From b6a7719aedd7e5c0f2df7641aa47386111682df4 Mon Sep 17 00:00:00 2001 From: Hannes Frederic Sowa Date: Wed, 25 Mar 2015 17:07:44 +0100 Subject: [PATCH] ipv4: hash net ptr into fragmentation bucket selection As namespaces are sometimes used with overlapping ip address ranges, we should also use the namespace as input to the hash to select the ip fragmentation counter bucket. Cc: Eric Dumazet Cc: Flavio Leitner Signed-off-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- drivers/net/ppp/pptp.c | 2 +- include/net/ip.h | 12 +++++++----- net/ipv4/igmp.c | 4 ++-- net/ipv4/ip_output.c | 7 ++++--- net/ipv4/ip_tunnel_core.c | 2 +- net/ipv4/ipmr.c | 7 ++++--- net/ipv4/raw.c | 2 +- net/ipv4/route.c | 4 ++-- net/ipv4/xfrm4_mode_tunnel.c | 2 +- net/netfilter/ipvs/ip_vs_xmit.c | 5 +++-- 10 files changed, 26 insertions(+), 21 deletions(-) diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c index 1dc628ffce2b..e3bfbd4d0136 100644 --- a/drivers/net/ppp/pptp.c +++ b/drivers/net/ppp/pptp.c @@ -281,7 +281,7 @@ static int pptp_xmit(struct ppp_channel *chan, struct sk_buff *skb) nf_reset(skb); skb->ip_summed = CHECKSUM_NONE; - ip_select_ident(skb, NULL); + ip_select_ident(sock_net(sk), skb, NULL); ip_send_check(iph); ip_local_out(skb); diff --git a/include/net/ip.h b/include/net/ip.h index 025c61c0dffb..d0808a323763 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -318,9 +318,10 @@ static inline unsigned int ip_skb_dst_mtu(const struct sk_buff *skb) } u32 ip_idents_reserve(u32 hash, int segs); -void __ip_select_ident(struct iphdr *iph, int segs); +void __ip_select_ident(struct net *net, struct iphdr *iph, int segs); -static inline void ip_select_ident_segs(struct sk_buff *skb, struct sock *sk, int segs) +static inline void ip_select_ident_segs(struct net *net, struct sk_buff *skb, + struct sock *sk, int segs) { struct iphdr *iph = ip_hdr(skb); @@ -337,13 +338,14 @@ static inline void ip_select_ident_segs(struct sk_buff *skb, struct sock *sk, in iph->id = 0; } } else { - __ip_select_ident(iph, segs); + __ip_select_ident(net, iph, segs); } } -static inline void ip_select_ident(struct sk_buff *skb, struct sock *sk) +static inline void ip_select_ident(struct net *net, struct sk_buff *skb, + struct sock *sk) { - ip_select_ident_segs(skb, sk, 1); + ip_select_ident_segs(net, skb, sk, 1); } static inline __wsum inet_compute_pseudo(struct sk_buff *skb, int proto) diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index ad3f866085de..ad09213ac5b2 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -370,7 +370,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, unsigned int mtu) pip->saddr = fl4.saddr; pip->protocol = IPPROTO_IGMP; pip->tot_len = 0; /* filled in later */ - ip_select_ident(skb, NULL); + ip_select_ident(net, skb, NULL); ((u8 *)&pip[1])[0] = IPOPT_RA; ((u8 *)&pip[1])[1] = 4; ((u8 *)&pip[1])[2] = 0; @@ -714,7 +714,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc, iph->daddr = dst; iph->saddr = fl4.saddr; iph->protocol = IPPROTO_IGMP; - ip_select_ident(skb, NULL); + ip_select_ident(net, skb, NULL); ((u8 *)&iph[1])[0] = IPOPT_RA; ((u8 *)&iph[1])[1] = 4; ((u8 *)&iph[1])[2] = 0; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 90b49e88e84a..8259e777b249 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -148,7 +148,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk, iph->daddr = (opt && opt->opt.srr ? opt->opt.faddr : daddr); iph->saddr = saddr; iph->protocol = sk->sk_protocol; - ip_select_ident(skb, sk); + ip_select_ident(sock_net(sk), skb, sk); if (opt && opt->opt.optlen) { iph->ihl += opt->opt.optlen>>2; @@ -430,7 +430,8 @@ packet_routed: ip_options_build(skb, &inet_opt->opt, inet->inet_daddr, rt, 0); } - ip_select_ident_segs(skb, sk, skb_shinfo(skb)->gso_segs ?: 1); + ip_select_ident_segs(sock_net(sk), skb, sk, + skb_shinfo(skb)->gso_segs ?: 1); /* TODO : should we use skb->sk here instead of sk ? */ skb->priority = sk->sk_priority; @@ -1379,7 +1380,7 @@ struct sk_buff *__ip_make_skb(struct sock *sk, iph->ttl = ttl; iph->protocol = sk->sk_protocol; ip_copy_addrs(iph, fl4); - ip_select_ident(skb, sk); + ip_select_ident(net, skb, sk); if (opt) { iph->ihl += opt->optlen>>2; diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 88c386cf7d85..8c4dcc46acd2 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -74,7 +74,7 @@ int iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, iph->daddr = dst; iph->saddr = src; iph->ttl = ttl; - __ip_select_ident(iph, skb_shinfo(skb)->gso_segs ?: 1); + __ip_select_ident(sock_net(sk), iph, skb_shinfo(skb)->gso_segs ?: 1); err = ip_local_out_sk(sk, skb); if (unlikely(net_xmit_eval(err))) diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 5b188832800f..c688cd1b2110 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1642,7 +1642,8 @@ static struct notifier_block ip_mr_notifier = { * important for multicast video. */ -static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr) +static void ip_encap(struct net *net, struct sk_buff *skb, + __be32 saddr, __be32 daddr) { struct iphdr *iph; const struct iphdr *old_iph = ip_hdr(skb); @@ -1661,7 +1662,7 @@ static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr) iph->protocol = IPPROTO_IPIP; iph->ihl = 5; iph->tot_len = htons(skb->len); - ip_select_ident(skb, NULL); + ip_select_ident(net, skb, NULL); ip_send_check(iph); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); @@ -1758,7 +1759,7 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, * What do we do with netfilter? -- RR */ if (vif->flags & VIFF_TUNNEL) { - ip_encap(skb, vif->local, vif->remote); + ip_encap(net, skb, vif->local, vif->remote); /* FIXME: extra output firewall step used to be here. --RR */ vif->dev->stats.tx_packets++; vif->dev->stats.tx_bytes += skb->len; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 923cf538fce1..56946f47d446 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -404,7 +404,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4, iph->check = 0; iph->tot_len = htons(length); if (!iph->id) - ip_select_ident(skb, NULL); + ip_select_ident(net, skb, NULL); iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl); } diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 649c8a3f0189..be8703d02ef0 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -482,7 +482,7 @@ u32 ip_idents_reserve(u32 hash, int segs) } EXPORT_SYMBOL(ip_idents_reserve); -void __ip_select_ident(struct iphdr *iph, int segs) +void __ip_select_ident(struct net *net, struct iphdr *iph, int segs) { static u32 ip_idents_hashrnd __read_mostly; u32 hash, id; @@ -491,7 +491,7 @@ void __ip_select_ident(struct iphdr *iph, int segs) hash = jhash_3words((__force u32)iph->daddr, (__force u32)iph->saddr, - iph->protocol, + iph->protocol ^ net_hash_mix(net), ip_idents_hashrnd); id = ip_idents_reserve(hash, segs); iph->id = htons(id); diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c index 91771a7c802f..35feda676464 100644 --- a/net/ipv4/xfrm4_mode_tunnel.c +++ b/net/ipv4/xfrm4_mode_tunnel.c @@ -63,7 +63,7 @@ static int xfrm4_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) top_iph->saddr = x->props.saddr.a4; top_iph->daddr = x->id.daddr.a4; - ip_select_ident(skb, NULL); + ip_select_ident(dev_net(dst->dev), skb, NULL); return 0; } diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index f35c15b0de6b..bf02932b7188 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -924,7 +924,8 @@ int ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, struct ip_vs_protocol *pp, struct ip_vs_iphdr *ipvsh) { - struct netns_ipvs *ipvs = net_ipvs(skb_net(skb)); + struct net *net = skb_net(skb); + struct netns_ipvs *ipvs = net_ipvs(net); struct rtable *rt; /* Route to the other host */ __be32 saddr; /* Source for tunnel */ struct net_device *tdev; /* Device to other host */ @@ -991,7 +992,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, iph->daddr = cp->daddr.ip; iph->saddr = saddr; iph->ttl = ttl; - ip_select_ident(skb, NULL); + ip_select_ident(net, skb, NULL); /* Another hack: avoid icmp_send in ip_fragment */ skb->ignore_df = 1;