aboutsummaryrefslogtreecommitdiff
path: root/net/tipc/link.c
diff options
context:
space:
mode:
Diffstat (limited to 'net/tipc/link.c')
-rw-r--r--net/tipc/link.c1498
1 files changed, 606 insertions, 892 deletions
diff --git a/net/tipc/link.c b/net/tipc/link.c
index a79c755cb41..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, 2010-2011, 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
@@ -37,10 +37,13 @@
#include "core.h"
#include "link.h"
#include "port.h"
+#include "socket.h"
#include "name_distr.h"
#include "discover.h"
#include "config.h"
+#include <linux/pkt_sched.h>
+
/*
* Error message prefixes
*/
@@ -73,36 +76,21 @@ static const char *link_unk_evt = "Unknown link event ";
*/
#define START_CHANGEOVER 100000u
-/**
- * struct tipc_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 tipc_link_name {
- u32 addr_local;
- char if_local[TIPC_MAX_IF_NAME];
- u32 addr_peer;
- char if_peer[TIPC_MAX_IF_NAME];
-};
-
static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr,
struct sk_buff *buf);
-static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf);
-static int link_recv_changeover_msg(struct tipc_link **l_ptr,
- struct sk_buff **buf);
+static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf);
+static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr,
+ struct sk_buff **buf);
static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tolerance);
-static int link_send_sections_long(struct tipc_port *sender,
- struct iovec const *msg_sect,
- u32 num_sect, unsigned int total_len,
- u32 destnode);
-static void link_check_defragm_bufs(struct tipc_link *l_ptr);
+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 void link_start(struct tipc_link *l_ptr);
-static int link_send_long_buf(struct tipc_link *l_ptr, struct sk_buff *buf);
+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
@@ -114,9 +102,18 @@ static unsigned int align(unsigned int i)
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->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;
@@ -158,79 +155,8 @@ int tipc_link_is_active(struct tipc_link *l_ptr)
}
/**
- * link_name_validate - validate & (optionally) deconstruct tipc_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 tipc_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_local = strchr(addr_local, ':');
- if (if_local == NULL)
- return 0;
- *(if_local++) = 0;
- addr_peer = strchr(if_local, '-');
- if (addr_peer == NULL)
- return 0;
- *(addr_peer++) = 0;
- if_local_len = addr_peer - if_local;
- if_peer = strchr(addr_peer, ':');
- if (if_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))
- 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;
-}
-
-/**
* 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 tipc_link *l_ptr)
{
@@ -269,7 +195,6 @@ static void link_timeout(struct tipc_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);
@@ -293,8 +218,8 @@ static void link_set_timer(struct tipc_link *l_ptr, u32 time)
* Returns pointer to link.
*/
struct tipc_link *tipc_link_create(struct tipc_node *n_ptr,
- struct tipc_bearer *b_ptr,
- const struct tipc_media_addr *media_addr)
+ struct tipc_bearer *b_ptr,
+ const struct tipc_media_addr *media_addr)
{
struct tipc_link *l_ptr;
struct tipc_msg *msg;
@@ -333,7 +258,7 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr,
l_ptr->owner = n_ptr;
l_ptr->checkpoint = 1;
l_ptr->peer_session = INVALID_SESSION;
- l_ptr->b_ptr = b_ptr;
+ l_ptr->bearer_id = b_ptr->identity;
link_set_supervision_props(l_ptr, b_ptr->tolerance);
l_ptr->state = RESET_UNKNOWN;
@@ -348,6 +273,7 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr,
l_ptr->priority = b_ptr->priority;
tipc_link_set_queue_limits(l_ptr, b_ptr->window);
+ l_ptr->net_plane = b_ptr->net_plane;
link_init_max_pkt(l_ptr);
l_ptr->next_out_no = 1;
@@ -357,45 +283,44 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr,
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)link_start, (unsigned long)l_ptr);
+ k_init_timer(&l_ptr->timer, (Handler)link_timeout,
+ (unsigned long)l_ptr);
+
+ link_state_event(l_ptr, STARTING_EVT);
return l_ptr;
}
-/**
- * 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 tipc_link *l_ptr)
+void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down)
{
- if (!l_ptr) {
- pr_err("Attempt to delete non-existent link\n");
- return;
- }
-
- k_cancel_timer(&l_ptr->timer);
+ struct tipc_link *l_ptr;
+ struct tipc_node *n_ptr;
- 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);
- 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);
-static void link_start(struct tipc_link *l_ptr)
-{
- tipc_node_lock(l_ptr->owner);
- link_state_event(l_ptr, STARTING_EVT);
- tipc_node_unlock(l_ptr->owner);
+ /* 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();
}
/**
@@ -414,8 +339,6 @@ static int link_schedule_port(struct tipc_link *l_ptr, u32 origport, u32 sz)
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 = 1;
@@ -450,7 +373,7 @@ void tipc_link_wakeup_ports(struct tipc_link *l_ptr, int all)
list_del_init(&p_ptr->wait_list);
spin_lock_bh(p_ptr->lock);
p_ptr->congested = 0;
- p_ptr->wakeup(p_ptr);
+ tipc_port_wakeup(p_ptr);
win -= p_ptr->waiting_pkts;
spin_unlock_bh(p_ptr->lock);
}
@@ -465,14 +388,7 @@ exit:
*/
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;
- kfree_skb(buf);
- buf = next;
- }
+ kfree_skb_list(l_ptr->first_out);
l_ptr->first_out = NULL;
l_ptr->out_queue_size = 0;
}
@@ -483,49 +399,25 @@ static void link_release_outqueue(struct tipc_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;
- kfree_skb(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 tipc_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;
- kfree_skb(buf);
- buf = next;
- }
-
- buf = l_ptr->first_out;
- while (buf) {
- next = buf->next;
- kfree_skb(buf);
- buf = next;
- }
-
+ kfree_skb_list(l_ptr->oldest_deferred_in);
+ kfree_skb_list(l_ptr->first_out);
tipc_link_reset_fragments(l_ptr);
-
kfree_skb(l_ptr->proto_msg_queue);
l_ptr->proto_msg_queue = NULL;
}
void tipc_link_reset(struct tipc_link *l_ptr)
{
- 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);
@@ -544,10 +436,9 @@ void tipc_link_reset(struct tipc_link *l_ptr)
return;
tipc_node_link_down(l_ptr->owner, l_ptr);
- tipc_bearer_remove_dest(l_ptr->b_ptr, l_ptr->addr);
+ tipc_bearer_remove_dest(l_ptr->bearer_id, l_ptr->addr);
- if (was_active_link && tipc_node_active_links(l_ptr->owner) &&
- l_ptr->owner->permit_changeover) {
+ if (was_active_link && tipc_node_active_links(l_ptr->owner)) {
l_ptr->reset_checkpoint = checkpoint;
l_ptr->exp_msg_count = START_CHANGEOVER;
}
@@ -556,12 +447,7 @@ void tipc_link_reset(struct tipc_link *l_ptr)
link_release_outqueue(l_ptr);
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;
- kfree_skb(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);
@@ -581,12 +467,27 @@ void tipc_link_reset(struct tipc_link *l_ptr)
link_reset_statistics(l_ptr);
}
+void tipc_link_reset_list(unsigned int bearer_id)
+{
+ struct tipc_link *l_ptr;
+ struct tipc_node *n_ptr;
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) {
+ tipc_node_lock(n_ptr);
+ l_ptr = n_ptr->links[bearer_id];
+ if (l_ptr)
+ tipc_link_reset(l_ptr);
+ tipc_node_unlock(n_ptr);
+ }
+ rcu_read_unlock();
+}
static void link_activate(struct tipc_link *l_ptr)
{
l_ptr->next_in_no = l_ptr->stats.recv_info = 1;
tipc_node_link_up(l_ptr->owner, l_ptr);
- tipc_bearer_add_dest(l_ptr->b_ptr, l_ptr->addr);
+ tipc_bearer_add_dest(l_ptr->bearer_id, l_ptr->addr);
}
/**
@@ -599,13 +500,17 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
struct tipc_link *other;
u32 cont_intv = l_ptr->continuity_interval;
- if (!l_ptr->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)) {
+ /* 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;
}
switch (l_ptr->state) {
@@ -618,12 +523,12 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
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);
@@ -631,7 +536,7 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
}
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;
@@ -641,7 +546,8 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
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;
@@ -663,7 +569,8 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
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;
@@ -673,14 +580,14 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
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) {
- 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 */
@@ -689,8 +596,8 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
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);
}
@@ -710,22 +617,25 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
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:
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:
- l_ptr->started = 1;
+ l_ptr->flags |= LINK_STARTED;
/* fall through */
case TIMEOUT_EVT:
- 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;
@@ -743,14 +653,17 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
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:
break;
case TIMEOUT_EVT:
- 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;
@@ -767,8 +680,7 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
* link_bundle_buf(): Append contents of a buffer to
* the tail of an existing one.
*/
-static int link_bundle_buf(struct tipc_link *l_ptr,
- struct sk_buff *bundler,
+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);
@@ -837,11 +749,11 @@ static void link_add_chain_to_outqueue(struct tipc_link *l_ptr,
}
/*
- * 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 tipc_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);
@@ -869,20 +781,14 @@ int tipc_link_send_buf(struct tipc_link *l_ptr, struct sk_buff *buf)
/* Fragmentation needed ? */
if (size > max_packet)
- return link_send_long_buf(l_ptr, buf);
+ return tipc_link_frag_xmit(l_ptr, buf);
/* Packet can be queued or sent. */
- if (likely(!tipc_bearer_congested(l_ptr->b_ptr, l_ptr) &&
- !link_congested(l_ptr))) {
+ 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 ? */
@@ -891,10 +797,8 @@ int tipc_link_send_buf(struct tipc_link *l_ptr, struct sk_buff *buf)
/* 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) {
@@ -917,40 +821,78 @@ int tipc_link_send_buf(struct tipc_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 tipc_link *l_ptr;
struct tipc_node *n_ptr;
int res = -ELINKCONG;
- read_lock_bh(&tipc_net_lock);
n_ptr = tipc_node_find(dest);
if (n_ptr) {
tipc_node_lock(n_ptr);
l_ptr = n_ptr->active_links[selector & 1];
if (l_ptr)
- res = tipc_link_send_buf(l_ptr, buf);
+ res = __tipc_link_xmit(l_ptr, buf);
else
kfree_skb(buf);
tipc_node_unlock(n_ptr);
} else {
kfree_skb(buf);
}
- read_unlock_bh(&tipc_net_lock);
return res;
}
-/**
- * tipc_link_send_names - send name table entries to new neighbor
+/*
+ * tipc_link_sync_xmit - synchronize broadcast link endpoints.
+ *
+ * Give a newly added peer node the sequence number where it should
+ * start receiving and acking broadcast packets.
+ *
+ * Called with node locked
+ */
+static void tipc_link_sync_xmit(struct tipc_link *l)
+{
+ struct sk_buff *buf;
+ struct tipc_msg *msg;
+
+ buf = tipc_buf_acquire(INT_H_SIZE);
+ if (!buf)
+ return;
+
+ msg = buf_msg(buf);
+ tipc_msg_init(msg, BCAST_PROTOCOL, STATE_MSG, INT_H_SIZE, l->addr);
+ msg_set_last_bcast(msg, l->owner->bclink.acked);
+ link_add_chain_to_outqueue(l, buf, 0);
+ tipc_link_push_queue(l);
+}
+
+/*
+ * tipc_link_sync_rcv - synchronize broadcast link endpoints.
+ * Receive the sequence number where we should start receiving and
+ * acking broadcast packets from a newly added peer node, and open
+ * up for reception of such packets.
+ *
+ * Called with node locked
+ */
+static void tipc_link_sync_rcv(struct tipc_node *n, struct sk_buff *buf)
+{
+ struct tipc_msg *msg = buf_msg(buf);
+
+ n->bclink.last_sent = n->bclink.last_in = msg_last_bcast(msg);
+ n->bclink.recv_permitted = true;
+ kfree_skb(buf);
+}
+
+/*
+ * tipc_link_names_xmit - send name table entries to new neighbor
*
* Send routine for bulk delivery of name table messages when contact
* with a new neighbor occurs. No link congestion checking is performed
@@ -958,7 +900,7 @@ int tipc_link_send(struct sk_buff *buf, u32 dest, u32 selector)
* small enough not to require fragmentation.
* Called without any locks held.
*/
-void tipc_link_send_names(struct list_head *message_list, u32 dest)
+void tipc_link_names_xmit(struct list_head *message_list, u32 dest)
{
struct tipc_node *n_ptr;
struct tipc_link *l_ptr;
@@ -968,7 +910,6 @@ void tipc_link_send_names(struct list_head *message_list, u32 dest)
if (list_empty(message_list))
return;
- read_lock_bh(&tipc_net_lock);
n_ptr = tipc_node_find(dest);
if (n_ptr) {
tipc_node_lock(n_ptr);
@@ -983,7 +924,6 @@ void tipc_link_send_names(struct list_head *message_list, u32 dest)
}
tipc_node_unlock(n_ptr);
}
- read_unlock_bh(&tipc_net_lock);
/* discard the messages if they couldn't be sent */
list_for_each_safe(buf, temp_buf, ((struct sk_buff *)message_list)) {
@@ -993,82 +933,40 @@ void tipc_link_send_names(struct list_head *message_list, u32 dest)
}
/*
- * link_send_buf_fast: Entry for data messages where the
+ * 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 link_send_buf_fast(struct tipc_link *l_ptr, struct sk_buff *buf,
- u32 *used_max_pkt)
+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)) {
- 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;
- return res;
- }
- 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 = l_ptr->max_pkt;
- }
- return tipc_link_send_buf(l_ptr, buf); /* All other cases */
-}
-
-/*
- * 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.
- */
-int tipc_send_buf_fast(struct sk_buff *buf, u32 destnode)
-{
- struct tipc_link *l_ptr;
- struct tipc_node *n_ptr;
- int res;
- u32 selector = msg_origport(buf_msg(buf)) & 1;
- u32 dummy;
-
- read_lock_bh(&tipc_net_lock);
- n_ptr = tipc_node_find(destnode);
- if (likely(n_ptr)) {
- tipc_node_lock(n_ptr);
- l_ptr = n_ptr->active_links[selector];
- if (likely(l_ptr)) {
- res = link_send_buf_fast(l_ptr, buf, &dummy);
- tipc_node_unlock(n_ptr);
- read_unlock_bh(&tipc_net_lock);
+ 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;
}
- tipc_node_unlock(n_ptr);
+ else
+ *used_max_pkt = l_ptr->max_pkt;
}
- read_unlock_bh(&tipc_net_lock);
- res = msg_data_sz(buf_msg(buf));
- tipc_reject_msg(buf, TIPC_ERR_NO_NODE);
- return res;
+ 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 tipc_port *sender,
- struct iovec const *msg_sect,
- const u32 num_sect,
- unsigned int total_len,
- 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->phdr;
struct tipc_link *l_ptr;
@@ -1082,31 +980,26 @@ 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, num_sect, total_len,
- sender->max_pkt, !sender->user_port, &buf);
+ res = tipc_msg_build(hdr, msg_sect, len, sender->max_pkt, &buf);
+ /* Exit if build request was invalid */
+ if (unlikely(res < 0))
+ return res;
- read_lock_bh(&tipc_net_lock);
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->max_pkt);
+ 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->ref, res);
goto exit;
@@ -1118,31 +1011,25 @@ exit:
*/
sender->max_pkt = l_ptr->max_pkt;
tipc_node_unlock(node);
- read_unlock_bh(&tipc_net_lock);
if ((msg_hdr_sz(hdr) + res) <= sender->max_pkt)
goto again;
- return link_send_sections_long(sender, msg_sect,
- num_sect, total_len,
- 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,
- total_len, 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,
@@ -1155,23 +1042,22 @@ exit:
*
* Returns user data length or errno.
*/
-static int link_send_sections_long(struct tipc_port *sender,
- struct iovec const *msg_sect,
- u32 num_sect,
- unsigned int total_len,
- u32 destaddr)
+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 = total_len;
+ 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;
+ const unchar __user *sect_crs;
int curr_sect;
u32 fragm_no;
+ int res = 0;
again:
fragm_no = 1;
@@ -1210,7 +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)
@@ -1218,18 +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;
- kfree_skb(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;
@@ -1249,8 +1129,10 @@ error:
msg_set_fragm_no(&fragm_hdr, ++fragm_no);
prev = buf;
buf = tipc_buf_acquire(fragm_sz + INT_H_SIZE);
- if (!buf)
+ if (!buf) {
+ res = -ENOMEM;
goto error;
+ }
buf->next = NULL;
prev->next = buf;
@@ -1275,20 +1157,15 @@ error:
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;
- kfree_skb(buf_chain);
- }
+ kfree_skb_list(buf_chain);
goto again;
}
} else {
reject:
- for (; buf_chain; buf_chain = buf) {
- buf = buf_chain->next;
- kfree_skb(buf_chain);
- }
- return tipc_port_reject_sections(sender, hdr, msg_sect, num_sect,
- total_len, 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 chain of fragments to send queue & send them */
@@ -1304,7 +1181,7 @@ reject:
/*
* tipc_link_push_packet: Push one unsent packet to the media
*/
-u32 tipc_link_push_packet(struct tipc_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;
@@ -1329,15 +1206,11 @@ u32 tipc_link_push_packet(struct tipc_link *l_ptr)
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)) {
- l_ptr->retransm_queue_head = mod(++r_q_head);
- l_ptr->retransm_queue_size = --r_q_size;
- l_ptr->stats.retransmitted++;
- return 0;
- } else {
- l_ptr->stats.bearer_congs++;
- 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: */
@@ -1345,15 +1218,11 @@ u32 tipc_link_push_packet(struct tipc_link *l_ptr)
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)) {
- l_ptr->unacked_window = 0;
- kfree_skb(buf);
- l_ptr->proto_msg_queue = NULL;
- return 0;
- } else {
- l_ptr->stats.bearer_congs++;
- return PUSH_FAILED;
- }
+ 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: */
@@ -1366,18 +1235,15 @@ u32 tipc_link_push_packet(struct tipc_link *l_ptr)
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);
- l_ptr->next_out = buf->next;
- return 0;
- } else {
- 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;
}
/*
@@ -1388,48 +1254,33 @@ 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);
-
- if (res == PUSH_FAILED)
- tipc_bearer_schedule(l_ptr->b_ptr, l_ptr);
}
-static void link_reset_all(unsigned long addr)
+void tipc_link_reset_all(struct tipc_node *node)
{
- struct tipc_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);
pr_warn("Resetting all links to %s\n",
- tipc_addr_string_fill(addr_string, n_ptr->addr));
+ 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], "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 tipc_link *l_ptr,
- struct sk_buff *buf)
+ struct sk_buff *buf)
{
struct tipc_msg *msg = buf_msg(buf);
@@ -1454,19 +1305,17 @@ static void link_retransmit_failure(struct tipc_link *l_ptr,
tipc_addr_string_fill(addr_string, n_ptr->addr);
pr_info("Broadcast link info for %s\n", addr_string);
- pr_info("Supportable: %d, Supported: %d, Acked: %u\n",
- n_ptr->bclink.supportable,
- n_ptr->bclink.supported,
+ 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_k_signal((Handler)link_reset_all, (unsigned long)n_ptr->addr);
-
tipc_node_unlock(n_ptr);
+ tipc_bclink_set_flags(TIPC_BCLINK_RESET);
l_ptr->stale_count = 0;
}
}
@@ -1481,43 +1330,25 @@ void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *buf,
msg = buf_msg(buf);
- if (tipc_bearer_congested(l_ptr->b_ptr, l_ptr)) {
- if (l_ptr->retransm_queue_size == 0) {
- l_ptr->retransm_queue_head = msg_seqno(msg);
- l_ptr->retransm_queue_size = retransmits;
- } else {
- pr_err("Unexpected retransmit on link %s (qsize=%d)\n",
- l_ptr->name, l_ptr->retransm_queue_size);
+ /* 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;
}
- return;
} 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) {
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)) {
- 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 = buf_seqno(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;
@@ -1569,6 +1400,12 @@ static int link_recv_buf_validate(struct sk_buff *buf)
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;
@@ -1594,16 +1431,15 @@ static int link_recv_buf_validate(struct sk_buff *buf)
}
/**
- * tipc_recv_msg - process TIPC messages arriving from off-node
+ * tipc_rcv - process TIPC packets/messages arriving from off-node
* @head: pointer to message buffer chain
- * @tb_ptr: pointer to bearer message arrived on
+ * @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_recv_msg(struct sk_buff *head, struct tipc_bearer *b_ptr)
+void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr)
{
- read_lock_bh(&tipc_net_lock);
while (head) {
struct tipc_node *n_ptr;
struct tipc_link *l_ptr;
@@ -1613,77 +1449,68 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *b_ptr)
u32 seq_no;
u32 ackd;
u32 released = 0;
- int type;
head = head->next;
-
- /* Ensure bearer is still enabled */
- if (unlikely(!b_ptr->active))
- goto cont;
+ buf->next = NULL;
/* Ensure message is well-formed */
if (unlikely(!link_recv_buf_validate(buf)))
- goto cont;
+ goto discard;
/* Ensure message data is a single contiguous unit */
if (unlikely(skb_linearize(buf)))
- goto cont;
+ goto discard;
/* Handle arrival of a non-unicast link message */
msg = buf_msg(buf);
if (unlikely(msg_non_seq(msg))) {
if (msg_user(msg) == LINK_CONFIG)
- tipc_disc_recv_msg(buf, b_ptr);
+ tipc_disc_rcv(buf, b_ptr);
else
- tipc_bclink_recv_pkt(buf);
+ 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;
- }
+ if (unlikely(!l_ptr))
+ goto unlock_discard;
/* Verify that communication with node is currently allowed */
- if ((n_ptr->block_setup & WAIT_PEER_DOWN) &&
- msg_user(msg) == LINK_PROTOCOL &&
- (msg_type(msg) == RESET_MSG ||
- msg_type(msg) == ACTIVATE_MSG) &&
- !msg_redundant_link(msg))
- n_ptr->block_setup &= ~WAIT_PEER_DOWN;
-
- if (n_ptr->block_setup) {
- tipc_node_unlock(n_ptr);
- goto cont;
- }
+ 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.supported)
+ if (n_ptr->bclink.recv_permitted)
tipc_bclink_acknowledge(n_ptr, msg_bcast_ack(msg));
crs = l_ptr->first_out;
while ((crs != l_ptr->next_out) &&
less_eq(buf_seqno(crs), ackd)) {
struct sk_buff *next = crs->next;
-
kfree_skb(crs);
crs = next;
released++;
@@ -1696,102 +1523,105 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *b_ptr)
/* 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);
}
- /* Now (finally!) process the incoming message */
-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);
-deliver:
- if (likely(msg_isdata(msg))) {
- tipc_node_unlock(n_ptr);
- tipc_port_recv_msg(buf);
- continue;
- }
- switch (msg_user(msg)) {
- int ret;
- 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 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++;
- ret = tipc_link_recv_fragment(
- &l_ptr->defragm_buf,
- &buf, &msg);
- if (ret == 1) {
- l_ptr->stats.recv_fragmented++;
- goto deliver;
- }
- if (ret == -1)
- l_ptr->next_in_no--;
- 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);
- if (type == ORIGINAL_MSG)
- goto deliver;
- goto protocol_check;
- }
- break;
- default:
- kfree_skb(buf);
- buf = NULL;
- 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);
- tipc_net_route_msg(buf);
continue;
}
+
+ /* Traffic message. Conditionally activate link */
+ link_state_event(l_ptr, TRAFFIC_MSG_EVT);
+
+ if (link_working_working(l_ptr)) {
+ /* Re-insert buffer in front of queue */
+ buf->next = head;
+ head = buf;
+ tipc_node_unlock(n_ptr);
+ continue;
+ }
+ goto unlock_discard;
+ }
+
+ /* Link is now in state WORKING_WORKING */
+ if (unlikely(seq_no != mod(l_ptr->next_in_no))) {
link_handle_out_of_seq_msg(l_ptr, buf);
head = link_insert_deferred_queue(l_ptr, head);
tipc_node_unlock(n_ptr);
continue;
}
-
- 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;
+ }
}
- link_state_event(l_ptr, TRAFFIC_MSG_EVT);
- if (link_working_working(l_ptr)) {
- /* Re-insert in front of queue */
- 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:
+ continue;
+unlock_discard:
+ tipc_node_unlock(n_ptr);
+discard:
kfree_skb(buf);
}
- read_unlock_bh(&tipc_net_lock);
}
/**
@@ -1853,7 +1683,7 @@ static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr,
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;
}
@@ -1874,8 +1704,9 @@ static void link_handle_out_of_seq_msg(struct tipc_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++;
}
@@ -1883,9 +1714,8 @@ static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr,
/*
* Send protocol message to the other endpoint.
*/
-void tipc_link_send_proto_msg(struct tipc_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;
@@ -1898,16 +1728,17 @@ void tipc_link_send_proto_msg(struct tipc_link *l_ptr, u32 msg_typ,
l_ptr->proto_msg_queue = NULL;
}
- if (link_blocked(l_ptr))
+ /* Don't send protocol message during link changeover */
+ if (l_ptr->exp_msg_count)
return;
/* Abort non-RESET send if communication with node is prohibited */
- if ((l_ptr->owner->block_setup) && (msg_typ != RESET_MSG))
+ 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_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());
@@ -1971,22 +1802,9 @@ void tipc_link_send_proto_msg(struct tipc_link *l_ptr, u32 msg_typ,
return;
skb_copy_to_linear_data(buf, msg, sizeof(l_ptr->proto_msg));
+ buf->priority = TC_PRIO_CONTROL;
- /* Defer message if bearer is already congested */
- if (tipc_bearer_congested(l_ptr->b_ptr, l_ptr)) {
- l_ptr->proto_msg_queue = buf;
- return;
- }
-
- /* Defer message if attempting to send results in bearer congestion */
- if (!tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) {
- tipc_bearer_schedule(l_ptr->b_ptr, l_ptr);
- l_ptr->proto_msg_queue = buf;
- l_ptr->stats.bearer_congs++;
- return;
- }
-
- /* Discard message if it was sent successfully */
+ tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr);
l_ptr->unacked_window = 0;
kfree_skb(buf);
}
@@ -1996,7 +1814,7 @@ void tipc_link_send_proto_msg(struct tipc_link *l_ptr, u32 msg_typ,
* 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 tipc_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;
@@ -2004,17 +1822,13 @@ static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf)
u32 msg_tol;
struct tipc_msg *msg = buf_msg(buf);
- 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)) {
@@ -2031,7 +1845,7 @@ static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf)
* peer has lost contact -- don't allow peer's links
* to reactivate before we recognize loss & clean up
*/
- l_ptr->owner->block_setup = WAIT_NODE_DOWN;
+ l_ptr->owner->action_flags |= TIPC_WAIT_OWN_LINKS_DOWN;
}
link_state_event(l_ptr, RESET_MSG);
@@ -2057,7 +1871,6 @@ static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf)
} else {
l_ptr->max_pkt = l_ptr->max_pkt_target;
}
- l_ptr->owner->bclink.supportable = (max_pkt_info != 0);
/* Synchronize broadcast link info, if not done previously */
if (!tipc_node_is_up(l_ptr->owner)) {
@@ -2088,6 +1901,10 @@ static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf)
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))
@@ -2112,13 +1929,13 @@ static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf)
}
/* Protocol message before retransmits, reduce loss risk */
- if (l_ptr->owner->bclink.supported)
+ 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)) {
l_ptr->stats.recv_nacks++;
@@ -2132,14 +1949,13 @@ exit:
}
-/*
- * 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.
*/
-static void tipc_link_tunnel(struct tipc_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 tipc_link *tunnel;
struct sk_buff *buf;
@@ -2158,16 +1974,17 @@ static void tipc_link_tunnel(struct tipc_link *l_ptr,
}
skb_copy_to_linear_data(buf, tunnel_hdr, INT_H_SIZE);
skb_copy_to_linear_data_offset(buf, INT_H_SIZE, msg, length);
- tipc_link_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 tipc_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;
@@ -2178,11 +1995,6 @@ void tipc_link_changeover(struct tipc_link *l_ptr)
if (!tunnel)
return;
- if (!l_ptr->owner->permit_changeover) {
- pr_warn("%speer did not permit changeover\n", link_co_err);
- return;
- }
-
tipc_msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL,
ORIGINAL_MSG, INT_H_SIZE, l_ptr->addr);
msg_set_bearer_id(&tunnel_hdr, l_ptr->peer_bearer_id);
@@ -2195,7 +2007,7 @@ void tipc_link_changeover(struct tipc_link *l_ptr)
if (buf) {
skb_copy_to_linear_data(buf, &tunnel_hdr, INT_H_SIZE);
msg_set_size(&tunnel_hdr, INT_H_SIZE);
- tipc_link_send_buf(tunnel, buf);
+ __tipc_link_xmit(tunnel, buf);
} else {
pr_warn("%sunable to send changeover msg\n",
link_co_err);
@@ -2216,20 +2028,30 @@ void tipc_link_changeover(struct tipc_link *l_ptr)
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));
+ 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 tipc_link *l_ptr, struct tipc_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;
@@ -2258,7 +2080,7 @@ void tipc_link_send_duplicate(struct tipc_link *l_ptr, struct tipc_link *tunnel)
skb_copy_to_linear_data(outbuf, &tunnel_hdr, INT_H_SIZE);
skb_copy_to_linear_data_offset(outbuf, INT_H_SIZE, iter->data,
length);
- tipc_link_send_buf(tunnel, outbuf);
+ __tipc_link_xmit(tunnel, outbuf);
if (!tipc_link_is_up(l_ptr))
return;
iter = iter->next;
@@ -2285,84 +2107,112 @@ static struct sk_buff *buf_extract(struct sk_buff *skb, u32 from_pos)
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 int link_recv_changeover_msg(struct tipc_link **l_ptr,
- struct sk_buff **buf)
+static void tipc_link_dup_rcv(struct tipc_link *l_ptr,
+ struct sk_buff *t_buf)
{
- struct sk_buff *tunnel_buf = *buf;
- struct tipc_link *dest_link;
- 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);
+ struct sk_buff *buf;
- dest_link = (*l_ptr)->owner->links[msg_bearer_id(tunnel_msg)];
- if (!dest_link)
- goto exit;
- if (dest_link == *l_ptr) {
- pr_err("Unexpected changeover message on link <%s>\n",
- (*l_ptr)->name);
- goto exit;
+ 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;
}
- *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)))
- goto exit;
- *buf = buf_extract(tunnel_buf, INT_H_SIZE);
- if (*buf == NULL) {
- pr_warn("%sduplicate msg dropped\n", link_co_err);
+ /* Add buffer to deferred queue, if applicable: */
+ link_handle_out_of_seq_msg(l_ptr, buf);
+}
+
+/* tipc_link_failover_rcv(): Receive a tunnelled ORIGINAL_MSG packet
+ * Owner node is locked.
+ */
+static struct sk_buff *tipc_link_failover_rcv(struct tipc_link *l_ptr,
+ struct sk_buff *t_buf)
+{
+ struct tipc_msg *t_msg = buf_msg(t_buf);
+ struct sk_buff *buf = NULL;
+ struct tipc_msg *msg;
+
+ if (tipc_link_is_up(l_ptr))
+ tipc_link_reset(l_ptr);
+
+ /* First failover packet? */
+ if (l_ptr->exp_msg_count == START_CHANGEOVER)
+ l_ptr->exp_msg_count = msg_msgcnt(t_msg);
+
+ /* Should there be an inner packet? */
+ if (l_ptr->exp_msg_count) {
+ l_ptr->exp_msg_count--;
+ buf = buf_extract(t_buf, INT_H_SIZE);
+ if (buf == NULL) {
+ pr_warn("%sno inner failover pkt\n", link_co_err);
goto exit;
}
- kfree_skb(tunnel_buf);
- return 1;
- }
+ msg = buf_msg(buf);
- /* First original message ?: */
- if (tipc_link_is_up(dest_link)) {
- pr_info("%s<%s>, changeover initiated by peer\n", link_rst_msg,
- dest_link->name);
- tipc_link_reset(dest_link);
- dest_link->exp_msg_count = msg_count;
- if (!msg_count)
- goto exit;
- } else if (dest_link->exp_msg_count == START_CHANGEOVER) {
- dest_link->exp_msg_count = msg_count;
- if (!msg_count)
+ if (less(msg_seqno(msg), l_ptr->reset_checkpoint)) {
+ kfree_skb(buf);
+ buf = NULL;
goto exit;
+ }
+ if (msg_user(msg) == MSG_FRAGMENTER) {
+ l_ptr->stats.recv_fragments++;
+ tipc_buf_append(&l_ptr->reasm_buf, &buf);
+ }
+ }
+exit:
+ if ((l_ptr->exp_msg_count == 0) && (l_ptr->flags & LINK_STOPPED)) {
+ tipc_node_detach_link(l_ptr->owner, l_ptr);
+ kfree(l_ptr);
}
+ return buf;
+}
- /* Receive original message */
- if (dest_link->exp_msg_count == 0) {
- pr_warn("%sgot too many tunnelled messages\n", link_co_err);
+/* tipc_link_tunnel_rcv(): Receive a tunnelled packet, sent
+ * via other link as result of a failover (ORIGINAL_MSG) or
+ * a new active link (DUPLICATE_MSG). Failover packets are
+ * returned to the active link for delivery upwards.
+ * Owner node is locked.
+ */
+static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr,
+ struct sk_buff **buf)
+{
+ struct sk_buff *t_buf = *buf;
+ struct tipc_link *l_ptr;
+ struct tipc_msg *t_msg = buf_msg(t_buf);
+ u32 bearer_id = msg_bearer_id(t_msg);
+
+ *buf = NULL;
+
+ if (bearer_id >= MAX_BEARERS)
goto exit;
- }
- dest_link->exp_msg_count--;
- if (less(msg_seqno(msg), dest_link->reset_checkpoint)) {
+
+ l_ptr = n_ptr->links[bearer_id];
+ if (!l_ptr)
goto exit;
- } else {
- *buf = buf_extract(tunnel_buf, INT_H_SIZE);
- if (*buf != NULL) {
- kfree_skb(tunnel_buf);
- return 1;
- } else {
- pr_warn("%soriginal msg dropped\n", link_co_err);
- }
- }
+
+ 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;
- kfree_skb(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;
@@ -2385,11 +2235,11 @@ void tipc_link_recv_bundle(struct sk_buff *buf)
*/
/*
- * 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.
*/
-static int link_send_long_buf(struct tipc_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;
@@ -2424,11 +2274,7 @@ static int link_send_long_buf(struct tipc_link *l_ptr, struct sk_buff *buf)
fragm = tipc_buf_acquire(fragm_sz + INT_H_SIZE);
if (fragm == NULL) {
kfree_skb(buf);
- while (buf_chain) {
- buf = buf_chain;
- buf_chain = buf_chain->next;
- kfree_skb(buf);
- }
+ kfree_skb_list(buf_chain);
return -ENOMEM;
}
msg_set_size(&fragm_hdr, fragm_sz + INT_H_SIZE);
@@ -2456,157 +2302,6 @@ static int link_send_long_buf(struct tipc_link *l_ptr, struct sk_buff *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 unnecessary.
- */
-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);
-
- *fb = NULL;
-
- /* Is there an incomplete message waiting for this fragment? */
- while (pbuf && ((buf_seqno(pbuf) != long_msg_seq_no) ||
- (msg_orignode(fragm) != msg_orignode(buf_msg(pbuf))))) {
- prev = pbuf;
- pbuf = pbuf->next;
- }
-
- 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 + NAMED_H_SIZE;
- if (msg_type(imsg) == TIPC_MCAST_MSG)
- max = TIPC_MAX_USER_MSG_SIZE + MCAST_H_SIZE;
- if (msg_size(imsg) > max) {
- kfree_skb(fbuf);
- return 0;
- }
- pbuf = tipc_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 {
- pr_debug("Link unable to reassemble fragmented message\n");
- kfree_skb(fbuf);
- return -1;
- }
- kfree_skb(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);
- kfree_skb(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;
- }
- kfree_skb(fbuf);
- return 0;
-}
-
-/**
- * link_check_defragm_bufs - flush stale incoming message fragments
- * @l_ptr: pointer to link
- */
-static void link_check_defragm_bufs(struct tipc_link *l_ptr)
-{
- 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))
- return;
- while (buf) {
- u32 cnt = get_timer_cnt(buf);
-
- next = buf->next;
- if (cnt < 4) {
- incr_timer_cnt(buf);
- prev = buf;
- } else {
- if (prev)
- prev->next = buf->next;
- else
- l_ptr->defragm_buf = buf->next;
- kfree_skb(buf);
- }
- buf = next;
- }
-}
-
static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tolerance)
{
if ((tolerance < TIPC_MIN_LINK_TOL) || (tolerance > TIPC_MAX_LINK_TOL))
@@ -2637,39 +2332,39 @@ void tipc_link_set_queue_limits(struct tipc_link *l_ptr, u32 window)
l_ptr->queue_limit[MSG_FRAGMENTER] = 4000;
}
-/**
- * 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
- *
- * Caller must hold 'tipc_net_lock' to ensure node and bearer are not deleted;
- * this also prevents link deletion.
+/* 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 link (or 0 if invalid link name).
+ * Returns pointer to node owning the link, or 0 if no matching link is found.
*/
-static struct tipc_link *link_find_link(const char *name,
- struct tipc_node **node)
+static struct tipc_node *tipc_link_find_owner(const char *link_name,
+ unsigned int *bearer_id)
{
- struct tipc_link_name link_name_parts;
- struct tipc_bearer *b_ptr;
struct tipc_link *l_ptr;
+ struct tipc_node *n_ptr;
+ struct tipc_node *found_node = 0;
+ int i;
- if (!link_name_validate(name, &link_name_parts))
- return NULL;
-
- b_ptr = tipc_bearer_find_interface(link_name_parts.if_local);
- if (!b_ptr)
- return NULL;
-
- *node = tipc_node_find(link_name_parts.addr_peer);
- if (!*node)
- return NULL;
-
- l_ptr = (*node)->links[b_ptr->identity];
- if (!l_ptr || strcmp(l_ptr->name, name))
- return NULL;
+ *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 l_ptr;
+ return found_node;
}
/**
@@ -2701,7 +2396,7 @@ static int link_value_is_valid(u16 cmd, u32 new_value)
* @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 'tipc_net_lock' to ensure link/bearer/media is not deleted.
+ * Caller must hold RTNL lock to ensure link/bearer/media is not deleted.
*
* Returns 0 if value updated and negative value on error.
*/
@@ -2711,31 +2406,36 @@ static int link_cmd_set_value(const char *name, u32 new_value, u16 cmd)
struct tipc_link *l_ptr;
struct tipc_bearer *b_ptr;
struct tipc_media *m_ptr;
+ int bearer_id;
+ int res = 0;
- l_ptr = link_find_link(name, &node);
- if (l_ptr) {
- /*
- * acquire node lock for tipc_link_send_proto_msg().
- * see "TIPC locking policy" in net.c.
- */
+ node = tipc_link_find_owner(name, &bearer_id);
+ if (node) {
tipc_node_lock(node);
- switch (cmd) {
- case TIPC_CMD_SET_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);
- break;
- case TIPC_CMD_SET_LINK_PRI:
- l_ptr->priority = new_value;
- tipc_link_send_proto_msg(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;
+ l_ptr = node->links[bearer_id];
+
+ if (l_ptr) {
+ switch (cmd) {
+ case TIPC_CMD_SET_LINK_TOL:
+ link_set_supervision_props(l_ptr, new_value);
+ tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0,
+ new_value, 0, 0);
+ break;
+ case TIPC_CMD_SET_LINK_PRI:
+ l_ptr->priority = new_value;
+ tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0,
+ 0, new_value, 0);
+ break;
+ case TIPC_CMD_SET_LINK_WINDOW:
+ tipc_link_set_queue_limits(l_ptr, new_value);
+ break;
+ default:
+ res = -EINVAL;
+ break;
+ }
}
tipc_node_unlock(node);
- return 0;
+ return res;
}
b_ptr = tipc_bearer_find(name);
@@ -2743,15 +2443,18 @@ static int link_cmd_set_value(const char *name, u32 new_value, u16 cmd)
switch (cmd) {
case TIPC_CMD_SET_LINK_TOL:
b_ptr->tolerance = new_value;
- return 0;
+ break;
case TIPC_CMD_SET_LINK_PRI:
b_ptr->priority = new_value;
- return 0;
+ break;
case TIPC_CMD_SET_LINK_WINDOW:
b_ptr->window = new_value;
- return 0;
+ break;
+ default:
+ res = -EINVAL;
+ break;
}
- return -EINVAL;
+ return res;
}
m_ptr = tipc_media_find(name);
@@ -2760,15 +2463,18 @@ static int link_cmd_set_value(const char *name, u32 new_value, u16 cmd)
switch (cmd) {
case TIPC_CMD_SET_LINK_TOL:
m_ptr->tolerance = new_value;
- return 0;
+ break;
case TIPC_CMD_SET_LINK_PRI:
m_ptr->priority = new_value;
- return 0;
+ break;
case TIPC_CMD_SET_LINK_WINDOW:
m_ptr->window = new_value;
- return 0;
+ break;
+ default:
+ res = -EINVAL;
+ break;
}
- return -EINVAL;
+ return res;
}
struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space,
@@ -2796,9 +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);
res = link_cmd_set_value(args->name, new_value, cmd);
- read_unlock_bh(&tipc_net_lock);
if (res)
return tipc_cfg_reply_error_string("cannot change link setting");
@@ -2821,6 +2525,7 @@ struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_
char *link_name;
struct tipc_link *l_ptr;
struct tipc_node *node;
+ unsigned int bearer_id;
if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME))
return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR);
@@ -2831,18 +2536,18 @@ 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();
}
@@ -2869,18 +2574,24 @@ static int tipc_link_stats(const char *name, char *buf, const u32 buf_size)
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);
- read_lock_bh(&tipc_net_lock);
- l = link_find_link(name, &node);
+ node = tipc_link_find_owner(name, &bearer_id);
+ if (!node)
+ return 0;
+
+ tipc_node_lock(node);
+
+ l = node->links[bearer_id];
if (!l) {
- read_unlock_bh(&tipc_net_lock);
+ tipc_node_unlock(node);
return 0;
}
- tipc_node_lock(node);
+
s = &l->stats;
if (tipc_link_is_active(l))
@@ -2937,13 +2648,12 @@ static int tipc_link_stats(const char *name, char *buf, const u32 buf_size)
s->sent_nacks, s->sent_acks, s->retransmitted);
ret += tipc_snprintf(buf + ret, buf_size - ret,
- " Congestion bearer:%u link:%u Send queue"
- " max:%u avg:%u\n", s->bearer_congs, s->link_congs,
+ " 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 ret;
}
@@ -2994,7 +2704,6 @@ u32 tipc_link_get_max_pkt(u32 dest, u32 selector)
if (dest == tipc_own_addr)
return MAX_MSG_SIZE;
- read_lock_bh(&tipc_net_lock);
n_ptr = tipc_node_find(dest);
if (n_ptr) {
tipc_node_lock(n_ptr);
@@ -3003,13 +2712,18 @@ u32 tipc_link_get_max_pkt(u32 dest, u32 selector)
res = l_ptr->max_pkt;
tipc_node_unlock(n_ptr);
}
- read_unlock_bh(&tipc_net_lock);
return res;
}
static void link_print(struct tipc_link *l_ptr, const char *str)
{
- pr_info("%s Link %x<%s>:", str, l_ptr->addr, l_ptr->b_ptr->name);
+ struct tipc_bearer *b_ptr;
+
+ rcu_read_lock();
+ b_ptr = rcu_dereference_rtnl(bearer_list[l_ptr->bearer_id]);
+ if (b_ptr)
+ pr_info("%s Link %x<%s>:", str, l_ptr->addr, b_ptr->name);
+ rcu_read_unlock();
if (link_working_unknown(l_ptr))
pr_cont(":WU\n");
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-10 11:26:46 +0000