summaryrefslogtreecommitdiffstats
path: root/src/seastar/dpdk/drivers/net/szedata2
diff options
context:
space:
mode:
Diffstat (limited to 'src/seastar/dpdk/drivers/net/szedata2')
-rw-r--r--src/seastar/dpdk/drivers/net/szedata2/Makefile57
-rw-r--r--src/seastar/dpdk/drivers/net/szedata2/rte_eth_szedata2.c1613
-rw-r--r--src/seastar/dpdk/drivers/net/szedata2/rte_eth_szedata2.h450
-rw-r--r--src/seastar/dpdk/drivers/net/szedata2/rte_pmd_szedata2_version.map3
4 files changed, 2123 insertions, 0 deletions
diff --git a/src/seastar/dpdk/drivers/net/szedata2/Makefile b/src/seastar/dpdk/drivers/net/szedata2/Makefile
new file mode 100644
index 00000000..836c3b2a
--- /dev/null
+++ b/src/seastar/dpdk/drivers/net/szedata2/Makefile
@@ -0,0 +1,57 @@
+# 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 += $(WERROR_FLAGS)
+LDLIBS += -lsze2
+
+EXPORT_MAP := rte_pmd_szedata2_version.map
+
+LIBABIVER := 1
+
+#
+# all source are stored in SRCS-y
+#
+SRCS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2) += rte_eth_szedata2.c
+
+#
+# Export include files
+#
+SYMLINK-y-include +=
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/src/seastar/dpdk/drivers/net/szedata2/rte_eth_szedata2.c b/src/seastar/dpdk/drivers/net/szedata2/rte_eth_szedata2.c
new file mode 100644
index 00000000..54212b71
--- /dev/null
+++ b/src/seastar/dpdk/drivers/net/szedata2/rte_eth_szedata2.c
@@ -0,0 +1,1613 @@
+/*-
+ * BSD LICENSE
+ *
+ * Copyright (c) 2015 - 2016 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 <sys/types.h>
+#include <dirent.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <sys/mman.h>
+
+#include <libsze2.h>
+
+#include <rte_mbuf.h>
+#include <rte_ethdev.h>
+#include <rte_ethdev_pci.h>
+#include <rte_malloc.h>
+#include <rte_memcpy.h>
+#include <rte_kvargs.h>
+#include <rte_dev.h>
+#include <rte_atomic.h>
+
+#include "rte_eth_szedata2.h"
+
+#define RTE_ETH_SZEDATA2_MAX_RX_QUEUES 32
+#define RTE_ETH_SZEDATA2_MAX_TX_QUEUES 32
+#define RTE_ETH_SZEDATA2_TX_LOCK_SIZE (32 * 1024 * 1024)
+
+/**
+ * size of szedata2_packet header with alignment
+ */
+#define RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED 8
+
+#define RTE_SZEDATA2_DRIVER_NAME net_szedata2
+#define RTE_SZEDATA2_PCI_DRIVER_NAME "rte_szedata2_pmd"
+
+#define SZEDATA2_DEV_PATH_FMT "/dev/szedataII%u"
+
+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 tx_bytes;
+ volatile uint64_t err_pkts;
+};
+
+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];
+ uint16_t max_rx_queues;
+ uint16_t max_tx_queues;
+ char sze_dev[PATH_MAX];
+ struct rte_mem_resource *pci_rsc;
+};
+
+static struct ether_addr eth_addr = {
+ .addr_bytes = { 0x00, 0x11, 0x17, 0x00, 0x00, 0x00 }
+};
+
+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++) {
+ mbuf = rte_pktmbuf_alloc(sze_q->mb_pool);
+
+ if (unlikely(mbuf == NULL))
+ break;
+
+ /* 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 = szedata_rx_lock_data(sze_q->sze, ~0U);
+ sze->ct_rx_lck_orig = sze->ct_rx_lck;
+
+ if (sze->ct_rx_lck == NULL) {
+ /* nothing to lock */
+ rte_pktmbuf_free(mbuf);
+ 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) {
+ rte_pktmbuf_free(mbuf);
+ 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) {
+ rte_pktmbuf_free(mbuf);
+ 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 */
+ 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 */
+ 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)) {
+ rte_pktmbuf_free(mbuf);
+ 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 one 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 = packet_size;
+ mbuf->port = sze_q->in_port;
+ bufs[num_rx] = mbuf;
+ num_rx++;
+ num_bytes += packet_size;
+ } else {
+ /*
+ * sze packet will not fit in one mbuf,
+ * scattered mode is not enabled, drop packet
+ */
+ RTE_LOG(ERR, PMD,
+ "SZE segment %d bytes will not fit in one mbuf "
+ "(%d bytes), scattered mode is not enabled, "
+ "drop packet!!\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_rx_scattered(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++) {
+ const struct szedata_lock *ct_rx_lck_backup;
+ unsigned int ct_rx_rem_bytes_backup;
+ unsigned char *ct_rx_cur_ptr_backup;
+
+ /* 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;
+ }
+
+ /*
+ * Store items from sze structure which can be changed
+ * before mbuf allocating. Use these items in case of mbuf
+ * allocating failure.
+ */
+ ct_rx_lck_backup = sze->ct_rx_lck;
+ ct_rx_rem_bytes_backup = sze->ct_rx_rem_bytes;
+ ct_rx_cur_ptr_backup = sze->ct_rx_cur_ptr;
+
+ if (!sze->ct_rx_rem_bytes && sze->ct_rx_lck_orig == NULL) {
+ /* nothing to read, lock new data */
+ sze->ct_rx_lck = szedata_rx_lock_data(sze_q->sze, ~0U);
+ sze->ct_rx_lck_orig = sze->ct_rx_lck;
+
+ /*
+ * Backup items from sze structure must be updated
+ * after locking to contain pointers to new locks.
+ */
+ ct_rx_lck_backup = sze->ct_rx_lck;
+ ct_rx_rem_bytes_backup = sze->ct_rx_rem_bytes;
+ ct_rx_cur_ptr_backup = sze->ct_rx_cur_ptr;
+
+ 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 */
+ 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 */
+ 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;
+
+ mbuf = rte_pktmbuf_alloc(sze_q->mb_pool);
+
+ if (unlikely(mbuf == NULL)) {
+ /*
+ * Restore items from sze structure to state after
+ * unlocking (eventually locking).
+ */
+ sze->ct_rx_lck = ct_rx_lck_backup;
+ sze->ct_rx_rem_bytes = ct_rx_rem_bytes_backup;
+ sze->ct_rx_cur_ptr = ct_rx_cur_ptr_backup;
+ 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 one 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;
+ } else {
+ /*
+ * sze packet will not fit in one mbuf,
+ * scatter packet into more mbufs
+ */
+ struct rte_mbuf *m = mbuf;
+ uint16_t len = rte_pktmbuf_tailroom(mbuf);
+
+ /* copy first part of packet */
+ /* fill first mbuf */
+ rte_memcpy(rte_pktmbuf_append(mbuf, len), packet_ptr1,
+ len);
+ packet_len1 -= len;
+ packet_ptr1 = ((uint8_t *)packet_ptr1) + len;
+
+ while (packet_len1 > 0) {
+ /* fill new mbufs */
+ m->next = rte_pktmbuf_alloc(sze_q->mb_pool);
+
+ if (unlikely(m->next == NULL)) {
+ rte_pktmbuf_free(mbuf);
+ /*
+ * Restore items from sze structure
+ * to state after unlocking (eventually
+ * locking).
+ */
+ sze->ct_rx_lck = ct_rx_lck_backup;
+ sze->ct_rx_rem_bytes =
+ ct_rx_rem_bytes_backup;
+ sze->ct_rx_cur_ptr =
+ ct_rx_cur_ptr_backup;
+ goto finish;
+ }
+
+ m = m->next;
+
+ len = RTE_MIN(rte_pktmbuf_tailroom(m),
+ packet_len1);
+ rte_memcpy(rte_pktmbuf_append(mbuf, len),
+ packet_ptr1, len);
+
+ (mbuf->nb_segs)++;
+ packet_len1 -= len;
+ packet_ptr1 = ((uint8_t *)packet_ptr1) + len;
+ }
+
+ if (packet_ptr2 != NULL) {
+ /* copy second part of packet, if exists */
+ /* fill the rest of currently last mbuf */
+ len = rte_pktmbuf_tailroom(m);
+ rte_memcpy(rte_pktmbuf_append(mbuf, len),
+ packet_ptr2, len);
+ packet_len2 -= len;
+ packet_ptr2 = ((uint8_t *)packet_ptr2) + len;
+
+ while (packet_len2 > 0) {
+ /* fill new mbufs */
+ m->next = rte_pktmbuf_alloc(
+ sze_q->mb_pool);
+
+ if (unlikely(m->next == NULL)) {
+ rte_pktmbuf_free(mbuf);
+ /*
+ * Restore items from sze
+ * structure to state after
+ * unlocking (eventually
+ * locking).
+ */
+ sze->ct_rx_lck =
+ ct_rx_lck_backup;
+ sze->ct_rx_rem_bytes =
+ ct_rx_rem_bytes_backup;
+ sze->ct_rx_cur_ptr =
+ ct_rx_cur_ptr_backup;
+ goto finish;
+ }
+
+ m = m->next;
+
+ len = RTE_MIN(rte_pktmbuf_tailroom(m),
+ packet_len2);
+ rte_memcpy(
+ rte_pktmbuf_append(mbuf, len),
+ packet_ptr2, len);
+
+ (mbuf->nb_segs)++;
+ packet_len2 -= len;
+ packet_ptr2 = ((uint8_t *)packet_ptr2) +
+ len;
+ }
+ }
+ }
+ mbuf->pkt_len = packet_size;
+ mbuf->port = sze_q->in_port;
+ bufs[num_rx] = mbuf;
+ num_rx++;
+ num_bytes += packet_size;
+ }
+
+finish:
+ 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)
+{
+ 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;
+ uint8_t mbuf_segs;
+ uint16_t pkt_left = nb_pkts;
+
+ if (sze_q->sze == NULL || nb_pkts == 0)
+ return 0;
+
+ while (pkt_left > 0) {
+ unlock_size = 0;
+ lck = szedata_tx_lock_data(sze_q->sze,
+ RTE_ETH_SZEDATA2_TX_LOCK_SIZE,
+ sze_q->tx_channel);
+ if (lck == NULL)
+ continue;
+
+ dst = lck->start;
+ lock_size = lck->len;
+ lock_size2 = lck->next ? lck->next->len : 0;
+
+next_packet:
+ mbuf = bufs[nb_pkts - pkt_left];
+
+ pkt_len = mbuf->pkt_len;
+ mbuf_segs = mbuf->nb_segs;
+
+ 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);
+ continue;
+ }
+
+ num_bytes += pkt_len;
+
+ if (lock_size > hwpkt_len) {
+ void *tmp_dst;
+
+ 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 */
+ tmp_dst = ((uint8_t *)(dst)) +
+ RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED;
+ if (mbuf_segs == 1) {
+ /*
+ * non-scattered packet,
+ * transmit from one mbuf
+ */
+ rte_memcpy(tmp_dst,
+ rte_pktmbuf_mtod(mbuf, const void *),
+ pkt_len);
+ } else {
+ /* scattered packet, transmit from more mbufs */
+ struct rte_mbuf *m = mbuf;
+ while (m) {
+ rte_memcpy(tmp_dst,
+ rte_pktmbuf_mtod(m,
+ const void *),
+ m->data_len);
+ tmp_dst = ((uint8_t *)(tmp_dst)) +
+ m->data_len;
+ m = m->next;
+ }
+ }
+
+
+ dst = ((uint8_t *)dst) + hwpkt_len;
+ unlock_size += hwpkt_len;
+ lock_size -= hwpkt_len;
+
+ rte_pktmbuf_free(mbuf);
+ num_tx++;
+ pkt_left--;
+ if (pkt_left == 0) {
+ szedata_tx_unlock_data(sze_q->sze, lck,
+ unlock_size);
+ break;
+ }
+ goto next_packet;
+ } else if (lock_size + lock_size2 >= hwpkt_len) {
+ void *tmp_dst;
+ uint16_t write_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;
+
+ rem_len = hwpkt_len - lock_size;
+
+ tmp_dst = ((uint8_t *)(dst)) +
+ RTE_SZE2_PACKET_HEADER_SIZE_ALIGNED;
+ if (mbuf_segs == 1) {
+ /*
+ * non-scattered packet,
+ * transmit from one mbuf
+ */
+ /* copy part of packet to first area */
+ rte_memcpy(tmp_dst,
+ rte_pktmbuf_mtod(mbuf, const void *),
+ write_len);
+
+ if (lck->next)
+ dst = lck->next->start;
+
+ /* copy part of packet to second area */
+ rte_memcpy(dst,
+ (const void *)(rte_pktmbuf_mtod(mbuf,
+ const uint8_t *) +
+ write_len), pkt_len - write_len);
+ } else {
+ /* scattered packet, transmit from more mbufs */
+ struct rte_mbuf *m = mbuf;
+ uint16_t written = 0;
+ uint16_t to_write = 0;
+ bool new_mbuf = true;
+ uint16_t write_off = 0;
+
+ /* copy part of packet to first area */
+ while (m && written < write_len) {
+ to_write = RTE_MIN(m->data_len,
+ write_len - written);
+ rte_memcpy(tmp_dst,
+ rte_pktmbuf_mtod(m,
+ const void *),
+ to_write);
+
+ tmp_dst = ((uint8_t *)(tmp_dst)) +
+ to_write;
+ if (m->data_len <= write_len -
+ written) {
+ m = m->next;
+ new_mbuf = true;
+ } else {
+ new_mbuf = false;
+ }
+ written += to_write;
+ }
+
+ if (lck->next)
+ dst = lck->next->start;
+
+ tmp_dst = dst;
+ written = 0;
+ write_off = new_mbuf ? 0 : to_write;
+
+ /* copy part of packet to second area */
+ while (m && written < pkt_len - write_len) {
+ rte_memcpy(tmp_dst, (const void *)
+ (rte_pktmbuf_mtod(m,
+ uint8_t *) + write_off),
+ m->data_len - write_off);
+
+ tmp_dst = ((uint8_t *)(tmp_dst)) +
+ (m->data_len - write_off);
+ written += m->data_len - write_off;
+ m = m->next;
+ write_off = 0;
+ }
+ }
+
+ 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);
+ pkt_left--;
+ }
+
+ 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_rx_queue_start(struct rte_eth_dev *dev, uint16_t rxq_id)
+{
+ struct szedata2_rx_queue *rxq = dev->data->rx_queues[rxq_id];
+ int ret;
+ struct pmd_internals *internals = (struct pmd_internals *)
+ dev->data->dev_private;
+
+ if (rxq->sze == NULL) {
+ uint32_t rx = 1 << rxq->rx_channel;
+ uint32_t tx = 0;
+ rxq->sze = szedata_open(internals->sze_dev);
+ if (rxq->sze == NULL)
+ return -EINVAL;
+ ret = szedata_subscribe3(rxq->sze, &rx, &tx);
+ if (ret != 0 || rx == 0)
+ goto err;
+ }
+
+ ret = szedata_start(rxq->sze);
+ if (ret != 0)
+ goto err;
+ dev->data->rx_queue_state[rxq_id] = RTE_ETH_QUEUE_STATE_STARTED;
+ return 0;
+
+err:
+ szedata_close(rxq->sze);
+ rxq->sze = NULL;
+ return -EINVAL;
+}
+
+static int
+eth_rx_queue_stop(struct rte_eth_dev *dev, uint16_t rxq_id)
+{
+ struct szedata2_rx_queue *rxq = dev->data->rx_queues[rxq_id];
+
+ if (rxq->sze != NULL) {
+ szedata_close(rxq->sze);
+ rxq->sze = NULL;
+ }
+
+ dev->data->rx_queue_state[rxq_id] = RTE_ETH_QUEUE_STATE_STOPPED;
+ return 0;
+}
+
+static int
+eth_tx_queue_start(struct rte_eth_dev *dev, uint16_t txq_id)
+{
+ struct szedata2_tx_queue *txq = dev->data->tx_queues[txq_id];
+ int ret;
+ struct pmd_internals *internals = (struct pmd_internals *)
+ dev->data->dev_private;
+
+ if (txq->sze == NULL) {
+ uint32_t rx = 0;
+ uint32_t tx = 1 << txq->tx_channel;
+ txq->sze = szedata_open(internals->sze_dev);
+ if (txq->sze == NULL)
+ return -EINVAL;
+ ret = szedata_subscribe3(txq->sze, &rx, &tx);
+ if (ret != 0 || tx == 0)
+ goto err;
+ }
+
+ ret = szedata_start(txq->sze);
+ if (ret != 0)
+ goto err;
+ dev->data->tx_queue_state[txq_id] = RTE_ETH_QUEUE_STATE_STARTED;
+ return 0;
+
+err:
+ szedata_close(txq->sze);
+ txq->sze = NULL;
+ return -EINVAL;
+}
+
+static int
+eth_tx_queue_stop(struct rte_eth_dev *dev, uint16_t txq_id)
+{
+ struct szedata2_tx_queue *txq = dev->data->tx_queues[txq_id];
+
+ if (txq->sze != NULL) {
+ szedata_close(txq->sze);
+ txq->sze = NULL;
+ }
+
+ dev->data->tx_queue_state[txq_id] = RTE_ETH_QUEUE_STATE_STOPPED;
+ return 0;
+}
+
+static int
+eth_dev_start(struct rte_eth_dev *dev)
+{
+ int ret;
+ uint16_t i;
+ uint16_t nb_rx = dev->data->nb_rx_queues;
+ uint16_t nb_tx = dev->data->nb_tx_queues;
+
+ for (i = 0; i < nb_rx; i++) {
+ ret = eth_rx_queue_start(dev, i);
+ if (ret != 0)
+ goto err_rx;
+ }
+
+ for (i = 0; i < nb_tx; i++) {
+ ret = eth_tx_queue_start(dev, i);
+ if (ret != 0)
+ goto err_tx;
+ }
+
+ return 0;
+
+err_tx:
+ for (i = 0; i < nb_tx; i++)
+ eth_tx_queue_stop(dev, i);
+err_rx:
+ for (i = 0; i < nb_rx; i++)
+ eth_rx_queue_stop(dev, i);
+ return ret;
+}
+
+static void
+eth_dev_stop(struct rte_eth_dev *dev)
+{
+ uint16_t i;
+ uint16_t nb_rx = dev->data->nb_rx_queues;
+ uint16_t nb_tx = dev->data->nb_tx_queues;
+
+ for (i = 0; i < nb_tx; i++)
+ eth_tx_queue_stop(dev, i);
+
+ for (i = 0; i < nb_rx; i++)
+ eth_rx_queue_stop(dev, i);
+}
+
+static int
+eth_dev_configure(struct rte_eth_dev *dev)
+{
+ struct rte_eth_dev_data *data = dev->data;
+ if (data->dev_conf.rxmode.enable_scatter == 1) {
+ dev->rx_pkt_burst = eth_szedata2_rx_scattered;
+ data->scattered_rx = 1;
+ } else {
+ dev->rx_pkt_burst = eth_szedata2_rx;
+ data->scattered_rx = 0;
+ }
+ 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->pci_dev = RTE_DEV_TO_PCI(dev->device);
+ dev_info->if_index = 0;
+ dev_info->max_mac_addrs = 1;
+ dev_info->max_rx_pktlen = (uint32_t)-1;
+ dev_info->max_rx_queues = internals->max_rx_queues;
+ dev_info->max_tx_queues = internals->max_tx_queues;
+ dev_info->min_rx_bufsize = 0;
+ dev_info->speed_capa = ETH_LINK_SPEED_100G;
+}
+
+static void
+eth_stats_get(struct rte_eth_dev *dev,
+ struct rte_eth_stats *stats)
+{
+ uint16_t i;
+ uint16_t nb_rx = dev->data->nb_rx_queues;
+ uint16_t nb_tx = dev->data->nb_tx_queues;
+ 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 *internals = dev->data->dev_private;
+
+ for (i = 0; i < RTE_ETHDEV_QUEUE_STAT_CNTRS && i < nb_rx; i++) {
+ stats->q_ipackets[i] = internals->rx_queue[i].rx_pkts;
+ stats->q_ibytes[i] = internals->rx_queue[i].rx_bytes;
+ rx_total += stats->q_ipackets[i];
+ rx_total_bytes += stats->q_ibytes[i];
+ }
+
+ for (i = 0; i < RTE_ETHDEV_QUEUE_STAT_CNTRS && i < nb_tx; i++) {
+ stats->q_opackets[i] = internals->tx_queue[i].tx_pkts;
+ stats->q_obytes[i] = internals->tx_queue[i].tx_bytes;
+ stats->q_errors[i] = internals->tx_queue[i].err_pkts;
+ tx_total += stats->q_opackets[i];
+ tx_total_bytes += stats->q_obytes[i];
+ tx_err_total += stats->q_errors[i];
+ }
+
+ stats->ipackets = rx_total;
+ stats->opackets = tx_total;
+ stats->ibytes = rx_total_bytes;
+ stats->obytes = tx_total_bytes;
+ stats->oerrors = tx_err_total;
+}
+
+static void
+eth_stats_reset(struct rte_eth_dev *dev)
+{
+ uint16_t i;
+ uint16_t nb_rx = dev->data->nb_rx_queues;
+ uint16_t nb_tx = dev->data->nb_tx_queues;
+ struct pmd_internals *internals = dev->data->dev_private;
+
+ for (i = 0; i < nb_rx; i++) {
+ internals->rx_queue[i].rx_pkts = 0;
+ internals->rx_queue[i].rx_bytes = 0;
+ internals->rx_queue[i].err_pkts = 0;
+ }
+ for (i = 0; i < nb_tx; i++) {
+ internals->tx_queue[i].tx_pkts = 0;
+ internals->tx_queue[i].tx_bytes = 0;
+ internals->tx_queue[i].err_pkts = 0;
+ }
+}
+
+static void
+eth_rx_queue_release(void *q)
+{
+ struct szedata2_rx_queue *rxq = (struct szedata2_rx_queue *)q;
+ if (rxq->sze != NULL) {
+ szedata_close(rxq->sze);
+ rxq->sze = NULL;
+ }
+}
+
+static void
+eth_tx_queue_release(void *q)
+{
+ struct szedata2_tx_queue *txq = (struct szedata2_tx_queue *)q;
+ if (txq->sze != NULL) {
+ szedata_close(txq->sze);
+ txq->sze = NULL;
+ }
+}
+
+static void
+eth_dev_close(struct rte_eth_dev *dev)
+{
+ uint16_t i;
+ uint16_t nb_rx = dev->data->nb_rx_queues;
+ uint16_t nb_tx = dev->data->nb_tx_queues;
+
+ eth_dev_stop(dev);
+
+ for (i = 0; i < nb_rx; i++) {
+ eth_rx_queue_release(dev->data->rx_queues[i]);
+ dev->data->rx_queues[i] = NULL;
+ }
+ dev->data->nb_rx_queues = 0;
+ for (i = 0; i < nb_tx; i++) {
+ eth_tx_queue_release(dev->data->tx_queues[i]);
+ dev->data->tx_queues[i] = NULL;
+ }
+ dev->data->nb_tx_queues = 0;
+}
+
+static int
+eth_link_update(struct rte_eth_dev *dev,
+ int wait_to_complete __rte_unused)
+{
+ struct rte_eth_link link;
+ struct rte_eth_link *link_ptr = &link;
+ struct rte_eth_link *dev_link = &dev->data->dev_link;
+ struct pmd_internals *internals = (struct pmd_internals *)
+ dev->data->dev_private;
+ volatile struct szedata2_cgmii_ibuf *ibuf = SZEDATA2_PCI_RESOURCE_PTR(
+ internals->pci_rsc, SZEDATA2_CGMII_IBUF_BASE_OFF,
+ volatile struct szedata2_cgmii_ibuf *);
+
+ switch (cgmii_link_speed(ibuf)) {
+ case SZEDATA2_LINK_SPEED_10G:
+ link.link_speed = ETH_SPEED_NUM_10G;
+ break;
+ case SZEDATA2_LINK_SPEED_40G:
+ link.link_speed = ETH_SPEED_NUM_40G;
+ break;
+ case SZEDATA2_LINK_SPEED_100G:
+ link.link_speed = ETH_SPEED_NUM_100G;
+ break;
+ default:
+ link.link_speed = ETH_SPEED_NUM_10G;
+ break;
+ }
+
+ /* szedata2 uses only full duplex */
+ link.link_duplex = ETH_LINK_FULL_DUPLEX;
+
+ link.link_status = (cgmii_ibuf_is_enabled(ibuf) &&
+ cgmii_ibuf_is_link_up(ibuf)) ? ETH_LINK_UP : ETH_LINK_DOWN;
+
+ link.link_autoneg = ETH_LINK_SPEED_FIXED;
+
+ rte_atomic64_cmpset((uint64_t *)dev_link, *(uint64_t *)dev_link,
+ *(uint64_t *)link_ptr);
+
+ return 0;
+}
+
+static int
+eth_dev_set_link_up(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internals = (struct pmd_internals *)
+ dev->data->dev_private;
+ volatile struct szedata2_cgmii_ibuf *ibuf = SZEDATA2_PCI_RESOURCE_PTR(
+ internals->pci_rsc, SZEDATA2_CGMII_IBUF_BASE_OFF,
+ volatile struct szedata2_cgmii_ibuf *);
+ volatile struct szedata2_cgmii_obuf *obuf = SZEDATA2_PCI_RESOURCE_PTR(
+ internals->pci_rsc, SZEDATA2_CGMII_OBUF_BASE_OFF,
+ volatile struct szedata2_cgmii_obuf *);
+
+ cgmii_ibuf_enable(ibuf);
+ cgmii_obuf_enable(obuf);
+ return 0;
+}
+
+static int
+eth_dev_set_link_down(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internals = (struct pmd_internals *)
+ dev->data->dev_private;
+ volatile struct szedata2_cgmii_ibuf *ibuf = SZEDATA2_PCI_RESOURCE_PTR(
+ internals->pci_rsc, SZEDATA2_CGMII_IBUF_BASE_OFF,
+ volatile struct szedata2_cgmii_ibuf *);
+ volatile struct szedata2_cgmii_obuf *obuf = SZEDATA2_PCI_RESOURCE_PTR(
+ internals->pci_rsc, SZEDATA2_CGMII_OBUF_BASE_OFF,
+ volatile struct szedata2_cgmii_obuf *);
+
+ cgmii_ibuf_disable(ibuf);
+ cgmii_obuf_disable(obuf);
+ 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 *rxq = &internals->rx_queue[rx_queue_id];
+ int ret;
+ uint32_t rx = 1 << rx_queue_id;
+ uint32_t tx = 0;
+
+ rxq->sze = szedata_open(internals->sze_dev);
+ if (rxq->sze == NULL)
+ return -EINVAL;
+ ret = szedata_subscribe3(rxq->sze, &rx, &tx);
+ if (ret != 0 || rx == 0) {
+ szedata_close(rxq->sze);
+ rxq->sze = NULL;
+ return -EINVAL;
+ }
+ rxq->rx_channel = rx_queue_id;
+ rxq->in_port = dev->data->port_id;
+ rxq->mb_pool = mb_pool;
+ rxq->rx_pkts = 0;
+ rxq->rx_bytes = 0;
+ rxq->err_pkts = 0;
+
+ dev->data->rx_queues[rx_queue_id] = rxq;
+ 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;
+ struct szedata2_tx_queue *txq = &internals->tx_queue[tx_queue_id];
+ int ret;
+ uint32_t rx = 0;
+ uint32_t tx = 1 << tx_queue_id;
+
+ txq->sze = szedata_open(internals->sze_dev);
+ if (txq->sze == NULL)
+ return -EINVAL;
+ ret = szedata_subscribe3(txq->sze, &rx, &tx);
+ if (ret != 0 || tx == 0) {
+ szedata_close(txq->sze);
+ txq->sze = NULL;
+ return -EINVAL;
+ }
+ txq->tx_channel = tx_queue_id;
+ txq->tx_pkts = 0;
+ txq->tx_bytes = 0;
+ txq->err_pkts = 0;
+
+ dev->data->tx_queues[tx_queue_id] = txq;
+ return 0;
+}
+
+static void
+eth_mac_addr_set(struct rte_eth_dev *dev __rte_unused,
+ struct ether_addr *mac_addr __rte_unused)
+{
+}
+
+static void
+eth_promiscuous_enable(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internals = (struct pmd_internals *)
+ dev->data->dev_private;
+ volatile struct szedata2_cgmii_ibuf *ibuf = SZEDATA2_PCI_RESOURCE_PTR(
+ internals->pci_rsc, SZEDATA2_CGMII_IBUF_BASE_OFF,
+ volatile struct szedata2_cgmii_ibuf *);
+ cgmii_ibuf_mac_mode_write(ibuf, SZEDATA2_MAC_CHMODE_PROMISC);
+}
+
+static void
+eth_promiscuous_disable(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internals = (struct pmd_internals *)
+ dev->data->dev_private;
+ volatile struct szedata2_cgmii_ibuf *ibuf = SZEDATA2_PCI_RESOURCE_PTR(
+ internals->pci_rsc, SZEDATA2_CGMII_IBUF_BASE_OFF,
+ volatile struct szedata2_cgmii_ibuf *);
+ cgmii_ibuf_mac_mode_write(ibuf, SZEDATA2_MAC_CHMODE_ONLY_VALID);
+}
+
+static void
+eth_allmulticast_enable(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internals = (struct pmd_internals *)
+ dev->data->dev_private;
+ volatile struct szedata2_cgmii_ibuf *ibuf = SZEDATA2_PCI_RESOURCE_PTR(
+ internals->pci_rsc, SZEDATA2_CGMII_IBUF_BASE_OFF,
+ volatile struct szedata2_cgmii_ibuf *);
+ cgmii_ibuf_mac_mode_write(ibuf, SZEDATA2_MAC_CHMODE_ALL_MULTICAST);
+}
+
+static void
+eth_allmulticast_disable(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internals = (struct pmd_internals *)
+ dev->data->dev_private;
+ volatile struct szedata2_cgmii_ibuf *ibuf = SZEDATA2_PCI_RESOURCE_PTR(
+ internals->pci_rsc, SZEDATA2_CGMII_IBUF_BASE_OFF,
+ volatile struct szedata2_cgmii_ibuf *);
+ cgmii_ibuf_mac_mode_write(ibuf, SZEDATA2_MAC_CHMODE_ONLY_VALID);
+}
+
+static const struct eth_dev_ops ops = {
+ .dev_start = eth_dev_start,
+ .dev_stop = eth_dev_stop,
+ .dev_set_link_up = eth_dev_set_link_up,
+ .dev_set_link_down = eth_dev_set_link_down,
+ .dev_close = eth_dev_close,
+ .dev_configure = eth_dev_configure,
+ .dev_infos_get = eth_dev_info,
+ .promiscuous_enable = eth_promiscuous_enable,
+ .promiscuous_disable = eth_promiscuous_disable,
+ .allmulticast_enable = eth_allmulticast_enable,
+ .allmulticast_disable = eth_allmulticast_disable,
+ .rx_queue_start = eth_rx_queue_start,
+ .rx_queue_stop = eth_rx_queue_stop,
+ .tx_queue_start = eth_tx_queue_start,
+ .tx_queue_stop = eth_tx_queue_stop,
+ .rx_queue_setup = eth_rx_queue_setup,
+ .tx_queue_setup = eth_tx_queue_setup,
+ .rx_queue_release = eth_rx_queue_release,
+ .tx_queue_release = eth_tx_queue_release,
+ .link_update = eth_link_update,
+ .stats_get = eth_stats_get,
+ .stats_reset = eth_stats_reset,
+ .mac_addr_set = eth_mac_addr_set,
+};
+
+/*
+ * This function goes through sysfs and looks for an index of szedata2
+ * device file (/dev/szedataIIX, where X is the index).
+ *
+ * @return
+ * 0 on success
+ * -1 on error
+ */
+static int
+get_szedata2_index(const struct rte_pci_addr *pcislot_addr, uint32_t *index)
+{
+ DIR *dir;
+ struct dirent *entry;
+ int ret;
+ uint32_t tmp_index;
+ FILE *fd;
+ char pcislot_path[PATH_MAX];
+ uint32_t domain;
+ uint32_t bus;
+ uint32_t devid;
+ uint32_t function;
+
+ dir = opendir("/sys/class/combo");
+ if (dir == NULL)
+ return -1;
+
+ /*
+ * Iterate through all combosixX directories.
+ * When the value in /sys/class/combo/combosixX/device/pcislot
+ * file is the location of the ethernet device dev, "X" is the
+ * index of the device.
+ */
+ while ((entry = readdir(dir)) != NULL) {
+ ret = sscanf(entry->d_name, "combosix%u", &tmp_index);
+ if (ret != 1)
+ continue;
+
+ snprintf(pcislot_path, PATH_MAX,
+ "/sys/class/combo/combosix%u/device/pcislot",
+ tmp_index);
+
+ fd = fopen(pcislot_path, "r");
+ if (fd == NULL)
+ continue;
+
+ ret = fscanf(fd, "%4" PRIx16 ":%2" PRIx8 ":%2" PRIx8 ".%" PRIx8,
+ &domain, &bus, &devid, &function);
+ fclose(fd);
+ if (ret != 4)
+ continue;
+
+ if (pcislot_addr->domain == domain &&
+ pcislot_addr->bus == bus &&
+ pcislot_addr->devid == devid &&
+ pcislot_addr->function == function) {
+ *index = tmp_index;
+ closedir(dir);
+ return 0;
+ }
+ }
+
+ closedir(dir);
+ return -1;
+}
+
+static int
+rte_szedata2_eth_dev_init(struct rte_eth_dev *dev)
+{
+ struct rte_eth_dev_data *data = dev->data;
+ struct pmd_internals *internals = (struct pmd_internals *)
+ data->dev_private;
+ struct szedata *szedata_temp;
+ int ret;
+ uint32_t szedata2_index;
+ struct rte_pci_device *pci_dev = RTE_DEV_TO_PCI(dev->device);
+ struct rte_pci_addr *pci_addr = &pci_dev->addr;
+ struct rte_mem_resource *pci_rsc =
+ &pci_dev->mem_resource[PCI_RESOURCE_NUMBER];
+ char rsc_filename[PATH_MAX];
+ void *pci_resource_ptr = NULL;
+ int fd;
+
+ RTE_LOG(INFO, PMD, "Initializing szedata2 device (" PCI_PRI_FMT ")\n",
+ pci_addr->domain, pci_addr->bus, pci_addr->devid,
+ pci_addr->function);
+
+ /* Get index of szedata2 device file and create path to device file */
+ ret = get_szedata2_index(pci_addr, &szedata2_index);
+ if (ret != 0) {
+ RTE_LOG(ERR, PMD, "Failed to get szedata2 device index!\n");
+ return -ENODEV;
+ }
+ snprintf(internals->sze_dev, PATH_MAX, SZEDATA2_DEV_PATH_FMT,
+ szedata2_index);
+
+ RTE_LOG(INFO, PMD, "SZEDATA2 path: %s\n", internals->sze_dev);
+
+ /*
+ * Get number of available DMA RX and TX channels, which is maximum
+ * number of queues that can be created and store it in private device
+ * data structure.
+ */
+ szedata_temp = szedata_open(internals->sze_dev);
+ if (szedata_temp == NULL) {
+ RTE_LOG(ERR, PMD, "szedata_open(): failed to open %s",
+ internals->sze_dev);
+ return -EINVAL;
+ }
+ internals->max_rx_queues = szedata_ifaces_available(szedata_temp,
+ SZE2_DIR_RX);
+ internals->max_tx_queues = szedata_ifaces_available(szedata_temp,
+ SZE2_DIR_TX);
+ szedata_close(szedata_temp);
+
+ RTE_LOG(INFO, PMD, "Available DMA channels RX: %u TX: %u\n",
+ internals->max_rx_queues, internals->max_tx_queues);
+
+ /* Set rx, tx burst functions */
+ if (data->dev_conf.rxmode.enable_scatter == 1 ||
+ data->scattered_rx == 1) {
+ dev->rx_pkt_burst = eth_szedata2_rx_scattered;
+ data->scattered_rx = 1;
+ } else {
+ dev->rx_pkt_burst = eth_szedata2_rx;
+ data->scattered_rx = 0;
+ }
+ dev->tx_pkt_burst = eth_szedata2_tx;
+
+ /* Set function callbacks for Ethernet API */
+ dev->dev_ops = &ops;
+
+ rte_eth_copy_pci_info(dev, pci_dev);
+
+ /* mmap pci resource0 file to rte_mem_resource structure */
+ if (pci_dev->mem_resource[PCI_RESOURCE_NUMBER].phys_addr ==
+ 0) {
+ RTE_LOG(ERR, PMD, "Missing resource%u file\n",
+ PCI_RESOURCE_NUMBER);
+ return -EINVAL;
+ }
+ snprintf(rsc_filename, PATH_MAX,
+ "%s/" PCI_PRI_FMT "/resource%u", pci_get_sysfs_path(),
+ pci_addr->domain, pci_addr->bus,
+ pci_addr->devid, pci_addr->function, PCI_RESOURCE_NUMBER);
+ fd = open(rsc_filename, O_RDWR);
+ if (fd < 0) {
+ RTE_LOG(ERR, PMD, "Could not open file %s\n", rsc_filename);
+ return -EINVAL;
+ }
+
+ pci_resource_ptr = mmap(0,
+ pci_dev->mem_resource[PCI_RESOURCE_NUMBER].len,
+ PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
+ close(fd);
+ if (pci_resource_ptr == NULL) {
+ RTE_LOG(ERR, PMD, "Could not mmap file %s (fd = %d)\n",
+ rsc_filename, fd);
+ return -EINVAL;
+ }
+ pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr = pci_resource_ptr;
+ internals->pci_rsc = pci_rsc;
+
+ RTE_LOG(DEBUG, PMD, "resource%u phys_addr = 0x%llx len = %llu "
+ "virt addr = %llx\n", PCI_RESOURCE_NUMBER,
+ (unsigned long long)pci_rsc->phys_addr,
+ (unsigned long long)pci_rsc->len,
+ (unsigned long long)pci_rsc->addr);
+
+ /* Get link state */
+ eth_link_update(dev, 0);
+
+ /* Allocate space for one mac address */
+ data->mac_addrs = rte_zmalloc(data->name, sizeof(struct ether_addr),
+ RTE_CACHE_LINE_SIZE);
+ if (data->mac_addrs == NULL) {
+ RTE_LOG(ERR, PMD, "Could not alloc space for MAC address!\n");
+ munmap(pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr,
+ pci_dev->mem_resource[PCI_RESOURCE_NUMBER].len);
+ return -EINVAL;
+ }
+
+ ether_addr_copy(&eth_addr, data->mac_addrs);
+
+ /* At initial state COMBO card is in promiscuous mode so disable it */
+ eth_promiscuous_disable(dev);
+
+ RTE_LOG(INFO, PMD, "szedata2 device ("
+ PCI_PRI_FMT ") successfully initialized\n",
+ pci_addr->domain, pci_addr->bus, pci_addr->devid,
+ pci_addr->function);
+
+ return 0;
+}
+
+static int
+rte_szedata2_eth_dev_uninit(struct rte_eth_dev *dev)
+{
+ struct rte_pci_device *pci_dev = RTE_DEV_TO_PCI(dev->device);
+ struct rte_pci_addr *pci_addr = &pci_dev->addr;
+
+ rte_free(dev->data->mac_addrs);
+ dev->data->mac_addrs = NULL;
+ munmap(pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr,
+ pci_dev->mem_resource[PCI_RESOURCE_NUMBER].len);
+
+ RTE_LOG(INFO, PMD, "szedata2 device ("
+ PCI_PRI_FMT ") successfully uninitialized\n",
+ pci_addr->domain, pci_addr->bus, pci_addr->devid,
+ pci_addr->function);
+
+ return 0;
+}
+
+static const struct rte_pci_id rte_szedata2_pci_id_table[] = {
+ {
+ RTE_PCI_DEVICE(PCI_VENDOR_ID_NETCOPE,
+ PCI_DEVICE_ID_NETCOPE_COMBO80G)
+ },
+ {
+ RTE_PCI_DEVICE(PCI_VENDOR_ID_NETCOPE,
+ PCI_DEVICE_ID_NETCOPE_COMBO100G)
+ },
+ {
+ RTE_PCI_DEVICE(PCI_VENDOR_ID_NETCOPE,
+ PCI_DEVICE_ID_NETCOPE_COMBO100G2)
+ },
+ {
+ .vendor_id = 0,
+ }
+};
+
+static int szedata2_eth_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
+ struct rte_pci_device *pci_dev)
+{
+ return rte_eth_dev_pci_generic_probe(pci_dev,
+ sizeof(struct pmd_internals), rte_szedata2_eth_dev_init);
+}
+
+static int szedata2_eth_pci_remove(struct rte_pci_device *pci_dev)
+{
+ return rte_eth_dev_pci_generic_remove(pci_dev,
+ rte_szedata2_eth_dev_uninit);
+}
+
+static struct rte_pci_driver szedata2_eth_driver = {
+ .id_table = rte_szedata2_pci_id_table,
+ .probe = szedata2_eth_pci_probe,
+ .remove = szedata2_eth_pci_remove,
+};
+
+RTE_PMD_REGISTER_PCI(RTE_SZEDATA2_DRIVER_NAME, szedata2_eth_driver);
+RTE_PMD_REGISTER_PCI_TABLE(RTE_SZEDATA2_DRIVER_NAME, rte_szedata2_pci_id_table);
+RTE_PMD_REGISTER_KMOD_DEP(RTE_SZEDATA2_DRIVER_NAME,
+ "* combo6core & combov3 & szedata2 & szedata2_cv3");
diff --git a/src/seastar/dpdk/drivers/net/szedata2/rte_eth_szedata2.h b/src/seastar/dpdk/drivers/net/szedata2/rte_eth_szedata2.h
new file mode 100644
index 00000000..afe8a383
--- /dev/null
+++ b/src/seastar/dpdk/drivers/net/szedata2/rte_eth_szedata2.h
@@ -0,0 +1,450 @@
+/*-
+ * BSD LICENSE
+ *
+ * Copyright (c) 2015 - 2016 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_
+
+#include <stdbool.h>
+
+#include <rte_byteorder.h>
+
+/* PCI Vendor ID */
+#define PCI_VENDOR_ID_NETCOPE 0x1b26
+
+/* PCI Device IDs */
+#define PCI_DEVICE_ID_NETCOPE_COMBO80G 0xcb80
+#define PCI_DEVICE_ID_NETCOPE_COMBO100G 0xc1c1
+#define PCI_DEVICE_ID_NETCOPE_COMBO100G2 0xc2c1
+
+/* number of PCI resource used by COMBO card */
+#define PCI_RESOURCE_NUMBER 0
+
+/* 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;
+};
+
+/*
+ * @return Byte from PCI resource at offset "offset".
+ */
+static inline uint8_t
+pci_resource_read8(struct rte_mem_resource *rsc, uint32_t offset)
+{
+ return *((uint8_t *)((uint8_t *)rsc->addr + offset));
+}
+
+/*
+ * @return Two bytes from PCI resource starting at offset "offset".
+ */
+static inline uint16_t
+pci_resource_read16(struct rte_mem_resource *rsc, uint32_t offset)
+{
+ return rte_le_to_cpu_16(*((uint16_t *)((uint8_t *)rsc->addr +
+ offset)));
+}
+
+/*
+ * @return Four bytes from PCI resource starting at offset "offset".
+ */
+static inline uint32_t
+pci_resource_read32(struct rte_mem_resource *rsc, uint32_t offset)
+{
+ return rte_le_to_cpu_32(*((uint32_t *)((uint8_t *)rsc->addr +
+ offset)));
+}
+
+/*
+ * @return Eight bytes from PCI resource starting at offset "offset".
+ */
+static inline uint64_t
+pci_resource_read64(struct rte_mem_resource *rsc, uint32_t offset)
+{
+ return rte_le_to_cpu_64(*((uint64_t *)((uint8_t *)rsc->addr +
+ offset)));
+}
+
+/*
+ * Write one byte to PCI resource address space at offset "offset".
+ */
+static inline void
+pci_resource_write8(struct rte_mem_resource *rsc, uint32_t offset, uint8_t val)
+{
+ *((uint8_t *)((uint8_t *)rsc->addr + offset)) = val;
+}
+
+/*
+ * Write two bytes to PCI resource address space at offset "offset".
+ */
+static inline void
+pci_resource_write16(struct rte_mem_resource *rsc, uint32_t offset,
+ uint16_t val)
+{
+ *((uint16_t *)((uint8_t *)rsc->addr + offset)) = rte_cpu_to_le_16(val);
+}
+
+/*
+ * Write four bytes to PCI resource address space at offset "offset".
+ */
+static inline void
+pci_resource_write32(struct rte_mem_resource *rsc, uint32_t offset,
+ uint32_t val)
+{
+ *((uint32_t *)((uint8_t *)rsc->addr + offset)) = rte_cpu_to_le_32(val);
+}
+
+/*
+ * Write eight bytes to PCI resource address space at offset "offset".
+ */
+static inline void
+pci_resource_write64(struct rte_mem_resource *rsc, uint32_t offset,
+ uint64_t val)
+{
+ *((uint64_t *)((uint8_t *)rsc->addr + offset)) = rte_cpu_to_le_64(val);
+}
+
+#define SZEDATA2_PCI_RESOURCE_PTR(rsc, offset, type) \
+ ((type)(((uint8_t *)(rsc)->addr) + (offset)))
+
+enum szedata2_link_speed {
+ SZEDATA2_LINK_SPEED_DEFAULT = 0,
+ SZEDATA2_LINK_SPEED_10G,
+ SZEDATA2_LINK_SPEED_40G,
+ SZEDATA2_LINK_SPEED_100G,
+};
+
+enum szedata2_mac_check_mode {
+ SZEDATA2_MAC_CHMODE_PROMISC = 0x0,
+ SZEDATA2_MAC_CHMODE_ONLY_VALID = 0x1,
+ SZEDATA2_MAC_CHMODE_ALL_BROADCAST = 0x2,
+ SZEDATA2_MAC_CHMODE_ALL_MULTICAST = 0x3,
+};
+
+/*
+ * Structure describes CGMII IBUF address space
+ */
+struct szedata2_cgmii_ibuf {
+ /** Total Received Frames Counter low part */
+ uint32_t trfcl;
+ /** Correct Frames Counter low part */
+ uint32_t cfcl;
+ /** Discarded Frames Counter low part */
+ uint32_t dfcl;
+ /** Counter of frames discarded due to buffer overflow low part */
+ uint32_t bodfcl;
+ /** Total Received Frames Counter high part */
+ uint32_t trfch;
+ /** Correct Frames Counter high part */
+ uint32_t cfch;
+ /** Discarded Frames Counter high part */
+ uint32_t dfch;
+ /** Counter of frames discarded due to buffer overflow high part */
+ uint32_t bodfch;
+ /** IBUF enable register */
+ uint32_t ibuf_en;
+ /** Error mask register */
+ uint32_t err_mask;
+ /** IBUF status register */
+ uint32_t ibuf_st;
+ /** IBUF command register */
+ uint32_t ibuf_cmd;
+ /** Minimum frame length allowed */
+ uint32_t mfla;
+ /** Frame MTU */
+ uint32_t mtu;
+ /** MAC address check mode */
+ uint32_t mac_chmode;
+ /** Octets Received OK Counter low part */
+ uint32_t orocl;
+ /** Octets Received OK Counter high part */
+ uint32_t oroch;
+} __rte_packed;
+
+/* Offset of CGMII IBUF memory for MAC addresses */
+#define SZEDATA2_CGMII_IBUF_MAC_MEM_OFF 0x80
+
+/*
+ * @return
+ * true if IBUF is enabled
+ * false if IBUF is disabled
+ */
+static inline bool
+cgmii_ibuf_is_enabled(volatile struct szedata2_cgmii_ibuf *ibuf)
+{
+ return ((rte_le_to_cpu_32(ibuf->ibuf_en) & 0x1) != 0) ? true : false;
+}
+
+/*
+ * Enables IBUF.
+ */
+static inline void
+cgmii_ibuf_enable(volatile struct szedata2_cgmii_ibuf *ibuf)
+{
+ ibuf->ibuf_en =
+ rte_cpu_to_le_32(rte_le_to_cpu_32(ibuf->ibuf_en) | 0x1);
+}
+
+/*
+ * Disables IBUF.
+ */
+static inline void
+cgmii_ibuf_disable(volatile struct szedata2_cgmii_ibuf *ibuf)
+{
+ ibuf->ibuf_en =
+ rte_cpu_to_le_32(rte_le_to_cpu_32(ibuf->ibuf_en) & ~0x1);
+}
+
+/*
+ * @return
+ * true if ibuf link is up
+ * false if ibuf link is down
+ */
+static inline bool
+cgmii_ibuf_is_link_up(volatile struct szedata2_cgmii_ibuf *ibuf)
+{
+ return ((rte_le_to_cpu_32(ibuf->ibuf_st) & 0x80) != 0) ? true : false;
+}
+
+/*
+ * @return
+ * MAC address check mode
+ */
+static inline enum szedata2_mac_check_mode
+cgmii_ibuf_mac_mode_read(volatile struct szedata2_cgmii_ibuf *ibuf)
+{
+ switch (rte_le_to_cpu_32(ibuf->mac_chmode) & 0x3) {
+ case 0x0:
+ return SZEDATA2_MAC_CHMODE_PROMISC;
+ case 0x1:
+ return SZEDATA2_MAC_CHMODE_ONLY_VALID;
+ case 0x2:
+ return SZEDATA2_MAC_CHMODE_ALL_BROADCAST;
+ case 0x3:
+ return SZEDATA2_MAC_CHMODE_ALL_MULTICAST;
+ default:
+ return SZEDATA2_MAC_CHMODE_PROMISC;
+ }
+}
+
+/*
+ * Writes "mode" in MAC address check mode register.
+ */
+static inline void
+cgmii_ibuf_mac_mode_write(volatile struct szedata2_cgmii_ibuf *ibuf,
+ enum szedata2_mac_check_mode mode)
+{
+ ibuf->mac_chmode = rte_cpu_to_le_32(
+ (rte_le_to_cpu_32(ibuf->mac_chmode) & ~0x3) | mode);
+}
+
+/*
+ * Structure describes CGMII OBUF address space
+ */
+struct szedata2_cgmii_obuf {
+ /** Total Sent Frames Counter low part */
+ uint32_t tsfcl;
+ /** Octets Sent Counter low part */
+ uint32_t oscl;
+ /** Total Discarded Frames Counter low part */
+ uint32_t tdfcl;
+ /** reserved */
+ uint32_t reserved1;
+ /** Total Sent Frames Counter high part */
+ uint32_t tsfch;
+ /** Octets Sent Counter high part */
+ uint32_t osch;
+ /** Total Discarded Frames Counter high part */
+ uint32_t tdfch;
+ /** reserved */
+ uint32_t reserved2;
+ /** OBUF enable register */
+ uint32_t obuf_en;
+ /** reserved */
+ uint64_t reserved3;
+ /** OBUF control register */
+ uint32_t ctrl;
+ /** OBUF status register */
+ uint32_t obuf_st;
+} __rte_packed;
+
+/*
+ * @return
+ * true if OBUF is enabled
+ * false if OBUF is disabled
+ */
+static inline bool
+cgmii_obuf_is_enabled(volatile struct szedata2_cgmii_obuf *obuf)
+{
+ return ((rte_le_to_cpu_32(obuf->obuf_en) & 0x1) != 0) ? true : false;
+}
+
+/*
+ * Enables OBUF.
+ */
+static inline void
+cgmii_obuf_enable(volatile struct szedata2_cgmii_obuf *obuf)
+{
+ obuf->obuf_en =
+ rte_cpu_to_le_32(rte_le_to_cpu_32(obuf->obuf_en) | 0x1);
+}
+
+/*
+ * Disables OBUF.
+ */
+static inline void
+cgmii_obuf_disable(volatile struct szedata2_cgmii_obuf *obuf)
+{
+ obuf->obuf_en =
+ rte_cpu_to_le_32(rte_le_to_cpu_32(obuf->obuf_en) & ~0x1);
+}
+
+/*
+ * Function takes value from IBUF status register. Values in IBUF and OBUF
+ * should be same.
+ *
+ * @return Link speed constant.
+ */
+static inline enum szedata2_link_speed
+cgmii_link_speed(volatile struct szedata2_cgmii_ibuf *ibuf)
+{
+ uint32_t speed = (rte_le_to_cpu_32(ibuf->ibuf_st) & 0x70) >> 4;
+ switch (speed) {
+ case 0x03:
+ return SZEDATA2_LINK_SPEED_10G;
+ case 0x04:
+ return SZEDATA2_LINK_SPEED_40G;
+ case 0x05:
+ return SZEDATA2_LINK_SPEED_100G;
+ default:
+ return SZEDATA2_LINK_SPEED_DEFAULT;
+ }
+}
+
+/*
+ * IBUFs and OBUFs can generally be located at different offsets in different
+ * firmwares.
+ * This part defines base offsets of IBUFs and OBUFs through various firmwares.
+ * Currently one firmware type is supported.
+ * Type of firmware is set through configuration option
+ * CONFIG_RTE_LIBRTE_PMD_SZEDATA_AS.
+ * Possible values are:
+ * 0 - for firmwares:
+ * NIC_100G1_LR4
+ * HANIC_100G1_LR4
+ * HANIC_100G1_SR10
+ */
+#if !defined(RTE_LIBRTE_PMD_SZEDATA2_AS)
+#error "RTE_LIBRTE_PMD_SZEDATA2_AS has to be defined"
+#elif RTE_LIBRTE_PMD_SZEDATA2_AS == 0
+
+/*
+ * CGMII IBUF offset from the beginning of PCI resource address space.
+ */
+#define SZEDATA2_CGMII_IBUF_BASE_OFF 0x8000
+/*
+ * Size of CGMII IBUF.
+ */
+#define SZEDATA2_CGMII_IBUF_SIZE 0x200
+
+/*
+ * GCMII OBUF offset from the beginning of PCI resource address space.
+ */
+#define SZEDATA2_CGMII_OBUF_BASE_OFF 0x9000
+/*
+ * Size of CGMII OBUF.
+ */
+#define SZEDATA2_CGMII_OBUF_SIZE 0x100
+
+#else
+#error "RTE_LIBRTE_PMD_SZEDATA2_AS has wrong value, see comments in config file"
+#endif
+
+#endif
diff --git a/src/seastar/dpdk/drivers/net/szedata2/rte_pmd_szedata2_version.map b/src/seastar/dpdk/drivers/net/szedata2/rte_pmd_szedata2_version.map
new file mode 100644
index 00000000..ad607bbe
--- /dev/null
+++ b/src/seastar/dpdk/drivers/net/szedata2/rte_pmd_szedata2_version.map
@@ -0,0 +1,3 @@
+DPDK_2.2 {
+ local: *;
+};