summaryrefslogtreecommitdiff
path: root/net/ipv4
diff options
context:
space:
mode:
Diffstat (limited to 'net/ipv4')
-rw-r--r--net/ipv4/fib_frontend.c1
-rw-r--r--net/ipv4/inet_diag.c20
-rw-r--r--net/ipv4/ip_output.c3
-rw-r--r--net/ipv4/ip_tunnel_core.c1
-rw-r--r--net/ipv4/ip_vti.c2
-rw-r--r--net/ipv4/proc.c1
-rw-r--r--net/ipv4/route.c14
-rw-r--r--net/ipv4/syncookies.c2
-rw-r--r--net/ipv4/tcp.c3
-rw-r--r--net/ipv4/tcp_input.c32
-rw-r--r--net/ipv4/tcp_ipv4.c6
11 files changed, 62 insertions, 23 deletions
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 41079490a118..86a23e4a6a50 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -362,6 +362,7 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst,
fl4.flowi4_tun_key.tun_id = 0;
fl4.flowi4_flags = 0;
fl4.flowi4_uid = sock_net_uid(net, NULL);
+ fl4.flowi4_multipath_hash = 0;
no_addr = idev->ifa_list == NULL;
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index 4a98dd736270..f1bd95f243b3 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -186,8 +186,8 @@ errout:
}
EXPORT_SYMBOL_GPL(inet_diag_msg_attrs_fill);
-static void inet_diag_parse_attrs(const struct nlmsghdr *nlh, int hdrlen,
- struct nlattr **req_nlas)
+static int inet_diag_parse_attrs(const struct nlmsghdr *nlh, int hdrlen,
+ struct nlattr **req_nlas)
{
struct nlattr *nla;
int remaining;
@@ -195,9 +195,13 @@ static void inet_diag_parse_attrs(const struct nlmsghdr *nlh, int hdrlen,
nlmsg_for_each_attr(nla, nlh, hdrlen, remaining) {
int type = nla_type(nla);
+ if (type == INET_DIAG_REQ_PROTOCOL && nla_len(nla) != sizeof(u32))
+ return -EINVAL;
+
if (type < __INET_DIAG_REQ_MAX)
req_nlas[type] = nla;
}
+ return 0;
}
static int inet_diag_get_protocol(const struct inet_diag_req_v2 *req,
@@ -574,7 +578,10 @@ static int inet_diag_cmd_exact(int cmd, struct sk_buff *in_skb,
int err, protocol;
memset(&dump_data, 0, sizeof(dump_data));
- inet_diag_parse_attrs(nlh, hdrlen, dump_data.req_nlas);
+ err = inet_diag_parse_attrs(nlh, hdrlen, dump_data.req_nlas);
+ if (err)
+ return err;
+
protocol = inet_diag_get_protocol(req, &dump_data);
handler = inet_diag_lock_handler(protocol);
@@ -1180,8 +1187,11 @@ static int __inet_diag_dump_start(struct netlink_callback *cb, int hdrlen)
if (!cb_data)
return -ENOMEM;
- inet_diag_parse_attrs(nlh, hdrlen, cb_data->req_nlas);
-
+ err = inet_diag_parse_attrs(nlh, hdrlen, cb_data->req_nlas);
+ if (err) {
+ kfree(cb_data);
+ return err;
+ }
nla = cb_data->inet_diag_nla_bc;
if (nla) {
err = inet_diag_bc_audit(nla, skb);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 61f802d5350c..e6f2ada9e7d5 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -74,6 +74,7 @@
#include <net/icmp.h>
#include <net/checksum.h>
#include <net/inetpeer.h>
+#include <net/inet_ecn.h>
#include <net/lwtunnel.h>
#include <linux/bpf-cgroup.h>
#include <linux/igmp.h>
@@ -1703,7 +1704,7 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb,
if (IS_ERR(rt))
return;
- inet_sk(sk)->tos = arg->tos;
+ inet_sk(sk)->tos = arg->tos & ~INET_ECN_MASK;
sk->sk_protocol = ip_hdr(skb)->protocol;
sk->sk_bound_dev_if = arg->bound_dev_if;
diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c
index 75c6013ff9a4..b2ea1a8c5fd6 100644
--- a/net/ipv4/ip_tunnel_core.c
+++ b/net/ipv4/ip_tunnel_core.c
@@ -554,6 +554,7 @@ static int ip_tun_parse_opts_vxlan(struct nlattr *attr,
attr = tb[LWTUNNEL_IP_OPT_VXLAN_GBP];
md->gbp = nla_get_u32(attr);
+ md->gbp &= VXLAN_GBP_MASK;
info->key.tun_flags |= TUNNEL_VXLAN_OPT;
}
diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c
index 49daaed89764..f687abb069fa 100644
--- a/net/ipv4/ip_vti.c
+++ b/net/ipv4/ip_vti.c
@@ -490,6 +490,7 @@ static struct xfrm_tunnel vti_ipip_handler __read_mostly = {
.priority = 0,
};
+#if IS_ENABLED(CONFIG_IPV6)
static struct xfrm_tunnel vti_ipip6_handler __read_mostly = {
.handler = vti_rcv_tunnel,
.cb_handler = vti_rcv_cb,
@@ -497,6 +498,7 @@ static struct xfrm_tunnel vti_ipip6_handler __read_mostly = {
.priority = 0,
};
#endif
+#endif
static int __net_init vti_init_net(struct net *net)
{
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 1074df726ec0..8d5e1695b9aa 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -293,6 +293,7 @@ static const struct snmp_mib snmp4_net_list[] = {
SNMP_MIB_ITEM("TcpTimeoutRehash", LINUX_MIB_TCPTIMEOUTREHASH),
SNMP_MIB_ITEM("TcpDuplicateDataRehash", LINUX_MIB_TCPDUPLICATEDATAREHASH),
SNMP_MIB_ITEM("TCPDSACKRecvSegs", LINUX_MIB_TCPDSACKRECVSEGS),
+ SNMP_MIB_ITEM("TCPDSACKIgnoredDubious", LINUX_MIB_TCPDSACKIGNOREDDUBIOUS),
SNMP_MIB_SENTINEL
};
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 8ca6bcab7b03..58642b29a499 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -786,8 +786,10 @@ static void __ip_do_redirect(struct rtable *rt, struct sk_buff *skb, struct flow
neigh_event_send(n, NULL);
} else {
if (fib_lookup(net, fl4, &res, 0) == 0) {
- struct fib_nh_common *nhc = FIB_RES_NHC(res);
+ struct fib_nh_common *nhc;
+ fib_select_path(net, &res, fl4, skb);
+ nhc = FIB_RES_NHC(res);
update_or_create_fnhe(nhc, fl4->daddr, new_gw,
0, false,
jiffies + ip_rt_gc_timeout);
@@ -1013,6 +1015,7 @@ out: kfree_skb(skb);
static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu)
{
struct dst_entry *dst = &rt->dst;
+ struct net *net = dev_net(dst->dev);
u32 old_mtu = ipv4_mtu(dst);
struct fib_result res;
bool lock = false;
@@ -1033,9 +1036,11 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu)
return;
rcu_read_lock();
- if (fib_lookup(dev_net(dst->dev), fl4, &res, 0) == 0) {
- struct fib_nh_common *nhc = FIB_RES_NHC(res);
+ if (fib_lookup(net, fl4, &res, 0) == 0) {
+ struct fib_nh_common *nhc;
+ fib_select_path(net, &res, fl4, NULL);
+ nhc = FIB_RES_NHC(res);
update_or_create_fnhe(nhc, fl4->daddr, 0, mtu, lock,
jiffies + ip_rt_mtu_expires);
}
@@ -2147,6 +2152,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr,
fl4.daddr = daddr;
fl4.saddr = saddr;
fl4.flowi4_uid = sock_net_uid(net, NULL);
+ fl4.flowi4_multipath_hash = 0;
if (fib4_rules_early_flow_dissect(net, skb, &fl4, &_flkeys)) {
flkeys = &_flkeys;
@@ -2667,8 +2673,6 @@ struct rtable *ip_route_output_key_hash_rcu(struct net *net, struct flowi4 *fl4,
fib_select_path(net, res, fl4, skb);
dev_out = FIB_RES_DEV(*res);
- fl4->flowi4_oif = dev_out->ifindex;
-
make_route:
rth = __mkroute_output(res, fl4, orig_oif, dev_out, flags);
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index f0794f0232ba..e03756631541 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -214,7 +214,7 @@ struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
sock_rps_save_rxhash(child, skb);
if (rsk_drop_req(req)) {
- refcount_set(&req->rsk_refcnt, 2);
+ reqsk_put(req);
return child;
}
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 31f3b858db81..2135ee7c806d 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -970,7 +970,8 @@ ssize_t do_tcp_sendpages(struct sock *sk, struct page *page, int offset,
long timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
if (IS_ENABLED(CONFIG_DEBUG_VM) &&
- WARN_ONCE(PageSlab(page), "page must not be a Slab one"))
+ WARN_ONCE(!sendpage_ok(page),
+ "page must not be a Slab one and have page_count > 0"))
return -EINVAL;
/* Wait for a connection to finish. One exception is TCP Fast Open
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 184ea556f50e..b1ce2054291d 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -885,21 +885,34 @@ struct tcp_sacktag_state {
struct rate_sample *rate;
};
-/* Take a notice that peer is sending D-SACKs */
+/* Take a notice that peer is sending D-SACKs. Skip update of data delivery
+ * and spurious retransmission information if this DSACK is unlikely caused by
+ * sender's action:
+ * - DSACKed sequence range is larger than maximum receiver's window.
+ * - Total no. of DSACKed segments exceed the total no. of retransmitted segs.
+ */
static u32 tcp_dsack_seen(struct tcp_sock *tp, u32 start_seq,
u32 end_seq, struct tcp_sacktag_state *state)
{
u32 seq_len, dup_segs = 1;
- if (before(start_seq, end_seq)) {
- seq_len = end_seq - start_seq;
- if (seq_len > tp->mss_cache)
- dup_segs = DIV_ROUND_UP(seq_len, tp->mss_cache);
- }
+ if (!before(start_seq, end_seq))
+ return 0;
+
+ seq_len = end_seq - start_seq;
+ /* Dubious DSACK: DSACKed range greater than maximum advertised rwnd */
+ if (seq_len > tp->max_window)
+ return 0;
+ if (seq_len > tp->mss_cache)
+ dup_segs = DIV_ROUND_UP(seq_len, tp->mss_cache);
+
+ tp->dsack_dups += dup_segs;
+ /* Skip the DSACK if dup segs weren't retransmitted by sender */
+ if (tp->dsack_dups > tp->total_retrans)
+ return 0;
tp->rx_opt.sack_ok |= TCP_DSACK_SEEN;
tp->rack.dsack_seen = 1;
- tp->dsack_dups += dup_segs;
state->flag |= FLAG_DSACKING_ACK;
/* A spurious retransmission is delivered */
@@ -1153,6 +1166,11 @@ static bool tcp_check_dsack(struct sock *sk, const struct sk_buff *ack_skb,
}
dup_segs = tcp_dsack_seen(tp, start_seq_0, end_seq_0, state);
+ if (!dup_segs) { /* Skip dubious DSACK */
+ NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPDSACKIGNOREDDUBIOUS);
+ return false;
+ }
+
NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPDSACKRECVSEGS, dup_segs);
/* D-SACK for already forgotten data... Do dumb counting. */
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 5084333b5ab6..592c73962723 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1788,12 +1788,12 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb)
__skb_pull(skb, hdrlen);
if (skb_try_coalesce(tail, skb, &fragstolen, &delta)) {
- thtail->window = th->window;
-
TCP_SKB_CB(tail)->end_seq = TCP_SKB_CB(skb)->end_seq;
- if (after(TCP_SKB_CB(skb)->ack_seq, TCP_SKB_CB(tail)->ack_seq))
+ if (likely(!before(TCP_SKB_CB(skb)->ack_seq, TCP_SKB_CB(tail)->ack_seq))) {
TCP_SKB_CB(tail)->ack_seq = TCP_SKB_CB(skb)->ack_seq;
+ thtail->window = th->window;
+ }
/* We have to update both TCP_SKB_CB(tail)->tcp_flags and
* thtail->fin, so that the fast path in tcp_rcv_established()
generated by cgit (git 2.34.1) at 2024-05-25 11:31:31 +0000