diff options
58 files changed, 2195 insertions, 964 deletions
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index a1489e14f8ee..58f3c1041759 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2470,6 +2470,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted. nfsrootdebug [NFS] enable nfsroot debugging messages. See Documentation/filesystems/nfs/nfsroot.txt. + nfs.callback_nr_threads= + [NFSv4] set the total number of threads that the + NFS client will assign to service NFSv4 callback + requests. + nfs.callback_tcpport= [NFS] set the TCP port on which the NFSv4 callback channel should listen. @@ -2493,6 +2498,13 @@ bytes respectively. Such letter suffixes can also be entirely omitted. of returning the full 64-bit number. The default is to return 64-bit inode numbers. + nfs.max_session_cb_slots= + [NFSv4.1] Sets the maximum number of session + slots the client will assign to the callback + channel. This determines the maximum number of + callbacks the client will process in parallel for + a particular server. + nfs.max_session_slots= [NFSv4.1] Sets the maximum number of session slots the client will attempt to negotiate with the server. diff --git a/fs/nfs/cache_lib.c b/fs/nfs/cache_lib.c index 5f7b053720ee..6de15709d024 100644 --- a/fs/nfs/cache_lib.c +++ b/fs/nfs/cache_lib.c @@ -76,7 +76,7 @@ static void nfs_dns_cache_revisit(struct cache_deferred_req *d, int toomany) dreq = container_of(d, struct nfs_cache_defer_req, deferred_req); - complete_all(&dreq->completion); + complete(&dreq->completion); nfs_cache_defer_req_put(dreq); } diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 52a28311e2a4..532d8e242d4d 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -31,8 +31,6 @@ struct nfs_callback_data { unsigned int users; struct svc_serv *serv; - struct svc_rqst *rqst; - struct task_struct *task; }; static struct nfs_callback_data nfs_callback_info[NFS4_MAX_MINOR_VERSION + 1]; @@ -89,15 +87,6 @@ nfs4_callback_svc(void *vrqstp) return 0; } -/* - * Prepare to bring up the NFSv4 callback service - */ -static struct svc_rqst * -nfs4_callback_up(struct svc_serv *serv) -{ - return svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE); -} - #if defined(CONFIG_NFS_V4_1) /* * The callback service for NFSv4.1 callbacks @@ -139,29 +128,6 @@ nfs41_callback_svc(void *vrqstp) return 0; } -/* - * Bring up the NFSv4.1 callback service - */ -static struct svc_rqst * -nfs41_callback_up(struct svc_serv *serv) -{ - struct svc_rqst *rqstp; - - INIT_LIST_HEAD(&serv->sv_cb_list); - spin_lock_init(&serv->sv_cb_lock); - init_waitqueue_head(&serv->sv_cb_waitq); - rqstp = svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE); - dprintk("--> %s return %d\n", __func__, PTR_ERR_OR_ZERO(rqstp)); - return rqstp; -} - -static void nfs_minorversion_callback_svc_setup(struct svc_serv *serv, - struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp)) -{ - *rqstpp = nfs41_callback_up(serv); - *callback_svc = nfs41_callback_svc; -} - static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt, struct svc_serv *serv) { @@ -173,13 +139,6 @@ static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt, xprt->bc_serv = serv; } #else -static void nfs_minorversion_callback_svc_setup(struct svc_serv *serv, - struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp)) -{ - *rqstpp = ERR_PTR(-ENOTSUPP); - *callback_svc = ERR_PTR(-ENOTSUPP); -} - static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt, struct svc_serv *serv) { @@ -189,45 +148,22 @@ static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt, static int nfs_callback_start_svc(int minorversion, struct rpc_xprt *xprt, struct svc_serv *serv) { - struct svc_rqst *rqstp; - int (*callback_svc)(void *vrqstp); - struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; + int nrservs = nfs_callback_nr_threads; int ret; nfs_callback_bc_serv(minorversion, xprt, serv); - if (cb_info->task) - return 0; + if (nrservs < NFS4_MIN_NR_CALLBACK_THREADS) + nrservs = NFS4_MIN_NR_CALLBACK_THREADS; - switch (minorversion) { - case 0: - /* v4.0 callback setup */ - rqstp = nfs4_callback_up(serv); - callback_svc = nfs4_callback_svc; - break; - default: - nfs_minorversion_callback_svc_setup(serv, - &rqstp, &callback_svc); - } - - if (IS_ERR(rqstp)) - return PTR_ERR(rqstp); - - svc_sock_update_bufs(serv); + if (serv->sv_nrthreads-1 == nrservs) + return 0; - cb_info->serv = serv; - cb_info->rqst = rqstp; - cb_info->task = kthread_create(callback_svc, cb_info->rqst, - "nfsv4.%u-svc", minorversion); - if (IS_ERR(cb_info->task)) { - ret = PTR_ERR(cb_info->task); - svc_exit_thread(cb_info->rqst); - cb_info->rqst = NULL; - cb_info->task = NULL; + ret = serv->sv_ops->svo_setup(serv, NULL, nrservs); + if (ret) { + serv->sv_ops->svo_setup(serv, NULL, 0); return ret; } - rqstp->rq_task = cb_info->task; - wake_up_process(cb_info->task); dprintk("nfs_callback_up: service started\n"); return 0; } @@ -281,19 +217,41 @@ err_bind: return ret; } -static struct svc_serv_ops nfs_cb_sv_ops = { +static struct svc_serv_ops nfs40_cb_sv_ops = { + .svo_function = nfs4_callback_svc, .svo_enqueue_xprt = svc_xprt_do_enqueue, + .svo_setup = svc_set_num_threads, + .svo_module = THIS_MODULE, +}; +#if defined(CONFIG_NFS_V4_1) +static struct svc_serv_ops nfs41_cb_sv_ops = { + .svo_function = nfs41_callback_svc, + .svo_enqueue_xprt = svc_xprt_do_enqueue, + .svo_setup = svc_set_num_threads, + .svo_module = THIS_MODULE, +}; + +struct svc_serv_ops *nfs4_cb_sv_ops[] = { + [0] = &nfs40_cb_sv_ops, + [1] = &nfs41_cb_sv_ops, +}; +#else +struct svc_serv_ops *nfs4_cb_sv_ops[] = { + [0] = &nfs40_cb_sv_ops, + [1] = NULL, }; +#endif static struct svc_serv *nfs_callback_create_svc(int minorversion) { struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; struct svc_serv *serv; + struct svc_serv_ops *sv_ops; /* * Check whether we're already up and running. */ - if (cb_info->task) { + if (cb_info->serv) { /* * Note: increase service usage, because later in case of error * svc_destroy() will be called. @@ -302,6 +260,17 @@ static struct svc_serv *nfs_callback_create_svc(int minorversion) return cb_info->serv; } + switch (minorversion) { + case 0: + sv_ops = nfs4_cb_sv_ops[0]; + break; + default: + sv_ops = nfs4_cb_sv_ops[1]; + } + + if (sv_ops == NULL) + return ERR_PTR(-ENOTSUPP); + /* * Sanity check: if there's no task, * we should be the first user ... @@ -310,11 +279,12 @@ static struct svc_serv *nfs_callback_create_svc(int minorversion) printk(KERN_WARNING "nfs_callback_create_svc: no kthread, %d users??\n", cb_info->users); - serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, &nfs_cb_sv_ops); + serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, sv_ops); if (!serv) { printk(KERN_ERR "nfs_callback_create_svc: create service failed\n"); return ERR_PTR(-ENOMEM); } + cb_info->serv = serv; /* As there is only one thread we need to over-ride the * default maximum of 80 connections */ @@ -357,6 +327,8 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt) * thread exits. */ err_net: + if (!cb_info->users) + cb_info->serv = NULL; svc_destroy(serv); err_create: mutex_unlock(&nfs_callback_mutex); @@ -374,18 +346,18 @@ err_start: void nfs_callback_down(int minorversion, struct net *net) { struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion]; + struct svc_serv *serv; mutex_lock(&nfs_callback_mutex); - nfs_callback_down_net(minorversion, cb_info->serv, net); + serv = cb_info->serv; + nfs_callback_down_net(minorversion, serv, net); cb_info->users--; - if (cb_info->users == 0 && cb_info->task != NULL) { - kthread_stop(cb_info->task); - dprintk("nfs_callback_down: service stopped\n"); - svc_exit_thread(cb_info->rqst); + if (cb_info->users == 0) { + svc_get(serv); + serv->sv_ops->svo_setup(serv, NULL, 0); + svc_destroy(serv); dprintk("nfs_callback_down: service destroyed\n"); cb_info->serv = NULL; - cb_info->rqst = NULL; - cb_info->task = NULL; } mutex_unlock(&nfs_callback_mutex); } diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index 5fe1cecbf9f0..c701c308fac5 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -179,6 +179,15 @@ extern __be32 nfs4_callback_devicenotify( struct cb_devicenotifyargs *args, void *dummy, struct cb_process_state *cps); +struct cb_notify_lock_args { + struct nfs_fh cbnl_fh; + struct nfs_lowner cbnl_owner; + bool cbnl_valid; +}; + +extern __be32 nfs4_callback_notify_lock(struct cb_notify_lock_args *args, + void *dummy, + struct cb_process_state *cps); #endif /* CONFIG_NFS_V4_1 */ extern int check_gss_callback_principal(struct nfs_client *, struct svc_rqst *); extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, @@ -198,6 +207,9 @@ extern void nfs_callback_down(int minorversion, struct net *net); #define NFS41_BC_MIN_CALLBACKS 1 #define NFS41_BC_MAX_CALLBACKS 1 +#define NFS4_MIN_NR_CALLBACK_THREADS 1 + extern unsigned int nfs_callback_set_tcpport; +extern unsigned short nfs_callback_nr_threads; #endif /* __LINUX_FS_NFS_CALLBACK_H */ diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index f953ef6b2f2e..e9aa235e9d10 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -628,4 +628,20 @@ out: dprintk("%s: exit with status = %d\n", __func__, ntohl(status)); return status; } + +__be32 nfs4_callback_notify_lock(struct cb_notify_lock_args *args, void *dummy, + struct cb_process_state *cps) +{ + if (!cps->clp) /* set in cb_sequence */ + return htonl(NFS4ERR_OP_NOT_IN_SESSION); + + dprintk_rcu("NFS: CB_NOTIFY_LOCK request from %s\n", + rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR)); + + /* Don't wake anybody if the string looked bogus */ + if (args->cbnl_valid) + __wake_up(&cps->clp->cl_lock_waitq, TASK_NORMAL, 0, args); + + return htonl(NFS4_OK); +} #endif /* CONFIG_NFS_V4_1 */ diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 656f68f7fe53..eb094c6011d8 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -35,6 +35,7 @@ (1 + 3) * 4) // seqid, 3 slotids #define CB_OP_RECALLANY_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) #define CB_OP_RECALLSLOT_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) +#define CB_OP_NOTIFY_LOCK_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) #endif /* CONFIG_NFS_V4_1 */ #define NFSDBG_FACILITY NFSDBG_CALLBACK @@ -72,7 +73,7 @@ static int nfs4_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) return xdr_ressize_check(rqstp, p); } -static __be32 *read_buf(struct xdr_stream *xdr, int nbytes) +static __be32 *read_buf(struct xdr_stream *xdr, size_t nbytes) { __be32 *p; @@ -534,6 +535,49 @@ static __be32 decode_recallslot_args(struct svc_rqst *rqstp, return 0; } +static __be32 decode_lockowner(struct xdr_stream *xdr, struct cb_notify_lock_args *args) +{ + __be32 *p; + unsigned int len; + + p = read_buf(xdr, 12); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_BADXDR); + + p = xdr_decode_hyper(p, &args->cbnl_owner.clientid); + len = be32_to_cpu(*p); + + p = read_buf(xdr, len); + if (unlikely(p == NULL)) + return htonl(NFS4ERR_BADXDR); + + /* Only try to decode if the length is right */ + if (len == 20) { + p += 2; /* skip "lock id:" */ + args->cbnl_owner.s_dev = be32_to_cpu(*p++); + xdr_decode_hyper(p, &args->cbnl_owner.id); + args->cbnl_valid = true; + } else { + args->cbnl_owner.s_dev = 0; + args->cbnl_owner.id = 0; + args->cbnl_valid = false; + } + return 0; +} + +static __be32 decode_notify_lock_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_notify_lock_args *args) +{ + __be32 status; + + status = decode_fh(xdr, &args->cbnl_fh); + if (unlikely(status != 0)) + goto out; + status = decode_lockowner(xdr, args); +out: + dprintk("%s: exit with status = %d\n", __func__, ntohl(status)); + return status; +} + #endif /* CONFIG_NFS_V4_1 */ static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) @@ -746,6 +790,7 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op) case OP_CB_RECALL_SLOT: case OP_CB_LAYOUTRECALL: case OP_CB_NOTIFY_DEVICEID: + case OP_CB_NOTIFY_LOCK: *op = &callback_ops[op_nr]; break; @@ -753,7 +798,6 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op) case OP_CB_PUSH_DELEG: case OP_CB_RECALLABLE_OBJ_AVAIL: case OP_CB_WANTS_CANCELLED: - case OP_CB_NOTIFY_LOCK: return htonl(NFS4ERR_NOTSUPP); default: @@ -1006,6 +1050,11 @@ static struct callback_op callback_ops[] = { .decode_args = (callback_decode_arg_t)decode_recallslot_args, .res_maxsize = CB_OP_RECALLSLOT_RES_MAXSZ, }, + [OP_CB_NOTIFY_LOCK] = { + .process_op = (callback_process_op_t)nfs4_callback_notify_lock, + .decode_args = (callback_decode_arg_t)decode_notify_lock_args, + .res_maxsize = CB_OP_NOTIFY_LOCK_RES_MAXSZ, + }, #endif /* CONFIG_NFS_V4_1 */ }; diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 1e106780a237..7555ba889d1f 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -313,7 +313,10 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat continue; /* Match the full socket address */ if (!rpc_cmp_addr_port(sap, clap)) - continue; + /* Match all xprt_switch full socket addresses */ + if (!rpc_clnt_xprt_switch_has_addr(clp->cl_rpcclient, + sap)) + continue; atomic_inc(&clp->cl_count); return clp; @@ -785,7 +788,8 @@ int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, struct nfs } fsinfo.fattr = fattr; - fsinfo.layouttype = 0; + fsinfo.nlayouttypes = 0; + memset(fsinfo.layouttype, 0, sizeof(fsinfo.layouttype)); error = clp->rpc_ops->fsinfo(server, mntfh, &fsinfo); if (error < 0) goto out_error; @@ -1078,7 +1082,7 @@ void nfs_clients_init(struct net *net) idr_init(&nn->cb_ident_idr); #endif spin_lock_init(&nn->nfs_client_lock); - nn->boot_time = CURRENT_TIME; + nn->boot_time = ktime_get_real(); } #ifdef CONFIG_PROC_FS diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 322c2585bc34..dff600ae0d74 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -41,6 +41,17 @@ void nfs_mark_delegation_referenced(struct nfs_delegation *delegation) set_bit(NFS_DELEGATION_REFERENCED, &delegation->flags); } +static bool +nfs4_is_valid_delegation(const struct nfs_delegation *delegation, + fmode_t flags) +{ + if (delegation != NULL && (delegation->type & flags) == flags && + !test_bit(NFS_DELEGATION_REVOKED, &delegation->flags) && + !test_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) + return true; + return false; +} + static int nfs4_do_check_delegation(struct inode *inode, fmode_t flags, bool mark) { @@ -50,8 +61,7 @@ nfs4_do_check_delegation(struct inode *inode, fmode_t flags, bool mark) flags &= FMODE_READ|FMODE_WRITE; rcu_read_lock(); delegation = rcu_dereference(NFS_I(inode)->delegation); - if (delegation != NULL && (delegation->type & flags) == flags && - !test_bit(NFS_DELEGATION_RETURNING, &delegation->flags)) { + if (nfs4_is_valid_delegation(delegation, flags)) { if (mark) nfs_mark_delegation_referenced(delegation); ret = 1; @@ -185,15 +195,13 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, rcu_read_unlock(); put_rpccred(oldcred); trace_nfs4_reclaim_delegation(inode, res->delegation_type); - } else { - /* We appear to have raced with a delegation return. */ - spin_unlock(&delegation->lock); - rcu_read_unlock(); - nfs_inode_set_delegation(inode, cred, res); + return; } - } else { - rcu_read_unlock(); + /* We appear to have raced with a delegation return. */ + spin_unlock(&delegation->lock); } + rcu_read_unlock(); + nfs_inode_set_delegation(inode, cred, res); } static int nfs_do_return_delegation(struct inode *inode, struct nfs_delegation *delegation, int issync) @@ -642,28 +650,49 @@ static void nfs_client_mark_return_unused_delegation_types(struct nfs_client *cl rcu_read_unlock(); } -static void nfs_revoke_delegation(struct inode *inode) +static void nfs_mark_delegation_revoked(struct nfs_server *server, + struct nfs_delegation *delegation) +{ + set_bit(NFS_DELEGATION_REVOKED, &delegation->flags); + delegation->stateid.type = NFS4_INVALID_STATEID_TYPE; + nfs_mark_return_delegation(server, delegation); +} + +static bool nfs_revoke_delegation(struct inode *inode, + const nfs4_stateid *stateid) { struct nfs_delegation *delegation; + nfs4_stateid tmp; + bool ret = false; + rcu_read_lock(); delegation = rcu_dereference(NFS_I(inode)->delegation); - if (delegation != NULL) { - set_bit(NFS_DELEGATION_REVOKED, &delegation->flags); - nfs_mark_return_delegation(NFS_SERVER(inode), delegation); - } + if (delegation == NULL) + goto out; + if (stateid == NULL) { + nfs4_stateid_copy(&tmp, &delegation->stateid); + stateid = &tmp; + } else if (!nfs4_stateid_match(stateid, &delegation->stateid)) + goto out; + nfs_mark_delegation_revoked(NFS_SERVER(inode), delegation); + ret = true; +out: rcu_read_unlock(); + if (ret) + nfs_inode_find_state_and_recover(inode, stateid); + return ret; } -void nfs_remove_bad_delegation(struct inode *inode) +void nfs_remove_bad_delegation(struct inode *inode, + const nfs4_stateid *stateid) { struct nfs_delegation *delegation; - nfs_revoke_delegation(inode); + if (!nfs_revoke_delegation(inode, stateid)) + return; delegation = nfs_inode_detach_delegation(inode); - if (delegation) { - nfs_inode_find_state_and_recover(inode, &delegation->stateid); + if (delegation) nfs_free_delegation(delegation); - } } EXPORT_SYMBOL_GPL(nfs_remove_bad_delegation); @@ -786,8 +815,15 @@ static void nfs_delegation_mark_reclaim_server(struct nfs_server *server) { struct nfs_delegation *delegation; - list_for_each_entry_rcu(delegation, &server->delegations, super_list) + list_for_each_entry_rcu(delegation, &server->delegations, super_list) { + /* + * If the delegation may have been admin revoked, then we + * cannot reclaim it. + */ + if (test_bit(NFS_DELEGATION_TEST_EXPIRED, &delegation->flags)) + continue; set_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags); + } } /** @@ -851,6 +887,141 @@ restart: rcu_read_unlock(); } +static inline bool nfs4_server_rebooted(const struct nfs_client *clp) +{ + return (clp->cl_state & (BIT(NFS4CLNT_CHECK_LEASE) | + BIT(NFS4CLNT_LEASE_EXPIRED) | + BIT(NFS4CLNT_SESSION_RESET))) != 0; +} + +static void nfs_mark_test_expired_delegation(struct nfs_server *server, + struct nfs_delegation *delegation) +{ + if (delegation->stateid.type == NFS4_INVALID_STATEID_TYPE) + return; + clear_bit(NFS_DELEGATION_NEED_RECLAIM, &delegation->flags); + set_bit(NFS_DELEGATION_TEST_EXPIRED, &delegation->flags); + set_bit(NFS4CLNT_DELEGATION_EXPIRED, &server->nfs_client->cl_state); +} + +static void nfs_inode_mark_test_expired_delegation(struct nfs_server *server, + struct inode *inode) +{ + struct nfs_delegation *delegation; + + rcu_read_lock(); + delegation = rcu_dereference(NFS_I(inode)->delegation); + if (delegation) + nfs_mark_test_expired_delegation(server, delegation); + rcu_read_unlock(); + +} + +static void nfs_delegation_mark_test_expired_server(struct nfs_server *server) +{ + struct nfs_delegation *delegation; + + list_for_each_entry_rcu(delegation, &server->delegations, super_list) + nfs_mark_test_expired_delegation(server, delegation); +} + +/** + * nfs_mark_test_expired_all_delegations - mark all delegations for testing + * @clp: nfs_client to process + * + * Iterates through all the delegations associated with this server and + * marks them as needing to be checked for validity. + */ +void nfs_mark_test_expired_all_delegations(struct nfs_client *clp) +{ + struct nfs_server *server; + + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + nfs_delegation_mark_test_expired_server(server); + rcu_read_unlock(); +} + |