i40e: add AF_XDP zero-copy Rx support

This patch adds zero-copy Rx support for AF_XDP sockets. Instead of
allocating buffers of type MEM_TYPE_PAGE_SHARED, the Rx frames are
allocated as MEM_TYPE_ZERO_COPY when AF_XDP is enabled for a certain
queue.

All AF_XDP specific functions are added to a new file, i40e_xsk.c.

Note that when AF_XDP zero-copy is enabled, the XDP action XDP_PASS
will allocate a new buffer and copy the zero-copy frame prior passing
it to the kernel stack.

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

drivers/net/ethernet/intel/i40e/Makefile

@@ -22,6 +22,7 @@ i40e-objs := i40e_main.o \
 	i40e_txrx.o	\
 	i40e_ptp.o	\
 	i40e_client.o   \
-	i40e_virtchnl_pf.o
+	i40e_virtchnl_pf.o \
+	i40e_xsk.o
 
 i40e-$(CONFIG_I40E_DCB) += i40e_dcb.o i40e_dcb_nl.o

drivers/net/ethernet/intel/i40e/i40e.h

@@ -786,6 +786,11 @@ struct i40e_vsi {
 
 	/* VSI specific handlers */
 	irqreturn_t (*irq_handler)(int irq, void *data);
+
+	/* AF_XDP zero-copy */
+	struct xdp_umem **xsk_umems;
+	u16 num_xsk_umems_used;
+	u16 num_xsk_umems;
 } ____cacheline_internodealigned_in_smp;
 
 struct i40e_netdev_priv {
@@ -1090,6 +1095,20 @@ static inline bool i40e_enabled_xdp_vsi(struct i40e_vsi *vsi)
 	return !!vsi->xdp_prog;
 }
 
+static inline struct xdp_umem *i40e_xsk_umem(struct i40e_ring *ring)
+{
+	bool xdp_on = i40e_enabled_xdp_vsi(ring->vsi);
+	int qid = ring->queue_index;
+
+	if (ring_is_xdp(ring))
+		qid -= ring->vsi->alloc_queue_pairs;
+
+	if (!ring->vsi->xsk_umems || !ring->vsi->xsk_umems[qid] || !xdp_on)
+		return NULL;
+
+	return ring->vsi->xsk_umems[qid];
+}
+
 int i40e_create_queue_channel(struct i40e_vsi *vsi, struct i40e_channel *ch);
 int i40e_set_bw_limit(struct i40e_vsi *vsi, u16 seid, u64 max_tx_rate);
 int i40e_add_del_cloud_filter(struct i40e_vsi *vsi,

drivers/net/ethernet/intel/i40e/i40e_main.c

@@ -9,7 +9,9 @@
 /* Local includes */
 #include "i40e.h"
 #include "i40e_diag.h"
+#include "i40e_xsk.h"
 #include <net/udp_tunnel.h>
+#include <net/xdp_sock.h>
 /* All i40e tracepoints are defined by the include below, which
  * must be included exactly once across the whole kernel with
  * CREATE_TRACE_POINTS defined
@@ -3181,13 +3183,46 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring)
 	struct i40e_hw *hw = &vsi->back->hw;
 	struct i40e_hmc_obj_rxq rx_ctx;
 	i40e_status err = 0;
+	bool ok;
+	int ret;
 
 	bitmap_zero(ring->state, __I40E_RING_STATE_NBITS);
 
 	/* clear the context structure first */
 	memset(&rx_ctx, 0, sizeof(rx_ctx));
 
-	ring->rx_buf_len = vsi->rx_buf_len;
+	if (ring->vsi->type == I40E_VSI_MAIN)
+		xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq);
+
+	ring->xsk_umem = i40e_xsk_umem(ring);
+	if (ring->xsk_umem) {
+		ring->rx_buf_len = ring->xsk_umem->chunk_size_nohr -
+				   XDP_PACKET_HEADROOM;
+		/* For AF_XDP ZC, we disallow packets to span on
+		 * multiple buffers, thus letting us skip that
+		 * handling in the fast-path.
+		 */
+		chain_len = 1;
+		ring->zca.free = i40e_zca_free;
+		ret = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq,
+						 MEM_TYPE_ZERO_COPY,
+						 &ring->zca);
+		if (ret)
+			return ret;
+		dev_info(&vsi->back->pdev->dev,
+			 "Registered XDP mem model MEM_TYPE_ZERO_COPY on Rx ring %d\n",
+			 ring->queue_index);
+
+	} else {
+		ring->rx_buf_len = vsi->rx_buf_len;
+		if (ring->vsi->type == I40E_VSI_MAIN) {
+			ret = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq,
+							 MEM_TYPE_PAGE_SHARED,
+							 NULL);
+			if (ret)
+				return ret;
+		}
+	}
 
 	rx_ctx.dbuff = DIV_ROUND_UP(ring->rx_buf_len,
 				    BIT_ULL(I40E_RXQ_CTX_DBUFF_SHIFT));
@@ -3243,7 +3278,15 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring)
 	ring->tail = hw->hw_addr + I40E_QRX_TAIL(pf_q);
 	writel(0, ring->tail);
 
-	i40e_alloc_rx_buffers(ring, I40E_DESC_UNUSED(ring));
+	ok = ring->xsk_umem ?
+	     i40e_alloc_rx_buffers_zc(ring, I40E_DESC_UNUSED(ring)) :
+	     !i40e_alloc_rx_buffers(ring, I40E_DESC_UNUSED(ring));
+	if (!ok) {
+		dev_info(&vsi->back->pdev->dev,
+			 "Failed allocate some buffers on %sRx ring %d (pf_q %d)\n",
+			 ring->xsk_umem ? "UMEM enabled " : "",
+			 ring->queue_index, pf_q);
+	}
 
 	return 0;
 }
@@ -12097,6 +12140,12 @@ static int i40e_xdp(struct net_device *dev,
 	case XDP_QUERY_PROG:
 		xdp->prog_id = vsi->xdp_prog ? vsi->xdp_prog->aux->id : 0;
 		return 0;
+	case XDP_QUERY_XSK_UMEM:
+		return i40e_xsk_umem_query(vsi, &xdp->xsk.umem,
+					   xdp->xsk.queue_id);
+	case XDP_SETUP_XSK_UMEM:
+		return i40e_xsk_umem_setup(vsi, xdp->xsk.umem,
+					   xdp->xsk.queue_id);
 	default:
 		return -EINVAL;
 	}

drivers/net/ethernet/intel/i40e/i40e_txrx.c

@@ -9,6 +9,7 @@
 #include "i40e_trace.h"
 #include "i40e_prototype.h"
 #include "i40e_txrx_common.h"
+#include "i40e_xsk.h"
 
 static inline __le64 build_ctob(u32 td_cmd, u32 td_offset, unsigned int size,
 				u32 td_tag)
@@ -1380,6 +1381,9 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring)
 		rx_ring->skb = NULL;
 	}
 
+	if (rx_ring->xsk_umem)
+		goto skip_free;
+
 	/* Free all the Rx ring sk_buffs */
 	for (i = 0; i < rx_ring->count; i++) {
 		struct i40e_rx_buffer *rx_bi = &rx_ring->rx_bi[i];
@@ -1408,6 +1412,7 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring)
 		rx_bi->page_offset = 0;
 	}
 
+skip_free:
 	bi_size = sizeof(struct i40e_rx_buffer) * rx_ring->count;
 	memset(rx_ring->rx_bi, 0, bi_size);
 
@@ -2641,7 +2646,9 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)
 	budget_per_ring = max(budget/q_vector->num_ringpairs, 1);
 
 	i40e_for_each_ring(ring, q_vector->rx) {
-		int cleaned = i40e_clean_rx_irq(ring, budget_per_ring);
+		int cleaned = ring->xsk_umem ?
+			      i40e_clean_rx_irq_zc(ring, budget_per_ring) :
+			      i40e_clean_rx_irq(ring, budget_per_ring);
 
 		work_done += cleaned;
 		/* if we clean as many as budgeted, we must not be done */

drivers/net/ethernet/intel/i40e/i40e_txrx.h

@@ -296,13 +296,17 @@ struct i40e_tx_buffer {
 
 struct i40e_rx_buffer {
 	dma_addr_t dma;
-	struct page *page;
-#if (BITS_PER_LONG > 32) || (PAGE_SIZE >= 65536)
-	__u32 page_offset;
-#else
-	__u16 page_offset;
-#endif
-	__u16 pagecnt_bias;
+	union {
+		struct {
+			struct page *page;
+			__u32 page_offset;
+			__u16 pagecnt_bias;
+		};
+		struct {
+			void *addr;
+			u64 handle;
+		};
+	};
 };
 
 struct i40e_queue_stats {
@@ -414,6 +418,8 @@ struct i40e_ring {
 
 	struct i40e_channel *ch;
 	struct xdp_rxq_info xdp_rxq;
+	struct xdp_umem *xsk_umem;
+	struct zero_copy_allocator zca; /* ZC allocator anchor */
 } ____cacheline_internodealigned_in_smp;
 
 static inline bool ring_uses_build_skb(struct i40e_ring *ring)

drivers/net/ethernet/intel/i40e/i40e_xsk.c

@@ -0,0 +1,661 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2018 Intel Corporation. */
+
+#include <linux/bpf_trace.h>
+#include <net/xdp_sock.h>
+#include <net/xdp.h>
+
+#include "i40e.h"
+#include "i40e_txrx_common.h"
+#include "i40e_xsk.h"
+
+/**
+ * i40e_alloc_xsk_umems - Allocate an array to store per ring UMEMs
+ * @vsi: Current VSI
+ *
+ * Returns 0 on success, <0 on failure
+ **/
+static int i40e_alloc_xsk_umems(struct i40e_vsi *vsi)
+{
+	if (vsi->xsk_umems)
+		return 0;
+
+	vsi->num_xsk_umems_used = 0;
+	vsi->num_xsk_umems = vsi->alloc_queue_pairs;
+	vsi->xsk_umems = kcalloc(vsi->num_xsk_umems, sizeof(*vsi->xsk_umems),
+				 GFP_KERNEL);
+	if (!vsi->xsk_umems) {
+		vsi->num_xsk_umems = 0;
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_add_xsk_umem - Store an UMEM for a certain ring/qid
+ * @vsi: Current VSI
+ * @umem: UMEM to store
+ * @qid: Ring/qid to associate with the UMEM
+ *
+ * Returns 0 on success, <0 on failure
+ **/
+static int i40e_add_xsk_umem(struct i40e_vsi *vsi, struct xdp_umem *umem,
+			     u16 qid)
+{
+	int err;
+
+	err = i40e_alloc_xsk_umems(vsi);
+	if (err)
+		return err;
+
+	vsi->xsk_umems[qid] = umem;
+	vsi->num_xsk_umems_used++;
+
+	return 0;
+}
+
+/**
+ * i40e_remove_xsk_umem - Remove an UMEM for a certain ring/qid
+ * @vsi: Current VSI
+ * @qid: Ring/qid associated with the UMEM
+ **/
+static void i40e_remove_xsk_umem(struct i40e_vsi *vsi, u16 qid)
+{
+	vsi->xsk_umems[qid] = NULL;
+	vsi->num_xsk_umems_used--;
+
+	if (vsi->num_xsk_umems == 0) {
+		kfree(vsi->xsk_umems);
+		vsi->xsk_umems = NULL;
+		vsi->num_xsk_umems = 0;
+	}
+}
+
+/**
+ * i40e_xsk_umem_dma_map - DMA maps all UMEM memory for the netdev
+ * @vsi: Current VSI
+ * @umem: UMEM to DMA map
+ *
+ * Returns 0 on success, <0 on failure
+ **/
+static int i40e_xsk_umem_dma_map(struct i40e_vsi *vsi, struct xdp_umem *umem)
+{
+	struct i40e_pf *pf = vsi->back;
+	struct device *dev;
+	unsigned int i, j;
+	dma_addr_t dma;
+
+	dev = &pf->pdev->dev;
+	for (i = 0; i < umem->npgs; i++) {
+		dma = dma_map_page_attrs(dev, umem->pgs[i], 0, PAGE_SIZE,
+					 DMA_BIDIRECTIONAL, I40E_RX_DMA_ATTR);
+		if (dma_mapping_error(dev, dma))
+			goto out_unmap;
+
+		umem->pages[i].dma = dma;
+	}
+
+	return 0;
+
+out_unmap:
+	for (j = 0; j < i; j++) {
+		dma_unmap_page_attrs(dev, umem->pages[i].dma, PAGE_SIZE,
+				     DMA_BIDIRECTIONAL, I40E_RX_DMA_ATTR);
+		umem->pages[i].dma = 0;
+	}
+
+	return -1;
+}
+
+/**
+ * i40e_xsk_umem_dma_unmap - DMA unmaps all UMEM memory for the netdev
+ * @vsi: Current VSI
+ * @umem: UMEM to DMA map
+ **/
+static void i40e_xsk_umem_dma_unmap(struct i40e_vsi *vsi, struct xdp_umem *umem)
+{
+	struct i40e_pf *pf = vsi->back;
+	struct device *dev;
+	unsigned int i;
+
+	dev = &pf->pdev->dev;
+
+	for (i = 0; i < umem->npgs; i++) {
+		dma_unmap_page_attrs(dev, umem->pages[i].dma, PAGE_SIZE,
+				     DMA_BIDIRECTIONAL, I40E_RX_DMA_ATTR);
+
+		umem->pages[i].dma = 0;
+	}
+}
+
+/**
+ * i40e_xsk_umem_enable - Enable/associate an UMEM to a certain ring/qid
+ * @vsi: Current VSI
+ * @umem: UMEM
+ * @qid: Rx ring to associate UMEM to
+ *
+ * Returns 0 on success, <0 on failure
+ **/
+static int i40e_xsk_umem_enable(struct i40e_vsi *vsi, struct xdp_umem *umem,
+				u16 qid)
+{
+	bool if_running;
+	int err;
+
+	if (vsi->type != I40E_VSI_MAIN)
+		return -EINVAL;
+
+	if (qid >= vsi->num_queue_pairs)
+		return -EINVAL;
+
+	if (vsi->xsk_umems) {
+		if (qid >= vsi->num_xsk_umems)
+			return -EINVAL;
+		if (vsi->xsk_umems[qid])
+			return -EBUSY;
+	}
+
+	err = i40e_xsk_umem_dma_map(vsi, umem);
+	if (err)
+		return err;
+
+	if_running = netif_running(vsi->netdev) && i40e_enabled_xdp_vsi(vsi);
+
+	if (if_running) {
+		err = i40e_queue_pair_disable(vsi, qid);
+		if (err)
+			return err;
+	}
+
+	err = i40e_add_xsk_umem(vsi, umem, qid);
+	if (err)
+		return err;
+
+	if (if_running) {
+		err = i40e_queue_pair_enable(vsi, qid);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_xsk_umem_disable - Diassociate an UMEM from a certain ring/qid
+ * @vsi: Current VSI
+ * @qid: Rx ring to associate UMEM to
+ *
+ * Returns 0 on success, <0 on failure
+ **/
+static int i40e_xsk_umem_disable(struct i40e_vsi *vsi, u16 qid)
+{
+	bool if_running;
+	int err;
+
+	if (!vsi->xsk_umems || qid >= vsi->num_xsk_umems ||
+	    !vsi->xsk_umems[qid])
+		return -EINVAL;
+
+	if_running = netif_running(vsi->netdev) && i40e_enabled_xdp_vsi(vsi);
+
+	if (if_running) {
+		err = i40e_queue_pair_disable(vsi, qid);
+		if (err)
+			return err;
+	}
+
+	i40e_xsk_umem_dma_unmap(vsi, vsi->xsk_umems[qid]);
+	i40e_remove_xsk_umem(vsi, qid);
+
+	if (if_running) {
+		err = i40e_queue_pair_enable(vsi, qid);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+/**
+ * i40e_xsk_umem_query - Queries a certain ring/qid for its UMEM
+ * @vsi: Current VSI
+ * @umem: UMEM associated to the ring, if any
+ * @qid: Rx ring to associate UMEM to
+ *
+ * This function will store, if any, the UMEM associated to certain ring.
+ *
+ * Returns 0 on success, <0 on failure
+ **/
+int i40e_xsk_umem_query(struct i40e_vsi *vsi, struct xdp_umem **umem,
+			u16 qid)
+{
+	if (vsi->type != I40E_VSI_MAIN)
+		return -EINVAL;
+
+	if (qid >= vsi->num_queue_pairs)
+		return -EINVAL;
+
+	if (vsi->xsk_umems) {
+		if (qid >= vsi->num_xsk_umems)
+			return -EINVAL;
+		*umem = vsi->xsk_umems[qid];
+		return 0;
+	}
+
+	*umem = NULL;
+	return 0;
+}
+
+/**
+ * i40e_xsk_umem_query - Queries a certain ring/qid for its UMEM
+ * @vsi: Current VSI
+ * @umem: UMEM to enable/associate to a ring, or NULL to disable
+ * @qid: Rx ring to (dis)associate UMEM (from)to
+ *
+ * This function enables or disables an UMEM to a certain ring.
+ *
+ * Returns 0 on success, <0 on failure
+ **/
+int i40e_xsk_umem_setup(struct i40e_vsi *vsi, struct xdp_umem *umem,
+			u16 qid)
+{
+	return umem ? i40e_xsk_umem_enable(vsi, umem, qid) :
+		i40e_xsk_umem_disable(vsi, qid);
+}
+
+/**
+ * i40e_run_xdp_zc - Executes an XDP program on an xdp_buff
+ * @rx_ring: Rx ring
+ * @xdp: xdp_buff used as input to the XDP program
+ *
+ * This function enables or disables an UMEM to a certain ring.
+ *
+ * Returns any of I40E_XDP_{PASS, CONSUMED, TX, REDIR}
+ **/
+static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp)
+{
+	int err, result = I40E_XDP_PASS;
+	struct i40e_ring *xdp_ring;
+	struct bpf_prog *xdp_prog;
+	u32 act;
+
+	rcu_read_lock();
+	/* NB! xdp_prog will always be !NULL, due to the fact that
+	 * this path is enabled by setting an XDP program.
+	 */
+	xdp_prog = READ_ONCE(rx_ring->xdp_prog);
+	act = bpf_prog_run_xdp(xdp_prog, xdp);
+	xdp->handle += xdp->data - xdp->data_hard_start;
+	switch (act) {
+	case XDP_PASS:
+		break;
+	case XDP_TX:
+		xdp_ring = rx_ring->vsi->xdp_rings[rx_ring->queue_index];
+		result = i40e_xmit_xdp_tx_ring(xdp, xdp_ring);
+		break;
+	case XDP_REDIRECT:
+		err = xdp_do_redirect(rx_ring->netdev, xdp, xdp_prog);
+		result = !err ? I40E_XDP_REDIR : I40E_XDP_CONSUMED;
+		break;
+	default:
+		bpf_warn_invalid_xdp_action(act);
+	case XDP_ABORTED:
+		trace_xdp_exception(rx_ring->netdev, xdp_prog, act);
+		/* fallthrough -- handle aborts by dropping packet */
+	case XDP_DROP:
+		result = I40E_XDP_CONSUMED;
+		break;
+	}
+	rcu_read_unlock();
+	return result;
+}
+
+/**
+ * i40e_alloc_buffer_zc - Allocates an i40e_rx_buffer
+ * @rx_ring: Rx ring
+ * @bi: Rx buffer to populate
+ *
+ * This function allocates an Rx buffer. The buffer can come from fill
+ * queue, or via the recycle queue (next_to_alloc).
+ *
+ * Returns true for a successful allocation, false otherwise
+ **/
+static bool i40e_alloc_buffer_zc(struct i40e_ring *rx_ring,
+				 struct i40e_rx_buffer *bi)
+{
+	struct xdp_umem *umem = rx_ring->xsk_umem;
+	void *addr = bi->addr;
+	u64 handle, hr;
+
+	if (addr) {
+		rx_ring->rx_stats.page_reuse_count++;
+		return true;
+	}
+
+	if (!xsk_umem_peek_addr(umem, &handle)) {
+		rx_ring->rx_stats.alloc_page_failed++;
+		return false;
+	}
+
+	hr = umem->headroom + XDP_PACKET_HEADROOM;
+
+	bi->dma = xdp_umem_get_dma(umem, handle);
+	bi->dma += hr;
+
+	bi->addr = xdp_umem_get_data(umem, handle);
+	bi->addr += hr;
+
+	bi->handle = handle + umem->headroom;
+
+	xsk_umem_discard_addr(umem);
+	return true;
+}
+
+/**
+ * i40e_alloc_rx_buffers_zc - Allocates a number of Rx buffers
+ * @rx_ring: Rx ring
+ * @count: The number of buffers to allocate
+ *
+ * This function allocates a number of Rx buffers and places them on
+ * the Rx ring.
+ *
+ * Returns true for a successful allocation, false otherwise
+ **/
+bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 count)
+{
+	u16 ntu = rx_ring->next_to_use;
+	union i40e_rx_desc *rx_desc;
+	struct i40e_rx_buffer *bi;
+	bool ok = true;
+
+	rx_desc = I40E_RX_DESC(rx_ring, ntu);
+	bi = &rx_ring->rx_bi[ntu];
+	do {
+		if (!i40e_alloc_buffer_zc(rx_ring, bi)) {
+			ok = false;
+			goto no_buffers;
+		}
+
+		dma_sync_single_range_for_device(rx_ring->dev, bi->dma, 0,
+						 rx_ring->rx_buf_len,
+						 DMA_BIDIRECTIONAL);
+
+		rx_desc->read.pkt_addr = cpu_to_le64(bi->dma);
+
+		rx_desc++;
+		bi++;
+		ntu++;
+
+		if (unlikely(ntu == rx_ring->count)) {
+			rx_desc = I40E_RX_DESC(rx_ring, 0);
+			bi = rx_ring->rx_bi;
+			ntu = 0;
+		}
+
+		rx_desc->wb.qword1.status_error_len = 0;
+		count--;
+	} while (count);
+
+no_buffers:
+	if (rx_ring->next_to_use != ntu)
+		i40e_release_rx_desc(rx_ring, ntu);
+
+	return ok;
+}
+
+/**
+ * i40e_get_rx_buffer_zc - Return the current Rx buffer
+ * @rx_ring: Rx ring
+ * @size: The size of the rx buffer (read from descriptor)
+ *
+ * This function returns the current, received Rx buffer, and also
+ * does DMA synchronization.  the Rx ring.
+ *
+ * Returns the received Rx buffer
+ **/
+static struct i40e_rx_buffer *i40e_get_rx_buffer_zc(struct i40e_ring *rx_ring,
+						    const unsigned int size)
+{
+	struct i40e_rx_buffer *bi;
+
+	bi = &rx_ring->rx_bi[rx_ring->next_to_clean];
+
+	/* we are reusing so sync this buffer for CPU use */
+	dma_sync_single_range_for_cpu(rx_ring->dev,
+				      bi->dma, 0,
+				      size,
+				      DMA_BIDIRECTIONAL);
+
+	return bi;
+}
+
+/**
+ * i40e_reuse_rx_buffer_zc - Recycle an Rx buffer
+ * @rx_ring: Rx ring
+ * @old_bi: The Rx buffer to recycle
+ *
+ * This function recycles a finished Rx buffer, and places it on the
+ * recycle queue (next_to_alloc).
+ **/
+static void i40e_reuse_rx_buffer_zc(struct i40e_ring *rx_ring,
+				    struct i40e_rx_buffer *old_bi)
+{
+	struct i40e_rx_buffer *new_bi = &rx_ring->rx_bi[rx_ring->next_to_alloc];
+	unsigned long mask = (unsigned long)rx_ring->xsk_umem->props.chunk_mask;
+	u64 hr = rx_ring->xsk_umem->headroom + XDP_PACKET_HEADROOM;
+	u16 nta = rx_ring->next_to_alloc;
+
+	/* update, and store next to alloc */
+	nta++;
+	rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0;
+
+	/* transfer page from old buffer to new buffer */
+	new_bi->dma = old_bi->dma & mask;
+	new_bi->dma += hr;
+
+	new_bi->addr = (void *)((unsigned long)old_bi->addr & mask);
+	new_bi->addr += hr;
+
+	new_bi->handle = old_bi->handle & mask;
+	new_bi->handle += rx_ring->xsk_umem->headroom;
+
+	old_bi->addr = NULL;
+}
+
+/**
+ * i40e_zca_free - Free callback for MEM_TYPE_ZERO_COPY allocations
+ * @alloc: Zero-copy allocator
+ * @handle: Buffer handle
+ **/
+void i40e_zca_free(struct zero_copy_allocator *alloc, unsigned long handle)
+{
+	struct i40e_rx_buffer *bi;
+	struct i40e_ring *rx_ring;
+	u64 hr, mask;
+	u16 nta;
+
+	rx_ring = container_of(alloc, struct i40e_ring, zca);
+	hr = rx_ring->xsk_umem->headroom + XDP_PACKET_HEADROOM;
+	mask = rx_ring->xsk_umem->props.chunk_mask;
+
+	nta = rx_ring->next_to_alloc;
+	bi = &rx_ring->rx_bi[nta];
+
+	nta++;
+	rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0;
+
+	handle &= mask;
+
+	bi->dma = xdp_umem_get_dma(rx_ring->xsk_umem, handle);
+	bi->dma += hr;
+
+	bi->addr = xdp_umem_get_data(rx_ring->xsk_umem, handle);
+	bi->addr += hr;
+
+	bi->handle = (u64)handle + rx_ring->xsk_umem->headroom;
+}
+
+/**
+ * i40e_construct_skb_zc - Create skbufff from zero-copy Rx buffer
+ * @rx_ring: Rx ring
+ * @bi: Rx buffer
+ * @xdp: xdp_buff
+ *
+ * This functions allocates a new skb from a zero-copy Rx buffer.
+ *
+ * Returns the skb, or NULL on failure.
+ **/
+static struct sk_buff *i40e_construct_skb_zc(struct i40e_ring *rx_ring,
+					     struct i40e_rx_buffer *bi,
+					     struct xdp_buff *xdp)
+{
+	unsigned int metasize = xdp->data - xdp->data_meta;
+	unsigned int datasize = xdp->data_end - xdp->data;
+	struct sk_buff *skb;
+
+	/* allocate a skb to store the frags */
+	skb = __napi_alloc_skb(&rx_ring->q_vector->napi,
+			       xdp->data_end - xdp->data_hard_start,
+			       GFP_ATOMIC | __GFP_NOWARN);
+	if (unlikely(!skb))
+		return NULL;
+
+	skb_reserve(skb, xdp->data - xdp->data_hard_start);
+	memcpy(__skb_put(skb, datasize), xdp->data, datasize);
+	if (metasize)
+		skb_metadata_set(skb, metasize);
+
+	i40e_reuse_rx_buffer_zc(rx_ring, bi);
+	return skb;
+}
+
+/**
+ * i40e_inc_ntc: Advance the next_to_clean index
+ * @rx_ring: Rx ring
+ **/
+static void i40e_inc_ntc(struct i40e_ring *rx_ring)
+{
+	u32 ntc = rx_ring->next_to_clean + 1;
+
+	ntc = (ntc < rx_ring->count) ? ntc : 0;
+	rx_ring->next_to_clean = ntc;
+	prefetch(I40E_RX_DESC(rx_ring, ntc));
+}
+
+/**
+ * i40e_clean_rx_irq_zc - Consumes Rx packets from the hardware ring
+ * @rx_ring: Rx ring
+ * @budget: NAPI budget
+ *
+ * Returns amount of work completed
+ **/
+int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget)
+{
+	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
+	u16 cleaned_count = I40E_DESC_UNUSED(rx_ring);
+	unsigned int xdp_res, xdp_xmit = 0;
+	bool failure = false;
+	struct sk_buff *skb;
+	struct xdp_buff xdp;
+
+	xdp.rxq = &rx_ring->xdp_rxq;
+
+	while (likely(total_rx_packets < (unsigned int)budget)) {
+		struct i40e_rx_buffer *bi;
+		union i40e_rx_desc *rx_desc;
+		unsigned int size;
+		u16 vlan_tag;
+		u8 rx_ptype;
+		u64 qword;
+
+		if (cleaned_count >= I40E_RX_BUFFER_WRITE) {
+			failure = failure ||
+				  !i40e_alloc_rx_buffers_zc(rx_ring,
+							    cleaned_count);
+			cleaned_count = 0;
+		}
+
+		rx_desc = I40E_RX_DESC(rx_ring, rx_ring->next_to_clean);
+		qword = le64_to_cpu(rx_desc->wb.qword1.status_error_len);
+
+		/* This memory barrier is needed to keep us from reading
+		 * any other fields out of the rx_desc until we have
+		 * verified the descriptor has been written back.
+		 */
+		dma_rmb();
+
+		bi = i40e_clean_programming_status(rx_ring, rx_desc,
+						   qword);
+		if (unlikely(bi)) {
+			i40e_reuse_rx_buffer_zc(rx_ring, bi);
+			cleaned_count++;
+			continue;
+		}
+
+		size = (qword & I40E_RXD_QW1_LENGTH_PBUF_MASK) >>
+		       I40E_RXD_QW1_LENGTH_PBUF_SHIFT;
+		if (!size)
+			break;
+
+		bi = i40e_get_rx_buffer_zc(rx_ring, size);
+		xdp.data = bi->addr;
+		xdp.data_meta = xdp.data;
+		xdp.data_hard_start = xdp.data - XDP_PACKET_HEADROOM;
+		xdp.data_end = xdp.data + size;
+		xdp.handle = bi->handle;
+
+		xdp_res = i40e_run_xdp_zc(rx_ring, &xdp);
+		if (xdp_res) {
+			if (xdp_res & (I40E_XDP_TX | I40E_XDP_REDIR)) {
+				xdp_xmit |= xdp_res;
+				bi->addr = NULL;
+			} else {
+				i40e_reuse_rx_buffer_zc(rx_ring, bi);
+			}
+
+			total_rx_bytes += size;
+			total_rx_packets++;
+
+			cleaned_count++;
+			i40e_inc_ntc(rx_ring);
+			continue;
+		}
+
+		/* XDP_PASS path */
+
+		/* NB! We are not checking for errors using
+		 * i40e_test_staterr with
+		 * BIT(I40E_RXD_QW1_ERROR_SHIFT). This is due to that
+		 * SBP is *not* set in PRT_SBPVSI (default not set).
+		 */
+		skb = i40e_construct_skb_zc(rx_ring, bi, &xdp);
+		if (!skb) {
+			rx_ring->rx_stats.alloc_buff_failed++;
+			break;
+		}
+
+		cleaned_count++;
+		i40e_inc_ntc(rx_ring);
+
+		if (eth_skb_pad(skb))
+			continue;
+
+		total_rx_bytes += skb->len;
+		total_rx_packets++;
+
+		qword = le64_to_cpu(rx_desc->wb.qword1.status_error_len);
+		rx_ptype = (qword & I40E_RXD_QW1_PTYPE_MASK) >>
+			   I40E_RXD_QW1_PTYPE_SHIFT;
+		i40e_process_skb_fields(rx_ring, rx_desc, skb, rx_ptype);
+
+		vlan_tag = (qword & BIT(I40E_RX_DESC_STATUS_L2TAG1P_SHIFT)) ?
+			   le16_to_cpu(rx_desc->wb.qword0.lo_dword.l2tag1) : 0;
+		i40e_receive_skb(rx_ring, skb, vlan_tag);
+	}
+
+	i40e_finalize_xdp_rx(rx_ring, xdp_xmit);
+	i40e_update_rx_stats(rx_ring, total_rx_bytes, total_rx_packets);
+	return failure ? budget : (int)total_rx_packets;
+}
+

drivers/net/ethernet/intel/i40e/i40e_xsk.h

@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2018 Intel Corporation. */
+
+#ifndef _I40E_XSK_H_
+#define _I40E_XSK_H_
+
+struct i40e_vsi;
+struct xdp_umem;
+struct zero_copy_allocator;
+
+int i40e_queue_pair_disable(struct i40e_vsi *vsi, int queue_pair);
+int i40e_queue_pair_enable(struct i40e_vsi *vsi, int queue_pair);
+int i40e_xsk_umem_query(struct i40e_vsi *vsi, struct xdp_umem **umem,
+			u16 qid);
+int i40e_xsk_umem_setup(struct i40e_vsi *vsi, struct xdp_umem *umem,
+			u16 qid);
+void i40e_zca_free(struct zero_copy_allocator *alloc, unsigned long handle);
+bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 cleaned_count);
+int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget);
+
+#endif /* _I40E_XSK_H_ */