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
  ...

8 files changed, 1207 insertions, 334 deletions

diff --git a/drivers/net/ethernet/amd/Kconfig b/drivers/net/ethernet/amd/Kconfig
index 8319c99..7a5e4aa 100644
--- a/drivers/net/ethernet/amd/Kconfig
+++ b/drivers/net/ethernet/amd/Kconfig
@@ -179,7 +179,7 @@ config SUNLANCE
 
 config AMD_XGBE
 	tristate "AMD 10GbE Ethernet driver"
-	depends on OF_NET
+	depends on OF_NET && HAS_IOMEM
 	select PHYLIB
 	select AMD_XGBE_PHY
 	select BITREVERSE
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-common.h b/drivers/net/ethernet/amd/xgbe/xgbe-common.h
index caade30..75b08c6 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-common.h
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-common.h
@@ -125,9 +125,6 @@
 #define DMA_AXIAWCR			0x3018
 #define DMA_DSR0			0x3020
 #define DMA_DSR1			0x3024
-#define DMA_DSR2			0x3028
-#define DMA_DSR3			0x302c
-#define DMA_DSR4			0x3030
 
 /* DMA register entry bit positions and sizes */
 #define DMA_AXIARCR_DRC_INDEX		0
@@ -158,10 +155,6 @@
 #define DMA_AXIAWCR_TDC_WIDTH		4
 #define DMA_AXIAWCR_TDD_INDEX		28
 #define DMA_AXIAWCR_TDD_WIDTH		2
-#define DMA_DSR0_RPS_INDEX		8
-#define DMA_DSR0_RPS_WIDTH		4
-#define DMA_DSR0_TPS_INDEX		12
-#define DMA_DSR0_TPS_WIDTH		4
 #define DMA_ISR_MACIS_INDEX		17
 #define DMA_ISR_MACIS_WIDTH		1
 #define DMA_ISR_MTLIS_INDEX		16
@@ -175,6 +168,20 @@
 #define DMA_SBMR_UNDEF_INDEX		0
 #define DMA_SBMR_UNDEF_WIDTH		1
 
+/* DMA register values */
+#define DMA_DSR_RPS_WIDTH		4
+#define DMA_DSR_TPS_WIDTH		4
+#define DMA_DSR_Q_WIDTH			(DMA_DSR_RPS_WIDTH + DMA_DSR_TPS_WIDTH)
+#define DMA_DSR0_RPS_START		8
+#define DMA_DSR0_TPS_START		12
+#define DMA_DSRX_FIRST_QUEUE		3
+#define DMA_DSRX_INC			4
+#define DMA_DSRX_QPR			4
+#define DMA_DSRX_RPS_START		0
+#define DMA_DSRX_TPS_START		4
+#define DMA_TPS_STOPPED			0x00
+#define DMA_TPS_SUSPENDED		0x06
+
 /* DMA channel register offsets
  *   Multiple channels can be active.  The first channel has registers
  *   that begin at 0x3100.  Each subsequent channel has registers that
@@ -207,6 +214,8 @@
 /* DMA channel register entry bit positions and sizes */
 #define DMA_CH_CR_PBLX8_INDEX		16
 #define DMA_CH_CR_PBLX8_WIDTH		1
+#define DMA_CH_CR_SPH_INDEX		24
+#define DMA_CH_CR_SPH_WIDTH		1
 #define DMA_CH_IER_AIE_INDEX		15
 #define DMA_CH_IER_AIE_WIDTH		1
 #define DMA_CH_IER_FBEE_INDEX		12
@@ -306,6 +315,9 @@
 #define MAC_MACA0LR			0x0304
 #define MAC_MACA1HR			0x0308
 #define MAC_MACA1LR			0x030c
+#define MAC_RSSCR			0x0c80
+#define MAC_RSSAR			0x0c88
+#define MAC_RSSDR			0x0c8c
 #define MAC_TSCR			0x0d00
 #define MAC_SSIR			0x0d04
 #define MAC_STSR			0x0d08
@@ -429,6 +441,8 @@
 #define MAC_RCR_CST_WIDTH		1
 #define MAC_RCR_DCRCC_INDEX		3
 #define MAC_RCR_DCRCC_WIDTH		1
+#define MAC_RCR_HDSMS_INDEX		12
+#define MAC_RCR_HDSMS_WIDTH		3
 #define MAC_RCR_IPC_INDEX		9
 #define MAC_RCR_IPC_WIDTH		1
 #define MAC_RCR_JE_INDEX		8
@@ -445,6 +459,24 @@
 #define MAC_RFCR_UP_WIDTH		1
 #define MAC_RQC0R_RXQ0EN_INDEX		0
 #define MAC_RQC0R_RXQ0EN_WIDTH		2
+#define MAC_RSSAR_ADDRT_INDEX		2
+#define MAC_RSSAR_ADDRT_WIDTH		1
+#define MAC_RSSAR_CT_INDEX		1
+#define MAC_RSSAR_CT_WIDTH		1
+#define MAC_RSSAR_OB_INDEX		0
+#define MAC_RSSAR_OB_WIDTH		1
+#define MAC_RSSAR_RSSIA_INDEX		8
+#define MAC_RSSAR_RSSIA_WIDTH		8
+#define MAC_RSSCR_IP2TE_INDEX		1
+#define MAC_RSSCR_IP2TE_WIDTH		1
+#define MAC_RSSCR_RSSE_INDEX		0
+#define MAC_RSSCR_RSSE_WIDTH		1
+#define MAC_RSSCR_TCP4TE_INDEX		2
+#define MAC_RSSCR_TCP4TE_WIDTH		1
+#define MAC_RSSCR_UDP4TE_INDEX		3
+#define MAC_RSSCR_UDP4TE_WIDTH		1
+#define MAC_RSSDR_DMCH_INDEX		0
+#define MAC_RSSDR_DMCH_WIDTH		4
 #define MAC_SSIR_SNSINC_INDEX		8
 #define MAC_SSIR_SNSINC_WIDTH		8
 #define MAC_SSIR_SSINC_INDEX		16
@@ -844,9 +876,13 @@
 #define RX_PACKET_ATTRIBUTES_CONTEXT_WIDTH	1
 #define RX_PACKET_ATTRIBUTES_RX_TSTAMP_INDEX	5
 #define RX_PACKET_ATTRIBUTES_RX_TSTAMP_WIDTH	1
+#define RX_PACKET_ATTRIBUTES_RSS_HASH_INDEX	6
+#define RX_PACKET_ATTRIBUTES_RSS_HASH_WIDTH	1
 
 #define RX_NORMAL_DESC0_OVT_INDEX		0
 #define RX_NORMAL_DESC0_OVT_WIDTH		16
+#define RX_NORMAL_DESC2_HL_INDEX		0
+#define RX_NORMAL_DESC2_HL_WIDTH		10
 #define RX_NORMAL_DESC3_CDA_INDEX		27
 #define RX_NORMAL_DESC3_CDA_WIDTH		1
 #define RX_NORMAL_DESC3_CTXT_INDEX		30
@@ -855,14 +891,27 @@
 #define RX_NORMAL_DESC3_ES_WIDTH		1
 #define RX_NORMAL_DESC3_ETLT_INDEX		16
 #define RX_NORMAL_DESC3_ETLT_WIDTH		4
+#define RX_NORMAL_DESC3_FD_INDEX		29
+#define RX_NORMAL_DESC3_FD_WIDTH		1
 #define RX_NORMAL_DESC3_INTE_INDEX		30
 #define RX_NORMAL_DESC3_INTE_WIDTH		1
+#define RX_NORMAL_DESC3_L34T_INDEX		20
+#define RX_NORMAL_DESC3_L34T_WIDTH		4
 #define RX_NORMAL_DESC3_LD_INDEX		28
 #define RX_NORMAL_DESC3_LD_WIDTH		1
 #define RX_NORMAL_DESC3_OWN_INDEX		31
 #define RX_NORMAL_DESC3_OWN_WIDTH		1
 #define RX_NORMAL_DESC3_PL_INDEX		0
 #define RX_NORMAL_DESC3_PL_WIDTH		14
+#define RX_NORMAL_DESC3_RSV_INDEX		26
+#define RX_NORMAL_DESC3_RSV_WIDTH		1
+
+#define RX_DESC3_L34T_IPV4_TCP			1
+#define RX_DESC3_L34T_IPV4_UDP			2
+#define RX_DESC3_L34T_IPV4_ICMP			3
+#define RX_DESC3_L34T_IPV6_TCP			9
+#define RX_DESC3_L34T_IPV6_UDP			10
+#define RX_DESC3_L34T_IPV6_ICMP			11
 
 #define RX_CONTEXT_DESC3_TSA_INDEX		4
 #define RX_CONTEXT_DESC3_TSA_WIDTH		1
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
index 6fc5da0..a50891f5 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
@@ -117,7 +117,7 @@
 #include "xgbe.h"
 #include "xgbe-common.h"
 
-static void xgbe_unmap_skb(struct xgbe_prv_data *, struct xgbe_ring_data *);
+static void xgbe_unmap_rdata(struct xgbe_prv_data *, struct xgbe_ring_data *);
 
 static void xgbe_free_ring(struct xgbe_prv_data *pdata,
 			   struct xgbe_ring *ring)
@@ -131,13 +131,35 @@ static void xgbe_free_ring(struct xgbe_prv_data *pdata,
 	if (ring->rdata) {
 		for (i = 0; i < ring->rdesc_count; i++) {
 			rdata = XGBE_GET_DESC_DATA(ring, i);
-			xgbe_unmap_skb(pdata, rdata);
+			xgbe_unmap_rdata(pdata, rdata);
 		}
 
 		kfree(ring->rdata);
 		ring->rdata = NULL;
 	}
 
+	if (ring->rx_hdr_pa.pages) {
+		dma_unmap_page(pdata->dev, ring->rx_hdr_pa.pages_dma,
+			       ring->rx_hdr_pa.pages_len, DMA_FROM_DEVICE);
+		put_page(ring->rx_hdr_pa.pages);
+
+		ring->rx_hdr_pa.pages = NULL;
+		ring->rx_hdr_pa.pages_len = 0;
+		ring->rx_hdr_pa.pages_offset = 0;
+		ring->rx_hdr_pa.pages_dma = 0;
+	}
+
+	if (ring->rx_buf_pa.pages) {
+		dma_unmap_page(pdata->dev, ring->rx_buf_pa.pages_dma,
+			       ring->rx_buf_pa.pages_len, DMA_FROM_DEVICE);
+		put_page(ring->rx_buf_pa.pages);
+
+		ring->rx_buf_pa.pages = NULL;
+		ring->rx_buf_pa.pages_len = 0;
+		ring->rx_buf_pa.pages_offset = 0;
+		ring->rx_buf_pa.pages_dma = 0;
+	}
+
 	if (ring->rdesc) {
 		dma_free_coherent(pdata->dev,
 				  (sizeof(struct xgbe_ring_desc) *
@@ -233,6 +255,96 @@ err_ring:
 	return ret;
 }
 
+static int xgbe_alloc_pages(struct xgbe_prv_data *pdata,
+			    struct xgbe_page_alloc *pa, gfp_t gfp, int order)
+{
+	struct page *pages = NULL;
+	dma_addr_t pages_dma;
+	int ret;
+
+	/* Try to obtain pages, decreasing order if necessary */
+	gfp |= __GFP_COLD | __GFP_COMP;
+	while (order >= 0) {
+		pages = alloc_pages(gfp, order);
+		if (pages)
+			break;
+
+		order--;
+	}
+	if (!pages)
+		return -ENOMEM;
+
+	/* Map the pages */
+	pages_dma = dma_map_page(pdata->dev, pages, 0,
+				 PAGE_SIZE << order, DMA_FROM_DEVICE);
+	ret = dma_mapping_error(pdata->dev, pages_dma);
+	if (ret) {
+		put_page(pages);
+		return ret;
+	}
+
+	pa->pages = pages;
+	pa->pages_len = PAGE_SIZE << order;
+	pa->pages_offset = 0;
+	pa->pages_dma = pages_dma;
+
+	return 0;
+}
+
+static void xgbe_set_buffer_data(struct xgbe_buffer_data *bd,
+				 struct xgbe_page_alloc *pa,
+				 unsigned int len)
+{
+	get_page(pa->pages);
+	bd->pa = *pa;
+
+	bd->dma = pa->pages_dma + pa->pages_offset;
+	bd->dma_len = len;
+
+	pa->pages_offset += len;
+	if ((pa->pages_offset + len) > pa->pages_len) {
+		/* This data descriptor is responsible for unmapping page(s) */
+		bd->pa_unmap = *pa;
+
+		/* Get a new allocation next time */
+		pa->pages = NULL;
+		pa->pages_len = 0;
+		pa->pages_offset = 0;
+		pa->pages_dma = 0;
+	}
+}
+
+static int xgbe_map_rx_buffer(struct xgbe_prv_data *pdata,
+			      struct xgbe_ring *ring,
+			      struct xgbe_ring_data *rdata)
+{
+	int order, ret;
+
+	if (!ring->rx_hdr_pa.pages) {
+		ret = xgbe_alloc_pages(pdata, &ring->rx_hdr_pa, GFP_ATOMIC, 0);
+		if (ret)
+			return ret;
+	}
+
+	if (!ring->rx_buf_pa.pages) {
+		order = max_t(int, PAGE_ALLOC_COSTLY_ORDER - 1, 0);
+		ret = xgbe_alloc_pages(pdata, &ring->rx_buf_pa, GFP_ATOMIC,
+				       order);
+		if (ret)
+			return ret;
+	}
+
+	/* Set up the header page info */
+	xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa,
+			     XGBE_SKB_ALLOC_SIZE);
+
+	/* Set up the buffer page info */
+	xgbe_set_buffer_data(&rdata->rx.buf, &ring->rx_buf_pa,
+			     pdata->rx_buf_size);
+
+	return 0;
+}
+
 static void xgbe_wrapper_tx_descriptor_init(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
@@ -266,7 +378,7 @@ static void xgbe_wrapper_tx_descriptor_init(struct xgbe_prv_data *pdata)
 
 		ring->cur = 0;
 		ring->dirty = 0;
-		ring->tx.queue_stopped = 0;
+		memset(&ring->tx, 0, sizeof(ring->tx));
 
 		hw_if->tx_desc_init(channel);
 	}
@@ -281,8 +393,7 @@ static void xgbe_wrapper_rx_descriptor_init(struct xgbe_prv_data *pdata)
 	struct xgbe_ring *ring;
 	struct xgbe_ring_desc *rdesc;
 	struct xgbe_ring_data *rdata;
-	dma_addr_t rdesc_dma, skb_dma;
-	struct sk_buff *skb = NULL;
+	dma_addr_t rdesc_dma;
 	unsigned int i, j;
 
 	DBGPR("-->xgbe_wrapper_rx_descriptor_init\n");
@@ -302,22 +413,8 @@ static void xgbe_wrapper_rx_descriptor_init(struct xgbe_prv_data *pdata)
 			rdata->rdesc = rdesc;
 			rdata->rdesc_dma = rdesc_dma;
 
-			/* Allocate skb & assign to each rdesc */
-			skb = dev_alloc_skb(pdata->rx_buf_size);
-			if (skb == NULL)
-				break;
-			skb_dma = dma_map_single(pdata->dev, skb->data,
-						 pdata->rx_buf_size,
-						 DMA_FROM_DEVICE);
-			if (dma_mapping_error(pdata->dev, skb_dma)) {
-				netdev_alert(pdata->netdev,
-					     "failed to do the dma map\n");
-				dev_kfree_skb_any(skb);
+			if (xgbe_map_rx_buffer(pdata, ring, rdata))
 				break;
-			}
-			rdata->skb = skb;
-			rdata->skb_dma = skb_dma;
-			rdata->skb_dma_len = pdata->rx_buf_size;
 
 			rdesc++;
 			rdesc_dma += sizeof(struct xgbe_ring_desc);
@@ -325,8 +422,7 @@ static void xgbe_wrapper_rx_descriptor_init(struct xgbe_prv_data *pdata)
 
 		ring->cur = 0;
 		ring->dirty = 0;
-		ring->rx.realloc_index = 0;
-		ring->rx.realloc_threshold = 0;
+		memset(&ring->rx, 0, sizeof(ring->rx));
 
 		hw_if->rx_desc_init(channel);
 	}
@@ -334,8 +430,8 @@ static void xgbe_wrapper_rx_descriptor_init(struct xgbe_prv_data *pdata)
 	DBGPR("<--xgbe_wrapper_rx_descriptor_init\n");
 }
 
-static void xgbe_unmap_skb(struct xgbe_prv_data *pdata,
-			   struct xgbe_ring_data *rdata)
+static void xgbe_unmap_rdata(struct xgbe_prv_data *pdata,
+			     struct xgbe_ring_data *rdata)
 {
 	if (rdata->skb_dma) {
 		if (rdata->mapped_as_page) {
@@ -354,9 +450,29 @@ static void xgbe_unmap_skb(struct xgbe_prv_data *pdata,
 		rdata->skb = NULL;
 	}
 
-	rdata->tso_header = 0;
-	rdata->len = 0;
-	rdata->interrupt = 0;
+	if (rdata->rx.hdr.pa.pages)
+		put_page(rdata->rx.hdr.pa.pages);
+
+	if (rdata->rx.hdr.pa_unmap.pages) {
+		dma_unmap_page(pdata->dev, rdata->rx.hdr.pa_unmap.pages_dma,
+			       rdata->rx.hdr.pa_unmap.pages_len,
+			       DMA_FROM_DEVICE);
+		put_page(rdata->rx.hdr.pa_unmap.pages);
+	}
+
+	if (rdata->rx.buf.pa.pages)
+		put_page(rdata->rx.buf.pa.pages);
+
+	if (rdata->rx.buf.pa_unmap.pages) {
+		dma_unmap_page(pdata->dev, rdata->rx.buf.pa_unmap.pages_dma,
+			       rdata->rx.buf.pa_unmap.pages_len,
+			       DMA_FROM_DEVICE);
+		put_page(rdata->rx.buf.pa_unmap.pages);
+	}
+
+	memset(&rdata->tx, 0, sizeof(rdata->tx));
+	memset(&rdata->rx, 0, sizeof(rdata->rx));
+
 	rdata->mapped_as_page = 0;
 
 	if (rdata->state_saved) {
@@ -414,7 +530,6 @@ static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb)
 		}
 		rdata->skb_dma = skb_dma;
 		rdata->skb_dma_len = packet->header_len;
-		rdata->tso_header = 1;
 
 		offset = packet->header_len;
 
@@ -481,7 +596,11 @@ static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb)
 		}
 	}
 
-	/* Save the skb address in the last entry */
+	/* Save the skb address in the last entry. We always have some data
+	 * that has been mapped so rdata is always advanced past the last
+	 * piece of mapped data - use the entry pointed to by cur_index - 1.
+	 */
+	rdata = XGBE_GET_DESC_DATA(ring, cur_index - 1);
 	rdata->skb = skb;
 
 	/* Save the number of descriptor entries used */
@@ -494,7 +613,7 @@ static int xgbe_map_tx_skb(struct xgbe_channel *channel, struct sk_buff *skb)
 err_out:
 	while (start_index < cur_index) {
 		rdata = XGBE_GET_DESC_DATA(ring, start_index++);
-		xgbe_unmap_skb(pdata, rdata);
+		xgbe_unmap_rdata(pdata, rdata);
 	}
 
 	DBGPR("<--xgbe_map_tx_skb: count=0\n");
@@ -502,40 +621,25 @@ err_out:
 	return 0;
 }
 
-static void xgbe_realloc_skb(struct xgbe_channel *channel)
+static void xgbe_realloc_rx_buffer(struct xgbe_channel *channel)
 {
 	struct xgbe_prv_data *pdata = channel->pdata;
 	struct xgbe_hw_if *hw_if = &pdata->hw_if;
 	struct xgbe_ring *ring = channel->rx_ring;
 	struct xgbe_ring_data *rdata;
-	struct sk_buff *skb = NULL;
-	dma_addr_t skb_dma;
 	int i;
 
-	DBGPR("-->xgbe_realloc_skb: rx_ring->rx.realloc_index = %u\n",
+	DBGPR("-->xgbe_realloc_rx_buffer: rx_ring->rx.realloc_index = %u\n",
 	      ring->rx.realloc_index);
 
 	for (i = 0; i < ring->dirty; i++) {
 		rdata = XGBE_GET_DESC_DATA(ring, ring->rx.realloc_index);
 
 		/* Reset rdata values */
-		xgbe_unmap_skb(pdata, rdata);
+		xgbe_unmap_rdata(pdata, rdata);
 
-		/* Allocate skb & assign to each rdesc */
-		skb = dev_alloc_skb(pdata->rx_buf_size);
-		if (skb == NULL)
+		if (xgbe_map_rx_buffer(pdata, ring, rdata))
 			break;
-		skb_dma = dma_map_single(pdata->dev, skb->data,
-					 pdata->rx_buf_size, DMA_FROM_DEVICE);
-		if (dma_mapping_error(pdata->dev, skb_dma)) {
-			netdev_alert(pdata->netdev,
-				     "failed to do the dma map\n");
-			dev_kfree_skb_any(skb);
-			break;
-		}
-		rdata->skb = skb;
-		rdata->skb_dma = skb_dma;
-		rdata->skb_dma_len = pdata->rx_buf_size;
 
 		hw_if->rx_desc_reset(rdata);
 
@@ -543,7 +647,7 @@ static void xgbe_realloc_skb(struct xgbe_channel *channel)
 	}
 	ring->dirty = 0;
 
-	DBGPR("<--xgbe_realloc_skb\n");
+	DBGPR("<--xgbe_realloc_rx_buffer\n");
 }
 
 void xgbe_init_function_ptrs_desc(struct xgbe_desc_if *desc_if)
@@ -553,8 +657,8 @@ void xgbe_init_function_ptrs_desc(struct xgbe_desc_if *desc_if)
 	desc_if->alloc_ring_resources = xgbe_alloc_ring_resources;
 	desc_if->free_ring_resources = xgbe_free_ring_resources;
 	desc_if->map_tx_skb = xgbe_map_tx_skb;
-	desc_if->realloc_skb = xgbe_realloc_skb;
-	desc_if->unmap_skb = xgbe_unmap_skb;
+	desc_if->realloc_rx_buffer = xgbe_realloc_rx_buffer;
+	desc_if->unmap_rdata = xgbe_unmap_rdata;
 	desc_if->wrapper_tx_desc_init = xgbe_wrapper_tx_descriptor_init;
 	desc_if->wrapper_rx_desc_init = xgbe_wrapper_rx_descriptor_init;
 
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
index 9da3a03..53f5f66 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -335,6 +335,161 @@ static void xgbe_config_tso_mode(struct xgbe_prv_data *pdata)
 	}
 }
 
+static void xgbe_config_sph_mode(struct xgbe_prv_data *pdata)
+{
+	struct xgbe_channel *channel;
+	unsigned int i;
+
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(channel, DMA_CH_CR, SPH, 1);
+	}
+
+	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, HDSMS, XGBE_SPH_HDSMS_SIZE);
+}
+
+static int xgbe_write_rss_reg(struct xgbe_prv_data *pdata, unsigned int type,
+			      unsigned int index, unsigned int val)
+{
+	unsigned int wait;
+	int ret = 0;
+
+	mutex_lock(&pdata->rss_mutex);
+
+	if (XGMAC_IOREAD_BITS(pdata, MAC_RSSAR, OB)) {
+		ret = -EBUSY;
+		goto unlock;
+	}
+
+	XGMAC_IOWRITE(pdata, MAC_RSSDR, val);
+
+	XGMAC_IOWRITE_BITS(pdata, MAC_RSSAR, RSSIA, index);
+	XGMAC_IOWRITE_BITS(pdata, MAC_RSSAR, ADDRT, type);
+	XGMAC_IOWRITE_BITS(pdata, MAC_RSSAR, CT, 0);
+	XGMAC_IOWRITE_BITS(pdata, MAC_RSSAR, OB, 1);
+
+	wait = 1000;
+	while (wait--) {
+		if (!XGMAC_IOREAD_BITS(pdata, MAC_RSSAR, OB))
+			goto unlock;
+
+		usleep_range(1000, 1500);
+	}
+
+	ret = -EBUSY;
+
+unlock:
+	mutex_unlock(&pdata->rss_mutex);
+
+	return ret;
+}
+
+static int xgbe_write_rss_hash_key(struct xgbe_prv_data *pdata)
+{
+	unsigned int key_regs = sizeof(pdata->rss_key) / sizeof(u32);
+	unsigned int *key = (unsigned int *)&pdata->rss_key;
+	int ret;
+
+	while (key_regs--) {
+		ret = xgbe_write_rss_reg(pdata, XGBE_RSS_HASH_KEY_TYPE,
+					 key_regs, *key++);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int xgbe_write_rss_lookup_table(struct xgbe_prv_data *pdata)
+{
+	unsigned int i;
+	int ret;
+
+	for (i = 0; i < ARRAY_SIZE(pdata->rss_table); i++) {
+		ret = xgbe_write_rss_reg(pdata,
+					 XGBE_RSS_LOOKUP_TABLE_TYPE, i,
+					 pdata->rss_table[i]);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int xgbe_set_rss_hash_key(struct xgbe_prv_data *pdata, const u8 *key)
+{
+	memcpy(pdata->rss_key, key, sizeof(pdata->rss_key));
+
+	return xgbe_write_rss_hash_key(pdata);
+}
+
+static int xgbe_set_rss_lookup_table(struct xgbe_prv_data *pdata,
+				     const u32 *table)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(pdata->rss_table); i++)
+		XGMAC_SET_BITS(pdata->rss_table[i], MAC_RSSDR, DMCH, table[i]);
+
+	return xgbe_write_rss_lookup_table(pdata);
+}
+
+static int xgbe_enable_rss(struct xgbe_prv_data *pdata)
+{
+	int ret;
+
+	if (!pdata->hw_feat.rss)
+		return -EOPNOTSUPP;
+
+	/* Program the hash key */
+	ret = xgbe_write_rss_hash_key(pdata);
+	if (ret)
+		return ret;
+
+	/* Program the lookup table */
+	ret = xgbe_write_rss_lookup_table(pdata);
+	if (ret)
+		return ret;
+
+	/* Set the RSS options */
+	XGMAC_IOWRITE(pdata, MAC_RSSCR, pdata->rss_options);
+
+	/* Enable RSS */
+	XGMAC_IOWRITE_BITS(pdata, MAC_RSSCR, RSSE, 1);
+
+	return 0;
+}
+
+static int xgbe_disable_rss(struct xgbe_prv_data *pdata)
+{
+	if (!pdata->hw_feat.rss)
+		return -EOPNOTSUPP;
+
+	XGMAC_IOWRITE_BITS(pdata, MAC_RSSCR, RSSE, 0);
+
+	return 0;
+}
+
+static void xgbe_config_rss(struct xgbe_prv_data *pdata)
+{
+	int ret;
+
+	if (!pdata->hw_feat.rss)
+		return;
+
+	if (pdata->netdev->features & NETIF_F_RXHASH)
+		ret = xgbe_enable_rss(pdata);
+	else
+		ret = xgbe_disable_rss(pdata);
+
+	if (ret)
+		netdev_err(pdata->netdev,
+			   "error configuring RSS, RSS disabled\n");
+}
+
 static int xgbe_disable_tx_flow_control(struct xgbe_prv_data *pdata)
 {
 	unsigned int max_q_count, q_count;
@@ -465,17 +620,21 @@ static void xgbe_enable_dma_interrupts(struct xgbe_prv_data *pdata)
 
 		if (channel->tx_ring) {
 			/* Enable the following Tx interrupts
-			 *   TIE  - Transmit Interrupt Enable (unless polling)
+			 *   TIE  - Transmit Interrupt Enable (unless using
+			 *          per channel interrupts)
 			 */
-			XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TIE, 1);
+			if (!pdata->per_channel_irq)
+				XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, TIE, 1);
 		}
 		if (channel->rx_ring) {
 			/* Enable following Rx interrupts
 			 *   RBUE - Receive Buffer Unavailable Enable
-			 *   RIE  - Receive Interrupt Enable
+			 *   RIE  - Receive Interrupt Enable (unless using
+			 *          per channel interrupts)
 			 */
 			XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RBUE, 1);
-			XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RIE, 1);
+			if (!pdata->per_channel_irq)
+				XGMAC_SET_BITS(dma_ch_ier, DMA_CH_IER, RIE, 1);
 		}
 
 		XGMAC_DMA_IOWRITE(channel, DMA_CH_IER, dma_ch_ier);
@@ -880,13 +1039,15 @@ static void xgbe_tx_desc_reset(struct xgbe_ring_data *rdata)
 	rdesc->desc1 = 0;
 	rdesc->desc2 = 0;
 	rdesc->desc3 = 0;
+
+	/* Make sure ownership is written to the descriptor */
+	wmb();
 }
 
 static void xgbe_tx_desc_init(struct xgbe_channel *channel)
 {
 	struct xgbe_ring *ring = channel->tx_ring;
 	struct xgbe_ring_data *rdata;
-	struct xgbe_ring_desc *rdesc;
 	int i;
 	int start_index = ring->cur;
 
@@ -895,26 +1056,11 @@ static void xgbe_tx_desc_init(struct xgbe_channel *channel)
 	/* Initialze all descriptors */
 	for (i = 0; i < ring->rdesc_count; i++) {
 		rdata = XGBE_GET_DESC_DATA(ring, i);
-		rdesc = rdata->rdesc;
 
-		/* Initialize Tx descriptor
-		 *   Set buffer 1 (lo) address to zero
-		 *   Set buffer 1 (hi) address to zero
-		 *   Reset all other control bits (IC, TTSE, B2L & B1L)
-		 *   Reset all other control bits (OWN, CTXT, FD, LD, CPC, CIC,
-		 *     etc)
-		 */
-		rdesc->desc0 = 0;
-		rdesc->desc1 = 0;
-		rdesc->desc2 = 0;
-		rdesc->desc3 = 0;
+		/* Initialize Tx descriptor */
+		xgbe_tx_desc_reset(rdata);
 	}
 
-	/* Make sure everything is written to the descriptor(s) before
-	 * telling the device about them
-	 */
-	wmb();
-
 	/* Update the total number of Tx descriptors */
 	XGMAC_DMA_IOWRITE(channel, DMA_CH_TDRLR, ring->rdesc_count - 1);
 
@@ -933,19 +1079,19 @@ static void xgbe_rx_desc_reset(struct xgbe_ring_data *rdata)
 	struct xgbe_ring_desc *rdesc = rdata->rdesc;
 
 	/* Reset the Rx descriptor
-	 *   Set buffer 1 (lo) address to dma address (lo)
-	 *   Set buffer 1 (hi) address to dma address (hi)
-	 *   Set buffer 2 (lo) address to zero
-	 *   Set buffer 2 (hi) address to zero and set control bits
-	 *     OWN and INTE
+	 *   Set buffer 1 (lo) address to header dma address (lo)
+	 *   Set buffer 1 (hi) address to header dma address (hi)
+	 *   Set buffer 2 (lo) address to buffer dma address (lo)
+	 *   Set buffer 2 (hi) address to buffer dma address (hi) and
+	 *     set control bits OWN and INTE
 	 */
-	rdesc->desc0 = cpu_to_le32(lower_32_bits(rdata->skb_dma));
-	rdesc->desc1 = cpu_to_le32(upper_32_bits(rdata->skb_dma));
-	rdesc->desc2 = 0;
+	rdesc->desc0 = cpu_to_le32(lower_32_bits(rdata->rx.hdr.dma));
+	rdesc->desc1 = cpu_to_le32(upper_32_bits(rdata->rx.hdr.dma));
+	rdesc->desc2 = cpu_to_le32(lower_32_bits(rdata->rx.buf.dma));
+	rdesc->desc3 = cpu_to_le32(upper_32_bits(rdata->rx.buf.dma));
 
-	rdesc->desc3 = 0;
-	if (rdata->interrupt)
-		XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, INTE, 1);
+	XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, INTE,
+			  rdata->interrupt ? 1 : 0);
 
 	/* Since the Rx DMA engine is likely running, make sure everything
 	 * is written to the descriptor(s) before setting the OWN bit
@@ -964,7 +1110,6 @@ static void xgbe_rx_desc_init(struct xgbe_channel *channel)
 	struct xgbe_prv_data *pdata = channel->pdata;
 	struct xgbe_ring *ring = channel->rx_ring;
 	struct xgbe_ring_data *rdata;
-	struct xgbe_ring_desc *rdesc;
 	unsigned int start_index = ring->cur;
 	unsigned int rx_coalesce, rx_frames;
 	unsigned int i;
@@ -977,34 +1122,16 @@ static void xgbe_rx_desc_init(struct xgbe_channel *channel)
 	/* Initialize all descriptors */
 	for (i = 0; i < ring->rdesc_count; i++) {
 		rdata = XGBE_GET_DESC_DATA(ring, i);
-		rdesc = rdata->rdesc;
 
-		/* Initialize Rx descriptor
-		 *   Set buffer 1 (lo) address to dma address (lo)
-		 *   Set buffer 1 (hi) address to dma address (hi)
-		 *   Set buffer 2 (lo) address to zero
-		 *   Set buffer 2 (hi) address to zero and set control
-		 *     bits OWN and INTE appropriateley
-		 */
-		rdesc->desc0 = cpu_to_le32(lower_32_bits(rdata->skb_dma));
-		rdesc->desc1 = cpu_to_le32(upper_32_bits(rdata->skb_dma));
-		rdesc->desc2 = 0;
-		rdesc->desc3 = 0;
-		XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, OWN, 1);
-		XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, INTE, 1);
-		rdata->interrupt = 1;
-		if (rx_coalesce && (!rx_frames || ((i + 1) % rx_frames))) {
-			/* Clear interrupt on completion bit */
-			XGMAC_SET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, INTE,
-					  0);
+		/* Set interrupt on completion bit as appropriate */
+		if (rx_coalesce && (!rx_frames || ((i + 1) % rx_frames)))
 			rdata->interrupt = 0;
-		}
-	}
+		else
+			rdata->interrupt = 1;
 
-	/* Make sure everything is written to the descriptors before
-	 * telling the device about them
-	 */
-	wmb();
+		/* Initialize Rx descriptor */
+		xgbe_rx_desc_reset(rdata);
+	}
 
 	/* Update the total number of Rx descriptors */
 	XGMAC_DMA_IOWRITE(channel, DMA_CH_RDRLR, ring->rdesc_count - 1);
@@ -1198,7 +1325,30 @@ static void xgbe_config_dcb_pfc(struct xgbe_prv_data *pdata)
 	xgbe_config_flow_control(pdata);
 }
 
-static void xgbe_pre_xmit(struct xgbe_channel *channel)
+static void xgbe_tx_start_xmit(struct xgbe_channel *channel,
+			       struct xgbe_ring *ring)
+{
+	struct xgbe_prv_data *pdata = channel->pdata;
+	struct xgbe_ring_data *rdata;
+
+	/* Issue a poll command to Tx DMA by writing address
+	 * of next immediate free descriptor */
+	rdata = XGBE_GET_DESC_DATA(ring, ring->cur);
+	XGMAC_DMA_IOWRITE(channel, DMA_CH_TDTR_LO,
+			  lower_32_bits(rdata->rdesc_dma));
+
+	/* Start the Tx coalescing timer */
+	if (pdata->tx_usecs && !channel->tx_timer_active) {
+		channel->tx_timer_active = 1;
+		hrtimer_start(&channel->tx_timer,
+			      ktime_set(0, pdata->tx_usecs * NSEC_PER_USEC),
+			      HRTIMER_MODE_REL);
+	}
+
+	ring->tx.xmit_more = 0;
+}
+
+static void xgbe_dev_xmit(struct xgbe_channel *channel)
 {
 	struct xgbe_prv_data *pdata = channel->pdata;
 	struct xgbe_ring *ring = channel->tx_ring;
@@ -1207,11 +1357,11 @@ static void xgbe_pre_xmit(struct xgbe_channel *channel)
 	struct xgbe_packet_data *packet = &ring->packet_data;
 	unsigned int csum, tso, vlan;
 	unsigned int tso_context, vlan_context;
-	unsigned int tx_coalesce, tx_frames;
+	unsigned int tx_set_ic;
 	int start_index = ring->cur;
 	int i;
 
-	DBGPR("-->xgbe_pre_xmit\n");
+	DBGPR("-->xgbe_dev_xmit\n");
 
 	csum = XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES,
 			      CSUM_ENABLE);
@@ -1230,10 +1380,26 @@ static void xgbe_pre_xmit(struct xgbe_channel *channel)
 	else
 		vlan_context = 0;
 
-	tx_coalesce = (pdata->tx_usecs || pdata->tx_frames) ? 1 : 0;
-	tx_frames = pdata->tx_frames;
-	if (tx_coalesce && !channel->tx_timer_active)
-		ring->coalesce_count = 0;
+	/* Determine if an interrupt should be generated for this Tx:
+	 *   Interrupt:
+	 *     - Tx frame count exceeds the frame count setting
+	 *     - Addition of Tx frame count to the frame count since the
+	 *       last interrupt was set exceeds the frame count setting
+	 *   No interrupt:
+	 *     - No frame count setting specified (ethtool -C ethX tx-frames 0)
+	 *     - Addition of Tx frame count to the frame count since the
+	 *       last interrupt was set does not exceed the frame count setting
+	 */
+	ring->coalesce_count += packet->tx_packets;
+	if (!pdata->tx_frames)
+		tx_set_ic = 0;
+	else if (packet->tx_packets > pdata->tx_frames)
+		tx_set_ic = 1;
+	else if ((ring->coalesce_count % pdata->tx_frames) <
+		 packet->tx_packets)
+		tx_set_ic = 1;
+	else
+		tx_set_ic = 0;
 
 	rdata = XGBE_GET_DESC_DATA(ring, ring->cur);
 	rdesc = rdata->rdesc;
@@ -1300,13 +1466,6 @@ static void xgbe_pre_xmit(struct xgbe_channel *channel)
 	if (XGMAC_GET_BITS(packet->attributes, TX_PACKET_ATTRIBUTES, PTP))
 		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, TTSE, 1);
 
-	/* Set IC bit based on Tx coalescing settings */
-	XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 1);
-	if (tx_coalesce && (!tx_frames ||
-			    (++ring->coalesce_count % tx_frames)))
-		/* Clear IC bit */
-		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 0);
-
 	/* Mark it as First Descriptor */
 	XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, FD, 1);
 
@@ -1351,13 +1510,6 @@ static void xgbe_pre_xmit(struct xgbe_channel *channel)
 		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, HL_B1L,
 				  rdata->skb_dma_len);
 
-		/* Set IC bit based on Tx coalescing settings */
-		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 1);
-		if (tx_coalesce && (!tx_frames ||
-				    (++ring->coalesce_count % tx_frames)))
-			/* Clear IC bit */
-			XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 0);
-
 		/* Set OWN bit */
 		XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, OWN, 1);
 
@@ -1373,6 +1525,14 @@ static void xgbe_pre_xmit(struct xgbe_channel *channel)
 	/* Set LAST bit for the last descriptor */
 	XGMAC_SET_BITS_LE(rdesc->desc3, TX_NORMAL_DESC3, LD, 1);
 
+	/* Set IC bit based on Tx coalescing settings */
+	if (tx_set_ic)
+		XGMAC_SET_BITS_LE(rdesc->desc2, TX_NORMAL_DESC2, IC, 1);
+
+	/* Save the Tx info to report back during cleanup */
+	rdata->tx.packets = packet->tx_packets;
+	rdata->tx.bytes = packet->tx_bytes;
+
 	/* In case the Tx DMA engine is running, make sure everything
 	 * is written to the descriptor(s) before setting the OWN bit
 	 * for the first descriptor
@@ -1391,26 +1551,19 @@ static void xgbe_pre_xmit(struct xgbe_channel *channel)
 	/* Make sure ownership is written to the descriptor */
 	wmb();
 
-	/* Issue a poll command to Tx DMA by writing address
-	 * of next immediate free descriptor */
 	ring->cur++;
-	rdata = XGBE_GET_DESC_DATA(ring, ring->cur);
-	XGMAC_DMA_IOWRITE(channel, DMA_CH_TDTR_LO,
-			  lower_32_bits(rdata->rdesc_dma));
-
-	/* Start the Tx coalescing timer */
-	if (tx_coalesce && !channel->tx_timer_active) {
-		channel->tx_timer_active = 1;
-		hrtimer_start(&channel->tx_timer,
-			      ktime_set(0, pdata->tx_usecs * NSEC_PER_USEC),
-			      HRTIMER_MODE_REL);
-	}
+	if (!packet->skb->xmit_more ||
+	    netif_xmit_stopped(netdev_get_tx_queue(pdata->netdev,
+						   channel->queue_index)))
+		xgbe_tx_start_xmit(channel, ring);
+	else
+		ring->tx.xmit_more = 1;
 
 	DBGPR("  %s: descriptors %u to %u written\n",
 	      channel->name, start_index & (ring->rdesc_count - 1),
 	      (ring->cur - 1) & (ring->rdesc_count - 1));
 
-	DBGPR("<--xgbe_pre_xmit\n");
+	DBGPR("<--xgbe_dev_xmit\n");
 }
 
 static int xgbe_dev_read(struct xgbe_channel *channel)
@@ -1420,7 +1573,7 @@ static int xgbe_dev_read(struct xgbe_channel *channel)
 	struct xgbe_ring_desc *rdesc;
 	struct xgbe_packet_data *packet = &ring->packet_data;
 	struct net_device *netdev = channel->pdata->netdev;
-	unsigned int err, etlt;
+	unsigned int err, etlt, l34t;
 
 	DBGPR("-->xgbe_dev_read: cur = %d\n", ring->cur);
 
@@ -1431,6 +1584,9 @@ static int xgbe_dev_read(struct xgbe_channel *channel)
 	if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, OWN))
 		return 1;
 
+	/* Make sure descriptor fields are read after reading the OWN bit */
+	rmb();
+
 #ifdef XGMAC_ENABLE_RX_DESC_DUMP
 	xgbe_dump_rx_desc(ring, rdesc, ring->cur);
 #endif
@@ -1454,8 +1610,33 @@ static int xgbe_dev_read(struct xgbe_channel *channel)
 		XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
 			       CONTEXT_NEXT, 1);
 
+	/* Get the header length */
+	if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, FD))
+		rdata->rx.hdr_len = XGMAC_GET_BITS_LE(rdesc->desc2,
+						      RX_NORMAL_DESC2, HL);
+
+	/* Get the RSS hash */
+	if (XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, RSV)) {
+		XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
+			       RSS_HASH, 1);
+
+		packet->rss_hash = le32_to_cpu(rdesc->desc1);
+
+		l34t = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, L34T);
+		switch (l34t) {
+		case RX_DESC3_L34T_IPV4_TCP:
+		case RX_DESC3_L34T_IPV4_UDP:
+		case RX_DESC3_L34T_IPV6_TCP:
+		case RX_DESC3_L34T_IPV6_UDP:
+			packet->rss_hash_type = PKT_HASH_TYPE_L4;
+			break;
+		default:
+			packet->rss_hash_type = PKT_HASH_TYPE_L3;
+		}
+	}
+
 	/* Get the packet length */
-	rdata->len = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, PL);
+	rdata->rx.len = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, PL);
 
 	if (!XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, LD)) {
 		/* Not all the data has been transferred for this packet */
@@ -1478,7 +1659,8 @@ static int xgbe_dev_read(struct xgbe_channel *channel)
 	etlt = XGMAC_GET_BITS_LE(rdesc->desc3, RX_NORMAL_DESC3, ETLT);
 	DBGPR("  err=%u, etlt=%#x\n", err, etlt);
 
-	if (!err || (err && !etlt)) {
+	if (!err || !etlt) {
+		/* No error if err is 0 or etlt is 0 */
 		if ((etlt == 0x09) &&
 		    (netdev->features & NETIF_F_HW_VLAN_CTAG_RX)) {
 			XGMAC_SET_BITS(packet->attributes, RX_PACKET_ATTRIBUTES,
@@ -2298,6 +2480,47 @@ static void xgbe_config_mmc(struct xgbe_prv_data *pdata)
 	XGMAC_IOWRITE_BITS(pdata, MMC_CR, CR, 1);
 }
 
+static void xgbe_prepare_tx_stop(struct xgbe_prv_data *pdata,
+				 struct xgbe_channel *channel)
+{
+	unsigned int tx_dsr, tx_pos, tx_qidx;
+	unsigned int tx_status;
+	unsigned long tx_timeout;
+
+	/* Calculate the status register to read and the position within */
+	if (channel->queue_index < DMA_DSRX_FIRST_QUEUE) {
+		tx_dsr = DMA_DSR0;
+		tx_pos = (channel->queue_index * DMA_DSR_Q_WIDTH) +
+			 DMA_DSR0_TPS_START;
+	} else {
+		tx_qidx = channel->queue_index - DMA_DSRX_FIRST_QUEUE;
+
+		tx_dsr = DMA_DSR1 + ((tx_qidx / DMA_DSRX_QPR) * DMA_DSRX_INC);
+		tx_pos = ((tx_qidx % DMA_DSRX_QPR) * DMA_DSR_Q_WIDTH) +
+			 DMA_DSRX_TPS_START;
+	}
+
+	/* The Tx engine cannot be stopped if it is actively processing
+	 * descriptors. Wait for the Tx engine to enter the stopped or
+	 * suspended state.  Don't wait forever though...
+	 */
+	tx_timeout = jiffies + (XGBE_DMA_STOP_TIMEOUT * HZ);
+	while (time_before(jiffies, tx_timeout)) {
+		tx_status = XGMAC_IOREAD(pdata, tx_dsr);
+		tx_status = GET_BITS(tx_status, tx_pos, DMA_DSR_TPS_WIDTH);
+		if ((tx_status == DMA_TPS_STOPPED) ||
+		    (tx_status == DMA_TPS_SUSPENDED))
+			break;
+
+		usleep_range(500, 1000);
+	}
+
+	if (!time_before(jiffies, tx_timeout))
+		netdev_info(pdata->netdev,
+			    "timed out waiting for Tx DMA channel %u to stop\n",
+			    channel->queue_index);
+}
+
 static void xgbe_enable_tx(struct xgbe_prv_data *pdata)
 {
 	struct xgbe_channel *channel;
@@ -2326,6 +2549,15 @@ static void xgbe_disable_tx(struct xgbe_prv_data *pdata)
 	struct xgbe_channel *channel;
 	unsigned int i;
 
+	/* Prepare for Tx DMA channel stop */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		xgbe_prepare_tx_stop(pdata, channel);
+	}
+
 	/* Disable MAC Tx */
 	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, TE, 0);
 
@@ -2417,6 +2649,15 @@ static void xgbe_powerdown_tx(struct xgbe_prv_data *pdata)
 	struct xgbe_channel *channel;
 	unsigned int i;
 
+	/* Prepare for Tx DMA channel stop */
+	channel = pdata->channel;
+	for (i = 0; i < pdata->channel_count; i++, channel++) {
+		if (!channel->tx_ring)
+			break;
+
+		xgbe_prepare_tx_stop(pdata, channel);
+	}
+
 	/* Disable MAC Tx */
 	XGMAC_IOWRITE_BITS(pdata, MAC_TCR, TE, 0);
 
@@ -2485,6 +2726,8 @@ static int xgbe_init(struct xgbe_prv_data *pdata)
 	xgbe_config_tx_coalesce(pdata);
 	xgbe_config_rx_buffer_size(pdata);
 	xgbe_config_tso_mode(pdata);
+	xgbe_config_sph_mode(pdata);
+	xgbe_config_rss(pdata);
 	desc_if->wrapper_tx_desc_init(pdata);
 	desc_if->wrapper_rx_desc_init(pdata);
 	xgbe_enable_dma_interrupts(pdata);
@@ -2561,7 +2804,7 @@ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if)
 	hw_if->powerup_rx = xgbe_powerup_rx;
 	hw_if->powerdown_rx = xgbe_powerdown_rx;
 
-	hw_if->pre_xmit = xgbe_pre_xmit;
+	hw_if->dev_xmit = xgbe_dev_xmit;
 	hw_if->dev_read = xgbe_dev_read;
 	hw_if->enable_int = xgbe_enable_int;
 	hw_if->disable_int = xgbe_disable_int;
@@ -2575,6 +2818,7 @@ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if)
 	hw_if->rx_desc_reset = xgbe_rx_desc_reset;
 	hw_if->is_last_desc = xgbe_is_last_desc;
 	hw_if->is_context_desc = xgbe_is_context_desc;
+	hw_if->tx_start_xmit = xgbe_tx_start_xmit;
 
 	/* For FLOW ctrl */
 	hw_if->config_tx_flow_control = xgbe_config_tx_flow_control;
@@ -2620,5 +2864,11 @@ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if)
 	hw_if->config_dcb_tc = xgbe_config_dcb_tc;
 	hw_if->config_dcb_pfc = xgbe_config_dcb_pfc;
 
+	/* For Receive Side Scaling */
+	hw_if->enable_rss = xgbe_enable_rss;
+	hw_if->disable_rss = xgbe_disable_rss;
+	hw_if->set_rss_hash_key = xgbe_set_rss_hash_key;
+	hw_if->set_rss_lookup_table = xgbe_set_rss_lookup_table;
+
 	DBGPR("<--xgbe_init_function_ptrs\n");
 }
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 2349ea9..7bb5f07 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)
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
index 49508ec..ebf4893 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
@@ -452,9 +452,9 @@ static int xgbe_set_coalesce(struct net_device *netdev,
 			     rx_usecs);
 		return -EINVAL;
 	}
-	if (rx_frames > pdata->channel->rx_ring->rdesc_count) {
+	if (rx_frames > pdata->rx_desc_count) {
 		netdev_alert(netdev, "rx-frames is limited to %d frames\n",
-			     pdata->channel->rx_ring->rdesc_count);
+			     pdata->rx_desc_count);
 		return -EINVAL;
 	}
 
@@ -462,9 +462,9 @@ static int xgbe_set_coalesce(struct net_device *netdev,
 	tx_frames = ec->tx_max_coalesced_frames;
 
 	/* Check the bounds of values for Tx */
-	if (tx_frames > pdata->channel->tx_ring->rdesc_count) {
+	if (tx_frames > pdata->tx_desc_count) {
 		netdev_alert(netdev, "tx-frames is limited to %d frames\n",
-			     pdata->channel->tx_ring->rdesc_count);
+			     pdata->tx_desc_count);
 		return -EINVAL;
 	}
 
@@ -481,6 +481,82 @@ static int xgbe_set_coalesce(struct net_device *netdev,
 	return 0;
 }
 
+static int xgbe_get_rxnfc(struct net_device *netdev,
+			  struct ethtool_rxnfc *rxnfc, u32 *rule_locs)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	switch (rxnfc->cmd) {
+	case ETHTOOL_GRXRINGS:
+		rxnfc->data = pdata->rx_ring_count;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static u32 xgbe_get_rxfh_key_size(struct net_device *netdev)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	return sizeof(pdata->rss_key);
+}
+
+static u32 xgbe_get_rxfh_indir_size(struct net_device *netdev)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+
+	return ARRAY_SIZE(pdata->rss_table);
+}
+
+static int xgbe_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
+			 u8 *hfunc)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	unsigned int i;
+
+	if (indir) {
+		for (i = 0; i < ARRAY_SIZE(pdata->rss_table); i++)
+			indir[i] = XGMAC_GET_BITS(pdata->rss_table[i],
+						  MAC_RSSDR, DMCH);
+	}
+
+	if (key)
+		memcpy(key, pdata->rss_key, sizeof(pdata->rss_key));
+
+	if (hfunc)
+		*hfunc = ETH_RSS_HASH_TOP;
+
+	return 0;
+}
+
+static int xgbe_set_rxfh(struct net_device *netdev, const u32 *indir,
+			 const u8 *key, const u8 hfunc)
+{
+	struct xgbe_prv_data *pdata = netdev_priv(netdev);
+	struct xgbe_hw_if *hw_if = &pdata->hw_if;
+	unsigned int ret;
+
+	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)
+		return -EOPNOTSUPP;
+
+	if (indir) {
+		ret = hw_if->set_rss_lookup_table(pdata, indir);
+		if (ret)
+			return ret;
+	}
+
+	if (key) {
+		ret = hw_if->set_rss_hash_key(pdata, key);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static int xgbe_get_ts_info(struct net_device *netdev,
 			    struct ethtool_ts_info *ts_info)
 {
@@ -526,6 +602,11 @@ static const struct ethtool_ops xgbe_ethtool_ops = {
 	.get_strings = xgbe_get_strings,
 	.get_ethtool_stats = xgbe_get_ethtool_stats,
 	.get_sset_count = xgbe_get_sset_count,
+	.get_rxnfc = xgbe_get_rxnfc,
+	.get_rxfh_key_size = xgbe_get_rxfh_key_size,
+	.get_rxfh_indir_size = xgbe_get_rxfh_indir_size,
+	.get_rxfh = xgbe_get_rxfh,
+	.set_rxfh = xgbe_set_rxfh,
 	.get_ts_info = xgbe_get_ts_info,
 };
 
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-main.c b/drivers/net/ethernet/amd/xgbe/xgbe-main.c
index f5a8fa0..dbd3850 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-main.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-main.c
@@ -133,60 +133,6 @@ MODULE_LICENSE("Dual BSD/GPL");
 MODULE_VERSION(XGBE_DRV_VERSION);
 MODULE_DESCRIPTION(XGBE_DRV_DESC);
 
-static struct xgbe_channel *xgbe_alloc_rings(struct xgbe_prv_data *pdata)
-{
-	struct xgbe_channel *channel_mem, *channel;
-	struct xgbe_ring *tx_ring, *rx_ring;
-	unsigned int count, i;
-
-	DBGPR("-->xgbe_alloc_rings\n");
-
-	count = max_t(unsigned int, pdata->tx_ring_count, pdata->rx_ring_count);
-
-	channel_mem = devm_kcalloc(pdata->dev, count,
-				   sizeof(struct xgbe_channel), GFP_KERNEL);
-	if (!channel_mem)
-		return NULL;
-
-	tx_ring = devm_kcalloc(pdata->dev, pdata->tx_ring_count,
-			       sizeof(struct xgbe_ring), GFP_KERNEL);
-	if (!tx_ring)
-		return NULL;
-
-	rx_ring = devm_kcalloc(pdata->dev, pdata->rx_ring_count,
-			       sizeof(struct xgbe_ring), GFP_KERNEL);
-	if (!rx_ring)
-		return NULL;
-
-	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 (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_index=%u, dma_regs=%p, tx=%p, rx=%p\n",
-		      channel->name, channel->queue_index, channel->dma_regs,
-		      channel->tx_ring, channel->rx_ring);
-	}
-
-	pdata->channel_count = count;
-
-	DBGPR("<--xgbe_alloc_rings\n");
-
-	return channel_mem;
-}
-
 static void xgbe_default_config(struct xgbe_prv_data *pdata)
 {
 	DBGPR("-->xgbe_default_config\n");
@@ -224,6 +170,7 @@ static int xgbe_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct resource *res;
 	const u8 *mac_addr;
+	unsigned int i;
 	int ret;
 
 	DBGPR("--> xgbe_probe\n");
@@ -244,6 +191,7 @@ static int xgbe_probe(struct platform_device *pdev)
 
 	spin_lock_init(&pdata->lock);
 	mutex_init(&pdata->xpcs_mutex);
+	mutex_init(&pdata->rss_mutex);
 	spin_lock_init(&pdata->tstamp_lock);
 
 	/* Set and validate the number of descriptors for a ring */
@@ -318,12 +266,18 @@ static int xgbe_probe(struct platform_device *pdev)
 		pdata->awcache = XGBE_DMA_SYS_AWCACHE;
 	}
 
+	/* Check for per channel interrupt support */
+	if (of_property_read_bool(dev->of_node, XGBE_DMA_IRQS))
+		pdata->per_channel_irq = 1;
+
 	ret = platform_get_irq(pdev, 0);
 	if (ret < 0) {
-		dev_err(dev, "platform_get_irq failed\n");
+		dev_err(dev, "platform_get_irq 0 failed\n");
 		goto err_io;
 	}
-	netdev->irq = ret;
+	pdata->dev_irq = ret;
+
+	netdev->irq = pdata->dev_irq;
 	netdev->base_addr = (unsigned long)pdata->xgmac_regs;
 
 	/* Set all the function pointers */
@@ -383,13 +337,16 @@ static int xgbe_probe(struct platform_device *pdev)
 		goto err_io;
 	}
 
-	/* Allocate the rings for the DMA channels */
-	pdata->channel = xgbe_alloc_rings(pdata);
-	if (!pdata->channel) {
-		dev_err(dev, "ring allocation failed\n");
-		ret = -ENOMEM;
-		goto err_io;
-	}
+	/* Initialize RSS hash key and lookup table */
+	netdev_rss_key_fill(pdata->rss_key, sizeof(pdata->rss_key));
+
+	for (i = 0; i < XGBE_RSS_MAX_TABLE_SIZE; i++)
+		XGMAC_SET_BITS(pdata->rss_table[i], MAC_RSSDR, DMCH,
+			       i % pdata->rx_ring_count);
+
+	XGMAC_SET_BITS(pdata->rss_options, MAC_RSSCR, IP2TE, 1);
+	XGMAC_SET_BITS(pdata->rss_options, MAC_RSSCR, TCP4TE, 1);
+	XGMAC_SET_BITS(pdata->rss_options, MAC_RSSCR, UDP4TE, 1);
 
 	/* Prepare to regsiter with MDIO */
 	pdata->mii_bus_id = kasprintf(GFP_KERNEL, "%s", pdev->name);
@@ -421,6 +378,9 @@ static int xgbe_probe(struct platform_device *pdev)
 			      NETIF_F_HW_VLAN_CTAG_TX |
 			      NETIF_F_HW_VLAN_CTAG_FILTER;
 
+	if (pdata->hw_feat.rss)
+		netdev->hw_features |= NETIF_F_RXHASH;
+
 	netdev->vlan_features |= NETIF_F_SG |
 				 NETIF_F_IP_CSUM |
 				 NETIF_F_IPV6_CSUM |
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h
index 789957d..f9ec762 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe.h
+++ b/drivers/net/ethernet/amd/xgbe/xgbe.h
@@ -140,11 +140,25 @@
 
 #define XGBE_TX_MAX_BUF_SIZE	(0x3fff & ~(64 - 1))
 
+/* Descriptors required for maximum contigous TSO/GSO packet */
+#define XGBE_TX_MAX_SPLIT	((GSO_MAX_SIZE / XGBE_TX_MAX_BUF_SIZE) + 1)
+
+/* Maximum possible descriptors needed for an SKB:
+ * - Maximum number of SKB frags
+ * - Maximum descriptors for contiguous TSO/GSO packet
+ * - Possible context descriptor
+ * - Possible TSO header descriptor
+ */
+#define XGBE_TX_MAX_DESCS	(MAX_SKB_FRAGS + XGBE_TX_MAX_SPLIT + 2)
+
 #define XGBE_RX_MIN_BUF_SIZE	(ETH_FRAME_LEN + ETH_FCS_LEN + VLAN_HLEN)
 #define XGBE_RX_BUF_ALIGN	64
+#define XGBE_SKB_ALLOC_SIZE	256
+#define XGBE_SPH_HDSMS_SIZE	2	/* Keep in sync with SKB_ALLOC_SIZE */
 
 #define XGBE_MAX_DMA_CHANNELS	16
 #define XGBE_MAX_QUEUES		16
+#define XGBE_DMA_STOP_TIMEOUT	5
 
 /* DMA cache settings - Outer sharable, write-back, write-allocate */
 #define XGBE_DMA_OS_AXDOMAIN	0x2
@@ -171,6 +185,7 @@
 /* Device-tree clock names */
 #define XGBE_DMA_CLOCK		"dma_clk"
 #define XGBE_PTP_CLOCK		"ptp_clk"
+#define XGBE_DMA_IRQS		"amd,per-channel-interrupt"
 
 /* Timestamp support - values based on 50MHz PTP clock
  *   50MHz => 20 nsec
@@ -212,9 +227,17 @@
 /* Maximum MAC address hash table size (256 bits = 8 bytes) */
 #define XGBE_MAC_HASH_TABLE_SIZE	8
 
+/* Receive Side Scaling */
+#define XGBE_RSS_HASH_KEY_SIZE		40
+#define XGBE_RSS_MAX_TABLE_SIZE		256
+#define XGBE_RSS_LOOKUP_TABLE_TYPE	0
+#define XGBE_RSS_HASH_KEY_TYPE		1
+
 struct xgbe_prv_data;
 
 struct xgbe_packet_data {
+	struct sk_buff *skb;
+
 	unsigned int attributes;
 
 	unsigned int errors;
@@ -230,14 +253,53 @@ struct xgbe_packet_data {
 	unsigned short vlan_ctag;
 
 	u64 rx_tstamp;
+
+	u32 rss_hash;
+	enum pkt_hash_types rss_hash_type;
+
+	unsigned int tx_packets;
+	unsigned int tx_bytes;
 };
 
 /* Common Rx and Tx descriptor mapping */
 struct xgbe_ring_desc {
-	unsigned int desc0;
-	unsigned int desc1;
-	unsigned int desc2;
-	unsigned int desc3;
+	__le32 desc0;
+	__le32 desc1;
+	__le32 desc2;
+	__le32 desc3;
+};
+
+/* Page allocation related values */
+struct xgbe_page_alloc {
+	struct page *pages;
+	unsigned int pages_len;
+	unsigned int pages_offset;
+
+	dma_addr_t pages_dma;
+};
+
+/* Ring entry buffer data */
+struct xgbe_buffer_data {
+	struct xgbe_page_alloc pa;
+	struct xgbe_page_alloc pa_unmap;
+
+	dma_addr_t dma;
+	unsigned int dma_len;
+};
+
+/* Tx-related ring data */
+struct xgbe_tx_ring_data {
+	unsigned int packets;		/* BQL packet count */
+	unsigned int bytes;		/* BQL byte count */
+};
+
+/* Rx-related ring data */
+struct xgbe_rx_ring_data {
+	struct xgbe_buffer_data hdr;	/* Header locations */
+	struct xgbe_buffer_data buf;	/* Payload locations */
+
+	unsigned short hdr_len;		/* Length of received header */
+	unsigned short len;		/* Length of received packet */
 };
 
 /* Structure used to hold information related to the descriptor
@@ -251,9 +313,9 @@ struct xgbe_ring_data {
 	struct sk_buff *skb;		/* Virtual address of SKB */
 	dma_addr_t skb_dma;		/* DMA address of SKB data */
 	unsigned int skb_dma_len;	/* Length of SKB DMA area */
-	unsigned int tso_header;        /* TSO header indicator */
 
-	unsigned short len;		/* Length of received Rx packet */
+	struct xgbe_tx_ring_data tx;	/* Tx-related data */
+	struct xgbe_rx_ring_data rx;	/* Rx-related data */
 
 	unsigned int interrupt;		/* Interrupt indicator */
 
@@ -291,6 +353,10 @@ struct xgbe_ring {
 	 */
 	struct xgbe_ring_data *rdata;
 
+	/* Page allocation for RX buffers */
+	struct xgbe_page_alloc rx_hdr_pa;
+	struct xgbe_page_alloc rx_buf_pa;
+
 	/* Ring index values
 	 *  cur   - Tx: index of descriptor to be used for current transfer
 	 *          Rx: index of descriptor to check for packet availability
@@ -307,6 +373,7 @@ struct xgbe_ring {
 	union {
 		struct {
 			unsigned int queue_stopped;
+			unsigned int xmit_more;
 			unsigned short cur_mss;
 			unsigned short cur_vlan_ctag;
 		} tx;
@@ -331,6 +398,13 @@ struct xgbe_channel {
 	unsigned int queue_index;
 	void __iomem *dma_regs;
 
+	/* Per channel interrupt irq number */
+	int dma_irq;
+	char dma_irq_name[IFNAMSIZ + 32];
+
+	/* Netdev related settings */
+	struct napi_struct napi;
+
 	unsigned int saved_ier;
 
 	unsigned int tx_timer_active;
@@ -456,7 +530,7 @@ struct xgbe_hw_if {
 
 	int (*enable_int)(struct xgbe_channel *, enum xgbe_int);
 	int (*disable_int)(struct xgbe_channel *, enum xgbe_int);
-	void (*pre_xmit)(struct xgbe_channel *);
+	void (*dev_xmit)(struct xgbe_channel *);
 	int (*dev_read)(struct xgbe_channel *);
 	void (*tx_desc_init)(struct xgbe_channel *);
 	void (*rx_desc_init)(struct xgbe_channel *);
@@ -464,6 +538,7 @@ struct xgbe_hw_if {
 	void (*tx_desc_reset)(struct xgbe_ring_data *);
 	int (*is_last_desc)(struct xgbe_ring_desc *);
 	int (*is_context_desc)(struct xgbe_ring_desc *);
+	void (*tx_start_xmit)(struct xgbe_channel *, struct xgbe_ring *);
 
 	/* For FLOW ctrl */
 	int (*config_tx_flow_control)(struct xgbe_prv_data *);
@@ -509,14 +584,20 @@ struct xgbe_hw_if {
 	/* For Data Center Bridging config */
 	void (*config_dcb_tc)(struct xgbe_prv_data *);
 	void (*config_dcb_pfc)(struct xgbe_prv_data *);
+
+	/* For Receive Side Scaling */
+	int (*enable_rss)(struct xgbe_prv_data *);
+	int (*disable_rss)(struct xgbe_prv_data *);
+	int (*set_rss_hash_key)(struct xgbe_prv_data *, const u8 *);
+	int (*set_rss_lookup_table)(struct xgbe_prv_data *, const u32 *);
 };
 
 struct xgbe_desc_if {
 	int (*alloc_ring_resources)(struct xgbe_prv_data *);
 	void (*free_ring_resources)(struct xgbe_prv_data *);
 	int (*map_tx_skb)(struct xgbe_channel *, struct sk_buff *);
-	void (*realloc_skb)(struct xgbe_channel *);
-	void (*unmap_skb)(struct xgbe_prv_data *, struct xgbe_ring_data *);
+	void (*realloc_rx_buffer)(struct xgbe_channel *);
+	void (*unmap_rdata)(struct xgbe_prv_data *, struct xgbe_ring_data *);
 	void (*wrapper_tx_desc_init)(struct xgbe_prv_data *);
 	void (*wrapper_rx_desc_init)(struct xgbe_prv_data *);
 };
@@ -581,7 +662,11 @@ struct xgbe_prv_data {
 	/* XPCS indirect addressing mutex */
 	struct mutex xpcs_mutex;
 
-	int irq_number;
+	/* RSS addressing mutex */
+	struct mutex rss_mutex;
+
+	int dev_irq;
+	unsigned int per_channel_irq;
 
 	struct xgbe_hw_if hw_if;
 	struct xgbe_desc_if desc_if;
@@ -624,7 +709,7 @@ struct xgbe_prv_data {
 	unsigned int rx_riwt;
 	unsigned int rx_frames;
 
-	/* Current MTU */
+	/* Current Rx buffer size */
 	unsigned int rx_buf_size;
 
 	/* Flow control settings */
@@ -632,6 +717,11 @@ struct xgbe_prv_data {
 	unsigned int tx_pause;
 	unsigned int rx_pause;
 
+	/* Receive Side Scaling settings */
+	u8 rss_key[XGBE_RSS_HASH_KEY_SIZE];
+	u32 rss_table[XGBE_RSS_MAX_TABLE_SIZE];
+	u32 rss_options;
+
 	/* MDIO settings */
 	struct module *phy_module;
 	char *mii_bus_id;