diff options
Diffstat (limited to 'net/tipc')
49 files changed, 7216 insertions, 10839 deletions
diff --git a/net/tipc/Kconfig b/net/tipc/Kconfig index 3b30d1130b6..c890848f9d5 100644 --- a/net/tipc/Kconfig +++ b/net/tipc/Kconfig @@ -3,14 +3,14 @@ # menuconfig TIPC - tristate "The TIPC Protocol (EXPERIMENTAL)" - depends on INET && EXPERIMENTAL + 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 @@ -20,96 +20,21 @@ menuconfig TIPC If in doubt, say N. -if TIPC - -config TIPC_ADVANCED - bool "TIPC: Advanced configuration" - default n - help - Saying Y here will open some advanced configuration - for TIPC. Most users do not need to bother, so if - unsure, just say N. - -config TIPC_ZONES - int "Maximum number of zones in network" - depends on TIPC_ADVANCED - default "3" - help - Max number of zones inside TIPC network. Max supported value - is 255 zones, minimum is 1 - - Default is 3 zones in a network; setting this to higher - allows more zones but might use more memory. - -config TIPC_CLUSTERS - int "Maximum number of clusters in a zone" - depends on TIPC_ADVANCED - default "1" - help - ***Only 1 (one cluster in a zone) is supported by current code. - Any value set here will be overridden.*** - - (Max number of clusters inside TIPC zone. Max supported - value is 4095 clusters, minimum is 1. - - Default is 1; setting this to smaller value might save - some memory, setting it to higher - allows more clusters and might consume more memory.) - -config TIPC_NODES - int "Maximum number of nodes in cluster" - depends on TIPC_ADVANCED - default "255" - help - Maximum number of nodes inside a TIPC cluster. Maximum - supported value is 2047 nodes, minimum is 8. - - Setting this to a smaller value saves some memory, - setting it to higher allows more nodes. - -config TIPC_SLAVE_NODES - int "Maximum number of slave nodes in cluster" - depends on TIPC_ADVANCED - default "0" - help - ***This capability is not supported by current code.*** - - Maximum number of slave nodes inside a TIPC cluster. Maximum - supported value is 2047 nodes, minimum is 0. - - Setting this to a smaller value saves some memory, - setting it to higher allows more nodes. - config TIPC_PORTS int "Maximum number of ports in a node" - depends on TIPC_ADVANCED + depends on TIPC + range 127 65535 default "8191" help - Maximum number of ports within a node. Maximum - supported value is 64535 nodes, minimum is 127. + 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 more ports. + Setting this to a smaller value saves some memory, + setting it to higher allows for more ports. -config TIPC_LOG - int "Size of log buffer" - depends on TIPC_ADVANCED - default 0 +config TIPC_MEDIA_IB + bool "InfiniBand media type support" + depends on TIPC && INFINIBAND_IPOIB help - Size (in bytes) of TIPC's internal log buffer, which records the - occurrence of significant events. Maximum supported value - is 32768 bytes, minimum is 0. - - There is no need to enable the log buffer unless the node will be - managed remotely via TIPC. - -config TIPC_DEBUG - bool "Enable debugging support" - default n - help - This will enable debugging of TIPC. - - Only say Y here if you are having trouble with TIPC. It will - enable the display of detailed information about what is going on. - -endif # TIPC + Saying Y here will enable support for running TIPC on + IP-over-InfiniBand devices. diff --git a/net/tipc/Makefile b/net/tipc/Makefile index dceb7027946..a080c66d819 100644 --- a/net/tipc/Makefile +++ b/net/tipc/Makefile @@ -4,10 +4,11 @@ obj-$(CONFIG_TIPC) := tipc.o -tipc-y += addr.o bcast.o bearer.o config.o cluster.o \ - core.o handler.o link.o discover.o msg.o \ +tipc-y += addr.o bcast.o bearer.o config.o \ + core.o link.o discover.o msg.o \ name_distr.o subscr.o name_table.o net.o \ netlink.o node.o node_subscr.o port.o ref.o \ - socket.o user_reg.o zone.o dbg.o eth_media.o + socket.o log.o eth_media.o server.o -# End of file +tipc-$(CONFIG_TIPC_MEDIA_IB) += ib_media.o +tipc-$(CONFIG_SYSCTL) += sysctl.o diff --git a/net/tipc/addr.c b/net/tipc/addr.c index e5207a11edf..357b74b26f9 100644 --- a/net/tipc/addr.c +++ b/net/tipc/addr.c @@ -2,7 +2,7 @@ * net/tipc/addr.c: TIPC address utility routines * * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2004-2005, Wind River Systems + * Copyright (c) 2004-2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -35,41 +35,21 @@ */ #include "core.h" -#include "dbg.h" #include "addr.h" -#include "zone.h" -#include "cluster.h" -#include "net.h" - -u32 tipc_get_addr(void) -{ - return tipc_own_addr; -} /** * 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 and do not exceed the configured limits. + * 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); - u32 max_nodes = tipc_max_nodes; - - if (is_slave(addr)) - max_nodes = LOWEST_SLAVE + tipc_max_slaves; - if (n > max_nodes) - return 0; - if (c > tipc_max_clusters) - return 0; - if (z > tipc_max_zones) - return 0; if (n && (!z || !c)) return 0; @@ -81,14 +61,43 @@ int tipc_addr_domain_valid(u32 addr) /** * tipc_addr_node_valid - validates a proposed network address for this node * - * Accepts <Z.C.N>, where Z, C, and N are non-zero and do not exceed - * the configured limits. + * 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)); + 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 index 3ba67e6ce03..a74acf9ee80 100644 --- a/net/tipc/addr.h +++ b/net/tipc/addr.h @@ -37,60 +37,40 @@ #ifndef _TIPC_ADDR_H #define _TIPC_ADDR_H -static inline u32 own_node(void) -{ - return tipc_node(tipc_own_addr); -} - -static inline u32 own_cluster(void) -{ - return tipc_cluster(tipc_own_addr); -} +#include "core.h" -static inline u32 own_zone(void) -{ - return tipc_zone(tipc_own_addr); -} +#define TIPC_ZONE_MASK 0xff000000u +#define TIPC_CLUSTER_MASK 0xfffff000u -static inline int in_own_cluster(u32 addr) +static inline u32 tipc_zone_mask(u32 addr) { - return !((addr ^ tipc_own_addr) >> 12); + return addr & TIPC_ZONE_MASK; } -static inline int is_slave(u32 addr) +static inline u32 tipc_cluster_mask(u32 addr) { - return addr & 0x800; + return addr & TIPC_CLUSTER_MASK; } -static inline int may_route(u32 addr) +static inline int in_own_cluster_exact(u32 addr) { - return(addr ^ tipc_own_addr) >> 11; + return !((addr ^ tipc_own_addr) >> 12); } -static inline int in_scope(u32 domain, u32 addr) +/** + * in_own_node - test for node inclusion; <0.0.0> always matches + */ +static inline int in_own_node(u32 addr) { - if (!domain || (domain == addr)) - return 1; - if (domain == (addr & 0xfffff000u)) /* domain <Z.C.0> */ - return 1; - if (domain == (addr & 0xff000000u)) /* domain <Z.0.0> */ - return 1; - return 0; + return (addr == tipc_own_addr) || !addr; } /** - * addr_scope - convert message lookup domain to equivalent 2-bit scope value + * in_own_cluster - test for cluster inclusion; <0.0.0> always matches */ - -static inline int addr_scope(u32 domain) +static inline int in_own_cluster(u32 addr) { - 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; + return in_own_cluster_exact(addr) || !addr; } /** @@ -99,25 +79,18 @@ static inline int addr_scope(u32 domain) * Needed when address of a named message must be looked up a second time * after a network hop. */ - -static inline int addr_domain(int sc) +static inline u32 addr_domain(u32 sc) { if (likely(sc == TIPC_NODE_SCOPE)) return tipc_own_addr; if (sc == TIPC_CLUSTER_SCOPE) - return tipc_addr(tipc_zone(tipc_own_addr), - tipc_cluster(tipc_own_addr), 0); - return tipc_addr(tipc_zone(tipc_own_addr), 0, 0); -} - -static inline char *addr_string_fill(char *string, u32 addr) -{ - snprintf(string, 16, "<%u.%u.%u>", - tipc_zone(addr), tipc_cluster(addr), tipc_node(addr)); - return string; + 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 index e7880172ef1..55c6c9d3e1c 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -3,7 +3,7 @@ * * Copyright (c) 2004-2006, Ericsson AB * Copyright (c) 2004, Intel Corporation. - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -36,34 +36,17 @@ */ #include "core.h" -#include "msg.h" -#include "dbg.h" #include "link.h" -#include "net.h" -#include "node.h" #include "port.h" -#include "addr.h" -#include "node_subscr.h" -#include "name_distr.h" -#include "bearer.h" -#include "name_table.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 BCLINK_LOG_BUF_SIZE 0 - -/* - * Loss rate for incoming broadcast frames; used to test retransmission code. - * Set to N to cause every N'th frame to be discarded; 0 => don't discard any. - */ - -#define TIPC_BCAST_LOSS_RATE 0 +#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 bcbearer_pair - a pair of bearers used by broadcast link + * struct tipc_bcbearer_pair - a pair of bearers used by broadcast link * @primary: pointer to primary bearer * @secondary: pointer to secondary bearer * @@ -71,13 +54,13 @@ * to be paired. */ -struct bcbearer_pair { - struct bearer *primary; - struct bearer *secondary; +struct tipc_bcbearer_pair { + struct tipc_bearer *primary; + struct tipc_bearer *secondary; }; /** - * struct bcbearer - bearer used by broadcast link + * 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 @@ -88,43 +71,76 @@ struct bcbearer_pair { * 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 "bc_lock". + * prevented through use of the spinlock "bclink_lock". */ - -struct bcbearer { - struct bearer bearer; - struct media media; - struct bcbearer_pair bpairs[MAX_BEARERS]; - struct bcbearer_pair bpairs_temp[TIPC_MAX_LINK_PRI + 1]; - struct node_map remains; - struct node_map remains_new; +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 bclink - link used for broadcast messages + * 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 bclink { - struct link link; - struct node node; +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; -static struct bcbearer *bcbearer = NULL; -static struct bclink *bclink = NULL; -static struct link *bcl = NULL; -static DEFINE_SPINLOCK(bc_lock); + if (likely(!bclink->flags)) { + spin_unlock_bh(&bclink->lock); + return; + } -char tipc_bclink_name[] = "multicast-link"; + 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); +} -static u32 buf_seqno(struct sk_buff *buf) +void tipc_bclink_set_flags(unsigned int flags) { - return msg_seqno(buf_msg(buf)); + bclink->flags |= flags; } static u32 bcbuf_acks(struct sk_buff *buf) @@ -142,55 +158,64 @@ 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(); +} -/** - * bclink_set_gap - set gap according to contents of current deferred pkt queue - * - * Called with 'node' locked, bc_lock unlocked - */ +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); +} -static void bclink_set_gap(struct node *n_ptr) +u32 tipc_bclink_get_last_sent(void) { - struct sk_buff *buf = n_ptr->bclink.deferred_head; + return bcl->fsm_msg_cnt; +} - n_ptr->bclink.gap_after = n_ptr->bclink.gap_to = - mod(n_ptr->bclink.last_in); - if (unlikely(buf != NULL)) - n_ptr->bclink.gap_to = mod(buf_seqno(buf) - 1); +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; } + /** - * bclink_ack_allowed - test if ACK or NACK message can be sent at this moment - * - * This mechanism endeavours to prevent all nodes in network from trying - * to ACK or NACK at the same time. + * tipc_bclink_retransmit_to - get most recent node to request retransmission * - * Note: TIPC uses a different trigger to distribute ACKs than it does to - * distribute NACKs, but tries to use the same spacing (divide by 16). + * Called with bclink_lock locked */ - -static int bclink_ack_allowed(u32 n) +struct tipc_node *tipc_bclink_retransmit_to(void) { - return((n % TIPC_MIN_LINK_WIN) == tipc_own_tag); + 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 bc_lock locked + * 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)) { + while (buf && less_eq(buf_seqno(buf), after)) buf = buf->next; - } tipc_link_retransmit(bcl, buf, mod(to - after)); } @@ -199,36 +224,63 @@ static void bclink_retransmit_pkt(u32 after, u32 to) * @n_ptr: node that sent acknowledgement info * @acked: broadcast sequence # that has been acknowledged * - * Node is locked, bc_lock unlocked. + * Node is locked, bclink_lock unlocked. */ - -void tipc_bclink_acknowledge(struct node *n_ptr, u32 acked) +void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) { struct sk_buff *crs; struct sk_buff *next; unsigned int released = 0; - if (less_eq(acked, n_ptr->bclink.acked)) - return; + tipc_bclink_lock(); + /* Bail out if tx queue is empty (no clean up is required) */ + crs = bcl->first_out; + if (!crs) + goto exit; - spin_lock_bh(&bc_lock); + /* 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 */ - - crs = bcl->first_out; - while (crs && less_eq(buf_seqno(crs), n_ptr->bclink.acked)) { + 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; - bcbuf_decr_acks(crs); + + 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--; - buf_discard(crs); + kfree_skb(crs); released = 1; } crs = next; @@ -237,310 +289,311 @@ void tipc_bclink_acknowledge(struct node *n_ptr, u32 acked) /* Try resolving broadcast link congestion, if necessary */ - if (unlikely(bcl->next_out)) + 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); - spin_unlock_bh(&bc_lock); +exit: + tipc_bclink_unlock(); } /** - * bclink_send_ack - unicast an ACK msg + * tipc_bclink_update_link_state - update broadcast link state * - * tipc_net_lock and node lock set + * RCU and node lock set */ - -static void bclink_send_ack(struct node *n_ptr) +void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) { - struct link *l_ptr = n_ptr->active_links[n_ptr->addr & 1]; + struct sk_buff *buf; - if (l_ptr != NULL) - tipc_link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); -} + /* Ignore "stale" link state info */ -/** - * bclink_send_nack- broadcast a NACK msg - * - * tipc_net_lock and node lock set - */ + if (less_eq(last_sent, n_ptr->bclink.last_in)) + return; -static void bclink_send_nack(struct node *n_ptr) -{ - struct sk_buff *buf; - struct tipc_msg *msg; + /* Update link synchronization state; quit if in sync */ - if (!less(n_ptr->bclink.gap_after, n_ptr->bclink.gap_to)) - return; + bclink_update_last_sent(n_ptr, last_sent); - buf = buf_acquire(INT_H_SIZE); - if (buf) { - msg = buf_msg(buf); - msg_init(msg, BCAST_PROTOCOL, STATE_MSG, - TIPC_OK, INT_H_SIZE, n_ptr->addr); - msg_set_mc_netid(msg, tipc_net_id); - msg_set_bcast_ack(msg, mod(n_ptr->bclink.last_in)); - msg_set_bcgap_after(msg, n_ptr->bclink.gap_after); - msg_set_bcgap_to(msg, n_ptr->bclink.gap_to); - msg_set_bcast_tag(msg, tipc_own_tag); - - if (tipc_bearer_send(&bcbearer->bearer, buf, NULL)) { - bcl->stats.sent_nacks++; - buf_discard(buf); - } else { - tipc_bearer_schedule(bcl->b_ptr, bcl); - bcl->proto_msg_queue = buf; - bcl->stats.bearer_congs++; - } + if (n_ptr->bclink.last_sent == n_ptr->bclink.last_in) + return; - /* - * Ensure we doesn't send another NACK msg to the node - * until 16 more deferred messages arrive from it - * (i.e. helps prevent all nodes from NACK'ing at same time) - */ + /* Update out-of-sync state; quit if loss is still unconfirmed */ - n_ptr->bclink.nack_sync = tipc_own_tag; + if ((++n_ptr->bclink.oos_state) == 1) { + if (n_ptr->bclink.deferred_size < (TIPC_MIN_LINK_WIN / 2)) + return; + n_ptr->bclink.oos_state++; } -} -/** - * tipc_bclink_check_gap - send a NACK if a sequence gap exists - * - * tipc_net_lock and node lock set - */ + /* Don't NACK if one has been recently sent (or seen) */ -void tipc_bclink_check_gap(struct node *n_ptr, u32 last_sent) -{ - if (!n_ptr->bclink.supported || - less_eq(last_sent, mod(n_ptr->bclink.last_in))) + if (n_ptr->bclink.oos_state & 0x1) return; - bclink_set_gap(n_ptr); - if (n_ptr->bclink.gap_after == n_ptr->bclink.gap_to) - n_ptr->bclink.gap_to = last_sent; - bclink_send_nack(n_ptr); + /* 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++; + } } /** - * tipc_bclink_peek_nack - process a NACK msg meant for another node + * bclink_peek_nack - monitor retransmission requests sent by other nodes * - * Only tipc_net_lock set. + * 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 tipc_bclink_peek_nack(u32 dest, u32 sender_tag, u32 gap_after, u32 gap_to) +static void bclink_peek_nack(struct tipc_msg *msg) { - struct node *n_ptr = tipc_node_find(dest); - u32 my_after, my_to; + struct tipc_node *n_ptr = tipc_node_find(msg_destnode(msg)); - if (unlikely(!n_ptr || !tipc_node_is_up(n_ptr))) + if (unlikely(!n_ptr)) return; + tipc_node_lock(n_ptr); - /* - * Modify gap to suppress unnecessary NACKs from this node - */ - my_after = n_ptr->bclink.gap_after; - my_to = n_ptr->bclink.gap_to; - - if (less_eq(gap_after, my_after)) { - if (less(my_after, gap_to) && less(gap_to, my_to)) - n_ptr->bclink.gap_after = gap_to; - else if (less_eq(my_to, gap_to)) - n_ptr->bclink.gap_to = n_ptr->bclink.gap_after; - } else if (less_eq(gap_after, my_to)) { - if (less_eq(my_to, gap_to)) - n_ptr->bclink.gap_to = gap_after; - } else { - /* - * Expand gap if missing bufs not in deferred queue: - */ - struct sk_buff *buf = n_ptr->bclink.deferred_head; - u32 prev = n_ptr->bclink.gap_to; - for (; buf; buf = buf->next) { - u32 seqno = buf_seqno(buf); + 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; - if (mod(seqno - prev) != 1) { - buf = NULL; - break; - } - if (seqno == gap_after) - break; - prev = seqno; - } - if (buf == NULL) - n_ptr->bclink.gap_to = gap_after; - } - /* - * Some nodes may send a complementary NACK now: - */ - if (bclink_ack_allowed(sender_tag + 1)) { - if (n_ptr->bclink.gap_to != n_ptr->bclink.gap_after) { - bclink_send_nack(n_ptr); - bclink_set_gap(n_ptr); - } - } tipc_node_unlock(n_ptr); } -/** - * tipc_bclink_send_msg - broadcast a packet to all nodes in cluster +/* + * tipc_bclink_xmit - broadcast a packet to all nodes in cluster */ - -int tipc_bclink_send_msg(struct sk_buff *buf) +int tipc_bclink_xmit(struct sk_buff *buf) { int res; - spin_lock_bh(&bc_lock); - - res = tipc_link_send_buf(bcl, buf); - if (unlikely(res == -ELINKCONG)) - buf_discard(buf); - else - bcl->stats.sent_info++; + tipc_bclink_lock(); - if (bcl->out_queue_size > bcl->stats.max_queue_sz) - bcl->stats.max_queue_sz = bcl->out_queue_size; - bcl->stats.queue_sz_counts++; - bcl->stats.accu_queue_sz += bcl->out_queue_size; + if (!bclink->bcast_nodes.count) { + res = msg_data_sz(buf_msg(buf)); + kfree_skb(buf); + goto exit; + } - spin_unlock_bh(&bc_lock); + 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; } /** - * tipc_bclink_recv_pkt - receive a broadcast packet, and deliver upwards + * bclink_accept_pkt - accept an incoming, in-sequence broadcast packet * - * tipc_net_lock is read_locked, no other locks set + * 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++; -void tipc_bclink_recv_pkt(struct sk_buff *buf) + /* + * 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) { -#if (TIPC_BCAST_LOSS_RATE) - static int rx_count = 0; -#endif struct tipc_msg *msg = buf_msg(buf); - struct node* node = tipc_node_find(msg_prevnode(msg)); + struct tipc_node *node; u32 next_in; u32 seqno; - struct sk_buff *deferred; + int deferred; - msg_dbg(msg, "<BC<<<"); + /* Screen out unwanted broadcast messages */ - if (unlikely(!node || !tipc_node_is_up(node) || !node->bclink.supported || - (msg_mc_netid(msg) != tipc_net_id))) { - buf_discard(buf); - return; - } + 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)) { - msg_dbg(msg, "<BCNACK<<<"); + if (msg_type(msg) != STATE_MSG) + goto unlock; if (msg_destnode(msg) == tipc_own_addr) { - tipc_node_lock(node); tipc_bclink_acknowledge(node, msg_bcast_ack(msg)); tipc_node_unlock(node); - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bcl->stats.recv_nacks++; - bcl->owner->next = node; /* remember requestor */ + bclink->retransmit_to = node; bclink_retransmit_pkt(msg_bcgap_after(msg), msg_bcgap_to(msg)); - bcl->owner->next = NULL; - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } else { - tipc_bclink_peek_nack(msg_destnode(msg), - msg_bcast_tag(msg), - msg_bcgap_after(msg), - msg_bcgap_to(msg)); + tipc_node_unlock(node); + bclink_peek_nack(msg); } - buf_discard(buf); - return; + goto exit; } -#if (TIPC_BCAST_LOSS_RATE) - if (++rx_count == TIPC_BCAST_LOSS_RATE) { - rx_count = 0; - buf_discard(buf); - return; - } -#endif + /* Handle in-sequence broadcast message */ - tipc_node_lock(node); -receive: - deferred = node->bclink.deferred_head; - next_in = mod(node->bclink.last_in + 1); seqno = msg_seqno(msg); + next_in = mod(node->bclink.last_in + 1); if (likely(seqno == next_in)) { - bcl->stats.recv_info++; - node->bclink.last_in++; - bclink_set_gap(node); - if (unlikely(bclink_ack_allowed(seqno))) { - bclink_send_ack(node); - bcl->stats.sent_acks++; - } +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); - tipc_port_recv_mcast(buf, NULL); + 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_recv_bundle(buf); + 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 (tipc_link_recv_fragment(&node->bclink.defragm, - &buf, &msg)) + 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_net_route_msg(buf); + tipc_named_rcv(buf); } else { + tipc_bclink_lock(); + bclink_accept_pkt(node, seqno); + tipc_bclink_unlock(); tipc_node_unlock(node); - tipc_net_route_msg(buf); + kfree_skb(buf); } - if (deferred && (buf_seqno(deferred) == mod(next_in + 1))) { - tipc_node_lock(node); - buf = deferred; - msg = buf_msg(buf); - node->bclink.deferred_head = deferred->next; - goto receive; - } - return; - } else if (less(next_in, seqno)) { - u32 gap_after = node->bclink.gap_after; - u32 gap_to = node->bclink.gap_to; - - if (tipc_link_defer_pkt(&node->bclink.deferred_head, - &node->bclink.deferred_tail, - buf)) { - node->bclink.nack_sync++; - bcl->stats.deferred_recv++; - if (seqno == mod(gap_after + 1)) - node->bclink.gap_after = seqno; - else if (less(gap_after, seqno) && less(seqno, gap_to)) - node->bclink.gap_to = seqno; - } - if (bclink_ack_allowed(node->bclink.nack_sync)) { - if (gap_to != gap_after) - bclink_send_nack(node); - bclink_set_gap(node); + 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; } - } else { - bcl->stats.duplicates++; - buf_discard(buf); + + 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; } - tipc_node_unlock(node); -} -u32 tipc_bclink_get_last_sent(void) -{ - u32 last_sent = mod(bcl->next_out_no - 1); + /* Handle out-of-sequence broadcast message */ - if (bcl->next_out) - last_sent = mod(buf_seqno(bcl->next_out) - 1); - return last_sent; + 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 node *n_ptr) +u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr) { - return (n_ptr->bclink.supported && + return (n_ptr->bclink.recv_permitted && (tipc_bclink_get_last_sent() != n_ptr->bclink.acked)); } @@ -548,102 +601,106 @@ u32 tipc_bclink_acks_missing(struct node *n_ptr) /** * tipc_bcbearer_send - send a packet through the broadcast pseudo-bearer * - * Send through as many bearers as necessary to reach all nodes - * that support TIPC multicasting. + * Send packet over as many bearers as necessary to reach all nodes + * that have joined the broadcast link. * - * Returns 0 if packet sent successfully, non-zero if not + * 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, +static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, struct tipc_media_addr *unused2) { - static int send_count = 0; - int bp_index; - int swap_time; - - /* Prepare buffer for broadcasting (if first time trying to send it) */ + /* 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; - assert(tipc_cltr_bcast_nodes.count != 0); - bcbuf_set_acks(buf, tipc_cltr_bcast_nodes.count); + bcbuf_set_acks(buf, bclink->bcast_nodes.count); msg = buf_msg(buf); - msg_set_non_seq(msg); + msg_set_non_seq(msg, 1); msg_set_mc_netid(msg, tipc_net_id); - } - - /* Determine if bearer pairs should be swapped following this attempt */ + bcl->stats.sent_info++; - if ((swap_time = (++send_count >= 10))) - send_count = 0; + if (WARN_ON(!bclink->bcast_nodes.count)) { + dump_stack(); + return 0; + } + } /* Send buffer over bearers until all targets reached */ - - bcbearer->remains = tipc_cltr_bcast_nodes; + bcbearer->remains = bclink->bcast_nodes; for (bp_index = 0; bp_index < MAX_BEARERS; bp_index++) { - struct bearer *p = bcbearer->bpairs[bp_index].primary; - struct bearer *s = bcbearer->bpairs[bp_index].secondary; + 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 */ + break; /* No more bearers to try */ - tipc_nmap_diff(&bcbearer->remains, &p->nodes, &bcbearer->remains_new); + tipc_nmap_diff(&bcbearer->remains, &b->nodes, + &bcbearer->remains_new); if (bcbearer->remains_new.count == bcbearer->remains.count) - continue; /* bearer pair doesn't add anything */ + continue; /* Nothing added by bearer pair */ - if (!p->publ.blocked && - !p->media->send_msg(buf, &p->publ, &p->media->bcast_addr)) { - if (swap_time && s && !s->publ.blocked) - goto swap; - else - goto update; + 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 (!s || s->publ.blocked || - s->media->send_msg(buf, &s->publ, &s->media->bcast_addr)) - continue; /* unable to send using bearer pair */ -swap: - bcbearer->bpairs[bp_index].primary = s; - bcbearer->bpairs[bp_index].secondary = p; -update: if (bcbearer->remains_new.count == 0) - return TIPC_OK; + break; /* All targets reached */ bcbearer->remains = bcbearer->remains_new; } - /* Unable to reach all targets */ - - bcbearer->bearer.publ.blocked = 1; - bcl->stats.bearer_congs++; - return ~TIPC_OK; + return 0; } /** * tipc_bcbearer_sort - create sets of bearer pairs used by broadcast bearer */ - -void tipc_bcbearer_sort(void) +void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action) { - struct bcbearer_pair *bp_temp = bcbearer->bpairs_temp; - struct bcbearer_pair *bp_curr; + struct tipc_bcbearer_pair *bp_temp = bcbearer->bpairs_temp; + struct tipc_bcbearer_pair *bp_curr; + struct tipc_bearer *b; int b_index; int pri; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); - /* Group bearers by priority (can assume max of two per priority) */ + 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++) { - struct bearer *b = &tipc_bearers[b_index]; - - if (!b->active || !b->nodes.count) + b = rcu_dereference_rtnl(bearer_list[b_index]); + if (!b || !b->nodes.count) continue; if (!bp_temp[b->priority].primary) @@ -651,9 +708,9 @@ void tipc_bcbearer_sort(void) 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)); @@ -677,75 +734,49 @@ void tipc_bcbearer_sort(void) bp_curr++; } - spin_unlock_bh(&bc_lock); -} - -/** - * tipc_bcbearer_push - resolve bearer congestion - * - * Forces bclink to push out any unsent packets, until all packets are gone - * or congestion reoccurs. - * No locks set when function called - */ - -void tipc_bcbearer_push(void) -{ - struct bearer *b_ptr; - - spin_lock_bh(&bc_lock); - b_ptr = &bcbearer->bearer; - if (b_ptr->publ.blocked) { - b_ptr->publ.blocked = 0; - tipc_bearer_lock_push(b_ptr); - } - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } int tipc_bclink_stats(char *buf, const u32 buf_size) { - struct print_buf pb; + int ret; + struct tipc_stats *s; if (!bcl) return 0; - tipc_printbuf_init(&pb, buf, buf_size); - - spin_lock_bh(&bc_lock); - - tipc_printf(&pb, "Link <%s>\n" - " Window:%u packets\n", - bcl->name, bcl->queue_limit[0]); - tipc_printf(&pb, " RX packets:%u fragments:%u/%u bundles:%u/%u\n", - bcl->stats.recv_info, - bcl->stats.recv_fragments, - bcl->stats.recv_fragmented, - bcl->stats.recv_bundles, - bcl->stats.recv_bundled); - tipc_printf(&pb, " TX packets:%u fragments:%u/%u bundles:%u/%u\n", - bcl->stats.sent_info, - bcl->stats.sent_fragments, - bcl->stats.sent_fragmented, - bcl->stats.sent_bundles, - bcl->stats.sent_bundled); - tipc_printf(&pb, " RX naks:%u defs:%u dups:%u\n", - bcl->stats.recv_nacks, - bcl->stats.deferred_recv, - bcl->stats.duplicates); - tipc_printf(&pb, " TX naks:%u acks:%u dups:%u\n", - bcl->stats.sent_nacks, - bcl->stats.sent_acks, - bcl->stats.retransmitted); - tipc_printf(&pb, " Congestion bearer:%u link:%u Send queue max:%u avg:%u\n", - bcl->stats.bearer_congs, - bcl->stats.link_congs, - bcl->stats.max_queue_sz, - bcl->stats.queue_sz_counts - ? (bcl->stats.accu_queue_sz / bcl->stats.queue_sz_counts) - : 0); - - spin_unlock_bh(&bc_lock); - return tipc_printbuf_validate(&pb); + 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) @@ -753,10 +784,10 @@ int tipc_bclink_reset_stats(void) if (!bcl) return -ENOPROTOOPT; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); memset(&bcl->stats, 0, sizeof(bcl->stats)); - spin_unlock_bh(&bc_lock); - return TIPC_OK; + tipc_bclink_unlock(); + return 0; } int tipc_bclink_set_queue_limits(u32 limit) @@ -766,66 +797,156 @@ int tipc_bclink_set_queue_limits(u32 limit) if ((limit < TIPC_MIN_LINK_WIN) || (limit > TIPC_MAX_LINK_WIN)) return -EINVAL; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); tipc_link_set_queue_limits(bcl, limit); - spin_unlock_bh(&bc_lock); - return TIPC_OK; + 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 (!bcbearer || !bclink) { - nomem: - warn("Multicast link creation failed, no memory\n"); + if (!bclink) { kfree(bcbearer); - bcbearer = NULL; - kfree(bclink); - bclink = NULL; return -ENOMEM; } - INIT_LIST_HEAD(&bcbearer->bearer.cong_links); + bcl = &bclink->link; bcbearer->bearer.media = &bcbearer->media; bcbearer->media.send_msg = tipc_bcbearer_send; - sprintf(bcbearer->media.name, "tipc-multicast"); + sprintf(bcbearer->media.name, "tipc-broadcast"); - bcl = &bclink->link; + 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->b_ptr = &bcbearer->bearer; + bcl->bearer_id = MAX_BEARERS; + rcu_assign_pointer(bearer_list[MAX_BEARERS], &bcbearer->bearer); bcl->state = WORKING_WORKING; - sprintf(bcl->name, tipc_bclink_name); + strlcpy(bcl->name, tipc_bclink_name, TIPC_MAX_LINK_NAME); + return 0; +} - if (BCLINK_LOG_BUF_SIZE) { - char *pb = kmalloc(BCLINK_LOG_BUF_SIZE, GFP_ATOMIC); +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); +} - if (!pb) - goto nomem; - tipc_printbuf_init(&bcl->print_buf, pb, BCLINK_LOG_BUF_SIZE); +/** + * 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)); - return TIPC_OK; + if ((nm_ptr->map[w] & mask) != 0) { + nm_ptr->map[w] &= ~mask; + nm_ptr->count--; + } } -void tipc_bclink_stop(void) +/** + * 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) { - spin_lock_bh(&bc_lock); - if (bcbearer) { - tipc_link_stop(bcl); - if (BCLINK_LOG_BUF_SIZE) - kfree(bcl->print_buf.buf); - bcl = NULL; - kfree(bclink); - bclink = NULL; - kfree(bcbearer); - bcbearer = NULL; + 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++; + } + } } - spin_unlock_bh(&bc_lock); } +/** + * 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 index a2416fa6b90..00330c45df3 100644 --- a/net/tipc/bcast.h +++ b/net/tipc/bcast.h @@ -2,7 +2,7 @@ * net/tipc/bcast.h: Include file for TIPC broadcast code * * Copyright (c) 2003-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -39,172 +39,64 @@ #define MAX_NODES 4096 #define WSIZE 32 +#define TIPC_BCLINK_RESET 1 /** - * struct node_map - set of node identifiers + * struct tipc_node_map - set of node identifiers * @count: # of nodes in set * @map: bitmap of node identifiers that are in the set */ - -struct node_map { +struct tipc_node_map { u32 count; u32 map[MAX_NODES / WSIZE]; }; - #define PLSIZE 32 /** - * struct port_list - set of node local destination ports + * 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 port_list { +struct tipc_port_list { int count; - struct port_list *next; + struct tipc_port_list *next; u32 ports[PLSIZE]; }; -struct node; - -extern char tipc_bclink_name[]; +struct tipc_node; +extern const char tipc_bclink_name[]; /** - * nmap_add - add a node to a node map + * tipc_nmap_equal - test for equality of node maps */ - -static inline void tipc_nmap_add(struct 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; - } -} - -/** - * nmap_remove - remove a node from a node map - */ - -static inline void tipc_nmap_remove(struct 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--; - } -} - -/** - * nmap_equal - test for equality of node maps - */ - -static inline int tipc_nmap_equal(struct node_map *nm_a, struct node_map *nm_b) +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)); } -/** - * 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 inline void tipc_nmap_diff(struct node_map *nm_a, struct node_map *nm_b, - struct node_map *nm_diff) -{ - int stop = sizeof(nm_a->map) / sizeof(u32); - 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++; - } - } - } -} - -/** - * port_list_add - add a port to a port list, ensuring no duplicates - */ - -static inline void tipc_port_list_add(struct port_list *pl_ptr, u32 port) -{ - struct 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) { - warn("Incomplete multicast delivery, no memory\n"); - return; - } - item->next->next = NULL; - } - } -} - -/** - * port_list_free - free dynamically created entries in port_list chain - * - * Note: First item is on stack, so it doesn't need to be released - */ - -static inline void tipc_port_list_free(struct port_list *pl_ptr) -{ - struct port_list *item; - struct port_list *next; - - for (item = pl_ptr->next; item; item = next) { - next = item->next; - kfree(item); - } -} - +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); +int tipc_bclink_init(void); void tipc_bclink_stop(void); -void tipc_bclink_acknowledge(struct node *n_ptr, u32 acked); -int tipc_bclink_send_msg(struct sk_buff *buf); -void tipc_bclink_recv_pkt(struct sk_buff *buf); +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 node *n_ptr); -void tipc_bclink_check_gap(struct node *n_ptr, u32 seqno); +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(void); -void tipc_bcbearer_push(void); +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 index 271a375b49b..264474394f9 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -1,8 +1,8 @@ /* * net/tipc/bearer.c: TIPC bearer code * - * Copyright (c) 1996-2006, Ericsson AB - * Copyright (c) 2004-2006, Wind River Systems + * 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 @@ -36,206 +36,103 @@ #include "core.h" #include "config.h" -#include "dbg.h" #include "bearer.h" -#include "link.h" -#include "port.h" #include "discover.h" -#include "bcast.h" -#define MAX_ADDR_STR 32 +#define MAX_ADDR_STR 60 -static struct media *media_list = NULL; -static u32 media_count = 0; +static struct tipc_media * const media_info_array[] = { + ð_media_info, +#ifdef CONFIG_TIPC_MEDIA_IB + &ib_media_info, +#endif + NULL +}; -struct bearer *tipc_bearers = NULL; +struct tipc_bearer __rcu *bearer_list[MAX_BEARERS + 1]; -/** - * media_name_valid - validate media name - * - * Returns 1 if media name is valid, otherwise 0. - */ - -static int media_name_valid(const char *name) -{ - u32 len; - - len = strlen(name); - if ((len + 1) > TIPC_MAX_MEDIA_NAME) - return 0; - return (strspn(name, tipc_alphabet) == len); -} +static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down); /** - * media_find - locates specified media object by name + * tipc_media_find - locates specified media object by name */ - -static struct media *media_find(const char *name) +struct tipc_media *tipc_media_find(const char *name) { - struct media *m_ptr; u32 i; - for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { - if (!strcmp(m_ptr->name, name)) - return m_ptr; + for (i = 0; media_info_array[i] != NULL; i++) { + if (!strcmp(media_info_array[i]->name, name)) + break; } - return NULL; + return media_info_array[i]; } /** - * tipc_register_media - register a media type - * - * Bearers for this media type must be activated separately at a later stage. + * media_find_id - locates specified media object by type identifier */ - -int tipc_register_media(u32 media_type, - char *name, - int (*enable)(struct tipc_bearer *), - void (*disable)(struct tipc_bearer *), - int (*send_msg)(struct sk_buff *, - struct tipc_bearer *, - struct tipc_media_addr *), - char *(*addr2str)(struct tipc_media_addr *a, - char *str_buf, int str_size), - struct tipc_media_addr *bcast_addr, - const u32 bearer_priority, - const u32 link_tolerance, /* [ms] */ - const u32 send_window_limit) +static struct tipc_media *media_find_id(u8 type) { - struct media *m_ptr; - u32 media_id; u32 i; - int res = -EINVAL; - - write_lock_bh(&tipc_net_lock); - if (!media_list) - goto exit; - - if (!media_name_valid(name)) { - warn("Media <%s> rejected, illegal name\n", name); - goto exit; - } - if (!bcast_addr) { - warn("Media <%s> rejected, no broadcast address\n", name); - goto exit; - } - if ((bearer_priority < TIPC_MIN_LINK_PRI) && - (bearer_priority > TIPC_MAX_LINK_PRI)) { - warn("Media <%s> rejected, illegal priority (%u)\n", name, - bearer_priority); - goto exit; - } - if ((link_tolerance < TIPC_MIN_LINK_TOL) || - (link_tolerance > TIPC_MAX_LINK_TOL)) { - warn("Media <%s> rejected, illegal tolerance (%u)\n", name, - link_tolerance); - goto exit; - } - media_id = media_count++; - if (media_id >= MAX_MEDIA) { - warn("Media <%s> rejected, media limit reached (%u)\n", name, - MAX_MEDIA); - media_count--; - goto exit; - } - for (i = 0; i < media_id; i++) { - if (media_list[i].type_id == media_type) { - warn("Media <%s> rejected, duplicate type (%u)\n", name, - media_type); - media_count--; - goto exit; - } - if (!strcmp(name, media_list[i].name)) { - warn("Media <%s> rejected, duplicate name\n", name); - media_count--; - goto exit; - } + for (i = 0; media_info_array[i] != NULL; i++) { + if (media_info_array[i]->type_id == type) + break; } - - m_ptr = &media_list[media_id]; - m_ptr->type_id = media_type; - m_ptr->send_msg = send_msg; - m_ptr->enable_bearer = enable; - m_ptr->disable_bearer = disable; - m_ptr->addr2str = addr2str; - memcpy(&m_ptr->bcast_addr, bcast_addr, sizeof(*bcast_addr)); - m_ptr->bcast = 1; - strcpy(m_ptr->name, name); - m_ptr->priority = bearer_priority; - m_ptr->tolerance = link_tolerance; - m_ptr->window = send_window_limit; - dbg("Media <%s> registered\n", name); - res = 0; -exit: - write_unlock_bh(&tipc_net_lock); - return res; + return media_info_array[i]; } /** * tipc_media_addr_printf - record media address in print buffer */ - -void tipc_media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a) +void tipc_media_addr_printf(char *buf, int len, struct tipc_media_addr *a) { - struct media *m_ptr; - u32 media_type; - u32 i; - - media_type = ntohl(a->type); - for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { - if (m_ptr->type_id == media_type) - break; - } + char addr_str[MAX_ADDR_STR]; + struct tipc_media *m_ptr; + int ret; - if ((i < media_count) && (m_ptr->addr2str != NULL)) { - char addr_str[MAX_ADDR_STR]; + m_ptr = media_find_id(a->media_id); - tipc_printf(pb, "%s(%s)", m_ptr->name, - m_ptr->addr2str(a, addr_str, sizeof(addr_str))); - } else { - unchar *addr = (unchar *)&a->dev_addr; + 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; - tipc_printf(pb, "UNKNOWN(%u)", media_type); - for (i = 0; i < (sizeof(*a) - sizeof(a->type)); i++) { - tipc_printf(pb, "-%02x", addr[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; - struct media *m_ptr; int i; buf = tipc_cfg_reply_alloc(MAX_MEDIA * TLV_SPACE(TIPC_MAX_MEDIA_NAME)); if (!buf) return NULL; - read_lock_bh(&tipc_net_lock); - for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { - tipc_cfg_append_tlv(buf, TIPC_TLV_MEDIA_NAME, m_ptr->name, - strlen(m_ptr->name) + 1); + 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); } - read_unlock_bh(&tipc_net_lock); 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) + * @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 bearer_name *name_parts) + struct tipc_bearer_names *name_parts) { char name_copy[TIPC_MAX_BEARER_NAME]; char *media_name; @@ -244,7 +141,6 @@ static int bearer_name_validate(const char *name, 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); @@ -252,24 +148,20 @@ static int bearer_name_validate(const char *name, return 0; /* ensure all component parts of bearer name are present */ - media_name = name_copy; - if ((if_name = strchr(media_name, ':')) == NULL) + 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) || - (strspn(media_name, tipc_alphabet) != (media_len - 1)) || - (strspn(if_name, tipc_alphabet) != (if_len - 1))) + (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); @@ -278,39 +170,16 @@ static int bearer_name_validate(const char *name, } /** - * bearer_find - locates bearer object with matching bearer name + * tipc_bearer_find - locates bearer object with matching bearer name */ - -static struct bearer *bearer_find(const char *name) +struct tipc_bearer *tipc_bearer_find(const char *name) { - struct bearer *b_ptr; + struct tipc_bearer *b_ptr; u32 i; - if (tipc_mode != TIPC_NET_MODE) - return NULL; - - for (i = 0, b_ptr = tipc_bearers; i < MAX_BEARERS; i++, b_ptr++) { - if (b_ptr->active && (!strcmp(b_ptr->publ.name, name))) - return b_ptr; - } - return NULL; -} - -/** - * tipc_bearer_find_interface - locates bearer object with matching interface name - */ - -struct bearer *tipc_bearer_find_interface(const char *if_name) -{ - struct bearer *b_ptr; - char *b_if_name; - u32 i; - - for (i = 0, b_ptr = tipc_bearers; i < MAX_BEARERS; i++, b_ptr++) { - if (!b_ptr->active) - continue; - b_if_name = strchr(b_ptr->publ.name, ':') + 1; - if (!strcmp(b_if_name, if_name)) + 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; @@ -319,196 +188,104 @@ struct bearer *tipc_bearer_find_interface(const char *if_name) /** * tipc_bearer_get_names - record names of bearers in buffer */ - struct sk_buff *tipc_bearer_get_names(void) { struct sk_buff *buf; - struct media *m_ptr; - struct bearer *b_ptr; + struct tipc_bearer *b; int i, j; buf = tipc_cfg_reply_alloc(MAX_BEARERS * TLV_SPACE(TIPC_MAX_BEARER_NAME)); if (!buf) return NULL; - read_lock_bh(&tipc_net_lock); - for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { + for (i = 0; media_info_array[i] != NULL; i++) { for (j = 0; j < MAX_BEARERS; j++) { - b_ptr = &tipc_bearers[j]; - if (b_ptr->active && (b_ptr->media == m_ptr)) { + 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_ptr->publ.name, - strlen(b_ptr->publ.name) + 1); + b->name, + strlen(b->name) + 1); } } } - read_unlock_bh(&tipc_net_lock); return buf; } -void tipc_bearer_add_dest(struct bearer *b_ptr, u32 dest) -{ - tipc_nmap_add(&b_ptr->nodes, dest); - tipc_disc_update_link_req(b_ptr->link_req); - tipc_bcbearer_sort(); -} - -void tipc_bearer_remove_dest(struct bearer *b_ptr, u32 dest) -{ - tipc_nmap_remove(&b_ptr->nodes, dest); - tipc_disc_update_link_req(b_ptr->link_req); - tipc_bcbearer_sort(); -} - -/* - * bearer_push(): Resolve bearer congestion. Force the waiting - * links to push out their unsent packets, one packet per link - * per iteration, until all packets are gone or congestion reoccurs. - * 'tipc_net_lock' is read_locked when this function is called - * bearer.lock must be taken before calling - * Returns binary true(1) ore false(0) - */ -static int bearer_push(struct bearer *b_ptr) +void tipc_bearer_add_dest(u32 bearer_id, u32 dest) { - u32 res = TIPC_OK; - struct link *ln, *tln; - - if (b_ptr->publ.blocked) - return 0; + struct tipc_bearer *b_ptr; - while (!list_empty(&b_ptr->cong_links) && (res != PUSH_FAILED)) { - list_for_each_entry_safe(ln, tln, &b_ptr->cong_links, link_list) { - res = tipc_link_push_packet(ln); - if (res == PUSH_FAILED) - break; - if (res == PUSH_FINISHED) - list_move_tail(&ln->link_list, &b_ptr->links); - } + 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); } - return list_empty(&b_ptr->cong_links); + rcu_read_unlock(); } -void tipc_bearer_lock_push(struct bearer *b_ptr) +void tipc_bearer_remove_dest(u32 bearer_id, u32 dest) { - int res; - - spin_lock_bh(&b_ptr->publ.lock); - res = bearer_push(b_ptr); - spin_unlock_bh(&b_ptr->publ.lock); - if (res) - tipc_bcbearer_push(); -} - + struct tipc_bearer *b_ptr; -/* - * Interrupt enabling new requests after bearer congestion or blocking: - * See bearer_send(). - */ -void tipc_continue(struct tipc_bearer *tb_ptr) -{ - struct bearer *b_ptr = (struct bearer *)tb_ptr; - - spin_lock_bh(&b_ptr->publ.lock); - b_ptr->continue_count++; - if (!list_empty(&b_ptr->cong_links)) - tipc_k_signal((Handler)tipc_bearer_lock_push, (unsigned long)b_ptr); - b_ptr->publ.blocked = 0; - spin_unlock_bh(&b_ptr->publ.lock); -} - -/* - * Schedule link for sending of messages after the bearer - * has been deblocked by 'continue()'. This method is called - * when somebody tries to send a message via this link while - * the bearer is congested. 'tipc_net_lock' is in read_lock here - * bearer.lock is busy - */ - -static void tipc_bearer_schedule_unlocked(struct bearer *b_ptr, struct link *l_ptr) -{ - list_move_tail(&l_ptr->link_list, &b_ptr->cong_links); -} - -/* - * Schedule link for sending of messages after the bearer - * has been deblocked by 'continue()'. This method is called - * when somebody tries to send a message via this link while - * the bearer is congested. 'tipc_net_lock' is in read_lock here, - * bearer.lock is free - */ - -void tipc_bearer_schedule(struct bearer *b_ptr, struct link *l_ptr) -{ - spin_lock_bh(&b_ptr->publ.lock); - tipc_bearer_schedule_unlocked(b_ptr, l_ptr); - spin_unlock_bh(&b_ptr->publ.lock); -} - - -/* - * tipc_bearer_resolve_congestion(): Check if there is bearer congestion, - * and if there is, try to resolve it before returning. - * 'tipc_net_lock' is read_locked when this function is called - */ -int tipc_bearer_resolve_congestion(struct bearer *b_ptr, struct link *l_ptr) -{ - int res = 1; - - if (list_empty(&b_ptr->cong_links)) - return 1; - spin_lock_bh(&b_ptr->publ.lock); - if (!bearer_push(b_ptr)) { - tipc_bearer_schedule_unlocked(b_ptr, l_ptr); - res = 0; + 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); } - spin_unlock_bh(&b_ptr->publ.lock); - return res; + rcu_read_unlock(); } - /** * tipc_enable_bearer - enable bearer with the given name */ - -int tipc_enable_bearer(const char *name, u32 bcast_scope, u32 priority) +int tipc_enable_bearer(const char *name, u32 disc_domain, u32 priority) { - struct bearer *b_ptr; - struct media *m_ptr; - struct bearer_name b_name; + 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_mode != TIPC_NET_MODE) { - warn("Bearer <%s> rejected, not supported in standalone mode\n", - name); + if (!tipc_own_addr) { + pr_warn("Bearer <%s> rejected, not supported in standalone mode\n", + name); return -ENOPROTOOPT; } - if (!bearer_name_validate(name, &b_name)) { - warn("Bearer <%s> rejected, illegal name\n", name); + if (!bearer_name_validate(name, &b_names)) { + pr_warn("Bearer <%s> rejected, illegal name\n", name); return -EINVAL; } - if (!tipc_addr_domain_valid(bcast_scope) || - !in_scope(bcast_scope, tipc_own_addr)) { - warn("Bearer <%s> rejected, illegal broadcast scope\n", name); + 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_MIN_LINK_PRI || - priority > TIPC_MAX_LINK_PRI) && + if ((priority > TIPC_MAX_LINK_PRI) && (priority != TIPC_MEDIA_LINK_PRI)) { - warn("Bearer <%s> rejected, illegal priority\n", name); + pr_warn("Bearer <%s> rejected, illegal priority\n", name); return -EINVAL; } - write_lock_bh(&tipc_net_lock); - - m_ptr = media_find(b_name.media_name); + m_ptr = tipc_media_find(b_names.media_name); if (!m_ptr) { - warn("Bearer <%s> rejected, media <%s> not registered\n", name, - b_name.media_name); - goto failed; + pr_warn("Bearer <%s> rejected, media <%s> not registered\n", + name, b_names.media_name); + return -EINVAL; } if (priority == TIPC_MEDIA_LINK_PRI) @@ -518,188 +295,335 @@ restart: bearer_id = MAX_BEARERS; with_this_prio = 1; for (i = MAX_BEARERS; i-- != 0; ) { - if (!tipc_bearers[i].active) { + b_ptr = rtnl_dereference(bearer_list[i]); + if (!b_ptr) { bearer_id = i; continue; } - if (!strcmp(name, tipc_bearers[i].publ.name)) { - warn("Bearer <%s> rejected, already enabled\n", name); - goto failed; + if (!strcmp(name, b_ptr->name)) { + pr_warn("Bearer <%s> rejected, already enabled\n", + name); + return -EINVAL; } - if ((tipc_bearers[i].priority == priority) && + if ((b_ptr->priority == priority) && (++with_this_prio > 2)) { if (priority-- == 0) { - warn("Bearer <%s> rejected, duplicate priority\n", - name); - goto failed; + pr_warn("Bearer <%s> rejected, duplicate priority\n", + name); + return -EINVAL; } - warn("Bearer <%s> priority adjustment required %u->%u\n", - name, priority + 1, priority); + pr_warn("Bearer <%s> priority adjustment required %u->%u\n", + name, priority + 1, priority); goto restart; } } if (bearer_id >= MAX_BEARERS) { - warn("Bearer <%s> rejected, bearer limit reached (%u)\n", - name, MAX_BEARERS); - goto failed; + pr_warn("Bearer <%s> rejected, bearer limit reached (%u)\n", + name, MAX_BEARERS); + return -EINVAL; } - b_ptr = &tipc_bearers[bearer_id]; - memset(b_ptr, 0, sizeof(struct bearer)); + b_ptr = kzalloc(sizeof(*b_ptr), GFP_ATOMIC); + if (!b_ptr) + return -ENOMEM; - strcpy(b_ptr->publ.name, name); - res = m_ptr->enable_bearer(&b_ptr->publ); + strcpy(b_ptr->name, name); + b_ptr->media = m_ptr; + res = m_ptr->enable_media(b_ptr); if (res) { - warn("Bearer <%s> rejected, enable failure (%d)\n", name, -res); - goto failed; + pr_warn("Bearer <%s> rejected, enable failure (%d)\n", + name, -res); + return -EINVAL; } b_ptr->identity = bearer_id; - b_ptr->media = m_ptr; + 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->active = 1; - b_ptr->detect_scope = bcast_scope; b_ptr->priority = priority; - INIT_LIST_HEAD(&b_ptr->cong_links); - INIT_LIST_HEAD(&b_ptr->links); - if (m_ptr->bcast) { - b_ptr->link_req = tipc_disc_init_link_req(b_ptr, &m_ptr->bcast_addr, - bcast_scope, 2); + + 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; } - spin_lock_init(&b_ptr->publ.lock); - write_unlock_bh(&tipc_net_lock); - info("Enabled bearer <%s>, discovery domain %s, priority %u\n", - name, addr_string_fill(addr_string, bcast_scope), priority); - return 0; -failed: - write_unlock_bh(&tipc_net_lock); + + 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_block_bearer(): Block the bearer with the given name, - * and reset all its links + * 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; +} -int tipc_block_bearer(const char *name) +/** + * bearer_disable + * + * Note: This routine assumes caller holds RTNL lock. + */ +static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down) { - struct bearer *b_ptr = NULL; - struct link *l_ptr; - struct link *temp_l_ptr; - - read_lock_bh(&tipc_net_lock); - b_ptr = bearer_find(name); - if (!b_ptr) { - warn("Attempt to block unknown bearer <%s>\n", name); - read_unlock_bh(&tipc_net_lock); - return -EINVAL; + 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); +} - info("Blocking bearer <%s>\n", name); - spin_lock_bh(&b_ptr->publ.lock); - b_ptr->publ.blocked = 1; - list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) { - struct node *n_ptr = l_ptr->owner; +int tipc_disable_bearer(const char *name) +{ + struct tipc_bearer *b_ptr; + int res; - spin_lock_bh(&n_ptr->lock); - tipc_link_reset(l_ptr); - spin_unlock_bh(&n_ptr->lock); + 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; } - spin_unlock_bh(&b_ptr->publ.lock); - read_unlock_bh(&tipc_net_lock); - return TIPC_OK; + return res; } -/** - * bearer_disable - +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 * - * Note: This routine assumes caller holds tipc_net_lock. + * 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); +} -static int bearer_disable(const char *name) +/** + * 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 bearer *b_ptr; - struct link *l_ptr; - struct link *temp_l_ptr; + struct sk_buff *clone; + struct net_device *dev; + int delta; - b_ptr = bearer_find(name); - if (!b_ptr) { - warn("Attempt to disable unknown bearer <%s>\n", name); - return -EINVAL; + 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; } - info("Disabling bearer <%s>\n", name); - tipc_disc_stop_link_req(b_ptr->link_req); - spin_lock_bh(&b_ptr->publ.lock); - b_ptr->link_req = NULL; - b_ptr->publ.blocked = 1; - if (b_ptr->media->disable_bearer) { - spin_unlock_bh(&b_ptr->publ.lock); - write_unlock_bh(&tipc_net_lock); - b_ptr->media->disable_bearer(&b_ptr->publ); - write_lock_bh(&tipc_net_lock); - spin_lock_bh(&b_ptr->publ.lock); + 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; } - list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) { - tipc_link_delete(l_ptr); + + 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; + } } - spin_unlock_bh(&b_ptr->publ.lock); - memset(b_ptr, 0, sizeof(struct bearer)); - return TIPC_OK; + rcu_read_unlock(); + + kfree_skb(buf); + return NET_RX_DROP; } -int tipc_disable_bearer(const char *name) +/** + * 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) { - int res; + struct tipc_bearer *b_ptr; + struct net_device *dev = netdev_notifier_info_to_dev(ptr); - write_lock_bh(&tipc_net_lock); - res = bearer_disable(name); - write_unlock_bh(&tipc_net_lock); - return res; + 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_init(void) +int tipc_bearer_setup(void) { - int res; + int err; - write_lock_bh(&tipc_net_lock); - tipc_bearers = kcalloc(MAX_BEARERS, sizeof(struct bearer), GFP_ATOMIC); - media_list = kcalloc(MAX_MEDIA, sizeof(struct media), GFP_ATOMIC); - if (tipc_bearers && media_list) { - res = TIPC_OK; - } else { - kfree(tipc_bearers); - kfree(media_list); - tipc_bearers = NULL; - media_list = NULL; - res = -ENOMEM; - } - write_unlock_bh(&tipc_net_lock); - return res; + 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; - if (!tipc_bearers) - return; - for (i = 0; i < MAX_BEARERS; i++) { - if (tipc_bearers[i].active) - tipc_bearers[i].publ.blocked = 1; - } - for (i = 0; i < MAX_BEARERS; i++) { - if (tipc_bearers[i].active) - bearer_disable(tipc_bearers[i].publ.name); + b_ptr = rtnl_dereference(bearer_list[i]); + if (b_ptr) { + bearer_disable(b_ptr, true); + bearer_list[i] = NULL; + } } - kfree(tipc_bearers); - kfree(media_list); - tipc_bearers = NULL; - media_list = NULL; - media_count = 0; } - - diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index 6a36b6600e6..78fccc49de2 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -1,8 +1,8 @@ /* * net/tipc/bearer.h: Include file for TIPC bearer code * - * Copyright (c) 1996-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * 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 @@ -37,134 +37,163 @@ #ifndef _TIPC_BEARER_H #define _TIPC_BEARER_H -#include "core.h" #include "bcast.h" -#define MAX_BEARERS 8 -#define MAX_MEDIA 4 +#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 media - TIPC media information available to internal users + * struct tipc_media - Media specific info exposed to generic bearer layer * @send_msg: routine which handles buffer transmission - * @enable_bearer: routine which enables a bearer - * @disable_bearer: routine which disables a bearer - * @addr2str: routine which converts bearer's address to string form - * @bcast_addr: media address used in broadcasting - * @bcast: non-zero if media supports broadcasting [currently mandatory] + * @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 [defined in tipc_bearer.h] + * @type_id: TIPC media identifier + * @hwaddr_len: TIPC media address len * @name: media name */ - -struct media { +struct tipc_media { int (*send_msg)(struct sk_buff *buf, struct tipc_bearer *b_ptr, struct tipc_media_addr *dest); - int (*enable_bearer)(struct tipc_bearer *b_ptr); - void (*disable_bearer)(struct tipc_bearer *b_ptr); - char *(*addr2str)(struct tipc_media_addr *a, - char *str_buf, int str_size); - struct tipc_media_addr bcast_addr; - int bcast; + 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 bearer - TIPC bearer information available to internal users - * @publ: bearer information available to privileged users + * 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 - * @detect_scope: network address mask used during automatic link creation + * @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 - * @links: list of non-congested links associated with bearer - * @cong_links: list of congested links associated with bearer - * @continue_count: # of times bearer has resumed after congestion or blocking - * @active: non-zero if bearer structure is represents a bearer * @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 bearer { - struct tipc_bearer publ; - struct media *media; +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 detect_scope; + u32 window; + u32 tolerance; + u32 domain; u32 identity; - struct link_req *link_req; - struct list_head links; - struct list_head cong_links; - u32 continue_count; - int active; + struct tipc_link_req *link_req; char net_plane; - struct node_map nodes; + struct tipc_node_map nodes; }; -struct bearer_name { +struct tipc_bearer_names { char media_name[TIPC_MAX_MEDIA_NAME]; char if_name[TIPC_MAX_IF_NAME]; }; -struct link; +struct tipc_link; -extern struct bearer *tipc_bearers; +extern struct tipc_bearer __rcu *bearer_list[]; -void tipc_media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a); -struct sk_buff *tipc_media_get_names(void); - -struct sk_buff *tipc_bearer_get_names(void); -void tipc_bearer_add_dest(struct bearer *b_ptr, u32 dest); -void tipc_bearer_remove_dest(struct bearer *b_ptr, u32 dest); -void tipc_bearer_schedule(struct bearer *b_ptr, struct link *l_ptr); -struct bearer *tipc_bearer_find_interface(const char *if_name); -int tipc_bearer_resolve_congestion(struct bearer *b_ptr, struct link *l_ptr); -int tipc_bearer_init(void); -void tipc_bearer_stop(void); -void tipc_bearer_lock_push(struct bearer *b_ptr); +/* + * 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); -/** - * tipc_bearer_send- sends buffer to destination over bearer - * - * Returns true (1) if successful, or false (0) if unable to send - * - * IMPORTANT: - * The media send routine must not alter the buffer being passed in - * as it may be needed for later retransmission! - * - * If the media send routine returns a non-zero value (indicating that - * it was unable to send the buffer), it must: - * 1) mark the bearer as blocked, - * 2) call tipc_continue() once the bearer is able to send again. - * Media types that are unable to meet these two critera must ensure their - * send routine always returns success -- even if the buffer was not sent -- - * and let TIPC's link code deal with the undelivered message. +/* + * Routines made available to TIPC by supported media types */ +extern struct tipc_media eth_media_info; -static inline int tipc_bearer_send(struct bearer *b_ptr, struct sk_buff *buf, - struct tipc_media_addr *dest) -{ - return !b_ptr->media->send_msg(buf, &b_ptr->publ, dest); -} +#ifdef CONFIG_TIPC_MEDIA_IB +extern struct tipc_media ib_media_info; +#endif -/** - * tipc_bearer_congested - determines if bearer is currently congested - */ +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); -static inline int tipc_bearer_congested(struct bearer *b_ptr, struct link *l_ptr) -{ - if (unlikely(b_ptr->publ.blocked)) - return 1; - if (likely(list_empty(&b_ptr->cong_links))) - return 0; - return !tipc_bearer_resolve_congestion(b_ptr, l_ptr); -} +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 +#endif /* _TIPC_BEARER_H */ diff --git a/net/tipc/cluster.c b/net/tipc/cluster.c deleted file mode 100644 index 4bb3404f610..00000000000 --- a/net/tipc/cluster.c +++ /dev/null @@ -1,576 +0,0 @@ -/* - * net/tipc/cluster.c: TIPC cluster management routines - * - * 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. - */ - -#include "core.h" -#include "cluster.h" -#include "addr.h" -#include "node_subscr.h" -#include "link.h" -#include "node.h" -#include "net.h" -#include "msg.h" -#include "bearer.h" - -static void tipc_cltr_multicast(struct cluster *c_ptr, struct sk_buff *buf, - u32 lower, u32 upper); -static struct sk_buff *tipc_cltr_prepare_routing_msg(u32 data_size, u32 dest); - -struct node **tipc_local_nodes = NULL; -struct node_map tipc_cltr_bcast_nodes = {0,{0,}}; -u32 tipc_highest_allowed_slave = 0; - -struct cluster *tipc_cltr_create(u32 addr) -{ - struct _zone *z_ptr; - struct cluster *c_ptr; - int max_nodes; - - c_ptr = kzalloc(sizeof(*c_ptr), GFP_ATOMIC); - if (c_ptr == NULL) { - warn("Cluster creation failure, no memory\n"); - return NULL; - } - - c_ptr->addr = tipc_addr(tipc_zone(addr), tipc_cluster(addr), 0); - if (in_own_cluster(addr)) - max_nodes = LOWEST_SLAVE + tipc_max_slaves; - else - max_nodes = tipc_max_nodes + 1; - - c_ptr->nodes = kcalloc(max_nodes + 1, sizeof(void*), GFP_ATOMIC); - if (c_ptr->nodes == NULL) { - warn("Cluster creation failure, no memory for node area\n"); - kfree(c_ptr); - return NULL; - } - - if (in_own_cluster(addr)) - tipc_local_nodes = c_ptr->nodes; - c_ptr->highest_slave = LOWEST_SLAVE - 1; - c_ptr->highest_node = 0; - - z_ptr = tipc_zone_find(tipc_zone(addr)); - if (!z_ptr) { - z_ptr = tipc_zone_create(addr); - } - if (!z_ptr) { - kfree(c_ptr->nodes); - kfree(c_ptr); - return NULL; - } - - tipc_zone_attach_cluster(z_ptr, c_ptr); - c_ptr->owner = z_ptr; - return c_ptr; -} - -void tipc_cltr_delete(struct cluster *c_ptr) -{ - u32 n_num; - - if (!c_ptr) - return; - for (n_num = 1; n_num <= c_ptr->highest_node; n_num++) { - tipc_node_delete(c_ptr->nodes[n_num]); - } - for (n_num = LOWEST_SLAVE; n_num <= c_ptr->highest_slave; n_num++) { - tipc_node_delete(c_ptr->nodes[n_num]); - } - kfree(c_ptr->nodes); - kfree(c_ptr); -} - -u32 tipc_cltr_next_node(struct cluster *c_ptr, u32 addr) -{ - struct node *n_ptr; - u32 n_num = tipc_node(addr) + 1; - - if (!c_ptr) - return addr; - for (; n_num <= c_ptr->highest_node; n_num++) { - n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && tipc_node_has_active_links(n_ptr)) - return n_ptr->addr; - } - for (n_num = 1; n_num < tipc_node(addr); n_num++) { - n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && tipc_node_has_active_links(n_ptr)) - return n_ptr->addr; - } - return 0; -} - -void tipc_cltr_attach_node(struct cluster *c_ptr, struct node *n_ptr) -{ - u32 n_num = tipc_node(n_ptr->addr); - u32 max_n_num = tipc_max_nodes; - - if (in_own_cluster(n_ptr->addr)) - max_n_num = tipc_highest_allowed_slave; - assert(n_num > 0); - assert(n_num <= max_n_num); - assert(c_ptr->nodes[n_num] == NULL); - c_ptr->nodes[n_num] = n_ptr; - if (n_num > c_ptr->highest_node) - c_ptr->highest_node = n_num; -} - -/** - * tipc_cltr_select_router - select router to a cluster - * - * Uses deterministic and fair algorithm. - */ - -u32 tipc_cltr_select_router(struct cluster *c_ptr, u32 ref) -{ - u32 n_num; - u32 ulim = c_ptr->highest_node; - u32 mask; - u32 tstart; - - assert(!in_own_cluster(c_ptr->addr)); - if (!ulim) - return 0; - - /* Start entry must be random */ - mask = tipc_max_nodes; - while (mask > ulim) - mask >>= 1; - tstart = ref & mask; - n_num = tstart; - - /* Lookup upwards with wrap-around */ - do { - if (tipc_node_is_up(c_ptr->nodes[n_num])) - break; - } while (++n_num <= ulim); - if (n_num > ulim) { - n_num = 1; - do { - if (tipc_node_is_up(c_ptr->nodes[n_num])) - break; - } while (++n_num < tstart); - if (n_num == tstart) - return 0; - } - assert(n_num <= ulim); - return tipc_node_select_router(c_ptr->nodes[n_num], ref); -} - -/** - * tipc_cltr_select_node - select destination node within a remote cluster - * - * Uses deterministic and fair algorithm. - */ - -struct node *tipc_cltr_select_node(struct cluster *c_ptr, u32 selector) -{ - u32 n_num; - u32 mask = tipc_max_nodes; - u32 start_entry; - - assert(!in_own_cluster(c_ptr->addr)); - if (!c_ptr->highest_node) - return NULL; - - /* Start entry must be random */ - while (mask > c_ptr->highest_node) { - mask >>= 1; - } - start_entry = (selector & mask) ? selector & mask : 1u; - assert(start_entry <= c_ptr->highest_node); - - /* Lookup upwards with wrap-around */ - for (n_num = start_entry; n_num <= c_ptr->highest_node; n_num++) { - if (tipc_node_has_active_links(c_ptr->nodes[n_num])) - return c_ptr->nodes[n_num]; - } - for (n_num = 1; n_num < start_entry; n_num++) { - if (tipc_node_has_active_links(c_ptr->nodes[n_num])) - return c_ptr->nodes[n_num]; - } - return NULL; -} - -/* - * Routing table management: See description in node.c - */ - -static struct sk_buff *tipc_cltr_prepare_routing_msg(u32 data_size, u32 dest) -{ - u32 size = INT_H_SIZE + data_size; - struct sk_buff *buf = buf_acquire(size); - struct tipc_msg *msg; - - if (buf) { - msg = buf_msg(buf); - memset((char *)msg, 0, size); - msg_init(msg, ROUTE_DISTRIBUTOR, 0, TIPC_OK, INT_H_SIZE, dest); - } - return buf; -} - -void tipc_cltr_bcast_new_route(struct cluster *c_ptr, u32 dest, - u32 lower, u32 upper) -{ - struct sk_buff *buf = tipc_cltr_prepare_routing_msg(0, c_ptr->addr); - struct tipc_msg *msg; - - if (buf) { - msg = buf_msg(buf); - msg_set_remote_node(msg, dest); - msg_set_type(msg, ROUTE_ADDITION); - tipc_cltr_multicast(c_ptr, buf, lower, upper); - } else { - warn("Memory squeeze: broadcast of new route failed\n"); - } -} - -void tipc_cltr_bcast_lost_route(struct cluster *c_ptr, u32 dest, - u32 lower, u32 upper) -{ - struct sk_buff *buf = tipc_cltr_prepare_routing_msg(0, c_ptr->addr); - struct tipc_msg *msg; - - if (buf) { - msg = buf_msg(buf); - msg_set_remote_node(msg, dest); - msg_set_type(msg, ROUTE_REMOVAL); - tipc_cltr_multicast(c_ptr, buf, lower, upper); - } else { - warn("Memory squeeze: broadcast of lost route failed\n"); - } -} - -void tipc_cltr_send_slave_routes(struct cluster *c_ptr, u32 dest) -{ - struct sk_buff *buf; - struct tipc_msg *msg; - u32 highest = c_ptr->highest_slave; - u32 n_num; - int send = 0; - - assert(!is_slave(dest)); - assert(in_own_cluster(dest)); - assert(in_own_cluster(c_ptr->addr)); - if (highest <= LOWEST_SLAVE) - return; - buf = tipc_cltr_prepare_routing_msg(highest - LOWEST_SLAVE + 1, - c_ptr->addr); - if (buf) { - msg = buf_msg(buf); - msg_set_remote_node(msg, c_ptr->addr); - msg_set_type(msg, SLAVE_ROUTING_TABLE); - for (n_num = LOWEST_SLAVE; n_num <= highest; n_num++) { - if (c_ptr->nodes[n_num] && - tipc_node_has_active_links(c_ptr->nodes[n_num])) { - send = 1; - msg_set_dataoctet(msg, n_num); - } - } - if (send) - tipc_link_send(buf, dest, dest); - else - buf_discard(buf); - } else { - warn("Memory squeeze: broadcast of lost route failed\n"); - } -} - -void tipc_cltr_send_ext_routes(struct cluster *c_ptr, u32 dest) -{ - struct sk_buff *buf; - struct tipc_msg *msg; - u32 highest = c_ptr->highest_node; - u32 n_num; - int send = 0; - - if (in_own_cluster(c_ptr->addr)) - return; - assert(!is_slave(dest)); - assert(in_own_cluster(dest)); - highest = c_ptr->highest_node; - buf = tipc_cltr_prepare_routing_msg(highest + 1, c_ptr->addr); - if (buf) { - msg = buf_msg(buf); - msg_set_remote_node(msg, c_ptr->addr); - msg_set_type(msg, EXT_ROUTING_TABLE); - for (n_num = 1; n_num <= highest; n_num++) { - if (c_ptr->nodes[n_num] && - tipc_node_has_active_links(c_ptr->nodes[n_num])) { - send = 1; - msg_set_dataoctet(msg, n_num); - } - } - if (send) - tipc_link_send(buf, dest, dest); - else - buf_discard(buf); - } else { - warn("Memory squeeze: broadcast of external route failed\n"); - } -} - -void tipc_cltr_send_local_routes(struct cluster *c_ptr, u32 dest) -{ - struct sk_buff *buf; - struct tipc_msg *msg; - u32 highest = c_ptr->highest_node; - u32 n_num; - int send = 0; - - assert(is_slave(dest)); - assert(in_own_cluster(c_ptr->addr)); - buf = tipc_cltr_prepare_routing_msg(highest, c_ptr->addr); - if (buf) { - msg = buf_msg(buf); - msg_set_remote_node(msg, c_ptr->addr); - msg_set_type(msg, LOCAL_ROUTING_TABLE); - for (n_num = 1; n_num <= highest; n_num++) { - if (c_ptr->nodes[n_num] && - tipc_node_has_active_links(c_ptr->nodes[n_num])) { - send = 1; - msg_set_dataoctet(msg, n_num); - } - } - if (send) - tipc_link_send(buf, dest, dest); - else - buf_discard(buf); - } else { - warn("Memory squeeze: broadcast of local route failed\n"); - } -} - -void tipc_cltr_recv_routing_table(struct sk_buff *buf) -{ - struct tipc_msg *msg = buf_msg(buf); - struct cluster *c_ptr; - struct node *n_ptr; - unchar *node_table; - u32 table_size; - u32 router; - u32 rem_node = msg_remote_node(msg); - u32 z_num; - u32 c_num; - u32 n_num; - - c_ptr = tipc_cltr_find(rem_node); - if (!c_ptr) { - c_ptr = tipc_cltr_create(rem_node); - if (!c_ptr) { - buf_discard(buf); - return; - } - } - - node_table = buf->data + msg_hdr_sz(msg); - table_size = msg_size(msg) - msg_hdr_sz(msg); - router = msg_prevnode(msg); - z_num = tipc_zone(rem_node); - c_num = tipc_cluster(rem_node); - - switch (msg_type(msg)) { - case LOCAL_ROUTING_TABLE: - assert(is_slave(tipc_own_addr)); - case EXT_ROUTING_TABLE: - for (n_num = 1; n_num < table_size; n_num++) { - if (node_table[n_num]) { - u32 addr = tipc_addr(z_num, c_num, n_num); - n_ptr = c_ptr->nodes[n_num]; - if (!n_ptr) { - n_ptr = tipc_node_create(addr); - } - if (n_ptr) - tipc_node_add_router(n_ptr, router); - } - } - break; - case SLAVE_ROUTING_TABLE: - assert(!is_slave(tipc_own_addr)); - assert(in_own_cluster(c_ptr->addr)); - for (n_num = 1; n_num < table_size; n_num++) { - if (node_table[n_num]) { - u32 slave_num = n_num + LOWEST_SLAVE; - u32 addr = tipc_addr(z_num, c_num, slave_num); - n_ptr = c_ptr->nodes[slave_num]; - if (!n_ptr) { - n_ptr = tipc_node_create(addr); - } - if (n_ptr) - tipc_node_add_router(n_ptr, router); - } - } - break; - case ROUTE_ADDITION: - if (!is_slave(tipc_own_addr)) { - assert(!in_own_cluster(c_ptr->addr) - || is_slave(rem_node)); - } else { - assert(in_own_cluster(c_ptr->addr) - && !is_slave(rem_node)); - } - n_ptr = c_ptr->nodes[tipc_node(rem_node)]; - if (!n_ptr) - n_ptr = tipc_node_create(rem_node); - if (n_ptr) - tipc_node_add_router(n_ptr, router); - break; - case ROUTE_REMOVAL: - if (!is_slave(tipc_own_addr)) { - assert(!in_own_cluster(c_ptr->addr) - || is_slave(rem_node)); - } else { - assert(in_own_cluster(c_ptr->addr) - && !is_slave(rem_node)); - } - n_ptr = c_ptr->nodes[tipc_node(rem_node)]; - if (n_ptr) - tipc_node_remove_router(n_ptr, router); - break; - default: - assert(!"Illegal routing manager message received\n"); - } - buf_discard(buf); -} - -void tipc_cltr_remove_as_router(struct cluster *c_ptr, u32 router) -{ - u32 start_entry; - u32 tstop; - u32 n_num; - - if (is_slave(router)) - return; /* Slave nodes can not be routers */ - - if (in_own_cluster(c_ptr->addr)) { - start_entry = LOWEST_SLAVE; - tstop = c_ptr->highest_slave; - } else { - start_entry = 1; - tstop = c_ptr->highest_node; - } - - for (n_num = start_entry; n_num <= tstop; n_num++) { - if (c_ptr->nodes[n_num]) { - tipc_node_remove_router(c_ptr->nodes[n_num], router); - } - } -} - -/** - * tipc_cltr_multicast - multicast message to local nodes - */ - -static void tipc_cltr_multicast(struct cluster *c_ptr, struct sk_buff *buf, - u32 lower, u32 upper) -{ - struct sk_buff *buf_copy; - struct node *n_ptr; - u32 n_num; - u32 tstop; - - assert(lower <= upper); - assert(((lower >= 1) && (lower <= tipc_max_nodes)) || - ((lower >= LOWEST_SLAVE) && (lower <= tipc_highest_allowed_slave))); - assert(((upper >= 1) && (upper <= tipc_max_nodes)) || - ((upper >= LOWEST_SLAVE) && (upper <= tipc_highest_allowed_slave))); - assert(in_own_cluster(c_ptr->addr)); - - tstop = is_slave(upper) ? c_ptr->highest_slave : c_ptr->highest_node; - if (tstop > upper) - tstop = upper; - for (n_num = lower; n_num <= tstop; n_num++) { - n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && tipc_node_has_active_links(n_ptr)) { - buf_copy = skb_copy(buf, GFP_ATOMIC); - if (buf_copy == NULL) - break; - msg_set_destnode(buf_msg(buf_copy), n_ptr->addr); - tipc_link_send(buf_copy, n_ptr->addr, n_ptr->addr); - } - } - buf_discard(buf); -} - -/** - * tipc_cltr_broadcast - broadcast message to all nodes within cluster - */ - -void tipc_cltr_broadcast(struct sk_buff *buf) -{ - struct sk_buff *buf_copy; - struct cluster *c_ptr; - struct node *n_ptr; - u32 n_num; - u32 tstart; - u32 tstop; - u32 node_type; - - if (tipc_mode == TIPC_NET_MODE) { - c_ptr = tipc_cltr_find(tipc_own_addr); - assert(in_own_cluster(c_ptr->addr)); /* For now */ - - /* Send to standard nodes, then repeat loop sending to slaves */ - tstart = 1; - tstop = c_ptr->highest_node; - for (node_type = 1; node_type <= 2; node_type++) { - for (n_num = tstart; n_num <= tstop; n_num++) { - n_ptr = c_ptr->nodes[n_num]; - if (n_ptr && tipc_node_has_active_links(n_ptr)) { - buf_copy = skb_copy(buf, GFP_ATOMIC); - if (buf_copy == NULL) - goto exit; - msg_set_destnode(buf_msg(buf_copy), - n_ptr->addr); - tipc_link_send(buf_copy, n_ptr->addr, - n_ptr->addr); - } - } - tstart = LOWEST_SLAVE; - tstop = c_ptr->highest_slave; - } - } -exit: - buf_discard(buf); -} - -int tipc_cltr_init(void) -{ - tipc_highest_allowed_slave = LOWEST_SLAVE + tipc_max_slaves; - return tipc_cltr_create(tipc_own_addr) ? TIPC_OK : -ENOMEM; -} - diff --git a/net/tipc/cluster.h b/net/tipc/cluster.h deleted file mode 100644 index 62df074afae..00000000000 --- a/net/tipc/cluster.h +++ /dev/null @@ -1,92 +0,0 @@ -/* - * net/tipc/cluster.h: Include file for TIPC cluster management routines - * - * 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_CLUSTER_H -#define _TIPC_CLUSTER_H - -#include "addr.h" -#include "zone.h" - -#define LOWEST_SLAVE 2048u - -/** - * struct cluster - TIPC cluster structure - * @addr: network address of cluster - * @owner: pointer to zone that cluster belongs to - * @nodes: array of pointers to all nodes within cluster - * @highest_node: id of highest numbered node within cluster - * @highest_slave: (used for secondary node support) - */ - -struct cluster { - u32 addr; - struct _zone *owner; - struct node **nodes; - u32 highest_node; - u32 highest_slave; -}; - - -extern struct node **tipc_local_nodes; -extern u32 tipc_highest_allowed_slave; -extern struct node_map tipc_cltr_bcast_nodes; - -void tipc_cltr_remove_as_router(struct cluster *c_ptr, u32 router); -void tipc_cltr_send_ext_routes(struct cluster *c_ptr, u32 dest); -struct node *tipc_cltr_select_node(struct cluster *c_ptr, u32 selector); -u32 tipc_cltr_select_router(struct cluster *c_ptr, u32 ref); -void tipc_cltr_recv_routing_table(struct sk_buff *buf); -struct cluster *tipc_cltr_create(u32 addr); -void tipc_cltr_delete(struct cluster *c_ptr); -void tipc_cltr_attach_node(struct cluster *c_ptr, struct node *n_ptr); -void tipc_cltr_send_slave_routes(struct cluster *c_ptr, u32 dest); -void tipc_cltr_broadcast(struct sk_buff *buf); -int tipc_cltr_init(void); -u32 tipc_cltr_next_node(struct cluster *c_ptr, u32 addr); -void tipc_cltr_bcast_new_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); -void tipc_cltr_send_local_routes(struct cluster *c_ptr, u32 dest); -void tipc_cltr_bcast_lost_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); - -static inline struct cluster *tipc_cltr_find(u32 addr) -{ - struct _zone *z_ptr = tipc_zone_find(addr); - - if (z_ptr) - return z_ptr->clusters[1]; - return NULL; -} - -#endif diff --git a/net/tipc/config.c b/net/tipc/config.c index c71337a22d3..2b42403ad33 100644 --- a/net/tipc/config.c +++ b/net/tipc/config.c @@ -2,7 +2,7 @@ * net/tipc/config.c: TIPC configuration management code * * Copyright (c) 2002-2006, Ericsson AB - * Copyright (c) 2004-2006, Wind River Systems + * Copyright (c) 2004-2007, 2010-2013, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -35,47 +35,17 @@ */ #include "core.h" -#include "dbg.h" -#include "bearer.h" #include "port.h" -#include "link.h" -#include "zone.h" -#include "addr.h" #include "name_table.h" -#include "node.h" #include "config.h" -#include "discover.h" +#include "server.h" -struct subscr_data { - char usr_handle[8]; - u32 domain; - u32 port_ref; - struct list_head subd_list; -}; - -struct manager { - u32 user_ref; - u32 port_ref; - u32 subscr_ref; - u32 link_subscriptions; - struct list_head link_subscribers; -}; - -static struct manager mng = { 0}; - -static DEFINE_SPINLOCK(config_lock); +#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 */ - -void tipc_cfg_link_event(u32 addr, char *name, int up) -{ - /* TIPC DOESN'T HANDLE LINK EVENT SUBSCRIPTIONS AT THE MOMENT */ -} - - struct sk_buff *tipc_cfg_reply_alloc(int payload_size) { struct sk_buff *buf; @@ -92,10 +62,8 @@ int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, 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) { - dbg("tipc_cfg_append_tlv unable to append TLV\n"); + 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)); @@ -104,7 +72,7 @@ int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, return 1; } -struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) +static struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) { struct sk_buff *buf; __be32 value_net; @@ -118,6 +86,11 @@ struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) 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; @@ -129,126 +102,37 @@ struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string) return buf; } - - - -#if 0 - -/* Now obsolete code for handling commands not yet implemented the new way */ - -int tipc_cfg_cmd(const struct tipc_cmd_msg * msg, - char *data, - u32 sz, - u32 *ret_size, - struct tipc_portid *orig) +static struct sk_buff *tipc_show_stats(void) { - int rv = -EINVAL; - u32 cmd = msg->cmd; + struct sk_buff *buf; + struct tlv_desc *rep_tlv; + char *pb; + int pb_len; + int str_len; + u32 value; - *ret_size = 0; - switch (cmd) { - case TIPC_REMOVE_LINK: - case TIPC_CMD_BLOCK_LINK: - case TIPC_CMD_UNBLOCK_LINK: - if (!cfg_check_connection(orig)) - rv = link_control(msg->argv.link_name, msg->cmd, 0); - break; - case TIPC_ESTABLISH: - { - int connected; - - tipc_isconnected(mng.conn_port_ref, &connected); - if (connected || !orig) { - rv = TIPC_FAILURE; - break; - } - rv = tipc_connect2port(mng.conn_port_ref, orig); - if (rv == TIPC_OK) - orig = 0; - break; - } - case TIPC_GET_PEER_ADDRESS: - *ret_size = link_peer_addr(msg->argv.link_name, data, sz); - break; - case TIPC_GET_ROUTES: - rv = TIPC_OK; - break; - default: {} - } - if (*ret_size) - rv = TIPC_OK; - return rv; -} + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); -static void cfg_cmd_event(struct tipc_cmd_msg *msg, - char *data, - u32 sz, - struct tipc_portid const *orig) -{ - int rv = -EINVAL; - struct tipc_cmd_result_msg rmsg; - struct iovec msg_sect[2]; - int *arg; + value = ntohl(*(u32 *)TLV_DATA(req_tlv_area)); + if (value != 0) + return tipc_cfg_reply_error_string("unsupported argument"); - msg->cmd = ntohl(msg->cmd); + buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); + if (buf == NULL) + return NULL; - cfg_prepare_res_msg(msg->cmd, msg->usr_handle, rv, &rmsg, msg_sect, - data, 0); - if (ntohl(msg->magic) != TIPC_MAGIC) - goto exit; + rep_tlv = (struct tlv_desc *)buf->data; + pb = TLV_DATA(rep_tlv); + pb_len = ULTRA_STRING_MAX_LEN; - switch (msg->cmd) { - case TIPC_CREATE_LINK: - if (!cfg_check_connection(orig)) - rv = disc_create_link(&msg->argv.create_link); - break; - case TIPC_LINK_SUBSCRIBE: - { - struct subscr_data *sub; - - if (mng.link_subscriptions > 64) - break; - sub = kmalloc(sizeof(*sub), - GFP_ATOMIC); - if (sub == NULL) { - warn("Memory squeeze; dropped remote link subscription\n"); - break; - } - INIT_LIST_HEAD(&sub->subd_list); - tipc_createport(mng.user_ref, - (void *)sub, - TIPC_HIGH_IMPORTANCE, - 0, - 0, - (tipc_conn_shutdown_event)cfg_linksubscr_cancel, - 0, - 0, - (tipc_conn_msg_event)cfg_linksubscr_cancel, - 0, - &sub->port_ref); - if (!sub->port_ref) { - kfree(sub); - break; - } - memcpy(sub->usr_handle,msg->usr_handle, - sizeof(sub->usr_handle)); - sub->domain = msg->argv.domain; - list_add_tail(&sub->subd_list, &mng.link_subscribers); - tipc_connect2port(sub->port_ref, orig); - rmsg.retval = TIPC_OK; - tipc_send(sub->port_ref, 2u, msg_sect); - mng.link_subscriptions++; - return; - } - default: - rv = tipc_cfg_cmd(msg, data, sz, (u32 *)&msg_sect[1].iov_len, orig); - } - exit: - rmsg.result_len = htonl(msg_sect[1].iov_len); - rmsg.retval = htonl(rv); - tipc_cfg_respond(msg_sect, 2u, orig); + 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; } -#endif static struct sk_buff *cfg_enable_bearer(void) { @@ -259,7 +143,7 @@ static struct sk_buff *cfg_enable_bearer(void) args = (struct tipc_bearer_config *)TLV_DATA(req_tlv_area); if (tipc_enable_bearer(args->name, - ntohl(args->detect_scope), + ntohl(args->disc_domain), ntohl(args->priority))) return tipc_cfg_reply_error_string("unable to enable bearer"); @@ -290,67 +174,13 @@ static struct sk_buff *cfg_set_own_addr(void) if (!tipc_addr_node_valid(addr)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (node address)"); - if (tipc_mode == TIPC_NET_MODE) + if (tipc_own_addr) return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED " (cannot change node address once assigned)"); - tipc_own_addr = addr; - - /* - * Must release all spinlocks before calling start_net() because - * Linux version of TIPC calls eth_media_start() which calls - * register_netdevice_notifier() which may block! - * - * Temporarily releasing the lock should be harmless for non-Linux TIPC, - * but Linux version of eth_media_start() should really be reworked - * so that it can be called with spinlocks held. - */ - - spin_unlock_bh(&config_lock); - tipc_core_start_net(); - spin_lock_bh(&config_lock); - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_set_remote_mng(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)); - tipc_remote_management = (value != 0); - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_set_max_publications(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 != delimit(value, 1, 65535)) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max publications must be 1-65535)"); - tipc_max_publications = value; - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_set_max_subscriptions(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); + if (!tipc_net_start(addr)) + return tipc_cfg_reply_none(); - value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); - if (value != delimit(value, 1, 65535)) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max subscriptions must be 1-65535"); - tipc_max_subscriptions = value; - 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) @@ -362,78 +192,11 @@ static struct sk_buff *cfg_set_max_ports(void) value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_max_ports) return tipc_cfg_reply_none(); - if (value != delimit(value, 127, 65535)) + if (value < 127 || value > 65535) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (max ports must be 127-65535)"); - if (tipc_mode != TIPC_NOT_RUNNING) - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change max ports while TIPC is active)"); - tipc_max_ports = value; - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_set_max_zones(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_zones) - return tipc_cfg_reply_none(); - if (value != delimit(value, 1, 255)) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max zones must be 1-255)"); - if (tipc_mode == TIPC_NET_MODE) - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change max zones once TIPC has joined a network)"); - tipc_max_zones = value; - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_set_max_clusters(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 != delimit(value, 1, 1)) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max clusters fixed at 1)"); - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_set_max_nodes(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_nodes) - return tipc_cfg_reply_none(); - if (value != delimit(value, 8, 2047)) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max nodes must be 8-2047)"); - if (tipc_mode == TIPC_NET_MODE) - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change max nodes once TIPC has joined a network)"); - tipc_max_nodes = value; - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_set_max_slaves(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 != 0) - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (max secondary nodes fixed at 0)"); - return tipc_cfg_reply_none(); + 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) @@ -445,10 +208,10 @@ static struct sk_buff *cfg_set_netid(void) value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_net_id) return tipc_cfg_reply_none(); - if (value != delimit(value, 1, 9999)) + if (value < 1 || value > 9999) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (network id must be 1-9999)"); - if (tipc_mode == TIPC_NET_MODE) + 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; @@ -460,38 +223,23 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area { struct sk_buff *rep_tlv_buf; - spin_lock_bh(&config_lock); + 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(orig_node == tipc_own_addr)) { + if (likely(in_own_node(orig_node))) { /* command is permitted */ - } else if (cmd >= 0x8000) { + } else { rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED " (cannot be done remotely)"); goto exit; - } else if (!tipc_remote_management) { - rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NO_REMOTE); - goto exit; - } - else if (cmd >= 0x4000) { - u32 domain = 0; - - if ((tipc_nametbl_translate(TIPC_ZM_SRV, 0, &domain) == 0) || - (domain != orig_node)) { - rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_ZONE_MSTR); - goto exit; - } } /* Call appropriate processing routine */ - switch (cmd) { case TIPC_CMD_NOOP: rep_tlv_buf = tipc_cfg_reply_none(); @@ -520,19 +268,8 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area case TIPC_CMD_SHOW_PORTS: rep_tlv_buf = tipc_port_get_ports(); break; -#if 0 - case TIPC_CMD_SHOW_PORT_STATS: - rep_tlv_buf = port_show_stats(req_tlv_area, req_tlv_space); - break; - case TIPC_CMD_RESET_PORT_STATS: - rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED); - break; -#endif - case TIPC_CMD_SET_LOG_SIZE: - rep_tlv_buf = tipc_log_resize(req_tlv_area, req_tlv_space); - break; - case TIPC_CMD_DUMP_LOG: - rep_tlv_buf = tipc_log_dump(); + case TIPC_CMD_SHOW_STATS: + rep_tlv_buf = tipc_show_stats(); break; case TIPC_CMD_SET_LINK_TOL: case TIPC_CMD_SET_LINK_PRI: @@ -548,59 +285,40 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area case TIPC_CMD_SET_NODE_ADDR: rep_tlv_buf = cfg_set_own_addr(); break; - case TIPC_CMD_SET_REMOTE_MNG: - rep_tlv_buf = cfg_set_remote_mng(); - break; case TIPC_CMD_SET_MAX_PORTS: rep_tlv_buf = cfg_set_max_ports(); break; - case TIPC_CMD_SET_MAX_PUBL: - rep_tlv_buf = cfg_set_max_publications(); - break; - case TIPC_CMD_SET_MAX_SUBSCR: - rep_tlv_buf = cfg_set_max_subscriptions(); - break; - case TIPC_CMD_SET_MAX_ZONES: - rep_tlv_buf = cfg_set_max_zones(); - break; - case TIPC_CMD_SET_MAX_CLUSTERS: - rep_tlv_buf = cfg_set_max_clusters(); - break; - case TIPC_CMD_SET_MAX_NODES: - rep_tlv_buf = cfg_set_max_nodes(); - break; - case TIPC_CMD_SET_MAX_SLAVES: - rep_tlv_buf = cfg_set_max_slaves(); - break; case TIPC_CMD_SET_NETID: rep_tlv_buf = cfg_set_netid(); break; - case TIPC_CMD_GET_REMOTE_MNG: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_remote_management); - break; case TIPC_CMD_GET_MAX_PORTS: rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_ports); break; - case TIPC_CMD_GET_MAX_PUBL: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_publications); + case TIPC_CMD_GET_NETID: + rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_net_id); break; - case TIPC_CMD_GET_MAX_SUBSCR: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_subscriptions); + 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: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_zones); - break; + case TIPC_CMD_SET_MAX_SLAVES: + case TIPC_CMD_GET_MAX_SLAVES: + case TIPC_CMD_SET_MAX_CLUSTERS: case TIPC_CMD_GET_MAX_CLUSTERS: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_clusters); - break; + case TIPC_CMD_SET_MAX_NODES: case TIPC_CMD_GET_MAX_NODES: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_nodes); - break; - case TIPC_CMD_GET_MAX_SLAVES: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_slaves); - break; - case TIPC_CMD_GET_NETID: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_net_id); + 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 @@ -608,95 +326,17 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area break; } - /* Return reply buffer */ -exit: - spin_unlock_bh(&config_lock); - return rep_tlv_buf; -} + WARN_ON(rep_tlv_buf->len > TLV_SPACE(ULTRA_STRING_MAX_LEN)); -static void cfg_named_msg_event(void *userdata, - u32 port_ref, - struct sk_buff **buf, - const unchar *msg, - u32 size, - u32 importance, - struct tipc_portid const *orig, - struct tipc_name_seq const *dest) -{ - struct tipc_cfg_msg_hdr *req_hdr; - struct tipc_cfg_msg_hdr *rep_hdr; - struct sk_buff *rep_buf; - - /* Validate configuration message header (ignore invalid message) */ - - req_hdr = (struct tipc_cfg_msg_hdr *)msg; - if ((size < sizeof(*req_hdr)) || - (size != TCM_ALIGN(ntohl(req_hdr->tcm_len))) || - (ntohs(req_hdr->tcm_flags) != TCM_F_REQUEST)) { - warn("Invalid configuration message discarded\n"); - return; + /* 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); } - /* Generate reply for request (if can't, return request) */ - - rep_buf = tipc_cfg_do_cmd(orig->node, - ntohs(req_hdr->tcm_type), - msg + sizeof(*req_hdr), - size - sizeof(*req_hdr), - BUF_HEADROOM + MAX_H_SIZE + sizeof(*rep_hdr)); - if (rep_buf) { - skb_push(rep_buf, sizeof(*rep_hdr)); - rep_hdr = (struct tipc_cfg_msg_hdr *)rep_buf->data; - memcpy(rep_hdr, req_hdr, sizeof(*rep_hdr)); - rep_hdr->tcm_len = htonl(rep_buf->len); - rep_hdr->tcm_flags &= htons(~TCM_F_REQUEST); - } else { - rep_buf = *buf; - *buf = NULL; - } - - /* NEED TO ADD CODE TO HANDLE FAILED SEND (SUCH AS CONGESTION) */ - tipc_send_buf2port(port_ref, orig, rep_buf, rep_buf->len); -} - -int tipc_cfg_init(void) -{ - struct tipc_name_seq seq; - int res; - - memset(&mng, 0, sizeof(mng)); - INIT_LIST_HEAD(&mng.link_subscribers); - - res = tipc_attach(&mng.user_ref, NULL, NULL); - if (res) - goto failed; - - res = tipc_createport(mng.user_ref, NULL, TIPC_CRITICAL_IMPORTANCE, - NULL, NULL, NULL, - NULL, cfg_named_msg_event, NULL, - NULL, &mng.port_ref); - if (res) - goto failed; - - seq.type = TIPC_CFG_SRV; - seq.lower = seq.upper = tipc_own_addr; - res = tipc_nametbl_publish_rsv(mng.port_ref, TIPC_ZONE_SCOPE, &seq); - if (res) - goto failed; - - return 0; - -failed: - err("Unable to create configuration service\n"); - tipc_detach(mng.user_ref); - mng.user_ref = 0; - return res; -} - -void tipc_cfg_stop(void) -{ - if (mng.user_ref) { - tipc_detach(mng.user_ref); - mng.user_ref = 0; - } + /* Return reply buffer */ +exit: + rtnl_unlock(); + return rep_tlv_buf; } diff --git a/net/tipc/config.h b/net/tipc/config.h index 5cd7cc56c54..47b1bf18161 100644 --- a/net/tipc/config.h +++ b/net/tipc/config.h @@ -39,13 +39,11 @@ /* ---------------------------------------------------------------------- */ -#include "core.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_unsigned_type(u16 tlv_type, u32 value); struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string); static inline struct sk_buff *tipc_cfg_reply_none(void) @@ -53,11 +51,6 @@ static inline struct sk_buff *tipc_cfg_reply_none(void) return tipc_cfg_reply_alloc(0); } -static inline struct sk_buff *tipc_cfg_reply_unsigned(u32 value) -{ - return tipc_cfg_reply_unsigned_type(TIPC_TLV_UNSIGNED, value); -} - static inline struct sk_buff *tipc_cfg_reply_error_string(char *string) { return tipc_cfg_reply_string_type(TIPC_TLV_ERROR_STRING, string); @@ -71,9 +64,4 @@ static inline struct sk_buff *tipc_cfg_reply_ultra_string(char *string) struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *req_tlv_area, int req_tlv_space, int headroom); - -void tipc_cfg_link_event(u32 addr, char *name, int up); -int tipc_cfg_init(void); -void tipc_cfg_stop(void); - #endif diff --git a/net/tipc/core.c b/net/tipc/core.c index d2d7d32c02c..676d18015dd 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -1,8 +1,8 @@ /* * net/tipc/core.c: TIPC module code * - * Copyright (c) 2003-2006, Ericsson AB - * Copyright (c) 2005-2006, Wind River Systems + * 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 @@ -34,190 +34,144 @@ * POSSIBILITY OF SUCH DAMAGE. */ -#include <linux/init.h> -#include <linux/module.h> -#include <linux/kernel.h> -#include <linux/random.h> - #include "core.h" -#include "dbg.h" #include "ref.h" -#include "net.h" -#include "user_reg.h" #include "name_table.h" #include "subscr.h" #include "config.h" +#include "port.h" -int tipc_eth_media_start(void); -void tipc_eth_media_stop(void); -int tipc_handler_start(void); -void tipc_handler_stop(void); -int tipc_socket_init(void); -void tipc_socket_stop(void); -int tipc_netlink_start(void); -void tipc_netlink_stop(void); - -#define TIPC_MOD_VER "1.6.2" - -#ifndef CONFIG_TIPC_ZONES -#define CONFIG_TIPC_ZONES 3 -#endif - -#ifndef CONFIG_TIPC_CLUSTERS -#define CONFIG_TIPC_CLUSTERS 1 -#endif - -#ifndef CONFIG_TIPC_NODES -#define CONFIG_TIPC_NODES 255 -#endif - -#ifndef CONFIG_TIPC_SLAVE_NODES -#define CONFIG_TIPC_SLAVE_NODES 0 -#endif - -#ifndef CONFIG_TIPC_PORTS -#define CONFIG_TIPC_PORTS 8191 -#endif - -#ifndef CONFIG_TIPC_LOG -#define CONFIG_TIPC_LOG 0 -#endif +#include <linux/module.h> /* global variables used by multiple sub-systems within TIPC */ - -int tipc_mode = TIPC_NOT_RUNNING; -int tipc_random; -atomic_t tipc_user_count = ATOMIC_INIT(0); - -const char tipc_alphabet[] = - "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789_."; +int tipc_random __read_mostly; /* configurable TIPC parameters */ - -u32 tipc_own_addr; -int tipc_max_zones; -int tipc_max_clusters; -int tipc_max_nodes; -int tipc_max_slaves; -int tipc_max_ports; -int tipc_max_subscriptions; -int tipc_max_publications; -int tipc_net_id; -int tipc_remote_management; - - -int tipc_get_mode(void) -{ - return tipc_mode; -} +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_core_stop_net - shut down TIPC networking sub-systems - */ - -void tipc_core_stop_net(void) -{ - tipc_eth_media_stop(); - tipc_net_stop(); -} - -/** - * start_net - start TIPC networking sub-systems + * 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. */ - -int tipc_core_start_net(void) +struct sk_buff *tipc_buf_acquire(u32 size) { - int res; - - if ((res = tipc_net_start()) || - (res = tipc_eth_media_start())) { - tipc_core_stop_net(); + 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 res; + return skb; } /** * tipc_core_stop - switch TIPC from SINGLE NODE to NOT RUNNING mode */ - -void tipc_core_stop(void) +static void tipc_core_stop(void) { - if (tipc_mode != TIPC_NODE_MODE) - return; - - tipc_mode = TIPC_NOT_RUNNING; - + tipc_net_stop(); + tipc_bearer_cleanup(); tipc_netlink_stop(); - tipc_handler_stop(); - tipc_cfg_stop(); tipc_subscr_stop(); - tipc_reg_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 */ - -int tipc_core_start(void) +static int tipc_core_start(void) { - int res; - - if (tipc_mode != TIPC_NOT_RUNNING) - return -ENOPROTOOPT; + int err; get_random_bytes(&tipc_random, sizeof(tipc_random)); - tipc_mode = TIPC_NODE_MODE; - if ((res = tipc_handler_start()) || - (res = tipc_ref_table_init(tipc_max_ports + tipc_max_subscriptions, - tipc_random)) || - (res = tipc_reg_start()) || - (res = tipc_nametbl_init()) || - (res = tipc_k_signal((Handler)tipc_subscr_start, 0)) || - (res = tipc_k_signal((Handler)tipc_cfg_init, 0)) || - (res = tipc_netlink_start()) || - (res = tipc_socket_init())) { - tipc_core_stop(); - } - return res; -} + 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; - tipc_log_reinit(CONFIG_TIPC_LOG); - info("Activated (version " TIPC_MOD_VER - " compiled " __DATE__ " " __TIME__ ")\n"); + pr_info("Activated (version " TIPC_MOD_VER ")\n"); tipc_own_addr = 0; - tipc_remote_management = 1; - tipc_max_publications = 10000; - tipc_max_subscriptions = 2000; - tipc_max_ports = delimit(CONFIG_TIPC_PORTS, 127, 65536); - tipc_max_zones = delimit(CONFIG_TIPC_ZONES, 1, 255); - tipc_max_clusters = delimit(CONFIG_TIPC_CLUSTERS, 1, 1); - tipc_max_nodes = delimit(CONFIG_TIPC_NODES, 8, 2047); - tipc_max_slaves = delimit(CONFIG_TIPC_SLAVE_NODES, 0, 2047); + tipc_max_ports = CONFIG_TIPC_PORTS; tipc_net_id = 4711; - if ((res = tipc_core_start())) - err("Unable to start in single node mode\n"); + 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 - info("Started in single node mode\n"); + pr_info("Started in single node mode\n"); return res; } static void __exit tipc_exit(void) { - tipc_core_stop_net(); tipc_core_stop(); - info("Deactivated\n"); - tipc_log_stop(); + pr_info("Deactivated\n"); } module_init(tipc_init); @@ -226,61 +180,3 @@ module_exit(tipc_exit); MODULE_DESCRIPTION("TIPC: Transparent Inter Process Communication"); MODULE_LICENSE("Dual BSD/GPL"); MODULE_VERSION(TIPC_MOD_VER); - -/* Native TIPC API for kernel-space applications (see tipc.h) */ - -EXPORT_SYMBOL(tipc_attach); -EXPORT_SYMBOL(tipc_detach); -EXPORT_SYMBOL(tipc_get_addr); -EXPORT_SYMBOL(tipc_get_mode); -EXPORT_SYMBOL(tipc_createport); -EXPORT_SYMBOL(tipc_deleteport); -EXPORT_SYMBOL(tipc_ownidentity); -EXPORT_SYMBOL(tipc_portimportance); -EXPORT_SYMBOL(tipc_set_portimportance); -EXPORT_SYMBOL(tipc_portunreliable); -EXPORT_SYMBOL(tipc_set_portunreliable); -EXPORT_SYMBOL(tipc_portunreturnable); -EXPORT_SYMBOL(tipc_set_portunreturnable); -EXPORT_SYMBOL(tipc_publish); -EXPORT_SYMBOL(tipc_withdraw); -EXPORT_SYMBOL(tipc_connect2port); -EXPORT_SYMBOL(tipc_disconnect); -EXPORT_SYMBOL(tipc_shutdown); -EXPORT_SYMBOL(tipc_isconnected); -EXPORT_SYMBOL(tipc_peer); -EXPORT_SYMBOL(tipc_ref_valid); -EXPORT_SYMBOL(tipc_send); -EXPORT_SYMBOL(tipc_send_buf); -EXPORT_SYMBOL(tipc_send2name); -EXPORT_SYMBOL(tipc_forward2name); -EXPORT_SYMBOL(tipc_send_buf2name); -EXPORT_SYMBOL(tipc_forward_buf2name); -EXPORT_SYMBOL(tipc_send2port); -EXPORT_SYMBOL(tipc_forward2port); -EXPORT_SYMBOL(tipc_send_buf2port); -EXPORT_SYMBOL(tipc_forward_buf2port); -EXPORT_SYMBOL(tipc_multicast); -/* EXPORT_SYMBOL(tipc_multicast_buf); not available yet */ -EXPORT_SYMBOL(tipc_ispublished); -EXPORT_SYMBOL(tipc_available_nodes); - -/* TIPC API for external bearers (see tipc_bearer.h) */ - -EXPORT_SYMBOL(tipc_block_bearer); -EXPORT_SYMBOL(tipc_continue); -EXPORT_SYMBOL(tipc_disable_bearer); -EXPORT_SYMBOL(tipc_enable_bearer); -EXPORT_SYMBOL(tipc_recv_msg); -EXPORT_SYMBOL(tipc_register_media); - -/* TIPC API for external APIs (see tipc_port.h) */ - -EXPORT_SYMBOL(tipc_createport_raw); -EXPORT_SYMBOL(tipc_set_msg_option); -EXPORT_SYMBOL(tipc_reject_msg); -EXPORT_SYMBOL(tipc_send_buf_fast); -EXPORT_SYMBOL(tipc_acknowledge); -EXPORT_SYMBOL(tipc_get_port); -EXPORT_SYMBOL(tipc_get_handle); - diff --git a/net/tipc/core.h b/net/tipc/core.h index feabca58082..bb26ed1ee96 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -1,8 +1,8 @@ /* * net/tipc/core.h: Include file for TIPC global declarations * - * Copyright (c) 2005-2006, Ericsson AB - * Copyright (c) 2005-2006, Wind River Systems + * 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 @@ -37,168 +37,81 @@ #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 <net/tipc/tipc_msg.h> -#include <net/tipc/tipc_port.h> -#include <net/tipc/tipc_bearer.h> -#include <net/tipc/tipc.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 <asm/uaccess.h> +#include <linux/uaccess.h> #include <linux/interrupt.h> -#include <asm/atomic.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> -/* - * TIPC debugging code - */ - -#define assert(i) BUG_ON(!(i)) - -struct tipc_msg; -extern struct print_buf *TIPC_NULL, *TIPC_CONS, *TIPC_LOG; -extern struct print_buf *TIPC_TEE(struct print_buf *, struct print_buf *); -void tipc_msg_print(struct print_buf*,struct tipc_msg *,const char*); -void tipc_printf(struct print_buf *, const char *fmt, ...); -void tipc_dump(struct print_buf*,const char *fmt, ...); - -#ifdef CONFIG_TIPC_DEBUG - -/* - * TIPC debug support included: - * - system messages are printed to TIPC_OUTPUT print buffer - * - debug messages are printed to DBG_OUTPUT print buffer - */ - -#define err(fmt, arg...) tipc_printf(TIPC_OUTPUT, KERN_ERR "TIPC: " fmt, ## arg) -#define warn(fmt, arg...) tipc_printf(TIPC_OUTPUT, KERN_WARNING "TIPC: " fmt, ## arg) -#define info(fmt, arg...) tipc_printf(TIPC_OUTPUT, KERN_NOTICE "TIPC: " fmt, ## arg) - -#define dbg(fmt, arg...) do {if (DBG_OUTPUT != TIPC_NULL) tipc_printf(DBG_OUTPUT, fmt, ## arg);} while(0) -#define msg_dbg(msg, txt) do {if (DBG_OUTPUT != TIPC_NULL) tipc_msg_print(DBG_OUTPUT, msg, txt);} while(0) -#define dump(fmt, arg...) do {if (DBG_OUTPUT != TIPC_NULL) tipc_dump(DBG_OUTPUT, fmt, ##arg);} while(0) - - -/* - * By default, TIPC_OUTPUT is defined to be system console and TIPC log buffer, - * while DBG_OUTPUT is the null print buffer. These defaults can be changed - * here, or on a per .c file basis, by redefining these symbols. The following - * print buffer options are available: - * - * TIPC_NULL : null buffer (i.e. print nowhere) - * TIPC_CONS : system console - * TIPC_LOG : TIPC log buffer - * &buf : user-defined buffer (struct print_buf *) - * TIPC_TEE(&buf_a,&buf_b) : list of buffers (eg. TIPC_TEE(TIPC_CONS,TIPC_LOG)) - */ - -#ifndef TIPC_OUTPUT -#define TIPC_OUTPUT TIPC_TEE(TIPC_CONS,TIPC_LOG) -#endif +#define TIPC_MOD_VER "2.0.0" -#ifndef DBG_OUTPUT -#define DBG_OUTPUT TIPC_NULL -#endif +#define ULTRA_STRING_MAX_LEN 32768 +#define TIPC_MAX_SUBSCRIPTIONS 65535 +#define TIPC_MAX_PUBLICATIONS 65535 -#else - -/* - * TIPC debug support not included: - * - system messages are printed to system console - * - debug messages are not printed - */ - -#define err(fmt, arg...) printk(KERN_ERR "TIPC: " fmt , ## arg) -#define info(fmt, arg...) printk(KERN_INFO "TIPC: " fmt , ## arg) -#define warn(fmt, arg...) printk(KERN_WARNING "TIPC: " fmt , ## arg) - -#define dbg(fmt, arg...) do {} while (0) -#define msg_dbg(msg,txt) do {} while (0) -#define dump(fmt,arg...) do {} while (0) - - -/* - * TIPC_OUTPUT is defined to be the system console, while DBG_OUTPUT is - * the null print buffer. Thes ensures that any system or debug messages - * that are generated without using the above macros are handled correctly. - */ - -#undef TIPC_OUTPUT -#define TIPC_OUTPUT TIPC_CONS - -#undef DBG_OUTPUT -#define DBG_OUTPUT TIPC_NULL - -#endif +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; -extern int tipc_max_zones; -extern int tipc_max_clusters; -extern int tipc_max_nodes; -extern int tipc_max_slaves; -extern int tipc_max_ports; -extern int tipc_max_subscriptions; -extern int tipc_max_publications; -extern int tipc_net_id; -extern int tipc_remote_management; +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_mode; -extern int tipc_random; -extern const char tipc_alphabet[]; -extern atomic_t tipc_user_count; - +extern int tipc_random __read_mostly; /* * Routines available to privileged subsystems */ - -extern int tipc_core_start(void); -extern void tipc_core_stop(void); -extern int tipc_core_start_net(void); -extern void tipc_core_stop_net(void); - -static inline int delimit(int val, int min, int max) -{ - if (val > max) - return max; - if (val < min) - return min; - return val; -} - +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 and signal code + * TIPC timer code */ - typedef void (*Handler) (unsigned long); -u32 tipc_k_signal(Handler routine, unsigned long argument); - /** * k_init_timer - initialize a timer * @timer: pointer to timer structure @@ -207,11 +120,9 @@ u32 tipc_k_signal(Handler routine, unsigned long argument); * * 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) { - dbg("initializing timer %p\n", timer); setup_timer(timer, routine, argument); } @@ -228,10 +139,8 @@ static inline void k_init_timer(struct timer_list *timer, Handler routine, * 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) { - dbg("starting timer %p for %u\n", timer, msec); mod_timer(timer, jiffies + msecs_to_jiffies(msec) + 1); } @@ -245,10 +154,8 @@ static inline void k_start_timer(struct timer_list *timer, unsigned long msec) * 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) { - dbg("cancelling timer %p\n", timer); del_timer_sync(timer); } @@ -263,73 +170,34 @@ static inline void k_cancel_timer(struct timer_list *timer) * (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) { - dbg("terminating timer %p\n", timer); } - /* * TIPC message buffer code * - * TIPC message buffer headroom reserves space for a link-level header - * (in case the message is sent off-node), - * while ensuring TIPC header is word aligned for quicker access + * TIPC message buffer headroom reserves space for the worst-case + * link-level device header (in case the message is sent off-node). * - * The largest header currently supported is 18 bytes, which is used when - * the standard 14 byte Ethernet header has 4 added bytes for VLAN info + * Note: Headroom should be a multiple of 4 to ensure the TIPC header fields + * are word aligned for quicker access */ - -#define BUF_HEADROOM 20u +#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; } -/** - * 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. - */ - -static inline struct sk_buff *buf_acquire(u32 size) -{ - struct sk_buff *skb; - unsigned int buf_size = (BUF_HEADROOM + size + 3) & ~3u; - - skb = alloc_skb(buf_size, GFP_ATOMIC); - if (skb) { - skb_reserve(skb, BUF_HEADROOM); - skb_put(skb, size); - skb->next = NULL; - } - return skb; -} - -/** - * buf_discard - frees a TIPC message buffer - * @skb: message buffer - * - * Frees a message buffer. If passed NULL, just returns. - */ - -static inline void buf_discard(struct sk_buff *skb) -{ - if (likely(skb != NULL)) - kfree_skb(skb); -} +struct sk_buff *tipc_buf_acquire(u32 size); #endif diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c deleted file mode 100644 index e809d2a2ce0..00000000000 --- a/net/tipc/dbg.c +++ /dev/null @@ -1,436 +0,0 @@ -/* - * net/tipc/dbg.c: TIPC print buffer routines for debugging - * - * Copyright (c) 1996-2006, 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. - */ - -#include "core.h" -#include "config.h" -#include "dbg.h" - -static char print_string[TIPC_PB_MAX_STR]; -static DEFINE_SPINLOCK(print_lock); - -static struct print_buf null_buf = { NULL, 0, NULL, NULL }; -struct print_buf *TIPC_NULL = &null_buf; - -static struct print_buf cons_buf = { NULL, 0, NULL, NULL }; -struct print_buf *TIPC_CONS = &cons_buf; - -static struct print_buf log_buf = { NULL, 0, NULL, NULL }; -struct print_buf *TIPC_LOG = &log_buf; - - -#define FORMAT(PTR,LEN,FMT) \ -{\ - va_list args;\ - va_start(args, FMT);\ - LEN = vsprintf(PTR, FMT, args);\ - va_end(args);\ - *(PTR + LEN) = '\0';\ -} - -/* - * Locking policy when using print buffers. - * - * The following routines use 'print_lock' for protection: - * 1) tipc_printf() - to protect its print buffer(s) and 'print_string' - * 2) TIPC_TEE() - to protect its print buffer(s) - * 3) tipc_dump() - to protect its print buffer(s) and 'print_string' - * 4) tipc_log_XXX() - to protect TIPC_LOG - * - * All routines of the form tipc_printbuf_XXX() rely on the caller to prevent - * simultaneous use of the print buffer(s) being manipulated. - */ - -/** - * tipc_printbuf_init - initialize print buffer to empty - * @pb: pointer to print buffer structure - * @raw: pointer to character array used by print buffer - * @size: size of character array - * - * Makes the print buffer a null device that discards anything written to it - * if the character array is too small (or absent). - */ - -void tipc_printbuf_init(struct print_buf *pb, char *raw, u32 size) -{ - pb->buf = raw; - pb->crs = raw; - pb->size = size; - pb->next = NULL; - - if (size < TIPC_PB_MIN_SIZE) { - pb->buf = NULL; - } else if (raw) { - pb->buf[0] = 0; - pb->buf[size-1] = ~0; - } -} - -/** - * tipc_printbuf_reset - reinitialize print buffer to empty state - * @pb: pointer to print buffer structure - */ - -void tipc_printbuf_reset(struct print_buf *pb) -{ - tipc_printbuf_init(pb, pb->buf, pb->size); -} - -/** - * tipc_printbuf_empty - test if print buffer is in empty state - * @pb: pointer to print buffer structure - * - * Returns non-zero if print buffer is empty. - */ - -int tipc_printbuf_empty(struct print_buf *pb) -{ - return (!pb->buf || (pb->crs == pb->buf)); -} - -/** - * tipc_printbuf_validate - check for print buffer overflow - * @pb: pointer to print buffer structure - * - * Verifies that a print buffer has captured all data written to it. - * If data has been lost, linearize buffer and prepend an error message - * - * Returns length of print buffer data string (including trailing NUL) - */ - -int tipc_printbuf_validate(struct print_buf *pb) -{ - char *err = "\n\n*** PRINT BUFFER OVERFLOW ***\n\n"; - char *cp_buf; - struct print_buf cb; - - if (!pb->buf) - return 0; - - if (pb->buf[pb->size - 1] == 0) { - cp_buf = kmalloc(pb->size, GFP_ATOMIC); - if (cp_buf != NULL){ - tipc_printbuf_init(&cb, cp_buf, pb->size); - tipc_printbuf_move(&cb, pb); - tipc_printbuf_move(pb, &cb); - kfree(cp_buf); - memcpy(pb->buf, err, strlen(err)); - } else { - tipc_printbuf_reset(pb); - tipc_printf(pb, err); - } - } - return (pb->crs - pb->buf + 1); -} - -/** - * tipc_printbuf_move - move print buffer contents to another print buffer - * @pb_to: pointer to destination print buffer structure - * @pb_from: pointer to source print buffer structure - * - * Current contents of destination print buffer (if any) are discarded. - * Source print buffer becomes empty if a successful move occurs. - */ - -void tipc_printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) -{ - int len; - - /* Handle the cases where contents can't be moved */ - - if (!pb_to->buf) - return; - - if (!pb_from->buf) { - tipc_printbuf_reset(pb_to); - return; - } - - if (pb_to->size < pb_from->size) { - tipc_printbuf_reset(pb_to); - tipc_printf(pb_to, "*** PRINT BUFFER MOVE ERROR ***"); - return; - } - - /* Copy data from char after cursor to end (if used) */ - - len = pb_from->buf + pb_from->size - pb_from->crs - 2; - if ((pb_from->buf[pb_from->size-1] == 0) && (len > 0)) { - strcpy(pb_to->buf, pb_from->crs + 1); - pb_to->crs = pb_to->buf + len; - } else - pb_to->crs = pb_to->buf; - - /* Copy data from start to cursor (always) */ - - len = pb_from->crs - pb_from->buf; - strcpy(pb_to->crs, pb_from->buf); - pb_to->crs += len; - - tipc_printbuf_reset(pb_from); -} - -/** - * tipc_printf - append formatted output to print buffer chain - * @pb: pointer to chain of print buffers (may be NULL) - * @fmt: formatted info to be printed - */ - -void tipc_printf(struct print_buf *pb, const char *fmt, ...) -{ - int chars_to_add; - int chars_left; - char save_char; - struct print_buf *pb_next; - - spin_lock_bh(&print_lock); - FORMAT(print_string, chars_to_add, fmt); - if (chars_to_add >= TIPC_PB_MAX_STR) - strcpy(print_string, "*** PRINT BUFFER STRING TOO LONG ***"); - - while (pb) { - if (pb == TIPC_CONS) - printk(print_string); - else if (pb->buf) { - chars_left = pb->buf + pb->size - pb->crs - 1; - if (chars_to_add <= chars_left) { - strcpy(pb->crs, print_string); - pb->crs += chars_to_add; - } else if (chars_to_add >= (pb->size - 1)) { - strcpy(pb->buf, print_string + chars_to_add + 1 - - pb->size); - pb->crs = pb->buf + pb->size - 1; - } else { - strcpy(pb->buf, print_string + chars_left); - save_char = print_string[chars_left]; - print_string[chars_left] = 0; - strcpy(pb->crs, print_string); - print_string[chars_left] = save_char; - pb->crs = pb->buf + chars_to_add - chars_left; - } - } - pb_next = pb->next; - pb->next = NULL; - pb = pb_next; - } - spin_unlock_bh(&print_lock); -} - -/** - * TIPC_TEE - perform next output operation on both print buffers - * @b0: pointer to chain of print buffers (may be NULL) - * @b1: pointer to print buffer to add to chain - * - * Returns pointer to print buffer chain. - */ - -struct print_buf *TIPC_TEE(struct print_buf *b0, struct print_buf *b1) -{ - struct print_buf *pb = b0; - - if (!b0 || (b0 == b1)) - return b1; - - spin_lock_bh(&print_lock); - while (pb->next) { - if ((pb->next == b1) || (pb->next == b0)) - pb->next = pb->next->next; - else - pb = pb->next; - } - pb->next = b1; - spin_unlock_bh(&print_lock); - return b0; -} - -/** - * print_to_console - write string of bytes to console in multiple chunks - */ - -static void print_to_console(char *crs, int len) -{ - int rest = len; - - while (rest > 0) { - int sz = rest < TIPC_PB_MAX_STR ? rest : TIPC_PB_MAX_STR; - char c = crs[sz]; - - crs[sz] = 0; - printk((const char *)crs); - crs[sz] = c; - rest -= sz; - crs += sz; - } -} - -/** - * printbuf_dump - write print buffer contents to console - */ - -static void printbuf_dump(struct print_buf *pb) -{ - int len; - - if (!pb->buf) { - printk("*** PRINT BUFFER NOT ALLOCATED ***"); - return; - } - - /* Dump print buffer from char after cursor to end (if used) */ - - len = pb->buf + pb->size - pb->crs - 2; - if ((pb->buf[pb->size - 1] == 0) && (len > 0)) - print_to_console(pb->crs + 1, len); - - /* Dump print buffer from start to cursor (always) */ - - len = pb->crs - pb->buf; - print_to_console(pb->buf, len); -} - -/** - * tipc_dump - dump non-console print buffer(s) to console - * @pb: pointer to chain of print buffers - */ - -void tipc_dump(struct print_buf *pb, const char *fmt, ...) -{ - struct print_buf *pb_next; - int len; - - spin_lock_bh(&print_lock); - FORMAT(print_string, len, fmt); - printk(print_string); - - for (; pb; pb = pb->next) { - if (pb != TIPC_CONS) { - printk("\n---- Start of %s log dump ----\n\n", - (pb == TIPC_LOG) ? "global" : "local"); - printbuf_dump(pb); - tipc_printbuf_reset(pb); - printk("\n---- End of dump ----\n"); - } - pb_next = pb->next; - pb->next = NULL; - pb = pb_next; - } - spin_unlock_bh(&print_lock); -} - -/** - * tipc_log_stop - free up TIPC log print buffer - */ - -void tipc_log_stop(void) -{ - spin_lock_bh(&print_lock); - if (TIPC_LOG->buf) { - kfree(TIPC_LOG->buf); - TIPC_LOG->buf = NULL; - } - spin_unlock_bh(&print_lock); -} - -/** - * tipc_log_reinit - (re)initialize TIPC log print buffer - * @log_size: print buffer size to use - */ - -void tipc_log_reinit(int log_size) -{ - tipc_log_stop(); - - if (log_size) { - if (log_size < TIPC_PB_MIN_SIZE) - log_size = TIPC_PB_MIN_SIZE; - spin_lock_bh(&print_lock); - tipc_printbuf_init(TIPC_LOG, kmalloc(log_size, GFP_ATOMIC), - log_size); - spin_unlock_bh(&print_lock); - } -} - -/** - * tipc_log_resize - reconfigure size of TIPC log buffer - */ - -struct sk_buff *tipc_log_resize(const void *req_tlv_area, int req_tlv_space) -{ - 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 != delimit(value, 0, 32768)) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (log size must be 0-32768)"); - tipc_log_reinit(value); - return tipc_cfg_reply_none(); -} - -/** - * tipc_log_dump - capture TIPC log buffer contents in configuration message - */ - -struct sk_buff *tipc_log_dump(void) -{ - struct sk_buff *reply; - - spin_lock_bh(&print_lock); - if (!TIPC_LOG->buf) - reply = tipc_cfg_reply_ultra_string("log not activated\n"); - else if (tipc_printbuf_empty(TIPC_LOG)) - reply = tipc_cfg_reply_ultra_string("log is empty\n"); - else { - struct tlv_desc *rep_tlv; - struct print_buf pb; - int str_len; - - str_len = min(TIPC_LOG->size, 32768u); - reply = tipc_cfg_reply_alloc(TLV_SPACE(str_len)); - if (reply) { - rep_tlv = (struct tlv_desc *)reply->data; - tipc_printbuf_init(&pb, TLV_DATA(rep_tlv), str_len); - tipc_printbuf_move(&pb, TIPC_LOG); - str_len = strlen(TLV_DATA(rep_tlv)) + 1; - skb_put(reply, TLV_SPACE(str_len)); - TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); - } - } - spin_unlock_bh(&print_lock); - return reply; -} - diff --git a/net/tipc/discover.c b/net/tipc/discover.c index 5d643e5721e..aa722a42ef8 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -1,8 +1,8 @@ /* * net/tipc/discover.c * - * Copyright (c) 2003-2006, Ericsson AB - * Copyright (c) 2005-2006, Wind River Systems + * 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 @@ -35,100 +35,55 @@ */ #include "core.h" -#include "dbg.h" #include "link.h" -#include "zone.h" #include "discover.h" -#include "port.h" -#include "name_table.h" #define TIPC_LINK_REQ_INIT 125 /* min delay during bearer start up */ -#define TIPC_LINK_REQ_FAST 2000 /* normal delay if bearer has no links */ -#define TIPC_LINK_REQ_SLOW 600000 /* normal delay if bearer has links */ - -#if 0 -#define GET_NODE_INFO 300 -#define GET_NODE_INFO_RESULT 301 -#define FORWARD_LINK_PROBE 302 -#define LINK_REQUEST_REJECTED 303 -#define LINK_REQUEST_ACCEPTED 304 -#define DROP_LINK_REQUEST 305 -#define CHECK_LINK_COUNT 306 -#endif - -/* - * TODO: Most of the inter-cluster setup stuff should be - * rewritten, and be made conformant with specification. - */ +#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 link_req - information about an ongoing link setup request - * @bearer: bearer issuing requests + * 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 link_req { - struct bearer *bearer; +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; }; - -#if 0 -int disc_create_link(const struct tipc_link_create *argv) -{ - /* - * Code for inter cluster link setup here - */ - return TIPC_OK; -} -#endif - -/* - * disc_lost_link(): A link has lost contact - */ - -void tipc_disc_link_event(u32 addr, char *name, int up) -{ - if (in_own_cluster(addr)) - return; - /* - * Code for inter cluster link setup here - */ -} - /** * tipc_disc_init_msg - initialize a link setup message * @type: message type (request or response) - * @req_links: number of links associated with message - * @dest_domain: network domain of node(s) which should respond to message * @b_ptr: ptr to bearer issuing message */ - -static struct sk_buff *tipc_disc_init_msg(u32 type, - u32 req_links, - u32 dest_domain, - struct bearer *b_ptr) +static void tipc_disc_init_msg(struct sk_buff *buf, u32 type, + struct tipc_bearer *b_ptr) { - struct sk_buff *buf = buf_acquire(DSC_H_SIZE); struct tipc_msg *msg; - - if (buf) { - msg = buf_msg(buf); - msg_init(msg, LINK_CONFIG, type, TIPC_OK, DSC_H_SIZE, - dest_domain); - msg_set_non_seq(msg); - msg_set_req_links(msg, req_links); - msg_set_dest_domain(msg, dest_domain); - msg_set_bc_netid(msg, tipc_net_id); - msg_set_media_addr(msg, &b_ptr->publ.addr); - } - return buf; + 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); } /** @@ -137,147 +92,205 @@ static struct sk_buff *tipc_disc_init_msg(u32 type, * @node_addr: duplicated node address * @media_addr: media address advertised by duplicated node */ - -static void disc_dupl_alert(struct bearer *b_ptr, u32 node_addr, +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]; - struct print_buf pb; - - addr_string_fill(node_addr_str, node_addr); - tipc_printbuf_init(&pb, media_addr_str, sizeof(media_addr_str)); - tipc_media_addr_printf(&pb, media_addr); - tipc_printbuf_validate(&pb); - warn("Duplicate %s using %s seen on <%s>\n", - node_addr_str, media_addr_str, b_ptr->publ.name); + + 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_recv_msg - handle incoming link setup message (request or response) + * tipc_disc_rcv - handle incoming discovery message (request or response) * @buf: buffer containing message + * @bearer: bearer that message arrived on */ - -void tipc_disc_recv_msg(struct sk_buff *buf) +void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *bearer) { - struct bearer *b_ptr = (struct bearer *)TIPC_SKB_CB(buf)->handle; - struct link *link; - struct tipc_media_addr media_addr; + 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 dest = msg_dest_domain(msg); - u32 orig = msg_prevnode(msg); + u32 ddom = msg_dest_domain(msg); + u32 onode = msg_prevnode(msg); u32 net_id = msg_bc_netid(msg); - u32 type = msg_type(msg); - - msg_get_media_addr(msg,&media_addr); - msg_dbg(msg, "RECV:"); - buf_discard(buf); - + 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 (!tipc_addr_domain_valid(dest)) + if (maddr.broadcast) return; - if (!tipc_addr_node_valid(orig)) + if (!tipc_addr_domain_valid(ddom)) return; - if (orig == tipc_own_addr) { - if (memcmp(&media_addr, &b_ptr->publ.addr, sizeof(media_addr))) - disc_dupl_alert(b_ptr, tipc_own_addr, &media_addr); + 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 (!in_scope(dest, tipc_own_addr)) + if (!tipc_in_scope(ddom, tipc_own_addr)) return; - if (is_slave(tipc_own_addr) && is_slave(orig)) + if (!tipc_in_scope(bearer->domain, onode)) return; - if (is_slave(orig) && !in_own_cluster(orig)) + + /* Locate, or if necessary, create, node: */ + node = tipc_node_find(onode); + if (!node) + node = tipc_node_create(onode); + if (!node) return; - if (in_own_cluster(orig)) { - /* Always accept link here */ - struct sk_buff *rbuf; - struct tipc_media_addr *addr; - struct node *n_ptr = tipc_node_find(orig); - int link_fully_up; - - dbg(" in own cluster\n"); - if (n_ptr == NULL) { - n_ptr = tipc_node_create(orig); - } - if (n_ptr == NULL) { - return; - } - spin_lock_bh(&n_ptr->lock); - link = n_ptr->links[b_ptr->identity]; - if (!link) { - dbg("creating link\n"); - link = tipc_link_create(b_ptr, orig, &media_addr); - if (!link) { - spin_unlock_bh(&n_ptr->lock); - return; - } - } - addr = &link->media_addr; - if (memcmp(addr, &media_addr, sizeof(*addr))) { - if (tipc_link_is_up(link) || (!link->started)) { - disc_dupl_alert(b_ptr, orig, &media_addr); - spin_unlock_bh(&n_ptr->lock); - return; - } - warn("Resetting link <%s>, peer interface address changed\n", - link->name); - memcpy(addr, &media_addr, sizeof(*addr)); + + 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; } - link_fully_up = (link->state == WORKING_WORKING); - spin_unlock_bh(&n_ptr->lock); - if ((type == DSC_RESP_MSG) || link_fully_up) - return; - rbuf = tipc_disc_init_msg(DSC_RESP_MSG, 1, orig, b_ptr); - if (rbuf != NULL) { - msg_dbg(buf_msg(rbuf),"SEND:"); - b_ptr->media->send_msg(rbuf, &b_ptr->publ, &media_addr); - buf_discard(rbuf); + } + + /* 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); } /** - * tipc_disc_stop_link_req - stop sending periodic link setup requests + * 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 */ - -void tipc_disc_stop_link_req(struct link_req *req) +static void disc_update(struct tipc_link_req *req) { - if (!req) - return; - - k_cancel_timer(&req->timer); - k_term_timer(&req->timer); - buf_discard(req->buf); - kfree(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_update_link_req - update frequency of periodic link setup requests + * tipc_disc_add_dest - increment set of discovered nodes * @req: ptr to link request structure */ - -void tipc_disc_update_link_req(struct link_req *req) +void tipc_disc_add_dest(struct tipc_link_req *req) { - if (!req) - return; + spin_lock_bh(&req->lock); + req->num_nodes++; + spin_unlock_bh(&req->lock); +} - if (req->timer_intv == TIPC_LINK_REQ_SLOW) { - if (!req->bearer->nodes.count) { - req->timer_intv = TIPC_LINK_REQ_FAST; - k_start_timer(&req->timer, req->timer_intv); - } - } else if (req->timer_intv == TIPC_LINK_REQ_FAST) { - if (req->bearer->nodes.count) { - req->timer_intv = TIPC_LINK_REQ_SLOW; - k_start_timer(&req->timer, req->timer_intv); - } - } else { - /* leave timer "as is" if haven't yet reached a "normal" rate */ - } +/** + * 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); } /** @@ -286,61 +299,105 @@ void tipc_disc_update_link_req(struct link_req *req) * * Called whenever a link setup request timer associated with a bearer expires. */ - -static void disc_timeout(struct link_req *req) +static void disc_timeout(struct tipc_link_req *req) { - spin_lock_bh(&req->bearer->publ.lock); - - req->bearer->media->send_msg(req->buf, &req->bearer->publ, &req->dest); - - if ((req->timer_intv == TIPC_LINK_REQ_SLOW) || - (req->timer_intv == TIPC_LINK_REQ_FAST)) { - /* leave timer interval "as is" if already at a "normal" rate */ - } else { - req->timer_intv *= 2; - if (req->timer_intv > TIPC_LINK_REQ_FAST) - req->timer_intv = TIPC_LINK_REQ_FAST; - if ((req->timer_intv == TIPC_LINK_REQ_FAST) && - (req->bearer->nodes.count)) - req->timer_intv = TIPC_LINK_REQ_SLOW; + 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; } - k_start_timer(&req->timer, req->timer_intv); - spin_unlock_bh(&req->bearer->publ.lock); + /* + * 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_init_link_req - start sending periodic link setup requests + * 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 of node(s) which should respond to message - * @req_links: max number of desired links + * @dest_domain: network domain to which links can be established * - * Returns pointer to link request structure, or NULL if unable to create. + * Returns 0 if successful, otherwise -errno. */ - -struct link_req *tipc_disc_init_link_req(struct bearer *b_ptr, - const struct tipc_media_addr *dest, - u32 dest_domain, - u32 req_links) +int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest) { - struct link_req *req; + struct tipc_link_req *req; req = kmalloc(sizeof(*req), GFP_ATOMIC); if (!req) - return NULL; + return -ENOMEM; - req->buf = tipc_disc_init_msg(DSC_REQ_MSG, req_links, dest_domain, b_ptr); + req->buf = tipc_buf_acquire(INT_H_SIZE); if (!req->buf) { kfree(req); - return NULL; + return -ENOMEM; } + tipc_disc_init_msg(req->buf, DSC_REQ_MSG, b_ptr); memcpy(&req->dest, dest, sizeof(*dest)); - req->bearer = b_ptr; + 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); - return req; + 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 index 9fd7587b143..515b57392f4 100644 --- a/net/tipc/discover.h +++ b/net/tipc/discover.h @@ -2,7 +2,7 @@ * net/tipc/discover.h * * Copyright (c) 2003-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -37,22 +37,13 @@ #ifndef _TIPC_DISCOVER_H #define _TIPC_DISCOVER_H -#include "core.h" +struct tipc_link_req; -struct link_req; - -struct link_req *tipc_disc_init_link_req(struct bearer *b_ptr, - const struct tipc_media_addr *dest, - u32 dest_domain, - u32 req_links); -void tipc_disc_update_link_req(struct link_req *req); -void tipc_disc_stop_link_req(struct link_req *req); - -void tipc_disc_recv_msg(struct sk_buff *buf); - -void tipc_disc_link_event(u32 addr, char *name, int up); -#if 0 -int disc_create_link(const struct tipc_link_create *argv); -#endif +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 index 3bbef2ab22a..5e1426f1751 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -1,8 +1,8 @@ /* * net/tipc/eth_media.c: Ethernet bearer support for TIPC * - * Copyright (c) 2001-2007, Ericsson AB - * Copyright (c) 2005-2007, Wind River Systems + * 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 @@ -34,281 +34,68 @@ * POSSIBILITY OF SUCH DAMAGE. */ -#include <net/tipc/tipc.h> -#include <net/tipc/tipc_bearer.h> -#include <net/tipc/tipc_msg.h> -#include <linux/netdevice.h> -#include <net/net_namespace.h> +#include "core.h" +#include "bearer.h" -#define MAX_ETH_BEARERS 2 -#define ETH_LINK_PRIORITY TIPC_DEF_LINK_PRI -#define ETH_LINK_TOLERANCE TIPC_DEF_LINK_TOL -#define ETH_LINK_WINDOW TIPC_DEF_LINK_WIN +#define ETH_ADDR_OFFSET 4 /* MAC addr position inside address field */ -/** - * struct eth_bearer - Ethernet bearer data structure - * @bearer: ptr to associated "generic" bearer structure - * @dev: ptr to associated Ethernet network device - * @tipc_packet_type: used in binding TIPC to Ethernet driver - */ - -struct eth_bearer { - struct tipc_bearer *bearer; - struct net_device *dev; - struct packet_type tipc_packet_type; -}; - -static struct eth_bearer eth_bearers[MAX_ETH_BEARERS]; -static int eth_started = 0; -static struct notifier_block notifier; - -/** - * send_msg - send a TIPC message out over an Ethernet interface - */ - -static int send_msg(struct sk_buff *buf, struct tipc_bearer *tb_ptr, - struct tipc_media_addr *dest) +/* Convert Ethernet address (media address format) to string */ +static int tipc_eth_addr2str(struct tipc_media_addr *addr, + char *strbuf, int bufsz) { - struct sk_buff *clone; - struct net_device *dev; - - clone = skb_clone(buf, GFP_ATOMIC); - if (clone) { - skb_reset_network_header(clone); - dev = ((struct eth_bearer *)(tb_ptr->usr_handle))->dev; - clone->dev = dev; - dev_hard_header(clone, dev, ETH_P_TIPC, - &dest->dev_addr.eth_addr, - dev->dev_addr, clone->len); - dev_queue_xmit(clone); - } - return TIPC_OK; -} - -/** - * recv_msg - handle incoming TIPC message from an Ethernet interface - * - * Accept only packets explicitly sent to this node, or broadcast packets; - * ignores packets sent using Ethernet multicast, and traffic sent to other - * nodes (which can happen if interface is running in promiscuous mode). - * Routine truncates any Ethernet padding/CRC appended to the message, - * and ensures message size matches actual length - */ - -static int recv_msg(struct sk_buff *buf, struct net_device *dev, - struct packet_type *pt, struct net_device *orig_dev) -{ - struct eth_bearer *eb_ptr = (struct eth_bearer *)pt->af_packet_priv; - u32 size; - - if (dev->nd_net != &init_net) { - kfree_skb(buf); - return 0; - } + if (bufsz < 18) /* 18 = strlen("aa:bb:cc:dd:ee:ff\0") */ + return 1; - if (likely(eb_ptr->bearer)) { - if (likely(buf->pkt_type <= PACKET_BROADCAST)) { - size = msg_size((struct tipc_msg *)buf->data); - skb_trim(buf, size); - if (likely(buf->len == size)) { - buf->next = NULL; - tipc_recv_msg(buf, eb_ptr->bearer); - return TIPC_OK; - } - } - } - kfree_skb(buf); - return TIPC_OK; -} - -/** - * enable_bearer - attach TIPC bearer to an Ethernet interface - */ - -static int enable_bearer(struct tipc_bearer *tb_ptr) -{ - struct net_device *dev = NULL; - struct net_device *pdev = NULL; - struct eth_bearer *eb_ptr = ð_bearers[0]; - struct eth_bearer *stop = ð_bearers[MAX_ETH_BEARERS]; - char *driver_name = strchr((const char *)tb_ptr->name, ':') + 1; - - /* Find device with specified name */ - - for_each_netdev(&init_net, pdev){ - if (!strncmp(pdev->name, driver_name, IFNAMSIZ)) { - dev = pdev; - break; - } - } - if (!dev) - return -ENODEV; - - /* Find Ethernet bearer for device (or create one) */ - - for (;(eb_ptr != stop) && eb_ptr->dev && (eb_ptr->dev != dev); eb_ptr++); - if (eb_ptr == stop) - return -EDQUOT; - if (!eb_ptr->dev) { - eb_ptr->dev = dev; - eb_ptr->tipc_packet_type.type = htons(ETH_P_TIPC); - eb_ptr->tipc_packet_type.dev = dev; - eb_ptr->tipc_packet_type.func = recv_msg; - eb_ptr->tipc_packet_type.af_packet_priv = eb_ptr; - INIT_LIST_HEAD(&(eb_ptr->tipc_packet_type.list)); - dev_hold(dev); - dev_add_pack(&eb_ptr->tipc_packet_type); - } - - /* Associate TIPC bearer with Ethernet bearer */ - - eb_ptr->bearer = tb_ptr; - tb_ptr->usr_handle = (void *)eb_ptr; - tb_ptr->mtu = dev->mtu; - tb_ptr->blocked = 0; - tb_ptr->addr.type = htonl(TIPC_MEDIA_TYPE_ETH); - memcpy(&tb_ptr->addr.dev_addr, &dev->dev_addr, ETH_ALEN); + sprintf(strbuf, "%pM", addr->value); return 0; } -/** - * disable_bearer - detach TIPC bearer from an Ethernet interface - * - * We really should do dev_remove_pack() here, but this function can not be - * called at tasklet level. => Use eth_bearer->bearer as a flag to throw away - * incoming buffers, & postpone dev_remove_pack() to eth_media_stop() on exit. - */ - -static void disable_bearer(struct tipc_bearer *tb_ptr) -{ - ((struct eth_bearer *)tb_ptr->usr_handle)->bearer = NULL; -} - -/** - * recv_notification - handle device updates from OS - * - * Change the state of the Ethernet bearer (if any) associated with the - * specified device. - */ - -static int recv_notification(struct notifier_block *nb, unsigned long evt, - void *dv) +/* Convert from media address format to discovery message addr format */ +static int tipc_eth_addr2msg(char *msg, struct tipc_media_addr *addr) { - struct net_device *dev = (struct net_device *)dv; - struct eth_bearer *eb_ptr = ð_bearers[0]; - struct eth_bearer *stop = ð_bearers[MAX_ETH_BEARERS]; - - if (dev->nd_net != &init_net) - return NOTIFY_DONE; - - while ((eb_ptr->dev != dev)) { - if (++eb_ptr == stop) - return NOTIFY_DONE; /* couldn't find device */ - } - if (!eb_ptr->bearer) - return NOTIFY_DONE; /* bearer had been disabled */ - - eb_ptr->bearer->mtu = dev->mtu; - - switch (evt) { - case NETDEV_CHANGE: - if (netif_carrier_ok(dev)) - tipc_continue(eb_ptr->bearer); - else - tipc_block_bearer(eb_ptr->bearer->name); - break; - case NETDEV_UP: - tipc_continue(eb_ptr->bearer); - break; - case NETDEV_DOWN: - tipc_block_bearer(eb_ptr->bearer->name); - break; - case NETDEV_CHANGEMTU: - case NETDEV_CHANGEADDR: - tipc_block_bearer(eb_ptr->bearer->name); - tipc_continue(eb_ptr->bearer); - break; - case NETDEV_UNREGISTER: - case NETDEV_CHANGENAME: - tipc_disable_bearer(eb_ptr->bearer->name); - break; - } - return NOTIFY_OK; + 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; } -/** - * eth_addr2str - convert Ethernet address to string - */ - -static char *eth_addr2str(struct tipc_media_addr *a, char *str_buf, int str_size) +/* 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) { - unchar *addr = (unchar *)&a->dev_addr; - DECLARE_MAC_BUF(mac); + char bcast_mac[ETH_ALEN] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; - if (str_size < 18) - *str_buf = '\0'; - else - sprintf(str_buf, "%s", print_mac(mac, addr)); - return str_buf; + 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; } -/** - * tipc_eth_media_start - activate Ethernet bearer support - * - * Register Ethernet media type with TIPC bearer code. Also register - * with OS for notifications about device state changes. - */ - -int tipc_eth_media_start(void) +/* 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) { - struct tipc_media_addr bcast_addr; - int res; - - if (eth_started) - return -EINVAL; - - bcast_addr.type = htonl(TIPC_MEDIA_TYPE_ETH); - memset(&bcast_addr.dev_addr, 0xff, ETH_ALEN); - - memset(eth_bearers, 0, sizeof(eth_bearers)); - - res = tipc_register_media(TIPC_MEDIA_TYPE_ETH, "eth", - enable_bearer, disable_bearer, send_msg, - eth_addr2str, &bcast_addr, ETH_LINK_PRIORITY, - ETH_LINK_TOLERANCE, ETH_LINK_WINDOW); - if (res) - return res; - - notifier.notifier_call = &recv_notification; - notifier.priority = 0; - res = register_netdevice_notifier(¬ifier); - if (!res) - eth_started = 1; - return res; + /* Skip past preamble: */ + msg += ETH_ADDR_OFFSET; + return tipc_eth_raw2addr(b, addr, msg); } -/** - * tipc_eth_media_stop - deactivate Ethernet bearer support - */ - -void tipc_eth_media_stop(void) -{ - int i; - - if (!eth_started) - return; - - unregister_netdevice_notifier(¬ifier); - for (i = 0; i < MAX_ETH_BEARERS ; i++) { - if (eth_bearers[i].bearer) { - eth_bearers[i].bearer->blocked = 1; - eth_bearers[i].bearer = NULL; - } - if (eth_bearers[i].dev) { - dev_remove_pack(ð_bearers[i].tipc_packet_type); - dev_put(eth_bearers[i].dev); - } - } - memset(ð_bearers, 0, sizeof(eth_bearers)); - eth_started = 0; -} +/* 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/handler.c b/net/tipc/handler.c deleted file mode 100644 index 0c70010a7df..00000000000 --- a/net/tipc/handler.c +++ /dev/null @@ -1,132 +0,0 @@ -/* - * net/tipc/handler.c: TIPC signal handling - * - * 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. - */ - -#include "core.h" - -struct queue_item { - struct list_head next_signal; - void (*handler) (unsigned long); - unsigned long data; -}; - -static struct kmem_cache *tipc_queue_item_cache; -static struct list_head signal_queue_head; -static DEFINE_SPINLOCK(qitem_lock); -static int handler_enabled = 0; - -static void process_signal_queue(unsigned long dummy); - -static DECLARE_TASKLET_DISABLED(tipc_tasklet, process_signal_queue, 0); - - -unsigned int tipc_k_signal(Handler routine, unsigned long argument) -{ - struct queue_item *item; - - if (!handler_enabled) { - err("Signal request ignored by handler\n"); - return -ENOPROTOOPT; - } - - spin_lock_bh(&qitem_lock); - item = kmem_cache_alloc(tipc_queue_item_cache, GFP_ATOMIC); - if (!item) { - err("Signal queue out of memory\n"); - spin_unlock_bh(&qitem_lock); - return -ENOMEM; - } - item->handler = routine; - item->data = argument; - list_add_tail(&item->next_signal, &signal_queue_head); - spin_unlock_bh(&qitem_lock); - tasklet_schedule(&tipc_tasklet); - return 0; -} - -static void process_signal_queue(unsigned long dummy) -{ - struct queue_item *__volatile__ item; - struct list_head *l, *n; - - spin_lock_bh(&qitem_lock); - list_for_each_safe(l, n, &signal_queue_head) { - item = list_entry(l, struct queue_item, next_signal); - list_del(&item->next_signal); - spin_unlock_bh(&qitem_lock); - item->handler(item->data); - spin_lock_bh(&qitem_lock); - kmem_cache_free(tipc_queue_item_cache, item); - } - spin_unlock_bh(&qitem_lock); -} - -int tipc_handler_start(void) -{ - tipc_queue_item_cache = - kmem_cache_create("tipc_queue_items", sizeof(struct queue_item), - 0, SLAB_HWCACHE_ALIGN, NULL); - if (!tipc_queue_item_cache) - return -ENOMEM; - - INIT_LIST_HEAD(&signal_queue_head); - tasklet_enable(&tipc_tasklet); - handler_enabled = 1; - return 0; -} - -void tipc_handler_stop(void) -{ - struct list_head *l, *n; - struct queue_item *item; - - if (!handler_enabled) - return; - - handler_enabled = 0; - tasklet_disable(&tipc_tasklet); - tasklet_kill(&tipc_tasklet); - - spin_lock_bh(&qitem_lock); - list_for_each_safe(l, n, &signal_queue_head) { - item = list_entry(l, struct queue_item, next_signal); - list_del(&item->next_signal); - kmem_cache_free(tipc_queue_item_cache, item); - } - spin_unlock_bh(&qitem_lock); - - kmem_cache_destroy(tipc_queue_item_cache); -} - 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 index cefa99824c5..ad2c57f5868 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -1,8 +1,8 @@ /* * net/tipc/link.c: TIPC link code * - * Copyright (c) 1996-2007, Ericsson AB - * Copyright (c) 2004-2007, Wind River Systems + * 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 @@ -35,31 +35,30 @@ */ #include "core.h" -#include "dbg.h" #include "link.h" -#include "net.h" -#include "node.h" #include "port.h" -#include "addr.h" -#include "node_subscr.h" +#include "socket.h" #include "name_distr.h" -#include "bearer.h" -#include "name_table.h" #include "discover.h" #include "config.h" -#include "bcast.h" +#include <linux/pkt_sched.h> /* - * Limit for deferred reception queue: + * 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 "; -#define DEF_QUEUE_LIMIT 256u +/* + * 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 */ @@ -75,155 +74,46 @@ /* * State value stored in 'exp_msg_count' */ - #define START_CHANGEOVER 100000u -/** - * struct link_name - deconstructed link name - * @addr_local: network address of node at this end - * @if_local: name of interface at this end - * @addr_peer: network address of node at far end - * @if_peer: name of interface at far end - */ - -struct link_name { - u32 addr_local; - char if_local[TIPC_MAX_IF_NAME]; - u32 addr_peer; - char if_peer[TIPC_MAX_IF_NAME]; -}; - -#if 0 - -/* LINK EVENT CODE IS NOT SUPPORTED AT PRESENT */ - -/** - * struct link_event - link up/down event notification - */ - -struct link_event { - u32 addr; - int up; - void (*fcn)(u32, char *, int); - char name[TIPC_MAX_LINK_NAME]; -}; - -#endif - -static void link_handle_out_of_seq_msg(struct link *l_ptr, +static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, struct sk_buff *buf); -static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf); -static int link_recv_changeover_msg(struct link **l_ptr, struct sk_buff **buf); -static void link_set_supervision_props(struct link *l_ptr, u32 tolerance); -static int link_send_sections_long(struct port *sender, - struct iovec const *msg_sect, - u32 num_sect, u32 destnode); -static void link_check_defragm_bufs(struct link *l_ptr); -static void link_state_event(struct link *l_ptr, u32 event); -static void link_reset_statistics(struct link *l_ptr); -static void link_print(struct link *l_ptr, struct print_buf *buf, - const char *str); - -/* - * Debugging code used by link routines only - * - * When debugging link problems on a system that has multiple links, - * the standard TIPC debugging routines may not be useful since they - * allow the output from multiple links to be intermixed. For this reason - * routines of the form "dbg_link_XXX()" have been created that will capture - * debug info into a link's personal print buffer, which can then be dumped - * into the TIPC system log (TIPC_LOG) upon request. - * - * To enable per-link debugging, use LINK_LOG_BUF_SIZE to specify the size - * of the print buffer used by each link. If LINK_LOG_BUF_SIZE is set to 0, - * the dbg_link_XXX() routines simply send their output to the standard - * debug print buffer (DBG_OUTPUT), if it has been defined; this can be useful - * when there is only a single link in the system being debugged. - * - * Notes: - * - When enabled, LINK_LOG_BUF_SIZE should be set to at least TIPC_PB_MIN_SIZE - * - "l_ptr" must be valid when using dbg_link_XXX() macros - */ - -#define LINK_LOG_BUF_SIZE 0 - -#define dbg_link(fmt, arg...) do {if (LINK_LOG_BUF_SIZE) tipc_printf(&l_ptr->print_buf, fmt, ## arg); } while(0) -#define dbg_link_msg(msg, txt) do {if (LINK_LOG_BUF_SIZE) tipc_msg_print(&l_ptr->print_buf, msg, txt); } while(0) -#define dbg_link_state(txt) do {if (LINK_LOG_BUF_SIZE) link_print(l_ptr, &l_ptr->print_buf, txt); } while(0) -#define dbg_link_dump() do { \ - if (LINK_LOG_BUF_SIZE) { \ - tipc_printf(LOG, "\n\nDumping link <%s>:\n", l_ptr->name); \ - tipc_printbuf_move(LOG, &l_ptr->print_buf); \ - } \ -} while (0) - -static void dbg_print_link(struct link *l_ptr, const char *str) -{ - if (DBG_OUTPUT != TIPC_NULL) - link_print(l_ptr, DBG_OUTPUT, str); -} - -static void dbg_print_buf_chain(struct sk_buff *root_buf) -{ - if (DBG_OUTPUT != TIPC_NULL) { - struct sk_buff *buf = root_buf; - - while (buf) { - msg_dbg(buf_msg(buf), "In chain: "); - buf = buf->next; - } - } -} +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 int link_working_working(struct link *l_ptr) -{ - return (l_ptr->state == WORKING_WORKING); -} - -static int link_working_unknown(struct link *l_ptr) -{ - return (l_ptr->state == WORKING_UNKNOWN); -} - -static int link_reset_unknown(struct link *l_ptr) -{ - return (l_ptr->state == RESET_UNKNOWN); -} - -static int link_reset_reset(struct link *l_ptr) -{ - return (l_ptr->state == RESET_RESET); -} - -static int link_blocked(struct link *l_ptr) -{ - return (l_ptr->exp_msg_count || l_ptr->blocked); -} - -static int link_congested(struct link *l_ptr) -{ - return (l_ptr->out_queue_size >= l_ptr->queue_limit[0]); -} - -static u32 link_max_pkt(struct link *l_ptr) -{ - return l_ptr->max_pkt; -} - -static void link_init_max_pkt(struct link *l_ptr) +static void link_init_max_pkt(struct tipc_link *l_ptr) { + struct tipc_bearer *b_ptr; u32 max_pkt; - max_pkt = (l_ptr->b_ptr->publ.mtu & ~3); + 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; @@ -236,14 +126,14 @@ static void link_init_max_pkt(struct link *l_ptr) l_ptr->max_pkt_probes = 0; } -static u32 link_next_sent(struct link *l_ptr) +static u32 link_next_sent(struct tipc_link *l_ptr) { if (l_ptr->next_out) - return msg_seqno(buf_msg(l_ptr->next_out)); + return buf_seqno(l_ptr->next_out); return mod(l_ptr->next_out_no); } -static u32 link_last_sent(struct link *l_ptr) +static u32 link_last_sent(struct tipc_link *l_ptr) { return mod(link_next_sent(l_ptr) - 1); } @@ -251,117 +141,37 @@ static u32 link_last_sent(struct link *l_ptr) /* * Simple non-static link routines (i.e. referenced outside this file) */ - -int tipc_link_is_up(struct link *l_ptr) +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)); + return link_working_working(l_ptr) || link_working_unknown(l_ptr); } -int tipc_link_is_active(struct link *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_name_validate - validate & (optionally) deconstruct link name - * @name - ptr to link name string - * @name_parts - ptr to area for link name components (or NULL if not needed) - * - * Returns 1 if link name is valid, otherwise 0. - */ - -static int link_name_validate(const char *name, struct link_name *name_parts) -{ - char name_copy[TIPC_MAX_LINK_NAME]; - char *addr_local; - char *if_local; - char *addr_peer; - char *if_peer; - char dummy; - u32 z_local, c_local, n_local; - u32 z_peer, c_peer, n_peer; - u32 if_local_len; - u32 if_peer_len; - - /* copy link name & ensure length is OK */ - - name_copy[TIPC_MAX_LINK_NAME - 1] = 0; - /* need above in case non-Posix strncpy() doesn't pad with nulls */ - strncpy(name_copy, name, TIPC_MAX_LINK_NAME); - if (name_copy[TIPC_MAX_LINK_NAME - 1] != 0) - return 0; - - /* ensure all component parts of link name are present */ - - addr_local = name_copy; - if ((if_local = strchr(addr_local, ':')) == NULL) - return 0; - *(if_local++) = 0; - if ((addr_peer = strchr(if_local, '-')) == NULL) - return 0; - *(addr_peer++) = 0; - if_local_len = addr_peer - if_local; - if ((if_peer = strchr(addr_peer, ':')) == NULL) - return 0; - *(if_peer++) = 0; - if_peer_len = strlen(if_peer) + 1; - - /* validate component parts of link name */ - - if ((sscanf(addr_local, "%u.%u.%u%c", - &z_local, &c_local, &n_local, &dummy) != 3) || - (sscanf(addr_peer, "%u.%u.%u%c", - &z_peer, &c_peer, &n_peer, &dummy) != 3) || - (z_local > 255) || (c_local > 4095) || (n_local > 4095) || - (z_peer > 255) || (c_peer > 4095) || (n_peer > 4095) || - (if_local_len <= 1) || (if_local_len > TIPC_MAX_IF_NAME) || - (if_peer_len <= 1) || (if_peer_len > TIPC_MAX_IF_NAME) || - (strspn(if_local, tipc_alphabet) != (if_local_len - 1)) || - (strspn(if_peer, tipc_alphabet) != (if_peer_len - 1))) - return 0; - - /* return link name components, if necessary */ - - if (name_parts) { - name_parts->addr_local = tipc_addr(z_local, c_local, n_local); - strcpy(name_parts->if_local, if_local); - name_parts->addr_peer = tipc_addr(z_peer, c_peer, n_peer); - strcpy(name_parts->if_peer, if_peer); - } - return 1; + 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 - * - * This routine must not grab "tipc_net_lock" to avoid a potential deadlock conflict - * with tipc_link_delete(). (There is no risk that the node will be deleted by - * another thread because tipc_link_delete() always cancels the link timer before - * tipc_node_delete() is called.) */ - -static void link_timeout(struct link *l_ptr) +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->out_queue_size > l_ptr->stats.max_queue_sz) - l_ptr->stats.max_queue_sz = l_ptr->out_queue_size; - 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)) { + if ((msg_user(msg) == MSG_FRAGMENTER) && + (msg_type(msg) == FIRST_FRAGMENT)) { length = msg_size(msg_get_wrapped(msg)); } if (length) { @@ -386,8 +196,6 @@ static void link_timeout(struct link *l_ptr) /* do all other link processing performed on a periodic basis */ - link_check_defragm_bufs(l_ptr); - link_state_event(l_ptr, TIMEOUT_EVT); if (l_ptr->next_out) @@ -396,69 +204,76 @@ static void link_timeout(struct link *l_ptr) tipc_node_unlock(l_ptr->owner); } -static void link_set_timer(struct link *l_ptr, u32 time) +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 - * @peer: network address of node at other end of link * @media_addr: media address to use when sending messages over link * * Returns pointer to link. */ - -struct link *tipc_link_create(struct bearer *b_ptr, const u32 peer, - const struct tipc_media_addr *media_addr) +struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, + struct tipc_bearer *b_ptr, + const struct tipc_media_addr *media_addr) { - struct link *l_ptr; + struct tipc_link *l_ptr; struct tipc_msg *msg; char *if_name; + char addr_string[16]; + u32 peer = n_ptr->addr; - l_ptr = kzalloc(sizeof(*l_ptr), GFP_ATOMIC); - if (!l_ptr) { - warn("Link creation failed, no memory\n"); + 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 (LINK_LOG_BUF_SIZE) { - char *pb = kmalloc(LINK_LOG_BUF_SIZE, GFP_ATOMIC); + 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; + } - if (!pb) { - kfree(l_ptr); - warn("Link creation failed, no memory for print buffer\n"); - return NULL; - } - tipc_printbuf_init(&l_ptr->print_buf, pb, LINK_LOG_BUF_SIZE); + 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->publ.name, ':') + 1; - sprintf(l_ptr->name, "%u.%u.%u:%s-%u.%u.%u:", + 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 is appended to link name by reset/activate */ + /* 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->b_ptr = b_ptr; - link_set_supervision_props(l_ptr, b_ptr->media->tolerance); + 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; - msg_init(msg, LINK_PROTOCOL, RESET_MSG, TIPC_OK, INT_H_SIZE, l_ptr->addr); + 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); + 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->media->window); + 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; @@ -466,60 +281,46 @@ struct link *tipc_link_create(struct bearer *b_ptr, const u32 peer, link_reset_statistics(l_ptr); - l_ptr->owner = tipc_node_attach_link(l_ptr); - if (!l_ptr->owner) { - if (LINK_LOG_BUF_SIZE) - kfree(l_ptr->print_buf.buf); - kfree(l_ptr); - return NULL; - } + tipc_node_attach_link(n_ptr, l_ptr); - k_init_timer(&l_ptr->timer, (Handler)link_timeout, (unsigned long)l_ptr); - list_add_tail(&l_ptr->link_list, &b_ptr->links); - tipc_k_signal((Handler)tipc_link_start, (unsigned long)l_ptr); + k_init_timer(&l_ptr->timer, (Handler)link_timeout, + (unsigned long)l_ptr); - dbg("tipc_link_create(): tolerance = %u,cont intv = %u, abort_limit = %u\n", - l_ptr->tolerance, l_ptr->continuity_interval, l_ptr->abort_limit); + link_state_event(l_ptr, STARTING_EVT); return l_ptr; } -/** - * tipc_link_delete - delete a link - * @l_ptr: pointer to link - * - * Note: 'tipc_net_lock' is write_locked, bearer is locked. - * This routine must not grab the node lock until after link timer cancellation - * to avoid a potential deadlock situation. - */ - -void tipc_link_delete(struct link *l_ptr) +void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down) { - if (!l_ptr) { - err("Attempt to delete non-existent link\n"); - return; - } + struct tipc_link *l_ptr; + struct tipc_node *n_ptr; - dbg("tipc_link_delete()\n"); - - k_cancel_timer(&l_ptr->timer); - - tipc_node_lock(l_ptr->owner); - tipc_link_reset(l_ptr); - tipc_node_detach_link(l_ptr->owner, l_ptr); - tipc_link_stop(l_ptr); - list_del_init(&l_ptr->link_list); - if (LINK_LOG_BUF_SIZE) - kfree(l_ptr->print_buf.buf); - tipc_node_unlock(l_ptr->owner); - k_term_timer(&l_ptr->timer); - kfree(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->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); -void tipc_link_start(struct link *l_ptr) -{ - dbg("tipc_link_start %x\n", l_ptr); - link_state_event(l_ptr, STARTING_EVT); + /* 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(); } /** @@ -531,21 +332,17 @@ void tipc_link_start(struct link *l_ptr) * Schedules port for renewed sending of messages after link congestion * has abated. */ - -static int link_schedule_port(struct link *l_ptr, u32 origport, u32 sz) +static int link_schedule_port(struct tipc_link *l_ptr, u32 origport, u32 sz) { - struct port *p_ptr; + struct tipc_port *p_ptr; spin_lock_bh(&tipc_port_list_lock); p_ptr = tipc_port_lock(origport); if (p_ptr) { - if (!p_ptr->wakeup) - goto exit; if (!list_empty(&p_ptr->wait_list)) goto exit; - p_ptr->congested_link = l_ptr; - p_ptr->publ.congested = 1; - p_ptr->waiting_pkts = 1 + ((sz - 1) / link_max_pkt(l_ptr)); + 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: @@ -555,10 +352,10 @@ exit: return -ELINKCONG; } -void tipc_link_wakeup_ports(struct link *l_ptr, int all) +void tipc_link_wakeup_ports(struct tipc_link *l_ptr, int all) { - struct port *p_ptr; - struct port *temp_p_ptr; + 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) @@ -574,12 +371,11 @@ void tipc_link_wakeup_ports(struct link *l_ptr, int all) if (win <= 0) break; list_del_init(&p_ptr->wait_list); - p_ptr->congested_link = NULL; - spin_lock_bh(p_ptr->publ.lock); - p_ptr->publ.congested = 0; - p_ptr->wakeup(&p_ptr->publ); + 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->publ.lock); + spin_unlock_bh(p_ptr->lock); } exit: @@ -590,17 +386,9 @@ exit: * link_release_outqueue - purge link's outbound message queue * @l_ptr: pointer to link */ - -static void link_release_outqueue(struct link *l_ptr) +static void link_release_outqueue(struct tipc_link *l_ptr) { - struct sk_buff *buf = l_ptr->first_out; - struct sk_buff *next; - - while (buf) { - next = buf->next; - buf_discard(buf); - buf = next; - } + kfree_skb_list(l_ptr->first_out); l_ptr->first_out = NULL; l_ptr->out_queue_size = 0; } @@ -609,127 +397,57 @@ static void link_release_outqueue(struct link *l_ptr) * tipc_link_reset_fragments - purge link's inbound message fragments queue * @l_ptr: pointer to link */ - -void tipc_link_reset_fragments(struct link *l_ptr) +void tipc_link_reset_fragments(struct tipc_link *l_ptr) { - struct sk_buff *buf = l_ptr->defragm_buf; - struct sk_buff *next; - - while (buf) { - next = buf->next; - buf_discard(buf); - buf = next; - } - l_ptr->defragm_buf = NULL; + kfree_skb(l_ptr->reasm_buf); + l_ptr->reasm_buf = NULL; } /** - * tipc_link_stop - purge all inbound and outbound messages associated with link + * tipc_link_purge_queues - purge all pkt queues associated with link * @l_ptr: pointer to link */ - -void tipc_link_stop(struct link *l_ptr) +void tipc_link_purge_queues(struct tipc_link *l_ptr) { - struct sk_buff *buf; - struct sk_buff *next; - - buf = l_ptr->oldest_deferred_in; - while (buf) { - next = buf->next; - buf_discard(buf); - buf = next; - } - - buf = l_ptr->first_out; - while (buf) { - next = buf->next; - buf_discard(buf); - buf = next; - } - + kfree_skb_list(l_ptr->oldest_deferred_in); + kfree_skb_list(l_ptr->first_out); tipc_link_reset_fragments(l_ptr); - - buf_discard(l_ptr->proto_msg_queue); + kfree_skb(l_ptr->proto_msg_queue); l_ptr->proto_msg_queue = NULL; } -#if 0 - -/* LINK EVENT CODE IS NOT SUPPORTED AT PRESENT */ - -static void link_recv_event(struct link_event *ev) +void tipc_link_reset(struct tipc_link *l_ptr) { - ev->fcn(ev->addr, ev->name, ev->up); - kfree(ev); -} - -static void link_send_event(void (*fcn)(u32 a, char *n, int up), - struct link *l_ptr, int up) -{ - struct link_event *ev; - - ev = kmalloc(sizeof(*ev), GFP_ATOMIC); - if (!ev) { - warn("Link event allocation failure\n"); - return; - } - ev->addr = l_ptr->addr; - ev->up = up; - ev->fcn = fcn; - memcpy(ev->name, l_ptr->name, TIPC_MAX_LINK_NAME); - tipc_k_signal((Handler)link_recv_event, (unsigned long)ev); -} - -#else - -#define link_send_event(fcn, l_ptr, up) do { } while (0) - -#endif - -void tipc_link_reset(struct link *l_ptr) -{ - struct sk_buff *buf; 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); + msg_set_session(l_ptr->pmsg, ((msg_session(l_ptr->pmsg) + 1) & 0xffff)); - /* Link is down, accept any session: */ - l_ptr->peer_session = 0; + /* 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; - dbg_link_state("Resetting Link\n"); 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->b_ptr, l_ptr->addr); -#if 0 - tipc_printf(TIPC_CONS, "\nReset link <%s>\n", l_ptr->name); - dbg_link_dump(); -#endif - if (was_active_link && tipc_node_has_active_links(l_ptr->owner) && - l_ptr->owner->permit_changeover) { + 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); - buf_discard(l_ptr->proto_msg_queue); + kfree_skb(l_ptr->proto_msg_queue); l_ptr->proto_msg_queue = NULL; - buf = l_ptr->oldest_deferred_in; - while (buf) { - struct sk_buff *next = buf->next; - buf_discard(buf); - buf = next; - } + kfree_skb_list(l_ptr->oldest_deferred_in); if (!list_empty(&l_ptr->waiting_ports)) tipc_link_wakeup_ports(l_ptr, 1); @@ -747,21 +465,29 @@ void tipc_link_reset(struct link *l_ptr) l_ptr->fsm_msg_cnt = 0; l_ptr->stale_count = 0; link_reset_statistics(l_ptr); - - link_send_event(tipc_cfg_link_event, l_ptr, 0); - if (!in_own_cluster(l_ptr->addr)) - link_send_event(tipc_disc_link_event, l_ptr, 0); } +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 link *l_ptr) +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->b_ptr, l_ptr->addr); - link_send_event(tipc_cfg_link_event, l_ptr, 1); - if (!in_own_cluster(l_ptr->addr)) - link_send_event(tipc_disc_link_event, l_ptr, 1); + tipc_bearer_add_dest(l_ptr->bearer_id, l_ptr->addr); } /** @@ -769,211 +495,184 @@ static void link_activate(struct link *l_ptr) * @l_ptr: pointer to link * @event: state machine event to process */ - -static void link_state_event(struct link *l_ptr, unsigned event) +static void link_state_event(struct tipc_link *l_ptr, unsigned int event) { - struct link *other; + struct tipc_link *other; u32 cont_intv = l_ptr->continuity_interval; - if (!l_ptr->started && (event != STARTING_EVT)) + if (l_ptr->flags & LINK_STOPPED) + return; + + if (!(l_ptr->flags & LINK_STARTED) && (event != STARTING_EVT)) return; /* Not yet. */ - if (link_blocked(l_ptr)) { - if (event == TIMEOUT_EVT) { + /* Check whether changeover is going on */ + if (l_ptr->exp_msg_count) { + if (event == TIMEOUT_EVT) link_set_timer(l_ptr, cont_intv); - } - return; /* Changeover going on */ + return; } - dbg_link("STATE_EV: <%s> ", l_ptr->name); switch (l_ptr->state) { case WORKING_WORKING: - dbg_link("WW/"); switch (event) { case TRAFFIC_MSG_EVT: - dbg_link("TRF-"); - /* fall through */ case ACTIVATE_MSG: - dbg_link("ACT\n"); break; case TIMEOUT_EVT: - dbg_link("TIM "); 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_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, 0, 0, 0); + 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_send_proto_msg(l_ptr, STATE_MSG, - 1, 0, 0, 0, 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); break; } - dbg_link(" -> WU\n"); l_ptr->state = WORKING_UNKNOWN; l_ptr->fsm_msg_cnt = 0; - tipc_link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 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: - dbg_link("RES -> RR\n"); - info("Resetting link <%s>, requested by peer\n", - l_ptr->name); + 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_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 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: - err("Unknown link event %u in WW state\n", event); + pr_err("%s%u in WW state\n", link_unk_evt, event); } break; case WORKING_UNKNOWN: - dbg_link("WU/"); switch (event) { case TRAFFIC_MSG_EVT: - dbg_link("TRF-"); case ACTIVATE_MSG: - dbg_link("ACT -> WW\n"); l_ptr->state = WORKING_WORKING; l_ptr->fsm_msg_cnt = 0; link_set_timer(l_ptr, cont_intv); break; case RESET_MSG: - dbg_link("RES -> RR\n"); - info("Resetting link <%s>, requested by peer " - "while probing\n", l_ptr->name); + 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_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 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: - dbg_link("TIM "); if (l_ptr->next_in_no != l_ptr->checkpoint) { - dbg_link("-> WW \n"); 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_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, 0, 0, 0); + 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) { - dbg_link("Probing %u/%u,timer = %u ms)\n", - l_ptr->fsm_msg_cnt, l_ptr->abort_limit, - cont_intv / 4); - tipc_link_send_proto_msg(l_ptr, STATE_MSG, - 1, 0, 0, 0, 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); } else { /* Link has failed */ - dbg_link("-> RU (%u probes unanswered)\n", - l_ptr->fsm_msg_cnt); - warn("Resetting link <%s>, peer not responding\n", - l_ptr->name); + 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_send_proto_msg(l_ptr, RESET_MSG, - 0, 0, 0, 0, 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: - err("Unknown link event %u in WU state\n", event); + pr_err("%s%u in WU state\n", link_unk_evt, event); } break; case RESET_UNKNOWN: - dbg_link("RU/"); switch (event) { case TRAFFIC_MSG_EVT: - dbg_link("TRF-\n"); break; case ACTIVATE_MSG: other = l_ptr->owner->active_links[0]; - if (other && link_working_unknown(other)) { - dbg_link("ACT\n"); + if (other && link_working_unknown(other)) break; - } - dbg_link("ACT -> WW\n"); l_ptr->state = WORKING_WORKING; l_ptr->fsm_msg_cnt = 0; link_activate(l_ptr); - tipc_link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + 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: - dbg_link("RES \n"); - dbg_link(" -> RR\n"); l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; - tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 1, 0, 0, 0, 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: - dbg_link("START-"); - l_ptr->started = 1; + l_ptr->flags |= LINK_STARTED; /* fall through */ case TIMEOUT_EVT: - dbg_link("TIM \n"); - tipc_link_send_proto_msg(l_ptr, RESET_MSG, 0, 0, 0, 0, 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: - err("Unknown link event %u in RU state\n", event); + pr_err("%s%u in RU state\n", link_unk_evt, event); } break; case RESET_RESET: - dbg_link("RR/ "); switch (event) { case TRAFFIC_MSG_EVT: - dbg_link("TRF-"); - /* fall through */ case ACTIVATE_MSG: other = l_ptr->owner->active_links[0]; - if (other && link_working_unknown(other)) { - dbg_link("ACT\n"); + if (other && link_working_unknown(other)) break; - } - dbg_link("ACT -> WW\n"); l_ptr->state = WORKING_WORKING; l_ptr->fsm_msg_cnt = 0; link_activate(l_ptr); - tipc_link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + 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: - dbg_link("RES\n"); break; case TIMEOUT_EVT: - dbg_link("TIM\n"); - tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 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); - dbg_link("fsm_msg_cnt %u\n", l_ptr->fsm_msg_cnt); break; default: - err("Unknown link event %u in RR state\n", event); + pr_err("%s%u in RR state\n", link_unk_evt, event); } break; default: - err("Unknown link state %u/%u\n", l_ptr->state, event); + pr_err("Unknown link state %u/%u\n", l_ptr->state, event); } } @@ -981,9 +680,7 @@ static void link_state_event(struct link *l_ptr, unsigned event) * link_bundle_buf(): Append contents of a buffer to * the tail of an existing one. */ - -static int link_bundle_buf(struct link *l_ptr, - struct sk_buff *bundler, +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); @@ -999,22 +696,19 @@ static int link_bundle_buf(struct link *l_ptr, return 0; if (skb_tailroom(bundler) < (pad + size)) return 0; - if (link_max_pkt(l_ptr) < (to_pos + size)) + 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); - dbg("Packed msg # %u(%u octets) into pos %u in buf(#%u)\n", - msg_msgcnt(bundler_msg), size, to_pos, msg_seqno(bundler_msg)); - msg_dbg(msg, "PACKD:"); - buf_discard(buf); + kfree_skb(buf); l_ptr->stats.sent_bundled++; return 1; } -static void link_add_to_outqueue(struct link *l_ptr, +static void link_add_to_outqueue(struct tipc_link *l_ptr, struct sk_buff *buf, struct tipc_msg *msg) { @@ -1029,88 +723,91 @@ static void link_add_to_outqueue(struct link *l_ptr, 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_send_buf() is the 'full path' for messages, called from - * inside TIPC when the 'fast path' in tipc_send_buf + * 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_send_buf(struct link *l_ptr, struct sk_buff *buf) +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 = msg_tot_importance(msg); + u32 imp = tipc_msg_tot_importance(msg); u32 queue_limit = l_ptr->queue_limit[imp]; - u32 max_packet = link_max_pkt(l_ptr); - - msg_set_prevnode(msg, tipc_own_addr); /* If routed message */ + u32 max_packet = l_ptr->max_pkt; /* Match msg importance against queue limits: */ - if (unlikely(queue_size >= queue_limit)) { if (imp <= TIPC_CRITICAL_IMPORTANCE) { - return link_schedule_port(l_ptr, msg_origport(msg), - size); + link_schedule_port(l_ptr, msg_origport(msg), size); + kfree_skb(buf); + return -ELINKCONG; } - msg_dbg(msg, "TIPC: Congestion, throwing away\n"); - buf_discard(buf); + kfree_skb(buf); if (imp > CONN_MANAGER) { - warn("Resetting link <%s>, send queue full", l_ptr->name); + 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_send_long_buf(l_ptr, buf); - - /* Packet can be queued or sent: */ + return tipc_link_frag_xmit(l_ptr, buf); - if (queue_size > l_ptr->stats.max_queue_sz) - l_ptr->stats.max_queue_sz = queue_size; - - if (likely(!tipc_bearer_congested(l_ptr->b_ptr, l_ptr) && - !link_congested(l_ptr))) { + /* Packet can be queued or sent. */ + if (likely(!link_congested(l_ptr))) { link_add_to_outqueue(l_ptr, buf, msg); - if (likely(tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr))) { - l_ptr->unacked_window = 0; - } else { - tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); - l_ptr->stats.bearer_congs++; - l_ptr->next_out = buf; - } + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); + l_ptr->unacked_window = 0; return dsz; } - /* Congestion: can message be bundled ?: */ - + /* 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)) { - tipc_bearer_resolve_congestion(l_ptr->b_ptr, l_ptr); + 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 = buf_acquire(max_packet); + struct sk_buff *bundler = tipc_buf_acquire(max_packet); struct tipc_msg bundler_hdr; if (bundler) { - msg_init(&bundler_hdr, MSG_BUNDLER, OPEN_MSG, - TIPC_OK, INT_H_SIZE, l_ptr->addr); + 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); @@ -1124,135 +821,157 @@ int tipc_link_send_buf(struct link *l_ptr, struct sk_buff *buf) if (!l_ptr->next_out) l_ptr->next_out = buf; link_add_to_outqueue(l_ptr, buf, msg); - tipc_bearer_resolve_congestion(l_ptr->b_ptr, l_ptr); return dsz; } /* - * tipc_link_send(): same as tipc_link_send_buf(), but the link to use has - * not been selected yet, and the the owner node is not locked + * 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_send(struct sk_buff *buf, u32 dest, u32 selector) +int tipc_link_xmit(struct sk_buff *buf, u32 dest, u32 selector) { - struct link *l_ptr; - struct node *n_ptr; + struct tipc_link *l_ptr; + struct tipc_node *n_ptr; int res = -ELINKCONG; - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_select(dest, selector); + 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) { - dbg("tipc_link_send: found link %x for dest %x\n", l_ptr, dest); - res = tipc_link_send_buf(l_ptr, buf); - } else { - dbg("Attempt to send msg to unreachable node:\n"); - msg_dbg(buf_msg(buf),">>>"); - buf_discard(buf); - } + if (l_ptr) + res = __tipc_link_xmit(l_ptr, buf); + else + kfree_skb(buf); tipc_node_unlock(n_ptr); } else { - dbg("Attempt to send msg to unknown node:\n"); - msg_dbg(buf_msg(buf),">>>"); - buf_discard(buf); + kfree_skb(buf); } - read_unlock_bh(&tipc_net_lock); return res; } /* - * link_send_buf_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. + * 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); +} -static int link_send_buf_fast(struct link *l_ptr, struct sk_buff *buf, - u32 *used_max_pkt) +/* + * 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); - int res = msg_data_sz(msg); - if (likely(!link_congested(l_ptr))) { - if (likely(msg_size(msg) <= link_max_pkt(l_ptr))) { - if (likely(list_empty(&l_ptr->b_ptr->cong_links))) { - link_add_to_outqueue(l_ptr, buf, msg); - if (likely(tipc_bearer_send(l_ptr->b_ptr, buf, - &l_ptr->media_addr))) { - l_ptr->unacked_window = 0; - msg_dbg(msg,"SENT_FAST:"); - return res; - } - dbg("failed sent fast...\n"); - tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); - l_ptr->stats.bearer_congs++; - l_ptr->next_out = buf; - return res; - } - } - else - *used_max_pkt = link_max_pkt(l_ptr); - } - return tipc_link_send_buf(l_ptr, buf); /* All other cases */ + n->bclink.last_sent = n->bclink.last_in = msg_last_bcast(msg); + n->bclink.recv_permitted = true; + kfree_skb(buf); } /* - * tipc_send_buf_fast: Entry for data messages where the - * destination node is known and the header is complete, - * inclusive total message length. - * Returns user data length. + * 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. */ -int tipc_send_buf_fast(struct sk_buff *buf, u32 destnode) +void tipc_link_names_xmit(struct list_head *message_list, u32 dest) { - struct link *l_ptr; - struct node *n_ptr; - int res; - u32 selector = msg_origport(buf_msg(buf)) & 1; - u32 dummy; + struct tipc_node *n_ptr; + struct tipc_link *l_ptr; + struct sk_buff *buf; + struct sk_buff *temp_buf; - if (destnode == tipc_own_addr) - return tipc_port_recv_msg(buf); + if (list_empty(message_list)) + return; - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_select(destnode, selector); - if (likely(n_ptr)) { + n_ptr = tipc_node_find(dest); + if (n_ptr) { tipc_node_lock(n_ptr); - l_ptr = n_ptr->active_links[selector]; - dbg("send_fast: buf %x selected %x, destnode = %x\n", - buf, l_ptr, destnode); - if (likely(l_ptr)) { - res = link_send_buf_fast(l_ptr, buf, &dummy); - tipc_node_unlock(n_ptr); - read_unlock_bh(&tipc_net_lock); - return res; + 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); } - read_unlock_bh(&tipc_net_lock); - res = msg_data_sz(buf_msg(buf)); - tipc_reject_msg(buf, TIPC_ERR_NO_NODE); - return res; + + /* 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_send_sections_fast: Entry for messages where the + * 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_send_sections_fast(struct port *sender, - struct iovec const *msg_sect, - const u32 num_sect, - u32 destaddr) +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->publ.phdr; - struct link *l_ptr; + struct tipc_msg *hdr = &sender->phdr; + struct tipc_link *l_ptr; struct sk_buff *buf; - struct node *node; + struct tipc_node *node; int res; u32 selector = msg_origport(hdr) & 1; @@ -1261,38 +980,28 @@ 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; - res = msg_build(hdr, msg_sect, num_sect, sender->publ.max_pkt, - !sender->user_port, &buf); - - read_lock_bh(&tipc_net_lock); - node = tipc_node_select(destaddr, selector); + 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 = link_send_buf_fast(l_ptr, buf, - &sender->publ.max_pkt); - if (unlikely(res < 0)) - buf_discard(buf); + res = tipc_link_xmit_fast(l_ptr, buf, + &sender->max_pkt); exit: tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); return res; } - /* Exit if build request was invalid */ - - if (unlikely(res < 0)) - goto exit; - /* Exit if link (or bearer) is congested */ - - if (link_congested(l_ptr) || - !list_empty(&l_ptr->b_ptr->cong_links)) { + if (link_congested(l_ptr)) { res = link_schedule_port(l_ptr, - sender->publ.ref, res); + sender->ref, res); goto exit; } @@ -1300,34 +1009,27 @@ exit: * Message size exceeds max_pkt hint; update hint, * then re-try fast path or fragment the message */ - - sender->publ.max_pkt = link_max_pkt(l_ptr); + sender->max_pkt = l_ptr->max_pkt; tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); - if ((msg_hdr_sz(hdr) + res) <= sender->publ.max_pkt) + if ((msg_hdr_sz(hdr) + res) <= sender->max_pkt) goto again; - return link_send_sections_long(sender, msg_sect, - num_sect, destaddr); + return tipc_link_iovec_long_xmit(sender, msg_sect, + len, destaddr); } tipc_node_unlock(node); } - read_unlock_bh(&tipc_net_lock); /* Couldn't find a link to the destination node */ - - if (buf) - return tipc_reject_msg(buf, TIPC_ERR_NO_NODE); - if (res >= 0) - return tipc_port_reject_sections(sender, hdr, msg_sect, num_sect, - TIPC_ERR_NO_NODE); - return res; + kfree_skb(buf); + tipc_port_iovec_reject(sender, hdr, msg_sect, len, TIPC_ERR_NO_NODE); + return -ENETUNREACH; } /* - * link_send_sections_long(): Entry for long messages where the + * 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, @@ -1340,26 +1042,26 @@ exit: * * Returns user data length or errno. */ -static int link_send_sections_long(struct port *sender, - struct iovec const *msg_sect, - u32 num_sect, - u32 destaddr) -{ - struct link *l_ptr; - struct node *node; - struct tipc_msg *hdr = &sender->publ.phdr; - u32 dsz = msg_data_sz(hdr); - u32 max_pkt,fragm_sz,rest; +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 *sect_crs; + 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->publ.max_pkt - INT_H_SIZE; + 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 */ @@ -1370,28 +1072,22 @@ again: sect_crs = NULL; curr_sect = -1; - /* Prepare reusable fragment header: */ - - msg_dbg(hdr, ">FRAGMENTING>"); - msg_init(&fragm_hdr, MSG_FRAGMENTER, FIRST_FRAGMENT, - TIPC_OK, INT_H_SIZE, msg_destnode(hdr)); - msg_set_link_selector(&fragm_hdr, sender->publ.ref); + /* 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 = buf_acquire(max_pkt); + /* 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); - msg_dbg(buf_msg(buf), ">BUILD>"); - - /* Chop up message: */ + /* Chop up message */ fragm_crs = INT_H_SIZE + hsz; fragm_rest = fragm_sz - hsz; @@ -1400,7 +1096,7 @@ again: if (!sect_rest) { sect_rest = msg_sect[++curr_sect].iov_len; - sect_crs = (const unchar *)msg_sect[curr_sect].iov_base; + sect_crs = msg_sect[curr_sect].iov_base; } if (sect_rest < fragm_rest) @@ -1408,18 +1104,12 @@ again: else sz = fragm_rest; - if (likely(!sender->user_port)) { - if (copy_from_user(buf->data + fragm_crs, sect_crs, sz)) { + if (copy_from_user(buf->data + fragm_crs, sect_crs, sz)) { + res = -EFAULT; error: - for (; buf_chain; buf_chain = buf) { - buf = buf_chain->next; - buf_discard(buf_chain); - } - return -EFAULT; - } - } else - skb_copy_to_linear_data_offset(buf, fragm_crs, - sect_crs, sz); + kfree_skb_list(buf_chain); + return res; + } sect_crs += sz; sect_rest -= sz; fragm_crs += sz; @@ -1431,78 +1121,58 @@ error: /* Initiate new fragment: */ if (rest <= fragm_sz) { fragm_sz = rest; - msg_set_type(&fragm_hdr,LAST_FRAGMENT); + 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 = buf_acquire(fragm_sz + INT_H_SIZE); - if (!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; - msg_dbg(buf_msg(buf)," >BUILD>"); } - } - while (rest > 0); + } while (rest > 0); /* * Now we have a buffer chain. Select a link and check * that packet size is still OK */ - node = tipc_node_select(destaddr, sender->publ.ref & 1); + node = tipc_node_find(destaddr); if (likely(node)) { tipc_node_lock(node); - l_ptr = node->active_links[sender->publ.ref & 1]; + l_ptr = node->active_links[sender->ref & 1]; if (!l_ptr) { tipc_node_unlock(node); goto reject; } - if (link_max_pkt(l_ptr) < max_pkt) { - sender->publ.max_pkt = link_max_pkt(l_ptr); + if (l_ptr->max_pkt < max_pkt) { + sender->max_pkt = l_ptr->max_pkt; tipc_node_unlock(node); - for (; buf_chain; buf_chain = buf) { - buf = buf_chain->next; - buf_discard(buf_chain); - } + kfree_skb_list(buf_chain); goto again; } } else { reject: - for (; buf_chain; buf_chain = buf) { - buf = buf_chain->next; - buf_discard(buf_chain); - } - return tipc_port_reject_sections(sender, hdr, msg_sect, num_sect, - TIPC_ERR_NO_NODE); + kfree_skb_list(buf_chain); + tipc_port_iovec_reject(sender, hdr, msg_sect, len, + TIPC_ERR_NO_NODE); + return -ENETUNREACH; } - /* Append whole chain to send queue: */ - - buf = buf_chain; - l_ptr->long_msg_seq_no = mod(l_ptr->long_msg_seq_no + 1); - if (!l_ptr->next_out) - l_ptr->next_out = buf_chain; + /* 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++; - while (buf) { - struct sk_buff *next = buf->next; - struct tipc_msg *msg = buf_msg(buf); - - l_ptr->stats.sent_fragments++; - msg_set_long_msgno(msg, l_ptr->long_msg_seq_no); - link_add_to_outqueue(l_ptr, buf, msg); - msg_dbg(msg, ">ADD>"); - buf = next; - } - - /* Send it, if possible: */ - tipc_link_push_queue(l_ptr); tipc_node_unlock(node); return dsz; @@ -1511,7 +1181,7 @@ reject: /* * tipc_link_push_packet: Push one unsent packet to the media */ -u32 tipc_link_push_packet(struct link *l_ptr) +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; @@ -1519,11 +1189,10 @@ u32 tipc_link_push_packet(struct link *l_ptr) /* 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 = msg_seqno(buf_msg(buf)); + u32 first = buf_seqno(buf); while (buf && less(first, r_q_head)) { first = mod(first + 1); @@ -1534,164 +1203,124 @@ u32 tipc_link_push_packet(struct link *l_ptr) } /* Continue retransmission now, if there is anything: */ - - if (r_q_size && buf && !skb_cloned(buf)) { + 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); - if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { - msg_dbg(buf_msg(buf), ">DEF-RETR>"); - l_ptr->retransm_queue_head = mod(++r_q_head); - l_ptr->retransm_queue_size = --r_q_size; - l_ptr->stats.retransmitted++; - return TIPC_OK; - } else { - l_ptr->stats.bearer_congs++; - msg_dbg(buf_msg(buf), "|>DEF-RETR>"); - return PUSH_FAILED; - } + 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); - if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { - msg_dbg(buf_msg(buf), ">DEF-PROT>"); - l_ptr->unacked_window = 0; - buf_discard(buf); - l_ptr->proto_msg_queue = NULL; - return TIPC_OK; - } else { - msg_dbg(buf_msg(buf), "|>DEF-PROT>"); - l_ptr->stats.bearer_congs++; - return PUSH_FAILED; - } + 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 = msg_seqno(buf_msg(l_ptr->first_out)); + 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); - if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { - if (msg_user(msg) == MSG_BUNDLER) - msg_set_type(msg, CLOSED_MSG); - msg_dbg(msg, ">PUSH-DATA>"); - l_ptr->next_out = buf->next; - return TIPC_OK; - } else { - msg_dbg(msg, "|PUSH-DATA|"); - l_ptr->stats.bearer_congs++; - return PUSH_FAILED; - } + 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 PUSH_FINISHED; + 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 link *l_ptr) +void tipc_link_push_queue(struct tipc_link *l_ptr) { u32 res; - if (tipc_bearer_congested(l_ptr->b_ptr, l_ptr)) - return; - do { res = tipc_link_push_packet(l_ptr); - } - while (res == TIPC_OK); - if (res == PUSH_FAILED) - tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); + } while (!res); } -static void link_reset_all(unsigned long addr) +void tipc_link_reset_all(struct tipc_node *node) { - struct node *n_ptr; char addr_string[16]; u32 i; - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_find((u32)addr); - if (!n_ptr) { - read_unlock_bh(&tipc_net_lock); - return; /* node no longer exists */ - } - - tipc_node_lock(n_ptr); + tipc_node_lock(node); - warn("Resetting all links to %s\n", - addr_string_fill(addr_string, n_ptr->addr)); + pr_warn("Resetting all links to %s\n", + tipc_addr_string_fill(addr_string, node->addr)); for (i = 0; i < MAX_BEARERS; i++) { - if (n_ptr->links[i]) { - link_print(n_ptr->links[i], TIPC_OUTPUT, - "Resetting link\n"); - tipc_link_reset(n_ptr->links[i]); + if (node->links[i]) { + link_print(node->links[i], "Resetting link\n"); + tipc_link_reset(node->links[i]); } } - tipc_node_unlock(n_ptr); - read_unlock_bh(&tipc_net_lock); + tipc_node_unlock(node); } -static void link_retransmit_failure(struct link *l_ptr, struct sk_buff *buf) +static void link_retransmit_failure(struct tipc_link *l_ptr, + struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); - warn("Retransmission failure on link <%s>\n", l_ptr->name); - tipc_msg_print(TIPC_OUTPUT, msg, ">RETR-FAIL>"); + pr_warn("Retransmission failure on link <%s>\n", l_ptr->name); if (l_ptr->addr) { - /* Handle failure on standard link */ - - link_print(l_ptr, TIPC_OUTPUT, "Resetting link\n"); + link_print(l_ptr, "Resetting link\n"); tipc_link_reset(l_ptr); } else { - /* Handle failure on broadcast link */ - - struct node *n_ptr; + struct tipc_node *n_ptr; char addr_string[16]; - tipc_printf(TIPC_OUTPUT, "Msg seq number: %u, ", msg_seqno(msg)); - tipc_printf(TIPC_OUTPUT, "Outstanding acks: %lu\n", - (unsigned long) TIPC_SKB_CB(buf)->handle); + pr_info("Msg seq number: %u, ", msg_seqno(msg)); + pr_cont("Outstanding acks: %lu\n", + (unsigned long) TIPC_SKB_CB(buf)->handle); - n_ptr = l_ptr->owner->next; + n_ptr = tipc_bclink_retransmit_to(); tipc_node_lock(n_ptr); - addr_string_fill(addr_string, n_ptr->addr); - tipc_printf(TIPC_OUTPUT, "Multicast link info for %s\n", addr_string); - tipc_printf(TIPC_OUTPUT, "Supported: %d, ", n_ptr->bclink.supported); - tipc_printf(TIPC_OUTPUT, "Acked: %u\n", n_ptr->bclink.acked); - tipc_printf(TIPC_OUTPUT, "Last in: %u, ", n_ptr->bclink.last_in); - tipc_printf(TIPC_OUTPUT, "Gap after: %u, ", n_ptr->bclink.gap_after); - tipc_printf(TIPC_OUTPUT, "Gap to: %u\n", n_ptr->bclink.gap_to); - tipc_printf(TIPC_OUTPUT, "Nack sync: %u\n\n", n_ptr->bclink.nack_sync); - - tipc_k_signal((Handler)link_reset_all, (unsigned long)n_ptr->addr); + 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 link *l_ptr, struct sk_buff *buf, +void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *buf, u32 retransmits) { struct tipc_msg *msg; @@ -1701,73 +1330,34 @@ void tipc_link_retransmit(struct link *l_ptr, struct sk_buff *buf, msg = buf_msg(buf); - dbg("Retransmitting %u in link %x\n", retransmits, l_ptr); - - if (tipc_bearer_congested(l_ptr->b_ptr, l_ptr)) { - if (!skb_cloned(buf)) { - msg_dbg(msg, ">NO_RETR->BCONG>"); - dbg_print_link(l_ptr, " "); - l_ptr->retransm_queue_head = msg_seqno(msg); - l_ptr->retransm_queue_size = retransmits; + /* 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 { - /* Don't retransmit if driver already has the buffer */ } } else { - /* Detect repeated retransmit failures on uncongested bearer */ - - 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; - } + l_ptr->last_retransmitted = msg_seqno(msg); + l_ptr->stale_count = 1; } - while (retransmits && (buf != l_ptr->next_out) && buf && !skb_cloned(buf)) { + 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); - if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { - msg_dbg(buf_msg(buf), ">RETR>"); - buf = buf->next; - retransmits--; - l_ptr->stats.retransmitted++; - } else { - tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); - l_ptr->stats.bearer_congs++; - l_ptr->retransm_queue_head = msg_seqno(buf_msg(buf)); - l_ptr->retransm_queue_size = retransmits; - return; - } + 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_recv_non_seq: Receive packets which are outside - * the link sequence flow - */ - -static void link_recv_non_seq(struct sk_buff *buf) -{ - struct tipc_msg *msg = buf_msg(buf); - - if (msg_user(msg) == LINK_CONFIG) - tipc_disc_recv_msg(buf); - else - tipc_bclink_recv_pkt(buf); -} - /** * link_insert_deferred_queue - insert deferred messages back into receive chain */ - -static struct sk_buff *link_insert_deferred_queue(struct link *l_ptr, +static struct sk_buff *link_insert_deferred_queue(struct tipc_link *l_ptr, struct sk_buff *buf) { u32 seq_no; @@ -1775,7 +1365,7 @@ static struct sk_buff *link_insert_deferred_queue(struct link *l_ptr, if (l_ptr->oldest_deferred_in == NULL) return buf; - seq_no = msg_seqno(buf_msg(l_ptr->oldest_deferred_in)); + 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; @@ -1785,65 +1375,143 @@ static struct sk_buff *link_insert_deferred_queue(struct link *l_ptr, return buf; } -void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) +/** + * 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) { - read_lock_bh(&tipc_net_lock); while (head) { - struct bearer *b_ptr; - struct node *n_ptr; - struct link *l_ptr; + struct tipc_node *n_ptr; + struct tipc_link *l_ptr; struct sk_buff *crs; struct sk_buff *buf = head; - struct tipc_msg *msg = buf_msg(buf); - u32 seq_no = msg_seqno(msg); - u32 ackd = msg_ack(msg); + struct tipc_msg *msg; + u32 seq_no; + u32 ackd; u32 released = 0; - int type; - - b_ptr = (struct bearer *)tb_ptr; - TIPC_SKB_CB(buf)->handle = b_ptr; head = head->next; - if (unlikely(msg_version(msg) != TIPC_VERSION)) - goto cont; -#if 0 - if (msg_user(msg) != LINK_PROTOCOL) -#endif - msg_dbg(msg,"<REC<"); + 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))) { - link_recv_non_seq(buf); + 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 cont; + goto discard; + /* Locate neighboring node that sent message */ n_ptr = tipc_node_find(msg_prevnode(msg)); if (unlikely(!n_ptr)) - goto cont; - + 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)) { - tipc_node_unlock(n_ptr); - goto cont; - } - /* - * Release acked messages - */ - if (less(n_ptr->bclink.acked, msg_bcast_ack(msg))) { - if (tipc_node_is_up(n_ptr) && n_ptr->bclink.supported) - tipc_bclink_acknowledge(n_ptr, msg_bcast_ack(msg)); - } + 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(msg_seqno(buf_msg(crs)), ackd)) { + less_eq(buf_seqno(crs), ackd)) { struct sk_buff *next = crs->next; - - buf_discard(crs); + kfree_skb(crs); crs = next; released++; } @@ -1851,118 +1519,122 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) 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_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); } -protocol_check: - if (likely(link_working_working(l_ptr))) { - if (likely(seq_no == mod(l_ptr->next_in_no))) { - l_ptr->next_in_no++; - if (unlikely(l_ptr->oldest_deferred_in)) - head = link_insert_deferred_queue(l_ptr, - head); - if (likely(msg_is_dest(msg, tipc_own_addr))) { -deliver: - if (likely(msg_isdata(msg))) { - tipc_node_unlock(n_ptr); - tipc_port_recv_msg(buf); - continue; - } - switch (msg_user(msg)) { - case MSG_BUNDLER: - l_ptr->stats.recv_bundles++; - l_ptr->stats.recv_bundled += - msg_msgcnt(msg); - tipc_node_unlock(n_ptr); - tipc_link_recv_bundle(buf); - continue; - case ROUTE_DISTRIBUTOR: - tipc_node_unlock(n_ptr); - tipc_cltr_recv_routing_table(buf); - continue; - case NAME_DISTRIBUTOR: - tipc_node_unlock(n_ptr); - tipc_named_recv(buf); - continue; - case CONN_MANAGER: - tipc_node_unlock(n_ptr); - tipc_port_recv_proto_msg(buf); - continue; - case MSG_FRAGMENTER: - l_ptr->stats.recv_fragments++; - if (tipc_link_recv_fragment(&l_ptr->defragm_buf, - &buf, &msg)) { - l_ptr->stats.recv_fragmented++; - goto deliver; - } - break; - case CHANGEOVER_PROTOCOL: - type = msg_type(msg); - if (link_recv_changeover_msg(&l_ptr, &buf)) { - msg = buf_msg(buf); - seq_no = msg_seqno(msg); - TIPC_SKB_CB(buf)->handle - = b_ptr; - if (type == ORIGINAL_MSG) - goto deliver; - goto protocol_check; - } - break; - } - } + /* 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); - tipc_net_route_msg(buf); 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; } - - if (msg_user(msg) == LINK_PROTOCOL) { - link_recv_proto_msg(l_ptr, buf); + l_ptr->next_in_no++; + if (unlikely(l_ptr->oldest_deferred_in)) head = link_insert_deferred_queue(l_ptr, head); - tipc_node_unlock(n_ptr); - continue; + + /* 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; + } } - msg_dbg(msg,"NSEQ<REC<"); - link_state_event(l_ptr, TRAFFIC_MSG_EVT); - if (link_working_working(l_ptr)) { - /* Re-insert in front of queue */ - msg_dbg(msg,"RECV-REINS:"); - buf->next = head; - head = buf; + 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); -cont: - buf_discard(buf); + continue; +unlock_discard: + tipc_node_unlock(n_ptr); +discard: + kfree_skb(buf); } - read_unlock_bh(&tipc_net_lock); } -/* - * link_defer_buf(): Sort a received out-of-sequence packet - * into the deferred reception queue. - * Returns the increase of the queue length,i.e. 0 or 1 +/** + * 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, +u32 tipc_link_defer_pkt(struct sk_buff **head, struct sk_buff **tail, struct sk_buff *buf) { - struct sk_buff *prev = NULL; - struct sk_buff *crs = *head; - u32 seq_no = msg_seqno(buf_msg(buf)); + struct sk_buff *queue_buf; + struct sk_buff **prev; + u32 seq_no = buf_seqno(buf); buf->next = NULL; @@ -1973,67 +1645,58 @@ u32 tipc_link_defer_pkt(struct sk_buff **head, } /* Last ? */ - if (less(msg_seqno(buf_msg(*tail)), seq_no)) { + if (less(buf_seqno(*tail), seq_no)) { (*tail)->next = buf; *tail = buf; return 1; } - /* Scan through queue and sort it in */ - do { - struct tipc_msg *msg = buf_msg(crs); + /* Locate insertion point in queue, then insert; discard if duplicate */ + prev = head; + queue_buf = *head; + for (;;) { + u32 curr_seqno = buf_seqno(queue_buf); - if (less(seq_no, msg_seqno(msg))) { - buf->next = crs; - if (prev) - prev->next = buf; - else - *head = buf; - return 1; + if (seq_no == curr_seqno) { + kfree_skb(buf); + return 0; } - if (seq_no == msg_seqno(msg)) { + + if (less(seq_no, curr_seqno)) break; - } - prev = crs; - crs = crs->next; - } - while (crs); - /* Message is a duplicate of an existing message */ + prev = &queue_buf->next; + queue_buf = queue_buf->next; + } - buf_discard(buf); - return 0; + 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 link *l_ptr, +static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, struct sk_buff *buf) { - u32 seq_no = msg_seqno(buf_msg(buf)); + u32 seq_no = buf_seqno(buf); if (likely(msg_user(buf_msg(buf)) == LINK_PROTOCOL)) { - link_recv_proto_msg(l_ptr, buf); + tipc_link_proto_rcv(l_ptr, buf); return; } - dbg("rx OOS msg: seq_no %u, expecting %u (%u)\n", - seq_no, mod(l_ptr->next_in_no), l_ptr->next_in_no); - /* 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++; - buf_discard(buf); + kfree_skb(buf); return; } @@ -2041,8 +1704,9 @@ static void link_handle_out_of_seq_msg(struct link *l_ptr, &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_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); } else l_ptr->stats.duplicates++; } @@ -2050,18 +1714,32 @@ static void link_handle_out_of_seq_msg(struct link *l_ptr, /* * Send protocol message to the other endpoint. */ -void tipc_link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, - u32 gap, u32 tolerance, u32 priority, u32 ack_mtu) +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; - if (link_blocked(l_ptr)) + /* 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->b_ptr->net_plane); - msg_set_bcast_ack(msg, mod(l_ptr->owner->bclink.last_in)); + 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) { @@ -2070,10 +1748,10 @@ void tipc_link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, if (!tipc_link_is_up(l_ptr)) return; if (l_ptr->next_out) - next_sent = msg_seqno(buf_msg(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 = msg_seqno(buf_msg(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); @@ -2106,58 +1784,29 @@ void tipc_link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_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); } - if (tipc_node_has_redundant_links(l_ptr->owner)) { - msg_set_redundant_link(msg); - } else { - msg_clear_redundant_link(msg); - } + 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); - - /* Ensure sequence number will not fit : */ + msg_set_size(msg, msg_size); msg_set_seqno(msg, mod(l_ptr->next_out_no + (0xffff/2))); - /* Congestion? */ - - if (tipc_bearer_congested(l_ptr->b_ptr, l_ptr)) { - if (!l_ptr->proto_msg_queue) { - l_ptr->proto_msg_queue = - buf_acquire(sizeof(l_ptr->proto_msg)); - } - buf = l_ptr->proto_msg_queue; - if (!buf) - return; - skb_copy_to_linear_data(buf, msg, sizeof(l_ptr->proto_msg)); - return; - } - msg_set_timestamp(msg, jiffies_to_msecs(jiffies)); - - /* Message can be sent */ - - msg_dbg(msg, ">>"); - - buf = buf_acquire(msg_size); + buf = tipc_buf_acquire(msg_size); if (!buf) return; skb_copy_to_linear_data(buf, msg, sizeof(l_ptr->proto_msg)); - msg_set_size(buf_msg(buf), msg_size); - - if (tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { - l_ptr->unacked_window = 0; - buf_discard(buf); - return; - } + buf->priority = TC_PRIO_CONTROL; - /* New congestion */ - tipc_bearer_schedule(l_ptr->b_ptr, l_ptr); - l_ptr->proto_msg_queue = buf; - l_ptr->stats.bearer_congs++; + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); + l_ptr->unacked_window = 0; + kfree_skb(buf); } /* @@ -2165,8 +1814,7 @@ void tipc_link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, * Note that network plane id propagates through the network, and may * change at any time. The node with lowest address rules */ - -static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf) +static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) { u32 rec_gap = 0; u32 max_pkt_info; @@ -2174,39 +1822,41 @@ static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf) u32 msg_tol; struct tipc_msg *msg = buf_msg(buf); - dbg("AT(%u):", jiffies_to_msecs(jiffies)); - msg_dbg(msg, "<<"); - if (link_blocked(l_ptr)) + /* Discard protocol message during link changeover */ + if (l_ptr->exp_msg_count) goto exit; - /* record unnumbered packet arrival (force mismatch on next timeout) */ - - l_ptr->checkpoint--; - - if (l_ptr->b_ptr->net_plane != msg_net_plane(msg)) + if (l_ptr->net_plane != msg_net_plane(msg)) if (tipc_own_addr > msg_prevnode(msg)) - l_ptr->b_ptr->net_plane = msg_net_plane(msg); - - l_ptr->owner->permit_changeover = msg_redundant_link(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) { - if (msg_session(msg) == l_ptr->peer_session) { - dbg("Duplicate RESET: %u<->%u\n", - msg_session(msg), l_ptr->peer_session); - break; /* duplicate: ignore */ - } + 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)); - if ((msg_tol = msg_link_tolerance(msg)) && - (msg_tol > l_ptr->tolerance)) + 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) @@ -2221,31 +1871,40 @@ static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf) } else { l_ptr->max_pkt = l_ptr->max_pkt_target; } - l_ptr->owner->bclink.supported = (max_pkt_info != 0); - link_state_event(l_ptr, msg_type(msg)); + /* 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); - /* Synchronize broadcast sequence numbers */ - if (!tipc_node_has_redundant_links(l_ptr->owner)) { - l_ptr->owner->bclink.last_in = mod(msg_last_bcast(msg)); - } + if (msg_type(msg) == ACTIVATE_MSG) + link_state_event(l_ptr, ACTIVATE_MSG); break; case STATE_MSG: - if ((msg_tol = msg_link_tolerance(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)) { - warn("Resetting link <%s>, priority change %u->%u\n", - l_ptr->name, l_ptr->priority, msg_linkprio(msg)); + 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)) @@ -2258,8 +1917,6 @@ static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf) max_pkt_ack = msg_max_pkt(msg); if (max_pkt_ack > l_ptr->max_pkt) { - dbg("Link <%s> updated MTU %u -> %u\n", - l_ptr->name, l_ptr->max_pkt, max_pkt_ack); l_ptr->max_pkt = max_pkt_ack; l_ptr->max_pkt_probes = 0; } @@ -2267,111 +1924,93 @@ static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf) max_pkt_ack = 0; if (msg_probe(msg)) { l_ptr->stats.recv_probes++; - if (msg_size(msg) > sizeof(l_ptr->proto_msg)) { + if (msg_size(msg) > sizeof(l_ptr->proto_msg)) max_pkt_ack = msg_size(msg); - } } /* Protocol message before retransmits, reduce loss risk */ - - tipc_bclink_check_gap(l_ptr->owner, msg_last_bcast(msg)); + 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_send_proto_msg(l_ptr, STATE_MSG, - 0, rec_gap, 0, 0, max_pkt_ack); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, rec_gap, 0, + 0, max_pkt_ack); } if (msg_seq_gap(msg)) { - msg_dbg(msg, "With Gap:"); l_ptr->stats.recv_nacks++; tipc_link_retransmit(l_ptr, l_ptr->first_out, msg_seq_gap(msg)); } break; - default: - msg_dbg(buf_msg(buf), "<DISCARDING UNKNOWN<"); } exit: - buf_discard(buf); + kfree_skb(buf); } -/* - * tipc_link_tunnel(): Send one message via a link belonging to - * another bearer. Owner node is locked. +/* tipc_link_tunnel_xmit(): Tunnel one packet via a link belonging to + * a different bearer. Owner node is locked. */ -void tipc_link_tunnel(struct link *l_ptr, - struct tipc_msg *tunnel_hdr, - struct tipc_msg *msg, - u32 selector) +static void tipc_link_tunnel_xmit(struct tipc_link *l_ptr, + struct tipc_msg *tunnel_hdr, + struct tipc_msg *msg, + u32 selector) { - struct link *tunnel; + 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)) { - warn("Link changeover error, " - "tunnel link no longer available\n"); + pr_warn("%stunnel link no longer available\n", link_co_err); return; } msg_set_size(tunnel_hdr, length + INT_H_SIZE); - buf = buf_acquire(length + INT_H_SIZE); + buf = tipc_buf_acquire(length + INT_H_SIZE); if (!buf) { - warn("Link changeover error, " - "unable to send tunnel msg\n"); + 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); - dbg("%c->%c:", l_ptr->b_ptr->net_plane, tunnel->b_ptr->net_plane); - msg_dbg(buf_msg(buf), ">SEND>"); - tipc_link_send_buf(tunnel, buf); + __tipc_link_xmit(tunnel, buf); } - -/* - * changeover(): Send whole message queue via the remaining link - * Owner node is locked. +/* 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_changeover(struct link *l_ptr) +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 link *tunnel = l_ptr->owner->active_links[0]; + struct tipc_link *tunnel = l_ptr->owner->active_links[0]; struct tipc_msg tunnel_hdr; int split_bundles; if (!tunnel) return; - if (!l_ptr->owner->permit_changeover) { - warn("Link changeover error, " - "peer did not permit changeover\n"); - return; - } - - msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL, - ORIGINAL_MSG, TIPC_OK, INT_H_SIZE, l_ptr->addr); + 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); - dbg("Link changeover requires %u tunnel messages\n", msgcount); if (!l_ptr->first_out) { struct sk_buff *buf; - buf = buf_acquire(INT_H_SIZE); + 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); - dbg("%c->%c:", l_ptr->b_ptr->net_plane, - tunnel->b_ptr->net_plane); - msg_dbg(&tunnel_hdr, "EMPTY>SEND>"); - tipc_link_send_buf(tunnel, buf); + __tipc_link_xmit(tunnel, buf); } else { - warn("Link changeover error, " - "unable to send changeover msg\n"); + pr_warn("%sunable to send changeover msg\n", + link_co_err); } return; } @@ -2384,31 +2023,41 @@ void tipc_link_changeover(struct link *l_ptr) if ((msg_user(msg) == MSG_BUNDLER) && split_bundles) { struct tipc_msg *m = msg_get_wrapped(msg); - unchar* pos = (unchar*)m; + unchar *pos = (unchar *)m; msgcount = msg_msgcnt(msg); while (msgcount--) { - msg_set_seqno(m,msg_seqno(msg)); - tipc_link_tunnel(l_ptr, &tunnel_hdr, m, - msg_link_selector(m)); + 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(l_ptr, &tunnel_hdr, msg, - msg_link_selector(msg)); + tipc_link_tunnel_xmit(l_ptr, &tunnel_hdr, msg, + msg_link_selector(msg)); } crs = crs->next; } } -void tipc_link_send_duplicate(struct link *l_ptr, struct link *tunnel) +/* 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; - msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL, - DUPLICATE_MSG, TIPC_OK, INT_H_SIZE, l_ptr->addr); + 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; @@ -2422,27 +2071,22 @@ void tipc_link_send_duplicate(struct link *l_ptr, struct link *tunnel) 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 = buf_acquire(length + INT_H_SIZE); + outbuf = tipc_buf_acquire(length + INT_H_SIZE); if (outbuf == NULL) { - warn("Link changeover error, " - "unable to send duplicate msg\n"); + 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); - dbg("%c->%c:", l_ptr->b_ptr->net_plane, - tunnel->b_ptr->net_plane); - msg_dbg(buf_msg(outbuf), ">SEND>"); - tipc_link_send_buf(tunnel, outbuf); + __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 @@ -2451,175 +2095,175 @@ void tipc_link_send_duplicate(struct link *l_ptr, struct link *tunnel) * 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 = buf_acquire(size); + eb = tipc_buf_acquire(size); if (eb) skb_copy_to_linear_data(eb, msg, size); return eb; } -/* - * link_recv_changeover_msg(): Receive tunneled packet sent - * via other link. Node is locked. Return extracted buffer. + + +/* 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); +} -static int link_recv_changeover_msg(struct link **l_ptr, - struct sk_buff **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 sk_buff *tunnel_buf = *buf; - struct link *dest_link; + struct tipc_msg *t_msg = buf_msg(t_buf); + struct sk_buff *buf = NULL; struct tipc_msg *msg; - struct tipc_msg *tunnel_msg = buf_msg(tunnel_buf); - u32 msg_typ = msg_type(tunnel_msg); - u32 msg_count = msg_msgcnt(tunnel_msg); - dest_link = (*l_ptr)->owner->links[msg_bearer_id(tunnel_msg)]; - if (!dest_link) { - msg_dbg(tunnel_msg, "NOLINK/<REC<"); - goto exit; - } - if (dest_link == *l_ptr) { - err("Unexpected changeover message on link <%s>\n", - (*l_ptr)->name); - goto exit; - } - dbg("%c<-%c:", dest_link->b_ptr->net_plane, - (*l_ptr)->b_ptr->net_plane); - *l_ptr = dest_link; - msg = msg_get_wrapped(tunnel_msg); - - if (msg_typ == DUPLICATE_MSG) { - if (less(msg_seqno(msg), mod(dest_link->next_in_no))) { - msg_dbg(tunnel_msg, "DROP/<REC<"); + 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; } - *buf = buf_extract(tunnel_buf,INT_H_SIZE); - if (*buf == NULL) { - warn("Link changeover error, duplicate msg dropped\n"); + msg = buf_msg(buf); + + if (less(msg_seqno(msg), l_ptr->reset_checkpoint)) { + kfree_skb(buf); + buf = NULL; goto exit; } - msg_dbg(tunnel_msg, "TNL<REC<"); - buf_discard(tunnel_buf); - return 1; + if (msg_user(msg) == MSG_FRAGMENTER) { + l_ptr->stats.recv_fragments++; + tipc_buf_append(&l_ptr->reasm_buf, &buf); + } } - - /* First original message ?: */ - - if (tipc_link_is_up(dest_link)) { - msg_dbg(tunnel_msg, "UP/FIRST/<REC<"); - info("Resetting link <%s>, changeover initiated by peer\n", - dest_link->name); - tipc_link_reset(dest_link); - dest_link->exp_msg_count = msg_count; - dbg("Expecting %u tunnelled messages\n", msg_count); - if (!msg_count) - goto exit; - } else if (dest_link->exp_msg_count == START_CHANGEOVER) { - msg_dbg(tunnel_msg, "BLK/FIRST/<REC<"); - dest_link->exp_msg_count = msg_count; - dbg("Expecting %u tunnelled messages\n", msg_count); - if (!msg_count) - goto exit; +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); - /* Receive original message */ + *buf = NULL; - if (dest_link->exp_msg_count == 0) { - warn("Link switchover error, " - "got too many tunnelled messages\n"); - msg_dbg(tunnel_msg, "OVERDUE/DROP/<REC<"); - dbg_print_link(dest_link, "LINK:"); + if (bearer_id >= MAX_BEARERS) goto exit; - } - dest_link->exp_msg_count--; - if (less(msg_seqno(msg), dest_link->reset_checkpoint)) { - msg_dbg(tunnel_msg, "DROP/DUPL/<REC<"); + + l_ptr = n_ptr->links[bearer_id]; + if (!l_ptr) goto exit; - } else { - *buf = buf_extract(tunnel_buf, INT_H_SIZE); - if (*buf != NULL) { - msg_dbg(tunnel_msg, "TNL<REC<"); - buf_discard(tunnel_buf); - return 1; - } else { - warn("Link changeover error, original msg dropped\n"); - } - } + + 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: - *buf = NULL; - buf_discard(tunnel_buf); - return 0; + kfree_skb(t_buf); + return *buf != NULL; } /* * Bundler functionality: */ -void tipc_link_recv_bundle(struct sk_buff *buf) +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; - msg_dbg(buf_msg(buf), "<BNDL<: "); while (msgcount--) { obuf = buf_extract(buf, pos); if (obuf == NULL) { - warn("Link unable to unbundle message(s)\n"); + pr_warn("Link unable to unbundle message(s)\n"); break; } pos += align(msg_size(buf_msg(obuf))); - msg_dbg(buf_msg(obuf), " /"); tipc_net_route_msg(obuf); } - buf_discard(buf); + kfree_skb(buf); } /* * Fragmentation/defragmentation: */ - /* - * tipc_link_send_long_buf: Entry for buffers needing fragmentation. + * tipc_link_frag_xmit: Entry for buffers needing fragmentation. * The buffer is complete, inclusive total message length. * Returns user data length. */ -int tipc_link_send_long_buf(struct link *l_ptr, struct sk_buff *buf) +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 = link_max_pkt(l_ptr); + u32 pack_sz = l_ptr->max_pkt; u32 fragm_sz = pack_sz - INT_H_SIZE; - u32 fragm_no = 1; - u32 destaddr = msg_destnode(inmsg); + u32 fragm_no = 0; + u32 destaddr; if (msg_short(inmsg)) destaddr = l_ptr->addr; - - if (msg_routed(inmsg)) - msg_set_prevnode(inmsg, tipc_own_addr); + else + destaddr = msg_destnode(inmsg); /* Prepare reusable fragment header: */ - - msg_init(&fragm_hdr, MSG_FRAGMENTER, FIRST_FRAGMENT, - TIPC_OK, INT_H_SIZE, destaddr); - msg_set_link_selector(&fragm_hdr, msg_link_selector(inmsg)); - msg_set_long_msgno(&fragm_hdr, mod(l_ptr->long_msg_seq_no++)); - msg_set_fragm_no(&fragm_hdr, fragm_no); - l_ptr->stats.sent_fragmented++; + tipc_msg_init(&fragm_hdr, MSG_FRAGMENTER, FIRST_FRAGMENT, + INT_H_SIZE, destaddr); /* Chop up message: */ - while (rest > 0) { struct sk_buff *fragm; @@ -2627,261 +2271,210 @@ int tipc_link_send_long_buf(struct link *l_ptr, struct sk_buff *buf) fragm_sz = rest; msg_set_type(&fragm_hdr, LAST_FRAGMENT); } - fragm = buf_acquire(fragm_sz + INT_H_SIZE); + fragm = tipc_buf_acquire(fragm_sz + INT_H_SIZE); if (fragm == NULL) { - warn("Link unable to fragment message\n"); - dsz = -ENOMEM; - goto exit; + 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); - /* Send queued messages first, if any: */ + buf_chain_tail->next = fragm; + buf_chain_tail = fragm; - l_ptr->stats.sent_fragments++; - tipc_link_send_buf(l_ptr, fragm); - if (!tipc_link_is_up(l_ptr)) - return dsz; - msg_set_fragm_no(&fragm_hdr, ++fragm_no); rest -= fragm_sz; crs += fragm_sz; msg_set_type(&fragm_hdr, FRAGMENT); } -exit: - buf_discard(buf); - return dsz; -} - -/* - * A pending message being re-assembled must store certain values - * to handle subsequent fragments correctly. The following functions - * help storing these values in unused, available fields in the - * pending message. This makes dynamic memory allocation unecessary. - */ - -static void set_long_msg_seqno(struct sk_buff *buf, u32 seqno) -{ - msg_set_seqno(buf_msg(buf), seqno); -} - -static u32 get_fragm_size(struct sk_buff *buf) -{ - return msg_ack(buf_msg(buf)); -} - -static void set_fragm_size(struct sk_buff *buf, u32 sz) -{ - msg_set_ack(buf_msg(buf), sz); -} - -static u32 get_expected_frags(struct sk_buff *buf) -{ - return msg_bcast_ack(buf_msg(buf)); -} - -static void set_expected_frags(struct sk_buff *buf, u32 exp) -{ - msg_set_bcast_ack(buf_msg(buf), exp); -} - -static u32 get_timer_cnt(struct sk_buff *buf) -{ - return msg_reroute_cnt(buf_msg(buf)); -} - -static void incr_timer_cnt(struct sk_buff *buf) -{ - msg_incr_reroute_cnt(buf_msg(buf)); -} - -/* - * tipc_link_recv_fragment(): Called with node lock on. Returns - * the reassembled buffer if message is complete. - */ -int tipc_link_recv_fragment(struct sk_buff **pending, struct sk_buff **fb, - struct tipc_msg **m) -{ - struct sk_buff *prev = NULL; - struct sk_buff *fbuf = *fb; - struct tipc_msg *fragm = buf_msg(fbuf); - struct sk_buff *pbuf = *pending; - u32 long_msg_seq_no = msg_long_msgno(fragm); + kfree_skb(buf); - *fb = NULL; - msg_dbg(fragm,"FRG<REC<"); - - /* Is there an incomplete message waiting for this fragment? */ - - while (pbuf && ((msg_seqno(buf_msg(pbuf)) != long_msg_seq_no) - || (msg_orignode(fragm) != msg_orignode(buf_msg(pbuf))))) { - prev = pbuf; - pbuf = pbuf->next; - } + /* 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); - if (!pbuf && (msg_type(fragm) == FIRST_FRAGMENT)) { - struct tipc_msg *imsg = (struct tipc_msg *)msg_data(fragm); - u32 msg_sz = msg_size(imsg); - u32 fragm_sz = msg_data_sz(fragm); - u32 exp_fragm_cnt = msg_sz/fragm_sz + !!(msg_sz % fragm_sz); - u32 max = TIPC_MAX_USER_MSG_SIZE + LONG_H_SIZE; - if (msg_type(imsg) == TIPC_MCAST_MSG) - max = TIPC_MAX_USER_MSG_SIZE + MCAST_H_SIZE; - if (msg_size(imsg) > max) { - msg_dbg(fragm,"<REC<Oversized: "); - buf_discard(fbuf); - return 0; - } - pbuf = buf_acquire(msg_size(imsg)); - if (pbuf != NULL) { - pbuf->next = *pending; - *pending = pbuf; - skb_copy_to_linear_data(pbuf, imsg, - msg_data_sz(fragm)); - /* Prepare buffer for subsequent fragments. */ - - set_long_msg_seqno(pbuf, long_msg_seq_no); - set_fragm_size(pbuf,fragm_sz); - set_expected_frags(pbuf,exp_fragm_cnt - 1); - } else { - warn("Link unable to reassemble fragmented message\n"); - } - buf_discard(fbuf); - return 0; - } else if (pbuf && (msg_type(fragm) != FIRST_FRAGMENT)) { - u32 dsz = msg_data_sz(fragm); - u32 fsz = get_fragm_size(pbuf); - u32 crs = ((msg_fragm_no(fragm) - 1) * fsz); - u32 exp_frags = get_expected_frags(pbuf) - 1; - skb_copy_to_linear_data_offset(pbuf, crs, - msg_data(fragm), dsz); - buf_discard(fbuf); - - /* Is message complete? */ - - if (exp_frags == 0) { - if (prev) - prev->next = pbuf->next; - else - *pending = pbuf->next; - msg_reset_reroute_cnt(buf_msg(pbuf)); - *fb = pbuf; - *m = buf_msg(pbuf); - return 1; - } - set_expected_frags(pbuf,exp_frags); - return 0; - } - dbg(" Discarding orphan fragment %x\n",fbuf); - msg_dbg(fragm,"ORPHAN:"); - dbg("Pending long buffers:\n"); - dbg_print_buf_chain(*pending); - buf_discard(fbuf); - return 0; + return dsz; } -/** - * link_check_defragm_bufs - flush stale incoming message fragments - * @l_ptr: pointer to link - */ - -static void link_check_defragm_bufs(struct link *l_ptr) +static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tolerance) { - struct sk_buff *prev = NULL; - struct sk_buff *next = NULL; - struct sk_buff *buf = l_ptr->defragm_buf; - - if (!buf) - return; - if (!link_working_working(l_ptr)) + if ((tolerance < TIPC_MIN_LINK_TOL) || (tolerance > TIPC_MAX_LINK_TOL)) return; - while (buf) { - u32 cnt = get_timer_cnt(buf); - - next = buf->next; - if (cnt < 4) { - incr_timer_cnt(buf); - prev = buf; - } else { - dbg(" Discarding incomplete long buffer\n"); - msg_dbg(buf_msg(buf), "LONG:"); - dbg_print_link(l_ptr, "curr:"); - dbg("Pending long buffers:\n"); - dbg_print_buf_chain(l_ptr->defragm_buf); - if (prev) - prev->next = buf->next; - else - l_ptr->defragm_buf = buf->next; - buf_discard(buf); - } - buf = next; - } -} - - -static void link_set_supervision_props(struct link *l_ptr, u32 tolerance) -{ 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 link *l_ptr, u32 window) +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[DATA_LOW] = window; - l_ptr->queue_limit[DATA_MEDIUM] = (window / 3) * 4; - l_ptr->queue_limit[DATA_HIGH] = (window / 3) * 5; - l_ptr->queue_limit[DATA_CRITICAL] = (window / 3) * 6; + 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[DATA_LOW + 4] = 300; - l_ptr->queue_limit[DATA_MEDIUM + 4] = 600; - l_ptr->queue_limit[DATA_HIGH + 4] = 900; - l_ptr->queue_limit[DATA_CRITICAL + 4] = 1200; + 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[ROUTE_DISTRIBUTOR] = 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_find_link - locate link by name - * @name - ptr to link name string - * @node - ptr to area to be filled with ptr to associated node + * link_value_is_valid -- validate proposed link tolerance/priority/window * - * Caller must hold 'tipc_net_lock' to ensure node and bearer are not deleted; - * this also prevents link deletion. + * @cmd: value type (TIPC_CMD_SET_LINK_*) + * @new_value: the new value * - * Returns pointer to link (or 0 if invalid link name). + * Returns 1 if value is within range, 0 if not. */ - -static struct link *link_find_link(const char *name, struct node **node) +static int link_value_is_valid(u16 cmd, u32 new_value) { - struct link_name link_name_parts; - struct bearer *b_ptr; - struct link *l_ptr; + 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; +} - if (!link_name_validate(name, &link_name_parts)) - return NULL; +/** + * 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; - b_ptr = tipc_bearer_find_interface(link_name_parts.if_local); - if (!b_ptr) - return NULL; + node = tipc_link_find_owner(name, &bearer_id); + if (node) { + tipc_node_lock(node); + l_ptr = node->links[bearer_id]; - *node = tipc_node_find(link_name_parts.addr_peer); - if (!*node) - return NULL; + 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; + } - l_ptr = (*node)->links[b_ptr->identity]; - if (!l_ptr || strcmp(l_ptr->name, name)) - return NULL; + 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; + } - return l_ptr; + 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, @@ -2889,8 +2482,6 @@ struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space { struct tipc_link_config *args; u32 new_value; - struct link *l_ptr; - struct node *node; int res; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_CONFIG)) @@ -2899,6 +2490,10 @@ struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space 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)) @@ -2907,45 +2502,7 @@ struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space " (cannot change setting on broadcast link)"); } - read_lock_bh(&tipc_net_lock); - l_ptr = link_find_link(args->name, &node); - if (!l_ptr) { - read_unlock_bh(&tipc_net_lock); - return tipc_cfg_reply_error_string("link not found"); - } - - tipc_node_lock(node); - res = -EINVAL; - switch (cmd) { - case TIPC_CMD_SET_LINK_TOL: - if ((new_value >= TIPC_MIN_LINK_TOL) && - (new_value <= TIPC_MAX_LINK_TOL)) { - link_set_supervision_props(l_ptr, new_value); - tipc_link_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, new_value, 0, 0); - res = TIPC_OK; - } - break; - case TIPC_CMD_SET_LINK_PRI: - if ((new_value >= TIPC_MIN_LINK_PRI) && - (new_value <= TIPC_MAX_LINK_PRI)) { - l_ptr->priority = new_value; - tipc_link_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, 0, new_value, 0); - res = TIPC_OK; - } - break; - case TIPC_CMD_SET_LINK_WINDOW: - if ((new_value >= TIPC_MIN_LINK_WIN) && - (new_value <= TIPC_MAX_LINK_WIN)) { - tipc_link_set_queue_limits(l_ptr, new_value); - res = TIPC_OK; - } - break; - } - tipc_node_unlock(node); - - read_unlock_bh(&tipc_net_lock); + res = link_cmd_set_value(args->name, new_value, cmd); if (res) return tipc_cfg_reply_error_string("cannot change link setting"); @@ -2956,8 +2513,7 @@ struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space * link_reset_statistics - reset link statistics * @l_ptr: pointer to link */ - -static void link_reset_statistics(struct link *l_ptr) +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; @@ -2967,8 +2523,9 @@ static void link_reset_statistics(struct link *l_ptr) struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space) { char *link_name; - struct link *l_ptr; - struct node *node; + 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); @@ -2979,25 +2536,24 @@ struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_ 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"); - read_lock_bh(&tipc_net_lock); - l_ptr = link_find_link(link_name, &node); + tipc_node_lock(node); + l_ptr = node->links[bearer_id]; if (!l_ptr) { - read_unlock_bh(&tipc_net_lock); + tipc_node_unlock(node); return tipc_cfg_reply_error_string("link not found"); } - - tipc_node_lock(node); link_reset_statistics(l_ptr); tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); 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; @@ -3011,159 +2567,127 @@ static u32 percent(u32 count, u32 total) * * 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 print_buf pb; - struct link *l_ptr; - struct node *node; + 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); - tipc_printbuf_init(&pb, buf, buf_size); + node = tipc_link_find_owner(name, &bearer_id); + if (!node) + return 0; - read_lock_bh(&tipc_net_lock); - l_ptr = link_find_link(name, &node); - if (!l_ptr) { - read_unlock_bh(&tipc_net_lock); + tipc_node_lock(node); + + l = node->links[bearer_id]; + if (!l) { + tipc_node_unlock(node); return 0; } - tipc_node_lock(node); - if (tipc_link_is_active(l_ptr)) + s = &l->stats; + + if (tipc_link_is_active(l)) status = "ACTIVE"; - else if (tipc_link_is_up(l_ptr)) + else if (tipc_link_is_up(l)) status = "STANDBY"; else status = "DEFUNCT"; - tipc_printf(&pb, "Link <%s>\n" - " %s MTU:%u Priority:%u Tolerance:%u ms" - " Window:%u packets\n", - l_ptr->name, status, link_max_pkt(l_ptr), - l_ptr->priority, l_ptr->tolerance, l_ptr->queue_limit[0]); - tipc_printf(&pb, " RX packets:%u fragments:%u/%u bundles:%u/%u\n", - l_ptr->next_in_no - l_ptr->stats.recv_info, - l_ptr->stats.recv_fragments, - l_ptr->stats.recv_fragmented, - l_ptr->stats.recv_bundles, - l_ptr->stats.recv_bundled); - tipc_printf(&pb, " TX packets:%u fragments:%u/%u bundles:%u/%u\n", - l_ptr->next_out_no - l_ptr->stats.sent_info, - l_ptr->stats.sent_fragments, - l_ptr->stats.sent_fragmented, - l_ptr->stats.sent_bundles, - l_ptr->stats.sent_bundled); - profile_total = l_ptr->stats.msg_length_counts; + + 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; - tipc_printf(&pb, " TX profile sample:%u packets average:%u octets\n" - " 0-64:%u%% -256:%u%% -1024:%u%% -4096:%u%% " - "-16354:%u%% -32768:%u%% -66000:%u%%\n", - l_ptr->stats.msg_length_counts, - l_ptr->stats.msg_lengths_total / profile_total, - percent(l_ptr->stats.msg_length_profile[0], profile_total), - percent(l_ptr->stats.msg_length_profile[1], profile_total), - percent(l_ptr->stats.msg_length_profile[2], profile_total), - percent(l_ptr->stats.msg_length_profile[3], profile_total), - percent(l_ptr->stats.msg_length_profile[4], profile_total), - percent(l_ptr->stats.msg_length_profile[5], profile_total), - percent(l_ptr->stats.msg_length_profile[6], profile_total)); - tipc_printf(&pb, " RX states:%u probes:%u naks:%u defs:%u dups:%u\n", - l_ptr->stats.recv_states, - l_ptr->stats.recv_probes, - l_ptr->stats.recv_nacks, - l_ptr->stats.deferred_recv, - l_ptr->stats.duplicates); - tipc_printf(&pb, " TX states:%u probes:%u naks:%u acks:%u dups:%u\n", - l_ptr->stats.sent_states, - l_ptr->stats.sent_probes, - l_ptr->stats.sent_nacks, - l_ptr->stats.sent_acks, - l_ptr->stats.retransmitted); - tipc_printf(&pb, " Congestion bearer:%u link:%u Send queue max:%u avg:%u\n", - l_ptr->stats.bearer_congs, - l_ptr->stats.link_congs, - l_ptr->stats.max_queue_sz, - l_ptr->stats.queue_sz_counts - ? (l_ptr->stats.accu_queue_sz / l_ptr->stats.queue_sz_counts) - : 0); + + 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); - read_unlock_bh(&tipc_net_lock); - return tipc_printbuf_validate(&pb); + return ret; } -#define MAX_LINK_STATS_INFO 2000 - 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(MAX_LINK_STATS_INFO)); + 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), - (char *)TLV_DATA(rep_tlv), MAX_LINK_STATS_INFO); + pb, pb_len); if (!str_len) { - buf_discard(buf); + 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; } -#if 0 -int link_control(const char *name, u32 op, u32 val) -{ - int res = -EINVAL; - struct link *l_ptr; - u32 bearer_id; - struct node * node; - u32 a; - - a = link_name2addr(name, &bearer_id); - read_lock_bh(&tipc_net_lock); - node = tipc_node_find(a); - if (node) { - tipc_node_lock(node); - l_ptr = node->links[bearer_id]; - if (l_ptr) { - if (op == TIPC_REMOVE_LINK) { - struct bearer *b_ptr = l_ptr->b_ptr; - spin_lock_bh(&b_ptr->publ.lock); - tipc_link_delete(l_ptr); - spin_unlock_bh(&b_ptr->publ.lock); - } - if (op == TIPC_CMD_BLOCK_LINK) { - tipc_link_reset(l_ptr); - l_ptr->blocked = 1; - } - if (op == TIPC_CMD_UNBLOCK_LINK) { - l_ptr->blocked = 0; - } - res = TIPC_OK; - } - tipc_node_unlock(node); - } - read_unlock_bh(&tipc_net_lock); - return res; -} -#endif - /** * tipc_link_get_max_pkt - get maximum packet size to use when sending to destination * @dest: network address of destination node @@ -3171,113 +2695,44 @@ int link_control(const char *name, u32 op, u32 val) * * If no active link can be found, uses default maximum packet size. */ - u32 tipc_link_get_max_pkt(u32 dest, u32 selector) { - struct node *n_ptr; - struct link *l_ptr; + struct tipc_node *n_ptr; + struct tipc_link *l_ptr; u32 res = MAX_PKT_DEFAULT; if (dest == tipc_own_addr) return MAX_MSG_SIZE; - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_select(dest, selector); + 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 = link_max_pkt(l_ptr); + res = l_ptr->max_pkt; tipc_node_unlock(n_ptr); } - read_unlock_bh(&tipc_net_lock); return res; } -#if 0 -static void link_dump_rec_queue(struct link *l_ptr) +static void link_print(struct tipc_link *l_ptr, const char *str) { - struct sk_buff *crs; + struct tipc_bearer *b_ptr; - if (!l_ptr->oldest_deferred_in) { - info("Reception queue empty\n"); - return; - } - info("Contents of Reception queue:\n"); - crs = l_ptr->oldest_deferred_in; - while (crs) { - if (crs->data == (void *)0x0000a3a3) { - info("buffer %x invalid\n", crs); - return; - } - msg_dbg(buf_msg(crs), "In rec queue: \n"); - crs = crs->next; - } -} -#endif + 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(); -static void link_dump_send_queue(struct link *l_ptr) -{ - if (l_ptr->next_out) { - info("\nContents of unsent queue:\n"); - dbg_print_buf_chain(l_ptr->next_out); - } - info("\nContents of send queue:\n"); - if (l_ptr->first_out) { - dbg_print_buf_chain(l_ptr->first_out); - } - info("Empty send queue\n"); -} - -static void link_print(struct link *l_ptr, struct print_buf *buf, - const char *str) -{ - tipc_printf(buf, str); - if (link_reset_reset(l_ptr) || link_reset_unknown(l_ptr)) - return; - tipc_printf(buf, "Link %x<%s>:", - l_ptr->addr, l_ptr->b_ptr->publ.name); - tipc_printf(buf, ": NXO(%u):", mod(l_ptr->next_out_no)); - tipc_printf(buf, "NXI(%u):", mod(l_ptr->next_in_no)); - tipc_printf(buf, "SQUE"); - if (l_ptr->first_out) { - tipc_printf(buf, "[%u..", msg_seqno(buf_msg(l_ptr->first_out))); - if (l_ptr->next_out) - tipc_printf(buf, "%u..", - msg_seqno(buf_msg(l_ptr->next_out))); - tipc_printf(buf, "%u]", - msg_seqno(buf_msg - (l_ptr->last_out)), l_ptr->out_queue_size); - if ((mod(msg_seqno(buf_msg(l_ptr->last_out)) - - msg_seqno(buf_msg(l_ptr->first_out))) - != (l_ptr->out_queue_size - 1)) - || (l_ptr->last_out->next != NULL)) { - tipc_printf(buf, "\nSend queue inconsistency\n"); - tipc_printf(buf, "first_out= %x ", l_ptr->first_out); - tipc_printf(buf, "next_out= %x ", l_ptr->next_out); - tipc_printf(buf, "last_out= %x ", l_ptr->last_out); - link_dump_send_queue(l_ptr); - } - } else - tipc_printf(buf, "[]"); - tipc_printf(buf, "SQSIZ(%u)", l_ptr->out_queue_size); - if (l_ptr->oldest_deferred_in) { - u32 o = msg_seqno(buf_msg(l_ptr->oldest_deferred_in)); - u32 n = msg_seqno(buf_msg(l_ptr->newest_deferred_in)); - tipc_printf(buf, ":RQUE[%u..%u]", o, n); - if (l_ptr->deferred_inqueue_sz != mod((n + 1) - o)) { - tipc_printf(buf, ":RQSIZ(%u)", - l_ptr->deferred_inqueue_sz); - } - } if (link_working_unknown(l_ptr)) - tipc_printf(buf, ":WU"); - if (link_reset_reset(l_ptr)) - tipc_printf(buf, ":RR"); - if (link_reset_unknown(l_ptr)) - tipc_printf(buf, ":RU"); - if (link_working_working(l_ptr)) - tipc_printf(buf, ":WW"); - tipc_printf(buf, "\n"); + 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 index 52f3e7c1871..200d518b218 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -1,8 +1,8 @@ /* * net/tipc/link.h: Include file for TIPC link code * - * Copyright (c) 1995-2006, Ericsson AB - * Copyright (c) 2004-2005, Wind River Systems + * 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 @@ -37,52 +37,81 @@ #ifndef _TIPC_LINK_H #define _TIPC_LINK_H -#include "dbg.h" #include "msg.h" -#include "bearer.h" #include "node.h" -#define PUSH_FAILED 1 -#define PUSH_FINISHED 2 - -/* - * Link states +/* 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 -/* - * Starting value for maximum packet size negotiation on unicast links - * (unless bearer MTU is less) +/* 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 link - TIPC link data structure + * 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 - * @link_list: adjacent links in bearer's list of links - * @started: indicates if link has been started + * @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 - * @b_ptr: pointer to bearer used by link + * @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 - * @blocked: indicates if link has been administratively blocked * @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 @@ -106,30 +135,26 @@ * @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 - * @defragm_buf: list of partially reassembled inbound message fragments + * @reasm_buf: head of partially reassembled inbound message fragments * @stats: collects statistics regarding link activity - * @print_buf: print buffer used to log link activity */ - -struct link { +struct tipc_link { u32 addr; char name[TIPC_MAX_LINK_NAME]; struct tipc_media_addr media_addr; struct timer_list timer; - struct node *owner; - struct list_head link_list; + struct tipc_node *owner; /* Management and link supervision data */ - int started; + unsigned int flags; u32 checkpoint; u32 peer_session; u32 peer_bearer_id; - struct bearer *b_ptr; + u32 bearer_id; u32 tolerance; u32 continuity_interval; u32 abort_limit; int state; - int blocked; u32 fsm_msg_cnt; struct { unchar hdr[INT_H_SIZE]; @@ -137,6 +162,7 @@ struct link { } proto_msg; struct tipc_msg *pmsg; u32 priority; + char net_plane; u32 queue_limit[15]; /* queue_limit[0]==window limit */ /* Changeover */ @@ -170,98 +196,62 @@ struct link { struct sk_buff *next_out; struct list_head waiting_ports; - /* Fragmentation/defragmentation */ + /* Fragmentation/reassembly */ u32 long_msg_seq_no; - struct sk_buff *defragm_buf; + struct sk_buff *reasm_buf; /* Statistics */ - struct { - 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 bearer_congs; - u32 deferred_recv; - u32 duplicates; - - /* for statistical profiling of send queue size */ - - u32 max_queue_sz; - u32 accu_queue_sz; - u32 queue_sz_counts; - - /* for statistical profiling of message lengths */ - - u32 msg_length_counts; - u32 msg_lengths_total; - u32 msg_length_profile[7]; -#if 0 - u32 sent_tunneled; - u32 recv_tunneled; -#endif - } stats; - - struct print_buf print_buf; + struct tipc_stats stats; }; -struct port; +struct tipc_port; -struct link *tipc_link_create(struct bearer *b_ptr, const u32 peer, +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(struct link *l_ptr); -void tipc_link_changeover(struct link *l_ptr); -void tipc_link_send_duplicate(struct link *l_ptr, struct link *dest); -void tipc_link_reset_fragments(struct link *l_ptr); -int tipc_link_is_up(struct link *l_ptr); -int tipc_link_is_active(struct link *l_ptr); -void tipc_link_start(struct link *l_ptr); -u32 tipc_link_push_packet(struct link *l_ptr); -void tipc_link_stop(struct 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(struct link *l_ptr); -int tipc_link_send(struct sk_buff *buf, u32 dest, u32 selector); -int tipc_link_send_buf(struct link *l_ptr, struct sk_buff *buf); -u32 tipc_link_get_max_pkt(u32 dest,u32 selector); -int tipc_link_send_sections_fast(struct port* sender, - struct iovec const *msg_sect, - const u32 num_sect, - u32 destnode); -int tipc_link_send_long_buf(struct link *l_ptr, struct sk_buff *buf); -void tipc_link_tunnel(struct link *l_ptr, struct tipc_msg *tnl_hdr, - struct tipc_msg *msg, u32 selector); -void tipc_link_recv_bundle(struct sk_buff *buf); -int tipc_link_recv_fragment(struct sk_buff **pending, - struct sk_buff **fb, - struct tipc_msg **msg); -void tipc_link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int prob, u32 gap, - u32 tolerance, u32 priority, u32 acked_mtu); -void tipc_link_push_queue(struct link *l_ptr); +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 link *l_ptr, int all); -void tipc_link_set_queue_limits(struct link *l_ptr, u32 window); -void tipc_link_retransmit(struct link *l_ptr, struct sk_buff *start, u32 retransmits); + 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) { @@ -279,12 +269,12 @@ static inline int between(u32 lower, u32 upper, u32 n) static inline int less_eq(u32 left, u32 right) { - return (mod(right - left) < 32768u); + return mod(right - left) < 32768u; } static inline int less(u32 left, u32 right) { - return (less_eq(left, right) && (mod(right) != mod(left))); + return less_eq(left, right) && (mod(right) != mod(left)); } static inline u32 lesser(u32 left, u32 right) @@ -292,4 +282,33 @@ 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/user_reg.h b/net/tipc/log.c index 81dc12e2882..abef644f27d 100644 --- a/net/tipc/user_reg.h +++ b/net/tipc/log.c @@ -1,8 +1,8 @@ /* - * net/tipc/user_reg.h: Include file for TIPC user registry code + * net/tipc/log.c: TIPC print buffer routines for debugging * - * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * 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 @@ -34,15 +34,22 @@ * POSSIBILITY OF SUCH DAMAGE. */ -#ifndef _TIPC_USER_REG_H -#define _TIPC_USER_REG_H +#include "core.h" +#include "config.h" -#include "port.h" - -int tipc_reg_start(void); -void tipc_reg_stop(void); - -int tipc_reg_add_port(struct user_port *up_ptr); -int tipc_reg_remove_port(struct user_port *up_ptr); +/** + * 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; -#endif + 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 index 782485468fb..0a37a472c29 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -1,8 +1,8 @@ /* * net/tipc/msg.c: TIPC message header routines * - * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * 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 @@ -35,289 +35,125 @@ */ #include "core.h" -#include "addr.h" -#include "dbg.h" #include "msg.h" -#include "bearer.h" - -void tipc_msg_print(struct print_buf *buf, struct tipc_msg *msg, const char *str) +u32 tipc_msg_tot_importance(struct tipc_msg *m) { - u32 usr = msg_user(msg); - tipc_printf(buf, str); - - switch (usr) { - case MSG_BUNDLER: - tipc_printf(buf, "BNDL::"); - tipc_printf(buf, "MSGS(%u):", msg_msgcnt(msg)); - break; - case BCAST_PROTOCOL: - tipc_printf(buf, "BCASTP::"); - break; - case MSG_FRAGMENTER: - tipc_printf(buf, "FRAGM::"); - switch (msg_type(msg)) { - case FIRST_FRAGMENT: - tipc_printf(buf, "FIRST:"); - break; - case FRAGMENT: - tipc_printf(buf, "BODY:"); - break; - case LAST_FRAGMENT: - tipc_printf(buf, "LAST:"); - break; - default: - tipc_printf(buf, "UNKNOWN:%x",msg_type(msg)); - - } - tipc_printf(buf, "NO(%u/%u):",msg_long_msgno(msg), - msg_fragm_no(msg)); - break; - case DATA_LOW: - case DATA_MEDIUM: - case DATA_HIGH: - case DATA_CRITICAL: - tipc_printf(buf, "DAT%u:", msg_user(msg)); - if (msg_short(msg)) { - tipc_printf(buf, "CON:"); - break; - } - switch (msg_type(msg)) { - case TIPC_CONN_MSG: - tipc_printf(buf, "CON:"); - break; - case TIPC_MCAST_MSG: - tipc_printf(buf, "MCST:"); - break; - case TIPC_NAMED_MSG: - tipc_printf(buf, "NAM:"); - break; - case TIPC_DIRECT_MSG: - tipc_printf(buf, "DIR:"); - break; - default: - tipc_printf(buf, "UNKNOWN TYPE %u",msg_type(msg)); - } - if (msg_routed(msg) && !msg_non_seq(msg)) - tipc_printf(buf, "ROUT:"); - if (msg_reroute_cnt(msg)) - tipc_printf(buf, "REROUTED(%u):", - msg_reroute_cnt(msg)); - break; - case NAME_DISTRIBUTOR: - tipc_printf(buf, "NMD::"); - switch (msg_type(msg)) { - case PUBLICATION: - tipc_printf(buf, "PUBL(%u):", (msg_size(msg) - msg_hdr_sz(msg)) / 20); /* Items */ - break; - case WITHDRAWAL: - tipc_printf(buf, "WDRW:"); - break; - default: - tipc_printf(buf, "UNKNOWN:%x",msg_type(msg)); - } - if (msg_routed(msg)) - tipc_printf(buf, "ROUT:"); - if (msg_reroute_cnt(msg)) - tipc_printf(buf, "REROUTED(%u):", - msg_reroute_cnt(msg)); - break; - case CONN_MANAGER: - tipc_printf(buf, "CONN_MNG:"); - switch (msg_type(msg)) { - case CONN_PROBE: - tipc_printf(buf, "PROBE:"); - break; - case CONN_PROBE_REPLY: - tipc_printf(buf, "PROBE_REPLY:"); - break; - case CONN_ACK: - tipc_printf(buf, "CONN_ACK:"); - tipc_printf(buf, "ACK(%u):",msg_msgcnt(msg)); - break; - default: - tipc_printf(buf, "UNKNOWN TYPE:%x",msg_type(msg)); - } - if (msg_routed(msg)) - tipc_printf(buf, "ROUT:"); - if (msg_reroute_cnt(msg)) - tipc_printf(buf, "REROUTED(%u):",msg_reroute_cnt(msg)); - break; - case LINK_PROTOCOL: - tipc_printf(buf, "PROT:TIM(%u):",msg_timestamp(msg)); - switch (msg_type(msg)) { - case STATE_MSG: - tipc_printf(buf, "STATE:"); - tipc_printf(buf, "%s:",msg_probe(msg) ? "PRB" :""); - tipc_printf(buf, "NXS(%u):",msg_next_sent(msg)); - tipc_printf(buf, "GAP(%u):",msg_seq_gap(msg)); - tipc_printf(buf, "LSTBC(%u):",msg_last_bcast(msg)); - break; - case RESET_MSG: - tipc_printf(buf, "RESET:"); - if (msg_size(msg) != msg_hdr_sz(msg)) - tipc_printf(buf, "BEAR:%s:",msg_data(msg)); - break; - case ACTIVATE_MSG: - tipc_printf(buf, "ACTIVATE:"); - break; - default: - tipc_printf(buf, "UNKNOWN TYPE:%x",msg_type(msg)); - } - tipc_printf(buf, "PLANE(%c):",msg_net_plane(msg)); - tipc_printf(buf, "SESS(%u):",msg_session(msg)); - break; - case CHANGEOVER_PROTOCOL: - tipc_printf(buf, "TUNL:"); - switch (msg_type(msg)) { - case DUPLICATE_MSG: - tipc_printf(buf, "DUPL:"); - break; - case ORIGINAL_MSG: - tipc_printf(buf, "ORIG:"); - tipc_printf(buf, "EXP(%u)",msg_msgcnt(msg)); - break; - default: - tipc_printf(buf, "UNKNOWN TYPE:%x",msg_type(msg)); - } - break; - case ROUTE_DISTRIBUTOR: - tipc_printf(buf, "ROUTING_MNG:"); - switch (msg_type(msg)) { - case EXT_ROUTING_TABLE: - tipc_printf(buf, "EXT_TBL:"); - tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); - break; - case LOCAL_ROUTING_TABLE: - tipc_printf(buf, "LOCAL_TBL:"); - tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); - break; - case SLAVE_ROUTING_TABLE: - tipc_printf(buf, "DP_TBL:"); - tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); - break; - case ROUTE_ADDITION: - tipc_printf(buf, "ADD:"); - tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); - break; - case ROUTE_REMOVAL: - tipc_printf(buf, "REMOVE:"); - tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); - break; - default: - tipc_printf(buf, "UNKNOWN TYPE:%x",msg_type(msg)); - } - break; - case LINK_CONFIG: - tipc_printf(buf, "CFG:"); - switch (msg_type(msg)) { - case DSC_REQ_MSG: - tipc_printf(buf, "DSC_REQ:"); - break; - case DSC_RESP_MSG: - tipc_printf(buf, "DSC_RESP:"); - break; - default: - tipc_printf(buf, "UNKNOWN TYPE:%x:",msg_type(msg)); - break; - } - break; - default: - tipc_printf(buf, "UNKNOWN USER:"); + 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); +} - switch (usr) { - case CONN_MANAGER: - case NAME_DISTRIBUTOR: - case DATA_LOW: - case DATA_MEDIUM: - case DATA_HIGH: - case DATA_CRITICAL: - if (msg_short(msg)) - break; /* No error */ - switch (msg_errcode(msg)) { - case TIPC_OK: - break; - case TIPC_ERR_NO_NAME: - tipc_printf(buf, "NO_NAME:"); - break; - case TIPC_ERR_NO_PORT: - tipc_printf(buf, "NO_PORT:"); - break; - case TIPC_ERR_NO_NODE: - tipc_printf(buf, "NO_PROC:"); - break; - case TIPC_ERR_OVERLOAD: - tipc_printf(buf, "OVERLOAD:"); - break; - case TIPC_CONN_SHUTDOWN: - tipc_printf(buf, "SHUTDOWN:"); - break; - default: - tipc_printf(buf, "UNKNOWN ERROR(%x):", - msg_errcode(msg)); - } - default:{} - } - tipc_printf(buf, "HZ(%u):", msg_hdr_sz(msg)); - tipc_printf(buf, "SZ(%u):", msg_size(msg)); - tipc_printf(buf, "SQNO(%u):", msg_seqno(msg)); +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); +} - if (msg_non_seq(msg)) - tipc_printf(buf, "NOSEQ:"); - else { - tipc_printf(buf, "ACK(%u):", msg_ack(msg)); - } - tipc_printf(buf, "BACK(%u):", msg_bcast_ack(msg)); - tipc_printf(buf, "PRND(%x)", msg_prevnode(msg)); +/** + * 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; - if (msg_isdata(msg)) { - if (msg_named(msg)) { - tipc_printf(buf, "NTYP(%u):", msg_nametype(msg)); - tipc_printf(buf, "NINST(%u)", msg_nameinst(msg)); - } + dsz = len; + hsz = msg_hdr_sz(hdr); + sz = hsz + dsz; + msg_set_size(hdr, sz); + if (unlikely(sz > max_size)) { + *buf = NULL; + return dsz; } - if ((usr != LINK_PROTOCOL) && (usr != LINK_CONFIG) && - (usr != MSG_BUNDLER)) { - if (!msg_short(msg)) { - tipc_printf(buf, ":ORIG(%x:%u):", - msg_orignode(msg), msg_origport(msg)); - tipc_printf(buf, ":DEST(%x:%u):", - msg_destnode(msg), msg_destport(msg)); - } else { - tipc_printf(buf, ":OPRT(%u):", msg_origport(msg)); - tipc_printf(buf, ":DPRT(%u):", msg_destport(msg)); - } - if (msg_routed(msg) && !msg_non_seq(msg)) - tipc_printf(buf, ":TSEQN(%u)", msg_transp_seqno(msg)); - } - if (msg_user(msg) == NAME_DISTRIBUTOR) { - tipc_printf(buf, ":ONOD(%x):", msg_orignode(msg)); - tipc_printf(buf, ":DNOD(%x):", msg_destnode(msg)); - if (msg_routed(msg)) { - tipc_printf(buf, ":CSEQN(%u)", msg_transp_seqno(msg)); - } + *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; +} - if (msg_user(msg) == LINK_CONFIG) { - u32* raw = (u32*)msg; - struct tipc_media_addr* orig = (struct tipc_media_addr*)&raw[5]; - tipc_printf(buf, ":REQL(%u):", msg_req_links(msg)); - tipc_printf(buf, ":DDOM(%x):", msg_dest_domain(msg)); - tipc_printf(buf, ":NETID(%u):", msg_bc_netid(msg)); - tipc_media_addr_printf(buf, orig); - } - if (msg_user(msg) == BCAST_PROTOCOL) { - tipc_printf(buf, "BCNACK:AFTER(%u):", msg_bcgap_after(msg)); - tipc_printf(buf, "TO(%u):", msg_bcgap_to(msg)); +/* 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; } - tipc_printf(buf, "\n"); - if ((usr == CHANGEOVER_PROTOCOL) && (msg_msgcnt(msg))) { - tipc_msg_print(buf,msg_get_wrapped(msg)," /"); + 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 ((usr == MSG_FRAGMENTER) && (msg_type(msg) == FIRST_FRAGMENT)) { - tipc_msg_print(buf,msg_get_wrapped(msg)," /"); + 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 index e9ef6df2656..503511903d1 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -1,8 +1,8 @@ /* * net/tipc/msg.h: Include file for TIPC message header routines * - * Copyright (c) 2000-2007, Ericsson AB - * Copyright (c) 2005-2007, Wind River Systems + * 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 @@ -37,49 +37,87 @@ #ifndef _TIPC_MSG_H #define _TIPC_MSG_H -#include "core.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 -#define DATA_LOW TIPC_LOW_IMPORTANCE -#define DATA_MEDIUM TIPC_MEDIUM_IMPORTANCE -#define DATA_HIGH TIPC_HIGH_IMPORTANCE -#define DATA_CRITICAL TIPC_CRITICAL_IMPORTANCE -#define SHORT_H_SIZE 24 /* Connected,in cluster */ -#define DIR_MSG_H_SIZE 32 /* Directly addressed messages */ -#define CONN_MSG_H_SIZE 36 /* Routed connected msgs*/ -#define LONG_H_SIZE 40 /* Named Messages */ -#define MCAST_H_SIZE 44 /* Multicast messages */ -#define MAX_H_SIZE 60 /* Inclusive full options */ -#define MAX_MSG_SIZE (MAX_H_SIZE + TIPC_MAX_USER_MSG_SIZE) -#define LINK_CONFIG 13 +/* + * Payload message users are defined in TIPC's public API: + * - TIPC_LOW_IMPORTANCE + * - TIPC_MEDIUM_IMPORTANCE + * - TIPC_HIGH_IMPORTANCE + * - TIPC_CRITICAL_IMPORTANCE + */ /* - TIPC user data message header format, version 2 + * 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 - - Fundamental definitions available to privileged TIPC users - are located in tipc_msg.h. - - Remaining definitions available to TIPC internal users appear below. -*/ +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; - m->hdr[w] &= ~htonl(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); @@ -87,7 +125,7 @@ static inline u32 msg_version(struct tipc_msg *m) static inline void msg_set_version(struct tipc_msg *m) { - msg_set_bits(m, 0, 29, 0xf, TIPC_VERSION); + msg_set_bits(m, 0, 29, 7, TIPC_VERSION); } static inline u32 msg_user(struct tipc_msg *m) @@ -97,7 +135,7 @@ static inline u32 msg_user(struct tipc_msg *m) static inline u32 msg_isdata(struct tipc_msg *m) { - return (msg_user(m) <= DATA_CRITICAL); + return msg_user(m) <= TIPC_CRITICAL_IMPORTANCE; } static inline void msg_set_user(struct tipc_msg *m, u32 n) @@ -105,24 +143,44 @@ 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 void msg_set_hdr_sz(struct tipc_msg *m,u32 n) +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) +static inline void msg_set_non_seq(struct tipc_msg *m, u32 n) { - msg_set_bits(m, 0, 20, 1, 1); + msg_set_bits(m, 0, 20, 1, n); } static inline int msg_dest_droppable(struct tipc_msg *m) @@ -154,12 +212,36 @@ static inline void msg_set_size(struct tipc_msg *m, u32 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); @@ -190,18 +272,6 @@ static inline void msg_set_lookup_scope(struct tipc_msg *m, u32 n) msg_set_bits(m, 1, 19, 0x3, n); } -static inline void msg_set_options(struct tipc_msg *m, const char *opt, u32 sz) -{ - u32 hsz = msg_hdr_sz(m); - char *to = (char *)&m->hdr[hsz/4]; - - if ((hsz < DIR_MSG_H_SIZE) || ((hsz + sz) > MAX_H_SIZE)) - return; - msg_set_bits(m, 1, 16, 0x7, (hsz - 28)/4); - msg_set_hdr_sz(m, hsz + sz); - memcpy(to, opt, sz); -} - static inline u32 msg_bcast_ack(struct tipc_msg *m) { return msg_bits(m, 1, 0, 0xffff); @@ -216,7 +286,6 @@ static inline void msg_set_bcast_ack(struct tipc_msg *m, u32 n) /* * Word 2 */ - static inline u32 msg_ack(struct tipc_msg *m) { return msg_bits(m, 2, 16, 0xffff); @@ -237,77 +306,94 @@ 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 void msg_set_orignode(struct tipc_msg *m, u32 a) +static inline int msg_short(struct tipc_msg *m) { - msg_set_word(m, 6, a); + return msg_hdr_sz(m) == SHORT_H_SIZE; } -static inline void msg_set_destnode(struct tipc_msg *m, u32 a) +static inline u32 msg_orignode(struct tipc_msg *m) { - msg_set_word(m, 7, a); + if (likely(msg_short(m))) + return msg_prevnode(m); + return msg_word(m, 6); } -static inline int msg_is_dest(struct tipc_msg *m, u32 d) +static inline void msg_set_orignode(struct tipc_msg *m, u32 a) { - return(msg_short(m) || (msg_destnode(m) == d)); + msg_set_word(m, 6, a); } -static inline u32 msg_routed(struct tipc_msg *m) +static inline u32 msg_destnode(struct tipc_msg *m) { - if (likely(msg_short(m))) - return 0; - return(msg_destnode(m) ^ msg_orignode(m)) >> 11; + return msg_word(m, 7); } -static inline void msg_set_nametype(struct tipc_msg *m, u32 n) +static inline void msg_set_destnode(struct tipc_msg *m, u32 a) { - msg_set_word(m, 8, n); + msg_set_word(m, 7, a); } -static inline u32 msg_transp_seqno(struct tipc_msg *m) +static inline u32 msg_nametype(struct tipc_msg *m) { return msg_word(m, 8); } -static inline void msg_set_timestamp(struct tipc_msg *m, u32 n) +static inline void msg_set_nametype(struct tipc_msg *m, u32 n) { msg_set_word(m, 8, n); } -static inline u32 msg_timestamp(struct tipc_msg *m) +static inline u32 msg_nameinst(struct tipc_msg *m) { - return msg_word(m, 8); + return msg_word(m, 9); } -static inline void msg_set_transp_seqno(struct tipc_msg *m, u32 n) +static inline u32 msg_namelower(struct tipc_msg *m) { - msg_set_word(m, 8, n); + return msg_nameinst(m); } static inline void msg_set_namelower(struct tipc_msg *m, u32 n) @@ -320,122 +406,110 @@ 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 struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) +static inline unchar *msg_data(struct tipc_msg *m) { - return (struct tipc_msg *)msg_data(m); + return ((unchar *)m) + msg_hdr_sz(m); } -static inline void msg_expand(struct tipc_msg *m, u32 destnode) +static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) { - if (!msg_short(m)) - return; - msg_set_hdr_sz(m, LONG_H_SIZE); - msg_set_orignode(m, msg_prevnode(m)); - msg_set_destnode(m, destnode); - memset(&m->hdr[8], 0, 12); + return (struct tipc_msg *)msg_data(m); } - - /* - TIPC internal message header format, version 2 - - 1 0 9 8 7 6 5 4|3 2 1 0 9 8 7 6|5 4 3 2 1 0 9 8|7 6 5 4 3 2 1 0 - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w0:|vers |msg usr|hdr sz |n|resrv| packet size | - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w1:|m typ|rsv=0| sequence gap | broadcast ack no | - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w2:| link level ack no/bc_gap_from | seq no / bcast_gap_to | - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w3:| previous node | - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w4:| next sent broadcast/fragm no | next sent pkt/ fragm msg no | - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w5:| session no |rsv=0|r|berid|link prio|netpl|p| - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w6:| originating node | - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w7:| destination node | - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w8:| transport sequence number | - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - w9:| msg count / bcast tag | link tolerance | - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - \ \ - / User Specific Data / - \ \ - +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - - NB: CONN_MANAGER use data message format. LINK_CONFIG has own format. -*/ + * Constants and routines used to read and write TIPC internal message headers + */ /* - * Internal users + * Internal message users */ - #define BCAST_PROTOCOL 5 #define MSG_BUNDLER 6 #define LINK_PROTOCOL 7 #define CONN_MANAGER 8 -#define ROUTE_DISTRIBUTOR 9 +#define ROUTE_DISTRIBUTOR 9 /* obsoleted */ #define CHANGEOVER_PROTOCOL 10 #define NAME_DISTRIBUTOR 11 #define MSG_FRAGMENTER 12 #define LINK_CONFIG 13 -#define INT_H_SIZE 40 -#define DSC_H_SIZE 40 /* - * Connection management protocol messages + * Connection management protocol message types */ - #define CONN_PROBE 0 #define CONN_PROBE_REPLY 1 #define CONN_ACK 2 /* - * Name distributor messages + * Name distributor message types */ - #define PUBLICATION 0 #define WITHDRAWAL 1 +/* + * Segmentation message types + */ +#define FIRST_FRAGMENT 0 +#define FRAGMENT 1 +#define LAST_FRAGMENT 2 /* - * Word 1 + * 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, 0xff); + 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, 0xff, n); + msg_set_bits(m, 1, 16, 0x1fff, n); } -static inline u32 msg_req_links(struct tipc_msg *m) +static inline u32 msg_node_sig(struct tipc_msg *m) { - return msg_bits(m, 1, 16, 0xfff); + return msg_bits(m, 1, 0, 0xffff); } -static inline void msg_set_req_links(struct tipc_msg *m, u32 n) +static inline void msg_set_node_sig(struct tipc_msg *m, u32 n) { - msg_set_bits(m, 1, 16, 0xfff, 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); @@ -470,7 +544,6 @@ static inline void msg_set_bcgap_to(struct tipc_msg *m, u32 n) /* * Word 4 */ - static inline u32 msg_last_bcast(struct tipc_msg *m) { return msg_bits(m, 4, 16, 0xffff); @@ -481,12 +554,6 @@ static inline void msg_set_last_bcast(struct tipc_msg *m, u32 n) msg_set_bits(m, 4, 16, 0xffff, n); } - -static inline u32 msg_fragm_no(struct tipc_msg *m) -{ - return msg_bits(m, 4, 16, 0xffff); -} - static inline void msg_set_fragm_no(struct tipc_msg *m, u32 n) { msg_set_bits(m, 4, 16, 0xffff, n); @@ -503,12 +570,6 @@ static inline void msg_set_next_sent(struct tipc_msg *m, u32 n) msg_set_bits(m, 4, 0, 0xffff, n); } - -static inline u32 msg_long_msgno(struct tipc_msg *m) -{ - return msg_bits(m, 4, 0, 0xffff); -} - static inline void msg_set_long_msgno(struct tipc_msg *m, u32 n) { msg_set_bits(m, 4, 0, 0xffff, n); @@ -531,13 +592,12 @@ static inline u32 msg_link_selector(struct tipc_msg *m) static inline void msg_set_link_selector(struct tipc_msg *m, u32 n) { - msg_set_bits(m, 4, 0, 1, (n & 1)); + 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); @@ -555,7 +615,7 @@ static inline u32 msg_probe(struct tipc_msg *m) static inline void msg_set_probe(struct tipc_msg *m, u32 val) { - msg_set_bits(m, 5, 0, 1, (val & 1)); + msg_set_bits(m, 5, 0, 1, val); } static inline char msg_net_plane(struct tipc_msg *m) @@ -593,21 +653,19 @@ 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) +static inline void msg_set_redundant_link(struct tipc_msg *m, u32 r) { - msg_set_bits(m, 5, 12, 0x1, 1); + msg_set_bits(m, 5, 12, 0x1, r); } -static inline void msg_clear_redundant_link(struct tipc_msg *m) +static inline char *msg_media_addr(struct tipc_msg *m) { - msg_set_bits(m, 5, 12, 0x1, 0); + 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); @@ -630,7 +688,7 @@ static inline void msg_set_bcast_tag(struct tipc_msg *m, u32 n) static inline u32 msg_max_pkt(struct tipc_msg *m) { - return (msg_bits(m, 9, 16, 0xffff) * 4); + return msg_bits(m, 9, 16, 0xffff) * 4; } static inline void msg_set_max_pkt(struct tipc_msg *m, u32 n) @@ -648,168 +706,12 @@ static inline void msg_set_link_tolerance(struct tipc_msg *m, u32 n) msg_set_bits(m, 9, 0, 0xffff, n); } -/* - * Routing table message data - */ - - -static inline u32 msg_remote_node(struct tipc_msg *m) -{ - return msg_word(m, msg_hdr_sz(m)/4); -} - -static inline void msg_set_remote_node(struct tipc_msg *m, u32 a) -{ - msg_set_word(m, msg_hdr_sz(m)/4, a); -} - -static inline void msg_set_dataoctet(struct tipc_msg *m, u32 pos) -{ - msg_data(m)[pos + 4] = 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 - -/* - * Routing table message types - */ -#define EXT_ROUTING_TABLE 0 -#define LOCAL_ROUTING_TABLE 1 -#define SLAVE_ROUTING_TABLE 2 -#define ROUTE_ADDITION 3 -#define ROUTE_REMOVAL 4 - -/* - * Config protocol message types - */ - -#define DSC_REQ_MSG 0 -#define DSC_RESP_MSG 1 - -static inline u32 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); -} - +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); -static inline void msg_init(struct tipc_msg *m, u32 user, u32 type, - u32 err, 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_errcode(m, err); - if (!msg_short(m)) { - msg_set_orignode(m, tipc_own_addr); - msg_set_destnode(m, destnode); - } -} - -/** - * msg_calc_data_size - determine total data size for message - */ - -static inline int msg_calc_data_size(struct iovec const *msg_sect, u32 num_sect) -{ - int dsz = 0; - int i; - - for (i = 0; i < num_sect; i++) - dsz += msg_sect[i].iov_len; - return dsz; -} - -/** - * 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 - */ - -static inline int msg_build(struct tipc_msg *hdr, - struct iovec const *msg_sect, u32 num_sect, - int max_size, int usrmem, struct sk_buff** buf) -{ - int dsz, sz, hsz, pos, res, cnt; - - dsz = msg_calc_data_size(msg_sect, num_sect); - if (unlikely(dsz > TIPC_MAX_USER_MSG_SIZE)) { - *buf = NULL; - return -EINVAL; - } - - pos = hsz = msg_hdr_sz(hdr); - sz = hsz + dsz; - msg_set_size(hdr, sz); - if (unlikely(sz > max_size)) { - *buf = NULL; - return dsz; - } - - *buf = buf_acquire(sz); - if (!(*buf)) - return -ENOMEM; - skb_copy_to_linear_data(*buf, hdr, hsz); - for (res = 1, cnt = 0; res && (cnt < num_sect); cnt++) { - if (likely(usrmem)) - res = !copy_from_user((*buf)->data + pos, - msg_sect[cnt].iov_base, - msg_sect[cnt].iov_len); - else - skb_copy_to_linear_data_offset(*buf, pos, - msg_sect[cnt].iov_base, - msg_sect[cnt].iov_len); - pos += msg_sect[cnt].iov_len; - } - if (likely(res)) - return dsz; - - buf_discard(*buf); - *buf = NULL; - return -EFAULT; -} - -static inline void msg_set_media_addr(struct tipc_msg *m, struct tipc_media_addr *a) -{ - memcpy(&((int *)m)[5], a, sizeof(*a)); -} - -static inline void msg_get_media_addr(struct tipc_msg *m, struct tipc_media_addr *a) -{ - memcpy(a, &((int*)m)[5], sizeof(*a)); -} +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 index 39fd1619feb..8ce730984aa 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -2,7 +2,7 @@ * net/tipc/name_distr.c: TIPC name distribution code * * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -35,56 +35,45 @@ */ #include "core.h" -#include "cluster.h" -#include "dbg.h" #include "link.h" -#include "msg.h" #include "name_distr.h" -#undef DBG_OUTPUT -#define DBG_OUTPUT NULL - -#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 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; +}; -struct distr_item { - __be32 type; - __be32 lower; - __be32 upper; - __be32 ref; - __be32 key; +static struct publ_list publ_zone = { + .list = LIST_HEAD_INIT(publ_zone.list), + .size = 0, }; -/** - * List of externally visible publications by this node -- - * that is, all publications having scope > TIPC_NODE_SCOPE. - */ +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] */ +}; -static LIST_HEAD(publ_root); -static u32 publ_cnt = 0; /** * 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); @@ -92,101 +81,120 @@ static void publ_to_item(struct distr_item *i, struct publication *p) i->upper = htonl(p->upper); i->ref = htonl(p->ref); i->key = htonl(p->key); - dbg("publ_to_item: %u, %u, %u\n", p->type, p->lower, p->upper); } /** * 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 = buf_acquire(LONG_H_SIZE + size); + struct sk_buff *buf = tipc_buf_acquire(INT_H_SIZE + size); struct tipc_msg *msg; if (buf != NULL) { msg = buf_msg(buf); - msg_init(msg, NAME_DISTRIBUTOR, type, TIPC_OK, - LONG_H_SIZE, dest); - msg_set_size(msg, LONG_H_SIZE + size); + 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 */ - -void tipc_named_publish(struct publication *publ) +struct sk_buff *tipc_named_publish(struct publication *publ) { struct sk_buff *buf; struct distr_item *item; - list_add_tail(&publ->local_list, &publ_root); - publ_cnt++; + 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) { - warn("Publication distribution failure\n"); - return; + pr_warn("Publication distribution failure\n"); + return NULL; } item = (struct distr_item *)msg_data(buf_msg(buf)); publ_to_item(item, publ); - dbg("tipc_named_withdraw: broadcasting publish msg\n"); - tipc_cltr_broadcast(buf); + return buf; } /** * tipc_named_withdraw - tell other nodes about a withdrawn publication by this node */ - -void tipc_named_withdraw(struct publication *publ) +struct sk_buff *tipc_named_withdraw(struct publication *publ) { struct sk_buff *buf; struct distr_item *item; list_del(&publ->local_list); - publ_cnt--; + publ_lists[publ->scope]->size--; + + if (publ->scope == TIPC_NODE_SCOPE) + return NULL; buf = named_prepare_buf(WITHDRAWAL, ITEM_SIZE, 0); if (!buf) { - warn("Withdrawl distribution failure\n"); - return; + pr_warn("Withdrawal distribution failure\n"); + return NULL; } item = (struct distr_item *)msg_data(buf_msg(buf)); publ_to_item(item, publ); - dbg("tipc_named_withdraw: broadcasting withdraw msg\n"); - tipc_cltr_broadcast(buf); + return buf; } -/** - * tipc_named_node_up - tell specified node about all publications by this node +/* + * named_distribute - prepare name info for bulk distribution to another node */ - -void tipc_named_node_up(unsigned long node) +static void named_distribute(struct list_head *message_list, u32 node, + struct publ_list *pls, u32 max_item_buf) { struct publication *publ; - struct distr_item *item = NULL; struct sk_buff *buf = NULL; + struct distr_item *item = NULL; u32 left = 0; - u32 rest; - u32 max_item_buf; - - read_lock_bh(&tipc_nametbl_lock); - max_item_buf = TIPC_MAX_USER_MSG_SIZE / ITEM_SIZE; - max_item_buf *= ITEM_SIZE; - rest = publ_cnt * ITEM_SIZE; + u32 rest = pls->size * ITEM_SIZE; - list_for_each_entry(publ, &publ_root, local_list) { + 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) { - warn("Bulk publication distribution failure\n"); - goto exit; + pr_warn("Bulk publication failure\n"); + return; } item = (struct distr_item *)msg_data(buf_msg(buf)); } @@ -194,57 +202,58 @@ void tipc_named_node_up(unsigned long node) item++; left -= ITEM_SIZE; if (!left) { - msg_set_link_selector(buf_msg(buf), node); - dbg("tipc_named_node_up: sending publish msg to " - "<%u.%u.%u>\n", tipc_zone(node), - tipc_cluster(node), tipc_node(node)); - tipc_link_send(buf, node, node); + list_add_tail((struct list_head *)buf, message_list); buf = NULL; } } -exit: +} + +/** + * 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); } /** - * node_is_down - remove publication associated with a failed 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. - * In rare cases the link may have come back up again when this - * function is called, and we have two items representing the same - * publication. Nudge this item's key to distinguish it from the other. - * (Note: Publication's node subscription is already unsubscribed.) */ - -static void node_is_down(struct publication *publ) +static void named_purge_publ(struct publication *publ) { struct publication *p; write_lock_bh(&tipc_nametbl_lock); - dbg("node_is_down: withdrawing %u, %u, %u\n", - publ->type, publ->lower, publ->upper); - publ->key += 1222345; 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) { - 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); + 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); } - if (p) { - kfree(p); - } + kfree(p); } /** - * tipc_named_recv - process name table update message sent by another node + * tipc_named_rcv - process name table update message sent by another node */ - -void tipc_named_recv(struct sk_buff *buf) +void tipc_named_rcv(struct sk_buff *buf) { struct publication *publ; struct tipc_msg *msg = buf_msg(buf); @@ -254,9 +263,6 @@ void tipc_named_recv(struct sk_buff *buf) write_lock_bh(&tipc_nametbl_lock); while (count--) { if (msg_type(msg) == PUBLICATION) { - dbg("tipc_named_recv: got publication for %u, %u, %u\n", - ntohl(item->type), ntohl(item->lower), - ntohl(item->upper)); publ = tipc_nametbl_insert_publ(ntohl(item->type), ntohl(item->lower), ntohl(item->upper), @@ -268,12 +274,10 @@ void tipc_named_recv(struct sk_buff *buf) tipc_nodesub_subscribe(&publ->subscr, msg_orignode(msg), publ, - (net_ev_handler)node_is_down); + (net_ev_handler) + named_purge_publ); } } else if (msg_type(msg) == WITHDRAWAL) { - dbg("tipc_named_recv: got withdrawl for %u, %u, %u\n", - ntohl(item->type), ntohl(item->lower), - ntohl(item->upper)); publ = tipc_nametbl_remove_publ(ntohl(item->type), ntohl(item->lower), msg_orignode(msg), @@ -284,39 +288,38 @@ void tipc_named_recv(struct sk_buff *buf) tipc_nodesub_unsubscribe(&publ->subscr); kfree(publ); } else { - 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)); + 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 { - warn("Unrecognized name table message received\n"); + pr_warn("Unrecognized name table message received\n"); } item++; } write_unlock_bh(&tipc_nametbl_lock); - buf_discard(buf); + kfree_skb(buf); } /** - * tipc_named_reinit - re-initialize local publication list + * tipc_named_reinit - re-initialize local publications * - * This routine is called whenever TIPC networking is (re)enabled. - * All existing publications by this node that have "cluster" or "zone" scope - * are updated to reflect the node's current network address. - * (If the node's address is unchanged, the update loop terminates immediately.) + * 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); - list_for_each_entry(publ, &publ_root, local_list) { - if (publ->node == tipc_own_addr) - break; - publ->node = tipc_own_addr; - } + + 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 index 1e41bdd4f25..b2eed4ec152 100644 --- a/net/tipc/name_distr.h +++ b/net/tipc/name_distr.h @@ -39,10 +39,39 @@ #include "name_table.h" -void tipc_named_publish(struct publication *publ); -void tipc_named_withdraw(struct publication *publ); -void tipc_named_node_up(unsigned long node); -void tipc_named_recv(struct sk_buff *buf); +#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 index ac7dfdda797..9d7d37d9518 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -2,7 +2,7 @@ * net/tipc/name_table.c: TIPC name table code * * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2004-2005, Wind River Systems + * Copyright (c) 2004-2008, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -36,33 +36,45 @@ #include "core.h" #include "config.h" -#include "dbg.h" #include "name_table.h" #include "name_distr.h" -#include "addr.h" -#include "node_subscr.h" #include "subscr.h" #include "port.h" -#include "cluster.h" -#include "bcast.h" -static int tipc_nametbl_size = 1024; /* must be a power of 2 */ +#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 - * @node_list: circular list of matching publications with >= node scope - * @cluster_list: circular list of matching publications with >= cluster scope - * @zone_list: circular list of matching publications with >= zone scope + * @info: pointer to name sequence publication info */ - struct sub_seq { u32 lower; u32 upper; - struct publication *node_list; - struct publication *cluster_list; - struct publication *zone_list; + struct name_info *info; }; /** @@ -74,9 +86,8 @@ struct sub_seq { * @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 name sequence structure + * @lock: spinlock controlling access to publication lists of all sub-sequences */ - struct name_seq { u32 type; struct sub_seq *sseqs; @@ -93,33 +104,29 @@ struct name_seq { * 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 = { NULL } ; -static atomic_t rsv_publ_ok = ATOMIC_INIT(0); +static struct name_table table; DEFINE_RWLOCK(tipc_nametbl_lock); - static int hash(int x) { - return(x & (tipc_nametbl_size - 1)); + 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) { - warn("Publication creation failure, no memory\n"); + pr_warn("Publication creation failure, no memory\n"); return NULL; } @@ -139,11 +146,9 @@ static struct publication *publ_create(u32 type, u32 lower, u32 upper, /** * tipc_subseq_alloc - allocate a specified number of sub-sequence structures */ - static struct sub_seq *tipc_subseq_alloc(u32 cnt) { - struct sub_seq *sseq = kcalloc(cnt, sizeof(struct sub_seq), GFP_ATOMIC); - return sseq; + return kcalloc(cnt, sizeof(struct sub_seq), GFP_ATOMIC); } /** @@ -151,14 +156,13 @@ static struct sub_seq *tipc_subseq_alloc(u32 cnt) * * 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) { - warn("Name sequence creation failed, no memory\n"); + pr_warn("Name sequence creation failed, no memory\n"); kfree(nseq); kfree(sseq); return NULL; @@ -167,8 +171,6 @@ static struct name_seq *tipc_nameseq_create(u32 type, struct hlist_head *seq_hea spin_lock_init(&nseq->lock); nseq->type = type; nseq->sseqs = sseq; - dbg("tipc_nameseq_create(): nseq = %p, type %u, ssseqs %p, ff: %u\n", - nseq, type, nseq->sseqs, nseq->first_free); nseq->alloc = 1; INIT_HLIST_NODE(&nseq->ns_list); INIT_LIST_HEAD(&nseq->subscriptions); @@ -176,12 +178,23 @@ static struct name_seq *tipc_nameseq_create(u32 type, struct hlist_head *seq_hea 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) { @@ -211,7 +224,6 @@ static struct sub_seq *nameseq_find_subseq(struct name_seq *nseq, * * 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; @@ -232,122 +244,111 @@ static u32 nameseq_locate_subseq(struct name_seq *nseq, u32 instance) } /** - * tipc_nameseq_insert_publ - + * 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 subscription *s; - struct subscription *st; + 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); - dbg("nameseq_ins: for seq %p, {%u,%u}, found sseq %p\n", - nseq, type, lower, sseq); if (sseq) { /* Lower end overlaps existing entry => need an exact match */ - if ((sseq->lower != lower) || (sseq->upper != upper)) { - warn("Cannot publish {%u,%u,%u}, overlap error\n", - type, lower, 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)) { - warn("Cannot publish {%u,%u,%u}, overlap error\n", - type, lower, upper); + 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) { - warn("Cannot publish {%u,%u,%u}, no memory\n", - type, lower, upper); + pr_warn("Cannot publish {%u,%u,%u}, no memory\n", + type, lower, upper); return NULL; } - dbg("Allocated %u more sseqs\n", nseq->alloc); memcpy(sseqs, nseq->sseqs, nseq->alloc * sizeof(struct sub_seq)); kfree(nseq->sseqs); nseq->sseqs = sseqs; nseq->alloc *= 2; } - dbg("Have %u sseqs for type %u\n", nseq->alloc, type); - /* Insert new sub-sequence */ + 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); - dbg("ins in pos %u, ff = %u\n", inspos, nseq->first_free); + /* 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)); + 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; } - dbg("inserting {%u,%u,%u} from <0x%x:%u> into sseq %p(%u,%u) of seq %p\n", - type, lower, upper, node, port, sseq, - sseq->lower, sseq->upper, nseq); - - /* Insert a publication: */ + /* Insert a publication */ publ = publ_create(type, lower, upper, scope, node, port, key); if (!publ) return NULL; - dbg("inserting publ %p, node=0x%x publ->node=0x%x, subscr->node=%p\n", - publ, node, publ->node, publ->subscr.node); - - if (!sseq->zone_list) - sseq->zone_list = publ->zone_list_next = publ; - else { - publ->zone_list_next = sseq->zone_list->zone_list_next; - sseq->zone_list->zone_list_next = publ; - } + + list_add(&publ->zone_list, &info->zone_list); + info->zone_list_size++; if (in_own_cluster(node)) { - if (!sseq->cluster_list) - sseq->cluster_list = publ->cluster_list_next = publ; - else { - publ->cluster_list_next = - sseq->cluster_list->cluster_list_next; - sseq->cluster_list->cluster_list_next = publ; - } + list_add(&publ->cluster_list, &info->cluster_list); + info->cluster_list_size++; } - if (node == tipc_own_addr) { - if (!sseq->node_list) - sseq->node_list = publ->node_list_next = publ; - else { - publ->node_list_next = sseq->node_list->node_list_next; - sseq->node_list->node_list_next = publ; - } + if (in_own_node(node)) { + list_add(&publ->node_list, &info->node_list); + info->node_list_size++; } - /* - * Any subscriptions waiting for notification? - */ + /* Any subscriptions waiting for notification? */ list_for_each_entry_safe(s, st, &nseq->subscriptions, nameseq_list) { - dbg("calling report_overlap()\n"); tipc_subscr_report_overlap(s, publ->lower, publ->upper, @@ -360,7 +361,7 @@ static struct publication *tipc_nameseq_insert_publ(struct name_seq *nseq, } /** - * tipc_nameseq_remove_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 @@ -370,118 +371,55 @@ static struct publication *tipc_nameseq_insert_publ(struct name_seq *nseq, * 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 publication *curr; - struct publication *prev; struct sub_seq *sseq = nameseq_find_subseq(nseq, inst); + struct name_info *info; struct sub_seq *free; - struct subscription *s, *st; + struct tipc_subscription *s, *st; int removed_subseq = 0; if (!sseq) return NULL; - dbg("tipc_nameseq_remove_publ: seq: %p, sseq %p, {%u,%u}, key %u\n", - nseq, sseq, nseq->type, inst, key); - - /* Remove publication from zone scope list */ - - prev = sseq->zone_list; - publ = sseq->zone_list->zone_list_next; - while ((publ->key != key) || (publ->ref != ref) || - (publ->node && (publ->node != node))) { - prev = publ; - publ = publ->zone_list_next; - if (prev == sseq->zone_list) { - - /* Prevent endless loop if publication not found */ + info = sseq->info; - return NULL; - } - } - if (publ != sseq->zone_list) - prev->zone_list_next = publ->zone_list_next; - else if (publ->zone_list_next != publ) { - prev->zone_list_next = publ->zone_list_next; - sseq->zone_list = publ->zone_list_next; - } else { - sseq->zone_list = NULL; + /* 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; - /* Remove publication from cluster scope list, if present */ +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)) { - prev = sseq->cluster_list; - curr = sseq->cluster_list->cluster_list_next; - while (curr != publ) { - prev = curr; - curr = curr->cluster_list_next; - if (prev == sseq->cluster_list) { - - /* Prevent endless loop for malformed list */ - - err("Unable to de-list cluster publication\n" - "{%u%u}, node=0x%x, ref=%u, key=%u)\n", - publ->type, publ->lower, publ->node, - publ->ref, publ->key); - goto end_cluster; - } - } - if (publ != sseq->cluster_list) - prev->cluster_list_next = publ->cluster_list_next; - else if (publ->cluster_list_next != publ) { - prev->cluster_list_next = publ->cluster_list_next; - sseq->cluster_list = publ->cluster_list_next; - } else { - sseq->cluster_list = NULL; - } + list_del(&publ->cluster_list); + info->cluster_list_size--; } -end_cluster: /* Remove publication from node scope list, if present */ - - if (node == tipc_own_addr) { - prev = sseq->node_list; - curr = sseq->node_list->node_list_next; - while (curr != publ) { - prev = curr; - curr = curr->node_list_next; - if (prev == sseq->node_list) { - - /* Prevent endless loop for malformed list */ - - err("Unable to de-list node publication\n" - "{%u%u}, node=0x%x, ref=%u, key=%u)\n", - publ->type, publ->lower, publ->node, - publ->ref, publ->key); - goto end_node; - } - } - if (publ != sseq->node_list) - prev->node_list_next = publ->node_list_next; - else if (publ->node_list_next != publ) { - prev->node_list_next = publ->node_list_next; - sseq->node_list = publ->node_list_next; - } else { - sseq->node_list = NULL; - } + if (in_own_node(node)) { + list_del(&publ->node_list); + info->node_list_size--; } -end_node: /* Contract subseq list if no more publications for that subseq */ - - if (!sseq->zone_list) { + if (list_empty(&info->zone_list)) { + kfree(info); free = &nseq->sseqs[nseq->first_free--]; - memmove(sseq, sseq + 1, (free - (sseq + 1)) * sizeof (*sseq)); + 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, @@ -496,12 +434,12 @@ end_node: } /** - * tipc_nameseq_subscribe: attach a subscription, and issue + * 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 subscription *s) +static void tipc_nameseq_subscribe(struct name_seq *nseq, + struct tipc_subscription *s) { struct sub_seq *sseq = nseq->sseqs; @@ -511,12 +449,12 @@ static void tipc_nameseq_subscribe(struct name_seq *nseq, struct subscription *s return; while (sseq != &nseq->sseqs[nseq->first_free]) { - struct publication *zl = sseq->zone_list; - if (zl && tipc_subscr_overlap(s,sseq->lower,sseq->upper)) { - struct publication *crs = zl; + if (tipc_subscr_overlap(s, sseq->lower, sseq->upper)) { + struct publication *crs; + struct name_info *info = sseq->info; int must_report = 1; - do { + list_for_each_entry(crs, &info->zone_list, zone_list) { tipc_subscr_report_overlap(s, sseq->lower, sseq->upper, @@ -525,8 +463,7 @@ static void tipc_nameseq_subscribe(struct name_seq *nseq, struct subscription *s crs->node, must_report); must_report = 0; - crs = crs->zone_list_next; - } while (crs != zl); + } } sseq++; } @@ -535,18 +472,12 @@ static void tipc_nameseq_subscribe(struct name_seq *nseq, struct subscription *s static struct name_seq *nametbl_find_seq(u32 type) { struct hlist_head *seq_head; - struct hlist_node *seq_node; struct name_seq *ns; - dbg("find_seq %u,(%u,0x%x) table = %p, hash[type] = %u\n", - type, ntohl(type), type, table.types, hash(type)); - seq_head = &table.types[hash(type)]; - hlist_for_each_entry(ns, seq_node, seq_head, ns_list) { - if (ns->type == type) { - dbg("found %p\n", ns); + hlist_for_each_entry(ns, seq_head, ns_list) { + if (ns->type == type) return ns; - } } return NULL; @@ -557,18 +488,15 @@ struct publication *tipc_nametbl_insert_publ(u32 type, u32 lower, u32 upper, { struct name_seq *seq = nametbl_find_seq(type); - dbg("tipc_nametbl_insert_publ: {%u,%u,%u} found %p\n", type, lower, upper, seq); - if (lower > upper) { - warn("Failed to publish illegal {%u,%u,%u}\n", - type, lower, upper); + 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; } - dbg("Publishing {%u,%u,%u} from 0x%x\n", type, lower, upper, node); - if (!seq) { + if (!seq) seq = tipc_nameseq_create(type, &table.types[hash(type)]); - dbg("tipc_nametbl_insert_publ: created %p\n", seq); - } if (!seq) return NULL; @@ -585,33 +513,34 @@ struct publication *tipc_nametbl_remove_publ(u32 type, u32 lower, if (!seq) return NULL; - dbg("Withdrawing {%u,%u} from 0x%x\n", type, lower, node); publ = tipc_nameseq_remove_publ(seq, lower, node, ref, key); - - if (!seq->first_free && list_empty(&seq->subscriptions)) { - hlist_del_init(&seq->ns_list); - kfree(seq->sseqs); - kfree(seq); - } + nameseq_delete_empty(seq); return publ; } -/* - * tipc_nametbl_translate(): Translate tipc_name -> tipc_portid. - * Very time-critical. +/** + * tipc_nametbl_translate - perform name translation * - * Note: on entry 'destnode' is the search domain used during translation; - * on exit it passes back the node address of the matching port (if any) + * 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 publication *publ = NULL; + struct name_info *info; + struct publication *publ; struct name_seq *seq; - u32 ref; + u32 ref = 0; + u32 node = 0; - if (!in_scope(*destnode, tipc_own_addr)) + if (!tipc_in_scope(*destnode, tipc_own_addr)) return 0; read_lock_bh(&tipc_nametbl_lock); @@ -622,56 +551,58 @@ u32 tipc_nametbl_translate(u32 type, u32 instance, u32 *destnode) if (unlikely(!sseq)) goto not_found; spin_lock_bh(&seq->lock); + info = sseq->info; - /* Closest-First Algorithm: */ + /* Closest-First Algorithm */ if (likely(!*destnode)) { - publ = sseq->node_list; - if (publ) { - sseq->node_list = publ->node_list_next; -found: - ref = publ->ref; - *destnode = publ->node; - spin_unlock_bh(&seq->lock); - read_unlock_bh(&tipc_nametbl_lock); - return ref; - } - publ = sseq->cluster_list; - if (publ) { - sseq->cluster_list = publ->cluster_list_next; - goto found; - } - publ = sseq->zone_list; - if (publ) { - sseq->zone_list = publ->zone_list_next; - goto found; + 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: */ + /* Round-Robin Algorithm */ else if (*destnode == tipc_own_addr) { - publ = sseq->node_list; - if (publ) { - sseq->node_list = publ->node_list_next; - goto found; - } - } else if (in_own_cluster(*destnode)) { - publ = sseq->cluster_list; - if (publ) { - sseq->cluster_list = publ->cluster_list_next; - goto found; - } + 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 = sseq->zone_list; - if (publ) { - sseq->zone_list = publ->zone_list_next; - goto found; - } + 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: - *destnode = 0; read_unlock_bh(&tipc_nametbl_lock); - return 0; + *destnode = node; + return ref; } /** @@ -686,13 +617,13 @@ not_found: * * Returns non-zero if any off-node ports overlap */ - int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, - struct port_list *dports) + 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); @@ -709,15 +640,15 @@ int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, if (sseq->lower > upper) break; - publ = sseq->cluster_list; - if (publ && (publ->scope <= limit)) - do { - if (publ->node == tipc_own_addr) - tipc_port_list_add(dports, publ->ref); - else - res = 1; - publ = publ->cluster_list_next; - } while (publ != sseq->cluster_list); + + 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); @@ -726,102 +657,82 @@ exit: return res; } -/** - * tipc_nametbl_publish_rsv - publish port name using a reserved name type - */ - -int tipc_nametbl_publish_rsv(u32 ref, unsigned int scope, - struct tipc_name_seq const *seq) -{ - int res; - - atomic_inc(&rsv_publ_ok); - res = tipc_publish(ref, scope, seq); - atomic_dec(&rsv_publ_ok); - 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) + u32 scope, u32 port_ref, u32 key) { struct publication *publ; + struct sk_buff *buf = NULL; - if (table.local_publ_count >= tipc_max_publications) { - warn("Publication failed, local publication limit reached (%u)\n", - tipc_max_publications); - return NULL; - } - if ((type < TIPC_RESERVED_TYPES) && !atomic_read(&rsv_publ_ok)) { - warn("Publication failed, reserved name {%u,%u,%u}\n", - type, lower, upper); + 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); - table.local_publ_count++; publ = tipc_nametbl_insert_publ(type, lower, upper, scope, tipc_own_addr, port_ref, key); - if (publ && (scope != TIPC_NODE_SCOPE)) { - tipc_named_publish(publ); + 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; - dbg("tipc_nametbl_withdraw: {%u,%u}, key=%u\n", type, lower, key); 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--; - if (publ->scope != TIPC_NODE_SCOPE) - tipc_named_withdraw(publ); + 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); - err("Unable to remove local publication\n" - "(type=%u, lower=%u, ref=%u, key=%u)\n", - type, lower, ref, key); + 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 subscription *s) +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) { + if (!seq) seq = tipc_nameseq_create(type, &table.types[hash(type)]); - } - if (seq){ + if (seq) { spin_lock_bh(&seq->lock); - dbg("tipc_nametbl_subscribe:found %p for {%u,%u,%u}\n", - seq, type, s->seq.lower, s->seq.upper); tipc_nameseq_subscribe(seq, s); spin_unlock_bh(&seq->lock); } else { - warn("Failed to create subscription for {%u,%u,%u}\n", - s->seq.type, s->seq.lower, s->seq.upper); + 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); } @@ -829,136 +740,128 @@ void tipc_nametbl_subscribe(struct subscription *s) /** * tipc_nametbl_unsubscribe - remove a subscription object from name table */ - -void tipc_nametbl_unsubscribe(struct subscription *s) +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){ + if (seq != NULL) { spin_lock_bh(&seq->lock); list_del_init(&s->nameseq_list); spin_unlock_bh(&seq->lock); - if ((seq->first_free == 0) && list_empty(&seq->subscriptions)) { - hlist_del_init(&seq->ns_list); - kfree(seq->sseqs); - kfree(seq); - } + nameseq_delete_empty(seq); } write_unlock_bh(&tipc_nametbl_lock); } /** - * subseq_list: print specified sub-sequence contents into the given buffer + * subseq_list - print specified sub-sequence contents into the given buffer */ - -static void subseq_list(struct sub_seq *sseq, struct print_buf *buf, u32 depth, - u32 index) +static int subseq_list(struct sub_seq *sseq, char *buf, int len, u32 depth, + u32 index) { char portIdStr[27]; - char *scopeStr; - struct publication *publ = sseq->zone_list; + const char *scope_str[] = {"", " zone", " cluster", " node"}; + struct publication *publ; + struct name_info *info; + int ret; - tipc_printf(buf, "%-10u %-10u ", sseq->lower, sseq->upper); + ret = tipc_snprintf(buf, len, "%-10u %-10u ", sseq->lower, sseq->upper); - if (depth == 2 || !publ) { - tipc_printf(buf, "\n"); - return; + if (depth == 2) { + ret += tipc_snprintf(buf - ret, len + ret, "\n"); + return ret; } - do { - sprintf (portIdStr, "<%u.%u.%u:%u>", + 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); - tipc_printf(buf, "%-26s ", portIdStr); + ret += tipc_snprintf(buf + ret, len - ret, "%-26s ", portIdStr); if (depth > 3) { - if (publ->node != tipc_own_addr) - scopeStr = ""; - else if (publ->scope == TIPC_NODE_SCOPE) - scopeStr = "node"; - else if (publ->scope == TIPC_CLUSTER_SCOPE) - scopeStr = "cluster"; - else - scopeStr = "zone"; - tipc_printf(buf, "%-10u %s", publ->key, scopeStr); + 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", " "); + } - publ = publ->zone_list_next; - if (publ == sseq->zone_list) - break; - - tipc_printf(buf, "\n%33s", " "); - } while (1); - - tipc_printf(buf, "\n"); + ret += tipc_snprintf(buf + ret, len - ret, "\n"); + return ret; } /** - * nameseq_list: print specified name sequence contents into the given buffer + * nameseq_list - print specified name sequence contents into the given buffer */ - -static void nameseq_list(struct name_seq *seq, struct print_buf *buf, u32 depth, - u32 type, u32 lowbound, u32 upbound, u32 index) +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) { - tipc_printf(buf, "%s\n", typearea); - return; + 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)) { - tipc_printf(buf, "%s ", typearea); - subseq_list(sseq, buf, depth, index); + 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 void nametbl_header(struct print_buf *buf, u32 depth) +static int nametbl_header(char *buf, int len, u32 depth) { - tipc_printf(buf, "Type "); - - if (depth > 1) - tipc_printf(buf, "Lower Upper "); - if (depth > 2) - tipc_printf(buf, "Port Identity "); - if (depth > 3) - tipc_printf(buf, "Publication"); - - tipc_printf(buf, "\n-----------"); - - if (depth > 1) - tipc_printf(buf, "--------------------- "); - if (depth > 2) - tipc_printf(buf, "-------------------------- "); - if (depth > 3) - tipc_printf(buf, "------------------"); - - tipc_printf(buf, "\n"); + 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 void nametbl_list(struct print_buf *buf, u32 depth_info, - u32 type, u32 lowbound, u32 upbound) +static int nametbl_list(char *buf, int len, u32 depth_info, + u32 type, u32 lowbound, u32 upbound) { struct hlist_head *seq_head; - struct hlist_node *seq_node; struct name_seq *seq; int all_types; + int ret = 0; u32 depth; u32 i; @@ -966,118 +869,130 @@ static void nametbl_list(struct print_buf *buf, u32 depth_info, depth = (depth_info & ~TIPC_NTQ_ALLTYPES); if (depth == 0) - return; + return 0; if (all_types) { /* display all entries in name table to specified depth */ - nametbl_header(buf, depth); + ret += nametbl_header(buf, len, depth); lowbound = 0; upbound = ~0; - for (i = 0; i < tipc_nametbl_size; i++) { + for (i = 0; i < TIPC_NAMETBL_SIZE; i++) { seq_head = &table.types[i]; - hlist_for_each_entry(seq, seq_node, seq_head, ns_list) { - nameseq_list(seq, buf, depth, seq->type, - lowbound, upbound, 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) { - tipc_printf(buf, "invalid name sequence specified\n"); - return; + ret += tipc_snprintf(buf + ret, len - ret, + "invalid name sequence specified\n"); + return ret; } - nametbl_header(buf, depth); + ret += nametbl_header(buf + ret, len - ret, depth); i = hash(type); seq_head = &table.types[i]; - hlist_for_each_entry(seq, seq_node, seq_head, ns_list) { + hlist_for_each_entry(seq, seq_head, ns_list) { if (seq->type == type) { - nameseq_list(seq, buf, depth, type, - lowbound, upbound, i); + ret += nameseq_list(seq, buf + ret, len - ret, + depth, type, + lowbound, upbound, i); break; } } } + return ret; } -#if 0 -void tipc_nametbl_print(struct print_buf *buf, const char *str) -{ - tipc_printf(buf, str); - read_lock_bh(&tipc_nametbl_lock); - nametbl_list(buf, 0, 0, 0, 0); - read_unlock_bh(&tipc_nametbl_lock); -} -#endif - -#define MAX_NAME_TBL_QUERY 32768 - 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; - struct print_buf b; + 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(MAX_NAME_TBL_QUERY)); + buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); if (!buf) return NULL; rep_tlv = (struct tlv_desc *)buf->data; - tipc_printbuf_init(&b, TLV_DATA(rep_tlv), MAX_NAME_TBL_QUERY); + 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); - nametbl_list(&b, ntohl(argv->depth), ntohl(argv->type), - ntohl(argv->lowbound), ntohl(argv->upbound)); + 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 = tipc_printbuf_validate(&b); - + 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; } -#if 0 -void tipc_nametbl_dump(void) -{ - nametbl_list(TIPC_CONS, 0, 0, 0, 0); -} -#endif - int tipc_nametbl_init(void) { - int array_size = sizeof(struct hlist_head) * tipc_nametbl_size; - - table.types = kzalloc(array_size, GFP_ATOMIC); + table.types = kcalloc(TIPC_NAMETBL_SIZE, sizeof(struct hlist_head), + GFP_ATOMIC); if (!table.types) return -ENOMEM; - write_lock_bh(&tipc_nametbl_lock); table.local_publ_count = 0; - write_unlock_bh(&tipc_nametbl_lock); return 0; } -void tipc_nametbl_stop(void) +/** + * 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) { - u32 i; + struct publication *publ, *safe; + struct sub_seq *sseq; + struct name_info *info; - if (!table.types) + 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); + } +} - /* Verify name table is empty, then release it */ +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])) - err("tipc_nametbl_stop(): hash chain %u is non-null\n", i); + 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 index b9e7cd336d7..f02f48b9a21 100644 --- a/net/tipc/name_table.h +++ b/net/tipc/name_table.h @@ -2,7 +2,7 @@ * net/tipc/name_table.h: Include file for TIPC name table code * * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2004-2005, Wind River Systems + * Copyright (c) 2004-2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -39,15 +39,13 @@ #include "node_subscr.h" -struct subscription; -struct port_list; +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 */ - +#define TIPC_ZM_SRV 3 /* zone master service name type */ /** * struct publication - info about a published (name or) name sequence @@ -61,13 +59,12 @@ struct port_list; * @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: next matching name seq publication with >= node scope - * @cluster_list: next matching name seq publication with >= cluster scope - * @zone_list: next matching name seq publication with >= zone scope + * @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; @@ -76,12 +73,12 @@ struct publication { u32 node; u32 ref; u32 key; - struct node_subscr subscr; + struct tipc_node_subscr subscr; struct list_head local_list; struct list_head pport_list; - struct publication *node_list_next; - struct publication *cluster_list_next; - struct publication *zone_list_next; + struct list_head node_list; + struct list_head cluster_list; + struct list_head zone_list; }; @@ -90,18 +87,17 @@ 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 port_list *dports); -int tipc_nametbl_publish_rsv(u32 ref, unsigned int scope, - struct tipc_name_seq const *seq); + struct tipc_port_list *dports); struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, - u32 scope, u32 port_ref, u32 key); + 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 subscription *s); -void tipc_nametbl_unsubscribe(struct subscription *s); + 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); diff --git a/net/tipc/net.c b/net/tipc/net.c index c39c76201e8..f64375e7f99 100644 --- a/net/tipc/net.c +++ b/net/tipc/net.c @@ -2,7 +2,7 @@ * net/tipc/net.c: TIPC network routing code * * Copyright (c) 1995-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -35,56 +35,45 @@ */ #include "core.h" -#include "bearer.h" #include "net.h" -#include "zone.h" -#include "addr.h" -#include "name_table.h" #include "name_distr.h" #include "subscr.h" -#include "link.h" -#include "msg.h" #include "port.h" -#include "bcast.h" -#include "discover.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 three major + * port and node/link instances. The code consists of four major * locking domains, each protected with their own disjunct set of locks. * - * 1: The routing hierarchy. - * Comprises the structures 'zone', 'cluster', 'node', 'link' - * and 'bearer'. The whole hierarchy is protected by a big - * read/write lock, tipc_net_lock, to enssure that nothing is added - * or removed while code is accessing any of these structures. - * This layer must not be called from the two others while they - * hold any of their own locks. - * Neither must it itself do any upcalls to the other two before - * it has released tipc_net_lock and other protective 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. * - * Within the tipc_net_lock domain there are two sub-domains;'node' and - * 'bearer', where local write operations are permitted, - * provided that those are protected by individual spin_locks - * per instance. Code holding tipc_net_lock(read) and a node spin_lock - * is permitted to poke around in both the node itself and its - * subordinate links. I.e, it can update link counters and queues, - * change link state, send protocol messages, and alter the - * "active_links" array in the node; but it can _not_ remove a link - * or a node from the overall structure. - * Correspondingly, individual bearers may change status within a - * tipc_net_lock(read), protected by an individual spin_lock ber bearer - * instance, but it needs tipc_net_lock(write) to remove/add any bearers. + * 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. * - * 2: 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). + * 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: + * 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 @@ -103,7 +92,7 @@ * There are two such lists; 'port_list', which is used for management, * and 'wait_list', which is used to queue ports during congestion. * - * 3: The name table (name_table.c, name_distr.c, subscription.c) + * 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. @@ -115,73 +104,6 @@ * - A local spin_lock protecting the queue of subscriber events. */ -DEFINE_RWLOCK(tipc_net_lock); -struct network tipc_net = { NULL }; - -struct node *tipc_net_select_remote_node(u32 addr, u32 ref) -{ - return tipc_zone_select_remote_node(tipc_net.zones[tipc_zone(addr)], addr, ref); -} - -u32 tipc_net_select_router(u32 addr, u32 ref) -{ - return tipc_zone_select_router(tipc_net.zones[tipc_zone(addr)], addr, ref); -} - -#if 0 -u32 tipc_net_next_node(u32 a) -{ - if (tipc_net.zones[tipc_zone(a)]) - return tipc_zone_next_node(a); - return 0; -} -#endif - -void tipc_net_remove_as_router(u32 router) -{ - u32 z_num; - - for (z_num = 1; z_num <= tipc_max_zones; z_num++) { - if (!tipc_net.zones[z_num]) - continue; - tipc_zone_remove_as_router(tipc_net.zones[z_num], router); - } -} - -void tipc_net_send_external_routes(u32 dest) -{ - u32 z_num; - - for (z_num = 1; z_num <= tipc_max_zones; z_num++) { - if (tipc_net.zones[z_num]) - tipc_zone_send_external_routes(tipc_net.zones[z_num], dest); - } -} - -static int net_init(void) -{ - memset(&tipc_net, 0, sizeof(tipc_net)); - tipc_net.zones = kcalloc(tipc_max_zones + 1, sizeof(struct _zone *), GFP_ATOMIC); - if (!tipc_net.zones) { - return -ENOMEM; - } - return TIPC_OK; -} - -static void net_stop(void) -{ - u32 z_num; - - if (!tipc_net.zones) - return; - - for (z_num = 1; z_num <= tipc_max_zones; z_num++) { - tipc_zone_delete(tipc_net.zones[z_num]); - } - kfree(tipc_net.zones); - tipc_net.zones = NULL; -} - static void net_route_named_msg(struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); @@ -189,22 +111,18 @@ static void net_route_named_msg(struct sk_buff *buf) u32 dport; if (!msg_named(msg)) { - msg_dbg(msg, "tipc_net->drop_nam:"); - buf_discard(buf); + kfree_skb(buf); return; } dnode = addr_domain(msg_lookup_scope(msg)); dport = tipc_nametbl_translate(msg_nametype(msg), msg_nameinst(msg), &dnode); - dbg("tipc_net->lookup<%u,%u>-><%u,%x>\n", - msg_nametype(msg), msg_nameinst(msg), dport, dnode); if (dport) { msg_set_destnode(msg, dnode); msg_set_destport(msg, dport); tipc_net_route_msg(buf); return; } - msg_dbg(msg, "tipc_net->rej:NO NAME: "); tipc_reject_msg(buf, TIPC_ERR_NO_NAME); } @@ -217,93 +135,68 @@ void tipc_net_route_msg(struct sk_buff *buf) return; msg = buf_msg(buf); - msg_incr_reroute_cnt(msg); - if (msg_reroute_cnt(msg) > 6) { - if (msg_errcode(msg)) { - msg_dbg(msg, "NET>DISC>:"); - buf_discard(buf); - } else { - msg_dbg(msg, "NET>REJ>:"); - tipc_reject_msg(buf, msg_destport(msg) ? - TIPC_ERR_NO_PORT : TIPC_ERR_NO_NAME); - } - return; - } - - msg_dbg(msg, "tipc_net->rout: "); - /* Handle message for this node */ dnode = msg_short(msg) ? tipc_own_addr : msg_destnode(msg); - if (in_scope(dnode, tipc_own_addr)) { + if (tipc_in_scope(dnode, tipc_own_addr)) { if (msg_isdata(msg)) { if (msg_mcast(msg)) - tipc_port_recv_mcast(buf, NULL); + tipc_port_mcast_rcv(buf, NULL); else if (msg_destport(msg)) - tipc_port_recv_msg(buf); + tipc_sk_rcv(buf); else net_route_named_msg(buf); return; } switch (msg_user(msg)) { - case ROUTE_DISTRIBUTOR: - tipc_cltr_recv_routing_table(buf); - break; case NAME_DISTRIBUTOR: - tipc_named_recv(buf); + tipc_named_rcv(buf); break; case CONN_MANAGER: - tipc_port_recv_proto_msg(buf); + tipc_port_proto_rcv(buf); break; default: - msg_dbg(msg,"DROP/NET/<REC<"); - buf_discard(buf); + kfree_skb(buf); } return; } /* Handle message for another node */ - msg_dbg(msg, "NET>SEND>: "); - tipc_link_send(buf, dnode, msg_link_selector(msg)); + skb_trim(buf, msg_size(msg)); + tipc_link_xmit(buf, dnode, msg_link_selector(msg)); } -int tipc_net_start(void) +int tipc_net_start(u32 addr) { char addr_string[16]; int res; - if (tipc_mode != TIPC_NODE_MODE) - return -ENOPROTOOPT; - - tipc_mode = TIPC_NET_MODE; + tipc_own_addr = addr; tipc_named_reinit(); tipc_port_reinit(); - - if ((res = tipc_bearer_init()) || - (res = net_init()) || - (res = tipc_cltr_init()) || - (res = tipc_bclink_init())) { + res = tipc_bclink_init(); + if (res) return res; - } - tipc_subscr_stop(); - tipc_cfg_stop(); - tipc_k_signal((Handler)tipc_subscr_start, 0); - tipc_k_signal((Handler)tipc_cfg_init, 0); - info("Started in network mode\n"); - info("Own node address %s, network identity %u\n", - addr_string_fill(addr_string, tipc_own_addr), tipc_net_id); - return TIPC_OK; + + 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_mode != TIPC_NET_MODE) + if (!tipc_own_addr) return; - write_lock_bh(&tipc_net_lock); + + tipc_nametbl_withdraw(TIPC_CFG_SRV, tipc_own_addr, 0, tipc_own_addr); + rtnl_lock(); tipc_bearer_stop(); - tipc_mode = TIPC_NODE_MODE; tipc_bclink_stop(); - net_stop(); - write_unlock_bh(&tipc_net_lock); - info("Left network mode \n"); -} + tipc_node_stop(); + rtnl_unlock(); + pr_info("Left network mode\n"); +} diff --git a/net/tipc/net.h b/net/tipc/net.h index a6a0e9976ac..c6c2b46f7c2 100644 --- a/net/tipc/net.h +++ b/net/tipc/net.h @@ -2,7 +2,7 @@ * net/tipc/net.h: Include file for TIPC network routing code * * Copyright (c) 1995-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -37,28 +37,9 @@ #ifndef _TIPC_NET_H #define _TIPC_NET_H -struct _zone; - -/** - * struct network - TIPC network structure - * @zones: array of pointers to all zones within network - */ - -struct network { - struct _zone **zones; -}; - - -extern struct network tipc_net; -extern rwlock_t tipc_net_lock; - -void tipc_net_remove_as_router(u32 router); -void tipc_net_send_external_routes(u32 dest); void tipc_net_route_msg(struct sk_buff *buf); -struct node *tipc_net_select_remote_node(u32 addr, u32 ref); -u32 tipc_net_select_router(u32 addr, u32 ref); -int tipc_net_start(void); +int tipc_net_start(u32 addr); void tipc_net_stop(void); #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 6a7f7b4c259..ad844d36534 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -2,7 +2,7 @@ * net/tipc/netlink.c: TIPC configuration handling * * Copyright (c) 2005-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005-2007, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -44,29 +44,31 @@ static int handle_cmd(struct sk_buff *skb, struct genl_info *info) struct nlmsghdr *rep_nlh; struct nlmsghdr *req_nlh = info->nlhdr; struct tipc_genlmsghdr *req_userhdr = info->userhdr; - int hdr_space = NLMSG_SPACE(GENL_HDRLEN + TIPC_GENL_HDRLEN); + int hdr_space = nlmsg_total_size(GENL_HDRLEN + TIPC_GENL_HDRLEN); + u16 cmd; - if ((req_userhdr->cmd & 0xC000) && (!capable(CAP_NET_ADMIN))) - rep_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN); + if ((req_userhdr->cmd & 0xC000) && (!netlink_capable(skb, CAP_NET_ADMIN))) + cmd = TIPC_CMD_NOT_NET_ADMIN; else - rep_buf = tipc_cfg_do_cmd(req_userhdr->dest, - req_userhdr->cmd, - NLMSG_DATA(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN, - NLMSG_PAYLOAD(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN), - hdr_space); + 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(rep_buf, NETLINK_CB(skb).pid); + genlmsg_unicast(&init_net, rep_buf, NETLINK_CB(skb).portid); } return 0; } -static struct genl_family family = { +static struct genl_family tipc_genl_family = { .id = GENL_ID_GENERATE, .name = TIPC_GENL_NAME, .version = TIPC_GENL_VERSION, @@ -74,39 +76,26 @@ static struct genl_family family = { .maxattr = 0, }; -static struct genl_ops ops = { - .cmd = TIPC_GENL_CMD, - .doit = handle_cmd, +static struct genl_ops tipc_genl_ops[] = { + { + .cmd = TIPC_GENL_CMD, + .doit = handle_cmd, + }, }; -static int family_registered = 0; - int tipc_netlink_start(void) { + int res; - - if (genl_register_family(&family)) - goto err; - - family_registered = 1; - - if (genl_register_ops(&family, &ops)) - goto err_unregister; - + 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; - - err_unregister: - genl_unregister_family(&family); - family_registered = 0; - err: - err("Failed to register netlink interface\n"); - return -EFAULT; } void tipc_netlink_stop(void) { - if (family_registered) { - genl_unregister_family(&family); - family_registered = 0; - } + genl_unregister_family(&tipc_genl_family); } diff --git a/net/tipc/node.c b/net/tipc/node.c index 598f4d3a009..5b44c3041be 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1,8 +1,8 @@ /* * net/tipc/node.c: TIPC node management routines * - * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005-2006, Wind River Systems + * 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 @@ -37,133 +37,154 @@ #include "core.h" #include "config.h" #include "node.h" -#include "cluster.h" -#include "net.h" -#include "addr.h" -#include "node_subscr.h" -#include "link.h" -#include "port.h" -#include "bearer.h" #include "name_distr.h" -void node_print(struct print_buf *buf, struct node *n_ptr, char *str); -static void node_lost_contact(struct node *n_ptr); -static void node_established_contact(struct node *n_ptr); +#define NODE_HTABLE_SIZE 512 -struct node *tipc_nodes = NULL; /* sorted list of nodes within cluster */ +static void node_lost_contact(struct tipc_node *n_ptr); +static void node_established_contact(struct tipc_node *n_ptr); -u32 tipc_own_tag = 0; +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); -struct node *tipc_node_create(u32 addr) +/* + * 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) { - struct cluster *c_ptr; - struct node *n_ptr; - struct node **curr_node; + return addr & (NODE_HTABLE_SIZE - 1); +} - n_ptr = kzalloc(sizeof(*n_ptr),GFP_ATOMIC); - if (!n_ptr) { - warn("Node creation failed, no memory\n"); +/* + * 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; - } - c_ptr = tipc_cltr_find(addr); - if (!c_ptr) { - c_ptr = tipc_cltr_create(addr); + rcu_read_lock(); + hlist_for_each_entry_rcu(node, &node_htable[tipc_hashfn(addr)], hash) { + if (node->addr == addr) { + rcu_read_unlock(); + return node; + } } - if (!c_ptr) { - kfree(n_ptr); + 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); + spin_lock_init(&n_ptr->lock); + INIT_HLIST_NODE(&n_ptr->hash); + INIT_LIST_HEAD(&n_ptr->list); INIT_LIST_HEAD(&n_ptr->nsub); - n_ptr->owner = c_ptr; - tipc_cltr_attach_node(c_ptr, n_ptr); - n_ptr->last_router = -1; - - /* Insert node into ordered list */ - for (curr_node = &tipc_nodes; *curr_node; - curr_node = &(*curr_node)->next) { - if (addr < (*curr_node)->addr) { - n_ptr->next = *curr_node; + + 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; - } } - (*curr_node) = n_ptr; + 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; } -void tipc_node_delete(struct node *n_ptr) +static void tipc_node_delete(struct tipc_node *n_ptr) { - if (!n_ptr) - return; - -#if 0 - /* Not needed because links are already deleted via tipc_bearer_stop() */ + list_del_rcu(&n_ptr->list); + hlist_del_rcu(&n_ptr->hash); + kfree_rcu(n_ptr, rcu); - u32 l_num; + tipc_num_nodes--; +} - for (l_num = 0; l_num < MAX_BEARERS; l_num++) { - link_delete(n_ptr->links[l_num]); - } -#endif +void tipc_node_stop(void) +{ + struct tipc_node *node, *t_node; - dbg("node %x deleted\n", n_ptr->addr); - kfree(n_ptr); + 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 node *n_ptr, struct link *l_ptr) +void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { - struct link **active = &n_ptr->active_links[0]; + struct tipc_link **active = &n_ptr->active_links[0]; + u32 addr = n_ptr->addr; n_ptr->working_links++; - - info("Established link <%s> on network plane %c\n", - l_ptr->name, l_ptr->b_ptr->net_plane); + 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]) { - dbg(" link %x into %x/%x\n", l_ptr, &active[0], &active[1]); active[0] = active[1] = l_ptr; node_established_contact(n_ptr); return; } if (l_ptr->priority < active[0]->priority) { - info("New link <%s> becomes standby\n", l_ptr->name); + pr_info("New link <%s> becomes standby\n", l_ptr->name); return; } - tipc_link_send_duplicate(active[0], l_ptr); + tipc_link_dup_queue_xmit(active[0], l_ptr); if (l_ptr->priority == active[0]->priority) { active[0] = l_ptr; return; } - info("Old link <%s> becomes standby\n", active[0]->name); + pr_info("Old link <%s> becomes standby\n", active[0]->name); if (active[1] != active[0]) - info("Old link <%s> becomes standby\n", active[1]->name); + 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 node *n_ptr) +static void node_select_active_links(struct tipc_node *n_ptr) { - struct link **active = &n_ptr->active_links[0]; + 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 link *l_ptr = n_ptr->links[i]; + struct tipc_link *l_ptr = n_ptr->links[i]; if (!l_ptr || !tipc_link_is_up(l_ptr) || (l_ptr->priority < highest_prio)) @@ -181,20 +202,21 @@ static void node_select_active_links(struct node *n_ptr) /** * tipc_node_link_down - handle loss of link */ - -void tipc_node_link_down(struct node *n_ptr, struct link *l_ptr) +void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { - struct link **active; + 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)) { - info("Lost standby link <%s> on network plane %c\n", - l_ptr->name, l_ptr->b_ptr->net_plane); + pr_info("Lost standby link <%s> on network plane %c\n", + l_ptr->name, l_ptr->net_plane); return; } - info("Lost link <%s> on network plane %c\n", - l_ptr->name, l_ptr->b_ptr->net_plane); + 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) @@ -204,220 +226,80 @@ void tipc_node_link_down(struct node *n_ptr, struct link *l_ptr) if (active[0] == l_ptr) node_select_active_links(n_ptr); if (tipc_node_is_up(n_ptr)) - tipc_link_changeover(l_ptr); + tipc_link_failover_send_queue(l_ptr); else node_lost_contact(n_ptr); } -int tipc_node_has_active_links(struct node *n_ptr) +int tipc_node_active_links(struct tipc_node *n_ptr) { - return (n_ptr && - ((n_ptr->active_links[0]) || (n_ptr->active_links[1]))); + return n_ptr->active_links[0] != NULL; } -int tipc_node_has_redundant_links(struct node *n_ptr) +int tipc_node_is_up(struct tipc_node *n_ptr) { - return (n_ptr->working_links > 1); + return tipc_node_active_links(n_ptr); } -static int tipc_node_has_active_routes(struct node *n_ptr) +void tipc_node_attach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { - return (n_ptr && (n_ptr->last_router >= 0)); + 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++; } -int tipc_node_is_up(struct node *n_ptr) +void tipc_node_detach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { - return (tipc_node_has_active_links(n_ptr) || tipc_node_has_active_routes(n_ptr)); -} - -struct node *tipc_node_attach_link(struct link *l_ptr) -{ - struct node *n_ptr = tipc_node_find(l_ptr->addr); - - if (!n_ptr) - n_ptr = tipc_node_create(l_ptr->addr); - if (n_ptr) { - u32 bearer_id = l_ptr->b_ptr->identity; - char addr_string[16]; - - if (n_ptr->link_cnt >= 2) { - err("Attempt to create third link to %s\n", - addr_string_fill(addr_string, n_ptr->addr)); - return NULL; - } + int i; - if (!n_ptr->links[bearer_id]) { - n_ptr->links[bearer_id] = l_ptr; - tipc_net.zones[tipc_zone(l_ptr->addr)]->links++; - n_ptr->link_cnt++; - return n_ptr; - } - err("Attempt to establish second link on <%s> to %s \n", - l_ptr->b_ptr->publ.name, - addr_string_fill(addr_string, l_ptr->addr)); + 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--; } - return NULL; -} - -void tipc_node_detach_link(struct node *n_ptr, struct link *l_ptr) -{ - n_ptr->links[l_ptr->b_ptr->identity] = NULL; - tipc_net.zones[tipc_zone(l_ptr->addr)]->links--; - n_ptr->link_cnt--; } -/* - * Routing table management - five cases to handle: - * - * 1: A link towards a zone/cluster external node comes up. - * => Send a multicast message updating routing tables of all - * system nodes within own cluster that the new destination - * can be reached via this node. - * (node.establishedContact()=>cluster.multicastNewRoute()) - * - * 2: A link towards a slave node comes up. - * => Send a multicast message updating routing tables of all - * system nodes within own cluster that the new destination - * can be reached via this node. - * (node.establishedContact()=>cluster.multicastNewRoute()) - * => Send a message to the slave node about existence - * of all system nodes within cluster: - * (node.establishedContact()=>cluster.sendLocalRoutes()) - * - * 3: A new cluster local system node becomes available. - * => Send message(s) to this particular node containing - * information about all cluster external and slave - * nodes which can be reached via this node. - * (node.establishedContact()==>network.sendExternalRoutes()) - * (node.establishedContact()==>network.sendSlaveRoutes()) - * => Send messages to all directly connected slave nodes - * containing information about the existence of the new node - * (node.establishedContact()=>cluster.multicastNewRoute()) - * - * 4: The link towards a zone/cluster external node or slave - * node goes down. - * => Send a multcast message updating routing tables of all - * nodes within cluster that the new destination can not any - * longer be reached via this node. - * (node.lostAllLinks()=>cluster.bcastLostRoute()) - * - * 5: A cluster local system node becomes unavailable. - * => Remove all references to this node from the local - * routing tables. Note: This is a completely node - * local operation. - * (node.lostAllLinks()=>network.removeAsRouter()) - * => Send messages to all directly connected slave nodes - * containing information about loss of the node - * (node.establishedContact()=>cluster.multicastLostRoute()) - * - */ - -static void node_established_contact(struct node *n_ptr) +static void node_established_contact(struct tipc_node *n_ptr) { - struct cluster *c_ptr; - - dbg("node_established_contact:-> %x\n", n_ptr->addr); - if (!tipc_node_has_active_routes(n_ptr) && in_own_cluster(n_ptr->addr)) { - tipc_k_signal((Handler)tipc_named_node_up, n_ptr->addr); - } - - /* Syncronize broadcast acks */ + n_ptr->action_flags |= TIPC_NOTIFY_NODE_UP; + n_ptr->bclink.oos_state = 0; n_ptr->bclink.acked = tipc_bclink_get_last_sent(); - - if (is_slave(tipc_own_addr)) - return; - if (!in_own_cluster(n_ptr->addr)) { - /* Usage case 1 (see above) */ - c_ptr = tipc_cltr_find(tipc_own_addr); - if (!c_ptr) - c_ptr = tipc_cltr_create(tipc_own_addr); - if (c_ptr) - tipc_cltr_bcast_new_route(c_ptr, n_ptr->addr, 1, - tipc_max_nodes); - return; - } - - c_ptr = n_ptr->owner; - if (is_slave(n_ptr->addr)) { - /* Usage case 2 (see above) */ - tipc_cltr_bcast_new_route(c_ptr, n_ptr->addr, 1, tipc_max_nodes); - tipc_cltr_send_local_routes(c_ptr, n_ptr->addr); - return; - } - - if (n_ptr->bclink.supported) { - tipc_nmap_add(&tipc_cltr_bcast_nodes, n_ptr->addr); - if (n_ptr->addr < tipc_own_addr) - tipc_own_tag++; - } - - /* Case 3 (see above) */ - tipc_net_send_external_routes(n_ptr->addr); - tipc_cltr_send_slave_routes(c_ptr, n_ptr->addr); - tipc_cltr_bcast_new_route(c_ptr, n_ptr->addr, LOWEST_SLAVE, - tipc_highest_allowed_slave); + tipc_bclink_add_node(n_ptr->addr); } -static void node_lost_contact(struct node *n_ptr) +static void node_lost_contact(struct tipc_node *n_ptr) { - struct cluster *c_ptr; - struct node_subscr *ns, *tns; char addr_string[16]; u32 i; - /* Clean up broadcast reception remains */ - n_ptr->bclink.gap_after = n_ptr->bclink.gap_to = 0; - while (n_ptr->bclink.deferred_head) { - struct sk_buff* buf = n_ptr->bclink.deferred_head; - n_ptr->bclink.deferred_head = buf->next; - buf_discard(buf); - } - if (n_ptr->bclink.defragm) { - buf_discard(n_ptr->bclink.defragm); - n_ptr->bclink.defragm = NULL; - } - if (in_own_cluster(n_ptr->addr) && n_ptr->bclink.supported) { - tipc_bclink_acknowledge(n_ptr, mod(n_ptr->bclink.acked + 10000)); - } + pr_info("Lost contact with %s\n", + tipc_addr_string_fill(addr_string, n_ptr->addr)); - /* Update routing tables */ - if (is_slave(tipc_own_addr)) { - tipc_net_remove_as_router(n_ptr->addr); - } else { - if (!in_own_cluster(n_ptr->addr)) { - /* Case 4 (see above) */ - c_ptr = tipc_cltr_find(tipc_own_addr); - tipc_cltr_bcast_lost_route(c_ptr, n_ptr->addr, 1, - tipc_max_nodes); - } else { - /* Case 5 (see above) */ - c_ptr = tipc_cltr_find(n_ptr->addr); - if (is_slave(n_ptr->addr)) { - tipc_cltr_bcast_lost_route(c_ptr, n_ptr->addr, 1, - tipc_max_nodes); - } else { - if (n_ptr->bclink.supported) { - tipc_nmap_remove(&tipc_cltr_bcast_nodes, - n_ptr->addr); - if (n_ptr->addr < tipc_own_addr) - tipc_own_tag--; - } - tipc_net_remove_as_router(n_ptr->addr); - tipc_cltr_bcast_lost_route(c_ptr, n_ptr->addr, - LOWEST_SLAVE, - tipc_highest_allowed_slave); - } + /* 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; } - } - if (tipc_node_has_active_routes(n_ptr)) - return; - info("Lost contact with %s\n", - addr_string_fill(addr_string, n_ptr->addr)); + 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 link *l_ptr = n_ptr->links[i]; + struct tipc_link *l_ptr = n_ptr->links[i]; if (!l_ptr) continue; l_ptr->reset_checkpoint = l_ptr->next_in_no; @@ -425,169 +307,20 @@ static void node_lost_contact(struct node *n_ptr) tipc_link_reset_fragments(l_ptr); } - /* Notify subscribers */ - list_for_each_entry_safe(ns, tns, &n_ptr->nsub, nodesub_list) { - ns->node = NULL; - list_del_init(&ns->nodesub_list); - tipc_k_signal((Handler)ns->handle_node_down, - (unsigned long)ns->usr_handle); - } -} - -/** - * tipc_node_select_next_hop - find the next-hop node for a message - * - * Called by when cluster local lookup has failed. - */ - -struct node *tipc_node_select_next_hop(u32 addr, u32 selector) -{ - struct node *n_ptr; - u32 router_addr; + n_ptr->action_flags &= ~TIPC_WAIT_OWN_LINKS_DOWN; - if (!tipc_addr_domain_valid(addr)) - return NULL; - - /* Look for direct link to destination processsor */ - n_ptr = tipc_node_find(addr); - if (n_ptr && tipc_node_has_active_links(n_ptr)) - return n_ptr; - - /* Cluster local system nodes *must* have direct links */ - if (!is_slave(addr) && in_own_cluster(addr)) - return NULL; - - /* Look for cluster local router with direct link to node */ - router_addr = tipc_node_select_router(n_ptr, selector); - if (router_addr) - return tipc_node_select(router_addr, selector); - - /* Slave nodes can only be accessed within own cluster via a - known router with direct link -- if no router was found,give up */ - if (is_slave(addr)) - return NULL; - - /* Inter zone/cluster -- find any direct link to remote cluster */ - addr = tipc_addr(tipc_zone(addr), tipc_cluster(addr), 0); - n_ptr = tipc_net_select_remote_node(addr, selector); - if (n_ptr && tipc_node_has_active_links(n_ptr)) - return n_ptr; - - /* Last resort -- look for any router to anywhere in remote zone */ - router_addr = tipc_net_select_router(addr, selector); - if (router_addr) - return tipc_node_select(router_addr, selector); - - return NULL; -} - -/** - * tipc_node_select_router - select router to reach specified node - * - * Uses a deterministic and fair algorithm for selecting router node. - */ - -u32 tipc_node_select_router(struct node *n_ptr, u32 ref) -{ - u32 ulim; - u32 mask; - u32 start; - u32 r; - - if (!n_ptr) - return 0; - - if (n_ptr->last_router < 0) - return 0; - ulim = ((n_ptr->last_router + 1) * 32) - 1; - - /* Start entry must be random */ - mask = tipc_max_nodes; - while (mask > ulim) - mask >>= 1; - start = ref & mask; - r = start; - - /* Lookup upwards with wrap-around */ - do { - if (((n_ptr->routers[r / 32]) >> (r % 32)) & 1) - break; - } while (++r <= ulim); - if (r > ulim) { - r = 1; - do { - if (((n_ptr->routers[r / 32]) >> (r % 32)) & 1) - break; - } while (++r < start); - assert(r != start); - } - assert(r && (r <= ulim)); - return tipc_addr(own_zone(), own_cluster(), r); -} - -void tipc_node_add_router(struct node *n_ptr, u32 router) -{ - u32 r_num = tipc_node(router); - - n_ptr->routers[r_num / 32] = - ((1 << (r_num % 32)) | n_ptr->routers[r_num / 32]); - n_ptr->last_router = tipc_max_nodes / 32; - while ((--n_ptr->last_router >= 0) && - !n_ptr->routers[n_ptr->last_router]); -} - -void tipc_node_remove_router(struct node *n_ptr, u32 router) -{ - u32 r_num = tipc_node(router); - - if (n_ptr->last_router < 0) - return; /* No routes */ - - n_ptr->routers[r_num / 32] = - ((~(1 << (r_num % 32))) & (n_ptr->routers[r_num / 32])); - n_ptr->last_router = tipc_max_nodes / 32; - while ((--n_ptr->last_router >= 0) && - !n_ptr->routers[n_ptr->last_router]); - - if (!tipc_node_is_up(n_ptr)) - node_lost_contact(n_ptr); -} - -#if 0 -void node_print(struct print_buf *buf, struct node *n_ptr, char *str) -{ - u32 i; - - tipc_printf(buf, "\n\n%s", str); - for (i = 0; i < MAX_BEARERS; i++) { - if (!n_ptr->links[i]) - continue; - tipc_printf(buf, "Links[%u]: %x, ", i, n_ptr->links[i]); - } - tipc_printf(buf, "Active links: [%x,%x]\n", - n_ptr->active_links[0], n_ptr->active_links[1]); -} -#endif - -u32 tipc_available_nodes(const u32 domain) -{ - struct node *n_ptr; - u32 cnt = 0; - - for (n_ptr = tipc_nodes; n_ptr; n_ptr = n_ptr->next) { - if (!in_scope(domain, n_ptr->addr)) - continue; - if (tipc_node_is_up(n_ptr)) - cnt++; - } - return cnt; + /* 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 node *n_ptr; + struct tipc_node *n_ptr; struct tipc_node_info node_info; u32 payload_size; @@ -599,31 +332,36 @@ struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (network address)"); - if (!tipc_nodes) + 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 - (will need to modify this when slave nodes are supported */ - - payload_size = TLV_SPACE(sizeof(node_info)) * (tipc_max_nodes - 1); - if (payload_size > 32768u) + /* 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 */ - - for (n_ptr = tipc_nodes; n_ptr; n_ptr = n_ptr->next) { - if (!in_scope(domain, n_ptr->addr)) + 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; } @@ -631,7 +369,7 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) { u32 domain; struct sk_buff *buf; - struct node *n_ptr; + struct tipc_node *n_ptr; struct tipc_link_info link_info; u32 payload_size; @@ -643,34 +381,37 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (network address)"); - if (tipc_mode != TIPC_NET_MODE) + if (!tipc_own_addr) return tipc_cfg_reply_none(); - /* Get space for all unicast links + multicast link */ - - payload_size = TLV_SPACE(sizeof(link_info)) * - (tipc_net.zones[tipc_zone(tipc_own_addr)]->links + 1); - if (payload_size > 32768u) + 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_own_addr & 0xfffff00); + link_info.dest = htonl(tipc_cluster_mask(tipc_own_addr)); link_info.up = htonl(1); - sprintf(link_info.str, tipc_bclink_name); + 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 */ - - for (n_ptr = tipc_nodes; n_ptr; n_ptr = n_ptr->next) { + rcu_read_lock(); + list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { u32 i; - if (!in_scope(domain, n_ptr->addr)) + 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; @@ -680,7 +421,68 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) 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 index cd1882654bb..9087063793f 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -2,7 +2,7 @@ * net/tipc/node.h: Include file for TIPC node management routines * * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005, 2010-2014, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -39,108 +39,108 @@ #include "node_subscr.h" #include "addr.h" -#include "cluster.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 node - TIPC node structure + * 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 - * @owner: pointer to cluster that node belongs to - * @next: pointer to next node in sorted list of cluster's nodes - * @nsub: list of "node down" subscriptions monitoring node + * @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 - * @permit_changeover: non-zero if node has redundant links to this system - * @routers: bitmap (used for multicluster communication) - * @last_router: (used for multicluster communication) - * @bclink: broadcast-related info - * @supported: non-zero if node supports TIPC b'cast capability - * @acked: sequence # of last outbound b'cast message acknowledged by node - * @last_in: sequence # of last in-sequence b'cast message received from node - * @gap_after: sequence # of last message not requiring a NAK request - * @gap_to: sequence # of last message requiring a NAK request - * @nack_sync: counter that determines when NAK requests should be sent - * @deferred_head: oldest OOS b'cast message received from node - * @deferred_tail: newest OOS b'cast message received from node - * @defragm: list of partially reassembled b'cast message fragments from node + * @signature: node instance identifier + * @nsub: list of "node down" subscriptions monitoring node + * @rcu: rcu struct for tipc_node */ - -struct node { +struct tipc_node { u32 addr; spinlock_t lock; - struct cluster *owner; - struct node *next; - struct list_head nsub; - struct link *active_links[2]; - struct link *links[MAX_BEARERS]; + 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; - int permit_changeover; - u32 routers[512/32]; - int last_router; - struct { - int supported; - u32 acked; - u32 last_in; - u32 gap_after; - u32 gap_to; - u32 nack_sync; - struct sk_buff *deferred_head; - struct sk_buff *deferred_tail; - struct sk_buff *defragm; - } bclink; + u32 signature; + struct list_head nsub; + struct rcu_head rcu; }; -extern struct node *tipc_nodes; -extern u32 tipc_own_tag; +extern struct list_head tipc_node_list; -struct node *tipc_node_create(u32 addr); -void tipc_node_delete(struct node *n_ptr); -struct node *tipc_node_attach_link(struct link *l_ptr); -void tipc_node_detach_link(struct node *n_ptr, struct link *l_ptr); -void tipc_node_link_down(struct node *n_ptr, struct link *l_ptr); -void tipc_node_link_up(struct node *n_ptr, struct link *l_ptr); -int tipc_node_has_active_links(struct node *n_ptr); -int tipc_node_has_redundant_links(struct node *n_ptr); -u32 tipc_node_select_router(struct node *n_ptr, u32 ref); -struct node *tipc_node_select_next_hop(u32 addr, u32 selector); -int tipc_node_is_up(struct node *n_ptr); -void tipc_node_add_router(struct node *n_ptr, u32 router); -void tipc_node_remove_router(struct node *n_ptr, u32 router); +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 struct node *tipc_node_find(u32 addr) -{ - if (likely(in_own_cluster(addr))) - return tipc_local_nodes[tipc_node(addr)]; - else if (tipc_addr_domain_valid(addr)) { - struct cluster *c_ptr = tipc_cltr_find(addr); - - if (c_ptr) - return c_ptr->nodes[tipc_node(addr)]; - } - return NULL; -} - -static inline struct node *tipc_node_select(u32 addr, u32 selector) -{ - if (likely(in_own_cluster(addr))) - return tipc_local_nodes[tipc_node(addr)]; - return tipc_node_select_next_hop(addr, selector); -} - -static inline void tipc_node_lock(struct node *n_ptr) +static inline void tipc_node_lock(struct tipc_node *node) { - spin_lock_bh(&n_ptr->lock); + spin_lock_bh(&node->lock); } -static inline void tipc_node_unlock(struct node *n_ptr) +static inline bool tipc_node_blocked(struct tipc_node *node) { - spin_unlock_bh(&n_ptr->lock); + 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 index 8ecbd0fb610..7c59ab1d6ec 100644 --- a/net/tipc/node_subscr.c +++ b/net/tipc/node_subscr.c @@ -2,7 +2,7 @@ * net/tipc/node_subscr.c: TIPC "node down" subscription handling * * Copyright (c) 1995-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -35,26 +35,24 @@ */ #include "core.h" -#include "dbg.h" #include "node_subscr.h" #include "node.h" -#include "addr.h" /** * tipc_nodesub_subscribe - create "node down" subscription for specified node */ - -void tipc_nodesub_subscribe(struct node_subscr *node_sub, u32 addr, - void *usr_handle, net_ev_handler handle_down) +void tipc_nodesub_subscribe(struct tipc_node_subscr *node_sub, u32 addr, + void *usr_handle, net_ev_handler handle_down) { - if (addr == tipc_own_addr) { + if (in_own_node(addr)) { node_sub->node = NULL; return; } node_sub->node = tipc_node_find(addr); if (!node_sub->node) { - warn("Node subscription rejected, unknown node 0x%x\n", addr); + pr_warn("Node subscription rejected, unknown node 0x%x\n", + addr); return; } node_sub->handle_node_down = handle_down; @@ -68,8 +66,7 @@ void tipc_nodesub_subscribe(struct node_subscr *node_sub, u32 addr, /** * tipc_nodesub_unsubscribe - cancel "node down" subscription (if any) */ - -void tipc_nodesub_unsubscribe(struct node_subscr *node_sub) +void tipc_nodesub_unsubscribe(struct tipc_node_subscr *node_sub) { if (!node_sub->node) return; @@ -78,3 +75,20 @@ void tipc_nodesub_unsubscribe(struct node_subscr *node_sub) 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 index 5f3f5859b84..d91b8cc81e3 100644 --- a/net/tipc/node_subscr.h +++ b/net/tipc/node_subscr.h @@ -2,7 +2,7 @@ * net/tipc/node_subscr.h: Include file for TIPC "node down" subscription handling * * Copyright (c) 1995-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -42,22 +42,22 @@ typedef void (*net_ev_handler) (void *usr_handle); /** - * struct node_subscr - "node down" subscription entry + * 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 node_subscr { - struct node *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 node_subscr *node_sub, u32 addr, +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 node_subscr *node_sub); +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 index f508614ca59..5fd7acce01e 100644 --- a/net/tipc/port.c +++ b/net/tipc/port.c @@ -1,8 +1,8 @@ /* * net/tipc/port.c: TIPC port code * - * Copyright (c) 1992-2007, Ericsson AB - * Copyright (c) 2004-2007, Wind River Systems + * 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 @@ -36,15 +36,9 @@ #include "core.h" #include "config.h" -#include "dbg.h" #include "port.h" -#include "addr.h" -#include "link.h" -#include "node.h" #include "name_table.h" -#include "user_reg.h" -#include "msg.h" -#include "bcast.h" +#include "socket.h" /* Connection management: */ #define PROBING_INTERVAL 3600000 /* [ms] => 1 h */ @@ -53,101 +47,89 @@ #define MAX_REJECT_SIZE 1024 -static struct sk_buff *msg_queue_head = NULL; -static struct sk_buff *msg_queue_tail = NULL; - DEFINE_SPINLOCK(tipc_port_list_lock); -static DEFINE_SPINLOCK(queue_lock); static LIST_HEAD(ports); static void port_handle_node_down(unsigned long ref); -static struct sk_buff* port_build_self_abort_msg(struct port *,u32 err); -static struct sk_buff* port_build_peer_abort_msg(struct port *,u32 err); +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); - -static u32 port_peernode(struct port *p_ptr) +/** + * 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) { - return msg_destnode(&p_ptr->publ.phdr); -} + u32 peernode; + u32 orignode; -static u32 port_peerport(struct port *p_ptr) -{ - return msg_destport(&p_ptr->publ.phdr); -} + if (msg_origport(msg) != tipc_port_peerport(p_ptr)) + return 0; -static u32 port_out_seqno(struct port *p_ptr) -{ - return msg_transp_seqno(&p_ptr->publ.phdr); -} - -static void port_incr_out_seqno(struct port *p_ptr) -{ - struct tipc_msg *m = &p_ptr->publ.phdr; - - if (likely(!msg_routed(m))) - return; - msg_set_transp_seqno(m, (msg_transp_seqno(m) + 1)); + orignode = msg_orignode(msg); + peernode = tipc_port_peernode(p_ptr); + return (orignode == peernode) || + (!orignode && (peernode == tipc_own_addr)) || + (!peernode && (orignode == tipc_own_addr)); } /** - * tipc_multicast - send a multicast message to local and remote destinations + * tipc_port_mcast_xmit - send a multicast message to local and remote + * destinations */ - -int tipc_multicast(u32 ref, struct tipc_name_seq const *seq, u32 domain, - u32 num_sect, struct iovec const *msg_sect) +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 port_list dports = {0, NULL, }; - struct port *oport = tipc_port_deref(ref); + struct tipc_port_list dports = {0, NULL, }; int ext_targets; int res; - if (unlikely(!oport)) - return -EINVAL; - /* Create multicast message */ - - hdr = &oport->publ.phdr; + 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 = msg_build(hdr, msg_sect, num_sect, MAX_MSG_SIZE, - !oport->user_port, &buf); + 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); - buf_discard(buf); + kfree_skb(buf); return -ENOMEM; } } - res = tipc_bclink_send_msg(buf); - if ((res < 0) && (dports.count != 0)) { - buf_discard(ibuf); - } + res = tipc_bclink_xmit(buf); + if ((res < 0) && (dports.count != 0)) + kfree_skb(ibuf); } else { ibuf = buf; } if (res >= 0) { if (ibuf) - tipc_port_recv_mcast(ibuf, &dports); + tipc_port_mcast_rcv(ibuf, &dports); } else { tipc_port_list_free(&dports); } @@ -155,22 +137,20 @@ int tipc_multicast(u32 ref, struct tipc_name_seq const *seq, u32 domain, } /** - * tipc_port_recv_mcast - deliver multicast message to all destination ports + * 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_recv_mcast(struct sk_buff *buf, struct port_list *dp) +void tipc_port_mcast_rcv(struct sk_buff *buf, struct tipc_port_list *dp) { - struct tipc_msg* msg; - struct port_list dports = {0, NULL, }; - struct port_list *item = 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), @@ -181,11 +161,11 @@ void tipc_port_recv_mcast(struct sk_buff *buf, struct port_list *dp) } /* 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_port_recv_msg(buf); + tipc_sk_rcv(buf); tipc_port_list_free(dp); return; } @@ -194,305 +174,191 @@ void tipc_port_recv_mcast(struct sk_buff *buf, struct port_list *dp) struct sk_buff *b = skb_clone(buf, GFP_ATOMIC); if (b == NULL) { - warn("Unable to deliver multicast message(s)\n"); - msg_dbg(msg, "LOST:"); + pr_warn("Unable to deliver multicast message(s)\n"); goto exit; } - if ((index == 0) && (cnt != 0)) { + if ((index == 0) && (cnt != 0)) item = item->next; - } - msg_set_destport(buf_msg(b),item->ports[index]); - tipc_port_recv_msg(b); + msg_set_destport(buf_msg(b), item->ports[index]); + tipc_sk_rcv(b); } } exit: - buf_discard(buf); + kfree_skb(buf); tipc_port_list_free(dp); } -/** - * tipc_createport_raw - create a native TIPC port + +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 local port reference + * Returns obtained reference if initialization is successful, zero otherwise */ - -u32 tipc_createport_raw(void *usr_handle, - u32 (*dispatcher)(struct tipc_port *, struct sk_buff *), - void (*wakeup)(struct tipc_port *), - const u32 importance) +u32 tipc_port_init(struct tipc_port *p_ptr, + const unsigned int importance) { - struct port *p_ptr; struct tipc_msg *msg; u32 ref; - p_ptr = kzalloc(sizeof(*p_ptr), GFP_ATOMIC); - if (!p_ptr) { - warn("Port creation failed, no memory\n"); - return 0; - } - ref = tipc_ref_acquire(p_ptr, &p_ptr->publ.lock); + ref = tipc_ref_acquire(p_ptr, &p_ptr->lock); if (!ref) { - warn("Port creation failed, reference table exhausted\n"); - kfree(p_ptr); + pr_warn("Port registration failed, ref. table exhausted\n"); return 0; } - tipc_port_lock(ref); - p_ptr->publ.usr_handle = usr_handle; - p_ptr->publ.max_pkt = MAX_PKT_DEFAULT; - p_ptr->publ.ref = ref; - msg = &p_ptr->publ.phdr; - msg_init(msg, DATA_LOW, TIPC_NAMED_MSG, TIPC_OK, LONG_H_SIZE, 0); - msg_set_orignode(msg, tipc_own_addr); - msg_set_prevnode(msg, tipc_own_addr); - msg_set_origport(msg, ref); - msg_set_importance(msg,importance); - p_ptr->last_in_seqno = 41; - p_ptr->sent = 1; + 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); - p_ptr->congested_link = NULL; - p_ptr->dispatcher = dispatcher; - p_ptr->wakeup = wakeup; - p_ptr->user_port = NULL; k_init_timer(&p_ptr->timer, (Handler)port_timeout, ref); - spin_lock_bh(&tipc_port_list_lock); 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); - tipc_port_unlock(p_ptr); return ref; } -int tipc_deleteport(u32 ref) +void tipc_port_destroy(struct tipc_port *p_ptr) { - struct port *p_ptr; struct sk_buff *buf = NULL; - tipc_withdraw(ref, 0, NULL); - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; + tipc_withdraw(p_ptr, 0, NULL); - tipc_ref_discard(ref); - tipc_port_unlock(p_ptr); + 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->publ.connected) { + if (p_ptr->connected) { buf = port_build_peer_abort_msg(p_ptr, TIPC_ERR_NO_PORT); tipc_nodesub_unsubscribe(&p_ptr->subscription); } - if (p_ptr->user_port) { - tipc_reg_remove_port(p_ptr->user_port); - kfree(p_ptr->user_port); - } 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); - kfree(p_ptr); - dbg("Deleted port %u\n", ref); tipc_net_route_msg(buf); - return TIPC_OK; -} - -/** - * tipc_get_port() - return port associated with 'ref' - * - * Note: Port is not locked. - */ - -struct tipc_port *tipc_get_port(const u32 ref) -{ - return (struct tipc_port *)tipc_ref_deref(ref); -} - -/** - * tipc_get_handle - return user handle associated to port 'ref' - */ - -void *tipc_get_handle(const u32 ref) -{ - struct port *p_ptr; - void * handle; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return NULL; - handle = p_ptr->publ.usr_handle; - tipc_port_unlock(p_ptr); - return handle; -} - -static int port_unreliable(struct port *p_ptr) -{ - return msg_src_droppable(&p_ptr->publ.phdr); -} - -int tipc_portunreliable(u32 ref, unsigned int *isunreliable) -{ - struct port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - *isunreliable = port_unreliable(p_ptr); - tipc_port_unlock(p_ptr); - return TIPC_OK; -} - -int tipc_set_portunreliable(u32 ref, unsigned int isunreliable) -{ - struct port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - msg_set_src_droppable(&p_ptr->publ.phdr, (isunreliable != 0)); - tipc_port_unlock(p_ptr); - return TIPC_OK; -} - -static int port_unreturnable(struct port *p_ptr) -{ - return msg_dest_droppable(&p_ptr->publ.phdr); -} - -int tipc_portunreturnable(u32 ref, unsigned int *isunrejectable) -{ - struct port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - *isunrejectable = port_unreturnable(p_ptr); - tipc_port_unlock(p_ptr); - return TIPC_OK; -} - -int tipc_set_portunreturnable(u32 ref, unsigned int isunrejectable) -{ - struct port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - msg_set_dest_droppable(&p_ptr->publ.phdr, (isunrejectable != 0)); - tipc_port_unlock(p_ptr); - return TIPC_OK; } /* - * port_build_proto_msg(): build a port level protocol - * or a connection abortion message. Called with - * tipc_port lock on. + * 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(u32 destport, u32 destnode, - u32 origport, u32 orignode, - u32 usr, u32 type, u32 err, - u32 seqno, u32 ack) +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 = buf_acquire(LONG_H_SIZE); + buf = tipc_buf_acquire(INT_H_SIZE); if (buf) { msg = buf_msg(buf); - msg_init(msg, usr, type, err, LONG_H_SIZE, destnode); - msg_set_destport(msg, destport); - msg_set_origport(msg, origport); - msg_set_destnode(msg, destnode); - msg_set_orignode(msg, orignode); - msg_set_transp_seqno(msg, seqno); + 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); - msg_dbg(msg, "PORT>SEND>:"); } return buf; } -int tipc_set_msg_option(struct tipc_port *tp_ptr, const char *opt, const u32 sz) -{ - msg_expand(&tp_ptr->phdr, msg_destnode(&tp_ptr->phdr)); - msg_set_options(&tp_ptr->phdr, opt, sz); - return TIPC_OK; -} - 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 = msg_importance(msg); + u32 imp; u32 data_sz = msg_data_sz(msg); - - if (data_sz > MAX_REJECT_SIZE) - data_sz = MAX_REJECT_SIZE; - if (msg_connected(msg) && (imp < TIPC_CRITICAL_IMPORTANCE)) - imp++; - msg_dbg(msg, "port->rej: "); + u32 src_node; + u32 rmsg_sz; /* discard rejected message if it shouldn't be returned to sender */ - if (msg_errcode(msg) || msg_dest_droppable(msg)) { - buf_discard(buf); - return data_sz; + 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; - /* construct rejected message */ - if (msg_mcast(msg)) - hdr_sz = MCAST_H_SIZE; - else - hdr_sz = LONG_H_SIZE; - rbuf = buf_acquire(data_sz + hdr_sz); - if (rbuf == NULL) { - buf_discard(buf); - return data_sz; - } rmsg = buf_msg(rbuf); - msg_init(rmsg, imp, msg_type(msg), err, hdr_sz, msg_orignode(msg)); - msg_set_destport(rmsg, msg_origport(msg)); + 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_set_origport(rmsg, msg_destport(msg)); - if (msg_short(msg)) - msg_set_orignode(rmsg, tipc_own_addr); - else - msg_set_orignode(rmsg, msg_destnode(msg)); - msg_set_size(rmsg, data_sz + hdr_sz); - msg_set_nametype(rmsg, msg_nametype(msg)); - msg_set_nameinst(rmsg, msg_nameinst(msg)); - skb_copy_to_linear_data_offset(rbuf, hdr_sz, msg_data(msg), data_sz); + 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 sk_buff *abuf = NULL; - struct port *p_ptr = tipc_port_lock(msg_destport(msg)); + struct tipc_port *p_ptr = tipc_port_lock(msg_destport(msg)); if (p_ptr) { - if (p_ptr->publ.connected) + 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); } - tipc_net_route_msg(abuf); } - /* send rejected message */ - buf_discard(buf); - tipc_net_route_msg(rbuf); + /* 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_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, - struct iovec const *msg_sect, u32 num_sect, - int err) +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 = msg_build(hdr, msg_sect, num_sect, MAX_MSG_SIZE, - !p_ptr->user_port, &buf); + res = tipc_msg_build(hdr, msg_sect, len, MAX_MSG_SIZE, &buf); if (!buf) return res; @@ -501,13 +367,13 @@ int tipc_port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, static void port_timeout(unsigned long ref) { - struct port *p_ptr = tipc_port_lock(ref); + struct tipc_port *p_ptr = tipc_port_lock(ref); struct sk_buff *buf = NULL; if (!p_ptr) return; - if (!p_ptr->publ.connected) { + if (!p_ptr->connected) { tipc_port_unlock(p_ptr); return; } @@ -516,16 +382,7 @@ static void port_timeout(unsigned long ref) if (p_ptr->probing_state == PROBING) { buf = port_build_self_abort_msg(p_ptr, TIPC_ERR_NO_PORT); } else { - buf = port_build_proto_msg(port_peerport(p_ptr), - port_peernode(p_ptr), - p_ptr->publ.ref, - tipc_own_addr, - CONN_MANAGER, - CONN_PROBE, - TIPC_OK, - port_out_seqno(p_ptr), - 0); - port_incr_out_seqno(p_ptr); + 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); } @@ -536,8 +393,8 @@ static void port_timeout(unsigned long ref) static void port_handle_node_down(unsigned long ref) { - struct port *p_ptr = tipc_port_lock(ref); - struct sk_buff* buf = NULL; + struct tipc_port *p_ptr = tipc_port_lock(ref); + struct sk_buff *buf = NULL; if (!p_ptr) return; @@ -547,641 +404,232 @@ static void port_handle_node_down(unsigned long ref) } -static struct sk_buff *port_build_self_abort_msg(struct port *p_ptr, u32 err) +static struct sk_buff *port_build_self_abort_msg(struct tipc_port *p_ptr, u32 err) { - u32 imp = msg_importance(&p_ptr->publ.phdr); + struct sk_buff *buf = port_build_peer_abort_msg(p_ptr, err); - if (!p_ptr->publ.connected) - return NULL; - if (imp < TIPC_CRITICAL_IMPORTANCE) - imp++; - return port_build_proto_msg(p_ptr->publ.ref, - tipc_own_addr, - port_peerport(p_ptr), - port_peernode(p_ptr), - imp, - TIPC_CONN_MSG, - err, - p_ptr->last_in_seqno + 1, - 0); + 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 port *p_ptr, u32 err) +static struct sk_buff *port_build_peer_abort_msg(struct tipc_port *p_ptr, u32 err) { - u32 imp = msg_importance(&p_ptr->publ.phdr); + struct sk_buff *buf; + struct tipc_msg *msg; + u32 imp; - if (!p_ptr->publ.connected) + if (!p_ptr->connected) return NULL; - if (imp < TIPC_CRITICAL_IMPORTANCE) - imp++; - return port_build_proto_msg(port_peerport(p_ptr), - port_peernode(p_ptr), - p_ptr->publ.ref, - tipc_own_addr, - imp, - TIPC_CONN_MSG, - err, - port_out_seqno(p_ptr), - 0); + + 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_recv_proto_msg(struct sk_buff *buf) +void tipc_port_proto_rcv(struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); - struct port *p_ptr = tipc_port_lock(msg_destport(msg)); - u32 err = TIPC_OK; + struct tipc_port *p_ptr; struct sk_buff *r_buf = NULL; - struct sk_buff *abort_buf = NULL; - - msg_dbg(msg, "PORT<RECV<:"); - - if (!p_ptr) { - err = TIPC_ERR_NO_PORT; - } else if (p_ptr->publ.connected) { - if (port_peernode(p_ptr) != msg_orignode(msg)) - err = TIPC_ERR_NO_PORT; - if (port_peerport(p_ptr) != msg_origport(msg)) - err = TIPC_ERR_NO_PORT; - if (!err && msg_routed(msg)) { - u32 seqno = msg_transp_seqno(msg); - u32 myno = ++p_ptr->last_in_seqno; - if (seqno != myno) { - err = TIPC_ERR_NO_PORT; - abort_buf = port_build_self_abort_msg(p_ptr, err); - } - } - if (msg_type(msg) == CONN_ACK) { - int wakeup = tipc_port_congested(p_ptr) && - p_ptr->publ.congested && - p_ptr->wakeup; - p_ptr->acked += msg_msgcnt(msg); - if (tipc_port_congested(p_ptr)) - goto exit; - p_ptr->publ.congested = 0; - if (!wakeup) - goto exit; - p_ptr->wakeup(&p_ptr->publ); - goto exit; + 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)); } - } else if (p_ptr->publ.published) { - err = TIPC_ERR_NO_PORT; - } - if (err) { - r_buf = port_build_proto_msg(msg_origport(msg), - msg_orignode(msg), - msg_destport(msg), - tipc_own_addr, - DATA_HIGH, - TIPC_CONN_MSG, - err, - 0, - 0); + if (p_ptr) + tipc_port_unlock(p_ptr); goto exit; } - /* All is fine */ - if (msg_type(msg) == CONN_PROBE) { - r_buf = port_build_proto_msg(msg_origport(msg), - msg_orignode(msg), - msg_destport(msg), - tipc_own_addr, - CONN_MANAGER, - CONN_PROBE_REPLY, - TIPC_OK, - port_out_seqno(p_ptr), - 0); + /* 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; - port_incr_out_seqno(p_ptr); + tipc_port_unlock(p_ptr); exit: - if (p_ptr) - tipc_port_unlock(p_ptr); tipc_net_route_msg(r_buf); - tipc_net_route_msg(abort_buf); - buf_discard(buf); + kfree_skb(buf); } -static void port_print(struct port *p_ptr, struct print_buf *buf, int full_id) +static int port_print(struct tipc_port *p_ptr, char *buf, int len, int full_id) { struct publication *publ; + int ret; if (full_id) - tipc_printf(buf, "<%u.%u.%u:%u>:", - tipc_zone(tipc_own_addr), tipc_cluster(tipc_own_addr), - tipc_node(tipc_own_addr), p_ptr->publ.ref); + 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 - tipc_printf(buf, "%-10u:", p_ptr->publ.ref); - - if (p_ptr->publ.connected) { - u32 dport = port_peerport(p_ptr); - u32 destnode = port_peernode(p_ptr); - - tipc_printf(buf, " connected to <%u.%u.%u:%u>", - tipc_zone(destnode), tipc_cluster(destnode), - tipc_node(destnode), dport); - if (p_ptr->publ.conn_type != 0) - tipc_printf(buf, " via {%u,%u}", - p_ptr->publ.conn_type, - p_ptr->publ.conn_instance); - } - else if (p_ptr->publ.published) { - tipc_printf(buf, " bound to"); + 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) - tipc_printf(buf, " {%u,%u}", publ->type, - publ->lower); + ret += tipc_snprintf(buf + ret, len - ret, + " {%u,%u}", publ->type, + publ->lower); else - tipc_printf(buf, " {%u,%u,%u}", publ->type, - publ->lower, publ->upper); + ret += tipc_snprintf(buf + ret, len - ret, + " {%u,%u,%u}", publ->type, + publ->lower, publ->upper); } } - tipc_printf(buf, "\n"); + ret += tipc_snprintf(buf + ret, len - ret, "\n"); + return ret; } -#define MAX_PORT_QUERY 32768 - struct sk_buff *tipc_port_get_ports(void) { struct sk_buff *buf; struct tlv_desc *rep_tlv; - struct print_buf pb; - struct port *p_ptr; - int str_len; + char *pb; + int pb_len; + struct tipc_port *p_ptr; + int str_len = 0; - buf = tipc_cfg_reply_alloc(TLV_SPACE(MAX_PORT_QUERY)); + 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; - tipc_printbuf_init(&pb, TLV_DATA(rep_tlv), MAX_PORT_QUERY); spin_lock_bh(&tipc_port_list_lock); list_for_each_entry(p_ptr, &ports, port_list) { - spin_lock_bh(p_ptr->publ.lock); - port_print(p_ptr, &pb, 0); - spin_unlock_bh(p_ptr->publ.lock); + 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 = tipc_printbuf_validate(&pb); - - skb_put(buf, TLV_SPACE(str_len)); - TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); - - return buf; -} - -#if 0 - -#define MAX_PORT_STATS 2000 - -struct sk_buff *port_show_stats(const void *req_tlv_area, int req_tlv_space) -{ - u32 ref; - struct port *p_ptr; - struct sk_buff *buf; - struct tlv_desc *rep_tlv; - struct print_buf pb; - int str_len; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_PORT_REF)) - return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - ref = *(u32 *)TLV_DATA(req_tlv_area); - ref = ntohl(ref); - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return cfg_reply_error_string("port not found"); - - buf = tipc_cfg_reply_alloc(TLV_SPACE(MAX_PORT_STATS)); - if (!buf) { - tipc_port_unlock(p_ptr); - return NULL; - } - rep_tlv = (struct tlv_desc *)buf->data; - - tipc_printbuf_init(&pb, TLV_DATA(rep_tlv), MAX_PORT_STATS); - port_print(p_ptr, &pb, 1); - /* NEED TO FILL IN ADDITIONAL PORT STATISTICS HERE */ - tipc_port_unlock(p_ptr); - str_len = tipc_printbuf_validate(&pb); - + 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; } -#endif - void tipc_port_reinit(void) { - struct port *p_ptr; + 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->publ.phdr; - if (msg_orignode(msg) == tipc_own_addr) - break; + 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); } - -/* - * port_dispatcher_sigh(): Signal handler for messages destinated - * to the tipc_port interface. - */ - -static void port_dispatcher_sigh(void *dummy) -{ - struct sk_buff *buf; - - spin_lock_bh(&queue_lock); - buf = msg_queue_head; - msg_queue_head = NULL; - spin_unlock_bh(&queue_lock); - - while (buf) { - struct port *p_ptr; - struct user_port *up_ptr; - struct tipc_portid orig; - struct tipc_name_seq dseq; - void *usr_handle; - int connected; - int published; - u32 message_type; - - struct sk_buff *next = buf->next; - struct tipc_msg *msg = buf_msg(buf); - u32 dref = msg_destport(msg); - - message_type = msg_type(msg); - if (message_type > TIPC_DIRECT_MSG) - goto reject; /* Unsupported message type */ - - p_ptr = tipc_port_lock(dref); - if (!p_ptr) - goto reject; /* Port deleted while msg in queue */ - - orig.ref = msg_origport(msg); - orig.node = msg_orignode(msg); - up_ptr = p_ptr->user_port; - usr_handle = up_ptr->usr_handle; - connected = p_ptr->publ.connected; - published = p_ptr->publ.published; - - if (unlikely(msg_errcode(msg))) - goto err; - - switch (message_type) { - - case TIPC_CONN_MSG:{ - tipc_conn_msg_event cb = up_ptr->conn_msg_cb; - u32 peer_port = port_peerport(p_ptr); - u32 peer_node = port_peernode(p_ptr); - - tipc_port_unlock(p_ptr); - if (unlikely(!connected)) { - if (unlikely(published)) - goto reject; - tipc_connect2port(dref,&orig); - } - if (unlikely(msg_origport(msg) != peer_port)) - goto reject; - if (unlikely(msg_orignode(msg) != peer_node)) - goto reject; - if (unlikely(!cb)) - goto reject; - if (unlikely(++p_ptr->publ.conn_unacked >= - TIPC_FLOW_CONTROL_WIN)) - tipc_acknowledge(dref, - p_ptr->publ.conn_unacked); - skb_pull(buf, msg_hdr_sz(msg)); - cb(usr_handle, dref, &buf, msg_data(msg), - msg_data_sz(msg)); - break; - } - case TIPC_DIRECT_MSG:{ - tipc_msg_event cb = up_ptr->msg_cb; - - tipc_port_unlock(p_ptr); - if (unlikely(connected)) - goto reject; - if (unlikely(!cb)) - goto reject; - skb_pull(buf, msg_hdr_sz(msg)); - cb(usr_handle, dref, &buf, msg_data(msg), - msg_data_sz(msg), msg_importance(msg), - &orig); - break; - } - case TIPC_MCAST_MSG: - case TIPC_NAMED_MSG:{ - tipc_named_msg_event cb = up_ptr->named_msg_cb; - - tipc_port_unlock(p_ptr); - if (unlikely(connected)) - goto reject; - if (unlikely(!cb)) - goto reject; - if (unlikely(!published)) - goto reject; - dseq.type = msg_nametype(msg); - dseq.lower = msg_nameinst(msg); - dseq.upper = (message_type == TIPC_NAMED_MSG) - ? dseq.lower : msg_nameupper(msg); - skb_pull(buf, msg_hdr_sz(msg)); - cb(usr_handle, dref, &buf, msg_data(msg), - msg_data_sz(msg), msg_importance(msg), - &orig, &dseq); - break; - } - } - if (buf) - buf_discard(buf); - buf = next; - continue; -err: - switch (message_type) { - - case TIPC_CONN_MSG:{ - tipc_conn_shutdown_event cb = - up_ptr->conn_err_cb; - u32 peer_port = port_peerport(p_ptr); - u32 peer_node = port_peernode(p_ptr); - - tipc_port_unlock(p_ptr); - if (!connected || !cb) - break; - if (msg_origport(msg) != peer_port) - break; - if (msg_orignode(msg) != peer_node) - break; - tipc_disconnect(dref); - skb_pull(buf, msg_hdr_sz(msg)); - cb(usr_handle, dref, &buf, msg_data(msg), - msg_data_sz(msg), msg_errcode(msg)); - break; - } - case TIPC_DIRECT_MSG:{ - tipc_msg_err_event cb = up_ptr->err_cb; - - tipc_port_unlock(p_ptr); - if (connected || !cb) - break; - skb_pull(buf, msg_hdr_sz(msg)); - cb(usr_handle, dref, &buf, msg_data(msg), - msg_data_sz(msg), msg_errcode(msg), &orig); - break; - } - case TIPC_MCAST_MSG: - case TIPC_NAMED_MSG:{ - tipc_named_msg_err_event cb = - up_ptr->named_err_cb; - - tipc_port_unlock(p_ptr); - if (connected || !cb) - break; - dseq.type = msg_nametype(msg); - dseq.lower = msg_nameinst(msg); - dseq.upper = (message_type == TIPC_NAMED_MSG) - ? dseq.lower : msg_nameupper(msg); - skb_pull(buf, msg_hdr_sz(msg)); - cb(usr_handle, dref, &buf, msg_data(msg), - msg_data_sz(msg), msg_errcode(msg), &dseq); - break; - } - } - if (buf) - buf_discard(buf); - buf = next; - continue; -reject: - tipc_reject_msg(buf, TIPC_ERR_NO_PORT); - buf = next; - } -} - -/* - * port_dispatcher(): Dispatcher for messages destinated - * to the tipc_port interface. Called with port locked. - */ - -static u32 port_dispatcher(struct tipc_port *dummy, struct sk_buff *buf) -{ - buf->next = NULL; - spin_lock_bh(&queue_lock); - if (msg_queue_head) { - msg_queue_tail->next = buf; - msg_queue_tail = buf; - } else { - msg_queue_tail = msg_queue_head = buf; - tipc_k_signal((Handler)port_dispatcher_sigh, 0); - } - spin_unlock_bh(&queue_lock); - return TIPC_OK; -} - -/* - * Wake up port after congestion: Called with port locked, - * - */ - -static void port_wakeup_sh(unsigned long ref) -{ - struct port *p_ptr; - struct user_port *up_ptr; - tipc_continue_event cb = NULL; - void *uh = NULL; - - p_ptr = tipc_port_lock(ref); - if (p_ptr) { - up_ptr = p_ptr->user_port; - if (up_ptr) { - cb = up_ptr->continue_event_cb; - uh = up_ptr->usr_handle; - } - tipc_port_unlock(p_ptr); - } - if (cb) - cb(uh, ref); -} - - -static void port_wakeup(struct tipc_port *p_ptr) -{ - tipc_k_signal((Handler)port_wakeup_sh, p_ptr->ref); -} - void tipc_acknowledge(u32 ref, u32 ack) { - struct port *p_ptr; + struct tipc_port *p_ptr; struct sk_buff *buf = NULL; p_ptr = tipc_port_lock(ref); if (!p_ptr) return; - if (p_ptr->publ.connected) { - p_ptr->publ.conn_unacked -= ack; - buf = port_build_proto_msg(port_peerport(p_ptr), - port_peernode(p_ptr), - ref, - tipc_own_addr, - CONN_MANAGER, - CONN_ACK, - TIPC_OK, - port_out_seqno(p_ptr), - ack); + 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); } -/* - * tipc_createport(): user level call. Will add port to - * registry if non-zero user_ref. - */ - -int tipc_createport(u32 user_ref, - void *usr_handle, - unsigned int importance, - tipc_msg_err_event error_cb, - tipc_named_msg_err_event named_error_cb, - tipc_conn_shutdown_event conn_error_cb, - tipc_msg_event msg_cb, - tipc_named_msg_event named_msg_cb, - tipc_conn_msg_event conn_msg_cb, - tipc_continue_event continue_event_cb,/* May be zero */ - u32 *portref) -{ - struct user_port *up_ptr; - struct port *p_ptr; - u32 ref; - - up_ptr = kmalloc(sizeof(*up_ptr), GFP_ATOMIC); - if (!up_ptr) { - warn("Port creation failed, no memory\n"); - return -ENOMEM; - } - ref = tipc_createport_raw(NULL, port_dispatcher, port_wakeup, importance); - p_ptr = tipc_port_lock(ref); - if (!p_ptr) { - kfree(up_ptr); - return -ENOMEM; - } - - p_ptr->user_port = up_ptr; - up_ptr->user_ref = user_ref; - up_ptr->usr_handle = usr_handle; - up_ptr->ref = p_ptr->publ.ref; - up_ptr->err_cb = error_cb; - up_ptr->named_err_cb = named_error_cb; - up_ptr->conn_err_cb = conn_error_cb; - up_ptr->msg_cb = msg_cb; - up_ptr->named_msg_cb = named_msg_cb; - up_ptr->conn_msg_cb = conn_msg_cb; - up_ptr->continue_event_cb = continue_event_cb; - INIT_LIST_HEAD(&up_ptr->uport_list); - tipc_reg_add_port(up_ptr); - *portref = p_ptr->publ.ref; - dbg(" tipc_createport: %x with ref %u\n", p_ptr, p_ptr->publ.ref); - tipc_port_unlock(p_ptr); - return TIPC_OK; -} - -int tipc_ownidentity(u32 ref, struct tipc_portid *id) -{ - id->ref = ref; - id->node = tipc_own_addr; - return TIPC_OK; -} - -int tipc_portimportance(u32 ref, unsigned int *importance) -{ - struct port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - *importance = (unsigned int)msg_importance(&p_ptr->publ.phdr); - tipc_port_unlock(p_ptr); - return TIPC_OK; -} - -int tipc_set_portimportance(u32 ref, unsigned int imp) +int tipc_publish(struct tipc_port *p_ptr, unsigned int scope, + struct tipc_name_seq const *seq) { - struct port *p_ptr; - - if (imp > TIPC_CRITICAL_IMPORTANCE) - return -EINVAL; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - msg_set_importance(&p_ptr->publ.phdr, (u32)imp); - tipc_port_unlock(p_ptr); - return TIPC_OK; -} - - -int tipc_publish(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) -{ - struct port *p_ptr; struct publication *publ; u32 key; - int res = -EINVAL; - p_ptr = tipc_port_lock(ref); - if (!p_ptr) + if (p_ptr->connected) return -EINVAL; + key = p_ptr->ref + p_ptr->pub_count + 1; + if (key == p_ptr->ref) + return -EADDRINUSE; - dbg("tipc_publ %u, p_ptr = %x, conn = %x, scope = %x, " - "lower = %u, upper = %u\n", - ref, p_ptr, p_ptr->publ.connected, scope, seq->lower, seq->upper); - if (p_ptr->publ.connected) - goto exit; - if (seq->lower > seq->upper) - goto exit; - if ((scope < TIPC_ZONE_SCOPE) || (scope > TIPC_NODE_SCOPE)) - goto exit; - key = ref + p_ptr->pub_count + 1; - if (key == ref) { - res = -EADDRINUSE; - goto exit; - } publ = tipc_nametbl_publish(seq->type, seq->lower, seq->upper, - scope, p_ptr->publ.ref, key); + scope, p_ptr->ref, key); if (publ) { list_add(&publ->pport_list, &p_ptr->publications); p_ptr->pub_count++; - p_ptr->publ.published = 1; - res = TIPC_OK; + p_ptr->published = 1; + return 0; } -exit: - tipc_port_unlock(p_ptr); - return res; + return -EINVAL; } -int tipc_withdraw(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) +int tipc_withdraw(struct tipc_port *p_ptr, unsigned int scope, + struct tipc_name_seq const *seq) { - struct port *p_ptr; struct publication *publ; struct publication *tpubl; int res = -EINVAL; - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -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 = TIPC_OK; + res = 0; } else { list_for_each_entry_safe(publ, tpubl, &p_ptr->publications, pport_list) { @@ -1195,516 +643,256 @@ int tipc_withdraw(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) break; tipc_nametbl_withdraw(publ->type, publ->lower, publ->ref, publ->key); - res = TIPC_OK; + res = 0; break; } } if (list_empty(&p_ptr->publications)) - p_ptr->publ.published = 0; - tipc_port_unlock(p_ptr); + p_ptr->published = 0; return res; } -int tipc_connect2port(u32 ref, struct tipc_portid const *peer) +int tipc_port_connect(u32 ref, struct tipc_portid const *peer) { - struct port *p_ptr; - struct tipc_msg *msg; - int res = -EINVAL; + struct tipc_port *p_ptr; + int res; p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; - if (p_ptr->publ.published || p_ptr->publ.connected) + 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->publ.phdr; + msg = &p_ptr->phdr; msg_set_destnode(msg, peer->node); msg_set_destport(msg, peer->ref); - msg_set_orignode(msg, tipc_own_addr); - msg_set_origport(msg, p_ptr->publ.ref); - msg_set_transp_seqno(msg, 42); msg_set_type(msg, TIPC_CONN_MSG); - if (!may_route(peer->node)) - msg_set_hdr_sz(msg, SHORT_H_SIZE); - else - msg_set_hdr_sz(msg, LONG_H_SIZE); + 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->publ.connected = 1; + p_ptr->connected = 1; k_start_timer(&p_ptr->timer, p_ptr->probing_interval); - tipc_nodesub_subscribe(&p_ptr->subscription,peer->node, + tipc_nodesub_subscribe(&p_ptr->subscription, peer->node, (void *)(unsigned long)ref, (net_ev_handler)port_handle_node_down); - res = TIPC_OK; + res = 0; exit: - tipc_port_unlock(p_ptr); - p_ptr->publ.max_pkt = tipc_link_get_max_pkt(peer->node, ref); + p_ptr->max_pkt = tipc_link_get_max_pkt(peer->node, ref); return res; } /* - * tipc_disconnect(): Disconnect port form peer. - * This is a node local operation. + * __tipc_disconnect - disconnect port from peer + * + * Port must be locked. */ - -int tipc_disconnect(u32 ref) +int __tipc_port_disconnect(struct tipc_port *tp_ptr) { - struct port *p_ptr; - int res = -ENOTCONN; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - if (p_ptr->publ.connected) { - p_ptr->publ.connected = 0; + if (tp_ptr->connected) { + tp_ptr->connected = 0; /* let timer expire on it's own to avoid deadlock! */ - tipc_nodesub_unsubscribe(&p_ptr->subscription); - res = TIPC_OK; + tipc_nodesub_unsubscribe(&tp_ptr->subscription); + return 0; } - tipc_port_unlock(p_ptr); - return res; + + return -ENOTCONN; } /* - * tipc_shutdown(): Send a SHUTDOWN msg to peer and disconnect + * tipc_port_disconnect(): Disconnect port form peer. + * This is a node local operation. */ -int tipc_shutdown(u32 ref) +int tipc_port_disconnect(u32 ref) { - struct port *p_ptr; - struct sk_buff *buf = NULL; + struct tipc_port *p_ptr; + int res; p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; - - if (p_ptr->publ.connected) { - u32 imp = msg_importance(&p_ptr->publ.phdr); - if (imp < TIPC_CRITICAL_IMPORTANCE) - imp++; - buf = port_build_proto_msg(port_peerport(p_ptr), - port_peernode(p_ptr), - ref, - tipc_own_addr, - imp, - TIPC_CONN_MSG, - TIPC_CONN_SHUTDOWN, - port_out_seqno(p_ptr), - 0); - } + res = __tipc_port_disconnect(p_ptr); tipc_port_unlock(p_ptr); - tipc_net_route_msg(buf); - return tipc_disconnect(ref); + return res; } -int tipc_isconnected(u32 ref, int *isconnected) +/* + * tipc_port_shutdown(): Send a SHUTDOWN msg to peer and disconnect + */ +int tipc_port_shutdown(u32 ref) { - struct port *p_ptr; + struct tipc_port *p_ptr; + struct sk_buff *buf = NULL; p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; - *isconnected = p_ptr->publ.connected; - tipc_port_unlock(p_ptr); - return TIPC_OK; -} - -int tipc_peer(u32 ref, struct tipc_portid *peer) -{ - struct port *p_ptr; - int res; - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - if (p_ptr->publ.connected) { - peer->ref = port_peerport(p_ptr); - peer->node = port_peernode(p_ptr); - res = TIPC_OK; - } else - res = -ENOTCONN; + buf = port_build_peer_abort_msg(p_ptr, TIPC_CONN_SHUTDOWN); tipc_port_unlock(p_ptr); - return res; -} - -int tipc_ref_valid(u32 ref) -{ - /* Works irrespective of type */ - return !!tipc_ref_deref(ref); + tipc_net_route_msg(buf); + return tipc_port_disconnect(ref); } - /* - * tipc_port_recv_sections(): Concatenate and deliver sectioned - * message for this node. + * tipc_port_iovec_rcv: Concatenate and deliver sectioned + * message for this node. */ - -int tipc_port_recv_sections(struct port *sender, unsigned int num_sect, - struct iovec const *msg_sect) +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 = msg_build(&sender->publ.phdr, msg_sect, num_sect, - MAX_MSG_SIZE, !sender->user_port, &buf); + res = tipc_msg_build(&sender->phdr, msg_sect, len, MAX_MSG_SIZE, &buf); if (likely(buf)) - tipc_port_recv_msg(buf); + tipc_sk_rcv(buf); return res; } /** * tipc_send - send message sections on connection */ - -int tipc_send(u32 ref, unsigned int num_sect, struct iovec const *msg_sect) +int tipc_send(struct tipc_port *p_ptr, + struct iovec const *msg_sect, + unsigned int len) { - struct port *p_ptr; u32 destnode; int res; - p_ptr = tipc_port_deref(ref); - if (!p_ptr || !p_ptr->publ.connected) + if (!p_ptr->connected) return -EINVAL; - p_ptr->publ.congested = 1; + p_ptr->congested = 1; if (!tipc_port_congested(p_ptr)) { - destnode = port_peernode(p_ptr); - if (likely(destnode != tipc_own_addr)) - res = tipc_link_send_sections_fast(p_ptr, msg_sect, num_sect, - destnode); + 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_recv_sections(p_ptr, num_sect, msg_sect); + res = tipc_port_iovec_rcv(p_ptr, msg_sect, len); if (likely(res != -ELINKCONG)) { - port_incr_out_seqno(p_ptr); - p_ptr->publ.congested = 0; - p_ptr->sent++; + p_ptr->congested = 0; + if (res > 0) + p_ptr->sent++; return res; } } - if (port_unreliable(p_ptr)) { - p_ptr->publ.congested = 0; - /* Just calculate msg length and return */ - return msg_calc_data_size(msg_sect, num_sect); + if (tipc_port_unreliable(p_ptr)) { + p_ptr->congested = 0; + return len; } return -ELINKCONG; } /** - * tipc_send_buf - send message buffer on connection - */ - -int tipc_send_buf(u32 ref, struct sk_buff *buf, unsigned int dsz) -{ - struct port *p_ptr; - struct tipc_msg *msg; - u32 destnode; - u32 hsz; - u32 sz; - u32 res; - - p_ptr = tipc_port_deref(ref); - if (!p_ptr || !p_ptr->publ.connected) - return -EINVAL; - - msg = &p_ptr->publ.phdr; - hsz = msg_hdr_sz(msg); - sz = hsz + dsz; - msg_set_size(msg, sz); - if (skb_cow(buf, hsz)) - return -ENOMEM; - - skb_push(buf, hsz); - skb_copy_to_linear_data(buf, msg, hsz); - destnode = msg_destnode(msg); - p_ptr->publ.congested = 1; - if (!tipc_port_congested(p_ptr)) { - if (likely(destnode != tipc_own_addr)) - res = tipc_send_buf_fast(buf, destnode); - else { - tipc_port_recv_msg(buf); - res = sz; - } - if (likely(res != -ELINKCONG)) { - port_incr_out_seqno(p_ptr); - p_ptr->sent++; - p_ptr->publ.congested = 0; - return res; - } - } - if (port_unreliable(p_ptr)) { - p_ptr->publ.congested = 0; - return dsz; - } - return -ELINKCONG; -} - -/** - * tipc_forward2name - forward message sections to port name - */ - -int tipc_forward2name(u32 ref, - struct tipc_name const *name, - u32 domain, - u32 num_sect, - struct iovec const *msg_sect, - struct tipc_portid const *orig, - unsigned int importance) -{ - struct port *p_ptr; - struct tipc_msg *msg; - u32 destnode = domain; - u32 destport = 0; - int res; - - p_ptr = tipc_port_deref(ref); - if (!p_ptr || p_ptr->publ.connected) - return -EINVAL; - - msg = &p_ptr->publ.phdr; - msg_set_type(msg, TIPC_NAMED_MSG); - msg_set_orignode(msg, orig->node); - msg_set_origport(msg, orig->ref); - msg_set_hdr_sz(msg, LONG_H_SIZE); - msg_set_nametype(msg, name->type); - msg_set_nameinst(msg, name->instance); - msg_set_lookup_scope(msg, addr_scope(domain)); - if (importance <= TIPC_CRITICAL_IMPORTANCE) - msg_set_importance(msg,importance); - destport = tipc_nametbl_translate(name->type, name->instance, &destnode); - msg_set_destnode(msg, destnode); - msg_set_destport(msg, destport); - - if (likely(destport || destnode)) { - p_ptr->sent++; - if (likely(destnode == tipc_own_addr)) - return tipc_port_recv_sections(p_ptr, num_sect, msg_sect); - res = tipc_link_send_sections_fast(p_ptr, msg_sect, num_sect, - destnode); - if (likely(res != -ELINKCONG)) - return res; - if (port_unreliable(p_ptr)) { - /* Just calculate msg length and return */ - return msg_calc_data_size(msg_sect, num_sect); - } - return -ELINKCONG; - } - return tipc_port_reject_sections(p_ptr, msg, msg_sect, num_sect, - TIPC_ERR_NO_NAME); -} - -/** * tipc_send2name - send message sections to port name */ - -int tipc_send2name(u32 ref, +int tipc_send2name(struct tipc_port *p_ptr, struct tipc_name const *name, unsigned int domain, - unsigned int num_sect, - struct iovec const *msg_sect) + struct iovec const *msg_sect, + unsigned int len) { - struct tipc_portid orig; - - orig.ref = ref; - orig.node = tipc_own_addr; - return tipc_forward2name(ref, name, domain, num_sect, msg_sect, &orig, - TIPC_PORT_IMPORTANCE); -} - -/** - * tipc_forward_buf2name - forward message buffer to port name - */ - -int tipc_forward_buf2name(u32 ref, - struct tipc_name const *name, - u32 domain, - struct sk_buff *buf, - unsigned int dsz, - struct tipc_portid const *orig, - unsigned int importance) -{ - struct port *p_ptr; struct tipc_msg *msg; u32 destnode = domain; - u32 destport = 0; + u32 destport; int res; - p_ptr = (struct port *)tipc_ref_deref(ref); - if (!p_ptr || p_ptr->publ.connected) + if (p_ptr->connected) return -EINVAL; - msg = &p_ptr->publ.phdr; - if (importance <= TIPC_CRITICAL_IMPORTANCE) - msg_set_importance(msg, importance); + msg = &p_ptr->phdr; msg_set_type(msg, TIPC_NAMED_MSG); - msg_set_orignode(msg, orig->node); - msg_set_origport(msg, orig->ref); + 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, addr_scope(domain)); - msg_set_hdr_sz(msg, LONG_H_SIZE); - msg_set_size(msg, LONG_H_SIZE + dsz); + 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); - msg_dbg(msg, "forw2name ==> "); - if (skb_cow(buf, LONG_H_SIZE)) - return -ENOMEM; - skb_push(buf, LONG_H_SIZE); - skb_copy_to_linear_data(buf, msg, LONG_H_SIZE); - msg_dbg(buf_msg(buf),"PREP:"); + if (likely(destport || destnode)) { - p_ptr->sent++; - if (destnode == tipc_own_addr) - return tipc_port_recv_msg(buf); - res = tipc_send_buf_fast(buf, destnode); - if (likely(res != -ELINKCONG)) + 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 (port_unreliable(p_ptr)) - return dsz; - return -ELINKCONG; - } - return tipc_reject_msg(buf, TIPC_ERR_NO_NAME); -} - -/** - * tipc_send_buf2name - send message buffer to port name - */ - -int tipc_send_buf2name(u32 ref, - struct tipc_name const *dest, - u32 domain, - struct sk_buff *buf, - unsigned int dsz) -{ - struct tipc_portid orig; - - orig.ref = ref; - orig.node = tipc_own_addr; - return tipc_forward_buf2name(ref, dest, domain, buf, dsz, &orig, - TIPC_PORT_IMPORTANCE); -} - -/** - * tipc_forward2port - forward message sections to port identity - */ - -int tipc_forward2port(u32 ref, - struct tipc_portid const *dest, - unsigned int num_sect, - struct iovec const *msg_sect, - struct tipc_portid const *orig, - unsigned int importance) -{ - struct port *p_ptr; - struct tipc_msg *msg; - int res; - - p_ptr = tipc_port_deref(ref); - if (!p_ptr || p_ptr->publ.connected) - return -EINVAL; + } + if (tipc_port_unreliable(p_ptr)) + return len; - msg = &p_ptr->publ.phdr; - msg_set_type(msg, TIPC_DIRECT_MSG); - msg_set_orignode(msg, orig->node); - msg_set_origport(msg, orig->ref); - msg_set_destnode(msg, dest->node); - msg_set_destport(msg, dest->ref); - msg_set_hdr_sz(msg, DIR_MSG_H_SIZE); - if (importance <= TIPC_CRITICAL_IMPORTANCE) - msg_set_importance(msg, importance); - p_ptr->sent++; - if (dest->node == tipc_own_addr) - return tipc_port_recv_sections(p_ptr, num_sect, msg_sect); - res = tipc_link_send_sections_fast(p_ptr, msg_sect, num_sect, dest->node); - if (likely(res != -ELINKCONG)) - return res; - if (port_unreliable(p_ptr)) { - /* Just calculate msg length and return */ - return msg_calc_data_size(msg_sect, num_sect); + return -ELINKCONG; } - 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(u32 ref, +int tipc_send2port(struct tipc_port *p_ptr, struct tipc_portid const *dest, - unsigned int num_sect, - struct iovec const *msg_sect) -{ - struct tipc_portid orig; - - orig.ref = ref; - orig.node = tipc_own_addr; - return tipc_forward2port(ref, dest, num_sect, msg_sect, &orig, - TIPC_PORT_IMPORTANCE); -} - -/** - * tipc_forward_buf2port - forward message buffer to port identity - */ -int tipc_forward_buf2port(u32 ref, - struct tipc_portid const *dest, - struct sk_buff *buf, - unsigned int dsz, - struct tipc_portid const *orig, - unsigned int importance) + struct iovec const *msg_sect, + unsigned int len) { - struct port *p_ptr; struct tipc_msg *msg; int res; - p_ptr = (struct port *)tipc_ref_deref(ref); - if (!p_ptr || p_ptr->publ.connected) + if (p_ptr->connected) return -EINVAL; - msg = &p_ptr->publ.phdr; + msg = &p_ptr->phdr; msg_set_type(msg, TIPC_DIRECT_MSG); - msg_set_orignode(msg, orig->node); - msg_set_origport(msg, orig->ref); + msg_set_lookup_scope(msg, 0); msg_set_destnode(msg, dest->node); msg_set_destport(msg, dest->ref); - msg_set_hdr_sz(msg, DIR_MSG_H_SIZE); - if (importance <= TIPC_CRITICAL_IMPORTANCE) - msg_set_importance(msg, importance); - msg_set_size(msg, DIR_MSG_H_SIZE + dsz); - if (skb_cow(buf, DIR_MSG_H_SIZE)) - return -ENOMEM; - - skb_push(buf, DIR_MSG_H_SIZE); - skb_copy_to_linear_data(buf, msg, DIR_MSG_H_SIZE); - msg_dbg(msg, "buf2port: "); - p_ptr->sent++; - if (dest->node == tipc_own_addr) - return tipc_port_recv_msg(buf); - res = tipc_send_buf_fast(buf, dest->node); - if (likely(res != -ELINKCONG)) - return res; - if (port_unreliable(p_ptr)) - return dsz; - return -ELINKCONG; -} + msg_set_hdr_sz(msg, BASIC_H_SIZE); -/** - * tipc_send_buf2port - send message buffer to port identity - */ - -int tipc_send_buf2port(u32 ref, - struct tipc_portid const *dest, - struct sk_buff *buf, - unsigned int dsz) -{ - struct tipc_portid orig; + 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; - orig.ref = ref; - orig.node = tipc_own_addr; - return tipc_forward_buf2port(ref, dest, buf, dsz, &orig, - TIPC_PORT_IMPORTANCE); + return -ELINKCONG; } - diff --git a/net/tipc/port.h b/net/tipc/port.h index e5f8c16429b..cf4ca5b1d9a 100644 --- a/net/tipc/port.h +++ b/net/tipc/port.h @@ -1,8 +1,8 @@ /* * net/tipc/port.h: Include file for TIPC port code * - * Copyright (c) 1994-2007, Ericsson AB - * Copyright (c) 2004-2007, Wind River Systems + * 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 @@ -37,65 +37,53 @@ #ifndef _TIPC_PORT_H #define _TIPC_PORT_H -#include "core.h" #include "ref.h" #include "net.h" #include "msg.h" -#include "dbg.h" #include "node_subscr.h" -/** - * struct user_port - TIPC user port (used with native API) - * @user_ref: id of user who created user port - * @usr_handle: user-specified field - * @ref: object reference to associated TIPC port - * <various callback routines> - * @uport_list: adjacent user ports in list of ports held by user - */ - -struct user_port { - u32 user_ref; - void *usr_handle; - u32 ref; - tipc_msg_err_event err_cb; - tipc_named_msg_err_event named_err_cb; - tipc_conn_shutdown_event conn_err_cb; - tipc_msg_event msg_cb; - tipc_named_msg_event named_msg_cb; - tipc_conn_msg_event conn_msg_cb; - tipc_continue_event continue_event_cb; - struct list_head uport_list; -}; +#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 port - TIPC port structure - * @publ: TIPC port info available to privileged users + * 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 - * @dispatcher: ptr to routine which handles received messages - * @wakeup: ptr to routine to call when port is no longer congested - * @user_port: ptr to user port associated with port (if any) * @wait_list: adjacent ports in list of ports waiting on link congestion - * @congested_link: ptr to congested link port is waiting on * @waiting_pkts: - * @sent: - * @acked: + * @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: - * @last_in_seqno: * @timer_ref: * @subscription: "node down" subscription used to terminate failed connections */ - -struct port { - struct tipc_port publ; +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; - u32 (*dispatcher)(struct tipc_port *, struct sk_buff *); - void (*wakeup)(struct tipc_port *); - struct user_port *user_port; struct list_head wait_list; - struct link *congested_link; u32 waiting_pkts; u32 sent; u32 acked; @@ -103,32 +91,88 @@ struct port { u32 pub_count; u32 probing_state; u32 probing_interval; - u32 last_in_seqno; struct timer_list timer; - struct node_subscr subscription; + struct tipc_node_subscr subscription; }; extern spinlock_t tipc_port_list_lock; -struct port_list; +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); -int tipc_port_recv_sections(struct port *p_ptr, u32 num_sect, - struct iovec const *msg_sect); -int tipc_port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, - struct iovec const *msg_sect, u32 num_sect, - int err); struct sk_buff *tipc_port_get_ports(void); -struct sk_buff *port_show_stats(const void *req_tlv_area, int req_tlv_space); -void tipc_port_recv_proto_msg(struct sk_buff *buf); -void tipc_port_recv_mcast(struct sk_buff *buf, struct port_list *dp); +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 port *tipc_port_lock(u32 ref) +static inline struct tipc_port *tipc_port_lock(u32 ref) { - return (struct port *)tipc_ref_lock(ref); + return (struct tipc_port *)tipc_ref_lock(ref); } /** @@ -136,72 +180,58 @@ static inline struct port *tipc_port_lock(u32 ref) * * 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 void tipc_port_unlock(struct port *p_ptr) +static inline u32 tipc_port_peernode(struct tipc_port *p_ptr) { - spin_unlock_bh(p_ptr->publ.lock); + return msg_destnode(&p_ptr->phdr); } -static inline struct port* tipc_port_deref(u32 ref) +static inline u32 tipc_port_peerport(struct tipc_port *p_ptr) { - return (struct port *)tipc_ref_deref(ref); + return msg_destport(&p_ptr->phdr); } -static inline u32 tipc_peer_port(struct port *p_ptr) +static inline bool tipc_port_unreliable(struct tipc_port *port) { - return msg_destport(&p_ptr->publ.phdr); + return msg_src_droppable(&port->phdr) != 0; } -static inline u32 tipc_peer_node(struct port *p_ptr) +static inline void tipc_port_set_unreliable(struct tipc_port *port, + bool unreliable) { - return msg_destnode(&p_ptr->publ.phdr); + msg_set_src_droppable(&port->phdr, unreliable ? 1 : 0); } -static inline int tipc_port_congested(struct port *p_ptr) +static inline bool tipc_port_unreturnable(struct tipc_port *port) { - return((p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2)); + return msg_dest_droppable(&port->phdr) != 0; } -/** - * tipc_port_recv_msg - receive message from lower layer and deliver to port user - */ +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 int tipc_port_recv_msg(struct sk_buff *buf) +static inline void tipc_port_set_importance(struct tipc_port *port, int imp) { - struct port *p_ptr; - struct tipc_msg *msg = buf_msg(buf); - u32 destport = msg_destport(msg); - u32 dsz = msg_data_sz(msg); - u32 err; - - /* forward unresolved named message */ - if (unlikely(!destport)) { - tipc_net_route_msg(buf); - return dsz; - } - - /* validate destination & pass to port, otherwise reject message */ - p_ptr = tipc_port_lock(destport); - if (likely(p_ptr)) { - if (likely(p_ptr->publ.connected)) { - if ((unlikely(msg_origport(msg) != tipc_peer_port(p_ptr))) || - (unlikely(msg_orignode(msg) != tipc_peer_node(p_ptr))) || - (unlikely(!msg_connected(msg)))) { - err = TIPC_ERR_NO_PORT; - tipc_port_unlock(p_ptr); - goto reject; - } - } - err = p_ptr->dispatcher(&p_ptr->publ, buf); - tipc_port_unlock(p_ptr); - if (likely(!err)) - return dsz; - } else { - err = TIPC_ERR_NO_PORT; - } -reject: - dbg("port->rejecting, err = %x..\n",err); - return tipc_reject_msg(buf, err); + msg_set_importance(&port->phdr, (u32)imp); } #endif diff --git a/net/tipc/ref.c b/net/tipc/ref.c index c38744c96ed..3d4ecd754ee 100644 --- a/net/tipc/ref.c +++ b/net/tipc/ref.c @@ -2,7 +2,7 @@ * net/tipc/ref.c: TIPC object registry code * * Copyright (c) 1991-2006, Ericsson AB - * Copyright (c) 2004-2005, Wind River Systems + * Copyright (c) 2004-2007, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -36,77 +36,96 @@ #include "core.h" #include "ref.h" -#include "port.h" -#include "subscr.h" -#include "name_distr.h" -#include "name_table.h" -#include "config.h" -#include "discover.h" -#include "bearer.h" -#include "node.h" -#include "bcast.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. * - * A used entry has object ptr != 0, reference == XXXX|own index - * (XXXX changes each time entry is acquired) - * A free entry has object ptr == 0, reference == YYYY|next free index - * (YYYY is one more than last used XXXX) + * 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 * - * Free list is initially chained from entry (2**N)-1 to entry 1. - * Entry 0 is not used to allow index 0 to indicate the end of the free list. + * Entry 0 is not used; this allows index 0 to denote the end of the free list. * - * Note: Any accidental reference of the form XXXX|0--0 won't match entry 0 - * because entry 0's reference field has the form XXXX|1--1. + * 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). */ -struct ref_table tipc_ref_table = { NULL }; +static struct ref_table tipc_ref_table; -static DEFINE_RWLOCK(ref_table_lock); +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 sz = 1 << 4; - u32 index_mask; - int i; + u32 actual_size; - while (sz < requested_size) { - sz <<= 1; - } - table = vmalloc(sz * sizeof(*table)); + /* 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; - write_lock_bh(&ref_table_lock); - index_mask = sz - 1; - for (i = sz - 1; i >= 0; i--) { - table[i].object = NULL; - spin_lock_init(&table[i].lock); - table[i].data.next_plus_upper = (start & ~index_mask) + i - 1; - } tipc_ref_table.entries = table; - tipc_ref_table.index_mask = index_mask; - tipc_ref_table.first_free = sz - 1; - tipc_ref_table.last_free = 1; - write_unlock_bh(&ref_table_lock); - return TIPC_OK; + 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) { - if (!tipc_ref_table.entries) - return; - vfree(tipc_ref_table.entries); tipc_ref_table.entries = NULL; } @@ -114,46 +133,66 @@ void tipc_ref_table_stop(void) /** * tipc_ref_acquire - create reference to an object * - * Return a unique reference value which can be translated back to the pointer - * 'object' at a later time. Also, pass back a pointer to the lock protecting - * the object, but without locking it. + * 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) { - struct reference *entry; u32 index; u32 index_mask; u32 next_plus_upper; - u32 reference = 0; + u32 ref; + struct reference *entry = NULL; if (!object) { - err("Attempt to acquire reference to non-existent object\n"); + pr_err("Attempt to acquire ref. to non-existent obj\n"); return 0; } if (!tipc_ref_table.entries) { - err("Reference table not found during acquisition attempt\n"); + pr_err("Ref. table not found in acquisition attempt\n"); return 0; } - write_lock_bh(&ref_table_lock); + /* 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; - /* take lock in case a previous user of entry still holds it */ - spin_lock_bh(&entry->lock); - next_plus_upper = entry->data.next_plus_upper; + next_plus_upper = entry->ref; tipc_ref_table.first_free = next_plus_upper & index_mask; - reference = (next_plus_upper & ~index_mask) + index; - entry->data.reference = reference; + 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; - if (lock != NULL) - *lock = &entry->lock; - spin_unlock_bh(&entry->lock); + *lock = &entry->lock; + /* + * keep it locked, the caller is responsible + * for unlocking this when they're done with it + */ } - write_unlock_bh(&ref_table_lock); - return reference; + + return ref; } /** @@ -162,49 +201,66 @@ u32 tipc_ref_acquire(void *object, spinlock_t **lock) * 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 (!ref) { - err("Attempt to discard reference 0\n"); - return; - } if (!tipc_ref_table.entries) { - err("Reference table not found during discard attempt\n"); + pr_err("Ref. table not found during discard attempt\n"); return; } - write_lock_bh(&ref_table_lock); 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) { - err("Attempt to discard reference to non-existent object\n"); + pr_err("Attempt to discard ref. to non-existent obj\n"); goto exit; } - if (entry->data.reference != ref) { - err("Attempt to discard non-existent reference\n"); + if (entry->ref != ref) { + pr_err("Attempt to discard non-existent reference\n"); goto exit; } - /* mark entry as unused */ + /* + * 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 - /* next_plus_upper is always XXXX|0--0 for last free entry */ - tipc_ref_table.entries[tipc_ref_table.last_free].data.next_plus_upper - |= index; + tipc_ref_table.entries[tipc_ref_table.last_free].ref |= index; tipc_ref_table.last_free = index; - /* increment upper bits of entry to invalidate subsequent references */ - entry->data.next_plus_upper = (ref & ~index_mask) + (index_mask + 1); exit: - write_unlock_bh(&ref_table_lock); + 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 index 38f3a7f4a78..d01aa1df63b 100644 --- a/net/tipc/ref.h +++ b/net/tipc/ref.h @@ -2,7 +2,7 @@ * net/tipc/ref.h: Include file for TIPC object registry code * * Copyright (c) 1991-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005-2006, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -37,95 +37,12 @@ #ifndef _TIPC_REF_H #define _TIPC_REF_H -/** - * struct reference - TIPC object reference entry - * @object: pointer to object associated with reference entry - * @lock: spinlock controlling access to object - * @data: reference value associated with object (or link to next unused entry) - */ - -struct reference { - void *object; - spinlock_t lock; - union { - u32 next_plus_upper; - u32 reference; - } data; -}; - -/** - * struct tipc_ref_table - table of TIPC object reference entries - * @entries: pointer to array of reference entries - * @index_mask: bitmask for array index portion of reference values - * @first_free: array index of first unused object reference entry - * @last_free: array index of last unused object reference entry - */ - -struct ref_table { - struct reference *entries; - u32 index_mask; - u32 first_free; - u32 last_free; -}; - -extern struct ref_table tipc_ref_table; - 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); - -/** - * tipc_ref_lock - lock referenced object and return pointer to it - */ - -static inline void *tipc_ref_lock(u32 ref) -{ - if (likely(tipc_ref_table.entries)) { - struct reference *r = - &tipc_ref_table.entries[ref & tipc_ref_table.index_mask]; - - spin_lock_bh(&r->lock); - if (likely(r->data.reference == ref)) - return r->object; - spin_unlock_bh(&r->lock); - } - return NULL; -} - -/** - * tipc_ref_unlock - unlock referenced object - */ - -static inline void tipc_ref_unlock(u32 ref) -{ - if (likely(tipc_ref_table.entries)) { - struct reference *r = - &tipc_ref_table.entries[ref & tipc_ref_table.index_mask]; - - if (likely(r->data.reference == ref)) - spin_unlock_bh(&r->lock); - else - err("tipc_ref_unlock() invoked using obsolete reference\n"); - } -} - -/** - * tipc_ref_deref - return pointer referenced object (without locking it) - */ - -static inline void *tipc_ref_deref(u32 ref) -{ - if (likely(tipc_ref_table.entries)) { - struct reference *r = - &tipc_ref_table.entries[ref & tipc_ref_table.index_mask]; - - if (likely(r->data.reference == ref)) - return r->object; - } - return NULL; -} +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 index 22909036b9b..ef0475568f9 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1,8 +1,8 @@ /* * net/tipc/socket.c: TIPC socket API * - * Copyright (c) 2001-2007, Ericsson AB - * Copyright (c) 2004-2007, Wind River Systems + * 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 @@ -34,200 +34,255 @@ * POSSIBILITY OF SUCH DAMAGE. */ -#include <linux/module.h> -#include <linux/types.h> -#include <linux/net.h> -#include <linux/socket.h> -#include <linux/errno.h> -#include <linux/mm.h> -#include <linux/slab.h> -#include <linux/poll.h> -#include <linux/fcntl.h> -#include <asm/semaphore.h> -#include <asm/string.h> -#include <asm/atomic.h> -#include <net/sock.h> - -#include <linux/tipc.h> -#include <linux/tipc_config.h> -#include <net/tipc/tipc_msg.h> -#include <net/tipc/tipc_port.h> - #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 OVERLOAD_LIMIT_BASE 5000 - -struct tipc_sock { - struct sock sk; - struct tipc_port *p; - struct semaphore sem; -}; - -#define tipc_sk(sk) ((struct tipc_sock*)sk) +#define CONN_TIMEOUT_DEFAULT 8000 /* default connect timeout = 8s */ -static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf); -static void wakeupdispatch(struct tipc_port *tport); +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 int sockets_enabled = 0; - -static atomic_t tipc_queue_size = ATOMIC_INIT(0); - +static struct proto tipc_proto_kern; /* - * sock_lock(): Lock a port/socket pair. lock_sock() can - * not be used here, since the same lock must protect ports - * with non-socket interfaces. - * See net.c for description of locking policy. + * 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 */ -static void sock_lock(struct tipc_sock* tsock) -{ - spin_lock_bh(tsock->p->lock); -} -/* - * sock_unlock(): Unlock a port/socket pair - */ -static void sock_unlock(struct tipc_sock* tsock) -{ - spin_unlock_bh(tsock->p->lock); -} +#include "socket.h" /** - * pollmask - determine the current set of poll() events for a socket - * @sock: socket structure - * - * TIPC sets the returned events as follows: - * a) POLLRDNORM and POLLIN are set if the socket's receive queue is non-empty - * or if a connection-oriented socket is does not have an active connection - * (i.e. a read operation will not block). - * b) POLLOUT is set except when a socket's connection has been terminated - * (i.e. a write operation will not block). - * c) POLLHUP is set when a socket's connection has been terminated. - * - * IMPORTANT: The fact that a read or write operation will not block does NOT - * imply that the operation will succeed! + * advance_rx_queue - discard first buffer in socket receive queue * - * Returns pollmask value + * Caller must hold socket lock */ - -static u32 pollmask(struct socket *sock) +static void advance_rx_queue(struct sock *sk) { - u32 mask; - - if ((skb_queue_len(&sock->sk->sk_receive_queue) != 0) || - (sock->state == SS_UNCONNECTED) || - (sock->state == SS_DISCONNECTING)) - mask = (POLLRDNORM | POLLIN); - else - mask = 0; - - if (sock->state == SS_DISCONNECTING) - mask |= POLLHUP; - else - mask |= POLLOUT; - - return mask; + kfree_skb(__skb_dequeue(&sk->sk_receive_queue)); } - /** - * advance_queue - discard first buffer in queue - * @tsock: TIPC socket + * reject_rx_queue - reject all buffers in socket receive queue + * + * Caller must hold socket lock */ - -static void advance_queue(struct tipc_sock *tsock) +static void reject_rx_queue(struct sock *sk) { - sock_lock(tsock); - buf_discard(skb_dequeue(&tsock->sk.sk_receive_queue)); - sock_unlock(tsock); - atomic_dec(&tipc_queue_size); + struct sk_buff *buf; + + while ((buf = __skb_dequeue(&sk->sk_receive_queue))) + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); } /** - * tipc_create - create a TIPC socket + * 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 and attaches a 'struct sock' to the 'struct socket', - * then create and attaches a TIPC port to the 'struct sock' part. + * 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_create(struct net *net, struct socket *sock, int protocol) +static int tipc_sk_create(struct net *net, struct socket *sock, + int protocol, int kern) { - struct tipc_sock *tsock; - struct tipc_port *port; + const struct proto_ops *ops; + socket_state state; struct sock *sk; + struct tipc_sock *tsk; + struct tipc_port *port; u32 ref; - if (net != &init_net) - return -EAFNOSUPPORT; - + /* Validate arguments */ if (unlikely(protocol != 0)) return -EPROTONOSUPPORT; - ref = tipc_createport_raw(NULL, &dispatch, &wakeupdispatch, TIPC_LOW_IMPORTANCE); - if (unlikely(!ref)) - return -ENOMEM; - - sock->state = SS_UNCONNECTED; - switch (sock->type) { case SOCK_STREAM: - sock->ops = &stream_ops; + ops = &stream_ops; + state = SS_UNCONNECTED; break; case SOCK_SEQPACKET: - sock->ops = &packet_ops; + ops = &packet_ops; + state = SS_UNCONNECTED; break; case SOCK_DGRAM: - tipc_set_portunreliable(ref, 1); - /* fall through */ case SOCK_RDM: - tipc_set_portunreturnable(ref, 1); - sock->ops = &msg_ops; - sock->state = SS_READY; + ops = &msg_ops; + state = SS_READY; break; default: - tipc_deleteport(ref); return -EPROTOTYPE; } - sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto); - if (!sk) { - tipc_deleteport(ref); + /* 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); - init_waitqueue_head(sk->sk_sleep); - sk->sk_rcvtimeo = 8 * HZ; /* default connect timeout = 8s */ + 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; +} - tsock = tipc_sk(sk); - port = tipc_get_port(ref); +/** + * 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; - tsock->p = port; - port->usr_handle = tsock; + 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); - init_MUTEX(&tsock->sem); + return 0; +} - dbg("sock_create: %x\n",tsock); +/** + * 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); +} - atomic_inc(&tipc_user_count); +/** + * 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 + */ - return 0; +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; } /** - * release - destroy a TIPC socket + * tipc_release - destroy a TIPC socket * @sock: socket to destroy * * This routine cleans up any messages that are still queued on the socket. @@ -242,62 +297,64 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol) * * Returns 0 on success, errno otherwise */ - -static int release(struct socket *sock) +static int tipc_release(struct socket *sock) { - struct tipc_sock *tsock = tipc_sk(sock->sk); struct sock *sk = sock->sk; - int res = TIPC_OK; + struct tipc_sock *tsk; + struct tipc_port *port; struct sk_buff *buf; - dbg("sock_delete: %x\n",tsock); - if (!tsock) - return 0; - down(&tsock->sem); - if (!sock->sk) { - up(&tsock->sem); + /* + * 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; - } - /* Reject unreceived messages, unless no longer connected */ + 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) { - sock_lock(tsock); - buf = skb_dequeue(&sk->sk_receive_queue); - if (!buf) - tsock->p->usr_handle = NULL; - sock_unlock(tsock); - if (!buf) + buf = __skb_dequeue(&sk->sk_receive_queue); + if (buf == NULL) break; - if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf))) - buf_discard(buf); - else + 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); - atomic_dec(&tipc_queue_size); + } } - /* Delete TIPC port */ - - res = tipc_deleteport(tsock->p->ref); - sock->sk = NULL; + /* Destroy TIPC port; also disconnects an active connection and + * sends a 'FIN-' to peer. + */ + tipc_port_destroy(port); - /* Discard any remaining messages */ + /* Discard any remaining (connection-based) messages in receive queue */ + __skb_queue_purge(&sk->sk_receive_queue); - while ((buf = skb_dequeue(&sk->sk_receive_queue))) { - buf_discard(buf); - atomic_dec(&tipc_queue_size); - } - - up(&tsock->sem); + /* Reject any messages that accumulated in backlog queue */ + sock->state = SS_DISCONNECTING; + release_sock(sk); sock_put(sk); + sock->sk = NULL; - atomic_dec(&tipc_user_count); - return res; + return 0; } /** - * bind - associate or disassocate TIPC name(s) with a socket + * 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 @@ -307,19 +364,21 @@ static int release(struct socket *sock) * (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 bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len) +static int tipc_bind(struct socket *sock, struct sockaddr *uaddr, + int uaddr_len) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; - int res; - - if (down_interruptible(&tsock->sem)) - return -ERESTARTSYS; + struct tipc_sock *tsk = tipc_sk(sk); + int res = -EINVAL; + lock_sock(sk); if (unlikely(!uaddr_len)) { - res = tipc_withdraw(tsock->p->ref, 0, NULL); + res = tipc_withdraw(&tsk->port, 0, NULL); goto exit; } @@ -327,11 +386,11 @@ static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len) 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) { @@ -339,66 +398,131 @@ static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len) goto exit; } - if (addr->scope > 0) - res = tipc_publish(tsock->p->ref, addr->scope, - &addr->addr.nameseq); - else - res = tipc_withdraw(tsock->p->ref, -addr->scope, - &addr->addr.nameseq); + 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: - up(&tsock->sem); + release_sock(sk); return res; } /** - * get_name - get port ID of socket or peer socket + * 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 to obtain socket name, 1 to obtain peer socket name + * @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 get_name(struct socket *sock, struct sockaddr *uaddr, - int *uaddr_len, int peer) +static int tipc_getname(struct socket *sock, struct sockaddr *uaddr, + int *uaddr_len, int peer) { - struct tipc_sock *tsock = tipc_sk(sock->sk); struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; - u32 res; + struct tipc_sock *tsk = tipc_sk(sock->sk); - if (down_interruptible(&tsock->sem)) - return -ERESTARTSYS; + 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; - if (peer) - res = tipc_peer(tsock->p->ref, &addr->addr.id); - else - res = tipc_ownidentity(tsock->p->ref, &addr->addr.id); addr->addr.name.domain = 0; - up(&tsock->sem); - return res; + return 0; } /** - * poll - read and possibly block on pollmask + * 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 the pollmask + * 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 poll(struct file *file, struct socket *sock, - poll_table *wait) +static unsigned int tipc_poll(struct file *file, struct socket *sock, + poll_table *wait) { - poll_wait(file, sock->sk->sk_sleep, wait); - /* NEED LOCK HERE? */ - return pollmask(sock); + 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; } /** @@ -411,7 +535,6 @@ static unsigned int poll(struct file *file, struct socket *sock, * * 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; @@ -420,10 +543,11 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) 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))) @@ -432,9 +556,35 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) 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; +} + + /** - * send_msg - send message in connectionless manner - * @iocb: (unused) + * 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 @@ -446,14 +596,15 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) * * Returns the number of bytes sent on success, or errno otherwise */ - -static int send_msg(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t total_len) +static int tipc_sendmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) { - struct tipc_sock *tsock = tipc_sk(sock->sk); - struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name; - struct sk_buff *buf; + 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)) @@ -461,86 +612,112 @@ static int send_msg(struct kiocb *iocb, struct socket *sock, 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) - return -EPIPE; - if (sock->state != SS_UNCONNECTED) - return -EISCONN; - if ((tsock->p->published) || - ((sock->type == SOCK_STREAM) && (total_len != 0))) - return -EOPNOTSUPP; + 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) { - tsock->p->conn_type = dest->addr.name.name.type; - tsock->p->conn_instance = dest->addr.name.name.instance; + tsk->port.conn_type = dest->addr.name.name.type; + tsk->port.conn_instance = dest->addr.name.name.instance; } - } - - if (down_interruptible(&tsock->sem)) - return -ERESTARTSYS; - - if (needs_conn) { /* Abort any pending connection attempts (very unlikely) */ - - while ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { - tipc_reject_msg(buf, TIPC_ERR_NO_PORT); - atomic_dec(&tipc_queue_size); - } - - sock->state = SS_CONNECTING; + reject_rx_queue(sk); } + timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); do { if (dest->addrtype == TIPC_ADDR_NAME) { - if ((res = dest_name_check(dest, m))) - goto exit; - res = tipc_send2name(tsock->p->ref, + res = dest_name_check(dest, m); + if (res) + break; + res = tipc_send2name(port, &dest->addr.name.name, dest->addr.name.domain, - m->msg_iovlen, - m->msg_iov); - } - else if (dest->addrtype == TIPC_ADDR_ID) { - res = tipc_send2port(tsock->p->ref, + m->msg_iov, + total_len); + } else if (dest->addrtype == TIPC_ADDR_ID) { + res = tipc_send2port(port, &dest->addr.id, - m->msg_iovlen, - m->msg_iov); - } - else if (dest->addrtype == TIPC_ADDR_MCAST) { + m->msg_iov, + total_len); + } else if (dest->addrtype == TIPC_ADDR_MCAST) { if (needs_conn) { res = -EOPNOTSUPP; - goto exit; + break; } - if ((res = dest_name_check(dest, m))) - goto exit; - res = tipc_multicast(tsock->p->ref, - &dest->addr.nameseq, - 0, - m->msg_iovlen, - m->msg_iov); + 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)) { -exit: - up(&tsock->sem); - return res; - } - if (m->msg_flags & MSG_DONTWAIT) { - res = -EWOULDBLOCK; - goto exit; - } - if (wait_event_interruptible(*sock->sk->sk_sleep, - !tsock->p->congested)) { - res = -ERESTARTSYS; - goto exit; + 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; } /** - * send_packet - send a connection-oriented message - * @iocb: (unused) + * 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 @@ -549,52 +726,50 @@ exit: * * Returns the number of bytes sent on success, or errno otherwise */ - -static int send_packet(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t total_len) +static int tipc_send_packet(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) { - struct tipc_sock *tsock = tipc_sk(sock->sk); - struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name; - int res; + 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 send_msg(iocb, sock, m, total_len); + return tipc_sendmsg(iocb, sock, m, total_len); + + if (total_len > TIPC_MAX_USER_MSG_SIZE) + return -EMSGSIZE; - if (down_interruptible(&tsock->sem)) { - return -ERESTARTSYS; + 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 { - if (unlikely(sock->state != SS_CONNECTED)) { - if (sock->state == SS_DISCONNECTING) - res = -EPIPE; - else - res = -ENOTCONN; - goto exit; - } - - res = tipc_send(tsock->p->ref, m->msg_iovlen, m->msg_iov); - if (likely(res != -ELINKCONG)) { -exit: - up(&tsock->sem); - return res; - } - if (m->msg_flags & MSG_DONTWAIT) { - res = -EWOULDBLOCK; - goto exit; - } - if (wait_event_interruptible(*sock->sk->sk_sleep, - !tsock->p->congested)) { - res = -ERESTARTSYS; - goto exit; - } + 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; } /** - * send_stream - send stream-oriented data + * tipc_send_stream - send stream-oriented data * @iocb: (unused) * @sock: socket structure * @m: data to send @@ -605,12 +780,11 @@ exit: * Returns the number of bytes sent on success (or partial success), * or errno if no data sent */ - - -static int send_stream(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t total_len) +static int tipc_send_stream(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) { - struct tipc_port *tport; + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); struct msghdr my_msg; struct iovec my_iov; struct iovec *curr_iov; @@ -622,19 +796,26 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, int bytes_sent; int res; - /* Handle special cases where there is no connection */ + lock_sock(sk); + /* Handle special cases where there is no connection */ if (unlikely(sock->state != SS_CONNECTED)) { if (sock->state == SS_UNCONNECTED) - return send_packet(iocb, sock, m, total_len); - else if (sock->state == SS_DISCONNECTING) - return -EPIPE; + res = tipc_send_packet(NULL, sock, m, total_len); else - return -ENOTCONN; + res = sock->state == SS_DISCONNECTING ? -EPIPE : -ENOTCONN; + goto exit; } - if (unlikely(m->msg_name)) - return -EISCONN; + 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 @@ -643,7 +824,6 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, * (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; @@ -652,25 +832,26 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, my_msg.msg_name = NULL; bytes_sent = 0; - tport = tipc_sk(sock->sk)->p; - hdr_size = msg_hdr_sz(&tport->phdr); + 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 = tport->max_pkt - hdr_size; + 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; - if ((res = send_packet(iocb, sock, &my_msg, 0)) < 0) { - if (bytes_sent != 0) + res = tipc_send_packet(NULL, sock, &my_msg, + bytes_to_send); + if (res < 0) { + if (bytes_sent) res = bytes_sent; - return res; + goto exit; } curr_left -= bytes_to_send; curr_start += bytes_to_send; @@ -679,33 +860,33 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, curr_iov++; } - - return bytes_sent; + res = bytes_sent; +exit: + release_sock(sk); + return res; } /** * auto_connect - complete connection setup to a remote port - * @sock: socket structure - * @tsock: TIPC-specific socket structure + * @tsk: tipc socket structure * @msg: peer's response message * * Returns 0 on success, errno otherwise */ - -static int auto_connect(struct socket *sock, struct tipc_sock *tsock, - struct tipc_msg *msg) +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; - if (msg_errcode(msg)) { - sock->state = SS_DISCONNECTING; - return -ECONNREFUSED; - } - peer.ref = msg_origport(msg); peer.node = msg_orignode(msg); - tipc_connect2port(tsock->p->ref, &peer); - tipc_set_portimportance(tsock->p->ref, msg_importance(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; } @@ -717,18 +898,18 @@ static int auto_connect(struct socket *sock, struct tipc_sock *tsock, * * Note: Address is not captured if not requested by receiver. */ - static void set_orig_addr(struct msghdr *m, struct tipc_msg *msg) { - struct sockaddr_tipc *addr = (struct sockaddr_tipc *)m->msg_name; + 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 */ + addr->addr.name.domain = 0; /* could leave uninitialized */ + addr->scope = 0; /* could leave uninitialized */ m->msg_namelen = sizeof(struct sockaddr_tipc); } } @@ -743,9 +924,8 @@ static void set_orig_addr(struct msghdr *m, struct tipc_msg *msg) * * Returns 0 if successful, otherwise errno */ - static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, - struct tipc_port *tport) + struct tipc_port *tport) { u32 anc_data[3]; u32 err; @@ -757,21 +937,22 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, 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); - if ((res = put_cmsg(m, SOL_TIPC, TIPC_ERRINFO, 8, anc_data))) - return res; - if (anc_data[1] && - (res = put_cmsg(m, SOL_TIPC, TIPC_RETDATA, anc_data[1], - msg_data(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: @@ -795,15 +976,50 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, default: has_name = 0; } - if (has_name && - (res = put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, anc_data))) - return res; + 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; +} + /** - * recv_msg - receive packet-oriented message + * tipc_recvmsg - receive packet-oriented message * @iocb: (unused) * @m: descriptor for message info * @buf_len: total size of user buffer area @@ -814,105 +1030,68 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, * * Returns size of returned message data, errno otherwise */ - -static int recv_msg(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t buf_len, int flags) +static int tipc_recvmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t buf_len, int flags) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + 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; - unsigned int q_len; + long timeo; unsigned int sz; u32 err; int res; - /* Currently doesn't support receiving into multiple iovec entries */ - - if (m->msg_iovlen != 1) - return -EOPNOTSUPP; - - /* Catch invalid receive attempts */ - + /* Catch invalid receive requests */ if (unlikely(!buf_len)) return -EINVAL; - if (sock->type == SOCK_SEQPACKET) { - if (unlikely(sock->state == SS_UNCONNECTED)) - return -ENOTCONN; - if (unlikely((sock->state == SS_DISCONNECTING) && - (skb_queue_len(&sock->sk->sk_receive_queue) == 0))) - return -ENOTCONN; - } - - /* Look for a message in receive queue; wait if necessary */ - - if (unlikely(down_interruptible(&tsock->sem))) - return -ERESTARTSYS; - -restart: - if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && - (flags & MSG_DONTWAIT))) { - res = -EWOULDBLOCK; - goto exit; - } + lock_sock(sk); - if ((res = wait_event_interruptible( - *sock->sk->sk_sleep, - ((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) || - (sock->state == SS_DISCONNECTING))) )) { + if (unlikely(sock->state == SS_UNCONNECTED)) { + res = -ENOTCONN; goto exit; } - /* Catch attempt to receive on an already terminated connection */ - /* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */ + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); +restart: - if (!q_len) { - res = -ENOTCONN; + /* Look for a message in receive queue; wait if necessary */ + res = tipc_wait_for_rcvmsg(sock, &timeo); + if (res) goto exit; - } - - /* Get access to first message in receive queue */ - buf = skb_peek(&sock->sk->sk_receive_queue); + /* 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); - /* Complete connection setup for an implied connect */ - - if (unlikely(sock->state == SS_CONNECTING)) { - if ((res = auto_connect(sock, tsock, msg))) - goto exit; - } - /* Discard an empty non-errored message & try again */ - if ((!sz) && (!err)) { - advance_queue(tsock); + advance_rx_queue(sk); goto restart; } /* Capture sender's address (optional) */ - set_orig_addr(m, msg); /* Capture ancillary data (optional) */ - - if ((res = anc_data_recv(m, msg, tsock->p))) + 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; } - if (unlikely(copy_to_user(m->msg_iov->iov_base, msg_data(msg), - sz))) { - res = -EFAULT; + 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) || @@ -923,19 +1102,19 @@ restart: } /* Consume received message (optional) */ - if (likely(!(flags & MSG_PEEK))) { - if (unlikely(++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) - tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked); - advance_queue(tsock); + if ((sock->state != SS_READY) && + (++port->conn_unacked >= TIPC_CONNACK_INTV)) + tipc_acknowledge(port->ref, port->conn_unacked); + advance_rx_queue(sk); } exit: - up(&tsock->sem); + release_sock(sk); return res; } /** - * recv_stream - receive stream-oriented data + * tipc_recv_stream - receive stream-oriented data * @iocb: (unused) * @m: descriptor for message info * @buf_len: total size of user buffer area @@ -946,109 +1125,82 @@ exit: * * Returns size of returned message data, errno otherwise */ - -static int recv_stream(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t buf_len, int flags) +static int tipc_recv_stream(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t buf_len, int flags) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + 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; - unsigned int q_len; + long timeo; unsigned int sz; - int sz_to_copy; + int sz_to_copy, target, needed; int sz_copied = 0; - int needed; - char __user *crs = m->msg_iov->iov_base; - unsigned char *buf_crs; u32 err; - int res; - - /* Currently doesn't support receiving into multiple iovec entries */ - - if (m->msg_iovlen != 1) - return -EOPNOTSUPP; + int res = 0; /* Catch invalid receive attempts */ - if (unlikely(!buf_len)) return -EINVAL; - if (unlikely(sock->state == SS_DISCONNECTING)) { - if (skb_queue_len(&sock->sk->sk_receive_queue) == 0) - return -ENOTCONN; - } else if (unlikely(sock->state != SS_CONNECTED)) - return -ENOTCONN; + lock_sock(sk); - /* Look for a message in receive queue; wait if necessary */ - - if (unlikely(down_interruptible(&tsock->sem))) - return -ERESTARTSYS; - -restart: - if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && - (flags & MSG_DONTWAIT))) { - res = -EWOULDBLOCK; - goto exit; - } - - if ((res = wait_event_interruptible( - *sock->sk->sk_sleep, - ((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) || - (sock->state == SS_DISCONNECTING))) )) { + if (unlikely(sock->state == SS_UNCONNECTED)) { + res = -ENOTCONN; goto exit; } - /* Catch attempt to receive on an already terminated connection */ - /* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */ + target = sock_rcvlowat(sk, flags & MSG_WAITALL, buf_len); + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); - if (!q_len) { - res = -ENOTCONN; +restart: + /* Look for a message in receive queue; wait if necessary */ + res = tipc_wait_for_rcvmsg(sock, &timeo); + if (res) goto exit; - } - /* Get access to first message in receive queue */ - - buf = skb_peek(&sock->sk->sk_receive_queue); + /* 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_queue(tsock); + 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); - if ((res = anc_data_recv(m, msg, tsock->p))) + res = anc_data_recv(m, msg, port); + if (res) goto exit; } /* Capture message data (if valid) & compute return value (always) */ - if (!err) { - buf_crs = (unsigned char *)(TIPC_SKB_CB(buf)->handle); - sz = skb_tail_pointer(buf) - buf_crs; + u32 offset = (u32)(unsigned long)(TIPC_SKB_CB(buf)->handle); + sz -= offset; needed = (buf_len - sz_copied); sz_to_copy = (sz <= needed) ? sz : needed; - if (unlikely(copy_to_user(crs, buf_crs, sz_to_copy))) { - res = -EFAULT; + + 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 = buf_crs + sz_to_copy; + TIPC_SKB_CB(buf)->handle = + (void *)(unsigned long)(offset + sz_to_copy); goto exit; } - - crs += sz_to_copy; } else { if (sz_copied != 0) goto exit; /* can't add error msg to valid data */ @@ -1060,421 +1212,588 @@ restart: } /* Consume received message (optional) */ - if (likely(!(flags & MSG_PEEK))) { - if (unlikely(++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) - tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked); - advance_queue(tsock); + 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 */ - && (flags & MSG_WAITALL) /* ... and need to wait for more */ - && (!(flags & MSG_PEEK)) /* ... and aren't just peeking at data */ - && (!err) /* ... and haven't reached a FIN */ - ) + 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: - up(&tsock->sem); + release_sock(sk); return sz_copied ? sz_copied : res; } /** - * queue_overloaded - test if queue overload condition exists - * @queue_size: current size of queue - * @base: nominal maximum size of queue - * @msg: message to be added to queue - * - * Returns 1 if queue is currently overloaded, 0 otherwise + * 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(); +} -static int queue_overloaded(u32 queue_size, u32 base, struct tipc_msg *msg) +/** + * 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) { - u32 threshold; - u32 imp = msg_importance(msg); - - if (imp == TIPC_LOW_IMPORTANCE) - threshold = base; - else if (imp == TIPC_MEDIUM_IMPORTANCE) - threshold = base * 2; - else if (imp == TIPC_HIGH_IMPORTANCE) - threshold = base * 100; - else - return 0; + struct sock *sk = &tsk->sk; + struct tipc_port *port = &tsk->port; + struct socket *sock = sk->sk_socket; + struct tipc_msg *msg = buf_msg(*buf); - if (msg_connected(msg)) - threshold *= 4; + 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; - return (queue_size > threshold); + 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; } /** - * async_disconnect - wrapper function used to disconnect port - * @portref: TIPC port reference (passed as pointer-sized value) + * 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 void async_disconnect(unsigned long portref) +static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *buf) { - tipc_disconnect((u32)portref); + 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); } /** - * dispatch - handle arriving message - * @tport: TIPC port that received message + * filter_rcv - validate incoming message + * @sk: socket * @buf: message * - * Called with port locked. Must not take socket lock to avoid deadlock risk. + * 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 dispatch(struct tipc_port *tport, struct sk_buff *buf) +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); - struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle; - struct socket *sock; - u32 recv_q_len; - - /* Reject message if socket is closing */ - - if (!tsock) - return TIPC_ERR_NO_PORT; + 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; - /* - * WOULD IT BE BETTER TO JUST DISCARD THESE MESSAGES INSTEAD? - * "NO PORT" ISN'T REALLY THE RIGHT ERROR CODE, AND THERE MAY - * BE SECURITY IMPLICATIONS INHERENT IN REJECTING INVALID TRAFFIC - */ - sock = tsock->sk.sk_socket; if (sock->state == SS_READY) { - if (msg_connected(msg)) { - msg_dbg(msg, "dispatch filter 1\n"); + if (msg_connected(msg)) return TIPC_ERR_NO_PORT; - } } else { - if (msg_mcast(msg)) { - msg_dbg(msg, "dispatch filter 2\n"); - return TIPC_ERR_NO_PORT; - } - if (sock->state == SS_CONNECTED) { - if (!msg_connected(msg)) { - msg_dbg(msg, "dispatch filter 3\n"); - return TIPC_ERR_NO_PORT; - } - } - else if (sock->state == SS_CONNECTING) { - if (!msg_connected(msg) && (msg_errcode(msg) == 0)) { - msg_dbg(msg, "dispatch filter 4\n"); - return TIPC_ERR_NO_PORT; - } - } - else if (sock->state == SS_LISTENING) { - if (msg_connected(msg) || msg_errcode(msg)) { - msg_dbg(msg, "dispatch filter 5\n"); - return TIPC_ERR_NO_PORT; - } - } - else if (sock->state == SS_DISCONNECTING) { - msg_dbg(msg, "dispatch filter 6\n"); - return TIPC_ERR_NO_PORT; - } - else /* (sock->state == SS_UNCONNECTED) */ { - if (msg_connected(msg) || msg_errcode(msg)) { - msg_dbg(msg, "dispatch filter 7\n"); - return TIPC_ERR_NO_PORT; - } - } + 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; - if (unlikely((u32)atomic_read(&tipc_queue_size) > - OVERLOAD_LIMIT_BASE)) { - if (queue_overloaded(atomic_read(&tipc_queue_size), - OVERLOAD_LIMIT_BASE, msg)) - return TIPC_ERR_OVERLOAD; - } - recv_q_len = skb_queue_len(&tsock->sk.sk_receive_queue); - if (unlikely(recv_q_len > (OVERLOAD_LIMIT_BASE / 2))) { - if (queue_overloaded(recv_q_len, - OVERLOAD_LIMIT_BASE / 2, msg)) - 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); - /* Initiate connection termination for an incoming 'FIN' */ + sk->sk_data_ready(sk); + return TIPC_OK; +} - if (unlikely(msg_errcode(msg) && (sock->state == SS_CONNECTED))) { - sock->state = SS_DISCONNECTING; - /* Note: Use signal since port lock is already taken! */ - tipc_k_signal((Handler)async_disconnect, tport->ref); - } +/** + * 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; - /* Enqueue message (finally!) */ + res = filter_rcv(sk, buf); + if (unlikely(res)) + tipc_reject_msg(buf, res); - msg_dbg(msg,"<DISP<: "); - TIPC_SKB_CB(buf)->handle = msg_data(msg); - atomic_inc(&tipc_queue_size); - skb_queue_tail(&sock->sk->sk_receive_queue, buf); + if (atomic_read(&tsk->dupl_rcvcnt) < TIPC_CONN_OVERLOAD_LIMIT) + atomic_add(truesize, &tsk->dupl_rcvcnt); - if (waitqueue_active(sock->sk->sk_sleep)) - wake_up_interruptible(sock->sk->sk_sleep); - return TIPC_OK; + return 0; } /** - * wakeupdispatch - wake up port after congestion - * @tport: port to wakeup - * - * Called with port lock on. + * 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 void wakeupdispatch(struct tipc_port *tport) +static int tipc_wait_for_connect(struct socket *sock, long *timeo_p) { - struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle; + struct sock *sk = sock->sk; + DEFINE_WAIT(wait); + int done; - if (waitqueue_active(tsock->sk.sk_sleep)) - wake_up_interruptible(tsock->sk.sk_sleep); + 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; } /** - * connect - establish a connection to another TIPC port + * 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: (unused) + * @flags: file-related flags associated with socket * * Returns 0 on success, errno otherwise */ - -static int connect(struct socket *sock, struct sockaddr *dest, int destlen, - int flags) +static int tipc_connect(struct socket *sock, struct sockaddr *dest, + int destlen, int flags) { - struct tipc_sock *tsock = tipc_sk(sock->sk); - struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest; - struct msghdr m = {NULL,}; - struct sk_buff *buf; - struct tipc_msg *msg; - int res; - - /* For now, TIPC does not allow use of connect() with DGRAM or RDM types */ - - if (sock->state == SS_READY) - return -EOPNOTSUPP; - - /* Issue Posix-compliant error code if socket is in the wrong state */ - - if (sock->state == SS_LISTENING) - return -EOPNOTSUPP; - if (sock->state == SS_CONNECTING) - return -EALREADY; - if (sock->state != SS_UNCONNECTED) - return -EISCONN; - - /* - * 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) - return -EINVAL; - - /* Send a 'SYN-' to destination */ - - m.msg_name = dest; - m.msg_namelen = destlen; - if ((res = send_msg(NULL, sock, &m, 0)) < 0) { - sock->state = SS_DISCONNECTING; - return res; - } - - if (down_interruptible(&tsock->sem)) - return -ERESTARTSYS; - - /* Wait for destination's 'ACK' response */ - - res = wait_event_interruptible_timeout(*sock->sk->sk_sleep, - skb_queue_len(&sock->sk->sk_receive_queue), - sock->sk->sk_rcvtimeo); - buf = skb_peek(&sock->sk->sk_receive_queue); - if (res > 0) { - msg = buf_msg(buf); - res = auto_connect(sock, tsock, msg); - if (!res) { - if (!msg_data_sz(msg)) - advance_queue(tsock); - } - } else { - if (res == 0) { - res = -ETIMEDOUT; - } else - { /* leave "res" unchanged */ } - sock->state = SS_DISCONNECTING; - } - - up(&tsock->sem); - return res; + 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; } /** - * listen - allow socket to listen for incoming connections + * tipc_listen - allow socket to listen for incoming connections * @sock: socket structure * @len: (unused) * * Returns 0 on success, errno otherwise */ - -static int listen(struct socket *sock, int len) +static int tipc_listen(struct socket *sock, int len) { - /* REQUIRES SOCKET LOCKING OF SOME SORT? */ + struct sock *sk = sock->sk; + int res; + + lock_sock(sk); - if (sock->state == SS_READY) - return -EOPNOTSUPP; if (sock->state != SS_UNCONNECTED) - return -EINVAL; - sock->state = SS_LISTENING; - return 0; + 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; } /** - * accept - wait for connection request + * 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 accept(struct socket *sock, struct socket *newsock, int flags) +static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *new_sk, *sk = sock->sk; struct sk_buff *buf; - int res = -EFAULT; + struct tipc_port *new_port; + struct tipc_msg *msg; + struct tipc_portid peer; + u32 new_ref; + long timeo; + int res; - if (sock->state == SS_READY) - return -EOPNOTSUPP; - if (sock->state != SS_LISTENING) - return -EINVAL; + lock_sock(sk); - if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && - (flags & O_NONBLOCK))) - return -EWOULDBLOCK; + 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; - if (down_interruptible(&tsock->sem)) - return -ERESTARTSYS; + buf = skb_peek(&sk->sk_receive_queue); - if (wait_event_interruptible(*sock->sk->sk_sleep, - skb_queue_len(&sock->sk->sk_receive_queue))) { - res = -ERESTARTSYS; + res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, 1); + if (res) goto exit; - } - buf = skb_peek(&sock->sk->sk_receive_queue); - - res = tipc_create(sock->sk->sk_net, newsock, 0); - if (!res) { - struct tipc_sock *new_tsock = tipc_sk(newsock->sk); - struct tipc_portid id; - struct tipc_msg *msg = buf_msg(buf); - u32 new_ref = new_tsock->p->ref; - - id.ref = msg_origport(msg); - id.node = msg_orignode(msg); - tipc_connect2port(new_ref, &id); - newsock->state = SS_CONNECTED; - - tipc_set_portimportance(new_ref, msg_importance(msg)); - if (msg_named(msg)) { - new_tsock->p->conn_type = msg_nametype(msg); - new_tsock->p->conn_instance = msg_nameinst(msg); - } - /* - * Respond to 'SYN-' by discarding it & returning 'ACK'-. - * Respond to 'SYN+' by queuing it on new socket. - */ + new_sk = new_sock->sk; + new_port = &tipc_sk(new_sk)->port; + new_ref = new_port->ref; + msg = buf_msg(buf); - msg_dbg(msg,"<ACC<: "); - if (!msg_data_sz(msg)) { - struct msghdr m = {NULL,}; + /* we lock on new_sk; but lockdep sees the lock on sk */ + lock_sock_nested(new_sk, SINGLE_DEPTH_NESTING); - send_packet(NULL, newsock, &m, 0); - advance_queue(tsock); - } else { - sock_lock(tsock); - skb_dequeue(&sock->sk->sk_receive_queue); - sock_unlock(tsock); - skb_queue_head(&newsock->sk->sk_receive_queue, buf); - } + /* + * 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: - up(&tsock->sem); + release_sock(sk); return res; } /** - * shutdown - shutdown socket connection + * tipc_shutdown - shutdown socket connection * @sock: socket structure - * @how: direction to close (unused; always treated as read + write) + * @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 shutdown(struct socket *sock, int how) +static int tipc_shutdown(struct socket *sock, int how) { - struct tipc_sock* tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; struct sk_buff *buf; int res; - /* Could return -EINVAL for an invalid "how", but why bother? */ - - if (down_interruptible(&tsock->sem)) - return -ERESTARTSYS; + if (how != SHUT_RDWR) + return -EINVAL; - sock_lock(tsock); + lock_sock(sk); switch (sock->state) { + case SS_CONNECTING: case SS_CONNECTED: - /* Send 'FIN+' or 'FIN-' message to peer */ - - sock_unlock(tsock); restart: - if ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { - atomic_dec(&tipc_queue_size); - if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf))) { - buf_discard(buf); + /* 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); } - else { - tipc_shutdown(tsock->p->ref); - } - sock_lock(tsock); + + sock->state = SS_DISCONNECTING; /* fall through */ case SS_DISCONNECTING: /* Discard any unreceived messages */ + __skb_queue_purge(&sk->sk_receive_queue); - while ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { - atomic_dec(&tipc_queue_size); - buf_discard(buf); - } - tsock->p->conn_unacked = 0; - - /* fall through */ - - case SS_CONNECTING: - sock->state = SS_DISCONNECTING; + /* Wake up anyone sleeping in poll */ + sk->sk_state_change(sk); res = 0; break; @@ -1482,14 +1801,12 @@ restart: res = -ENOTCONN; } - sock_unlock(tsock); - - up(&tsock->sem); + release_sock(sk); return res; } /** - * setsockopt - set socket option + * tipc_setsockopt - set socket option * @sock: socket structure * @lvl: option level * @opt: option identifier @@ -1501,11 +1818,12 @@ restart: * * Returns 0 on success, errno otherwise */ - -static int setsockopt(struct socket *sock, - int lvl, int opt, char __user *ov, int ol) +static int tipc_setsockopt(struct socket *sock, int lvl, int opt, + char __user *ov, unsigned int ol) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; u32 value; int res; @@ -1515,38 +1833,40 @@ static int setsockopt(struct socket *sock, return -ENOPROTOOPT; if (ol < sizeof(value)) return -EINVAL; - if ((res = get_user(value, (u32 __user *)ov))) + res = get_user(value, (u32 __user *)ov); + if (res) return res; - if (down_interruptible(&tsock->sem)) - return -ERESTARTSYS; + lock_sock(sk); switch (opt) { case TIPC_IMPORTANCE: - res = tipc_set_portimportance(tsock->p->ref, value); + tipc_port_set_importance(port, value); break; case TIPC_SRC_DROPPABLE: if (sock->type != SOCK_STREAM) - res = tipc_set_portunreliable(tsock->p->ref, value); + tipc_port_set_unreliable(port, value); else res = -ENOPROTOOPT; break; case TIPC_DEST_DROPPABLE: - res = tipc_set_portunreturnable(tsock->p->ref, value); + tipc_port_set_unreturnable(port, value); break; case TIPC_CONN_TIMEOUT: - sock->sk->sk_rcvtimeo = (value * HZ / 1000); + tipc_sk(sk)->conn_timeout = value; + /* no need to set "res", since already 0 at this point */ break; default: res = -EINVAL; } - up(&tsock->sem); + release_sock(sk); + return res; } /** - * getsockopt - get socket option + * tipc_getsockopt - get socket option * @sock: socket structure * @lvl: option level * @opt: option identifier @@ -1558,11 +1878,12 @@ static int setsockopt(struct socket *sock, * * Returns 0 on success, errno otherwise */ - -static int getsockopt(struct socket *sock, - int lvl, int opt, char __user *ov, int __user *ol) +static int tipc_getsockopt(struct socket *sock, int lvl, int opt, + char __user *ov, int __user *ol) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; int len; u32 value; int res; @@ -1571,123 +1892,154 @@ static int getsockopt(struct socket *sock, return put_user(0, ol); if (lvl != SOL_TIPC) return -ENOPROTOOPT; - if ((res = get_user(len, ol))) + res = get_user(len, ol); + if (res) return res; - if (down_interruptible(&tsock->sem)) - return -ERESTARTSYS; + lock_sock(sk); switch (opt) { case TIPC_IMPORTANCE: - res = tipc_portimportance(tsock->p->ref, &value); + value = tipc_port_importance(port); break; case TIPC_SRC_DROPPABLE: - res = tipc_portunreliable(tsock->p->ref, &value); + value = tipc_port_unreliable(port); break; case TIPC_DEST_DROPPABLE: - res = tipc_portunreturnable(tsock->p->ref, &value); + value = tipc_port_unreturnable(port); break; case TIPC_CONN_TIMEOUT: - value = (sock->sk->sk_rcvtimeo * 1000) / HZ; + 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; } - if (res) { - /* "get" failed */ - } - else if (len < sizeof(value)) { - res = -EINVAL; - } - else if ((res = copy_to_user(ov, &value, sizeof(value)))) { - /* couldn't return value */ - } - else { - res = put_user(sizeof(value), ol); - } + release_sock(sk); - up(&tsock->sem); - return res; + 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); } -/** - * Protocol switches for the various types of TIPC sockets - */ +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, + .owner = THIS_MODULE, .family = AF_TIPC, - .release = release, - .bind = bind, - .connect = connect, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_connect, .socketpair = sock_no_socketpair, - .accept = accept, - .getname = get_name, - .poll = poll, - .ioctl = sock_no_ioctl, - .listen = listen, - .shutdown = shutdown, - .setsockopt = setsockopt, - .getsockopt = getsockopt, - .sendmsg = send_msg, - .recvmsg = recv_msg, + .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, + .owner = THIS_MODULE, .family = AF_TIPC, - .release = release, - .bind = bind, - .connect = connect, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_connect, .socketpair = sock_no_socketpair, - .accept = accept, - .getname = get_name, - .poll = poll, - .ioctl = sock_no_ioctl, - .listen = listen, - .shutdown = shutdown, - .setsockopt = setsockopt, - .getsockopt = getsockopt, - .sendmsg = send_packet, - .recvmsg = recv_msg, + .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, + .owner = THIS_MODULE, .family = AF_TIPC, - .release = release, - .bind = bind, - .connect = connect, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_connect, .socketpair = sock_no_socketpair, - .accept = accept, - .getname = get_name, - .poll = poll, - .ioctl = sock_no_ioctl, - .listen = listen, - .shutdown = shutdown, - .setsockopt = setsockopt, - .getsockopt = getsockopt, - .sendmsg = send_stream, - .recvmsg = recv_stream, + .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, + .owner = THIS_MODULE, .family = AF_TIPC, - .create = tipc_create + .create = tipc_sk_create }; static struct proto tipc_proto = { .name = "TIPC", .owner = THIS_MODULE, - .obj_size = sizeof(struct tipc_sock) + .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 }; /** @@ -1701,18 +2053,16 @@ int tipc_socket_init(void) res = proto_register(&tipc_proto, 1); if (res) { - err("Failed to register TIPC protocol type\n"); + pr_err("Failed to register TIPC protocol type\n"); goto out; } res = sock_register(&tipc_family_ops); if (res) { - err("Failed to register TIPC socket type\n"); + pr_err("Failed to register TIPC socket type\n"); proto_unregister(&tipc_proto); goto out; } - - sockets_enabled = 1; out: return res; } @@ -1722,11 +2072,6 @@ int tipc_socket_init(void) */ void tipc_socket_stop(void) { - if (!sockets_enabled) - return; - - sockets_enabled = 0; sock_unregister(tipc_family_ops.family); proto_unregister(&tipc_proto); } - diff --git a/net/tipc/zone.h b/net/tipc/socket.h index 6e7a08df8af..3afcd2a70b3 100644 --- a/net/tipc/zone.h +++ b/net/tipc/socket.h @@ -1,8 +1,6 @@ -/* - * net/tipc/zone.h: Include file for TIPC zone management routines +/* net/tipc/socket.h: Include file for TIPC socket code * - * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005-2006, Wind River Systems + * Copyright (c) 2014, Ericsson AB * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -34,38 +32,43 @@ * POSSIBILITY OF SUCH DAMAGE. */ -#ifndef _TIPC_ZONE_H -#define _TIPC_ZONE_H - -#include "node_subscr.h" -#include "net.h" +#ifndef _TIPC_SOCK_H +#define _TIPC_SOCK_H +#include "port.h" +#include <net/sock.h> /** - * struct _zone - TIPC zone structure - * @addr: network address of zone - * @clusters: array of pointers to all clusters within zone - * @links: number of (unicast) links to zone + * 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 _zone { - u32 addr; - struct cluster *clusters[2]; /* currently limited to just 1 cluster */ - u32 links; +struct tipc_sock { + struct sock sk; + struct tipc_port port; + unsigned int conn_timeout; + atomic_t dupl_rcvcnt; }; -struct node *tipc_zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref); -u32 tipc_zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref); -void tipc_zone_remove_as_router(struct _zone *z_ptr, u32 router); -void tipc_zone_send_external_routes(struct _zone *z_ptr, u32 dest); -struct _zone *tipc_zone_create(u32 addr); -void tipc_zone_delete(struct _zone *z_ptr); -void tipc_zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr); -u32 tipc_zone_next_node(u32 addr); +static inline struct tipc_sock *tipc_sk(const struct sock *sk) +{ + return container_of(sk, struct tipc_sock, sk); +} -static inline struct _zone *tipc_zone_find(u32 addr) +static inline struct tipc_sock *tipc_port_to_sock(const struct tipc_port *port) { - return tipc_net.zones[tipc_zone(addr)]; + 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 index 8c01ccd3626..642437231ad 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -1,8 +1,8 @@ /* - * net/tipc/subscr.c: TIPC subscription service + * net/tipc/subscr.c: TIPC network topology service * * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005-2007, 2010-2013, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -35,48 +35,48 @@ */ #include "core.h" -#include "dbg.h" -#include "subscr.h" #include "name_table.h" -#include "ref.h" +#include "port.h" +#include "subscr.h" /** - * struct subscriber - TIPC network topology subscriber - * @ref: object reference to subscriber object itself - * @lock: pointer to spinlock controlling access to subscriber object - * @subscriber_list: adjacent subscribers in top. server's list of subscribers + * 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 - * @port_ref: object reference to port used to communicate with subscriber - * @swap: indicates if subscriber uses opposite endianness in its messages */ - -struct subscriber { - u32 ref; - spinlock_t *lock; - struct list_head subscriber_list; +struct tipc_subscriber { + int conid; + spinlock_t lock; struct list_head subscription_list; - u32 port_ref; - int swap; }; -/** - * struct top_srv - TIPC network topology subscription service - * @user_ref: TIPC userid of subscription service - * @setup_port: reference to TIPC port that handles subscription requests - * @subscription_count: number of active subscriptions (not subscribers!) - * @subscriber_list: list of ports subscribing to service - * @lock: spinlock govering access to subscriber 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); -struct top_srv { - u32 user_ref; - u32 setup_port; - atomic_t subscription_count; - struct list_head subscriber_list; - spinlock_t lock; +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 top_srv topsrv = { 0 }; +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 @@ -85,36 +85,27 @@ static struct top_srv topsrv = { 0 }; * * Returns converted value */ - static u32 htohl(u32 in, int swap) { - char *c = (char *)∈ - - return swap ? ((c[3] << 3) + (c[2] << 2) + (c[1] << 1) + c[0]) : in; + return swap ? swab32(in) : in; } -/** - * subscr_send_event - send a message containing a tipc_event to the subscriber - */ - -static void subscr_send_event(struct subscription *sub, - u32 found_lower, - u32 found_upper, - u32 event, - u32 port_ref, +static void subscr_send_event(struct tipc_subscription *sub, u32 found_lower, + u32 found_upper, u32 event, u32 port_ref, u32 node) { - struct iovec msg_sect; + 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->owner->swap); - sub->evt.found_lower = htohl(found_lower, sub->owner->swap); - sub->evt.found_upper = htohl(found_upper, sub->owner->swap); - sub->evt.port.ref = htohl(port_ref, sub->owner->swap); - sub->evt.port.node = htohl(node, sub->owner->swap); - tipc_send(sub->owner->port_ref, 1, &msg_sect); + 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); } /** @@ -122,11 +113,8 @@ static void subscr_send_event(struct subscription *sub, * * Returns 1 if there is overlap, otherwise 0. */ - -int tipc_subscr_overlap(struct subscription *sub, - u32 found_lower, +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; @@ -142,153 +130,113 @@ int tipc_subscr_overlap(struct subscription *sub, * * Protected by nameseq.lock in name_table.c */ - -void tipc_subscr_report_overlap(struct subscription *sub, - u32 found_lower, - u32 found_upper, - u32 event, - u32 port_ref, - u32 node, - int must) +void tipc_subscr_report_overlap(struct tipc_subscription *sub, u32 found_lower, + u32 found_upper, u32 event, u32 port_ref, + u32 node, int must) { - dbg("Rep overlap %u:%u,%u<->%u,%u\n", sub->seq.type, sub->seq.lower, - sub->seq.upper, found_lower, found_upper); 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); } -/** - * subscr_timeout - subscription timeout has occurred - */ - -static void subscr_timeout(struct subscription *sub) +static void subscr_timeout(struct tipc_subscription *sub) { - struct subscriber *subscriber; - u32 subscriber_ref; - - /* Validate subscriber reference (in case subscriber is terminating) */ + struct tipc_subscriber *subscriber = sub->subscriber; - subscriber_ref = sub->owner->ref; - subscriber = (struct subscriber *)tipc_ref_lock(subscriber_ref); - if (subscriber == NULL) - return; + /* 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) { - tipc_ref_unlock(subscriber_ref); + spin_unlock_bh(&subscriber->lock); return; } /* Unlink subscription from name table */ - tipc_nametbl_unsubscribe(sub); - /* Notify subscriber of timeout, then unlink subscription */ - - subscr_send_event(sub, - sub->evt.s.seq.lower, - sub->evt.s.seq.upper, - TIPC_SUBSCR_TIMEOUT, - 0, - 0); + /* Unlink subscription from subscriber */ list_del(&sub->subscription_list); - /* Now destroy subscription */ + spin_unlock_bh(&subscriber->lock); - tipc_ref_unlock(subscriber_ref); + /* 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(&topsrv.subscription_count); + atomic_dec(&subscription_count); } /** * subscr_del - delete a subscription within a subscription list * - * Called with subscriber locked. + * Called with subscriber lock held. */ - -static void subscr_del(struct subscription *sub) +static void subscr_del(struct tipc_subscription *sub) { tipc_nametbl_unsubscribe(sub); list_del(&sub->subscription_list); kfree(sub); - atomic_dec(&topsrv.subscription_count); + atomic_dec(&subscription_count); } /** * subscr_terminate - terminate communication with a subscriber * - * Called with subscriber locked. Routine must temporarily release this lock - * to enable subscription timeout routine(s) to finish without deadlocking; - * the lock is then reclaimed to allow caller to release it upon return. - * (This should work even in the unlikely event some other thread creates - * a new object reference in the interim that uses this lock; this routine will - * simply wait for it to be released, then claim it.) + * Note: Must call it in process context since it might sleep. */ - -static void subscr_terminate(struct subscriber *subscriber) +static void subscr_terminate(struct tipc_subscriber *subscriber) { - struct subscription *sub; - struct subscription *sub_temp; + tipc_conn_terminate(&topsrv, subscriber->conid); +} - /* Invalidate subscriber reference */ +static void subscr_release(struct tipc_subscriber *subscriber) +{ + struct tipc_subscription *sub; + struct tipc_subscription *sub_temp; - tipc_ref_discard(subscriber->ref); - spin_unlock_bh(subscriber->lock); + 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); } - dbg("Term: Removing sub %u,%u,%u from subscriber %x list\n", - sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber); subscr_del(sub); } - - /* Sever connection to subscriber */ - - tipc_shutdown(subscriber->port_ref); - tipc_deleteport(subscriber->port_ref); - - /* Remove subscriber from topology server's subscriber list */ - - spin_lock_bh(&topsrv.lock); - list_del(&subscriber->subscriber_list); - spin_unlock_bh(&topsrv.lock); + spin_unlock_bh(&subscriber->lock); /* Now destroy subscriber */ - - spin_lock_bh(subscriber->lock); kfree(subscriber); } /** * subscr_cancel - handle subscription cancellation request * - * Called with subscriber locked. Routine must temporarily release this lock + * 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 subscriber *subscriber) + struct tipc_subscriber *subscriber) { - struct subscription *sub; - struct subscription *sub_temp; + 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))) { @@ -300,295 +248,128 @@ static void subscr_cancel(struct tipc_subscr *s, 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); + spin_unlock_bh(&subscriber->lock); k_cancel_timer(&sub->timer); k_term_timer(&sub->timer); - spin_lock_bh(subscriber->lock); + spin_lock_bh(&subscriber->lock); } - dbg("Cancel: removing sub %u,%u,%u from subscriber %x list\n", - sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber); subscr_del(sub); } /** * subscr_subscribe - create subscription for subscriber * - * Called with subscriber locked + * 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; -static void subscr_subscribe(struct tipc_subscr *s, - struct subscriber *subscriber) -{ - struct subscription *sub; - - /* Determine/update subscriber's endianness */ - - if (s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE)) - subscriber->swap = 0; - else - subscriber->swap = 1; + /* 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, subscriber->swap)) { - s->filter &= ~htohl(TIPC_SUB_CANCEL, subscriber->swap); + if (s->filter & htohl(TIPC_SUB_CANCEL, swap)) { + s->filter &= ~htohl(TIPC_SUB_CANCEL, swap); subscr_cancel(s, subscriber); - return; + return 0; } /* Refuse subscription if global limit exceeded */ - - if (atomic_read(&topsrv.subscription_count) >= tipc_max_subscriptions) { - warn("Subscription rejected, subscription limit reached (%u)\n", - tipc_max_subscriptions); - subscr_terminate(subscriber); - return; + 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 = kzalloc(sizeof(*sub), GFP_ATOMIC); + sub = kmalloc(sizeof(*sub), GFP_ATOMIC); if (!sub) { - warn("Subscription rejected, no memory\n"); - subscr_terminate(subscriber); - return; + pr_warn("Subscription rejected, no memory\n"); + return -ENOMEM; } /* Initialize subscription object */ - - sub->seq.type = htohl(s->seq.type, subscriber->swap); - sub->seq.lower = htohl(s->seq.lower, subscriber->swap); - sub->seq.upper = htohl(s->seq.upper, subscriber->swap); - sub->timeout = htohl(s->timeout, subscriber->swap); - sub->filter = htohl(s->filter, subscriber->swap); - if ((!(sub->filter & TIPC_SUB_PORTS) - == !(sub->filter & TIPC_SUB_SERVICE)) - || (sub->seq.lower > sub->seq.upper)) { - warn("Subscription rejected, illegal request\n"); + 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); - subscr_terminate(subscriber); - return; + return -EINVAL; } - memcpy(&sub->evt.s, s, sizeof(struct tipc_subscr)); - INIT_LIST_HEAD(&sub->subscription_list); INIT_LIST_HEAD(&sub->nameseq_list); list_add(&sub->subscription_list, &subscriber->subscription_list); - atomic_inc(&topsrv.subscription_count); + 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->owner = subscriber; - tipc_nametbl_subscribe(sub); + *sub_p = sub; + return 0; } -/** - * subscr_conn_shutdown_event - handle termination request from subscriber - */ - -static void subscr_conn_shutdown_event(void *usr_handle, - u32 portref, - struct sk_buff **buf, - unsigned char const *data, - unsigned int size, - int reason) +/* Handle one termination request for the subscriber */ +static void subscr_conn_shutdown_event(int conid, void *usr_data) { - struct subscriber *subscriber; - spinlock_t *subscriber_lock; - - subscriber = tipc_ref_lock((u32)(unsigned long)usr_handle); - if (subscriber == NULL) - return; - - subscriber_lock = subscriber->lock; - subscr_terminate(subscriber); - spin_unlock_bh(subscriber_lock); + subscr_release((struct tipc_subscriber *)usr_data); } -/** - * subscr_conn_msg_event - handle new subscription request from subscriber - */ - -static void subscr_conn_msg_event(void *usr_handle, - u32 port_ref, - struct sk_buff **buf, - const unchar *data, - u32 size) +/* 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 subscriber *subscriber; - spinlock_t *subscriber_lock; + struct tipc_subscriber *subscriber = usr_data; + struct tipc_subscription *sub = NULL; - subscriber = tipc_ref_lock((u32)(unsigned long)usr_handle); - if (subscriber == NULL) - return; - - subscriber_lock = subscriber->lock; - if (size != sizeof(struct tipc_subscr)) + spin_lock_bh(&subscriber->lock); + if (subscr_subscribe((struct tipc_subscr *)buf, subscriber, &sub) < 0) { + spin_unlock_bh(&subscriber->lock); subscr_terminate(subscriber); - else - subscr_subscribe((struct tipc_subscr *)data, subscriber); - - spin_unlock_bh(subscriber_lock); + return; + } + if (sub) + tipc_nametbl_subscribe(sub); + spin_unlock_bh(&subscriber->lock); } -/** - * subscr_named_msg_event - handle request to establish a new subscriber - */ -static void subscr_named_msg_event(void *usr_handle, - u32 port_ref, - struct sk_buff **buf, - const unchar *data, - u32 size, - u32 importance, - struct tipc_portid const *orig, - struct tipc_name_seq const *dest) +/* Handle one request to establish a new subscriber */ +static void *subscr_named_msg_event(int conid) { - struct subscriber *subscriber; - struct iovec msg_sect = {NULL, 0}; - spinlock_t *subscriber_lock; - - dbg("subscr_named_msg_event: orig = %x own = %x,\n", - orig->node, tipc_own_addr); - if (size && (size != sizeof(struct tipc_subscr))) { - warn("Subscriber rejected, invalid subscription size\n"); - return; - } + struct tipc_subscriber *subscriber; /* Create subscriber object */ - - subscriber = kzalloc(sizeof(struct subscriber), GFP_ATOMIC); + subscriber = kzalloc(sizeof(struct tipc_subscriber), GFP_ATOMIC); if (subscriber == NULL) { - warn("Subscriber rejected, no memory\n"); - return; + pr_warn("Subscriber rejected, no memory\n"); + return NULL; } INIT_LIST_HEAD(&subscriber->subscription_list); - INIT_LIST_HEAD(&subscriber->subscriber_list); - subscriber->ref = tipc_ref_acquire(subscriber, &subscriber->lock); - if (subscriber->ref == 0) { - warn("Subscriber rejected, reference table exhausted\n"); - kfree(subscriber); - return; - } - - /* Establish a connection to subscriber */ - - tipc_createport(topsrv.user_ref, - (void *)(unsigned long)subscriber->ref, - importance, - NULL, - NULL, - subscr_conn_shutdown_event, - NULL, - NULL, - subscr_conn_msg_event, - NULL, - &subscriber->port_ref); - if (subscriber->port_ref == 0) { - warn("Subscriber rejected, unable to create port\n"); - tipc_ref_discard(subscriber->ref); - kfree(subscriber); - return; - } - tipc_connect2port(subscriber->port_ref, orig); - - - /* Add subscriber to topology server's subscriber list */ - - tipc_ref_lock(subscriber->ref); - spin_lock_bh(&topsrv.lock); - list_add(&subscriber->subscriber_list, &topsrv.subscriber_list); - spin_unlock_bh(&topsrv.lock); - - /* - * Subscribe now if message contains a subscription, - * otherwise send an empty response to complete connection handshaking - */ + subscriber->conid = conid; + spin_lock_init(&subscriber->lock); - subscriber_lock = subscriber->lock; - if (size) - subscr_subscribe((struct tipc_subscr *)data, subscriber); - else - tipc_send(subscriber->port_ref, 1, &msg_sect); - - spin_unlock_bh(subscriber_lock); + return (void *)subscriber; } int tipc_subscr_start(void) { - struct tipc_name_seq seq = {TIPC_TOP_SRV, TIPC_TOP_SRV, TIPC_TOP_SRV}; - int res = -1; - - memset(&topsrv, 0, sizeof (topsrv)); - spin_lock_init(&topsrv.lock); - INIT_LIST_HEAD(&topsrv.subscriber_list); - - spin_lock_bh(&topsrv.lock); - res = tipc_attach(&topsrv.user_ref, NULL, NULL); - if (res) { - spin_unlock_bh(&topsrv.lock); - return res; - } - - res = tipc_createport(topsrv.user_ref, - NULL, - TIPC_CRITICAL_IMPORTANCE, - NULL, - NULL, - NULL, - NULL, - subscr_named_msg_event, - NULL, - NULL, - &topsrv.setup_port); - if (res) - goto failed; - - res = tipc_nametbl_publish_rsv(topsrv.setup_port, TIPC_NODE_SCOPE, &seq); - if (res) - goto failed; - - spin_unlock_bh(&topsrv.lock); - return 0; - -failed: - err("Failed to create subscription service\n"); - tipc_detach(topsrv.user_ref); - topsrv.user_ref = 0; - spin_unlock_bh(&topsrv.lock); - return res; + return tipc_server_start(&topsrv); } void tipc_subscr_stop(void) { - struct subscriber *subscriber; - struct subscriber *subscriber_temp; - spinlock_t *subscriber_lock; - - if (topsrv.user_ref) { - tipc_deleteport(topsrv.setup_port); - list_for_each_entry_safe(subscriber, subscriber_temp, - &topsrv.subscriber_list, - subscriber_list) { - tipc_ref_lock(subscriber->ref); - subscriber_lock = subscriber->lock; - subscr_terminate(subscriber); - spin_unlock_bh(subscriber_lock); - } - tipc_detach(topsrv.user_ref); - topsrv.user_ref = 0; - } + tipc_server_stop(&topsrv); } - - -int tipc_ispublished(struct tipc_name const *name) -{ - u32 domain = 0; - - return(tipc_nametbl_translate(name->type, name->instance,&domain) != 0); -} - diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h index 93a8e674fac..393e417bee3 100644 --- a/net/tipc/subscr.h +++ b/net/tipc/subscr.h @@ -1,8 +1,8 @@ /* - * net/tipc/subscr.h: Include file for TIPC subscription service + * net/tipc/subscr.h: Include file for TIPC network topology service * * Copyright (c) 2003-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems + * Copyright (c) 2005-2007, 2012-2013, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -37,44 +37,45 @@ #ifndef _TIPC_SUBSCR_H #define _TIPC_SUBSCR_H +#include "server.h" + +struct tipc_subscription; +struct tipc_subscriber; + /** - * struct subscription - TIPC network topology subscription object + * 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 - * @evt: template for events generated by subscription - * @subscription_list: adjacent subscriptions in subscriber's subscription list + * @timer: timer governing subscription duration (optional) * @nameseq_list: adjacent subscriptions in name sequence's subscription list - * @timer_ref: reference to timer governing subscription duration (may be NULL) - * @owner: pointer to subscriber object associated with this subscription + * @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 subscription { +struct tipc_subscription { + struct tipc_subscriber *subscriber; struct tipc_name_seq seq; u32 timeout; u32 filter; - struct tipc_event evt; - struct list_head subscription_list; - struct list_head nameseq_list; struct timer_list timer; - struct subscriber *owner; + struct list_head nameseq_list; + struct list_head subscription_list; + int swap; + struct tipc_event evt; }; -int tipc_subscr_overlap(struct subscription * sub, - u32 found_lower, +int tipc_subscr_overlap(struct tipc_subscription *sub, u32 found_lower, u32 found_upper); -void tipc_subscr_report_overlap(struct subscription * sub, - u32 found_lower, - u32 found_upper, - u32 event, - u32 port_ref, - u32 node, - int must_report); +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/dbg.h b/net/tipc/sysctl.c index c01b085000e..f3fef93325a 100644 --- a/net/tipc/dbg.h +++ b/net/tipc/sysctl.c @@ -1,8 +1,7 @@ /* - * net/tipc/dbg.h: Include file for TIPC print buffer routines + * net/tipc/sysctl.c: sysctl interface to TIPC subsystem * - * Copyright (c) 1997-2006, Ericsson AB - * Copyright (c) 2005-2006, Wind River Systems + * Copyright (c) 2013, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -34,37 +33,32 @@ * POSSIBILITY OF SUCH DAMAGE. */ -#ifndef _TIPC_DBG_H -#define _TIPC_DBG_H +#include "core.h" -/** - * struct print_buf - TIPC print buffer structure - * @buf: pointer to character array containing print buffer contents - * @size: size of character array - * @crs: pointer to first unused space in character array (i.e. final NUL) - * @next: used to link print buffers when printing to more than one at a time - */ - -struct print_buf { - char *buf; - u32 size; - char *crs; - struct print_buf *next; -}; +#include <linux/sysctl.h> -#define TIPC_PB_MIN_SIZE 64 /* minimum size for a print buffer's array */ -#define TIPC_PB_MAX_STR 512 /* max printable string (with trailing NUL) */ +static struct ctl_table_header *tipc_ctl_hdr; -void tipc_printbuf_init(struct print_buf *pb, char *buf, u32 size); -void tipc_printbuf_reset(struct print_buf *pb); -int tipc_printbuf_empty(struct print_buf *pb); -int tipc_printbuf_validate(struct print_buf *pb); -void tipc_printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from); - -void tipc_log_reinit(int log_size); -void tipc_log_stop(void); +static struct ctl_table tipc_table[] = { + { + .procname = "tipc_rmem", + .data = &sysctl_tipc_rmem, + .maxlen = sizeof(sysctl_tipc_rmem), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + {} +}; -struct sk_buff *tipc_log_resize(const void *req_tlv_area, int req_tlv_space); -struct sk_buff *tipc_log_dump(void); +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; +} -#endif +void tipc_unregister_sysctl(void) +{ + unregister_net_sysctl_table(tipc_ctl_hdr); +} diff --git a/net/tipc/user_reg.c b/net/tipc/user_reg.c deleted file mode 100644 index 4146c40cd20..00000000000 --- a/net/tipc/user_reg.c +++ /dev/null @@ -1,264 +0,0 @@ -/* - * net/tipc/user_reg.c: TIPC user registry code - * - * 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. - */ - -#include "core.h" -#include "user_reg.h" - -/* - * TIPC user registry keeps track of users of the tipc_port interface. - * - * The registry utilizes an array of "TIPC user" entries; - * a user's ID is the index of their associated array entry. - * Array entry 0 is not used, so userid 0 is not valid; - * TIPC sometimes uses this value to denote an anonymous user. - * The list of free entries is initially chained from last entry to entry 1. - */ - -/** - * struct tipc_user - registered TIPC user info - * @next: index of next free registry entry (or -1 for an allocated entry) - * @callback: ptr to routine to call when TIPC mode changes (NULL if none) - * @usr_handle: user-defined value passed to callback routine - * @ports: list of user ports owned by the user - */ - -struct tipc_user { - int next; - tipc_mode_event callback; - void *usr_handle; - struct list_head ports; -}; - -#define MAX_USERID 64 -#define USER_LIST_SIZE ((MAX_USERID + 1) * sizeof(struct tipc_user)) - -static struct tipc_user *users = NULL; -static u32 next_free_user = MAX_USERID + 1; -static DEFINE_SPINLOCK(reg_lock); - -/** - * reg_init - create TIPC user registry (but don't activate it) - * - * If registry has been pre-initialized it is left "as is". - * NOTE: This routine may be called when TIPC is inactive. - */ - -static int reg_init(void) -{ - u32 i; - - spin_lock_bh(®_lock); - if (!users) { - users = kzalloc(USER_LIST_SIZE, GFP_ATOMIC); - if (users) { - for (i = 1; i <= MAX_USERID; i++) { - users[i].next = i - 1; - } - next_free_user = MAX_USERID; - } - } - spin_unlock_bh(®_lock); - return users ? TIPC_OK : -ENOMEM; -} - -/** - * reg_callback - inform TIPC user about current operating mode - */ - -static void reg_callback(struct tipc_user *user_ptr) -{ - tipc_mode_event cb; - void *arg; - - spin_lock_bh(®_lock); - cb = user_ptr->callback; - arg = user_ptr->usr_handle; - spin_unlock_bh(®_lock); - - if (cb) - cb(arg, tipc_mode, tipc_own_addr); -} - -/** - * tipc_reg_start - activate TIPC user registry - */ - -int tipc_reg_start(void) -{ - u32 u; - int res; - - if ((res = reg_init())) - return res; - - for (u = 1; u <= MAX_USERID; u++) { - if (users[u].callback) - tipc_k_signal((Handler)reg_callback, - (unsigned long)&users[u]); - } - return TIPC_OK; -} - -/** - * tipc_reg_stop - shut down & delete TIPC user registry - */ - -void tipc_reg_stop(void) -{ - int id; - - if (!users) - return; - - for (id = 1; id <= MAX_USERID; id++) { - if (users[id].callback) - reg_callback(&users[id]); - } - kfree(users); - users = NULL; -} - -/** - * tipc_attach - register a TIPC user - * - * NOTE: This routine may be called when TIPC is inactive. - */ - -int tipc_attach(u32 *userid, tipc_mode_event cb, void *usr_handle) -{ - struct tipc_user *user_ptr; - - if ((tipc_mode == TIPC_NOT_RUNNING) && !cb) - return -ENOPROTOOPT; - if (!users) - reg_init(); - - spin_lock_bh(®_lock); - if (!next_free_user) { - spin_unlock_bh(®_lock); - return -EBUSY; - } - user_ptr = &users[next_free_user]; - *userid = next_free_user; - next_free_user = user_ptr->next; - user_ptr->next = -1; - spin_unlock_bh(®_lock); - - user_ptr->callback = cb; - user_ptr->usr_handle = usr_handle; - INIT_LIST_HEAD(&user_ptr->ports); - atomic_inc(&tipc_user_count); - - if (cb && (tipc_mode != TIPC_NOT_RUNNING)) - tipc_k_signal((Handler)reg_callback, (unsigned long)user_ptr); - return TIPC_OK; -} - -/** - * tipc_detach - deregister a TIPC user - */ - -void tipc_detach(u32 userid) -{ - struct tipc_user *user_ptr; - struct list_head ports_temp; - struct user_port *up_ptr, *temp_up_ptr; - - if ((userid == 0) || (userid > MAX_USERID)) - return; - - spin_lock_bh(®_lock); - if ((!users) || (users[userid].next >= 0)) { - spin_unlock_bh(®_lock); - return; - } - - user_ptr = &users[userid]; - user_ptr->callback = NULL; - INIT_LIST_HEAD(&ports_temp); - list_splice(&user_ptr->ports, &ports_temp); - user_ptr->next = next_free_user; - next_free_user = userid; - spin_unlock_bh(®_lock); - - atomic_dec(&tipc_user_count); - - list_for_each_entry_safe(up_ptr, temp_up_ptr, &ports_temp, uport_list) { - tipc_deleteport(up_ptr->ref); - } -} - -/** - * tipc_reg_add_port - register a user's driver port - */ - -int tipc_reg_add_port(struct user_port *up_ptr) -{ - struct tipc_user *user_ptr; - - if (up_ptr->user_ref == 0) - return TIPC_OK; - if (up_ptr->user_ref > MAX_USERID) - return -EINVAL; - if ((tipc_mode == TIPC_NOT_RUNNING) || !users ) - return -ENOPROTOOPT; - - spin_lock_bh(®_lock); - user_ptr = &users[up_ptr->user_ref]; - list_add(&up_ptr->uport_list, &user_ptr->ports); - spin_unlock_bh(®_lock); - return TIPC_OK; -} - -/** - * tipc_reg_remove_port - deregister a user's driver port - */ - -int tipc_reg_remove_port(struct user_port *up_ptr) -{ - if (up_ptr->user_ref == 0) - return TIPC_OK; - if (up_ptr->user_ref > MAX_USERID) - return -EINVAL; - if (!users ) - return -ENOPROTOOPT; - - spin_lock_bh(®_lock); - list_del_init(&up_ptr->uport_list); - spin_unlock_bh(®_lock); - return TIPC_OK; -} - diff --git a/net/tipc/zone.c b/net/tipc/zone.c deleted file mode 100644 index 3506f856344..00000000000 --- a/net/tipc/zone.c +++ /dev/null @@ -1,173 +0,0 @@ -/* - * net/tipc/zone.c: TIPC zone management routines - * - * 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. - */ - -#include "core.h" -#include "zone.h" -#include "net.h" -#include "addr.h" -#include "node_subscr.h" -#include "cluster.h" -#include "node.h" - -struct _zone *tipc_zone_create(u32 addr) -{ - struct _zone *z_ptr; - u32 z_num; - - if (!tipc_addr_domain_valid(addr)) { - err("Zone creation failed, invalid domain 0x%x\n", addr); - return NULL; - } - - z_ptr = kzalloc(sizeof(*z_ptr), GFP_ATOMIC); - if (!z_ptr) { - warn("Zone creation failed, insufficient memory\n"); - return NULL; - } - - z_num = tipc_zone(addr); - z_ptr->addr = tipc_addr(z_num, 0, 0); - tipc_net.zones[z_num] = z_ptr; - return z_ptr; -} - -void tipc_zone_delete(struct _zone *z_ptr) -{ - u32 c_num; - - if (!z_ptr) - return; - for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { - tipc_cltr_delete(z_ptr->clusters[c_num]); - } - kfree(z_ptr); -} - -void tipc_zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr) -{ - u32 c_num = tipc_cluster(c_ptr->addr); - - assert(c_ptr->addr); - assert(c_num <= tipc_max_clusters); - assert(z_ptr->clusters[c_num] == NULL); - z_ptr->clusters[c_num] = c_ptr; -} - -void tipc_zone_remove_as_router(struct _zone *z_ptr, u32 router) -{ - u32 c_num; - - for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { - if (z_ptr->clusters[c_num]) { - tipc_cltr_remove_as_router(z_ptr->clusters[c_num], - router); - } - } -} - -void tipc_zone_send_external_routes(struct _zone *z_ptr, u32 dest) -{ - u32 c_num; - - for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { - if (z_ptr->clusters[c_num]) { - if (in_own_cluster(z_ptr->addr)) - continue; - tipc_cltr_send_ext_routes(z_ptr->clusters[c_num], dest); - } - } -} - -struct node *tipc_zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref) -{ - struct cluster *c_ptr; - struct node *n_ptr; - u32 c_num; - - if (!z_ptr) - return NULL; - c_ptr = z_ptr->clusters[tipc_cluster(addr)]; - if (!c_ptr) - return NULL; - n_ptr = tipc_cltr_select_node(c_ptr, ref); - if (n_ptr) - return n_ptr; - - /* Links to any other clusters within this zone ? */ - for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { - c_ptr = z_ptr->clusters[c_num]; - if (!c_ptr) - return NULL; - n_ptr = tipc_cltr_select_node(c_ptr, ref); - if (n_ptr) - return n_ptr; - } - return NULL; -} - -u32 tipc_zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref) -{ - struct cluster *c_ptr; - u32 c_num; - u32 router; - - if (!z_ptr) - return 0; - c_ptr = z_ptr->clusters[tipc_cluster(addr)]; - router = c_ptr ? tipc_cltr_select_router(c_ptr, ref) : 0; - if (router) - return router; - - /* Links to any other clusters within the zone? */ - for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { - c_ptr = z_ptr->clusters[c_num]; - router = c_ptr ? tipc_cltr_select_router(c_ptr, ref) : 0; - if (router) - return router; - } - return 0; -} - - -u32 tipc_zone_next_node(u32 addr) -{ - struct cluster *c_ptr = tipc_cltr_find(addr); - - if (c_ptr) - return tipc_cltr_next_node(c_ptr, addr); - return 0; -} - |