[dpdk-dev] [PATCH 1/2] szedata2: new poll mode driver

Matej Vido vido at cesnet.cz
Fri Jun 19 10:25:00 CEST 2015


Added virtual PMD which communicates with COMBO cards through sze2
layer using libsze2 library.

Signed-off-by: Matej Vido <vido at cesnet.cz>
---
 config/common_bsdapp                              |    5 +
 config/common_linuxapp                            |    5 +
 drivers/net/Makefile                              |    1 +
 drivers/net/szedata2/Makefile                     |   62 ++
 drivers/net/szedata2/rte_eth_szedata2.c           | 1120 +++++++++++++++++++++
 drivers/net/szedata2/rte_eth_szedata2.h           |   96 ++
 drivers/net/szedata2/rte_pmd_szedata2_version.map |    4 +
 mk/rte.app.mk                                     |    3 +
 8 files changed, 1296 insertions(+)
 create mode 100644 drivers/net/szedata2/Makefile
 create mode 100644 drivers/net/szedata2/rte_eth_szedata2.c
 create mode 100644 drivers/net/szedata2/rte_eth_szedata2.h
 create mode 100644 drivers/net/szedata2/rte_pmd_szedata2_version.map

diff --git a/config/common_bsdapp b/config/common_bsdapp
index 0b169c8..d5f5970 100644
--- a/config/common_bsdapp
+++ b/config/common_bsdapp
@@ -249,6 +249,11 @@ CONFIG_RTE_PMD_RING_MAX_TX_RINGS=16
 CONFIG_RTE_LIBRTE_PMD_PCAP=y
 
 #
+# Compile software PMD backed by SZEDATA2 device
+#
+CONFIG_RTE_LIBRTE_PMD_SZEDATA2=n
+
+#
 # Compile link bonding PMD library
 #
 CONFIG_RTE_LIBRTE_PMD_BOND=y
diff --git a/config/common_linuxapp b/config/common_linuxapp
index 5deb55a..3bec728 100644
--- a/config/common_linuxapp
+++ b/config/common_linuxapp
@@ -246,6 +246,11 @@ CONFIG_RTE_PMD_RING_MAX_TX_RINGS=16
 CONFIG_RTE_LIBRTE_PMD_PCAP=n
 
 #
+# Compile software PMD backed by SZEDATA2 device
+#
+CONFIG_RTE_LIBRTE_PMD_SZEDATA2=n
+
+#
 # Compile link bonding PMD library
 #
 CONFIG_RTE_LIBRTE_PMD_BOND=y
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 1e6648a..3312b13 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -45,6 +45,7 @@ DIRS-$(CONFIG_RTE_LIBRTE_PMD_RING) += ring
 DIRS-$(CONFIG_RTE_LIBRTE_VIRTIO_PMD) += virtio
 DIRS-$(CONFIG_RTE_LIBRTE_VMXNET3_PMD) += vmxnet3
 DIRS-$(CONFIG_RTE_LIBRTE_PMD_XENVIRT) += xenvirt
+DIRS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2) += szedata2
 
 include $(RTE_SDK)/mk/rte.sharelib.mk
 include $(RTE_SDK)/mk/rte.subdir.mk
diff --git a/drivers/net/szedata2/Makefile b/drivers/net/szedata2/Makefile
new file mode 100644
index 0000000..c3c42e5
--- /dev/null
+++ b/drivers/net/szedata2/Makefile
@@ -0,0 +1,62 @@
+#   BSD LICENSE
+#
+#   Copyright (c) 2015 CESNET
+#   All rights reserved.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+#     * Redistributions of source code must retain the above copyright
+#       notice, this list of conditions and the following disclaimer.
+#     * Redistributions in binary form must reproduce the above copyright
+#       notice, this list of conditions and the following disclaimer in
+#       the documentation and/or other materials provided with the
+#       distribution.
+#     * Neither the name of CESNET nor the names of its
+#       contributors may be used to endorse or promote products derived
+#       from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+#
+# library name
+#
+LIB = librte_pmd_szedata2.a
+
+CFLAGS += -O3
+CFLAGS += $(WERROR_FLAGS)
+
+EXPORT_MAP := rte_pmd_szedata2_version.map
+
+LIBABIVER := 1
+
+#
+# all source are stored in SRCS-y
+#
+SRCS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2) += rte_eth_szedata2.c
+
+#
+# Export include files
+#
+SYMLINK-y-include +=
+
+# this lib depends upon:
+DEPDIRS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2) += lib/librte_mbuf
+DEPDIRS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2) += lib/librte_ether
+DEPDIRS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2) += lib/librte_malloc
+DEPDIRS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2) += lib/librte_kvargs
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/szedata2/rte_eth_szedata2.c b/drivers/net/szedata2/rte_eth_szedata2.c
new file mode 100644
index 0000000..cb3b2a2
--- /dev/null
+++ b/drivers/net/szedata2/rte_eth_szedata2.c
@@ -0,0 +1,1120 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright (c) 2015 CESNET
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of CESNET nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <stdint.h>
+#include <unistd.h>
+#include <stdbool.h>
+#include <err.h>
+
+#include <libsze2.h>
+
+#include <rte_mbuf.h>
+#include <rte_ethdev.h>
+#include <rte_malloc.h>
+#include <rte_memcpy.h>
+#include <rte_kvargs.h>
+#include <rte_dev.h>
+
+#include "rte_eth_szedata2.h"
+
+#define RTE_ETH_SZEDATA2_DEV_PATH_ARG "dev_path"
+#define RTE_ETH_SZEDATA2_RX_IFACES_ARG "rx_ifaces"
+#define RTE_ETH_SZEDATA2_TX_IFACES_ARG "tx_ifaces"
+
+#define RTE_ETH_SZEDATA2_MAX_RX_QUEUES 32
+#define RTE_ETH_SZEDATA2_MAX_TX_QUEUES 32
+#define RTE_ETH_SZEDATA2_TX_LOCK_SIZE 33554432 /**< 32*1024*1024 */
+
+/**
+ * size of szedata2_packet header with alignment
+ */
+#define RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED 8
+
+struct szedata2_rx_queue {
+	struct szedata * sze;
+	uint8_t rx_channel;
+	uint8_t in_port;
+	struct rte_mempool *mb_pool;
+	volatile uint64_t rx_pkts;
+	volatile uint64_t rx_bytes;
+	volatile uint64_t err_pkts;
+};
+
+struct szedata2_tx_queue {
+	struct szedata * sze;
+	uint8_t tx_channel;
+	volatile uint64_t tx_pkts;
+	volatile uint64_t err_pkts;
+	volatile uint64_t tx_bytes;
+};
+
+struct rxtx_szedata2 {
+	uint32_t num_of_rx;
+	uint32_t num_of_tx;
+	uint32_t sze_rx_mask_req;
+	uint32_t sze_tx_mask_req;
+	char * sze_dev;
+};
+
+struct pmd_internals {
+	struct szedata2_rx_queue rx_queue[RTE_ETH_SZEDATA2_MAX_RX_QUEUES];
+	struct szedata2_tx_queue tx_queue[RTE_ETH_SZEDATA2_MAX_TX_QUEUES];
+	unsigned nb_rx_queues;
+	unsigned nb_tx_queues;
+	uint32_t num_of_rx;
+	uint32_t num_of_tx;
+	uint32_t sze_rx_req;
+	uint32_t sze_tx_req;
+	int if_index;
+	char * sze_dev;
+};
+
+static const char *valid_arguments[] = {
+	RTE_ETH_SZEDATA2_DEV_PATH_ARG,
+	RTE_ETH_SZEDATA2_RX_IFACES_ARG,
+	RTE_ETH_SZEDATA2_TX_IFACES_ARG,
+	NULL
+};
+
+static struct ether_addr eth_addr = {
+	.addr_bytes = { 0, 0, 0, 0x1, 0x2, 0x3 }
+};
+static const char *drivername = "SZEdata2 PMD";
+static struct rte_eth_link pmd_link = {
+		.link_speed = ETH_LINK_SPEED_10G,
+		.link_duplex = ETH_LINK_FULL_DUPLEX,
+		.link_status = 0
+};
+
+
+static uint32_t
+count_set_bits(uint32_t num)
+{
+	num = num - ((num >> 1) & 0x55555555); /* reuse input as temporary */
+	num = (num & 0x33333333) + ((num >> 2) & 0x33333333);        /* temp */
+	return (((num + (num >> 4)) & 0xF0F0F0F) * 0x1010101) >> 24; /* count */
+}
+
+static uint16_t
+eth_szedata2_rx(void *queue,
+		struct rte_mbuf **bufs,
+		uint16_t nb_pkts)
+{
+	unsigned int i;
+	struct rte_mbuf *mbuf;
+	struct szedata2_rx_queue *sze_q = queue;
+	struct rte_pktmbuf_pool_private *mbp_priv;
+	uint16_t num_rx = 0;
+	uint16_t buf_size;
+	uint16_t sg_size;
+	uint16_t hw_size;
+	uint16_t packet_size;
+	uint64_t num_bytes = 0;
+	struct szedata * sze = sze_q->sze;
+	uint8_t * header_ptr = NULL; /* header of packet */
+	uint8_t * packet_ptr1 = NULL;
+	uint8_t * packet_ptr2 = NULL;
+	uint16_t packet_len1 = 0;
+	uint16_t packet_len2 = 0;
+	uint16_t hw_data_align;
+
+	if (unlikely(sze_q->sze == NULL || nb_pkts == 0)) {
+		return 0;
+	}
+
+	/*
+	 * Reads the given number of packets from szedata2 channel given by queue
+	 * and copies the packet data into a newly allocated mbuf to return.
+	 */
+	for (i = 0; i < nb_pkts; i++) {
+		/* get the next sze packet */
+		if (sze->ct_rx_lck != NULL && !sze->ct_rx_rem_bytes &&
+				sze->ct_rx_lck->next == NULL) {
+			/* unlock old data */
+			szedata_rx_unlock_data(sze_q->sze, sze->ct_rx_lck_orig);
+			sze->ct_rx_lck_orig = NULL;
+			sze->ct_rx_lck = NULL;
+		}
+
+		if (!sze->ct_rx_rem_bytes && sze->ct_rx_lck_orig == NULL) {
+			/* nothing to read, lock new data */
+			sze->ct_rx_lck_orig = sze->ct_rx_lck =
+				szedata_rx_lock_data(sze_q->sze, ~0U);
+
+			if (sze->ct_rx_lck == NULL) {
+				/* nothing to lock */
+				break;
+			}
+
+			sze->ct_rx_cur_ptr = sze->ct_rx_lck->start;
+			sze->ct_rx_rem_bytes = sze->ct_rx_lck->len;
+
+			if (!sze->ct_rx_rem_bytes) {
+				break;
+			}
+		}
+
+		if (sze->ct_rx_rem_bytes < RTE_SZE2_PACKET_HEADER_SIZE) {
+			/* cut in header - copy parts of header to merge buffer */
+			if (sze->ct_rx_lck->next == NULL) {
+				break;
+			}
+
+			/* copy first part of header */
+			rte_memcpy(sze->ct_rx_buffer, sze->ct_rx_cur_ptr,
+					sze->ct_rx_rem_bytes);
+
+			/* copy second part of header */
+			sze->ct_rx_lck = sze->ct_rx_lck->next;
+			sze->ct_rx_cur_ptr = sze->ct_rx_lck->start;
+			rte_memcpy(sze->ct_rx_buffer + sze->ct_rx_rem_bytes,
+					sze->ct_rx_cur_ptr,
+					RTE_SZE2_PACKET_HEADER_SIZE - sze->ct_rx_rem_bytes);
+
+			sze->ct_rx_cur_ptr += RTE_SZE2_PACKET_HEADER_SIZE -
+				sze->ct_rx_rem_bytes;
+			sze->ct_rx_rem_bytes = sze->ct_rx_lck->len -
+				RTE_SZE2_PACKET_HEADER_SIZE + sze->ct_rx_rem_bytes;
+
+			header_ptr = (uint8_t *) sze->ct_rx_buffer;
+		} else {
+			/* not cut */
+			header_ptr = (uint8_t *) sze->ct_rx_cur_ptr;
+			sze->ct_rx_cur_ptr += RTE_SZE2_PACKET_HEADER_SIZE;
+			sze->ct_rx_rem_bytes -= RTE_SZE2_PACKET_HEADER_SIZE;
+		}
+
+		sg_size = le16toh(*((uint16_t *)header_ptr));
+		hw_size = le16toh(*(((uint16_t *)header_ptr)+1));
+		packet_size = sg_size -
+			RTE_SZE2_ALIGN8(RTE_SZE2_PACKET_HEADER_SIZE + hw_size);
+
+
+		/* checks if packet all right */
+		if (!sg_size) {
+			errx(5, "Zero segsize");
+		}
+
+		/* check sg_size and hwsize */
+		if (hw_size > sg_size - RTE_SZE2_PACKET_HEADER_SIZE) {
+			errx(10, "Hwsize bigger than expected. Segsize: %d, hwsize: %d",
+					sg_size, hw_size);
+		}
+
+		hw_data_align =
+			RTE_SZE2_ALIGN8((RTE_SZE2_PACKET_HEADER_SIZE + hw_size)) -
+			RTE_SZE2_PACKET_HEADER_SIZE;
+
+		if (sze->ct_rx_rem_bytes >=
+				(uint16_t)(sg_size - RTE_SZE2_PACKET_HEADER_SIZE)) {
+			/* no cut */
+			/* one packet ready - go to another */
+			packet_ptr1 = sze->ct_rx_cur_ptr + hw_data_align;
+			packet_len1 = packet_size;
+			packet_ptr2 = NULL;
+			packet_len2 = 0;
+
+			sze->ct_rx_cur_ptr += RTE_SZE2_ALIGN8(sg_size) -
+				RTE_SZE2_PACKET_HEADER_SIZE;
+			sze->ct_rx_rem_bytes -= RTE_SZE2_ALIGN8(sg_size) -
+				RTE_SZE2_PACKET_HEADER_SIZE;
+		} else {
+			/* cut in data */
+			if (sze->ct_rx_lck->next == NULL) {
+				errx(6, "Need \"next\" lock, but it is missing: %u",
+						sze->ct_rx_rem_bytes);
+			}
+
+			/* skip hw data */
+			if (sze->ct_rx_rem_bytes <= hw_data_align) {
+				uint16_t rem_size = hw_data_align - sze->ct_rx_rem_bytes;
+
+				/* MOVE to next lock */
+				sze->ct_rx_lck = sze->ct_rx_lck->next;
+				sze->ct_rx_cur_ptr =
+					(void *) (((uint8_t *)(sze->ct_rx_lck->start)) + rem_size);
+
+				packet_ptr1 = sze->ct_rx_cur_ptr;
+				packet_len1 = packet_size;
+				packet_ptr2 = NULL;
+				packet_len2 = 0;
+
+				sze->ct_rx_cur_ptr += RTE_SZE2_ALIGN8(packet_size);
+				sze->ct_rx_rem_bytes = sze->ct_rx_lck->len -
+					rem_size - RTE_SZE2_ALIGN8(packet_size);
+			} else {
+				/* get pointer and length from first part of data */
+				packet_ptr1 = sze->ct_rx_cur_ptr + hw_data_align;
+				packet_len1 = sze->ct_rx_rem_bytes - hw_data_align;
+
+				/* MOVE to next lock */
+				sze->ct_rx_lck = sze->ct_rx_lck->next;
+				sze->ct_rx_cur_ptr = sze->ct_rx_lck->start;
+
+				/* get pointer and length from second part of data */
+				packet_ptr2 = sze->ct_rx_cur_ptr;
+				packet_len2 = packet_size - packet_len1;
+
+				sze->ct_rx_cur_ptr += RTE_SZE2_ALIGN8(packet_size) -
+					packet_len1;
+				sze->ct_rx_rem_bytes = sze->ct_rx_lck->len -
+					(RTE_SZE2_ALIGN8(packet_size) - packet_len1);
+			}
+		}
+
+		if (unlikely(packet_ptr1 == NULL)) {
+			break;
+		} else {
+			mbuf = rte_pktmbuf_alloc(sze_q->mb_pool);
+		}
+
+		if (unlikely(mbuf == NULL)) {
+			break;
+		}
+
+		/* get the space available for data in the mbuf */
+		mbp_priv = rte_mempool_get_priv(sze_q->mb_pool);
+		buf_size = (uint16_t) (mbp_priv->mbuf_data_room_size -
+				RTE_PKTMBUF_HEADROOM);
+
+		if (packet_size <= buf_size) {
+			/* sze packet will fit in the mbuf, go ahead and copy */
+			rte_memcpy(rte_pktmbuf_mtod(mbuf, void *),
+					packet_ptr1, packet_len1);
+			if (packet_ptr2 != NULL) {
+				rte_memcpy((void *)(rte_pktmbuf_mtod(mbuf, uint8_t *) +
+							packet_len1), packet_ptr2, packet_len2);
+			}
+			mbuf->data_len = (uint16_t)packet_size;
+			mbuf->pkt_len = mbuf->data_len;
+			mbuf->port = sze_q->in_port;
+			bufs[num_rx] = mbuf;
+			num_rx++;
+			num_bytes += packet_size;
+		} else {
+			/* sze packet will not fit in the mbuf, drop it */
+			RTE_LOG(ERR, PMD,
+					"SZE segment %d bytes will not fit in mbuf (%d bytes)\n",
+					packet_size, buf_size);
+			rte_pktmbuf_free(mbuf);
+		}
+	}
+
+	sze_q->rx_pkts += num_rx;
+	sze_q->rx_bytes += num_bytes;
+	return num_rx;
+}
+
+static uint16_t
+eth_szedata2_tx(void *queue,
+		struct rte_mbuf **bufs,
+		uint16_t nb_pkts)
+{
+	unsigned i;
+	struct rte_mbuf *mbuf;
+	struct szedata2_tx_queue *sze_q = queue;
+	uint16_t num_tx = 0;
+	uint64_t num_bytes = 0;
+
+	const struct szedata_lock *lck;
+	uint32_t lock_size;
+	uint32_t lock_size2;
+	void *dst;
+	uint32_t pkt_len;
+	uint32_t hwpkt_len;
+	uint32_t unlock_size;
+	uint32_t rem_len;
+	uint32_t write_len;
+
+	if (sze_q->sze == NULL || nb_pkts == 0)
+		return 0;
+
+
+	for (i = 0; i < nb_pkts; i++) {
+lock_again:
+		unlock_size = 0;
+		lck = szedata_tx_lock_data(sze_q->sze, RTE_ETH_SZEDATA2_TX_LOCK_SIZE,
+				sze_q->tx_channel);
+		if (lck == NULL) {
+			goto lock_again;
+		}
+
+		dst = lck->start;
+		lock_size = lck->len;
+		lock_size2 = lck->next ? lck->next->len : 0;
+
+next_packet:
+		mbuf = bufs[i];
+
+		pkt_len = mbuf->data_len;
+
+		hwpkt_len = RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED +
+			RTE_SZE2_ALIGN8(pkt_len);
+
+		if (lock_size + lock_size2 < hwpkt_len) {
+			szedata_tx_unlock_data(sze_q->sze, lck, unlock_size);
+			goto lock_again;
+		}
+
+		num_bytes += pkt_len;
+
+		if (lock_size > hwpkt_len) {
+			rem_len = 0;
+			/* write packet length at first 2 bytes in 8B header */
+			*((uint16_t *) dst) = htole16(RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED +
+					pkt_len);
+			*(((uint16_t *) dst) + 1) = htole16(0);
+			/* copy packet from mbuf */
+			rte_memcpy((void *)(((uint8_t *)(dst)) +
+						RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED),
+					rte_pktmbuf_mtod(mbuf, const void *), pkt_len);
+
+
+			dst = ((uint8_t *)dst) + hwpkt_len;
+			unlock_size += hwpkt_len;
+			lock_size -= hwpkt_len;
+
+			rte_pktmbuf_free(mbuf);
+			num_tx++;
+			i++;
+			if (i == nb_pkts) {
+				szedata_tx_unlock_data(sze_q->sze, lck, unlock_size);
+				break;
+			}
+			goto next_packet;
+		} else if (lock_size + lock_size2 >= hwpkt_len) {
+			/* write packet length at first 2 bytes in 8B header */
+			*((uint16_t *) dst) = htole16(RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED +
+					pkt_len);
+			*(((uint16_t *) dst) + 1) = htole16(0);
+
+			/*
+			 * If the raw packet (pkt_len) is smaller than lock_size,
+			 * get the correct length for memcpy
+			 */
+			write_len =
+				pkt_len < lock_size - RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED ?
+				pkt_len : lock_size - RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED;
+
+			/* copy first part of packet */
+			rte_memcpy((void *)(((uint8_t *)(dst)) +
+						RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED),
+					rte_pktmbuf_mtod(mbuf, const void *), write_len);
+
+			rem_len = hwpkt_len - lock_size;
+			if (lck->next)
+				dst = lck->next->start;
+
+			rte_memcpy(dst,
+					(const void *) (rte_pktmbuf_mtod(mbuf, const uint8_t *) +
+						write_len), pkt_len - write_len);
+
+			dst = ((uint8_t *)dst) + rem_len;
+			unlock_size += hwpkt_len;
+			lock_size = lock_size2 - rem_len;
+			lock_size2 = 0;
+
+			rte_pktmbuf_free(mbuf);
+			num_tx++;
+		}
+
+		szedata_tx_unlock_data(sze_q->sze, lck, unlock_size);
+	}
+
+	sze_q->tx_pkts += num_tx;
+	sze_q->err_pkts += nb_pkts - num_tx;
+	sze_q->tx_bytes += num_bytes;
+	return num_tx;
+}
+
+static int
+eth_dev_start(struct rte_eth_dev *dev)
+{
+	struct pmd_internals *internals = dev->data->dev_private;
+	int ret;
+	unsigned i;
+	uint32_t x;
+	uint32_t rx;
+	uint32_t tx;
+	uint32_t num_rx_sub = 0;
+	uint32_t num_tx_sub = 0;
+
+	if (internals->nb_rx_queues == 0) {
+		rx = internals->sze_rx_req;
+		tx = 0;
+
+		for (i = 0; i < internals->num_of_rx; i++) {
+			/*
+			 * Open, subscribe rx,tx channels and start device
+			 */
+			internals->rx_queue[num_rx_sub].sze =
+				szedata_open(internals->sze_dev);
+			if (internals->rx_queue[num_rx_sub].sze == NULL)
+				goto err_rx;
+
+			x = rx & ((~rx)+1); /* separate least significant non-zero bit */
+
+			ret = szedata_subscribe3(internals->rx_queue[num_rx_sub].sze,
+					&x, &tx);
+			if (ret)
+				goto err_rx;
+
+			if (x) {
+				ret = szedata_start(internals->rx_queue[num_rx_sub].sze);
+				if (ret)
+					goto err_rx;
+
+				/*
+				 * set to 1 all bits lower than bit set to 1
+				 * and that bit to 0
+				 */
+				x -= 1;
+				internals->rx_queue[num_rx_sub].rx_channel = count_set_bits(x);
+				num_rx_sub++;
+			} else {
+				szedata_close(internals->rx_queue[num_rx_sub].sze);
+				internals->rx_queue[num_rx_sub].sze = NULL;
+			}
+
+			rx = rx & (rx-1); /* set least significant non-zero bit to zero */
+		}
+
+		internals->nb_rx_queues = num_rx_sub;
+		dev->data->nb_rx_queues = (uint16_t) num_rx_sub;
+	}
+
+	if (internals->nb_tx_queues == 0) {
+		rx = 0;
+		tx = internals->sze_tx_req;
+
+		for (i = 0; i < internals->num_of_tx; i++) {
+			/*
+			 * Open, subscribe rx,tx channels and start device
+			 */
+			internals->tx_queue[num_tx_sub].sze =
+				szedata_open(internals->sze_dev);
+			if (internals->tx_queue[num_tx_sub].sze == NULL)
+				goto err_tx;
+
+			x = tx & ((~tx)+1); /* separate least significant non-zero bit */
+
+			ret = szedata_subscribe3(internals->tx_queue[num_tx_sub].sze,
+					&rx, &x);
+			if (ret)
+				goto err_tx;
+
+			if (x) {
+				ret = szedata_start(internals->tx_queue[num_tx_sub].sze);
+				if (ret)
+					goto err_tx;
+
+				/*
+				 * set to 1 all bits lower than bit set to 1
+				 * and that bit to 0
+				 */
+				x -= 1;
+				internals->tx_queue[num_tx_sub].tx_channel = count_set_bits(x);
+				num_tx_sub++;
+			} else {
+				szedata_close(internals->tx_queue[num_tx_sub].sze);
+				internals->tx_queue[num_tx_sub].sze = NULL;
+			}
+
+			tx = tx & (tx-1); /* set least significant non-zero bit to zero */
+		}
+
+		internals->nb_tx_queues = num_tx_sub;
+		dev->data->nb_tx_queues = (uint16_t) num_tx_sub;
+	}
+
+	dev->data->dev_link.link_status = 1;
+	return 0;
+err_tx: /* close sze for tx and rx queues */
+	for (i = 0; i < num_tx_sub; i++) {
+		if (internals->tx_queue[num_tx_sub].sze != NULL) {
+			szedata_close(internals->tx_queue[num_tx_sub].sze);
+			internals->tx_queue[num_tx_sub].sze = NULL;
+		}
+	}
+	/* set number of rx queues to zero */
+	internals->nb_rx_queues = 0;
+	dev->data->nb_rx_queues = (uint16_t) 0;
+err_rx: /* close sze only for rx queues */
+	for (i = 0; i < num_rx_sub; i++) {
+		if (internals->rx_queue[num_rx_sub].sze != NULL) {
+			szedata_close(internals->rx_queue[num_rx_sub].sze);
+			internals->tx_queue[num_rx_sub].sze = NULL;
+		}
+	}
+	return -1;
+}
+
+static void
+eth_dev_stop(struct rte_eth_dev *dev)
+{
+	unsigned i;
+	struct pmd_internals *internals = dev->data->dev_private;
+
+	for (i = 0; i < internals->nb_rx_queues; i++) {
+		if (internals->rx_queue[i].sze != NULL) {
+			szedata_close(internals->rx_queue[i].sze);
+			internals->rx_queue[i].sze = NULL;
+		}
+	}
+
+	for (i = 0; i < internals->nb_tx_queues; i++) {
+		if (internals->tx_queue[i].sze != NULL) {
+			szedata_close(internals->tx_queue[i].sze);
+			internals->tx_queue[i].sze = NULL;
+		}
+	}
+
+	internals->nb_rx_queues = 0;
+	internals->nb_tx_queues = 0;
+
+	dev->data->nb_rx_queues = (uint16_t) 0;
+	dev->data->nb_tx_queues = (uint16_t) 0;
+
+	dev->data->dev_link.link_status = 0;
+}
+
+static int
+eth_dev_configure(struct rte_eth_dev *dev __rte_unused)
+{
+	return 0;
+}
+
+static void
+eth_dev_info(struct rte_eth_dev *dev,
+		struct rte_eth_dev_info *dev_info)
+{
+	struct pmd_internals *internals = dev->data->dev_private;
+	dev_info->driver_name = drivername;
+	dev_info->if_index = internals->if_index;
+	dev_info->max_mac_addrs = 1;
+	dev_info->max_rx_pktlen = (uint32_t) -1;
+	dev_info->max_rx_queues = (uint16_t)internals->nb_rx_queues;
+	dev_info->max_tx_queues = (uint16_t)internals->nb_tx_queues;
+	dev_info->min_rx_bufsize = 0;
+	dev_info->pci_dev = NULL;
+}
+
+static void
+eth_stats_get(struct rte_eth_dev *dev,
+		struct rte_eth_stats *igb_stats)
+{
+	unsigned i;
+	uint64_t rx_total = 0;
+	uint64_t tx_total = 0;
+	uint64_t tx_err_total = 0;
+	uint64_t rx_total_bytes = 0;
+	uint64_t tx_total_bytes = 0;
+	const struct pmd_internals *internal = dev->data->dev_private;
+
+	for (i = 0; i < RTE_ETHDEV_QUEUE_STAT_CNTRS && i < internal->nb_rx_queues;
+			i++) {
+		igb_stats->q_ipackets[i] = internal->rx_queue[i].rx_pkts;
+		igb_stats->q_ibytes[i] = internal->rx_queue[i].rx_bytes;
+		rx_total += igb_stats->q_ipackets[i];
+		rx_total_bytes += igb_stats->q_ibytes[i];
+	}
+
+	for (i = 0; i < RTE_ETHDEV_QUEUE_STAT_CNTRS && i < internal->nb_tx_queues;
+			i++) {
+		igb_stats->q_opackets[i] = internal->tx_queue[i].tx_pkts;
+		igb_stats->q_errors[i] = internal->tx_queue[i].err_pkts;
+		igb_stats->q_obytes[i] = internal->tx_queue[i].tx_bytes;
+		tx_total += igb_stats->q_opackets[i];
+		tx_err_total += igb_stats->q_errors[i];
+		tx_total_bytes += igb_stats->q_obytes[i];
+	}
+
+	igb_stats->ipackets = rx_total;
+	igb_stats->opackets = tx_total;
+	igb_stats->ibytes = rx_total_bytes;
+	igb_stats->obytes = tx_total_bytes;
+	igb_stats->oerrors = tx_err_total;
+}
+
+static void
+eth_stats_reset(struct rte_eth_dev *dev)
+{
+	unsigned i;
+	struct pmd_internals *internal = dev->data->dev_private;
+	for (i = 0; i < internal->nb_rx_queues; i++) {
+		internal->rx_queue[i].rx_pkts = 0;
+		internal->rx_queue[i].rx_bytes = 0;
+	}
+	for (i = 0; i < internal->nb_tx_queues; i++) {
+		internal->tx_queue[i].tx_pkts = 0;
+		internal->tx_queue[i].err_pkts = 0;
+		internal->tx_queue[i].tx_bytes = 0;
+	}
+}
+
+static void
+eth_dev_close(struct rte_eth_dev *dev)
+{
+	unsigned i;
+	struct pmd_internals *internals = dev->data->dev_private;
+
+	for (i = 0; i < internals->nb_rx_queues; i++) {
+		if (internals->rx_queue[i].sze != NULL) {
+			szedata_close(internals->rx_queue[i].sze);
+			internals->rx_queue[i].sze = NULL;
+		}
+	}
+
+	for (i = 0; i < internals->nb_tx_queues; i++) {
+		if (internals->tx_queue[i].sze != NULL) {
+			szedata_close(internals->tx_queue[i].sze);
+			internals->tx_queue[i].sze = NULL;
+		}
+	}
+
+	internals->nb_rx_queues = 0;
+	internals->nb_tx_queues = 0;
+
+	dev->data->nb_rx_queues = (uint16_t) 0;
+	dev->data->nb_tx_queues = (uint16_t) 0;
+}
+
+static void
+eth_queue_release(void *q __rte_unused)
+{
+}
+
+static int
+eth_link_update(struct rte_eth_dev *dev __rte_unused,
+		int wait_to_complete __rte_unused)
+{
+	return 0;
+}
+
+static int
+eth_rx_queue_setup(struct rte_eth_dev *dev,
+		uint16_t rx_queue_id,
+		uint16_t nb_rx_desc __rte_unused,
+		unsigned int socket_id __rte_unused,
+		const struct rte_eth_rxconf *rx_conf __rte_unused,
+		struct rte_mempool *mb_pool)
+{
+	struct pmd_internals *internals = dev->data->dev_private;
+	struct szedata2_rx_queue *szedata2_q = &internals->rx_queue[rx_queue_id];
+	szedata2_q->mb_pool = mb_pool;
+	dev->data->rx_queues[rx_queue_id] = szedata2_q;
+	szedata2_q->in_port = dev->data->port_id;
+	return 0;
+}
+
+static int
+eth_tx_queue_setup(struct rte_eth_dev *dev,
+		uint16_t tx_queue_id,
+		uint16_t nb_tx_desc __rte_unused,
+		unsigned int socket_id __rte_unused,
+		const struct rte_eth_txconf *tx_conf __rte_unused)
+{
+
+	struct pmd_internals *internals = dev->data->dev_private;
+	dev->data->tx_queues[tx_queue_id] = &internals->tx_queue[tx_queue_id];
+	return 0;
+}
+
+static struct eth_dev_ops ops = {
+		.dev_start          = eth_dev_start,
+		.dev_stop           = eth_dev_stop,
+		.dev_close          = eth_dev_close,
+		.dev_configure      = eth_dev_configure,
+		.dev_infos_get      = eth_dev_info,
+		.rx_queue_setup     = eth_rx_queue_setup,
+		.tx_queue_setup     = eth_tx_queue_setup,
+		.rx_queue_release   = eth_queue_release,
+		.tx_queue_release   = eth_queue_release,
+		.link_update        = eth_link_update,
+		.stats_get          = eth_stats_get,
+		.stats_reset        = eth_stats_reset,
+};
+
+static int
+get_mask(const char *mask_str, uint32_t *mask_num)
+{
+	char *endptr;
+	long int value;
+
+	value = strtol(mask_str, &endptr, 0);
+	if (*endptr != '\0' || value > UINT32_MAX || value < 0)
+		return -1;
+
+	*mask_num = (uint32_t) value;
+	return 0;
+}
+
+static int
+add_rx_mask(const char *key __rte_unused, const char *value, void *extra_args)
+{
+	struct rxtx_szedata2 *szedata2 = extra_args;
+	uint32_t mask;
+
+	if (get_mask(value, &mask) != 0)
+		return -1;
+
+	szedata2->sze_rx_mask_req |= mask;
+	return 0;
+}
+
+static int
+add_tx_mask(const char *key __rte_unused, const char *value, void *extra_args)
+{
+	struct rxtx_szedata2 *szedata2 = extra_args;
+	uint32_t mask;
+
+	if (get_mask(value, &mask) != 0)
+		return -1;
+
+	szedata2->sze_tx_mask_req |= mask;
+	return 0;
+}
+
+static int
+rte_pmd_init_internals(const char *name, const unsigned nb_rx_queues,
+		const unsigned nb_tx_queues,
+		const unsigned numa_node,
+		struct pmd_internals **internals,
+		struct rte_eth_dev **eth_dev)
+{
+	struct rte_eth_dev_data *data = NULL;
+	struct rte_pci_device *pci_dev = NULL;
+
+	RTE_LOG(INFO, PMD,
+			"Creating szedata2-backed ethdev on numa socket %u\n", numa_node);
+
+	/*
+	 * now do all data allocation - for eth_dev structure, dummy pci driver
+	 * and internal (private) data
+	 */
+	data = rte_zmalloc_socket(name, sizeof(*data), 0, numa_node);
+	if (data == NULL)
+		goto error;
+
+	pci_dev = rte_zmalloc_socket(name, sizeof(*pci_dev), 0, numa_node);
+	if (pci_dev == NULL)
+		goto error;
+
+	*internals = rte_zmalloc_socket(name, sizeof(**internals), 0, numa_node);
+	if (*internals == NULL)
+		goto error;
+
+	/* reserve an ethdev entry */
+	*eth_dev = rte_eth_dev_allocate(name, RTE_ETH_DEV_VIRTUAL);
+	if (*eth_dev == NULL)
+		goto error;
+
+	/*
+	 * now put it all together
+	 * - store queue data in internals,
+	 * - store numa_node info in pci_driver
+	 * - point eth_dev_data to internals and pci_driver
+	 * - and point eth_dev structure to new eth_dev_data structure
+	 *
+	 * NOTE: we'll replace the data element, of originally allocated eth_dev
+	 * so the rings are local per-process
+	 */
+
+	(*internals)->nb_rx_queues = nb_rx_queues;
+	(*internals)->nb_tx_queues = nb_tx_queues;
+
+	(*internals)->if_index = 0;
+
+	pci_dev->numa_node = numa_node;
+
+	data->dev_private = *internals;
+	data->port_id = (*eth_dev)->data->port_id;
+	snprintf(data->name, sizeof(data->name), "%s", (*eth_dev)->data->name);
+	data->nb_rx_queues = (uint16_t)nb_rx_queues;
+	data->nb_tx_queues = (uint16_t)nb_tx_queues;
+	data->dev_link = pmd_link;
+	data->mac_addrs = &eth_addr;
+
+	(*eth_dev)->data = data;
+	(*eth_dev)->dev_ops = &ops;
+	(*eth_dev)->pci_dev = pci_dev;
+
+	return 0;
+
+error:
+	if (data)
+		rte_free(data);
+	if (pci_dev)
+		rte_free(pci_dev);
+	if (*internals)
+		rte_free(*internals);
+	return -1;
+}
+
+static int
+rte_eth_from_szedata2(const char *name,
+		struct rxtx_szedata2 *szedata2,
+		const unsigned numa_node)
+{
+	struct pmd_internals *internals = NULL;
+	struct rte_eth_dev *eth_dev = NULL;
+	int ret;
+	unsigned i;
+	uint32_t x;
+	uint32_t rx;
+	uint32_t tx;
+	uint32_t num_rx_sub = 0;
+	uint32_t num_tx_sub = 0;
+
+	if (rte_pmd_init_internals(name, 0, 0, numa_node,
+			&internals, &eth_dev) < 0)
+		return -1;
+
+	internals->sze_dev = szedata2->sze_dev;
+	internals->sze_rx_req = szedata2->sze_rx_mask_req;
+	internals->sze_tx_req = szedata2->sze_tx_mask_req;
+	internals->num_of_rx = szedata2->num_of_rx;
+	internals->num_of_tx = szedata2->num_of_tx;
+
+	RTE_LOG(INFO, PMD, "Number of rx channels to open: %u mask: 0x%x\n",
+			internals->num_of_rx, internals->sze_rx_req);
+	RTE_LOG(INFO, PMD, "Number of tx channels to open: %u mask: 0x%x\n",
+			internals->num_of_tx, internals->sze_tx_req);
+
+	rx = internals->sze_rx_req;
+	tx = 0;
+
+	for (i = 0; i < internals->num_of_rx; i++) {
+		/*
+		 * Open, subscribe rx,tx channels and start device
+		 */
+		RTE_LOG(INFO, PMD, "Opening SZE device %u. time\n", i);
+
+		internals->rx_queue[num_rx_sub].sze = szedata_open(internals->sze_dev);
+		if (internals->rx_queue[num_rx_sub].sze == NULL)
+			return -1;
+
+		x = rx & ((~rx)+1); /* separate least significant non-zero bit */
+
+		RTE_LOG(INFO, PMD, "Subscribing rx channel: 0x%x tx channel: 0x%x\n",
+				x, tx);
+
+		ret = szedata_subscribe3(internals->rx_queue[num_rx_sub].sze, &x, &tx);
+		if (ret) {
+			szedata_close(internals->rx_queue[num_rx_sub].sze);
+			internals->rx_queue[num_rx_sub].sze = NULL;
+			return -1;
+		}
+
+		RTE_LOG(INFO, PMD, "Subscribed rx channel: 0x%x tx channel: 0x%x\n",
+				x, tx);
+
+		if (x) {
+			RTE_LOG(INFO, PMD, "Starting SZE device for rx queue: %u\n",
+					num_rx_sub);
+
+			ret = szedata_start(internals->rx_queue[num_rx_sub].sze);
+			if (ret) {
+				szedata_close(internals->rx_queue[num_rx_sub].sze);
+				internals->rx_queue[num_rx_sub].sze = NULL;
+				return -1;
+			}
+
+			/*
+			 * set to 1 all bits lower than bit set to 1
+			 * and that bit to 0
+			 */
+			x -= 1;
+			internals->rx_queue[num_rx_sub].rx_channel = count_set_bits(x);
+			RTE_LOG(INFO, PMD, "Subscribed rx channel no: %u\n",
+					internals->rx_queue[num_rx_sub].rx_channel);
+			num_rx_sub++;
+		} else {
+			RTE_LOG(INFO, PMD,
+				"Could not subscribe any rx channel. Closing SZE device\n");
+
+			szedata_close(internals->rx_queue[num_rx_sub].sze);
+			internals->rx_queue[num_rx_sub].sze = NULL;
+		}
+
+		rx = rx & (rx-1); /* set least significant non-zero bit to zero */
+	}
+
+	rx = 0;
+	tx = internals->sze_tx_req;
+
+	for (i = 0; i < internals->num_of_tx; i++) {
+		/*
+		 * Open, subscribe rx,tx channels and start device
+		 */
+		RTE_LOG(INFO, PMD, "Opening SZE device %u. time\n",
+				i + internals->num_of_rx);
+
+		internals->tx_queue[num_tx_sub].sze = szedata_open(internals->sze_dev);
+		if (internals->tx_queue[num_tx_sub].sze == NULL)
+			return -1;
+
+		x = tx & ((~tx)+1); /* separate least significant non-zero bit */
+
+		RTE_LOG(INFO, PMD, "Subscribing rx channel: 0x%x tx channel: 0x%x\n",
+				rx, x);
+
+		ret = szedata_subscribe3(internals->tx_queue[num_tx_sub].sze, &rx, &x);
+		if (ret) {
+			szedata_close(internals->tx_queue[num_tx_sub].sze);
+			internals->tx_queue[num_tx_sub].sze = NULL;
+			return -1;
+		}
+
+		RTE_LOG(INFO, PMD, "Subscribed rx channel: 0x%x tx channel: 0x%x\n",
+				rx, x);
+
+		if (x) {
+			RTE_LOG(INFO, PMD, "Starting SZE device for tx queue: %u\n",
+					num_tx_sub);
+
+			ret = szedata_start(internals->tx_queue[num_tx_sub].sze);
+			if (ret) {
+				szedata_close(internals->tx_queue[num_tx_sub].sze);
+				internals->tx_queue[num_tx_sub].sze = NULL;
+				return -1;
+			}
+
+			/*
+			 * set to 1 all bits lower than bit set to 1
+			 * and that bit to 0
+			 */
+			x -= 1;
+			internals->tx_queue[num_tx_sub].tx_channel = count_set_bits(x);
+			num_tx_sub++;
+		} else {
+			RTE_LOG(INFO, PMD,
+				"Could not subscribe any tx channel. Closing SZE device\n");
+
+			szedata_close(internals->tx_queue[num_tx_sub].sze);
+			internals->tx_queue[num_tx_sub].sze = NULL;
+		}
+
+		tx = tx & (tx-1); /* set least significant non-zero bit to zero */
+	}
+
+	RTE_LOG(INFO, PMD, "Successfully opened rx channels: %u\n",
+			num_rx_sub);
+	RTE_LOG(INFO, PMD, "Successfully opened tx channels: %u\n",
+			num_tx_sub);
+
+	internals->nb_rx_queues = num_rx_sub;
+	internals->nb_tx_queues = num_tx_sub;
+
+	eth_dev->data->nb_rx_queues = (uint16_t) num_rx_sub;
+	eth_dev->data->nb_tx_queues = (uint16_t) num_tx_sub;
+
+	eth_dev->rx_pkt_burst = eth_szedata2_rx;
+	eth_dev->tx_pkt_burst = eth_szedata2_tx;
+
+	return 0;
+}
+
+
+static int
+rte_pmd_szedata2_devinit(const char *name, const char *params)
+{
+	unsigned numa_node;
+	int ret;
+	struct rte_kvargs *kvlist;
+	unsigned k_idx;
+	struct rte_kvargs_pair *pair = NULL;
+	struct rxtx_szedata2 szedata2 = { 0, 0, 0, 0, NULL };
+	bool dev_path_missing = true;
+
+	RTE_LOG(INFO, PMD, "Initializing pmd_szedata2 for %s\n", name);
+
+	numa_node = rte_socket_id();
+
+	kvlist = rte_kvargs_parse(params, valid_arguments);
+	if (kvlist == NULL)
+		return -1;
+
+	/*
+	 * Get szedata2 device path and rx,tx channels from passed arguments.
+	 */
+
+	if (rte_kvargs_count(kvlist, RTE_ETH_SZEDATA2_DEV_PATH_ARG) != 1)
+		goto err;
+
+	if (rte_kvargs_count(kvlist, RTE_ETH_SZEDATA2_RX_IFACES_ARG) < 1)
+		goto err;
+
+	if (rte_kvargs_count(kvlist, RTE_ETH_SZEDATA2_TX_IFACES_ARG) < 1)
+		goto err;
+
+	for (k_idx = 0; k_idx < kvlist->count; k_idx++) {
+		pair = &kvlist->pairs[k_idx];
+		if (strstr(pair->key, RTE_ETH_SZEDATA2_DEV_PATH_ARG) != NULL) {
+			szedata2.sze_dev = pair->value;
+			dev_path_missing = false;
+			break;
+		}
+	}
+
+	if (dev_path_missing)
+		goto err;
+
+	ret = rte_kvargs_process(kvlist, RTE_ETH_SZEDATA2_RX_IFACES_ARG,
+			&add_rx_mask, &szedata2);
+	if (ret < 0)
+		goto err;
+
+	ret = rte_kvargs_process(kvlist, RTE_ETH_SZEDATA2_TX_IFACES_ARG,
+			&add_tx_mask, &szedata2);
+	if (ret < 0)
+		goto err;
+
+	szedata2.num_of_rx = count_set_bits(szedata2.sze_rx_mask_req);
+	szedata2.num_of_tx = count_set_bits(szedata2.sze_tx_mask_req);
+
+	RTE_LOG(INFO, PMD, "SZE device found at path %s\n", szedata2.sze_dev);
+
+	return rte_eth_from_szedata2(name, &szedata2, numa_node);
+err:
+	rte_kvargs_free(kvlist);
+	return -1;
+}
+
+static struct rte_driver pmd_szedata2_drv = {
+	.name = "eth_szedata2",
+	.type = PMD_VDEV,
+	.init = rte_pmd_szedata2_devinit,
+};
+
+PMD_REGISTER_DRIVER(pmd_szedata2_drv);
diff --git a/drivers/net/szedata2/rte_eth_szedata2.h b/drivers/net/szedata2/rte_eth_szedata2.h
new file mode 100644
index 0000000..cf4f1c9
--- /dev/null
+++ b/drivers/net/szedata2/rte_eth_szedata2.h
@@ -0,0 +1,96 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright (c) 2015 CESNET
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of CESNET nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef RTE_PMD_SZEDATA2_H_
+#define RTE_PMD_SZEDATA2_H_
+
+/* szedata2_packet header length == 4 bytes == 2B segment size + 2B hw size */
+#define RTE_SZE2_PACKET_HEADER_SIZE 4
+
+#define RTE_SZE2_MMIO_MAX 10
+
+/*!
+ * Round 'what' to the nearest larger (or equal) multiple of '8'
+ * (szedata2 packet is aligned to 8 bytes)
+ */
+#define RTE_SZE2_ALIGN8(what) (((what) + ((8)-1)) & (~((8)-1)))
+
+/*! main handle structure */
+struct szedata {
+	int fd;
+	struct sze2_instance_info *info;
+	uint32_t *write_size;
+	void *space[RTE_SZE2_MMIO_MAX];
+	struct szedata_lock lock[2][2];
+
+	__u32 *rx_asize, *tx_asize;
+
+	/* szedata_read_next variables - to keep context (ct) */
+
+	/*
+	 * rx
+	 */
+	/** initial sze lock ptr */
+	const struct szedata_lock   *ct_rx_lck_orig;
+	/** current sze lock ptr (initial or next) */
+	const struct szedata_lock   *ct_rx_lck;
+	/** remaining bytes (not read) within current lock */
+	unsigned int                ct_rx_rem_bytes;
+	/** current pointer to locked memory */
+	unsigned char               *ct_rx_cur_ptr;
+	/** allocated buffer to store RX packet if it was split into 2 buffers */
+	unsigned char               *ct_rx_buffer;
+	/** registered function to provide filtering based on hwdata */
+	int (* ct_rx_filter)(u_int16_t hwdata_len, u_char *hwdata);
+
+	/*
+	 * tx
+	 */
+	/** buffer for tx - packet is prepared here (in future for burst write) */
+	unsigned char               *ct_tx_buffer;
+	/** initial sze TX lock ptrs - number according to TX interfaces */
+	const struct szedata_lock   **ct_tx_lck_orig;
+	/** current sze TX lock ptrs - number according to TX interfaces */
+	const struct szedata_lock   **ct_tx_lck;
+	/** already written bytes in both locks */
+	unsigned int                *ct_tx_written_bytes;
+	/** remaining bytes (not written) within current lock */
+	unsigned int                *ct_tx_rem_bytes;
+	/** current pointers to locked memory */
+	unsigned char               **ct_tx_cur_ptr;
+	/** NUMA node closest to PCIe device, or -1 */
+	int                         numa_node;
+};
+
+
+#endif
diff --git a/drivers/net/szedata2/rte_pmd_szedata2_version.map b/drivers/net/szedata2/rte_pmd_szedata2_version.map
new file mode 100644
index 0000000..ef35398
--- /dev/null
+++ b/drivers/net/szedata2/rte_pmd_szedata2_version.map
@@ -0,0 +1,4 @@
+DPDK_2.0 {
+
+	local: *;
+};
diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index 1a2043a..f9ef4b1 100644
--- a/mk/rte.app.mk
+++ b/mk/rte.app.mk
@@ -92,6 +92,8 @@ endif # ! CONFIG_RTE_BUILD_COMBINE_LIBS
 
 _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_PCAP)       += -lpcap
 
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2)   += -lsze2
+
 ifeq ($(CONFIG_RTE_LIBRTE_VHOST_USER),n)
 _LDLIBS-$(CONFIG_RTE_LIBRTE_VHOST)          += -lfuse
 endif
@@ -132,6 +134,7 @@ _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_RING)       += -lrte_pmd_ring
 _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_PCAP)       += -lrte_pmd_pcap
 _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_AF_PACKET)  += -lrte_pmd_af_packet
 _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_NULL)       += -lrte_pmd_null
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2)   += -lrte_pmd_szedata2
 
 endif # ! $(CONFIG_RTE_BUILD_SHARED_LIB)
 
-- 
Matej Vido
Software Developer
CESNET, a. l. e.



More information about the dev mailing list