diff options
Diffstat (limited to 'fs/nfs')
34 files changed, 5168 insertions, 2322 deletions
diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile index ec61fd56a1a..f4580b44eef 100644 --- a/fs/nfs/Makefile +++ b/fs/nfs/Makefile @@ -4,14 +4,16 @@ obj-$(CONFIG_NFS_FS) += nfs.o -nfs-y := dir.o file.o inode.o nfs2xdr.o pagelist.o \ - proc.o read.o symlink.o unlink.o write.o +nfs-y := client.o dir.o file.o getroot.o inode.o super.o nfs2xdr.o \ + pagelist.o proc.o read.o symlink.o unlink.o \ + write.o namespace.o nfs-$(CONFIG_ROOT_NFS) += nfsroot.o mount_clnt.o nfs-$(CONFIG_NFS_V3) += nfs3proc.o nfs3xdr.o nfs-$(CONFIG_NFS_V3_ACL) += nfs3acl.o nfs-$(CONFIG_NFS_V4) += nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o \ delegation.o idmap.o \ - callback.o callback_xdr.o callback_proc.o + callback.o callback_xdr.o callback_proc.o \ + nfs4namespace.o nfs-$(CONFIG_NFS_DIRECTIO) += direct.o nfs-$(CONFIG_SYSCTL) += sysctl.o nfs-objs := $(nfs-y) diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 90c95adc8c1..a3ee11364db 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -6,7 +6,6 @@ * NFSv4 callback handling */ -#include <linux/config.h> #include <linux/completion.h> #include <linux/ip.h> #include <linux/module.h> @@ -20,6 +19,7 @@ #include "nfs4_fs.h" #include "callback.h" +#include "internal.h" #define NFSDBG_FACILITY NFSDBG_CALLBACK @@ -37,6 +37,21 @@ static struct svc_program nfs4_callback_program; unsigned int nfs_callback_set_tcpport; unsigned short nfs_callback_tcpport; +static const int nfs_set_port_min = 0; +static const int nfs_set_port_max = 65535; + +static int param_set_port(const char *val, struct kernel_param *kp) +{ + char *endp; + int num = simple_strtol(val, &endp, 0); + if (endp == val || *endp || num < nfs_set_port_min || num > nfs_set_port_max) + return -EINVAL; + *((int *)kp->arg) = num; + return 0; +} + +module_param_call(callback_tcpport, param_set_port, param_get_int, + &nfs_callback_set_tcpport, 0644); /* * This is the callback kernel thread. @@ -135,10 +150,8 @@ out_err: /* * Kill the server process if it is not already up. */ -int nfs_callback_down(void) +void nfs_callback_down(void) { - int ret = 0; - lock_kernel(); mutex_lock(&nfs_callback_mutex); nfs_callback_info.users--; @@ -150,20 +163,19 @@ int nfs_callback_down(void) } while (wait_for_completion_timeout(&nfs_callback_info.stopped, 5*HZ) == 0); mutex_unlock(&nfs_callback_mutex); unlock_kernel(); - return ret; } static int nfs_callback_authenticate(struct svc_rqst *rqstp) { - struct in_addr *addr = &rqstp->rq_addr.sin_addr; - struct nfs4_client *clp; + struct sockaddr_in *addr = &rqstp->rq_addr; + struct nfs_client *clp; /* Don't talk to strangers */ - clp = nfs4_find_client(addr); + clp = nfs_find_client(addr, 4); if (clp == NULL) return SVC_DROP; - dprintk("%s: %u.%u.%u.%u NFSv4 callback!\n", __FUNCTION__, NIPQUAD(addr)); - nfs4_put_client(clp); + dprintk("%s: %u.%u.%u.%u NFSv4 callback!\n", __FUNCTION__, NIPQUAD(addr->sin_addr)); + nfs_put_client(clp); switch (rqstp->rq_authop->flavour) { case RPC_AUTH_NULL: if (rqstp->rq_proc != CB_NULL) @@ -182,8 +194,6 @@ static int nfs_callback_authenticate(struct svc_rqst *rqstp) /* * Define NFS4 callback program */ -extern struct svc_version nfs4_callback_version1; - static struct svc_version *nfs4_callback_version[] = { [1] = &nfs4_callback_version1, }; diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index b252e7fe53a..5676163d26e 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -62,8 +62,13 @@ struct cb_recallargs { extern unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res); extern unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy); +#ifdef CONFIG_NFS_V4 extern int nfs_callback_up(void); -extern int nfs_callback_down(void); +extern void nfs_callback_down(void); +#else +#define nfs_callback_up() (0) +#define nfs_callback_down() do {} while(0) +#endif extern unsigned int nfs_callback_set_tcpport; extern unsigned short nfs_callback_tcpport; diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 462cfceb50c..97cf8f71451 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -5,25 +5,25 @@ * * NFSv4 callback procedures */ -#include <linux/config.h> #include <linux/nfs4.h> #include <linux/nfs_fs.h> #include "nfs4_fs.h" #include "callback.h" #include "delegation.h" +#include "internal.h" #define NFSDBG_FACILITY NFSDBG_CALLBACK unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res) { - struct nfs4_client *clp; + struct nfs_client *clp; struct nfs_delegation *delegation; struct nfs_inode *nfsi; struct inode *inode; res->bitmap[0] = res->bitmap[1] = 0; res->status = htonl(NFS4ERR_BADHANDLE); - clp = nfs4_find_client(&args->addr->sin_addr); + clp = nfs_find_client(args->addr, 4); if (clp == NULL) goto out; inode = nfs_delegation_find_inode(clp, &args->fh); @@ -49,7 +49,7 @@ out_iput: up_read(&nfsi->rwsem); iput(inode); out_putclient: - nfs4_put_client(clp); + nfs_put_client(clp); out: dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(res->status)); return res->status; @@ -57,12 +57,12 @@ out: unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy) { - struct nfs4_client *clp; + struct nfs_client *clp; struct inode *inode; unsigned res; res = htonl(NFS4ERR_BADHANDLE); - clp = nfs4_find_client(&args->addr->sin_addr); + clp = nfs_find_client(args->addr, 4); if (clp == NULL) goto out; inode = nfs_delegation_find_inode(clp, &args->fh); @@ -81,7 +81,7 @@ unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy) } iput(inode); out_putclient: - nfs4_put_client(clp); + nfs_put_client(clp); out: dprintk("%s: exit with status = %d\n", __FUNCTION__, ntohl(res)); return res; diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 05c38cf40b6..29f93219205 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -5,7 +5,6 @@ * * NFSv4 callback encode/decode procedures */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/sunrpc/svc.h> #include <linux/nfs4.h> @@ -202,7 +201,7 @@ static unsigned decode_recall_args(struct svc_rqst *rqstp, struct xdr_stream *xd status = decode_fh(xdr, &args->fh); out: dprintk("%s: exit with status = %d\n", __FUNCTION__, status); - return 0; + return status; } static unsigned encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) diff --git a/fs/nfs/client.c b/fs/nfs/client.c new file mode 100644 index 00000000000..ec1938d4b81 --- /dev/null +++ b/fs/nfs/client.c @@ -0,0 +1,1448 @@ +/* client.c: NFS client sharing and management code + * + * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/init.h> + +#include <linux/time.h> +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/string.h> +#include <linux/stat.h> +#include <linux/errno.h> +#include <linux/unistd.h> +#include <linux/sunrpc/clnt.h> +#include <linux/sunrpc/stats.h> +#include <linux/sunrpc/metrics.h> +#include <linux/nfs_fs.h> +#include <linux/nfs_mount.h> +#include <linux/nfs4_mount.h> +#include <linux/lockd/bind.h> +#include <linux/smp_lock.h> +#include <linux/seq_file.h> +#include <linux/mount.h> +#include <linux/nfs_idmap.h> +#include <linux/vfs.h> +#include <linux/inet.h> +#include <linux/nfs_xdr.h> + +#include <asm/system.h> + +#include "nfs4_fs.h" +#include "callback.h" +#include "delegation.h" +#include "iostat.h" +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_CLIENT + +static DEFINE_SPINLOCK(nfs_client_lock); +static LIST_HEAD(nfs_client_list); +static LIST_HEAD(nfs_volume_list); +static DECLARE_WAIT_QUEUE_HEAD(nfs_client_active_wq); + +/* + * RPC cruft for NFS + */ +static struct rpc_version *nfs_version[5] = { + [2] = &nfs_version2, +#ifdef CONFIG_NFS_V3 + [3] = &nfs_version3, +#endif +#ifdef CONFIG_NFS_V4 + [4] = &nfs_version4, +#endif +}; + +struct rpc_program nfs_program = { + .name = "nfs", + .number = NFS_PROGRAM, + .nrvers = ARRAY_SIZE(nfs_version), + .version = nfs_version, + .stats = &nfs_rpcstat, + .pipe_dir_name = "/nfs", +}; + +struct rpc_stat nfs_rpcstat = { + .program = &nfs_program +}; + + +#ifdef CONFIG_NFS_V3_ACL +static struct rpc_stat nfsacl_rpcstat = { &nfsacl_program }; +static struct rpc_version * nfsacl_version[] = { + [3] = &nfsacl_version3, +}; + +struct rpc_program nfsacl_program = { + .name = "nfsacl", + .number = NFS_ACL_PROGRAM, + .nrvers = ARRAY_SIZE(nfsacl_version), + .version = nfsacl_version, + .stats = &nfsacl_rpcstat, +}; +#endif /* CONFIG_NFS_V3_ACL */ + +/* + * Allocate a shared client record + * + * Since these are allocated/deallocated very rarely, we don't + * bother putting them in a slab cache... + */ +static struct nfs_client *nfs_alloc_client(const char *hostname, + const struct sockaddr_in *addr, + int nfsversion) +{ + struct nfs_client *clp; + int error; + + if ((clp = kzalloc(sizeof(*clp), GFP_KERNEL)) == NULL) + goto error_0; + + error = rpciod_up(); + if (error < 0) { + dprintk("%s: couldn't start rpciod! Error = %d\n", + __FUNCTION__, error); + goto error_1; + } + __set_bit(NFS_CS_RPCIOD, &clp->cl_res_state); + + if (nfsversion == 4) { + if (nfs_callback_up() < 0) + goto error_2; + __set_bit(NFS_CS_CALLBACK, &clp->cl_res_state); + } + + atomic_set(&clp->cl_count, 1); + clp->cl_cons_state = NFS_CS_INITING; + + clp->cl_nfsversion = nfsversion; + memcpy(&clp->cl_addr, addr, sizeof(clp->cl_addr)); + + if (hostname) { + clp->cl_hostname = kstrdup(hostname, GFP_KERNEL); + if (!clp->cl_hostname) + goto error_3; + } + + INIT_LIST_HEAD(&clp->cl_superblocks); + clp->cl_rpcclient = ERR_PTR(-EINVAL); + +#ifdef CONFIG_NFS_V4 + init_rwsem(&clp->cl_sem); + INIT_LIST_HEAD(&clp->cl_delegations); + INIT_LIST_HEAD(&clp->cl_state_owners); + INIT_LIST_HEAD(&clp->cl_unused); + spin_lock_init(&clp->cl_lock); + INIT_WORK(&clp->cl_renewd, nfs4_renew_state, clp); + rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client"); + clp->cl_boot_time = CURRENT_TIME; + clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED; +#endif + + return clp; + +error_3: + if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state)) + nfs_callback_down(); +error_2: + rpciod_down(); + __clear_bit(NFS_CS_RPCIOD, &clp->cl_res_state); +error_1: + kfree(clp); +error_0: + return NULL; +} + +static void nfs4_shutdown_client(struct nfs_client *clp) +{ +#ifdef CONFIG_NFS_V4 + if (__test_and_clear_bit(NFS_CS_RENEWD, &clp->cl_res_state)) + nfs4_kill_renewd(clp); + while (!list_empty(&clp->cl_unused)) { + struct nfs4_state_owner *sp; + + sp = list_entry(clp->cl_unused.next, + struct nfs4_state_owner, + so_list); + list_del(&sp->so_list); + kfree(sp); + } + BUG_ON(!list_empty(&clp->cl_state_owners)); + if (__test_and_clear_bit(NFS_CS_IDMAP, &clp->cl_res_state)) + nfs_idmap_delete(clp); +#endif +} + +/* + * Destroy a shared client record + */ +static void nfs_free_client(struct nfs_client *clp) +{ + dprintk("--> nfs_free_client(%d)\n", clp->cl_nfsversion); + + nfs4_shutdown_client(clp); + + /* -EIO all pending I/O */ + if (!IS_ERR(clp->cl_rpcclient)) + rpc_shutdown_client(clp->cl_rpcclient); + + if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state)) + nfs_callback_down(); + + if (__test_and_clear_bit(NFS_CS_RPCIOD, &clp->cl_res_state)) + rpciod_down(); + + kfree(clp->cl_hostname); + kfree(clp); + + dprintk("<-- nfs_free_client()\n"); +} + +/* + * Release a reference to a shared client record + */ +void nfs_put_client(struct nfs_client *clp) +{ + if (!clp) + return; + + dprintk("--> nfs_put_client({%d})\n", atomic_read(&clp->cl_count)); + + if (atomic_dec_and_lock(&clp->cl_count, &nfs_client_lock)) { + list_del(&clp->cl_share_link); + spin_unlock(&nfs_client_lock); + + BUG_ON(!list_empty(&clp->cl_superblocks)); + + nfs_free_client(clp); + } +} + +/* + * Find a client by address + * - caller must hold nfs_client_lock + */ +static struct nfs_client *__nfs_find_client(const struct sockaddr_in *addr, int nfsversion) +{ + struct nfs_client *clp; + + list_for_each_entry(clp, &nfs_client_list, cl_share_link) { + /* Different NFS versions cannot share the same nfs_client */ + if (clp->cl_nfsversion != nfsversion) + continue; + + if (memcmp(&clp->cl_addr.sin_addr, &addr->sin_addr, + sizeof(clp->cl_addr.sin_addr)) != 0) + continue; + + if (clp->cl_addr.sin_port == addr->sin_port) + goto found; + } + + return NULL; + +found: + atomic_inc(&clp->cl_count); + return clp; +} + +/* + * Find a client by IP address and protocol version + * - returns NULL if no such client + */ +struct nfs_client *nfs_find_client(const struct sockaddr_in *addr, int nfsversion) +{ + struct nfs_client *clp; + + spin_lock(&nfs_client_lock); + clp = __nfs_find_client(addr, nfsversion); + spin_unlock(&nfs_client_lock); + + BUG_ON(clp && clp->cl_cons_state == 0); + + return clp; +} + +/* + * Look up a client by IP address and protocol version + * - creates a new record if one doesn't yet exist + */ +static struct nfs_client *nfs_get_client(const char *hostname, + const struct sockaddr_in *addr, + int nfsversion) +{ + struct nfs_client *clp, *new = NULL; + int error; + + dprintk("--> nfs_get_client(%s,"NIPQUAD_FMT":%d,%d)\n", + hostname ?: "", NIPQUAD(addr->sin_addr), + addr->sin_port, nfsversion); + + /* see if the client already exists */ + do { + spin_lock(&nfs_client_lock); + + clp = __nfs_find_client(addr, nfsversion); + if (clp) + goto found_client; + if (new) + goto install_client; + + spin_unlock(&nfs_client_lock); + + new = nfs_alloc_client(hostname, addr, nfsversion); + } while (new); + + return ERR_PTR(-ENOMEM); + + /* install a new client and return with it unready */ +install_client: + clp = new; + list_add(&clp->cl_share_link, &nfs_client_list); + spin_unlock(&nfs_client_lock); + dprintk("--> nfs_get_client() = %p [new]\n", clp); + return clp; + + /* found an existing client + * - make sure it's ready before returning + */ +found_client: + spin_unlock(&nfs_client_lock); + + if (new) + nfs_free_client(new); + + if (clp->cl_cons_state == NFS_CS_INITING) { + DECLARE_WAITQUEUE(myself, current); + + add_wait_queue(&nfs_client_active_wq, &myself); + + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + if (signal_pending(current) || + clp->cl_cons_state > NFS_CS_READY) + break; + schedule(); + } + + remove_wait_queue(&nfs_client_active_wq, &myself); + + if (signal_pending(current)) { + nfs_put_client(clp); + return ERR_PTR(-ERESTARTSYS); + } + } + + if (clp->cl_cons_state < NFS_CS_READY) { + error = clp->cl_cons_state; + nfs_put_client(clp); + return ERR_PTR(error); + } + + BUG_ON(clp->cl_cons_state != NFS_CS_READY); + + dprintk("--> nfs_get_client() = %p [share]\n", clp); + return clp; +} + +/* + * Mark a server as ready or failed + */ +static void nfs_mark_client_ready(struct nfs_client *clp, int state) +{ + clp->cl_cons_state = state; + wake_up_all(&nfs_client_active_wq); +} + +/* + * Initialise the timeout values for a connection + */ +static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, + unsigned int timeo, unsigned int retrans) +{ + to->to_initval = timeo * HZ / 10; + to->to_retries = retrans; + if (!to->to_retries) + to->to_retries = 2; + + switch (proto) { + case IPPROTO_TCP: + if (!to->to_initval) + to->to_initval = 60 * HZ; + if (to->to_initval > NFS_MAX_TCP_TIMEOUT) + to->to_initval = NFS_MAX_TCP_TIMEOUT; + to->to_increment = to->to_initval; + to->to_maxval = to->to_initval + (to->to_increment * to->to_retries); + to->to_exponential = 0; + break; + case IPPROTO_UDP: + default: + if (!to->to_initval) + to->to_initval = 11 * HZ / 10; + if (to->to_initval > NFS_MAX_UDP_TIMEOUT) + to->to_initval = NFS_MAX_UDP_TIMEOUT; + to->to_maxval = NFS_MAX_UDP_TIMEOUT; + to->to_exponential = 1; + break; + } +} + +/* + * Create an RPC client handle + */ +static int nfs_create_rpc_client(struct nfs_client *clp, int proto, + unsigned int timeo, + unsigned int retrans, + rpc_authflavor_t flavor) +{ + struct rpc_timeout timeparms; + struct rpc_clnt *clnt = NULL; + struct rpc_create_args args = { + .protocol = proto, + .address = (struct sockaddr *)&clp->cl_addr, + .addrsize = sizeof(clp->cl_addr), + .timeout = &timeparms, + .servername = clp->cl_hostname, + .program = &nfs_program, + .version = clp->rpc_ops->version, + .authflavor = flavor, + }; + + if (!IS_ERR(clp->cl_rpcclient)) + return 0; + + nfs_init_timeout_values(&timeparms, proto, timeo, retrans); + clp->retrans_timeo = timeparms.to_initval; + clp->retrans_count = timeparms.to_retries; + + clnt = rpc_create(&args); + if (IS_ERR(clnt)) { + dprintk("%s: cannot create RPC client. Error = %ld\n", + __FUNCTION__, PTR_ERR(clnt)); + return PTR_ERR(clnt); + } + + clp->cl_rpcclient = clnt; + return 0; +} + +/* + * Version 2 or 3 client destruction + */ +static void nfs_destroy_server(struct nfs_server *server) +{ + if (!IS_ERR(server->client_acl)) + rpc_shutdown_client(server->client_acl); + + if (!(server->flags & NFS_MOUNT_NONLM)) + lockd_down(); /* release rpc.lockd */ +} + +/* + * Version 2 or 3 lockd setup + */ +static int nfs_start_lockd(struct nfs_server *server) +{ + int error = 0; + + if (server->nfs_client->cl_nfsversion > 3) + goto out; + if (server->flags & NFS_MOUNT_NONLM) + goto out; + error = lockd_up(); + if (error < 0) + server->flags |= NFS_MOUNT_NONLM; + else + server->destroy = nfs_destroy_server; +out: + return error; +} + +/* + * Initialise an NFSv3 ACL client connection + */ +#ifdef CONFIG_NFS_V3_ACL +static void nfs_init_server_aclclient(struct nfs_server *server) +{ + if (server->nfs_client->cl_nfsversion != 3) + goto out_noacl; + if (server->flags & NFS_MOUNT_NOACL) + goto out_noacl; + + server->client_acl = rpc_bind_new_program(server->client, &nfsacl_program, 3); + if (IS_ERR(server->client_acl)) + goto out_noacl; + + /* No errors! Assume that Sun nfsacls are supported */ + server->caps |= NFS_CAP_ACLS; + return; + +out_noacl: + server->caps &= ~NFS_CAP_ACLS; +} +#else +static inline void nfs_init_server_aclclient(struct nfs_server *server) +{ + server->flags &= ~NFS_MOUNT_NOACL; + server->caps &= ~NFS_CAP_ACLS; +} +#endif + +/* + * Create a general RPC client + */ +static int nfs_init_server_rpcclient(struct nfs_server *server, rpc_authflavor_t pseudoflavour) +{ + struct nfs_client *clp = server->nfs_client; + + server->client = rpc_clone_client(clp->cl_rpcclient); + if (IS_ERR(server->client)) { + dprintk("%s: couldn't create rpc_client!\n", __FUNCTION__); + return PTR_ERR(server->client); + } + + if (pseudoflavour != clp->cl_rpcclient->cl_auth->au_flavor) { + struct rpc_auth *auth; + + auth = rpcauth_create(pseudoflavour, server->client); + if (IS_ERR(auth)) { + dprintk("%s: couldn't create credcache!\n", __FUNCTION__); + return PTR_ERR(auth); + } + } + server->client->cl_softrtry = 0; + if (server->flags & NFS_MOUNT_SOFT) + server->client->cl_softrtry = 1; + + server->client->cl_intr = 0; + if (server->flags & NFS4_MOUNT_INTR) + server->client->cl_intr = 1; + + return 0; +} + +/* + * Initialise an NFS2 or NFS3 client + */ +static int nfs_init_client(struct nfs_client *clp, const struct nfs_mount_data *data) +{ + int proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP; + int error; + + if (clp->cl_cons_state == NFS_CS_READY) { + /* the client is already initialised */ + dprintk("<-- nfs_init_client() = 0 [already %p]\n", clp); + return 0; + } + + /* Check NFS protocol revision and initialize RPC op vector */ + clp->rpc_ops = &nfs_v2_clientops; +#ifdef CONFIG_NFS_V3 + if (clp->cl_nfsversion == 3) + clp->rpc_ops = &nfs_v3_clientops; +#endif + /* + * Create a client RPC handle for doing FSSTAT with UNIX auth only + * - RFC 2623, sec 2.3.2 + */ + error = nfs_create_rpc_client(clp, proto, data->timeo, data->retrans, + RPC_AUTH_UNIX); + if (error < 0) + goto error; + nfs_mark_client_ready(clp, NFS_CS_READY); + return 0; + +error: + nfs_mark_client_ready(clp, error); + dprintk("<-- nfs_init_client() = xerror %d\n", error); + return error; +} + +/* + * Create a version 2 or 3 client + */ +static int nfs_init_server(struct nfs_server *server, const struct nfs_mount_data *data) +{ + struct nfs_client *clp; + int error, nfsvers = 2; + + dprintk("--> nfs_init_server()\n"); + +#ifdef CONFIG_NFS_V3 + if (data->flags & NFS_MOUNT_VER3) + nfsvers = 3; +#endif + + /* Allocate or find a client reference we can use */ + clp = nfs_get_client(data->hostname, &data->addr, nfsvers); + if (IS_ERR(clp)) { + dprintk("<-- nfs_init_server() = error %ld\n", PTR_ERR(clp)); + return PTR_ERR(clp); + } + + error = nfs_init_client(clp, data); + if (error < 0) + goto error; + + server->nfs_client = clp; + + /* Initialise the client representation from the mount data */ + server->flags = data->flags & NFS_MOUNT_FLAGMASK; + + if (data->rsize) + server->rsize = nfs_block_size(data->rsize, NULL); + if (data->wsize) + server->wsize = nfs_block_size(data->wsize, NULL); + + server->acregmin = data->acregmin * HZ; + server->acregmax = data->acregmax * HZ; + server->acdirmin = data->acdirmin * HZ; + server->acdirmax = data->acdirmax * HZ; + + /* Start lockd here, before we might error out */ + error = nfs_start_lockd(server); + if (error < 0) + goto error; + + error = nfs_init_server_rpcclient(server, data->pseudoflavor); + if (error < 0) + goto error; + + server->namelen = data->namlen; + /* Create a client RPC handle for the NFSv3 ACL management interface */ + nfs_init_server_aclclient(server); + if (clp->cl_nfsversion == 3) { + if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN) + server->namelen = NFS3_MAXNAMLEN; + server->caps |= NFS_CAP_READDIRPLUS; + } else { + if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN) + server->namelen = NFS2_MAXNAMLEN; + } + + dprintk("<-- nfs_init_server() = 0 [new %p]\n", clp); + return 0; + +error: + server->nfs_client = NULL; + nfs_put_client(clp); + dprintk("<-- nfs_init_server() = xerror %d\n", error); + return error; +} + +/* + * Load up the server record from information gained in an fsinfo record + */ +static void nfs_server_set_fsinfo(struct nfs_server *server, struct nfs_fsinfo *fsinfo) +{ + unsigned long max_rpc_payload; + + /* Work out a lot of parameters */ + if (server->rsize == 0) + server->rsize = nfs_block_size(fsinfo->rtpref, NULL); + if (server->wsize == 0) + server->wsize = nfs_block_size(fsinfo->wtpref, NULL); + + if (fsinfo->rtmax >= 512 && server->rsize > fsinfo->rtmax) + server->rsize = nfs_block_size(fsinfo->rtmax, NULL); + if (fsinfo->wtmax >= 512 && server->wsize > fsinfo->wtmax) + server->wsize = nfs_block_size(fsinfo->wtmax, NULL); + + max_rpc_payload = nfs_block_size(rpc_max_payload(server->client), NULL); + if (server->rsize > max_rpc_payload) + server->rsize = max_rpc_payload; + if (server->rsize > NFS_MAX_FILE_IO_SIZE) + server->rsize = NFS_MAX_FILE_IO_SIZE; + server->rpages = (server->rsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + server->backing_dev_info.ra_pages = server->rpages * NFS_MAX_READAHEAD; + + if (server->wsize > max_rpc_payload) + server->wsize = max_rpc_payload; + if (server->wsize > NFS_MAX_FILE_IO_SIZE) + server->wsize = NFS_MAX_FILE_IO_SIZE; + server->wpages = (server->wsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + server->wtmult = nfs_block_bits(fsinfo->wtmult, NULL); + + server->dtsize = nfs_block_size(fsinfo->dtpref, NULL); + if (server->dtsize > PAGE_CACHE_SIZE) + server->dtsize = PAGE_CACHE_SIZE; + if (server->dtsize > server->rsize) + server->dtsize = server->rsize; + + if (server->flags & NFS_MOUNT_NOAC) { + server->acregmin = server->acregmax = 0; + server->acdirmin = server->acdirmax = 0; + } + + server->maxfilesize = fsinfo->maxfilesize; + + /* We're airborne Set socket buffersize */ + rpc_setbufsize(server->client, server->wsize + 100, server->rsize + 100); +} + +/* + * Probe filesystem information, including the FSID on v2/v3 + */ +static int nfs_probe_fsinfo(struct nfs_server *server, struct nfs_fh *mntfh, struct nfs_fattr *fattr) +{ + struct nfs_fsinfo fsinfo; + struct nfs_client *clp = server->nfs_client; + int error; + + dprintk("--> nfs_probe_fsinfo()\n"); + + if (clp->rpc_ops->set_capabilities != NULL) { + error = clp->rpc_ops->set_capabilities(server, mntfh); + if (error < 0) + goto out_error; + } + + fsinfo.fattr = fattr; + nfs_fattr_init(fattr); + error = clp->rpc_ops->fsinfo(server, mntfh, &fsinfo); + if (error < 0) + goto out_error; + + nfs_server_set_fsinfo(server, &fsinfo); + + /* Get some general file system info */ + if (server->namelen == 0) { + struct nfs_pathconf pathinfo; + + pathinfo.fattr = fattr; + nfs_fattr_init(fattr); + + if (clp->rpc_ops->pathconf(server, mntfh, &pathinfo) >= 0) + server->namelen = pathinfo.max_namelen; + } + + dprintk("<-- nfs_probe_fsinfo() = 0\n"); + return 0; + +out_error: + dprintk("nfs_probe_fsinfo: error = %d\n", -error); + return error; +} + +/* + * Copy useful information when duplicating a server record + */ +static void nfs_server_copy_userdata(struct nfs_server *target, struct nfs_server *source) +{ + target->flags = source->flags; + target->acregmin = source->acregmin; + target->acregmax = source->acregmax; + target->acdirmin = source->acdirmin; + target->acdirmax = source->acdirmax; + target->caps = source->caps; +} + +/* + * Allocate and initialise a server record + */ +static struct nfs_server *nfs_alloc_server(void) +{ + struct nfs_server *server; + + server = kzalloc(sizeof(struct nfs_server), GFP_KERNEL); + if (!server) + return NULL; + + server->client = server->client_acl = ERR_PTR(-EINVAL); + + /* Zero out the NFS state stuff */ + INIT_LIST_HEAD(&server->client_link); + INIT_LIST_HEAD(&server->master_link); + + server->io_stats = nfs_alloc_iostats(); + if (!server->io_stats) { + kfree(server); + return NULL; + } + + return server; +} + +/* + * Free up a server record + */ +void nfs_free_server(struct nfs_server *server) +{ + dprintk("--> nfs_free_server()\n"); + + spin_lock(&nfs_client_lock); + list_del(&server->client_link); + list_del(&server->master_link); + spin_unlock(&nfs_client_lock); + + if (server->destroy != NULL) + server->destroy(server); + if (!IS_ERR(server->client)) + rpc_shutdown_client(server->client); + + nfs_put_client(server->nfs_client); + + nfs_free_iostats(server->io_stats); + kfree(server); + nfs_release_automount_timer(); + dprintk("<-- nfs_free_server()\n"); +} + +/* + * Create a version 2 or 3 volume record + * - keyed on server and FSID + */ +struct nfs_server *nfs_create_server(const struct nfs_mount_data *data, + struct nfs_fh *mntfh) +{ + struct nfs_server *server; + struct nfs_fattr fattr; + int error; + + server = nfs_alloc_server(); + if (!server) + return ERR_PTR(-ENOMEM); + + /* Get a client representation */ + error = nfs_init_server(server, data); + if (error < 0) + goto error; + + BUG_ON(!server->nfs_client); + BUG_ON(!server->nfs_client->rpc_ops); + BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops); + + /* Probe the root fh to retrieve its FSID */ + error = nfs_probe_fsinfo(server, mntfh, &fattr); + if (error < 0) + goto error; + if (!(fattr.valid & NFS_ATTR_FATTR)) { + error = server->nfs_client->rpc_ops->getattr(server, mntfh, &fattr); + if (error < 0) { + dprintk("nfs_create_server: getattr error = %d\n", -error); + goto error; + } + } + memcpy(&server->fsid, &fattr.fsid, sizeof(server->fsid)); + + dprintk("Server FSID: %llx:%llx\n", + (unsigned long long) server->fsid.major, + (unsigned long long) server->fsid.minor); + + BUG_ON(!server->nfs_client); + BUG_ON(!server->nfs_client->rpc_ops); + BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops); + + spin_lock(&nfs_client_lock); + list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks); + list_add_tail(&server->master_link, &nfs_volume_list); + spin_unlock(&nfs_client_lock); + + server->mount_time = jiffies; + return server; + +error: + nfs_free_server(server); + return ERR_PTR(error); +} + +#ifdef CONFIG_NFS_V4 +/* + * Initialise an NFS4 client record + */ +static int nfs4_init_client(struct nfs_client *clp, + int proto, int timeo, int retrans, + rpc_authflavor_t authflavour) +{ + int error; + + if (clp->cl_cons_state == NFS_CS_READY) { + /* the client is initialised already */ + dprintk("<-- nfs4_init_client() = 0 [already %p]\n", clp); + return 0; + } + + /* Check NFS protocol revision and initialize RPC op vector */ + clp->rpc_ops = &nfs_v4_clientops; + + error = nfs_create_rpc_client(clp, proto, timeo, retrans, authflavour); + if (error < 0) + goto error; + + error = nfs_idmap_new(clp); + if (error < 0) { + dprintk("%s: failed to create idmapper. Error = %d\n", + __FUNCTION__, error); + goto error; + } + __set_bit(NFS_CS_IDMAP, &clp->cl_res_state); + + nfs_mark_client_ready(clp, NFS_CS_READY); + return 0; + +error: + nfs_mark_client_ready(clp, error); + dprintk("<-- nfs4_init_client() = xerror %d\n", error); + return error; +} + +/* + * Set up an NFS4 client + */ +static int nfs4_set_client(struct nfs_server *server, + const char *hostname, const struct sockaddr_in *addr, + rpc_authflavor_t authflavour, + int proto, int timeo, int retrans) +{ + struct nfs_client *clp; + int error; + + dprintk("--> nfs4_set_client()\n"); + + /* Allocate or find a client reference we can use */ + clp = nfs_get_client(hostname, addr, 4); + if (IS_ERR(clp)) { + error = PTR_ERR(clp); + goto error; + } + error = nfs4_init_client(clp, proto, timeo, retrans, authflavour); + if (error < 0) + goto error_put; + + server->nfs_client = clp; + dprintk("<-- nfs4_set_client() = 0 [new %p]\n", clp); + return 0; + +error_put: + nfs_put_client(clp); +error: + dprintk("<-- nfs4_set_client() = xerror %d\n", error); + return error; +} + +/* + * Create a version 4 volume record + */ +static int nfs4_init_server(struct nfs_server *server, + const struct nfs4_mount_data *data, rpc_authflavor_t authflavour) +{ + int error; + + dprintk("--> nfs4_init_server()\n"); + + /* Initialise the client representation from the mount data */ + server->flags = data->flags & NFS_MOUNT_FLAGMASK; + server->caps |= NFS_CAP_ATOMIC_OPEN; + + if (data->rsize) + server->rsize = nfs_block_size(data->rsize, NULL); + if (data->wsize) + server->wsize = nfs_block_size(data->wsize, NULL); + + server->acregmin = data->acregmin * HZ; + server->acregmax = data->acregmax * HZ; + server->acdirmin = data->acdirmin * HZ; + server->acdirmax = data->acdirmax * HZ; + + error = nfs_init_server_rpcclient(server, authflavour); + + /* Done */ + dprintk("<-- nfs4_init_server() = %d\n", error); + return error; +} + +/* + * Create a version 4 volume record + * - keyed on server and FSID + */ +struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *data, + const char *hostname, + const struct sockaddr_in *addr, + const char *mntpath, + const char *ip_addr, + rpc_authflavor_t authflavour, + struct nfs_fh *mntfh) +{ + struct nfs_fattr fattr; + struct nfs_server *server; + int error; + + dprintk("--> nfs4_create_server()\n"); + + server = nfs_alloc_server(); + if (!server) + return ERR_PTR(-ENOMEM); + + /* Get a client record */ + error = nfs4_set_client(server, hostname, addr, authflavour, + data->proto, data->timeo, data->retrans); + if (error < 0) + goto error; + + /* set up the general RPC client */ + error = nfs4_init_server(server, data, authflavour); + if (error < 0) + goto error; + + BUG_ON(!server->nfs_client); + BUG_ON(!server->nfs_client->rpc_ops); + BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops); + + /* Probe the root fh to retrieve its FSID */ + error = nfs4_path_walk(server, mntfh, mntpath); + if (error < 0) + goto error; + + dprintk("Server FSID: %llx:%llx\n", + (unsigned long long) server->fsid.major, + (unsigned long long) server->fsid.minor); + dprintk("Mount FH: %d\n", mntfh->size); + + error = nfs_probe_fsinfo(server, mntfh, &fattr); + if (error < 0) + goto error; + + BUG_ON(!server->nfs_client); + BUG_ON(!server->nfs_client->rpc_ops); + BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops); + + spin_lock(&nfs_client_lock); + list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks); + list_add_tail(&server->master_link, &nfs_volume_list); + spin_unlock(&nfs_client_lock); + + server->mount_time = jiffies; + dprintk("<-- nfs4_create_server() = %p\n", server); + return server; + +error: + nfs_free_server(server); + dprintk("<-- nfs4_create_server() = error %d\n", error); + return ERR_PTR(error); +} + +/* + * Create an NFS4 referral server record + */ +struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, + struct nfs_fh *fh) +{ + struct nfs_client *parent_client; + struct nfs_server *server, *parent_server; + struct nfs_fattr fattr; + int error; + + dprintk("--> nfs4_create_referral_server()\n"); + + server = nfs_alloc_server(); + if (!server) + return ERR_PTR(-ENOMEM); + + parent_server = NFS_SB(data->sb); + parent_client = parent_server->nfs_client; + + /* Get a client representation. + * Note: NFSv4 always uses TCP, */ + error = nfs4_set_client(server, data->hostname, data->addr, + data->authflavor, + parent_server->client->cl_xprt->prot, + parent_client->retrans_timeo, + parent_client->retrans_count); + if (error < 0) + goto error; + + /* Initialise the client representation from the parent server */ + nfs_server_copy_userdata(server, parent_server); + server->caps |= NFS_CAP_ATOMIC_OPEN; + + error = nfs_init_server_rpcclient(server, data->authflavor); + if (error < 0) + goto error; + + BUG_ON(!server->nfs_client); + BUG_ON(!server->nfs_client->rpc_ops); + BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops); + + /* probe the filesystem info for this server filesystem */ + error = nfs_probe_fsinfo(server, fh, &fattr); + if (error < 0) + goto error; + + dprintk("Referral FSID: %llx:%llx\n", + (unsigned long long) server->fsid.major, + (unsigned long long) server->fsid.minor); + + spin_lock(&nfs_client_lock); + list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks); + list_add_tail(&server->master_link, &nfs_volume_list); + spin_unlock(&nfs_client_lock); + + server->mount_time = jiffies; + + dprintk("<-- nfs_create_referral_server() = %p\n", server); + return server; + +error: + nfs_free_server(server); + dprintk("<-- nfs4_create_referral_server() = error %d\n", error); + return ERR_PTR(error); +} + +#endif /* CONFIG_NFS_V4 */ + +/* + * Clone an NFS2, NFS3 or NFS4 server record + */ +struct nfs_server *nfs_clone_server(struct nfs_server *source, + struct nfs_fh *fh, + struct nfs_fattr *fattr) +{ + struct nfs_server *server; + struct nfs_fattr fattr_fsinfo; + int error; + + dprintk("--> nfs_clone_server(,%llx:%llx,)\n", + (unsigned long long) fattr->fsid.major, + (unsigned long long) fattr->fsid.minor); + + server = nfs_alloc_server(); + if (!server) + return ERR_PTR(-ENOMEM); + + /* Copy data from the source */ + server->nfs_client = source->nfs_client; + atomic_inc(&server->nfs_client->cl_count); + nfs_server_copy_userdata(server, source); + + server->fsid = fattr->fsid; + + error = nfs_init_server_rpcclient(server, source->client->cl_auth->au_flavor); + if (error < 0) + goto out_free_server; + if (!IS_ERR(source->client_acl)) + nfs_init_server_aclclient(server); + + /* probe the filesystem info for this server filesystem */ + error = nfs_probe_fsinfo(server, fh, &fattr_fsinfo); + if (error < 0) + goto out_free_server; + + dprintk("Cloned FSID: %llx:%llx\n", + (unsigned long long) server->fsid.major, + (unsigned long long) server->fsid.minor); + + error = nfs_start_lockd(server); + if (error < 0) + goto out_free_server; + + spin_lock(&nfs_client_lock); + list_add_tail(&server->client_link, &server->nfs_client->cl_superblocks); + list_add_tail(&server->master_link, &nfs_volume_list); + spin_unlock(&nfs_client_lock); + + server->mount_time = jiffies; + + dprintk("<-- nfs_clone_server() = %p\n", server); + return server; + +out_free_server: + nfs_free_server(server); + dprintk("<-- nfs_clone_server() = error %d\n", error); + return ERR_PTR(error); +} + +#ifdef CONFIG_PROC_FS +static struct proc_dir_entry *proc_fs_nfs; + +static int nfs_server_list_open(struct inode *inode, struct file *file); +static void *nfs_server_list_start(struct seq_file *p, loff_t *pos); +static void *nfs_server_list_next(struct seq_file *p, void *v, loff_t *pos); +static void nfs_server_list_stop(struct seq_file *p, void *v); +static int nfs_server_list_show(struct seq_file *m, void *v); + +static struct seq_operations nfs_server_list_ops = { + .start = nfs_server_list_start, + .next = nfs_server_list_next, + .stop = nfs_server_list_stop, + .show = nfs_server_list_show, +}; + +static struct file_operations nfs_server_list_fops = { + .open = nfs_server_list_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static int nfs_volume_list_open(struct inode *inode, struct file *file); +static void *nfs_volume_list_start(struct seq_file *p, loff_t *pos); +static void *nfs_volume_list_next(struct seq_file *p, void *v, loff_t *pos); +static void nfs_volume_list_stop(struct seq_file *p, void *v); +static int nfs_volume_list_show(struct seq_file *m, void *v); + +static struct seq_operations nfs_volume_list_ops = { + .start = nfs_volume_list_start, + .next = nfs_volume_list_next, + .stop = nfs_volume_list_stop, + .show = nfs_volume_list_show, +}; + +static struct file_operations nfs_volume_list_fops = { + .open = nfs_volume_list_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +/* + * open "/proc/fs/nfsfs/servers" which provides a summary of servers with which + * we're dealing + */ +static int nfs_server_list_open(struct inode *inode, struct file *file) +{ + struct seq_file *m; + int ret; + + ret = seq_open(file, &nfs_server_list_ops); + if (ret < 0) + return ret; + + m = file->private_data; + m->private = PDE(inode)->data; + + return 0; +} + +/* + * set up the iterator to start reading from the server list and return the first item + */ +static void *nfs_server_list_start(struct seq_file *m, loff_t *_pos) +{ + struct list_head *_p; + loff_t pos = *_pos; + + /* lock the list against modification */ + spin_lock(&nfs_client_lock); + + /* allow for the header line */ + if (!pos) + return SEQ_START_TOKEN; + pos--; + + /* find the n'th element in the list */ + list_for_each(_p, &nfs_client_list) + if (!pos--) + break; + + return _p != &nfs_client_list ? _p : NULL; +} + +/* + * move to next server + */ +static void *nfs_server_list_next(struct seq_file *p, void *v, loff_t *pos) +{ + struct list_head *_p; + + (*pos)++; + + _p = v; + _p = (v == SEQ_START_TOKEN) ? nfs_client_list.next : _p->next; + + return _p != &nfs_client_list ? _p : NULL; +} + +/* + * clean up after reading from the transports list + */ +static void nfs_server_list_stop(struct seq_file *p, void *v) +{ + spin_unlock(&nfs_client_lock); +} + +/* + * display a header line followed by a load of call lines + */ +static int nfs_server_list_show(struct seq_file *m, void *v) +{ + struct nfs_client *clp; + + /* display header on line 1 */ + if (v == SEQ_START_TOKEN) { + seq_puts(m, "NV SERVER PORT USE HOSTNAME\n"); + return 0; + } + + /* display one transport per line on subsequent lines */ + clp = list_entry(v, struct nfs_client, cl_share_link); + + seq_printf(m, "v%d %02x%02x%02x%02x %4hx %3d %s\n", + clp->cl_nfsversion, + NIPQUAD(clp->cl_addr.sin_addr), + ntohs(clp->cl_addr.sin_port), + atomic_read(&clp->cl_count), + clp->cl_hostname); + + return 0; +} + +/* + * open "/proc/fs/nfsfs/volumes" which provides a summary of extant volumes + */ +static int nfs_volume_list_open(struct inode *inode, struct file *file) +{ + struct seq_file *m; + int ret; + + ret = seq_open(file, &nfs_volume_list_ops); + if (ret < 0) + return ret; + + m = file->private_data; + m->private = PDE(inode)->data; + + return 0; +} + +/* + * set up the iterator to start reading from the volume list and return the first item + */ +static void *nfs_volume_list_start(struct seq_file *m, loff_t *_pos) +{ + struct list_head *_p; + loff_t pos = *_pos; + + /* lock the list against modification */ + spin_lock(&nfs_client_lock); + + /* allow for the header line */ + if (!pos) + return SEQ_START_TOKEN; + pos--; + + /* find the n'th element in the list */ + list_for_each(_p, &nfs_volume_list) + if (!pos--) + break; + + return _p != &nfs_volume_list ? _p : NULL; +} + +/* + * move to next volume + */ +static void *nfs_volume_list_next(struct seq_file *p, void *v, loff_t *pos) +{ + struct list_head *_p; + + (*pos)++; + + _p = v; + _p = (v == SEQ_START_TOKEN) ? nfs_volume_list.next : _p->next; + + return _p != &nfs_volume_list ? _p : NULL; +} + +/* + * clean up after reading from the transports list + */ +static void nfs_volume_list_stop(struct seq_file *p, void *v) +{ + spin_unlock(&nfs_client_lock); +} + +/* + * display a header line followed by a load of call lines + */ +static int nfs_volume_list_show(struct seq_file *m, void *v) +{ + struct nfs_server *server; + struct nfs_client *clp; + char dev[8], fsid[17]; + + /* display header on line 1 */ + if (v == SEQ_START_TOKEN) { + seq_puts(m, "NV SERVER PORT DEV FSID\n"); + return 0; + } + /* display one transport per line on subsequent lines */ + server = list_entry(v, struct nfs_server, master_link); + clp = server->nfs_client; + + snprintf(dev, 8, "%u:%u", + MAJOR(server->s_dev), MINOR(server->s_dev)); + + snprintf(fsid, 17, "%llx:%llx", + (unsigned long long) server->fsid.major, + (unsigned long long) server->fsid.minor); + + seq_printf(m, "v%d %02x%02x%02x%02x %4hx %-7s %-17s\n", + clp->cl_nfsversion, + NIPQUAD(clp->cl_addr.sin_addr), + ntohs(clp->cl_addr.sin_port), + dev, + fsid); + + return 0; +} + +/* + * initialise the /proc/fs/nfsfs/ directory + */ +int __init nfs_fs_proc_init(void) +{ + struct proc_dir_entry *p; + + proc_fs_nfs = proc_mkdir("nfsfs", proc_root_fs); + if (!proc_fs_nfs) + goto error_0; + + proc_fs_nfs->owner = THIS_MODULE; + + /* a file of servers with which we're dealing */ + p = create_proc_entry("servers", S_IFREG|S_IRUGO, proc_fs_nfs); + if (!p) + goto error_1; + + p->proc_fops = &nfs_server_list_fops; + p->owner = THIS_MODULE; + + /* a file of volumes that we have mounted */ + p = create_proc_entry("volumes", S_IFREG|S_IRUGO, proc_fs_nfs); + if (!p) + goto error_2; + + p->proc_fops = &nfs_volume_list_fops; + p->owner = THIS_MODULE; + return 0; + +error_2: + remove_proc_entry("servers", proc_fs_nfs); +error_1: + remove_proc_entry("nfsfs", proc_root_fs); +error_0: + return -ENOMEM; +} + +/* + * clean up the /proc/fs/nfsfs/ directory + */ +void nfs_fs_proc_exit(void) +{ + remove_proc_entry("volumes", proc_fs_nfs); + remove_proc_entry("servers", proc_fs_nfs); + remove_proc_entry("nfsfs", proc_root_fs); +} + +#endif /* CONFIG_PROC_FS */ diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index d3be923d4e4..841c99a9b11 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -6,7 +6,6 @@ * NFS file delegation management * */ -#include <linux/config.h> #include <linux/completion.h> #include <linux/kthread.h> #include <linux/module.h> @@ -19,11 +18,7 @@ #include "nfs4_fs.h" #include "delegation.h" - -static struct nfs_delegation *nfs_alloc_delegation(void) -{ - return (struct nfs_delegation *)kmalloc(sizeof(struct nfs_delegation), GFP_KERNEL); -} +#include "internal.h" static void nfs_free_delegation(struct nfs_delegation *delegation) { @@ -53,7 +48,7 @@ static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_ case -NFS4ERR_EXPIRED: /* kill_proc(fl->fl_pid, SIGLOST, 1); */ case -NFS4ERR_STALE_CLIENTID: - nfs4_schedule_state_recovery(NFS_SERVER(inode)->nfs4_state); + nfs4_schedule_state_recovery(NFS_SERVER(inode)->nfs_client); goto out_err; } } @@ -115,7 +110,7 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, st */ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res) { - struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state; + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; struct nfs_inode *nfsi = NFS_I(inode); struct nfs_delegation *delegation; int status = 0; @@ -124,7 +119,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct if ((nfsi->cache_validity & (NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_ATTR))) __nfs_revalidate_inode(NFS_SERVER(inode), inode); - delegation = nfs_alloc_delegation(); + delegation = kmalloc(sizeof(*delegation), GFP_KERNEL); if (delegation == NULL) return -ENOMEM; memcpy(delegation->stateid.data, res->delegation.data, @@ -146,7 +141,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct sizeof(delegation->stateid)) != 0 || delegation->type != nfsi->delegation->type) { printk("%s: server %u.%u.%u.%u, handed out a duplicate delegation!\n", - __FUNCTION__, NIPQUAD(clp->cl_addr)); + __FUNCTION__, NIPQUAD(clp->cl_addr.sin_addr)); status = -EIO; } } @@ -177,7 +172,7 @@ static void nfs_msync_inode(struct inode *inode) */ int __nfs_inode_return_delegation(struct inode *inode) { - struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state; + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; struct nfs_inode *nfsi = NFS_I(inode); struct nfs_delegation *delegation; int res = 0; @@ -209,7 +204,7 @@ int __nfs_inode_return_delegation(struct inode *inode) */ void nfs_return_all_delegations(struct super_block *sb) { - struct nfs4_client *clp = NFS_SB(sb)->nfs4_state; + struct nfs_client *clp = NFS_SB(sb)->nfs_client; struct nfs_delegation *delegation; struct inode *inode; @@ -233,7 +228,7 @@ restart: int nfs_do_expire_all_delegations(void *ptr) { - struct nfs4_client *clp = ptr; + struct nfs_client *clp = ptr; struct nfs_delegation *delegation; struct inode *inode; @@ -255,11 +250,11 @@ restart: } out: spin_unlock(&clp->cl_lock); - nfs4_put_client(clp); + nfs_put_client(clp); module_put_and_exit(0); } -void nfs_expire_all_delegations(struct nfs4_client *clp) +void nfs_expire_all_delegations(struct nfs_client *clp) { struct task_struct *task; @@ -267,17 +262,17 @@ void nfs_expire_all_delegations(struct nfs4_client *clp) atomic_inc(&clp->cl_count); task = kthread_run(nfs_do_expire_all_delegations, clp, "%u.%u.%u.%u-delegreturn", - NIPQUAD(clp->cl_addr)); + NIPQUAD(clp->cl_addr.sin_addr)); if (!IS_ERR(task)) return; - nfs4_put_client(clp); + nfs_put_client(clp); module_put(THIS_MODULE); } /* * Return all delegations following an NFS4ERR_CB_PATH_DOWN error. */ -void nfs_handle_cb_pathdown(struct nfs4_client *clp) +void nfs_handle_cb_pathdown(struct nfs_client *clp) { struct nfs_delegation *delegation; struct inode *inode; @@ -300,7 +295,7 @@ restart: struct recall_threadargs { struct inode *inode; - struct nfs4_client *clp; + struct nfs_client *clp; const nfs4_stateid *stateid; struct completion started; @@ -311,7 +306,7 @@ static int recall_thread(void *data) { struct recall_threadargs *args = (struct recall_threadargs *)data; struct inode *inode = igrab(args->inode); - struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state; + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; struct nfs_inode *nfsi = NFS_I(inode); struct nfs_delegation *delegation; @@ -372,7 +367,7 @@ out_module_put: /* * Retrieve the inode associated with a delegation */ -struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nfs_fh *fhandle) +struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle) { struct nfs_delegation *delegation; struct inode *res = NULL; @@ -390,7 +385,7 @@ struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nf /* * Mark all delegations as needing to be reclaimed */ -void nfs_delegation_mark_reclaim(struct nfs4_client *clp) +void nfs_delegation_mark_reclaim(struct nfs_client *clp) { struct nfs_delegation *delegation; spin_lock(&clp->cl_lock); @@ -402,7 +397,7 @@ void nfs_delegation_mark_reclaim(struct nfs4_client *clp) /* * Reap all unclaimed delegations after reboot recovery is done */ -void nfs_delegation_reap_unclaimed(struct nfs4_client *clp) +void nfs_delegation_reap_unclaimed(struct nfs_client *clp) { struct nfs_delegation *delegation, *n; LIST_HEAD(head); @@ -424,7 +419,7 @@ void nfs_delegation_reap_unclaimed(struct nfs4_client *clp) int nfs4_copy_delegation_stateid(nfs4_stateid *dst, struct inode *inode) { - struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state; + struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; struct nfs_inode *nfsi = NFS_I(inode); struct nfs_delegation *delegation; int res = 0; diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h index 3858694652f..2cfd4b24c7f 100644 --- a/fs/nfs/delegation.h +++ b/fs/nfs/delegation.h @@ -29,13 +29,13 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, st int __nfs_inode_return_delegation(struct inode *inode); int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid); -struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nfs_fh *fhandle); +struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle); void nfs_return_all_delegations(struct super_block *sb); -void nfs_expire_all_delegations(struct nfs4_client *clp); -void nfs_handle_cb_pathdown(struct nfs4_client *clp); +void nfs_expire_all_delegations(struct nfs_client *clp); +void nfs_handle_cb_pathdown(struct nfs_client *clp); -void nfs_delegation_mark_reclaim(struct nfs4_client *clp); -void nfs_delegation_reap_unclaimed(struct nfs4_client *clp); +void nfs_delegation_mark_reclaim(struct nfs_client *clp); +void nfs_delegation_reap_unclaimed(struct nfs_client *clp); /* NFSv4 delegation-related procedures */ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid); diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index cae74dd4c7f..7432f1a43f3 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -30,7 +30,9 @@ #include <linux/nfs_mount.h> #include <linux/pagemap.h> #include <linux/smp_lock.h> +#include <linux/pagevec.h> #include <linux/namei.h> +#include <linux/mount.h> #include "nfs4_fs.h" #include "delegation.h" @@ -528,7 +530,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) lock_kernel(); - res = nfs_revalidate_inode(NFS_SERVER(inode), inode); + res = nfs_revalidate_mapping(inode, filp->f_mapping); if (res < 0) { unlock_kernel(); return res; @@ -690,7 +692,9 @@ int nfs_lookup_verify_inode(struct inode *inode, struct nameidata *nd) goto out_force; /* This is an open(2) */ if (nfs_lookup_check_intent(nd, LOOKUP_OPEN) != 0 && - !(server->flags & NFS_MOUNT_NOCTO)) + !(server->flags & NFS_MOUNT_NOCTO) && + (S_ISREG(inode->i_mode) || + S_ISDIR(inode->i_mode))) goto out_force; } return nfs_revalidate_inode(server, inode); @@ -868,6 +872,17 @@ int nfs_is_exclusive_create(struct inode *dir, struct nameidata *nd) return (nd->intent.open.flags & O_EXCL) != 0; } +static inline int nfs_reval_fsid(struct vfsmount *mnt, struct inode *dir, + struct nfs_fh *fh, struct nfs_fattr *fattr) +{ + struct nfs_server *server = NFS_SERVER(dir); + + if (!nfs_fsid_equal(&server->fsid, &fattr->fsid)) + /* Revalidate fsid on root dir */ + return __nfs_revalidate_inode(server, mnt->mnt_root->d_inode); + return 0; +} + static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd) { struct dentry *res; @@ -889,9 +904,15 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru lock_kernel(); - /* If we're doing an exclusive create, optimize away the lookup */ - if (nfs_is_exclusive_create(dir, nd)) - goto no_entry; + /* + * If we're doing an exclusive create, optimize away the lookup + * but don't hash the dentry. + */ + if (nfs_is_exclusive_create(dir, nd)) { + d_instantiate(dentry, NULL); + res = NULL; + goto out_unlock; + } error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, &fhandle, &fattr); if (error == -ENOENT) @@ -900,12 +921,18 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru res = ERR_PTR(error); goto out_unlock; } + error = nfs_reval_fsid(nd->mnt, dir, &fhandle, &fattr); + if (error < 0) { + res = ERR_PTR(error); + goto out_unlock; + } inode = nfs_fhget(dentry->d_sb, &fhandle, &fattr); res = (struct dentry *)inode; if (IS_ERR(res)) goto out_unlock; + no_entry: - res = d_add_unique(dentry, inode); + res = d_materialise_unique(dentry, inode); if (res != NULL) dentry = res; nfs_renew_times(dentry); @@ -1099,11 +1126,13 @@ static struct dentry *nfs_readdir_lookup(nfs_readdir_descriptor_t *desc) dput(dentry); return NULL; } - alias = d_add_unique(dentry, inode); + + alias = d_materialise_unique(dentry, inode); if (alias != NULL) { dput(dentry); dentry = alias; } + nfs_renew_times(dentry); nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); return dentry; @@ -1125,23 +1154,22 @@ int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle, struct inode *dir = dentry->d_parent->d_inode; error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr); if (error) - goto out_err; + return error; } if (!(fattr->valid & NFS_ATTR_FATTR)) { struct nfs_server *server = NFS_SB(dentry->d_sb); - error = server->rpc_ops->getattr(server, fhandle, fattr); + error = server->nfs_client->rpc_ops->getattr(server, fhandle, fattr); if (error < 0) - goto out_err; + return error; } inode = nfs_fhget(dentry->d_sb, fhandle, fattr); error = PTR_ERR(inode); if (IS_ERR(inode)) - goto out_err; + return error; d_instantiate(dentry, inode); + if (d_unhashed(dentry)) + d_rehash(dentry); return 0; -out_err: - d_drop(dentry); - return error; } /* @@ -1422,48 +1450,82 @@ static int nfs_unlink(struct inode *dir, struct dentry *dentry) return error; } -static int -nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) +/* + * To create a symbolic link, most file systems instantiate a new inode, + * add a page to it containing the path, then write it out to the disk + * using prepare_write/commit_write. + * + * Unfortunately the NFS client can't create the in-core inode first + * because it needs a file handle to create an in-core inode (see + * fs/nfs/inode.c:nfs_fhget). We only have a file handle *after* the + * symlink request has completed on the server. + * + * So instead we allocate a raw page, copy the symname into it, then do + * the SYMLINK request with the page as the buffer. If it succeeds, we + * now have a new file handle and can instantiate an in-core NFS inode + * and move the raw page into its mapping. + */ +static int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) { + struct pagevec lru_pvec; + struct page *page; + char *kaddr; struct iattr attr; - struct nfs_fattr sym_attr; - struct nfs_fh sym_fh; - struct qstr qsymname; + unsigned int pathlen = strlen(symname); int error; dfprintk(VFS, "NFS: symlink(%s/%ld, %s, %s)\n", dir->i_sb->s_id, dir->i_ino, dentry->d_name.name, symname); -#ifdef NFS_PARANOIA -if (dentry->d_inode) -printk("nfs_proc_symlink: %s/%s not negative!\n", -dentry->d_parent->d_name.name, dentry->d_name.name); -#endif - /* - * Fill in the sattr for the call. - * Note: SunOS 4.1.2 crashes if the mode isn't initialized! - */ - attr.ia_valid = ATTR_MODE; - attr.ia_mode = S_IFLNK | S_IRWXUGO; + if (pathlen > PAGE_SIZE) + return -ENAMETOOLONG; - qsymname.name = symname; - qsymname.len = strlen(symname); + attr.ia_mode = S_IFLNK | S_IRWXUGO; + attr.ia_valid = ATTR_MODE; lock_kernel(); + + page = alloc_page(GFP_KERNEL); + if (!page) { + unlock_kernel(); + return -ENOMEM; + } + + kaddr = kmap_atomic(page, KM_USER0); + memcpy(kaddr, symname, pathlen); + if (pathlen < PAGE_SIZE) + memset(kaddr + pathlen, 0, PAGE_SIZE - pathlen); + kunmap_atomic(kaddr, KM_USER0); + nfs_begin_data_update(dir); - error = NFS_PROTO(dir)->symlink(dir, &dentry->d_name, &qsymname, - &attr, &sym_fh, &sym_attr); + error = NFS_PROTO(dir)->symlink(dir, dentry, page, pathlen, &attr); nfs_end_data_update(dir); - if (!error) { - error = nfs_instantiate(dentry, &sym_fh, &sym_attr); - } else { - if (error == -EEXIST) - printk("nfs_proc_symlink: %s/%s already exists??\n", - dentry->d_parent->d_name.name, dentry->d_name.name); + if (error != 0) { + dfprintk(VFS, "NFS: symlink(%s/%ld, %s, %s) error %d\n", + dir->i_sb->s_id, dir->i_ino, + dentry->d_name.name, symname, error); d_drop(dentry); + __free_page(page); + unlock_kernel(); + return error; } + + /* + * No big deal if we can't add this page to the page cache here. + * READLINK will get the missing page from the server if needed. + */ + pagevec_init(&lru_pvec, 0); + if (!add_to_page_cache(page, dentry->d_inode->i_mapping, 0, + GFP_KERNEL)) { + if (!pagevec_add(&lru_pvec, page)) + __pagevec_lru_add(&lru_pvec); + SetPageUptodate(page); + unlock_page(page); + } else + __free_page(page); + unlock_kernel(); - return error; + return 0; } static int @@ -1607,8 +1669,7 @@ out: if (rehash) d_rehash(rehash); if (!error) { - if (!S_ISDIR(old_inode->i_mode)) - d_move(old_dentry, new_dentry); + d_move(old_dentry, new_dentry); nfs_renew_times(new_dentry); nfs_set_verifier(new_dentry, nfs_save_change_attribute(new_dir)); } @@ -1620,35 +1681,211 @@ out: return error; } +static DEFINE_SPINLOCK(nfs_access_lru_lock); +static LIST_HEAD(nfs_access_lru_list); +static atomic_long_t nfs_access_nr_entries; + +static void nfs_access_free_entry(struct nfs_access_entry *entry) +{ + put_rpccred(entry->cred); + kfree(entry); + smp_mb__before_atomic_dec(); + atomic_long_dec(&nfs_access_nr_entries); + smp_mb__after_atomic_dec(); +} + +int nfs_access_cache_shrinker(int nr_to_scan, gfp_t gfp_mask) +{ + LIST_HEAD(head); + struct nfs_inode *nfsi; + struct nfs_access_entry *cache; + + spin_lock(&nfs_access_lru_lock); +restart: + list_for_each_entry(nfsi, &nfs_access_lru_list, access_cache_inode_lru) { + struct inode *inode; + + if (nr_to_scan-- == 0) + break; + inode = igrab(&nfsi->vfs_inode); + if (inode == NULL) + continue; + spin_lock(&inode->i_lock); + if (list_empty(&nfsi->access_cache_entry_lru)) + goto remove_lru_entry; + cache = list_entry(nfsi->access_cache_entry_lru.next, + struct nfs_access_entry, lru); + list_move(&cache->lru, &head); + rb_erase(&cache->rb_node, &nfsi->access_cache); + if (!list_empty(&nfsi->access_cache_entry_lru)) + list_move_tail(&nfsi->access_cache_inode_lru, + &nfs_access_lru_list); + else { +remove_lru_entry: + list_del_init(&nfsi->access_cache_inode_lru); + clear_bit(NFS_INO_ACL_LRU_SET, &nfsi->flags); + } + spin_unlock(&inode->i_lock); + iput(inode); + goto restart; + } + spin_unlock(&nfs_access_lru_lock); + while (!list_empty(&head)) { + cache = list_entry(head.next, struct nfs_access_entry, lru); + list_del(&cache->lru); + nfs_access_free_entry(cache); + } + return (atomic_long_read(&nfs_access_nr_entries) / 100) * sysctl_vfs_cache_pressure; +} + +static void __nfs_access_zap_cache(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + struct rb_root *root_node = &nfsi->access_cache; + struct rb_node *n, *dispose = NULL; + struct nfs_access_entry *entry; + + /* Unhook entries from the cache */ + while ((n = rb_first(root_node)) != NULL) { + entry = rb_entry(n, struct nfs_access_entry, rb_node); + rb_erase(n, root_node); + list_del(&entry->lru); + n->rb_left = dispose; + dispose = n; + } + nfsi->cache_validity &= ~NFS_INO_INVALID_ACCESS; + spin_unlock(&inode->i_lock); + + /* Now kill them all! */ + while (dispose != NULL) { + n = dispose; + dispose = n->rb_left; + nfs_access_free_entry(rb_entry(n, struct nfs_access_entry, rb_node)); + } +} + +void nfs_access_zap_cache(struct inode *inode) +{ + /* Remove from global LRU init */ + if (test_and_clear_bit(NFS_INO_ACL_LRU_SET, &NFS_FLAGS(inode))) { + spin_lock(&nfs_access_lru_lock); + list_del_init(&NFS_I(inode)->access_cache_inode_lru); + spin_unlock(&nfs_access_lru_lock); + } + + spin_lock(&inode->i_lock); + /* This will release the spinlock */ + __nfs_access_zap_cache(inode); +} + +static struct nfs_access_entry *nfs_access_search_rbtree(struct inode *inode, struct rpc_cred *cred) +{ + struct rb_node *n = NFS_I(inode)->access_cache.rb_node; + struct nfs_access_entry *entry; + + while (n != NULL) { + entry = rb_entry(n, struct nfs_access_entry, rb_node); + + if (cred < entry->cred) + n = n->rb_left; + else if (cred > entry->cred) + n = n->rb_right; + else + return entry; + } + return NULL; +} + int nfs_access_get_cached(struct inode *inode, struct rpc_cred *cred, struct nfs_access_entry *res) { struct nfs_inode *nfsi = NFS_I(inode); - struct nfs_access_entry *cache = &nfsi->cache_access; + struct nfs_access_entry *cache; + int err = -ENOENT; - if (cache->cred != cred - || time_after(jiffies, cache->jiffies + NFS_ATTRTIMEO(inode)) - || (nfsi->cache_validity & NFS_INO_INVALID_ACCESS)) - return -ENOENT; - memcpy(res, cache, sizeof(*res)); - return 0; + spin_lock(&inode->i_lock); + if (nfsi->cache_validity & NFS_INO_INVALID_ACCESS) + goto out_zap; + cache = nfs_access_search_rbtree(inode, cred); + if (cache == NULL) + goto out; + if (time_after(jiffies, cache->jiffies + NFS_ATTRTIMEO(inode))) + goto out_stale; + res->jiffies = cache->jiffies; + res->cred = cache->cred; + res->mask = cache->mask; + list_move_tail(&cache->lru, &nfsi->access_cache_entry_lru); + err = 0; +out: + spin_unlock(&inode->i_lock); + return err; +out_stale: + rb_erase(&cache->rb_node, &nfsi->access_cache); + list_del(&cache->lru); + spin_unlock(&inode->i_lock); + nfs_access_free_entry(cache); + return -ENOENT; +out_zap: + /* This will release the spinlock */ + __nfs_access_zap_cache(inode); + return -ENOENT; } -void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) +static void nfs_access_add_rbtree(struct inode *inode, struct nfs_access_entry *set) { struct nfs_inode *nfsi = NFS_I(inode); - struct nfs_access_entry *cache = &nfsi->cache_access; + struct rb_root *root_node = &nfsi->access_cache; + struct rb_node **p = &root_node->rb_node; + struct rb_node *parent = NULL; + struct nfs_access_entry *entry; - if (cache->cred != set->cred) { - if (cache->cred) - put_rpccred(cache->cred); - cache->cred = get_rpccred(set->cred); - } - /* FIXME: replace current access_cache BKL reliance with inode->i_lock */ spin_lock(&inode->i_lock); - nfsi->cache_validity &= ~NFS_INO_INVALID_ACCESS; + while (*p != NULL) { + parent = *p; + entry = rb_entry(parent, struct nfs_access_entry, rb_node); + + if (set->cred < entry->cred) + p = &parent->rb_left; + else if (set->cred > entry->cred) + p = &parent->rb_right; + else + goto found; + } + rb_link_node(&set->rb_node, parent, p); + rb_insert_color(&set->rb_node, root_node); + list_add_tail(&set->lru, &nfsi->access_cache_entry_lru); spin_unlock(&inode->i_lock); + return; +found: + rb_replace_node(parent, &set->rb_node, root_node); + list_add_tail(&set->lru, &nfsi->access_cache_entry_lru); + list_del(&entry->lru); + spin_unlock(&inode->i_lock); + nfs_access_free_entry(entry); +} + +void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) +{ + struct nfs_access_entry *cache = kmalloc(sizeof(*cache), GFP_KERNEL); + if (cache == NULL) + return; + RB_CLEAR_NODE(&cache->rb_node); cache->jiffies = set->jiffies; + cache->cred = get_rpccred(set->cred); cache->mask = set->mask; + + nfs_access_add_rbtree(inode, cache); + + /* Update accounting */ + smp_mb__before_atomic_inc(); + atomic_long_inc(&nfs_access_nr_entries); + smp_mb__after_atomic_inc(); + + /* Add inode to global LRU list */ + if (!test_and_set_bit(NFS_INO_ACL_LRU_SET, &NFS_FLAGS(inode))) { + spin_lock(&nfs_access_lru_lock); + list_add_tail(&NFS_I(inode)->access_cache_inode_lru, &nfs_access_lru_list); + spin_unlock(&nfs_access_lru_lock); + } } static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask) diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 3c72b0c0728..377839bed17 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -38,7 +38,6 @@ * */ -#include <linux/config.h> #include <linux/errno.h> #include <linux/sched.h> #include <linux/kernel.h> @@ -68,25 +67,19 @@ struct nfs_direct_req { struct kref kref; /* release manager */ /* I/O parameters */ - struct list_head list, /* nfs_read/write_data structs */ - rewrite_list; /* saved nfs_write_data structs */ struct nfs_open_context *ctx; /* file open context info */ struct kiocb * iocb; /* controlling i/o request */ struct inode * inode; /* target file of i/o */ - unsigned long user_addr; /* location of user's buffer */ - size_t user_count; /* total bytes to move */ - loff_t pos; /* starting offset in file */ - struct page ** pages; /* pages in our buffer */ - unsigned int npages; /* count of pages */ /* completion state */ + atomic_t io_count; /* i/os we're waiting for */ spinlock_t lock; /* protect completion state */ - int outstanding; /* i/os we're waiting for */ ssize_t count, /* bytes actually processed */ error; /* any reported error */ struct completion completion; /* wait for i/o completion */ /* commit state */ + struct list_head rewrite_list; /* saved nfs_write_data structs */ struct nfs_write_data * commit_data; /* special write_data for commits */ int flags; #define NFS_ODIRECT_DO_COMMIT (1) /* an unstable reply was received */ @@ -94,8 +87,18 @@ struct nfs_direct_req { struct nfs_writeverf verf; /* unstable write verifier */ }; -static void nfs_direct_write_schedule(struct nfs_direct_req *dreq, int sync); static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode *inode); +static const struct rpc_call_ops nfs_write_direct_ops; + +static inline void get_dreq(struct nfs_direct_req *dreq) +{ + atomic_inc(&dreq->io_count); +} + +static inline int put_dreq(struct nfs_direct_req *dreq) +{ + return atomic_dec_and_test(&dreq->io_count); +} /** * nfs_direct_IO - NFS address space operation for direct I/O @@ -119,50 +122,21 @@ ssize_t nfs_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_ return -EINVAL; } -static void nfs_free_user_pages(struct page **pages, int npages, int do_dirty) +static void nfs_direct_dirty_pages(struct page **pages, int npages) { int i; for (i = 0; i < npages; i++) { struct page *page = pages[i]; - if (do_dirty && !PageCompound(page)) + if (!PageCompound(page)) set_page_dirty_lock(page); - page_cache_release(page); } - kfree(pages); } -static inline int nfs_get_user_pages(int rw, unsigned long user_addr, size_t size, struct page ***pages) +static void nfs_direct_release_pages(struct page **pages, int npages) { - int result = -ENOMEM; - unsigned long page_count; - size_t array_size; - - page_count = (user_addr + size + PAGE_SIZE - 1) >> PAGE_SHIFT; - page_count -= user_addr >> PAGE_SHIFT; - - array_size = (page_count * sizeof(struct page *)); - *pages = kmalloc(array_size, GFP_KERNEL); - if (*pages) { - down_read(¤t->mm->mmap_sem); - result = get_user_pages(current, current->mm, user_addr, - page_count, (rw == READ), 0, - *pages, NULL); - up_read(¤t->mm->mmap_sem); - if (result != page_count) { - /* - * If we got fewer pages than expected from - * get_user_pages(), the user buffer runs off the - * end of a mapping; return EFAULT. - */ - if (result >= 0) { - nfs_free_user_pages(*pages, result, 0); - result = -EFAULT; - } else - kfree(*pages); - *pages = NULL; - } - } - return result; + int i; + for (i = 0; i < npages; i++) + page_cache_release(pages[i]); } static inline struct nfs_direct_req *nfs_direct_req_alloc(void) @@ -174,13 +148,13 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void) return NULL; kref_init(&dreq->kref); + kref_get(&dreq->kref); init_completion(&dreq->completion); - INIT_LIST_HEAD(&dreq->list); INIT_LIST_HEAD(&dreq->rewrite_list); dreq->iocb = NULL; dreq->ctx = NULL; spin_lock_init(&dreq->lock); - dreq->outstanding = 0; + atomic_set(&dreq->io_count, 0); dreq->count = 0; dreq->error = 0; dreq->flags = 0; @@ -221,18 +195,11 @@ out: } /* - * We must hold a reference to all the pages in this direct read request - * until the RPCs complete. This could be long *after* we are woken up in - * nfs_direct_wait (for instance, if someone hits ^C on a slow server). - * - * In addition, synchronous I/O uses a stack-allocated iocb. Thus we - * can't trust the iocb is still valid here if this is a synchronous - * request. If the waiter is woken prematurely, the iocb is long gone. + * Synchronous I/O uses a stack-allocated iocb. Thus we can't trust + * the iocb is still valid here if this is a synchronous request. */ static void nfs_direct_complete(struct nfs_direct_req *dreq) { - nfs_free_user_pages(dreq->pages, dreq->npages, 1); - if (dreq->iocb) { long res = (long) dreq->error; if (!res) @@ -245,48 +212,10 @@ static void nfs_direct_complete(struct nfs_direct_req *dreq) } /* - * Note we also set the number of requests we have in the dreq when we are - * done. This prevents races with I/O completion so we will always wait - * until all requests have been dispatched and completed. + * We must hold a reference to all the pages in this direct read request + * until the RPCs complete. This could be long *after* we are woken up in + * nfs_direct_wait (for instance, if someone hits ^C on a slow server). */ -static struct nfs_direct_req *nfs_direct_read_alloc(size_t nbytes, size_t rsize) -{ - struct list_head *list; - struct nfs_direct_req *dreq; - unsigned int rpages = (rsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - - dreq = nfs_direct_req_alloc(); - if (!dreq) - return NULL; - - list = &dreq->list; - for(;;) { - struct nfs_read_data *data = nfs_readdata_alloc(rpages); - - if (unlikely(!data)) { - while (!list_empty(list)) { - data = list_entry(list->next, - struct nfs_read_data, pages); - list_del(&data->pages); - nfs_readdata_free(data); - } - kref_put(&dreq->kref, nfs_direct_req_release); - return NULL; - } - - INIT_LIST_HEAD(&data->pages); - list_add(&data->pages, list); - - data->req = (struct nfs_page *) dreq; - dreq->outstanding++; - if (nbytes <= rsize) - break; - nbytes -= rsize; - } - kref_get(&dreq->kref); - return dreq; -} - static void nfs_direct_read_result(struct rpc_task *task, void *calldata) { struct nfs_read_data *data = calldata; @@ -295,6 +224,9 @@ static void nfs_direct_read_result(struct rpc_task *task, void *calldata) if (nfs_readpage_result(task, data) != 0) return; + nfs_direct_dirty_pages(data->pagevec, data->npages); + nfs_direct_release_pages(data->pagevec, data->npages); + spin_lock(&dreq->lock); if (likely(task->tk_status >= 0)) @@ -302,13 +234,10 @@ static void nfs_direct_read_result(struct rpc_task *task, void *calldata) else dreq->error = task->tk_status; - if (--dreq->outstanding) { - spin_unlock(&dreq->lock); - return; - } - spin_unlock(&dreq->lock); - nfs_direct_complete(dreq); + + if (put_dreq(dreq)) + nfs_direct_complete(dreq); } static const struct rpc_call_ops nfs_read_direct_ops = { @@ -317,41 +246,56 @@ static const struct rpc_call_ops nfs_read_direct_ops = { }; /* - * For each nfs_read_data struct that was allocated on the list, dispatch - * an NFS READ operation + * For each rsize'd chunk of the user's buffer, dispatch an NFS READ + * operation. If nfs_readdata_alloc() or get_user_pages() fails, + * bail and stop sending more reads. Read length accounting is + * handled automatically by nfs_direct_read_result(). Otherwise, if + * no requests have been sent, just return an error. */ -static void nfs_direct_read_schedule(struct nfs_direct_req *dreq) +static ssize_t nfs_direct_read_schedule(struct nfs_direct_req *dreq, unsigned long user_addr, size_t count, loff_t pos) { struct nfs_open_context *ctx = dreq->ctx; struct inode *inode = ctx->dentry->d_inode; - struct list_head *list = &dreq->list; - struct page **pages = dreq->pages; - size_t count = dreq->user_count; - loff_t pos = dreq->pos; size_t rsize = NFS_SERVER(inode)->rsize; - unsigned int curpage, pgbase; + unsigned int pgbase; + int result; + ssize_t started = 0; + + get_dreq(dreq); - curpage = 0; - pgbase = dreq->user_addr & ~PAGE_MASK; do { struct nfs_read_data *data; size_t bytes; - bytes = rsize; - if (count < rsize) - bytes = count; + pgbase = user_addr & ~PAGE_MASK; + bytes = min(rsize,count); - BUG_ON(list_empty(list)); - data = list_entry(list->next, struct nfs_read_data, pages); - list_del_init(&data->pages); + result = -ENOMEM; + data = nfs_readdata_alloc(pgbase + bytes); + if (unlikely(!data)) + break; + + down_read(¤t->mm->mmap_sem); + result = get_user_pages(current, current->mm, user_addr, + data->npages, 1, 0, data->pagevec, NULL); + up_read(¤t->mm->mmap_sem); + if (unlikely(result < data->npages)) { + if (result > 0) + nfs_direct_release_pages(data->pagevec, result); + nfs_readdata_release(data); + break; + } + + get_dreq(dreq); + data->req = (struct nfs_page *) dreq; data->inode = inode; data->cred = ctx->cred; data->args.fh = NFS_FH(inode); data->args.context = ctx; data->args.offset = pos; data->args.pgbase = pgbase; - data->args.pages = &pages[curpage]; + data->args.pages = data->pagevec; data->args.count = bytes; data->res.fattr = &data->fattr; data->res.eof = 0; @@ -374,33 +318,37 @@ static void nfs_direct_read_schedule(struct nfs_direct_req *dreq) bytes, (unsigned long long)data->args.offset); + started += bytes; + user_addr += bytes; pos += bytes; + /* FIXME: Remove this unnecessary math from final patch */ pgbase += bytes; - curpage += pgbase >> PAGE_SHIFT; pgbase &= ~PAGE_MASK; + BUG_ON(pgbase != (user_addr & ~PAGE_MASK)); count -= bytes; } while (count != 0); - BUG_ON(!list_empty(list)); + + if (put_dreq(dreq)) + nfs_direct_complete(dreq); + + if (started) + return 0; + return result < 0 ? (ssize_t) result : -EFAULT; } -static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size_t count, loff_t pos, struct page **pages, unsigned int nr_pages) +static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size_t count, loff_t pos) { - ssize_t result; + ssize_t result = 0; sigset_t oldset; struct inode *inode = iocb->ki_filp->f_mapping->host; struct rpc_clnt *clnt = NFS_CLIENT(inode); struct nfs_direct_req *dreq; - dreq = nfs_direct_read_alloc(count, NFS_SERVER(inode)->rsize); + dreq = nfs_direct_req_alloc(); if (!dreq) return -ENOMEM; - dreq->user_addr = user_addr; - dreq->user_count = count; - dreq->pos = pos; - dreq->pages = pages; - dreq->npages = nr_pages; dreq->inode = inode; dreq->ctx = get_nfs_open_context((struct nfs_open_context *)iocb->ki_filp->private_data); if (!is_sync_kiocb(iocb)) @@ -408,8 +356,9 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size nfs_add_stats(inode, NFSIOS_DIRECTREADBYTES, count); rpc_clnt_sigmask(clnt, &oldset); - nfs_direct_read_schedule(dreq); - result = nfs_direct_wait(dreq); + result = nfs_direct_read_schedule(dreq, user_addr, count, pos); + if (!result) + result = nfs_direct_wait(dreq); rpc_clnt_sigunmask(clnt, &oldset); return result; @@ -417,10 +366,10 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, unsigned long user_addr, size static void nfs_direct_free_writedata(struct nfs_direct_req *dreq) { - list_splice_init(&dreq->rewrite_list, &dreq->list); - while (!list_empty(&dreq->list)) { - struct nfs_write_data *data = list_entry(dreq->list.next, struct nfs_write_data, pages); + while (!list_empty(&dreq->rewrite_list)) { + struct nfs_write_data *data = list_entry(dreq->rewrite_list.next, struct nfs_write_data, pages); list_del(&data->pages); + nfs_direct_release_pages(data->pagevec, data->npages); nfs_writedata_release(data); } } @@ -428,14 +377,51 @@ static void nfs_direct_free_writedata(struct nfs_direct_req *dreq) #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) { - struct list_head *pos; + struct inode *inode = dreq->inode; + struct list_head *p; + struct nfs_write_data *data; - list_splice_init(&dreq->rewrite_list, &dreq->list); - list_for_each(pos, &dreq->list) - dreq->outstanding++; dreq->count = 0; + get_dreq(dreq); + + list_for_each(p, &dreq->rewrite_list) { + data = list_entry(p, struct nfs_write_data, pages); + + get_dreq(dreq); - nfs_direct_write_schedule(dreq, FLUSH_STABLE); + /* + * Reset data->res. + */ + nfs_fattr_init(&data->fattr); + data->res.count = data->args.count; + memset(&data->verf, 0, sizeof(data->verf)); + + /* + * Reuse data->task; data->args should not have changed + * since the original request was sent. + */ + rpc_init_task(&data->task, NFS_CLIENT(inode), RPC_TASK_ASYNC, + &nfs_write_direct_ops, data); + NFS_PROTO(inode)->write_setup(data, FLUSH_STABLE); + + data->task.tk_priority = RPC_PRIORITY_NORMAL; + data->task.tk_cookie = (unsigned long) inode; + + /* + * We're called via an RPC callback, so BKL is already held. + */ + rpc_execute(&data->task); + + dprintk("NFS: %5u rescheduled direct write call (req %s/%Ld, %u bytes @ offset %Lu)\n", + data->task.tk_pid, + inode->i_sb->s_id, + (long long)NFS_FILEID(inode), + data->args.count, + (unsigned long long)data->args.offset); + } + + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq, inode); } static void nfs_direct_commit_result(struct rpc_task *task, void *calldata) @@ -472,8 +458,8 @@ static void nfs_direct_commit_schedule(struct nfs_direct_req *dreq) data->cred = dreq->ctx->cred; data->args.fh = NFS_FH(data->inode); - data->args.offset = dreq->pos; - data->args.count = dreq->user_count; + data->args.offset = 0; + data->args.count = 0; data->res.count = 0; data->res.fattr = &data->fattr; data->res.verf = &data->verf; @@ -517,7 +503,7 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode static void nfs_alloc_commit_data(struct nfs_direct_req *dreq) { - dreq->commit_data = nfs_commit_alloc(0); + dreq->commit_data = nfs_commit_alloc(); if (dreq->commit_data != NULL) dreq->commit_data->req = (struct nfs_page *) dreq; } @@ -535,47 +521,6 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode } #endif -static struct nfs_direct_req *nfs_direct_write_alloc(size_t nbytes, size_t wsize) -{ - struct list_head *list; - struct nfs_direct_req *dreq; - unsigned int wpages = (wsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - - dreq = nfs_direct_req_alloc(); - if (!dreq) - return NULL; - - list = &dreq->list; - for(;;) { - struct nfs_write_data *data = nfs_writedata_alloc(wpages); - - if (unlikely(!data)) { - while (!list_empty(list)) { - data = list_entry(list->next, - struct nfs_write_data, pages); - list_del(&data->pages); - nfs_writedata_free(data); - } - kref_put(&dreq->kref, nfs_direct_req_release); - return NULL; - } - - INIT_LIST_HEAD(&data->pages); - list_add(&data->pages, list); - - data->req = (struct nfs_page *) dreq; - dreq->outstanding++; - if (nbytes <= wsize) - break; - nbytes -= wsize; - } - - nfs_alloc_commit_data(dreq); - - kref_get(&dreq->kref); - return dreq; -} - static void nfs_direct_write_result(struct rpc_task *task, void *calldata) { struct nfs_write_data *data = calldata; @@ -605,8 +550,6 @@ static void nfs_direct_write_result(struct rpc_task *task, void *calldata) } } } - /* In case we have to resend */ - data->args.stable = NFS_FILE_SYNC; spin_unlock(&dreq->lock); } @@ -620,14 +563,8 @@ static void nfs_direct_write_release(void *calldata) struct nfs_write_data *data = calldata; struct nfs_direct_req *dreq = (struct nfs_direct_req *) data->req; - spin_lock(&dreq->lock); - if (--dreq->outstanding) { - spin_unlock(&dreq->lock); - return; - } - spin_unlock(&dreq->lock); - - nfs_direct_write_complete(dreq, data->inode); + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq, data->inode); } static const struct rpc_call_ops nfs_write_direct_ops = { @@ -636,41 +573,58 @@ static const struct rpc_call_ops nfs_write_direct_ops = { }; /* - * For each nfs_write_data struct that was allocated on the list, dispatch - * an NFS WRITE operation + * For each wsize'd chunk of the user's buffer, dispatch an NFS WRITE + * operation. If nfs_writedata_alloc() or get_user_pages() fails, + * bail and stop sending more writes. Write length accounting is + * handled automatically by nfs_direct_write_result(). Otherwise, if + * no requests have been sent, just return an error. */ -static void nfs_direct_write_schedule(struct nfs_direct_req *dreq, int sync) +static ssize_t nfs_direct_write_schedule(struct nfs_direct_req *dreq, unsigned long user_addr, size_t count, loff_t pos, int sync) { struct nfs_open_context *ctx = dreq->ctx; struct inode *inode = ctx->dentry->d_inode; - struct list_head *list = &dreq->list; - struct page **pages = dreq->pages; - size_t count = dreq->user_count; - loff_t pos = dreq->pos; size_t wsize = NFS_SERVER(inode)->wsize; - unsigned int curpage, pgbase; + unsigned int pgbase; + int result; + ssize_t started = 0; + + get_dreq(dreq); - curpage = 0; - pgbase = dreq->user_addr & ~PAGE_MASK; do { struct nfs_write_data *data; size_t bytes; - bytes = wsize; - if (count < wsize) - bytes = count; + pgbase = user_addr & ~PAGE_MASK; + bytes = min(wsize,count); + + result = -ENOMEM; + data = nfs_writedata_alloc(pgbase + bytes); + if (unlikely(!data)) + break; + + down_read(¤t->mm->mmap_sem); + result = get_user_pages(current, current->mm, user_addr, + data->npages, 0, 0, data->pagevec, NULL); + up_read(¤t->mm->mmap_sem); + if (unlikely(result < data->npages)) { + if (result > 0) + nfs_direct_release_pages(data->pagevec, result); + nfs_writedata_release(data); + break; + } + + get_dreq(dreq); - BUG_ON(list_empty(list)); - data = list_entry(list->next, struct nfs_write_data, pages); list_move_tail(&data->pages, &dreq->rewrite_list); + data->req = (struct nfs_page *) dreq; data->inode = inode; data->cred = ctx->cred; data->args.fh = NFS_FH(inode); data->args.context = ctx; data->args.offset = pos; data->args.pgbase = pgbase; - data->args.pages = &pages[curpage]; + data->args.pages = data->pagevec; data->args.count = bytes; data->res.fattr = &data->fattr; data->res.count = bytes; @@ -694,19 +648,29 @@ static void nfs_direct_write_schedule(struct nfs_direct_req *dreq, int sync) bytes, (unsigned long long)data->args.offset); + started += bytes; + user_addr += bytes; pos += bytes; + + /* FIXME: Remove this useless math from the final patch */ pgbase += bytes; - curpage += pgbase >> PAGE_SHIFT; pgbase &= ~PAGE_MASK; + BUG_ON(pgbase != (user_addr & ~PAGE_MASK)); count -= bytes; } while (count != 0); - BUG_ON(!list_empty(list)); + + if (put_dreq(dreq)) + nfs_direct_write_complete(dreq, inode); + + if (started) + return 0; + return result < 0 ? (ssize_t) result : -EFAULT; } -static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, size_t count, loff_t pos, struct page **pages, int nr_pages) +static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, size_t count, loff_t pos) { - ssize_t result; + ssize_t result = 0; sigset_t oldset; struct inode *inode = iocb->ki_filp->f_mapping->host; struct rpc_clnt *clnt = NFS_CLIENT(inode); @@ -714,17 +678,14 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz size_t wsize = NFS_SERVER(inode)->wsize; int sync = 0; - dreq = nfs_direct_write_alloc(count, wsize); + dreq = nfs_direct_req_alloc(); if (!dreq) return -ENOMEM; + nfs_alloc_commit_data(dreq); + if (dreq->commit_data == NULL || count < wsize) sync = FLUSH_STABLE; - dreq->user_addr = user_addr; - dreq->user_count = count; - dreq->pos = pos; - dreq->pages = pages; - dreq->npages = nr_pages; dreq->inode = inode; dreq->ctx = get_nfs_open_context((struct nfs_open_context *)iocb->ki_filp->private_data); if (!is_sync_kiocb(iocb)) @@ -735,8 +696,9 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz nfs_begin_data_update(inode); rpc_clnt_sigmask(clnt, &oldset); - nfs_direct_write_schedule(dreq, sync); - result = nfs_direct_wait(dreq); + result = nfs_direct_write_schedule(dreq, user_addr, count, pos, sync); + if (!result) + result = nfs_direct_wait(dreq); rpc_clnt_sigunmask(clnt, &oldset); return result; @@ -766,8 +728,6 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, unsigned long user_addr, siz ssize_t nfs_file_direct_read(struct kiocb *iocb, char __user *buf, size_t count, loff_t pos) { ssize_t retval = -EINVAL; - int page_count; - struct page **pages; struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -789,14 +749,7 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, char __user *buf, size_t count, if (retval) goto out; - retval = nfs_get_user_pages(READ, (unsigned long) buf, - count, &pages); - if (retval < 0) - goto out; - page_count = retval; - - retval = nfs_direct_read(iocb, (unsigned long) buf, count, pos, - pages, page_count); + retval = nfs_direct_read(iocb, (unsigned long) buf, count, pos); if (retval > 0) iocb->ki_pos = pos + retval; @@ -832,8 +785,6 @@ out: ssize_t nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t pos) { ssize_t retval; - int page_count; - struct page **pages; struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -861,14 +812,7 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, size_t if (retval) goto out; - retval = nfs_get_user_pages(WRITE, (unsigned long) buf, - count, &pages); - if (retval < 0) - goto out; - page_count = retval; - - retval = nfs_direct_write(iocb, (unsigned long) buf, count, - pos, pages, page_count); + retval = nfs_direct_write(iocb, (unsigned long) buf, count, pos); /* * XXX: nfs_end_data_update() already ensures this file's @@ -892,7 +836,7 @@ out: * nfs_init_directcache - create a slab cache for nfs_direct_req structures * */ -int nfs_init_directcache(void) +int __init nfs_init_directcache(void) { nfs_direct_cachep = kmem_cache_create("nfs_direct_cache", sizeof(struct nfs_direct_req), @@ -906,11 +850,10 @@ int nfs_init_directcache(void) } /** - * nfs_init_directcache - destroy the slab cache for nfs_direct_req structures + * nfs_destroy_directcache - destroy the slab cache for nfs_direct_req structures * */ void nfs_destroy_directcache(void) { - if (kmem_cache_destroy(nfs_direct_cachep)) - printk(KERN_INFO "nfs_direct_cache: not all structures were freed\n"); + kmem_cache_destroy(nfs_direct_cachep); } diff --git a/fs/nfs/file.c b/fs/nfs/file.c index fa05c027ea1..be997d64912 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -111,7 +111,7 @@ nfs_file_open(struct inode *inode, struct file *filp) nfs_inc_stats(inode, NFSIOS_VFSOPEN); lock_kernel(); - res = NFS_SERVER(inode)->rpc_ops->file_open(inode, filp); + res = NFS_PROTO(inode)->file_open(inode, filp); unlock_kernel(); return res; } @@ -127,23 +127,6 @@ nfs_file_release(struct inode *inode, struct file *filp) } /** - * nfs_revalidate_file - Revalidate the page cache & related metadata - * @inode - pointer to inode struct - * @file - pointer to file - */ -static int nfs_revalidate_file(struct inode *inode, struct file *filp) -{ - struct nfs_inode *nfsi = NFS_I(inode); - int retval = 0; - - if ((nfsi->cache_validity & (NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_ATTR)) - || nfs_attribute_timeout(inode)) - retval = __nfs_revalidate_inode(NFS_SERVER(inode), inode); - nfs_revalidate_mapping(inode, filp->f_mapping); - return 0; -} - -/** * nfs_revalidate_size - Revalidate the file size * @inode - pointer to inode struct * @file - pointer to struct file @@ -174,7 +157,7 @@ force_reval: static loff_t nfs_file_llseek(struct file *filp, loff_t offset, int origin) { /* origin == SEEK_END => we must revalidate the cached file length */ - if (origin == 2) { + if (origin == SEEK_END) { struct inode *inode = filp->f_mapping->host; int retval = nfs_revalidate_file_size(inode, filp); if (retval < 0) @@ -228,7 +211,7 @@ nfs_file_read(struct kiocb *iocb, char __user * buf, size_t count, loff_t pos) dentry->d_parent->d_name.name, dentry->d_name.name, (unsigned long) count, (unsigned long) pos); - result = nfs_revalidate_file(inode, iocb->ki_filp); + result = nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping); nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, count); if (!result) result = generic_file_aio_read(iocb, buf, count, pos); @@ -247,7 +230,7 @@ nfs_file_sendfile(struct file *filp, loff_t *ppos, size_t count, dentry->d_parent->d_name.name, dentry->d_name.name, (unsigned long) count, (unsigned long long) *ppos); - res = nfs_revalidate_file(inode, filp); + res = nfs_revalidate_mapping(inode, filp->f_mapping); if (!res) res = generic_file_sendfile(filp, ppos, count, actor, target); return res; @@ -263,7 +246,7 @@ nfs_file_mmap(struct file * file, struct vm_area_struct * vma) dfprintk(VFS, "nfs: mmap(%s/%s)\n", dentry->d_parent->d_name.name, dentry->d_name.name); - status = nfs_revalidate_file(inode, file); + status = nfs_revalidate_mapping(inode, file->f_mapping); if (!status) status = generic_file_mmap(file, vma); return status; @@ -320,15 +303,25 @@ static int nfs_commit_write(struct file *file, struct page *page, unsigned offse static void nfs_invalidate_page(struct page *page, unsigned long offset) { - /* FIXME: we really should cancel any unstarted writes on this page */ + struct inode *inode = page->mapping->host; + + /* Cancel any unstarted writes on this page */ + if (offset == 0) + nfs_sync_inode_wait(inode, page->index, 1, FLUSH_INVALIDATE); } static int nfs_release_page(struct page *page, gfp_t gfp) { - return !nfs_wb_page(page->mapping->host, page); + if (gfp & __GFP_FS) + return !nfs_wb_page(page->mapping->host, page); + else + /* + * Avoid deadlock on nfs_wait_on_request(). + */ + return 0; } -struct address_space_operations nfs_file_aops = { +const struct address_space_operations nfs_file_aops = { .readpage = nfs_readpage, .readpages = nfs_readpages, .set_page_dirty = __set_page_dirty_nobuffers, @@ -373,7 +366,6 @@ nfs_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t if (result) goto out; } - nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping); result = count; if (!count) diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c new file mode 100644 index 00000000000..76b08ae9ed8 --- /dev/null +++ b/fs/nfs/getroot.c @@ -0,0 +1,311 @@ +/* getroot.c: get the root dentry for an NFS mount + * + * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/init.h> + +#include <linux/time.h> +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/string.h> +#include <linux/stat.h> +#include <linux/errno.h> +#include <linux/unistd.h> +#include <linux/sunrpc/clnt.h> +#include <linux/sunrpc/stats.h> +#include <linux/nfs_fs.h> +#include <linux/nfs_mount.h> +#include <linux/nfs4_mount.h> +#include <linux/lockd/bind.h> +#include <linux/smp_lock.h> +#include <linux/seq_file.h> +#include <linux/mount.h> +#include <linux/nfs_idmap.h> +#include <linux/vfs.h> +#include <linux/namei.h> +#include <linux/namespace.h> +#include <linux/security.h> + +#include <asm/system.h> +#include <asm/uaccess.h> + +#include "nfs4_fs.h" +#include "delegation.h" +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_CLIENT +#define NFS_PARANOIA 1 + +/* + * get an NFS2/NFS3 root dentry from the root filehandle + */ +struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh) +{ + struct nfs_server *server = NFS_SB(sb); + struct nfs_fsinfo fsinfo; + struct nfs_fattr fattr; + struct dentry *mntroot; + struct inode *inode; + int error; + + /* create a dummy root dentry with dummy inode for this superblock */ + if (!sb->s_root) { + struct nfs_fh dummyfh; + struct dentry *root; + struct inode *iroot; + + memset(&dummyfh, 0, sizeof(dummyfh)); + memset(&fattr, 0, sizeof(fattr)); + nfs_fattr_init(&fattr); + fattr.valid = NFS_ATTR_FATTR; + fattr.type = NFDIR; + fattr.mode = S_IFDIR | S_IRUSR | S_IWUSR; + fattr.nlink = 2; + + iroot = nfs_fhget(sb, &dummyfh, &fattr); + if (IS_ERR(iroot)) + return ERR_PTR(PTR_ERR(iroot)); + + root = d_alloc_root(iroot); + if (!root) { + iput(iroot); + return ERR_PTR(-ENOMEM); + } + + sb->s_root = root; + } + + /* get the actual root for this mount */ + fsinfo.fattr = &fattr; + + error = server->nfs_client->rpc_ops->getroot(server, mntfh, &fsinfo); + if (error < 0) { + dprintk("nfs_get_root: getattr error = %d\n", -error); + return ERR_PTR(error); + } + + inode = nfs_fhget(sb, mntfh, fsinfo.fattr); + if (IS_ERR(inode)) { + dprintk("nfs_get_root: get root inode failed\n"); + return ERR_PTR(PTR_ERR(inode)); + } + + /* root dentries normally start off anonymous and get spliced in later + * if the dentry tree reaches them; however if the dentry already + * exists, we'll pick it up at this point and use it as the root + */ + mntroot = d_alloc_anon(inode); + if (!mntroot) { + iput(inode); + dprintk("nfs_get_root: get root dentry failed\n"); + return ERR_PTR(-ENOMEM); + } + + security_d_instantiate(mntroot, inode); + + if (!mntroot->d_op) + mntroot->d_op = server->nfs_client->rpc_ops->dentry_ops; + + return mntroot; +} + +#ifdef CONFIG_NFS_V4 + +/* + * Do a simple pathwalk from the root FH of the server to the nominated target + * of the mountpoint + * - give error on symlinks + * - give error on ".." occurring in the path + * - follow traversals + */ +int nfs4_path_walk(struct nfs_server *server, + struct nfs_fh *mntfh, + const char *path) +{ + struct nfs_fsinfo fsinfo; + struct nfs_fattr fattr; + struct nfs_fh lastfh; + struct qstr name; + int ret; + //int referral_count = 0; + + dprintk("--> nfs4_path_walk(,,%s)\n", path); + + fsinfo.fattr = &fattr; + nfs_fattr_init(&fattr); + + if (*path++ != '/') { + dprintk("nfs4_get_root: Path does not begin with a slash\n"); + return -EINVAL; + } + + /* Start by getting the root filehandle from the server */ + ret = server->nfs_client->rpc_ops->getroot(server, mntfh, &fsinfo); + if (ret < 0) { + dprintk("nfs4_get_root: getroot error = %d\n", -ret); + return ret; + } + + if (fattr.type != NFDIR) { + printk(KERN_ERR "nfs4_get_root:" + " getroot encountered non-directory\n"); + return -ENOTDIR; + } + + if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL) { + printk(KERN_ERR "nfs4_get_root:" + " getroot obtained referral\n"); + return -EREMOTE; + } + +next_component: + dprintk("Next: %s\n", path); + + /* extract the next bit of the path */ + if (!*path) + goto path_walk_complete; + + name.name = path; + while (*path && *path != '/') + path++; + name.len = path - (const char *) name.name; + +eat_dot_dir: + while (*path == '/') + path++; + + if (path[0] == '.' && (path[1] == '/' || !path[1])) { + path += 2; + goto eat_dot_dir; + } + + if (path[0] == '.' && path[1] == '.' && (path[2] == '/' || !path[2]) + ) { + printk(KERN_ERR "nfs4_get_root:" + " Mount path contains reference to \"..\"\n"); + return -EINVAL; + } + + /* lookup the next FH in the sequence */ + memcpy(&lastfh, mntfh, sizeof(lastfh)); + + dprintk("LookupFH: %*.*s [%s]\n", name.len, name.len, name.name, path); + + ret = server->nfs_client->rpc_ops->lookupfh(server, &lastfh, &name, + mntfh, &fattr); + if (ret < 0) { + dprintk("nfs4_get_root: getroot error = %d\n", -ret); + return ret; + } + + if (fattr.type != NFDIR) { + printk(KERN_ERR "nfs4_get_root:" + " lookupfh encountered non-directory\n"); + return -ENOTDIR; + } + + if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL) { + printk(KERN_ERR "nfs4_get_root:" + " lookupfh obtained referral\n"); + return -EREMOTE; + } + + goto next_component; + +path_walk_complete: + memcpy(&server->fsid, &fattr.fsid, sizeof(server->fsid)); + dprintk("<-- nfs4_path_walk() = 0\n"); + return 0; +} + +/* + * get an NFS4 root dentry from the root filehandle + */ +struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh) +{ + struct nfs_server *server = NFS_SB(sb); + struct nfs_fattr fattr; + struct dentry *mntroot; + struct inode *inode; + int error; + + dprintk("--> nfs4_get_root()\n"); + + /* create a dummy root dentry with dummy inode for this superblock */ + if (!sb->s_root) { + struct nfs_fh dummyfh; + struct dentry *root; + struct inode *iroot; + + memset(&dummyfh, 0, sizeof(dummyfh)); + memset(&fattr, 0, sizeof(fattr)); + nfs_fattr_init(&fattr); + fattr.valid = NFS_ATTR_FATTR; + fattr.type = NFDIR; + fattr.mode = S_IFDIR | S_IRUSR | S_IWUSR; + fattr.nlink = 2; + + iroot = nfs_fhget(sb, &dummyfh, &fattr); + if (IS_ERR(iroot)) + return ERR_PTR(PTR_ERR(iroot)); + + root = d_alloc_root(iroot); + if (!root) { + iput(iroot); + return ERR_PTR(-ENOMEM); + } + + sb->s_root = root; + } + + /* get the info about the server and filesystem */ + error = nfs4_server_capabilities(server, mntfh); + if (error < 0) { + dprintk("nfs_get_root: getcaps error = %d\n", + -error); + return ERR_PTR(error); + } + + /* get the actual root for this mount */ + error = server->nfs_client->rpc_ops->getattr(server, mntfh, &fattr); + if (error < 0) { + dprintk("nfs_get_root: getattr error = %d\n", -error); + return ERR_PTR(error); + } + + inode = nfs_fhget(sb, mntfh, &fattr); + if (IS_ERR(inode)) { + dprintk("nfs_get_root: get root inode failed\n"); + return ERR_PTR(PTR_ERR(inode)); + } + + /* root dentries normally start off anonymous and get spliced in later + * if the dentry tree reaches them; however if the dentry already + * exists, we'll pick it up at this point and use it as the root + */ + mntroot = d_alloc_anon(inode); + if (!mntroot) { + iput(inode); + dprintk("nfs_get_root: get root dentry failed\n"); + return ERR_PTR(-ENOMEM); + } + + security_d_instantiate(mntroot, inode); + + if (!mntroot->d_op) + mntroot->d_op = server->nfs_client->rpc_ops->dentry_ops; + + dprintk("<-- nfs4_get_root()\n"); + return mntroot; +} + +#endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c index 3fab5b0cfc5..82ad7110a1c 100644 --- a/fs/nfs/idmap.c +++ b/fs/nfs/idmap.c @@ -47,7 +47,6 @@ #include <linux/workqueue.h> #include <linux/sunrpc/rpc_pipe_fs.h> -#include <linux/nfs_fs_sb.h> #include <linux/nfs_fs.h> #include <linux/nfs_idmap.h> @@ -58,6 +57,20 @@ /* Default cache timeout is 10 minutes */ unsigned int nfs_idmap_cache_timeout = 600 * HZ; +static int param_set_idmap_timeout(const char *val, struct kernel_param *kp) +{ + char *endp; + int num = simple_strtol(val, &endp, 0); + int jif = num * HZ; + if (endp == val || *endp || num < 0 || jif < num) + return -EINVAL; + *((int *)kp->arg) = jif; + return 0; +} + +module_param_call(idmap_cache_timeout, param_set_idmap_timeout, param_get_int, + &nfs_idmap_cache_timeout, 0644); + struct idmap_hashent { unsigned long ih_expires; __u32 ih_id; @@ -71,7 +84,6 @@ struct idmap_hashtable { }; struct idmap { - char idmap_path[48]; struct dentry *idmap_dentry; wait_queue_head_t idmap_wq; struct idmap_msg idmap_im; @@ -95,24 +107,23 @@ static struct rpc_pipe_ops idmap_upcall_ops = { .destroy_msg = idmap_pipe_destroy_msg, }; -void -nfs_idmap_new(struct nfs4_client *clp) +int +nfs_idmap_new(struct nfs_client *clp) { struct idmap *idmap; + int error; - if (clp->cl_idmap != NULL) - return; - if ((idmap = kzalloc(sizeof(*idmap), GFP_KERNEL)) == NULL) - return; + BUG_ON(clp->cl_idmap != NULL); - snprintf(idmap->idmap_path, sizeof(idmap->idmap_path), - "%s/idmap", clp->cl_rpcclient->cl_pathname); + if ((idmap = kzalloc(sizeof(*idmap), GFP_KERNEL)) == NULL) + return -ENOMEM; - idmap->idmap_dentry = rpc_mkpipe(idmap->idmap_path, + idmap->idmap_dentry = rpc_mkpipe(clp->cl_rpcclient->cl_dentry, "idmap", idmap, &idmap_upcall_ops, 0); if (IS_ERR(idmap->idmap_dentry)) { + error = PTR_ERR(idmap->idmap_dentry); kfree(idmap); - return; + return error; } mutex_init(&idmap->idmap_lock); @@ -122,18 +133,17 @@ nfs_idmap_new(struct nfs4_client *clp) idmap->idmap_group_hash.h_type = IDMAP_TYPE_GROUP; clp->cl_idmap = idmap; + return 0; } void -nfs_idmap_delete(struct nfs4_client *clp) +nfs_idmap_delete(struct nfs_client *clp) { struct idmap *idmap = clp->cl_idmap; if (!idmap) return; - dput(idmap->idmap_dentry); - idmap->idmap_dentry = NULL; - rpc_unlink(idmap->idmap_path); + rpc_unlink(idmap->idmap_dentry); clp->cl_idmap = NULL; kfree(idmap); } @@ -480,27 +490,27 @@ static unsigned int fnvhash32(const void *buf, size_t buflen) return (hash); } -int nfs_map_name_to_uid(struct nfs4_client *clp, const char *name, size_t namelen, __u32 *uid) +int nfs_map_name_to_uid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid) { struct idmap *idmap = clp->cl_idmap; return nfs_idmap_id(idmap, &idmap->idmap_user_hash, name, namelen, uid); } -int nfs_map_group_to_gid(struct nfs4_client *clp, const char *name, size_t namelen, __u32 *uid) +int nfs_map_group_to_gid(struct nfs_client *clp, const char *name, size_t namelen, __u32 *uid) { struct idmap *idmap = clp->cl_idmap; return nfs_idmap_id(idmap, &idmap->idmap_group_hash, name, namelen, uid); } -int nfs_map_uid_to_name(struct nfs4_client *clp, __u32 uid, char *buf) +int nfs_map_uid_to_name(struct nfs_client *clp, __u32 uid, char *buf) { struct idmap *idmap = clp->cl_idmap; return nfs_idmap_name(idmap, &idmap->idmap_user_hash, uid, buf); } -int nfs_map_gid_to_group(struct nfs4_client *clp, __u32 uid, char *buf) +int nfs_map_gid_to_group(struct nfs_client *clp, __u32 uid, char *buf) { struct idmap *idmap = clp->cl_idmap; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 937fbfc381b..bc9376ca86c 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -13,7 +13,6 @@ * */ -#include <linux/config.h> #include <linux/module.h> #include <linux/init.h> @@ -36,6 +35,8 @@ #include <linux/mount.h> #include <linux/nfs_idmap.h> #include <linux/vfs.h> +#include <linux/inet.h> +#include <linux/nfs_xdr.h> #include <asm/system.h> #include <asm/uaccess.h> @@ -44,89 +45,17 @@ #include "callback.h" #include "delegation.h" #include "iostat.h" +#include "internal.h" #define NFSDBG_FACILITY NFSDBG_VFS #define NFS_PARANOIA 1 -/* Maximum number of readahead requests - * FIXME: this should really be a sysctl so that users may tune it to suit - * their needs. People that do NFS over a slow network, might for - * instance want to reduce it to something closer to 1 for improved - * interactive response. - */ -#define NFS_MAX_READAHEAD (RPC_DEF_SLOT_TABLE - 1) - static void nfs_invalidate_inode(struct inode *); static int nfs_update_inode(struct inode *, struct nfs_fattr *); -static struct inode *nfs_alloc_inode(struct super_block *sb); -static void nfs_destroy_inode(struct inode *); -static int nfs_write_inode(struct inode *,int); -static void nfs_delete_inode(struct inode *); -static void nfs_clear_inode(struct inode *); -static void nfs_umount_begin(struct super_block *); -static int nfs_statfs(struct dentry *, struct kstatfs *); -static int nfs_show_options(struct seq_file *, struct vfsmount *); -static int nfs_show_stats(struct seq_file *, struct vfsmount *); static void nfs_zap_acl_cache(struct inode *); -static struct rpc_program nfs_program; - -static struct super_operations nfs_sops = { - .alloc_inode = nfs_alloc_inode, - .destroy_inode = nfs_destroy_inode, - .write_inode = nfs_write_inode, - .delete_inode = nfs_delete_inode, - .statfs = nfs_statfs, - .clear_inode = nfs_clear_inode, - .umount_begin = nfs_umount_begin, - .show_options = nfs_show_options, - .show_stats = nfs_show_stats, -}; - -/* - * RPC cruft for NFS - */ -static struct rpc_stat nfs_rpcstat = { - .program = &nfs_program -}; -static struct rpc_version * nfs_version[] = { - NULL, - NULL, - &nfs_version2, -#if defined(CONFIG_NFS_V3) - &nfs_version3, -#elif defined(CONFIG_NFS_V4) - NULL, -#endif -#if defined(CONFIG_NFS_V4) - &nfs_version4, -#endif -}; - -static struct rpc_program nfs_program = { - .name = "nfs", - .number = NFS_PROGRAM, - .nrvers = ARRAY_SIZE(nfs_version), - .version = nfs_version, - .stats = &nfs_rpcstat, - .pipe_dir_name = "/nfs", -}; - -#ifdef CONFIG_NFS_V3_ACL -static struct rpc_stat nfsacl_rpcstat = { &nfsacl_program }; -static struct rpc_version * nfsacl_version[] = { - [3] = &nfsacl_version3, -}; - -struct rpc_program nfsacl_program = { - .name = "nfsacl", - .number = NFS_ACL_PROGRAM, - .nrvers = ARRAY_SIZE(nfsacl_version), - .version = nfsacl_version, - .stats = &nfsacl_rpcstat, -}; -#endif /* CONFIG_NFS_V3_ACL */ +static kmem_cache_t * nfs_inode_cachep; static inline unsigned long nfs_fattr_to_ino_t(struct nfs_fattr *fattr) @@ -134,8 +63,7 @@ nfs_fattr_to_ino_t(struct nfs_fattr *fattr) return nfs_fileid_to_ino_t(fattr->fileid); } -static int -nfs_write_inode(struct inode *inode, int sync) +int nfs_write_inode(struct inode *inode, int sync) { int flags = sync ? FLUSH_SYNC : 0; int ret; @@ -146,586 +74,16 @@ nfs_write_inode(struct inode *inode, int sync) return 0; } -static void -nfs_delete_inode(struct inode * inode) +void nfs_clear_inode(struct inode *inode) { - dprintk("NFS: delete_inode(%s/%ld)\n", inode->i_sb->s_id, inode->i_ino); - - truncate_inode_pages(&inode->i_data, 0); - - nfs_wb_all(inode); /* * The following should never happen... */ - if (nfs_have_writebacks(inode)) { - printk(KERN_ERR "nfs_delete_inode: inode %ld has pending RPC requests\n", inode->i_ino); - } - - clear_inode(inode); -} - -static void -nfs_clear_inode(struct inode *inode) -{ - struct nfs_inode *nfsi = NFS_I(inode); - struct rpc_cred *cred; - - nfs_wb_all(inode); - BUG_ON (!list_empty(&nfsi->open_files)); + BUG_ON(nfs_have_writebacks(inode)); + BUG_ON(!list_empty(&NFS_I(inode)->open_files)); + BUG_ON(atomic_read(&NFS_I(inode)->data_updates) != 0); nfs_zap_acl_cache(inode); - cred = nfsi->cache_access.cred; - if (cred) - put_rpccred(cred); - BUG_ON(atomic_read(&nfsi->data_updates) != 0); -} - -void -nfs_umount_begin(struct super_block *sb) -{ - struct rpc_clnt *rpc = NFS_SB(sb)->client; - - /* -EIO all pending I/O */ - if (!IS_ERR(rpc)) - rpc_killall_tasks(rpc); - rpc = NFS_SB(sb)->client_acl; - if (!IS_ERR(rpc)) - rpc_killall_tasks(rpc); -} - - -static inline unsigned long -nfs_block_bits(unsigned long bsize, unsigned char *nrbitsp) -{ - /* make sure blocksize is a power of two */ - if ((bsize & (bsize - 1)) || nrbitsp) { - unsigned char nrbits; - - for (nrbits = 31; nrbits && !(bsize & (1 << nrbits)); nrbits--) - ; - bsize = 1 << nrbits; - if (nrbitsp) - *nrbitsp = nrbits; - } - - return bsize; -} - -/* - * Calculate the number of 512byte blocks used. - */ -static inline unsigned long -nfs_calc_block_size(u64 tsize) -{ - loff_t used = (tsize + 511) >> 9; - return (used > ULONG_MAX) ? ULONG_MAX : used; -} - -/* - * Compute and set NFS server blocksize - */ -static inline unsigned long -nfs_block_size(unsigned long bsize, unsigned char *nrbitsp) -{ - if (bsize < NFS_MIN_FILE_IO_SIZE) - bsize = NFS_DEF_FILE_IO_SIZE; - else if (bsize >= NFS_MAX_FILE_IO_SIZE) - bsize = NFS_MAX_FILE_IO_SIZE; - - return nfs_block_bits(bsize, nrbitsp); -} - -/* - * Obtain the root inode of the file system. - */ -static struct inode * -nfs_get_root(struct super_block *sb, struct nfs_fh *rootfh, struct nfs_fsinfo *fsinfo) -{ - struct nfs_server *server = NFS_SB(sb); - int error; - - error = server->rpc_ops->getroot(server, rootfh, fsinfo); - if (error < 0) { - dprintk("nfs_get_root: getattr error = %d\n", -error); - return ERR_PTR(error); - } - - return nfs_fhget(sb, rootfh, fsinfo->fattr); -} - -/* - * Do NFS version-independent mount processing, and sanity checking - */ -static int -nfs_sb_init(struct super_block *sb, rpc_authflavor_t authflavor) -{ - struct nfs_server *server; - struct inode *root_inode; - struct nfs_fattr fattr; - struct nfs_fsinfo fsinfo = { - .fattr = &fattr, - }; - struct nfs_pathconf pathinfo = { - .fattr = &fattr, - }; - int no_root_error = 0; - unsigned long max_rpc_payload; - - /* We probably want something more informative here */ - snprintf(sb->s_id, sizeof(sb->s_id), "%x:%x", MAJOR(sb->s_dev), MINOR(sb->s_dev)); - - server = NFS_SB(sb); - - sb->s_magic = NFS_SUPER_MAGIC; - - server->io_stats = nfs_alloc_iostats(); - if (server->io_stats == NULL) - return -ENOMEM; - - root_inode = nfs_get_root(sb, &server->fh, &fsinfo); - /* Did getting the root inode fail? */ - if (IS_ERR(root_inode)) { - no_root_error = PTR_ERR(root_inode); - goto out_no_root; - } - sb->s_root = d_alloc_root(root_inode); - if (!sb->s_root) { - no_root_error = -ENOMEM; - goto out_no_root; - } - sb->s_root->d_op = server->rpc_ops->dentry_ops; - - /* mount time stamp, in seconds */ - server->mount_time = jiffies; - - /* Get some general file system info */ - if (server->namelen == 0 && - server->rpc_ops->pathconf(server, &server->fh, &pathinfo) >= 0) - server->namelen = pathinfo.max_namelen; - /* Work out a lot of parameters */ - if (server->rsize == 0) - server->rsize = nfs_block_size(fsinfo.rtpref, NULL); - if (server->wsize == 0) - server->wsize = nfs_block_size(fsinfo.wtpref, NULL); - - if (fsinfo.rtmax >= 512 && server->rsize > fsinfo.rtmax) - server->rsize = nfs_block_size(fsinfo.rtmax, NULL); - if (fsinfo.wtmax >= 512 && server->wsize > fsinfo.wtmax) - server->wsize = nfs_block_size(fsinfo.wtmax, NULL); - - max_rpc_payload = nfs_block_size(rpc_max_payload(server->client), NULL); - if (server->rsize > max_rpc_payload) - server->rsize = max_rpc_payload; - if (server->rsize > NFS_MAX_FILE_IO_SIZE) - server->rsize = NFS_MAX_FILE_IO_SIZE; - server->rpages = (server->rsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - - if (server->wsize > max_rpc_payload) - server->wsize = max_rpc_payload; - if (server->wsize > NFS_MAX_FILE_IO_SIZE) - server->wsize = NFS_MAX_FILE_IO_SIZE; - server->wpages = (server->wsize + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - - if (sb->s_blocksize == 0) - sb->s_blocksize = nfs_block_bits(server->wsize, - &sb->s_blocksize_bits); - server->wtmult = nfs_block_bits(fsinfo.wtmult, NULL); - - server->dtsize = nfs_block_size(fsinfo.dtpref, NULL); - if (server->dtsize > PAGE_CACHE_SIZE) - server->dtsize = PAGE_CACHE_SIZE; - if (server->dtsize > server->rsize) - server->dtsize = server->rsize; - - if (server->flags & NFS_MOUNT_NOAC) { - server->acregmin = server->acregmax = 0; - server->acdirmin = server->acdirmax = 0; - sb->s_flags |= MS_SYNCHRONOUS; - } - server->backing_dev_info.ra_pages = server->rpages * NFS_MAX_READAHEAD; - - sb->s_maxbytes = fsinfo.maxfilesize; - if (sb->s_maxbytes > MAX_LFS_FILESIZE) - sb->s_maxbytes = MAX_LFS_FILESIZE; - - server->client->cl_intr = (server->flags & NFS_MOUNT_INTR) ? 1 : 0; - server->client->cl_softrtry = (server->flags & NFS_MOUNT_SOFT) ? 1 : 0; - - /* We're airborne Set socket buffersize */ - rpc_setbufsize(server->client, server->wsize + 100, server->rsize + 100); - return 0; - /* Yargs. It didn't work out. */ -out_no_root: - dprintk("nfs_sb_init: get root inode failed: errno %d\n", -no_root_error); - if (!IS_ERR(root_inode)) - iput(root_inode); - return no_root_error; -} - -static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, unsigned int timeo, unsigned int retrans) -{ - to->to_initval = timeo * HZ / 10; - to->to_retries = retrans; - if (!to->to_retries) - to->to_retries = 2; - - switch (proto) { - case IPPROTO_TCP: - if (!to->to_initval) - to->to_initval = 60 * HZ; - if (to->to_initval > NFS_MAX_TCP_TIMEOUT) - to->to_initval = NFS_MAX_TCP_TIMEOUT; - to->to_increment = to->to_initval; - to->to_maxval = to->to_initval + (to->to_increment * to->to_retries); - to->to_exponential = 0; - break; - case IPPROTO_UDP: - default: - if (!to->to_initval) - to->to_initval = 11 * HZ / 10; - if (to->to_initval > NFS_MAX_UDP_TIMEOUT) - to->to_initval = NFS_MAX_UDP_TIMEOUT; - to->to_maxval = NFS_MAX_UDP_TIMEOUT; - to->to_exponential = 1; - break; - } -} - -/* - * Create an RPC client handle. - */ -static struct rpc_clnt * -nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data) -{ - struct rpc_timeout timeparms; - struct rpc_xprt *xprt = NULL; - struct rpc_clnt *clnt = NULL; - int proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP; - - nfs_init_timeout_values(&timeparms, proto, data->timeo, data->retrans); - - server->retrans_timeo = timeparms.to_initval; - server->retrans_count = timeparms.to_retries; - - /* create transport and client */ - xprt = xprt_create_proto(proto, &server->addr, &timeparms); - if (IS_ERR(xprt)) { - dprintk("%s: cannot create RPC transport. Error = %ld\n", - __FUNCTION__, PTR_ERR(xprt)); - return (struct rpc_clnt *)xprt; - } - clnt = rpc_create_client(xprt, server->hostname, &nfs_program, - server->rpc_ops->version, data->pseudoflavor); - if (IS_ERR(clnt)) { - dprintk("%s: cannot create RPC client. Error = %ld\n", - __FUNCTION__, PTR_ERR(xprt)); - goto out_fail; - } - - clnt->cl_intr = 1; - clnt->cl_softrtry = 1; - - return clnt; - -out_fail: - return clnt; -} - -/* - * The way this works is that the mount process passes a structure - * in the data argument which contains the server's IP address - * and the root file handle obtained from the server's mount - * daemon. We stash these away in the private superblock fields. - */ -static int -nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data, int silent) -{ - struct nfs_server *server; - rpc_authflavor_t authflavor; - - server = NFS_SB(sb); - sb->s_blocksize_bits = 0; - sb->s_blocksize = 0; - if (data->bsize) - sb->s_blocksize = nfs_block_size(data->bsize, &sb->s_blocksize_bits); - if (data->rsize) - server->rsize = nfs_block_size(data->rsize, NULL); - if (data->wsize) - server->wsize = nfs_block_size(data->wsize, NULL); - server->flags = data->flags & NFS_MOUNT_FLAGMASK; - - server->acregmin = data->acregmin*HZ; - server->acregmax = data->acregmax*HZ; - server->acdirmin = data->acdirmin*HZ; - server->acdirmax = data->acdirmax*HZ; - - /* Start lockd here, before we might error out */ - if (!(server->flags & NFS_MOUNT_NONLM)) - lockd_up(); - - server->namelen = data->namlen; - server->hostname = kmalloc(strlen(data->hostname) + 1, GFP_KERNEL); - if (!server->hostname) - return -ENOMEM; - strcpy(server->hostname, data->hostname); - - /* Check NFS protocol revision and initialize RPC op vector - * and file handle pool. */ -#ifdef CONFIG_NFS_V3 - if (server->flags & NFS_MOUNT_VER3) { - server->rpc_ops = &nfs_v3_clientops; - server->caps |= NFS_CAP_READDIRPLUS; - } else { - server->rpc_ops = &nfs_v2_clientops; - } -#else - server->rpc_ops = &nfs_v2_clientops; -#endif - - /* Fill in pseudoflavor for mount version < 5 */ - if (!(data->flags & NFS_MOUNT_SECFLAVOUR)) - data->pseudoflavor = RPC_AUTH_UNIX; - authflavor = data->pseudoflavor; /* save for sb_init() */ - /* XXX maybe we want to add a server->pseudoflavor field */ - - /* Create RPC client handles */ - server->client = nfs_create_client(server, data); - if (IS_ERR(server->client)) - return PTR_ERR(server->client); - /* RFC 2623, sec 2.3.2 */ - if (authflavor != RPC_AUTH_UNIX) { - struct rpc_auth *auth; - - server->client_sys = rpc_clone_client(server->client); - if (IS_ERR(server->client_sys)) - return PTR_ERR(server->client_sys); - auth = rpcauth_create(RPC_AUTH_UNIX, server->client_sys); - if (IS_ERR(auth)) - return PTR_ERR(auth); - } else { - atomic_inc(&server->client->cl_count); - server->client_sys = server->client; - } - if (server->flags & NFS_MOUNT_VER3) { -#ifdef CONFIG_NFS_V3_ACL - if (!(server->flags & NFS_MOUNT_NOACL)) { - server->client_acl = rpc_bind_new_program(server->client, &nfsacl_program, 3); - /* No errors! Assume that Sun nfsacls are supported */ - if (!IS_ERR(server->client_acl)) - server->caps |= NFS_CAP_ACLS; - } -#else - server->flags &= ~NFS_MOUNT_NOACL; -#endif /* CONFIG_NFS_V3_ACL */ - /* - * The VFS shouldn't apply the umask to mode bits. We will - * do so ourselves when necessary. - */ - sb->s_flags |= MS_POSIXACL; - if (server->namelen == 0 || server->namelen > NFS3_MAXNAMLEN) - server->namelen = NFS3_MAXNAMLEN; - sb->s_time_gran = 1; - } else { - if (server->namelen == 0 || server->namelen > NFS2_MAXNAMLEN) - server->namelen = NFS2_MAXNAMLEN; - } - - sb->s_op = &nfs_sops; - return nfs_sb_init(sb, authflavor); -} - -static int -nfs_statfs(struct dentry *dentry, struct kstatfs *buf) -{ - struct super_block *sb = dentry->d_sb; - struct nfs_server *server = NFS_SB(sb); - unsigned char blockbits; - unsigned long blockres; - struct nfs_fh *rootfh = NFS_FH(sb->s_root->d_inode); - struct nfs_fattr fattr; - struct nfs_fsstat res = { - .fattr = &fattr, - }; - int error; - - lock_kernel(); - - error = server->rpc_ops->statfs(server, rootfh, &res); - buf->f_type = NFS_SUPER_MAGIC; - if (error < 0) - goto out_err; - - /* - * Current versions of glibc do not correctly handle the - * case where f_frsize != f_bsize. Eventually we want to - * report the value of wtmult in this field. - */ - buf->f_frsize = sb->s_blocksize; - - /* - * On most *nix systems, f_blocks, f_bfree, and f_bavail - * are reported in units of f_frsize. Linux hasn't had - * an f_frsize field in its statfs struct until recently, - * thus historically Linux's sys_statfs reports these - * fields in units of f_bsize. - */ - buf->f_bsize = sb->s_blocksize; - blockbits = sb->s_blocksize_bits; - blockres = (1 << blockbits) - 1; - buf->f_blocks = (res.tbytes + blockres) >> blockbits; - buf->f_bfree = (res.fbytes + blockres) >> blockbits; - buf->f_bavail = (res.abytes + blockres) >> blockbits; - - buf->f_files = res.tfiles; - buf->f_ffree = res.afiles; - - buf->f_namelen = server->namelen; - out: - unlock_kernel(); - return 0; - - out_err: - dprintk("%s: statfs error = %d\n", __FUNCTION__, -error); - buf->f_bsize = buf->f_blocks = buf->f_bfree = buf->f_bavail = -1; - goto out; - -} - -static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, int showdefaults) -{ - static struct proc_nfs_info { - int flag; - char *str; - char *nostr; - } nfs_info[] = { - { NFS_MOUNT_SOFT, ",soft", ",hard" }, - { NFS_MOUNT_INTR, ",intr", "" }, - { NFS_MOUNT_NOCTO, ",nocto", "" }, - { NFS_MOUNT_NOAC, ",noac", "" }, - { NFS_MOUNT_NONLM, ",nolock", "" }, - { NFS_MOUNT_NOACL, ",noacl", "" }, - { 0, NULL, NULL } - }; - struct proc_nfs_info *nfs_infop; - char buf[12]; - char *proto; - - seq_printf(m, ",vers=%d", nfss->rpc_ops->version); - seq_printf(m, ",rsize=%d", nfss->rsize); - seq_printf(m, ",wsize=%d", nfss->wsize); - if (nfss->acregmin != 3*HZ || showdefaults) - seq_printf(m, ",acregmin=%d", nfss->acregmin/HZ); - if (nfss->acregmax != 60*HZ || showdefaults) - seq_printf(m, ",acregmax=%d", nfss->acregmax/HZ); - if (nfss->acdirmin != 30*HZ || showdefaults) - seq_printf(m, ",acdirmin=%d", nfss->acdirmin/HZ); - if (nfss->acdirmax != 60*HZ || showdefaults) - seq_printf(m, ",acdirmax=%d", nfss->acdirmax/HZ); - for (nfs_infop = nfs_info; nfs_infop->flag; nfs_infop++) { - if (nfss->flags & nfs_infop->flag) - seq_puts(m, nfs_infop->str); - else - seq_puts(m, nfs_infop->nostr); - } - switch (nfss->client->cl_xprt->prot) { - case IPPROTO_TCP: - proto = "tcp"; - break; - case IPPROTO_UDP: - proto = "udp"; - break; - default: - snprintf(buf, sizeof(buf), "%u", nfss->client->cl_xprt->prot); - proto = buf; - } - seq_printf(m, ",proto=%s", proto); - seq_printf(m, ",timeo=%lu", 10U * nfss->retrans_timeo / HZ); - seq_printf(m, ",retrans=%u", nfss->retrans_count); -} - -static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) -{ - struct nfs_server *nfss = NFS_SB(mnt->mnt_sb); - - nfs_show_mount_options(m, nfss, 0); - - seq_puts(m, ",addr="); - seq_escape(m, nfss->hostname, " \t\n\\"); - - return 0; -} - -static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt) -{ - int i, cpu; - struct nfs_server *nfss = NFS_SB(mnt->mnt_sb); - struct rpc_auth *auth = nfss->client->cl_auth; - struct nfs_iostats totals = { }; - - seq_printf(m, "statvers=%s", NFS_IOSTAT_VERS); - - /* - * Display all mount option settings - */ - seq_printf(m, "\n\topts:\t"); - seq_puts(m, mnt->mnt_sb->s_flags & MS_RDONLY ? "ro" : "rw"); - seq_puts(m, mnt->mnt_sb->s_flags & MS_SYNCHRONOUS ? ",sync" : ""); - seq_puts(m, mnt->mnt_sb->s_flags & MS_NOATIME ? ",noatime" : ""); - seq_puts(m, mnt->mnt_sb->s_flags & MS_NODIRATIME ? ",nodiratime" : ""); - nfs_show_mount_options(m, nfss, 1); - - seq_printf(m, "\n\tage:\t%lu", (jiffies - nfss->mount_time) / HZ); - - seq_printf(m, "\n\tcaps:\t"); - seq_printf(m, "caps=0x%x", nfss->caps); - seq_printf(m, ",wtmult=%d", nfss->wtmult); - seq_printf(m, ",dtsize=%d", nfss->dtsize); - seq_printf(m, ",bsize=%d", nfss->bsize); - seq_printf(m, ",namelen=%d", nfss->namelen); - -#ifdef CONFIG_NFS_V4 - if (nfss->rpc_ops->version == 4) { - seq_printf(m, "\n\tnfsv4:\t"); - seq_printf(m, "bm0=0x%x", nfss->attr_bitmask[0]); - seq_printf(m, ",bm1=0x%x", nfss->attr_bitmask[1]); - seq_printf(m, ",acl=0x%x", nfss->acl_bitmask); - } -#endif - - /* - * Display security flavor in effect for this mount - */ - seq_printf(m, "\n\tsec:\tflavor=%d", auth->au_ops->au_flavor); - if (auth->au_flavor) - seq_printf(m, ",pseudoflavor=%d", auth->au_flavor); - - /* - * Display superblock I/O counters - */ - for_each_possible_cpu(cpu) { - struct nfs_iostats *stats; - - preempt_disable(); - stats = per_cpu_ptr(nfss->io_stats, cpu); - - for (i = 0; i < __NFSIOS_COUNTSMAX; i++) - totals.events[i] += stats->events[i]; - for (i = 0; i < __NFSIOS_BYTESMAX; i++) - totals.bytes[i] += stats->bytes[i]; - - preempt_enable(); - } - - seq_printf(m, "\n\tevents:\t"); - for (i = 0; i < __NFSIOS_COUNTSMAX; i++) - seq_printf(m, "%lu ", totals.events[i]); - seq_printf(m, "\n\tbytes:\t"); - for (i = 0; i < __NFSIOS_BYTESMAX; i++) - seq_printf(m, "%Lu ", totals.bytes[i]); - seq_printf(m, "\n"); - - rpc_print_iostats(m, nfss->client); - - return 0; + nfs_access_zap_cache(inode); } /** @@ -879,17 +237,25 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) /* Why so? Because we want revalidate for devices/FIFOs, and * that's precisely what we have in nfs_file_inode_operations. */ - inode->i_op = NFS_SB(sb)->rpc_ops->file_inode_ops; + inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->file_inode_ops; if (S_ISREG(inode->i_mode)) { inode->i_fop = &nfs_file_operations; inode->i_data.a_ops = &nfs_file_aops; inode->i_data.backing_dev_info = &NFS_SB(sb)->backing_dev_info; } else if (S_ISDIR(inode->i_mode)) { - inode->i_op = NFS_SB(sb)->rpc_ops->dir_inode_ops; + inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->dir_inode_ops; inode->i_fop = &nfs_dir_operations; if (nfs_server_capable(inode, NFS_CAP_READDIRPLUS) && fattr->size <= NFS_LIMIT_READDIRPLUS) set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode)); + /* Deal with crossing mountpoints */ + if (!nfs_fsid_equal(&NFS_SB(sb)->fsid, &fattr->fsid)) { + if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) + inode->i_op = &nfs_referral_inode_operations; + else + inode->i_op = &nfs_mountpoint_inode_operations; + inode->i_fop = NULL; + } } else if (S_ISLNK(inode->i_mode)) inode->i_op = &nfs_symlink_inode_operations; else @@ -911,15 +277,13 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) * report the blocks in 512byte units */ inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); - inode->i_blksize = inode->i_sb->s_blocksize; } else { inode->i_blocks = fattr->du.nfs2.blocks; - inode->i_blksize = fattr->du.nfs2.blocksize; } nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); nfsi->attrtimeo_timestamp = jiffies; memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); - nfsi->cache_access.cred = NULL; + nfsi->access_cache = RB_ROOT; unlock_new_inode(inode); } else @@ -1077,7 +441,7 @@ static struct nfs_open_context *alloc_nfs_open_context(struct vfsmount *mnt, str { struct nfs_open_context *ctx; - ctx = (struct nfs_open_context *)kmalloc(sizeof(*ctx), GFP_KERNEL); + ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); if (ctx != NULL) { atomic_set(&ctx->count, 1); ctx->dentry = dget(dentry); @@ -1208,6 +572,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) dfprintk(PAGECACHE, "NFS: revalidating (%s/%Ld)\n", inode->i_sb->s_id, (long long)NFS_FILEID(inode)); + nfs_inc_stats(inode, NFSIOS_INODEREVALIDATE); lock_kernel(); if (!inode || is_bad_inode(inode)) goto out_nowait; @@ -1221,7 +586,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) status = -ESTALE; /* Do we trust the cached ESTALE? */ if (NFS_ATTRTIMEO(inode) != 0) { - if (nfsi->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ATIME)) { + if (nfsi->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME)) { /* no */ } else goto out; @@ -1252,8 +617,6 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) } spin_unlock(&inode->i_lock); - nfs_revalidate_mapping(inode, inode->i_mapping); - if (nfsi->cache_validity & NFS_INO_INVALID_ACL) nfs_zap_acl_cache(inode); @@ -1287,8 +650,7 @@ int nfs_attribute_timeout(struct inode *inode) */ int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) { - nfs_inc_stats(inode, NFSIOS_INODEREVALIDATE); - if (!(NFS_I(inode)->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA)) + if (!(NFS_I(inode)->cache_validity & NFS_INO_INVALID_ATTR) && !nfs_attribute_timeout(inode)) return NFS_STALE(inode) ? -ESTALE : 0; return __nfs_revalidate_inode(server, inode); @@ -1299,9 +661,16 @@ int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) * @inode - pointer to host inode * @mapping - pointer to mapping */ -void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) +int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) { struct nfs_inode *nfsi = NFS_I(inode); + int ret = 0; + + if (NFS_STALE(inode)) + ret = -ESTALE; + if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) + || nfs_attribute_timeout(inode)) + ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode); if (nfsi->cache_validity & NFS_INO_INVALID_DATA) { nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); @@ -1322,6 +691,7 @@ void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) inode->i_sb->s_id, (long long)NFS_FILEID(inode)); } + return ret; } /** @@ -1345,13 +715,11 @@ void nfs_end_data_update(struct inode *inode) { struct nfs_inode *nfsi = NFS_I(inode); - if (!nfs_have_delegation(inode, FMODE_READ)) { - /* Directories and symlinks: invalidate page cache */ - if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) { - spin_lock(&inode->i_lock); - nfsi->cache_validity |= NFS_INO_INVALID_DATA; - spin_unlock(&inode->i_lock); - } + /* Directories: invalidate page cache */ + if (S_ISDIR(inode->i_mode)) { + spin_lock(&inode->i_lock); + nfsi->cache_validity |= NFS_INO_INVALID_DATA; + spin_unlock(&inode->i_lock); } nfsi->cache_change_attribute = jiffies; atomic_dec(&nfsi->data_updates); @@ -1361,12 +729,6 @@ static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr) { struct nfs_inode *nfsi = NFS_I(inode); - if ((fattr->valid & NFS_ATTR_PRE_CHANGE) != 0 - && nfsi->change_attr == fattr->pre_change_attr) { - nfsi->change_attr = fattr->change_attr; - nfsi->cache_change_attribute = jiffies; - } - /* If we have atomic WCC data, we may update some attributes */ if ((fattr->valid & NFS_ATTR_WCC) != 0) { if (timespec_equal(&inode->i_ctime, &fattr->pre_ctime)) { @@ -1400,9 +762,6 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat int data_unstable; - if ((fattr->valid & NFS_ATTR_FATTR) == 0) - return 0; - /* Has the inode gone and changed behind our back? */ if (nfsi->fileid != fattr->fileid || (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) { @@ -1415,20 +774,13 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat /* Do atomic weak cache consistency updates */ nfs_wcc_update_inode(inode, fattr); - if ((fattr->valid & NFS_ATTR_FATTR_V4) != 0) { - if (nfsi->change_attr == fattr->change_attr) - goto out; - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; - if (!data_unstable) - nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE; - } + if ((fattr->valid & NFS_ATTR_FATTR_V4) != 0 && + nfsi->change_attr != fattr->change_attr) + nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; /* Verify a few of the more important attributes */ - if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) { - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; - if (!data_unstable) - nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE; - } + if (!timespec_equal(&inode->i_mtime, &fattr->mtime)) + nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; cur_size = i_size_read(inode); new_isize = nfs_size_to_loff_t(fattr->size); @@ -1445,7 +797,6 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat if (inode->i_nlink != fattr->nlink) nfsi->cache_validity |= NFS_INO_INVALID_ATTR; -out: if (!timespec_equal(&inode->i_atime, &fattr->atime)) nfsi->cache_validity |= NFS_INO_INVALID_ATIME; @@ -1471,7 +822,6 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) if ((fattr->valid & NFS_ATTR_FATTR) == 0) return 0; spin_lock(&inode->i_lock); - nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE; if (time_after(fattr->time_start, nfsi->last_updated)) status = nfs_update_inode(inode, fattr); else @@ -1488,6 +838,12 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) * * After an operation that has changed the inode metadata, mark the * attribute cache as being invalid, then try to update it. + * + * NB: if the server didn't return any post op attributes, this + * function will force the retrieval of attributes before the next + * NFS request. Thus it should be used only for operations that + * are expected to change one or more attributes, to avoid + * unnecessary NFS requests and trips through nfs_update_inode(). */ int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr) { @@ -1496,7 +852,7 @@ int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr) spin_lock(&inode->i_lock); if (unlikely((fattr->valid & NFS_ATTR_FATTR) == 0)) { - nfsi->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS; + nfsi->cache_validity |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; goto out; } status = nfs_update_inode(inode, fattr); @@ -1519,6 +875,7 @@ out: */ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) { + struct nfs_server *server; struct nfs_inode *nfsi = NFS_I(inode); loff_t cur_isize, new_isize; unsigned int invalid = 0; @@ -1528,9 +885,6 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) __FUNCTION__, inode->i_sb->s_id, inode->i_ino, atomic_read(&inode->i_count), fattr->valid); - if ((fattr->valid & NFS_ATTR_FATTR) == 0) - return 0; - if (nfsi->fileid != fattr->fileid) goto out_fileid; @@ -1540,6 +894,12 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) goto out_changed; + server = NFS_SERVER(inode); + /* Update the fsid if and only if this is the root directory */ + if (inode == inode->i_sb->s_root->d_inode + && !nfs_fsid_equal(&server->fsid, &fattr->fsid)) + server->fsid = fattr->fsid; + /* * Update the read time so we don't revalidate too often. */ @@ -1549,7 +909,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) /* Are we racing with known updates of the metadata on the server? */ data_stable = nfs_verify_change_attribute(inode, fattr->time_start); if (data_stable) - nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME); + nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_ATIME); /* Do atomic weak cache consistency updates */ nfs_wcc_update_inode(inode, fattr); @@ -1607,21 +967,17 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) * report the blocks in 512byte units */ inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); - inode->i_blksize = inode->i_sb->s_blocksize; } else { inode->i_blocks = fattr->du.nfs2.blocks; - inode->i_blksize = fattr->du.nfs2.blocksize; } - if ((fattr->valid & NFS_ATTR_FATTR_V4)) { - if (nfsi->change_attr != fattr->change_attr) { - dprintk("NFS: change_attr change on server for file %s/%ld\n", - inode->i_sb->s_id, inode->i_ino); - nfsi->change_attr = fattr->change_attr; - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; - nfsi->cache_change_attribute = jiffies; - } else - invalid &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA); + if ((fattr->valid & NFS_ATTR_FATTR_V4) != 0 && + nfsi->change_attr != fattr->change_attr) { + dprintk("NFS: change_attr change on server for file %s/%ld\n", + inode->i_sb->s_id, inode->i_ino); + nfsi->change_attr = fattr->change_attr; + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + nfsi->cache_change_attribute = jiffies; } /* Update attrtimeo value if we're out of the unstable period */ @@ -1664,207 +1020,20 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) out_fileid: printk(KERN_ERR "NFS: server %s error: fileid changed\n" "fsid %s: expected fileid 0x%Lx, got 0x%Lx\n", - NFS_SERVER(inode)->hostname, inode->i_sb->s_id, + NFS_SERVER(inode)->nfs_client->cl_hostname, inode->i_sb->s_id, (long long)nfsi->fileid, (long long)fattr->fileid); goto out_err; } -/* - * File system information - */ - -static int nfs_set_super(struct super_block *s, void *data) -{ - s->s_fs_info = data; - return set_anon_super(s, data); -} - -static int nfs_compare_super(struct super_block *sb, void *data) -{ - struct nfs_server *server = data; - struct nfs_server *old = NFS_SB(sb); - - if (old->addr.sin_addr.s_addr != server->addr.sin_addr.s_addr) - return 0; - if (old->addr.sin_port != server->addr.sin_port) - return 0; - return !nfs_compare_fh(&old->fh, &server->fh); -} - -static int nfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt) -{ - int error; - struct nfs_server *server = NULL; - struct super_block *s; - struct nfs_fh *root; - struct nfs_mount_data *data = raw_data; - - error = -EINVAL; - if (data == NULL) { - dprintk("%s: missing data argument\n", __FUNCTION__); - goto out_err_noserver; - } - if (data->version <= 0 || data->version > NFS_MOUNT_VERSION) { - dprintk("%s: bad mount version\n", __FUNCTION__); - goto out_err_noserver; - } - switch (data->version) { - case 1: - data->namlen = 0; - case 2: - data->bsize = 0; - case 3: - if (data->flags & NFS_MOUNT_VER3) { - dprintk("%s: mount structure version %d does not support NFSv3\n", - __FUNCTION__, - data->version); - goto out_err_noserver; - } - data->root.size = NFS2_FHSIZE; - memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE); - case 4: - if (data->flags & NFS_MOUNT_SECFLAVOUR) { - dprintk("%s: mount structure version %d does not support strong security\n", - __FUNCTION__, - data->version); - goto out_err_noserver; - } - case 5: - memset(data->context, 0, sizeof(data->context)); - } -#ifndef CONFIG_NFS_V3 - /* If NFSv3 is not compiled in, return -EPROTONOSUPPORT */ - error = -EPROTONOSUPPORT; - if (data->flags & NFS_MOUNT_VER3) { - dprintk("%s: NFSv3 not compiled into kernel\n", __FUNCTION__); - goto out_err_noserver; - } -#endif /* CONFIG_NFS_V3 */ - - error = -ENOMEM; - server = kzalloc(sizeof(struct nfs_server), GFP_KERNEL); - if (!server) - goto out_err_noserver; - /* Zero out the NFS state stuff */ - init_nfsv4_state(server); - server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL); - - root = &server->fh; - if (data->flags & NFS_MOUNT_VER3) - root->size = data->root.size; - else - root->size = NFS2_FHSIZE; - error = -EINVAL; - if (root->size > sizeof(root->data)) { - dprintk("%s: invalid root filehandle\n", __FUNCTION__); - goto out_err; - } - memcpy(root->data, data->root.data, root->size); - - /* We now require that the mount process passes the remote address */ - memcpy(&server->addr, &data->addr, sizeof(server->addr)); - if (server->addr.sin_addr.s_addr == INADDR_ANY) { - dprintk("%s: mount program didn't pass remote address!\n", - __FUNCTION__); - goto out_err; - } - - /* Fire up rpciod if not yet running */ - error = rpciod_up(); - if (error < 0) { - dprintk("%s: couldn't start rpciod! Error = %d\n", - __FUNCTION__, error); - goto out_err; - } - - s = sget(fs_type, nfs_compare_super, nfs_set_super, server); - if (IS_ERR(s)) { - error = PTR_ERR(s); - goto out_err_rpciod; - } - - if (s->s_root) - goto out_rpciod_down; - - s->s_flags = flags; - - error = nfs_fill_super(s, data, flags & MS_SILENT ? 1 : 0); - if (error) { - up_write(&s->s_umount); - deactivate_super(s); - return error; - } - s->s_flags |= MS_ACTIVE; - return simple_set_mnt(mnt, s); - -out_rpciod_down: - rpciod_down(); - kfree(server); - return simple_set_mnt(mnt, s); - -out_err_rpciod: - rpciod_down(); -out_err: - kfree(server); -out_err_noserver: - return error; -} - -static void nfs_kill_super(struct super_block *s) -{ - struct nfs_server *server = NFS_SB(s); - - kill_anon_super(s); - - if (!IS_ERR(server->client)) - rpc_shutdown_client(server->client); - if (!IS_ERR(server->client_sys)) - rpc_shutdown_client(server->client_sys); - if (!IS_ERR(server->client_acl)) - rpc_shutdown_client(server->client_acl); - - if (!(server->flags & NFS_MOUNT_NONLM)) - lockd_down(); /* release rpc.lockd */ - - rpciod_down(); /* release rpciod */ - - nfs_free_iostats(server->io_stats); - kfree(server->hostname); - kfree(server); -} - -static struct file_system_type nfs_fs_type = { - .owner = THIS_MODULE, - .name = "nfs", - .get_sb = nfs_get_sb, - .kill_sb = nfs_kill_super, - .fs_flags = FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, -}; #ifdef CONFIG_NFS_V4 -static void nfs4_clear_inode(struct inode *); - - -static struct super_operations nfs4_sops = { - .alloc_inode = nfs_alloc_inode, - .destroy_inode = nfs_destroy_inode, - .write_inode = nfs_write_inode, - .delete_inode = nfs_delete_inode, - .statfs = nfs_statfs, - .clear_inode = nfs4_clear_inode, - .umount_begin = nfs_umount_begin, - .show_options = nfs_show_options, - .show_stats = nfs_show_stats, -}; - /* * Clean out any remaining NFSv4 state that might be left over due * to open() calls that passed nfs_atomic_lookup, but failed to call * nfs_open(). */ -static void nfs4_clear_inode(struct inode *inode) +void nfs4_clear_inode(struct inode *inode) { struct nfs_inode *nfsi = NFS_I(inode); @@ -1888,365 +1057,9 @@ static void nfs4_clear_inode(struct inode *inode) nfs4_close_state(state, state->state); } } - - -static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data, int silent) -{ - struct nfs_server *server; - struct nfs4_client *clp = NULL; - struct rpc_xprt *xprt = NULL; - struct rpc_clnt *clnt = NULL; - struct rpc_timeout timeparms; - rpc_authflavor_t authflavour; - int err = -EIO; - - sb->s_blocksize_bits = 0; - sb->s_blocksize = 0; - server = NFS_SB(sb); - if (data->rsize != 0) - server->rsize = nfs_block_size(data->rsize, NULL); - if (data->wsize != 0) - server->wsize = nfs_block_size(data->wsize, NULL); - server->flags = data->flags & NFS_MOUNT_FLAGMASK; - server->caps = NFS_CAP_ATOMIC_OPEN; - - server->acregmin = data->acregmin*HZ; - server->acregmax = data->acregmax*HZ; - server->acdirmin = data->acdirmin*HZ; - server->acdirmax = data->acdirmax*HZ; - - server->rpc_ops = &nfs_v4_clientops; - - nfs_init_timeout_values(&timeparms, data->proto, data->timeo, data->retrans); - - server->retrans_timeo = timeparms.to_initval; - server->retrans_count = timeparms.to_retries; - - clp = nfs4_get_client(&server->addr.sin_addr); - if (!clp) { - dprintk("%s: failed to create NFS4 client.\n", __FUNCTION__); - return -EIO; - } - - /* Now create transport and client */ - authflavour = RPC_AUTH_UNIX; - if (data->auth_flavourlen != 0) { - if (data->auth_flavourlen != 1) { - dprintk("%s: Invalid number of RPC auth flavours %d.\n", - __FUNCTION__, data->auth_flavourlen); - err = -EINVAL; - goto out_fail; - } - if (copy_from_user(&authflavour, data->auth_flavours, sizeof(authflavour))) { - err = -EFAULT; - goto out_fail; - } - } - - down_write(&clp->cl_sem); - if (IS_ERR(clp->cl_rpcclient)) { - xprt = xprt_create_proto(data->proto, &server->addr, &timeparms); - if (IS_ERR(xprt)) { - up_write(&clp->cl_sem); - err = PTR_ERR(xprt); - dprintk("%s: cannot create RPC transport. Error = %d\n", - __FUNCTION__, err); - goto out_fail; - } - clnt = rpc_create_client(xprt, server->hostname, &nfs_program, - server->rpc_ops->version, authflavour); - if (IS_ERR(clnt)) { - up_write(&clp->cl_sem); - err = PTR_ERR(clnt); - dprintk("%s: cannot create RPC client. Error = %d\n", - __FUNCTION__, err); - goto out_fail; - } - clnt->cl_intr = 1; - clnt->cl_softrtry = 1; - clp->cl_rpcclient = clnt; - memcpy(clp->cl_ipaddr, server->ip_addr, sizeof(clp->cl_ipaddr)); - nfs_idmap_new(clp); - } - list_add_tail(&server->nfs4_siblings, &clp->cl_superblocks); - clnt = rpc_clone_client(clp->cl_rpcclient); - if (!IS_ERR(clnt)) - server->nfs4_state = clp; - up_write(&clp->cl_sem); - clp = NULL; - - if (IS_ERR(clnt)) { - err = PTR_ERR(clnt); - dprintk("%s: cannot create RPC client. Error = %d\n", - __FUNCTION__, err); - return err; - } - - server->client = clnt; - - if (server->nfs4_state->cl_idmap == NULL) { - dprintk("%s: failed to create idmapper.\n", __FUNCTION__); - return -ENOMEM; - } - - if (clnt->cl_auth->au_flavor != authflavour) { - struct rpc_auth *auth; - - auth = rpcauth_create(authflavour, clnt); - if (IS_ERR(auth)) { - dprintk("%s: couldn't create credcache!\n", __FUNCTION__); - return PTR_ERR(auth); - } - } - - sb->s_time_gran = 1; - - sb->s_op = &nfs4_sops; - err = nfs_sb_init(sb, authflavour); - if (err == 0) - return 0; -out_fail: - if (clp) - nfs4_put_client(clp); - return err; -} - -static int nfs4_compare_super(struct super_block *sb, void *data) -{ - struct nfs_server *server = data; - struct nfs_server *old = NFS_SB(sb); - - if (strcmp(server->hostname, old->hostname) != 0) - return 0; - if (strcmp(server->mnt_path, old->mnt_path) != 0) - return 0; - return 1; -} - -static void * -nfs_copy_user_string(char *dst, struct nfs_string *src, int maxlen) -{ - void *p = NULL; - - if (!src->len) - return ERR_PTR(-EINVAL); - if (src->len < maxlen) - maxlen = src->len; - if (dst == NULL) { - p = dst = kmalloc(maxlen + 1, GFP_KERNEL); - if (p == NULL) - return ERR_PTR(-ENOMEM); - } - if (copy_from_user(dst, src->data, maxlen)) { - kfree(p); - return ERR_PTR(-EFAULT); - } - dst[maxlen] = '\0'; - return dst; -} - -static int nfs4_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt) -{ - int error; - struct nfs_server *server; - struct super_block *s; - struct nfs4_mount_data *data = raw_data; - void *p; - - if (data == NULL) { - dprintk("%s: missing data argument\n", __FUNCTION__); - return -EINVAL; - } - if (data->version <= 0 || data->version > NFS4_MOUNT_VERSION) { - dprintk("%s: bad mount version\n", __FUNCTION__); - return -EINVAL; - } - - server = kzalloc(sizeof(struct nfs_server), GFP_KERNEL); - if (!server) - return -ENOMEM; - /* Zero out the NFS state stuff */ - init_nfsv4_state(server); - server->client = server->client_sys = server->client_acl = ERR_PTR(-EINVAL); - - p = nfs_copy_user_string(NULL, &data->hostname, 256); - if (IS_ERR(p)) - goto out_err; - server->hostname = p; - - p = nfs_copy_user_string(NULL, &data->mnt_path, 1024); - if (IS_ERR(p)) - goto out_err; - server->mnt_path = p; - - p = nfs_copy_user_string(server->ip_addr, &data->client_addr, - sizeof(server->ip_addr) - 1); - if (IS_ERR(p)) - goto out_err; - - /* We now require that the mount process passes the remote address */ - if (data->host_addrlen != sizeof(server->addr)) { - error = -EINVAL; - goto out_free; - } - if (copy_from_user(&server->addr, data->host_addr, sizeof(server->addr))) { - error = -EFAULT; - goto out_free; - } - if (server->addr.sin_family != AF_INET || - server->addr.sin_addr.s_addr == INADDR_ANY) { - dprintk("%s: mount program didn't pass remote IP address!\n", - __FUNCTION__); - error = -EINVAL; - goto out_free; - } - - /* Fire up rpciod if not yet running */ - error = rpciod_up(); - if (error < 0) { - dprintk("%s: couldn't start rpciod! Error = %d\n", - __FUNCTION__, error); - goto out_free; - } - - s = sget(fs_type, nfs4_compare_super, nfs_set_super, server); - if (IS_ERR(s)) { - error = PTR_ERR(s); - goto out_free; - } - - if (s->s_root) { - kfree(server->mnt_path); - kfree(server->hostname); - kfree(server); - return simple_set_mnt(mnt, s); - } - - s->s_flags = flags; - - error = nfs4_fill_super(s, data, flags & MS_SILENT ? 1 : 0); - if (error) { - up_write(&s->s_umount); - deactivate_super(s); - return error; - } - s->s_flags |= MS_ACTIVE; - return simple_set_mnt(mnt, s); -out_err: - error = PTR_ERR(p); -out_free: - kfree(server->mnt_path); - kfree(server->hostname); - kfree(server); - return error; -} - -static void nfs4_kill_super(struct super_block *sb) -{ - struct nfs_server *server = NFS_SB(sb); - - nfs_return_all_delegations(sb); - kill_anon_super(sb); - - nfs4_renewd_prepare_shutdown(server); - - if (server->client != NULL && !IS_ERR(server->client)) - rpc_shutdown_client(server->client); - - destroy_nfsv4_state(server); - - rpciod_down(); - - nfs_free_iostats(server->io_stats); - kfree(server->hostname); - kfree(server); -} - -static struct file_system_type nfs4_fs_type = { - .owner = THIS_MODULE, - .name = "nfs4", - .get_sb = nfs4_get_sb, - .kill_sb = nfs4_kill_super, - .fs_flags = FS_ODD_RENAME|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, -}; - -static const int nfs_set_port_min = 0; -static const int nfs_set_port_max = 65535; -static int param_set_port(const char *val, struct kernel_param *kp) -{ - char *endp; - int num = simple_strtol(val, &endp, 0); - if (endp == val || *endp || num < nfs_set_port_min || num > nfs_set_port_max) - return -EINVAL; - *((int *)kp->arg) = num; - return 0; -} - -module_param_call(callback_tcpport, param_set_port, param_get_int, - &nfs_callback_set_tcpport, 0644); - -static int param_set_idmap_timeout(const char *val, struct kernel_param *kp) -{ - char *endp; - int num = simple_strtol(val, &endp, 0); - int jif = num * HZ; - if (endp == val || *endp || num < 0 || jif < num) - return -EINVAL; - *((int *)kp->arg) = jif; - return 0; -} - -module_param_call(idmap_cache_timeout, param_set_idmap_timeout, param_get_int, - &nfs_idmap_cache_timeout, 0644); - -#define nfs4_init_once(nfsi) \ - do { \ - INIT_LIST_HEAD(&(nfsi)->open_states); \ - nfsi->delegation = NULL; \ - nfsi->delegation_state = 0; \ - init_rwsem(&nfsi->rwsem); \ - } while(0) - -static inline int register_nfs4fs(void) -{ - int ret; - - ret = nfs_register_sysctl(); - if (ret != 0) - return ret; - ret = register_filesystem(&nfs4_fs_type); - if (ret != 0) - nfs_unregister_sysctl(); - return ret; -} - -static inline void unregister_nfs4fs(void) -{ - unregister_filesystem(&nfs4_fs_type); - nfs_unregister_sysctl(); -} -#else -#define nfs4_init_once(nfsi) \ - do { } while (0) -#define register_nfs4fs() (0) -#define unregister_nfs4fs() -#endif - -extern int nfs_init_nfspagecache(void); -extern void nfs_destroy_nfspagecache(void); -extern int nfs_init_readpagecache(void); -extern void nfs_destroy_readpagecache(void); -extern int nfs_init_writepagecache(void); -extern void nfs_destroy_writepagecache(void); -#ifdef CONFIG_NFS_DIRECTIO -extern int nfs_init_directcache(void); -extern void nfs_destroy_directcache(void); #endif -static kmem_cache_t * nfs_inode_cachep; - -static struct inode *nfs_alloc_inode(struct super_block *sb) +struct inode *nfs_alloc_inode(struct super_block *sb) { struct nfs_inode *nfsi; nfsi = (struct nfs_inode *)kmem_cache_alloc(nfs_inode_cachep, SLAB_KERNEL); @@ -2265,11 +1078,21 @@ static struct inode *nfs_alloc_inode(struct super_block *sb) return &nfsi->vfs_inode; } -static void nfs_destroy_inode(struct inode *inode) +void nfs_destroy_inode(struct inode *inode) { kmem_cache_free(nfs_inode_cachep, NFS_I(inode)); } +static inline void nfs4_init_once(struct nfs_inode *nfsi) +{ +#ifdef CONFIG_NFS_V4 + INIT_LIST_HEAD(&nfsi->open_states); + nfsi->delegation = NULL; + nfsi->delegation_state = 0; + init_rwsem(&nfsi->rwsem); +#endif +} + static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) { struct nfs_inode *nfsi = (struct nfs_inode *) foo; @@ -2281,6 +1104,8 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) INIT_LIST_HEAD(&nfsi->dirty); INIT_LIST_HEAD(&nfsi->commit); INIT_LIST_HEAD(&nfsi->open_files); + INIT_LIST_HEAD(&nfsi->access_cache_entry_lru); + INIT_LIST_HEAD(&nfsi->access_cache_inode_lru); INIT_RADIX_TREE(&nfsi->nfs_page_tree, GFP_ATOMIC); atomic_set(&nfsi->data_updates, 0); nfsi->ndirty = 0; @@ -2290,7 +1115,7 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) } } -static int nfs_init_inodecache(void) +static int __init nfs_init_inodecache(void) { nfs_inode_cachep = kmem_cache_create("nfs_inode_cache", sizeof(struct nfs_inode), @@ -2305,8 +1130,7 @@ static int nfs_init_inodecache(void) static void nfs_destroy_inodecache(void) { - if (kmem_cache_destroy(nfs_inode_cachep)) - printk(KERN_INFO "nfs_inode_cache: not all structures were freed\n"); + kmem_cache_destroy(nfs_inode_cachep); } /* @@ -2316,6 +1140,10 @@ static int __init init_nfs_fs(void) { int err; + err = nfs_fs_proc_init(); + if (err) + goto out5; + err = nfs_init_nfspagecache(); if (err) goto out4; @@ -2332,29 +1160,22 @@ static int __init init_nfs_fs(void) if (err) goto out1; -#ifdef CONFIG_NFS_DIRECTIO err = nfs_init_directcache(); if (err) goto out0; -#endif #ifdef CONFIG_PROC_FS rpc_proc_register(&nfs_rpcstat); #endif - err = register_filesystem(&nfs_fs_type); - if (err) - goto out; - if ((err = register_nfs4fs()) != 0) + if ((err = register_nfs_fs()) != 0) goto out; return 0; out: #ifdef CONFIG_PROC_FS rpc_proc_unregister("nfs"); #endif -#ifdef CONFIG_NFS_DIRECTIO nfs_destroy_directcache(); out0: -#endif nfs_destroy_writepagecache(); out1: nfs_destroy_readpagecache(); @@ -2363,14 +1184,14 @@ out2: out3: nfs_destroy_nfspagecache(); out4: + nfs_fs_proc_exit(); +out5: return err; } static void __exit exit_nfs_fs(void) { -#ifdef CONFIG_NFS_DIRECTIO nfs_destroy_directcache(); -#endif nfs_destroy_writepagecache(); nfs_destroy_readpagecache(); nfs_destroy_inodecache(); @@ -2378,8 +1199,8 @@ static void __exit exit_nfs_fs(void) #ifdef CONFIG_PROC_FS rpc_proc_unregister("nfs"); #endif - unregister_filesystem(&nfs_fs_type); - unregister_nfs4fs(); + unregister_nfs_fs(); + nfs_fs_proc_exit(); } /* Not quite true; I just maintain it */ diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h new file mode 100644 index 00000000000..bea0b016bd7 --- /dev/null +++ b/fs/nfs/internal.h @@ -0,0 +1,219 @@ +/* + * NFS internal definitions + */ + +#include <linux/mount.h> + +struct nfs_string; +struct nfs_mount_data; +struct nfs4_mount_data; + +/* Maximum number of readahead requests + * FIXME: this should really be a sysctl so that users may tune it to suit + * their needs. People that do NFS over a slow network, might for + * instance want to reduce it to something closer to 1 for improved + * interactive response. + */ +#define NFS_MAX_READAHEAD (RPC_DEF_SLOT_TABLE - 1) + +struct nfs_clone_mount { + const struct super_block *sb; + const struct dentry *dentry; + struct nfs_fh *fh; + struct nfs_fattr *fattr; + char *hostname; + char *mnt_path; + struct sockaddr_in *addr; + rpc_authflavor_t authflavor; +}; + +/* client.c */ +extern struct rpc_program nfs_program; + +extern void nfs_put_client(struct nfs_client *); +extern struct nfs_client *nfs_find_client(const struct sockaddr_in *, int); +extern struct nfs_server *nfs_create_server(const struct nfs_mount_data *, + struct nfs_fh *); +extern struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *, + const char *, + const struct sockaddr_in *, + const char *, + const char *, + rpc_authflavor_t, + struct nfs_fh *); +extern struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *, + struct nfs_fh *); +extern void nfs_free_server(struct nfs_server *server); +extern struct nfs_server *nfs_clone_server(struct nfs_server *, + struct nfs_fh *, + struct nfs_fattr *); +#ifdef CONFIG_PROC_FS +extern int __init nfs_fs_proc_init(void); +extern void nfs_fs_proc_exit(void); +#else +static inline int nfs_fs_proc_init(void) +{ + return 0; +} +static inline void nfs_fs_proc_exit(void) +{ +} +#endif + +/* nfs4namespace.c */ +#ifdef CONFIG_NFS_V4 +extern struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry); +#else +static inline +struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry) +{ + return ERR_PTR(-ENOENT); +} +#endif + +/* callback_xdr.c */ +extern struct svc_version nfs4_callback_version1; + +/* pagelist.c */ +extern int __init nfs_init_nfspagecache(void); +extern void nfs_destroy_nfspagecache(void); +extern int __init nfs_init_readpagecache(void); +extern void nfs_destroy_readpagecache(void); +extern int __init nfs_init_writepagecache(void); +extern void nfs_destroy_writepagecache(void); + +#ifdef CONFIG_NFS_DIRECTIO +extern int __init nfs_init_directcache(void); +extern void nfs_destroy_directcache(void); +#else +#define nfs_init_directcache() (0) +#define nfs_destroy_directcache() do {} while(0) +#endif + +/* nfs2xdr.c */ +extern int nfs_stat_to_errno(int); +extern struct rpc_procinfo nfs_procedures[]; +extern u32 * nfs_decode_dirent(u32 *, struct nfs_entry *, int); + +/* nfs3xdr.c */ +extern struct rpc_procinfo nfs3_procedures[]; +extern u32 *nfs3_decode_dirent(u32 *, struct nfs_entry *, int); + +/* nfs4xdr.c */ +#ifdef CONFIG_NFS_V4 +extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus); +#endif + +/* nfs4proc.c */ +#ifdef CONFIG_NFS_V4 +extern struct rpc_procinfo nfs4_procedures[]; + +extern int nfs4_proc_fs_locations(struct inode *dir, struct dentry *dentry, + struct nfs4_fs_locations *fs_locations, + struct page *page); +#endif + +/* dir.c */ +extern int nfs_access_cache_shrinker(int nr_to_scan, gfp_t gfp_mask); + +/* inode.c */ +extern struct inode *nfs_alloc_inode(struct super_block *sb); +extern void nfs_destroy_inode(struct inode *); +extern int nfs_write_inode(struct inode *,int); +extern void nfs_clear_inode(struct inode *); +#ifdef CONFIG_NFS_V4 +extern void nfs4_clear_inode(struct inode *); +#endif + +/* super.c */ +extern struct file_system_type nfs_xdev_fs_type; +#ifdef CONFIG_NFS_V4 +extern struct file_system_type nfs4_xdev_fs_type; +extern struct file_system_type nfs4_referral_fs_type; +#endif + +extern struct rpc_stat nfs_rpcstat; + +extern int __init register_nfs_fs(void); +extern void __exit unregister_nfs_fs(void); + +/* namespace.c */ +extern char *nfs_path(const char *base, + const struct dentry *droot, + const struct dentry *dentry, + char *buffer, ssize_t buflen); + +/* getroot.c */ +extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *); +#ifdef CONFIG_NFS_V4 +extern struct dentry *nfs4_get_root(struct super_block *, struct nfs_fh *); + +extern int nfs4_path_walk(struct nfs_server *server, + struct nfs_fh *mntfh, + const char *path); +#endif + +/* + * Determine the device name as a string + */ +static inline char *nfs_devname(const struct vfsmount *mnt_parent, + const struct dentry *dentry, + char *buffer, ssize_t buflen) +{ + return nfs_path(mnt_parent->mnt_devname, mnt_parent->mnt_root, + dentry, buffer, buflen); +} + +/* + * Determine the actual block size (and log2 thereof) + */ +static inline +unsigned long nfs_block_bits(unsigned long bsize, unsigned char *nrbitsp) +{ + /* make sure blocksize is a power of two */ + if ((bsize & (bsize - 1)) || nrbitsp) { + unsigned char nrbits; + + for (nrbits = 31; nrbits && !(bsize & (1 << nrbits)); nrbits--) + ; + bsize = 1 << nrbits; + if (nrbitsp) + *nrbitsp = nrbits; + } + + return bsize; +} + +/* + * Calculate the number of 512byte blocks used. + */ +static inline unsigned long nfs_calc_block_size(u64 tsize) +{ + loff_t used = (tsize + 511) >> 9; + return (used > ULONG_MAX) ? ULONG_MAX : used; +} + +/* + * Compute and set NFS server blocksize + */ +static inline +unsigned long nfs_block_size(unsigned long bsize, unsigned char *nrbitsp) +{ + if (bsize < NFS_MIN_FILE_IO_SIZE) + bsize = NFS_DEF_FILE_IO_SIZE; + else if (bsize >= NFS_MAX_FILE_IO_SIZE) + bsize = NFS_MAX_FILE_IO_SIZE; + + return nfs_block_bits(bsize, nrbitsp); +} + +/* + * Determine the maximum file size for a superblock + */ +static inline +void nfs_super_set_maxbytes(struct super_block *sb, __u64 maxfilesize) +{ + sb->s_maxbytes = (loff_t)maxfilesize; + if (sb->s_maxbytes > MAX_LFS_FILESIZE || sb->s_maxbytes <= 0) + sb->s_maxbytes = MAX_LFS_FILESIZE; +} diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c index 445abb4d421..d507b021207 100644 --- a/fs/nfs/mount_clnt.c +++ b/fs/nfs/mount_clnt.c @@ -14,7 +14,6 @@ #include <linux/net.h> #include <linux/in.h> #include <linux/sunrpc/clnt.h> -#include <linux/sunrpc/xprt.h> #include <linux/sunrpc/sched.h> #include <linux/nfs_fs.h> @@ -77,22 +76,19 @@ static struct rpc_clnt * mnt_create(char *hostname, struct sockaddr_in *srvaddr, int version, int protocol) { - struct rpc_xprt *xprt; - struct rpc_clnt *clnt; - - xprt = xprt_create_proto(protocol, srvaddr, NULL); - if (IS_ERR(xprt)) - return (struct rpc_clnt *)xprt; - - clnt = rpc_create_client(xprt, hostname, - &mnt_program, version, - RPC_AUTH_UNIX); - if (!IS_ERR(clnt)) { - clnt->cl_softrtry = 1; - clnt->cl_oneshot = 1; - clnt->cl_intr = 1; - } - return clnt; + struct rpc_create_args args = { + .protocol = protocol, + .address = (struct sockaddr *)srvaddr, + .addrsize = sizeof(*srvaddr), + .servername = hostname, + .program = &mnt_program, + .version = version, + .authflavor = RPC_AUTH_UNIX, + .flags = (RPC_CLNT_CREATE_ONESHOT | + RPC_CLNT_CREATE_INTR), + }; + + return rpc_create(&args); } /* diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c new file mode 100644 index 00000000000..60408646176 --- /dev/null +++ b/fs/nfs/namespace.c @@ -0,0 +1,251 @@ +/* + * linux/fs/nfs/namespace.c + * + * Copyright (C) 2005 Trond Myklebust <Trond.Myklebust@netapp.com> + * - Modified by David Howells <dhowells@redhat.com> + * + * NFS namespace + */ + +#include <linux/config.h> + +#include <linux/dcache.h> +#include <linux/mount.h> +#include <linux/namei.h> +#include <linux/nfs_fs.h> +#include <linux/string.h> +#include <linux/sunrpc/clnt.h> +#include <linux/vfs.h> +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +static void nfs_expire_automounts(void *list); + +LIST_HEAD(nfs_automount_list); +static DECLARE_WORK(nfs_automount_task, nfs_expire_automounts, &nfs_automount_list); +int nfs_mountpoint_expiry_timeout = 500 * HZ; + +static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent, + const struct dentry *dentry, + struct nfs_fh *fh, + struct nfs_fattr *fattr); + +/* + * nfs_path - reconstruct the path given an arbitrary dentry + * @base - arbitrary string to prepend to the path + * @droot - pointer to root dentry for mountpoint + * @dentry - pointer to dentry + * @buffer - result buffer + * @buflen - length of buffer + * + * Helper function for constructing the path from the + * root dentry to an arbitrary hashed dentry. + * + * This is mainly for use in figuring out the path on the + * server side when automounting on top of an existing partition. + */ +char *nfs_path(const char *base, + const struct dentry *droot, + const struct dentry *dentry, + char *buffer, ssize_t buflen) +{ + char *end = buffer+buflen; + int namelen; + + *--end = '\0'; + buflen--; + spin_lock(&dcache_lock); + while (!IS_ROOT(dentry) && dentry != droot) { + namelen = dentry->d_name.len; + buflen -= namelen + 1; + if (buflen < 0) + goto Elong_unlock; + end -= namelen; + memcpy(end, dentry->d_name.name, namelen); + *--end = '/'; + dentry = dentry->d_parent; + } + spin_unlock(&dcache_lock); + namelen = strlen(base); + /* Strip off excess slashes in base string */ + while (namelen > 0 && base[namelen - 1] == '/') + namelen--; + buflen -= namelen; + if (buflen < 0) + goto Elong; + end -= namelen; + memcpy(end, base, namelen); + return end; +Elong_unlock: + spin_unlock(&dcache_lock); +Elong: + return ERR_PTR(-ENAMETOOLONG); +} + +/* + * nfs_follow_mountpoint - handle crossing a mountpoint on the server + * @dentry - dentry of mountpoint + * @nd - nameidata info + * + * When we encounter a mountpoint on the server, we want to set up + * a mountpoint on the client too, to prevent inode numbers from + * colliding, and to allow "df" to work properly. + * On NFSv4, we also want to allow for the fact that different + * filesystems may be migrated to different servers in a failover + * situation, and that different filesystems may want to use + * different security flavours. + */ +static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd) +{ + struct vfsmount *mnt; + struct nfs_server *server = NFS_SERVER(dentry->d_inode); + struct dentry *parent; + struct nfs_fh fh; + struct nfs_fattr fattr; + int err; + + dprintk("--> nfs_follow_mountpoint()\n"); + + BUG_ON(IS_ROOT(dentry)); + dprintk("%s: enter\n", __FUNCTION__); + dput(nd->dentry); + nd->dentry = dget(dentry); + + /* Look it up again */ + parent = dget_parent(nd->dentry); + err = server->nfs_client->rpc_ops->lookup(parent->d_inode, + &nd->dentry->d_name, + &fh, &fattr); + dput(parent); + if (err != 0) + goto out_err; + + if (fattr.valid & NFS_ATTR_FATTR_V4_REFERRAL) + mnt = nfs_do_refmount(nd->mnt, nd->dentry); + else + mnt = nfs_do_submount(nd->mnt, nd->dentry, &fh, &fattr); + err = PTR_ERR(mnt); + if (IS_ERR(mnt)) + goto out_err; + + mntget(mnt); + err = do_add_mount(mnt, nd, nd->mnt->mnt_flags|MNT_SHRINKABLE, &nfs_automount_list); + if (err < 0) { + mntput(mnt); + if (err == -EBUSY) + goto out_follow; + goto out_err; + } + mntput(nd->mnt); + dput(nd->dentry); + nd->mnt = mnt; + nd->dentry = dget(mnt->mnt_root); + schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout); +out: + dprintk("%s: done, returned %d\n", __FUNCTION__, err); + + dprintk("<-- nfs_follow_mountpoint() = %d\n", err); + return ERR_PTR(err); +out_err: + path_release(nd); + goto out; +out_follow: + while(d_mountpoint(nd->dentry) && follow_down(&nd->mnt, &nd->dentry)) + ; + err = 0; + goto out; +} + +struct inode_operations nfs_mountpoint_inode_operations = { + .follow_link = nfs_follow_mountpoint, + .getattr = nfs_getattr, +}; + +struct inode_operations nfs_referral_inode_operations = { + .follow_link = nfs_follow_mountpoint, +}; + +static void nfs_expire_automounts(void *data) +{ + struct list_head *list = (struct list_head *)data; + + mark_mounts_for_expiry(list); + if (!list_empty(list)) + schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout); +} + +void nfs_release_automount_timer(void) +{ + if (list_empty(&nfs_automount_list)) { + cancel_delayed_work(&nfs_automount_task); + flush_scheduled_work(); + } +} + +/* + * Clone a mountpoint of the appropriate type + */ +static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server, + const char *devname, + struct nfs_clone_mount *mountdata) +{ +#ifdef CONFIG_NFS_V4 + struct vfsmount *mnt = NULL; + switch (server->nfs_client->cl_nfsversion) { + case 2: + case 3: + mnt = vfs_kern_mount(&nfs_xdev_fs_type, 0, devname, mountdata); + break; + case 4: + mnt = vfs_kern_mount(&nfs4_xdev_fs_type, 0, devname, mountdata); + } + return mnt; +#else + return vfs_kern_mount(&nfs_xdev_fs_type, 0, devname, mountdata); +#endif +} + +/** + * nfs_do_submount - set up mountpoint when crossing a filesystem boundary + * @mnt_parent - mountpoint of parent directory + * @dentry - parent directory + * @fh - filehandle for new root dentry + * @fattr - attributes for new root inode + * + */ +static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent, + const struct dentry *dentry, + struct nfs_fh *fh, + struct nfs_fattr *fattr) +{ + struct nfs_clone_mount mountdata = { + .sb = mnt_parent->mnt_sb, + .dentry = dentry, + .fh = fh, + .fattr = fattr, + }; + struct vfsmount *mnt = ERR_PTR(-ENOMEM); + char *page = (char *) __get_free_page(GFP_USER); + char *devname; + + dprintk("--> nfs_do_submount()\n"); + + dprintk("%s: submounting on %s/%s\n", __FUNCTION__, + dentry->d_parent->d_name.name, + dentry->d_name.name); + if (page == NULL) + goto out; + devname = nfs_devname(mnt_parent, dentry, page, PAGE_SIZE); + mnt = (struct vfsmount *)devname; + if (IS_ERR(devname)) + goto free_page; + mnt = nfs_do_clone_mount(NFS_SB(mnt_parent->mnt_sb), devname, &mountdata); +free_page: + free_page((unsigned long)page); +out: + dprintk("%s: done\n", __FUNCTION__); + + dprintk("<-- nfs_do_submount() = %p\n", mnt); + return mnt; +} diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index f0015fa876e..b49501fc0a7 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -23,12 +23,11 @@ #include <linux/nfs.h> #include <linux/nfs2.h> #include <linux/nfs_fs.h> +#include "internal.h" #define NFSDBG_FACILITY NFSDBG_XDR /* #define NFS_PARANOIA 1 */ -extern int nfs_stat_to_errno(int stat); - /* Mapping from NFS error code to "errno" error code. */ #define errno_NFSERR_IO EIO @@ -52,7 +51,7 @@ extern int nfs_stat_to_errno(int stat); #define NFS_createargs_sz (NFS_diropargs_sz+NFS_sattr_sz) #define NFS_renameargs_sz (NFS_diropargs_sz+NFS_diropargs_sz) #define NFS_linkargs_sz (NFS_fhandle_sz+NFS_diropargs_sz) -#define NFS_symlinkargs_sz (NFS_diropargs_sz+NFS_path_sz+NFS_sattr_sz) +#define NFS_symlinkargs_sz (NFS_diropargs_sz+1+NFS_sattr_sz) #define NFS_readdirargs_sz (NFS_fhandle_sz+2) #define NFS_attrstat_sz (1+NFS_fattr_sz) @@ -131,7 +130,8 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) fattr->du.nfs2.blocksize = ntohl(*p++); rdev = ntohl(*p++); fattr->du.nfs2.blocks = ntohl(*p++); - fattr->fsid_u.nfs3 = ntohl(*p++); + fattr->fsid.major = ntohl(*p++); + fattr->fsid.minor = 0; fattr->fileid = ntohl(*p++); p = xdr_decode_time(p, &fattr->atime); p = xdr_decode_time(p, &fattr->mtime); @@ -351,11 +351,26 @@ nfs_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs_linkargs *args) static int nfs_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_symlinkargs *args) { + struct xdr_buf *sndbuf = &req->rq_snd_buf; + size_t pad; + p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_array(p, args->fromname, args->fromlen); - p = xdr_encode_array(p, args->topath, args->tolen); + *p++ = htonl(args->pathlen); + sndbuf->len = xdr_adjust_iovec(sndbuf->head, p); + + xdr_encode_pages(sndbuf, args->pages, 0, args->pathlen); + + /* + * xdr_encode_pages may have added a few bytes to ensure the + * pathname ends on a 4-byte boundary. Start encoding the + * attributes after the pad bytes. + */ + pad = sndbuf->tail->iov_len; + if (pad > 0) + p++; p = xdr_encode_sattr(p, args->sattr); - req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); + sndbuf->len += xdr_adjust_iovec(sndbuf->tail, p) - pad; return 0; } diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index 33287879bd2..7322da4d205 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -172,8 +172,10 @@ static void nfs3_cache_acls(struct inode *inode, struct posix_acl *acl, inode->i_ino, acl, dfacl); spin_lock(&inode->i_lock); __nfs3_forget_cached_acls(NFS_I(inode)); - nfsi->acl_access = posix_acl_dup(acl); - nfsi->acl_default = posix_acl_dup(dfacl); + if (!IS_ERR(acl)) + nfsi->acl_access = posix_acl_dup(acl); + if (!IS_ERR(dfacl)) + nfsi->acl_default = posix_acl_dup(dfacl); spin_unlock(&inode->i_lock); } @@ -254,7 +256,9 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type) res.acl_access = NULL; } } - nfs3_cache_acls(inode, res.acl_access, res.acl_default); + nfs3_cache_acls(inode, + (res.mask & NFS_ACL) ? res.acl_access : ERR_PTR(-EINVAL), + (res.mask & NFS_DFACL) ? res.acl_default : ERR_PTR(-EINVAL)); switch(type) { case ACL_TYPE_ACCESS: @@ -329,6 +333,7 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, switch (status) { case 0: status = nfs_refresh_inode(inode, &fattr); + nfs3_cache_acls(inode, acl, dfacl); break; case -EPFNOSUPPORT: case -EPROTONOSUPPORT: diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index cf186f0d2b3..3b234d4601e 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -20,11 +20,10 @@ #include <linux/nfs_mount.h> #include "iostat.h" +#include "internal.h" #define NFSDBG_FACILITY NFSDBG_PROC -extern struct rpc_procinfo nfs3_procedures[]; - /* A wrapper to handle the EJUKEBOX error message */ static int nfs3_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) @@ -82,7 +81,7 @@ do_proc_get_root(struct rpc_clnt *client, struct nfs_fh *fhandle, } /* - * Bare-bones access to getattr: this is for nfs_read_super. + * Bare-bones access to getattr: this is for nfs_get_root/nfs_get_sb */ static int nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, @@ -91,8 +90,8 @@ nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, int status; status = do_proc_get_root(server->client, fhandle, info); - if (status && server->client_sys != server->client) - status = do_proc_get_root(server->client_sys, fhandle, info); + if (status && server->nfs_client->cl_rpcclient != server->client) + status = do_proc_get_root(server->nfs_client->cl_rpcclient, fhandle, info); return status; } @@ -450,7 +449,7 @@ nfs3_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, struct qstr struct nfs_fattr res; } *ptr; - ptr = (struct unlinkxdr *)kmalloc(sizeof(*ptr), GFP_KERNEL); + ptr = kmalloc(sizeof(*ptr), GFP_KERNEL); if (!ptr) return -ENOMEM; ptr->arg.fh = NFS_FH(dir->d_inode); @@ -545,23 +544,23 @@ nfs3_proc_link(struct inode *inode, struct inode *dir, struct qstr *name) } static int -nfs3_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path, - struct iattr *sattr, struct nfs_fh *fhandle, - struct nfs_fattr *fattr) +nfs3_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page, + unsigned int len, struct iattr *sattr) { - struct nfs_fattr dir_attr; + struct nfs_fh fhandle; + struct nfs_fattr fattr, dir_attr; struct nfs3_symlinkargs arg = { .fromfh = NFS_FH(dir), - .fromname = name->name, - .fromlen = name->len, - .topath = path->name, - .tolen = path->len, + .fromname = dentry->d_name.name, + .fromlen = dentry->d_name.len, + .pages = &page, + .pathlen = len, .sattr = sattr }; struct nfs3_diropres res = { .dir_attr = &dir_attr, - .fh = fhandle, - .fattr = fattr + .fh = &fhandle, + .fattr = &fattr }; struct rpc_message msg = { .rpc_proc = &nfs3_procedures[NFS3PROC_SYMLINK], @@ -570,13 +569,19 @@ nfs3_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path, }; int status; - if (path->len > NFS3_MAXPATHLEN) + if (len > NFS3_MAXPATHLEN) return -ENAMETOOLONG; - dprintk("NFS call symlink %s -> %s\n", name->name, path->name); + + dprintk("NFS call symlink %s\n", dentry->d_name.name); + nfs_fattr_init(&dir_attr); - nfs_fattr_init(fattr); + nfs_fattr_init(&fattr); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); nfs_post_op_update_inode(dir, &dir_attr); + if (status != 0) + goto out; + status = nfs_instantiate(dentry, &fhandle, &fattr); +out: dprintk("NFS reply symlink: %d\n", status); return status; } @@ -786,7 +791,7 @@ nfs3_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, dprintk("NFS call fsinfo\n"); nfs_fattr_init(info->fattr); - status = rpc_call_sync(server->client_sys, &msg, 0); + status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0); dprintk("NFS reply fsinfo: %d\n", status); return status; } @@ -809,8 +814,6 @@ nfs3_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, return status; } -extern u32 *nfs3_decode_dirent(u32 *, struct nfs_entry *, int); - static int nfs3_read_done(struct rpc_task *task, struct nfs_read_data *data) { if (nfs3_async_handle_jukebox(task, data->inode)) @@ -889,7 +892,7 @@ nfs3_proc_lock(struct file *filp, int cmd, struct file_lock *fl) return nlmclnt_proc(filp->f_dentry->d_inode, cmd, fl); } -struct nfs_rpc_ops nfs_v3_clientops = { +const struct nfs_rpc_ops nfs_v3_clientops = { .version = 3, /* protocol version */ .dentry_ops = &nfs_dentry_operations, .dir_inode_ops = &nfs3_dir_inode_operations, diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index ec233619687..16556fa4eff 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -22,14 +22,13 @@ #include <linux/nfs3.h> #include <linux/nfs_fs.h> #include <linux/nfsacl.h> +#include "internal.h" #define NFSDBG_FACILITY NFSDBG_XDR /* Mapping from NFS error code to "errno" error code. */ #define errno_NFSERR_IO EIO -extern int nfs_stat_to_errno(int); - /* * Declare the space requirements for NFS arguments and replies as * number of 32bit-words @@ -57,7 +56,7 @@ extern int nfs_stat_to_errno(int); #define NFS3_writeargs_sz (NFS3_fh_sz+5) #define NFS3_createargs_sz (NFS3_diropargs_sz+NFS3_sattr_sz) #define NFS3_mkdirargs_sz (NFS3_diropargs_sz+NFS3_sattr_sz) -#define NFS3_symlinkargs_sz (NFS3_diropargs_sz+NFS3_path_sz+NFS3_sattr_sz) +#define NFS3_symlinkargs_sz (NFS3_diropargs_sz+1+NFS3_sattr_sz) #define NFS3_mknodargs_sz (NFS3_diropargs_sz+2+NFS3_sattr_sz) #define NFS3_renameargs_sz (NFS3_diropargs_sz+NFS3_diropargs_sz) #define NFS3_linkargs_sz (NFS3_fh_sz+NFS3_diropargs_sz) @@ -166,7 +165,8 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) if (MAJOR(fattr->rdev) != major || MINOR(fattr->rdev) != minor) fattr->rdev = 0; - p = xdr_decode_hyper(p, &fattr->fsid_u.nfs3); + p = xdr_decode_hyper(p, &fattr->fsid.major); + fattr->fsid.minor = 0; p = xdr_decode_hyper(p, &fattr->fileid); p = xdr_decode_time3(p, &fattr->atime); p = xdr_decode_time3(p, &fattr->mtime); @@ -398,8 +398,11 @@ nfs3_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_symlinkargs *args p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_array(p, args->fromname, args->fromlen); p = xdr_encode_sattr(p, args->sattr); - p = xdr_encode_array(p, args->topath, args->tolen); + *p++ = htonl(args->pathlen); req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); + + /* Copy the page */ + xdr_encode_pages(&req->rq_snd_buf, args->pages, 0, args->pathlen); return 0; } diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 0f5e4e7cdde..61095fe4b5c 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -43,55 +43,6 @@ enum nfs4_client_state { }; /* - * The nfs4_client identifies our client state to the server. - */ -struct nfs4_client { - struct list_head cl_servers; /* Global list of servers */ - struct in_addr cl_addr; /* Server identifier */ - u64 cl_clientid; /* constant */ - nfs4_verifier cl_confirm; - unsigned long cl_state; - - u32 cl_lockowner_id; - - /* - * The following rwsem ensures exclusive access to the server - * while we recover the state following a lease expiration. - */ - struct rw_semaphore cl_sem; - - struct list_head cl_delegations; - struct list_head cl_state_owners; - struct list_head cl_unused; - int cl_nunused; - spinlock_t cl_lock; - atomic_t cl_count; - - struct rpc_clnt * cl_rpcclient; - - struct list_head cl_superblocks; /* List of nfs_server structs */ - - unsigned long cl_lease_time; - unsigned long cl_last_renewal; - struct work_struct cl_renewd; - struct work_struct cl_recoverd; - - struct rpc_wait_queue cl_rpcwaitq; - - /* used for the setclientid verifier */ - struct timespec cl_boot_time; - - /* idmapper */ - struct idmap * cl_idmap; - - /* Our own IP address, as a null-terminated string. - * This is used to generate the clientid, and the callback address. - */ - char cl_ipaddr[16]; - unsigned char cl_id_uniquifier; -}; - -/* * struct rpc_sequence ensures that RPC calls are sent in the exact * order that they appear on the list. */ @@ -127,7 +78,7 @@ static inline void nfs_confirm_seqid(struct nfs_seqid_counter *seqid, int status struct nfs4_state_owner { spinlock_t so_lock; struct list_head so_list; /* per-clientid list of state_owners */ - struct nfs4_client *so_client; + struct nfs_client *so_client; u32 so_id; /* 32-bit identifier, unique */ atomic_t so_count; @@ -210,13 +161,16 @@ extern ssize_t nfs4_listxattr(struct dentry *, char *, size_t); /* nfs4proc.c */ extern int nfs4_map_errors(int err); -extern int nfs4_proc_setclientid(struct nfs4_client *, u32, unsigned short, struct rpc_cred *); -extern int nfs4_proc_setclientid_confirm(struct nfs4_client *, struct rpc_cred *); -extern int nfs4_proc_async_renew(struct nfs4_client *, struct rpc_cred *); -extern int nfs4_proc_renew(struct nfs4_client *, struct rpc_cred *); +extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *); +extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct rpc_cred *); +extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *); +extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *); extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state); extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *); extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *); +extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle); +extern int nfs4_proc_fs_locations(struct inode *dir, struct dentry *dentry, + struct nfs4_fs_locations *fs_locations, struct page *page); extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops; extern struct nfs4_state_recovery_ops nfs4_network_partition_recovery_ops; @@ -225,21 +179,17 @@ extern const u32 nfs4_fattr_bitmap[2]; extern const u32 nfs4_statfs_bitmap[2]; extern const u32 nfs4_pathconf_bitmap[2]; extern const u32 nfs4_fsinfo_bitmap[2]; +extern const u32 nfs4_fs_locations_bitmap[2]; /* nfs4renewd.c */ -extern void nfs4_schedule_state_renewal(struct nfs4_client *); +extern void nfs4_schedule_state_renewal(struct nfs_client *); extern void nfs4_renewd_prepare_shutdown(struct nfs_server *); -extern void nfs4_kill_renewd(struct nfs4_client *); +extern void nfs4_kill_renewd(struct nfs_client *); extern void nfs4_renew_state(void *); /* nfs4state.c */ -extern void init_nfsv4_state(struct nfs_server *); -extern void destroy_nfsv4_state(struct nfs_server *); -extern struct nfs4_client *nfs4_get_client(struct in_addr *); -extern void nfs4_put_client(struct nfs4_client *clp); -extern struct nfs4_client *nfs4_find_client(struct in_addr *); -struct rpc_cred *nfs4_get_renew_cred(struct nfs4_client *clp); -extern u32 nfs4_alloc_lockowner_id(struct nfs4_client *); +struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp); +extern u32 nfs4_alloc_lockowner_id(struct nfs_client *); extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *); extern void nfs4_put_state_owner(struct nfs4_state_owner *); @@ -248,7 +198,7 @@ extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state extern void nfs4_put_open_state(struct nfs4_state *); extern void nfs4_close_state(struct nfs4_state *, mode_t); extern void nfs4_state_set_mode_locked(struct nfs4_state *, mode_t); -extern void nfs4_schedule_state_recovery(struct nfs4_client *); +extern void nfs4_schedule_state_recovery(struct nfs_client *); 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_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t); @@ -272,10 +222,6 @@ extern struct svc_version nfs4_callback_version1; #else -#define init_nfsv4_state(server) do { } while (0) -#define destroy_nfsv4_state(server) do { } while (0) -#define nfs4_put_state_owner(inode, owner) do { } while (0) -#define nfs4_put_open_state(state) do { } while (0) #define nfs4_close_state(a, b) do { } while (0) #endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c new file mode 100644 index 00000000000..24e47f3bbd1 --- /dev/null +++ b/fs/nfs/nfs4namespace.c @@ -0,0 +1,261 @@ +/* + * linux/fs/nfs/nfs4namespace.c + * + * Copyright (C) 2005 Trond Myklebust <Trond.Myklebust@netapp.com> + * - Modified by David Howells <dhowells@redhat.com> + * + * NFSv4 namespace + */ + +#include <linux/config.h> + +#include <linux/dcache.h> +#include <linux/mount.h> +#include <linux/namei.h> +#include <linux/nfs_fs.h> +#include <linux/string.h> +#include <linux/sunrpc/clnt.h> +#include <linux/vfs.h> +#include <linux/inet.h> +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +/* + * Check if fs_root is valid + */ +static inline char *nfs4_pathname_string(const struct nfs4_pathname *pathname, + char *buffer, ssize_t buflen) +{ + char *end = buffer + buflen; + int n; + + *--end = '\0'; + buflen--; + + n = pathname->ncomponents; + while (--n >= 0) { + const struct nfs4_string *component = &pathname->components[n]; + buflen -= component->len + 1; + if (buflen < 0) + goto Elong; + end -= component->len; + memcpy(end, component->data, component->len); + *--end = '/'; + } + return end; +Elong: + return ERR_PTR(-ENAMETOOLONG); +} + +/* + * Determine the mount path as a string + */ +static char *nfs4_path(const struct vfsmount *mnt_parent, + const struct dentry *dentry, + char *buffer, ssize_t buflen) +{ + const char *srvpath; + + srvpath = strchr(mnt_parent->mnt_devname, ':'); + if (srvpath) + srvpath++; + else + srvpath = mnt_parent->mnt_devname; + + return nfs_path(srvpath, mnt_parent->mnt_root, dentry, buffer, buflen); +} + +/* + * Check that fs_locations::fs_root [RFC3530 6.3] is a prefix for what we + * believe to be the server path to this dentry + */ +static int nfs4_validate_fspath(const struct vfsmount *mnt_parent, + const struct dentry *dentry, + const struct nfs4_fs_locations *locations, + char *page, char *page2) +{ + const char *path, *fs_path; + + path = nfs4_path(mnt_parent, dentry, page, PAGE_SIZE); + if (IS_ERR(path)) + return PTR_ERR(path); + + fs_path = nfs4_pathname_string(&locations->fs_path, page2, PAGE_SIZE); + if (IS_ERR(fs_path)) + return PTR_ERR(fs_path); + + if (strncmp(path, fs_path, strlen(fs_path)) != 0) { + dprintk("%s: path %s does not begin with fsroot %s\n", + __FUNCTION__, path, fs_path); + return -ENOENT; + } + + return 0; +} + +/* + * Check if the string represents a "valid" IPv4 address + */ +static inline int valid_ipaddr4(const char *buf) +{ + int rc, count, in[4]; + + rc = sscanf(buf, "%d.%d.%d.%d", &in[0], &in[1], &in[2], &in[3]); + if (rc != 4) + return -EINVAL; + for (count = 0; count < 4; count++) { + if (in[count] > 255) + return -EINVAL; + } + return 0; +} + +/** + * nfs_follow_referral - set up mountpoint when hitting a referral on moved error + * @mnt_parent - mountpoint of parent directory + * @dentry - parent directory + * @fspath - fs path returned in fs_locations + * @mntpath - mount path to new server + * @hostname - hostname of new server + * @addr - host addr of new server + * + */ +static struct vfsmount *nfs_follow_referral(const struct vfsmount *mnt_parent, + const struct dentry *dentry, + const struct nfs4_fs_locations *locations) +{ + struct vfsmount *mnt = ERR_PTR(-ENOENT); + struct nfs_clone_mount mountdata = { + .sb = mnt_parent->mnt_sb, + .dentry = dentry, + .authflavor = NFS_SB(mnt_parent->mnt_sb)->client->cl_auth->au_flavor, + }; + char *page = NULL, *page2 = NULL; + char *devname; + int loc, s, error; + + if (locations == NULL || locations->nlocations <= 0) + goto out; + + dprintk("%s: referral at %s/%s\n", __FUNCTION__, + dentry->d_parent->d_name.name, dentry->d_name.name); + + page = (char *) __get_free_page(GFP_USER); + if (!page) + goto out; + + page2 = (char *) __get_free_page(GFP_USER); + if (!page2) + goto out; + + /* Ensure fs path is a prefix of current dentry path */ + error = nfs4_validate_fspath(mnt_parent, dentry, locations, page, page2); + if (error < 0) { + mnt = ERR_PTR(error); + goto out; + } + + devname = nfs_devname(mnt_parent, dentry, page, PAGE_SIZE); + if (IS_ERR(devname)) { + mnt = (struct vfsmount *)devname; + goto out; + } + + loc = 0; + while (loc < locations->nlocations && IS_ERR(mnt)) { + const struct nfs4_fs_location *location = &locations->locations[loc]; + char *mnt_path; + + if (location == NULL || location->nservers <= 0 || + location->rootpath.ncomponents == 0) { + loc++; + continue; + } + + mnt_path = nfs4_pathname_string(&location->rootpath, page2, PAGE_SIZE); + if (IS_ERR(mnt_path)) { + loc++; + continue; + } + mountdata.mnt_path = mnt_path; + + s = 0; + while (s < location->nservers) { + struct sockaddr_in addr = {}; + + if (location->servers[s].len <= 0 || + valid_ipaddr4(location->servers[s].data) < 0) { + s++; + continue; + } + + mountdata.hostname = location->servers[s].data; + addr.sin_addr.s_addr = in_aton(mountdata.hostname); + addr.sin_family = AF_INET; + addr.sin_port = htons(NFS_PORT); + mountdata.addr = &addr; + + mnt = vfs_kern_mount(&nfs4_referral_fs_type, 0, devname, &mountdata); + if (!IS_ERR(mnt)) { + break; + } + s++; + } + loc++; + } + +out: + free_page((unsigned long) page); + free_page((unsigned long) page2); + dprintk("%s: done\n", __FUNCTION__); + return mnt; +} + +/* + * nfs_do_refmount - handle crossing a referral on server + * @dentry - dentry of referral + * @nd - nameidata info + * + */ +struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentry *dentry) +{ + struct vfsmount *mnt = ERR_PTR(-ENOMEM); + struct dentry *parent; + struct nfs4_fs_locations *fs_locations = NULL; + struct page *page; + int err; + + /* BUG_ON(IS_ROOT(dentry)); */ + dprintk("%s: enter\n", __FUNCTION__); + + page = alloc_page(GFP_KERNEL); + if (page == NULL) + goto out; + + fs_locations = kmalloc(sizeof(struct nfs4_fs_locations), GFP_KERNEL); + if (fs_locations == NULL) + goto out_free; + + /* Get locations */ + mnt = ERR_PTR(-ENOENT); + + parent = dget_parent(dentry); + dprintk("%s: getting locations for %s/%s\n", + __FUNCTION__, parent->d_name.name, dentry->d_name.name); + + err = nfs4_proc_fs_locations(parent->d_inode, dentry, fs_locations, page); + dput(parent); + if (err != 0 || + fs_locations->nlocations <= 0 || + fs_locations->fs_path.ncomponents <= 0) + goto out_free; + + mnt = nfs_follow_referral(mnt_parent, dentry, fs_locations); +out_free: + __free_page(page); + kfree(fs_locations); +out: + dprintk("%s: done\n", __FUNCTION__); + return mnt; +} diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index d86c0db7b1e..47c7e6e3910 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -55,7 +55,7 @@ #define NFSDBG_FACILITY NFSDBG_PROC -#define NFS4_POLL_RETRY_MIN (1*HZ) +#define NFS4_POLL_RETRY_MIN (HZ/10) #define NFS4_POLL_RETRY_MAX (15*HZ) struct nfs4_opendata; @@ -64,9 +64,7 @@ static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinf static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *); static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry); static int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception); -static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs4_client *clp); -extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus); -extern struct rpc_procinfo nfs4_procedures[]; +static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp); /* Prevent leaks of NFSv4 errors into userland */ int nfs4_map_errors(int err) @@ -121,6 +119,25 @@ const u32 nfs4_fsinfo_bitmap[2] = { FATTR4_WORD0_MAXFILESIZE 0 }; +const u32 nfs4_fs_locations_bitmap[2] = { + FATTR4_WORD0_TYPE + | FATTR4_WORD0_CHANGE + | FATTR4_WORD0_SIZE + | FATTR4_WORD0_FSID + | FATTR4_WORD0_FILEID + | FATTR4_WORD0_FS_LOCATIONS, + FATTR4_WORD1_MODE + | FATTR4_WORD1_NUMLINKS + | FATTR4_WORD1_OWNER + | FATTR4_WORD1_OWNER_GROUP + | FATTR4_WORD1_RAWDEV + | FATTR4_WORD1_SPACE_USED + | FATTR4_WORD1_TIME_ACCESS + | FATTR4_WORD1_TIME_METADATA + | FATTR4_WORD1_TIME_MODIFY + | FATTR4_WORD1_MOUNTED_ON_FILEID +}; + static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, struct nfs4_readdir_arg *readdir) { @@ -178,22 +195,22 @@ static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, static void renew_lease(const struct nfs_server *server, unsigned long timestamp) { - struct nfs4_client *clp = server->nfs4_state; + struct nfs_client *clp = server->nfs_client; spin_lock(&clp->cl_lock); if (time_before(clp->cl_last_renewal,timestamp)) clp->cl_last_renewal = timestamp; spin_unlock(&clp->cl_lock); } -static void update_changeattr(struct inode *inode, struct nfs4_change_info *cinfo) +static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo) { - struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_inode *nfsi = NFS_I(dir); - spin_lock(&inode->i_lock); - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; + spin_lock(&dir->i_lock); + nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_DATA; if (cinfo->before == nfsi->change_attr && cinfo->atomic) nfsi->change_attr = cinfo->after; - spin_unlock(&inode->i_lock); + spin_unlock(&dir->i_lock); } struct nfs4_opendata { @@ -235,7 +252,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct dentry *dentry, atomic_inc(&sp->so_count); p->o_arg.fh = NFS_FH(dir); p->o_arg.open_flags = flags, - p->o_arg.clientid = server->nfs4_state->cl_clientid; + p->o_arg.clientid = server->nfs_client->cl_clientid; p->o_arg.id = sp->so_id; p->o_arg.name = &dentry->d_name; p->o_arg.server = server; @@ -533,7 +550,7 @@ int nfs4_open_delegation_recall(struct dentry *dentry, struct nfs4_state *state) case -NFS4ERR_STALE_STATEID: case -NFS4ERR_EXPIRED: /* Don't recall a delegation if it was lost */ - nfs4_schedule_state_recovery(server->nfs4_state); + nfs4_schedule_state_recovery(server->nfs_client); return err; } err = nfs4_handle_exception(server, err, &exception); @@ -741,7 +758,7 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) } nfs_confirm_seqid(&data->owner->so_seqid, 0); if (!(o_res->f_attr->valid & NFS_ATTR_FATTR)) - return server->rpc_ops->getattr(server, &o_res->fh, o_res->f_attr); + return server->nfs_client->rpc_ops->getattr(server, &o_res->fh, o_res->f_attr); return 0; } @@ -775,11 +792,18 @@ out: int nfs4_recover_expired_lease(struct nfs_server *server) { - struct nfs4_client *clp = server->nfs4_state; + struct nfs_client *clp = server->nfs_client; + int ret; - if (test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) + for (;;) { + ret = nfs4_wait_clnt_recover(server->client, clp); + if (ret != 0) + return ret; + if (!test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) + break; nfs4_schedule_state_recovery(clp); - return nfs4_wait_clnt_recover(server->client, clp); + } + return 0; } /* @@ -850,7 +874,7 @@ static int _nfs4_open_delegated(struct inode *inode, int flags, struct rpc_cred { struct nfs_delegation *delegation; struct nfs_server *server = NFS_SERVER(inode); - struct nfs4_client *clp = server->nfs4_state; + struct nfs_client *clp = server->nfs_client; struct nfs_inode *nfsi = NFS_I(inode); struct nfs4_state_owner *sp = NULL; struct nfs4_state *state = NULL; @@ -936,7 +960,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st struct nfs4_state_owner *sp; struct nfs4_state *state = NULL; struct nfs_server *server = NFS_SERVER(dir); - struct nfs4_client *clp = server->nfs4_state; + struct nfs_client *clp = server->nfs_client; struct nfs4_opendata *opendata; int status; @@ -953,7 +977,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st status = -ENOMEM; opendata = nfs4_opendata_alloc(dentry, sp, flags, sattr); if (opendata == NULL) - goto err_put_state_owner; + goto err_release_rwsem; status = _nfs4_proc_open(opendata); if (status != 0) @@ -972,11 +996,11 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st return 0; err_opendata_free: nfs4_opendata_free(opendata); +err_release_rwsem: + up_read(&clp->cl_sem); err_put_state_owner: nfs4_put_state_owner(sp); out_err: - /* Note: clp->cl_sem must be released before nfs4_put_open_state()! */ - up_read(&clp->cl_sem); *res = NULL; return status; } @@ -1116,7 +1140,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data) break; case -NFS4ERR_STALE_STATEID: case -NFS4ERR_EXPIRED: - nfs4_schedule_state_recovery(server->nfs4_state); + nfs4_schedule_state_recovery(server->nfs_client); break; default: if (nfs4_async_handle_error(task, server) == -EAGAIN) { @@ -1251,7 +1275,7 @@ nfs4_atomic_open(struct inode *dir, struct dentry *dentry, struct nameidata *nd) BUG_ON(nd->intent.open.flags & O_CREAT); } - cred = rpcauth_lookupcred(NFS_SERVER(dir)->client->cl_auth, 0); + cred = rpcauth_lookupcred(NFS_CLIENT(dir)->cl_auth, 0); if (IS_ERR(cred)) return (struct dentry *)cred; state = nfs4_do_open(dir, dentry, nd->intent.open.flags, &attr, cred); @@ -1274,7 +1298,7 @@ nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags, st struct rpc_cred *cred; struct nfs4_state *state; - cred = rpcauth_lookupcred(NFS_SERVER(dir)->client->cl_auth, 0); + cred = rpcauth_lookupcred(NFS_CLIENT(dir)->cl_auth, 0); if (IS_ERR(cred)) return PTR_ERR(cred); state = nfs4_open_delegated(dentry->d_inode, openflags, cred); @@ -1331,7 +1355,7 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f return status; } -static int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) +int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) { struct nfs4_exception exception = { }; int err; @@ -1376,73 +1400,66 @@ static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, return err; } +/* + * get the file handle for the "/" directory on the server + */ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fsinfo *info) + struct nfs_fsinfo *info) { - struct nfs_fattr * fattr = info->fattr; - unsigned char * p; - struct qstr q; - struct nfs4_lookup_arg args = { - .dir_fh = fhandle, - .name = &q, - .bitmask = nfs4_fattr_bitmap, - }; - struct nfs4_lookup_res res = { - .server = server, - .fattr = fattr, - .fh = fhandle, - }; - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP], - .rpc_argp = &args, - .rpc_resp = &res, - }; int status; - /* - * Now we do a separate LOOKUP for each component of the mount path. - * The LOOKUPs are done separately so that we can conveniently - * catch an ERR_WRONGSEC if it occurs along the way... - */ status = nfs4_lookup_root(server, fhandle, info); - if (status) - goto out; - - p = server->mnt_path; - for (;;) { - struct nfs4_exception exception = { }; - - while (*p == '/') - p++; - if (!*p) - break; - q.name = p; - while (*p && (*p != '/')) - p++; - q.len = p - q.name; - - do { - nfs_fattr_init(fattr); - status = nfs4_handle_exception(server, - rpc_call_sync(server->client, &msg, 0), - &exception); - } while (exception.retry); - if (status == 0) - continue; - if (status == -ENOENT) { - printk(KERN_NOTICE "NFS: mount path %s does not exist!\n", server->mnt_path); - printk(KERN_NOTICE "NFS: suggestion: try mounting '/' instead.\n"); - } - break; - } if (status == 0) status = nfs4_server_capabilities(server, fhandle); if (status == 0) status = nfs4_do_fsinfo(server, fhandle, info); -out: return nfs4_map_errors(status); } +/* + * Get locations and (maybe) other attributes of a referral. + * Note that we'll actually follow the referral later when + * we detect fsid mismatch in inode revalidation + */ +static int nfs4_get_referral(struct inode *dir, struct qstr *name, struct nfs_fattr *fattr, struct nfs_fh *fhandle) +{ + int status = -ENOMEM; + struct page *page = NULL; + struct nfs4_fs_locations *locations = NULL; + struct dentry dentry = {}; + + page = alloc_page(GFP_KERNEL); + if (page == NULL) + goto out; + locations = kmalloc(sizeof(struct nfs4_fs_locations), GFP_KERNEL); + if (locations == NULL) + goto out; + + dentry.d_name.name = name->name; + dentry.d_name.len = name->len; + status = nfs4_proc_fs_locations(dir, &dentry, locations, page); + if (status != 0) + goto out; + /* Make sure server returned a different fsid for the referral */ + if (nfs_fsid_equal(&NFS_SERVER(dir)->fsid, &locations->fattr.fsid)) { + dprintk("%s: server did not return a different fsid for a referral at %s\n", __FUNCTION__, name->name); + status = -EIO; + goto out; + } + + memcpy(fattr, &locations->fattr, sizeof(struct nfs_fattr)); + fattr->valid |= NFS_ATTR_FATTR_V4_REFERRAL; + if (!fattr->mode) + fattr->mode = S_IFDIR; + memset(fhandle, 0, sizeof(struct nfs_fh)); +out: + if (page) + __free_page(page); + if (locations) + kfree(locations); + return status; +} + static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr) { struct nfs4_getattr_arg args = { @@ -1504,7 +1521,7 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, nfs_fattr_init(fattr); - cred = rpcauth_lookupcred(NFS_SERVER(inode)->client->cl_auth, 0); + cred = rpcauth_lookupcred(NFS_CLIENT(inode)->cl_auth, 0); if (IS_ERR(cred)) return PTR_ERR(cred); @@ -1522,6 +1539,52 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, return status; } +static int _nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh, + struct qstr *name, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + int status; + struct nfs4_lookup_arg args = { + .bitmask = server->attr_bitmask, + .dir_fh = dirfh, + .name = name, + }; + struct nfs4_lookup_res res = { + .server = server, + .fattr = fattr, + .fh = fhandle, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOOKUP], + .rpc_argp = &args, + .rpc_resp = &res, + }; + + nfs_fattr_init(fattr); + + dprintk("NFS call lookupfh %s\n", name->name); + status = rpc_call_sync(server->client, &msg, 0); + dprintk("NFS reply lookupfh: %d\n", status); + if (status == -NFS4ERR_MOVED) + status = -EREMOTE; + return status; +} + +static int nfs4_proc_lookupfh(struct nfs_server *server, struct nfs_fh *dirfh, + struct qstr *name, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + struct nfs4_exception exception = { }; + int err; + do { + err = nfs4_handle_exception(server, + _nfs4_proc_lookupfh(server, dirfh, name, + fhandle, fattr), + &exception); + } while (exception.retry); + return err; +} + static int _nfs4_proc_lookup(struct inode *dir, struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr) { @@ -1547,6 +1610,8 @@ static int _nfs4_proc_lookup(struct inode *dir, struct qstr *name, dprintk("NFS call lookup %s\n", name->name); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + if (status == -NFS4ERR_MOVED) + status = nfs4_get_referral(dir, name, fattr, fhandle); dprintk("NFS reply lookup: %d\n", status); return status; } @@ -1818,7 +1883,7 @@ nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, struct rpc_cred *cred; int status = 0; - cred = rpcauth_lookupcred(NFS_SERVER(dir)->client->cl_auth, 0); + cred = rpcauth_lookupcred(NFS_CLIENT(dir)->cl_auth, 0); if (IS_ERR(cred)) { status = PTR_ERR(cred); goto out; @@ -2008,7 +2073,7 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr * if (!status) { update_changeattr(dir, &res.cinfo); nfs_post_op_update_inode(dir, res.dir_attr); - nfs_refresh_inode(inode, res.fattr); + nfs_post_op_update_inode(inode, res.fattr); } return status; @@ -2026,24 +2091,24 @@ static int nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *n return err; } -static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name, - struct qstr *path, struct iattr *sattr, struct nfs_fh *fhandle, - struct nfs_fattr *fattr) +static int _nfs4_proc_symlink(struct inode *dir, struct dentry *dentry, + struct page *page, unsigned int len, struct iattr *sattr) { struct nfs_server *server = NFS_SERVER(dir); - struct nfs_fattr dir_fattr; + struct nfs_fh fhandle; + struct nfs_fattr fattr, dir_fattr; struct nfs4_create_arg arg = { .dir_fh = NFS_FH(dir), .server = server, - .name = name, + .name = &dentry->d_name, .attrs = sattr, .ftype = NF4LNK, .bitmask = server->attr_bitmask, }; struct nfs4_create_res res = { .server = server, - .fh = fhandle, - .fattr = fattr, + .fh = &fhandle, + .fattr = &fattr, .dir_fattr = &dir_fattr, }; struct rpc_message msg = { @@ -2053,29 +2118,32 @@ static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name, }; int status; - if (path->len > NFS4_MAXPATHLEN) + if (len > NFS4_MAXPATHLEN) return -ENAMETOOLONG; - arg.u.symlink = path; - nfs_fattr_init(fattr); + + arg.u.symlink.pages = &page; + arg.u.symlink.len = len; + nfs_fattr_init(&fattr); nfs_fattr_init(&dir_fattr); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); - if (!status) + if (!status) { update_changeattr(dir, &res.dir_cinfo); - nfs_post_op_update_inode(dir, res.dir_fattr); + nfs_post_op_update_inode(dir, res.dir_fattr); + status = nfs_instantiate(dentry, &fhandle, &fattr); + } return status; } -static int nfs4_proc_symlink(struct inode *dir, struct qstr *name, - struct qstr *path, struct iattr *sattr, struct nfs_fh *fhandle, - struct nfs_fattr *fattr) +static int nfs4_proc_symlink(struct inode *dir, struct dentry *dentry, + struct page *page, unsigned int len, struct iattr *sattr) { struct nfs4_exception exception = { }; int err; do { err = nfs4_handle_exception(NFS_SERVER(dir), - _nfs4_proc_symlink(dir, name, path, sattr, - fhandle, fattr), + _nfs4_proc_symlink(dir, dentry, page, + len, sattr), &exception); } while (exception.retry); return err; @@ -2458,7 +2526,7 @@ static void nfs4_proc_commit_setup(struct nfs_write_data *data, int how) */ static void nfs4_renew_done(struct rpc_task *task, void *data) { - struct nfs4_client *clp = (struct nfs4_client *)task->tk_msg.rpc_argp; + struct nfs_client *clp = (struct nfs_client *)task->tk_msg.rpc_argp; unsigned long timestamp = (unsigned long)data; if (task->tk_status < 0) { @@ -2480,7 +2548,7 @@ static const struct rpc_call_ops nfs4_renew_ops = { .rpc_call_done = nfs4_renew_done, }; -int nfs4_proc_async_renew(struct nfs4_client *clp, struct rpc_cred *cred) +int nfs4_proc_async_renew(struct nfs_client *clp, struct rpc_cred *cred) { struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENEW], @@ -2492,7 +2560,7 @@ int nfs4_proc_async_renew(struct nfs4_client *clp, struct rpc_cred *cred) &nfs4_renew_ops, (void *)jiffies); } -int nfs4_proc_renew(struct nfs4_client *clp, struct rpc_cred *cred) +int nfs4_proc_renew(struct nfs_client *clp, struct rpc_cred *cred) { struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENEW], @@ -2605,7 +2673,7 @@ out: nfs4_set_cached_acl(inode, acl); } -static inline ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) +static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) { struct page *pages[NFS4ACL_MAXPAGES]; struct nfs_getaclargs args = { @@ -2658,6 +2726,19 @@ out_free: return ret; } +static ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) +{ + struct nfs4_exception exception = { }; + ssize_t ret; + do { + ret = __nfs4_get_acl_uncached(inode, buf, buflen); + if (ret >= 0) + break; + ret = nfs4_handle_exception(NFS_SERVER(inode), ret, &exception); + } while (exception.retry); + return ret; +} + static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen) { struct nfs_server *server = NFS_SERVER(inode); @@ -2674,7 +2755,7 @@ static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen) return nfs4_get_acl_uncached(inode, buf, buflen); } -static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen) +static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen) { struct nfs_server *server = NFS_SERVER(inode); struct page *pages[NFS4ACL_MAXPAGES]; @@ -2694,16 +2775,28 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen return -EOPNOTSUPP; nfs_inode_return_delegation(inode); buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase); - ret = rpc_call_sync(NFS_SERVER(inode)->client, &msg, 0); + ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); if (ret == 0) nfs4_write_cached_acl(inode, buf, buflen); return ret; } +static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen) +{ + struct nfs4_exception exception = { }; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(inode), + __nfs4_proc_set_acl(inode, buf, buflen), + &exception); + } while (exception.retry); + return err; +} + static int nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server) { - struct nfs4_client *clp = server->nfs4_state; + struct nfs_client *clp = server->nfs_client; if (!clp || task->tk_status >= 0) return 0; @@ -2740,7 +2833,7 @@ static int nfs4_wait_bit_interruptible(void *word) return 0; } -static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs4_client *clp) +static int nfs4_wait_clnt_recover(struct rpc_clnt *clnt, struct nfs_client *clp) { sigset_t oldset; int res; @@ -2783,7 +2876,7 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout) */ int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception) { - struct nfs4_client *clp = server->nfs4_state; + struct nfs_client *clp = server->nfs_client; int ret = errorcode; exception->retry = 0; @@ -2798,6 +2891,7 @@ int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct if (ret == 0) exception->retry = 1; break; + case -NFS4ERR_FILE_OPEN: case -NFS4ERR_GRACE: case -NFS4ERR_DELAY: ret = nfs4_delay(server->client, &exception->timeout); @@ -2810,7 +2904,7 @@ int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct return nfs4_map_errors(ret); } -int nfs4_proc_setclientid(struct nfs4_client *clp, u32 program, unsigned short port, struct rpc_cred *cred) +int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short port, struct rpc_cred *cred) { nfs4_verifier sc_verifier; struct nfs4_setclientid setclientid = { @@ -2834,7 +2928,7 @@ int nfs4_proc_setclientid(struct nfs4_client *clp, u32 program, unsigned short p for(;;) { setclientid.sc_name_len = scnprintf(setclientid.sc_name, sizeof(setclientid.sc_name), "%s/%u.%u.%u.%u %s %u", - clp->cl_ipaddr, NIPQUAD(clp->cl_addr.s_addr), + clp->cl_ipaddr, NIPQUAD(clp->cl_addr.sin_addr), cred->cr_ops->cr_name, clp->cl_id_uniquifier); setclientid.sc_netid_len = scnprintf(setclientid.sc_netid, @@ -2857,7 +2951,7 @@ int nfs4_proc_setclientid(struct nfs4_client *clp, u32 program, unsigned short p return status; } -static int _nfs4_proc_setclientid_confirm(struct nfs4_client *clp, struct rpc_cred *cred) +static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cred *cred) { struct nfs_fsinfo fsinfo; struct rpc_message msg = { @@ -2881,7 +2975,7 @@ static int _nfs4_proc_setclientid_confirm(struct nfs4_client *clp, struct rpc_cr return status; } -int nfs4_proc_setclientid_confirm(struct nfs4_client *clp, struct rpc_cred *cred) +int nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct rpc_cred *cred) { long timeout; int err; @@ -2989,7 +3083,7 @@ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4 switch (err) { case -NFS4ERR_STALE_STATEID: case -NFS4ERR_EXPIRED: - nfs4_schedule_state_recovery(server->nfs4_state); + nfs4_schedule_state_recovery(server->nfs_client); case 0: return 0; } @@ -3018,7 +3112,7 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock { struct inode *inode = state->inode; struct nfs_server *server = NFS_SERVER(inode); - struct nfs4_client *clp = server->nfs4_state; + struct nfs_client *clp = server->nfs_client; struct nfs_lockt_args arg = { .fh = NFS_FH(inode), .fl = request, @@ -3081,9 +3175,6 @@ static int do_vfs_lock(struct file *file, struct file_lock *fl) default: BUG(); } - if (res < 0) - printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", - __FUNCTION__); return res; } @@ -3146,7 +3237,7 @@ static void nfs4_locku_done(struct rpc_task *task, void *data) break; case -NFS4ERR_STALE_STATEID: case -NFS4ERR_EXPIRED: - nfs4_schedule_state_recovery(calldata->server->nfs4_state); + nfs4_schedule_state_recovery(calldata->server->nfs_client); break; default: if (nfs4_async_handle_error(task, calldata->server) == -EAGAIN) { @@ -3195,8 +3286,6 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl, return ERR_PTR(-ENOMEM); } - /* Unlock _before_ we do the RPC call */ - do_vfs_lock(fl->fl_file, fl); return rpc_run_task(NFS_CLIENT(lsp->ls_state->inode), RPC_TASK_ASYNC, &nfs4_locku_ops, data); } @@ -3207,30 +3296,28 @@ static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock * struct rpc_task *task; int status = 0; - /* Is this a delegated lock? */ - if (test_bit(NFS_DELEGATED_STATE, &state->flags)) - goto out_unlock; - /* Is this open_owner holding any locks on the server? */ - if (test_bit(LK_STATE_IN_USE, &state->flags) == 0) - goto out_unlock; - status = nfs4_set_lock_state(state, request); + /* Unlock _before_ we do the RPC call */ + request->fl_flags |= FL_EXISTS; + if (do_vfs_lock(request->fl_file, request) == -ENOENT) + goto out; if (status != 0) - goto out_unlock; + goto out; + /* Is this a delegated lock? */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) + goto out; lsp = request->fl_u.nfs4_fl.owner; - status = -ENOMEM; seqid = nfs_alloc_seqid(&lsp->ls_seqid); + status = -ENOMEM; if (seqid == NULL) - goto out_unlock; + goto out; task = nfs4_do_unlck(request, request->fl_file->private_data, lsp, seqid); status = PTR_ERR(task); if (IS_ERR(task)) - goto out_unlock; + goto out; status = nfs4_wait_for_completion_rpc_task(task); rpc_release_task(task); - return status; -out_unlock: - do_vfs_lock(request->fl_file, request); +out: return status; } @@ -3262,7 +3349,7 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl, if (p->arg.lock_seqid == NULL) goto out_free; p->arg.lock_stateid = &lsp->ls_stateid; - p->arg.lock_owner.clientid = server->nfs4_state->cl_clientid; + p->arg.lock_owner.clientid = server->nfs_client->cl_clientid; p->arg.lock_owner.id = lsp->ls_id; p->lsp = lsp; atomic_inc(&lsp->ls_count); @@ -3398,10 +3485,10 @@ static int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request struct nfs4_exception exception = { }; int err; - /* Cache the lock if possible... */ - if (test_bit(NFS_DELEGATED_STATE, &state->flags)) - return 0; do { + /* Cache the lock if possible... */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) + return 0; err = _nfs4_do_setlk(state, F_SETLK, request, 1); if (err != -NFS4ERR_DELAY) break; @@ -3420,6 +3507,8 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request if (err != 0) return err; do { + if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0) + return 0; err = _nfs4_do_setlk(state, F_SETLK, request, 0); if (err != -NFS4ERR_DELAY) break; @@ -3430,30 +3519,43 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request) { - struct nfs4_client *clp = state->owner->so_client; + struct nfs_client *clp = state->owner->so_client; + unsigned char fl_flags = request->fl_flags; int status; /* Is this a delegated open? */ - if (NFS_I(state->inode)->delegation_state != 0) { - /* Yes: cache locks! */ - status = do_vfs_lock(request->fl_file, request); - /* ...but avoid races with delegation recall... */ - if (status < 0 || test_bit(NFS_DELEGATED_STATE, &state->flags)) - return status; - } - down_read(&clp->cl_sem); status = nfs4_set_lock_state(state, request); if (status != 0) goto out; + request->fl_flags |= FL_ACCESS; + status = do_vfs_lock(request->fl_file, request); + if (status < 0) + goto out; + down_read(&clp->cl_sem); + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) { + struct nfs_inode *nfsi = NFS_I(state->inode); + /* Yes: cache locks! */ + down_read(&nfsi->rwsem); + /* ...but avoid races with delegation recall... */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) { + request->fl_flags = fl_flags & ~FL_SLEEP; + status = do_vfs_lock(request->fl_file, request); + up_read(&nfsi->rwsem); + goto out_unlock; + } + up_read(&nfsi->rwsem); + } status = _nfs4_do_setlk(state, cmd, request, 0); if (status != 0) - goto out; + goto out_unlock; /* Note: we always want to sleep here! */ - request->fl_flags |= FL_SLEEP; + request->fl_flags = fl_flags | FL_SLEEP; if (do_vfs_lock(request->fl_file, request) < 0) printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", __FUNCTION__); -out: +out_unlock: up_read(&clp->cl_sem); +out: + request->fl_flags = fl_flags; return status; } @@ -3570,6 +3672,36 @@ ssize_t nfs4_listxattr(struct dentry *dentry, char *buf, size_t buflen) return len; } +int nfs4_proc_fs_locations(struct inode *dir, struct dentry *dentry, + struct nfs4_fs_locations *fs_locations, struct page *page) +{ + struct nfs_server *server = NFS_SERVER(dir); + u32 bitmask[2] = { + [0] = FATTR4_WORD0_FSID | FATTR4_WORD0_FS_LOCATIONS, + [1] = FATTR4_WORD1_MOUNTED_ON_FILEID, + }; + struct nfs4_fs_locations_arg args = { + .dir_fh = NFS_FH(dir), + .name = &dentry->d_name, + .page = page, + .bitmask = bitmask, + }; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FS_LOCATIONS], + .rpc_argp = &args, + .rpc_resp = fs_locations, + }; + int status; + + dprintk("%s: start\n", __FUNCTION__); + fs_locations->fattr.valid = 0; + fs_locations->server = server; + fs_locations->nlocations = 0; + status = rpc_call_sync(server->client, &msg, 0); + dprintk("%s: returned status = %d\n", __FUNCTION__, status); + return status; +} + struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = { .recover_open = nfs4_open_reclaim, .recover_lock = nfs4_lock_reclaim, @@ -3589,7 +3721,7 @@ static struct inode_operations nfs4_file_inode_operations = { .listxattr = nfs4_listxattr, }; -struct nfs_rpc_ops nfs_v4_clientops = { +const struct nfs_rpc_ops nfs_v4_clientops = { .version = 4, /* protocol version */ .dentry_ops = &nfs4_dentry_operations, .dir_inode_ops = &nfs4_dir_inode_operations, @@ -3597,6 +3729,7 @@ struct nfs_rpc_ops nfs_v4_clientops = { .getroot = nfs4_proc_get_root, .getattr = nfs4_proc_getattr, .setattr = nfs4_proc_setattr, + .lookupfh = nfs4_proc_lookupfh, .lookup = nfs4_proc_lookup, .access = nfs4_proc_access, .readlink = nfs4_proc_readlink, @@ -3617,6 +3750,7 @@ struct nfs_rpc_ops nfs_v4_clientops = { .statfs = nfs4_proc_statfs, .fsinfo = nfs4_proc_fsinfo, .pathconf = nfs4_proc_pathconf, + .set_capabilities = nfs4_server_capabilities, .decode_dirent = nfs4_decode_dirent, .read_setup = nfs4_proc_read_setup, .read_done = nfs4_read_done, diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c index 5d764d8e6d8..7b6df1852e7 100644 --- a/fs/nfs/nfs4renewd.c +++ b/fs/nfs/nfs4renewd.c @@ -61,7 +61,7 @@ void nfs4_renew_state(void *data) { - struct nfs4_client *clp = (struct nfs4_client *)data; + struct nfs_client *clp = (struct nfs_client *)data; struct rpc_cred *cred; long lease, timeout; unsigned long last, now; @@ -108,7 +108,7 @@ out: /* Must be called with clp->cl_sem locked for writes */ void -nfs4_schedule_state_renewal(struct nfs4_client *clp) +nfs4_schedule_state_renewal(struct nfs_client *clp) { long timeout; @@ -121,32 +121,20 @@ nfs4_schedule_state_renewal(struct nfs4_client *clp) __FUNCTION__, (timeout + HZ - 1) / HZ); cancel_delayed_work(&clp->cl_renewd); schedule_delayed_work(&clp->cl_renewd, timeout); + set_bit(NFS_CS_RENEWD, &clp->cl_res_state); spin_unlock(&clp->cl_lock); } void nfs4_renewd_prepare_shutdown(struct nfs_server *server) { - struct nfs4_client *clp = server->nfs4_state; - - if (!clp) - return; flush_scheduled_work(); - down_write(&clp->cl_sem); - if (!list_empty(&server->nfs4_siblings)) - list_del_init(&server->nfs4_siblings); - up_write(&clp->cl_sem); } -/* Must be called with clp->cl_sem locked for writes */ void -nfs4_kill_renewd(struct nfs4_client *clp) +nfs4_kill_renewd(struct nfs_client *clp) { down_read(&clp->cl_sem); - if (!list_empty(&clp->cl_superblocks)) { - up_read(&clp->cl_sem); - return; - } cancel_delayed_work(&clp->cl_renewd); up_read(&clp->cl_sem); flush_scheduled_work(); diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 96e5b82c153..5fffbdfa971 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -38,7 +38,6 @@ * subsequent patch. */ -#include <linux/config.h> #include <linux/slab.h> #include <linux/smp_lock.h> #include <linux/nfs_fs.h> @@ -51,149 +50,15 @@ #include "nfs4_fs.h" #include "callback.h" #include "delegation.h" +#include "internal.h" #define OPENOWNER_POOL_SIZE 8 const nfs4_stateid zero_stateid; -static DEFINE_SPINLOCK(state_spinlock); static LIST_HEAD(nfs4_clientid_list); -void -init_nfsv4_state(struct nfs_server *server) -{ - server->nfs4_state = NULL; - INIT_LIST_HEAD(&server->nfs4_siblings); -} - -void -destroy_nfsv4_state(struct nfs_server *server) -{ - kfree(server->mnt_path); - server->mnt_path = NULL; - if (server->nfs4_state) { - nfs4_put_client(server->nfs4_state); - server->nfs4_state = NULL; - } -} - -/* - * nfs4_get_client(): returns an empty client structure - * nfs4_put_client(): drops reference to client structure - * - * Since these are allocated/deallocated very rarely, we don't - * bother putting them in a slab cache... - */ -static struct nfs4_client * -nfs4_alloc_client(struct in_addr *addr) -{ - struct nfs4_client *clp; - - if (nfs_callback_up() < 0) - return NULL; - if ((clp = kzalloc(sizeof(*clp), GFP_KERNEL)) == NULL) { - nfs_callback_down(); - return NULL; - } - memcpy(&clp->cl_addr, addr, sizeof(clp->cl_addr)); - init_rwsem(&clp->cl_sem); - INIT_LIST_HEAD(&clp->cl_delegations); - INIT_LIST_HEAD(&clp->cl_state_owners); - INIT_LIST_HEAD(&clp->cl_unused); - spin_lock_init(&clp->cl_lock); - atomic_set(&clp->cl_count, 1); - INIT_WORK(&clp->cl_renewd, nfs4_renew_state, clp); - INIT_LIST_HEAD(&clp->cl_superblocks); - rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS4 client"); - clp->cl_rpcclient = ERR_PTR(-EINVAL); - clp->cl_boot_time = CURRENT_TIME; - clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED; - return clp; -} - -static void -nfs4_free_client(struct nfs4_client *clp) -{ - struct nfs4_state_owner *sp; - - while (!list_empty(&clp->cl_unused)) { - sp = list_entry(clp->cl_unused.next, - struct nfs4_state_owner, - so_list); - list_del(&sp->so_list); - kfree(sp); - } - BUG_ON(!list_empty(&clp->cl_state_owners)); - nfs_idmap_delete(clp); - if (!IS_ERR(clp->cl_rpcclient)) - rpc_shutdown_client(clp->cl_rpcclient); - kfree(clp); - nfs_callback_down(); -} - -static struct nfs4_client *__nfs4_find_client(struct in_addr *addr) -{ - struct nfs4_client *clp; - list_for_each_entry(clp, &nfs4_clientid_list, cl_servers) { - if (memcmp(&clp->cl_addr, addr, sizeof(clp->cl_addr)) == 0) { - atomic_inc(&clp->cl_count); - return clp; - } - } - return NULL; -} - -struct nfs4_client *nfs4_find_client(struct in_addr *addr) -{ - struct nfs4_client *clp; - spin_lock(&state_spinlock); - clp = __nfs4_find_client(addr); - spin_unlock(&state_spinlock); - return clp; -} - -struct nfs4_client * -nfs4_get_client(struct in_addr *addr) -{ - struct nfs4_client *clp, *new = NULL; - - spin_lock(&state_spinlock); - for (;;) { - clp = __nfs4_find_client(addr); - if (clp != NULL) - break; - clp = new; - if (clp != NULL) { - list_add(&clp->cl_servers, &nfs4_clientid_list); - new = NULL; - break; - } - spin_unlock(&state_spinlock); - new = nfs4_alloc_client(addr); - spin_lock(&state_spinlock); - if (new == NULL) - break; - } - spin_unlock(&state_spinlock); - if (new) - nfs4_free_client(new); - return clp; -} - -void -nfs4_put_client(struct nfs4_client *clp) -{ - if (!atomic_dec_and_lock(&clp->cl_count, &state_spinlock)) - return; - list_del(&clp->cl_servers); - spin_unlock(&state_spinlock); - BUG_ON(!list_empty(&clp->cl_superblocks)); - rpc_wake_up(&clp->cl_rpcwaitq); - nfs4_kill_renewd(clp); - nfs4_free_client(clp); -} - -static int nfs4_init_client(struct nfs4_client *clp, struct rpc_cred *cred) +static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred) { int status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, nfs_callback_tcpport, cred); @@ -205,13 +70,13 @@ static int nfs4_init_client(struct nfs4_client *clp, struct rpc_cred *cred) } u32 -nfs4_alloc_lockowner_id(struct nfs4_client *clp) +nfs4_alloc_lockowner_id(struct nfs_client *clp) { return clp->cl_lockowner_id ++; } static struct nfs4_state_owner * -nfs4_client_grab_unused(struct nfs4_client *clp, struct rpc_cred *cred) +nfs4_client_grab_unused(struct nfs_client *clp, struct rpc_cred *cred) { struct nfs4_state_owner *sp = NULL; @@ -225,7 +90,7 @@ nfs4_client_grab_unused(struct nfs4_client *clp, struct rpc_cred *cred) return sp; } -struct rpc_cred *nfs4_get_renew_cred(struct nfs4_client *clp) +struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp) { struct nfs4_state_owner *sp; struct rpc_cred *cred = NULL; @@ -239,7 +104,7 @@ struct rpc_cred *nfs4_get_renew_cred(struct nfs4_client *clp) return cred; } -struct rpc_cred *nfs4_get_setclientid_cred(struct nfs4_client *clp) +struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp) { struct nfs4_state_owner *sp; @@ -252,7 +117,7 @@ struct rpc_cred *nfs4_get_setclientid_cred(struct nfs4_client *clp) } static struct nfs4_state_owner * -nfs4_find_state_owner(struct nfs4_client *clp, struct rpc_cred *cred) +nfs4_find_state_owner(struct nfs_client *clp, struct rpc_cred *cred) { struct nfs4_state_owner *sp, *res = NULL; @@ -295,7 +160,7 @@ nfs4_alloc_state_owner(void) void nfs4_drop_state_owner(struct nfs4_state_owner *sp) { - struct nfs4_client *clp = sp->so_client; + struct nfs_client *clp = sp->so_client; spin_lock(&clp->cl_lock); list_del_init(&sp->so_list); spin_unlock(&clp->cl_lock); @@ -307,7 +172,7 @@ nfs4_drop_state_owner(struct nfs4_state_owner *sp) */ struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, struct rpc_cred *cred) { - struct nfs4_client *clp = server->nfs4_state; + struct nfs_client *clp = server->nfs_client; struct nfs4_state_owner *sp, *new; get_rpccred(cred); @@ -338,7 +203,7 @@ struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, struct */ void nfs4_put_state_owner(struct nfs4_state_owner *sp) { - struct nfs4_client *clp = sp->so_client; + struct nfs_client *clp = sp->so_client; struct rpc_cred *cred = sp->so_cred; if (!atomic_dec_and_lock(&sp->so_count, &clp->cl_lock)) @@ -541,7 +406,7 @@ __nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner) static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, fl_owner_t fl_owner) { struct nfs4_lock_state *lsp; - struct nfs4_client *clp = state->owner->so_client; + struct nfs_client *clp = state->owner->so_client; lsp = kzalloc(sizeof(*lsp), GFP_KERNEL); if (lsp == NULL) @@ -753,7 +618,7 @@ out: static int reclaimer(void *); -static inline void nfs4_clear_recover_bit(struct nfs4_client *clp) +static inline void nfs4_clear_recover_bit(struct nfs_client *clp) { smp_mb__before_clear_bit(); clear_bit(NFS4CLNT_STATE_RECOVER, &clp->cl_state); @@ -765,25 +630,25 @@ static inline void nfs4_clear_recover_bit(struct nfs4_client *clp) /* * State recovery routine */ -static void nfs4_recover_state(struct nfs4_client *clp) +static void nfs4_recover_state(struct nfs_client *clp) { struct task_struct *task; __module_get(THIS_MODULE); atomic_inc(&clp->cl_count); task = kthread_run(reclaimer, clp, "%u.%u.%u.%u-reclaim", - NIPQUAD(clp->cl_addr)); + NIPQUAD(clp->cl_addr.sin_addr)); if (!IS_ERR(task)) return; nfs4_clear_recover_bit(clp); - nfs4_put_client(clp); + nfs_put_client(clp); module_put(THIS_MODULE); } /* * Schedule a state recovery attempt */ -void nfs4_schedule_state_recovery(struct nfs4_client *clp) +void nfs4_schedule_state_recovery(struct nfs_client *clp) { if (!clp) return; @@ -880,7 +745,7 @@ out_err: return status; } -static void nfs4_state_mark_reclaim(struct nfs4_client *clp) +static void nfs4_state_mark_reclaim(struct nfs_client *clp) { struct nfs4_state_owner *sp; struct nfs4_state *state; @@ -904,7 +769,7 @@ static void nfs4_state_mark_reclaim(struct nfs4_client *clp) static int reclaimer(void *ptr) { - struct nfs4_client *clp = ptr; + struct nfs_client *clp = ptr; struct nfs4_state_owner *sp; struct nfs4_state_recovery_ops *ops; struct rpc_cred *cred; @@ -971,12 +836,12 @@ out: if (status == -NFS4ERR_CB_PATH_DOWN) nfs_handle_cb_pathdown(clp); nfs4_clear_recover_bit(clp); - nfs4_put_client(clp); + nfs_put_client(clp); module_put_and_exit(0); return 0; out_error: printk(KERN_WARNING "Error: state recovery failed on NFSv4 server %u.%u.%u.%u with error %d\n", - NIPQUAD(clp->cl_addr.s_addr), -status); + NIPQUAD(clp->cl_addr.sin_addr), -status); set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); goto out; } diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 7c5d70efe72..3dd413f52da 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -58,7 +58,7 @@ /* Mapping from NFS error code to "errno" error code. */ #define errno_NFSERR_IO EIO -static int nfs_stat_to_errno(int); +static int nfs4_stat_to_errno(int); /* NFSv4 COMPOUND tags are only wanted for debugging purposes */ #ifdef DEBUG @@ -128,7 +128,7 @@ static int nfs_stat_to_errno(int); #define decode_link_maxsz (op_decode_hdr_maxsz + 5) #define encode_symlink_maxsz (op_encode_hdr_maxsz + \ 1 + nfs4_name_maxsz + \ - nfs4_path_maxsz + \ + 1 + \ nfs4_fattr_maxsz) #define decode_symlink_maxsz (op_decode_hdr_maxsz + 8) #define encode_create_maxsz (op_encode_hdr_maxsz + \ @@ -411,6 +411,15 @@ static int nfs_stat_to_errno(int); #define NFS4_dec_setacl_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ op_decode_hdr_maxsz + nfs4_fattr_bitmap_maxsz) +#define NFS4_enc_fs_locations_sz \ + (compound_encode_hdr_maxsz + \ + encode_putfh_maxsz + \ + encode_getattr_maxsz) +#define NFS4_dec_fs_locations_sz \ + (compound_decode_hdr_maxsz + \ + decode_putfh_maxsz + \ + op_decode_hdr_maxsz + \ + nfs4_fattr_bitmap_maxsz) static struct { unsigned int mode; @@ -520,7 +529,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s if (iap->ia_valid & ATTR_MODE) len += 4; if (iap->ia_valid & ATTR_UID) { - owner_namelen = nfs_map_uid_to_name(server->nfs4_state, iap->ia_uid, owner_name); + owner_namelen = nfs_map_uid_to_name(server->nfs_client, iap->ia_uid, owner_name); if (owner_namelen < 0) { printk(KERN_WARNING "nfs: couldn't resolve uid %d to string\n", iap->ia_uid); @@ -532,7 +541,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s len += 4 + (XDR_QUADLEN(owner_namelen) << 2); } if (iap->ia_valid & ATTR_GID) { - owner_grouplen = nfs_map_gid_to_group(server->nfs4_state, iap->ia_gid, owner_group); + owner_grouplen = nfs_map_gid_to_group(server->nfs_client, iap->ia_gid, owner_group); if (owner_grouplen < 0) { printk(KERN_WARNING "nfs4: couldn't resolve gid %d to string\n", iap->ia_gid); @@ -664,9 +673,9 @@ static int encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *c switch (create->ftype) { case NF4LNK: - RESERVE_SPACE(4 + create->u.symlink->len); - WRITE32(create->u.symlink->len); - WRITEMEM(create->u.symlink->name, create->u.symlink->len); + RESERVE_SPACE(4); + WRITE32(create->u.symlink.len); + xdr_write_pages(xdr, create->u.symlink.pages, 0, create->u.symlink.len); break; case NF4BLK: case NF4CHR: @@ -722,6 +731,13 @@ static int encode_fsinfo(struct xdr_stream *xdr, const u32* bitmask) bitmask[1] & nfs4_fsinfo_bitmap[1]); } +static int encode_fs_locations(struct xdr_stream *xdr, const u32* bitmask) +{ + return encode_getattr_two(xdr, + bitmask[0] & nfs4_fs_locations_bitmap[0], + bitmask[1] & nfs4_fs_locations_bitmap[1]); +} + static int encode_getfh(struct xdr_stream *xdr) { uint32_t *p; @@ -1144,7 +1160,7 @@ static int encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, con return 0; } -static int encode_renew(struct xdr_stream *xdr, const struct nfs4_client *client_stateid) +static int encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_stateid) { uint32_t *p; @@ -1230,7 +1246,7 @@ static int encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclien return 0; } -static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs4_client *client_state) +static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_client *client_state) { uint32_t *p; @@ -1929,7 +1945,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, uint32_t *p, const str /* * a RENEW request */ -static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs4_client *clp) +static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1959,7 +1975,7 @@ static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, uint32_t *p, struct nf /* * a SETCLIENTID_CONFIRM request */ -static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs4_client *clp) +static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -2003,6 +2019,38 @@ out: } /* + * Encode FS_LOCATIONS request + */ +static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs_locations_arg *args) +{ + struct xdr_stream xdr; + struct compound_hdr hdr = { + .nops = 3, + }; + struct rpc_auth *auth = req->rq_task->tk_auth; + int replen; + int status; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + encode_compound_hdr(&xdr, &hdr); + if ((status = encode_putfh(&xdr, args->dir_fh)) != 0) + goto out; + if ((status = encode_lookup(&xdr, args->name)) != 0) + goto out; + if ((status = encode_fs_locations(&xdr, args->bitmask)) != 0) + goto out; + /* set up reply + * toplevel_status + OP_PUTFH + status + * + OP_LOOKUP + status + OP_GETATTR + status = 7 + */ + replen = (RPC_REPHDRSIZE + auth->au_rslack + 7) << 2; + xdr_inline_pages(&req->rq_rcv_buf, replen, &args->page, + 0, PAGE_SIZE); +out: + return status; +} + +/* * START OF "GENERIC" DECODE ROUTINES. * These may look a little ugly since they are imported from a "generic" * set of XDR encode/decode routines which are intended to be shared by @@ -2036,7 +2084,7 @@ out: } \ } while (0) -static int decode_opaque_inline(struct xdr_stream *xdr, uint32_t *len, char **string) +static int decode_opaque_inline(struct xdr_stream *xdr, unsigned int *len, char **string) { uint32_t *p; @@ -2079,15 +2127,15 @@ static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) } READ32(nfserr); if (nfserr != NFS_OK) - return -nfs_stat_to_errno(nfserr); + return -nfs4_stat_to_errno(nfserr); return 0; } /* Dummy routine */ -static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs4_client *clp) +static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs_client *clp) { uint32_t *p; - uint32_t strlen; + unsigned int strlen; char *str; READ_BUF(12); @@ -2217,7 +2265,7 @@ static int decode_attr_symlink_support(struct xdr_stream *xdr, uint32_t *bitmap, return 0; } -static int decode_attr_fsid(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_fsid *fsid) +static int decode_attr_fsid(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_fsid *fsid) { uint32_t *p; @@ -2285,6 +2333,22 @@ static int decode_attr_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t return 0; } +static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *fileid) +{ + uint32_t *p; + + *fileid = 0; + if (unlikely(bitmap[1] & (FATTR4_WORD1_MOUNTED_ON_FILEID - 1U))) + return -EIO; + if (likely(bitmap[1] & FATTR4_WORD1_MOUNTED_ON_FILEID)) { + READ_BUF(8); + READ64(*fileid); + bitmap[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID; + } + dprintk("%s: fileid=%Lu\n", __FUNCTION__, (unsigned long long)*fileid); + return 0; +} + static int decode_attr_files_avail(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { uint32_t *p; @@ -2336,6 +2400,116 @@ static int decode_attr_files_total(struct xdr_stream *xdr, uint32_t *bitmap, uin return status; } +static int decode_pathname(struct xdr_stream *xdr, struct nfs4_pathname *path) +{ + int n; + uint32_t *p; + int status = 0; + + READ_BUF(4); + READ32(n); + if (n < 0) + goto out_eio; + if (n == 0) + goto root_path; + dprintk("path "); + path->ncomponents = 0; + while (path->ncomponents < n) { + struct nfs4_string *component = &path->components[path->ncomponents]; + status = decode_opaque_inline(xdr, &component->len, &component->data); + if (unlikely(status != 0)) + goto out_eio; + if (path->ncomponents != n) + dprintk("/"); + dprintk("%s", component->data); + if (path->ncomponents < NFS4_PATHNAME_MAXCOMPONENTS) + path->ncomponents++; + else { + dprintk("cannot parse %d components in path\n", n); + goto out_eio; + } + } +out: + dprintk("\n"); + return status; +root_path: +/* a root pathname is sent as a zero component4 */ + path->ncomponents = 1; + path->components[0].len=0; + path->components[0].data=NULL; + dprintk("path /\n"); + goto out; +out_eio: + dprintk(" status %d", status); + status = -EIO; + goto out; +} + +static int decode_attr_fs_locations(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_fs_locations *res) +{ + int n; + uint32_t *p; + int status = -EIO; + + if (unlikely(bitmap[0] & (FATTR4_WORD0_FS_LOCATIONS -1U))) + goto out; + status = 0; + if (unlikely(!(bitmap[0] & FATTR4_WORD0_FS_LOCATIONS))) + goto out; + dprintk("%s: fsroot ", __FUNCTION__); + status = decode_pathname(xdr, &res->fs_path); + if (unlikely(status != 0)) + goto out; + READ_BUF(4); + READ32(n); + if (n <= 0) + goto out_eio; + res->nlocations = 0; + while (res->nlocations < n) { + int m; + struct nfs4_fs_location *loc = &res->locations[res->nlocations]; + + READ_BUF(4); + READ32(m); + if (m <= 0) + goto out_eio; + + loc->nservers = 0; + dprintk("%s: servers ", __FUNCTION__); + while (loc->nservers < m) { + struct nfs4_string *server = &loc->servers[loc->nservers]; + status = decode_opaque_inline(xdr, &server->len, &server->data); + if (unlikely(status != 0)) + goto out_eio; + dprintk("%s ", server->data); + if (loc->nservers < NFS4_FS_LOCATION_MAXSERVERS) + loc->nservers++; + else { + int i; + dprintk("%s: using first %d of %d servers returned for location %d\n", __FUNCTION__, NFS4_FS_LOCATION_MAXSERVERS, m, res->nlocations); + for (i = loc->nservers; i < m; i++) { + int len; + char *data; + status = decode_opaque_inline(xdr, &len, &data); + if (unlikely(status != 0)) + goto out_eio; + } + } + } + status = decode_pathname(xdr, &loc->rootpath); + if (unlikely(status != 0)) + goto out_eio; + if (res->nlocations < NFS4_FS_LOCATIONS_MAXENTRIES) + res->nlocations++; + } +out: + dprintk("%s: fs_locations done, error = %d\n", __FUNCTION__, status); + return status; +out_eio: + status = -EIO; + goto out; +} + static int decode_attr_maxfilesize(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { uint32_t *p; @@ -2462,7 +2636,7 @@ static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t return 0; } -static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_client *clp, int32_t *uid) +static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *uid) { uint32_t len, *p; @@ -2486,7 +2660,7 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nf return 0; } -static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_client *clp, int32_t *gid) +static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *gid) { uint32_t len, *p; @@ -2841,6 +3015,7 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons bitmap[2] = {0}, type; int status, fmode = 0; + uint64_t fileid; if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) goto xdr_error; @@ -2863,18 +3038,22 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons goto xdr_error; if ((status = decode_attr_size(xdr, bitmap, &fattr->size)) != 0) goto xdr_error; - if ((status = decode_attr_fsid(xdr, bitmap, &fattr->fsid_u.nfs4)) != 0) + if ((status = decode_attr_fsid(xdr, bitmap, &fattr->fsid)) != 0) goto xdr_error; if ((status = decode_attr_fileid(xdr, bitmap, &fattr->fileid)) != 0) goto xdr_error; + if ((status = decode_attr_fs_locations(xdr, bitmap, container_of(fattr, + struct nfs4_fs_locations, + fattr))) != 0) + goto xdr_error; if ((status = decode_attr_mode(xdr, bitmap, &fattr->mode)) != 0) goto xdr_error; fattr->mode |= fmode; if ((status = decode_attr_nlink(xdr, bitmap, &fattr->nlink)) != 0) goto xdr_error; - if ((status = decode_attr_owner(xdr, bitmap, server->nfs4_state, &fattr->uid)) != 0) + if ((status = decode_attr_owner(xdr, bitmap, server->nfs_client, &fattr->uid)) != 0) goto xdr_error; - if ((status = decode_attr_group(xdr, bitmap, server->nfs4_state, &fattr->gid)) != 0) + if ((status = decode_attr_group(xdr, bitmap, server->nfs_client, &fattr->gid)) != 0) goto xdr_error; if ((status = decode_attr_rdev(xdr, bitmap, &fattr->rdev)) != 0) goto xdr_error; @@ -2886,6 +3065,10 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons goto xdr_error; if ((status = decode_attr_time_modify(xdr, bitmap, &fattr->mtime)) != 0) goto xdr_error; + if ((status = decode_attr_mounted_on_fileid(xdr, bitmap, &fileid)) != 0) + goto xdr_error; + if (fattr->fileid == 0 && fileid != 0) + fattr->fileid = fileid; if ((status = verify_attr_len(xdr, savep, attrlen)) == 0) fattr->valid = NFS_ATTR_FATTR | NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4; xdr_error: @@ -3071,7 +3254,7 @@ static int decode_delegation(struct xdr_stream *xdr, struct nfs_openres *res) if (decode_space_limit(xdr, &res->maxsize) < 0) return -EIO; } - return decode_ace(xdr, NULL, res->server->nfs4_state); + return decode_ace(xdr, NULL, res->server->nfs_client); } static int decode_open(struct xdr_stream *xdr, struct nfs_openres *res) @@ -3172,7 +3355,7 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n struct kvec *iov = rcvbuf->head; unsigned int nr, pglen = rcvbuf->page_len; uint32_t *end, *entry, *p, *kaddr; - uint32_t len, attrlen; + uint32_t len, attrlen, xlen; int hdrlen, recvd, status; status = decode_op_hdr(xdr, OP_READDIR); @@ -3194,10 +3377,10 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n BUG_ON(pglen + readdir->pgbase > PAGE_CACHE_SIZE); kaddr = p = (uint32_t *) kmap_atomic(page, KM_USER0); - end = (uint32_t *) ((char *)p + pglen + readdir->pgbase); + end = p + ((pglen + readdir->pgbase) >> 2); entry = p; for (nr = 0; *p++; nr++) { - if (p + 3 > end) + if (end - p < 3) goto short_pkt; dprintk("cookie = %Lu, ", *((unsigned long long *)p)); p += 2; /* cookie */ @@ -3206,18 +3389,19 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n printk(KERN_WARNING "NFS: giant filename in readdir (len 0x%x)\n", len); goto err_unmap; } - dprintk("filename = %*s\n", len, (char *)p); - p += XDR_QUADLEN(len); - if (p + 1 > end) + xlen = XDR_QUADLEN(len); + if (end - p < xlen + 1) goto short_pkt; + dprintk("filename = %*s\n", len, (char *)p); + p += xlen; len = ntohl(*p++); /* bitmap length */ - p += len; - if (p + 1 > end) + if (end - p < len + 1) goto short_pkt; + p += len; attrlen = XDR_QUADLEN(ntohl(*p++)); - p += attrlen; /* attributes */ - if (p + 2 > end) + if (end - p < attrlen + 2) goto short_pkt; + p += attrlen; /* attributes */ entry = p; } if (!nr && (entry[0] != 0 || entry[1] == 0)) @@ -3350,8 +3534,7 @@ static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, attrlen, recvd); return -EINVAL; } - if (attrlen <= *acl_len) - xdr_read_pages(xdr, attrlen); + xdr_read_pages(xdr, attrlen); *acl_len = attrlen; } else status = -EOPNOTSUPP; @@ -3382,7 +3565,7 @@ static int decode_setattr(struct xdr_stream *xdr, struct nfs_setattrres *res) return 0; } -static int decode_setclientid(struct xdr_stream *xdr, struct nfs4_client *clp) +static int decode_setclientid(struct xdr_stream *xdr, struct nfs_client *clp) { uint32_t *p; uint32_t opnum; @@ -3415,7 +3598,7 @@ static int decode_setclientid(struct xdr_stream *xdr, struct nfs4_client *clp) READ_BUF(len); return -NFSERR_CLID_INUSE; } else - return -nfs_stat_to_errno(nfserr); + return -nfs4_stat_to_errno(nfserr); return 0; } @@ -4073,7 +4256,7 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs_fsi if (!status) status = decode_fsinfo(&xdr, fsinfo); if (!status) - status = -nfs_stat_to_errno(hdr.status); + status = -nfs4_stat_to_errno(hdr.status); return status; } @@ -4152,7 +4335,7 @@ static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, uint32_t *p, void *dummy) * a SETCLIENTID request */ static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, uint32_t *p, - struct nfs4_client *clp) + struct nfs_client *clp) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4163,7 +4346,7 @@ static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, uint32_t *p, if (!status) status = decode_setclientid(&xdr, clp); if (!status) - status = -nfs_stat_to_errno(hdr.status); + status = -nfs4_stat_to_errno(hdr.status); return status; } @@ -4185,7 +4368,7 @@ static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, s if (!status) status = decode_fsinfo(&xdr, fsinfo); if (!status) - status = -nfs_stat_to_errno(hdr.status); + status = -nfs4_stat_to_errno(hdr.status); return status; } @@ -4211,6 +4394,29 @@ out: return status; } +/* + * FS_LOCATIONS request + */ +static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs_locations *res) +{ + struct xdr_stream xdr; + struct compound_hdr hdr; + int status; + + xdr_init_decode(&xdr, &req->rq_rcv_buf, p); + status = decode_compound_hdr(&xdr, &hdr); + if (status != 0) + goto out; + if ((status = decode_putfh(&xdr)) != 0) + goto out; + if ((status = decode_lookup(&xdr)) != 0) + goto out; + xdr_enter_page(&xdr, PAGE_SIZE); + status = decode_getfattr(&xdr, &res->fattr, res->server); +out: + return status; +} + uint32_t *nfs4_decode_dirent(uint32_t *p, struct nfs_entry *entry, int plus) { uint32_t bitmap[2] = {0}; @@ -4315,7 +4521,7 @@ static struct { * This one is used jointly by NFSv2 and NFSv3. */ static int -nfs_stat_to_errno(int stat) +nfs4_stat_to_errno(int stat) { int i; for (i = 0; nfs_errtbl[i].stat != -1; i++) { @@ -4382,6 +4588,7 @@ struct rpc_procinfo nfs4_procedures[] = { PROC(DELEGRETURN, enc_delegreturn, dec_delegreturn), PROC(GETACL, enc_getacl, dec_getacl), PROC(SETACL, enc_setacl, dec_setacl), + PROC(FS_LOCATIONS, enc_fs_locations, dec_fs_locations), }; struct rpc_version nfs_version4 = { diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 106aca388eb..829af323f28 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -9,7 +9,6 @@ * */ -#include <linux/config.h> #include <linux/slab.h> #include <linux/file.h> #include <linux/sunrpc/clnt.h> @@ -315,6 +314,7 @@ nfs_scan_lock_dirty(struct nfs_inode *nfsi, struct list_head *dst, req->wb_index, NFS_PAGE_TAG_DIRTY); nfs_list_remove_request(req); nfs_list_add_request(req, dst); + dec_zone_page_state(req->wb_page, NR_FILE_DIRTY); res++; } } @@ -325,6 +325,7 @@ out: /** * nfs_scan_list - Scan a list for matching requests + * @nfsi: NFS inode * @head: One of the NFS inode request lists * @dst: Destination list * @idx_start: lower bound of page->index to scan @@ -336,14 +337,15 @@ out: * The requests are *not* checked to ensure that they form a contiguous set. * You must be holding the inode's req_lock when calling this function */ -int -nfs_scan_list(struct list_head *head, struct list_head *dst, - unsigned long idx_start, unsigned int npages) +int nfs_scan_list(struct nfs_inode *nfsi, struct list_head *head, + struct list_head *dst, unsigned long idx_start, + unsigned int npages) { - struct list_head *pos, *tmp; - struct nfs_page *req; - unsigned long idx_end; - int res; + struct nfs_page *pgvec[NFS_SCAN_MAXENTRIES]; + struct nfs_page *req; + unsigned long idx_end; + int found, i; + int res; res = 0; if (npages == 0) @@ -351,25 +353,32 @@ nfs_scan_list(struct list_head *head, struct list_head *dst, else idx_end = idx_start + npages - 1; - list_for_each_safe(pos, tmp, head) { - - req = nfs_list_entry(pos); - - if (req->wb_index < idx_start) - continue; - if (req->wb_index > idx_end) + for (;;) { + found = radix_tree_gang_lookup(&nfsi->nfs_page_tree, + (void **)&pgvec[0], idx_start, + NFS_SCAN_MAXENTRIES); + if (found <= 0) break; + for (i = 0; i < found; i++) { + req = pgvec[i]; + if (req->wb_index > idx_end) + goto out; + idx_start = req->wb_index + 1; + if (req->wb_list_head != head) + continue; + if (nfs_set_page_writeback_locked(req)) { + nfs_list_remove_request(req); + nfs_list_add_request(req, dst); + res++; + } + } - if (!nfs_set_page_writeback_locked(req)) - continue; - nfs_list_remove_request(req); - nfs_list_add_request(req, dst); - res++; } +out: return res; } -int nfs_init_nfspagecache(void) +int __init nfs_init_nfspagecache(void) { nfs_page_cachep = kmem_cache_create("nfs_page", sizeof(struct nfs_page), @@ -383,7 +392,6 @@ int nfs_init_nfspagecache(void) void nfs_destroy_nfspagecache(void) { - if (kmem_cache_destroy(nfs_page_cachep)) - printk(KERN_INFO "nfs_page: not all structures were freed\n"); + kmem_cache_destroy(nfs_page_cachep); } diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index 9dd85cac2df..4529cc4f3f8 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -44,11 +44,10 @@ #include <linux/nfs_page.h> #include <linux/lockd/bind.h> #include <linux/smp_lock.h> +#include "internal.h" #define NFSDBG_FACILITY NFSDBG_PROC -extern struct rpc_procinfo nfs_procedures[]; - /* * Bare-bones access to getattr: this is for nfs_read_super. */ @@ -67,14 +66,14 @@ nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, dprintk("%s: call getattr\n", __FUNCTION__); nfs_fattr_init(fattr); - status = rpc_call_sync(server->client_sys, &msg, 0); + status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0); dprintk("%s: reply getattr: %d\n", __FUNCTION__, status); if (status) return status; dprintk("%s: call statfs\n", __FUNCTION__); msg.rpc_proc = &nfs_procedures[NFSPROC_STATFS]; msg.rpc_resp = &fsinfo; - status = rpc_call_sync(server->client_sys, &msg, 0); + status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0); dprintk("%s: reply statfs: %d\n", __FUNCTION__, status); if (status) return status; @@ -353,7 +352,7 @@ nfs_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, struct qstr * { struct nfs_diropargs *arg; - arg = (struct nfs_diropargs *)kmalloc(sizeof(*arg), GFP_KERNEL); + arg = kmalloc(sizeof(*arg), GFP_KERNEL); if (!arg) return -ENOMEM; arg->fh = NFS_FH(dir->d_inode); @@ -426,16 +425,17 @@ nfs_proc_link(struct inode *inode, struct inode *dir, struct qstr *name) } static int -nfs_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path, - struct iattr *sattr, struct nfs_fh *fhandle, - struct nfs_fattr *fattr) +nfs_proc_symlink(struct inode *dir, struct dentry *dentry, struct page *page, + unsigned int len, struct iattr *sattr) { + struct nfs_fh fhandle; + struct nfs_fattr fattr; struct nfs_symlinkargs arg = { .fromfh = NFS_FH(dir), - .fromname = name->name, - .fromlen = name->len, - .topath = path->name, - .tolen = path->len, + .fromname = dentry->d_name.name, + .fromlen = dentry->d_name.len, + .pages = &page, + .pathlen = len, .sattr = sattr }; struct rpc_message msg = { @@ -444,13 +444,25 @@ nfs_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path, }; int status; - if (path->len > NFS2_MAXPATHLEN) + if (len > NFS2_MAXPATHLEN) return -ENAMETOOLONG; - dprintk("NFS call symlink %s -> %s\n", name->name, path->name); - nfs_fattr_init(fattr); - fhandle->size = 0; + + dprintk("NFS call symlink %s\n", dentry->d_name.name); + status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); nfs_mark_for_revalidate(dir); + + /* + * V2 SYMLINK requests don't return any attributes. Setting the + * filehandle size to zero indicates to nfs_instantiate that it + * should fill in the data with a LOOKUP call on the wire. + */ + if (status == 0) { + nfs_fattr_init(&fattr); + fhandle.size = 0; + status = nfs_instantiate(dentry, &fhandle, &fattr); + } + dprintk("NFS reply symlink: %d\n", status); return status; } @@ -611,8 +623,6 @@ nfs_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, return 0; } -extern u32 * nfs_decode_dirent(u32 *, struct nfs_entry *, int); - static int nfs_read_done(struct rpc_task *task, struct nfs_read_data *data) { if (task->tk_status >= 0) { @@ -674,7 +684,7 @@ nfs_proc_lock(struct file *filp, int cmd, struct file_lock *fl) } -struct nfs_rpc_ops nfs_v2_clientops = { +const struct nfs_rpc_ops nfs_v2_clientops = { .version = 2, /* protocol version */ .dentry_ops = &nfs_dentry_operations, .dir_inode_ops = &nfs_dir_inode_operations, diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 624ca7146b6..c2e49c397a2 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -15,7 +15,6 @@ * within the RPC code when root squashing is suspected. */ -#include <linux/config.h> #include <linux/time.h> #include <linux/kernel.h> #include <linux/errno.h> @@ -44,21 +43,20 @@ static mempool_t *nfs_rdata_mempool; #define MIN_POOL_READ (32) -struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount) +struct nfs_read_data *nfs_readdata_alloc(size_t len) { + unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, SLAB_NOFS); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); - if (pagecount < NFS_PAGEVEC_SIZE) - p->pagevec = &p->page_array[0]; + p->npages = pagecount; + if (pagecount <= ARRAY_SIZE(p->page_array)) + p->pagevec = p->page_array; else { - size_t size = ++pagecount * sizeof(struct page *); - p->pagevec = kmalloc(size, GFP_NOFS); - if (p->pagevec) { - memset(p->pagevec, 0, size); - } else { + p->pagevec = kcalloc(pagecount, sizeof(struct page *), GFP_NOFS); + if (!p->pagevec) { mempool_free(p, nfs_rdata_mempool); p = NULL; } @@ -67,7 +65,7 @@ struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount) return p; } -void nfs_readdata_free(struct nfs_read_data *p) +static void nfs_readdata_free(struct nfs_read_data *p) { if (p && (p->pagevec != &p->page_array[0])) kfree(p->pagevec); @@ -104,6 +102,35 @@ int nfs_return_empty_page(struct page *page) return 0; } +static void nfs_readpage_truncate_uninitialised_page(struct nfs_read_data *data) +{ + unsigned int remainder = data->args.count - data->res.count; + unsigned int base = data->args.pgbase + data->res.count; + unsigned int pglen; + struct page **pages; + + if (data->res.eof == 0 || remainder == 0) + return; + /* + * Note: "remainder" can never be negative, since we check for + * this in the XDR code. + */ + pages = &data->args.pages[base >> PAGE_CACHE_SHIFT]; + base &= ~PAGE_CACHE_MASK; + pglen = PAGE_CACHE_SIZE - base; + for (;;) { + if (remainder <= pglen) { + memclear_highpage_flush(*pages, base, remainder); + break; + } + memclear_highpage_flush(*pages, base, pglen); + pages++; + remainder -= pglen; + pglen = PAGE_CACHE_SIZE; + base = 0; + } +} + /* * Read a page synchronously. */ @@ -115,7 +142,7 @@ static int nfs_readpage_sync(struct nfs_open_context *ctx, struct inode *inode, int result; struct nfs_read_data *rdata; - rdata = nfs_readdata_alloc(1); + rdata = nfs_readdata_alloc(count); if (!rdata) return -ENOMEM; @@ -144,7 +171,7 @@ static int nfs_readpage_sync(struct nfs_open_context *ctx, struct inode *inode, rdata->args.offset = page_offset(page) + rdata->args.pgbase; dprintk("NFS: nfs_proc_read(%s, (%s/%Ld), %Lu, %u)\n", - NFS_SERVER(inode)->hostname, + NFS_SERVER(inode)->nfs_client->cl_hostname, inode->i_sb->s_id, (long long)NFS_FILEID(inode), (unsigned long long)rdata->args.pgbase, @@ -177,11 +204,11 @@ static int nfs_readpage_sync(struct nfs_open_context *ctx, struct inode *inode, NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATIME; spin_unlock(&inode->i_lock); - if (count) - memclear_highpage_flush(page, rdata->args.pgbase, count); - SetPageUptodate(page); - if (PageError(page)) - ClearPageError(page); + if (rdata->res.eof || rdata->res.count == rdata->args.count) { + SetPageUptodate(page); + if (rdata->res.eof && count != 0) + memclear_highpage_flush(page, rdata->args.pgbase, count); + } result = 0; io_error: @@ -313,25 +340,25 @@ static int nfs_pagein_multi(struct list_head *head, struct inode *inode) struct nfs_page *req = nfs_list_entry(head->next); struct page *page = req->wb_page; struct nfs_read_data *data; - unsigned int rsize = NFS_SERVER(inode)->rsize; - unsigned int nbytes, offset; + size_t rsize = NFS_SERVER(inode)->rsize, nbytes; + unsigned int offset; int requests = 0; LIST_HEAD(list); nfs_list_remove_request(req); nbytes = req->wb_bytes; - for(;;) { - data = nfs_readdata_alloc(1); + do { + size_t len = min(nbytes,rsize); + + data = nfs_readdata_alloc(len); if (!data) goto out_bad; INIT_LIST_HEAD(&data->pages); list_add(&data->pages, &list); requests++; - if (nbytes <= rsize) - break; - nbytes -= rsize; - } + nbytes -= len; + } while(nbytes != 0); atomic_set(&req->wb_complete, requests); ClearPageError(page); @@ -379,7 +406,7 @@ static int nfs_pagein_one(struct list_head *head, struct inode *inode) if (NFS_SERVER(inode)->rsize < PAGE_CACHE_SIZE) return nfs_pagein_multi(head, inode); - data = nfs_readdata_alloc(NFS_SERVER(inode)->rpages); + data = nfs_readdata_alloc(NFS_SERVER(inode)->rsize); if (!data) goto out_bad; @@ -436,20 +463,12 @@ static void nfs_readpage_result_partial(struct rpc_task *task, void *calldata) struct nfs_page *req = data->req; struct page *page = req->wb_page; + if (likely(task->tk_status >= 0)) + nfs_readpage_truncate_uninitialised_page(data); + else + SetPageError(page); if (nfs_readpage_result(task, data) != 0) return; - if (task->tk_status >= 0) { - unsigned int request = data->args.count; - unsigned int result = data->res.count; - - if (result < request) { - memclear_highpage_flush(page, - data->args.pgbase + result, - request - result); - } - } else - SetPageError(page); - if (atomic_dec_and_test(&req->wb_complete)) { if (!PageError(page)) SetPageUptodate(page); @@ -462,6 +481,40 @@ static const struct rpc_call_ops nfs_read_partial_ops = { .rpc_release = nfs_readdata_release, }; +static void nfs_readpage_set_pages_uptodate(struct nfs_read_data *data) +{ + unsigned int count = data->res.count; + unsigned int base = data->args.pgbase; + struct page **pages; + + if (data->res.eof) + count = data->args.count; + if (unlikely(count == 0)) + return; + pages = &data->args.pages[base >> PAGE_CACHE_SHIFT]; + base &= ~PAGE_CACHE_MASK; + count += base; + for (;count >= PAGE_CACHE_SIZE; count -= PAGE_CACHE_SIZE, pages++) + SetPageUptodate(*pages); + if (count != 0) + SetPageUptodate(*pages); +} + +static void nfs_readpage_set_pages_error(struct nfs_read_data *data) +{ + unsigned int count = data->args.count; + unsigned int base = data->args.pgbase; + struct page **pages; + + pages = &data->args.pages[base >> PAGE_CACHE_SHIFT]; + base &= ~PAGE_CACHE_MASK; + count += base; + for (;count >= PAGE_CACHE_SIZE; count -= PAGE_CACHE_SIZE, pages++) + SetPageError(*pages); + if (count != 0) + SetPageError(*pages); +} + /* * This is the callback from RPC telling us whether a reply was * received or some error occurred (timeout or socket shutdown). @@ -469,27 +522,24 @@ static const struct rpc_call_ops nfs_read_partial_ops = { static void nfs_readpage_result_full(struct rpc_task *task, void *calldata) { struct nfs_read_data *data = calldata; - unsigned int count = data->res.count; + /* + * Note: nfs_readpage_result may change the values of + * data->args. In the multi-page case, we therefore need + * to ensure that we call the next nfs_readpage_set_page_uptodate() + * first in the multi-page case. + */ + if (likely(task->tk_status >= 0)) { + nfs_readpage_truncate_uninitialised_page(data); + nfs_readpage_set_pages_uptodate(data); + } else + nfs_readpage_set_pages_error(data); if (nfs_readpage_result(task, data) != 0) return; while (!list_empty(&data->pages)) { struct nfs_page *req = nfs_list_entry(data->pages.next); - struct page *page = req->wb_page; - nfs_list_remove_request(req); - if (task->tk_status >= 0) { - if (count < PAGE_CACHE_SIZE) { - if (count < req->wb_bytes) - memclear_highpage_flush(page, - req->wb_pgbase + count, - req->wb_bytes - count); - count = 0; - } else - count -= PAGE_CACHE_SIZE; - SetPageUptodate(page); - } else - SetPageError(page); + nfs_list_remove_request(req); nfs_readpage_release(req); } } @@ -518,8 +568,13 @@ int nfs_readpage_result(struct rpc_task *task, struct nfs_read_data *data) nfs_add_stats(data->inode, NFSIOS_SERVERREADBYTES, resp->count); - /* Is this a short read? */ - if (task->tk_status >= 0 && resp->count < argp->count && !resp->eof) { + if (task->tk_status < 0) { + if (task->tk_status == -ESTALE) { + set_bit(NFS_INO_STALE, &NFS_FLAGS(data->inode)); + nfs_mark_for_revalidate(data->inode); + } + } else if (resp->count < argp->count && !resp->eof) { + /* This is a short read! */ nfs_inc_stats(data->inode, NFSIOS_SHORTREAD); /* Has the server at least made some progress? */ if (resp->count != 0) { @@ -566,6 +621,10 @@ int nfs_readpage(struct file *file, struct page *page) if (error) goto out_error; + error = -ESTALE; + if (NFS_STALE(inode)) + goto out_error; + if (file == NULL) { ctx = nfs_find_open_context(inode, NULL, FMODE_READ); if (ctx == NULL) @@ -628,7 +687,7 @@ int nfs_readpages(struct file *filp, struct address_space *mapping, }; struct inode *inode = mapping->host; struct nfs_server *server = NFS_SERVER(inode); - int ret; + int ret = -ESTALE; dprintk("NFS: nfs_readpages (%s/%Ld %d)\n", inode->i_sb->s_id, @@ -636,6 +695,9 @@ int nfs_readpages(struct file *filp, struct address_space *mapping, nr_pages); nfs_inc_stats(inode, NFSIOS_VFSREADPAGES); + if (NFS_STALE(inode)) + goto out; + if (filp == NULL) { desc.ctx = nfs_find_open_context(inode, NULL, FMODE_READ); if (desc.ctx == NULL) @@ -651,10 +713,11 @@ int nfs_readpages(struct file *filp, struct address_space *mapping, ret = err; } put_nfs_open_context(desc.ctx); +out: return ret; } -int nfs_init_readpagecache(void) +int __init nfs_init_readpagecache(void) { nfs_rdata_cachep = kmem_cache_create("nfs_read_data", sizeof(struct nfs_read_data), @@ -674,6 +737,5 @@ int nfs_init_readpagecache(void) void nfs_destroy_readpagecache(void) { mempool_destroy(nfs_rdata_mempool); - if (kmem_cache_destroy(nfs_rdata_cachep)) - printk(KERN_INFO "nfs_read_data: not all structures were freed\n"); + kmem_cache_destroy(nfs_rdata_cachep); } diff --git a/fs/nfs/super.c b/fs/nfs/super.c new file mode 100644 index 00000000000..e8d40030cab --- /dev/null +++ b/fs/nfs/super.c @@ -0,0 +1,1074 @@ +/* + * linux/fs/nfs/super.c + * + * Copyright (C) 1992 Rick Sladkey + * + * nfs superblock handling functions + * + * Modularised by Alan Cox <Alan.Cox@linux.org>, while hacking some + * experimental NFS changes. Modularisation taken straight from SYS5 fs. + * + * Change to nfs_read_super() to permit NFS mounts to multi-homed hosts. + * J.S.Peatfield@damtp.cam.ac.uk + * + * Split from inode.c by David Howells <dhowells@redhat.com> + * + * - superblocks are indexed on server only - all inodes, dentries, etc. associated with a + * particular server are held in the same superblock + * - NFS superblocks can have several effective roots to the dentry tree + * - directory type roots are spliced into the tree when a path from one root reaches the root + * of another (see nfs_lookup()) + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/init.h> + +#include <linux/time.h> +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/string.h> +#include <linux/stat.h> +#include <linux/errno.h> +#include <linux/unistd.h> +#include <linux/sunrpc/clnt.h> +#include <linux/sunrpc/stats.h> +#include <linux/sunrpc/metrics.h> +#include <linux/nfs_fs.h> +#include <linux/nfs_mount.h> +#include <linux/nfs4_mount.h> +#include <linux/lockd/bind.h> +#include <linux/smp_lock.h> +#include <linux/seq_file.h> +#include <linux/mount.h> +#include <linux/nfs_idmap.h> +#include <linux/vfs.h> +#include <linux/inet.h> +#include <linux/nfs_xdr.h> + +#include <asm/system.h> +#include <asm/uaccess.h> + +#include "nfs4_fs.h" +#include "callback.h" +#include "delegation.h" +#include "iostat.h" +#include "internal.h" + +#define NFSDBG_FACILITY NFSDBG_VFS + +static void nfs_umount_begin(struct vfsmount *, int); +static int nfs_statfs(struct dentry *, struct kstatfs *); +static int nfs_show_options(struct seq_file *, struct vfsmount *); +static int nfs_show_stats(struct seq_file *, struct vfsmount *); +static int nfs_get_sb(struct file_system_type *, int, const char *, void *, struct vfsmount *); +static int nfs_xdev_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); +static void nfs_kill_super(struct super_block *); + +static struct file_system_type nfs_fs_type = { + .owner = THIS_MODULE, + .name = "nfs", + .get_sb = nfs_get_sb, + .kill_sb = nfs_kill_super, + .fs_flags = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, +}; + +struct file_system_type nfs_xdev_fs_type = { + .owner = THIS_MODULE, + .name = "nfs", + .get_sb = nfs_xdev_get_sb, + .kill_sb = nfs_kill_super, + .fs_flags = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, +}; + +static struct super_operations nfs_sops = { + .alloc_inode = nfs_alloc_inode, + .destroy_inode = nfs_destroy_inode, + .write_inode = nfs_write_inode, + .statfs = nfs_statfs, + .clear_inode = nfs_clear_inode, + .umount_begin = nfs_umount_begin, + .show_options = nfs_show_options, + .show_stats = nfs_show_stats, +}; + +#ifdef CONFIG_NFS_V4 +static int nfs4_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); +static int nfs4_xdev_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); +static int nfs4_referral_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); +static void nfs4_kill_super(struct super_block *sb); + +static struct file_system_type nfs4_fs_type = { + .owner = THIS_MODULE, + .name = "nfs4", + .get_sb = nfs4_get_sb, + .kill_sb = nfs4_kill_super, + .fs_flags = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, +}; + +struct file_system_type nfs4_xdev_fs_type = { + .owner = THIS_MODULE, + .name = "nfs4", + .get_sb = nfs4_xdev_get_sb, + .kill_sb = nfs4_kill_super, + .fs_flags = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, +}; + +struct file_system_type nfs4_referral_fs_type = { + .owner = THIS_MODULE, + .name = "nfs4", + .get_sb = nfs4_referral_get_sb, + .kill_sb = nfs4_kill_super, + .fs_flags = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, +}; + +static struct super_operations nfs4_sops = { + .alloc_inode = nfs_alloc_inode, + .destroy_inode = nfs_destroy_inode, + .write_inode = nfs_write_inode, + .statfs = nfs_statfs, + .clear_inode = nfs4_clear_inode, + .umount_begin = nfs_umount_begin, + .show_options = nfs_show_options, + .show_stats = nfs_show_stats, +}; +#endif + +static struct shrinker *acl_shrinker; + +/* + * Register the NFS filesystems + */ +int __init register_nfs_fs(void) +{ + int ret; + + ret = register_filesystem(&nfs_fs_type); + if (ret < 0) + goto error_0; + +#ifdef CONFIG_NFS_V4 + ret = nfs_register_sysctl(); + if (ret < 0) + goto error_1; + ret = register_filesystem(&nfs4_fs_type); + if (ret < 0) + goto error_2; +#endif + acl_shrinker = set_shrinker(DEFAULT_SEEKS, nfs_access_cache_shrinker); + return 0; + +#ifdef CONFIG_NFS_V4 +error_2: + nfs_unregister_sysctl(); +error_1: + unregister_filesystem(&nfs_fs_type); +#endif +error_0: + return ret; +} + +/* + * Unregister the NFS filesystems + */ +void __exit unregister_nfs_fs(void) +{ + if (acl_shrinker != NULL) + remove_shrinker(acl_shrinker); +#ifdef CONFIG_NFS_V4 + unregister_filesystem(&nfs4_fs_type); + nfs_unregister_sysctl(); +#endif + unregister_filesystem(&nfs_fs_type); +} + +/* + * Deliver file system statistics to userspace + */ +static int nfs_statfs(struct dentry *dentry, struct kstatfs *buf) +{ + struct nfs_server *server = NFS_SB(dentry->d_sb); + unsigned char blockbits; + unsigned long blockres; + struct nfs_fh *fh = NFS_FH(dentry->d_inode); + struct nfs_fattr fattr; + struct nfs_fsstat res = { + .fattr = &fattr, + }; + int error; + + lock_kernel(); + + error = server->nfs_client->rpc_ops->statfs(server, fh, &res); + buf->f_type = NFS_SUPER_MAGIC; + if (error < 0) + goto out_err; + + /* + * Current versions of glibc do not correctly handle the + * case where f_frsize != f_bsize. Eventually we want to + * report the value of wtmult in this field. + */ + buf->f_frsize = dentry->d_sb->s_blocksize; + + /* + * On most *nix systems, f_blocks, f_bfree, and f_bavail + * are reported in units of f_frsize. Linux hasn't had + * an f_frsize field in its statfs struct until recently, + * thus historically Linux's sys_statfs reports these + * fields in units of f_bsize. + */ + buf->f_bsize = dentry->d_sb->s_blocksize; + blockbits = dentry->d_sb->s_blocksize_bits; + blockres = (1 << blockbits) - 1; + buf->f_blocks = (res.tbytes + blockres) >> blockbits; + buf->f_bfree = (res.fbytes + blockres) >> blockbits; + buf->f_bavail = (res.abytes + blockres) >> blockbits; + + buf->f_files = res.tfiles; + buf->f_ffree = res.afiles; + + buf->f_namelen = server->namelen; + out: + unlock_kernel(); + return 0; + + out_err: + dprintk("%s: statfs error = %d\n", __FUNCTION__, -error); + buf->f_bsize = buf->f_blocks = buf->f_bfree = buf->f_bavail = -1; + goto out; + +} + +/* + * Map the security flavour number to a name + */ +static const char *nfs_pseudoflavour_to_name(rpc_authflavor_t flavour) +{ + static const struct { + rpc_authflavor_t flavour; + const char *str; + } sec_flavours[] = { + { RPC_AUTH_NULL, "null" }, + { RPC_AUTH_UNIX, "sys" }, + { RPC_AUTH_GSS_KRB5, "krb5" }, + { RPC_AUTH_GSS_KRB5I, "krb5i" }, + { RPC_AUTH_GSS_KRB5P, "krb5p" }, + { RPC_AUTH_GSS_LKEY, "lkey" }, + { RPC_AUTH_GSS_LKEYI, "lkeyi" }, + { RPC_AUTH_GSS_LKEYP, "lkeyp" }, + { RPC_AUTH_GSS_SPKM, "spkm" }, + { RPC_AUTH_GSS_SPKMI, "spkmi" }, + { RPC_AUTH_GSS_SPKMP, "spkmp" }, + { -1, "unknown" } + }; + int i; + + for (i=0; sec_flavours[i].flavour != -1; i++) { + if (sec_flavours[i].flavour == flavour) + break; + } + return sec_flavours[i].str; +} + +/* + * Describe the mount options in force on this server representation + */ +static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, int showdefaults) +{ + static const struct proc_nfs_info { + int flag; + const char *str; + const char *nostr; + } nfs_info[] = { + { NFS_MOUNT_SOFT, ",soft", ",hard" }, + { NFS_MOUNT_INTR, ",intr", "" }, + { NFS_MOUNT_NOCTO, ",nocto", "" }, + { NFS_MOUNT_NOAC, ",noac", "" }, + { NFS_MOUNT_NONLM, ",nolock", "" }, + { NFS_MOUNT_NOACL, ",noacl", "" }, + { 0, NULL, NULL } + }; + const struct proc_nfs_info *nfs_infop; + struct nfs_client *clp = nfss->nfs_client; + char buf[12]; + const char *proto; + + seq_printf(m, ",vers=%d", clp->rpc_ops->version); + seq_printf(m, ",rsize=%d", nfss->rsize); + seq_printf(m, ",wsize=%d", nfss->wsize); + if (nfss->acregmin != 3*HZ || showdefaults) + seq_printf(m, ",acregmin=%d", nfss->acregmin/HZ); + if (nfss->acregmax != 60*HZ || showdefaults) + seq_printf(m, ",acregmax=%d", nfss->acregmax/HZ); + if (nfss->acdirmin != 30*HZ || showdefaults) + seq_printf(m, ",acdirmin=%d", nfss->acdirmin/HZ); + if (nfss->acdirmax != 60*HZ || showdefaults) + seq_printf(m, ",acdirmax=%d", nfss->acdirmax/HZ); + for (nfs_infop = nfs_info; nfs_infop->flag; nfs_infop++) { + if (nfss->flags & nfs_infop->flag) + seq_puts(m, nfs_infop->str); + else + seq_puts(m, nfs_infop->nostr); + } + switch (nfss->client->cl_xprt->prot) { + case IPPROTO_TCP: + proto = "tcp"; + break; + case IPPROTO_UDP: + proto = "udp"; + break; + default: + snprintf(buf, sizeof(buf), "%u", nfss->client->cl_xprt->prot); + proto = buf; + } + seq_printf(m, ",proto=%s", proto); + seq_printf(m, ",timeo=%lu", 10U * clp->retrans_timeo / HZ); + seq_printf(m, ",retrans=%u", clp->retrans_count); + seq_printf(m, ",sec=%s", nfs_pseudoflavour_to_name(nfss->client->cl_auth->au_flavor)); +} + +/* + * Describe the mount options on this VFS mountpoint + */ +static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) +{ + struct nfs_server *nfss = NFS_SB(mnt->mnt_sb); + + nfs_show_mount_options(m, nfss, 0); + + seq_puts(m, ",addr="); + seq_escape(m, nfss->nfs_client->cl_hostname, " \t\n\\"); + + return 0; +} + +/* + * Present statistical information for this VFS mountpoint + */ +static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt) +{ + int i, cpu; + struct nfs_server *nfss = NFS_SB(mnt->mnt_sb); + struct rpc_auth *auth = nfss->client->cl_auth; + struct nfs_iostats totals = { }; + + seq_printf(m, "statvers=%s", NFS_IOSTAT_VERS); + + /* + * Display all mount option settings + */ + seq_printf(m, "\n\topts:\t"); + seq_puts(m, mnt->mnt_sb->s_flags & MS_RDONLY ? "ro" : "rw"); + seq_puts(m, mnt->mnt_sb->s_flags & MS_SYNCHRONOUS ? ",sync" : ""); + seq_puts(m, mnt->mnt_sb->s_flags & MS_NOATIME ? ",noatime" : ""); + seq_puts(m, mnt->mnt_sb->s_flags & MS_NODIRATIME ? ",nodiratime" : ""); + nfs_show_mount_options(m, nfss, 1); + + seq_printf(m, "\n\tage:\t%lu", (jiffies - nfss->mount_time) / HZ); + + seq_printf(m, "\n\tcaps:\t"); + seq_printf(m, "caps=0x%x", nfss->caps); + seq_printf(m, ",wtmult=%d", nfss->wtmult); + seq_printf(m, ",dtsize=%d", nfss->dtsize); + seq_printf(m, ",bsize=%d", nfss->bsize); + seq_printf(m, ",namelen=%d", nfss->namelen); + +#ifdef CONFIG_NFS_V4 + if (nfss->nfs_client->cl_nfsversion == 4) { + seq_printf(m, "\n\tnfsv4:\t"); + seq_printf(m, "bm0=0x%x", nfss->attr_bitmask[0]); + seq_printf(m, ",bm1=0x%x", nfss->attr_bitmask[1]); + seq_printf(m, ",acl=0x%x", nfss->acl_bitmask); + } +#endif + + /* + * Display security flavor in effect for this mount + */ + seq_printf(m, "\n\tsec:\tflavor=%d", auth->au_ops->au_flavor); + if (auth->au_flavor) + seq_printf(m, ",pseudoflavor=%d", auth->au_flavor); + + /* + * Display superblock I/O counters + */ + for_each_possible_cpu(cpu) { + struct nfs_iostats *stats; + + preempt_disable(); + stats = per_cpu_ptr(nfss->io_stats, cpu); + + for (i = 0; i < __NFSIOS_COUNTSMAX; i++) + totals.events[i] += stats->events[i]; + for (i = 0; i < __NFSIOS_BYTESMAX; i++) + totals.bytes[i] += stats->bytes[i]; + + preempt_enable(); + } + + seq_printf(m, "\n\tevents:\t"); + for (i = 0; i < __NFSIOS_COUNTSMAX; i++) + seq_printf(m, "%lu ", totals.events[i]); + seq_printf(m, "\n\tbytes:\t"); + for (i = 0; i < __NFSIOS_BYTESMAX; i++) + seq_printf(m, "%Lu ", totals.bytes[i]); + seq_printf(m, "\n"); + + rpc_print_iostats(m, nfss->client); + + return 0; +} + +/* + * Begin unmount by attempting to remove all automounted mountpoints we added + * in response to xdev traversals and referrals + */ +static void nfs_umount_begin(struct vfsmount *vfsmnt, int flags) +{ + shrink_submounts(vfsmnt, &nfs_automount_list); +} + +/* + * Validate the NFS2/NFS3 mount data + * - fills in the mount root filehandle + */ +static int nfs_validate_mount_data(struct nfs_mount_data *data, + struct nfs_fh *mntfh) +{ + if (data == NULL) { + dprintk("%s: missing data argument\n", __FUNCTION__); + return -EINVAL; + } + + if (data->version <= 0 || data->version > NFS_MOUNT_VERSION) { + dprintk("%s: bad mount version\n", __FUNCTION__); + return -EINVAL; + } + + switch (data->version) { + case 1: + data->namlen = 0; + case 2: + data->bsize = 0; + case 3: + if (data->flags & NFS_MOUNT_VER3) { + dprintk("%s: mount structure version %d does not support NFSv3\n", + __FUNCTION__, + data->version); + return -EINVAL; + } + data->root.size = NFS2_FHSIZE; + memcpy(data->root.data, data->old_root.data, NFS2_FHSIZE); + case 4: + if (data->flags & NFS_MOUNT_SECFLAVOUR) { + dprintk("%s: mount structure version %d does not support strong security\n", + __FUNCTION__, + data->version); + return -EINVAL; + } + case 5: + memset(data->context, 0, sizeof(data->context)); + } + + /* Set the pseudoflavor */ + if (!(data->flags & NFS_MOUNT_SECFLAVOUR)) + data->pseudoflavor = RPC_AUTH_UNIX; + +#ifndef CONFIG_NFS_V3 + /* If NFSv3 is not compiled in, return -EPROTONOSUPPORT */ + if (data->flags & NFS_MOUNT_VER3) { + dprintk("%s: NFSv3 not compiled into kernel\n", __FUNCTION__); + return -EPROTONOSUPPORT; + } +#endif /* CONFIG_NFS_V3 */ + + /* We now require that the mount process passes the remote address */ + if (data->addr.sin_addr.s_addr == INADDR_ANY) { + dprintk("%s: mount program didn't pass remote address!\n", + __FUNCTION__); + return -EINVAL; + } + + /* Prepare the root filehandle */ + if (data->flags & NFS_MOUNT_VER3) + mntfh->size = data->root.size; + else + mntfh->size = NFS2_FHSIZE; + + if (mntfh->size > sizeof(mntfh->data)) { + dprintk("%s: invalid root filehandle\n", __FUNCTION__); + return -EINVAL; + } + + memcpy(mntfh->data, data->root.data, mntfh->size); + if (mntfh->size < sizeof(mntfh->data)) + memset(mntfh->data + mntfh->size, 0, + sizeof(mntfh->data) - mntfh->size); + + return 0; +} + +/* + * Initialise the common bits of the superblock + */ +static inline void nfs_initialise_sb(struct super_block *sb) +{ + struct nfs_server *server = NFS_SB(sb); + + sb->s_magic = NFS_SUPER_MAGIC; + + /* We probably want something more informative here */ + snprintf(sb->s_id, sizeof(sb->s_id), + "%x:%x", MAJOR(sb->s_dev), MINOR(sb->s_dev)); + + if (sb->s_blocksize == 0) + sb->s_blocksize = nfs_block_bits(server->wsize, + &sb->s_blocksize_bits); + + if (server->flags & NFS_MOUNT_NOAC) + sb->s_flags |= MS_SYNCHRONOUS; + + nfs_super_set_maxbytes(sb, server->maxfilesize); +} + +/* + * Finish setting up an NFS2/3 superblock + */ +static void nfs_fill_super(struct super_block *sb, struct nfs_mount_data *data) +{ + struct nfs_server *server = NFS_SB(sb); + + sb->s_blocksize_bits = 0; + sb->s_blocksize = 0; + if (data->bsize) + sb->s_blocksize = nfs_block_size(data->bsize, &sb->s_blocksize_bits); + + if (server->flags & NFS_MOUNT_VER3) { + /* The VFS shouldn't apply the umask to mode bits. We will do + * so ourselves when necessary. + */ + sb->s_flags |= MS_POSIXACL; + sb->s_time_gran = 1; + } + + sb->s_op = &nfs_sops; + nfs_initialise_sb(sb); +} + +/* + * Finish setting up a cloned NFS2/3 superblock + */ +static void nfs_clone_super(struct super_block *sb, + const struct super_block *old_sb) +{ + struct nfs_server *server = NFS_SB(sb); + + sb->s_blocksize_bits = old_sb->s_blocksize_bits; + sb->s_blocksize = old_sb->s_blocksize; + sb->s_maxbytes = old_sb->s_maxbytes; + + if (server->flags & NFS_MOUNT_VER3) { + /* The VFS shouldn't apply the umask to mode bits. We will do + * so ourselves when necessary. + */ + sb->s_flags |= MS_POSIXACL; + sb->s_time_gran = 1; + } + + sb->s_op = old_sb->s_op; + nfs_initialise_sb(sb); +} + +static int nfs_set_super(struct super_block *s, void *_server) +{ + struct nfs_server *server = _server; + int ret; + + s->s_fs_info = server; + ret = set_anon_super(s, server); + if (ret == 0) + server->s_dev = s->s_dev; + return ret; +} + +static int nfs_compare_super(struct super_block *sb, void *data) +{ + struct nfs_server *server = data, *old = NFS_SB(sb); + + if (old->nfs_client != server->nfs_client) + return 0; + if (memcmp(&old->fsid, &server->fsid, sizeof(old->fsid)) != 0) + return 0; + return 1; +} + +static int nfs_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt) +{ + struct nfs_server *server = NULL; + struct super_block *s; + struct nfs_fh mntfh; + struct nfs_mount_data *data = raw_data; + struct dentry *mntroot; + int error; + + /* Validate the mount data */ + error = nfs_validate_mount_data(data, &mntfh); + if (error < 0) + return error; + + /* Get a volume representation */ + server = nfs_create_server(data, &mntfh); + if (IS_ERR(server)) { + error = PTR_ERR(server); + goto out_err_noserver; + } + + /* Get a superblock - note that we may end up sharing one that already exists */ + s = sget(fs_type, nfs_compare_super, nfs_set_super, server); + if (IS_ERR(s)) { + error = PTR_ERR(s); + goto out_err_nosb; + } + + if (s->s_fs_info != server) { + nfs_free_server(server); + server = NULL; + } + + if (!s->s_root) { + /* initial superblock/root creation */ + s->s_flags = flags; + nfs_fill_super(s, data); + } + + mntroot = nfs_get_root(s, &mntfh); + if (IS_ERR(mntroot)) { + error = PTR_ERR(mntroot); + goto error_splat_super; + } + + s->s_flags |= MS_ACTIVE; + mnt->mnt_sb = s; + mnt->mnt_root = mntroot; + return 0; + +out_err_nosb: + nfs_free_server(server); +out_err_noserver: + return error; + +error_splat_super: + up_write(&s->s_umount); + deactivate_super(s); + return error; +} + +/* + * Destroy an NFS2/3 superblock + */ +static void nfs_kill_super(struct super_block *s) +{ + struct nfs_server *server = NFS_SB(s); + + kill_anon_super(s); + nfs_free_server(server); +} + +/* + * Clone an NFS2/3 server record on xdev traversal (FSID-change) + */ +static int nfs_xdev_get_sb(struct file_system_type *fs_type, int flags, + const char *dev_name, void *raw_data, + struct vfsmount *mnt) +{ + struct nfs_clone_mount *data = raw_data; + struct super_block *s; + struct nfs_server *server; + struct dentry *mntroot; + int error; + + dprintk("--> nfs_xdev_get_sb()\n"); + + /* create a new volume representation */ + server = nfs_clone_server(NFS_SB(data->sb), data->fh, data->fattr); + if (IS_ERR(server)) { + error = PTR_ERR(server); + goto out_err_noserver; + } + + /* Get a superblock - note that we may end up sharing one that already exists */ + s = sget(&nfs_fs_type, nfs_compare_super, nfs_set_super, server); + if (IS_ERR(s)) { + error = PTR_ERR(s); + goto out_err_nosb; + } + + if (s->s_fs_info != server) { + nfs_free_server(server); + server = NULL; + } + + if (!s->s_root) { + /* initial superblock/root creation */ + s->s_flags = flags; + nfs_clone_super(s, data->sb); + } + + mntroot = nfs_get_root(s, data->fh); + if (IS_ERR(mntroot)) { + error = PTR_ERR(mntroot); + goto error_splat_super; + } + + s->s_flags |= MS_ACTIVE; + mnt->mnt_sb = s; + mnt->mnt_root = mntroot; + + dprintk("<-- nfs_xdev_get_sb() = 0\n"); + return 0; + +out_err_nosb: + nfs_free_server(server); +out_err_noserver: + dprintk("<-- nfs_xdev_get_sb() = %d [error]\n", error); + return error; + +error_splat_super: + up_write(&s->s_umount); + deactivate_super(s); + dprintk("<-- nfs_xdev_get_sb() = %d [splat]\n", error); + return error; +} + +#ifdef CONFIG_NFS_V4 + +/* + * Finish setting up a cloned NFS4 superblock + */ +static void nfs4_clone_super(struct super_block *sb, + const struct super_block *old_sb) +{ + sb->s_blocksize_bits = old_sb->s_blocksize_bits; + sb->s_blocksize = old_sb->s_blocksize; + sb->s_maxbytes = old_sb->s_maxbytes; + sb->s_time_gran = 1; + sb->s_op = old_sb->s_op; + nfs_initialise_sb(sb); +} + +/* + * Set up an NFS4 superblock + */ +static void nfs4_fill_super(struct super_block *sb) +{ + sb->s_time_gran = 1; + sb->s_op = &nfs4_sops; + nfs_initialise_sb(sb); +} + +static void *nfs_copy_user_string(char *dst, struct nfs_string *src, int maxlen) +{ + void *p = NULL; + + if (!src->len) + return ERR_PTR(-EINVAL); + if (src->len < maxlen) + maxlen = src->len; + if (dst == NULL) { + p = dst = kmalloc(maxlen + 1, GFP_KERNEL); + if (p == NULL) + return ERR_PTR(-ENOMEM); + } + if (copy_from_user(dst, src->data, maxlen)) { + kfree(p); + return ERR_PTR(-EFAULT); + } + dst[maxlen] = '\0'; + return dst; +} + +/* + * Get the superblock for an NFS4 mountpoint + */ +static int nfs4_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt) +{ + struct nfs4_mount_data *data = raw_data; + struct super_block *s; + struct nfs_server *server; + struct sockaddr_in addr; + rpc_authflavor_t authflavour; + struct nfs_fh mntfh; + struct dentry *mntroot; + char *mntpath = NULL, *hostname = NULL, ip_addr[16]; + void *p; + int error; + + if (data == NULL) { + dprintk("%s: missing data argument\n", __FUNCTION__); + return -EINVAL; + } + if (data->version <= 0 || data->version > NFS4_MOUNT_VERSION) { + dprintk("%s: bad mount version\n", __FUNCTION__); + return -EINVAL; + } + + /* We now require that the mount process passes the remote address */ + if (data->host_addrlen != sizeof(addr)) + return -EINVAL; + + if (copy_from_user(&addr, data->host_addr, sizeof(addr))) + return -EFAULT; + + if (addr.sin_family != AF_INET || + addr.sin_addr.s_addr == INADDR_ANY + ) { + dprintk("%s: mount program didn't pass remote IP address!\n", + __FUNCTION__); + return -EINVAL; + } + /* RFC3530: The default port for NFS is 2049 */ + if (addr.sin_port == 0) + addr.sin_port = NFS_PORT; + + /* Grab the authentication type */ + authflavour = RPC_AUTH_UNIX; + if (data->auth_flavourlen != 0) { + if (data->auth_flavourlen != 1) { + dprintk("%s: Invalid number of RPC auth flavours %d.\n", + __FUNCTION__, data->auth_flavourlen); + error = -EINVAL; + goto out_err_noserver; + } + + if (copy_from_user(&authflavour, data->auth_flavours, + sizeof(authflavour))) { + error = -EFAULT; + goto out_err_noserver; + } + } + + p = nfs_copy_user_string(NULL, &data->hostname, 256); + if (IS_ERR(p)) + goto out_err; + hostname = p; + + p = nfs_copy_user_string(NULL, &data->mnt_path, 1024); + if (IS_ERR(p)) + goto out_err; + mntpath = p; + + dprintk("MNTPATH: %s\n", mntpath); + + p = nfs_copy_user_string(ip_addr, &data->client_addr, + sizeof(ip_addr) - 1); + if (IS_ERR(p)) + goto out_err; + + /* Get a volume representation */ + server = nfs4_create_server(data, hostname, &addr, mntpath, ip_addr, + authflavour, &mntfh); + if (IS_ERR(server)) { + error = PTR_ERR(server); + goto out_err_noserver; + } + + /* Get a superblock - note that we may end up sharing one that already exists */ + s = sget(fs_type, nfs_compare_super, nfs_set_super, server); + if (IS_ERR(s)) { + error = PTR_ERR(s); + goto out_free; + } + + if (s->s_fs_info != server) { + nfs_free_server(server); + server = NULL; + } + + if (!s->s_root) { + /* initial superblock/root creation */ + s->s_flags = flags; + nfs4_fill_super(s); + } + + mntroot = nfs4_get_root(s, &mntfh); + if (IS_ERR(mntroot)) { + error = PTR_ERR(mntroot); + goto error_splat_super; + } + + s->s_flags |= MS_ACTIVE; + mnt->mnt_sb = s; + mnt->mnt_root = mntroot; + kfree(mntpath); + kfree(hostname); + return 0; + +out_err: + error = PTR_ERR(p); + goto out_err_noserver; + +out_free: + nfs_free_server(server); +out_err_noserver: + kfree(mntpath); + kfree(hostname); + return error; + +error_splat_super: + up_write(&s->s_umount); + deactivate_super(s); + goto out_err_noserver; +} + +static void nfs4_kill_super(struct super_block *sb) +{ + struct nfs_server *server = NFS_SB(sb); + + nfs_return_all_delegations(sb); + kill_anon_super(sb); + + nfs4_renewd_prepare_shutdown(server); + nfs_free_server(server); +} + +/* + * Clone an NFS4 server record on xdev traversal (FSID-change) + */ +static int nfs4_xdev_get_sb(struct file_system_type *fs_type, int flags, + const char *dev_name, void *raw_data, + struct vfsmount *mnt) +{ + struct nfs_clone_mount *data = raw_data; + struct super_block *s; + struct nfs_server *server; + struct dentry *mntroot; + int error; + + dprintk("--> nfs4_xdev_get_sb()\n"); + + /* create a new volume representation */ + server = nfs_clone_server(NFS_SB(data->sb), data->fh, data->fattr); + if (IS_ERR(server)) { + error = PTR_ERR(server); + goto out_err_noserver; + } + + /* Get a superblock - note that we may end up sharing one that already exists */ + s = sget(&nfs_fs_type, nfs_compare_super, nfs_set_super, server); + if (IS_ERR(s)) { + error = PTR_ERR(s); + goto out_err_nosb; + } + + if (s->s_fs_info != server) { + nfs_free_server(server); + server = NULL; + } + + if (!s->s_root) { + /* initial superblock/root creation */ + s->s_flags = flags; + nfs4_clone_super(s, data->sb); + } + + mntroot = nfs4_get_root(s, data->fh); + if (IS_ERR(mntroot)) { + error = PTR_ERR(mntroot); + goto error_splat_super; + } + + s->s_flags |= MS_ACTIVE; + mnt->mnt_sb = s; + mnt->mnt_root = mntroot; + + dprintk("<-- nfs4_xdev_get_sb() = 0\n"); + return 0; + +out_err_nosb: + nfs_free_server(server); +out_err_noserver: + dprintk("<-- nfs4_xdev_get_sb() = %d [error]\n", error); + return error; + +error_splat_super: + up_write(&s->s_umount); + deactivate_super(s); + dprintk("<-- nfs4_xdev_get_sb() = %d [splat]\n", error); + return error; +} + +/* + * Create an NFS4 server record on referral traversal + */ +static int nfs4_referral_get_sb(struct file_system_type *fs_type, int flags, + const char *dev_name, void *raw_data, + struct vfsmount *mnt) +{ + struct nfs_clone_mount *data = raw_data; + struct super_block *s; + struct nfs_server *server; + struct dentry *mntroot; + struct nfs_fh mntfh; + int error; + + dprintk("--> nfs4_referral_get_sb()\n"); + + /* create a new volume representation */ + server = nfs4_create_referral_server(data, &mntfh); + if (IS_ERR(server)) { + error = PTR_ERR(server); + goto out_err_noserver; + } + + /* Get a superblock - note that we may end up sharing one that already exists */ + s = sget(&nfs_fs_type, nfs_compare_super, nfs_set_super, server); + if (IS_ERR(s)) { + error = PTR_ERR(s); + goto out_err_nosb; + } + + if (s->s_fs_info != server) { + nfs_free_server(server); + server = NULL; + } + + if (!s->s_root) { + /* initial superblock/root creation */ + s->s_flags = flags; + nfs4_fill_super(s); + } + + mntroot = nfs4_get_root(s, data->fh); + if (IS_ERR(mntroot)) { + error = PTR_ERR(mntroot); + goto error_splat_super; + } + + s->s_flags |= MS_ACTIVE; + mnt->mnt_sb = s; + mnt->mnt_root = mntroot; + + dprintk("<-- nfs4_referral_get_sb() = 0\n"); + return 0; + +out_err_nosb: + nfs_free_server(server); +out_err_noserver: + dprintk("<-- nfs4_referral_get_sb() = %d [error]\n", error); + return error; + +error_splat_super: + up_write(&s->s_umount); + deactivate_super(s); + dprintk("<-- nfs4_referral_get_sb() = %d [splat]\n", error); + return error; +} + +#endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/symlink.c b/fs/nfs/symlink.c index 18dc95b0b64..600bbe630ab 100644 --- a/fs/nfs/symlink.c +++ b/fs/nfs/symlink.c @@ -52,7 +52,7 @@ static void *nfs_follow_link(struct dentry *dentry, struct nameidata *nd) { struct inode *inode = dentry->d_inode; struct page *page; - void *err = ERR_PTR(nfs_revalidate_inode(NFS_SERVER(inode), inode)); + void *err = ERR_PTR(nfs_revalidate_mapping(inode, inode->i_mapping)); if (err) goto read_failed; page = read_cache_page(&inode->i_data, 0, @@ -75,22 +75,13 @@ read_failed: return NULL; } -static void nfs_put_link(struct dentry *dentry, struct nameidata *nd, void *cookie) -{ - if (cookie) { - struct page *page = cookie; - kunmap(page); - page_cache_release(page); - } -} - /* * symlinks can't do much... */ struct inode_operations nfs_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = nfs_follow_link, - .put_link = nfs_put_link, + .put_link = page_put_link, .getattr = nfs_getattr, .setattr = nfs_setattr, }; diff --git a/fs/nfs/sysctl.c b/fs/nfs/sysctl.c index 4c486eb867c..2fe3403c240 100644 --- a/fs/nfs/sysctl.c +++ b/fs/nfs/sysctl.c @@ -3,7 +3,6 @@ * * Sysctl interface to NFS parameters */ -#include <linux/config.h> #include <linux/types.h> #include <linux/linkage.h> #include <linux/ctype.h> @@ -12,6 +11,7 @@ #include <linux/module.h> #include <linux/nfs4.h> #include <linux/nfs_idmap.h> +#include <linux/nfs_fs.h> #include "callback.h" @@ -46,6 +46,15 @@ static ctl_table nfs_cb_sysctls[] = { .strategy = &sysctl_jiffies, }, #endif + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nfs_mountpoint_timeout", + .data = &nfs_mountpoint_expiry_timeout, + .maxlen = sizeof(nfs_mountpoint_expiry_timeout), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + .strategy = &sysctl_jiffies, + }, { .ctl_name = 0 } }; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 4cfada2cc09..b674462793d 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -46,7 +46,6 @@ * Copyright (C) 1996, 1997, Olaf Kirch <okir@monad.swb.de> */ -#include <linux/config.h> #include <linux/types.h> #include <linux/slab.h> #include <linux/mm.h> @@ -91,23 +90,13 @@ static mempool_t *nfs_commit_mempool; static DECLARE_WAIT_QUEUE_HEAD(nfs_write_congestion); -struct nfs_write_data *nfs_commit_alloc(unsigned int pagecount) +struct nfs_write_data *nfs_commit_alloc(void) { struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, SLAB_NOFS); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); - if (pagecount < NFS_PAGEVEC_SIZE) - p->pagevec = &p->page_array[0]; - else { - size_t size = ++pagecount * sizeof(struct page *); - p->pagevec = kzalloc(size, GFP_NOFS); - if (!p->pagevec) { - mempool_free(p, nfs_commit_mempool); - p = NULL; - } - } } return p; } @@ -119,21 +108,20 @@ void nfs_commit_free(struct nfs_write_data *p) mempool_free(p, nfs_commit_mempool); } -struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount) +struct nfs_write_data *nfs_writedata_alloc(size_t len) { + unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; struct nfs_write_data *p = mempool_alloc(nfs_wdata_mempool, SLAB_NOFS); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); - if (pagecount < NFS_PAGEVEC_SIZE) - p->pagevec = &p->page_array[0]; + p->npages = pagecount; + if (pagecount <= ARRAY_SIZE(p->page_array)) + p->pagevec = p->page_array; else { - size_t size = ++pagecount * sizeof(struct page *); - p->pagevec = kmalloc(size, GFP_NOFS); - if (p->pagevec) { - memset(p->pagevec, 0, size); - } else { + p->pagevec = kcalloc(pagecount, sizeof(struct page *), GFP_NOFS); + if (!p->pagevec) { mempool_free(p, nfs_wdata_mempool); p = NULL; } @@ -142,7 +130,7 @@ struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount) return p; } -void nfs_writedata_free(struct nfs_write_data *p) +static void nfs_writedata_free(struct nfs_write_data *p) { if (p && (p->pagevec != &p->page_array[0])) kfree(p->pagevec); @@ -213,7 +201,7 @@ static int nfs_writepage_sync(struct nfs_open_context *ctx, struct inode *inode, int result, written = 0; struct nfs_write_data *wdata; - wdata = nfs_writedata_alloc(1); + wdata = nfs_writedata_alloc(wsize); if (!wdata) return -ENOMEM; @@ -408,6 +396,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) out: clear_bit(BDI_write_congested, &bdi->state); wake_up_all(&nfs_write_congestion); + writeback_congestion_end(); return err; } @@ -501,7 +490,7 @@ nfs_mark_request_dirty(struct nfs_page *req) nfs_list_add_request(req, &nfsi->dirty); nfsi->ndirty++; spin_unlock(&nfsi->req_lock); - inc_page_state(nr_dirty); + inc_zone_page_state(req->wb_page, NR_FILE_DIRTY); mark_inode_dirty(inode); } @@ -529,7 +518,7 @@ nfs_mark_request_commit(struct nfs_page *req) nfs_list_add_request(req, &nfsi->commit); nfsi->ncommit++; spin_unlock(&nfsi->req_lock); - inc_page_state(nr_unstable); + inc_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); mark_inode_dirty(inode); } #endif @@ -583,6 +572,30 @@ static int nfs_wait_on_requests(struct inode *inode, unsigned long idx_start, un return ret; } +static void nfs_cancel_dirty_list(struct list_head *head) +{ + struct nfs_page *req; + while(!list_empty(head)) { + req = nfs_list_entry(head->next); + nfs_list_remove_request(req); + nfs_inode_remove_request(req); + nfs_clear_page_writeback(req); + } +} + +static void nfs_cancel_commit_list(struct list_head *head) +{ + struct nfs_page *req; + + while(!list_empty(head)) { + req = nfs_list_entry(head->next); + nfs_list_remove_request(req); + nfs_inode_remove_request(req); + dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); + nfs_clear_page_writeback(req); + } +} + /* * nfs_scan_dirty - Scan an inode for dirty requests * @inode: NFS inode to scan @@ -602,7 +615,6 @@ nfs_scan_dirty(struct inode *inode, struct list_head *dst, unsigned long idx_sta if (nfsi->ndirty != 0) { res = nfs_scan_lock_dirty(nfsi, dst, idx_start, npages); nfsi->ndirty -= res; - sub_page_state(nr_dirty,res); if ((nfsi->ndirty == 0) != list_empty(&nfsi->dirty)) printk(KERN_ERR "NFS: desynchronized value of nfs_i.ndirty.\n"); } @@ -627,7 +639,7 @@ nfs_scan_commit(struct inode *inode, struct list_head *dst, unsigned long idx_st int res = 0; if (nfsi->ncommit != 0) { - res = nfs_scan_list(&nfsi->commit, dst, idx_start, npages); + res = nfs_scan_list(nfsi, &nfsi->commit, dst, idx_start, npages); nfsi->ncommit -= res; if ((nfsi->ncommit == 0) != list_empty(&nfsi->commit)) printk(KERN_ERR "NFS: desynchronized value of nfs_i.ncommit.\n"); @@ -981,24 +993,24 @@ static int nfs_flush_multi(struct inode *inode, struct list_head *head, int how) struct nfs_page *req = nfs_list_entry(head->next); struct page *page = req->wb_page; struct nfs_write_data *data; - unsigned int wsize = NFS_SERVER(inode)->wsize; - unsigned int nbytes, offset; + size_t wsize = NFS_SERVER(inode)->wsize, nbytes; + unsigned int offset; int requests = 0; LIST_HEAD(list); nfs_list_remove_request(req); nbytes = req->wb_bytes; - for (;;) { - data = nfs_writedata_alloc(1); + do { + size_t len = min(nbytes, wsize); + + data = nfs_writedata_alloc(len); if (!data) goto out_bad; list_add(&data->pages, &list); requests++; - if (nbytes <= wsize) - break; - nbytes -= wsize; - } + nbytes -= len; + } while (nbytes != 0); atomic_set(&req->wb_complete, requests); ClearPageError(page); @@ -1052,7 +1064,7 @@ static int nfs_flush_one(struct inode *inode, struct list_head *head, int how) struct nfs_write_data *data; unsigned int count; - data = nfs_writedata_alloc(NFS_SERVER(inode)->wpages); + data = nfs_writedata_alloc(NFS_SERVER(inode)->wsize); if (!data) goto out_bad; @@ -1241,7 +1253,13 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) dprintk("NFS: %4d nfs_writeback_done (status %d)\n", task->tk_pid, task->tk_status); - /* Call the NFS version-specific code */ + /* + * ->write_done will attempt to use post-op attributes to detect + * conflicting writes by other clients. A strict interpretation + * of close-to-open would allow us to continue caching even if + * another writer had changed the file, but some applications + * depend on tighter cache coherency when writing. + */ status = NFS_PROTO(data->inode)->write_done(task, data); if (status != 0) return status; @@ -1262,7 +1280,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) if (time_before(complain, jiffies)) { dprintk("NFS: faulty NFS server %s:" " (committed = %d) != (stable = %d)\n", - NFS_SERVER(data->inode)->hostname, + NFS_SERVER(data->inode)->nfs_client->cl_hostname, resp->verf->committed, argp->stable); complain = jiffies + 300 * HZ; } @@ -1360,7 +1378,7 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how) struct nfs_write_data *data; struct nfs_page *req; - data = nfs_commit_alloc(NFS_SERVER(inode)->wpages); + data = nfs_commit_alloc(); if (!data) goto out_bad; @@ -1375,6 +1393,7 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how) req = nfs_list_entry(head->next); nfs_list_remove_request(req); nfs_mark_request_commit(req); + dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); nfs_clear_page_writeback(req); } return -ENOMEM; @@ -1387,7 +1406,6 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata) { struct nfs_write_data *data = calldata; struct nfs_page *req; - int res = 0; dprintk("NFS: %4d nfs_commit_done (status %d)\n", task->tk_pid, task->tk_status); @@ -1399,6 +1417,7 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata) while (!list_empty(&data->pages)) { req = nfs_list_entry(data->pages.next); nfs_list_remove_request(req); + dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); dprintk("NFS: commit (%s/%Ld %d@%Ld)", req->wb_context->dentry->d_inode->i_sb->s_id, @@ -1425,9 +1444,7 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata) nfs_mark_request_dirty(req); next: nfs_clear_page_writeback(req); - res++; } - sub_page_state(nr_unstable,res); } static const struct rpc_call_ops nfs_commit_ops = { @@ -1495,15 +1512,25 @@ int nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, pages = nfs_scan_dirty(inode, &head, idx_start, npages); if (pages != 0) { spin_unlock(&nfsi->req_lock); - ret = nfs_flush_list(inode, &head, pages, how); + if (how & FLUSH_INVALIDATE) + nfs_cancel_dirty_list(&head); + else + ret = nfs_flush_list(inode, &head, pages, how); spin_lock(&nfsi->req_lock); continue; } if (nocommit) break; - pages = nfs_scan_commit(inode, &head, 0, 0); + pages = nfs_scan_commit(inode, &head, idx_start, npages); if (pages == 0) break; + if (how & FLUSH_INVALIDATE) { + spin_unlock(&nfsi->req_lock); + nfs_cancel_commit_list(&head); + spin_lock(&nfsi->req_lock); + continue; + } + pages += nfs_scan_commit(inode, &head, 0, 0); spin_unlock(&nfsi->req_lock); ret = nfs_commit_list(inode, &head, how); spin_lock(&nfsi->req_lock); @@ -1512,7 +1539,7 @@ int nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, return ret; } -int nfs_init_writepagecache(void) +int __init nfs_init_writepagecache(void) { nfs_wdata_cachep = kmem_cache_create("nfs_write_data", sizeof(struct nfs_write_data), @@ -1538,7 +1565,6 @@ void nfs_destroy_writepagecache(void) { mempool_destroy(nfs_commit_mempool); mempool_destroy(nfs_wdata_mempool); - if (kmem_cache_destroy(nfs_wdata_cachep)) - printk(KERN_INFO "nfs_write_data: not all structures were freed\n"); + kmem_cache_destroy(nfs_wdata_cachep); } |