Merge branch 'sfc-tc-encap-actions-offload'

Edward Cree says:

====================
sfc: TC encap actions offload

This series adds support for offloading TC tunnel_key set actions to the
 EF100 driver, supporting VxLAN and GENEVE tunnels over IPv4 or IPv6.
====================

Link: https://lore.kernel.org/r/cover.1686240142.git.ecree.xilinx@gmail.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

drivers/net/ethernet/sfc/Makefile

@@ -10,7 +10,8 @@ sfc-y			+= efx.o efx_common.o efx_channels.o nic.o \
 			   efx_devlink.o
 sfc-$(CONFIG_SFC_MTD)	+= mtd.o
 sfc-$(CONFIG_SFC_SRIOV)	+= sriov.o ef10_sriov.o ef100_sriov.o ef100_rep.o \
-                           mae.o tc.o tc_bindings.o tc_counters.o
+                           mae.o tc.o tc_bindings.o tc_counters.o \
+                           tc_encap_actions.o
 
 obj-$(CONFIG_SFC)	+= sfc.o
 

drivers/net/ethernet/sfc/ef100_netdev.c

@@ -24,6 +24,7 @@
 #include "rx_common.h"
 #include "ef100_sriov.h"
 #include "tc_bindings.h"
+#include "tc_encap_actions.h"
 #include "efx_devlink.h"
 
 static void ef100_update_name(struct efx_nic *efx)
@@ -300,14 +301,38 @@ int ef100_netdev_event(struct notifier_block *this,
 {
 	struct efx_nic *efx = container_of(this, struct efx_nic, netdev_notifier);
 	struct net_device *net_dev = netdev_notifier_info_to_dev(ptr);
+	struct ef100_nic_data *nic_data = efx->nic_data;
+	int err;
 
 	if (efx->net_dev == net_dev &&
 	    (event == NETDEV_CHANGENAME || event == NETDEV_REGISTER))
 		ef100_update_name(efx);
 
+	if (!nic_data->grp_mae)
+		return NOTIFY_DONE;
+	err = efx_tc_netdev_event(efx, event, net_dev);
+	if (err & NOTIFY_STOP_MASK)
+		return err;
+
 	return NOTIFY_DONE;
 }
 
+static int ef100_netevent_event(struct notifier_block *this,
+				unsigned long event, void *ptr)
+{
+	struct efx_nic *efx = container_of(this, struct efx_nic, netevent_notifier);
+	struct ef100_nic_data *nic_data = efx->nic_data;
+	int err;
+
+	if (!nic_data->grp_mae)
+		return NOTIFY_DONE;
+	err = efx_tc_netevent_event(efx, event, ptr);
+	if (err & NOTIFY_STOP_MASK)
+		return err;
+
+	return NOTIFY_DONE;
+};
+
 static int ef100_register_netdev(struct efx_nic *efx)
 {
 	struct net_device *net_dev = efx->net_dev;
@@ -367,6 +392,7 @@ void ef100_remove_netdev(struct efx_probe_data *probe_data)
 	rtnl_unlock();
 
 	unregister_netdevice_notifier(&efx->netdev_notifier);
+	unregister_netevent_notifier(&efx->netevent_notifier);
 #if defined(CONFIG_SFC_SRIOV)
 	if (!efx->type->is_vf)
 		efx_ef100_pci_sriov_disable(efx, true);
@@ -487,6 +513,14 @@ int ef100_probe_netdev(struct efx_probe_data *probe_data)
 		goto fail;
 	}
 
+	efx->netevent_notifier.notifier_call = ef100_netevent_event;
+	rc = register_netevent_notifier(&efx->netevent_notifier);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "Failed to register netevent notifier, rc=%d\n", rc);
+		goto fail;
+	}
+
 	efx_probe_devlink_unlock(efx);
 	return rc;
 fail:

drivers/net/ethernet/sfc/mae.c

@@ -15,6 +15,7 @@
 #include "mcdi.h"
 #include "mcdi_pcol.h"
 #include "mcdi_pcol_mae.h"
+#include "tc_encap_actions.h"
 
 int efx_mae_allocate_mport(struct efx_nic *efx, u32 *id, u32 *label)
 {
@@ -610,6 +611,87 @@ static int efx_mae_encap_type_to_mae_type(enum efx_encap_type type)
 	}
 }
 
+int efx_mae_allocate_encap_md(struct efx_nic *efx,
+			      struct efx_tc_encap_action *encap)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_MAE_ENCAP_HEADER_ALLOC_IN_LEN(EFX_TC_MAX_ENCAP_HDR));
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_MAE_ENCAP_HEADER_ALLOC_OUT_LEN);
+	size_t inlen, outlen;
+	int rc;
+
+	rc = efx_mae_encap_type_to_mae_type(encap->type);
+	if (rc < 0)
+		return rc;
+	MCDI_SET_DWORD(inbuf, MAE_ENCAP_HEADER_ALLOC_IN_ENCAP_TYPE, rc);
+	inlen = MC_CMD_MAE_ENCAP_HEADER_ALLOC_IN_LEN(encap->encap_hdr_len);
+	if (WARN_ON(inlen > sizeof(inbuf))) /* can't happen */
+		return -EINVAL;
+	memcpy(MCDI_PTR(inbuf, MAE_ENCAP_HEADER_ALLOC_IN_HDR_DATA),
+	       encap->encap_hdr,
+	       encap->encap_hdr_len);
+	rc = efx_mcdi_rpc(efx, MC_CMD_MAE_ENCAP_HEADER_ALLOC, inbuf,
+			  inlen, outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		return rc;
+	if (outlen < sizeof(outbuf))
+		return -EIO;
+	encap->fw_id = MCDI_DWORD(outbuf, MAE_ENCAP_HEADER_ALLOC_OUT_ENCAP_HEADER_ID);
+	return 0;
+}
+
+int efx_mae_update_encap_md(struct efx_nic *efx,
+			    struct efx_tc_encap_action *encap)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_MAE_ENCAP_HEADER_UPDATE_IN_LEN(EFX_TC_MAX_ENCAP_HDR));
+	size_t inlen;
+	int rc;
+
+	rc = efx_mae_encap_type_to_mae_type(encap->type);
+	if (rc < 0)
+		return rc;
+	MCDI_SET_DWORD(inbuf, MAE_ENCAP_HEADER_UPDATE_IN_ENCAP_TYPE, rc);
+	MCDI_SET_DWORD(inbuf, MAE_ENCAP_HEADER_UPDATE_IN_EH_ID,
+		       encap->fw_id);
+	inlen = MC_CMD_MAE_ENCAP_HEADER_UPDATE_IN_LEN(encap->encap_hdr_len);
+	if (WARN_ON(inlen > sizeof(inbuf))) /* can't happen */
+		return -EINVAL;
+	memcpy(MCDI_PTR(inbuf, MAE_ENCAP_HEADER_UPDATE_IN_HDR_DATA),
+	       encap->encap_hdr,
+	       encap->encap_hdr_len);
+
+	BUILD_BUG_ON(MC_CMD_MAE_ENCAP_HEADER_UPDATE_OUT_LEN != 0);
+	return efx_mcdi_rpc(efx, MC_CMD_MAE_ENCAP_HEADER_UPDATE, inbuf,
+			    inlen, NULL, 0, NULL);
+}
+
+int efx_mae_free_encap_md(struct efx_nic *efx,
+			  struct efx_tc_encap_action *encap)
+{
+	MCDI_DECLARE_BUF(outbuf, MC_CMD_MAE_ENCAP_HEADER_FREE_OUT_LEN(1));
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_MAE_ENCAP_HEADER_FREE_IN_LEN(1));
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, MAE_ENCAP_HEADER_FREE_IN_EH_ID, encap->fw_id);
+	rc = efx_mcdi_rpc(efx, MC_CMD_MAE_ENCAP_HEADER_FREE, inbuf,
+			  sizeof(inbuf), outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		return rc;
+	if (outlen < sizeof(outbuf))
+		return -EIO;
+	/* FW freed a different ID than we asked for, should also never happen.
+	 * Warn because it means we've now got a different idea to the FW of
+	 * what encap_mds exist, which could cause mayhem later.
+	 */
+	if (WARN_ON(MCDI_DWORD(outbuf, MAE_ENCAP_HEADER_FREE_OUT_FREED_EH_ID) != encap->fw_id))
+		return -EIO;
+	/* We're probably about to free @encap, but let's just make sure its
+	 * fw_id is blatted so that it won't look valid if it leaks out.
+	 */
+	encap->fw_id = MC_CMD_MAE_ENCAP_HEADER_ALLOC_OUT_ENCAP_HEADER_ID_NULL;
+	return 0;
+}
+
 int efx_mae_lookup_mport(struct efx_nic *efx, u32 vf_idx, u32 *id)
 {
 	struct ef100_nic_data *nic_data = efx->nic_data;
@@ -833,8 +915,12 @@ int efx_mae_alloc_action_set(struct efx_nic *efx, struct efx_tc_action_set *act)
 		MCDI_SET_WORD_BE(inbuf, MAE_ACTION_SET_ALLOC_IN_VLAN1_PROTO_BE,
 				 act->vlan_proto[1]);
 	}
-	MCDI_SET_DWORD(inbuf, MAE_ACTION_SET_ALLOC_IN_ENCAP_HEADER_ID,
-		       MC_CMD_MAE_ENCAP_HEADER_ALLOC_OUT_ENCAP_HEADER_ID_NULL);
+	if (act->encap_md)
+		MCDI_SET_DWORD(inbuf, MAE_ACTION_SET_ALLOC_IN_ENCAP_HEADER_ID,
+			       act->encap_md->fw_id);
+	else
+		MCDI_SET_DWORD(inbuf, MAE_ACTION_SET_ALLOC_IN_ENCAP_HEADER_ID,
+			       MC_CMD_MAE_ENCAP_HEADER_ALLOC_OUT_ENCAP_HEADER_ID_NULL);
 	if (act->deliver)
 		MCDI_SET_DWORD(inbuf, MAE_ACTION_SET_ALLOC_IN_DELIVER,
 			       act->dest_mport);
@@ -1229,6 +1315,29 @@ int efx_mae_insert_rule(struct efx_nic *efx, const struct efx_tc_match *match,
 	return 0;
 }
 
+int efx_mae_update_rule(struct efx_nic *efx, u32 acts_id, u32 id)
+{
+	MCDI_DECLARE_BUF(inbuf, MC_CMD_MAE_ACTION_RULE_UPDATE_IN_LEN);
+	MCDI_DECLARE_STRUCT_PTR(response);
+
+	BUILD_BUG_ON(MC_CMD_MAE_ACTION_RULE_UPDATE_OUT_LEN);
+	response = _MCDI_DWORD(inbuf, MAE_ACTION_RULE_UPDATE_IN_RESPONSE);
+
+	MCDI_SET_DWORD(inbuf, MAE_ACTION_RULE_UPDATE_IN_AR_ID, id);
+	if (efx_mae_asl_id(acts_id)) {
+		MCDI_STRUCT_SET_DWORD(response, MAE_ACTION_RULE_RESPONSE_ASL_ID, acts_id);
+		MCDI_STRUCT_SET_DWORD(response, MAE_ACTION_RULE_RESPONSE_AS_ID,
+				      MC_CMD_MAE_ACTION_SET_ALLOC_OUT_ACTION_SET_ID_NULL);
+	} else {
+		/* We only had one AS, so we didn't wrap it in an ASL */
+		MCDI_STRUCT_SET_DWORD(response, MAE_ACTION_RULE_RESPONSE_ASL_ID,
+				      MC_CMD_MAE_ACTION_SET_LIST_ALLOC_OUT_ACTION_SET_LIST_ID_NULL);
+		MCDI_STRUCT_SET_DWORD(response, MAE_ACTION_RULE_RESPONSE_AS_ID, acts_id);
+	}
+	return efx_mcdi_rpc(efx, MC_CMD_MAE_ACTION_RULE_UPDATE, inbuf, sizeof(inbuf),
+			    NULL, 0, NULL);
+}
+
 int efx_mae_delete_rule(struct efx_nic *efx, u32 id)
 {
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_MAE_ACTION_RULE_DELETE_OUT_LEN(1));

drivers/net/ethernet/sfc/mae.h

@@ -90,6 +90,13 @@ int efx_mae_check_encap_type_supported(struct efx_nic *efx,
 int efx_mae_allocate_counter(struct efx_nic *efx, struct efx_tc_counter *cnt);
 int efx_mae_free_counter(struct efx_nic *efx, struct efx_tc_counter *cnt);
 
+int efx_mae_allocate_encap_md(struct efx_nic *efx,
+			      struct efx_tc_encap_action *encap);
+int efx_mae_update_encap_md(struct efx_nic *efx,
+			    struct efx_tc_encap_action *encap);
+int efx_mae_free_encap_md(struct efx_nic *efx,
+			  struct efx_tc_encap_action *encap);
+
 int efx_mae_alloc_action_set(struct efx_nic *efx, struct efx_tc_action_set *act);
 int efx_mae_free_action_set(struct efx_nic *efx, u32 fw_id);
 
@@ -105,6 +112,7 @@ int efx_mae_unregister_encap_match(struct efx_nic *efx,
 
 int efx_mae_insert_rule(struct efx_nic *efx, const struct efx_tc_match *match,
 			u32 prio, u32 acts_id, u32 *id);
+int efx_mae_update_rule(struct efx_nic *efx, u32 acts_id, u32 id);
 int efx_mae_delete_rule(struct efx_nic *efx, u32 id);
 
 int efx_init_mae(struct efx_nic *efx);

drivers/net/ethernet/sfc/net_driver.h

@@ -27,6 +27,7 @@
 #include <linux/mtd/mtd.h>
 #include <net/busy_poll.h>
 #include <net/xdp.h>
+#include <net/netevent.h>
 
 #include "enum.h"
 #include "bitfield.h"
@@ -996,6 +997,7 @@ struct efx_mae;
  * @xdp_rxq_info_failed: Have any of the rx queues failed to initialise their
  *      xdp_rxq_info structures?
  * @netdev_notifier: Netdevice notifier.
+ * @netevent_notifier: Netevent notifier (for neighbour updates).
  * @tc: state for TC offload (EF100).
  * @devlink: reference to devlink structure owned by this device
  * @dl_port: devlink port associated with the PF
@@ -1183,6 +1185,7 @@ struct efx_nic {
 	bool xdp_rxq_info_failed;
 
 	struct notifier_block netdev_notifier;
+	struct notifier_block netevent_notifier;
 	struct efx_tc_state *tc;
 
 	struct devlink *devlink;

drivers/net/ethernet/sfc/tc.c

@@ -14,11 +14,12 @@
 #include <net/geneve.h>
 #include "tc.h"
 #include "tc_bindings.h"
+#include "tc_encap_actions.h"
 #include "mae.h"
 #include "ef100_rep.h"
 #include "efx.h"
 
-static enum efx_encap_type efx_tc_indr_netdev_type(struct net_device *net_dev)
+enum efx_encap_type efx_tc_indr_netdev_type(struct net_device *net_dev)
 {
 	if (netif_is_vxlan(net_dev))
 		return EFX_ENCAP_TYPE_VXLAN;
@@ -33,8 +34,8 @@ static enum efx_encap_type efx_tc_indr_netdev_type(struct net_device *net_dev)
  * May return NULL for the PF (us), or an error pointer for a device that
  * isn't supported as a TC offload endpoint
  */
-static struct efx_rep *efx_tc_flower_lookup_efv(struct efx_nic *efx,
-						struct net_device *dev)
+struct efx_rep *efx_tc_flower_lookup_efv(struct efx_nic *efx,
+					 struct net_device *dev)
 {
 	struct efx_rep *efv;
 
@@ -70,7 +71,7 @@ static s64 efx_tc_flower_internal_mport(struct efx_nic *efx, struct efx_rep *efv
 }
 
 /* Convert a driver-internal vport ID into an external device (wire or VF) */
-static s64 efx_tc_flower_external_mport(struct efx_nic *efx, struct efx_rep *efv)
+s64 efx_tc_flower_external_mport(struct efx_nic *efx, struct efx_rep *efv)
 {
 	u32 mport;
 
@@ -111,6 +112,10 @@ static void efx_tc_free_action_set(struct efx_nic *efx,
 	}
 	if (act->count)
 		efx_tc_flower_put_counter_index(efx, act->count);
+	if (act->encap_md) {
+		list_del(&act->encap_user);
+		efx_tc_flower_release_encap_md(efx, act->encap_md);
+	}
 	kfree(act);
 }
 
@@ -594,6 +599,7 @@ enum efx_tc_action_order {
 	EFX_TC_AO_VLAN_POP,
 	EFX_TC_AO_VLAN_PUSH,
 	EFX_TC_AO_COUNT,
+	EFX_TC_AO_ENCAP,
 	EFX_TC_AO_DELIVER
 };
 /* Determine whether we can add @new action without violating order */
@@ -623,6 +629,10 @@ static bool efx_tc_flower_action_order_ok(const struct efx_tc_action_set *act,
 		if (act->count)
 			return false;
 		fallthrough;
+	case EFX_TC_AO_ENCAP:
+		if (act->encap_md)
+			return false;
+		fallthrough;
 	case EFX_TC_AO_DELIVER:
 		return !act->deliver;
 	default:
@@ -918,11 +928,13 @@ static int efx_tc_flower_replace(struct efx_nic *efx,
 {
 	struct flow_rule *fr = flow_cls_offload_flow_rule(tc);
 	struct netlink_ext_ack *extack = tc->common.extack;
+	const struct ip_tunnel_info *encap_info = NULL;
 	struct efx_tc_flow_rule *rule = NULL, *old;
 	struct efx_tc_action_set *act = NULL;
 	const struct flow_action_entry *fa;
 	struct efx_rep *from_efv, *to_efv;
 	struct efx_tc_match match;
+	u32 acts_id;
 	s64 rc;
 	int i;
 
@@ -1087,6 +1099,48 @@ static int efx_tc_flower_replace(struct efx_nic *efx,
 		case FLOW_ACTION_MIRRED:
 			save = *act;
 
+			if (encap_info) {
+				struct efx_tc_encap_action *encap;
+
+				if (!efx_tc_flower_action_order_ok(act,
+								   EFX_TC_AO_ENCAP)) {
+					rc = -EOPNOTSUPP;
+					NL_SET_ERR_MSG_MOD(extack, "Encap action violates action order");
+					goto release;
+				}
+				encap = efx_tc_flower_create_encap_md(
+						efx, encap_info, fa->dev, extack);
+				if (IS_ERR_OR_NULL(encap)) {
+					rc = PTR_ERR(encap);
+					if (!rc)
+						rc = -EIO; /* arbitrary */
+					goto release;
+				}
+				act->encap_md = encap;
+				list_add_tail(&act->encap_user, &encap->users);
+				act->dest_mport = encap->dest_mport;
+				act->deliver = 1;
+				rc = efx_mae_alloc_action_set(efx, act);
+				if (rc) {
+					NL_SET_ERR_MSG_MOD(extack, "Failed to write action set to hw (encap)");
+					goto release;
+				}
+				list_add_tail(&act->list, &rule->acts.list);
+				act->user = &rule->acts;
+				act = NULL;
+				if (fa->id == FLOW_ACTION_REDIRECT)
+					break; /* end of the line */
+				/* Mirror, so continue on with saved act */
+				save.count = NULL;
+				act = kzalloc(sizeof(*act), GFP_USER);
+				if (!act) {
+					rc = -ENOMEM;
+					goto release;
+				}
+				*act = save;
+				break;
+			}
+
 			if (!efx_tc_flower_action_order_ok(act, EFX_TC_AO_DELIVER)) {
 				/* can't happen */
 				rc = -EOPNOTSUPP;
@@ -1150,6 +1204,37 @@ static int efx_tc_flower_replace(struct efx_nic *efx,
 			act->vlan_proto[act->vlan_push] = fa->vlan.proto;
 			act->vlan_push++;
 			break;
+		case FLOW_ACTION_TUNNEL_ENCAP:
+			if (encap_info) {
+				/* Can't specify encap multiple times.
+				 * If you want to overwrite an existing
+				 * encap_info, use an intervening
+				 * FLOW_ACTION_TUNNEL_DECAP to clear it.
+				 */
+				NL_SET_ERR_MSG_MOD(extack, "Tunnel key set when already set");
+				rc = -EINVAL;
+				goto release;
+			}
+			if (!fa->tunnel) {
+				NL_SET_ERR_MSG_MOD(extack, "Tunnel key set is missing key");
+				rc = -EOPNOTSUPP;
+				goto release;
+			}
+			encap_info = fa->tunnel;
+			break;
+		case FLOW_ACTION_TUNNEL_DECAP:
+			if (encap_info) {
+				encap_info = NULL;
+				break;
+			}
+			/* Since we don't support enc_key matches on ingress
+			 * (and if we did there'd be no tunnel-device to give
+			 * us a type), we can't offload a decap that's not
+			 * just undoing a previous encap action.
+			 */
+			NL_SET_ERR_MSG_MOD(extack, "Cannot offload tunnel decap action without tunnel device");
+			rc = -EOPNOTSUPP;
+			goto release;
 		default:
 			NL_SET_ERR_MSG_FMT_MOD(extack, "Unhandled action %u",
 					       fa->id);
@@ -1193,8 +1278,21 @@ static int efx_tc_flower_replace(struct efx_nic *efx,
 		NL_SET_ERR_MSG_MOD(extack, "Failed to write action set list to hw");
 		goto release;
 	}
+	if (from_efv == EFX_EFV_PF)
+		/* PF netdev, so rule applies to traffic from wire */
+		rule->fallback = &efx->tc->facts.pf;
+	else
+		/* repdev, so rule applies to traffic from representee */
+		rule->fallback = &efx->tc->facts.reps;
+	if (!efx_tc_check_ready(efx, rule)) {
+		netif_dbg(efx, drv, efx->net_dev, "action not ready for hw\n");
+		acts_id = rule->fallback->fw_id;
+	} else {
+		netif_dbg(efx, drv, efx->net_dev, "ready for hw\n");
+		acts_id = rule->acts.fw_id;
+	}
 	rc = efx_mae_insert_rule(efx, &rule->match, EFX_TC_PRIO_TC,
-				 rule->acts.fw_id, &rule->fw_id);
+				 acts_id, &rule->fw_id);
 	if (rc) {
 		NL_SET_ERR_MSG_MOD(extack, "Failed to insert rule in hw");
 		goto release_acts;
@@ -1391,6 +1489,58 @@ void efx_tc_deconfigure_default_rule(struct efx_nic *efx,
 	rule->fw_id = MC_CMD_MAE_ACTION_RULE_INSERT_OUT_ACTION_RULE_ID_NULL;
 }
 
+static int efx_tc_configure_fallback_acts(struct efx_nic *efx, u32 eg_port,
+					  struct efx_tc_action_set_list *acts)
+{
+	struct efx_tc_action_set *act;
+	int rc;
+
+	act = kzalloc(sizeof(*act), GFP_KERNEL);
+	if (!act)
+		return -ENOMEM;
+	act->deliver = 1;
+	act->dest_mport = eg_port;
+	rc = efx_mae_alloc_action_set(efx, act);
+	if (rc)
+		goto fail1;
+	EFX_WARN_ON_PARANOID(!list_empty(&acts->list));
+	list_add_tail(&act->list, &acts->list);
+	rc = efx_mae_alloc_action_set_list(efx, acts);
+	if (rc)
+		goto fail2;
+	return 0;
+fail2:
+	list_del(&act->list);
+	efx_mae_free_action_set(efx, act->fw_id);
+fail1:
+	kfree(act);
+	return rc;
+}
+
+static int efx_tc_configure_fallback_acts_pf(struct efx_nic *efx)
+{
+	struct efx_tc_action_set_list *acts = &efx->tc->facts.pf;
+	u32 eg_port;
+
+	efx_mae_mport_uplink(efx, &eg_port);
+	return efx_tc_configure_fallback_acts(efx, eg_port, acts);
+}
+
+static int efx_tc_configure_fallback_acts_reps(struct efx_nic *efx)
+{
+	struct efx_tc_action_set_list *acts = &efx->tc->facts.reps;
+	u32 eg_port;
+
+	efx_mae_mport_mport(efx, efx->tc->reps_mport_id, &eg_port);
+	return efx_tc_configure_fallback_acts(efx, eg_port, acts);
+}
+
+static void efx_tc_deconfigure_fallback_acts(struct efx_nic *efx,
+					     struct efx_tc_action_set_list *acts)
+{
+	efx_tc_free_action_set_list(efx, acts, true);
+}
+
 static int efx_tc_configure_rep_mport(struct efx_nic *efx)
 {
 	u32 rep_mport_label;
@@ -1481,6 +1631,12 @@ int efx_init_tc(struct efx_nic *efx)
 	if (rc)
 		return rc;
 	rc = efx_tc_configure_rep_mport(efx);
+	if (rc)
+		return rc;
+	rc = efx_tc_configure_fallback_acts_pf(efx);
+	if (rc)
+		return rc;
+	rc = efx_tc_configure_fallback_acts_reps(efx);
 	if (rc)
 		return rc;
 	efx->tc->up = true;
@@ -1500,6 +1656,8 @@ void efx_fini_tc(struct efx_nic *efx)
 	efx_tc_deconfigure_rep_mport(efx);
 	efx_tc_deconfigure_default_rule(efx, &efx->tc->dflt.pf);
 	efx_tc_deconfigure_default_rule(efx, &efx->tc->dflt.wire);
+	efx_tc_deconfigure_fallback_acts(efx, &efx->tc->facts.pf);
+	efx_tc_deconfigure_fallback_acts(efx, &efx->tc->facts.reps);
 	efx->tc->up = false;
 }
 
@@ -1549,6 +1707,9 @@ int efx_init_struct_tc(struct efx_nic *efx)
 
 	mutex_init(&efx->tc->mutex);
 	init_waitqueue_head(&efx->tc->flush_wq);
+	rc = efx_tc_init_encap_actions(efx);
+	if (rc < 0)
+		goto fail_encap_actions;
 	rc = efx_tc_init_counters(efx);
 	if (rc < 0)
 		goto fail_counters;
@@ -1564,6 +1725,10 @@ int efx_init_struct_tc(struct efx_nic *efx)
 	efx->tc->dflt.pf.fw_id = MC_CMD_MAE_ACTION_RULE_INSERT_OUT_ACTION_RULE_ID_NULL;
 	INIT_LIST_HEAD(&efx->tc->dflt.wire.acts.list);
 	efx->tc->dflt.wire.fw_id = MC_CMD_MAE_ACTION_RULE_INSERT_OUT_ACTION_RULE_ID_NULL;
+	INIT_LIST_HEAD(&efx->tc->facts.pf.list);
+	efx->tc->facts.pf.fw_id = MC_CMD_MAE_ACTION_SET_ALLOC_OUT_ACTION_SET_ID_NULL;
+	INIT_LIST_HEAD(&efx->tc->facts.reps.list);
+	efx->tc->facts.reps.fw_id = MC_CMD_MAE_ACTION_SET_ALLOC_OUT_ACTION_SET_ID_NULL;
 	efx->extra_channel_type[EFX_EXTRA_CHANNEL_TC] = &efx_tc_channel_type;
 	return 0;
 fail_match_action_ht:
@@ -1571,6 +1736,8 @@ int efx_init_struct_tc(struct efx_nic *efx)
 fail_encap_match_ht:
 	efx_tc_destroy_counters(efx);
 fail_counters:
+	efx_tc_destroy_encap_actions(efx);
+fail_encap_actions:
 	mutex_destroy(&efx->tc->mutex);
 	kfree(efx->tc->caps);
 fail_alloc_caps:
@@ -1589,11 +1756,16 @@ void efx_fini_struct_tc(struct efx_nic *efx)
 			     MC_CMD_MAE_ACTION_RULE_INSERT_OUT_ACTION_RULE_ID_NULL);
 	EFX_WARN_ON_PARANOID(efx->tc->dflt.wire.fw_id !=
 			     MC_CMD_MAE_ACTION_RULE_INSERT_OUT_ACTION_RULE_ID_NULL);
+	EFX_WARN_ON_PARANOID(efx->tc->facts.pf.fw_id !=
+			     MC_CMD_MAE_ACTION_SET_LIST_ALLOC_OUT_ACTION_SET_LIST_ID_NULL);
+	EFX_WARN_ON_PARANOID(efx->tc->facts.reps.fw_id !=
+			     MC_CMD_MAE_ACTION_SET_LIST_ALLOC_OUT_ACTION_SET_LIST_ID_NULL);
 	rhashtable_free_and_destroy(&efx->tc->match_action_ht, efx_tc_flow_free,
 				    efx);
 	rhashtable_free_and_destroy(&efx->tc->encap_match_ht,
 				    efx_tc_encap_match_free, NULL);
 	efx_tc_fini_counters(efx);
+	efx_tc_fini_encap_actions(efx);
 	mutex_unlock(&efx->tc->mutex);
 	mutex_destroy(&efx->tc->mutex);
 	kfree(efx->tc->caps);

drivers/net/ethernet/sfc/tc.h

@@ -25,6 +25,8 @@ static inline bool efx_ipv6_addr_all_ones(struct in6_addr *addr)
 }
 #endif
 
+struct efx_tc_encap_action; /* see tc_encap_actions.h */
+
 struct efx_tc_action_set {
 	u16 vlan_push:2;
 	u16 vlan_pop:2;
@@ -33,6 +35,9 @@ struct efx_tc_action_set {
 	__be16 vlan_tci[2]; /* TCIs for vlan_push */
 	__be16 vlan_proto[2]; /* Ethertypes for vlan_push */
 	struct efx_tc_counter_index *count;
+	struct efx_tc_encap_action *encap_md; /* entry in tc_encap_ht table */
+	struct list_head encap_user; /* entry on encap_md->users list */
+	struct efx_tc_action_set_list *user; /* Only populated if encap_md */
 	u32 dest_mport;
 	u32 fw_id; /* index of this entry in firmware actions table */
 	struct list_head list;
@@ -127,6 +132,7 @@ struct efx_tc_flow_rule {
 	struct rhash_head linkage;
 	struct efx_tc_match match;
 	struct efx_tc_action_set_list acts;
+	struct efx_tc_action_set_list *fallback; /* what to use when unready? */
 	u32 fw_id;
 };
 
@@ -144,8 +150,10 @@ enum efx_tc_rule_prios {
  * @mutex: Used to serialise operations on TC hashtables
  * @counter_ht: Hashtable of TC counters (FW IDs and counter values)
  * @counter_id_ht: Hashtable mapping TC counter cookies to counters
+ * @encap_ht: Hashtable of TC encap actions
  * @encap_match_ht: Hashtable of TC encap matches
  * @match_action_ht: Hashtable of TC match-action rules
+ * @neigh_ht: Hashtable of neighbour watches (&struct efx_neigh_binder)
  * @reps_mport_id: MAE port allocated for representor RX
  * @reps_filter_uc: VNIC filter for representor unicast RX (promisc)
  * @reps_filter_mc: VNIC filter for representor multicast RX (allmulti)
@@ -160,6 +168,11 @@ enum efx_tc_rule_prios {
  *	%EFX_TC_PRIO_DFLT.  Named by *ingress* port
  * @dflt.pf: rule for traffic ingressing from PF (egresses to wire)
  * @dflt.wire: rule for traffic ingressing from wire (egresses to PF)
+ * @facts: Fallback action-set-lists for unready rules.  Named by *egress* port
+ * @facts.pf: action-set-list for unready rules on PF netdev, hence applying to
+ *	traffic from wire, and egressing to PF
+ * @facts.reps: action-set-list for unready rules on representors, hence
+ *	applying to traffic from representees, and egressing to the reps mport
  * @up: have TC datastructures been set up?
  */
 struct efx_tc_state {
@@ -168,8 +181,10 @@ struct efx_tc_state {
 	struct mutex mutex;
 	struct rhashtable counter_ht;
 	struct rhashtable counter_id_ht;
+	struct rhashtable encap_ht;
 	struct rhashtable encap_match_ht;
 	struct rhashtable match_action_ht;
+	struct rhashtable neigh_ht;
 	u32 reps_mport_id, reps_mport_vport_id;
 	s32 reps_filter_uc, reps_filter_mc;
 	bool flush_counters;
@@ -180,11 +195,19 @@ struct efx_tc_state {
 		struct efx_tc_flow_rule pf;
 		struct efx_tc_flow_rule wire;
 	} dflt;
+	struct {
+		struct efx_tc_action_set_list pf;
+		struct efx_tc_action_set_list reps;
+	} facts;
 	bool up;
 };
 
 struct efx_rep;
 
+enum efx_encap_type efx_tc_indr_netdev_type(struct net_device *net_dev);
+struct efx_rep *efx_tc_flower_lookup_efv(struct efx_nic *efx,
+					 struct net_device *dev);
+s64 efx_tc_flower_external_mport(struct efx_nic *efx, struct efx_rep *efv);
 int efx_tc_configure_default_rule_rep(struct efx_rep *efv);
 void efx_tc_deconfigure_default_rule(struct efx_nic *efx,
 				     struct efx_tc_flow_rule *rule);

drivers/net/ethernet/sfc/tc_bindings.c

@@ -10,6 +10,7 @@
 
 #include "tc_bindings.h"
 #include "tc.h"
+#include "tc_encap_actions.h"
 
 struct efx_tc_block_binding {
 	struct list_head list;
@@ -226,3 +227,15 @@ int efx_tc_setup(struct net_device *net_dev, enum tc_setup_type type,
 
 	return -EOPNOTSUPP;
 }
+
+int efx_tc_netdev_event(struct efx_nic *efx, unsigned long event,
+			struct net_device *net_dev)
+{
+	if (efx->type->is_vf)
+		return NOTIFY_DONE;
+
+	if (event == NETDEV_UNREGISTER)
+		efx_tc_unregister_egdev(efx, net_dev);
+
+	return NOTIFY_OK;
+}

drivers/net/ethernet/sfc/tc_bindings.h

@@ -26,4 +26,6 @@ int efx_tc_indr_setup_cb(struct net_device *net_dev, struct Qdisc *sch,
 			 void *cb_priv, enum tc_setup_type type,
 			 void *type_data, void *data,
 			 void (*cleanup)(struct flow_block_cb *block_cb));
+int efx_tc_netdev_event(struct efx_nic *efx, unsigned long event,
+			struct net_device *net_dev);
 #endif /* EFX_TC_BINDINGS_H */

drivers/net/ethernet/sfc/tc_encap_actions.c

+// SPDX-License-Identifier: GPL-2.0-only
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2023, Advanced Micro Devices, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include "tc_encap_actions.h"
+#include "tc.h"
+#include "mae.h"
+#include <net/vxlan.h>
+#include <net/geneve.h>
+#include <net/netevent.h>
+#include <net/arp.h>
+
+static const struct rhashtable_params efx_neigh_ht_params = {
+	.key_len	= offsetof(struct efx_neigh_binder, ha),
+	.key_offset	= 0,
+	.head_offset	= offsetof(struct efx_neigh_binder, linkage),
+};
+
+static const struct rhashtable_params efx_tc_encap_ht_params = {
+	.key_len	= offsetofend(struct efx_tc_encap_action, key),
+	.key_offset	= 0,
+	.head_offset	= offsetof(struct efx_tc_encap_action, linkage),
+};
+
+static void efx_tc_encap_free(void *ptr, void *__unused)
+{
+	struct efx_tc_encap_action *enc = ptr;
+
+	WARN_ON(refcount_read(&enc->ref));
+	kfree(enc);
+}
+
+static void efx_neigh_free(void *ptr, void *__unused)
+{
+	struct efx_neigh_binder *neigh = ptr;
+
+	WARN_ON(refcount_read(&neigh->ref));
+	WARN_ON(!list_empty(&neigh->users));
+	put_net_track(neigh->net, &neigh->ns_tracker);
+	netdev_put(neigh->egdev, &neigh->dev_tracker);
+	kfree(neigh);
+}
+
+int efx_tc_init_encap_actions(struct efx_nic *efx)
+{
+	int rc;
+
+	rc = rhashtable_init(&efx->tc->neigh_ht, &efx_neigh_ht_params);
+	if (rc < 0)
+		goto fail_neigh_ht;
+	rc = rhashtable_init(&efx->tc->encap_ht, &efx_tc_encap_ht_params);
+	if (rc < 0)
+		goto fail_encap_ht;
+	return 0;
+fail_encap_ht:
+	rhashtable_destroy(&efx->tc->neigh_ht);
+fail_neigh_ht:
+	return rc;
+}
+
+/* Only call this in init failure teardown.
+ * Normal exit should fini instead as there may be entries in the table.
+ */
+void efx_tc_destroy_encap_actions(struct efx_nic *efx)
+{
+	rhashtable_destroy(&efx->tc->encap_ht);
+	rhashtable_destroy(&efx->tc->neigh_ht);
+}
+
+void efx_tc_fini_encap_actions(struct efx_nic *efx)
+{
+	rhashtable_free_and_destroy(&efx->tc->encap_ht, efx_tc_encap_free, NULL);
+	rhashtable_free_and_destroy(&efx->tc->neigh_ht, efx_neigh_free, NULL);
+}
+
+static void efx_neigh_update(struct work_struct *work);
+
+static int efx_bind_neigh(struct efx_nic *efx,
+			  struct efx_tc_encap_action *encap, struct net *net,
+			  struct netlink_ext_ack *extack)
+{
+	struct efx_neigh_binder *neigh, *old;
+	struct flowi6 flow6 = {};
+	struct flowi4 flow4 = {};
+	int rc;
+
+	/* GCC stupidly thinks that only values explicitly listed in the enum
+	 * definition can _possibly_ be sensible case values, so without this
+	 * cast it complains about the IPv6 versions.
+	 */
+	switch ((int)encap->type) {
+	case EFX_ENCAP_TYPE_VXLAN:
+	case EFX_ENCAP_TYPE_GENEVE:
+		flow4.flowi4_proto = IPPROTO_UDP;
+		flow4.fl4_dport = encap->key.tp_dst;
+		flow4.flowi4_tos = encap->key.tos;
+		flow4.daddr = encap->key.u.ipv4.dst;
+		flow4.saddr = encap->key.u.ipv4.src;
+		break;
+	case EFX_ENCAP_TYPE_VXLAN | EFX_ENCAP_FLAG_IPV6:
+	case EFX_ENCAP_TYPE_GENEVE | EFX_ENCAP_FLAG_IPV6:
+		flow6.flowi6_proto = IPPROTO_UDP;
+		flow6.fl6_dport = encap->key.tp_dst;
+		flow6.flowlabel = ip6_make_flowinfo(encap->key.tos,
+						    encap->key.label);
+		flow6.daddr = encap->key.u.ipv6.dst;
+		flow6.saddr = encap->key.u.ipv6.src;
+		break;
+	default:
+		NL_SET_ERR_MSG_FMT_MOD(extack, "Unsupported encap type %d",
+				       (int)encap->type);
+		return -EOPNOTSUPP;
+	}
+
+	neigh = kzalloc(sizeof(*neigh), GFP_KERNEL_ACCOUNT);
+	if (!neigh)
+		return -ENOMEM;
+	neigh->net = get_net_track(net, &neigh->ns_tracker, GFP_KERNEL_ACCOUNT);
+	neigh->dst_ip = flow4.daddr;
+	neigh->dst_ip6 = flow6.daddr;
+
+	old = rhashtable_lookup_get_insert_fast(&efx->tc->neigh_ht,
+						&neigh->linkage,
+						efx_neigh_ht_params);
+	if (old) {
+		/* don't need our new entry */
+		put_net_track(neigh->net, &neigh->ns_tracker);
+		kfree(neigh);
+		if (!refcount_inc_not_zero(&old->ref))
+			return -EAGAIN;
+		/* existing entry found, ref taken */
+		neigh = old;
+	} else {
+		/* New entry.  We need to initiate a lookup */
+		struct neighbour *n;
+		struct rtable *rt;
+
+		if (encap->type & EFX_ENCAP_FLAG_IPV6) {
+#if IS_ENABLED(CONFIG_IPV6)
+			struct dst_entry *dst;
+
+			dst = ipv6_stub->ipv6_dst_lookup_flow(net, NULL, &flow6,
+							      NULL);
+			rc = PTR_ERR_OR_ZERO(dst);
+			if (rc) {
+				NL_SET_ERR_MSG_MOD(extack, "Failed to lookup route for IPv6 encap");
+				goto out_free;
+			}
+			neigh->egdev = dst->dev;
+			netdev_hold(neigh->egdev, &neigh->dev_tracker,
+				    GFP_KERNEL_ACCOUNT);
+			neigh->ttl = ip6_dst_hoplimit(dst);
+			n = dst_neigh_lookup(dst, &flow6.daddr);
+			dst_release(dst);
+#else
+			/* We shouldn't ever get here, because if IPv6 isn't
+			 * enabled how did someone create an IPv6 tunnel_key?
+			 */
+			rc = -EOPNOTSUPP;
+			NL_SET_ERR_MSG_MOD(extack, "No IPv6 support (neigh bind)");
+#endif
+		} else {
+			rt = ip_route_output_key(net, &flow4);
+			if (IS_ERR_OR_NULL(rt)) {
+				rc = PTR_ERR_OR_ZERO(rt);
+				if (!rc)
+					rc = -EIO;
+				NL_SET_ERR_MSG_MOD(extack, "Failed to lookup route for encap");
+				goto out_free;
+			}
+			neigh->egdev = rt->dst.dev;
+			netdev_hold(neigh->egdev, &neigh->dev_tracker,
+				    GFP_KERNEL_ACCOUNT);
+			neigh->ttl = ip4_dst_hoplimit(&rt->dst);
+			n = dst_neigh_lookup(&rt->dst, &flow4.daddr);
+			ip_rt_put(rt);
+		}
+		if (!n) {
+			rc = -ENETUNREACH;
+			NL_SET_ERR_MSG_MOD(extack, "Failed to lookup neighbour for encap");
+			netdev_put(neigh->egdev, &neigh->dev_tracker);
+			goto out_free;
+		}
+		refcount_set(&neigh->ref, 1);
+		INIT_LIST_HEAD(&neigh->users);
+		read_lock_bh(&n->lock);
+		ether_addr_copy(neigh->ha, n->ha);
+		neigh->n_valid = n->nud_state & NUD_VALID;
+		read_unlock_bh(&n->lock);
+		rwlock_init(&neigh->lock);
+		INIT_WORK(&neigh->work, efx_neigh_update);
+		neigh->efx = efx;
+		neigh->used = jiffies;
+		if (!neigh->n_valid)
+			/* Prod ARP to find us a neighbour */
+			neigh_event_send(n, NULL);
+		neigh_release(n);
+	}
+	/* Add us to this neigh */
+	encap->neigh = neigh;
+	list_add_tail(&encap->list, &neigh->users);
+	return 0;
+
+out_free:
+	/* cleanup common to several error paths */
+	rhashtable_remove_fast(&efx->tc->neigh_ht, &neigh->linkage,
+			       efx_neigh_ht_params);
+	synchronize_rcu();
+	put_net_track(net, &neigh->ns_tracker);
+	kfree(neigh);
+	return rc;
+}
+
+static void efx_free_neigh(struct efx_neigh_binder *neigh)
+{
+	struct efx_nic *efx = neigh->efx;
+
+	rhashtable_remove_fast(&efx->tc->neigh_ht, &neigh->linkage,
+			       efx_neigh_ht_params);
+	synchronize_rcu();
+	netdev_put(neigh->egdev, &neigh->dev_tracker);
+	put_net_track(neigh->net, &neigh->ns_tracker);
+	kfree(neigh);
+}
+
+static void efx_release_neigh(struct efx_nic *efx,
+			      struct efx_tc_encap_action *encap)
+{
+	struct efx_neigh_binder *neigh = encap->neigh;
+
+	if (!neigh)
+		return;
+	list_del(&encap->list);
+	encap->neigh = NULL;
+	if (!refcount_dec_and_test(&neigh->ref))
+		return; /* still in use */
+	efx_free_neigh(neigh);
+}
+
+static void efx_gen_tun_header_eth(struct efx_tc_encap_action *encap, u16 proto)
+{
+	struct efx_neigh_binder *neigh = encap->neigh;
+	struct ethhdr *eth;
+
+	encap->encap_hdr_len = sizeof(*eth);
+	eth = (struct ethhdr *)encap->encap_hdr;
+
+	if (encap->neigh->n_valid)
+		ether_addr_copy(eth->h_dest, neigh->ha);
+	else
+		eth_zero_addr(eth->h_dest);
+	ether_addr_copy(eth->h_source, neigh->egdev->dev_addr);
+	eth->h_proto = htons(proto);
+}
+
+static void efx_gen_tun_header_ipv4(struct efx_tc_encap_action *encap, u8 ipproto, u8 len)
+{
+	struct efx_neigh_binder *neigh = encap->neigh;
+	struct ip_tunnel_key *key = &encap->key;
+	struct iphdr *ip;
+
+	ip = (struct iphdr *)(encap->encap_hdr + encap->encap_hdr_len);
+	encap->encap_hdr_len += sizeof(*ip);
+
+	ip->daddr = key->u.ipv4.dst;
+	ip->saddr = key->u.ipv4.src;
+	ip->ttl = neigh->ttl;
+	ip->protocol = ipproto;
+	ip->version = 0x4;
+	ip->ihl = 0x5;
+	ip->tot_len = cpu_to_be16(ip->ihl * 4 + len);
+	ip_send_check(ip);
+}
+
+#ifdef CONFIG_IPV6
+static void efx_gen_tun_header_ipv6(struct efx_tc_encap_action *encap, u8 ipproto, u8 len)
+{
+	struct efx_neigh_binder *neigh = encap->neigh;
+	struct ip_tunnel_key *key = &encap->key;
+	struct ipv6hdr *ip;
+
+	ip = (struct ipv6hdr *)(encap->encap_hdr + encap->encap_hdr_len);
+	encap->encap_hdr_len += sizeof(*ip);
+
+	ip6_flow_hdr(ip, key->tos, key->label);
+	ip->daddr = key->u.ipv6.dst;
+	ip->saddr = key->u.ipv6.src;
+	ip->hop_limit = neigh->ttl;
+	ip->nexthdr = ipproto;
+	ip->version = 0x6;
+	ip->payload_len = cpu_to_be16(len);
+}
+#endif
+
+static void efx_gen_tun_header_udp(struct efx_tc_encap_action *encap, u8 len)
+{
+	struct ip_tunnel_key *key = &encap->key;
+	struct udphdr *udp;
+
+	udp = (struct udphdr *)(encap->encap_hdr + encap->encap_hdr_len);
+	encap->encap_hdr_len += sizeof(*udp);
+
+	udp->dest = key->tp_dst;
+	udp->len = cpu_to_be16(sizeof(*udp) + len);
+}
+
+static void efx_gen_tun_header_vxlan(struct efx_tc_encap_action *encap)
+{
+	struct ip_tunnel_key *key = &encap->key;
+	struct vxlanhdr *vxlan;
+
+	vxlan = (struct vxlanhdr *)(encap->encap_hdr + encap->encap_hdr_len);
+	encap->encap_hdr_len += sizeof(*vxlan);
+
+	vxlan->vx_flags = VXLAN_HF_VNI;
+	vxlan->vx_vni = vxlan_vni_field(tunnel_id_to_key32(key->tun_id));
+}
+
+static void efx_gen_tun_header_geneve(struct efx_tc_encap_action *encap)
+{
+	struct ip_tunnel_key *key = &encap->key;
+	struct genevehdr *geneve;
+	u32 vni;
+
+	geneve = (struct genevehdr *)(encap->encap_hdr + encap->encap_hdr_len);
+	encap->encap_hdr_len += sizeof(*geneve);
+
+	geneve->proto_type = htons(ETH_P_TEB);
+	/* convert tun_id to host-endian so we can use host arithmetic to
+	 * extract individual bytes.
+	 */
+	vni = ntohl(tunnel_id_to_key32(key->tun_id));
+	geneve->vni[0] = vni >> 16;
+	geneve->vni[1] = vni >> 8;
+	geneve->vni[2] = vni;
+}
+
+#define vxlan_header_l4_len	(sizeof(struct udphdr) + sizeof(struct vxlanhdr))
+#define vxlan4_header_len	(sizeof(struct ethhdr) + sizeof(struct iphdr) + vxlan_header_l4_len)
+static void efx_gen_vxlan_header_ipv4(struct efx_tc_encap_action *encap)
+{
+	BUILD_BUG_ON(sizeof(encap->encap_hdr) < vxlan4_header_len);
+	efx_gen_tun_header_eth(encap, ETH_P_IP);
+	efx_gen_tun_header_ipv4(encap, IPPROTO_UDP, vxlan_header_l4_len);
+	efx_gen_tun_header_udp(encap, sizeof(struct vxlanhdr));
+	efx_gen_tun_header_vxlan(encap);
+}
+
+#define geneve_header_l4_len	(sizeof(struct udphdr) + sizeof(struct genevehdr))
+#define geneve4_header_len	(sizeof(struct ethhdr) + sizeof(struct iphdr) + geneve_header_l4_len)
+static void efx_gen_geneve_header_ipv4(struct efx_tc_encap_action *encap)
+{
+	BUILD_BUG_ON(sizeof(encap->encap_hdr) < geneve4_header_len);
+	efx_gen_tun_header_eth(encap, ETH_P_IP);
+	efx_gen_tun_header_ipv4(encap, IPPROTO_UDP, geneve_header_l4_len);
+	efx_gen_tun_header_udp(encap, sizeof(struct genevehdr));
+	efx_gen_tun_header_geneve(encap);
+}
+
+#ifdef CONFIG_IPV6
+#define vxlan6_header_len	(sizeof(struct ethhdr) + sizeof(struct ipv6hdr) + vxlan_header_l4_len)
+static void efx_gen_vxlan_header_ipv6(struct efx_tc_encap_action *encap)
+{
+	BUILD_BUG_ON(sizeof(encap->encap_hdr) < vxlan6_header_len);
+	efx_gen_tun_header_eth(encap, ETH_P_IPV6);
+	efx_gen_tun_header_ipv6(encap, IPPROTO_UDP, vxlan_header_l4_len);
+	efx_gen_tun_header_udp(encap, sizeof(struct vxlanhdr));
+	efx_gen_tun_header_vxlan(encap);
+}
+
+#define geneve6_header_len	(sizeof(struct ethhdr) + sizeof(struct ipv6hdr) + geneve_header_l4_len)
+static void efx_gen_geneve_header_ipv6(struct efx_tc_encap_action *encap)
+{
+	BUILD_BUG_ON(sizeof(encap->encap_hdr) < geneve6_header_len);
+	efx_gen_tun_header_eth(encap, ETH_P_IPV6);
+	efx_gen_tun_header_ipv6(encap, IPPROTO_UDP, geneve_header_l4_len);
+	efx_gen_tun_header_udp(encap, sizeof(struct genevehdr));
+	efx_gen_tun_header_geneve(encap);
+}
+#endif
+
+static void efx_gen_encap_header(struct efx_nic *efx,
+				 struct efx_tc_encap_action *encap)
+{
+	encap->n_valid = encap->neigh->n_valid;
+
+	/* GCC stupidly thinks that only values explicitly listed in the enum
+	 * definition can _possibly_ be sensible case values, so without this
+	 * cast it complains about the IPv6 versions.
+	 */
+	switch ((int)encap->type) {
+	case EFX_ENCAP_TYPE_VXLAN:
+		efx_gen_vxlan_header_ipv4(encap);
+		break;
+	case EFX_ENCAP_TYPE_GENEVE:
+		efx_gen_geneve_header_ipv4(encap);
+		break;
+#ifdef CONFIG_IPV6
+	case EFX_ENCAP_TYPE_VXLAN | EFX_ENCAP_FLAG_IPV6:
+		efx_gen_vxlan_header_ipv6(encap);
+		break;
+	case EFX_ENCAP_TYPE_GENEVE | EFX_ENCAP_FLAG_IPV6:
+		efx_gen_geneve_header_ipv6(encap);
+		break;
+#endif
+	default:
+		/* unhandled encap type, can't happen */
+		if (net_ratelimit())
+			netif_err(efx, drv, efx->net_dev,
+				  "Bogus encap type %d, can't generate\n",
+				  encap->type);
+
+		/* Use fallback action. */
+		encap->n_valid = false;
+		break;
+	}
+}
+
+static void efx_tc_update_encap(struct efx_nic *efx,
+				struct efx_tc_encap_action *encap)
+{
+	struct efx_tc_action_set_list *acts, *fallback;
+	struct efx_tc_flow_rule *rule;
+	struct efx_tc_action_set *act;
+	int rc;
+
+	if (encap->n_valid) {
+		/* Make sure no rules are using this encap while we change it */
+		list_for_each_entry(act, &encap->users, encap_user) {
+			acts = act->user;
+			if (WARN_ON(!acts)) /* can't happen */
+				continue;
+			rule = container_of(acts, struct efx_tc_flow_rule, acts);
+			if (rule->fallback)
+				fallback = rule->fallback;
+			else /* fallback fallback: deliver to PF */
+				fallback = &efx->tc->facts.pf;
+			rc = efx_mae_update_rule(efx, fallback->fw_id,
+						 rule->fw_id);
+			if (rc)
+				netif_err(efx, drv, efx->net_dev,
+					  "Failed to update (f) rule %08x rc %d\n",
+					  rule->fw_id, rc);
+			else
+				netif_dbg(efx, drv, efx->net_dev, "Updated (f) rule %08x\n",
+					  rule->fw_id);
+		}
+	}
+
+	/* Make sure we don't leak arbitrary bytes on the wire;
+	 * set an all-0s ethernet header.  A successful call to
+	 * efx_gen_encap_header() will overwrite this.
+	 */
+	memset(encap->encap_hdr, 0, sizeof(encap->encap_hdr));
+	encap->encap_hdr_len = ETH_HLEN;
+
+	if (encap->neigh) {
+		read_lock_bh(&encap->neigh->lock);
+		efx_gen_encap_header(efx, encap);
+		read_unlock_bh(&encap->neigh->lock);
+	} else {
+		encap->n_valid = false;
+	}
+
+	rc = efx_mae_update_encap_md(efx, encap);
+	if (rc) {
+		netif_err(efx, drv, efx->net_dev,
+			  "Failed to update encap hdr %08x rc %d\n",
+			  encap->fw_id, rc);
+		return;
+	}
+	netif_dbg(efx, drv, efx->net_dev, "Updated encap hdr %08x\n",
+		  encap->fw_id);
+	if (!encap->n_valid)
+		return;
+	/* Update rule users: use the action if they are now ready */
+	list_for_each_entry(act, &encap->users, encap_user) {
+		acts = act->user;
+		if (WARN_ON(!acts)) /* can't happen */
+			continue;
+		rule = container_of(acts, struct efx_tc_flow_rule, acts);
+		if (!efx_tc_check_ready(efx, rule))
+			continue;
+		rc = efx_mae_update_rule(efx, acts->fw_id, rule->fw_id);
+		if (rc)
+			netif_err(efx, drv, efx->net_dev,
+				  "Failed to update rule %08x rc %d\n",
+				  rule->fw_id, rc);
+		else
+			netif_dbg(efx, drv, efx->net_dev, "Updated rule %08x\n",
+				  rule->fw_id);
+	}
+}
+
+static void efx_neigh_update(struct work_struct *work)
+{
+	struct efx_neigh_binder *neigh = container_of(work, struct efx_neigh_binder, work);
+	struct efx_tc_encap_action *encap;
+	struct efx_nic *efx = neigh->efx;
+
+	mutex_lock(&efx->tc->mutex);
+	list_for_each_entry(encap, &neigh->users, list)
+		efx_tc_update_encap(neigh->efx, encap);
+	/* release ref taken in efx_neigh_event() */
+	if (refcount_dec_and_test(&neigh->ref))
+		efx_free_neigh(neigh);
+	mutex_unlock(&efx->tc->mutex);
+}
+
+static int efx_neigh_event(struct efx_nic *efx, struct neighbour *n)
+{
+	struct efx_neigh_binder keys = {NULL}, *neigh;
+	bool n_valid, ipv6 = false;
+	char ha[ETH_ALEN];
+	size_t keysize;
+
+	if (WARN_ON(!efx->tc))
+		return NOTIFY_DONE;
+
+	if (n->tbl == &arp_tbl) {
+		keysize = sizeof(keys.dst_ip);
+#if IS_ENABLED(CONFIG_IPV6)
+	} else if (n->tbl == ipv6_stub->nd_tbl) {
+		ipv6 = true;
+		keysize = sizeof(keys.dst_ip6);
+#endif
+	} else {
+		return NOTIFY_DONE;
+	}
+	if (!n->parms) {
+		netif_warn(efx, drv, efx->net_dev, "neigh_event with no parms!\n");
+		return NOTIFY_DONE;
+	}
+	keys.net = read_pnet(&n->parms->net);
+	if (n->tbl->key_len != keysize) {
+		netif_warn(efx, drv, efx->net_dev, "neigh_event with bad key_len %u\n",
+			   n->tbl->key_len);
+		return NOTIFY_DONE;
+	}
+	read_lock_bh(&n->lock); /* Get a consistent view */
+	memcpy(ha, n->ha, ETH_ALEN);
+	n_valid = (n->nud_state & NUD_VALID) && !n->dead;
+	read_unlock_bh(&n->lock);
+	if (ipv6)
+		memcpy(&keys.dst_ip6, n->primary_key, n->tbl->key_len);
+	else
+		memcpy(&keys.dst_ip, n->primary_key, n->tbl->key_len);
+	rcu_read_lock();
+	neigh = rhashtable_lookup_fast(&efx->tc->neigh_ht, &keys,
+				       efx_neigh_ht_params);
+	if (!neigh || neigh->dying)
+		/* We're not interested in this neighbour */
+		goto done;
+	write_lock_bh(&neigh->lock);
+	if (n_valid == neigh->n_valid && !memcmp(ha, neigh->ha, ETH_ALEN)) {
+		write_unlock_bh(&neigh->lock);
+		/* Nothing has changed; no work to do */
+		goto done;
+	}
+	neigh->n_valid = n_valid;
+	memcpy(neigh->ha, ha, ETH_ALEN);
+	write_unlock_bh(&neigh->lock);
+	if (refcount_inc_not_zero(&neigh->ref)) {
+		rcu_read_unlock();
+		if (!schedule_work(&neigh->work))
+			/* failed to schedule, release the ref we just took */
+			if (refcount_dec_and_test(&neigh->ref))
+				efx_free_neigh(neigh);
+	} else {
+done:
+		rcu_read_unlock();
+	}
+	return NOTIFY_DONE;
+}
+
+bool efx_tc_check_ready(struct efx_nic *efx, struct efx_tc_flow_rule *rule)
+{
+	struct efx_tc_action_set *act;
+
+	/* Encap actions can only be offloaded if they have valid
+	 * neighbour info for the outer Ethernet header.
+	 */
+	list_for_each_entry(act, &rule->acts.list, list)
+		if (act->encap_md && !act->encap_md->n_valid)
+			return false;
+	return true;
+}
+
+struct efx_tc_encap_action *efx_tc_flower_create_encap_md(
+			struct efx_nic *efx, const struct ip_tunnel_info *info,
+			struct net_device *egdev, struct netlink_ext_ack *extack)
+{
+	enum efx_encap_type type = efx_tc_indr_netdev_type(egdev);
+	struct efx_tc_encap_action *encap, *old;
+	struct efx_rep *to_efv;
+	s64 rc;
+
+	if (type == EFX_ENCAP_TYPE_NONE) {
+		/* dest is not an encap device */
+		NL_SET_ERR_MSG_MOD(extack, "Not a (supported) tunnel device but tunnel_key is set");
+		return ERR_PTR(-EOPNOTSUPP);
+	}
+	rc = efx_mae_check_encap_type_supported(efx, type);
+	if (rc < 0) {
+		NL_SET_ERR_MSG_MOD(extack, "Firmware reports no support for this tunnel type");
+		return ERR_PTR(rc);
+	}
+	/* No support yet for Geneve options */
+	if (info->options_len) {
+		NL_SET_ERR_MSG_MOD(extack, "Unsupported tunnel options");
+		return ERR_PTR(-EOPNOTSUPP);
+	}
+	switch (info->mode) {
+	case IP_TUNNEL_INFO_TX:
+		break;
+	case IP_TUNNEL_INFO_TX | IP_TUNNEL_INFO_IPV6:
+		type |= EFX_ENCAP_FLAG_IPV6;
+		break;
+	default:
+		NL_SET_ERR_MSG_FMT_MOD(extack, "Unsupported tunnel mode %u",
+				       info->mode);
+		return ERR_PTR(-EOPNOTSUPP);
+	}
+	encap = kzalloc(sizeof(*encap), GFP_KERNEL_ACCOUNT);
+	if (!encap)
+		return ERR_PTR(-ENOMEM);
+	encap->type = type;
+	encap->key = info->key;
+	INIT_LIST_HEAD(&encap->users);
+	old = rhashtable_lookup_get_insert_fast(&efx->tc->encap_ht,
+						&encap->linkage,
+						efx_tc_encap_ht_params);
+	if (old) {
+		/* don't need our new entry */
+		kfree(encap);
+		if (!refcount_inc_not_zero(&old->ref))
+			return ERR_PTR(-EAGAIN);
+		/* existing entry found, ref taken */
+		return old;
+	}
+
+	rc = efx_bind_neigh(efx, encap, dev_net(egdev), extack);
+	if (rc < 0)
+		goto out_remove;
+	to_efv = efx_tc_flower_lookup_efv(efx, encap->neigh->egdev);
+	if (IS_ERR(to_efv)) {
+		/* neigh->egdev isn't ours */
+		NL_SET_ERR_MSG_MOD(extack, "Tunnel egress device not on switch");
+		rc = PTR_ERR(to_efv);
+		goto out_release;
+	}
+	rc = efx_tc_flower_external_mport(efx, to_efv);
+	if (rc < 0) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to identify tunnel egress m-port");
+		goto out_release;
+	}
+	encap->dest_mport = rc;
+	read_lock_bh(&encap->neigh->lock);
+	efx_gen_encap_header(efx, encap);
+	read_unlock_bh(&encap->neigh->lock);
+
+	rc = efx_mae_allocate_encap_md(efx, encap);
+	if (rc < 0) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to write tunnel header to hw");
+		goto out_release;
+	}
+
+	/* ref and return */
+	refcount_set(&encap->ref, 1);
+	return encap;
+out_release:
+	efx_release_neigh(efx, encap);
+out_remove:
+	rhashtable_remove_fast(&efx->tc->encap_ht, &encap->linkage,
+			       efx_tc_encap_ht_params);
+	kfree(encap);
+	return ERR_PTR(rc);
+}
+
+void efx_tc_flower_release_encap_md(struct efx_nic *efx,
+				    struct efx_tc_encap_action *encap)
+{
+	if (!refcount_dec_and_test(&encap->ref))
+		return; /* still in use */
+	efx_release_neigh(efx, encap);
+	rhashtable_remove_fast(&efx->tc->encap_ht, &encap->linkage,
+			       efx_tc_encap_ht_params);
+	efx_mae_free_encap_md(efx, encap);
+	kfree(encap);
+}
+
+static void efx_tc_remove_neigh_users(struct efx_nic *efx, struct efx_neigh_binder *neigh)
+{
+	struct efx_tc_encap_action *encap, *next;
+
+	list_for_each_entry_safe(encap, next, &neigh->users, list) {
+		/* Should cause neigh usage count to fall to zero, freeing it */
+		efx_release_neigh(efx, encap);
+		/* The encap has lost its neigh, so it's now unready */
+		efx_tc_update_encap(efx, encap);
+	}
+}
+
+void efx_tc_unregister_egdev(struct efx_nic *efx, struct net_device *net_dev)
+{
+	struct efx_neigh_binder *neigh;
+	struct rhashtable_iter walk;
+
+	mutex_lock(&efx->tc->mutex);
+	rhashtable_walk_enter(&efx->tc->neigh_ht, &walk);
+	rhashtable_walk_start(&walk);
+	while ((neigh = rhashtable_walk_next(&walk)) != NULL) {
+		if (IS_ERR(neigh))
+			continue;
+		if (neigh->egdev != net_dev)
+			continue;
+		neigh->dying = true;
+		rhashtable_walk_stop(&walk);
+		synchronize_rcu(); /* Make sure any updates see dying flag */
+		efx_tc_remove_neigh_users(efx, neigh); /* might sleep */
+		rhashtable_walk_start(&walk);
+	}
+	rhashtable_walk_stop(&walk);
+	rhashtable_walk_exit(&walk);
+	mutex_unlock(&efx->tc->mutex);
+}
+
+int efx_tc_netevent_event(struct efx_nic *efx, unsigned long event,
+			  void *ptr)
+{
+	if (efx->type->is_vf)
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETEVENT_NEIGH_UPDATE:
+		return efx_neigh_event(efx, ptr);
+	default:
+		return NOTIFY_DONE;
+	}
+}

drivers/net/ethernet/sfc/tc_encap_actions.h

+/* SPDX-License-Identifier: GPL-2.0-only */
+/****************************************************************************
+ * Driver for Solarflare network controllers and boards
+ * Copyright 2023, Advanced Micro Devices, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_TC_ENCAP_ACTIONS_H
+#define EFX_TC_ENCAP_ACTIONS_H
+#include "net_driver.h"
+
+#include <linux/refcount.h>
+#include <net/tc_act/tc_tunnel_key.h>
+
+/**
+ * struct efx_neigh_binder - driver state for a neighbour entry
+ * @net: the network namespace in which this neigh resides
+ * @dst_ip: the IPv4 destination address resolved by this neigh
+ * @dst_ip6: the IPv6 destination address resolved by this neigh
+ * @ha: the hardware (Ethernet) address of the neighbour
+ * @n_valid: true if the neighbour is in NUD_VALID state
+ * @lock: protects @ha and @n_valid
+ * @ttl: Time To Live associated with the route used
+ * @dying: set when egdev is going away, to skip further updates
+ * @egdev: egress device from the route lookup.  Holds a reference
+ * @dev_tracker: reference tracker entry for @egdev
+ * @ns_tracker: reference tracker entry for @ns
+ * @ref: counts encap actions referencing this entry
+ * @used: jiffies of last time traffic hit any encap action using this.
+ *      When counter reads update this, a new neighbour event is sent to
+ *      indicate that the neighbour entry is still in use.
+ * @users: list of &struct efx_tc_encap_action
+ * @linkage: entry in efx->neigh_ht (keys are @net, @dst_ip, @dst_ip6).
+ * @work: processes neighbour state changes, updates the encap actions
+ * @efx: owning NIC instance.
+ *
+ * Associates a neighbour entry with the encap actions that are
+ * interested in it, allowing the latter to be updated when the
+ * neighbour details change.
+ * Whichever of @dst_ip and @dst_ip6 is not in use will be all-zeroes,
+ * this distinguishes IPv4 from IPv6 entries.
+ */
+struct efx_neigh_binder {
+	struct net *net;
+	__be32 dst_ip;
+	struct in6_addr dst_ip6;
+	char ha[ETH_ALEN];
+	bool n_valid;
+	rwlock_t lock;
+	u8 ttl;
+	bool dying;
+	struct net_device *egdev;
+	netdevice_tracker dev_tracker;
+	netns_tracker ns_tracker;
+	refcount_t ref;
+	unsigned long used;
+	struct list_head users;
+	struct rhash_head linkage;
+	struct work_struct work;
+	struct efx_nic *efx;
+};
+
+/* This limit is arbitrary; current hardware (SN1022) handles encap headers
+ * of up to 126 bytes, but that limit is not enshrined in the MCDI protocol.
+ */
+#define EFX_TC_MAX_ENCAP_HDR	126
+struct efx_tc_encap_action {
+	enum efx_encap_type type;
+	struct ip_tunnel_key key; /* 52 bytes */
+	u32 dest_mport; /* is copied into struct efx_tc_action_set */
+	u8 encap_hdr_len;
+	bool n_valid;
+	u8 encap_hdr[EFX_TC_MAX_ENCAP_HDR];
+	struct efx_neigh_binder *neigh;
+	struct list_head list; /* entry on neigh->users list */
+	struct list_head users; /* action sets using this encap_md */
+	struct rhash_head linkage; /* efx->tc_encap_ht */
+	refcount_t ref;
+	u32 fw_id; /* index of this entry in firmware encap table */
+};
+
+/* create/uncreate/teardown hashtables */
+int efx_tc_init_encap_actions(struct efx_nic *efx);
+void efx_tc_destroy_encap_actions(struct efx_nic *efx);
+void efx_tc_fini_encap_actions(struct efx_nic *efx);
+
+struct efx_tc_flow_rule;
+bool efx_tc_check_ready(struct efx_nic *efx, struct efx_tc_flow_rule *rule);
+
+struct efx_tc_encap_action *efx_tc_flower_create_encap_md(
+			struct efx_nic *efx, const struct ip_tunnel_info *info,
+			struct net_device *egdev, struct netlink_ext_ack *extack);
+void efx_tc_flower_release_encap_md(struct efx_nic *efx,
+				    struct efx_tc_encap_action *encap);
+
+void efx_tc_unregister_egdev(struct efx_nic *efx, struct net_device *net_dev);
+int efx_tc_netevent_event(struct efx_nic *efx, unsigned long event,
+			  void *ptr);
+
+#endif /* EFX_TC_ENCAP_ACTIONS_H */