 12f4c5e..988c8b9 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -206,6 +206,9 @@ extern int nfs4_path_walk(struct nfs_server *server,
 			  const char *path);
 #endif
 
+/* read.c */
+extern void nfs_read_prepare(struct rpc_task *task, void *calldata);
+
 /* nfs4proc.c */
 extern int _nfs4_call_sync(struct nfs_server *server,
 			   struct rpc_message *msg,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 54172c2..a3e19e9 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2864,6 +2864,11 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
 {
 	struct nfs_server *server = NFS_SERVER(data->inode);
 
+	dprintk("--> %s\n", __func__);
+
+	/* nfs4_sequence_free_slot called in the read rpc_call_done */
+	nfs4_sequence_done(server, &data->res.seq_res, task->tk_status);
+
 	if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) {
 		rpc_restart_call(task);
 		return -EAGAIN;
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 70ba2b4..d71f0d5 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -22,6 +22,7 @@
 
 #include <asm/system.h>
 
+#include "nfs4_fs.h"
 #include "internal.h"
 #include "iostat.h"
 #include "fscache.h"
@@ -358,19 +359,25 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data
 	struct nfs_readres *resp = &data->res;
 
 	if (resp->eof || resp->count == argp->count)
-		return;
+		goto out;
 
 	/* This is a short read! */
 	nfs_inc_stats(data->inode, NFSIOS_SHORTREAD);
 	/* Has the server at least made some progress? */
 	if (resp->count == 0)
-		return;
+		goto out;
 
 	/* Yes, so retry the read at the end of the data */
 	argp->offset += resp->count;
 	argp->pgbase += resp->count;
 	argp->count -= resp->count;
 	rpc_restart_call(task);
+	return;
+out:
+	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
+				&data->res.seq_res);
+	return;
+
 }
 
 /*
@@ -407,7 +414,23 @@ static void nfs_readpage_release_partial(void *calldata)
 	nfs_readdata_release(calldata);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+void nfs_read_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs_read_data *data = calldata;
+
+	if (nfs4_setup_sequence(NFS_SERVER(data->inode)->nfs_client,
+				&data->args.seq_args, &data->res.seq_res,
+				0, task))
+		return;
+	rpc_call_start(task);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static const struct rpc_call_ops nfs_read_partial_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_read_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_readpage_result_partial,
 	.rpc_release = nfs_readpage_release_partial,
 };
@@ -471,6 +494,9 @@ static void nfs_readpage_release_full(void *calldata)
 }
 
 static const struct rpc_call_ops nfs_read_full_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_read_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_readpage_result_full,
 	.rpc_release = nfs_readpage_release_full,
 };
-- 
1.6.2.1

[PATCH v2 41/67] nfs41 write sequence setup done support

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Separate write calls from nfs41: sequence setup/done support

Implement the write rpc_call_prepare method for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move the nfs4_sequence_free_slot call in nfs_readpage_retry from]
[nfs41: separate free slot from sequence done
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Support sessions with O_DIRECT.]
Signed-off-by: Dean Hildebrand <dhildeb@us.ibm.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/direct.c   |    3 +++
 fs/nfs/internal.h |    3 +++
 fs/nfs/nfs4proc.c |    4 ++++
 fs/nfs/write.c    |   22 ++++++++++++++++++++++
 4 files changed, 32 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index c8c53a0..1955bfe 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -676,6 +676,9 @@ out_unlock:
 }
 
 static const struct rpc_call_ops nfs_write_direct_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_direct_write_result,
 	.rpc_release = nfs_direct_write_release,
 };
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 988c8b9..f62bc52 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -209,6 +209,9 @@ extern int nfs4_path_walk(struct nfs_server *server,
 /* read.c */
 extern void nfs_read_prepare(struct rpc_task *task, void *calldata);
 
+/* write.c */
+extern void nfs_write_prepare(struct rpc_task *task, void *calldata);
+
 /* nfs4proc.c */
 extern int _nfs4_call_sync(struct nfs_server *server,
 			   struct rpc_message *msg,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a3e19e9..ba8e419 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2890,6 +2890,10 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	struct inode *inode = data->inode;
 	
+	/* slot is freed in nfs_writeback_done */
+	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
+			   task->tk_status);
+
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) {
 		rpc_restart_call(task);
 		return -EAGAIN;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 035e6fb..195fe76 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -25,6 +25,7 @@
 #include "delegation.h"
 #include "internal.h"
 #include "iostat.h"
+#include "nfs4_fs.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PAGECACHE
 
@@ -1050,7 +1051,23 @@ out:
 	nfs_writedata_release(calldata);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+void nfs_write_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs_write_data *data = calldata;
+	struct nfs_client *clp = (NFS_SERVER(data->inode))->nfs_client;
+
+	if (nfs4_setup_sequence(clp, &data->args.seq_args,
+				&data->res.seq_res, 1, task))
+		return;
+	rpc_call_start(task);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static const struct rpc_call_ops nfs_write_partial_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_writeback_done_partial,
 	.rpc_release = nfs_writeback_release_partial,
 };
@@ -1113,6 +1130,9 @@ remove_request:
 }
 
 static const struct rpc_call_ops nfs_write_full_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_writeback_done_full,
 	.rpc_release = nfs_writeback_release_full,
 };
@@ -1195,6 +1215,8 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 		/* Can't do anything about it except throw an error. */
 		task->tk_status = -EIO;
 	}
+	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
+				&data->res.seq_res);
 	return 0;
 }
 
-- 
1.6.2.1

[PATCH v2 42/67] nfs41 commit sequence setup done support

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Separate commit calls from nfs41: sequence setup/done support

Implement the commit rpc_call_prepare method for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Support sessions with O_DIRECT.]
Signed-off-by: Dean Hildebrand <dhildeb@us.ibm.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: separate free slot from sequence done]
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/direct.c   |    3 +++
 fs/nfs/nfs4proc.c |    4 ++++
 fs/nfs/write.c    |    3 +++
 3 files changed, 10 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 1955bfe..489fc01 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -538,6 +538,9 @@ static void nfs_direct_commit_release(void *calldata)
 }
 
 static const struct rpc_call_ops nfs_commit_direct_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_direct_commit_result,
 	.rpc_release = nfs_direct_commit_release,
 };
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index ba8e419..1132eda 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2920,10 +2920,14 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	struct inode *inode = data->inode;
 	
+	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
+			   task->tk_status);
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) {
 		rpc_restart_call(task);
 		return -EAGAIN;
 	}
+	nfs4_sequence_free_slot(NFS_SERVER(inode)->nfs_client,
+				&data->res.seq_res);
 	nfs_refresh_inode(inode, data->res.fattr);
 	return 0;
 }
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 195fe76..85a7640 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1373,6 +1373,9 @@ static void nfs_commit_release(void *calldata)
 }
 
 static const struct rpc_call_ops nfs_commit_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_commit_done,
 	.rpc_release = nfs_commit_release,
 };
-- 
1.6.2.1

[PATCH v2 43/67] nfs41 delegreturn sequence setup done support

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Separate delegreturn calls from nfs41: sequence setup/done support

Implement the delegreturn rpc_call_prepare method for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   24 +++++++++++++++++++++++-
 1 files changed, 23 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 1132eda..ffdc7e8 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3363,6 +3363,10 @@ struct nfs4_delegreturndata {
 static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_delegreturndata *data = calldata;
+
+	nfs4_sequence_done_free_slot(data->res.server, &data->res.seq_res,
+				     task->tk_status);
+
 	data->rpc_status = task->tk_status;
 	if (data->rpc_status == 0)
 		renew_lease(data->res.server, data->timestamp);
@@ -3373,7 +3377,25 @@ static void nfs4_delegreturn_release(void *calldata)
 	kfree(calldata);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+static void nfs4_delegreturn_prepare(struct rpc_task *task, void *data)
+{
+	struct nfs4_delegreturndata *d_data;
+
+	d_data = (struct nfs4_delegreturndata *)data;
+
+	if (nfs4_setup_sequence(d_data->res.server->nfs_client,
+				&d_data->args.seq_args,
+				&d_data->res.seq_res, 1, task))
+		return;
+	rpc_call_start(task);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static const struct rpc_call_ops nfs4_delegreturn_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs4_delegreturn_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs4_delegreturn_done,
 	.rpc_release = nfs4_delegreturn_release,
 };
@@ -3395,7 +3417,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	};
 	int status = 0;
 
-	data = kmalloc(sizeof(*data), GFP_KERNEL);
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	if (data == NULL)
 		return -ENOMEM;
 	data->args.fhandle = &data->fh;
-- 
1.6.2.1

[PATCH v2 44/67] nfs41: exchange_id operation

[Benny Halevy]

Implement the exchange_id operation conforming to
http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26

Unlike NFSv4.0, NFSv4.1 requires machine credentials. RPC_AUTH_GSS machine
credentials will be passed into the kernel at mount time to be available for
the exchange_id operation.

RPC_AUTH_UNIX root mounts can use the UNIX root credential. Store the root
credential in the nfs_client struct.

Without a credential, NFSv4.1 state renewal fails.

[nfs41: establish clientid via exchange id only if cred != NULL]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfsd41: move nfstime4 from under CONFIG_NFS_V4_1]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: do not wait a lease time in exchange id]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c          |   61 +++++++++++++++++++
 fs/nfs/nfs4xdr.c           |  141 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/nfs4.h       |    1 +
 include/linux/nfs_fs_sb.h  |    6 ++
 include/linux/nfs_xdr.h    |   43 +++++++++++++
 include/linux/nfsd/state.h |    1 -
 6 files changed, 252 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index ffdc7e8..1391ce1 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -48,6 +48,7 @@
 #include <linux/smp_lock.h>
 #include <linux/namei.h>
 #include <linux/mount.h>
+#include <linux/module.h>
 
 #include "nfs4_fs.h"
 #include "delegation.h"
@@ -433,11 +434,13 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 	spin_unlock(&tbl->slot_tbl_lock);
 
 	slot = tbl->slots + slotid;
+	args->sa_session = session;
 	args->sa_slotid = slotid;
 	args->sa_cache_this = cache_reply;
 
 	dprintk("<-- %s slotid=%d seqid=%d\n", __func__, slotid, slot->seq_nr);
 
+	res->sr_session = session;
 	res->sr_slotid = slotid;
 	res->sr_renewal_time = jiffies;
 	/*
@@ -4131,6 +4134,64 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 }
 
 #ifdef CONFIG_NFS_V4_1
+/*
+ * nfs4_proc_exchange_id()
+ *
+ * Since the clientid has expired, all compounds using sessions
+ * associated with the stale clientid will be returning
+ * NFS4ERR_BADSESSION in the sequence operation, and will therefore
+ * be in some phase of session reset.
+ */
+static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
+{
+	nfs4_verifier verifier;
+	struct nfs41_exchange_id_args args = {
+		.client = clp,
+		.flags = clp->cl_exchange_flags,
+	};
+	struct nfs41_exchange_id_res res = {
+		.client = clp,
+	};
+	int status;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_EXCHANGE_ID],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+		.rpc_cred = cred,
+	};
+	__be32 *p;
+
+	dprintk("--> %s\n", __func__);
+	BUG_ON(clp == NULL);
+	p = (u32 *)verifier.data;
+	*p++ = htonl((u32)clp->cl_boot_time.tv_sec);
+	*p = htonl((u32)clp->cl_boot_time.tv_nsec);
+	args.verifier = &verifier;
+
+	while (1) {
+		args.id_len = scnprintf(args.id, sizeof(args.id),
+					"%s/%s %u",
+					clp->cl_ipaddr,
+					rpc_peeraddr2str(clp->cl_rpcclient,
+							 RPC_DISPLAY_ADDR),
+					clp->cl_id_uniquifier);
+
+		status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
+
+		if (status != NFS4ERR_CLID_INUSE)
+			break;
+
+		if (signalled())
+			break;
+
+		if (++clp->cl_id_uniquifier == 0)
+			break;
+	}
+
+	dprintk("<-- %s status= %d\n", __func__, status);
+	return status;
+}
+
 /* Destroy the slot table */
 static void nfs4_destroy_slot_table(struct nfs4_session *session)
 {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 5b944cd..31e9998 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -246,6 +246,27 @@ static int nfs4_stat_to_errno(int);
 				(0)
 
 #if defined(CONFIG_NFS_V4_1)
+#define encode_exchange_id_maxsz (op_encode_hdr_maxsz + \
+				encode_verifier_maxsz + \
+				1 /* co_ownerid.len */ + \
+				XDR_QUADLEN(NFS4_EXCHANGE_ID_LEN) + \
+				1 /* flags */ + \
+				1 /* spa_how */ + \
+				0 /* SP4_NONE (for now) */ + \
+				1 /* zero implemetation id array */)
+#define decode_exchange_id_maxsz (op_decode_hdr_maxsz + \
+				2 /* eir_clientid */ + \
+				1 /* eir_sequenceid */ + \
+				1 /* eir_flags */ + \
+				1 /* spr_how */ + \
+				0 /* SP4_NONE (for now) */ + \
+				2 /* eir_server_owner.so_minor_id */ + \
+				/* eir_server_owner.so_major_id<> */ \
+				XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \
+				/* eir_server_scope<> */ \
+				XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \
+				1 /* eir_server_impl_id array length */ + \
+				0 /* ignored eir_server_impl_id contents */)
 #define encode_sequence_maxsz	0 /* stub */
 #define decode_sequence_maxsz	0 /* stub */
 #else /* CONFIG_NFS_V4_1 */
@@ -594,6 +615,14 @@ static int nfs4_stat_to_errno(int);
 				 decode_putfh_maxsz + \
 				 decode_lookup_maxsz + \
 				 decode_fs_locations_maxsz)
+#if defined(CONFIG_NFS_V4_1)
+#define NFS4_enc_exchange_id_sz \
+				(compound_encode_hdr_maxsz + \
+				 encode_exchange_id_maxsz)
+#define NFS4_dec_exchange_id_sz \
+				(compound_decode_hdr_maxsz + \
+				 decode_exchange_id_maxsz)
+#endif /* CONFIG_NFS_V4_1 */
 
 static const umode_t nfs_type2fmt[] = {
 	[NF4BAD] = 0,
@@ -1455,7 +1484,29 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state
 	hdr->replen += decode_delegreturn_maxsz;
 }
 
+#if defined(CONFIG_NFS_V4_1)
 /* NFSv4.1 operations */
+static void encode_exchange_id(struct xdr_stream *xdr,
+			       struct nfs41_exchange_id_args *args,
+			       struct compound_hdr *hdr)
+{
+	uint32_t *p;
+
+	RESERVE_SPACE(4 + sizeof(args->verifier->data));
+	WRITE32(OP_EXCHANGE_ID);
+	WRITEMEM(args->verifier->data, sizeof(args->verifier->data));
+
+	encode_string(xdr, args->id_len, args->id);
+
+	RESERVE_SPACE(12);
+	WRITE32(args->flags);
+	WRITE32(0);	/* zero length state_protect4_a */
+	WRITE32(0);	/* zero length implementation id array */
+	hdr->nops++;
+	hdr->replen += decode_exchange_id_maxsz;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static void encode_sequence(struct xdr_stream *xdr,
 			    const struct nfs4_sequence_args *args,
 			    struct compound_hdr *hdr)
@@ -2162,6 +2213,26 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 	return 0;
 }
 
+#if defined(CONFIG_NFS_V4_1)
+/*
+ * EXCHANGE_ID request
+ */
+static int nfs4_xdr_enc_exchange_id(struct rpc_rqst *req, uint32_t *p,
+				    struct nfs41_exchange_id_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = args->client->cl_minorversion,
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_exchange_id(&xdr, args, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 /*
  * START OF "GENERIC" DECODE ROUTINES.
  *   These may look a little ugly since they are imported from a "generic"
@@ -3877,6 +3948,54 @@ static int decode_delegreturn(struct xdr_stream *xdr)
 	return decode_op_hdr(xdr, OP_DELEGRETURN);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+static int decode_exchange_id(struct xdr_stream *xdr,
+			      struct nfs41_exchange_id_res *res)
+{
+	uint32_t *p;
+	int status, dummy;
+	struct nfs_client *clp = res->client;
+
+	status = decode_op_hdr(xdr, OP_EXCHANGE_ID);
+	if (status)
+		return status;
+
+	READ_BUF(8);
+	READ64(clp->cl_ex_clid);
+	READ_BUF(12);
+	READ32(clp->cl_seqid);
+	READ32(clp->cl_exchange_flags);
+
+	/* We ask for SP4_NONE */
+	READ32(dummy);
+	if (dummy != SP4_NONE)
+		return -EIO;
+
+	/* minor_id */
+	READ_BUF(8);
+	READ64(res->server_owner.minor_id);
+
+	/* Major id */
+	READ_BUF(4);
+	READ32(res->server_owner.major_id_sz);
+	READ_BUF(res->server_owner.major_id_sz);
+	COPYMEM(res->server_owner.major_id, res->server_owner.major_id_sz);
+
+	/* server_scope */
+	READ_BUF(4);
+	READ32(res->server_scope.server_scope_sz);
+	READ_BUF(res->server_scope.server_scope_sz);
+	COPYMEM(res->server_scope.server_scope,
+		res->server_scope.server_scope_sz);
+	/* Throw away Implementation id array */
+	READ_BUF(4);
+	READ32(dummy);
+	p += XDR_QUADLEN(dummy);
+
+	return 0;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static int decode_sequence(struct xdr_stream *xdr,
 			   struct nfs4_sequence_res *res,
 			   struct rpc_rqst *rqstp)
@@ -4774,6 +4893,25 @@ out:
 	return status;
 }
 
+#if defined(CONFIG_NFS_V4_1)
+/*
+ * EXCHANGE_ID request
+ */
+static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
+				    void *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_exchange_id(&xdr, res);
+	return status;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus)
 {
 	uint32_t bitmap[2] = {0};
@@ -4943,6 +5081,9 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(GETACL,		enc_getacl,	dec_getacl),
   PROC(SETACL,		enc_setacl,	dec_setacl),
   PROC(FS_LOCATIONS,	enc_fs_locations, dec_fs_locations),
+#if defined(CONFIG_NFS_V4_1)
+  PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
+#endif /* CONFIG_NFS_V4_1 */
 };
 
 struct rpc_version		nfs_version4 = {
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index c0c1e8e..7dc6fa0 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -21,6 +21,7 @@
 #define NFS4_FHSIZE		128
 #define NFS4_MAXPATHLEN		PATH_MAX
 #define NFS4_MAXNAMLEN		NAME_MAX
+#define NFS4_OPAQUE_LIMIT	1024
 #define NFS4_MAX_SESSIONID_LEN	16
 
 #define NFS4_ACCESS_READ        0x0001
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index d541a49..876e5ff 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -78,6 +78,12 @@ struct nfs_client {
 #endif /* CONFIG_NFS_V4 */
 
 #ifdef CONFIG_NFS_V4_1
+	/* clientid returned from EXCHANGE_ID, used by session operations */
+	u64			cl_ex_clid;
+	/* The sequence id to use for the next CREATE_SESSION */
+	u32			cl_seqid;
+	/* The flags used for obtaining the clientid during EXCHANGE_ID */
+	u32			cl_exchange_flags;
 	struct nfs4_session	*cl_session; 	/* sharred session */
 #endif /* CONFIG_NFS_V4_1 */
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 1e00b79..1f22fcd 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -879,6 +879,49 @@ struct nfs4_fs_locations_res {
 
 #endif /* CONFIG_NFS_V4 */
 
+struct nfstime4 {
+	u64	seconds;
+	u32	nseconds;
+};
+
+#ifdef CONFIG_NFS_V4_1
+struct nfs_impl_id4 {
+	u32		domain_len;
+	char		*domain;
+	u32		name_len;
+	char		*name;
+	struct nfstime4	date;
+};
+
+#define NFS4_EXCHANGE_ID_LEN	(48)
+struct nfs41_exchange_id_args {
+	struct nfs_client		*client;
+	nfs4_verifier			*verifier;
+	unsigned int 			id_len;
+	char 				id[NFS4_EXCHANGE_ID_LEN];
+	u32				flags;
+};
+
+struct server_owner {
+	uint64_t			minor_id;
+	uint32_t			major_id_sz;
+	char				major_id[NFS4_OPAQUE_LIMIT];
+};
+
+struct server_scope {
+	uint32_t			server_scope_sz;
+	char 				server_scope[NFS4_OPAQUE_LIMIT];
+};
+
+struct nfs41_exchange_id_res {
+	struct nfs_client		*client;
+	u32				flags;
+	struct server_owner		server_owner;
+	struct server_scope		server_scope;
+	struct nfs_impl_id4		impl_id;
+};
+#endif /* CONFIG_NFS_V4_1 */
+
 struct nfs_page;
 
 #define NFS_PAGEVEC_SIZE	(8U)
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 128298c..ec64274 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -41,7 +41,6 @@
 #include <linux/kref.h>
 #include <linux/sunrpc/clnt.h>
 
-#define NFS4_OPAQUE_LIMIT 1024
 typedef struct {
 	u32             cl_boot;
 	u32             cl_id;
-- 
1.6.2.1

[PATCH v2 45/67] nfs41: get_lease_time

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

get_lease_time uses the FSINFO rpc operation to
get the lease time attribute.

nfs4_get_lease_time() is only called from the state manager on session setup
so don't recover from clientid or sequence level errors.

We do need to recover from NFS4ERR_DELAY or NFS4ERR_GRACE.
Use NFS4_POLL_RETRY_MIN - the Linux server returns NFS4ERR_DELAY when an
upcall is needed to resolve an uncached export referenced by a file handle.

[nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove extraneous rpc_clnt pointer]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: have get_lease_time work on nfs_client]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: get_lease_time recover from NFS4ERR_DELAY]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
[define nfs4_get_lease_time_{args,res}]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |   94 +++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/nfs4xdr.c        |   51 +++++++++++++++++++++++++
 include/linux/nfs_xdr.h |    9 ++++
 3 files changed, 154 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 1391ce1..926a9d6 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4192,6 +4192,100 @@ static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 	return status;
 }
 
+struct nfs4_get_lease_time_data {
+	struct nfs4_get_lease_time_args *args;
+	struct nfs4_get_lease_time_res *res;
+	struct nfs_client *clp;
+};
+
+static void nfs4_get_lease_time_prepare(struct rpc_task *task,
+					void *calldata)
+{
+	int ret;
+	struct nfs4_get_lease_time_data *data =
+			(struct nfs4_get_lease_time_data *)calldata;
+
+	dprintk("--> %s\n", __func__);
+	/* just setup sequence, do not trigger session recovery
+	   since we're invoked within one */
+	ret = nfs41_setup_sequence(data->clp->cl_session,
+					&data->args->la_seq_args,
+					&data->res->lr_seq_res, 0, task);
+
+	BUG_ON(ret == -EAGAIN);
+	rpc_call_start(task);
+	dprintk("<-- %s\n", __func__);
+}
+
+/*
+ * Called from nfs4_state_manager thread for session setup, so don't recover
+ * from sequence operation or clientid errors.
+ */
+static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs4_get_lease_time_data *data =
+			(struct nfs4_get_lease_time_data *)calldata;
+
+	dprintk("--> %s\n", __func__);
+	nfs41_sequence_done(data->clp, &data->res->lr_seq_res, task->tk_status);
+	switch (task->tk_status) {
+	case -NFS4ERR_DELAY:
+	case -NFS4ERR_GRACE:
+		dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status);
+		rpc_delay(task, NFS4_POLL_RETRY_MIN);
+		task->tk_status = 0;
+		rpc_restart_call(task);
+		return;
+	}
+	nfs41_sequence_free_slot(data->clp, &data->res->lr_seq_res);
+	dprintk("<-- %s\n", __func__);
+}
+
+struct rpc_call_ops nfs4_get_lease_time_ops = {
+	.rpc_call_prepare = nfs4_get_lease_time_prepare,
+	.rpc_call_done = nfs4_get_lease_time_done,
+};
+
+int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
+{
+	struct rpc_task *task;
+	struct nfs4_get_lease_time_args args;
+	struct nfs4_get_lease_time_res res = {
+		.lr_fsinfo = fsinfo,
+	};
+	struct nfs4_get_lease_time_data data = {
+		.args = &args,
+		.res = &res,
+		.clp = clp,
+	};
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GET_LEASE_TIME],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+	struct rpc_task_setup task_setup = {
+		.rpc_client = clp->cl_rpcclient,
+		.rpc_message = &msg,
+		.callback_ops = &nfs4_get_lease_time_ops,
+		.callback_data = &data
+	};
+	int status;
+
+	res.lr_seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
+	dprintk("--> %s\n", __func__);
+	task = rpc_run_task(&task_setup);
+
+	if (IS_ERR(task))
+		status = PTR_ERR(task);
+	else {
+		status = task->tk_status;
+		rpc_put_task(task);
+	}
+	dprintk("<-- %s return %d\n", __func__, status);
+
+	return status;
+}
+
 /* Destroy the slot table */
 static void nfs4_destroy_slot_table(struct nfs4_session *session)
 {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 31e9998..3f88a20 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -622,6 +622,14 @@ static int nfs4_stat_to_errno(int);
 #define NFS4_dec_exchange_id_sz \
 				(compound_decode_hdr_maxsz + \
 				 decode_exchange_id_maxsz)
+#define NFS4_enc_get_lease_time_sz	(compound_encode_hdr_maxsz + \
+					 encode_sequence_maxsz + \
+					 encode_putrootfh_maxsz + \
+					 encode_fsinfo_maxsz)
+#define NFS4_dec_get_lease_time_sz	(compound_decode_hdr_maxsz + \
+					 decode_sequence_maxsz + \
+					 decode_putrootfh_maxsz + \
+					 decode_fsinfo_maxsz)
 #endif /* CONFIG_NFS_V4_1 */
 
 static const umode_t nfs_type2fmt[] = {
@@ -2231,6 +2239,27 @@ static int nfs4_xdr_enc_exchange_id(struct rpc_rqst *req, uint32_t *p,
 	encode_nops(&hdr);
 	return 0;
 }
+
+/*
+ * a GET_LEASE_TIME request
+ */
+static int nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, uint32_t *p,
+				       struct nfs4_get_lease_time_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->la_seq_args),
+	};
+	const u32 lease_bitmap[2] = { FATTR4_WORD0_LEASE_TIME, 0 };
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->la_seq_args, &hdr);
+	encode_putrootfh(&xdr, &hdr);
+	encode_fsinfo(&xdr, lease_bitmap, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 /*
@@ -4910,6 +4939,27 @@ static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
 		status = decode_exchange_id(&xdr, res);
 	return status;
 }
+
+/*
+ * a GET_LEASE_TIME request
+ */
+static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p,
+				       struct nfs4_get_lease_time_res *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_sequence(&xdr, &res->lr_seq_res, rqstp);
+	if (!status)
+		status = decode_putrootfh(&xdr);
+	if (!status)
+		status = decode_fsinfo(&xdr, res->lr_fsinfo);
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus)
@@ -5083,6 +5133,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(FS_LOCATIONS,	enc_fs_locations, dec_fs_locations),
 #if defined(CONFIG_NFS_V4_1)
   PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
+  PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
 #endif /* CONFIG_NFS_V4_1 */
 };
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 1f22fcd..6c75c63 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -174,6 +174,15 @@ struct nfs4_sequence_res {
 	int			sr_status;	/* sequence operation status */
 };
 
+struct nfs4_get_lease_time_args {
+	struct nfs4_sequence_args	la_seq_args;
+};
+
+struct nfs4_get_lease_time_res {
+	struct nfs_fsinfo	       *lr_fsinfo;
+	struct nfs4_sequence_res	lr_seq_res;
+};
+
 /*
  * Arguments to the open call.
  */
-- 
1.6.2.1

[PATCH v2 46/67] nfs41: create_session operation

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Implement the create_session operation conforming to
http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26

Set the real fore channel max operations to preserve server resources.
Note: If the server returns < NFS4_MAX_OPS, the client will very soon
get an NFS4ERR_TOO_MANY_OPS. A later patch will handle this.

Set the max_rqst_sz and max_resp_sz to PAGE_SIZE - we preallocate the buffers.

Set the back channel max_resp_sz_cached to zero to force the client to
always set csa_cachethis to FALSE because the current implementation
of the back channel DRC only supports caching the CB_SEQUENCE operation.

The client back channel server supports one slot, and desires 2 operations
per compound.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove extraneous rpc_clnt pointer]
Use the struct nfs_client cl_rpcclient.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_init_channel_attrs, just use nfs41_create_session_args]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use rsize and wsize for session channel attributes]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: set channel max operations]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: set back channel attributes]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: obliterate nfs4_adjust_channel_attrs]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: have create_session work on nfs_client]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: move CONFIG_NFS_V4_1 endif]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
[moved nfs4_init_slot_table definition here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use kcalloc to allocate slot table]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       |  172 ++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/nfs4xdr.c        |  189 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/nfs4.h    |   10 +++
 include/linux/nfs_xdr.h |   12 +++
 4 files changed, 383 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 926a9d6..4381939 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -54,6 +54,7 @@
 #include "delegation.h"
 #include "internal.h"
 #include "iostat.h"
+#include "callback.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PROC
 
@@ -4286,6 +4287,50 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
 	return status;
 }
 
+/*
+ * Initialize slot table
+ */
+static int nfs4_init_slot_table(struct nfs4_session *session)
+{
+	struct nfs4_slot_table *tbl = &session->fc_slot_table;
+	int i, max_slots = session->fc_attrs.max_reqs;
+	struct nfs4_slot *slot;
+	int ret = -ENOMEM;
+
+	BUG_ON(max_slots > NFS4_MAX_SLOT_TABLE);
+
+	dprintk("--> %s: max_reqs=%u\n", __func__,
+		session->fc_attrs.max_reqs);
+
+	slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_KERNEL);
+	if (!slot)
+		goto out;
+	for (i = 0; i < max_slots; ++i)
+		slot[i].seq_nr = 1;
+	ret = 0;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	if (tbl->slots != NULL) {
+		spin_unlock(&tbl->slot_tbl_lock);
+		dprintk("%s: slot table already initialized. tbl=%p slots=%p\n",
+			__func__, tbl, tbl->slots);
+		WARN_ON(1);
+		goto out_free;
+	}
+	tbl->max_slots = max_slots;
+	tbl->slots = slot;
+	tbl->highest_used_slotid = -1;  /* no slot is currently used */
+	spin_unlock(&tbl->slot_tbl_lock);
+	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
+		tbl, tbl->slots, tbl->max_slots);
+out:
+	dprintk("<-- %s: return %d\n", __func__, ret);
+	return ret;
+out_free:
+	kfree(slot);
+	goto out;
+}
+
 /* Destroy the slot table */
 static void nfs4_destroy_slot_table(struct nfs4_session *session)
 {
@@ -4317,6 +4362,133 @@ void nfs4_destroy_session(struct nfs4_session *session)
 	kfree(session);
 }
 
+/*
+ * Initialize the values to be used by the client in CREATE_SESSION
+ * If nfs4_init_session set the fore channel request and response sizes,
+ * use them.
+ *
+ * Set the back channel max_resp_sz_cached to zero to force the client to
+ * always set csa_cachethis to FALSE because the current implementation
+ * of the back channel DRC only supports caching the CB_SEQUENCE operation.
+ */
+static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args)
+{
+	struct nfs4_session *session = args->client->cl_session;
+	unsigned int mxrqst_sz = session->fc_attrs.max_rqst_sz,
+		     mxresp_sz = session->fc_attrs.max_resp_sz;
+
+	if (mxrqst_sz == 0)
+		mxrqst_sz = NFS_MAX_FILE_IO_SIZE;
+	if (mxresp_sz == 0)
+		mxresp_sz = NFS_MAX_FILE_IO_SIZE;
+	/* Fore channel attributes */
+	args->fc_attrs.headerpadsz = 0;
+	args->fc_attrs.max_rqst_sz = mxrqst_sz;
+	args->fc_attrs.max_resp_sz = mxresp_sz;
+	args->fc_attrs.max_resp_sz_cached = mxresp_sz;
+	args->fc_attrs.max_ops = NFS4_MAX_OPS;
+	args->fc_attrs.max_reqs = session->clp->cl_rpcclient->cl_xprt->max_reqs;
+
+	dprintk("%s: Fore Channel : max_rqst_sz=%u max_resp_sz=%u "
+		"max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n",
+		__func__,
+		args->fc_attrs.max_rqst_sz, args->fc_attrs.max_resp_sz,
+		args->fc_attrs.max_resp_sz_cached, args->fc_attrs.max_ops,
+		args->fc_attrs.max_reqs);
+
+	/* Back channel attributes */
+	args->bc_attrs.headerpadsz = 0;
+	args->bc_attrs.max_rqst_sz = PAGE_SIZE;
+	args->bc_attrs.max_resp_sz = PAGE_SIZE;
+	args->bc_attrs.max_resp_sz_cached = 0;
+	args->bc_attrs.max_ops = NFS4_MAX_BACK_CHANNEL_OPS;
+	args->bc_attrs.max_reqs = 1;
+
+	dprintk("%s: Back Channel : max_rqst_sz=%u max_resp_sz=%u "
+		"max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n",
+		__func__,
+		args->bc_attrs.max_rqst_sz, args->bc_attrs.max_resp_sz,
+		args->bc_attrs.max_resp_sz_cached, args->bc_attrs.max_ops,
+		args->bc_attrs.max_reqs);
+}
+
+static int _nfs4_proc_create_session(struct nfs_client *clp)
+{
+	struct nfs4_session *session = clp->cl_session;
+	struct nfs41_create_session_args args = {
+		.client = clp,
+		.cb_program = NFS4_CALLBACK,
+	};
+	struct nfs41_create_session_res res = {
+		.client = clp,
+	};
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE_SESSION],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+	int status;
+
+	nfs4_init_channel_attrs(&args);
+	args.flags = (SESSION4_PERSIST);
+
+	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
+
+	/* Set the negotiated values in the session's channel_attrs struct */
+
+	if (!status) {
+		/* Increment the clientid slot sequence id */
+		clp->cl_seqid++;
+	}
+
+	return status;
+}
+
+/*
+ * Issues a CREATE_SESSION operation to the server.
+ * It is the responsibility of the caller to verify the session is
+ * expired before calling this routine.
+ */
+int nfs4_proc_create_session(struct nfs_client *clp)
+{
+	int status;
+	unsigned *ptr;
+	struct nfs_fsinfo fsinfo;
+	struct nfs4_session *session = clp->cl_session;
+
+	dprintk("--> %s clp=%p session=%p\n", __func__, clp, session);
+
+	status = _nfs4_proc_create_session(clp);
+	if (status)
+		goto out;
+
+	/* Init the fore channel */
+	status = nfs4_init_slot_table(session);
+	dprintk("fore channel slot table initialization returned %d\n", status);
+	if (status)
+		goto out;
+
+	ptr = (unsigned *)&session->sess_id.data[0];
+	dprintk("%s client>seqid %d sessionid %u:%u:%u:%u\n", __func__,
+		clp->cl_seqid, ptr[0], ptr[1], ptr[2], ptr[3]);
+
+	/* Get the lease time */
+	status = nfs4_proc_get_lease_time(clp, &fsinfo);
+	if (status == 0) {
+		/* Update lease time and schedule renewal */
+		spin_lock(&clp->cl_lock);
+		clp->cl_lease_time = fsinfo.lease_time * HZ;
+		clp->cl_last_renewal = jiffies;
+		clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+		spin_unlock(&clp->cl_lock);
+
+		nfs4_schedule_state_renewal(clp);
+	}
+out:
+	dprintk("<-- %s\n", __func__);
+	return status;
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 3f88a20..11c210f 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -246,6 +246,8 @@ static int nfs4_stat_to_errno(int);
 				(0)
 
 #if defined(CONFIG_NFS_V4_1)
+#define NFS4_MAX_MACHINE_NAME_LEN (64)
+
 #define encode_exchange_id_maxsz (op_encode_hdr_maxsz + \
 				encode_verifier_maxsz + \
 				1 /* co_ownerid.len */ + \
@@ -267,6 +269,31 @@ static int nfs4_stat_to_errno(int);
 				XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \
 				1 /* eir_server_impl_id array length */ + \
 				0 /* ignored eir_server_impl_id contents */)
+#define encode_channel_attrs_maxsz  (6 + 1 /* ca_rdma_ird.len (0) */)
+#define decode_channel_attrs_maxsz  (6 + \
+				     1 /* ca_rdma_ird.len */ + \
+				     1 /* ca_rdma_ird */)
+#define encode_create_session_maxsz  (op_encode_hdr_maxsz + \
+				     2 /* csa_clientid */ + \
+				     1 /* csa_sequence */ + \
+				     1 /* csa_flags */ + \
+				     encode_channel_attrs_maxsz + \
+				     encode_channel_attrs_maxsz + \
+				     1 /* csa_cb_program */ + \
+				     1 /* csa_sec_parms.len (1) */ + \
+				     1 /* cb_secflavor (AUTH_SYS) */ + \
+				     1 /* stamp */ + \
+				     1 /* machinename.len */ + \
+				     XDR_QUADLEN(NFS4_MAX_MACHINE_NAME_LEN) + \
+				     1 /* uid */ + \
+				     1 /* gid */ + \
+				     1 /* gids.len (0) */)
+#define decode_create_session_maxsz  (op_decode_hdr_maxsz +	\
+				     XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + \
+				     1 /* csr_sequence */ + \
+				     1 /* csr_flags */ + \
+				     decode_channel_attrs_maxsz + \
+				     decode_channel_attrs_maxsz)
 #define encode_sequence_maxsz	0 /* stub */
 #define decode_sequence_maxsz	0 /* stub */
 #else /* CONFIG_NFS_V4_1 */
@@ -622,6 +649,12 @@ static int nfs4_stat_to_errno(int);
 #define NFS4_dec_exchange_id_sz \
 				(compound_decode_hdr_maxsz + \
 				 decode_exchange_id_maxsz)
+#define NFS4_enc_create_session_sz \
+				(compound_encode_hdr_maxsz + \
+				 encode_create_session_maxsz)
+#define NFS4_dec_create_session_sz \
+				(compound_decode_hdr_maxsz + \
+				 decode_create_session_maxsz)
 #define NFS4_enc_get_lease_time_sz	(compound_encode_hdr_maxsz + \
 					 encode_sequence_maxsz + \
 					 encode_putrootfh_maxsz + \
@@ -712,6 +745,7 @@ static void encode_compound_hdr(struct xdr_stream *xdr,
 
 static void encode_nops(struct compound_hdr *hdr)
 {
+	BUG_ON(hdr->nops > NFS4_MAX_OPS);
 	*hdr->nops_p = htonl(hdr->nops);
 }
 
@@ -1513,6 +1547,68 @@ static void encode_exchange_id(struct xdr_stream *xdr,
 	hdr->nops++;
 	hdr->replen += decode_exchange_id_maxsz;
 }
+
+static void encode_create_session(struct xdr_stream *xdr,
+				  struct nfs41_create_session_args *args,
+				  struct compound_hdr *hdr)
+{
+	uint32_t *p;
+	char machine_name[NFS4_MAX_MACHINE_NAME_LEN];
+	uint32_t len;
+	struct nfs_client *clp = args->client;
+
+	RESERVE_SPACE(4);
+	WRITE32(OP_CREATE_SESSION);
+
+	RESERVE_SPACE(8);
+	WRITE64(clp->cl_ex_clid);
+
+	RESERVE_SPACE(8);
+	WRITE32(clp->cl_seqid);			/*Sequence id */
+	WRITE32(args->flags);			/*flags */
+
+	RESERVE_SPACE(2*28);			/* 2 channel_attrs */
+	/* Fore Channel */
+	WRITE32(args->fc_attrs.headerpadsz);	/* header padding size */
+	WRITE32(args->fc_attrs.max_rqst_sz);	/* max req size */
+	WRITE32(args->fc_attrs.max_resp_sz);	/* max resp size */
+	WRITE32(args->fc_attrs.max_resp_sz_cached);	/* Max resp sz cached */
+	WRITE32(args->fc_attrs.max_ops);	/* max operations */
+	WRITE32(args->fc_attrs.max_reqs);	/* max requests */
+	WRITE32(0);				/* rdmachannel_attrs */
+
+	/* Back Channel */
+	WRITE32(args->fc_attrs.headerpadsz);	/* header padding size */
+	WRITE32(args->bc_attrs.max_rqst_sz);	/* max req size */
+	WRITE32(args->bc_attrs.max_resp_sz);	/* max resp size */
+	WRITE32(args->bc_attrs.max_resp_sz_cached);	/* Max resp sz cached */
+	WRITE32(args->bc_attrs.max_ops);	/* max operations */
+	WRITE32(args->bc_attrs.max_reqs);	/* max requests */
+	WRITE32(0);				/* rdmachannel_attrs */
+
+	RESERVE_SPACE(4);
+	WRITE32(args->cb_program);		/* cb_program */
+
+	RESERVE_SPACE(4);			/* # of security flavors */
+	WRITE32(1);
+
+	RESERVE_SPACE(4);
+	WRITE32(RPC_AUTH_UNIX);			/* auth_sys */
+
+	/* authsys_parms rfc1831 */
+	RESERVE_SPACE(4);
+	WRITE32((u32)clp->cl_boot_time.tv_nsec);	/* stamp */
+	len = scnprintf(machine_name, sizeof(machine_name), "%s",
+			clp->cl_ipaddr);
+	RESERVE_SPACE(16 + len);
+	WRITE32(len);
+	WRITEMEM(machine_name, len);
+	WRITE32(0);				/* UID */
+	WRITE32(0);				/* GID */
+	WRITE32(0);				/* No more gids */
+	hdr->nops++;
+	hdr->replen += decode_create_session_maxsz;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static void encode_sequence(struct xdr_stream *xdr,
@@ -2241,6 +2337,24 @@ static int nfs4_xdr_enc_exchange_id(struct rpc_rqst *req, uint32_t *p,
 }
 
 /*
+ * a CREATE_SESSION request
+ */
+static int nfs4_xdr_enc_create_session(struct rpc_rqst *req, uint32_t *p,
+				       struct nfs41_create_session_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = args->client->cl_minorversion,
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_create_session(&xdr, args, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+
+/*
  * a GET_LEASE_TIME request
  */
 static int nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, uint32_t *p,
@@ -4023,6 +4137,63 @@ static int decode_exchange_id(struct xdr_stream *xdr,
 
 	return 0;
 }
+
+static int decode_create_session(struct xdr_stream *xdr,
+				 struct nfs41_create_session_res *res)
+{
+	uint32_t *p;
+	int status;
+	u32 nr_attrs;
+	struct nfs_client *clp = res->client;
+	struct nfs4_session *session = clp->cl_session;
+
+	status = decode_op_hdr(xdr, OP_CREATE_SESSION);
+
+	if (status)
+		return status;
+
+	/* sessionid */
+	READ_BUF(NFS4_MAX_SESSIONID_LEN);
+	COPYMEM(&session->sess_id, NFS4_MAX_SESSIONID_LEN);
+
+	/* seqid, flags */
+	READ_BUF(8);
+	READ32(clp->cl_seqid);
+	READ32(session->flags);
+
+	/* Channel attributes */
+	/* fore channel */
+	READ_BUF(24);
+	READ32(session->fc_attrs.headerpadsz);
+	READ32(session->fc_attrs.max_rqst_sz);
+	READ32(session->fc_attrs.max_resp_sz);
+	READ32(session->fc_attrs.max_resp_sz_cached);
+	READ32(session->fc_attrs.max_ops);
+	READ32(session->fc_attrs.max_reqs);
+	READ_BUF(4);
+	READ32(nr_attrs);
+	if (nr_attrs == 1) {
+		READ_BUF(4);
+		p++; /* skip rdma_attrs */
+	}
+
+	/* back channel */
+	READ_BUF(24);
+	READ32(session->bc_attrs.headerpadsz);
+	READ32(session->bc_attrs.max_rqst_sz);
+	READ32(session->bc_attrs.max_resp_sz);
+	READ32(session->bc_attrs.max_resp_sz_cached);
+	READ32(session->bc_attrs.max_ops);
+	READ32(session->bc_attrs.max_reqs);
+	READ_BUF(4);
+	READ32(nr_attrs);
+	if (nr_attrs == 1) {
+		READ_BUF(4);
+		p++; /* skip rdma_attrs */
+	}
+
+	return 0;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static int decode_sequence(struct xdr_stream *xdr,
@@ -4941,6 +5112,23 @@ static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
+ * a CREATE_SESSION request
+ */
+static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, uint32_t *p,
+				       struct nfs41_create_session_res *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_create_session(&xdr, res);
+	return status;
+}
+
+/*
  * a GET_LEASE_TIME request
  */
 static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p,
@@ -5133,6 +5321,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(FS_LOCATIONS,	enc_fs_locations, dec_fs_locations),
 #if defined(CONFIG_NFS_V4_1)
   PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
+  PROC(CREATE_SESSION,	enc_create_session,	dec_create_session),
   PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
 #endif /* CONFIG_NFS_V4_1 */
 };
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index 7dc6fa0..d2d6a1b 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -131,6 +131,16 @@
 
 #define NFS4_MAX_UINT64	(~(u64)0)
 
+/* An NFS4 sessions server must support at least NFS4_MAX_OPS operations.
+ * If a compound requires more operations, adjust NFS4_MAX_OPS accordingly.
+ */
+#define NFS4_MAX_OPS   8
+
+/* Our NFS4 client back channel server only wants the cb_sequene and the
+ * actual operation per compound
+ */
+#define NFS4_MAX_BACK_CHANNEL_OPS 2
+
 enum nfs4_acl_whotype {
 	NFS4_ACL_WHO_NAMED = 0,
 	NFS4_ACL_WHO_OWNER,
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 6c75c63..d8bc445 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -929,6 +929,18 @@ struct nfs41_exchange_id_res {
 	struct server_scope		server_scope;
 	struct nfs_impl_id4		impl_id;
 };
+
+struct nfs41_create_session_args {
+	struct nfs_client	       *client;
+	uint32_t			flags;
+	uint32_t			cb_program;
+	struct nfs4_channel_attrs	fc_attrs;	/* Fore Channel */
+	struct nfs4_channel_attrs	bc_attrs;	/* Back Channel */
+};
+
+struct nfs41_create_session_res {
+	struct nfs_client	       *client;
+};
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs_page;
-- 
1.6.2.1

[PATCH v2 47/67] nfs41: verify session channel attribues

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Invalidate the session if the server returns invalid fore or back channel
attributes.

Use a KERN_WARNING to report the fatal session estabishment error.

Signed-off-by: Andy Adamson <andros@netapp.com>
[refactor nfs4_verify_channel_attrs]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   50 ++++++++++++++++++++++++++++++++++++++++++++++++--
 1 files changed, 48 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4381939..5ce8999 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4412,6 +4412,51 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args)
 		args->bc_attrs.max_reqs);
 }
 
+static int _verify_channel_attr(char *chan, char *attr_name, u32 sent, u32 rcvd)
+{
+	if (rcvd <= sent)
+		return 0;
+	printk(KERN_WARNING "%s: Session INVALID: %s channel %s increased. "
+		"sent=%u rcvd=%u\n", __func__, chan, attr_name, sent, rcvd);
+	return -EINVAL;
+}
+
+#define _verify_fore_channel_attr(_name_) \
+	_verify_channel_attr("fore", #_name_, \
+			     args->fc_attrs._name_, \
+			     session->fc_attrs._name_)
+
+#define _verify_back_channel_attr(_name_) \
+	_verify_channel_attr("back", #_name_, \
+			     args->bc_attrs._name_, \
+			     session->bc_attrs._name_)
+
+/*
+ * The server is not allowed to increase the fore channel header pad size,
+ * maximum response size, or maximum number of operations.
+ *
+ * The back channel attributes are only negotiatied down: We send what the
+ * (back channel) server insists upon.
+ */
+static int nfs4_verify_channel_attrs(struct nfs41_create_session_args *args,
+				     struct nfs4_session *session)
+{
+	int ret = 0;
+
+	ret |= _verify_fore_channel_attr(headerpadsz);
+	ret |= _verify_fore_channel_attr(max_resp_sz);
+	ret |= _verify_fore_channel_attr(max_ops);
+
+	ret |= _verify_back_channel_attr(headerpadsz);
+	ret |= _verify_back_channel_attr(max_rqst_sz);
+	ret |= _verify_back_channel_attr(max_resp_sz);
+	ret |= _verify_back_channel_attr(max_resp_sz_cached);
+	ret |= _verify_back_channel_attr(max_ops);
+	ret |= _verify_back_channel_attr(max_reqs);
+
+	return ret;
+}
+
 static int _nfs4_proc_create_session(struct nfs_client *clp)
 {
 	struct nfs4_session *session = clp->cl_session;
@@ -4434,8 +4479,9 @@ static int _nfs4_proc_create_session(struct nfs_client *clp)
 
 	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
 
-	/* Set the negotiated values in the session's channel_attrs struct */
-
+	if (!status)
+		/* Verify the session's negotiated channel_attrs values */
+		status = nfs4_verify_channel_attrs(&args, session);
 	if (!status) {
 		/* Increment the clientid slot sequence id */
 		clp->cl_seqid++;
-- 
1.6.2.1

[PATCH v2 48/67] nfs41: use session attributes for rsize and wsize

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Set the mount points rsize and wsize to the negotiated session fore channel
maximum response and requeset size. These values will be bound checked in
nfs_server_set_fsinfo.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/client.c |   17 +++++++++++++++++
 1 files changed, 17 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 562f658..fda2c53 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -527,6 +527,21 @@ static void nfs_mark_client_ready(struct nfs_client *clp, int state)
 }
 
 /*
+ * Session has been established, and the client marked ready.
+ * Set the mount rsize and wsize with negotiated fore channel
+ * attributes which will be bound checked in nfs_server_set_fsinfo.
+ */
+static void nfs4_session_set_rwsize(struct nfs_server *server)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (!nfs4_has_session(server->nfs_client))
+		return;
+	server->rsize = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
+	server->wsize = server->nfs_client->cl_session->fc_attrs.max_rqst_sz;
+#endif /* CONFIG_NFS_V4_1 */
+}
+
+/*
  * Initialise the timeout values for a connection
  */
 static void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
@@ -1296,6 +1311,8 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 		(unsigned long long) server->fsid.minor);
 	dprintk("Mount FH: %d\n", mntfh->size);
 
+	nfs4_session_set_rwsize(server);
+
 	error = nfs_probe_fsinfo(server, mntfh, &fattr);
 	if (error < 0)
 		goto error;
-- 
1.6.2.1

[PATCH v2 49/67] nfs41: destroy_session operation

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Implement the destroy_session operation conforming to
http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove extraneous rpc_clnt pointer]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41; NFS_CS_READY required for DESTROY_SESSION]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   30 +++++++++++++++++++++++++++
 fs/nfs/nfs4xdr.c  |   59 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/super.c    |    2 +
 3 files changed, 91 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 5ce8999..27b00c0 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4535,6 +4535,36 @@ out:
 	return status;
 }
 
+/*
+ * Issue the over-the-wire RPC DESTROY_SESSION.
+ * The caller must serialize access to this routine.
+ */
+int nfs4_proc_destroy_session(struct nfs4_session *session)
+{
+	int status = 0;
+	struct rpc_message msg;
+
+	dprintk("--> nfs4_proc_destroy_session\n");
+
+	/* session is still being setup */
+	if (session->clp->cl_cons_state != NFS_CS_READY)
+		return status;
+
+	msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DESTROY_SESSION];
+	msg.rpc_argp = session;
+	msg.rpc_resp = NULL;
+	msg.rpc_cred = NULL;
+	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
+
+	if (status)
+		printk(KERN_WARNING
+			"Got error %d from the server on DESTROY_SESSION. "
+			"Session has been destroyed regardless...\n", status);
+
+	dprintk("<-- nfs4_proc_destroy_session\n");
+	return status;
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 11c210f..1ad51b5 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -294,6 +294,8 @@ static int nfs4_stat_to_errno(int);
 				     1 /* csr_flags */ + \
 				     decode_channel_attrs_maxsz + \
 				     decode_channel_attrs_maxsz)
+#define encode_destroy_session_maxsz    (op_encode_hdr_maxsz + 4)
+#define decode_destroy_session_maxsz    (op_decode_hdr_maxsz)
 #define encode_sequence_maxsz	0 /* stub */
 #define decode_sequence_maxsz	0 /* stub */
 #else /* CONFIG_NFS_V4_1 */
@@ -655,6 +657,10 @@ static int nfs4_stat_to_errno(int);
 #define NFS4_dec_create_session_sz \
 				(compound_decode_hdr_maxsz + \
 				 decode_create_session_maxsz)
+#define NFS4_enc_destroy_session_sz	(compound_encode_hdr_maxsz + \
+					 encode_destroy_session_maxsz)
+#define NFS4_dec_destroy_session_sz	(compound_decode_hdr_maxsz + \
+					 decode_destroy_session_maxsz)
 #define NFS4_enc_get_lease_time_sz	(compound_encode_hdr_maxsz + \
 					 encode_sequence_maxsz + \
 					 encode_putrootfh_maxsz + \
@@ -1609,6 +1615,18 @@ static void encode_create_session(struct xdr_stream *xdr,
 	hdr->nops++;
 	hdr->replen += decode_create_session_maxsz;
 }
+
+static void encode_destroy_session(struct xdr_stream *xdr,
+				   struct nfs4_session *session,
+				   struct compound_hdr *hdr)
+{
+	uint32_t *p;
+	RESERVE_SPACE(4 + NFS4_MAX_SESSIONID_LEN);
+	WRITE32(OP_DESTROY_SESSION);
+	WRITEMEM(session->sess_id.data, NFS4_MAX_SESSIONID_LEN);
+	hdr->nops++;
+	hdr->replen += decode_destroy_session_maxsz;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static void encode_sequence(struct xdr_stream *xdr,
@@ -2355,6 +2373,24 @@ static int nfs4_xdr_enc_create_session(struct rpc_rqst *req, uint32_t *p,
 }
 
 /*
+ * a DESTROY_SESSION request
+ */
+static int nfs4_xdr_enc_destroy_session(struct rpc_rqst *req, uint32_t *p,
+					struct nfs4_session *session)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = session->clp->cl_minorversion,
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_destroy_session(&xdr, session, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+
+/*
  * a GET_LEASE_TIME request
  */
 static int nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, uint32_t *p,
@@ -4194,6 +4230,11 @@ static int decode_create_session(struct xdr_stream *xdr,
 
 	return 0;
 }
+
+static int decode_destroy_session(struct xdr_stream *xdr, void *dummy)
+{
+	return decode_op_hdr(xdr, OP_DESTROY_SESSION);
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static int decode_sequence(struct xdr_stream *xdr,
@@ -5129,6 +5170,23 @@ static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
+ * a DESTROY_SESSION request
+ */
+static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, uint32_t *p,
+					void *dummy)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_destroy_session(&xdr, dummy);
+	return status;
+}
+
+/*
  * a GET_LEASE_TIME request
  */
 static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p,
@@ -5322,6 +5380,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
 #if defined(CONFIG_NFS_V4_1)
   PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
   PROC(CREATE_SESSION,	enc_create_session,	dec_create_session),
+  PROC(DESTROY_SESSION,	enc_destroy_session,	dec_destroy_session),
   PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
 #endif /* CONFIG_NFS_V4_1 */
 };
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 4d22b18..6ae7088 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -2604,12 +2604,14 @@ static void nfs4_kill_super(struct super_block *sb)
 {
 	struct nfs_server *server = NFS_SB(sb);
 
+	dprintk("--> %s\n", __func__);
 	nfs_super_return_all_delegations(sb);
 	kill_anon_super(sb);
 
 	nfs4_renewd_prepare_shutdown(server);
 	nfs_fscache_release_super_cookie(sb);
 	nfs_free_server(server);
+	dprintk("<-- %s\n", __func__);
 }
 
 /*
-- 
1.6.2.1

[PATCH v2 50/67] nfs41: enable nfs_client only nfs4_async_handle_error

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

The session is per struct nfs_client, not per nfs_server. Allow the handler
to be called with no nfs_server which simplifies the nfs4_proc_async_sequence session renewal call and will let it be used by pnfs file layout data servers.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   13 +++++++++----
 1 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 27b00c0..c03d775 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3224,10 +3224,8 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen
 }
 
 static int
-nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, struct nfs4_state *state)
+_nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, struct nfs_client *clp, struct nfs4_state *state)
 {
-	struct nfs_client *clp = server->nfs_client;
-
 	if (!clp || task->tk_status >= 0)
 		return 0;
 	switch(task->tk_status) {
@@ -3247,7 +3245,8 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 			task->tk_status = 0;
 			return -EAGAIN;
 		case -NFS4ERR_DELAY:
-			nfs_inc_server_stats(server, NFSIOS_DELAY);
+			if (server)
+				nfs_inc_server_stats(server, NFSIOS_DELAY);
 		case -NFS4ERR_GRACE:
 			rpc_delay(task, NFS4_POLL_RETRY_MAX);
 			task->tk_status = 0;
@@ -3260,6 +3259,12 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 	return 0;
 }
 
+static int
+nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, struct nfs4_state *state)
+{
+	return _nfs4_async_handle_error(task, server, server->nfs_client, state);
+}
+
 int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short port, struct rpc_cred *cred)
 {
 	nfs4_verifier sc_verifier;
-- 
1.6.2.1

[PATCH v2 51/67] nfs41: sequence operation

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Implement the sequence operation conforming to
http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26

Check returned sessionid, slotid and slot sequenceid in decode_sequence.

If the server returns different values for sessionID, slotID or slot sequence
number than what was sent, the server is looney tunes.

Pass the sequence operation status to nfs41_sequence_done in order to
determine when to increment the slot sequence ID.

Free slot is separated from sequence done.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Andy Adamson<andros@umich.edu>
[nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: deref slot table in decode_sequence only for minorversion!=0]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_call_sync]
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
[nfs41: return ESERVERFAULT in decode_sequence]
[no sr_session, no sr_flags]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use nfs4_call_sync_sequence to renew session lease]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove nfs4_call_sync_sequence forward definition]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: use struct nfs_client for nfs41_proc_async_sequence]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41 nfs41_sequence_call_done update error checking]
[nfs41 nfs41_sequence_done update error checking]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove switch on error from nfs41_sequence_call_done]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   92 ++++++++++++++++++++++++++++++++++++++
 fs/nfs/nfs4xdr.c  |  127 ++++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 214 insertions(+), 5 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index c03d775..e1b942e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4570,6 +4570,98 @@ int nfs4_proc_destroy_session(struct nfs4_session *session)
 	return status;
 }
 
+/*
+ * Renew the cl_session lease.
+ */
+static int nfs4_proc_sequence(struct nfs_client *clp, struct rpc_cred *cred)
+{
+	struct nfs4_sequence_args args;
+	struct nfs4_sequence_res res;
+
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SEQUENCE],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+		.rpc_cred = cred,
+	};
+
+	args.sa_cache_this = 0;
+
+	return nfs4_call_sync_sequence(clp, clp->cl_rpcclient, &msg, &args,
+				       &res, 0);
+}
+
+void nfs41_sequence_call_done(struct rpc_task *task, void *data)
+{
+	struct nfs_client *clp = (struct nfs_client *)data;
+
+	nfs41_sequence_done(clp, task->tk_msg.rpc_resp, task->tk_status);
+
+	if (task->tk_status < 0) {
+		dprintk("%s ERROR %d\n", __func__, task->tk_status);
+
+		if (_nfs4_async_handle_error(task, NULL, clp, NULL)
+								== -EAGAIN) {
+			rpc_restart_call(task);
+			return;
+		}
+	}
+	nfs41_sequence_free_slot(clp, task->tk_msg.rpc_resp);
+	dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred);
+
+	put_rpccred(task->tk_msg.rpc_cred);
+	kfree(task->tk_msg.rpc_argp);
+	kfree(task->tk_msg.rpc_resp);
+
+	dprintk("<-- %s\n", __func__);
+}
+
+static void nfs41_sequence_prepare(struct rpc_task *task, void *data)
+{
+	struct nfs_client *clp;
+	struct nfs4_sequence_args *args;
+	struct nfs4_sequence_res *res;
+
+	clp = (struct nfs_client *)data;
+	args = task->tk_msg.rpc_argp;
+	res = task->tk_msg.rpc_resp;
+
+	if (nfs4_setup_sequence(clp, args, res, 0, task))
+		return;
+	rpc_call_start(task);
+}
+
+static const struct rpc_call_ops nfs41_sequence_ops = {
+	.rpc_call_done = nfs41_sequence_call_done,
+	.rpc_call_prepare = nfs41_sequence_prepare,
+};
+
+static int nfs41_proc_async_sequence(struct nfs_client *clp,
+				     struct rpc_cred *cred)
+{
+	struct nfs4_sequence_args *args;
+	struct nfs4_sequence_res *res;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SEQUENCE],
+		.rpc_cred = cred,
+	};
+
+	args = kzalloc(sizeof(*args), GFP_KERNEL);
+	if (!args)
+		return -ENOMEM;
+	res = kzalloc(sizeof(*res), GFP_KERNEL);
+	if (!res) {
+		kfree(args);
+		return -ENOMEM;
+	}
+	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+	msg.rpc_argp = args;
+	msg.rpc_resp = res;
+
+	return rpc_call_async(clp->cl_rpcclient, &msg, RPC_TASK_SOFT,
+			      &nfs41_sequence_ops, (void *)clp);
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 1ad51b5..80af0ae 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -296,8 +296,10 @@ static int nfs4_stat_to_errno(int);
 				     decode_channel_attrs_maxsz)
 #define encode_destroy_session_maxsz    (op_encode_hdr_maxsz + 4)
 #define decode_destroy_session_maxsz    (op_decode_hdr_maxsz)
-#define encode_sequence_maxsz	0 /* stub */
-#define decode_sequence_maxsz	0 /* stub */
+#define encode_sequence_maxsz	(op_encode_hdr_maxsz + \
+				XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 4)
+#define decode_sequence_maxsz	(op_decode_hdr_maxsz + \
+				XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 5)
 #else /* CONFIG_NFS_V4_1 */
 #define encode_sequence_maxsz	0
 #define decode_sequence_maxsz	0
@@ -661,6 +663,12 @@ static int nfs4_stat_to_errno(int);
 					 encode_destroy_session_maxsz)
 #define NFS4_dec_destroy_session_sz	(compound_decode_hdr_maxsz + \
 					 decode_destroy_session_maxsz)
+#define NFS4_enc_sequence_sz \
+				(compound_decode_hdr_maxsz + \
+				 encode_sequence_maxsz)
+#define NFS4_dec_sequence_sz \
+				(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz)
 #define NFS4_enc_get_lease_time_sz	(compound_encode_hdr_maxsz + \
 					 encode_sequence_maxsz + \
 					 encode_putrootfh_maxsz + \
@@ -1635,11 +1643,39 @@ static void encode_sequence(struct xdr_stream *xdr,
 {
 #if defined(CONFIG_NFS_V4_1)
 	struct nfs4_session *session = args->sa_session;
+	struct nfs4_slot_table *tp;
+	struct nfs4_slot *slot;
+	__be32 *p;
 
 	if (!session)
 		return;
 
-	/* stub */
+	tp = &session->fc_slot_table;
+
+	WARN_ON(args->sa_slotid == NFS4_MAX_SLOT_TABLE);
+	slot = tp->slots + args->sa_slotid;
+
+	RESERVE_SPACE(4);
+	WRITE32(OP_SEQUENCE);
+
+	/*
+	 * Sessionid + seqid + slotid + max slotid + cache_this
+	 */
+	dprintk("%s: sessionid=%u:%u:%u:%u seqid=%d slotid=%d "
+		"max_slotid=%d cache_this=%d\n",
+		__func__,
+		((u32 *)session->sess_id.data)[0],
+		((u32 *)session->sess_id.data)[1],
+		((u32 *)session->sess_id.data)[2],
+		((u32 *)session->sess_id.data)[3],
+		slot->seq_nr, args->sa_slotid,
+		tp->highest_used_slotid, args->sa_cache_this);
+	RESERVE_SPACE(NFS4_MAX_SESSIONID_LEN + 16);
+	WRITEMEM(session->sess_id.data, NFS4_MAX_SESSIONID_LEN);
+	WRITE32(slot->seq_nr);
+	WRITE32(args->sa_slotid);
+	WRITE32(tp->highest_used_slotid);
+	WRITE32(args->sa_cache_this);
 	hdr->nops++;
 	hdr->replen += decode_sequence_maxsz;
 #endif /* CONFIG_NFS_V4_1 */
@@ -2391,6 +2427,24 @@ static int nfs4_xdr_enc_destroy_session(struct rpc_rqst *req, uint32_t *p,
 }
 
 /*
+ * a SEQUENCE request
+ */
+static int nfs4_xdr_enc_sequence(struct rpc_rqst *req, uint32_t *p,
+				 struct nfs4_sequence_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(args),
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, args, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+
+/*
  * a GET_LEASE_TIME request
  */
 static int nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, uint32_t *p,
@@ -4242,13 +4296,58 @@ static int decode_sequence(struct xdr_stream *xdr,
 			   struct rpc_rqst *rqstp)
 {
 #if defined(CONFIG_NFS_V4_1)
+	struct nfs4_slot *slot;
+	struct nfs4_sessionid id;
+	u32 dummy;
+	int status;
+	__be32 *p;
+
 	if (!res->sr_session)
 		return 0;
 
-	/* stub */
-#endif /* CONFIG_NFS_V4_1 */
+	status = decode_op_hdr(xdr, OP_SEQUENCE);
+	if (status)
+		goto out_err;
 
+	/*
+	 * If the server returns different values for sessionID, slotID or
+	 * sequence number, the server is looney tunes.
+	 */
+	status = -ESERVERFAULT;
+
+	slot = &res->sr_session->fc_slot_table.slots[res->sr_slotid];
+	READ_BUF(NFS4_MAX_SESSIONID_LEN + 20);
+	COPYMEM(id.data, NFS4_MAX_SESSIONID_LEN);
+	if (memcmp(id.data, res->sr_session->sess_id.data,
+		   NFS4_MAX_SESSIONID_LEN)) {
+		dprintk("%s Invalid session id\n", __func__);
+		goto out_err;
+	}
+	/* seqid */
+	READ32(dummy);
+	if (dummy != slot->seq_nr) {
+		dprintk("%s Invalid sequence number\n", __func__);
+		goto out_err;
+	}
+	/* slot id */
+	READ32(dummy);
+	if (dummy != res->sr_slotid) {
+		dprintk("%s Invalid slot id\n", __func__);
+		goto out_err;
+	}
+	/* highest slot id - currently not processed */
+	READ32(dummy);
+	/* target highest slot id - currently not processed */
+	READ32(dummy);
+	/* result flags - currently not processed */
+	READ32(dummy);
+	status = 0;
+out_err:
+	res->sr_status = status;
+	return status;
+#else  /* CONFIG_NFS_V4_1 */
 	return 0;
+#endif /* CONFIG_NFS_V4_1 */
 }
 
 /*
@@ -5187,6 +5286,23 @@ static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
+ * a SEQUENCE request
+ */
+static int nfs4_xdr_dec_sequence(struct rpc_rqst *rqstp, uint32_t *p,
+				 struct nfs4_sequence_res *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_sequence(&xdr, res, rqstp);
+	return status;
+}
+
+/*
  * a GET_LEASE_TIME request
  */
 static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p,
@@ -5381,6 +5497,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
   PROC(CREATE_SESSION,	enc_create_session,	dec_create_session),
   PROC(DESTROY_SESSION,	enc_destroy_session,	dec_destroy_session),
+  PROC(SEQUENCE,	enc_sequence,	dec_sequence),
   PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
 #endif /* CONFIG_NFS_V4_1 */
 };
-- 
1.6.2.1

[PATCH v2 52/67] nfs41: reset the session slot table

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Separated from nfs41: schedule async session reset

Do not kfree the session slot table upon session reset, just re-initialize it.
Add a boolean to nfs4_proc_create_session to inidicate if this is a
session reset or a session initialization.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   44 +++++++++++++++++++++++++++++++++++++++++---
 1 files changed, 41 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index e1b942e..ba68d43 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4292,6 +4292,37 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
 	return status;
 }
 
+/* Reset a slot table */
+static int nfs4_reset_slot_table(struct nfs4_session *session)
+{
+	struct nfs4_slot_table *tbl = &session->fc_slot_table;
+	int i, max_slots = session->fc_attrs.max_reqs;
+	int old_max_slots = session->fc_slot_table.max_slots;
+	int ret = 0;
+
+	dprintk("--> %s: max_reqs=%u, tbl %p\n", __func__,
+		session->fc_attrs.max_reqs, tbl);
+
+	/* Until we have dynamic slot table adjustment, insist
+	 * upon the same slot table size */
+	if (max_slots != old_max_slots) {
+		dprintk("%s reset slot table does't match old\n",
+			__func__);
+		ret = -EINVAL; /*XXX NFS4ERR_REQ_TOO_BIG ? */
+		goto out;
+	}
+	spin_lock(&tbl->slot_tbl_lock);
+	for (i = 0; i < max_slots; ++i)
+		tbl->slots[i].seq_nr = 1;
+	tbl->highest_used_slotid = -1;
+	spin_unlock(&tbl->slot_tbl_lock);
+	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
+		tbl, tbl->slots, tbl->max_slots);
+out:
+	dprintk("<-- %s: return %d\n", __func__, ret);
+	return ret;
+}
+
 /*
  * Initialize slot table
  */
@@ -4500,7 +4531,7 @@ static int _nfs4_proc_create_session(struct nfs_client *clp)
  * It is the responsibility of the caller to verify the session is
  * expired before calling this routine.
  */
-int nfs4_proc_create_session(struct nfs_client *clp)
+int nfs4_proc_create_session(struct nfs_client *clp, int reset)
 {
 	int status;
 	unsigned *ptr;
@@ -4513,8 +4544,11 @@ int nfs4_proc_create_session(struct nfs_client *clp)
 	if (status)
 		goto out;
 
-	/* Init the fore channel */
-	status = nfs4_init_slot_table(session);
+	/* Init or reset the fore channel */
+	if (reset)
+		status = nfs4_reset_slot_table(session);
+	else
+		status = nfs4_init_slot_table(session);
 	dprintk("fore channel slot table initialization returned %d\n", status);
 	if (status)
 		goto out;
@@ -4523,6 +4557,10 @@ int nfs4_proc_create_session(struct nfs_client *clp)
 	dprintk("%s client>seqid %d sessionid %u:%u:%u:%u\n", __func__,
 		clp->cl_seqid, ptr[0], ptr[1], ptr[2], ptr[3]);
 
+	if (reset)
+		/* Lease time is aleady set */
+		goto out;
+
 	/* Get the lease time */
 	status = nfs4_proc_get_lease_time(clp, &fsinfo);
 	if (status == 0) {
-- 
1.6.2.1

[PATCH v2 53/67] nfs41: add session setup to the state manager

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

At mount, nfs_alloc_client sets the cl_state NFS4CLNT_LEASE_EXPIRED bit
and nfs4_alloc_session sets the NFS4CLNT_SESSION_SETUP bit, so both bits are
set when nfs4_lookup_root calls nfs4_recover_expired_lease which schedules
the nfs4_state_manager and waits for it to complete.

Place the session setup after the clientid establishment in nfs4_state_manager
so that the session is setup right after the clientid has been established
without rescheduling the state manager.

Unlike nfsv4.0, the nfs_client struct is not ready to use until the session
has been established.  Postpone marking the nfs_client struct to NFS_CS_READY
until after a successful CREATE_SESSION call so that other threads cannot use
the client until the session is established.

If the EXCHANGE_ID call fails and the session has not been setup (the
NFS4CLNT_SESSION_SETUP bit is set), mark the client with the error and return.

If the session setup CREATE_SESSION call fails with NFS4ERR_STALE_CLIENTID
which could occur due to server reboot or network partition inbetween the
EXCHANGE_ID and CREATE_SESSION call, reset the NFS4CLNT_LEASE_EXPIRED and
NFS4CLNT_SESSION_SETUP bits and try again.

If the CREATE_SESSION call fails with other errors, mark the client with
the error and return.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>

[nfs41: NFS_CS_SESSION_SETUP cl_cons_state for back channel setup]
  On session setup, the CREATE_SESSION reply races with the server back channel
  probe which needs to succeed to setup the back channel. Set a new
  cl_cons_state NFS_CS_SESSION_SETUP just prior to the CREATE_SESSION call
  and add it as a valid state to nfs_find_client so that the client back channel
  can find the nfs_client struct and won't drop the server backchannel probe.
  Use a new cl_cons_state so that NFSv4.0 back channel behaviour which only
  sets NFS_CS_READY is unchanged.
  Adjust waiting on the nfs_client_active_wq accordingly.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>

[nfs41: rename NFS_CS_SESSION_SETUP to NFS_CS_SESSION_INITING]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: set NFS_CL_SESSION_INITING in alloc_session]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: move session setup into a function]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[moved nfs4_proc_create_session declaration here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/client.c           |   10 ++++++----
 fs/nfs/internal.h         |    1 +
 fs/nfs/nfs4_fs.h          |    2 ++
 fs/nfs/nfs4proc.c         |   10 ++++++++++
 fs/nfs/nfs4state.c        |   35 ++++++++++++++++++++++++++++++++++-
 include/linux/nfs_fs_sb.h |    1 +
 6 files changed, 54 insertions(+), 5 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index fda2c53..8a3aa9a 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -366,7 +366,8 @@ struct nfs_client *nfs_find_client(const struct sockaddr *addr, u32 nfsversion)
 		struct sockaddr *clap = (struct sockaddr *)&clp->cl_addr;
 
 		/* Don't match clients that failed to initialise properly */
-		if (clp->cl_cons_state != NFS_CS_READY)
+		if (!(clp->cl_cons_state == NFS_CS_READY ||
+		      clp->cl_cons_state == NFS_CS_SESSION_INITING))
 			continue;
 
 		/* Different NFS versions cannot share the same nfs_client */
@@ -499,7 +500,7 @@ found_client:
 		nfs_free_client(new);
 
 	error = wait_event_killable(nfs_client_active_wq,
-				clp->cl_cons_state != NFS_CS_INITING);
+				clp->cl_cons_state < NFS_CS_INITING);
 	if (error < 0) {
 		nfs_put_client(clp);
 		return ERR_PTR(-ERESTARTSYS);
@@ -520,7 +521,7 @@ found_client:
 /*
  * Mark a server as ready or failed
  */
-static void nfs_mark_client_ready(struct nfs_client *clp, int state)
+void nfs_mark_client_ready(struct nfs_client *clp, int state)
 {
 	clp->cl_cons_state = state;
 	wake_up_all(&nfs_client_active_wq);
@@ -1150,7 +1151,8 @@ static int nfs4_init_client(struct nfs_client *clp,
 	if (error < 0)
 		goto error;
 
-	nfs_mark_client_ready(clp, NFS_CS_READY);
+	if (!nfs4_has_session(clp))
+		nfs_mark_client_ready(clp, NFS_CS_READY);
 	return 0;
 
 error:
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index f62bc52..f3b310e 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -100,6 +100,7 @@ extern void nfs_free_server(struct nfs_server *server);
 extern struct nfs_server *nfs_clone_server(struct nfs_server *,
 					   struct nfs_fh *,
 					   struct nfs_fattr *);
+extern void nfs_mark_client_ready(struct nfs_client *clp, int state);
 #ifdef CONFIG_PROC_FS
 extern int __init nfs_fs_proc_init(void);
 extern void nfs_fs_proc_exit(void);
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index c2ef040..f9195d7 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -44,6 +44,7 @@ enum nfs4_client_state {
 	NFS4CLNT_RECLAIM_REBOOT,
 	NFS4CLNT_RECLAIM_NOGRACE,
 	NFS4CLNT_DELEGRETURN,
+	NFS4CLNT_SESSION_SETUP,
 };
 
 /*
@@ -208,6 +209,7 @@ extern int nfs4_setup_sequence(struct nfs_client *clp,
 		int cache_reply, struct rpc_task *task);
 extern void nfs4_destroy_session(struct nfs4_session *session);
 extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
+extern int nfs4_proc_create_session(struct nfs_client *, int reset);
 #else /* CONFIG_NFS_v4_1 */
 static inline int nfs4_setup_sequence(struct nfs_client *clp,
 		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index ba68d43..8e185eb 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4385,6 +4385,16 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 	session = kzalloc(sizeof(struct nfs4_session), GFP_KERNEL);
 	if (!session)
 		return NULL;
+
+	set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+	/*
+	 * The create session reply races with the server back
+	 * channel probe. Mark the client NFS_CS_SESSION_INITING
+	 * so that the client back channel can find the
+	 * nfs_client struct
+	 */
+	clp->cl_cons_state = NFS_CS_SESSION_INITING;
+
 	tbl = &session->fc_slot_table;
 	spin_lock_init(&tbl->slot_tbl_lock);
 	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "Slot table");
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index bc683ed..df5b480 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1113,6 +1113,27 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 	return status;
 }
 
+#ifdef CONFIG_NFS_V4_1
+
+static int nfs4_initialize_session(struct nfs_client *clp)
+{
+	int status;
+
+	status = nfs4_proc_create_session(clp, 0);
+	if (!status) {
+		nfs_mark_client_ready(clp, NFS_CS_READY);
+	} else if (status == -NFS4ERR_STALE_CLIENTID) {
+		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+		set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+	} else {
+		nfs_mark_client_ready(clp, status);
+	}
+	return status;
+}
+#else /* CONFIG_NFS_V4_1 */
+static int nfs4_initialize_session(struct nfs_client *clp) { return 0; }
+#endif /* CONFIG_NFS_V4_1 */
+
 static void nfs4_state_manager(struct nfs_client *clp)
 {
 	int status = 0;
@@ -1126,6 +1147,9 @@ static void nfs4_state_manager(struct nfs_client *clp)
 				set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
 				if (status == -EAGAIN)
 					continue;
+				if (clp->cl_cons_state ==
+							NFS_CS_SESSION_INITING)
+					nfs_mark_client_ready(clp, status);
 				goto out_error;
 			}
 			clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state);
@@ -1136,7 +1160,16 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			if (status != 0)
 				continue;
 		}
-
+		/* Setup the session */
+		if (nfs4_has_session(clp) &&
+		   test_and_clear_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) {
+			status = nfs4_initialize_session(clp);
+			if (status) {
+				if (status == -NFS4ERR_STALE_CLIENTID)
+					continue;
+				goto out_error;
+			}
+		}
 		/* First recover reboot state... */
 		if (test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) {
 			status = nfs4_do_reclaim(clp, &nfs4_reboot_recovery_ops);
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 876e5ff..bc90ad8 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -24,6 +24,7 @@ struct nfs_client {
 	int			cl_cons_state;	/* current construction state (-ve: init error) */
 #define NFS_CS_READY		0		/* ready to be used */
 #define NFS_CS_INITING		1		/* busy initialising */
+#define NFS_CS_SESSION_INITING	2		/* busy initialising  session */
 	unsigned long		cl_res_state;	/* NFS resources state */
 #define NFS_CS_CALLBACK		1		/* - callback started */
 #define NFS_CS_IDMAP		2		/* - idmap started */
-- 
1.6.2.1

[PATCH v2 54/67] nfs41: add session reset to state manager

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Move the code to reset a session from the session_reclaimer to the
nfs4_state_manager.  Destroy the session, and create a new one. Treat
NFS4ERR_BADSESSION and NFS4ERR_DEADSESSION as a successful
nfs4_proc_destroy_session. Signal nfs4_proc_create_session that this is a
session reset so that the session slot table is re-used.

If the clientid is stale, set both NFS4CLNT_LEASE_EXPIRED and
NFS4CLNT_SESSION_SETUP bits and retry.

Use a switch statement in nfs4_session_recovery_handle_error for future
patche which will add handling for other errors.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>

[nfs41: session reset in nfs4_recovery_handle_error]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: reset session on nfs4_do_reclaim session reset error]
    If nfs4_do_reclaim gets a session reset error, nfs4_recovery_handle_error
    will set the NFS4CLNT_SESSION_SETUP bit, and the state manager should
    continue processing to reset the session.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move nfs4_proc_destroy_session declaration here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4_fs.h   |    1 +
 fs/nfs/nfs4state.c |   51 +++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 50 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index f9195d7..ea59327 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -210,6 +210,7 @@ extern int nfs4_setup_sequence(struct nfs_client *clp,
 extern void nfs4_destroy_session(struct nfs4_session *session);
 extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
 extern int nfs4_proc_create_session(struct nfs_client *, int reset);
+extern int nfs4_proc_destroy_session(struct nfs4_session *);
 #else /* CONFIG_NFS_v4_1 */
 static inline int nfs4_setup_sequence(struct nfs_client *clp,
 		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index df5b480..8b7f007 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1042,6 +1042,14 @@ static void nfs4_recovery_handle_error(struct nfs_client *clp, int error)
 		case -NFS4ERR_EXPIRED:
 			set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
 			nfs4_state_start_reclaim_nograce(clp);
+		case -NFS4ERR_BADSESSION:
+		case -NFS4ERR_BADSLOT:
+		case -NFS4ERR_BAD_HIGH_SLOT:
+		case -NFS4ERR_DEADSESSION:
+		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+		case -NFS4ERR_SEQ_FALSE_RETRY:
+		case -NFS4ERR_SEQ_MISORDERED:
+			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
 	}
 }
 
@@ -1114,6 +1122,36 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 }
 
 #ifdef CONFIG_NFS_V4_1
+static void nfs4_session_recovery_handle_error(struct nfs_client *clp, int err)
+{
+	switch (err) {
+	case -NFS4ERR_STALE_CLIENTID:
+		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+		set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+	}
+}
+
+static int nfs4_reset_session(struct nfs_client *clp)
+{
+	int status;
+
+	status = nfs4_proc_destroy_session(clp->cl_session);
+	if (status && status != -NFS4ERR_BADSESSION &&
+	    status != -NFS4ERR_DEADSESSION) {
+		nfs4_session_recovery_handle_error(clp, status);
+		goto out;
+	}
+
+	memset(clp->cl_session->sess_id.data, 0, NFS4_MAX_SESSIONID_LEN);
+	status = nfs4_proc_create_session(clp, 1);
+	if (status)
+		nfs4_session_recovery_handle_error(clp, status);
+		/* fall through*/
+out:
+	/* Wake up the next rpc task even on error */
+	rpc_wake_up_next(&clp->cl_session->fc_slot_table.slot_tbl_waitq);
+	return status;
+}
 
 static int nfs4_initialize_session(struct nfs_client *clp)
 {
@@ -1131,6 +1169,7 @@ static int nfs4_initialize_session(struct nfs_client *clp)
 	return status;
 }
 #else /* CONFIG_NFS_V4_1 */
+static int nfs4_reset_session(struct nfs_client *clp) { return 0; }
 static int nfs4_initialize_session(struct nfs_client *clp) { return 0; }
 #endif /* CONFIG_NFS_V4_1 */
 
@@ -1160,10 +1199,13 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			if (status != 0)
 				continue;
 		}
-		/* Setup the session */
+		/* Initialize or reset the session */
 		if (nfs4_has_session(clp) &&
 		   test_and_clear_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) {
-			status = nfs4_initialize_session(clp);
+			if (clp->cl_cons_state == NFS_CS_SESSION_INITING)
+				status = nfs4_initialize_session(clp);
+			else
+				status = nfs4_reset_session(clp);
 			if (status) {
 				if (status == -NFS4ERR_STALE_CLIENTID)
 					continue;
@@ -1175,6 +1217,8 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			status = nfs4_do_reclaim(clp, &nfs4_reboot_recovery_ops);
 			if (status == -NFS4ERR_STALE_CLIENTID)
 				continue;
+			if (test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
+				continue;
 			nfs4_state_end_reclaim_reboot(clp);
 			continue;
 		}
@@ -1188,6 +1232,9 @@ static void nfs4_state_manager(struct nfs_client *clp)
 					continue;
 				if (status == -NFS4ERR_EXPIRED)
 					continue;
+				if (test_bit(NFS4CLNT_SESSION_SETUP,
+								&clp->cl_state))
+					continue;
 				goto out_error;
 			} else
 				nfs4_state_end_reclaim_nograce(clp);
-- 
1.6.2.1

[PATCH v2 55/67] nfs41: sunrpc: Export the call prepare state for session reset

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/clnt.h  |    1 +
 include/linux/sunrpc/sched.h |    1 +
 net/sunrpc/clnt.c            |   13 +++++++++++++
 net/sunrpc/sched.c           |    2 +-
 4 files changed, 16 insertions(+), 1 deletions(-)

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index c39a210..37881f1 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -143,6 +143,7 @@ int		rpc_call_sync(struct rpc_clnt *clnt,
 			      const struct rpc_message *msg, int flags);
 struct rpc_task *rpc_call_null(struct rpc_clnt *clnt, struct rpc_cred *cred,
 			       int flags);
+void		rpc_restart_call_prepare(struct rpc_task *);
 void		rpc_restart_call(struct rpc_task *);
 void		rpc_setbufsize(struct rpc_clnt *, unsigned int, unsigned int);
 size_t		rpc_max_payload(struct rpc_clnt *);
diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 64981a2..1773768 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -237,6 +237,7 @@ void		rpc_show_tasks(void);
 int		rpc_init_mempool(void);
 void		rpc_destroy_mempool(void);
 extern struct workqueue_struct *rpciod_workqueue;
+void		rpc_prepare_task(struct rpc_task *task);
 
 static inline void rpc_exit(struct rpc_task *task, int status)
 {
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 5abab09..d00e813 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -695,6 +695,19 @@ void rpc_force_rebind(struct rpc_clnt *clnt)
 EXPORT_SYMBOL_GPL(rpc_force_rebind);
 
 /*
+ * Restart an (async) RPC call from the call_prepare state.
+ * Usually called from within the exit handler.
+ */
+void
+rpc_restart_call_prepare(struct rpc_task *task)
+{
+	if (RPC_ASSASSINATED(task))
+		return;
+	task->tk_action = rpc_prepare_task;
+}
+EXPORT_SYMBOL_GPL(rpc_restart_call_prepare);
+
+/*
  * Restart an (async) RPC call. Usually called from within the
  * exit handler.
  */
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index ff50a05..1102ce1 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -569,7 +569,7 @@ EXPORT_SYMBOL_GPL(rpc_delay);
 /*
  * Helper to call task->tk_ops->rpc_call_prepare
  */
-static void rpc_prepare_task(struct rpc_task *task)
+void rpc_prepare_task(struct rpc_task *task)
 {
 	task->tk_ops->rpc_call_prepare(task, task->tk_calldata);
 }
-- 
1.6.2.1

[PATCH v2 56/67] nfs41: use rpc prepare call state for session reset

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

[nfs41: change nfs4_restart_rpc argument]
[nfs41: check for session not minorversion]
[nfs41: trigger the state manager for session reset]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/internal.h |    4 ++++
 fs/nfs/nfs4proc.c |   27 ++++++++++++++++++++-------
 fs/nfs/read.c     |    2 +-
 fs/nfs/unlink.c   |    2 +-
 fs/nfs/write.c    |    8 ++++----
 5 files changed, 30 insertions(+), 13 deletions(-)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index f3b310e..dc92be7 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -148,6 +148,10 @@ extern __be32 * nfs_decode_dirent(__be32 *, struct nfs_entry *, int);
 extern struct rpc_procinfo nfs3_procedures[];
 extern __be32 *nfs3_decode_dirent(__be32 *, struct nfs_entry *, int);
 
+/* nfs4proc.c */
+extern void nfs4_restart_rpc(struct rpc_task *task,
+			     const struct nfs_client *clp);
+
 /* nfs4xdr.c */
 #ifdef CONFIG_NFS_V4
 extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 8e185eb..984a13a 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -586,6 +586,18 @@ static void nfs4_sequence_done_free_slot(const struct nfs_server *server,
 	nfs4_sequence_free_slot(server->nfs_client, res);
 }
 
+void nfs4_restart_rpc(struct rpc_task *task, const struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(clp) &&
+	    test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) {
+		rpc_restart_call_prepare(task);
+		return;
+	}
+#endif /* CONFIG_NFS_V4_1 */
+	rpc_restart_call(task);
+}
+
 static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
@@ -1661,7 +1673,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 				break;
 		default:
 			if (nfs4_async_handle_error(task, server, state) == -EAGAIN) {
-				rpc_restart_call(task);
+				nfs4_restart_rpc(task, server->nfs_client);
 				return;
 			}
 	}
@@ -2874,7 +2886,7 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
 	nfs4_sequence_done(server, &data->res.seq_res, task->tk_status);
 
 	if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) {
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, server->nfs_client);
 		return -EAGAIN;
 	}
 
@@ -2899,7 +2911,7 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
 			   task->tk_status);
 
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) {
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
 		return -EAGAIN;
 	}
 	if (task->tk_status >= 0) {
@@ -2927,7 +2939,7 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
 			   task->tk_status);
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) {
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
 		return -EAGAIN;
 	}
 	nfs4_sequence_free_slot(NFS_SERVER(inode)->nfs_client,
@@ -3631,7 +3643,8 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 			break;
 		default:
 			if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN)
-				rpc_restart_call(task);
+				nfs4_restart_rpc(task,
+						calldata->server->nfs_client);
 	}
 	nfs4_sequence_free_slot(calldata->server->nfs_client,
 				&calldata->res.seq_res);
@@ -4240,7 +4253,7 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata)
 		dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status);
 		rpc_delay(task, NFS4_POLL_RETRY_MIN);
 		task->tk_status = 0;
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, data->clp);
 		return;
 	}
 	nfs41_sequence_free_slot(data->clp, &data->res->lr_seq_res);
@@ -4650,7 +4663,7 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data)
 
 		if (_nfs4_async_handle_error(task, NULL, clp, NULL)
 								== -EAGAIN) {
-			rpc_restart_call(task);
+			nfs4_restart_rpc(task, clp);
 			return;
 		}
 	}
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index d71f0d5..96c4ebf 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -371,7 +371,7 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data
 	argp->offset += resp->count;
 	argp->pgbase += resp->count;
 	argp->count -= resp->count;
-	rpc_restart_call(task);
+	nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client);
 	return;
 out:
 	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index 089a21b..1064c91 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -83,7 +83,7 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata)
 	struct inode *dir = data->dir;
 
 	if (!NFS_PROTO(dir)->unlink_done(task, dir))
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client);
 }
 
 /**
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 85a7640..ce72882 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1145,6 +1145,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	struct nfs_writeargs	*argp = &data->args;
 	struct nfs_writeres	*resp = &data->res;
+	struct nfs_server	*server = NFS_SERVER(data->inode);
 	int status;
 
 	dprintk("NFS: %5u nfs_writeback_done (status %d)\n",
@@ -1177,7 +1178,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 		if (time_before(complain, jiffies)) {
 			dprintk("NFS:       faulty NFS server %s:"
 				" (committed = %d) != (stable = %d)\n",
-				NFS_SERVER(data->inode)->nfs_client->cl_hostname,
+				server->nfs_client->cl_hostname,
 				resp->verf->committed, argp->stable);
 			complain = jiffies + 300 * HZ;
 		}
@@ -1203,7 +1204,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 				 */
 				argp->stable = NFS_FILE_SYNC;
 			}
-			rpc_restart_call(task);
+			nfs4_restart_rpc(task, server->nfs_client);
 			return -EAGAIN;
 		}
 		if (time_before(complain, jiffies)) {
@@ -1215,8 +1216,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 		/* Can't do anything about it except throw an error. */
 		task->tk_status = -EIO;
 	}
-	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
-				&data->res.seq_res);
+	nfs4_sequence_free_slot(server->nfs_client, &data->res.seq_res);
 	return 0;
 }
 
-- 
1.6.2.1

[PATCH v2 57/67] nfs41: kick start nfs41 session recovery when handling errors

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Remove checking for any errors that the SEQUENCE operation does not return.
-NFS4ERR_STALE_CLIENTID, NFS4ERR_EXPIRED, NFS4ERR_CB_PATH_DOWN, NFS4ERR_BACK_CHAN_BUSY, NFS4ERR_OP_NOT_IN_SESSION.

SEQUENCE operation error recovery is very primative, we only reset the session.

Remove checking for any errors that are returned by the SEQUENCE operation, but
that resetting the session won't address.
NFS4ERR_RETRY_UNCACHED_REP, NFS4ERR_SEQUENCE_POS,NFS4ERR_TOO_MANY_OPS.

Add error checking for missing SEQUENCE errors that a session reset will
address.
NFS4ERR_BAD_HIGH_SLOT, NFS4ERR_DEADSESSION, NFS4ERR_SEQ_FALSE_RETRY.

A reset of the session is currently our only response to a SEQUENCE operation
error. Don't reset the session on errors where a new session won't help.

Don't reset the session on errors where a new session won't help.

[nfs41: nfs4_async_handle_error update error checking]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: trigger the state manager for session reset]
    Replace session state bit with nfs_client state bit.  Set the
    NFS4CLNT_SESSION_SETUP bit upon a session related error in the sync/async
    error handlers.
[nfs41: _nfs4_async_handle_error fix session reset error list]
Sequence operation errors that session reset could help.
NFS4ERR_BADSESSION
NFS4ERR_BADSLOT
NFS4ERR_BAD_HIGH_SLOT
NFS4ERR_DEADSESSION
NFS4ERR_CONN_NOT_BOUND_TO_SESSION
NFS4ERR_SEQ_FALSE_RETRY
NFS4ERR_SEQ_MISORDERED

Sequence operation errors that a session reset would not help

NFS4ERR_BADXDR
NFS4ERR_DELAY
NFS4ERR_REP_TOO_BIG
NFS4ERR_REP_TOO_BIG_TO_CACHE
NFS4ERR_REQ_TOO_BIG
NFS4ERR_RETRY_UNCACHED_REP
NFS4ERR_SEQUENCE_POS
NFS4ERR_TOO_MANY_OPS

Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41 nfs4_handle_exception fix session reset error list]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[moved nfs41_sequece_call_done code to nfs41: sequence operation]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   32 ++++++++++++++++++++++++++++++++
 1 files changed, 32 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 984a13a..a9c5b00 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -249,7 +249,25 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 			ret = nfs4_wait_clnt_recover(clp);
 			if (ret == 0)
 				exception->retry = 1;
+#if !defined(CONFIG_NFS_V4_1)
 			break;
+#else /* !defined(CONFIG_NFS_V4_1) */
+			if (!nfs4_has_session(server->nfs_client))
+				break;
+			/* FALLTHROUGH */
+		case -NFS4ERR_BADSESSION:
+		case -NFS4ERR_BADSLOT:
+		case -NFS4ERR_BAD_HIGH_SLOT:
+		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+		case -NFS4ERR_DEADSESSION:
+		case -NFS4ERR_SEQ_FALSE_RETRY:
+		case -NFS4ERR_SEQ_MISORDERED:
+			dprintk("%s ERROR: %d Reset session\n", __func__,
+				errorcode);
+			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			exception->retry = 1;
+			/* FALLTHROUGH */
+#endif /* !defined(CONFIG_NFS_V4_1) */
 		case -NFS4ERR_FILE_OPEN:
 		case -NFS4ERR_GRACE:
 		case -NFS4ERR_DELAY:
@@ -3256,6 +3274,20 @@ _nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 				rpc_wake_up_queued_task(&clp->cl_rpcwaitq, task);
 			task->tk_status = 0;
 			return -EAGAIN;
+#if defined(CONFIG_NFS_V4_1)
+		case -NFS4ERR_BADSESSION:
+		case -NFS4ERR_BADSLOT:
+		case -NFS4ERR_BAD_HIGH_SLOT:
+		case -NFS4ERR_DEADSESSION:
+		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+		case -NFS4ERR_SEQ_FALSE_RETRY:
+		case -NFS4ERR_SEQ_MISORDERED:
+			dprintk("%s ERROR %d, Reset session\n", __func__,
+				task->tk_status);
+			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			task->tk_status = 0;
+			return -EAGAIN;
+#endif /* CONFIG_NFS_V4_1 */
 		case -NFS4ERR_DELAY:
 			if (server)
 				nfs_inc_server_stats(server, NFSIOS_DELAY);
-- 
1.6.2.1

[PATCH v2 58/67] nfs41: schedule async session reset

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Define a new session reset state which is set upon a sequence operation error
in both the sync and async error handlers.

Place all new requests and all but the last outstanding rpc on the
slot_tbl_waitq. Spawn the recovery thread when the last slot is free.
Call nfs4_proc_destroy_session, reinitialize the session, call
nfs4_proc_create_session, clear the session reset state, and wake up the next
task on the slot_tbl_waitq.

Return the nfs4_proc_destroy_session status to the session reclaimer and
check for NFS4ERR_BADSESSION and NFS4ERR_DEADSESSION. Other destroy session
errors should be handled in nfs4_proc_destroy_session where the call can
be retried with adjusted arguments.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
nfs41: make nfs4_wait_bit_killable public]
    nfs4_wait_bit_killable to be used by NFSv4.1 session recover logic.
Signed-off-by: Rahul Iyer <iyer@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: have create_session work on nfs_client]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: trigger the state manager for session reset]
    Replace the session reset state with the NFS4CLNT_SESSION_SETUP cl_state.
    Place all rpc tasks to sleep on the slot table waitqueue until the slot
    table is drained, then schedule state recovery and wait for it to complete.
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: remove nfs41_session_recovery [ch]
Replaced by using the nfs4_state_manager.
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: nfs4_wait_bit_killable only used locally]
[nfs41: keep nfs4_wait_bit_killable static]
[nfs41: keep const nfs_server in nfs4_handle_exception]
[nfs41: remove session parameter from nfs4_find_slot]
Signed-off-by: Andy Adamson <andros@netapp.com
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: resset the session from nfs41_setup_sequence]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |   34 ++++++++++++++++++++++++++++++++++
 1 files changed, 34 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a9c5b00..3b3d1df 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -423,6 +423,22 @@ out:
 	return ret_id;
 }
 
+static int nfs4_recover_session(struct nfs4_session *session)
+{
+	struct nfs_client *clp = session->clp;
+	int ret;
+
+	for (;;) {
+		ret = nfs4_wait_clnt_recover(clp);
+		if (ret != 0)
+				return ret;
+		if (!test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
+			break;
+		nfs4_schedule_state_manager(clp);
+	}
+	return 0;
+}
+
 static int nfs41_setup_sequence(struct nfs4_session *session,
 				struct nfs4_sequence_args *args,
 				struct nfs4_sequence_res *res,
@@ -431,6 +447,7 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 {
 	struct nfs4_slot *slot;
 	struct nfs4_slot_table *tbl;
+	int status = 0;
 	u8 slotid;
 
 	dprintk("--> %s\n", __func__);
@@ -443,6 +460,23 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 	tbl = &session->fc_slot_table;
 
 	spin_lock(&tbl->slot_tbl_lock);
+	if (test_bit(NFS4CLNT_SESSION_SETUP, &session->clp->cl_state)) {
+		if (tbl->highest_used_slotid != -1) {
+			rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
+			spin_unlock(&tbl->slot_tbl_lock);
+			dprintk("<-- %s: Session reset: draining\n", __func__);
+			return -EAGAIN;
+		}
+
+		/* The slot table is empty; start the reset thread */
+		dprintk("%s Session Reset\n", __func__);
+		spin_unlock(&tbl->slot_tbl_lock);
+		status = nfs4_recover_session(session);
+		if (status)
+			return status;
+		spin_lock(&tbl->slot_tbl_lock);
+	}
+
 	slotid = nfs4_find_slot(tbl, task);
 	if (slotid == NFS4_MAX_SLOT_TABLE) {
 		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
-- 
1.6.2.1

[PATCH v2 59/67] nfs41: lease renewal

[Benny Halevy]

Send a NFSv4.1 SEQUENCE op rather than RENEW that was deprecated in
minorversion 1.
Use the nfs_client minorversion to select reboot_recover/
network_partition_recovery/state_renewal ops.

Note: we use reclaimer to create the nfs41 session before there are any
cl_superblocks for the nfs_client.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[revamped patch for new nfs4_state_manager design]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: obliterate nfs4_state_recovery_ops.renew_lease method]
    moved to nfs4_state_maintenance_ops
[also undid per-minorversion nfs4_state_recovery_ops here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4_fs.h    |    6 ++++++
 fs/nfs/nfs4proc.c   |   24 ++++++++++++++++++++++++
 fs/nfs/nfs4renewd.c |    4 +++-
 3 files changed, 33 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index ea59327..43310ee 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -180,6 +180,10 @@ struct nfs4_state_recovery_ops {
 	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
 };
 
+struct nfs4_state_maintenance_ops {
+	int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *);
+};
+
 extern struct dentry_operations nfs4_dentry_operations;
 extern const struct inode_operations nfs4_dir_inode_operations;
 
@@ -220,6 +224,8 @@ static inline int nfs4_setup_sequence(struct nfs_client *clp,
 }
 #endif /* CONFIG_NFS_V4_1 */
 
+extern struct nfs4_state_maintenance_ops *nfs4_state_renewal_ops[];
+
 extern const u32 nfs4_fattr_bitmap[2];
 extern const u32 nfs4_statfs_bitmap[2];
 extern const u32 nfs4_pathconf_bitmap[2];
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 3b3d1df..f047185 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3031,6 +3031,9 @@ static void nfs4_renew_done(struct rpc_task *task, void *data)
 	if (time_before(clp->cl_last_renewal,timestamp))
 		clp->cl_last_renewal = timestamp;
 	spin_unlock(&clp->cl_lock);
+	dprintk("%s calling put_rpccred on rpc_cred %p\n", __func__,
+				task->tk_msg.rpc_cred);
+	put_rpccred(task->tk_msg.rpc_cred);
 }
 
 static const struct rpc_call_ops nfs4_renew_ops = {
@@ -4805,6 +4808,27 @@ struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops = {
 	.recover_lock	= nfs4_lock_expired,
 };
 
+struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = {
+	.sched_state_renewal = nfs4_proc_async_renew,
+};
+
+#if defined(CONFIG_NFS_V4_1)
+struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = {
+	.sched_state_renewal = nfs41_proc_async_sequence,
+};
+#endif
+
+/*
+ * Per minor version reboot and network partition recovery ops
+ */
+
+struct nfs4_state_maintenance_ops *nfs4_state_renewal_ops[] = {
+	&nfs40_state_renewal_ops,
+#if defined(CONFIG_NFS_V4_1)
+	&nfs41_state_renewal_ops,
+#endif
+};
+
 static const struct inode_operations nfs4_file_inode_operations = {
 	.permission	= nfs_permission,
 	.getattr	= nfs_getattr,
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index f524e93..92b43e8 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -59,12 +59,14 @@
 void
 nfs4_renew_state(struct work_struct *work)
 {
+	struct nfs4_state_maintenance_ops *ops;
 	struct nfs_client *clp =
 		container_of(work, struct nfs_client, cl_renewd.work);
 	struct rpc_cred *cred;
 	long lease, timeout;
 	unsigned long last, now;
 
+	ops = nfs4_state_renewal_ops[clp->cl_minorversion];
 	dprintk("%s: start\n", __func__);
 	/* Are there any active superblocks? */
 	if (list_empty(&clp->cl_superblocks))
@@ -86,7 +88,7 @@ nfs4_renew_state(struct work_struct *work)
 			nfs_expire_all_delegations(clp);
 		} else {
 			/* Queue an asynchronous RENEW. */
-			nfs4_proc_async_renew(clp, cred);
+			ops->sched_state_renewal(clp, cred);
 			put_rpccred(cred);
 		}
 		timeout = (2 * lease) / 3;
-- 
1.6.2.1

[PATCH v2 60/67] nfs41: support minorversion 1 for nfs4_check_lease

[Benny Halevy]

[moved nfs4_get_renew_cred related changes to
 "nfs41: introduce get_state_renewal_cred"]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4_fs.h   |    1 +
 fs/nfs/nfs4proc.c  |    2 ++
 fs/nfs/nfs4state.c |    4 +++-
 3 files changed, 6 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 43310ee..c2c54e1 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -182,6 +182,7 @@ struct nfs4_state_recovery_ops {
 
 struct nfs4_state_maintenance_ops {
 	int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *);
+	int (*renew_lease)(struct nfs_client *, struct rpc_cred *);
 };
 
 extern struct dentry_operations nfs4_dentry_operations;
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index f047185..9d25c0d 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4810,11 +4810,13 @@ struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops = {
 
 struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = {
 	.sched_state_renewal = nfs4_proc_async_renew,
+	.renew_lease = nfs4_proc_renew,
 };
 
 #if defined(CONFIG_NFS_V4_1)
 struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = {
 	.sched_state_renewal = nfs41_proc_async_sequence,
+	.renew_lease = nfs4_proc_sequence,
 };
 #endif
 
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 8b7f007..dc5599c 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1083,6 +1083,8 @@ restart:
 static int nfs4_check_lease(struct nfs_client *clp)
 {
 	struct rpc_cred *cred;
+	struct nfs4_state_maintenance_ops *ops =
+		nfs4_state_renewal_ops[clp->cl_minorversion];
 	int status = -NFS4ERR_EXPIRED;
 
 	/* Is the client already known to have an expired lease? */
@@ -1094,7 +1096,7 @@ static int nfs4_check_lease(struct nfs_client *clp)
 		if (cred == NULL)
 			goto out;
 	}
-	status = nfs4_proc_renew(clp, cred);
+	status = ops->renew_lease(clp, cred);
 	put_rpccred(cred);
 out:
 	nfs4_recovery_handle_error(clp, status);
-- 
1.6.2.1

[PATCH v2 61/67] nfs41: introduce get_state_renewal_cred

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Use the machine cred for sending SEQUENCE to renew
the client's lease.

[revamp patch for new state management design starting 2.6.29]
[nfs41: support minorversion 1 for nfs4_check_lease]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: get cred in exchange_id when cred arg is NULL]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use cl_machined_cred instead of cl_ex_cred]
    Since EXCHANGE_ID insists on using the machine credential, cl_ex_cred is
    not needed. nfs4_proc_exchange_id() is only called if the machine credential
    is available. Remove the credential logic from nfs4_proc_exchange_id.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4_fs.h    |    5 +++++
 fs/nfs/nfs4proc.c   |    3 +++
 fs/nfs/nfs4renewd.c |    2 +-
 fs/nfs/nfs4state.c  |   18 +++++-------------
 4 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index c2c54e1..757245e 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -182,6 +182,7 @@ struct nfs4_state_recovery_ops {
 
 struct nfs4_state_maintenance_ops {
 	int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *);
+	struct rpc_cred * (*get_state_renewal_cred_locked)(struct nfs_client *);
 	int (*renew_lease)(struct nfs_client *, struct rpc_cred *);
 };
 
@@ -240,7 +241,11 @@ extern void nfs4_kill_renewd(struct nfs_client *);
 extern void nfs4_renew_state(struct work_struct *);
 
 /* nfs4state.c */
+struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp);
 struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp);
+#if defined(CONFIG_NFS_V4_1)
+struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp);
+#endif /* CONFIG_NFS_V4_1 */
 
 extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
 extern void nfs4_put_state_owner(struct nfs4_state_owner *);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 9d25c0d..fd8a950 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4251,6 +4251,7 @@ static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 
 	dprintk("--> %s\n", __func__);
 	BUG_ON(clp == NULL);
+
 	p = (u32 *)verifier.data;
 	*p++ = htonl((u32)clp->cl_boot_time.tv_sec);
 	*p = htonl((u32)clp->cl_boot_time.tv_nsec);
@@ -4810,12 +4811,14 @@ struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops = {
 
 struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = {
 	.sched_state_renewal = nfs4_proc_async_renew,
+	.get_state_renewal_cred_locked = nfs4_get_renew_cred_locked,
 	.renew_lease = nfs4_proc_renew,
 };
 
 #if defined(CONFIG_NFS_V4_1)
 struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = {
 	.sched_state_renewal = nfs41_proc_async_sequence,
+	.get_state_renewal_cred_locked = nfs4_get_machine_cred_locked,
 	.renew_lease = nfs4_proc_sequence,
 };
 #endif
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index 92b43e8..e27c6ce 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -78,7 +78,7 @@ nfs4_renew_state(struct work_struct *work)
 	timeout = (2 * lease) / 3 + (long)last - (long)now;
 	/* Are we close to a lease timeout? */
 	if (time_after(now, last + lease/3)) {
-		cred = nfs4_get_renew_cred_locked(clp);
+		cred = ops->get_state_renewal_cred_locked(clp);
 		spin_unlock(&clp->cl_lock);
 		if (cred == NULL) {
 			if (list_empty(&clp->cl_delegations)) {
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index dc5599c..da940ab 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -77,7 +77,7 @@ static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred)
 	return status;
 }
 
-static struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp)
+struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp)
 {
 	struct rpc_cred *cred = NULL;
 
@@ -114,17 +114,7 @@ struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp)
 	return cred;
 }
 
-static struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp)
-{
-	struct rpc_cred *cred;
-
-	spin_lock(&clp->cl_lock);
-	cred = nfs4_get_renew_cred_locked(clp);
-	spin_unlock(&clp->cl_lock);
-	return cred;
-}
-
-static struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp)
+struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
 	struct rb_node *pos;
@@ -1090,7 +1080,9 @@ static int nfs4_check_lease(struct nfs_client *clp)
 	/* Is the client already known to have an expired lease? */
 	if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
 		return 0;
-	cred = nfs4_get_renew_cred(clp);
+	spin_lock(&clp->cl_lock);
+	cred = ops->get_state_renewal_cred_locked(clp);
+	spin_unlock(&clp->cl_lock);
 	if (cred == NULL) {
 		cred = nfs4_get_setclientid_cred(clp);
 		if (cred == NULL)
-- 
1.6.2.1

[PATCH v2 62/67] nfs41: establish sessions-based clientid

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

nfsv4.1 clientid is established via EXCHANGE_ID rather than
SETCLIENTID{,_CONFIRM}

This is implemented using a new establish_clid method in
nfs4_state_recovery_ops.

nfs41: establish clientid via exchange id only if cred != NULL

>From 2.6.26 reclaimer() uses machine cred for setting up the client id
therefore it is never expected to be NULL.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
[removed dprintk]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: lease renewal]
[revamped patch for new nfs4_state_manager design]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4_fs.h   |    6 ++++--
 fs/nfs/nfs4proc.c  |   40 ++++++++++++++++++++++++++++++++++++++--
 fs/nfs/nfs4state.c |   12 ++++++++----
 3 files changed, 50 insertions(+), 8 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 757245e..e74d310 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -178,6 +178,7 @@ struct nfs4_state_recovery_ops {
 	int state_flag_bit;
 	int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *);
 	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
+	int (*establish_clid)(struct nfs_client *, struct rpc_cred *);
 };
 
 struct nfs4_state_maintenance_ops {
@@ -200,6 +201,7 @@ extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struc
 extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait);
 extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
 extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *);
@@ -207,8 +209,8 @@ extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fh
 extern int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 		struct nfs4_fs_locations *fs_locations, struct page *page);
 
-extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops;
-extern struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops;
+extern struct nfs4_state_recovery_ops *nfs4_reboot_recovery_ops[];
+extern struct nfs4_state_recovery_ops *nfs4_nograce_recovery_ops[];
 #if defined(CONFIG_NFS_V4_1)
 extern int nfs4_setup_sequence(struct nfs_client *clp,
 		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index fd8a950..7ad40fa 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4795,19 +4795,41 @@ static int nfs41_proc_async_sequence(struct nfs_client *clp,
 
 #endif /* CONFIG_NFS_V4_1 */
 
-struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
+struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = {
 	.owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT,
 	.state_flag_bit	= NFS_STATE_RECLAIM_REBOOT,
 	.recover_open	= nfs4_open_reclaim,
 	.recover_lock	= nfs4_lock_reclaim,
+	.establish_clid = nfs4_init_clientid,
 };
 
-struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops = {
+#if defined(CONFIG_NFS_V4_1)
+struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = {
+	.owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT,
+	.state_flag_bit	= NFS_STATE_RECLAIM_REBOOT,
+	.recover_open	= nfs4_open_reclaim,
+	.recover_lock	= nfs4_lock_reclaim,
+	.establish_clid = nfs4_proc_exchange_id,
+};
+#endif /* CONFIG_NFS_V4_1 */
+
+struct nfs4_state_recovery_ops nfs40_nograce_recovery_ops = {
+	.owner_flag_bit = NFS_OWNER_RECLAIM_NOGRACE,
+	.state_flag_bit	= NFS_STATE_RECLAIM_NOGRACE,
+	.recover_open	= nfs4_open_expired,
+	.recover_lock	= nfs4_lock_expired,
+	.establish_clid = nfs4_init_clientid,
+};
+
+#if defined(CONFIG_NFS_V4_1)
+struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = {
 	.owner_flag_bit = NFS_OWNER_RECLAIM_NOGRACE,
 	.state_flag_bit	= NFS_STATE_RECLAIM_NOGRACE,
 	.recover_open	= nfs4_open_expired,
 	.recover_lock	= nfs4_lock_expired,
+	.establish_clid = nfs4_proc_exchange_id,
 };
+#endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = {
 	.sched_state_renewal = nfs4_proc_async_renew,
@@ -4827,6 +4849,20 @@ struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = {
  * Per minor version reboot and network partition recovery ops
  */
 
+struct nfs4_state_recovery_ops *nfs4_reboot_recovery_ops[] = {
+	&nfs40_reboot_recovery_ops,
+#if defined(CONFIG_NFS_V4_1)
+	&nfs41_reboot_recovery_ops,
+#endif
+};
+
+struct nfs4_state_recovery_ops *nfs4_nograce_recovery_ops[] = {
+	&nfs40_nograce_recovery_ops,
+#if defined(CONFIG_NFS_V4_1)
+	&nfs41_nograce_recovery_ops,
+#endif
+};
+
 struct nfs4_state_maintenance_ops *nfs4_state_renewal_ops[] = {
 	&nfs40_state_renewal_ops,
 #if defined(CONFIG_NFS_V4_1)
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index da940ab..e17bd44 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -60,7 +60,7 @@ const nfs4_stateid zero_stateid;
 
 static LIST_HEAD(nfs4_clientid_list);
 
-static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred)
+int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	unsigned short port;
 	int status;
@@ -1098,11 +1098,13 @@ out:
 static int nfs4_reclaim_lease(struct nfs_client *clp)
 {
 	struct rpc_cred *cred;
+	struct nfs4_state_recovery_ops *ops =
+		nfs4_reboot_recovery_ops[clp->cl_minorversion];
 	int status = -ENOENT;
 
 	cred = nfs4_get_setclientid_cred(clp);
 	if (cred != NULL) {
-		status = nfs4_init_client(clp, cred);
+		status = ops->establish_clid(clp, cred);
 		put_rpccred(cred);
 		/* Handle case where the user hasn't set up machine creds */
 		if (status == -EACCES && cred == clp->cl_machine_cred) {
@@ -1208,7 +1210,8 @@ static void nfs4_state_manager(struct nfs_client *clp)
 		}
 		/* First recover reboot state... */
 		if (test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) {
-			status = nfs4_do_reclaim(clp, &nfs4_reboot_recovery_ops);
+			status = nfs4_do_reclaim(clp,
+				nfs4_reboot_recovery_ops[clp->cl_minorversion]);
 			if (status == -NFS4ERR_STALE_CLIENTID)
 				continue;
 			if (test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
@@ -1219,7 +1222,8 @@ static void nfs4_state_manager(struct nfs_client *clp)
 
 		/* Now recover expired state... */
 		if (test_and_clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state)) {
-			status = nfs4_do_reclaim(clp, &nfs4_nograce_recovery_ops);
+			status = nfs4_do_reclaim(clp,
+				nfs4_nograce_recovery_ops[clp->cl_minorversion]);
 			if (status < 0) {
 				set_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state);
 				if (status == -NFS4ERR_STALE_CLIENTID)
-- 
1.6.2.1

[PATCH v2 63/67] nfs41: add a get_clid_cred function to nfs4_state_recovery_ops

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

EXCHANGE_ID has different credential requirements than SETCLIENTID.
Prepare for a separate credential function.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4_fs.h   |    1 +
 fs/nfs/nfs4proc.c  |    2 ++
 fs/nfs/nfs4state.c |    2 +-
 3 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index e74d310..90eebff 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -179,6 +179,7 @@ struct nfs4_state_recovery_ops {
 	int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *);
 	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
 	int (*establish_clid)(struct nfs_client *, struct rpc_cred *);
+	struct rpc_cred * (*get_clid_cred)(struct nfs_client *);
 };
 
 struct nfs4_state_maintenance_ops {
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 7ad40fa..d7a320f 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4801,6 +4801,7 @@ struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = {
 	.recover_open	= nfs4_open_reclaim,
 	.recover_lock	= nfs4_lock_reclaim,
 	.establish_clid = nfs4_init_clientid,
+	.get_clid_cred	= nfs4_get_setclientid_cred,
 };
 
 #if defined(CONFIG_NFS_V4_1)
@@ -4819,6 +4820,7 @@ struct nfs4_state_recovery_ops nfs40_nograce_recovery_ops = {
 	.recover_open	= nfs4_open_expired,
 	.recover_lock	= nfs4_lock_expired,
 	.establish_clid = nfs4_init_clientid,
+	.get_clid_cred	= nfs4_get_setclientid_cred,
 };
 
 #if defined(CONFIG_NFS_V4_1)
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index e17bd44..90e5607 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1102,7 +1102,7 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 		nfs4_reboot_recovery_ops[clp->cl_minorversion];
 	int status = -ENOENT;
 
-	cred = nfs4_get_setclientid_cred(clp);
+	cred = ops->get_clid_cred(clp);
 	if (cred != NULL) {
 		status = ops->establish_clid(clp, cred);
 		put_rpccred(cred);
-- 
1.6.2.1

[PATCH v2 64/67] nfs41: get_clid_cred for EXCHANGE_ID

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

Unlike SETCLIENTID, EXCHANGE_ID requires a machine credential. Do not search
for credentials other than the machine credential.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4_fs.h   |    1 +
 fs/nfs/nfs4proc.c  |    2 ++
 fs/nfs/nfs4state.c |   14 ++++++++++++++
 3 files changed, 17 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 90eebff..72a4a0a 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -248,6 +248,7 @@ struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp);
 struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp);
 #if defined(CONFIG_NFS_V4_1)
 struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp);
+struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp);
 #endif /* CONFIG_NFS_V4_1 */
 
 extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index d7a320f..cd773c1 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4811,6 +4811,7 @@ struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = {
 	.recover_open	= nfs4_open_reclaim,
 	.recover_lock	= nfs4_lock_reclaim,
 	.establish_clid = nfs4_proc_exchange_id,
+	.get_clid_cred	= nfs4_get_exchange_id_cred,
 };
 #endif /* CONFIG_NFS_V4_1 */
 
@@ -4830,6 +4831,7 @@ struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = {
 	.recover_open	= nfs4_open_expired,
 	.recover_lock	= nfs4_lock_expired,
 	.establish_clid = nfs4_proc_exchange_id,
+	.get_clid_cred	= nfs4_get_exchange_id_cred,
 };
 #endif /* CONFIG_NFS_V4_1 */
 
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 90e5607..7dc9713 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -114,6 +114,20 @@ struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp)
 	return cred;
 }
 
+#if defined(CONFIG_NFS_V4_1)
+
+struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp)
+{
+	struct rpc_cred *cred;
+
+	spin_lock(&clp->cl_lock);
+	cred = nfs4_get_machine_cred_locked(clp);
+	spin_unlock(&clp->cl_lock);
+	return cred;
+}
+
+#endif /* CONFIG_NFS_V4_1 */
+
 struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
-- 
1.6.2.1

[PATCH v2 65/67] nfs41: recover lease in _nfs4_lookup_root

[Benny Halevy]

This creates the nfsv4.1 session on mount.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c |    7 ++++++-
 1 files changed, 6 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index cd773c1..e0d8698 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2036,8 +2036,13 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
 		.rpc_argp = &args,
 		.rpc_resp = &res,
 	};
+	int status;
+
 	nfs_fattr_init(info->fattr);
-	return nfs4_call_sync(server, &msg, &args, &res, 0);
+	status = nfs4_recover_expired_lease(server);
+	if (!status)
+		status = nfs4_call_sync(server, &msg, &args, &res, 0);
+	return status;
 }
 
 static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
-- 
1.6.2.1

[PATCH v2 66/67] nfs41: fail mount on EXCHANGE_ID, CREATE_SESSION failure

[Benny Halevy]

From: Andy Adamson <andros@netapp.com>

The nfs4_recover_expired_lease() call schedules the state manager thread and
waits for it to complete. A negative cl_cons_state indicates that either the
EXCHANGE_ID or the CREATE_SESSION failed. In this case return -EPROTONOSUPPORT
which will be returned to the caller (mount).

Note that session setup either succeeds or fails prior to calling
a SEQUENCE operation, so unlike the session_reclaimer NFS41_SESSION_ALLOC
state, there is no need to test for session alloc (setup) in
nfs4_setup_sequence.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/client.c   |   16 ++++++++++++++++
 fs/nfs/internal.h |    1 +
 fs/nfs/nfs4proc.c |    2 ++
 3 files changed, 19 insertions(+), 0 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 8a3aa9a..0c18a18 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -528,6 +528,22 @@ void nfs_mark_client_ready(struct nfs_client *clp, int state)
 }
 
 /*
+ * With sessions, the client is not marked ready until after a
+ * successful EXCHANGE_ID and CREATE_SESSION.
+ *
+ * Map errors cl_cons_state errors to EPROTONOSUPPORT to indicate
+ * other versions of NFS can be tried.
+ */
+int nfs4_check_client_ready(struct nfs_client *clp)
+{
+	if (!nfs4_has_session(clp))
+		return 0;
+	if (clp->cl_cons_state < NFS_CS_READY)
+		return -EPROTONOSUPPORT;
+	return 0;
+}
+
+/*
  * Session has been established, and the client marked ready.
  * Set the mount rsize and wsize with negotiated fore channel
  * attributes which will be bound checked in nfs_server_set_fsinfo.
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index dc92be7..7b5cb33 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -101,6 +101,7 @@ extern struct nfs_server *nfs_clone_server(struct nfs_server *,
 					   struct nfs_fh *,
 					   struct nfs_fattr *);
 extern void nfs_mark_client_ready(struct nfs_client *clp, int state);
+extern int nfs4_check_client_ready(struct nfs_client *clp);
 #ifdef CONFIG_PROC_FS
 extern int __init nfs_fs_proc_init(void);
 extern void nfs_fs_proc_exit(void);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index e0d8698..f6db7a5 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2041,6 +2041,8 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
 	nfs_fattr_init(info->fattr);
 	status = nfs4_recover_expired_lease(server);
 	if (!status)
+		status = nfs4_check_client_ready(server->nfs_client);
+	if (!status)
 		status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	return status;
 }
-- 
1.6.2.1

[PATCH v2 67/67] nfs41: increment_{open,lock}_seqid

[Benny Halevy]

Unlike minorversion0, in nfsv4.1 the open and lock seqids need
not be incremented by the client and should always be set to zero.

This is implemented using a new nfs_rpc_ops methods -
increment_open_seqid and increment_lock_seqid

Signed-off-by: Rahul Iyer <iyer@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c  |    2 ++
 fs/nfs/nfs4state.c |   12 +++++++-----
 2 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index f6db7a5..ef3dca3 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1820,6 +1820,8 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	calldata->state = state;
 	calldata->arg.fh = NFS_FH(state->inode);
 	calldata->arg.stateid = &state->open_stateid;
+	if (nfs4_has_session(server->nfs_client))
+		memset(calldata->arg.stateid->data, 0, 4);    /* clear seqid */
 	/* Serialization for the sequence id */
 	calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid);
 	if (calldata->arg.seqid == NULL)
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 7dc9713..c16f24d 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -742,12 +742,14 @@ static void nfs_increment_seqid(int status, struct nfs_seqid *seqid)
 
 void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid)
 {
-	if (status == -NFS4ERR_BAD_SEQID) {
-		struct nfs4_state_owner *sp = container_of(seqid->sequence,
-				struct nfs4_state_owner, so_seqid);
+	struct nfs4_state_owner *sp = container_of(seqid->sequence,
+					struct nfs4_state_owner, so_seqid);
+	struct nfs_server *server = sp->so_server;
+
+	if (status == -NFS4ERR_BAD_SEQID)
 		nfs4_drop_state_owner(sp);
-	}
-	nfs_increment_seqid(status, seqid);
+	if (!nfs4_has_session(server->nfs_client))
+		nfs_increment_seqid(status, seqid);
 }
 
 /*
-- 
1.6.2.1

Re: [pnfs] [PATCH v2 01/67] nfs41: common protocol definitions

[J. Bruce Fields]

On Thu, Apr 02, 2009 at 04:50:33PM +0300, Benny Halevy wrote:
> Define all NFSv4.1 common operation and error code constants.
> 
> Note that some of the definitions are used by both the nfs41 client
> and the server code. This patch is duplicated in the nfs41 and nfsd41
> sessions patchset.

Oh, and, for what it's worth (and not being sure which route this may go
in by):

	Acked-by: J. Bruce Fields <bfields@citi.umich.edu>

--b.

> Signed-off-by: Andy Adamson<andros@netapp.com>
> Signed-off-by: Benny Halevy <bhalevy@panasas.com>
> [nfs41: add exchange id flags]
> Signed-off-by: Mike Sager <sager@netapp.com>
> Signed-off-by: Benny Halevy <bhalevy@panasas.com>
> [removed server-only hunk changing NFSERR_REPLAY_ME]
> Signed-off-by: Benny Halevy <bhalevy@panasas.com>
> [nfs41: add SEQ4_XX to nfs41-common-protocol]
> Signed-off-by: Andy Adamson <andros@netapp.com>
> Signed-off-by: Benny Halevy <bhalevy@panasas.com>
> [nfs41: generic error code update]
> [nfs41: reverse EXCHGID4_INVAL_FLAG_MASK_{A,R}]
> Signed-off-by: Benny Halevy <bhalevy@panasas.com>
> ---
>  include/linux/nfs4.h |  128 +++++++++++++++++++++++++++++++++++++++++++++++++-
>  1 files changed, 127 insertions(+), 1 deletions(-)
> 
> diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
> index b912311..29ed600 100644
> --- a/include/linux/nfs4.h
> +++ b/include/linux/nfs4.h
> @@ -21,6 +21,7 @@
>  #define NFS4_FHSIZE		128
>  #define NFS4_MAXPATHLEN		PATH_MAX
>  #define NFS4_MAXNAMLEN		NAME_MAX
> +#define NFS4_MAX_SESSIONID_LEN	16
>  
>  #define NFS4_ACCESS_READ        0x0001
>  #define NFS4_ACCESS_LOOKUP      0x0002
> @@ -38,6 +39,7 @@
>  #define NFS4_OPEN_RESULT_CONFIRM 0x0002
>  #define NFS4_OPEN_RESULT_LOCKTYPE_POSIX 0x0004
>  
> +#define NFS4_SHARE_ACCESS_MASK	0x000F
>  #define NFS4_SHARE_ACCESS_READ	0x0001
>  #define NFS4_SHARE_ACCESS_WRITE	0x0002
>  #define NFS4_SHARE_ACCESS_BOTH	0x0003
> @@ -45,6 +47,19 @@
>  #define NFS4_SHARE_DENY_WRITE	0x0002
>  #define NFS4_SHARE_DENY_BOTH	0x0003
>  
> +/* nfs41 */
> +#define NFS4_SHARE_WANT_MASK		0xFF00
> +#define NFS4_SHARE_WANT_NO_PREFERENCE	0x0000
> +#define NFS4_SHARE_WANT_READ_DELEG	0x0100
> +#define NFS4_SHARE_WANT_WRITE_DELEG	0x0200
> +#define NFS4_SHARE_WANT_ANY_DELEG	0x0300
> +#define NFS4_SHARE_WANT_NO_DELEG	0x0400
> +#define NFS4_SHARE_WANT_CANCEL		0x0500
> +
> +#define NFS4_SHARE_WHEN_MASK		0xF0000
> +#define NFS4_SHARE_SIGNAL_DELEG_WHEN_RESRC_AVAIL	0x10000
> +#define NFS4_SHARE_PUSH_DELEG_WHEN_UNCONTENDED		0x20000
> +
>  #define NFS4_SET_TO_SERVER_TIME	0
>  #define NFS4_SET_TO_CLIENT_TIME	1
>  
> @@ -88,6 +103,31 @@
>  #define NFS4_ACE_GENERIC_EXECUTE              0x001200A0
>  #define NFS4_ACE_MASK_ALL                     0x001F01FF
>  
> +#define EXCHGID4_FLAG_SUPP_MOVED_REFER		0x00000001
> +#define EXCHGID4_FLAG_SUPP_MOVED_MIGR		0x00000002
> +#define EXCHGID4_FLAG_USE_NON_PNFS		0x00010000
> +#define EXCHGID4_FLAG_USE_PNFS_MDS		0x00020000
> +#define EXCHGID4_FLAG_USE_PNFS_DS		0x00040000
> +#define EXCHGID4_FLAG_UPD_CONFIRMED_REC_A	0x40000000
> +#define EXCHGID4_FLAG_CONFIRMED_R		0x80000000
> +/*
> + * Since the validity of these bits depends on whether
> + * they're set in the argument or response, have separate
> + * invalid flag masks for arg (_A) and resp (_R).
> + */
> +#define EXCHGID4_FLAG_MASK_A			0x40070003
> +#define EXCHGID4_FLAG_MASK_R			0x80070003
> +
> +#define SEQ4_STATUS_CB_PATH_DOWN		0x00000001
> +#define SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRING	0x00000002
> +#define SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRED	0x00000004
> +#define SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED	0x00000008
> +#define SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED	0x00000010
> +#define SEQ4_STATUS_ADMIN_STATE_REVOKED		0x00000020
> +#define SEQ4_STATUS_RECALLABLE_STATE_REVOKED	0x00000040
> +#define SEQ4_STATUS_LEASE_MOVED			0x00000080
> +#define SEQ4_STATUS_RESTART_RECLAIM_NEEDED	0x00000100
> +
>  #define NFS4_MAX_UINT64	(~(u64)0)
>  
>  enum nfs4_acl_whotype {
> @@ -154,6 +194,28 @@ enum nfs_opnum4 {
>  	OP_VERIFY = 37,
>  	OP_WRITE = 38,
>  	OP_RELEASE_LOCKOWNER = 39,
> +
> +	/* nfs41 */
> +	OP_BACKCHANNEL_CTL = 40,
> +	OP_BIND_CONN_TO_SESSION = 41,
> +	OP_EXCHANGE_ID = 42,
> +	OP_CREATE_SESSION = 43,
> +	OP_DESTROY_SESSION = 44,
> +	OP_FREE_STATEID = 45,
> +	OP_GET_DIR_DELEGATION = 46,
> +	OP_GETDEVICEINFO = 47,
> +	OP_GETDEVICELIST = 48,
> +	OP_LAYOUTCOMMIT = 49,
> +	OP_LAYOUTGET = 50,
> +	OP_LAYOUTRETURN = 51,
> +	OP_SECINFO_NO_NAME = 52,
> +	OP_SEQUENCE = 53,
> +	OP_SET_SSV = 54,
> +	OP_TEST_STATEID = 55,
> +	OP_WANT_DELEGATION = 56,
> +	OP_DESTROY_CLIENTID = 57,
> +	OP_RECLAIM_COMPLETE = 58,
> +
>  	OP_ILLEGAL = 10044,
>  };
>  
> @@ -230,7 +292,48 @@ enum nfsstat4 {
>  	NFS4ERR_DEADLOCK = 10045,
>  	NFS4ERR_FILE_OPEN = 10046,
>  	NFS4ERR_ADMIN_REVOKED = 10047,
> -	NFS4ERR_CB_PATH_DOWN = 10048
> +	NFS4ERR_CB_PATH_DOWN = 10048,
> +
> +	/* nfs41 */
> +	NFS4ERR_BADIOMODE	= 10049,
> +	NFS4ERR_BADLAYOUT	= 10050,
> +	NFS4ERR_BAD_SESSION_DIGEST = 10051,
> +	NFS4ERR_BADSESSION	= 10052,
> +	NFS4ERR_BADSLOT		= 10053,
> +	NFS4ERR_COMPLETE_ALREADY = 10054,
> +	NFS4ERR_CONN_NOT_BOUND_TO_SESSION = 10055,
> +	NFS4ERR_DELEG_ALREADY_WANTED = 10056,
> +	NFS4ERR_BACK_CHAN_BUSY	= 10057,	/* backchan reqs outstanding */
> +	NFS4ERR_LAYOUTTRYLATER	= 10058,
> +	NFS4ERR_LAYOUTUNAVAILABLE = 10059,
> +	NFS4ERR_NOMATCHING_LAYOUT = 10060,
> +	NFS4ERR_RECALLCONFLICT	= 10061,
> +	NFS4ERR_UNKNOWN_LAYOUTTYPE = 10062,
> +	NFS4ERR_SEQ_MISORDERED = 10063, 	/* unexpected seq.id in req */
> +	NFS4ERR_SEQUENCE_POS	= 10064,	/* [CB_]SEQ. op not 1st op */
> +	NFS4ERR_REQ_TOO_BIG	= 10065,	/* request too big */
> +	NFS4ERR_REP_TOO_BIG	= 10066,	/* reply too big */
> +	NFS4ERR_REP_TOO_BIG_TO_CACHE = 10067,	/* rep. not all cached */
> +	NFS4ERR_RETRY_UNCACHED_REP = 10068,	/* retry & rep. uncached */
> +	NFS4ERR_UNSAFE_COMPOUND = 10069,	/* retry/recovery too hard */
> +	NFS4ERR_TOO_MANY_OPS	= 10070,	/* too many ops in [CB_]COMP */
> +	NFS4ERR_OP_NOT_IN_SESSION = 10071,	/* op needs [CB_]SEQ. op */
> +	NFS4ERR_HASH_ALG_UNSUPP = 10072,	/* hash alg. not supp. */
> +						/* Error 10073 is unused. */
> +	NFS4ERR_CLIENTID_BUSY	= 10074,	/* clientid has state */
> +	NFS4ERR_PNFS_IO_HOLE	= 10075,	/* IO to _SPARSE file hole */
> +	NFS4ERR_SEQ_FALSE_RETRY	= 10076,	/* retry not origional */
> +	NFS4ERR_BAD_HIGH_SLOT	= 10077,	/* sequence arg bad */
> +	NFS4ERR_DEADSESSION	= 10078,	/* persistent session dead */
> +	NFS4ERR_ENCR_ALG_UNSUPP = 10079,	/* SSV alg mismatch */
> +	NFS4ERR_PNFS_NO_LAYOUT	= 10080,	/* direct I/O with no layout */
> +	NFS4ERR_NOT_ONLY_OP	= 10081,	/* bad compound */
> +	NFS4ERR_WRONG_CRED	= 10082,	/* permissions:state change */
> +	NFS4ERR_WRONG_TYPE	= 10083,	/* current operation mismatch */
> +	NFS4ERR_DIRDELEG_UNAVAIL = 10084,	/* no directory delegation */
> +	NFS4ERR_REJECT_DELEG	= 10085,	/* on callback */
> +	NFS4ERR_RETURNCONFLICT	= 10086,	/* outstanding layoutreturn */
> +	NFS4ERR_DELEG_REVOKED	= 10087,	/* deleg./layout revoked */
>  };
>  
>  /*
> @@ -391,6 +494,29 @@ enum {
>  	NFSPROC4_CLNT_GETACL,
>  	NFSPROC4_CLNT_SETACL,
>  	NFSPROC4_CLNT_FS_LOCATIONS,
> +
> +	/* nfs41 */
> +	NFSPROC4_CLNT_EXCHANGE_ID,
> +	NFSPROC4_CLNT_CREATE_SESSION,
> +	NFSPROC4_CLNT_DESTROY_SESSION,
> +	NFSPROC4_CLNT_SEQUENCE,
> +	NFSPROC4_CLNT_GET_LEASE_TIME,
> +};
> +
> +/* nfs41 types */
> +struct nfs4_sessionid {
> +	unsigned char data[NFS4_MAX_SESSIONID_LEN];
> +};
> +
> +/* Create Session Flags */
> +#define SESSION4_PERSIST	 0x001
> +#define SESSION4_BACK_CHAN 	 0x002
> +#define SESSION4_RDMA		 0x004
> +
> +enum state_protect_how4 {
> +	SP4_NONE	= 0,
> +	SP4_MACH_CRED	= 1,
> +	SP4_SSV		= 2
>  };
>  
>  #endif
> -- 
> 1.6.2.1
> 
> _______________________________________________
> pNFS mailing list
> pNFS@linux-nfs.org
> http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs

Re: [PATCH v2 0/67] nfs41 client patch for 2.6.30

[Trond Myklebust]

On Thu, 2009-04-02 at 16:46 +0300, Benny Halevy wrote:
> Trond,
> 
> I've integrated the latest client patches from Andy
> (with a few minor changes listed below) and updated
> git://linux-nfs.org/~bhalevy/linux-pnfs.git
> branches:
> 	nfs41-for-next
> 	nfs41-for-2.6.30
> 	nfs41
> 
> nfs41-for-{next,2.6.30} contain all non-backchannel patches
> rebased onto {v2.6.29,nfs-2.6/for-next} respectively.


I've rebased it onto the latest mainline tree and put the result out on
git://git.linux-nfs.org/projects/trondmy/nfs-2.6.git#nfsv41

-- 
Trond Myklebust
Linux NFS client maintainer

NetApp
Trond.Myklebust@netapp.com
www.netapp.com

Re: [PATCH v2 0/67] nfs41 client patch for 2.6.30

[Benny Halevy]

On Apr. 04, 2009, 22:37 +0300, Trond Myklebust <Trond.Myklebust@netapp.com> wrote:
> On Thu, 2009-04-02 at 16:46 +0300, Benny Halevy wrote:
>> Trond,
>>
>> I've integrated the latest client patches from Andy
>> (with a few minor changes listed below) and updated
>> git://linux-nfs.org/~bhalevy/linux-pnfs.git
>> branches:
>> 	nfs41-for-next
>> 	nfs41-for-2.6.30
>> 	nfs41
>>
>> nfs41-for-{next,2.6.30} contain all non-backchannel patches
>> rebased onto {v2.6.29,nfs-2.6/for-next} respectively.
> 
> 
> I've rebased it onto the latest mainline tree and put the result out on
> git://git.linux-nfs.org/projects/trondmy/nfs-2.6.git#nfsv41
> 

Thanks!

Benny