diff options
Diffstat (limited to 'fs/lockd/svc.c')
| -rw-r--r-- | fs/lockd/svc.c | 502 |
1 files changed, 306 insertions, 196 deletions
diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 1ed8bd4de94..8f27c93f8d2 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -21,10 +21,9 @@ #include <linux/errno.h> #include <linux/in.h> #include <linux/uio.h> -#include <linux/slab.h> #include <linux/smp.h> -#include <linux/smp_lock.h> #include <linux/mutex.h> +#include <linux/kthread.h> #include <linux/freezer.h> #include <linux/sunrpc/types.h> @@ -34,9 +33,10 @@ #include <linux/sunrpc/svcsock.h> #include <net/ip.h> #include <linux/lockd/lockd.h> -#include <linux/lockd/sm_inter.h> #include <linux/nfs.h> +#include "netns.h" + #define NLMDBG_FACILITY NLMDBG_SVC #define LOCKD_BUFSIZE (1024 + NLMSVC_XDRSIZE) #define ALLOWED_SIGS (sigmask(SIGKILL)) @@ -44,17 +44,15 @@ static struct svc_program nlmsvc_program; struct nlmsvc_binding * nlmsvc_ops; -EXPORT_SYMBOL(nlmsvc_ops); +EXPORT_SYMBOL_GPL(nlmsvc_ops); static DEFINE_MUTEX(nlmsvc_mutex); static unsigned int nlmsvc_users; -static pid_t nlmsvc_pid; -static struct svc_serv *nlmsvc_serv; -int nlmsvc_grace_period; +static struct task_struct *nlmsvc_task; +static struct svc_rqst *nlmsvc_rqst; unsigned long nlmsvc_timeout; -static DECLARE_COMPLETION(lockd_start_done); -static DECLARE_WAIT_QUEUE_HEAD(lockd_exit); +int lockd_net_id; /* * These can be set at insmod time (useful for NFS as root filesystem), @@ -63,7 +61,9 @@ static DECLARE_WAIT_QUEUE_HEAD(lockd_exit); static unsigned long nlm_grace_period; static unsigned long nlm_timeout = LOCKD_DFLT_TIMEO; static int nlm_udpport, nlm_tcpport; -int nsm_use_hostnames = 0; + +/* RLIM_NOFILE defaults to 1024. That seems like a reasonable default here. */ +static unsigned int nlm_max_connections = 1024; /* * Constants needed for the sysctl interface. @@ -74,7 +74,9 @@ static const unsigned long nlm_timeout_min = 3; static const unsigned long nlm_timeout_max = 20; static const int nlm_port_min = 0, nlm_port_max = 65535; +#ifdef CONFIG_SYSCTL static struct ctl_table_header * nlm_sysctl_table; +#endif static unsigned long get_lockd_grace_period(void) { @@ -85,57 +87,52 @@ static unsigned long get_lockd_grace_period(void) return nlm_timeout * 5 * HZ; } -unsigned long get_nfs_grace_period(void) +static void grace_ender(struct work_struct *grace) { - unsigned long lockdgrace = get_lockd_grace_period(); - unsigned long nfsdgrace = 0; + struct delayed_work *dwork = container_of(grace, struct delayed_work, + work); + struct lockd_net *ln = container_of(dwork, struct lockd_net, + grace_period_end); - if (nlmsvc_ops) - nfsdgrace = nlmsvc_ops->get_grace_period(); - - return max(lockdgrace, nfsdgrace); + locks_end_grace(&ln->lockd_manager); } -EXPORT_SYMBOL(get_nfs_grace_period); -static unsigned long set_grace_period(void) +static void set_grace_period(struct net *net) { - nlmsvc_grace_period = 1; - return get_nfs_grace_period() + jiffies; + unsigned long grace_period = get_lockd_grace_period(); + struct lockd_net *ln = net_generic(net, lockd_net_id); + + locks_start_grace(net, &ln->lockd_manager); + cancel_delayed_work_sync(&ln->grace_period_end); + schedule_delayed_work(&ln->grace_period_end, grace_period); } -static inline void clear_grace_period(void) +static void restart_grace(void) { - nlmsvc_grace_period = 0; + if (nlmsvc_ops) { + struct net *net = &init_net; + struct lockd_net *ln = net_generic(net, lockd_net_id); + + cancel_delayed_work_sync(&ln->grace_period_end); + locks_end_grace(&ln->lockd_manager); + nlmsvc_invalidate_all(); + set_grace_period(net); + } } /* * This is the lockd kernel thread */ -static void -lockd(struct svc_rqst *rqstp) +static int +lockd(void *vrqstp) { int err = 0; - unsigned long grace_period_expire; + struct svc_rqst *rqstp = vrqstp; - /* Lock module and set up kernel thread */ - /* lockd_up is waiting for us to startup, so will - * be holding a reference to this module, so it - * is safe to just claim another reference - */ - __module_get(THIS_MODULE); - lock_kernel(); - - /* - * Let our maker know we're running. - */ - nlmsvc_pid = current->pid; - nlmsvc_serv = rqstp->rq_server; - complete(&lockd_start_done); - - daemonize("lockd"); + /* try_to_freeze() is called from svc_recv() */ set_freezable(); - /* Process request with signals blocked, but allow SIGKILL. */ + /* Allow SIGKILL to tell lockd to drop all of its locks */ allow_signal(SIGKILL); dprintk("NFS locking service started (ver " LOCKD_VERSION ").\n"); @@ -144,35 +141,24 @@ lockd(struct svc_rqst *rqstp) nlm_timeout = LOCKD_DFLT_TIMEO; nlmsvc_timeout = nlm_timeout * HZ; - grace_period_expire = set_grace_period(); - /* * The main request loop. We don't terminate until the last - * NFS mount or NFS daemon has gone away, and we've been sent a - * signal, or else another process has taken over our job. + * NFS mount or NFS daemon has gone away. */ - while ((nlmsvc_users || !signalled()) && nlmsvc_pid == current->pid) { + while (!kthread_should_stop()) { long timeout = MAX_SCHEDULE_TIMEOUT; RPC_IFDEBUG(char buf[RPC_MAX_ADDRBUFLEN]); + /* update sv_maxconn if it has changed */ + rqstp->rq_server->sv_maxconn = nlm_max_connections; + if (signalled()) { flush_signals(current); - if (nlmsvc_ops) { - nlmsvc_invalidate_all(); - grace_period_expire = set_grace_period(); - } + restart_grace(); + continue; } - /* - * Retry any blocked locks that have been notified by - * the VFS. Don't do this during grace period. - * (Theoretically, there shouldn't even be blocked locks - * during grace period). - */ - if (!nlmsvc_grace_period) { - timeout = nlmsvc_retry_blocked(); - } else if (time_before(grace_period_expire, jiffies)) - clear_grace_period(); + timeout = nlmsvc_retry_blocked(); /* * Find a socket with data available and call its @@ -181,97 +167,177 @@ lockd(struct svc_rqst *rqstp) err = svc_recv(rqstp, timeout); if (err == -EAGAIN || err == -EINTR) continue; - if (err < 0) { - printk(KERN_WARNING - "lockd: terminating on error %d\n", - -err); - break; - } - dprintk("lockd: request from %s\n", svc_print_addr(rqstp, buf, sizeof(buf))); svc_process(rqstp); } - flush_signals(current); + if (nlmsvc_ops) + nlmsvc_invalidate_all(); + nlm_shutdown_hosts(); + return 0; +} - /* - * Check whether there's a new lockd process before - * shutting down the hosts and clearing the slot. - */ - if (!nlmsvc_pid || current->pid == nlmsvc_pid) { - if (nlmsvc_ops) - nlmsvc_invalidate_all(); - nlm_shutdown_hosts(); - nlmsvc_pid = 0; - nlmsvc_serv = NULL; - } else - printk(KERN_DEBUG - "lockd: new process, skipping host shutdown\n"); - wake_up(&lockd_exit); - - /* Exit the RPC thread */ - svc_exit_thread(rqstp); - - /* Release module */ - unlock_kernel(); - module_put_and_exit(0); +static int create_lockd_listener(struct svc_serv *serv, const char *name, + struct net *net, const int family, + const unsigned short port) +{ + struct svc_xprt *xprt; + + xprt = svc_find_xprt(serv, name, net, family, 0); + if (xprt == NULL) + return svc_create_xprt(serv, name, net, family, port, + SVC_SOCK_DEFAULTS); + svc_xprt_put(xprt); + return 0; +} + +static int create_lockd_family(struct svc_serv *serv, struct net *net, + const int family) +{ + int err; + + err = create_lockd_listener(serv, "udp", net, family, nlm_udpport); + if (err < 0) + return err; + + return create_lockd_listener(serv, "tcp", net, family, nlm_tcpport); } /* - * Make any sockets that are needed but not present. - * If nlm_udpport or nlm_tcpport were set as module - * options, make those sockets unconditionally + * Ensure there are active UDP and TCP listeners for lockd. + * + * Even if we have only TCP NFS mounts and/or TCP NFSDs, some + * local services (such as rpc.statd) still require UDP, and + * some NFS servers do not yet support NLM over TCP. + * + * Returns zero if all listeners are available; otherwise a + * negative errno value is returned. */ -static int make_socks(struct svc_serv *serv, int proto) +static int make_socks(struct svc_serv *serv, struct net *net) { static int warned; - struct svc_xprt *xprt; - int err = 0; - - if (proto == IPPROTO_UDP || nlm_udpport) { - xprt = svc_find_xprt(serv, "udp", 0, 0); - if (!xprt) - err = svc_create_xprt(serv, "udp", nlm_udpport, - SVC_SOCK_DEFAULTS); - else - svc_xprt_put(xprt); + int err; + + err = create_lockd_family(serv, net, PF_INET); + if (err < 0) + goto out_err; + + err = create_lockd_family(serv, net, PF_INET6); + if (err < 0 && err != -EAFNOSUPPORT) + goto out_err; + + warned = 0; + return 0; + +out_err: + if (warned++ == 0) + printk(KERN_WARNING + "lockd_up: makesock failed, error=%d\n", err); + svc_shutdown_net(serv, net); + return err; +} + +static int lockd_up_net(struct svc_serv *serv, struct net *net) +{ + struct lockd_net *ln = net_generic(net, lockd_net_id); + int error; + + if (ln->nlmsvc_users++) + return 0; + + error = svc_bind(serv, net); + if (error) + goto err_bind; + + error = make_socks(serv, net); + if (error < 0) + goto err_socks; + set_grace_period(net); + dprintk("lockd_up_net: per-net data created; net=%p\n", net); + return 0; + +err_socks: + svc_rpcb_cleanup(serv, net); +err_bind: + ln->nlmsvc_users--; + return error; +} + +static void lockd_down_net(struct svc_serv *serv, struct net *net) +{ + struct lockd_net *ln = net_generic(net, lockd_net_id); + + if (ln->nlmsvc_users) { + if (--ln->nlmsvc_users == 0) { + nlm_shutdown_hosts_net(net); + cancel_delayed_work_sync(&ln->grace_period_end); + locks_end_grace(&ln->lockd_manager); + svc_shutdown_net(serv, net); + dprintk("lockd_down_net: per-net data destroyed; net=%p\n", net); + } + } else { + printk(KERN_ERR "lockd_down_net: no users! task=%p, net=%p\n", + nlmsvc_task, net); + BUG(); } - if (err >= 0 && (proto == IPPROTO_TCP || nlm_tcpport)) { - xprt = svc_find_xprt(serv, "tcp", 0, 0); - if (!xprt) - err = svc_create_xprt(serv, "tcp", nlm_tcpport, - SVC_SOCK_DEFAULTS); - else - svc_xprt_put(xprt); +} + +static int lockd_start_svc(struct svc_serv *serv) +{ + int error; + + if (nlmsvc_rqst) + return 0; + + /* + * Create the kernel thread and wait for it to start. + */ + nlmsvc_rqst = svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE); + if (IS_ERR(nlmsvc_rqst)) { + error = PTR_ERR(nlmsvc_rqst); + printk(KERN_WARNING + "lockd_up: svc_rqst allocation failed, error=%d\n", + error); + goto out_rqst; } - if (err >= 0) { - warned = 0; - err = 0; - } else if (warned++ == 0) + + svc_sock_update_bufs(serv); + serv->sv_maxconn = nlm_max_connections; + + nlmsvc_task = kthread_run(lockd, nlmsvc_rqst, "%s", serv->sv_name); + if (IS_ERR(nlmsvc_task)) { + error = PTR_ERR(nlmsvc_task); printk(KERN_WARNING - "lockd_up: makesock failed, error=%d\n", err); - return err; + "lockd_up: kthread_run failed, error=%d\n", error); + goto out_task; + } + dprintk("lockd_up: service started\n"); + return 0; + +out_task: + svc_exit_thread(nlmsvc_rqst); + nlmsvc_task = NULL; +out_rqst: + nlmsvc_rqst = NULL; + return error; } -/* - * Bring up the lockd process if it's not already up. - */ -int -lockd_up(int proto) /* Maybe add a 'family' option when IPv6 is supported ?? */ +static struct svc_serv *lockd_create_svc(void) { - struct svc_serv * serv; - int error = 0; + struct svc_serv *serv; - mutex_lock(&nlmsvc_mutex); /* * Check whether we're already up and running. */ - if (nlmsvc_pid) { - if (proto) - error = make_socks(nlmsvc_serv, proto); - goto out; + if (nlmsvc_rqst) { + /* + * Note: increase service usage, because later in case of error + * svc_destroy() will be called. + */ + svc_get(nlmsvc_rqst->rq_server); + return nlmsvc_rqst->rq_server; } /* @@ -282,167 +348,168 @@ lockd_up(int proto) /* Maybe add a 'family' option when IPv6 is supported ?? */ printk(KERN_WARNING "lockd_up: no pid, %d users??\n", nlmsvc_users); - error = -ENOMEM; serv = svc_create(&nlmsvc_program, LOCKD_BUFSIZE, NULL); if (!serv) { printk(KERN_WARNING "lockd_up: create service failed\n"); - goto out; + return ERR_PTR(-ENOMEM); } + dprintk("lockd_up: service created\n"); + return serv; +} + +/* + * Bring up the lockd process if it's not already up. + */ +int lockd_up(struct net *net) +{ + struct svc_serv *serv; + int error; - if ((error = make_socks(serv, proto)) < 0) - goto destroy_and_out; + mutex_lock(&nlmsvc_mutex); - /* - * Create the kernel thread and wait for it to start. - */ - error = svc_create_thread(lockd, serv); - if (error) { - printk(KERN_WARNING - "lockd_up: create thread failed, error=%d\n", error); - goto destroy_and_out; + serv = lockd_create_svc(); + if (IS_ERR(serv)) { + error = PTR_ERR(serv); + goto err_create; } - wait_for_completion(&lockd_start_done); + error = lockd_up_net(serv, net); + if (error < 0) + goto err_net; + + error = lockd_start_svc(serv); + if (error < 0) + goto err_start; + + nlmsvc_users++; /* * Note: svc_serv structures have an initial use count of 1, * so we exit through here on both success and failure. */ -destroy_and_out: +err_net: svc_destroy(serv); -out: - if (!error) - nlmsvc_users++; +err_create: mutex_unlock(&nlmsvc_mutex); return error; + +err_start: + lockd_down_net(serv, net); + goto err_net; } -EXPORT_SYMBOL(lockd_up); +EXPORT_SYMBOL_GPL(lockd_up); /* * Decrement the user count and bring down lockd if we're the last. */ void -lockd_down(void) +lockd_down(struct net *net) { - static int warned; - mutex_lock(&nlmsvc_mutex); + lockd_down_net(nlmsvc_rqst->rq_server, net); if (nlmsvc_users) { if (--nlmsvc_users) goto out; - } else - printk(KERN_WARNING "lockd_down: no users! pid=%d\n", nlmsvc_pid); - - if (!nlmsvc_pid) { - if (warned++ == 0) - printk(KERN_WARNING "lockd_down: no lockd running.\n"); - goto out; + } else { + printk(KERN_ERR "lockd_down: no users! task=%p\n", + nlmsvc_task); + BUG(); } - warned = 0; - kill_proc(nlmsvc_pid, SIGKILL, 1); - /* - * Wait for the lockd process to exit, but since we're holding - * the lockd semaphore, we can't wait around forever ... - */ - clear_thread_flag(TIF_SIGPENDING); - interruptible_sleep_on_timeout(&lockd_exit, HZ); - if (nlmsvc_pid) { - printk(KERN_WARNING - "lockd_down: lockd failed to exit, clearing pid\n"); - nlmsvc_pid = 0; + if (!nlmsvc_task) { + printk(KERN_ERR "lockd_down: no lockd running.\n"); + BUG(); } - spin_lock_irq(¤t->sighand->siglock); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); + kthread_stop(nlmsvc_task); + dprintk("lockd_down: service stopped\n"); + svc_exit_thread(nlmsvc_rqst); + dprintk("lockd_down: service destroyed\n"); + nlmsvc_task = NULL; + nlmsvc_rqst = NULL; out: mutex_unlock(&nlmsvc_mutex); } -EXPORT_SYMBOL(lockd_down); +EXPORT_SYMBOL_GPL(lockd_down); + +#ifdef CONFIG_SYSCTL /* * Sysctl parameters (same as module parameters, different interface). */ -static ctl_table nlm_sysctls[] = { +static struct ctl_table nlm_sysctls[] = { { - .ctl_name = CTL_UNNUMBERED, .procname = "nlm_grace_period", .data = &nlm_grace_period, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = (unsigned long *) &nlm_grace_period_min, .extra2 = (unsigned long *) &nlm_grace_period_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "nlm_timeout", .data = &nlm_timeout, .maxlen = sizeof(unsigned long), .mode = 0644, - .proc_handler = &proc_doulongvec_minmax, + .proc_handler = proc_doulongvec_minmax, .extra1 = (unsigned long *) &nlm_timeout_min, .extra2 = (unsigned long *) &nlm_timeout_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "nlm_udpport", .data = &nlm_udpport, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, + .proc_handler = proc_dointvec_minmax, .extra1 = (int *) &nlm_port_min, .extra2 = (int *) &nlm_port_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "nlm_tcpport", .data = &nlm_tcpport, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_minmax, + .proc_handler = proc_dointvec_minmax, .extra1 = (int *) &nlm_port_min, .extra2 = (int *) &nlm_port_max, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "nsm_use_hostnames", .data = &nsm_use_hostnames, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, { - .ctl_name = CTL_UNNUMBERED, .procname = "nsm_local_state", .data = &nsm_local_state, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = proc_dointvec, }, - { .ctl_name = 0 } + { } }; -static ctl_table nlm_sysctl_dir[] = { +static struct ctl_table nlm_sysctl_dir[] = { { - .ctl_name = CTL_UNNUMBERED, .procname = "nfs", .mode = 0555, .child = nlm_sysctls, }, - { .ctl_name = 0 } + { } }; -static ctl_table nlm_sysctl_root[] = { +static struct ctl_table nlm_sysctl_root[] = { { - .ctl_name = CTL_FS, .procname = "fs", .mode = 0555, .child = nlm_sysctl_dir, }, - { .ctl_name = 0 } + { } }; +#endif /* CONFIG_SYSCTL */ + /* * Module (and sysfs) parameters. */ @@ -454,7 +521,7 @@ static int param_set_##name(const char *val, struct kernel_param *kp) \ __typeof__(type) num = which_strtol(val, &endp, 0); \ if (endp == val || *endp || num < (min) || num > (max)) \ return -EINVAL; \ - *((int *) kp->arg) = num; \ + *((type *) kp->arg) = num; \ return 0; \ } @@ -509,6 +576,29 @@ module_param_call(nlm_udpport, param_set_port, param_get_int, module_param_call(nlm_tcpport, param_set_port, param_get_int, &nlm_tcpport, 0644); module_param(nsm_use_hostnames, bool, 0644); +module_param(nlm_max_connections, uint, 0644); + +static int lockd_init_net(struct net *net) +{ + struct lockd_net *ln = net_generic(net, lockd_net_id); + + INIT_DELAYED_WORK(&ln->grace_period_end, grace_ender); + INIT_LIST_HEAD(&ln->grace_list); + spin_lock_init(&ln->nsm_clnt_lock); + return 0; +} + +static void lockd_exit_net(struct net *net) +{ +} + +static struct pernet_operations lockd_net_ops = { + .init = lockd_init_net, + .exit = lockd_exit_net, + .id = &lockd_net_id, + .size = sizeof(struct lockd_net), +}; + /* * Initialising and terminating the module. @@ -516,15 +606,35 @@ module_param(nsm_use_hostnames, bool, 0644); static int __init init_nlm(void) { + int err; + +#ifdef CONFIG_SYSCTL + err = -ENOMEM; nlm_sysctl_table = register_sysctl_table(nlm_sysctl_root); - return nlm_sysctl_table ? 0 : -ENOMEM; + if (nlm_sysctl_table == NULL) + goto err_sysctl; +#endif + err = register_pernet_subsys(&lockd_net_ops); + if (err) + goto err_pernet; + return 0; + +err_pernet: +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nlm_sysctl_table); +err_sysctl: +#endif + return err; } static void __exit exit_nlm(void) { /* FIXME: delete all NLM clients */ nlm_shutdown_hosts(); + unregister_pernet_subsys(&lockd_net_ops); +#ifdef CONFIG_SYSCTL unregister_sysctl_table(nlm_sysctl_table); +#endif } module_init(init_nlm); |
