/* * Copyright (c) 2001 The Regents of the University of Michigan. * All rights reserved. * * Kendrick Smith * Andy Adamson * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. Neither the name of the University nor the names of its * contributors may be used to endorse or promote products derived * from this software without specific prior written permission. * * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * */ #include #include #include #include #include #include #include #include #include "xdr4.h" #include "vfs.h" #include "current_stateid.h" #define NFSDDBG_FACILITY NFSDDBG_PROC /* Globals */ time_t nfsd4_lease = 90; /* default lease time */ time_t nfsd4_grace = 90; static time_t boot_time; #define all_ones {{~0,~0},~0} static const stateid_t one_stateid = { .si_generation = ~0, .si_opaque = all_ones, }; static const stateid_t zero_stateid = { /* all fields zero */ }; static const stateid_t currentstateid = { .si_generation = 1, }; static u64 current_sessionid = 1; #define ZERO_STATEID(stateid) (!memcmp((stateid), &zero_stateid, sizeof(stateid_t))) #define ONE_STATEID(stateid) (!memcmp((stateid), &one_stateid, sizeof(stateid_t))) #define CURRENT_STATEID(stateid) (!memcmp((stateid), ¤tstateid, sizeof(stateid_t))) /* forward declarations */ static int check_for_locks(struct nfs4_file *filp, struct nfs4_lockowner *lowner); /* Locking: */ /* Currently used for almost all code touching nfsv4 state: */ static DEFINE_MUTEX(client_mutex); /* * Currently used for the del_recall_lru and file hash table. In an * effort to decrease the scope of the client_mutex, this spinlock may * eventually cover more: */ static DEFINE_SPINLOCK(recall_lock); static struct kmem_cache *openowner_slab = NULL; static struct kmem_cache *lockowner_slab = NULL; static struct kmem_cache *file_slab = NULL; static struct kmem_cache *stateid_slab = NULL; static struct kmem_cache *deleg_slab = NULL; void nfs4_lock_state(void) { mutex_lock(&client_mutex); } static void free_session(struct kref *); /* Must be called under the client_lock */ static void nfsd4_put_session_locked(struct nfsd4_session *ses) { kref_put(&ses->se_ref, free_session); } static void nfsd4_get_session(struct nfsd4_session *ses) { kref_get(&ses->se_ref); } void nfs4_unlock_state(void) { mutex_unlock(&client_mutex); } static inline u32 opaque_hashval(const void *ptr, int nbytes) { unsigned char *cptr = (unsigned char *) ptr; u32 x = 0; while (nbytes--) { x *= 37; x += *cptr++; } return x; } static struct list_head del_recall_lru; static void nfsd4_free_file(struct nfs4_file *f) { kmem_cache_free(file_slab, f); } static inline void put_nfs4_file(struct nfs4_file *fi) { if (atomic_dec_and_lock(&fi->fi_ref, &recall_lock)) { list_del(&fi->fi_hash); spin_unlock(&recall_lock); iput(fi->fi_inode); nfsd4_free_file(fi); } } static inline void get_nfs4_file(struct nfs4_file *fi) { atomic_inc(&fi->fi_ref); } static int num_delegations; unsigned int max_delegations; /* * Open owner state (share locks) */ /* hash tables for lock and open owners */ #define OWNER_HASH_BITS 8 #define OWNER_HASH_SIZE (1 << OWNER_HASH_BITS) #define OWNER_HASH_MASK (OWNER_HASH_SIZE - 1) static unsigned int ownerstr_hashval(u32 clientid, struct xdr_netobj *ownername) { unsigned int ret; ret = opaque_hashval(ownername->data, ownername->len); ret += clientid; return ret & OWNER_HASH_MASK; } static struct list_head ownerstr_hashtbl[OWNER_HASH_SIZE]; /* hash table for nfs4_file */ #define FILE_HASH_BITS 8 #define FILE_HASH_SIZE (1 << FILE_HASH_BITS) static unsigned int file_hashval(struct inode *ino) { /* XXX: why are we hashing on inode pointer, anyway? */ return hash_ptr(ino, FILE_HASH_BITS); } static struct list_head file_hashtbl[FILE_HASH_SIZE]; static void __nfs4_file_get_access(struct nfs4_file *fp, int oflag) { BUG_ON(!(fp->fi_fds[oflag] || fp->fi_fds[O_RDWR])); atomic_inc(&fp->fi_access[oflag]); } static void nfs4_file_get_access(struct nfs4_file *fp, int oflag) { if (oflag == O_RDWR) { __nfs4_file_get_access(fp, O_RDONLY); __nfs4_file_get_access(fp, O_WRONLY); } else __nfs4_file_get_access(fp, oflag); } static void nfs4_file_put_fd(struct nfs4_file *fp, int oflag) { if (fp->fi_fds[oflag]) { fput(fp->fi_fds[oflag]); fp->fi_fds[oflag] = NULL; } } static void __nfs4_file_put_access(struct nfs4_file *fp, int oflag) { if (atomic_dec_and_test(&fp->fi_access[oflag])) { nfs4_file_put_fd(fp, oflag); /* * It's also safe to get rid of the RDWR open *if* * we no longer have need of the other kind of access * or if we already have the other kind of open: */ if (fp->fi_fds[1-oflag] || atomic_read(&fp->fi_access[1 - oflag]) == 0) nfs4_file_put_fd(fp, O_RDWR); } } static void nfs4_file_put_access(struct nfs4_file *fp, int oflag) { if (oflag == O_RDWR) { __nfs4_file_put_access(fp, O_RDONLY); __nfs4_file_put_access(fp, O_WRONLY); } else __nfs4_file_put_access(fp, oflag); } static inline int get_new_stid(struct nfs4_stid *stid) { static int min_stateid = 0; struct idr *stateids = &stid->sc_client->cl_stateids; int new_stid; int error; error = idr_get_new_above(stateids, stid, min_stateid, &new_stid); /* * Note: the necessary preallocation was done in * nfs4_alloc_stateid(). The idr code caps the number of * preallocations that can exist at a time, but the state lock * prevents anyone from using ours before we get here: */ BUG_ON(error); /* * It shouldn't be a problem to reuse an opaque stateid value. * I don't think it is for 4.1. But with 4.0 I worry that, for * example, a stray write retransmission could be accepted by * the server when it should have been rejected. Therefore, * adopt a trick from the sctp code to attempt to maximize the * amount of time until an id is reused, by ensuring they always * "increase" (mod INT_MAX): */ min_stateid = new_stid+1; if (min_stateid == INT_MAX) min_stateid = 0; return new_stid; } static void init_stid(struct nfs4_stid *stid, struct nfs4_client *cl, unsigned char type) { stateid_t *s = &stid->sc_stateid; int new_id; stid->sc_type = type; stid->sc_client = cl; s->si_opaque.so_clid = cl->cl_clientid; new_id = get_new_stid(stid); s->si_opaque.so_id = (u32)new_id; /* Will be incremented before return to client: */ s->si_generation = 0; } static struct nfs4_stid *nfs4_alloc_stid(struct nfs4_client *cl, struct kmem_cache *slab) { struct idr *stateids = &cl->cl_stateids; if (!idr_pre_get(stateids, GFP_KERNEL)) return NULL; /* * Note: if we fail here (or any time between now and the time * we actually get the new idr), we won't need to undo the idr * preallocation, since the idr code caps the number of * preallocated entries. */ return kmem_cache_alloc(slab, GFP_KERNEL); } static struct nfs4_ol_stateid * nfs4_alloc_stateid(struct nfs4_client *clp) { return openlockstateid(nfs4_alloc_stid(clp, stateid_slab)); } static struct nfs4_delegation * alloc_init_deleg(struct nfs4_client *clp, struct nfs4_ol_stateid *stp, struct svc_fh *current_fh, u32 type) { struct nfs4_delegation *dp; struct nfs4_file *fp = stp->st_file; dprintk("NFSD alloc_init_deleg\n"); /* * Major work on the lease subsystem (for example, to support * calbacks on stat) will be required before we can support * write delegations properly. */ if (type != NFS4_OPEN_DELEGATE_READ) return NULL; if (fp->fi_had_conflict) return NULL; if (num_delegations > max_delegations) return NULL; dp = delegstateid(nfs4_alloc_stid(clp, deleg_slab)); if (dp == NULL) return dp; init_stid(&dp->dl_stid, clp, NFS4_DELEG_STID); /* * delegation seqid's are never incremented. The 4.1 special * meaning of seqid 0 isn't meaningful, really, but let's avoid * 0 anyway just for consistency and use 1: */ dp->dl_stid.sc_stateid.si_generation = 1; num_delegations++; INIT_LIST_HEAD(&dp->dl_perfile); INIT_LIST_HEAD(&dp->dl_perclnt); INIT_LIST_HEAD(&dp->dl_recall_lru); get_nfs4_file(fp); dp->dl_file = fp; dp->dl_type = type; fh_copy_shallow(&dp->dl_fh, ¤t_fh->fh_handle); dp->dl_time = 0; atomic_set(&dp->dl_count, 1); INIT_WORK(&dp->dl_recall.cb_work, nfsd4_do_callback_rpc); return dp; } void nfs4_put_delegation(struct nfs4_delegation *dp) { if (atomic_dec_and_test(&dp->dl_count)) { dprintk("NFSD: freeing dp %p\n",dp); put_nfs4_file(dp->dl_file); kmem_cache_free(deleg_slab, dp); num_delegations--; } } static void nfs4_put_deleg_lease(struct nfs4_file *fp) { if (atomic_dec_and_test(&fp->fi_delegees)) { vfs_setlease(fp->fi_deleg_file, F_UNLCK, &fp->fi_lease); fp->fi_lease = NULL; fput(fp->fi_deleg_file); fp->fi_deleg_file = NULL; } } static void unhash_stid(struct nfs4_stid *s) { struct idr *stateids = &s->sc_client->cl_stateids; idr_remove(stateids, s->sc_stateid.si_opaque.so_id); } /* Called under the state lock. */ static void unhash_delegation(struct nfs4_delegation *dp) { unhash_stid(&dp->dl_stid); list_del_init(&dp->dl_perclnt); spin_lock(&recall_lock); list_del_init(&dp->dl_perfile); list_del_init(&dp->dl_recall_lru); spin_unlock(&recall_lock); nfs4_put_deleg_lease(dp->dl_file); nfs4_put_delegation(dp); } /* * SETCLIENTID state */ /* client_lock protects the client lru list and session hash table */ static DEFINE_SPINLOCK(client_lock); /* Hash tables for nfs4_clientid state */ #define CLIENT_HASH_BITS 4 #define CLIENT_HASH_SIZE (1 << CLIENT_HASH_BITS) #define CLIENT_HASH_MASK (CLIENT_HASH_SIZE - 1) static unsigned int clientid_hashval(u32 id) { return id & CLIENT_HASH_MASK; } static unsigned int clientstr_hashval(const char *name) { return opaque_hashval(name, 8) & CLIENT_HASH_MASK; } /* * reclaim_str_hashtbl[] holds known client info from previous reset/reboot * used in reboot/reset lease grace period processing * * conf_id_hashtbl[], and conf_str_hashtbl[] hold confirmed * setclientid_confirmed info. * * unconf_str_hastbl[] and unconf_id_hashtbl[] hold unconfirmed * setclientid info. * * client_lru holds client queue ordered by nfs4_client.cl_time * for lease renewal. * * close_lru holds (open) stateowner queue ordered by nfs4_stateowner.so_time * for last close replay. */ static struct list_head reclaim_str_hashtbl[CLIENT_HASH_SIZE]; static int reclaim_str_hashtbl_size = 0; static struct list_head conf_id_hashtbl[CLIENT_HASH_SIZE]; static struct list_head conf_str_hashtbl[CLIENT_HASH_SIZE]; static struct list_head unconf_str_hashtbl[CLIENT_HASH_SIZE]; static struct list_head unconf_id_hashtbl[CLIENT_HASH_SIZE]; static struct list_head client_lru; static struct list_head close_lru; /* * We store the NONE, READ, WRITE, and BOTH bits separately in the * st_{access,deny}_bmap field of the stateid, in order to track not * only what share bits are currently in force, but also what * combinations of share bits previous opens have used. This allows us * to enforce the recommendation of rfc 3530 14.2.19 that the server * return an error if the client attempt to downgrade to a combination * of share bits not explicable by closing some of its previous opens. * * XXX: This enforcement is actually incomplete, since we don't keep * track of access/deny bit combinations; so, e.g., we allow: * * OPEN allow read, deny write * OPEN allow both, deny none * DOWNGRADE allow read, deny none * * which we should reject. */ static unsigned int bmap_to_share_mode(unsigned long bmap) { int i; unsigned int access = 0; for (i = 1; i < 4; i++) { if (test_bit(i, &bmap)) access |= i; } return access; } static bool test_share(struct nfs4_ol_stateid *stp, struct nfsd4_open *open) { unsigned int access, deny; access = bmap_to_share_mode(stp->st_access_bmap); deny = bmap_to_share_mode(stp->st_deny_bmap); if ((access & open->op_share_deny) || (deny & open->op_share_access)) return false; return true; } /* set share access for a given stateid */ static inline void set_access(u32 access, struct nfs4_ol_stateid *stp) { __set_bit(access, &stp->st_access_bmap); } /* clear share access for a given stateid */ static inline void clear_access(u32 access, struct nfs4_ol_stateid *stp) { __clear_bit(access, &stp->st_access_bmap); } /* test whether a given stateid has access */ static inline bool test_access(u32 access, struct nfs4_ol_stateid *stp) { return test_bit(access, &stp->st_access_bmap); } /* set share deny for a given stateid */ static inline void set_deny(u32 access, struct nfs4_ol_stateid *stp) { __set_bit(access, &stp->st_deny_bmap); } /* clear share deny for a given stateid */ static inline void clear_deny(u32 access, struct nfs4_ol_stateid *stp) { __clear_bit(access, &stp->st_deny_bmap); } /* test whether a given stateid is denying specific access */ static inline bool test_deny(u32 access, struct nfs4_ol_stateid *stp) { return test_bit(access, &stp->st_deny_bmap); } static int nfs4_access_to_omode(u32 access) { switch (access & NFS4_SHARE_ACCESS_BOTH) { case NFS4_SHARE_ACCESS_READ: return O_RDONLY; case NFS4_SHARE_ACCESS_WRITE: return O_WRONLY; case NFS4_SHARE_ACCESS_BOTH: return O_RDWR; } BUG(); } /* release all access and file references for a given stateid */ static void release_all_access(struct nfs4_ol_stateid *stp) { int i; for (i = 1; i < 4; i++) { if (test_access(i, stp)) nfs4_file_put_access(stp->st_file, nfs4_access_to_omode(i)); clear_access(i, stp); } } static void unhash_generic_stateid(struct nfs4_ol_stateid *stp) { list_del(&stp->st_perfile); list_del(&stp->st_perstateowner); } static void close_generic_stateid(struct nfs4_ol_stateid *stp) { release_all_access(stp); put_nfs4_file(stp->st_file); stp->st_file = NULL; } static void free_generic_stateid(struct nfs4_ol_stateid *stp) { kmem_cache_free(stateid_slab, stp); } static void release_lock_stateid(struct nfs4_ol_stateid *stp) { struct file *file; unhash_generic_stateid(stp); unhash_stid(&stp->st_stid); file = find_any_file(stp->st_file); if (file) locks_remove_posix(file, (fl_owner_t)lockowner(stp->st_stateowner)); close_generic_stateid(stp); free_generic_stateid(stp); } static void unhash_lockowner(struct nfs4_lockowner *lo) { struct nfs4_ol_stateid *stp; list_del(&lo->lo_owner.so_strhash); list_del(&lo->lo_perstateid); list_del(&lo->lo_owner_ino_hash); while (!list_empty(&lo->lo_owner.so_stateids)) { stp = list_first_entry(&lo->lo_owner.so_stateids, struct nfs4_ol_stateid, st_perstateowner); release_lock_stateid(stp); } } static void release_lockowner(struct nfs4_lockowner *lo) { unhash_lockowner(lo); nfs4_free_lockowner(lo); } static void release_stateid_lockowners(struct nfs4_ol_stateid *open_stp) { struct nfs4_lockowner *lo; while (!list_empty(&open_stp->st_lockowners)) { lo = list_entry(open_stp->st_lockowners.next, struct nfs4_lockowner, lo_perstateid); release_lockowner(lo); } } static void unhash_open_stateid(struct nfs4_ol_stateid *stp) { unhash_generic_stateid(stp); release_stateid_lockowners(stp); close_generic_stateid(stp); } static void release_open_stateid(struct nfs4_ol_stateid *stp) { unhash_open_stateid(stp); unhash_stid(&stp->st_stid); free_generic_stateid(stp); } static void unhash_openowner(struct nfs4_openowner *oo) { struct nfs4_ol_stateid *stp; list_del(&oo->oo_owner.so_strhash); list_del(&oo->oo_perclient); while (!list_empty(&oo->oo_owner.so_stateids)) { stp = list_first_entry(&oo->oo_owner.so_stateids, struct nfs4_ol_stateid, st_perstateowner); release_open_stateid(stp); } } static void release_last_closed_stateid(struct nfs4_openowner *oo) { struct nfs4_ol_stateid *s = oo->oo_last_closed_stid; if (s) { unhash_stid(&s->st_stid); free_generic_stateid(s); oo->oo_last_closed_stid = NULL; } } static void release_openowner(struct nfs4_openowner *oo) { unhash_openowner(oo); list_del(&oo->oo_close_lru); release_last_closed_stateid(oo); nfs4_free_openowner(oo); } #define SESSION_HASH_SIZE 512 static struct list_head sessionid_hashtbl[SESSION_HASH_SIZE]; static inline int hash_sessionid(struct nfs4_sessionid *sessionid) { struct nfsd4_sessionid *sid = (struct nfsd4_sessionid *)sessionid; return sid->sequence % SESSION_HASH_SIZE; } #ifdef NFSD_DEBUG static inline void dump_sessionid(const char *fn, struct nfs4_sessionid *sessionid) { u32 *ptr = (u32 *)(&sessionid->data[0]); dprintk("%s: %u:%u:%u:%u\n", fn, ptr[0], ptr[1], ptr[2], ptr[3]); } #else static inline void dump_sessionid(const char *fn, struct nfs4_sessionid *sessionid) { } #endif static void gen_sessionid(struct nfsd4_session *ses) { struct nfs4_client *clp = ses->se_client; struct nfsd4_sessionid *sid; sid = (struct nfsd4_sessionid *)ses->se_sessionid.data; sid->clientid = clp->cl_clientid; sid->sequence = current_sessionid++; sid->reserved = 0; } /* * The protocol defines ca_maxresponssize_cached to include the size of * the rpc header, but all we need to cache is the data starting after * the end of the initial SEQUENCE operation--the rest we regenerate * each time. Therefore we can advertise a ca_maxresponssize_cached * value that is the number of bytes in our cache plus a few additional * bytes. In order to stay on the safe side, and not promise more than * we can cache, those additional bytes must be the minimum possible: 24 * bytes of rpc header (xid through accept state, with AUTH_NULL * verifier), 12 for the compound header (with zero-length tag), and 44 * for the SEQUENCE op response: */ #define NFSD_MIN_HDR_SEQ_SZ (24 + 12 + 44) static void free_session_slots(struct nfsd4_session *ses) { int i; for (i = 0; i < ses->se_fchannel.maxreqs; i++) kfree(ses->se_slots[i]); } /* * We don't actually need to cache the rpc and session headers, so we * can allocate a little less for each slot: */ static inline int slot_bytes(struct nfsd4_channel_attrs *ca) { return ca->maxresp_cached - NFSD_MIN_HDR_SEQ_SZ; } static int nfsd4_sanitize_slot_size(u32 size) { size -= NFSD_MIN_HDR_SEQ_SZ; /* We don't cache the rpc header */ size = min_t(u32, size, NFSD_SLOT_CACHE_SIZE); return size; } /* * XXX: If we run out of reserved DRC memory we could (up to a point) * re-negotiate active sessions and reduce their slot usage to make * room for new connections. For now we just fail the create session. */ static int nfsd4_get_drc_mem(int slotsize, u32 num) { int avail; num = min_t(u32, num, NFSD_MAX_SLOTS_PER_SESSION); spin_lock(&nfsd_drc_lock); avail = min_t(int, NFSD_MAX_MEM_PER_SESSION, nfsd_drc_max_mem - nfsd_drc_mem_used); num = min_t(int, num, avail / slotsize); nfsd_drc_mem_used += num * slotsize; spin_unlock(&nfsd_drc_lock); return num; } static void nfsd4_put_drc_mem(int slotsize, int num) { spin_lock(&nfsd_drc_lock); nfsd_drc_mem_used -= slotsize * num; spin_unlock(&nfsd_drc_lock); } static struct nfsd4_session *alloc_session(int slotsize, int numslots) { struct nfsd4_session *new; int mem, i; BUILD_BUG_ON(NFSD_MAX_SLOTS_PER_SESSION * sizeof(struct nfsd4_slot *) + sizeof(struct nfsd4_session) > PAGE_SIZE); mem = numslots * sizeof(struct nfsd4_slot *); new = kzalloc(sizeof(*new) + mem, GFP_KERNEL); if (!new) return NULL; /* allocate each struct nfsd4_slot and data cache in one piece */ for (i = 0; i < numslots; i++) { mem = sizeof(struct nfsd4_slot) + slotsize; new->se_slots[i] = kzalloc(mem, GFP_KERNEL); if (!new->se_slots[i]) goto out_free; } return new; out_free: while (i--) kfree(new->se_slots[i]); kfree(new); return NULL; } static void init_forechannel_attrs(struct nfsd4_channel_attrs *new, struct nfsd4_channel_attrs *req, int numslots, int slotsize) { u32 maxrpc = nfsd_serv->sv_max_mesg; new->maxreqs = numslots; new->maxresp_cached = min_t(u32, req->maxresp_cached, slotsize + NFSD_MIN_HDR_SEQ_SZ); new->maxreq_sz = min_t(u32, req->maxreq_sz, maxrpc); new->maxresp_sz = min_t(u32, req->maxresp_sz, maxrpc); new->maxops = min_t(u32, req->maxops, NFSD_MAX_OPS_PER_COMPOUND); } static void free_conn(struct nfsd4_conn *c) { svc_xprt_put(c->cn_xprt); kfree(c); } static void nfsd4_conn_lost(struct svc_xpt_user *u) { struct nfsd4_conn *c = container_of(u, struct nfsd4_conn, cn_xpt_user); struct nfs4_client *clp = c->cn_session->se_client; spin_lock(&clp->cl_lock); if (!list_empty(&c->cn_persession)) { list_del(&c->cn_persession); free_conn(c); } spin_unlock(&clp->cl_lock); nfsd4_probe_callback(clp); } static struct nfsd4_conn *alloc_conn(struct svc_rqst *rqstp, u32 flags) { struct nfsd4_conn *conn; conn = kmalloc(sizeof(struct nfsd4_conn), GFP_KERNEL); if (!conn) return NULL; svc_xprt_get(rqstp->rq_xprt); conn->cn_xprt = rqstp->rq_xprt; conn->cn_flags = flags; INIT_LIST_HEAD(&conn->cn_xpt_user.list); return conn; } static void __nfsd4_hash_conn(struct nfsd4_conn *conn, struct nfsd4_session *ses) { conn->cn_session = ses; list_add(&conn->cn_persession, &ses->se_conns); } static void nfsd4_hash_conn(struct nfsd4_conn *conn, struct nfsd4_session *ses) { struct nfs4_client *clp = ses->se_client; spin_lock(&clp->cl_lock); __nfsd4_hash_conn(conn, ses); spin_unlock(&clp->cl_lock); } static int nfsd4_register_conn(struct nfsd4_conn *conn) { conn->cn_xpt_user.callback = nfsd4_conn_lost; return register_xpt_user(conn->cn_xprt, &conn->cn_xpt_user); } static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses, u32 dir) { struct nfsd4_conn *conn; int ret; conn = alloc_conn(rqstp, dir); if (!conn) return nfserr_jukebox; nfsd4_hash_conn(conn, ses); ret = nfsd4_register_conn(conn); if (ret) /* oops; xprt is already down: */ nfsd4_conn_lost(&conn->cn_xpt_user); return nfs_ok; } static __be32 nfsd4_new_conn_from_crses(struct svc_rqst *rqstp, struct nfsd4_session *ses) { u32 dir = NFS4_CDFC4_FORE; if (ses->se_flags & SESSION4_BACK_CHAN) dir |= NFS4_CDFC4_BACK; return nfsd4_new_conn(rqstp, ses, dir); } /* must be called under client_lock */ static void nfsd4_del_conns(struct nfsd4_session *s) { struct nfs4_client *clp = s->se_client; struct nfsd4_conn *c; spin_lock(&clp->cl_lock); while (!list_empty(&s->se_conns)) { c = list_first_entry(&s->se_conns, struct nfsd4_conn, cn_persession); list_del_init(&c->cn_persession); spin_unlock(&clp->cl_lock); unregister_xpt_user(c->cn_xprt, &c->cn_xpt_user); free_conn(c); spin_lock(&clp->cl_lock); } spin_unlock(&clp->cl_lock); } static void free_session(struct kref *kref) { struct nfsd4_session *ses; int mem; BUG_ON(!spin_is_locked(&client_lock)); ses = container_of(kref, struct nfsd4_session, se_ref); nfsd4_del_conns(ses); spin_lock(&nfsd_drc_lock); mem = ses->se_fchannel.maxreqs * slot_bytes(&ses->se_fchannel); nfsd_drc_mem_used -= mem; spin_unlock(&nfsd_drc_lock); free_session_slots(ses); kfree(ses); } void nfsd4_put_session(struct nfsd4_session *ses) { spin_lock(&client_lock); nfsd4_put_session_locked(ses); spin_unlock(&client_lock); } static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct nfs4_client *clp, struct nfsd4_create_session *cses) { struct nfsd4_session *new; struct nfsd4_channel_attrs *fchan = &cses->fore_channel; int numslots, slotsize; __be32 status; int idx; /* * Note decreasing slot size below client's request may * make it difficult for client to function correctly, whereas * decreasing the number of slots will (just?) affect * performance. When short on memory we therefore prefer to * decrease number of slots instead of their size. */ slotsize = nfsd4_sanitize_slot_size(fchan->maxresp_cached); numslots = nfsd4_get_drc_mem(slotsize, fchan->maxreqs); if (numslots < 1) return NULL; new = alloc_session(slotsize, numslots); if (!new) { nfsd4_put_drc_mem(slotsize, fchan->maxreqs); return NULL; } init_forechannel_attrs(&new->se_fchannel, fchan, numslots, slotsize); new->se_client = clp; gen_sessionid(new); INIT_LIST_HEAD(&new->se_conns); new->se_cb_seq_nr = 1; new->se_flags = cses->flags; new->se_cb_prog = cses->callback_prog; kref_init(&new->se_ref); idx = hash_sessionid(&new->se_sessionid); spin_lock(&client_lock); list_add(&new->se_hash, &sessionid_hashtbl[idx]); spin_lock(&clp->cl_lock); list_add(&new->se_perclnt, &clp->cl_sessions); spin_unlock(&clp->cl_lock); spin_unlock(&client_lock); status = nfsd4_new_conn_from_crses(rqstp, new); /* whoops: benny points out, status is ignored! (err, or bogus) */ if (status) { spin_lock(&client_lock); free_session(&new->se_ref); spin_unlock(&client_lock); return NULL; } if (cses->flags & SESSION4_BACK_CHAN) { struct sockaddr *sa = svc_addr(rqstp); /* * This is a little silly; with sessions there's no real * use for the callback address. Use the peer address * as a reasonable default for now, but consider fixing * the rpc client not to require an address in the * future: */ rpc_copy_addr((struct sockaddr *)&clp->cl_cb_conn.cb_addr, sa); clp->cl_cb_conn.cb_addrlen = svc_addr_len(sa); } nfsd4_probe_callback(clp); return new; } /* caller must hold client_lock */ static struct nfsd4_session * find_in_sessionid_hashtbl(struct nfs4_sessionid *sessionid) { struct nfsd4_session *elem; int idx; dump_sessionid(__func__, sessionid); idx = hash_sessionid(sessionid); /* Search in the appropriate list */ list_for_each_entry(elem, &sessionid_hashtbl[idx], se_hash) { if (!memcmp(elem->se_sessionid.data, sessionid->data, NFS4_MAX_SESSIONID_LEN)) { return elem; } } dprintk("%s: session not found\n", __func__); return NULL; } /* caller must hold client_lock */ static void unhash_session(struct nfsd4_session *ses) { list_del(&ses->se_hash); spin_lock(&ses->se_client->cl_lock); list_del(&ses->se_perclnt); spin_unlock(&ses->se_client->cl_lock); } /* must be called under the client_lock */ static inline void renew_client_locked(struct nfs4_client *clp) { if (is_client_expired(clp)) { WARN_ON(1); printk("%s: client (clientid %08x/%08x) already expired\n", __func__, clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id); return; } dprintk("renewing client (clientid %08x/%08x)\n", clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id); list_move_tail(&clp->cl_lru, &client_lru); clp->cl_time = get_seconds(); } static inline void renew_client(struct nfs4_client *clp) { spin_lock(&client_lock); renew_client_locked(clp); spin_unlock(&client_lock); } /* SETCLIENTID and SETCLIENTID_CONFIRM Helper functions */ static int STALE_CLIENTID(clientid_t *clid) { if (clid->cl_boot == boot_time) return 0; dprintk("NFSD stale clientid (%08x/%08x) boot_time %08lx\n", clid->cl_boot, clid->cl_id, boot_time); return 1; } /* * XXX Should we use a slab cache ? * This type of memory management is somewhat inefficient, but we use it * anyway since SETCLIENTID is not a common operation. */ static struct nfs4_client *alloc_client(struct xdr_netobj name) { struct nfs4_client *clp; clp = kzalloc(sizeof(struct nfs4_client), GFP_KERNEL); if (clp == NULL) return NULL; clp->cl_name.data = kmemdup(name.data, name.len, GFP_KERNEL); if (clp->cl_name.data == NULL) { kfree(clp); return NULL; } clp->cl_name.len = name.len; return clp; } static inline void free_client(struct nfs4_client *clp) { BUG_ON(!spin_is_locked(&client_lock)); while (!list_empty(&clp->cl_sessions)) { struct nfsd4_session *ses; ses = list_entry(clp->cl_sessions.next, struct nfsd4_session, se_perclnt); list_del(&ses->se_perclnt); nfsd4_put_session_locked(ses); } free_svc_cred(&clp->cl_cred); kfree(clp->cl_name.data); kfree(clp); } void release_session_client(struct nfsd4_session *session) { struct nfs4_client *clp = session->se_client; if (!atomic_dec_and_lock(&clp->cl_refcount, &client_lock)) return; if (is_client_expired(clp)) { free_client(clp); session->se_client = NULL; } else renew_client_locked(clp); spin_unlock(&client_lock); } /* must be called under the client_lock */ static inline void unhash_client_locked(struct nfs4_client *clp) { struct nfsd4_session *ses; mark_client_expired(clp); list_del(&clp->cl_lru); spin_lock(&clp->cl_lock); list_for_each_entry(ses, &clp->cl_sessions, se_perclnt) list_del_init(&ses->se_hash); spin_unlock(&clp->cl_lock); } static void expire_client(struct nfs4_client *clp) { struct nfs4_openowner *oo; struct nfs4_delegation *dp; struct list_head reaplist; INIT_LIST_HEAD(&reaplist); spin_lock(&recall_lock); while (!list_empty(&clp->cl_delegations)) { dp = list_entry(clp->cl_delegations.next, struct nfs4_delegation, dl_perclnt); list_del_init(&dp->dl_perclnt); list_move(&dp->dl_recall_lru, &reaplist); } spin_unlock(&recall_lock); while (!list_empty(&reaplist)) { dp = list_entry(reaplist.next, struct nfs4_delegation, dl_recall_lru); unhash_delegation(dp); } while (!list_empty(&clp->cl_openowners)) { oo = list_entry(clp->cl_openowners.next, struct nfs4_openowner, oo_perclient); release_openowner(oo); } nfsd4_shutdown_callback(clp); if (clp->cl_cb_conn.cb_xprt) svc_xprt_put(clp->cl_cb_conn.cb_xprt); list_del(&clp->cl_idhash); list_del(&clp->cl_strhash); spin_lock(&client_lock); unhash_client_locked(clp); if (atomic_read(&clp->cl_refcount) == 0) free_client(clp); spin_unlock(&client_lock); } static void copy_verf(struct nfs4_client *target, nfs4_verifier *source) { memcpy(target->cl_verifier.data, source->data, sizeof(target->cl_verifier.data)); } static void copy_clid(struct nfs4_client *target, struct nfs4_client *source) { target->cl_clientid.cl_boot = source->cl_clientid.cl_boot; target->cl_clientid.cl_id = source->cl_clientid.cl_id; } static int copy_cred(struct svc_cred *target, struct svc_cred *source) { if (source->cr_principal) { target->cr_principal = kstrdup(source->cr_principal, GFP_KERNEL); if (target->cr_principal == NULL) return -ENOMEM; } else target->cr_principal = NULL; target->cr_flavor = source->cr_flavor; target->cr_uid = source->cr_uid; target->cr_gid = source->cr_gid; target->cr_group_info = source->cr_group_info; get_group_info(target->cr_group_info); return 0; } static int same_name(const char *n1, const char *n2) { return 0 == memcmp(n1, n2, HEXDIR_LEN); } static int same_verf(nfs4_verifier *v1, nfs4_verifier *v2) { return 0 == memcmp(v1->data, v2->data, sizeof(v1->data)); } static int same_clid(clientid_t *cl1, clientid_t *cl2) { return (cl1->cl_boot == cl2->cl_boot) && (cl1->cl_id == cl2->cl_id); } static bool groups_equal(struct group_info *g1, struct group_info *g2) { int i; if (g1->ngroups != g2->ngroups) return false; for (i=0; ingroups; i++) if (GROUP_AT(g1, i) != GROUP_AT(g2, i)) return false; return true; } static int same_creds(struct svc_cred *cr1, struct svc_cred *cr2) { if ((cr1->cr_flavor != cr2->cr_flavor) || (cr1->cr_uid != cr2->cr_uid) || (cr1->cr_gid != cr2->cr_gid) || !groups_equal(cr1->cr_group_info, cr2->cr_group_info)) return false; if (cr1->cr_principal == cr2->cr_principal) return true; if (!cr1->cr_principal || !cr2->cr_principal) return false; return 0 == strcmp(cr1->cr_principal, cr1->cr_principal); } static void gen_clid(struct nfs4_client *clp) { static u32 current_clientid = 1; clp->cl_clientid.cl_boot = boot_time; clp->cl_clientid.cl_id = current_clientid++; } static void gen_confirm(struct nfs4_client *clp) { __be32 verf[2]; static u32 i; verf[0] = (__be32)get_seconds(); verf[1] = (__be32)i++; memcpy(clp->cl_confirm.data, verf, sizeof(clp->cl_confirm.data)); } static struct nfs4_stid *find_stateid(struct nfs4_client *cl, stateid_t *t) { return idr_find(&cl->cl_stateids, t->si_opaque.so_id); } static struct nfs4_stid *find_stateid_by_type(struct nfs4_client *cl, stateid_t *t, char typemask) { struct nfs4_stid *s; s = find_stateid(cl, t); if (!s) return NULL; if (typemask & s->sc_type) return s; return NULL; } static struct nfs4_client *create_client(struct xdr_netobj name, char *recdir, struct svc_rqst *rqstp, nfs4_verifier *verf) { struct nfs4_client *clp; struct sockaddr *sa = svc_addr(rqstp); int ret; clp = alloc_client(name); if (clp == NULL) return NULL; INIT_LIST_HEAD(&clp->cl_sessions); ret = copy_cred(&clp->cl_cred, &rqstp->rq_cred); if (ret) { spin_lock(&client_lock); free_client(clp); spin_unlock(&client_lock); return NULL; } idr_init(&clp->cl_stateids); memcpy(clp->cl_recdir, recdir, HEXDIR_LEN); atomic_set(&clp->cl_refcount, 0); clp->cl_cb_state = NFSD4_CB_UNKNOWN; INIT_LIST_HEAD(&clp->cl_idhash); INIT_LIST_HEAD(&clp->cl_strhash); INIT_LIST_HEAD(&clp->cl_openowners); INIT_LIST_HEAD(&clp->cl_delegations); INIT_LIST_HEAD(&clp->cl_lru); INIT_LIST_HEAD(&clp->cl_callbacks); spin_lock_init(&clp->cl_lock); INIT_WORK(&clp->cl_cb_null.cb_work, nfsd4_do_callback_rpc); clp->cl_time = get_seconds(); clear_bit(0, &clp->cl_cb_slot_busy); rpc_init_wait_queue(&clp->cl_cb_waitq, "Backchannel slot table"); copy_verf(clp, verf); rpc_copy_addr((struct sockaddr *) &clp->cl_addr, sa); gen_confirm(clp); clp->cl_cb_session = NULL; return clp; } static void add_to_unconfirmed(struct nfs4_client *clp, unsigned int strhashval) { unsigned int idhashval; list_add(&clp->cl_strhash, &unconf_str_hashtbl[strhashval]); idhashval = clientid_hashval(clp->cl_clientid.cl_id); list_add(&clp->cl_idhash, &unconf_id_hashtbl[idhashval]); renew_client(clp); } static void move_to_confirmed(struct nfs4_client *clp) { unsigned int idhashval = clientid_hashval(clp->cl_clientid.cl_id); unsigned int strhashval; dprintk("NFSD: move_to_confirm nfs4_client %p\n", clp); list_move(&clp->cl_idhash, &conf_id_hashtbl[idhashval]); strhashval = clientstr_hashval(clp->cl_recdir); list_move(&clp->cl_strhash, &conf_str_hashtbl[strhashval]); renew_client(clp); } static struct nfs4_client * find_confirmed_client(clientid_t *clid) { struct nfs4_client *clp; unsigned int idhashval = clientid_hashval(clid->cl_id); list_for_each_entry(clp, &conf_id_hashtbl[idhashval], cl_idhash) { if (same_clid(&clp->cl_clientid, clid)) { renew_client(clp); return clp; } } return NULL; } static struct nfs4_client * find_unconfirmed_client(clientid_t *clid) { struct nfs4_client *clp; unsigned int idhashval = clientid_hashval(clid->cl_id); list_for_each_entry(clp, &unconf_id_hashtbl[idhashval], cl_idhash) { if (same_clid(&clp->cl_clientid, clid)) return clp; } return NULL; } static bool clp_used_exchangeid(struct nfs4_client *clp) { return clp->cl_exchange_flags != 0; } static struct nfs4_client * find_confirmed_client_by_str(const char *dname, unsigned int hashval) { struct nfs4_client *clp; list_for_each_entry(clp, &conf_str_hashtbl[hashval], cl_strhash) { if (same_name(clp->cl_recdir, dname)) return clp; } return NULL; } static struct nfs4_client * find_unconfirmed_client_by_str(const char *dname, unsigned int hashval) { struct nfs4_client *clp; list_for_each_entry(clp, &unconf_str_hashtbl[hashval], cl_strhash) { if (same_name(clp->cl_recdir, dname)) return clp; } return NULL; } static void gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se, struct svc_rqst *rqstp) { struct nfs4_cb_conn *conn = &clp->cl_cb_conn; struct sockaddr *sa = svc_addr(rqstp); u32 scopeid = rpc_get_scope_id(sa); unsigned short expected_family; /* Currently, we only support tcp and tcp6 for the callback channel */ if (se->se_callback_netid_len == 3 && !memcmp(se->se_callback_netid_val, "tcp", 3)) expected_family = AF_INET; else if (se->se_callback_netid_len == 4 && !memcmp(se->se_callback_netid_val, "tcp6", 4)) expected_family = AF_INET6; else goto out_err; conn->cb_addrlen = rpc_uaddr2sockaddr(&init_net, se->se_callback_addr_val, se->se_callback_addr_len, (struct sockaddr *)&conn->cb_addr, sizeof(conn->cb_addr)); if (!conn->cb_addrlen || conn->cb_addr.ss_family != expected_family) goto out_err; if (conn->cb_addr.ss_family == AF_INET6) ((struct sockaddr_in6 *)&conn->cb_addr)->sin6_scope_id = scopeid; conn->cb_prog = se->se_callback_prog; conn->cb_ident = se->se_callback_ident; memcpy(&conn->cb_saddr, &rqstp->rq_daddr, rqstp->rq_daddrlen); return; out_err: conn->cb_addr.ss_family = AF_UNSPEC; conn->cb_addrlen = 0; dprintk(KERN_INFO "NFSD: this client (clientid %08x/%08x) " "will not receive delegations\n", clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id); return; } /* * Cache a reply. nfsd4_check_drc_limit() has bounded the cache size. */ void nfsd4_store_cache_entry(struct nfsd4_compoundres *resp) { struct nfsd4_slot *slot = resp->cstate.slot; unsigned int base; dprintk("--> %s slot %p\n", __func__, slot); slot->sl_opcnt = resp->opcnt; slot->sl_status = resp->cstate.status; slot->sl_flags |= NFSD4_SLOT_INITIALIZED; if (nfsd4_not_cached(resp)) { slot->sl_datalen = 0; return; } slot->sl_datalen = (char *)resp->p - (char *)resp->cstate.datap; base = (char *)resp->cstate.datap - (char *)resp->xbuf->head[0].iov_base; if (read_bytes_from_xdr_buf(resp->xbuf, base, slot->sl_data, slot->sl_datalen)) WARN("%s: sessions DRC could not cache compound\n", __func__); return; } /* * Encode the replay sequence operation from the slot values. * If cachethis is FALSE encode the uncached rep error on the next * operation which sets resp->p and increments resp->opcnt for * nfs4svc_encode_compoundres. * */ static __be32 nfsd4_enc_sequence_replay(struct nfsd4_compoundargs *args, struct nfsd4_compoundres *resp) { struct nfsd4_op *op; struct nfsd4_slot *slot = resp->cstate.slot; /* Encode the replayed sequence operation */ op = &args->ops[resp->opcnt - 1]; nfsd4_encode_operation(resp, op); /* Return nfserr_retry_uncached_rep in next operation. */ if (args->opcnt > 1 && !(slot->sl_flags & NFSD4_SLOT_CACHETHIS)) { op = &args->ops[resp->opcnt++]; op->status = nfserr_retry_uncached_rep; nfsd4_encode_operation(resp, op); } return op->status; } /* * The sequence operation is not cached because we can use the slot and * session values. */ __be32 nfsd4_replay_cache_entry(struct nfsd4_compoundres *resp, struct nfsd4_sequence *seq) { struct nfsd4_slot *slot = resp->cstate.slot; __be32 status; dprintk("--> %s slot %p\n", __func__, slot); /* Either returns 0 or nfserr_retry_uncached */ status = nfsd4_enc_sequence_replay(resp->rqstp->rq_argp, resp); if (status == nfserr_retry_uncached_rep) return status; /* The sequence operation has been encoded, cstate->datap set. */ memcpy(resp->cstate.datap, slot->sl_data, slot->sl_datalen); resp->opcnt = slot->sl_opcnt; resp->p = resp->cstate.datap + XDR_QUADLEN(slot->sl_datalen); status = slot->sl_status; return status; } /* * Set the exchange_id flags returned by the server. */ static void nfsd4_set_ex_flags(struct nfs4_client *new, struct nfsd4_exchange_id *clid) { /* pNFS is not supported */ new->cl_exchange_flags |= EXCHGID4_FLAG_USE_NON_PNFS; /* Referrals are supported, Migration is not. */ new->cl_exchange_flags |= EXCHGID4_FLAG_SUPP_MOVED_REFER; /* set the wire flags to return to client. */ clid->flags = new->cl_exchange_flags; } static bool client_has_state(struct nfs4_client *clp) { /* * Note clp->cl_openowners check isn't quite right: there's no * need to count owners without stateid's. * * Also note we should probably be using this in 4.0 case too. */ return !list_empty(&clp->cl_openowners) || !list_empty(&clp->cl_delegations) || !list_empty(&clp->cl_sessions); } __be32 nfsd4_exchange_id(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_exchange_id *exid) { struct nfs4_client *unconf, *conf, *new; __be32 status; unsigned int strhashval; char dname[HEXDIR_LEN]; char addr_str[INET6_ADDRSTRLEN]; nfs4_verifier verf = exid->verifier; struct sockaddr *sa = svc_addr(rqstp); bool update = exid->flags & EXCHGID4_FLAG_UPD_CONFIRMED_REC_A; rpc_ntop(sa, addr_str, sizeof(addr_str)); dprintk("%s rqstp=%p exid=%p clname.len=%u clname.data=%p " "ip_addr=%s flags %x, spa_how %d\n", __func__, rqstp, exid, exid->clname.len, exid->clname.data, addr_str, exid->flags, exid->spa_how); if (exid->flags & ~EXCHGID4_FLAG_MASK_A) return nfserr_inval; /* Currently only support SP4_NONE */ switch (exid->spa_how) { case SP4_NONE: break; case SP4_SSV: return nfserr_serverfault; default: BUG(); /* checked by xdr code */ case SP4_MACH_CRED: return nfserr_serverfault; /* no excuse :-/ */ } status = nfs4_make_rec_clidname(dname, &exid->clname); if (status) return status; strhashval = clientstr_hashval(dname); /* Cases below refer to rfc 5661 section 18.35.4: */ nfs4_lock_state(); conf = find_confirmed_client_by_str(dname, strhashval); if (conf) { bool creds_match = same_creds(&conf->cl_cred, &rqstp->rq_cred); bool verfs_match = same_verf(&verf, &conf->cl_verifier); if (update) { if (!clp_used_exchangeid(conf)) { /* buggy client */ status = nfserr_inval; goto out; } if (!creds_match) { /* case 9 */ status = nfserr_perm; goto out; } if (!verfs_match) { /* case 8 */ status = nfserr_not_same; goto out; } /* case 6 */ exid->flags |= EXCHGID4_FLAG_CONFIRMED_R; new = conf; goto out_copy; } if (!creds_match) { /* case 3 */ if (client_has_state(conf)) { status = nfserr_clid_inuse; goto out; } expire_client(conf); goto out_new; } if (verfs_match) { /* case 2 */ conf->cl_exchange_flags |= EXCHGID4_FLAG_CONFIRMED_R; new = conf; goto out_copy; } /* case 5, client reboot */ goto out_new; } if (update) { /* case 7 */ status = nfserr_noent; goto out; } unconf = find_unconfirmed_client_by_str(dname, strhashval); if (unconf) /* case 4, possible retry or client restart */ expire_client(unconf); /* case 1 (normal case) */ out_new: new = create_client(exid->clname, dname, rqstp, &verf); if (new == NULL) { status = nfserr_jukebox; goto out; } gen_clid(new); add_to_unconfirmed(new, strhashval); out_copy: exid->clientid.cl_boot = new->cl_clientid.cl_boot; exid->clientid.cl_id = new->cl_clientid.cl_id; exid->seqid = new->cl_cs_slot.sl_seqid + 1; nfsd4_set_ex_flags(new, exid); dprintk("nfsd4_exchange_id seqid %d flags %x\n", new->cl_cs_slot.sl_seqid, new->cl_exchange_flags); status = nfs_ok; out: nfs4_unlock_state(); return status; } static __be32 check_slot_seqid(u32 seqid, u32 slot_seqid, int slot_inuse) { dprintk("%s enter. seqid %d slot_seqid %d\n", __func__, seqid, slot_seqid); /* The slot is in use, and no response has been sent. */ if (slot_inuse) { if (seqid == slot_seqid) return nfserr_jukebox; else return nfserr_seq_misordered; } /* Note unsigned 32-bit arithmetic handles wraparound: */ if (likely(seqid == slot_seqid + 1)) return nfs_ok; if (seqid == slot_seqid) return nfserr_replay_cache; return nfserr_seq_misordered; } /* * Cache the create session result into the create session single DRC * slot cache by saving the xdr structure. sl_seqid has been set. * Do this for solo or embedded create session operations. */ static void nfsd4_cache_create_session(struct nfsd4_create_session *cr_ses, struct nfsd4_clid_slot *slot, __be32 nfserr) { slot->sl_status = nfserr; memcpy(&slot->sl_cr_ses, cr_ses, sizeof(*cr_ses)); } static __be32 nfsd4_replay_create_session(struct nfsd4_create_session *cr_ses, struct nfsd4_clid_slot *slot) { memcpy(cr_ses, &slot->sl_cr_ses, sizeof(*cr_ses)); return slot->sl_status; } #define NFSD_MIN_REQ_HDR_SEQ_SZ ((\ 2 * 2 + /* credential,verifier: AUTH_NULL, length 0 */ \ 1 + /* MIN tag is length with zero, only length */ \ 3 + /* version, opcount, opcode */ \ XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + \ /* seqid, slotID, slotID, cache */ \ 4 ) * sizeof(__be32)) #define NFSD_MIN_RESP_HDR_SEQ_SZ ((\ 2 + /* verifier: AUTH_NULL, length 0 */\ 1 + /* status */ \ 1 + /* MIN tag is length with zero, only length */ \ 3 + /* opcount, opcode, opstatus*/ \ XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + \ /* seqid, slotID, slotID, slotID, status */ \ 5 ) * sizeof(__be32)) static bool check_forechannel_attrs(struct nfsd4_channel_attrs fchannel) { return fchannel.maxreq_sz < NFSD_MIN_REQ_HDR_SEQ_SZ || fchannel.maxresp_sz < NFSD_MIN_RESP_HDR_SEQ_SZ; } __be32 nfsd4_create_session(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_create_session *cr_ses) { struct sockaddr *sa = svc_addr(rqstp); struct nfs4_client *conf, *unconf; struct nfsd4_session *new; struct nfsd4_clid_slot *cs_slot = NULL; bool confirm_me = false; __be32 status = 0; if (cr_ses->flags & ~SESSION4_FLAG_MASK_A) return nfserr_inval; nfs4_lock_state(); unconf = find_unconfirmed_client(&cr_ses->clientid); conf = find_confirmed_client(&cr_ses->clientid); if (conf) { cs_slot = &conf->cl_cs_slot; status = check_slot_seqid(cr_ses->seqid, cs_slot->sl_seqid, 0); if (status == nfserr_replay_cache) { status = nfsd4_replay_create_session(cr_ses, cs_slot); goto out; } else if (cr_ses->seqid != cs_slot->sl_seqid + 1) { status = nfserr_seq_misordered; goto out; } } else if (unconf) { if (!same_creds(&unconf->cl_cred, &rqstp->rq_cred) || !rpc_cmp_addr(sa, (struct sockaddr *) &unconf->cl_addr)) { status = nfserr_clid_inuse; goto out; } cs_slot = &unconf->cl_cs_slot; status = check_slot_seqid(cr_ses->seqid, cs_slot->sl_seqid, 0); if (status) { /* an unconfirmed replay returns misordered */ status = nfserr_seq_misordered; goto out; } confirm_me = true; conf = unconf; } else { status = nfserr_stale_clientid; goto out; } /* * XXX: we should probably set this at creation time, and check * for consistent minorversion use throughout: */ conf->cl_minorversion = 1; /* * We do not support RDMA or persistent sessions */ cr_ses->flags &= ~SESSION4_PERSIST; cr_ses->flags &= ~SESSION4_RDMA; status = nfserr_toosmall; if (check_forechannel_attrs(cr_ses->fore_channel)) goto out; status = nfserr_jukebox; new = alloc_init_session(rqstp, conf, cr_ses); if (!new) goto out; status = nfs_ok; memcpy(cr_ses->sessionid.data, new->se_sessionid.data, NFS4_MAX_SESSIONID_LEN); memcpy(&cr_ses->fore_channel, &new->se_fchannel, sizeof(struct nfsd4_channel_attrs)); cs_slot->sl_seqid++; cr_ses->seqid = cs_slot->sl_seqid; /* cache solo and embedded create sessions under the state lock */ nfsd4_cache_create_session(cr_ses, cs_slot, status); if (confirm_me) { unsigned int hash = clientstr_hashval(unconf->cl_recdir); struct nfs4_client *old = find_confirmed_client_by_str(conf->cl_recdir, hash); if (old) expire_client(old); move_to_confirmed(conf); } out: nfs4_unlock_state(); dprintk("%s returns %d\n", __func__, ntohl(status)); return status; } static bool nfsd4_last_compound_op(struct svc_rqst *rqstp) { struct nfsd4_compoundres *resp = rqstp->rq_resp; struct nfsd4_compoundargs *argp = rqstp->rq_argp; return argp->opcnt == resp->opcnt; } static __be32 nfsd4_map_bcts_dir(u32 *dir) { switch (*dir) { case NFS4_CDFC4_FORE: case NFS4_CDFC4_BACK: return nfs_ok; case NFS4_CDFC4_FORE_OR_BOTH: case NFS4_CDFC4_BACK_OR_BOTH: *dir = NFS4_CDFC4_BOTH; return nfs_ok; }; return nfserr_inval; } __be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_bind_conn_to_session *bcts) { __be32 status; if (!nfsd4_last_compound_op(rqstp)) return nfserr_not_only_op; spin_lock(&client_lock); cstate->session = find_in_sessionid_hashtbl(&bcts->sessionid); /* Sorta weird: we only need the refcnt'ing because new_conn acquires * client_lock iself: */ if (cstate->session) { nfsd4_get_session(cstate->session); atomic_inc(&cstate->session->se_client->cl_refcount); } spin_unlock(&client_lock); if (!cstate->session) return nfserr_badsession; status = nfsd4_map_bcts_dir(&bcts->dir); if (!status) nfsd4_new_conn(rqstp, cstate->session, bcts->dir); return status; } static bool nfsd4_compound_in_session(struct nfsd4_session *session, struct nfs4_sessionid *sid) { if (!session) return 0; return !memcmp(sid, &session->se_sessionid, sizeof(*sid)); } __be32 nfsd4_destroy_session(struct svc_rqst *r, struct nfsd4_compound_state *cstate, struct nfsd4_destroy_session *sessionid) { struct nfsd4_session *ses; __be32 status = nfserr_badsession; /* Notes: * - The confirmed nfs4_client->cl_sessionid holds destroyed sessinid * - Should we return nfserr_back_chan_busy if waiting for * callbacks on to-be-destroyed session? * - Do we need to clear any callback info from previous session? */ if (nfsd4_compound_in_session(cstate->session, &sessionid->sessionid)) { if (!nfsd4_last_compound_op(r)) return nfserr_not_only_op; } dump_sessionid(__func__, &sessionid->sessionid); spin_lock(&client_lock); ses = find_in_sessionid_hashtbl(&sessionid->sessionid); if (!ses) { spin_unlock(&client_lock); goto out; } unhash_session(ses); spin_unlock(&client_lock); nfs4_lock_state(); nfsd4_probe_callback_sync(ses->se_client); nfs4_unlock_state(); spin_lock(&client_lock); nfsd4_del_conns(ses); nfsd4_put_session_locked(ses); spin_unlock(&client_lock); status = nfs_ok; out: dprintk("%s returns %d\n", __func__, ntohl(status)); return status; } static struct nfsd4_conn *__nfsd4_find_conn(struct svc_xprt *xpt, struct nfsd4_session *s) { struct nfsd4_conn *c; list_for_each_entry(c, &s->se_conns, cn_persession) { if (c->cn_xprt == xpt) { return c; } } return NULL; } static void nfsd4_sequence_check_conn(struct nfsd4_conn *new, struct nfsd4_session *ses) { struct nfs4_client *clp = ses->se_client; struct nfsd4_conn *c; int ret; spin_lock(&clp->cl_lock); c = __nfsd4_find_conn(new->cn_xprt, ses); if (c) { spin_unlock(&clp->cl_lock); free_conn(new); return; } __nfsd4_hash_conn(new, ses); spin_unlock(&clp->cl_lock); ret = nfsd4_register_conn(new); if (ret) /* oops; xprt is already down: */ nfsd4_conn_lost(&new->cn_xpt_user); return; } static bool nfsd4_session_too_many_ops(struct svc_rqst *rqstp, struct nfsd4_session *session) { struct nfsd4_compoundargs *args = rqstp->rq_argp; return args->opcnt > session->se_fchannel.maxops; } static bool nfsd4_request_too_big(struct svc_rqst *rqstp, struct nfsd4_session *session) { struct xdr_buf *xb = &rqstp->rq_arg; return xb->len > session->se_fchannel.maxreq_sz; } __be32 nfsd4_sequence(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_sequence *seq) { struct nfsd4_compoundres *resp = rqstp->rq_resp; struct nfsd4_session *session; struct nfsd4_slot *slot; struct nfsd4_conn *conn; __be32 status; if (resp->opcnt != 1) return nfserr_sequence_pos; /* * Will be either used or freed by nfsd4_sequence_check_conn * below. */ conn = alloc_conn(rqstp, NFS4_CDFC4_FORE); if (!conn) return nfserr_jukebox; spin_lock(&client_lock); status = nfserr_badsession; session = find_in_sessionid_hashtbl(&seq->sessionid); if (!session) goto out; status = nfserr_too_many_ops; if (nfsd4_session_too_many_ops(rqstp, session)) goto out; status = nfserr_req_too_big; if (nfsd4_request_too_big(rqstp, session)) goto out; status = nfserr_badslot; if (seq->slotid >= session->se_fchannel.maxreqs) goto out; slot = session->se_slots[seq->slotid]; dprintk("%s: slotid %d\n", __func__, seq->slotid); /* We do not negotiate the number of slots yet, so set the * maxslots to the session maxreqs which is used to encode * sr_highest_slotid and the sr_target_slot id to maxslots */ seq->maxslots = session->se_fchannel.maxreqs; status = check_slot_seqid(seq->seqid, slot->sl_seqid, slot->sl_flags & NFSD4_SLOT_INUSE); if (status == nfserr_replay_cache) { status = nfserr_seq_misordered; if (!(slot->sl_flags & NFSD4_SLOT_INITIALIZED)) goto out; cstate->slot = slot; cstate->session = session; /* Return the cached reply status and set cstate->status * for nfsd4_proc_compound processing */ status = nfsd4_replay_cache_entry(resp, seq); cstate->status = nfserr_replay_cache; goto out; } if (status) goto out; nfsd4_sequence_check_conn(conn, session); conn = NULL; /* Success! bump slot seqid */ slot->sl_seqid = seq->seqid; slot->sl_flags |= NFSD4_SLOT_INUSE; if (seq->cachethis) slot->sl_flags |= NFSD4_SLOT_CACHETHIS; else slot->sl_flags &= ~NFSD4_SLOT_CACHETHIS; cstate->slot = slot; cstate->session = session; out: /* Hold a session reference until done processing the compound. */ if (cstate->session) { struct nfs4_client *clp = session->se_client; nfsd4_get_session(cstate->session); atomic_inc(&clp->cl_refcount); switch (clp->cl_cb_state) { case NFSD4_CB_DOWN: seq->status_flags = SEQ4_STATUS_CB_PATH_DOWN; break; case NFSD4_CB_FAULT: seq->status_flags = SEQ4_STATUS_BACKCHANNEL_FAULT; break; default: seq->status_flags = 0; } } kfree(conn); spin_unlock(&client_lock); dprintk("%s: return %d\n", __func__, ntohl(status)); return status; } __be32 nfsd4_destroy_clientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_destroy_clientid *dc) { struct nfs4_client *conf, *unconf, *clp; __be32 status = 0; nfs4_lock_state(); unconf = find_unconfirmed_client(&dc->clientid); conf = find_confirmed_client(&dc->clientid); if (conf) { clp = conf; if (!is_client_expired(conf) && client_has_state(conf)) { status = nfserr_clientid_busy; goto out; } /* rfc5661 18.50.3 */ if (cstate->session && conf == cstate->session->se_client) { status = nfserr_clientid_busy; goto out; } } else if (unconf) clp = unconf; else { status = nfserr_stale_clientid; goto out; } expire_client(clp); out: nfs4_unlock_state(); dprintk("%s return %d\n", __func__, ntohl(status)); return status; } __be32 nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_reclaim_complete *rc) { __be32 status = 0; if (rc->rca_one_fs) { if (!cstate->current_fh.fh_dentry) return nfserr_nofilehandle; /* * We don't take advantage of the rca_one_fs case. * That's OK, it's optional, we can safely ignore it. */ return nfs_ok; } nfs4_lock_state(); status = nfserr_complete_already; if (test_and_set_bit(NFSD4_CLIENT_RECLAIM_COMPLETE, &cstate->session->se_client->cl_flags)) goto out; status = nfserr_stale_clientid; if (is_client_expired(cstate->session->se_client)) /* * The following error isn't really legal. * But we only get here if the client just explicitly * destroyed the client. Surely it no longer cares what * error it gets back on an operation for the dead * client. */ goto out; status = nfs_ok; nfsd4_client_record_create(cstate->session->se_client); out: nfs4_unlock_state(); return status; } __be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_setclientid *setclid) { struct xdr_netobj clname = setclid->se_name; nfs4_verifier clverifier = setclid->se_verf; unsigned int strhashval; struct nfs4_client *conf, *unconf, *new; __be32 status; char dname[HEXDIR_LEN]; status = nfs4_make_rec_clidname(dname, &clname); if (status) return status; strhashval = clientstr_hashval(dname); /* Cases below refer to rfc 3530 section 14.2.33: */ nfs4_lock_state(); conf = find_confirmed_client_by_str(dname, strhashval); if (conf) { /* case 0: */ status = nfserr_clid_inuse; if (clp_used_exchangeid(conf)) goto out; if (!same_creds(&conf->cl_cred, &rqstp->rq_cred)) { char addr_str[INET6_ADDRSTRLEN]; rpc_ntop((struct sockaddr *) &conf->cl_addr, addr_str, sizeof(addr_str)); dprintk("NFSD: setclientid: string in use by client " "at %s\n", addr_str); goto out; } } unconf = find_unconfirmed_client_by_str(dname, strhashval); if (unconf) expire_client(unconf); status = nfserr_jukebox; new = create_client(clname, dname, rqstp, &clverifier); if (new == NULL) goto out; if (conf && same_verf(&conf->cl_verifier, &clverifier)) /* case 1: probable callback update */ copy_clid(new, conf); else /* case 4 (new client) or cases 2, 3 (client reboot): */ gen_clid(new); /* * XXX: we should probably set this at creation time, and check * for consistent minorversion use throughout: */ new->cl_minorversion = 0; gen_callback(new, setclid, rqstp); add_to_unconfirmed(new, strhashval); setclid->se_clientid.cl_boot = new->cl_clientid.cl_boot; setclid->se_clientid.cl_id = new->cl_clientid.cl_id; memcpy(setclid->se_confirm.data, new->cl_confirm.data, sizeof(setclid->se_confirm.data)); status = nfs_ok; out: nfs4_unlock_state(); return status; } __be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_setclientid_confirm *setclientid_confirm) { struct nfs4_client *conf, *unconf; nfs4_verifier confirm = setclientid_confirm->sc_confirm; clientid_t * clid = &setclientid_confirm->sc_clientid; __be32 status; if (STALE_CLIENTID(clid)) return nfserr_stale_clientid; nfs4_lock_state(); conf = find_confirmed_client(clid); unconf = find_unconfirmed_client(clid); /* * We try hard to give out unique clientid's, so if we get an * attempt to confirm the same clientid with a different cred, * there's a bug somewhere. Let's charitably assume it's our * bug. */ status = nfserr_serverfault; if (unconf && !same_creds(&unconf->cl_cred, &rqstp->rq_cred)) goto out; if (conf && !same_creds(&conf->cl_cred, &rqstp->rq_cred)) goto out; /* cases below refer to rfc 3530 section 14.2.34: */ if (!unconf || !same_verf(&confirm, &unconf->cl_confirm)) { if (conf && !unconf) /* case 2: probable retransmit */ status = nfs_ok; else /* case 4: client hasn't noticed we rebooted yet? */ status = nfserr_stale_clientid; goto out; } status = nfs_ok; if (conf) { /* case 1: callback update */ nfsd4_change_callback(conf, &unconf->cl_cb_conn); nfsd4_probe_callback(conf); expire_client(unconf); } else { /* case 3: normal case; new or rebooted client */ unsigned int hash = clientstr_hashval(unconf->cl_recdir); conf = find_confirmed_client_by_str(unconf->cl_recdir, hash); if (conf) { nfsd4_client_record_remove(conf); expire_client(conf); } move_to_confirmed(unconf); nfsd4_probe_callback(unconf); } out: nfs4_unlock_state(); return status; } static struct nfs4_file *nfsd4_alloc_file(void) { return kmem_cache_alloc(file_slab, GFP_KERNEL); } /* OPEN Share state helper functions */ static void nfsd4_init_file(struct nfs4_file *fp, struct inode *ino) { unsigned int hashval = file_hashval(ino); atomic_set(&fp->fi_ref, 1); INIT_LIST_HEAD(&fp->fi_hash); INIT_LIST_HEAD(&fp->fi_stateids); INIT_LIST_HEAD(&fp->fi_delegations); fp->fi_inode = igrab(ino); fp->fi_had_conflict = false; fp->fi_lease = NULL; memset(fp->fi_fds, 0, sizeof(fp->fi_fds)); memset(fp->fi_access, 0, sizeof(fp->fi_access)); spin_lock(&recall_lock); list_add(&fp->fi_hash, &file_hashtbl[hashval]); spin_unlock(&recall_lock); } static void nfsd4_free_slab(struct kmem_cache **slab) { if (*slab == NULL) return; kmem_cache_destroy(*slab); *slab = NULL; } void nfsd4_free_slabs(void) { nfsd4_free_slab(&openowner_slab); nfsd4_free_slab(&lockowner_slab); nfsd4_free_slab(&file_slab); nfsd4_free_slab(&stateid_slab); nfsd4_free_slab(&deleg_slab); } int nfsd4_init_slabs(void) { openowner_slab = kmem_cache_create("nfsd4_openowners", sizeof(struct nfs4_openowner), 0, 0, NULL); if (openowner_slab == NULL) goto out_nomem; lockowner_slab = kmem_cache_create("nfsd4_lockowners", sizeof(struct nfs4_openowner), 0, 0, NULL); if (lockowner_slab == NULL) goto out_nomem; file_slab = kmem_cache_create("nfsd4_files", sizeof(struct nfs4_file), 0, 0, NULL); if (file_slab == NULL) goto out_nomem; stateid_slab = kmem_cache_create("nfsd4_stateids", sizeof(struct nfs4_ol_stateid), 0, 0, NULL); if (stateid_slab == NULL) goto out_nomem; deleg_slab = kmem_cache_create("nfsd4_delegations", sizeof(struct nfs4_delegation), 0, 0, NULL); if (deleg_slab == NULL) goto out_nomem; return 0; out_nomem: nfsd4_free_slabs(); dprintk("nfsd4: out of memory while initializing nfsv4\n"); return -ENOMEM; } void nfs4_free_openowner(struct nfs4_openowner *oo) { kfree(oo->oo_owner.so_owner.data); kmem_cache_free(openowner_slab, oo); } void nfs4_free_lockowner(struct nfs4_lockowner *lo) { kfree(lo->lo_owner.so_owner.data); kmem_cache_free(lockowner_slab, lo); } static void init_nfs4_replay(struct nfs4_replay *rp) { rp->rp_status = nfserr_serverfault; rp->rp_buflen = 0; rp->rp_buf = rp->rp_ibuf; } static inline void *alloc_stateowner(struct kmem_cache *slab, struct xdr_netobj *owner, struct nfs4_client *clp) { struct nfs4_stateowner *sop; sop = kmem_cache_alloc(slab, GFP_KERNEL); if (!sop) return NULL; sop->so_owner.data = kmemdup(owner->data, owner->len, GFP_KERNEL); if (!sop->so_owner.data) { kmem_cache_free(slab, sop); return NULL; } sop->so_owner.len = owner->len; INIT_LIST_HEAD(&sop->so_stateids); sop->so_client = clp; init_nfs4_replay(&sop->so_replay); return sop; } static void hash_openowner(struct nfs4_openowner *oo, struct nfs4_client *clp, unsigned int strhashval) { list_add(&oo->oo_owner.so_strhash, &ownerstr_hashtbl[strhashval]); list_add(&oo->oo_perclient, &clp->cl_openowners); } static struct nfs4_openowner * alloc_init_open_stateowner(unsigned int strhashval, struct nfs4_client *clp, struct nfsd4_open *open) { struct nfs4_openowner *oo; oo = alloc_stateowner(openowner_slab, &open->op_owner, clp); if (!oo) return NULL; oo->oo_owner.so_is_open_owner = 1; oo->oo_owner.so_seqid = open->op_seqid; oo->oo_flags = NFS4_OO_NEW; oo->oo_time = 0; oo->oo_last_closed_stid = NULL; INIT_LIST_HEAD(&oo->oo_close_lru); hash_openowner(oo, clp, strhashval); return oo; } static void init_open_stateid(struct nfs4_ol_stateid *stp, struct nfs4_file *fp, struct nfsd4_open *open) { struct nfs4_openowner *oo = open->op_openowner; struct nfs4_client *clp = oo->oo_owner.so_client; init_stid(&stp->st_stid, clp, NFS4_OPEN_STID); INIT_LIST_HEAD(&stp->st_lockowners); list_add(&stp->st_perstateowner, &oo->oo_owner.so_stateids); list_add(&stp->st_perfile, &fp->fi_stateids); stp->st_stateowner = &oo->oo_owner; get_nfs4_file(fp); stp->st_file = fp; stp->st_access_bmap = 0; stp->st_deny_bmap = 0; set_access(open->op_share_access, stp); set_deny(open->op_share_deny, stp); stp->st_openstp = NULL; } static void move_to_close_lru(struct nfs4_openowner *oo) { dprintk("NFSD: move_to_close_lru nfs4_openowner %p\n", oo); list_move_tail(&oo->oo_close_lru, &close_lru); oo->oo_time = get_seconds(); } static int same_owner_str(struct nfs4_stateowner *sop, struct xdr_netobj *owner, clientid_t *clid) { return (sop->so_owner.len == owner->len) && 0 == memcmp(sop->so_owner.data, owner->data, owner->len) && (sop->so_client->cl_clientid.cl_id == clid->cl_id); } static struct nfs4_openowner * find_openstateowner_str(unsigned int hashval, struct nfsd4_open *open) { struct nfs4_stateowner *so; struct nfs4_openowner *oo; list_for_each_entry(so, &ownerstr_hashtbl[hashval], so_strhash) { if (!so->so_is_open_owner) continue; if (same_owner_str(so, &open->op_owner, &open->op_cl