From 2244d07bfa2097cb00600da91c715a8aa547917e Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Tue, 17 Aug 2010 08:59:14 +0000 Subject: net: simplify flags for tx timestamping This patch removes the abstraction introduced by the union skb_shared_tx in the shared skb data. The access of the different union elements at several places led to some confusion about accessing the shared tx_flags e.g. in skb_orphan_try(). http://marc.info/?l=linux-netdev&m=128084897415886&w=2 Signed-off-by: Oliver Hartkopp Signed-off-by: David S. Miller --- include/net/ip.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index 890f9725d681..7691aca133db 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -53,7 +53,7 @@ struct ipcm_cookie { __be32 addr; int oif; struct ip_options *opt; - union skb_shared_tx shtx; + __u8 tx_flags; }; #define IPCB(skb) ((struct inet_skb_parm*)((skb)->cb)) -- cgit v1.2.2 From a02cec2155fbea457eca8881870fd2de1a4c4c76 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 22 Sep 2010 20:43:57 +0000 Subject: net: return operator cleanup Change "return (EXPR);" to "return EXPR;" return is not a function, parentheses are not required. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index 7691aca133db..dbee3fe260e1 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -238,9 +238,9 @@ int ip_decrease_ttl(struct iphdr *iph) static inline int ip_dont_fragment(struct sock *sk, struct dst_entry *dst) { - return (inet_sk(sk)->pmtudisc == IP_PMTUDISC_DO || + return inet_sk(sk)->pmtudisc == IP_PMTUDISC_DO || (inet_sk(sk)->pmtudisc == IP_PMTUDISC_WANT && - !(dst_metric_locked(dst, RTAX_MTU)))); + !(dst_metric_locked(dst, RTAX_MTU))); } extern void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst, int more); -- cgit v1.2.2 From 43a951e9994fd218ab4e641f94a2fc53556c3675 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 25 Oct 2010 03:32:44 +0000 Subject: ipv4: add __rcu annotations to ip_ra_chain Add __rcu annotations to : (struct ip_ra_chain)->next struct ip_ra_chain *ip_ra_chain; And use appropriate rcu primitives. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index dbee3fe260e1..86e2b182a0c0 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -59,7 +59,7 @@ struct ipcm_cookie { #define IPCB(skb) ((struct inet_skb_parm*)((skb)->cb)) struct ip_ra_chain { - struct ip_ra_chain *next; + struct ip_ra_chain __rcu *next; struct sock *sk; union { void (*destructor)(struct sock *); @@ -68,7 +68,7 @@ struct ip_ra_chain { struct rcu_head rcu; }; -extern struct ip_ra_chain *ip_ra_chain; +extern struct ip_ra_chain __rcu *ip_ra_chain; /* IP flags. */ #define IP_CE 0x8000 /* Flag: "Congestion" */ -- cgit v1.2.2 From 323e126f0c5995f779d7df7fd035f6e8fed8764d Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 12 Dec 2010 21:55:08 -0800 Subject: ipv4: Don't pre-seed hoplimit metric. Always go through a new ip4_dst_hoplimit() helper, just like ipv6. This allowed several simplifications: 1) The interim dst_metric_hoplimit() can go as it's no longer userd. 2) The sysctl_ip_default_ttl entry no longer needs to use ipv4_doint_and_flush, since the sysctl is not cached in routing cache metrics any longer. 3) ipv4_doint_and_flush no longer needs to be exported and therefore can be marked static. When ipv4_doint_and_flush_strategy was removed some time ago, the external declaration in ip.h was mistakenly left around so kill that off too. We have to move the sysctl_ip_default_ttl declaration into ipv4's route cache definition header net/route.h, because currently net/ip.h (where the declaration lives now) has a back dependency on net/route.h Signed-off-by: David S. Miller --- include/net/ip.h | 10 ---------- 1 file changed, 10 deletions(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index 86e2b182a0c0..67fac78a186b 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -201,7 +201,6 @@ static inline int inet_is_reserved_local_port(int port) return test_bit(port, sysctl_local_reserved_ports); } -extern int sysctl_ip_default_ttl; extern int sysctl_ip_nonlocal_bind; extern struct ctl_path net_core_path[]; @@ -428,15 +427,6 @@ extern void ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err, extern void ip_local_error(struct sock *sk, int err, __be32 daddr, __be16 dport, u32 info); -/* sysctl helpers - any sysctl which holds a value that ends up being - * fed into the routing cache should use these handlers. - */ -int ipv4_doint_and_flush(ctl_table *ctl, int write, - void __user *buffer, - size_t *lenp, loff_t *ppos); -int ipv4_doint_and_flush_strategy(ctl_table *table, - void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen); #ifdef CONFIG_PROC_FS extern int ip_misc_proc_init(void); #endif -- cgit v1.2.2 From 1c32c5ad6fac8cee1a77449f5abf211e911ff830 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 1 Mar 2011 02:36:47 +0000 Subject: inet: Add ip_make_skb and ip_finish_skb This patch adds the helper ip_make_skb which is like ip_append_data and ip_push_pending_frames all rolled into one, except that it does not send the skb produced. The sending part is carried out by ip_send_skb, which the transport protocol can call after it has tweaked the skb. It is meant to be called in cases where corking is not used should have a one-to-one correspondence to sendmsg. This patch also adds the helper ip_finish_skb which is meant to be replace ip_push_pending_frames when corking is required. Previously the protocol stack would peek at the socket write queue and add its header to the first packet. With ip_finish_skb, the protocol stack can directly operate on the final skb instead, just like the non-corking case with ip_make_skb. Signed-off-by: Herbert Xu Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip.h | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index 67fac78a186b..a4f631108c54 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -116,8 +116,24 @@ extern int ip_append_data(struct sock *sk, extern int ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb); extern ssize_t ip_append_page(struct sock *sk, struct page *page, int offset, size_t size, int flags); +extern struct sk_buff *__ip_make_skb(struct sock *sk, + struct sk_buff_head *queue, + struct inet_cork *cork); +extern int ip_send_skb(struct sk_buff *skb); extern int ip_push_pending_frames(struct sock *sk); extern void ip_flush_pending_frames(struct sock *sk); +extern struct sk_buff *ip_make_skb(struct sock *sk, + int getfrag(void *from, char *to, int offset, int len, + int odd, struct sk_buff *skb), + void *from, int length, int transhdrlen, + struct ipcm_cookie *ipc, + struct rtable **rtp, + unsigned int flags); + +static inline struct sk_buff *ip_finish_skb(struct sock *sk) +{ + return __ip_make_skb(sk, &sk->sk_write_queue, &inet_sk(sk)->cork); +} /* datagram.c */ extern int ip4_datagram_connect(struct sock *sk, -- cgit v1.2.2 From 93ca3bb5df9bc8b2c60485e1cc6507c3d7c8e1fa Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Timo=20Ter=C3=A4s?= Date: Mon, 28 Mar 2011 22:40:53 +0000 Subject: net: gre: provide multicast mappings for ipv4 and ipv6 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit My commit 6d55cb91a0020ac0 (gre: fix hard header destination address checking) broke multicast. The reason is that ip_gre used to get ipgre_header() calls with zero destination if we have NOARP or multicast destination. Instead the actual target was decided at ipgre_tunnel_xmit() time based on per-protocol dissection. Instead of allowing the "abuse" of ->header() calls with invalid destination, this creates multicast mappings for ip_gre. This also fixes "ip neigh show nud noarp" to display the proper multicast mappings used by the gre device. Reported-by: Doug Kehn Signed-off-by: Timo Teräs Acked-by: Doug Kehn Signed-off-by: David S. Miller --- include/net/ip.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index a4f631108c54..7c416583b710 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -339,6 +339,14 @@ static inline void ip_ib_mc_map(__be32 naddr, const unsigned char *broadcast, ch buf[16] = addr & 0x0f; } +static inline void ip_ipgre_mc_map(__be32 naddr, const unsigned char *broadcast, char *buf) +{ + if ((broadcast[0] | broadcast[1] | broadcast[2] | broadcast[3]) != 0) + memcpy(buf, broadcast, 4); + else + memcpy(buf, &naddr, sizeof(naddr)); +} + #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) #include #endif -- cgit v1.2.2 From f6d8bd051c391c1c0458a30b2a7abcd939329259 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 21 Apr 2011 09:45:37 +0000 Subject: inet: add RCU protection to inet->opt We lack proper synchronization to manipulate inet->opt ip_options Problem is ip_make_skb() calls ip_setup_cork() and ip_setup_cork() possibly makes a copy of ipc->opt (struct ip_options), without any protection against another thread manipulating inet->opt. Another thread can change inet->opt pointer and free old one under us. Use RCU to protect inet->opt (changed to inet->inet_opt). Instead of handling atomic refcounts, just copy ip_options when necessary, to avoid cache line dirtying. We cant insert an rcu_head in struct ip_options since its included in skb->cb[], so this patch is large because I had to introduce a new ip_options_rcu structure. Signed-off-by: Eric Dumazet Cc: Herbert Xu Signed-off-by: David S. Miller --- include/net/ip.h | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index 7c416583b710..3a59bf99aa3a 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -52,7 +52,7 @@ static inline unsigned int ip_hdrlen(const struct sk_buff *skb) struct ipcm_cookie { __be32 addr; int oif; - struct ip_options *opt; + struct ip_options_rcu *opt; __u8 tx_flags; }; @@ -92,7 +92,7 @@ extern int igmp_mc_proc_init(void); extern int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk, __be32 saddr, __be32 daddr, - struct ip_options *opt); + struct ip_options_rcu *opt); extern int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev); extern int ip_local_deliver(struct sk_buff *skb); @@ -416,14 +416,15 @@ extern int ip_forward(struct sk_buff *skb); * Functions provided by ip_options.c */ -extern void ip_options_build(struct sk_buff *skb, struct ip_options *opt, __be32 daddr, struct rtable *rt, int is_frag); +extern void ip_options_build(struct sk_buff *skb, struct ip_options *opt, + __be32 daddr, struct rtable *rt, int is_frag); extern int ip_options_echo(struct ip_options *dopt, struct sk_buff *skb); extern void ip_options_fragment(struct sk_buff *skb); extern int ip_options_compile(struct net *net, struct ip_options *opt, struct sk_buff *skb); -extern int ip_options_get(struct net *net, struct ip_options **optp, +extern int ip_options_get(struct net *net, struct ip_options_rcu **optp, unsigned char *data, int optlen); -extern int ip_options_get_from_user(struct net *net, struct ip_options **optp, +extern int ip_options_get_from_user(struct net *net, struct ip_options_rcu **optp, unsigned char __user *data, int optlen); extern void ip_options_undo(struct ip_options * opt); extern void ip_forward_options(struct sk_buff *skb); -- cgit v1.2.2 From bdc712b4c2baf9515887de3a52e7ecd89fafc0c7 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 6 May 2011 15:02:07 -0700 Subject: inet: Decrease overhead of on-stack inet_cork. When we fast path datagram sends to avoid locking by putting the inet_cork on the stack we use up lots of space that isn't necessary. This is because inet_cork contains a "struct flowi" which isn't used in these code paths. Split inet_cork to two parts, "inet_cork" and "inet_cork_full". Only the latter of which has the "struct flowi" and is what is stored in inet_sock. Signed-off-by: David S. Miller Acked-by: Eric Dumazet --- include/net/ip.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index 3a59bf99aa3a..095e392d5f16 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -132,7 +132,7 @@ extern struct sk_buff *ip_make_skb(struct sock *sk, static inline struct sk_buff *ip_finish_skb(struct sock *sk) { - return __ip_make_skb(sk, &sk->sk_write_queue, &inet_sk(sk)->cork); + return __ip_make_skb(sk, &sk->sk_write_queue, &inet_sk(sk)->cork.base); } /* datagram.c */ -- cgit v1.2.2 From d9d8da805dcb503ef8ee49918a94d49085060f23 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 6 May 2011 22:23:20 -0700 Subject: inet: Pass flowi to ->queue_xmit(). This allows us to acquire the exact route keying information from the protocol, however that might be managed. It handles all of the possibilities, from the simplest case of storing the key in inet->cork.fl to the more complex setup SCTP has where individual transports determine the flow. Signed-off-by: David S. Miller --- include/net/ip.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index 095e392d5f16..acf8b7814c4e 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -104,7 +104,7 @@ extern int ip_do_nat(struct sk_buff *skb); extern void ip_send_check(struct iphdr *ip); extern int __ip_local_out(struct sk_buff *skb); extern int ip_local_out(struct sk_buff *skb); -extern int ip_queue_xmit(struct sk_buff *skb); +extern int ip_queue_xmit(struct sk_buff *skb, struct flowi *fl); extern void ip_init(void); extern int ip_append_data(struct sock *sk, int getfrag(void *from, char *to, int offset, int len, -- cgit v1.2.2 From 77968b78242ee25e2a4d759f0fca8dd52df6d479 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 8 May 2011 17:12:19 -0700 Subject: ipv4: Pass flow keys down into datagram packet building engine. This way ip_output.c no longer needs rt->rt_{src,dst}. We already have these keys sitting, ready and waiting, on the stack or in a socket structure. Signed-off-by: David S. Miller --- include/net/ip.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index acf8b7814c4e..a4253795c5c5 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -117,12 +117,14 @@ extern int ip_generic_getfrag(void *from, char *to, int offset, int len, int od extern ssize_t ip_append_page(struct sock *sk, struct page *page, int offset, size_t size, int flags); extern struct sk_buff *__ip_make_skb(struct sock *sk, + struct flowi4 *fl4, struct sk_buff_head *queue, struct inet_cork *cork); extern int ip_send_skb(struct sk_buff *skb); -extern int ip_push_pending_frames(struct sock *sk); +extern int ip_push_pending_frames(struct sock *sk, struct flowi4 *fl4); extern void ip_flush_pending_frames(struct sock *sk); extern struct sk_buff *ip_make_skb(struct sock *sk, + struct flowi4 *fl4, int getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb), void *from, int length, int transhdrlen, @@ -130,9 +132,9 @@ extern struct sk_buff *ip_make_skb(struct sock *sk, struct rtable **rtp, unsigned int flags); -static inline struct sk_buff *ip_finish_skb(struct sock *sk) +static inline struct sk_buff *ip_finish_skb(struct sock *sk, struct flowi4 *fl4) { - return __ip_make_skb(sk, &sk->sk_write_queue, &inet_sk(sk)->cork.base); + return __ip_make_skb(sk, fl4, &sk->sk_write_queue, &inet_sk(sk)->cork.base); } /* datagram.c */ -- cgit v1.2.2 From f5fca6086511294653a9e821f8e22f041415ba7b Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 8 May 2011 17:24:10 -0700 Subject: ipv4: Pass flow key down into ip_append_*(). This way rt->rt_dst accesses are unnecessary. Signed-off-by: David S. Miller --- include/net/ip.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index a4253795c5c5..0b30d3ab4a30 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -106,7 +106,7 @@ extern int __ip_local_out(struct sk_buff *skb); extern int ip_local_out(struct sk_buff *skb); extern int ip_queue_xmit(struct sk_buff *skb, struct flowi *fl); extern void ip_init(void); -extern int ip_append_data(struct sock *sk, +extern int ip_append_data(struct sock *sk, struct flowi4 *fl4, int getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb), void *from, int len, int protolen, @@ -114,7 +114,7 @@ extern int ip_append_data(struct sock *sk, struct rtable **rt, unsigned int flags); extern int ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb); -extern ssize_t ip_append_page(struct sock *sk, struct page *page, +extern ssize_t ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page, int offset, size_t size, int flags); extern struct sk_buff *__ip_make_skb(struct sock *sk, struct flowi4 *fl4, -- cgit v1.2.2 From 0a5ebb8000c5362be368df9d197943deb06b6916 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 9 May 2011 13:22:43 -0700 Subject: ipv4: Pass explicit daddr arg to ip_send_reply(). This eliminates an access to rt->rt_src. Signed-off-by: David S. Miller --- include/net/ip.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net/ip.h') diff --git a/include/net/ip.h b/include/net/ip.h index 0b30d3ab4a30..66dd49149208 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -174,8 +174,8 @@ static inline __u8 ip_reply_arg_flowi_flags(const struct ip_reply_arg *arg) return (arg->flags & IP_REPLY_ARG_NOSRCCHECK) ? FLOWI_FLAG_ANYSRC : 0; } -void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *arg, - unsigned int len); +void ip_send_reply(struct sock *sk, struct sk_buff *skb, __be32 daddr, + struct ip_reply_arg *arg, unsigned int len); struct ipv4_config { int log_martians; -- cgit v1.2.2