1 files changed, 92 insertions, 69 deletions

diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index a7aea2048a0d..6bf89a6312bc 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -83,6 +83,10 @@
 int sysctl_ip_default_ttl __read_mostly = IPDEFTTL;
 EXPORT_SYMBOL(sysctl_ip_default_ttl);
 
+static int ip_fragment(struct sock *sk, struct sk_buff *skb,
+		       unsigned int mtu,
+		       int (*output)(struct sock *, struct sk_buff *));
+
 /* Generate a checksum for an outgoing IP datagram. */
 void ip_send_check(struct iphdr *iph)
 {
@@ -91,14 +95,19 @@ void ip_send_check(struct iphdr *iph)
 }
 EXPORT_SYMBOL(ip_send_check);
 
-int __ip_local_out(struct sk_buff *skb)
+static int __ip_local_out_sk(struct sock *sk, struct sk_buff *skb)
 {
 	struct iphdr *iph = ip_hdr(skb);
 
 	iph->tot_len = htons(skb->len);
 	ip_send_check(iph);
-	return nf_hook(NFPROTO_IPV4, NF_INET_LOCAL_OUT, skb, NULL,
-		       skb_dst(skb)->dev, dst_output);
+	return nf_hook(NFPROTO_IPV4, NF_INET_LOCAL_OUT, sk, skb, NULL,
+		       skb_dst(skb)->dev, dst_output_sk);
+}
+
+int __ip_local_out(struct sk_buff *skb)
+{
+	return __ip_local_out_sk(skb->sk, skb);
 }
 
 int ip_local_out_sk(struct sock *sk, struct sk_buff *skb)
@@ -148,7 +157,7 @@ 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;
@@ -163,7 +172,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk,
 }
 EXPORT_SYMBOL_GPL(ip_build_and_send_pkt);
 
-static inline int ip_finish_output2(struct sk_buff *skb)
+static int ip_finish_output2(struct sock *sk, struct sk_buff *skb)
 {
 	struct dst_entry *dst = skb_dst(skb);
 	struct rtable *rt = (struct rtable *)dst;
@@ -182,7 +191,7 @@ static inline int ip_finish_output2(struct sk_buff *skb)
 		struct sk_buff *skb2;
 
 		skb2 = skb_realloc_headroom(skb, LL_RESERVED_SPACE(dev));
-		if (skb2 == NULL) {
+		if (!skb2) {
 			kfree_skb(skb);
 			return -ENOMEM;
 		}
@@ -211,7 +220,8 @@ static inline int ip_finish_output2(struct sk_buff *skb)
 	return -EINVAL;
 }
 
-static int ip_finish_output_gso(struct sk_buff *skb)
+static int ip_finish_output_gso(struct sock *sk, struct sk_buff *skb,
+				unsigned int mtu)
 {
 	netdev_features_t features;
 	struct sk_buff *segs;
@@ -219,8 +229,8 @@ static int ip_finish_output_gso(struct sk_buff *skb)
 
 	/* common case: locally created skb or seglen is <= mtu */
 	if (((IPCB(skb)->flags & IPSKB_FORWARDED) == 0) ||
-	      skb_gso_network_seglen(skb) <= ip_skb_dst_mtu(skb))
-		return ip_finish_output2(skb);
+	      skb_gso_network_seglen(skb) <= mtu)
+		return ip_finish_output2(sk, skb);
 
 	/* Slowpath -  GSO segment length is exceeding the dst MTU.
 	 *
@@ -243,7 +253,7 @@ static int ip_finish_output_gso(struct sk_buff *skb)
 		int err;
 
 		segs->next = NULL;
-		err = ip_fragment(segs, ip_finish_output2);
+		err = ip_fragment(sk, segs, mtu, ip_finish_output2);
 
 		if (err && ret == 0)
 			ret = err;
@@ -253,22 +263,25 @@ static int ip_finish_output_gso(struct sk_buff *skb)
 	return ret;
 }
 
-static int ip_finish_output(struct sk_buff *skb)
+static int ip_finish_output(struct sock *sk, struct sk_buff *skb)
 {
+	unsigned int mtu;
+
 #if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM)
 	/* Policy lookup after SNAT yielded a new policy */
-	if (skb_dst(skb)->xfrm != NULL) {
+	if (skb_dst(skb)->xfrm) {
 		IPCB(skb)->flags |= IPSKB_REROUTED;
-		return dst_output(skb);
+		return dst_output_sk(sk, skb);
 	}
 #endif
+	mtu = ip_skb_dst_mtu(skb);
 	if (skb_is_gso(skb))
-		return ip_finish_output_gso(skb);
+		return ip_finish_output_gso(sk, skb, mtu);
 
-	if (skb->len > ip_skb_dst_mtu(skb))
-		return ip_fragment(skb, ip_finish_output2);
+	if (skb->len > mtu || (IPCB(skb)->flags & IPSKB_FRAG_PMTU))
+		return ip_fragment(sk, skb, mtu, ip_finish_output2);
 
-	return ip_finish_output2(skb);
+	return ip_finish_output2(sk, skb);
 }
 
 int ip_mc_output(struct sock *sk, struct sk_buff *skb)
@@ -307,7 +320,7 @@ int ip_mc_output(struct sock *sk, struct sk_buff *skb)
 			struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC);
 			if (newskb)
 				NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING,
-					newskb, NULL, newskb->dev,
+					sk, newskb, NULL, newskb->dev,
 					dev_loopback_xmit);
 		}
 
@@ -322,11 +335,11 @@ int ip_mc_output(struct sock *sk, struct sk_buff *skb)
 	if (rt->rt_flags&RTCF_BROADCAST) {
 		struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC);
 		if (newskb)
-			NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING, newskb,
+			NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, newskb,
 				NULL, newskb->dev, dev_loopback_xmit);
 	}
 
-	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, skb, NULL,
+	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb, NULL,
 			    skb->dev, ip_finish_output,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
@@ -340,7 +353,8 @@ int ip_output(struct sock *sk, struct sk_buff *skb)
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_IP);
 
-	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, skb, NULL, dev,
+	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, sk, skb,
+			    NULL, dev,
 			    ip_finish_output,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
@@ -376,12 +390,12 @@ int ip_queue_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl)
 	inet_opt = rcu_dereference(inet->inet_opt);
 	fl4 = &fl->u.ip4;
 	rt = skb_rtable(skb);
-	if (rt != NULL)
+	if (rt)
 		goto packet_routed;
 
 	/* Make sure we can route this packet. */
 	rt = (struct rtable *)__sk_dst_check(sk, 0);
-	if (rt == NULL) {
+	if (!rt) {
 		__be32 daddr;
 
 		/* Use correct destination address if we have options. */
@@ -430,7 +444,8 @@ 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;
@@ -448,7 +463,6 @@ no_route:
 }
 EXPORT_SYMBOL(ip_queue_xmit);
 
-
 static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from)
 {
 	to->pkt_type = from->pkt_type;
@@ -472,6 +486,31 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from)
 	skb_copy_secmark(to, from);
 }
 
+static int ip_fragment(struct sock *sk, struct sk_buff *skb,
+		       unsigned int mtu,
+		       int (*output)(struct sock *, struct sk_buff *))
+{
+	struct iphdr *iph = ip_hdr(skb);
+
+	if ((iph->frag_off & htons(IP_DF)) == 0)
+		return ip_do_fragment(sk, skb, output);
+
+	if (unlikely(!skb->ignore_df ||
+		     (IPCB(skb)->frag_max_size &&
+		      IPCB(skb)->frag_max_size > mtu))) {
+		struct rtable *rt = skb_rtable(skb);
+		struct net_device *dev = rt->dst.dev;
+
+		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
+		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
+			  htonl(mtu));
+		kfree_skb(skb);
+		return -EMSGSIZE;
+	}
+
+	return ip_do_fragment(sk, skb, output);
+}
+
 /*
  *	This IP datagram is too large to be sent in one piece.  Break it up into
  *	smaller pieces (each of size equal to IP header plus
@@ -479,7 +518,8 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from)
  *	single device frame, and queue such a frame for sending.
  */
 
-int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
+int ip_do_fragment(struct sock *sk, struct sk_buff *skb,
+		   int (*output)(struct sock *, struct sk_buff *))
 {
 	struct iphdr *iph;
 	int ptr;
@@ -500,15 +540,8 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 	iph = ip_hdr(skb);
 
 	mtu = ip_skb_dst_mtu(skb);
-	if (unlikely(((iph->frag_off & htons(IP_DF)) && !skb->ignore_df) ||
-		     (IPCB(skb)->frag_max_size &&
-		      IPCB(skb)->frag_max_size > mtu))) {
-		IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
-		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
-			  htonl(mtu));
-		kfree_skb(skb);
-		return -EMSGSIZE;
-	}
+	if (IPCB(skb)->frag_max_size && IPCB(skb)->frag_max_size < mtu)
+		mtu = IPCB(skb)->frag_max_size;
 
 	/*
 	 *	Setup starting values.
@@ -516,10 +549,6 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 
 	hlen = iph->ihl * 4;
 	mtu = mtu - hlen;	/* Size of data space */
-#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
-	if (skb->nf_bridge)
-		mtu -= nf_bridge_mtu_reduction(skb);
-#endif
 	IPCB(skb)->flags |= IPSKB_FRAG_COMPLETE;
 
 	/* When frag_list is given, use it. First, check its validity:
@@ -586,13 +615,13 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 					ip_options_fragment(frag);
 				offset += skb->len - hlen;
 				iph->frag_off = htons(offset>>3);
-				if (frag->next != NULL)
+				if (frag->next)
 					iph->frag_off |= htons(IP_MF);
 				/* Ready, complete checksum */
 				ip_send_check(iph);
 			}
 
-			err = output(skb);
+			err = output(sk, skb);
 
 			if (!err)
 				IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGCREATES);
@@ -636,10 +665,7 @@ slow_path:
 	left = skb->len - hlen;		/* Space per frame */
 	ptr = hlen;		/* Where to start from */
 
-	/* for bridged IP traffic encapsulated inside f.e. a vlan header,
-	 * we need to make room for the encapsulating header
-	 */
-	ll_rs = LL_RESERVED_SPACE_EXTRA(rt->dst.dev, nf_bridge_pad(skb));
+	ll_rs = LL_RESERVED_SPACE(rt->dst.dev);
 
 	/*
 	 *	Fragment the datagram.
@@ -707,6 +733,9 @@ slow_path:
 		iph = ip_hdr(skb2);
 		iph->frag_off = htons((offset >> 3));
 
+		if (IPCB(skb)->flags & IPSKB_FRAG_PMTU)
+			iph->frag_off |= htons(IP_DF);
+
 		/* ANK: dirty, but effective trick. Upgrade options only if
 		 * the segment to be fragmented was THE FIRST (otherwise,
 		 * options are already fixed) and make it ONCE
@@ -732,7 +761,7 @@ slow_path:
 
 		ip_send_check(iph);
 
-		err = output(skb2);
+		err = output(sk, skb2);
 		if (err)
 			goto fail;
 
@@ -747,7 +776,7 @@ fail:
 	IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS);
 	return err;
 }
-EXPORT_SYMBOL(ip_fragment);
+EXPORT_SYMBOL(ip_do_fragment);
 
 int
 ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb)
@@ -792,12 +821,13 @@ static inline int ip_ufo_append_data(struct sock *sk,
 	 * device, so create one single skb packet containing complete
 	 * udp datagram
 	 */
-	if ((skb = skb_peek_tail(queue)) == NULL) {
+	skb = skb_peek_tail(queue);
+	if (!skb) {
 		skb = sock_alloc_send_skb(sk,
 			hh_len + fragheaderlen + transhdrlen + 20,
 			(flags & MSG_DONTWAIT), &err);
 
-		if (skb == NULL)
+		if (!skb)
 			return err;
 
 		/* reserve space for Hardware header */
@@ -814,7 +844,6 @@ static inline int ip_ufo_append_data(struct sock *sk,
 
 		skb->csum = 0;
 
-
 		__skb_queue_tail(queue, skb);
 	} else if (skb_is_gso(skb)) {
 		goto append;
@@ -963,10 +992,10 @@ alloc_new_skb:
 					skb = sock_wmalloc(sk,
 							   alloclen + hh_len + 15, 1,
 							   sk->sk_allocation);
-				if (unlikely(skb == NULL))
+				if (unlikely(!skb))
 					err = -ENOBUFS;
 			}
-			if (skb == NULL)
+			if (!skb)
 				goto error;
 
 			/*
@@ -1090,10 +1119,10 @@ static int ip_setup_cork(struct sock *sk, struct inet_cork *cork,
 	 */
 	opt = ipc->opt;
 	if (opt) {
-		if (cork->opt == NULL) {
+		if (!cork->opt) {
 			cork->opt = kmalloc(sizeof(struct ip_options) + 40,
 					    sk->sk_allocation);
-			if (unlikely(cork->opt == NULL))
+			if (unlikely(!cork->opt))
 				return -ENOBUFS;
 		}
 		memcpy(cork->opt, &opt->opt, sizeof(struct ip_options) + opt->opt.optlen);
@@ -1200,7 +1229,8 @@ ssize_t	ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page,
 		return -EMSGSIZE;
 	}
 
-	if ((skb = skb_peek_tail(&sk->sk_write_queue)) == NULL)
+	skb = skb_peek_tail(&sk->sk_write_queue);
+	if (!skb)
 		return -EINVAL;
 
 	cork->length += size;
@@ -1211,13 +1241,10 @@ ssize_t	ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page,
 		skb_shinfo(skb)->gso_type = SKB_GSO_UDP;
 	}
 
-
 	while (size > 0) {
-		int i;
-
-		if (skb_is_gso(skb))
+		if (skb_is_gso(skb)) {
 			len = size;
-		else {
+		} else {
 
 			/* Check if the remaining data fits into current packet. */
 			len = mtu - skb->len;
@@ -1269,15 +1296,10 @@ ssize_t	ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page,
 			continue;
 		}
 
-		i = skb_shinfo(skb)->nr_frags;
 		if (len > size)
 			len = size;
-		if (skb_can_coalesce(skb, i, page, offset)) {
-			skb_frag_size_add(&skb_shinfo(skb)->frags[i-1], len);
-		} else if (i < MAX_SKB_FRAGS) {
-			get_page(page);
-			skb_fill_page_desc(skb, i, page, offset, len);
-		} else {
+
+		if (skb_append_pagefrags(skb, page, offset, len)) {
 			err = -EMSGSIZE;
 			goto error;
 		}
@@ -1331,7 +1353,8 @@ struct sk_buff *__ip_make_skb(struct sock *sk,
 	__be16 df = 0;
 	__u8 ttl;
 
-	if ((skb = __skb_dequeue(queue)) == NULL)
+	skb = __skb_dequeue(queue);
+	if (!skb)
 		goto out;
 	tail_skb = &(skb_shinfo(skb)->frag_list);
 
@@ -1382,7 +1405,7 @@ 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;