From 43dff98b022ded593e73c3784bac03bc9fc7ec55 Mon Sep 17 00:00:00 2001 From: Shaun Pereira Date: Fri, 28 Apr 2006 12:00:17 -0700 Subject: [X25]: fix for spinlock recurse and spinlock lockup with timer handler When the sk_timer function x25_heartbeat_expiry() is called by the kernel in a running/terminating process, spinlock-recursion and spinlock-lockup locks up the kernel. This has happened with testing on some distro's and the patch below fixed it. Signed-off-by: Shaun Pereira Signed-off-by: David S. Miller --- net/x25/x25_timer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/x25/x25_timer.c b/net/x25/x25_timer.c index 0a92e1da392..71ff3088f6f 100644 --- a/net/x25/x25_timer.c +++ b/net/x25/x25_timer.c @@ -114,8 +114,9 @@ static void x25_heartbeat_expiry(unsigned long param) if (sock_flag(sk, SOCK_DESTROY) || (sk->sk_state == TCP_LISTEN && sock_flag(sk, SOCK_DEAD))) { + bh_unlock_sock(sk); x25_destroy_socket(sk); - goto unlock; + return; } break; @@ -128,7 +129,6 @@ static void x25_heartbeat_expiry(unsigned long param) } restart_heartbeat: x25_start_heartbeat(sk); -unlock: bh_unlock_sock(sk); } -- cgit v1.2.3 From 89bbb0a361cdae50eec863f10a876b58abf7d312 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 28 Apr 2006 12:11:36 -0700 Subject: [PKT_SCHED] netem: fix loss The following one line fix is needed to make loss function of netem work right when doing loss on the local host. Otherwise, higher layers just recover. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/sched/sch_netem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 7228d30512c..5a4a4d0ae50 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -167,7 +167,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (count == 0) { sch->qstats.drops++; kfree_skb(skb); - return NET_XMIT_DROP; + return NET_XMIT_BYPASS; } /* -- cgit v1.2.3 From 09493abfdbe8144ce0805efac7a8a3c4eb869c12 Mon Sep 17 00:00:00 2001 From: Soyoung Park Date: Fri, 28 Apr 2006 14:59:44 -0700 Subject: [NETLINK]: cleanup unused macro in net/netlink/af_netlink.c 1 line removal, of unused macro. ran 'egrep -r' from linux-2.6.16/ for Nprintk and didn't see it anywhere else but here, in #define... Signed-off-by: Soyoung Park Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 2a233ffcf61..b8ea61f357e 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -61,7 +61,6 @@ #include #include -#define Nprintk(a...) #define NLGRPSZ(x) (ALIGN(x, sizeof(unsigned long) * 8) / 8) struct netlink_sock { -- cgit v1.2.3 From a536e0778781c1f2ce38cf8e1d82ce258f0193c1 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 28 Apr 2006 15:19:17 -0700 Subject: [IPV4]: inet_init() -> fs_initcall Convert inet_init to an fs_initcall to make sure its called before any device driver's initcall. Signed-off-by: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- net/ipv4/af_inet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index dc206f1f914..0a277453526 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1257,7 +1257,7 @@ out_unregister_udp_proto: goto out; } -module_init(inet_init); +fs_initcall(inet_init); /* ------------------------------------------------------------------------ */ -- cgit v1.2.3 From a76e07acd0de635122c5e60ccd5e55f9d5082391 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 28 Apr 2006 15:22:13 -0700 Subject: [IPSEC]: Fix IP ID selection I was looking through the xfrm input/output code in order to abstract out the address family specific encapsulation/decapsulation code. During that process I found this bug in the IP ID selection code in xfrm4_output.c. At that point dst is still the xfrm_dst for the current SA which represents an internal flow as far as the IPsec tunnel is concerned. Since the IP ID is going to sit on the outside of the encapsulated packet, we obviously want the external flow which is just dst->child. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/xfrm4_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c index 32ad229b4fe..4ef8efaf6a6 100644 --- a/net/ipv4/xfrm4_output.c +++ b/net/ipv4/xfrm4_output.c @@ -62,7 +62,7 @@ static void xfrm4_encap(struct sk_buff *skb) top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ? 0 : (iph->frag_off & htons(IP_DF)); if (!top_iph->frag_off) - __ip_select_ident(top_iph, dst, 0); + __ip_select_ident(top_iph, dst->child, 0); top_iph->ttl = dst_metric(dst->child, RTAX_HOPLIMIT); -- cgit v1.2.3 From 8dff7c29707b7514043539f5ab5e0a6eb7bd9dcd Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 28 Apr 2006 15:23:59 -0700 Subject: [XFRM]: fix softirq-unsafe xfrm typemap->lock use xfrm typemap->lock may be used in softirq context, so all write_lock() uses must be softirq-safe. Signed-off-by: Ingo Molnar Signed-off-by: David S. Miller --- net/xfrm/xfrm_policy.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index c3725fe2a8f..e5b0afc94f1 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -57,12 +57,12 @@ int xfrm_register_type(struct xfrm_type *type, unsigned short family) return -EAFNOSUPPORT; typemap = afinfo->type_map; - write_lock(&typemap->lock); + write_lock_bh(&typemap->lock); if (likely(typemap->map[type->proto] == NULL)) typemap->map[type->proto] = type; else err = -EEXIST; - write_unlock(&typemap->lock); + write_unlock_bh(&typemap->lock); xfrm_policy_put_afinfo(afinfo); return err; } @@ -78,12 +78,12 @@ int xfrm_unregister_type(struct xfrm_type *type, unsigned short family) return -EAFNOSUPPORT; typemap = afinfo->type_map; - write_lock(&typemap->lock); + write_lock_bh(&typemap->lock); if (unlikely(typemap->map[type->proto] != type)) err = -ENOENT; else typemap->map[type->proto] = NULL; - write_unlock(&typemap->lock); + write_unlock_bh(&typemap->lock); xfrm_policy_put_afinfo(afinfo); return err; } -- cgit v1.2.3 From 83de47cd0c5738105f40e65191b0761dfa7431ac Mon Sep 17 00:00:00 2001 From: Hua Zhong Date: Fri, 28 Apr 2006 15:26:50 -0700 Subject: [TCP]: Fix unlikely usage in tcp_transmit_skb() The following unlikely should be replaced by likely because the condition happens every time unless there is a hard error to transmit a packet. Signed-off-by: Hua Zhong Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index a28ae593b97..743016baa04 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -465,7 +465,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, TCP_INC_STATS(TCP_MIB_OUTSEGS); err = icsk->icsk_af_ops->queue_xmit(skb, 0); - if (unlikely(err <= 0)) + if (likely(err <= 0)) return err; tcp_enter_cwr(sk); -- cgit v1.2.3 From f3111502c065d32dcb021f55e30398aaebd8fb0f Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 28 Apr 2006 15:30:03 -0700 Subject: [XFRM]: fix incorrect xfrm_state_afinfo_lock use xfrm_state_afinfo_lock can be read-locked from bh context, so take it in a bh-safe manner in xfrm_state_register_afinfo() and xfrm_state_unregister_afinfo(). Found by the lock validator. Signed-off-by: Ingo Molnar Signed-off-by: David S. Miller --- net/xfrm/xfrm_state.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 3dc3e1f3b7a..93a2f36ad3d 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -1061,7 +1061,7 @@ int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo) return -EINVAL; if (unlikely(afinfo->family >= NPROTO)) return -EAFNOSUPPORT; - write_lock(&xfrm_state_afinfo_lock); + write_lock_bh(&xfrm_state_afinfo_lock); if (unlikely(xfrm_state_afinfo[afinfo->family] != NULL)) err = -ENOBUFS; else { @@ -1069,7 +1069,7 @@ int xfrm_state_register_afinfo(struct xfrm_state_afinfo *afinfo) afinfo->state_byspi = xfrm_state_byspi; xfrm_state_afinfo[afinfo->family] = afinfo; } - write_unlock(&xfrm_state_afinfo_lock); + write_unlock_bh(&xfrm_state_afinfo_lock); return err; } EXPORT_SYMBOL(xfrm_state_register_afinfo); @@ -1081,7 +1081,7 @@ int xfrm_state_unregister_afinfo(struct xfrm_state_afinfo *afinfo) return -EINVAL; if (unlikely(afinfo->family >= NPROTO)) return -EAFNOSUPPORT; - write_lock(&xfrm_state_afinfo_lock); + write_lock_bh(&xfrm_state_afinfo_lock); if (likely(xfrm_state_afinfo[afinfo->family] != NULL)) { if (unlikely(xfrm_state_afinfo[afinfo->family] != afinfo)) err = -EINVAL; @@ -1091,7 +1091,7 @@ int xfrm_state_unregister_afinfo(struct xfrm_state_afinfo *afinfo) afinfo->state_bydst = NULL; } } - write_unlock(&xfrm_state_afinfo_lock); + write_unlock_bh(&xfrm_state_afinfo_lock); return err; } EXPORT_SYMBOL(xfrm_state_unregister_afinfo); -- cgit v1.2.3 From e959d8121fcbfee6ec049cc617e9423d1799f2e4 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 28 Apr 2006 15:32:29 -0700 Subject: [XFRM]: fix incorrect xfrm_policy_afinfo_lock use xfrm_policy_afinfo_lock can be taken in bh context, at: [] lockdep_acquire_read+0x54/0x6d [] _read_lock+0x15/0x22 [] xfrm_policy_get_afinfo+0x1a/0x3d [] xfrm_decode_session+0x12/0x32 [] ip_route_me_harder+0x1c9/0x25b [] ip_nat_local_fn+0x94/0xad [] nf_iterate+0x2e/0x7a [] nf_hook_slow+0x3c/0x9e [] ip_push_pending_frames+0x2de/0x3a7 [] icmp_push_reply+0x136/0x141 [] icmp_reply+0x118/0x1a0 [] icmp_echo+0x44/0x46 [] icmp_rcv+0x111/0x138 [] ip_local_deliver+0x150/0x1f9 [] ip_rcv+0x3d5/0x413 [] netif_receive_skb+0x337/0x356 [] process_backlog+0x95/0x110 [] net_rx_action+0xa5/0x16d [] __do_softirq+0x6f/0xe6 [] do_softirq+0x52/0xb1 this means that all write-locking of xfrm_policy_afinfo_lock must be bh-safe. This patch fixes xfrm_policy_register_afinfo() and xfrm_policy_unregister_afinfo(). Signed-off-by: Ingo Molnar Signed-off-by: David S. Miller --- net/xfrm/xfrm_policy.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index e5b0afc94f1..b469c8b5461 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -1251,7 +1251,7 @@ int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo) return -EINVAL; if (unlikely(afinfo->family >= NPROTO)) return -EAFNOSUPPORT; - write_lock(&xfrm_policy_afinfo_lock); + write_lock_bh(&xfrm_policy_afinfo_lock); if (unlikely(xfrm_policy_afinfo[afinfo->family] != NULL)) err = -ENOBUFS; else { @@ -1268,7 +1268,7 @@ int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo) afinfo->garbage_collect = __xfrm_garbage_collect; xfrm_policy_afinfo[afinfo->family] = afinfo; } - write_unlock(&xfrm_policy_afinfo_lock); + write_unlock_bh(&xfrm_policy_afinfo_lock); return err; } EXPORT_SYMBOL(xfrm_policy_register_afinfo); @@ -1280,7 +1280,7 @@ int xfrm_policy_unregister_afinfo(struct xfrm_policy_afinfo *afinfo) return -EINVAL; if (unlikely(afinfo->family >= NPROTO)) return -EAFNOSUPPORT; - write_lock(&xfrm_policy_afinfo_lock); + write_lock_bh(&xfrm_policy_afinfo_lock); if (likely(xfrm_policy_afinfo[afinfo->family] != NULL)) { if (unlikely(xfrm_policy_afinfo[afinfo->family] != afinfo)) err = -EINVAL; @@ -1294,7 +1294,7 @@ int xfrm_policy_unregister_afinfo(struct xfrm_policy_afinfo *afinfo) afinfo->garbage_collect = NULL; } } - write_unlock(&xfrm_policy_afinfo_lock); + write_unlock_bh(&xfrm_policy_afinfo_lock); return err; } EXPORT_SYMBOL(xfrm_policy_unregister_afinfo); -- cgit v1.2.3 From c302e6d54e468ee9b1e18152e2e9da06953f3473 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Fri, 28 Apr 2006 15:59:15 -0700 Subject: [IPV6]: Fix race in route selection. We eliminated rt6_dflt_lock (to protect default router pointer) at 2.6.17-rc1, and introduced rt6_select() for general router selection. The function is called in the context of rt6_lock read-lock held, but this means, we have some race conditions when we do round-robin. Signed-off-by; YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/route.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 79078747a64..0190e39096b 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -317,7 +317,7 @@ static struct rt6_info *rt6_select(struct rt6_info **head, int oif, __FUNCTION__, head, head ? *head : NULL, oif); for (rt = rt0, metric = rt0->rt6i_metric; - rt && rt->rt6i_metric == metric; + rt && rt->rt6i_metric == metric && (!last || rt != rt0); rt = rt->u.next) { int m; @@ -343,9 +343,12 @@ static struct rt6_info *rt6_select(struct rt6_info **head, int oif, (strict & RT6_SELECT_F_REACHABLE) && last && last != rt0) { /* no entries matched; do round-robin */ + static spinlock_t lock = SPIN_LOCK_UNLOCKED; + spin_lock(&lock); *head = rt0->u.next; rt0->u.next = last->u.next; last->u.next = rt0; + spin_unlock(&lock); } RT6_TRACE("%s() => %p, score=%d\n", -- cgit v1.2.3 From d6fe3945b42d09a1eca7ad180a1646e585b8594f Mon Sep 17 00:00:00 2001 From: Steve Grubb Date: Thu, 30 Mar 2006 12:20:22 -0500 Subject: [PATCH] sockaddr patch On Thursday 23 March 2006 09:08, John D. Ramsdell wrote: > I noticed that a socketcall(bind) and socketcall(connect) event contain a > record of type=SOCKADDR, but I cannot see one for a system call event > associated with socketcall(accept). Recording the sockaddr of an accepted > socket is important for cross platform information flow analys Thanks for pointing this out. The following patch should address this. Signed-off-by: Steve Grubb Signed-off-by: Al Viro --- net/socket.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/socket.c b/net/socket.c index 0ce12dfc7a7..02948b622bd 100644 --- a/net/socket.c +++ b/net/socket.c @@ -267,6 +267,8 @@ int move_addr_to_user(void *kaddr, int klen, void __user *uaddr, int __user *ule return -EINVAL; if(len) { + if (audit_sockaddr(klen, kaddr)) + return -ENOMEM; if(copy_to_user(uaddr,kaddr,len)) return -EFAULT; } -- cgit v1.2.3 From e7c3497013a7e5496ce3d5fd3c73b5cf5af7a56e Mon Sep 17 00:00:00 2001 From: Steve Grubb Date: Mon, 3 Apr 2006 09:08:13 -0400 Subject: [PATCH] Reworked patch for labels on user space messages The below patch should be applied after the inode and ipc sid patches. This patch is a reworking of Tim's patch that has been updated to match the inode and ipc patches since its similar. [updated: > Stephen Smalley also wanted to change a variable from isec to tsec in the > user sid patch. ] Signed-off-by: Steve Grubb Signed-off-by: Al Viro --- net/netlink/af_netlink.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 2a233ffcf61..09fbc4bc708 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -56,6 +56,7 @@ #include #include #include +#include #include #include @@ -1157,6 +1158,7 @@ static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock, NETLINK_CB(skb).dst_pid = dst_pid; NETLINK_CB(skb).dst_group = dst_group; NETLINK_CB(skb).loginuid = audit_get_loginuid(current->audit_context); + selinux_get_task_sid(current, &(NETLINK_CB(skb).sid)); memcpy(NETLINK_CREDS(skb), &siocb->scm->creds, sizeof(struct ucred)); /* What can I do? Netlink is asynchronous, so that -- cgit v1.2.3 From 46c5ea3c9ae7fbc6e52a13c92e59d4fc7f4ca80a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 2 May 2006 05:12:22 +0200 Subject: [NETFILTER] x_tables: fix compat related crash on non-x86 When iptables userspace adds an ipt_standard_target, it calculates the size of the entire entry as: sizeof(struct ipt_entry) + XT_ALIGN(sizeof(struct ipt_standard_target)) ipt_standard_target looks like this: struct xt_standard_target { struct xt_entry_target target; int verdict; }; xt_entry_target contains a pointer, so when compiled for 64 bit the structure gets an extra 4 byte of padding at the end. On 32 bit architectures where iptables aligns to 8 byte it will also have 4 byte padding at the end because it is only 36 bytes large. The compat_ipt_standard_fn in the kernel adjusts the offsets by sizeof(struct ipt_standard_target) - sizeof(struct compat_ipt_standard_target), which will always result in 4, even if the structure from userspace was already padded to a multiple of 8. On x86 this works out by accident because userspace only aligns to 4, on all other architectures this is broken and causes incorrect adjustments to the size and following offsets. Thanks to Linus for lots of debugging help and testing. Signed-off-by: Patrick McHardy Signed-off-by: Linus Torvalds --- net/ipv4/netfilter/ip_tables.c | 33 ++++++++++++++------------------- 1 file changed, 14 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index d25ac8ba6eb..6d1c1156394 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -956,15 +956,16 @@ struct compat_ipt_standard_target compat_int_t verdict; }; -#define IPT_ST_OFFSET (sizeof(struct ipt_standard_target) - \ - sizeof(struct compat_ipt_standard_target)) - struct compat_ipt_standard { struct compat_ipt_entry entry; struct compat_ipt_standard_target target; }; +#define IPT_ST_LEN XT_ALIGN(sizeof(struct ipt_standard_target)) +#define IPT_ST_COMPAT_LEN COMPAT_XT_ALIGN(sizeof(struct compat_ipt_standard_target)) +#define IPT_ST_OFFSET (IPT_ST_LEN - IPT_ST_COMPAT_LEN) + static int compat_ipt_standard_fn(void *target, void **dstptr, int *size, int convert) { @@ -975,35 +976,29 @@ static int compat_ipt_standard_fn(void *target, ret = 0; switch (convert) { case COMPAT_TO_USER: - pst = (struct ipt_standard_target *)target; + pst = target; memcpy(&compat_st.target, &pst->target, - sizeof(struct ipt_entry_target)); + sizeof(compat_st.target)); compat_st.verdict = pst->verdict; if (compat_st.verdict > 0) compat_st.verdict -= compat_calc_jump(compat_st.verdict); - compat_st.target.u.user.target_size = - sizeof(struct compat_ipt_standard_target); - if (__copy_to_user(*dstptr, &compat_st, - sizeof(struct compat_ipt_standard_target))) + compat_st.target.u.user.target_size = IPT_ST_COMPAT_LEN; + if (copy_to_user(*dstptr, &compat_st, IPT_ST_COMPAT_LEN)) ret = -EFAULT; *size -= IPT_ST_OFFSET; - *dstptr += sizeof(struct compat_ipt_standard_target); + *dstptr += IPT_ST_COMPAT_LEN; break; case COMPAT_FROM_USER: - pcompat_st = - (struct compat_ipt_standard_target *)target; - memcpy(&st.target, &pcompat_st->target, - sizeof(struct ipt_entry_target)); + pcompat_st = target; + memcpy(&st.target, &pcompat_st->target, IPT_ST_COMPAT_LEN); st.verdict = pcompat_st->verdict; if (st.verdict > 0) st.verdict += compat_calc_jump(st.verdict); - st.target.u.user.target_size = - sizeof(struct ipt_standard_target); - memcpy(*dstptr, &st, - sizeof(struct ipt_standard_target)); + st.target.u.user.target_size = IPT_ST_LEN; + memcpy(*dstptr, &st, IPT_ST_LEN); *size += IPT_ST_OFFSET; - *dstptr += sizeof(struct ipt_standard_target); + *dstptr += IPT_ST_LEN; break; case COMPAT_CALC_SIZE: *size += IPT_ST_OFFSET; -- cgit v1.2.3 From e17df688f7064dae1417ce425dd1e4b71d24d63b Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 2 May 2006 23:23:07 +0200 Subject: [NETFILTER] SCTP conntrack: fix infinite loop fix infinite loop in the SCTP-netfilter code: check SCTP chunk size to guarantee progress of for_each_sctp_chunk(). (all other uses of for_each_sctp_chunk() are preceded by do_basic_checks(), so this fix should be complete.) Based on patch from Ingo Molnar CVE-2006-1527 Signed-off-by: Patrick McHardy Signed-off-by: Linus Torvalds --- net/ipv4/netfilter/ip_conntrack_proto_sctp.c | 11 +++++++---- net/netfilter/nf_conntrack_proto_sctp.c | 11 +++++++---- 2 files changed, 14 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c index 5259abd0fb4..0416073c560 100644 --- a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c +++ b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c @@ -235,12 +235,15 @@ static int do_basic_checks(struct ip_conntrack *conntrack, flag = 1; } - /* Cookie Ack/Echo chunks not the first OR - Init / Init Ack / Shutdown compl chunks not the only chunks */ - if ((sch->type == SCTP_CID_COOKIE_ACK + /* + * Cookie Ack/Echo chunks not the first OR + * Init / Init Ack / Shutdown compl chunks not the only chunks + * OR zero-length. + */ + if (((sch->type == SCTP_CID_COOKIE_ACK || sch->type == SCTP_CID_COOKIE_ECHO || flag) - && count !=0 ) { + && count !=0) || !sch->length) { DEBUGP("Basic checks failed\n"); return 1; } diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index 9cccc325b68..0c6da496cfa 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -240,12 +240,15 @@ static int do_basic_checks(struct nf_conn *conntrack, flag = 1; } - /* Cookie Ack/Echo chunks not the first OR - Init / Init Ack / Shutdown compl chunks not the only chunks */ - if ((sch->type == SCTP_CID_COOKIE_ACK + /* + * Cookie Ack/Echo chunks not the first OR + * Init / Init Ack / Shutdown compl chunks not the only chunks + * OR zero-length. + */ + if (((sch->type == SCTP_CID_COOKIE_ACK || sch->type == SCTP_CID_COOKIE_ECHO || flag) - && count !=0 ) { + && count !=0) || !sch->length) { DEBUGP("Basic checks failed\n"); return 1; } -- cgit v1.2.3 From 6fd737031eb6869430d0f3cf6bf1440adf7aedf5 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 May 2006 23:16:29 -0700 Subject: [NETFILTER]: H.323 helper: fix endless loop caused by invalid TPKT len When the TPKT len included in the packet is below the lowest valid value of 4 an underflow occurs which results in an endless loop. Found by testcase 0000058 from the PROTOS c07-h2250v4 testsuite. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_helper_h323.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323.c b/net/ipv4/netfilter/ip_conntrack_helper_h323.c index 2c2fb700d83..518f581d39e 100644 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323.c +++ b/net/ipv4/netfilter/ip_conntrack_helper_h323.c @@ -162,6 +162,8 @@ static int get_tpkt_data(struct sk_buff **pskb, struct ip_conntrack *ct, /* Validate TPKT length */ tpktlen = tpkt[2] * 256 + tpkt[3]; + if (tpktlen < 4) + goto clear_out; if (tpktlen > tcpdatalen) { if (tcpdatalen == 4) { /* Separate TPKT header */ /* Netmeeting sends TPKT header and data separately */ -- cgit v1.2.3 From 4228e2a9890cd01b0c8cc58af6fd9e08a4b5e8a7 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 May 2006 23:17:11 -0700 Subject: [NETFILTER]: H.323 helper: fix use of uninitialized data When a Choice element contains an unsupported choice no error is returned and parsing continues normally, but the choice value is not set and contains data from the last parsed message. This may in turn lead to parsing of more stale data and following crashes. Fixes a crash triggered by testcase 0003243 from the PROTOS c07-h2250v4 testsuite following random other testcases: CPU: 0 EIP: 0060:[] Not tainted VLI EFLAGS: 00210646 (2.6.17-rc2 #3) EIP is at memmove+0x19/0x22 eax: d7be0307 ebx: d7be0307 ecx: e841fcf9 edx: d7be0307 esi: bfffffff edi: bfffffff ebp: da5eb980 esp: c0347e2c ds: 007b es: 007b ss: 0068 Process events/0 (pid: 4, threadinfo=c0347000 task=dff86a90) Stack: <0>00000006 c0347ea6 d7be0301 e09a6b2c 00000006 da5eb980 d7be003e d7be0052 c0347f6c e09a6d9c 00000006 c0347ea6 00000006 00000000 d7b9a548 00000000 c0347f6c d7b9a548 00000004 e0a1a119 0000028f 00000006 c0347ea6 00000006 Call Trace: [] mangle_contents+0x40/0xd8 [ip_nat] [] ip_nat_mangle_tcp_packet+0xa1/0x191 [ip_nat] [] set_addr+0x60/0x14d [ip_nat_h323] [] q931_help+0x2da/0x71a [ip_conntrack_h323] [] q931_help+0x30c/0x71a [ip_conntrack_h323] [] ip_conntrack_help+0x22/0x2f [ip_conntrack] [] nf_iterate+0x2e/0x5f [] xfrm4_output_finish+0x0/0x39f [] nf_hook_slow+0x42/0xb0 [] xfrm4_output_finish+0x0/0x39f [] xfrm4_output+0x3c/0x4e [] xfrm4_output_finish+0x0/0x39f [] ip_forward+0x1c2/0x1fa [] ip_rcv+0x388/0x3b5 [] netif_receive_skb+0x2bc/0x2ec [] process_backlog+0x6b/0xd0 [] net_rx_action+0x4b/0xb7 [] __do_softirq+0x35/0x7d [] do_softirq+0x38/0x3f Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c b/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c index 48078002e45..f52d2c45a6a 100644 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c +++ b/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c @@ -703,6 +703,10 @@ int decode_choice(bitstr_t * bs, field_t * f, char *base, int level) type = get_bits(bs, f->sz); } + /* Write Type */ + if (base) + *(unsigned *) base = type; + /* Check Range */ if (type >= f->ub) { /* Newer version? */ BYTE_ALIGN(bs); @@ -712,10 +716,6 @@ int decode_choice(bitstr_t * bs, field_t * f, char *base, int level) return H323_ERROR_NONE; } - /* Write Type */ - if (base) - *(unsigned *) base = type; - /* Transfer to son level */ son = &f->fields[type]; if (son->attr & STOP) { -- cgit v1.2.3 From 2354feaeb2acb78f6aabdf8410d55b44492a7949 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 May 2006 23:19:26 -0700 Subject: [NETFILTER]: NAT: silence unused variable warnings with CONFIG_XFRM=n net/ipv4/netfilter/ip_nat_standalone.c: In function 'ip_nat_out': net/ipv4/netfilter/ip_nat_standalone.c:223: warning: unused variable 'ctinfo' net/ipv4/netfilter/ip_nat_standalone.c:222: warning: unused variable 'ct' Surprisingly no complaints so far .. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_nat_standalone.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c index 8f760b28617..67e676783da 100644 --- a/net/ipv4/netfilter/ip_nat_standalone.c +++ b/net/ipv4/netfilter/ip_nat_standalone.c @@ -219,8 +219,10 @@ ip_nat_out(unsigned int hooknum, const struct net_device *out, int (*okfn)(struct sk_buff *)) { +#ifdef CONFIG_XFRM struct ip_conntrack *ct; enum ip_conntrack_info ctinfo; +#endif unsigned int ret; /* root is playing with raw sockets. */ -- cgit v1.2.3 From 7582e9d17edbabab6cbe59467c5d1b5e8c04fca8 Mon Sep 17 00:00:00 2001 From: Jing Min Zhao Date: Wed, 3 May 2006 23:19:59 -0700 Subject: [NETFILTER]: H.323 helper: Change author's email address Signed-off-by: Jing Min Zhao Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c b/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c index f52d2c45a6a..355a53a5b6c 100644 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c +++ b/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c @@ -2,7 +2,7 @@ * ip_conntrack_helper_h323_asn1.c - BER and PER decoding library for H.323 * conntrack/NAT module. * - * Copyright (c) 2006 by Jing Min Zhao + * Copyright (c) 2006 by Jing Min Zhao * * This source code is licensed under General Public License version 2. * -- cgit v1.2.3 From 7800007c1e2d42cd4120b87b0ba3f3480f17f30a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 3 May 2006 23:20:27 -0700 Subject: [NETFILTER]: x_tables: don't use __copy_{from,to}_user on unchecked memory in compat layer Noticed by Linus Torvalds Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_tables.c | 6 +++--- net/netfilter/x_tables.c | 4 ++-- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 6d1c1156394..cee3397ec27 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1441,7 +1441,7 @@ static int compat_copy_entry_to_user(struct ipt_entry *e, ret = -EFAULT; origsize = *size; ce = (struct compat_ipt_entry __user *)*dstptr; - if (__copy_to_user(ce, e, sizeof(struct ipt_entry))) + if (copy_to_user(ce, e, sizeof(struct ipt_entry))) goto out; *dstptr += sizeof(struct compat_ipt_entry); @@ -1459,9 +1459,9 @@ static int compat_copy_entry_to_user(struct ipt_entry *e, goto out; ret = -EFAULT; next_offset = e->next_offset - (origsize - *size); - if (__put_user(target_offset, &ce->target_offset)) + if (put_user(target_offset, &ce->target_offset)) goto out; - if (__put_user(next_offset, &ce->next_offset)) + if (put_user(next_offset, &ce->next_offset)) goto out; return 0; out: diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 17abf60f957..99293c63ff7 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -289,7 +289,7 @@ int xt_compat_match(void *match, void **dstptr, int *size, int convert) case COMPAT_TO_USER: pm = (struct xt_entry_match *)match; msize = pm->u.user.match_size; - if (__copy_to_user(*dstptr, pm, msize)) { + if (copy_to_user(*dstptr, pm, msize)) { ret = -EFAULT; break; } @@ -366,7 +366,7 @@ int xt_compat_target(void *target, void **dstptr, int *size, int convert) case COMPAT_TO_USER: pt = (struct xt_entry_target *)target; tsize = pt->u.user.target_size; - if (__copy_to_user(*dstptr, pt, tsize)) { + if (copy_to_user(*dstptr, pt, tsize)) { ret = -EFAULT; break; } -- cgit v1.2.3 From 0cc5ae24af08abe8e2a467f45b54c48a0f52670f Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 3 May 2006 23:22:01 -0700 Subject: [ROSE]: Remove useless prototype for rose_remove_neigh(). Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- net/rose/rose_route.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net') diff --git a/net/rose/rose_route.c b/net/rose/rose_route.c index 8631b65a731..4cb6bfa6fcb 100644 --- a/net/rose/rose_route.c +++ b/net/rose/rose_route.c @@ -48,8 +48,6 @@ static DEFINE_SPINLOCK(rose_route_list_lock); struct rose_neigh *rose_loopback_neigh; -static void rose_remove_neigh(struct rose_neigh *); - /* * Add a new route to a node, and in the process add the node and the * neighbour if it is new. -- cgit v1.2.3 From 3f072310d0ca85891323e9d325c37c76de389387 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 3 May 2006 23:22:36 -0700 Subject: [AX.25]: Spelling fix Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- net/ax25/ax25_route.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ax25/ax25_route.c b/net/ax25/ax25_route.c index f04f8630fd2..5ac98250797 100644 --- a/net/ax25/ax25_route.c +++ b/net/ax25/ax25_route.c @@ -360,7 +360,7 @@ struct file_operations ax25_route_fops = { /* * Find AX.25 route * - * Only routes with a refernce rout of zero can be destroyed. + * Only routes with a reference count of zero can be destroyed. */ static ax25_route *ax25_get_route(ax25_address *addr, struct net_device *dev) { -- cgit v1.2.3 From 86cfcb95ec60e910d7efcb35ae89bf3403befaad Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 3 May 2006 23:23:48 -0700 Subject: [AX25, ROSE]: Remove useless SET_MODULE_OWNER calls. Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- net/netrom/nr_dev.c | 1 - net/rose/rose_dev.c | 1 - 2 files changed, 2 deletions(-) (limited to 'net') diff --git a/net/netrom/nr_dev.c b/net/netrom/nr_dev.c index 509afddae56..621e5586ab0 100644 --- a/net/netrom/nr_dev.c +++ b/net/netrom/nr_dev.c @@ -185,7 +185,6 @@ static struct net_device_stats *nr_get_stats(struct net_device *dev) void nr_setup(struct net_device *dev) { - SET_MODULE_OWNER(dev); dev->mtu = NR_MAX_PACKET_SIZE; dev->hard_start_xmit = nr_xmit; dev->open = nr_open; diff --git a/net/rose/rose_dev.c b/net/rose/rose_dev.c index d297af737d1..2a1bf8e119e 100644 --- a/net/rose/rose_dev.c +++ b/net/rose/rose_dev.c @@ -135,7 +135,6 @@ static struct net_device_stats *rose_get_stats(struct net_device *dev) void rose_setup(struct net_device *dev) { - SET_MODULE_OWNER(dev); dev->mtu = ROSE_MAX_PACKET_SIZE - 2; dev->hard_start_xmit = rose_xmit; dev->open = rose_open; -- cgit v1.2.3 From 70868eace5031298c6f6e991a40a2106957f582c Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 3 May 2006 23:25:17 -0700 Subject: [AX.25]: Move AX.25 symbol exports Move AX.25 symbol exports to next to their definitions where they're supposed to be these days. Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- net/ax25/af_ax25.c | 20 ++------------------ net/ax25/ax25_addr.c | 9 +++++++++ net/ax25/ax25_iface.c | 13 +++++++++++++ net/ax25/ax25_ip.c | 3 +++ net/ax25/ax25_out.c | 3 +++ net/ax25/ax25_timer.c | 3 +++ net/ax25/ax25_uid.c | 4 ++++ 7 files changed, 37 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index dbf9b47681f..a9f13dfde07 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -228,6 +228,8 @@ ax25_cb *ax25_find_cb(ax25_address *src_addr, ax25_address *dest_addr, return NULL; } +EXPORT_SYMBOL(ax25_find_cb); + void ax25_send_to_raw(ax25_address *addr, struct sk_buff *skb, int proto) { ax25_cb *s; @@ -1979,24 +1981,6 @@ static struct notifier_block ax25_dev_notifier = { .notifier_call =ax25_device_event, }; -EXPORT_SYMBOL(ax25_hard_header); -EXPORT_SYMBOL(ax25_rebuild_header); -EXPORT_SYMBOL(ax25_findbyuid); -EXPORT_SYMBOL(ax25_find_cb); -EXPORT_SYMBOL(ax25_linkfail_register); -EXPORT_SYMBOL(ax25_linkfail_release); -EXPORT_SYMBOL(ax25_listen_register); -EXPORT_SYMBOL(ax25_listen_release); -EXPORT_SYMBOL(ax25_protocol_register); -EXPORT_SYMBOL(ax25_protocol_release); -EXPORT_SYMBOL(ax25_send_frame); -EXPORT_SYMBOL(ax25_uid_policy); -EXPORT_SYMBOL(ax25cmp); -EXPORT_SYMBOL(ax2asc); -EXPORT_SYMBOL(asc2ax); -EXPORT_SYMBOL(null_ax25_address); -EXPORT_SYMBOL(ax25_display_timer); - static int __init ax25_init(void) { int rc = proto_register(&ax25_proto, 0); diff --git a/net/ax25/ax25_addr.c b/net/ax25/ax25_addr.c index 0164a155b8c..5f0896ad004 100644 --- a/net/ax25/ax25_addr.c +++ b/net/ax25/ax25_addr.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -33,6 +34,8 @@ */ ax25_address null_ax25_address = {{0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x00}}; +EXPORT_SYMBOL(null_ax25_address); + /* * ax25 -> ascii conversion */ @@ -64,6 +67,8 @@ char *ax2asc(char *buf, ax25_address *a) } +EXPORT_SYMBOL(ax2asc); + /* * ascii -> ax25 conversion */ @@ -97,6 +102,8 @@ void asc2ax(ax25_address *addr, char *callsign) addr->ax25_call[6] &= 0x1E; } +EXPORT_SYMBOL(asc2ax); + /* * Compare two ax.25 addresses */ @@ -116,6 +123,8 @@ int ax25cmp(ax25_address *a, ax25_address *b) return 2; /* Partial match */ } +EXPORT_SYMBOL(ax25cmp); + /* * Compare two AX.25 digipeater paths. */ diff --git a/net/ax25/ax25_iface.c b/net/ax25/ax25_iface.c index d68aff10072..3bb152710b7 100644 --- a/net/ax25/ax25_iface.c +++ b/net/ax25/ax25_iface.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -74,6 +75,8 @@ int ax25_protocol_register(unsigned int pid, return 1; } +EXPORT_SYMBOL(ax25_protocol_register); + void ax25_protocol_release(unsigned int pid) { struct protocol_struct *s, *protocol; @@ -106,6 +109,8 @@ void ax25_protocol_release(unsigned int pid) write_unlock(&protocol_list_lock); } +EXPORT_SYMBOL(ax25_protocol_release); + int ax25_linkfail_register(void (*func)(ax25_cb *, int)) { struct linkfail_struct *linkfail; @@ -123,6 +128,8 @@ int ax25_linkfail_register(void (*func)(ax25_cb *, int)) return 1; } +EXPORT_SYMBOL(ax25_linkfail_register); + void ax25_linkfail_release(void (*func)(ax25_cb *, int)) { struct linkfail_struct *s, *linkfail; @@ -155,6 +162,8 @@ void ax25_linkfail_release(void (*func)(ax25_cb *, int)) spin_unlock_bh(&linkfail_lock); } +EXPORT_SYMBOL(ax25_linkfail_release); + int ax25_listen_register(ax25_address *callsign, struct net_device *dev) { struct listen_struct *listen; @@ -176,6 +185,8 @@ int ax25_listen_register(ax25_address *callsign, struct net_device *dev) return 1; } +EXPORT_SYMBOL(ax25_listen_register); + void ax25_listen_release(ax25_address *callsign, struct net_device *dev) { struct listen_struct *s, *listen; @@ -208,6 +219,8 @@ void ax25_listen_release(ax25_address *callsign, struct net_device *dev) spin_unlock_bh(&listen_lock); } +EXPORT_SYMBOL(ax25_listen_release); + int (*ax25_protocol_function(unsigned int pid))(struct sk_buff *, ax25_cb *) { int (*res)(struct sk_buff *, ax25_cb *) = NULL; diff --git a/net/ax25/ax25_ip.c b/net/ax25/ax25_ip.c index d643dac3ecc..a0b534f80f1 100644 --- a/net/ax25/ax25_ip.c +++ b/net/ax25/ax25_ip.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -221,3 +222,5 @@ int ax25_rebuild_header(struct sk_buff *skb) #endif +EXPORT_SYMBOL(ax25_hard_header); +EXPORT_SYMBOL(ax25_rebuild_header); diff --git a/net/ax25/ax25_out.c b/net/ax25/ax25_out.c index 5fc048dcd39..5d99852b239 100644 --- a/net/ax25/ax25_out.c +++ b/net/ax25/ax25_out.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -104,6 +105,8 @@ ax25_cb *ax25_send_frame(struct sk_buff *skb, int paclen, ax25_address *src, ax2 return ax25; /* We had to create it */ } +EXPORT_SYMBOL(ax25_send_frame); + /* * All outgoing AX.25 I frames pass via this routine. Therefore this is * where the fragmentation of frames takes place. If fragment is set to diff --git a/net/ax25/ax25_timer.c b/net/ax25/ax25_timer.c index 7a6b50a1455..ec254057f21 100644 --- a/net/ax25/ax25_timer.c +++ b/net/ax25/ax25_timer.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -137,6 +138,8 @@ unsigned long ax25_display_timer(struct timer_list *timer) return timer->expires - jiffies; } +EXPORT_SYMBOL(ax25_display_timer); + static void ax25_heartbeat_expiry(unsigned long param) { int proto = AX25_PROTO_STD_SIMPLEX; diff --git a/net/ax25/ax25_uid.c b/net/ax25/ax25_uid.c index b8b5854bce9..5e9a81e8b21 100644 --- a/net/ax25/ax25_uid.c +++ b/net/ax25/ax25_uid.c @@ -49,6 +49,8 @@ static DEFINE_RWLOCK(ax25_uid_lock); int ax25_uid_policy = 0; +EXPORT_SYMBOL(ax25_uid_policy); + ax25_uid_assoc *ax25_findbyuid(uid_t uid) { ax25_uid_assoc *ax25_uid, *res = NULL; @@ -67,6 +69,8 @@ ax25_uid_assoc *ax25_findbyuid(uid_t uid) return res; } +EXPORT_SYMBOL(ax25_findbyuid); + int ax25_uid_ioctl(int cmd, struct sockaddr_ax25 *sax) { ax25_uid_assoc *ax25_uid; -- cgit v1.2.3 From 4cc7c2734e2b4032103e47d8f3e8b6fa3360d3f1 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 3 May 2006 23:26:20 -0700 Subject: [ROSE]: Fix routing table locking in rose_remove_neigh. The locking rule for rose_remove_neigh() are that the caller needs to hold rose_neigh_list_lock, so we better don't take it yet again in rose_neigh_list_lock. Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- net/rose/rose_route.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'net') diff --git a/net/rose/rose_route.c b/net/rose/rose_route.c index 4cb6bfa6fcb..a22542fa1bc 100644 --- a/net/rose/rose_route.c +++ b/net/rose/rose_route.c @@ -233,11 +233,8 @@ static void rose_remove_neigh(struct rose_neigh *rose_neigh) skb_queue_purge(&rose_neigh->queue); - spin_lock_bh(&rose_neigh_list_lock); - if ((s = rose_neigh_list) == rose_neigh) { rose_neigh_list = rose_neigh->next; - spin_unlock_bh(&rose_neigh_list_lock); kfree(rose_neigh->digipeat); kfree(rose_neigh); return; @@ -246,7 +243,6 @@ static void rose_remove_neigh(struct rose_neigh *rose_neigh) while (s != NULL && s->next != NULL) { if (s->next == rose_neigh) { s->next = rose_neigh->next; - spin_unlock_bh(&rose_neigh_list_lock); kfree(rose_neigh->digipeat); kfree(rose_neigh); return; @@ -254,7 +250,6 @@ static void rose_remove_neigh(struct rose_neigh *rose_neigh) s = s->next; } - spin_unlock_bh(&rose_neigh_list_lock); } /* -- cgit v1.2.3 From e1fdb5b39656ea2be8cadde565e543649a988af9 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 3 May 2006 23:27:16 -0700 Subject: [AX.25]: Eleminate HZ from AX.25 kernel interfaces Convert all AX.25 sysctl time values from jiffies to ms as units. Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- net/ax25/af_ax25.c | 73 ++++++++++++++++++++++++++-------------------- net/ax25/ax25_ds_timer.c | 3 +- net/ax25/sysctl_net_ax25.c | 10 +++---- 3 files changed, 48 insertions(+), 38 deletions(-) (limited to 'net') diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index a9f13dfde07..a2e0dd047e9 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -426,6 +426,26 @@ static int ax25_ctl_ioctl(const unsigned int cmd, void __user *arg) return 0; } +static void ax25_fillin_cb_from_dev(ax25_cb *ax25, ax25_dev *ax25_dev) +{ + ax25->rtt = msecs_to_jiffies(ax25_dev->values[AX25_VALUES_T1]) / 2; + ax25->t1 = msecs_to_jiffies(ax25_dev->values[AX25_VALUES_T1]); + ax25->t2 = msecs_to_jiffies(ax25_dev->values[AX25_VALUES_T2]); + ax25->t3 = msecs_to_jiffies(ax25_dev->values[AX25_VALUES_T3]); + ax25->n2 = ax25_dev->values[AX25_VALUES_N2]; + ax25->paclen = ax25_dev->values[AX25_VALUES_PACLEN]; + ax25->idle = msecs_to_jiffies(ax25_dev->values[AX25_VALUES_IDLE]); + ax25->backoff = ax25_dev->values[AX25_VALUES_BACKOFF]; + + if (ax25_dev->values[AX25_VALUES_AXDEFMODE]) { + ax25->modulus = AX25_EMODULUS; + ax25->window = ax25_dev->values[AX25_VALUES_EWINDOW]; + } else { + ax25->modulus = AX25_MODULUS; + ax25->window = ax25_dev->values[AX25_VALUES_WINDOW]; + } +} + /* * Fill in a created AX.25 created control block with the default * values for a particular device. @@ -435,39 +455,28 @@ void ax25_fillin_cb(ax25_cb *ax25, ax25_dev *ax25_dev) ax25->ax25_dev = ax25_dev; if (ax25->ax25_dev != NULL) { - ax25->rtt = ax25_dev->values[AX25_VALUES_T1] / 2; - ax25->t1 = ax25_dev->values[AX25_VALUES_T1]; - ax25->t2 = ax25_dev->values[AX25_VALUES_T2]; - ax25->t3 = ax25_dev->values[AX25_VALUES_T3]; - ax25->n2 = ax25_dev->values[AX25_VALUES_N2]; - ax25->paclen = ax25_dev->values[AX25_VALUES_PACLEN]; - ax25->idle = ax25_dev->values[AX25_VALUES_IDLE]; - ax25->backoff = ax25_dev->values[AX25_VALUES_BACKOFF]; - - if (ax25_dev->values[AX25_VALUES_AXDEFMODE]) { - ax25->modulus = AX25_EMODULUS; - ax25->window = ax25_dev->values[AX25_VALUES_EWINDOW]; - } else { - ax25->modulus = AX25_MODULUS; - ax25->window = ax25_dev->values[AX25_VALUES_WINDOW]; - } + ax25_fillin_cb_from_dev(ax25, ax25_dev); + return; + } + + /* + * No device, use kernel / AX.25 spec default values + */ + ax25->rtt = msecs_to_jiffies(AX25_DEF_T1) / 2; + ax25->t1 = msecs_to_jiffies(AX25_DEF_T1); + ax25->t2 = msecs_to_jiffies(AX25_DEF_T2); + ax25->t3 = msecs_to_jiffies(AX25_DEF_T3); + ax25->n2 = AX25_DEF_N2; + ax25->paclen = AX25_DEF_PACLEN; + ax25->idle = msecs_to_jiffies(AX25_DEF_IDLE); + ax25->backoff = AX25_DEF_BACKOFF; + + if (AX25_DEF_AXDEFMODE) { + ax25->modulus = AX25_EMODULUS; + ax25->window = AX25_DEF_EWINDOW; } else { - ax25->rtt = AX25_DEF_T1 / 2; - ax25->t1 = AX25_DEF_T1; - ax25->t2 = AX25_DEF_T2; - ax25->t3 = AX25_DEF_T3; - ax25->n2 = AX25_DEF_N2; - ax25->paclen = AX25_DEF_PACLEN; - ax25->idle = AX25_DEF_IDLE; - ax25->backoff = AX25_DEF_BACKOFF; - - if (AX25_DEF_AXDEFMODE) { - ax25->modulus = AX25_EMODULUS; - ax25->window = AX25_DEF_EWINDOW; - } else { - ax25->modulus = AX25_MODULUS; - ax25->window = AX25_DEF_WINDOW; - } + ax25->modulus = AX25_MODULUS; + ax25->window = AX25_DEF_WINDOW; } } diff --git a/net/ax25/ax25_ds_timer.c b/net/ax25/ax25_ds_timer.c index 061083efc1d..5961459935e 100644 --- a/net/ax25/ax25_ds_timer.c +++ b/net/ax25/ax25_ds_timer.c @@ -61,7 +61,8 @@ void ax25_ds_set_timer(ax25_dev *ax25_dev) return; del_timer(&ax25_dev->dama.slave_timer); - ax25_dev->dama.slave_timeout = ax25_dev->values[AX25_VALUES_DS_TIMEOUT] / 10; + ax25_dev->dama.slave_timeout = + msecs_to_jiffies(ax25_dev->values[AX25_VALUES_DS_TIMEOUT]) / 10; ax25_ds_add_timer(ax25_dev); } diff --git a/net/ax25/sysctl_net_ax25.c b/net/ax25/sysctl_net_ax25.c index 894a22558d9..bdb64c36df1 100644 --- a/net/ax25/sysctl_net_ax25.c +++ b/net/ax25/sysctl_net_ax25.c @@ -18,14 +18,14 @@ static int min_backoff[1], max_backoff[] = {2}; static int min_conmode[1], max_conmode[] = {2}; static int min_window[] = {1}, max_window[] = {7}; static int min_ewindow[] = {1}, max_ewindow[] = {63}; -static int min_t1[] = {1}, max_t1[] = {30 * HZ}; -static int min_t2[] = {1}, max_t2[] = {20 * HZ}; -static int min_t3[1], max_t3[] = {3600 * HZ}; -static int min_idle[1], max_idle[] = {65535 * HZ}; +static int min_t1[] = {1}, max_t1[] = {30000}; +static int min_t2[] = {1}, max_t2[] = {20000}; +static int min_t3[1], max_t3[] = {3600000}; +static int min_idle[1], max_idle[] = {65535000}; static int min_n2[] = {1}, max_n2[] = {31}; static int min_paclen[] = {1}, max_paclen[] = {512}; static int min_proto[1], max_proto[] = { AX25_PROTO_MAX }; -static int min_ds_timeout[1], max_ds_timeout[] = {65535 * HZ}; +static int min_ds_timeout[1], max_ds_timeout[] = {65535000}; static struct ctl_table_header *ax25_table_header; -- cgit v1.2.3 From 4d8937d0b113e8ec39f7d18cf13804f3b5fb8fd4 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 3 May 2006 23:27:47 -0700 Subject: [NETROM]: Eleminate HZ from NET/ROM kernel interfaces Convert all NET/ROM sysctl time values from jiffies to ms as units. Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- net/netrom/af_netrom.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c index d44981f5a61..ecd288beca7 100644 --- a/net/netrom/af_netrom.c +++ b/net/netrom/af_netrom.c @@ -425,11 +425,16 @@ static int nr_create(struct socket *sock, int protocol) nr_init_timers(sk); - nr->t1 = sysctl_netrom_transport_timeout; - nr->t2 = sysctl_netrom_transport_acknowledge_delay; - nr->n2 = sysctl_netrom_transport_maximum_tries; - nr->t4 = sysctl_netrom_transport_busy_delay; - nr->idle = sysctl_netrom_transport_no_activity_timeout; + nr->t1 = + msecs_to_jiffies(sysctl_netrom_transport_timeout); + nr->t2 = + msecs_to_jiffies(sysctl_netrom_transport_acknowledge_delay); + nr->n2 = + msecs_to_jiffies(sysctl_netrom_transport_maximum_tries); + nr->t4 = + msecs_to_jiffies(sysctl_netrom_transport_busy_delay); + nr->idle = + msecs_to_jiffies(sysctl_netrom_transport_no_activity_timeout); nr->window = sysctl_netrom_transport_requested_window_size; nr->bpqext = 1; -- cgit v1.2.3 From 82e84249f0ee098e004c8bd6d90a1640bd56cfbb Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 3 May 2006 23:28:20 -0700 Subject: [ROSE]: Eleminate HZ from ROSE kernel interfaces Convert all ROSE sysctl time values from jiffies to ms as units. Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- net/rose/af_rose.c | 10 +++++----- net/rose/rose_link.c | 6 ++++-- 2 files changed, 9 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index ea65396d161..ef4538ac84f 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -518,11 +518,11 @@ static int rose_create(struct socket *sock, int protocol) init_timer(&rose->timer); init_timer(&rose->idletimer); - rose->t1 = sysctl_rose_call_request_timeout; - rose->t2 = sysctl_rose_reset_request_timeout; - rose->t3 = sysctl_rose_clear_request_timeout; - rose->hb = sysctl_rose_ack_hold_back_timeout; - rose->idle = sysctl_rose_no_activity_timeout; + rose->t1 = msecs_to_jiffies(sysctl_rose_call_request_timeout); + rose->t2 = msecs_to_jiffies(sysctl_rose_reset_request_timeout); + rose->t3 = msecs_to_jiffies(sysctl_rose_clear_request_timeout); + rose->hb = msecs_to_jiffies(sysctl_rose_ack_hold_back_timeout); + rose->idle = msecs_to_jiffies(sysctl_rose_no_activity_timeout); rose->state = ROSE_STATE_0; diff --git a/net/rose/rose_link.c b/net/rose/rose_link.c index 09e9e9d04d9..bd86a63960c 100644 --- a/net/rose/rose_link.c +++ b/net/rose/rose_link.c @@ -40,7 +40,8 @@ void rose_start_ftimer(struct rose_neigh *neigh) neigh->ftimer.data = (unsigned long)neigh; neigh->ftimer.function = &rose_ftimer_expiry; - neigh->ftimer.expires = jiffies + sysctl_rose_link_fail_timeout; + neigh->ftimer.expires = + jiffies + msecs_to_jiffies(sysctl_rose_link_fail_timeout); add_timer(&neigh->ftimer); } @@ -51,7 +52,8 @@ static void rose_start_t0timer(struct rose_neigh *neigh) neigh->t0timer.data = (unsigned long)neigh; neigh->t0timer.function = &rose_t0timer_expiry; - neigh->t0timer.expires = jiffies + sysctl_rose_restart_request_timeout; + neigh->t0timer.expires = + jiffies + msecs_to_jiffies(sysctl_rose_restart_request_timeout); add_timer(&neigh->t0timer); } -- cgit v1.2.3 From 75c2d9077c63ac21488129cc23561d4f4fd0f5e5 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 3 May 2006 23:31:35 -0700 Subject: [TCP]: Fix sock_orphan dead lock Calling sock_orphan inside bh_lock_sock in tcp_close can lead to dead locks. For example, the inet_diag code holds sk_callback_lock without disabling BH. If an inbound packet arrives during that admittedly tiny window, it will cause a dead lock on bh_lock_sock. Another possible path would be through sock_wfree if the network device driver frees the tx skb in process context with BH enabled. We can fix this by moving sock_orphan out of bh_lock_sock. The tricky bit is to work out when we need to destroy the socket ourselves and when it has already been destroyed by someone else. By moving sock_orphan before the release_sock we can solve this problem. This is because as long as we own the socket lock its state cannot change. So we simply record the socket state before the release_sock and then check the state again after we regain the socket lock. If the socket state has transitioned to TCP_CLOSE in the time being, we know that the socket has been destroyed. Otherwise the socket is still ours to keep. Note that I've also moved the increment on the orphan count forward. This may look like a problem as we're increasing it even if the socket is just about to be destroyed where it'll be decreased again. However, this simply enlarges a window that already exists. This also changes the orphan count test by one. Considering what the orphan count is meant to do this is no big deal. This problem was discoverd by Ingo Molnar using his lock validator. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 87f68e787d0..e2b7b805503 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1468,6 +1468,7 @@ void tcp_close(struct sock *sk, long timeout) { struct sk_buff *skb; int data_was_unread = 0; + int state; lock_sock(sk); sk->sk_shutdown = SHUTDOWN_MASK; @@ -1544,6 +1545,11 @@ void tcp_close(struct sock *sk, long timeout) sk_stream_wait_close(sk, timeout); adjudge_to_death: + state = sk->sk_state; + sock_hold(sk); + sock_orphan(sk); + atomic_inc(sk->sk_prot->orphan_count); + /* It is the last release_sock in its life. It will remove backlog. */ release_sock(sk); @@ -1555,8 +1561,9 @@ adjudge_to_death: bh_lock_sock(sk); BUG_TRAP(!sock_owned_by_user(sk)); - sock_hold(sk); - sock_orphan(sk); + /* Have we already been destroyed by a softirq or backlog? */ + if (state != TCP_CLOSE && sk->sk_state == TCP_CLOSE) + goto out; /* This is a (useful) BSD violating of the RFC. There is a * problem with TCP as specified in that the other end could @@ -1584,7 +1591,6 @@ adjudge_to_death: if (tmo > TCP_TIMEWAIT_LEN) { inet_csk_reset_keepalive_timer(sk, tcp_fin_time(sk)); } else { - atomic_inc(sk->sk_prot->orphan_count); tcp_time_wait(sk, TCP_FIN_WAIT2, tmo); goto out; } @@ -1603,7 +1609,6 @@ adjudge_to_death: NET_INC_STATS_BH(LINUX_MIB_TCPABORTONMEMORY); } } - atomic_inc(sk->sk_prot->orphan_count); if (sk->sk_state == TCP_CLOSE) inet_csk_destroy_sock(sk); -- cgit v1.2.3 From d1a649838802edd94b6335834919463c6ae61f40 Mon Sep 17 00:00:00 2001 From: Patrick Caulfield Date: Wed, 3 May 2006 23:36:23 -0700 Subject: [DECNET]: Fix level1 router hello This patch fixes hello messages sent when a node is a level 1 router. Slightly contrary to the spec (maybe) VMS ignores hello messages that do not name level2 routers that it also knows about. So, here we simply name all the routers that the node knows about rather just other level1 routers. (I hope the patch is clearer than the description. sorry). Signed-off-by: Patrick Caulfield Signed-off-by: David S. Miller --- net/decnet/dn_neigh.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'net') diff --git a/net/decnet/dn_neigh.c b/net/decnet/dn_neigh.c index 7c8692c26bf..66e230c3b32 100644 --- a/net/decnet/dn_neigh.c +++ b/net/decnet/dn_neigh.c @@ -493,7 +493,6 @@ struct elist_cb_state { static void neigh_elist_cb(struct neighbour *neigh, void *_info) { struct elist_cb_state *s = _info; - struct dn_dev *dn_db; struct dn_neigh *dn; if (neigh->dev != s->dev) @@ -503,10 +502,6 @@ static void neigh_elist_cb(struct neighbour *neigh, void *_info) if (!(dn->flags & (DN_NDFLAG_R1|DN_NDFLAG_R2))) return; - dn_db = (struct dn_dev *) s->dev->dn_ptr; - if (dn_db->parms.forwarding == 1 && (dn->flags & DN_NDFLAG_R2)) - return; - if (s->t == s->n) s->rs = dn_find_slot(s->ptr, s->n, dn->priority); else -- cgit v1.2.3 From 995c99268e0b12eb3c8939211ba5368dd92d98d9 Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Sun, 30 Apr 2006 19:49:30 +0100 Subject: [PATCH] softmac: don't reassociate if user asked for deauthentication When wpa_supplicant exits, it uses SIOCSIWMLME to request deauthentication. softmac then tries to reassociate without any user intervention, which isn't the desired behaviour of this signal. This change makes softmac only attempt reassociation if the remote network itself deauthenticated us. Signed-off-by: Daniel Drake Acked-by: Johannes Berg Signed-off-by: John W. Linville --- net/ieee80211/softmac/ieee80211softmac_auth.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ieee80211/softmac/ieee80211softmac_auth.c b/net/ieee80211/softmac/ieee80211softmac_auth.c index 9a0eac6c61e..d6a04f3ab86 100644 --- a/net/ieee80211/softmac/ieee80211softmac_auth.c +++ b/net/ieee80211/softmac/ieee80211softmac_auth.c @@ -298,8 +298,6 @@ ieee80211softmac_deauth_from_net(struct ieee80211softmac_device *mac, /* can't transmit data right now... */ netif_carrier_off(mac->dev); - /* let's try to re-associate */ - schedule_work(&mac->associnfo.work); spin_unlock_irqrestore(&mac->lock, flags); } @@ -360,5 +358,8 @@ ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *de } ieee80211softmac_deauth_from_net(mac, net); + + /* let's try to re-associate */ + schedule_work(&mac->associnfo.work); return 0; } -- cgit v1.2.3 From d57336e3f2dd7c2d1fbe4a8323029869fb6e1f00 Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Sun, 30 Apr 2006 22:09:07 +0100 Subject: [PATCH] softmac: make non-operational after being stopped zd1211 with softmac and wpa_supplicant revealed an issue with softmac and the use of workqueues. Some of the work functions actually reschedule themselves, so this meant that there could still be pending work after flush_scheduled_work() had been called during ieee80211softmac_stop(). This patch introduces a "running" flag which is used to ensure that rescheduling does not happen in this situation. I also used this flag to ensure that softmac's hooks into ieee80211 are non-operational once the stop operation has been started. This simply makes softmac a little more robust, because I could crash it easily by receiving frames in the short timeframe after shutting down softmac and before turning off the ZD1211 radio. (ZD1211 is now fixed as well!) Signed-off-by: Daniel Drake Acked-by: Johannes Berg Signed-off-by: John W. Linville --- net/ieee80211/softmac/ieee80211softmac_assoc.c | 17 +++++++++++++++-- net/ieee80211/softmac/ieee80211softmac_auth.c | 11 +++++++++++ net/ieee80211/softmac/ieee80211softmac_module.c | 4 ++++ net/ieee80211/softmac/ieee80211softmac_scan.c | 8 ++++++++ 4 files changed, 38 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ieee80211/softmac/ieee80211softmac_assoc.c b/net/ieee80211/softmac/ieee80211softmac_assoc.c index fb79ce7d643..57ea9f6f465 100644 --- a/net/ieee80211/softmac/ieee80211softmac_assoc.c +++ b/net/ieee80211/softmac/ieee80211softmac_assoc.c @@ -51,11 +51,12 @@ ieee80211softmac_assoc(struct ieee80211softmac_device *mac, struct ieee80211soft spin_lock_irqsave(&mac->lock, flags); mac->associnfo.associating = 1; mac->associated = 0; /* just to make sure */ - spin_unlock_irqrestore(&mac->lock, flags); /* Set a timer for timeout */ /* FIXME: make timeout configurable */ - schedule_delayed_work(&mac->associnfo.timeout, 5 * HZ); + if (likely(mac->running)) + schedule_delayed_work(&mac->associnfo.timeout, 5 * HZ); + spin_unlock_irqrestore(&mac->lock, flags); } void @@ -319,6 +320,9 @@ ieee80211softmac_handle_assoc_response(struct net_device * dev, u16 status = le16_to_cpup(&resp->status); struct ieee80211softmac_network *network = NULL; unsigned long flags; + + if (unlikely(!mac->running)) + return -ENODEV; spin_lock_irqsave(&mac->lock, flags); @@ -377,10 +381,16 @@ ieee80211softmac_handle_disassoc(struct net_device * dev, { struct ieee80211softmac_device *mac = ieee80211_priv(dev); unsigned long flags; + + if (unlikely(!mac->running)) + return -ENODEV; + if (memcmp(disassoc->header.addr2, mac->associnfo.bssid, ETH_ALEN)) return 0; + if (memcmp(disassoc->header.addr1, mac->dev->dev_addr, ETH_ALEN)) return 0; + dprintk(KERN_INFO PFX "got disassoc frame\n"); netif_carrier_off(dev); spin_lock_irqsave(&mac->lock, flags); @@ -400,6 +410,9 @@ ieee80211softmac_handle_reassoc_req(struct net_device * dev, struct ieee80211softmac_device *mac = ieee80211_priv(dev); struct ieee80211softmac_network *network; + if (unlikely(!mac->running)) + return -ENODEV; + network = ieee80211softmac_get_network_by_bssid(mac, resp->header.addr3); if (!network) { dprintkl(KERN_INFO PFX "reassoc request from unknown network\n"); diff --git a/net/ieee80211/softmac/ieee80211softmac_auth.c b/net/ieee80211/softmac/ieee80211softmac_auth.c index d6a04f3ab86..06e33262466 100644 --- a/net/ieee80211/softmac/ieee80211softmac_auth.c +++ b/net/ieee80211/softmac/ieee80211softmac_auth.c @@ -86,6 +86,11 @@ ieee80211softmac_auth_queue(void *data) /* Lock and set flags */ spin_lock_irqsave(&mac->lock, flags); + if (unlikely(!mac->running)) { + /* Prevent reschedule on workqueue flush */ + spin_unlock_irqrestore(&mac->lock, flags); + return; + } net->authenticated = 0; net->authenticating = 1; /* add a timeout call so we eventually give up waiting for an auth reply */ @@ -124,6 +129,9 @@ ieee80211softmac_auth_resp(struct net_device *dev, struct ieee80211_auth *auth) unsigned long flags; u8 * data; + if (unlikely(!mac->running)) + return -ENODEV; + /* Find correct auth queue item */ spin_lock_irqsave(&mac->lock, flags); list_for_each(list_ptr, &mac->auth_queue) { @@ -336,6 +344,9 @@ ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *de struct ieee80211softmac_network *net = NULL; struct ieee80211softmac_device *mac = ieee80211_priv(dev); + if (unlikely(!mac->running)) + return -ENODEV; + if (!deauth) { dprintk("deauth without deauth packet. eek!\n"); return 0; diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c index be83bdc1644..6252be2c0db 100644 --- a/net/ieee80211/softmac/ieee80211softmac_module.c +++ b/net/ieee80211/softmac/ieee80211softmac_module.c @@ -89,6 +89,8 @@ ieee80211softmac_clear_pending_work(struct ieee80211softmac_device *sm) ieee80211softmac_wait_for_scan(sm); spin_lock_irqsave(&sm->lock, flags); + sm->running = 0; + /* Free all pending assoc work items */ cancel_delayed_work(&sm->associnfo.work); @@ -204,6 +206,8 @@ void ieee80211softmac_start(struct net_device *dev) assert(0); if (mac->txrates_change) mac->txrates_change(dev, change, &oldrates); + + mac->running = 1; } EXPORT_SYMBOL_GPL(ieee80211softmac_start); diff --git a/net/ieee80211/softmac/ieee80211softmac_scan.c b/net/ieee80211/softmac/ieee80211softmac_scan.c index 2b9e7edfa3c..d31cf77498c 100644 --- a/net/ieee80211/softmac/ieee80211softmac_scan.c +++ b/net/ieee80211/softmac/ieee80211softmac_scan.c @@ -115,7 +115,15 @@ void ieee80211softmac_scan(void *d) // TODO: is this if correct, or should we do this only if scanning from assoc request? if (sm->associnfo.req_essid.len) ieee80211softmac_send_mgt_frame(sm, &sm->associnfo.req_essid, IEEE80211_STYPE_PROBE_REQ, 0); + + spin_lock_irqsave(&sm->lock, flags); + if (unlikely(!sm->running)) { + /* Prevent reschedule on workqueue flush */ + spin_unlock_irqrestore(&sm->lock, flags); + break; + } schedule_delayed_work(&si->softmac_scan, IEEE80211SOFTMAC_PROBE_DELAY); + spin_unlock_irqrestore(&sm->lock, flags); return; } else { dprintk(PFX "Not probing Channel %d (not allowed here)\n", si->channels[current_channel_idx].channel); -- cgit v1.2.3 From 7c3ceb4fb9667f34f1599a062efecf4cdc4a4ce5 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Fri, 5 May 2006 17:02:09 -0700 Subject: [SCTP]: Allow spillover of receive buffer to avoid deadlock. This patch fixes a deadlock situation in the receive path by allowing temporary spillover of the receive buffer. - If the chunk we receive has a tsn that immediately follows the ctsn, accept it even if we run out of receive buffer space and renege data with higher TSNs. - Once we accept one chunk in a packet, accept all the remaining chunks even if we run out of receive buffer space. Signed-off-by: Neil Horman Acked-by: Mark Butler Acked-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- net/sctp/inqueue.c | 1 + net/sctp/sm_statefuns.c | 46 ++++++++++++++++++++++++++++++++++++---------- 2 files changed, 37 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index 297b8951463..cf0c767d43a 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -149,6 +149,7 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) /* This is the first chunk in the packet. */ chunk->singleton = 1; ch = (sctp_chunkhdr_t *) chunk->skb->data; + chunk->data_accepted = 0; } chunk->chunk_hdr = ch; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 2b9a832b29a..f5d131f52a7 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -5151,7 +5151,9 @@ static int sctp_eat_data(const struct sctp_association *asoc, int tmp; __u32 tsn; int account_value; + struct sctp_tsnmap *map = (struct sctp_tsnmap *)&asoc->peer.tsn_map; struct sock *sk = asoc->base.sk; + int rcvbuf_over = 0; data_hdr = chunk->subh.data_hdr = (sctp_datahdr_t *)chunk->skb->data; skb_pull(chunk->skb, sizeof(sctp_datahdr_t)); @@ -5162,10 +5164,16 @@ static int sctp_eat_data(const struct sctp_association *asoc, /* ASSERT: Now skb->data is really the user data. */ /* - * if we are established, and we have used up our receive - * buffer memory, drop the frame - */ - if (asoc->state == SCTP_STATE_ESTABLISHED) { + * If we are established, and we have used up our receive buffer + * memory, think about droping the frame. + * Note that we have an opportunity to improve performance here. + * If we accept one chunk from an skbuff, we have to keep all the + * memory of that skbuff around until the chunk is read into user + * space. Therefore, once we accept 1 chunk we may as well accept all + * remaining chunks in the skbuff. The data_accepted flag helps us do + * that. + */ + if ((asoc->state == SCTP_STATE_ESTABLISHED) && (!chunk->data_accepted)) { /* * If the receive buffer policy is 1, then each * association can allocate up to sk_rcvbuf bytes @@ -5176,9 +5184,25 @@ static int sctp_eat_data(const struct sctp_association *asoc, account_value = atomic_read(&asoc->rmem_alloc); else account_value = atomic_read(&sk->sk_rmem_alloc); - - if (account_value > sk->sk_rcvbuf) - return SCTP_IERROR_IGNORE_TSN; + if (account_value > sk->sk_rcvbuf) { + /* + * We need to make forward progress, even when we are + * under memory pressure, so we always allow the + * next tsn after the ctsn ack point to be accepted. + * This lets us avoid deadlocks in which we have to + * drop frames that would otherwise let us drain the + * receive queue. + */ + if ((sctp_tsnmap_get_ctsn(map) + 1) != tsn) + return SCTP_IERROR_IGNORE_TSN; + + /* + * We're going to accept the frame but we should renege + * to make space for it. This will send us down that + * path later in this function. + */ + rcvbuf_over = 1; + } } /* Process ECN based congestion. @@ -5226,6 +5250,7 @@ static int sctp_eat_data(const struct sctp_association *asoc, datalen -= sizeof(sctp_data_chunk_t); deliver = SCTP_CMD_CHUNK_ULP; + chunk->data_accepted = 1; /* Think about partial delivery. */ if ((datalen >= asoc->rwnd) && (!asoc->ulpq.pd_mode)) { @@ -5242,7 +5267,8 @@ static int sctp_eat_data(const struct sctp_association *asoc, * large spill over. */ if (!asoc->rwnd || asoc->rwnd_over || - (datalen > asoc->rwnd + asoc->frag_point)) { + (datalen > asoc->rwnd + asoc->frag_point) || + rcvbuf_over) { /* If this is the next TSN, consider reneging to make * room. Note: Playing nice with a confused sender. A @@ -5250,8 +5276,8 @@ static int sctp_eat_data(const struct sctp_association *asoc, * space and in the future we may want to detect and * do more drastic reneging. */ - if (sctp_tsnmap_has_gap(&asoc->peer.tsn_map) && - (sctp_tsnmap_get_ctsn(&asoc->peer.tsn_map) + 1) == tsn) { + if (sctp_tsnmap_has_gap(map) && + (sctp_tsnmap_get_ctsn(map) + 1) == tsn) { SCTP_DEBUG_PRINTK("Reneging for tsn:%u\n", tsn); deliver = SCTP_CMD_RENEGE; } else { -- cgit v1.2.3 From 672e7cca17ed6036a1756ed34cf20dbd72d5e5f6 Mon Sep 17 00:00:00 2001 From: Vladislav Yasevich Date: Fri, 5 May 2006 17:03:49 -0700 Subject: [SCTP]: Prevent possible infinite recursion with multiple bundled DATA. There is a rare situation that causes lksctp to go into infinite recursion and crash the system. The trigger is a packet that contains at least the first two DATA fragments of a message bundled together. The recursion is triggered when the user data buffer is smaller that the full data message. The problem is that we clone the skb for every fragment in the message. When reassembling the full message, we try to link skbs from the "first fragment" clone using the frag_list. However, since the frag_list is shared between two clones in this rare situation, we end up setting the frag_list pointer of the second fragment to point to itself. This causes sctp_skb_pull() to potentially recurse indefinitely. Proposed solution is to make a copy of the skb when attempting to link things using frag_list. Signed-off-by: Vladislav Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- net/sctp/ulpqueue.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sctp/ulpqueue.c b/net/sctp/ulpqueue.c index 2080b2d28c9..575e556aeb3 100644 --- a/net/sctp/ulpqueue.c +++ b/net/sctp/ulpqueue.c @@ -279,6 +279,7 @@ static inline void sctp_ulpq_store_reasm(struct sctp_ulpq *ulpq, static struct sctp_ulpevent *sctp_make_reassembled_event(struct sk_buff_head *queue, struct sk_buff *f_frag, struct sk_buff *l_frag) { struct sk_buff *pos; + struct sk_buff *new = NULL; struct sctp_ulpevent *event; struct sk_buff *pnext, *last; struct sk_buff *list = skb_shinfo(f_frag)->frag_list; @@ -297,11 +298,33 @@ static struct sctp_ulpevent *sctp_make_reassembled_event(struct sk_buff_head *qu */ if (last) last->next = pos; - else - skb_shinfo(f_frag)->frag_list = pos; + else { + if (skb_cloned(f_frag)) { + /* This is a cloned skb, we can't just modify + * the frag_list. We need a new skb to do that. + * Instead of calling skb_unshare(), we'll do it + * ourselves since we need to delay the free. + */ + new = skb_copy(f_frag, GFP_ATOMIC); + if (!new) + return NULL; /* try again later */ + + new->sk = f_frag->sk; + + skb_shinfo(new)->frag_list = pos; + } else + skb_shinfo(f_frag)->frag_list = pos; + } /* Remove the first fragment from the reassembly queue. */ __skb_unlink(f_frag, queue); + + /* if we did unshare, then free the old skb and re-assign */ + if (new) { + kfree_skb(f_frag); + f_frag = new; + } + while (pos) { pnext = pos->next; -- cgit v1.2.3 From 62b08083ec3dbfd7e533c8d230dd1d8191a6e813 Mon Sep 17 00:00:00 2001 From: Sridhar Samudrala Date: Fri, 5 May 2006 17:04:43 -0700 Subject: [SCTP]: Fix panic's when receiving fragmented SCTP control chunks. Use pskb_pull() to handle incoming COOKIE_ECHO and HEARTBEAT chunks that are received as skb's with fragment list. Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- net/sctp/sm_statefuns.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index f5d131f52a7..8cdba51ec07 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -636,8 +636,9 @@ sctp_disposition_t sctp_sf_do_5_1D_ce(const struct sctp_endpoint *ep, */ chunk->subh.cookie_hdr = (struct sctp_signed_cookie *)chunk->skb->data; - skb_pull(chunk->skb, - ntohs(chunk->chunk_hdr->length) - sizeof(sctp_chunkhdr_t)); + if (!pskb_pull(chunk->skb, ntohs(chunk->chunk_hdr->length) - + sizeof(sctp_chunkhdr_t))) + goto nomem; /* 5.1 D) Upon reception of the COOKIE ECHO chunk, Endpoint * "Z" will reply with a COOKIE ACK chunk after building a TCB @@ -965,7 +966,8 @@ sctp_disposition_t sctp_sf_beat_8_3(const struct sctp_endpoint *ep, */ chunk->subh.hb_hdr = (sctp_heartbeathdr_t *) chunk->skb->data; paylen = ntohs(chunk->chunk_hdr->length) - sizeof(sctp_chunkhdr_t); - skb_pull(chunk->skb, paylen); + if (!pskb_pull(chunk->skb, paylen)) + goto nomem; reply = sctp_make_heartbeat_ack(asoc, chunk, chunk->subh.hb_hdr, paylen); @@ -1860,8 +1862,9 @@ sctp_disposition_t sctp_sf_do_5_2_4_dupcook(const struct sctp_endpoint *ep, * are in good shape. */ chunk->subh.cookie_hdr = (struct sctp_signed_cookie *)chunk->skb->data; - skb_pull(chunk->skb, ntohs(chunk->chunk_hdr->length) - - sizeof(sctp_chunkhdr_t)); + if (!pskb_pull(chunk->skb, ntohs(chunk->chunk_hdr->length) - + sizeof(sctp_chunkhdr_t))) + goto nomem; /* In RFC 2960 5.2.4 3, if both Verification Tags in the State Cookie * of a duplicate COOKIE ECHO match the Verification Tags of the -- cgit v1.2.3 From 35d63edb1c807bc5317e49592260e84637bc432e Mon Sep 17 00:00:00 2001 From: Sridhar Samudrala Date: Fri, 5 May 2006 17:05:23 -0700 Subject: [SCTP]: Fix state table entries for chunks received in CLOSED state. Discard an unexpected chunk in CLOSED state rather can calling BUG(). Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- net/sctp/sm_statetable.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c index 75ef1040876..8bcca567615 100644 --- a/net/sctp/sm_statetable.c +++ b/net/sctp/sm_statetable.c @@ -366,9 +366,9 @@ const sctp_sm_table_entry_t *sctp_sm_lookup_event(sctp_event_t event_type, /* SCTP_STATE_EMPTY */ \ {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ /* SCTP_STATE_CLOSED */ \ - {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ /* SCTP_STATE_COOKIE_WAIT */ \ - {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ /* SCTP_STATE_COOKIE_ECHOED */ \ {.fn = sctp_sf_do_ecne, .name = "sctp_sf_do_ecne"}, \ /* SCTP_STATE_ESTABLISHED */ \ @@ -380,7 +380,7 @@ const sctp_sm_table_entry_t *sctp_sm_lookup_event(sctp_event_t event_type, /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ {.fn = sctp_sf_do_ecne, .name = "sctp_sf_do_ecne"}, \ /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ - {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ } /* TYPE_SCTP_ECN_ECNE */ #define TYPE_SCTP_ECN_CWR { \ @@ -401,7 +401,7 @@ const sctp_sm_table_entry_t *sctp_sm_lookup_event(sctp_event_t event_type, /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ - {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ } /* TYPE_SCTP_ECN_CWR */ #define TYPE_SCTP_SHUTDOWN_COMPLETE { \ @@ -647,7 +647,7 @@ chunk_event_table_unknown[SCTP_STATE_NUM_STATES] = { /* SCTP_STATE_EMPTY */ \ {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ /* SCTP_STATE_CLOSED */ \ - {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + {.fn = sctp_sf_error_closed, .name = "sctp_sf_error_closed"}, \ /* SCTP_STATE_COOKIE_WAIT */ \ {.fn = sctp_sf_do_prm_requestheartbeat, \ .name = "sctp_sf_do_prm_requestheartbeat"}, \ -- cgit v1.2.3 From 1c29fc4989bc2a3838b2837adc12b8aeb0feeede Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 5 May 2006 17:07:13 -0700 Subject: [BRIDGE]: keep track of received multicast packets It makes sense to add this simple statistic to keep track of received multicast packets. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/bridge/br_input.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index b0b7f55c1ed..bfa4d8c333f 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -66,6 +66,7 @@ int br_handle_frame_finish(struct sk_buff *skb) } if (is_multicast_ether_addr(dest)) { + br->statistics.multicast++; br_flood_forward(br, skb, !passedup); if (!passedup) br_pass_frame_up(br, skb); -- cgit v1.2.3 From 134af34632a7b3b0a98a79a2e56bf9cc927e0eac Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 5 May 2006 17:09:13 -0700 Subject: [DCCP]: Fix sock_orphan dead lock Calling sock_orphan inside bh_lock_sock in dccp_close can lead to dead locks. For example, the inet_diag code holds sk_callback_lock without disabling BH. If an inbound packet arrives during that admittedly tiny window, it will cause a dead lock on bh_lock_sock. Another possible path would be through sock_wfree if the network device driver frees the tx skb in process context with BH enabled. We can fix this by moving sock_orphan out of bh_lock_sock. The tricky bit is to work out when we need to destroy the socket ourselves and when it has already been destroyed by someone else. By moving sock_orphan before the release_sock we can solve this problem. This is because as long as we own the socket lock its state cannot change. So we simply record the socket state before the release_sock and then check the state again after we regain the socket lock. If the socket state has transitioned to DCCP_CLOSED in the time being, we know that the socket has been destroyed. Otherwise the socket is still ours to keep. This problem was discoverd by Ingo Molnar using his lock validator. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/dccp/proto.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 1ff7328b0e1..2e0ee8355c4 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -848,6 +848,7 @@ static int dccp_close_state(struct sock *sk) void dccp_close(struct sock *sk, long timeout) { struct sk_buff *skb; + int state; lock_sock(sk); @@ -882,6 +883,11 @@ void dccp_close(struct sock *sk, long timeout) sk_stream_wait_close(sk, timeout); adjudge_to_death: + state = sk->sk_state; + sock_hold(sk); + sock_orphan(sk); + atomic_inc(sk->sk_prot->orphan_count); + /* * It is the last release_sock in its life. It will remove backlog. */ @@ -894,8 +900,9 @@ adjudge_to_death: bh_lock_sock(sk); BUG_TRAP(!sock_owned_by_user(sk)); - sock_hold(sk); - sock_orphan(sk); + /* Have we already been destroyed by a softirq or backlog? */ + if (state != DCCP_CLOSED && sk->sk_state == DCCP_CLOSED) + goto out; /* * The last release_sock may have processed the CLOSE or RESET @@ -915,12 +922,12 @@ adjudge_to_death: #endif } - atomic_inc(sk->sk_prot->orphan_count); if (sk->sk_state == DCCP_CLOSED) inet_csk_destroy_sock(sk); /* Otherwise, socket is reprieved until protocol close. */ +out: bh_unlock_sock(sk); local_bh_enable(); sock_put(sk); -- cgit v1.2.3 From f530937b2cccdb131cb459977943c98421ab09b3 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 5 May 2006 17:19:26 -0700 Subject: [NETROM/ROSE]: Kill module init version kernel log messages. There are out of date and don't tell the user anything useful. The similar messages which IPV4 and the core networking used to output were killed a long time ago. Signed-off-by: Ralf Baechle Signed-off-by: David S. Miller --- net/netrom/af_netrom.c | 3 --- net/rose/af_rose.c | 3 --- 2 files changed, 6 deletions(-) (limited to 'net') diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c index ecd288beca7..3669cb953e6 100644 --- a/net/netrom/af_netrom.c +++ b/net/netrom/af_netrom.c @@ -1370,8 +1370,6 @@ static struct notifier_block nr_dev_notifier = { static struct net_device **dev_nr; -static char banner[] __initdata = KERN_INFO "G4KLX NET/ROM for Linux. Version 0.7 for AX25.037 Linux 2.4\n"; - static int __init nr_proto_init(void) { int i; @@ -1419,7 +1417,6 @@ static int __init nr_proto_init(void) } register_netdevice_notifier(&nr_dev_notifier); - printk(banner); ax25_protocol_register(AX25_P_NETROM, nr_route_frame); ax25_linkfail_register(nr_link_failed); diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index ef4538ac84f..55564efccf1 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -1469,8 +1469,6 @@ static struct notifier_block rose_dev_notifier = { static struct net_device **dev_rose; -static const char banner[] = KERN_INFO "F6FBB/G4KLX ROSE for Linux. Version 0.62 for AX25.037 Linux 2.4\n"; - static int __init rose_proto_init(void) { int i; @@ -1519,7 +1517,6 @@ static int __init rose_proto_init(void) sock_register(&rose_family_ops); register_netdevice_notifier(&rose_dev_notifier); - printk(banner); ax25_protocol_register(AX25_P_ROSE, rose_route_frame); ax25_linkfail_register(rose_link_failed); -- cgit v1.2.3 From 5528e568a760442e0ec8fd2dea1f0791875a066b Mon Sep 17 00:00:00 2001 From: John Heffner Date: Fri, 5 May 2006 17:41:44 -0700 Subject: [TCP]: Fix snd_cwnd adjustments in tcp_highspeed.c Xiaoliang (David) Wei wrote: > Hi gurus, > > I am reading the code of tcp_highspeed.c in the kernel and have a > question on the hstcp_cong_avoid function, specifically the following > AI part (line 136~143 in net/ipv4/tcp_highspeed.c ): > > /* Do additive increase */ > if (tp->snd_cwnd < tp->snd_cwnd_clamp) { > tp->snd_cwnd_cnt += ca->ai; > if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { > tp->snd_cwnd++; > tp->snd_cwnd_cnt -= tp->snd_cwnd; > } > } > > In this part, when (tp->snd_cwnd_cnt == tp->snd_cwnd), > snd_cwnd_cnt will be -1... snd_cwnd_cnt is defined as u16, will this > small chance of getting -1 becomes a problem? > Shall we change it by reversing the order of the cwnd++ and cwnd_cnt -= > cwnd? Absolutely correct. Thanks. Signed-off-by: John Heffner Signed-off-by: David S. Miller --- net/ipv4/tcp_highspeed.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp_highspeed.c b/net/ipv4/tcp_highspeed.c index e0e9d1383c7..b72fa55dfb8 100644 --- a/net/ipv4/tcp_highspeed.c +++ b/net/ipv4/tcp_highspeed.c @@ -137,8 +137,8 @@ static void hstcp_cong_avoid(struct sock *sk, u32 adk, u32 rtt, if (tp->snd_cwnd < tp->snd_cwnd_clamp) { tp->snd_cwnd_cnt += ca->ai; if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { - tp->snd_cwnd++; tp->snd_cwnd_cnt -= tp->snd_cwnd; + tp->snd_cwnd++; } } } -- cgit v1.2.3 From fe9925b551a95fae6ec61470c79f8b701a2fe928 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Sat, 6 May 2006 17:56:03 -0700 Subject: [NET]: Create netdev attribute_groups with class_device_add Atomically create attributes when class device is added. This avoids the race between registering class_device (which generates hotplug event), and the creation of attribute groups. Signed-off-by: Stephen Hemminger Signed-off-by: Greg Kroah-Hartman Signed-off-by: David S. Miller --- net/core/dev.c | 2 +- net/core/net-sysfs.c | 49 ++++++++++++------------------------------------- 2 files changed, 13 insertions(+), 38 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 3bad1afc89f..9ab3cfa5846 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3043,11 +3043,11 @@ void netdev_run_todo(void) switch(dev->reg_state) { case NETREG_REGISTERING: - dev->reg_state = NETREG_REGISTERED; err = netdev_register_sysfs(dev); if (err) printk(KERN_ERR "%s: failed sysfs registration (%d)\n", dev->name, err); + dev->reg_state = NETREG_REGISTERED; break; case NETREG_UNREGISTERING: diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index c12990c9c60..47a6fceb677 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -29,7 +29,7 @@ static const char fmt_ulong[] = "%lu\n"; static inline int dev_isalive(const struct net_device *dev) { - return dev->reg_state == NETREG_REGISTERED; + return dev->reg_state <= NETREG_REGISTERED; } /* use same locking rules as GIF* ioctl's */ @@ -445,58 +445,33 @@ static struct class net_class = { void netdev_unregister_sysfs(struct net_device * net) { - struct class_device * class_dev = &(net->class_dev); - - if (net->get_stats) - sysfs_remove_group(&class_dev->kobj, &netstat_group); - -#ifdef WIRELESS_EXT - if (net->get_wireless_stats || (net->wireless_handlers && - net->wireless_handlers->get_wireless_stats)) - sysfs_remove_group(&class_dev->kobj, &wireless_group); -#endif - class_device_del(class_dev); - + class_device_del(&(net->class_dev)); } /* Create sysfs entries for network device. */ int netdev_register_sysfs(struct net_device *net) { struct class_device *class_dev = &(net->class_dev); - int ret; + struct attribute_group **groups = net->sysfs_groups; + class_device_initialize(class_dev); class_dev->class = &net_class; class_dev->class_data = net; + class_dev->groups = groups; + BUILD_BUG_ON(BUS_ID_SIZE < IFNAMSIZ); strlcpy(class_dev->class_id, net->name, BUS_ID_SIZE); - if ((ret = class_device_register(class_dev))) - goto out; - if (net->get_stats && - (ret = sysfs_create_group(&class_dev->kobj, &netstat_group))) - goto out_unreg; + if (net->get_stats) + *groups++ = &netstat_group; #ifdef WIRELESS_EXT - if (net->get_wireless_stats || (net->wireless_handlers && - net->wireless_handlers->get_wireless_stats)) { - ret = sysfs_create_group(&class_dev->kobj, &wireless_group); - if (ret) - goto out_cleanup; - } - return 0; -out_cleanup: - if (net->get_stats) - sysfs_remove_group(&class_dev->kobj, &netstat_group); -#else - return 0; + if (net->get_wireless_stats + || (net->wireless_handlers && net->wireless_handlers->get_wireless_stats)) + *groups++ = &wireless_group; #endif -out_unreg: - printk(KERN_WARNING "%s: sysfs attribute registration failed %d\n", - net->name, ret); - class_device_unregister(class_dev); -out: - return ret; + return class_device_add(class_dev); } int netdev_sysfs_init(void) -- cgit v1.2.3 From 0182bd2b1e2fb45a55f110795bfdb9aa5f6c6b0b Mon Sep 17 00:00:00 2001 From: Hua Zhong Date: Sat, 6 May 2006 18:11:39 -0700 Subject: [IPV4]: Remove likely in ip_rcv_finish() This is another result from my likely profiling tool (dwalker@mvista.com just sent the patch of the profiling tool to linux-kernel mailing list, which is similar to what I use). On my system (not very busy, normal development machine within a VMWare workstation), I see a 6/5 miss/hit ratio for this "likely". Signed-off-by: Hua Zhong Signed-off-by: David S. Miller --- net/ipv4/ip_input.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 18d7fad474d..c9026dbf4c9 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -337,7 +337,7 @@ static inline int ip_rcv_finish(struct sk_buff *skb) * Initialise the virtual path cache for the packet. It describes * how the packet travels inside Linux networking. */ - if (likely(skb->dst == NULL)) { + if (skb->dst == NULL) { int err = ip_route_input(skb, iph->daddr, iph->saddr, iph->tos, skb->dev); if (unlikely(err)) { -- cgit v1.2.3 From 63cbd2fda38f3d1f107c4fd6261e5660be3eccf9 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 9 May 2006 15:18:50 -0700 Subject: [IPV4]: ip_options_fragment() has no effect on fragmentation Fix error point to options in ip_options_fragment(). optptr get a error pointer to the ipv4 header, correct is pointer to ipv4 options. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller --- net/ipv4/ip_options.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index 9bebad07bf2..cbcae654462 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -209,7 +209,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb) void ip_options_fragment(struct sk_buff * skb) { - unsigned char * optptr = skb->nh.raw; + unsigned char * optptr = skb->nh.raw + sizeof(struct iphdr); struct ip_options * opt = &(IPCB(skb)->opt); int l = opt->optlen; int optlen; -- cgit v1.2.3 From f07d5b946510a54937a75a3654941e855ffdc4c2 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 9 May 2006 15:23:03 -0700 Subject: [NET]: Make netdev_chain a raw notifier. From: Alan Stern This chain does it's own locking via the RTNL semaphore, and can also run recursively so adding a new mutex here was causing deadlocks. Signed-off-by: David S. Miller --- net/core/dev.c | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 9ab3cfa5846..ced57430f6d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -193,7 +193,7 @@ static inline struct hlist_head *dev_index_hash(int ifindex) * Our notifier list */ -static BLOCKING_NOTIFIER_HEAD(netdev_chain); +static RAW_NOTIFIER_HEAD(netdev_chain); /* * Device drivers call our routines to queue packets here. We empty the @@ -736,7 +736,7 @@ int dev_change_name(struct net_device *dev, char *newname) if (!err) { hlist_del(&dev->name_hlist); hlist_add_head(&dev->name_hlist, dev_name_hash(dev->name)); - blocking_notifier_call_chain(&netdev_chain, + raw_notifier_call_chain(&netdev_chain, NETDEV_CHANGENAME, dev); } @@ -751,7 +751,7 @@ int dev_change_name(struct net_device *dev, char *newname) */ void netdev_features_change(struct net_device *dev) { - blocking_notifier_call_chain(&netdev_chain, NETDEV_FEAT_CHANGE, dev); + raw_notifier_call_chain(&netdev_chain, NETDEV_FEAT_CHANGE, dev); } EXPORT_SYMBOL(netdev_features_change); @@ -766,7 +766,7 @@ EXPORT_SYMBOL(netdev_features_change); void netdev_state_change(struct net_device *dev) { if (dev->flags & IFF_UP) { - blocking_notifier_call_chain(&netdev_chain, + raw_notifier_call_chain(&netdev_chain, NETDEV_CHANGE, dev); rtmsg_ifinfo(RTM_NEWLINK, dev, 0); } @@ -864,7 +864,7 @@ int dev_open(struct net_device *dev) /* * ... and announce new interface. */ - blocking_notifier_call_chain(&netdev_chain, NETDEV_UP, dev); + raw_notifier_call_chain(&netdev_chain, NETDEV_UP, dev); } return ret; } @@ -887,7 +887,7 @@ int dev_close(struct net_device *dev) * Tell people we are going down, so that they can * prepare to death, when device is still operating. */ - blocking_notifier_call_chain(&netdev_chain, NETDEV_GOING_DOWN, dev); + raw_notifier_call_chain(&netdev_chain, NETDEV_GOING_DOWN, dev); dev_deactivate(dev); @@ -924,7 +924,7 @@ int dev_close(struct net_device *dev) /* * Tell people we are down */ - blocking_notifier_call_chain(&netdev_chain, NETDEV_DOWN, dev); + raw_notifier_call_chain(&netdev_chain, NETDEV_DOWN, dev); return 0; } @@ -955,7 +955,7 @@ int register_netdevice_notifier(struct notifier_block *nb) int err; rtnl_lock(); - err = blocking_notifier_chain_register(&netdev_chain, nb); + err = raw_notifier_chain_register(&netdev_chain, nb); if (!err) { for (dev = dev_base; dev; dev = dev->next) { nb->notifier_call(nb, NETDEV_REGISTER, dev); @@ -983,7 +983,7 @@ int unregister_netdevice_notifier(struct notifier_block *nb) int err; rtnl_lock(); - err = blocking_notifier_chain_unregister(&netdev_chain, nb); + err = raw_notifier_chain_unregister(&netdev_chain, nb); rtnl_unlock(); return err; } @@ -994,12 +994,12 @@ int unregister_netdevice_notifier(struct notifier_block *nb) * @v: pointer passed unmodified to notifier function * * Call all network notifier blocks. Parameters and return value - * are as for blocking_notifier_call_chain(). + * are as for raw_notifier_call_chain(). */ int call_netdevice_notifiers(unsigned long val, void *v) { - return blocking_notifier_call_chain(&netdev_chain, val, v); + return raw_notifier_call_chain(&netdev_chain, val, v); } /* When > 0 there are consumers of rx skb time stamps */ @@ -2308,7 +2308,7 @@ int dev_change_flags(struct net_device *dev, unsigned flags) if (dev->flags & IFF_UP && ((old_flags ^ dev->flags) &~ (IFF_UP | IFF_PROMISC | IFF_ALLMULTI | IFF_VOLATILE))) - blocking_notifier_call_chain(&netdev_chain, + raw_notifier_call_chain(&netdev_chain, NETDEV_CHANGE, dev); if ((flags ^ dev->gflags) & IFF_PROMISC) { @@ -2353,7 +2353,7 @@ int dev_set_mtu(struct net_device *dev, int new_mtu) else dev->mtu = new_mtu; if (!err && dev->flags & IFF_UP) - blocking_notifier_call_chain(&netdev_chain, + raw_notifier_call_chain(&netdev_chain, NETDEV_CHANGEMTU, dev); return err; } @@ -2370,7 +2370,7 @@ int dev_set_mac_address(struct net_device *dev, struct sockaddr *sa) return -ENODEV; err = dev->set_mac_address(dev, sa); if (!err) - blocking_notifier_call_chain(&netdev_chain, + raw_notifier_call_chain(&netdev_chain, NETDEV_CHANGEADDR, dev); return err; } @@ -2427,7 +2427,7 @@ static int dev_ifsioc(struct ifreq *ifr, unsigned int cmd) return -EINVAL; memcpy(dev->broadcast, ifr->ifr_hwaddr.sa_data, min(sizeof ifr->ifr_hwaddr.sa_data, (size_t) dev->addr_len)); - blocking_notifier_call_chain(&netdev_chain, + raw_notifier_call_chain(&netdev_chain, NETDEV_CHANGEADDR, dev); return 0; @@ -2882,7 +2882,7 @@ int register_netdevice(struct net_device *dev) write_unlock_bh(&dev_base_lock); /* Notify protocols, that a new device appeared. */ - blocking_notifier_call_chain(&netdev_chain, NETDEV_REGISTER, dev); + raw_notifier_call_chain(&netdev_chain, NETDEV_REGISTER, dev); /* Finish registration after unlock */ net_set_todo(dev); @@ -2961,7 +2961,7 @@ static void netdev_wait_allrefs(struct net_device *dev) rtnl_lock(); /* Rebroadcast unregister notification */ - blocking_notifier_call_chain(&netdev_chain, + raw_notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev); if (test_bit(__LINK_STATE_LINKWATCH_PENDING, @@ -3216,7 +3216,7 @@ int unregister_netdevice(struct net_device *dev) /* Notify protocols, that we are about to destroy this device. They should clean all the things. */ - blocking_notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev); + raw_notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev); /* * Flush the multicast chain -- cgit v1.2.3 From 11766199a0bb9a7ba57510119e7340140e7c3e24 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 9 May 2006 15:25:25 -0700 Subject: [IRDA]: Removing unused EXPORT_SYMBOLs This patch removes the following unused EXPORT_SYMBOL's: - irias_find_attrib - irias_new_string_value - irias_new_octseq_value Signed-off-by: Adrian Bunk Signed-off-by: Samuel Ortiz Signed-off-by: David S. Miller --- net/irda/irias_object.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'net') diff --git a/net/irda/irias_object.c b/net/irda/irias_object.c index c6d169fbdce..82e665c7999 100644 --- a/net/irda/irias_object.c +++ b/net/irda/irias_object.c @@ -257,7 +257,6 @@ struct ias_attrib *irias_find_attrib(struct ias_object *obj, char *name) /* Unsafe (locking), attrib might change */ return attrib; } -EXPORT_SYMBOL(irias_find_attrib); /* * Function irias_add_attribute (obj, attrib) @@ -484,7 +483,6 @@ struct ias_value *irias_new_string_value(char *string) return value; } -EXPORT_SYMBOL(irias_new_string_value); /* * Function irias_new_octseq_value (octets, len) @@ -519,7 +517,6 @@ struct ias_value *irias_new_octseq_value(__u8 *octseq , int len) memcpy(value->t.oct_seq, octseq , len); return value; } -EXPORT_SYMBOL(irias_new_octseq_value); struct ias_value *irias_new_missing_value(void) { -- cgit v1.2.3 From 8c1056839e808aad728db86d739ffec71d2d1db8 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 9 May 2006 15:27:54 -0700 Subject: [NET] linkwatch: Handle jiffies wrap-around The test used in the linkwatch does not handle wrap-arounds correctly. Since the intention of the code is to eliminate bursts of messages we can afford to delay things up to a second. Using that fact we can easily handle wrap-arounds by making sure that we don't delay things by more than one second. This is based on diagnosis and a patch by Stefan Rompf. Signed-off-by: Herbert Xu Acked-by: Stefan Rompf Signed-off-by: David S. Miller --- net/core/link_watch.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/link_watch.c b/net/core/link_watch.c index 341de44c7ed..646937cc2d8 100644 --- a/net/core/link_watch.c +++ b/net/core/link_watch.c @@ -170,13 +170,13 @@ void linkwatch_fire_event(struct net_device *dev) spin_unlock_irqrestore(&lweventlist_lock, flags); if (!test_and_set_bit(LW_RUNNING, &linkwatch_flags)) { - unsigned long thisevent = jiffies; + unsigned long delay = linkwatch_nextevent - jiffies; - if (thisevent >= linkwatch_nextevent) { + /* If we wrap around we'll delay it by at most HZ. */ + if (!delay || delay > HZ) schedule_work(&linkwatch_work); - } else { - schedule_delayed_work(&linkwatch_work, linkwatch_nextevent - thisevent); - } + else + schedule_delayed_work(&linkwatch_work, delay); } } } -- cgit v1.2.3 From b17a7c179dd3ce7d04373fddf660eda21efc9db9 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 10 May 2006 13:21:17 -0700 Subject: [NET]: Do sysfs registration as part of register_netdevice. The last step of netdevice registration was being done by a delayed call, but because it was delayed, it was impossible to return any error code if the class_device registration failed. Side effects: * one state in registration process is unnecessary. * register_netdevice can sleep inside class_device registration/hotplug * code in netdev_run_todo only does unregistration so it is simpler. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/core/dev.c | 63 +++++++++++++++++++++++++--------------------------------- 1 file changed, 27 insertions(+), 36 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index ced57430f6d..2dce673a039 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2777,6 +2777,8 @@ int register_netdevice(struct net_device *dev) BUG_ON(dev_boot_phase); ASSERT_RTNL(); + might_sleep(); + /* When net_device's are persistent, this will be fatal. */ BUG_ON(dev->reg_state != NETREG_UNINITIALIZED); @@ -2863,6 +2865,11 @@ int register_netdevice(struct net_device *dev) if (!dev->rebuild_header) dev->rebuild_header = default_rebuild_header; + ret = netdev_register_sysfs(dev); + if (ret) + goto out_err; + dev->reg_state = NETREG_REGISTERED; + /* * Default initial state at registry is that the * device is present. @@ -2878,14 +2885,11 @@ int register_netdevice(struct net_device *dev) hlist_add_head(&dev->name_hlist, head); hlist_add_head(&dev->index_hlist, dev_index_hash(dev->ifindex)); dev_hold(dev); - dev->reg_state = NETREG_REGISTERING; write_unlock_bh(&dev_base_lock); /* Notify protocols, that a new device appeared. */ raw_notifier_call_chain(&netdev_chain, NETDEV_REGISTER, dev); - /* Finish registration after unlock */ - net_set_todo(dev); ret = 0; out: @@ -3008,7 +3012,7 @@ static void netdev_wait_allrefs(struct net_device *dev) * * We are invoked by rtnl_unlock() after it drops the semaphore. * This allows us to deal with problems: - * 1) We can create/delete sysfs objects which invoke hotplug + * 1) We can delete sysfs objects which invoke hotplug * without deadlocking with linkwatch via keventd. * 2) Since we run with the RTNL semaphore not held, we can sleep * safely in order to wait for the netdev refcnt to drop to zero. @@ -3017,8 +3021,6 @@ static DEFINE_MUTEX(net_todo_run_mutex); void netdev_run_todo(void) { struct list_head list = LIST_HEAD_INIT(list); - int err; - /* Need to guard against multiple cpu's getting out of order. */ mutex_lock(&net_todo_run_mutex); @@ -3041,40 +3043,29 @@ void netdev_run_todo(void) = list_entry(list.next, struct net_device, todo_list); list_del(&dev->todo_list); - switch(dev->reg_state) { - case NETREG_REGISTERING: - err = netdev_register_sysfs(dev); - if (err) - printk(KERN_ERR "%s: failed sysfs registration (%d)\n", - dev->name, err); - dev->reg_state = NETREG_REGISTERED; - break; - - case NETREG_UNREGISTERING: - netdev_unregister_sysfs(dev); - dev->reg_state = NETREG_UNREGISTERED; - - netdev_wait_allrefs(dev); + if (unlikely(dev->reg_state != NETREG_UNREGISTERING)) { + printk(KERN_ERR "network todo '%s' but state %d\n", + dev->name, dev->reg_state); + dump_stack(); + continue; + } - /* paranoia */ - BUG_ON(atomic_read(&dev->refcnt)); - BUG_TRAP(!dev->ip_ptr); - BUG_TRAP(!dev->ip6_ptr); - BUG_TRAP(!dev->dn_ptr); + netdev_unregister_sysfs(dev); + dev->reg_state = NETREG_UNREGISTERED; + netdev_wait_allrefs(dev); - /* It must be the very last action, - * after this 'dev' may point to freed up memory. - */ - if (dev->destructor) - dev->destructor(dev); - break; + /* paranoia */ + BUG_ON(atomic_read(&dev->refcnt)); + BUG_TRAP(!dev->ip_ptr); + BUG_TRAP(!dev->ip6_ptr); + BUG_TRAP(!dev->dn_ptr); - default: - printk(KERN_ERR "network todo '%s' but state %d\n", - dev->name, dev->reg_state); - break; - } + /* It must be the very last action, + * after this 'dev' may point to freed up memory. + */ + if (dev->destructor) + dev->destructor(dev); } out: -- cgit v1.2.3 From ac05202e8b83594bf6797d241371e6c752f371e6 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 10 May 2006 13:21:53 -0700 Subject: [BRIDGE]: Do sysfs registration inside rtnl. Now that netdevice sysfs registration is done as part of register_netdevice; bridge code no longer has to be tricky when adding it's kobjects to bridges. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/bridge/br_if.c | 21 +++++++-------------- 1 file changed, 7 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 59eef42d4a4..ad1c7af65ec 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -308,26 +308,19 @@ int br_add_bridge(const char *name) if (ret) goto err2; - /* network device kobject is not setup until - * after rtnl_unlock does it's hotplug magic. - * so hold reference to avoid race. - */ - dev_hold(dev); - rtnl_unlock(); - ret = br_sysfs_addbr(dev); - dev_put(dev); - - if (ret) - unregister_netdev(dev); - out: - return ret; + if (ret) + goto err3; + rtnl_unlock(); + return 0; + err3: + unregister_netdev(dev); err2: free_netdev(dev); err1: rtnl_unlock(); - goto out; + return ret; } int br_del_bridge(const char *name) -- cgit v1.2.3 From b0013fd47b14fc26eec07a6b2cec0c2a8954e1d7 Mon Sep 17 00:00:00 2001 From: Alexey Kuznetsov Date: Wed, 10 May 2006 13:24:38 -0700 Subject: [IPV6]: skb leakage in inet6_csk_xmit inet6_csk_xit does not free skb when routing fails. Signed-off-by: Alexey Kuznetsov Signed-off-by: David S. Miller --- net/ipv6/inet6_connection_sock.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index f8f3a37a149..eb2865d5ae2 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -173,6 +173,7 @@ int inet6_csk_xmit(struct sk_buff *skb, int ipfragok) if (err) { sk->sk_err_soft = -err; + kfree_skb(skb); return err; } @@ -181,6 +182,7 @@ int inet6_csk_xmit(struct sk_buff *skb, int ipfragok) if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0) { sk->sk_route_caps = 0; + kfree_skb(skb); return err; } -- cgit v1.2.3 From 210525d65d33d17eb6bea6c965ce442d60d9aa8d Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 11 May 2006 12:22:03 -0700 Subject: [NET_SCHED]: HFSC: fix thinko in hfsc_adjust_levels() When deleting the last child the level of a class should drop to zero. Noticed by Andreas Mueller Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/sched/sch_hfsc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 91132f6871d..f1c7bd29f2c 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -974,10 +974,10 @@ hfsc_adjust_levels(struct hfsc_class *cl) do { level = 0; list_for_each_entry(p, &cl->children, siblings) { - if (p->level > level) - level = p->level; + if (p->level >= level) + level = p->level + 1; } - cl->level = level + 1; + cl->level = level; } while ((cl = cl->cl_parent) != NULL); } -- cgit v1.2.3 From bd89efc532fe41f867f848144cc8b42054ddf6f9 Mon Sep 17 00:00:00 2001 From: Simon Kelley Date: Fri, 12 May 2006 14:56:08 -0700 Subject: [NEIGH]: Fix IP-over-ATM and ARP interaction. The classical IP over ATM code maintains its own IPv4 <-> ARP table, using the standard neighbour-table code. The neigh_table_init function adds this neighbour table to a linked list of all neighbor tables which is used by the functions neigh_delete() neigh_add() and neightbl_set(), all called by the netlink code. Once the ATM neighbour table is added to the list, there are two tables with family == AF_INET there, and ARP entries sent via netlink go into the first table with matching family. This is indeterminate and often wrong. To see the bug, on a kernel with CLIP enabled, create a standard IPv4 ARP entry by pinging an unused address on a local subnet. Then attempt to complete that entry by doing ip neigh replace lladdr nud reachable Looking at the ARP tables by using ip neigh show will reveal two ARP entries for the same address. One of these can be found in /proc/net/arp, and the other in /proc/net/atm/arp. This patch adds a new function, neigh_table_init_no_netlink() which does everything the neigh_table_init() does, except add the table to the netlink all-arp-tables chain. In addition neigh_table_init() has a check that all tables on the chain have a distinct address family. The init call in clip.c is changed to call neigh_table_init_no_netlink(). Since ATM ARP tables are rather more complicated than can currently be handled by the available rtattrs in the netlink protocol, no functionality is lost by this patch, and non-ATM ARP manipulation via netlink is rescued. A more complete solution would involve a rtattr for ATM ARP entries and some way for the netlink code to give neigh_add and friends more information than just address family with which to find the correct ARP table. [ I've changed the assertion checking in neigh_table_init() to not use BUG_ON() while holding neigh_tbl_lock. Instead we remember that we found an existing tbl with the same family, and after dropping the lock we'll give a diagnostic kernel log message and a stack dump. -DaveM ] Signed-off-by: Simon Kelley Signed-off-by: David S. Miller --- net/atm/clip.c | 2 +- net/core/neighbour.c | 21 +++++++++++++++++++-- 2 files changed, 20 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/atm/clip.c b/net/atm/clip.c index 1a786bfaa41..72d85298266 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -963,7 +963,7 @@ static struct file_operations arp_seq_fops = { static int __init atm_clip_init(void) { struct proc_dir_entry *p; - neigh_table_init(&clip_tbl); + neigh_table_init_no_netlink(&clip_tbl); clip_tbl_hook = &clip_tbl; register_atm_ioctl(&clip_ioctl_ops); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 4cf878efdb4..50a8c73caf9 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1326,8 +1326,7 @@ void neigh_parms_destroy(struct neigh_parms *parms) kfree(parms); } - -void neigh_table_init(struct neigh_table *tbl) +void neigh_table_init_no_netlink(struct neigh_table *tbl) { unsigned long now = jiffies; unsigned long phsize; @@ -1383,10 +1382,27 @@ void neigh_table_init(struct neigh_table *tbl) tbl->last_flush = now; tbl->last_rand = now + tbl->parms.reachable_time * 20; +} + +void neigh_table_init(struct neigh_table *tbl) +{ + struct neigh_table *tmp; + + neigh_table_init_no_netlink(tbl); write_lock(&neigh_tbl_lock); + for (tmp = neigh_tables; tmp; tmp = tmp->next) { + if (tmp->family == tbl->family) + break; + } tbl->next = neigh_tables; neigh_tables = tbl; write_unlock(&neigh_tbl_lock); + + if (unlikely(tmp)) { + printk(KERN_ERR "NEIGH: Registering multiple tables for " + "family %d\n", tbl->family); + dump_stack(); + } } int neigh_table_clear(struct neigh_table *tbl) @@ -2657,6 +2673,7 @@ EXPORT_SYMBOL(neigh_rand_reach_time); EXPORT_SYMBOL(neigh_resolve_output); EXPORT_SYMBOL(neigh_table_clear); EXPORT_SYMBOL(neigh_table_init); +EXPORT_SYMBOL(neigh_table_init_no_netlink); EXPORT_SYMBOL(neigh_update); EXPORT_SYMBOL(neigh_update_hhs); EXPORT_SYMBOL(pneigh_enqueue); -- cgit v1.2.3