diff options
Diffstat (limited to 'net')
40 files changed, 470 insertions, 313 deletions
diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index 35b8911b1c8..fd05c81cb34 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -39,6 +39,7 @@ #include <linux/inet.h> #include <linux/idr.h> #include <linux/file.h> +#include <linux/highmem.h> #include <linux/slab.h> #include <net/9p/9p.h> #include <linux/parser.h> @@ -325,7 +326,7 @@ static int p9_get_mapped_pages(struct virtio_chan *chan, int count = nr_pages; while (nr_pages) { s = rest_of_page(data); - pages[index++] = virt_to_page(data); + pages[index++] = kmap_to_page(data); data += s; nr_pages--; } diff --git a/net/atm/atm_sysfs.c b/net/atm/atm_sysfs.c index f49da5814bc..350bf62b2ae 100644 --- a/net/atm/atm_sysfs.c +++ b/net/atm/atm_sysfs.c @@ -14,49 +14,45 @@ static ssize_t show_type(struct device *cdev, struct device_attribute *attr, char *buf) { struct atm_dev *adev = to_atm_dev(cdev); - return sprintf(buf, "%s\n", adev->type); + + return scnprintf(buf, PAGE_SIZE, "%s\n", adev->type); } static ssize_t show_address(struct device *cdev, struct device_attribute *attr, char *buf) { - char *pos = buf; struct atm_dev *adev = to_atm_dev(cdev); - int i; - - for (i = 0; i < (ESI_LEN - 1); i++) - pos += sprintf(pos, "%02x:", adev->esi[i]); - pos += sprintf(pos, "%02x\n", adev->esi[i]); - return pos - buf; + return scnprintf(buf, PAGE_SIZE, "%pM\n", adev->esi); } static ssize_t show_atmaddress(struct device *cdev, struct device_attribute *attr, char *buf) { unsigned long flags; - char *pos = buf; struct atm_dev *adev = to_atm_dev(cdev); struct atm_dev_addr *aaddr; int bin[] = { 1, 2, 10, 6, 1 }, *fmt = bin; - int i, j; + int i, j, count = 0; spin_lock_irqsave(&adev->lock, flags); list_for_each_entry(aaddr, &adev->local, entry) { for (i = 0, j = 0; i < ATM_ESA_LEN; ++i, ++j) { if (j == *fmt) { - pos += sprintf(pos, "."); + count += scnprintf(buf + count, + PAGE_SIZE - count, "."); ++fmt; j = 0; } - pos += sprintf(pos, "%02x", - aaddr->addr.sas_addr.prv[i]); + count += scnprintf(buf + count, + PAGE_SIZE - count, "%02x", + aaddr->addr.sas_addr.prv[i]); } - pos += sprintf(pos, "\n"); + count += scnprintf(buf + count, PAGE_SIZE - count, "\n"); } spin_unlock_irqrestore(&adev->lock, flags); - return pos - buf; + return count; } static ssize_t show_atmindex(struct device *cdev, @@ -64,25 +60,21 @@ static ssize_t show_atmindex(struct device *cdev, { struct atm_dev *adev = to_atm_dev(cdev); - return sprintf(buf, "%d\n", adev->number); + return scnprintf(buf, PAGE_SIZE, "%d\n", adev->number); } static ssize_t show_carrier(struct device *cdev, struct device_attribute *attr, char *buf) { - char *pos = buf; struct atm_dev *adev = to_atm_dev(cdev); - pos += sprintf(pos, "%d\n", - adev->signal == ATM_PHY_SIG_LOST ? 0 : 1); - - return pos - buf; + return scnprintf(buf, PAGE_SIZE, "%d\n", + adev->signal == ATM_PHY_SIG_LOST ? 0 : 1); } static ssize_t show_link_rate(struct device *cdev, struct device_attribute *attr, char *buf) { - char *pos = buf; struct atm_dev *adev = to_atm_dev(cdev); int link_rate; @@ -100,9 +92,7 @@ static ssize_t show_link_rate(struct device *cdev, default: link_rate = adev->link_rate * 8 * 53; } - pos += sprintf(pos, "%d\n", link_rate); - - return pos - buf; + return scnprintf(buf, PAGE_SIZE, "%d\n", link_rate); } static DEVICE_ATTR(address, S_IRUGO, show_address, NULL); diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c index 6f0a2eebcb2..acc9f4cc18f 100644 --- a/net/bridge/br_mdb.c +++ b/net/bridge/br_mdb.c @@ -83,9 +83,12 @@ static int br_mdb_fill_info(struct sk_buff *skb, struct netlink_callback *cb, if (port) { struct br_mdb_entry e; e.ifindex = port->dev->ifindex; - e.addr.u.ip4 = p->addr.u.ip4; + e.state = p->state; + if (p->addr.proto == htons(ETH_P_IP)) + e.addr.u.ip4 = p->addr.u.ip4; #if IS_ENABLED(CONFIG_IPV6) - e.addr.u.ip6 = p->addr.u.ip6; + if (p->addr.proto == htons(ETH_P_IPV6)) + e.addr.u.ip6 = p->addr.u.ip6; #endif e.addr.proto = p->addr.proto; if (nla_put(skb, MDBA_MDB_ENTRY_INFO, sizeof(e), &e)) { @@ -253,6 +256,8 @@ static bool is_valid_mdb_entry(struct br_mdb_entry *entry) #endif } else return false; + if (entry->state != MDB_PERMANENT && entry->state != MDB_TEMPORARY) + return false; return true; } @@ -310,7 +315,7 @@ static int br_mdb_parse(struct sk_buff *skb, struct nlmsghdr *nlh, } static int br_mdb_add_group(struct net_bridge *br, struct net_bridge_port *port, - struct br_ip *group) + struct br_ip *group, unsigned char state) { struct net_bridge_mdb_entry *mp; struct net_bridge_port_group *p; @@ -336,7 +341,7 @@ static int br_mdb_add_group(struct net_bridge *br, struct net_bridge_port *port, break; } - p = br_multicast_new_port_group(port, group, *pp); + p = br_multicast_new_port_group(port, group, *pp, state); if (unlikely(!p)) return -ENOMEM; rcu_assign_pointer(*pp, p); @@ -373,7 +378,7 @@ static int __br_mdb_add(struct net *net, struct net_bridge *br, #endif spin_lock_bh(&br->multicast_lock); - ret = br_mdb_add_group(br, p, &ip); + ret = br_mdb_add_group(br, p, &ip, entry->state); spin_unlock_bh(&br->multicast_lock); return ret; } @@ -479,3 +484,10 @@ void br_mdb_init(void) rtnl_register(PF_BRIDGE, RTM_NEWMDB, br_mdb_add, NULL, NULL); rtnl_register(PF_BRIDGE, RTM_DELMDB, br_mdb_del, NULL, NULL); } + +void br_mdb_uninit(void) +{ + rtnl_unregister(PF_BRIDGE, RTM_GETMDB); + rtnl_unregister(PF_BRIDGE, RTM_NEWMDB); + rtnl_unregister(PF_BRIDGE, RTM_DELMDB); +} diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 1093c89095d..5391ca43336 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -279,7 +279,7 @@ static void br_multicast_port_group_expired(unsigned long data) spin_lock(&br->multicast_lock); if (!netif_running(br->dev) || timer_pending(&pg->timer) || - hlist_unhashed(&pg->mglist)) + hlist_unhashed(&pg->mglist) || pg->state & MDB_PERMANENT) goto out; br_multicast_del_pg(br, pg); @@ -622,7 +622,8 @@ out: struct net_bridge_port_group *br_multicast_new_port_group( struct net_bridge_port *port, struct br_ip *group, - struct net_bridge_port_group __rcu *next) + struct net_bridge_port_group __rcu *next, + unsigned char state) { struct net_bridge_port_group *p; @@ -632,6 +633,7 @@ struct net_bridge_port_group *br_multicast_new_port_group( p->addr = *group; p->port = port; + p->state = state; rcu_assign_pointer(p->next, next); hlist_add_head(&p->mglist, &port->mglist); setup_timer(&p->timer, br_multicast_port_group_expired, @@ -674,7 +676,7 @@ static int br_multicast_add_group(struct net_bridge *br, break; } - p = br_multicast_new_port_group(port, group, *pp); + p = br_multicast_new_port_group(port, group, *pp, MDB_TEMPORARY); if (unlikely(!p)) goto err; rcu_assign_pointer(*pp, p); @@ -1165,7 +1167,6 @@ static int br_ip6_multicast_query(struct net_bridge *br, if (max_delay) group = &mld->mld_mca; } else if (skb->len >= sizeof(*mld2q)) { - u16 mrc; if (!pskb_may_pull(skb, sizeof(*mld2q))) { err = -EINVAL; goto out; @@ -1173,8 +1174,7 @@ static int br_ip6_multicast_query(struct net_bridge *br, mld2q = (struct mld2_query *)icmp6_hdr(skb); if (!mld2q->mld2q_nsrcs) group = &mld2q->mld2q_mca; - mrc = ntohs(mld2q->mld2q_mrc); - max_delay = mrc ? MLDV2_MRC(mrc) : 1; + max_delay = mld2q->mld2q_mrc ? MLDV2_MRC(ntohs(mld2q->mld2q_mrc)) : 1; } if (!group) @@ -1633,6 +1633,7 @@ void br_multicast_stop(struct net_bridge *br) del_timer_sync(&br->multicast_querier_timer); del_timer_sync(&br->multicast_query_timer); + br_mdb_uninit(); spin_lock_bh(&br->multicast_lock); mdb = mlock_dereference(br->mdb, br); if (!mdb) diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index dead9dfe865..97ba0189c6f 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -305,5 +305,4 @@ int __init br_netlink_init(void) void __exit br_netlink_fini(void) { rtnl_link_unregister(&br_link_ops); - rtnl_unregister_all(PF_BRIDGE); } diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index f21a739a618..8d83be5ffed 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -83,6 +83,7 @@ struct net_bridge_port_group { struct rcu_head rcu; struct timer_list timer; struct br_ip addr; + unsigned char state; }; struct net_bridge_mdb_entry @@ -443,8 +444,10 @@ extern void br_multicast_free_pg(struct rcu_head *head); extern struct net_bridge_port_group *br_multicast_new_port_group( struct net_bridge_port *port, struct br_ip *group, - struct net_bridge_port_group *next); + struct net_bridge_port_group *next, + unsigned char state); extern void br_mdb_init(void); +extern void br_mdb_uninit(void); extern void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port, struct br_ip *group, int type); diff --git a/net/ceph/ceph_common.c b/net/ceph/ceph_common.c index a8020293f34..ee71ea26777 100644 --- a/net/ceph/ceph_common.c +++ b/net/ceph/ceph_common.c @@ -305,7 +305,6 @@ ceph_parse_options(char *options, const char *dev_name, /* start with defaults */ opt->flags = CEPH_OPT_DEFAULT; - opt->osd_timeout = CEPH_OSD_TIMEOUT_DEFAULT; opt->osd_keepalive_timeout = CEPH_OSD_KEEPALIVE_DEFAULT; opt->mount_timeout = CEPH_MOUNT_TIMEOUT_DEFAULT; /* seconds */ opt->osd_idle_ttl = CEPH_OSD_IDLE_TTL_DEFAULT; /* seconds */ @@ -391,7 +390,7 @@ ceph_parse_options(char *options, const char *dev_name, /* misc */ case Opt_osdtimeout: - opt->osd_timeout = intval; + pr_warning("ignoring deprecated osdtimeout option\n"); break; case Opt_osdkeepalivetimeout: opt->osd_keepalive_timeout = intval; diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c index 3ef1759403b..4d111fd2b49 100644 --- a/net/ceph/messenger.c +++ b/net/ceph/messenger.c @@ -2244,22 +2244,62 @@ bad_tag: /* - * Atomically queue work on a connection. Bump @con reference to - * avoid races with connection teardown. + * Atomically queue work on a connection after the specified delay. + * Bump @con reference to avoid races with connection teardown. + * Returns 0 if work was queued, or an error code otherwise. */ -static void queue_con(struct ceph_connection *con) +static int queue_con_delay(struct ceph_connection *con, unsigned long delay) { if (!con->ops->get(con)) { - dout("queue_con %p ref count 0\n", con); - return; + dout("%s %p ref count 0\n", __func__, con); + + return -ENOENT; } - if (!queue_delayed_work(ceph_msgr_wq, &con->work, 0)) { - dout("queue_con %p - already queued\n", con); + if (!queue_delayed_work(ceph_msgr_wq, &con->work, delay)) { + dout("%s %p - already queued\n", __func__, con); con->ops->put(con); - } else { - dout("queue_con %p\n", con); + + return -EBUSY; } + + dout("%s %p %lu\n", __func__, con, delay); + + return 0; +} + +static void queue_con(struct ceph_connection *con) +{ + (void) queue_con_delay(con, 0); +} + +static bool con_sock_closed(struct ceph_connection *con) +{ + if (!test_and_clear_bit(CON_FLAG_SOCK_CLOSED, &con->flags)) + return false; + +#define CASE(x) \ + case CON_STATE_ ## x: \ + con->error_msg = "socket closed (con state " #x ")"; \ + break; + + switch (con->state) { + CASE(CLOSED); + CASE(PREOPEN); + CASE(CONNECTING); + CASE(NEGOTIATING); + CASE(OPEN); + CASE(STANDBY); + default: + pr_warning("%s con %p unrecognized state %lu\n", + __func__, con, con->state); + con->error_msg = "unrecognized con state"; + BUG(); + break; + } +#undef CASE + + return true; } /* @@ -2273,35 +2313,16 @@ static void con_work(struct work_struct *work) mutex_lock(&con->mutex); restart: - if (test_and_clear_bit(CON_FLAG_SOCK_CLOSED, &con->flags)) { - switch (con->state) { - case CON_STATE_CONNECTING: - con->error_msg = "connection failed"; - break; - case CON_STATE_NEGOTIATING: - con->error_msg = "negotiation failed"; - break; - case CON_STATE_OPEN: - con->error_msg = "socket closed"; - break; - default: - dout("unrecognized con state %d\n", (int)con->state); - con->error_msg = "unrecognized con state"; - BUG(); - } + if (con_sock_closed(con)) goto fault; - } if (test_and_clear_bit(CON_FLAG_BACKOFF, &con->flags)) { dout("con_work %p backing off\n", con); - if (queue_delayed_work(ceph_msgr_wq, &con->work, - round_jiffies_relative(con->delay))) { - dout("con_work %p backoff %lu\n", con, con->delay); - mutex_unlock(&con->mutex); - return; - } else { + ret = queue_con_delay(con, round_jiffies_relative(con->delay)); + if (ret) { dout("con_work %p FAILED to back off %lu\n", con, con->delay); + BUG_ON(ret == -ENOENT); set_bit(CON_FLAG_BACKOFF, &con->flags); } goto done; @@ -2356,7 +2377,7 @@ fault: static void ceph_fault(struct ceph_connection *con) __releases(con->mutex) { - pr_err("%s%lld %s %s\n", ENTITY_NAME(con->peer_name), + pr_warning("%s%lld %s %s\n", ENTITY_NAME(con->peer_name), ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg); dout("fault %p state %lu to peer %s\n", con, con->state, ceph_pr_addr(&con->peer_addr.in_addr)); @@ -2398,24 +2419,8 @@ static void ceph_fault(struct ceph_connection *con) con->delay = BASE_DELAY_INTERVAL; else if (con->delay < MAX_DELAY_INTERVAL) con->delay *= 2; - con->ops->get(con); - if (queue_delayed_work(ceph_msgr_wq, &con->work, - round_jiffies_relative(con->delay))) { - dout("fault queued %p delay %lu\n", con, con->delay); - } else { - con->ops->put(con); - dout("fault failed to queue %p delay %lu, backoff\n", - con, con->delay); - /* - * In many cases we see a socket state change - * while con_work is running and end up - * queuing (non-delayed) work, such that we - * can't backoff with a delay. Set a flag so - * that when con_work restarts we schedule the - * delay then. - */ - set_bit(CON_FLAG_BACKOFF, &con->flags); - } + set_bit(CON_FLAG_BACKOFF, &con->flags); + queue_con(con); } out_unlock: diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index c1d756cc744..780caf6b049 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -221,6 +221,7 @@ struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc, kref_init(&req->r_kref); init_completion(&req->r_completion); init_completion(&req->r_safe_completion); + RB_CLEAR_NODE(&req->r_node); INIT_LIST_HEAD(&req->r_unsafe_item); INIT_LIST_HEAD(&req->r_linger_item); INIT_LIST_HEAD(&req->r_linger_osd); @@ -580,7 +581,7 @@ static void __kick_osd_requests(struct ceph_osd_client *osdc, dout("__kick_osd_requests osd%d\n", osd->o_osd); err = __reset_osd(osdc, osd); - if (err == -EAGAIN) + if (err) return; list_for_each_entry(req, &osd->o_requests, r_osd_item) { @@ -607,14 +608,6 @@ static void __kick_osd_requests(struct ceph_osd_client *osdc, } } -static void kick_osd_requests(struct ceph_osd_client *osdc, - struct ceph_osd *kickosd) -{ - mutex_lock(&osdc->request_mutex); - __kick_osd_requests(osdc, kickosd); - mutex_unlock(&osdc->request_mutex); -} - /* * If the osd connection drops, we need to resubmit all requests. */ @@ -628,7 +621,9 @@ static void osd_reset(struct ceph_connection *con) dout("osd_reset osd%d\n", osd->o_osd); osdc = osd->o_osdc; down_read(&osdc->map_sem); - kick_osd_requests(osdc, osd); + mutex_lock(&osdc->request_mutex); + __kick_osd_requests(osdc, osd); + mutex_unlock(&osdc->request_mutex); send_queued(osdc); up_read(&osdc->map_sem); } @@ -647,6 +642,7 @@ static struct ceph_osd *create_osd(struct ceph_osd_client *osdc, int onum) atomic_set(&osd->o_ref, 1); osd->o_osdc = osdc; osd->o_osd = onum; + RB_CLEAR_NODE(&osd->o_node); INIT_LIST_HEAD(&osd->o_requests); INIT_LIST_HEAD(&osd->o_linger_requests); INIT_LIST_HEAD(&osd->o_osd_lru); @@ -750,6 +746,7 @@ static int __reset_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd) if (list_empty(&osd->o_requests) && list_empty(&osd->o_linger_requests)) { __remove_osd(osdc, osd); + ret = -ENODEV; } else if (memcmp(&osdc->osdmap->osd_addr[osd->o_osd], &osd->o_con.peer_addr, sizeof(osd->o_con.peer_addr)) == 0 && @@ -876,9 +873,9 @@ static void __unregister_request(struct ceph_osd_client *osdc, req->r_osd = NULL; } + list_del_init(&req->r_req_lru_item); ceph_osdc_put_request(req); - list_del_init(&req->r_req_lru_item); if (osdc->num_requests == 0) { dout(" no requests, canceling timeout\n"); __cancel_osd_timeout(osdc); @@ -910,8 +907,8 @@ static void __unregister_linger_request(struct ceph_osd_client *osdc, struct ceph_osd_request *req) { dout("__unregister_linger_request %p\n", req); + list_del_init(&req->r_linger_item); if (req->r_osd) { - list_del_init(&req->r_linger_item); list_del_init(&req->r_linger_osd); if (list_empty(&req->r_osd->o_requests) && @@ -1090,12 +1087,10 @@ static void handle_timeout(struct work_struct *work) { struct ceph_osd_client *osdc = container_of(work, struct ceph_osd_client, timeout_work.work); - struct ceph_osd_request *req, *last_req = NULL; + struct ceph_osd_request *req; struct ceph_osd *osd; - unsigned long timeout = osdc->client->options->osd_timeout * HZ; unsigned long keepalive = osdc->client->options->osd_keepalive_timeout * HZ; - unsigned long last_stamp = 0; struct list_head slow_osds; dout("timeout\n"); down_read(&osdc->map_sem); @@ -1105,37 +1100,6 @@ static void handle_timeout(struct work_struct *work) mutex_lock(&osdc->request_mutex); /* - * reset osds that appear to be _really_ unresponsive. this - * is a failsafe measure.. we really shouldn't be getting to - * this point if the system is working properly. the monitors - * should mark the osd as failed and we should find out about - * it from an updated osd map. - */ - while (timeout && !list_empty(&osdc->req_lru)) { - req = list_entry(osdc->req_lru.next, struct ceph_osd_request, - r_req_lru_item); - - /* hasn't been long enough since we sent it? */ - if (time_before(jiffies, req->r_stamp + timeout)) - break; - - /* hasn't been long enough since it was acked? */ - if (req->r_request->ack_stamp == 0 || - time_before(jiffies, req->r_request->ack_stamp + timeout)) - break; - - BUG_ON(req == last_req && req->r_stamp == last_stamp); - last_req = req; - last_stamp = req->r_stamp; - - osd = req->r_osd; - BUG_ON(!osd); - pr_warning(" tid %llu timed out on osd%d, will reset osd\n", - req->r_tid, osd->o_osd); - __kick_osd_requests(osdc, osd); - } - - /* * ping osds that are a bit slow. this ensures that if there * is a break in the TCP connection we will notice, and reopen * a connection with that osd (from the fault callback). @@ -1364,8 +1328,8 @@ static void kick_requests(struct ceph_osd_client *osdc, int force_resend) dout("kicking lingering %p tid %llu osd%d\n", req, req->r_tid, req->r_osd ? req->r_osd->o_osd : -1); - __unregister_linger_request(osdc, req); __register_request(osdc, req); + __unregister_linger_request(osdc, req); } mutex_unlock(&osdc->request_mutex); @@ -1599,6 +1563,7 @@ int ceph_osdc_create_event(struct ceph_osd_client *osdc, event->data = data; event->osdc = osdc; INIT_LIST_HEAD(&event->osd_node); + RB_CLEAR_NODE(&event->node); kref_init(&event->kref); /* one ref for us */ kref_get(&event->kref); /* one ref for the caller */ init_completion(&event->completion); diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c index 5433fb0eb3c..de73214b5d2 100644 --- a/net/ceph/osdmap.c +++ b/net/ceph/osdmap.c @@ -469,6 +469,22 @@ static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, int id) return NULL; } +const char *ceph_pg_pool_name_by_id(struct ceph_osdmap *map, u64 id) +{ + struct ceph_pg_pool_info *pi; + + if (id == CEPH_NOPOOL) + return NULL; + + if (WARN_ON_ONCE(id > (u64) INT_MAX)) + return NULL; + + pi = __lookup_pg_pool(&map->pg_pools, (int) id); + + return pi ? pi->name : NULL; +} +EXPORT_SYMBOL(ceph_pg_pool_name_by_id); + int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name) { struct rb_node *rbp; @@ -645,10 +661,12 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end) ceph_decode_32_safe(p, end, max, bad); while (max--) { ceph_decode_need(p, end, 4 + 1 + sizeof(pi->v), bad); + err = -ENOMEM; pi = kzalloc(sizeof(*pi), GFP_NOFS); if (!pi) goto bad; pi->id = ceph_decode_32(p); + err = -EINVAL; ev = ceph_decode_8(p); /* encoding version */ if (ev > CEPH_PG_POOL_VERSION) { pr_warning("got unknown v %d > %d of ceph_pg_pool\n", @@ -664,8 +682,13 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end) __insert_pg_pool(&map->pg_pools, pi); } - if (version >= 5 && __decode_pool_names(p, end, map) < 0) - goto bad; + if (version >= 5) { + err = __decode_pool_names(p, end, map); + if (err < 0) { + dout("fail to decode pool names"); + goto bad; + } + } ceph_decode_32_safe(p, end, map->pool_max, bad); @@ -745,7 +768,7 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end) return map; bad: - dout("osdmap_decode fail\n"); + dout("osdmap_decode fail err %d\n", err); ceph_osdmap_destroy(map); return ERR_PTR(err); } @@ -839,6 +862,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, if (ev > CEPH_PG_POOL_VERSION) { pr_warning("got unknown v %d > %d of ceph_pg_pool\n", ev, CEPH_PG_POOL_VERSION); + err = -EINVAL; goto bad; } pi = __lookup_pg_pool(&map->pg_pools, pool); @@ -855,8 +879,11 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, if (err < 0) goto bad; } - if (version >= 5 && __decode_pool_names(p, end, map) < 0) - goto bad; + if (version >= 5) { + err = __decode_pool_names(p, end, map); + if (err < 0) + goto bad; + } /* old_pool */ ceph_decode_32_safe(p, end, len, bad); @@ -932,15 +959,13 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end, (void) __remove_pg_mapping(&map->pg_temp, pgid); /* insert */ - if (pglen > (UINT_MAX - sizeof(*pg)) / sizeof(u32)) { - err = -EINVAL; + err = -EINVAL; + if (pglen > (UINT_MAX - sizeof(*pg)) / sizeof(u32)) goto bad; - } + err = -ENOMEM; pg = kmalloc(sizeof(*pg) + sizeof(u32)*pglen, GFP_NOFS); - if (!pg) { - err = -ENOMEM; + if (!pg) goto bad; - } pg->pgid = pgid; pg->len = pglen; for (j = 0; j < pglen; j++) diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 6456439cbbd..8acce01b6da 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -381,6 +381,21 @@ struct net *get_net_ns_by_pid(pid_t pid) } EXPORT_SYMBOL_GPL(get_net_ns_by_pid); +static __net_init int net_ns_net_init(struct net *net) +{ + return proc_alloc_inum(&net->proc_inum); +} + +static __net_exit void net_ns_net_exit(struct net *net) +{ + proc_free_inum(net->proc_inum); +} + +static struct pernet_operations __net_initdata net_ns_ops = { + .init = net_ns_net_init, + .exit = net_ns_net_exit, +}; + static int __init net_ns_init(void) { struct net_generic *ng; @@ -412,6 +427,8 @@ static int __init net_ns_init(void) mutex_unlock(&net_mutex); + register_pernet_subsys(&net_ns_ops); + return 0; } @@ -630,16 +647,29 @@ static void netns_put(void *ns) static int netns_install(struct nsproxy *nsproxy, void *ns) { + struct net *net = ns; + + if (!ns_capable(net->user_ns, CAP_SYS_ADMIN) || + !nsown_capable(CAP_SYS_ADMIN)) + return -EPERM; + put_net(nsproxy->net_ns); - nsproxy->net_ns = get_net(ns); + nsproxy->net_ns = get_net(net); return 0; } +static unsigned int netns_inum(void *ns) +{ + struct net *net = ns; + return net->proc_inum; +} + const struct proc_ns_operations netns_operations = { .name = "net", .type = CLONE_NEWNET, .get = netns_get, .put = netns_put, .install = netns_install, + .inum = netns_inum, }; #endif diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 176ecdba4a2..4f9f5eb478f 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -439,8 +439,8 @@ exit: NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS); return NULL; put_and_exit: - bh_unlock_sock(newsk); - sock_put(newsk); + inet_csk_prepare_forced_close(newsk); + dccp_done(newsk); goto exit; } diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 56840b249f3..6e05981f271 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -585,7 +585,8 @@ static struct sock *dccp_v6_request_recv_sock(struct sock *sk, newinet->inet_rcv_saddr = LOOPBACK4_IPV6; if (__inet_inherit_port(sk, newsk) < 0) { - sock_put(newsk); + inet_csk_prepare_forced_close(newsk); + dccp_done(newsk); goto out; } __inet6_hash(newsk, NULL); diff --git a/net/dns_resolver/dns_key.c b/net/dns_resolver/dns_key.c index 8aa4b111538..0a69d075779 100644 --- a/net/dns_resolver/dns_key.c +++ b/net/dns_resolver/dns_key.c @@ -259,20 +259,16 @@ static int __init init_dns_resolver(void) if (!cred) return -ENOMEM; - keyring = key_alloc(&key_type_keyring, ".dns_resolver", - GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, cred, - (KEY_POS_ALL & ~KEY_POS_SETATTR) | - KEY_USR_VIEW | KEY_USR_READ, - KEY_ALLOC_NOT_IN_QUOTA); + keyring = keyring_alloc(".dns_resolver", + GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, cred, + (KEY_POS_ALL & ~KEY_POS_SETATTR) | + KEY_USR_VIEW | KEY_USR_READ, + KEY_ALLOC_NOT_IN_QUOTA, NULL); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); goto failed_put_cred; } - ret = key_instantiate_and_link(keyring, NULL, 0, NULL, NULL); - if (ret < 0) - goto failed_put_key; - ret = register_key_type(&key_type_dns_resolver); if (ret < 0) goto failed_put_key; @@ -304,3 +300,4 @@ static void __exit exit_dns_resolver(void) module_init(init_dns_resolver) module_exit(exit_dns_resolver) MODULE_LICENSE("GPL"); + diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 2026542d683..d0670f00d52 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -710,6 +710,22 @@ void inet_csk_destroy_sock(struct sock *sk) } EXPORT_SYMBOL(inet_csk_destroy_sock); +/* This function allows to force a closure of a socket after the call to + * tcp/dccp_create_openreq_child(). + */ +void inet_csk_prepare_forced_close(struct sock *sk) +{ + /* sk_clone_lock locked the socket and set refcnt to 2 */ + bh_unlock_sock(sk); + sock_put(sk); + + /* The below has to be done to allow calling inet_csk_destroy_sock */ + sock_set_flag(sk, SOCK_DEAD); + percpu_counter_inc(sk->sk_prot->orphan_count); + inet_sk(sk)->inet_num = 0; +} +EXPORT_SYMBOL(inet_csk_prepare_forced_close); + int inet_csk_listen_start(struct sock *sk, const int nr_table_entries) { struct inet_sock *inet = inet_sk(sk); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 1ed230716d5..54139fa514e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1767,10 +1767,8 @@ exit: NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS); return NULL; put_and_exit: - tcp_clear_xmit_timers(newsk); - tcp_cleanup_congestion_control(newsk); - bh_unlock_sock(newsk); - sock_put(newsk); + inet_csk_prepare_forced_close(newsk); + tcp_done(newsk); goto exit; } EXPORT_SYMBOL(tcp_v4_syn_recv_sock); diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index 2068ac4fbda..4ea244891b5 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -41,6 +41,6 @@ obj-$(CONFIG_IPV6_TUNNEL) += ip6_tunnel.o obj-$(CONFIG_IPV6_GRE) += ip6_gre.o obj-y += addrconf_core.o exthdrs_core.o -obj-$(CONFIG_INET) += output_core.o protocol.o $(ipv6_offload) +obj-$(CONFIG_INET) += output_core.o protocol.o $(ipv6-offload) obj-$(subst m,y,$(CONFIG_IPV6)) += inet6_hashtables.o diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 6fca01f136a..408cac4ae00 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -534,8 +534,7 @@ void inet6_netconf_notify_devconf(struct net *net, int type, int ifindex, rtnl_notify(skb, net, 0, RTNLGRP_IPV6_NETCONF, NULL, GFP_ATOMIC); return; errout: - if (err < 0) - rtnl_set_sk_err(net, RTNLGRP_IPV6_NETCONF, err); + rtnl_set_sk_err(net, RTNLGRP_IPV6_NETCONF, err); } static const struct nla_policy devconf_ipv6_policy[NETCONFA_MAX+1] = { diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index f2a007b7bde..6574175795d 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1314,6 +1314,12 @@ out: static void ndisc_redirect_rcv(struct sk_buff *skb) { + u8 *hdr; + struct ndisc_options ndopts; + struct rd_msg *msg = (struct rd_msg *)skb_transport_header(skb); + u32 ndoptlen = skb->tail - (skb->transport_header + + offsetof(struct rd_msg, opt)); + #ifdef CONFIG_IPV6_NDISC_NODETYPE switch (skb->ndisc_nodetype) { case NDISC_NODETYPE_HOST: @@ -1330,6 +1336,17 @@ static void ndisc_redirect_rcv(struct sk_buff *skb) return; } + if (!ndisc_parse_options(msg->opt, ndoptlen, &ndopts)) + return; + + if (!ndopts.nd_opts_rh) + return; + + hdr = (u8 *)ndopts.nd_opts_rh; + hdr += 8; + if (!pskb_pull(skb, hdr - skb_transport_header(skb))) + return; + icmpv6_notify(skb, NDISC_REDIRECT, 0, 0); } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 6565cf55eb1..93825dd3a7c 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1288,7 +1288,8 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, #endif if (__inet_inherit_port(sk, newsk) < 0) { - sock_put(newsk); + inet_csk_prepare_forced_close(newsk); + tcp_done(newsk); goto out; } __inet6_hash(newsk, NULL); diff --git a/net/mac802154/ieee802154_dev.c b/net/mac802154/ieee802154_dev.c index e748aed290a..b7c7f815dea 100644 --- a/net/mac802154/ieee802154_dev.c +++ b/net/mac802154/ieee802154_dev.c @@ -224,9 +224,9 @@ void ieee802154_free_device(struct ieee802154_dev *hw) BUG_ON(!list_empty(&priv->slaves)); - wpan_phy_free(priv->phy); - mutex_destroy(&priv->slaves_mtx); + + wpan_phy_free(priv->phy); } EXPORT_SYMBOL(ieee802154_free_device); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index c8a1eb6eca2..c0353d55d56 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -669,6 +669,9 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, struct sockaddr_nl *nladdr = (struct sockaddr_nl *)addr; int err; + if (addr_len < sizeof(struct sockaddr_nl)) + return -EINVAL; + if (nladdr->nl_family != AF_NETLINK) return -EINVAL; @@ -2059,7 +2062,7 @@ static int netlink_seq_show(struct seq_file *seq, void *v) struct sock *s = v; struct netlink_sock *nlk = nlk_sk(s); - seq_printf(seq, "%pK %-3d %-6d %08x %-8d %-8d %pK %-8d %-8d %-8lu\n", + seq_printf(seq, "%pK %-3d %-6u %08x %-8d %-8d %pK %-8d %-8d %-8lu\n", s, s->sk_protocol, nlk->portid, diff --git a/net/sctp/Kconfig b/net/sctp/Kconfig index a9edd2e205f..c26210618e1 100644 --- a/net/sctp/Kconfig +++ b/net/sctp/Kconfig @@ -66,12 +66,36 @@ config SCTP_DBG_OBJCNT 'cat /proc/net/sctp/sctp_dbg_objcnt' If unsure, say N +choice + prompt "Default SCTP cookie HMAC encoding" + default SCTP_COOKIE_HMAC_MD5 + help + This option sets the default sctp cookie hmac algorithm + when in doubt select 'md5' + +config SCTP_DEFAULT_COOKIE_HMAC_MD5 + bool "Enable optional MD5 hmac cookie generation" + help + Enable optional MD5 hmac based SCTP cookie generation + select SCTP_COOKIE_HMAC_MD5 + +config SCTP_DEFAULT_COOKIE_HMAC_SHA1 + bool "Enable optional SHA1 hmac cookie generation" + help + Enable optional SHA1 hmac based SCTP cookie generation + select SCTP_COOKIE_HMAC_SHA1 + +config SCTP_DEFAULT_COOKIE_HMAC_NONE + bool "Use no hmac alg in SCTP cookie generation" + help + Use no hmac algorithm in SCTP cookie generation + +endchoice config SCTP_COOKIE_HMAC_MD5 bool "Enable optional MD5 hmac cookie generation" help Enable optional MD5 hmac based SCTP cookie generation - default y select CRYPTO_HMAC if SCTP_COOKIE_HMAC_MD5 select CRYPTO_MD5 if SCTP_COOKIE_HMAC_MD5 @@ -79,7 +103,6 @@ config SCTP_COOKIE_HMAC_SHA1 bool "Enable optional SHA1 hmac cookie generation" help Enable optional SHA1 hmac based SCTP cookie generation - default y select CRYPTO_HMAC if SCTP_COOKIE_HMAC_SHA1 select CRYPTO_SHA1 if SCTP_COOKIE_HMAC_SHA1 diff --git a/net/sctp/probe.c b/net/sctp/probe.c index bc6cd75cc1d..5f7518de2fd 100644 --- a/net/sctp/probe.c +++ b/net/sctp/probe.c @@ -122,7 +122,8 @@ static const struct file_operations sctpprobe_fops = { .llseek = noop_llseek, }; -sctp_disposition_t jsctp_sf_eat_sack(const struct sctp_endpoint *ep, +sctp_disposition_t jsctp_sf_eat_sack(struct net *net, + const struct sctp_endpoint *ep, const struct sctp_association *asoc, const sctp_subtype_t type, void *arg, diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 2c7785bacf7..f898b1c58bd 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1191,9 +1191,9 @@ static int __net_init sctp_net_init(struct net *net) net->sctp.cookie_preserve_enable = 1; /* Default sctp sockets to use md5 as their hmac alg */ -#if defined (CONFIG_CRYPTO_MD5) +#if defined (CONFIG_SCTP_DEFAULT_COOKIE_HMAC_MD5) net->sctp.sctp_hmac_alg = "md5"; -#elif defined (CONFIG_CRYPTO_SHA1) +#elif defined (CONFIG_SCTP_DEFAULT_COOKIE_HMAC_SHA1) net->sctp.sctp_hmac_alg = "sha1"; #else net->sctp.sctp_hmac_alg = NULL; diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index 909dc0c31aa..6e5c824b040 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -192,17 +192,23 @@ gss_fill_context(const void *p, const void *end, struct gss_cl_ctx *ctx, struct const void *q; unsigned int seclen; unsigned int timeout; + unsigned long now = jiffies; u32 window_size; int ret; - /* First unsigned int gives the lifetime (in seconds) of the cred */ + /* First unsigned int gives the remaining lifetime in seconds of the + * credential - e.g. the remaining TGT lifetime for Kerberos or + * the -t value passed to GSSD. + */ p = simple_get_bytes(p, end, &timeout, sizeof(timeout)); if (IS_ERR(p)) goto err; if (timeout == 0) timeout = GSSD_MIN_TIMEOUT; - ctx->gc_expiry = jiffies + (unsigned long)timeout * HZ * 3 / 4; - /* Sequence number window. Determines the maximum number of simultaneous requests */ + ctx->gc_expiry = now + ((unsigned long)timeout * HZ); + /* Sequence number window. Determines the maximum number of + * simultaneous requests + */ p = simple_get_bytes(p, end, &window_size, sizeof(window_size)); if (IS_ERR(p)) goto err; @@ -237,9 +243,12 @@ gss_fill_context(const void *p, const void *end, struct gss_cl_ctx *ctx, struct p = ERR_PTR(ret); goto err; } + dprintk("RPC: %s Success. gc_expiry %lu now %lu timeout %u\n", + __func__, ctx->gc_expiry, now, timeout); return q; err: - dprintk("RPC: %s returning %ld\n", __func__, -PTR_ERR(p)); + dprintk("RPC: %s returns %ld gc_expiry %lu now %lu timeout %u\n", + __func__, -PTR_ERR(p), ctx->gc_expiry, now, timeout); return p; } diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c index a9c0bbccad6..890a29912d5 100644 --- a/net/sunrpc/backchannel_rqst.c +++ b/net/sunrpc/backchannel_rqst.c @@ -59,7 +59,7 @@ static void xprt_free_allocation(struct rpc_rqst *req) struct xdr_buf *xbufp; dprintk("RPC: free allocations for req= %p\n", req); - BUG_ON(test_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state)); + WARN_ON_ONCE(test_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state)); xbufp = &req->rq_private_buf; free_page((unsigned long)xbufp->head[0].iov_base); xbufp = &req->rq_snd_buf; @@ -191,7 +191,9 @@ void xprt_destroy_backchannel(struct rpc_xprt *xprt, unsigned int max_reqs) dprintk("RPC: destroy backchannel transport\n"); - BUG_ON(max_reqs == 0); + if (max_reqs == 0) + goto out; + spin_lock_bh(&xprt->bc_pa_lock); xprt_dec_alloc_count(xprt, max_reqs); list_for_each_entry_safe(req, tmp, &xprt->bc_pa_list, rq_bc_pa_list) { @@ -202,6 +204,7 @@ void xprt_destroy_backchannel(struct rpc_xprt *xprt, unsigned int max_reqs) } spin_unlock_bh(&xprt->bc_pa_lock); +out: dprintk("RPC: backchannel list empty= %s\n", list_empty(&xprt->bc_pa_list) ? "true" : "false"); } @@ -255,7 +258,7 @@ void xprt_free_bc_request(struct rpc_rqst *req) dprintk("RPC: free backchannel req=%p\n", req); smp_mb__before_clear_bit(); - BUG_ON(!test_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state)); + WARN_ON_ONCE(!test_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state)); clear_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state); smp_mb__after_clear_bit(); diff --git a/net/sunrpc/bc_svc.c b/net/sunrpc/bc_svc.c index 0b2eb388cbd..15c7a8a1c24 100644 --- a/net/sunrpc/bc_svc.c +++ b/net/sunrpc/bc_svc.c @@ -53,7 +53,7 @@ int bc_send(struct rpc_rqst *req) if (IS_ERR(task)) ret = PTR_ERR(task); else { - BUG_ON(atomic_read(&task->tk_count) != 1); + WARN_ON_ONCE(atomic_read(&task->tk_count) != 1); ret = task->tk_status; rpc_put_task(task); } diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index fc2f7aa4dca..9afa4393c21 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -775,11 +775,11 @@ static ssize_t cache_read(struct file *filp, char __user *buf, size_t count, if (rp->q.list.next == &cd->queue) { spin_unlock(&queue_lock); mutex_unlock(&inode->i_mutex); - BUG_ON(rp->offset); + WARN_ON_ONCE(rp->offset); return 0; } rq = container_of(rp->q.list.next, struct cache_request, q.list); - BUG_ON(rq->q.reader); + WARN_ON_ONCE(rq->q.reader); if (rp->offset == 0) rq->readers++; spin_unlock(&queue_lock); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index cdc7564b451..822f020fa7f 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -132,8 +132,10 @@ static struct dentry *rpc_setup_pipedir_sb(struct super_block *sb, int error; dir = rpc_d_lookup_sb(sb, dir_name); - if (dir == NULL) + if (dir == NULL) { + pr_info("RPC: pipefs directory doesn't exist: %s\n", dir_name); return dir; + } for (;;) { q.len = snprintf(name, sizeof(name), "clnt%x", (unsigned int)clntid++); name[sizeof(name) - 1] = '\0'; @@ -192,7 +194,8 @@ static int __rpc_clnt_handle_event(struct rpc_clnt *clnt, unsigned long event, case RPC_PIPEFS_MOUNT: dentry = rpc_setup_pipedir_sb(sb, clnt, clnt->cl_program->pipe_dir_name); - BUG_ON(dentry == NULL); + if (!dentry) + return -ENOENT; if (IS_ERR(dentry)) return PTR_ERR(dentry); clnt->cl_dentry = dentry; @@ -234,7 +237,7 @@ static struct rpc_clnt *rpc_get_client_for_event(struct net *net, int event) spin_lock(&sn->rpc_client_lock); list_for_each_entry(clnt, &sn->all_clients, cl_clients) { if (clnt->cl_program->pipe_dir_name == NULL) - break; + continue; if (rpc_clnt_skip_event(clnt, event)) continue; if (atomic_inc_not_zero(&clnt->cl_count) == 0) @@ -607,6 +610,13 @@ EXPORT_SYMBOL_GPL(rpc_killall_tasks); */ void rpc_shutdown_client(struct rpc_clnt *clnt) { + /* + * To avoid deadlock, never call rpc_shutdown_client from a + * workqueue context! + */ + WARN_ON_ONCE(current->flags & PF_WQ_WORKER); + might_sleep(); + dprintk_rcu("RPC: shutting down %s client for %s\n", clnt->cl_protname, rcu_dereference(clnt->cl_xprt)->servername); @@ -693,21 +703,19 @@ struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *old, const struct rpc_program *program, u32 vers) { + struct rpc_create_args args = { + .program = program, + .prognumber = program->number, + .version = vers, + .authflavor = old->cl_auth->au_flavor, + .client_name = old->cl_principal, + }; struct rpc_clnt *clnt; - const struct rpc_version *version; int err; - BUG_ON(vers >= program->nrvers || !program->version[vers]); - version = program->version[vers]; - clnt = rpc_clone_client(old); + clnt = __rpc_clone_client(&args, old); if (IS_ERR(clnt)) goto out; - clnt->cl_procinfo = version->procs; - clnt->cl_maxproc = version->nrprocs; - clnt->cl_protname = program->name; - clnt->cl_prog = program->number; - clnt->cl_vers = version->number; - clnt->cl_stats = program->stats; err = rpc_ping(clnt); if (err != 0) { rpc_shutdown_client(clnt); @@ -832,7 +840,12 @@ int rpc_call_sync(struct rpc_clnt *clnt, const struct rpc_message *msg, int flag }; int status; - BUG_ON(flags & RPC_TASK_ASYNC); + WARN_ON_ONCE(flags & RPC_TASK_ASYNC); + if (flags & RPC_TASK_ASYNC) { + rpc_release_calldata(task_setup_data.callback_ops, + task_setup_data.callback_data); + return -EINVAL; + } task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) @@ -908,7 +921,7 @@ struct rpc_task *rpc_run_bc_task(struct rpc_rqst *req, task->tk_action = call_bc_transmit; atomic_inc(&task->tk_count); - BUG_ON(atomic_read(&task->tk_count) != 2); + WARN_ON_ONCE(atomic_read(&task->tk_count) != 2); rpc_execute(task); out: @@ -1368,6 +1381,7 @@ call_refreshresult(struct rpc_task *task) return; case -ETIMEDOUT: rpc_delay(task, 3*HZ); + case -EKEYEXPIRED: case -EAGAIN: status = -EACCES; if (!task->tk_cred_retry) @@ -1654,7 +1668,6 @@ call_transmit(struct rpc_task *task) task->tk_action = call_transmit_status; /* Encode here so that rpcsec_gss can use correct sequence number. */ if (rpc_task_need_encode(task)) { - BUG_ON(task->tk_rqstp->rq_bytes_sent != 0); rpc_xdr_encode(task); /* Did the encode result in an error condition? */ if (task->tk_status != 0) { @@ -1738,7 +1751,6 @@ call_bc_transmit(struct rpc_task *task) { struct rpc_rqst *req = task->tk_rqstp; - BUG_ON(task->tk_status != 0); task->tk_status = xprt_prepare_transmit(task); if (task->tk_status == -EAGAIN) { /* @@ -1785,7 +1797,7 @@ call_bc_transmit(struct rpc_task *task) * We were unable to reply and will have to drop the * request. The server should reconnect and retransmit. */ - BUG_ON(task->tk_status == -EAGAIN); + WARN_ON_ONCE(task->tk_status == -EAGAIN); printk(KERN_NOTICE "RPC: Could not send backchannel reply " "error: %d\n", task->tk_status); break; diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 80f5dd23417..fd10981ea79 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -1093,7 +1093,7 @@ void rpc_put_sb_net(const struct net *net) { struct sunrpc_net *sn = net_generic(net, sunrpc_net_id); - BUG_ON(sn->pipefs_sb == NULL); + WARN_ON(sn->pipefs_sb == NULL); mutex_unlock(&sn->pipefs_sb_lock); } EXPORT_SYMBOL_GPL(rpc_put_sb_net); @@ -1152,14 +1152,19 @@ static void rpc_kill_sb(struct super_block *sb) struct sunrpc_net *sn = net_generic(net, sunrpc_net_id); mutex_lock(&sn->pipefs_sb_lock); + if (sn->pipefs_sb != sb) { + mutex_unlock(&sn->pipefs_sb_lock); + goto out; + } sn->pipefs_sb = NULL; mutex_unlock(&sn->pipefs_sb_lock); - put_net(net); dprintk("RPC: sending pipefs UMOUNT notification for net %p%s\n", net, NET_NAME(net)); blocking_notifier_call_chain(&rpc_pipefs_notifier_list, RPC_PIPEFS_UMOUNT, sb); + put_net(net); +out: kill_litter_super(sb); } diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index a70acae496e..795a0f4e920 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -23,7 +23,6 @@ #include <linux/errno.h> #include <linux/mutex.h> #include <linux/slab.h> -#include <linux/nsproxy.h> #include <net/ipv6.h> #include <linux/sunrpc/clnt.h> @@ -884,7 +883,10 @@ static void encode_rpcb_string(struct xdr_stream *xdr, const char *string, u32 len; len = strlen(string); - BUG_ON(len > maxstrlen); + WARN_ON_ONCE(len > maxstrlen); + if (len > maxstrlen) + /* truncate and hope for the best */ + len = maxstrlen; p = xdr_reserve_space(xdr, 4 + len); xdr_encode_opaque(p, string, len); } diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index 6357fcb00c7..d17a704aaf5 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -98,6 +98,23 @@ __rpc_add_timer(struct rpc_wait_queue *queue, struct rpc_task *task) list_add(&task->u.tk_wait.timer_list, &queue->timer_list.list); } +static void rpc_set_waitqueue_priority(struct rpc_wait_queue *queue, int priority) +{ + queue->priority = priority; +} + +static void rpc_set_waitqueue_owner(struct rpc_wait_queue *queue, pid_t pid) +{ + queue->owner = pid; + queue->nr = RPC_BATCH_COUNT; +} + +static void rpc_reset_waitqueue_priority(struct rpc_wait_queue *queue) +{ + rpc_set_waitqueue_priority(queue, queue->maxpriority); + rpc_set_waitqueue_owner(queue, 0); +} + /* * Add new request to a priority queue. */ @@ -109,9 +126,11 @@ static void __rpc_add_wait_queue_priority(struct rpc_wait_queue *queue, struct rpc_task *t; INIT_LIST_HEAD(&task->u.tk_wait.links); - q = &queue->tasks[queue_priority]; if (unlikely(queue_priority > queue->maxpriority)) - q = &queue->tasks[queue->maxpriority]; + queue_priority = queue->maxpriority; + if (queue_priority > queue->priority) + rpc_set_waitqueue_priority(queue, queue_priority); + q = &queue->tasks[queue_priority]; list_for_each_entry(t, q, u.tk_wait.list) { if (t->tk_owner == task->tk_owner) { list_add_tail(&task->u.tk_wait.list, &t->u.tk_wait.links); @@ -133,7 +152,9 @@ static void __rpc_add_wait_queue(struct rpc_wait_queue *queue, struct rpc_task *task, unsigned char queue_priority) { - BUG_ON (RPC_IS_QUEUED(task)); + WARN_ON_ONCE(RPC_IS_QUEUED(task)); + if (RPC_IS_QUEUED(task)) + return; if (RPC_IS_PRIORITY(queue)) __rpc_add_wait_queue_priority(queue, task, queue_priority); @@ -178,24 +199,6 @@ static void __rpc_remove_wait_queue(struct rpc_wait_queue *queue, struct rpc_tas task->tk_pid, queue, rpc_qname(queue)); } -static inline void rpc_set_waitqueue_priority(struct rpc_wait_queue *queue, int priority) -{ - queue->priority = priority; - queue->count = 1 << (priority * 2); -} - -static inline void rpc_set_waitqueue_owner(struct rpc_wait_queue *queue, pid_t pid) -{ - queue->owner = pid; - queue->nr = RPC_BATCH_COUNT; -} - -static inline void rpc_reset_waitqueue_priority(struct rpc_wait_queue *queue) -{ - rpc_set_waitqueue_priority(queue, queue->maxpriority); - rpc_set_waitqueue_owner(queue, 0); -} - static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qname, unsigned char nr_queues) { int i; @@ -334,7 +337,7 @@ static void __rpc_sleep_on_priority(struct rpc_wait_queue *q, __rpc_add_wait_queue(q, task, queue_priority); - BUG_ON(task->tk_callback != NULL); + WARN_ON_ONCE(task->tk_callback != NULL); task->tk_callback = action; __rpc_add_timer(q, task); } @@ -343,7 +346,12 @@ void rpc_sleep_on(struct rpc_wait_queue *q, struct rpc_task *task, rpc_action action) { /* We shouldn't ever put an inactive task to sleep */ - BUG_ON(!RPC_IS_ACTIVATED(task)); + WARN_ON_ONCE(!RPC_IS_ACTIVATED(task)); + if (!RPC_IS_ACTIVATED(task)) { + task->tk_status = -EIO; + rpc_put_task_async(task); + return; + } /* * Protect the queue operations. @@ -358,7 +366,12 @@ void rpc_sleep_on_priority(struct rpc_wait_queue *q, struct rpc_task *task, rpc_action action, int priority) { /* We shouldn't ever put an inactive task to sleep */ - BUG_ON(!RPC_IS_ACTIVATED(task)); + WARN_ON_ONCE(!RPC_IS_ACTIVATED(task)); + if (!RPC_IS_ACTIVATED(task)) { + task->tk_status = -EIO; + rpc_put_task_async(task); + return; + } /* * Protect the queue operations. @@ -367,6 +380,7 @@ void rpc_sleep_on_priority(struct rpc_wait_queue *q, struct rpc_task *task, __rpc_sleep_on_priority(q, task, action, priority - RPC_PRIORITY_LOW); spin_unlock_bh(&q->lock); } +EXPORT_SYMBOL_GPL(rpc_sleep_on_priority); /** * __rpc_do_wake_up_task - wake up a single rpc_task @@ -451,8 +465,7 @@ static struct rpc_task *__rpc_find_next_queued_priority(struct rpc_wait_queue *q /* * Check if we need to switch queues. */ - if (--queue->count) - goto new_owner; + goto new_owner; } /* @@ -697,7 +710,9 @@ static void __rpc_execute(struct rpc_task *task) dprintk("RPC: %5u __rpc_execute flags=0x%x\n", task->tk_pid, task->tk_flags); - BUG_ON(RPC_IS_QUEUED(task)); + WARN_ON_ONCE(RPC_IS_QUEUED(task)); + if (RPC_IS_QUEUED(task)) + return; for (;;) { void (*do_action)(struct rpc_task *); @@ -981,7 +996,7 @@ static void rpc_release_task(struct rpc_task *task) { dprintk("RPC: %5u release task\n", task->tk_pid); - BUG_ON (RPC_IS_QUEUED(task)); + WARN_ON_ONCE(RPC_IS_QUEUED(task)); rpc_release_resources_task(task); diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 3ee7461926d..dbf12ac5ecb 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -20,7 +20,6 @@ #include <linux/module.h> #include <linux/kthread.h> #include <linux/slab.h> -#include <linux/nsproxy.h> #include <linux/sunrpc/types.h> #include <linux/sunrpc/xdr.h> @@ -324,7 +323,9 @@ svc_pool_map_set_cpumask(struct task_struct *task, unsigned int pidx) * The caller checks for sv_nrpools > 1, which * implies that we've been initialized. */ - BUG_ON(m->count == 0); + WARN_ON_ONCE(m->count == 0); + if (m->count == 0) + return; switch (m->mode) { case SVC_POOL_PERCPU: @@ -585,7 +586,9 @@ svc_init_buffer(struct svc_rqst *rqstp, unsigned int size, int node) * We assume one is at most one page */ arghi = 0; - BUG_ON(pages > RPCSVC_MAXPAGES); + WARN_ON_ONCE(pages > RPCSVC_MAXPAGES); + if (pages > RPCSVC_MAXPAGES) + pages = RPCSVC_MAXPAGES; while (pages) { struct page *p = alloc_pages_node(node, GFP_KERNEL, 0); if (!p) @@ -946,7 +949,9 @@ int svc_register(const struct svc_serv *serv, struct net *net, unsigned int i; int error = 0; - BUG_ON(proto == 0 && port == 0); + WARN_ON_ONCE(proto == 0 && port == 0); + if (proto == 0 && port == 0) + return -EINVAL; for (progp = serv->sv_program; progp; progp = progp->pg_next) { for (i = 0; i < progp->pg_nvers; i++) { @@ -1035,7 +1040,7 @@ static void svc_unregister(const struct svc_serv *serv, struct net *net) } /* - * Printk the given error with the address of the client that caused it. + * dprintk the given error with the address of the client that caused it. */ static __printf(2, 3) void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...) @@ -1049,8 +1054,7 @@ void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...) vaf.fmt = fmt; vaf.va = &args; - net_warn_ratelimited("svc: %s: %pV", - svc_print_addr(rqstp, buf, sizeof(buf)), &vaf); + dprintk("svc: %s: %pV", svc_print_addr(rqstp, buf, sizeof(buf)), &vaf); va_end(args); } @@ -1299,7 +1303,7 @@ svc_process(struct svc_rqst *rqstp) * Setup response xdr_buf. * Initially it has just one page */ - rqstp->rq_resused = 1; + rqstp->rq_next_page = &rqstp->rq_respages[1]; resv->iov_base = page_address(rqstp->rq_respages[0]); resv->iov_len = 0; rqstp->rq_res.pages = rqstp->rq_respages + 1; diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 194d865fae7..b8e47fac731 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -218,7 +218,9 @@ static struct svc_xprt *__svc_xpo_create(struct svc_xprt_class *xcl, */ static void svc_xprt_received(struct svc_xprt *xprt) { - BUG_ON(!test_bit(XPT_BUSY, &xprt->xpt_flags)); + WARN_ON_ONCE(!test_bit(XPT_BUSY, &xprt->xpt_flags)); + if (!test_bit(XPT_BUSY, &xprt->xpt_flags)) + return; /* As soon as we clear busy, the xprt could be closed and * 'put', so we need a reference to call svc_xprt_enqueue with: */ @@ -577,7 +579,10 @@ int svc_alloc_arg(struct svc_rqst *rqstp) /* now allocate needed pages. If we get a failure, sleep briefly */ pages = (serv->sv_max_mesg + PAGE_SIZE) / PAGE_SIZE; - BUG_ON(pages >= RPCSVC_MAXPAGES); + WARN_ON_ONCE(pages >= RPCSVC_MAXPAGES); + if (pages >= RPCSVC_MAXPAGES) + /* use as many pages as possible */ + pages = RPCSVC_MAXPAGES - 1; for (i = 0; i < pages ; i++) while (rqstp->rq_pages[i] == NULL) { struct page *p = alloc_page(GFP_KERNEL); @@ -926,7 +931,7 @@ static void svc_delete_xprt(struct svc_xprt *xprt) spin_lock_bh(&serv->sv_lock); if (!test_and_set_bit(XPT_DETACHED, &xprt->xpt_flags)) list_del_init(&xprt->xpt_list); - BUG_ON(!list_empty(&xprt->xpt_ready)); + WARN_ON_ONCE(!list_empty(&xprt->xpt_ready)); if (test_bit(XPT_TEMP, &xprt->xpt_flags)) serv->sv_tmpcnt--; spin_unlock_bh(&serv->sv_lock); diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 03827cef1fa..0a148c9d2a5 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -84,7 +84,11 @@ static struct lock_class_key svc_slock_key[2]; static void svc_reclassify_socket(struct socket *sock) { struct sock *sk = sock->sk; - BUG_ON(sock_owned_by_user(sk)); + + WARN_ON_ONCE(sock_owned_by_user(sk)); + if (sock_owned_by_user(sk)) + return; + switch (sk->sk_family) { case AF_INET: sock_lock_init_class_and_name(sk, "slock-AF_INET-NFSD", @@ -601,6 +605,7 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp) rqstp->rq_respages = rqstp->rq_pages + 1 + DIV_ROUND_UP(rqstp->rq_arg.page_len, PAGE_SIZE); } + rqstp->rq_next_page = rqstp->rq_respages+1; if (serv->sv_stats) serv->sv_stats->netudpcnt++; @@ -874,9 +879,9 @@ static unsigned int svc_tcp_restore_pages(struct svc_sock *svsk, struct svc_rqst { unsigned int i, len, npages; - if (svsk->sk_tcplen <= sizeof(rpc_fraghdr)) + if (svsk->sk_datalen == 0) return 0; - len = svsk->sk_tcplen - sizeof(rpc_fraghdr); + len = svsk->sk_datalen; npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; for (i = 0; i < npages; i++) { if (rqstp->rq_pages[i] != NULL) @@ -893,9 +898,9 @@ static void svc_tcp_save_pages(struct svc_sock *svsk, struct svc_rqst *rqstp) { unsigned int i, len, npages; - if (svsk->sk_tcplen <= sizeof(rpc_fraghdr)) + if (svsk->sk_datalen == 0) return; - len = svsk->sk_tcplen - sizeof(rpc_fraghdr); + len = svsk->sk_datalen; npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; for (i = 0; i < npages; i++) { svsk->sk_pages[i] = rqstp->rq_pages[i]; @@ -907,9 +912,9 @@ static void svc_tcp_clear_pages(struct svc_sock *svsk) { unsigned int i, len, npages; - if (svsk->sk_tcplen <= sizeof(rpc_fraghdr)) + if (svsk->sk_datalen == 0) goto out; - len = svsk->sk_tcplen - sizeof(rpc_fraghdr); + len = svsk->sk_datalen; npages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; for (i = 0; i < npages; i++) { BUG_ON(svsk->sk_pages[i] == NULL); @@ -918,13 +923,12 @@ static void svc_tcp_clear_pages(struct svc_sock *svsk) } out: svsk->sk_tcplen = 0; + svsk->sk_datalen = 0; } /* - * Receive data. + * Receive fragment record header. * If we haven't gotten the record length yet, get the next four bytes. - * Otherwise try to gobble up as much as possible up to the complete - * record length. */ static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp) { @@ -950,32 +954,16 @@ static int svc_tcp_recv_record(struct svc_sock *svsk, struct svc_rqst *rqstp) return -EAGAIN; } - svsk->sk_reclen = ntohl(svsk->sk_reclen); - if (!(svsk->sk_reclen & RPC_LAST_STREAM_FRAGMENT)) { - /* FIXME: technically, a record can be fragmented, - * and non-terminal fragments will not have the top - * bit set in the fragment length header. - * But apparently no known nfs clients send fragmented - * records. */ - net_notice_ratelimited("RPC: multiple fragments per record not supported\n"); - goto err_delete; - } - - svsk->sk_reclen &= RPC_FRAGMENT_SIZE_MASK; - dprintk("svc: TCP record, %d bytes\n", svsk->sk_reclen); - if (svsk->sk_reclen > serv->sv_max_mesg) { - net_notice_ratelimited("RPC: fragment too large: 0x%08lx\n", - (unsigned long)svsk->sk_reclen); + dprintk("svc: TCP record, %d bytes\n", svc_sock_reclen(svsk)); + if (svc_sock_reclen(svsk) + svsk->sk_datalen > + serv->sv_max_mesg) { + net_notice_ratelimited("RPC: fragment too large: %d\n", + svc_sock_reclen(svsk)); goto err_delete; } } - if (svsk->sk_reclen < 8) - goto err_delete; /* client is nuts. */ - - len = svsk->sk_reclen; - - return len; + return svc_sock_reclen(svsk); error: dprintk("RPC: TCP recv_record got %d\n", len); return len; @@ -1019,7 +1007,7 @@ static int receive_cb_reply(struct svc_sock *svsk, struct svc_rqst *rqstp) if (dst->iov_len < src->iov_len) return -EAGAIN; /* whatever; just giving up. */ memcpy(dst->iov_base, src->iov_base, src->iov_len); - xprt_complete_rqst(req->rq_task, svsk->sk_reclen); + xprt_complete_rqst(req->rq_task, rqstp->rq_arg.len); rqstp->rq_arg.len = 0; return 0; } @@ -1038,6 +1026,17 @@ static int copy_pages_to_kvecs(struct kvec *vec, struct page **pages, int len) return i; } +static void svc_tcp_fragment_received(struct svc_sock *svsk) +{ + /* If we have more data, signal svc_xprt_enqueue() to try again */ + if (svc_recv_available(svsk) > sizeof(rpc_fraghdr)) + set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags); + dprintk("svc: TCP %s record (%d bytes)\n", + svc_sock_final_rec(svsk) ? "final" : "nonfinal", + svc_sock_reclen(svsk)); + svsk->sk_tcplen = 0; + svsk->sk_reclen = 0; +} /* * Receive data from a TCP socket. @@ -1064,29 +1063,39 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp) goto error; base = svc_tcp_restore_pages(svsk, rqstp); - want = svsk->sk_reclen - base; + want = svc_sock_reclen(svsk) - (svsk->sk_tcplen - sizeof(rpc_fraghdr)); vec = rqstp->rq_vec; pnum = copy_pages_to_kvecs(&vec[0], &rqstp->rq_pages[0], - svsk->sk_reclen); + svsk->sk_datalen + want); rqstp->rq_respages = &rqstp->rq_pages[pnum]; + rqstp->rq_next_page = rqstp->rq_respages + 1; /* Now receive data */ len = svc_partial_recvfrom(rqstp, vec, pnum, want, base); - if (len >= 0) + if (len >= 0) { svsk->sk_tcplen += len; - if (len != want) { + svsk->sk_datalen += len; + } + if (len != want || !svc_sock_final_rec(svsk)) { svc_tcp_save_pages(svsk, rqstp); if (len < 0 && len != -EAGAIN) - goto err_other; - dprintk("svc: incomplete TCP record (%d of %d)\n", - svsk->sk_tcplen, svsk->sk_reclen); + goto err_delete; + if (len == want) + svc_tcp_fragment_received(svsk); + else + dprintk("svc: incomplete TCP record (%d of %d)\n", + (int)(svsk->sk_tcplen - sizeof(rpc_fraghdr)), + svc_sock_reclen(svsk)); goto err_noclose; } - rqstp->rq_arg.len = svsk->sk_reclen; + if (svc_sock_reclen(svsk) < 8) + goto err_delete; /* client is nuts. */ + + rqstp->rq_arg.len = svsk->sk_datalen; rqstp->rq_arg.page_base = 0; if (rqstp->rq_arg.len <= rqstp->rq_arg.head[0].iov_len) { rqstp->rq_arg.head[0].iov_len = rqstp->rq_arg.len; @@ -1103,11 +1112,8 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp) len = receive_cb_reply(svsk, rqstp); /* Reset TCP read info */ - svsk->sk_reclen = 0; - svsk->sk_tcplen = 0; - /* If we have more data, signal svc_xprt_enqueue() to try again */ - if (svc_recv_available(svsk) > sizeof(rpc_fraghdr)) - set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags); + svsk->sk_datalen = 0; + svc_tcp_fragment_received(svsk); if (len < 0) goto error; @@ -1116,15 +1122,14 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp) if (serv->sv_stats) serv->sv_stats->nettcpcnt++; - dprintk("svc: TCP complete record (%d bytes)\n", rqstp->rq_arg.len); return rqstp->rq_arg.len; error: if (len != -EAGAIN) - goto err_other; + goto err_delete; dprintk("RPC: TCP recvfrom got EAGAIN\n"); return 0; -err_other: +err_delete: printk(KERN_NOTICE "%s: recvfrom returned errno %d\n", svsk->sk_xprt.xpt_server->sv_name, -len); set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags); @@ -1301,6 +1306,7 @@ static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv) svsk->sk_reclen = 0; svsk->sk_tcplen = 0; + svsk->sk_datalen = 0; memset(&svsk->sk_pages[0], 0, sizeof(svsk->sk_pages)); tcp_sk(sk)->nonagle |= TCP_NAGLE_OFF; diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c index 08f50afd5f2..56055632f15 100644 --- a/net/sunrpc/xdr.c +++ b/net/sunrpc/xdr.c @@ -318,7 +318,10 @@ xdr_shrink_bufhead(struct xdr_buf *buf, size_t len) tail = buf->tail; head = buf->head; - BUG_ON (len > head->iov_len); + + WARN_ON_ONCE(len > head->iov_len); + if (len > head->iov_len) + len = head->iov_len; /* Shift the tail first */ if (tail->iov_len != 0) { diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c index 41cb63b623d..0ce75524ed2 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c +++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c @@ -521,11 +521,11 @@ next_sge: rqstp->rq_pages[ch_no] = NULL; /* - * Detach res pages. svc_release must see a resused count of - * zero or it will attempt to put them. + * Detach res pages. If svc_release sees any it will attempt to + * put them. */ - while (rqstp->rq_resused) - rqstp->rq_respages[--rqstp->rq_resused] = NULL; + while (rqstp->rq_next_page != rqstp->rq_respages) + *(--rqstp->rq_next_page) = NULL; return err; } @@ -550,7 +550,7 @@ static int rdma_read_complete(struct svc_rqst *rqstp, /* rq_respages starts after the last arg page */ rqstp->rq_respages = &rqstp->rq_arg.pages[page_no]; - rqstp->rq_resused = 0; + rqstp->rq_next_page = &rqstp->rq_arg.pages[page_no]; /* Rebuild rq_arg head and tail. */ rqstp->rq_arg.head[0] = head->arg.head[0]; diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c index 42eb7ba0b90..c1d124dc772 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c +++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c @@ -548,6 +548,7 @@ static int send_reply(struct svcxprt_rdma *rdma, int sge_no; int sge_bytes; int page_no; + int pages; int ret; /* Post a recv buffer to handle another request. */ @@ -611,7 +612,8 @@ static int send_reply(struct svcxprt_rdma *rdma, * respages array. They are our pages until the I/O * completes. */ - for (page_no = 0; page_no < rqstp->rq_resused; page_no++) { + pages = rqstp->rq_next_page - rqstp->rq_respages; + for (page_no = 0; page_no < pages; page_no++) { ctxt->pages[page_no+1] = rqstp->rq_respages[page_no]; ctxt->count++; rqstp->rq_respages[page_no] = NULL; diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 75853cabf4c..68b0a81c31d 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1746,7 +1746,6 @@ static inline void xs_reclassify_socketu(struct socket *sock) { struct sock *sk = sock->sk; - BUG_ON(sock_owned_by_user(sk)); sock_lock_init_class_and_name(sk, "slock-AF_LOCAL-RPC", &xs_slock_key[1], "sk_lock-AF_LOCAL-RPC", &xs_key[1]); } @@ -1755,7 +1754,6 @@ static inline void xs_reclassify_socket4(struct socket *sock) { struct sock *sk = sock->sk; - BUG_ON(sock_owned_by_user(sk)); sock_lock_init_class_and_name(sk, "slock-AF_INET-RPC", &xs_slock_key[0], "sk_lock-AF_INET-RPC", &xs_key[0]); } @@ -1764,13 +1762,16 @@ static inline void xs_reclassify_socket6(struct socket *sock) { struct sock *sk = sock->sk; - BUG_ON(sock_owned_by_user(sk)); sock_lock_init_class_and_name(sk, "slock-AF_INET6-RPC", &xs_slock_key[1], "sk_lock-AF_INET6-RPC", &xs_key[1]); } static inline void xs_reclassify_socket(int family, struct socket *sock) { + WARN_ON_ONCE(sock_owned_by_user(sock->sk)); + if (sock_owned_by_user(sock->sk)) + return; + switch (family) { case AF_LOCAL: xs_reclassify_socketu(sock); @@ -1901,6 +1902,10 @@ static void xs_local_setup_socket(struct work_struct *work) dprintk("RPC: xprt %p: socket %s does not exist\n", xprt, xprt->address_strings[RPC_DISPLAY_ADDR]); break; + case -ECONNREFUSED: + dprintk("RPC: xprt %p: connection refused for %s\n", + xprt, xprt->address_strings[RPC_DISPLAY_ADDR]); + break; default: printk(KERN_ERR "%s: unhandled error (%d) connecting to %s\n", __func__, -status, @@ -2329,9 +2334,11 @@ static void *bc_malloc(struct rpc_task *task, size_t size) struct page *page; struct rpc_buffer *buf; - BUG_ON(size > PAGE_SIZE - sizeof(struct rpc_buffer)); - page = alloc_page(GFP_KERNEL); + WARN_ON_ONCE(size > PAGE_SIZE - sizeof(struct rpc_buffer)); + if (size > PAGE_SIZE - sizeof(struct rpc_buffer)) + return NULL; + page = alloc_page(GFP_KERNEL); if (!page) return NULL; @@ -2393,7 +2400,6 @@ static int bc_send_request(struct rpc_task *task) { struct rpc_rqst *req = task->tk_rqstp; struct svc_xprt *xprt; - struct svc_sock *svsk; u32 len; dprintk("sending request with xid: %08x\n", ntohl(req->rq_xid)); @@ -2401,7 +2407,6 @@ static int bc_send_request(struct rpc_task *task) * Get the server socket associated with this callback xprt */ xprt = req->rq_xprt->bc_xprt; - svsk = container_of(xprt, struct svc_sock, sk_xprt); /* * Grab the mutex to serialize data as the connection is shared |