diff options
Diffstat (limited to 'net/tipc')
43 files changed, 15021 insertions, 0 deletions
diff --git a/net/tipc/Kconfig b/net/tipc/Kconfig new file mode 100644 index 00000000000..c890848f9d5 --- /dev/null +++ b/net/tipc/Kconfig @@ -0,0 +1,40 @@ +# +# TIPC configuration +# + +menuconfig TIPC + tristate "The TIPC Protocol" + depends on INET + ---help--- + The Transparent Inter Process Communication (TIPC) protocol is + specially designed for intra cluster communication. This protocol + originates from Ericsson where it has been used in carrier grade + cluster applications for many years. + + For more information about TIPC, see http://tipc.sourceforge.net. + + This protocol support is also available as a module ( = code which + can be inserted in and removed from the running kernel whenever you + want). The module will be called tipc. If you want to compile it + as a module, say M here and read <file:Documentation/kbuild/modules.txt>. + + If in doubt, say N. + +config TIPC_PORTS + int "Maximum number of ports in a node" + depends on TIPC + range 127 65535 + default "8191" + help + Specifies how many ports can be supported by a node. + Can range from 127 to 65535 ports; default is 8191. + + Setting this to a smaller value saves some memory, + setting it to higher allows for more ports. + +config TIPC_MEDIA_IB + bool "InfiniBand media type support" + depends on TIPC && INFINIBAND_IPOIB + help + Saying Y here will enable support for running TIPC on + IP-over-InfiniBand devices. diff --git a/net/tipc/Makefile b/net/tipc/Makefile new file mode 100644 index 00000000000..a080c66d819 --- /dev/null +++ b/net/tipc/Makefile @@ -0,0 +1,14 @@ +# +# Makefile for the Linux TIPC layer +# + +obj-$(CONFIG_TIPC) := tipc.o + +tipc-y += addr.o bcast.o bearer.o config.o \ + core.o link.o discover.o msg.o \ + name_distr.o subscr.o name_table.o net.o \ + netlink.o node.o node_subscr.o port.o ref.o \ + socket.o log.o eth_media.o server.o + +tipc-$(CONFIG_TIPC_MEDIA_IB) += ib_media.o +tipc-$(CONFIG_SYSCTL) += sysctl.o diff --git a/net/tipc/addr.c b/net/tipc/addr.c new file mode 100644 index 00000000000..357b74b26f9 --- /dev/null +++ b/net/tipc/addr.c @@ -0,0 +1,103 @@ +/* + * net/tipc/addr.c: TIPC address utility routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2004-2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "addr.h" + +/** + * tipc_addr_domain_valid - validates a network domain address + * + * Accepts <Z.C.N>, <Z.C.0>, <Z.0.0>, and <0.0.0>, + * where Z, C, and N are non-zero. + * + * Returns 1 if domain address is valid, otherwise 0 + */ +int tipc_addr_domain_valid(u32 addr) +{ + u32 n = tipc_node(addr); + u32 c = tipc_cluster(addr); + u32 z = tipc_zone(addr); + + if (n && (!z || !c)) + return 0; + if (c && !z) + return 0; + return 1; +} + +/** + * tipc_addr_node_valid - validates a proposed network address for this node + * + * Accepts <Z.C.N>, where Z, C, and N are non-zero. + * + * Returns 1 if address can be used, otherwise 0 + */ +int tipc_addr_node_valid(u32 addr) +{ + return tipc_addr_domain_valid(addr) && tipc_node(addr); +} + +int tipc_in_scope(u32 domain, u32 addr) +{ + if (!domain || (domain == addr)) + return 1; + if (domain == tipc_cluster_mask(addr)) /* domain <Z.C.0> */ + return 1; + if (domain == tipc_zone_mask(addr)) /* domain <Z.0.0> */ + return 1; + return 0; +} + +/** + * tipc_addr_scope - convert message lookup domain to a 2-bit scope value + */ +int tipc_addr_scope(u32 domain) +{ + if (likely(!domain)) + return TIPC_ZONE_SCOPE; + if (tipc_node(domain)) + return TIPC_NODE_SCOPE; + if (tipc_cluster(domain)) + return TIPC_CLUSTER_SCOPE; + return TIPC_ZONE_SCOPE; +} + +char *tipc_addr_string_fill(char *string, u32 addr) +{ + snprintf(string, 16, "<%u.%u.%u>", + tipc_zone(addr), tipc_cluster(addr), tipc_node(addr)); + return string; +} diff --git a/net/tipc/addr.h b/net/tipc/addr.h new file mode 100644 index 00000000000..a74acf9ee80 --- /dev/null +++ b/net/tipc/addr.h @@ -0,0 +1,96 @@ +/* + * net/tipc/addr.h: Include file for TIPC address utility routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_ADDR_H +#define _TIPC_ADDR_H + +#include "core.h" + +#define TIPC_ZONE_MASK 0xff000000u +#define TIPC_CLUSTER_MASK 0xfffff000u + +static inline u32 tipc_zone_mask(u32 addr) +{ + return addr & TIPC_ZONE_MASK; +} + +static inline u32 tipc_cluster_mask(u32 addr) +{ + return addr & TIPC_CLUSTER_MASK; +} + +static inline int in_own_cluster_exact(u32 addr) +{ + return !((addr ^ tipc_own_addr) >> 12); +} + +/** + * in_own_node - test for node inclusion; <0.0.0> always matches + */ +static inline int in_own_node(u32 addr) +{ + return (addr == tipc_own_addr) || !addr; +} + +/** + * in_own_cluster - test for cluster inclusion; <0.0.0> always matches + */ +static inline int in_own_cluster(u32 addr) +{ + return in_own_cluster_exact(addr) || !addr; +} + +/** + * addr_domain - convert 2-bit scope value to equivalent message lookup domain + * + * Needed when address of a named message must be looked up a second time + * after a network hop. + */ +static inline u32 addr_domain(u32 sc) +{ + if (likely(sc == TIPC_NODE_SCOPE)) + return tipc_own_addr; + if (sc == TIPC_CLUSTER_SCOPE) + return tipc_cluster_mask(tipc_own_addr); + return tipc_zone_mask(tipc_own_addr); +} + +int tipc_addr_domain_valid(u32); +int tipc_addr_node_valid(u32 addr); +int tipc_in_scope(u32 domain, u32 addr); +int tipc_addr_scope(u32 domain); +char *tipc_addr_string_fill(char *string, u32 addr); +#endif diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c new file mode 100644 index 00000000000..55c6c9d3e1c --- /dev/null +++ b/net/tipc/bcast.c @@ -0,0 +1,952 @@ +/* + * net/tipc/bcast.c: TIPC broadcast code + * + * Copyright (c) 2004-2006, Ericsson AB + * Copyright (c) 2004, Intel Corporation. + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "link.h" +#include "port.h" +#include "bcast.h" +#include "name_distr.h" + +#define MAX_PKT_DEFAULT_MCAST 1500 /* bcast link max packet size (fixed) */ +#define BCLINK_WIN_DEFAULT 20 /* bcast link window size (default) */ +#define BCBEARER MAX_BEARERS + +/** + * struct tipc_bcbearer_pair - a pair of bearers used by broadcast link + * @primary: pointer to primary bearer + * @secondary: pointer to secondary bearer + * + * Bearers must have same priority and same set of reachable destinations + * to be paired. + */ + +struct tipc_bcbearer_pair { + struct tipc_bearer *primary; + struct tipc_bearer *secondary; +}; + +/** + * struct tipc_bcbearer - bearer used by broadcast link + * @bearer: (non-standard) broadcast bearer structure + * @media: (non-standard) broadcast media structure + * @bpairs: array of bearer pairs + * @bpairs_temp: temporary array of bearer pairs used by tipc_bcbearer_sort() + * @remains: temporary node map used by tipc_bcbearer_send() + * @remains_new: temporary node map used tipc_bcbearer_send() + * + * Note: The fields labelled "temporary" are incorporated into the bearer + * to avoid consuming potentially limited stack space through the use of + * large local variables within multicast routines. Concurrent access is + * prevented through use of the spinlock "bclink_lock". + */ +struct tipc_bcbearer { + struct tipc_bearer bearer; + struct tipc_media media; + struct tipc_bcbearer_pair bpairs[MAX_BEARERS]; + struct tipc_bcbearer_pair bpairs_temp[TIPC_MAX_LINK_PRI + 1]; + struct tipc_node_map remains; + struct tipc_node_map remains_new; +}; + +/** + * struct tipc_bclink - link used for broadcast messages + * @lock: spinlock governing access to structure + * @link: (non-standard) broadcast link structure + * @node: (non-standard) node structure representing b'cast link's peer node + * @flags: represent bclink states + * @bcast_nodes: map of broadcast-capable nodes + * @retransmit_to: node that most recently requested a retransmit + * + * Handles sequence numbering, fragmentation, bundling, etc. + */ +struct tipc_bclink { + spinlock_t lock; + struct tipc_link link; + struct tipc_node node; + unsigned int flags; + struct tipc_node_map bcast_nodes; + struct tipc_node *retransmit_to; +}; + +static struct tipc_bcbearer *bcbearer; +static struct tipc_bclink *bclink; +static struct tipc_link *bcl; + +const char tipc_bclink_name[] = "broadcast-link"; + +static void tipc_nmap_diff(struct tipc_node_map *nm_a, + struct tipc_node_map *nm_b, + struct tipc_node_map *nm_diff); +static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node); +static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node); + +static void tipc_bclink_lock(void) +{ + spin_lock_bh(&bclink->lock); +} + +static void tipc_bclink_unlock(void) +{ + struct tipc_node *node = NULL; + + if (likely(!bclink->flags)) { + spin_unlock_bh(&bclink->lock); + return; + } + + if (bclink->flags & TIPC_BCLINK_RESET) { + bclink->flags &= ~TIPC_BCLINK_RESET; + node = tipc_bclink_retransmit_to(); + } + spin_unlock_bh(&bclink->lock); + + if (node) + tipc_link_reset_all(node); +} + +void tipc_bclink_set_flags(unsigned int flags) +{ + bclink->flags |= flags; +} + +static u32 bcbuf_acks(struct sk_buff *buf) +{ + return (u32)(unsigned long)TIPC_SKB_CB(buf)->handle; +} + +static void bcbuf_set_acks(struct sk_buff *buf, u32 acks) +{ + TIPC_SKB_CB(buf)->handle = (void *)(unsigned long)acks; +} + +static void bcbuf_decr_acks(struct sk_buff *buf) +{ + bcbuf_set_acks(buf, bcbuf_acks(buf) - 1); +} + +void tipc_bclink_add_node(u32 addr) +{ + tipc_bclink_lock(); + tipc_nmap_add(&bclink->bcast_nodes, addr); + tipc_bclink_unlock(); +} + +void tipc_bclink_remove_node(u32 addr) +{ + tipc_bclink_lock(); + tipc_nmap_remove(&bclink->bcast_nodes, addr); + tipc_bclink_unlock(); +} + +static void bclink_set_last_sent(void) +{ + if (bcl->next_out) + bcl->fsm_msg_cnt = mod(buf_seqno(bcl->next_out) - 1); + else + bcl->fsm_msg_cnt = mod(bcl->next_out_no - 1); +} + +u32 tipc_bclink_get_last_sent(void) +{ + return bcl->fsm_msg_cnt; +} + +static void bclink_update_last_sent(struct tipc_node *node, u32 seqno) +{ + node->bclink.last_sent = less_eq(node->bclink.last_sent, seqno) ? + seqno : node->bclink.last_sent; +} + + +/** + * tipc_bclink_retransmit_to - get most recent node to request retransmission + * + * Called with bclink_lock locked + */ +struct tipc_node *tipc_bclink_retransmit_to(void) +{ + return bclink->retransmit_to; +} + +/** + * bclink_retransmit_pkt - retransmit broadcast packets + * @after: sequence number of last packet to *not* retransmit + * @to: sequence number of last packet to retransmit + * + * Called with bclink_lock locked + */ +static void bclink_retransmit_pkt(u32 after, u32 to) +{ + struct sk_buff *buf; + + buf = bcl->first_out; + while (buf && less_eq(buf_seqno(buf), after)) + buf = buf->next; + tipc_link_retransmit(bcl, buf, mod(to - after)); +} + +/** + * tipc_bclink_acknowledge - handle acknowledgement of broadcast packets + * @n_ptr: node that sent acknowledgement info + * @acked: broadcast sequence # that has been acknowledged + * + * Node is locked, bclink_lock unlocked. + */ +void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) +{ + struct sk_buff *crs; + struct sk_buff *next; + unsigned int released = 0; + + tipc_bclink_lock(); + /* Bail out if tx queue is empty (no clean up is required) */ + crs = bcl->first_out; + if (!crs) + goto exit; + + /* Determine which messages need to be acknowledged */ + if (acked == INVALID_LINK_SEQ) { + /* + * Contact with specified node has been lost, so need to + * acknowledge sent messages only (if other nodes still exist) + * or both sent and unsent messages (otherwise) + */ + if (bclink->bcast_nodes.count) + acked = bcl->fsm_msg_cnt; + else + acked = bcl->next_out_no; + } else { + /* + * Bail out if specified sequence number does not correspond + * to a message that has been sent and not yet acknowledged + */ + if (less(acked, buf_seqno(crs)) || + less(bcl->fsm_msg_cnt, acked) || + less_eq(acked, n_ptr->bclink.acked)) + goto exit; + } + + /* Skip over packets that node has previously acknowledged */ + while (crs && less_eq(buf_seqno(crs), n_ptr->bclink.acked)) + crs = crs->next; + + /* Update packets that node is now acknowledging */ + + while (crs && less_eq(buf_seqno(crs), acked)) { + next = crs->next; + + if (crs != bcl->next_out) + bcbuf_decr_acks(crs); + else { + bcbuf_set_acks(crs, 0); + bcl->next_out = next; + bclink_set_last_sent(); + } + + if (bcbuf_acks(crs) == 0) { + bcl->first_out = next; + bcl->out_queue_size--; + kfree_skb(crs); + released = 1; + } + crs = next; + } + n_ptr->bclink.acked = acked; + + /* Try resolving broadcast link congestion, if necessary */ + + if (unlikely(bcl->next_out)) { + tipc_link_push_queue(bcl); + bclink_set_last_sent(); + } + if (unlikely(released && !list_empty(&bcl->waiting_ports))) + tipc_link_wakeup_ports(bcl, 0); +exit: + tipc_bclink_unlock(); +} + +/** + * tipc_bclink_update_link_state - update broadcast link state + * + * RCU and node lock set + */ +void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) +{ + struct sk_buff *buf; + + /* Ignore "stale" link state info */ + + if (less_eq(last_sent, n_ptr->bclink.last_in)) + return; + + /* Update link synchronization state; quit if in sync */ + + bclink_update_last_sent(n_ptr, last_sent); + + if (n_ptr->bclink.last_sent == n_ptr->bclink.last_in) + return; + + /* Update out-of-sync state; quit if loss is still unconfirmed */ + + if ((++n_ptr->bclink.oos_state) == 1) { + if (n_ptr->bclink.deferred_size < (TIPC_MIN_LINK_WIN / 2)) + return; + n_ptr->bclink.oos_state++; + } + + /* Don't NACK if one has been recently sent (or seen) */ + + if (n_ptr->bclink.oos_state & 0x1) + return; + + /* Send NACK */ + + buf = tipc_buf_acquire(INT_H_SIZE); + if (buf) { + struct tipc_msg *msg = buf_msg(buf); + + tipc_msg_init(msg, BCAST_PROTOCOL, STATE_MSG, + INT_H_SIZE, n_ptr->addr); + msg_set_non_seq(msg, 1); + msg_set_mc_netid(msg, tipc_net_id); + msg_set_bcast_ack(msg, n_ptr->bclink.last_in); + msg_set_bcgap_after(msg, n_ptr->bclink.last_in); + msg_set_bcgap_to(msg, n_ptr->bclink.deferred_head + ? buf_seqno(n_ptr->bclink.deferred_head) - 1 + : n_ptr->bclink.last_sent); + + tipc_bclink_lock(); + tipc_bearer_send(MAX_BEARERS, buf, NULL); + bcl->stats.sent_nacks++; + tipc_bclink_unlock(); + kfree_skb(buf); + + n_ptr->bclink.oos_state++; + } +} + +/** + * bclink_peek_nack - monitor retransmission requests sent by other nodes + * + * Delay any upcoming NACK by this node if another node has already + * requested the first message this node is going to ask for. + */ +static void bclink_peek_nack(struct tipc_msg *msg) +{ + struct tipc_node *n_ptr = tipc_node_find(msg_destnode(msg)); + + if (unlikely(!n_ptr)) + return; + + tipc_node_lock(n_ptr); + + if (n_ptr->bclink.recv_permitted && + (n_ptr->bclink.last_in != n_ptr->bclink.last_sent) && + (n_ptr->bclink.last_in == msg_bcgap_after(msg))) + n_ptr->bclink.oos_state = 2; + + tipc_node_unlock(n_ptr); +} + +/* + * tipc_bclink_xmit - broadcast a packet to all nodes in cluster + */ +int tipc_bclink_xmit(struct sk_buff *buf) +{ + int res; + + tipc_bclink_lock(); + + if (!bclink->bcast_nodes.count) { + res = msg_data_sz(buf_msg(buf)); + kfree_skb(buf); + goto exit; + } + + res = __tipc_link_xmit(bcl, buf); + if (likely(res >= 0)) { + bclink_set_last_sent(); + bcl->stats.queue_sz_counts++; + bcl->stats.accu_queue_sz += bcl->out_queue_size; + } +exit: + tipc_bclink_unlock(); + return res; +} + +/** + * bclink_accept_pkt - accept an incoming, in-sequence broadcast packet + * + * Called with both sending node's lock and bclink_lock taken. + */ +static void bclink_accept_pkt(struct tipc_node *node, u32 seqno) +{ + bclink_update_last_sent(node, seqno); + node->bclink.last_in = seqno; + node->bclink.oos_state = 0; + bcl->stats.recv_info++; + + /* + * Unicast an ACK periodically, ensuring that + * all nodes in the cluster don't ACK at the same time + */ + + if (((seqno - tipc_own_addr) % TIPC_MIN_LINK_WIN) == 0) { + tipc_link_proto_xmit(node->active_links[node->addr & 1], + STATE_MSG, 0, 0, 0, 0, 0); + bcl->stats.sent_acks++; + } +} + +/** + * tipc_bclink_rcv - receive a broadcast packet, and deliver upwards + * + * RCU is locked, no other locks set + */ +void tipc_bclink_rcv(struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + struct tipc_node *node; + u32 next_in; + u32 seqno; + int deferred; + + /* Screen out unwanted broadcast messages */ + + if (msg_mc_netid(msg) != tipc_net_id) + goto exit; + + node = tipc_node_find(msg_prevnode(msg)); + if (unlikely(!node)) + goto exit; + + tipc_node_lock(node); + if (unlikely(!node->bclink.recv_permitted)) + goto unlock; + + /* Handle broadcast protocol message */ + + if (unlikely(msg_user(msg) == BCAST_PROTOCOL)) { + if (msg_type(msg) != STATE_MSG) + goto unlock; + if (msg_destnode(msg) == tipc_own_addr) { + tipc_bclink_acknowledge(node, msg_bcast_ack(msg)); + tipc_node_unlock(node); + tipc_bclink_lock(); + bcl->stats.recv_nacks++; + bclink->retransmit_to = node; + bclink_retransmit_pkt(msg_bcgap_after(msg), + msg_bcgap_to(msg)); + tipc_bclink_unlock(); + } else { + tipc_node_unlock(node); + bclink_peek_nack(msg); + } + goto exit; + } + + /* Handle in-sequence broadcast message */ + + seqno = msg_seqno(msg); + next_in = mod(node->bclink.last_in + 1); + + if (likely(seqno == next_in)) { +receive: + /* Deliver message to destination */ + + if (likely(msg_isdata(msg))) { + tipc_bclink_lock(); + bclink_accept_pkt(node, seqno); + tipc_bclink_unlock(); + tipc_node_unlock(node); + if (likely(msg_mcast(msg))) + tipc_port_mcast_rcv(buf, NULL); + else + kfree_skb(buf); + } else if (msg_user(msg) == MSG_BUNDLER) { + tipc_bclink_lock(); + bclink_accept_pkt(node, seqno); + bcl->stats.recv_bundles++; + bcl->stats.recv_bundled += msg_msgcnt(msg); + tipc_bclink_unlock(); + tipc_node_unlock(node); + tipc_link_bundle_rcv(buf); + } else if (msg_user(msg) == MSG_FRAGMENTER) { + tipc_buf_append(&node->bclink.reasm_buf, &buf); + if (unlikely(!buf && !node->bclink.reasm_buf)) + goto unlock; + tipc_bclink_lock(); + bclink_accept_pkt(node, seqno); + bcl->stats.recv_fragments++; + if (buf) { + bcl->stats.recv_fragmented++; + msg = buf_msg(buf); + tipc_bclink_unlock(); + goto receive; + } + tipc_bclink_unlock(); + tipc_node_unlock(node); + } else if (msg_user(msg) == NAME_DISTRIBUTOR) { + tipc_bclink_lock(); + bclink_accept_pkt(node, seqno); + tipc_bclink_unlock(); + tipc_node_unlock(node); + tipc_named_rcv(buf); + } else { + tipc_bclink_lock(); + bclink_accept_pkt(node, seqno); + tipc_bclink_unlock(); + tipc_node_unlock(node); + kfree_skb(buf); + } + buf = NULL; + + /* Determine new synchronization state */ + + tipc_node_lock(node); + if (unlikely(!tipc_node_is_up(node))) + goto unlock; + + if (node->bclink.last_in == node->bclink.last_sent) + goto unlock; + + if (!node->bclink.deferred_head) { + node->bclink.oos_state = 1; + goto unlock; + } + + msg = buf_msg(node->bclink.deferred_head); + seqno = msg_seqno(msg); + next_in = mod(next_in + 1); + if (seqno != next_in) + goto unlock; + + /* Take in-sequence message from deferred queue & deliver it */ + + buf = node->bclink.deferred_head; + node->bclink.deferred_head = buf->next; + buf->next = NULL; + node->bclink.deferred_size--; + goto receive; + } + + /* Handle out-of-sequence broadcast message */ + + if (less(next_in, seqno)) { + deferred = tipc_link_defer_pkt(&node->bclink.deferred_head, + &node->bclink.deferred_tail, + buf); + node->bclink.deferred_size += deferred; + bclink_update_last_sent(node, seqno); + buf = NULL; + } else + deferred = 0; + + tipc_bclink_lock(); + + if (deferred) + bcl->stats.deferred_recv++; + else + bcl->stats.duplicates++; + + tipc_bclink_unlock(); + +unlock: + tipc_node_unlock(node); +exit: + kfree_skb(buf); +} + +u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr) +{ + return (n_ptr->bclink.recv_permitted && + (tipc_bclink_get_last_sent() != n_ptr->bclink.acked)); +} + + +/** + * tipc_bcbearer_send - send a packet through the broadcast pseudo-bearer + * + * Send packet over as many bearers as necessary to reach all nodes + * that have joined the broadcast link. + * + * Returns 0 (packet sent successfully) under all circumstances, + * since the broadcast link's pseudo-bearer never blocks + */ +static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, + struct tipc_media_addr *unused2) +{ + int bp_index; + + /* Prepare broadcast link message for reliable transmission, + * if first time trying to send it; + * preparation is skipped for broadcast link protocol messages + * since they are sent in an unreliable manner and don't need it + */ + if (likely(!msg_non_seq(buf_msg(buf)))) { + struct tipc_msg *msg; + + bcbuf_set_acks(buf, bclink->bcast_nodes.count); + msg = buf_msg(buf); + msg_set_non_seq(msg, 1); + msg_set_mc_netid(msg, tipc_net_id); + bcl->stats.sent_info++; + + if (WARN_ON(!bclink->bcast_nodes.count)) { + dump_stack(); + return 0; + } + } + + /* Send buffer over bearers until all targets reached */ + bcbearer->remains = bclink->bcast_nodes; + + for (bp_index = 0; bp_index < MAX_BEARERS; bp_index++) { + struct tipc_bearer *p = bcbearer->bpairs[bp_index].primary; + struct tipc_bearer *s = bcbearer->bpairs[bp_index].secondary; + struct tipc_bearer *b = p; + struct sk_buff *tbuf; + + if (!p) + break; /* No more bearers to try */ + + tipc_nmap_diff(&bcbearer->remains, &b->nodes, + &bcbearer->remains_new); + if (bcbearer->remains_new.count == bcbearer->remains.count) + continue; /* Nothing added by bearer pair */ + + if (bp_index == 0) { + /* Use original buffer for first bearer */ + tipc_bearer_send(b->identity, buf, &b->bcast_addr); + } else { + /* Avoid concurrent buffer access */ + tbuf = pskb_copy_for_clone(buf, GFP_ATOMIC); + if (!tbuf) + break; + tipc_bearer_send(b->identity, tbuf, &b->bcast_addr); + kfree_skb(tbuf); /* Bearer keeps a clone */ + } + + /* Swap bearers for next packet */ + if (s) { + bcbearer->bpairs[bp_index].primary = s; + bcbearer->bpairs[bp_index].secondary = p; + } + + if (bcbearer->remains_new.count == 0) + break; /* All targets reached */ + + bcbearer->remains = bcbearer->remains_new; + } + + return 0; +} + +/** + * tipc_bcbearer_sort - create sets of bearer pairs used by broadcast bearer + */ +void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action) +{ + struct tipc_bcbearer_pair *bp_temp = bcbearer->bpairs_temp; + struct tipc_bcbearer_pair *bp_curr; + struct tipc_bearer *b; + int b_index; + int pri; + + tipc_bclink_lock(); + + if (action) + tipc_nmap_add(nm_ptr, node); + else + tipc_nmap_remove(nm_ptr, node); + + /* Group bearers by priority (can assume max of two per priority) */ + memset(bp_temp, 0, sizeof(bcbearer->bpairs_temp)); + + rcu_read_lock(); + for (b_index = 0; b_index < MAX_BEARERS; b_index++) { + b = rcu_dereference_rtnl(bearer_list[b_index]); + if (!b || !b->nodes.count) + continue; + + if (!bp_temp[b->priority].primary) + bp_temp[b->priority].primary = b; + else + bp_temp[b->priority].secondary = b; + } + rcu_read_unlock(); + + /* Create array of bearer pairs for broadcasting */ + bp_curr = bcbearer->bpairs; + memset(bcbearer->bpairs, 0, sizeof(bcbearer->bpairs)); + + for (pri = TIPC_MAX_LINK_PRI; pri >= 0; pri--) { + + if (!bp_temp[pri].primary) + continue; + + bp_curr->primary = bp_temp[pri].primary; + + if (bp_temp[pri].secondary) { + if (tipc_nmap_equal(&bp_temp[pri].primary->nodes, + &bp_temp[pri].secondary->nodes)) { + bp_curr->secondary = bp_temp[pri].secondary; + } else { + bp_curr++; + bp_curr->primary = bp_temp[pri].secondary; + } + } + + bp_curr++; + } + + tipc_bclink_unlock(); +} + + +int tipc_bclink_stats(char *buf, const u32 buf_size) +{ + int ret; + struct tipc_stats *s; + + if (!bcl) + return 0; + + tipc_bclink_lock(); + + s = &bcl->stats; + + ret = tipc_snprintf(buf, buf_size, "Link <%s>\n" + " Window:%u packets\n", + bcl->name, bcl->queue_limit[0]); + ret += tipc_snprintf(buf + ret, buf_size - ret, + " RX packets:%u fragments:%u/%u bundles:%u/%u\n", + s->recv_info, s->recv_fragments, + s->recv_fragmented, s->recv_bundles, + s->recv_bundled); + ret += tipc_snprintf(buf + ret, buf_size - ret, + " TX packets:%u fragments:%u/%u bundles:%u/%u\n", + s->sent_info, s->sent_fragments, + s->sent_fragmented, s->sent_bundles, + s->sent_bundled); + ret += tipc_snprintf(buf + ret, buf_size - ret, + " RX naks:%u defs:%u dups:%u\n", + s->recv_nacks, s->deferred_recv, s->duplicates); + ret += tipc_snprintf(buf + ret, buf_size - ret, + " TX naks:%u acks:%u dups:%u\n", + s->sent_nacks, s->sent_acks, s->retransmitted); + ret += tipc_snprintf(buf + ret, buf_size - ret, + " Congestion link:%u Send queue max:%u avg:%u\n", + s->link_congs, s->max_queue_sz, + s->queue_sz_counts ? + (s->accu_queue_sz / s->queue_sz_counts) : 0); + + tipc_bclink_unlock(); + return ret; +} + +int tipc_bclink_reset_stats(void) +{ + if (!bcl) + return -ENOPROTOOPT; + + tipc_bclink_lock(); + memset(&bcl->stats, 0, sizeof(bcl->stats)); + tipc_bclink_unlock(); + return 0; +} + +int tipc_bclink_set_queue_limits(u32 limit) +{ + if (!bcl) + return -ENOPROTOOPT; + if ((limit < TIPC_MIN_LINK_WIN) || (limit > TIPC_MAX_LINK_WIN)) + return -EINVAL; + + tipc_bclink_lock(); + tipc_link_set_queue_limits(bcl, limit); + tipc_bclink_unlock(); + return 0; +} + +int tipc_bclink_init(void) +{ + bcbearer = kzalloc(sizeof(*bcbearer), GFP_ATOMIC); + if (!bcbearer) + return -ENOMEM; + + bclink = kzalloc(sizeof(*bclink), GFP_ATOMIC); + if (!bclink) { + kfree(bcbearer); + return -ENOMEM; + } + + bcl = &bclink->link; + bcbearer->bearer.media = &bcbearer->media; + bcbearer->media.send_msg = tipc_bcbearer_send; + sprintf(bcbearer->media.name, "tipc-broadcast"); + + spin_lock_init(&bclink->lock); + INIT_LIST_HEAD(&bcl->waiting_ports); + bcl->next_out_no = 1; + spin_lock_init(&bclink->node.lock); + bcl->owner = &bclink->node; + bcl->max_pkt = MAX_PKT_DEFAULT_MCAST; + tipc_link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT); + bcl->bearer_id = MAX_BEARERS; + rcu_assign_pointer(bearer_list[MAX_BEARERS], &bcbearer->bearer); + bcl->state = WORKING_WORKING; + strlcpy(bcl->name, tipc_bclink_name, TIPC_MAX_LINK_NAME); + return 0; +} + +void tipc_bclink_stop(void) +{ + tipc_bclink_lock(); + tipc_link_purge_queues(bcl); + tipc_bclink_unlock(); + + RCU_INIT_POINTER(bearer_list[BCBEARER], NULL); + synchronize_net(); + kfree(bcbearer); + kfree(bclink); +} + +/** + * tipc_nmap_add - add a node to a node map + */ +static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node) +{ + int n = tipc_node(node); + int w = n / WSIZE; + u32 mask = (1 << (n % WSIZE)); + + if ((nm_ptr->map[w] & mask) == 0) { + nm_ptr->count++; + nm_ptr->map[w] |= mask; + } +} + +/** + * tipc_nmap_remove - remove a node from a node map + */ +static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node) +{ + int n = tipc_node(node); + int w = n / WSIZE; + u32 mask = (1 << (n % WSIZE)); + + if ((nm_ptr->map[w] & mask) != 0) { + nm_ptr->map[w] &= ~mask; + nm_ptr->count--; + } +} + +/** + * tipc_nmap_diff - find differences between node maps + * @nm_a: input node map A + * @nm_b: input node map B + * @nm_diff: output node map A-B (i.e. nodes of A that are not in B) + */ +static void tipc_nmap_diff(struct tipc_node_map *nm_a, + struct tipc_node_map *nm_b, + struct tipc_node_map *nm_diff) +{ + int stop = ARRAY_SIZE(nm_a->map); + int w; + int b; + u32 map; + + memset(nm_diff, 0, sizeof(*nm_diff)); + for (w = 0; w < stop; w++) { + map = nm_a->map[w] ^ (nm_a->map[w] & nm_b->map[w]); + nm_diff->map[w] = map; + if (map != 0) { + for (b = 0 ; b < WSIZE; b++) { + if (map & (1 << b)) + nm_diff->count++; + } + } + } +} + +/** + * tipc_port_list_add - add a port to a port list, ensuring no duplicates + */ +void tipc_port_list_add(struct tipc_port_list *pl_ptr, u32 port) +{ + struct tipc_port_list *item = pl_ptr; + int i; + int item_sz = PLSIZE; + int cnt = pl_ptr->count; + + for (; ; cnt -= item_sz, item = item->next) { + if (cnt < PLSIZE) + item_sz = cnt; + for (i = 0; i < item_sz; i++) + if (item->ports[i] == port) + return; + if (i < PLSIZE) { + item->ports[i] = port; + pl_ptr->count++; + return; + } + if (!item->next) { + item->next = kmalloc(sizeof(*item), GFP_ATOMIC); + if (!item->next) { + pr_warn("Incomplete multicast delivery, no memory\n"); + return; + } + item->next->next = NULL; + } + } +} + +/** + * tipc_port_list_free - free dynamically created entries in port_list chain + * + */ +void tipc_port_list_free(struct tipc_port_list *pl_ptr) +{ + struct tipc_port_list *item; + struct tipc_port_list *next; + + for (item = pl_ptr->next; item; item = next) { + next = item->next; + kfree(item); + } +} diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h new file mode 100644 index 00000000000..00330c45df3 --- /dev/null +++ b/net/tipc/bcast.h @@ -0,0 +1,102 @@ +/* + * net/tipc/bcast.h: Include file for TIPC broadcast code + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_BCAST_H +#define _TIPC_BCAST_H + +#define MAX_NODES 4096 +#define WSIZE 32 +#define TIPC_BCLINK_RESET 1 + +/** + * struct tipc_node_map - set of node identifiers + * @count: # of nodes in set + * @map: bitmap of node identifiers that are in the set + */ +struct tipc_node_map { + u32 count; + u32 map[MAX_NODES / WSIZE]; +}; + +#define PLSIZE 32 + +/** + * struct tipc_port_list - set of node local destination ports + * @count: # of ports in set (only valid for first entry in list) + * @next: pointer to next entry in list + * @ports: array of port references + */ +struct tipc_port_list { + int count; + struct tipc_port_list *next; + u32 ports[PLSIZE]; +}; + + +struct tipc_node; + +extern const char tipc_bclink_name[]; + +/** + * tipc_nmap_equal - test for equality of node maps + */ +static inline int tipc_nmap_equal(struct tipc_node_map *nm_a, + struct tipc_node_map *nm_b) +{ + return !memcmp(nm_a, nm_b, sizeof(*nm_a)); +} + +void tipc_port_list_add(struct tipc_port_list *pl_ptr, u32 port); +void tipc_port_list_free(struct tipc_port_list *pl_ptr); + +int tipc_bclink_init(void); +void tipc_bclink_stop(void); +void tipc_bclink_set_flags(unsigned int flags); +void tipc_bclink_add_node(u32 addr); +void tipc_bclink_remove_node(u32 addr); +struct tipc_node *tipc_bclink_retransmit_to(void); +void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked); +int tipc_bclink_xmit(struct sk_buff *buf); +void tipc_bclink_rcv(struct sk_buff *buf); +u32 tipc_bclink_get_last_sent(void); +u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr); +void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent); +int tipc_bclink_stats(char *stats_buf, const u32 buf_size); +int tipc_bclink_reset_stats(void); +int tipc_bclink_set_queue_limits(u32 limit); +void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action); + +#endif diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c new file mode 100644 index 00000000000..264474394f9 --- /dev/null +++ b/net/tipc/bearer.c @@ -0,0 +1,629 @@ +/* + * net/tipc/bearer.c: TIPC bearer code + * + * Copyright (c) 1996-2006, 2013, Ericsson AB + * Copyright (c) 2004-2006, 2010-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include "bearer.h" +#include "discover.h" + +#define MAX_ADDR_STR 60 + +static struct tipc_media * const media_info_array[] = { + ð_media_info, +#ifdef CONFIG_TIPC_MEDIA_IB + &ib_media_info, +#endif + NULL +}; + +struct tipc_bearer __rcu *bearer_list[MAX_BEARERS + 1]; + +static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down); + +/** + * tipc_media_find - locates specified media object by name + */ +struct tipc_media *tipc_media_find(const char *name) +{ + u32 i; + + for (i = 0; media_info_array[i] != NULL; i++) { + if (!strcmp(media_info_array[i]->name, name)) + break; + } + return media_info_array[i]; +} + +/** + * media_find_id - locates specified media object by type identifier + */ +static struct tipc_media *media_find_id(u8 type) +{ + u32 i; + + for (i = 0; media_info_array[i] != NULL; i++) { + if (media_info_array[i]->type_id == type) + break; + } + return media_info_array[i]; +} + +/** + * tipc_media_addr_printf - record media address in print buffer + */ +void tipc_media_addr_printf(char *buf, int len, struct tipc_media_addr *a) +{ + char addr_str[MAX_ADDR_STR]; + struct tipc_media *m_ptr; + int ret; + + m_ptr = media_find_id(a->media_id); + + if (m_ptr && !m_ptr->addr2str(a, addr_str, sizeof(addr_str))) + ret = tipc_snprintf(buf, len, "%s(%s)", m_ptr->name, addr_str); + else { + u32 i; + + ret = tipc_snprintf(buf, len, "UNKNOWN(%u)", a->media_id); + for (i = 0; i < sizeof(a->value); i++) + ret += tipc_snprintf(buf - ret, len + ret, + "-%02x", a->value[i]); + } +} + +/** + * tipc_media_get_names - record names of registered media in buffer + */ +struct sk_buff *tipc_media_get_names(void) +{ + struct sk_buff *buf; + int i; + + buf = tipc_cfg_reply_alloc(MAX_MEDIA * TLV_SPACE(TIPC_MAX_MEDIA_NAME)); + if (!buf) + return NULL; + + for (i = 0; media_info_array[i] != NULL; i++) { + tipc_cfg_append_tlv(buf, TIPC_TLV_MEDIA_NAME, + media_info_array[i]->name, + strlen(media_info_array[i]->name) + 1); + } + return buf; +} + +/** + * bearer_name_validate - validate & (optionally) deconstruct bearer name + * @name: ptr to bearer name string + * @name_parts: ptr to area for bearer name components (or NULL if not needed) + * + * Returns 1 if bearer name is valid, otherwise 0. + */ +static int bearer_name_validate(const char *name, + struct tipc_bearer_names *name_parts) +{ + char name_copy[TIPC_MAX_BEARER_NAME]; + char *media_name; + char *if_name; + u32 media_len; + u32 if_len; + + /* copy bearer name & ensure length is OK */ + name_copy[TIPC_MAX_BEARER_NAME - 1] = 0; + /* need above in case non-Posix strncpy() doesn't pad with nulls */ + strncpy(name_copy, name, TIPC_MAX_BEARER_NAME); + if (name_copy[TIPC_MAX_BEARER_NAME - 1] != 0) + return 0; + + /* ensure all component parts of bearer name are present */ + media_name = name_copy; + if_name = strchr(media_name, ':'); + if (if_name == NULL) + return 0; + *(if_name++) = 0; + media_len = if_name - media_name; + if_len = strlen(if_name) + 1; + + /* validate component parts of bearer name */ + if ((media_len <= 1) || (media_len > TIPC_MAX_MEDIA_NAME) || + (if_len <= 1) || (if_len > TIPC_MAX_IF_NAME)) + return 0; + + /* return bearer name components, if necessary */ + if (name_parts) { + strcpy(name_parts->media_name, media_name); + strcpy(name_parts->if_name, if_name); + } + return 1; +} + +/** + * tipc_bearer_find - locates bearer object with matching bearer name + */ +struct tipc_bearer *tipc_bearer_find(const char *name) +{ + struct tipc_bearer *b_ptr; + u32 i; + + for (i = 0; i < MAX_BEARERS; i++) { + b_ptr = rtnl_dereference(bearer_list[i]); + if (b_ptr && (!strcmp(b_ptr->name, name))) + return b_ptr; + } + return NULL; +} + +/** + * tipc_bearer_get_names - record names of bearers in buffer + */ +struct sk_buff *tipc_bearer_get_names(void) +{ + struct sk_buff *buf; + struct tipc_bearer *b; + int i, j; + + buf = tipc_cfg_reply_alloc(MAX_BEARERS * TLV_SPACE(TIPC_MAX_BEARER_NAME)); + if (!buf) + return NULL; + + for (i = 0; media_info_array[i] != NULL; i++) { + for (j = 0; j < MAX_BEARERS; j++) { + b = rtnl_dereference(bearer_list[j]); + if (!b) + continue; + if (b->media == media_info_array[i]) { + tipc_cfg_append_tlv(buf, TIPC_TLV_BEARER_NAME, + b->name, + strlen(b->name) + 1); + } + } + } + return buf; +} + +void tipc_bearer_add_dest(u32 bearer_id, u32 dest) +{ + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + if (b_ptr) { + tipc_bcbearer_sort(&b_ptr->nodes, dest, true); + tipc_disc_add_dest(b_ptr->link_req); + } + rcu_read_unlock(); +} + +void tipc_bearer_remove_dest(u32 bearer_id, u32 dest) +{ + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + if (b_ptr) { + tipc_bcbearer_sort(&b_ptr->nodes, dest, false); + tipc_disc_remove_dest(b_ptr->link_req); + } + rcu_read_unlock(); +} + +/** + * tipc_enable_bearer - enable bearer with the given name + */ +int tipc_enable_bearer(const char *name, u32 disc_domain, u32 priority) +{ + struct tipc_bearer *b_ptr; + struct tipc_media *m_ptr; + struct tipc_bearer_names b_names; + char addr_string[16]; + u32 bearer_id; + u32 with_this_prio; + u32 i; + int res = -EINVAL; + + if (!tipc_own_addr) { + pr_warn("Bearer <%s> rejected, not supported in standalone mode\n", + name); + return -ENOPROTOOPT; + } + if (!bearer_name_validate(name, &b_names)) { + pr_warn("Bearer <%s> rejected, illegal name\n", name); + return -EINVAL; + } + if (tipc_addr_domain_valid(disc_domain) && + (disc_domain != tipc_own_addr)) { + if (tipc_in_scope(disc_domain, tipc_own_addr)) { + disc_domain = tipc_own_addr & TIPC_CLUSTER_MASK; + res = 0; /* accept any node in own cluster */ + } else if (in_own_cluster_exact(disc_domain)) + res = 0; /* accept specified node in own cluster */ + } + if (res) { + pr_warn("Bearer <%s> rejected, illegal discovery domain\n", + name); + return -EINVAL; + } + if ((priority > TIPC_MAX_LINK_PRI) && + (priority != TIPC_MEDIA_LINK_PRI)) { + pr_warn("Bearer <%s> rejected, illegal priority\n", name); + return -EINVAL; + } + + m_ptr = tipc_media_find(b_names.media_name); + if (!m_ptr) { + pr_warn("Bearer <%s> rejected, media <%s> not registered\n", + name, b_names.media_name); + return -EINVAL; + } + + if (priority == TIPC_MEDIA_LINK_PRI) + priority = m_ptr->priority; + +restart: + bearer_id = MAX_BEARERS; + with_this_prio = 1; + for (i = MAX_BEARERS; i-- != 0; ) { + b_ptr = rtnl_dereference(bearer_list[i]); + if (!b_ptr) { + bearer_id = i; + continue; + } + if (!strcmp(name, b_ptr->name)) { + pr_warn("Bearer <%s> rejected, already enabled\n", + name); + return -EINVAL; + } + if ((b_ptr->priority == priority) && + (++with_this_prio > 2)) { + if (priority-- == 0) { + pr_warn("Bearer <%s> rejected, duplicate priority\n", + name); + return -EINVAL; + } + pr_warn("Bearer <%s> priority adjustment required %u->%u\n", + name, priority + 1, priority); + goto restart; + } + } + if (bearer_id >= MAX_BEARERS) { + pr_warn("Bearer <%s> rejected, bearer limit reached (%u)\n", + name, MAX_BEARERS); + return -EINVAL; + } + + b_ptr = kzalloc(sizeof(*b_ptr), GFP_ATOMIC); + if (!b_ptr) + return -ENOMEM; + + strcpy(b_ptr->name, name); + b_ptr->media = m_ptr; + res = m_ptr->enable_media(b_ptr); + if (res) { + pr_warn("Bearer <%s> rejected, enable failure (%d)\n", + name, -res); + return -EINVAL; + } + + b_ptr->identity = bearer_id; + b_ptr->tolerance = m_ptr->tolerance; + b_ptr->window = m_ptr->window; + b_ptr->domain = disc_domain; + b_ptr->net_plane = bearer_id + 'A'; + b_ptr->priority = priority; + + res = tipc_disc_create(b_ptr, &b_ptr->bcast_addr); + if (res) { + bearer_disable(b_ptr, false); + pr_warn("Bearer <%s> rejected, discovery object creation failed\n", + name); + return -EINVAL; + } + + rcu_assign_pointer(bearer_list[bearer_id], b_ptr); + + pr_info("Enabled bearer <%s>, discovery domain %s, priority %u\n", + name, + tipc_addr_string_fill(addr_string, disc_domain), priority); + return res; +} + +/** + * tipc_reset_bearer - Reset all links established over this bearer + */ +static int tipc_reset_bearer(struct tipc_bearer *b_ptr) +{ + pr_info("Resetting bearer <%s>\n", b_ptr->name); + tipc_link_reset_list(b_ptr->identity); + tipc_disc_reset(b_ptr); + return 0; +} + +/** + * bearer_disable + * + * Note: This routine assumes caller holds RTNL lock. + */ +static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down) +{ + u32 i; + + pr_info("Disabling bearer <%s>\n", b_ptr->name); + b_ptr->media->disable_media(b_ptr); + + tipc_link_delete_list(b_ptr->identity, shutting_down); + if (b_ptr->link_req) + tipc_disc_delete(b_ptr->link_req); + + for (i = 0; i < MAX_BEARERS; i++) { + if (b_ptr == rtnl_dereference(bearer_list[i])) { + RCU_INIT_POINTER(bearer_list[i], NULL); + break; + } + } + kfree_rcu(b_ptr, rcu); +} + +int tipc_disable_bearer(const char *name) +{ + struct tipc_bearer *b_ptr; + int res; + + b_ptr = tipc_bearer_find(name); + if (b_ptr == NULL) { + pr_warn("Attempt to disable unknown bearer <%s>\n", name); + res = -EINVAL; + } else { + bearer_disable(b_ptr, false); + res = 0; + } + return res; +} + +int tipc_enable_l2_media(struct tipc_bearer *b) +{ + struct net_device *dev; + char *driver_name = strchr((const char *)b->name, ':') + 1; + + /* Find device with specified name */ + dev = dev_get_by_name(&init_net, driver_name); + if (!dev) + return -ENODEV; + + /* Associate TIPC bearer with L2 bearer */ + rcu_assign_pointer(b->media_ptr, dev); + memset(&b->bcast_addr, 0, sizeof(b->bcast_addr)); + memcpy(b->bcast_addr.value, dev->broadcast, b->media->hwaddr_len); + b->bcast_addr.media_id = b->media->type_id; + b->bcast_addr.broadcast = 1; + b->mtu = dev->mtu; + b->media->raw2addr(b, &b->addr, (char *)dev->dev_addr); + rcu_assign_pointer(dev->tipc_ptr, b); + return 0; +} + +/* tipc_disable_l2_media - detach TIPC bearer from an L2 interface + * + * Mark L2 bearer as inactive so that incoming buffers are thrown away, + * then get worker thread to complete bearer cleanup. (Can't do cleanup + * here because cleanup code needs to sleep and caller holds spinlocks.) + */ +void tipc_disable_l2_media(struct tipc_bearer *b) +{ + struct net_device *dev; + + dev = (struct net_device *)rtnl_dereference(b->media_ptr); + RCU_INIT_POINTER(b->media_ptr, NULL); + RCU_INIT_POINTER(dev->tipc_ptr, NULL); + synchronize_net(); + dev_put(dev); +} + +/** + * tipc_l2_send_msg - send a TIPC packet out over an L2 interface + * @buf: the packet to be sent + * @b_ptr: the bearer through which the packet is to be sent + * @dest: peer destination address + */ +int tipc_l2_send_msg(struct sk_buff *buf, struct tipc_bearer *b, + struct tipc_media_addr *dest) +{ + struct sk_buff *clone; + struct net_device *dev; + int delta; + + dev = (struct net_device *)rcu_dereference_rtnl(b->media_ptr); + if (!dev) + return 0; + + clone = skb_clone(buf, GFP_ATOMIC); + if (!clone) + return 0; + + delta = dev->hard_header_len - skb_headroom(buf); + if ((delta > 0) && + pskb_expand_head(clone, SKB_DATA_ALIGN(delta), 0, GFP_ATOMIC)) { + kfree_skb(clone); + return 0; + } + + skb_reset_network_header(clone); + clone->dev = dev; + clone->protocol = htons(ETH_P_TIPC); + dev_hard_header(clone, dev, ETH_P_TIPC, dest->value, + dev->dev_addr, clone->len); + dev_queue_xmit(clone); + return 0; +} + +/* tipc_bearer_send- sends buffer to destination over bearer + * + * IMPORTANT: + * The media send routine must not alter the buffer being passed in + * as it may be needed for later retransmission! + */ +void tipc_bearer_send(u32 bearer_id, struct sk_buff *buf, + struct tipc_media_addr *dest) +{ + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + if (likely(b_ptr)) + b_ptr->media->send_msg(buf, b_ptr, dest); + rcu_read_unlock(); +} + +/** + * tipc_l2_rcv_msg - handle incoming TIPC message from an interface + * @buf: the received packet + * @dev: the net device that the packet was received on + * @pt: the packet_type structure which was used to register this handler + * @orig_dev: the original receive net device in case the device is a bond + * + * Accept only packets explicitly sent to this node, or broadcast packets; + * ignores packets sent using interface multicast, and traffic sent to other + * nodes (which can happen if interface is running in promiscuous mode). + */ +static int tipc_l2_rcv_msg(struct sk_buff *buf, struct net_device *dev, + struct packet_type *pt, struct net_device *orig_dev) +{ + struct tipc_bearer *b_ptr; + + if (!net_eq(dev_net(dev), &init_net)) { + kfree_skb(buf); + return NET_RX_DROP; + } + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(dev->tipc_ptr); + if (likely(b_ptr)) { + if (likely(buf->pkt_type <= PACKET_BROADCAST)) { + buf->next = NULL; + tipc_rcv(buf, b_ptr); + rcu_read_unlock(); + return NET_RX_SUCCESS; + } + } + rcu_read_unlock(); + + kfree_skb(buf); + return NET_RX_DROP; +} + +/** + * tipc_l2_device_event - handle device events from network device + * @nb: the context of the notification + * @evt: the type of event + * @ptr: the net device that the event was on + * + * This function is called by the Ethernet driver in case of link + * change event. + */ +static int tipc_l2_device_event(struct notifier_block *nb, unsigned long evt, + void *ptr) +{ + struct tipc_bearer *b_ptr; + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + + if (!net_eq(dev_net(dev), &init_net)) + return NOTIFY_DONE; + + b_ptr = rtnl_dereference(dev->tipc_ptr); + if (!b_ptr) + return NOTIFY_DONE; + + b_ptr->mtu = dev->mtu; + + switch (evt) { + case NETDEV_CHANGE: + if (netif_carrier_ok(dev)) + break; + case NETDEV_DOWN: + case NETDEV_CHANGEMTU: + tipc_reset_bearer(b_ptr); + break; + case NETDEV_CHANGEADDR: + b_ptr->media->raw2addr(b_ptr, &b_ptr->addr, + (char *)dev->dev_addr); + tipc_reset_bearer(b_ptr); + break; + case NETDEV_UNREGISTER: + case NETDEV_CHANGENAME: + bearer_disable(b_ptr, false); + break; + } + return NOTIFY_OK; +} + +static struct packet_type tipc_packet_type __read_mostly = { + .type = htons(ETH_P_TIPC), + .func = tipc_l2_rcv_msg, +}; + +static struct notifier_block notifier = { + .notifier_call = tipc_l2_device_event, + .priority = 0, +}; + +int tipc_bearer_setup(void) +{ + int err; + + err = register_netdevice_notifier(¬ifier); + if (err) + return err; + dev_add_pack(&tipc_packet_type); + return 0; +} + +void tipc_bearer_cleanup(void) +{ + unregister_netdevice_notifier(¬ifier); + dev_remove_pack(&tipc_packet_type); +} + +void tipc_bearer_stop(void) +{ + struct tipc_bearer *b_ptr; + u32 i; + + for (i = 0; i < MAX_BEARERS; i++) { + b_ptr = rtnl_dereference(bearer_list[i]); + if (b_ptr) { + bearer_disable(b_ptr, true); + bearer_list[i] = NULL; + } + } +} diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h new file mode 100644 index 00000000000..78fccc49de2 --- /dev/null +++ b/net/tipc/bearer.h @@ -0,0 +1,199 @@ +/* + * net/tipc/bearer.h: Include file for TIPC bearer code + * + * Copyright (c) 1996-2006, 2013, Ericsson AB + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_BEARER_H +#define _TIPC_BEARER_H + +#include "bcast.h" + +#define MAX_BEARERS 2 +#define MAX_MEDIA 2 + +/* Identifiers associated with TIPC message header media address info + * - address info field is 32 bytes long + * - the field's actual content and length is defined per media + * - remaining unused bytes in the field are set to zero + */ +#define TIPC_MEDIA_ADDR_SIZE 32 +#define TIPC_MEDIA_TYPE_OFFSET 3 + +/* + * Identifiers of supported TIPC media types + */ +#define TIPC_MEDIA_TYPE_ETH 1 +#define TIPC_MEDIA_TYPE_IB 2 + +/** + * struct tipc_media_addr - destination address used by TIPC bearers + * @value: address info (format defined by media) + * @media_id: TIPC media type identifier + * @broadcast: non-zero if address is a broadcast address + */ +struct tipc_media_addr { + u8 value[TIPC_MEDIA_ADDR_SIZE]; + u8 media_id; + u8 broadcast; +}; + +struct tipc_bearer; + +/** + * struct tipc_media - Media specific info exposed to generic bearer layer + * @send_msg: routine which handles buffer transmission + * @enable_media: routine which enables a media + * @disable_media: routine which disables a media + * @addr2str: convert media address format to string + * @addr2msg: convert from media addr format to discovery msg addr format + * @msg2addr: convert from discovery msg addr format to media addr format + * @raw2addr: convert from raw addr format to media addr format + * @priority: default link (and bearer) priority + * @tolerance: default time (in ms) before declaring link failure + * @window: default window (in packets) before declaring link congestion + * @type_id: TIPC media identifier + * @hwaddr_len: TIPC media address len + * @name: media name + */ +struct tipc_media { + int (*send_msg)(struct sk_buff *buf, + struct tipc_bearer *b_ptr, + struct tipc_media_addr *dest); + int (*enable_media)(struct tipc_bearer *b_ptr); + void (*disable_media)(struct tipc_bearer *b_ptr); + int (*addr2str)(struct tipc_media_addr *addr, + char *strbuf, + int bufsz); + int (*addr2msg)(char *msg, struct tipc_media_addr *addr); + int (*msg2addr)(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg); + int (*raw2addr)(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *raw); + u32 priority; + u32 tolerance; + u32 window; + u32 type_id; + u32 hwaddr_len; + char name[TIPC_MAX_MEDIA_NAME]; +}; + +/** + * struct tipc_bearer - Generic TIPC bearer structure + * @media_ptr: pointer to additional media-specific information about bearer + * @mtu: max packet size bearer can support + * @addr: media-specific address associated with bearer + * @name: bearer name (format = media:interface) + * @media: ptr to media structure associated with bearer + * @bcast_addr: media address used in broadcasting + * @rcu: rcu struct for tipc_bearer + * @priority: default link priority for bearer + * @window: default window size for bearer + * @tolerance: default link tolerance for bearer + * @domain: network domain to which links can be established + * @identity: array index of this bearer within TIPC bearer array + * @link_req: ptr to (optional) structure making periodic link setup requests + * @net_plane: network plane ('A' through 'H') currently associated with bearer + * @nodes: indicates which nodes in cluster can be reached through bearer + * + * Note: media-specific code is responsible for initialization of the fields + * indicated below when a bearer is enabled; TIPC's generic bearer code takes + * care of initializing all other fields. + */ +struct tipc_bearer { + void __rcu *media_ptr; /* initalized by media */ + u32 mtu; /* initalized by media */ + struct tipc_media_addr addr; /* initalized by media */ + char name[TIPC_MAX_BEARER_NAME]; + struct tipc_media *media; + struct tipc_media_addr bcast_addr; + struct rcu_head rcu; + u32 priority; + u32 window; + u32 tolerance; + u32 domain; + u32 identity; + struct tipc_link_req *link_req; + char net_plane; + struct tipc_node_map nodes; +}; + +struct tipc_bearer_names { + char media_name[TIPC_MAX_MEDIA_NAME]; + char if_name[TIPC_MAX_IF_NAME]; +}; + +struct tipc_link; + +extern struct tipc_bearer __rcu *bearer_list[]; + +/* + * TIPC routines available to supported media types + */ + +void tipc_rcv(struct sk_buff *buf, struct tipc_bearer *tb_ptr); +int tipc_enable_bearer(const char *bearer_name, u32 disc_domain, u32 priority); +int tipc_disable_bearer(const char *name); + +/* + * Routines made available to TIPC by supported media types + */ +extern struct tipc_media eth_media_info; + +#ifdef CONFIG_TIPC_MEDIA_IB +extern struct tipc_media ib_media_info; +#endif + +int tipc_media_set_priority(const char *name, u32 new_value); +int tipc_media_set_window(const char *name, u32 new_value); +void tipc_media_addr_printf(char *buf, int len, struct tipc_media_addr *a); +struct sk_buff *tipc_media_get_names(void); +int tipc_enable_l2_media(struct tipc_bearer *b); +void tipc_disable_l2_media(struct tipc_bearer *b); +int tipc_l2_send_msg(struct sk_buff *buf, struct tipc_bearer *b, + struct tipc_media_addr *dest); + +struct sk_buff *tipc_bearer_get_names(void); +void tipc_bearer_add_dest(u32 bearer_id, u32 dest); +void tipc_bearer_remove_dest(u32 bearer_id, u32 dest); +struct tipc_bearer *tipc_bearer_find(const char *name); +struct tipc_media *tipc_media_find(const char *name); +int tipc_bearer_setup(void); +void tipc_bearer_cleanup(void); +void tipc_bearer_stop(void); +void tipc_bearer_send(u32 bearer_id, struct sk_buff *buf, + struct tipc_media_addr *dest); + +#endif /* _TIPC_BEARER_H */ diff --git a/net/tipc/config.c b/net/tipc/config.c new file mode 100644 index 00000000000..2b42403ad33 --- /dev/null +++ b/net/tipc/config.c @@ -0,0 +1,342 @@ +/* + * net/tipc/config.c: TIPC configuration management code + * + * Copyright (c) 2002-2006, Ericsson AB + * Copyright (c) 2004-2007, 2010-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "port.h" +#include "name_table.h" +#include "config.h" +#include "server.h" + +#define REPLY_TRUNCATED "<truncated>\n" + +static const void *req_tlv_area; /* request message TLV area */ +static int req_tlv_space; /* request message TLV area size */ +static int rep_headroom; /* reply message headroom to use */ + +struct sk_buff *tipc_cfg_reply_alloc(int payload_size) +{ + struct sk_buff *buf; + + buf = alloc_skb(rep_headroom + payload_size, GFP_ATOMIC); + if (buf) + skb_reserve(buf, rep_headroom); + return buf; +} + +int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, + void *tlv_data, int tlv_data_size) +{ + struct tlv_desc *tlv = (struct tlv_desc *)skb_tail_pointer(buf); + int new_tlv_space = TLV_SPACE(tlv_data_size); + + if (skb_tailroom(buf) < new_tlv_space) + return 0; + skb_put(buf, new_tlv_space); + tlv->tlv_type = htons(tlv_type); + tlv->tlv_len = htons(TLV_LENGTH(tlv_data_size)); + if (tlv_data_size && tlv_data) + memcpy(TLV_DATA(tlv), tlv_data, tlv_data_size); + return 1; +} + +static struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) +{ + struct sk_buff *buf; + __be32 value_net; + + buf = tipc_cfg_reply_alloc(TLV_SPACE(sizeof(value))); + if (buf) { + value_net = htonl(value); + tipc_cfg_append_tlv(buf, tlv_type, &value_net, + sizeof(value_net)); + } + return buf; +} + +static struct sk_buff *tipc_cfg_reply_unsigned(u32 value) +{ + return tipc_cfg_reply_unsigned_type(TIPC_TLV_UNSIGNED, value); +} + +struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string) +{ + struct sk_buff *buf; + int string_len = strlen(string) + 1; + + buf = tipc_cfg_reply_alloc(TLV_SPACE(string_len)); + if (buf) + tipc_cfg_append_tlv(buf, tlv_type, string, string_len); + return buf; +} + +static struct sk_buff *tipc_show_stats(void) +{ + struct sk_buff *buf; + struct tlv_desc *rep_tlv; + char *pb; + int pb_len; + int str_len; + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + value = ntohl(*(u32 *)TLV_DATA(req_tlv_area)); + if (value != 0) + return tipc_cfg_reply_error_string("unsupported argument"); + + buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); + if (buf == NULL) + return NULL; + + rep_tlv = (struct tlv_desc *)buf->data; + pb = TLV_DATA(rep_tlv); + pb_len = ULTRA_STRING_MAX_LEN; + + str_len = tipc_snprintf(pb, pb_len, "TIPC version " TIPC_MOD_VER "\n"); + str_len += 1; /* for "\0" */ + skb_put(buf, TLV_SPACE(str_len)); + TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); + + return buf; +} + +static struct sk_buff *cfg_enable_bearer(void) +{ + struct tipc_bearer_config *args; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_BEARER_CONFIG)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + args = (struct tipc_bearer_config *)TLV_DATA(req_tlv_area); + if (tipc_enable_bearer(args->name, + ntohl(args->disc_domain), + ntohl(args->priority))) + return tipc_cfg_reply_error_string("unable to enable bearer"); + + return tipc_cfg_reply_none(); +} + +static struct sk_buff *cfg_disable_bearer(void) +{ + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_BEARER_NAME)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + if (tipc_disable_bearer((char *)TLV_DATA(req_tlv_area))) + return tipc_cfg_reply_error_string("unable to disable bearer"); + + return tipc_cfg_reply_none(); +} + +static struct sk_buff *cfg_set_own_addr(void) +{ + u32 addr; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + addr = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); + if (addr == tipc_own_addr) + return tipc_cfg_reply_none(); + if (!tipc_addr_node_valid(addr)) + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (node address)"); + if (tipc_own_addr) + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change node address once assigned)"); + if (!tipc_net_start(addr)) + return tipc_cfg_reply_none(); + + return tipc_cfg_reply_error_string("cannot change to network mode"); +} + +static struct sk_buff *cfg_set_max_ports(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); + if (value == tipc_max_ports) + return tipc_cfg_reply_none(); + if (value < 127 || value > 65535) + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max ports must be 127-65535)"); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change max ports while TIPC is active)"); +} + +static struct sk_buff *cfg_set_netid(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); + if (value == tipc_net_id) + return tipc_cfg_reply_none(); + if (value < 1 || value > 9999) + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (network id must be 1-9999)"); + if (tipc_own_addr) + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change network id once TIPC has joined a network)"); + tipc_net_id = value; + return tipc_cfg_reply_none(); +} + +struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area, + int request_space, int reply_headroom) +{ + struct sk_buff *rep_tlv_buf; + + rtnl_lock(); + + /* Save request and reply details in a well-known location */ + req_tlv_area = request_area; + req_tlv_space = request_space; + rep_headroom = reply_headroom; + + /* Check command authorization */ + if (likely(in_own_node(orig_node))) { + /* command is permitted */ + } else { + rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot be done remotely)"); + goto exit; + } + + /* Call appropriate processing routine */ + switch (cmd) { + case TIPC_CMD_NOOP: + rep_tlv_buf = tipc_cfg_reply_none(); + break; + case TIPC_CMD_GET_NODES: + rep_tlv_buf = tipc_node_get_nodes(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_GET_LINKS: + rep_tlv_buf = tipc_node_get_links(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_SHOW_LINK_STATS: + rep_tlv_buf = tipc_link_cmd_show_stats(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_RESET_LINK_STATS: + rep_tlv_buf = tipc_link_cmd_reset_stats(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_SHOW_NAME_TABLE: + rep_tlv_buf = tipc_nametbl_get(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_GET_BEARER_NAMES: + rep_tlv_buf = tipc_bearer_get_names(); + break; + case TIPC_CMD_GET_MEDIA_NAMES: + rep_tlv_buf = tipc_media_get_names(); + break; + case TIPC_CMD_SHOW_PORTS: + rep_tlv_buf = tipc_port_get_ports(); + break; + case TIPC_CMD_SHOW_STATS: + rep_tlv_buf = tipc_show_stats(); + break; + case TIPC_CMD_SET_LINK_TOL: + case TIPC_CMD_SET_LINK_PRI: + case TIPC_CMD_SET_LINK_WINDOW: + rep_tlv_buf = tipc_link_cmd_config(req_tlv_area, req_tlv_space, cmd); + break; + case TIPC_CMD_ENABLE_BEARER: + rep_tlv_buf = cfg_enable_bearer(); + break; + case TIPC_CMD_DISABLE_BEARER: + rep_tlv_buf = cfg_disable_bearer(); + break; + case TIPC_CMD_SET_NODE_ADDR: + rep_tlv_buf = cfg_set_own_addr(); + break; + case TIPC_CMD_SET_MAX_PORTS: + rep_tlv_buf = cfg_set_max_ports(); + break; + case TIPC_CMD_SET_NETID: + rep_tlv_buf = cfg_set_netid(); + break; + case TIPC_CMD_GET_MAX_PORTS: + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_ports); + break; + case TIPC_CMD_GET_NETID: + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_net_id); + break; + case TIPC_CMD_NOT_NET_ADMIN: + rep_tlv_buf = + tipc_cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN); + break; + case TIPC_CMD_SET_MAX_ZONES: + case TIPC_CMD_GET_MAX_ZONES: + case TIPC_CMD_SET_MAX_SLAVES: + case TIPC_CMD_GET_MAX_SLAVES: + case TIPC_CMD_SET_MAX_CLUSTERS: + case TIPC_CMD_GET_MAX_CLUSTERS: + case TIPC_CMD_SET_MAX_NODES: + case TIPC_CMD_GET_MAX_NODES: + case TIPC_CMD_SET_MAX_SUBSCR: + case TIPC_CMD_GET_MAX_SUBSCR: + case TIPC_CMD_SET_MAX_PUBL: + case TIPC_CMD_GET_MAX_PUBL: + case TIPC_CMD_SET_LOG_SIZE: + case TIPC_CMD_SET_REMOTE_MNG: + case TIPC_CMD_GET_REMOTE_MNG: + case TIPC_CMD_DUMP_LOG: + rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (obsolete command)"); + break; + default: + rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (unknown command)"); + break; + } + + WARN_ON(rep_tlv_buf->len > TLV_SPACE(ULTRA_STRING_MAX_LEN)); + + /* Append an error message if we cannot return all requested data */ + if (rep_tlv_buf->len == TLV_SPACE(ULTRA_STRING_MAX_LEN)) { + if (*(rep_tlv_buf->data + ULTRA_STRING_MAX_LEN) != '\0') + sprintf(rep_tlv_buf->data + rep_tlv_buf->len - + sizeof(REPLY_TRUNCATED) - 1, REPLY_TRUNCATED); + } + + /* Return reply buffer */ +exit: + rtnl_unlock(); + return rep_tlv_buf; +} diff --git a/net/tipc/config.h b/net/tipc/config.h new file mode 100644 index 00000000000..47b1bf18161 --- /dev/null +++ b/net/tipc/config.h @@ -0,0 +1,67 @@ +/* + * net/tipc/config.h: Include file for TIPC configuration service code + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_CONFIG_H +#define _TIPC_CONFIG_H + +/* ---------------------------------------------------------------------- */ + +#include "link.h" + +struct sk_buff *tipc_cfg_reply_alloc(int payload_size); +int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, + void *tlv_data, int tlv_data_size); +struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string); + +static inline struct sk_buff *tipc_cfg_reply_none(void) +{ + return tipc_cfg_reply_alloc(0); +} + +static inline struct sk_buff *tipc_cfg_reply_error_string(char *string) +{ + return tipc_cfg_reply_string_type(TIPC_TLV_ERROR_STRING, string); +} + +static inline struct sk_buff *tipc_cfg_reply_ultra_string(char *string) +{ + return tipc_cfg_reply_string_type(TIPC_TLV_ULTRA_STRING, string); +} + +struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, + const void *req_tlv_area, int req_tlv_space, + int headroom); +#endif diff --git a/net/tipc/core.c b/net/tipc/core.c new file mode 100644 index 00000000000..676d18015dd --- /dev/null +++ b/net/tipc/core.c @@ -0,0 +1,182 @@ +/* + * net/tipc/core.c: TIPC module code + * + * Copyright (c) 2003-2006, 2013, Ericsson AB + * Copyright (c) 2005-2006, 2010-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "ref.h" +#include "name_table.h" +#include "subscr.h" +#include "config.h" +#include "port.h" + +#include <linux/module.h> + +/* global variables used by multiple sub-systems within TIPC */ +int tipc_random __read_mostly; + +/* configurable TIPC parameters */ +u32 tipc_own_addr __read_mostly; +int tipc_max_ports __read_mostly; +int tipc_net_id __read_mostly; +int sysctl_tipc_rmem[3] __read_mostly; /* min/default/max */ + +/** + * tipc_buf_acquire - creates a TIPC message buffer + * @size: message size (including TIPC header) + * + * Returns a new buffer with data pointers set to the specified size. + * + * NOTE: Headroom is reserved to allow prepending of a data link header. + * There may also be unrequested tailroom present at the buffer's end. + */ +struct sk_buff *tipc_buf_acquire(u32 size) +{ + struct sk_buff *skb; + unsigned int buf_size = (BUF_HEADROOM + size + 3) & ~3u; + + skb = alloc_skb_fclone(buf_size, GFP_ATOMIC); + if (skb) { + skb_reserve(skb, BUF_HEADROOM); + skb_put(skb, size); + skb->next = NULL; + } + return skb; +} + +/** + * tipc_core_stop - switch TIPC from SINGLE NODE to NOT RUNNING mode + */ +static void tipc_core_stop(void) +{ + tipc_net_stop(); + tipc_bearer_cleanup(); + tipc_netlink_stop(); + tipc_subscr_stop(); + tipc_nametbl_stop(); + tipc_ref_table_stop(); + tipc_socket_stop(); + tipc_unregister_sysctl(); +} + +/** + * tipc_core_start - switch TIPC from NOT RUNNING to SINGLE NODE mode + */ +static int tipc_core_start(void) +{ + int err; + + get_random_bytes(&tipc_random, sizeof(tipc_random)); + + err = tipc_ref_table_init(tipc_max_ports, tipc_random); + if (err) + goto out_reftbl; + + err = tipc_nametbl_init(); + if (err) + goto out_nametbl; + + err = tipc_netlink_start(); + if (err) + goto out_netlink; + + err = tipc_socket_init(); + if (err) + goto out_socket; + + err = tipc_register_sysctl(); + if (err) + goto out_sysctl; + + err = tipc_subscr_start(); + if (err) + goto out_subscr; + + err = tipc_bearer_setup(); + if (err) + goto out_bearer; + + return 0; +out_bearer: + tipc_subscr_stop(); +out_subscr: + tipc_unregister_sysctl(); +out_sysctl: + tipc_socket_stop(); +out_socket: + tipc_netlink_stop(); +out_netlink: + tipc_nametbl_stop(); +out_nametbl: + tipc_ref_table_stop(); +out_reftbl: + return err; +} + +static int __init tipc_init(void) +{ + int res; + + pr_info("Activated (version " TIPC_MOD_VER ")\n"); + + tipc_own_addr = 0; + tipc_max_ports = CONFIG_TIPC_PORTS; + tipc_net_id = 4711; + + sysctl_tipc_rmem[0] = TIPC_CONN_OVERLOAD_LIMIT >> 4 << + TIPC_LOW_IMPORTANCE; + sysctl_tipc_rmem[1] = TIPC_CONN_OVERLOAD_LIMIT >> 4 << + TIPC_CRITICAL_IMPORTANCE; + sysctl_tipc_rmem[2] = TIPC_CONN_OVERLOAD_LIMIT; + + res = tipc_core_start(); + if (res) + pr_err("Unable to start in single node mode\n"); + else + pr_info("Started in single node mode\n"); + return res; +} + +static void __exit tipc_exit(void) +{ + tipc_core_stop(); + pr_info("Deactivated\n"); +} + +module_init(tipc_init); +module_exit(tipc_exit); + +MODULE_DESCRIPTION("TIPC: Transparent Inter Process Communication"); +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_VERSION(TIPC_MOD_VER); diff --git a/net/tipc/core.h b/net/tipc/core.h new file mode 100644 index 00000000000..bb26ed1ee96 --- /dev/null +++ b/net/tipc/core.h @@ -0,0 +1,203 @@ +/* + * net/tipc/core.h: Include file for TIPC global declarations + * + * Copyright (c) 2005-2006, 2013 Ericsson AB + * Copyright (c) 2005-2007, 2010-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_CORE_H +#define _TIPC_CORE_H + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/tipc.h> +#include <linux/tipc_config.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/mm.h> +#include <linux/timer.h> +#include <linux/string.h> +#include <linux/uaccess.h> +#include <linux/interrupt.h> +#include <linux/atomic.h> +#include <asm/hardirq.h> +#include <linux/netdevice.h> +#include <linux/in.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include <linux/rtnetlink.h> +#include <linux/etherdevice.h> + +#define TIPC_MOD_VER "2.0.0" + +#define ULTRA_STRING_MAX_LEN 32768 +#define TIPC_MAX_SUBSCRIPTIONS 65535 +#define TIPC_MAX_PUBLICATIONS 65535 + +struct tipc_msg; /* msg.h */ + +int tipc_snprintf(char *buf, int len, const char *fmt, ...); + +/* + * TIPC-specific error codes + */ +#define ELINKCONG EAGAIN /* link congestion <=> resource unavailable */ + +/* + * Global configuration variables + */ +extern u32 tipc_own_addr __read_mostly; +extern int tipc_max_ports __read_mostly; +extern int tipc_net_id __read_mostly; +extern int sysctl_tipc_rmem[3] __read_mostly; + +/* + * Other global variables + */ +extern int tipc_random __read_mostly; + +/* + * Routines available to privileged subsystems + */ +int tipc_netlink_start(void); +void tipc_netlink_stop(void); +int tipc_socket_init(void); +void tipc_socket_stop(void); +int tipc_sock_create_local(int type, struct socket **res); +void tipc_sock_release_local(struct socket *sock); +int tipc_sock_accept_local(struct socket *sock, struct socket **newsock, + int flags); + +#ifdef CONFIG_SYSCTL +int tipc_register_sysctl(void); +void tipc_unregister_sysctl(void); +#else +#define tipc_register_sysctl() 0 +#define tipc_unregister_sysctl() +#endif + +/* + * TIPC timer code + */ +typedef void (*Handler) (unsigned long); + +/** + * k_init_timer - initialize a timer + * @timer: pointer to timer structure + * @routine: pointer to routine to invoke when timer expires + * @argument: value to pass to routine when timer expires + * + * Timer must be initialized before use (and terminated when no longer needed). + */ +static inline void k_init_timer(struct timer_list *timer, Handler routine, + unsigned long argument) +{ + setup_timer(timer, routine, argument); +} + +/** + * k_start_timer - start a timer + * @timer: pointer to timer structure + * @msec: time to delay (in ms) + * + * Schedules a previously initialized timer for later execution. + * If timer is already running, the new timeout overrides the previous request. + * + * To ensure the timer doesn't expire before the specified delay elapses, + * the amount of delay is rounded up when converting to the jiffies + * then an additional jiffy is added to account for the fact that + * the starting time may be in the middle of the current jiffy. + */ +static inline void k_start_timer(struct timer_list *timer, unsigned long msec) +{ + mod_timer(timer, jiffies + msecs_to_jiffies(msec) + 1); +} + +/** + * k_cancel_timer - cancel a timer + * @timer: pointer to timer structure + * + * Cancels a previously initialized timer. + * Can be called safely even if the timer is already inactive. + * + * WARNING: Must not be called when holding locks required by the timer's + * timeout routine, otherwise deadlock can occur on SMP systems! + */ +static inline void k_cancel_timer(struct timer_list *timer) +{ + del_timer_sync(timer); +} + +/** + * k_term_timer - terminate a timer + * @timer: pointer to timer structure + * + * Prevents further use of a previously initialized timer. + * + * WARNING: Caller must ensure timer isn't currently running. + * + * (Do not "enhance" this routine to automatically cancel an active timer, + * otherwise deadlock can arise when a timeout routine calls k_term_timer.) + */ +static inline void k_term_timer(struct timer_list *timer) +{ +} + +/* + * TIPC message buffer code + * + * TIPC message buffer headroom reserves space for the worst-case + * link-level device header (in case the message is sent off-node). + * + * Note: Headroom should be a multiple of 4 to ensure the TIPC header fields + * are word aligned for quicker access + */ +#define BUF_HEADROOM LL_MAX_HEADER + +struct tipc_skb_cb { + void *handle; + bool deferred; + struct sk_buff *tail; +}; + +#define TIPC_SKB_CB(__skb) ((struct tipc_skb_cb *)&((__skb)->cb[0])) + +static inline struct tipc_msg *buf_msg(struct sk_buff *skb) +{ + return (struct tipc_msg *)skb->data; +} + +struct sk_buff *tipc_buf_acquire(u32 size); + +#endif diff --git a/net/tipc/discover.c b/net/tipc/discover.c new file mode 100644 index 00000000000..aa722a42ef8 --- /dev/null +++ b/net/tipc/discover.c @@ -0,0 +1,403 @@ +/* + * net/tipc/discover.c + * + * Copyright (c) 2003-2006, 2014, Ericsson AB + * Copyright (c) 2005-2006, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "link.h" +#include "discover.h" + +#define TIPC_LINK_REQ_INIT 125 /* min delay during bearer start up */ +#define TIPC_LINK_REQ_FAST 1000 /* max delay if bearer has no links */ +#define TIPC_LINK_REQ_SLOW 60000 /* max delay if bearer has links */ +#define TIPC_LINK_REQ_INACTIVE 0xffffffff /* indicates no timer in use */ + + +/** + * struct tipc_link_req - information about an ongoing link setup request + * @bearer_id: identity of bearer issuing requests + * @dest: destination address for request messages + * @domain: network domain to which links can be established + * @num_nodes: number of nodes currently discovered (i.e. with an active link) + * @lock: spinlock for controlling access to requests + * @buf: request message to be (repeatedly) sent + * @timer: timer governing period between requests + * @timer_intv: current interval between requests (in ms) + */ +struct tipc_link_req { + u32 bearer_id; + struct tipc_media_addr dest; + u32 domain; + int num_nodes; + spinlock_t lock; + struct sk_buff *buf; + struct timer_list timer; + unsigned int timer_intv; +}; + +/** + * tipc_disc_init_msg - initialize a link setup message + * @type: message type (request or response) + * @b_ptr: ptr to bearer issuing message + */ +static void tipc_disc_init_msg(struct sk_buff *buf, u32 type, + struct tipc_bearer *b_ptr) +{ + struct tipc_msg *msg; + u32 dest_domain = b_ptr->domain; + + msg = buf_msg(buf); + tipc_msg_init(msg, LINK_CONFIG, type, INT_H_SIZE, dest_domain); + msg_set_non_seq(msg, 1); + msg_set_node_sig(msg, tipc_random); + msg_set_dest_domain(msg, dest_domain); + msg_set_bc_netid(msg, tipc_net_id); + b_ptr->media->addr2msg(msg_media_addr(msg), &b_ptr->addr); +} + +/** + * disc_dupl_alert - issue node address duplication alert + * @b_ptr: pointer to bearer detecting duplication + * @node_addr: duplicated node address + * @media_addr: media address advertised by duplicated node + */ +static void disc_dupl_alert(struct tipc_bearer *b_ptr, u32 node_addr, + struct tipc_media_addr *media_addr) +{ + char node_addr_str[16]; + char media_addr_str[64]; + + tipc_addr_string_fill(node_addr_str, node_addr); + tipc_media_addr_printf(media_addr_str, sizeof(media_addr_str), + media_addr); + pr_warn("Duplicate %s using %s seen on <%s>\n", node_addr_str, + media_addr_str, b_ptr->name); +} + +/** + * tipc_disc_rcv - handle incoming discovery message (request or response) + * @buf: buffer containing message + * @bearer: bearer that message arrived on + */ +void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *bearer) +{ + struct tipc_node *node; + struct tipc_link *link; + struct tipc_media_addr maddr; + struct sk_buff *rbuf; + struct tipc_msg *msg = buf_msg(buf); + u32 ddom = msg_dest_domain(msg); + u32 onode = msg_prevnode(msg); + u32 net_id = msg_bc_netid(msg); + u32 mtyp = msg_type(msg); + u32 signature = msg_node_sig(msg); + bool addr_match = false; + bool sign_match = false; + bool link_up = false; + bool accept_addr = false; + bool accept_sign = false; + bool respond = false; + + bearer->media->msg2addr(bearer, &maddr, msg_media_addr(msg)); + kfree_skb(buf); + + /* Ensure message from node is valid and communication is permitted */ + if (net_id != tipc_net_id) + return; + if (maddr.broadcast) + return; + if (!tipc_addr_domain_valid(ddom)) + return; + if (!tipc_addr_node_valid(onode)) + return; + + if (in_own_node(onode)) { + if (memcmp(&maddr, &bearer->addr, sizeof(maddr))) + disc_dupl_alert(bearer, tipc_own_addr, &maddr); + return; + } + if (!tipc_in_scope(ddom, tipc_own_addr)) + return; + if (!tipc_in_scope(bearer->domain, onode)) + return; + + /* Locate, or if necessary, create, node: */ + node = tipc_node_find(onode); + if (!node) + node = tipc_node_create(onode); + if (!node) + return; + + tipc_node_lock(node); + link = node->links[bearer->identity]; + + /* Prepare to validate requesting node's signature and media address */ + sign_match = (signature == node->signature); + addr_match = link && !memcmp(&link->media_addr, &maddr, sizeof(maddr)); + link_up = link && tipc_link_is_up(link); + + + /* These three flags give us eight permutations: */ + + if (sign_match && addr_match && link_up) { + /* All is fine. Do nothing. */ + } else if (sign_match && addr_match && !link_up) { + /* Respond. The link will come up in due time */ + respond = true; + } else if (sign_match && !addr_match && link_up) { + /* Peer has changed i/f address without rebooting. + * If so, the link will reset soon, and the next + * discovery will be accepted. So we can ignore it. + * It may also be an cloned or malicious peer having + * chosen the same node address and signature as an + * existing one. + * Ignore requests until the link goes down, if ever. + */ + disc_dupl_alert(bearer, onode, &maddr); + } else if (sign_match && !addr_match && !link_up) { + /* Peer link has changed i/f address without rebooting. + * It may also be a cloned or malicious peer; we can't + * distinguish between the two. + * The signature is correct, so we must accept. + */ + accept_addr = true; + respond = true; + } else if (!sign_match && addr_match && link_up) { + /* Peer node rebooted. Two possibilities: + * - Delayed re-discovery; this link endpoint has already + * reset and re-established contact with the peer, before + * receiving a discovery message from that node. + * (The peer happened to receive one from this node first). + * - The peer came back so fast that our side has not + * discovered it yet. Probing from this side will soon + * reset the link, since there can be no working link + * endpoint at the peer end, and the link will re-establish. + * Accept the signature, since it comes from a known peer. + */ + accept_sign = true; + } else if (!sign_match && addr_match && !link_up) { + /* The peer node has rebooted. + * Accept signature, since it is a known peer. + */ + accept_sign = true; + respond = true; + } else if (!sign_match && !addr_match && link_up) { + /* Peer rebooted with new address, or a new/duplicate peer. + * Ignore until the link goes down, if ever. + */ + disc_dupl_alert(bearer, onode, &maddr); + } else if (!sign_match && !addr_match && !link_up) { + /* Peer rebooted with new address, or it is a new peer. + * Accept signature and address. + */ + accept_sign = true; + accept_addr = true; + respond = true; + } + + if (accept_sign) + node->signature = signature; + + if (accept_addr) { + if (!link) + link = tipc_link_create(node, bearer, &maddr); + if (link) { + memcpy(&link->media_addr, &maddr, sizeof(maddr)); + tipc_link_reset(link); + } else { + respond = false; + } + } + + /* Send response, if necessary */ + if (respond && (mtyp == DSC_REQ_MSG)) { + rbuf = tipc_buf_acquire(INT_H_SIZE); + if (rbuf) { + tipc_disc_init_msg(rbuf, DSC_RESP_MSG, bearer); + tipc_bearer_send(bearer->identity, rbuf, &maddr); + kfree_skb(rbuf); + } + } + tipc_node_unlock(node); +} + +/** + * disc_update - update frequency of periodic link setup requests + * @req: ptr to link request structure + * + * Reinitiates discovery process if discovery object has no associated nodes + * and is either not currently searching or is searching at a slow rate + */ +static void disc_update(struct tipc_link_req *req) +{ + if (!req->num_nodes) { + if ((req->timer_intv == TIPC_LINK_REQ_INACTIVE) || + (req->timer_intv > TIPC_LINK_REQ_FAST)) { + req->timer_intv = TIPC_LINK_REQ_INIT; + k_start_timer(&req->timer, req->timer_intv); + } + } +} + +/** + * tipc_disc_add_dest - increment set of discovered nodes + * @req: ptr to link request structure + */ +void tipc_disc_add_dest(struct tipc_link_req *req) +{ + spin_lock_bh(&req->lock); + req->num_nodes++; + spin_unlock_bh(&req->lock); +} + +/** + * tipc_disc_remove_dest - decrement set of discovered nodes + * @req: ptr to link request structure + */ +void tipc_disc_remove_dest(struct tipc_link_req *req) +{ + spin_lock_bh(&req->lock); + req->num_nodes--; + disc_update(req); + spin_unlock_bh(&req->lock); +} + +/** + * disc_timeout - send a periodic link setup request + * @req: ptr to link request structure + * + * Called whenever a link setup request timer associated with a bearer expires. + */ +static void disc_timeout(struct tipc_link_req *req) +{ + int max_delay; + + spin_lock_bh(&req->lock); + + /* Stop searching if only desired node has been found */ + if (tipc_node(req->domain) && req->num_nodes) { + req->timer_intv = TIPC_LINK_REQ_INACTIVE; + goto exit; + } + + /* + * Send discovery message, then update discovery timer + * + * Keep doubling time between requests until limit is reached; + * hold at fast polling rate if don't have any associated nodes, + * otherwise hold at slow polling rate + */ + tipc_bearer_send(req->bearer_id, req->buf, &req->dest); + + + req->timer_intv *= 2; + if (req->num_nodes) + max_delay = TIPC_LINK_REQ_SLOW; + else + max_delay = TIPC_LINK_REQ_FAST; + if (req->timer_intv > max_delay) + req->timer_intv = max_delay; + + k_start_timer(&req->timer, req->timer_intv); +exit: + spin_unlock_bh(&req->lock); +} + +/** + * tipc_disc_create - create object to send periodic link setup requests + * @b_ptr: ptr to bearer issuing requests + * @dest: destination address for request messages + * @dest_domain: network domain to which links can be established + * + * Returns 0 if successful, otherwise -errno. + */ +int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest) +{ + struct tipc_link_req *req; + + req = kmalloc(sizeof(*req), GFP_ATOMIC); + if (!req) + return -ENOMEM; + + req->buf = tipc_buf_acquire(INT_H_SIZE); + if (!req->buf) { + kfree(req); + return -ENOMEM; + } + + tipc_disc_init_msg(req->buf, DSC_REQ_MSG, b_ptr); + memcpy(&req->dest, dest, sizeof(*dest)); + req->bearer_id = b_ptr->identity; + req->domain = b_ptr->domain; + req->num_nodes = 0; + req->timer_intv = TIPC_LINK_REQ_INIT; + spin_lock_init(&req->lock); + k_init_timer(&req->timer, (Handler)disc_timeout, (unsigned long)req); + k_start_timer(&req->timer, req->timer_intv); + b_ptr->link_req = req; + tipc_bearer_send(req->bearer_id, req->buf, &req->dest); + return 0; +} + +/** + * tipc_disc_delete - destroy object sending periodic link setup requests + * @req: ptr to link request structure + */ +void tipc_disc_delete(struct tipc_link_req *req) +{ + k_cancel_timer(&req->timer); + k_term_timer(&req->timer); + kfree_skb(req->buf); + kfree(req); +} + +/** + * tipc_disc_reset - reset object to send periodic link setup requests + * @b_ptr: ptr to bearer issuing requests + * @dest_domain: network domain to which links can be established + */ +void tipc_disc_reset(struct tipc_bearer *b_ptr) +{ + struct tipc_link_req *req = b_ptr->link_req; + + spin_lock_bh(&req->lock); + tipc_disc_init_msg(req->buf, DSC_REQ_MSG, b_ptr); + req->bearer_id = b_ptr->identity; + req->domain = b_ptr->domain; + req->num_nodes = 0; + req->timer_intv = TIPC_LINK_REQ_INIT; + k_start_timer(&req->timer, req->timer_intv); + tipc_bearer_send(req->bearer_id, req->buf, &req->dest); + spin_unlock_bh(&req->lock); +} diff --git a/net/tipc/discover.h b/net/tipc/discover.h new file mode 100644 index 00000000000..515b57392f4 --- /dev/null +++ b/net/tipc/discover.h @@ -0,0 +1,49 @@ +/* + * net/tipc/discover.h + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_DISCOVER_H +#define _TIPC_DISCOVER_H + +struct tipc_link_req; + +int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest); +void tipc_disc_delete(struct tipc_link_req *req); +void tipc_disc_reset(struct tipc_bearer *b_ptr); +void tipc_disc_add_dest(struct tipc_link_req *req); +void tipc_disc_remove_dest(struct tipc_link_req *req); +void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *b_ptr); + +#endif diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c new file mode 100644 index 00000000000..5e1426f1751 --- /dev/null +++ b/net/tipc/eth_media.c @@ -0,0 +1,101 @@ +/* + * net/tipc/eth_media.c: Ethernet bearer support for TIPC + * + * Copyright (c) 2001-2007, 2013-2014, Ericsson AB + * Copyright (c) 2005-2008, 2011-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "bearer.h" + +#define ETH_ADDR_OFFSET 4 /* MAC addr position inside address field */ + +/* Convert Ethernet address (media address format) to string */ +static int tipc_eth_addr2str(struct tipc_media_addr *addr, + char *strbuf, int bufsz) +{ + if (bufsz < 18) /* 18 = strlen("aa:bb:cc:dd:ee:ff\0") */ + return 1; + + sprintf(strbuf, "%pM", addr->value); + return 0; +} + +/* Convert from media address format to discovery message addr format */ +static int tipc_eth_addr2msg(char *msg, struct tipc_media_addr *addr) +{ + memset(msg, 0, TIPC_MEDIA_ADDR_SIZE); + msg[TIPC_MEDIA_TYPE_OFFSET] = TIPC_MEDIA_TYPE_ETH; + memcpy(msg + ETH_ADDR_OFFSET, addr->value, ETH_ALEN); + return 0; +} + +/* Convert raw mac address format to media addr format */ +static int tipc_eth_raw2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) +{ + char bcast_mac[ETH_ALEN] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; + + memset(addr, 0, sizeof(*addr)); + ether_addr_copy(addr->value, msg); + addr->media_id = TIPC_MEDIA_TYPE_ETH; + addr->broadcast = !memcmp(addr->value, bcast_mac, ETH_ALEN); + return 0; +} + +/* Convert discovery msg addr format to Ethernet media addr format */ +static int tipc_eth_msg2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) +{ + /* Skip past preamble: */ + msg += ETH_ADDR_OFFSET; + return tipc_eth_raw2addr(b, addr, msg); +} + +/* Ethernet media registration info */ +struct tipc_media eth_media_info = { + .send_msg = tipc_l2_send_msg, + .enable_media = tipc_enable_l2_media, + .disable_media = tipc_disable_l2_media, + .addr2str = tipc_eth_addr2str, + .addr2msg = tipc_eth_addr2msg, + .msg2addr = tipc_eth_msg2addr, + .raw2addr = tipc_eth_raw2addr, + .priority = TIPC_DEF_LINK_PRI, + .tolerance = TIPC_DEF_LINK_TOL, + .window = TIPC_DEF_LINK_WIN, + .type_id = TIPC_MEDIA_TYPE_ETH, + .hwaddr_len = ETH_ALEN, + .name = "eth" +}; diff --git a/net/tipc/ib_media.c b/net/tipc/ib_media.c new file mode 100644 index 00000000000..8522eef9c13 --- /dev/null +++ b/net/tipc/ib_media.c @@ -0,0 +1,101 @@ +/* + * net/tipc/ib_media.c: Infiniband bearer support for TIPC + * + * Copyright (c) 2013 Patrick McHardy <kaber@trash.net> + * + * Based on eth_media.c, which carries the following copyright notice: + * + * Copyright (c) 2001-2007, Ericsson AB + * Copyright (c) 2005-2008, 2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/if_infiniband.h> +#include "core.h" +#include "bearer.h" + +/* convert InfiniBand address (media address format) media address to string */ +static int tipc_ib_addr2str(struct tipc_media_addr *a, char *str_buf, + int str_size) +{ + if (str_size < 60) /* 60 = 19 * strlen("xx:") + strlen("xx\0") */ + return 1; + + sprintf(str_buf, "%20phC", a->value); + + return 0; +} + +/* Convert from media address format to discovery message addr format */ +static int tipc_ib_addr2msg(char *msg, struct tipc_media_addr *addr) +{ + memset(msg, 0, TIPC_MEDIA_ADDR_SIZE); + memcpy(msg, addr->value, INFINIBAND_ALEN); + return 0; +} + +/* Convert raw InfiniBand address format to media addr format */ +static int tipc_ib_raw2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) +{ + memset(addr, 0, sizeof(*addr)); + memcpy(addr->value, msg, INFINIBAND_ALEN); + addr->media_id = TIPC_MEDIA_TYPE_IB; + addr->broadcast = !memcmp(msg, b->bcast_addr.value, + INFINIBAND_ALEN); + return 0; +} + +/* Convert discovery msg addr format to InfiniBand media addr format */ +static int tipc_ib_msg2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) +{ + return tipc_ib_raw2addr(b, addr, msg); +} + +/* InfiniBand media registration info */ +struct tipc_media ib_media_info = { + .send_msg = tipc_l2_send_msg, + .enable_media = tipc_enable_l2_media, + .disable_media = tipc_disable_l2_media, + .addr2str = tipc_ib_addr2str, + .addr2msg = tipc_ib_addr2msg, + .msg2addr = tipc_ib_msg2addr, + .raw2addr = tipc_ib_raw2addr, + .priority = TIPC_DEF_LINK_PRI, + .tolerance = TIPC_DEF_LINK_TOL, + .window = TIPC_DEF_LINK_WIN, + .type_id = TIPC_MEDIA_TYPE_IB, + .hwaddr_len = INFINIBAND_ALEN, + .name = "ib" +}; diff --git a/net/tipc/link.c b/net/tipc/link.c new file mode 100644 index 00000000000..ad2c57f5868 --- /dev/null +++ b/net/tipc/link.c @@ -0,0 +1,2738 @@ +/* + * net/tipc/link.c: TIPC link code + * + * Copyright (c) 1996-2007, 2012-2014, Ericsson AB + * Copyright (c) 2004-2007, 2010-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "link.h" +#include "port.h" +#include "socket.h" +#include "name_distr.h" +#include "discover.h" +#include "config.h" + +#include <linux/pkt_sched.h> + +/* + * Error message prefixes + */ +static const char *link_co_err = "Link changeover error, "; +static const char *link_rst_msg = "Resetting link "; +static const char *link_unk_evt = "Unknown link event "; + +/* + * Out-of-range value for link session numbers + */ +#define INVALID_SESSION 0x10000 + +/* + * Link state events: + */ +#define STARTING_EVT 856384768 /* link processing trigger */ +#define TRAFFIC_MSG_EVT 560815u /* rx'd ??? */ +#define TIMEOUT_EVT 560817u /* link timer expired */ + +/* + * The following two 'message types' is really just implementation + * data conveniently stored in the message header. + * They must not be considered part of the protocol + */ +#define OPEN_MSG 0 +#define CLOSED_MSG 1 + +/* + * State value stored in 'exp_msg_count' + */ +#define START_CHANGEOVER 100000u + +static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, + struct sk_buff *buf); +static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf); +static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr, + struct sk_buff **buf); +static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tolerance); +static int tipc_link_iovec_long_xmit(struct tipc_port *sender, + struct iovec const *msg_sect, + unsigned int len, u32 destnode); +static void link_state_event(struct tipc_link *l_ptr, u32 event); +static void link_reset_statistics(struct tipc_link *l_ptr); +static void link_print(struct tipc_link *l_ptr, const char *str); +static int tipc_link_frag_xmit(struct tipc_link *l_ptr, struct sk_buff *buf); +static void tipc_link_sync_xmit(struct tipc_link *l); +static void tipc_link_sync_rcv(struct tipc_node *n, struct sk_buff *buf); + +/* + * Simple link routines + */ +static unsigned int align(unsigned int i) +{ + return (i + 3) & ~3u; +} + +static void link_init_max_pkt(struct tipc_link *l_ptr) +{ + struct tipc_bearer *b_ptr; + u32 max_pkt; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[l_ptr->bearer_id]); + if (!b_ptr) { + rcu_read_unlock(); + return; + } + max_pkt = (b_ptr->mtu & ~3); + rcu_read_unlock(); + + if (max_pkt > MAX_MSG_SIZE) + max_pkt = MAX_MSG_SIZE; + + l_ptr->max_pkt_target = max_pkt; + if (l_ptr->max_pkt_target < MAX_PKT_DEFAULT) + l_ptr->max_pkt = l_ptr->max_pkt_target; + else + l_ptr->max_pkt = MAX_PKT_DEFAULT; + + l_ptr->max_pkt_probes = 0; +} + +static u32 link_next_sent(struct tipc_link *l_ptr) +{ + if (l_ptr->next_out) + return buf_seqno(l_ptr->next_out); + return mod(l_ptr->next_out_no); +} + +static u32 link_last_sent(struct tipc_link *l_ptr) +{ + return mod(link_next_sent(l_ptr) - 1); +} + +/* + * Simple non-static link routines (i.e. referenced outside this file) + */ +int tipc_link_is_up(struct tipc_link *l_ptr) +{ + if (!l_ptr) + return 0; + return link_working_working(l_ptr) || link_working_unknown(l_ptr); +} + +int tipc_link_is_active(struct tipc_link *l_ptr) +{ + return (l_ptr->owner->active_links[0] == l_ptr) || + (l_ptr->owner->active_links[1] == l_ptr); +} + +/** + * link_timeout - handle expiration of link timer + * @l_ptr: pointer to link + */ +static void link_timeout(struct tipc_link *l_ptr) +{ + tipc_node_lock(l_ptr->owner); + + /* update counters used in statistical profiling of send traffic */ + l_ptr->stats.accu_queue_sz += l_ptr->out_queue_size; + l_ptr->stats.queue_sz_counts++; + + if (l_ptr->first_out) { + struct tipc_msg *msg = buf_msg(l_ptr->first_out); + u32 length = msg_size(msg); + + if ((msg_user(msg) == MSG_FRAGMENTER) && + (msg_type(msg) == FIRST_FRAGMENT)) { + length = msg_size(msg_get_wrapped(msg)); + } + if (length) { + l_ptr->stats.msg_lengths_total += length; + l_ptr->stats.msg_length_counts++; + if (length <= 64) + l_ptr->stats.msg_length_profile[0]++; + else if (length <= 256) + l_ptr->stats.msg_length_profile[1]++; + else if (length <= 1024) + l_ptr->stats.msg_length_profile[2]++; + else if (length <= 4096) + l_ptr->stats.msg_length_profile[3]++; + else if (length <= 16384) + l_ptr->stats.msg_length_profile[4]++; + else if (length <= 32768) + l_ptr->stats.msg_length_profile[5]++; + else + l_ptr->stats.msg_length_profile[6]++; + } + } + + /* do all other link processing performed on a periodic basis */ + + link_state_event(l_ptr, TIMEOUT_EVT); + + if (l_ptr->next_out) + tipc_link_push_queue(l_ptr); + + tipc_node_unlock(l_ptr->owner); +} + +static void link_set_timer(struct tipc_link *l_ptr, u32 time) +{ + k_start_timer(&l_ptr->timer, time); +} + +/** + * tipc_link_create - create a new link + * @n_ptr: pointer to associated node + * @b_ptr: pointer to associated bearer + * @media_addr: media address to use when sending messages over link + * + * Returns pointer to link. + */ +struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, + struct tipc_bearer *b_ptr, + const struct tipc_media_addr *media_addr) +{ + struct tipc_link *l_ptr; + struct tipc_msg *msg; + char *if_name; + char addr_string[16]; + u32 peer = n_ptr->addr; + + if (n_ptr->link_cnt >= 2) { + tipc_addr_string_fill(addr_string, n_ptr->addr); + pr_err("Attempt to establish third link to %s\n", addr_string); + return NULL; + } + + if (n_ptr->links[b_ptr->identity]) { + tipc_addr_string_fill(addr_string, n_ptr->addr); + pr_err("Attempt to establish second link on <%s> to %s\n", + b_ptr->name, addr_string); + return NULL; + } + + l_ptr = kzalloc(sizeof(*l_ptr), GFP_ATOMIC); + if (!l_ptr) { + pr_warn("Link creation failed, no memory\n"); + return NULL; + } + + l_ptr->addr = peer; + if_name = strchr(b_ptr->name, ':') + 1; + sprintf(l_ptr->name, "%u.%u.%u:%s-%u.%u.%u:unknown", + tipc_zone(tipc_own_addr), tipc_cluster(tipc_own_addr), + tipc_node(tipc_own_addr), + if_name, + tipc_zone(peer), tipc_cluster(peer), tipc_node(peer)); + /* note: peer i/f name is updated by reset/activate message */ + memcpy(&l_ptr->media_addr, media_addr, sizeof(*media_addr)); + l_ptr->owner = n_ptr; + l_ptr->checkpoint = 1; + l_ptr->peer_session = INVALID_SESSION; + l_ptr->bearer_id = b_ptr->identity; + link_set_supervision_props(l_ptr, b_ptr->tolerance); + l_ptr->state = RESET_UNKNOWN; + + l_ptr->pmsg = (struct tipc_msg *)&l_ptr->proto_msg; + msg = l_ptr->pmsg; + tipc_msg_init(msg, LINK_PROTOCOL, RESET_MSG, INT_H_SIZE, l_ptr->addr); + msg_set_size(msg, sizeof(l_ptr->proto_msg)); + msg_set_session(msg, (tipc_random & 0xffff)); + msg_set_bearer_id(msg, b_ptr->identity); + strcpy((char *)msg_data(msg), if_name); + + l_ptr->priority = b_ptr->priority; + tipc_link_set_queue_limits(l_ptr, b_ptr->window); + + l_ptr->net_plane = b_ptr->net_plane; + link_init_max_pkt(l_ptr); + + l_ptr->next_out_no = 1; + INIT_LIST_HEAD(&l_ptr->waiting_ports); + + link_reset_statistics(l_ptr); + + tipc_node_attach_link(n_ptr, l_ptr); + + k_init_timer(&l_ptr->timer, (Handler)link_timeout, + (unsigned long)l_ptr); + + link_state_event(l_ptr, STARTING_EVT); + + return l_ptr; +} + +void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down) +{ + struct tipc_link *l_ptr; + struct tipc_node *n_ptr; + + rcu_read_lock(); + list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { + tipc_node_lock(n_ptr); + l_ptr = n_ptr->links[bearer_id]; + if (l_ptr) { + tipc_link_reset(l_ptr); + if (shutting_down || !tipc_node_is_up(n_ptr)) { + tipc_node_detach_link(l_ptr->owner, l_ptr); + tipc_link_reset_fragments(l_ptr); + tipc_node_unlock(n_ptr); + + /* Nobody else can access this link now: */ + del_timer_sync(&l_ptr->timer); + kfree(l_ptr); + } else { + /* Detach/delete when failover is finished: */ + l_ptr->flags |= LINK_STOPPED; + tipc_node_unlock(n_ptr); + del_timer_sync(&l_ptr->timer); + } + continue; + } + tipc_node_unlock(n_ptr); + } + rcu_read_unlock(); +} + +/** + * link_schedule_port - schedule port for deferred sending + * @l_ptr: pointer to link + * @origport: reference to sending port + * @sz: amount of data to be sent + * + * Schedules port for renewed sending of messages after link congestion + * has abated. + */ +static int link_schedule_port(struct tipc_link *l_ptr, u32 origport, u32 sz) +{ + struct tipc_port *p_ptr; + + spin_lock_bh(&tipc_port_list_lock); + p_ptr = tipc_port_lock(origport); + if (p_ptr) { + if (!list_empty(&p_ptr->wait_list)) + goto exit; + p_ptr->congested = 1; + p_ptr->waiting_pkts = 1 + ((sz - 1) / l_ptr->max_pkt); + list_add_tail(&p_ptr->wait_list, &l_ptr->waiting_ports); + l_ptr->stats.link_congs++; +exit: + tipc_port_unlock(p_ptr); + } + spin_unlock_bh(&tipc_port_list_lock); + return -ELINKCONG; +} + +void tipc_link_wakeup_ports(struct tipc_link *l_ptr, int all) +{ + struct tipc_port *p_ptr; + struct tipc_port *temp_p_ptr; + int win = l_ptr->queue_limit[0] - l_ptr->out_queue_size; + + if (all) + win = 100000; + if (win <= 0) + return; + if (!spin_trylock_bh(&tipc_port_list_lock)) + return; + if (link_congested(l_ptr)) + goto exit; + list_for_each_entry_safe(p_ptr, temp_p_ptr, &l_ptr->waiting_ports, + wait_list) { + if (win <= 0) + break; + list_del_init(&p_ptr->wait_list); + spin_lock_bh(p_ptr->lock); + p_ptr->congested = 0; + tipc_port_wakeup(p_ptr); + win -= p_ptr->waiting_pkts; + spin_unlock_bh(p_ptr->lock); + } + +exit: + spin_unlock_bh(&tipc_port_list_lock); +} + +/** + * link_release_outqueue - purge link's outbound message queue + * @l_ptr: pointer to link + */ +static void link_release_outqueue(struct tipc_link *l_ptr) +{ + kfree_skb_list(l_ptr->first_out); + l_ptr->first_out = NULL; + l_ptr->out_queue_size = 0; +} + +/** + * tipc_link_reset_fragments - purge link's inbound message fragments queue + * @l_ptr: pointer to link + */ +void tipc_link_reset_fragments(struct tipc_link *l_ptr) +{ + kfree_skb(l_ptr->reasm_buf); + l_ptr->reasm_buf = NULL; +} + +/** + * tipc_link_purge_queues - purge all pkt queues associated with link + * @l_ptr: pointer to link + */ +void tipc_link_purge_queues(struct tipc_link *l_ptr) +{ + kfree_skb_list(l_ptr->oldest_deferred_in); + kfree_skb_list(l_ptr->first_out); + tipc_link_reset_fragments(l_ptr); + kfree_skb(l_ptr->proto_msg_queue); + l_ptr->proto_msg_queue = NULL; +} + +void tipc_link_reset(struct tipc_link *l_ptr) +{ + u32 prev_state = l_ptr->state; + u32 checkpoint = l_ptr->next_in_no; + int was_active_link = tipc_link_is_active(l_ptr); + + msg_set_session(l_ptr->pmsg, ((msg_session(l_ptr->pmsg) + 1) & 0xffff)); + + /* Link is down, accept any session */ + l_ptr->peer_session = INVALID_SESSION; + + /* Prepare for max packet size negotiation */ + link_init_max_pkt(l_ptr); + + l_ptr->state = RESET_UNKNOWN; + + if ((prev_state == RESET_UNKNOWN) || (prev_state == RESET_RESET)) + return; + + tipc_node_link_down(l_ptr->owner, l_ptr); + tipc_bearer_remove_dest(l_ptr->bearer_id, l_ptr->addr); + + if (was_active_link && tipc_node_active_links(l_ptr->owner)) { + l_ptr->reset_checkpoint = checkpoint; + l_ptr->exp_msg_count = START_CHANGEOVER; + } + + /* Clean up all queues: */ + link_release_outqueue(l_ptr); + kfree_skb(l_ptr->proto_msg_queue); + l_ptr->proto_msg_queue = NULL; + kfree_skb_list(l_ptr->oldest_deferred_in); + if (!list_empty(&l_ptr->waiting_ports)) + tipc_link_wakeup_ports(l_ptr, 1); + + l_ptr->retransm_queue_head = 0; + l_ptr->retransm_queue_size = 0; + l_ptr->last_out = NULL; + l_ptr->first_out = NULL; + l_ptr->next_out = NULL; + l_ptr->unacked_window = 0; + l_ptr->checkpoint = 1; + l_ptr->next_out_no = 1; + l_ptr->deferred_inqueue_sz = 0; + l_ptr->oldest_deferred_in = NULL; + l_ptr->newest_deferred_in = NULL; + l_ptr->fsm_msg_cnt = 0; + l_ptr->stale_count = 0; + link_reset_statistics(l_ptr); +} + +void tipc_link_reset_list(unsigned int bearer_id) +{ + struct tipc_link *l_ptr; + struct tipc_node *n_ptr; + + rcu_read_lock(); + list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { + tipc_node_lock(n_ptr); + l_ptr = n_ptr->links[bearer_id]; + if (l_ptr) + tipc_link_reset(l_ptr); + tipc_node_unlock(n_ptr); + } + rcu_read_unlock(); +} + +static void link_activate(struct tipc_link *l_ptr) +{ + l_ptr->next_in_no = l_ptr->stats.recv_info = 1; + tipc_node_link_up(l_ptr->owner, l_ptr); + tipc_bearer_add_dest(l_ptr->bearer_id, l_ptr->addr); +} + +/** + * link_state_event - link finite state machine + * @l_ptr: pointer to link + * @event: state machine event to process + */ +static void link_state_event(struct tipc_link *l_ptr, unsigned int event) +{ + struct tipc_link *other; + u32 cont_intv = l_ptr->continuity_interval; + + if (l_ptr->flags & LINK_STOPPED) + return; + + if (!(l_ptr->flags & LINK_STARTED) && (event != STARTING_EVT)) + return; /* Not yet. */ + + /* Check whether changeover is going on */ + if (l_ptr->exp_msg_count) { + if (event == TIMEOUT_EVT) + link_set_timer(l_ptr, cont_intv); + return; + } + + switch (l_ptr->state) { + case WORKING_WORKING: + switch (event) { + case TRAFFIC_MSG_EVT: + case ACTIVATE_MSG: + break; + case TIMEOUT_EVT: + if (l_ptr->next_in_no != l_ptr->checkpoint) { + l_ptr->checkpoint = l_ptr->next_in_no; + if (tipc_bclink_acks_missing(l_ptr->owner)) { + tipc_link_proto_xmit(l_ptr, STATE_MSG, + 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + } else if (l_ptr->max_pkt < l_ptr->max_pkt_target) { + tipc_link_proto_xmit(l_ptr, STATE_MSG, + 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + } + link_set_timer(l_ptr, cont_intv); + break; + } + l_ptr->state = WORKING_UNKNOWN; + l_ptr->fsm_msg_cnt = 0; + tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv / 4); + break; + case RESET_MSG: + pr_info("%s<%s>, requested by peer\n", link_rst_msg, + l_ptr->name); + tipc_link_reset(l_ptr); + l_ptr->state = RESET_RESET; + l_ptr->fsm_msg_cnt = 0; + tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, + 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + default: + pr_err("%s%u in WW state\n", link_unk_evt, event); + } + break; + case WORKING_UNKNOWN: + switch (event) { + case TRAFFIC_MSG_EVT: + case ACTIVATE_MSG: + l_ptr->state = WORKING_WORKING; + l_ptr->fsm_msg_cnt = 0; + link_set_timer(l_ptr, cont_intv); + break; + case RESET_MSG: + pr_info("%s<%s>, requested by peer while probing\n", + link_rst_msg, l_ptr->name); + tipc_link_reset(l_ptr); + l_ptr->state = RESET_RESET; + l_ptr->fsm_msg_cnt = 0; + tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, + 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + case TIMEOUT_EVT: + if (l_ptr->next_in_no != l_ptr->checkpoint) { + l_ptr->state = WORKING_WORKING; + l_ptr->fsm_msg_cnt = 0; + l_ptr->checkpoint = l_ptr->next_in_no; + if (tipc_bclink_acks_missing(l_ptr->owner)) { + tipc_link_proto_xmit(l_ptr, STATE_MSG, + 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + } + link_set_timer(l_ptr, cont_intv); + } else if (l_ptr->fsm_msg_cnt < l_ptr->abort_limit) { + tipc_link_proto_xmit(l_ptr, STATE_MSG, + 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv / 4); + } else { /* Link has failed */ + pr_warn("%s<%s>, peer not responding\n", + link_rst_msg, l_ptr->name); + tipc_link_reset(l_ptr); + l_ptr->state = RESET_UNKNOWN; + l_ptr->fsm_msg_cnt = 0; + tipc_link_proto_xmit(l_ptr, RESET_MSG, + 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + } + break; + default: + pr_err("%s%u in WU state\n", link_unk_evt, event); + } + break; + case RESET_UNKNOWN: + switch (event) { + case TRAFFIC_MSG_EVT: + break; + case ACTIVATE_MSG: + other = l_ptr->owner->active_links[0]; + if (other && link_working_unknown(other)) + break; + l_ptr->state = WORKING_WORKING; + l_ptr->fsm_msg_cnt = 0; + link_activate(l_ptr); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + if (l_ptr->owner->working_links == 1) + tipc_link_sync_xmit(l_ptr); + link_set_timer(l_ptr, cont_intv); + break; + case RESET_MSG: + l_ptr->state = RESET_RESET; + l_ptr->fsm_msg_cnt = 0; + tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, + 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + case STARTING_EVT: + l_ptr->flags |= LINK_STARTED; + /* fall through */ + case TIMEOUT_EVT: + tipc_link_proto_xmit(l_ptr, RESET_MSG, 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + default: + pr_err("%s%u in RU state\n", link_unk_evt, event); + } + break; + case RESET_RESET: + switch (event) { + case TRAFFIC_MSG_EVT: + case ACTIVATE_MSG: + other = l_ptr->owner->active_links[0]; + if (other && link_working_unknown(other)) + break; + l_ptr->state = WORKING_WORKING; + l_ptr->fsm_msg_cnt = 0; + link_activate(l_ptr); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + if (l_ptr->owner->working_links == 1) + tipc_link_sync_xmit(l_ptr); + link_set_timer(l_ptr, cont_intv); + break; + case RESET_MSG: + break; + case TIMEOUT_EVT: + tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, + 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + default: + pr_err("%s%u in RR state\n", link_unk_evt, event); + } + break; + default: + pr_err("Unknown link state %u/%u\n", l_ptr->state, event); + } +} + +/* + * link_bundle_buf(): Append contents of a buffer to + * the tail of an existing one. + */ +static int link_bundle_buf(struct tipc_link *l_ptr, struct sk_buff *bundler, + struct sk_buff *buf) +{ + struct tipc_msg *bundler_msg = buf_msg(bundler); + struct tipc_msg *msg = buf_msg(buf); + u32 size = msg_size(msg); + u32 bundle_size = msg_size(bundler_msg); + u32 to_pos = align(bundle_size); + u32 pad = to_pos - bundle_size; + + if (msg_user(bundler_msg) != MSG_BUNDLER) + return 0; + if (msg_type(bundler_msg) != OPEN_MSG) + return 0; + if (skb_tailroom(bundler) < (pad + size)) + return 0; + if (l_ptr->max_pkt < (to_pos + size)) + return 0; + + skb_put(bundler, pad + size); + skb_copy_to_linear_data_offset(bundler, to_pos, buf->data, size); + msg_set_size(bundler_msg, to_pos + size); + msg_set_msgcnt(bundler_msg, msg_msgcnt(bundler_msg) + 1); + kfree_skb(buf); + l_ptr->stats.sent_bundled++; + return 1; +} + +static void link_add_to_outqueue(struct tipc_link *l_ptr, + struct sk_buff *buf, + struct tipc_msg *msg) +{ + u32 ack = mod(l_ptr->next_in_no - 1); + u32 seqno = mod(l_ptr->next_out_no++); + + msg_set_word(msg, 2, ((ack << 16) | seqno)); + msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); + buf->next = NULL; + if (l_ptr->first_out) { + l_ptr->last_out->next = buf; + l_ptr->last_out = buf; + } else + l_ptr->first_out = l_ptr->last_out = buf; + + l_ptr->out_queue_size++; + if (l_ptr->out_queue_size > l_ptr->stats.max_queue_sz) + l_ptr->stats.max_queue_sz = l_ptr->out_queue_size; +} + +static void link_add_chain_to_outqueue(struct tipc_link *l_ptr, + struct sk_buff *buf_chain, + u32 long_msgno) +{ + struct sk_buff *buf; + struct tipc_msg *msg; + + if (!l_ptr->next_out) + l_ptr->next_out = buf_chain; + while (buf_chain) { + buf = buf_chain; + buf_chain = buf_chain->next; + + msg = buf_msg(buf); + msg_set_long_msgno(msg, long_msgno); + link_add_to_outqueue(l_ptr, buf, msg); + } +} + +/* + * tipc_link_xmit() is the 'full path' for messages, called from + * inside TIPC when the 'fast path' in tipc_send_xmit + * has failed, and from link_send() + */ +int __tipc_link_xmit(struct tipc_link *l_ptr, struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + u32 size = msg_size(msg); + u32 dsz = msg_data_sz(msg); + u32 queue_size = l_ptr->out_queue_size; + u32 imp = tipc_msg_tot_importance(msg); + u32 queue_limit = l_ptr->queue_limit[imp]; + u32 max_packet = l_ptr->max_pkt; + + /* Match msg importance against queue limits: */ + if (unlikely(queue_size >= queue_limit)) { + if (imp <= TIPC_CRITICAL_IMPORTANCE) { + link_schedule_port(l_ptr, msg_origport(msg), size); + kfree_skb(buf); + return -ELINKCONG; + } + kfree_skb(buf); + if (imp > CONN_MANAGER) { + pr_warn("%s<%s>, send queue full", link_rst_msg, + l_ptr->name); + tipc_link_reset(l_ptr); + } + return dsz; + } + + /* Fragmentation needed ? */ + if (size > max_packet) + return tipc_link_frag_xmit(l_ptr, buf); + + /* Packet can be queued or sent. */ + if (likely(!link_congested(l_ptr))) { + link_add_to_outqueue(l_ptr, buf, msg); + + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); + l_ptr->unacked_window = 0; + return dsz; + } + /* Congestion: can message be bundled ? */ + if ((msg_user(msg) != CHANGEOVER_PROTOCOL) && + (msg_user(msg) != MSG_FRAGMENTER)) { + + /* Try adding message to an existing bundle */ + if (l_ptr->next_out && + link_bundle_buf(l_ptr, l_ptr->last_out, buf)) + return dsz; + + /* Try creating a new bundle */ + if (size <= max_packet * 2 / 3) { + struct sk_buff *bundler = tipc_buf_acquire(max_packet); + struct tipc_msg bundler_hdr; + + if (bundler) { + tipc_msg_init(&bundler_hdr, MSG_BUNDLER, OPEN_MSG, + INT_H_SIZE, l_ptr->addr); + skb_copy_to_linear_data(bundler, &bundler_hdr, + INT_H_SIZE); + skb_trim(bundler, INT_H_SIZE); + link_bundle_buf(l_ptr, bundler, buf); + buf = bundler; + msg = buf_msg(buf); + l_ptr->stats.sent_bundles++; + } + } + } + if (!l_ptr->next_out) + l_ptr->next_out = buf; + link_add_to_outqueue(l_ptr, buf, msg); + return dsz; +} + +/* + * tipc_link_xmit(): same as __tipc_link_xmit(), but the link to use + * has not been selected yet, and the the owner node is not locked + * Called by TIPC internal users, e.g. the name distributor + */ +int tipc_link_xmit(struct sk_buff *buf, u32 dest, u32 selector) +{ + struct tipc_link *l_ptr; + struct tipc_node *n_ptr; + int res = -ELINKCONG; + + n_ptr = tipc_node_find(dest); + if (n_ptr) { + tipc_node_lock(n_ptr); + l_ptr = n_ptr->active_links[selector & 1]; + if (l_ptr) + res = __tipc_link_xmit(l_ptr, buf); + else + kfree_skb(buf); + tipc_node_unlock(n_ptr); + } else { + kfree_skb(buf); + } + return res; +} + +/* + * tipc_link_sync_xmit - synchronize broadcast link endpoints. + * + * Give a newly added peer node the sequence number where it should + * start receiving and acking broadcast packets. + * + * Called with node locked + */ +static void tipc_link_sync_xmit(struct tipc_link *l) +{ + struct sk_buff *buf; + struct tipc_msg *msg; + + buf = tipc_buf_acquire(INT_H_SIZE); + if (!buf) + return; + + msg = buf_msg(buf); + tipc_msg_init(msg, BCAST_PROTOCOL, STATE_MSG, INT_H_SIZE, l->addr); + msg_set_last_bcast(msg, l->owner->bclink.acked); + link_add_chain_to_outqueue(l, buf, 0); + tipc_link_push_queue(l); +} + +/* + * tipc_link_sync_rcv - synchronize broadcast link endpoints. + * Receive the sequence number where we should start receiving and + * acking broadcast packets from a newly added peer node, and open + * up for reception of such packets. + * + * Called with node locked + */ +static void tipc_link_sync_rcv(struct tipc_node *n, struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + + n->bclink.last_sent = n->bclink.last_in = msg_last_bcast(msg); + n->bclink.recv_permitted = true; + kfree_skb(buf); +} + +/* + * tipc_link_names_xmit - send name table entries to new neighbor + * + * Send routine for bulk delivery of name table messages when contact + * with a new neighbor occurs. No link congestion checking is performed + * because name table messages *must* be delivered. The messages must be + * small enough not to require fragmentation. + * Called without any locks held. + */ +void tipc_link_names_xmit(struct list_head *message_list, u32 dest) +{ + struct tipc_node *n_ptr; + struct tipc_link *l_ptr; + struct sk_buff *buf; + struct sk_buff *temp_buf; + + if (list_empty(message_list)) + return; + + n_ptr = tipc_node_find(dest); + if (n_ptr) { + tipc_node_lock(n_ptr); + l_ptr = n_ptr->active_links[0]; + if (l_ptr) { + /* convert circular list to linear list */ + ((struct sk_buff *)message_list->prev)->next = NULL; + link_add_chain_to_outqueue(l_ptr, + (struct sk_buff *)message_list->next, 0); + tipc_link_push_queue(l_ptr); + INIT_LIST_HEAD(message_list); + } + tipc_node_unlock(n_ptr); + } + + /* discard the messages if they couldn't be sent */ + list_for_each_safe(buf, temp_buf, ((struct sk_buff *)message_list)) { + list_del((struct list_head *)buf); + kfree_skb(buf); + } +} + +/* + * tipc_link_xmit_fast: Entry for data messages where the + * destination link is known and the header is complete, + * inclusive total message length. Very time critical. + * Link is locked. Returns user data length. + */ +static int tipc_link_xmit_fast(struct tipc_link *l_ptr, struct sk_buff *buf, + u32 *used_max_pkt) +{ + struct tipc_msg *msg = buf_msg(buf); + int res = msg_data_sz(msg); + + if (likely(!link_congested(l_ptr))) { + if (likely(msg_size(msg) <= l_ptr->max_pkt)) { + link_add_to_outqueue(l_ptr, buf, msg); + tipc_bearer_send(l_ptr->bearer_id, buf, + &l_ptr->media_addr); + l_ptr->unacked_window = 0; + return res; + } + else + *used_max_pkt = l_ptr->max_pkt; + } + return __tipc_link_xmit(l_ptr, buf); /* All other cases */ +} + +/* + * tipc_link_iovec_xmit_fast: Entry for messages where the + * destination processor is known and the header is complete, + * except for total message length. + * Returns user data length or errno. + */ +int tipc_link_iovec_xmit_fast(struct tipc_port *sender, + struct iovec const *msg_sect, + unsigned int len, u32 destaddr) +{ + struct tipc_msg *hdr = &sender->phdr; + struct tipc_link *l_ptr; + struct sk_buff *buf; + struct tipc_node *node; + int res; + u32 selector = msg_origport(hdr) & 1; + +again: + /* + * Try building message using port's max_pkt hint. + * (Must not hold any locks while building message.) + */ + res = tipc_msg_build(hdr, msg_sect, len, sender->max_pkt, &buf); + /* Exit if build request was invalid */ + if (unlikely(res < 0)) + return res; + + node = tipc_node_find(destaddr); + if (likely(node)) { + tipc_node_lock(node); + l_ptr = node->active_links[selector]; + if (likely(l_ptr)) { + if (likely(buf)) { + res = tipc_link_xmit_fast(l_ptr, buf, + &sender->max_pkt); +exit: + tipc_node_unlock(node); + return res; + } + + /* Exit if link (or bearer) is congested */ + if (link_congested(l_ptr)) { + res = link_schedule_port(l_ptr, + sender->ref, res); + goto exit; + } + + /* + * Message size exceeds max_pkt hint; update hint, + * then re-try fast path or fragment the message + */ + sender->max_pkt = l_ptr->max_pkt; + tipc_node_unlock(node); + + + if ((msg_hdr_sz(hdr) + res) <= sender->max_pkt) + goto again; + + return tipc_link_iovec_long_xmit(sender, msg_sect, + len, destaddr); + } + tipc_node_unlock(node); + } + + /* Couldn't find a link to the destination node */ + kfree_skb(buf); + tipc_port_iovec_reject(sender, hdr, msg_sect, len, TIPC_ERR_NO_NODE); + return -ENETUNREACH; +} + +/* + * tipc_link_iovec_long_xmit(): Entry for long messages where the + * destination node is known and the header is complete, + * inclusive total message length. + * Link and bearer congestion status have been checked to be ok, + * and are ignored if they change. + * + * Note that fragments do not use the full link MTU so that they won't have + * to undergo refragmentation if link changeover causes them to be sent + * over another link with an additional tunnel header added as prefix. + * (Refragmentation will still occur if the other link has a smaller MTU.) + * + * Returns user data length or errno. + */ +static int tipc_link_iovec_long_xmit(struct tipc_port *sender, + struct iovec const *msg_sect, + unsigned int len, u32 destaddr) +{ + struct tipc_link *l_ptr; + struct tipc_node *node; + struct tipc_msg *hdr = &sender->phdr; + u32 dsz = len; + u32 max_pkt, fragm_sz, rest; + struct tipc_msg fragm_hdr; + struct sk_buff *buf, *buf_chain, *prev; + u32 fragm_crs, fragm_rest, hsz, sect_rest; + const unchar __user *sect_crs; + int curr_sect; + u32 fragm_no; + int res = 0; + +again: + fragm_no = 1; + max_pkt = sender->max_pkt - INT_H_SIZE; + /* leave room for tunnel header in case of link changeover */ + fragm_sz = max_pkt - INT_H_SIZE; + /* leave room for fragmentation header in each fragment */ + rest = dsz; + fragm_crs = 0; + fragm_rest = 0; + sect_rest = 0; + sect_crs = NULL; + curr_sect = -1; + + /* Prepare reusable fragment header */ + tipc_msg_init(&fragm_hdr, MSG_FRAGMENTER, FIRST_FRAGMENT, + INT_H_SIZE, msg_destnode(hdr)); + msg_set_size(&fragm_hdr, max_pkt); + msg_set_fragm_no(&fragm_hdr, 1); + + /* Prepare header of first fragment */ + buf_chain = buf = tipc_buf_acquire(max_pkt); + if (!buf) + return -ENOMEM; + buf->next = NULL; + skb_copy_to_linear_data(buf, &fragm_hdr, INT_H_SIZE); + hsz = msg_hdr_sz(hdr); + skb_copy_to_linear_data_offset(buf, INT_H_SIZE, hdr, hsz); + + /* Chop up message */ + fragm_crs = INT_H_SIZE + hsz; + fragm_rest = fragm_sz - hsz; + + do { /* For all sections */ + u32 sz; + + if (!sect_rest) { + sect_rest = msg_sect[++curr_sect].iov_len; + sect_crs = msg_sect[curr_sect].iov_base; + } + + if (sect_rest < fragm_rest) + sz = sect_rest; + else + sz = fragm_rest; + + if (copy_from_user(buf->data + fragm_crs, sect_crs, sz)) { + res = -EFAULT; +error: + kfree_skb_list(buf_chain); + return res; + } + sect_crs += sz; + sect_rest -= sz; + fragm_crs += sz; + fragm_rest -= sz; + rest -= sz; + + if (!fragm_rest && rest) { + + /* Initiate new fragment: */ + if (rest <= fragm_sz) { + fragm_sz = rest; + msg_set_type(&fragm_hdr, LAST_FRAGMENT); + } else { + msg_set_type(&fragm_hdr, FRAGMENT); + } + msg_set_size(&fragm_hdr, fragm_sz + INT_H_SIZE); + msg_set_fragm_no(&fragm_hdr, ++fragm_no); + prev = buf; + buf = tipc_buf_acquire(fragm_sz + INT_H_SIZE); + if (!buf) { + res = -ENOMEM; + goto error; + } + + buf->next = NULL; + prev->next = buf; + skb_copy_to_linear_data(buf, &fragm_hdr, INT_H_SIZE); + fragm_crs = INT_H_SIZE; + fragm_rest = fragm_sz; + } + } while (rest > 0); + + /* + * Now we have a buffer chain. Select a link and check + * that packet size is still OK + */ + node = tipc_node_find(destaddr); + if (likely(node)) { + tipc_node_lock(node); + l_ptr = node->active_links[sender->ref & 1]; + if (!l_ptr) { + tipc_node_unlock(node); + goto reject; + } + if (l_ptr->max_pkt < max_pkt) { + sender->max_pkt = l_ptr->max_pkt; + tipc_node_unlock(node); + kfree_skb_list(buf_chain); + goto again; + } + } else { +reject: + kfree_skb_list(buf_chain); + tipc_port_iovec_reject(sender, hdr, msg_sect, len, + TIPC_ERR_NO_NODE); + return -ENETUNREACH; + } + + /* Append chain of fragments to send queue & send them */ + l_ptr->long_msg_seq_no++; + link_add_chain_to_outqueue(l_ptr, buf_chain, l_ptr->long_msg_seq_no); + l_ptr->stats.sent_fragments += fragm_no; + l_ptr->stats.sent_fragmented++; + tipc_link_push_queue(l_ptr); + tipc_node_unlock(node); + return dsz; +} + +/* + * tipc_link_push_packet: Push one unsent packet to the media + */ +static u32 tipc_link_push_packet(struct tipc_link *l_ptr) +{ + struct sk_buff *buf = l_ptr->first_out; + u32 r_q_size = l_ptr->retransm_queue_size; + u32 r_q_head = l_ptr->retransm_queue_head; + + /* Step to position where retransmission failed, if any, */ + /* consider that buffers may have been released in meantime */ + if (r_q_size && buf) { + u32 last = lesser(mod(r_q_head + r_q_size), + link_last_sent(l_ptr)); + u32 first = buf_seqno(buf); + + while (buf && less(first, r_q_head)) { + first = mod(first + 1); + buf = buf->next; + } + l_ptr->retransm_queue_head = r_q_head = first; + l_ptr->retransm_queue_size = r_q_size = mod(last - first); + } + + /* Continue retransmission now, if there is anything: */ + if (r_q_size && buf) { + msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); + msg_set_bcast_ack(buf_msg(buf), l_ptr->owner->bclink.last_in); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); + l_ptr->retransm_queue_head = mod(++r_q_head); + l_ptr->retransm_queue_size = --r_q_size; + l_ptr->stats.retransmitted++; + return 0; + } + + /* Send deferred protocol message, if any: */ + buf = l_ptr->proto_msg_queue; + if (buf) { + msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); + msg_set_bcast_ack(buf_msg(buf), l_ptr->owner->bclink.last_in); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); + l_ptr->unacked_window = 0; + kfree_skb(buf); + l_ptr->proto_msg_queue = NULL; + return 0; + } + + /* Send one deferred data message, if send window not full: */ + buf = l_ptr->next_out; + if (buf) { + struct tipc_msg *msg = buf_msg(buf); + u32 next = msg_seqno(msg); + u32 first = buf_seqno(l_ptr->first_out); + + if (mod(next - first) < l_ptr->queue_limit[0]) { + msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); + msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); + tipc_bearer_send(l_ptr->bearer_id, buf, + &l_ptr->media_addr); + if (msg_user(msg) == MSG_BUNDLER) + msg_set_type(msg, CLOSED_MSG); + l_ptr->next_out = buf->next; + return 0; + } + } + return 1; +} + +/* + * push_queue(): push out the unsent messages of a link where + * congestion has abated. Node is locked + */ +void tipc_link_push_queue(struct tipc_link *l_ptr) +{ + u32 res; + + do { + res = tipc_link_push_packet(l_ptr); + } while (!res); +} + +void tipc_link_reset_all(struct tipc_node *node) +{ + char addr_string[16]; + u32 i; + + tipc_node_lock(node); + + pr_warn("Resetting all links to %s\n", + tipc_addr_string_fill(addr_string, node->addr)); + + for (i = 0; i < MAX_BEARERS; i++) { + if (node->links[i]) { + link_print(node->links[i], "Resetting link\n"); + tipc_link_reset(node->links[i]); + } + } + + tipc_node_unlock(node); +} + +static void link_retransmit_failure(struct tipc_link *l_ptr, + struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + + pr_warn("Retransmission failure on link <%s>\n", l_ptr->name); + + if (l_ptr->addr) { + /* Handle failure on standard link */ + link_print(l_ptr, "Resetting link\n"); + tipc_link_reset(l_ptr); + + } else { + /* Handle failure on broadcast link */ + struct tipc_node *n_ptr; + char addr_string[16]; + + pr_info("Msg seq number: %u, ", msg_seqno(msg)); + pr_cont("Outstanding acks: %lu\n", + (unsigned long) TIPC_SKB_CB(buf)->handle); + + n_ptr = tipc_bclink_retransmit_to(); + tipc_node_lock(n_ptr); + + tipc_addr_string_fill(addr_string, n_ptr->addr); + pr_info("Broadcast link info for %s\n", addr_string); + pr_info("Reception permitted: %d, Acked: %u\n", + n_ptr->bclink.recv_permitted, + n_ptr->bclink.acked); + pr_info("Last in: %u, Oos state: %u, Last sent: %u\n", + n_ptr->bclink.last_in, + n_ptr->bclink.oos_state, + n_ptr->bclink.last_sent); + + tipc_node_unlock(n_ptr); + + tipc_bclink_set_flags(TIPC_BCLINK_RESET); + l_ptr->stale_count = 0; + } +} + +void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *buf, + u32 retransmits) +{ + struct tipc_msg *msg; + + if (!buf) + return; + + msg = buf_msg(buf); + + /* Detect repeated retransmit failures */ + if (l_ptr->last_retransmitted == msg_seqno(msg)) { + if (++l_ptr->stale_count > 100) { + link_retransmit_failure(l_ptr, buf); + return; + } + } else { + l_ptr->last_retransmitted = msg_seqno(msg); + l_ptr->stale_count = 1; + } + + while (retransmits && (buf != l_ptr->next_out) && buf) { + msg = buf_msg(buf); + msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); + msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); + buf = buf->next; + retransmits--; + l_ptr->stats.retransmitted++; + } + + l_ptr->retransm_queue_head = l_ptr->retransm_queue_size = 0; +} + +/** + * link_insert_deferred_queue - insert deferred messages back into receive chain + */ +static struct sk_buff *link_insert_deferred_queue(struct tipc_link *l_ptr, + struct sk_buff *buf) +{ + u32 seq_no; + + if (l_ptr->oldest_deferred_in == NULL) + return buf; + + seq_no = buf_seqno(l_ptr->oldest_deferred_in); + if (seq_no == mod(l_ptr->next_in_no)) { + l_ptr->newest_deferred_in->next = buf; + buf = l_ptr->oldest_deferred_in; + l_ptr->oldest_deferred_in = NULL; + l_ptr->deferred_inqueue_sz = 0; + } + return buf; +} + +/** + * link_recv_buf_validate - validate basic format of received message + * + * This routine ensures a TIPC message has an acceptable header, and at least + * as much data as the header indicates it should. The routine also ensures + * that the entire message header is stored in the main fragment of the message + * buffer, to simplify future access to message header fields. + * + * Note: Having extra info present in the message header or data areas is OK. + * TIPC will ignore the excess, under the assumption that it is optional info + * introduced by a later release of the protocol. + */ +static int link_recv_buf_validate(struct sk_buff *buf) +{ + static u32 min_data_hdr_size[8] = { + SHORT_H_SIZE, MCAST_H_SIZE, NAMED_H_SIZE, BASIC_H_SIZE, + MAX_H_SIZE, MAX_H_SIZE, MAX_H_SIZE, MAX_H_SIZE + }; + + struct tipc_msg *msg; + u32 tipc_hdr[2]; + u32 size; + u32 hdr_size; + u32 min_hdr_size; + + /* If this packet comes from the defer queue, the skb has already + * been validated + */ + if (unlikely(TIPC_SKB_CB(buf)->deferred)) + return 1; + + if (unlikely(buf->len < MIN_H_SIZE)) + return 0; + + msg = skb_header_pointer(buf, 0, sizeof(tipc_hdr), tipc_hdr); + if (msg == NULL) + return 0; + + if (unlikely(msg_version(msg) != TIPC_VERSION)) + return 0; + + size = msg_size(msg); + hdr_size = msg_hdr_sz(msg); + min_hdr_size = msg_isdata(msg) ? + min_data_hdr_size[msg_type(msg)] : INT_H_SIZE; + + if (unlikely((hdr_size < min_hdr_size) || + (size < hdr_size) || + (buf->len < size) || + (size - hdr_size > TIPC_MAX_USER_MSG_SIZE))) + return 0; + + return pskb_may_pull(buf, hdr_size); +} + +/** + * tipc_rcv - process TIPC packets/messages arriving from off-node + * @head: pointer to message buffer chain + * @b_ptr: pointer to bearer message arrived on + * + * Invoked with no locks held. Bearer pointer must point to a valid bearer + * structure (i.e. cannot be NULL), but bearer can be inactive. + */ +void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr) +{ + while (head) { + struct tipc_node *n_ptr; + struct tipc_link *l_ptr; + struct sk_buff *crs; + struct sk_buff *buf = head; + struct tipc_msg *msg; + u32 seq_no; + u32 ackd; + u32 released = 0; + + head = head->next; + buf->next = NULL; + + /* Ensure message is well-formed */ + if (unlikely(!link_recv_buf_validate(buf))) + goto discard; + + /* Ensure message data is a single contiguous unit */ + if (unlikely(skb_linearize(buf))) + goto discard; + + /* Handle arrival of a non-unicast link message */ + msg = buf_msg(buf); + + if (unlikely(msg_non_seq(msg))) { + if (msg_user(msg) == LINK_CONFIG) + tipc_disc_rcv(buf, b_ptr); + else + tipc_bclink_rcv(buf); + continue; + } + + /* Discard unicast link messages destined for another node */ + if (unlikely(!msg_short(msg) && + (msg_destnode(msg) != tipc_own_addr))) + goto discard; + + /* Locate neighboring node that sent message */ + n_ptr = tipc_node_find(msg_prevnode(msg)); + if (unlikely(!n_ptr)) + goto discard; + tipc_node_lock(n_ptr); + + /* Locate unicast link endpoint that should handle message */ + l_ptr = n_ptr->links[b_ptr->identity]; + if (unlikely(!l_ptr)) + goto unlock_discard; + + /* Verify that communication with node is currently allowed */ + if ((n_ptr->action_flags & TIPC_WAIT_PEER_LINKS_DOWN) && + msg_user(msg) == LINK_PROTOCOL && + (msg_type(msg) == RESET_MSG || + msg_type(msg) == ACTIVATE_MSG) && + !msg_redundant_link(msg)) + n_ptr->action_flags &= ~TIPC_WAIT_PEER_LINKS_DOWN; + + if (tipc_node_blocked(n_ptr)) + goto unlock_discard; + + /* Validate message sequence number info */ + seq_no = msg_seqno(msg); + ackd = msg_ack(msg); + + /* Release acked messages */ + if (n_ptr->bclink.recv_permitted) + tipc_bclink_acknowledge(n_ptr, msg_bcast_ack(msg)); + + crs = l_ptr->first_out; + while ((crs != l_ptr->next_out) && + less_eq(buf_seqno(crs), ackd)) { + struct sk_buff *next = crs->next; + kfree_skb(crs); + crs = next; + released++; + } + if (released) { + l_ptr->first_out = crs; + l_ptr->out_queue_size -= released; + } + + /* Try sending any messages link endpoint has pending */ + if (unlikely(l_ptr->next_out)) + tipc_link_push_queue(l_ptr); + + if (unlikely(!list_empty(&l_ptr->waiting_ports))) + tipc_link_wakeup_ports(l_ptr, 0); + + if (unlikely(++l_ptr->unacked_window >= TIPC_MIN_LINK_WIN)) { + l_ptr->stats.sent_acks++; + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + } + + /* Process the incoming packet */ + if (unlikely(!link_working_working(l_ptr))) { + if (msg_user(msg) == LINK_PROTOCOL) { + tipc_link_proto_rcv(l_ptr, buf); + head = link_insert_deferred_queue(l_ptr, head); + tipc_node_unlock(n_ptr); + continue; + } + + /* Traffic message. Conditionally activate link */ + link_state_event(l_ptr, TRAFFIC_MSG_EVT); + + if (link_working_working(l_ptr)) { + /* Re-insert buffer in front of queue */ + buf->next = head; + head = buf; + tipc_node_unlock(n_ptr); + continue; + } + goto unlock_discard; + } + + /* Link is now in state WORKING_WORKING */ + if (unlikely(seq_no != mod(l_ptr->next_in_no))) { + link_handle_out_of_seq_msg(l_ptr, buf); + head = link_insert_deferred_queue(l_ptr, head); + tipc_node_unlock(n_ptr); + continue; + } + l_ptr->next_in_no++; + if (unlikely(l_ptr->oldest_deferred_in)) + head = link_insert_deferred_queue(l_ptr, head); + + /* Deliver packet/message to correct user: */ + if (unlikely(msg_user(msg) == CHANGEOVER_PROTOCOL)) { + if (!tipc_link_tunnel_rcv(n_ptr, &buf)) { + tipc_node_unlock(n_ptr); + continue; + } + msg = buf_msg(buf); + } else if (msg_user(msg) == MSG_FRAGMENTER) { + l_ptr->stats.recv_fragments++; + if (tipc_buf_append(&l_ptr->reasm_buf, &buf)) { + l_ptr->stats.recv_fragmented++; + msg = buf_msg(buf); + } else { + if (!l_ptr->reasm_buf) + tipc_link_reset(l_ptr); + tipc_node_unlock(n_ptr); + continue; + } + } + + switch (msg_user(msg)) { + case TIPC_LOW_IMPORTANCE: + case TIPC_MEDIUM_IMPORTANCE: + case TIPC_HIGH_IMPORTANCE: + case TIPC_CRITICAL_IMPORTANCE: + tipc_node_unlock(n_ptr); + tipc_sk_rcv(buf); + continue; + case MSG_BUNDLER: + l_ptr->stats.recv_bundles++; + l_ptr->stats.recv_bundled += msg_msgcnt(msg); + tipc_node_unlock(n_ptr); + tipc_link_bundle_rcv(buf); + continue; + case NAME_DISTRIBUTOR: + n_ptr->bclink.recv_permitted = true; + tipc_node_unlock(n_ptr); + tipc_named_rcv(buf); + continue; + case CONN_MANAGER: + tipc_node_unlock(n_ptr); + tipc_port_proto_rcv(buf); + continue; + case BCAST_PROTOCOL: + tipc_link_sync_rcv(n_ptr, buf); + break; + default: + kfree_skb(buf); + break; + } + tipc_node_unlock(n_ptr); + continue; +unlock_discard: + tipc_node_unlock(n_ptr); +discard: + kfree_skb(buf); + } +} + +/** + * tipc_link_defer_pkt - Add out-of-sequence message to deferred reception queue + * + * Returns increase in queue length (i.e. 0 or 1) + */ +u32 tipc_link_defer_pkt(struct sk_buff **head, struct sk_buff **tail, + struct sk_buff *buf) +{ + struct sk_buff *queue_buf; + struct sk_buff **prev; + u32 seq_no = buf_seqno(buf); + + buf->next = NULL; + + /* Empty queue ? */ + if (*head == NULL) { + *head = *tail = buf; + return 1; + } + + /* Last ? */ + if (less(buf_seqno(*tail), seq_no)) { + (*tail)->next = buf; + *tail = buf; + return 1; + } + + /* Locate insertion point in queue, then insert; discard if duplicate */ + prev = head; + queue_buf = *head; + for (;;) { + u32 curr_seqno = buf_seqno(queue_buf); + + if (seq_no == curr_seqno) { + kfree_skb(buf); + return 0; + } + + if (less(seq_no, curr_seqno)) + break; + + prev = &queue_buf->next; + queue_buf = queue_buf->next; + } + + buf->next = queue_buf; + *prev = buf; + return 1; +} + +/* + * link_handle_out_of_seq_msg - handle arrival of out-of-sequence packet + */ +static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, + struct sk_buff *buf) +{ + u32 seq_no = buf_seqno(buf); + + if (likely(msg_user(buf_msg(buf)) == LINK_PROTOCOL)) { + tipc_link_proto_rcv(l_ptr, buf); + return; + } + + /* Record OOS packet arrival (force mismatch on next timeout) */ + l_ptr->checkpoint--; + + /* + * Discard packet if a duplicate; otherwise add it to deferred queue + * and notify peer of gap as per protocol specification + */ + if (less(seq_no, mod(l_ptr->next_in_no))) { + l_ptr->stats.duplicates++; + kfree_skb(buf); + return; + } + + if (tipc_link_defer_pkt(&l_ptr->oldest_deferred_in, + &l_ptr->newest_deferred_in, buf)) { + l_ptr->deferred_inqueue_sz++; + l_ptr->stats.deferred_recv++; + TIPC_SKB_CB(buf)->deferred = true; + if ((l_ptr->deferred_inqueue_sz % 16) == 1) + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + } else + l_ptr->stats.duplicates++; +} + +/* + * Send protocol message to the other endpoint. + */ +void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int probe_msg, + u32 gap, u32 tolerance, u32 priority, u32 ack_mtu) +{ + struct sk_buff *buf = NULL; + struct tipc_msg *msg = l_ptr->pmsg; + u32 msg_size = sizeof(l_ptr->proto_msg); + int r_flag; + + /* Discard any previous message that was deferred due to congestion */ + if (l_ptr->proto_msg_queue) { + kfree_skb(l_ptr->proto_msg_queue); + l_ptr->proto_msg_queue = NULL; + } + + /* Don't send protocol message during link changeover */ + if (l_ptr->exp_msg_count) + return; + + /* Abort non-RESET send if communication with node is prohibited */ + if ((tipc_node_blocked(l_ptr->owner)) && (msg_typ != RESET_MSG)) + return; + + /* Create protocol message with "out-of-sequence" sequence number */ + msg_set_type(msg, msg_typ); + msg_set_net_plane(msg, l_ptr->net_plane); + msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); + msg_set_last_bcast(msg, tipc_bclink_get_last_sent()); + + if (msg_typ == STATE_MSG) { + u32 next_sent = mod(l_ptr->next_out_no); + + if (!tipc_link_is_up(l_ptr)) + return; + if (l_ptr->next_out) + next_sent = buf_seqno(l_ptr->next_out); + msg_set_next_sent(msg, next_sent); + if (l_ptr->oldest_deferred_in) { + u32 rec = buf_seqno(l_ptr->oldest_deferred_in); + gap = mod(rec - mod(l_ptr->next_in_no)); + } + msg_set_seq_gap(msg, gap); + if (gap) + l_ptr->stats.sent_nacks++; + msg_set_link_tolerance(msg, tolerance); + msg_set_linkprio(msg, priority); + msg_set_max_pkt(msg, ack_mtu); + msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); + msg_set_probe(msg, probe_msg != 0); + if (probe_msg) { + u32 mtu = l_ptr->max_pkt; + + if ((mtu < l_ptr->max_pkt_target) && + link_working_working(l_ptr) && + l_ptr->fsm_msg_cnt) { + msg_size = (mtu + (l_ptr->max_pkt_target - mtu)/2 + 2) & ~3; + if (l_ptr->max_pkt_probes == 10) { + l_ptr->max_pkt_target = (msg_size - 4); + l_ptr->max_pkt_probes = 0; + msg_size = (mtu + (l_ptr->max_pkt_target - mtu)/2 + 2) & ~3; + } + l_ptr->max_pkt_probes++; + } + + l_ptr->stats.sent_probes++; + } + l_ptr->stats.sent_states++; + } else { /* RESET_MSG or ACTIVATE_MSG */ + msg_set_ack(msg, mod(l_ptr->reset_checkpoint - 1)); + msg_set_seq_gap(msg, 0); + msg_set_next_sent(msg, 1); + msg_set_probe(msg, 0); + msg_set_link_tolerance(msg, l_ptr->tolerance); + msg_set_linkprio(msg, l_ptr->priority); + msg_set_max_pkt(msg, l_ptr->max_pkt_target); + } + + r_flag = (l_ptr->owner->working_links > tipc_link_is_up(l_ptr)); + msg_set_redundant_link(msg, r_flag); + msg_set_linkprio(msg, l_ptr->priority); + msg_set_size(msg, msg_size); + + msg_set_seqno(msg, mod(l_ptr->next_out_no + (0xffff/2))); + + buf = tipc_buf_acquire(msg_size); + if (!buf) + return; + + skb_copy_to_linear_data(buf, msg, sizeof(l_ptr->proto_msg)); + buf->priority = TC_PRIO_CONTROL; + + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); + l_ptr->unacked_window = 0; + kfree_skb(buf); +} + +/* + * Receive protocol message : + * Note that network plane id propagates through the network, and may + * change at any time. The node with lowest address rules + */ +static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) +{ + u32 rec_gap = 0; + u32 max_pkt_info; + u32 max_pkt_ack; + u32 msg_tol; + struct tipc_msg *msg = buf_msg(buf); + + /* Discard protocol message during link changeover */ + if (l_ptr->exp_msg_count) + goto exit; + + if (l_ptr->net_plane != msg_net_plane(msg)) + if (tipc_own_addr > msg_prevnode(msg)) + l_ptr->net_plane = msg_net_plane(msg); + + switch (msg_type(msg)) { + + case RESET_MSG: + if (!link_working_unknown(l_ptr) && + (l_ptr->peer_session != INVALID_SESSION)) { + if (less_eq(msg_session(msg), l_ptr->peer_session)) + break; /* duplicate or old reset: ignore */ + } + + if (!msg_redundant_link(msg) && (link_working_working(l_ptr) || + link_working_unknown(l_ptr))) { + /* + * peer has lost contact -- don't allow peer's links + * to reactivate before we recognize loss & clean up + */ + l_ptr->owner->action_flags |= TIPC_WAIT_OWN_LINKS_DOWN; + } + + link_state_event(l_ptr, RESET_MSG); + + /* fall thru' */ + case ACTIVATE_MSG: + /* Update link settings according other endpoint's values */ + strcpy((strrchr(l_ptr->name, ':') + 1), (char *)msg_data(msg)); + + msg_tol = msg_link_tolerance(msg); + if (msg_tol > l_ptr->tolerance) + link_set_supervision_props(l_ptr, msg_tol); + + if (msg_linkprio(msg) > l_ptr->priority) + l_ptr->priority = msg_linkprio(msg); + + max_pkt_info = msg_max_pkt(msg); + if (max_pkt_info) { + if (max_pkt_info < l_ptr->max_pkt_target) + l_ptr->max_pkt_target = max_pkt_info; + if (l_ptr->max_pkt > l_ptr->max_pkt_target) + l_ptr->max_pkt = l_ptr->max_pkt_target; + } else { + l_ptr->max_pkt = l_ptr->max_pkt_target; + } + + /* Synchronize broadcast link info, if not done previously */ + if (!tipc_node_is_up(l_ptr->owner)) { + l_ptr->owner->bclink.last_sent = + l_ptr->owner->bclink.last_in = + msg_last_bcast(msg); + l_ptr->owner->bclink.oos_state = 0; + } + + l_ptr->peer_session = msg_session(msg); + l_ptr->peer_bearer_id = msg_bearer_id(msg); + + if (msg_type(msg) == ACTIVATE_MSG) + link_state_event(l_ptr, ACTIVATE_MSG); + break; + case STATE_MSG: + + msg_tol = msg_link_tolerance(msg); + if (msg_tol) + link_set_supervision_props(l_ptr, msg_tol); + + if (msg_linkprio(msg) && + (msg_linkprio(msg) != l_ptr->priority)) { + pr_warn("%s<%s>, priority change %u->%u\n", + link_rst_msg, l_ptr->name, l_ptr->priority, + msg_linkprio(msg)); + l_ptr->priority = msg_linkprio(msg); + tipc_link_reset(l_ptr); /* Enforce change to take effect */ + break; + } + + /* Record reception; force mismatch at next timeout: */ + l_ptr->checkpoint--; + + link_state_event(l_ptr, TRAFFIC_MSG_EVT); + l_ptr->stats.recv_states++; + if (link_reset_unknown(l_ptr)) + break; + + if (less_eq(mod(l_ptr->next_in_no), msg_next_sent(msg))) { + rec_gap = mod(msg_next_sent(msg) - + mod(l_ptr->next_in_no)); + } + + max_pkt_ack = msg_max_pkt(msg); + if (max_pkt_ack > l_ptr->max_pkt) { + l_ptr->max_pkt = max_pkt_ack; + l_ptr->max_pkt_probes = 0; + } + + max_pkt_ack = 0; + if (msg_probe(msg)) { + l_ptr->stats.recv_probes++; + if (msg_size(msg) > sizeof(l_ptr->proto_msg)) + max_pkt_ack = msg_size(msg); + } + + /* Protocol message before retransmits, reduce loss risk */ + if (l_ptr->owner->bclink.recv_permitted) + tipc_bclink_update_link_state(l_ptr->owner, + msg_last_bcast(msg)); + + if (rec_gap || (msg_probe(msg))) { + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, rec_gap, 0, + 0, max_pkt_ack); + } + if (msg_seq_gap(msg)) { + l_ptr->stats.recv_nacks++; + tipc_link_retransmit(l_ptr, l_ptr->first_out, + msg_seq_gap(msg)); + } + break; + } +exit: + kfree_skb(buf); +} + + +/* tipc_link_tunnel_xmit(): Tunnel one packet via a link belonging to + * a different bearer. Owner node is locked. + */ +static void tipc_link_tunnel_xmit(struct tipc_link *l_ptr, + struct tipc_msg *tunnel_hdr, + struct tipc_msg *msg, + u32 selector) +{ + struct tipc_link *tunnel; + struct sk_buff *buf; + u32 length = msg_size(msg); + + tunnel = l_ptr->owner->active_links[selector & 1]; + if (!tipc_link_is_up(tunnel)) { + pr_warn("%stunnel link no longer available\n", link_co_err); + return; + } + msg_set_size(tunnel_hdr, length + INT_H_SIZE); + buf = tipc_buf_acquire(length + INT_H_SIZE); + if (!buf) { + pr_warn("%sunable to send tunnel msg\n", link_co_err); + return; + } + skb_copy_to_linear_data(buf, tunnel_hdr, INT_H_SIZE); + skb_copy_to_linear_data_offset(buf, INT_H_SIZE, msg, length); + __tipc_link_xmit(tunnel, buf); +} + + +/* tipc_link_failover_send_queue(): A link has gone down, but a second + * link is still active. We can do failover. Tunnel the failing link's + * whole send queue via the remaining link. This way, we don't lose + * any packets, and sequence order is preserved for subsequent traffic + * sent over the remaining link. Owner node is locked. + */ +void tipc_link_failover_send_queue(struct tipc_link *l_ptr) +{ + u32 msgcount = l_ptr->out_queue_size; + struct sk_buff *crs = l_ptr->first_out; + struct tipc_link *tunnel = l_ptr->owner->active_links[0]; + struct tipc_msg tunnel_hdr; + int split_bundles; + + if (!tunnel) + return; + + tipc_msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL, + ORIGINAL_MSG, INT_H_SIZE, l_ptr->addr); + msg_set_bearer_id(&tunnel_hdr, l_ptr->peer_bearer_id); + msg_set_msgcnt(&tunnel_hdr, msgcount); + + if (!l_ptr->first_out) { + struct sk_buff *buf; + + buf = tipc_buf_acquire(INT_H_SIZE); + if (buf) { + skb_copy_to_linear_data(buf, &tunnel_hdr, INT_H_SIZE); + msg_set_size(&tunnel_hdr, INT_H_SIZE); + __tipc_link_xmit(tunnel, buf); + } else { + pr_warn("%sunable to send changeover msg\n", + link_co_err); + } + return; + } + + split_bundles = (l_ptr->owner->active_links[0] != + l_ptr->owner->active_links[1]); + + while (crs) { + struct tipc_msg *msg = buf_msg(crs); + + if ((msg_user(msg) == MSG_BUNDLER) && split_bundles) { + struct tipc_msg *m = msg_get_wrapped(msg); + unchar *pos = (unchar *)m; + + msgcount = msg_msgcnt(msg); + while (msgcount--) { + msg_set_seqno(m, msg_seqno(msg)); + tipc_link_tunnel_xmit(l_ptr, &tunnel_hdr, m, + msg_link_selector(m)); + pos += align(msg_size(m)); + m = (struct tipc_msg *)pos; + } + } else { + tipc_link_tunnel_xmit(l_ptr, &tunnel_hdr, msg, + msg_link_selector(msg)); + } + crs = crs->next; + } +} + +/* tipc_link_dup_queue_xmit(): A second link has become active. Tunnel a + * duplicate of the first link's send queue via the new link. This way, we + * are guaranteed that currently queued packets from a socket are delivered + * before future traffic from the same socket, even if this is using the + * new link. The last arriving copy of each duplicate packet is dropped at + * the receiving end by the regular protocol check, so packet cardinality + * and sequence order is preserved per sender/receiver socket pair. + * Owner node is locked. + */ +void tipc_link_dup_queue_xmit(struct tipc_link *l_ptr, + struct tipc_link *tunnel) +{ + struct sk_buff *iter; + struct tipc_msg tunnel_hdr; + + tipc_msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL, + DUPLICATE_MSG, INT_H_SIZE, l_ptr->addr); + msg_set_msgcnt(&tunnel_hdr, l_ptr->out_queue_size); + msg_set_bearer_id(&tunnel_hdr, l_ptr->peer_bearer_id); + iter = l_ptr->first_out; + while (iter) { + struct sk_buff *outbuf; + struct tipc_msg *msg = buf_msg(iter); + u32 length = msg_size(msg); + + if (msg_user(msg) == MSG_BUNDLER) + msg_set_type(msg, CLOSED_MSG); + msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); /* Update */ + msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); + msg_set_size(&tunnel_hdr, length + INT_H_SIZE); + outbuf = tipc_buf_acquire(length + INT_H_SIZE); + if (outbuf == NULL) { + pr_warn("%sunable to send duplicate msg\n", + link_co_err); + return; + } + skb_copy_to_linear_data(outbuf, &tunnel_hdr, INT_H_SIZE); + skb_copy_to_linear_data_offset(outbuf, INT_H_SIZE, iter->data, + length); + __tipc_link_xmit(tunnel, outbuf); + if (!tipc_link_is_up(l_ptr)) + return; + iter = iter->next; + } +} + +/** + * buf_extract - extracts embedded TIPC message from another message + * @skb: encapsulating message buffer + * @from_pos: offset to extract from + * + * Returns a new message buffer containing an embedded message. The + * encapsulating message itself is left unchanged. + */ +static struct sk_buff *buf_extract(struct sk_buff *skb, u32 from_pos) +{ + struct tipc_msg *msg = (struct tipc_msg *)(skb->data + from_pos); + u32 size = msg_size(msg); + struct sk_buff *eb; + + eb = tipc_buf_acquire(size); + if (eb) + skb_copy_to_linear_data(eb, msg, size); + return eb; +} + + + +/* tipc_link_dup_rcv(): Receive a tunnelled DUPLICATE_MSG packet. + * Owner node is locked. + */ +static void tipc_link_dup_rcv(struct tipc_link *l_ptr, + struct sk_buff *t_buf) +{ + struct sk_buff *buf; + + if (!tipc_link_is_up(l_ptr)) + return; + + buf = buf_extract(t_buf, INT_H_SIZE); + if (buf == NULL) { + pr_warn("%sfailed to extract inner dup pkt\n", link_co_err); + return; + } + + /* Add buffer to deferred queue, if applicable: */ + link_handle_out_of_seq_msg(l_ptr, buf); +} + +/* tipc_link_failover_rcv(): Receive a tunnelled ORIGINAL_MSG packet + * Owner node is locked. + */ +static struct sk_buff *tipc_link_failover_rcv(struct tipc_link *l_ptr, + struct sk_buff *t_buf) +{ + struct tipc_msg *t_msg = buf_msg(t_buf); + struct sk_buff *buf = NULL; + struct tipc_msg *msg; + + if (tipc_link_is_up(l_ptr)) + tipc_link_reset(l_ptr); + + /* First failover packet? */ + if (l_ptr->exp_msg_count == START_CHANGEOVER) + l_ptr->exp_msg_count = msg_msgcnt(t_msg); + + /* Should there be an inner packet? */ + if (l_ptr->exp_msg_count) { + l_ptr->exp_msg_count--; + buf = buf_extract(t_buf, INT_H_SIZE); + if (buf == NULL) { + pr_warn("%sno inner failover pkt\n", link_co_err); + goto exit; + } + msg = buf_msg(buf); + + if (less(msg_seqno(msg), l_ptr->reset_checkpoint)) { + kfree_skb(buf); + buf = NULL; + goto exit; + } + if (msg_user(msg) == MSG_FRAGMENTER) { + l_ptr->stats.recv_fragments++; + tipc_buf_append(&l_ptr->reasm_buf, &buf); + } + } +exit: + if ((l_ptr->exp_msg_count == 0) && (l_ptr->flags & LINK_STOPPED)) { + tipc_node_detach_link(l_ptr->owner, l_ptr); + kfree(l_ptr); + } + return buf; +} + +/* tipc_link_tunnel_rcv(): Receive a tunnelled packet, sent + * via other link as result of a failover (ORIGINAL_MSG) or + * a new active link (DUPLICATE_MSG). Failover packets are + * returned to the active link for delivery upwards. + * Owner node is locked. + */ +static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr, + struct sk_buff **buf) +{ + struct sk_buff *t_buf = *buf; + struct tipc_link *l_ptr; + struct tipc_msg *t_msg = buf_msg(t_buf); + u32 bearer_id = msg_bearer_id(t_msg); + + *buf = NULL; + + if (bearer_id >= MAX_BEARERS) + goto exit; + + l_ptr = n_ptr->links[bearer_id]; + if (!l_ptr) + goto exit; + + if (msg_type(t_msg) == DUPLICATE_MSG) + tipc_link_dup_rcv(l_ptr, t_buf); + else if (msg_type(t_msg) == ORIGINAL_MSG) + *buf = tipc_link_failover_rcv(l_ptr, t_buf); + else + pr_warn("%sunknown tunnel pkt received\n", link_co_err); +exit: + kfree_skb(t_buf); + return *buf != NULL; +} + +/* + * Bundler functionality: + */ +void tipc_link_bundle_rcv(struct sk_buff *buf) +{ + u32 msgcount = msg_msgcnt(buf_msg(buf)); + u32 pos = INT_H_SIZE; + struct sk_buff *obuf; + + while (msgcount--) { + obuf = buf_extract(buf, pos); + if (obuf == NULL) { + pr_warn("Link unable to unbundle message(s)\n"); + break; + } + pos += align(msg_size(buf_msg(obuf))); + tipc_net_route_msg(obuf); + } + kfree_skb(buf); +} + +/* + * Fragmentation/defragmentation: + */ + +/* + * tipc_link_frag_xmit: Entry for buffers needing fragmentation. + * The buffer is complete, inclusive total message length. + * Returns user data length. + */ +static int tipc_link_frag_xmit(struct tipc_link *l_ptr, struct sk_buff *buf) +{ + struct sk_buff *buf_chain = NULL; + struct sk_buff *buf_chain_tail = (struct sk_buff *)&buf_chain; + struct tipc_msg *inmsg = buf_msg(buf); + struct tipc_msg fragm_hdr; + u32 insize = msg_size(inmsg); + u32 dsz = msg_data_sz(inmsg); + unchar *crs = buf->data; + u32 rest = insize; + u32 pack_sz = l_ptr->max_pkt; + u32 fragm_sz = pack_sz - INT_H_SIZE; + u32 fragm_no = 0; + u32 destaddr; + + if (msg_short(inmsg)) + destaddr = l_ptr->addr; + else + destaddr = msg_destnode(inmsg); + + /* Prepare reusable fragment header: */ + tipc_msg_init(&fragm_hdr, MSG_FRAGMENTER, FIRST_FRAGMENT, + INT_H_SIZE, destaddr); + + /* Chop up message: */ + while (rest > 0) { + struct sk_buff *fragm; + + if (rest <= fragm_sz) { + fragm_sz = rest; + msg_set_type(&fragm_hdr, LAST_FRAGMENT); + } + fragm = tipc_buf_acquire(fragm_sz + INT_H_SIZE); + if (fragm == NULL) { + kfree_skb(buf); + kfree_skb_list(buf_chain); + return -ENOMEM; + } + msg_set_size(&fragm_hdr, fragm_sz + INT_H_SIZE); + fragm_no++; + msg_set_fragm_no(&fragm_hdr, fragm_no); + skb_copy_to_linear_data(fragm, &fragm_hdr, INT_H_SIZE); + skb_copy_to_linear_data_offset(fragm, INT_H_SIZE, crs, + fragm_sz); + buf_chain_tail->next = fragm; + buf_chain_tail = fragm; + + rest -= fragm_sz; + crs += fragm_sz; + msg_set_type(&fragm_hdr, FRAGMENT); + } + kfree_skb(buf); + + /* Append chain of fragments to send queue & send them */ + l_ptr->long_msg_seq_no++; + link_add_chain_to_outqueue(l_ptr, buf_chain, l_ptr->long_msg_seq_no); + l_ptr->stats.sent_fragments += fragm_no; + l_ptr->stats.sent_fragmented++; + tipc_link_push_queue(l_ptr); + + return dsz; +} + +static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tolerance) +{ + if ((tolerance < TIPC_MIN_LINK_TOL) || (tolerance > TIPC_MAX_LINK_TOL)) + return; + + l_ptr->tolerance = tolerance; + l_ptr->continuity_interval = + ((tolerance / 4) > 500) ? 500 : tolerance / 4; + l_ptr->abort_limit = tolerance / (l_ptr->continuity_interval / 4); +} + +void tipc_link_set_queue_limits(struct tipc_link *l_ptr, u32 window) +{ + /* Data messages from this node, inclusive FIRST_FRAGM */ + l_ptr->queue_limit[TIPC_LOW_IMPORTANCE] = window; + l_ptr->queue_limit[TIPC_MEDIUM_IMPORTANCE] = (window / 3) * 4; + l_ptr->queue_limit[TIPC_HIGH_IMPORTANCE] = (window / 3) * 5; + l_ptr->queue_limit[TIPC_CRITICAL_IMPORTANCE] = (window / 3) * 6; + /* Transiting data messages,inclusive FIRST_FRAGM */ + l_ptr->queue_limit[TIPC_LOW_IMPORTANCE + 4] = 300; + l_ptr->queue_limit[TIPC_MEDIUM_IMPORTANCE + 4] = 600; + l_ptr->queue_limit[TIPC_HIGH_IMPORTANCE + 4] = 900; + l_ptr->queue_limit[TIPC_CRITICAL_IMPORTANCE + 4] = 1200; + l_ptr->queue_limit[CONN_MANAGER] = 1200; + l_ptr->queue_limit[CHANGEOVER_PROTOCOL] = 2500; + l_ptr->queue_limit[NAME_DISTRIBUTOR] = 3000; + /* FRAGMENT and LAST_FRAGMENT packets */ + l_ptr->queue_limit[MSG_FRAGMENTER] = 4000; +} + +/* tipc_link_find_owner - locate owner node of link by link's name + * @name: pointer to link name string + * @bearer_id: pointer to index in 'node->links' array where the link was found. + * + * Returns pointer to node owning the link, or 0 if no matching link is found. + */ +static struct tipc_node *tipc_link_find_owner(const char *link_name, + unsigned int *bearer_id) +{ + struct tipc_link *l_ptr; + struct tipc_node *n_ptr; + struct tipc_node *found_node = 0; + int i; + + *bearer_id = 0; + rcu_read_lock(); + list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { + tipc_node_lock(n_ptr); + for (i = 0; i < MAX_BEARERS; i++) { + l_ptr = n_ptr->links[i]; + if (l_ptr && !strcmp(l_ptr->name, link_name)) { + *bearer_id = i; + found_node = n_ptr; + break; + } + } + tipc_node_unlock(n_ptr); + if (found_node) + break; + } + rcu_read_unlock(); + + return found_node; +} + +/** + * link_value_is_valid -- validate proposed link tolerance/priority/window + * + * @cmd: value type (TIPC_CMD_SET_LINK_*) + * @new_value: the new value + * + * Returns 1 if value is within range, 0 if not. + */ +static int link_value_is_valid(u16 cmd, u32 new_value) +{ + switch (cmd) { + case TIPC_CMD_SET_LINK_TOL: + return (new_value >= TIPC_MIN_LINK_TOL) && + (new_value <= TIPC_MAX_LINK_TOL); + case TIPC_CMD_SET_LINK_PRI: + return (new_value <= TIPC_MAX_LINK_PRI); + case TIPC_CMD_SET_LINK_WINDOW: + return (new_value >= TIPC_MIN_LINK_WIN) && + (new_value <= TIPC_MAX_LINK_WIN); + } + return 0; +} + +/** + * link_cmd_set_value - change priority/tolerance/window for link/bearer/media + * @name: ptr to link, bearer, or media name + * @new_value: new value of link, bearer, or media setting + * @cmd: which link, bearer, or media attribute to set (TIPC_CMD_SET_LINK_*) + * + * Caller must hold RTNL lock to ensure link/bearer/media is not deleted. + * + * Returns 0 if value updated and negative value on error. + */ +static int link_cmd_set_value(const char *name, u32 new_value, u16 cmd) +{ + struct tipc_node *node; + struct tipc_link *l_ptr; + struct tipc_bearer *b_ptr; + struct tipc_media *m_ptr; + int bearer_id; + int res = 0; + + node = tipc_link_find_owner(name, &bearer_id); + if (node) { + tipc_node_lock(node); + l_ptr = node->links[bearer_id]; + + if (l_ptr) { + switch (cmd) { + case TIPC_CMD_SET_LINK_TOL: + link_set_supervision_props(l_ptr, new_value); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, + new_value, 0, 0); + break; + case TIPC_CMD_SET_LINK_PRI: + l_ptr->priority = new_value; + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, + 0, new_value, 0); + break; + case TIPC_CMD_SET_LINK_WINDOW: + tipc_link_set_queue_limits(l_ptr, new_value); + break; + default: + res = -EINVAL; + break; + } + } + tipc_node_unlock(node); + return res; + } + + b_ptr = tipc_bearer_find(name); + if (b_ptr) { + switch (cmd) { + case TIPC_CMD_SET_LINK_TOL: + b_ptr->tolerance = new_value; + break; + case TIPC_CMD_SET_LINK_PRI: + b_ptr->priority = new_value; + break; + case TIPC_CMD_SET_LINK_WINDOW: + b_ptr->window = new_value; + break; + default: + res = -EINVAL; + break; + } + return res; + } + + m_ptr = tipc_media_find(name); + if (!m_ptr) + return -ENODEV; + switch (cmd) { + case TIPC_CMD_SET_LINK_TOL: + m_ptr->tolerance = new_value; + break; + case TIPC_CMD_SET_LINK_PRI: + m_ptr->priority = new_value; + break; + case TIPC_CMD_SET_LINK_WINDOW: + m_ptr->window = new_value; + break; + default: + res = -EINVAL; + break; + } + return res; +} + +struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space, + u16 cmd) +{ + struct tipc_link_config *args; + u32 new_value; + int res; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_CONFIG)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + args = (struct tipc_link_config *)TLV_DATA(req_tlv_area); + new_value = ntohl(args->value); + + if (!link_value_is_valid(cmd, new_value)) + return tipc_cfg_reply_error_string( + "cannot change, value invalid"); + + if (!strcmp(args->name, tipc_bclink_name)) { + if ((cmd == TIPC_CMD_SET_LINK_WINDOW) && + (tipc_bclink_set_queue_limits(new_value) == 0)) + return tipc_cfg_reply_none(); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change setting on broadcast link)"); + } + + res = link_cmd_set_value(args->name, new_value, cmd); + if (res) + return tipc_cfg_reply_error_string("cannot change link setting"); + + return tipc_cfg_reply_none(); +} + +/** + * link_reset_statistics - reset link statistics + * @l_ptr: pointer to link + */ +static void link_reset_statistics(struct tipc_link *l_ptr) +{ + memset(&l_ptr->stats, 0, sizeof(l_ptr->stats)); + l_ptr->stats.sent_info = l_ptr->next_out_no; + l_ptr->stats.recv_info = l_ptr->next_in_no; +} + +struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space) +{ + char *link_name; + struct tipc_link *l_ptr; + struct tipc_node *node; + unsigned int bearer_id; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + link_name = (char *)TLV_DATA(req_tlv_area); + if (!strcmp(link_name, tipc_bclink_name)) { + if (tipc_bclink_reset_stats()) + return tipc_cfg_reply_error_string("link not found"); + return tipc_cfg_reply_none(); + } + node = tipc_link_find_owner(link_name, &bearer_id); + if (!node) + return tipc_cfg_reply_error_string("link not found"); + + tipc_node_lock(node); + l_ptr = node->links[bearer_id]; + if (!l_ptr) { + tipc_node_unlock(node); + return tipc_cfg_reply_error_string("link not found"); + } + link_reset_statistics(l_ptr); + tipc_node_unlock(node); + return tipc_cfg_reply_none(); +} + +/** + * percent - convert count to a percentage of total (rounding up or down) + */ +static u32 percent(u32 count, u32 total) +{ + return (count * 100 + (total / 2)) / total; +} + +/** + * tipc_link_stats - print link statistics + * @name: link name + * @buf: print buffer area + * @buf_size: size of print buffer area + * + * Returns length of print buffer data string (or 0 if error) + */ +static int tipc_link_stats(const char *name, char *buf, const u32 buf_size) +{ + struct tipc_link *l; + struct tipc_stats *s; + struct tipc_node *node; + char *status; + u32 profile_total = 0; + unsigned int bearer_id; + int ret; + + if (!strcmp(name, tipc_bclink_name)) + return tipc_bclink_stats(buf, buf_size); + + node = tipc_link_find_owner(name, &bearer_id); + if (!node) + return 0; + + tipc_node_lock(node); + + l = node->links[bearer_id]; + if (!l) { + tipc_node_unlock(node); + return 0; + } + + s = &l->stats; + + if (tipc_link_is_active(l)) + status = "ACTIVE"; + else if (tipc_link_is_up(l)) + status = "STANDBY"; + else + status = "DEFUNCT"; + + ret = tipc_snprintf(buf, buf_size, "Link <%s>\n" + " %s MTU:%u Priority:%u Tolerance:%u ms" + " Window:%u packets\n", + l->name, status, l->max_pkt, l->priority, + l->tolerance, l->queue_limit[0]); + + ret += tipc_snprintf(buf + ret, buf_size - ret, + " RX packets:%u fragments:%u/%u bundles:%u/%u\n", + l->next_in_no - s->recv_info, s->recv_fragments, + s->recv_fragmented, s->recv_bundles, + s->recv_bundled); + + ret += tipc_snprintf(buf + ret, buf_size - ret, + " TX packets:%u fragments:%u/%u bundles:%u/%u\n", + l->next_out_no - s->sent_info, s->sent_fragments, + s->sent_fragmented, s->sent_bundles, + s->sent_bundled); + + profile_total = s->msg_length_counts; + if (!profile_total) + profile_total = 1; + + ret += tipc_snprintf(buf + ret, buf_size - ret, + " TX profile sample:%u packets average:%u octets\n" + " 0-64:%u%% -256:%u%% -1024:%u%% -4096:%u%% " + "-16384:%u%% -32768:%u%% -66000:%u%%\n", + s->msg_length_counts, + s->msg_lengths_total / profile_total, + percent(s->msg_length_profile[0], profile_total), + percent(s->msg_length_profile[1], profile_total), + percent(s->msg_length_profile[2], profile_total), + percent(s->msg_length_profile[3], profile_total), + percent(s->msg_length_profile[4], profile_total), + percent(s->msg_length_profile[5], profile_total), + percent(s->msg_length_profile[6], profile_total)); + + ret += tipc_snprintf(buf + ret, buf_size - ret, + " RX states:%u probes:%u naks:%u defs:%u" + " dups:%u\n", s->recv_states, s->recv_probes, + s->recv_nacks, s->deferred_recv, s->duplicates); + + ret += tipc_snprintf(buf + ret, buf_size - ret, + " TX states:%u probes:%u naks:%u acks:%u" + " dups:%u\n", s->sent_states, s->sent_probes, + s->sent_nacks, s->sent_acks, s->retransmitted); + + ret += tipc_snprintf(buf + ret, buf_size - ret, + " Congestion link:%u Send queue" + " max:%u avg:%u\n", s->link_congs, + s->max_queue_sz, s->queue_sz_counts ? + (s->accu_queue_sz / s->queue_sz_counts) : 0); + + tipc_node_unlock(node); + return ret; +} + +struct sk_buff *tipc_link_cmd_show_stats(const void *req_tlv_area, int req_tlv_space) +{ + struct sk_buff *buf; + struct tlv_desc *rep_tlv; + int str_len; + int pb_len; + char *pb; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); + if (!buf) + return NULL; + + rep_tlv = (struct tlv_desc *)buf->data; + pb = TLV_DATA(rep_tlv); + pb_len = ULTRA_STRING_MAX_LEN; + str_len = tipc_link_stats((char *)TLV_DATA(req_tlv_area), + pb, pb_len); + if (!str_len) { + kfree_skb(buf); + return tipc_cfg_reply_error_string("link not found"); + } + str_len += 1; /* for "\0" */ + skb_put(buf, TLV_SPACE(str_len)); + TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); + + return buf; +} + +/** + * tipc_link_get_max_pkt - get maximum packet size to use when sending to destination + * @dest: network address of destination node + * @selector: used to select from set of active links + * + * If no active link can be found, uses default maximum packet size. + */ +u32 tipc_link_get_max_pkt(u32 dest, u32 selector) +{ + struct tipc_node *n_ptr; + struct tipc_link *l_ptr; + u32 res = MAX_PKT_DEFAULT; + + if (dest == tipc_own_addr) + return MAX_MSG_SIZE; + + n_ptr = tipc_node_find(dest); + if (n_ptr) { + tipc_node_lock(n_ptr); + l_ptr = n_ptr->active_links[selector & 1]; + if (l_ptr) + res = l_ptr->max_pkt; + tipc_node_unlock(n_ptr); + } + return res; +} + +static void link_print(struct tipc_link *l_ptr, const char *str) +{ + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[l_ptr->bearer_id]); + if (b_ptr) + pr_info("%s Link %x<%s>:", str, l_ptr->addr, b_ptr->name); + rcu_read_unlock(); + + if (link_working_unknown(l_ptr)) + pr_cont(":WU\n"); + else if (link_reset_reset(l_ptr)) + pr_cont(":RR\n"); + else if (link_reset_unknown(l_ptr)) + pr_cont(":RU\n"); + else if (link_working_working(l_ptr)) + pr_cont(":WW\n"); + else + pr_cont("\n"); +} diff --git a/net/tipc/link.h b/net/tipc/link.h new file mode 100644 index 00000000000..200d518b218 --- /dev/null +++ b/net/tipc/link.h @@ -0,0 +1,314 @@ +/* + * net/tipc/link.h: Include file for TIPC link code + * + * Copyright (c) 1995-2006, 2013, Ericsson AB + * Copyright (c) 2004-2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_LINK_H +#define _TIPC_LINK_H + +#include "msg.h" +#include "node.h" + +/* Out-of-range value for link sequence numbers + */ +#define INVALID_LINK_SEQ 0x10000 + +/* Link working states + */ +#define WORKING_WORKING 560810u +#define WORKING_UNKNOWN 560811u +#define RESET_UNKNOWN 560812u +#define RESET_RESET 560813u + +/* Link endpoint execution states + */ +#define LINK_STARTED 0x0001 +#define LINK_STOPPED 0x0002 + +/* Starting value for maximum packet size negotiation on unicast links + * (unless bearer MTU is less) + */ +#define MAX_PKT_DEFAULT 1500 + +struct tipc_stats { + u32 sent_info; /* used in counting # sent packets */ + u32 recv_info; /* used in counting # recv'd packets */ + u32 sent_states; + u32 recv_states; + u32 sent_probes; + u32 recv_probes; + u32 sent_nacks; + u32 recv_nacks; + u32 sent_acks; + u32 sent_bundled; + u32 sent_bundles; + u32 recv_bundled; + u32 recv_bundles; + u32 retransmitted; + u32 sent_fragmented; + u32 sent_fragments; + u32 recv_fragmented; + u32 recv_fragments; + u32 link_congs; /* # port sends blocked by congestion */ + u32 deferred_recv; + u32 duplicates; + u32 max_queue_sz; /* send queue size high water mark */ + u32 accu_queue_sz; /* used for send queue size profiling */ + u32 queue_sz_counts; /* used for send queue size profiling */ + u32 msg_length_counts; /* used for message length profiling */ + u32 msg_lengths_total; /* used for message length profiling */ + u32 msg_length_profile[7]; /* used for msg. length profiling */ +}; + +/** + * struct tipc_link - TIPC link data structure + * @addr: network address of link's peer node + * @name: link name character string + * @media_addr: media address to use when sending messages over link + * @timer: link timer + * @owner: pointer to peer node + * @flags: execution state flags for link endpoint instance + * @checkpoint: reference point for triggering link continuity checking + * @peer_session: link session # being used by peer end of link + * @peer_bearer_id: bearer id used by link's peer endpoint + * @bearer_id: local bearer id used by link + * @tolerance: minimum link continuity loss needed to reset link [in ms] + * @continuity_interval: link continuity testing interval [in ms] + * @abort_limit: # of unacknowledged continuity probes needed to reset link + * @state: current state of link FSM + * @fsm_msg_cnt: # of protocol messages link FSM has sent in current state + * @proto_msg: template for control messages generated by link + * @pmsg: convenience pointer to "proto_msg" field + * @priority: current link priority + * @net_plane: current link network plane ('A' through 'H') + * @queue_limit: outbound message queue congestion thresholds (indexed by user) + * @exp_msg_count: # of tunnelled messages expected during link changeover + * @reset_checkpoint: seq # of last acknowledged message at time of link reset + * @max_pkt: current maximum packet size for this link + * @max_pkt_target: desired maximum packet size for this link + * @max_pkt_probes: # of probes based on current (max_pkt, max_pkt_target) + * @out_queue_size: # of messages in outbound message queue + * @first_out: ptr to first outbound message in queue + * @last_out: ptr to last outbound message in queue + * @next_out_no: next sequence number to use for outbound messages + * @last_retransmitted: sequence number of most recently retransmitted message + * @stale_count: # of identical retransmit requests made by peer + * @next_in_no: next sequence number to expect for inbound messages + * @deferred_inqueue_sz: # of messages in inbound message queue + * @oldest_deferred_in: ptr to first inbound message in queue + * @newest_deferred_in: ptr to last inbound message in queue + * @unacked_window: # of inbound messages rx'd without ack'ing back to peer + * @proto_msg_queue: ptr to (single) outbound control message + * @retransm_queue_size: number of messages to retransmit + * @retransm_queue_head: sequence number of first message to retransmit + * @next_out: ptr to first unsent outbound message in queue + * @waiting_ports: linked list of ports waiting for link congestion to abate + * @long_msg_seq_no: next identifier to use for outbound fragmented messages + * @reasm_buf: head of partially reassembled inbound message fragments + * @stats: collects statistics regarding link activity + */ +struct tipc_link { + u32 addr; + char name[TIPC_MAX_LINK_NAME]; + struct tipc_media_addr media_addr; + struct timer_list timer; + struct tipc_node *owner; + + /* Management and link supervision data */ + unsigned int flags; + u32 checkpoint; + u32 peer_session; + u32 peer_bearer_id; + u32 bearer_id; + u32 tolerance; + u32 continuity_interval; + u32 abort_limit; + int state; + u32 fsm_msg_cnt; + struct { + unchar hdr[INT_H_SIZE]; + unchar body[TIPC_MAX_IF_NAME]; + } proto_msg; + struct tipc_msg *pmsg; + u32 priority; + char net_plane; + u32 queue_limit[15]; /* queue_limit[0]==window limit */ + + /* Changeover */ + u32 exp_msg_count; + u32 reset_checkpoint; + + /* Max packet negotiation */ + u32 max_pkt; + u32 max_pkt_target; + u32 max_pkt_probes; + + /* Sending */ + u32 out_queue_size; + struct sk_buff *first_out; + struct sk_buff *last_out; + u32 next_out_no; + u32 last_retransmitted; + u32 stale_count; + + /* Reception */ + u32 next_in_no; + u32 deferred_inqueue_sz; + struct sk_buff *oldest_deferred_in; + struct sk_buff *newest_deferred_in; + u32 unacked_window; + + /* Congestion handling */ + struct sk_buff *proto_msg_queue; + u32 retransm_queue_size; + u32 retransm_queue_head; + struct sk_buff *next_out; + struct list_head waiting_ports; + + /* Fragmentation/reassembly */ + u32 long_msg_seq_no; + struct sk_buff *reasm_buf; + + /* Statistics */ + struct tipc_stats stats; +}; + +struct tipc_port; + +struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, + struct tipc_bearer *b_ptr, + const struct tipc_media_addr *media_addr); +void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down); +void tipc_link_failover_send_queue(struct tipc_link *l_ptr); +void tipc_link_dup_queue_xmit(struct tipc_link *l_ptr, struct tipc_link *dest); +void tipc_link_reset_fragments(struct tipc_link *l_ptr); +int tipc_link_is_up(struct tipc_link *l_ptr); +int tipc_link_is_active(struct tipc_link *l_ptr); +void tipc_link_purge_queues(struct tipc_link *l_ptr); +struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, + int req_tlv_space, + u16 cmd); +struct sk_buff *tipc_link_cmd_show_stats(const void *req_tlv_area, + int req_tlv_space); +struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, + int req_tlv_space); +void tipc_link_reset_all(struct tipc_node *node); +void tipc_link_reset(struct tipc_link *l_ptr); +void tipc_link_reset_list(unsigned int bearer_id); +int tipc_link_xmit(struct sk_buff *buf, u32 dest, u32 selector); +void tipc_link_names_xmit(struct list_head *message_list, u32 dest); +int __tipc_link_xmit(struct tipc_link *l_ptr, struct sk_buff *buf); +int tipc_link_send_buf(struct tipc_link *l_ptr, struct sk_buff *buf); +u32 tipc_link_get_max_pkt(u32 dest, u32 selector); +int tipc_link_iovec_xmit_fast(struct tipc_port *sender, + struct iovec const *msg_sect, + unsigned int len, u32 destnode); +void tipc_link_bundle_rcv(struct sk_buff *buf); +void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int prob, + u32 gap, u32 tolerance, u32 priority, u32 acked_mtu); +void tipc_link_push_queue(struct tipc_link *l_ptr); +u32 tipc_link_defer_pkt(struct sk_buff **head, struct sk_buff **tail, + struct sk_buff *buf); +void tipc_link_wakeup_ports(struct tipc_link *l_ptr, int all); +void tipc_link_set_queue_limits(struct tipc_link *l_ptr, u32 window); +void tipc_link_retransmit(struct tipc_link *l_ptr, + struct sk_buff *start, u32 retransmits); + +/* + * Link sequence number manipulation routines (uses modulo 2**16 arithmetic) + */ +static inline u32 buf_seqno(struct sk_buff *buf) +{ + return msg_seqno(buf_msg(buf)); +} + +static inline u32 mod(u32 x) +{ + return x & 0xffffu; +} + +static inline int between(u32 lower, u32 upper, u32 n) +{ + if ((lower < n) && (n < upper)) + return 1; + if ((upper < lower) && ((n > lower) || (n < upper))) + return 1; + return 0; +} + +static inline int less_eq(u32 left, u32 right) +{ + return mod(right - left) < 32768u; +} + +static inline int less(u32 left, u32 right) +{ + return less_eq(left, right) && (mod(right) != mod(left)); +} + +static inline u32 lesser(u32 left, u32 right) +{ + return less_eq(left, right) ? left : right; +} + + +/* + * Link status checking routines + */ +static inline int link_working_working(struct tipc_link *l_ptr) +{ + return l_ptr->state == WORKING_WORKING; +} + +static inline int link_working_unknown(struct tipc_link *l_ptr) +{ + return l_ptr->state == WORKING_UNKNOWN; +} + +static inline int link_reset_unknown(struct tipc_link *l_ptr) +{ + return l_ptr->state == RESET_UNKNOWN; +} + +static inline int link_reset_reset(struct tipc_link *l_ptr) +{ + return l_ptr->state == RESET_RESET; +} + +static inline int link_congested(struct tipc_link *l_ptr) +{ + return l_ptr->out_queue_size >= l_ptr->queue_limit[0]; +} + +#endif diff --git a/net/tipc/log.c b/net/tipc/log.c new file mode 100644 index 00000000000..abef644f27d --- /dev/null +++ b/net/tipc/log.c @@ -0,0 +1,55 @@ +/* + * net/tipc/log.c: TIPC print buffer routines for debugging + * + * Copyright (c) 1996-2006, Ericsson AB + * Copyright (c) 2005-2007, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" + +/** + * tipc_snprintf - append formatted output to print buffer + * @buf: pointer to print buffer + * @len: buffer length + * @fmt: formatted info to be printed + */ +int tipc_snprintf(char *buf, int len, const char *fmt, ...) +{ + int i; + va_list args; + + va_start(args, fmt); + i = vscnprintf(buf, len, fmt, args); + va_end(args); + return i; +} diff --git a/net/tipc/msg.c b/net/tipc/msg.c new file mode 100644 index 00000000000..0a37a472c29 --- /dev/null +++ b/net/tipc/msg.c @@ -0,0 +1,159 @@ +/* + * net/tipc/msg.c: TIPC message header routines + * + * Copyright (c) 2000-2006, 2014, Ericsson AB + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "msg.h" + +u32 tipc_msg_tot_importance(struct tipc_msg *m) +{ + if (likely(msg_isdata(m))) { + if (likely(msg_orignode(m) == tipc_own_addr)) + return msg_importance(m); + return msg_importance(m) + 4; + } + if ((msg_user(m) == MSG_FRAGMENTER) && + (msg_type(m) == FIRST_FRAGMENT)) + return msg_importance(msg_get_wrapped(m)); + return msg_importance(m); +} + + +void tipc_msg_init(struct tipc_msg *m, u32 user, u32 type, u32 hsize, + u32 destnode) +{ + memset(m, 0, hsize); + msg_set_version(m); + msg_set_user(m, user); + msg_set_hdr_sz(m, hsize); + msg_set_size(m, hsize); + msg_set_prevnode(m, tipc_own_addr); + msg_set_type(m, type); + msg_set_orignode(m, tipc_own_addr); + msg_set_destnode(m, destnode); +} + +/** + * tipc_msg_build - create message using specified header and data + * + * Note: Caller must not hold any locks in case copy_from_user() is interrupted! + * + * Returns message data size or errno + */ +int tipc_msg_build(struct tipc_msg *hdr, struct iovec const *msg_sect, + unsigned int len, int max_size, struct sk_buff **buf) +{ + int dsz, sz, hsz; + unsigned char *to; + + dsz = len; + hsz = msg_hdr_sz(hdr); + sz = hsz + dsz; + msg_set_size(hdr, sz); + if (unlikely(sz > max_size)) { + *buf = NULL; + return dsz; + } + + *buf = tipc_buf_acquire(sz); + if (!(*buf)) + return -ENOMEM; + skb_copy_to_linear_data(*buf, hdr, hsz); + to = (*buf)->data + hsz; + if (len && memcpy_fromiovecend(to, msg_sect, 0, dsz)) { + kfree_skb(*buf); + *buf = NULL; + return -EFAULT; + } + return dsz; +} + +/* tipc_buf_append(): Append a buffer to the fragment list of another buffer + * @*headbuf: in: NULL for first frag, otherwise value returned from prev call + * out: set when successful non-complete reassembly, otherwise NULL + * @*buf: in: the buffer to append. Always defined + * out: head buf after sucessful complete reassembly, otherwise NULL + * Returns 1 when reassembly complete, otherwise 0 + */ +int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf) +{ + struct sk_buff *head = *headbuf; + struct sk_buff *frag = *buf; + struct sk_buff *tail; + struct tipc_msg *msg = buf_msg(frag); + u32 fragid = msg_type(msg); + bool headstolen; + int delta; + + skb_pull(frag, msg_hdr_sz(msg)); + + if (fragid == FIRST_FRAGMENT) { + if (head || skb_unclone(frag, GFP_ATOMIC)) + goto out_free; + head = *headbuf = frag; + skb_frag_list_init(head); + *buf = NULL; + return 0; + } + if (!head) + goto out_free; + tail = TIPC_SKB_CB(head)->tail; + if (skb_try_coalesce(head, frag, &headstolen, &delta)) { + kfree_skb_partial(frag, headstolen); + } else { + if (!skb_has_frag_list(head)) + skb_shinfo(head)->frag_list = frag; + else + tail->next = frag; + head->truesize += frag->truesize; + head->data_len += frag->len; + head->len += frag->len; + TIPC_SKB_CB(head)->tail = frag; + } + if (fragid == LAST_FRAGMENT) { + *buf = head; + TIPC_SKB_CB(head)->tail = NULL; + *headbuf = NULL; + return 1; + } + *buf = NULL; + return 0; +out_free: + pr_warn_ratelimited("Unable to build fragment list\n"); + kfree_skb(*buf); + kfree_skb(*headbuf); + *buf = *headbuf = NULL; + return 0; +} diff --git a/net/tipc/msg.h b/net/tipc/msg.h new file mode 100644 index 00000000000..503511903d1 --- /dev/null +++ b/net/tipc/msg.h @@ -0,0 +1,717 @@ +/* + * net/tipc/msg.h: Include file for TIPC message header routines + * + * Copyright (c) 2000-2007, 2014, Ericsson AB + * Copyright (c) 2005-2008, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_MSG_H +#define _TIPC_MSG_H + +#include "bearer.h" + +/* + * Constants and routines used to read and write TIPC payload message headers + * + * Note: Some items are also used with TIPC internal message headers + */ +#define TIPC_VERSION 2 + +/* + * Payload message users are defined in TIPC's public API: + * - TIPC_LOW_IMPORTANCE + * - TIPC_MEDIUM_IMPORTANCE + * - TIPC_HIGH_IMPORTANCE + * - TIPC_CRITICAL_IMPORTANCE + */ + +/* + * Payload message types + */ +#define TIPC_CONN_MSG 0 +#define TIPC_MCAST_MSG 1 +#define TIPC_NAMED_MSG 2 +#define TIPC_DIRECT_MSG 3 + +/* + * Message header sizes + */ +#define SHORT_H_SIZE 24 /* In-cluster basic payload message */ +#define BASIC_H_SIZE 32 /* Basic payload message */ +#define NAMED_H_SIZE 40 /* Named payload message */ +#define MCAST_H_SIZE 44 /* Multicast payload message */ +#define INT_H_SIZE 40 /* Internal messages */ +#define MIN_H_SIZE 24 /* Smallest legal TIPC header size */ +#define MAX_H_SIZE 60 /* Largest possible TIPC header size */ + +#define MAX_MSG_SIZE (MAX_H_SIZE + TIPC_MAX_USER_MSG_SIZE) + +#define TIPC_MEDIA_ADDR_OFFSET 5 + + +struct tipc_msg { + __be32 hdr[15]; +}; + + +static inline u32 msg_word(struct tipc_msg *m, u32 pos) +{ + return ntohl(m->hdr[pos]); +} + +static inline void msg_set_word(struct tipc_msg *m, u32 w, u32 val) +{ + m->hdr[w] = htonl(val); +} + +static inline u32 msg_bits(struct tipc_msg *m, u32 w, u32 pos, u32 mask) +{ + return (msg_word(m, w) >> pos) & mask; +} + +static inline void msg_set_bits(struct tipc_msg *m, u32 w, + u32 pos, u32 mask, u32 val) +{ + val = (val & mask) << pos; + mask = mask << pos; + m->hdr[w] &= ~htonl(mask); + m->hdr[w] |= htonl(val); +} + +static inline void msg_swap_words(struct tipc_msg *msg, u32 a, u32 b) +{ + u32 temp = msg->hdr[a]; + + msg->hdr[a] = msg->hdr[b]; + msg->hdr[b] = temp; +} + +/* + * Word 0 + */ +static inline u32 msg_version(struct tipc_msg *m) +{ + return msg_bits(m, 0, 29, 7); +} + +static inline void msg_set_version(struct tipc_msg *m) +{ + msg_set_bits(m, 0, 29, 7, TIPC_VERSION); +} + +static inline u32 msg_user(struct tipc_msg *m) +{ + return msg_bits(m, 0, 25, 0xf); +} + +static inline u32 msg_isdata(struct tipc_msg *m) +{ + return msg_user(m) <= TIPC_CRITICAL_IMPORTANCE; +} + +static inline void msg_set_user(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 0, 25, 0xf, n); +} + +static inline u32 msg_importance(struct tipc_msg *m) +{ + return msg_bits(m, 0, 25, 0xf); +} + +static inline void msg_set_importance(struct tipc_msg *m, u32 i) +{ + msg_set_user(m, i); +} + +static inline u32 msg_hdr_sz(struct tipc_msg *m) +{ + return msg_bits(m, 0, 21, 0xf) << 2; +} + +static inline void msg_set_hdr_sz(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 0, 21, 0xf, n>>2); +} + +static inline u32 msg_size(struct tipc_msg *m) +{ + return msg_bits(m, 0, 0, 0x1ffff); +} + +static inline u32 msg_data_sz(struct tipc_msg *m) +{ + return msg_size(m) - msg_hdr_sz(m); +} + +static inline int msg_non_seq(struct tipc_msg *m) +{ + return msg_bits(m, 0, 20, 1); +} + +static inline void msg_set_non_seq(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 0, 20, 1, n); +} + +static inline int msg_dest_droppable(struct tipc_msg *m) +{ + return msg_bits(m, 0, 19, 1); +} + +static inline void msg_set_dest_droppable(struct tipc_msg *m, u32 d) +{ + msg_set_bits(m, 0, 19, 1, d); +} + +static inline int msg_src_droppable(struct tipc_msg *m) +{ + return msg_bits(m, 0, 18, 1); +} + +static inline void msg_set_src_droppable(struct tipc_msg *m, u32 d) +{ + msg_set_bits(m, 0, 18, 1, d); +} + +static inline void msg_set_size(struct tipc_msg *m, u32 sz) +{ + m->hdr[0] = htonl((msg_word(m, 0) & ~0x1ffff) | sz); +} + + +/* + * Word 1 + */ +static inline u32 msg_type(struct tipc_msg *m) +{ + return msg_bits(m, 1, 29, 0x7); +} + +static inline void msg_set_type(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 29, 0x7, n); +} + +static inline u32 msg_named(struct tipc_msg *m) +{ + return msg_type(m) == TIPC_NAMED_MSG; +} + +static inline u32 msg_mcast(struct tipc_msg *m) +{ + return msg_type(m) == TIPC_MCAST_MSG; +} + +static inline u32 msg_connected(struct tipc_msg *m) +{ + return msg_type(m) == TIPC_CONN_MSG; +} + +static inline u32 msg_errcode(struct tipc_msg *m) +{ + return msg_bits(m, 1, 25, 0xf); +} + +static inline void msg_set_errcode(struct tipc_msg *m, u32 err) +{ + msg_set_bits(m, 1, 25, 0xf, err); +} + +static inline u32 msg_reroute_cnt(struct tipc_msg *m) +{ + return msg_bits(m, 1, 21, 0xf); +} + +static inline void msg_incr_reroute_cnt(struct tipc_msg *m) +{ + msg_set_bits(m, 1, 21, 0xf, msg_reroute_cnt(m) + 1); +} + +static inline void msg_reset_reroute_cnt(struct tipc_msg *m) +{ + msg_set_bits(m, 1, 21, 0xf, 0); +} + +static inline u32 msg_lookup_scope(struct tipc_msg *m) +{ + return msg_bits(m, 1, 19, 0x3); +} + +static inline void msg_set_lookup_scope(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 19, 0x3, n); +} + +static inline u32 msg_bcast_ack(struct tipc_msg *m) +{ + return msg_bits(m, 1, 0, 0xffff); +} + +static inline void msg_set_bcast_ack(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 0, 0xffff, n); +} + + +/* + * Word 2 + */ +static inline u32 msg_ack(struct tipc_msg *m) +{ + return msg_bits(m, 2, 16, 0xffff); +} + +static inline void msg_set_ack(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 2, 16, 0xffff, n); +} + +static inline u32 msg_seqno(struct tipc_msg *m) +{ + return msg_bits(m, 2, 0, 0xffff); +} + +static inline void msg_set_seqno(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 2, 0, 0xffff, n); +} + +/* + * Words 3-10 + */ +static inline u32 msg_prevnode(struct tipc_msg *m) +{ + return msg_word(m, 3); +} + +static inline void msg_set_prevnode(struct tipc_msg *m, u32 a) +{ + msg_set_word(m, 3, a); +} + +static inline u32 msg_origport(struct tipc_msg *m) +{ + return msg_word(m, 4); +} + +static inline void msg_set_origport(struct tipc_msg *m, u32 p) +{ + msg_set_word(m, 4, p); +} + +static inline u32 msg_destport(struct tipc_msg *m) +{ + return msg_word(m, 5); +} + +static inline void msg_set_destport(struct tipc_msg *m, u32 p) +{ + msg_set_word(m, 5, p); +} + +static inline u32 msg_mc_netid(struct tipc_msg *m) +{ + return msg_word(m, 5); +} + +static inline void msg_set_mc_netid(struct tipc_msg *m, u32 p) +{ + msg_set_word(m, 5, p); +} + +static inline int msg_short(struct tipc_msg *m) +{ + return msg_hdr_sz(m) == SHORT_H_SIZE; +} + +static inline u32 msg_orignode(struct tipc_msg *m) +{ + if (likely(msg_short(m))) + return msg_prevnode(m); + return msg_word(m, 6); +} + +static inline void msg_set_orignode(struct tipc_msg *m, u32 a) +{ + msg_set_word(m, 6, a); +} + +static inline u32 msg_destnode(struct tipc_msg *m) +{ + return msg_word(m, 7); +} + +static inline void msg_set_destnode(struct tipc_msg *m, u32 a) +{ + msg_set_word(m, 7, a); +} + +static inline u32 msg_nametype(struct tipc_msg *m) +{ + return msg_word(m, 8); +} + +static inline void msg_set_nametype(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 8, n); +} + +static inline u32 msg_nameinst(struct tipc_msg *m) +{ + return msg_word(m, 9); +} + +static inline u32 msg_namelower(struct tipc_msg *m) +{ + return msg_nameinst(m); +} + +static inline void msg_set_namelower(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 9, n); +} + +static inline void msg_set_nameinst(struct tipc_msg *m, u32 n) +{ + msg_set_namelower(m, n); +} + +static inline u32 msg_nameupper(struct tipc_msg *m) +{ + return msg_word(m, 10); +} + +static inline void msg_set_nameupper(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 10, n); +} + +static inline unchar *msg_data(struct tipc_msg *m) +{ + return ((unchar *)m) + msg_hdr_sz(m); +} + +static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) +{ + return (struct tipc_msg *)msg_data(m); +} + +/* + * Constants and routines used to read and write TIPC internal message headers + */ + +/* + * Internal message users + */ +#define BCAST_PROTOCOL 5 +#define MSG_BUNDLER 6 +#define LINK_PROTOCOL 7 +#define CONN_MANAGER 8 +#define ROUTE_DISTRIBUTOR 9 /* obsoleted */ +#define CHANGEOVER_PROTOCOL 10 +#define NAME_DISTRIBUTOR 11 +#define MSG_FRAGMENTER 12 +#define LINK_CONFIG 13 + +/* + * Connection management protocol message types + */ +#define CONN_PROBE 0 +#define CONN_PROBE_REPLY 1 +#define CONN_ACK 2 + +/* + * Name distributor message types + */ +#define PUBLICATION 0 +#define WITHDRAWAL 1 + +/* + * Segmentation message types + */ +#define FIRST_FRAGMENT 0 +#define FRAGMENT 1 +#define LAST_FRAGMENT 2 + +/* + * Link management protocol message types + */ +#define STATE_MSG 0 +#define RESET_MSG 1 +#define ACTIVATE_MSG 2 + +/* + * Changeover tunnel message types + */ +#define DUPLICATE_MSG 0 +#define ORIGINAL_MSG 1 + +/* + * Config protocol message types + */ +#define DSC_REQ_MSG 0 +#define DSC_RESP_MSG 1 + + +/* + * Word 1 + */ +static inline u32 msg_seq_gap(struct tipc_msg *m) +{ + return msg_bits(m, 1, 16, 0x1fff); +} + +static inline void msg_set_seq_gap(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 16, 0x1fff, n); +} + +static inline u32 msg_node_sig(struct tipc_msg *m) +{ + return msg_bits(m, 1, 0, 0xffff); +} + +static inline void msg_set_node_sig(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 0, 0xffff, n); +} + + +/* + * Word 2 + */ +static inline u32 msg_dest_domain(struct tipc_msg *m) +{ + return msg_word(m, 2); +} + +static inline void msg_set_dest_domain(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 2, n); +} + +static inline u32 msg_bcgap_after(struct tipc_msg *m) +{ + return msg_bits(m, 2, 16, 0xffff); +} + +static inline void msg_set_bcgap_after(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 2, 16, 0xffff, n); +} + +static inline u32 msg_bcgap_to(struct tipc_msg *m) +{ + return msg_bits(m, 2, 0, 0xffff); +} + +static inline void msg_set_bcgap_to(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 2, 0, 0xffff, n); +} + + +/* + * Word 4 + */ +static inline u32 msg_last_bcast(struct tipc_msg *m) +{ + return msg_bits(m, 4, 16, 0xffff); +} + +static inline void msg_set_last_bcast(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 16, 0xffff, n); +} + +static inline void msg_set_fragm_no(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 16, 0xffff, n); +} + + +static inline u32 msg_next_sent(struct tipc_msg *m) +{ + return msg_bits(m, 4, 0, 0xffff); +} + +static inline void msg_set_next_sent(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 0, 0xffff, n); +} + +static inline void msg_set_long_msgno(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 0, 0xffff, n); +} + +static inline u32 msg_bc_netid(struct tipc_msg *m) +{ + return msg_word(m, 4); +} + +static inline void msg_set_bc_netid(struct tipc_msg *m, u32 id) +{ + msg_set_word(m, 4, id); +} + +static inline u32 msg_link_selector(struct tipc_msg *m) +{ + return msg_bits(m, 4, 0, 1); +} + +static inline void msg_set_link_selector(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 0, 1, n); +} + +/* + * Word 5 + */ +static inline u32 msg_session(struct tipc_msg *m) +{ + return msg_bits(m, 5, 16, 0xffff); +} + +static inline void msg_set_session(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 5, 16, 0xffff, n); +} + +static inline u32 msg_probe(struct tipc_msg *m) +{ + return msg_bits(m, 5, 0, 1); +} + +static inline void msg_set_probe(struct tipc_msg *m, u32 val) +{ + msg_set_bits(m, 5, 0, 1, val); +} + +static inline char msg_net_plane(struct tipc_msg *m) +{ + return msg_bits(m, 5, 1, 7) + 'A'; +} + +static inline void msg_set_net_plane(struct tipc_msg *m, char n) +{ + msg_set_bits(m, 5, 1, 7, (n - 'A')); +} + +static inline u32 msg_linkprio(struct tipc_msg *m) +{ + return msg_bits(m, 5, 4, 0x1f); +} + +static inline void msg_set_linkprio(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 5, 4, 0x1f, n); +} + +static inline u32 msg_bearer_id(struct tipc_msg *m) +{ + return msg_bits(m, 5, 9, 0x7); +} + +static inline void msg_set_bearer_id(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 5, 9, 0x7, n); +} + +static inline u32 msg_redundant_link(struct tipc_msg *m) +{ + return msg_bits(m, 5, 12, 0x1); +} + +static inline void msg_set_redundant_link(struct tipc_msg *m, u32 r) +{ + msg_set_bits(m, 5, 12, 0x1, r); +} + +static inline char *msg_media_addr(struct tipc_msg *m) +{ + return (char *)&m->hdr[TIPC_MEDIA_ADDR_OFFSET]; +} + +/* + * Word 9 + */ +static inline u32 msg_msgcnt(struct tipc_msg *m) +{ + return msg_bits(m, 9, 16, 0xffff); +} + +static inline void msg_set_msgcnt(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 9, 16, 0xffff, n); +} + +static inline u32 msg_bcast_tag(struct tipc_msg *m) +{ + return msg_bits(m, 9, 16, 0xffff); +} + +static inline void msg_set_bcast_tag(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 9, 16, 0xffff, n); +} + +static inline u32 msg_max_pkt(struct tipc_msg *m) +{ + return msg_bits(m, 9, 16, 0xffff) * 4; +} + +static inline void msg_set_max_pkt(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 9, 16, 0xffff, (n / 4)); +} + +static inline u32 msg_link_tolerance(struct tipc_msg *m) +{ + return msg_bits(m, 9, 0, 0xffff); +} + +static inline void msg_set_link_tolerance(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 9, 0, 0xffff, n); +} + +u32 tipc_msg_tot_importance(struct tipc_msg *m); +void tipc_msg_init(struct tipc_msg *m, u32 user, u32 type, u32 hsize, + u32 destnode); +int tipc_msg_build(struct tipc_msg *hdr, struct iovec const *msg_sect, + unsigned int len, int max_size, struct sk_buff **buf); + +int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf); + +#endif diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c new file mode 100644 index 00000000000..8ce730984aa --- /dev/null +++ b/net/tipc/name_distr.c @@ -0,0 +1,325 @@ +/* + * net/tipc/name_distr.c: TIPC name distribution code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "link.h" +#include "name_distr.h" + +/** + * struct publ_list - list of publications made by this node + * @list: circular list of publications + * @list_size: number of entries in list + */ +struct publ_list { + struct list_head list; + u32 size; +}; + +static struct publ_list publ_zone = { + .list = LIST_HEAD_INIT(publ_zone.list), + .size = 0, +}; + +static struct publ_list publ_cluster = { + .list = LIST_HEAD_INIT(publ_cluster.list), + .size = 0, +}; + +static struct publ_list publ_node = { + .list = LIST_HEAD_INIT(publ_node.list), + .size = 0, +}; + +static struct publ_list *publ_lists[] = { + NULL, + &publ_zone, /* publ_lists[TIPC_ZONE_SCOPE] */ + &publ_cluster, /* publ_lists[TIPC_CLUSTER_SCOPE] */ + &publ_node /* publ_lists[TIPC_NODE_SCOPE] */ +}; + + +/** + * publ_to_item - add publication info to a publication message + */ +static void publ_to_item(struct distr_item *i, struct publication *p) +{ + i->type = htonl(p->type); + i->lower = htonl(p->lower); + i->upper = htonl(p->upper); + i->ref = htonl(p->ref); + i->key = htonl(p->key); +} + +/** + * named_prepare_buf - allocate & initialize a publication message + */ +static struct sk_buff *named_prepare_buf(u32 type, u32 size, u32 dest) +{ + struct sk_buff *buf = tipc_buf_acquire(INT_H_SIZE + size); + struct tipc_msg *msg; + + if (buf != NULL) { + msg = buf_msg(buf); + tipc_msg_init(msg, NAME_DISTRIBUTOR, type, INT_H_SIZE, dest); + msg_set_size(msg, INT_H_SIZE + size); + } + return buf; +} + +void named_cluster_distribute(struct sk_buff *buf) +{ + struct sk_buff *buf_copy; + struct tipc_node *n_ptr; + struct tipc_link *l_ptr; + + rcu_read_lock(); + list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { + tipc_node_lock(n_ptr); + l_ptr = n_ptr->active_links[n_ptr->addr & 1]; + if (l_ptr) { + buf_copy = skb_copy(buf, GFP_ATOMIC); + if (!buf_copy) { + tipc_node_unlock(n_ptr); + break; + } + msg_set_destnode(buf_msg(buf_copy), n_ptr->addr); + __tipc_link_xmit(l_ptr, buf_copy); + } + tipc_node_unlock(n_ptr); + } + rcu_read_unlock(); + + kfree_skb(buf); +} + +/** + * tipc_named_publish - tell other nodes about a new publication by this node + */ +struct sk_buff *tipc_named_publish(struct publication *publ) +{ + struct sk_buff *buf; + struct distr_item *item; + + list_add_tail(&publ->local_list, &publ_lists[publ->scope]->list); + publ_lists[publ->scope]->size++; + + if (publ->scope == TIPC_NODE_SCOPE) + return NULL; + + buf = named_prepare_buf(PUBLICATION, ITEM_SIZE, 0); + if (!buf) { + pr_warn("Publication distribution failure\n"); + return NULL; + } + + item = (struct distr_item *)msg_data(buf_msg(buf)); + publ_to_item(item, publ); + return buf; +} + +/** + * tipc_named_withdraw - tell other nodes about a withdrawn publication by this node + */ +struct sk_buff *tipc_named_withdraw(struct publication *publ) +{ + struct sk_buff *buf; + struct distr_item *item; + + list_del(&publ->local_list); + publ_lists[publ->scope]->size--; + + if (publ->scope == TIPC_NODE_SCOPE) + return NULL; + + buf = named_prepare_buf(WITHDRAWAL, ITEM_SIZE, 0); + if (!buf) { + pr_warn("Withdrawal distribution failure\n"); + return NULL; + } + + item = (struct distr_item *)msg_data(buf_msg(buf)); + publ_to_item(item, publ); + return buf; +} + +/* + * named_distribute - prepare name info for bulk distribution to another node + */ +static void named_distribute(struct list_head *message_list, u32 node, + struct publ_list *pls, u32 max_item_buf) +{ + struct publication *publ; + struct sk_buff *buf = NULL; + struct distr_item *item = NULL; + u32 left = 0; + u32 rest = pls->size * ITEM_SIZE; + + list_for_each_entry(publ, &pls->list, local_list) { + if (!buf) { + left = (rest <= max_item_buf) ? rest : max_item_buf; + rest -= left; + buf = named_prepare_buf(PUBLICATION, left, node); + if (!buf) { + pr_warn("Bulk publication failure\n"); + return; + } + item = (struct distr_item *)msg_data(buf_msg(buf)); + } + publ_to_item(item, publ); + item++; + left -= ITEM_SIZE; + if (!left) { + list_add_tail((struct list_head *)buf, message_list); + buf = NULL; + } + } +} + +/** + * tipc_named_node_up - tell specified node about all publications by this node + */ +void tipc_named_node_up(u32 max_item_buf, u32 node) +{ + LIST_HEAD(message_list); + + read_lock_bh(&tipc_nametbl_lock); + named_distribute(&message_list, node, &publ_cluster, max_item_buf); + named_distribute(&message_list, node, &publ_zone, max_item_buf); + read_unlock_bh(&tipc_nametbl_lock); + + tipc_link_names_xmit(&message_list, node); +} + +/** + * named_purge_publ - remove publication associated with a failed node + * + * Invoked for each publication issued by a newly failed node. + * Removes publication structure from name table & deletes it. + */ +static void named_purge_publ(struct publication *publ) +{ + struct publication *p; + + write_lock_bh(&tipc_nametbl_lock); + p = tipc_nametbl_remove_publ(publ->type, publ->lower, + publ->node, publ->ref, publ->key); + if (p) + tipc_nodesub_unsubscribe(&p->subscr); + write_unlock_bh(&tipc_nametbl_lock); + + if (p != publ) { + pr_err("Unable to remove publication from failed node\n" + " (type=%u, lower=%u, node=0x%x, ref=%u, key=%u)\n", + publ->type, publ->lower, publ->node, publ->ref, + publ->key); + } + + kfree(p); +} + +/** + * tipc_named_rcv - process name table update message sent by another node + */ +void tipc_named_rcv(struct sk_buff *buf) +{ + struct publication *publ; + struct tipc_msg *msg = buf_msg(buf); + struct distr_item *item = (struct distr_item *)msg_data(msg); + u32 count = msg_data_sz(msg) / ITEM_SIZE; + + write_lock_bh(&tipc_nametbl_lock); + while (count--) { + if (msg_type(msg) == PUBLICATION) { + publ = tipc_nametbl_insert_publ(ntohl(item->type), + ntohl(item->lower), + ntohl(item->upper), + TIPC_CLUSTER_SCOPE, + msg_orignode(msg), + ntohl(item->ref), + ntohl(item->key)); + if (publ) { + tipc_nodesub_subscribe(&publ->subscr, + msg_orignode(msg), + publ, + (net_ev_handler) + named_purge_publ); + } + } else if (msg_type(msg) == WITHDRAWAL) { + publ = tipc_nametbl_remove_publ(ntohl(item->type), + ntohl(item->lower), + msg_orignode(msg), + ntohl(item->ref), + ntohl(item->key)); + + if (publ) { + tipc_nodesub_unsubscribe(&publ->subscr); + kfree(publ); + } else { + pr_err("Unable to remove publication by node 0x%x\n" + " (type=%u, lower=%u, ref=%u, key=%u)\n", + msg_orignode(msg), ntohl(item->type), + ntohl(item->lower), ntohl(item->ref), + ntohl(item->key)); + } + } else { + pr_warn("Unrecognized name table message received\n"); + } + item++; + } + write_unlock_bh(&tipc_nametbl_lock); + kfree_skb(buf); +} + +/** + * tipc_named_reinit - re-initialize local publications + * + * This routine is called whenever TIPC networking is enabled. + * All name table entries published by this node are updated to reflect + * the node's new network address. + */ +void tipc_named_reinit(void) +{ + struct publication *publ; + int scope; + + write_lock_bh(&tipc_nametbl_lock); + + for (scope = TIPC_ZONE_SCOPE; scope <= TIPC_NODE_SCOPE; scope++) + list_for_each_entry(publ, &publ_lists[scope]->list, local_list) + publ->node = tipc_own_addr; + + write_unlock_bh(&tipc_nametbl_lock); +} diff --git a/net/tipc/name_distr.h b/net/tipc/name_distr.h new file mode 100644 index 00000000000..b2eed4ec152 --- /dev/null +++ b/net/tipc/name_distr.h @@ -0,0 +1,77 @@ +/* + * net/tipc/name_distr.h: Include file for TIPC name distribution code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NAME_DISTR_H +#define _TIPC_NAME_DISTR_H + +#include "name_table.h" + +#define ITEM_SIZE sizeof(struct distr_item) + +/** + * struct distr_item - publication info distributed to other nodes + * @type: name sequence type + * @lower: name sequence lower bound + * @upper: name sequence upper bound + * @ref: publishing port reference + * @key: publication key + * + * ===> All fields are stored in network byte order. <=== + * + * First 3 fields identify (name or) name sequence being published. + * Reference field uniquely identifies port that published name sequence. + * Key field uniquely identifies publication, in the event a port has + * multiple publications of the same name sequence. + * + * Note: There is no field that identifies the publishing node because it is + * the same for all items contained within a publication message. + */ +struct distr_item { + __be32 type; + __be32 lower; + __be32 upper; + __be32 ref; + __be32 key; +}; + +struct sk_buff *tipc_named_publish(struct publication *publ); +struct sk_buff *tipc_named_withdraw(struct publication *publ); +void named_cluster_distribute(struct sk_buff *buf); +void tipc_named_node_up(u32 max_item_buf, u32 node); +void tipc_named_rcv(struct sk_buff *buf); +void tipc_named_reinit(void); + +#endif diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c new file mode 100644 index 00000000000..9d7d37d9518 --- /dev/null +++ b/net/tipc/name_table.c @@ -0,0 +1,998 @@ +/* + * net/tipc/name_table.c: TIPC name table code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2004-2008, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include "name_table.h" +#include "name_distr.h" +#include "subscr.h" +#include "port.h" + +#define TIPC_NAMETBL_SIZE 1024 /* must be a power of 2 */ + +/** + * struct name_info - name sequence publication info + * @node_list: circular list of publications made by own node + * @cluster_list: circular list of publications made by own cluster + * @zone_list: circular list of publications made by own zone + * @node_list_size: number of entries in "node_list" + * @cluster_list_size: number of entries in "cluster_list" + * @zone_list_size: number of entries in "zone_list" + * + * Note: The zone list always contains at least one entry, since all + * publications of the associated name sequence belong to it. + * (The cluster and node lists may be empty.) + */ +struct name_info { + struct list_head node_list; + struct list_head cluster_list; + struct list_head zone_list; + u32 node_list_size; + u32 cluster_list_size; + u32 zone_list_size; +}; + +/** + * struct sub_seq - container for all published instances of a name sequence + * @lower: name sequence lower bound + * @upper: name sequence upper bound + * @info: pointer to name sequence publication info + */ +struct sub_seq { + u32 lower; + u32 upper; + struct name_info *info; +}; + +/** + * struct name_seq - container for all published instances of a name type + * @type: 32 bit 'type' value for name sequence + * @sseq: pointer to dynamically-sized array of sub-sequences of this 'type'; + * sub-sequences are sorted in ascending order + * @alloc: number of sub-sequences currently in array + * @first_free: array index of first unused sub-sequence entry + * @ns_list: links to adjacent name sequences in hash chain + * @subscriptions: list of subscriptions for this 'type' + * @lock: spinlock controlling access to publication lists of all sub-sequences + */ +struct name_seq { + u32 type; + struct sub_seq *sseqs; + u32 alloc; + u32 first_free; + struct hlist_node ns_list; + struct list_head subscriptions; + spinlock_t lock; +}; + +/** + * struct name_table - table containing all existing port name publications + * @types: pointer to fixed-sized array of name sequence lists, + * accessed via hashing on 'type'; name sequence lists are *not* sorted + * @local_publ_count: number of publications issued by this node + */ +struct name_table { + struct hlist_head *types; + u32 local_publ_count; +}; + +static struct name_table table; +DEFINE_RWLOCK(tipc_nametbl_lock); + +static int hash(int x) +{ + return x & (TIPC_NAMETBL_SIZE - 1); +} + +/** + * publ_create - create a publication structure + */ +static struct publication *publ_create(u32 type, u32 lower, u32 upper, + u32 scope, u32 node, u32 port_ref, + u32 key) +{ + struct publication *publ = kzalloc(sizeof(*publ), GFP_ATOMIC); + if (publ == NULL) { + pr_warn("Publication creation failure, no memory\n"); + return NULL; + } + + publ->type = type; + publ->lower = lower; + publ->upper = upper; + publ->scope = scope; + publ->node = node; + publ->ref = port_ref; + publ->key = key; + INIT_LIST_HEAD(&publ->local_list); + INIT_LIST_HEAD(&publ->pport_list); + INIT_LIST_HEAD(&publ->subscr.nodesub_list); + return publ; +} + +/** + * tipc_subseq_alloc - allocate a specified number of sub-sequence structures + */ +static struct sub_seq *tipc_subseq_alloc(u32 cnt) +{ + return kcalloc(cnt, sizeof(struct sub_seq), GFP_ATOMIC); +} + +/** + * tipc_nameseq_create - create a name sequence structure for the specified 'type' + * + * Allocates a single sub-sequence structure and sets it to all 0's. + */ +static struct name_seq *tipc_nameseq_create(u32 type, struct hlist_head *seq_head) +{ + struct name_seq *nseq = kzalloc(sizeof(*nseq), GFP_ATOMIC); + struct sub_seq *sseq = tipc_subseq_alloc(1); + + if (!nseq || !sseq) { + pr_warn("Name sequence creation failed, no memory\n"); + kfree(nseq); + kfree(sseq); + return NULL; + } + + spin_lock_init(&nseq->lock); + nseq->type = type; + nseq->sseqs = sseq; + nseq->alloc = 1; + INIT_HLIST_NODE(&nseq->ns_list); + INIT_LIST_HEAD(&nseq->subscriptions); + hlist_add_head(&nseq->ns_list, seq_head); + return nseq; +} + +/* + * nameseq_delete_empty - deletes a name sequence structure if now unused + */ +static void nameseq_delete_empty(struct name_seq *seq) +{ + if (!seq->first_free && list_empty(&seq->subscriptions)) { + hlist_del_init(&seq->ns_list); + kfree(seq->sseqs); + kfree(seq); + } +} + +/** + * nameseq_find_subseq - find sub-sequence (if any) matching a name instance + * + * Very time-critical, so binary searches through sub-sequence array. + */ +static struct sub_seq *nameseq_find_subseq(struct name_seq *nseq, + u32 instance) +{ + struct sub_seq *sseqs = nseq->sseqs; + int low = 0; + int high = nseq->first_free - 1; + int mid; + + while (low <= high) { + mid = (low + high) / 2; + if (instance < sseqs[mid].lower) + high = mid - 1; + else if (instance > sseqs[mid].upper) + low = mid + 1; + else + return &sseqs[mid]; + } + return NULL; +} + +/** + * nameseq_locate_subseq - determine position of name instance in sub-sequence + * + * Returns index in sub-sequence array of the entry that contains the specified + * instance value; if no entry contains that value, returns the position + * where a new entry for it would be inserted in the array. + * + * Note: Similar to binary search code for locating a sub-sequence. + */ +static u32 nameseq_locate_subseq(struct name_seq *nseq, u32 instance) +{ + struct sub_seq *sseqs = nseq->sseqs; + int low = 0; + int high = nseq->first_free - 1; + int mid; + + while (low <= high) { + mid = (low + high) / 2; + if (instance < sseqs[mid].lower) + high = mid - 1; + else if (instance > sseqs[mid].upper) + low = mid + 1; + else + return mid; + } + return low; +} + +/** + * tipc_nameseq_insert_publ + */ +static struct publication *tipc_nameseq_insert_publ(struct name_seq *nseq, + u32 type, u32 lower, u32 upper, + u32 scope, u32 node, u32 port, u32 key) +{ + struct tipc_subscription *s; + struct tipc_subscription *st; + struct publication *publ; + struct sub_seq *sseq; + struct name_info *info; + int created_subseq = 0; + + sseq = nameseq_find_subseq(nseq, lower); + if (sseq) { + + /* Lower end overlaps existing entry => need an exact match */ + if ((sseq->lower != lower) || (sseq->upper != upper)) { + pr_warn("Cannot publish {%u,%u,%u}, overlap error\n", + type, lower, upper); + return NULL; + } + + info = sseq->info; + + /* Check if an identical publication already exists */ + list_for_each_entry(publ, &info->zone_list, zone_list) { + if ((publ->ref == port) && (publ->key == key) && + (!publ->node || (publ->node == node))) + return NULL; + } + } else { + u32 inspos; + struct sub_seq *freesseq; + + /* Find where lower end should be inserted */ + inspos = nameseq_locate_subseq(nseq, lower); + + /* Fail if upper end overlaps into an existing entry */ + if ((inspos < nseq->first_free) && + (upper >= nseq->sseqs[inspos].lower)) { + pr_warn("Cannot publish {%u,%u,%u}, overlap error\n", + type, lower, upper); + return NULL; + } + + /* Ensure there is space for new sub-sequence */ + if (nseq->first_free == nseq->alloc) { + struct sub_seq *sseqs = tipc_subseq_alloc(nseq->alloc * 2); + + if (!sseqs) { + pr_warn("Cannot publish {%u,%u,%u}, no memory\n", + type, lower, upper); + return NULL; + } + memcpy(sseqs, nseq->sseqs, + nseq->alloc * sizeof(struct sub_seq)); + kfree(nseq->sseqs); + nseq->sseqs = sseqs; + nseq->alloc *= 2; + } + + info = kzalloc(sizeof(*info), GFP_ATOMIC); + if (!info) { + pr_warn("Cannot publish {%u,%u,%u}, no memory\n", + type, lower, upper); + return NULL; + } + + INIT_LIST_HEAD(&info->node_list); + INIT_LIST_HEAD(&info->cluster_list); + INIT_LIST_HEAD(&info->zone_list); + + /* Insert new sub-sequence */ + sseq = &nseq->sseqs[inspos]; + freesseq = &nseq->sseqs[nseq->first_free]; + memmove(sseq + 1, sseq, (freesseq - sseq) * sizeof(*sseq)); + memset(sseq, 0, sizeof(*sseq)); + nseq->first_free++; + sseq->lower = lower; + sseq->upper = upper; + sseq->info = info; + created_subseq = 1; + } + + /* Insert a publication */ + publ = publ_create(type, lower, upper, scope, node, port, key); + if (!publ) + return NULL; + + list_add(&publ->zone_list, &info->zone_list); + info->zone_list_size++; + + if (in_own_cluster(node)) { + list_add(&publ->cluster_list, &info->cluster_list); + info->cluster_list_size++; + } + + if (in_own_node(node)) { + list_add(&publ->node_list, &info->node_list); + info->node_list_size++; + } + + /* Any subscriptions waiting for notification? */ + list_for_each_entry_safe(s, st, &nseq->subscriptions, nameseq_list) { + tipc_subscr_report_overlap(s, + publ->lower, + publ->upper, + TIPC_PUBLISHED, + publ->ref, + publ->node, + created_subseq); + } + return publ; +} + +/** + * tipc_nameseq_remove_publ + * + * NOTE: There may be cases where TIPC is asked to remove a publication + * that is not in the name table. For example, if another node issues a + * publication for a name sequence that overlaps an existing name sequence + * the publication will not be recorded, which means the publication won't + * be found when the name sequence is later withdrawn by that node. + * A failed withdraw request simply returns a failure indication and lets the + * caller issue any error or warning messages associated with such a problem. + */ +static struct publication *tipc_nameseq_remove_publ(struct name_seq *nseq, u32 inst, + u32 node, u32 ref, u32 key) +{ + struct publication *publ; + struct sub_seq *sseq = nameseq_find_subseq(nseq, inst); + struct name_info *info; + struct sub_seq *free; + struct tipc_subscription *s, *st; + int removed_subseq = 0; + + if (!sseq) + return NULL; + + info = sseq->info; + + /* Locate publication, if it exists */ + list_for_each_entry(publ, &info->zone_list, zone_list) { + if ((publ->key == key) && (publ->ref == ref) && + (!publ->node || (publ->node == node))) + goto found; + } + return NULL; + +found: + /* Remove publication from zone scope list */ + list_del(&publ->zone_list); + info->zone_list_size--; + + /* Remove publication from cluster scope list, if present */ + if (in_own_cluster(node)) { + list_del(&publ->cluster_list); + info->cluster_list_size--; + } + + /* Remove publication from node scope list, if present */ + if (in_own_node(node)) { + list_del(&publ->node_list); + info->node_list_size--; + } + + /* Contract subseq list if no more publications for that subseq */ + if (list_empty(&info->zone_list)) { + kfree(info); + free = &nseq->sseqs[nseq->first_free--]; + memmove(sseq, sseq + 1, (free - (sseq + 1)) * sizeof(*sseq)); + removed_subseq = 1; + } + + /* Notify any waiting subscriptions */ + list_for_each_entry_safe(s, st, &nseq->subscriptions, nameseq_list) { + tipc_subscr_report_overlap(s, + publ->lower, + publ->upper, + TIPC_WITHDRAWN, + publ->ref, + publ->node, + removed_subseq); + } + + return publ; +} + +/** + * tipc_nameseq_subscribe - attach a subscription, and issue + * the prescribed number of events if there is any sub- + * sequence overlapping with the requested sequence + */ +static void tipc_nameseq_subscribe(struct name_seq *nseq, + struct tipc_subscription *s) +{ + struct sub_seq *sseq = nseq->sseqs; + + list_add(&s->nameseq_list, &nseq->subscriptions); + + if (!sseq) + return; + + while (sseq != &nseq->sseqs[nseq->first_free]) { + if (tipc_subscr_overlap(s, sseq->lower, sseq->upper)) { + struct publication *crs; + struct name_info *info = sseq->info; + int must_report = 1; + + list_for_each_entry(crs, &info->zone_list, zone_list) { + tipc_subscr_report_overlap(s, + sseq->lower, + sseq->upper, + TIPC_PUBLISHED, + crs->ref, + crs->node, + must_report); + must_report = 0; + } + } + sseq++; + } +} + +static struct name_seq *nametbl_find_seq(u32 type) +{ + struct hlist_head *seq_head; + struct name_seq *ns; + + seq_head = &table.types[hash(type)]; + hlist_for_each_entry(ns, seq_head, ns_list) { + if (ns->type == type) + return ns; + } + + return NULL; +}; + +struct publication *tipc_nametbl_insert_publ(u32 type, u32 lower, u32 upper, + u32 scope, u32 node, u32 port, u32 key) +{ + struct name_seq *seq = nametbl_find_seq(type); + + if ((scope < TIPC_ZONE_SCOPE) || (scope > TIPC_NODE_SCOPE) || + (lower > upper)) { + pr_debug("Failed to publish illegal {%u,%u,%u} with scope %u\n", + type, lower, upper, scope); + return NULL; + } + + if (!seq) + seq = tipc_nameseq_create(type, &table.types[hash(type)]); + if (!seq) + return NULL; + + return tipc_nameseq_insert_publ(seq, type, lower, upper, + scope, node, port, key); +} + +struct publication *tipc_nametbl_remove_publ(u32 type, u32 lower, + u32 node, u32 ref, u32 key) +{ + struct publication *publ; + struct name_seq *seq = nametbl_find_seq(type); + + if (!seq) + return NULL; + + publ = tipc_nameseq_remove_publ(seq, lower, node, ref, key); + nameseq_delete_empty(seq); + return publ; +} + +/** + * tipc_nametbl_translate - perform name translation + * + * On entry, 'destnode' is the search domain used during translation. + * + * On exit: + * - if name translation is deferred to another node/cluster/zone, + * leaves 'destnode' unchanged (will be non-zero) and returns 0 + * - if name translation is attempted and succeeds, sets 'destnode' + * to publishing node and returns port reference (will be non-zero) + * - if name translation is attempted and fails, sets 'destnode' to 0 + * and returns 0 + */ +u32 tipc_nametbl_translate(u32 type, u32 instance, u32 *destnode) +{ + struct sub_seq *sseq; + struct name_info *info; + struct publication *publ; + struct name_seq *seq; + u32 ref = 0; + u32 node = 0; + + if (!tipc_in_scope(*destnode, tipc_own_addr)) + return 0; + + read_lock_bh(&tipc_nametbl_lock); + seq = nametbl_find_seq(type); + if (unlikely(!seq)) + goto not_found; + sseq = nameseq_find_subseq(seq, instance); + if (unlikely(!sseq)) + goto not_found; + spin_lock_bh(&seq->lock); + info = sseq->info; + + /* Closest-First Algorithm */ + if (likely(!*destnode)) { + if (!list_empty(&info->node_list)) { + publ = list_first_entry(&info->node_list, + struct publication, + node_list); + list_move_tail(&publ->node_list, + &info->node_list); + } else if (!list_empty(&info->cluster_list)) { + publ = list_first_entry(&info->cluster_list, + struct publication, + cluster_list); + list_move_tail(&publ->cluster_list, + &info->cluster_list); + } else { + publ = list_first_entry(&info->zone_list, + struct publication, + zone_list); + list_move_tail(&publ->zone_list, + &info->zone_list); + } + } + + /* Round-Robin Algorithm */ + else if (*destnode == tipc_own_addr) { + if (list_empty(&info->node_list)) + goto no_match; + publ = list_first_entry(&info->node_list, struct publication, + node_list); + list_move_tail(&publ->node_list, &info->node_list); + } else if (in_own_cluster_exact(*destnode)) { + if (list_empty(&info->cluster_list)) + goto no_match; + publ = list_first_entry(&info->cluster_list, struct publication, + cluster_list); + list_move_tail(&publ->cluster_list, &info->cluster_list); + } else { + publ = list_first_entry(&info->zone_list, struct publication, + zone_list); + list_move_tail(&publ->zone_list, &info->zone_list); + } + + ref = publ->ref; + node = publ->node; +no_match: + spin_unlock_bh(&seq->lock); +not_found: + read_unlock_bh(&tipc_nametbl_lock); + *destnode = node; + return ref; +} + +/** + * tipc_nametbl_mc_translate - find multicast destinations + * + * Creates list of all local ports that overlap the given multicast address; + * also determines if any off-node ports overlap. + * + * Note: Publications with a scope narrower than 'limit' are ignored. + * (i.e. local node-scope publications mustn't receive messages arriving + * from another node, even if the multcast link brought it here) + * + * Returns non-zero if any off-node ports overlap + */ +int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, + struct tipc_port_list *dports) +{ + struct name_seq *seq; + struct sub_seq *sseq; + struct sub_seq *sseq_stop; + struct name_info *info; + int res = 0; + + read_lock_bh(&tipc_nametbl_lock); + seq = nametbl_find_seq(type); + if (!seq) + goto exit; + + spin_lock_bh(&seq->lock); + + sseq = seq->sseqs + nameseq_locate_subseq(seq, lower); + sseq_stop = seq->sseqs + seq->first_free; + for (; sseq != sseq_stop; sseq++) { + struct publication *publ; + + if (sseq->lower > upper) + break; + + info = sseq->info; + list_for_each_entry(publ, &info->node_list, node_list) { + if (publ->scope <= limit) + tipc_port_list_add(dports, publ->ref); + } + + if (info->cluster_list_size != info->node_list_size) + res = 1; + } + + spin_unlock_bh(&seq->lock); +exit: + read_unlock_bh(&tipc_nametbl_lock); + return res; +} + +/* + * tipc_nametbl_publish - add name publication to network name tables + */ +struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, + u32 scope, u32 port_ref, u32 key) +{ + struct publication *publ; + struct sk_buff *buf = NULL; + + if (table.local_publ_count >= TIPC_MAX_PUBLICATIONS) { + pr_warn("Publication failed, local publication limit reached (%u)\n", + TIPC_MAX_PUBLICATIONS); + return NULL; + } + + write_lock_bh(&tipc_nametbl_lock); + publ = tipc_nametbl_insert_publ(type, lower, upper, scope, + tipc_own_addr, port_ref, key); + if (likely(publ)) { + table.local_publ_count++; + buf = tipc_named_publish(publ); + } + write_unlock_bh(&tipc_nametbl_lock); + + if (buf) + named_cluster_distribute(buf); + return publ; +} + +/** + * tipc_nametbl_withdraw - withdraw name publication from network name tables + */ +int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key) +{ + struct publication *publ; + struct sk_buff *buf; + + write_lock_bh(&tipc_nametbl_lock); + publ = tipc_nametbl_remove_publ(type, lower, tipc_own_addr, ref, key); + if (likely(publ)) { + table.local_publ_count--; + buf = tipc_named_withdraw(publ); + write_unlock_bh(&tipc_nametbl_lock); + list_del_init(&publ->pport_list); + kfree(publ); + + if (buf) + named_cluster_distribute(buf); + return 1; + } + write_unlock_bh(&tipc_nametbl_lock); + pr_err("Unable to remove local publication\n" + "(type=%u, lower=%u, ref=%u, key=%u)\n", + type, lower, ref, key); + return 0; +} + +/** + * tipc_nametbl_subscribe - add a subscription object to the name table + */ +void tipc_nametbl_subscribe(struct tipc_subscription *s) +{ + u32 type = s->seq.type; + struct name_seq *seq; + + write_lock_bh(&tipc_nametbl_lock); + seq = nametbl_find_seq(type); + if (!seq) + seq = tipc_nameseq_create(type, &table.types[hash(type)]); + if (seq) { + spin_lock_bh(&seq->lock); + tipc_nameseq_subscribe(seq, s); + spin_unlock_bh(&seq->lock); + } else { + pr_warn("Failed to create subscription for {%u,%u,%u}\n", + s->seq.type, s->seq.lower, s->seq.upper); + } + write_unlock_bh(&tipc_nametbl_lock); +} + +/** + * tipc_nametbl_unsubscribe - remove a subscription object from name table + */ +void tipc_nametbl_unsubscribe(struct tipc_subscription *s) +{ + struct name_seq *seq; + + write_lock_bh(&tipc_nametbl_lock); + seq = nametbl_find_seq(s->seq.type); + if (seq != NULL) { + spin_lock_bh(&seq->lock); + list_del_init(&s->nameseq_list); + spin_unlock_bh(&seq->lock); + nameseq_delete_empty(seq); + } + write_unlock_bh(&tipc_nametbl_lock); +} + + +/** + * subseq_list - print specified sub-sequence contents into the given buffer + */ +static int subseq_list(struct sub_seq *sseq, char *buf, int len, u32 depth, + u32 index) +{ + char portIdStr[27]; + const char *scope_str[] = {"", " zone", " cluster", " node"}; + struct publication *publ; + struct name_info *info; + int ret; + + ret = tipc_snprintf(buf, len, "%-10u %-10u ", sseq->lower, sseq->upper); + + if (depth == 2) { + ret += tipc_snprintf(buf - ret, len + ret, "\n"); + return ret; + } + + info = sseq->info; + + list_for_each_entry(publ, &info->zone_list, zone_list) { + sprintf(portIdStr, "<%u.%u.%u:%u>", + tipc_zone(publ->node), tipc_cluster(publ->node), + tipc_node(publ->node), publ->ref); + ret += tipc_snprintf(buf + ret, len - ret, "%-26s ", portIdStr); + if (depth > 3) { + ret += tipc_snprintf(buf + ret, len - ret, "%-10u %s", + publ->key, scope_str[publ->scope]); + } + if (!list_is_last(&publ->zone_list, &info->zone_list)) + ret += tipc_snprintf(buf + ret, len - ret, + "\n%33s", " "); + } + + ret += tipc_snprintf(buf + ret, len - ret, "\n"); + return ret; +} + +/** + * nameseq_list - print specified name sequence contents into the given buffer + */ +static int nameseq_list(struct name_seq *seq, char *buf, int len, u32 depth, + u32 type, u32 lowbound, u32 upbound, u32 index) +{ + struct sub_seq *sseq; + char typearea[11]; + int ret = 0; + + if (seq->first_free == 0) + return 0; + + sprintf(typearea, "%-10u", seq->type); + + if (depth == 1) { + ret += tipc_snprintf(buf, len, "%s\n", typearea); + return ret; + } + + for (sseq = seq->sseqs; sseq != &seq->sseqs[seq->first_free]; sseq++) { + if ((lowbound <= sseq->upper) && (upbound >= sseq->lower)) { + ret += tipc_snprintf(buf + ret, len - ret, "%s ", + typearea); + spin_lock_bh(&seq->lock); + ret += subseq_list(sseq, buf + ret, len - ret, + depth, index); + spin_unlock_bh(&seq->lock); + sprintf(typearea, "%10s", " "); + } + } + return ret; +} + +/** + * nametbl_header - print name table header into the given buffer + */ +static int nametbl_header(char *buf, int len, u32 depth) +{ + const char *header[] = { + "Type ", + "Lower Upper ", + "Port Identity ", + "Publication Scope" + }; + + int i; + int ret = 0; + + if (depth > 4) + depth = 4; + for (i = 0; i < depth; i++) + ret += tipc_snprintf(buf + ret, len - ret, header[i]); + ret += tipc_snprintf(buf + ret, len - ret, "\n"); + return ret; +} + +/** + * nametbl_list - print specified name table contents into the given buffer + */ +static int nametbl_list(char *buf, int len, u32 depth_info, + u32 type, u32 lowbound, u32 upbound) +{ + struct hlist_head *seq_head; + struct name_seq *seq; + int all_types; + int ret = 0; + u32 depth; + u32 i; + + all_types = (depth_info & TIPC_NTQ_ALLTYPES); + depth = (depth_info & ~TIPC_NTQ_ALLTYPES); + + if (depth == 0) + return 0; + + if (all_types) { + /* display all entries in name table to specified depth */ + ret += nametbl_header(buf, len, depth); + lowbound = 0; + upbound = ~0; + for (i = 0; i < TIPC_NAMETBL_SIZE; i++) { + seq_head = &table.types[i]; + hlist_for_each_entry(seq, seq_head, ns_list) { + ret += nameseq_list(seq, buf + ret, len - ret, + depth, seq->type, + lowbound, upbound, i); + } + } + } else { + /* display only the sequence that matches the specified type */ + if (upbound < lowbound) { + ret += tipc_snprintf(buf + ret, len - ret, + "invalid name sequence specified\n"); + return ret; + } + ret += nametbl_header(buf + ret, len - ret, depth); + i = hash(type); + seq_head = &table.types[i]; + hlist_for_each_entry(seq, seq_head, ns_list) { + if (seq->type == type) { + ret += nameseq_list(seq, buf + ret, len - ret, + depth, type, + lowbound, upbound, i); + break; + } + } + } + return ret; +} + +struct sk_buff *tipc_nametbl_get(const void *req_tlv_area, int req_tlv_space) +{ + struct sk_buff *buf; + struct tipc_name_table_query *argv; + struct tlv_desc *rep_tlv; + char *pb; + int pb_len; + int str_len; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NAME_TBL_QUERY)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); + if (!buf) + return NULL; + + rep_tlv = (struct tlv_desc *)buf->data; + pb = TLV_DATA(rep_tlv); + pb_len = ULTRA_STRING_MAX_LEN; + argv = (struct tipc_name_table_query *)TLV_DATA(req_tlv_area); + read_lock_bh(&tipc_nametbl_lock); + str_len = nametbl_list(pb, pb_len, ntohl(argv->depth), + ntohl(argv->type), + ntohl(argv->lowbound), ntohl(argv->upbound)); + read_unlock_bh(&tipc_nametbl_lock); + str_len += 1; /* for "\0" */ + skb_put(buf, TLV_SPACE(str_len)); + TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); + + return buf; +} + +int tipc_nametbl_init(void) +{ + table.types = kcalloc(TIPC_NAMETBL_SIZE, sizeof(struct hlist_head), + GFP_ATOMIC); + if (!table.types) + return -ENOMEM; + + table.local_publ_count = 0; + return 0; +} + +/** + * tipc_purge_publications - remove all publications for a given type + * + * tipc_nametbl_lock must be held when calling this function + */ +static void tipc_purge_publications(struct name_seq *seq) +{ + struct publication *publ, *safe; + struct sub_seq *sseq; + struct name_info *info; + + if (!seq->sseqs) { + nameseq_delete_empty(seq); + return; + } + sseq = seq->sseqs; + info = sseq->info; + list_for_each_entry_safe(publ, safe, &info->zone_list, zone_list) { + tipc_nametbl_remove_publ(publ->type, publ->lower, publ->node, + publ->ref, publ->key); + kfree(publ); + } +} + +void tipc_nametbl_stop(void) +{ + u32 i; + struct name_seq *seq; + struct hlist_head *seq_head; + struct hlist_node *safe; + + /* Verify name table is empty and purge any lingering + * publications, then release the name table + */ + write_lock_bh(&tipc_nametbl_lock); + for (i = 0; i < TIPC_NAMETBL_SIZE; i++) { + if (hlist_empty(&table.types[i])) + continue; + seq_head = &table.types[i]; + hlist_for_each_entry_safe(seq, safe, seq_head, ns_list) { + tipc_purge_publications(seq); + } + } + kfree(table.types); + table.types = NULL; + write_unlock_bh(&tipc_nametbl_lock); +} diff --git a/net/tipc/name_table.h b/net/tipc/name_table.h new file mode 100644 index 00000000000..f02f48b9a21 --- /dev/null +++ b/net/tipc/name_table.h @@ -0,0 +1,104 @@ +/* + * net/tipc/name_table.h: Include file for TIPC name table code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2004-2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NAME_TABLE_H +#define _TIPC_NAME_TABLE_H + +#include "node_subscr.h" + +struct tipc_subscription; +struct tipc_port_list; + +/* + * TIPC name types reserved for internal TIPC use (both current and planned) + */ +#define TIPC_ZM_SRV 3 /* zone master service name type */ + +/** + * struct publication - info about a published (name or) name sequence + * @type: name sequence type + * @lower: name sequence lower bound + * @upper: name sequence upper bound + * @scope: scope of publication + * @node: network address of publishing port's node + * @ref: publishing port + * @key: publication key + * @subscr: subscription to "node down" event (for off-node publications only) + * @local_list: adjacent entries in list of publications made by this node + * @pport_list: adjacent entries in list of publications made by this port + * @node_list: adjacent matching name seq publications with >= node scope + * @cluster_list: adjacent matching name seq publications with >= cluster scope + * @zone_list: adjacent matching name seq publications with >= zone scope + * + * Note that the node list, cluster list, and zone list are circular lists. + */ +struct publication { + u32 type; + u32 lower; + u32 upper; + u32 scope; + u32 node; + u32 ref; + u32 key; + struct tipc_node_subscr subscr; + struct list_head local_list; + struct list_head pport_list; + struct list_head node_list; + struct list_head cluster_list; + struct list_head zone_list; +}; + + +extern rwlock_t tipc_nametbl_lock; + +struct sk_buff *tipc_nametbl_get(const void *req_tlv_area, int req_tlv_space); +u32 tipc_nametbl_translate(u32 type, u32 instance, u32 *node); +int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, + struct tipc_port_list *dports); +struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, + u32 scope, u32 port_ref, u32 key); +int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key); +struct publication *tipc_nametbl_insert_publ(u32 type, u32 lower, u32 upper, + u32 scope, u32 node, u32 ref, + u32 key); +struct publication *tipc_nametbl_remove_publ(u32 type, u32 lower, u32 node, + u32 ref, u32 key); +void tipc_nametbl_subscribe(struct tipc_subscription *s); +void tipc_nametbl_unsubscribe(struct tipc_subscription *s); +int tipc_nametbl_init(void); +void tipc_nametbl_stop(void); + +#endif diff --git a/net/tipc/net.c b/net/tipc/net.c new file mode 100644 index 00000000000..f64375e7f99 --- /dev/null +++ b/net/tipc/net.c @@ -0,0 +1,202 @@ +/* + * net/tipc/net.c: TIPC network routing code + * + * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "net.h" +#include "name_distr.h" +#include "subscr.h" +#include "port.h" +#include "socket.h" +#include "node.h" +#include "config.h" + +/* + * The TIPC locking policy is designed to ensure a very fine locking + * granularity, permitting complete parallel access to individual + * port and node/link instances. The code consists of four major + * locking domains, each protected with their own disjunct set of locks. + * + * 1: The bearer level. + * RTNL lock is used to serialize the process of configuring bearer + * on update side, and RCU lock is applied on read side to make + * bearer instance valid on both paths of message transmission and + * reception. + * + * 2: The node and link level. + * All node instances are saved into two tipc_node_list and node_htable + * lists. The two lists are protected by node_list_lock on write side, + * and they are guarded with RCU lock on read side. Especially node + * instance is destroyed only when TIPC module is removed, and we can + * confirm that there has no any user who is accessing the node at the + * moment. Therefore, Except for iterating the two lists within RCU + * protection, it's no needed to hold RCU that we access node instance + * in other places. + * + * In addition, all members in node structure including link instances + * are protected by node spin lock. + * + * 3: The transport level of the protocol. + * This consists of the structures port, (and its user level + * representations, such as user_port and tipc_sock), reference and + * tipc_user (port.c, reg.c, socket.c). + * + * This layer has four different locks: + * - The tipc_port spin_lock. This is protecting each port instance + * from parallel data access and removal. Since we can not place + * this lock in the port itself, it has been placed in the + * corresponding reference table entry, which has the same life + * cycle as the module. This entry is difficult to access from + * outside the TIPC core, however, so a pointer to the lock has + * been added in the port instance, -to be used for unlocking + * only. + * - A read/write lock to protect the reference table itself (teg.c). + * (Nobody is using read-only access to this, so it can just as + * well be changed to a spin_lock) + * - A spin lock to protect the registry of kernel/driver users (reg.c) + * - A global spin_lock (tipc_port_lock), which only task is to ensure + * consistency where more than one port is involved in an operation, + * i.e., whe a port is part of a linked list of ports. + * There are two such lists; 'port_list', which is used for management, + * and 'wait_list', which is used to queue ports during congestion. + * + * 4: The name table (name_table.c, name_distr.c, subscription.c) + * - There is one big read/write-lock (tipc_nametbl_lock) protecting the + * overall name table structure. Nothing must be added/removed to + * this structure without holding write access to it. + * - There is one local spin_lock per sub_sequence, which can be seen + * as a sub-domain to the tipc_nametbl_lock domain. It is used only + * for translation operations, and is needed because a translation + * steps the root of the 'publication' linked list between each lookup. + * This is always used within the scope of a tipc_nametbl_lock(read). + * - A local spin_lock protecting the queue of subscriber events. +*/ + +static void net_route_named_msg(struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + u32 dnode; + u32 dport; + + if (!msg_named(msg)) { + kfree_skb(buf); + return; + } + + dnode = addr_domain(msg_lookup_scope(msg)); + dport = tipc_nametbl_translate(msg_nametype(msg), msg_nameinst(msg), &dnode); + if (dport) { + msg_set_destnode(msg, dnode); + msg_set_destport(msg, dport); + tipc_net_route_msg(buf); + return; + } + tipc_reject_msg(buf, TIPC_ERR_NO_NAME); +} + +void tipc_net_route_msg(struct sk_buff *buf) +{ + struct tipc_msg *msg; + u32 dnode; + + if (!buf) + return; + msg = buf_msg(buf); + + /* Handle message for this node */ + dnode = msg_short(msg) ? tipc_own_addr : msg_destnode(msg); + if (tipc_in_scope(dnode, tipc_own_addr)) { + if (msg_isdata(msg)) { + if (msg_mcast(msg)) + tipc_port_mcast_rcv(buf, NULL); + else if (msg_destport(msg)) + tipc_sk_rcv(buf); + else + net_route_named_msg(buf); + return; + } + switch (msg_user(msg)) { + case NAME_DISTRIBUTOR: + tipc_named_rcv(buf); + break; + case CONN_MANAGER: + tipc_port_proto_rcv(buf); + break; + default: + kfree_skb(buf); + } + return; + } + + /* Handle message for another node */ + skb_trim(buf, msg_size(msg)); + tipc_link_xmit(buf, dnode, msg_link_selector(msg)); +} + +int tipc_net_start(u32 addr) +{ + char addr_string[16]; + int res; + + tipc_own_addr = addr; + tipc_named_reinit(); + tipc_port_reinit(); + res = tipc_bclink_init(); + if (res) + return res; + + tipc_nametbl_publish(TIPC_CFG_SRV, tipc_own_addr, tipc_own_addr, + TIPC_ZONE_SCOPE, 0, tipc_own_addr); + + pr_info("Started in network mode\n"); + pr_info("Own node address %s, network identity %u\n", + tipc_addr_string_fill(addr_string, tipc_own_addr), tipc_net_id); + return 0; +} + +void tipc_net_stop(void) +{ + if (!tipc_own_addr) + return; + + tipc_nametbl_withdraw(TIPC_CFG_SRV, tipc_own_addr, 0, tipc_own_addr); + rtnl_lock(); + tipc_bearer_stop(); + tipc_bclink_stop(); + tipc_node_stop(); + rtnl_unlock(); + + pr_info("Left network mode\n"); +} diff --git a/net/tipc/net.h b/net/tipc/net.h new file mode 100644 index 00000000000..c6c2b46f7c2 --- /dev/null +++ b/net/tipc/net.h @@ -0,0 +1,45 @@ +/* + * net/tipc/net.h: Include file for TIPC network routing code + * + * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NET_H +#define _TIPC_NET_H + +void tipc_net_route_msg(struct sk_buff *buf); + +int tipc_net_start(u32 addr); +void tipc_net_stop(void); + +#endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c new file mode 100644 index 00000000000..ad844d36534 --- /dev/null +++ b/net/tipc/netlink.c @@ -0,0 +1,101 @@ +/* + * net/tipc/netlink.c: TIPC configuration handling + * + * Copyright (c) 2005-2006, Ericsson AB + * Copyright (c) 2005-2007, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include <net/genetlink.h> + +static int handle_cmd(struct sk_buff *skb, struct genl_info *info) +{ + struct sk_buff *rep_buf; + struct nlmsghdr *rep_nlh; + struct nlmsghdr *req_nlh = info->nlhdr; + struct tipc_genlmsghdr *req_userhdr = info->userhdr; + int hdr_space = nlmsg_total_size(GENL_HDRLEN + TIPC_GENL_HDRLEN); + u16 cmd; + + if ((req_userhdr->cmd & 0xC000) && (!netlink_capable(skb, CAP_NET_ADMIN))) + cmd = TIPC_CMD_NOT_NET_ADMIN; + else + cmd = req_userhdr->cmd; + + rep_buf = tipc_cfg_do_cmd(req_userhdr->dest, cmd, + nlmsg_data(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN, + nlmsg_attrlen(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN), + hdr_space); + + if (rep_buf) { + skb_push(rep_buf, hdr_space); + rep_nlh = nlmsg_hdr(rep_buf); + memcpy(rep_nlh, req_nlh, hdr_space); + rep_nlh->nlmsg_len = rep_buf->len; + genlmsg_unicast(&init_net, rep_buf, NETLINK_CB(skb).portid); + } + + return 0; +} + +static struct genl_family tipc_genl_family = { + .id = GENL_ID_GENERATE, + .name = TIPC_GENL_NAME, + .version = TIPC_GENL_VERSION, + .hdrsize = TIPC_GENL_HDRLEN, + .maxattr = 0, +}; + +static struct genl_ops tipc_genl_ops[] = { + { + .cmd = TIPC_GENL_CMD, + .doit = handle_cmd, + }, +}; + +int tipc_netlink_start(void) +{ + int res; + + res = genl_register_family_with_ops(&tipc_genl_family, tipc_genl_ops); + if (res) { + pr_err("Failed to register netlink interface\n"); + return res; + } + return 0; +} + +void tipc_netlink_stop(void) +{ + genl_unregister_family(&tipc_genl_family); +} diff --git a/net/tipc/node.c b/net/tipc/node.c new file mode 100644 index 00000000000..5b44c3041be --- /dev/null +++ b/net/tipc/node.c @@ -0,0 +1,488 @@ +/* + * net/tipc/node.c: TIPC node management routines + * + * Copyright (c) 2000-2006, 2012 Ericsson AB + * Copyright (c) 2005-2006, 2010-2014, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include "node.h" +#include "name_distr.h" + +#define NODE_HTABLE_SIZE 512 + +static void node_lost_contact(struct tipc_node *n_ptr); +static void node_established_contact(struct tipc_node *n_ptr); + +static struct hlist_head node_htable[NODE_HTABLE_SIZE]; +LIST_HEAD(tipc_node_list); +static u32 tipc_num_nodes; +static u32 tipc_num_links; +static DEFINE_SPINLOCK(node_list_lock); + +/* + * A trivial power-of-two bitmask technique is used for speed, since this + * operation is done for every incoming TIPC packet. The number of hash table + * entries has been chosen so that no hash chain exceeds 8 nodes and will + * usually be much smaller (typically only a single node). + */ +static unsigned int tipc_hashfn(u32 addr) +{ + return addr & (NODE_HTABLE_SIZE - 1); +} + +/* + * tipc_node_find - locate specified node object, if it exists + */ +struct tipc_node *tipc_node_find(u32 addr) +{ + struct tipc_node *node; + + if (unlikely(!in_own_cluster_exact(addr))) + return NULL; + + rcu_read_lock(); + hlist_for_each_entry_rcu(node, &node_htable[tipc_hashfn(addr)], hash) { + if (node->addr == addr) { + rcu_read_unlock(); + return node; + } + } + rcu_read_unlock(); + return NULL; +} + +struct tipc_node *tipc_node_create(u32 addr) +{ + struct tipc_node *n_ptr, *temp_node; + + spin_lock_bh(&node_list_lock); + + n_ptr = kzalloc(sizeof(*n_ptr), GFP_ATOMIC); + if (!n_ptr) { + spin_unlock_bh(&node_list_lock); + pr_warn("Node creation failed, no memory\n"); + return NULL; + } + + n_ptr->addr = addr; + spin_lock_init(&n_ptr->lock); + INIT_HLIST_NODE(&n_ptr->hash); + INIT_LIST_HEAD(&n_ptr->list); + INIT_LIST_HEAD(&n_ptr->nsub); + + hlist_add_head_rcu(&n_ptr->hash, &node_htable[tipc_hashfn(addr)]); + + list_for_each_entry_rcu(temp_node, &tipc_node_list, list) { + if (n_ptr->addr < temp_node->addr) + break; + } + list_add_tail_rcu(&n_ptr->list, &temp_node->list); + n_ptr->action_flags = TIPC_WAIT_PEER_LINKS_DOWN; + n_ptr->signature = INVALID_NODE_SIG; + + tipc_num_nodes++; + + spin_unlock_bh(&node_list_lock); + return n_ptr; +} + +static void tipc_node_delete(struct tipc_node *n_ptr) +{ + list_del_rcu(&n_ptr->list); + hlist_del_rcu(&n_ptr->hash); + kfree_rcu(n_ptr, rcu); + + tipc_num_nodes--; +} + +void tipc_node_stop(void) +{ + struct tipc_node *node, *t_node; + + spin_lock_bh(&node_list_lock); + list_for_each_entry_safe(node, t_node, &tipc_node_list, list) + tipc_node_delete(node); + spin_unlock_bh(&node_list_lock); +} + +/** + * tipc_node_link_up - handle addition of link + * + * Link becomes active (alone or shared) or standby, depending on its priority. + */ +void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr) +{ + struct tipc_link **active = &n_ptr->active_links[0]; + u32 addr = n_ptr->addr; + + n_ptr->working_links++; + tipc_nametbl_publish(TIPC_LINK_STATE, addr, addr, TIPC_NODE_SCOPE, + l_ptr->bearer_id, addr); + pr_info("Established link <%s> on network plane %c\n", + l_ptr->name, l_ptr->net_plane); + + if (!active[0]) { + active[0] = active[1] = l_ptr; + node_established_contact(n_ptr); + return; + } + if (l_ptr->priority < active[0]->priority) { + pr_info("New link <%s> becomes standby\n", l_ptr->name); + return; + } + tipc_link_dup_queue_xmit(active[0], l_ptr); + if (l_ptr->priority == active[0]->priority) { + active[0] = l_ptr; + return; + } + pr_info("Old link <%s> becomes standby\n", active[0]->name); + if (active[1] != active[0]) + pr_info("Old link <%s> becomes standby\n", active[1]->name); + active[0] = active[1] = l_ptr; +} + +/** + * node_select_active_links - select active link + */ +static void node_select_active_links(struct tipc_node *n_ptr) +{ + struct tipc_link **active = &n_ptr->active_links[0]; + u32 i; + u32 highest_prio = 0; + + active[0] = active[1] = NULL; + + for (i = 0; i < MAX_BEARERS; i++) { + struct tipc_link *l_ptr = n_ptr->links[i]; + + if (!l_ptr || !tipc_link_is_up(l_ptr) || + (l_ptr->priority < highest_prio)) + continue; + + if (l_ptr->priority > highest_prio) { + highest_prio = l_ptr->priority; + active[0] = active[1] = l_ptr; + } else { + active[1] = l_ptr; + } + } +} + +/** + * tipc_node_link_down - handle loss of link + */ +void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr) +{ + struct tipc_link **active; + u32 addr = n_ptr->addr; + + n_ptr->working_links--; + tipc_nametbl_withdraw(TIPC_LINK_STATE, addr, l_ptr->bearer_id, addr); + + if (!tipc_link_is_active(l_ptr)) { + pr_info("Lost standby link <%s> on network plane %c\n", + l_ptr->name, l_ptr->net_plane); + return; + } + pr_info("Lost link <%s> on network plane %c\n", + l_ptr->name, l_ptr->net_plane); + + active = &n_ptr->active_links[0]; + if (active[0] == l_ptr) + active[0] = active[1]; + if (active[1] == l_ptr) + active[1] = active[0]; + if (active[0] == l_ptr) + node_select_active_links(n_ptr); + if (tipc_node_is_up(n_ptr)) + tipc_link_failover_send_queue(l_ptr); + else + node_lost_contact(n_ptr); +} + +int tipc_node_active_links(struct tipc_node *n_ptr) +{ + return n_ptr->active_links[0] != NULL; +} + +int tipc_node_is_up(struct tipc_node *n_ptr) +{ + return tipc_node_active_links(n_ptr); +} + +void tipc_node_attach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr) +{ + n_ptr->links[l_ptr->bearer_id] = l_ptr; + spin_lock_bh(&node_list_lock); + tipc_num_links++; + spin_unlock_bh(&node_list_lock); + n_ptr->link_cnt++; +} + +void tipc_node_detach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr) +{ + int i; + + for (i = 0; i < MAX_BEARERS; i++) { + if (l_ptr != n_ptr->links[i]) + continue; + n_ptr->links[i] = NULL; + spin_lock_bh(&node_list_lock); + tipc_num_links--; + spin_unlock_bh(&node_list_lock); + n_ptr->link_cnt--; + } +} + +static void node_established_contact(struct tipc_node *n_ptr) +{ + n_ptr->action_flags |= TIPC_NOTIFY_NODE_UP; + n_ptr->bclink.oos_state = 0; + n_ptr->bclink.acked = tipc_bclink_get_last_sent(); + tipc_bclink_add_node(n_ptr->addr); +} + +static void node_lost_contact(struct tipc_node *n_ptr) +{ + char addr_string[16]; + u32 i; + + pr_info("Lost contact with %s\n", + tipc_addr_string_fill(addr_string, n_ptr->addr)); + + /* Flush broadcast link info associated with lost node */ + if (n_ptr->bclink.recv_permitted) { + kfree_skb_list(n_ptr->bclink.deferred_head); + n_ptr->bclink.deferred_size = 0; + + if (n_ptr->bclink.reasm_buf) { + kfree_skb(n_ptr->bclink.reasm_buf); + n_ptr->bclink.reasm_buf = NULL; + } + + tipc_bclink_remove_node(n_ptr->addr); + tipc_bclink_acknowledge(n_ptr, INVALID_LINK_SEQ); + + n_ptr->bclink.recv_permitted = false; + } + + /* Abort link changeover */ + for (i = 0; i < MAX_BEARERS; i++) { + struct tipc_link *l_ptr = n_ptr->links[i]; + if (!l_ptr) + continue; + l_ptr->reset_checkpoint = l_ptr->next_in_no; + l_ptr->exp_msg_count = 0; + tipc_link_reset_fragments(l_ptr); + } + + n_ptr->action_flags &= ~TIPC_WAIT_OWN_LINKS_DOWN; + + /* Notify subscribers and prevent re-contact with node until + * cleanup is done. + */ + n_ptr->action_flags |= TIPC_WAIT_PEER_LINKS_DOWN | + TIPC_NOTIFY_NODE_DOWN; +} + +struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) +{ + u32 domain; + struct sk_buff *buf; + struct tipc_node *n_ptr; + struct tipc_node_info node_info; + u32 payload_size; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + domain = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); + if (!tipc_addr_domain_valid(domain)) + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (network address)"); + + spin_lock_bh(&node_list_lock); + if (!tipc_num_nodes) { + spin_unlock_bh(&node_list_lock); + return tipc_cfg_reply_none(); + } + + /* For now, get space for all other nodes */ + payload_size = TLV_SPACE(sizeof(node_info)) * tipc_num_nodes; + if (payload_size > 32768u) { + spin_unlock_bh(&node_list_lock); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (too many nodes)"); + } + spin_unlock_bh(&node_list_lock); + + buf = tipc_cfg_reply_alloc(payload_size); + if (!buf) + return NULL; + + /* Add TLVs for all nodes in scope */ + rcu_read_lock(); + list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { + if (!tipc_in_scope(domain, n_ptr->addr)) + continue; + node_info.addr = htonl(n_ptr->addr); + node_info.up = htonl(tipc_node_is_up(n_ptr)); + tipc_cfg_append_tlv(buf, TIPC_TLV_NODE_INFO, + &node_info, sizeof(node_info)); + } + rcu_read_unlock(); + return buf; +} + +struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) +{ + u32 domain; + struct sk_buff *buf; + struct tipc_node *n_ptr; + struct tipc_link_info link_info; + u32 payload_size; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + domain = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); + if (!tipc_addr_domain_valid(domain)) + return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (network address)"); + + if (!tipc_own_addr) + return tipc_cfg_reply_none(); + + spin_lock_bh(&node_list_lock); + /* Get space for all unicast links + broadcast link */ + payload_size = TLV_SPACE((sizeof(link_info)) * (tipc_num_links + 1)); + if (payload_size > 32768u) { + spin_unlock_bh(&node_list_lock); + return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (too many links)"); + } + spin_unlock_bh(&node_list_lock); + + buf = tipc_cfg_reply_alloc(payload_size); + if (!buf) + return NULL; + + /* Add TLV for broadcast link */ + link_info.dest = htonl(tipc_cluster_mask(tipc_own_addr)); + link_info.up = htonl(1); + strlcpy(link_info.str, tipc_bclink_name, TIPC_MAX_LINK_NAME); + tipc_cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, &link_info, sizeof(link_info)); + + /* Add TLVs for any other links in scope */ + rcu_read_lock(); + list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { + u32 i; + + if (!tipc_in_scope(domain, n_ptr->addr)) + continue; + tipc_node_lock(n_ptr); + for (i = 0; i < MAX_BEARERS; i++) { + if (!n_ptr->links[i]) + continue; + link_info.dest = htonl(n_ptr->addr); + link_info.up = htonl(tipc_link_is_up(n_ptr->links[i])); + strcpy(link_info.str, n_ptr->links[i]->name); + tipc_cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, + &link_info, sizeof(link_info)); + } + tipc_node_unlock(n_ptr); + } + rcu_read_unlock(); + return buf; +} + +/** + * tipc_node_get_linkname - get the name of a link + * + * @bearer_id: id of the bearer + * @node: peer node address + * @linkname: link name output buffer + * + * Returns 0 on success + */ +int tipc_node_get_linkname(u32 bearer_id, u32 addr, char *linkname, size_t len) +{ + struct tipc_link *link; + struct tipc_node *node = tipc_node_find(addr); + + if ((bearer_id >= MAX_BEARERS) || !node) + return -EINVAL; + tipc_node_lock(node); + link = node->links[bearer_id]; + if (link) { + strncpy(linkname, link->name, len); + tipc_node_unlock(node); + return 0; + } + tipc_node_unlock(node); + return -EINVAL; +} + +void tipc_node_unlock(struct tipc_node *node) +{ + LIST_HEAD(nsub_list); + struct tipc_link *link; + int pkt_sz = 0; + u32 addr = 0; + + if (likely(!node->action_flags)) { + spin_unlock_bh(&node->lock); + return; + } + + if (node->action_flags & TIPC_NOTIFY_NODE_DOWN) { + list_replace_init(&node->nsub, &nsub_list); + node->action_flags &= ~TIPC_NOTIFY_NODE_DOWN; + } + if (node->action_flags & TIPC_NOTIFY_NODE_UP) { + link = node->active_links[0]; + node->action_flags &= ~TIPC_NOTIFY_NODE_UP; + if (link) { + pkt_sz = ((link->max_pkt - INT_H_SIZE) / ITEM_SIZE) * + ITEM_SIZE; + addr = node->addr; + } + } + spin_unlock_bh(&node->lock); + + if (!list_empty(&nsub_list)) + tipc_nodesub_notify(&nsub_list); + if (pkt_sz) + tipc_named_node_up(pkt_sz, addr); +} diff --git a/net/tipc/node.h b/net/tipc/node.h new file mode 100644 index 00000000000..9087063793f --- /dev/null +++ b/net/tipc/node.h @@ -0,0 +1,146 @@ +/* + * net/tipc/node.h: Include file for TIPC node management routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, 2010-2014, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NODE_H +#define _TIPC_NODE_H + +#include "node_subscr.h" +#include "addr.h" +#include "net.h" +#include "bearer.h" + +/* + * Out-of-range value for node signature + */ +#define INVALID_NODE_SIG 0x10000 + +/* Flags used to take different actions according to flag type + * TIPC_WAIT_PEER_LINKS_DOWN: wait to see that peer's links are down + * TIPC_WAIT_OWN_LINKS_DOWN: wait until peer node is declared down + * TIPC_NOTIFY_NODE_DOWN: notify node is down + * TIPC_NOTIFY_NODE_UP: notify node is up + */ +enum { + TIPC_WAIT_PEER_LINKS_DOWN = (1 << 1), + TIPC_WAIT_OWN_LINKS_DOWN = (1 << 2), + TIPC_NOTIFY_NODE_DOWN = (1 << 3), + TIPC_NOTIFY_NODE_UP = (1 << 4) +}; + +/** + * struct tipc_node_bclink - TIPC node bclink structure + * @acked: sequence # of last outbound b'cast message acknowledged by node + * @last_in: sequence # of last in-sequence b'cast message received from node + * @last_sent: sequence # of last b'cast message sent by node + * @oos_state: state tracker for handling OOS b'cast messages + * @deferred_size: number of OOS b'cast messages in deferred queue + * @deferred_head: oldest OOS b'cast message received from node + * @deferred_tail: newest OOS b'cast message received from node + * @reasm_buf: broadcast reassembly queue head from node + * @recv_permitted: true if node is allowed to receive b'cast messages + */ +struct tipc_node_bclink { + u32 acked; + u32 last_in; + u32 last_sent; + u32 oos_state; + u32 deferred_size; + struct sk_buff *deferred_head; + struct sk_buff *deferred_tail; + struct sk_buff *reasm_buf; + bool recv_permitted; +}; + +/** + * struct tipc_node - TIPC node structure + * @addr: network address of node + * @lock: spinlock governing access to structure + * @hash: links to adjacent nodes in unsorted hash chain + * @active_links: pointers to active links to node + * @links: pointers to all links to node + * @action_flags: bit mask of different types of node actions + * @bclink: broadcast-related info + * @list: links to adjacent nodes in sorted list of cluster's nodes + * @working_links: number of working links to node (both active and standby) + * @link_cnt: number of links to node + * @signature: node instance identifier + * @nsub: list of "node down" subscriptions monitoring node + * @rcu: rcu struct for tipc_node + */ +struct tipc_node { + u32 addr; + spinlock_t lock; + struct hlist_node hash; + struct tipc_link *active_links[2]; + struct tipc_link *links[MAX_BEARERS]; + unsigned int action_flags; + struct tipc_node_bclink bclink; + struct list_head list; + int link_cnt; + int working_links; + u32 signature; + struct list_head nsub; + struct rcu_head rcu; +}; + +extern struct list_head tipc_node_list; + +struct tipc_node *tipc_node_find(u32 addr); +struct tipc_node *tipc_node_create(u32 addr); +void tipc_node_stop(void); +void tipc_node_attach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr); +void tipc_node_detach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr); +void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr); +void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr); +int tipc_node_active_links(struct tipc_node *n_ptr); +int tipc_node_is_up(struct tipc_node *n_ptr); +struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space); +struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space); +int tipc_node_get_linkname(u32 bearer_id, u32 node, char *linkname, size_t len); +void tipc_node_unlock(struct tipc_node *node); + +static inline void tipc_node_lock(struct tipc_node *node) +{ + spin_lock_bh(&node->lock); +} + +static inline bool tipc_node_blocked(struct tipc_node *node) +{ + return (node->action_flags & (TIPC_WAIT_PEER_LINKS_DOWN | + TIPC_NOTIFY_NODE_DOWN | TIPC_WAIT_OWN_LINKS_DOWN)); +} + +#endif diff --git a/net/tipc/node_subscr.c b/net/tipc/node_subscr.c new file mode 100644 index 00000000000..7c59ab1d6ec --- /dev/null +++ b/net/tipc/node_subscr.c @@ -0,0 +1,94 @@ +/* + * net/tipc/node_subscr.c: TIPC "node down" subscription handling + * + * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "node_subscr.h" +#include "node.h" + +/** + * tipc_nodesub_subscribe - create "node down" subscription for specified node + */ +void tipc_nodesub_subscribe(struct tipc_node_subscr *node_sub, u32 addr, + void *usr_handle, net_ev_handler handle_down) +{ + if (in_own_node(addr)) { + node_sub->node = NULL; + return; + } + + node_sub->node = tipc_node_find(addr); + if (!node_sub->node) { + pr_warn("Node subscription rejected, unknown node 0x%x\n", + addr); + return; + } + node_sub->handle_node_down = handle_down; + node_sub->usr_handle = usr_handle; + + tipc_node_lock(node_sub->node); + list_add_tail(&node_sub->nodesub_list, &node_sub->node->nsub); + tipc_node_unlock(node_sub->node); +} + +/** + * tipc_nodesub_unsubscribe - cancel "node down" subscription (if any) + */ +void tipc_nodesub_unsubscribe(struct tipc_node_subscr *node_sub) +{ + if (!node_sub->node) + return; + + tipc_node_lock(node_sub->node); + list_del_init(&node_sub->nodesub_list); + tipc_node_unlock(node_sub->node); +} + +/** + * tipc_nodesub_notify - notify subscribers that a node is unreachable + * + * Note: node is locked by caller + */ +void tipc_nodesub_notify(struct list_head *nsub_list) +{ + struct tipc_node_subscr *ns, *safe; + + list_for_each_entry_safe(ns, safe, nsub_list, nodesub_list) { + if (ns->handle_node_down) { + ns->handle_node_down(ns->usr_handle); + ns->handle_node_down = NULL; + } + } +} diff --git a/net/tipc/node_subscr.h b/net/tipc/node_subscr.h new file mode 100644 index 00000000000..d91b8cc81e3 --- /dev/null +++ b/net/tipc/node_subscr.h @@ -0,0 +1,63 @@ +/* + * net/tipc/node_subscr.h: Include file for TIPC "node down" subscription handling + * + * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 2005, 2010-2011, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NODE_SUBSCR_H +#define _TIPC_NODE_SUBSCR_H + +#include "addr.h" + +typedef void (*net_ev_handler) (void *usr_handle); + +/** + * struct tipc_node_subscr - "node down" subscription entry + * @node: ptr to node structure of interest (or NULL, if none) + * @handle_node_down: routine to invoke when node fails + * @usr_handle: argument to pass to routine when node fails + * @nodesub_list: adjacent entries in list of subscriptions for the node + */ +struct tipc_node_subscr { + struct tipc_node *node; + net_ev_handler handle_node_down; + void *usr_handle; + struct list_head nodesub_list; +}; + +void tipc_nodesub_subscribe(struct tipc_node_subscr *node_sub, u32 addr, + void *usr_handle, net_ev_handler handle_down); +void tipc_nodesub_unsubscribe(struct tipc_node_subscr *node_sub); +void tipc_nodesub_notify(struct list_head *nsub_list); + +#endif diff --git a/net/tipc/port.c b/net/tipc/port.c new file mode 100644 index 00000000000..5fd7acce01e --- /dev/null +++ b/net/tipc/port.c @@ -0,0 +1,898 @@ +/* + * net/tipc/port.c: TIPC port code + * + * Copyright (c) 1992-2007, 2014, Ericsson AB + * Copyright (c) 2004-2008, 2010-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include "port.h" +#include "name_table.h" +#include "socket.h" + +/* Connection management: */ +#define PROBING_INTERVAL 3600000 /* [ms] => 1 h */ +#define CONFIRMED 0 +#define PROBING 1 + +#define MAX_REJECT_SIZE 1024 + +DEFINE_SPINLOCK(tipc_port_list_lock); + +static LIST_HEAD(ports); +static void port_handle_node_down(unsigned long ref); +static struct sk_buff *port_build_self_abort_msg(struct tipc_port *, u32 err); +static struct sk_buff *port_build_peer_abort_msg(struct tipc_port *, u32 err); +static void port_timeout(unsigned long ref); + +/** + * tipc_port_peer_msg - verify message was sent by connected port's peer + * + * Handles cases where the node's network address has changed from + * the default of <0.0.0> to its configured setting. + */ +int tipc_port_peer_msg(struct tipc_port *p_ptr, struct tipc_msg *msg) +{ + u32 peernode; + u32 orignode; + + if (msg_origport(msg) != tipc_port_peerport(p_ptr)) + return 0; + + orignode = msg_orignode(msg); + peernode = tipc_port_peernode(p_ptr); + return (orignode == peernode) || + (!orignode && (peernode == tipc_own_addr)) || + (!peernode && (orignode == tipc_own_addr)); +} + +/** + * tipc_port_mcast_xmit - send a multicast message to local and remote + * destinations + */ +int tipc_port_mcast_xmit(struct tipc_port *oport, + struct tipc_name_seq const *seq, + struct iovec const *msg_sect, + unsigned int len) +{ + struct tipc_msg *hdr; + struct sk_buff *buf; + struct sk_buff *ibuf = NULL; + struct tipc_port_list dports = {0, NULL, }; + int ext_targets; + int res; + + /* Create multicast message */ + hdr = &oport->phdr; + msg_set_type(hdr, TIPC_MCAST_MSG); + msg_set_lookup_scope(hdr, TIPC_CLUSTER_SCOPE); + msg_set_destport(hdr, 0); + msg_set_destnode(hdr, 0); + msg_set_nametype(hdr, seq->type); + msg_set_namelower(hdr, seq->lower); + msg_set_nameupper(hdr, seq->upper); + msg_set_hdr_sz(hdr, MCAST_H_SIZE); + res = tipc_msg_build(hdr, msg_sect, len, MAX_MSG_SIZE, &buf); + if (unlikely(!buf)) + return res; + + /* Figure out where to send multicast message */ + ext_targets = tipc_nametbl_mc_translate(seq->type, seq->lower, seq->upper, + TIPC_NODE_SCOPE, &dports); + + /* Send message to destinations (duplicate it only if necessary) */ + if (ext_targets) { + if (dports.count != 0) { + ibuf = skb_copy(buf, GFP_ATOMIC); + if (ibuf == NULL) { + tipc_port_list_free(&dports); + kfree_skb(buf); + return -ENOMEM; + } + } + res = tipc_bclink_xmit(buf); + if ((res < 0) && (dports.count != 0)) + kfree_skb(ibuf); + } else { + ibuf = buf; + } + + if (res >= 0) { + if (ibuf) + tipc_port_mcast_rcv(ibuf, &dports); + } else { + tipc_port_list_free(&dports); + } + return res; +} + +/** + * tipc_port_mcast_rcv - deliver multicast message to all destination ports + * + * If there is no port list, perform a lookup to create one + */ +void tipc_port_mcast_rcv(struct sk_buff *buf, struct tipc_port_list *dp) +{ + struct tipc_msg *msg; + struct tipc_port_list dports = {0, NULL, }; + struct tipc_port_list *item = dp; + int cnt = 0; + + msg = buf_msg(buf); + + /* Create destination port list, if one wasn't supplied */ + if (dp == NULL) { + tipc_nametbl_mc_translate(msg_nametype(msg), + msg_namelower(msg), + msg_nameupper(msg), + TIPC_CLUSTER_SCOPE, + &dports); + item = dp = &dports; + } + + /* Deliver a copy of message to each destination port */ + if (dp->count != 0) { + msg_set_destnode(msg, tipc_own_addr); + if (dp->count == 1) { + msg_set_destport(msg, dp->ports[0]); + tipc_sk_rcv(buf); + tipc_port_list_free(dp); + return; + } + for (; cnt < dp->count; cnt++) { + int index = cnt % PLSIZE; + struct sk_buff *b = skb_clone(buf, GFP_ATOMIC); + + if (b == NULL) { + pr_warn("Unable to deliver multicast message(s)\n"); + goto exit; + } + if ((index == 0) && (cnt != 0)) + item = item->next; + msg_set_destport(buf_msg(b), item->ports[index]); + tipc_sk_rcv(b); + } + } +exit: + kfree_skb(buf); + tipc_port_list_free(dp); +} + + +void tipc_port_wakeup(struct tipc_port *port) +{ + tipc_sock_wakeup(tipc_port_to_sock(port)); +} + +/* tipc_port_init - intiate TIPC port and lock it + * + * Returns obtained reference if initialization is successful, zero otherwise + */ +u32 tipc_port_init(struct tipc_port *p_ptr, + const unsigned int importance) +{ + struct tipc_msg *msg; + u32 ref; + + ref = tipc_ref_acquire(p_ptr, &p_ptr->lock); + if (!ref) { + pr_warn("Port registration failed, ref. table exhausted\n"); + return 0; + } + + p_ptr->max_pkt = MAX_PKT_DEFAULT; + p_ptr->ref = ref; + INIT_LIST_HEAD(&p_ptr->wait_list); + INIT_LIST_HEAD(&p_ptr->subscription.nodesub_list); + k_init_timer(&p_ptr->timer, (Handler)port_timeout, ref); + INIT_LIST_HEAD(&p_ptr->publications); + INIT_LIST_HEAD(&p_ptr->port_list); + + /* + * Must hold port list lock while initializing message header template + * to ensure a change to node's own network address doesn't result + * in template containing out-dated network address information + */ + spin_lock_bh(&tipc_port_list_lock); + msg = &p_ptr->phdr; + tipc_msg_init(msg, importance, TIPC_NAMED_MSG, NAMED_H_SIZE, 0); + msg_set_origport(msg, ref); + list_add_tail(&p_ptr->port_list, &ports); + spin_unlock_bh(&tipc_port_list_lock); + return ref; +} + +void tipc_port_destroy(struct tipc_port *p_ptr) +{ + struct sk_buff *buf = NULL; + + tipc_withdraw(p_ptr, 0, NULL); + + spin_lock_bh(p_ptr->lock); + tipc_ref_discard(p_ptr->ref); + spin_unlock_bh(p_ptr->lock); + + k_cancel_timer(&p_ptr->timer); + if (p_ptr->connected) { + buf = port_build_peer_abort_msg(p_ptr, TIPC_ERR_NO_PORT); + tipc_nodesub_unsubscribe(&p_ptr->subscription); + } + + spin_lock_bh(&tipc_port_list_lock); + list_del(&p_ptr->port_list); + list_del(&p_ptr->wait_list); + spin_unlock_bh(&tipc_port_list_lock); + k_term_timer(&p_ptr->timer); + tipc_net_route_msg(buf); +} + +/* + * port_build_proto_msg(): create connection protocol message for port + * + * On entry the port must be locked and connected. + */ +static struct sk_buff *port_build_proto_msg(struct tipc_port *p_ptr, + u32 type, u32 ack) +{ + struct sk_buff *buf; + struct tipc_msg *msg; + + buf = tipc_buf_acquire(INT_H_SIZE); + if (buf) { + msg = buf_msg(buf); + tipc_msg_init(msg, CONN_MANAGER, type, INT_H_SIZE, + tipc_port_peernode(p_ptr)); + msg_set_destport(msg, tipc_port_peerport(p_ptr)); + msg_set_origport(msg, p_ptr->ref); + msg_set_msgcnt(msg, ack); + } + return buf; +} + +int tipc_reject_msg(struct sk_buff *buf, u32 err) +{ + struct tipc_msg *msg = buf_msg(buf); + struct sk_buff *rbuf; + struct tipc_msg *rmsg; + int hdr_sz; + u32 imp; + u32 data_sz = msg_data_sz(msg); + u32 src_node; + u32 rmsg_sz; + + /* discard rejected message if it shouldn't be returned to sender */ + if (WARN(!msg_isdata(msg), + "attempt to reject message with user=%u", msg_user(msg))) { + dump_stack(); + goto exit; + } + if (msg_errcode(msg) || msg_dest_droppable(msg)) + goto exit; + + /* + * construct returned message by copying rejected message header and + * data (or subset), then updating header fields that need adjusting + */ + hdr_sz = msg_hdr_sz(msg); + rmsg_sz = hdr_sz + min_t(u32, data_sz, MAX_REJECT_SIZE); + + rbuf = tipc_buf_acquire(rmsg_sz); + if (rbuf == NULL) + goto exit; + + rmsg = buf_msg(rbuf); + skb_copy_to_linear_data(rbuf, msg, rmsg_sz); + + if (msg_connected(rmsg)) { + imp = msg_importance(rmsg); + if (imp < TIPC_CRITICAL_IMPORTANCE) + msg_set_importance(rmsg, ++imp); + } + msg_set_non_seq(rmsg, 0); + msg_set_size(rmsg, rmsg_sz); + msg_set_errcode(rmsg, err); + msg_set_prevnode(rmsg, tipc_own_addr); + msg_swap_words(rmsg, 4, 5); + if (!msg_short(rmsg)) + msg_swap_words(rmsg, 6, 7); + + /* send self-abort message when rejecting on a connected port */ + if (msg_connected(msg)) { + struct tipc_port *p_ptr = tipc_port_lock(msg_destport(msg)); + + if (p_ptr) { + struct sk_buff *abuf = NULL; + + if (p_ptr->connected) + abuf = port_build_self_abort_msg(p_ptr, err); + tipc_port_unlock(p_ptr); + tipc_net_route_msg(abuf); + } + } + + /* send returned message & dispose of rejected message */ + src_node = msg_prevnode(msg); + if (in_own_node(src_node)) + tipc_sk_rcv(rbuf); + else + tipc_link_xmit(rbuf, src_node, msg_link_selector(rmsg)); +exit: + kfree_skb(buf); + return data_sz; +} + +int tipc_port_iovec_reject(struct tipc_port *p_ptr, struct tipc_msg *hdr, + struct iovec const *msg_sect, unsigned int len, + int err) +{ + struct sk_buff *buf; + int res; + + res = tipc_msg_build(hdr, msg_sect, len, MAX_MSG_SIZE, &buf); + if (!buf) + return res; + + return tipc_reject_msg(buf, err); +} + +static void port_timeout(unsigned long ref) +{ + struct tipc_port *p_ptr = tipc_port_lock(ref); + struct sk_buff *buf = NULL; + + if (!p_ptr) + return; + + if (!p_ptr->connected) { + tipc_port_unlock(p_ptr); + return; + } + + /* Last probe answered ? */ + if (p_ptr->probing_state == PROBING) { + buf = port_build_self_abort_msg(p_ptr, TIPC_ERR_NO_PORT); + } else { + buf = port_build_proto_msg(p_ptr, CONN_PROBE, 0); + p_ptr->probing_state = PROBING; + k_start_timer(&p_ptr->timer, p_ptr->probing_interval); + } + tipc_port_unlock(p_ptr); + tipc_net_route_msg(buf); +} + + +static void port_handle_node_down(unsigned long ref) +{ + struct tipc_port *p_ptr = tipc_port_lock(ref); + struct sk_buff *buf = NULL; + + if (!p_ptr) + return; + buf = port_build_self_abort_msg(p_ptr, TIPC_ERR_NO_NODE); + tipc_port_unlock(p_ptr); + tipc_net_route_msg(buf); +} + + +static struct sk_buff *port_build_self_abort_msg(struct tipc_port *p_ptr, u32 err) +{ + struct sk_buff *buf = port_build_peer_abort_msg(p_ptr, err); + + if (buf) { + struct tipc_msg *msg = buf_msg(buf); + msg_swap_words(msg, 4, 5); + msg_swap_words(msg, 6, 7); + } + return buf; +} + + +static struct sk_buff *port_build_peer_abort_msg(struct tipc_port *p_ptr, u32 err) +{ + struct sk_buff *buf; + struct tipc_msg *msg; + u32 imp; + + if (!p_ptr->connected) + return NULL; + + buf = tipc_buf_acquire(BASIC_H_SIZE); + if (buf) { + msg = buf_msg(buf); + memcpy(msg, &p_ptr->phdr, BASIC_H_SIZE); + msg_set_hdr_sz(msg, BASIC_H_SIZE); + msg_set_size(msg, BASIC_H_SIZE); + imp = msg_importance(msg); + if (imp < TIPC_CRITICAL_IMPORTANCE) + msg_set_importance(msg, ++imp); + msg_set_errcode(msg, err); + } + return buf; +} + +void tipc_port_proto_rcv(struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + struct tipc_port *p_ptr; + struct sk_buff *r_buf = NULL; + u32 destport = msg_destport(msg); + int wakeable; + + /* Validate connection */ + p_ptr = tipc_port_lock(destport); + if (!p_ptr || !p_ptr->connected || !tipc_port_peer_msg(p_ptr, msg)) { + r_buf = tipc_buf_acquire(BASIC_H_SIZE); + if (r_buf) { + msg = buf_msg(r_buf); + tipc_msg_init(msg, TIPC_HIGH_IMPORTANCE, TIPC_CONN_MSG, + BASIC_H_SIZE, msg_orignode(msg)); + msg_set_errcode(msg, TIPC_ERR_NO_PORT); + msg_set_origport(msg, destport); + msg_set_destport(msg, msg_origport(msg)); + } + if (p_ptr) + tipc_port_unlock(p_ptr); + goto exit; + } + + /* Process protocol message sent by peer */ + switch (msg_type(msg)) { + case CONN_ACK: + wakeable = tipc_port_congested(p_ptr) && p_ptr->congested; + p_ptr->acked += msg_msgcnt(msg); + if (!tipc_port_congested(p_ptr)) { + p_ptr->congested = 0; + if (wakeable) + tipc_port_wakeup(p_ptr); + } + break; + case CONN_PROBE: + r_buf = port_build_proto_msg(p_ptr, CONN_PROBE_REPLY, 0); + break; + default: + /* CONN_PROBE_REPLY or unrecognized - no action required */ + break; + } + p_ptr->probing_state = CONFIRMED; + tipc_port_unlock(p_ptr); +exit: + tipc_net_route_msg(r_buf); + kfree_skb(buf); +} + +static int port_print(struct tipc_port *p_ptr, char *buf, int len, int full_id) +{ + struct publication *publ; + int ret; + + if (full_id) + ret = tipc_snprintf(buf, len, "<%u.%u.%u:%u>:", + tipc_zone(tipc_own_addr), + tipc_cluster(tipc_own_addr), + tipc_node(tipc_own_addr), p_ptr->ref); + else + ret = tipc_snprintf(buf, len, "%-10u:", p_ptr->ref); + + if (p_ptr->connected) { + u32 dport = tipc_port_peerport(p_ptr); + u32 destnode = tipc_port_peernode(p_ptr); + + ret += tipc_snprintf(buf + ret, len - ret, + " connected to <%u.%u.%u:%u>", + tipc_zone(destnode), + tipc_cluster(destnode), + tipc_node(destnode), dport); + if (p_ptr->conn_type != 0) + ret += tipc_snprintf(buf + ret, len - ret, + " via {%u,%u}", p_ptr->conn_type, + p_ptr->conn_instance); + } else if (p_ptr->published) { + ret += tipc_snprintf(buf + ret, len - ret, " bound to"); + list_for_each_entry(publ, &p_ptr->publications, pport_list) { + if (publ->lower == publ->upper) + ret += tipc_snprintf(buf + ret, len - ret, + " {%u,%u}", publ->type, + publ->lower); + else + ret += tipc_snprintf(buf + ret, len - ret, + " {%u,%u,%u}", publ->type, + publ->lower, publ->upper); + } + } + ret += tipc_snprintf(buf + ret, len - ret, "\n"); + return ret; +} + +struct sk_buff *tipc_port_get_ports(void) +{ + struct sk_buff *buf; + struct tlv_desc *rep_tlv; + char *pb; + int pb_len; + struct tipc_port *p_ptr; + int str_len = 0; + + buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); + if (!buf) + return NULL; + rep_tlv = (struct tlv_desc *)buf->data; + pb = TLV_DATA(rep_tlv); + pb_len = ULTRA_STRING_MAX_LEN; + + spin_lock_bh(&tipc_port_list_lock); + list_for_each_entry(p_ptr, &ports, port_list) { + spin_lock_bh(p_ptr->lock); + str_len += port_print(p_ptr, pb, pb_len, 0); + spin_unlock_bh(p_ptr->lock); + } + spin_unlock_bh(&tipc_port_list_lock); + str_len += 1; /* for "\0" */ + skb_put(buf, TLV_SPACE(str_len)); + TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); + + return buf; +} + +void tipc_port_reinit(void) +{ + struct tipc_port *p_ptr; + struct tipc_msg *msg; + + spin_lock_bh(&tipc_port_list_lock); + list_for_each_entry(p_ptr, &ports, port_list) { + msg = &p_ptr->phdr; + msg_set_prevnode(msg, tipc_own_addr); + msg_set_orignode(msg, tipc_own_addr); + } + spin_unlock_bh(&tipc_port_list_lock); +} + +void tipc_acknowledge(u32 ref, u32 ack) +{ + struct tipc_port *p_ptr; + struct sk_buff *buf = NULL; + + p_ptr = tipc_port_lock(ref); + if (!p_ptr) + return; + if (p_ptr->connected) { + p_ptr->conn_unacked -= ack; + buf = port_build_proto_msg(p_ptr, CONN_ACK, ack); + } + tipc_port_unlock(p_ptr); + tipc_net_route_msg(buf); +} + +int tipc_publish(struct tipc_port *p_ptr, unsigned int scope, + struct tipc_name_seq const *seq) +{ + struct publication *publ; + u32 key; + + if (p_ptr->connected) + return -EINVAL; + key = p_ptr->ref + p_ptr->pub_count + 1; + if (key == p_ptr->ref) + return -EADDRINUSE; + + publ = tipc_nametbl_publish(seq->type, seq->lower, seq->upper, + scope, p_ptr->ref, key); + if (publ) { + list_add(&publ->pport_list, &p_ptr->publications); + p_ptr->pub_count++; + p_ptr->published = 1; + return 0; + } + return -EINVAL; +} + +int tipc_withdraw(struct tipc_port *p_ptr, unsigned int scope, + struct tipc_name_seq const *seq) +{ + struct publication *publ; + struct publication *tpubl; + int res = -EINVAL; + + if (!seq) { + list_for_each_entry_safe(publ, tpubl, + &p_ptr->publications, pport_list) { + tipc_nametbl_withdraw(publ->type, publ->lower, + publ->ref, publ->key); + } + res = 0; + } else { + list_for_each_entry_safe(publ, tpubl, + &p_ptr->publications, pport_list) { + if (publ->scope != scope) + continue; + if (publ->type != seq->type) + continue; + if (publ->lower != seq->lower) + continue; + if (publ->upper != seq->upper) + break; + tipc_nametbl_withdraw(publ->type, publ->lower, + publ->ref, publ->key); + res = 0; + break; + } + } + if (list_empty(&p_ptr->publications)) + p_ptr->published = 0; + return res; +} + +int tipc_port_connect(u32 ref, struct tipc_portid const *peer) +{ + struct tipc_port *p_ptr; + int res; + + p_ptr = tipc_port_lock(ref); + if (!p_ptr) + return -EINVAL; + res = __tipc_port_connect(ref, p_ptr, peer); + tipc_port_unlock(p_ptr); + return res; +} + +/* + * __tipc_port_connect - connect to a remote peer + * + * Port must be locked. + */ +int __tipc_port_connect(u32 ref, struct tipc_port *p_ptr, + struct tipc_portid const *peer) +{ + struct tipc_msg *msg; + int res = -EINVAL; + + if (p_ptr->published || p_ptr->connected) + goto exit; + if (!peer->ref) + goto exit; + + msg = &p_ptr->phdr; + msg_set_destnode(msg, peer->node); + msg_set_destport(msg, peer->ref); + msg_set_type(msg, TIPC_CONN_MSG); + msg_set_lookup_scope(msg, 0); + msg_set_hdr_sz(msg, SHORT_H_SIZE); + + p_ptr->probing_interval = PROBING_INTERVAL; + p_ptr->probing_state = CONFIRMED; + p_ptr->connected = 1; + k_start_timer(&p_ptr->timer, p_ptr->probing_interval); + + tipc_nodesub_subscribe(&p_ptr->subscription, peer->node, + (void *)(unsigned long)ref, + (net_ev_handler)port_handle_node_down); + res = 0; +exit: + p_ptr->max_pkt = tipc_link_get_max_pkt(peer->node, ref); + return res; +} + +/* + * __tipc_disconnect - disconnect port from peer + * + * Port must be locked. + */ +int __tipc_port_disconnect(struct tipc_port *tp_ptr) +{ + if (tp_ptr->connected) { + tp_ptr->connected = 0; + /* let timer expire on it's own to avoid deadlock! */ + tipc_nodesub_unsubscribe(&tp_ptr->subscription); + return 0; + } + + return -ENOTCONN; +} + +/* + * tipc_port_disconnect(): Disconnect port form peer. + * This is a node local operation. + */ +int tipc_port_disconnect(u32 ref) +{ + struct tipc_port *p_ptr; + int res; + + p_ptr = tipc_port_lock(ref); + if (!p_ptr) + return -EINVAL; + res = __tipc_port_disconnect(p_ptr); + tipc_port_unlock(p_ptr); + return res; +} + +/* + * tipc_port_shutdown(): Send a SHUTDOWN msg to peer and disconnect + */ +int tipc_port_shutdown(u32 ref) +{ + struct tipc_port *p_ptr; + struct sk_buff *buf = NULL; + + p_ptr = tipc_port_lock(ref); + if (!p_ptr) + return -EINVAL; + + buf = port_build_peer_abort_msg(p_ptr, TIPC_CONN_SHUTDOWN); + tipc_port_unlock(p_ptr); + tipc_net_route_msg(buf); + return tipc_port_disconnect(ref); +} + +/* + * tipc_port_iovec_rcv: Concatenate and deliver sectioned + * message for this node. + */ +static int tipc_port_iovec_rcv(struct tipc_port *sender, + struct iovec const *msg_sect, + unsigned int len) +{ + struct sk_buff *buf; + int res; + + res = tipc_msg_build(&sender->phdr, msg_sect, len, MAX_MSG_SIZE, &buf); + if (likely(buf)) + tipc_sk_rcv(buf); + return res; +} + +/** + * tipc_send - send message sections on connection + */ +int tipc_send(struct tipc_port *p_ptr, + struct iovec const *msg_sect, + unsigned int len) +{ + u32 destnode; + int res; + + if (!p_ptr->connected) + return -EINVAL; + + p_ptr->congested = 1; + if (!tipc_port_congested(p_ptr)) { + destnode = tipc_port_peernode(p_ptr); + if (likely(!in_own_node(destnode))) + res = tipc_link_iovec_xmit_fast(p_ptr, msg_sect, len, + destnode); + else + res = tipc_port_iovec_rcv(p_ptr, msg_sect, len); + + if (likely(res != -ELINKCONG)) { + p_ptr->congested = 0; + if (res > 0) + p_ptr->sent++; + return res; + } + } + if (tipc_port_unreliable(p_ptr)) { + p_ptr->congested = 0; + return len; + } + return -ELINKCONG; +} + +/** + * tipc_send2name - send message sections to port name + */ +int tipc_send2name(struct tipc_port *p_ptr, + struct tipc_name const *name, + unsigned int domain, + struct iovec const *msg_sect, + unsigned int len) +{ + struct tipc_msg *msg; + u32 destnode = domain; + u32 destport; + int res; + + if (p_ptr->connected) + return -EINVAL; + + msg = &p_ptr->phdr; + msg_set_type(msg, TIPC_NAMED_MSG); + msg_set_hdr_sz(msg, NAMED_H_SIZE); + msg_set_nametype(msg, name->type); + msg_set_nameinst(msg, name->instance); + msg_set_lookup_scope(msg, tipc_addr_scope(domain)); + destport = tipc_nametbl_translate(name->type, name->instance, &destnode); + msg_set_destnode(msg, destnode); + msg_set_destport(msg, destport); + + if (likely(destport || destnode)) { + if (likely(in_own_node(destnode))) + res = tipc_port_iovec_rcv(p_ptr, msg_sect, len); + else if (tipc_own_addr) + res = tipc_link_iovec_xmit_fast(p_ptr, msg_sect, len, + destnode); + else + res = tipc_port_iovec_reject(p_ptr, msg, msg_sect, + len, TIPC_ERR_NO_NODE); + if (likely(res != -ELINKCONG)) { + if (res > 0) + p_ptr->sent++; + return res; + } + if (tipc_port_unreliable(p_ptr)) + return len; + + return -ELINKCONG; + } + return tipc_port_iovec_reject(p_ptr, msg, msg_sect, len, + TIPC_ERR_NO_NAME); +} + +/** + * tipc_send2port - send message sections to port identity + */ +int tipc_send2port(struct tipc_port *p_ptr, + struct tipc_portid const *dest, + struct iovec const *msg_sect, + unsigned int len) +{ + struct tipc_msg *msg; + int res; + + if (p_ptr->connected) + return -EINVAL; + + msg = &p_ptr->phdr; + msg_set_type(msg, TIPC_DIRECT_MSG); + msg_set_lookup_scope(msg, 0); + msg_set_destnode(msg, dest->node); + msg_set_destport(msg, dest->ref); + msg_set_hdr_sz(msg, BASIC_H_SIZE); + + if (in_own_node(dest->node)) + res = tipc_port_iovec_rcv(p_ptr, msg_sect, len); + else if (tipc_own_addr) + res = tipc_link_iovec_xmit_fast(p_ptr, msg_sect, len, + dest->node); + else + res = tipc_port_iovec_reject(p_ptr, msg, msg_sect, len, + TIPC_ERR_NO_NODE); + if (likely(res != -ELINKCONG)) { + if (res > 0) + p_ptr->sent++; + return res; + } + if (tipc_port_unreliable(p_ptr)) + return len; + + return -ELINKCONG; +} diff --git a/net/tipc/port.h b/net/tipc/port.h new file mode 100644 index 00000000000..cf4ca5b1d9a --- /dev/null +++ b/net/tipc/port.h @@ -0,0 +1,237 @@ +/* + * net/tipc/port.h: Include file for TIPC port code + * + * Copyright (c) 1994-2007, 2014, Ericsson AB + * Copyright (c) 2004-2007, 2010-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_PORT_H +#define _TIPC_PORT_H + +#include "ref.h" +#include "net.h" +#include "msg.h" +#include "node_subscr.h" + +#define TIPC_CONNACK_INTV 256 +#define TIPC_FLOWCTRL_WIN (TIPC_CONNACK_INTV * 2) +#define TIPC_CONN_OVERLOAD_LIMIT ((TIPC_FLOWCTRL_WIN * 2 + 1) * \ + SKB_TRUESIZE(TIPC_MAX_USER_MSG_SIZE)) + +/** + * struct tipc_port - TIPC port structure + * @lock: pointer to spinlock for controlling access to port + * @connected: non-zero if port is currently connected to a peer port + * @conn_type: TIPC type used when connection was established + * @conn_instance: TIPC instance used when connection was established + * @conn_unacked: number of unacknowledged messages received from peer port + * @published: non-zero if port has one or more associated names + * @congested: non-zero if cannot send because of link or port congestion + * @max_pkt: maximum packet size "hint" used when building messages sent by port + * @ref: unique reference to port in TIPC object registry + * @phdr: preformatted message header used when sending messages + * @port_list: adjacent ports in TIPC's global list of ports + * @wait_list: adjacent ports in list of ports waiting on link congestion + * @waiting_pkts: + * @sent: # of non-empty messages sent by port + * @acked: # of non-empty message acknowledgements from connected port's peer + * @publications: list of publications for port + * @pub_count: total # of publications port has made during its lifetime + * @probing_state: + * @probing_interval: + * @timer_ref: + * @subscription: "node down" subscription used to terminate failed connections + */ +struct tipc_port { + spinlock_t *lock; + int connected; + u32 conn_type; + u32 conn_instance; + u32 conn_unacked; + int published; + u32 congested; + u32 max_pkt; + u32 ref; + struct tipc_msg phdr; + struct list_head port_list; + struct list_head wait_list; + u32 waiting_pkts; + u32 sent; + u32 acked; + struct list_head publications; + u32 pub_count; + u32 probing_state; + u32 probing_interval; + struct timer_list timer; + struct tipc_node_subscr subscription; +}; + +extern spinlock_t tipc_port_list_lock; +struct tipc_port_list; + +/* + * TIPC port manipulation routines + */ +u32 tipc_port_init(struct tipc_port *p_ptr, + const unsigned int importance); + +int tipc_reject_msg(struct sk_buff *buf, u32 err); + +void tipc_acknowledge(u32 port_ref, u32 ack); + +void tipc_port_destroy(struct tipc_port *p_ptr); + +int tipc_publish(struct tipc_port *p_ptr, unsigned int scope, + struct tipc_name_seq const *name_seq); + +int tipc_withdraw(struct tipc_port *p_ptr, unsigned int scope, + struct tipc_name_seq const *name_seq); + +int tipc_port_connect(u32 portref, struct tipc_portid const *port); + +int tipc_port_disconnect(u32 portref); + +int tipc_port_shutdown(u32 ref); + +void tipc_port_wakeup(struct tipc_port *port); + +/* + * The following routines require that the port be locked on entry + */ +int __tipc_port_disconnect(struct tipc_port *tp_ptr); +int __tipc_port_connect(u32 ref, struct tipc_port *p_ptr, + struct tipc_portid const *peer); +int tipc_port_peer_msg(struct tipc_port *p_ptr, struct tipc_msg *msg); + +/* + * TIPC messaging routines + */ + +int tipc_send(struct tipc_port *port, + struct iovec const *msg_sect, + unsigned int len); + +int tipc_send2name(struct tipc_port *port, + struct tipc_name const *name, + u32 domain, + struct iovec const *msg_sect, + unsigned int len); + +int tipc_send2port(struct tipc_port *port, + struct tipc_portid const *dest, + struct iovec const *msg_sect, + unsigned int len); + +int tipc_port_mcast_xmit(struct tipc_port *port, + struct tipc_name_seq const *seq, + struct iovec const *msg, + unsigned int len); + +int tipc_port_iovec_reject(struct tipc_port *p_ptr, + struct tipc_msg *hdr, + struct iovec const *msg_sect, + unsigned int len, + int err); + +struct sk_buff *tipc_port_get_ports(void); +void tipc_port_proto_rcv(struct sk_buff *buf); +void tipc_port_mcast_rcv(struct sk_buff *buf, struct tipc_port_list *dp); +void tipc_port_reinit(void); + +/** + * tipc_port_lock - lock port instance referred to and return its pointer + */ +static inline struct tipc_port *tipc_port_lock(u32 ref) +{ + return (struct tipc_port *)tipc_ref_lock(ref); +} + +/** + * tipc_port_unlock - unlock a port instance + * + * Can use pointer instead of tipc_ref_unlock() since port is already locked. + */ +static inline void tipc_port_unlock(struct tipc_port *p_ptr) +{ + spin_unlock_bh(p_ptr->lock); +} + +static inline int tipc_port_congested(struct tipc_port *p_ptr) +{ + return ((p_ptr->sent - p_ptr->acked) >= TIPC_FLOWCTRL_WIN); +} + + +static inline u32 tipc_port_peernode(struct tipc_port *p_ptr) +{ + return msg_destnode(&p_ptr->phdr); +} + +static inline u32 tipc_port_peerport(struct tipc_port *p_ptr) +{ + return msg_destport(&p_ptr->phdr); +} + +static inline bool tipc_port_unreliable(struct tipc_port *port) +{ + return msg_src_droppable(&port->phdr) != 0; +} + +static inline void tipc_port_set_unreliable(struct tipc_port *port, + bool unreliable) +{ + msg_set_src_droppable(&port->phdr, unreliable ? 1 : 0); +} + +static inline bool tipc_port_unreturnable(struct tipc_port *port) +{ + return msg_dest_droppable(&port->phdr) != 0; +} + +static inline void tipc_port_set_unreturnable(struct tipc_port *port, + bool unreturnable) +{ + msg_set_dest_droppable(&port->phdr, unreturnable ? 1 : 0); +} + + +static inline int tipc_port_importance(struct tipc_port *port) +{ + return msg_importance(&port->phdr); +} + +static inline void tipc_port_set_importance(struct tipc_port *port, int imp) +{ + msg_set_importance(&port->phdr, (u32)imp); +} + +#endif diff --git a/net/tipc/ref.c b/net/tipc/ref.c new file mode 100644 index 00000000000..3d4ecd754ee --- /dev/null +++ b/net/tipc/ref.c @@ -0,0 +1,266 @@ +/* + * net/tipc/ref.c: TIPC object registry code + * + * Copyright (c) 1991-2006, Ericsson AB + * Copyright (c) 2004-2007, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "ref.h" + +/** + * struct reference - TIPC object reference entry + * @object: pointer to object associated with reference entry + * @lock: spinlock controlling access to object + * @ref: reference value for object (combines instance & array index info) + */ +struct reference { + void *object; + spinlock_t lock; + u32 ref; +}; + +/** + * struct tipc_ref_table - table of TIPC object reference entries + * @entries: pointer to array of reference entries + * @capacity: array index of first unusable entry + * @init_point: array index of first uninitialized entry + * @first_free: array index of first unused object reference entry + * @last_free: array index of last unused object reference entry + * @index_mask: bitmask for array index portion of reference values + * @start_mask: initial value for instance value portion of reference values + */ +struct ref_table { + struct reference *entries; + u32 capacity; + u32 init_point; + u32 first_free; + u32 last_free; + u32 index_mask; + u32 start_mask; +}; + +/* + * Object reference table consists of 2**N entries. + * + * State Object ptr Reference + * ----- ---------- --------- + * In use non-NULL XXXX|own index + * (XXXX changes each time entry is acquired) + * Free NULL YYYY|next free index + * (YYYY is one more than last used XXXX) + * Uninitialized NULL 0 + * + * Entry 0 is not used; this allows index 0 to denote the end of the free list. + * + * Note that a reference value of 0 does not necessarily indicate that an + * entry is uninitialized, since the last entry in the free list could also + * have a reference value of 0 (although this is unlikely). + */ + +static struct ref_table tipc_ref_table; + +static DEFINE_SPINLOCK(ref_table_lock); + +/** + * tipc_ref_table_init - create reference table for objects + */ +int tipc_ref_table_init(u32 requested_size, u32 start) +{ + struct reference *table; + u32 actual_size; + + /* account for unused entry, then round up size to a power of 2 */ + + requested_size++; + for (actual_size = 16; actual_size < requested_size; actual_size <<= 1) + /* do nothing */ ; + + /* allocate table & mark all entries as uninitialized */ + table = vzalloc(actual_size * sizeof(struct reference)); + if (table == NULL) + return -ENOMEM; + + tipc_ref_table.entries = table; + tipc_ref_table.capacity = requested_size; + tipc_ref_table.init_point = 1; + tipc_ref_table.first_free = 0; + tipc_ref_table.last_free = 0; + tipc_ref_table.index_mask = actual_size - 1; + tipc_ref_table.start_mask = start & ~tipc_ref_table.index_mask; + + return 0; +} + +/** + * tipc_ref_table_stop - destroy reference table for objects + */ +void tipc_ref_table_stop(void) +{ + vfree(tipc_ref_table.entries); + tipc_ref_table.entries = NULL; +} + +/** + * tipc_ref_acquire - create reference to an object + * + * Register an object pointer in reference table and lock the object. + * Returns a unique reference value that is used from then on to retrieve the + * object pointer, or to determine that the object has been deregistered. + * + * Note: The object is returned in the locked state so that the caller can + * register a partially initialized object, without running the risk that + * the object will be accessed before initialization is complete. + */ +u32 tipc_ref_acquire(void *object, spinlock_t **lock) +{ + u32 index; + u32 index_mask; + u32 next_plus_upper; + u32 ref; + struct reference *entry = NULL; + + if (!object) { + pr_err("Attempt to acquire ref. to non-existent obj\n"); + return 0; + } + if (!tipc_ref_table.entries) { + pr_err("Ref. table not found in acquisition attempt\n"); + return 0; + } + + /* take a free entry, if available; otherwise initialize a new entry */ + spin_lock_bh(&ref_table_lock); + if (tipc_ref_table.first_free) { + index = tipc_ref_table.first_free; + entry = &(tipc_ref_table.entries[index]); + index_mask = tipc_ref_table.index_mask; + next_plus_upper = entry->ref; + tipc_ref_table.first_free = next_plus_upper & index_mask; + ref = (next_plus_upper & ~index_mask) + index; + } else if (tipc_ref_table.init_point < tipc_ref_table.capacity) { + index = tipc_ref_table.init_point++; + entry = &(tipc_ref_table.entries[index]); + spin_lock_init(&entry->lock); + ref = tipc_ref_table.start_mask + index; + } else { + ref = 0; + } + spin_unlock_bh(&ref_table_lock); + + /* + * Grab the lock so no one else can modify this entry + * While we assign its ref value & object pointer + */ + if (entry) { + spin_lock_bh(&entry->lock); + entry->ref = ref; + entry->object = object; + *lock = &entry->lock; + /* + * keep it locked, the caller is responsible + * for unlocking this when they're done with it + */ + } + + return ref; +} + +/** + * tipc_ref_discard - invalidate references to an object + * + * Disallow future references to an object and free up the entry for re-use. + * Note: The entry's spin_lock may still be busy after discard + */ +void tipc_ref_discard(u32 ref) +{ + struct reference *entry; + u32 index; + u32 index_mask; + + if (!tipc_ref_table.entries) { + pr_err("Ref. table not found during discard attempt\n"); + return; + } + + index_mask = tipc_ref_table.index_mask; + index = ref & index_mask; + entry = &(tipc_ref_table.entries[index]); + + spin_lock_bh(&ref_table_lock); + + if (!entry->object) { + pr_err("Attempt to discard ref. to non-existent obj\n"); + goto exit; + } + if (entry->ref != ref) { + pr_err("Attempt to discard non-existent reference\n"); + goto exit; + } + + /* + * mark entry as unused; increment instance part of entry's reference + * to invalidate any subsequent references + */ + entry->object = NULL; + entry->ref = (ref & ~index_mask) + (index_mask + 1); + + /* append entry to free entry list */ + if (tipc_ref_table.first_free == 0) + tipc_ref_table.first_free = index; + else + tipc_ref_table.entries[tipc_ref_table.last_free].ref |= index; + tipc_ref_table.last_free = index; + +exit: + spin_unlock_bh(&ref_table_lock); +} + +/** + * tipc_ref_lock - lock referenced object and return pointer to it + */ +void *tipc_ref_lock(u32 ref) +{ + if (likely(tipc_ref_table.entries)) { + struct reference *entry; + + entry = &tipc_ref_table.entries[ref & + tipc_ref_table.index_mask]; + if (likely(entry->ref != 0)) { + spin_lock_bh(&entry->lock); + if (likely((entry->ref == ref) && (entry->object))) + return entry->object; + spin_unlock_bh(&entry->lock); + } + } + return NULL; +} diff --git a/net/tipc/ref.h b/net/tipc/ref.h new file mode 100644 index 00000000000..d01aa1df63b --- /dev/null +++ b/net/tipc/ref.h @@ -0,0 +1,48 @@ +/* + * net/tipc/ref.h: Include file for TIPC object registry code + * + * Copyright (c) 1991-2006, Ericsson AB + * Copyright (c) 2005-2006, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_REF_H +#define _TIPC_REF_H + +int tipc_ref_table_init(u32 requested_size, u32 start); +void tipc_ref_table_stop(void); + +u32 tipc_ref_acquire(void *object, spinlock_t **lock); +void tipc_ref_discard(u32 ref); + +void *tipc_ref_lock(u32 ref); + +#endif diff --git a/net/tipc/server.c b/net/tipc/server.c new file mode 100644 index 00000000000..a538a02f869 --- /dev/null +++ b/net/tipc/server.c @@ -0,0 +1,600 @@ +/* + * net/tipc/server.c: TIPC server infrastructure + * + * Copyright (c) 2012-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "server.h" +#include "core.h" +#include <net/sock.h> + +/* Number of messages to send before rescheduling */ +#define MAX_SEND_MSG_COUNT 25 +#define MAX_RECV_MSG_COUNT 25 +#define CF_CONNECTED 1 + +#define sock2con(x) ((struct tipc_conn *)(x)->sk_user_data) + +/** + * struct tipc_conn - TIPC connection structure + * @kref: reference counter to connection object + * @conid: connection identifier + * @sock: socket handler associated with connection + * @flags: indicates connection state + * @server: pointer to connected server + * @rwork: receive work item + * @usr_data: user-specified field + * @rx_action: what to do when connection socket is active + * @outqueue: pointer to first outbound message in queue + * @outqueue_lock: control access to the outqueue + * @outqueue: list of connection objects for its server + * @swork: send work item + */ +struct tipc_conn { + struct kref kref; + int conid; + struct socket *sock; + unsigned long flags; + struct tipc_server *server; + struct work_struct rwork; + int (*rx_action) (struct tipc_conn *con); + void *usr_data; + struct list_head outqueue; + spinlock_t outqueue_lock; + struct work_struct swork; +}; + +/* An entry waiting to be sent */ +struct outqueue_entry { + struct list_head list; + struct kvec iov; + struct sockaddr_tipc dest; +}; + +static void tipc_recv_work(struct work_struct *work); +static void tipc_send_work(struct work_struct *work); +static void tipc_clean_outqueues(struct tipc_conn *con); + +static void tipc_conn_kref_release(struct kref *kref) +{ + struct tipc_conn *con = container_of(kref, struct tipc_conn, kref); + + if (con->sock) { + tipc_sock_release_local(con->sock); + con->sock = NULL; + } + + tipc_clean_outqueues(con); + kfree(con); +} + +static void conn_put(struct tipc_conn *con) +{ + kref_put(&con->kref, tipc_conn_kref_release); +} + +static void conn_get(struct tipc_conn *con) +{ + kref_get(&con->kref); +} + +static struct tipc_conn *tipc_conn_lookup(struct tipc_server *s, int conid) +{ + struct tipc_conn *con; + + spin_lock_bh(&s->idr_lock); + con = idr_find(&s->conn_idr, conid); + if (con) + conn_get(con); + spin_unlock_bh(&s->idr_lock); + return con; +} + +static void sock_data_ready(struct sock *sk) +{ + struct tipc_conn *con; + + read_lock(&sk->sk_callback_lock); + con = sock2con(sk); + if (con && test_bit(CF_CONNECTED, &con->flags)) { + conn_get(con); + if (!queue_work(con->server->rcv_wq, &con->rwork)) + conn_put(con); + } + read_unlock(&sk->sk_callback_lock); +} + +static void sock_write_space(struct sock *sk) +{ + struct tipc_conn *con; + + read_lock(&sk->sk_callback_lock); + con = sock2con(sk); + if (con && test_bit(CF_CONNECTED, &con->flags)) { + conn_get(con); + if (!queue_work(con->server->send_wq, &con->swork)) + conn_put(con); + } + read_unlock(&sk->sk_callback_lock); +} + +static void tipc_register_callbacks(struct socket *sock, struct tipc_conn *con) +{ + struct sock *sk = sock->sk; + + write_lock_bh(&sk->sk_callback_lock); + + sk->sk_data_ready = sock_data_ready; + sk->sk_write_space = sock_write_space; + sk->sk_user_data = con; + + con->sock = sock; + + write_unlock_bh(&sk->sk_callback_lock); +} + +static void tipc_unregister_callbacks(struct tipc_conn *con) +{ + struct sock *sk = con->sock->sk; + + write_lock_bh(&sk->sk_callback_lock); + sk->sk_user_data = NULL; + write_unlock_bh(&sk->sk_callback_lock); +} + +static void tipc_close_conn(struct tipc_conn *con) +{ + struct tipc_server *s = con->server; + + if (test_and_clear_bit(CF_CONNECTED, &con->flags)) { + if (con->conid) + s->tipc_conn_shutdown(con->conid, con->usr_data); + + spin_lock_bh(&s->idr_lock); + idr_remove(&s->conn_idr, con->conid); + s->idr_in_use--; + spin_unlock_bh(&s->idr_lock); + + tipc_unregister_callbacks(con); + + /* We shouldn't flush pending works as we may be in the + * thread. In fact the races with pending rx/tx work structs + * are harmless for us here as we have already deleted this + * connection from server connection list and set + * sk->sk_user_data to 0 before releasing connection object. + */ + kernel_sock_shutdown(con->sock, SHUT_RDWR); + + conn_put(con); + } +} + +static struct tipc_conn *tipc_alloc_conn(struct tipc_server *s) +{ + struct tipc_conn *con; + int ret; + + con = kzalloc(sizeof(struct tipc_conn), GFP_ATOMIC); + if (!con) + return ERR_PTR(-ENOMEM); + + kref_init(&con->kref); + INIT_LIST_HEAD(&con->outqueue); + spin_lock_init(&con->outqueue_lock); + INIT_WORK(&con->swork, tipc_send_work); + INIT_WORK(&con->rwork, tipc_recv_work); + + spin_lock_bh(&s->idr_lock); + ret = idr_alloc(&s->conn_idr, con, 0, 0, GFP_ATOMIC); + if (ret < 0) { + kfree(con); + spin_unlock_bh(&s->idr_lock); + return ERR_PTR(-ENOMEM); + } + con->conid = ret; + s->idr_in_use++; + spin_unlock_bh(&s->idr_lock); + + set_bit(CF_CONNECTED, &con->flags); + con->server = s; + + return con; +} + +static int tipc_receive_from_sock(struct tipc_conn *con) +{ + struct msghdr msg = {}; + struct tipc_server *s = con->server; + struct sockaddr_tipc addr; + struct kvec iov; + void *buf; + int ret; + + buf = kmem_cache_alloc(s->rcvbuf_cache, GFP_ATOMIC); + if (!buf) { + ret = -ENOMEM; + goto out_close; + } + + iov.iov_base = buf; + iov.iov_len = s->max_rcvbuf_size; + msg.msg_name = &addr; + ret = kernel_recvmsg(con->sock, &msg, &iov, 1, iov.iov_len, + MSG_DONTWAIT); + if (ret <= 0) { + kmem_cache_free(s->rcvbuf_cache, buf); + goto out_close; + } + + s->tipc_conn_recvmsg(con->conid, &addr, con->usr_data, buf, ret); + + kmem_cache_free(s->rcvbuf_cache, buf); + + return 0; + +out_close: + if (ret != -EWOULDBLOCK) + tipc_close_conn(con); + else if (ret == 0) + /* Don't return success if we really got EOF */ + ret = -EAGAIN; + + return ret; +} + +static int tipc_accept_from_sock(struct tipc_conn *con) +{ + struct tipc_server *s = con->server; + struct socket *sock = con->sock; + struct socket *newsock; + struct tipc_conn *newcon; + int ret; + + ret = tipc_sock_accept_local(sock, &newsock, O_NONBLOCK); + if (ret < 0) + return ret; + + newcon = tipc_alloc_conn(con->server); + if (IS_ERR(newcon)) { + ret = PTR_ERR(newcon); + sock_release(newsock); + return ret; + } + + newcon->rx_action = tipc_receive_from_sock; + tipc_register_callbacks(newsock, newcon); + + /* Notify that new connection is incoming */ + newcon->usr_data = s->tipc_conn_new(newcon->conid); + + /* Wake up receive process in case of 'SYN+' message */ + newsock->sk->sk_data_ready(newsock->sk); + return ret; +} + +static struct socket *tipc_create_listen_sock(struct tipc_conn *con) +{ + struct tipc_server *s = con->server; + struct socket *sock = NULL; + int ret; + + ret = tipc_sock_create_local(s->type, &sock); + if (ret < 0) + return NULL; + ret = kernel_setsockopt(sock, SOL_TIPC, TIPC_IMPORTANCE, + (char *)&s->imp, sizeof(s->imp)); + if (ret < 0) + goto create_err; + ret = kernel_bind(sock, (struct sockaddr *)s->saddr, sizeof(*s->saddr)); + if (ret < 0) + goto create_err; + + switch (s->type) { + case SOCK_STREAM: + case SOCK_SEQPACKET: + con->rx_action = tipc_accept_from_sock; + + ret = kernel_listen(sock, 0); + if (ret < 0) + goto create_err; + break; + case SOCK_DGRAM: + case SOCK_RDM: + con->rx_action = tipc_receive_from_sock; + break; + default: + pr_err("Unknown socket type %d\n", s->type); + goto create_err; + } + return sock; + +create_err: + sock_release(sock); + con->sock = NULL; + return NULL; +} + +static int tipc_open_listening_sock(struct tipc_server *s) +{ + struct socket *sock; + struct tipc_conn *con; + + con = tipc_alloc_conn(s); + if (IS_ERR(con)) + return PTR_ERR(con); + + sock = tipc_create_listen_sock(con); + if (!sock) { + idr_remove(&s->conn_idr, con->conid); + s->idr_in_use--; + kfree(con); + return -EINVAL; + } + + tipc_register_callbacks(sock, con); + return 0; +} + +static struct outqueue_entry *tipc_alloc_entry(void *data, int len) +{ + struct outqueue_entry *entry; + void *buf; + + entry = kmalloc(sizeof(struct outqueue_entry), GFP_ATOMIC); + if (!entry) + return NULL; + + buf = kmalloc(len, GFP_ATOMIC); + if (!buf) { + kfree(entry); + return NULL; + } + + memcpy(buf, data, len); + entry->iov.iov_base = buf; + entry->iov.iov_len = len; + + return entry; +} + +static void tipc_free_entry(struct outqueue_entry *e) +{ + kfree(e->iov.iov_base); + kfree(e); +} + +static void tipc_clean_outqueues(struct tipc_conn *con) +{ + struct outqueue_entry *e, *safe; + + spin_lock_bh(&con->outqueue_lock); + list_for_each_entry_safe(e, safe, &con->outqueue, list) { + list_del(&e->list); + tipc_free_entry(e); + } + spin_unlock_bh(&con->outqueue_lock); +} + +int tipc_conn_sendmsg(struct tipc_server *s, int conid, + struct sockaddr_tipc *addr, void *data, size_t len) +{ + struct outqueue_entry *e; + struct tipc_conn *con; + + con = tipc_conn_lookup(s, conid); + if (!con) + return -EINVAL; + + e = tipc_alloc_entry(data, len); + if (!e) { + conn_put(con); + return -ENOMEM; + } + + if (addr) + memcpy(&e->dest, addr, sizeof(struct sockaddr_tipc)); + + spin_lock_bh(&con->outqueue_lock); + list_add_tail(&e->list, &con->outqueue); + spin_unlock_bh(&con->outqueue_lock); + + if (test_bit(CF_CONNECTED, &con->flags)) { + if (!queue_work(s->send_wq, &con->swork)) + conn_put(con); + } else { + conn_put(con); + } + return 0; +} + +void tipc_conn_terminate(struct tipc_server *s, int conid) +{ + struct tipc_conn *con; + + con = tipc_conn_lookup(s, conid); + if (con) { + tipc_close_conn(con); + conn_put(con); + } +} + +static void tipc_send_to_sock(struct tipc_conn *con) +{ + int count = 0; + struct tipc_server *s = con->server; + struct outqueue_entry *e; + struct msghdr msg; + int ret; + + spin_lock_bh(&con->outqueue_lock); + while (1) { + e = list_entry(con->outqueue.next, struct outqueue_entry, + list); + if ((struct list_head *) e == &con->outqueue) + break; + spin_unlock_bh(&con->outqueue_lock); + + memset(&msg, 0, sizeof(msg)); + msg.msg_flags = MSG_DONTWAIT; + + if (s->type == SOCK_DGRAM || s->type == SOCK_RDM) { + msg.msg_name = &e->dest; + msg.msg_namelen = sizeof(struct sockaddr_tipc); + } + ret = kernel_sendmsg(con->sock, &msg, &e->iov, 1, + e->iov.iov_len); + if (ret == -EWOULDBLOCK || ret == 0) { + cond_resched(); + goto out; + } else if (ret < 0) { + goto send_err; + } + + /* Don't starve users filling buffers */ + if (++count >= MAX_SEND_MSG_COUNT) { + cond_resched(); + count = 0; + } + + spin_lock_bh(&con->outqueue_lock); + list_del(&e->list); + tipc_free_entry(e); + } + spin_unlock_bh(&con->outqueue_lock); +out: + return; + +send_err: + tipc_close_conn(con); +} + +static void tipc_recv_work(struct work_struct *work) +{ + struct tipc_conn *con = container_of(work, struct tipc_conn, rwork); + int count = 0; + + while (test_bit(CF_CONNECTED, &con->flags)) { + if (con->rx_action(con)) + break; + + /* Don't flood Rx machine */ + if (++count >= MAX_RECV_MSG_COUNT) { + cond_resched(); + count = 0; + } + } + conn_put(con); +} + +static void tipc_send_work(struct work_struct *work) +{ + struct tipc_conn *con = container_of(work, struct tipc_conn, swork); + + if (test_bit(CF_CONNECTED, &con->flags)) + tipc_send_to_sock(con); + + conn_put(con); +} + +static void tipc_work_stop(struct tipc_server *s) +{ + destroy_workqueue(s->rcv_wq); + destroy_workqueue(s->send_wq); +} + +static int tipc_work_start(struct tipc_server *s) +{ + s->rcv_wq = alloc_workqueue("tipc_rcv", WQ_UNBOUND, 1); + if (!s->rcv_wq) { + pr_err("can't start tipc receive workqueue\n"); + return -ENOMEM; + } + + s->send_wq = alloc_workqueue("tipc_send", WQ_UNBOUND, 1); + if (!s->send_wq) { + pr_err("can't start tipc send workqueue\n"); + destroy_workqueue(s->rcv_wq); + return -ENOMEM; + } + + return 0; +} + +int tipc_server_start(struct tipc_server *s) +{ + int ret; + + spin_lock_init(&s->idr_lock); + idr_init(&s->conn_idr); + s->idr_in_use = 0; + + s->rcvbuf_cache = kmem_cache_create(s->name, s->max_rcvbuf_size, + 0, SLAB_HWCACHE_ALIGN, NULL); + if (!s->rcvbuf_cache) + return -ENOMEM; + + ret = tipc_work_start(s); + if (ret < 0) { + kmem_cache_destroy(s->rcvbuf_cache); + return ret; + } + ret = tipc_open_listening_sock(s); + if (ret < 0) { + tipc_work_stop(s); + kmem_cache_destroy(s->rcvbuf_cache); + return ret; + } + return ret; +} + +void tipc_server_stop(struct tipc_server *s) +{ + struct tipc_conn *con; + int total = 0; + int id; + + spin_lock_bh(&s->idr_lock); + for (id = 0; total < s->idr_in_use; id++) { + con = idr_find(&s->conn_idr, id); + if (con) { + total++; + spin_unlock_bh(&s->idr_lock); + tipc_close_conn(con); + spin_lock_bh(&s->idr_lock); + } + } + spin_unlock_bh(&s->idr_lock); + + tipc_work_stop(s); + kmem_cache_destroy(s->rcvbuf_cache); + idr_destroy(&s->conn_idr); +} diff --git a/net/tipc/server.h b/net/tipc/server.h new file mode 100644 index 00000000000..be817b0b547 --- /dev/null +++ b/net/tipc/server.h @@ -0,0 +1,92 @@ +/* + * net/tipc/server.h: Include file for TIPC server code + * + * Copyright (c) 2012-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_SERVER_H +#define _TIPC_SERVER_H + +#include "core.h" + +#define TIPC_SERVER_NAME_LEN 32 + +/** + * struct tipc_server - TIPC server structure + * @conn_idr: identifier set of connection + * @idr_lock: protect the connection identifier set + * @idr_in_use: amount of allocated identifier entry + * @rcvbuf_cache: memory cache of server receive buffer + * @rcv_wq: receive workqueue + * @send_wq: send workqueue + * @max_rcvbuf_size: maximum permitted receive message length + * @tipc_conn_new: callback will be called when new connection is incoming + * @tipc_conn_shutdown: callback will be called when connection is shut down + * @tipc_conn_recvmsg: callback will be called when message arrives + * @saddr: TIPC server address + * @name: server name + * @imp: message importance + * @type: socket type + */ +struct tipc_server { + struct idr conn_idr; + spinlock_t idr_lock; + int idr_in_use; + struct kmem_cache *rcvbuf_cache; + struct workqueue_struct *rcv_wq; + struct workqueue_struct *send_wq; + int max_rcvbuf_size; + void *(*tipc_conn_new) (int conid); + void (*tipc_conn_shutdown) (int conid, void *usr_data); + void (*tipc_conn_recvmsg) (int conid, struct sockaddr_tipc *addr, + void *usr_data, void *buf, size_t len); + struct sockaddr_tipc *saddr; + const char name[TIPC_SERVER_NAME_LEN]; + int imp; + int type; +}; + +int tipc_conn_sendmsg(struct tipc_server *s, int conid, + struct sockaddr_tipc *addr, void *data, size_t len); + +/** + * tipc_conn_terminate - terminate connection with server + * + * Note: Must call it in process context since it might sleep + */ +void tipc_conn_terminate(struct tipc_server *s, int conid); + +int tipc_server_start(struct tipc_server *s); + +void tipc_server_stop(struct tipc_server *s); + +#endif diff --git a/net/tipc/socket.c b/net/tipc/socket.c new file mode 100644 index 00000000000..ef0475568f9 --- /dev/null +++ b/net/tipc/socket.c @@ -0,0 +1,2077 @@ +/* + * net/tipc/socket.c: TIPC socket API + * + * Copyright (c) 2001-2007, 2012-2014, Ericsson AB + * Copyright (c) 2004-2008, 2010-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "port.h" +#include "node.h" + +#include <linux/export.h> + +#define SS_LISTENING -1 /* socket is listening */ +#define SS_READY -2 /* socket is connectionless */ + +#define CONN_TIMEOUT_DEFAULT 8000 /* default connect timeout = 8s */ + +static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *skb); +static void tipc_data_ready(struct sock *sk); +static void tipc_write_space(struct sock *sk); +static int tipc_release(struct socket *sock); +static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags); + +static const struct proto_ops packet_ops; +static const struct proto_ops stream_ops; +static const struct proto_ops msg_ops; + +static struct proto tipc_proto; +static struct proto tipc_proto_kern; + +/* + * Revised TIPC socket locking policy: + * + * Most socket operations take the standard socket lock when they start + * and hold it until they finish (or until they need to sleep). Acquiring + * this lock grants the owner exclusive access to the fields of the socket + * data structures, with the exception of the backlog queue. A few socket + * operations can be done without taking the socket lock because they only + * read socket information that never changes during the life of the socket. + * + * Socket operations may acquire the lock for the associated TIPC port if they + * need to perform an operation on the port. If any routine needs to acquire + * both the socket lock and the port lock it must take the socket lock first + * to avoid the risk of deadlock. + * + * The dispatcher handling incoming messages cannot grab the socket lock in + * the standard fashion, since invoked it runs at the BH level and cannot block. + * Instead, it checks to see if the socket lock is currently owned by someone, + * and either handles the message itself or adds it to the socket's backlog + * queue; in the latter case the queued message is processed once the process + * owning the socket lock releases it. + * + * NOTE: Releasing the socket lock while an operation is sleeping overcomes + * the problem of a blocked socket operation preventing any other operations + * from occurring. However, applications must be careful if they have + * multiple threads trying to send (or receive) on the same socket, as these + * operations might interfere with each other. For example, doing a connect + * and a receive at the same time might allow the receive to consume the + * ACK message meant for the connect. While additional work could be done + * to try and overcome this, it doesn't seem to be worthwhile at the present. + * + * NOTE: Releasing the socket lock while an operation is sleeping also ensures + * that another operation that must be performed in a non-blocking manner is + * not delayed for very long because the lock has already been taken. + * + * NOTE: This code assumes that certain fields of a port/socket pair are + * constant over its lifetime; such fields can be examined without taking + * the socket lock and/or port lock, and do not need to be re-read even + * after resuming processing after waiting. These fields include: + * - socket type + * - pointer to socket sk structure (aka tipc_sock structure) + * - pointer to port structure + * - port reference + */ + +#include "socket.h" + +/** + * advance_rx_queue - discard first buffer in socket receive queue + * + * Caller must hold socket lock + */ +static void advance_rx_queue(struct sock *sk) +{ + kfree_skb(__skb_dequeue(&sk->sk_receive_queue)); +} + +/** + * reject_rx_queue - reject all buffers in socket receive queue + * + * Caller must hold socket lock + */ +static void reject_rx_queue(struct sock *sk) +{ + struct sk_buff *buf; + + while ((buf = __skb_dequeue(&sk->sk_receive_queue))) + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); +} + +/** + * tipc_sk_create - create a TIPC socket + * @net: network namespace (must be default network) + * @sock: pre-allocated socket structure + * @protocol: protocol indicator (must be 0) + * @kern: caused by kernel or by userspace? + * + * This routine creates additional data structures used by the TIPC socket, + * initializes them, and links them together. + * + * Returns 0 on success, errno otherwise + */ +static int tipc_sk_create(struct net *net, struct socket *sock, + int protocol, int kern) +{ + const struct proto_ops *ops; + socket_state state; + struct sock *sk; + struct tipc_sock *tsk; + struct tipc_port *port; + u32 ref; + + /* Validate arguments */ + if (unlikely(protocol != 0)) + return -EPROTONOSUPPORT; + + switch (sock->type) { + case SOCK_STREAM: + ops = &stream_ops; + state = SS_UNCONNECTED; + break; + case SOCK_SEQPACKET: + ops = &packet_ops; + state = SS_UNCONNECTED; + break; + case SOCK_DGRAM: + case SOCK_RDM: + ops = &msg_ops; + state = SS_READY; + break; + default: + return -EPROTOTYPE; + } + + /* Allocate socket's protocol area */ + if (!kern) + sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto); + else + sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto_kern); + + if (sk == NULL) + return -ENOMEM; + + tsk = tipc_sk(sk); + port = &tsk->port; + + ref = tipc_port_init(port, TIPC_LOW_IMPORTANCE); + if (!ref) { + pr_warn("Socket registration failed, ref. table exhausted\n"); + sk_free(sk); + return -ENOMEM; + } + + /* Finish initializing socket data structures */ + sock->ops = ops; + sock->state = state; + + sock_init_data(sock, sk); + sk->sk_backlog_rcv = tipc_backlog_rcv; + sk->sk_rcvbuf = sysctl_tipc_rmem[1]; + sk->sk_data_ready = tipc_data_ready; + sk->sk_write_space = tipc_write_space; + tsk->conn_timeout = CONN_TIMEOUT_DEFAULT; + atomic_set(&tsk->dupl_rcvcnt, 0); + tipc_port_unlock(port); + + if (sock->state == SS_READY) { + tipc_port_set_unreturnable(port, true); + if (sock->type == SOCK_DGRAM) + tipc_port_set_unreliable(port, true); + } + return 0; +} + +/** + * tipc_sock_create_local - create TIPC socket from inside TIPC module + * @type: socket type - SOCK_RDM or SOCK_SEQPACKET + * + * We cannot use sock_creat_kern here because it bumps module user count. + * Since socket owner and creator is the same module we must make sure + * that module count remains zero for module local sockets, otherwise + * we cannot do rmmod. + * + * Returns 0 on success, errno otherwise + */ +int tipc_sock_create_local(int type, struct socket **res) +{ + int rc; + + rc = sock_create_lite(AF_TIPC, type, 0, res); + if (rc < 0) { + pr_err("Failed to create kernel socket\n"); + return rc; + } + tipc_sk_create(&init_net, *res, 0, 1); + + return 0; +} + +/** + * tipc_sock_release_local - release socket created by tipc_sock_create_local + * @sock: the socket to be released. + * + * Module reference count is not incremented when such sockets are created, + * so we must keep it from being decremented when they are released. + */ +void tipc_sock_release_local(struct socket *sock) +{ + tipc_release(sock); + sock->ops = NULL; + sock_release(sock); +} + +/** + * tipc_sock_accept_local - accept a connection on a socket created + * with tipc_sock_create_local. Use this function to avoid that + * module reference count is inadvertently incremented. + * + * @sock: the accepting socket + * @newsock: reference to the new socket to be created + * @flags: socket flags + */ + +int tipc_sock_accept_local(struct socket *sock, struct socket **newsock, + int flags) +{ + struct sock *sk = sock->sk; + int ret; + + ret = sock_create_lite(sk->sk_family, sk->sk_type, + sk->sk_protocol, newsock); + if (ret < 0) + return ret; + + ret = tipc_accept(sock, *newsock, flags); + if (ret < 0) { + sock_release(*newsock); + return ret; + } + (*newsock)->ops = sock->ops; + return ret; +} + +/** + * tipc_release - destroy a TIPC socket + * @sock: socket to destroy + * + * This routine cleans up any messages that are still queued on the socket. + * For DGRAM and RDM socket types, all queued messages are rejected. + * For SEQPACKET and STREAM socket types, the first message is rejected + * and any others are discarded. (If the first message on a STREAM socket + * is partially-read, it is discarded and the next one is rejected instead.) + * + * NOTE: Rejected messages are not necessarily returned to the sender! They + * are returned or discarded according to the "destination droppable" setting + * specified for the message by the sender. + * + * Returns 0 on success, errno otherwise + */ +static int tipc_release(struct socket *sock) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk; + struct tipc_port *port; + struct sk_buff *buf; + + /* + * Exit if socket isn't fully initialized (occurs when a failed accept() + * releases a pre-allocated child socket that was never used) + */ + if (sk == NULL) + return 0; + + tsk = tipc_sk(sk); + port = &tsk->port; + lock_sock(sk); + + /* + * Reject all unreceived messages, except on an active connection + * (which disconnects locally & sends a 'FIN+' to peer) + */ + while (sock->state != SS_DISCONNECTING) { + buf = __skb_dequeue(&sk->sk_receive_queue); + if (buf == NULL) + break; + if (TIPC_SKB_CB(buf)->handle != NULL) + kfree_skb(buf); + else { + if ((sock->state == SS_CONNECTING) || + (sock->state == SS_CONNECTED)) { + sock->state = SS_DISCONNECTING; + tipc_port_disconnect(port->ref); + } + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + } + } + + /* Destroy TIPC port; also disconnects an active connection and + * sends a 'FIN-' to peer. + */ + tipc_port_destroy(port); + + /* Discard any remaining (connection-based) messages in receive queue */ + __skb_queue_purge(&sk->sk_receive_queue); + + /* Reject any messages that accumulated in backlog queue */ + sock->state = SS_DISCONNECTING; + release_sock(sk); + + sock_put(sk); + sock->sk = NULL; + + return 0; +} + +/** + * tipc_bind - associate or disassocate TIPC name(s) with a socket + * @sock: socket structure + * @uaddr: socket address describing name(s) and desired operation + * @uaddr_len: size of socket address data structure + * + * Name and name sequence binding is indicated using a positive scope value; + * a negative scope value unbinds the specified name. Specifying no name + * (i.e. a socket address length of 0) unbinds all names from the socket. + * + * Returns 0 on success, errno otherwise + * + * NOTE: This routine doesn't need to take the socket lock since it doesn't + * access any non-constant socket information. + */ +static int tipc_bind(struct socket *sock, struct sockaddr *uaddr, + int uaddr_len) +{ + struct sock *sk = sock->sk; + struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; + struct tipc_sock *tsk = tipc_sk(sk); + int res = -EINVAL; + + lock_sock(sk); + if (unlikely(!uaddr_len)) { + res = tipc_withdraw(&tsk->port, 0, NULL); + goto exit; + } + + if (uaddr_len < sizeof(struct sockaddr_tipc)) { + res = -EINVAL; + goto exit; + } + if (addr->family != AF_TIPC) { + res = -EAFNOSUPPORT; + goto exit; + } + + if (addr->addrtype == TIPC_ADDR_NAME) + addr->addr.nameseq.upper = addr->addr.nameseq.lower; + else if (addr->addrtype != TIPC_ADDR_NAMESEQ) { + res = -EAFNOSUPPORT; + goto exit; + } + + if ((addr->addr.nameseq.type < TIPC_RESERVED_TYPES) && + (addr->addr.nameseq.type != TIPC_TOP_SRV) && + (addr->addr.nameseq.type != TIPC_CFG_SRV)) { + res = -EACCES; + goto exit; + } + + res = (addr->scope > 0) ? + tipc_publish(&tsk->port, addr->scope, &addr->addr.nameseq) : + tipc_withdraw(&tsk->port, -addr->scope, &addr->addr.nameseq); +exit: + release_sock(sk); + return res; +} + +/** + * tipc_getname - get port ID of socket or peer socket + * @sock: socket structure + * @uaddr: area for returned socket address + * @uaddr_len: area for returned length of socket address + * @peer: 0 = own ID, 1 = current peer ID, 2 = current/former peer ID + * + * Returns 0 on success, errno otherwise + * + * NOTE: This routine doesn't need to take the socket lock since it only + * accesses socket information that is unchanging (or which changes in + * a completely predictable manner). + */ +static int tipc_getname(struct socket *sock, struct sockaddr *uaddr, + int *uaddr_len, int peer) +{ + struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; + struct tipc_sock *tsk = tipc_sk(sock->sk); + + memset(addr, 0, sizeof(*addr)); + if (peer) { + if ((sock->state != SS_CONNECTED) && + ((peer != 2) || (sock->state != SS_DISCONNECTING))) + return -ENOTCONN; + addr->addr.id.ref = tipc_port_peerport(&tsk->port); + addr->addr.id.node = tipc_port_peernode(&tsk->port); + } else { + addr->addr.id.ref = tsk->port.ref; + addr->addr.id.node = tipc_own_addr; + } + + *uaddr_len = sizeof(*addr); + addr->addrtype = TIPC_ADDR_ID; + addr->family = AF_TIPC; + addr->scope = 0; + addr->addr.name.domain = 0; + + return 0; +} + +/** + * tipc_poll - read and possibly block on pollmask + * @file: file structure associated with the socket + * @sock: socket for which to calculate the poll bits + * @wait: ??? + * + * Returns pollmask value + * + * COMMENTARY: + * It appears that the usual socket locking mechanisms are not useful here + * since the pollmask info is potentially out-of-date the moment this routine + * exits. TCP and other protocols seem to rely on higher level poll routines + * to handle any preventable race conditions, so TIPC will do the same ... + * + * TIPC sets the returned events as follows: + * + * socket state flags set + * ------------ --------- + * unconnected no read flags + * POLLOUT if port is not congested + * + * connecting POLLIN/POLLRDNORM if ACK/NACK in rx queue + * no write flags + * + * connected POLLIN/POLLRDNORM if data in rx queue + * POLLOUT if port is not congested + * + * disconnecting POLLIN/POLLRDNORM/POLLHUP + * no write flags + * + * listening POLLIN if SYN in rx queue + * no write flags + * + * ready POLLIN/POLLRDNORM if data in rx queue + * [connectionless] POLLOUT (since port cannot be congested) + * + * IMPORTANT: The fact that a read or write operation is indicated does NOT + * imply that the operation will succeed, merely that it should be performed + * and will not block. + */ +static unsigned int tipc_poll(struct file *file, struct socket *sock, + poll_table *wait) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + u32 mask = 0; + + sock_poll_wait(file, sk_sleep(sk), wait); + + switch ((int)sock->state) { + case SS_UNCONNECTED: + if (!tsk->port.congested) + mask |= POLLOUT; + break; + case SS_READY: + case SS_CONNECTED: + if (!tsk->port.congested) + mask |= POLLOUT; + /* fall thru' */ + case SS_CONNECTING: + case SS_LISTENING: + if (!skb_queue_empty(&sk->sk_receive_queue)) + mask |= (POLLIN | POLLRDNORM); + break; + case SS_DISCONNECTING: + mask = (POLLIN | POLLRDNORM | POLLHUP); + break; + } + + return mask; +} + +/** + * dest_name_check - verify user is permitted to send to specified port name + * @dest: destination address + * @m: descriptor for message to be sent + * + * Prevents restricted configuration commands from being issued by + * unauthorized users. + * + * Returns 0 if permission is granted, otherwise errno + */ +static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) +{ + struct tipc_cfg_msg_hdr hdr; + + if (likely(dest->addr.name.name.type >= TIPC_RESERVED_TYPES)) + return 0; + if (likely(dest->addr.name.name.type == TIPC_TOP_SRV)) + return 0; + if (likely(dest->addr.name.name.type != TIPC_CFG_SRV)) + return -EACCES; + + if (!m->msg_iovlen || (m->msg_iov[0].iov_len < sizeof(hdr))) + return -EMSGSIZE; + if (copy_from_user(&hdr, m->msg_iov[0].iov_base, sizeof(hdr))) + return -EFAULT; + if ((ntohs(hdr.tcm_type) & 0xC000) && (!capable(CAP_NET_ADMIN))) + return -EACCES; + + return 0; +} + +static int tipc_wait_for_sndmsg(struct socket *sock, long *timeo_p) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + DEFINE_WAIT(wait); + int done; + + do { + int err = sock_error(sk); + if (err) + return err; + if (sock->state == SS_DISCONNECTING) + return -EPIPE; + if (!*timeo_p) + return -EAGAIN; + if (signal_pending(current)) + return sock_intr_errno(*timeo_p); + + prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + done = sk_wait_event(sk, timeo_p, !tsk->port.congested); + finish_wait(sk_sleep(sk), &wait); + } while (!done); + return 0; +} + + +/** + * tipc_sendmsg - send message in connectionless manner + * @iocb: if NULL, indicates that socket lock is already held + * @sock: socket structure + * @m: message to send + * @total_len: length of message + * + * Message must have an destination specified explicitly. + * Used for SOCK_RDM and SOCK_DGRAM messages, + * and for 'SYN' messages on SOCK_SEQPACKET and SOCK_STREAM connections. + * (Note: 'SYN+' is prohibited on SOCK_STREAM.) + * + * Returns the number of bytes sent on success, or errno otherwise + */ +static int tipc_sendmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; + DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); + int needs_conn; + long timeo; + int res = -EINVAL; + + if (unlikely(!dest)) + return -EDESTADDRREQ; + if (unlikely((m->msg_namelen < sizeof(*dest)) || + (dest->family != AF_TIPC))) + return -EINVAL; + if (total_len > TIPC_MAX_USER_MSG_SIZE) + return -EMSGSIZE; + + if (iocb) + lock_sock(sk); + + needs_conn = (sock->state != SS_READY); + if (unlikely(needs_conn)) { + if (sock->state == SS_LISTENING) { + res = -EPIPE; + goto exit; + } + if (sock->state != SS_UNCONNECTED) { + res = -EISCONN; + goto exit; + } + if (tsk->port.published) { + res = -EOPNOTSUPP; + goto exit; + } + if (dest->addrtype == TIPC_ADDR_NAME) { + tsk->port.conn_type = dest->addr.name.name.type; + tsk->port.conn_instance = dest->addr.name.name.instance; + } + + /* Abort any pending connection attempts (very unlikely) */ + reject_rx_queue(sk); + } + + timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); + do { + if (dest->addrtype == TIPC_ADDR_NAME) { + res = dest_name_check(dest, m); + if (res) + break; + res = tipc_send2name(port, + &dest->addr.name.name, + dest->addr.name.domain, + m->msg_iov, + total_len); + } else if (dest->addrtype == TIPC_ADDR_ID) { + res = tipc_send2port(port, + &dest->addr.id, + m->msg_iov, + total_len); + } else if (dest->addrtype == TIPC_ADDR_MCAST) { + if (needs_conn) { + res = -EOPNOTSUPP; + break; + } + res = dest_name_check(dest, m); + if (res) + break; + res = tipc_port_mcast_xmit(port, + &dest->addr.nameseq, + m->msg_iov, + total_len); + } + if (likely(res != -ELINKCONG)) { + if (needs_conn && (res >= 0)) + sock->state = SS_CONNECTING; + break; + } + res = tipc_wait_for_sndmsg(sock, &timeo); + if (res) + break; + } while (1); + +exit: + if (iocb) + release_sock(sk); + return res; +} + +static int tipc_wait_for_sndpkt(struct socket *sock, long *timeo_p) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; + DEFINE_WAIT(wait); + int done; + + do { + int err = sock_error(sk); + if (err) + return err; + if (sock->state == SS_DISCONNECTING) + return -EPIPE; + else if (sock->state != SS_CONNECTED) + return -ENOTCONN; + if (!*timeo_p) + return -EAGAIN; + if (signal_pending(current)) + return sock_intr_errno(*timeo_p); + + prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + done = sk_wait_event(sk, timeo_p, + (!port->congested || !port->connected)); + finish_wait(sk_sleep(sk), &wait); + } while (!done); + return 0; +} + +/** + * tipc_send_packet - send a connection-oriented message + * @iocb: if NULL, indicates that socket lock is already held + * @sock: socket structure + * @m: message to send + * @total_len: length of message + * + * Used for SOCK_SEQPACKET messages and SOCK_STREAM data. + * + * Returns the number of bytes sent on success, or errno otherwise + */ +static int tipc_send_packet(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); + int res = -EINVAL; + long timeo; + + /* Handle implied connection establishment */ + if (unlikely(dest)) + return tipc_sendmsg(iocb, sock, m, total_len); + + if (total_len > TIPC_MAX_USER_MSG_SIZE) + return -EMSGSIZE; + + if (iocb) + lock_sock(sk); + + if (unlikely(sock->state != SS_CONNECTED)) { + if (sock->state == SS_DISCONNECTING) + res = -EPIPE; + else + res = -ENOTCONN; + goto exit; + } + + timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); + do { + res = tipc_send(&tsk->port, m->msg_iov, total_len); + if (likely(res != -ELINKCONG)) + break; + res = tipc_wait_for_sndpkt(sock, &timeo); + if (res) + break; + } while (1); +exit: + if (iocb) + release_sock(sk); + return res; +} + +/** + * tipc_send_stream - send stream-oriented data + * @iocb: (unused) + * @sock: socket structure + * @m: data to send + * @total_len: total length of data to be sent + * + * Used for SOCK_STREAM data. + * + * Returns the number of bytes sent on success (or partial success), + * or errno if no data sent + */ +static int tipc_send_stream(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct msghdr my_msg; + struct iovec my_iov; + struct iovec *curr_iov; + int curr_iovlen; + char __user *curr_start; + u32 hdr_size; + int curr_left; + int bytes_to_send; + int bytes_sent; + int res; + + lock_sock(sk); + + /* Handle special cases where there is no connection */ + if (unlikely(sock->state != SS_CONNECTED)) { + if (sock->state == SS_UNCONNECTED) + res = tipc_send_packet(NULL, sock, m, total_len); + else + res = sock->state == SS_DISCONNECTING ? -EPIPE : -ENOTCONN; + goto exit; + } + + if (unlikely(m->msg_name)) { + res = -EISCONN; + goto exit; + } + + if (total_len > (unsigned int)INT_MAX) { + res = -EMSGSIZE; + goto exit; + } + + /* + * Send each iovec entry using one or more messages + * + * Note: This algorithm is good for the most likely case + * (i.e. one large iovec entry), but could be improved to pass sets + * of small iovec entries into send_packet(). + */ + curr_iov = m->msg_iov; + curr_iovlen = m->msg_iovlen; + my_msg.msg_iov = &my_iov; + my_msg.msg_iovlen = 1; + my_msg.msg_flags = m->msg_flags; + my_msg.msg_name = NULL; + bytes_sent = 0; + + hdr_size = msg_hdr_sz(&tsk->port.phdr); + + while (curr_iovlen--) { + curr_start = curr_iov->iov_base; + curr_left = curr_iov->iov_len; + + while (curr_left) { + bytes_to_send = tsk->port.max_pkt - hdr_size; + if (bytes_to_send > TIPC_MAX_USER_MSG_SIZE) + bytes_to_send = TIPC_MAX_USER_MSG_SIZE; + if (curr_left < bytes_to_send) + bytes_to_send = curr_left; + my_iov.iov_base = curr_start; + my_iov.iov_len = bytes_to_send; + res = tipc_send_packet(NULL, sock, &my_msg, + bytes_to_send); + if (res < 0) { + if (bytes_sent) + res = bytes_sent; + goto exit; + } + curr_left -= bytes_to_send; + curr_start += bytes_to_send; + bytes_sent += bytes_to_send; + } + + curr_iov++; + } + res = bytes_sent; +exit: + release_sock(sk); + return res; +} + +/** + * auto_connect - complete connection setup to a remote port + * @tsk: tipc socket structure + * @msg: peer's response message + * + * Returns 0 on success, errno otherwise + */ +static int auto_connect(struct tipc_sock *tsk, struct tipc_msg *msg) +{ + struct tipc_port *port = &tsk->port; + struct socket *sock = tsk->sk.sk_socket; + struct tipc_portid peer; + + peer.ref = msg_origport(msg); + peer.node = msg_orignode(msg); + + __tipc_port_connect(port->ref, port, &peer); + + if (msg_importance(msg) > TIPC_CRITICAL_IMPORTANCE) + return -EINVAL; + msg_set_importance(&port->phdr, (u32)msg_importance(msg)); + sock->state = SS_CONNECTED; + return 0; +} + +/** + * set_orig_addr - capture sender's address for received message + * @m: descriptor for message info + * @msg: received message header + * + * Note: Address is not captured if not requested by receiver. + */ +static void set_orig_addr(struct msghdr *m, struct tipc_msg *msg) +{ + DECLARE_SOCKADDR(struct sockaddr_tipc *, addr, m->msg_name); + + if (addr) { + addr->family = AF_TIPC; + addr->addrtype = TIPC_ADDR_ID; + memset(&addr->addr, 0, sizeof(addr->addr)); + addr->addr.id.ref = msg_origport(msg); + addr->addr.id.node = msg_orignode(msg); + addr->addr.name.domain = 0; /* could leave uninitialized */ + addr->scope = 0; /* could leave uninitialized */ + m->msg_namelen = sizeof(struct sockaddr_tipc); + } +} + +/** + * anc_data_recv - optionally capture ancillary data for received message + * @m: descriptor for message info + * @msg: received message header + * @tport: TIPC port associated with message + * + * Note: Ancillary data is not captured if not requested by receiver. + * + * Returns 0 if successful, otherwise errno + */ +static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, + struct tipc_port *tport) +{ + u32 anc_data[3]; + u32 err; + u32 dest_type; + int has_name; + int res; + + if (likely(m->msg_controllen == 0)) + return 0; + + /* Optionally capture errored message object(s) */ + err = msg ? msg_errcode(msg) : 0; + if (unlikely(err)) { + anc_data[0] = err; + anc_data[1] = msg_data_sz(msg); + res = put_cmsg(m, SOL_TIPC, TIPC_ERRINFO, 8, anc_data); + if (res) + return res; + if (anc_data[1]) { + res = put_cmsg(m, SOL_TIPC, TIPC_RETDATA, anc_data[1], + msg_data(msg)); + if (res) + return res; + } + } + + /* Optionally capture message destination object */ + dest_type = msg ? msg_type(msg) : TIPC_DIRECT_MSG; + switch (dest_type) { + case TIPC_NAMED_MSG: + has_name = 1; + anc_data[0] = msg_nametype(msg); + anc_data[1] = msg_namelower(msg); + anc_data[2] = msg_namelower(msg); + break; + case TIPC_MCAST_MSG: + has_name = 1; + anc_data[0] = msg_nametype(msg); + anc_data[1] = msg_namelower(msg); + anc_data[2] = msg_nameupper(msg); + break; + case TIPC_CONN_MSG: + has_name = (tport->conn_type != 0); + anc_data[0] = tport->conn_type; + anc_data[1] = tport->conn_instance; + anc_data[2] = tport->conn_instance; + break; + default: + has_name = 0; + } + if (has_name) { + res = put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, anc_data); + if (res) + return res; + } + + return 0; +} + +static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) +{ + struct sock *sk = sock->sk; + DEFINE_WAIT(wait); + long timeo = *timeop; + int err; + + for (;;) { + prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + if (timeo && skb_queue_empty(&sk->sk_receive_queue)) { + if (sock->state == SS_DISCONNECTING) { + err = -ENOTCONN; + break; + } + release_sock(sk); + timeo = schedule_timeout(timeo); + lock_sock(sk); + } + err = 0; + if (!skb_queue_empty(&sk->sk_receive_queue)) + break; + err = sock_intr_errno(timeo); + if (signal_pending(current)) + break; + err = -EAGAIN; + if (!timeo) + break; + } + finish_wait(sk_sleep(sk), &wait); + *timeop = timeo; + return err; +} + +/** + * tipc_recvmsg - receive packet-oriented message + * @iocb: (unused) + * @m: descriptor for message info + * @buf_len: total size of user buffer area + * @flags: receive flags + * + * Used for SOCK_DGRAM, SOCK_RDM, and SOCK_SEQPACKET messages. + * If the complete message doesn't fit in user area, truncate it. + * + * Returns size of returned message data, errno otherwise + */ +static int tipc_recvmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t buf_len, int flags) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; + struct sk_buff *buf; + struct tipc_msg *msg; + long timeo; + unsigned int sz; + u32 err; + int res; + + /* Catch invalid receive requests */ + if (unlikely(!buf_len)) + return -EINVAL; + + lock_sock(sk); + + if (unlikely(sock->state == SS_UNCONNECTED)) { + res = -ENOTCONN; + goto exit; + } + + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); +restart: + + /* Look for a message in receive queue; wait if necessary */ + res = tipc_wait_for_rcvmsg(sock, &timeo); + if (res) + goto exit; + + /* Look at first message in receive queue */ + buf = skb_peek(&sk->sk_receive_queue); + msg = buf_msg(buf); + sz = msg_data_sz(msg); + err = msg_errcode(msg); + + /* Discard an empty non-errored message & try again */ + if ((!sz) && (!err)) { + advance_rx_queue(sk); + goto restart; + } + + /* Capture sender's address (optional) */ + set_orig_addr(m, msg); + + /* Capture ancillary data (optional) */ + res = anc_data_recv(m, msg, port); + if (res) + goto exit; + + /* Capture message data (if valid) & compute return value (always) */ + if (!err) { + if (unlikely(buf_len < sz)) { + sz = buf_len; + m->msg_flags |= MSG_TRUNC; + } + res = skb_copy_datagram_iovec(buf, msg_hdr_sz(msg), + m->msg_iov, sz); + if (res) + goto exit; + res = sz; + } else { + if ((sock->state == SS_READY) || + ((err == TIPC_CONN_SHUTDOWN) || m->msg_control)) + res = 0; + else + res = -ECONNRESET; + } + + /* Consume received message (optional) */ + if (likely(!(flags & MSG_PEEK))) { + if ((sock->state != SS_READY) && + (++port->conn_unacked >= TIPC_CONNACK_INTV)) + tipc_acknowledge(port->ref, port->conn_unacked); + advance_rx_queue(sk); + } +exit: + release_sock(sk); + return res; +} + +/** + * tipc_recv_stream - receive stream-oriented data + * @iocb: (unused) + * @m: descriptor for message info + * @buf_len: total size of user buffer area + * @flags: receive flags + * + * Used for SOCK_STREAM messages only. If not enough data is available + * will optionally wait for more; never truncates data. + * + * Returns size of returned message data, errno otherwise + */ +static int tipc_recv_stream(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t buf_len, int flags) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; + struct sk_buff *buf; + struct tipc_msg *msg; + long timeo; + unsigned int sz; + int sz_to_copy, target, needed; + int sz_copied = 0; + u32 err; + int res = 0; + + /* Catch invalid receive attempts */ + if (unlikely(!buf_len)) + return -EINVAL; + + lock_sock(sk); + + if (unlikely(sock->state == SS_UNCONNECTED)) { + res = -ENOTCONN; + goto exit; + } + + target = sock_rcvlowat(sk, flags & MSG_WAITALL, buf_len); + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + +restart: + /* Look for a message in receive queue; wait if necessary */ + res = tipc_wait_for_rcvmsg(sock, &timeo); + if (res) + goto exit; + + /* Look at first message in receive queue */ + buf = skb_peek(&sk->sk_receive_queue); + msg = buf_msg(buf); + sz = msg_data_sz(msg); + err = msg_errcode(msg); + + /* Discard an empty non-errored message & try again */ + if ((!sz) && (!err)) { + advance_rx_queue(sk); + goto restart; + } + + /* Optionally capture sender's address & ancillary data of first msg */ + if (sz_copied == 0) { + set_orig_addr(m, msg); + res = anc_data_recv(m, msg, port); + if (res) + goto exit; + } + + /* Capture message data (if valid) & compute return value (always) */ + if (!err) { + u32 offset = (u32)(unsigned long)(TIPC_SKB_CB(buf)->handle); + + sz -= offset; + needed = (buf_len - sz_copied); + sz_to_copy = (sz <= needed) ? sz : needed; + + res = skb_copy_datagram_iovec(buf, msg_hdr_sz(msg) + offset, + m->msg_iov, sz_to_copy); + if (res) + goto exit; + + sz_copied += sz_to_copy; + + if (sz_to_copy < sz) { + if (!(flags & MSG_PEEK)) + TIPC_SKB_CB(buf)->handle = + (void *)(unsigned long)(offset + sz_to_copy); + goto exit; + } + } else { + if (sz_copied != 0) + goto exit; /* can't add error msg to valid data */ + + if ((err == TIPC_CONN_SHUTDOWN) || m->msg_control) + res = 0; + else + res = -ECONNRESET; + } + + /* Consume received message (optional) */ + if (likely(!(flags & MSG_PEEK))) { + if (unlikely(++port->conn_unacked >= TIPC_CONNACK_INTV)) + tipc_acknowledge(port->ref, port->conn_unacked); + advance_rx_queue(sk); + } + + /* Loop around if more data is required */ + if ((sz_copied < buf_len) && /* didn't get all requested data */ + (!skb_queue_empty(&sk->sk_receive_queue) || + (sz_copied < target)) && /* and more is ready or required */ + (!(flags & MSG_PEEK)) && /* and aren't just peeking at data */ + (!err)) /* and haven't reached a FIN */ + goto restart; + +exit: + release_sock(sk); + return sz_copied ? sz_copied : res; +} + +/** + * tipc_write_space - wake up thread if port congestion is released + * @sk: socket + */ +static void tipc_write_space(struct sock *sk) +{ + struct socket_wq *wq; + + rcu_read_lock(); + wq = rcu_dereference(sk->sk_wq); + if (wq_has_sleeper(wq)) + wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | + POLLWRNORM | POLLWRBAND); + rcu_read_unlock(); +} + +/** + * tipc_data_ready - wake up threads to indicate messages have been received + * @sk: socket + * @len: the length of messages + */ +static void tipc_data_ready(struct sock *sk) +{ + struct socket_wq *wq; + + rcu_read_lock(); + wq = rcu_dereference(sk->sk_wq); + if (wq_has_sleeper(wq)) + wake_up_interruptible_sync_poll(&wq->wait, POLLIN | + POLLRDNORM | POLLRDBAND); + rcu_read_unlock(); +} + +/** + * filter_connect - Handle all incoming messages for a connection-based socket + * @tsk: TIPC socket + * @msg: message + * + * Returns TIPC error status code and socket error status code + * once it encounters some errors + */ +static u32 filter_connect(struct tipc_sock *tsk, struct sk_buff **buf) +{ + struct sock *sk = &tsk->sk; + struct tipc_port *port = &tsk->port; + struct socket *sock = sk->sk_socket; + struct tipc_msg *msg = buf_msg(*buf); + + u32 retval = TIPC_ERR_NO_PORT; + int res; + + if (msg_mcast(msg)) + return retval; + + switch ((int)sock->state) { + case SS_CONNECTED: + /* Accept only connection-based messages sent by peer */ + if (msg_connected(msg) && tipc_port_peer_msg(port, msg)) { + if (unlikely(msg_errcode(msg))) { + sock->state = SS_DISCONNECTING; + __tipc_port_disconnect(port); + } + retval = TIPC_OK; + } + break; + case SS_CONNECTING: + /* Accept only ACK or NACK message */ + if (unlikely(msg_errcode(msg))) { + sock->state = SS_DISCONNECTING; + sk->sk_err = ECONNREFUSED; + retval = TIPC_OK; + break; + } + + if (unlikely(!msg_connected(msg))) + break; + + res = auto_connect(tsk, msg); + if (res) { + sock->state = SS_DISCONNECTING; + sk->sk_err = -res; + retval = TIPC_OK; + break; + } + + /* If an incoming message is an 'ACK-', it should be + * discarded here because it doesn't contain useful + * data. In addition, we should try to wake up + * connect() routine if sleeping. + */ + if (msg_data_sz(msg) == 0) { + kfree_skb(*buf); + *buf = NULL; + if (waitqueue_active(sk_sleep(sk))) + wake_up_interruptible(sk_sleep(sk)); + } + retval = TIPC_OK; + break; + case SS_LISTENING: + case SS_UNCONNECTED: + /* Accept only SYN message */ + if (!msg_connected(msg) && !(msg_errcode(msg))) + retval = TIPC_OK; + break; + case SS_DISCONNECTING: + break; + default: + pr_err("Unknown socket state %u\n", sock->state); + } + return retval; +} + +/** + * rcvbuf_limit - get proper overload limit of socket receive queue + * @sk: socket + * @buf: message + * + * For all connection oriented messages, irrespective of importance, + * the default overload value (i.e. 67MB) is set as limit. + * + * For all connectionless messages, by default new queue limits are + * as belows: + * + * TIPC_LOW_IMPORTANCE (4 MB) + * TIPC_MEDIUM_IMPORTANCE (8 MB) + * TIPC_HIGH_IMPORTANCE (16 MB) + * TIPC_CRITICAL_IMPORTANCE (32 MB) + * + * Returns overload limit according to corresponding message importance + */ +static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + + if (msg_connected(msg)) + return sysctl_tipc_rmem[2]; + + return sk->sk_rcvbuf >> TIPC_CRITICAL_IMPORTANCE << + msg_importance(msg); +} + +/** + * filter_rcv - validate incoming message + * @sk: socket + * @buf: message + * + * Enqueues message on receive queue if acceptable; optionally handles + * disconnect indication for a connected socket. + * + * Called with socket lock already taken; port lock may also be taken. + * + * Returns TIPC error status code (TIPC_OK if message is not to be rejected) + */ +static u32 filter_rcv(struct sock *sk, struct sk_buff *buf) +{ + struct socket *sock = sk->sk_socket; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_msg *msg = buf_msg(buf); + unsigned int limit = rcvbuf_limit(sk, buf); + u32 res = TIPC_OK; + + /* Reject message if it is wrong sort of message for socket */ + if (msg_type(msg) > TIPC_DIRECT_MSG) + return TIPC_ERR_NO_PORT; + + if (sock->state == SS_READY) { + if (msg_connected(msg)) + return TIPC_ERR_NO_PORT; + } else { + res = filter_connect(tsk, &buf); + if (res != TIPC_OK || buf == NULL) + return res; + } + + /* Reject message if there isn't room to queue it */ + if (sk_rmem_alloc_get(sk) + buf->truesize >= limit) + return TIPC_ERR_OVERLOAD; + + /* Enqueue message */ + TIPC_SKB_CB(buf)->handle = NULL; + __skb_queue_tail(&sk->sk_receive_queue, buf); + skb_set_owner_r(buf, sk); + + sk->sk_data_ready(sk); + return TIPC_OK; +} + +/** + * tipc_backlog_rcv - handle incoming message from backlog queue + * @sk: socket + * @buf: message + * + * Caller must hold socket lock, but not port lock. + * + * Returns 0 + */ +static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *buf) +{ + u32 res; + struct tipc_sock *tsk = tipc_sk(sk); + uint truesize = buf->truesize; + + res = filter_rcv(sk, buf); + if (unlikely(res)) + tipc_reject_msg(buf, res); + + if (atomic_read(&tsk->dupl_rcvcnt) < TIPC_CONN_OVERLOAD_LIMIT) + atomic_add(truesize, &tsk->dupl_rcvcnt); + + return 0; +} + +/** + * tipc_sk_rcv - handle incoming message + * @buf: buffer containing arriving message + * Consumes buffer + * Returns 0 if success, or errno: -EHOSTUNREACH + */ +int tipc_sk_rcv(struct sk_buff *buf) +{ + struct tipc_sock *tsk; + struct tipc_port *port; + struct sock *sk; + u32 dport = msg_destport(buf_msg(buf)); + int err = TIPC_OK; + uint limit; + + /* Forward unresolved named message */ + if (unlikely(!dport)) { + tipc_net_route_msg(buf); + return 0; + } + + /* Validate destination */ + port = tipc_port_lock(dport); + if (unlikely(!port)) { + err = TIPC_ERR_NO_PORT; + goto exit; + } + + tsk = tipc_port_to_sock(port); + sk = &tsk->sk; + + /* Queue message */ + bh_lock_sock(sk); + + if (!sock_owned_by_user(sk)) { + err = filter_rcv(sk, buf); + } else { + if (sk->sk_backlog.len == 0) + atomic_set(&tsk->dupl_rcvcnt, 0); + limit = rcvbuf_limit(sk, buf) + atomic_read(&tsk->dupl_rcvcnt); + if (sk_add_backlog(sk, buf, limit)) + err = TIPC_ERR_OVERLOAD; + } + + bh_unlock_sock(sk); + tipc_port_unlock(port); + + if (likely(!err)) + return 0; +exit: + tipc_reject_msg(buf, err); + return -EHOSTUNREACH; +} + +static int tipc_wait_for_connect(struct socket *sock, long *timeo_p) +{ + struct sock *sk = sock->sk; + DEFINE_WAIT(wait); + int done; + + do { + int err = sock_error(sk); + if (err) + return err; + if (!*timeo_p) + return -ETIMEDOUT; + if (signal_pending(current)) + return sock_intr_errno(*timeo_p); + + prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + done = sk_wait_event(sk, timeo_p, sock->state != SS_CONNECTING); + finish_wait(sk_sleep(sk), &wait); + } while (!done); + return 0; +} + +/** + * tipc_connect - establish a connection to another TIPC port + * @sock: socket structure + * @dest: socket address for destination port + * @destlen: size of socket address data structure + * @flags: file-related flags associated with socket + * + * Returns 0 on success, errno otherwise + */ +static int tipc_connect(struct socket *sock, struct sockaddr *dest, + int destlen, int flags) +{ + struct sock *sk = sock->sk; + struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest; + struct msghdr m = {NULL,}; + long timeout = (flags & O_NONBLOCK) ? 0 : tipc_sk(sk)->conn_timeout; + socket_state previous; + int res; + + lock_sock(sk); + + /* For now, TIPC does not allow use of connect() with DGRAM/RDM types */ + if (sock->state == SS_READY) { + res = -EOPNOTSUPP; + goto exit; + } + + /* + * Reject connection attempt using multicast address + * + * Note: send_msg() validates the rest of the address fields, + * so there's no need to do it here + */ + if (dst->addrtype == TIPC_ADDR_MCAST) { + res = -EINVAL; + goto exit; + } + + previous = sock->state; + switch (sock->state) { + case SS_UNCONNECTED: + /* Send a 'SYN-' to destination */ + m.msg_name = dest; + m.msg_namelen = destlen; + + /* If connect is in non-blocking case, set MSG_DONTWAIT to + * indicate send_msg() is never blocked. + */ + if (!timeout) + m.msg_flags = MSG_DONTWAIT; + + res = tipc_sendmsg(NULL, sock, &m, 0); + if ((res < 0) && (res != -EWOULDBLOCK)) + goto exit; + + /* Just entered SS_CONNECTING state; the only + * difference is that return value in non-blocking + * case is EINPROGRESS, rather than EALREADY. + */ + res = -EINPROGRESS; + case SS_CONNECTING: + if (previous == SS_CONNECTING) + res = -EALREADY; + if (!timeout) + goto exit; + timeout = msecs_to_jiffies(timeout); + /* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */ + res = tipc_wait_for_connect(sock, &timeout); + break; + case SS_CONNECTED: + res = -EISCONN; + break; + default: + res = -EINVAL; + break; + } +exit: + release_sock(sk); + return res; +} + +/** + * tipc_listen - allow socket to listen for incoming connections + * @sock: socket structure + * @len: (unused) + * + * Returns 0 on success, errno otherwise + */ +static int tipc_listen(struct socket *sock, int len) +{ + struct sock *sk = sock->sk; + int res; + + lock_sock(sk); + + if (sock->state != SS_UNCONNECTED) + res = -EINVAL; + else { + sock->state = SS_LISTENING; + res = 0; + } + + release_sock(sk); + return res; +} + +static int tipc_wait_for_accept(struct socket *sock, long timeo) +{ + struct sock *sk = sock->sk; + DEFINE_WAIT(wait); + int err; + + /* True wake-one mechanism for incoming connections: only + * one process gets woken up, not the 'whole herd'. + * Since we do not 'race & poll' for established sockets + * anymore, the common case will execute the loop only once. + */ + for (;;) { + prepare_to_wait_exclusive(sk_sleep(sk), &wait, + TASK_INTERRUPTIBLE); + if (timeo && skb_queue_empty(&sk->sk_receive_queue)) { + release_sock(sk); + timeo = schedule_timeout(timeo); + lock_sock(sk); + } + err = 0; + if (!skb_queue_empty(&sk->sk_receive_queue)) + break; + err = -EINVAL; + if (sock->state != SS_LISTENING) + break; + err = sock_intr_errno(timeo); + if (signal_pending(current)) + break; + err = -EAGAIN; + if (!timeo) + break; + } + finish_wait(sk_sleep(sk), &wait); + return err; +} + +/** + * tipc_accept - wait for connection request + * @sock: listening socket + * @newsock: new socket that is to be connected + * @flags: file-related flags associated with socket + * + * Returns 0 on success, errno otherwise + */ +static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags) +{ + struct sock *new_sk, *sk = sock->sk; + struct sk_buff *buf; + struct tipc_port *new_port; + struct tipc_msg *msg; + struct tipc_portid peer; + u32 new_ref; + long timeo; + int res; + + lock_sock(sk); + + if (sock->state != SS_LISTENING) { + res = -EINVAL; + goto exit; + } + timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + res = tipc_wait_for_accept(sock, timeo); + if (res) + goto exit; + + buf = skb_peek(&sk->sk_receive_queue); + + res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, 1); + if (res) + goto exit; + + new_sk = new_sock->sk; + new_port = &tipc_sk(new_sk)->port; + new_ref = new_port->ref; + msg = buf_msg(buf); + + /* we lock on new_sk; but lockdep sees the lock on sk */ + lock_sock_nested(new_sk, SINGLE_DEPTH_NESTING); + + /* + * Reject any stray messages received by new socket + * before the socket lock was taken (very, very unlikely) + */ + reject_rx_queue(new_sk); + + /* Connect new socket to it's peer */ + peer.ref = msg_origport(msg); + peer.node = msg_orignode(msg); + tipc_port_connect(new_ref, &peer); + new_sock->state = SS_CONNECTED; + + tipc_port_set_importance(new_port, msg_importance(msg)); + if (msg_named(msg)) { + new_port->conn_type = msg_nametype(msg); + new_port->conn_instance = msg_nameinst(msg); + } + + /* + * Respond to 'SYN-' by discarding it & returning 'ACK'-. + * Respond to 'SYN+' by queuing it on new socket. + */ + if (!msg_data_sz(msg)) { + struct msghdr m = {NULL,}; + + advance_rx_queue(sk); + tipc_send_packet(NULL, new_sock, &m, 0); + } else { + __skb_dequeue(&sk->sk_receive_queue); + __skb_queue_head(&new_sk->sk_receive_queue, buf); + skb_set_owner_r(buf, new_sk); + } + release_sock(new_sk); +exit: + release_sock(sk); + return res; +} + +/** + * tipc_shutdown - shutdown socket connection + * @sock: socket structure + * @how: direction to close (must be SHUT_RDWR) + * + * Terminates connection (if necessary), then purges socket's receive queue. + * + * Returns 0 on success, errno otherwise + */ +static int tipc_shutdown(struct socket *sock, int how) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; + struct sk_buff *buf; + int res; + + if (how != SHUT_RDWR) + return -EINVAL; + + lock_sock(sk); + + switch (sock->state) { + case SS_CONNECTING: + case SS_CONNECTED: + +restart: + /* Disconnect and send a 'FIN+' or 'FIN-' message to peer */ + buf = __skb_dequeue(&sk->sk_receive_queue); + if (buf) { + if (TIPC_SKB_CB(buf)->handle != NULL) { + kfree_skb(buf); + goto restart; + } + tipc_port_disconnect(port->ref); + tipc_reject_msg(buf, TIPC_CONN_SHUTDOWN); + } else { + tipc_port_shutdown(port->ref); + } + + sock->state = SS_DISCONNECTING; + + /* fall through */ + + case SS_DISCONNECTING: + + /* Discard any unreceived messages */ + __skb_queue_purge(&sk->sk_receive_queue); + + /* Wake up anyone sleeping in poll */ + sk->sk_state_change(sk); + res = 0; + break; + + default: + res = -ENOTCONN; + } + + release_sock(sk); + return res; +} + +/** + * tipc_setsockopt - set socket option + * @sock: socket structure + * @lvl: option level + * @opt: option identifier + * @ov: pointer to new option value + * @ol: length of option value + * + * For stream sockets only, accepts and ignores all IPPROTO_TCP options + * (to ease compatibility). + * + * Returns 0 on success, errno otherwise + */ +static int tipc_setsockopt(struct socket *sock, int lvl, int opt, + char __user *ov, unsigned int ol) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; + u32 value; + int res; + + if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM)) + return 0; + if (lvl != SOL_TIPC) + return -ENOPROTOOPT; + if (ol < sizeof(value)) + return -EINVAL; + res = get_user(value, (u32 __user *)ov); + if (res) + return res; + + lock_sock(sk); + + switch (opt) { + case TIPC_IMPORTANCE: + tipc_port_set_importance(port, value); + break; + case TIPC_SRC_DROPPABLE: + if (sock->type != SOCK_STREAM) + tipc_port_set_unreliable(port, value); + else + res = -ENOPROTOOPT; + break; + case TIPC_DEST_DROPPABLE: + tipc_port_set_unreturnable(port, value); + break; + case TIPC_CONN_TIMEOUT: + tipc_sk(sk)->conn_timeout = value; + /* no need to set "res", since already 0 at this point */ + break; + default: + res = -EINVAL; + } + + release_sock(sk); + + return res; +} + +/** + * tipc_getsockopt - get socket option + * @sock: socket structure + * @lvl: option level + * @opt: option identifier + * @ov: receptacle for option value + * @ol: receptacle for length of option value + * + * For stream sockets only, returns 0 length result for all IPPROTO_TCP options + * (to ease compatibility). + * + * Returns 0 on success, errno otherwise + */ +static int tipc_getsockopt(struct socket *sock, int lvl, int opt, + char __user *ov, int __user *ol) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; + int len; + u32 value; + int res; + + if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM)) + return put_user(0, ol); + if (lvl != SOL_TIPC) + return -ENOPROTOOPT; + res = get_user(len, ol); + if (res) + return res; + + lock_sock(sk); + + switch (opt) { + case TIPC_IMPORTANCE: + value = tipc_port_importance(port); + break; + case TIPC_SRC_DROPPABLE: + value = tipc_port_unreliable(port); + break; + case TIPC_DEST_DROPPABLE: + value = tipc_port_unreturnable(port); + break; + case TIPC_CONN_TIMEOUT: + value = tipc_sk(sk)->conn_timeout; + /* no need to set "res", since already 0 at this point */ + break; + case TIPC_NODE_RECVQ_DEPTH: + value = 0; /* was tipc_queue_size, now obsolete */ + break; + case TIPC_SOCK_RECVQ_DEPTH: + value = skb_queue_len(&sk->sk_receive_queue); + break; + default: + res = -EINVAL; + } + + release_sock(sk); + + if (res) + return res; /* "get" failed */ + + if (len < sizeof(value)) + return -EINVAL; + + if (copy_to_user(ov, &value, sizeof(value))) + return -EFAULT; + + return put_user(sizeof(value), ol); +} + +int tipc_ioctl(struct socket *sk, unsigned int cmd, unsigned long arg) +{ + struct tipc_sioc_ln_req lnr; + void __user *argp = (void __user *)arg; + + switch (cmd) { + case SIOCGETLINKNAME: + if (copy_from_user(&lnr, argp, sizeof(lnr))) + return -EFAULT; + if (!tipc_node_get_linkname(lnr.bearer_id, lnr.peer, + lnr.linkname, TIPC_MAX_LINK_NAME)) { + if (copy_to_user(argp, &lnr, sizeof(lnr))) + return -EFAULT; + return 0; + } + return -EADDRNOTAVAIL; + break; + default: + return -ENOIOCTLCMD; + } +} + +/* Protocol switches for the various types of TIPC sockets */ + +static const struct proto_ops msg_ops = { + .owner = THIS_MODULE, + .family = AF_TIPC, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_connect, + .socketpair = sock_no_socketpair, + .accept = sock_no_accept, + .getname = tipc_getname, + .poll = tipc_poll, + .ioctl = tipc_ioctl, + .listen = sock_no_listen, + .shutdown = tipc_shutdown, + .setsockopt = tipc_setsockopt, + .getsockopt = tipc_getsockopt, + .sendmsg = tipc_sendmsg, + .recvmsg = tipc_recvmsg, + .mmap = sock_no_mmap, + .sendpage = sock_no_sendpage +}; + +static const struct proto_ops packet_ops = { + .owner = THIS_MODULE, + .family = AF_TIPC, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_connect, + .socketpair = sock_no_socketpair, + .accept = tipc_accept, + .getname = tipc_getname, + .poll = tipc_poll, + .ioctl = tipc_ioctl, + .listen = tipc_listen, + .shutdown = tipc_shutdown, + .setsockopt = tipc_setsockopt, + .getsockopt = tipc_getsockopt, + .sendmsg = tipc_send_packet, + .recvmsg = tipc_recvmsg, + .mmap = sock_no_mmap, + .sendpage = sock_no_sendpage +}; + +static const struct proto_ops stream_ops = { + .owner = THIS_MODULE, + .family = AF_TIPC, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_connect, + .socketpair = sock_no_socketpair, + .accept = tipc_accept, + .getname = tipc_getname, + .poll = tipc_poll, + .ioctl = tipc_ioctl, + .listen = tipc_listen, + .shutdown = tipc_shutdown, + .setsockopt = tipc_setsockopt, + .getsockopt = tipc_getsockopt, + .sendmsg = tipc_send_stream, + .recvmsg = tipc_recv_stream, + .mmap = sock_no_mmap, + .sendpage = sock_no_sendpage +}; + +static const struct net_proto_family tipc_family_ops = { + .owner = THIS_MODULE, + .family = AF_TIPC, + .create = tipc_sk_create +}; + +static struct proto tipc_proto = { + .name = "TIPC", + .owner = THIS_MODULE, + .obj_size = sizeof(struct tipc_sock), + .sysctl_rmem = sysctl_tipc_rmem +}; + +static struct proto tipc_proto_kern = { + .name = "TIPC", + .obj_size = sizeof(struct tipc_sock), + .sysctl_rmem = sysctl_tipc_rmem +}; + +/** + * tipc_socket_init - initialize TIPC socket interface + * + * Returns 0 on success, errno otherwise + */ +int tipc_socket_init(void) +{ + int res; + + res = proto_register(&tipc_proto, 1); + if (res) { + pr_err("Failed to register TIPC protocol type\n"); + goto out; + } + + res = sock_register(&tipc_family_ops); + if (res) { + pr_err("Failed to register TIPC socket type\n"); + proto_unregister(&tipc_proto); + goto out; + } + out: + return res; +} + +/** + * tipc_socket_stop - stop TIPC socket interface + */ +void tipc_socket_stop(void) +{ + sock_unregister(tipc_family_ops.family); + proto_unregister(&tipc_proto); +} diff --git a/net/tipc/socket.h b/net/tipc/socket.h new file mode 100644 index 00000000000..3afcd2a70b3 --- /dev/null +++ b/net/tipc/socket.h @@ -0,0 +1,74 @@ +/* net/tipc/socket.h: Include file for TIPC socket code + * + * Copyright (c) 2014, Ericsson AB + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_SOCK_H +#define _TIPC_SOCK_H + +#include "port.h" +#include <net/sock.h> + +/** + * struct tipc_sock - TIPC socket structure + * @sk: socket - interacts with 'port' and with user via the socket API + * @port: port - interacts with 'sk' and with the rest of the TIPC stack + * @peer_name: the peer of the connection, if any + * @conn_timeout: the time we can wait for an unresponded setup request + * @dupl_rcvcnt: number of bytes counted twice, in both backlog and rcv queue + */ + +struct tipc_sock { + struct sock sk; + struct tipc_port port; + unsigned int conn_timeout; + atomic_t dupl_rcvcnt; +}; + +static inline struct tipc_sock *tipc_sk(const struct sock *sk) +{ + return container_of(sk, struct tipc_sock, sk); +} + +static inline struct tipc_sock *tipc_port_to_sock(const struct tipc_port *port) +{ + return container_of(port, struct tipc_sock, port); +} + +static inline void tipc_sock_wakeup(struct tipc_sock *tsk) +{ + tsk->sk.sk_write_space(&tsk->sk); +} + +int tipc_sk_rcv(struct sk_buff *buf); + +#endif diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c new file mode 100644 index 00000000000..642437231ad --- /dev/null +++ b/net/tipc/subscr.c @@ -0,0 +1,375 @@ +/* + * net/tipc/subscr.c: TIPC network topology service + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005-2007, 2010-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "name_table.h" +#include "port.h" +#include "subscr.h" + +/** + * struct tipc_subscriber - TIPC network topology subscriber + * @conid: connection identifier to server connecting to subscriber + * @lock: control access to subscriber + * @subscription_list: list of subscription objects for this subscriber + */ +struct tipc_subscriber { + int conid; + spinlock_t lock; + struct list_head subscription_list; +}; + +static void subscr_conn_msg_event(int conid, struct sockaddr_tipc *addr, + void *usr_data, void *buf, size_t len); +static void *subscr_named_msg_event(int conid); +static void subscr_conn_shutdown_event(int conid, void *usr_data); + +static atomic_t subscription_count = ATOMIC_INIT(0); + +static struct sockaddr_tipc topsrv_addr __read_mostly = { + .family = AF_TIPC, + .addrtype = TIPC_ADDR_NAMESEQ, + .addr.nameseq.type = TIPC_TOP_SRV, + .addr.nameseq.lower = TIPC_TOP_SRV, + .addr.nameseq.upper = TIPC_TOP_SRV, + .scope = TIPC_NODE_SCOPE +}; + +static struct tipc_server topsrv __read_mostly = { + .saddr = &topsrv_addr, + .imp = TIPC_CRITICAL_IMPORTANCE, + .type = SOCK_SEQPACKET, + .max_rcvbuf_size = sizeof(struct tipc_subscr), + .name = "topology_server", + .tipc_conn_recvmsg = subscr_conn_msg_event, + .tipc_conn_new = subscr_named_msg_event, + .tipc_conn_shutdown = subscr_conn_shutdown_event, +}; + +/** + * htohl - convert value to endianness used by destination + * @in: value to convert + * @swap: non-zero if endianness must be reversed + * + * Returns converted value + */ +static u32 htohl(u32 in, int swap) +{ + return swap ? swab32(in) : in; +} + +static void subscr_send_event(struct tipc_subscription *sub, u32 found_lower, + u32 found_upper, u32 event, u32 port_ref, + u32 node) +{ + struct tipc_subscriber *subscriber = sub->subscriber; + struct kvec msg_sect; + + msg_sect.iov_base = (void *)&sub->evt; + msg_sect.iov_len = sizeof(struct tipc_event); + sub->evt.event = htohl(event, sub->swap); + sub->evt.found_lower = htohl(found_lower, sub->swap); + sub->evt.found_upper = htohl(found_upper, sub->swap); + sub->evt.port.ref = htohl(port_ref, sub->swap); + sub->evt.port.node = htohl(node, sub->swap); + tipc_conn_sendmsg(&topsrv, subscriber->conid, NULL, msg_sect.iov_base, + msg_sect.iov_len); +} + +/** + * tipc_subscr_overlap - test for subscription overlap with the given values + * + * Returns 1 if there is overlap, otherwise 0. + */ +int tipc_subscr_overlap(struct tipc_subscription *sub, u32 found_lower, + u32 found_upper) +{ + if (found_lower < sub->seq.lower) + found_lower = sub->seq.lower; + if (found_upper > sub->seq.upper) + found_upper = sub->seq.upper; + if (found_lower > found_upper) + return 0; + return 1; +} + +/** + * tipc_subscr_report_overlap - issue event if there is subscription overlap + * + * Protected by nameseq.lock in name_table.c + */ +void tipc_subscr_report_overlap(struct tipc_subscription *sub, u32 found_lower, + u32 found_upper, u32 event, u32 port_ref, + u32 node, int must) +{ + if (!tipc_subscr_overlap(sub, found_lower, found_upper)) + return; + if (!must && !(sub->filter & TIPC_SUB_PORTS)) + return; + + subscr_send_event(sub, found_lower, found_upper, event, port_ref, node); +} + +static void subscr_timeout(struct tipc_subscription *sub) +{ + struct tipc_subscriber *subscriber = sub->subscriber; + + /* The spin lock per subscriber is used to protect its members */ + spin_lock_bh(&subscriber->lock); + + /* Validate timeout (in case subscription is being cancelled) */ + if (sub->timeout == TIPC_WAIT_FOREVER) { + spin_unlock_bh(&subscriber->lock); + return; + } + + /* Unlink subscription from name table */ + tipc_nametbl_unsubscribe(sub); + + /* Unlink subscription from subscriber */ + list_del(&sub->subscription_list); + + spin_unlock_bh(&subscriber->lock); + + /* Notify subscriber of timeout */ + subscr_send_event(sub, sub->evt.s.seq.lower, sub->evt.s.seq.upper, + TIPC_SUBSCR_TIMEOUT, 0, 0); + + /* Now destroy subscription */ + k_term_timer(&sub->timer); + kfree(sub); + atomic_dec(&subscription_count); +} + +/** + * subscr_del - delete a subscription within a subscription list + * + * Called with subscriber lock held. + */ +static void subscr_del(struct tipc_subscription *sub) +{ + tipc_nametbl_unsubscribe(sub); + list_del(&sub->subscription_list); + kfree(sub); + atomic_dec(&subscription_count); +} + +/** + * subscr_terminate - terminate communication with a subscriber + * + * Note: Must call it in process context since it might sleep. + */ +static void subscr_terminate(struct tipc_subscriber *subscriber) +{ + tipc_conn_terminate(&topsrv, subscriber->conid); +} + +static void subscr_release(struct tipc_subscriber *subscriber) +{ + struct tipc_subscription *sub; + struct tipc_subscription *sub_temp; + + spin_lock_bh(&subscriber->lock); + + /* Destroy any existing subscriptions for subscriber */ + list_for_each_entry_safe(sub, sub_temp, &subscriber->subscription_list, + subscription_list) { + if (sub->timeout != TIPC_WAIT_FOREVER) { + spin_unlock_bh(&subscriber->lock); + k_cancel_timer(&sub->timer); + k_term_timer(&sub->timer); + spin_lock_bh(&subscriber->lock); + } + subscr_del(sub); + } + spin_unlock_bh(&subscriber->lock); + + /* Now destroy subscriber */ + kfree(subscriber); +} + +/** + * subscr_cancel - handle subscription cancellation request + * + * Called with subscriber lock held. Routine must temporarily release lock + * to enable the subscription timeout routine to finish without deadlocking; + * the lock is then reclaimed to allow caller to release it upon return. + * + * Note that fields of 's' use subscriber's endianness! + */ +static void subscr_cancel(struct tipc_subscr *s, + struct tipc_subscriber *subscriber) +{ + struct tipc_subscription *sub; + struct tipc_subscription *sub_temp; + int found = 0; + + /* Find first matching subscription, exit if not found */ + list_for_each_entry_safe(sub, sub_temp, &subscriber->subscription_list, + subscription_list) { + if (!memcmp(s, &sub->evt.s, sizeof(struct tipc_subscr))) { + found = 1; + break; + } + } + if (!found) + return; + + /* Cancel subscription timer (if used), then delete subscription */ + if (sub->timeout != TIPC_WAIT_FOREVER) { + sub->timeout = TIPC_WAIT_FOREVER; + spin_unlock_bh(&subscriber->lock); + k_cancel_timer(&sub->timer); + k_term_timer(&sub->timer); + spin_lock_bh(&subscriber->lock); + } + subscr_del(sub); +} + +/** + * subscr_subscribe - create subscription for subscriber + * + * Called with subscriber lock held. + */ +static int subscr_subscribe(struct tipc_subscr *s, + struct tipc_subscriber *subscriber, + struct tipc_subscription **sub_p) { + struct tipc_subscription *sub; + int swap; + + /* Determine subscriber's endianness */ + swap = !(s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE)); + + /* Detect & process a subscription cancellation request */ + if (s->filter & htohl(TIPC_SUB_CANCEL, swap)) { + s->filter &= ~htohl(TIPC_SUB_CANCEL, swap); + subscr_cancel(s, subscriber); + return 0; + } + + /* Refuse subscription if global limit exceeded */ + if (atomic_read(&subscription_count) >= TIPC_MAX_SUBSCRIPTIONS) { + pr_warn("Subscription rejected, limit reached (%u)\n", + TIPC_MAX_SUBSCRIPTIONS); + return -EINVAL; + } + + /* Allocate subscription object */ + sub = kmalloc(sizeof(*sub), GFP_ATOMIC); + if (!sub) { + pr_warn("Subscription rejected, no memory\n"); + return -ENOMEM; + } + + /* Initialize subscription object */ + sub->seq.type = htohl(s->seq.type, swap); + sub->seq.lower = htohl(s->seq.lower, swap); + sub->seq.upper = htohl(s->seq.upper, swap); + sub->timeout = htohl(s->timeout, swap); + sub->filter = htohl(s->filter, swap); + if ((!(sub->filter & TIPC_SUB_PORTS) == + !(sub->filter & TIPC_SUB_SERVICE)) || + (sub->seq.lower > sub->seq.upper)) { + pr_warn("Subscription rejected, illegal request\n"); + kfree(sub); + return -EINVAL; + } + INIT_LIST_HEAD(&sub->nameseq_list); + list_add(&sub->subscription_list, &subscriber->subscription_list); + sub->subscriber = subscriber; + sub->swap = swap; + memcpy(&sub->evt.s, s, sizeof(struct tipc_subscr)); + atomic_inc(&subscription_count); + if (sub->timeout != TIPC_WAIT_FOREVER) { + k_init_timer(&sub->timer, + (Handler)subscr_timeout, (unsigned long)sub); + k_start_timer(&sub->timer, sub->timeout); + } + *sub_p = sub; + return 0; +} + +/* Handle one termination request for the subscriber */ +static void subscr_conn_shutdown_event(int conid, void *usr_data) +{ + subscr_release((struct tipc_subscriber *)usr_data); +} + +/* Handle one request to create a new subscription for the subscriber */ +static void subscr_conn_msg_event(int conid, struct sockaddr_tipc *addr, + void *usr_data, void *buf, size_t len) +{ + struct tipc_subscriber *subscriber = usr_data; + struct tipc_subscription *sub = NULL; + + spin_lock_bh(&subscriber->lock); + if (subscr_subscribe((struct tipc_subscr *)buf, subscriber, &sub) < 0) { + spin_unlock_bh(&subscriber->lock); + subscr_terminate(subscriber); + return; + } + if (sub) + tipc_nametbl_subscribe(sub); + spin_unlock_bh(&subscriber->lock); +} + + +/* Handle one request to establish a new subscriber */ +static void *subscr_named_msg_event(int conid) +{ + struct tipc_subscriber *subscriber; + + /* Create subscriber object */ + subscriber = kzalloc(sizeof(struct tipc_subscriber), GFP_ATOMIC); + if (subscriber == NULL) { + pr_warn("Subscriber rejected, no memory\n"); + return NULL; + } + INIT_LIST_HEAD(&subscriber->subscription_list); + subscriber->conid = conid; + spin_lock_init(&subscriber->lock); + + return (void *)subscriber; +} + +int tipc_subscr_start(void) +{ + return tipc_server_start(&topsrv); +} + +void tipc_subscr_stop(void) +{ + tipc_server_stop(&topsrv); +} diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h new file mode 100644 index 00000000000..393e417bee3 --- /dev/null +++ b/net/tipc/subscr.h @@ -0,0 +1,81 @@ +/* + * net/tipc/subscr.h: Include file for TIPC network topology service + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005-2007, 2012-2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_SUBSCR_H +#define _TIPC_SUBSCR_H + +#include "server.h" + +struct tipc_subscription; +struct tipc_subscriber; + +/** + * struct tipc_subscription - TIPC network topology subscription object + * @subscriber: pointer to its subscriber + * @seq: name sequence associated with subscription + * @timeout: duration of subscription (in ms) + * @filter: event filtering to be done for subscription + * @timer: timer governing subscription duration (optional) + * @nameseq_list: adjacent subscriptions in name sequence's subscription list + * @subscription_list: adjacent subscriptions in subscriber's subscription list + * @server_ref: object reference of server port associated with subscription + * @swap: indicates if subscriber uses opposite endianness in its messages + * @evt: template for events generated by subscription + */ +struct tipc_subscription { + struct tipc_subscriber *subscriber; + struct tipc_name_seq seq; + u32 timeout; + u32 filter; + struct timer_list timer; + struct list_head nameseq_list; + struct list_head subscription_list; + int swap; + struct tipc_event evt; +}; + +int tipc_subscr_overlap(struct tipc_subscription *sub, u32 found_lower, + u32 found_upper); + +void tipc_subscr_report_overlap(struct tipc_subscription *sub, u32 found_lower, + u32 found_upper, u32 event, u32 port_ref, + u32 node, int must); + +int tipc_subscr_start(void); + +void tipc_subscr_stop(void); + +#endif diff --git a/net/tipc/sysctl.c b/net/tipc/sysctl.c new file mode 100644 index 00000000000..f3fef93325a --- /dev/null +++ b/net/tipc/sysctl.c @@ -0,0 +1,64 @@ +/* + * net/tipc/sysctl.c: sysctl interface to TIPC subsystem + * + * Copyright (c) 2013, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" + +#include <linux/sysctl.h> + +static struct ctl_table_header *tipc_ctl_hdr; + +static struct ctl_table tipc_table[] = { + { + .procname = "tipc_rmem", + .data = &sysctl_tipc_rmem, + .maxlen = sizeof(sysctl_tipc_rmem), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + {} +}; + +int tipc_register_sysctl(void) +{ + tipc_ctl_hdr = register_net_sysctl(&init_net, "net/tipc", tipc_table); + if (tipc_ctl_hdr == NULL) + return -ENOMEM; + return 0; +} + +void tipc_unregister_sysctl(void) +{ + unregister_net_sysctl_table(tipc_ctl_hdr); +} |