ixgbevf: Add pseudo header split

This patch introduces pseudo header split support in the ixgbevf
driver, specifically targeting ixgbe_mac_82599_vf.

On older hardware (e.g. ixgbe_mac_82599_vf), RX DMA write size can only be
limited in 1K increments. This causes issues when attempting to fit
multiple packets per page, as a DMA write may overwrite the
headroom of the next packet.

To address this, introduce pseudo header split support, where the hardware
copies the full L2 header into a dedicated header buffer. This avoids the
need for HR/TR alignment and allows safe skb construction from the header
buffer without risking overwrites.

Signed-off-by: Natalia Wochtman <natalia.wochtman@intel.com>

Author: nwochtma

drivers/net/ethernet/intel/ixgbevf/ixgbevf.h

@@ -89,6 +89,7 @@ struct ixgbevf_ring {
 	dma_addr_t dma;			/* phys. address of descriptor ring */
 	unsigned int size;		/* length in bytes */
 	u32 truesize;			/* Rx buffer full size */
+	u32 hdr_truesize;		/* Rx header buffer full size */
 	u16 count;			/* amount of descriptors */
 	u16 next_to_use;
 	u16 next_to_clean;
@@ -104,6 +105,8 @@ struct ixgbevf_ring {
 		struct ixgbevf_tx_queue_stats tx_stats;
 		struct ixgbevf_rx_queue_stats rx_stats;
 	};
+	struct libeth_fqe *hdr_fqes;
+	struct page_pool *hdr_pp;
 	struct xdp_rxq_info xdp_rxq;
 	u64 hw_csum_rx_error;
 	u8 __iomem *tail;
@@ -113,6 +116,7 @@ struct ixgbevf_ring {
 	 */
 	u16 reg_idx;
 	int queue_index; /* needed for multiqueue queue management */
+	u32 hdr_buf_len;
 	u32 rx_buf_len;
 	struct libeth_xdp_buff_stash xdp_stash;
 } ____cacheline_internodealigned_in_smp;
@@ -147,6 +151,8 @@ struct ixgbevf_ring {
 #define IXGBEVF_RX_PAGE_LEN(hr)		(ALIGN_DOWN(LIBETH_RX_PAGE_LEN(hr), \
 					 IXGBE_SRRCTL_BSIZEPKT_STEP))
 
+#define IXGBEVF_FLAG_HSPLIT	BIT(0)
+
 #define IXGBE_TX_FLAGS_CSUM		BIT(0)
 #define IXGBE_TX_FLAGS_VLAN		BIT(1)
 #define IXGBE_TX_FLAGS_TSO		BIT(2)

drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c

@@ -568,6 +568,12 @@ static void ixgbevf_alloc_rx_buffers(struct ixgbevf_ring *rx_ring,
 		.truesize	= rx_ring->truesize,
 		.count		= rx_ring->count,
 	};
+	const struct libeth_fq_fp hdr_fq = {
+		.pp		= rx_ring->hdr_pp,
+		.fqes		= rx_ring->hdr_fqes,
+		.truesize	= rx_ring->hdr_truesize,
+		.count		= rx_ring->count,
+	};
 	u16 ntu = rx_ring->next_to_use;
 
 	/* nothing to do or no valid netdev defined */
@@ -584,6 +590,14 @@ static void ixgbevf_alloc_rx_buffers(struct ixgbevf_ring *rx_ring,
 
 		rx_desc->read.pkt_addr = cpu_to_le64(addr);
 
+		if (!hdr_fq.pp)
+			goto next;
+
+		addr = libeth_rx_alloc(&hdr_fq, ntu);
+		if (addr == DMA_MAPPING_ERROR)
+			return;
+
+next:
 		rx_desc++;
 		ntu++;
 		if (unlikely(ntu == fq.count)) {
@@ -781,6 +795,32 @@ static int ixgbevf_run_xdp(struct ixgbevf_adapter *adapter,
 	return result;
 }
 
+static u32 ixgbevf_rx_hsplit_wa(const struct libeth_fqe *hdr,
+				struct libeth_fqe *buf, u32 data_len)
+{
+	u32 copy = data_len <= L1_CACHE_BYTES ? data_len : ETH_HLEN;
+	struct page *hdr_page, *buf_page;
+	const void *src;
+	void *dst;
+
+	if (unlikely(netmem_is_net_iov(buf->netmem)) ||
+	    !libeth_rx_sync_for_cpu(buf, copy))
+		return 0;
+
+	hdr_page = __netmem_to_page(hdr->netmem);
+	buf_page = __netmem_to_page(buf->netmem);
+
+	dst = page_address(hdr_page) + hdr->offset +
+	      pp_page_to_nmdesc(hdr_page)->pp->p.offset;
+	src = page_address(buf_page) + buf->offset +
+	      pp_page_to_nmdesc(buf_page)->pp->p.offset;
+
+	memcpy(dst, src, LARGEST_ALIGN(copy));
+	buf->offset += copy;
+
+	return copy;
+}
+
 static int ixgbevf_clean_rx_irq(struct ixgbevf_q_vector *q_vector,
 				struct ixgbevf_ring *rx_ring,
 				int budget)
@@ -818,6 +858,23 @@ static int ixgbevf_clean_rx_irq(struct ixgbevf_q_vector *q_vector,
 		rmb();
 
 		rx_buffer = &rx_ring->rx_fqes[rx_ring->next_to_clean];
+
+		if (unlikely(rx_ring->hdr_pp)) {
+			struct libeth_fqe *hdr_buff;
+			unsigned int hdr_size = 0;
+
+			hdr_buff = &rx_ring->hdr_fqes[rx_ring->next_to_clean];
+
+			if (!xdp->data) {
+				hdr_size = ixgbevf_rx_hsplit_wa(hdr_buff,
+								rx_buffer,
+								size);
+				size -= hdr_size ? : size;
+			}
+
+			libeth_xdp_process_buff(xdp, hdr_buff, hdr_size);
+		}
+
 		libeth_xdp_process_buff(xdp, rx_buffer, size);
 
 		cleaned_count++;
@@ -3054,19 +3111,38 @@ static int ixgbevf_setup_all_tx_resources(struct ixgbevf_adapter *adapter)
 	return err;
 }
 
-/**
- * ixgbevf_setup_rx_resources - allocate Rx resources (Descriptors)
- * @adapter: board private structure
- * @rx_ring: Rx descriptor ring (for a specific queue) to setup
- *
- * Returns 0 on success, negative on failure
- **/
-int ixgbevf_setup_rx_resources(struct ixgbevf_adapter *adapter,
-			       struct ixgbevf_ring *rx_ring)
+static void ixgbvf_rx_destroy_pp(struct ixgbevf_ring *rx_ring)
 {
+	struct libeth_fq fq = {
+		.pp	= rx_ring->pp,
+		.fqes	= rx_ring->rx_fqes,
+	};
+
+	libeth_rx_fq_destroy(&fq);
+	rx_ring->rx_fqes = NULL;
+	rx_ring->pp = NULL;
+
+	if (!rx_ring->hdr_pp)
+		return;
+
+	fq = (struct libeth_fq) {
+		.pp	= rx_ring->hdr_pp,
+		.fqes	= rx_ring->hdr_fqes,
+	};
+
+	libeth_rx_fq_destroy(&fq);
+	rx_ring->hdr_fqes = NULL;
+	rx_ring->hdr_pp = NULL;
+}
+
+static int ixgbvf_rx_create_pp(struct ixgbevf_ring *rx_ring)
+{
+	u32 adapter_flags = rx_ring->q_vector->adapter->flags;
+
 	struct libeth_fq fq = {
 		.count		= rx_ring->count,
 		.nid		= NUMA_NO_NODE,
+		.hsplit		= adapter_flags & IXGBEVF_FLAG_HSPLIT,
 		.type		= LIBETH_FQE_MTU,
 		.xdp		= !!rx_ring->xdp_prog,
 		.buf_len	= IXGBEVF_RX_PAGE_LEN(rx_ring->xdp_prog ?
@@ -3084,34 +3160,75 @@ int ixgbevf_setup_rx_resources(struct ixgbevf_adapter *adapter,
 	rx_ring->truesize = fq.truesize;
 	rx_ring->rx_buf_len = fq.buf_len;
 
+	if (!fq.hsplit)
+		return 0;
+
+	fq = (struct libeth_fq) {
+		.count		= rx_ring->count,
+		.nid		= NUMA_NO_NODE,
+		.type		= LIBETH_FQE_HDR,
+		.xdp		= !!rx_ring->xdp_prog,
+	};
+
+	ret = libeth_rx_fq_create(&fq, &rx_ring->q_vector->napi);
+	if (ret)
+		goto err;
+
+	rx_ring->hdr_pp = fq.pp;
+	rx_ring->hdr_fqes = fq.fqes;
+	rx_ring->hdr_truesize = fq.truesize;
+	rx_ring->hdr_buf_len = fq.buf_len;
+
+	return 0;
+
+err:
+	ixgbvf_rx_destroy_pp(rx_ring);
+	return ret;
+}
+
+/**
+ * ixgbevf_setup_rx_resources - allocate Rx resources
+ * @adapter: board private structure
+ * @rx_ring: Rx descriptor ring (for a specific queue) to setup
+ *
+ * Returns: 0 on success, negative on failure.
+ **/
+int ixgbevf_setup_rx_resources(struct ixgbevf_adapter *adapter,
+			       struct ixgbevf_ring *rx_ring)
+{
+	int ret;
+
+	ret = ixgbvf_rx_create_pp(rx_ring);
+	if (ret)
+		return ret;
+
 	u64_stats_init(&rx_ring->syncp);
 
 	/* Round up to nearest 4K */
 	rx_ring->size = rx_ring->count * sizeof(union ixgbe_adv_rx_desc);
 	rx_ring->size = ALIGN(rx_ring->size, 4096);
 
-	rx_ring->desc = dma_alloc_coherent(fq.pp->p.dev, rx_ring->size,
+	rx_ring->desc = dma_alloc_coherent(rx_ring->pp->p.dev, rx_ring->size,
 					   &rx_ring->dma, GFP_KERNEL);
 
 	if (!rx_ring->desc)
 		goto err;
 
 	/* XDP RX-queue info */
 	ret = __xdp_rxq_info_reg(&rx_ring->xdp_rxq, adapter->netdev,
-				 rx_ring->queue_index, 0, fq.buf_len);
+				 rx_ring->queue_index, 0, rx_ring->rx_buf_len);
 	if (ret)
 		goto err;
 
-	xdp_rxq_info_attach_page_pool(&rx_ring->xdp_rxq, fq.pp);
+	xdp_rxq_info_attach_page_pool(&rx_ring->xdp_rxq, rx_ring->pp);
 
 	rx_ring->xdp_prog = adapter->xdp_prog;
 
 	return 0;
 err:
-	libeth_rx_fq_destroy(&fq);
-	rx_ring->rx_fqes = NULL;
-	rx_ring->pp = NULL;
+	ixgbvf_rx_destroy_pp(rx_ring);
 	dev_err(rx_ring->dev, "Unable to allocate memory for the Rx descriptor ring\n");
+
 	return ret;
 }
 
@@ -4222,6 +4339,9 @@ static int ixgbevf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 	netdev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_RX_SG;
 
+	if (adapter->hw.mac.type == ixgbe_mac_82599_vf)
+		adapter->flags |= IXGBEVF_FLAG_HSPLIT;
+
 	/* MTU range: 68 - 1504 or 9710 */
 	netdev->min_mtu = ETH_MIN_MTU;
 	switch (adapter->hw.api_version) {