Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:

 1) New offloading infrastructure and example 'rocker' driver for
    offloading of switching and routing to hardware.

    This work was done by a large group of dedicated individuals, not
    limited to: Scott Feldman, Jiri Pirko, Thomas Graf, John Fastabend,
    Jamal Hadi Salim, Andy Gospodarek, Florian Fainelli, Roopa Prabhu

 2) Start making the networking operate on IOV iterators instead of
    modifying iov objects in-situ during transfers.  Thanks to Al Viro
    and Herbert Xu.

 3) A set of new netlink interfaces for the TIPC stack, from Richard
    Alpe.

 4) Remove unnecessary looping during ipv6 routing lookups, from Martin
    KaFai Lau.

 5) Add PAUSE frame generation support to gianfar driver, from Matei
    Pavaluca.

 6) Allow for larger reordering levels in TCP, which are easily
    achievable in the real world right now, from Eric Dumazet.

 7) Add a variable of napi_schedule that doesn't need to disable cpu
    interrupts, from Eric Dumazet.

 8) Use a doubly linked list to optimize neigh_parms_release(), from
    Nicolas Dichtel.

 9) Various enhancements to the kernel BPF verifier, and allow eBPF
    programs to actually be attached to sockets.  From Alexei
    Starovoitov.

10) Support TSO/LSO in sunvnet driver, from David L Stevens.

11) Allow controlling ECN usage via routing metrics, from Florian
    Westphal.

12) Remote checksum offload, from Tom Herbert.

13) Add split-header receive, BQL, and xmit_more support to amd-xgbe
    driver, from Thomas Lendacky.

14) Add MPLS support to openvswitch, from Simon Horman.

15) Support wildcard tunnel endpoints in ipv6 tunnels, from Steffen
    Klassert.

16) Do gro flushes on a per-device basis using a timer, from Eric
    Dumazet.  This tries to resolve the conflicting goals between the
    desired handling of bulk vs.  RPC-like traffic.

17) Allow userspace to ask for the CPU upon what a packet was
    received/steered, via SO_INCOMING_CPU.  From Eric Dumazet.

18) Limit GSO packets to half the current congestion window, from Eric
    Dumazet.

19) Add a generic helper so that all drivers set their RSS keys in a
    consistent way, from Eric Dumazet.

20) Add xmit_more support to enic driver, from Govindarajulu
    Varadarajan.

21) Add VLAN packet scheduler action, from Jiri Pirko.

22) Support configurable RSS hash functions via ethtool, from Eyal
    Perry.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1820 commits)
  Fix race condition between vxlan_sock_add and vxlan_sock_release
  net/macb: fix compilation warning for print_hex_dump() called with skb->mac_header
  net/mlx4: Add support for A0 steering
  net/mlx4: Refactor QUERY_PORT
  net/mlx4_core: Add explicit error message when rule doesn't meet configuration
  net/mlx4: Add A0 hybrid steering
  net/mlx4: Add mlx4_bitmap zone allocator
  net/mlx4: Add a check if there are too many reserved QPs
  net/mlx4: Change QP allocation scheme
  net/mlx4_core: Use tasklet for user-space CQ completion events
  net/mlx4_core: Mask out host side virtualization features for guests
  net/mlx4_en: Set csum level for encapsulated packets
  be2net: Export tunnel offloads only when a VxLAN tunnel is created
  gianfar: Fix dma check map error when DMA_API_DEBUG is enabled
  cxgb4/csiostor: Don't use MASTER_MUST for fw_hello call
  net: fec: only enable mdio interrupt before phy device link up
  net: fec: clear all interrupt events to support i.MX6SX
  net: fec: reset fep link status in suspend function
  net: sock: fix access via invalid file descriptor
  net: introduce helper macro for_each_cmsghdr
  ...

1 files changed, 437 insertions, 98 deletions

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 2349ea970255..7bb5f07dbeef 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -114,6 +114,7 @@
  *     THE POSSIBILITY OF SUCH DAMAGE.
  */
 
+#include <linux/platform_device.h>
 #include <linux/spinlock.h>
 #include <linux/tcp.h>
 #include <linux/if_vlan.h>
@@ -126,14 +127,126 @@
 #include "xgbe.h"
 #include "xgbe-common.h"
 
-static int xgbe_poll(struct napi_struct *, int);
+static int xgbe_one_poll(struct napi_struct *, int);
+static int xgbe_all_poll(struct napi_struct *, int);
 static void xgbe_set_rx_mode(struct net_device *);
 
+static int xgbe_alloc_channels(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel_mem, *channel;
+	struct xgbe_ring *tx_ring, *rx_ring;
+	unsigned int count, i;
+	int ret = -ENOMEM;
+
+	count = max_t(unsigned int, pdata->tx_ring_count, pdata->rx_ring_count);
+
+	channel_mem = kcalloc(count, sizeof(struct xgbe_channel), GFP_KERNEL);
+	if (!channel_mem)
+		goto err_channel;
+
+	tx_ring = kcalloc(pdata->tx_ring_count, sizeof(struct xgbe_ring),
+			  GFP_KERNEL);
+	if (!tx_ring)
+		goto err_tx_ring;
+
+	rx_ring = kcalloc(pdata->rx_ring_count, sizeof(struct xgbe_ring),
+			  GFP_KERNEL);
+	if (!rx_ring)
+		goto err_rx_ring;
+
+	for (i = 0, channel = channel_mem; i < count; i++, channel++) {
+		snprintf(channel->name, sizeof(channel->name), "channel-%d", i);
+		channel->pdata = pdata;
+		channel->queue_index = i;
+		channel->dma_regs = pdata->xgmac_regs + DMA_CH_BASE +
+				    (DMA_CH_INC * i);
+
+		if (pdata->per_channel_irq) {
+			/* Get the DMA interrupt (offset 1) */
+			ret = platform_get_irq(pdata->pdev, i + 1);
+			if (ret < 0) {
+				netdev_err(pdata->netdev,
+					   "platform_get_irq %u failed\n",
+					   i + 1);
+				goto err_irq;
+			}
+
+			channel->dma_irq = ret;
+		}
+
+		if (i < pdata->tx_ring_count) {
+			spin_lock_init(&tx_ring->lock);
+			channel->tx_ring = tx_ring++;
+		}
+
+		if (i < pdata->rx_ring_count) {
+			spin_lock_init(&rx_ring->lock);
+			channel->rx_ring = rx_ring++;
+		}
+
+		DBGPR("  %s: queue=%u, dma_regs=%p, dma_irq=%d, tx=%p, rx=%p\n",
+		      channel->name, channel->queue_index, channel->dma_regs,
+		      channel->dma_irq, channel->tx_ring, channel->rx_ring);
+	}
+
+	pdata->channel = channel_mem;
+	pdata->channel_count = count;
+
+	return 0;
+
+err_irq:
+	kfree(rx_ring);
+
+err_rx_ring:
+	kfree(tx_ring);
+
+err_tx_ring:
+	kfree(channel_mem);
+
+err_channel:
+	return ret;
+}
+
+static void xgbe_free_channels(struct xgbe_prv_data *pdata)
+{
+	if (!pdata->channel)
+		return;
+
+	kfree(pdata->channel->rx_ring);
+	kfree(pdata->channel->tx_ring);
+	kfree(pdata->channel);
+
+	pdata->channel = NULL;
+	pdata->channel_count = 0;
+}
+
 static inline unsigned int xgbe_tx_avail_desc(struct xgbe_ring *ring)
 {
 	return (ring->rdesc_count - (ring->cur - ring->dirty));
 }
 
+static int xgbe_maybe_stop_tx_queue(struct xgbe_channel *channel,
+				    struct xgbe_ring *ring, unsigned int count)
+{
+	struct xgbe_prv_data *pdata = channel->pdata;
+
+	if (count > xgbe_tx_avail_desc(ring)) {
+		DBGPR("  Tx queue stopped, not enough descriptors available\n");
+		netif_stop_subqueue(pdata->netdev, channel->queue_index);
+		ring->tx.queue_stopped = 1;
+
+		/* If we haven't notified the hardware because of xmit_more
+		 * support, tell it now
+		 */
+		if (ring->tx.xmit_more)
+			pdata->hw_if.tx_start_xmit(channel, ring);
+
+		return NETDEV_TX_BUSY;
+	}
+
+	return 0;
+}
+
 static int xgbe_calc_rx_buf_size(struct net_device *netdev, unsigned int mtu)
 {
 	unsigned int rx_buf_size;
@@ -144,8 +257,8 @@ static int xgbe_calc_rx_buf_size(struct net_device *netdev, unsigned int mtu)
 	}
 
 	rx_buf_size = mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN;
-	if (rx_buf_size < XGBE_RX_MIN_BUF_SIZE)
-		rx_buf_size = XGBE_RX_MIN_BUF_SIZE;
+	rx_buf_size = clamp_val(rx_buf_size, XGBE_RX_MIN_BUF_SIZE, PAGE_SIZE);
+
 	rx_buf_size = (rx_buf_size + XGBE_RX_BUF_ALIGN - 1) &
 		      ~(XGBE_RX_BUF_ALIGN - 1);
 
@@ -213,11 +326,7 @@ static irqreturn_t xgbe_isr(int irq, void *data)
 	if (!dma_isr)
 		goto isr_done;
 
-	DBGPR("-->xgbe_isr\n");
-
 	DBGPR("  DMA_ISR = %08x\n", dma_isr);
-	DBGPR("  DMA_DS0 = %08x\n", XGMAC_IOREAD(pdata, DMA_DSR0));
-	DBGPR("  DMA_DS1 = %08x\n", XGMAC_IOREAD(pdata, DMA_DSR1));
 
 	for (i = 0; i < pdata->channel_count; i++) {
 		if (!(dma_isr & (1 << i)))
@@ -228,6 +337,10 @@ static irqreturn_t xgbe_isr(int irq, void *data)
 		dma_ch_isr = XGMAC_DMA_IOREAD(channel, DMA_CH_SR);
 		DBGPR("  DMA_CH%u_ISR = %08x\n", i, dma_ch_isr);
 
+		/* If we get a TI or RI interrupt that means per channel DMA
+		 * interrupts are not enabled, so we use the private data napi
+		 * structure, not the per channel napi structure
+		 */
 		if (XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, TI) ||
 		    XGMAC_GET_BITS(dma_ch_isr, DMA_CH_SR, RI)) {
 			if (napi_schedule_prep(&pdata->napi)) {
@@ -270,12 +383,28 @@ static irqreturn_t xgbe_isr(int irq, void *data)
 
 	DBGPR("  DMA_ISR = %08x\n", XGMAC_IOREAD(pdata, DMA_ISR));
 
-	DBGPR("<--xgbe_isr\n");
-
 isr_done:
 	return IRQ_HANDLED;
 }
 
+static irqreturn_t xgbe_dma_isr(int irq, void *data)
+{
+	struct xgbe_channel *channel = data;
+
+	/* Per channel DMA interrupts are enabled, so we use the per
+	 * channel napi structure and not the private data napi structure
+	 */
+	if (napi_schedule_prep(&channel->napi)) {
+		/* Disable Tx and Rx interrupts */
+		disable_irq_nosync(channel->dma_irq);
+
+		/* Turn on polling */
+		__napi_schedule(&channel->napi);
+	}
+
+	return IRQ_HANDLED;
+}
+
 static enum hrtimer_restart xgbe_tx_timer(struct hrtimer *timer)
 {
 	struct xgbe_channel *channel = container_of(timer,
@@ -283,18 +412,24 @@ static enum hrtimer_restart xgbe_tx_timer(struct hrtimer *timer)
 						    tx_timer);
 	struct xgbe_ring *ring = channel->tx_ring;
 	struct xgbe_prv_data *pdata = channel->pdata;
+	struct napi_struct *napi;
 	unsigned long flags;
 
 	DBGPR("-->xgbe_tx_timer\n");
 
+	napi = (pdata->per_channel_irq) ? &channel->napi : &pdata->napi;
+
 	spin_lock_irqsave(&ring->lock, flags);
 
-	if (napi_schedule_prep(&pdata->napi)) {
+	if (napi_schedule_prep(napi)) {
 		/* Disable Tx and Rx interrupts */
-		xgbe_disable_rx_tx_ints(pdata);
+		if (pdata->per_channel_irq)
+			disable_irq(channel->dma_irq);
+		else
+			xgbe_disable_rx_tx_ints(pdata);
 
 		/* Turn on polling */
-		__napi_schedule(&pdata->napi);
+		__napi_schedule(napi);
 	}
 
 	channel->tx_timer_active = 0;
@@ -430,18 +565,46 @@ void xgbe_get_all_hw_features(struct xgbe_prv_data *pdata)
 
 static void xgbe_napi_enable(struct xgbe_prv_data *pdata, unsigned int add)
 {
-	if (add)
-		netif_napi_add(pdata->netdev, &pdata->napi, xgbe_poll,
-			       NAPI_POLL_WEIGHT);
-	napi_enable(&pdata->napi);
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	if (pdata->per_channel_irq) {
+		channel = pdata->channel;
+		for (i = 0; i < pdata->channel_count; i++, channel++) {
+			if (add)
+				netif_napi_add(pdata->netdev, &channel->napi,
+					       xgbe_one_poll, NAPI_POLL_WEIGHT);
+
+			napi_enable(&channel->napi);
+		}
+	} else {
+		if (add)
+			netif_napi_add(pdata->netdev, &pdata->napi,
+				       xgbe_all_poll, NAPI_POLL_WEIGHT);
+
+		napi_enable(&pdata->napi);
+	}
 }
 
 static void xgbe_napi_disable(struct xgbe_prv_data *pdata, unsigned int del)
 {
-	napi_disable(&pdata->napi);
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	if (pdata->per_channel_irq) {
+		channel = pdata->channel;
+		for (i = 0; i < pdata->channel_count; i++, channel++) {
+			napi_disable(&channel->napi);
 
-	if (del)
-		netif_napi_del(&pdata->napi);
+			if (del)
+				netif_napi_del(&channel->napi);
+		}
+	} else {
+		napi_disable(&pdata->napi);
+
+		if (del)
+			netif_napi_del(&pdata->napi);
+	}
 }
 
 void xgbe_init_tx_coalesce(struct xgbe_prv_data *pdata)
@@ -472,7 +635,7 @@ void xgbe_init_rx_coalesce(struct xgbe_prv_data *pdata)
 	DBGPR("<--xgbe_init_rx_coalesce\n");
 }
 
-static void xgbe_free_tx_skbuff(struct xgbe_prv_data *pdata)
+static void xgbe_free_tx_data(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_desc_if *desc_if = &pdata->desc_if;
 	struct xgbe_channel *channel;
@@ -480,7 +643,7 @@ static void xgbe_free_tx_skbuff(struct xgbe_prv_data *pdata)
 	struct xgbe_ring_data *rdata;
 	unsigned int i, j;
 
-	DBGPR("-->xgbe_free_tx_skbuff\n");
+	DBGPR("-->xgbe_free_tx_data\n");
 
 	channel = pdata->channel;
 	for (i = 0; i < pdata->channel_count; i++, channel++) {
@@ -490,14 +653,14 @@ static void xgbe_free_tx_skbuff(struct xgbe_prv_data *pdata)
 
 		for (j = 0; j < ring->rdesc_count; j++) {
 			rdata = XGBE_GET_DESC_DATA(ring, j);
-			desc_if->unmap_skb(pdata, rdata);
+			desc_if->unmap_rdata(pdata, rdata);
 		}
 	}
 
-	DBGPR("<--xgbe_free_tx_skbuff\n");
+	DBGPR("<--xgbe_free_tx_data\n");
 }
 
-static void xgbe_free_rx_skbuff(struct xgbe_prv_data *pdata)
+static void xgbe_free_rx_data(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_desc_if *desc_if = &pdata->desc_if;
 	struct xgbe_channel *channel;
@@ -505,7 +668,7 @@ static void xgbe_free_rx_skbuff(struct xgbe_prv_data *pdata)
 	struct xgbe_ring_data *rdata;
 	unsigned int i, j;
 
-	DBGPR("-->xgbe_free_rx_skbuff\n");
+	DBGPR("-->xgbe_free_rx_data\n");
 
 	channel = pdata->channel;
 	for (i = 0; i < pdata->channel_count; i++, channel++) {
@@ -515,11 +678,11 @@ static void xgbe_free_rx_skbuff(struct xgbe_prv_data *pdata)
 
 		for (j = 0; j < ring->rdesc_count; j++) {
 			rdata = XGBE_GET_DESC_DATA(ring, j);
-			desc_if->unmap_skb(pdata, rdata);
+			desc_if->unmap_rdata(pdata, rdata);
 		}
 	}
 
-	DBGPR("<--xgbe_free_rx_skbuff\n");
+	DBGPR("<--xgbe_free_rx_data\n");
 }
 
 static void xgbe_adjust_link(struct net_device *netdev)
@@ -735,7 +898,10 @@ static int xgbe_start(struct xgbe_prv_data *pdata)
 static void xgbe_stop(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	struct xgbe_channel *channel;
 	struct net_device *netdev = pdata->netdev;
+	struct netdev_queue *txq;
+	unsigned int i;
 
 	DBGPR("-->xgbe_stop\n");
 
@@ -749,12 +915,23 @@ static void xgbe_stop(struct xgbe_prv_data *pdata)
 	hw_if->disable_tx(pdata);
 	hw_if->disable_rx(pdata);
 
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			continue;
+
+		txq = netdev_get_tx_queue(netdev, channel->queue_index);
+		netdev_tx_reset_queue(txq);
+	}
+
 	DBGPR("<--xgbe_stop\n");
 }
 
 static void xgbe_restart_dev(struct xgbe_prv_data *pdata, unsigned int reset)
 {
+	struct xgbe_channel *channel;
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	unsigned int i;
 
 	DBGPR("-->xgbe_restart_dev\n");
 
@@ -763,10 +940,15 @@ static void xgbe_restart_dev(struct xgbe_prv_data *pdata, unsigned int reset)
 		return;
 
 	xgbe_stop(pdata);
-	synchronize_irq(pdata->irq_number);
+	synchronize_irq(pdata->dev_irq);
+	if (pdata->per_channel_irq) {
+		channel = pdata->channel;
+		for (i = 0; i < pdata->channel_count; i++, channel++)
+			synchronize_irq(channel->dma_irq);
+	}
 
-	xgbe_free_tx_skbuff(pdata);
-	xgbe_free_rx_skbuff(pdata);
+	xgbe_free_tx_data(pdata);
+	xgbe_free_rx_data(pdata);
 
 	/* Issue software reset to device if requested */
 	if (reset)
@@ -1008,6 +1190,12 @@ static int xgbe_prep_tso(struct sk_buff *skb, struct xgbe_packet_data *packet)
 	      packet->tcp_header_len, packet->tcp_payload_len);
 	DBGPR("  packet->mss=%u\n", packet->mss);
 
+	/* Update the number of packets that will ultimately be transmitted
+	 * along with the extra bytes for each extra packet
+	 */
+	packet->tx_packets = skb_shinfo(skb)->gso_segs;
+	packet->tx_bytes += (packet->tx_packets - 1) * packet->header_len;
+
 	return 0;
 }
 
@@ -1033,17 +1221,22 @@ static void xgbe_packet_info(struct xgbe_prv_data *pdata,
 	unsigned int len;
 	unsigned int i;
 
+	packet->skb = skb;
+
 	context_desc = 0;
 	packet->rdesc_count = 0;
 
+	packet->tx_packets = 1;
+	packet->tx_bytes = skb->len;
+
 	if (xgbe_is_tso(skb)) {
-		/* TSO requires an extra desriptor if mss is different */
+		/* TSO requires an extra descriptor if mss is different */
 		if (skb_shinfo(skb)->gso_size != ring->tx.cur_mss) {
 			context_desc = 1;
 			packet->rdesc_count++;
 		}
 
-		/* TSO requires an extra desriptor for TSO header */
+		/* TSO requires an extra descriptor for TSO header */
 		packet->rdesc_count++;
 
 		XGMAC_SET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
@@ -1091,6 +1284,8 @@ static int xgbe_open(struct net_device *netdev)
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
 	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	struct xgbe_channel *channel = NULL;
+	unsigned int i = 0;
 	int ret;
 
 	DBGPR("-->xgbe_open\n");
@@ -1119,24 +1314,48 @@ static int xgbe_open(struct net_device *netdev)
 		goto err_ptpclk;
 	pdata->rx_buf_size = ret;
 
+	/* Allocate the channel and ring structures */
+	ret = xgbe_alloc_channels(pdata);
+	if (ret)
+		goto err_ptpclk;
+
 	/* Allocate the ring descriptors and buffers */
 	ret = desc_if->alloc_ring_resources(pdata);
 	if (ret)
-		goto err_ptpclk;
+		goto err_channels;
 
 	/* Initialize the device restart and Tx timestamp work struct */
 	INIT_WORK(&pdata->restart_work, xgbe_restart);
 	INIT_WORK(&pdata->tx_tstamp_work, xgbe_tx_tstamp);
 
 	/* Request interrupts */
-	ret = devm_request_irq(pdata->dev, netdev->irq, xgbe_isr, 0,
+	ret = devm_request_irq(pdata->dev, pdata->dev_irq, xgbe_isr, 0,
 			       netdev->name, pdata);
 	if (ret) {
 		netdev_alert(netdev, "error requesting irq %d\n",
-			     pdata->irq_number);
-		goto err_irq;
+			     pdata->dev_irq);
+		goto err_rings;
+	}
+
+	if (pdata->per_channel_irq) {
+		channel = pdata->channel;
+		for (i = 0; i < pdata->channel_count; i++, channel++) {
+			snprintf(channel->dma_irq_name,
+				 sizeof(channel->dma_irq_name) - 1,
+				 "%s-TxRx-%u", netdev_name(netdev),
+				 channel->queue_index);
+
+			ret = devm_request_irq(pdata->dev, channel->dma_irq,
+					       xgbe_dma_isr, 0,
+					       channel->dma_irq_name, channel);
+			if (ret) {
+				netdev_alert(netdev,
+					     "error requesting irq %d\n",
+					     channel->dma_irq);
+				goto err_irq;
+			}
+		}
 	}
-	pdata->irq_number = netdev->irq;
 
 	ret = xgbe_start(pdata);
 	if (ret)
@@ -1149,12 +1368,21 @@ static int xgbe_open(struct net_device *netdev)
 err_start:
 	hw_if->exit(pdata);
 
-	devm_free_irq(pdata->dev, pdata->irq_number, pdata);
-	pdata->irq_number = 0;
-
 err_irq:
+	if (pdata->per_channel_irq) {
+		/* Using an unsigned int, 'i' will go to UINT_MAX and exit */
+		for (i--, channel--; i < pdata->channel_count; i--, channel--)
+			devm_free_irq(pdata->dev, channel->dma_irq, channel);
+	}
+
+	devm_free_irq(pdata->dev, pdata->dev_irq, pdata);
+
+err_rings:
 	desc_if->free_ring_resources(pdata);
 
+err_channels:
+	xgbe_free_channels(pdata);
+
 err_ptpclk:
 	clk_disable_unprepare(pdata->ptpclk);
 
@@ -1172,6 +1400,8 @@ static int xgbe_close(struct net_device *netdev)
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
 	struct xgbe_desc_if *desc_if = &pdata->desc_if;
+	struct xgbe_channel *channel;
+	unsigned int i;
 
 	DBGPR("-->xgbe_close\n");
 
@@ -1181,15 +1411,20 @@ static int xgbe_close(struct net_device *netdev)
 	/* Issue software reset to device */
 	hw_if->exit(pdata);
 
-	/* Free all the ring data */
+	/* Free the ring descriptors and buffers */
 	desc_if->free_ring_resources(pdata);
 
-	/* Release the interrupt */
-	if (pdata->irq_number != 0) {
-		devm_free_irq(pdata->dev, pdata->irq_number, pdata);
-		pdata->irq_number = 0;
+	/* Release the interrupts */
+	devm_free_irq(pdata->dev, pdata->dev_irq, pdata);
+	if (pdata->per_channel_irq) {
+		channel = pdata->channel;
+		for (i = 0; i < pdata->channel_count; i++, channel++)
+			devm_free_irq(pdata->dev, channel->dma_irq, channel);
 	}
 
+	/* Free the channel and ring structures */
+	xgbe_free_channels(pdata);
+
 	/* Disable the clocks */
 	clk_disable_unprepare(pdata->ptpclk);
 	clk_disable_unprepare(pdata->sysclk);
@@ -1210,12 +1445,14 @@ static int xgbe_xmit(struct sk_buff *skb, struct net_device *netdev)
 	struct xgbe_channel *channel;
 	struct xgbe_ring *ring;
 	struct xgbe_packet_data *packet;
+	struct netdev_queue *txq;
 	unsigned long flags;
 	int ret;
 
 	DBGPR("-->xgbe_xmit: skb->len = %d\n", skb->len);
 
 	channel = pdata->channel + skb->queue_mapping;
+	txq = netdev_get_tx_queue(netdev, channel->queue_index);
 	ring = channel->tx_ring;
 	packet = &ring->packet_data;
 
@@ -1234,13 +1471,9 @@ static int xgbe_xmit(struct sk_buff *skb, struct net_device *netdev)
 	xgbe_packet_info(pdata, ring, skb, packet);
 
 	/* Check that there are enough descriptors available */
-	if (packet->rdesc_count > xgbe_tx_avail_desc(ring)) {
-		DBGPR("  Tx queue stopped, not enough descriptors available\n");
-		netif_stop_subqueue(netdev, channel->queue_index);
-		ring->tx.queue_stopped = 1;
-		ret = NETDEV_TX_BUSY;
+	ret = xgbe_maybe_stop_tx_queue(channel, ring, packet->rdesc_count);
+	if (ret)
 		goto tx_netdev_return;
-	}
 
 	ret = xgbe_prep_tso(skb, packet);
 	if (ret) {
@@ -1257,13 +1490,21 @@ static int xgbe_xmit(struct sk_buff *skb, struct net_device *netdev)
 
 	xgbe_prep_tx_tstamp(pdata, skb, packet);
 
+	/* Report on the actual number of bytes (to be) sent */
+	netdev_tx_sent_queue(txq, packet->tx_bytes);
+
 	/* Configure required descriptor fields for transmission */
-	hw_if->pre_xmit(channel);
+	hw_if->dev_xmit(channel);
 
 #ifdef XGMAC_ENABLE_TX_PKT_DUMP
 	xgbe_print_pkt(netdev, skb, true);
 #endif
 
+	/* Stop the queue in advance if there may not be enough descriptors */
+	xgbe_maybe_stop_tx_queue(channel, ring, XGBE_TX_MAX_DESCS);
+
+	ret = NETDEV_TX_OK;
+
 tx_netdev_return:
 	spin_unlock_irqrestore(&ring->lock, flags);
 
@@ -1420,14 +1661,20 @@ static int xgbe_vlan_rx_kill_vid(struct net_device *netdev, __be16 proto,
 static void xgbe_poll_controller(struct net_device *netdev)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_channel *channel;
+	unsigned int i;
 
 	DBGPR("-->xgbe_poll_controller\n");
 
-	disable_irq(pdata->irq_number);
-
-	xgbe_isr(pdata->irq_number, pdata);
-
-	enable_irq(pdata->irq_number);
+	if (pdata->per_channel_irq) {
+		channel = pdata->channel;
+		for (i = 0; i < pdata->channel_count; i++, channel++)
+			xgbe_dma_isr(channel->dma_irq, channel);
+	} else {
+		disable_irq(pdata->dev_irq);
+		xgbe_isr(pdata->dev_irq, pdata);
+		enable_irq(pdata->dev_irq);
+	}
 
 	DBGPR("<--xgbe_poll_controller\n");
 }
@@ -1465,12 +1712,21 @@ static int xgbe_set_features(struct net_device *netdev,
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
-	netdev_features_t rxcsum, rxvlan, rxvlan_filter;
+	netdev_features_t rxhash, rxcsum, rxvlan, rxvlan_filter;
+	int ret = 0;
 
+	rxhash = pdata->netdev_features & NETIF_F_RXHASH;
 	rxcsum = pdata->netdev_features & NETIF_F_RXCSUM;
 	rxvlan = pdata->netdev_features & NETIF_F_HW_VLAN_CTAG_RX;
 	rxvlan_filter = pdata->netdev_features & NETIF_F_HW_VLAN_CTAG_FILTER;
 
+	if ((features & NETIF_F_RXHASH) && !rxhash)
+		ret = hw_if->enable_rss(pdata);
+	else if (!(features & NETIF_F_RXHASH) && rxhash)
+		ret = hw_if->disable_rss(pdata);
+	if (ret)
+		return ret;
+
 	if ((features & NETIF_F_RXCSUM) && !rxcsum)
 		hw_if->enable_rx_csum(pdata);
 	else if (!(features & NETIF_F_RXCSUM) && rxcsum)
@@ -1524,7 +1780,7 @@ static void xgbe_rx_refresh(struct xgbe_channel *channel)
 	struct xgbe_ring *ring = channel->rx_ring;
 	struct xgbe_ring_data *rdata;
 
-	desc_if->realloc_skb(channel);
+	desc_if->realloc_rx_buffer(channel);
 
 	/* Update the Rx Tail Pointer Register with address of
 	 * the last cleaned entry */
@@ -1533,6 +1789,31 @@ static void xgbe_rx_refresh(struct xgbe_channel *channel)
 			  lower_32_bits(rdata->rdesc_dma));
 }
 
+static struct sk_buff *xgbe_create_skb(struct xgbe_prv_data *pdata,
+				       struct xgbe_ring_data *rdata,
+				       unsigned int *len)
+{
+	struct net_device *netdev = pdata->netdev;
+	struct sk_buff *skb;
+	u8 *packet;
+	unsigned int copy_len;
+
+	skb = netdev_alloc_skb_ip_align(netdev, rdata->rx.hdr.dma_len);
+	if (!skb)
+		return NULL;
+
+	packet = page_address(rdata->rx.hdr.pa.pages) +
+		 rdata->rx.hdr.pa.pages_offset;
+	copy_len = (rdata->rx.hdr_len) ? rdata->rx.hdr_len : *len;
+	copy_len = min(rdata->rx.hdr.dma_len, copy_len);
+	skb_copy_to_linear_data(skb, packet, copy_len);
+	skb_put(skb, copy_len);
+
+	*len -= copy_len;
+
+	return skb;
+}
+
 static int xgbe_tx_poll(struct xgbe_channel *channel)
 {
 	struct xgbe_prv_data *pdata = channel->pdata;
@@ -1542,8 +1823,10 @@ static int xgbe_tx_poll(struct xgbe_channel *channel)
 	struct xgbe_ring_data *rdata;
 	struct xgbe_ring_desc *rdesc;
 	struct net_device *netdev = pdata->netdev;
+	struct netdev_queue *txq;
 	unsigned long flags;
 	int processed = 0;
+	unsigned int tx_packets = 0, tx_bytes = 0;
 
 	DBGPR("-->xgbe_tx_poll\n");
 
@@ -1551,36 +1834,53 @@ static int xgbe_tx_poll(struct xgbe_channel *channel)
 	if (!ring)
 		return 0;
 
+	txq = netdev_get_tx_queue(netdev, channel->queue_index);
+
 	spin_lock_irqsave(&ring->lock, flags);
 
 	while ((processed < XGBE_TX_DESC_MAX_PROC) &&
-	       (ring->dirty < ring->cur)) {
+	       (ring->dirty != ring->cur)) {
 		rdata = XGBE_GET_DESC_DATA(ring, ring->dirty);
 		rdesc = rdata->rdesc;
 
 		if (!hw_if->tx_complete(rdesc))
 			break;
 
+		/* Make sure descriptor fields are read after reading the OWN
+		 * bit */
+		rmb();
+
 #ifdef XGMAC_ENABLE_TX_DESC_DUMP
 		xgbe_dump_tx_desc(ring, ring->dirty, 1, 0);
 #endif
 
+		if (hw_if->is_last_desc(rdesc)) {
+			tx_packets += rdata->tx.packets;
+			tx_bytes += rdata->tx.bytes;
+		}
+
 		/* Free the SKB and reset the descriptor for re-use */
-		desc_if->unmap_skb(pdata, rdata);
+		desc_if->unmap_rdata(pdata, rdata);
 		hw_if->tx_desc_reset(rdata);
 
 		processed++;
 		ring->dirty++;
 	}
 
+	if (!processed)
+		goto unlock;
+
+	netdev_tx_completed_queue(txq, tx_packets, tx_bytes);
+
 	if ((ring->tx.queue_stopped == 1) &&
 	    (xgbe_tx_avail_desc(ring) > XGBE_TX_DESC_MIN_FREE)) {
 		ring->tx.queue_stopped = 0;
-		netif_wake_subqueue(netdev, channel->queue_index);
+		netif_tx_wake_queue(txq);
 	}
 
 	DBGPR("<--xgbe_tx_poll: processed=%d\n", processed);
 
+unlock:
 	spin_unlock_irqrestore(&ring->lock, flags);
 
 	return processed;
@@ -1594,6 +1894,7 @@ static int xgbe_rx_poll(struct xgbe_channel *channel, int budget)
 	struct xgbe_ring_data *rdata;
 	struct xgbe_packet_data *packet;
 	struct net_device *netdev = pdata->netdev;
+	struct napi_struct *napi;
 	struct sk_buff *skb;
 	struct skb_shared_hwtstamps *hwtstamps;
 	unsigned int incomplete, error, context_next, context;
@@ -1607,6 +1908,8 @@ static int xgbe_rx_poll(struct xgbe_channel *channel, int budget)
 	if (!ring)
 		return 0;
 
+	napi = (pdata->per_channel_irq) ? &channel->napi : &pdata->napi;
+
 	rdata = XGBE_GET_DESC_DATA(ring, ring->cur);
 	packet = &ring->packet_data;
 	while (packet_count < budget) {
@@ -1641,10 +1944,6 @@ read_again:
 		ring->cur++;
 		ring->dirty++;
 
-		dma_unmap_single(pdata->dev, rdata->skb_dma,
-				 rdata->skb_dma_len, DMA_FROM_DEVICE);
-		rdata->skb_dma = 0;
-
 		incomplete = XGMAC_GET_BITS(packet->attributes,
 					    RX_PACKET_ATTRIBUTES,
 					    INCOMPLETE);
@@ -1667,33 +1966,40 @@ read_again:
 		}
 
 		if (!context) {
-			put_len = rdata->len - len;
-			if (skb) {
-				if (pskb_expand_head(skb, 0, put_len,
-						     GFP_ATOMIC)) {
-					DBGPR("pskb_expand_head error\n");
-					if (incomplete) {
-						error = 1;
-						goto read_again;
-					}
-
-					dev_kfree_skb(skb);
-					goto next_packet;
+			put_len = rdata->rx.len - len;
+			len += put_len;
+
+			if (!skb) {
+				dma_sync_single_for_cpu(pdata->dev,
+							rdata->rx.hdr.dma,
+							rdata->rx.hdr.dma_len,
+							DMA_FROM_DEVICE);
+
+				skb = xgbe_create_skb(pdata, rdata, &put_len);
+				if (!skb) {
+					error = 1;
+					goto skip_data;
 				}
-				memcpy(skb_tail_pointer(skb), rdata->skb->data,
-				       put_len);
-			} else {
-				skb = rdata->skb;
-				rdata->skb = NULL;
 			}
-			skb_put(skb, put_len);
-			len += put_len;
+
+			if (put_len) {
+				dma_sync_single_for_cpu(pdata->dev,
+							rdata->rx.buf.dma,
+							rdata->rx.buf.dma_len,
+							DMA_FROM_DEVICE);
+
+				skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
+						rdata->rx.buf.pa.pages,
+						rdata->rx.buf.pa.pages_offset,
+						put_len, rdata->rx.buf.dma_len);
+				rdata->rx.buf.pa.pages = NULL;
+			}
 		}
 
+skip_data:
 		if (incomplete || context_next)
 			goto read_again;
 
-		/* Stray Context Descriptor? */
 		if (!skb)
 			goto next_packet;
 
@@ -1733,13 +2039,18 @@ read_again:
 			hwtstamps->hwtstamp = ns_to_ktime(nsec);
 		}
 
+		if (XGMAC_GET_BITS(packet->attributes,
+				   RX_PACKET_ATTRIBUTES, RSS_HASH))
+			skb_set_hash(skb, packet->rss_hash,
+				     packet->rss_hash_type);
+
 		skb->dev = netdev;
 		skb->protocol = eth_type_trans(skb, netdev);
 		skb_record_rx_queue(skb, channel->queue_index);
-		skb_mark_napi_id(skb, &pdata->napi);
+		skb_mark_napi_id(skb, napi);
 
 		netdev->last_rx = jiffies;
-		napi_gro_receive(&pdata->napi, skb);
+		napi_gro_receive(napi, skb);
 
 next_packet:
 		packet_count++;
@@ -1761,7 +2072,35 @@ next_packet:
 	return packet_count;
 }
 
-static int xgbe_poll(struct napi_struct *napi, int budget)
+static int xgbe_one_poll(struct napi_struct *napi, int budget)
+{
+	struct xgbe_channel *channel = container_of(napi, struct xgbe_channel,
+						    napi);
+	int processed = 0;
+
+	DBGPR("-->xgbe_one_poll: budget=%d\n", budget);
+
+	/* Cleanup Tx ring first */
+	xgbe_tx_poll(channel);
+
+	/* Process Rx ring next */
+	processed = xgbe_rx_poll(channel, budget);
+
+	/* If we processed everything, we are done */
+	if (processed < budget) {
+		/* Turn off polling */
+		napi_complete(napi);
+
+		/* Enable Tx and Rx interrupts */
+		enable_irq(channel->dma_irq);
+	}
+
+	DBGPR("<--xgbe_one_poll: received = %d\n", processed);
+
+	return processed;
+}
+
+static int xgbe_all_poll(struct napi_struct *napi, int budget)
 {
 	struct xgbe_prv_data *pdata = container_of(napi, struct xgbe_prv_data,
 						   napi);
@@ -1770,7 +2109,7 @@ static int xgbe_poll(struct napi_struct *napi, int budget)
 	int processed, last_processed;
 	unsigned int i;
 
-	DBGPR("-->xgbe_poll: budget=%d\n", budget);
+	DBGPR("-->xgbe_all_poll: budget=%d\n", budget);
 
 	processed = 0;
 	ring_budget = budget / pdata->rx_ring_count;
@@ -1798,7 +2137,7 @@ static int xgbe_poll(struct napi_struct *napi, int budget)
 		xgbe_enable_rx_tx_ints(pdata);
 	}
 
-	DBGPR("<--xgbe_poll: received = %d\n", processed);
+	DBGPR("<--xgbe_all_poll: received = %d\n", processed);
 
 	return processed;
 }
@@ -1812,10 +2151,10 @@ void xgbe_dump_tx_desc(struct xgbe_ring *ring, unsigned int idx,
 	while (count--) {
 		rdata = XGBE_GET_DESC_DATA(ring, idx);
 		rdesc = rdata->rdesc;
-		DBGPR("TX_NORMAL_DESC[%d %s] = %08x:%08x:%08x:%08x\n", idx,
-		      (flag == 1) ? "QUEUED FOR TX" : "TX BY DEVICE",
-		      le32_to_cpu(rdesc->desc0), le32_to_cpu(rdesc->desc1),
-		      le32_to_cpu(rdesc->desc2), le32_to_cpu(rdesc->desc3));
+		pr_alert("TX_NORMAL_DESC[%d %s] = %08x:%08x:%08x:%08x\n", idx,
+			 (flag == 1) ? "QUEUED FOR TX" : "TX BY DEVICE",
+			 le32_to_cpu(rdesc->desc0), le32_to_cpu(rdesc->desc1),
+			 le32_to_cpu(rdesc->desc2), le32_to_cpu(rdesc->desc3));
 		idx++;
 	}
 }
@@ -1823,9 +2162,9 @@ void xgbe_dump_tx_desc(struct xgbe_ring *ring, unsigned int idx,
 void xgbe_dump_rx_desc(struct xgbe_ring *ring, struct xgbe_ring_desc *desc,
 		       unsigned int idx)
 {
-	DBGPR("RX_NORMAL_DESC[%d RX BY DEVICE] = %08x:%08x:%08x:%08x\n", idx,
-	      le32_to_cpu(desc->desc0), le32_to_cpu(desc->desc1),
-	      le32_to_cpu(desc->desc2), le32_to_cpu(desc->desc3));
+	pr_alert("RX_NORMAL_DESC[%d RX BY DEVICE] = %08x:%08x:%08x:%08x\n", idx,
+		 le32_to_cpu(desc->desc0), le32_to_cpu(desc->desc1),
+		 le32_to_cpu(desc->desc2), le32_to_cpu(desc->desc3));
 }
 
 void xgbe_print_pkt(struct net_device *netdev, struct sk_buff *skb, bool tx_rx)