diff options
Diffstat (limited to 'fs/lockd/mon.c')
| -rw-r--r-- | fs/lockd/mon.c | 656 |
1 files changed, 512 insertions, 144 deletions
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index 2d144abe84a..1812f026960 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -7,239 +7,607 @@ */ #include <linux/types.h> -#include <linux/utsname.h> #include <linux/kernel.h> +#include <linux/ktime.h> +#include <linux/slab.h> + #include <linux/sunrpc/clnt.h> +#include <linux/sunrpc/addr.h> +#include <linux/sunrpc/xprtsock.h> #include <linux/sunrpc/svc.h> #include <linux/lockd/lockd.h> -#include <linux/lockd/sm_inter.h> +#include <asm/unaligned.h> + +#include "netns.h" #define NLMDBG_FACILITY NLMDBG_MONITOR +#define NSM_PROGRAM 100024 +#define NSM_VERSION 1 + +enum { + NSMPROC_NULL, + NSMPROC_STAT, + NSMPROC_MON, + NSMPROC_UNMON, + NSMPROC_UNMON_ALL, + NSMPROC_SIMU_CRASH, + NSMPROC_NOTIFY, +}; -static struct rpc_clnt * nsm_create(void); +struct nsm_args { + struct nsm_private *priv; + u32 prog; /* RPC callback info */ + u32 vers; + u32 proc; -static struct rpc_program nsm_program; + char *mon_name; + char *nodename; +}; + +struct nsm_res { + u32 status; + u32 state; +}; + +static const struct rpc_program nsm_program; +static LIST_HEAD(nsm_handles); +static DEFINE_SPINLOCK(nsm_lock); /* * Local NSM state */ -u32 nsm_local_state; +u32 __read_mostly nsm_local_state; +bool __read_mostly nsm_use_hostnames; -/* - * Common procedure for SM_MON/SM_UNMON calls - */ -static int -nsm_mon_unmon(struct nlm_host *host, u32 proc, struct nsm_res *res) +static inline struct sockaddr *nsm_addr(const struct nsm_handle *nsm) { - struct rpc_clnt *clnt; - int status; - struct nsm_args args; + return (struct sockaddr *)&nsm->sm_addr; +} - clnt = nsm_create(); - if (IS_ERR(clnt)) { - status = PTR_ERR(clnt); +static struct rpc_clnt *nsm_create(struct net *net) +{ + struct sockaddr_in sin = { + .sin_family = AF_INET, + .sin_addr.s_addr = htonl(INADDR_LOOPBACK), + }; + struct rpc_create_args args = { + .net = net, + .protocol = XPRT_TRANSPORT_TCP, + .address = (struct sockaddr *)&sin, + .addrsize = sizeof(sin), + .servername = "rpc.statd", + .program = &nsm_program, + .version = NSM_VERSION, + .authflavor = RPC_AUTH_NULL, + .flags = RPC_CLNT_CREATE_NOPING, + }; + + return rpc_create(&args); +} + +static struct rpc_clnt *nsm_client_set(struct lockd_net *ln, + struct rpc_clnt *clnt) +{ + spin_lock(&ln->nsm_clnt_lock); + if (ln->nsm_users == 0) { + if (clnt == NULL) + goto out; + ln->nsm_clnt = clnt; + } + clnt = ln->nsm_clnt; + ln->nsm_users++; +out: + spin_unlock(&ln->nsm_clnt_lock); + return clnt; +} + +static struct rpc_clnt *nsm_client_get(struct net *net) +{ + struct rpc_clnt *clnt, *new; + struct lockd_net *ln = net_generic(net, lockd_net_id); + + clnt = nsm_client_set(ln, NULL); + if (clnt != NULL) goto out; + + clnt = new = nsm_create(net); + if (IS_ERR(clnt)) + goto out; + + clnt = nsm_client_set(ln, new); + if (clnt != new) + rpc_shutdown_client(new); +out: + return clnt; +} + +static void nsm_client_put(struct net *net) +{ + struct lockd_net *ln = net_generic(net, lockd_net_id); + struct rpc_clnt *clnt = NULL; + + spin_lock(&ln->nsm_clnt_lock); + ln->nsm_users--; + if (ln->nsm_users == 0) { + clnt = ln->nsm_clnt; + ln->nsm_clnt = NULL; } + spin_unlock(&ln->nsm_clnt_lock); + if (clnt != NULL) + rpc_shutdown_client(clnt); +} + +static int nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res, + struct rpc_clnt *clnt) +{ + int status; + struct nsm_args args = { + .priv = &nsm->sm_priv, + .prog = NLM_PROGRAM, + .vers = 3, + .proc = NLMPROC_NSM_NOTIFY, + .mon_name = nsm->sm_mon_name, + .nodename = clnt->cl_nodename, + }; + struct rpc_message msg = { + .rpc_argp = &args, + .rpc_resp = res, + }; - args.addr = host->h_addr.sin_addr.s_addr; - args.proto= (host->h_proto<<1) | host->h_server; - args.prog = NLM_PROGRAM; - args.vers = host->h_version; - args.proc = NLMPROC_NSM_NOTIFY; memset(res, 0, sizeof(*res)); - status = rpc_call(clnt, proc, &args, res, 0); + msg.rpc_proc = &clnt->cl_procinfo[proc]; + status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFTCONN); if (status < 0) - printk(KERN_DEBUG "nsm_mon_unmon: rpc failed, status=%d\n", - status); + dprintk("lockd: NSM upcall RPC failed, status=%d\n", + status); else status = 0; - out: return status; } -/* - * Set up monitoring of a remote host +/** + * nsm_monitor - Notify a peer in case we reboot + * @host: pointer to nlm_host of peer to notify + * + * If this peer is not already monitored, this function sends an + * upcall to the local rpc.statd to record the name/address of + * the peer to notify in case we reboot. + * + * Returns zero if the peer is monitored by the local rpc.statd; + * otherwise a negative errno value is returned. */ -int -nsm_monitor(struct nlm_host *host) +int nsm_monitor(const struct nlm_host *host) { + struct nsm_handle *nsm = host->h_nsmhandle; struct nsm_res res; int status; + struct rpc_clnt *clnt; - dprintk("lockd: nsm_monitor(%s)\n", host->h_name); + dprintk("lockd: nsm_monitor(%s)\n", nsm->sm_name); - status = nsm_mon_unmon(host, SM_MON, &res); + if (nsm->sm_monitored) + return 0; - if (status < 0 || res.status != 0) - printk(KERN_NOTICE "lockd: cannot monitor %s\n", host->h_name); - else - host->h_monitored = 1; - return status; + /* + * Choose whether to record the caller_name or IP address of + * this peer in the local rpc.statd's database. + */ + nsm->sm_mon_name = nsm_use_hostnames ? nsm->sm_name : nsm->sm_addrbuf; + + clnt = nsm_client_get(host->net); + if (IS_ERR(clnt)) { + status = PTR_ERR(clnt); + dprintk("lockd: failed to create NSM upcall transport, " + "status=%d, net=%p\n", status, host->net); + return status; + } + + status = nsm_mon_unmon(nsm, NSMPROC_MON, &res, clnt); + if (unlikely(res.status != 0)) + status = -EIO; + if (unlikely(status < 0)) { + printk(KERN_NOTICE "lockd: cannot monitor %s\n", nsm->sm_name); + return status; + } + + nsm->sm_monitored = 1; + if (unlikely(nsm_local_state != res.state)) { + nsm_local_state = res.state; + dprintk("lockd: NSM state changed to %d\n", nsm_local_state); + } + return 0; } -/* - * Cease to monitor remote host +/** + * nsm_unmonitor - Unregister peer notification + * @host: pointer to nlm_host of peer to stop monitoring + * + * If this peer is monitored, this function sends an upcall to + * tell the local rpc.statd not to send this peer a notification + * when we reboot. */ -int -nsm_unmonitor(struct nlm_host *host) +void nsm_unmonitor(const struct nlm_host *host) { + struct nsm_handle *nsm = host->h_nsmhandle; struct nsm_res res; - int status; + int status; - dprintk("lockd: nsm_unmonitor(%s)\n", host->h_name); + if (atomic_read(&nsm->sm_count) == 1 + && nsm->sm_monitored && !nsm->sm_sticky) { + struct lockd_net *ln = net_generic(host->net, lockd_net_id); - status = nsm_mon_unmon(host, SM_UNMON, &res); - if (status < 0) - printk(KERN_NOTICE "lockd: cannot unmonitor %s\n", host->h_name); - else - host->h_monitored = 0; - return status; + dprintk("lockd: nsm_unmonitor(%s)\n", nsm->sm_name); + + status = nsm_mon_unmon(nsm, NSMPROC_UNMON, &res, ln->nsm_clnt); + if (res.status != 0) + status = -EIO; + if (status < 0) + printk(KERN_NOTICE "lockd: cannot unmonitor %s\n", + nsm->sm_name); + else + nsm->sm_monitored = 0; + + nsm_client_put(host->net); + } +} + +static struct nsm_handle *nsm_lookup_hostname(const char *hostname, + const size_t len) +{ + struct nsm_handle *nsm; + + list_for_each_entry(nsm, &nsm_handles, sm_link) + if (strlen(nsm->sm_name) == len && + memcmp(nsm->sm_name, hostname, len) == 0) + return nsm; + return NULL; +} + +static struct nsm_handle *nsm_lookup_addr(const struct sockaddr *sap) +{ + struct nsm_handle *nsm; + + list_for_each_entry(nsm, &nsm_handles, sm_link) + if (rpc_cmp_addr(nsm_addr(nsm), sap)) + return nsm; + return NULL; +} + +static struct nsm_handle *nsm_lookup_priv(const struct nsm_private *priv) +{ + struct nsm_handle *nsm; + + list_for_each_entry(nsm, &nsm_handles, sm_link) + if (memcmp(nsm->sm_priv.data, priv->data, + sizeof(priv->data)) == 0) + return nsm; + return NULL; } /* - * Create NSM client for the local host + * Construct a unique cookie to match this nsm_handle to this monitored + * host. It is passed to the local rpc.statd via NSMPROC_MON, and + * returned via NLMPROC_SM_NOTIFY, in the "priv" field of these + * requests. + * + * The NSM protocol requires that these cookies be unique while the + * system is running. We prefer a stronger requirement of making them + * unique across reboots. If user space bugs cause a stale cookie to + * be sent to the kernel, it could cause the wrong host to lose its + * lock state if cookies were not unique across reboots. + * + * The cookies are exposed only to local user space via loopback. They + * do not appear on the physical network. If we want greater security + * for some reason, nsm_init_private() could perform a one-way hash to + * obscure the contents of the cookie. */ -static struct rpc_clnt * -nsm_create(void) +static void nsm_init_private(struct nsm_handle *nsm) { - struct rpc_xprt *xprt; - struct rpc_clnt *clnt; - struct sockaddr_in sin; + u64 *p = (u64 *)&nsm->sm_priv.data; + struct timespec ts; + s64 ns; + + ktime_get_ts(&ts); + ns = timespec_to_ns(&ts); + put_unaligned(ns, p); + put_unaligned((unsigned long)nsm, p + 1); +} - sin.sin_family = AF_INET; - sin.sin_addr.s_addr = htonl(INADDR_LOOPBACK); - sin.sin_port = 0; +static struct nsm_handle *nsm_create_handle(const struct sockaddr *sap, + const size_t salen, + const char *hostname, + const size_t hostname_len) +{ + struct nsm_handle *new; + + new = kzalloc(sizeof(*new) + hostname_len + 1, GFP_KERNEL); + if (unlikely(new == NULL)) + return NULL; + + atomic_set(&new->sm_count, 1); + new->sm_name = (char *)(new + 1); + memcpy(nsm_addr(new), sap, salen); + new->sm_addrlen = salen; + nsm_init_private(new); + + if (rpc_ntop(nsm_addr(new), new->sm_addrbuf, + sizeof(new->sm_addrbuf)) == 0) + (void)snprintf(new->sm_addrbuf, sizeof(new->sm_addrbuf), + "unsupported address family"); + memcpy(new->sm_name, hostname, hostname_len); + new->sm_name[hostname_len] = '\0'; + + return new; +} + +/** + * nsm_get_handle - Find or create a cached nsm_handle + * @sap: pointer to socket address of handle to find + * @salen: length of socket address + * @hostname: pointer to C string containing hostname to find + * @hostname_len: length of C string + * + * Behavior is modulated by the global nsm_use_hostnames variable. + * + * Returns a cached nsm_handle after bumping its ref count, or + * returns a fresh nsm_handle if a handle that matches @sap and/or + * @hostname cannot be found in the handle cache. Returns NULL if + * an error occurs. + */ +struct nsm_handle *nsm_get_handle(const struct sockaddr *sap, + const size_t salen, const char *hostname, + const size_t hostname_len) +{ + struct nsm_handle *cached, *new = NULL; + + if (hostname && memchr(hostname, '/', hostname_len) != NULL) { + if (printk_ratelimit()) { + printk(KERN_WARNING "Invalid hostname \"%.*s\" " + "in NFS lock request\n", + (int)hostname_len, hostname); + } + return NULL; + } - xprt = xprt_create_proto(IPPROTO_UDP, &sin, NULL); - if (IS_ERR(xprt)) - return (struct rpc_clnt *)xprt; - xprt->resvport = 1; /* NSM requires a reserved port */ +retry: + spin_lock(&nsm_lock); - clnt = rpc_create_client(xprt, "localhost", - &nsm_program, SM_VERSION, - RPC_AUTH_NULL); - if (IS_ERR(clnt)) - goto out_err; - clnt->cl_softrtry = 1; - clnt->cl_chatty = 1; - clnt->cl_oneshot = 1; - return clnt; + if (nsm_use_hostnames && hostname != NULL) + cached = nsm_lookup_hostname(hostname, hostname_len); + else + cached = nsm_lookup_addr(sap); + + if (cached != NULL) { + atomic_inc(&cached->sm_count); + spin_unlock(&nsm_lock); + kfree(new); + dprintk("lockd: found nsm_handle for %s (%s), " + "cnt %d\n", cached->sm_name, + cached->sm_addrbuf, + atomic_read(&cached->sm_count)); + return cached; + } -out_err: - return clnt; + if (new != NULL) { + list_add(&new->sm_link, &nsm_handles); + spin_unlock(&nsm_lock); + dprintk("lockd: created nsm_handle for %s (%s)\n", + new->sm_name, new->sm_addrbuf); + return new; + } + + spin_unlock(&nsm_lock); + + new = nsm_create_handle(sap, salen, hostname, hostname_len); + if (unlikely(new == NULL)) + return NULL; + goto retry; +} + +/** + * nsm_reboot_lookup - match NLMPROC_SM_NOTIFY arguments to an nsm_handle + * @info: pointer to NLMPROC_SM_NOTIFY arguments + * + * Returns a matching nsm_handle if found in the nsm cache. The returned + * nsm_handle's reference count is bumped. Otherwise returns NULL if some + * error occurred. + */ +struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info) +{ + struct nsm_handle *cached; + + spin_lock(&nsm_lock); + + cached = nsm_lookup_priv(&info->priv); + if (unlikely(cached == NULL)) { + spin_unlock(&nsm_lock); + dprintk("lockd: never saw rebooted peer '%.*s' before\n", + info->len, info->mon); + return cached; + } + + atomic_inc(&cached->sm_count); + spin_unlock(&nsm_lock); + + dprintk("lockd: host %s (%s) rebooted, cnt %d\n", + cached->sm_name, cached->sm_addrbuf, + atomic_read(&cached->sm_count)); + return cached; +} + +/** + * nsm_release - Release an NSM handle + * @nsm: pointer to handle to be released + * + */ +void nsm_release(struct nsm_handle *nsm) +{ + if (atomic_dec_and_lock(&nsm->sm_count, &nsm_lock)) { + list_del(&nsm->sm_link); + spin_unlock(&nsm_lock); + dprintk("lockd: destroyed nsm_handle for %s (%s)\n", + nsm->sm_name, nsm->sm_addrbuf); + kfree(nsm); + } } /* * XDR functions for NSM. + * + * See http://www.opengroup.org/ for details on the Network + * Status Monitor wire protocol. */ -static u32 * -xdr_encode_common(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +static void encode_nsm_string(struct xdr_stream *xdr, const char *string) { - char buffer[20]; + const u32 len = strlen(string); + __be32 *p; - /* - * Use the dotted-quad IP address of the remote host as - * identifier. Linux statd always looks up the canonical - * hostname first for whatever remote hostname it receives, - * so this works alright. - */ - sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(argp->addr)); - if (!(p = xdr_encode_string(p, buffer)) - || !(p = xdr_encode_string(p, system_utsname.nodename))) - return ERR_PTR(-EIO); - *p++ = htonl(argp->prog); - *p++ = htonl(argp->vers); - *p++ = htonl(argp->proc); - - return p; -} - -static int -xdr_encode_mon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) -{ - p = xdr_encode_common(rqstp, p, argp); - if (IS_ERR(p)) - return PTR_ERR(p); - *p++ = argp->addr; - *p++ = argp->vers; - *p++ = argp->proto; - *p++ = 0; - rqstp->rq_slen = xdr_adjust_iovec(rqstp->rq_svec, p); - return 0; + p = xdr_reserve_space(xdr, 4 + len); + xdr_encode_opaque(p, string, len); } -static int -xdr_encode_unmon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +/* + * "mon_name" specifies the host to be monitored. + */ +static void encode_mon_name(struct xdr_stream *xdr, const struct nsm_args *argp) { - p = xdr_encode_common(rqstp, p, argp); - if (IS_ERR(p)) - return PTR_ERR(p); - rqstp->rq_slen = xdr_adjust_iovec(rqstp->rq_svec, p); - return 0; + encode_nsm_string(xdr, argp->mon_name); +} + +/* + * The "my_id" argument specifies the hostname and RPC procedure + * to be called when the status manager receives notification + * (via the NLMPROC_SM_NOTIFY call) that the state of host "mon_name" + * has changed. + */ +static void encode_my_id(struct xdr_stream *xdr, const struct nsm_args *argp) +{ + __be32 *p; + + encode_nsm_string(xdr, argp->nodename); + p = xdr_reserve_space(xdr, 4 + 4 + 4); + *p++ = cpu_to_be32(argp->prog); + *p++ = cpu_to_be32(argp->vers); + *p = cpu_to_be32(argp->proc); } -static int -xdr_decode_stat_res(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) +/* + * The "mon_id" argument specifies the non-private arguments + * of an NSMPROC_MON or NSMPROC_UNMON call. + */ +static void encode_mon_id(struct xdr_stream *xdr, const struct nsm_args *argp) { - resp->status = ntohl(*p++); - resp->state = ntohl(*p++); - dprintk("nsm: xdr_decode_stat_res status %d state %d\n", - resp->status, resp->state); + encode_mon_name(xdr, argp); + encode_my_id(xdr, argp); +} + +/* + * The "priv" argument may contain private information required + * by the NSMPROC_MON call. This information will be supplied in the + * NLMPROC_SM_NOTIFY call. + */ +static void encode_priv(struct xdr_stream *xdr, const struct nsm_args *argp) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, SM_PRIV_SIZE); + xdr_encode_opaque_fixed(p, argp->priv->data, SM_PRIV_SIZE); +} + +static void nsm_xdr_enc_mon(struct rpc_rqst *req, struct xdr_stream *xdr, + const struct nsm_args *argp) +{ + encode_mon_id(xdr, argp); + encode_priv(xdr, argp); +} + +static void nsm_xdr_enc_unmon(struct rpc_rqst *req, struct xdr_stream *xdr, + const struct nsm_args *argp) +{ + encode_mon_id(xdr, argp); +} + +static int nsm_xdr_dec_stat_res(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + struct nsm_res *resp) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, 4 + 4); + if (unlikely(p == NULL)) + return -EIO; + resp->status = be32_to_cpup(p++); + resp->state = be32_to_cpup(p); + + dprintk("lockd: %s status %d state %d\n", + __func__, resp->status, resp->state); return 0; } -static int -xdr_decode_stat(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) +static int nsm_xdr_dec_stat(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + struct nsm_res *resp) { - resp->state = ntohl(*p++); + __be32 *p; + + p = xdr_inline_decode(xdr, 4); + if (unlikely(p == NULL)) + return -EIO; + resp->state = be32_to_cpup(p); + + dprintk("lockd: %s state %d\n", __func__, resp->state); return 0; } #define SM_my_name_sz (1+XDR_QUADLEN(SM_MAXSTRLEN)) -#define SM_my_id_sz (3+1+SM_my_name_sz) -#define SM_mon_id_sz (1+XDR_QUADLEN(20)+SM_my_id_sz) -#define SM_mon_sz (SM_mon_id_sz+4) +#define SM_my_id_sz (SM_my_name_sz+3) +#define SM_mon_name_sz (1+XDR_QUADLEN(SM_MAXSTRLEN)) +#define SM_mon_id_sz (SM_mon_name_sz+SM_my_id_sz) +#define SM_priv_sz (XDR_QUADLEN(SM_PRIV_SIZE)) +#define SM_mon_sz (SM_mon_id_sz+SM_priv_sz) #define SM_monres_sz 2 #define SM_unmonres_sz 1 -#ifndef MAX -# define MAX(a, b) (((a) > (b))? (a) : (b)) -#endif - static struct rpc_procinfo nsm_procedures[] = { -[SM_MON] = { - .p_proc = SM_MON, - .p_encode = (kxdrproc_t) xdr_encode_mon, - .p_decode = (kxdrproc_t) xdr_decode_stat_res, - .p_bufsiz = MAX(SM_mon_sz, SM_monres_sz) << 2, +[NSMPROC_MON] = { + .p_proc = NSMPROC_MON, + .p_encode = (kxdreproc_t)nsm_xdr_enc_mon, + .p_decode = (kxdrdproc_t)nsm_xdr_dec_stat_res, + .p_arglen = SM_mon_sz, + .p_replen = SM_monres_sz, + .p_statidx = NSMPROC_MON, + .p_name = "MONITOR", }, -[SM_UNMON] = { - .p_proc = SM_UNMON, - .p_encode = (kxdrproc_t) xdr_encode_unmon, - .p_decode = (kxdrproc_t) xdr_decode_stat, - .p_bufsiz = MAX(SM_mon_id_sz, SM_unmonres_sz) << 2, +[NSMPROC_UNMON] = { + .p_proc = NSMPROC_UNMON, + .p_encode = (kxdreproc_t)nsm_xdr_enc_unmon, + .p_decode = (kxdrdproc_t)nsm_xdr_dec_stat, + .p_arglen = SM_mon_id_sz, + .p_replen = SM_unmonres_sz, + .p_statidx = NSMPROC_UNMON, + .p_name = "UNMONITOR", }, }; -static struct rpc_version nsm_version1 = { - .number = 1, - .nrprocs = sizeof(nsm_procedures)/sizeof(nsm_procedures[0]), +static const struct rpc_version nsm_version1 = { + .number = 1, + .nrprocs = ARRAY_SIZE(nsm_procedures), .procs = nsm_procedures }; -static struct rpc_version * nsm_version[] = { +static const struct rpc_version *nsm_version[] = { [1] = &nsm_version1, }; static struct rpc_stat nsm_stats; -static struct rpc_program nsm_program = { +static const struct rpc_program nsm_program = { .name = "statd", - .number = SM_PROGRAM, - .nrvers = sizeof(nsm_version)/sizeof(nsm_version[0]), + .number = NSM_PROGRAM, + .nrvers = ARRAY_SIZE(nsm_version), .version = nsm_version, .stats = &nsm_stats }; |
