Merge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue

Tony Nguyen says:

====================
100GbE Intel Wired LAN Driver Updates 2022-03-09

This series contains updates to ice driver only.

Martyna implements switchdev filtering on inner EtherType field for
tunnels.

Marcin adds reporting of slowpath statistics for port representors.

Jonathan Toppins changes a non-fatal link error message from warning to
debug.

Maciej removes unnecessary checks in ice_clean_tx_irq().

Amritha adds support for ADQ to match outer destination MAC for tunnels.

* '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue:
  ice: Add support for outer dest MAC for ADQ tunnels
  ice: avoid XDP checks in ice_clean_tx_irq()
  ice: change "can't set link" message to dbg level
  ice: Add slow path offload stats on port representor in switchdev
  ice: Add support for inner etype in switchdev
====================

Link: https://lore.kernel.org/r/20220309190315.1380414-1-anthony.l.nguyen@intel.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

drivers/net/ethernet/intel/ice/ice.h

@@ -834,6 +834,9 @@ u16 ice_get_avail_rxq_count(struct ice_pf *pf);
 int ice_vsi_recfg_qs(struct ice_vsi *vsi, int new_rx, int new_tx);
 void ice_update_vsi_stats(struct ice_vsi *vsi);
 void ice_update_pf_stats(struct ice_pf *pf);
+void
+ice_fetch_u64_stats_per_ring(struct u64_stats_sync *syncp,
+			     struct ice_q_stats stats, u64 *pkts, u64 *bytes);
 int ice_up(struct ice_vsi *vsi);
 int ice_down(struct ice_vsi *vsi);
 int ice_vsi_cfg(struct ice_vsi *vsi);

drivers/net/ethernet/intel/ice/ice_lib.c

@@ -3960,9 +3960,9 @@ int ice_set_link(struct ice_vsi *vsi, bool ena)
 	 */
 	if (status == -EIO) {
 		if (hw->adminq.sq_last_status == ICE_AQ_RC_EMODE)
-			dev_warn(dev, "can't set link to %s, err %d aq_err %s. not fatal, continuing\n",
-				 (ena ? "ON" : "OFF"), status,
-				 ice_aq_str(hw->adminq.sq_last_status));
+			dev_dbg(dev, "can't set link to %s, err %d aq_err %s. not fatal, continuing\n",
+				(ena ? "ON" : "OFF"), status,
+				ice_aq_str(hw->adminq.sq_last_status));
 	} else if (status) {
 		dev_err(dev, "can't set link to %s, err %d aq_err %s\n",
 			(ena ? "ON" : "OFF"), status,

drivers/net/ethernet/intel/ice/ice_main.c

@@ -6134,9 +6134,9 @@ int ice_up(struct ice_vsi *vsi)
  * This function fetches stats from the ring considering the atomic operations
  * that needs to be performed to read u64 values in 32 bit machine.
  */
-static void
-ice_fetch_u64_stats_per_ring(struct u64_stats_sync *syncp, struct ice_q_stats stats,
-			     u64 *pkts, u64 *bytes)
+void
+ice_fetch_u64_stats_per_ring(struct u64_stats_sync *syncp,
+			     struct ice_q_stats stats, u64 *pkts, u64 *bytes)
 {
 	unsigned int start;
 

drivers/net/ethernet/intel/ice/ice_protocol_type.h

@@ -29,6 +29,7 @@ enum ice_protocol_type {
 	ICE_MAC_OFOS = 0,
 	ICE_MAC_IL,
 	ICE_ETYPE_OL,
+	ICE_ETYPE_IL,
 	ICE_VLAN_OFOS,
 	ICE_IPV4_OFOS,
 	ICE_IPV4_IL,
@@ -92,6 +93,7 @@ enum ice_prot_id {
 #define ICE_MAC_OFOS_HW		1
 #define ICE_MAC_IL_HW		4
 #define ICE_ETYPE_OL_HW		9
+#define ICE_ETYPE_IL_HW		10
 #define ICE_VLAN_OF_HW		16
 #define ICE_VLAN_OL_HW		17
 #define ICE_IPV4_OFOS_HW	32

drivers/net/ethernet/intel/ice/ice_repr.c

@@ -142,6 +142,59 @@ ice_repr_get_devlink_port(struct net_device *netdev)
 	return &repr->vf->devlink_port;
 }
 
+/**
+ * ice_repr_sp_stats64 - get slow path stats for port representor
+ * @dev: network interface device structure
+ * @stats: netlink stats structure
+ *
+ * RX/TX stats are being swapped here to be consistent with VF stats. In slow
+ * path, port representor receives data when the corresponding VF is sending it
+ * (and vice versa), TX and RX bytes/packets are effectively swapped on port
+ * representor.
+ */
+static int
+ice_repr_sp_stats64(const struct net_device *dev,
+		    struct rtnl_link_stats64 *stats)
+{
+	struct ice_netdev_priv *np = netdev_priv(dev);
+	int vf_id = np->repr->vf->vf_id;
+	struct ice_tx_ring *tx_ring;
+	struct ice_rx_ring *rx_ring;
+	u64 pkts, bytes;
+
+	tx_ring = np->vsi->tx_rings[vf_id];
+	ice_fetch_u64_stats_per_ring(&tx_ring->syncp, tx_ring->stats,
+				     &pkts, &bytes);
+	stats->rx_packets = pkts;
+	stats->rx_bytes = bytes;
+
+	rx_ring = np->vsi->rx_rings[vf_id];
+	ice_fetch_u64_stats_per_ring(&rx_ring->syncp, rx_ring->stats,
+				     &pkts, &bytes);
+	stats->tx_packets = pkts;
+	stats->tx_bytes = bytes;
+	stats->tx_dropped = rx_ring->rx_stats.alloc_page_failed +
+			    rx_ring->rx_stats.alloc_buf_failed;
+
+	return 0;
+}
+
+static bool
+ice_repr_ndo_has_offload_stats(const struct net_device *dev, int attr_id)
+{
+	return attr_id == IFLA_OFFLOAD_XSTATS_CPU_HIT;
+}
+
+static int
+ice_repr_ndo_get_offload_stats(int attr_id, const struct net_device *dev,
+			       void *sp)
+{
+	if (attr_id == IFLA_OFFLOAD_XSTATS_CPU_HIT)
+		return ice_repr_sp_stats64(dev, (struct rtnl_link_stats64 *)sp);
+
+	return -EINVAL;
+}
+
 static int
 ice_repr_setup_tc_cls_flower(struct ice_repr *repr,
 			     struct flow_cls_offload *flower)
@@ -199,6 +252,8 @@ static const struct net_device_ops ice_repr_netdev_ops = {
 	.ndo_start_xmit = ice_eswitch_port_start_xmit,
 	.ndo_get_devlink_port = ice_repr_get_devlink_port,
 	.ndo_setup_tc = ice_repr_setup_tc,
+	.ndo_has_offload_stats = ice_repr_ndo_has_offload_stats,
+	.ndo_get_offload_stats = ice_repr_ndo_get_offload_stats,
 };
 
 /**

drivers/net/ethernet/intel/ice/ice_tc_lib.c

@@ -24,6 +24,9 @@ ice_tc_count_lkups(u32 flags, struct ice_tc_flower_lyr_2_4_hdrs *headers,
 	if (flags & ICE_TC_FLWR_FIELD_TENANT_ID)
 		lkups_cnt++;
 
+	if (flags & ICE_TC_FLWR_FIELD_ENC_DST_MAC)
+		lkups_cnt++;
+
 	if (flags & (ICE_TC_FLWR_FIELD_ENC_SRC_IPV4 |
 		     ICE_TC_FLWR_FIELD_ENC_DEST_IPV4 |
 		     ICE_TC_FLWR_FIELD_ENC_SRC_IPV6 |
@@ -33,9 +36,7 @@ ice_tc_count_lkups(u32 flags, struct ice_tc_flower_lyr_2_4_hdrs *headers,
 	if (flags & ICE_TC_FLWR_FIELD_ENC_DEST_L4_PORT)
 		lkups_cnt++;
 
-	/* currently inner etype filter isn't supported */
-	if ((flags & ICE_TC_FLWR_FIELD_ETH_TYPE_ID) &&
-	    fltr->tunnel_type == TNL_LAST)
+	if (flags & ICE_TC_FLWR_FIELD_ETH_TYPE_ID)
 		lkups_cnt++;
 
 	/* are MAC fields specified? */
@@ -64,6 +65,11 @@ static enum ice_protocol_type ice_proto_type_from_mac(bool inner)
 	return inner ? ICE_MAC_IL : ICE_MAC_OFOS;
 }
 
+static enum ice_protocol_type ice_proto_type_from_etype(bool inner)
+{
+	return inner ? ICE_ETYPE_IL : ICE_ETYPE_OL;
+}
+
 static enum ice_protocol_type ice_proto_type_from_ipv4(bool inner)
 {
 	return inner ? ICE_IPV4_IL : ICE_IPV4_OFOS;
@@ -145,6 +151,15 @@ ice_tc_fill_tunnel_outer(u32 flags, struct ice_tc_flower_fltr *fltr,
 		}
 	}
 
+	if (flags & ICE_TC_FLWR_FIELD_ENC_DST_MAC) {
+		list[i].type = ice_proto_type_from_mac(false);
+		ether_addr_copy(list[i].h_u.eth_hdr.dst_addr,
+				hdr->l2_key.dst_mac);
+		ether_addr_copy(list[i].m_u.eth_hdr.dst_addr,
+				hdr->l2_mask.dst_mac);
+		i++;
+	}
+
 	if (flags & (ICE_TC_FLWR_FIELD_ENC_SRC_IPV4 |
 		     ICE_TC_FLWR_FIELD_ENC_DEST_IPV4)) {
 		list[i].type = ice_proto_type_from_ipv4(false);
@@ -224,8 +239,10 @@ ice_tc_fill_rules(struct ice_hw *hw, u32 flags,
 
 		headers = &tc_fltr->inner_headers;
 		inner = true;
-	} else if (flags & ICE_TC_FLWR_FIELD_ETH_TYPE_ID) {
-		list[i].type = ICE_ETYPE_OL;
+	}
+
+	if (flags & ICE_TC_FLWR_FIELD_ETH_TYPE_ID) {
+		list[i].type = ice_proto_type_from_etype(inner);
 		list[i].h_u.ethertype.ethtype_id = headers->l2_key.n_proto;
 		list[i].m_u.ethertype.ethtype_id = headers->l2_mask.n_proto;
 		i++;
@@ -1059,12 +1076,24 @@ ice_handle_tclass_action(struct ice_vsi *vsi,
 	 * this code won't do anything
 	 * 2. For non-tunnel, if user didn't specify MAC address, add implicit
 	 * dest MAC to be lower netdev's active unicast MAC address
+	 * 3. For tunnel,  as of now TC-filter through flower classifier doesn't
+	 * have provision for user to specify outer DMAC, hence driver to
+	 * implicitly add outer dest MAC to be lower netdev's active unicast
+	 * MAC address.
 	 */
-	if (!(fltr->flags & ICE_TC_FLWR_FIELD_DST_MAC)) {
-		ether_addr_copy(fltr->outer_headers.l2_key.dst_mac,
-				main_vsi->netdev->dev_addr);
-		eth_broadcast_addr(fltr->outer_headers.l2_mask.dst_mac);
+	if (fltr->tunnel_type != TNL_LAST &&
+	    !(fltr->flags & ICE_TC_FLWR_FIELD_ENC_DST_MAC))
+		fltr->flags |= ICE_TC_FLWR_FIELD_ENC_DST_MAC;
+
+	if (fltr->tunnel_type == TNL_LAST &&
+	    !(fltr->flags & ICE_TC_FLWR_FIELD_DST_MAC))
 		fltr->flags |= ICE_TC_FLWR_FIELD_DST_MAC;
+
+	if (fltr->flags & (ICE_TC_FLWR_FIELD_DST_MAC |
+			   ICE_TC_FLWR_FIELD_ENC_DST_MAC)) {
+		ether_addr_copy(fltr->outer_headers.l2_key.dst_mac,
+				vsi->netdev->dev_addr);
+		memset(fltr->outer_headers.l2_mask.dst_mac, 0xff, ETH_ALEN);
 	}
 
 	/* validate specified dest MAC address, make sure either it belongs to

drivers/net/ethernet/intel/ice/ice_txrx.c

@@ -223,8 +223,7 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget)
 	struct ice_tx_buf *tx_buf;
 
 	/* get the bql data ready */
-	if (!ice_ring_is_xdp(tx_ring))
-		netdev_txq_bql_complete_prefetchw(txring_txq(tx_ring));
+	netdev_txq_bql_complete_prefetchw(txring_txq(tx_ring));
 
 	tx_buf = &tx_ring->tx_buf[i];
 	tx_desc = ICE_TX_DESC(tx_ring, i);
@@ -313,10 +312,6 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget)
 	tx_ring->next_to_clean = i;
 
 	ice_update_tx_ring_stats(tx_ring, total_pkts, total_bytes);
-
-	if (ice_ring_is_xdp(tx_ring))
-		return !!budget;
-
 	netdev_tx_completed_queue(txring_txq(tx_ring), total_pkts, total_bytes);
 
 #define TX_WAKE_THRESHOLD ((s16)(DESC_NEEDED * 2))