Skip to contentSkip to conversation

ice: PPPoE offload support

RFC PATCH Superseded Marcin Szycik 2 participants 6 messages Jul 15, 2022 – Jul 17, 2022

Subsystems (4)

v5 of 6 Superseded Latest is v6 →

lore
Share & tools
lore
#0 lore raw

v5 cover ice: PPPoE offload support

Add support for dissecting PPPoE and PPP-specific fields in flow dissector:
PPPoE session id and PPP protocol type. Add support for those fields in tc-flower and support offloading PPPoE. Finally, add support for hardware offload of PPPoE packets in switchdev mode in ice driver.

Example filter:
tc filter add dev $PF1 ingress protocol ppp_ses prio 1 flower pppoe_sid \
1234 ppp_proto ip skip_sw action mirred egress redirect dev $VF1_PR

Changes in iproute2 are required to use the new fields (will be submitted soon).

ICE COMMS DDP package is required to create a filter in ice.

v5: fix endianness when processing compressed protocols
v4:
* PPPoE header validation
* added MPLS support
* added support for compressed PPP protocol field
* flow_dissector_key_pppoe::session_id stored in __be16
* new field: flow_dissector_key_pppoe::type
* always add an ethtype lookup if PPP/PPPoE options are provided (to prevent setting incorrect ethtype)
* rebase
v3:
* revert byte order changes in is_ppp_proto_supported from previous version
* add kernel-doc for is_ppp_proto_supported
* add more CC
v2: cosmetic changes

Marcin Szycik (1):
ice: Add support for PPPoE hardware offload

Wojciech Drewek (3):
flow_dissector: Add PPPoE dissectors
net/sched: flower: Add PPPoE filter
flow_offload: Introduce flow_match_pppoe

 drivers/net/ethernet/intel/ice/ice.h          |   1 +
 .../net/ethernet/intel/ice/ice_flex_pipe.c    |   5 +-
 .../ethernet/intel/ice/ice_protocol_type.h    |  11 ++
 drivers/net/ethernet/intel/ice/ice_switch.c   | 165 ++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_tc_lib.c   |  71 +++++++-
 drivers/net/ethernet/intel/ice/ice_tc_lib.h   |   8 +
 include/net/flow_dissector.h                  |  13 ++
 include/net/flow_offload.h                    |   6 +
 include/uapi/linux/pkt_cls.h                  |   3 +
 net/core/flow_dissector.c                     |  85 ++++++++-
 net/core/flow_offload.c                       |   7 +
 net/sched/cls_flower.c                        |  61 +++++++
 12 files changed, 427 insertions(+), 9 deletions(-)
Signature · 1 line
2.35.1
#1 lore raw

RFC PATCH v5 1/4 flow_dissector: Add PPPoE dissectors

From: Wojciech Drewek <[email protected]>

Allow to dissect PPPoE specific fields which are:
- session ID (16 bits)
- ppp protocol (16 bits)
- type (16 bits) - this is PPPoE ethertype, for now only ETH_P_PPP_SES is supported, possible ETH_P_PPP_DISC in the future

The goal is to make the following TC command possible:

  # tc filter add dev ens6f0 ingress prio 1 protocol ppp_ses \
      flower \
        pppoe_sid 12 \
        ppp_proto ip \
      action drop

Note that only PPPoE Session is supported.

---
v5: fix endianness when processing compressed protocols v4:
* pppoe header validation
* added MPLS dissection
* added support for compressed ppp protocol field
* flow_dissector_key_pppoe::session_id stored in __be16
* new field: flow_dissector_key_pppoe::type
v3: revert byte order changes in is_ppp_proto_supported from
previous version
v2: ntohs instead of htons in is_ppp_proto_supported

 include/net/flow_dissector.h | 13 ++++++
 net/core/flow_dissector.c    | 85 +++++++++++++++++++++++++++++++++---
 2 files changed, 91 insertions(+), 7 deletions(-)
include/net/flow_dissector.h +13 -0
index 0f9544a9bb9e..6c74812d64b2 100644
@@ -277,6 +277,18 @@ struct flow_dissector_key_num_of_vlans {
 	u8 num_of_vlans;
 };
 
+/**
+ * struct flow_dissector_key_pppoe:
+ * @session_id: pppoe session id
+ * @ppp_proto: ppp protocol
+ * @type: pppoe eth type
+ */
+struct flow_dissector_key_pppoe {
+	__be16 session_id;
+	__be16 ppp_proto;
+	__be16 type;
+};
+
 enum flow_dissector_key_id {
 	FLOW_DISSECTOR_KEY_CONTROL, /* struct flow_dissector_key_control */
 	FLOW_DISSECTOR_KEY_BASIC, /* struct flow_dissector_key_basic */
@@ -307,6 +319,7 @@ enum flow_dissector_key_id {
 	FLOW_DISSECTOR_KEY_CT, /* struct flow_dissector_key_ct */
 	FLOW_DISSECTOR_KEY_HASH, /* struct flow_dissector_key_hash */
 	FLOW_DISSECTOR_KEY_NUM_OF_VLANS, /* struct flow_dissector_key_num_of_vlans */
+	FLOW_DISSECTOR_KEY_PPPOE, /* struct flow_dissector_key_pppoe */
 
 	FLOW_DISSECTOR_KEY_MAX,
 };
net/core/flow_dissector.c +78 -7

110 lines not on this page · raw message

Signature · 1 line
2.35.1
#2 lore raw

Re: [RFC PATCH net-next v5 1/4] flow_dissector: Add PPPoE dissectors

Quoted 20 lines · Marcin Szycik
net/core/flow_dissector.c +20 -0
+/**
+ * is_ppp_proto_supported - checks if inner PPP protocol should be dissected
+ * @proto: protocol type (PPP proto field)
+ */
+static bool is_ppp_proto_supported(__be16 proto)
+{
+	switch (proto) {
+	case htons(PPP_AT):
+	case htons(PPP_IPX):
+	case htons(PPP_VJC_COMP):
+	case htons(PPP_VJC_UNCOMP):
+	case htons(PPP_MP):
+	case htons(PPP_COMPFRAG):
+	case htons(PPP_COMP):
+	case htons(PPP_IPCP):
+	case htons(PPP_ATCP):
+	case htons(PPP_IPXCP):
+	case htons(PPP_IPV6CP):
+	case htons(PPP_CCPFRAG):
+	case htons(PPP_MPLSCP):

net/core/flow_dissector.c bool bpf_flow_dissect(struct bpf_prog *prog, struct bpf_flow_dissector *ctx, · line 927

Quoted 10 lines
net/core/flow_dissector.c +10 -0
+	case htons(PPP_LCP):
+	case htons(PPP_PAP):
+	case htons(PPP_LQR):
+	case htons(PPP_CHAP):
+	case htons(PPP_CBCP):
+		return true;
+	default:
+		return false;
+	}
+}

The more I think about it, the more I believe we should make this function more generic by following the Protocol field definition found in section 2 of RFC 1661:
https://datatracker.ietf.org/doc/html/rfc1661#section-2

I mean, it's very surprising that is_ppp_proto_supported() returns false for protocols like PPP_IP or PPP_IPV6, which certainly are supported. Of course, in the context of your patch, PPP_IP and PPP_IPV6 have been tested first, but that makes the code a bit unclear.

We could define a simpler and more generic helper function (probably in a ppp header file). Something like (unstested):

/* Assumes proto isn't compressed. */
static bool ppp_proto_is_valid(u16 proto) /*

{
	/* Protocol must be odd and the least significant bit of the
	 * most significant octet must be 0 (see RFC 1661, section 2).
	 */
	return !!(proto & 0X0101 == 0x0001)
}

BTW, we don't have to restrict the list of supported protocols to the PPP_* numbers defined in the kernel as we have no indication that the PPP frame is going to be handled locally.

Quoted 19 lines
net/core/flow_dissector.c +7 -0
+static bool is_pppoe_ses_hdr_valid(struct pppoe_hdr hdr)
+{
+	return hdr.ver == 1 && hdr.type == 1 && hdr.code == 0;
+}
+
 /**
  * __skb_flow_dissect - extract the flow_keys struct and return it
  * @net: associated network namespace, derived from @skb if NULL
@@ -1214,26 +1250,61 @@ bool __skb_flow_dissect(const struct net *net,
 			struct pppoe_hdr hdr;
 			__be16 proto;
 		} *hdr, _hdr;
+		u16 ppp_proto;
+
 		hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), data, hlen, &_hdr);
 		if (!hdr) {
 			fdret = FLOW_DISSECT_RET_OUT_BAD;
 			break;
 		}

net/core/flow_dissector.c bool __skb_flow_dissect(const struct net *net, · line 1266

Quoted 10 lines
net/core/flow_dissector.c +6 -3
 
-		nhoff += PPPOE_SES_HLEN;
-		switch (hdr->proto) {
-		case htons(PPP_IP):
+		if (!is_pppoe_ses_hdr_valid(hdr->hdr)) {
+			fdret = FLOW_DISSECT_RET_OUT_BAD;
+			break;
+		}
+
+		/* least significant bit of the least significant octet

That's the least significant bit of the _most significant_ octet (the one of the least significant octet must always be 1).

net/core/flow_dissector.c bool __skb_flow_dissect(const struct net *net, · line 1270

Quoted 4 lines
net/core/flow_dissector.c +4 -0
+		 * indicates if protocol field was compressed
+		 */
+		ppp_proto = ntohs(hdr->proto);
+		if (ppp_proto & 256) {

Using hex would improve readability in my opinion (that is, s/256/0x0100/).

net/core/flow_dissector.c bool __skb_flow_dissect(const struct net *net, · line 1271

Quoted 1 line
net/core/flow_dissector.c +1 -0
+			ppp_proto = htons(ppp_proto >> 8);

I don't get why you convert ppp_proto back to network byte order. That contradicts the type annotation (u16).

net/core/flow_dissector.c bool __skb_flow_dissect(const struct net *net, · line 1274

Quoted 3 lines
net/core/flow_dissector.c +3 -0
+			nhoff += PPPOE_SES_HLEN - 1;
+		} else {
+			ppp_proto = htons(ppp_proto);

Same here. We could leave ppp_proto untouched in this branch.

net/core/flow_dissector.c bool __skb_flow_dissect(const struct net *net, · line 1278

Quoted 4 lines
net/core/flow_dissector.c +4 -0
+			nhoff += PPPOE_SES_HLEN;
+		}
+
+		if (ppp_proto == htons(PPP_IP)) {

With ppp_proto kept in host byte order, the htons() call would need to go.

net/core/flow_dissector.c bool __skb_flow_dissect(const struct net *net, · line 1281

Quoted 5 lines
net/core/flow_dissector.c +1 -2
 			proto = htons(ETH_P_IP);
 			fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
-			break;
-		case htons(PPP_IPV6):
+		} else if (ppp_proto == htons(PPP_IPV6)) {

Same here, and in the following 'if' branches.

Quoted 16 lines
net/core/flow_dissector.c +9 -2
 			proto = htons(ETH_P_IPV6);
 			fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
-			break;
-		default:
+		} else if (ppp_proto == htons(PPP_MPLS_UC)) {
+			proto = htons(ETH_P_MPLS_UC);
+			fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
+		} else if (ppp_proto == htons(PPP_MPLS_MC)) {
+			proto = htons(ETH_P_MPLS_MC);
+			fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
+		} else if (is_ppp_proto_supported(ppp_proto)) {
+			fdret = FLOW_DISSECT_RET_OUT_GOOD;
+		} else {
 			fdret = FLOW_DISSECT_RET_OUT_BAD;
 			break;
 		}

net/core/flow_dissector.c bool __skb_flow_dissect(const struct net *net, · line 1305

Quoted 10 lines
net/core/flow_dissector.c +10 -0
+
+		if (dissector_uses_key(flow_dissector,
+				       FLOW_DISSECTOR_KEY_PPPOE)) {
+			struct flow_dissector_key_pppoe *key_pppoe;
+
+			key_pppoe = skb_flow_dissector_target(flow_dissector,
+							      FLOW_DISSECTOR_KEY_PPPOE,
+							      target_container);
+			key_pppoe->session_id = hdr->hdr.sid;
+			key_pppoe->ppp_proto = ppp_proto;

With ppp_proto being u16, we'd now need to call htons(ppp_proto).

Quoted 8 lines
net/core/flow_dissector.c +2 -0
+			key_pppoe->type = htons(ETH_P_PPP_SES);
+		}
 		break;
 	}
 	case htons(ETH_P_TIPC): {
Signature · 1 line
2.35.1
#3 lore raw

RFC PATCH v5 2/4 net/sched: flower: Add PPPoE filter

In reply to Marcin Szycik

From: Wojciech Drewek <[email protected]>

Add support for PPPoE specific fields for tc-flower. Those fields can be provided only when protocol was set to ETH_P_PPP_SES. Defines, dump, load and set are being done here.

Overwrite basic.n_proto only in case of PPP_IP and PPP_IPV6, otherwise leave it as ETH_P_PPP_SES.

---
v4:
* support of MPLS inner fields
* session_id stored in __be16

 include/uapi/linux/pkt_cls.h |  3 ++
 net/sched/cls_flower.c       | 61 ++++++++++++++++++++++++++++++++++++
 2 files changed, 64 insertions(+)
include/uapi/linux/pkt_cls.h +3 -0
index 9a2ee1e39fad..c142c0f8ed8a 100644
@@ -589,6 +589,9 @@ enum {
 
 	TCA_FLOWER_KEY_NUM_OF_VLANS,    /* u8 */
 
+	TCA_FLOWER_KEY_PPPOE_SID,	/* be16 */
+	TCA_FLOWER_KEY_PPP_PROTO,	/* be16 */
+
 	__TCA_FLOWER_MAX,
 };
 
net/sched/cls_flower.c +61 -0

103 lines not on this page · raw message

Signature · 1 line
2.35.1
#4 lore raw

RFC PATCH v5 3/4 flow_offload: Introduce flow_match_pppoe

In reply to Marcin Szycik

From: Wojciech Drewek <[email protected]>

Allow to offload PPPoE filters by adding flow_rule_match_pppoe. Drivers can extract PPPoE specific fields from now on.

---
 include/net/flow_offload.h | 6 ++++++
 net/core/flow_offload.c    | 7 +++++++
 2 files changed, 13 insertions(+)
include/net/flow_offload.h +6 -0
index a8d8512b7059..2a9a9e42e7fd 100644
@@ -76,6 +76,10 @@ struct flow_match_ct {
 	struct flow_dissector_key_ct *key, *mask;
 };
 
+struct flow_match_pppoe {
+	struct flow_dissector_key_pppoe *key, *mask;
+};
+
 struct flow_rule;
 
 void flow_rule_match_meta(const struct flow_rule *rule,
@@ -122,6 +126,8 @@ void flow_rule_match_enc_opts(const struct flow_rule *rule,
 			      struct flow_match_enc_opts *out);
 void flow_rule_match_ct(const struct flow_rule *rule,
 			struct flow_match_ct *out);
+void flow_rule_match_pppoe(const struct flow_rule *rule,
+			   struct flow_match_pppoe *out);
 
 enum flow_action_id {
 	FLOW_ACTION_ACCEPT		= 0,
net/core/flow_offload.c +7 -0
index 0d3075d3c8fb..8cfb63528d18 100644
@@ -230,6 +230,13 @@ void flow_rule_match_ct(const struct flow_rule *rule,
 }
 EXPORT_SYMBOL(flow_rule_match_ct);
 
+void flow_rule_match_pppoe(const struct flow_rule *rule,
+			   struct flow_match_pppoe *out)
+{
+	FLOW_DISSECTOR_MATCH(rule, FLOW_DISSECTOR_KEY_PPPOE, out);
+}
+EXPORT_SYMBOL(flow_rule_match_pppoe);
+
 struct flow_block_cb *flow_block_cb_alloc(flow_setup_cb_t *cb,
 					  void *cb_ident, void *cb_priv,
 					  void (*release)(void *cb_priv))
Signature · 1 line
2.35.1
#5 lore raw

RFC PATCH v5 4/4 ice: Add support for PPPoE hardware offload

In reply to Marcin Szycik

Add support for creating PPPoE filters in switchdev mode. Add support for parsing PPPoE and PPP-specific tc options: pppoe_sid and ppp_proto.

Example filter:
tc filter add dev $PF1 ingress protocol ppp_ses prio 1 flower pppoe_sid \
1234 ppp_proto ip skip_sw action mirred egress redirect dev $VF1_PR

Changes in iproute2 are required to use the new fields.

ICE COMMS DDP package is required to create a filter as it contains PPPoE profiles. Added a warning message when loaded DDP package does not contain required profiles.

---
v4:
* adapted code to the redefined ice_tc_pppoe_hdr (session_id is now

    stored in __be16)

* removed note about missing support for vlan fields:
263957263a00 ("ice: switch: dynamically add VLAN headers to dummy packets") is already accepted
* always add an ethtype lookup if PPP/PPPoE options are provided (to prevent setting incorrect ethtype)
* rebase
v2: wrap one long line

 drivers/net/ethernet/intel/ice/ice.h          |   1 +
 .../net/ethernet/intel/ice/ice_flex_pipe.c    |   5 +-
 .../ethernet/intel/ice/ice_protocol_type.h    |  11 ++
 drivers/net/ethernet/intel/ice/ice_switch.c   | 165 ++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_tc_lib.c   |  71 +++++++-
 drivers/net/ethernet/intel/ice/ice_tc_lib.h   |   8 +
 6 files changed, 259 insertions(+), 2 deletions(-)
drivers/net/ethernet/intel/ice/ice.h +1 -0
index 60453b3b8d23..0e48a930f004 100644
@@ -52,6 +52,7 @@
 #include <net/udp_tunnel.h>
 #include <net/vxlan.h>
 #include <net/gtp.h>
+#include <linux/ppp_defs.h>
 #include "ice_devids.h"
 #include "ice_type.h"
 #include "ice_txrx.h"
drivers/net/ethernet/intel/ice/ice_flex_pipe.c +4 -1
index ada5198b5b16..4b3bb19e1d06 100644
@@ -1964,8 +1964,11 @@ ice_get_sw_fv_list(struct ice_hw *hw, struct ice_prot_lkup_ext *lkups,
 			}
 		}
 	} while (fv);
-	if (list_empty(fv_list))
+	if (list_empty(fv_list)) {
+		dev_warn(ice_hw_to_dev(hw), "Required profiles not found in currently loaded DDP package");
 		return -EIO;
+	}
+
 	return 0;
 
 err:
drivers/net/ethernet/intel/ice/ice_protocol_type.h +11 -0
index d4a0d089649c..560efc7654c7 100644
@@ -43,6 +43,7 @@ enum ice_protocol_type {
 	ICE_NVGRE,
 	ICE_GTP,
 	ICE_GTP_NO_PAY,
+	ICE_PPPOE,
 	ICE_VLAN_EX,
 	ICE_VLAN_IN,
 	ICE_VXLAN_GPE,
@@ -109,6 +110,7 @@ enum ice_prot_id {
 #define ICE_TCP_IL_HW		49
 #define ICE_UDP_ILOS_HW		53
 #define ICE_GRE_OF_HW		64
+#define ICE_PPPOE_HW		103
 
 #define ICE_UDP_OF_HW	52 /* UDP Tunnels */
 #define ICE_META_DATA_ID_HW 255 /* this is used for tunnel and VLAN type */
@@ -207,6 +209,14 @@ struct ice_udp_gtp_hdr {
 	u8 rsvrd;
 };
 
+struct ice_pppoe_hdr {
+	u8 rsrvd_ver_type;
+	u8 rsrvd_code;
+	__be16 session_id;
+	__be16 length;
+	__be16 ppp_prot_id; /* control and data only */
+};
+
 struct ice_nvgre_hdr {
 	__be16 flags;
 	__be16 protocol;
@@ -224,6 +234,7 @@ union ice_prot_hdr {
 	struct ice_udp_tnl_hdr tnl_hdr;
 	struct ice_nvgre_hdr nvgre_hdr;
 	struct ice_udp_gtp_hdr gtp_hdr;
+	struct ice_pppoe_hdr pppoe_hdr;
 };
 
 /* This is mapping table entry that maps every word within a given protocol
drivers/net/ethernet/intel/ice/ice_switch.c +165 -0

207 lines not on this page · raw message

drivers/net/ethernet/intel/ice/ice_tc_lib.c +70 -1

101 lines not on this page · raw message

drivers/net/ethernet/intel/ice/ice_tc_lib.h +8 -0
index 0193874cd203..91cd3d3778c7 100644
@@ -24,6 +24,8 @@
 #define ICE_TC_FLWR_FIELD_ETH_TYPE_ID		BIT(17)
 #define ICE_TC_FLWR_FIELD_ENC_OPTS		BIT(18)
 #define ICE_TC_FLWR_FIELD_CVLAN			BIT(19)
+#define ICE_TC_FLWR_FIELD_PPPOE_SESSID		BIT(20)
+#define ICE_TC_FLWR_FIELD_PPP_PROTO		BIT(21)
 
 #define ICE_TC_FLOWER_MASK_32   0xFFFFFFFF
 
@@ -44,6 +46,11 @@ struct ice_tc_vlan_hdr {
 	__be16 vlan_tpid;
 };
 
+struct ice_tc_pppoe_hdr {
+	__be16 session_id;
+	__be16 ppp_proto;
+};
+
 struct ice_tc_l2_hdr {
 	u8 dst_mac[ETH_ALEN];
 	u8 src_mac[ETH_ALEN];
@@ -84,6 +91,7 @@ struct ice_tc_flower_lyr_2_4_hdrs {
 	struct ice_tc_l2_hdr l2_mask;
 	struct ice_tc_vlan_hdr vlan_hdr;
 	struct ice_tc_vlan_hdr cvlan_hdr;
+	struct ice_tc_pppoe_hdr pppoe_hdr;
 	/* L3 (IPv4[6]) layer fields with their mask */
 	struct ice_tc_l3_hdr l3_key;
 	struct ice_tc_l3_hdr l3_mask;
Signature · 1 line
2.35.1