linux/fs/nfs/nfs4_fs.h

437 lines
14 KiB
C
Raw Normal View History

/*
* linux/fs/nfs/nfs4_fs.h
*
* Copyright (C) 2005 Trond Myklebust
*
* NFSv4-specific filesystem definitions and declarations
*/
#ifndef __LINUX_FS_NFS_NFS4_FS_H
#define __LINUX_FS_NFS_NFS4_FS_H
#if IS_ENABLED(CONFIG_NFS_V4)
struct idmap;
enum nfs4_client_state {
NFS4CLNT_MANAGER_RUNNING = 0,
NFS4CLNT_CHECK_LEASE,
NFS4CLNT_LEASE_EXPIRED,
NFS4CLNT_RECLAIM_REBOOT,
NFS4CLNT_RECLAIM_NOGRACE,
NFS4CLNT_DELEGRETURN,
NFS4CLNT_SESSION_RESET,
NFS4CLNT_LEASE_CONFIRM,
NFS4CLNT_SERVER_SCOPE_MISMATCH,
NFS4CLNT_PURGE_STATE,
NFS4CLNT_BIND_CONN_TO_SESSION,
};
enum nfs4_session_state {
NFS4_SESSION_INITING,
NFS4_SESSION_DRAINING,
};
#define NFS4_RENEW_TIMEOUT 0x01
#define NFS4_RENEW_DELEGATION_CB 0x02
struct nfs4_minor_version_ops {
u32 minor_version;
int (*call_sync)(struct rpc_clnt *clnt,
struct nfs_server *server,
struct rpc_message *msg,
struct nfs4_sequence_args *args,
struct nfs4_sequence_res *res,
int cache_reply);
bool (*match_stateid)(const nfs4_stateid *,
const nfs4_stateid *);
int (*find_root_sec)(struct nfs_server *, struct nfs_fh *,
struct nfs_fsinfo *);
const struct nfs4_state_recovery_ops *reboot_recovery_ops;
const struct nfs4_state_recovery_ops *nograce_recovery_ops;
const struct nfs4_state_maintenance_ops *state_renewal_ops;
};
#define NFS_SEQID_CONFIRMED 1
struct nfs_seqid_counter {
ktime_t create_time;
int owner_id;
int flags;
u32 counter;
spinlock_t lock; /* Protects the list */
struct list_head list; /* Defines sequence of RPC calls */
struct rpc_wait_queue wait; /* RPC call delay queue */
};
struct nfs_seqid {
struct nfs_seqid_counter *sequence;
struct list_head list;
struct rpc_task *task;
};
static inline void nfs_confirm_seqid(struct nfs_seqid_counter *seqid, int status)
{
if (seqid_mutating_err(-status))
seqid->flags |= NFS_SEQID_CONFIRMED;
}
/*
* NFS4 state_owners and lock_owners are simply labels for ordered
* sequences of RPC calls. Their sole purpose is to provide once-only
* semantics by allowing the server to identify replayed requests.
*/
struct nfs4_state_owner {
struct nfs_server *so_server;
NFS: Cache state owners after files are closed Servers have a finite amount of memory to store NFSv4 open and lock owners. Moreover, servers may have a difficult time determining when they can reap their state owner table, thanks to gray areas in the NFSv4 protocol specification. Thus clients should be careful to reuse state owners when possible. Currently Linux is not too careful. When a user has closed all her files on one mount point, the state owner's reference count goes to zero, and it is released. The next OPEN allocates a new one. A workload that serially opens and closes files can run through a large number of open owners this way. When a state owner's reference count goes to zero, slap it onto a free list for that nfs_server, with an expiry time. Garbage collect before looking for a state owner. This makes state owners for active users available for re-use. Now that there can be unused state owners remaining at umount time, purge the state owner free list when a server is destroyed. Also be sure not to reclaim unused state owners during state recovery. This change has benefits for the client as well. For some workloads, this approach drops the number of OPEN_CONFIRM calls from the same as the number of OPEN calls, down to just one. This reduces wire traffic and thus open(2) latency. Before this patch, untarring a kernel source tarball shows the OPEN_CONFIRM call counter steadily increasing through the test. With the patch, the OPEN_CONFIRM count remains at 1 throughout the entire untar. As long as the expiry time is kept short, I don't think garbage collection should be terribly expensive, although it does bounce the clp->cl_lock around a bit. [ At some point we should rationalize the use of the nfs_server ->destroy method. ] Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [Trond: Fixed a garbage collection race and a few efficiency issues] Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2011-12-06 21:13:48 +00:00
struct list_head so_lru;
unsigned long so_expires;
struct rb_node so_server_node;
struct rpc_cred *so_cred; /* Associated cred */
spinlock_t so_lock;
atomic_t so_count;
unsigned long so_flags;
struct list_head so_states;
struct nfs_seqid_counter so_seqid;
};
enum {
NFS_OWNER_RECLAIM_REBOOT,
NFS_OWNER_RECLAIM_NOGRACE
};
#define NFS_LOCK_NEW 0
#define NFS_LOCK_RECLAIM 1
#define NFS_LOCK_EXPIRED 2
/*
* struct nfs4_state maintains the client-side state for a given
* (state_owner,inode) tuple (OPEN) or state_owner (LOCK).
*
* OPEN:
* In order to know when to OPEN_DOWNGRADE or CLOSE the state on the server,
* we need to know how many files are open for reading or writing on a
* given inode. This information too is stored here.
*
* LOCK: one nfs4_state (LOCK) to hold the lock stateid nfs4_state(OPEN)
*/
struct nfs4_lock_owner {
unsigned int lo_type;
#define NFS4_ANY_LOCK_TYPE (0U)
#define NFS4_FLOCK_LOCK_TYPE (1U << 0)
#define NFS4_POSIX_LOCK_TYPE (1U << 1)
union {
fl_owner_t posix_owner;
pid_t flock_owner;
} lo_u;
};
struct nfs4_lock_state {
struct list_head ls_locks; /* Other lock stateids */
struct nfs4_state * ls_state; /* Pointer to open state */
#define NFS_LOCK_INITIALIZED 0
unsigned long ls_flags;
struct nfs_seqid_counter ls_seqid;
nfs4_stateid ls_stateid;
atomic_t ls_count;
struct nfs4_lock_owner ls_owner;
};
/* bits for nfs4_state->flags */
enum {
LK_STATE_IN_USE,
NFS_DELEGATED_STATE, /* Current stateid is delegation */
NFS_O_RDONLY_STATE, /* OPEN stateid has read-only state */
NFS_O_WRONLY_STATE, /* OPEN stateid has write-only state */
NFS_O_RDWR_STATE, /* OPEN stateid has read/write state */
NFS_STATE_RECLAIM_REBOOT, /* OPEN stateid server rebooted */
NFS_STATE_RECLAIM_NOGRACE, /* OPEN stateid needs to recover state */
NFS_STATE_POSIX_LOCKS, /* Posix locks are supported */
};
struct nfs4_state {
struct list_head open_states; /* List of states for the same state_owner */
struct list_head inode_states; /* List of states for the same inode */
struct list_head lock_states; /* List of subservient lock stateids */
struct nfs4_state_owner *owner; /* Pointer to the open owner */
struct inode *inode; /* Pointer to the inode */
unsigned long flags; /* Do we hold any locks? */
spinlock_t state_lock; /* Protects the lock_states list */
seqlock_t seqlock; /* Protects the stateid/open_stateid */
nfs4_stateid stateid; /* Current stateid: may be delegation */
nfs4_stateid open_stateid; /* OPEN stateid */
/* The following 3 fields are protected by owner->so_lock */
unsigned int n_rdonly; /* Number of read-only references */
unsigned int n_wronly; /* Number of write-only references */
unsigned int n_rdwr; /* Number of read/write references */
fmode_t state; /* State on the server (R,W, or RW) */
atomic_t count;
};
struct nfs4_exception {
long timeout;
int retry;
struct nfs4_state *state;
struct inode *inode;
};
struct nfs4_state_recovery_ops {
int owner_flag_bit;
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 rpc_cred * (*get_clid_cred)(struct nfs_client *);
int (*reclaim_complete)(struct nfs_client *);
NFS: Discover NFSv4 server trunking when mounting "Server trunking" is a fancy named for a multi-homed NFS server. Trunking might occur if a client sends NFS requests for a single workload to multiple network interfaces on the same server. There are some implications for NFSv4 state management that make it useful for a client to know if a single NFSv4 server instance is multi-homed. (Note this is only a consideration for NFSv4, not for legacy versions of NFS, which are stateless). If a client cares about server trunking, no NFSv4 operations can proceed until that client determines who it is talking to. Thus server IP trunking discovery must be done when the client first encounters an unfamiliar server IP address. The nfs_get_client() function walks the nfs_client_list and matches on server IP address. The outcome of that walk tells us immediately if we have an unfamiliar server IP address. It invokes nfs_init_client() in this case. Thus, nfs4_init_client() is a good spot to perform trunking discovery. Discovery requires a client to establish a fresh client ID, so our client will now send SETCLIENTID or EXCHANGE_ID as the first NFS operation after a successful ping, rather than waiting for an application to perform an operation that requires NFSv4 state. The exact process for detecting trunking is different for NFSv4.0 and NFSv4.1, so a minorversion-specific init_client callout method is introduced. CLID_INUSE recovery is important for the trunking discovery process. CLID_INUSE is a sign the server recognizes the client's nfs_client_id4 id string, but the client is using the wrong principal this time for the SETCLIENTID operation. The SETCLIENTID must be retried with a series of different principals until one works, and then the rest of trunking discovery can proceed. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2012-09-14 21:24:32 +00:00
int (*detect_trunking)(struct nfs_client *, struct nfs_client **,
struct rpc_cred *);
};
struct nfs4_state_maintenance_ops {
int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *, unsigned);
struct rpc_cred * (*get_state_renewal_cred_locked)(struct nfs_client *);
int (*renew_lease)(struct nfs_client *, struct rpc_cred *);
};
extern const struct dentry_operations nfs4_dentry_operations;
/* dir.c */
int nfs_atomic_open(struct inode *, struct dentry *, struct file *,
unsigned, umode_t, int *);
/* super.c */
extern struct file_system_type nfs4_fs_type;
/* nfs4namespace.c */
rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *);
struct rpc_clnt *nfs4_create_sec_client(struct rpc_clnt *, struct inode *, struct qstr *);
struct vfsmount *nfs4_submount(struct nfs_server *, struct dentry *,
struct nfs_fh *, struct nfs_fattr *);
/* nfs4proc.c */
extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *, struct nfs4_setclientid_res *);
extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct nfs4_setclientid_res *arg, struct rpc_cred *);
extern int nfs4_proc_get_rootfh(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
extern int nfs4_proc_bind_conn_to_session(struct nfs_client *, struct rpc_cred *cred);
extern int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred);
extern int nfs4_destroy_clientid(struct nfs_client *clp);
extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *);
extern int nfs41_init_clientid(struct nfs_client *, struct rpc_cred *);
extern int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait);
extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle);
extern int nfs4_proc_fs_locations(struct rpc_clnt *, struct inode *, const struct qstr *,
struct nfs4_fs_locations *, struct page *);
extern struct rpc_clnt *nfs4_proc_lookup_mountpoint(struct inode *, struct qstr *,
struct nfs_fh *, struct nfs_fattr *);
extern int nfs4_proc_secinfo(struct inode *, const struct qstr *, struct nfs4_secinfo_flavors *);
extern int nfs4_release_lockowner(struct nfs4_lock_state *);
extern const struct xattr_handler *nfs4_xattr_handlers[];
nfs41: sessions client infrastructure 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> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2009-04-01 13:21:53 +00:00
#if defined(CONFIG_NFS_V4_1)
static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server)
{
return server->nfs_client->cl_session;
}
extern bool nfs4_set_task_privileged(struct rpc_task *task, void *dummy);
extern int nfs4_setup_sequence(const struct nfs_server *server,
nfs41: setup_sequence method 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> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2009-04-01 13:22:15 +00:00
struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
struct rpc_task *task);
extern int nfs41_setup_sequence(struct nfs4_session *session,
struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
struct rpc_task *task);
nfs41: sessions client infrastructure 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> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2009-04-01 13:21:53 +00:00
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 *, struct rpc_cred *);
extern int nfs4_proc_destroy_session(struct nfs4_session *, struct rpc_cred *);
extern int nfs4_init_session(struct nfs_server *server);
extern int nfs4_proc_get_lease_time(struct nfs_client *clp,
struct nfs_fsinfo *fsinfo);
extern int nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data,
bool sync);
extern void nfs41_set_target_slotid(struct nfs4_slot_table *tbl,
u32 target_highest_slotid);
static inline bool
is_ds_only_client(struct nfs_client *clp)
{
return (clp->cl_exchange_flags & EXCHGID4_FLAG_MASK_PNFS) ==
EXCHGID4_FLAG_USE_PNFS_DS;
}
static inline bool
is_ds_client(struct nfs_client *clp)
{
return clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_DS;
}
nfs41: setup_sequence method 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> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2009-04-01 13:22:15 +00:00
#else /* CONFIG_NFS_v4_1 */
static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *server)
{
return NULL;
}
static inline int nfs4_setup_sequence(const struct nfs_server *server,
nfs41: setup_sequence method 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> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2009-04-01 13:22:15 +00:00
struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
struct rpc_task *task)
nfs41: setup_sequence method 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> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2009-04-01 13:22:15 +00:00
{
return 0;
}
static inline int nfs4_init_session(struct nfs_server *server)
{
return 0;
}
static inline bool
is_ds_only_client(struct nfs_client *clp)
{
return false;
}
static inline bool
is_ds_client(struct nfs_client *clp)
{
return false;
}
nfs41: sessions client infrastructure 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> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2009-04-01 13:21:53 +00:00
#endif /* CONFIG_NFS_V4_1 */
extern const struct nfs4_minor_version_ops *nfs_v4_minor_ops[];
extern const u32 nfs4_fattr_bitmap[3];
extern const u32 nfs4_statfs_bitmap[2];
extern const u32 nfs4_pathconf_bitmap[2];
extern const u32 nfs4_fsinfo_bitmap[3];
extern const u32 nfs4_fs_locations_bitmap[2];
void nfs4_free_client(struct nfs_client *);
struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *);
/* nfs4renewd.c */
extern void nfs4_schedule_state_renewal(struct nfs_client *);
extern void nfs4_renewd_prepare_shutdown(struct nfs_server *);
extern void nfs4_kill_renewd(struct nfs_client *);
2006-11-22 14:55:48 +00:00
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);
NFS: Discover NFSv4 server trunking when mounting "Server trunking" is a fancy named for a multi-homed NFS server. Trunking might occur if a client sends NFS requests for a single workload to multiple network interfaces on the same server. There are some implications for NFSv4 state management that make it useful for a client to know if a single NFSv4 server instance is multi-homed. (Note this is only a consideration for NFSv4, not for legacy versions of NFS, which are stateless). If a client cares about server trunking, no NFSv4 operations can proceed until that client determines who it is talking to. Thus server IP trunking discovery must be done when the client first encounters an unfamiliar server IP address. The nfs_get_client() function walks the nfs_client_list and matches on server IP address. The outcome of that walk tells us immediately if we have an unfamiliar server IP address. It invokes nfs_init_client() in this case. Thus, nfs4_init_client() is a good spot to perform trunking discovery. Discovery requires a client to establish a fresh client ID, so our client will now send SETCLIENTID or EXCHANGE_ID as the first NFS operation after a successful ping, rather than waiting for an application to perform an operation that requires NFSv4 state. The exact process for detecting trunking is different for NFSv4.0 and NFSv4.1, so a minorversion-specific init_client callout method is introduced. CLID_INUSE recovery is important for the trunking discovery process. CLID_INUSE is a sign the server recognizes the client's nfs_client_id4 id string, but the client is using the wrong principal this time for the SETCLIENTID operation. The SETCLIENTID must be retried with a series of different principals until one works, and then the rest of trunking discovery can proceed. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2012-09-14 21:24:32 +00:00
int nfs4_discover_server_trunking(struct nfs_client *clp,
struct nfs_client **);
int nfs40_discover_server_trunking(struct nfs_client *clp,
struct nfs_client **, struct rpc_cred *);
#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);
NFS: Discover NFSv4 server trunking when mounting "Server trunking" is a fancy named for a multi-homed NFS server. Trunking might occur if a client sends NFS requests for a single workload to multiple network interfaces on the same server. There are some implications for NFSv4 state management that make it useful for a client to know if a single NFSv4 server instance is multi-homed. (Note this is only a consideration for NFSv4, not for legacy versions of NFS, which are stateless). If a client cares about server trunking, no NFSv4 operations can proceed until that client determines who it is talking to. Thus server IP trunking discovery must be done when the client first encounters an unfamiliar server IP address. The nfs_get_client() function walks the nfs_client_list and matches on server IP address. The outcome of that walk tells us immediately if we have an unfamiliar server IP address. It invokes nfs_init_client() in this case. Thus, nfs4_init_client() is a good spot to perform trunking discovery. Discovery requires a client to establish a fresh client ID, so our client will now send SETCLIENTID or EXCHANGE_ID as the first NFS operation after a successful ping, rather than waiting for an application to perform an operation that requires NFSv4 state. The exact process for detecting trunking is different for NFSv4.0 and NFSv4.1, so a minorversion-specific init_client callout method is introduced. CLID_INUSE recovery is important for the trunking discovery process. CLID_INUSE is a sign the server recognizes the client's nfs_client_id4 id string, but the client is using the wrong principal this time for the SETCLIENTID operation. The SETCLIENTID must be retried with a series of different principals until one works, and then the rest of trunking discovery can proceed. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2012-09-14 21:24:32 +00:00
int nfs41_discover_server_trunking(struct nfs_client *clp,
struct nfs_client **, struct rpc_cred *);
extern void nfs4_schedule_session_recovery(struct nfs4_session *, int);
extern void nfs41_server_notify_target_slotid_update(struct nfs_client *clp);
extern void nfs4_session_drain_complete(struct nfs4_session *session,
struct nfs4_slot_table *tbl);
static inline bool nfs4_session_draining(struct nfs4_session *session)
{
return !!test_bit(NFS4_SESSION_DRAINING, &session->session_state);
}
#else
static inline void nfs4_schedule_session_recovery(struct nfs4_session *session, int err)
{
}
#endif /* CONFIG_NFS_V4_1 */
extern struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *, gfp_t);
extern void nfs4_put_state_owner(struct nfs4_state_owner *);
NFS: Cache state owners after files are closed Servers have a finite amount of memory to store NFSv4 open and lock owners. Moreover, servers may have a difficult time determining when they can reap their state owner table, thanks to gray areas in the NFSv4 protocol specification. Thus clients should be careful to reuse state owners when possible. Currently Linux is not too careful. When a user has closed all her files on one mount point, the state owner's reference count goes to zero, and it is released. The next OPEN allocates a new one. A workload that serially opens and closes files can run through a large number of open owners this way. When a state owner's reference count goes to zero, slap it onto a free list for that nfs_server, with an expiry time. Garbage collect before looking for a state owner. This makes state owners for active users available for re-use. Now that there can be unused state owners remaining at umount time, purge the state owner free list when a server is destroyed. Also be sure not to reclaim unused state owners during state recovery. This change has benefits for the client as well. For some workloads, this approach drops the number of OPEN_CONFIRM calls from the same as the number of OPEN calls, down to just one. This reduces wire traffic and thus open(2) latency. Before this patch, untarring a kernel source tarball shows the OPEN_CONFIRM call counter steadily increasing through the test. With the patch, the OPEN_CONFIRM count remains at 1 throughout the entire untar. As long as the expiry time is kept short, I don't think garbage collection should be terribly expensive, although it does bounce the clp->cl_lock around a bit. [ At some point we should rationalize the use of the nfs_server ->destroy method. ] Signed-off-by: Chuck Lever <chuck.lever@oracle.com> [Trond: Fixed a garbage collection race and a few efficiency issues] Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2011-12-06 21:13:48 +00:00
extern void nfs4_purge_state_owners(struct nfs_server *);
extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *);
extern void nfs4_put_open_state(struct nfs4_state *);
extern void nfs4_close_state(struct nfs4_state *, fmode_t);
extern void nfs4_close_sync(struct nfs4_state *, fmode_t);
extern void nfs4_state_set_mode_locked(struct nfs4_state *, fmode_t);
extern void nfs_inode_find_state_and_recover(struct inode *inode,
const nfs4_stateid *stateid);
extern void nfs4_schedule_lease_recovery(struct nfs_client *);
extern void nfs4_schedule_state_manager(struct nfs_client *);
extern void nfs4_schedule_path_down_recovery(struct nfs_client *clp);
extern void nfs4_schedule_stateid_recovery(const struct nfs_server *, struct nfs4_state *);
extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags);
extern void nfs41_handle_server_scope(struct nfs_client *,
struct nfs41_server_scope **);
extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
extern void nfs4_select_rw_stateid(nfs4_stateid *, struct nfs4_state *,
fmode_t, const struct nfs_lockowner *);
extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter, gfp_t gfp_mask);
extern int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task);
extern void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid);
extern void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid);
extern void nfs_release_seqid(struct nfs_seqid *seqid);
extern void nfs_free_seqid(struct nfs_seqid *seqid);
extern void nfs4_free_lock_state(struct nfs_server *server, struct nfs4_lock_state *lsp);
extern const nfs4_stateid zero_stateid;
/* nfs4super.c */
struct nfs_mount_info;
extern struct nfs_subversion nfs_v4;
struct dentry *nfs4_try_mount(int, const char *, struct nfs_mount_info *, struct nfs_subversion *);
extern bool nfs4_disable_idmapping;
extern unsigned short max_session_slots;
extern unsigned short send_implementation_id;
#define NFS4_CLIENT_ID_UNIQ_LEN (64)
extern char nfs4_client_id_uniquifier[NFS4_CLIENT_ID_UNIQ_LEN];
/* nfs4sysctl.c */
#ifdef CONFIG_SYSCTL
int nfs4_register_sysctl(void);
void nfs4_unregister_sysctl(void);
#else
static inline int nfs4_register_sysctl(void)
{
return 0;
}
static inline void nfs4_unregister_sysctl(void)
{
}
#endif
/* nfs4xdr.c */
extern struct rpc_procinfo nfs4_procedures[];
struct nfs4_mount_data;
/* callback_xdr.c */
extern struct svc_version nfs4_callback_version1;
extern struct svc_version nfs4_callback_version4;
static inline void nfs4_stateid_copy(nfs4_stateid *dst, const nfs4_stateid *src)
{
memcpy(dst, src, sizeof(*dst));
}
static inline bool nfs4_stateid_match(const nfs4_stateid *dst, const nfs4_stateid *src)
{
return memcmp(dst, src, sizeof(*dst)) == 0;
}
#else
#define nfs4_close_state(a, b) do { } while (0)
#define nfs4_close_sync(a, b) do { } while (0)
#endif /* CONFIG_NFS_V4 */
#endif /* __LINUX_FS_NFS_NFS4_FS.H */