[PATCH net] cls_flower: Fix inability to match GRE/IPIP packets

* [PATCH net] cls_flower: Fix inability to match GRE/IPIP packets
@ 2021-10-29  9:21 Yoshiki Komachi
  2021-10-29 13:10 ` patchwork-bot+netdevbpf
  0 siblings, 1 reply; 2+ messages in thread
From: Yoshiki Komachi @ 2021-10-29  9:21 UTC (permalink / raw)
  To: David S. Miller, Jakub Kicinski, Jamal Hadi Salim, Cong Wang,
	Jiri Pirko, Pravin B Shelar
  Cc: Yoshiki Komachi, netdev, dev, toshiaki.makita1

When a packet of a new flow arrives in openvswitch kernel module, it dissects
the packet and passes the extracted flow key to ovs-vswtichd daemon. If hw-
offload configuration is enabled, the daemon creates a new TC flower entry to
bypass openvswitch kernel module for the flow (TC flower can also offload flows
to NICs but this time that does not matter).

In this processing flow, I found the following issue in cases of GRE/IPIP
packets.

When ovs_flow_key_extract() in openvswitch module parses a packet of a new
GRE (or IPIP) flow received on non-tunneling vports, it extracts information
of the outer IP header for ip_proto/src_ip/dst_ip match keys.

This means ovs-vswitchd creates a TC flower entry with IP protocol/addresses
match keys whose values are those of the outer IP header. OTOH, TC flower,
which uses flow_dissector (different parser from openvswitch module), extracts
information of the inner IP header.

The following flow is an example to describe the issue in more detail.

   <----------- Outer IP -----------------> <---------- Inner IP ---------->
  +----------+--------------+--------------+----------+----------+----------+
  | ip_proto | src_ip       | dst_ip       | ip_proto | src_ip   | dst_ip   |
  | 47 (GRE) | 192.168.10.1 | 192.168.10.2 | 6 (TCP)  | 10.0.0.1 | 10.0.0.2 |
  +----------+--------------+--------------+----------+----------+----------+

In this case, TC flower entry and extracted information are shown as below:

  - ovs-vswitchd creates TC flower entry with:
      - ip_proto: 47
      - src_ip: 192.168.10.1
      - dst_ip: 192.168.10.2

  - TC flower extracts below for IP header matches:
      - ip_proto: 6
      - src_ip: 10.0.0.1
      - dst_ip: 10.0.0.2

Thus, GRE or IPIP packets never match the TC flower entry, as each
dissector behaves differently.

IMHO, the behavior of TC flower (flow dissector) does not look correct,
as ip_proto/src_ip/dst_ip in TC flower match means the outermost IP
header information except for GRE/IPIP cases. This patch adds a new
flow_dissector flag FLOW_DISSECTOR_F_STOP_BEFORE_ENCAP which skips
dissection of the encapsulated inner GRE/IPIP header in TC flower
classifier.

Signed-off-by: Yoshiki Komachi <komachi.yoshiki@gmail.com>
---
 include/net/flow_dissector.h |  1 +
 net/core/flow_dissector.c    | 15 +++++++++++++++
 net/sched/cls_flower.c       |  3 ++-
 3 files changed, 18 insertions(+), 1 deletion(-)

diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h
index ffd386ea0dbb..aa33e1092e2c 100644
--- a/include/net/flow_dissector.h
+++ b/include/net/flow_dissector.h
@@ -287,6 +287,7 @@ enum flow_dissector_key_id {
 #define FLOW_DISSECTOR_F_PARSE_1ST_FRAG		BIT(0)
 #define FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL	BIT(1)
 #define FLOW_DISSECTOR_F_STOP_AT_ENCAP		BIT(2)
+#define FLOW_DISSECTOR_F_STOP_BEFORE_ENCAP	BIT(3)
 
 struct flow_dissector_key {
 	enum flow_dissector_key_id key_id;
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index bac0184cf3de..0d4bbf534c7d 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -1307,6 +1307,11 @@ bool __skb_flow_dissect(const struct net *net,
 
 	switch (ip_proto) {
 	case IPPROTO_GRE:
+		if (flags & FLOW_DISSECTOR_F_STOP_BEFORE_ENCAP) {
+			fdret = FLOW_DISSECT_RET_OUT_GOOD;
+			break;
+		}
+
 		fdret = __skb_flow_dissect_gre(skb, key_control, flow_dissector,
 					       target_container, data,
 					       &proto, &nhoff, &hlen, flags);
@@ -1364,6 +1369,11 @@ bool __skb_flow_dissect(const struct net *net,
 		break;
 	}
 	case IPPROTO_IPIP:
+		if (flags & FLOW_DISSECTOR_F_STOP_BEFORE_ENCAP) {
+			fdret = FLOW_DISSECT_RET_OUT_GOOD;
+			break;
+		}
+
 		proto = htons(ETH_P_IP);
 
 		key_control->flags |= FLOW_DIS_ENCAPSULATION;
@@ -1376,6 +1386,11 @@ bool __skb_flow_dissect(const struct net *net,
 		break;
 
 	case IPPROTO_IPV6:
+		if (flags & FLOW_DISSECTOR_F_STOP_BEFORE_ENCAP) {
+			fdret = FLOW_DISSECT_RET_OUT_GOOD;
+			break;
+		}
+
 		proto = htons(ETH_P_IPV6);
 
 		key_control->flags |= FLOW_DIS_ENCAPSULATION;
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index eb6345a027e1..aab13ba11767 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -329,7 +329,8 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 				    ARRAY_SIZE(fl_ct_info_to_flower_map),
 				    post_ct);
 		skb_flow_dissect_hash(skb, &mask->dissector, &skb_key);
-		skb_flow_dissect(skb, &mask->dissector, &skb_key, 0);
+		skb_flow_dissect(skb, &mask->dissector, &skb_key,
+				 FLOW_DISSECTOR_F_STOP_BEFORE_ENCAP);
 
 		f = fl_mask_lookup(mask, &skb_key);
 		if (f && !tc_skip_sw(f->flags)) {
-- 
2.30.1 (Apple Git-130)


^ permalink raw reply related	[flat|nested] 2+ messages in thread