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

Jeff Kirsher says:

====================
100GbE Intel Wired LAN Driver Updates 2019-11-04

This series contains updates to the ice driver only.

Anirudh refactors the code to reduce the kernel configuration flags and
introduces ice_base.c file.

Maciej does additional refactoring on the configuring of transmit
rings so that we are not configuring per each traffic class flow.
Added support for XDP in the ice driver.  Provides additional
re-organizing of the code in preparation for adding build_skb() support
in the driver.  Adjusted the computational padding logic for headroom
and tailroom to better support build_skb(), which also aligns with the
logic in other Intel LAN drivers.  Added build_skb support and make use
of the XDP's data_meta.

Krzysztof refactors the driver to prepare for AF_XDP support in the
driver and then adds support for AF_XDP.

v2: Updated patch 3 of the series based on community feedback with the
    following changes...
    - return -EOPNOTSUPP instead of ENOTSUPP for too large MTU which makes
      it impossible to attach XDP prog
    - don't check for case when there's no XDP prog currently on interface
      and ice_xdp() is called with NULL bpf_prog; this happens when user
      does "ip link set eth0 xdp off" and no prog is present on VSI; no need
      for that as it is handled by higher layer
    - drop the extack message for unknown xdp->command
    - use the smp_processor_id() for accessing the XDP Tx ring for XDP_TX
      action
    - don't leave the interface in downed state in case of any failure
      during the XDP Tx resources handling
    - undo rename of ice_build_ctob
    The above changes caused a ripple effect in patches 4 & 5 to update
    references to ice_build_ctob() which are now build_ctob()
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/intel/ice/Makefile

@@ -13,9 +13,12 @@ ice-y := ice_main.o	\
 	 ice_nvm.o	\
 	 ice_switch.o	\
 	 ice_sched.o	\
+	 ice_base.o	\
 	 ice_lib.o	\
+	 ice_txrx_lib.o	\
 	 ice_txrx.o	\
 	 ice_flex_pipe.o	\
 	 ice_ethtool.o
 ice-$(CONFIG_PCI_IOV) += ice_virtchnl_pf.o ice_sriov.o
 ice-$(CONFIG_DCB) += ice_dcb.o ice_dcb_lib.o
+ice-$(CONFIG_XDP_SOCKETS) += ice_xsk.o

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

@@ -29,10 +29,13 @@
 #include <linux/ip.h>
 #include <linux/sctp.h>
 #include <linux/ipv6.h>
+#include <linux/pkt_sched.h>
 #include <linux/if_bridge.h>
 #include <linux/ctype.h>
+#include <linux/bpf.h>
 #include <linux/avf/virtchnl.h>
 #include <net/ipv6.h>
+#include <net/xdp_sock.h>
 #include "ice_devids.h"
 #include "ice_type.h"
 #include "ice_txrx.h"
@@ -42,6 +45,7 @@
 #include "ice_sched.h"
 #include "ice_virtchnl_pf.h"
 #include "ice_sriov.h"
+#include "ice_xsk.h"
 
 extern const char ice_drv_ver[];
 #define ICE_BAR0		0
@@ -78,8 +82,7 @@ extern const char ice_drv_ver[];
 
 #define ICE_DFLT_NETIF_M (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK)
 
-#define ICE_MAX_MTU	(ICE_AQ_SET_MAC_FRAME_SIZE_MAX - \
-			(ETH_HLEN + ETH_FCS_LEN + (VLAN_HLEN * 2)))
+#define ICE_MAX_MTU	(ICE_AQ_SET_MAC_FRAME_SIZE_MAX - ICE_ETH_PKT_HDR_PAD)
 
 #define ICE_UP_TABLE_TRANSLATE(val, i) \
 		(((val) << ICE_AQ_VSI_UP_TABLE_UP##i##_S) & \
@@ -127,6 +130,14 @@ extern const char ice_drv_ver[];
 				     ICE_PROMISC_VLAN_TX  | \
 				     ICE_PROMISC_VLAN_RX)
 
+struct ice_txq_meta {
+	u32 q_teid;	/* Tx-scheduler element identifier */
+	u16 q_id;	/* Entry in VSI's txq_map bitmap */
+	u16 q_handle;	/* Relative index of Tx queue within TC */
+	u16 vsi_idx;	/* VSI index that Tx queue belongs to */
+	u8 tc;		/* TC number that Tx queue belongs to */
+};
+
 struct ice_tc_info {
 	u16 qoffset;
 	u16 qcount_tx;
@@ -274,6 +285,13 @@ struct ice_vsi {
 	u16 num_rx_desc;
 	u16 num_tx_desc;
 	struct ice_tc_cfg tc_cfg;
+	struct bpf_prog *xdp_prog;
+	struct ice_ring **xdp_rings;	 /* XDP ring array */
+	u16 num_xdp_txq;		 /* Used XDP queues */
+	u8 xdp_mapping_mode;		 /* ICE_MAP_MODE_[CONTIG|SCATTER] */
+	struct xdp_umem **xsk_umems;
+	u16 num_xsk_umems_used;
+	u16 num_xsk_umems;
 } ____cacheline_internodealigned_in_smp;
 
 /* struct that defines an interrupt vector */
@@ -313,6 +331,7 @@ enum ice_pf_flags {
 	ICE_FLAG_NO_MEDIA,
 	ICE_FLAG_FW_LLDP_AGENT,
 	ICE_FLAG_ETHTOOL_CTXT,		/* set when ethtool holds RTNL lock */
+	ICE_FLAG_LEGACY_RX,
 	ICE_PF_FLAGS_NBITS		/* must be last */
 };
 
@@ -417,6 +436,37 @@ static inline struct ice_pf *ice_netdev_to_pf(struct net_device *netdev)
 	return np->vsi->back;
 }
 
+static inline bool ice_is_xdp_ena_vsi(struct ice_vsi *vsi)
+{
+	return !!vsi->xdp_prog;
+}
+
+static inline void ice_set_ring_xdp(struct ice_ring *ring)
+{
+	ring->flags |= ICE_TX_FLAGS_RING_XDP;
+}
+
+/**
+ * ice_xsk_umem - get XDP UMEM bound to a ring
+ * @ring - ring to use
+ *
+ * Returns a pointer to xdp_umem structure if there is an UMEM present,
+ * NULL otherwise.
+ */
+static inline struct xdp_umem *ice_xsk_umem(struct ice_ring *ring)
+{
+	struct xdp_umem **umems = ring->vsi->xsk_umems;
+	int qid = ring->q_index;
+
+	if (ice_ring_is_xdp(ring))
+		qid -= ring->vsi->num_xdp_txq;
+
+	if (!umems || !umems[qid] || !ice_is_xdp_ena_vsi(ring->vsi))
+		return NULL;
+
+	return umems[qid];
+}
+
 /**
  * ice_get_main_vsi - Get the PF VSI
  * @pf: PF instance
@@ -443,6 +493,11 @@ int ice_up(struct ice_vsi *vsi);
 int ice_down(struct ice_vsi *vsi);
 int ice_vsi_cfg(struct ice_vsi *vsi);
 struct ice_vsi *ice_lb_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi);
+int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog);
+int ice_destroy_xdp_rings(struct ice_vsi *vsi);
+int
+ice_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
+	     u32 flags);
 int ice_set_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size);
 int ice_get_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size);
 void ice_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size);

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

+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2019, Intel Corporation. */
+
+#ifndef _ICE_BASE_H_
+#define _ICE_BASE_H_
+
+#include "ice.h"
+
+int ice_setup_rx_ctx(struct ice_ring *ring);
+int __ice_vsi_get_qs(struct ice_qs_cfg *qs_cfg);
+int ice_vsi_ctrl_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx);
+int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi);
+void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi);
+void ice_vsi_free_q_vectors(struct ice_vsi *vsi);
+int
+ice_vsi_cfg_txq(struct ice_vsi *vsi, struct ice_ring *ring,
+		struct ice_aqc_add_tx_qgrp *qg_buf);
+void ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector);
+void
+ice_cfg_txq_interrupt(struct ice_vsi *vsi, u16 txq, u16 msix_idx, u16 itr_idx);
+void
+ice_cfg_rxq_interrupt(struct ice_vsi *vsi, u16 rxq, u16 msix_idx, u16 itr_idx);
+void ice_trigger_sw_intr(struct ice_hw *hw, struct ice_q_vector *q_vector);
+int
+ice_vsi_stop_tx_ring(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src,
+		     u16 rel_vmvf_num, struct ice_ring *ring,
+		     struct ice_txq_meta *txq_meta);
+void
+ice_fill_txq_meta(struct ice_vsi *vsi, struct ice_ring *ring,
+		  struct ice_txq_meta *txq_meta);
+#endif /* _ICE_BASE_H_ */

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

@@ -5,6 +5,7 @@
 #define _ICE_DCB_LIB_H_
 
 #include "ice.h"
+#include "ice_base.h"
 #include "ice_lib.h"
 
 #ifdef CONFIG_DCB

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

@@ -156,6 +156,7 @@ struct ice_priv_flag {
 static const struct ice_priv_flag ice_gstrings_priv_flags[] = {
 	ICE_PRIV_FLAG("link-down-on-close", ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA),
 	ICE_PRIV_FLAG("fw-lldp-agent", ICE_FLAG_FW_LLDP_AGENT),
+	ICE_PRIV_FLAG("legacy-rx", ICE_FLAG_LEGACY_RX),
 };
 
 #define ICE_PRIV_FLAG_ARRAY_SIZE	ARRAY_SIZE(ice_gstrings_priv_flags)
@@ -623,7 +624,7 @@ static int ice_lbtest_receive_frames(struct ice_ring *rx_ring)
 			continue;
 
 		rx_buf = &rx_ring->rx_buf[i];
-		received_buf = page_address(rx_buf->page);
+		received_buf = page_address(rx_buf->page) + rx_buf->page_offset;
 
 		if (ice_lbtest_check_frame(received_buf))
 			valid_frames++;
@@ -1256,6 +1257,11 @@ static int ice_set_priv_flags(struct net_device *netdev, u32 flags)
 					"Fail to enable MIB change events\n");
 		}
 	}
+	if (test_bit(ICE_FLAG_LEGACY_RX, change_flags)) {
+		/* down and up VSI so that changes of Rx cfg are reflected. */
+		ice_down(vsi);
+		ice_up(vsi);
+	}
 	clear_bit(ICE_FLAG_ETHTOOL_CTXT, pf->flags);
 	return ret;
 }
@@ -2577,6 +2583,7 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring)
 {
 	struct ice_ring *tx_rings = NULL, *rx_rings = NULL;
 	struct ice_netdev_priv *np = netdev_priv(netdev);
+	struct ice_ring *xdp_rings = NULL;
 	struct ice_vsi *vsi = np->vsi;
 	struct ice_pf *pf = vsi->back;
 	int i, timeout = 50, err = 0;
@@ -2611,6 +2618,13 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring)
 		return 0;
 	}
 
+	/* If there is a AF_XDP UMEM attached to any of Rx rings,
+	 * disallow changing the number of descriptors -- regardless
+	 * if the netdev is running or not.
+	 */
+	if (ice_xsk_any_rx_ring_ena(vsi))
+		return -EBUSY;
+
 	while (test_and_set_bit(__ICE_CFG_BUSY, pf->state)) {
 		timeout--;
 		if (!timeout)
@@ -2624,6 +2638,11 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring)
 			vsi->tx_rings[i]->count = new_tx_cnt;
 		for (i = 0; i < vsi->alloc_rxq; i++)
 			vsi->rx_rings[i]->count = new_rx_cnt;
+		if (ice_is_xdp_ena_vsi(vsi))
+			for (i = 0; i < vsi->num_xdp_txq; i++)
+				vsi->xdp_rings[i]->count = new_tx_cnt;
+		vsi->num_tx_desc = new_tx_cnt;
+		vsi->num_rx_desc = new_rx_cnt;
 		netdev_dbg(netdev, "Link is down, descriptor count change happens when link is brought up\n");
 		goto done;
 	}
@@ -2650,15 +2669,43 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring)
 		tx_rings[i].tx_buf = NULL;
 		err = ice_setup_tx_ring(&tx_rings[i]);
 		if (err) {
-			while (i) {
-				i--;
+			while (i--)
 				ice_clean_tx_ring(&tx_rings[i]);
-			}
 			devm_kfree(&pf->pdev->dev, tx_rings);
 			goto done;
 		}
 	}
 
+	if (!ice_is_xdp_ena_vsi(vsi))
+		goto process_rx;
+
+	/* alloc updated XDP resources */
+	netdev_info(netdev, "Changing XDP descriptor count from %d to %d\n",
+		    vsi->xdp_rings[0]->count, new_tx_cnt);
+
+	xdp_rings = devm_kcalloc(&pf->pdev->dev, vsi->num_xdp_txq,
+				 sizeof(*xdp_rings), GFP_KERNEL);
+	if (!xdp_rings) {
+		err = -ENOMEM;
+		goto free_tx;
+	}
+
+	for (i = 0; i < vsi->num_xdp_txq; i++) {
+		/* clone ring and setup updated count */
+		xdp_rings[i] = *vsi->xdp_rings[i];
+		xdp_rings[i].count = new_tx_cnt;
+		xdp_rings[i].desc = NULL;
+		xdp_rings[i].tx_buf = NULL;
+		err = ice_setup_tx_ring(&xdp_rings[i]);
+		if (err) {
+			while (i--)
+				ice_clean_tx_ring(&xdp_rings[i]);
+			devm_kfree(&pf->pdev->dev, xdp_rings);
+			goto free_tx;
+		}
+		ice_set_ring_xdp(&xdp_rings[i]);
+	}
+
 process_rx:
 	if (new_rx_cnt == vsi->rx_rings[0]->count)
 		goto process_link;
@@ -2737,6 +2784,16 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring)
 			devm_kfree(&pf->pdev->dev, rx_rings);
 		}
 
+		if (xdp_rings) {
+			for (i = 0; i < vsi->num_xdp_txq; i++) {
+				ice_free_tx_ring(vsi->xdp_rings[i]);
+				*vsi->xdp_rings[i] = xdp_rings[i];
+			}
+			devm_kfree(&pf->pdev->dev, xdp_rings);
+		}
+
+		vsi->num_tx_desc = new_tx_cnt;
+		vsi->num_rx_desc = new_rx_cnt;
 		ice_up(vsi);
 	}
 	goto done;

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

@@ -6,19 +6,6 @@
 
 #include "ice.h"
 
-struct ice_txq_meta {
-	/* Tx-scheduler element identifier */
-	u32 q_teid;
-	/* Entry in VSI's txq_map bitmap */
-	u16 q_id;
-	/* Relative index of Tx queue within TC */
-	u16 q_handle;
-	/* VSI index that Tx queue belongs to */
-	u16 vsi_idx;
-	/* TC number that Tx queue belongs to */
-	u8 tc;
-};
-
 int
 ice_add_mac_to_list(struct ice_vsi *vsi, struct list_head *add_list,
 		    const u8 *macaddr);
@@ -33,24 +20,6 @@ int ice_vsi_cfg_lan_txqs(struct ice_vsi *vsi);
 
 void ice_vsi_cfg_msix(struct ice_vsi *vsi);
 
-#ifdef CONFIG_PCI_IOV
-void
-ice_cfg_txq_interrupt(struct ice_vsi *vsi, u16 txq, u16 msix_idx, u16 itr_idx);
-
-void
-ice_cfg_rxq_interrupt(struct ice_vsi *vsi, u16 rxq, u16 msix_idx, u16 itr_idx);
-
-int
-ice_vsi_stop_tx_ring(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src,
-		     u16 rel_vmvf_num, struct ice_ring *ring,
-		     struct ice_txq_meta *txq_meta);
-
-void ice_fill_txq_meta(struct ice_vsi *vsi, struct ice_ring *ring,
-		       struct ice_txq_meta *txq_meta);
-
-int ice_vsi_ctrl_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx);
-#endif /* CONFIG_PCI_IOV */
-
 int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid);
 
 int ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid);
@@ -67,6 +36,10 @@ int
 ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src,
 			  u16 rel_vmvf_num);
 
+int ice_vsi_cfg_xdp_txqs(struct ice_vsi *vsi);
+
+int ice_vsi_stop_xdp_tx_rings(struct ice_vsi *vsi);
+
 int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc);
 
 void ice_cfg_sw_lldp(struct ice_vsi *vsi, bool tx, bool create);
@@ -98,16 +71,8 @@ int ice_vsi_rebuild(struct ice_vsi *vsi);
 
 bool ice_is_reset_in_progress(unsigned long *state);
 
-void ice_vsi_free_q_vectors(struct ice_vsi *vsi);
-
-void ice_trigger_sw_intr(struct ice_hw *hw, struct ice_q_vector *q_vector);
-
 void ice_vsi_put_qs(struct ice_vsi *vsi);
 
-#ifdef CONFIG_DCB
-void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi);
-#endif /* CONFIG_DCB */
-
 void ice_vsi_dis_irq(struct ice_vsi *vsi);
 
 void ice_vsi_free_irq(struct ice_vsi *vsi);
@@ -118,6 +83,12 @@ void ice_vsi_free_tx_rings(struct ice_vsi *vsi);
 
 int ice_vsi_manage_rss_lut(struct ice_vsi *vsi, bool ena);
 
+void ice_update_tx_ring_stats(struct ice_ring *ring, u64 pkts, u64 bytes);
+
+void ice_update_rx_ring_stats(struct ice_ring *ring, u64 pkts, u64 bytes);
+
+void ice_vsi_cfg_frame_size(struct ice_vsi *vsi);
+
 u32 ice_intrl_usec_to_reg(u8 intrl, u8 gran);
 
 char *ice_nvm_version_str(struct ice_hw *hw);

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

@@ -4,8 +4,12 @@
 #ifndef _ICE_TXRX_H_
 #define _ICE_TXRX_H_
 
+#include "ice_type.h"
+
 #define ICE_DFLT_IRQ_WORK	256
+#define ICE_RXBUF_3072		3072
 #define ICE_RXBUF_2048		2048
+#define ICE_RXBUF_1536		1536
 #define ICE_MAX_CHAINED_RX_BUFS	5
 #define ICE_MAX_BUF_TXD		8
 #define ICE_MIN_TX_LEN		17
@@ -22,6 +26,71 @@
 #define ICE_RX_BUF_WRITE	16	/* Must be power of 2 */
 #define ICE_MAX_TXQ_PER_TXQG	128
 
+/* Attempt to maximize the headroom available for incoming frames. We use a 2K
+ * buffer for MTUs <= 1500 and need 1536/1534 to store the data for the frame.
+ * This leaves us with 512 bytes of room.  From that we need to deduct the
+ * space needed for the shared info and the padding needed to IP align the
+ * frame.
+ *
+ * Note: For cache line sizes 256 or larger this value is going to end
+ *       up negative.  In these cases we should fall back to the legacy
+ *       receive path.
+ */
+#if (PAGE_SIZE < 8192)
+#define ICE_2K_TOO_SMALL_WITH_PADDING \
+((NET_SKB_PAD + ICE_RXBUF_1536) > SKB_WITH_OVERHEAD(ICE_RXBUF_2048))
+
+/**
+ * ice_compute_pad - compute the padding
+ * rx_buf_len: buffer length
+ *
+ * Figure out the size of half page based on given buffer length and
+ * then subtract the skb_shared_info followed by subtraction of the
+ * actual buffer length; this in turn results in the actual space that
+ * is left for padding usage
+ */
+static inline int ice_compute_pad(int rx_buf_len)
+{
+	int half_page_size;
+
+	half_page_size = ALIGN(rx_buf_len, PAGE_SIZE / 2);
+	return SKB_WITH_OVERHEAD(half_page_size) - rx_buf_len;
+}
+
+/**
+ * ice_skb_pad - determine the padding that we can supply
+ *
+ * Figure out the right Rx buffer size and based on that calculate the
+ * padding
+ */
+static inline int ice_skb_pad(void)
+{
+	int rx_buf_len;
+
+	/* If a 2K buffer cannot handle a standard Ethernet frame then
+	 * optimize padding for a 3K buffer instead of a 1.5K buffer.
+	 *
+	 * For a 3K buffer we need to add enough padding to allow for
+	 * tailroom due to NET_IP_ALIGN possibly shifting us out of
+	 * cache-line alignment.
+	 */
+	if (ICE_2K_TOO_SMALL_WITH_PADDING)
+		rx_buf_len = ICE_RXBUF_3072 + SKB_DATA_ALIGN(NET_IP_ALIGN);
+	else
+		rx_buf_len = ICE_RXBUF_1536;
+
+	/* if needed make room for NET_IP_ALIGN */
+	rx_buf_len -= NET_IP_ALIGN;
+
+	return ice_compute_pad(rx_buf_len);
+}
+
+#define ICE_SKB_PAD ice_skb_pad()
+#else
+#define ICE_2K_TOO_SMALL_WITH_PADDING false
+#define ICE_SKB_PAD (NET_SKB_PAD + NET_IP_ALIGN)
+#endif
+
 /* We are assuming that the cache line is always 64 Bytes here for ice.
  * In order to make sure that is a correct assumption there is a check in probe
  * to print a warning if the read from GLPCI_CNF2 tells us that the cache line
@@ -49,12 +118,24 @@
 #define ICE_TX_FLAGS_VLAN_PR_S	29
 #define ICE_TX_FLAGS_VLAN_S	16
 
+#define ICE_XDP_PASS		0
+#define ICE_XDP_CONSUMED	BIT(0)
+#define ICE_XDP_TX		BIT(1)
+#define ICE_XDP_REDIR		BIT(2)
+
 #define ICE_RX_DMA_ATTR \
 	(DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)
 
+#define ICE_ETH_PKT_HDR_PAD	(ETH_HLEN + ETH_FCS_LEN + (VLAN_HLEN * 2))
+
+#define ICE_TXD_LAST_DESC_CMD (ICE_TX_DESC_CMD_EOP | ICE_TX_DESC_CMD_RS)
+
 struct ice_tx_buf {
 	struct ice_tx_desc *next_to_watch;
-	struct sk_buff *skb;
+	union {
+		struct sk_buff *skb;
+		void *raw_buf; /* used for XDP */
+	};
 	unsigned int bytecount;
 	unsigned short gso_segs;
 	u32 tx_flags;
@@ -76,9 +157,17 @@ struct ice_tx_offload_params {
 struct ice_rx_buf {
 	struct sk_buff *skb;
 	dma_addr_t dma;
-	struct page *page;
-	unsigned int page_offset;
-	u16 pagecnt_bias;
+	union {
+		struct {
+			struct page *page;
+			unsigned int page_offset;
+			u16 pagecnt_bias;
+		};
+		struct {
+			void *addr;
+			u64 handle;
+		};
+	};
 };
 
 struct ice_q_stats {
@@ -198,18 +287,44 @@ struct ice_ring {
 	};
 
 	struct rcu_head rcu;		/* to avoid race on free */
+	struct bpf_prog *xdp_prog;
+	struct xdp_umem *xsk_umem;
+	struct zero_copy_allocator zca;
+	/* CL3 - 3rd cacheline starts here */
+	struct xdp_rxq_info xdp_rxq;
 	/* CLX - the below items are only accessed infrequently and should be
 	 * in their own cache line if possible
 	 */
+#define ICE_TX_FLAGS_RING_XDP		BIT(0)
+#define ICE_RX_FLAGS_RING_BUILD_SKB	BIT(1)
+	u8 flags;
 	dma_addr_t dma;			/* physical address of ring */
 	unsigned int size;		/* length of descriptor ring in bytes */
 	u32 txq_teid;			/* Added Tx queue TEID */
 	u16 rx_buf_len;
-#ifdef CONFIG_DCB
 	u8 dcb_tc;			/* Traffic class of ring */
-#endif /* CONFIG_DCB */
 } ____cacheline_internodealigned_in_smp;
 
+static inline bool ice_ring_uses_build_skb(struct ice_ring *ring)
+{
+	return !!(ring->flags & ICE_RX_FLAGS_RING_BUILD_SKB);
+}
+
+static inline void ice_set_ring_build_skb_ena(struct ice_ring *ring)
+{
+	ring->flags |= ICE_RX_FLAGS_RING_BUILD_SKB;
+}
+
+static inline void ice_clear_ring_build_skb_ena(struct ice_ring *ring)
+{
+	ring->flags &= ~ICE_RX_FLAGS_RING_BUILD_SKB;
+}
+
+static inline bool ice_ring_is_xdp(struct ice_ring *ring)
+{
+	return !!(ring->flags & ICE_TX_FLAGS_RING_XDP);
+}
+
 struct ice_ring_container {
 	/* head of linked-list of rings */
 	struct ice_ring *ring;
@@ -230,6 +345,19 @@ struct ice_ring_container {
 #define ice_for_each_ring(pos, head) \
 	for (pos = (head).ring; pos; pos = pos->next)
 
+static inline unsigned int ice_rx_pg_order(struct ice_ring *ring)
+{
+#if (PAGE_SIZE < 8192)
+	if (ring->rx_buf_len > (PAGE_SIZE / 2))
+		return 1;
+#endif
+	return 0;
+}
+
+#define ice_rx_pg_size(_ring) (PAGE_SIZE << ice_rx_pg_order(_ring))
+
+union ice_32b_rx_flex_desc;
+
 bool ice_alloc_rx_bufs(struct ice_ring *rxr, u16 cleaned_count);
 netdev_tx_t ice_start_xmit(struct sk_buff *skb, struct net_device *netdev);
 void ice_clean_tx_ring(struct ice_ring *tx_ring);

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

+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019, Intel Corporation. */
+
+#include "ice_txrx_lib.h"
+
+/**
+ * ice_release_rx_desc - Store the new tail and head values
+ * @rx_ring: ring to bump
+ * @val: new head index
+ */
+void ice_release_rx_desc(struct ice_ring *rx_ring, u32 val)
+{
+	u16 prev_ntu = rx_ring->next_to_use;
+
+	rx_ring->next_to_use = val;
+
+	/* update next to alloc since we have filled the ring */
+	rx_ring->next_to_alloc = val;
+
+	/* QRX_TAIL will be updated with any tail value, but hardware ignores
+	 * the lower 3 bits. This makes it so we only bump tail on meaningful
+	 * boundaries. Also, this allows us to bump tail on intervals of 8 up to
+	 * the budget depending on the current traffic load.
+	 */
+	val &= ~0x7;
+	if (prev_ntu != val) {
+		/* Force memory writes to complete before letting h/w
+		 * know there are new descriptors to fetch. (Only
+		 * applicable for weak-ordered memory model archs,
+		 * such as IA-64).
+		 */
+		wmb();
+		writel(val, rx_ring->tail);
+	}
+}
+
+/**
+ * ice_ptype_to_htype - get a hash type
+ * @ptype: the ptype value from the descriptor
+ *
+ * Returns a hash type to be used by skb_set_hash
+ */
+static enum pkt_hash_types ice_ptype_to_htype(u8 __always_unused ptype)
+{
+	return PKT_HASH_TYPE_NONE;
+}
+
+/**
+ * ice_rx_hash - set the hash value in the skb
+ * @rx_ring: descriptor ring
+ * @rx_desc: specific descriptor
+ * @skb: pointer to current skb
+ * @rx_ptype: the ptype value from the descriptor
+ */
+static void
+ice_rx_hash(struct ice_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc,
+	    struct sk_buff *skb, u8 rx_ptype)
+{
+	struct ice_32b_rx_flex_desc_nic *nic_mdid;
+	u32 hash;
+
+	if (!(rx_ring->netdev->features & NETIF_F_RXHASH))
+		return;
+
+	if (rx_desc->wb.rxdid != ICE_RXDID_FLEX_NIC)
+		return;
+
+	nic_mdid = (struct ice_32b_rx_flex_desc_nic *)rx_desc;
+	hash = le32_to_cpu(nic_mdid->rss_hash);
+	skb_set_hash(skb, hash, ice_ptype_to_htype(rx_ptype));
+}
+
+/**
+ * ice_rx_csum - Indicate in skb if checksum is good
+ * @ring: the ring we care about
+ * @skb: skb currently being received and modified
+ * @rx_desc: the receive descriptor
+ * @ptype: the packet type decoded by hardware
+ *
+ * skb->protocol must be set before this function is called
+ */
+static void
+ice_rx_csum(struct ice_ring *ring, struct sk_buff *skb,
+	    union ice_32b_rx_flex_desc *rx_desc, u8 ptype)
+{
+	struct ice_rx_ptype_decoded decoded;
+	u32 rx_error, rx_status;
+	bool ipv4, ipv6;
+
+	rx_status = le16_to_cpu(rx_desc->wb.status_error0);
+	rx_error = rx_status;
+
+	decoded = ice_decode_rx_desc_ptype(ptype);
+
+	/* Start with CHECKSUM_NONE and by default csum_level = 0 */
+	skb->ip_summed = CHECKSUM_NONE;
+	skb_checksum_none_assert(skb);
+
+	/* check if Rx checksum is enabled */
+	if (!(ring->netdev->features & NETIF_F_RXCSUM))
+		return;
+
+	/* check if HW has decoded the packet and checksum */
+	if (!(rx_status & BIT(ICE_RX_FLEX_DESC_STATUS0_L3L4P_S)))
+		return;
+
+	if (!(decoded.known && decoded.outer_ip))
+		return;
+
+	ipv4 = (decoded.outer_ip == ICE_RX_PTYPE_OUTER_IP) &&
+	       (decoded.outer_ip_ver == ICE_RX_PTYPE_OUTER_IPV4);
+	ipv6 = (decoded.outer_ip == ICE_RX_PTYPE_OUTER_IP) &&
+	       (decoded.outer_ip_ver == ICE_RX_PTYPE_OUTER_IPV6);
+
+	if (ipv4 && (rx_error & (BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_IPE_S) |
+				 BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_EIPE_S))))
+		goto checksum_fail;
+	else if (ipv6 && (rx_status &
+		 (BIT(ICE_RX_FLEX_DESC_STATUS0_IPV6EXADD_S))))
+		goto checksum_fail;
+
+	/* check for L4 errors and handle packets that were not able to be
+	 * checksummed due to arrival speed
+	 */
+	if (rx_error & BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_L4E_S))
+		goto checksum_fail;
+
+	/* Only report checksum unnecessary for TCP, UDP, or SCTP */
+	switch (decoded.inner_prot) {
+	case ICE_RX_PTYPE_INNER_PROT_TCP:
+	case ICE_RX_PTYPE_INNER_PROT_UDP:
+	case ICE_RX_PTYPE_INNER_PROT_SCTP:
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	default:
+		break;
+	}
+	return;
+
+checksum_fail:
+	ring->vsi->back->hw_csum_rx_error++;
+}
+
+/**
+ * ice_process_skb_fields - Populate skb header fields from Rx descriptor
+ * @rx_ring: Rx descriptor ring packet is being transacted on
+ * @rx_desc: pointer to the EOP Rx descriptor
+ * @skb: pointer to current skb being populated
+ * @ptype: the packet type decoded by hardware
+ *
+ * This function checks the ring, descriptor, and packet information in
+ * order to populate the hash, checksum, VLAN, protocol, and
+ * other fields within the skb.
+ */
+void
+ice_process_skb_fields(struct ice_ring *rx_ring,
+		       union ice_32b_rx_flex_desc *rx_desc,
+		       struct sk_buff *skb, u8 ptype)
+{
+	ice_rx_hash(rx_ring, rx_desc, skb, ptype);
+
+	/* modifies the skb - consumes the enet header */
+	skb->protocol = eth_type_trans(skb, rx_ring->netdev);
+
+	ice_rx_csum(rx_ring, skb, rx_desc, ptype);
+}
+
+/**
+ * ice_receive_skb - Send a completed packet up the stack
+ * @rx_ring: Rx ring in play
+ * @skb: packet to send up
+ * @vlan_tag: VLAN tag for packet
+ *
+ * This function sends the completed packet (via. skb) up the stack using
+ * gro receive functions (with/without VLAN tag)
+ */
+void
+ice_receive_skb(struct ice_ring *rx_ring, struct sk_buff *skb, u16 vlan_tag)
+{
+	if ((rx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+	    (vlan_tag & VLAN_VID_MASK))
+		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tag);
+	napi_gro_receive(&rx_ring->q_vector->napi, skb);
+}
+
+/**
+ * ice_xmit_xdp_ring - submit single packet to XDP ring for transmission
+ * @data: packet data pointer
+ * @size: packet data size
+ * @xdp_ring: XDP ring for transmission
+ */
+int ice_xmit_xdp_ring(void *data, u16 size, struct ice_ring *xdp_ring)
+{
+	u16 i = xdp_ring->next_to_use;
+	struct ice_tx_desc *tx_desc;
+	struct ice_tx_buf *tx_buf;
+	dma_addr_t dma;
+
+	if (!unlikely(ICE_DESC_UNUSED(xdp_ring))) {
+		xdp_ring->tx_stats.tx_busy++;
+		return ICE_XDP_CONSUMED;
+	}
+
+	dma = dma_map_single(xdp_ring->dev, data, size, DMA_TO_DEVICE);
+	if (dma_mapping_error(xdp_ring->dev, dma))
+		return ICE_XDP_CONSUMED;
+
+	tx_buf = &xdp_ring->tx_buf[i];
+	tx_buf->bytecount = size;
+	tx_buf->gso_segs = 1;
+	tx_buf->raw_buf = data;
+
+	/* record length, and DMA address */
+	dma_unmap_len_set(tx_buf, len, size);
+	dma_unmap_addr_set(tx_buf, dma, dma);
+
+	tx_desc = ICE_TX_DESC(xdp_ring, i);
+	tx_desc->buf_addr = cpu_to_le64(dma);
+	tx_desc->cmd_type_offset_bsz = build_ctob(ICE_TXD_LAST_DESC_CMD, 0,
+						  size, 0);
+
+	/* Make certain all of the status bits have been updated
+	 * before next_to_watch is written.
+	 */
+	smp_wmb();
+
+	i++;
+	if (i == xdp_ring->count)
+		i = 0;
+
+	tx_buf->next_to_watch = tx_desc;
+	xdp_ring->next_to_use = i;
+
+	return ICE_XDP_TX;
+}
+
+/**
+ * ice_xmit_xdp_buff - convert an XDP buffer to an XDP frame and send it
+ * @xdp: XDP buffer
+ * @xdp_ring: XDP Tx ring
+ *
+ * Returns negative on failure, 0 on success.
+ */
+int ice_xmit_xdp_buff(struct xdp_buff *xdp, struct ice_ring *xdp_ring)
+{
+	struct xdp_frame *xdpf = convert_to_xdp_frame(xdp);
+
+	if (unlikely(!xdpf))
+		return ICE_XDP_CONSUMED;
+
+	return ice_xmit_xdp_ring(xdpf->data, xdpf->len, xdp_ring);
+}
+
+/**
+ * ice_finalize_xdp_rx - Bump XDP Tx tail and/or flush redirect map
+ * @rx_ring: Rx ring
+ * @xdp_res: Result of the receive batch
+ *
+ * This function bumps XDP Tx tail and/or flush redirect map, and
+ * should be called when a batch of packets has been processed in the
+ * napi loop.
+ */
+void ice_finalize_xdp_rx(struct ice_ring *rx_ring, unsigned int xdp_res)
+{
+	if (xdp_res & ICE_XDP_REDIR)
+		xdp_do_flush_map();
+
+	if (xdp_res & ICE_XDP_TX) {
+		struct ice_ring *xdp_ring =
+			rx_ring->vsi->xdp_rings[rx_ring->q_index];
+
+		ice_xdp_ring_update_tail(xdp_ring);
+	}
+}

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

+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2019, Intel Corporation. */
+
+#ifndef _ICE_TXRX_LIB_H_
+#define _ICE_TXRX_LIB_H_
+#include "ice.h"
+
+/**
+ * ice_test_staterr - tests bits in Rx descriptor status and error fields
+ * @rx_desc: pointer to receive descriptor (in le64 format)
+ * @stat_err_bits: value to mask
+ *
+ * This function does some fast chicanery in order to return the
+ * value of the mask which is really only used for boolean tests.
+ * The status_error_len doesn't need to be shifted because it begins
+ * at offset zero.
+ */
+static inline bool
+ice_test_staterr(union ice_32b_rx_flex_desc *rx_desc, const u16 stat_err_bits)
+{
+	return !!(rx_desc->wb.status_error0 & cpu_to_le16(stat_err_bits));
+}
+
+static inline __le64
+build_ctob(u64 td_cmd, u64 td_offset, unsigned int size, u64 td_tag)
+{
+	return cpu_to_le64(ICE_TX_DESC_DTYPE_DATA |
+			   (td_cmd    << ICE_TXD_QW1_CMD_S) |
+			   (td_offset << ICE_TXD_QW1_OFFSET_S) |
+			   ((u64)size << ICE_TXD_QW1_TX_BUF_SZ_S) |
+			   (td_tag    << ICE_TXD_QW1_L2TAG1_S));
+}
+
+/**
+ * ice_xdp_ring_update_tail - Updates the XDP Tx ring tail register
+ * @xdp_ring: XDP Tx ring
+ *
+ * This function updates the XDP Tx ring tail register.
+ */
+static inline void ice_xdp_ring_update_tail(struct ice_ring *xdp_ring)
+{
+	/* Force memory writes to complete before letting h/w
+	 * know there are new descriptors to fetch.
+	 */
+	wmb();
+	writel_relaxed(xdp_ring->next_to_use, xdp_ring->tail);
+}
+
+void ice_finalize_xdp_rx(struct ice_ring *rx_ring, unsigned int xdp_res);
+int ice_xmit_xdp_buff(struct xdp_buff *xdp, struct ice_ring *xdp_ring);
+int ice_xmit_xdp_ring(void *data, u16 size, struct ice_ring *xdp_ring);
+void ice_release_rx_desc(struct ice_ring *rx_ring, u32 val);
+void
+ice_process_skb_fields(struct ice_ring *rx_ring,
+		       union ice_32b_rx_flex_desc *rx_desc,
+		       struct sk_buff *skb, u8 ptype);
+void
+ice_receive_skb(struct ice_ring *rx_ring, struct sk_buff *skb, u16 vlan_tag);
+#endif /* !_ICE_TXRX_LIB_H_ */

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

@@ -2,6 +2,7 @@
 /* Copyright (c) 2018, Intel Corporation. */
 
 #include "ice.h"
+#include "ice_base.h"
 #include "ice_lib.h"
 
 /**

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

+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2019, Intel Corporation. */
+
+#ifndef _ICE_XSK_H_
+#define _ICE_XSK_H_
+#include "ice_txrx.h"
+#include "ice.h"
+
+struct ice_vsi;
+
+#ifdef CONFIG_XDP_SOCKETS
+int ice_xsk_umem_setup(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid);
+void ice_zca_free(struct zero_copy_allocator *zca, unsigned long handle);
+int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget);
+bool ice_clean_tx_irq_zc(struct ice_ring *xdp_ring, int budget);
+int ice_xsk_wakeup(struct net_device *netdev, u32 queue_id, u32 flags);
+bool ice_alloc_rx_bufs_slow_zc(struct ice_ring *rx_ring, u16 count);
+bool ice_xsk_any_rx_ring_ena(struct ice_vsi *vsi);
+void ice_xsk_clean_rx_ring(struct ice_ring *rx_ring);
+void ice_xsk_clean_xdp_ring(struct ice_ring *xdp_ring);
+#else
+static inline int
+ice_xsk_umem_setup(struct ice_vsi __always_unused *vsi,
+		   struct xdp_umem __always_unused *umem,
+		   u16 __always_unused qid)
+{
+	return -ENOTSUPP;
+}
+
+static inline void
+ice_zca_free(struct zero_copy_allocator __always_unused *zca,
+	     unsigned long __always_unused handle)
+{
+}
+
+static inline int
+ice_clean_rx_irq_zc(struct ice_ring __always_unused *rx_ring,
+		    int __always_unused budget)
+{
+	return 0;
+}
+
+static inline bool
+ice_clean_tx_irq_zc(struct ice_ring __always_unused *xdp_ring,
+		    int __always_unused budget)
+{
+	return false;
+}
+
+static inline bool
+ice_alloc_rx_bufs_slow_zc(struct ice_ring __always_unused *rx_ring,
+			  u16 __always_unused count)
+{
+	return false;
+}
+
+static inline bool ice_xsk_any_rx_ring_ena(struct ice_vsi __always_unused *vsi)
+{
+	return false;
+}
+
+static inline int
+ice_xsk_wakeup(struct net_device __always_unused *netdev,
+	       u32 __always_unused queue_id, u32 __always_unused flags)
+{
+	return -ENOTSUPP;
+}
+
+#define ice_xsk_clean_rx_ring(rx_ring) do {} while (0)
+#define ice_xsk_clean_xdp_ring(xdp_ring) do {} while (0)
+#endif /* CONFIG_XDP_SOCKETS */
+#endif /* !_ICE_XSK_H_ */