Merge branch 'mlxsw-mc-route-offload'

Jiri Pirko says:

====================
mlxsw: Add support for partial multicast route offload

Yotam says:

Previous patchset introduced support for offloading multicast MFC routes to
the Spectrum hardware. As described in that patchset, no partial offloading
is supported, i.e if a route has one output interface which is not a valid
offloadable device (e.g. pimreg device, dummy device, management NIC), the
route is trapped to the CPU and the forwarding is done in slow-path.

Add support for partial offloading of multicast routes, by letting the
hardware to forward the packet to all the in-hardware devices, while the
kernel ipmr module will continue forwarding to all other interfaces.

Similarly to the bridge, the kernel ipmr module will forward a marked
packet to an interface only if the interface has a different parent ID than
the packet's ingress interfaces.

The first patch introduces the offload_mr_fwd_mark skb field, which can be
used by offloading drivers to indicate that a packet had already gone
through multicast forwarding in hardware, similarly to the offload_fwd_mark
field that indicates that a packet had already gone through L2 forwarding
in hardware.

Patches 2 and 3 change the ipmr module to not forward packets that had
already been forwarded by the hardware, i.e. packets that are marked with
offload_mr_fwd_mark and the ingress VIF shares the same parent ID with the
egress VIF.

Patches 4, 5, 6 and 7 add the support in the mlxsw Spectrum driver for trap
and forward routes, while marking the trapped packets with the
offload_mr_fwd_mark.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c

@@ -676,6 +676,7 @@ enum mlxsw_afa_trapdisc_trap_action {
 MLXSW_ITEM32(afa, trapdisc, trap_action, 0x00, 24, 4);
 
 enum mlxsw_afa_trapdisc_forward_action {
+	MLXSW_AFA_TRAPDISC_FORWARD_ACTION_FORWARD = 1,
 	MLXSW_AFA_TRAPDISC_FORWARD_ACTION_DISCARD = 3,
 };
 
@@ -729,6 +730,22 @@ int mlxsw_afa_block_append_trap(struct mlxsw_afa_block *block, u16 trap_id)
 }
 EXPORT_SYMBOL(mlxsw_afa_block_append_trap);
 
+int mlxsw_afa_block_append_trap_and_forward(struct mlxsw_afa_block *block,
+					    u16 trap_id)
+{
+	char *act = mlxsw_afa_block_append_action(block,
+						  MLXSW_AFA_TRAPDISC_CODE,
+						  MLXSW_AFA_TRAPDISC_SIZE);
+
+	if (!act)
+		return -ENOBUFS;
+	mlxsw_afa_trapdisc_pack(act, MLXSW_AFA_TRAPDISC_TRAP_ACTION_TRAP,
+				MLXSW_AFA_TRAPDISC_FORWARD_ACTION_FORWARD,
+				trap_id);
+	return 0;
+}
+EXPORT_SYMBOL(mlxsw_afa_block_append_trap_and_forward);
+
 /* Forwarding Action
  * -----------------
  * Forwarding Action can be used to implement Policy Based Switching (PBS)

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h

@@ -61,6 +61,8 @@ int mlxsw_afa_block_continue(struct mlxsw_afa_block *block);
 int mlxsw_afa_block_jump(struct mlxsw_afa_block *block, u16 group_id);
 int mlxsw_afa_block_append_drop(struct mlxsw_afa_block *block);
 int mlxsw_afa_block_append_trap(struct mlxsw_afa_block *block, u16 trap_id);
+int mlxsw_afa_block_append_trap_and_forward(struct mlxsw_afa_block *block,
+					    u16 trap_id);
 int mlxsw_afa_block_append_fwd(struct mlxsw_afa_block *block,
 			       u8 local_port, bool in_port);
 int mlxsw_afa_block_append_vlan_modify(struct mlxsw_afa_block *block,

drivers/net/ethernet/mellanox/mlxsw/spectrum.c

@@ -3312,6 +3312,14 @@ static void mlxsw_sp_rx_listener_mark_func(struct sk_buff *skb, u8 local_port,
 	return mlxsw_sp_rx_listener_no_mark_func(skb, local_port, priv);
 }
 
+static void mlxsw_sp_rx_listener_mr_mark_func(struct sk_buff *skb,
+					      u8 local_port, void *priv)
+{
+	skb->offload_mr_fwd_mark = 1;
+	skb->offload_fwd_mark = 1;
+	return mlxsw_sp_rx_listener_no_mark_func(skb, local_port, priv);
+}
+
 static void mlxsw_sp_rx_listener_sample_func(struct sk_buff *skb, u8 local_port,
 					     void *priv)
 {
@@ -3355,6 +3363,10 @@ static void mlxsw_sp_rx_listener_sample_func(struct sk_buff *skb, u8 local_port,
 	MLXSW_RXL(mlxsw_sp_rx_listener_mark_func, _trap_id, _action,	\
 		_is_ctrl, SP_##_trap_group, DISCARD)
 
+#define MLXSW_SP_RXL_MR_MARK(_trap_id, _action, _trap_group, _is_ctrl)	\
+	MLXSW_RXL(mlxsw_sp_rx_listener_mr_mark_func, _trap_id, _action,	\
+		_is_ctrl, SP_##_trap_group, DISCARD)
+
 #define MLXSW_SP_EVENTL(_func, _trap_id)		\
 	MLXSW_EVENTL(_func, _trap_id, SP_EVENT)
 
@@ -3425,6 +3437,7 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = {
 	MLXSW_SP_RXL_MARK(IPV4_PIM, TRAP_TO_CPU, PIM, false),
 	MLXSW_SP_RXL_MARK(RPF, TRAP_TO_CPU, RPF, false),
 	MLXSW_SP_RXL_MARK(ACL1, TRAP_TO_CPU, MULTICAST, false),
+	MLXSW_SP_RXL_MR_MARK(ACL2, TRAP_TO_CPU, MULTICAST, false),
 };
 
 static int mlxsw_sp_cpu_policers_set(struct mlxsw_core *mlxsw_core)

drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c

@@ -114,9 +114,9 @@ static bool mlxsw_sp_mr_vif_valid(const struct mlxsw_sp_mr_vif *vif)
 	return mlxsw_sp_mr_vif_regular(vif) && vif->dev && vif->rif;
 }
 
-static bool mlxsw_sp_mr_vif_rif_invalid(const struct mlxsw_sp_mr_vif *vif)
+static bool mlxsw_sp_mr_vif_exists(const struct mlxsw_sp_mr_vif *vif)
 {
-	return mlxsw_sp_mr_vif_regular(vif) && vif->dev && !vif->rif;
+	return vif->dev;
 }
 
 static bool
@@ -182,14 +182,13 @@ mlxsw_sp_mr_route_action(const struct mlxsw_sp_mr_route *mr_route)
 	if (!mlxsw_sp_mr_route_valid_evifs_num(mr_route))
 		return MLXSW_SP_MR_ROUTE_ACTION_TRAP;
 
-	/* If either one of the eVIFs is not regular (VIF of type pimreg or
-	 * tunnel) or one of the VIFs has no matching RIF, trap the packet.
+	/* If one of the eVIFs has no RIF, trap-and-forward the route as there
+	 * is some more routing to do in software too.
 	 */
-	list_for_each_entry(rve, &mr_route->evif_list, route_node) {
-		if (!mlxsw_sp_mr_vif_regular(rve->mr_vif) ||
-		    mlxsw_sp_mr_vif_rif_invalid(rve->mr_vif))
-			return MLXSW_SP_MR_ROUTE_ACTION_TRAP;
-	}
+	list_for_each_entry(rve, &mr_route->evif_list, route_node)
+		if (mlxsw_sp_mr_vif_exists(rve->mr_vif) && !rve->mr_vif->rif)
+			return MLXSW_SP_MR_ROUTE_ACTION_TRAP_AND_FORWARD;
+
 	return MLXSW_SP_MR_ROUTE_ACTION_FORWARD;
 }
 

drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.h

@@ -42,6 +42,7 @@
 enum mlxsw_sp_mr_route_action {
 	MLXSW_SP_MR_ROUTE_ACTION_FORWARD,
 	MLXSW_SP_MR_ROUTE_ACTION_TRAP,
+	MLXSW_SP_MR_ROUTE_ACTION_TRAP_AND_FORWARD,
 };
 
 enum mlxsw_sp_mr_route_prio {

drivers/net/ethernet/mellanox/mlxsw/spectrum_mr_tcam.c

@@ -253,6 +253,7 @@ mlxsw_sp_mr_tcam_afa_block_create(struct mlxsw_sp *mlxsw_sp,
 		if (err)
 			goto err;
 		break;
+	case MLXSW_SP_MR_ROUTE_ACTION_TRAP_AND_FORWARD:
 	case MLXSW_SP_MR_ROUTE_ACTION_FORWARD:
 		/* If we are about to append a multicast router action, commit
 		 * the erif_list.
@@ -266,6 +267,13 @@ mlxsw_sp_mr_tcam_afa_block_create(struct mlxsw_sp *mlxsw_sp,
 						      erif_list->kvdl_index);
 		if (err)
 			goto err;
+
+		if (route_action == MLXSW_SP_MR_ROUTE_ACTION_TRAP_AND_FORWARD) {
+			err = mlxsw_afa_block_append_trap_and_forward(afa_block,
+								      MLXSW_TRAP_ID_ACL2);
+			if (err)
+				goto err;
+		}
 		break;
 	default:
 		err = -EINVAL;

drivers/net/ethernet/mellanox/mlxsw/trap.h

@@ -93,6 +93,8 @@ enum {
 	MLXSW_TRAP_ID_ACL0 = 0x1C0,
 	/* Multicast trap used for routes with trap action */
 	MLXSW_TRAP_ID_ACL1 = 0x1C1,
+	/* Multicast trap used for routes with trap-and-forward action */
+	MLXSW_TRAP_ID_ACL2 = 0x1C2,
 
 	MLXSW_TRAP_ID_MAX = 0x1FF
 };

include/linux/mroute.h

@@ -57,6 +57,7 @@ static inline bool ipmr_rule_default(const struct fib_rule *rule)
 
 struct vif_device {
 	struct net_device 	*dev;			/* Device we are using */
+	struct netdev_phys_item_id dev_parent_id;	/* Device parent ID    */
 	unsigned long	bytes_in,bytes_out;
 	unsigned long	pkt_in,pkt_out;		/* Statistics 			*/
 	unsigned long	rate_limit;		/* Traffic shaping (NI) 	*/

include/linux/skbuff.h

@@ -772,6 +772,7 @@ struct sk_buff {
 	__u8			remcsum_offload:1;
 #ifdef CONFIG_NET_SWITCHDEV
 	__u8			offload_fwd_mark:1;
+	__u8			offload_mr_fwd_mark:1;
 #endif
 #ifdef CONFIG_NET_CLS_ACT
 	__u8			tc_skip_classify:1;

net/ipv4/ipmr.c

@@ -67,6 +67,7 @@
 #include <net/fib_rules.h>
 #include <linux/netconf.h>
 #include <net/nexthop.h>
+#include <net/switchdev.h>
 
 struct ipmr_rule {
 	struct fib_rule		common;
@@ -868,6 +869,9 @@ static int vif_add(struct net *net, struct mr_table *mrt,
 		   struct vifctl *vifc, int mrtsock)
 {
 	int vifi = vifc->vifc_vifi;
+	struct switchdev_attr attr = {
+		.id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
+	};
 	struct vif_device *v = &mrt->vif_table[vifi];
 	struct net_device *dev;
 	struct in_device *in_dev;
@@ -942,6 +946,13 @@ static int vif_add(struct net *net, struct mr_table *mrt,
 
 	/* Fill in the VIF structures */
 
+	attr.orig_dev = dev;
+	if (!switchdev_port_attr_get(dev, &attr)) {
+		memcpy(v->dev_parent_id.id, attr.u.ppid.id, attr.u.ppid.id_len);
+		v->dev_parent_id.id_len = attr.u.ppid.id_len;
+	} else {
+		v->dev_parent_id.id_len = 0;
+	}
 	v->rate_limit = vifc->vifc_rate_limit;
 	v->local = vifc->vifc_lcl_addr.s_addr;
 	v->remote = vifc->vifc_rmt_addr.s_addr;
@@ -1848,10 +1859,33 @@ static inline int ipmr_forward_finish(struct net *net, struct sock *sk,
 	return dst_output(net, sk, skb);
 }
 
+#ifdef CONFIG_NET_SWITCHDEV
+static bool ipmr_forward_offloaded(struct sk_buff *skb, struct mr_table *mrt,
+				   int in_vifi, int out_vifi)
+{
+	struct vif_device *out_vif = &mrt->vif_table[out_vifi];
+	struct vif_device *in_vif = &mrt->vif_table[in_vifi];
+
+	if (!skb->offload_mr_fwd_mark)
+		return false;
+	if (!out_vif->dev_parent_id.id_len || !in_vif->dev_parent_id.id_len)
+		return false;
+	return netdev_phys_item_id_same(&out_vif->dev_parent_id,
+					&in_vif->dev_parent_id);
+}
+#else
+static bool ipmr_forward_offloaded(struct sk_buff *skb, struct mr_table *mrt,
+				   int in_vifi, int out_vifi)
+{
+	return false;
+}
+#endif
+
 /* Processing handlers for ipmr_forward */
 
 static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt,
-			    struct sk_buff *skb, struct mfc_cache *c, int vifi)
+			    int in_vifi, struct sk_buff *skb,
+			    struct mfc_cache *c, int vifi)
 {
 	const struct iphdr *iph = ip_hdr(skb);
 	struct vif_device *vif = &mrt->vif_table[vifi];
@@ -1872,6 +1906,9 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt,
 		goto out_free;
 	}
 
+	if (ipmr_forward_offloaded(skb, mrt, in_vifi, vifi))
+		goto out_free;
+
 	if (vif->flags & VIFF_TUNNEL) {
 		rt = ip_route_output_ports(net, &fl4, NULL,
 					   vif->remote, vif->local,
@@ -2049,8 +2086,8 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt,
 				struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC);
 
 				if (skb2)
-					ipmr_queue_xmit(net, mrt, skb2, cache,
-							psend);
+					ipmr_queue_xmit(net, mrt, true_vifi,
+							skb2, cache, psend);
 			}
 			psend = ct;
 		}
@@ -2061,9 +2098,10 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt,
 			struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC);
 
 			if (skb2)
-				ipmr_queue_xmit(net, mrt, skb2, cache, psend);
+				ipmr_queue_xmit(net, mrt, true_vifi, skb2,
+						cache, psend);
 		} else {
-			ipmr_queue_xmit(net, mrt, skb, cache, psend);
+			ipmr_queue_xmit(net, mrt, true_vifi, skb, cache, psend);
 			return;
 		}
 	}