Merge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue

Tony Nguyen says:

====================
40GbE Intel Wired LAN Driver Updates 2022-04-12

This series contains updates to i40e and ice drivers.

Joe Damato adds TSO support for MPLS packets on i40e and ice drivers. He
also adds tracking and reporting of tx_stopped statistic for i40e.

Nabil S. Alramli adds reporting of tx_restart to ethtool for i40e.

Mateusz adds new device id support for i40e.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

10 files changed, 88 insertions, 30 deletions

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 55c6bce5da61e9..18558a01935319 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -852,6 +852,7 @@ struct i40e_vsi {
 	u64 tx_busy;
 	u64 tx_linearize;
 	u64 tx_force_wb;
+	u64 tx_stopped;
 	u64 rx_buf_failed;
 	u64 rx_page_failed;
 	u64 rx_page_reuse;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 6aefffd8361545..2819e261a126b6 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -47,6 +47,7 @@ i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 		case I40E_DEV_ID_1G_BASE_T_X722:
 		case I40E_DEV_ID_10G_BASE_T_X722:
 		case I40E_DEV_ID_SFP_I_X722:
+		case I40E_DEV_ID_SFP_X722_A:
 			hw->mac.type = I40E_MAC_X722;
 			break;
 		default:
diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index be7c6f34d45c66..c9dcd6d92c832c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -309,10 +309,11 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid)
 			 tx_ring->stats.bytes,
 			 tx_ring->tx_stats.restart_queue);
 		dev_info(&pf->pdev->dev,
-			 "    tx_rings[%i]: tx_stats: tx_busy = %lld, tx_done_old = %lld\n",
+			 "    tx_rings[%i]: tx_stats: tx_busy = %lld, tx_done_old = %lld, tx_stopped = %lld\n",
 			 i,
 			 tx_ring->tx_stats.tx_busy,
-			 tx_ring->tx_stats.tx_done_old);
+			 tx_ring->tx_stats.tx_done_old,
+			 tx_ring->tx_stats.tx_stopped);
 		dev_info(&pf->pdev->dev,
 			 "    tx_rings[%i]: size = %i\n",
 			 i, tx_ring->size);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_devids.h b/drivers/net/ethernet/intel/i40e/i40e_devids.h
index 1bcb0ec0f0c0eb..2610338002fee7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_devids.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_devids.h
@@ -33,6 +33,7 @@
 #define I40E_DEV_ID_1G_BASE_T_X722	0x37D1
 #define I40E_DEV_ID_10G_BASE_T_X722	0x37D2
 #define I40E_DEV_ID_SFP_I_X722		0x37D3
+#define I40E_DEV_ID_SFP_X722_A		0x0DDA
 
 
 #endif /* _I40E_DEVIDS_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index e48499624d22c6..610f00cbaff985 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -293,12 +293,14 @@ static const struct i40e_stats i40e_gstrings_misc_stats[] = {
 	I40E_VSI_STAT("tx_linearize", tx_linearize),
 	I40E_VSI_STAT("tx_force_wb", tx_force_wb),
 	I40E_VSI_STAT("tx_busy", tx_busy),
+	I40E_VSI_STAT("tx_stopped", tx_stopped),
 	I40E_VSI_STAT("rx_alloc_fail", rx_buf_failed),
 	I40E_VSI_STAT("rx_pg_alloc_fail", rx_page_failed),
 	I40E_VSI_STAT("rx_cache_reuse", rx_page_reuse),
 	I40E_VSI_STAT("rx_cache_alloc", rx_page_alloc),
 	I40E_VSI_STAT("rx_cache_waive", rx_page_waive),
 	I40E_VSI_STAT("rx_cache_busy", rx_page_busy),
+	I40E_VSI_STAT("tx_restart", tx_restart),
 };
 
 /* These PF_STATs might look like duplicates of some NETDEV_STATs,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 6778df2177a114..358c2edc118db8 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -77,6 +77,7 @@ static const struct pci_device_id i40e_pci_tbl[] = {
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_1G_BASE_T_X722), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T_X722), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_I_X722), 0},
+	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_X722_A), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2_A), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_X710_N3000), 0},
@@ -785,6 +786,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
 	unsigned int start;
 	u64 tx_linearize;
 	u64 tx_force_wb;
+	u64 tx_stopped;
 	u64 rx_p, rx_b;
 	u64 tx_p, tx_b;
 	u16 q;
@@ -804,6 +806,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
 	rx_b = rx_p = 0;
 	tx_b = tx_p = 0;
 	tx_restart = tx_busy = tx_linearize = tx_force_wb = 0;
+	tx_stopped = 0;
 	rx_page = 0;
 	rx_buf = 0;
 	rx_reuse = 0;
@@ -828,6 +831,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
 		tx_busy += p->tx_stats.tx_busy;
 		tx_linearize += p->tx_stats.tx_linearize;
 		tx_force_wb += p->tx_stats.tx_force_wb;
+		tx_stopped += p->tx_stats.tx_stopped;
 
 		/* locate Rx ring */
 		p = READ_ONCE(vsi->rx_rings[q]);
@@ -872,6 +876,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
 	vsi->tx_busy = tx_busy;
 	vsi->tx_linearize = tx_linearize;
 	vsi->tx_force_wb = tx_force_wb;
+	vsi->tx_stopped = tx_stopped;
 	vsi->rx_page_failed = rx_page;
 	vsi->rx_buf_failed = rx_buf;
 	vsi->rx_page_reuse = rx_reuse;
@@ -13436,8 +13441,7 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 	np->vsi = vsi;
 
 	hw_enc_features = NETIF_F_SG			|
-			  NETIF_F_IP_CSUM		|
-			  NETIF_F_IPV6_CSUM		|
+			  NETIF_F_HW_CSUM		|
 			  NETIF_F_HIGHDMA		|
 			  NETIF_F_SOFT_FEATURES		|
 			  NETIF_F_TSO			|
@@ -13468,6 +13472,23 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 	/* record features VLANs can make use of */
 	netdev->vlan_features |= hw_enc_features | NETIF_F_TSO_MANGLEID;
 
+#define I40E_GSO_PARTIAL_FEATURES (NETIF_F_GSO_GRE |		\
+				   NETIF_F_GSO_GRE_CSUM |	\
+				   NETIF_F_GSO_IPXIP4 |		\
+				   NETIF_F_GSO_IPXIP6 |		\
+				   NETIF_F_GSO_UDP_TUNNEL |	\
+				   NETIF_F_GSO_UDP_TUNNEL_CSUM)
+
+	netdev->gso_partial_features = I40E_GSO_PARTIAL_FEATURES;
+	netdev->features |= NETIF_F_GSO_PARTIAL |
+			    I40E_GSO_PARTIAL_FEATURES;
+
+	netdev->mpls_features |= NETIF_F_SG;
+	netdev->mpls_features |= NETIF_F_HW_CSUM;
+	netdev->mpls_features |= NETIF_F_TSO;
+	netdev->mpls_features |= NETIF_F_TSO6;
+	netdev->mpls_features |= I40E_GSO_PARTIAL_FEATURES;
+
 	/* enable macvlan offloads */
 	netdev->hw_features |= NETIF_F_HW_L2FW_DOFFLOAD;
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 0eae5858f2fe31..7bc1174edf6b9c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -3,6 +3,7 @@
 
 #include <linux/prefetch.h>
 #include <linux/bpf_trace.h>
+#include <net/mpls.h>
 #include <net/xdp.h>
 #include "i40e.h"
 #include "i40e_trace.h"
@@ -3015,6 +3016,7 @@ static int i40e_tso(struct i40e_tx_buffer *first, u8 *hdr_len,
 {
 	struct sk_buff *skb = first->skb;
 	u64 cd_cmd, cd_tso_len, cd_mss;
+	__be16 protocol;
 	union {
 		struct iphdr *v4;
 		struct ipv6hdr *v6;
@@ -3026,7 +3028,7 @@ static int i40e_tso(struct i40e_tx_buffer *first, u8 *hdr_len,
 		unsigned char *hdr;
 	} l4;
 	u32 paylen, l4_offset;
-	u16 gso_segs, gso_size;
+	u16 gso_size;
 	int err;
 
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
@@ -3039,15 +3041,23 @@ static int i40e_tso(struct i40e_tx_buffer *first, u8 *hdr_len,
 	if (err < 0)
 		return err;
 
-	ip.hdr = skb_network_header(skb);
-	l4.hdr = skb_transport_header(skb);
+	protocol = vlan_get_protocol(skb);
+
+	if (eth_p_mpls(protocol))
+		ip.hdr = skb_inner_network_header(skb);
+	else
+		ip.hdr = skb_network_header(skb);
+	l4.hdr = skb_checksum_start(skb);
 
 	/* initialize outer IP header fields */
 	if (ip.v4->version == 4) {
 		ip.v4->tot_len = 0;
 		ip.v4->check = 0;
+
+		first->tx_flags |= I40E_TX_FLAGS_TSO;
 	} else {
 		ip.v6->payload_len = 0;
+		first->tx_flags |= I40E_TX_FLAGS_TSO;
 	}
 
 	if (skb_shinfo(skb)->gso_type & (SKB_GSO_GRE |
@@ -3100,10 +3110,9 @@ static int i40e_tso(struct i40e_tx_buffer *first, u8 *hdr_len,
 
 	/* pull values out of skb_shinfo */
 	gso_size = skb_shinfo(skb)->gso_size;
-	gso_segs = skb_shinfo(skb)->gso_segs;
 
 	/* update GSO size and bytecount with header size */
-	first->gso_segs = gso_segs;
+	first->gso_segs = skb_shinfo(skb)->gso_segs;
 	first->bytecount += (first->gso_segs - 1) * *hdr_len;
 
 	/* find the field values */
@@ -3187,13 +3196,27 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 	unsigned char *exthdr;
 	u32 offset, cmd = 0;
 	__be16 frag_off;
+	__be16 protocol;
 	u8 l4_proto = 0;
 
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
 		return 0;
 
-	ip.hdr = skb_network_header(skb);
-	l4.hdr = skb_transport_header(skb);
+	protocol = vlan_get_protocol(skb);
+
+	if (eth_p_mpls(protocol))
+		ip.hdr = skb_inner_network_header(skb);
+	else
+		ip.hdr = skb_network_header(skb);
+	l4.hdr = skb_checksum_start(skb);
+
+	/* set the tx_flags to indicate the IP protocol type. this is
+	 * required so that checksum header computation below is accurate.
+	 */
+	if (ip.v4->version == 4)
+		*tx_flags |= I40E_TX_FLAGS_IPV4;
+	else
+		*tx_flags |= I40E_TX_FLAGS_IPV6;
 
 	/* compute outer L2 header size */
 	offset = ((ip.hdr - skb->data) / 2) << I40E_TX_DESC_LENGTH_MACLEN_SHIFT;
@@ -3373,6 +3396,8 @@ int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size)
 	/* Memory barrier before checking head and tail */
 	smp_mb();
 
+	++tx_ring->tx_stats.tx_stopped;
+
 	/* Check again in a case another CPU has just made room available. */
 	if (likely(I40E_DESC_UNUSED(tx_ring) < size))
 		return -EBUSY;
@@ -3749,7 +3774,6 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	struct i40e_tx_buffer *first;
 	u32 td_offset = 0;
 	u32 tx_flags = 0;
-	__be16 protocol;
 	u32 td_cmd = 0;
 	u8 hdr_len = 0;
 	int tso, count;
@@ -3791,15 +3815,6 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb,
 	if (i40e_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags))
 		goto out_drop;
 
-	/* obtain protocol of skb */
-	protocol = vlan_get_protocol(skb);
-
-	/* setup IPv4/IPv6 offloads */
-	if (protocol == htons(ETH_P_IP))
-		tx_flags |= I40E_TX_FLAGS_IPV4;
-	else if (protocol == htons(ETH_P_IPV6))
-		tx_flags |= I40E_TX_FLAGS_IPV6;
-
 	tso = i40e_tso(first, &hdr_len, &cd_type_cmd_tso_mss);
 
 	if (tso < 0)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index c471c2da313ce2..41f86e9535a00b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -290,6 +290,7 @@ struct i40e_tx_queue_stats {
 	u64 tx_done_old;
 	u64 tx_linearize;
 	u64 tx_force_wb;
+	u64 tx_stopped;
 	int prev_pkt_ctr;
 };
 
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index d768925785ca79..fbefcb08fa646b 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -3329,7 +3329,9 @@ static void ice_set_netdev_features(struct net_device *netdev)
 			      vlano_features | tso_features;
 
 	/* add support for HW_CSUM on packets with MPLS header */
-	netdev->mpls_features =  NETIF_F_HW_CSUM;
+	netdev->mpls_features =  NETIF_F_HW_CSUM |
+				 NETIF_F_TSO     |
+				 NETIF_F_TSO6;
 
 	/* enable features */
 	netdev->features |= netdev->hw_features;
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index f9bf008471c9ea..3f8b7274ed2f10 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -8,6 +8,7 @@
 #include <linux/prefetch.h>
 #include <linux/bpf_trace.h>
 #include <net/dsfield.h>
+#include <net/mpls.h>
 #include <net/xdp.h>
 #include "ice_txrx_lib.h"
 #include "ice_lib.h"
@@ -1748,18 +1749,24 @@ int ice_tx_csum(struct ice_tx_buf *first, struct ice_tx_offload_params *off)
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
 		return 0;
 
-	ip.hdr = skb_network_header(skb);
-	l4.hdr = skb_transport_header(skb);
+	protocol = vlan_get_protocol(skb);
+
+	if (eth_p_mpls(protocol))
+		ip.hdr = skb_inner_network_header(skb);
+	else
+		ip.hdr = skb_network_header(skb);
+	l4.hdr = skb_checksum_start(skb);
 
 	/* compute outer L2 header size */
 	l2_len = ip.hdr - skb->data;
 	offset = (l2_len / 2) << ICE_TX_DESC_LEN_MACLEN_S;
 
-	protocol = vlan_get_protocol(skb);
-
-	if (protocol == htons(ETH_P_IP))
+	/* set the tx_flags to indicate the IP protocol type. this is
+	 * required so that checksum header computation below is accurate.
+	 */
+	if (ip.v4->version == 4)
 		first->tx_flags |= ICE_TX_FLAGS_IPV4;
-	else if (protocol == htons(ETH_P_IPV6))
+	else if (ip.v6->version == 6)
 		first->tx_flags |= ICE_TX_FLAGS_IPV6;
 
 	if (skb->encapsulation) {
@@ -1957,6 +1964,7 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off)
 		unsigned char *hdr;
 	} l4;
 	u64 cd_mss, cd_tso_len;
+	__be16 protocol;
 	u32 paylen;
 	u8 l4_start;
 	int err;
@@ -1972,8 +1980,13 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off)
 		return err;
 
 	/* cppcheck-suppress unreadVariable */
-	ip.hdr = skb_network_header(skb);
-	l4.hdr = skb_transport_header(skb);
+	protocol = vlan_get_protocol(skb);
+
+	if (eth_p_mpls(protocol))
+		ip.hdr = skb_inner_network_header(skb);
+	else
+		ip.hdr = skb_network_header(skb);
+	l4.hdr = skb_checksum_start(skb);
 
 	/* initialize outer IP header fields */
 	if (ip.v4->version == 4) {