[net-next 00/15] 1GbE Intel Wired LAN Driver Updates 2020-09-28

[net-next 00/15] 1GbE Intel Wired LAN Driver Updates 2020-09-28

[Tony Nguyen]

This series contains updates to igb, igc and e1000e drivers.

Sven Auhagen adds XDP support for igb.

Gal Hammer allows for 82576 to display part number string correctly for
igb.

Sasha adds device IDs for i221 and i226 parts. Exposes LPI counters and
removes unused fields in structures for igc. He also adds Meteor Lake
support for e1000e.

For igc, Andre renames IGC_TSYNCTXCTL_VALID to IGC_TSYNCTXCTL_TXTT_0 to
match the datasheet and adds a warning if it's not set when expected.
Removes the PTP Tx timestamp check in igc_ptp_tx_work() as it's already
checked in the watchdog_task. Cleans up some code by removing invalid error
bits, renaming a bit to match datasheet naming, and removing a, now
unneeded, macro.

Vinicius makes changes for igc PTP: removes calling SYSTIMR to latch timer
value, stores PTP time before a reset, and rejects schedules with times in
the future.

The following are changes since commit 0675c285ea65540cccae64c87dfc7a00c7ede03a:
  net: vlan: Fixed signedness in vlan_group_prealloc_vid()
and are available in the git repository at:
  https://github.com/anguy11/next-queue.git 1GbE

Andre Guedes (4):
  igc: Rename IGC_TSYNCTXCTL_VALID macro
  igc: Don't reschedule ptp_tx work
  igc: Remove timeout check from ptp_tx work
  igc: Clean RX descriptor error flags

Gal Hammer (1):
  igb: read PBA number from flash

Sasha Neftin (5):
  igc: Add new device ID's
  igc: Expose LPI counters
  igc: Remove reset disable flag
  igc: Clean up nvm_info structure
  e1000e: Add support for Meteor Lake

Sven Auhagen (1):
  igb: add XDP support

Vinicius Costa Gomes (4):
  igc: Remove references to SYSTIMR register
  igc: Save PTP time before a reset
  igc: Export a way to read the PTP timer
  igc: Reject schedules with a base_time in the future

 drivers/net/ethernet/intel/e1000e/ethtool.c  |   2 +
 drivers/net/ethernet/intel/e1000e/hw.h       |   5 +
 drivers/net/ethernet/intel/e1000e/ich8lan.c  |   7 +
 drivers/net/ethernet/intel/e1000e/netdev.c   |   6 +
 drivers/net/ethernet/intel/e1000e/ptp.c      |   1 +
 drivers/net/ethernet/intel/igb/igb.h         |  80 +++-
 drivers/net/ethernet/intel/igb/igb_ethtool.c |   4 +
 drivers/net/ethernet/intel/igb/igb_main.c    | 437 +++++++++++++++++--
 drivers/net/ethernet/intel/igc/igc.h         |   3 +
 drivers/net/ethernet/intel/igc/igc_base.c    |   5 +
 drivers/net/ethernet/intel/igc/igc_defines.h |  16 +-
 drivers/net/ethernet/intel/igc/igc_ethtool.c |   3 +
 drivers/net/ethernet/intel/igc/igc_hw.h      |  11 +-
 drivers/net/ethernet/intel/igc/igc_main.c    |  39 +-
 drivers/net/ethernet/intel/igc/igc_ptp.c     |  58 +--
 15 files changed, 591 insertions(+), 86 deletions(-)

-- 
2.26.2

[net-next 01/15] igb: add XDP support

[Tony Nguyen]

From: Sven Auhagen <sven.auhagen@voleatech.de>

Add XDP support to the IGB driver.
The implementation follows the IXGBE XDP implementation
closely and I used the following patches as basis:

1. commit 924708081629 ("ixgbe: add XDP support for pass and drop actions")
2. commit 33fdc82f0883 ("ixgbe: add support for XDP_TX action")
3. commit ed93a3987128 ("ixgbe: tweak page counting for XDP_REDIRECT")

Due to the hardware constraints of the devices using the
IGB driver we must share the TX queues with XDP which
means locking the TX queue for XDP.

I ran tests on an older device to get better numbers.
Test machine:

Intel(R) Atom(TM) CPU C2338 @ 1.74GHz (2 Cores)
2x Intel I211

Routing Original Driver Network Stack: 382 Kpps

Routing XDP Redirect (xdp_fwd_kern): 1.48 Mpps
XDP Drop: 1.48 Mpps

Using XDP we can achieve line rate forwarding even on
an older Intel Atom CPU.

Signed-off-by: Sven Auhagen <sven.auhagen@voleatech.de>
Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igb/igb.h         |  80 +++-
 drivers/net/ethernet/intel/igb/igb_ethtool.c |   4 +
 drivers/net/ethernet/intel/igb/igb_main.c    | 433 +++++++++++++++++--
 3 files changed, 481 insertions(+), 36 deletions(-)

diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h
index 2f015b60a995..0286d2fceee4 100644
--- a/drivers/net/ethernet/intel/igb/igb.h
+++ b/drivers/net/ethernet/intel/igb/igb.h
@@ -19,6 +19,8 @@
 #include <linux/pci.h>
 #include <linux/mdio.h>
 
+#include <net/xdp.h>
+
 struct igb_adapter;
 
 #define E1000_PCS_CFG_IGN_SD	1
@@ -79,6 +81,12 @@ struct igb_adapter;
 #define IGB_I210_RX_LATENCY_100		2213
 #define IGB_I210_RX_LATENCY_1000	448
 
+/* XDP */
+#define IGB_XDP_PASS		0
+#define IGB_XDP_CONSUMED	BIT(0)
+#define IGB_XDP_TX		BIT(1)
+#define IGB_XDP_REDIR		BIT(2)
+
 struct vf_data_storage {
 	unsigned char vf_mac_addresses[ETH_ALEN];
 	u16 vf_mc_hashes[IGB_MAX_VF_MC_ENTRIES];
@@ -132,17 +140,62 @@ struct vf_mac_filter {
 
 /* Supported Rx Buffer Sizes */
 #define IGB_RXBUFFER_256	256
+#define IGB_RXBUFFER_1536	1536
 #define IGB_RXBUFFER_2048	2048
 #define IGB_RXBUFFER_3072	3072
 #define IGB_RX_HDR_LEN		IGB_RXBUFFER_256
 #define IGB_TS_HDR_LEN		16
 
-#define IGB_SKB_PAD		(NET_SKB_PAD + NET_IP_ALIGN)
+/* Attempt to maximize the headroom available for incoming frames.  We
+ * use a 2K buffer for receives 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 3K
+ *	 buffers.
+ */
 #if (PAGE_SIZE < 8192)
-#define IGB_MAX_FRAME_BUILD_SKB \
-	(SKB_WITH_OVERHEAD(IGB_RXBUFFER_2048) - IGB_SKB_PAD - IGB_TS_HDR_LEN)
+#define IGB_MAX_FRAME_BUILD_SKB (IGB_RXBUFFER_1536 - NET_IP_ALIGN)
+#define IGB_2K_TOO_SMALL_WITH_PADDING \
+((NET_SKB_PAD + IGB_TS_HDR_LEN + IGB_RXBUFFER_1536) > SKB_WITH_OVERHEAD(IGB_RXBUFFER_2048))
+
+static inline int igb_compute_pad(int rx_buf_len)
+{
+	int page_size, pad_size;
+
+	page_size = ALIGN(rx_buf_len, PAGE_SIZE / 2);
+	pad_size = SKB_WITH_OVERHEAD(page_size) - rx_buf_len;
+
+	return pad_size;
+}
+
+static inline int igb_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 (IGB_2K_TOO_SMALL_WITH_PADDING)
+		rx_buf_len = IGB_RXBUFFER_3072 + SKB_DATA_ALIGN(NET_IP_ALIGN);
+	else
+		rx_buf_len = IGB_RXBUFFER_1536;
+
+	/* if needed make room for NET_IP_ALIGN */
+	rx_buf_len -= NET_IP_ALIGN;
+
+	return igb_compute_pad(rx_buf_len);
+}
+
+#define IGB_SKB_PAD	igb_skb_pad()
 #else
-#define IGB_MAX_FRAME_BUILD_SKB (IGB_RXBUFFER_2048 - IGB_TS_HDR_LEN)
+#define IGB_SKB_PAD	(NET_SKB_PAD + NET_IP_ALIGN)
 #endif
 
 /* How many Rx Buffers do we bundle into one write to the hardware ? */
@@ -194,13 +247,22 @@ enum igb_tx_flags {
 #define IGB_SFF_ADDRESSING_MODE		0x4
 #define IGB_SFF_8472_UNSUP		0x00
 
+enum igb_tx_buf_type {
+	IGB_TYPE_SKB = 0,
+	IGB_TYPE_XDP,
+};
+
 /* wrapper around a pointer to a socket buffer,
  * so a DMA handle can be stored along with the buffer
  */
 struct igb_tx_buffer {
 	union e1000_adv_tx_desc *next_to_watch;
 	unsigned long time_stamp;
-	struct sk_buff *skb;
+	enum igb_tx_buf_type type;
+	union {
+		struct sk_buff *skb;
+		struct xdp_frame *xdpf;
+	};
 	unsigned int bytecount;
 	u16 gso_segs;
 	__be16 protocol;
@@ -248,6 +310,7 @@ struct igb_ring_container {
 struct igb_ring {
 	struct igb_q_vector *q_vector;	/* backlink to q_vector */
 	struct net_device *netdev;	/* back pointer to net_device */
+	struct bpf_prog *xdp_prog;
 	struct device *dev;		/* device pointer for dma mapping */
 	union {				/* array of buffer info structs */
 		struct igb_tx_buffer *tx_buffer_info;
@@ -288,6 +351,7 @@ struct igb_ring {
 			struct u64_stats_sync rx_syncp;
 		};
 	};
+	struct xdp_rxq_info xdp_rxq;
 } ____cacheline_internodealigned_in_smp;
 
 struct igb_q_vector {
@@ -339,7 +403,7 @@ static inline unsigned int igb_rx_bufsz(struct igb_ring *ring)
 		return IGB_RXBUFFER_3072;
 
 	if (ring_uses_build_skb(ring))
-		return IGB_MAX_FRAME_BUILD_SKB + IGB_TS_HDR_LEN;
+		return IGB_MAX_FRAME_BUILD_SKB;
 #endif
 	return IGB_RXBUFFER_2048;
 }
@@ -467,6 +531,7 @@ struct igb_adapter {
 	unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
 
 	struct net_device *netdev;
+	struct bpf_prog *xdp_prog;
 
 	unsigned long state;
 	unsigned int flags;
@@ -643,6 +708,9 @@ enum igb_boards {
 
 extern char igb_driver_name[];
 
+int igb_xmit_xdp_ring(struct igb_adapter *adapter,
+		      struct igb_ring *ring,
+		      struct xdp_frame *xdpf);
 int igb_open(struct net_device *netdev);
 int igb_close(struct net_device *netdev);
 int igb_up(struct igb_adapter *);
diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
index 6e8231c1ddf0..28baf203459a 100644
--- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
+++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
@@ -961,6 +961,10 @@ static int igb_set_ringparam(struct net_device *netdev,
 			memcpy(&temp_ring[i], adapter->rx_ring[i],
 			       sizeof(struct igb_ring));
 
+			/* Clear copied XDP RX-queue info */
+			memset(&temp_ring[i].xdp_rxq, 0,
+			       sizeof(temp_ring[i].xdp_rxq));
+
 			temp_ring[i].count = new_rx_count;
 			err = igb_setup_rx_resources(&temp_ring[i]);
 			if (err) {
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 44157fcd3cf7..73125d11460f 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -30,6 +30,8 @@
 #include <linux/if_ether.h>
 #include <linux/aer.h>
 #include <linux/prefetch.h>
+#include <linux/bpf.h>
+#include <linux/bpf_trace.h>
 #include <linux/pm_runtime.h>
 #include <linux/etherdevice.h>
 #ifdef CONFIG_IGB_DCA
@@ -2823,6 +2825,147 @@ static int igb_setup_tc(struct net_device *dev, enum tc_setup_type type,
 	}
 }
 
+static int igb_xdp_setup(struct net_device *dev, struct bpf_prog *prog)
+{
+	int i, frame_size = dev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
+	struct igb_adapter *adapter = netdev_priv(dev);
+	bool running = netif_running(dev);
+	struct bpf_prog *old_prog;
+	bool need_reset;
+
+	/* verify igb ring attributes are sufficient for XDP */
+	for (i = 0; i < adapter->num_rx_queues; i++) {
+		struct igb_ring *ring = adapter->rx_ring[i];
+
+		if (frame_size > igb_rx_bufsz(ring))
+			return -EINVAL;
+	}
+
+	old_prog = xchg(&adapter->xdp_prog, prog);
+	need_reset = (!!prog != !!old_prog);
+
+	/* device is up and bpf is added/removed, must setup the RX queues */
+	if (need_reset && running) {
+		igb_close(dev);
+	} else {
+		for (i = 0; i < adapter->num_rx_queues; i++)
+			(void)xchg(&adapter->rx_ring[i]->xdp_prog,
+			    adapter->xdp_prog);
+	}
+
+	if (old_prog)
+		bpf_prog_put(old_prog);
+
+	/* bpf is just replaced, RXQ and MTU are already setup */
+	if (!need_reset)
+		return 0;
+
+	if (running)
+		igb_open(dev);
+
+	return 0;
+}
+
+static int igb_xdp(struct net_device *dev, struct netdev_bpf *xdp)
+{
+	switch (xdp->command) {
+	case XDP_SETUP_PROG:
+		return igb_xdp_setup(dev, xdp->prog);
+	default:
+		return -EINVAL;
+	}
+}
+
+static void igb_xdp_ring_update_tail(struct igb_ring *ring)
+{
+	/* Force memory writes to complete before letting h/w know there
+	 * are new descriptors to fetch.
+	 */
+	wmb();
+	writel(ring->next_to_use, ring->tail);
+}
+
+static inline struct igb_ring *igb_xdp_tx_queue_mapping(struct igb_adapter *adapter)
+{
+	unsigned int r_idx = smp_processor_id();
+
+	if (r_idx >= adapter->num_tx_queues)
+		r_idx = r_idx % adapter->num_tx_queues;
+
+	return adapter->tx_ring[r_idx];
+}
+
+static int igb_xdp_xmit_back(struct igb_adapter *adapter, struct xdp_buff *xdp)
+{
+	struct xdp_frame *xdpf = xdp_convert_buff_to_frame(xdp);
+	int cpu = smp_processor_id();
+	struct igb_ring *tx_ring;
+	struct netdev_queue *nq;
+	u32 ret;
+
+	if (unlikely(!xdpf))
+		return IGB_XDP_CONSUMED;
+
+	/* During program transitions its possible adapter->xdp_prog is assigned
+	 * but ring has not been configured yet. In this case simply abort xmit.
+	 */
+	tx_ring = adapter->xdp_prog ? igb_xdp_tx_queue_mapping(adapter) : NULL;
+	if (unlikely(!tx_ring))
+		return -ENXIO;
+
+	nq = txring_txq(tx_ring);
+	__netif_tx_lock(nq, cpu);
+	ret = igb_xmit_xdp_ring(adapter, tx_ring, xdpf);
+	__netif_tx_unlock(nq);
+
+	return ret;
+}
+
+static int igb_xdp_xmit(struct net_device *dev, int n,
+			struct xdp_frame **frames, u32 flags)
+{
+	struct igb_adapter *adapter = netdev_priv(dev);
+	int cpu = smp_processor_id();
+	struct igb_ring *tx_ring;
+	struct netdev_queue *nq;
+	int drops = 0;
+	int i;
+
+	if (unlikely(test_bit(__IGB_DOWN, &adapter->state)))
+		return -ENETDOWN;
+
+	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+		return -EINVAL;
+
+	/* During program transitions its possible adapter->xdp_prog is assigned
+	 * but ring has not been configured yet. In this case simply abort xmit.
+	 */
+	tx_ring = adapter->xdp_prog ? igb_xdp_tx_queue_mapping(adapter) : NULL;
+	if (unlikely(!tx_ring))
+		return -ENXIO;
+
+	nq = txring_txq(tx_ring);
+	__netif_tx_lock(nq, cpu);
+
+	for (i = 0; i < n; i++) {
+		struct xdp_frame *xdpf = frames[i];
+		int err;
+
+		err = igb_xmit_xdp_ring(adapter, tx_ring, xdpf);
+		if (err != IGB_XDP_TX) {
+			xdp_return_frame_rx_napi(xdpf);
+			drops++;
+		}
+	}
+
+	__netif_tx_unlock(nq);
+
+	if (unlikely(flags & XDP_XMIT_FLUSH))
+		igb_xdp_ring_update_tail(tx_ring);
+
+	return n - drops;
+}
+
 static const struct net_device_ops igb_netdev_ops = {
 	.ndo_open		= igb_open,
 	.ndo_stop		= igb_close,
@@ -2847,6 +2990,8 @@ static const struct net_device_ops igb_netdev_ops = {
 	.ndo_fdb_add		= igb_ndo_fdb_add,
 	.ndo_features_check	= igb_features_check,
 	.ndo_setup_tc		= igb_setup_tc,
+	.ndo_bpf		= igb_xdp,
+	.ndo_xdp_xmit		= igb_xdp_xmit,
 };
 
 /**
@@ -4179,6 +4324,7 @@ static void igb_configure_tx(struct igb_adapter *adapter)
  **/
 int igb_setup_rx_resources(struct igb_ring *rx_ring)
 {
+	struct igb_adapter *adapter = netdev_priv(rx_ring->netdev);
 	struct device *dev = rx_ring->dev;
 	int size;
 
@@ -4201,6 +4347,13 @@ int igb_setup_rx_resources(struct igb_ring *rx_ring)
 	rx_ring->next_to_clean = 0;
 	rx_ring->next_to_use = 0;
 
+	rx_ring->xdp_prog = adapter->xdp_prog;
+
+	/* XDP RX-queue info */
+	if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
+			     rx_ring->queue_index) < 0)
+		goto err;
+
 	return 0;
 
 err:
@@ -4505,6 +4658,10 @@ void igb_configure_rx_ring(struct igb_adapter *adapter,
 	int reg_idx = ring->reg_idx;
 	u32 rxdctl = 0;
 
+	xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq);
+	WARN_ON(xdp_rxq_info_reg_mem_model(&ring->xdp_rxq,
+					   MEM_TYPE_PAGE_SHARED, NULL));
+
 	/* disable the queue */
 	wr32(E1000_RXDCTL(reg_idx), 0);
 
@@ -4709,6 +4866,8 @@ void igb_free_rx_resources(struct igb_ring *rx_ring)
 {
 	igb_clean_rx_ring(rx_ring);
 
+	rx_ring->xdp_prog = NULL;
+	xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
 	vfree(rx_ring->rx_buffer_info);
 	rx_ring->rx_buffer_info = NULL;
 
@@ -6078,6 +6237,80 @@ static int igb_tx_map(struct igb_ring *tx_ring,
 	return -1;
 }
 
+int igb_xmit_xdp_ring(struct igb_adapter *adapter,
+		      struct igb_ring *tx_ring,
+		      struct xdp_frame *xdpf)
+{
+	union e1000_adv_tx_desc *tx_desc;
+	u32 len, cmd_type, olinfo_status;
+	struct igb_tx_buffer *tx_buffer;
+	dma_addr_t dma;
+	u16 i;
+
+	len = xdpf->len;
+
+	if (unlikely(!igb_desc_unused(tx_ring)))
+		return IGB_XDP_CONSUMED;
+
+	dma = dma_map_single(tx_ring->dev, xdpf->data, len, DMA_TO_DEVICE);
+	if (dma_mapping_error(tx_ring->dev, dma))
+		return IGB_XDP_CONSUMED;
+
+	/* record the location of the first descriptor for this packet */
+	tx_buffer = &tx_ring->tx_buffer_info[tx_ring->next_to_use];
+	tx_buffer->bytecount = len;
+	tx_buffer->gso_segs = 1;
+	tx_buffer->protocol = 0;
+
+	i = tx_ring->next_to_use;
+	tx_desc = IGB_TX_DESC(tx_ring, i);
+
+	dma_unmap_len_set(tx_buffer, len, len);
+	dma_unmap_addr_set(tx_buffer, dma, dma);
+	tx_buffer->type = IGB_TYPE_XDP;
+	tx_buffer->xdpf = xdpf;
+
+	tx_desc->read.buffer_addr = cpu_to_le64(dma);
+
+	/* put descriptor type bits */
+	cmd_type = E1000_ADVTXD_DTYP_DATA |
+		   E1000_ADVTXD_DCMD_DEXT |
+		   E1000_ADVTXD_DCMD_IFCS;
+	cmd_type |= len | IGB_TXD_DCMD;
+	tx_desc->read.cmd_type_len = cpu_to_le32(cmd_type);
+
+	olinfo_status = cpu_to_le32(len << E1000_ADVTXD_PAYLEN_SHIFT);
+	/* 82575 requires a unique index per ring */
+	if (test_bit(IGB_RING_FLAG_TX_CTX_IDX, &tx_ring->flags))
+		olinfo_status |= tx_ring->reg_idx << 4;
+
+	tx_desc->read.olinfo_status = olinfo_status;
+
+	netdev_tx_sent_queue(txring_txq(tx_ring), tx_buffer->bytecount);
+
+	/* set the timestamp */
+	tx_buffer->time_stamp = jiffies;
+
+	/* Avoid any potential race with xdp_xmit and cleanup */
+	smp_wmb();
+
+	/* set next_to_watch value indicating a packet is present */
+	i++;
+	if (i == tx_ring->count)
+		i = 0;
+
+	tx_buffer->next_to_watch = tx_desc;
+	tx_ring->next_to_use = i;
+
+	/* Make sure there is space in the ring for the next send. */
+	igb_maybe_stop_tx(tx_ring, DESC_NEEDED);
+
+	if (netif_xmit_stopped(txring_txq(tx_ring)) || !netdev_xmit_more())
+		writel(i, tx_ring->tail);
+
+	return IGB_XDP_TX;
+}
+
 netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb,
 				struct igb_ring *tx_ring)
 {
@@ -6106,6 +6339,7 @@ netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb,
 
 	/* record the location of the first descriptor for this packet */
 	first = &tx_ring->tx_buffer_info[tx_ring->next_to_use];
+	first->type = IGB_TYPE_SKB;
 	first->skb = skb;
 	first->bytecount = skb->len;
 	first->gso_segs = 1;
@@ -6258,6 +6492,19 @@ static int igb_change_mtu(struct net_device *netdev, int new_mtu)
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
 
+	if (adapter->xdp_prog) {
+		int i;
+
+		for (i = 0; i < adapter->num_rx_queues; i++) {
+			struct igb_ring *ring = adapter->rx_ring[i];
+
+			if (max_frame > igb_rx_bufsz(ring)) {
+				netdev_warn(adapter->netdev, "Requested MTU size is not supported with XDP\n");
+				return -EINVAL;
+			}
+		}
+	}
+
 	/* adjust max frame to be at least the size of a standard frame */
 	if (max_frame < (ETH_FRAME_LEN + ETH_FCS_LEN))
 		max_frame = ETH_FRAME_LEN + ETH_FCS_LEN;
@@ -7811,7 +8058,10 @@ static bool igb_clean_tx_irq(struct igb_q_vector *q_vector, int napi_budget)
 		total_packets += tx_buffer->gso_segs;
 
 		/* free the skb */
-		napi_consume_skb(tx_buffer->skb, napi_budget);
+		if (tx_buffer->type == IGB_TYPE_SKB)
+			napi_consume_skb(tx_buffer->skb, napi_budget);
+		else
+			xdp_return_frame(tx_buffer->xdpf);
 
 		/* unmap skb header data */
 		dma_unmap_single(tx_ring->dev,
@@ -7995,8 +8245,8 @@ static bool igb_can_reuse_rx_page(struct igb_rx_buffer *rx_buffer)
 	 * the pagecnt_bias and page count so that we fully restock the
 	 * number of references the driver holds.
 	 */
-	if (unlikely(!pagecnt_bias)) {
-		page_ref_add(page, USHRT_MAX);
+	if (unlikely(pagecnt_bias == 1)) {
+		page_ref_add(page, USHRT_MAX - 1);
 		rx_buffer->pagecnt_bias = USHRT_MAX;
 	}
 
@@ -8035,20 +8285,21 @@ static void igb_add_rx_frag(struct igb_ring *rx_ring,
 
 static struct sk_buff *igb_construct_skb(struct igb_ring *rx_ring,
 					 struct igb_rx_buffer *rx_buffer,
-					 union e1000_adv_rx_desc *rx_desc,
-					 unsigned int size)
+					 struct xdp_buff *xdp,
+					 union e1000_adv_rx_desc *rx_desc)
 {
-	void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
 #if (PAGE_SIZE < 8192)
 	unsigned int truesize = igb_rx_pg_size(rx_ring) / 2;
 #else
-	unsigned int truesize = SKB_DATA_ALIGN(size);
+	unsigned int truesize = SKB_DATA_ALIGN(xdp->data_end -
+					       xdp->data_hard_start);
 #endif
+	unsigned int size = xdp->data_end - xdp->data;
 	unsigned int headlen;
 	struct sk_buff *skb;
 
 	/* prefetch first cache line of first page */
-	net_prefetch(va);
+	net_prefetch(xdp->data);
 
 	/* allocate a skb to store the frags */
 	skb = napi_alloc_skb(&rx_ring->q_vector->napi, IGB_RX_HDR_LEN);
@@ -8056,24 +8307,24 @@ static struct sk_buff *igb_construct_skb(struct igb_ring *rx_ring,
 		return NULL;
 
 	if (unlikely(igb_test_staterr(rx_desc, E1000_RXDADV_STAT_TSIP))) {
-		igb_ptp_rx_pktstamp(rx_ring->q_vector, va, skb);
-		va += IGB_TS_HDR_LEN;
+		igb_ptp_rx_pktstamp(rx_ring->q_vector, xdp->data, skb);
+		xdp->data += IGB_TS_HDR_LEN;
 		size -= IGB_TS_HDR_LEN;
 	}
 
 	/* Determine available headroom for copy */
 	headlen = size;
 	if (headlen > IGB_RX_HDR_LEN)
-		headlen = eth_get_headlen(skb->dev, va, IGB_RX_HDR_LEN);
+		headlen = eth_get_headlen(skb->dev, xdp->data, IGB_RX_HDR_LEN);
 
 	/* align pull length to size of long to optimize memcpy performance */
-	memcpy(__skb_put(skb, headlen), va, ALIGN(headlen, sizeof(long)));
+	memcpy(__skb_put(skb, headlen), xdp->data, ALIGN(headlen, sizeof(long)));
 
 	/* update all of the pointers */
 	size -= headlen;
 	if (size) {
 		skb_add_rx_frag(skb, 0, rx_buffer->page,
-				(va + headlen) - page_address(rx_buffer->page),
+				(xdp->data + headlen) - page_address(rx_buffer->page),
 				size, truesize);
 #if (PAGE_SIZE < 8192)
 		rx_buffer->page_offset ^= truesize;
@@ -8089,29 +8340,29 @@ static struct sk_buff *igb_construct_skb(struct igb_ring *rx_ring,
 
 static struct sk_buff *igb_build_skb(struct igb_ring *rx_ring,
 				     struct igb_rx_buffer *rx_buffer,
-				     union e1000_adv_rx_desc *rx_desc,
-				     unsigned int size)
+				     struct xdp_buff *xdp,
+				     union e1000_adv_rx_desc *rx_desc)
 {
-	void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
 #if (PAGE_SIZE < 8192)
 	unsigned int truesize = igb_rx_pg_size(rx_ring) / 2;
 #else
 	unsigned int truesize = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) +
-				SKB_DATA_ALIGN(IGB_SKB_PAD + size);
+				SKB_DATA_ALIGN(xdp->data_end -
+					       xdp->data_hard_start);
 #endif
 	struct sk_buff *skb;
 
 	/* prefetch first cache line of first page */
-	net_prefetch(va);
+	net_prefetch(xdp->data_meta);
 
 	/* build an skb around the page buffer */
-	skb = build_skb(va - IGB_SKB_PAD, truesize);
+	skb = build_skb(xdp->data_hard_start, truesize);
 	if (unlikely(!skb))
 		return NULL;
 
 	/* update pointers within the skb to store the data */
-	skb_reserve(skb, IGB_SKB_PAD);
-	__skb_put(skb, size);
+	skb_reserve(skb, xdp->data - xdp->data_hard_start);
+	__skb_put(skb, xdp->data_end - xdp->data);
 
 	/* pull timestamp out of packet data */
 	if (igb_test_staterr(rx_desc, E1000_RXDADV_STAT_TSIP)) {
@@ -8129,6 +8380,79 @@ static struct sk_buff *igb_build_skb(struct igb_ring *rx_ring,
 	return skb;
 }
 
+static struct sk_buff *igb_run_xdp(struct igb_adapter *adapter,
+				   struct igb_ring *rx_ring,
+				   struct xdp_buff *xdp)
+{
+	int err, result = IGB_XDP_PASS;
+	struct bpf_prog *xdp_prog;
+	u32 act;
+
+	rcu_read_lock();
+	xdp_prog = READ_ONCE(rx_ring->xdp_prog);
+
+	if (!xdp_prog)
+		goto xdp_out;
+
+	prefetchw(xdp->data_hard_start); /* xdp_frame write */
+
+	act = bpf_prog_run_xdp(xdp_prog, xdp);
+	switch (act) {
+	case XDP_PASS:
+		break;
+	case XDP_TX:
+		result = igb_xdp_xmit_back(adapter, xdp);
+		break;
+	case XDP_REDIRECT:
+		err = xdp_do_redirect(adapter->netdev, xdp, xdp_prog);
+		if (!err)
+			result = IGB_XDP_REDIR;
+		else
+			result = IGB_XDP_CONSUMED;
+		break;
+	default:
+		bpf_warn_invalid_xdp_action(act);
+		fallthrough;
+	case XDP_ABORTED:
+		trace_xdp_exception(rx_ring->netdev, xdp_prog, act);
+		fallthrough;
+	case XDP_DROP:
+		result = IGB_XDP_CONSUMED;
+		break;
+	}
+xdp_out:
+	rcu_read_unlock();
+	return ERR_PTR(-result);
+}
+
+static unsigned int igb_rx_frame_truesize(struct igb_ring *rx_ring,
+					  unsigned int size)
+{
+	unsigned int truesize;
+
+#if (PAGE_SIZE < 8192)
+	truesize = igb_rx_pg_size(rx_ring) / 2; /* Must be power-of-2 */
+#else
+	truesize = ring_uses_build_skb(rx_ring) ?
+		SKB_DATA_ALIGN(IGB_SKB_PAD + size) +
+		SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) :
+		SKB_DATA_ALIGN(size);
+#endif
+	return truesize;
+}
+
+static void igb_rx_buffer_flip(struct igb_ring *rx_ring,
+			       struct igb_rx_buffer *rx_buffer,
+			       unsigned int size)
+{
+	unsigned int truesize = igb_rx_frame_truesize(rx_ring, size);
+#if (PAGE_SIZE < 8192)
+	rx_buffer->page_offset ^= truesize;
+#else
+	rx_buffer->page_offset += truesize;
+#endif
+}
+
 static inline void igb_rx_checksum(struct igb_ring *ring,
 				   union e1000_adv_rx_desc *rx_desc,
 				   struct sk_buff *skb)
@@ -8224,6 +8548,10 @@ static bool igb_cleanup_headers(struct igb_ring *rx_ring,
 				union e1000_adv_rx_desc *rx_desc,
 				struct sk_buff *skb)
 {
+	/* XDP packets use error pointer so abort at this point */
+	if (IS_ERR(skb))
+		return true;
+
 	if (unlikely((igb_test_staterr(rx_desc,
 				       E1000_RXDEXT_ERR_FRAME_ERR_MASK)))) {
 		struct net_device *netdev = rx_ring->netdev;
@@ -8282,6 +8610,11 @@ static void igb_process_skb_fields(struct igb_ring *rx_ring,
 	skb->protocol = eth_type_trans(skb, rx_ring->netdev);
 }
 
+static inline unsigned int igb_rx_offset(struct igb_ring *rx_ring)
+{
+	return ring_uses_build_skb(rx_ring) ? IGB_SKB_PAD : 0;
+}
+
 static struct igb_rx_buffer *igb_get_rx_buffer(struct igb_ring *rx_ring,
 					       const unsigned int size)
 {
@@ -8325,10 +8658,20 @@ static void igb_put_rx_buffer(struct igb_ring *rx_ring,
 
 static int igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget)
 {
+	struct igb_adapter *adapter = q_vector->adapter;
 	struct igb_ring *rx_ring = q_vector->rx.ring;
 	struct sk_buff *skb = rx_ring->skb;
 	unsigned int total_bytes = 0, total_packets = 0;
 	u16 cleaned_count = igb_desc_unused(rx_ring);
+	unsigned int xdp_xmit = 0;
+	struct xdp_buff xdp;
+
+	xdp.rxq = &rx_ring->xdp_rxq;
+
+	/* Frame size depend on rx_ring setup when PAGE_SIZE=4K */
+#if (PAGE_SIZE < 8192)
+	xdp.frame_sz = igb_rx_frame_truesize(rx_ring, 0);
+#endif
 
 	while (likely(total_packets < budget)) {
 		union e1000_adv_rx_desc *rx_desc;
@@ -8355,13 +8698,38 @@ static int igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget)
 		rx_buffer = igb_get_rx_buffer(rx_ring, size);
 
 		/* retrieve a buffer from the ring */
-		if (skb)
+		if (!skb) {
+			xdp.data = page_address(rx_buffer->page) +
+				   rx_buffer->page_offset;
+			xdp.data_meta = xdp.data;
+			xdp.data_hard_start = xdp.data -
+					      igb_rx_offset(rx_ring);
+			xdp.data_end = xdp.data + size;
+#if (PAGE_SIZE > 4096)
+			/* At larger PAGE_SIZE, frame_sz depend on len size */
+			xdp.frame_sz = igb_rx_frame_truesize(rx_ring, size);
+#endif
+			skb = igb_run_xdp(adapter, rx_ring, &xdp);
+		}
+
+		if (IS_ERR(skb)) {
+			unsigned int xdp_res = -PTR_ERR(skb);
+
+			if (xdp_res & (IGB_XDP_TX | IGB_XDP_REDIR)) {
+				xdp_xmit |= xdp_res;
+				igb_rx_buffer_flip(rx_ring, rx_buffer, size);
+			} else {
+				rx_buffer->pagecnt_bias++;
+			}
+			total_packets++;
+			total_bytes += size;
+		} else if (skb)
 			igb_add_rx_frag(rx_ring, rx_buffer, skb, size);
 		else if (ring_uses_build_skb(rx_ring))
-			skb = igb_build_skb(rx_ring, rx_buffer, rx_desc, size);
+			skb = igb_build_skb(rx_ring, rx_buffer, &xdp, rx_desc);
 		else
 			skb = igb_construct_skb(rx_ring, rx_buffer,
-						rx_desc, size);
+						&xdp, rx_desc);
 
 		/* exit if we failed to retrieve a buffer */
 		if (!skb) {
@@ -8401,6 +8769,15 @@ static int igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget)
 	/* place incomplete frames back on ring for completion */
 	rx_ring->skb = skb;
 
+	if (xdp_xmit & IGB_XDP_REDIR)
+		xdp_do_flush_map();
+
+	if (xdp_xmit & IGB_XDP_TX) {
+		struct igb_ring *tx_ring = igb_xdp_tx_queue_mapping(adapter);
+
+		igb_xdp_ring_update_tail(tx_ring);
+	}
+
 	u64_stats_update_begin(&rx_ring->rx_syncp);
 	rx_ring->rx_stats.packets += total_packets;
 	rx_ring->rx_stats.bytes += total_bytes;
@@ -8414,11 +8791,6 @@ static int igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget)
 	return total_packets;
 }
 
-static inline unsigned int igb_rx_offset(struct igb_ring *rx_ring)
-{
-	return ring_uses_build_skb(rx_ring) ? IGB_SKB_PAD : 0;
-}
-
 static bool igb_alloc_mapped_page(struct igb_ring *rx_ring,
 				  struct igb_rx_buffer *bi)
 {
@@ -8455,7 +8827,8 @@ static bool igb_alloc_mapped_page(struct igb_ring *rx_ring,
 	bi->dma = dma;
 	bi->page = page;
 	bi->page_offset = igb_rx_offset(rx_ring);
-	bi->pagecnt_bias = 1;
+	page_ref_add(page, USHRT_MAX - 1);
+	bi->pagecnt_bias = USHRT_MAX;
 
 	return true;
 }
-- 
2.26.2

Re: [net-next 01/15] igb: add XDP support

[David Miller]

From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Mon, 28 Sep 2020 10:58:54 -0700

> From: Sven Auhagen <sven.auhagen@voleatech.de>
> 
> Add XDP support to the IGB driver.
> The implementation follows the IXGBE XDP implementation
> closely and I used the following patches as basis:
> 
> 1. commit 924708081629 ("ixgbe: add XDP support for pass and drop actions")
> 2. commit 33fdc82f0883 ("ixgbe: add support for XDP_TX action")
> 3. commit ed93a3987128 ("ixgbe: tweak page counting for XDP_REDIRECT")
> 
> Due to the hardware constraints of the devices using the
> IGB driver we must share the TX queues with XDP which
> means locking the TX queue for XDP.
> 
> I ran tests on an older device to get better numbers.
> Test machine:
> 
> Intel(R) Atom(TM) CPU C2338 @ 1.74GHz (2 Cores)
> 2x Intel I211
> 
> Routing Original Driver Network Stack: 382 Kpps
> 
> Routing XDP Redirect (xdp_fwd_kern): 1.48 Mpps
> XDP Drop: 1.48 Mpps
> 
> Using XDP we can achieve line rate forwarding even on
> an older Intel Atom CPU.
> 
> Signed-off-by: Sven Auhagen <sven.auhagen@voleatech.de>
> Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
> ---
>  drivers/net/ethernet/intel/igb/igb.h         |  80 +++-
>  drivers/net/ethernet/intel/igb/igb_ethtool.c |   4 +
>  drivers/net/ethernet/intel/igb/igb_main.c    | 433 +++++++++++++++++--
>  3 files changed, 481 insertions(+), 36 deletions(-)
> 
> diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h
> index 2f015b60a995..0286d2fceee4 100644
> --- a/drivers/net/ethernet/intel/igb/igb.h
> +++ b/drivers/net/ethernet/intel/igb/igb.h
> @@ -19,6 +19,8 @@
>  #include <linux/pci.h>
>  #include <linux/mdio.h>
>  
> +#include <net/xdp.h>
> +
>  struct igb_adapter;
>  
>  #define E1000_PCS_CFG_IGN_SD	1
> @@ -79,6 +81,12 @@ struct igb_adapter;
>  #define IGB_I210_RX_LATENCY_100		2213
>  #define IGB_I210_RX_LATENCY_1000	448
>  
> +/* XDP */
> +#define IGB_XDP_PASS		0
> +#define IGB_XDP_CONSUMED	BIT(0)
> +#define IGB_XDP_TX		BIT(1)
> +#define IGB_XDP_REDIR		BIT(2)
> +
>  struct vf_data_storage {
>  	unsigned char vf_mac_addresses[ETH_ALEN];
>  	u16 vf_mc_hashes[IGB_MAX_VF_MC_ENTRIES];
> @@ -132,17 +140,62 @@ struct vf_mac_filter {
>  
>  /* Supported Rx Buffer Sizes */
>  #define IGB_RXBUFFER_256	256
> +#define IGB_RXBUFFER_1536	1536
>  #define IGB_RXBUFFER_2048	2048
>  #define IGB_RXBUFFER_3072	3072
>  #define IGB_RX_HDR_LEN		IGB_RXBUFFER_256
>  #define IGB_TS_HDR_LEN		16
>  
> -#define IGB_SKB_PAD		(NET_SKB_PAD + NET_IP_ALIGN)
> +/* Attempt to maximize the headroom available for incoming frames.  We
> + * use a 2K buffer for receives 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 3K
> + *	 buffers.
> + */
>  #if (PAGE_SIZE < 8192)
> -#define IGB_MAX_FRAME_BUILD_SKB \
> -	(SKB_WITH_OVERHEAD(IGB_RXBUFFER_2048) - IGB_SKB_PAD - IGB_TS_HDR_LEN)
> +#define IGB_MAX_FRAME_BUILD_SKB (IGB_RXBUFFER_1536 - NET_IP_ALIGN)
> +#define IGB_2K_TOO_SMALL_WITH_PADDING \
> +((NET_SKB_PAD + IGB_TS_HDR_LEN + IGB_RXBUFFER_1536) > SKB_WITH_OVERHEAD(IGB_RXBUFFER_2048))
> +
> +static inline int igb_compute_pad(int rx_buf_len)
> +{
> +	int page_size, pad_size;
> +
> +	page_size = ALIGN(rx_buf_len, PAGE_SIZE / 2);
> +	pad_size = SKB_WITH_OVERHEAD(page_size) - rx_buf_len;
> +
> +	return pad_size;
> +}
> +
> +static inline int igb_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 (IGB_2K_TOO_SMALL_WITH_PADDING)
> +		rx_buf_len = IGB_RXBUFFER_3072 + SKB_DATA_ALIGN(NET_IP_ALIGN);
> +	else
> +		rx_buf_len = IGB_RXBUFFER_1536;
> +
> +	/* if needed make room for NET_IP_ALIGN */
> +	rx_buf_len -= NET_IP_ALIGN;
> +
> +	return igb_compute_pad(rx_buf_len);
> +}
> +
> +#define IGB_SKB_PAD	igb_skb_pad()
>  #else
> -#define IGB_MAX_FRAME_BUILD_SKB (IGB_RXBUFFER_2048 - IGB_TS_HDR_LEN)
> +#define IGB_SKB_PAD	(NET_SKB_PAD + NET_IP_ALIGN)
>  #endif
>  
>  /* How many Rx Buffers do we bundle into one write to the hardware ? */
> @@ -194,13 +247,22 @@ enum igb_tx_flags {
>  #define IGB_SFF_ADDRESSING_MODE		0x4
>  #define IGB_SFF_8472_UNSUP		0x00
>  
> +enum igb_tx_buf_type {
> +	IGB_TYPE_SKB = 0,
> +	IGB_TYPE_XDP,
> +};
> +
>  /* wrapper around a pointer to a socket buffer,
>   * so a DMA handle can be stored along with the buffer
>   */
>  struct igb_tx_buffer {
>  	union e1000_adv_tx_desc *next_to_watch;
>  	unsigned long time_stamp;
> -	struct sk_buff *skb;
> +	enum igb_tx_buf_type type;
> +	union {
> +		struct sk_buff *skb;
> +		struct xdp_frame *xdpf;
> +	};
>  	unsigned int bytecount;
>  	u16 gso_segs;
>  	__be16 protocol;
> @@ -248,6 +310,7 @@ struct igb_ring_container {
>  struct igb_ring {
>  	struct igb_q_vector *q_vector;	/* backlink to q_vector */
>  	struct net_device *netdev;	/* back pointer to net_device */
> +	struct bpf_prog *xdp_prog;
>  	struct device *dev;		/* device pointer for dma mapping */
>  	union {				/* array of buffer info structs */
>  		struct igb_tx_buffer *tx_buffer_info;
> @@ -288,6 +351,7 @@ struct igb_ring {
>  			struct u64_stats_sync rx_syncp;
>  		};
>  	};
> +	struct xdp_rxq_info xdp_rxq;
>  } ____cacheline_internodealigned_in_smp;
>  
>  struct igb_q_vector {
> @@ -339,7 +403,7 @@ static inline unsigned int igb_rx_bufsz(struct igb_ring *ring)
>  		return IGB_RXBUFFER_3072;
>  
>  	if (ring_uses_build_skb(ring))
> -		return IGB_MAX_FRAME_BUILD_SKB + IGB_TS_HDR_LEN;
> +		return IGB_MAX_FRAME_BUILD_SKB;
>  #endif
>  	return IGB_RXBUFFER_2048;
>  }
> @@ -467,6 +531,7 @@ struct igb_adapter {
>  	unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
>  
>  	struct net_device *netdev;
> +	struct bpf_prog *xdp_prog;
>  
>  	unsigned long state;
>  	unsigned int flags;
> @@ -643,6 +708,9 @@ enum igb_boards {
>  
>  extern char igb_driver_name[];
>  
> +int igb_xmit_xdp_ring(struct igb_adapter *adapter,
> +		      struct igb_ring *ring,
> +		      struct xdp_frame *xdpf);
>  int igb_open(struct net_device *netdev);
>  int igb_close(struct net_device *netdev);
>  int igb_up(struct igb_adapter *);
> diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
> index 6e8231c1ddf0..28baf203459a 100644
> --- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
> +++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
> @@ -961,6 +961,10 @@ static int igb_set_ringparam(struct net_device *netdev,
>  			memcpy(&temp_ring[i], adapter->rx_ring[i],
>  			       sizeof(struct igb_ring));
>  
> +			/* Clear copied XDP RX-queue info */
> +			memset(&temp_ring[i].xdp_rxq, 0,
> +			       sizeof(temp_ring[i].xdp_rxq));
> +
>  			temp_ring[i].count = new_rx_count;
>  			err = igb_setup_rx_resources(&temp_ring[i]);
>  			if (err) {
> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> index 44157fcd3cf7..73125d11460f 100644
> --- a/drivers/net/ethernet/intel/igb/igb_main.c
> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
 ...
> +static inline struct igb_ring *igb_xdp_tx_queue_mapping(struct igb_adapter *adapter)
> +{

Please do not use inline in foo.c files, let the compiler decide.

> @@ -8282,6 +8610,11 @@ static void igb_process_skb_fields(struct igb_ring *rx_ring,
>  	skb->protocol = eth_type_trans(skb, rx_ring->netdev);
>  }
>  
> +static inline unsigned int igb_rx_offset(struct igb_ring *rx_ring)

Likewise.

I know this function is just being moved, but nevertheless the inline tag should
be removed.

Thanks.

Re: [net-next 01/15] igb: add XDP support

[Jakub Kicinski]

On Mon, 28 Sep 2020 10:58:54 -0700 Tony Nguyen wrote:
> From: Sven Auhagen <sven.auhagen@voleatech.de>
> 
> Add XDP support to the IGB driver.
> The implementation follows the IXGBE XDP implementation
> closely and I used the following patches as basis:
> 
> 1. commit 924708081629 ("ixgbe: add XDP support for pass and drop actions")
> 2. commit 33fdc82f0883 ("ixgbe: add support for XDP_TX action")
> 3. commit ed93a3987128 ("ixgbe: tweak page counting for XDP_REDIRECT")
> 
> Due to the hardware constraints of the devices using the
> IGB driver we must share the TX queues with XDP which
> means locking the TX queue for XDP.
> 
> I ran tests on an older device to get better numbers.
> Test machine:
> 
> Intel(R) Atom(TM) CPU C2338 @ 1.74GHz (2 Cores)
> 2x Intel I211
> 
> Routing Original Driver Network Stack: 382 Kpps
> 
> Routing XDP Redirect (xdp_fwd_kern): 1.48 Mpps
> XDP Drop: 1.48 Mpps
> 
> Using XDP we can achieve line rate forwarding even on
> an older Intel Atom CPU.
> 
> Signed-off-by: Sven Auhagen <sven.auhagen@voleatech.de>
> Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com>
> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

Also some new sparse warnings added here:

drivers/net/ethernet/intel/igb/igb_main.c:6282:23: warning: incorrect type in assignment (different base types)
drivers/net/ethernet/intel/igb/igb_main.c:6282:23:    expected unsigned int [usertype] olinfo_status
drivers/net/ethernet/intel/igb/igb_main.c:6282:23:    got restricted __le32 [usertype]
drivers/net/ethernet/intel/igb/igb_main.c:6287:37: warning: incorrect type in assignment (different base types)
drivers/net/ethernet/intel/igb/igb_main.c:6287:37:    expected restricted __le32 [usertype] olinfo_status
drivers/net/ethernet/intel/igb/igb_main.c:6287:37:    got unsigned int [assigned] [usertype] olinfo_status

[net-next 02/15] igb: read PBA number from flash

[Tony Nguyen]

From: Gal Hammer <ghammer@redhat.com>

Fixed flash presence check for 82576 controllers so the part
number string is read and displayed correctly.

Signed-off-by: Gal Hammer <ghammer@redhat.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igb/igb_main.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 73125d11460f..d5c65850826e 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -3532,7 +3532,9 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			  "Width x1" : "unknown"), netdev->dev_addr);
 	}
 
-	if ((hw->mac.type >= e1000_i210 ||
+	if ((hw->mac.type == e1000_82576 &&
+	     rd32(E1000_EECD) & E1000_EECD_PRES) ||
+	    (hw->mac.type >= e1000_i210 ||
 	     igb_get_flash_presence_i210(hw))) {
 		ret_val = igb_read_part_string(hw, part_str,
 					       E1000_PBANUM_LENGTH);
-- 
2.26.2

[net-next 03/15] igc: Add new device ID's

[Tony Nguyen]

From: Sasha Neftin <sasha.neftin@intel.com>

Add new device ID's for the next step of the silicon and
reflect i221 and i226 parts

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_base.c | 5 +++++
 drivers/net/ethernet/intel/igc/igc_hw.h   | 5 +++++
 drivers/net/ethernet/intel/igc/igc_main.c | 5 +++++
 3 files changed, 15 insertions(+)

diff --git a/drivers/net/ethernet/intel/igc/igc_base.c b/drivers/net/ethernet/intel/igc/igc_base.c
index cc5a6cf531c7..fd37d2c203af 100644
--- a/drivers/net/ethernet/intel/igc/igc_base.c
+++ b/drivers/net/ethernet/intel/igc/igc_base.c
@@ -215,6 +215,11 @@ static s32 igc_get_invariants_base(struct igc_hw *hw)
 	case IGC_DEV_ID_I225_K2:
 	case IGC_DEV_ID_I225_LMVP:
 	case IGC_DEV_ID_I225_IT:
+	case IGC_DEV_ID_I226_LM:
+	case IGC_DEV_ID_I226_V:
+	case IGC_DEV_ID_I226_IT:
+	case IGC_DEV_ID_I221_V:
+	case IGC_DEV_ID_I226_BLANK_NVM:
 	case IGC_DEV_ID_I225_BLANK_NVM:
 		mac->type = igc_i225;
 		break;
diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h
index b9fe51b91c47..6defdb8a31fe 100644
--- a/drivers/net/ethernet/intel/igc/igc_hw.h
+++ b/drivers/net/ethernet/intel/igc/igc_hw.h
@@ -24,6 +24,11 @@
 #define IGC_DEV_ID_I225_K2			0x3101
 #define IGC_DEV_ID_I225_LMVP			0x5502
 #define IGC_DEV_ID_I225_IT			0x0D9F
+#define IGC_DEV_ID_I226_LM			0x125B
+#define IGC_DEV_ID_I226_V			0x125C
+#define IGC_DEV_ID_I226_IT			0x125D
+#define IGC_DEV_ID_I221_V			0x125E
+#define IGC_DEV_ID_I226_BLANK_NVM		0x125F
 #define IGC_DEV_ID_I225_BLANK_NVM		0x15FD
 
 /* Function pointers for the MAC. */
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 3183150c7995..b46bc8ded836 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -47,6 +47,11 @@ static const struct pci_device_id igc_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_K2), board_base },
 	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_LMVP), board_base },
 	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_IT), board_base },
+	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I226_LM), board_base },
+	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I226_V), board_base },
+	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I226_IT), board_base },
+	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I221_V), board_base },
+	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I226_BLANK_NVM), board_base },
 	{ PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_BLANK_NVM), board_base },
 	/* required last entry */
 	{0, }
-- 
2.26.2

[net-next 04/15] igc: Rename IGC_TSYNCTXCTL_VALID macro

[Tony Nguyen]

From: Andre Guedes <andre.guedes@intel.com>

Rename the IGC_TSYNCTXCTL_VALID macro to IGC_TSYNCTXCTL_TXTT_0 so it
matches the datasheet.

Signed-off-by: Andre Guedes <andre.guedes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_defines.h | 2 +-
 drivers/net/ethernet/intel/igc/igc_ptp.c     | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index f1f464967f87..3f5a201bda89 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -409,7 +409,7 @@
 #define IGC_IMIREXT_SIZE_BP	0x00001000  /* Packet size bypass */
 
 /* Time Sync Transmit Control bit definitions */
-#define IGC_TSYNCTXCTL_VALID			0x00000001  /* Tx timestamp valid */
+#define IGC_TSYNCTXCTL_TXTT_0			0x00000001  /* Tx timestamp reg 0 valid */
 #define IGC_TSYNCTXCTL_ENABLED			0x00000010  /* enable Tx timestamping */
 #define IGC_TSYNCTXCTL_MAX_ALLOWED_DLY_MASK	0x0000F000  /* max delay */
 #define IGC_TSYNCTXCTL_SYNC_COMP_ERR		0x20000000  /* sync err */
diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index e4b8f312f97c..dbe0776a7f2f 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -429,7 +429,7 @@ static void igc_ptp_tx_work(struct work_struct *work)
 	}
 
 	tsynctxctl = rd32(IGC_TSYNCTXCTL);
-	if (tsynctxctl & IGC_TSYNCTXCTL_VALID)
+	if (tsynctxctl & IGC_TSYNCTXCTL_TXTT_0)
 		igc_ptp_tx_hwtstamp(adapter);
 	else
 		/* reschedule to check later */
-- 
2.26.2

[net-next 05/15] igc: Don't reschedule ptp_tx work

[Tony Nguyen]

From: Andre Guedes <andre.guedes@intel.com>

The ptp_tx work is scheduled only if TSICR.TXTS bit is set, therefore
TSYNCTXCTL.TXTT_0 bit is expected to be set when we check it igc_ptp_tx_
work(). If it isn't, something is really off and rescheduling the ptp_tx
work to check it later doesn't help much. This patch changes the code to
WARN_ON_ONCE() if this situation ever happens.

Signed-off-by: Andre Guedes <andre.guedes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_ptp.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index dbe0776a7f2f..791f406f1314 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -429,11 +429,10 @@ static void igc_ptp_tx_work(struct work_struct *work)
 	}
 
 	tsynctxctl = rd32(IGC_TSYNCTXCTL);
-	if (tsynctxctl & IGC_TSYNCTXCTL_TXTT_0)
-		igc_ptp_tx_hwtstamp(adapter);
-	else
-		/* reschedule to check later */
-		schedule_work(&adapter->ptp_tx_work);
+	if (WARN_ON_ONCE(!(tsynctxctl & IGC_TSYNCTXCTL_TXTT_0)))
+		return;
+
+	igc_ptp_tx_hwtstamp(adapter);
 }
 
 /**
-- 
2.26.2

[net-next 06/15] igc: Remove timeout check from ptp_tx work

[Tony Nguyen]

From: Andre Guedes <andre.guedes@intel.com>

The Tx timestamp timeout is already checked by the watchdog_task
which runs periodically. In addition to that, from the ptp_tx work
perspective, if __IGC_PTP_TX_IN_PROGRESS flag is set we always want
handle the timestamp stored in hardware and update the skb. So remove
the timeout check in igc_ptp_tx_work() function.

Signed-off-by: Andre Guedes <andre.guedes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_ptp.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index 791f406f1314..61852c99815d 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -422,12 +422,6 @@ static void igc_ptp_tx_work(struct work_struct *work)
 	if (!test_bit(__IGC_PTP_TX_IN_PROGRESS, &adapter->state))
 		return;
 
-	if (time_is_before_jiffies(adapter->ptp_tx_start +
-				   IGC_PTP_TX_TIMEOUT)) {
-		igc_ptp_tx_timeout(adapter);
-		return;
-	}
-
 	tsynctxctl = rd32(IGC_TSYNCTXCTL);
 	if (WARN_ON_ONCE(!(tsynctxctl & IGC_TSYNCTXCTL_TXTT_0)))
 		return;
-- 
2.26.2

[net-next 07/15] igc: Clean RX descriptor error flags

[Tony Nguyen]

From: Andre Guedes <andre.guedes@intel.com>

i225 advanced receive descriptor doesn't have the following extend error
bits: CE, SE, SEQ, CXE. In addition to that, the bit TCPE is called L4E
in the datasheet.

Clean up the code accordingly, and get rid of the macro
IGC_RXDEXT_ERR_FRAME_ERR_MASK since it doesn't make much sense anymore.

Signed-off-by: Andre Guedes <andre.guedes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_defines.h | 14 +-------------
 drivers/net/ethernet/intel/igc/igc_main.c    |  5 ++---
 2 files changed, 3 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 3f5a201bda89..32f5fd684139 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -324,22 +324,10 @@
 /* Advanced Receive Descriptor bit definitions */
 #define IGC_RXDADV_STAT_TSIP	0x08000 /* timestamp in packet */
 
-#define IGC_RXDEXT_STATERR_CE		0x01000000
-#define IGC_RXDEXT_STATERR_SE		0x02000000
-#define IGC_RXDEXT_STATERR_SEQ		0x04000000
-#define IGC_RXDEXT_STATERR_CXE		0x10000000
-#define IGC_RXDEXT_STATERR_TCPE		0x20000000
+#define IGC_RXDEXT_STATERR_L4E		0x20000000
 #define IGC_RXDEXT_STATERR_IPE		0x40000000
 #define IGC_RXDEXT_STATERR_RXE		0x80000000
 
-/* Same mask, but for extended and packet split descriptors */
-#define IGC_RXDEXT_ERR_FRAME_ERR_MASK ( \
-	IGC_RXDEXT_STATERR_CE  |	\
-	IGC_RXDEXT_STATERR_SE  |	\
-	IGC_RXDEXT_STATERR_SEQ |	\
-	IGC_RXDEXT_STATERR_CXE |	\
-	IGC_RXDEXT_STATERR_RXE)
-
 #define IGC_MRQC_RSS_FIELD_IPV4_TCP	0x00010000
 #define IGC_MRQC_RSS_FIELD_IPV4		0x00020000
 #define IGC_MRQC_RSS_FIELD_IPV6_TCP_EX	0x00040000
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index b46bc8ded836..7a46b22413f2 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -1433,7 +1433,7 @@ static void igc_rx_checksum(struct igc_ring *ring,
 
 	/* TCP/UDP checksum error bit is set */
 	if (igc_test_staterr(rx_desc,
-			     IGC_RXDEXT_STATERR_TCPE |
+			     IGC_RXDEXT_STATERR_L4E |
 			     IGC_RXDEXT_STATERR_IPE)) {
 		/* work around errata with sctp packets where the TCPE aka
 		 * L4E bit is set incorrectly on 64 byte (60 byte w/o crc)
@@ -1742,8 +1742,7 @@ static bool igc_cleanup_headers(struct igc_ring *rx_ring,
 				union igc_adv_rx_desc *rx_desc,
 				struct sk_buff *skb)
 {
-	if (unlikely((igc_test_staterr(rx_desc,
-				       IGC_RXDEXT_ERR_FRAME_ERR_MASK)))) {
+	if (unlikely(igc_test_staterr(rx_desc, IGC_RXDEXT_STATERR_RXE))) {
 		struct net_device *netdev = rx_ring->netdev;
 
 		if (!(netdev->features & NETIF_F_RXALL)) {
-- 
2.26.2

[net-next 08/15] igc: Expose LPI counters

[Tony Nguyen]

From: Sasha Neftin <sasha.neftin@intel.com>

Completion to commit 900d1e8b346b ("igc: Add LPI counters")
LPI counters exposed by statistics update method.
A EEE TX LPI counter reflect the transmitter entries EEE (IEEE 802.3az)
into the LPI state. A EEE RX LPI counter reflect the receiver link
partner entries into EEE(IEEE 802.3az) LPI state.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_ethtool.c | 3 +++
 drivers/net/ethernet/intel/igc/igc_hw.h      | 2 ++
 drivers/net/ethernet/intel/igc/igc_main.c    | 2 ++
 3 files changed, 7 insertions(+)

diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c
index 44410c2265d6..61d331ce38cd 100644
--- a/drivers/net/ethernet/intel/igc/igc_ethtool.c
+++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c
@@ -321,6 +321,9 @@ static void igc_ethtool_get_regs(struct net_device *netdev,
 
 	for (i = 0; i < 8; i++)
 		regs_buff[205 + i] = rd32(IGC_ETQF(i));
+
+	regs_buff[213] = adapter->stats.tlpic;
+	regs_buff[214] = adapter->stats.rlpic;
 }
 
 static void igc_ethtool_get_wol(struct net_device *netdev,
diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h
index 6defdb8a31fe..b70253fb8ebc 100644
--- a/drivers/net/ethernet/intel/igc/igc_hw.h
+++ b/drivers/net/ethernet/intel/igc/igc_hw.h
@@ -244,6 +244,8 @@ struct igc_hw_stats {
 	u64 prc511;
 	u64 prc1023;
 	u64 prc1522;
+	u64 tlpic;
+	u64 rlpic;
 	u64 gprc;
 	u64 bprc;
 	u64 mprc;
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 7a46b22413f2..7576dbfdac99 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -3683,6 +3683,8 @@ void igc_update_stats(struct igc_adapter *adapter)
 	adapter->stats.prc511 += rd32(IGC_PRC511);
 	adapter->stats.prc1023 += rd32(IGC_PRC1023);
 	adapter->stats.prc1522 += rd32(IGC_PRC1522);
+	adapter->stats.tlpic += rd32(IGC_TLPIC);
+	adapter->stats.rlpic += rd32(IGC_RLPIC);
 
 	mpc = rd32(IGC_MPC);
 	adapter->stats.mpc += mpc;
-- 
2.26.2

[net-next 09/15] igc: Remove references to SYSTIMR register

[Tony Nguyen]

From: Vinicius Costa Gomes <vinicius.gomes@intel.com>

In i225, it's no longer necessary to use the SYSTIMR register to
latch the timer value, the timestamp is latched when SYSTIML is read.

Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_ptp.c | 12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index 61852c99815d..0300b45b36e2 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -22,11 +22,7 @@ static void igc_ptp_read_i225(struct igc_adapter *adapter,
 	struct igc_hw *hw = &adapter->hw;
 	u32 sec, nsec;
 
-	/* The timestamp latches on lowest register read. For I210/I211, the
-	 * lowest register is SYSTIMR. Since we only need to provide nanosecond
-	 * resolution, we can ignore it.
-	 */
-	rd32(IGC_SYSTIMR);
+	/* The timestamp is latched when SYSTIML is read. */
 	nsec = rd32(IGC_SYSTIML);
 	sec = rd32(IGC_SYSTIMH);
 
@@ -39,9 +35,6 @@ static void igc_ptp_write_i225(struct igc_adapter *adapter,
 {
 	struct igc_hw *hw = &adapter->hw;
 
-	/* Writing the SYSTIMR register is not necessary as it only
-	 * provides sub-nanosecond resolution.
-	 */
 	wr32(IGC_SYSTIML, ts->tv_nsec);
 	wr32(IGC_SYSTIMH, ts->tv_sec);
 }
@@ -102,10 +95,9 @@ static int igc_ptp_gettimex64_i225(struct ptp_clock_info *ptp,
 	spin_lock_irqsave(&igc->tmreg_lock, flags);
 
 	ptp_read_system_prets(sts);
-	rd32(IGC_SYSTIMR);
-	ptp_read_system_postts(sts);
 	ts->tv_nsec = rd32(IGC_SYSTIML);
 	ts->tv_sec = rd32(IGC_SYSTIMH);
+	ptp_read_system_postts(sts);
 
 	spin_unlock_irqrestore(&igc->tmreg_lock, flags);
 
-- 
2.26.2

[net-next 10/15] igc: Save PTP time before a reset

[Tony Nguyen]

From: Vinicius Costa Gomes <vinicius.gomes@intel.com>

Many TSN features depend on the internal PTP clock, so the internal
PTP jumping when the adapter is reset can cause problems, usually in
the form of "TX Hangs" warnings in the driver.

The solution is to save the PTP time before a reset and restore it
after the reset is done. The value of the PTP time is saved before a
reset and we use the difference from CLOCK_MONOTONIC from reset time
to now, to correct what's going to be the new PTP time.

This is heavily inspired by commit bf4bf09bdd91 ("i40e: save PTP time
before a device reset").

Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc.h      |  2 ++
 drivers/net/ethernet/intel/igc/igc_main.c |  2 ++
 drivers/net/ethernet/intel/igc/igc_ptp.c  | 28 ++++++++++++++++++++---
 3 files changed, 29 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index 2d566f3c827b..2e5720d34a16 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -215,6 +215,8 @@ struct igc_adapter {
 	spinlock_t tmreg_lock;
 	struct cyclecounter cc;
 	struct timecounter tc;
+	struct timespec64 prev_ptp_time; /* Pre-reset PTP clock */
+	ktime_t ptp_reset_start; /* Reset time in clock mono */
 };
 
 void igc_up(struct igc_adapter *adapter);
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 7576dbfdac99..1c16cd35c81c 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -3778,6 +3778,8 @@ void igc_down(struct igc_adapter *adapter)
 
 	set_bit(__IGC_DOWN, &adapter->state);
 
+	igc_ptp_suspend(adapter);
+
 	/* disable receives in the hardware */
 	rctl = rd32(IGC_RCTL);
 	wr32(IGC_RCTL, rctl & ~IGC_RCTL_EN);
diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index 0300b45b36e2..49abefdab26c 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -8,6 +8,7 @@
 #include <linux/pci.h>
 #include <linux/ptp_classify.h>
 #include <linux/clocksource.h>
+#include <linux/ktime.h>
 
 #define INCVALUE_MASK		0x7fffffff
 #define ISGN			0x80000000
@@ -500,6 +501,9 @@ void igc_ptp_init(struct igc_adapter *adapter)
 	adapter->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
 	adapter->tstamp_config.tx_type = HWTSTAMP_TX_OFF;
 
+	adapter->prev_ptp_time = ktime_to_timespec64(ktime_get_real());
+	adapter->ptp_reset_start = ktime_get();
+
 	adapter->ptp_clock = ptp_clock_register(&adapter->ptp_caps,
 						&adapter->pdev->dev);
 	if (IS_ERR(adapter->ptp_clock)) {
@@ -511,6 +515,24 @@ void igc_ptp_init(struct igc_adapter *adapter)
 	}
 }
 
+static void igc_ptp_time_save(struct igc_adapter *adapter)
+{
+	igc_ptp_read_i225(adapter, &adapter->prev_ptp_time);
+	adapter->ptp_reset_start = ktime_get();
+}
+
+static void igc_ptp_time_restore(struct igc_adapter *adapter)
+{
+	struct timespec64 ts = adapter->prev_ptp_time;
+	ktime_t delta;
+
+	delta = ktime_sub(ktime_get(), adapter->ptp_reset_start);
+
+	timespec64_add_ns(&ts, ktime_to_ns(delta));
+
+	igc_ptp_write_i225(adapter, &ts);
+}
+
 /**
  * igc_ptp_suspend - Disable PTP work items and prepare for suspend
  * @adapter: Board private structure
@@ -527,6 +549,8 @@ void igc_ptp_suspend(struct igc_adapter *adapter)
 	dev_kfree_skb_any(adapter->ptp_tx_skb);
 	adapter->ptp_tx_skb = NULL;
 	clear_bit_unlock(__IGC_PTP_TX_IN_PROGRESS, &adapter->state);
+
+	igc_ptp_time_save(adapter);
 }
 
 /**
@@ -576,9 +600,7 @@ void igc_ptp_reset(struct igc_adapter *adapter)
 
 	/* Re-initialize the timer. */
 	if (hw->mac.type == igc_i225) {
-		struct timespec64 ts64 = ktime_to_timespec64(ktime_get_real());
-
-		igc_ptp_write_i225(adapter, &ts64);
+		igc_ptp_time_restore(adapter);
 	} else {
 		timecounter_init(&adapter->tc, &adapter->cc,
 				 ktime_to_ns(ktime_get_real()));
-- 
2.26.2

[net-next 11/15] igc: Remove reset disable flag

[Tony Nguyen]

From: Sasha Neftin <sasha.neftin@intel.com>

Boolean reset disable flag not applicable for i225 device and
could be removed.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_hw.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h
index b70253fb8ebc..17d6669959db 100644
--- a/drivers/net/ethernet/intel/igc/igc_hw.h
+++ b/drivers/net/ethernet/intel/igc/igc_hw.h
@@ -158,7 +158,6 @@ struct igc_phy_info {
 	u8 mdix;
 
 	bool is_mdix;
-	bool reset_disable;
 	bool speed_downgraded;
 	bool autoneg_wait_to_complete;
 };
-- 
2.26.2

[net-next 12/15] igc: Export a way to read the PTP timer

[Tony Nguyen]

From: Vinicius Costa Gomes <vinicius.gomes@intel.com>

The next patch will need a way to retrieve the current timestamp from
the NIC's PTP clock.

The 'i225' suffix is removed, if anything model specific is needed,
those specifics should be hidden by this function.

Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc.h     | 1 +
 drivers/net/ethernet/intel/igc/igc_ptp.c | 7 +++----
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index 2e5720d34a16..35baae900c1f 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -550,6 +550,7 @@ void igc_ptp_rx_pktstamp(struct igc_q_vector *q_vector, void *va,
 int igc_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr);
 int igc_ptp_get_ts_config(struct net_device *netdev, struct ifreq *ifr);
 void igc_ptp_tx_hang(struct igc_adapter *adapter);
+void igc_ptp_read(struct igc_adapter *adapter, struct timespec64 *ts);
 
 #define igc_rx_pg_size(_ring) (PAGE_SIZE << igc_rx_pg_order(_ring))
 
diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index 49abefdab26c..ac0b9c85da7c 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -17,8 +17,7 @@
 #define IGC_PTP_TX_TIMEOUT		(HZ * 15)
 
 /* SYSTIM read access for I225 */
-static void igc_ptp_read_i225(struct igc_adapter *adapter,
-			      struct timespec64 *ts)
+void igc_ptp_read(struct igc_adapter *adapter, struct timespec64 *ts)
 {
 	struct igc_hw *hw = &adapter->hw;
 	u32 sec, nsec;
@@ -75,7 +74,7 @@ static int igc_ptp_adjtime_i225(struct ptp_clock_info *ptp, s64 delta)
 
 	spin_lock_irqsave(&igc->tmreg_lock, flags);
 
-	igc_ptp_read_i225(igc, &now);
+	igc_ptp_read(igc, &now);
 	now = timespec64_add(now, then);
 	igc_ptp_write_i225(igc, (const struct timespec64 *)&now);
 
@@ -517,7 +516,7 @@ void igc_ptp_init(struct igc_adapter *adapter)
 
 static void igc_ptp_time_save(struct igc_adapter *adapter)
 {
-	igc_ptp_read_i225(adapter, &adapter->prev_ptp_time);
+	igc_ptp_read(adapter, &adapter->prev_ptp_time);
 	adapter->ptp_reset_start = ktime_get();
 }
 
-- 
2.26.2

[net-next 13/15] igc: Reject schedules with a base_time in the future

[Tony Nguyen]

From: Vinicius Costa Gomes <vinicius.gomes@intel.com>

When we set the BASET registers of i225 with a base_time in the
future, i225 will "hold" all packets until that base_time is reached,
causing a lot of TX Hangs.

As this behaviour seems contrary to the expectations of the IEEE
802.1Q standard (section 8.6.9, especially 8.6.9.4.5), let's start by
rejecting these types of schedules. If this is too limiting, we can
for example, setup a timer to configure the BASET registers closer to
the start time, only blocking the packets for a "short" while.

Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_main.c | 25 +++++++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 1c16cd35c81c..569747bbefd8 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -4702,14 +4702,35 @@ static int igc_save_launchtime_params(struct igc_adapter *adapter, int queue,
 	return 0;
 }
 
-static bool validate_schedule(const struct tc_taprio_qopt_offload *qopt)
+static bool is_base_time_past(ktime_t base_time, const struct timespec64 *now)
+{
+	struct timespec64 b;
+
+	b = ktime_to_timespec64(base_time);
+
+	return timespec64_compare(now, &b) > 0;
+}
+
+static bool validate_schedule(struct igc_adapter *adapter,
+			      const struct tc_taprio_qopt_offload *qopt)
 {
 	int queue_uses[IGC_MAX_TX_QUEUES] = { };
+	struct timespec64 now;
 	size_t n;
 
 	if (qopt->cycle_time_extension)
 		return false;
 
+	igc_ptp_read(adapter, &now);
+
+	/* If we program the controller's BASET registers with a time
+	 * in the future, it will hold all the packets until that
+	 * time, causing a lot of TX Hangs, so to avoid that, we
+	 * reject schedules that would start in the future.
+	 */
+	if (!is_base_time_past(qopt->base_time, &now))
+		return false;
+
 	for (n = 0; n < qopt->num_entries; n++) {
 		const struct tc_taprio_sched_entry *e;
 		int i;
@@ -4764,7 +4785,7 @@ static int igc_save_qbv_schedule(struct igc_adapter *adapter,
 	if (adapter->base_time)
 		return -EALREADY;
 
-	if (!validate_schedule(qopt))
+	if (!validate_schedule(adapter, qopt))
 		return -EINVAL;
 
 	adapter->cycle_time = qopt->cycle_time;
-- 
2.26.2

[net-next 14/15] igc: Clean up nvm_info structure

[Tony Nguyen]

From: Sasha Neftin <sasha.neftin@intel.com>

flash_bank_size and flash_base_addr field not in use and can
be removed from a nvm_info structure

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc_hw.h | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h
index 17d6669959db..55dae7c4703f 100644
--- a/drivers/net/ethernet/intel/igc/igc_hw.h
+++ b/drivers/net/ethernet/intel/igc/igc_hw.h
@@ -130,9 +130,6 @@ struct igc_nvm_info {
 	struct igc_nvm_operations ops;
 	enum igc_nvm_type type;
 
-	u32 flash_bank_size;
-	u32 flash_base_addr;
-
 	u16 word_size;
 	u16 delay_usec;
 	u16 address_bits;
-- 
2.26.2

[net-next 15/15] e1000e: Add support for Meteor Lake

[Tony Nguyen]

From: Sasha Neftin <sasha.neftin@intel.com>

Add devices IDs for the next LOM generations that will be
available on the next Intel Client platform (Meteor Lake)
This patch provides the initial support for these devices

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/e1000e/ethtool.c | 2 ++
 drivers/net/ethernet/intel/e1000e/hw.h      | 5 +++++
 drivers/net/ethernet/intel/e1000e/ich8lan.c | 7 +++++++
 drivers/net/ethernet/intel/e1000e/netdev.c  | 6 ++++++
 drivers/net/ethernet/intel/e1000e/ptp.c     | 1 +
 5 files changed, 21 insertions(+)

diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index a8fc9208382c..03215b0aee4b 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -895,6 +895,7 @@ static int e1000_reg_test(struct e1000_adapter *adapter, u64 *data)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		mask |= BIT(18);
 		break;
 	default:
@@ -1560,6 +1561,7 @@ static void e1000_loopback_cleanup(struct e1000_adapter *adapter)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		fext_nvm11 = er32(FEXTNVM11);
 		fext_nvm11 &= ~E1000_FEXTNVM11_DISABLE_MULR_FIX;
 		ew32(FEXTNVM11, fext_nvm11);
diff --git a/drivers/net/ethernet/intel/e1000e/hw.h b/drivers/net/ethernet/intel/e1000e/hw.h
index b1447221669e..69a2329ea463 100644
--- a/drivers/net/ethernet/intel/e1000e/hw.h
+++ b/drivers/net/ethernet/intel/e1000e/hw.h
@@ -102,6 +102,10 @@ struct e1000_hw;
 #define E1000_DEV_ID_PCH_ADP_I219_V16		0x1A1F
 #define E1000_DEV_ID_PCH_ADP_I219_LM17		0x1A1C
 #define E1000_DEV_ID_PCH_ADP_I219_V17		0x1A1D
+#define E1000_DEV_ID_PCH_MTP_I219_LM18		0x550A
+#define E1000_DEV_ID_PCH_MTP_I219_V18		0x550B
+#define E1000_DEV_ID_PCH_MTP_I219_LM19		0x550C
+#define E1000_DEV_ID_PCH_MTP_I219_V19		0x550D
 
 #define E1000_REVISION_4	4
 
@@ -127,6 +131,7 @@ enum e1000_mac_type {
 	e1000_pch_cnp,
 	e1000_pch_tgp,
 	e1000_pch_adp,
+	e1000_pch_mtp,
 };
 
 enum e1000_media_type {
diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index ded74304e8cf..9aa6fad8ed47 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -320,6 +320,7 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		if (e1000_phy_is_accessible_pchlan(hw))
 			break;
 
@@ -464,6 +465,7 @@ static s32 e1000_init_phy_params_pchlan(struct e1000_hw *hw)
 		case e1000_pch_cnp:
 		case e1000_pch_tgp:
 		case e1000_pch_adp:
+		case e1000_pch_mtp:
 			/* In case the PHY needs to be in mdio slow mode,
 			 * set slow mode and try to get the PHY id again.
 			 */
@@ -708,6 +710,7 @@ static s32 e1000_init_mac_params_ich8lan(struct e1000_hw *hw)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 	case e1000_pchlan:
 		/* check management mode */
 		mac->ops.check_mng_mode = e1000_check_mng_mode_pchlan;
@@ -1648,6 +1651,7 @@ static s32 e1000_get_variants_ich8lan(struct e1000_adapter *adapter)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		rc = e1000_init_phy_params_pchlan(hw);
 		break;
 	default:
@@ -2102,6 +2106,7 @@ static s32 e1000_sw_lcd_config_ich8lan(struct e1000_hw *hw)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		sw_cfg_mask = E1000_FEXTNVM_SW_CONFIG_ICH8M;
 		break;
 	default:
@@ -3145,6 +3150,7 @@ static s32 e1000_valid_nvm_bank_detect_ich8lan(struct e1000_hw *hw, u32 *bank)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		bank1_offset = nvm->flash_bank_size;
 		act_offset = E1000_ICH_NVM_SIG_WORD;
 
@@ -4090,6 +4096,7 @@ static s32 e1000_validate_nvm_checksum_ich8lan(struct e1000_hw *hw)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		word = NVM_COMPAT;
 		valid_csum_mask = NVM_COMPAT_VALID_CSUM;
 		break;
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 99f4ec9b5696..b30f00891c03 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -3587,6 +3587,7 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		if (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI) {
 			/* Stable 24MHz frequency */
 			incperiod = INCPERIOD_24MHZ;
@@ -4104,6 +4105,7 @@ void e1000e_reset(struct e1000_adapter *adapter)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		fc->refresh_time = 0xFFFF;
 		fc->pause_time = 0xFFFF;
 
@@ -7877,6 +7879,10 @@ static const struct pci_device_id e1000_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V16), board_pch_cnp },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_LM17), board_pch_cnp },
 	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V17), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_LM18), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_V18), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_LM19), board_pch_cnp },
+	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_V19), board_pch_cnp },
 
 	{ 0, 0, 0, 0, 0, 0, 0 }	/* terminate list */
 };
diff --git a/drivers/net/ethernet/intel/e1000e/ptp.c b/drivers/net/ethernet/intel/e1000e/ptp.c
index 8d21bcb427ec..f3f671311855 100644
--- a/drivers/net/ethernet/intel/e1000e/ptp.c
+++ b/drivers/net/ethernet/intel/e1000e/ptp.c
@@ -297,6 +297,7 @@ void e1000e_ptp_init(struct e1000_adapter *adapter)
 	case e1000_pch_cnp:
 	case e1000_pch_tgp:
 	case e1000_pch_adp:
+	case e1000_pch_mtp:
 		if ((hw->mac.type < e1000_pch_lpt) ||
 		    (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI)) {
 			adapter->ptp_clock_info.max_adj = 24000000 - 1;
-- 
2.26.2