net: fix IPSTATS_MIB_OUTFORWDATAGRAMS increment after fragment check

Reproduce environment:
network with 3 VM linuxs is connected as below:
VM1<---->VM2(latest kernel 6.5.0-rc7)<---->VM3
VM1: eth0 ip: 192.168.122.207 MTU 1800
VM2: eth0 ip: 192.168.122.208, eth1 ip: 192.168.123.224 MTU 1500
VM3: eth0 ip: 192.168.123.240 MTU 1800

Reproduce:
VM1 send 1600 bytes UDP data to VM3 using tools scapy with flags='DF'.
scapy command:
send(IP(dst="192.168.123.240",flags='DF')/UDP()/str('0'*1600),count=1,
inter=1.000000)

Result:
Before IP data is sent.
----------------------------------------------------------------------
root@qemux86-64:~# cat /proc/net/snmp
Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors
    ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests
    OutDiscards OutNoRoutes ReasmTimeout ReasmReqdss
Ip: 1 64 6 0 2 2 0 0 2 4 0 0 0 0 0 0 0 0 0
......
root@qemux86-64:~#
----------------------------------------------------------------------
After IP data is sent.
----------------------------------------------------------------------
root@qemux86-64:~# cat /proc/net/snmp
Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors
    ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests
    OutDiscards OutNoRoutes ReasmTimeout ReasmReqdss
Ip: 1 64 7 0 2 2 0 0 2 5 0 0 0 0 0 0 0 1 0
......
root@qemux86-64:~#
----------------------------------------------------------------------
ForwDatagrams is always keeping 2 without increment.

Issue description and patch:
ip_exceeds_mtu() in ip_forward() drops this IP datagram because skb len
(1600 sending by scapy) is over MTU(1500 in VM2) if "DF" is set.
According to RFC 4293 "3.2.3. IP Statistics Tables",
  +-------+------>------+----->-----+----->-----+
  | InForwDatagrams (6) | OutForwDatagrams (6)  |
  |                     V                       +->-+ OutFragReqds
  |                 InNoRoutes                  |   | (packets)
  / (local packet (3)                           |   |
  |  IF is that of the address                  |   +--> OutFragFails
  |  and may not be the receiving IF)           |   |    (packets)
the IPSTATS_MIB_OUTFORWDATAGRAMS should be counted before fragment
check.
The existing implementation, instead, would incease the counter after
fragment check: ip_exceeds_mtu() in ipv4 and ip6_pkt_too_big() in ipv6.
So do patch to move IPSTATS_MIB_OUTFORWDATAGRAMS counter to ip_forward()
for ipv4 and ip6_forward() for ipv6.

Test result with patch:
Before IP data is sent.
----------------------------------------------------------------------
root@qemux86-64:~# cat /proc/net/snmp
Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors
    ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests
    OutDiscards OutNoRoutes ReasmTimeout ReasmReqdss
Ip: 1 64 6 0 2 2 0 0 2 4 0 0 0 0 0 0 0 0 0
......
root@qemux86-64:~#
----------------------------------------------------------------------
After IP data is sent.
----------------------------------------------------------------------
root@qemux86-64:~# cat /proc/net/snmp
Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors
    ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests
    OutDiscards OutNoRoutes ReasmTimeout ReasmReqdss
Ip: 1 64 7 0 2 3 0 0 2 5 0 0 0 0 0 0 0 1 0
......
root@qemux86-64:~#
----------------------------------------------------------------------
ForwDatagrams is updated from 2 to 3.

Reviewed-by: Filip Pudak <filip.pudak@windriver.com>
Signed-off-by: Heng Guo <heng.guo@windriver.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20231011015137.27262-1-heng.guo@windriver.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

2 files changed, 4 insertions, 6 deletions

diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 66fac1216d46..8b65f12583eb 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -66,8 +66,6 @@ static int ip_forward_finish(struct net *net, struct sock *sk, struct sk_buff *s
 {
 	struct ip_options *opt	= &(IPCB(skb)->opt);
 
-	__IP_INC_STATS(net, IPSTATS_MIB_OUTFORWDATAGRAMS);
-
 #ifdef CONFIG_NET_SWITCHDEV
 	if (skb->offload_l3_fwd_mark) {
 		consume_skb(skb);
@@ -130,6 +128,8 @@ int ip_forward(struct sk_buff *skb)
 	if (opt->is_strictroute && rt->rt_uses_gateway)
 		goto sr_failed;
 
+	__IP_INC_STATS(net, IPSTATS_MIB_OUTFORWDATAGRAMS);
+
 	IPCB(skb)->flags |= IPSKB_FORWARDED;
 	mtu = ip_dst_mtu_maybe_forward(&rt->dst, true);
 	if (ip_exceeds_mtu(skb, mtu)) {
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index cdaa9275e990..a471c7e91761 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -446,10 +446,6 @@ static int ip6_forward_proxy_check(struct sk_buff *skb)
 static inline int ip6_forward_finish(struct net *net, struct sock *sk,
 				     struct sk_buff *skb)
 {
-	struct dst_entry *dst = skb_dst(skb);
-
-	__IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTFORWDATAGRAMS);
-
 #ifdef CONFIG_NET_SWITCHDEV
 	if (skb->offload_l3_fwd_mark) {
 		consume_skb(skb);
@@ -617,6 +613,8 @@ int ip6_forward(struct sk_buff *skb)
 		}
 	}
 
+	__IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTFORWDATAGRAMS);
+
 	mtu = ip6_dst_mtu_maybe_forward(dst, true);
 	if (mtu < IPV6_MIN_MTU)
 		mtu = IPV6_MIN_MTU;