From 8123b421e8ed944671d7241323ed3198cccb4041 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 8 Aug 2008 23:23:39 -0700 Subject: pkt_sched: Fix ingress deletion and filter attachment. Based upon bug reports by Stephen Hemminger. We still had some cases using ->qdisc instead of ->qdisc_sleeping. Also, qdisc_lookup() should return ingress qdiscs. Signed-off-by: David S. Miller --- net/sched/sch_api.c | 36 +++++++++++++++++++++++------------- 1 file changed, 23 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index ba1d121f3127..bbf149dd7818 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -183,6 +183,21 @@ EXPORT_SYMBOL(unregister_qdisc); (root qdisc, all its children, children of children etc.) */ +struct Qdisc *qdisc_match_from_root(struct Qdisc *root, u32 handle) +{ + struct Qdisc *q; + + if (!(root->flags & TCQ_F_BUILTIN) && + root->handle == handle) + return root; + + list_for_each_entry(q, &root->list, list) { + if (q->handle == handle) + return q; + } + return NULL; +} + struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) { unsigned int i; @@ -191,16 +206,11 @@ struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) struct netdev_queue *txq = netdev_get_tx_queue(dev, i); struct Qdisc *q, *txq_root = txq->qdisc_sleeping; - if (!(txq_root->flags & TCQ_F_BUILTIN) && - txq_root->handle == handle) - return txq_root; - - list_for_each_entry(q, &txq_root->list, list) { - if (q->handle == handle) - return q; - } + q = qdisc_match_from_root(txq_root, handle); + if (q) + return q; } - return NULL; + return qdisc_match_from_root(dev->rx_queue.qdisc_sleeping, handle); } static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid) @@ -908,7 +918,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg) return -ENOENT; q = qdisc_leaf(p, clid); } else { /* ingress */ - q = dev->rx_queue.qdisc; + q = dev->rx_queue.qdisc_sleeping; } } else { struct netdev_queue *dev_queue; @@ -978,7 +988,7 @@ replay: return -ENOENT; q = qdisc_leaf(p, clid); } else { /*ingress */ - q = dev->rx_queue.qdisc; + q = dev->rx_queue.qdisc_sleeping; } } else { struct netdev_queue *dev_queue; @@ -1529,11 +1539,11 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb) t = 0; dev_queue = netdev_get_tx_queue(dev, 0); - if (tc_dump_tclass_root(dev_queue->qdisc, skb, tcm, cb, &t, s_t) < 0) + if (tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, &t, s_t) < 0) goto done; dev_queue = &dev->rx_queue; - if (tc_dump_tclass_root(dev_queue->qdisc, skb, tcm, cb, &t, s_t) < 0) + if (tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, &t, s_t) < 0) goto done; done: -- cgit v1.2.3 From d97106ea52aa57e63ff40d04479016836bbb5a4e Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sat, 9 Aug 2008 00:35:05 -0700 Subject: udp: Drop socket lock for encapsulated packets The socket lock is there to protect the normal UDP receive path. Encapsulation UDP sockets don't need that protection. In fact the locking is deadly for them as they may contain another UDP packet within, possibly with the same addresses. Also the nested bit was copied from TCP. TCP needs it because of accept(2) spawning sockets. This simply doesn't apply to UDP so I've removed it. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/udp.c | 6 ++++-- net/ipv6/udp.c | 6 +++--- 2 files changed, 7 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 383d17359d01..8e42fbbd5761 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -989,7 +989,9 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) up->encap_rcv != NULL) { int ret; + bh_unlock_sock(sk); ret = (*up->encap_rcv)(sk, skb); + bh_lock_sock(sk); if (ret <= 0) { UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_INDATAGRAMS, @@ -1092,7 +1094,7 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb, if (skb1) { int ret = 0; - bh_lock_sock_nested(sk); + bh_lock_sock(sk); if (!sock_owned_by_user(sk)) ret = udp_queue_rcv_skb(sk, skb1); else @@ -1194,7 +1196,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], if (sk != NULL) { int ret = 0; - bh_lock_sock_nested(sk); + bh_lock_sock(sk); if (!sock_owned_by_user(sk)) ret = udp_queue_rcv_skb(sk, skb); else diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index d1477b350f76..a6aecf76a71b 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -379,7 +379,7 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb, uh->source, saddr, dif))) { struct sk_buff *buff = skb_clone(skb, GFP_ATOMIC); if (buff) { - bh_lock_sock_nested(sk2); + bh_lock_sock(sk2); if (!sock_owned_by_user(sk2)) udpv6_queue_rcv_skb(sk2, buff); else @@ -387,7 +387,7 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb, bh_unlock_sock(sk2); } } - bh_lock_sock_nested(sk); + bh_lock_sock(sk); if (!sock_owned_by_user(sk)) udpv6_queue_rcv_skb(sk, skb); else @@ -508,7 +508,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], /* deliver */ - bh_lock_sock_nested(sk); + bh_lock_sock(sk); if (!sock_owned_by_user(sk)) udpv6_queue_rcv_skb(sk, skb); else -- cgit v1.2.3 From bc0fde2fad007a81ecffceb25a893a6c3f1ed767 Mon Sep 17 00:00:00 2001 From: Sven Wegener Date: Sun, 10 Aug 2008 09:14:05 +0000 Subject: ipvs: Fix possible deadlock in sync code Commit 998e7a76804b7a273a0460c2cdd5a51fa9856717 ("ipvs: Use kthread_run() instead of doing a double-fork via kernel_thread()") introduced a possible deadlock in the sync code. We need to use the _bh versions for the lock, as the lock is also accessed from a bottom half. Signed-off-by: Sven Wegener Acked-by: Simon Horman --- net/ipv4/ipvs/ip_vs_sync.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/ipvs/ip_vs_sync.c b/net/ipv4/ipvs/ip_vs_sync.c index 45e9bd96c286..a652da2c3200 100644 --- a/net/ipv4/ipvs/ip_vs_sync.c +++ b/net/ipv4/ipvs/ip_vs_sync.c @@ -904,9 +904,9 @@ int stop_sync_thread(int state) * progress of stopping the master sync daemon. */ - spin_lock(&ip_vs_sync_lock); + spin_lock_bh(&ip_vs_sync_lock); ip_vs_sync_state &= ~IP_VS_STATE_MASTER; - spin_unlock(&ip_vs_sync_lock); + spin_unlock_bh(&ip_vs_sync_lock); kthread_stop(sync_master_thread); sync_master_thread = NULL; } else if (state == IP_VS_STATE_BACKUP) { -- cgit v1.2.3 From 8ab19ea36c5c5340ff598e4d15fc084eb65671dc Mon Sep 17 00:00:00 2001 From: Sven Wegener Date: Sun, 10 Aug 2008 09:17:59 +0000 Subject: ipvs: Fix possible deadlock in estimator code There is a slight chance for a deadlock in the estimator code. We can't call del_timer_sync() while holding our lock, as the timer might be active and spinning for the lock on another cpu. Work around this issue by using try_to_del_timer_sync() and releasing the lock. We could actually delete the timer outside of our lock, as the add and kill functions are only every called from userspace via [gs]etsockopt() and are serialized by a mutex, but better make this explicit. Signed-off-by: Sven Wegener Cc: stable Acked-by: Simon Horman --- net/ipv4/ipvs/ip_vs_est.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/ipvs/ip_vs_est.c b/net/ipv4/ipvs/ip_vs_est.c index bc04eedd6dbb..1d6e58e502fd 100644 --- a/net/ipv4/ipvs/ip_vs_est.c +++ b/net/ipv4/ipvs/ip_vs_est.c @@ -170,8 +170,11 @@ void ip_vs_kill_estimator(struct ip_vs_stats *stats) kfree(est); killed++; } - if (killed && est_list == NULL) - del_timer_sync(&est_timer); + while (killed && !est_list && try_to_del_timer_sync(&est_timer) < 0) { + write_unlock_bh(&est_lock); + cpu_relax(); + write_lock_bh(&est_lock); + } write_unlock_bh(&est_lock); } -- cgit v1.2.3 From 66a0be47200fff30f8c482ea584052c6affb08cb Mon Sep 17 00:00:00 2001 From: Sven Wegener Date: Sun, 10 Aug 2008 09:18:02 +0000 Subject: ipvs: Use list_empty() instead of open-coding the same functionality Signed-off-by: Sven Wegener Acked-by: Simon Horman --- net/ipv4/ipvs/ip_vs_sched.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/ipvs/ip_vs_sched.c b/net/ipv4/ipvs/ip_vs_sched.c index b64767309855..a46ad9e35016 100644 --- a/net/ipv4/ipvs/ip_vs_sched.c +++ b/net/ipv4/ipvs/ip_vs_sched.c @@ -184,7 +184,7 @@ int register_ip_vs_scheduler(struct ip_vs_scheduler *scheduler) write_lock_bh(&__ip_vs_sched_lock); - if (scheduler->n_list.next != &scheduler->n_list) { + if (!list_empty(&scheduler->n_list)) { write_unlock_bh(&__ip_vs_sched_lock); ip_vs_use_count_dec(); IP_VS_ERR("register_ip_vs_scheduler(): [%s] scheduler " @@ -229,7 +229,7 @@ int unregister_ip_vs_scheduler(struct ip_vs_scheduler *scheduler) } write_lock_bh(&__ip_vs_sched_lock); - if (scheduler->n_list.next == &scheduler->n_list) { + if (list_empty(&scheduler->n_list)) { write_unlock_bh(&__ip_vs_sched_lock); IP_VS_ERR("unregister_ip_vs_scheduler(): [%s] scheduler " "is not in the list. failed\n", scheduler->name); -- cgit v1.2.3 From d149ccc9cf85cdf089c1b2189ade111305712b0c Mon Sep 17 00:00:00 2001 From: Sven Wegener Date: Sun, 10 Aug 2008 09:18:02 +0000 Subject: ipvs: Initialize schedulers' struct list_head at compile time No need to do it at runtime and this saves a couple of bytes in the text section. Signed-off-by: Sven Wegener Acked-by: Simon Horman --- net/ipv4/ipvs/ip_vs_dh.c | 2 +- net/ipv4/ipvs/ip_vs_lblc.c | 2 +- net/ipv4/ipvs/ip_vs_lblcr.c | 2 +- net/ipv4/ipvs/ip_vs_lc.c | 2 +- net/ipv4/ipvs/ip_vs_nq.c | 2 +- net/ipv4/ipvs/ip_vs_rr.c | 2 +- net/ipv4/ipvs/ip_vs_sed.c | 2 +- net/ipv4/ipvs/ip_vs_sh.c | 2 +- net/ipv4/ipvs/ip_vs_wlc.c | 2 +- net/ipv4/ipvs/ip_vs_wrr.c | 2 +- 10 files changed, 10 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/ipv4/ipvs/ip_vs_dh.c b/net/ipv4/ipvs/ip_vs_dh.c index 8afc1503ed20..fa66824d264f 100644 --- a/net/ipv4/ipvs/ip_vs_dh.c +++ b/net/ipv4/ipvs/ip_vs_dh.c @@ -233,6 +233,7 @@ static struct ip_vs_scheduler ip_vs_dh_scheduler = .name = "dh", .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_dh_scheduler.n_list), .init_service = ip_vs_dh_init_svc, .done_service = ip_vs_dh_done_svc, .update_service = ip_vs_dh_update_svc, @@ -242,7 +243,6 @@ static struct ip_vs_scheduler ip_vs_dh_scheduler = static int __init ip_vs_dh_init(void) { - INIT_LIST_HEAD(&ip_vs_dh_scheduler.n_list); return register_ip_vs_scheduler(&ip_vs_dh_scheduler); } diff --git a/net/ipv4/ipvs/ip_vs_lblc.c b/net/ipv4/ipvs/ip_vs_lblc.c index 0efa3db4b180..7a6a319f544a 100644 --- a/net/ipv4/ipvs/ip_vs_lblc.c +++ b/net/ipv4/ipvs/ip_vs_lblc.c @@ -539,6 +539,7 @@ static struct ip_vs_scheduler ip_vs_lblc_scheduler = .name = "lblc", .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_lblc_scheduler.n_list), .init_service = ip_vs_lblc_init_svc, .done_service = ip_vs_lblc_done_svc, .update_service = ip_vs_lblc_update_svc, @@ -550,7 +551,6 @@ static int __init ip_vs_lblc_init(void) { int ret; - INIT_LIST_HEAD(&ip_vs_lblc_scheduler.n_list); sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table); ret = register_ip_vs_scheduler(&ip_vs_lblc_scheduler); if (ret) diff --git a/net/ipv4/ipvs/ip_vs_lblcr.c b/net/ipv4/ipvs/ip_vs_lblcr.c index 8e3bbeb45138..c234e73968a6 100644 --- a/net/ipv4/ipvs/ip_vs_lblcr.c +++ b/net/ipv4/ipvs/ip_vs_lblcr.c @@ -728,6 +728,7 @@ static struct ip_vs_scheduler ip_vs_lblcr_scheduler = .name = "lblcr", .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_lblcr_scheduler.n_list), .init_service = ip_vs_lblcr_init_svc, .done_service = ip_vs_lblcr_done_svc, .update_service = ip_vs_lblcr_update_svc, @@ -739,7 +740,6 @@ static int __init ip_vs_lblcr_init(void) { int ret; - INIT_LIST_HEAD(&ip_vs_lblcr_scheduler.n_list); sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table); ret = register_ip_vs_scheduler(&ip_vs_lblcr_scheduler); if (ret) diff --git a/net/ipv4/ipvs/ip_vs_lc.c b/net/ipv4/ipvs/ip_vs_lc.c index ac9f08e065d5..ebcdbf75ac65 100644 --- a/net/ipv4/ipvs/ip_vs_lc.c +++ b/net/ipv4/ipvs/ip_vs_lc.c @@ -98,6 +98,7 @@ static struct ip_vs_scheduler ip_vs_lc_scheduler = { .name = "lc", .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_lc_scheduler.n_list), .init_service = ip_vs_lc_init_svc, .done_service = ip_vs_lc_done_svc, .update_service = ip_vs_lc_update_svc, @@ -107,7 +108,6 @@ static struct ip_vs_scheduler ip_vs_lc_scheduler = { static int __init ip_vs_lc_init(void) { - INIT_LIST_HEAD(&ip_vs_lc_scheduler.n_list); return register_ip_vs_scheduler(&ip_vs_lc_scheduler) ; } diff --git a/net/ipv4/ipvs/ip_vs_nq.c b/net/ipv4/ipvs/ip_vs_nq.c index a46bf258d420..92f3a6770031 100644 --- a/net/ipv4/ipvs/ip_vs_nq.c +++ b/net/ipv4/ipvs/ip_vs_nq.c @@ -136,6 +136,7 @@ static struct ip_vs_scheduler ip_vs_nq_scheduler = .name = "nq", .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_nq_scheduler.n_list), .init_service = ip_vs_nq_init_svc, .done_service = ip_vs_nq_done_svc, .update_service = ip_vs_nq_update_svc, @@ -145,7 +146,6 @@ static struct ip_vs_scheduler ip_vs_nq_scheduler = static int __init ip_vs_nq_init(void) { - INIT_LIST_HEAD(&ip_vs_nq_scheduler.n_list); return register_ip_vs_scheduler(&ip_vs_nq_scheduler); } diff --git a/net/ipv4/ipvs/ip_vs_rr.c b/net/ipv4/ipvs/ip_vs_rr.c index c8db12d39e61..358110d17e59 100644 --- a/net/ipv4/ipvs/ip_vs_rr.c +++ b/net/ipv4/ipvs/ip_vs_rr.c @@ -94,6 +94,7 @@ static struct ip_vs_scheduler ip_vs_rr_scheduler = { .name = "rr", /* name */ .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_rr_scheduler.n_list), .init_service = ip_vs_rr_init_svc, .done_service = ip_vs_rr_done_svc, .update_service = ip_vs_rr_update_svc, @@ -102,7 +103,6 @@ static struct ip_vs_scheduler ip_vs_rr_scheduler = { static int __init ip_vs_rr_init(void) { - INIT_LIST_HEAD(&ip_vs_rr_scheduler.n_list); return register_ip_vs_scheduler(&ip_vs_rr_scheduler); } diff --git a/net/ipv4/ipvs/ip_vs_sed.c b/net/ipv4/ipvs/ip_vs_sed.c index 2a7d31358181..77663d84cbd1 100644 --- a/net/ipv4/ipvs/ip_vs_sed.c +++ b/net/ipv4/ipvs/ip_vs_sed.c @@ -138,6 +138,7 @@ static struct ip_vs_scheduler ip_vs_sed_scheduler = .name = "sed", .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_sed_scheduler.n_list), .init_service = ip_vs_sed_init_svc, .done_service = ip_vs_sed_done_svc, .update_service = ip_vs_sed_update_svc, @@ -147,7 +148,6 @@ static struct ip_vs_scheduler ip_vs_sed_scheduler = static int __init ip_vs_sed_init(void) { - INIT_LIST_HEAD(&ip_vs_sed_scheduler.n_list); return register_ip_vs_scheduler(&ip_vs_sed_scheduler); } diff --git a/net/ipv4/ipvs/ip_vs_sh.c b/net/ipv4/ipvs/ip_vs_sh.c index b8fdfac65001..7b979e228056 100644 --- a/net/ipv4/ipvs/ip_vs_sh.c +++ b/net/ipv4/ipvs/ip_vs_sh.c @@ -230,6 +230,7 @@ static struct ip_vs_scheduler ip_vs_sh_scheduler = .name = "sh", .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_sh_scheduler.n_list), .init_service = ip_vs_sh_init_svc, .done_service = ip_vs_sh_done_svc, .update_service = ip_vs_sh_update_svc, @@ -239,7 +240,6 @@ static struct ip_vs_scheduler ip_vs_sh_scheduler = static int __init ip_vs_sh_init(void) { - INIT_LIST_HEAD(&ip_vs_sh_scheduler.n_list); return register_ip_vs_scheduler(&ip_vs_sh_scheduler); } diff --git a/net/ipv4/ipvs/ip_vs_wlc.c b/net/ipv4/ipvs/ip_vs_wlc.c index 772c3cb4eca1..9b0ef86bb1f7 100644 --- a/net/ipv4/ipvs/ip_vs_wlc.c +++ b/net/ipv4/ipvs/ip_vs_wlc.c @@ -126,6 +126,7 @@ static struct ip_vs_scheduler ip_vs_wlc_scheduler = .name = "wlc", .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_wlc_scheduler.n_list), .init_service = ip_vs_wlc_init_svc, .done_service = ip_vs_wlc_done_svc, .update_service = ip_vs_wlc_update_svc, @@ -135,7 +136,6 @@ static struct ip_vs_scheduler ip_vs_wlc_scheduler = static int __init ip_vs_wlc_init(void) { - INIT_LIST_HEAD(&ip_vs_wlc_scheduler.n_list); return register_ip_vs_scheduler(&ip_vs_wlc_scheduler); } diff --git a/net/ipv4/ipvs/ip_vs_wrr.c b/net/ipv4/ipvs/ip_vs_wrr.c index 1d6932d7dc97..0d86a79b87b5 100644 --- a/net/ipv4/ipvs/ip_vs_wrr.c +++ b/net/ipv4/ipvs/ip_vs_wrr.c @@ -212,6 +212,7 @@ static struct ip_vs_scheduler ip_vs_wrr_scheduler = { .name = "wrr", .refcnt = ATOMIC_INIT(0), .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_wrr_scheduler.n_list), .init_service = ip_vs_wrr_init_svc, .done_service = ip_vs_wrr_done_svc, .update_service = ip_vs_wrr_update_svc, @@ -220,7 +221,6 @@ static struct ip_vs_scheduler ip_vs_wrr_scheduler = { static int __init ip_vs_wrr_init(void) { - INIT_LIST_HEAD(&ip_vs_wrr_scheduler.n_list); return register_ip_vs_scheduler(&ip_vs_wrr_scheduler) ; } -- cgit v1.2.3 From 048cf48b897bcae9e6fa8b46b6976dab5e710e3c Mon Sep 17 00:00:00 2001 From: Sven Wegener Date: Sun, 10 Aug 2008 18:24:35 +0000 Subject: ipvs: Annotate init functions with __init Being able to discard these functions saves a couple of bytes at runtime. The cleanup functions can't be annotated with __exit as they are also called from init functions. Signed-off-by: Sven Wegener Acked-by: Simon Horman --- net/ipv4/ipvs/ip_vs_app.c | 2 +- net/ipv4/ipvs/ip_vs_conn.c | 2 +- net/ipv4/ipvs/ip_vs_ctl.c | 2 +- net/ipv4/ipvs/ip_vs_proto.c | 4 ++-- 4 files changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv4/ipvs/ip_vs_app.c b/net/ipv4/ipvs/ip_vs_app.c index 1f1897a1a702..201b8ea3020d 100644 --- a/net/ipv4/ipvs/ip_vs_app.c +++ b/net/ipv4/ipvs/ip_vs_app.c @@ -608,7 +608,7 @@ int ip_vs_skb_replace(struct sk_buff *skb, gfp_t pri, } -int ip_vs_app_init(void) +int __init ip_vs_app_init(void) { /* we will replace it with proc_net_ipvs_create() soon */ proc_net_fops_create(&init_net, "ip_vs_app", 0, &ip_vs_app_fops); diff --git a/net/ipv4/ipvs/ip_vs_conn.c b/net/ipv4/ipvs/ip_vs_conn.c index f8bdae47a77f..44a6872dc245 100644 --- a/net/ipv4/ipvs/ip_vs_conn.c +++ b/net/ipv4/ipvs/ip_vs_conn.c @@ -965,7 +965,7 @@ static void ip_vs_conn_flush(void) } -int ip_vs_conn_init(void) +int __init ip_vs_conn_init(void) { int idx; diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c index 9a5ace0b4dd6..df13333813ad 100644 --- a/net/ipv4/ipvs/ip_vs_ctl.c +++ b/net/ipv4/ipvs/ip_vs_ctl.c @@ -2306,7 +2306,7 @@ static struct nf_sockopt_ops ip_vs_sockopts = { }; -int ip_vs_control_init(void) +int __init ip_vs_control_init(void) { int ret; int idx; diff --git a/net/ipv4/ipvs/ip_vs_proto.c b/net/ipv4/ipvs/ip_vs_proto.c index 876714f23d65..6099a88fc200 100644 --- a/net/ipv4/ipvs/ip_vs_proto.c +++ b/net/ipv4/ipvs/ip_vs_proto.c @@ -43,7 +43,7 @@ static struct ip_vs_protocol *ip_vs_proto_table[IP_VS_PROTO_TAB_SIZE]; /* * register an ipvs protocol */ -static int __used register_ip_vs_protocol(struct ip_vs_protocol *pp) +static int __used __init register_ip_vs_protocol(struct ip_vs_protocol *pp) { unsigned hash = IP_VS_PROTO_HASH(pp->protocol); @@ -190,7 +190,7 @@ ip_vs_tcpudp_debug_packet(struct ip_vs_protocol *pp, } -int ip_vs_protocol_init(void) +int __init ip_vs_protocol_init(void) { char protocols[64]; #define REGISTER_PROTOCOL(p) \ -- cgit v1.2.3 From 5587da55fbf332ab8d1b37637536f94bc373867f Mon Sep 17 00:00:00 2001 From: Sven Wegener Date: Sun, 10 Aug 2008 18:24:40 +0000 Subject: ipvs: Mark net_vs_ctl_path const Signed-off-by: Sven Wegener Acked-by: Simon Horman --- include/net/ip_vs.h | 2 +- net/ipv4/ipvs/ip_vs_ctl.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index e980416bff81..c8ee9b89b023 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -620,7 +620,7 @@ extern int sysctl_ip_vs_expire_quiescent_template; extern int sysctl_ip_vs_sync_threshold[2]; extern int sysctl_ip_vs_nat_icmp_send; extern struct ip_vs_stats ip_vs_stats; -extern struct ctl_path net_vs_ctl_path[]; +extern const struct ctl_path net_vs_ctl_path[]; extern struct ip_vs_service * ip_vs_service_get(__u32 fwmark, __u16 protocol, __be32 vaddr, __be16 vport); diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c index df13333813ad..999d884e8862 100644 --- a/net/ipv4/ipvs/ip_vs_ctl.c +++ b/net/ipv4/ipvs/ip_vs_ctl.c @@ -1589,7 +1589,7 @@ static struct ctl_table vs_vars[] = { { .ctl_name = 0 } }; -struct ctl_path net_vs_ctl_path[] = { +const struct ctl_path net_vs_ctl_path[] = { { .procname = "net", .ctl_name = CTL_NET, }, { .procname = "ipv4", .ctl_name = NET_IPV4, }, { .procname = "vs", }, -- cgit v1.2.3 From 3a14a313f9b406c37ab7e3f855b060eb8587b8c7 Mon Sep 17 00:00:00 2001 From: Sven Wegener Date: Sun, 10 Aug 2008 18:24:41 +0000 Subject: ipvs: Embed estimator object into stats object There's no reason for dynamically allocating an estimator object for every stats object. Directly embed an estimator object into every stats object and switch to using the kernel-provided list implementation. This makes the code much simpler and faster, as we do not need to traverse the list of all estimators to find the one belonging to a stats object. There's no need to use an rwlock, as we only have one reader. Also reorder the members of the estimator structure slightly to avoid padding overhead. This can't be done with the stats object as the members are currently copied to our user space object via memcpy() and changing it would break ABI. Signed-off-by: Sven Wegener Acked-by: Simon Horman --- include/net/ip_vs.h | 28 ++++++++++- net/ipv4/ipvs/ip_vs_ctl.c | 2 +- net/ipv4/ipvs/ip_vs_est.c | 117 +++++++++++++++------------------------------- 3 files changed, 65 insertions(+), 82 deletions(-) (limited to 'net') diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index c8ee9b89b023..7312c3dd309f 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -140,8 +140,24 @@ struct ip_vs_seq { /* - * IPVS statistics object + * IPVS statistics objects */ +struct ip_vs_estimator { + struct list_head list; + + u64 last_inbytes; + u64 last_outbytes; + u32 last_conns; + u32 last_inpkts; + u32 last_outpkts; + + u32 cps; + u32 inpps; + u32 outpps; + u32 inbps; + u32 outbps; +}; + struct ip_vs_stats { __u32 conns; /* connections scheduled */ @@ -156,7 +172,15 @@ struct ip_vs_stats __u32 inbps; /* current in byte rate */ __u32 outbps; /* current out byte rate */ + /* + * Don't add anything before the lock, because we use memcpy() to copy + * the members before the lock to struct ip_vs_stats_user in + * ip_vs_ctl.c. + */ + spinlock_t lock; /* spin lock */ + + struct ip_vs_estimator est; /* estimator */ }; struct dst_entry; @@ -659,7 +683,7 @@ extern void ip_vs_sync_conn(struct ip_vs_conn *cp); /* * IPVS rate estimator prototypes (from ip_vs_est.c) */ -extern int ip_vs_new_estimator(struct ip_vs_stats *stats); +extern void ip_vs_new_estimator(struct ip_vs_stats *stats); extern void ip_vs_kill_estimator(struct ip_vs_stats *stats); extern void ip_vs_zero_estimator(struct ip_vs_stats *stats); diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c index 999d884e8862..d651bce05493 100644 --- a/net/ipv4/ipvs/ip_vs_ctl.c +++ b/net/ipv4/ipvs/ip_vs_ctl.c @@ -684,8 +684,8 @@ ip_vs_zero_stats(struct ip_vs_stats *stats) { spin_lock_bh(&stats->lock); memset(stats, 0, (char *)&stats->lock - (char *)stats); - spin_unlock_bh(&stats->lock); ip_vs_zero_estimator(stats); + spin_unlock_bh(&stats->lock); } /* diff --git a/net/ipv4/ipvs/ip_vs_est.c b/net/ipv4/ipvs/ip_vs_est.c index 1d6e58e502fd..5a20f93bd7f9 100644 --- a/net/ipv4/ipvs/ip_vs_est.c +++ b/net/ipv4/ipvs/ip_vs_est.c @@ -17,6 +17,7 @@ #include #include #include +#include #include @@ -44,28 +45,11 @@ */ -struct ip_vs_estimator -{ - struct ip_vs_estimator *next; - struct ip_vs_stats *stats; - - u32 last_conns; - u32 last_inpkts; - u32 last_outpkts; - u64 last_inbytes; - u64 last_outbytes; - - u32 cps; - u32 inpps; - u32 outpps; - u32 inbps; - u32 outbps; -}; +static void estimation_timer(unsigned long arg); - -static struct ip_vs_estimator *est_list = NULL; -static DEFINE_RWLOCK(est_lock); -static struct timer_list est_timer; +static LIST_HEAD(est_list); +static DEFINE_SPINLOCK(est_lock); +static DEFINE_TIMER(est_timer, estimation_timer, 0, 0); static void estimation_timer(unsigned long arg) { @@ -76,9 +60,9 @@ static void estimation_timer(unsigned long arg) u64 n_inbytes, n_outbytes; u32 rate; - read_lock(&est_lock); - for (e = est_list; e; e = e->next) { - s = e->stats; + spin_lock(&est_lock); + list_for_each_entry(e, &est_list, list) { + s = container_of(e, struct ip_vs_stats, est); spin_lock(&s->lock); n_conns = s->conns; @@ -114,19 +98,16 @@ static void estimation_timer(unsigned long arg) s->outbps = (e->outbps+0xF)>>5; spin_unlock(&s->lock); } - read_unlock(&est_lock); + spin_unlock(&est_lock); mod_timer(&est_timer, jiffies + 2*HZ); } -int ip_vs_new_estimator(struct ip_vs_stats *stats) +void ip_vs_new_estimator(struct ip_vs_stats *stats) { - struct ip_vs_estimator *est; + struct ip_vs_estimator *est = &stats->est; - est = kzalloc(sizeof(*est), GFP_KERNEL); - if (est == NULL) - return -ENOMEM; + INIT_LIST_HEAD(&est->list); - est->stats = stats; est->last_conns = stats->conns; est->cps = stats->cps<<10; @@ -142,62 +123,40 @@ int ip_vs_new_estimator(struct ip_vs_stats *stats) est->last_outbytes = stats->outbytes; est->outbps = stats->outbps<<5; - write_lock_bh(&est_lock); - est->next = est_list; - if (est->next == NULL) { - setup_timer(&est_timer, estimation_timer, 0); - est_timer.expires = jiffies + 2*HZ; - add_timer(&est_timer); - } - est_list = est; - write_unlock_bh(&est_lock); - return 0; + spin_lock_bh(&est_lock); + if (list_empty(&est_list)) + mod_timer(&est_timer, jiffies + 2 * HZ); + list_add(&est->list, &est_list); + spin_unlock_bh(&est_lock); } void ip_vs_kill_estimator(struct ip_vs_stats *stats) { - struct ip_vs_estimator *est, **pest; - int killed = 0; - - write_lock_bh(&est_lock); - pest = &est_list; - while ((est=*pest) != NULL) { - if (est->stats != stats) { - pest = &est->next; - continue; - } - *pest = est->next; - kfree(est); - killed++; - } - while (killed && !est_list && try_to_del_timer_sync(&est_timer) < 0) { - write_unlock_bh(&est_lock); + struct ip_vs_estimator *est = &stats->est; + + spin_lock_bh(&est_lock); + list_del(&est->list); + while (list_empty(&est_list) && try_to_del_timer_sync(&est_timer) < 0) { + spin_unlock_bh(&est_lock); cpu_relax(); - write_lock_bh(&est_lock); + spin_lock_bh(&est_lock); } - write_unlock_bh(&est_lock); + spin_unlock_bh(&est_lock); } void ip_vs_zero_estimator(struct ip_vs_stats *stats) { - struct ip_vs_estimator *e; - - write_lock_bh(&est_lock); - for (e = est_list; e; e = e->next) { - if (e->stats != stats) - continue; - - /* set counters zero */ - e->last_conns = 0; - e->last_inpkts = 0; - e->last_outpkts = 0; - e->last_inbytes = 0; - e->last_outbytes = 0; - e->cps = 0; - e->inpps = 0; - e->outpps = 0; - e->inbps = 0; - e->outbps = 0; - } - write_unlock_bh(&est_lock); + struct ip_vs_estimator *est = &stats->est; + + /* set counters zero, caller must hold the stats->lock lock */ + est->last_inbytes = 0; + est->last_outbytes = 0; + est->last_conns = 0; + est->last_inpkts = 0; + est->last_outpkts = 0; + est->cps = 0; + est->inpps = 0; + est->outpps = 0; + est->inbps = 0; + est->outbps = 0; } -- cgit v1.2.3 From 519e49e888458649dde453d36c08b7f3432525dc Mon Sep 17 00:00:00 2001 From: Sven Wegener Date: Sun, 10 Aug 2008 18:24:41 +0000 Subject: ipvs: No need to zero out ip_vs_stats during initialization It's a global variable and automatically initialized to zero. And now we can also initialize the lock at compile time. Signed-off-by: Sven Wegener Acked-by: Simon Horman --- net/ipv4/ipvs/ip_vs_ctl.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c index d651bce05493..cfb1d20993d1 100644 --- a/net/ipv4/ipvs/ip_vs_ctl.c +++ b/net/ipv4/ipvs/ip_vs_ctl.c @@ -1784,7 +1784,9 @@ static const struct file_operations ip_vs_info_fops = { #endif -struct ip_vs_stats ip_vs_stats; +struct ip_vs_stats ip_vs_stats = { + .lock = __SPIN_LOCK_UNLOCKED(ip_vs_stats.lock), +}; #ifdef CONFIG_PROC_FS static int ip_vs_stats_show(struct seq_file *seq, void *v) @@ -2333,8 +2335,6 @@ int __init ip_vs_control_init(void) INIT_LIST_HEAD(&ip_vs_rtable[idx]); } - memset(&ip_vs_stats, 0, sizeof(ip_vs_stats)); - spin_lock_init(&ip_vs_stats.lock); ip_vs_new_estimator(&ip_vs_stats); /* Hook the defense timer */ -- cgit v1.2.3 From e93615d0866a974afc7148172f8382e2af48c985 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Mon, 11 Aug 2008 17:19:14 +1000 Subject: ipvs: Explictly clear ip_vs_stats members In order to align the coding styles of ip_vs_zero_stats() and its child-function ip_vs_zero_estimator(), clear ip_vs_stats members explicitlty rather than doing a limited memset(). This was chosen over modifying ip_vs_zero_estimator() to use memset() as it is more robust against changes in members in the relevant structures. memset() would be prefered if all members of the structure were to be cleared. Cc: Sven Wegener Signed-off-by: Simon Horman Signed-off-by: Sven Wegener --- net/ipv4/ipvs/ip_vs_ctl.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c index cfb1d20993d1..6379705a8dcb 100644 --- a/net/ipv4/ipvs/ip_vs_ctl.c +++ b/net/ipv4/ipvs/ip_vs_ctl.c @@ -683,8 +683,21 @@ static void ip_vs_zero_stats(struct ip_vs_stats *stats) { spin_lock_bh(&stats->lock); - memset(stats, 0, (char *)&stats->lock - (char *)stats); + + stats->conns = 0; + stats->inpkts = 0; + stats->outpkts = 0; + stats->inbytes = 0; + stats->outbytes = 0; + + stats->cps = 0; + stats->inpps = 0; + stats->outpps = 0; + stats->inbps = 0; + stats->outbps = 0; + ip_vs_zero_estimator(stats); + spin_unlock_bh(&stats->lock); } -- cgit v1.2.3 From 1cfa26661a85549063e369e2b40275eeaa7b923c Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Mon, 11 Aug 2008 18:11:06 -0700 Subject: pkt_sched: Add BH protection for qdisc_stab_lock. Since qdisc_stab_lock is used in qdisc_put_stab(), which is called in BH context from __qdisc_destroy() RCU callback, softirq safe locking is needed. Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/sched/sch_api.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index bbf149dd7818..c25465e5607a 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -331,7 +331,7 @@ static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt) if (!s || tsize != s->tsize || (!tab && tsize > 0)) return ERR_PTR(-EINVAL); - spin_lock(&qdisc_stab_lock); + spin_lock_bh(&qdisc_stab_lock); list_for_each_entry(stab, &qdisc_stab_list, list) { if (memcmp(&stab->szopts, s, sizeof(*s))) @@ -339,11 +339,11 @@ static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt) if (tsize > 0 && memcmp(stab->data, tab, tsize * sizeof(u16))) continue; stab->refcnt++; - spin_unlock(&qdisc_stab_lock); + spin_unlock_bh(&qdisc_stab_lock); return stab; } - spin_unlock(&qdisc_stab_lock); + spin_unlock_bh(&qdisc_stab_lock); stab = kmalloc(sizeof(*stab) + tsize * sizeof(u16), GFP_KERNEL); if (!stab) @@ -354,9 +354,9 @@ static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt) if (tsize > 0) memcpy(stab->data, tab, tsize * sizeof(u16)); - spin_lock(&qdisc_stab_lock); + spin_lock_bh(&qdisc_stab_lock); list_add_tail(&stab->list, &qdisc_stab_list); - spin_unlock(&qdisc_stab_lock); + spin_unlock_bh(&qdisc_stab_lock); return stab; } @@ -366,14 +366,14 @@ void qdisc_put_stab(struct qdisc_size_table *tab) if (!tab) return; - spin_lock(&qdisc_stab_lock); + spin_lock_bh(&qdisc_stab_lock); if (--tab->refcnt == 0) { list_del(&tab->list); kfree(tab); } - spin_unlock(&qdisc_stab_lock); + spin_unlock_bh(&qdisc_stab_lock); } EXPORT_SYMBOL(qdisc_put_stab); -- cgit v1.2.3 From 5e0115e500fe9dd2ca11e6f92db9123204f1327a Mon Sep 17 00:00:00 2001 From: Brian Haley Date: Wed, 13 Aug 2008 01:58:57 -0700 Subject: ipv6: Fix OOPS, ip -f inet6 route get fec0::1, linux-2.6.26, ip6_route_output, rt6_fill_node+0x175 Alexey Dobriyan wrote: > On Thu, Aug 07, 2008 at 07:00:56PM +0200, John Gumb wrote: >> Scenario: no ipv6 default route set. > >> # ip -f inet6 route get fec0::1 >> >> BUG: unable to handle kernel NULL pointer dereference at 00000000 >> IP: [] rt6_fill_node+0x175/0x3b0 >> EIP is at rt6_fill_node+0x175/0x3b0 > > 0xffffffff80424dd3 is in rt6_fill_node (net/ipv6/route.c:2191). > 2186 } else > 2187 #endif > 2188 NLA_PUT_U32(skb, RTA_IIF, iif); > 2189 } else if (dst) { > 2190 struct in6_addr saddr_buf; > 2191 ====> if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev, > ^^^^^^^^^^^^^^^^^^^^^^^^ > NULL > > 2192 dst, 0, &saddr_buf) == 0) > 2193 NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf); > 2194 } The commit that changed this can't be reverted easily, but the patch below works for me. Fix NULL de-reference in rt6_fill_node() when there's no IPv6 input device present in the dst entry. Signed-off-by: Brian Haley Signed-off-by: David S. Miller --- net/ipv6/route.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 5a3e87e4b18f..41b165ffb369 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2187,8 +2187,9 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, #endif NLA_PUT_U32(skb, RTA_IIF, iif); } else if (dst) { + struct inet6_dev *idev = ip6_dst_idev(&rt->u.dst); struct in6_addr saddr_buf; - if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev, + if (ipv6_dev_get_saddr(idev ? idev->dev : NULL, dst, 0, &saddr_buf) == 0) NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf); } -- cgit v1.2.3 From 6ced0b3f1e1c089caf8798485423a093744b6a48 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 13 Aug 2008 02:32:06 -0700 Subject: net/tipc/subscr.c: don't use ___constant_swab32 It's an internal implementation detail which we _should_ be free to change. So we did, and it promptly broke. The compiler shold be able to work out when to use the __constant version anyway. Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- net/tipc/subscr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index 0326d3060bc7..0747d8a9232f 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -85,7 +85,7 @@ static struct top_srv topsrv = { 0 }; static u32 htohl(u32 in, int swap) { - return swap ? (u32)___constant_swab32(in) : in; + return swap ? swab32(in) : in; } /** -- cgit v1.2.3 From 317900cb010f4aca0e3cb14a02d0ddcc44ddafa7 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Wed, 13 Aug 2008 02:39:56 -0700 Subject: wext: Send name on events In the minimal the wireless extensions oughta send at least the name in addition to the ifindex. Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- net/wireless/wext.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/wireless/wext.c b/net/wireless/wext.c index df5b3886c36b..d98ffb75119a 100644 --- a/net/wireless/wext.c +++ b/net/wireless/wext.c @@ -1277,6 +1277,7 @@ static int rtnetlink_fill_iwinfo(struct sk_buff *skb, struct net_device *dev, r->ifi_flags = dev_get_flags(dev); r->ifi_change = 0; /* Wireless changes don't affect those flags */ + NLA_PUT_STRING(skb, IFLA_IFNAME, dev->name); /* Add the wireless events in the netlink packet */ NLA_PUT(skb, IFLA_WIRELESS, event_len, event); -- cgit v1.2.3 From 34093d055e09d1bb549efc11c8d448373437bbe4 Mon Sep 17 00:00:00 2001 From: Julien Brunel Date: Wed, 13 Aug 2008 02:40:48 -0700 Subject: net/rxrpc: Use an IS_ERR test rather than a NULL test In case of error, the function rxrpc_get_transport returns an ERR pointer, but never returns a NULL pointer. So after a call to this function, a NULL test should be replaced by an IS_ERR test. A simplified version of the semantic patch that makes this change is as follows: (http://www.emn.fr/x-info/coccinelle/) // @correct_null_test@ expression x,E; statement S1, S2; @@ x = rxrpc_get_transport(...) <... when != x = E if ( ( - x@p2 != NULL + ! IS_ERR ( x ) | - x@p2 == NULL + IS_ERR( x ) ) ) S1 else S2 ...> ? x = E; // Signed-off-by: Julien Brunel Signed-off-by: Julia Lawall Signed-off-by: David S. Miller --- net/rxrpc/ar-accept.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/rxrpc/ar-accept.c b/net/rxrpc/ar-accept.c index bdfb77417794..77228f28fa36 100644 --- a/net/rxrpc/ar-accept.c +++ b/net/rxrpc/ar-accept.c @@ -100,7 +100,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, trans = rxrpc_get_transport(local, peer, GFP_NOIO); rxrpc_put_peer(peer); - if (!trans) { + if (IS_ERR(trans)) { _debug("no trans"); ret = -EBUSY; goto error; -- cgit v1.2.3 From f97017cdefefdb6a0e19266024b0c6f9fd411eeb Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Wed, 13 Aug 2008 02:41:22 -0700 Subject: net-sched: Fix actions flushing Flushing of actions has been broken since we changed the semantics of netlink parsed tb[X] to mean X is an attribute type. This makes the flushing work. Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- net/sched/act_api.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 26c7e1f9a350..88b57331d130 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -790,6 +790,8 @@ static int tca_action_flush(struct nlattr *nla, struct nlmsghdr *n, u32 pid) err = a->ops->walk(skb, &dcb, RTM_DELACTION, a); if (err < 0) goto nla_put_failure; + if (err == 0) + goto noflush_out; nla_nest_end(skb, nest); @@ -807,6 +809,7 @@ nla_put_failure: nlmsg_failure: module_put(a->ops->owner); err_out: +noflush_out: kfree_skb(skb); kfree(a); return err; @@ -824,8 +827,10 @@ tca_action_gd(struct nlattr *nla, struct nlmsghdr *n, u32 pid, int event) return ret; if (event == RTM_DELACTION && n->nlmsg_flags&NLM_F_ROOT) { - if (tb[0] != NULL && tb[1] == NULL) - return tca_action_flush(tb[0], n, pid); + if (tb[1] != NULL) + return tca_action_flush(tb[1], n, pid); + else + return -EINVAL; } for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) { -- cgit v1.2.3 From 36723873b664fb6b5cfe06d291df948126e43f50 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Wed, 13 Aug 2008 02:41:45 -0700 Subject: net-sched: fix Action flushing return code Flushing must consistently return ENOMEM on failure of any allocation Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- net/sched/act_api.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 88b57331d130..9974b3f04f05 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -751,7 +751,7 @@ static int tca_action_flush(struct nlattr *nla, struct nlmsghdr *n, u32 pid) struct nlattr *tb[TCA_ACT_MAX+1]; struct nlattr *kind; struct tc_action *a = create_a(0); - int err = -EINVAL; + int err = -ENOMEM; if (a == NULL) { printk("tca_action_flush: couldnt create tc_action\n"); @@ -762,7 +762,7 @@ static int tca_action_flush(struct nlattr *nla, struct nlmsghdr *n, u32 pid) if (!skb) { printk("tca_action_flush: failed skb alloc\n"); kfree(a); - return -ENOBUFS; + return err; } b = skb_tail_pointer(skb); -- cgit v1.2.3 From c1e24df27fb1058739789126db6ad1b1ef719346 Mon Sep 17 00:00:00 2001 From: Jean-Christophe DUBOIS Date: Wed, 13 Aug 2008 13:35:37 -0700 Subject: xfrm: remove unnecessary variable in xfrm_output_resume() 2nd try Small fix removing an unnecessary intermediate variable. Signed-off-by: Jean-Christophe DUBOIS Signed-off-by: David S. Miller --- net/xfrm/xfrm_output.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c index 3f964db908a7..ac25b4c0e982 100644 --- a/net/xfrm/xfrm_output.c +++ b/net/xfrm/xfrm_output.c @@ -112,16 +112,13 @@ error_nolock: int xfrm_output_resume(struct sk_buff *skb, int err) { while (likely((err = xfrm_output_one(skb, err)) == 0)) { - struct xfrm_state *x; - nf_reset(skb); err = skb->dst->ops->local_out(skb); if (unlikely(err != 1)) goto out; - x = skb->dst->xfrm; - if (!x) + if (!skb->dst->xfrm) return dst_output(skb); err = nf_hook(skb->dst->ops->family, -- cgit v1.2.3 From 3e8a0a559c66ee9e7468195691a56fefc3589740 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 13 Aug 2008 13:48:39 -0700 Subject: dccp: change L/R must have at least one byte in the dccpsf_val field Thanks to Eugene Teo for reporting this problem. Signed-off-by: Eugene Teo Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- net/dccp/proto.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net') diff --git a/net/dccp/proto.c b/net/dccp/proto.c index b622d9744856..1ca3b26eed0f 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -474,6 +474,11 @@ static int dccp_setsockopt_change(struct sock *sk, int type, if (copy_from_user(&opt, optval, sizeof(opt))) return -EFAULT; + /* + * rfc4340: 6.1. Change Options + */ + if (opt.dccpsf_len < 1) + return -EINVAL; val = kmalloc(opt.dccpsf_len, GFP_KERNEL); if (!val) -- cgit v1.2.3 From 24b8b44780a2c53ecb738f4a1c08d114f5eda27c Mon Sep 17 00:00:00 2001 From: Tom Tucker Date: Wed, 13 Aug 2008 11:05:41 -0500 Subject: svcrdma: Fix race between svc_rdma_recvfrom thread and the dto_tasklet RDMA_READ completions are kept on a separate queue from the general I/O request queue. Since a separate lock is used to protect the RDMA_READ completion queue, a race exists between the dto_tasklet and the svc_rdma_recvfrom thread where the dto_tasklet sets the XPT_DATA bit and adds I/O to the read-completion queue. Concurrently, the recvfrom thread checks the generic queue, finds it empty and resets the XPT_DATA bit. A subsequent svc_xprt_enqueue will fail to enqueue the transport for I/O and cause the transport to "stall". The fix is to protect both lists with the same lock and set the XPT_DATA bit with this lock held. Signed-off-by: Tom Tucker Signed-off-by: J. Bruce Fields --- include/linux/sunrpc/svc_rdma.h | 1 - net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 8 ++++---- net/sunrpc/xprtrdma/svc_rdma_transport.c | 5 ++--- 3 files changed, 6 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/include/linux/sunrpc/svc_rdma.h b/include/linux/sunrpc/svc_rdma.h index ef2e3a20bf3b..dc05b54bd3a3 100644 --- a/include/linux/sunrpc/svc_rdma.h +++ b/include/linux/sunrpc/svc_rdma.h @@ -143,7 +143,6 @@ struct svcxprt_rdma { unsigned long sc_flags; struct list_head sc_dto_q; /* DTO tasklet I/O pending Q */ struct list_head sc_read_complete_q; - spinlock_t sc_read_complete_lock; struct work_struct sc_work; }; /* sc_flags */ diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c index b4b17f44cb29..74de31a06616 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c +++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c @@ -443,18 +443,18 @@ int svc_rdma_recvfrom(struct svc_rqst *rqstp) dprintk("svcrdma: rqstp=%p\n", rqstp); - spin_lock_bh(&rdma_xprt->sc_read_complete_lock); + spin_lock_bh(&rdma_xprt->sc_rq_dto_lock); if (!list_empty(&rdma_xprt->sc_read_complete_q)) { ctxt = list_entry(rdma_xprt->sc_read_complete_q.next, struct svc_rdma_op_ctxt, dto_q); list_del_init(&ctxt->dto_q); } - spin_unlock_bh(&rdma_xprt->sc_read_complete_lock); - if (ctxt) + if (ctxt) { + spin_unlock_bh(&rdma_xprt->sc_rq_dto_lock); return rdma_read_complete(rqstp, ctxt); + } - spin_lock_bh(&rdma_xprt->sc_rq_dto_lock); if (!list_empty(&rdma_xprt->sc_rq_dto_q)) { ctxt = list_entry(rdma_xprt->sc_rq_dto_q.next, struct svc_rdma_op_ctxt, diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c index 19ddc382b777..900cb69728c6 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_transport.c +++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c @@ -359,11 +359,11 @@ static void sq_cq_reap(struct svcxprt_rdma *xprt) if (test_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags)) { struct svc_rdma_op_ctxt *read_hdr = ctxt->read_hdr; BUG_ON(!read_hdr); + spin_lock_bh(&xprt->sc_rq_dto_lock); set_bit(XPT_DATA, &xprt->sc_xprt.xpt_flags); - spin_lock_bh(&xprt->sc_read_complete_lock); list_add_tail(&read_hdr->dto_q, &xprt->sc_read_complete_q); - spin_unlock_bh(&xprt->sc_read_complete_lock); + spin_unlock_bh(&xprt->sc_rq_dto_lock); svc_xprt_enqueue(&xprt->sc_xprt); } svc_rdma_put_context(ctxt, 0); @@ -428,7 +428,6 @@ static struct svcxprt_rdma *rdma_create_xprt(struct svc_serv *serv, init_waitqueue_head(&cma_xprt->sc_send_wait); spin_lock_init(&cma_xprt->sc_lock); - spin_lock_init(&cma_xprt->sc_read_complete_lock); spin_lock_init(&cma_xprt->sc_rq_dto_lock); cma_xprt->sc_ord = svcrdma_ord; -- cgit v1.2.3 From 64c00d81b5c2491bd140b3c8eb2e8c351513f971 Mon Sep 17 00:00:00 2001 From: Andrew Gallatin Date: Wed, 13 Aug 2008 15:16:00 -0700 Subject: pktgen: prevent pktgen from using bad tx queue With the new multi-queue transmit code, it is possible to accidentally make pktgen pick a non-existing tx queue simply by using a stale script to drive pktgen. Access to this non-existing tx queue will then trigger a bad memory access and kill the machine. For example, setting "queue_map_max 2" will cause my machine to die when accessing a garbage spinlock in the non-existing tx queue: BUG: spinlock bad magic on CPU#0, kpktgend_0/564 lock: ffff88001ddf6718, .magic: ffffffff, .owner: /-1, .owner_cpu: 0 Pid: 564, comm: kpktgend_0 Not tainted 2.6.27-rc3 #35 Call Trace: [] spin_bug+0xa4/0xac [] _raw_spin_lock+0x23/0x123 [] _spin_lock_bh+0x17/0x1b [] pktgen_thread_worker+0xa97/0x1002 [] ? finish_task_switch+0x38/0x97 [] ? autoremove_wake_function+0x0/0x36 [] ? autoremove_wake_function+0x0/0x36 [] ? pktgen_thread_worker+0x0/0x1002 [] kthread+0x44/0x6d [] child_rip+0xa/0x11 [] ? kthread+0x0/0x6d [] ? child_rip+0x0/0x11 The attached patch adds some sanity checking to prevent these sorts of configuration errors. Signed-off-by: Andrew Gallatin Signed-off-by: David S. Miller --- net/core/pktgen.c | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) (limited to 'net') diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 526236453908..a756847e3814 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -1961,6 +1961,8 @@ static int pktgen_setup_dev(struct pktgen_dev *pkt_dev, const char *ifname) */ static void pktgen_setup_inject(struct pktgen_dev *pkt_dev) { + int ntxq; + if (!pkt_dev->odev) { printk(KERN_ERR "pktgen: ERROR: pkt_dev->odev == NULL in " "setup_inject.\n"); @@ -1969,6 +1971,33 @@ static void pktgen_setup_inject(struct pktgen_dev *pkt_dev) return; } + /* make sure that we don't pick a non-existing transmit queue */ + ntxq = pkt_dev->odev->real_num_tx_queues; + if (ntxq <= num_online_cpus() && (pkt_dev->flags & F_QUEUE_MAP_CPU)) { + printk(KERN_WARNING "pktgen: WARNING: QUEUE_MAP_CPU " + "disabled because CPU count (%d) exceeds number ", + num_online_cpus()); + printk(KERN_WARNING "pktgen: WARNING: of tx queues " + "(%d) on %s \n", ntxq, pkt_dev->odev->name); + pkt_dev->flags &= ~F_QUEUE_MAP_CPU; + } + if (ntxq <= pkt_dev->queue_map_min) { + printk(KERN_WARNING "pktgen: WARNING: Requested " + "queue_map_min (%d) exceeds number of tx\n", + pkt_dev->queue_map_min); + printk(KERN_WARNING "pktgen: WARNING: queues (%d) on " + "%s, resetting\n", ntxq, pkt_dev->odev->name); + pkt_dev->queue_map_min = ntxq - 1; + } + if (ntxq <= pkt_dev->queue_map_max) { + printk(KERN_WARNING "pktgen: WARNING: Requested " + "queue_map_max (%d) exceeds number of tx\n", + pkt_dev->queue_map_max); + printk(KERN_WARNING "pktgen: WARNING: queues (%d) on " + "%s, resetting\n", ntxq, pkt_dev->odev->name); + pkt_dev->queue_map_max = ntxq - 1; + } + /* Default to the interface's mac if not explicitly set. */ if (is_zero_ether_addr(pkt_dev->src_mac)) -- cgit v1.2.3 From 26b284de54a5ca3dfbe2fd9a51ac1923e80085a2 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Wed, 13 Aug 2008 15:16:43 -0700 Subject: pkt_sched: Fix oops in htb_delete. Recent changes introduced a bug in htb_delete(): cl->parent->children counter update misses checking cl->parent for NULL, which is used for root classes, so deleting them causes an oops. Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/sched/sch_htb.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index be35422711a3..6febd245e62b 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1279,7 +1279,8 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg) /* delete from hash and active; remainder in destroy_class */ qdisc_class_hash_remove(&q->clhash, &cl->common); - cl->parent->children--; + if (cl->parent) + cl->parent->children--; if (cl->prio_activity) htb_deactivate(q, cl); -- cgit v1.2.3 From b9a3b1102bc80b4044224494100f67de132d5448 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 13 Aug 2008 15:18:38 -0700 Subject: pkt_sched: Fix queue quiescence testing in dev_deactivate(). Based upon discussions with Jarek P. and Herbert Xu. First, we're testing the wrong qdisc. We just reset the device queue qdiscs to &noop_qdisc and checking it's state is completely pointless here. We want to wait until the previous qdisc that was sitting at the ->qdisc pointer is not busy any more. And that would be ->qdisc_sleeping. Because of how we propagate the samples qdisc pointer down into qdisc_run and friends via per-cpu ->output_queue and netif_schedule, we have to wait also for the __QDISC_STATE_SCHED bit to clear as well. Signed-off-by: David S. Miller --- net/sched/sch_generic.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 7cf83b37459d..468574682caa 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -647,7 +647,7 @@ static void dev_deactivate_queue(struct net_device *dev, } } -static bool some_qdisc_is_running(struct net_device *dev, int lock) +static bool some_qdisc_is_busy(struct net_device *dev, int lock) { unsigned int i; @@ -658,13 +658,14 @@ static bool some_qdisc_is_running(struct net_device *dev, int lock) int val; dev_queue = netdev_get_tx_queue(dev, i); - q = dev_queue->qdisc; + q = dev_queue->qdisc_sleeping; root_lock = qdisc_lock(q); if (lock) spin_lock_bh(root_lock); - val = test_bit(__QDISC_STATE_RUNNING, &q->state); + val = (test_bit(__QDISC_STATE_RUNNING, &q->state) || + test_bit(__QDISC_STATE_SCHED, &q->state)); if (lock) spin_unlock_bh(root_lock); @@ -689,14 +690,14 @@ void dev_deactivate(struct net_device *dev) /* Wait for outstanding qdisc_run calls. */ do { - while (some_qdisc_is_running(dev, 0)) + while (some_qdisc_is_busy(dev, 0)) yield(); /* * Double-check inside queue lock to ensure that all effects * of the queue run are visible when we return. */ - running = some_qdisc_is_running(dev, 1); + running = some_qdisc_is_busy(dev, 1); /* * The running flag should never be set at this point because -- cgit v1.2.3 From d4766692e72422f3b0f0e9ac6773d92baad07d51 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Wed, 13 Aug 2008 15:20:24 -0700 Subject: pkt_sched: Protect gen estimators under est_lock. gen_kill_estimator() required rtnl_lock() protection, but since it is moved to an RCU callback __qdisc_destroy() let's use est_lock instead. Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/core/gen_estimator.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c index 57abe8266be1..a89f32fa94f6 100644 --- a/net/core/gen_estimator.c +++ b/net/core/gen_estimator.c @@ -99,7 +99,7 @@ struct gen_estimator_head static struct gen_estimator_head elist[EST_MAX_INTERVAL+1]; -/* Protects against NULL dereference */ +/* Protects against NULL dereference and RCU write-side */ static DEFINE_RWLOCK(est_lock); static void est_timer(unsigned long arg) @@ -185,6 +185,7 @@ int gen_new_estimator(struct gnet_stats_basic *bstats, est->last_packets = bstats->packets; est->avpps = rate_est->pps<<10; + write_lock_bh(&est_lock); if (!elist[idx].timer.function) { INIT_LIST_HEAD(&elist[idx].list); setup_timer(&elist[idx].timer, est_timer, idx); @@ -194,6 +195,7 @@ int gen_new_estimator(struct gnet_stats_basic *bstats, mod_timer(&elist[idx].timer, jiffies + ((HZ/4) << idx)); list_add_rcu(&est->list, &elist[idx].list); + write_unlock_bh(&est_lock); return 0; } @@ -212,7 +214,6 @@ static void __gen_kill_estimator(struct rcu_head *head) * Removes the rate estimator specified by &bstats and &rate_est * and deletes the timer. * - * NOTE: Called under rtnl_mutex */ void gen_kill_estimator(struct gnet_stats_basic *bstats, struct gnet_stats_rate_est *rate_est) @@ -226,17 +227,17 @@ void gen_kill_estimator(struct gnet_stats_basic *bstats, if (!elist[idx].timer.function) continue; + write_lock_bh(&est_lock); list_for_each_entry_safe(e, n, &elist[idx].list, list) { if (e->rate_est != rate_est || e->bstats != bstats) continue; - write_lock_bh(&est_lock); e->bstats = NULL; - write_unlock_bh(&est_lock); list_del_rcu(&e->list); call_rcu(&e->e_rcu, __gen_kill_estimator); } + write_unlock_bh(&est_lock); } } -- cgit v1.2.3 From 877acedc0d3ea07f7b36573ed2f1f479c2c1eefd Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Wed, 13 Aug 2008 16:15:57 -0700 Subject: netns: Fix crash by making igmp per namespace This patch makes the multicast socket to be per namespace. When a network namespace is created, other than the init_net and a multicast packet is received, the kernel goes to a hang or a kernel panic. How to reproduce ? * create a child network namespace * create a pair virtual device veth * ip link add type veth * move one side to the pair network device to the child namespace * ip link set netns dev veth1 * ping -I veth0 224.0.0.1 The bug appears because the function ip_mc_init_dev does not initialize the different multicast fields as it exits because it is not the init_net. BUG: soft lockup - CPU#0 stuck for 61s! [avahi-daemon:2695] Modules linked in: irq event stamp: 50350 hardirqs last enabled at (50349): [] _spin_unlock_irqrestore+0x34/0x39 hardirqs last disabled at (50350): [] schedule+0x9f/0x5ff softirqs last enabled at (45712): [] ip_setsockopt+0x8e7/0x909 softirqs last disabled at (45710): [] _spin_lock_bh+0x8/0x27 Pid: 2695, comm: avahi-daemon Not tainted (2.6.27-rc2-00029-g0872073 #3) EIP: 0060:[] EFLAGS: 00000297 CPU: 0 EIP is at __read_lock_failed+0x8/0x10 EAX: c4f38810 EBX: c4f38810 ECX: 00000000 EDX: c04cc22e ESI: fb0000e0 EDI: 00000011 EBP: 0f02000a ESP: c4e3faa0 DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 CR0: 8005003b CR2: 44618a40 CR3: 04e37000 CR4: 000006d0 DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 DR6: ffff0ff0 DR7: 00000400 [] ? _raw_read_lock+0x23/0x25 [] ? ip_check_mc+0x1c/0x83 [] ? ip_route_input+0x229/0xe92 [] ? trace_hardirqs_on_thunk+0xc/0x10 [] ? do_IRQ+0x69/0x7d [] ? restore_nocheck_notrace+0x0/0xe [] ? ip_rcv+0x227/0x505 [] ? netif_receive_skb+0xfe/0x2b3 [] ? netif_receive_skb+0x26c/0x2b3 [] ? process_backlog+0x73/0xbd [] ? net_rx_action+0xc1/0x1ae [] ? __do_softirq+0x7b/0xef [] ? do_softirq+0x37/0x4d [] ? dev_queue_xmit+0x3d4/0x40b [] ? local_bh_enable+0x96/0xab [] ? dev_queue_xmit+0x3d4/0x40b [] ? _local_bh_enable+0x79/0x88 [] ? neigh_resolve_output+0x20f/0x239 [] ? ip_finish_output+0x1df/0x209 [] ? ip_dev_loopback_xmit+0x62/0x66 [] ? ip_local_out+0x15/0x17 [] ? ip_push_pending_frames+0x25c/0x2bb [] ? udp_push_pending_frames+0x2bb/0x30e [] ? udp_sendmsg+0x413/0x51d [] ? udp_sendmsg+0x433/0x51d [] ? inet_sendmsg+0x35/0x3f [] ? sock_sendmsg+0xb8/0xd1 [] ? autoremove_wake_function+0x0/0x2b [] ? copy_from_user+0x32/0x5e [] ? copy_from_user+0x32/0x5e [] ? sys_sendmsg+0x18d/0x1f0 [] ? pipe_write+0x3cb/0x3d7 [] ? do_sync_write+0xbe/0x105 [] ? autoremove_wake_function+0x0/0x2b [] ? sys_socketcall+0x176/0x1b0 [] ? syscall_trace_enter+0x6c/0x7b [] ? syscall_call+0x7/0xb Signed-off-by: Daniel Lezcano Signed-off-by: David S. Miller --- net/ipv4/igmp.c | 71 ++++++++++++++++----------------------------------------- 1 file changed, 20 insertions(+), 51 deletions(-) (limited to 'net') diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 6203ece53606..f70fac612596 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -289,6 +289,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, int size) struct rtable *rt; struct iphdr *pip; struct igmpv3_report *pig; + struct net *net = dev_net(dev); skb = alloc_skb(size + LL_ALLOCATED_SPACE(dev), GFP_ATOMIC); if (skb == NULL) @@ -299,7 +300,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, int size) .nl_u = { .ip4_u = { .daddr = IGMPV3_ALL_MCR } }, .proto = IPPROTO_IGMP }; - if (ip_route_output_key(&init_net, &rt, &fl)) { + if (ip_route_output_key(net, &rt, &fl)) { kfree_skb(skb); return NULL; } @@ -629,6 +630,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc, struct igmphdr *ih; struct rtable *rt; struct net_device *dev = in_dev->dev; + struct net *net = dev_net(dev); __be32 group = pmc ? pmc->multiaddr : 0; __be32 dst; @@ -643,7 +645,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc, struct flowi fl = { .oif = dev->ifindex, .nl_u = { .ip4_u = { .daddr = dst } }, .proto = IPPROTO_IGMP }; - if (ip_route_output_key(&init_net, &rt, &fl)) + if (ip_route_output_key(net, &rt, &fl)) return -1; } if (rt->rt_src == 0) { @@ -1196,9 +1198,6 @@ void ip_mc_inc_group(struct in_device *in_dev, __be32 addr) ASSERT_RTNL(); - if (!net_eq(dev_net(in_dev->dev), &init_net)) - return; - for (im=in_dev->mc_list; im; im=im->next) { if (im->multiaddr == addr) { im->users++; @@ -1278,9 +1277,6 @@ void ip_mc_dec_group(struct in_device *in_dev, __be32 addr) ASSERT_RTNL(); - if (!net_eq(dev_net(in_dev->dev), &init_net)) - return; - for (ip=&in_dev->mc_list; (i=*ip)!=NULL; ip=&i->next) { if (i->multiaddr==addr) { if (--i->users == 0) { @@ -1308,9 +1304,6 @@ void ip_mc_down(struct in_device *in_dev) ASSERT_RTNL(); - if (!net_eq(dev_net(in_dev->dev), &init_net)) - return; - for (i=in_dev->mc_list; i; i=i->next) igmp_group_dropped(i); @@ -1331,9 +1324,6 @@ void ip_mc_init_dev(struct in_device *in_dev) { ASSERT_RTNL(); - if (!net_eq(dev_net(in_dev->dev), &init_net)) - return; - in_dev->mc_tomb = NULL; #ifdef CONFIG_IP_MULTICAST in_dev->mr_gq_running = 0; @@ -1357,9 +1347,6 @@ void ip_mc_up(struct in_device *in_dev) ASSERT_RTNL(); - if (!net_eq(dev_net(in_dev->dev), &init_net)) - return; - ip_mc_inc_group(in_dev, IGMP_ALL_HOSTS); for (i=in_dev->mc_list; i; i=i->next) @@ -1376,9 +1363,6 @@ void ip_mc_destroy_dev(struct in_device *in_dev) ASSERT_RTNL(); - if (!net_eq(dev_net(in_dev->dev), &init_net)) - return; - /* Deactivate timers */ ip_mc_down(in_dev); @@ -1395,7 +1379,7 @@ void ip_mc_destroy_dev(struct in_device *in_dev) write_unlock_bh(&in_dev->mc_list_lock); } -static struct in_device * ip_mc_find_dev(struct ip_mreqn *imr) +static struct in_device *ip_mc_find_dev(struct net *net, struct ip_mreqn *imr) { struct flowi fl = { .nl_u = { .ip4_u = { .daddr = imr->imr_multiaddr.s_addr } } }; @@ -1404,19 +1388,19 @@ static struct in_device * ip_mc_find_dev(struct ip_mreqn *imr) struct in_device *idev = NULL; if (imr->imr_ifindex) { - idev = inetdev_by_index(&init_net, imr->imr_ifindex); + idev = inetdev_by_index(net, imr->imr_ifindex); if (idev) __in_dev_put(idev); return idev; } if (imr->imr_address.s_addr) { - dev = ip_dev_find(&init_net, imr->imr_address.s_addr); + dev = ip_dev_find(net, imr->imr_address.s_addr); if (!dev) return NULL; dev_put(dev); } - if (!dev && !ip_route_output_key(&init_net, &rt, &fl)) { + if (!dev && !ip_route_output_key(net, &rt, &fl)) { dev = rt->u.dst.dev; ip_rt_put(rt); } @@ -1754,18 +1738,16 @@ int ip_mc_join_group(struct sock *sk , struct ip_mreqn *imr) struct ip_mc_socklist *iml=NULL, *i; struct in_device *in_dev; struct inet_sock *inet = inet_sk(sk); + struct net *net = sock_net(sk); int ifindex; int count = 0; if (!ipv4_is_multicast(addr)) return -EINVAL; - if (!net_eq(sock_net(sk), &init_net)) - return -EPROTONOSUPPORT; - rtnl_lock(); - in_dev = ip_mc_find_dev(imr); + in_dev = ip_mc_find_dev(net, imr); if (!in_dev) { iml = NULL; @@ -1827,15 +1809,13 @@ int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr) struct inet_sock *inet = inet_sk(sk); struct ip_mc_socklist *iml, **imlp; struct in_device *in_dev; + struct net *net = sock_net(sk); __be32 group = imr->imr_multiaddr.s_addr; u32 ifindex; int ret = -EADDRNOTAVAIL; - if (!net_eq(sock_net(sk), &init_net)) - return -EPROTONOSUPPORT; - rtnl_lock(); - in_dev = ip_mc_find_dev(imr); + in_dev = ip_mc_find_dev(net, imr); ifindex = imr->imr_ifindex; for (imlp = &inet->mc_list; (iml = *imlp) != NULL; imlp = &iml->next) { if (iml->multi.imr_multiaddr.s_addr != group) @@ -1873,21 +1853,19 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct struct in_device *in_dev = NULL; struct inet_sock *inet = inet_sk(sk); struct ip_sf_socklist *psl; + struct net *net = sock_net(sk); int leavegroup = 0; int i, j, rv; if (!ipv4_is_multicast(addr)) return -EINVAL; - if (!net_eq(sock_net(sk), &init_net)) - return -EPROTONOSUPPORT; - rtnl_lock(); imr.imr_multiaddr.s_addr = mreqs->imr_multiaddr; imr.imr_address.s_addr = mreqs->imr_interface; imr.imr_ifindex = ifindex; - in_dev = ip_mc_find_dev(&imr); + in_dev = ip_mc_find_dev(net, &imr); if (!in_dev) { err = -ENODEV; @@ -2007,6 +1985,7 @@ int ip_mc_msfilter(struct sock *sk, struct ip_msfilter *msf, int ifindex) struct in_device *in_dev; struct inet_sock *inet = inet_sk(sk); struct ip_sf_socklist *newpsl, *psl; + struct net *net = sock_net(sk); int leavegroup = 0; if (!ipv4_is_multicast(addr)) @@ -2015,15 +1994,12 @@ int ip_mc_msfilter(struct sock *sk, struct ip_msfilter *msf, int ifindex) msf->imsf_fmode != MCAST_EXCLUDE) return -EINVAL; - if (!net_eq(sock_net(sk), &init_net)) - return -EPROTONOSUPPORT; - rtnl_lock(); imr.imr_multiaddr.s_addr = msf->imsf_multiaddr; imr.imr_address.s_addr = msf->imsf_interface; imr.imr_ifindex = ifindex; - in_dev = ip_mc_find_dev(&imr); + in_dev = ip_mc_find_dev(net, &imr); if (!in_dev) { err = -ENODEV; @@ -2094,19 +2070,17 @@ int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf, struct in_device *in_dev; struct inet_sock *inet = inet_sk(sk); struct ip_sf_socklist *psl; + struct net *net = sock_net(sk); if (!ipv4_is_multicast(addr)) return -EINVAL; - if (!net_eq(sock_net(sk), &init_net)) - return -EPROTONOSUPPORT; - rtnl_lock(); imr.imr_multiaddr.s_addr = msf->imsf_multiaddr; imr.imr_address.s_addr = msf->imsf_interface; imr.imr_ifindex = 0; - in_dev = ip_mc_find_dev(&imr); + in_dev = ip_mc_find_dev(net, &imr); if (!in_dev) { err = -ENODEV; @@ -2163,9 +2137,6 @@ int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf, if (!ipv4_is_multicast(addr)) return -EINVAL; - if (!net_eq(sock_net(sk), &init_net)) - return -EPROTONOSUPPORT; - rtnl_lock(); err = -EADDRNOTAVAIL; @@ -2246,19 +2217,17 @@ void ip_mc_drop_socket(struct sock *sk) { struct inet_sock *inet = inet_sk(sk); struct ip_mc_socklist *iml; + struct net *net = sock_net(sk); if (inet->mc_list == NULL) return; - if (!net_eq(sock_net(sk), &init_net)) - return; - rtnl_lock(); while ((iml = inet->mc_list) != NULL) { struct in_device *in_dev; inet->mc_list = iml->next; - in_dev = inetdev_by_index(&init_net, iml->multi.imr_ifindex); + in_dev = inetdev_by_index(net, iml->multi.imr_ifindex); (void) ip_mc_leave_src(sk, iml, in_dev); if (in_dev != NULL) { ip_mc_dec_group(in_dev, iml->multi.imr_multiaddr.s_addr); -- cgit v1.2.3 From 191cd582500f49b32a63040fedeebb0168c720af Mon Sep 17 00:00:00 2001 From: Brian Haley Date: Thu, 14 Aug 2008 15:33:21 -0700 Subject: netns: Add network namespace argument to rt6_fill_node() and ipv6_dev_get_saddr() ipv6_dev_get_saddr() blindly de-references dst_dev to get the network namespace, but some callers might pass NULL. Change callers to pass a namespace pointer instead. Signed-off-by: Brian Haley Signed-off-by: David S. Miller --- include/net/addrconf.h | 3 ++- include/net/ip6_route.h | 1 + net/ipv6/addrconf.c | 3 +-- net/ipv6/fib6_rules.c | 3 ++- net/ipv6/ip6_fib.c | 1 + net/ipv6/ip6_output.c | 2 +- net/ipv6/ndisc.c | 2 +- net/ipv6/route.c | 12 +++++++----- net/ipv6/xfrm6_policy.c | 4 +++- net/sctp/ipv6.c | 3 ++- 10 files changed, 21 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 06b28142b3ab..c216de528b08 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -80,7 +80,8 @@ extern struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct net_device *dev, int strict); -extern int ipv6_dev_get_saddr(struct net_device *dev, +extern int ipv6_dev_get_saddr(struct net *net, + struct net_device *dev, const struct in6_addr *daddr, unsigned int srcprefs, struct in6_addr *saddr); diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index bc391ba101e9..5f53db7e4e57 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -107,6 +107,7 @@ struct rt6_rtnl_dump_arg { struct sk_buff *skb; struct netlink_callback *cb; + struct net *net; }; extern int rt6_dump_route(struct rt6_info *rt, void *p_arg); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index a7842c54f58a..e2d3b7580b76 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1106,13 +1106,12 @@ out: return ret; } -int ipv6_dev_get_saddr(struct net_device *dst_dev, +int ipv6_dev_get_saddr(struct net *net, struct net_device *dst_dev, const struct in6_addr *daddr, unsigned int prefs, struct in6_addr *saddr) { struct ipv6_saddr_score scores[2], *score = &scores[0], *hiscore = &scores[1]; - struct net *net = dev_net(dst_dev); struct ipv6_saddr_dst dst; struct net_device *dev; int dst_type; diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 8d05527524e3..f5de3f9dc692 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -93,7 +93,8 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, if (flags & RT6_LOOKUP_F_SRCPREF_COA) srcprefs |= IPV6_PREFER_SRC_COA; - if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev, + if (ipv6_dev_get_saddr(net, + ip6_dst_idev(&rt->u.dst)->dev, &flp->fl6_dst, srcprefs, &saddr)) goto again; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 52dddc25d3e6..29c7c99e69f7 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -378,6 +378,7 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) arg.skb = skb; arg.cb = cb; + arg.net = net; w->args = &arg; for (h = s_h; h < FIB_TABLE_HASHSZ; h++, s_e = 0) { diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index a4402de425d9..0e844c2736a7 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -934,7 +934,7 @@ static int ip6_dst_lookup_tail(struct sock *sk, goto out_err_release; if (ipv6_addr_any(&fl->fl6_src)) { - err = ipv6_dev_get_saddr(ip6_dst_idev(*dst)->dev, + err = ipv6_dev_get_saddr(net, ip6_dst_idev(*dst)->dev, &fl->fl6_dst, sk ? inet6_sk(sk)->srcprefs : 0, &fl->fl6_src); diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index beb48e3f038a..f1c62ba0f56b 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -549,7 +549,7 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh, override = 0; in6_ifa_put(ifp); } else { - if (ipv6_dev_get_saddr(dev, daddr, + if (ipv6_dev_get_saddr(dev_net(dev), dev, daddr, inet6_sk(dev_net(dev)->ipv6.ndisc_sk)->srcprefs, &tmpaddr)) return; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 41b165ffb369..9af6115f0f50 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2106,7 +2106,8 @@ static inline size_t rt6_nlmsg_size(void) + nla_total_size(sizeof(struct rta_cacheinfo)); } -static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, +static int rt6_fill_node(struct net *net, + struct sk_buff *skb, struct rt6_info *rt, struct in6_addr *dst, struct in6_addr *src, int iif, int type, u32 pid, u32 seq, int prefix, int nowait, unsigned int flags) @@ -2189,7 +2190,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, } else if (dst) { struct inet6_dev *idev = ip6_dst_idev(&rt->u.dst); struct in6_addr saddr_buf; - if (ipv6_dev_get_saddr(idev ? idev->dev : NULL, + if (ipv6_dev_get_saddr(net, idev ? idev->dev : NULL, dst, 0, &saddr_buf) == 0) NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf); } @@ -2234,7 +2235,8 @@ int rt6_dump_route(struct rt6_info *rt, void *p_arg) } else prefix = 0; - return rt6_fill_node(arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE, + return rt6_fill_node(arg->net, + arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE, NETLINK_CB(arg->cb->skb).pid, arg->cb->nlh->nlmsg_seq, prefix, 0, NLM_F_MULTI); } @@ -2300,7 +2302,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void rt = (struct rt6_info*) ip6_route_output(net, NULL, &fl); skb->dst = &rt->u.dst; - err = rt6_fill_node(skb, rt, &fl.fl6_dst, &fl.fl6_src, iif, + err = rt6_fill_node(net, skb, rt, &fl.fl6_dst, &fl.fl6_src, iif, RTM_NEWROUTE, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, 0, 0, 0); if (err < 0) { @@ -2327,7 +2329,7 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) if (skb == NULL) goto errout; - err = rt6_fill_node(skb, rt, NULL, NULL, 0, + err = rt6_fill_node(net, skb, rt, NULL, NULL, 0, event, info->pid, seq, 0, 0, 0); if (err < 0) { /* -EMSGSIZE implies BUG in rt6_nlmsg_size() */ diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 8f1e0543b3c4..08e4cbbe3f04 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -52,12 +52,14 @@ static struct dst_entry *xfrm6_dst_lookup(int tos, xfrm_address_t *saddr, static int xfrm6_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr) { struct dst_entry *dst; + struct net_device *dev; dst = xfrm6_dst_lookup(0, NULL, daddr); if (IS_ERR(dst)) return -EHOSTUNREACH; - ipv6_dev_get_saddr(ip6_dst_idev(dst)->dev, + dev = ip6_dst_idev(dst)->dev; + ipv6_dev_get_saddr(dev_net(dev), dev, (struct in6_addr *)&daddr->a6, 0, (struct in6_addr *)&saddr->a6); dst_release(dst); diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 483a01d0740a..47f91afa0211 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -319,7 +319,8 @@ static void sctp_v6_get_saddr(struct sctp_sock *sk, __func__, asoc, dst, NIP6(daddr->v6.sin6_addr)); if (!asoc) { - ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL, + ipv6_dev_get_saddr(sock_net(sctp_opt2sk(sk)), + dst ? ip6_dst_idev(dst)->dev : NULL, &daddr->v6.sin6_addr, inet6_sk(&sk->inet.sk)->srcprefs, &saddr->v6.sin6_addr); -- cgit v1.2.3 From 323c048836f73a11ded6f9743feda21c00465cb0 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Thu, 14 Aug 2008 17:01:10 -0700 Subject: pkt_sched: Fix unlocking in tc_ctl_tfilter() Fix a bug with spin_lock_bh() inserted instead of spin_unlock_bh() by some recent patch. Reported-by: Denys Fedoryshchenko Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/sched/cls_api.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index d2b6f54a6261..5cafdd4c8018 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -280,7 +280,7 @@ replay: if (n->nlmsg_type == RTM_DELTFILTER && t->tcm_handle == 0) { spin_lock_bh(root_lock); *back = tp->next; - spin_lock_bh(root_lock); + spin_unlock_bh(root_lock); tfilter_notify(skb, n, tp, fh, RTM_DELTFILTER); tcf_destroy(tp); -- cgit v1.2.3 From c6153b5b77650879d78dec76414213c76dd8d574 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 15 Aug 2008 13:44:31 -0700 Subject: ipv4: Disable route secret interval on zero interval Let me first state that disabling the route cache hash rebuild should not be done without extensive analysis on the risk profile and careful deliberation. However, there are times when this can be done safely or for testing. For example, when you have mechanisms for ensuring that offending parties do not exist in your network. This patch lets the user disable the rebuild if the interval is set to zero. This also incidentally fixes a divide-by-zero error with name-spaces. In addition, this patch makes the effect of an interval change immediate rather than it taking effect at the next rebuild as is currently the case. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/route.c | 76 +++++++++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 70 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 16fc6f454a31..cca921ea8550 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2914,6 +2914,68 @@ static int ipv4_sysctl_rtcache_flush_strategy(ctl_table *table, return 0; } +static void rt_secret_reschedule(int old) +{ + struct net *net; + int new = ip_rt_secret_interval; + int diff = new - old; + + if (!diff) + return; + + rtnl_lock(); + for_each_net(net) { + int deleted = del_timer_sync(&net->ipv4.rt_secret_timer); + + if (!new) + continue; + + if (deleted) { + long time = net->ipv4.rt_secret_timer.expires - jiffies; + + if (time <= 0 || (time += diff) <= 0) + time = 0; + + net->ipv4.rt_secret_timer.expires = time; + } else + net->ipv4.rt_secret_timer.expires = new; + + net->ipv4.rt_secret_timer.expires += jiffies; + add_timer(&net->ipv4.rt_secret_timer); + } + rtnl_unlock(); +} + +static int ipv4_sysctl_rt_secret_interval(ctl_table *ctl, int write, + struct file *filp, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + int old = ip_rt_secret_interval; + int ret = proc_dointvec_jiffies(ctl, write, filp, buffer, lenp, ppos); + + rt_secret_reschedule(old); + + return ret; +} + +static int ipv4_sysctl_rt_secret_interval_strategy(ctl_table *table, + int __user *name, + int nlen, + void __user *oldval, + size_t __user *oldlenp, + void __user *newval, + size_t newlen) +{ + int old = ip_rt_secret_interval; + int ret = sysctl_jiffies(table, name, nlen, oldval, oldlenp, newval, + newlen); + + rt_secret_reschedule(old); + + return ret; +} + static ctl_table ipv4_route_table[] = { { .ctl_name = NET_IPV4_ROUTE_GC_THRESH, @@ -3048,8 +3110,8 @@ static ctl_table ipv4_route_table[] = { .data = &ip_rt_secret_interval, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec_jiffies, - .strategy = &sysctl_jiffies, + .proc_handler = &ipv4_sysctl_rt_secret_interval, + .strategy = &ipv4_sysctl_rt_secret_interval_strategy, }, { .ctl_name = 0 } }; @@ -3126,10 +3188,12 @@ static __net_init int rt_secret_timer_init(struct net *net) net->ipv4.rt_secret_timer.data = (unsigned long)net; init_timer_deferrable(&net->ipv4.rt_secret_timer); - net->ipv4.rt_secret_timer.expires = - jiffies + net_random() % ip_rt_secret_interval + - ip_rt_secret_interval; - add_timer(&net->ipv4.rt_secret_timer); + if (ip_rt_secret_interval) { + net->ipv4.rt_secret_timer.expires = + jiffies + net_random() % ip_rt_secret_interval + + ip_rt_secret_interval; + add_timer(&net->ipv4.rt_secret_timer); + } return 0; } -- cgit v1.2.3 From e4119a43187139736750bad5d694c6a839df045d Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 15 Aug 2008 19:51:07 -0700 Subject: bridge: show offload settings Add more ethtool generic operations to dump the bridge offload settings. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/bridge/br_device.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 9b58d70b0e7d..4f52c3d50ebe 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -148,11 +148,16 @@ static int br_set_tx_csum(struct net_device *dev, u32 data) } static struct ethtool_ops br_ethtool_ops = { - .get_drvinfo = br_getinfo, - .get_link = ethtool_op_get_link, - .set_sg = br_set_sg, - .set_tx_csum = br_set_tx_csum, - .set_tso = br_set_tso, + .get_drvinfo = br_getinfo, + .get_link = ethtool_op_get_link, + .get_tx_csum = ethtool_op_get_tx_csum, + .set_tx_csum = br_set_tx_csum, + .get_sg = ethtool_op_get_sg, + .set_sg = br_set_sg, + .get_tso = ethtool_op_get_tso, + .set_tso = br_set_tso, + .get_ufo = ethtool_op_get_ufo, + .get_flags = ethtool_op_get_flags, }; void br_dev_setup(struct net_device *dev) -- cgit v1.2.3 From 6f85a124d819e1cf33b16d064a6a656fd448a735 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 15 Aug 2008 14:55:02 -0700 Subject: net: Preserve netfilter attributes in skb_gso_segment using __copy_skb_header skb_gso_segment didn't preserve some attributes in the original skb such as the netfilter fields. This was harmless until they were used which is the case for packets going through lo. This patch makes it call __copy_skb_header which also picks up some other missing attributes. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/core/skbuff.c | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 84640172d65d..ca1ccdf1ef76 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2256,14 +2256,7 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features) segs = nskb; tail = nskb; - nskb->dev = skb->dev; - skb_copy_queue_mapping(nskb, skb); - nskb->priority = skb->priority; - nskb->protocol = skb->protocol; - nskb->vlan_tci = skb->vlan_tci; - nskb->dst = dst_clone(skb->dst); - memcpy(nskb->cb, skb->cb, sizeof(skb->cb)); - nskb->pkt_type = skb->pkt_type; + __copy_skb_header(nskb, skb); nskb->mac_len = skb->mac_len; skb_reserve(nskb, headroom); @@ -2274,6 +2267,7 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features) skb_copy_from_linear_data(skb, skb_put(nskb, doffset), doffset); if (!sg) { + nskb->ip_summed = CHECKSUM_NONE; nskb->csum = skb_copy_and_csum_bits(skb, offset, skb_put(nskb, len), len, 0); @@ -2283,8 +2277,6 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features) frag = skb_shinfo(nskb)->frags; k = 0; - nskb->ip_summed = CHECKSUM_PARTIAL; - nskb->csum = skb->csum; skb_copy_from_linear_data_offset(skb, offset, skb_put(nskb, hsize), hsize); -- cgit v1.2.3 From db543c1f973cd1d557cc32ceee76737c1e4d2898 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Fri, 15 Aug 2008 15:13:53 -0700 Subject: net: skb_copy_datagram_from_iovec() There's an skb_copy_datagram_iovec() to copy out of a paged skb, but nothing the other way around (because we don't do that). We want to allocate big skbs in tun.c, so let's add the function. It's a carbon copy of skb_copy_datagram_iovec() with enough changes to be annoying. Signed-off-by: Rusty Russell Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 +++ net/core/datagram.c | 87 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 91 insertions(+) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 358661c9990e..909923717830 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1452,6 +1452,10 @@ extern int skb_copy_datagram_iovec(const struct sk_buff *from, extern int skb_copy_and_csum_datagram_iovec(struct sk_buff *skb, int hlen, struct iovec *iov); +extern int skb_copy_datagram_from_iovec(struct sk_buff *skb, + int offset, + struct iovec *from, + int len); extern void skb_free_datagram(struct sock *sk, struct sk_buff *skb); extern int skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags); diff --git a/net/core/datagram.c b/net/core/datagram.c index dd61dcad6019..52f577a0f544 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -339,6 +339,93 @@ fault: return -EFAULT; } +/** + * skb_copy_datagram_from_iovec - Copy a datagram from an iovec. + * @skb: buffer to copy + * @offset: offset in the buffer to start copying to + * @from: io vector to copy to + * @len: amount of data to copy to buffer from iovec + * + * Returns 0 or -EFAULT. + * Note: the iovec is modified during the copy. + */ +int skb_copy_datagram_from_iovec(struct sk_buff *skb, int offset, + struct iovec *from, int len) +{ + int start = skb_headlen(skb); + int i, copy = start - offset; + + /* Copy header. */ + if (copy > 0) { + if (copy > len) + copy = len; + if (memcpy_fromiovec(skb->data + offset, from, copy)) + goto fault; + if ((len -= copy) == 0) + return 0; + offset += copy; + } + + /* Copy paged appendix. Hmm... why does this look so complicated? */ + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + int end; + + WARN_ON(start > offset + len); + + end = start + skb_shinfo(skb)->frags[i].size; + if ((copy = end - offset) > 0) { + int err; + u8 *vaddr; + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + struct page *page = frag->page; + + if (copy > len) + copy = len; + vaddr = kmap(page); + err = memcpy_fromiovec(vaddr + frag->page_offset + + offset - start, from, copy); + kunmap(page); + if (err) + goto fault; + + if (!(len -= copy)) + return 0; + offset += copy; + } + start = end; + } + + if (skb_shinfo(skb)->frag_list) { + struct sk_buff *list = skb_shinfo(skb)->frag_list; + + for (; list; list = list->next) { + int end; + + WARN_ON(start > offset + len); + + end = start + list->len; + if ((copy = end - offset) > 0) { + if (copy > len) + copy = len; + if (skb_copy_datagram_from_iovec(list, + offset - start, + from, copy)) + goto fault; + if ((len -= copy) == 0) + return 0; + offset += copy; + } + start = end; + } + } + if (!len) + return 0; + +fault: + return -EFAULT; +} +EXPORT_SYMBOL(skb_copy_datagram_from_iovec); + static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset, u8 __user *to, int len, __wsum *csump) -- cgit v1.2.3 From a9312ae89324438b0edc554eb36c3ec6bf927d04 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 17 Aug 2008 21:51:03 -0700 Subject: pkt_sched: Add 'deactivated' state. This new state lets dev_deactivate() mark a qdisc as having been deactivated. dev_queue_xmit() and ing_filter() check for this bit and do not try to process the qdisc if the bit is set. dev_deactivate() polls the qdisc after setting the bit, waiting for both __QDISC_STATE_RUNNING and __QDISC_STATE_SCHED to clear. This isn't perfect yet, but subsequent changesets will make it so. This part is just one piece of the puzzle. Signed-off-by: David S. Miller --- include/net/sch_generic.h | 1 + net/core/dev.c | 9 ++++++++- net/sched/sch_generic.c | 6 ++++++ 3 files changed, 15 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index a7abfda3e447..757ab087adbf 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -27,6 +27,7 @@ enum qdisc_state_t { __QDISC_STATE_RUNNING, __QDISC_STATE_SCHED, + __QDISC_STATE_DEACTIVATED, }; struct qdisc_size_table { diff --git a/net/core/dev.c b/net/core/dev.c index 600bb23c4c2e..d9e31f63aded 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1800,6 +1800,12 @@ gso: spin_lock(root_lock); + if (unlikely(test_bit(__QDISC_STATE_DEACTIVATED, &q->state))) { + spin_unlock(root_lock); + rc = NET_XMIT_DROP; + goto out_kfree_skb; + } + rc = qdisc_enqueue_root(skb, q); qdisc_run(q); @@ -2084,7 +2090,8 @@ static int ing_filter(struct sk_buff *skb) q = rxq->qdisc; if (q != &noop_qdisc) { spin_lock(qdisc_lock(q)); - result = qdisc_enqueue_root(skb, q); + if (likely(!test_bit(__QDISC_STATE_DEACTIVATED, &q->state))) + result = qdisc_enqueue_root(skb, q); spin_unlock(qdisc_lock(q)); } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 468574682caa..ff1c4557e5f8 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -597,6 +597,9 @@ static void transition_one_qdisc(struct net_device *dev, struct Qdisc *new_qdisc = dev_queue->qdisc_sleeping; int *need_watchdog_p = _need_watchdog; + if (!(new_qdisc->flags & TCQ_F_BUILTIN)) + clear_bit(__QDISC_STATE_DEACTIVATED, &new_qdisc->state); + rcu_assign_pointer(dev_queue->qdisc, new_qdisc); if (need_watchdog_p && new_qdisc != &noqueue_qdisc) *need_watchdog_p = 1; @@ -640,6 +643,9 @@ static void dev_deactivate_queue(struct net_device *dev, if (qdisc) { spin_lock_bh(qdisc_lock(qdisc)); + if (!(qdisc->flags & TCQ_F_BUILTIN)) + set_bit(__QDISC_STATE_DEACTIVATED, &qdisc->state); + dev_queue->qdisc = qdisc_default; qdisc_reset(qdisc); -- cgit v1.2.3 From def82a1db1fdc4f861c77009e2ee86870c3743b0 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Sun, 17 Aug 2008 21:54:43 -0700 Subject: net: Change handling of the __QDISC_STATE_SCHED flag in net_tx_action(). Change handling of the __QDISC_STATE_SCHED flag in net_tx_action() to enable proper control in dev_deactivate(). Now, if this flag is seen as unset under root_lock means a qdisc can't be netif_scheduled. Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/core/dev.c | 34 +++++++++++++++++++--------------- 1 file changed, 19 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index d9e31f63aded..819f0175bdc9 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1339,19 +1339,23 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev) } -void __netif_schedule(struct Qdisc *q) +static inline void __netif_reschedule(struct Qdisc *q) { - if (!test_and_set_bit(__QDISC_STATE_SCHED, &q->state)) { - struct softnet_data *sd; - unsigned long flags; + struct softnet_data *sd; + unsigned long flags; - local_irq_save(flags); - sd = &__get_cpu_var(softnet_data); - q->next_sched = sd->output_queue; - sd->output_queue = q; - raise_softirq_irqoff(NET_TX_SOFTIRQ); - local_irq_restore(flags); - } + local_irq_save(flags); + sd = &__get_cpu_var(softnet_data); + q->next_sched = sd->output_queue; + sd->output_queue = q; + raise_softirq_irqoff(NET_TX_SOFTIRQ); + local_irq_restore(flags); +} + +void __netif_schedule(struct Qdisc *q) +{ + if (!test_and_set_bit(__QDISC_STATE_SCHED, &q->state)) + __netif_reschedule(q); } EXPORT_SYMBOL(__netif_schedule); @@ -1980,15 +1984,15 @@ static void net_tx_action(struct softirq_action *h) head = head->next_sched; - smp_mb__before_clear_bit(); - clear_bit(__QDISC_STATE_SCHED, &q->state); - root_lock = qdisc_lock(q); if (spin_trylock(root_lock)) { + smp_mb__before_clear_bit(); + clear_bit(__QDISC_STATE_SCHED, + &q->state); qdisc_run(q); spin_unlock(root_lock); } else { - __netif_schedule(q); + __netif_reschedule(q); } } } -- cgit v1.2.3 From 4335cd2da1e8986fa8aff21a91144d986cb0a5fc Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 17 Aug 2008 21:58:07 -0700 Subject: pkt_sched: Simplify dev_deactivate() polling loop. The condition under which the previous qdisc has no more references after we've attached &noop_qdisc is that both RUNNING and SCHED are both seen clear while holding the root lock. So just make specifically that check in the polling loop, instead of this overly complex "check without then check with lock held" sequence. Signed-off-by: David S. Miller --- net/sched/sch_generic.c | 31 +++++-------------------------- 1 file changed, 5 insertions(+), 26 deletions(-) (limited to 'net') diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index ff1c4557e5f8..30b76aec723b 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -653,7 +653,7 @@ static void dev_deactivate_queue(struct net_device *dev, } } -static bool some_qdisc_is_busy(struct net_device *dev, int lock) +static bool some_qdisc_is_busy(struct net_device *dev) { unsigned int i; @@ -667,14 +667,12 @@ static bool some_qdisc_is_busy(struct net_device *dev, int lock) q = dev_queue->qdisc_sleeping; root_lock = qdisc_lock(q); - if (lock) - spin_lock_bh(root_lock); + spin_lock_bh(root_lock); val = (test_bit(__QDISC_STATE_RUNNING, &q->state) || test_bit(__QDISC_STATE_SCHED, &q->state)); - if (lock) - spin_unlock_bh(root_lock); + spin_unlock_bh(root_lock); if (val) return true; @@ -684,8 +682,6 @@ static bool some_qdisc_is_busy(struct net_device *dev, int lock) void dev_deactivate(struct net_device *dev) { - bool running; - netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc); dev_deactivate_queue(dev, &dev->rx_queue, &noop_qdisc); @@ -695,25 +691,8 @@ void dev_deactivate(struct net_device *dev) synchronize_rcu(); /* Wait for outstanding qdisc_run calls. */ - do { - while (some_qdisc_is_busy(dev, 0)) - yield(); - - /* - * Double-check inside queue lock to ensure that all effects - * of the queue run are visible when we return. - */ - running = some_qdisc_is_busy(dev, 1); - - /* - * The running flag should never be set at this point because - * we've already set dev->qdisc to noop_qdisc *inside* the same - * pair of spin locks. That is, if any qdisc_run starts after - * our initial test it should see the noop_qdisc and then - * clear the RUNNING bit before dropping the queue lock. So - * if it is set here then we've found a bug. - */ - } while (WARN_ON_ONCE(running)); + while (some_qdisc_is_busy(dev)) + yield(); } static void dev_init_scheduler_queue(struct net_device *dev, -- cgit v1.2.3 From 3a76e3716b4e571f5d91a20b6afb412560599083 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Sun, 17 Aug 2008 22:02:11 -0700 Subject: pkt_sched: Grab correct lock in notify_and_destroy(). From: Jarek Poplawski When we are destroying non-root qdiscs, we need to lock the root of the qdisc tree not the the qdisc itself. Signed-off-by: David S. Miller --- net/sched/sch_api.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index c25465e5607a..c8dc72e12107 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -638,9 +638,9 @@ static void notify_and_destroy(struct sk_buff *skb, struct nlmsghdr *n, u32 clid qdisc_notify(skb, n, clid, old, new); if (old) { - spin_lock_bh(&old->q.lock); + sch_tree_lock(old); qdisc_destroy(old); - spin_unlock_bh(&old->q.lock); + sch_tree_unlock(old); } } -- cgit v1.2.3 From 1e0d5a5747772182d1bb2525d8153da640fdfb58 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 17 Aug 2008 22:31:26 -0700 Subject: pkt_sched: No longer destroy qdiscs from RCU. We can now kill them synchronously with all of the previous dev_deactivate() cures. This makes netdev destruction and shutdown saner as the qdiscs hold references to the device. Signed-off-by: David S. Miller --- include/net/sch_generic.h | 1 - net/sched/sch_generic.c | 27 +++++++++------------------ 2 files changed, 9 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 757ab087adbf..84d25f2e6188 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -61,7 +61,6 @@ struct Qdisc struct gnet_stats_basic bstats; struct gnet_stats_queue qstats; struct gnet_stats_rate_est rate_est; - struct rcu_head q_rcu; int (*reshape_fail)(struct sk_buff *skb, struct Qdisc *q); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 30b76aec723b..6f96b7bc0809 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -518,14 +518,19 @@ void qdisc_reset(struct Qdisc *qdisc) } EXPORT_SYMBOL(qdisc_reset); -/* this is the rcu callback function to clean up a qdisc when there - * are no further references to it */ +/* Under qdisc_lock(qdisc) and BH! */ -static void __qdisc_destroy(struct rcu_head *head) +void qdisc_destroy(struct Qdisc *qdisc) { - struct Qdisc *qdisc = container_of(head, struct Qdisc, q_rcu); const struct Qdisc_ops *ops = qdisc->ops; + if (qdisc->flags & TCQ_F_BUILTIN || + !atomic_dec_and_test(&qdisc->refcnt)) + return; + + if (qdisc->parent) + list_del(&qdisc->list); + #ifdef CONFIG_NET_SCHED qdisc_put_stab(qdisc->stab); #endif @@ -542,20 +547,6 @@ static void __qdisc_destroy(struct rcu_head *head) kfree((char *) qdisc - qdisc->padded); } - -/* Under qdisc_lock(qdisc) and BH! */ - -void qdisc_destroy(struct Qdisc *qdisc) -{ - if (qdisc->flags & TCQ_F_BUILTIN || - !atomic_dec_and_test(&qdisc->refcnt)) - return; - - if (qdisc->parent) - list_del(&qdisc->list); - - call_rcu(&qdisc->q_rcu, __qdisc_destroy); -} EXPORT_SYMBOL(qdisc_destroy); static bool dev_all_qdisc_sleeping_noop(struct net_device *dev) -- cgit v1.2.3 From 0d40b6e564bad2047b57a9afc48b701ef3243b89 Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Sun, 17 Aug 2008 22:43:56 -0700 Subject: sch_prio: Use return value from inner qdisc requeue Use return value from inner qdisc requeue when value returned isn't NET_XMIT_SUCCESS, instead of always returning NET_XMIT_DROP. Signed-off-by: Jussi Kivilinna Signed-off-by: David S. Miller --- net/sched/sch_prio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index eac197610edf..7cdc3e6b31e6 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -117,7 +117,7 @@ prio_requeue(struct sk_buff *skb, struct Qdisc* sch) } if (net_xmit_drop_count(ret)) sch->qstats.drops++; - return NET_XMIT_DROP; + return ret; } -- cgit v1.2.3 From 4cf7cb280e4fcfcd7ae7429e17d798d3a44087af Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 17 Aug 2008 22:45:17 -0700 Subject: sch_prio: Use NET_XMIT_SUCCESS instead of "0" constant. Signed-off-by: David S. Miller --- net/sched/sch_prio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index 7cdc3e6b31e6..a6697c686c7f 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -113,7 +113,7 @@ prio_requeue(struct sk_buff *skb, struct Qdisc* sch) if ((ret = qdisc->ops->requeue(skb, qdisc)) == NET_XMIT_SUCCESS) { sch->q.qlen++; sch->qstats.requeues++; - return 0; + return NET_XMIT_SUCCESS; } if (net_xmit_drop_count(ret)) sch->qstats.drops++; -- cgit v1.2.3 From 13601cd8e44aab332cedff1d6dc10786ec890b7b Mon Sep 17 00:00:00 2001 From: Yang Hongyang Date: Sun, 17 Aug 2008 23:21:52 -0700 Subject: ipv6: Fix the return interface index when get it while no message is received. When get receiving interface index while no message is received, the bounded device's index of the socket should be returned. RFC 3542: Issuing getsockopt() for the above options will return the sticky option value i.e., the value set with setsockopt(). If no sticky option value has been set getsockopt() will return the following values: - For the IPV6_PKTINFO option, it will return an in6_pktinfo structure with ipi6_addr being in6addr_any and ipi6_ifindex being zero. Signed-off-by: Yang Hongyang Signed-off-by: David S. Miller --- net/ipv6/ipv6_sockglue.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 741cfcd96f88..4e5eac301f91 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -911,7 +911,7 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, } else { if (np->rxopt.bits.rxinfo) { struct in6_pktinfo src_info; - src_info.ipi6_ifindex = np->mcast_oif; + src_info.ipi6_ifindex = np->mcast_oif ? np->mcast_oif : sk->sk_bound_dev_if; ipv6_addr_copy(&src_info.ipi6_addr, &np->daddr); put_cmsg(&msg, SOL_IPV6, IPV6_PKTINFO, sizeof(src_info), &src_info); } @@ -921,7 +921,7 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, } if (np->rxopt.bits.rxoinfo) { struct in6_pktinfo src_info; - src_info.ipi6_ifindex = np->mcast_oif; + src_info.ipi6_ifindex = np->mcast_oif ? np->mcast_oif : sk->sk_bound_dev_if; ipv6_addr_copy(&src_info.ipi6_addr, &np->daddr); put_cmsg(&msg, SOL_IPV6, IPV6_2292PKTINFO, sizeof(src_info), &src_info); } -- cgit v1.2.3 From 96d203169d1d851ac1468f7d4459a09581be364c Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 17 Aug 2008 23:37:16 -0700 Subject: pkt_sched: Fix missed RCU unlock in dev_queue_xmit() Noticed by Jarek Poplawski. Signed-off-by: David S. Miller --- net/core/dev.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 819f0175bdc9..8d133802372b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1805,14 +1805,12 @@ gso: spin_lock(root_lock); if (unlikely(test_bit(__QDISC_STATE_DEACTIVATED, &q->state))) { - spin_unlock(root_lock); + kfree_skb(skb); rc = NET_XMIT_DROP; - goto out_kfree_skb; + } else { + rc = qdisc_enqueue_root(skb, q); + qdisc_run(q); } - - rc = qdisc_enqueue_root(skb, q); - qdisc_run(q); - spin_unlock(root_lock); goto out; -- cgit v1.2.3 From 69747650c814a8a79fef412c7416adf823293a3e Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 17 Aug 2008 23:55:36 -0700 Subject: pkt_sched: Fix return value corruption in HTB and TBF. Based upon a bug report by Josip Rodin. Packet schedulers should only return NET_XMIT_DROP iff the packet really was dropped. If the packet does reach the device after we return NET_XMIT_DROP then TCP can crash because it depends upon the enqueue path return values being accurate. Signed-off-by: David S. Miller --- net/sched/sch_htb.c | 4 ++-- net/sched/sch_tbf.c | 11 ++--------- 2 files changed, 4 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 6febd245e62b..0df0df202ed0 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -577,7 +577,7 @@ static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch) sch->qstats.drops++; cl->qstats.drops++; } - return NET_XMIT_DROP; + return ret; } else { cl->bstats.packets += skb_is_gso(skb)?skb_shinfo(skb)->gso_segs:1; @@ -623,7 +623,7 @@ static int htb_requeue(struct sk_buff *skb, struct Qdisc *sch) sch->qstats.drops++; cl->qstats.drops++; } - return NET_XMIT_DROP; + return ret; } else htb_activate(q, cl); diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index 7d3b7ff3bf07..94c61598b86a 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -123,15 +123,8 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc* sch) struct tbf_sched_data *q = qdisc_priv(sch); int ret; - if (qdisc_pkt_len(skb) > q->max_size) { - sch->qstats.drops++; -#ifdef CONFIG_NET_CLS_ACT - if (sch->reshape_fail == NULL || sch->reshape_fail(skb, sch)) -#endif - kfree_skb(skb); - - return NET_XMIT_DROP; - } + if (qdisc_pkt_len(skb) > q->max_size) + return qdisc_reshape_fail(skb, sch); ret = qdisc_enqueue(skb, q->qdisc); if (ret != 0) { -- cgit v1.2.3 From 90855d7b725d764d6d70503bcc1b494cf10ddc98 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Mon, 18 Aug 2008 13:23:53 +0200 Subject: [Bluetooth] Fix userspace breakage due missing class links The Bluetooth adapters and connections are best presented via a class in sysfs. The removal of the links inside the Bluetooth class broke assumptions by userspace programs on how to find attached adapters. This patch creates adapters and connections as part of the Bluetooth class, but it uses different device types to distinguish them. The userspace programs can now easily navigate in the sysfs device tree. The unused platform device and bus have been removed to keep the code simple and clean. Signed-off-by: Marcel Holtmann --- net/bluetooth/hci_sysfs.c | 376 +++++++++++++++++++++++----------------------- 1 file changed, 189 insertions(+), 187 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index c85bf8f678dc..f4f6615cad9f 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -3,8 +3,6 @@ #include #include -#include - #include #include @@ -12,10 +10,164 @@ #undef BT_DBG #define BT_DBG(D...) #endif + +struct class *bt_class = NULL; +EXPORT_SYMBOL_GPL(bt_class); + static struct workqueue_struct *btaddconn; static struct workqueue_struct *btdelconn; -static inline char *typetostr(int type) +static inline char *link_typetostr(int type) +{ + switch (type) { + case ACL_LINK: + return "ACL"; + case SCO_LINK: + return "SCO"; + case ESCO_LINK: + return "eSCO"; + default: + return "UNKNOWN"; + } +} + +static ssize_t show_link_type(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct hci_conn *conn = dev_get_drvdata(dev); + return sprintf(buf, "%s\n", link_typetostr(conn->type)); +} + +static ssize_t show_link_address(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct hci_conn *conn = dev_get_drvdata(dev); + bdaddr_t bdaddr; + baswap(&bdaddr, &conn->dst); + return sprintf(buf, "%s\n", batostr(&bdaddr)); +} + +static ssize_t show_link_features(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct hci_conn *conn = dev_get_drvdata(dev); + + return sprintf(buf, "0x%02x%02x%02x%02x%02x%02x%02x%02x\n", + conn->features[0], conn->features[1], + conn->features[2], conn->features[3], + conn->features[4], conn->features[5], + conn->features[6], conn->features[7]); +} + +#define LINK_ATTR(_name,_mode,_show,_store) \ +struct device_attribute link_attr_##_name = __ATTR(_name,_mode,_show,_store) + +static LINK_ATTR(type, S_IRUGO, show_link_type, NULL); +static LINK_ATTR(address, S_IRUGO, show_link_address, NULL); +static LINK_ATTR(features, S_IRUGO, show_link_features, NULL); + +static struct attribute *bt_link_attrs[] = { + &link_attr_type.attr, + &link_attr_address.attr, + &link_attr_features.attr, + NULL +}; + +static struct attribute_group bt_link_group = { + .attrs = bt_link_attrs, +}; + +static struct attribute_group *bt_link_groups[] = { + &bt_link_group, + NULL +}; + +static void bt_link_release(struct device *dev) +{ + void *data = dev_get_drvdata(dev); + kfree(data); +} + +static struct device_type bt_link = { + .name = "link", + .groups = bt_link_groups, + .release = bt_link_release, +}; + +static void add_conn(struct work_struct *work) +{ + struct hci_conn *conn = container_of(work, struct hci_conn, work); + + flush_workqueue(btdelconn); + + if (device_add(&conn->dev) < 0) { + BT_ERR("Failed to register connection device"); + return; + } +} + +void hci_conn_add_sysfs(struct hci_conn *conn) +{ + struct hci_dev *hdev = conn->hdev; + + BT_DBG("conn %p", conn); + + conn->dev.type = &bt_link; + conn->dev.class = bt_class; + conn->dev.parent = &hdev->dev; + + snprintf(conn->dev.bus_id, BUS_ID_SIZE, "%s:%d", + hdev->name, conn->handle); + + dev_set_drvdata(&conn->dev, conn); + + device_initialize(&conn->dev); + + INIT_WORK(&conn->work, add_conn); + + queue_work(btaddconn, &conn->work); +} + +/* + * The rfcomm tty device will possibly retain even when conn + * is down, and sysfs doesn't support move zombie device, + * so we should move the device before conn device is destroyed. + */ +static int __match_tty(struct device *dev, void *data) +{ + return !strncmp(dev->bus_id, "rfcomm", 6); +} + +static void del_conn(struct work_struct *work) +{ + struct hci_conn *conn = container_of(work, struct hci_conn, work); + struct hci_dev *hdev = conn->hdev; + + while (1) { + struct device *dev; + + dev = device_find_child(&conn->dev, NULL, __match_tty); + if (!dev) + break; + device_move(dev, NULL); + put_device(dev); + } + + device_del(&conn->dev); + put_device(&conn->dev); + hci_dev_put(hdev); +} + +void hci_conn_del_sysfs(struct hci_conn *conn) +{ + BT_DBG("conn %p", conn); + + if (!device_is_registered(&conn->dev)) + return; + + INIT_WORK(&conn->work, del_conn); + + queue_work(btdelconn, &conn->work); +} + +static inline char *host_typetostr(int type) { switch (type) { case HCI_VIRTUAL: @@ -40,7 +192,7 @@ static inline char *typetostr(int type) static ssize_t show_type(struct device *dev, struct device_attribute *attr, char *buf) { struct hci_dev *hdev = dev_get_drvdata(dev); - return sprintf(buf, "%s\n", typetostr(hdev->type)); + return sprintf(buf, "%s\n", host_typetostr(hdev->type)); } static ssize_t show_name(struct device *dev, struct device_attribute *attr, char *buf) @@ -221,183 +373,62 @@ static DEVICE_ATTR(sniff_max_interval, S_IRUGO | S_IWUSR, static DEVICE_ATTR(sniff_min_interval, S_IRUGO | S_IWUSR, show_sniff_min_interval, store_sniff_min_interval); -static struct device_attribute *bt_attrs[] = { - &dev_attr_type, - &dev_attr_name, - &dev_attr_class, - &dev_attr_address, - &dev_attr_features, - &dev_attr_manufacturer, - &dev_attr_hci_version, - &dev_attr_hci_revision, - &dev_attr_inquiry_cache, - &dev_attr_idle_timeout, - &dev_attr_sniff_max_interval, - &dev_attr_sniff_min_interval, +static struct attribute *bt_host_attrs[] = { + &dev_attr_type.attr, + &dev_attr_name.attr, + &dev_attr_class.attr, + &dev_attr_address.attr, + &dev_attr_features.attr, + &dev_attr_manufacturer.attr, + &dev_attr_hci_version.attr, + &dev_attr_hci_revision.attr, + &dev_attr_inquiry_cache.attr, + &dev_attr_idle_timeout.attr, + &dev_attr_sniff_max_interval.attr, + &dev_attr_sniff_min_interval.attr, NULL }; -static ssize_t show_conn_type(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct hci_conn *conn = dev_get_drvdata(dev); - return sprintf(buf, "%s\n", conn->type == ACL_LINK ? "ACL" : "SCO"); -} - -static ssize_t show_conn_address(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct hci_conn *conn = dev_get_drvdata(dev); - bdaddr_t bdaddr; - baswap(&bdaddr, &conn->dst); - return sprintf(buf, "%s\n", batostr(&bdaddr)); -} - -static ssize_t show_conn_features(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct hci_conn *conn = dev_get_drvdata(dev); - - return sprintf(buf, "0x%02x%02x%02x%02x%02x%02x%02x%02x\n", - conn->features[0], conn->features[1], - conn->features[2], conn->features[3], - conn->features[4], conn->features[5], - conn->features[6], conn->features[7]); -} - -#define CONN_ATTR(_name,_mode,_show,_store) \ -struct device_attribute conn_attr_##_name = __ATTR(_name,_mode,_show,_store) - -static CONN_ATTR(type, S_IRUGO, show_conn_type, NULL); -static CONN_ATTR(address, S_IRUGO, show_conn_address, NULL); -static CONN_ATTR(features, S_IRUGO, show_conn_features, NULL); - -static struct device_attribute *conn_attrs[] = { - &conn_attr_type, - &conn_attr_address, - &conn_attr_features, - NULL +static struct attribute_group bt_host_group = { + .attrs = bt_host_attrs, }; -struct class *bt_class = NULL; -EXPORT_SYMBOL_GPL(bt_class); - -static struct bus_type bt_bus = { - .name = "bluetooth", +static struct attribute_group *bt_host_groups[] = { + &bt_host_group, + NULL }; -static struct platform_device *bt_platform; - -static void bt_release(struct device *dev) +static void bt_host_release(struct device *dev) { void *data = dev_get_drvdata(dev); kfree(data); } -static void add_conn(struct work_struct *work) -{ - struct hci_conn *conn = container_of(work, struct hci_conn, work); - int i; - - flush_workqueue(btdelconn); - - if (device_add(&conn->dev) < 0) { - BT_ERR("Failed to register connection device"); - return; - } - - for (i = 0; conn_attrs[i]; i++) - if (device_create_file(&conn->dev, conn_attrs[i]) < 0) - BT_ERR("Failed to create connection attribute"); -} - -void hci_conn_add_sysfs(struct hci_conn *conn) -{ - struct hci_dev *hdev = conn->hdev; - - BT_DBG("conn %p", conn); - - conn->dev.bus = &bt_bus; - conn->dev.parent = &hdev->dev; - - conn->dev.release = bt_release; - - snprintf(conn->dev.bus_id, BUS_ID_SIZE, "%s:%d", - hdev->name, conn->handle); - - dev_set_drvdata(&conn->dev, conn); - - device_initialize(&conn->dev); - - INIT_WORK(&conn->work, add_conn); - - queue_work(btaddconn, &conn->work); -} - -/* - * The rfcomm tty device will possibly retain even when conn - * is down, and sysfs doesn't support move zombie device, - * so we should move the device before conn device is destroyed. - */ -static int __match_tty(struct device *dev, void *data) -{ - return !strncmp(dev->bus_id, "rfcomm", 6); -} - -static void del_conn(struct work_struct *work) -{ - struct hci_conn *conn = container_of(work, struct hci_conn, work); - struct hci_dev *hdev = conn->hdev; - - while (1) { - struct device *dev; - - dev = device_find_child(&conn->dev, NULL, __match_tty); - if (!dev) - break; - device_move(dev, NULL); - put_device(dev); - } - - device_del(&conn->dev); - put_device(&conn->dev); - hci_dev_put(hdev); -} - -void hci_conn_del_sysfs(struct hci_conn *conn) -{ - BT_DBG("conn %p", conn); - - if (!device_is_registered(&conn->dev)) - return; - - INIT_WORK(&conn->work, del_conn); - - queue_work(btdelconn, &conn->work); -} +static struct device_type bt_host = { + .name = "host", + .groups = bt_host_groups, + .release = bt_host_release, +}; int hci_register_sysfs(struct hci_dev *hdev) { struct device *dev = &hdev->dev; - unsigned int i; int err; BT_DBG("%p name %s type %d", hdev, hdev->name, hdev->type); - dev->bus = &bt_bus; + dev->type = &bt_host; + dev->class = bt_class; dev->parent = hdev->parent; strlcpy(dev->bus_id, hdev->name, BUS_ID_SIZE); - dev->release = bt_release; - dev_set_drvdata(dev, hdev); err = device_register(dev); if (err < 0) return err; - for (i = 0; bt_attrs[i]; i++) - if (device_create_file(dev, bt_attrs[i]) < 0) - BT_ERR("Failed to create device attribute"); - return 0; } @@ -410,59 +441,30 @@ void hci_unregister_sysfs(struct hci_dev *hdev) int __init bt_sysfs_init(void) { - int err; - btaddconn = create_singlethread_workqueue("btaddconn"); - if (!btaddconn) { - err = -ENOMEM; - goto out; - } + if (!btaddconn) + return -ENOMEM; btdelconn = create_singlethread_workqueue("btdelconn"); if (!btdelconn) { - err = -ENOMEM; - goto out_del; - } - - bt_platform = platform_device_register_simple("bluetooth", -1, NULL, 0); - if (IS_ERR(bt_platform)) { - err = PTR_ERR(bt_platform); - goto out_platform; + destroy_workqueue(btaddconn); + return -ENOMEM; } - err = bus_register(&bt_bus); - if (err < 0) - goto out_bus; - bt_class = class_create(THIS_MODULE, "bluetooth"); if (IS_ERR(bt_class)) { - err = PTR_ERR(bt_class); - goto out_class; + destroy_workqueue(btdelconn); + destroy_workqueue(btaddconn); + return PTR_ERR(bt_class); } return 0; - -out_class: - bus_unregister(&bt_bus); -out_bus: - platform_device_unregister(bt_platform); -out_platform: - destroy_workqueue(btdelconn); -out_del: - destroy_workqueue(btaddconn); -out: - return err; } void bt_sysfs_cleanup(void) { destroy_workqueue(btaddconn); - destroy_workqueue(btdelconn); class_destroy(bt_class); - - bus_unregister(&bt_bus); - - platform_device_unregister(bt_platform); } -- cgit v1.2.3 From 63fbd24e5102eecfc9d049ed7f4be7f9a25f814f Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Mon, 18 Aug 2008 13:23:53 +0200 Subject: [Bluetooth] Consolidate maintainers information The Bluetooth entries for the MAINTAINERS file are a little bit too much. Consolidate them into two entries. One for Bluetooth drivers and another one for the Bluetooth subsystem. Also the MODULE_AUTHOR should indicate the current maintainer of the module and actually not the original author. Fix all Bluetooth modules to provide current maintainer information. Signed-off-by: Marcel Holtmann --- MAINTAINERS | 87 +++---------------------------------------- drivers/bluetooth/bt3c_cs.c | 2 +- drivers/bluetooth/hci_ldisc.c | 2 +- drivers/bluetooth/hci_usb.c | 2 +- drivers/bluetooth/hci_vhci.c | 2 +- net/bluetooth/af_bluetooth.c | 2 +- net/bluetooth/bnep/core.c | 2 +- net/bluetooth/l2cap.c | 2 +- net/bluetooth/rfcomm/core.c | 2 +- net/bluetooth/sco.c | 2 +- 10 files changed, 15 insertions(+), 90 deletions(-) (limited to 'net') diff --git a/MAINTAINERS b/MAINTAINERS index af6aa4e4b392..2676f9a1feee 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -936,94 +936,19 @@ M: joern@lazybastard.org L: linux-mtd@lists.infradead.org S: Maintained -BLUETOOTH SUBSYSTEM +BLUETOOTH DRIVERS P: Marcel Holtmann M: marcel@holtmann.org -P: Maxim Krasnyansky -M: maxk@qualcomm.com L: linux-bluetooth@vger.kernel.org -W: http://bluez.sf.net -W: http://www.bluez.org -W: http://www.holtmann.org/linux/bluetooth/ -T: git kernel.org:/pub/scm/linux/kernel/git/holtmann/bluetooth-2.6.git -S: Maintained - -BLUETOOTH RFCOMM LAYER -P: Marcel Holtmann -M: marcel@holtmann.org -P: Maxim Krasnyansky -M: maxk@qualcomm.com -S: Maintained - -BLUETOOTH BNEP LAYER -P: Marcel Holtmann -M: marcel@holtmann.org -P: Maxim Krasnyansky -M: maxk@qualcomm.com -S: Maintained - -BLUETOOTH CMTP LAYER -P: Marcel Holtmann -M: marcel@holtmann.org +W: http://www.bluez.org/ S: Maintained -BLUETOOTH HIDP LAYER -P: Marcel Holtmann -M: marcel@holtmann.org -S: Maintained - -BLUETOOTH HCI UART DRIVER -P: Marcel Holtmann -M: marcel@holtmann.org -P: Maxim Krasnyansky -M: maxk@qualcomm.com -S: Maintained - -BLUETOOTH HCI USB DRIVER -P: Marcel Holtmann -M: marcel@holtmann.org -P: Maxim Krasnyansky -M: maxk@qualcomm.com -S: Maintained - -BLUETOOTH HCI BCM203X DRIVER -P: Marcel Holtmann -M: marcel@holtmann.org -S: Maintained - -BLUETOOTH HCI BPA10X DRIVER -P: Marcel Holtmann -M: marcel@holtmann.org -S: Maintained - -BLUETOOTH HCI BFUSB DRIVER -P: Marcel Holtmann -M: marcel@holtmann.org -S: Maintained - -BLUETOOTH HCI DTL1 DRIVER -P: Marcel Holtmann -M: marcel@holtmann.org -S: Maintained - -BLUETOOTH HCI BLUECARD DRIVER -P: Marcel Holtmann -M: marcel@holtmann.org -S: Maintained - -BLUETOOTH HCI BT3C DRIVER -P: Marcel Holtmann -M: marcel@holtmann.org -S: Maintained - -BLUETOOTH HCI BTUART DRIVER +BLUETOOTH SUBSYSTEM P: Marcel Holtmann M: marcel@holtmann.org -S: Maintained - -BLUETOOTH HCI VHCI DRIVER -P: Maxim Krasnyansky -M: maxk@qualcomm.com +L: linux-bluetooth@vger.kernel.org +W: http://www.bluez.org/ +T: git kernel.org:/pub/scm/linux/kernel/git/holtmann/bluetooth-2.6.git S: Maintained BONDING DRIVER diff --git a/drivers/bluetooth/bt3c_cs.c b/drivers/bluetooth/bt3c_cs.c index 593b7c595038..27058477cc8b 100644 --- a/drivers/bluetooth/bt3c_cs.c +++ b/drivers/bluetooth/bt3c_cs.c @@ -60,7 +60,7 @@ /* ======================== Module parameters ======================== */ -MODULE_AUTHOR("Marcel Holtmann , Jose Orlando Pereira "); +MODULE_AUTHOR("Marcel Holtmann "); MODULE_DESCRIPTION("Bluetooth driver for the 3Com Bluetooth PCMCIA card"); MODULE_LICENSE("GPL"); MODULE_FIRMWARE("BT3CPCC.bin"); diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c index 69df187d74ce..8dfcf77cb717 100644 --- a/drivers/bluetooth/hci_ldisc.c +++ b/drivers/bluetooth/hci_ldisc.c @@ -577,7 +577,7 @@ module_exit(hci_uart_exit); module_param(reset, bool, 0644); MODULE_PARM_DESC(reset, "Send HCI reset command on initialization"); -MODULE_AUTHOR("Maxim Krasnyansky , Marcel Holtmann "); +MODULE_AUTHOR("Marcel Holtmann "); MODULE_DESCRIPTION("Bluetooth HCI UART driver ver " VERSION); MODULE_VERSION(VERSION); MODULE_LICENSE("GPL"); diff --git a/drivers/bluetooth/hci_usb.c b/drivers/bluetooth/hci_usb.c index e397572bf574..3c453924f838 100644 --- a/drivers/bluetooth/hci_usb.c +++ b/drivers/bluetooth/hci_usb.c @@ -1130,7 +1130,7 @@ module_param(isoc, int, 0644); MODULE_PARM_DESC(isoc, "Set isochronous transfers for SCO over HCI support"); #endif -MODULE_AUTHOR("Maxim Krasnyansky , Marcel Holtmann "); +MODULE_AUTHOR("Marcel Holtmann "); MODULE_DESCRIPTION("Bluetooth HCI USB driver ver " VERSION); MODULE_VERSION(VERSION); MODULE_LICENSE("GPL"); diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c index d97700aa54a9..7320a71b6368 100644 --- a/drivers/bluetooth/hci_vhci.c +++ b/drivers/bluetooth/hci_vhci.c @@ -377,7 +377,7 @@ module_exit(vhci_exit); module_param(minor, int, 0444); MODULE_PARM_DESC(minor, "Miscellaneous minor device number"); -MODULE_AUTHOR("Maxim Krasnyansky , Marcel Holtmann "); +MODULE_AUTHOR("Marcel Holtmann "); MODULE_DESCRIPTION("Bluetooth virtual HCI driver ver " VERSION); MODULE_VERSION(VERSION); MODULE_LICENSE("GPL"); diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 4e59df5f8e05..1edfdf4c095b 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -456,7 +456,7 @@ static void __exit bt_exit(void) subsys_initcall(bt_init); module_exit(bt_exit); -MODULE_AUTHOR("Maxim Krasnyansky , Marcel Holtmann "); +MODULE_AUTHOR("Marcel Holtmann "); MODULE_DESCRIPTION("Bluetooth Core ver " VERSION); MODULE_VERSION(VERSION); MODULE_LICENSE("GPL"); diff --git a/net/bluetooth/bnep/core.c b/net/bluetooth/bnep/core.c index 12bba6207a8d..80ba30cf4b68 100644 --- a/net/bluetooth/bnep/core.c +++ b/net/bluetooth/bnep/core.c @@ -736,7 +736,7 @@ MODULE_PARM_DESC(compress_src, "Compress sources headers"); module_param(compress_dst, bool, 0644); MODULE_PARM_DESC(compress_dst, "Compress destination headers"); -MODULE_AUTHOR("David Libault , Maxim Krasnyansky "); +MODULE_AUTHOR("Marcel Holtmann "); MODULE_DESCRIPTION("Bluetooth BNEP ver " VERSION); MODULE_VERSION(VERSION); MODULE_LICENSE("GPL"); diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index c1239852834a..3396d5bdef1c 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -2516,7 +2516,7 @@ EXPORT_SYMBOL(l2cap_load); module_init(l2cap_init); module_exit(l2cap_exit); -MODULE_AUTHOR("Maxim Krasnyansky , Marcel Holtmann "); +MODULE_AUTHOR("Marcel Holtmann "); MODULE_DESCRIPTION("Bluetooth L2CAP ver " VERSION); MODULE_VERSION(VERSION); MODULE_LICENSE("GPL"); diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index 6cfc7ba611b3..ba537fae0a4c 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -2115,7 +2115,7 @@ MODULE_PARM_DESC(channel_mtu, "Default MTU for the RFCOMM channel"); module_param(l2cap_mtu, uint, 0644); MODULE_PARM_DESC(l2cap_mtu, "Default MTU for the L2CAP connection"); -MODULE_AUTHOR("Maxim Krasnyansky , Marcel Holtmann "); +MODULE_AUTHOR("Marcel Holtmann "); MODULE_DESCRIPTION("Bluetooth RFCOMM ver " VERSION); MODULE_VERSION(VERSION); MODULE_LICENSE("GPL"); diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index 8cda49874868..a16011fedc1d 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -1002,7 +1002,7 @@ module_exit(sco_exit); module_param(disable_esco, bool, 0644); MODULE_PARM_DESC(disable_esco, "Disable eSCO connection creation"); -MODULE_AUTHOR("Maxim Krasnyansky , Marcel Holtmann "); +MODULE_AUTHOR("Marcel Holtmann "); MODULE_DESCRIPTION("Bluetooth SCO ver " VERSION); MODULE_VERSION(VERSION); MODULE_LICENSE("GPL"); -- cgit v1.2.3 From e10e0dfe3ba358cfb442cc3bf0d3f2068785bf5c Mon Sep 17 00:00:00 2001 From: Henrique de Moraes Holschuh Date: Sat, 2 Aug 2008 14:56:25 -0300 Subject: rfkill: protect suspended rfkill controllers Guard rfkill controllers attached to a rfkill class against state changes after class suspend has been issued. Signed-off-by: Henrique de Moraes Holschuh Acked-by: Ivo van Doorn Signed-off-by: John W. Linville --- Documentation/rfkill.txt | 5 +++++ net/rfkill/rfkill.c | 14 ++++++++++---- 2 files changed, 15 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/Documentation/rfkill.txt b/Documentation/rfkill.txt index 28b6ec87c642..6fcb3060dec5 100644 --- a/Documentation/rfkill.txt +++ b/Documentation/rfkill.txt @@ -363,6 +363,11 @@ This rule exists because users of the rfkill subsystem expect to get (and set, when possible) the overall transmitter rfkill state, not of a particular rfkill line. +5. During suspend, the rfkill class will attempt to soft-block the radio +through a call to rfkill->toggle_radio, and will try to restore its previous +state during resume. After a rfkill class is suspended, it will *not* call +rfkill->toggle_radio until it is resumed. + Example of a WLAN wireless driver connected to the rfkill subsystem: -------------------------------------------------------------------- diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c index d2d45655cd1a..35a9994e2339 100644 --- a/net/rfkill/rfkill.c +++ b/net/rfkill/rfkill.c @@ -150,6 +150,8 @@ static void update_rfkill_state(struct rfkill *rfkill) * calls and handling all the red tape such as issuing notifications * if the call is successful. * + * Suspended devices are not touched at all, and -EAGAIN is returned. + * * Note that the @force parameter cannot override a (possibly cached) * state of RFKILL_STATE_HARD_BLOCKED. Any device making use of * RFKILL_STATE_HARD_BLOCKED implements either get_state() or @@ -168,6 +170,9 @@ static int rfkill_toggle_radio(struct rfkill *rfkill, int retval = 0; enum rfkill_state oldstate, newstate; + if (unlikely(rfkill->dev.power.power_state.event & PM_EVENT_SLEEP)) + return -EBUSY; + oldstate = rfkill->state; if (rfkill->get_state && !force && @@ -214,7 +219,7 @@ static int rfkill_toggle_radio(struct rfkill *rfkill, * * This function toggles the state of all switches of given type, * unless a specific switch is claimed by userspace (in which case, - * that switch is left alone). + * that switch is left alone) or suspended. */ void rfkill_switch_all(enum rfkill_type type, enum rfkill_state state) { @@ -239,8 +244,8 @@ EXPORT_SYMBOL(rfkill_switch_all); /** * rfkill_epo - emergency power off all transmitters * - * This kicks all rfkill devices to RFKILL_STATE_SOFT_BLOCKED, ignoring - * everything in its path but rfkill_mutex and rfkill->mutex. + * This kicks all non-suspended rfkill devices to RFKILL_STATE_SOFT_BLOCKED, + * ignoring everything in its path but rfkill_mutex and rfkill->mutex. */ void rfkill_epo(void) { @@ -458,13 +463,14 @@ static int rfkill_resume(struct device *dev) if (dev->power.power_state.event != PM_EVENT_ON) { mutex_lock(&rfkill->mutex); + dev->power.power_state.event = PM_EVENT_ON; + /* restore radio state AND notify everybody */ rfkill_toggle_radio(rfkill, rfkill->state, 1); mutex_unlock(&rfkill->mutex); } - dev->power.power_state = PMSG_ON; return 0; } #else -- cgit v1.2.3 From a61dae1f784f0c4ced0d47721c0efc7033231522 Mon Sep 17 00:00:00 2001 From: Ron Rindjunsky Date: Sun, 10 Aug 2008 00:54:34 +0300 Subject: mac80211: update new sta's rx timestamp This patch fixes needless probe request caused by zero value in sta->last_rx inside ieee80211_associated flow Signed-off-by: Ron Rindjunsky Signed-off-by: Tomas Winkler Signed-off-by: John W. Linville --- net/mac80211/mlme.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index e1d11c9b6729..1e97fb9fb34b 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -2103,6 +2103,8 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, rcu_read_unlock(); return; } + /* update new sta with its last rx activity */ + sta->last_rx = jiffies; } /* -- cgit v1.2.3 From 8608db031b4d2932d645709e2cfe8fbcd91a7305 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 18 Aug 2008 20:51:18 -0700 Subject: pkt_sched: Never schedule non-root qdiscs. Based upon initial discovery and patch by Jarek Poplawski. The qdisc watchdogs can be attached to any qdisc, not just the root, so make sure we schedule the correct one. CBQ has a similar bug. Signed-off-by: David S. Miller --- net/sched/sch_api.c | 2 +- net/sched/sch_cbq.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index c8dc72e12107..98c00847a3d2 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -426,7 +426,7 @@ static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer) wd->qdisc->flags &= ~TCQ_F_THROTTLED; smp_wmb(); - __netif_schedule(wd->qdisc); + __netif_schedule(qdisc_root(wd->qdisc)); return HRTIMER_NORESTART; } diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 4e261ce62f48..47ef492c4ff4 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -654,7 +654,7 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer) } sch->flags &= ~TCQ_F_THROTTLED; - __netif_schedule(sch); + __netif_schedule(qdisc_root(sch)); return HRTIMER_NORESTART; } -- cgit v1.2.3 From 25bfcd5a78a377ea4c54a3c21e44590e2fc478a6 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Mon, 18 Aug 2008 20:53:34 -0700 Subject: pkt_sched: Add lockdep annotation for qdisc locks Qdisc locks are initialized in the same function, qdisc_alloc(), so lockdep can't distinguish tx qdisc lock from rx and reports "possible recursive locking detected" when both these locks are taken eg. while using act_mirred with ifb. This looks like a false positive. Anyway, after this patch these locks will be reported more exactly. Reported-by: Denys Fedoryshchenko Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/sched/sch_api.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 98c00847a3d2..7d7070b1eebd 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include @@ -707,6 +708,10 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, return err; } +/* lockdep annotation is needed for ingress; egress gets it only for name */ +static struct lock_class_key qdisc_tx_lock; +static struct lock_class_key qdisc_rx_lock; + /* Allocate and initialize new qdisc. @@ -767,6 +772,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, if (handle == TC_H_INGRESS) { sch->flags |= TCQ_F_INGRESS; handle = TC_H_MAKE(TC_H_INGRESS, 0); + lockdep_set_class(qdisc_lock(sch), &qdisc_rx_lock); } else { if (handle == 0) { handle = qdisc_alloc_handle(dev); @@ -774,6 +780,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, if (handle == 0) goto err_out3; } + lockdep_set_class(qdisc_lock(sch), &qdisc_tx_lock); } sch->handle = handle; -- cgit v1.2.3 From 4d8863a29c4755a0461cd31b6865026187d6c43a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 18 Aug 2008 21:03:15 -0700 Subject: pkt_sched: Don't hold qdisc lock over qdisc_destroy(). Based upon reports by Denys Fedoryshchenko, and feedback and help from Jarek Poplawski and Herbert Xu. We always either: 1) Never made an external reference to this qdisc. or 2) Did a dev_deactivate() which purged all asynchronous references. So do not lock the qdisc when we call qdisc_destroy(), it's illegal anyways as when we drop the lock this is free'd memory. Signed-off-by: David S. Miller --- net/sched/sch_api.c | 13 ++----------- net/sched/sch_generic.c | 6 ------ 2 files changed, 2 insertions(+), 17 deletions(-) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 7d7070b1eebd..d91a2338877c 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -638,11 +638,8 @@ static void notify_and_destroy(struct sk_buff *skb, struct nlmsghdr *n, u32 clid if (new || old) qdisc_notify(skb, n, clid, old, new); - if (old) { - sch_tree_lock(old); + if (old) qdisc_destroy(old); - sch_tree_unlock(old); - } } /* Graft qdisc "new" to class "classid" of qdisc "parent" or @@ -1092,16 +1089,10 @@ create_n_graft: graft: if (1) { - spinlock_t *root_lock; - err = qdisc_graft(dev, p, skb, n, clid, q, NULL); if (err) { - if (q) { - root_lock = qdisc_root_lock(q); - spin_lock_bh(root_lock); + if (q) qdisc_destroy(q); - spin_unlock_bh(root_lock); - } return err; } } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 6f96b7bc0809..c3ed4d44fc14 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -518,8 +518,6 @@ void qdisc_reset(struct Qdisc *qdisc) } EXPORT_SYMBOL(qdisc_reset); -/* Under qdisc_lock(qdisc) and BH! */ - void qdisc_destroy(struct Qdisc *qdisc) { const struct Qdisc_ops *ops = qdisc->ops; @@ -712,14 +710,10 @@ static void shutdown_scheduler_queue(struct net_device *dev, struct Qdisc *qdisc_default = _qdisc_default; if (qdisc) { - spinlock_t *root_lock = qdisc_lock(qdisc); - dev_queue->qdisc = qdisc_default; dev_queue->qdisc_sleeping = qdisc_default; - spin_lock_bh(root_lock); qdisc_destroy(qdisc); - spin_unlock_bh(root_lock); } } -- cgit v1.2.3 From d28934ad8a4e87203a95de9c376611de8bc2f013 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 18 Aug 2008 21:14:20 -0700 Subject: dccp: Fix panic caused by too early termination of retransmission mechanism Thanks is due to Wei Yongjun for the detailed analysis and description of this bug at http://marc.info/?l=dccp&m=121739364909199&w=2 The problem is that invalid packets received by a client in state REQUEST cause the retransmission timer for the DCCP-Request to be reset. This includes freeing the Request-skb ( in dccp_rcv_request_sent_state_process() ). As a consequence, * the arrival of further packets cause a double-free, triggering a panic(), * the connection then may hang, since further retransmissions are blocked. This patch changes the order of statements so that the retransmission timer is reset, and the pending Request freed, only if a valid Response has arrived (or the number of sysctl-retries has been exhausted). Further changes: ---------------- To be on the safe side, replaced __kfree_skb with kfree_skb so that if due to unexpected circumstances the sk_send_head is NULL the WARN_ON is used instead. Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- net/dccp/input.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/dccp/input.c b/net/dccp/input.c index df2f110df94a..803933ab396d 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -411,12 +411,6 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, struct dccp_sock *dp = dccp_sk(sk); long tstamp = dccp_timestamp(); - /* Stop the REQUEST timer */ - inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS); - WARN_ON(sk->sk_send_head == NULL); - __kfree_skb(sk->sk_send_head); - sk->sk_send_head = NULL; - if (!between48(DCCP_SKB_CB(skb)->dccpd_ack_seq, dp->dccps_awl, dp->dccps_awh)) { dccp_pr_debug("invalid ackno: S.AWL=%llu, " @@ -441,6 +435,12 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, DCCP_ACKVEC_STATE_RECEIVED)) goto out_invalid_packet; /* FIXME: change error code */ + /* Stop the REQUEST timer */ + inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS); + WARN_ON(sk->sk_send_head == NULL); + kfree_skb(sk->sk_send_head); + sk->sk_send_head = NULL; + dp->dccps_isr = DCCP_SKB_CB(skb)->dccpd_seq; dccp_update_gsr(sk, dp->dccps_isr); /* -- cgit v1.2.3 From 46faec9858e8943226464dac50e205bf210d9174 Mon Sep 17 00:00:00 2001 From: Anders Grafström Date: Mon, 18 Aug 2008 21:29:57 -0700 Subject: netfilter: ipt_addrtype: Fix matching of inverted destination address type MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch fixes matching of inverted destination address type. Signed-off-by: Anders Grafström Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ipt_addrtype.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/netfilter/ipt_addrtype.c b/net/ipv4/netfilter/ipt_addrtype.c index 49587a497229..462a22c97877 100644 --- a/net/ipv4/netfilter/ipt_addrtype.c +++ b/net/ipv4/netfilter/ipt_addrtype.c @@ -70,7 +70,7 @@ addrtype_mt_v1(const struct sk_buff *skb, const struct net_device *in, (info->flags & IPT_ADDRTYPE_INVERT_SOURCE); if (ret && info->dest) ret &= match_type(dev, iph->daddr, info->dest) ^ - (info->flags & IPT_ADDRTYPE_INVERT_DEST); + !!(info->flags & IPT_ADDRTYPE_INVERT_DEST); return ret; } -- cgit v1.2.3 From 1575e7ea018fec992b94a12a1a491ce693ae9eac Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 18 Aug 2008 21:30:55 -0700 Subject: netfilter: ctnetlink: fix double helper assignation for NAT'ed conntracks If we create a conntrack that has NAT handlings and a helper, the helper is assigned twice. This happens because nf_nat_setup_info() - via nf_conntrack_alter_reply() - sets the helper before ctnetlink, which indeed does not check if the conntrack already has a helper as it thinks that it is a brand new conntrack. The fix moves the helper assignation before the set of the status flags. This avoids a bogus assertion in __nf_ct_ext_add (if netfilter assertions are enabled) which checks that the conntrack must not be confirmed. This problem was introduced in 2.6.23 with the netfilter extension infrastructure. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_netlink.c | 34 +++++++++++++++++++--------------- 1 file changed, 19 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 105a616c5c78..d1fb2f8555e8 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1136,16 +1136,33 @@ ctnetlink_create_conntrack(struct nlattr *cda[], ct->timeout.expires = jiffies + ct->timeout.expires * HZ; ct->status |= IPS_CONFIRMED; + rcu_read_lock(); + helper = __nf_ct_helper_find(rtuple); + if (helper) { + help = nf_ct_helper_ext_add(ct, GFP_KERNEL); + if (help == NULL) { + rcu_read_unlock(); + err = -ENOMEM; + goto err; + } + /* not in hash table yet so not strictly necessary */ + rcu_assign_pointer(help->helper, helper); + } + if (cda[CTA_STATUS]) { err = ctnetlink_change_status(ct, cda); - if (err < 0) + if (err < 0) { + rcu_read_unlock(); goto err; + } } if (cda[CTA_PROTOINFO]) { err = ctnetlink_change_protoinfo(ct, cda); - if (err < 0) + if (err < 0) { + rcu_read_unlock(); goto err; + } } nf_ct_acct_ext_add(ct, GFP_KERNEL); @@ -1155,19 +1172,6 @@ ctnetlink_create_conntrack(struct nlattr *cda[], ct->mark = ntohl(nla_get_be32(cda[CTA_MARK])); #endif - rcu_read_lock(); - helper = __nf_ct_helper_find(rtuple); - if (helper) { - help = nf_ct_helper_ext_add(ct, GFP_KERNEL); - if (help == NULL) { - rcu_read_unlock(); - err = -ENOMEM; - goto err; - } - /* not in hash table yet so not strictly necessary */ - rcu_assign_pointer(help->helper, helper); - } - /* setup master conntrack: this is a confirmed expectation */ if (master_ct) { __set_bit(IPS_EXPECTED_BIT, &ct->status); -- cgit v1.2.3 From cb1cb5c47457ff2b604dac2da44cab4d39d11459 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 18 Aug 2008 21:31:24 -0700 Subject: netfilter: ctnetlink: fix sleep in read-side lock section Fix allocation with GFP_KERNEL in ctnetlink_create_conntrack() under read-side lock sections. This problem was introduced in 2.6.25. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/netfilter/nf_conntrack_netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index d1fb2f8555e8..a5b95ccb3ce7 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1139,7 +1139,7 @@ ctnetlink_create_conntrack(struct nlattr *cda[], rcu_read_lock(); helper = __nf_ct_helper_find(rtuple); if (helper) { - help = nf_ct_helper_ext_add(ct, GFP_KERNEL); + help = nf_ct_helper_ext_add(ct, GFP_ATOMIC); if (help == NULL) { rcu_read_unlock(); err = -ENOMEM; -- cgit v1.2.3 From fab00c5d15091546be681426c60b2ed2c10513bf Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 18 Aug 2008 21:31:46 -0700 Subject: netfilter: ctnetlink: sleepable allocation with spin lock bh This patch removes a GFP_KERNEL allocation while holding a spin lock with bottom halves disabled in ctnetlink_change_helper(). This problem was introduced in 2.6.23 with the netfilter extension infrastructure. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/netfilter/nf_conntrack_netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index a5b95ccb3ce7..a8752031adcb 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -968,7 +968,7 @@ ctnetlink_change_helper(struct nf_conn *ct, struct nlattr *cda[]) /* need to zero data of old helper */ memset(&help->help, 0, sizeof(help->help)); } else { - help = nf_ct_helper_ext_add(ct, GFP_KERNEL); + help = nf_ct_helper_ext_add(ct, GFP_ATOMIC); if (help == NULL) return -ENOMEM; } -- cgit v1.2.3 From 9f593653742d1dd816c4e94c6e5154a57ccba6d1 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Mon, 18 Aug 2008 21:32:32 -0700 Subject: nf_nat: use secure_ipv4_port_ephemeral() for NAT port randomization Use incoming network tuple as seed for NAT port randomization. This avoids concerns of leaking net_random() bits, and also gives better port distribution. Don't have NAT server, compile tested only. Signed-off-by: Stephen Hemminger [ added missing EXPORT_SYMBOL_GPL ] Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- drivers/char/random.c | 1 + net/ipv4/netfilter/nf_nat_proto_common.c | 8 ++++++-- 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/drivers/char/random.c b/drivers/char/random.c index e0d0e371909c..1838aa3d24fe 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1571,6 +1571,7 @@ u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport) return half_md4_transform(hash, keyptr->secret); } +EXPORT_SYMBOL_GPL(secure_ipv4_port_ephemeral); #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, diff --git a/net/ipv4/netfilter/nf_nat_proto_common.c b/net/ipv4/netfilter/nf_nat_proto_common.c index 91537f11273f..6c4f11f51446 100644 --- a/net/ipv4/netfilter/nf_nat_proto_common.c +++ b/net/ipv4/netfilter/nf_nat_proto_common.c @@ -73,9 +73,13 @@ bool nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, range_size = ntohs(range->max.all) - min + 1; } - off = *rover; if (range->flags & IP_NAT_RANGE_PROTO_RANDOM) - off = net_random(); + off = secure_ipv4_port_ephemeral(tuple->src.u3.ip, tuple->dst.u3.ip, + maniptype == IP_NAT_MANIP_SRC + ? tuple->dst.u.all + : tuple->src.u.all); + else + off = *rover; for (i = 0; i < range_size; i++, off++) { *portptr = htons(min + off % range_size); -- cgit v1.2.3 From e5befbd9525d92bb074b70192eb2c69aae65fc60 Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Mon, 18 Aug 2008 22:30:01 -0700 Subject: pkt_sched: remove bogus block (cleanup) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ...Last block local var got just deleted. Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- net/sched/sch_api.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index d91a2338877c..9372ec41ce84 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1088,14 +1088,13 @@ create_n_graft: } graft: - if (1) { - err = qdisc_graft(dev, p, skb, n, clid, q, NULL); - if (err) { - if (q) - qdisc_destroy(q); - return err; - } + err = qdisc_graft(dev, p, skb, n, clid, q, NULL); + if (err) { + if (q) + qdisc_destroy(q); + return err; } + return 0; } -- cgit v1.2.3 From deb3abf15fb92a608fba630da2e8719862731714 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 18 Aug 2008 22:32:10 -0700 Subject: Revert "pkt_sched: Protect gen estimators under est_lock." This reverts commit d4766692e72422f3b0f0e9ac6773d92baad07d51. qdisc_destroy() now runs in RTNL fully again, so this change is no longer needed. Signed-off-by: David S. Miller --- net/core/gen_estimator.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c index a89f32fa94f6..57abe8266be1 100644 --- a/net/core/gen_estimator.c +++ b/net/core/gen_estimator.c @@ -99,7 +99,7 @@ struct gen_estimator_head static struct gen_estimator_head elist[EST_MAX_INTERVAL+1]; -/* Protects against NULL dereference and RCU write-side */ +/* Protects against NULL dereference */ static DEFINE_RWLOCK(est_lock); static void est_timer(unsigned long arg) @@ -185,7 +185,6 @@ int gen_new_estimator(struct gnet_stats_basic *bstats, est->last_packets = bstats->packets; est->avpps = rate_est->pps<<10; - write_lock_bh(&est_lock); if (!elist[idx].timer.function) { INIT_LIST_HEAD(&elist[idx].list); setup_timer(&elist[idx].timer, est_timer, idx); @@ -195,7 +194,6 @@ int gen_new_estimator(struct gnet_stats_basic *bstats, mod_timer(&elist[idx].timer, jiffies + ((HZ/4) << idx)); list_add_rcu(&est->list, &elist[idx].list); - write_unlock_bh(&est_lock); return 0; } @@ -214,6 +212,7 @@ static void __gen_kill_estimator(struct rcu_head *head) * Removes the rate estimator specified by &bstats and &rate_est * and deletes the timer. * + * NOTE: Called under rtnl_mutex */ void gen_kill_estimator(struct gnet_stats_basic *bstats, struct gnet_stats_rate_est *rate_est) @@ -227,17 +226,17 @@ void gen_kill_estimator(struct gnet_stats_basic *bstats, if (!elist[idx].timer.function) continue; - write_lock_bh(&est_lock); list_for_each_entry_safe(e, n, &elist[idx].list, list) { if (e->rate_est != rate_est || e->bstats != bstats) continue; + write_lock_bh(&est_lock); e->bstats = NULL; + write_unlock_bh(&est_lock); list_del_rcu(&e->list); call_rcu(&e->e_rcu, __gen_kill_estimator); } - write_unlock_bh(&est_lock); } } -- cgit v1.2.3 From f3b9605d744df537dee10fd06630f35a62b343ec Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 18 Aug 2008 22:33:05 -0700 Subject: Revert "pkt_sched: Add BH protection for qdisc_stab_lock." This reverts commit 1cfa26661a85549063e369e2b40275eeaa7b923c. qdisc_destroy() runs fully under RTNL again and not from softint any longer, so this change is no longer needed. Signed-off-by: David S. Miller --- net/sched/sch_api.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 9372ec41ce84..ef0efeca6352 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -332,7 +332,7 @@ static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt) if (!s || tsize != s->tsize || (!tab && tsize > 0)) return ERR_PTR(-EINVAL); - spin_lock_bh(&qdisc_stab_lock); + spin_lock(&qdisc_stab_lock); list_for_each_entry(stab, &qdisc_stab_list, list) { if (memcmp(&stab->szopts, s, sizeof(*s))) @@ -340,11 +340,11 @@ static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt) if (tsize > 0 && memcmp(stab->data, tab, tsize * sizeof(u16))) continue; stab->refcnt++; - spin_unlock_bh(&qdisc_stab_lock); + spin_unlock(&qdisc_stab_lock); return stab; } - spin_unlock_bh(&qdisc_stab_lock); + spin_unlock(&qdisc_stab_lock); stab = kmalloc(sizeof(*stab) + tsize * sizeof(u16), GFP_KERNEL); if (!stab) @@ -355,9 +355,9 @@ static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt) if (tsize > 0) memcpy(stab->data, tab, tsize * sizeof(u16)); - spin_lock_bh(&qdisc_stab_lock); + spin_lock(&qdisc_stab_lock); list_add_tail(&stab->list, &qdisc_stab_list); - spin_unlock_bh(&qdisc_stab_lock); + spin_unlock(&qdisc_stab_lock); return stab; } @@ -367,14 +367,14 @@ void qdisc_put_stab(struct qdisc_size_table *tab) if (!tab) return; - spin_lock_bh(&qdisc_stab_lock); + spin_lock(&qdisc_stab_lock); if (--tab->refcnt == 0) { list_del(&tab->list); kfree(tab); } - spin_unlock_bh(&qdisc_stab_lock); + spin_unlock(&qdisc_stab_lock); } EXPORT_SYMBOL(qdisc_put_stab); -- cgit v1.2.3 From 195648bbc5ae0848e82f771ecf4cd7497054c212 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 19 Aug 2008 04:00:36 -0700 Subject: pkt_sched: Prevent livelock in TX queue running. If dev_deactivate() is trying to quiesce the queue, it is theoretically possible for another cpu to livelock trying to process that queue. This happens because dev_deactivate() grabs the queue spinlock as it checks the queue state, whereas net_tx_action() does a trylock and reschedules the qdisc if it hits the lock. This breaks the livelock by adding a check on __QDISC_STATE_DEACTIVATED to net_tx_action() when the trylock fails. Based upon feedback from Herbert Xu and Jarek Poplawski. Signed-off-by: David S. Miller --- net/core/dev.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 8d133802372b..60c51f765887 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1990,7 +1990,9 @@ static void net_tx_action(struct softirq_action *h) qdisc_run(q); spin_unlock(root_lock); } else { - __netif_reschedule(q); + if (!test_bit(__QDISC_STATE_DEACTIVATED, + &q->state)) + __netif_reschedule(q); } } } -- cgit v1.2.3 From 5e739d1752aca4e8f3e794d431503bfca3162df4 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Thu, 21 Aug 2008 03:34:25 -0700 Subject: sctp: fix potential panics in the SCTP-AUTH API. All of the SCTP-AUTH socket options could cause a panic if the extension is disabled and the API is envoked. Additionally, there were some additional assumptions that certain pointers would always be valid which may not always be the case. This patch hardens the API and address all of the crash scenarios. Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/sctp/endpointola.c | 4 +-- net/sctp/socket.c | 85 ++++++++++++++++++++++++++++++++++++++------------ 2 files changed, 67 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index e39a0cdef184..4c8d9f45ce09 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -103,6 +103,7 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, /* Initialize the CHUNKS parameter */ auth_chunks->param_hdr.type = SCTP_PARAM_CHUNKS; + auth_chunks->param_hdr.length = htons(sizeof(sctp_paramhdr_t)); /* If the Add-IP functionality is enabled, we must * authenticate, ASCONF and ASCONF-ACK chunks @@ -110,8 +111,7 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, if (sctp_addip_enable) { auth_chunks->chunks[0] = SCTP_CID_ASCONF; auth_chunks->chunks[1] = SCTP_CID_ASCONF_ACK; - auth_chunks->param_hdr.length = - htons(sizeof(sctp_paramhdr_t) + 2); + auth_chunks->param_hdr.length += htons(2); } } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index dbb79adf8f3c..bb5c9ef13046 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3055,6 +3055,9 @@ static int sctp_setsockopt_auth_chunk(struct sock *sk, { struct sctp_authchunk val; + if (!sctp_auth_enable) + return -EACCES; + if (optlen != sizeof(struct sctp_authchunk)) return -EINVAL; if (copy_from_user(&val, optval, optlen)) @@ -3085,6 +3088,9 @@ static int sctp_setsockopt_hmac_ident(struct sock *sk, struct sctp_hmacalgo *hmacs; int err; + if (!sctp_auth_enable) + return -EACCES; + if (optlen < sizeof(struct sctp_hmacalgo)) return -EINVAL; @@ -3123,6 +3129,9 @@ static int sctp_setsockopt_auth_key(struct sock *sk, struct sctp_association *asoc; int ret; + if (!sctp_auth_enable) + return -EACCES; + if (optlen <= sizeof(struct sctp_authkey)) return -EINVAL; @@ -3160,6 +3169,9 @@ static int sctp_setsockopt_active_key(struct sock *sk, struct sctp_authkeyid val; struct sctp_association *asoc; + if (!sctp_auth_enable) + return -EACCES; + if (optlen != sizeof(struct sctp_authkeyid)) return -EINVAL; if (copy_from_user(&val, optval, optlen)) @@ -3185,6 +3197,9 @@ static int sctp_setsockopt_del_key(struct sock *sk, struct sctp_authkeyid val; struct sctp_association *asoc; + if (!sctp_auth_enable) + return -EACCES; + if (optlen != sizeof(struct sctp_authkeyid)) return -EINVAL; if (copy_from_user(&val, optval, optlen)) @@ -5197,19 +5212,29 @@ static int sctp_getsockopt_maxburst(struct sock *sk, int len, static int sctp_getsockopt_hmac_ident(struct sock *sk, int len, char __user *optval, int __user *optlen) { + struct sctp_hmacalgo __user *p = (void __user *)optval; struct sctp_hmac_algo_param *hmacs; - __u16 param_len; + __u16 data_len = 0; + u32 num_idents; + + if (!sctp_auth_enable) + return -EACCES; hmacs = sctp_sk(sk)->ep->auth_hmacs_list; - param_len = ntohs(hmacs->param_hdr.length); + data_len = ntohs(hmacs->param_hdr.length) - sizeof(sctp_paramhdr_t); - if (len < param_len) + if (len < sizeof(struct sctp_hmacalgo) + data_len) return -EINVAL; + + len = sizeof(struct sctp_hmacalgo) + data_len; + num_idents = data_len / sizeof(u16); + if (put_user(len, optlen)) return -EFAULT; - if (copy_to_user(optval, hmacs->hmac_ids, len)) + if (put_user(num_idents, &p->shmac_num_idents)) + return -EFAULT; + if (copy_to_user(p->shmac_idents, hmacs->hmac_ids, data_len)) return -EFAULT; - return 0; } @@ -5219,6 +5244,9 @@ static int sctp_getsockopt_active_key(struct sock *sk, int len, struct sctp_authkeyid val; struct sctp_association *asoc; + if (!sctp_auth_enable) + return -EACCES; + if (len < sizeof(struct sctp_authkeyid)) return -EINVAL; if (copy_from_user(&val, optval, sizeof(struct sctp_authkeyid))) @@ -5233,6 +5261,12 @@ static int sctp_getsockopt_active_key(struct sock *sk, int len, else val.scact_keynumber = sctp_sk(sk)->ep->active_key_id; + len = sizeof(struct sctp_authkeyid); + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, len)) + return -EFAULT; + return 0; } @@ -5243,13 +5277,16 @@ static int sctp_getsockopt_peer_auth_chunks(struct sock *sk, int len, struct sctp_authchunks val; struct sctp_association *asoc; struct sctp_chunks_param *ch; - u32 num_chunks; + u32 num_chunks = 0; char __user *to; - if (len <= sizeof(struct sctp_authchunks)) + if (!sctp_auth_enable) + return -EACCES; + + if (len < sizeof(struct sctp_authchunks)) return -EINVAL; - if (copy_from_user(&val, p, sizeof(struct sctp_authchunks))) + if (copy_from_user(&val, optval, sizeof(struct sctp_authchunks))) return -EFAULT; to = p->gauth_chunks; @@ -5258,20 +5295,21 @@ static int sctp_getsockopt_peer_auth_chunks(struct sock *sk, int len, return -EINVAL; ch = asoc->peer.peer_chunks; + if (!ch) + goto num; /* See if the user provided enough room for all the data */ num_chunks = ntohs(ch->param_hdr.length) - sizeof(sctp_paramhdr_t); if (len < num_chunks) return -EINVAL; - len = num_chunks; - if (put_user(len, optlen)) + if (copy_to_user(to, ch->chunks, num_chunks)) return -EFAULT; +num: + len = sizeof(struct sctp_authchunks) + num_chunks; + if (put_user(len, optlen)) return -EFAULT; if (put_user(num_chunks, &p->gauth_number_of_chunks)) return -EFAULT; - if (copy_to_user(to, ch->chunks, len)) - return -EFAULT; - return 0; } @@ -5282,13 +5320,16 @@ static int sctp_getsockopt_local_auth_chunks(struct sock *sk, int len, struct sctp_authchunks val; struct sctp_association *asoc; struct sctp_chunks_param *ch; - u32 num_chunks; + u32 num_chunks = 0; char __user *to; - if (len <= sizeof(struct sctp_authchunks)) + if (!sctp_auth_enable) + return -EACCES; + + if (len < sizeof(struct sctp_authchunks)) return -EINVAL; - if (copy_from_user(&val, p, sizeof(struct sctp_authchunks))) + if (copy_from_user(&val, optval, sizeof(struct sctp_authchunks))) return -EFAULT; to = p->gauth_chunks; @@ -5301,17 +5342,21 @@ static int sctp_getsockopt_local_auth_chunks(struct sock *sk, int len, else ch = sctp_sk(sk)->ep->auth_chunk_list; + if (!ch) + goto num; + num_chunks = ntohs(ch->param_hdr.length) - sizeof(sctp_paramhdr_t); - if (len < num_chunks) + if (len < sizeof(struct sctp_authchunks) + num_chunks) return -EINVAL; - len = num_chunks; + if (copy_to_user(to, ch->chunks, num_chunks)) + return -EFAULT; +num: + len = sizeof(struct sctp_authchunks) + num_chunks; if (put_user(len, optlen)) return -EFAULT; if (put_user(num_chunks, &p->gauth_number_of_chunks)) return -EFAULT; - if (copy_to_user(to, ch->chunks, len)) - return -EFAULT; return 0; } -- cgit v1.2.3 From 2540e0511ea17e25831be543cdf9381e6209950d Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Thu, 21 Aug 2008 05:11:14 -0700 Subject: pkt_sched: Fix qdisc_watchdog() vs. dev_deactivate() race dev_deactivate() can skip rescheduling of a qdisc by qdisc_watchdog() or other timer calling netif_schedule() after dev_queue_deactivate(). We prevent this checking aliveness before scheduling the timer. Since during deactivation the root qdisc is available only as qdisc_sleeping additional accessor qdisc_root_sleeping() is created. With feedback from Herbert Xu Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- include/net/sch_generic.h | 5 +++++ net/sched/sch_api.c | 4 ++++ net/sched/sch_cbq.c | 4 ++++ 3 files changed, 13 insertions(+) (limited to 'net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 84d25f2e6188..b1d2cfea89c5 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -193,6 +193,11 @@ static inline struct Qdisc *qdisc_root(struct Qdisc *qdisc) return qdisc->dev_queue->qdisc; } +static inline struct Qdisc *qdisc_root_sleeping(struct Qdisc *qdisc) +{ + return qdisc->dev_queue->qdisc_sleeping; +} + /* The qdisc root lock is a mechanism by which to top level * of a qdisc tree can be locked from any qdisc node in the * forest. This allows changing the configuration of some diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index ef0efeca6352..45f442d7de47 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -444,6 +444,10 @@ void qdisc_watchdog_schedule(struct qdisc_watchdog *wd, psched_time_t expires) { ktime_t time; + if (test_bit(__QDISC_STATE_DEACTIVATED, + &qdisc_root_sleeping(wd->qdisc)->state)) + return; + wd->qdisc->flags |= TCQ_F_THROTTLED; time = ktime_set(0, 0); time = ktime_add_ns(time, PSCHED_US2NS(expires)); diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 47ef492c4ff4..8fa90d68ec6d 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -521,6 +521,10 @@ static void cbq_ovl_delay(struct cbq_class *cl) struct cbq_sched_data *q = qdisc_priv(cl->qdisc); psched_tdiff_t delay = cl->undertime - q->now; + if (test_bit(__QDISC_STATE_DEACTIVATED, + &qdisc_root_sleeping(cl->qdisc)->state)) + return; + if (!cl->delayed) { psched_time_t sched = q->now; ktime_t expires; -- cgit v1.2.3 From f6e0b239a2657ea8cb67f0d83d0bfdbfd19a481b Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Fri, 22 Aug 2008 03:24:05 -0700 Subject: pkt_sched: Fix qdisc list locking Since some qdiscs call qdisc_tree_decrease_qlen() (so qdisc_lookup()) without rtnl_lock(), adding and deleting from a qdisc list needs additional locking. This patch adds global spinlock qdisc_list_lock and wrapper functions for modifying the list. It is considered as a temporary solution until hfsc_dequeue(), netem_dequeue() and tbf_dequeue() (or qdisc_tree_decrease_qlen()) are redone. With feedback from Herbert Xu and David S. Miller. Signed-off-by: Jarek Poplawski Acked-by: Herbert Xu Signed-off-by: David S. Miller --- include/net/pkt_sched.h | 1 + net/sched/sch_api.c | 44 +++++++++++++++++++++++++++++++++++++++----- net/sched/sch_generic.c | 5 ++--- 3 files changed, 42 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 853fe83d9f37..b786a5b09253 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -78,6 +78,7 @@ extern struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops, extern int register_qdisc(struct Qdisc_ops *qops); extern int unregister_qdisc(struct Qdisc_ops *qops); +extern void qdisc_list_del(struct Qdisc *q); extern struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle); extern struct Qdisc *qdisc_lookup_class(struct net_device *dev, u32 handle); extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 45f442d7de47..e7fb9e0d21b4 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -199,19 +199,53 @@ struct Qdisc *qdisc_match_from_root(struct Qdisc *root, u32 handle) return NULL; } +/* + * This lock is needed until some qdiscs stop calling qdisc_tree_decrease_qlen() + * without rtnl_lock(); currently hfsc_dequeue(), netem_dequeue(), tbf_dequeue() + */ +static DEFINE_SPINLOCK(qdisc_list_lock); + +static void qdisc_list_add(struct Qdisc *q) +{ + if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { + spin_lock_bh(&qdisc_list_lock); + list_add_tail(&q->list, &qdisc_root_sleeping(q)->list); + spin_unlock_bh(&qdisc_list_lock); + } +} + +void qdisc_list_del(struct Qdisc *q) +{ + if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { + spin_lock_bh(&qdisc_list_lock); + list_del(&q->list); + spin_unlock_bh(&qdisc_list_lock); + } +} +EXPORT_SYMBOL(qdisc_list_del); + struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) { unsigned int i; + struct Qdisc *q; + + spin_lock_bh(&qdisc_list_lock); for (i = 0; i < dev->num_tx_queues; i++) { struct netdev_queue *txq = netdev_get_tx_queue(dev, i); - struct Qdisc *q, *txq_root = txq->qdisc_sleeping; + struct Qdisc *txq_root = txq->qdisc_sleeping; q = qdisc_match_from_root(txq_root, handle); if (q) - return q; + goto unlock; } - return qdisc_match_from_root(dev->rx_queue.qdisc_sleeping, handle); + + q = qdisc_match_from_root(dev->rx_queue.qdisc_sleeping, handle); + +unlock: + spin_unlock_bh(&qdisc_list_lock); + + return q; } static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid) @@ -810,8 +844,8 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, goto err_out3; } } - if ((parent != TC_H_ROOT) && !(sch->flags & TCQ_F_INGRESS)) - list_add_tail(&sch->list, &dev_queue->qdisc_sleeping->list); + + qdisc_list_add(sch); return sch; } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index c3ed4d44fc14..5f0ade7806a7 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -526,10 +526,9 @@ void qdisc_destroy(struct Qdisc *qdisc) !atomic_dec_and_test(&qdisc->refcnt)) return; - if (qdisc->parent) - list_del(&qdisc->list); - #ifdef CONFIG_NET_SCHED + qdisc_list_del(qdisc); + qdisc_put_stab(qdisc->stab); #endif gen_kill_estimator(&qdisc->bstats, &qdisc->rate_est); -- cgit v1.2.3 From fdc0bde90a689b9145f2b6f271c03f4c99d09667 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Sat, 23 Aug 2008 04:43:33 -0700 Subject: icmp: icmp_sk() should not use smp_processor_id() in preemptible code Pass namespace into icmp_xmit_lock, obtain socket inside and return it as a result for caller. Thanks Alexey Dobryan for this report: Steps to reproduce: CONFIG_PREEMPT=y CONFIG_DEBUG_PREEMPT=y tracepath BUG: using smp_processor_id() in preemptible [00000000] code: tracepath/3205 caller is icmp_sk+0x15/0x30 Pid: 3205, comm: tracepath Not tainted 2.6.27-rc4 #1 Call Trace: [] debug_smp_processor_id+0xe4/0xf0 [] icmp_sk+0x15/0x30 [] icmp_send+0x4b/0x3f0 [] ? trace_hardirqs_on_caller+0xd5/0x160 [] ? trace_hardirqs_on+0xd/0x10 [] ? local_bh_enable_ip+0x95/0x110 [] ? _spin_unlock_bh+0x39/0x40 [] ? mark_held_locks+0x4c/0x90 [] ? trace_hardirqs_on+0xd/0x10 [] ? trace_hardirqs_on_caller+0xd5/0x160 [] ip_fragment+0x8d4/0x900 [] ? ip_finish_output2+0x0/0x290 [] ? ip_finish_output+0x0/0x60 [] ? dst_output+0x0/0x10 [] ip_finish_output+0x4c/0x60 [] ip_output+0xa3/0xf0 [] ip_local_out+0x20/0x30 [] ip_push_pending_frames+0x27f/0x400 [] udp_push_pending_frames+0x233/0x3d0 [] udp_sendmsg+0x321/0x6f0 [] inet_sendmsg+0x45/0x80 [] sock_sendmsg+0xdf/0x110 [] ? autoremove_wake_function+0x0/0x40 [] ? validate_chain+0x415/0x1010 [] ? __do_fault+0x140/0x450 [] ? __lock_acquire+0x260/0x590 [] ? sockfd_lookup_light+0x45/0x80 [] sys_sendto+0xea/0x120 [] ? _spin_unlock_irqrestore+0x42/0x80 [] ? __up_read+0x4c/0xb0 [] ? up_read+0x26/0x30 [] system_call_fastpath+0x16/0x1b icmp6_sk() is similar. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- net/ipv4/icmp.c | 22 ++++++++++++++-------- net/ipv6/icmp.c | 23 ++++++++++++----------- 2 files changed, 26 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 860558633b2c..55c355e63234 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -204,18 +204,22 @@ static struct sock *icmp_sk(struct net *net) return net->ipv4.icmp_sk[smp_processor_id()]; } -static inline int icmp_xmit_lock(struct sock *sk) +static inline struct sock *icmp_xmit_lock(struct net *net) { + struct sock *sk; + local_bh_disable(); + sk = icmp_sk(net); + if (unlikely(!spin_trylock(&sk->sk_lock.slock))) { /* This can happen if the output path signals a * dst_link_failure() for an outgoing ICMP packet. */ local_bh_enable(); - return 1; + return NULL; } - return 0; + return sk; } static inline void icmp_xmit_unlock(struct sock *sk) @@ -354,15 +358,17 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) struct ipcm_cookie ipc; struct rtable *rt = skb->rtable; struct net *net = dev_net(rt->u.dst.dev); - struct sock *sk = icmp_sk(net); - struct inet_sock *inet = inet_sk(sk); + struct sock *sk; + struct inet_sock *inet; __be32 daddr; if (ip_options_echo(&icmp_param->replyopts, skb)) return; - if (icmp_xmit_lock(sk)) + sk = icmp_xmit_lock(net); + if (sk == NULL) return; + inet = inet_sk(sk); icmp_param->data.icmph.checksum = 0; @@ -419,7 +425,6 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) if (!rt) goto out; net = dev_net(rt->u.dst.dev); - sk = icmp_sk(net); /* * Find the original header. It is expected to be valid, of course. @@ -483,7 +488,8 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) } } - if (icmp_xmit_lock(sk)) + sk = icmp_xmit_lock(net); + if (sk == NULL) return; /* diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index abedf95fdf2d..b3157a0cc15d 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -91,19 +91,22 @@ static struct inet6_protocol icmpv6_protocol = { .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, }; -static __inline__ int icmpv6_xmit_lock(struct sock *sk) +static __inline__ struct sock *icmpv6_xmit_lock(struct net *net) { + struct sock *sk; + local_bh_disable(); + sk = icmpv6_sk(net); if (unlikely(!spin_trylock(&sk->sk_lock.slock))) { /* This can happen if the output path (f.e. SIT or * ip6ip6 tunnel) signals dst_link_failure() for an * outgoing ICMP6 packet. */ local_bh_enable(); - return 1; + return NULL; } - return 0; + return sk; } static __inline__ void icmpv6_xmit_unlock(struct sock *sk) @@ -392,11 +395,10 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, fl.fl_icmp_code = code; security_skb_classify_flow(skb, &fl); - sk = icmpv6_sk(net); - np = inet6_sk(sk); - - if (icmpv6_xmit_lock(sk)) + sk = icmpv6_xmit_lock(net); + if (sk == NULL) return; + np = inet6_sk(sk); if (!icmpv6_xrlim_allow(sk, type, &fl)) goto out; @@ -539,11 +541,10 @@ static void icmpv6_echo_reply(struct sk_buff *skb) fl.fl_icmp_type = ICMPV6_ECHO_REPLY; security_skb_classify_flow(skb, &fl); - sk = icmpv6_sk(net); - np = inet6_sk(sk); - - if (icmpv6_xmit_lock(sk)) + sk = icmpv6_xmit_lock(net); + if (sk == NULL) return; + np = inet6_sk(sk); if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst)) fl.oif = np->mcast_oif; -- cgit v1.2.3 From f410a1fba7afa79d2992620e874a343fdba28332 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Sat, 23 Aug 2008 05:16:46 -0700 Subject: ipv6: protocol for address routes This fixes a problem spotted with zebra, but not sure if it is necessary a kernel problem. With IPV6 when an address is added to an interface, Zebra creates a duplicate RIB entry, one as a connected route, and other as a kernel route. When an address is added to an interface the RTN_NEWADDR message causes Zebra to create a connected route. In IPV4 when an address is added to an interface a RTN_NEWROUTE message is set to user space with the protocol RTPROT_KERNEL. Zebra ignores these messages, because it already has the connected route. The problem is that route created in IPV6 has route protocol == RTPROT_BOOT. Was this a design decision or a bug? This fixes it. Same patch applies to both net-2.6 and stable. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index e2d3b7580b76..7b6a584b62dd 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1688,6 +1688,7 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev, .fc_dst_len = plen, .fc_flags = RTF_UP | flags, .fc_nlinfo.nl_net = dev_net(dev), + .fc_protocol = RTPROT_KERNEL, }; ipv6_addr_copy(&cfg.fc_dst, pfx); -- cgit v1.2.3 From 30c2235cbc477d4629983d440cdc4f496fec9246 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 25 Aug 2008 15:16:19 -0700 Subject: sctp: add verification checks to SCTP_AUTH_KEY option The structure used for SCTP_AUTH_KEY option contains a length that needs to be verfied to prevent buffer overflow conditions. Spoted by Eugene Teo . Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/sctp/auth.c | 4 ++++ net/sctp/socket.c | 5 +++++ 2 files changed, 9 insertions(+) (limited to 'net') diff --git a/net/sctp/auth.c b/net/sctp/auth.c index 675a5c3e68a6..1fcb4cf2f4c9 100644 --- a/net/sctp/auth.c +++ b/net/sctp/auth.c @@ -80,6 +80,10 @@ static struct sctp_auth_bytes *sctp_auth_create_key(__u32 key_len, gfp_t gfp) { struct sctp_auth_bytes *key; + /* Verify that we are not going to overflow INT_MAX */ + if ((INT_MAX - key_len) < sizeof(struct sctp_auth_bytes)) + return NULL; + /* Allocate the shared key */ key = kmalloc(sizeof(struct sctp_auth_bytes) + key_len, gfp); if (!key) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index bb5c9ef13046..afa952e726d7 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3144,6 +3144,11 @@ static int sctp_setsockopt_auth_key(struct sock *sk, goto out; } + if (authkey->sca_keylength > optlen) { + ret = -EINVAL; + goto out; + } + asoc = sctp_id2assoc(sk, authkey->sca_assoc_id); if (!asoc && authkey->sca_assoc_id && sctp_style(sk, UDP)) { ret = -EINVAL; -- cgit v1.2.3 From 2f4520d35d89ca6c5cd129c38e3b11f0283b7d1b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 25 Aug 2008 15:17:44 -0700 Subject: ipv4: sysctl fixes net.ipv4.neigh should be a part of skeleton to avoid ordering problems Signed-off-by: Al Viro Signed-off-by: David S. Miller --- net/ipv4/route.c | 24 ++++++++++++++++++++---- 1 file changed, 20 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/route.c b/net/ipv4/route.c index cca921ea8550..e91bafeb32f4 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -3116,14 +3116,23 @@ static ctl_table ipv4_route_table[] = { { .ctl_name = 0 } }; -static __net_initdata struct ctl_path ipv4_route_path[] = { +static struct ctl_table empty[1]; + +static struct ctl_table ipv4_skeleton[] = +{ + { .procname = "route", .ctl_name = NET_IPV4_ROUTE, + .child = ipv4_route_table}, + { .procname = "neigh", .ctl_name = NET_IPV4_NEIGH, + .child = empty}, + { } +}; + +static __net_initdata struct ctl_path ipv4_path[] = { { .procname = "net", .ctl_name = CTL_NET, }, { .procname = "ipv4", .ctl_name = NET_IPV4, }, - { .procname = "route", .ctl_name = NET_IPV4_ROUTE, }, { }, }; - static struct ctl_table ipv4_route_flush_table[] = { { .ctl_name = NET_IPV4_ROUTE_FLUSH, @@ -3136,6 +3145,13 @@ static struct ctl_table ipv4_route_flush_table[] = { { .ctl_name = 0 }, }; +static __net_initdata struct ctl_path ipv4_route_path[] = { + { .procname = "net", .ctl_name = CTL_NET, }, + { .procname = "ipv4", .ctl_name = NET_IPV4, }, + { .procname = "route", .ctl_name = NET_IPV4_ROUTE, }, + { }, +}; + static __net_init int sysctl_route_net_init(struct net *net) { struct ctl_table *tbl; @@ -3287,7 +3303,7 @@ int __init ip_rt_init(void) */ void __init ip_static_sysctl_init(void) { - register_sysctl_paths(ipv4_route_path, ipv4_route_table); + register_sysctl_paths(ipv4_path, ipv4_skeleton); } #endif -- cgit v1.2.3 From ce3113ec57abcd41cc5a2fed02474aee3f63d12c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 25 Aug 2008 15:18:15 -0700 Subject: ipv6: sysctl fixes Braino: net.ipv6 in ipv6 skeleton has no business in rotable class Signed-off-by: Al Viro Signed-off-by: David S. Miller --- net/ipv6/sysctl_net_ipv6.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c index e6dfaeac6be3..587f8f60c489 100644 --- a/net/ipv6/sysctl_net_ipv6.c +++ b/net/ipv6/sysctl_net_ipv6.c @@ -156,7 +156,7 @@ static struct ctl_table_header *ip6_base; int ipv6_static_sysctl_register(void) { static struct ctl_table empty[1]; - ip6_base = register_net_sysctl_rotable(net_ipv6_ctl_path, empty); + ip6_base = register_sysctl_paths(net_ipv6_ctl_path, empty); if (ip6_base == NULL) return -ENOMEM; return 0; -- cgit v1.2.3 From 988b02f1bf5b608ef91a9d98c7170d037d0f12e3 Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Mon, 18 Aug 2008 22:56:53 +0300 Subject: net: rfkill: add missing line break Trivial patch adding a missing line break on rfkill_claim_show(). Signed-off-by: Felipe Balbi Acked-by: Ivo van Doorn Signed-off-by: John W. Linville --- net/rfkill/rfkill.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c index 35a9994e2339..74aecc098bad 100644 --- a/net/rfkill/rfkill.c +++ b/net/rfkill/rfkill.c @@ -377,7 +377,7 @@ static ssize_t rfkill_claim_show(struct device *dev, { struct rfkill *rfkill = to_rfkill(dev); - return sprintf(buf, "%d", rfkill->user_claim); + return sprintf(buf, "%d\n", rfkill->user_claim); } static ssize_t rfkill_claim_store(struct device *dev, -- cgit v1.2.3 From 087d833e5a9f67ba933cb32eaf5a2279c1a5b47c Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Tue, 19 Aug 2008 10:54:32 +0300 Subject: mac80211: Use IWEVASSOCREQIE instead of IWEVCUSTOM The previous code was using IWEVCUSTOM to report IEs from AssocReq and AssocResp frames into user space. This can easily hit the 256 byte limit (IW_CUSTOM_MAX) with APs that include number of vendor IEs in AssocResp. This results in the event message not being sent and dmesg showing "wlan0 (WE) : Wireless Event too big (366)" type of errors. Convert mac80211 to use IWEVASSOCREQIE/IWEVASSOCRESPIE to avoid the issue of being unable to send association IEs as wireless events. These newer event types use binary encoding and larger maximum size (IW_GENERIC_IE_MAX = 1024), so the likelyhood of not being able to send the IEs is much smaller than with IWEVCUSTOM. As an extra benefit, the code is also quite a bit simpler since there is no need to allocate an extra buffer for hex encoding. Signed-off-by: Jouni Malinen Signed-off-by: John W. Linville --- net/mac80211/mlme.c | 48 +++++++++--------------------------------------- 1 file changed, 9 insertions(+), 39 deletions(-) (limited to 'net') diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 1e97fb9fb34b..09a56e24b799 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -478,51 +478,21 @@ int ieee80211_ht_addt_info_ie_to_ht_bss_info( static void ieee80211_sta_send_associnfo(struct net_device *dev, struct ieee80211_if_sta *ifsta) { - char *buf; - size_t len; - int i; union iwreq_data wrqu; - if (!ifsta->assocreq_ies && !ifsta->assocresp_ies) - return; - - buf = kmalloc(50 + 2 * (ifsta->assocreq_ies_len + - ifsta->assocresp_ies_len), GFP_KERNEL); - if (!buf) - return; - - len = sprintf(buf, "ASSOCINFO("); if (ifsta->assocreq_ies) { - len += sprintf(buf + len, "ReqIEs="); - for (i = 0; i < ifsta->assocreq_ies_len; i++) { - len += sprintf(buf + len, "%02x", - ifsta->assocreq_ies[i]); - } - } - if (ifsta->assocresp_ies) { - if (ifsta->assocreq_ies) - len += sprintf(buf + len, " "); - len += sprintf(buf + len, "RespIEs="); - for (i = 0; i < ifsta->assocresp_ies_len; i++) { - len += sprintf(buf + len, "%02x", - ifsta->assocresp_ies[i]); - } + memset(&wrqu, 0, sizeof(wrqu)); + wrqu.data.length = ifsta->assocreq_ies_len; + wireless_send_event(dev, IWEVASSOCREQIE, &wrqu, + ifsta->assocreq_ies); } - len += sprintf(buf + len, ")"); - if (len > IW_CUSTOM_MAX) { - len = sprintf(buf, "ASSOCRESPIE="); - for (i = 0; i < ifsta->assocresp_ies_len; i++) { - len += sprintf(buf + len, "%02x", - ifsta->assocresp_ies[i]); - } + if (ifsta->assocresp_ies) { + memset(&wrqu, 0, sizeof(wrqu)); + wrqu.data.length = ifsta->assocresp_ies_len; + wireless_send_event(dev, IWEVASSOCRESPIE, &wrqu, + ifsta->assocresp_ies); } - - memset(&wrqu, 0, sizeof(wrqu)); - wrqu.data.length = len; - wireless_send_event(dev, IWEVCUSTOM, &wrqu, buf); - - kfree(buf); } -- cgit v1.2.3 From 667d8af9af775a5fea203890978037ea750816cc Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sat, 23 Aug 2008 18:27:38 +0200 Subject: net/mac80211/mesh.c: correct the argument to __mesh_table_free In the function mesh_table_grow, it is the new table not the argument table that should be freed if the function fails (cf commit bd9b448f4c0a514559bdae4ca18ca3e8cd999c6d) The semantic match that detects this problem is as follows: (http://www.emn.fr/x-info/coccinelle/) // @r exists@ local idexpression x; expression E,f; position p1,p2,p3; identifier l; statement S; @@ x = mesh_table_alloc@p1(...) ... if (x == NULL) S ... when != E = x when != mesh_table_free(x) goto@p2 l; ... when != E = x when != f(...,x,...) when any ( return \(0\|x\); | return@p3 ...; ) @script:python@ p1 << r.p1; p2 << r.p2; p3 << r.p3; @@ print "%s: call on line %s not freed or saved before return on line %s via line %s" % (p1[0].file,p1[0].line,p3[0].line,p2[0].line) // Signed-off-by: Julia Lawall Signed-off-by: John W. Linville --- net/mac80211/mesh.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c index b5933b271491..35f2f95f2fa7 100644 --- a/net/mac80211/mesh.c +++ b/net/mac80211/mesh.c @@ -383,7 +383,7 @@ errcopy: hlist_for_each_safe(p, q, &newtbl->hash_buckets[i]) tbl->free_node(p, 0); } - __mesh_table_free(tbl); + __mesh_table_free(newtbl); endgrow: return NULL; } -- cgit v1.2.3 From 93015f0f34e81a47c4126329746ce5f364bafd11 Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Mon, 25 Aug 2008 11:57:06 +0300 Subject: mac80211: Fix debugfs file add/del for netdev Previous version was using incorrect union structures for non-AP interfaces when adding and removing max_ratectrl_rateidx and force_unicast_rateidx entries. Depending on the vif type, this ended up in corrupting debugfs entries since the dentries inside different union structures ended up going being on top of eachother.. As the end result, debugfs files were being left behind with references to freed data (instant kernel oops on access) and directories were not removed properly when unloading mac80211 drivers. This patch fixes those issues by using only a single union structure based on the vif type. Signed-off-by: Jouni Malinen Signed-off-by: John W. Linville --- net/mac80211/debugfs_netdev.c | 24 ++++++++++++------------ net/mac80211/ieee80211_i.h | 6 ++++++ 2 files changed, 18 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c index 475f89a8aee1..8165df578c92 100644 --- a/net/mac80211/debugfs_netdev.c +++ b/net/mac80211/debugfs_netdev.c @@ -248,8 +248,8 @@ IEEE80211_IF_WFILE(min_discovery_timeout, static void add_sta_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_ADD(drop_unencrypted, sta); - DEBUGFS_ADD(force_unicast_rateidx, ap); - DEBUGFS_ADD(max_ratectrl_rateidx, ap); + DEBUGFS_ADD(force_unicast_rateidx, sta); + DEBUGFS_ADD(max_ratectrl_rateidx, sta); DEBUGFS_ADD(state, sta); DEBUGFS_ADD(bssid, sta); @@ -283,8 +283,8 @@ static void add_ap_files(struct ieee80211_sub_if_data *sdata) static void add_wds_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_ADD(drop_unencrypted, wds); - DEBUGFS_ADD(force_unicast_rateidx, ap); - DEBUGFS_ADD(max_ratectrl_rateidx, ap); + DEBUGFS_ADD(force_unicast_rateidx, wds); + DEBUGFS_ADD(max_ratectrl_rateidx, wds); DEBUGFS_ADD(peer, wds); } @@ -292,8 +292,8 @@ static void add_wds_files(struct ieee80211_sub_if_data *sdata) static void add_vlan_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_ADD(drop_unencrypted, vlan); - DEBUGFS_ADD(force_unicast_rateidx, ap); - DEBUGFS_ADD(max_ratectrl_rateidx, ap); + DEBUGFS_ADD(force_unicast_rateidx, vlan); + DEBUGFS_ADD(max_ratectrl_rateidx, vlan); } static void add_monitor_files(struct ieee80211_sub_if_data *sdata) @@ -381,8 +381,8 @@ static void add_files(struct ieee80211_sub_if_data *sdata) static void del_sta_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_DEL(drop_unencrypted, sta); - DEBUGFS_DEL(force_unicast_rateidx, ap); - DEBUGFS_DEL(max_ratectrl_rateidx, ap); + DEBUGFS_DEL(force_unicast_rateidx, sta); + DEBUGFS_DEL(max_ratectrl_rateidx, sta); DEBUGFS_DEL(state, sta); DEBUGFS_DEL(bssid, sta); @@ -416,8 +416,8 @@ static void del_ap_files(struct ieee80211_sub_if_data *sdata) static void del_wds_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_DEL(drop_unencrypted, wds); - DEBUGFS_DEL(force_unicast_rateidx, ap); - DEBUGFS_DEL(max_ratectrl_rateidx, ap); + DEBUGFS_DEL(force_unicast_rateidx, wds); + DEBUGFS_DEL(max_ratectrl_rateidx, wds); DEBUGFS_DEL(peer, wds); } @@ -425,8 +425,8 @@ static void del_wds_files(struct ieee80211_sub_if_data *sdata) static void del_vlan_files(struct ieee80211_sub_if_data *sdata) { DEBUGFS_DEL(drop_unencrypted, vlan); - DEBUGFS_DEL(force_unicast_rateidx, ap); - DEBUGFS_DEL(max_ratectrl_rateidx, ap); + DEBUGFS_DEL(force_unicast_rateidx, vlan); + DEBUGFS_DEL(max_ratectrl_rateidx, vlan); } static void del_monitor_files(struct ieee80211_sub_if_data *sdata) diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index ec59345af65b..586a9b49b0fc 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -470,6 +470,8 @@ struct ieee80211_sub_if_data { struct dentry *auth_transaction; struct dentry *flags; struct dentry *num_beacons_sta; + struct dentry *force_unicast_rateidx; + struct dentry *max_ratectrl_rateidx; } sta; struct { struct dentry *drop_unencrypted; @@ -483,9 +485,13 @@ struct ieee80211_sub_if_data { struct { struct dentry *drop_unencrypted; struct dentry *peer; + struct dentry *force_unicast_rateidx; + struct dentry *max_ratectrl_rateidx; } wds; struct { struct dentry *drop_unencrypted; + struct dentry *force_unicast_rateidx; + struct dentry *max_ratectrl_rateidx; } vlan; struct { struct dentry *mode; -- cgit v1.2.3 From 8ab65b03b7893da4a49009e7e356e36e27b0c407 Mon Sep 17 00:00:00 2001 From: Jan-Espen Pettersen Date: Mon, 25 Aug 2008 20:29:22 +0200 Subject: mac80211: don't send empty extended rates IE The association request includes a list of supported data rates. 802.11b: 4 supported rates. 802.11g: 12 (8 + 4) supported rates. 802.11a: 8 supported rates. The rates tag of the assoc request has room for only 8 rates. In case of 802.11g an extended rate tag is appended. However in net/wireless/mlme.c an extended (empty) rate tag is also appended if the number of rates is exact 8. This empty (length=0) extended rates tag causes some APs to deny association with code 18 (unsupported rates). These APs include my ZyXEL G-570U, and according to Tomas Winkler som Cisco APs. 'If count == 8' has been used to check for the need for an extended rates tag. But count would also be equal to 8 if the for loop exited because of no more supported rates. Therefore a check for count being less than rates_len would seem more correct. Thanks to: * Dan Williams for newbie guidance * Tomas Winkler for confirming the problem Signed-off-by: Jan-Espen Pettersen Signed-off-by: John W. Linville --- net/mac80211/mlme.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 09a56e24b799..74777ade6b22 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -783,7 +783,7 @@ static void ieee80211_send_assoc(struct net_device *dev, } } - if (count == 8) { + if (rates_len > count) { pos = skb_put(skb, rates_len - count + 2); *pos++ = WLAN_EID_EXT_SUPP_RATES; *pos++ = rates_len - count; -- cgit v1.2.3 From 576fdeaef631976c3afcb1d78ef196051282611d Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Tue, 26 Aug 2008 20:33:34 -0400 Subject: mac80211: quiet chatty IBSS merge message It seems obvious that this #ifndef should be the opposite polarity... Signed-off-by: John W. Linville --- net/mac80211/mlme.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 74777ade6b22..9bb68c6a8f44 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -2838,7 +2838,7 @@ static void ieee80211_rx_bss_info(struct net_device *dev, jiffies); #endif /* CONFIG_MAC80211_IBSS_DEBUG */ if (beacon_timestamp > rx_timestamp) { -#ifndef CONFIG_MAC80211_IBSS_DEBUG +#ifdef CONFIG_MAC80211_IBSS_DEBUG printk(KERN_DEBUG "%s: beacon TSF higher than " "local TSF - IBSS merge with BSSID %s\n", dev->name, print_mac(mac, mgmt->bssid)); -- cgit v1.2.3 From 666d9bbedfff7c2c37eab92e715641922dee6864 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Wed, 27 Aug 2008 02:12:52 -0700 Subject: pkt_sched: Fix dev_graft_qdisc() locking During dev_graft_qdisc() dev is deactivated, so qdisc_root_lock() returns wrong lock of noop_qdisc instead of qdisc_sleeping. Signed-off-by: Jarek Poplawski Acked-by: Herbert Xu Signed-off-by: David S. Miller --- net/sched/sch_api.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index e7fb9e0d21b4..341d558b6e39 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -624,7 +624,7 @@ static struct Qdisc *dev_graft_qdisc(struct netdev_queue *dev_queue, struct Qdisc *oqdisc = dev_queue->qdisc_sleeping; spinlock_t *root_lock; - root_lock = qdisc_root_lock(oqdisc); + root_lock = qdisc_lock(oqdisc); spin_lock_bh(root_lock); /* Prune old scheduler */ -- cgit v1.2.3 From f7a54c13c7b072d9426bd5cec1cdb8306df5ef55 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Wed, 27 Aug 2008 02:22:07 -0700 Subject: pkt_sched: Use rcu_assign_pointer() to change dev_queue->qdisc These pointers are RCU protected, so proper primitives should be used. Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/sched/sch_api.c | 2 +- net/sched/sch_generic.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 341d558b6e39..ad9cda1b8c0a 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -635,7 +635,7 @@ static struct Qdisc *dev_graft_qdisc(struct netdev_queue *dev_queue, if (qdisc == NULL) qdisc = &noop_qdisc; dev_queue->qdisc_sleeping = qdisc; - dev_queue->qdisc = &noop_qdisc; + rcu_assign_pointer(dev_queue->qdisc, &noop_qdisc); spin_unlock_bh(root_lock); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 5f0ade7806a7..9634091ee2f0 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -634,7 +634,7 @@ static void dev_deactivate_queue(struct net_device *dev, if (!(qdisc->flags & TCQ_F_BUILTIN)) set_bit(__QDISC_STATE_DEACTIVATED, &qdisc->state); - dev_queue->qdisc = qdisc_default; + rcu_assign_pointer(dev_queue->qdisc, qdisc_default); qdisc_reset(qdisc); spin_unlock_bh(qdisc_lock(qdisc)); @@ -709,7 +709,7 @@ static void shutdown_scheduler_queue(struct net_device *dev, struct Qdisc *qdisc_default = _qdisc_default; if (qdisc) { - dev_queue->qdisc = qdisc_default; + rcu_assign_pointer(dev_queue->qdisc, qdisc_default); dev_queue->qdisc_sleeping = qdisc_default; qdisc_destroy(qdisc); -- cgit v1.2.3 From f6f9b93f1624206c802ac9162c9302edaf59bfd9 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Wed, 27 Aug 2008 02:25:17 -0700 Subject: pkt_sched: Fix gen_estimator locks While passing a qdisc root lock to gen_new_estimator() and gen_replace_estimator() dev could be deactivated or even before grafting proper root qdisc as qdisc_sleeping (e.g. qdisc_create), so using qdisc_root_lock() is not enough. This patch adds qdisc_root_sleeping_lock() for this, plus additional checks, where necessary. Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- include/net/sch_generic.h | 8 ++++++++ net/sched/sch_api.c | 14 +++++++++++--- net/sched/sch_cbq.c | 4 ++-- net/sched/sch_hfsc.c | 4 ++-- net/sched/sch_htb.c | 4 ++-- 5 files changed, 25 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index b1d2cfea89c5..ef8a7e2e12e7 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -217,6 +217,14 @@ static inline spinlock_t *qdisc_root_lock(struct Qdisc *qdisc) return qdisc_lock(root); } +static inline spinlock_t *qdisc_root_sleeping_lock(struct Qdisc *qdisc) +{ + struct Qdisc *root = qdisc_root_sleeping(qdisc); + + ASSERT_RTNL(); + return qdisc_lock(root); +} + static inline struct net_device *qdisc_dev(struct Qdisc *qdisc) { return qdisc->dev_queue->dev; diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index ad9cda1b8c0a..506b709510b6 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -830,9 +830,16 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, sch->stab = stab; } if (tca[TCA_RATE]) { + spinlock_t *root_lock; + + if ((sch->parent != TC_H_ROOT) && + !(sch->flags & TCQ_F_INGRESS)) + root_lock = qdisc_root_sleeping_lock(sch); + else + root_lock = qdisc_lock(sch); + err = gen_new_estimator(&sch->bstats, &sch->rate_est, - qdisc_root_lock(sch), - tca[TCA_RATE]); + root_lock, tca[TCA_RATE]); if (err) { /* * Any broken qdiscs that would require @@ -884,7 +891,8 @@ static int qdisc_change(struct Qdisc *sch, struct nlattr **tca) if (tca[TCA_RATE]) gen_replace_estimator(&sch->bstats, &sch->rate_est, - qdisc_root_lock(sch), tca[TCA_RATE]); + qdisc_root_sleeping_lock(sch), + tca[TCA_RATE]); return 0; } diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 8fa90d68ec6d..9b720adedead 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -1839,7 +1839,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t if (tca[TCA_RATE]) gen_replace_estimator(&cl->bstats, &cl->rate_est, - qdisc_root_lock(sch), + qdisc_root_sleeping_lock(sch), tca[TCA_RATE]); return 0; } @@ -1930,7 +1930,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t if (tca[TCA_RATE]) gen_new_estimator(&cl->bstats, &cl->rate_est, - qdisc_root_lock(sch), tca[TCA_RATE]); + qdisc_root_sleeping_lock(sch), tca[TCA_RATE]); *arg = (unsigned long)cl; return 0; diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index c2b8d9cce3d2..c1e77da8cd09 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1045,7 +1045,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (tca[TCA_RATE]) gen_replace_estimator(&cl->bstats, &cl->rate_est, - qdisc_root_lock(sch), + qdisc_root_sleeping_lock(sch), tca[TCA_RATE]); return 0; } @@ -1104,7 +1104,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (tca[TCA_RATE]) gen_new_estimator(&cl->bstats, &cl->rate_est, - qdisc_root_lock(sch), tca[TCA_RATE]); + qdisc_root_sleeping_lock(sch), tca[TCA_RATE]); *arg = (unsigned long)cl; return 0; } diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 0df0df202ed0..97d4761cc31e 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1372,7 +1372,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, goto failure; gen_new_estimator(&cl->bstats, &cl->rate_est, - qdisc_root_lock(sch), + qdisc_root_sleeping_lock(sch), tca[TCA_RATE] ? : &est.nla); cl->refcnt = 1; cl->children = 0; @@ -1427,7 +1427,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, } else { if (tca[TCA_RATE]) gen_replace_estimator(&cl->bstats, &cl->rate_est, - qdisc_root_lock(sch), + qdisc_root_sleeping_lock(sch), tca[TCA_RATE]); sch_tree_lock(sch); } -- cgit v1.2.3 From 7982d5e1b350acb96aa156916c44c25ef87bb809 Mon Sep 17 00:00:00 2001 From: Philip Love Date: Wed, 27 Aug 2008 02:33:50 -0700 Subject: tcp: fix tcp header size miscalculation when window scale is unused The size of the TCP header is miscalculated when the window scale ends up being 0. Additionally, this can be induced by sending a SYN to a passive open port with a window scale option with value 0. Signed-off-by: Philip Love Signed-off-by: Adam Langley Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index a00532de2a8c..8165f5aa8c71 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -468,7 +468,8 @@ static unsigned tcp_syn_options(struct sock *sk, struct sk_buff *skb, } if (likely(sysctl_tcp_window_scaling)) { opts->ws = tp->rx_opt.rcv_wscale; - size += TCPOLEN_WSCALE_ALIGNED; + if(likely(opts->ws)) + size += TCPOLEN_WSCALE_ALIGNED; } if (likely(sysctl_tcp_sack)) { opts->options |= OPTION_SACK_ADVERTISE; @@ -509,7 +510,8 @@ static unsigned tcp_synack_options(struct sock *sk, if (likely(ireq->wscale_ok)) { opts->ws = ireq->rcv_wscale; - size += TCPOLEN_WSCALE_ALIGNED; + if(likely(opts->ws)) + size += TCPOLEN_WSCALE_ALIGNED; } if (likely(doing_ts)) { opts->options |= OPTION_TS; -- cgit v1.2.3 From d994af0d50efc96b2077978fe9f066992639d525 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 27 Aug 2008 02:35:18 -0700 Subject: ipv4: mode 0555 in ipv4_skeleton vpnc on today's kernel says Cannot open "/proc/sys/net/ipv4/route/flush": d--------- 0 root root 0 2008-08-26 11:32 /proc/sys/net/ipv4/route d--------- 0 root root 0 2008-08-26 19:16 /proc/sys/net/ipv4/neigh Signed-off-by: Hugh Dickins Acked-by: Al Viro Signed-off-by: David S. Miller --- net/ipv4/route.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/route.c b/net/ipv4/route.c index e91bafeb32f4..6ee5354c9aa1 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -3121,9 +3121,9 @@ static struct ctl_table empty[1]; static struct ctl_table ipv4_skeleton[] = { { .procname = "route", .ctl_name = NET_IPV4_ROUTE, - .child = ipv4_route_table}, + .mode = 0555, .child = ipv4_route_table}, { .procname = "neigh", .ctl_name = NET_IPV4_NEIGH, - .child = empty}, + .mode = 0555, .child = empty}, { } }; -- cgit v1.2.3 From 328fc47ea0bcc27d9afa69c3ad6e52431cadd76c Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Wed, 27 Aug 2008 16:08:54 -0700 Subject: sctp: correct bounds check in sctp_setsockopt_auth_key The bonds check to prevent buffer overlflow was not exactly right. It still allowed overflow of up to 8 bytes which is sizeof(struct sctp_authkey). Since optlen is already checked against the size of that struct, we are guaranteed not to cause interger overflow either. Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/sctp/socket.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sctp/socket.c b/net/sctp/socket.c index afa952e726d7..9b9b2c31dd15 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3144,7 +3144,7 @@ static int sctp_setsockopt_auth_key(struct sock *sk, goto out; } - if (authkey->sca_keylength > optlen) { + if (authkey->sca_keylength > optlen - sizeof(struct sctp_authkey)) { ret = -EINVAL; goto out; } -- cgit v1.2.3 From d97240552cd98c4b07322f30f66fd9c3ba4171de Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Wed, 27 Aug 2008 16:09:49 -0700 Subject: sctp: fix random memory dereference with SCTP_HMAC_IDENT option. The number of identifiers needs to be checked against the option length. Also, the identifier index provided needs to be verified to make sure that it doesn't exceed the bounds of the array. Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller --- net/sctp/auth.c | 3 +++ net/sctp/socket.c | 6 ++++-- 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sctp/auth.c b/net/sctp/auth.c index 1fcb4cf2f4c9..52db5f60daa0 100644 --- a/net/sctp/auth.c +++ b/net/sctp/auth.c @@ -786,6 +786,9 @@ int sctp_auth_ep_set_hmacs(struct sctp_endpoint *ep, for (i = 0; i < hmacs->shmac_num_idents; i++) { id = hmacs->shmac_idents[i]; + if (id > SCTP_AUTH_HMAC_ID_MAX) + return -EOPNOTSUPP; + if (SCTP_AUTH_HMAC_ID_SHA1 == id) has_sha1 = 1; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 9b9b2c31dd15..5ffb9dec1c3f 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3086,6 +3086,7 @@ static int sctp_setsockopt_hmac_ident(struct sock *sk, int optlen) { struct sctp_hmacalgo *hmacs; + u32 idents; int err; if (!sctp_auth_enable) @@ -3103,8 +3104,9 @@ static int sctp_setsockopt_hmac_ident(struct sock *sk, goto out; } - if (hmacs->shmac_num_idents == 0 || - hmacs->shmac_num_idents > SCTP_AUTH_NUM_HMACS) { + idents = hmacs->shmac_num_idents; + if (idents == 0 || idents > SCTP_AUTH_NUM_HMACS || + (idents * sizeof(u16)) > (optlen - sizeof(struct sctp_hmacalgo))) { err = -EINVAL; goto out; } -- cgit v1.2.3 From 3cc76caa98b092a8fb3e7b4303c70f847db0651f Mon Sep 17 00:00:00 2001 From: Yang Hongyang Date: Fri, 29 Aug 2008 14:06:51 -0700 Subject: ipv6: When we droped a packet, we should return NET_RX_DROP instead of 0 Signed-off-by: Yang Hongyang Signed-off-by: David S. Miller --- net/ipv6/raw.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 01d47674f7e5..e53e493606c5 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -377,14 +377,14 @@ static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb) skb_checksum_complete(skb)) { atomic_inc(&sk->sk_drops); kfree_skb(skb); - return 0; + return NET_RX_DROP; } /* Charge it to the socket. */ if (sock_queue_rcv_skb(sk,skb)<0) { atomic_inc(&sk->sk_drops); kfree_skb(skb); - return 0; + return NET_RX_DROP; } return 0; @@ -429,7 +429,7 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb) if (skb_checksum_complete(skb)) { atomic_inc(&sk->sk_drops); kfree_skb(skb); - return 0; + return NET_RX_DROP; } } -- cgit v1.2.3 From 102396ae65108b026e4e1868e30fa013f45a169e Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Fri, 29 Aug 2008 14:21:52 -0700 Subject: pkt_sched: Fix locking of qdisc_root with qdisc_root_sleeping_lock() Use qdisc_root_sleeping_lock() instead of qdisc_root_lock() where appropriate. The only difference is while dev is deactivated, when currently we can use a sleeping qdisc with the lock of noop_qdisc. This shouldn't be dangerous since after deactivation root lock could be used only by gen_estimator code, but looks wrong anyway. Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/sched/cls_api.c | 2 +- net/sched/cls_route.c | 2 +- net/sched/sch_api.c | 8 ++++---- net/sched/sch_cbq.c | 2 +- net/sched/sch_htb.c | 4 ++-- net/sched/sch_netem.c | 2 +- net/sched/sch_teql.c | 2 +- 7 files changed, 11 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 5cafdd4c8018..8eb79e92e94c 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -205,7 +205,7 @@ replay: } } - root_lock = qdisc_root_lock(q); + root_lock = qdisc_root_sleeping_lock(q); if (tp == NULL) { /* Proto-tcf does not exist, create new one */ diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c index 481260a4f10f..e3d8455eebc2 100644 --- a/net/sched/cls_route.c +++ b/net/sched/cls_route.c @@ -75,7 +75,7 @@ static __inline__ int route4_fastmap_hash(u32 id, int iif) static inline void route4_reset_fastmap(struct Qdisc *q, struct route4_head *head, u32 id) { - spinlock_t *root_lock = qdisc_root_lock(q); + spinlock_t *root_lock = qdisc_root_sleeping_lock(q); spin_lock_bh(root_lock); memset(head->fastmap, 0, sizeof(head->fastmap)); diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 506b709510b6..1122c952aa99 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1169,8 +1169,8 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid, if (q->stab && qdisc_dump_stab(skb, q->stab) < 0) goto nla_put_failure; - if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, - TCA_XSTATS, qdisc_root_lock(q), &d) < 0) + if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, + qdisc_root_sleeping_lock(q), &d) < 0) goto nla_put_failure; if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0) @@ -1461,8 +1461,8 @@ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q, if (cl_ops->dump && cl_ops->dump(q, cl, skb, tcm) < 0) goto nla_put_failure; - if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, - TCA_XSTATS, qdisc_root_lock(q), &d) < 0) + if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, + qdisc_root_sleeping_lock(q), &d) < 0) goto nla_put_failure; if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0) diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 9b720adedead..8b06fa900482 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -1754,7 +1754,7 @@ static void cbq_put(struct Qdisc *sch, unsigned long arg) if (--cl->refcnt == 0) { #ifdef CONFIG_NET_CLS_ACT - spinlock_t *root_lock = qdisc_root_lock(sch); + spinlock_t *root_lock = qdisc_root_sleeping_lock(sch); struct cbq_sched_data *q = qdisc_priv(sch); spin_lock_bh(root_lock); diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 97d4761cc31e..d14f02056ae6 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1043,7 +1043,7 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt) static int htb_dump(struct Qdisc *sch, struct sk_buff *skb) { - spinlock_t *root_lock = qdisc_root_lock(sch); + spinlock_t *root_lock = qdisc_root_sleeping_lock(sch); struct htb_sched *q = qdisc_priv(sch); struct nlattr *nest; struct tc_htb_glob gopt; @@ -1075,7 +1075,7 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg, struct sk_buff *skb, struct tcmsg *tcm) { struct htb_class *cl = (struct htb_class *)arg; - spinlock_t *root_lock = qdisc_root_lock(sch); + spinlock_t *root_lock = qdisc_root_sleeping_lock(sch); struct nlattr *nest; struct tc_htb_opt opt; diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index fb0294d0b55e..3781e55046d0 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -341,7 +341,7 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr) for (i = 0; i < n; i++) d->table[i] = data[i]; - root_lock = qdisc_root_lock(sch); + root_lock = qdisc_root_sleeping_lock(sch); spin_lock_bh(root_lock); d = xchg(&q->delay_dist, d); diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index 2c35c678563b..d35ef059abb1 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -161,7 +161,7 @@ teql_destroy(struct Qdisc* sch) txq = netdev_get_tx_queue(master->dev, 0); master->slaves = NULL; - root_lock = qdisc_root_lock(txq->qdisc); + root_lock = qdisc_root_sleeping_lock(txq->qdisc); spin_lock_bh(root_lock); qdisc_reset(txq->qdisc); spin_unlock_bh(root_lock); -- cgit v1.2.3 From 27df6f25ff218072e0e879a96beeb398a79cdbc8 Mon Sep 17 00:00:00 2001 From: Cyrill Gorcunov Date: Sun, 31 Aug 2008 19:25:49 +0400 Subject: sunrpc: fix possible overrun on read of /proc/sys/sunrpc/transports Vegard Nossum reported ---------------------- > I noticed that something weird is going on with /proc/sys/sunrpc/transports. > This file is generated in net/sunrpc/sysctl.c, function proc_do_xprt(). When > I "cat" this file, I get the expected output: > $ cat /proc/sys/sunrpc/transports > tcp 1048576 > udp 32768 > But I think that it does not check the length of the buffer supplied by > userspace to read(). With my original program, I found that the stack was > being overwritten by the characters above, even when the length given to > read() was just 1. David Wagner added (among other things) that copy_to_user could be probably used here. Ingo Oeser suggested to use simple_read_from_buffer() here. The conclusion is that proc_do_xprt doesn't check for userside buffer size indeed so fix this by using Ingo's suggestion. Reported-by: Vegard Nossum Signed-off-by: Cyrill Gorcunov CC: Ingo Oeser Cc: Neil Brown Cc: Chuck Lever Cc: Greg Banks Cc: Tom Tucker Signed-off-by: J. Bruce Fields --- net/sunrpc/sysctl.c | 18 ++++-------------- 1 file changed, 4 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/sunrpc/sysctl.c b/net/sunrpc/sysctl.c index 0f8c439b848a..5231f7aaac0e 100644 --- a/net/sunrpc/sysctl.c +++ b/net/sunrpc/sysctl.c @@ -60,24 +60,14 @@ static int proc_do_xprt(ctl_table *table, int write, struct file *file, void __user *buffer, size_t *lenp, loff_t *ppos) { char tmpbuf[256]; - int len; + size_t len; + if ((*ppos && !write) || !*lenp) { *lenp = 0; return 0; } - if (write) - return -EINVAL; - else { - len = svc_print_xprts(tmpbuf, sizeof(tmpbuf)); - if (!access_ok(VERIFY_WRITE, buffer, len)) - return -EFAULT; - - if (__copy_to_user(buffer, tmpbuf, len)) - return -EFAULT; - } - *lenp -= len; - *ppos += len; - return 0; + len = svc_print_xprts(tmpbuf, sizeof(tmpbuf)); + return simple_read_from_buffer(buffer, *lenp, ppos, tmpbuf, len); } static int -- cgit v1.2.3 From d9664741e0e2216770d6e52646474d3982b8eb55 Mon Sep 17 00:00:00 2001 From: Florian Mickler Date: Tue, 2 Sep 2008 15:26:34 +0200 Subject: net/wireless/Kconfig: clarify the description for CONFIG_WIRELESS_EXT_SYSFS Current setup with hal and NetworkManager will fail to work without newest hal version with this config option disabled. Although this will solve itself by time, at the moment it is dishonest to say that we don't know any software that uses it, if there are many many people relying on old hal versions. Signed-off-by: Florian Mickler Signed-off-by: John W. Linville --- net/wireless/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/wireless/Kconfig b/net/wireless/Kconfig index ab015c62d561..833b024f8f66 100644 --- a/net/wireless/Kconfig +++ b/net/wireless/Kconfig @@ -39,4 +39,5 @@ config WIRELESS_EXT_SYSFS files in /sys/class/net/*/wireless/. The same information is available via the ioctls as well. - Say Y if you have programs using it (we don't know of any). + Say Y if you have programs using it, like old versions of + hal. -- cgit v1.2.3 From 2b58b209399844995ad48e421267e359e16c03db Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Thu, 28 Aug 2008 15:12:06 +0300 Subject: mac80211: Fix debugfs union misuse and pointer corruption debugfs union in struct ieee80211_sub_if_data is misused by including a common default_key dentry as a union member. This ends occupying the same memory area with the first dentry in other union members (structures; usually drop_unencrypted). Consequently, debugfs operations on default_key symlinks and drop_unencrypted entry are using the same dentry pointer even though they are supposed to be separate ones. This can lead to removing entries incorrectly or potentially leaving something behind since one of the dentry pointers gets lost. Fix this by moving the default_key dentry to a new struct (common_debugfs) that contains dentries (more to be added in future) that are shared by all vif types. The debugfs union must only be used for vif type-specific entries to avoid this type of pointer corruption. Signed-off-by: Jouni Malinen Acked-by: Johannes Berg Signed-off-by: John W. Linville --- net/mac80211/debugfs_key.c | 6 +++--- net/mac80211/ieee80211_i.h | 4 +++- 2 files changed, 6 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/mac80211/debugfs_key.c b/net/mac80211/debugfs_key.c index 7439b63df5d0..cf82acec913a 100644 --- a/net/mac80211/debugfs_key.c +++ b/net/mac80211/debugfs_key.c @@ -265,7 +265,7 @@ void ieee80211_debugfs_key_add_default(struct ieee80211_sub_if_data *sdata) key = sdata->default_key; if (key) { sprintf(buf, "../keys/%d", key->debugfs.cnt); - sdata->debugfs.default_key = + sdata->common_debugfs.default_key = debugfs_create_symlink("default_key", sdata->debugfsdir, buf); } else @@ -277,8 +277,8 @@ void ieee80211_debugfs_key_remove_default(struct ieee80211_sub_if_data *sdata) if (!sdata) return; - debugfs_remove(sdata->debugfs.default_key); - sdata->debugfs.default_key = NULL; + debugfs_remove(sdata->common_debugfs.default_key); + sdata->common_debugfs.default_key = NULL; } void ieee80211_debugfs_key_sta_del(struct ieee80211_key *key, diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 586a9b49b0fc..4498d8713652 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -496,8 +496,10 @@ struct ieee80211_sub_if_data { struct { struct dentry *mode; } monitor; - struct dentry *default_key; } debugfs; + struct { + struct dentry *default_key; + } common_debugfs; #ifdef CONFIG_MAC80211_MESH struct dentry *mesh_stats_dir; -- cgit v1.2.3 From 9d7d74029e0f5fde3b88b39892b9b9cfdf4ea10a Mon Sep 17 00:00:00 2001 From: Julien Brunel Date: Tue, 2 Sep 2008 17:24:28 -0700 Subject: net/xfrm: Use an IS_ERR test rather than a NULL test In case of error, the function xfrm_bundle_create returns an ERR pointer, but never returns a NULL pointer. So a NULL test that comes after an IS_ERR test should be deleted. The semantic match that finds this problem is as follows: (http://www.emn.fr/x-info/coccinelle/) // @match_bad_null_test@ expression x, E; statement S1,S2; @@ x = xfrm_bundle_create(...) ... when != x = E * if (x != NULL) S1 else S2 // Signed-off-by: Julien Brunel Signed-off-by: Julia Lawall Signed-off-by: David S. Miller --- net/xfrm/xfrm_policy.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 841b32a2e680..46914b79d850 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -1731,8 +1731,7 @@ restart: * We can't enlist stable bundles either. */ write_unlock_bh(&policy->lock); - if (dst) - dst_free(dst); + dst_free(dst); if (pol_dead) XFRM_INC_STATS(LINUX_MIB_XFRMOUTPOLDEAD); @@ -1748,8 +1747,7 @@ restart: err = xfrm_dst_update_origin(dst, fl); if (unlikely(err)) { write_unlock_bh(&policy->lock); - if (dst) - dst_free(dst); + dst_free(dst); XFRM_INC_STATS(LINUX_MIB_XFRMOUTBUNDLECHECKERROR); goto error; } -- cgit v1.2.3 From 06770843c2f0f929a6e0c758dc433902a01aabfb Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 2 Sep 2008 17:28:58 -0700 Subject: ipv: Re-enable IP when MTU > 68 Re-enable IP when the MTU gets back to a valid size. This patch just checks if the in_dev is NULL on a NETDEV_CHANGEMTU event and if MTU is valid (bigger than 68), then re-enable in_dev. Also a function that checks valid MTU size was created. Signed-off-by: Breno Leitao Signed-off-by: David S. Miller --- net/ipv4/devinet.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 91d3d96805d0..b12dae2b0b2d 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1029,6 +1029,11 @@ skip: } } +static inline bool inetdev_valid_mtu(unsigned mtu) +{ + return mtu >= 68; +} + /* Called only under RTNL semaphore */ static int inetdev_event(struct notifier_block *this, unsigned long event, @@ -1048,6 +1053,10 @@ static int inetdev_event(struct notifier_block *this, unsigned long event, IN_DEV_CONF_SET(in_dev, NOXFRM, 1); IN_DEV_CONF_SET(in_dev, NOPOLICY, 1); } + } else if (event == NETDEV_CHANGEMTU) { + /* Re-enabling IP */ + if (inetdev_valid_mtu(dev->mtu)) + in_dev = inetdev_init(dev); } goto out; } @@ -1058,7 +1067,7 @@ static int inetdev_event(struct notifier_block *this, unsigned long event, dev->ip_ptr = NULL; break; case NETDEV_UP: - if (dev->mtu < 68) + if (!inetdev_valid_mtu(dev->mtu)) break; if (dev->flags & IFF_LOOPBACK) { struct in_ifaddr *ifa; @@ -1080,9 +1089,9 @@ static int inetdev_event(struct notifier_block *this, unsigned long event, ip_mc_down(in_dev); break; case NETDEV_CHANGEMTU: - if (dev->mtu >= 68) + if (inetdev_valid_mtu(dev->mtu)) break; - /* MTU falled under 68, disable IP */ + /* disable IP when MTU is not enough */ case NETDEV_UNREGISTER: inetdev_destroy(in_dev); break; -- cgit v1.2.3 From 37b08e34a98c664bea86e3fae718ac45a46b7276 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 2 Sep 2008 20:14:15 -0700 Subject: ipsec: Fix deadlock in xfrm_state management. Ever since commit 4c563f7669c10a12354b72b518c2287ffc6ebfb3 ("[XFRM]: Speed up xfrm_policy and xfrm_state walking") it is illegal to call __xfrm_state_destroy (and thus xfrm_state_put()) with xfrm_state_lock held. If we do, we'll deadlock since we have the lock already and __xfrm_state_destroy() tries to take it again. Fix this by pushing the xfrm_state_put() calls after the lock is dropped. Signed-off-by: David S. Miller --- net/xfrm/xfrm_state.c | 32 +++++++++++++++++++++++--------- 1 file changed, 23 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 4c6914ef7d92..7bd62f61593f 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -780,11 +780,13 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, { unsigned int h; struct hlist_node *entry; - struct xfrm_state *x, *x0; + struct xfrm_state *x, *x0, *to_put; int acquire_in_progress = 0; int error = 0; struct xfrm_state *best = NULL; + to_put = NULL; + spin_lock_bh(&xfrm_state_lock); h = xfrm_dst_hash(daddr, saddr, tmpl->reqid, family); hlist_for_each_entry(x, entry, xfrm_state_bydst+h, bydst) { @@ -833,7 +835,7 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, if (tmpl->id.spi && (x0 = __xfrm_state_lookup(daddr, tmpl->id.spi, tmpl->id.proto, family)) != NULL) { - xfrm_state_put(x0); + to_put = x0; error = -EEXIST; goto out; } @@ -849,7 +851,7 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, error = security_xfrm_state_alloc_acquire(x, pol->security, fl->secid); if (error) { x->km.state = XFRM_STATE_DEAD; - xfrm_state_put(x); + to_put = x; x = NULL; goto out; } @@ -870,7 +872,7 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, xfrm_hash_grow_check(x->bydst.next != NULL); } else { x->km.state = XFRM_STATE_DEAD; - xfrm_state_put(x); + to_put = x; x = NULL; error = -ESRCH; } @@ -881,6 +883,8 @@ out: else *err = acquire_in_progress ? -EAGAIN : error; spin_unlock_bh(&xfrm_state_lock); + if (to_put) + xfrm_state_put(to_put); return x; } @@ -1067,18 +1071,20 @@ static struct xfrm_state *__xfrm_find_acq_byseq(u32 seq); int xfrm_state_add(struct xfrm_state *x) { - struct xfrm_state *x1; + struct xfrm_state *x1, *to_put; int family; int err; int use_spi = xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY); family = x->props.family; + to_put = NULL; + spin_lock_bh(&xfrm_state_lock); x1 = __xfrm_state_locate(x, use_spi, family); if (x1) { - xfrm_state_put(x1); + to_put = x1; x1 = NULL; err = -EEXIST; goto out; @@ -1088,7 +1094,7 @@ int xfrm_state_add(struct xfrm_state *x) x1 = __xfrm_find_acq_byseq(x->km.seq); if (x1 && ((x1->id.proto != x->id.proto) || xfrm_addr_cmp(&x1->id.daddr, &x->id.daddr, family))) { - xfrm_state_put(x1); + to_put = x1; x1 = NULL; } } @@ -1110,6 +1116,9 @@ out: xfrm_state_put(x1); } + if (to_put) + xfrm_state_put(to_put); + return err; } EXPORT_SYMBOL(xfrm_state_add); @@ -1269,10 +1278,12 @@ EXPORT_SYMBOL(xfrm_state_migrate); int xfrm_state_update(struct xfrm_state *x) { - struct xfrm_state *x1; + struct xfrm_state *x1, *to_put; int err; int use_spi = xfrm_id_proto_match(x->id.proto, IPSEC_PROTO_ANY); + to_put = NULL; + spin_lock_bh(&xfrm_state_lock); x1 = __xfrm_state_locate(x, use_spi, x->props.family); @@ -1281,7 +1292,7 @@ int xfrm_state_update(struct xfrm_state *x) goto out; if (xfrm_state_kern(x1)) { - xfrm_state_put(x1); + to_put = x1; err = -EEXIST; goto out; } @@ -1295,6 +1306,9 @@ int xfrm_state_update(struct xfrm_state *x) out: spin_unlock_bh(&xfrm_state_lock); + if (to_put) + xfrm_state_put(to_put); + if (err) return err; -- cgit v1.2.3 From 6c00055a819ce8a6e2c3af2f65d4ea1a8559c491 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 2 Sep 2008 23:38:32 -0700 Subject: tipc: Don't use structure names which easily globally conflict. Andrew Morton reported a build failure on sparc32, because TIPC uses names like "struct node" and there is a like named data structure defined in linux/node.h This just regexp replaces "struct node*" to "struct tipc_node*" to avoid this and any future similar problems. Signed-off-by: David S. Miller --- net/tipc/bcast.c | 22 +++++++++--------- net/tipc/bcast.h | 22 +++++++++--------- net/tipc/bearer.c | 2 +- net/tipc/bearer.h | 2 +- net/tipc/cluster.c | 16 +++++++------- net/tipc/cluster.h | 10 ++++----- net/tipc/discover.c | 2 +- net/tipc/link.c | 26 +++++++++++----------- net/tipc/link.h | 2 +- net/tipc/name_table.h | 2 +- net/tipc/net.c | 2 +- net/tipc/net.h | 2 +- net/tipc/node.c | 60 +++++++++++++++++++++++++------------------------- net/tipc/node.h | 42 +++++++++++++++++------------------ net/tipc/node_subscr.c | 4 ++-- net/tipc/node_subscr.h | 10 ++++----- net/tipc/port.h | 2 +- net/tipc/zone.c | 4 ++-- net/tipc/zone.h | 2 +- 19 files changed, 117 insertions(+), 117 deletions(-) (limited to 'net') diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index b1ff16aa4bdb..3ddaff42d1bb 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -96,8 +96,8 @@ struct bcbearer { 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_node_map remains; + struct tipc_node_map remains_new; }; /** @@ -110,7 +110,7 @@ struct bcbearer { struct bclink { struct link link; - struct node node; + struct tipc_node node; }; @@ -149,7 +149,7 @@ static void bcbuf_decr_acks(struct sk_buff *buf) * Called with 'node' locked, bc_lock unlocked */ -static void bclink_set_gap(struct node *n_ptr) +static void bclink_set_gap(struct tipc_node *n_ptr) { struct sk_buff *buf = n_ptr->bclink.deferred_head; @@ -202,7 +202,7 @@ static void bclink_retransmit_pkt(u32 after, u32 to) * Node is locked, bc_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; @@ -250,7 +250,7 @@ void tipc_bclink_acknowledge(struct node *n_ptr, u32 acked) * tipc_net_lock and node lock set */ -static void bclink_send_ack(struct node *n_ptr) +static void bclink_send_ack(struct tipc_node *n_ptr) { struct link *l_ptr = n_ptr->active_links[n_ptr->addr & 1]; @@ -264,7 +264,7 @@ static void bclink_send_ack(struct node *n_ptr) * tipc_net_lock and node lock set */ -static void bclink_send_nack(struct node *n_ptr) +static void bclink_send_nack(struct tipc_node *n_ptr) { struct sk_buff *buf; struct tipc_msg *msg; @@ -308,7 +308,7 @@ static void bclink_send_nack(struct node *n_ptr) * tipc_net_lock and node lock set */ -void tipc_bclink_check_gap(struct node *n_ptr, u32 last_sent) +void tipc_bclink_check_gap(struct tipc_node *n_ptr, u32 last_sent) { if (!n_ptr->bclink.supported || less_eq(last_sent, mod(n_ptr->bclink.last_in))) @@ -328,7 +328,7 @@ void tipc_bclink_check_gap(struct node *n_ptr, u32 last_sent) static void tipc_bclink_peek_nack(u32 dest, u32 sender_tag, u32 gap_after, u32 gap_to) { - struct node *n_ptr = tipc_node_find(dest); + struct tipc_node *n_ptr = tipc_node_find(dest); u32 my_after, my_to; if (unlikely(!n_ptr || !tipc_node_is_up(n_ptr))) @@ -418,7 +418,7 @@ void tipc_bclink_recv_pkt(struct sk_buff *buf) 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 = tipc_node_find(msg_prevnode(msg)); u32 next_in; u32 seqno; struct sk_buff *deferred; @@ -538,7 +538,7 @@ u32 tipc_bclink_get_last_sent(void) return last_sent; } -u32 tipc_bclink_acks_missing(struct node *n_ptr) +u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr) { return (n_ptr->bclink.supported && (tipc_bclink_get_last_sent() != n_ptr->bclink.acked)); diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h index a2416fa6b906..5aa024b99c55 100644 --- a/net/tipc/bcast.h +++ b/net/tipc/bcast.h @@ -41,12 +41,12 @@ #define WSIZE 32 /** - * 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]; }; @@ -68,7 +68,7 @@ struct port_list { }; -struct node; +struct tipc_node; extern char tipc_bclink_name[]; @@ -77,7 +77,7 @@ extern char tipc_bclink_name[]; * nmap_add - add a node to a node map */ -static inline void tipc_nmap_add(struct node_map *nm_ptr, u32 node) +static inline void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node) { int n = tipc_node(node); int w = n / WSIZE; @@ -93,7 +93,7 @@ static inline void tipc_nmap_add(struct node_map *nm_ptr, u32 node) * nmap_remove - remove a node from a node map */ -static inline void tipc_nmap_remove(struct node_map *nm_ptr, u32 node) +static inline void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node) { int n = tipc_node(node); int w = n / WSIZE; @@ -109,7 +109,7 @@ static inline void tipc_nmap_remove(struct node_map *nm_ptr, u32 node) * 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)); } @@ -121,8 +121,8 @@ static inline int tipc_nmap_equal(struct node_map *nm_a, struct node_map *nm_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) +static inline void tipc_nmap_diff(struct tipc_node_map *nm_a, struct tipc_node_map *nm_b, + struct tipc_node_map *nm_diff) { int stop = sizeof(nm_a->map) / sizeof(u32); int w; @@ -195,12 +195,12 @@ static inline void tipc_port_list_free(struct port_list *pl_ptr) int tipc_bclink_init(void); void tipc_bclink_stop(void); -void tipc_bclink_acknowledge(struct node *n_ptr, u32 acked); +void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked); int tipc_bclink_send_msg(struct sk_buff *buf); void tipc_bclink_recv_pkt(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_check_gap(struct tipc_node *n_ptr, u32 seqno); 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); diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 6a9aba3edd08..a7a36779b9b3 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -599,7 +599,7 @@ int tipc_block_bearer(const char *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; + struct tipc_node *n_ptr = l_ptr->owner; spin_lock_bh(&n_ptr->lock); tipc_link_reset(l_ptr); diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index 6a36b6600e6c..ca5734892713 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -104,7 +104,7 @@ struct bearer { u32 continue_count; int active; char net_plane; - struct node_map nodes; + struct tipc_node_map nodes; }; struct bearer_name { diff --git a/net/tipc/cluster.c b/net/tipc/cluster.c index 46ee6c58532d..689fdefe9d04 100644 --- a/net/tipc/cluster.c +++ b/net/tipc/cluster.c @@ -48,8 +48,8 @@ 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,}}; +struct tipc_node **tipc_local_nodes = NULL; +struct tipc_node_map tipc_cltr_bcast_nodes = {0,{0,}}; u32 tipc_highest_allowed_slave = 0; struct cluster *tipc_cltr_create(u32 addr) @@ -115,7 +115,7 @@ void tipc_cltr_delete(struct cluster *c_ptr) u32 tipc_cltr_next_node(struct cluster *c_ptr, u32 addr) { - struct node *n_ptr; + struct tipc_node *n_ptr; u32 n_num = tipc_node(addr) + 1; if (!c_ptr) @@ -133,7 +133,7 @@ u32 tipc_cltr_next_node(struct cluster *c_ptr, u32 addr) return 0; } -void tipc_cltr_attach_node(struct cluster *c_ptr, struct node *n_ptr) +void tipc_cltr_attach_node(struct cluster *c_ptr, struct tipc_node *n_ptr) { u32 n_num = tipc_node(n_ptr->addr); u32 max_n_num = tipc_max_nodes; @@ -196,7 +196,7 @@ u32 tipc_cltr_select_router(struct cluster *c_ptr, u32 ref) * Uses deterministic and fair algorithm. */ -struct node *tipc_cltr_select_node(struct cluster *c_ptr, u32 selector) +struct tipc_node *tipc_cltr_select_node(struct cluster *c_ptr, u32 selector) { u32 n_num; u32 mask = tipc_max_nodes; @@ -379,7 +379,7 @@ 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; + struct tipc_node *n_ptr; unchar *node_table; u32 table_size; u32 router; @@ -499,7 +499,7 @@ 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; + struct tipc_node *n_ptr; u32 n_num; u32 tstop; @@ -534,7 +534,7 @@ void tipc_cltr_broadcast(struct sk_buff *buf) { struct sk_buff *buf_copy; struct cluster *c_ptr; - struct node *n_ptr; + struct tipc_node *n_ptr; u32 n_num; u32 tstart; u32 tstop; diff --git a/net/tipc/cluster.h b/net/tipc/cluster.h index 62df074afaec..333efb0b9c44 100644 --- a/net/tipc/cluster.h +++ b/net/tipc/cluster.h @@ -54,24 +54,24 @@ struct cluster { u32 addr; struct _zone *owner; - struct node **nodes; + struct tipc_node **nodes; u32 highest_node; u32 highest_slave; }; -extern struct node **tipc_local_nodes; +extern struct tipc_node **tipc_local_nodes; extern u32 tipc_highest_allowed_slave; -extern struct node_map tipc_cltr_bcast_nodes; +extern struct tipc_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); +struct tipc_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_attach_node(struct cluster *c_ptr, struct tipc_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); diff --git a/net/tipc/discover.c b/net/tipc/discover.c index 1657f0e795ff..74b7d1e28aec 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -193,7 +193,7 @@ void tipc_disc_recv_msg(struct sk_buff *buf, struct bearer *b_ptr) /* Always accept link here */ struct sk_buff *rbuf; struct tipc_media_addr *addr; - struct node *n_ptr = tipc_node_find(orig); + struct tipc_node *n_ptr = tipc_node_find(orig); int link_fully_up; dbg(" in own cluster\n"); diff --git a/net/tipc/link.c b/net/tipc/link.c index d60113ba4b1b..dd4c18b9a35b 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -1155,7 +1155,7 @@ int tipc_link_send_buf(struct link *l_ptr, struct sk_buff *buf) int tipc_link_send(struct sk_buff *buf, u32 dest, u32 selector) { struct link *l_ptr; - struct node *n_ptr; + struct tipc_node *n_ptr; int res = -ELINKCONG; read_lock_bh(&tipc_net_lock); @@ -1226,7 +1226,7 @@ static int link_send_buf_fast(struct link *l_ptr, struct sk_buff *buf, int tipc_send_buf_fast(struct sk_buff *buf, u32 destnode) { struct link *l_ptr; - struct node *n_ptr; + struct tipc_node *n_ptr; int res; u32 selector = msg_origport(buf_msg(buf)) & 1; u32 dummy; @@ -1270,7 +1270,7 @@ int tipc_link_send_sections_fast(struct port *sender, struct tipc_msg *hdr = &sender->publ.phdr; struct link *l_ptr; struct sk_buff *buf; - struct node *node; + struct tipc_node *node; int res; u32 selector = msg_origport(hdr) & 1; @@ -1364,7 +1364,7 @@ static int link_send_sections_long(struct port *sender, u32 destaddr) { struct link *l_ptr; - struct node *node; + struct tipc_node *node; struct tipc_msg *hdr = &sender->publ.phdr; u32 dsz = msg_data_sz(hdr); u32 max_pkt,fragm_sz,rest; @@ -1636,7 +1636,7 @@ void tipc_link_push_queue(struct link *l_ptr) static void link_reset_all(unsigned long addr) { - struct node *n_ptr; + struct tipc_node *n_ptr; char addr_string[16]; u32 i; @@ -1682,7 +1682,7 @@ static void link_retransmit_failure(struct link *l_ptr, struct sk_buff *buf) /* 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)); @@ -1843,7 +1843,7 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) read_lock_bh(&tipc_net_lock); while (head) { struct bearer *b_ptr = (struct bearer *)tb_ptr; - struct node *n_ptr; + struct tipc_node *n_ptr; struct link *l_ptr; struct sk_buff *crs; struct sk_buff *buf = head; @@ -2935,7 +2935,7 @@ void tipc_link_set_queue_limits(struct link *l_ptr, u32 window) * Returns pointer to link (or 0 if invalid link name). */ -static struct link *link_find_link(const char *name, struct node **node) +static struct link *link_find_link(const char *name, struct tipc_node **node) { struct link_name link_name_parts; struct bearer *b_ptr; @@ -2965,7 +2965,7 @@ 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; + struct tipc_node *node; int res; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_CONFIG)) @@ -3043,7 +3043,7 @@ struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_ { char *link_name; struct link *l_ptr; - struct node *node; + struct tipc_node *node; if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); @@ -3091,7 +3091,7 @@ 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_node *node; char *status; u32 profile_total = 0; @@ -3207,7 +3207,7 @@ int link_control(const char *name, u32 op, u32 val) int res = -EINVAL; struct link *l_ptr; u32 bearer_id; - struct node * node; + struct tipc_node * node; u32 a; a = link_name2addr(name, &bearer_id); @@ -3249,7 +3249,7 @@ int link_control(const char *name, u32 op, u32 val) u32 tipc_link_get_max_pkt(u32 dest, u32 selector) { - struct node *n_ptr; + struct tipc_node *n_ptr; struct link *l_ptr; u32 res = MAX_PKT_DEFAULT; diff --git a/net/tipc/link.h b/net/tipc/link.h index 52f3e7c1871f..6a51e38ad25c 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -116,7 +116,7 @@ struct link { char name[TIPC_MAX_LINK_NAME]; struct tipc_media_addr media_addr; struct timer_list timer; - struct node *owner; + struct tipc_node *owner; struct list_head link_list; /* Management and link supervision data */ diff --git a/net/tipc/name_table.h b/net/tipc/name_table.h index b9e7cd336d76..139882d4ed00 100644 --- a/net/tipc/name_table.h +++ b/net/tipc/name_table.h @@ -76,7 +76,7 @@ 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; diff --git a/net/tipc/net.c b/net/tipc/net.c index ec7b04fbdc43..7906608bf510 100644 --- a/net/tipc/net.c +++ b/net/tipc/net.c @@ -118,7 +118,7 @@ DEFINE_RWLOCK(tipc_net_lock); struct network tipc_net = { NULL }; -struct node *tipc_net_select_remote_node(u32 addr, u32 ref) +struct tipc_node *tipc_net_select_remote_node(u32 addr, u32 ref) { return tipc_zone_select_remote_node(tipc_net.zones[tipc_zone(addr)], addr, ref); } diff --git a/net/tipc/net.h b/net/tipc/net.h index d154ac2bda9a..de2b9ad8f646 100644 --- a/net/tipc/net.h +++ b/net/tipc/net.h @@ -55,7 +55,7 @@ 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); +struct tipc_node *tipc_net_select_remote_node(u32 addr, u32 ref); u32 tipc_net_select_router(u32 addr, u32 ref); int tipc_net_start(u32 addr); diff --git a/net/tipc/node.c b/net/tipc/node.c index ee952ad60218..20d98c56e152 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -46,11 +46,11 @@ #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); +void node_print(struct print_buf *buf, struct tipc_node *n_ptr, char *str); +static void node_lost_contact(struct tipc_node *n_ptr); +static void node_established_contact(struct tipc_node *n_ptr); -struct node *tipc_nodes = NULL; /* sorted list of nodes within cluster */ +struct tipc_node *tipc_nodes = NULL; /* sorted list of nodes within cluster */ static DEFINE_SPINLOCK(node_create_lock); @@ -66,11 +66,11 @@ u32 tipc_own_tag = 0; * but this is a non-trivial change.) */ -struct node *tipc_node_create(u32 addr) +struct tipc_node *tipc_node_create(u32 addr) { struct cluster *c_ptr; - struct node *n_ptr; - struct node **curr_node; + struct tipc_node *n_ptr; + struct tipc_node **curr_node; spin_lock_bh(&node_create_lock); @@ -120,7 +120,7 @@ struct node *tipc_node_create(u32 addr) return n_ptr; } -void tipc_node_delete(struct node *n_ptr) +void tipc_node_delete(struct tipc_node *n_ptr) { if (!n_ptr) return; @@ -146,7 +146,7 @@ void tipc_node_delete(struct node *n_ptr) * 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 link *l_ptr) { struct link **active = &n_ptr->active_links[0]; @@ -180,7 +180,7 @@ void tipc_node_link_up(struct node *n_ptr, struct link *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]; u32 i; @@ -208,7 +208,7 @@ 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 link *l_ptr) { struct link **active; @@ -235,30 +235,30 @@ void tipc_node_link_down(struct node *n_ptr, struct link *l_ptr) node_lost_contact(n_ptr); } -int tipc_node_has_active_links(struct node *n_ptr) +int tipc_node_has_active_links(struct tipc_node *n_ptr) { return (n_ptr && ((n_ptr->active_links[0]) || (n_ptr->active_links[1]))); } -int tipc_node_has_redundant_links(struct node *n_ptr) +int tipc_node_has_redundant_links(struct tipc_node *n_ptr) { return (n_ptr->working_links > 1); } -static int tipc_node_has_active_routes(struct node *n_ptr) +static int tipc_node_has_active_routes(struct tipc_node *n_ptr) { return (n_ptr && (n_ptr->last_router >= 0)); } -int tipc_node_is_up(struct node *n_ptr) +int tipc_node_is_up(struct tipc_node *n_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 tipc_node *tipc_node_attach_link(struct link *l_ptr) { - struct node *n_ptr = tipc_node_find(l_ptr->addr); + struct tipc_node *n_ptr = tipc_node_find(l_ptr->addr); if (!n_ptr) n_ptr = tipc_node_create(l_ptr->addr); @@ -285,7 +285,7 @@ struct node *tipc_node_attach_link(struct link *l_ptr) return NULL; } -void tipc_node_detach_link(struct node *n_ptr, struct link *l_ptr) +void tipc_node_detach_link(struct tipc_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--; @@ -338,7 +338,7 @@ void tipc_node_detach_link(struct node *n_ptr, struct link *l_ptr) * */ -static void node_established_contact(struct node *n_ptr) +static void node_established_contact(struct tipc_node *n_ptr) { struct cluster *c_ptr; @@ -384,10 +384,10 @@ static void node_established_contact(struct node *n_ptr) tipc_highest_allowed_slave); } -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; + struct tipc_node_subscr *ns, *tns; char addr_string[16]; u32 i; @@ -466,9 +466,9 @@ static void node_lost_contact(struct node *n_ptr) * Called by when cluster local lookup has failed. */ -struct node *tipc_node_select_next_hop(u32 addr, u32 selector) +struct tipc_node *tipc_node_select_next_hop(u32 addr, u32 selector) { - struct node *n_ptr; + struct tipc_node *n_ptr; u32 router_addr; if (!tipc_addr_domain_valid(addr)) @@ -513,7 +513,7 @@ struct node *tipc_node_select_next_hop(u32 addr, u32 selector) * Uses a deterministic and fair algorithm for selecting router node. */ -u32 tipc_node_select_router(struct node *n_ptr, u32 ref) +u32 tipc_node_select_router(struct tipc_node *n_ptr, u32 ref) { u32 ulim; u32 mask; @@ -551,7 +551,7 @@ u32 tipc_node_select_router(struct node *n_ptr, u32 ref) return tipc_addr(own_zone(), own_cluster(), r); } -void tipc_node_add_router(struct node *n_ptr, u32 router) +void tipc_node_add_router(struct tipc_node *n_ptr, u32 router) { u32 r_num = tipc_node(router); @@ -562,7 +562,7 @@ void tipc_node_add_router(struct node *n_ptr, u32 router) !n_ptr->routers[n_ptr->last_router]); } -void tipc_node_remove_router(struct node *n_ptr, u32 router) +void tipc_node_remove_router(struct tipc_node *n_ptr, u32 router) { u32 r_num = tipc_node(router); @@ -580,7 +580,7 @@ void tipc_node_remove_router(struct node *n_ptr, u32 router) } #if 0 -void node_print(struct print_buf *buf, struct node *n_ptr, char *str) +void node_print(struct print_buf *buf, struct tipc_node *n_ptr, char *str) { u32 i; @@ -597,7 +597,7 @@ void node_print(struct print_buf *buf, struct node *n_ptr, char *str) u32 tipc_available_nodes(const u32 domain) { - struct node *n_ptr; + struct tipc_node *n_ptr; u32 cnt = 0; read_lock_bh(&tipc_net_lock); @@ -615,7 +615,7 @@ 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; @@ -667,7 +667,7 @@ 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; diff --git a/net/tipc/node.h b/net/tipc/node.h index cd1882654bbb..6f990da5d143 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -43,7 +43,7 @@ #include "bearer.h" /** - * struct node - TIPC node structure + * 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 @@ -68,11 +68,11 @@ * @defragm: list of partially reassembled b'cast message fragments from node */ -struct node { +struct tipc_node { u32 addr; spinlock_t lock; struct cluster *owner; - struct node *next; + struct tipc_node *next; struct list_head nsub; struct link *active_links[2]; struct link *links[MAX_BEARERS]; @@ -94,26 +94,26 @@ struct node { } bclink; }; -extern struct node *tipc_nodes; +extern struct tipc_node *tipc_nodes; extern u32 tipc_own_tag; -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_create(u32 addr); +void tipc_node_delete(struct tipc_node *n_ptr); +struct tipc_node *tipc_node_attach_link(struct link *l_ptr); +void tipc_node_detach_link(struct tipc_node *n_ptr, struct link *l_ptr); +void tipc_node_link_down(struct tipc_node *n_ptr, struct link *l_ptr); +void tipc_node_link_up(struct tipc_node *n_ptr, struct link *l_ptr); +int tipc_node_has_active_links(struct tipc_node *n_ptr); +int tipc_node_has_redundant_links(struct tipc_node *n_ptr); +u32 tipc_node_select_router(struct tipc_node *n_ptr, u32 ref); +struct tipc_node *tipc_node_select_next_hop(u32 addr, u32 selector); +int tipc_node_is_up(struct tipc_node *n_ptr); +void tipc_node_add_router(struct tipc_node *n_ptr, u32 router); +void tipc_node_remove_router(struct tipc_node *n_ptr, u32 router); 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); -static inline struct node *tipc_node_find(u32 addr) +static inline struct tipc_node *tipc_node_find(u32 addr) { if (likely(in_own_cluster(addr))) return tipc_local_nodes[tipc_node(addr)]; @@ -126,19 +126,19 @@ static inline struct node *tipc_node_find(u32 addr) return NULL; } -static inline struct node *tipc_node_select(u32 addr, u32 selector) +static inline struct tipc_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 *n_ptr) { spin_lock_bh(&n_ptr->lock); } -static inline void tipc_node_unlock(struct node *n_ptr) +static inline void tipc_node_unlock(struct tipc_node *n_ptr) { spin_unlock_bh(&n_ptr->lock); } diff --git a/net/tipc/node_subscr.c b/net/tipc/node_subscr.c index 8ecbd0fb6103..19194d476a9e 100644 --- a/net/tipc/node_subscr.c +++ b/net/tipc/node_subscr.c @@ -44,7 +44,7 @@ * tipc_nodesub_subscribe - create "node down" subscription for specified node */ -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) { if (addr == tipc_own_addr) { @@ -69,7 +69,7 @@ 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; diff --git a/net/tipc/node_subscr.h b/net/tipc/node_subscr.h index 5f3f5859b84c..006ed739f515 100644 --- a/net/tipc/node_subscr.h +++ b/net/tipc/node_subscr.h @@ -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); #endif diff --git a/net/tipc/port.h b/net/tipc/port.h index e5f8c16429bd..ff31ee4a1dc3 100644 --- a/net/tipc/port.h +++ b/net/tipc/port.h @@ -105,7 +105,7 @@ struct port { 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; diff --git a/net/tipc/zone.c b/net/tipc/zone.c index 3506f8563441..2c01ba2d86bf 100644 --- a/net/tipc/zone.c +++ b/net/tipc/zone.c @@ -111,10 +111,10 @@ void tipc_zone_send_external_routes(struct _zone *z_ptr, u32 dest) } } -struct node *tipc_zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref) +struct tipc_node *tipc_zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref) { struct cluster *c_ptr; - struct node *n_ptr; + struct tipc_node *n_ptr; u32 c_num; if (!z_ptr) diff --git a/net/tipc/zone.h b/net/tipc/zone.h index 6e7a08df8af5..7bdc3406ba9b 100644 --- a/net/tipc/zone.h +++ b/net/tipc/zone.h @@ -54,7 +54,7 @@ struct _zone { u32 links; }; -struct node *tipc_zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref); +struct tipc_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); -- cgit v1.2.3