From 3820c3f3e41786322c0bb225b9c77b8deff869d1 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Thu, 29 Jun 2006 20:11:25 -0700 Subject: [TCP]: Reset gso_segs if packet is dodgy I wasn't paranoid enough in verifying GSO information. A bogus gso_segs could upset drivers as much as a bogus header would. Let's reset it in the per-protocol gso_segment functions. I didn't verify gso_size because that can be verified by the source of the dodgy packets. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 0336422c88a..0bb0ac96d67 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2166,13 +2166,19 @@ struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features) if (!pskb_may_pull(skb, thlen)) goto out; - segs = NULL; - if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) - goto out; - oldlen = (u16)~skb->len; __skb_pull(skb, thlen); + if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) { + /* Packet is from an untrusted source, reset gso_segs. */ + int mss = skb_shinfo(skb)->gso_size; + + skb_shinfo(skb)->gso_segs = (skb->len + mss - 1) / mss; + + segs = NULL; + goto out; + } + segs = skb_segment(skb, features); if (IS_ERR(segs)) goto out; -- cgit v1.2.3 From dd7271feba61d5dc0fab1cb5365db9926d35ea3a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 29 Jun 2006 21:40:23 -0700 Subject: [NETFILTER]: SCTP conntrack: fix crash triggered by packet without chunks When a packet without any chunks is received, the newconntrack variable in sctp_packet contains an out of bounds value that is used to look up an pointer from the array of timeouts, which is then dereferenced, resulting in a crash. Make sure at least a single chunk is present. Problem noticed by George A. Theall Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_proto_sctp.c | 2 +- net/netfilter/nf_conntrack_proto_sctp.c | 2 +- 2 files changed, 2 insertions(+), 2 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 0416073c560..2d3612cd5f1 100644 --- a/net/ipv4/netfilter/ip_conntrack_proto_sctp.c +++ b/net/ipv4/netfilter/ip_conntrack_proto_sctp.c @@ -254,7 +254,7 @@ static int do_basic_checks(struct ip_conntrack *conntrack, } DEBUGP("Basic checks passed\n"); - return 0; + return count == 0; } static int new_state(enum ip_conntrack_dir dir, diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index 0839b701b93..9bd8a7877fd 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -261,7 +261,7 @@ static int do_basic_checks(struct nf_conn *conntrack, } DEBUGP("Basic checks passed\n"); - return 0; + return count == 0; } static int new_state(enum ip_conntrack_dir dir, -- cgit v1.2.3 From 2889139a6acd2945f6143eb85f7dc2a22a352e1a Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 30 Jun 2006 13:35:46 -0700 Subject: [IPV6]: Remove redundant length check on input We don't need to check skb->len when we're just about to call pskb_may_pull since that checks it for us. Signed-off-by: Herbert Xu Acked-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/ip6_input.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index aceee252503..df8f051c0fc 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -84,14 +84,9 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt */ IP6CB(skb)->iif = skb->dst ? ((struct rt6_info *)skb->dst)->rt6i_idev->dev->ifindex : dev->ifindex; - if (skb->len < sizeof(struct ipv6hdr)) + if (unlikely(!pskb_may_pull(skb, sizeof(*hdr)))) goto err; - if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) { - IP6_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); - goto drop; - } - hdr = skb->nh.ipv6h; if (hdr->version != 6) -- cgit v1.2.3 From adcfc7d0b4d7bc3c7edac6fdde9f3ae510bd6054 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 30 Jun 2006 13:36:15 -0700 Subject: [IPV6]: Added GSO support for TCPv6 This patch adds GSO support for IPv6 and TCPv6. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 1 + net/ipv6/exthdrs.c | 4 ++-- net/ipv6/ipv6_sockglue.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++++ net/ipv6/tcp_ipv6.c | 1 + 4 files changed, 66 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 0bb0ac96d67..b7cf26e0e5c 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2215,6 +2215,7 @@ struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features) out: return segs; } +EXPORT_SYMBOL(tcp_tso_segment); extern void __skb_cb_too_small_for_tcp(int, int); extern struct tcp_congestion_ops tcp_reno; diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index a18d4256372..9d0ee7f0eeb 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -179,7 +179,7 @@ static int ipv6_destopt_rcv(struct sk_buff **skbp) static struct inet6_protocol destopt_protocol = { .handler = ipv6_destopt_rcv, - .flags = INET6_PROTO_NOPOLICY, + .flags = INET6_PROTO_NOPOLICY | INET6_PROTO_GSO_EXTHDR, }; void __init ipv6_destopt_init(void) @@ -340,7 +340,7 @@ looped_back: static struct inet6_protocol rthdr_protocol = { .handler = ipv6_rthdr_rcv, - .flags = INET6_PROTO_NOPOLICY, + .flags = INET6_PROTO_NOPOLICY | INET6_PROTO_GSO_EXTHDR, }; void __init ipv6_rthdr_init(void) diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 4c20eeb3d56..25f8bf8ac49 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -58,9 +58,71 @@ DEFINE_SNMP_STAT(struct ipstats_mib, ipv6_statistics) __read_mostly; +static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb, int features) +{ + struct sk_buff *segs = ERR_PTR(-EINVAL); + struct ipv6hdr *ipv6h; + struct inet6_protocol *ops; + int proto; + + if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h)))) + goto out; + + ipv6h = skb->nh.ipv6h; + proto = ipv6h->nexthdr; + __skb_pull(skb, sizeof(*ipv6h)); + + rcu_read_lock(); + for (;;) { + struct ipv6_opt_hdr *opth; + int len; + + if (proto != NEXTHDR_HOP) { + ops = rcu_dereference(inet6_protos[proto]); + + if (unlikely(!ops)) + goto unlock; + + if (!(ops->flags & INET6_PROTO_GSO_EXTHDR)) + break; + } + + if (unlikely(!pskb_may_pull(skb, 8))) + goto unlock; + + opth = (void *)skb->data; + len = opth->hdrlen * 8 + 8; + + if (unlikely(!pskb_may_pull(skb, len))) + goto unlock; + + proto = opth->nexthdr; + __skb_pull(skb, len); + } + + skb->h.raw = skb->data; + if (likely(ops->gso_segment)) + segs = ops->gso_segment(skb, features); + +unlock: + rcu_read_unlock(); + + if (unlikely(IS_ERR(segs))) + goto out; + + for (skb = segs; skb; skb = skb->next) { + ipv6h = skb->nh.ipv6h; + ipv6h->payload_len = htons(skb->len - skb->mac_len); + } + +out: + return segs; +} + static struct packet_type ipv6_packet_type = { .type = __constant_htons(ETH_P_IPV6), .func = ipv6_rcv, + .gso_segment = ipv6_gso_segment, }; struct ip6_ra_chain *ip6_ra_chain; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index b36d5b2e7c3..bf7f8c26c48 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1606,6 +1606,7 @@ struct proto tcpv6_prot = { static struct inet6_protocol tcpv6_protocol = { .handler = tcp_v6_rcv, .err_handler = tcp_v6_err, + .gso_segment = tcp_tso_segment, .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, }; -- cgit v1.2.3 From bcd76111178ebccedd46a9b3eaff65c78e5a70af Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 30 Jun 2006 13:36:35 -0700 Subject: [NET]: Generalise TSO-specific bits from skb_setup_caps This patch generalises the TSO-specific bits from sk_setup_caps by adding the sk_gso_type member to struct sock. This makes sk_setup_caps generic so that it can be used by TCPv6 or UFO. The only catch is that whoever uses this must provide a GSO implementation for their protocol which I think is a fair deal :) For now UFO continues to live without a GSO implementation which is OK since it doesn't use the sock caps field at the moment. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 2 +- net/ipv4/tcp_ipv4.c | 2 ++ net/ipv4/tcp_output.c | 9 +++------ 3 files changed, 6 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b7cf26e0e5c..59e30bab060 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -643,7 +643,7 @@ static inline int select_size(struct sock *sk, struct tcp_sock *tp) int tmp = tp->mss_cache; if (sk->sk_route_caps & NETIF_F_SG) { - if (sk->sk_route_caps & NETIF_F_TSO) + if (sk_can_gso(sk)) tmp = 0; else { int pgbreak = SKB_MAX_HEAD(MAX_TCP_HEADER); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 4c6ef47eb1c..38e001076a5 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -242,6 +242,7 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) goto failure; /* OK, now commit destination to socket. */ + sk->sk_gso_type = SKB_GSO_TCPV4; sk_setup_caps(sk, &rt->u.dst); if (!tp->write_seq) @@ -884,6 +885,7 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, if (!newsk) goto exit; + newsk->sk_gso_type = SKB_GSO_TCPV4; sk_setup_caps(newsk, dst); newtp = tcp_sk(newsk); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 5a7cb4a9c86..5c08ea20a18 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -510,8 +510,7 @@ static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb) static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned int mss_now) { - if (skb->len <= mss_now || - !(sk->sk_route_caps & NETIF_F_TSO)) { + if (skb->len <= mss_now || !sk_can_gso(sk)) { /* Avoid the costly divide in the normal * non-TSO case. */ @@ -525,7 +524,7 @@ static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned factor /= mss_now; skb_shinfo(skb)->gso_segs = factor; skb_shinfo(skb)->gso_size = mss_now; - skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4; + skb_shinfo(skb)->gso_type = sk->sk_gso_type; } } @@ -824,9 +823,7 @@ unsigned int tcp_current_mss(struct sock *sk, int large_allowed) mss_now = tp->mss_cache; - if (large_allowed && - (sk->sk_route_caps & NETIF_F_TSO) && - !tp->urg_mode) + if (large_allowed && sk_can_gso(sk) && !tp->urg_mode) doing_tso = 1; if (dst) { -- cgit v1.2.3 From f83ef8c0b58dac17211a4c0b6df0e2b1bd6637b1 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 30 Jun 2006 13:37:03 -0700 Subject: [IPV6]: Added GSO support for TCPv6 This patch adds GSO support for IPv6 and TCPv6. This is based on a patch by Ananda Raju . His original description is: This patch enables TSO over IPv6. Currently Linux network stacks restricts TSO over IPv6 by clearing of the NETIF_F_TSO bit from "dev->features". This patch will remove this restriction. This patch will introduce a new flag NETIF_F_TSO6 which will be used to check whether device supports TSO over IPv6. If device support TSO over IPv6 then we don't clear of NETIF_F_TSO and which will make the TCP layer to create TSO packets. Any device supporting TSO over IPv6 will set NETIF_F_TSO6 flag in "dev->features" along with NETIF_F_TSO. In case when user disables TSO using ethtool, NETIF_F_TSO will get cleared from "dev->features". So even if we have NETIF_F_TSO6 we don't get TSO packets created by TCP layer. SKB_GSO_TCPV4 renamed to SKB_GSO_TCP to make it generic GSO packet. SKB_GSO_UDPV4 renamed to SKB_GSO_UDP as UFO is not a IPv4 feature. UFO is supported over IPv6 also The following table shows there is significant improvement in throughput with normal frames and CPU usage for both normal and jumbo. -------------------------------------------------- | | 1500 | 9600 | | ------------------|-------------------| | | thru CPU | thru CPU | -------------------------------------------------- | TSO OFF | 2.00 5.5% id | 5.66 20.0% id | -------------------------------------------------- | TSO ON | 2.63 78.0 id | 5.67 39.0% id | -------------------------------------------------- Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/ip_output.c | 4 ++-- net/ipv6/af_inet6.c | 2 -- net/ipv6/inet6_connection_sock.c | 2 -- net/ipv6/ip6_output.c | 4 ++-- net/ipv6/tcp_ipv6.c | 6 ++---- 5 files changed, 6 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 7624fd1d8f9..243d2a76336 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -744,7 +744,7 @@ static inline int ip_ufo_append_data(struct sock *sk, if (!err) { /* specify the length of each IP datagram fragment*/ skb_shinfo(skb)->gso_size = mtu - fragheaderlen; - skb_shinfo(skb)->gso_type = SKB_GSO_UDPV4; + skb_shinfo(skb)->gso_type = SKB_GSO_UDP; __skb_queue_tail(&sk->sk_write_queue, skb); return 0; @@ -1089,7 +1089,7 @@ ssize_t ip_append_page(struct sock *sk, struct page *page, if ((sk->sk_protocol == IPPROTO_UDP) && (rt->u.dst.dev->features & NETIF_F_UFO)) { skb_shinfo(skb)->gso_size = mtu - fragheaderlen; - skb_shinfo(skb)->gso_type = SKB_GSO_UDPV4; + skb_shinfo(skb)->gso_type = SKB_GSO_UDP; } diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index e19457fe4f6..0f26073117a 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -660,8 +660,6 @@ int inet6_sk_rebuild_header(struct sock *sk) } ip6_dst_store(sk, dst, NULL); - sk->sk_route_caps = dst->dev->features & - ~(NETIF_F_IP_CSUM | NETIF_F_TSO); } return 0; diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index eb2865d5ae2..5624f373164 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -187,8 +187,6 @@ int inet6_csk_xmit(struct sk_buff *skb, int ipfragok) } ip6_dst_store(sk, dst, NULL); - sk->sk_route_caps = dst->dev->features & - ~(NETIF_F_IP_CSUM | NETIF_F_TSO); } skb->dst = dst_clone(dst); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index abb94de3376..11007c75ae0 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -230,7 +230,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, skb->priority = sk->sk_priority; mtu = dst_mtu(dst); - if ((skb->len <= mtu) || ipfragok) { + if ((skb->len <= mtu) || ipfragok || skb_shinfo(skb)->gso_size) { IP6_INC_STATS(IPSTATS_MIB_OUTREQUESTS); return NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev, dst_output); @@ -835,7 +835,7 @@ static inline int ip6_ufo_append_data(struct sock *sk, /* specify the length of each IP datagram fragment*/ skb_shinfo(skb)->gso_size = mtu - fragheaderlen - sizeof(struct frag_hdr); - skb_shinfo(skb)->gso_type = SKB_GSO_UDPV4; + skb_shinfo(skb)->gso_type = SKB_GSO_UDP; ipv6_select_ident(skb, &fhdr); skb_shinfo(skb)->ip6_frag_id = fhdr.identification; __skb_queue_tail(&sk->sk_write_queue, skb); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index bf7f8c26c48..7ea5bea49aa 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -270,9 +270,8 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, ipv6_addr_copy(&np->saddr, saddr); inet->rcv_saddr = LOOPBACK4_IPV6; + sk->sk_gso_type = SKB_GSO_TCPV6; ip6_dst_store(sk, dst, NULL); - sk->sk_route_caps = dst->dev->features & - ~(NETIF_F_IP_CSUM | NETIF_F_TSO); icsk->icsk_ext_hdr_len = 0; if (np->opt) @@ -930,9 +929,8 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, * comment in that function for the gory details. -acme */ + sk->sk_gso_type = SKB_GSO_TCPV6; ip6_dst_store(newsk, dst, NULL); - newsk->sk_route_caps = dst->dev->features & - ~(NETIF_F_IP_CSUM | NETIF_F_TSO); newtcp6sk = (struct tcp6_sock *)newsk; inet_sk(newsk)->pinet6 = &newtcp6sk->inet6; -- cgit v1.2.3